diff --git a/.mailmap b/.mailmap index d1b3c4f370c2..fd55783680cd 100644 --- a/.mailmap +++ b/.mailmap @@ -154,6 +154,9 @@ Christian Brauner Christian Marangi Christophe Ricard Christoph Hellwig +Chuck Lever +Chuck Lever +Chuck Lever Claudiu Beznea Colin Ian King Corey Minyard diff --git a/Documentation/ABI/testing/sysfs-bus-pci b/Documentation/ABI/testing/sysfs-bus-pci index ecf47559f495..7f63c7e97773 100644 --- a/Documentation/ABI/testing/sysfs-bus-pci +++ b/Documentation/ABI/testing/sysfs-bus-pci @@ -500,3 +500,75 @@ Description: console drivers from the device. Raw users of pci-sysfs resourceN attributes must be terminated prior to resizing. Success of the resizing operation is not guaranteed. + +What: /sys/bus/pci/devices/.../leds/*:enclosure:*/brightness +What: /sys/class/leds/*:enclosure:*/brightness +Date: August 2024 +KernelVersion: 6.12 +Description: + LED indications on PCIe storage enclosures which are controlled + through the NPEM interface (Native PCIe Enclosure Management, + PCIe r6.1 sec 6.28) are accessible as led class devices, both + below /sys/class/leds and below NPEM-capable PCI devices. + + Although these led class devices could be manipulated manually, + in practice they are typically manipulated automatically by an + application such as ledmon(8). + + The name of a led class device is as follows: + :enclosure: + where: + + - is the domain, bus, device and function number + (e.g. 10000:02:05.0) + - is a short description of the LED indication + + Valid indications per PCIe r6.1 table 6-27 are: + + - ok (drive is functioning normally) + - locate (drive is being identified by an admin) + - fail (drive is not functioning properly) + - rebuild (drive is part of an array that is rebuilding) + - pfa (drive is predicted to fail soon) + - hotspare (drive is marked to be used as a replacement) + - ica (drive is part of an array that is degraded) + - ifa (drive is part of an array that is failed) + - idt (drive is not the right type for the connector) + - disabled (drive is disabled, removal is safe) + - specific0 to specific7 (enclosure-specific indications) + + Broadly, the indications fall into one of these categories: + + - to signify drive state (ok, locate, fail, idt, disabled) + - to signify drive role or state in a software RAID array + (rebuild, pfa, hotspare, ica, ifa) + - to signify any other role or state (specific0 to specific7) + + Mandatory indications per PCIe r6.1 sec 7.9.19.2 comprise: + ok, locate, fail, rebuild. All others are optional. + A led class device is only visible if the corresponding + indication is supported by the device. + + To manipulate the indications, write 0 (LED_OFF) or 1 (LED_ON) + to the "brightness" file. Note that manipulating an indication + may implicitly manipulate other indications at the vendor's + discretion. E.g. when the user lights up the "ok" indication, + the vendor may choose to automatically turn off the "fail" + indication. The current state of an indication can be + retrieved by reading its "brightness" file. + + The PCIe Base Specification allows vendors leeway to choose + different colors or blinking patterns for the indications, + but they typically follow the IBPI standard. E.g. the "locate" + indication is usually presented as one or two LEDs blinking at + 4 Hz frequency: + https://en.wikipedia.org/wiki/International_Blinking_Pattern_Interpretation + + PCI Firmware Specification r3.3 sec 4.7 defines a DSM interface + to facilitate shared access by operating system and platform + firmware to a device's NPEM registers. The kernel will use + this DSM interface where available, instead of accessing NPEM + registers directly. The DSM interface does not support the + enclosure-specific indications "specific0" to "specific7", + hence the corresponding led class devices are unavailable if + the DSM interface is used. diff --git a/Documentation/devicetree/bindings/pci/altera-pcie-msi.txt b/Documentation/devicetree/bindings/pci/altera-pcie-msi.txt deleted file mode 100644 index 9514c327d31b..000000000000 --- a/Documentation/devicetree/bindings/pci/altera-pcie-msi.txt +++ /dev/null @@ -1,27 +0,0 @@ -* Altera PCIe MSI controller - -Required properties: -- compatible: should contain "altr,msi-1.0" -- reg: specifies the physical base address of the controller and - the length of the memory mapped region. -- reg-names: must include the following entries: - "csr": CSR registers - "vector_slave": vectors slave port region -- interrupts: specifies the interrupt source of the parent interrupt - controller. The format of the interrupt specifier depends on the - parent interrupt controller. -- num-vectors: number of vectors, range 1 to 32. -- msi-controller: indicates that this is MSI controller node - - -Example -msi0: msi@0xFF200000 { - compatible = "altr,msi-1.0"; - reg = <0xFF200000 0x00000010 - 0xFF200010 0x00000080>; - reg-names = "csr", "vector_slave"; - interrupt-parent = <&hps_0_arm_gic_0>; - interrupts = <0 42 4>; - msi-controller; - num-vectors = <32>; -}; diff --git a/Documentation/devicetree/bindings/pci/altera-pcie.txt b/Documentation/devicetree/bindings/pci/altera-pcie.txt deleted file mode 100644 index 816b244a221e..000000000000 --- a/Documentation/devicetree/bindings/pci/altera-pcie.txt +++ /dev/null @@ -1,50 +0,0 @@ -* Altera PCIe controller - -Required properties: -- compatible : should contain "altr,pcie-root-port-1.0" or "altr,pcie-root-port-2.0" -- reg: a list of physical base address and length for TXS and CRA. - For "altr,pcie-root-port-2.0", additional HIP base address and length. -- reg-names: must include the following entries: - "Txs": TX slave port region - "Cra": Control register access region - "Hip": Hard IP region (if "altr,pcie-root-port-2.0") -- interrupts: specifies the interrupt source of the parent interrupt - controller. The format of the interrupt specifier depends - on the parent interrupt controller. -- device_type: must be "pci" -- #address-cells: set to <3> -- #size-cells: set to <2> -- #interrupt-cells: set to <1> -- ranges: describes the translation of addresses for root ports and - standard PCI regions. -- interrupt-map-mask and interrupt-map: standard PCI properties to define the - mapping of the PCIe interface to interrupt numbers. - -Optional properties: -- msi-parent: Link to the hardware entity that serves as the MSI controller - for this PCIe controller. -- bus-range: PCI bus numbers covered - -Example - pcie_0: pcie@c00000000 { - compatible = "altr,pcie-root-port-1.0"; - reg = <0xc0000000 0x20000000>, - <0xff220000 0x00004000>; - reg-names = "Txs", "Cra"; - interrupt-parent = <&hps_0_arm_gic_0>; - interrupts = <0 40 4>; - interrupt-controller; - #interrupt-cells = <1>; - bus-range = <0x0 0xFF>; - device_type = "pci"; - msi-parent = <&msi_to_gic_gen_0>; - #address-cells = <3>; - #size-cells = <2>; - interrupt-map-mask = <0 0 0 7>; - interrupt-map = <0 0 0 1 &pcie_0 1>, - <0 0 0 2 &pcie_0 2>, - <0 0 0 3 &pcie_0 3>, - <0 0 0 4 &pcie_0 4>; - ranges = <0x82000000 0x00000000 0x00000000 0xc0000000 0x00000000 0x10000000 - 0x82000000 0x00000000 0x10000000 0xd0000000 0x00000000 0x10000000>; - }; diff --git a/Documentation/devicetree/bindings/pci/altr,msi-controller.yaml b/Documentation/devicetree/bindings/pci/altr,msi-controller.yaml new file mode 100644 index 000000000000..98814862d006 --- /dev/null +++ b/Documentation/devicetree/bindings/pci/altr,msi-controller.yaml @@ -0,0 +1,65 @@ +# SPDX-License-Identifier: (GPL-2.0 OR BSD-2-Clause) +# Copyright (C) 2015, 2024, Intel Corporation +%YAML 1.2 +--- +$id: http://devicetree.org/schemas/altr,msi-controller.yaml# +$schema: http://devicetree.org/meta-schemas/core.yaml# + +title: Altera PCIe MSI controller + +maintainers: + - Matthew Gerlach + +properties: + compatible: + enum: + - altr,msi-1.0 + + reg: + items: + - description: CSR registers + - description: Vectors slave port region + + reg-names: + items: + - const: csr + - const: vector_slave + + interrupts: + maxItems: 1 + + msi-controller: true + + num-vectors: + description: number of vectors + $ref: /schemas/types.yaml#/definitions/uint32 + minimum: 1 + maximum: 32 + +required: + - compatible + - reg + - reg-names + - interrupts + - msi-controller + - num-vectors + +allOf: + - $ref: /schemas/interrupt-controller/msi-controller.yaml# + +unevaluatedProperties: false + +examples: + - | + #include + #include + msi@ff200000 { + compatible = "altr,msi-1.0"; + reg = <0xff200000 0x00000010>, + <0xff200010 0x00000080>; + reg-names = "csr", "vector_slave"; + interrupt-parent = <&hps_0_arm_gic_0>; + interrupts = ; + msi-controller; + num-vectors = <32>; + }; diff --git a/Documentation/devicetree/bindings/pci/altr,pcie-root-port.yaml b/Documentation/devicetree/bindings/pci/altr,pcie-root-port.yaml new file mode 100644 index 000000000000..52533fccc134 --- /dev/null +++ b/Documentation/devicetree/bindings/pci/altr,pcie-root-port.yaml @@ -0,0 +1,114 @@ +# SPDX-License-Identifier: (GPL-2.0 OR BSD-2-Clause) +# Copyright (C) 2015, 2019, 2024, Intel Corporation +%YAML 1.2 +--- +$id: http://devicetree.org/schemas/altr,pcie-root-port.yaml# +$schema: http://devicetree.org/meta-schemas/core.yaml# + +title: Altera PCIe Root Port + +maintainers: + - Matthew Gerlach + +properties: + compatible: + enum: + - altr,pcie-root-port-1.0 + - altr,pcie-root-port-2.0 + + reg: + items: + - description: TX slave port region + - description: Control register access region + - description: Hard IP region + minItems: 2 + + reg-names: + items: + - const: Txs + - const: Cra + - const: Hip + minItems: 2 + + interrupts: + maxItems: 1 + + interrupt-controller: true + + interrupt-map-mask: + items: + - const: 0 + - const: 0 + - const: 0 + - const: 7 + + interrupt-map: + maxItems: 4 + + "#interrupt-cells": + const: 1 + + msi-parent: true + +required: + - compatible + - reg + - reg-names + - interrupts + - "#interrupt-cells" + - interrupt-controller + - interrupt-map + - interrupt-map-mask + +allOf: + - $ref: /schemas/pci/pci-host-bridge.yaml# + - if: + properties: + compatible: + enum: + - altr,pcie-root-port-1.0 + then: + properties: + reg: + maxItems: 2 + + reg-names: + maxItems: 2 + + else: + properties: + reg: + minItems: 3 + + reg-names: + minItems: 3 + + +unevaluatedProperties: false + +examples: + - | + #include + #include + pcie_0: pcie@c00000000 { + compatible = "altr,pcie-root-port-1.0"; + reg = <0xc0000000 0x20000000>, + <0xff220000 0x00004000>; + reg-names = "Txs", "Cra"; + interrupt-parent = <&hps_0_arm_gic_0>; + interrupts = ; + interrupt-controller; + #interrupt-cells = <1>; + bus-range = <0x0 0xff>; + device_type = "pci"; + msi-parent = <&msi_to_gic_gen_0>; + #address-cells = <3>; + #size-cells = <2>; + interrupt-map-mask = <0 0 0 7>; + interrupt-map = <0 0 0 1 &pcie_0 0 0 0 1>, + <0 0 0 2 &pcie_0 0 0 0 2>, + <0 0 0 3 &pcie_0 0 0 0 3>, + <0 0 0 4 &pcie_0 0 0 0 4>; + ranges = <0x82000000 0x00000000 0x00000000 0xc0000000 0x00000000 0x10000000>, + <0x82000000 0x00000000 0x10000000 0xd0000000 0x00000000 0x10000000>; + }; diff --git a/Documentation/devicetree/bindings/pci/brcm,stb-pcie.yaml b/Documentation/devicetree/bindings/pci/brcm,stb-pcie.yaml index 11f8ea33240c..0925c520195a 100644 --- a/Documentation/devicetree/bindings/pci/brcm,stb-pcie.yaml +++ b/Documentation/devicetree/bindings/pci/brcm,stb-pcie.yaml @@ -7,7 +7,7 @@ $schema: http://devicetree.org/meta-schemas/core.yaml# title: Brcmstb PCIe Host Controller maintainers: - - Nicolas Saenz Julienne + - Jim Quinlan properties: compatible: @@ -16,11 +16,12 @@ properties: - brcm,bcm2711-pcie # The Raspberry Pi 4 - brcm,bcm4908-pcie - brcm,bcm7211-pcie # Broadcom STB version of RPi4 - - brcm,bcm7278-pcie # Broadcom 7278 Arm - brcm,bcm7216-pcie # Broadcom 7216 Arm - - brcm,bcm7445-pcie # Broadcom 7445 Arm + - brcm,bcm7278-pcie # Broadcom 7278 Arm - brcm,bcm7425-pcie # Broadcom 7425 MIPs - brcm,bcm7435-pcie # Broadcom 7435 MIPs + - brcm,bcm7445-pcie # Broadcom 7445 Arm + - brcm,bcm7712-pcie # Broadcom STB sibling of Rpi 5 reg: maxItems: 1 @@ -95,6 +96,14 @@ properties: minItems: 1 maxItems: 3 + resets: + minItems: 1 + maxItems: 3 + + reset-names: + minItems: 1 + maxItems: 3 + required: - compatible - reg @@ -118,8 +127,7 @@ allOf: then: properties: resets: - items: - - description: reset controller handling the PERST# signal + maxItems: 1 reset-names: items: @@ -136,8 +144,7 @@ allOf: then: properties: resets: - items: - - description: phandle pointing to the RESCAL reset controller + maxItems: 1 reset-names: items: @@ -147,6 +154,27 @@ allOf: - resets - reset-names + - if: + properties: + compatible: + contains: + const: brcm,bcm7712-pcie + then: + properties: + resets: + minItems: 3 + maxItems: 3 + + reset-names: + items: + - const: rescal + - const: bridge + - const: swinit + + required: + - resets + - reset-names + unevaluatedProperties: false examples: diff --git a/Documentation/devicetree/bindings/pci/fsl,imx6q-pcie-ep.yaml b/Documentation/devicetree/bindings/pci/fsl,imx6q-pcie-ep.yaml index a06f75df8458..84ca12e8b25b 100644 --- a/Documentation/devicetree/bindings/pci/fsl,imx6q-pcie-ep.yaml +++ b/Documentation/devicetree/bindings/pci/fsl,imx6q-pcie-ep.yaml @@ -65,12 +65,14 @@ allOf: then: properties: reg: - minItems: 2 - maxItems: 2 + minItems: 4 + maxItems: 4 reg-names: items: - const: dbi - const: addr_space + - const: dbi2 + - const: atu - if: properties: @@ -129,8 +131,11 @@ examples: pcie_ep: pcie-ep@33800000 { compatible = "fsl,imx8mp-pcie-ep"; - reg = <0x33800000 0x000400000>, <0x18000000 0x08000000>; - reg-names = "dbi", "addr_space"; + reg = <0x33800000 0x100000>, + <0x18000000 0x8000000>, + <0x33900000 0x100000>, + <0x33b00000 0x100000>; + reg-names = "dbi", "addr_space", "dbi2", "atu"; clocks = <&clk IMX8MP_CLK_HSIO_ROOT>, <&clk IMX8MP_CLK_HSIO_AXI>, <&clk IMX8MP_CLK_PCIE_ROOT>; diff --git a/Documentation/devicetree/bindings/pci/fsl,imx6q-pcie.yaml b/Documentation/devicetree/bindings/pci/fsl,imx6q-pcie.yaml index 8b8d77b1154b..1e05c560d797 100644 --- a/Documentation/devicetree/bindings/pci/fsl,imx6q-pcie.yaml +++ b/Documentation/devicetree/bindings/pci/fsl,imx6q-pcie.yaml @@ -30,6 +30,7 @@ properties: - fsl,imx8mm-pcie - fsl,imx8mp-pcie - fsl,imx95-pcie + - fsl,imx8q-pcie clocks: minItems: 3 @@ -184,6 +185,21 @@ allOf: - const: pcie_bus - const: pcie_aux + - if: + properties: + compatible: + enum: + - fsl,imx8q-pcie + then: + properties: + clocks: + maxItems: 3 + clock-names: + items: + - const: dbi + - const: mstr + - const: slv + unevaluatedProperties: false examples: diff --git a/Documentation/devicetree/bindings/pci/fsl,layerscape-pcie.yaml b/Documentation/devicetree/bindings/pci/fsl,layerscape-pcie.yaml index 793986c5af7f..be79712836c4 100644 --- a/Documentation/devicetree/bindings/pci/fsl,layerscape-pcie.yaml +++ b/Documentation/devicetree/bindings/pci/fsl,layerscape-pcie.yaml @@ -22,18 +22,20 @@ description: properties: compatible: - enum: - - fsl,ls1021a-pcie - - fsl,ls2080a-pcie - - fsl,ls2085a-pcie - - fsl,ls2088a-pcie - - fsl,ls1088a-pcie - - fsl,ls1046a-pcie - - fsl,ls1043a-pcie - - fsl,ls1012a-pcie - - fsl,ls1028a-pcie - - fsl,lx2160a-pcie - + oneOf: + - enum: + - fsl,ls1012a-pcie + - fsl,ls1021a-pcie + - fsl,ls1028a-pcie + - fsl,ls1043a-pcie + - fsl,ls1046a-pcie + - fsl,ls1088a-pcie + - fsl,ls2080a-pcie + - fsl,ls2085a-pcie + - fsl,ls2088a-pcie + - items: + - const: fsl,lx2160ar2-pcie + - const: fsl,ls2088a-pcie reg: maxItems: 2 @@ -43,10 +45,15 @@ properties: - const: config fsl,pcie-scfg: - $ref: /schemas/types.yaml#/definitions/phandle + $ref: /schemas/types.yaml#/definitions/phandle-array description: A phandle to the SCFG device node. The second entry is the physical PCIe controller index starting from '0'. This is used to get SCFG PEXN registers. + items: + items: + - description: A phandle to the SCFG device node + - description: PCIe controller index starting from '0' + maxItems: 1 big-endian: $ref: /schemas/types.yaml#/definitions/flag @@ -67,6 +74,14 @@ properties: minItems: 1 maxItems: 2 + num-viewport: + $ref: /schemas/types.yaml#/definitions/uint32 + deprecated: true + description: + Number of outbound view ports configured in hardware. It's the same as + the number of outbound AT windows. + maximum: 256 + required: - compatible - reg diff --git a/Documentation/devicetree/bindings/pci/hisilicon,kirin-pcie.yaml b/Documentation/devicetree/bindings/pci/hisilicon,kirin-pcie.yaml index c9f04999c9cf..e863519f3161 100644 --- a/Documentation/devicetree/bindings/pci/hisilicon,kirin-pcie.yaml +++ b/Documentation/devicetree/bindings/pci/hisilicon,kirin-pcie.yaml @@ -37,7 +37,8 @@ properties: minItems: 3 maxItems: 4 - clocks: true + clocks: + maxItems: 5 clock-names: items: diff --git a/Documentation/devicetree/bindings/pci/host-generic-pci.yaml b/Documentation/devicetree/bindings/pci/host-generic-pci.yaml index bcfbaf5582cc..420d551e9af9 100644 --- a/Documentation/devicetree/bindings/pci/host-generic-pci.yaml +++ b/Documentation/devicetree/bindings/pci/host-generic-pci.yaml @@ -102,8 +102,6 @@ properties: As described in IEEE Std 1275-1994, but must provide at least a definition of non-prefetchable memory. One or both of prefetchable Memory and IO Space may also be provided. - minItems: 1 - maxItems: 3 dma-coherent: true iommu-map: true diff --git a/Documentation/devicetree/bindings/pci/mediatek-pcie-gen3.yaml b/Documentation/devicetree/bindings/pci/mediatek-pcie-gen3.yaml index 76d742051f73..898c1be2d6a4 100644 --- a/Documentation/devicetree/bindings/pci/mediatek-pcie-gen3.yaml +++ b/Documentation/devicetree/bindings/pci/mediatek-pcie-gen3.yaml @@ -53,6 +53,7 @@ properties: - mediatek,mt8195-pcie - const: mediatek,mt8192-pcie - const: mediatek,mt8192-pcie + - const: airoha,en7581-pcie reg: maxItems: 1 @@ -76,20 +77,20 @@ properties: resets: minItems: 1 - maxItems: 2 + maxItems: 3 reset-names: minItems: 1 - maxItems: 2 + maxItems: 3 items: - enum: [ phy, mac ] + enum: [ phy, mac, phy-lane0, phy-lane1, phy-lane2 ] clocks: - minItems: 4 + minItems: 1 maxItems: 6 clock-names: - minItems: 4 + minItems: 1 maxItems: 6 assigned-clocks: @@ -147,6 +148,9 @@ allOf: const: mediatek,mt8192-pcie then: properties: + clocks: + minItems: 4 + clock-names: items: - const: pl_250m @@ -155,6 +159,15 @@ allOf: - const: tl_32k - const: peri_26m - const: top_133m + + resets: + minItems: 1 + maxItems: 2 + + reset-names: + minItems: 1 + maxItems: 2 + - if: properties: compatible: @@ -164,6 +177,9 @@ allOf: - mediatek,mt8195-pcie then: properties: + clocks: + minItems: 4 + clock-names: items: - const: pl_250m @@ -172,6 +188,15 @@ allOf: - const: tl_32k - const: peri_26m - const: peri_mem + + resets: + minItems: 1 + maxItems: 2 + + reset-names: + minItems: 1 + maxItems: 2 + - if: properties: compatible: @@ -180,6 +205,9 @@ allOf: - mediatek,mt7986-pcie then: properties: + clocks: + minItems: 4 + clock-names: items: - const: pl_250m @@ -187,6 +215,36 @@ allOf: - const: peri_26m - const: top_133m + resets: + minItems: 1 + maxItems: 2 + + reset-names: + minItems: 1 + maxItems: 2 + + - if: + properties: + compatible: + const: airoha,en7581-pcie + then: + properties: + clocks: + maxItems: 1 + + clock-names: + items: + - const: sys-ck + + resets: + minItems: 3 + + reset-names: + items: + - const: phy-lane0 + - const: phy-lane1 + - const: phy-lane2 + unevaluatedProperties: false examples: diff --git a/Documentation/devicetree/bindings/pci/pci-ep.yaml b/Documentation/devicetree/bindings/pci/pci-ep.yaml index d1eef4825207..f75000e3093d 100644 --- a/Documentation/devicetree/bindings/pci/pci-ep.yaml +++ b/Documentation/devicetree/bindings/pci/pci-ep.yaml @@ -10,7 +10,8 @@ description: | Common properties for PCI Endpoint Controller Nodes. maintainers: - - Kishon Vijay Abraham I + - Kishon Vijay Abraham I + - Manivannan Sadhasivam properties: $nodename: @@ -41,6 +42,17 @@ properties: default: 1 maximum: 16 + linux,pci-domain: + description: + If present this property assigns a fixed PCI domain number to a PCI + Endpoint Controller, otherwise an unstable (across boots) unique number + will be assigned. It is required to either not set this property at all + or set it for all PCI endpoint controllers in the system, otherwise + potentially conflicting domain numbers may be assigned to endpoint + controllers. The domain number for each endpoint controller in the system + must be unique. + $ref: /schemas/types.yaml#/definitions/uint32 + required: - compatible diff --git a/Documentation/devicetree/bindings/pci/qcom,pcie-common.yaml b/Documentation/devicetree/bindings/pci/qcom,pcie-common.yaml index 0a39bbfcb28b..e18900c41576 100644 --- a/Documentation/devicetree/bindings/pci/qcom,pcie-common.yaml +++ b/Documentation/devicetree/bindings/pci/qcom,pcie-common.yaml @@ -21,11 +21,11 @@ properties: interrupts: minItems: 1 - maxItems: 8 + maxItems: 9 interrupt-names: minItems: 1 - maxItems: 8 + maxItems: 9 iommu-map: minItems: 1 @@ -78,6 +78,9 @@ properties: description: GPIO controlled connection to WAKE# signal maxItems: 1 + vddpe-3v3-supply: + description: PCIe endpoint power supply + required: - reg - reg-names diff --git a/Documentation/devicetree/bindings/pci/qcom,pcie-ep.yaml b/Documentation/devicetree/bindings/pci/qcom,pcie-ep.yaml index 46802f7d9482..1226ee5d08d1 100644 --- a/Documentation/devicetree/bindings/pci/qcom,pcie-ep.yaml +++ b/Documentation/devicetree/bindings/pci/qcom,pcie-ep.yaml @@ -280,4 +280,5 @@ examples: phy-names = "pciephy"; max-link-speed = <3>; num-lanes = <2>; + linux,pci-domain = <0>; }; diff --git a/Documentation/devicetree/bindings/pci/qcom,pcie-sc7280.yaml b/Documentation/devicetree/bindings/pci/qcom,pcie-sc7280.yaml index 634da24ec3ed..76cb9fbfd476 100644 --- a/Documentation/devicetree/bindings/pci/qcom,pcie-sc7280.yaml +++ b/Documentation/devicetree/bindings/pci/qcom,pcie-sc7280.yaml @@ -53,11 +53,19 @@ properties: - const: aggre1 # Aggre NoC PCIe1 AXI clock interrupts: - maxItems: 1 + minItems: 8 + maxItems: 8 interrupt-names: items: - - const: msi + - const: msi0 + - const: msi1 + - const: msi2 + - const: msi3 + - const: msi4 + - const: msi5 + - const: msi6 + - const: msi7 resets: maxItems: 1 @@ -66,9 +74,6 @@ properties: items: - const: pci - vddpe-3v3-supply: - description: PCIe endpoint power supply - allOf: - $ref: qcom,pcie-common.yaml# @@ -137,8 +142,16 @@ examples: dma-coherent; - interrupts = ; - interrupt-names = "msi"; + interrupts = , + , + , + , + , + , + , + ; + interrupt-names = "msi0", "msi1", "msi2", "msi3", + "msi4", "msi5", "msi6", "msi7"; #interrupt-cells = <1>; interrupt-map-mask = <0 0 0 0x7>; interrupt-map = <0 0 0 1 &intc 0 0 0 434 IRQ_TYPE_LEVEL_HIGH>, diff --git a/Documentation/devicetree/bindings/pci/qcom,pcie-sc8280xp.yaml b/Documentation/devicetree/bindings/pci/qcom,pcie-sc8280xp.yaml index 25c9f13ae977..15ba2385eb73 100644 --- a/Documentation/devicetree/bindings/pci/qcom,pcie-sc8280xp.yaml +++ b/Documentation/devicetree/bindings/pci/qcom,pcie-sc8280xp.yaml @@ -58,9 +58,6 @@ properties: items: - const: pci - vddpe-3v3-supply: - description: A phandle to the PCIe endpoint power supply - required: - interconnects - interconnect-names diff --git a/Documentation/devicetree/bindings/pci/qcom,pcie-sm8450.yaml b/Documentation/devicetree/bindings/pci/qcom,pcie-sm8450.yaml index d8c0afaa4b19..46bd59eefadb 100644 --- a/Documentation/devicetree/bindings/pci/qcom,pcie-sm8450.yaml +++ b/Documentation/devicetree/bindings/pci/qcom,pcie-sm8450.yaml @@ -55,8 +55,8 @@ properties: - const: aggre1 # Aggre NoC PCIe1 AXI clock interrupts: - minItems: 8 - maxItems: 8 + minItems: 9 + maxItems: 9 interrupt-names: items: @@ -68,6 +68,7 @@ properties: - const: msi5 - const: msi6 - const: msi7 + - const: global operating-points-v2: true opp-table: @@ -149,9 +150,10 @@ examples: , , , - ; + , + ; interrupt-names = "msi0", "msi1", "msi2", "msi3", - "msi4", "msi5", "msi6", "msi7"; + "msi4", "msi5", "msi6", "msi7", "global"; #interrupt-cells = <1>; interrupt-map-mask = <0 0 0 0x7>; interrupt-map = <0 0 0 1 &intc 0 0 0 149 IRQ_TYPE_LEVEL_HIGH>, /* int_a */ diff --git a/Documentation/devicetree/bindings/pci/qcom,pcie.yaml b/Documentation/devicetree/bindings/pci/qcom,pcie.yaml index f867746b1ae5..ffabbac57fc1 100644 --- a/Documentation/devicetree/bindings/pci/qcom,pcie.yaml +++ b/Documentation/devicetree/bindings/pci/qcom,pcie.yaml @@ -91,6 +91,9 @@ properties: vdda_refclk-supply: description: A phandle to the core analog power supply for IC which generates reference clock + vddpe-3v3-supply: + description: A phandle to the PCIe endpoint power supply + phys: maxItems: 1 diff --git a/Documentation/devicetree/bindings/pci/rcar-gen4-pci-ep.yaml b/Documentation/devicetree/bindings/pci/rcar-gen4-pci-ep.yaml index 91b81ac75592..b23293314a6d 100644 --- a/Documentation/devicetree/bindings/pci/rcar-gen4-pci-ep.yaml +++ b/Documentation/devicetree/bindings/pci/rcar-gen4-pci-ep.yaml @@ -19,6 +19,7 @@ properties: - enum: - renesas,r8a779f0-pcie-ep # R-Car S4-8 - renesas,r8a779g0-pcie-ep # R-Car V4H + - renesas,r8a779h0-pcie-ep # R-Car V4M - const: renesas,rcar-gen4-pcie-ep # R-Car Gen4 reg: diff --git a/Documentation/devicetree/bindings/pci/rcar-gen4-pci-host.yaml b/Documentation/devicetree/bindings/pci/rcar-gen4-pci-host.yaml index 955c664f1fbb..bb3f843c59d9 100644 --- a/Documentation/devicetree/bindings/pci/rcar-gen4-pci-host.yaml +++ b/Documentation/devicetree/bindings/pci/rcar-gen4-pci-host.yaml @@ -19,6 +19,7 @@ properties: - enum: - renesas,r8a779f0-pcie # R-Car S4-8 - renesas,r8a779g0-pcie # R-Car V4H + - renesas,r8a779h0-pcie # R-Car V4M - const: renesas,rcar-gen4-pcie # R-Car Gen4 reg: diff --git a/Documentation/devicetree/bindings/pci/renesas,pci-rcar-gen2.yaml b/Documentation/devicetree/bindings/pci/renesas,pci-rcar-gen2.yaml index b288cdb1ec70..065b7508d288 100644 --- a/Documentation/devicetree/bindings/pci/renesas,pci-rcar-gen2.yaml +++ b/Documentation/devicetree/bindings/pci/renesas,pci-rcar-gen2.yaml @@ -42,9 +42,13 @@ properties: interrupts: maxItems: 1 - clocks: true + clocks: + minItems: 1 + maxItems: 3 - clock-names: true + clock-names: + minItems: 1 + maxItems: 3 resets: maxItems: 1 diff --git a/Documentation/devicetree/bindings/pci/socionext,uniphier-pcie-ep.yaml b/Documentation/devicetree/bindings/pci/socionext,uniphier-pcie-ep.yaml index f0d8e486a07d..93f3d0f4bb94 100644 --- a/Documentation/devicetree/bindings/pci/socionext,uniphier-pcie-ep.yaml +++ b/Documentation/devicetree/bindings/pci/socionext,uniphier-pcie-ep.yaml @@ -38,13 +38,17 @@ properties: minItems: 1 maxItems: 2 - clock-names: true + clock-names: + minItems: 1 + maxItems: 2 resets: minItems: 1 maxItems: 2 - reset-names: true + reset-names: + minItems: 1 + maxItems: 2 num-ib-windows: const: 16 diff --git a/Documentation/devicetree/bindings/pci/ti,j721e-pci-host.yaml b/Documentation/devicetree/bindings/pci/ti,j721e-pci-host.yaml index 15a2658ceeef..69b499c96c71 100644 --- a/Documentation/devicetree/bindings/pci/ti,j721e-pci-host.yaml +++ b/Documentation/devicetree/bindings/pci/ti,j721e-pci-host.yaml @@ -38,6 +38,16 @@ properties: - const: reg - const: cfg + ti,syscon-acspcie-proxy-ctrl: + $ref: /schemas/types.yaml#/definitions/phandle-array + items: + - items: + - description: Phandle to the ACSPCIE Proxy Control Register + - description: Bitmask corresponding to the PAD IO Buffer + output enable fields (Active Low). + description: Specifier for enabling the ACSPCIE PAD outputs to drive + the reference clock to the Endpoint device. + ti,syscon-pcie-ctrl: $ref: /schemas/types.yaml#/definitions/phandle-array items: diff --git a/Documentation/devicetree/bindings/pci/xlnx,nwl-pcie.yaml b/Documentation/devicetree/bindings/pci/xlnx,nwl-pcie.yaml index 9cad860c51a3..9de3c09efb6e 100644 --- a/Documentation/devicetree/bindings/pci/xlnx,nwl-pcie.yaml +++ b/Documentation/devicetree/bindings/pci/xlnx,nwl-pcie.yaml @@ -61,6 +61,11 @@ properties: interrupt-map: maxItems: 4 + phys: + minItems: 1 + maxItems: 4 + description: One phy per logical lane, in order + power-domains: maxItems: 1 @@ -110,6 +115,7 @@ examples: - | #include #include + #include #include soc { #address-cells = <2>; @@ -138,6 +144,7 @@ examples: <0x0 0x0 0x0 0x3 &pcie_intc 0x3>, <0x0 0x0 0x0 0x4 &pcie_intc 0x4>; msi-parent = <&nwl_pcie>; + phys = <&psgtr 0 PHY_TYPE_PCIE 0 0>; power-domains = <&zynqmp_firmware PD_PCIE>; iommus = <&smmu 0x4d0>; pcie_intc: legacy-interrupt-controller { diff --git a/Documentation/devicetree/bindings/pci/xlnx,xdma-host.yaml b/Documentation/devicetree/bindings/pci/xlnx,xdma-host.yaml index 2f59b3a73dd2..f1efd919c351 100644 --- a/Documentation/devicetree/bindings/pci/xlnx,xdma-host.yaml +++ b/Documentation/devicetree/bindings/pci/xlnx,xdma-host.yaml @@ -14,10 +14,21 @@ allOf: properties: compatible: - const: xlnx,xdma-host-3.00 + enum: + - xlnx,xdma-host-3.00 + - xlnx,qdma-host-3.00 reg: - maxItems: 1 + items: + - description: configuration region and XDMA bridge register. + - description: QDMA bridge register. + minItems: 1 + + reg-names: + items: + - const: cfg + - const: breg + minItems: 1 ranges: maxItems: 2 @@ -76,6 +87,27 @@ required: - "#interrupt-cells" - interrupt-controller +if: + properties: + compatible: + contains: + enum: + - xlnx,qdma-host-3.00 +then: + properties: + reg: + minItems: 2 + reg-names: + minItems: 2 + required: + - reg-names +else: + properties: + reg: + maxItems: 1 + reg-names: + maxItems: 1 + unevaluatedProperties: false examples: diff --git a/Documentation/filesystems/bcachefs/CodingStyle.rst b/Documentation/filesystems/bcachefs/CodingStyle.rst index 0c45829a4899..01de555e21d8 100644 --- a/Documentation/filesystems/bcachefs/CodingStyle.rst +++ b/Documentation/filesystems/bcachefs/CodingStyle.rst @@ -175,7 +175,7 @@ errors in our thinking by running our code and seeing what happens. If your time is being wasted because your tools are bad or too slow - don't accept it, fix it. -Put effort into your documentation, commmit messages, and code comments - but +Put effort into your documentation, commit messages, and code comments - but don't go overboard. A good commit message is wonderful - but if the information was important enough to go in a commit message, ask yourself if it would be even better as a code comment. diff --git a/MAINTAINERS b/MAINTAINERS index 0edf7ee267f9..4b545282b3af 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -2793,7 +2793,7 @@ F: drivers/iommu/msm* F: drivers/mfd/ssbi.c F: drivers/mmc/host/mmci_qcom* F: drivers/mmc/host/sdhci-msm.c -F: drivers/pci/controller/dwc/pcie-qcom.c +F: drivers/pci/controller/dwc/pcie-qcom* F: drivers/phy/qualcomm/ F: drivers/power/*/msm* F: drivers/reset/reset-qcom-* @@ -17551,7 +17551,7 @@ PCI DRIVER FOR ALTERA PCIE IP M: Joyce Ooi L: linux-pci@vger.kernel.org S: Supported -F: Documentation/devicetree/bindings/pci/altera-pcie.txt +F: Documentation/devicetree/bindings/pci/altr,pcie-root-port.yaml F: drivers/pci/controller/pcie-altera.c PCI DRIVER FOR APPLIEDMICRO XGENE @@ -17783,7 +17783,7 @@ PCI MSI DRIVER FOR ALTERA MSI IP M: Joyce Ooi L: linux-pci@vger.kernel.org S: Supported -F: Documentation/devicetree/bindings/pci/altera-pcie-msi.txt +F: Documentation/devicetree/bindings/pci/altr,msi-controller.yaml F: drivers/pci/controller/pcie-altera-msi.c PCI MSI DRIVER FOR APPLIEDMICRO XGENE @@ -17936,6 +17936,7 @@ M: Manivannan Sadhasivam L: linux-pci@vger.kernel.org L: linux-arm-msm@vger.kernel.org S: Maintained +F: drivers/pci/controller/dwc/pcie-qcom-common.c F: drivers/pci/controller/dwc/pcie-qcom.c PCIE DRIVER FOR ROCKCHIP @@ -17972,6 +17973,7 @@ L: linux-pci@vger.kernel.org L: linux-arm-msm@vger.kernel.org S: Maintained F: Documentation/devicetree/bindings/pci/qcom,pcie-ep.yaml +F: drivers/pci/controller/dwc/pcie-qcom-common.c F: drivers/pci/controller/dwc/pcie-qcom-ep.c PCMCIA SUBSYSTEM diff --git a/arch/arm64/boot/dts/xilinx/zynqmp-zcu102-revA.dts b/arch/arm64/boot/dts/xilinx/zynqmp-zcu102-revA.dts index ad8f23a0ec67..d2175f3dd099 100644 --- a/arch/arm64/boot/dts/xilinx/zynqmp-zcu102-revA.dts +++ b/arch/arm64/boot/dts/xilinx/zynqmp-zcu102-revA.dts @@ -941,6 +941,7 @@ &pcie { status = "okay"; + phys = <&psgtr 0 PHY_TYPE_PCIE 0 0>; }; &psgtr { diff --git a/arch/s390/include/asm/pci.h b/arch/s390/include/asm/pci.h index 30820a649e6e..9d920ced6047 100644 --- a/arch/s390/include/asm/pci.h +++ b/arch/s390/include/asm/pci.h @@ -191,7 +191,14 @@ static inline bool zdev_enabled(struct zpci_dev *zdev) return (zdev->fh & (1UL << 31)) ? true : false; } -extern const struct attribute_group *zpci_attr_groups[]; +extern const struct attribute_group zpci_attr_group; +extern const struct attribute_group pfip_attr_group; +extern const struct attribute_group zpci_ident_attr_group; + +#define ARCH_PCI_DEV_GROUPS &zpci_attr_group, \ + &pfip_attr_group, \ + &zpci_ident_attr_group, + extern unsigned int s390_pci_force_floating __initdata; extern unsigned int s390_pci_no_rid; diff --git a/arch/s390/pci/Makefile b/arch/s390/pci/Makefile index 0547a10406e7..2c21f0394c9a 100644 --- a/arch/s390/pci/Makefile +++ b/arch/s390/pci/Makefile @@ -3,7 +3,8 @@ # Makefile for the s390 PCI subsystem. # -obj-$(CONFIG_PCI) += pci.o pci_irq.o pci_clp.o pci_sysfs.o \ +obj-$(CONFIG_PCI) += pci.o pci_irq.o pci_clp.o \ pci_event.o pci_debug.o pci_insn.o pci_mmio.o \ pci_bus.o pci_kvm_hook.o obj-$(CONFIG_PCI_IOV) += pci_iov.o +obj-$(CONFIG_SYSFS) += pci_sysfs.o diff --git a/arch/s390/pci/pci.c b/arch/s390/pci/pci.c index cff4838fad21..bd9624c20b80 100644 --- a/arch/s390/pci/pci.c +++ b/arch/s390/pci/pci.c @@ -587,7 +587,6 @@ int pcibios_device_add(struct pci_dev *pdev) if (pdev->is_physfn) pdev->no_vf_scan = 1; - pdev->dev.groups = zpci_attr_groups; zpci_map_resources(pdev); for (i = 0; i < PCI_STD_NUM_BARS; i++) { diff --git a/arch/s390/pci/pci_sysfs.c b/arch/s390/pci/pci_sysfs.c index 0f4f1e8fc480..1f81f6ff7b95 100644 --- a/arch/s390/pci/pci_sysfs.c +++ b/arch/s390/pci/pci_sysfs.c @@ -197,7 +197,7 @@ static struct attribute *zpci_ident_attrs[] = { NULL, }; -static struct attribute_group zpci_ident_attr_group = { +const struct attribute_group zpci_ident_attr_group = { .attrs = zpci_ident_attrs, .is_visible = zpci_index_is_visible, }; @@ -223,7 +223,7 @@ static struct attribute *zpci_dev_attrs[] = { NULL, }; -static struct attribute_group zpci_attr_group = { +const struct attribute_group zpci_attr_group = { .attrs = zpci_dev_attrs, .bin_attrs = zpci_bin_attrs, }; @@ -235,14 +235,8 @@ static struct attribute *pfip_attrs[] = { &dev_attr_segment3.attr, NULL, }; -static struct attribute_group pfip_attr_group = { + +const struct attribute_group pfip_attr_group = { .name = "pfip", .attrs = pfip_attrs, }; - -const struct attribute_group *zpci_attr_groups[] = { - &zpci_attr_group, - &pfip_attr_group, - &zpci_ident_attr_group, - NULL, -}; diff --git a/arch/x86/pci/fixup.c b/arch/x86/pci/fixup.c index b33afb240601..98a9bb92d75c 100644 --- a/arch/x86/pci/fixup.c +++ b/arch/x86/pci/fixup.c @@ -980,7 +980,7 @@ static void amd_rp_pme_suspend(struct pci_dev *dev) return; rp = pcie_find_root_port(dev); - if (!rp->pm_cap) + if (!rp || !rp->pm_cap) return; rp->pme_support &= ~((PCI_PM_CAP_PME_D3hot|PCI_PM_CAP_PME_D3cold) >> @@ -994,7 +994,7 @@ static void amd_rp_pme_resume(struct pci_dev *dev) u16 pmc; rp = pcie_find_root_port(dev); - if (!rp->pm_cap) + if (!rp || !rp->pm_cap) return; pci_read_config_word(rp, rp->pm_cap + PCI_PM_PMC, &pmc); diff --git a/drivers/Makefile b/drivers/Makefile index fe9ceb0d2288..45d1c3e630f7 100644 --- a/drivers/Makefile +++ b/drivers/Makefile @@ -17,6 +17,9 @@ obj-$(CONFIG_PINCTRL) += pinctrl/ obj-$(CONFIG_GPIOLIB) += gpio/ obj-y += pwm/ +# LEDs must come before PCI, it is needed by NPEM driver +obj-y += leds/ + obj-y += pci/ obj-$(CONFIG_PARISC) += parisc/ @@ -130,7 +133,6 @@ obj-$(CONFIG_CPU_IDLE) += cpuidle/ obj-y += mmc/ obj-y += ufs/ obj-$(CONFIG_MEMSTICK) += memstick/ -obj-y += leds/ obj-$(CONFIG_INFINIBAND) += infiniband/ obj-y += firmware/ obj-$(CONFIG_CRYPTO) += crypto/ diff --git a/drivers/acpi/pci_mcfg.c b/drivers/acpi/pci_mcfg.c index 860014b89b8e..58e10a980114 100644 --- a/drivers/acpi/pci_mcfg.c +++ b/drivers/acpi/pci_mcfg.c @@ -181,6 +181,18 @@ static struct mcfg_fixup mcfg_quirks[] = { LOONGSON_ECAM_MCFG("LOONGSON", 0), LOONGSON_ECAM_MCFG("\0", 1), LOONGSON_ECAM_MCFG("LOONGSON", 1), + LOONGSON_ECAM_MCFG("\0", 2), + LOONGSON_ECAM_MCFG("LOONGSON", 2), + LOONGSON_ECAM_MCFG("\0", 3), + LOONGSON_ECAM_MCFG("LOONGSON", 3), + LOONGSON_ECAM_MCFG("\0", 4), + LOONGSON_ECAM_MCFG("LOONGSON", 4), + LOONGSON_ECAM_MCFG("\0", 5), + LOONGSON_ECAM_MCFG("LOONGSON", 5), + LOONGSON_ECAM_MCFG("\0", 6), + LOONGSON_ECAM_MCFG("LOONGSON", 6), + LOONGSON_ECAM_MCFG("\0", 7), + LOONGSON_ECAM_MCFG("LOONGSON", 7), #endif /* LOONGARCH */ }; diff --git a/drivers/bcma/driver_pci_host.c b/drivers/bcma/driver_pci_host.c index ed3be52ab63d..8540052d37c5 100644 --- a/drivers/bcma/driver_pci_host.c +++ b/drivers/bcma/driver_pci_host.c @@ -334,7 +334,7 @@ static u8 bcma_find_pci_capability(struct bcma_drv_pci *pc, unsigned int dev, } /* If the root port is capable of returning Config Request - * Retry Status (CRS) Completion Status to software then + * Retry Status (RRS) Completion Status to software then * enable the feature. */ static void bcma_core_pci_enable_crs(struct bcma_drv_pci *pc) @@ -348,10 +348,10 @@ static void bcma_core_pci_enable_crs(struct bcma_drv_pci *pc) NULL); root_cap = cap_ptr + PCI_EXP_RTCAP; bcma_extpci_read_config(pc, 0, 0, root_cap, &val16, sizeof(u16)); - if (val16 & BCMA_CORE_PCI_RC_CRS_VISIBILITY) { - /* Enable CRS software visibility */ + if (val16 & BCMA_CORE_PCI_RC_RRS_VISIBILITY) { + /* Enable Configuration RRS Software Visibility */ root_ctrl = cap_ptr + PCI_EXP_RTCTL; - val16 = PCI_EXP_RTCTL_CRSSVE; + val16 = PCI_EXP_RTCTL_RRS_SVE; bcma_extpci_read_config(pc, 0, 0, root_ctrl, &val16, sizeof(u16)); @@ -360,7 +360,7 @@ static void bcma_core_pci_enable_crs(struct bcma_drv_pci *pc) * 100 ms wait time from the end of Reset. If the device is * not done with its internal initialization, it must at * least return a completion TLP, with a completion status - * of "Configuration Request Retry Status (CRS)". The root + * of "Configuration Request Retry Status (RRS)". The root * complex must complete the request to the host by returning * a read-data value of 0001h for the Vendor ID field and * all 1s for any additional bytes included in the request. diff --git a/drivers/gpu/drm/ast/ast_drv.c b/drivers/gpu/drm/ast/ast_drv.c index 225817087b4d..3a908bb015fe 100644 --- a/drivers/gpu/drm/ast/ast_drv.c +++ b/drivers/gpu/drm/ast/ast_drv.c @@ -287,9 +287,9 @@ static int ast_pci_probe(struct pci_dev *pdev, const struct pci_device_id *ent) if (ret) return ret; - regs = pcim_iomap(pdev, 1, 0); - if (!regs) - return -EIO; + regs = pcim_iomap_region(pdev, 1, "ast"); + if (IS_ERR(regs)) + return PTR_ERR(regs); if (pdev->revision >= 0x40) { /* @@ -311,9 +311,9 @@ static int ast_pci_probe(struct pci_dev *pdev, const struct pci_device_id *ent) if (len < AST_IO_MM_LENGTH) return -EIO; - ioregs = pcim_iomap(pdev, 2, 0); - if (!ioregs) - return -EIO; + ioregs = pcim_iomap_region(pdev, 2, "ast"); + if (IS_ERR(ioregs)) + return PTR_ERR(ioregs); } else { /* * Anything else is best effort. diff --git a/drivers/gpu/drm/vboxvideo/vbox_main.c b/drivers/gpu/drm/vboxvideo/vbox_main.c index d4ade9325401..7f686a0190e6 100644 --- a/drivers/gpu/drm/vboxvideo/vbox_main.c +++ b/drivers/gpu/drm/vboxvideo/vbox_main.c @@ -114,6 +114,10 @@ int vbox_hw_init(struct vbox_private *vbox) DRM_INFO("VRAM %08x\n", vbox->full_vram_size); + ret = pcim_request_region(pdev, 0, "vboxvideo"); + if (ret) + return ret; + /* Map guest-heap at end of vram */ vbox->guest_heap = pcim_iomap_range(pdev, 0, GUEST_HEAP_OFFSET(vbox), GUEST_HEAP_SIZE); diff --git a/drivers/pci/Kconfig b/drivers/pci/Kconfig index aa4d1833f442..0d94e4a967d8 100644 --- a/drivers/pci/Kconfig +++ b/drivers/pci/Kconfig @@ -143,6 +143,15 @@ config PCI_IOV If unsure, say N. +config PCI_NPEM + bool "Native PCIe Enclosure Management" + depends on LEDS_CLASS=y + help + Support for Native PCIe Enclosure Management. It allows managing LED + indications in storage enclosures. Enclosure must support following + indications: OK, Locate, Fail, Rebuild, other indications are + optional. + config PCI_PRI bool "PCI PRI support" select PCI_ATS diff --git a/drivers/pci/Makefile b/drivers/pci/Makefile index 8ddad57934a6..374c5c06d92f 100644 --- a/drivers/pci/Makefile +++ b/drivers/pci/Makefile @@ -35,6 +35,7 @@ obj-$(CONFIG_XEN_PCIDEV_FRONTEND) += xen-pcifront.o obj-$(CONFIG_VGA_ARB) += vgaarb.o obj-$(CONFIG_PCI_DOE) += doe.o obj-$(CONFIG_PCI_DYNAMIC_OF_NODES) += of_property.o +obj-$(CONFIG_PCI_NPEM) += npem.o # Endpoint library must be initialized before its users obj-$(CONFIG_PCI_ENDPOINT) += endpoint/ diff --git a/drivers/pci/ats.c b/drivers/pci/ats.c index 87fa03540b8a..6afff1f1b143 100644 --- a/drivers/pci/ats.c +++ b/drivers/pci/ats.c @@ -488,8 +488,8 @@ void pci_restore_pasid_state(struct pci_dev *pdev) * pci_pasid_features - Check which PASID features are supported * @pdev: PCI device structure * - * Returns a negative value when no PASI capability is present. - * Otherwise is returns a bitmask with supported features. Current + * Return a negative value when no PASID capability is present. + * Otherwise return a bitmask with supported features. Current * features reported are: * PCI_PASID_CAP_EXEC - Execute permission supported * PCI_PASID_CAP_PRIV - Privileged mode supported diff --git a/drivers/pci/controller/Kconfig b/drivers/pci/controller/Kconfig index 4d2c188f5835..9800b7681054 100644 --- a/drivers/pci/controller/Kconfig +++ b/drivers/pci/controller/Kconfig @@ -196,7 +196,7 @@ config PCIE_MEDIATEK config PCIE_MEDIATEK_GEN3 tristate "MediaTek Gen3 PCIe controller" - depends on ARCH_MEDIATEK || COMPILE_TEST + depends on ARCH_AIROHA || ARCH_MEDIATEK || COMPILE_TEST depends on PCI_MSI help Adds support for PCIe Gen3 MAC controller for MediaTek SoCs. diff --git a/drivers/pci/controller/cadence/Kconfig b/drivers/pci/controller/cadence/Kconfig index 1d5a70c9055e..8a0044bb3989 100644 --- a/drivers/pci/controller/cadence/Kconfig +++ b/drivers/pci/controller/cadence/Kconfig @@ -38,7 +38,7 @@ config PCIE_CADENCE_PLAT_EP select PCIE_CADENCE_EP select PCIE_CADENCE_PLAT help - Say Y here if you want to support the Cadence PCIe platform controller in + Say Y here if you want to support the Cadence PCIe platform controller in endpoint mode. This PCIe controller may be embedded into many different vendors SoCs. diff --git a/drivers/pci/controller/cadence/pci-j721e.c b/drivers/pci/controller/cadence/pci-j721e.c index 85718246016b..284f2e0e4d26 100644 --- a/drivers/pci/controller/cadence/pci-j721e.c +++ b/drivers/pci/controller/cadence/pci-j721e.c @@ -7,6 +7,8 @@ */ #include +#include +#include #include #include #include @@ -22,6 +24,8 @@ #include "../../pci.h" #include "pcie-cadence.h" +#define cdns_pcie_to_rc(p) container_of(p, struct cdns_pcie_rc, pcie) + #define ENABLE_REG_SYS_2 0x108 #define STATUS_REG_SYS_2 0x508 #define STATUS_CLR_REG_SYS_2 0x708 @@ -44,6 +48,7 @@ enum link_status { #define J721E_MODE_RC BIT(7) #define LANE_COUNT(n) ((n) << 8) +#define ACSPCIE_PAD_DISABLE_MASK GENMASK(1, 0) #define GENERATION_SEL_MASK GENMASK(1, 0) struct j721e_pcie { @@ -52,6 +57,7 @@ struct j721e_pcie { u32 mode; u32 num_lanes; u32 max_lanes; + struct gpio_desc *reset_gpio; void __iomem *user_cfg_base; void __iomem *intd_cfg_base; u32 linkdown_irq_regfield; @@ -220,6 +226,36 @@ static int j721e_pcie_set_lane_count(struct j721e_pcie *pcie, return ret; } +static int j721e_enable_acspcie_refclk(struct j721e_pcie *pcie, + struct regmap *syscon) +{ + struct device *dev = pcie->cdns_pcie->dev; + struct device_node *node = dev->of_node; + u32 mask = ACSPCIE_PAD_DISABLE_MASK; + struct of_phandle_args args; + u32 val; + int ret; + + ret = of_parse_phandle_with_fixed_args(node, + "ti,syscon-acspcie-proxy-ctrl", + 1, 0, &args); + if (ret) { + dev_err(dev, + "ti,syscon-acspcie-proxy-ctrl has invalid arguments\n"); + return ret; + } + + /* Clear PAD IO disable bits to enable refclk output */ + val = ~(args.args[0]); + ret = regmap_update_bits(syscon, 0, mask, val); + if (ret) { + dev_err(dev, "failed to enable ACSPCIE refclk: %d\n", ret); + return ret; + } + + return 0; +} + static int j721e_pcie_ctrl_init(struct j721e_pcie *pcie) { struct device *dev = pcie->cdns_pcie->dev; @@ -259,7 +295,13 @@ static int j721e_pcie_ctrl_init(struct j721e_pcie *pcie) return ret; } - return 0; + /* Enable ACSPCIE refclk output if the optional property exists */ + syscon = syscon_regmap_lookup_by_phandle_optional(node, + "ti,syscon-acspcie-proxy-ctrl"); + if (!syscon) + return 0; + + return j721e_enable_acspcie_refclk(pcie, syscon); } static int cdns_ti_pcie_config_read(struct pci_bus *bus, unsigned int devfn, @@ -482,20 +524,20 @@ static int j721e_pcie_probe(struct platform_device *pdev) pm_runtime_enable(dev); ret = pm_runtime_get_sync(dev); if (ret < 0) { - dev_err(dev, "pm_runtime_get_sync failed\n"); + dev_err_probe(dev, ret, "pm_runtime_get_sync failed\n"); goto err_get_sync; } ret = j721e_pcie_ctrl_init(pcie); if (ret < 0) { - dev_err(dev, "pm_runtime_get_sync failed\n"); + dev_err_probe(dev, ret, "pm_runtime_get_sync failed\n"); goto err_get_sync; } ret = devm_request_irq(dev, irq, j721e_pcie_link_irq_handler, 0, "j721e-pcie-link-down-irq", pcie); if (ret < 0) { - dev_err(dev, "failed to request link state IRQ %d\n", irq); + dev_err_probe(dev, ret, "failed to request link state IRQ %d\n", irq); goto err_get_sync; } @@ -505,42 +547,40 @@ static int j721e_pcie_probe(struct platform_device *pdev) case PCI_MODE_RC: gpiod = devm_gpiod_get_optional(dev, "reset", GPIOD_OUT_LOW); if (IS_ERR(gpiod)) { - ret = PTR_ERR(gpiod); - if (ret != -EPROBE_DEFER) - dev_err(dev, "Failed to get reset GPIO\n"); + ret = dev_err_probe(dev, PTR_ERR(gpiod), "Failed to get reset GPIO\n"); goto err_get_sync; } + pcie->reset_gpio = gpiod; ret = cdns_pcie_init_phy(dev, cdns_pcie); if (ret) { - dev_err(dev, "Failed to init phy\n"); + dev_err_probe(dev, ret, "Failed to init phy\n"); goto err_get_sync; } clk = devm_clk_get_optional(dev, "pcie_refclk"); if (IS_ERR(clk)) { - ret = PTR_ERR(clk); - dev_err(dev, "failed to get pcie_refclk\n"); + ret = dev_err_probe(dev, PTR_ERR(clk), "failed to get pcie_refclk\n"); goto err_pcie_setup; } ret = clk_prepare_enable(clk); if (ret) { - dev_err(dev, "failed to enable pcie_refclk\n"); + dev_err_probe(dev, ret, "failed to enable pcie_refclk\n"); goto err_pcie_setup; } pcie->refclk = clk; /* - * "Power Sequencing and Reset Signal Timings" table in - * PCI EXPRESS CARD ELECTROMECHANICAL SPECIFICATION, REV. 3.0 - * indicates PERST# should be deasserted after minimum of 100us - * once REFCLK is stable. The REFCLK to the connector in RC - * mode is selected while enabling the PHY. So deassert PERST# - * after 100 us. + * The "Power Sequencing and Reset Signal Timings" table of the + * PCI Express Card Electromechanical Specification, Revision + * 5.1, Section 2.9.2, Symbol "T_PERST-CLK", indicates PERST# + * should be deasserted after minimum of 100us once REFCLK is + * stable. The REFCLK to the connector in RC mode is selected + * while enabling the PHY. So deassert PERST# after 100 us. */ if (gpiod) { - usleep_range(100, 200); + fsleep(PCIE_T_PERST_CLK_US); gpiod_set_value_cansleep(gpiod, 1); } @@ -554,7 +594,7 @@ static int j721e_pcie_probe(struct platform_device *pdev) case PCI_MODE_EP: ret = cdns_pcie_init_phy(dev, cdns_pcie); if (ret) { - dev_err(dev, "Failed to init phy\n"); + dev_err_probe(dev, ret, "Failed to init phy\n"); goto err_get_sync; } @@ -589,6 +629,87 @@ static void j721e_pcie_remove(struct platform_device *pdev) pm_runtime_disable(dev); } +static int j721e_pcie_suspend_noirq(struct device *dev) +{ + struct j721e_pcie *pcie = dev_get_drvdata(dev); + + if (pcie->mode == PCI_MODE_RC) { + gpiod_set_value_cansleep(pcie->reset_gpio, 0); + clk_disable_unprepare(pcie->refclk); + } + + cdns_pcie_disable_phy(pcie->cdns_pcie); + + return 0; +} + +static int j721e_pcie_resume_noirq(struct device *dev) +{ + struct j721e_pcie *pcie = dev_get_drvdata(dev); + struct cdns_pcie *cdns_pcie = pcie->cdns_pcie; + int ret; + + ret = j721e_pcie_ctrl_init(pcie); + if (ret < 0) + return ret; + + j721e_pcie_config_link_irq(pcie); + + /* + * This is not called explicitly in the probe, it is called by + * cdns_pcie_init_phy(). + */ + ret = cdns_pcie_enable_phy(pcie->cdns_pcie); + if (ret < 0) + return ret; + + if (pcie->mode == PCI_MODE_RC) { + struct cdns_pcie_rc *rc = cdns_pcie_to_rc(cdns_pcie); + + ret = clk_prepare_enable(pcie->refclk); + if (ret < 0) + return ret; + + /* + * The "Power Sequencing and Reset Signal Timings" table of the + * PCI Express Card Electromechanical Specification, Revision + * 5.1, Section 2.9.2, Symbol "T_PERST-CLK", indicates PERST# + * should be deasserted after minimum of 100us once REFCLK is + * stable. The REFCLK to the connector in RC mode is selected + * while enabling the PHY. So deassert PERST# after 100 us. + */ + if (pcie->reset_gpio) { + fsleep(PCIE_T_PERST_CLK_US); + gpiod_set_value_cansleep(pcie->reset_gpio, 1); + } + + ret = cdns_pcie_host_link_setup(rc); + if (ret < 0) { + clk_disable_unprepare(pcie->refclk); + return ret; + } + + /* + * Reset internal status of BARs to force reinitialization in + * cdns_pcie_host_init(). + */ + for (enum cdns_pcie_rp_bar bar = RP_BAR0; bar <= RP_NO_BAR; bar++) + rc->avail_ib_bar[bar] = true; + + ret = cdns_pcie_host_init(rc); + if (ret) { + clk_disable_unprepare(pcie->refclk); + return ret; + } + } + + return 0; +} + +static DEFINE_NOIRQ_DEV_PM_OPS(j721e_pcie_pm_ops, + j721e_pcie_suspend_noirq, + j721e_pcie_resume_noirq); + static struct platform_driver j721e_pcie_driver = { .probe = j721e_pcie_probe, .remove_new = j721e_pcie_remove, @@ -596,6 +717,7 @@ static struct platform_driver j721e_pcie_driver = { .name = "j721e-pcie", .of_match_table = of_j721e_pcie_match, .suppress_bind_attrs = true, + .pm = pm_sleep_ptr(&j721e_pcie_pm_ops), }, }; builtin_platform_driver(j721e_pcie_driver); diff --git a/drivers/pci/controller/cadence/pcie-cadence-host.c b/drivers/pci/controller/cadence/pcie-cadence-host.c index 5b14f7ee3c79..8af95e9da7ce 100644 --- a/drivers/pci/controller/cadence/pcie-cadence-host.c +++ b/drivers/pci/controller/cadence/pcie-cadence-host.c @@ -485,8 +485,7 @@ static int cdns_pcie_host_init_address_translation(struct cdns_pcie_rc *rc) return cdns_pcie_host_map_dma_ranges(rc); } -static int cdns_pcie_host_init(struct device *dev, - struct cdns_pcie_rc *rc) +int cdns_pcie_host_init(struct cdns_pcie_rc *rc) { int err; @@ -497,6 +496,30 @@ static int cdns_pcie_host_init(struct device *dev, return cdns_pcie_host_init_address_translation(rc); } +int cdns_pcie_host_link_setup(struct cdns_pcie_rc *rc) +{ + struct cdns_pcie *pcie = &rc->pcie; + struct device *dev = rc->pcie.dev; + int ret; + + if (rc->quirk_detect_quiet_flag) + cdns_pcie_detect_quiet_min_delay_set(&rc->pcie); + + cdns_pcie_host_enable_ptm_response(pcie); + + ret = cdns_pcie_start_link(pcie); + if (ret) { + dev_err(dev, "Failed to start link\n"); + return ret; + } + + ret = cdns_pcie_host_start_link(rc); + if (ret) + dev_dbg(dev, "PCIe link never came up\n"); + + return 0; +} + int cdns_pcie_host_setup(struct cdns_pcie_rc *rc) { struct device *dev = rc->pcie.dev; @@ -533,25 +556,14 @@ int cdns_pcie_host_setup(struct cdns_pcie_rc *rc) return PTR_ERR(rc->cfg_base); rc->cfg_res = res; - if (rc->quirk_detect_quiet_flag) - cdns_pcie_detect_quiet_min_delay_set(&rc->pcie); - - cdns_pcie_host_enable_ptm_response(pcie); - - ret = cdns_pcie_start_link(pcie); - if (ret) { - dev_err(dev, "Failed to start link\n"); - return ret; - } - - ret = cdns_pcie_host_start_link(rc); + ret = cdns_pcie_host_link_setup(rc); if (ret) - dev_dbg(dev, "PCIe link never came up\n"); + return ret; for (bar = RP_BAR0; bar <= RP_NO_BAR; bar++) rc->avail_ib_bar[bar] = true; - ret = cdns_pcie_host_init(dev, rc); + ret = cdns_pcie_host_init(rc); if (ret) return ret; diff --git a/drivers/pci/controller/cadence/pcie-cadence.h b/drivers/pci/controller/cadence/pcie-cadence.h index 7a66a2f815dc..f5eeff834ec1 100644 --- a/drivers/pci/controller/cadence/pcie-cadence.h +++ b/drivers/pci/controller/cadence/pcie-cadence.h @@ -314,7 +314,6 @@ struct cdns_pcie { /** * struct cdns_pcie_rc - private data for this PCIe Root Complex driver * @pcie: Cadence PCIe controller - * @dev: pointer to PCIe device * @cfg_res: start/end offsets in the physical system memory to map PCI * configuration space accesses * @cfg_base: IO mapped window to access the PCI configuration space of a @@ -521,10 +520,22 @@ static inline bool cdns_pcie_link_up(struct cdns_pcie *pcie) } #ifdef CONFIG_PCIE_CADENCE_HOST +int cdns_pcie_host_link_setup(struct cdns_pcie_rc *rc); +int cdns_pcie_host_init(struct cdns_pcie_rc *rc); int cdns_pcie_host_setup(struct cdns_pcie_rc *rc); void __iomem *cdns_pci_map_bus(struct pci_bus *bus, unsigned int devfn, int where); #else +static inline int cdns_pcie_host_link_setup(struct cdns_pcie_rc *rc) +{ + return 0; +} + +static inline int cdns_pcie_host_init(struct cdns_pcie_rc *rc) +{ + return 0; +} + static inline int cdns_pcie_host_setup(struct cdns_pcie_rc *rc) { return 0; diff --git a/drivers/pci/controller/dwc/Kconfig b/drivers/pci/controller/dwc/Kconfig index 4c38181acffa..b6d6778b0698 100644 --- a/drivers/pci/controller/dwc/Kconfig +++ b/drivers/pci/controller/dwc/Kconfig @@ -265,12 +265,16 @@ config PCIE_DW_PLAT_EP order to enable device-specific features PCI_DW_PLAT_EP must be selected. +config PCIE_QCOM_COMMON + bool + config PCIE_QCOM bool "Qualcomm PCIe controller (host mode)" depends on OF && (ARCH_QCOM || COMPILE_TEST) depends on PCI_MSI select PCIE_DW_HOST select CRC8 + select PCIE_QCOM_COMMON help Say Y here to enable PCIe controller support on Qualcomm SoCs. The PCIe controller uses the DesignWare core plus Qualcomm-specific @@ -281,6 +285,7 @@ config PCIE_QCOM_EP depends on OF && (ARCH_QCOM || COMPILE_TEST) depends on PCI_ENDPOINT select PCIE_DW_EP + select PCIE_QCOM_COMMON help Say Y here to enable support for the PCIe controllers on Qualcomm SoCs to work in endpoint mode. The PCIe controller uses the DesignWare core diff --git a/drivers/pci/controller/dwc/Makefile b/drivers/pci/controller/dwc/Makefile index ec215b3d6191..a8308d9ea986 100644 --- a/drivers/pci/controller/dwc/Makefile +++ b/drivers/pci/controller/dwc/Makefile @@ -12,6 +12,7 @@ obj-$(CONFIG_PCIE_SPEAR13XX) += pcie-spear13xx.o obj-$(CONFIG_PCI_KEYSTONE) += pci-keystone.o obj-$(CONFIG_PCI_LAYERSCAPE) += pci-layerscape.o obj-$(CONFIG_PCI_LAYERSCAPE_EP) += pci-layerscape-ep.o +obj-$(CONFIG_PCIE_QCOM_COMMON) += pcie-qcom-common.o obj-$(CONFIG_PCIE_QCOM) += pcie-qcom.o obj-$(CONFIG_PCIE_QCOM_EP) += pcie-qcom-ep.o obj-$(CONFIG_PCIE_ARMADA_8K) += pcie-armada8k.o diff --git a/drivers/pci/controller/dwc/pci-dra7xx.c b/drivers/pci/controller/dwc/pci-dra7xx.c index 4fe3b0cb72ec..5c62e1a3ba52 100644 --- a/drivers/pci/controller/dwc/pci-dra7xx.c +++ b/drivers/pci/controller/dwc/pci-dra7xx.c @@ -850,14 +850,21 @@ static int dra7xx_pcie_probe(struct platform_device *pdev) dra7xx->mode = mode; ret = devm_request_threaded_irq(dev, irq, NULL, dra7xx_pcie_irq_handler, - IRQF_SHARED, "dra7xx-pcie-main", dra7xx); + IRQF_SHARED | IRQF_ONESHOT, + "dra7xx-pcie-main", dra7xx); if (ret) { dev_err(dev, "failed to request irq\n"); - goto err_gpio; + goto err_deinit; } return 0; +err_deinit: + if (dra7xx->mode == DW_PCIE_RC_TYPE) + dw_pcie_host_deinit(&dra7xx->pci->pp); + else + dw_pcie_ep_deinit(&dra7xx->pci->ep); + err_gpio: err_get_sync: pm_runtime_put(dev); diff --git a/drivers/pci/controller/dwc/pci-imx6.c b/drivers/pci/controller/dwc/pci-imx6.c index 964d67756eb2..808d1f105417 100644 --- a/drivers/pci/controller/dwc/pci-imx6.c +++ b/drivers/pci/controller/dwc/pci-imx6.c @@ -28,6 +28,7 @@ #include #include #include +#include #include #include #include @@ -54,9 +55,9 @@ #define IMX95_PE0_GEN_CTRL_3 0x1058 #define IMX95_PCIE_LTSSM_EN BIT(0) -#define to_imx6_pcie(x) dev_get_drvdata((x)->dev) +#define to_imx_pcie(x) dev_get_drvdata((x)->dev) -enum imx6_pcie_variants { +enum imx_pcie_variants { IMX6Q, IMX6SX, IMX6QP, @@ -64,6 +65,7 @@ enum imx6_pcie_variants { IMX8MQ, IMX8MM, IMX8MP, + IMX8Q, IMX95, IMX8MQ_EP, IMX8MM_EP, @@ -71,25 +73,25 @@ enum imx6_pcie_variants { IMX95_EP, }; -#define IMX6_PCIE_FLAG_IMX6_PHY BIT(0) -#define IMX6_PCIE_FLAG_IMX6_SPEED_CHANGE BIT(1) -#define IMX6_PCIE_FLAG_SUPPORTS_SUSPEND BIT(2) -#define IMX6_PCIE_FLAG_HAS_PHYDRV BIT(3) -#define IMX6_PCIE_FLAG_HAS_APP_RESET BIT(4) -#define IMX6_PCIE_FLAG_HAS_PHY_RESET BIT(5) -#define IMX6_PCIE_FLAG_HAS_SERDES BIT(6) -#define IMX6_PCIE_FLAG_SUPPORT_64BIT BIT(7) +#define IMX_PCIE_FLAG_IMX_PHY BIT(0) +#define IMX_PCIE_FLAG_IMX_SPEED_CHANGE BIT(1) +#define IMX_PCIE_FLAG_SUPPORTS_SUSPEND BIT(2) +#define IMX_PCIE_FLAG_HAS_PHYDRV BIT(3) +#define IMX_PCIE_FLAG_HAS_APP_RESET BIT(4) +#define IMX_PCIE_FLAG_HAS_PHY_RESET BIT(5) +#define IMX_PCIE_FLAG_HAS_SERDES BIT(6) +#define IMX_PCIE_FLAG_SUPPORT_64BIT BIT(7) +#define IMX_PCIE_FLAG_CPU_ADDR_FIXUP BIT(8) -#define imx6_check_flag(pci, val) (pci->drvdata->flags & val) +#define imx_check_flag(pci, val) (pci->drvdata->flags & val) -#define IMX6_PCIE_MAX_CLKS 6 +#define IMX_PCIE_MAX_CLKS 6 +#define IMX_PCIE_MAX_INSTANCES 2 -#define IMX6_PCIE_MAX_INSTANCES 2 +struct imx_pcie; -struct imx6_pcie; - -struct imx6_pcie_drvdata { - enum imx6_pcie_variants variant; +struct imx_pcie_drvdata { + enum imx_pcie_variants variant; enum dw_pcie_device_mode mode; u32 flags; int dbi_length; @@ -98,17 +100,19 @@ struct imx6_pcie_drvdata { const u32 clks_cnt; const u32 ltssm_off; const u32 ltssm_mask; - const u32 mode_off[IMX6_PCIE_MAX_INSTANCES]; - const u32 mode_mask[IMX6_PCIE_MAX_INSTANCES]; + const u32 mode_off[IMX_PCIE_MAX_INSTANCES]; + const u32 mode_mask[IMX_PCIE_MAX_INSTANCES]; const struct pci_epc_features *epc_features; - int (*init_phy)(struct imx6_pcie *pcie); + int (*init_phy)(struct imx_pcie *pcie); + int (*enable_ref_clk)(struct imx_pcie *pcie, bool enable); + int (*core_reset)(struct imx_pcie *pcie, bool assert); }; -struct imx6_pcie { +struct imx_pcie { struct dw_pcie *pci; struct gpio_desc *reset_gpiod; bool link_is_up; - struct clk_bulk_data clks[IMX6_PCIE_MAX_CLKS]; + struct clk_bulk_data clks[IMX_PCIE_MAX_CLKS]; struct regmap *iomuxc_gpr; u16 msi_ctrl; u32 controller_id; @@ -129,7 +133,7 @@ struct imx6_pcie { /* power domain for pcie phy */ struct device *pd_pcie_phy; struct phy *phy; - const struct imx6_pcie_drvdata *drvdata; + const struct imx_pcie_drvdata *drvdata; }; /* Parameters for the waiting for PCIe PHY PLL to lock on i.MX7 */ @@ -184,28 +188,28 @@ struct imx6_pcie { #define PHY_RX_OVRD_IN_LO_RX_DATA_EN BIT(5) #define PHY_RX_OVRD_IN_LO_RX_PLL_EN BIT(3) -static unsigned int imx6_pcie_grp_offset(const struct imx6_pcie *imx6_pcie) +static unsigned int imx_pcie_grp_offset(const struct imx_pcie *imx_pcie) { - WARN_ON(imx6_pcie->drvdata->variant != IMX8MQ && - imx6_pcie->drvdata->variant != IMX8MQ_EP && - imx6_pcie->drvdata->variant != IMX8MM && - imx6_pcie->drvdata->variant != IMX8MM_EP && - imx6_pcie->drvdata->variant != IMX8MP && - imx6_pcie->drvdata->variant != IMX8MP_EP); - return imx6_pcie->controller_id == 1 ? IOMUXC_GPR16 : IOMUXC_GPR14; + WARN_ON(imx_pcie->drvdata->variant != IMX8MQ && + imx_pcie->drvdata->variant != IMX8MQ_EP && + imx_pcie->drvdata->variant != IMX8MM && + imx_pcie->drvdata->variant != IMX8MM_EP && + imx_pcie->drvdata->variant != IMX8MP && + imx_pcie->drvdata->variant != IMX8MP_EP); + return imx_pcie->controller_id == 1 ? IOMUXC_GPR16 : IOMUXC_GPR14; } -static int imx95_pcie_init_phy(struct imx6_pcie *imx6_pcie) +static int imx95_pcie_init_phy(struct imx_pcie *imx_pcie) { - regmap_update_bits(imx6_pcie->iomuxc_gpr, + regmap_update_bits(imx_pcie->iomuxc_gpr, IMX95_PCIE_SS_RW_REG_0, IMX95_PCIE_PHY_CR_PARA_SEL, IMX95_PCIE_PHY_CR_PARA_SEL); - regmap_update_bits(imx6_pcie->iomuxc_gpr, + regmap_update_bits(imx_pcie->iomuxc_gpr, IMX95_PCIE_PHY_GEN_CTRL, IMX95_PCIE_REF_USE_PAD, 0); - regmap_update_bits(imx6_pcie->iomuxc_gpr, + regmap_update_bits(imx_pcie->iomuxc_gpr, IMX95_PCIE_SS_RW_REG_0, IMX95_PCIE_REF_CLKEN, IMX95_PCIE_REF_CLKEN); @@ -213,9 +217,9 @@ static int imx95_pcie_init_phy(struct imx6_pcie *imx6_pcie) return 0; } -static void imx6_pcie_configure_type(struct imx6_pcie *imx6_pcie) +static void imx_pcie_configure_type(struct imx_pcie *imx_pcie) { - const struct imx6_pcie_drvdata *drvdata = imx6_pcie->drvdata; + const struct imx_pcie_drvdata *drvdata = imx_pcie->drvdata; unsigned int mask, val, mode, id; if (drvdata->mode == DW_PCIE_EP_TYPE) @@ -223,7 +227,11 @@ static void imx6_pcie_configure_type(struct imx6_pcie *imx6_pcie) else mode = PCI_EXP_TYPE_ROOT_PORT; - id = imx6_pcie->controller_id; + id = imx_pcie->controller_id; + + /* If mode_mask is 0, then generic PHY driver is used to set the mode */ + if (!drvdata->mode_mask[0]) + return; /* If mode_mask[id] is zero, means each controller have its individual gpr */ if (!drvdata->mode_mask[id]) @@ -232,12 +240,12 @@ static void imx6_pcie_configure_type(struct imx6_pcie *imx6_pcie) mask = drvdata->mode_mask[id]; val = mode << (ffs(mask) - 1); - regmap_update_bits(imx6_pcie->iomuxc_gpr, drvdata->mode_off[id], mask, val); + regmap_update_bits(imx_pcie->iomuxc_gpr, drvdata->mode_off[id], mask, val); } -static int pcie_phy_poll_ack(struct imx6_pcie *imx6_pcie, bool exp_val) +static int pcie_phy_poll_ack(struct imx_pcie *imx_pcie, bool exp_val) { - struct dw_pcie *pci = imx6_pcie->pci; + struct dw_pcie *pci = imx_pcie->pci; bool val; u32 max_iterations = 10; u32 wait_counter = 0; @@ -256,9 +264,9 @@ static int pcie_phy_poll_ack(struct imx6_pcie *imx6_pcie, bool exp_val) return -ETIMEDOUT; } -static int pcie_phy_wait_ack(struct imx6_pcie *imx6_pcie, int addr) +static int pcie_phy_wait_ack(struct imx_pcie *imx_pcie, int addr) { - struct dw_pcie *pci = imx6_pcie->pci; + struct dw_pcie *pci = imx_pcie->pci; u32 val; int ret; @@ -268,24 +276,24 @@ static int pcie_phy_wait_ack(struct imx6_pcie *imx6_pcie, int addr) val |= PCIE_PHY_CTRL_CAP_ADR; dw_pcie_writel_dbi(pci, PCIE_PHY_CTRL, val); - ret = pcie_phy_poll_ack(imx6_pcie, true); + ret = pcie_phy_poll_ack(imx_pcie, true); if (ret) return ret; val = PCIE_PHY_CTRL_DATA(addr); dw_pcie_writel_dbi(pci, PCIE_PHY_CTRL, val); - return pcie_phy_poll_ack(imx6_pcie, false); + return pcie_phy_poll_ack(imx_pcie, false); } /* Read from the 16-bit PCIe PHY control registers (not memory-mapped) */ -static int pcie_phy_read(struct imx6_pcie *imx6_pcie, int addr, u16 *data) +static int pcie_phy_read(struct imx_pcie *imx_pcie, int addr, u16 *data) { - struct dw_pcie *pci = imx6_pcie->pci; + struct dw_pcie *pci = imx_pcie->pci; u32 phy_ctl; int ret; - ret = pcie_phy_wait_ack(imx6_pcie, addr); + ret = pcie_phy_wait_ack(imx_pcie, addr); if (ret) return ret; @@ -293,7 +301,7 @@ static int pcie_phy_read(struct imx6_pcie *imx6_pcie, int addr, u16 *data) phy_ctl = PCIE_PHY_CTRL_RD; dw_pcie_writel_dbi(pci, PCIE_PHY_CTRL, phy_ctl); - ret = pcie_phy_poll_ack(imx6_pcie, true); + ret = pcie_phy_poll_ack(imx_pcie, true); if (ret) return ret; @@ -302,18 +310,18 @@ static int pcie_phy_read(struct imx6_pcie *imx6_pcie, int addr, u16 *data) /* deassert Read signal */ dw_pcie_writel_dbi(pci, PCIE_PHY_CTRL, 0x00); - return pcie_phy_poll_ack(imx6_pcie, false); + return pcie_phy_poll_ack(imx_pcie, false); } -static int pcie_phy_write(struct imx6_pcie *imx6_pcie, int addr, u16 data) +static int pcie_phy_write(struct imx_pcie *imx_pcie, int addr, u16 data) { - struct dw_pcie *pci = imx6_pcie->pci; + struct dw_pcie *pci = imx_pcie->pci; u32 var; int ret; /* write addr */ /* cap addr */ - ret = pcie_phy_wait_ack(imx6_pcie, addr); + ret = pcie_phy_wait_ack(imx_pcie, addr); if (ret) return ret; @@ -324,7 +332,7 @@ static int pcie_phy_write(struct imx6_pcie *imx6_pcie, int addr, u16 data) var |= PCIE_PHY_CTRL_CAP_DAT; dw_pcie_writel_dbi(pci, PCIE_PHY_CTRL, var); - ret = pcie_phy_poll_ack(imx6_pcie, true); + ret = pcie_phy_poll_ack(imx_pcie, true); if (ret) return ret; @@ -333,7 +341,7 @@ static int pcie_phy_write(struct imx6_pcie *imx6_pcie, int addr, u16 data) dw_pcie_writel_dbi(pci, PCIE_PHY_CTRL, var); /* wait for ack de-assertion */ - ret = pcie_phy_poll_ack(imx6_pcie, false); + ret = pcie_phy_poll_ack(imx_pcie, false); if (ret) return ret; @@ -342,7 +350,7 @@ static int pcie_phy_write(struct imx6_pcie *imx6_pcie, int addr, u16 data) dw_pcie_writel_dbi(pci, PCIE_PHY_CTRL, var); /* wait for ack */ - ret = pcie_phy_poll_ack(imx6_pcie, true); + ret = pcie_phy_poll_ack(imx_pcie, true); if (ret) return ret; @@ -351,7 +359,7 @@ static int pcie_phy_write(struct imx6_pcie *imx6_pcie, int addr, u16 data) dw_pcie_writel_dbi(pci, PCIE_PHY_CTRL, var); /* wait for ack de-assertion */ - ret = pcie_phy_poll_ack(imx6_pcie, false); + ret = pcie_phy_poll_ack(imx_pcie, false); if (ret) return ret; @@ -360,74 +368,74 @@ static int pcie_phy_write(struct imx6_pcie *imx6_pcie, int addr, u16 data) return 0; } -static int imx8mq_pcie_init_phy(struct imx6_pcie *imx6_pcie) +static int imx8mq_pcie_init_phy(struct imx_pcie *imx_pcie) { /* TODO: Currently this code assumes external oscillator is being used */ - regmap_update_bits(imx6_pcie->iomuxc_gpr, - imx6_pcie_grp_offset(imx6_pcie), + regmap_update_bits(imx_pcie->iomuxc_gpr, + imx_pcie_grp_offset(imx_pcie), IMX8MQ_GPR_PCIE_REF_USE_PAD, IMX8MQ_GPR_PCIE_REF_USE_PAD); /* * Regarding the datasheet, the PCIE_VPH is suggested to be 1.8V. If the PCIE_VPH is * supplied by 3.3V, the VREG_BYPASS should be cleared to zero. */ - if (imx6_pcie->vph && regulator_get_voltage(imx6_pcie->vph) > 3000000) - regmap_update_bits(imx6_pcie->iomuxc_gpr, - imx6_pcie_grp_offset(imx6_pcie), + if (imx_pcie->vph && regulator_get_voltage(imx_pcie->vph) > 3000000) + regmap_update_bits(imx_pcie->iomuxc_gpr, + imx_pcie_grp_offset(imx_pcie), IMX8MQ_GPR_PCIE_VREG_BYPASS, 0); return 0; } -static int imx7d_pcie_init_phy(struct imx6_pcie *imx6_pcie) +static int imx7d_pcie_init_phy(struct imx_pcie *imx_pcie) { - regmap_update_bits(imx6_pcie->iomuxc_gpr, IOMUXC_GPR12, IMX7D_GPR12_PCIE_PHY_REFCLK_SEL, 0); + regmap_update_bits(imx_pcie->iomuxc_gpr, IOMUXC_GPR12, IMX7D_GPR12_PCIE_PHY_REFCLK_SEL, 0); return 0; } -static int imx6_pcie_init_phy(struct imx6_pcie *imx6_pcie) +static int imx_pcie_init_phy(struct imx_pcie *imx_pcie) { - regmap_update_bits(imx6_pcie->iomuxc_gpr, IOMUXC_GPR12, + regmap_update_bits(imx_pcie->iomuxc_gpr, IOMUXC_GPR12, IMX6Q_GPR12_PCIE_CTL_2, 0 << 10); /* configure constant input signal to the pcie ctrl and phy */ - regmap_update_bits(imx6_pcie->iomuxc_gpr, IOMUXC_GPR12, + regmap_update_bits(imx_pcie->iomuxc_gpr, IOMUXC_GPR12, IMX6Q_GPR12_LOS_LEVEL, 9 << 4); - regmap_update_bits(imx6_pcie->iomuxc_gpr, IOMUXC_GPR8, + regmap_update_bits(imx_pcie->iomuxc_gpr, IOMUXC_GPR8, IMX6Q_GPR8_TX_DEEMPH_GEN1, - imx6_pcie->tx_deemph_gen1 << 0); - regmap_update_bits(imx6_pcie->iomuxc_gpr, IOMUXC_GPR8, + imx_pcie->tx_deemph_gen1 << 0); + regmap_update_bits(imx_pcie->iomuxc_gpr, IOMUXC_GPR8, IMX6Q_GPR8_TX_DEEMPH_GEN2_3P5DB, - imx6_pcie->tx_deemph_gen2_3p5db << 6); - regmap_update_bits(imx6_pcie->iomuxc_gpr, IOMUXC_GPR8, + imx_pcie->tx_deemph_gen2_3p5db << 6); + regmap_update_bits(imx_pcie->iomuxc_gpr, IOMUXC_GPR8, IMX6Q_GPR8_TX_DEEMPH_GEN2_6DB, - imx6_pcie->tx_deemph_gen2_6db << 12); - regmap_update_bits(imx6_pcie->iomuxc_gpr, IOMUXC_GPR8, + imx_pcie->tx_deemph_gen2_6db << 12); + regmap_update_bits(imx_pcie->iomuxc_gpr, IOMUXC_GPR8, IMX6Q_GPR8_TX_SWING_FULL, - imx6_pcie->tx_swing_full << 18); - regmap_update_bits(imx6_pcie->iomuxc_gpr, IOMUXC_GPR8, + imx_pcie->tx_swing_full << 18); + regmap_update_bits(imx_pcie->iomuxc_gpr, IOMUXC_GPR8, IMX6Q_GPR8_TX_SWING_LOW, - imx6_pcie->tx_swing_low << 25); + imx_pcie->tx_swing_low << 25); return 0; } -static int imx6sx_pcie_init_phy(struct imx6_pcie *imx6_pcie) +static int imx6sx_pcie_init_phy(struct imx_pcie *imx_pcie) { - regmap_update_bits(imx6_pcie->iomuxc_gpr, IOMUXC_GPR12, + regmap_update_bits(imx_pcie->iomuxc_gpr, IOMUXC_GPR12, IMX6SX_GPR12_PCIE_RX_EQ_MASK, IMX6SX_GPR12_PCIE_RX_EQ_2); - return imx6_pcie_init_phy(imx6_pcie); + return imx_pcie_init_phy(imx_pcie); } -static void imx7d_pcie_wait_for_phy_pll_lock(struct imx6_pcie *imx6_pcie) +static void imx7d_pcie_wait_for_phy_pll_lock(struct imx_pcie *imx_pcie) { u32 val; - struct device *dev = imx6_pcie->pci->dev; + struct device *dev = imx_pcie->pci->dev; - if (regmap_read_poll_timeout(imx6_pcie->iomuxc_gpr, + if (regmap_read_poll_timeout(imx_pcie->iomuxc_gpr, IOMUXC_GPR22, val, val & IMX7D_GPR22_PCIE_PHY_PLL_LOCKED, PHY_PLL_LOCK_WAIT_USLEEP_MAX, @@ -435,19 +443,19 @@ static void imx7d_pcie_wait_for_phy_pll_lock(struct imx6_pcie *imx6_pcie) dev_err(dev, "PCIe PLL lock timeout\n"); } -static int imx6_setup_phy_mpll(struct imx6_pcie *imx6_pcie) +static int imx_setup_phy_mpll(struct imx_pcie *imx_pcie) { unsigned long phy_rate = 0; int mult, div; u16 val; int i; - if (!(imx6_pcie->drvdata->flags & IMX6_PCIE_FLAG_IMX6_PHY)) + if (!(imx_pcie->drvdata->flags & IMX_PCIE_FLAG_IMX_PHY)) return 0; - for (i = 0; i < imx6_pcie->drvdata->clks_cnt; i++) - if (strncmp(imx6_pcie->clks[i].id, "pcie_phy", 8) == 0) - phy_rate = clk_get_rate(imx6_pcie->clks[i].clk); + for (i = 0; i < imx_pcie->drvdata->clks_cnt; i++) + if (strncmp(imx_pcie->clks[i].id, "pcie_phy", 8) == 0) + phy_rate = clk_get_rate(imx_pcie->clks[i].clk); switch (phy_rate) { case 125000000: @@ -465,46 +473,46 @@ static int imx6_setup_phy_mpll(struct imx6_pcie *imx6_pcie) div = 1; break; default: - dev_err(imx6_pcie->pci->dev, + dev_err(imx_pcie->pci->dev, "Unsupported PHY reference clock rate %lu\n", phy_rate); return -EINVAL; } - pcie_phy_read(imx6_pcie, PCIE_PHY_MPLL_OVRD_IN_LO, &val); + pcie_phy_read(imx_pcie, PCIE_PHY_MPLL_OVRD_IN_LO, &val); val &= ~(PCIE_PHY_MPLL_MULTIPLIER_MASK << PCIE_PHY_MPLL_MULTIPLIER_SHIFT); val |= mult << PCIE_PHY_MPLL_MULTIPLIER_SHIFT; val |= PCIE_PHY_MPLL_MULTIPLIER_OVRD; - pcie_phy_write(imx6_pcie, PCIE_PHY_MPLL_OVRD_IN_LO, val); + pcie_phy_write(imx_pcie, PCIE_PHY_MPLL_OVRD_IN_LO, val); - pcie_phy_read(imx6_pcie, PCIE_PHY_ATEOVRD, &val); + pcie_phy_read(imx_pcie, PCIE_PHY_ATEOVRD, &val); val &= ~(PCIE_PHY_ATEOVRD_REF_CLKDIV_MASK << PCIE_PHY_ATEOVRD_REF_CLKDIV_SHIFT); val |= div << PCIE_PHY_ATEOVRD_REF_CLKDIV_SHIFT; val |= PCIE_PHY_ATEOVRD_EN; - pcie_phy_write(imx6_pcie, PCIE_PHY_ATEOVRD, val); + pcie_phy_write(imx_pcie, PCIE_PHY_ATEOVRD, val); return 0; } -static void imx6_pcie_reset_phy(struct imx6_pcie *imx6_pcie) +static void imx_pcie_reset_phy(struct imx_pcie *imx_pcie) { u16 tmp; - if (!(imx6_pcie->drvdata->flags & IMX6_PCIE_FLAG_IMX6_PHY)) + if (!(imx_pcie->drvdata->flags & IMX_PCIE_FLAG_IMX_PHY)) return; - pcie_phy_read(imx6_pcie, PHY_RX_OVRD_IN_LO, &tmp); + pcie_phy_read(imx_pcie, PHY_RX_OVRD_IN_LO, &tmp); tmp |= (PHY_RX_OVRD_IN_LO_RX_DATA_EN | PHY_RX_OVRD_IN_LO_RX_PLL_EN); - pcie_phy_write(imx6_pcie, PHY_RX_OVRD_IN_LO, tmp); + pcie_phy_write(imx_pcie, PHY_RX_OVRD_IN_LO, tmp); usleep_range(2000, 3000); - pcie_phy_read(imx6_pcie, PHY_RX_OVRD_IN_LO, &tmp); + pcie_phy_read(imx_pcie, PHY_RX_OVRD_IN_LO, &tmp); tmp &= ~(PHY_RX_OVRD_IN_LO_RX_DATA_EN | PHY_RX_OVRD_IN_LO_RX_PLL_EN); - pcie_phy_write(imx6_pcie, PHY_RX_OVRD_IN_LO, tmp); + pcie_phy_write(imx_pcie, PHY_RX_OVRD_IN_LO, tmp); } #ifdef CONFIG_ARM @@ -543,22 +551,22 @@ static int imx6q_pcie_abort_handler(unsigned long addr, } #endif -static int imx6_pcie_attach_pd(struct device *dev) +static int imx_pcie_attach_pd(struct device *dev) { - struct imx6_pcie *imx6_pcie = dev_get_drvdata(dev); + struct imx_pcie *imx_pcie = dev_get_drvdata(dev); struct device_link *link; /* Do nothing when in a single power domain */ if (dev->pm_domain) return 0; - imx6_pcie->pd_pcie = dev_pm_domain_attach_by_name(dev, "pcie"); - if (IS_ERR(imx6_pcie->pd_pcie)) - return PTR_ERR(imx6_pcie->pd_pcie); + imx_pcie->pd_pcie = dev_pm_domain_attach_by_name(dev, "pcie"); + if (IS_ERR(imx_pcie->pd_pcie)) + return PTR_ERR(imx_pcie->pd_pcie); /* Do nothing when power domain missing */ - if (!imx6_pcie->pd_pcie) + if (!imx_pcie->pd_pcie) return 0; - link = device_link_add(dev, imx6_pcie->pd_pcie, + link = device_link_add(dev, imx_pcie->pd_pcie, DL_FLAG_STATELESS | DL_FLAG_PM_RUNTIME | DL_FLAG_RPM_ACTIVE); @@ -567,11 +575,11 @@ static int imx6_pcie_attach_pd(struct device *dev) return -EINVAL; } - imx6_pcie->pd_pcie_phy = dev_pm_domain_attach_by_name(dev, "pcie_phy"); - if (IS_ERR(imx6_pcie->pd_pcie_phy)) - return PTR_ERR(imx6_pcie->pd_pcie_phy); + imx_pcie->pd_pcie_phy = dev_pm_domain_attach_by_name(dev, "pcie_phy"); + if (IS_ERR(imx_pcie->pd_pcie_phy)) + return PTR_ERR(imx_pcie->pd_pcie_phy); - link = device_link_add(dev, imx6_pcie->pd_pcie_phy, + link = device_link_add(dev, imx_pcie->pd_pcie_phy, DL_FLAG_STATELESS | DL_FLAG_PM_RUNTIME | DL_FLAG_RPM_ACTIVE); @@ -583,21 +591,20 @@ static int imx6_pcie_attach_pd(struct device *dev) return 0; } -static int imx6_pcie_enable_ref_clk(struct imx6_pcie *imx6_pcie) +static int imx6sx_pcie_enable_ref_clk(struct imx_pcie *imx_pcie, bool enable) { - unsigned int offset; - int ret = 0; + if (enable) + regmap_clear_bits(imx_pcie->iomuxc_gpr, IOMUXC_GPR12, + IMX6SX_GPR12_PCIE_TEST_POWERDOWN); - switch (imx6_pcie->drvdata->variant) { - case IMX6SX: - regmap_update_bits(imx6_pcie->iomuxc_gpr, IOMUXC_GPR12, - IMX6SX_GPR12_PCIE_TEST_POWERDOWN, 0); - break; - case IMX6QP: - case IMX6Q: + return 0; +} + +static int imx6q_pcie_enable_ref_clk(struct imx_pcie *imx_pcie, bool enable) +{ + if (enable) { /* power up core phy and enable ref clock */ - regmap_update_bits(imx6_pcie->iomuxc_gpr, IOMUXC_GPR1, - IMX6Q_GPR1_PCIE_TEST_PD, 0 << 18); + regmap_clear_bits(imx_pcie->iomuxc_gpr, IOMUXC_GPR1, IMX6Q_GPR1_PCIE_TEST_PD); /* * the async reset input need ref clock to sync internally, * when the ref clock comes after reset, internal synced @@ -605,71 +612,51 @@ static int imx6_pcie_enable_ref_clk(struct imx6_pcie *imx6_pcie) * add one ~10us delay here. */ usleep_range(10, 100); - regmap_update_bits(imx6_pcie->iomuxc_gpr, IOMUXC_GPR1, - IMX6Q_GPR1_PCIE_REF_CLK_EN, 1 << 16); - break; - case IMX7D: - case IMX95: - case IMX95_EP: - break; - case IMX8MM: - case IMX8MM_EP: - case IMX8MQ: - case IMX8MQ_EP: - case IMX8MP: - case IMX8MP_EP: - offset = imx6_pcie_grp_offset(imx6_pcie); - /* - * Set the over ride low and enabled - * make sure that REF_CLK is turned on. - */ - regmap_update_bits(imx6_pcie->iomuxc_gpr, offset, - IMX8MQ_GPR_PCIE_CLK_REQ_OVERRIDE, - 0); - regmap_update_bits(imx6_pcie->iomuxc_gpr, offset, - IMX8MQ_GPR_PCIE_CLK_REQ_OVERRIDE_EN, - IMX8MQ_GPR_PCIE_CLK_REQ_OVERRIDE_EN); - break; + regmap_set_bits(imx_pcie->iomuxc_gpr, IOMUXC_GPR1, IMX6Q_GPR1_PCIE_REF_CLK_EN); + } else { + regmap_clear_bits(imx_pcie->iomuxc_gpr, IOMUXC_GPR1, IMX6Q_GPR1_PCIE_REF_CLK_EN); + regmap_set_bits(imx_pcie->iomuxc_gpr, IOMUXC_GPR1, IMX6Q_GPR1_PCIE_TEST_PD); } - return ret; + return 0; } -static void imx6_pcie_disable_ref_clk(struct imx6_pcie *imx6_pcie) +static int imx8mm_pcie_enable_ref_clk(struct imx_pcie *imx_pcie, bool enable) { - switch (imx6_pcie->drvdata->variant) { - case IMX6QP: - case IMX6Q: - regmap_update_bits(imx6_pcie->iomuxc_gpr, IOMUXC_GPR1, - IMX6Q_GPR1_PCIE_REF_CLK_EN, 0); - regmap_update_bits(imx6_pcie->iomuxc_gpr, IOMUXC_GPR1, - IMX6Q_GPR1_PCIE_TEST_PD, - IMX6Q_GPR1_PCIE_TEST_PD); - break; - case IMX7D: - regmap_update_bits(imx6_pcie->iomuxc_gpr, IOMUXC_GPR12, - IMX7D_GPR12_PCIE_PHY_REFCLK_SEL, - IMX7D_GPR12_PCIE_PHY_REFCLK_SEL); - break; - default: - break; + int offset = imx_pcie_grp_offset(imx_pcie); + + if (enable) { + regmap_clear_bits(imx_pcie->iomuxc_gpr, offset, IMX8MQ_GPR_PCIE_CLK_REQ_OVERRIDE); + regmap_set_bits(imx_pcie->iomuxc_gpr, offset, IMX8MQ_GPR_PCIE_CLK_REQ_OVERRIDE_EN); } + + return 0; } -static int imx6_pcie_clk_enable(struct imx6_pcie *imx6_pcie) +static int imx7d_pcie_enable_ref_clk(struct imx_pcie *imx_pcie, bool enable) { - struct dw_pcie *pci = imx6_pcie->pci; + if (!enable) + regmap_set_bits(imx_pcie->iomuxc_gpr, IOMUXC_GPR12, + IMX7D_GPR12_PCIE_PHY_REFCLK_SEL); + return 0; +} + +static int imx_pcie_clk_enable(struct imx_pcie *imx_pcie) +{ + struct dw_pcie *pci = imx_pcie->pci; struct device *dev = pci->dev; int ret; - ret = clk_bulk_prepare_enable(imx6_pcie->drvdata->clks_cnt, imx6_pcie->clks); + ret = clk_bulk_prepare_enable(imx_pcie->drvdata->clks_cnt, imx_pcie->clks); if (ret) return ret; - ret = imx6_pcie_enable_ref_clk(imx6_pcie); - if (ret) { - dev_err(dev, "unable to enable pcie ref clock\n"); - goto err_ref_clk; + if (imx_pcie->drvdata->enable_ref_clk) { + ret = imx_pcie->drvdata->enable_ref_clk(imx_pcie, true); + if (ret) { + dev_err(dev, "Failed to enable PCIe REFCLK\n"); + goto err_ref_clk; + } } /* allow the clocks to stabilize */ @@ -677,99 +664,120 @@ static int imx6_pcie_clk_enable(struct imx6_pcie *imx6_pcie) return 0; err_ref_clk: - clk_bulk_disable_unprepare(imx6_pcie->drvdata->clks_cnt, imx6_pcie->clks); + clk_bulk_disable_unprepare(imx_pcie->drvdata->clks_cnt, imx_pcie->clks); return ret; } -static void imx6_pcie_clk_disable(struct imx6_pcie *imx6_pcie) +static void imx_pcie_clk_disable(struct imx_pcie *imx_pcie) { - imx6_pcie_disable_ref_clk(imx6_pcie); - clk_bulk_disable_unprepare(imx6_pcie->drvdata->clks_cnt, imx6_pcie->clks); + if (imx_pcie->drvdata->enable_ref_clk) + imx_pcie->drvdata->enable_ref_clk(imx_pcie, false); + clk_bulk_disable_unprepare(imx_pcie->drvdata->clks_cnt, imx_pcie->clks); } -static void imx6_pcie_assert_core_reset(struct imx6_pcie *imx6_pcie) +static int imx6sx_pcie_core_reset(struct imx_pcie *imx_pcie, bool assert) { - reset_control_assert(imx6_pcie->pciephy_reset); - reset_control_assert(imx6_pcie->apps_reset); + if (assert) + regmap_set_bits(imx_pcie->iomuxc_gpr, IOMUXC_GPR12, + IMX6SX_GPR12_PCIE_TEST_POWERDOWN); - switch (imx6_pcie->drvdata->variant) { - case IMX6SX: - regmap_update_bits(imx6_pcie->iomuxc_gpr, IOMUXC_GPR12, - IMX6SX_GPR12_PCIE_TEST_POWERDOWN, - IMX6SX_GPR12_PCIE_TEST_POWERDOWN); - /* Force PCIe PHY reset */ - regmap_update_bits(imx6_pcie->iomuxc_gpr, IOMUXC_GPR5, - IMX6SX_GPR5_PCIE_BTNRST_RESET, - IMX6SX_GPR5_PCIE_BTNRST_RESET); - break; - case IMX6QP: - regmap_update_bits(imx6_pcie->iomuxc_gpr, IOMUXC_GPR1, - IMX6Q_GPR1_PCIE_SW_RST, - IMX6Q_GPR1_PCIE_SW_RST); - break; - case IMX6Q: - regmap_update_bits(imx6_pcie->iomuxc_gpr, IOMUXC_GPR1, - IMX6Q_GPR1_PCIE_TEST_PD, 1 << 18); - regmap_update_bits(imx6_pcie->iomuxc_gpr, IOMUXC_GPR1, - IMX6Q_GPR1_PCIE_REF_CLK_EN, 0 << 16); - break; - default: - break; - } - - /* Some boards don't have PCIe reset GPIO. */ - gpiod_set_value_cansleep(imx6_pcie->reset_gpiod, 1); + /* Force PCIe PHY reset */ + regmap_update_bits(imx_pcie->iomuxc_gpr, IOMUXC_GPR5, IMX6SX_GPR5_PCIE_BTNRST_RESET, + assert ? IMX6SX_GPR5_PCIE_BTNRST_RESET : 0); + return 0; } -static int imx6_pcie_deassert_core_reset(struct imx6_pcie *imx6_pcie) +static int imx6qp_pcie_core_reset(struct imx_pcie *imx_pcie, bool assert) { - struct dw_pcie *pci = imx6_pcie->pci; + regmap_update_bits(imx_pcie->iomuxc_gpr, IOMUXC_GPR1, IMX6Q_GPR1_PCIE_SW_RST, + assert ? IMX6Q_GPR1_PCIE_SW_RST : 0); + if (!assert) + usleep_range(200, 500); + + return 0; +} + +static int imx6q_pcie_core_reset(struct imx_pcie *imx_pcie, bool assert) +{ + if (!assert) + return 0; + + regmap_set_bits(imx_pcie->iomuxc_gpr, IOMUXC_GPR1, IMX6Q_GPR1_PCIE_TEST_PD); + regmap_set_bits(imx_pcie->iomuxc_gpr, IOMUXC_GPR1, IMX6Q_GPR1_PCIE_REF_CLK_EN); + + return 0; +} + +static int imx7d_pcie_core_reset(struct imx_pcie *imx_pcie, bool assert) +{ + struct dw_pcie *pci = imx_pcie->pci; struct device *dev = pci->dev; - reset_control_deassert(imx6_pcie->pciephy_reset); + if (assert) + return 0; - switch (imx6_pcie->drvdata->variant) { - case IMX7D: - /* Workaround for ERR010728, failure of PCI-e PLL VCO to - * oscillate, especially when cold. This turns off "Duty-cycle - * Corrector" and other mysterious undocumented things. - */ - if (likely(imx6_pcie->phy_base)) { - /* De-assert DCC_FB_EN */ - writel(PCIE_PHY_CMN_REG4_DCC_FB_EN, - imx6_pcie->phy_base + PCIE_PHY_CMN_REG4); - /* Assert RX_EQS and RX_EQS_SEL */ - writel(PCIE_PHY_CMN_REG24_RX_EQ_SEL - | PCIE_PHY_CMN_REG24_RX_EQ, - imx6_pcie->phy_base + PCIE_PHY_CMN_REG24); - /* Assert ATT_MODE */ - writel(PCIE_PHY_CMN_REG26_ATT_MODE, - imx6_pcie->phy_base + PCIE_PHY_CMN_REG26); - } else { - dev_warn(dev, "Unable to apply ERR010728 workaround. DT missing fsl,imx7d-pcie-phy phandle ?\n"); - } + /* + * Workaround for ERR010728 (IMX7DS_2N09P, Rev. 1.1, 4/2023): + * + * PCIe: PLL may fail to lock under corner conditions. + * + * Initial VCO oscillation may fail under corner conditions such as + * cold temperature which will cause the PCIe PLL fail to lock in the + * initialization phase. + * + * The Duty-cycle Corrector calibration must be disabled. + * + * 1. De-assert the G_RST signal by clearing + * SRC_PCIEPHY_RCR[PCIEPHY_G_RST]. + * 2. De-assert DCC_FB_EN by writing data “0x29” to the register + * address 0x306d0014 (PCIE_PHY_CMN_REG4). + * 3. Assert RX_EQS, RX_EQ_SEL by writing data “0x48” to the register + * address 0x306d0090 (PCIE_PHY_CMN_REG24). + * 4. Assert ATT_MODE by writing data “0xbc” to the register + * address 0x306d0098 (PCIE_PHY_CMN_REG26). + * 5. De-assert the CMN_RST signal by clearing register bit + * SRC_PCIEPHY_RCR[PCIEPHY_BTN] + */ - imx7d_pcie_wait_for_phy_pll_lock(imx6_pcie); - break; - case IMX6SX: - regmap_update_bits(imx6_pcie->iomuxc_gpr, IOMUXC_GPR5, - IMX6SX_GPR5_PCIE_BTNRST_RESET, 0); - break; - case IMX6QP: - regmap_update_bits(imx6_pcie->iomuxc_gpr, IOMUXC_GPR1, - IMX6Q_GPR1_PCIE_SW_RST, 0); - - usleep_range(200, 500); - break; - default: - break; + if (likely(imx_pcie->phy_base)) { + /* De-assert DCC_FB_EN */ + writel(PCIE_PHY_CMN_REG4_DCC_FB_EN, imx_pcie->phy_base + PCIE_PHY_CMN_REG4); + /* Assert RX_EQS and RX_EQS_SEL */ + writel(PCIE_PHY_CMN_REG24_RX_EQ_SEL | PCIE_PHY_CMN_REG24_RX_EQ, + imx_pcie->phy_base + PCIE_PHY_CMN_REG24); + /* Assert ATT_MODE */ + writel(PCIE_PHY_CMN_REG26_ATT_MODE, imx_pcie->phy_base + PCIE_PHY_CMN_REG26); + } else { + dev_warn(dev, "Unable to apply ERR010728 workaround. DT missing fsl,imx7d-pcie-phy phandle ?\n"); } + imx7d_pcie_wait_for_phy_pll_lock(imx_pcie); + return 0; +} + +static void imx_pcie_assert_core_reset(struct imx_pcie *imx_pcie) +{ + reset_control_assert(imx_pcie->pciephy_reset); + reset_control_assert(imx_pcie->apps_reset); + + if (imx_pcie->drvdata->core_reset) + imx_pcie->drvdata->core_reset(imx_pcie, true); /* Some boards don't have PCIe reset GPIO. */ - if (imx6_pcie->reset_gpiod) { + gpiod_set_value_cansleep(imx_pcie->reset_gpiod, 1); +} + +static int imx_pcie_deassert_core_reset(struct imx_pcie *imx_pcie) +{ + reset_control_deassert(imx_pcie->pciephy_reset); + + if (imx_pcie->drvdata->core_reset) + imx_pcie->drvdata->core_reset(imx_pcie, false); + + /* Some boards don't have PCIe reset GPIO. */ + if (imx_pcie->reset_gpiod) { msleep(100); - gpiod_set_value_cansleep(imx6_pcie->reset_gpiod, 0); + gpiod_set_value_cansleep(imx_pcie->reset_gpiod, 0); /* Wait for 100ms after PERST# deassertion (PCIe r5.0, 6.6.1) */ msleep(100); } @@ -777,9 +785,9 @@ static int imx6_pcie_deassert_core_reset(struct imx6_pcie *imx6_pcie) return 0; } -static int imx6_pcie_wait_for_speed_change(struct imx6_pcie *imx6_pcie) +static int imx_pcie_wait_for_speed_change(struct imx_pcie *imx_pcie) { - struct dw_pcie *pci = imx6_pcie->pci; + struct dw_pcie *pci = imx_pcie->pci; struct device *dev = pci->dev; u32 tmp; unsigned int retries; @@ -796,33 +804,38 @@ static int imx6_pcie_wait_for_speed_change(struct imx6_pcie *imx6_pcie) return -ETIMEDOUT; } -static void imx6_pcie_ltssm_enable(struct device *dev) +static void imx_pcie_ltssm_enable(struct device *dev) { - struct imx6_pcie *imx6_pcie = dev_get_drvdata(dev); - const struct imx6_pcie_drvdata *drvdata = imx6_pcie->drvdata; + struct imx_pcie *imx_pcie = dev_get_drvdata(dev); + const struct imx_pcie_drvdata *drvdata = imx_pcie->drvdata; + u8 offset = dw_pcie_find_capability(imx_pcie->pci, PCI_CAP_ID_EXP); + u32 tmp; + tmp = dw_pcie_readl_dbi(imx_pcie->pci, offset + PCI_EXP_LNKCAP); + phy_set_speed(imx_pcie->phy, FIELD_GET(PCI_EXP_LNKCAP_SLS, tmp)); if (drvdata->ltssm_mask) - regmap_update_bits(imx6_pcie->iomuxc_gpr, drvdata->ltssm_off, drvdata->ltssm_mask, + regmap_update_bits(imx_pcie->iomuxc_gpr, drvdata->ltssm_off, drvdata->ltssm_mask, drvdata->ltssm_mask); - reset_control_deassert(imx6_pcie->apps_reset); + reset_control_deassert(imx_pcie->apps_reset); } -static void imx6_pcie_ltssm_disable(struct device *dev) +static void imx_pcie_ltssm_disable(struct device *dev) { - struct imx6_pcie *imx6_pcie = dev_get_drvdata(dev); - const struct imx6_pcie_drvdata *drvdata = imx6_pcie->drvdata; + struct imx_pcie *imx_pcie = dev_get_drvdata(dev); + const struct imx_pcie_drvdata *drvdata = imx_pcie->drvdata; + phy_set_speed(imx_pcie->phy, 0); if (drvdata->ltssm_mask) - regmap_update_bits(imx6_pcie->iomuxc_gpr, drvdata->ltssm_off, + regmap_update_bits(imx_pcie->iomuxc_gpr, drvdata->ltssm_off, drvdata->ltssm_mask, 0); - reset_control_assert(imx6_pcie->apps_reset); + reset_control_assert(imx_pcie->apps_reset); } -static int imx6_pcie_start_link(struct dw_pcie *pci) +static int imx_pcie_start_link(struct dw_pcie *pci) { - struct imx6_pcie *imx6_pcie = to_imx6_pcie(pci); + struct imx_pcie *imx_pcie = to_imx_pcie(pci); struct device *dev = pci->dev; u8 offset = dw_pcie_find_capability(pci, PCI_CAP_ID_EXP); u32 tmp; @@ -841,18 +854,18 @@ static int imx6_pcie_start_link(struct dw_pcie *pci) dw_pcie_dbi_ro_wr_dis(pci); /* Start LTSSM. */ - imx6_pcie_ltssm_enable(dev); + imx_pcie_ltssm_enable(dev); ret = dw_pcie_wait_for_link(pci); if (ret) goto err_reset_phy; - if (pci->link_gen > 1) { + if (pci->max_link_speed > 1) { /* Allow faster modes after the link is up */ dw_pcie_dbi_ro_wr_en(pci); tmp = dw_pcie_readl_dbi(pci, offset + PCI_EXP_LNKCAP); tmp &= ~PCI_EXP_LNKCAP_SLS; - tmp |= pci->link_gen; + tmp |= pci->max_link_speed; dw_pcie_writel_dbi(pci, offset + PCI_EXP_LNKCAP, tmp); /* @@ -864,8 +877,8 @@ static int imx6_pcie_start_link(struct dw_pcie *pci) dw_pcie_writel_dbi(pci, PCIE_LINK_WIDTH_SPEED_CONTROL, tmp); dw_pcie_dbi_ro_wr_dis(pci); - if (imx6_pcie->drvdata->flags & - IMX6_PCIE_FLAG_IMX6_SPEED_CHANGE) { + if (imx_pcie->drvdata->flags & + IMX_PCIE_FLAG_IMX_SPEED_CHANGE) { /* * On i.MX7, DIRECT_SPEED_CHANGE behaves differently * from i.MX6 family when no link speed transition @@ -875,7 +888,7 @@ static int imx6_pcie_start_link(struct dw_pcie *pci) * failure. */ - ret = imx6_pcie_wait_for_speed_change(imx6_pcie); + ret = imx_pcie_wait_for_speed_change(imx_pcie); if (ret) { dev_err(dev, "Failed to bring link up!\n"); goto err_reset_phy; @@ -890,37 +903,37 @@ static int imx6_pcie_start_link(struct dw_pcie *pci) dev_info(dev, "Link: Only Gen1 is enabled\n"); } - imx6_pcie->link_is_up = true; + imx_pcie->link_is_up = true; tmp = dw_pcie_readw_dbi(pci, offset + PCI_EXP_LNKSTA); dev_info(dev, "Link up, Gen%i\n", tmp & PCI_EXP_LNKSTA_CLS); return 0; err_reset_phy: - imx6_pcie->link_is_up = false; + imx_pcie->link_is_up = false; dev_dbg(dev, "PHY DEBUG_R0=0x%08x DEBUG_R1=0x%08x\n", dw_pcie_readl_dbi(pci, PCIE_PORT_DEBUG0), dw_pcie_readl_dbi(pci, PCIE_PORT_DEBUG1)); - imx6_pcie_reset_phy(imx6_pcie); + imx_pcie_reset_phy(imx_pcie); return 0; } -static void imx6_pcie_stop_link(struct dw_pcie *pci) +static void imx_pcie_stop_link(struct dw_pcie *pci) { struct device *dev = pci->dev; /* Turn off PCIe LTSSM */ - imx6_pcie_ltssm_disable(dev); + imx_pcie_ltssm_disable(dev); } -static int imx6_pcie_host_init(struct dw_pcie_rp *pp) +static int imx_pcie_host_init(struct dw_pcie_rp *pp) { struct dw_pcie *pci = to_dw_pcie_from_pp(pp); struct device *dev = pci->dev; - struct imx6_pcie *imx6_pcie = to_imx6_pcie(pci); + struct imx_pcie *imx_pcie = to_imx_pcie(pci); int ret; - if (imx6_pcie->vpcie) { - ret = regulator_enable(imx6_pcie->vpcie); + if (imx_pcie->vpcie) { + ret = regulator_enable(imx_pcie->vpcie); if (ret) { dev_err(dev, "failed to enable vpcie regulator: %d\n", ret); @@ -928,83 +941,105 @@ static int imx6_pcie_host_init(struct dw_pcie_rp *pp) } } - imx6_pcie_assert_core_reset(imx6_pcie); + imx_pcie_assert_core_reset(imx_pcie); - if (imx6_pcie->drvdata->init_phy) - imx6_pcie->drvdata->init_phy(imx6_pcie); + if (imx_pcie->drvdata->init_phy) + imx_pcie->drvdata->init_phy(imx_pcie); - imx6_pcie_configure_type(imx6_pcie); + imx_pcie_configure_type(imx_pcie); - ret = imx6_pcie_clk_enable(imx6_pcie); + ret = imx_pcie_clk_enable(imx_pcie); if (ret) { dev_err(dev, "unable to enable pcie clocks: %d\n", ret); goto err_reg_disable; } - if (imx6_pcie->phy) { - ret = phy_init(imx6_pcie->phy); + if (imx_pcie->phy) { + ret = phy_init(imx_pcie->phy); if (ret) { dev_err(dev, "pcie PHY power up failed\n"); goto err_clk_disable; } - } - if (imx6_pcie->phy) { - ret = phy_power_on(imx6_pcie->phy); + ret = phy_set_mode_ext(imx_pcie->phy, PHY_MODE_PCIE, PHY_MODE_PCIE_RC); + if (ret) { + dev_err(dev, "unable to set PCIe PHY mode\n"); + goto err_phy_exit; + } + + ret = phy_power_on(imx_pcie->phy); if (ret) { dev_err(dev, "waiting for PHY ready timeout!\n"); - goto err_phy_off; + goto err_phy_exit; } } - ret = imx6_pcie_deassert_core_reset(imx6_pcie); + ret = imx_pcie_deassert_core_reset(imx_pcie); if (ret < 0) { dev_err(dev, "pcie deassert core reset failed: %d\n", ret); goto err_phy_off; } - imx6_setup_phy_mpll(imx6_pcie); + imx_setup_phy_mpll(imx_pcie); return 0; err_phy_off: - if (imx6_pcie->phy) - phy_exit(imx6_pcie->phy); + phy_power_off(imx_pcie->phy); +err_phy_exit: + phy_exit(imx_pcie->phy); err_clk_disable: - imx6_pcie_clk_disable(imx6_pcie); + imx_pcie_clk_disable(imx_pcie); err_reg_disable: - if (imx6_pcie->vpcie) - regulator_disable(imx6_pcie->vpcie); + if (imx_pcie->vpcie) + regulator_disable(imx_pcie->vpcie); return ret; } -static void imx6_pcie_host_exit(struct dw_pcie_rp *pp) +static void imx_pcie_host_exit(struct dw_pcie_rp *pp) { struct dw_pcie *pci = to_dw_pcie_from_pp(pp); - struct imx6_pcie *imx6_pcie = to_imx6_pcie(pci); + struct imx_pcie *imx_pcie = to_imx_pcie(pci); - if (imx6_pcie->phy) { - if (phy_power_off(imx6_pcie->phy)) + if (imx_pcie->phy) { + if (phy_power_off(imx_pcie->phy)) dev_err(pci->dev, "unable to power off PHY\n"); - phy_exit(imx6_pcie->phy); + phy_exit(imx_pcie->phy); } - imx6_pcie_clk_disable(imx6_pcie); + imx_pcie_clk_disable(imx_pcie); - if (imx6_pcie->vpcie) - regulator_disable(imx6_pcie->vpcie); + if (imx_pcie->vpcie) + regulator_disable(imx_pcie->vpcie); } -static const struct dw_pcie_host_ops imx6_pcie_host_ops = { - .init = imx6_pcie_host_init, - .deinit = imx6_pcie_host_exit, +static u64 imx_pcie_cpu_addr_fixup(struct dw_pcie *pcie, u64 cpu_addr) +{ + struct imx_pcie *imx_pcie = to_imx_pcie(pcie); + struct dw_pcie_rp *pp = &pcie->pp; + struct resource_entry *entry; + + if (!(imx_pcie->drvdata->flags & IMX_PCIE_FLAG_CPU_ADDR_FIXUP)) + return cpu_addr; + + entry = resource_list_first_type(&pp->bridge->windows, IORESOURCE_MEM); + if (!entry) + return cpu_addr; + + return cpu_addr - entry->offset; +} + +static const struct dw_pcie_host_ops imx_pcie_host_ops = { + .init = imx_pcie_host_init, + .deinit = imx_pcie_host_exit, }; static const struct dw_pcie_ops dw_pcie_ops = { - .start_link = imx6_pcie_start_link, - .stop_link = imx6_pcie_stop_link, + .start_link = imx_pcie_start_link, + .stop_link = imx_pcie_stop_link, + .cpu_addr_fixup = imx_pcie_cpu_addr_fixup, }; -static void imx6_pcie_ep_init(struct dw_pcie_ep *ep) +static void imx_pcie_ep_init(struct dw_pcie_ep *ep) { enum pci_barno bar; struct dw_pcie *pci = to_dw_pcie_from_ep(ep); @@ -1013,7 +1048,7 @@ static void imx6_pcie_ep_init(struct dw_pcie_ep *ep) dw_pcie_ep_reset_bar(pci, bar); } -static int imx6_pcie_ep_raise_irq(struct dw_pcie_ep *ep, u8 func_no, +static int imx_pcie_ep_raise_irq(struct dw_pcie_ep *ep, u8 func_no, unsigned int type, u16 interrupt_num) { struct dw_pcie *pci = to_dw_pcie_from_ep(ep); @@ -1060,35 +1095,35 @@ static const struct pci_epc_features imx95_pcie_epc_features = { }; static const struct pci_epc_features* -imx6_pcie_ep_get_features(struct dw_pcie_ep *ep) +imx_pcie_ep_get_features(struct dw_pcie_ep *ep) { struct dw_pcie *pci = to_dw_pcie_from_ep(ep); - struct imx6_pcie *imx6_pcie = to_imx6_pcie(pci); + struct imx_pcie *imx_pcie = to_imx_pcie(pci); - return imx6_pcie->drvdata->epc_features; + return imx_pcie->drvdata->epc_features; } static const struct dw_pcie_ep_ops pcie_ep_ops = { - .init = imx6_pcie_ep_init, - .raise_irq = imx6_pcie_ep_raise_irq, - .get_features = imx6_pcie_ep_get_features, + .init = imx_pcie_ep_init, + .raise_irq = imx_pcie_ep_raise_irq, + .get_features = imx_pcie_ep_get_features, }; -static int imx6_add_pcie_ep(struct imx6_pcie *imx6_pcie, +static int imx_add_pcie_ep(struct imx_pcie *imx_pcie, struct platform_device *pdev) { int ret; unsigned int pcie_dbi2_offset; struct dw_pcie_ep *ep; - struct dw_pcie *pci = imx6_pcie->pci; + struct dw_pcie *pci = imx_pcie->pci; struct dw_pcie_rp *pp = &pci->pp; struct device *dev = pci->dev; - imx6_pcie_host_init(pp); + imx_pcie_host_init(pp); ep = &pci->ep; ep->ops = &pcie_ep_ops; - switch (imx6_pcie->drvdata->variant) { + switch (imx_pcie->drvdata->variant) { case IMX8MQ_EP: case IMX8MM_EP: case IMX8MP_EP: @@ -1110,9 +1145,11 @@ static int imx6_add_pcie_ep(struct imx6_pcie *imx6_pcie, if (device_property_match_string(dev, "reg-names", "dbi2") >= 0) pci->dbi_base2 = NULL; - if (imx6_check_flag(imx6_pcie, IMX6_PCIE_FLAG_SUPPORT_64BIT)) + if (imx_check_flag(imx_pcie, IMX_PCIE_FLAG_SUPPORT_64BIT)) dma_set_mask_and_coherent(dev, DMA_BIT_MASK(64)); + ep->page_size = imx_pcie->drvdata->epc_features->align; + ret = dw_pcie_ep_init(ep); if (ret) { dev_err(dev, "failed to initialize endpoint\n"); @@ -1129,30 +1166,30 @@ static int imx6_add_pcie_ep(struct imx6_pcie *imx6_pcie, pci_epc_init_notify(ep->epc); /* Start LTSSM. */ - imx6_pcie_ltssm_enable(dev); + imx_pcie_ltssm_enable(dev); return 0; } -static void imx6_pcie_pm_turnoff(struct imx6_pcie *imx6_pcie) +static void imx_pcie_pm_turnoff(struct imx_pcie *imx_pcie) { - struct device *dev = imx6_pcie->pci->dev; + struct device *dev = imx_pcie->pci->dev; /* Some variants have a turnoff reset in DT */ - if (imx6_pcie->turnoff_reset) { - reset_control_assert(imx6_pcie->turnoff_reset); - reset_control_deassert(imx6_pcie->turnoff_reset); + if (imx_pcie->turnoff_reset) { + reset_control_assert(imx_pcie->turnoff_reset); + reset_control_deassert(imx_pcie->turnoff_reset); goto pm_turnoff_sleep; } /* Others poke directly at IOMUXC registers */ - switch (imx6_pcie->drvdata->variant) { + switch (imx_pcie->drvdata->variant) { case IMX6SX: case IMX6QP: - regmap_update_bits(imx6_pcie->iomuxc_gpr, IOMUXC_GPR12, + regmap_update_bits(imx_pcie->iomuxc_gpr, IOMUXC_GPR12, IMX6SX_GPR12_PCIE_PM_TURN_OFF, IMX6SX_GPR12_PCIE_PM_TURN_OFF); - regmap_update_bits(imx6_pcie->iomuxc_gpr, IOMUXC_GPR12, + regmap_update_bits(imx_pcie->iomuxc_gpr, IOMUXC_GPR12, IMX6SX_GPR12_PCIE_PM_TURN_OFF, 0); break; default: @@ -1171,73 +1208,73 @@ pm_turnoff_sleep: usleep_range(1000, 10000); } -static void imx6_pcie_msi_save_restore(struct imx6_pcie *imx6_pcie, bool save) +static void imx_pcie_msi_save_restore(struct imx_pcie *imx_pcie, bool save) { u8 offset; u16 val; - struct dw_pcie *pci = imx6_pcie->pci; + struct dw_pcie *pci = imx_pcie->pci; if (pci_msi_enabled()) { offset = dw_pcie_find_capability(pci, PCI_CAP_ID_MSI); if (save) { val = dw_pcie_readw_dbi(pci, offset + PCI_MSI_FLAGS); - imx6_pcie->msi_ctrl = val; + imx_pcie->msi_ctrl = val; } else { dw_pcie_dbi_ro_wr_en(pci); - val = imx6_pcie->msi_ctrl; + val = imx_pcie->msi_ctrl; dw_pcie_writew_dbi(pci, offset + PCI_MSI_FLAGS, val); dw_pcie_dbi_ro_wr_dis(pci); } } } -static int imx6_pcie_suspend_noirq(struct device *dev) +static int imx_pcie_suspend_noirq(struct device *dev) { - struct imx6_pcie *imx6_pcie = dev_get_drvdata(dev); - struct dw_pcie_rp *pp = &imx6_pcie->pci->pp; + struct imx_pcie *imx_pcie = dev_get_drvdata(dev); + struct dw_pcie_rp *pp = &imx_pcie->pci->pp; - if (!(imx6_pcie->drvdata->flags & IMX6_PCIE_FLAG_SUPPORTS_SUSPEND)) + if (!(imx_pcie->drvdata->flags & IMX_PCIE_FLAG_SUPPORTS_SUSPEND)) return 0; - imx6_pcie_msi_save_restore(imx6_pcie, true); - imx6_pcie_pm_turnoff(imx6_pcie); - imx6_pcie_stop_link(imx6_pcie->pci); - imx6_pcie_host_exit(pp); + imx_pcie_msi_save_restore(imx_pcie, true); + imx_pcie_pm_turnoff(imx_pcie); + imx_pcie_stop_link(imx_pcie->pci); + imx_pcie_host_exit(pp); return 0; } -static int imx6_pcie_resume_noirq(struct device *dev) +static int imx_pcie_resume_noirq(struct device *dev) { int ret; - struct imx6_pcie *imx6_pcie = dev_get_drvdata(dev); - struct dw_pcie_rp *pp = &imx6_pcie->pci->pp; + struct imx_pcie *imx_pcie = dev_get_drvdata(dev); + struct dw_pcie_rp *pp = &imx_pcie->pci->pp; - if (!(imx6_pcie->drvdata->flags & IMX6_PCIE_FLAG_SUPPORTS_SUSPEND)) + if (!(imx_pcie->drvdata->flags & IMX_PCIE_FLAG_SUPPORTS_SUSPEND)) return 0; - ret = imx6_pcie_host_init(pp); + ret = imx_pcie_host_init(pp); if (ret) return ret; - imx6_pcie_msi_save_restore(imx6_pcie, false); + imx_pcie_msi_save_restore(imx_pcie, false); dw_pcie_setup_rc(pp); - if (imx6_pcie->link_is_up) - imx6_pcie_start_link(imx6_pcie->pci); + if (imx_pcie->link_is_up) + imx_pcie_start_link(imx_pcie->pci); return 0; } -static const struct dev_pm_ops imx6_pcie_pm_ops = { - NOIRQ_SYSTEM_SLEEP_PM_OPS(imx6_pcie_suspend_noirq, - imx6_pcie_resume_noirq) +static const struct dev_pm_ops imx_pcie_pm_ops = { + NOIRQ_SYSTEM_SLEEP_PM_OPS(imx_pcie_suspend_noirq, + imx_pcie_resume_noirq) }; -static int imx6_pcie_probe(struct platform_device *pdev) +static int imx_pcie_probe(struct platform_device *pdev) { struct device *dev = &pdev->dev; struct dw_pcie *pci; - struct imx6_pcie *imx6_pcie; + struct imx_pcie *imx_pcie; struct device_node *np; struct resource *dbi_base; struct device_node *node = dev->of_node; @@ -1245,8 +1282,8 @@ static int imx6_pcie_probe(struct platform_device *pdev) u16 val; int i; - imx6_pcie = devm_kzalloc(dev, sizeof(*imx6_pcie), GFP_KERNEL); - if (!imx6_pcie) + imx_pcie = devm_kzalloc(dev, sizeof(*imx_pcie), GFP_KERNEL); + if (!imx_pcie) return -ENOMEM; pci = devm_kzalloc(dev, sizeof(*pci), GFP_KERNEL); @@ -1255,10 +1292,10 @@ static int imx6_pcie_probe(struct platform_device *pdev) pci->dev = dev; pci->ops = &dw_pcie_ops; - pci->pp.ops = &imx6_pcie_host_ops; + pci->pp.ops = &imx_pcie_host_ops; - imx6_pcie->pci = pci; - imx6_pcie->drvdata = of_device_get_match_data(dev); + imx_pcie->pci = pci; + imx_pcie->drvdata = of_device_get_match_data(dev); /* Find the PHY if one is defined, only imx7d uses it */ np = of_parse_phandle(node, "fsl,imx7d-pcie-phy", 0); @@ -1270,9 +1307,9 @@ static int imx6_pcie_probe(struct platform_device *pdev) dev_err(dev, "Unable to map PCIe PHY\n"); return ret; } - imx6_pcie->phy_base = devm_ioremap_resource(dev, &res); - if (IS_ERR(imx6_pcie->phy_base)) - return PTR_ERR(imx6_pcie->phy_base); + imx_pcie->phy_base = devm_ioremap_resource(dev, &res); + if (IS_ERR(imx_pcie->phy_base)) + return PTR_ERR(imx_pcie->phy_base); } pci->dbi_base = devm_platform_get_and_ioremap_resource(pdev, 0, &dbi_base); @@ -1280,72 +1317,72 @@ static int imx6_pcie_probe(struct platform_device *pdev) return PTR_ERR(pci->dbi_base); /* Fetch GPIOs */ - imx6_pcie->reset_gpiod = devm_gpiod_get_optional(dev, "reset", GPIOD_OUT_HIGH); - if (IS_ERR(imx6_pcie->reset_gpiod)) - return dev_err_probe(dev, PTR_ERR(imx6_pcie->reset_gpiod), + imx_pcie->reset_gpiod = devm_gpiod_get_optional(dev, "reset", GPIOD_OUT_HIGH); + if (IS_ERR(imx_pcie->reset_gpiod)) + return dev_err_probe(dev, PTR_ERR(imx_pcie->reset_gpiod), "unable to get reset gpio\n"); - gpiod_set_consumer_name(imx6_pcie->reset_gpiod, "PCIe reset"); + gpiod_set_consumer_name(imx_pcie->reset_gpiod, "PCIe reset"); - if (imx6_pcie->drvdata->clks_cnt >= IMX6_PCIE_MAX_CLKS) + if (imx_pcie->drvdata->clks_cnt >= IMX_PCIE_MAX_CLKS) return dev_err_probe(dev, -ENOMEM, "clks_cnt is too big\n"); - for (i = 0; i < imx6_pcie->drvdata->clks_cnt; i++) - imx6_pcie->clks[i].id = imx6_pcie->drvdata->clk_names[i]; + for (i = 0; i < imx_pcie->drvdata->clks_cnt; i++) + imx_pcie->clks[i].id = imx_pcie->drvdata->clk_names[i]; /* Fetch clocks */ - ret = devm_clk_bulk_get(dev, imx6_pcie->drvdata->clks_cnt, imx6_pcie->clks); + ret = devm_clk_bulk_get(dev, imx_pcie->drvdata->clks_cnt, imx_pcie->clks); if (ret) return ret; - if (imx6_check_flag(imx6_pcie, IMX6_PCIE_FLAG_HAS_PHYDRV)) { - imx6_pcie->phy = devm_phy_get(dev, "pcie-phy"); - if (IS_ERR(imx6_pcie->phy)) - return dev_err_probe(dev, PTR_ERR(imx6_pcie->phy), + if (imx_check_flag(imx_pcie, IMX_PCIE_FLAG_HAS_PHYDRV)) { + imx_pcie->phy = devm_phy_get(dev, "pcie-phy"); + if (IS_ERR(imx_pcie->phy)) + return dev_err_probe(dev, PTR_ERR(imx_pcie->phy), "failed to get pcie phy\n"); } - if (imx6_check_flag(imx6_pcie, IMX6_PCIE_FLAG_HAS_APP_RESET)) { - imx6_pcie->apps_reset = devm_reset_control_get_exclusive(dev, "apps"); - if (IS_ERR(imx6_pcie->apps_reset)) - return dev_err_probe(dev, PTR_ERR(imx6_pcie->apps_reset), + if (imx_check_flag(imx_pcie, IMX_PCIE_FLAG_HAS_APP_RESET)) { + imx_pcie->apps_reset = devm_reset_control_get_exclusive(dev, "apps"); + if (IS_ERR(imx_pcie->apps_reset)) + return dev_err_probe(dev, PTR_ERR(imx_pcie->apps_reset), "failed to get pcie apps reset control\n"); } - if (imx6_check_flag(imx6_pcie, IMX6_PCIE_FLAG_HAS_PHY_RESET)) { - imx6_pcie->pciephy_reset = devm_reset_control_get_exclusive(dev, "pciephy"); - if (IS_ERR(imx6_pcie->pciephy_reset)) - return dev_err_probe(dev, PTR_ERR(imx6_pcie->pciephy_reset), + if (imx_check_flag(imx_pcie, IMX_PCIE_FLAG_HAS_PHY_RESET)) { + imx_pcie->pciephy_reset = devm_reset_control_get_exclusive(dev, "pciephy"); + if (IS_ERR(imx_pcie->pciephy_reset)) + return dev_err_probe(dev, PTR_ERR(imx_pcie->pciephy_reset), "Failed to get PCIEPHY reset control\n"); } - switch (imx6_pcie->drvdata->variant) { + switch (imx_pcie->drvdata->variant) { case IMX8MQ: case IMX8MQ_EP: case IMX7D: if (dbi_base->start == IMX8MQ_PCIE2_BASE_ADDR) - imx6_pcie->controller_id = 1; + imx_pcie->controller_id = 1; break; default: break; } /* Grab turnoff reset */ - imx6_pcie->turnoff_reset = devm_reset_control_get_optional_exclusive(dev, "turnoff"); - if (IS_ERR(imx6_pcie->turnoff_reset)) { + imx_pcie->turnoff_reset = devm_reset_control_get_optional_exclusive(dev, "turnoff"); + if (IS_ERR(imx_pcie->turnoff_reset)) { dev_err(dev, "Failed to get TURNOFF reset control\n"); - return PTR_ERR(imx6_pcie->turnoff_reset); + return PTR_ERR(imx_pcie->turnoff_reset); } - if (imx6_pcie->drvdata->gpr) { + if (imx_pcie->drvdata->gpr) { /* Grab GPR config register range */ - imx6_pcie->iomuxc_gpr = - syscon_regmap_lookup_by_compatible(imx6_pcie->drvdata->gpr); - if (IS_ERR(imx6_pcie->iomuxc_gpr)) - return dev_err_probe(dev, PTR_ERR(imx6_pcie->iomuxc_gpr), + imx_pcie->iomuxc_gpr = + syscon_regmap_lookup_by_compatible(imx_pcie->drvdata->gpr); + if (IS_ERR(imx_pcie->iomuxc_gpr)) + return dev_err_probe(dev, PTR_ERR(imx_pcie->iomuxc_gpr), "unable to find iomuxc registers\n"); } - if (imx6_check_flag(imx6_pcie, IMX6_PCIE_FLAG_HAS_SERDES)) { + if (imx_check_flag(imx_pcie, IMX_PCIE_FLAG_HAS_SERDES)) { void __iomem *off = devm_platform_ioremap_resource_byname(pdev, "app"); if (IS_ERR(off)) @@ -1358,59 +1395,59 @@ static int imx6_pcie_probe(struct platform_device *pdev) .reg_stride = 4, }; - imx6_pcie->iomuxc_gpr = devm_regmap_init_mmio(dev, off, ®map_config); - if (IS_ERR(imx6_pcie->iomuxc_gpr)) - return dev_err_probe(dev, PTR_ERR(imx6_pcie->iomuxc_gpr), + imx_pcie->iomuxc_gpr = devm_regmap_init_mmio(dev, off, ®map_config); + if (IS_ERR(imx_pcie->iomuxc_gpr)) + return dev_err_probe(dev, PTR_ERR(imx_pcie->iomuxc_gpr), "unable to find iomuxc registers\n"); } /* Grab PCIe PHY Tx Settings */ if (of_property_read_u32(node, "fsl,tx-deemph-gen1", - &imx6_pcie->tx_deemph_gen1)) - imx6_pcie->tx_deemph_gen1 = 0; + &imx_pcie->tx_deemph_gen1)) + imx_pcie->tx_deemph_gen1 = 0; if (of_property_read_u32(node, "fsl,tx-deemph-gen2-3p5db", - &imx6_pcie->tx_deemph_gen2_3p5db)) - imx6_pcie->tx_deemph_gen2_3p5db = 0; + &imx_pcie->tx_deemph_gen2_3p5db)) + imx_pcie->tx_deemph_gen2_3p5db = 0; if (of_property_read_u32(node, "fsl,tx-deemph-gen2-6db", - &imx6_pcie->tx_deemph_gen2_6db)) - imx6_pcie->tx_deemph_gen2_6db = 20; + &imx_pcie->tx_deemph_gen2_6db)) + imx_pcie->tx_deemph_gen2_6db = 20; if (of_property_read_u32(node, "fsl,tx-swing-full", - &imx6_pcie->tx_swing_full)) - imx6_pcie->tx_swing_full = 127; + &imx_pcie->tx_swing_full)) + imx_pcie->tx_swing_full = 127; if (of_property_read_u32(node, "fsl,tx-swing-low", - &imx6_pcie->tx_swing_low)) - imx6_pcie->tx_swing_low = 127; + &imx_pcie->tx_swing_low)) + imx_pcie->tx_swing_low = 127; /* Limit link speed */ - pci->link_gen = 1; - of_property_read_u32(node, "fsl,max-link-speed", &pci->link_gen); + pci->max_link_speed = 1; + of_property_read_u32(node, "fsl,max-link-speed", &pci->max_link_speed); - imx6_pcie->vpcie = devm_regulator_get_optional(&pdev->dev, "vpcie"); - if (IS_ERR(imx6_pcie->vpcie)) { - if (PTR_ERR(imx6_pcie->vpcie) != -ENODEV) - return PTR_ERR(imx6_pcie->vpcie); - imx6_pcie->vpcie = NULL; + imx_pcie->vpcie = devm_regulator_get_optional(&pdev->dev, "vpcie"); + if (IS_ERR(imx_pcie->vpcie)) { + if (PTR_ERR(imx_pcie->vpcie) != -ENODEV) + return PTR_ERR(imx_pcie->vpcie); + imx_pcie->vpcie = NULL; } - imx6_pcie->vph = devm_regulator_get_optional(&pdev->dev, "vph"); - if (IS_ERR(imx6_pcie->vph)) { - if (PTR_ERR(imx6_pcie->vph) != -ENODEV) - return PTR_ERR(imx6_pcie->vph); - imx6_pcie->vph = NULL; + imx_pcie->vph = devm_regulator_get_optional(&pdev->dev, "vph"); + if (IS_ERR(imx_pcie->vph)) { + if (PTR_ERR(imx_pcie->vph) != -ENODEV) + return PTR_ERR(imx_pcie->vph); + imx_pcie->vph = NULL; } - platform_set_drvdata(pdev, imx6_pcie); + platform_set_drvdata(pdev, imx_pcie); - ret = imx6_pcie_attach_pd(dev); + ret = imx_pcie_attach_pd(dev); if (ret) return ret; - if (imx6_pcie->drvdata->mode == DW_PCIE_EP_TYPE) { - ret = imx6_add_pcie_ep(imx6_pcie, pdev); + if (imx_pcie->drvdata->mode == DW_PCIE_EP_TYPE) { + ret = imx_add_pcie_ep(imx_pcie, pdev); if (ret < 0) return ret; } else { @@ -1430,24 +1467,25 @@ static int imx6_pcie_probe(struct platform_device *pdev) return 0; } -static void imx6_pcie_shutdown(struct platform_device *pdev) +static void imx_pcie_shutdown(struct platform_device *pdev) { - struct imx6_pcie *imx6_pcie = platform_get_drvdata(pdev); + struct imx_pcie *imx_pcie = platform_get_drvdata(pdev); /* bring down link, so bootloader gets clean state in case of reboot */ - imx6_pcie_assert_core_reset(imx6_pcie); + imx_pcie_assert_core_reset(imx_pcie); } static const char * const imx6q_clks[] = {"pcie_bus", "pcie", "pcie_phy"}; static const char * const imx8mm_clks[] = {"pcie_bus", "pcie", "pcie_aux"}; static const char * const imx8mq_clks[] = {"pcie_bus", "pcie", "pcie_phy", "pcie_aux"}; static const char * const imx6sx_clks[] = {"pcie_bus", "pcie", "pcie_phy", "pcie_inbound_axi"}; +static const char * const imx8q_clks[] = {"mstr", "slv", "dbi"}; -static const struct imx6_pcie_drvdata drvdata[] = { +static const struct imx_pcie_drvdata drvdata[] = { [IMX6Q] = { .variant = IMX6Q, - .flags = IMX6_PCIE_FLAG_IMX6_PHY | - IMX6_PCIE_FLAG_IMX6_SPEED_CHANGE, + .flags = IMX_PCIE_FLAG_IMX_PHY | + IMX_PCIE_FLAG_IMX_SPEED_CHANGE, .dbi_length = 0x200, .gpr = "fsl,imx6q-iomuxc-gpr", .clk_names = imx6q_clks, @@ -1456,13 +1494,15 @@ static const struct imx6_pcie_drvdata drvdata[] = { .ltssm_mask = IMX6Q_GPR12_PCIE_CTL_2, .mode_off[0] = IOMUXC_GPR12, .mode_mask[0] = IMX6Q_GPR12_DEVICE_TYPE, - .init_phy = imx6_pcie_init_phy, + .init_phy = imx_pcie_init_phy, + .enable_ref_clk = imx6q_pcie_enable_ref_clk, + .core_reset = imx6q_pcie_core_reset, }, [IMX6SX] = { .variant = IMX6SX, - .flags = IMX6_PCIE_FLAG_IMX6_PHY | - IMX6_PCIE_FLAG_IMX6_SPEED_CHANGE | - IMX6_PCIE_FLAG_SUPPORTS_SUSPEND, + .flags = IMX_PCIE_FLAG_IMX_PHY | + IMX_PCIE_FLAG_IMX_SPEED_CHANGE | + IMX_PCIE_FLAG_SUPPORTS_SUSPEND, .gpr = "fsl,imx6q-iomuxc-gpr", .clk_names = imx6sx_clks, .clks_cnt = ARRAY_SIZE(imx6sx_clks), @@ -1471,12 +1511,14 @@ static const struct imx6_pcie_drvdata drvdata[] = { .mode_off[0] = IOMUXC_GPR12, .mode_mask[0] = IMX6Q_GPR12_DEVICE_TYPE, .init_phy = imx6sx_pcie_init_phy, + .enable_ref_clk = imx6sx_pcie_enable_ref_clk, + .core_reset = imx6sx_pcie_core_reset, }, [IMX6QP] = { .variant = IMX6QP, - .flags = IMX6_PCIE_FLAG_IMX6_PHY | - IMX6_PCIE_FLAG_IMX6_SPEED_CHANGE | - IMX6_PCIE_FLAG_SUPPORTS_SUSPEND, + .flags = IMX_PCIE_FLAG_IMX_PHY | + IMX_PCIE_FLAG_IMX_SPEED_CHANGE | + IMX_PCIE_FLAG_SUPPORTS_SUSPEND, .dbi_length = 0x200, .gpr = "fsl,imx6q-iomuxc-gpr", .clk_names = imx6q_clks, @@ -1485,24 +1527,28 @@ static const struct imx6_pcie_drvdata drvdata[] = { .ltssm_mask = IMX6Q_GPR12_PCIE_CTL_2, .mode_off[0] = IOMUXC_GPR12, .mode_mask[0] = IMX6Q_GPR12_DEVICE_TYPE, - .init_phy = imx6_pcie_init_phy, + .init_phy = imx_pcie_init_phy, + .enable_ref_clk = imx6q_pcie_enable_ref_clk, + .core_reset = imx6qp_pcie_core_reset, }, [IMX7D] = { .variant = IMX7D, - .flags = IMX6_PCIE_FLAG_SUPPORTS_SUSPEND | - IMX6_PCIE_FLAG_HAS_APP_RESET | - IMX6_PCIE_FLAG_HAS_PHY_RESET, + .flags = IMX_PCIE_FLAG_SUPPORTS_SUSPEND | + IMX_PCIE_FLAG_HAS_APP_RESET | + IMX_PCIE_FLAG_HAS_PHY_RESET, .gpr = "fsl,imx7d-iomuxc-gpr", .clk_names = imx6q_clks, .clks_cnt = ARRAY_SIZE(imx6q_clks), .mode_off[0] = IOMUXC_GPR12, .mode_mask[0] = IMX6Q_GPR12_DEVICE_TYPE, .init_phy = imx7d_pcie_init_phy, + .enable_ref_clk = imx7d_pcie_enable_ref_clk, + .core_reset = imx7d_pcie_core_reset, }, [IMX8MQ] = { .variant = IMX8MQ, - .flags = IMX6_PCIE_FLAG_HAS_APP_RESET | - IMX6_PCIE_FLAG_HAS_PHY_RESET, + .flags = IMX_PCIE_FLAG_HAS_APP_RESET | + IMX_PCIE_FLAG_HAS_PHY_RESET, .gpr = "fsl,imx8mq-iomuxc-gpr", .clk_names = imx8mq_clks, .clks_cnt = ARRAY_SIZE(imx8mq_clks), @@ -1511,32 +1557,42 @@ static const struct imx6_pcie_drvdata drvdata[] = { .mode_off[1] = IOMUXC_GPR12, .mode_mask[1] = IMX8MQ_GPR12_PCIE2_CTRL_DEVICE_TYPE, .init_phy = imx8mq_pcie_init_phy, + .enable_ref_clk = imx8mm_pcie_enable_ref_clk, }, [IMX8MM] = { .variant = IMX8MM, - .flags = IMX6_PCIE_FLAG_SUPPORTS_SUSPEND | - IMX6_PCIE_FLAG_HAS_PHYDRV | - IMX6_PCIE_FLAG_HAS_APP_RESET, + .flags = IMX_PCIE_FLAG_SUPPORTS_SUSPEND | + IMX_PCIE_FLAG_HAS_PHYDRV | + IMX_PCIE_FLAG_HAS_APP_RESET, .gpr = "fsl,imx8mm-iomuxc-gpr", .clk_names = imx8mm_clks, .clks_cnt = ARRAY_SIZE(imx8mm_clks), .mode_off[0] = IOMUXC_GPR12, .mode_mask[0] = IMX6Q_GPR12_DEVICE_TYPE, + .enable_ref_clk = imx8mm_pcie_enable_ref_clk, }, [IMX8MP] = { .variant = IMX8MP, - .flags = IMX6_PCIE_FLAG_SUPPORTS_SUSPEND | - IMX6_PCIE_FLAG_HAS_PHYDRV | - IMX6_PCIE_FLAG_HAS_APP_RESET, + .flags = IMX_PCIE_FLAG_SUPPORTS_SUSPEND | + IMX_PCIE_FLAG_HAS_PHYDRV | + IMX_PCIE_FLAG_HAS_APP_RESET, .gpr = "fsl,imx8mp-iomuxc-gpr", .clk_names = imx8mm_clks, .clks_cnt = ARRAY_SIZE(imx8mm_clks), .mode_off[0] = IOMUXC_GPR12, .mode_mask[0] = IMX6Q_GPR12_DEVICE_TYPE, + .enable_ref_clk = imx8mm_pcie_enable_ref_clk, + }, + [IMX8Q] = { + .variant = IMX8Q, + .flags = IMX_PCIE_FLAG_HAS_PHYDRV | + IMX_PCIE_FLAG_CPU_ADDR_FIXUP, + .clk_names = imx8q_clks, + .clks_cnt = ARRAY_SIZE(imx8q_clks), }, [IMX95] = { .variant = IMX95, - .flags = IMX6_PCIE_FLAG_HAS_SERDES, + .flags = IMX_PCIE_FLAG_HAS_SERDES, .clk_names = imx8mq_clks, .clks_cnt = ARRAY_SIZE(imx8mq_clks), .ltssm_off = IMX95_PE0_GEN_CTRL_3, @@ -1547,8 +1603,8 @@ static const struct imx6_pcie_drvdata drvdata[] = { }, [IMX8MQ_EP] = { .variant = IMX8MQ_EP, - .flags = IMX6_PCIE_FLAG_HAS_APP_RESET | - IMX6_PCIE_FLAG_HAS_PHY_RESET, + .flags = IMX_PCIE_FLAG_HAS_APP_RESET | + IMX_PCIE_FLAG_HAS_PHY_RESET, .mode = DW_PCIE_EP_TYPE, .gpr = "fsl,imx8mq-iomuxc-gpr", .clk_names = imx8mq_clks, @@ -1559,10 +1615,12 @@ static const struct imx6_pcie_drvdata drvdata[] = { .mode_mask[1] = IMX8MQ_GPR12_PCIE2_CTRL_DEVICE_TYPE, .epc_features = &imx8m_pcie_epc_features, .init_phy = imx8mq_pcie_init_phy, + .enable_ref_clk = imx8mm_pcie_enable_ref_clk, }, [IMX8MM_EP] = { .variant = IMX8MM_EP, - .flags = IMX6_PCIE_FLAG_HAS_PHYDRV, + .flags = IMX_PCIE_FLAG_HAS_APP_RESET | + IMX_PCIE_FLAG_HAS_PHYDRV, .mode = DW_PCIE_EP_TYPE, .gpr = "fsl,imx8mm-iomuxc-gpr", .clk_names = imx8mm_clks, @@ -1570,10 +1628,12 @@ static const struct imx6_pcie_drvdata drvdata[] = { .mode_off[0] = IOMUXC_GPR12, .mode_mask[0] = IMX6Q_GPR12_DEVICE_TYPE, .epc_features = &imx8m_pcie_epc_features, + .enable_ref_clk = imx8mm_pcie_enable_ref_clk, }, [IMX8MP_EP] = { .variant = IMX8MP_EP, - .flags = IMX6_PCIE_FLAG_HAS_PHYDRV, + .flags = IMX_PCIE_FLAG_HAS_APP_RESET | + IMX_PCIE_FLAG_HAS_PHYDRV, .mode = DW_PCIE_EP_TYPE, .gpr = "fsl,imx8mp-iomuxc-gpr", .clk_names = imx8mm_clks, @@ -1581,11 +1641,12 @@ static const struct imx6_pcie_drvdata drvdata[] = { .mode_off[0] = IOMUXC_GPR12, .mode_mask[0] = IMX6Q_GPR12_DEVICE_TYPE, .epc_features = &imx8m_pcie_epc_features, + .enable_ref_clk = imx8mm_pcie_enable_ref_clk, }, [IMX95_EP] = { .variant = IMX95_EP, - .flags = IMX6_PCIE_FLAG_HAS_SERDES | - IMX6_PCIE_FLAG_SUPPORT_64BIT, + .flags = IMX_PCIE_FLAG_HAS_SERDES | + IMX_PCIE_FLAG_SUPPORT_64BIT, .clk_names = imx8mq_clks, .clks_cnt = ARRAY_SIZE(imx8mq_clks), .ltssm_off = IMX95_PE0_GEN_CTRL_3, @@ -1598,7 +1659,7 @@ static const struct imx6_pcie_drvdata drvdata[] = { }, }; -static const struct of_device_id imx6_pcie_of_match[] = { +static const struct of_device_id imx_pcie_of_match[] = { { .compatible = "fsl,imx6q-pcie", .data = &drvdata[IMX6Q], }, { .compatible = "fsl,imx6sx-pcie", .data = &drvdata[IMX6SX], }, { .compatible = "fsl,imx6qp-pcie", .data = &drvdata[IMX6QP], }, @@ -1606,6 +1667,7 @@ static const struct of_device_id imx6_pcie_of_match[] = { { .compatible = "fsl,imx8mq-pcie", .data = &drvdata[IMX8MQ], }, { .compatible = "fsl,imx8mm-pcie", .data = &drvdata[IMX8MM], }, { .compatible = "fsl,imx8mp-pcie", .data = &drvdata[IMX8MP], }, + { .compatible = "fsl,imx8q-pcie", .data = &drvdata[IMX8Q], }, { .compatible = "fsl,imx95-pcie", .data = &drvdata[IMX95], }, { .compatible = "fsl,imx8mq-pcie-ep", .data = &drvdata[IMX8MQ_EP], }, { .compatible = "fsl,imx8mm-pcie-ep", .data = &drvdata[IMX8MM_EP], }, @@ -1614,19 +1676,19 @@ static const struct of_device_id imx6_pcie_of_match[] = { {}, }; -static struct platform_driver imx6_pcie_driver = { +static struct platform_driver imx_pcie_driver = { .driver = { .name = "imx6q-pcie", - .of_match_table = imx6_pcie_of_match, + .of_match_table = imx_pcie_of_match, .suppress_bind_attrs = true, - .pm = &imx6_pcie_pm_ops, + .pm = &imx_pcie_pm_ops, .probe_type = PROBE_PREFER_ASYNCHRONOUS, }, - .probe = imx6_pcie_probe, - .shutdown = imx6_pcie_shutdown, + .probe = imx_pcie_probe, + .shutdown = imx_pcie_shutdown, }; -static void imx6_pcie_quirk(struct pci_dev *dev) +static void imx_pcie_quirk(struct pci_dev *dev) { struct pci_bus *bus = dev->bus; struct dw_pcie_rp *pp = bus->sysdata; @@ -1636,33 +1698,33 @@ static void imx6_pcie_quirk(struct pci_dev *dev) return; /* Make sure we only quirk devices associated with this driver */ - if (bus->dev.parent->parent->driver != &imx6_pcie_driver.driver) + if (bus->dev.parent->parent->driver != &imx_pcie_driver.driver) return; if (pci_is_root_bus(bus)) { struct dw_pcie *pci = to_dw_pcie_from_pp(pp); - struct imx6_pcie *imx6_pcie = to_imx6_pcie(pci); + struct imx_pcie *imx_pcie = to_imx_pcie(pci); /* * Limit config length to avoid the kernel reading beyond * the register set and causing an abort on i.MX 6Quad */ - if (imx6_pcie->drvdata->dbi_length) { - dev->cfg_size = imx6_pcie->drvdata->dbi_length; + if (imx_pcie->drvdata->dbi_length) { + dev->cfg_size = imx_pcie->drvdata->dbi_length; dev_info(&dev->dev, "Limiting cfg_size to %d\n", dev->cfg_size); } } } DECLARE_PCI_FIXUP_CLASS_HEADER(PCI_VENDOR_ID_SYNOPSYS, 0xabcd, - PCI_CLASS_BRIDGE_PCI, 8, imx6_pcie_quirk); + PCI_CLASS_BRIDGE_PCI, 8, imx_pcie_quirk); -static int __init imx6_pcie_init(void) +static int __init imx_pcie_init(void) { #ifdef CONFIG_ARM struct device_node *np; - np = of_find_matching_node(NULL, imx6_pcie_of_match); + np = of_find_matching_node(NULL, imx_pcie_of_match); if (!np) return -ENODEV; of_node_put(np); @@ -1678,6 +1740,6 @@ static int __init imx6_pcie_init(void) "external abort on non-linefetch"); #endif - return platform_driver_register(&imx6_pcie_driver); + return platform_driver_register(&imx_pcie_driver); } -device_initcall(imx6_pcie_init); +device_initcall(imx_pcie_init); diff --git a/drivers/pci/controller/dwc/pci-keystone.c b/drivers/pci/controller/dwc/pci-keystone.c index 52c6420ae200..2219b1a866fa 100644 --- a/drivers/pci/controller/dwc/pci-keystone.c +++ b/drivers/pci/controller/dwc/pci-keystone.c @@ -189,12 +189,6 @@ static void ks_pcie_compose_msi_msg(struct irq_data *data, struct msi_msg *msg) (int)data->hwirq, msg->address_hi, msg->address_lo); } -static int ks_pcie_msi_set_affinity(struct irq_data *irq_data, - const struct cpumask *mask, bool force) -{ - return -EINVAL; -} - static void ks_pcie_msi_mask(struct irq_data *data) { struct dw_pcie_rp *pp = irq_data_get_irq_chip_data(data); @@ -247,7 +241,6 @@ static struct irq_chip ks_pcie_msi_irq_chip = { .name = "KEYSTONE-PCI-MSI", .irq_ack = ks_pcie_msi_irq_ack, .irq_compose_msi_msg = ks_pcie_compose_msi_msg, - .irq_set_affinity = ks_pcie_msi_set_affinity, .irq_mask = ks_pcie_msi_mask, .irq_unmask = ks_pcie_msi_unmask, }; @@ -577,7 +570,7 @@ static void ks_pcie_quirk(struct pci_dev *dev) */ if (pci_match_id(am6_pci_devids, bridge)) { bridge_dev = pci_get_host_bridge_device(dev); - if (!bridge_dev && !bridge_dev->parent) + if (!bridge_dev || !bridge_dev->parent) return; ks_pcie = dev_get_drvdata(bridge_dev->parent); diff --git a/drivers/pci/controller/dwc/pcie-designware-host.c b/drivers/pci/controller/dwc/pcie-designware-host.c index 8a1ff5c913d8..ffb52965e477 100644 --- a/drivers/pci/controller/dwc/pcie-designware-host.c +++ b/drivers/pci/controller/dwc/pcie-designware-host.c @@ -48,8 +48,9 @@ static struct irq_chip dw_pcie_msi_irq_chip = { }; static struct msi_domain_info dw_pcie_msi_domain_info = { - .flags = (MSI_FLAG_USE_DEF_DOM_OPS | MSI_FLAG_USE_DEF_CHIP_OPS | - MSI_FLAG_PCI_MSIX | MSI_FLAG_MULTI_PCI_MSI), + .flags = MSI_FLAG_USE_DEF_DOM_OPS | MSI_FLAG_USE_DEF_CHIP_OPS | + MSI_FLAG_NO_AFFINITY | MSI_FLAG_PCI_MSIX | + MSI_FLAG_MULTI_PCI_MSI, .chip = &dw_pcie_msi_irq_chip, }; @@ -117,12 +118,6 @@ static void dw_pci_setup_msi_msg(struct irq_data *d, struct msi_msg *msg) (int)d->hwirq, msg->address_hi, msg->address_lo); } -static int dw_pci_msi_set_affinity(struct irq_data *d, - const struct cpumask *mask, bool force) -{ - return -EINVAL; -} - static void dw_pci_bottom_mask(struct irq_data *d) { struct dw_pcie_rp *pp = irq_data_get_irq_chip_data(d); @@ -178,7 +173,6 @@ static struct irq_chip dw_pci_msi_bottom_irq_chip = { .name = "DWPCI-MSI", .irq_ack = dw_pci_bottom_ack, .irq_compose_msi_msg = dw_pci_setup_msi_msg, - .irq_set_affinity = dw_pci_msi_set_affinity, .irq_mask = dw_pci_bottom_mask, .irq_unmask = dw_pci_bottom_unmask, }; diff --git a/drivers/pci/controller/dwc/pcie-designware.c b/drivers/pci/controller/dwc/pcie-designware.c index 1b5aba1f0c92..6d6cbc8b5b2c 100644 --- a/drivers/pci/controller/dwc/pcie-designware.c +++ b/drivers/pci/controller/dwc/pcie-designware.c @@ -112,6 +112,7 @@ int dw_pcie_get_resources(struct dw_pcie *pci) pci->dbi_base = devm_pci_remap_cfg_resource(pci->dev, res); if (IS_ERR(pci->dbi_base)) return PTR_ERR(pci->dbi_base); + pci->dbi_phys_addr = res->start; } /* DBI2 is mainly useful for the endpoint controller */ @@ -134,6 +135,7 @@ int dw_pcie_get_resources(struct dw_pcie *pci) pci->atu_base = devm_ioremap_resource(pci->dev, res); if (IS_ERR(pci->atu_base)) return PTR_ERR(pci->atu_base); + pci->atu_phys_addr = res->start; } else { pci->atu_base = pci->dbi_base + DEFAULT_DBI_ATU_OFFSET; } @@ -166,8 +168,8 @@ int dw_pcie_get_resources(struct dw_pcie *pci) return ret; } - if (pci->link_gen < 1) - pci->link_gen = of_pci_get_max_link_speed(np); + if (pci->max_link_speed < 1) + pci->max_link_speed = of_pci_get_max_link_speed(np); of_property_read_u32(np, "num-lanes", &pci->num_lanes); @@ -687,16 +689,27 @@ void dw_pcie_upconfig_setup(struct dw_pcie *pci) } EXPORT_SYMBOL_GPL(dw_pcie_upconfig_setup); -static void dw_pcie_link_set_max_speed(struct dw_pcie *pci, u32 link_gen) +static void dw_pcie_link_set_max_speed(struct dw_pcie *pci) { u32 cap, ctrl2, link_speed; u8 offset = dw_pcie_find_capability(pci, PCI_CAP_ID_EXP); cap = dw_pcie_readl_dbi(pci, offset + PCI_EXP_LNKCAP); + + /* + * Even if the platform doesn't want to limit the maximum link speed, + * just cache the hardware default value so that the vendor drivers can + * use it to do any link specific configuration. + */ + if (pci->max_link_speed < 1) { + pci->max_link_speed = FIELD_GET(PCI_EXP_LNKCAP_SLS, cap); + return; + } + ctrl2 = dw_pcie_readl_dbi(pci, offset + PCI_EXP_LNKCTL2); ctrl2 &= ~PCI_EXP_LNKCTL2_TLS; - switch (pcie_link_speed[link_gen]) { + switch (pcie_link_speed[pci->max_link_speed]) { case PCIE_SPEED_2_5GT: link_speed = PCI_EXP_LNKCTL2_TLS_2_5GT; break; @@ -1058,8 +1071,7 @@ void dw_pcie_setup(struct dw_pcie *pci) { u32 val; - if (pci->link_gen > 0) - dw_pcie_link_set_max_speed(pci, pci->link_gen); + dw_pcie_link_set_max_speed(pci); /* Configure Gen1 N_FTS */ if (pci->n_fts[0]) { diff --git a/drivers/pci/controller/dwc/pcie-designware.h b/drivers/pci/controller/dwc/pcie-designware.h index 53c4c8f399c8..347ab74ac35a 100644 --- a/drivers/pci/controller/dwc/pcie-designware.h +++ b/drivers/pci/controller/dwc/pcie-designware.h @@ -125,6 +125,19 @@ #define GEN3_RELATED_OFF_GEN3_EQ_DISABLE BIT(16) #define GEN3_RELATED_OFF_RATE_SHADOW_SEL_SHIFT 24 #define GEN3_RELATED_OFF_RATE_SHADOW_SEL_MASK GENMASK(25, 24) +#define GEN3_RELATED_OFF_RATE_SHADOW_SEL_16_0GT 0x1 + +#define GEN3_EQ_CONTROL_OFF 0x8A8 +#define GEN3_EQ_CONTROL_OFF_FB_MODE GENMASK(3, 0) +#define GEN3_EQ_CONTROL_OFF_PHASE23_EXIT_MODE BIT(4) +#define GEN3_EQ_CONTROL_OFF_PSET_REQ_VEC GENMASK(23, 8) +#define GEN3_EQ_CONTROL_OFF_FOM_INC_INITIAL_EVAL BIT(24) + +#define GEN3_EQ_FB_MODE_DIR_CHANGE_OFF 0x8AC +#define GEN3_EQ_FMDC_T_MIN_PHASE23 GENMASK(4, 0) +#define GEN3_EQ_FMDC_N_EVALS GENMASK(9, 5) +#define GEN3_EQ_FMDC_MAX_PRE_CUSROR_DELTA GENMASK(13, 10) +#define GEN3_EQ_FMDC_MAX_POST_CUSROR_DELTA GENMASK(17, 14) #define PCIE_PORT_MULTI_LANE_CTRL 0x8C0 #define PORT_MLTI_UPCFG_SUPPORT BIT(7) @@ -197,6 +210,24 @@ #define PCIE_PL_CHK_REG_ERR_ADDR 0xB28 +/* + * 16.0 GT/s (Gen 4) lane margining register definitions + */ +#define GEN4_LANE_MARGINING_1_OFF 0xB80 +#define MARGINING_MAX_VOLTAGE_OFFSET GENMASK(29, 24) +#define MARGINING_NUM_VOLTAGE_STEPS GENMASK(22, 16) +#define MARGINING_MAX_TIMING_OFFSET GENMASK(13, 8) +#define MARGINING_NUM_TIMING_STEPS GENMASK(5, 0) + +#define GEN4_LANE_MARGINING_2_OFF 0xB84 +#define MARGINING_IND_ERROR_SAMPLER BIT(28) +#define MARGINING_SAMPLE_REPORTING_METHOD BIT(27) +#define MARGINING_IND_LEFT_RIGHT_TIMING BIT(26) +#define MARGINING_IND_UP_DOWN_VOLTAGE BIT(25) +#define MARGINING_VOLTAGE_SUPPORTED BIT(24) +#define MARGINING_MAXLANES GENMASK(20, 16) +#define MARGINING_SAMPLE_RATE_TIMING GENMASK(13, 8) +#define MARGINING_SAMPLE_RATE_VOLTAGE GENMASK(5, 0) /* * iATU Unroll-specific register definitions * From 4.80 core version the address translation will be made by unroll @@ -407,8 +438,10 @@ struct dw_pcie_ops { struct dw_pcie { struct device *dev; void __iomem *dbi_base; + resource_size_t dbi_phys_addr; void __iomem *dbi_base2; void __iomem *atu_base; + resource_size_t atu_phys_addr; size_t atu_size; u32 num_ib_windows; u32 num_ob_windows; @@ -421,7 +454,7 @@ struct dw_pcie { u32 type; unsigned long caps; int num_lanes; - int link_gen; + int max_link_speed; u8 n_fts[2]; struct dw_edma_chip edma; struct clk_bulk_data app_clks[DW_PCIE_NUM_APP_CLKS]; diff --git a/drivers/pci/controller/dwc/pcie-intel-gw.c b/drivers/pci/controller/dwc/pcie-intel-gw.c index acbe4f6d3291..676d2aba4fbd 100644 --- a/drivers/pci/controller/dwc/pcie-intel-gw.c +++ b/drivers/pci/controller/dwc/pcie-intel-gw.c @@ -132,7 +132,7 @@ static void intel_pcie_link_setup(struct intel_pcie *pcie) static void intel_pcie_init_n_fts(struct dw_pcie *pci) { - switch (pci->link_gen) { + switch (pci->max_link_speed) { case 3: pci->n_fts[1] = PORT_AFR_N_FTS_GEN3; break; @@ -252,7 +252,7 @@ static int intel_pcie_wait_l2(struct intel_pcie *pcie) int ret; struct dw_pcie *pci = &pcie->pci; - if (pci->link_gen < 3) + if (pci->max_link_speed < 3) return 0; /* Send PME_TURN_OFF message */ diff --git a/drivers/pci/controller/dwc/pcie-kirin.c b/drivers/pci/controller/dwc/pcie-kirin.c index 0a29136491b8..85a2c77b1835 100644 --- a/drivers/pci/controller/dwc/pcie-kirin.c +++ b/drivers/pci/controller/dwc/pcie-kirin.c @@ -420,11 +420,11 @@ static int kirin_pcie_parse_port(struct kirin_pcie *pcie, "unable to get a valid reset gpio\n"); } - pcie->num_slots++; - if (pcie->num_slots > MAX_PCI_SLOTS) { + if (pcie->num_slots + 1 >= MAX_PCI_SLOTS) { dev_err(dev, "Too many PCI slots!\n"); return -EINVAL; } + pcie->num_slots++; ret = of_pci_get_devfn(child); if (ret < 0) { diff --git a/drivers/pci/controller/dwc/pcie-qcom-common.c b/drivers/pci/controller/dwc/pcie-qcom-common.c new file mode 100644 index 000000000000..3aad19b56da8 --- /dev/null +++ b/drivers/pci/controller/dwc/pcie-qcom-common.c @@ -0,0 +1,78 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * Copyright (c) 2024 Qualcomm Innovation Center, Inc. All rights reserved. + */ + +#include + +#include "pcie-designware.h" +#include "pcie-qcom-common.h" + +void qcom_pcie_common_set_16gt_equalization(struct dw_pcie *pci) +{ + u32 reg; + + /* + * GEN3_RELATED_OFF register is repurposed to apply equalization + * settings at various data transmission rates through registers namely + * GEN3_EQ_*. The RATE_SHADOW_SEL bit field of GEN3_RELATED_OFF + * determines the data rate for which these equalization settings are + * applied. + */ + reg = dw_pcie_readl_dbi(pci, GEN3_RELATED_OFF); + reg &= ~GEN3_RELATED_OFF_GEN3_ZRXDC_NONCOMPL; + reg &= ~GEN3_RELATED_OFF_RATE_SHADOW_SEL_MASK; + reg |= FIELD_PREP(GEN3_RELATED_OFF_RATE_SHADOW_SEL_MASK, + GEN3_RELATED_OFF_RATE_SHADOW_SEL_16_0GT); + dw_pcie_writel_dbi(pci, GEN3_RELATED_OFF, reg); + + reg = dw_pcie_readl_dbi(pci, GEN3_EQ_FB_MODE_DIR_CHANGE_OFF); + reg &= ~(GEN3_EQ_FMDC_T_MIN_PHASE23 | + GEN3_EQ_FMDC_N_EVALS | + GEN3_EQ_FMDC_MAX_PRE_CUSROR_DELTA | + GEN3_EQ_FMDC_MAX_POST_CUSROR_DELTA); + reg |= FIELD_PREP(GEN3_EQ_FMDC_T_MIN_PHASE23, 0x1) | + FIELD_PREP(GEN3_EQ_FMDC_N_EVALS, 0xd) | + FIELD_PREP(GEN3_EQ_FMDC_MAX_PRE_CUSROR_DELTA, 0x5) | + FIELD_PREP(GEN3_EQ_FMDC_MAX_POST_CUSROR_DELTA, 0x5); + dw_pcie_writel_dbi(pci, GEN3_EQ_FB_MODE_DIR_CHANGE_OFF, reg); + + reg = dw_pcie_readl_dbi(pci, GEN3_EQ_CONTROL_OFF); + reg &= ~(GEN3_EQ_CONTROL_OFF_FB_MODE | + GEN3_EQ_CONTROL_OFF_PHASE23_EXIT_MODE | + GEN3_EQ_CONTROL_OFF_FOM_INC_INITIAL_EVAL | + GEN3_EQ_CONTROL_OFF_PSET_REQ_VEC); + dw_pcie_writel_dbi(pci, GEN3_EQ_CONTROL_OFF, reg); +} +EXPORT_SYMBOL_GPL(qcom_pcie_common_set_16gt_equalization); + +void qcom_pcie_common_set_16gt_lane_margining(struct dw_pcie *pci) +{ + u32 reg; + + reg = dw_pcie_readl_dbi(pci, GEN4_LANE_MARGINING_1_OFF); + reg &= ~(MARGINING_MAX_VOLTAGE_OFFSET | + MARGINING_NUM_VOLTAGE_STEPS | + MARGINING_MAX_TIMING_OFFSET | + MARGINING_NUM_TIMING_STEPS); + reg |= FIELD_PREP(MARGINING_MAX_VOLTAGE_OFFSET, 0x24) | + FIELD_PREP(MARGINING_NUM_VOLTAGE_STEPS, 0x78) | + FIELD_PREP(MARGINING_MAX_TIMING_OFFSET, 0x32) | + FIELD_PREP(MARGINING_NUM_TIMING_STEPS, 0x10); + dw_pcie_writel_dbi(pci, GEN4_LANE_MARGINING_1_OFF, reg); + + reg = dw_pcie_readl_dbi(pci, GEN4_LANE_MARGINING_2_OFF); + reg |= MARGINING_IND_ERROR_SAMPLER | + MARGINING_SAMPLE_REPORTING_METHOD | + MARGINING_IND_LEFT_RIGHT_TIMING | + MARGINING_VOLTAGE_SUPPORTED; + reg &= ~(MARGINING_IND_UP_DOWN_VOLTAGE | + MARGINING_MAXLANES | + MARGINING_SAMPLE_RATE_TIMING | + MARGINING_SAMPLE_RATE_VOLTAGE); + reg |= FIELD_PREP(MARGINING_MAXLANES, pci->num_lanes) | + FIELD_PREP(MARGINING_SAMPLE_RATE_TIMING, 0x3f) | + FIELD_PREP(MARGINING_SAMPLE_RATE_VOLTAGE, 0x3f); + dw_pcie_writel_dbi(pci, GEN4_LANE_MARGINING_2_OFF, reg); +} +EXPORT_SYMBOL_GPL(qcom_pcie_common_set_16gt_lane_margining); diff --git a/drivers/pci/controller/dwc/pcie-qcom-common.h b/drivers/pci/controller/dwc/pcie-qcom-common.h new file mode 100644 index 000000000000..7d88d29e4766 --- /dev/null +++ b/drivers/pci/controller/dwc/pcie-qcom-common.h @@ -0,0 +1,14 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* + * Copyright (c) 2024 Qualcomm Innovation Center, Inc. All rights reserved. + */ + +#ifndef _PCIE_QCOM_COMMON_H +#define _PCIE_QCOM_COMMON_H + +struct dw_pcie; + +void qcom_pcie_common_set_16gt_equalization(struct dw_pcie *pci); +void qcom_pcie_common_set_16gt_lane_margining(struct dw_pcie *pci); + +#endif diff --git a/drivers/pci/controller/dwc/pcie-qcom-ep.c b/drivers/pci/controller/dwc/pcie-qcom-ep.c index a9b263f749b6..e588fcc54589 100644 --- a/drivers/pci/controller/dwc/pcie-qcom-ep.c +++ b/drivers/pci/controller/dwc/pcie-qcom-ep.c @@ -25,6 +25,7 @@ #include "../../pci.h" #include "pcie-designware.h" +#include "pcie-qcom-common.h" /* PARF registers */ #define PARF_SYS_CTRL 0x00 @@ -498,6 +499,11 @@ static int qcom_pcie_perst_deassert(struct dw_pcie *pci) goto err_disable_resources; } + if (pcie_link_speed[pci->max_link_speed] == PCIE_SPEED_16_0GT) { + qcom_pcie_common_set_16gt_equalization(pci); + qcom_pcie_common_set_16gt_lane_margining(pci); + } + /* * The physical address of the MMIO region which is exposed as the BAR * should be written to MHI BASE registers. @@ -659,11 +665,9 @@ static irqreturn_t qcom_pcie_ep_global_irq_thread(int irq, void *data) struct dw_pcie *pci = &pcie_ep->pci; struct device *dev = pci->dev; u32 status = readl_relaxed(pcie_ep->parf + PARF_INT_ALL_STATUS); - u32 mask = readl_relaxed(pcie_ep->parf + PARF_INT_ALL_MASK); u32 dstate, val; writel_relaxed(status, pcie_ep->parf + PARF_INT_ALL_CLEAR); - status &= mask; if (FIELD_GET(PARF_INT_ALL_LINK_DOWN, status)) { dev_dbg(dev, "Received Linkdown event\n"); @@ -693,7 +697,8 @@ static irqreturn_t qcom_pcie_ep_global_irq_thread(int irq, void *data) dw_pcie_ep_linkup(&pci->ep); pcie_ep->link_status = QCOM_PCIE_EP_LINK_UP; } else { - dev_err(dev, "Received unknown event: %d\n", status); + dev_WARN_ONCE(dev, 1, "Received unknown event. INT_STATUS: 0x%08x\n", + status); } return IRQ_HANDLED; @@ -724,8 +729,15 @@ static irqreturn_t qcom_pcie_ep_perst_irq_thread(int irq, void *data) static int qcom_pcie_ep_enable_irq_resources(struct platform_device *pdev, struct qcom_pcie_ep *pcie_ep) { + struct device *dev = pcie_ep->pci.dev; + char *name; int ret; + name = devm_kasprintf(dev, GFP_KERNEL, "qcom_pcie_ep_global_irq%d", + pcie_ep->pci.ep.epc->domain_nr); + if (!name) + return -ENOMEM; + pcie_ep->global_irq = platform_get_irq_byname(pdev, "global"); if (pcie_ep->global_irq < 0) return pcie_ep->global_irq; @@ -733,18 +745,23 @@ static int qcom_pcie_ep_enable_irq_resources(struct platform_device *pdev, ret = devm_request_threaded_irq(&pdev->dev, pcie_ep->global_irq, NULL, qcom_pcie_ep_global_irq_thread, IRQF_ONESHOT, - "global_irq", pcie_ep); + name, pcie_ep); if (ret) { dev_err(&pdev->dev, "Failed to request Global IRQ\n"); return ret; } + name = devm_kasprintf(dev, GFP_KERNEL, "qcom_pcie_ep_perst_irq%d", + pcie_ep->pci.ep.epc->domain_nr); + if (!name) + return -ENOMEM; + pcie_ep->perst_irq = gpiod_to_irq(pcie_ep->reset); irq_set_status_flags(pcie_ep->perst_irq, IRQ_NOAUTOEN); ret = devm_request_threaded_irq(&pdev->dev, pcie_ep->perst_irq, NULL, qcom_pcie_ep_perst_irq_thread, IRQF_TRIGGER_HIGH | IRQF_ONESHOT, - "perst_irq", pcie_ep); + name, pcie_ep); if (ret) { dev_err(&pdev->dev, "Failed to request PERST IRQ\n"); disable_irq(pcie_ep->global_irq); @@ -858,21 +875,15 @@ static int qcom_pcie_ep_probe(struct platform_device *pdev) if (ret) return ret; - ret = qcom_pcie_enable_resources(pcie_ep); - if (ret) { - dev_err(dev, "Failed to enable resources: %d\n", ret); - return ret; - } - ret = dw_pcie_ep_init(&pcie_ep->pci.ep); if (ret) { dev_err(dev, "Failed to initialize endpoint: %d\n", ret); - goto err_disable_resources; + return ret; } ret = qcom_pcie_ep_enable_irq_resources(pdev, pcie_ep); if (ret) - goto err_disable_resources; + goto err_ep_deinit; name = devm_kasprintf(dev, GFP_KERNEL, "%pOFP", dev->of_node); if (!name) { @@ -889,8 +900,8 @@ err_disable_irqs: disable_irq(pcie_ep->global_irq); disable_irq(pcie_ep->perst_irq); -err_disable_resources: - qcom_pcie_disable_resources(pcie_ep); +err_ep_deinit: + dw_pcie_ep_deinit(&pcie_ep->pci.ep); return ret; } diff --git a/drivers/pci/controller/dwc/pcie-qcom.c b/drivers/pci/controller/dwc/pcie-qcom.c index 6f953e32d990..ef44a82be058 100644 --- a/drivers/pci/controller/dwc/pcie-qcom.c +++ b/drivers/pci/controller/dwc/pcie-qcom.c @@ -35,6 +35,7 @@ #include "../../pci.h" #include "pcie-designware.h" +#include "pcie-qcom-common.h" /* PARF registers */ #define PARF_SYS_CTRL 0x00 @@ -45,15 +46,24 @@ #define PARF_PHY_REFCLK 0x4c #define PARF_CONFIG_BITS 0x50 #define PARF_DBI_BASE_ADDR 0x168 +#define PARF_SLV_ADDR_SPACE_SIZE 0x16c #define PARF_MHI_CLOCK_RESET_CTRL 0x174 #define PARF_AXI_MSTR_WR_ADDR_HALT 0x178 #define PARF_AXI_MSTR_WR_ADDR_HALT_V2 0x1a8 #define PARF_Q2A_FLUSH 0x1ac #define PARF_LTSSM 0x1b0 +#define PARF_INT_ALL_STATUS 0x224 +#define PARF_INT_ALL_CLEAR 0x228 +#define PARF_INT_ALL_MASK 0x22c #define PARF_SID_OFFSET 0x234 #define PARF_BDF_TRANSLATE_CFG 0x24c -#define PARF_SLV_ADDR_SPACE_SIZE 0x358 +#define PARF_DBI_BASE_ADDR_V2 0x350 +#define PARF_DBI_BASE_ADDR_V2_HI 0x354 +#define PARF_SLV_ADDR_SPACE_SIZE_V2 0x358 +#define PARF_SLV_ADDR_SPACE_SIZE_V2_HI 0x35c #define PARF_NO_SNOOP_OVERIDE 0x3d4 +#define PARF_ATU_BASE_ADDR 0x634 +#define PARF_ATU_BASE_ADDR_HI 0x638 #define PARF_DEVICE_TYPE 0x1000 #define PARF_BDF_TO_SID_TABLE_N 0x2000 #define PARF_BDF_TO_SID_CFG 0x2c00 @@ -108,7 +118,7 @@ #define PHY_RX0_EQ(x) FIELD_PREP(GENMASK(26, 24), x) /* PARF_SLV_ADDR_SPACE_SIZE register value */ -#define SLV_ADDR_SPACE_SZ 0x10000000 +#define SLV_ADDR_SPACE_SZ 0x80000000 /* PARF_MHI_CLOCK_RESET_CTRL register fields */ #define AHB_CLK_EN BIT(0) @@ -121,6 +131,9 @@ /* PARF_LTSSM register fields */ #define LTSSM_EN BIT(8) +/* PARF_INT_ALL_{STATUS/CLEAR/MASK} register fields */ +#define PARF_INT_ALL_LINK_UP BIT(13) + /* PARF_NO_SNOOP_OVERIDE register fields */ #define WR_NO_SNOOP_OVERIDE_EN BIT(1) #define RD_NO_SNOOP_OVERIDE_EN BIT(3) @@ -284,6 +297,11 @@ static int qcom_pcie_start_link(struct dw_pcie *pci) { struct qcom_pcie *pcie = to_qcom_pcie(pci); + if (pcie_link_speed[pci->max_link_speed] == PCIE_SPEED_16_0GT) { + qcom_pcie_common_set_16gt_equalization(pci); + qcom_pcie_common_set_16gt_lane_margining(pci); + } + /* Enable Link Training state machine */ if (pcie->cfg->ops->ltssm_enable) pcie->cfg->ops->ltssm_enable(pcie); @@ -325,6 +343,50 @@ static void qcom_pcie_clear_hpc(struct dw_pcie *pci) dw_pcie_dbi_ro_wr_dis(pci); } +static void qcom_pcie_configure_dbi_base(struct qcom_pcie *pcie) +{ + struct dw_pcie *pci = pcie->pci; + + if (pci->dbi_phys_addr) { + /* + * PARF_DBI_BASE_ADDR register is in CPU domain and require to + * be programmed with CPU physical address. + */ + writel(lower_32_bits(pci->dbi_phys_addr), pcie->parf + + PARF_DBI_BASE_ADDR); + writel(SLV_ADDR_SPACE_SZ, pcie->parf + + PARF_SLV_ADDR_SPACE_SIZE); + } +} + +static void qcom_pcie_configure_dbi_atu_base(struct qcom_pcie *pcie) +{ + struct dw_pcie *pci = pcie->pci; + + if (pci->dbi_phys_addr) { + /* + * PARF_DBI_BASE_ADDR_V2 and PARF_ATU_BASE_ADDR registers are + * in CPU domain and require to be programmed with CPU + * physical addresses. + */ + writel(lower_32_bits(pci->dbi_phys_addr), pcie->parf + + PARF_DBI_BASE_ADDR_V2); + writel(upper_32_bits(pci->dbi_phys_addr), pcie->parf + + PARF_DBI_BASE_ADDR_V2_HI); + + if (pci->atu_phys_addr) { + writel(lower_32_bits(pci->atu_phys_addr), pcie->parf + + PARF_ATU_BASE_ADDR); + writel(upper_32_bits(pci->atu_phys_addr), pcie->parf + + PARF_ATU_BASE_ADDR_HI); + } + + writel(0x0, pcie->parf + PARF_SLV_ADDR_SPACE_SIZE_V2); + writel(SLV_ADDR_SPACE_SZ, pcie->parf + + PARF_SLV_ADDR_SPACE_SIZE_V2_HI); + } +} + static void qcom_pcie_2_1_0_ltssm_enable(struct qcom_pcie *pcie) { u32 val; @@ -541,8 +603,7 @@ err_assert_reset: static int qcom_pcie_post_init_1_0_0(struct qcom_pcie *pcie) { - /* change DBI base address */ - writel(0, pcie->parf + PARF_DBI_BASE_ADDR); + qcom_pcie_configure_dbi_base(pcie); if (IS_ENABLED(CONFIG_PCI_MSI)) { u32 val = readl(pcie->parf + PARF_AXI_MSTR_WR_ADDR_HALT); @@ -629,8 +690,7 @@ static int qcom_pcie_post_init_2_3_2(struct qcom_pcie *pcie) val &= ~PHY_TEST_PWR_DOWN; writel(val, pcie->parf + PARF_PHY_CTRL); - /* change DBI base address */ - writel(0, pcie->parf + PARF_DBI_BASE_ADDR); + qcom_pcie_configure_dbi_base(pcie); /* MAC PHY_POWERDOWN MUX DISABLE */ val = readl(pcie->parf + PARF_SYS_CTRL); @@ -812,13 +872,11 @@ static int qcom_pcie_post_init_2_3_3(struct qcom_pcie *pcie) u16 offset = dw_pcie_find_capability(pci, PCI_CAP_ID_EXP); u32 val; - writel(SLV_ADDR_SPACE_SZ, pcie->parf + PARF_SLV_ADDR_SPACE_SIZE); - val = readl(pcie->parf + PARF_PHY_CTRL); val &= ~PHY_TEST_PWR_DOWN; writel(val, pcie->parf + PARF_PHY_CTRL); - writel(0, pcie->parf + PARF_DBI_BASE_ADDR); + qcom_pcie_configure_dbi_atu_base(pcie); writel(MST_WAKEUP_EN | SLV_WAKEUP_EN | MSTR_ACLK_CGC_DIS | SLV_ACLK_CGC_DIS | CORE_CLK_CGC_DIS | @@ -914,8 +972,7 @@ static int qcom_pcie_init_2_7_0(struct qcom_pcie *pcie) val &= ~PHY_TEST_PWR_DOWN; writel(val, pcie->parf + PARF_PHY_CTRL); - /* change DBI base address */ - writel(0, pcie->parf + PARF_DBI_BASE_ADDR); + qcom_pcie_configure_dbi_atu_base(pcie); /* MAC PHY_POWERDOWN MUX DISABLE */ val = readl(pcie->parf + PARF_SYS_CTRL); @@ -1124,14 +1181,11 @@ static int qcom_pcie_post_init_2_9_0(struct qcom_pcie *pcie) u32 val; int i; - writel(SLV_ADDR_SPACE_SZ, - pcie->parf + PARF_SLV_ADDR_SPACE_SIZE); - val = readl(pcie->parf + PARF_PHY_CTRL); val &= ~PHY_TEST_PWR_DOWN; writel(val, pcie->parf + PARF_PHY_CTRL); - writel(0, pcie->parf + PARF_DBI_BASE_ADDR); + qcom_pcie_configure_dbi_atu_base(pcie); writel(DEVICE_TYPE_RC, pcie->parf + PARF_DEVICE_TYPE); writel(BYPASS | MSTR_AXI_CLK_EN | AHB_CLK_EN, @@ -1489,6 +1543,29 @@ static void qcom_pcie_init_debugfs(struct qcom_pcie *pcie) qcom_pcie_link_transition_count); } +static irqreturn_t qcom_pcie_global_irq_thread(int irq, void *data) +{ + struct qcom_pcie *pcie = data; + struct dw_pcie_rp *pp = &pcie->pci->pp; + struct device *dev = pcie->pci->dev; + u32 status = readl_relaxed(pcie->parf + PARF_INT_ALL_STATUS); + + writel_relaxed(status, pcie->parf + PARF_INT_ALL_CLEAR); + + if (FIELD_GET(PARF_INT_ALL_LINK_UP, status)) { + dev_dbg(dev, "Received Link up event. Starting enumeration!\n"); + /* Rescan the bus to enumerate endpoint devices */ + pci_lock_rescan_remove(); + pci_rescan_bus(pp->bridge->bus); + pci_unlock_rescan_remove(); + } else { + dev_WARN_ONCE(dev, 1, "Received unknown event. INT_STATUS: 0x%08x\n", + status); + } + + return IRQ_HANDLED; +} + static int qcom_pcie_probe(struct platform_device *pdev) { const struct qcom_pcie_cfg *pcie_cfg; @@ -1499,7 +1576,8 @@ static int qcom_pcie_probe(struct platform_device *pdev) struct dw_pcie_rp *pp; struct resource *res; struct dw_pcie *pci; - int ret; + int ret, irq; + char *name; pcie_cfg = of_device_get_match_data(dev); if (!pcie_cfg || !pcie_cfg->ops) { @@ -1620,6 +1698,27 @@ static int qcom_pcie_probe(struct platform_device *pdev) goto err_phy_exit; } + name = devm_kasprintf(dev, GFP_KERNEL, "qcom_pcie_global_irq%d", + pci_domain_nr(pp->bridge->bus)); + if (!name) { + ret = -ENOMEM; + goto err_host_deinit; + } + + irq = platform_get_irq_byname_optional(pdev, "global"); + if (irq > 0) { + ret = devm_request_threaded_irq(&pdev->dev, irq, NULL, + qcom_pcie_global_irq_thread, + IRQF_ONESHOT, name, pcie); + if (ret) { + dev_err_probe(&pdev->dev, ret, + "Failed to request Global IRQ\n"); + goto err_host_deinit; + } + + writel_relaxed(PARF_INT_ALL_LINK_UP, pcie->parf + PARF_INT_ALL_MASK); + } + qcom_pcie_icc_opp_update(pcie); if (pcie->mhi) @@ -1627,6 +1726,8 @@ static int qcom_pcie_probe(struct platform_device *pdev) return 0; +err_host_deinit: + dw_pcie_host_deinit(pp); err_phy_exit: phy_exit(pcie->phy); err_pm_runtime_put: diff --git a/drivers/pci/controller/dwc/pcie-rcar-gen4.c b/drivers/pci/controller/dwc/pcie-rcar-gen4.c index f0f3ebd1a033..3a5511c3f7d9 100644 --- a/drivers/pci/controller/dwc/pcie-rcar-gen4.c +++ b/drivers/pci/controller/dwc/pcie-rcar-gen4.c @@ -141,10 +141,10 @@ static int rcar_gen4_pcie_start_link(struct dw_pcie *dw) } /* - * Require direct speed change with retrying here if the link_gen is - * PCIe Gen2 or higher. + * Require direct speed change with retrying here if the max_link_speed + * is PCIe Gen2 or higher. */ - changes = min_not_zero(dw->link_gen, RCAR_MAX_LINK_SPEED) - 1; + changes = min_not_zero(dw->max_link_speed, RCAR_MAX_LINK_SPEED) - 1; /* * Since dw_pcie_setup_rc() sets it once, PCIe Gen2 will be trained. @@ -606,7 +606,12 @@ static int rcar_gen4_pcie_reg_test_bit(struct rcar_gen4_pcie *rcar, static int rcar_gen4_pcie_download_phy_firmware(struct rcar_gen4_pcie *rcar) { /* The check_addr values are magical numbers in the datasheet */ - const u32 check_addr[] = { 0x00101018, 0x00101118, 0x00101021, 0x00101121}; + static const u32 check_addr[] = { + 0x00101018, + 0x00101118, + 0x00101021, + 0x00101121, + }; struct dw_pcie *dw = &rcar->dw; const struct firmware *fw; unsigned int i, timeout; diff --git a/drivers/pci/controller/dwc/pcie-spear13xx.c b/drivers/pci/controller/dwc/pcie-spear13xx.c index 201dced209f0..ff986ced56b2 100644 --- a/drivers/pci/controller/dwc/pcie-spear13xx.c +++ b/drivers/pci/controller/dwc/pcie-spear13xx.c @@ -233,7 +233,7 @@ static int spear13xx_pcie_probe(struct platform_device *pdev) } if (of_property_read_bool(np, "st,pcie-is-gen1")) - pci->link_gen = 1; + pci->max_link_speed = 1; platform_set_drvdata(pdev, spear13xx_pcie); diff --git a/drivers/pci/controller/dwc/pcie-tegra194.c b/drivers/pci/controller/dwc/pcie-tegra194.c index 4bf7b433417a..c1394f2ab63f 100644 --- a/drivers/pci/controller/dwc/pcie-tegra194.c +++ b/drivers/pci/controller/dwc/pcie-tegra194.c @@ -177,17 +177,12 @@ #define N_FTS_VAL 52 #define FTS_VAL 52 -#define GEN3_EQ_CONTROL_OFF 0x8a8 -#define GEN3_EQ_CONTROL_OFF_PSET_REQ_VEC_SHIFT 8 -#define GEN3_EQ_CONTROL_OFF_PSET_REQ_VEC_MASK GENMASK(23, 8) -#define GEN3_EQ_CONTROL_OFF_FB_MODE_MASK GENMASK(3, 0) - #define PORT_LOGIC_AMBA_ERROR_RESPONSE_DEFAULT 0x8D0 -#define AMBA_ERROR_RESPONSE_CRS_SHIFT 3 -#define AMBA_ERROR_RESPONSE_CRS_MASK GENMASK(1, 0) -#define AMBA_ERROR_RESPONSE_CRS_OKAY 0 -#define AMBA_ERROR_RESPONSE_CRS_OKAY_FFFFFFFF 1 -#define AMBA_ERROR_RESPONSE_CRS_OKAY_FFFF0001 2 +#define AMBA_ERROR_RESPONSE_RRS_SHIFT 3 +#define AMBA_ERROR_RESPONSE_RRS_MASK GENMASK(1, 0) +#define AMBA_ERROR_RESPONSE_RRS_OKAY 0 +#define AMBA_ERROR_RESPONSE_RRS_OKAY_FFFFFFFF 1 +#define AMBA_ERROR_RESPONSE_RRS_OKAY_FFFF0001 2 #define MSIX_ADDR_MATCH_LOW_OFF 0x940 #define MSIX_ADDR_MATCH_LOW_OFF_EN BIT(0) @@ -861,9 +856,9 @@ static void config_gen3_gen4_eq_presets(struct tegra_pcie_dw *pcie) dw_pcie_writel_dbi(pci, GEN3_RELATED_OFF, val); val = dw_pcie_readl_dbi(pci, GEN3_EQ_CONTROL_OFF); - val &= ~GEN3_EQ_CONTROL_OFF_PSET_REQ_VEC_MASK; - val |= (0x3ff << GEN3_EQ_CONTROL_OFF_PSET_REQ_VEC_SHIFT); - val &= ~GEN3_EQ_CONTROL_OFF_FB_MODE_MASK; + val &= ~GEN3_EQ_CONTROL_OFF_PSET_REQ_VEC; + val |= FIELD_PREP(GEN3_EQ_CONTROL_OFF_PSET_REQ_VEC, 0x3ff); + val &= ~GEN3_EQ_CONTROL_OFF_FB_MODE; dw_pcie_writel_dbi(pci, GEN3_EQ_CONTROL_OFF, val); val = dw_pcie_readl_dbi(pci, GEN3_RELATED_OFF); @@ -872,10 +867,10 @@ static void config_gen3_gen4_eq_presets(struct tegra_pcie_dw *pcie) dw_pcie_writel_dbi(pci, GEN3_RELATED_OFF, val); val = dw_pcie_readl_dbi(pci, GEN3_EQ_CONTROL_OFF); - val &= ~GEN3_EQ_CONTROL_OFF_PSET_REQ_VEC_MASK; - val |= (pcie->of_data->gen4_preset_vec << - GEN3_EQ_CONTROL_OFF_PSET_REQ_VEC_SHIFT); - val &= ~GEN3_EQ_CONTROL_OFF_FB_MODE_MASK; + val &= ~GEN3_EQ_CONTROL_OFF_PSET_REQ_VEC; + val |= FIELD_PREP(GEN3_EQ_CONTROL_OFF_PSET_REQ_VEC, + pcie->of_data->gen4_preset_vec); + val &= ~GEN3_EQ_CONTROL_OFF_FB_MODE; dw_pcie_writel_dbi(pci, GEN3_EQ_CONTROL_OFF, val); val = dw_pcie_readl_dbi(pci, GEN3_RELATED_OFF); @@ -907,11 +902,11 @@ static int tegra_pcie_dw_host_init(struct dw_pcie_rp *pp) dw_pcie_writel_dbi(pci, PCI_BASE_ADDRESS_0, 0); - /* Enable as 0xFFFF0001 response for CRS */ + /* Enable as 0xFFFF0001 response for RRS */ val = dw_pcie_readl_dbi(pci, PORT_LOGIC_AMBA_ERROR_RESPONSE_DEFAULT); - val &= ~(AMBA_ERROR_RESPONSE_CRS_MASK << AMBA_ERROR_RESPONSE_CRS_SHIFT); - val |= (AMBA_ERROR_RESPONSE_CRS_OKAY_FFFF0001 << - AMBA_ERROR_RESPONSE_CRS_SHIFT); + val &= ~(AMBA_ERROR_RESPONSE_RRS_MASK << AMBA_ERROR_RESPONSE_RRS_SHIFT); + val |= (AMBA_ERROR_RESPONSE_RRS_OKAY_FFFF0001 << + AMBA_ERROR_RESPONSE_RRS_SHIFT); dw_pcie_writel_dbi(pci, PORT_LOGIC_AMBA_ERROR_RESPONSE_DEFAULT, val); /* Clear Slot Clock Configuration bit if SRNS configuration */ diff --git a/drivers/pci/controller/mobiveil/pcie-mobiveil-host.c b/drivers/pci/controller/mobiveil/pcie-mobiveil-host.c index 32951f7d6d6d..0e088e74155d 100644 --- a/drivers/pci/controller/mobiveil/pcie-mobiveil-host.c +++ b/drivers/pci/controller/mobiveil/pcie-mobiveil-host.c @@ -360,8 +360,8 @@ static struct irq_chip mobiveil_msi_irq_chip = { }; static struct msi_domain_info mobiveil_msi_domain_info = { - .flags = (MSI_FLAG_USE_DEF_DOM_OPS | MSI_FLAG_USE_DEF_CHIP_OPS | - MSI_FLAG_PCI_MSIX), + .flags = MSI_FLAG_USE_DEF_DOM_OPS | MSI_FLAG_USE_DEF_CHIP_OPS | + MSI_FLAG_NO_AFFINITY | MSI_FLAG_PCI_MSIX, .chip = &mobiveil_msi_irq_chip, }; @@ -378,16 +378,9 @@ static void mobiveil_compose_msi_msg(struct irq_data *data, struct msi_msg *msg) (int)data->hwirq, msg->address_hi, msg->address_lo); } -static int mobiveil_msi_set_affinity(struct irq_data *irq_data, - const struct cpumask *mask, bool force) -{ - return -EINVAL; -} - static struct irq_chip mobiveil_msi_bottom_irq_chip = { .name = "Mobiveil MSI", .irq_compose_msi_msg = mobiveil_compose_msi_msg, - .irq_set_affinity = mobiveil_msi_set_affinity, }; static int mobiveil_irq_msi_domain_alloc(struct irq_domain *domain, diff --git a/drivers/pci/controller/pci-aardvark.c b/drivers/pci/controller/pci-aardvark.c index 8b3e1a079cf3..a598a98247ce 100644 --- a/drivers/pci/controller/pci-aardvark.c +++ b/drivers/pci/controller/pci-aardvark.c @@ -50,7 +50,7 @@ #define PIO_COMPLETION_STATUS_MASK GENMASK(9, 7) #define PIO_COMPLETION_STATUS_OK 0 #define PIO_COMPLETION_STATUS_UR 1 -#define PIO_COMPLETION_STATUS_CRS 2 +#define PIO_COMPLETION_STATUS_RRS 2 #define PIO_COMPLETION_STATUS_CA 4 #define PIO_NON_POSTED_REQ BIT(10) #define PIO_ERR_STATUS BIT(11) @@ -262,7 +262,7 @@ enum { #define MSI_IRQ_NUM 32 -#define CFG_RD_CRS_VAL 0xffff0001 +#define CFG_RD_RRS_VAL 0xffff0001 struct advk_pcie { struct platform_device *pdev; @@ -649,7 +649,7 @@ static void advk_pcie_setup_hw(struct advk_pcie *pcie) advk_pcie_train_link(pcie); } -static int advk_pcie_check_pio_status(struct advk_pcie *pcie, bool allow_crs, u32 *val) +static int advk_pcie_check_pio_status(struct advk_pcie *pcie, bool allow_rrs, u32 *val) { struct device *dev = &pcie->pdev->dev; u32 reg; @@ -669,7 +669,7 @@ static int advk_pcie_check_pio_status(struct advk_pcie *pcie, bool allow_crs, u3 * 2) value Unsupported Request(1) of COMPLETION_STATUS(bit9:7) only * means a PIO write error, and for PIO read it is successful with * a read value of 0xFFFFFFFF. - * 3) value Completion Retry Status(CRS) of COMPLETION_STATUS(bit9:7) + * 3) value Config Request Retry Status(RRS) of COMPLETION_STATUS(bit9:7) * only means a PIO write error, and for PIO read it is successful * with a read value of 0xFFFF0001. * 4) value Completer Abort (CA) of COMPLETION_STATUS(bit9:7) means @@ -694,10 +694,10 @@ static int advk_pcie_check_pio_status(struct advk_pcie *pcie, bool allow_crs, u3 strcomp_status = "UR"; ret = -EOPNOTSUPP; break; - case PIO_COMPLETION_STATUS_CRS: - if (allow_crs && val) { - /* PCIe r4.0, sec 2.3.2, says: - * If CRS Software Visibility is enabled: + case PIO_COMPLETION_STATUS_RRS: + if (allow_rrs && val) { + /* PCIe r6.0, sec 2.3.2, says: + * If Configuration RRS Software Visibility is enabled: * For a Configuration Read Request that includes both * bytes of the Vendor ID field of a device Function's * Configuration Space Header, the Root Complex must @@ -706,22 +706,22 @@ static int advk_pcie_check_pio_status(struct advk_pcie *pcie, bool allow_crs, u3 * all '1's for any additional bytes included in the * request. * - * So CRS in this case is not an error status. + * So RRS in this case is not an error status. */ - *val = CFG_RD_CRS_VAL; + *val = CFG_RD_RRS_VAL; strcomp_status = NULL; ret = 0; break; } - /* PCIe r4.0, sec 2.3.2, says: - * If CRS Software Visibility is not enabled, the Root Complex + /* PCIe r6.0, sec 2.3.2, says: + * If RRS Software Visibility is not enabled, the Root Complex * must re-issue the Configuration Request as a new Request. - * If CRS Software Visibility is enabled: For a Configuration + * If RRS Software Visibility is enabled: For a Configuration * Write Request or for any other Configuration Read Request, * the Root Complex must re-issue the Configuration Request as * a new Request. * A Root Complex implementation may choose to limit the number - * of Configuration Request/CRS Completion Status loops before + * of Configuration Request/RRS Completion Status loops before * determining that something is wrong with the target of the * Request and taking appropriate action, e.g., complete the * Request to the host as a failed transaction. @@ -729,7 +729,7 @@ static int advk_pcie_check_pio_status(struct advk_pcie *pcie, bool allow_crs, u3 * So return -EAGAIN and caller (pci-aardvark.c driver) will * re-issue request again up to the PIO_RETRY_CNT retries. */ - strcomp_status = "CRS"; + strcomp_status = "RRS"; ret = -EAGAIN; break; case PIO_COMPLETION_STATUS_CA: @@ -920,8 +920,8 @@ advk_pci_bridge_emul_pcie_conf_write(struct pci_bridge_emul *bridge, case PCI_EXP_RTCTL: { u16 rootctl = le16_to_cpu(bridge->pcie_conf.rootctl); - /* Only emulation of PMEIE and CRSSVE bits is provided */ - rootctl &= PCI_EXP_RTCTL_PMEIE | PCI_EXP_RTCTL_CRSSVE; + /* Only emulation of PMEIE and RRS_SVE bits is provided */ + rootctl &= PCI_EXP_RTCTL_PMEIE | PCI_EXP_RTCTL_RRS_SVE; bridge->pcie_conf.rootctl = cpu_to_le16(rootctl); break; } @@ -1075,7 +1075,7 @@ static int advk_sw_pci_bridge_init(struct advk_pcie *pcie) bridge->pcie_conf.slotsta = cpu_to_le16(PCI_EXP_SLTSTA_PDS); /* Indicates supports for Completion Retry Status */ - bridge->pcie_conf.rootcap = cpu_to_le16(PCI_EXP_RTCAP_CRSVIS); + bridge->pcie_conf.rootcap = cpu_to_le16(PCI_EXP_RTCAP_RRS_SV); bridge->subsystem_vendor_id = advk_readl(pcie, PCIE_CORE_SSDEV_ID_REG) & 0xffff; bridge->subsystem_id = advk_readl(pcie, PCIE_CORE_SSDEV_ID_REG) >> 16; @@ -1141,7 +1141,7 @@ static int advk_pcie_rd_conf(struct pci_bus *bus, u32 devfn, { struct advk_pcie *pcie = bus->sysdata; int retry_count; - bool allow_crs; + bool allow_rrs; u32 reg; int ret; @@ -1153,16 +1153,16 @@ static int advk_pcie_rd_conf(struct pci_bus *bus, u32 devfn, size, val); /* - * Completion Retry Status is possible to return only when reading all - * 4 bytes from PCI_VENDOR_ID and PCI_DEVICE_ID registers at once and - * CRSSVE flag on Root Bridge is enabled. + * Configuration Request Retry Status (RRS) is possible to return + * only when reading both bytes from PCI_VENDOR_ID at once and + * RRS_SVE flag on Root Port is enabled. */ - allow_crs = (where == PCI_VENDOR_ID) && (size == 4) && + allow_rrs = (where == PCI_VENDOR_ID) && (size >= 2) && (le16_to_cpu(pcie->bridge.pcie_conf.rootctl) & - PCI_EXP_RTCTL_CRSSVE); + PCI_EXP_RTCTL_RRS_SVE); if (advk_pcie_pio_is_running(pcie)) - goto try_crs; + goto try_rrs; /* Program the control register */ reg = advk_readl(pcie, PIO_CTRL); @@ -1189,12 +1189,12 @@ static int advk_pcie_rd_conf(struct pci_bus *bus, u32 devfn, ret = advk_pcie_wait_pio(pcie); if (ret < 0) - goto try_crs; + goto try_rrs; retry_count += ret; /* Check PIO status and get the read result */ - ret = advk_pcie_check_pio_status(pcie, allow_crs, val); + ret = advk_pcie_check_pio_status(pcie, allow_rrs, val); } while (ret == -EAGAIN && retry_count < PIO_RETRY_CNT); if (ret < 0) @@ -1207,13 +1207,13 @@ static int advk_pcie_rd_conf(struct pci_bus *bus, u32 devfn, return PCIBIOS_SUCCESSFUL; -try_crs: +try_rrs: /* - * If it is possible, return Completion Retry Status so that caller - * tries to issue the request again instead of failing. + * If it is possible, return Configuration Request Retry Status so + * that caller tries to issue the request again instead of failing. */ - if (allow_crs) { - *val = CFG_RD_CRS_VAL; + if (allow_rrs) { + *val = CFG_RD_RRS_VAL; return PCIBIOS_SUCCESSFUL; } @@ -1304,12 +1304,6 @@ static void advk_msi_irq_compose_msi_msg(struct irq_data *data, msg->data = data->hwirq; } -static int advk_msi_set_affinity(struct irq_data *irq_data, - const struct cpumask *mask, bool force) -{ - return -EINVAL; -} - static void advk_msi_irq_mask(struct irq_data *d) { struct advk_pcie *pcie = d->domain->host_data; @@ -1353,7 +1347,6 @@ static void advk_msi_top_irq_unmask(struct irq_data *d) static struct irq_chip advk_msi_bottom_irq_chip = { .name = "MSI", .irq_compose_msi_msg = advk_msi_irq_compose_msi_msg, - .irq_set_affinity = advk_msi_set_affinity, .irq_mask = advk_msi_irq_mask, .irq_unmask = advk_msi_irq_unmask, }; @@ -1451,7 +1444,8 @@ static struct irq_chip advk_msi_irq_chip = { static struct msi_domain_info advk_msi_domain_info = { .flags = MSI_FLAG_USE_DEF_DOM_OPS | MSI_FLAG_USE_DEF_CHIP_OPS | - MSI_FLAG_MULTI_PCI_MSI | MSI_FLAG_PCI_MSIX, + MSI_FLAG_NO_AFFINITY | MSI_FLAG_MULTI_PCI_MSI | + MSI_FLAG_PCI_MSIX, .chip = &advk_msi_irq_chip, }; diff --git a/drivers/pci/controller/pci-tegra.c b/drivers/pci/controller/pci-tegra.c index 038d974a318e..d7517c3976e7 100644 --- a/drivers/pci/controller/pci-tegra.c +++ b/drivers/pci/controller/pci-tegra.c @@ -1629,11 +1629,6 @@ static void tegra_msi_irq_unmask(struct irq_data *d) spin_unlock_irqrestore(&msi->mask_lock, flags); } -static int tegra_msi_set_affinity(struct irq_data *d, const struct cpumask *mask, bool force) -{ - return -EINVAL; -} - static void tegra_compose_msi_msg(struct irq_data *data, struct msi_msg *msg) { struct tegra_msi *msi = irq_data_get_irq_chip_data(data); @@ -1648,7 +1643,6 @@ static struct irq_chip tegra_msi_bottom_chip = { .irq_ack = tegra_msi_irq_ack, .irq_mask = tegra_msi_irq_mask, .irq_unmask = tegra_msi_irq_unmask, - .irq_set_affinity = tegra_msi_set_affinity, .irq_compose_msi_msg = tegra_compose_msi_msg, }; @@ -1697,8 +1691,8 @@ static const struct irq_domain_ops tegra_msi_domain_ops = { }; static struct msi_domain_info tegra_msi_info = { - .flags = (MSI_FLAG_USE_DEF_DOM_OPS | MSI_FLAG_USE_DEF_CHIP_OPS | - MSI_FLAG_PCI_MSIX), + .flags = MSI_FLAG_USE_DEF_DOM_OPS | MSI_FLAG_USE_DEF_CHIP_OPS | + MSI_FLAG_NO_AFFINITY | MSI_FLAG_PCI_MSIX, .chip = &tegra_msi_top_chip, }; diff --git a/drivers/pci/controller/pci-xgene.c b/drivers/pci/controller/pci-xgene.c index 8e457fa450a2..1e2ebbfa36d1 100644 --- a/drivers/pci/controller/pci-xgene.c +++ b/drivers/pci/controller/pci-xgene.c @@ -171,17 +171,17 @@ static int xgene_pcie_config_read32(struct pci_bus *bus, unsigned int devfn, /* * The v1 controller has a bug in its Configuration Request Retry - * Status (CRS) logic: when CRS Software Visibility is enabled and + * Status (RRS) logic: when RRS Software Visibility is enabled and * we read the Vendor and Device ID of a non-existent device, the * controller fabricates return data of 0xFFFF0001 ("device exists * but is not ready") instead of 0xFFFFFFFF (PCI_ERROR_RESPONSE) * ("device does not exist"). This causes the PCI core to retry * the read until it times out. Avoid this by not claiming to - * support CRS SV. + * support RRS SV. */ if (pci_is_root_bus(bus) && (port->version == XGENE_PCIE_IP_VER_1) && ((where & ~0x3) == XGENE_V1_PCI_EXP_CAP + PCI_EXP_RTCTL)) - *val &= ~(PCI_EXP_RTCAP_CRSVIS << 16); + *val &= ~(PCI_EXP_RTCAP_RRS_SV << 16); if (size <= 2) *val = (*val >> (8 * (where & 3))) & ((1 << (size * 8)) - 1); diff --git a/drivers/pci/controller/pcie-altera-msi.c b/drivers/pci/controller/pcie-altera-msi.c index 16336a525c16..e36a6e158d23 100644 --- a/drivers/pci/controller/pcie-altera-msi.c +++ b/drivers/pci/controller/pcie-altera-msi.c @@ -81,8 +81,8 @@ static struct irq_chip altera_msi_irq_chip = { }; static struct msi_domain_info altera_msi_domain_info = { - .flags = (MSI_FLAG_USE_DEF_DOM_OPS | MSI_FLAG_USE_DEF_CHIP_OPS | - MSI_FLAG_PCI_MSIX), + .flags = MSI_FLAG_USE_DEF_DOM_OPS | MSI_FLAG_USE_DEF_CHIP_OPS | + MSI_FLAG_NO_AFFINITY | MSI_FLAG_PCI_MSIX, .chip = &altera_msi_irq_chip, }; @@ -99,16 +99,9 @@ static void altera_compose_msi_msg(struct irq_data *data, struct msi_msg *msg) (int)data->hwirq, msg->address_hi, msg->address_lo); } -static int altera_msi_set_affinity(struct irq_data *irq_data, - const struct cpumask *mask, bool force) -{ - return -EINVAL; -} - static struct irq_chip altera_msi_bottom_irq_chip = { .name = "Altera MSI", .irq_compose_msi_msg = altera_compose_msi_msg, - .irq_set_affinity = altera_msi_set_affinity, }; static int altera_irq_domain_alloc(struct irq_domain *domain, unsigned int virq, diff --git a/drivers/pci/controller/pcie-altera.c b/drivers/pci/controller/pcie-altera.c index ef73baefaeb9..650b2dd81c48 100644 --- a/drivers/pci/controller/pcie-altera.c +++ b/drivers/pci/controller/pcie-altera.c @@ -55,12 +55,11 @@ #define TLP_READ_TAG 0x1d #define TLP_WRITE_TAG 0x10 #define RP_DEVFN 0 -#define TLP_REQ_ID(bus, devfn) (((bus) << 8) | (devfn)) #define TLP_CFG_DW0(pcie, cfg) \ (((cfg) << 24) | \ TLP_PAYLOAD_SIZE) #define TLP_CFG_DW1(pcie, tag, be) \ - (((TLP_REQ_ID(pcie->root_bus_nr, RP_DEVFN)) << 16) | (tag << 8) | (be)) + (((PCI_DEVID(pcie->root_bus_nr, RP_DEVFN)) << 16) | (tag << 8) | (be)) #define TLP_CFG_DW2(bus, devfn, offset) \ (((bus) << 24) | ((devfn) << 16) | (offset)) #define TLP_COMP_STATUS(s) (((s) >> 13) & 7) diff --git a/drivers/pci/controller/pcie-brcmstb.c b/drivers/pci/controller/pcie-brcmstb.c index c08683febdd4..9321280f6edb 100644 --- a/drivers/pci/controller/pcie-brcmstb.c +++ b/drivers/pci/controller/pcie-brcmstb.c @@ -75,15 +75,19 @@ #define PCIE_MEM_WIN0_HI(win) \ PCIE_MISC_CPU_2_PCIE_MEM_WIN0_HI + ((win) * 8) +/* + * NOTE: You may see the term "BAR" in a number of register names used by + * this driver. The term is an artifact of when the HW core was an + * endpoint device (EP). Now it is a root complex (RC) and anywhere a + * register has the term "BAR" it is related to an inbound window. + */ + +#define PCIE_BRCM_MAX_INBOUND_WINS 16 #define PCIE_MISC_RC_BAR1_CONFIG_LO 0x402c #define PCIE_MISC_RC_BAR1_CONFIG_LO_SIZE_MASK 0x1f -#define PCIE_MISC_RC_BAR2_CONFIG_LO 0x4034 -#define PCIE_MISC_RC_BAR2_CONFIG_LO_SIZE_MASK 0x1f -#define PCIE_MISC_RC_BAR2_CONFIG_HI 0x4038 +#define PCIE_MISC_RC_BAR4_CONFIG_LO 0x40d4 -#define PCIE_MISC_RC_BAR3_CONFIG_LO 0x403c -#define PCIE_MISC_RC_BAR3_CONFIG_LO_SIZE_MASK 0x1f #define PCIE_MISC_MSI_BAR_CONFIG_LO 0x4044 #define PCIE_MISC_MSI_BAR_CONFIG_HI 0x4048 @@ -122,7 +126,6 @@ #define PCIE_MEM_WIN0_LIMIT_HI(win) \ PCIE_MISC_CPU_2_PCIE_MEM_WIN0_LIMIT_HI + ((win) * 8) -#define PCIE_MISC_HARD_PCIE_HARD_DEBUG 0x4204 #define PCIE_MISC_HARD_PCIE_HARD_DEBUG_CLKREQ_DEBUG_ENABLE_MASK 0x2 #define PCIE_MISC_HARD_PCIE_HARD_DEBUG_L1SS_ENABLE_MASK 0x200000 #define PCIE_MISC_HARD_PCIE_HARD_DEBUG_SERDES_IDDQ_MASK 0x08000000 @@ -131,9 +134,13 @@ (PCIE_MISC_HARD_PCIE_HARD_DEBUG_CLKREQ_DEBUG_ENABLE_MASK | \ PCIE_MISC_HARD_PCIE_HARD_DEBUG_L1SS_ENABLE_MASK) -#define PCIE_INTR2_CPU_BASE 0x4300 +#define PCIE_MISC_UBUS_BAR1_CONFIG_REMAP 0x40ac +#define PCIE_MISC_UBUS_BAR1_CONFIG_REMAP_ACCESS_EN_MASK BIT(0) +#define PCIE_MISC_UBUS_BAR4_CONFIG_REMAP 0x410c + #define PCIE_MSI_INTR2_BASE 0x4500 -/* Offsets from PCIE_INTR2_CPU_BASE and PCIE_MSI_INTR2_BASE */ + +/* Offsets from INTR2_CPU and MSI_INTR2 BASE offsets */ #define MSI_INT_STATUS 0x0 #define MSI_INT_CLR 0x8 #define MSI_INT_MASK_SET 0x10 @@ -184,9 +191,11 @@ #define SSC_STATUS_PLL_LOCK_MASK 0x800 #define PCIE_BRCM_MAX_MEMC 3 -#define IDX_ADDR(pcie) (pcie->reg_offsets[EXT_CFG_INDEX]) -#define DATA_ADDR(pcie) (pcie->reg_offsets[EXT_CFG_DATA]) -#define PCIE_RGR1_SW_INIT_1(pcie) (pcie->reg_offsets[RGR1_SW_INIT_1]) +#define IDX_ADDR(pcie) ((pcie)->reg_offsets[EXT_CFG_INDEX]) +#define DATA_ADDR(pcie) ((pcie)->reg_offsets[EXT_CFG_DATA]) +#define PCIE_RGR1_SW_INIT_1(pcie) ((pcie)->reg_offsets[RGR1_SW_INIT_1]) +#define HARD_DEBUG(pcie) ((pcie)->reg_offsets[PCIE_HARD_DEBUG]) +#define INTR2_CPU_BASE(pcie) ((pcie)->reg_offsets[PCIE_INTR2_CPU_BASE]) /* Rescal registers */ #define PCIE_DVT_PMU_PCIE_PHY_CTRL 0xc700 @@ -205,27 +214,33 @@ enum { RGR1_SW_INIT_1, EXT_CFG_INDEX, EXT_CFG_DATA, + PCIE_HARD_DEBUG, + PCIE_INTR2_CPU_BASE, }; -enum { - RGR1_SW_INIT_1_INIT_MASK, - RGR1_SW_INIT_1_INIT_SHIFT, -}; - -enum pcie_type { +enum pcie_soc_base { GENERIC, - BCM7425, - BCM7435, + BCM2711, BCM4908, BCM7278, - BCM2711, + BCM7425, + BCM7435, + BCM7712, +}; + +struct inbound_win { + u64 size; + u64 pci_offset; + u64 cpu_addr; }; struct pcie_cfg_data { const int *offsets; - const enum pcie_type type; - void (*perst_set)(struct brcm_pcie *pcie, u32 val); - void (*bridge_sw_init_set)(struct brcm_pcie *pcie, u32 val); + const enum pcie_soc_base soc_base; + const bool has_phy; + u8 num_inbound_wins; + int (*perst_set)(struct brcm_pcie *pcie, u32 val); + int (*bridge_sw_init_set)(struct brcm_pcie *pcie, u32 val); }; struct subdev_regulators { @@ -262,21 +277,25 @@ struct brcm_pcie { u64 msi_target_addr; struct brcm_msi *msi; const int *reg_offsets; - enum pcie_type type; + enum pcie_soc_base soc_base; struct reset_control *rescal; struct reset_control *perst_reset; + struct reset_control *bridge_reset; + struct reset_control *swinit_reset; int num_memc; u64 memc_size[PCIE_BRCM_MAX_MEMC]; u32 hw_rev; - void (*perst_set)(struct brcm_pcie *pcie, u32 val); - void (*bridge_sw_init_set)(struct brcm_pcie *pcie, u32 val); + int (*perst_set)(struct brcm_pcie *pcie, u32 val); + int (*bridge_sw_init_set)(struct brcm_pcie *pcie, u32 val); struct subdev_regulators *sr; bool ep_wakeup_capable; + bool has_phy; + u8 num_inbound_wins; }; static inline bool is_bmips(const struct brcm_pcie *pcie) { - return pcie->type == BCM7435 || pcie->type == BCM7425; + return pcie->soc_base == BCM7435 || pcie->soc_base == BCM7425; } /* @@ -394,7 +413,7 @@ static void brcm_pcie_set_gen(struct brcm_pcie *pcie, int gen) } static void brcm_pcie_set_outbound_win(struct brcm_pcie *pcie, - unsigned int win, u64 cpu_addr, + u8 win, u64 cpu_addr, u64 pcie_addr, u64 size) { u32 cpu_addr_mb_high, limit_addr_mb_high; @@ -445,8 +464,8 @@ static struct irq_chip brcm_msi_irq_chip = { }; static struct msi_domain_info brcm_msi_domain_info = { - .flags = (MSI_FLAG_USE_DEF_DOM_OPS | MSI_FLAG_USE_DEF_CHIP_OPS | - MSI_FLAG_MULTI_PCI_MSI), + .flags = MSI_FLAG_USE_DEF_DOM_OPS | MSI_FLAG_USE_DEF_CHIP_OPS | + MSI_FLAG_NO_AFFINITY | MSI_FLAG_MULTI_PCI_MSI, .chip = &brcm_msi_irq_chip, }; @@ -484,12 +503,6 @@ static void brcm_msi_compose_msi_msg(struct irq_data *data, struct msi_msg *msg) msg->data = (0xffff & PCIE_MISC_MSI_DATA_CONFIG_VAL_32) | data->hwirq; } -static int brcm_msi_set_affinity(struct irq_data *irq_data, - const struct cpumask *mask, bool force) -{ - return -EINVAL; -} - static void brcm_msi_ack_irq(struct irq_data *data) { struct brcm_msi *msi = irq_data_get_irq_chip_data(data); @@ -502,7 +515,6 @@ static void brcm_msi_ack_irq(struct irq_data *data) static struct irq_chip brcm_msi_bottom_irq_chip = { .name = "BRCM STB MSI", .irq_compose_msi_msg = brcm_msi_compose_msi_msg, - .irq_set_affinity = brcm_msi_set_affinity, .irq_ack = brcm_msi_ack_irq, }; @@ -649,7 +661,7 @@ static int brcm_pcie_enable_msi(struct brcm_pcie *pcie) BUILD_BUG_ON(BRCM_INT_PCI_MSI_LEGACY_NR > BRCM_INT_PCI_MSI_NR); if (msi->legacy) { - msi->intr_base = msi->base + PCIE_INTR2_CPU_BASE; + msi->intr_base = msi->base + INTR2_CPU_BASE(pcie); msi->nr = BRCM_INT_PCI_MSI_LEGACY_NR; msi->legacy_shift = 24; } else { @@ -730,17 +742,33 @@ static void __iomem *brcm7425_pcie_map_bus(struct pci_bus *bus, return base + DATA_ADDR(pcie); } -static void brcm_pcie_bridge_sw_init_set_generic(struct brcm_pcie *pcie, u32 val) +static int brcm_pcie_bridge_sw_init_set_generic(struct brcm_pcie *pcie, u32 val) { - u32 tmp, mask = RGR1_SW_INIT_1_INIT_GENERIC_MASK; + u32 tmp, mask = RGR1_SW_INIT_1_INIT_GENERIC_MASK; u32 shift = RGR1_SW_INIT_1_INIT_GENERIC_SHIFT; + int ret = 0; + + if (pcie->bridge_reset) { + if (val) + ret = reset_control_assert(pcie->bridge_reset); + else + ret = reset_control_deassert(pcie->bridge_reset); + + if (ret) + dev_err(pcie->dev, "failed to %s 'bridge' reset, err=%d\n", + val ? "assert" : "deassert", ret); + + return ret; + } tmp = readl(pcie->base + PCIE_RGR1_SW_INIT_1(pcie)); tmp = (tmp & ~mask) | ((val << shift) & mask); writel(tmp, pcie->base + PCIE_RGR1_SW_INIT_1(pcie)); + + return ret; } -static void brcm_pcie_bridge_sw_init_set_7278(struct brcm_pcie *pcie, u32 val) +static int brcm_pcie_bridge_sw_init_set_7278(struct brcm_pcie *pcie, u32 val) { u32 tmp, mask = RGR1_SW_INIT_1_INIT_7278_MASK; u32 shift = RGR1_SW_INIT_1_INIT_7278_SHIFT; @@ -748,20 +776,29 @@ static void brcm_pcie_bridge_sw_init_set_7278(struct brcm_pcie *pcie, u32 val) tmp = readl(pcie->base + PCIE_RGR1_SW_INIT_1(pcie)); tmp = (tmp & ~mask) | ((val << shift) & mask); writel(tmp, pcie->base + PCIE_RGR1_SW_INIT_1(pcie)); + + return 0; } -static void brcm_pcie_perst_set_4908(struct brcm_pcie *pcie, u32 val) +static int brcm_pcie_perst_set_4908(struct brcm_pcie *pcie, u32 val) { + int ret; + if (WARN_ONCE(!pcie->perst_reset, "missing PERST# reset controller\n")) - return; + return -EINVAL; if (val) - reset_control_assert(pcie->perst_reset); + ret = reset_control_assert(pcie->perst_reset); else - reset_control_deassert(pcie->perst_reset); + ret = reset_control_deassert(pcie->perst_reset); + + if (ret) + dev_err(pcie->dev, "failed to %s 'perst' reset, err=%d\n", + val ? "assert" : "deassert", ret); + return ret; } -static void brcm_pcie_perst_set_7278(struct brcm_pcie *pcie, u32 val) +static int brcm_pcie_perst_set_7278(struct brcm_pcie *pcie, u32 val) { u32 tmp; @@ -769,34 +806,77 @@ static void brcm_pcie_perst_set_7278(struct brcm_pcie *pcie, u32 val) tmp = readl(pcie->base + PCIE_MISC_PCIE_CTRL); u32p_replace_bits(&tmp, !val, PCIE_MISC_PCIE_CTRL_PCIE_PERSTB_MASK); writel(tmp, pcie->base + PCIE_MISC_PCIE_CTRL); + + return 0; } -static void brcm_pcie_perst_set_generic(struct brcm_pcie *pcie, u32 val) +static int brcm_pcie_perst_set_generic(struct brcm_pcie *pcie, u32 val) { u32 tmp; tmp = readl(pcie->base + PCIE_RGR1_SW_INIT_1(pcie)); u32p_replace_bits(&tmp, val, PCIE_RGR1_SW_INIT_1_PERST_MASK); writel(tmp, pcie->base + PCIE_RGR1_SW_INIT_1(pcie)); + + return 0; } -static int brcm_pcie_get_rc_bar2_size_and_offset(struct brcm_pcie *pcie, - u64 *rc_bar2_size, - u64 *rc_bar2_offset) +static void add_inbound_win(struct inbound_win *b, u8 *count, u64 size, + u64 cpu_addr, u64 pci_offset) +{ + b->size = size; + b->cpu_addr = cpu_addr; + b->pci_offset = pci_offset; + (*count)++; +} + +static int brcm_pcie_get_inbound_wins(struct brcm_pcie *pcie, + struct inbound_win inbound_wins[]) { struct pci_host_bridge *bridge = pci_host_bridge_from_priv(pcie); + u64 pci_offset, cpu_addr, size = 0, tot_size = 0; struct resource_entry *entry; struct device *dev = pcie->dev; u64 lowest_pcie_addr = ~(u64)0; int ret, i = 0; - u64 size = 0; + u8 n = 0; + + /* + * The HW registers (and PCIe) use order-1 numbering for BARs. As such, + * we have inbound_wins[0] unused and BAR1 starts at inbound_wins[1]. + */ + struct inbound_win *b_begin = &inbound_wins[1]; + struct inbound_win *b = b_begin; + + /* + * STB chips beside 7712 disable the first inbound window default. + * Rather being mapped to system memory it is mapped to the + * internal registers of the SoC. This feature is deprecated, has + * security considerations, and is not implemented in our modern + * SoCs. + */ + if (pcie->soc_base != BCM7712) + add_inbound_win(b++, &n, 0, 0, 0); resource_list_for_each_entry(entry, &bridge->dma_ranges) { - u64 pcie_beg = entry->res->start - entry->offset; + u64 pcie_start = entry->res->start - entry->offset; + u64 cpu_start = entry->res->start; - size += entry->res->end - entry->res->start + 1; - if (pcie_beg < lowest_pcie_addr) - lowest_pcie_addr = pcie_beg; + size = resource_size(entry->res); + tot_size += size; + if (pcie_start < lowest_pcie_addr) + lowest_pcie_addr = pcie_start; + /* + * 7712 and newer chips may have many BARs, with each + * offering a non-overlapping viewport to system memory. + * That being said, each BARs size must still be a power of + * two. + */ + if (pcie->soc_base == BCM7712) + add_inbound_win(b++, &n, size, cpu_start, pcie_start); + + if (n > pcie->num_inbound_wins) + break; } if (lowest_pcie_addr == ~(u64)0) { @@ -804,13 +884,20 @@ static int brcm_pcie_get_rc_bar2_size_and_offset(struct brcm_pcie *pcie, return -EINVAL; } + /* + * 7712 and newer chips do not have an internal memory mapping system + * that enables multiple memory controllers. As such, it can return + * now w/o doing special configuration. + */ + if (pcie->soc_base == BCM7712) + return n; + ret = of_property_read_variable_u64_array(pcie->np, "brcm,scb-sizes", pcie->memc_size, 1, PCIE_BRCM_MAX_MEMC); - if (ret <= 0) { /* Make an educated guess */ pcie->num_memc = 1; - pcie->memc_size[0] = 1ULL << fls64(size - 1); + pcie->memc_size[0] = 1ULL << fls64(tot_size - 1); } else { pcie->num_memc = ret; } @@ -819,10 +906,15 @@ static int brcm_pcie_get_rc_bar2_size_and_offset(struct brcm_pcie *pcie, for (i = 0, size = 0; i < pcie->num_memc; i++) size += pcie->memc_size[i]; - /* System memory starts at this address in PCIe-space */ - *rc_bar2_offset = lowest_pcie_addr; - /* The sum of all memc views must also be a power of 2 */ - *rc_bar2_size = 1ULL << fls64(size - 1); + /* Our HW mandates that the window size must be a power of 2 */ + size = 1ULL << fls64(size - 1); + + /* + * For STB chips, the BAR2 cpu_addr is hardwired to the start + * of system memory, so we set it to 0. + */ + cpu_addr = 0; + pci_offset = lowest_pcie_addr; /* * We validate the inbound memory view even though we should trust @@ -857,44 +949,119 @@ static int brcm_pcie_get_rc_bar2_size_and_offset(struct brcm_pcie *pcie, * outbound memory @ 3GB). So instead it will start at the 1x * multiple of its size */ - if (!*rc_bar2_size || (*rc_bar2_offset & (*rc_bar2_size - 1)) || - (*rc_bar2_offset < SZ_4G && *rc_bar2_offset > SZ_2G)) { - dev_err(dev, "Invalid rc_bar2_offset/size: size 0x%llx, off 0x%llx\n", - *rc_bar2_size, *rc_bar2_offset); + if (!size || (pci_offset & (size - 1)) || + (pci_offset < SZ_4G && pci_offset > SZ_2G)) { + dev_err(dev, "Invalid inbound_win2_offset/size: size 0x%llx, off 0x%llx\n", + size, pci_offset); return -EINVAL; } - return 0; + /* Enable inbound window 2, the main inbound window for STB chips */ + add_inbound_win(b++, &n, size, cpu_addr, pci_offset); + + /* + * Disable inbound window 3. On some chips presents the same + * window as #2 but the data appears in a settable endianness. + */ + add_inbound_win(b++, &n, 0, 0, 0); + + return n; +} + +static u32 brcm_bar_reg_offset(int bar) +{ + if (bar <= 3) + return PCIE_MISC_RC_BAR1_CONFIG_LO + 8 * (bar - 1); + else + return PCIE_MISC_RC_BAR4_CONFIG_LO + 8 * (bar - 4); +} + +static u32 brcm_ubus_reg_offset(int bar) +{ + if (bar <= 3) + return PCIE_MISC_UBUS_BAR1_CONFIG_REMAP + 8 * (bar - 1); + else + return PCIE_MISC_UBUS_BAR4_CONFIG_REMAP + 8 * (bar - 4); +} + +static void set_inbound_win_registers(struct brcm_pcie *pcie, + const struct inbound_win *inbound_wins, + u8 num_inbound_wins) +{ + void __iomem *base = pcie->base; + int i; + + for (i = 1; i <= num_inbound_wins; i++) { + u64 pci_offset = inbound_wins[i].pci_offset; + u64 cpu_addr = inbound_wins[i].cpu_addr; + u64 size = inbound_wins[i].size; + u32 reg_offset = brcm_bar_reg_offset(i); + u32 tmp = lower_32_bits(pci_offset); + + u32p_replace_bits(&tmp, brcm_pcie_encode_ibar_size(size), + PCIE_MISC_RC_BAR1_CONFIG_LO_SIZE_MASK); + + /* Write low */ + writel_relaxed(tmp, base + reg_offset); + /* Write high */ + writel_relaxed(upper_32_bits(pci_offset), base + reg_offset + 4); + + /* + * Most STB chips: + * Do nothing. + * 7712: + * All of their BARs need to be set. + */ + if (pcie->soc_base == BCM7712) { + /* BUS remap register settings */ + reg_offset = brcm_ubus_reg_offset(i); + tmp = lower_32_bits(cpu_addr) & ~0xfff; + tmp |= PCIE_MISC_UBUS_BAR1_CONFIG_REMAP_ACCESS_EN_MASK; + writel_relaxed(tmp, base + reg_offset); + tmp = upper_32_bits(cpu_addr); + writel_relaxed(tmp, base + reg_offset + 4); + } + } } static int brcm_pcie_setup(struct brcm_pcie *pcie) { - u64 rc_bar2_offset, rc_bar2_size; + struct inbound_win inbound_wins[PCIE_BRCM_MAX_INBOUND_WINS]; void __iomem *base = pcie->base; struct pci_host_bridge *bridge; struct resource_entry *entry; u32 tmp, burst, aspm_support; - int num_out_wins = 0; - int ret, memc; + u8 num_out_wins = 0; + int num_inbound_wins = 0; + int memc, ret; /* Reset the bridge */ - pcie->bridge_sw_init_set(pcie, 1); + ret = pcie->bridge_sw_init_set(pcie, 1); + if (ret) + return ret; /* Ensure that PERST# is asserted; some bootloaders may deassert it. */ - if (pcie->type == BCM2711) - pcie->perst_set(pcie, 1); + if (pcie->soc_base == BCM2711) { + ret = pcie->perst_set(pcie, 1); + if (ret) { + pcie->bridge_sw_init_set(pcie, 0); + return ret; + } + } usleep_range(100, 200); /* Take the bridge out of reset */ - pcie->bridge_sw_init_set(pcie, 0); + ret = pcie->bridge_sw_init_set(pcie, 0); + if (ret) + return ret; - tmp = readl(base + PCIE_MISC_HARD_PCIE_HARD_DEBUG); + tmp = readl(base + HARD_DEBUG(pcie)); if (is_bmips(pcie)) tmp &= ~PCIE_BMIPS_MISC_HARD_PCIE_HARD_DEBUG_SERDES_IDDQ_MASK; else tmp &= ~PCIE_MISC_HARD_PCIE_HARD_DEBUG_SERDES_IDDQ_MASK; - writel(tmp, base + PCIE_MISC_HARD_PCIE_HARD_DEBUG); + writel(tmp, base + HARD_DEBUG(pcie)); /* Wait for SerDes to be stable */ usleep_range(100, 200); @@ -905,9 +1072,9 @@ static int brcm_pcie_setup(struct brcm_pcie *pcie) */ if (is_bmips(pcie)) burst = 0x1; /* 256 bytes */ - else if (pcie->type == BCM2711) + else if (pcie->soc_base == BCM2711) burst = 0x0; /* 128 bytes */ - else if (pcie->type == BCM7278) + else if (pcie->soc_base == BCM7278) burst = 0x3; /* 512 bytes */ else burst = 0x2; /* 512 bytes */ @@ -924,17 +1091,16 @@ static int brcm_pcie_setup(struct brcm_pcie *pcie) u32p_replace_bits(&tmp, 1, PCIE_MISC_MISC_CTRL_PCIE_RCB_64B_MODE_MASK); writel(tmp, base + PCIE_MISC_MISC_CTRL); - ret = brcm_pcie_get_rc_bar2_size_and_offset(pcie, &rc_bar2_size, - &rc_bar2_offset); - if (ret) - return ret; + num_inbound_wins = brcm_pcie_get_inbound_wins(pcie, inbound_wins); + if (num_inbound_wins < 0) + return num_inbound_wins; - tmp = lower_32_bits(rc_bar2_offset); - u32p_replace_bits(&tmp, brcm_pcie_encode_ibar_size(rc_bar2_size), - PCIE_MISC_RC_BAR2_CONFIG_LO_SIZE_MASK); - writel(tmp, base + PCIE_MISC_RC_BAR2_CONFIG_LO); - writel(upper_32_bits(rc_bar2_offset), - base + PCIE_MISC_RC_BAR2_CONFIG_HI); + set_inbound_win_registers(pcie, inbound_wins, num_inbound_wins); + + if (!brcm_pcie_rc_mode(pcie)) { + dev_err(pcie->dev, "PCIe RC controller misconfigured as Endpoint\n"); + return -EINVAL; + } tmp = readl(base + PCIE_MISC_MISC_CTRL); for (memc = 0; memc < pcie->num_memc; memc++) { @@ -956,25 +1122,12 @@ static int brcm_pcie_setup(struct brcm_pcie *pcie) * 4GB or when the inbound area is smaller than 4GB (taking into * account the rounding-up we're forced to perform). */ - if (rc_bar2_offset >= SZ_4G || (rc_bar2_size + rc_bar2_offset) < SZ_4G) + if (inbound_wins[2].pci_offset >= SZ_4G || + (inbound_wins[2].size + inbound_wins[2].pci_offset) < SZ_4G) pcie->msi_target_addr = BRCM_MSI_TARGET_ADDR_LT_4GB; else pcie->msi_target_addr = BRCM_MSI_TARGET_ADDR_GT_4GB; - if (!brcm_pcie_rc_mode(pcie)) { - dev_err(pcie->dev, "PCIe RC controller misconfigured as Endpoint\n"); - return -EINVAL; - } - - /* disable the PCIe->GISB memory window (RC_BAR1) */ - tmp = readl(base + PCIE_MISC_RC_BAR1_CONFIG_LO); - tmp &= ~PCIE_MISC_RC_BAR1_CONFIG_LO_SIZE_MASK; - writel(tmp, base + PCIE_MISC_RC_BAR1_CONFIG_LO); - - /* disable the PCIe->SCB memory window (RC_BAR3) */ - tmp = readl(base + PCIE_MISC_RC_BAR3_CONFIG_LO); - tmp &= ~PCIE_MISC_RC_BAR3_CONFIG_LO_SIZE_MASK; - writel(tmp, base + PCIE_MISC_RC_BAR3_CONFIG_LO); /* Don't advertise L0s capability if 'aspm-no-l0s' */ aspm_support = PCIE_LINK_STATE_L1; @@ -1025,7 +1178,7 @@ static int brcm_pcie_setup(struct brcm_pcie *pcie) num_out_wins++; } - /* PCIe->SCB endian mode for BAR */ + /* PCIe->SCB endian mode for inbound window */ tmp = readl(base + PCIE_RC_CFG_VENDOR_VENDOR_SPECIFIC_REG1); u32p_replace_bits(&tmp, PCIE_RC_CFG_VENDOR_SPCIFIC_REG1_LITTLE_ENDIAN, PCIE_RC_CFG_VENDOR_VENDOR_SPECIFIC_REG1_ENDIAN_MODE_BAR2_MASK); @@ -1045,6 +1198,10 @@ static void brcm_extend_rbus_timeout(struct brcm_pcie *pcie) const unsigned int REG_OFFSET = PCIE_RGR1_SW_INIT_1(pcie) - 8; u32 timeout_us = 4000000; /* 4 seconds, our setting for L1SS */ + /* 7712 does not have this (RGR1) timer */ + if (pcie->soc_base == BCM7712) + return; + /* Each unit in timeout register is 1/216,000,000 seconds */ writel(216 * timeout_us, pcie->base + REG_OFFSET); } @@ -1063,7 +1220,7 @@ static void brcm_config_clkreq(struct brcm_pcie *pcie) } /* Start out assuming safe mode (both mode bits cleared) */ - clkreq_cntl = readl(pcie->base + PCIE_MISC_HARD_PCIE_HARD_DEBUG); + clkreq_cntl = readl(pcie->base + HARD_DEBUG(pcie)); clkreq_cntl &= ~PCIE_CLKREQ_MASK; if (strcmp(mode, "no-l1ss") == 0) { @@ -1106,7 +1263,7 @@ static void brcm_config_clkreq(struct brcm_pcie *pcie) dev_err(pcie->dev, err_msg); mode = "safe"; } - writel(clkreq_cntl, pcie->base + PCIE_MISC_HARD_PCIE_HARD_DEBUG); + writel(clkreq_cntl, pcie->base + HARD_DEBUG(pcie)); dev_info(pcie->dev, "clkreq-mode set to %s\n", mode); } @@ -1120,7 +1277,9 @@ static int brcm_pcie_start_link(struct brcm_pcie *pcie) int ret, i; /* Unassert the fundamental reset */ - pcie->perst_set(pcie, 0); + ret = pcie->perst_set(pcie, 0); + if (ret) + return ret; /* * Wait for 100ms after PERST# deassertion; see PCIe CEM specification @@ -1304,23 +1463,25 @@ static int brcm_phy_cntl(struct brcm_pcie *pcie, const int start) static inline int brcm_phy_start(struct brcm_pcie *pcie) { - return pcie->rescal ? brcm_phy_cntl(pcie, 1) : 0; + return pcie->has_phy ? brcm_phy_cntl(pcie, 1) : 0; } static inline int brcm_phy_stop(struct brcm_pcie *pcie) { - return pcie->rescal ? brcm_phy_cntl(pcie, 0) : 0; + return pcie->has_phy ? brcm_phy_cntl(pcie, 0) : 0; } -static void brcm_pcie_turn_off(struct brcm_pcie *pcie) +static int brcm_pcie_turn_off(struct brcm_pcie *pcie) { void __iomem *base = pcie->base; - int tmp; + int tmp, ret; if (brcm_pcie_link_up(pcie)) brcm_pcie_enter_l23(pcie); /* Assert fundamental reset */ - pcie->perst_set(pcie, 1); + ret = pcie->perst_set(pcie, 1); + if (ret) + return ret; /* Deassert request for L23 in case it was asserted */ tmp = readl(base + PCIE_MISC_PCIE_CTRL); @@ -1328,12 +1489,14 @@ static void brcm_pcie_turn_off(struct brcm_pcie *pcie) writel(tmp, base + PCIE_MISC_PCIE_CTRL); /* Turn off SerDes */ - tmp = readl(base + PCIE_MISC_HARD_PCIE_HARD_DEBUG); + tmp = readl(base + HARD_DEBUG(pcie)); u32p_replace_bits(&tmp, 1, PCIE_MISC_HARD_PCIE_HARD_DEBUG_SERDES_IDDQ_MASK); - writel(tmp, base + PCIE_MISC_HARD_PCIE_HARD_DEBUG); + writel(tmp, base + HARD_DEBUG(pcie)); /* Shutdown PCIe bridge */ - pcie->bridge_sw_init_set(pcie, 1); + ret = pcie->bridge_sw_init_set(pcie, 1); + + return ret; } static int pci_dev_may_wakeup(struct pci_dev *dev, void *data) @@ -1351,9 +1514,12 @@ static int brcm_pcie_suspend_noirq(struct device *dev) { struct brcm_pcie *pcie = dev_get_drvdata(dev); struct pci_host_bridge *bridge = pci_host_bridge_from_priv(pcie); - int ret; + int ret, rret; + + ret = brcm_pcie_turn_off(pcie); + if (ret) + return ret; - brcm_pcie_turn_off(pcie); /* * If brcm_phy_stop() returns an error, just dev_err(). If we * return the error it will cause the suspend to fail and this is a @@ -1382,7 +1548,10 @@ static int brcm_pcie_suspend_noirq(struct device *dev) pcie->sr->supplies); if (ret) { dev_err(dev, "Could not turn off regulators\n"); - reset_control_reset(pcie->rescal); + rret = reset_control_reset(pcie->rescal); + if (rret) + dev_err(dev, "failed to reset 'rascal' controller ret=%d\n", + rret); return ret; } } @@ -1397,7 +1566,7 @@ static int brcm_pcie_resume_noirq(struct device *dev) struct brcm_pcie *pcie = dev_get_drvdata(dev); void __iomem *base; u32 tmp; - int ret; + int ret, rret; base = pcie->base; ret = clk_prepare_enable(pcie->clk); @@ -1416,9 +1585,9 @@ static int brcm_pcie_resume_noirq(struct device *dev) pcie->bridge_sw_init_set(pcie, 0); /* SERDES_IDDQ = 0 */ - tmp = readl(base + PCIE_MISC_HARD_PCIE_HARD_DEBUG); + tmp = readl(base + HARD_DEBUG(pcie)); u32p_replace_bits(&tmp, 0, PCIE_MISC_HARD_PCIE_HARD_DEBUG_SERDES_IDDQ_MASK); - writel(tmp, base + PCIE_MISC_HARD_PCIE_HARD_DEBUG); + writel(tmp, base + HARD_DEBUG(pcie)); /* wait for serdes to be stable */ udelay(100); @@ -1459,7 +1628,9 @@ err_regulator: if (pcie->sr) regulator_bulk_disable(pcie->sr->num_supplies, pcie->sr->supplies); err_reset: - reset_control_rearm(pcie->rescal); + rret = reset_control_rearm(pcie->rescal); + if (rret) + dev_err(pcie->dev, "failed to rearm 'rescal' reset, err=%d\n", rret); err_disable_clk: clk_disable_unprepare(pcie->clk); return ret; @@ -1487,74 +1658,111 @@ static void brcm_pcie_remove(struct platform_device *pdev) } static const int pcie_offsets[] = { - [RGR1_SW_INIT_1] = 0x9210, - [EXT_CFG_INDEX] = 0x9000, - [EXT_CFG_DATA] = 0x9004, + [RGR1_SW_INIT_1] = 0x9210, + [EXT_CFG_INDEX] = 0x9000, + [EXT_CFG_DATA] = 0x9004, + [PCIE_HARD_DEBUG] = 0x4204, + [PCIE_INTR2_CPU_BASE] = 0x4300, }; -static const int pcie_offsets_bmips_7425[] = { - [RGR1_SW_INIT_1] = 0x8010, - [EXT_CFG_INDEX] = 0x8300, - [EXT_CFG_DATA] = 0x8304, +static const int pcie_offsets_bcm7278[] = { + [RGR1_SW_INIT_1] = 0xc010, + [EXT_CFG_INDEX] = 0x9000, + [EXT_CFG_DATA] = 0x9004, + [PCIE_HARD_DEBUG] = 0x4204, + [PCIE_INTR2_CPU_BASE] = 0x4300, +}; + +static const int pcie_offsets_bcm7425[] = { + [RGR1_SW_INIT_1] = 0x8010, + [EXT_CFG_INDEX] = 0x8300, + [EXT_CFG_DATA] = 0x8304, + [PCIE_HARD_DEBUG] = 0x4204, + [PCIE_INTR2_CPU_BASE] = 0x4300, +}; + +static const int pcie_offsets_bcm7712[] = { + [EXT_CFG_INDEX] = 0x9000, + [EXT_CFG_DATA] = 0x9004, + [PCIE_HARD_DEBUG] = 0x4304, + [PCIE_INTR2_CPU_BASE] = 0x4400, }; static const struct pcie_cfg_data generic_cfg = { .offsets = pcie_offsets, - .type = GENERIC, + .soc_base = GENERIC, .perst_set = brcm_pcie_perst_set_generic, .bridge_sw_init_set = brcm_pcie_bridge_sw_init_set_generic, -}; - -static const struct pcie_cfg_data bcm7425_cfg = { - .offsets = pcie_offsets_bmips_7425, - .type = BCM7425, - .perst_set = brcm_pcie_perst_set_generic, - .bridge_sw_init_set = brcm_pcie_bridge_sw_init_set_generic, -}; - -static const struct pcie_cfg_data bcm7435_cfg = { - .offsets = pcie_offsets, - .type = BCM7435, - .perst_set = brcm_pcie_perst_set_generic, - .bridge_sw_init_set = brcm_pcie_bridge_sw_init_set_generic, -}; - -static const struct pcie_cfg_data bcm4908_cfg = { - .offsets = pcie_offsets, - .type = BCM4908, - .perst_set = brcm_pcie_perst_set_4908, - .bridge_sw_init_set = brcm_pcie_bridge_sw_init_set_generic, -}; - -static const int pcie_offset_bcm7278[] = { - [RGR1_SW_INIT_1] = 0xc010, - [EXT_CFG_INDEX] = 0x9000, - [EXT_CFG_DATA] = 0x9004, -}; - -static const struct pcie_cfg_data bcm7278_cfg = { - .offsets = pcie_offset_bcm7278, - .type = BCM7278, - .perst_set = brcm_pcie_perst_set_7278, - .bridge_sw_init_set = brcm_pcie_bridge_sw_init_set_7278, + .num_inbound_wins = 3, }; static const struct pcie_cfg_data bcm2711_cfg = { .offsets = pcie_offsets, - .type = BCM2711, + .soc_base = BCM2711, .perst_set = brcm_pcie_perst_set_generic, .bridge_sw_init_set = brcm_pcie_bridge_sw_init_set_generic, + .num_inbound_wins = 3, +}; + +static const struct pcie_cfg_data bcm4908_cfg = { + .offsets = pcie_offsets, + .soc_base = BCM4908, + .perst_set = brcm_pcie_perst_set_4908, + .bridge_sw_init_set = brcm_pcie_bridge_sw_init_set_generic, + .num_inbound_wins = 3, +}; + +static const struct pcie_cfg_data bcm7278_cfg = { + .offsets = pcie_offsets_bcm7278, + .soc_base = BCM7278, + .perst_set = brcm_pcie_perst_set_7278, + .bridge_sw_init_set = brcm_pcie_bridge_sw_init_set_7278, + .num_inbound_wins = 3, +}; + +static const struct pcie_cfg_data bcm7425_cfg = { + .offsets = pcie_offsets_bcm7425, + .soc_base = BCM7425, + .perst_set = brcm_pcie_perst_set_generic, + .bridge_sw_init_set = brcm_pcie_bridge_sw_init_set_generic, + .num_inbound_wins = 3, +}; + +static const struct pcie_cfg_data bcm7435_cfg = { + .offsets = pcie_offsets, + .soc_base = BCM7435, + .perst_set = brcm_pcie_perst_set_generic, + .bridge_sw_init_set = brcm_pcie_bridge_sw_init_set_generic, + .num_inbound_wins = 3, +}; + +static const struct pcie_cfg_data bcm7216_cfg = { + .offsets = pcie_offsets_bcm7278, + .soc_base = BCM7278, + .perst_set = brcm_pcie_perst_set_7278, + .bridge_sw_init_set = brcm_pcie_bridge_sw_init_set_7278, + .has_phy = true, + .num_inbound_wins = 3, +}; + +static const struct pcie_cfg_data bcm7712_cfg = { + .offsets = pcie_offsets_bcm7712, + .perst_set = brcm_pcie_perst_set_7278, + .bridge_sw_init_set = brcm_pcie_bridge_sw_init_set_generic, + .soc_base = BCM7712, + .num_inbound_wins = 10, }; static const struct of_device_id brcm_pcie_match[] = { { .compatible = "brcm,bcm2711-pcie", .data = &bcm2711_cfg }, { .compatible = "brcm,bcm4908-pcie", .data = &bcm4908_cfg }, { .compatible = "brcm,bcm7211-pcie", .data = &generic_cfg }, + { .compatible = "brcm,bcm7216-pcie", .data = &bcm7216_cfg }, { .compatible = "brcm,bcm7278-pcie", .data = &bcm7278_cfg }, - { .compatible = "brcm,bcm7216-pcie", .data = &bcm7278_cfg }, - { .compatible = "brcm,bcm7445-pcie", .data = &generic_cfg }, - { .compatible = "brcm,bcm7435-pcie", .data = &bcm7435_cfg }, { .compatible = "brcm,bcm7425-pcie", .data = &bcm7425_cfg }, + { .compatible = "brcm,bcm7435-pcie", .data = &bcm7435_cfg }, + { .compatible = "brcm,bcm7445-pcie", .data = &generic_cfg }, + { .compatible = "brcm,bcm7712-pcie", .data = &bcm7712_cfg }, {}, }; @@ -1596,9 +1804,11 @@ static int brcm_pcie_probe(struct platform_device *pdev) pcie->dev = &pdev->dev; pcie->np = np; pcie->reg_offsets = data->offsets; - pcie->type = data->type; + pcie->soc_base = data->soc_base; pcie->perst_set = data->perst_set; pcie->bridge_sw_init_set = data->bridge_sw_init_set; + pcie->has_phy = data->has_phy; + pcie->num_inbound_wins = data->num_inbound_wins; pcie->base = devm_platform_ioremap_resource(pdev, 0); if (IS_ERR(pcie->base)) @@ -1613,25 +1823,52 @@ static int brcm_pcie_probe(struct platform_device *pdev) pcie->ssc = of_property_read_bool(np, "brcm,enable-ssc"); - ret = clk_prepare_enable(pcie->clk); - if (ret) { - dev_err(&pdev->dev, "could not enable clock\n"); - return ret; - } pcie->rescal = devm_reset_control_get_optional_shared(&pdev->dev, "rescal"); - if (IS_ERR(pcie->rescal)) { - clk_disable_unprepare(pcie->clk); + if (IS_ERR(pcie->rescal)) return PTR_ERR(pcie->rescal); - } + pcie->perst_reset = devm_reset_control_get_optional_exclusive(&pdev->dev, "perst"); - if (IS_ERR(pcie->perst_reset)) { - clk_disable_unprepare(pcie->clk); + if (IS_ERR(pcie->perst_reset)) return PTR_ERR(pcie->perst_reset); + + pcie->bridge_reset = devm_reset_control_get_optional_exclusive(&pdev->dev, "bridge"); + if (IS_ERR(pcie->bridge_reset)) + return PTR_ERR(pcie->bridge_reset); + + pcie->swinit_reset = devm_reset_control_get_optional_exclusive(&pdev->dev, "swinit"); + if (IS_ERR(pcie->swinit_reset)) + return PTR_ERR(pcie->swinit_reset); + + ret = clk_prepare_enable(pcie->clk); + if (ret) + return dev_err_probe(&pdev->dev, ret, "could not enable clock\n"); + + pcie->bridge_sw_init_set(pcie, 0); + + if (pcie->swinit_reset) { + ret = reset_control_assert(pcie->swinit_reset); + if (ret) { + clk_disable_unprepare(pcie->clk); + return dev_err_probe(&pdev->dev, ret, + "could not assert reset 'swinit'\n"); + } + + /* HW team recommends 1us for proper sync and propagation of reset */ + udelay(1); + + ret = reset_control_deassert(pcie->swinit_reset); + if (ret) { + clk_disable_unprepare(pcie->clk); + return dev_err_probe(&pdev->dev, ret, + "could not de-assert reset 'swinit'\n"); + } } ret = reset_control_reset(pcie->rescal); - if (ret) - dev_err(&pdev->dev, "failed to deassert 'rescal'\n"); + if (ret) { + clk_disable_unprepare(pcie->clk); + return dev_err_probe(&pdev->dev, ret, "failed to deassert 'rescal'\n"); + } ret = brcm_phy_start(pcie); if (ret) { @@ -1645,7 +1882,7 @@ static int brcm_pcie_probe(struct platform_device *pdev) goto fail; pcie->hw_rev = readl(pcie->base + PCIE_MISC_REVISION); - if (pcie->type == BCM4908 && pcie->hw_rev >= BRCM_PCIE_HW_REV_3_20) { + if (pcie->soc_base == BCM4908 && pcie->hw_rev >= BRCM_PCIE_HW_REV_3_20) { dev_err(pcie->dev, "hardware revision with unsupported PERST# setup\n"); ret = -ENODEV; goto fail; @@ -1660,7 +1897,7 @@ static int brcm_pcie_probe(struct platform_device *pdev) } } - bridge->ops = pcie->type == BCM7425 ? &brcm7425_pcie_ops : &brcm_pcie_ops; + bridge->ops = pcie->soc_base == BCM7425 ? &brcm7425_pcie_ops : &brcm_pcie_ops; bridge->sysdata = pcie; platform_set_drvdata(pdev, pcie); @@ -1678,6 +1915,7 @@ static int brcm_pcie_probe(struct platform_device *pdev) fail: __brcm_pcie_remove(pcie); + return ret; } diff --git a/drivers/pci/controller/pcie-iproc.c b/drivers/pci/controller/pcie-iproc.c index 97f739a2c9f8..22134e95574b 100644 --- a/drivers/pci/controller/pcie-iproc.c +++ b/drivers/pci/controller/pcie-iproc.c @@ -54,7 +54,7 @@ #define CFG_RD_SUCCESS 0 #define CFG_RD_UR 1 -#define CFG_RD_CRS 2 +#define CFG_RD_RRS 2 #define CFG_RD_CA 3 #define CFG_RETRY_STATUS 0xffff0001 #define CFG_RETRY_STATUS_TIMEOUT_US 500000 /* 500 milliseconds */ @@ -485,31 +485,31 @@ static unsigned int iproc_pcie_cfg_retry(struct iproc_pcie *pcie, u32 status; /* - * As per PCIe spec r3.1, sec 2.3.2, CRS Software Visibility only + * As per PCIe r6.0, sec 2.3.2, Config RRS Software Visibility only * affects config reads of the Vendor ID. For config writes or any * other config reads, the Root may automatically reissue the * configuration request again as a new request. * * For config reads, this hardware returns CFG_RETRY_STATUS data - * when it receives a CRS completion, regardless of the address of - * the read or the CRS Software Visibility Enable bit. As a + * when it receives a RRS completion, regardless of the address of + * the read or the RRS Software Visibility Enable bit. As a * partial workaround for this, we retry in software any read that * returns CFG_RETRY_STATUS. * * Note that a non-Vendor ID config register may have a value of * CFG_RETRY_STATUS. If we read that, we can't distinguish it from - * a CRS completion, so we will incorrectly retry the read and + * a RRS completion, so we will incorrectly retry the read and * eventually return the wrong data (0xffffffff). */ data = readl(cfg_data_p); while (data == CFG_RETRY_STATUS && timeout--) { /* - * CRS state is set in CFG_RD status register + * RRS state is set in CFG_RD status register * This will handle the case where CFG_RETRY_STATUS is * valid config data. */ status = iproc_pcie_read_reg(pcie, IPROC_PCIE_CFG_RD_STATUS); - if (status != CFG_RD_CRS) + if (status != CFG_RD_RRS) return data; udelay(1); @@ -556,8 +556,8 @@ static void iproc_pcie_fix_cap(struct iproc_pcie *pcie, int where, u32 *val) break; case IPROC_PCI_EXP_CAP + PCI_EXP_RTCTL: - /* Don't advertise CRS SV support */ - *val &= ~(PCI_EXP_RTCAP_CRSVIS << 16); + /* Don't advertise RRS SV support */ + *val &= ~(PCI_EXP_RTCAP_RRS_SV << 16); break; default: diff --git a/drivers/pci/controller/pcie-mediatek-gen3.c b/drivers/pci/controller/pcie-mediatek-gen3.c index b7e8e24f6a40..66ce4b5d309b 100644 --- a/drivers/pci/controller/pcie-mediatek-gen3.c +++ b/drivers/pci/controller/pcie-mediatek-gen3.c @@ -6,7 +6,9 @@ * Author: Jianjun Wang */ +#include #include +#include #include #include #include @@ -15,6 +17,8 @@ #include #include #include +#include +#include #include #include #include @@ -29,6 +33,12 @@ #define PCI_CLASS(class) (class << 8) #define PCIE_RC_MODE BIT(0) +#define PCIE_EQ_PRESET_01_REG 0x100 +#define PCIE_VAL_LN0_DOWNSTREAM GENMASK(6, 0) +#define PCIE_VAL_LN0_UPSTREAM GENMASK(14, 8) +#define PCIE_VAL_LN1_DOWNSTREAM GENMASK(22, 16) +#define PCIE_VAL_LN1_UPSTREAM GENMASK(30, 24) + #define PCIE_CFGNUM_REG 0x140 #define PCIE_CFG_DEVFN(devfn) ((devfn) & GENMASK(7, 0)) #define PCIE_CFG_BUS(bus) (((bus) << 8) & GENMASK(15, 8)) @@ -68,6 +78,14 @@ #define PCIE_MSI_SET_ENABLE_REG 0x190 #define PCIE_MSI_SET_ENABLE GENMASK(PCIE_MSI_SET_NUM - 1, 0) +#define PCIE_PIPE4_PIE8_REG 0x338 +#define PCIE_K_FINETUNE_MAX GENMASK(5, 0) +#define PCIE_K_FINETUNE_ERR GENMASK(7, 6) +#define PCIE_K_PRESET_TO_USE GENMASK(18, 8) +#define PCIE_K_PHYPARAM_QUERY BIT(19) +#define PCIE_K_QUERY_TIMEOUT BIT(20) +#define PCIE_K_PRESET_TO_USE_16G GENMASK(31, 21) + #define PCIE_MSI_SET_BASE_REG 0xc00 #define PCIE_MSI_SET_OFFSET 0x10 #define PCIE_MSI_SET_STATUS_OFFSET 0x04 @@ -100,6 +118,26 @@ #define PCIE_ATR_TLP_TYPE_MEM PCIE_ATR_TLP_TYPE(0) #define PCIE_ATR_TLP_TYPE_IO PCIE_ATR_TLP_TYPE(2) +#define MAX_NUM_PHY_RESETS 3 + +/* Time in ms needed to complete PCIe reset on EN7581 SoC */ +#define PCIE_EN7581_RESET_TIME_MS 100 + +struct mtk_gen3_pcie; + +/** + * struct mtk_gen3_pcie_pdata - differentiate between host generations + * @power_up: pcie power_up callback + * @phy_resets: phy reset lines SoC data. + */ +struct mtk_gen3_pcie_pdata { + int (*power_up)(struct mtk_gen3_pcie *pcie); + struct { + const char *id[MAX_NUM_PHY_RESETS]; + int num_resets; + } phy_resets; +}; + /** * struct mtk_msi_set - MSI information for each set * @base: IO mapped register base @@ -118,7 +156,7 @@ struct mtk_msi_set { * @base: IO mapped register base * @reg_base: physical register base * @mac_reset: MAC reset control - * @phy_reset: PHY reset control + * @phy_resets: PHY reset controllers * @phy: PHY controller block * @clks: PCIe clocks * @num_clks: PCIe clocks count for this port @@ -131,13 +169,14 @@ struct mtk_msi_set { * @msi_sets: MSI sets information * @lock: lock protecting IRQ bit map * @msi_irq_in_use: bit map for assigned MSI IRQ + * @soc: pointer to SoC-dependent operations */ struct mtk_gen3_pcie { struct device *dev; void __iomem *base; phys_addr_t reg_base; struct reset_control *mac_reset; - struct reset_control *phy_reset; + struct reset_control_bulk_data phy_resets[MAX_NUM_PHY_RESETS]; struct phy *phy; struct clk_bulk_data *clks; int num_clks; @@ -151,6 +190,8 @@ struct mtk_gen3_pcie { struct mtk_msi_set msi_sets[PCIE_MSI_SET_NUM]; struct mutex lock; DECLARE_BITMAP(msi_irq_in_use, PCIE_MSI_IRQS_NUM); + + const struct mtk_gen3_pcie_pdata *soc; }; /* LTSSM state in PCIE_LTSSM_STATUS_REG bit[28:24] */ @@ -424,12 +465,6 @@ static int mtk_pcie_startup_port(struct mtk_gen3_pcie *pcie) return 0; } -static int mtk_pcie_set_affinity(struct irq_data *data, - const struct cpumask *mask, bool force) -{ - return -EINVAL; -} - static void mtk_pcie_msi_irq_mask(struct irq_data *data) { pci_msi_mask_irq(data); @@ -450,8 +485,9 @@ static struct irq_chip mtk_msi_irq_chip = { }; static struct msi_domain_info mtk_msi_domain_info = { - .flags = (MSI_FLAG_USE_DEF_DOM_OPS | MSI_FLAG_USE_DEF_CHIP_OPS | - MSI_FLAG_PCI_MSIX | MSI_FLAG_MULTI_PCI_MSI), + .flags = MSI_FLAG_USE_DEF_DOM_OPS | MSI_FLAG_USE_DEF_CHIP_OPS | + MSI_FLAG_NO_AFFINITY | MSI_FLAG_PCI_MSIX | + MSI_FLAG_MULTI_PCI_MSI, .chip = &mtk_msi_irq_chip, }; @@ -517,7 +553,6 @@ static struct irq_chip mtk_msi_bottom_irq_chip = { .irq_mask = mtk_msi_bottom_irq_mask, .irq_unmask = mtk_msi_bottom_irq_unmask, .irq_compose_msi_msg = mtk_compose_msi_msg, - .irq_set_affinity = mtk_pcie_set_affinity, .name = "MSI", }; @@ -618,7 +653,6 @@ static struct irq_chip mtk_intx_irq_chip = { .irq_mask = mtk_intx_mask, .irq_unmask = mtk_intx_unmask, .irq_eoi = mtk_intx_eoi, - .irq_set_affinity = mtk_pcie_set_affinity, .name = "INTx", }; @@ -775,10 +809,10 @@ static int mtk_pcie_setup_irq(struct mtk_gen3_pcie *pcie) static int mtk_pcie_parse_port(struct mtk_gen3_pcie *pcie) { + int i, ret, num_resets = pcie->soc->phy_resets.num_resets; struct device *dev = pcie->dev; struct platform_device *pdev = to_platform_device(dev); struct resource *regs; - int ret; regs = platform_get_resource_byname(pdev, IORESOURCE_MEM, "pcie-mac"); if (!regs) @@ -791,12 +825,12 @@ static int mtk_pcie_parse_port(struct mtk_gen3_pcie *pcie) pcie->reg_base = regs->start; - pcie->phy_reset = devm_reset_control_get_optional_exclusive(dev, "phy"); - if (IS_ERR(pcie->phy_reset)) { - ret = PTR_ERR(pcie->phy_reset); - if (ret != -EPROBE_DEFER) - dev_err(dev, "failed to get PHY reset\n"); + for (i = 0; i < num_resets; i++) + pcie->phy_resets[i].id = pcie->soc->phy_resets.id[i]; + ret = devm_reset_control_bulk_get_optional_shared(dev, num_resets, pcie->phy_resets); + if (ret) { + dev_err(dev, "failed to get PHY bulk reset\n"); return ret; } @@ -827,13 +861,96 @@ static int mtk_pcie_parse_port(struct mtk_gen3_pcie *pcie) return 0; } +static int mtk_pcie_en7581_power_up(struct mtk_gen3_pcie *pcie) +{ + struct device *dev = pcie->dev; + int err; + u32 val; + + /* + * Wait for the time needed to complete the bulk assert in + * mtk_pcie_setup for EN7581 SoC. + */ + mdelay(PCIE_EN7581_RESET_TIME_MS); + + err = phy_init(pcie->phy); + if (err) { + dev_err(dev, "failed to initialize PHY\n"); + return err; + } + + err = phy_power_on(pcie->phy); + if (err) { + dev_err(dev, "failed to power on PHY\n"); + goto err_phy_on; + } + + err = reset_control_bulk_deassert(pcie->soc->phy_resets.num_resets, pcie->phy_resets); + if (err) { + dev_err(dev, "failed to deassert PHYs\n"); + goto err_phy_deassert; + } + + /* + * Wait for the time needed to complete the bulk de-assert above. + * This time is specific for EN7581 SoC. + */ + mdelay(PCIE_EN7581_RESET_TIME_MS); + + pm_runtime_enable(dev); + pm_runtime_get_sync(dev); + + err = clk_bulk_prepare(pcie->num_clks, pcie->clks); + if (err) { + dev_err(dev, "failed to prepare clock\n"); + goto err_clk_prepare; + } + + val = FIELD_PREP(PCIE_VAL_LN0_DOWNSTREAM, 0x47) | + FIELD_PREP(PCIE_VAL_LN1_DOWNSTREAM, 0x47) | + FIELD_PREP(PCIE_VAL_LN0_UPSTREAM, 0x41) | + FIELD_PREP(PCIE_VAL_LN1_UPSTREAM, 0x41); + writel_relaxed(val, pcie->base + PCIE_EQ_PRESET_01_REG); + + val = PCIE_K_PHYPARAM_QUERY | PCIE_K_QUERY_TIMEOUT | + FIELD_PREP(PCIE_K_PRESET_TO_USE_16G, 0x80) | + FIELD_PREP(PCIE_K_PRESET_TO_USE, 0x2) | + FIELD_PREP(PCIE_K_FINETUNE_MAX, 0xf); + writel_relaxed(val, pcie->base + PCIE_PIPE4_PIE8_REG); + + err = clk_bulk_enable(pcie->num_clks, pcie->clks); + if (err) { + dev_err(dev, "failed to prepare clock\n"); + goto err_clk_enable; + } + + return 0; + +err_clk_enable: + clk_bulk_unprepare(pcie->num_clks, pcie->clks); +err_clk_prepare: + pm_runtime_put_sync(dev); + pm_runtime_disable(dev); + reset_control_bulk_assert(pcie->soc->phy_resets.num_resets, pcie->phy_resets); +err_phy_deassert: + phy_power_off(pcie->phy); +err_phy_on: + phy_exit(pcie->phy); + + return err; +} + static int mtk_pcie_power_up(struct mtk_gen3_pcie *pcie) { struct device *dev = pcie->dev; int err; /* PHY power on and enable pipe clock */ - reset_control_deassert(pcie->phy_reset); + err = reset_control_bulk_deassert(pcie->soc->phy_resets.num_resets, pcie->phy_resets); + if (err) { + dev_err(dev, "failed to deassert PHYs\n"); + return err; + } err = phy_init(pcie->phy); if (err) { @@ -869,7 +986,7 @@ err_clk_init: err_phy_on: phy_exit(pcie->phy); err_phy_init: - reset_control_assert(pcie->phy_reset); + reset_control_bulk_assert(pcie->soc->phy_resets.num_resets, pcie->phy_resets); return err; } @@ -884,7 +1001,7 @@ static void mtk_pcie_power_down(struct mtk_gen3_pcie *pcie) phy_power_off(pcie->phy); phy_exit(pcie->phy); - reset_control_assert(pcie->phy_reset); + reset_control_bulk_assert(pcie->soc->phy_resets.num_resets, pcie->phy_resets); } static int mtk_pcie_setup(struct mtk_gen3_pcie *pcie) @@ -895,16 +1012,22 @@ static int mtk_pcie_setup(struct mtk_gen3_pcie *pcie) if (err) return err; + /* + * Deassert the line in order to avoid unbalance in deassert_count + * counter since the bulk is shared. + */ + reset_control_bulk_deassert(pcie->soc->phy_resets.num_resets, pcie->phy_resets); /* * The controller may have been left out of reset by the bootloader * so make sure that we get a clean start by asserting resets here. */ - reset_control_assert(pcie->phy_reset); + reset_control_bulk_assert(pcie->soc->phy_resets.num_resets, pcie->phy_resets); + reset_control_assert(pcie->mac_reset); usleep_range(10, 20); /* Don't touch the hardware registers before power up */ - err = mtk_pcie_power_up(pcie); + err = pcie->soc->power_up(pcie); if (err) return err; @@ -939,6 +1062,7 @@ static int mtk_pcie_probe(struct platform_device *pdev) pcie = pci_host_bridge_priv(host); pcie->dev = dev; + pcie->soc = device_get_match_data(dev); platform_set_drvdata(pdev, pcie); err = mtk_pcie_setup(pcie); @@ -1054,7 +1178,7 @@ static int mtk_pcie_resume_noirq(struct device *dev) struct mtk_gen3_pcie *pcie = dev_get_drvdata(dev); int err; - err = mtk_pcie_power_up(pcie); + err = pcie->soc->power_up(pcie); if (err) return err; @@ -1074,8 +1198,27 @@ static const struct dev_pm_ops mtk_pcie_pm_ops = { mtk_pcie_resume_noirq) }; +static const struct mtk_gen3_pcie_pdata mtk_pcie_soc_mt8192 = { + .power_up = mtk_pcie_power_up, + .phy_resets = { + .id[0] = "phy", + .num_resets = 1, + }, +}; + +static const struct mtk_gen3_pcie_pdata mtk_pcie_soc_en7581 = { + .power_up = mtk_pcie_en7581_power_up, + .phy_resets = { + .id[0] = "phy-lane0", + .id[1] = "phy-lane1", + .id[2] = "phy-lane2", + .num_resets = 3, + }, +}; + static const struct of_device_id mtk_pcie_of_match[] = { - { .compatible = "mediatek,mt8192-pcie" }, + { .compatible = "airoha,en7581-pcie", .data = &mtk_pcie_soc_en7581 }, + { .compatible = "mediatek,mt8192-pcie", .data = &mtk_pcie_soc_mt8192 }, {}, }; MODULE_DEVICE_TABLE(of, mtk_pcie_of_match); diff --git a/drivers/pci/controller/pcie-mediatek.c b/drivers/pci/controller/pcie-mediatek.c index 7fc0d7709b7f..7f7d04c2ea57 100644 --- a/drivers/pci/controller/pcie-mediatek.c +++ b/drivers/pci/controller/pcie-mediatek.c @@ -211,7 +211,6 @@ struct mtk_pcie_port { * @base: IO mapped register base * @cfg: IO mapped register map for PCIe config * @free_ck: free-run reference clock - * @mem: non-prefetchable memory resource * @ports: pointer to PCIe port information * @soc: pointer to SoC-dependent operations */ @@ -407,12 +406,6 @@ static void mtk_compose_msi_msg(struct irq_data *data, struct msi_msg *msg) (int)data->hwirq, msg->address_hi, msg->address_lo); } -static int mtk_msi_set_affinity(struct irq_data *irq_data, - const struct cpumask *mask, bool force) -{ - return -EINVAL; -} - static void mtk_msi_ack_irq(struct irq_data *data) { struct mtk_pcie_port *port = irq_data_get_irq_chip_data(data); @@ -424,7 +417,6 @@ static void mtk_msi_ack_irq(struct irq_data *data) static struct irq_chip mtk_msi_bottom_irq_chip = { .name = "MTK MSI", .irq_compose_msi_msg = mtk_compose_msi_msg, - .irq_set_affinity = mtk_msi_set_affinity, .irq_ack = mtk_msi_ack_irq, }; @@ -486,8 +478,8 @@ static struct irq_chip mtk_msi_irq_chip = { }; static struct msi_domain_info mtk_msi_domain_info = { - .flags = (MSI_FLAG_USE_DEF_DOM_OPS | MSI_FLAG_USE_DEF_CHIP_OPS | - MSI_FLAG_PCI_MSIX), + .flags = MSI_FLAG_USE_DEF_DOM_OPS | MSI_FLAG_USE_DEF_CHIP_OPS | + MSI_FLAG_NO_AFFINITY | MSI_FLAG_PCI_MSIX, .chip = &mtk_msi_irq_chip, }; diff --git a/drivers/pci/controller/pcie-rcar-host.c b/drivers/pci/controller/pcie-rcar-host.c index c01efc6ea64f..3dd653f3d784 100644 --- a/drivers/pci/controller/pcie-rcar-host.c +++ b/drivers/pci/controller/pcie-rcar-host.c @@ -658,11 +658,6 @@ static void rcar_msi_irq_unmask(struct irq_data *d) spin_unlock_irqrestore(&msi->mask_lock, flags); } -static int rcar_msi_set_affinity(struct irq_data *d, const struct cpumask *mask, bool force) -{ - return -EINVAL; -} - static void rcar_compose_msi_msg(struct irq_data *data, struct msi_msg *msg) { struct rcar_msi *msi = irq_data_get_irq_chip_data(data); @@ -678,7 +673,6 @@ static struct irq_chip rcar_msi_bottom_chip = { .irq_ack = rcar_msi_irq_ack, .irq_mask = rcar_msi_irq_mask, .irq_unmask = rcar_msi_irq_unmask, - .irq_set_affinity = rcar_msi_set_affinity, .irq_compose_msi_msg = rcar_compose_msi_msg, }; @@ -725,8 +719,8 @@ static const struct irq_domain_ops rcar_msi_domain_ops = { }; static struct msi_domain_info rcar_msi_info = { - .flags = (MSI_FLAG_USE_DEF_DOM_OPS | MSI_FLAG_USE_DEF_CHIP_OPS | - MSI_FLAG_MULTI_PCI_MSI), + .flags = MSI_FLAG_USE_DEF_DOM_OPS | MSI_FLAG_USE_DEF_CHIP_OPS | + MSI_FLAG_NO_AFFINITY | MSI_FLAG_MULTI_PCI_MSI, .chip = &rcar_msi_top_chip, }; diff --git a/drivers/pci/controller/pcie-xilinx-dma-pl.c b/drivers/pci/controller/pcie-xilinx-dma-pl.c index 5be5dfd8398f..dd117f07fc95 100644 --- a/drivers/pci/controller/pcie-xilinx-dma-pl.c +++ b/drivers/pci/controller/pcie-xilinx-dma-pl.c @@ -71,10 +71,24 @@ /* Phy Status/Control Register definitions */ #define XILINX_PCIE_DMA_REG_PSCR_LNKUP BIT(11) +#define QDMA_BRIDGE_BASE_OFF 0xcd8 /* Number of MSI IRQs */ #define XILINX_NUM_MSI_IRQS 64 +enum xilinx_pl_dma_version { + XDMA, + QDMA, +}; + +/** + * struct xilinx_pl_dma_variant - PL DMA PCIe variant information + * @version: DMA version + */ +struct xilinx_pl_dma_variant { + enum xilinx_pl_dma_version version; +}; + struct xilinx_msi { struct irq_domain *msi_domain; unsigned long *bitmap; @@ -88,6 +102,7 @@ struct xilinx_msi { * struct pl_dma_pcie - PCIe port information * @dev: Device pointer * @reg_base: IO Mapped Register Base + * @cfg_base: IO Mapped Configuration Base * @irq: Interrupt number * @cfg: Holds mappings of config space window * @phys_reg_base: Physical address of reg base @@ -97,10 +112,12 @@ struct xilinx_msi { * @msi: MSI information * @intx_irq: INTx error interrupt number * @lock: Lock protecting shared register access + * @variant: PL DMA PCIe version check pointer */ struct pl_dma_pcie { struct device *dev; void __iomem *reg_base; + void __iomem *cfg_base; int irq; struct pci_config_window *cfg; phys_addr_t phys_reg_base; @@ -110,16 +127,23 @@ struct pl_dma_pcie { struct xilinx_msi msi; int intx_irq; raw_spinlock_t lock; + const struct xilinx_pl_dma_variant *variant; }; static inline u32 pcie_read(struct pl_dma_pcie *port, u32 reg) { + if (port->variant->version == QDMA) + return readl(port->reg_base + reg + QDMA_BRIDGE_BASE_OFF); + return readl(port->reg_base + reg); } static inline void pcie_write(struct pl_dma_pcie *port, u32 val, u32 reg) { - writel(val, port->reg_base + reg); + if (port->variant->version == QDMA) + writel(val, port->reg_base + reg + QDMA_BRIDGE_BASE_OFF); + else + writel(val, port->reg_base + reg); } static inline bool xilinx_pl_dma_pcie_link_up(struct pl_dma_pcie *port) @@ -173,6 +197,9 @@ static void __iomem *xilinx_pl_dma_pcie_map_bus(struct pci_bus *bus, if (!xilinx_pl_dma_pcie_valid_device(bus, devfn)) return NULL; + if (port->variant->version == QDMA) + return port->cfg_base + PCIE_ECAM_OFFSET(bus->number, devfn, where); + return port->reg_base + PCIE_ECAM_OFFSET(bus->number, devfn, where); } @@ -355,8 +382,8 @@ static struct irq_chip xilinx_msi_irq_chip = { }; static struct msi_domain_info xilinx_msi_domain_info = { - .flags = (MSI_FLAG_USE_DEF_DOM_OPS | MSI_FLAG_USE_DEF_CHIP_OPS | - MSI_FLAG_MULTI_PCI_MSI), + .flags = MSI_FLAG_USE_DEF_DOM_OPS | MSI_FLAG_USE_DEF_CHIP_OPS | + MSI_FLAG_NO_AFFINITY | MSI_FLAG_MULTI_PCI_MSI, .chip = &xilinx_msi_irq_chip, }; @@ -370,16 +397,9 @@ static void xilinx_compose_msi_msg(struct irq_data *data, struct msi_msg *msg) msg->data = data->hwirq; } -static int xilinx_msi_set_affinity(struct irq_data *irq_data, - const struct cpumask *mask, bool force) -{ - return -EINVAL; -} - static struct irq_chip xilinx_irq_chip = { .name = "pl_dma:MSI", .irq_compose_msi_msg = xilinx_compose_msi_msg, - .irq_set_affinity = xilinx_msi_set_affinity, }; static int xilinx_irq_domain_alloc(struct irq_domain *domain, unsigned int virq, @@ -731,6 +751,15 @@ static int xilinx_pl_dma_pcie_parse_dt(struct pl_dma_pcie *port, port->reg_base = port->cfg->win; + if (port->variant->version == QDMA) { + port->cfg_base = port->cfg->win; + res = platform_get_resource_byname(pdev, IORESOURCE_MEM, "breg"); + port->reg_base = devm_ioremap_resource(dev, res); + if (IS_ERR(port->reg_base)) + return PTR_ERR(port->reg_base); + port->phys_reg_base = res->start; + } + err = xilinx_request_msi_irq(port); if (err) { pci_ecam_free(port->cfg); @@ -760,6 +789,8 @@ static int xilinx_pl_dma_pcie_probe(struct platform_device *pdev) if (!bus) return -ENODEV; + port->variant = of_device_get_match_data(dev); + err = xilinx_pl_dma_pcie_parse_dt(port, bus->res); if (err) { dev_err(dev, "Parsing DT failed\n"); @@ -791,9 +822,22 @@ err_irq_domain: return err; } +static const struct xilinx_pl_dma_variant xdma_host = { + .version = XDMA, +}; + +static const struct xilinx_pl_dma_variant qdma_host = { + .version = QDMA, +}; + static const struct of_device_id xilinx_pl_dma_pcie_of_match[] = { { .compatible = "xlnx,xdma-host-3.00", + .data = &xdma_host, + }, + { + .compatible = "xlnx,qdma-host-3.00", + .data = &qdma_host, }, {} }; diff --git a/drivers/pci/controller/pcie-xilinx-nwl.c b/drivers/pci/controller/pcie-xilinx-nwl.c index 0408f4d612b5..a8ae14474dd0 100644 --- a/drivers/pci/controller/pcie-xilinx-nwl.c +++ b/drivers/pci/controller/pcie-xilinx-nwl.c @@ -19,6 +19,7 @@ #include #include #include +#include #include #include @@ -80,8 +81,8 @@ #define MSGF_MISC_SR_NON_FATAL_DEV BIT(22) #define MSGF_MISC_SR_FATAL_DEV BIT(23) #define MSGF_MISC_SR_LINK_DOWN BIT(24) -#define MSGF_MSIC_SR_LINK_AUTO_BWIDTH BIT(25) -#define MSGF_MSIC_SR_LINK_BWIDTH BIT(26) +#define MSGF_MISC_SR_LINK_AUTO_BWIDTH BIT(25) +#define MSGF_MISC_SR_LINK_BWIDTH BIT(26) #define MSGF_MISC_SR_MASKALL (MSGF_MISC_SR_RXMSG_AVAIL | \ MSGF_MISC_SR_RXMSG_OVER | \ @@ -96,8 +97,8 @@ MSGF_MISC_SR_NON_FATAL_DEV | \ MSGF_MISC_SR_FATAL_DEV | \ MSGF_MISC_SR_LINK_DOWN | \ - MSGF_MSIC_SR_LINK_AUTO_BWIDTH | \ - MSGF_MSIC_SR_LINK_BWIDTH) + MSGF_MISC_SR_LINK_AUTO_BWIDTH | \ + MSGF_MISC_SR_LINK_BWIDTH) /* Legacy interrupt status mask bits */ #define MSGF_LEG_SR_INTA BIT(0) @@ -157,6 +158,7 @@ struct nwl_pcie { void __iomem *breg_base; void __iomem *pcireg_base; void __iomem *ecam_base; + struct phy *phy[4]; phys_addr_t phys_breg_base; /* Physical Bridge Register Base */ phys_addr_t phys_pcie_reg_base; /* Physical PCIe Controller Base */ phys_addr_t phys_ecam_base; /* Physical Configuration Base */ @@ -267,42 +269,42 @@ static irqreturn_t nwl_pcie_misc_handler(int irq, void *data) return IRQ_NONE; if (misc_stat & MSGF_MISC_SR_RXMSG_OVER) - dev_err(dev, "Received Message FIFO Overflow\n"); + dev_err_ratelimited(dev, "Received Message FIFO Overflow\n"); if (misc_stat & MSGF_MISC_SR_SLAVE_ERR) - dev_err(dev, "Slave error\n"); + dev_err_ratelimited(dev, "Slave error\n"); if (misc_stat & MSGF_MISC_SR_MASTER_ERR) - dev_err(dev, "Master error\n"); + dev_err_ratelimited(dev, "Master error\n"); if (misc_stat & MSGF_MISC_SR_I_ADDR_ERR) - dev_err(dev, "In Misc Ingress address translation error\n"); + dev_err_ratelimited(dev, "In Misc Ingress address translation error\n"); if (misc_stat & MSGF_MISC_SR_E_ADDR_ERR) - dev_err(dev, "In Misc Egress address translation error\n"); + dev_err_ratelimited(dev, "In Misc Egress address translation error\n"); if (misc_stat & MSGF_MISC_SR_FATAL_AER) - dev_err(dev, "Fatal Error in AER Capability\n"); + dev_err_ratelimited(dev, "Fatal Error in AER Capability\n"); if (misc_stat & MSGF_MISC_SR_NON_FATAL_AER) - dev_err(dev, "Non-Fatal Error in AER Capability\n"); + dev_err_ratelimited(dev, "Non-Fatal Error in AER Capability\n"); if (misc_stat & MSGF_MISC_SR_CORR_AER) - dev_err(dev, "Correctable Error in AER Capability\n"); + dev_err_ratelimited(dev, "Correctable Error in AER Capability\n"); if (misc_stat & MSGF_MISC_SR_UR_DETECT) - dev_err(dev, "Unsupported request Detected\n"); + dev_err_ratelimited(dev, "Unsupported request Detected\n"); if (misc_stat & MSGF_MISC_SR_NON_FATAL_DEV) - dev_err(dev, "Non-Fatal Error Detected\n"); + dev_err_ratelimited(dev, "Non-Fatal Error Detected\n"); if (misc_stat & MSGF_MISC_SR_FATAL_DEV) - dev_err(dev, "Fatal Error Detected\n"); + dev_err_ratelimited(dev, "Fatal Error Detected\n"); - if (misc_stat & MSGF_MSIC_SR_LINK_AUTO_BWIDTH) + if (misc_stat & MSGF_MISC_SR_LINK_AUTO_BWIDTH) dev_info(dev, "Link Autonomous Bandwidth Management Status bit set\n"); - if (misc_stat & MSGF_MSIC_SR_LINK_BWIDTH) + if (misc_stat & MSGF_MISC_SR_LINK_BWIDTH) dev_info(dev, "Link Bandwidth Management Status bit set\n"); /* Clear misc interrupt status */ @@ -371,7 +373,7 @@ static void nwl_mask_intx_irq(struct irq_data *data) u32 mask; u32 val; - mask = 1 << (data->hwirq - 1); + mask = 1 << data->hwirq; raw_spin_lock_irqsave(&pcie->leg_mask_lock, flags); val = nwl_bridge_readl(pcie, MSGF_LEG_MASK); nwl_bridge_writel(pcie, (val & (~mask)), MSGF_LEG_MASK); @@ -385,7 +387,7 @@ static void nwl_unmask_intx_irq(struct irq_data *data) u32 mask; u32 val; - mask = 1 << (data->hwirq - 1); + mask = 1 << data->hwirq; raw_spin_lock_irqsave(&pcie->leg_mask_lock, flags); val = nwl_bridge_readl(pcie, MSGF_LEG_MASK); nwl_bridge_writel(pcie, (val | mask), MSGF_LEG_MASK); @@ -425,8 +427,8 @@ static struct irq_chip nwl_msi_irq_chip = { }; static struct msi_domain_info nwl_msi_domain_info = { - .flags = (MSI_FLAG_USE_DEF_DOM_OPS | MSI_FLAG_USE_DEF_CHIP_OPS | - MSI_FLAG_MULTI_PCI_MSI), + .flags = MSI_FLAG_USE_DEF_DOM_OPS | MSI_FLAG_USE_DEF_CHIP_OPS | + MSI_FLAG_NO_AFFINITY | MSI_FLAG_MULTI_PCI_MSI, .chip = &nwl_msi_irq_chip, }; #endif @@ -441,16 +443,9 @@ static void nwl_compose_msi_msg(struct irq_data *data, struct msi_msg *msg) msg->data = data->hwirq; } -static int nwl_msi_set_affinity(struct irq_data *irq_data, - const struct cpumask *mask, bool force) -{ - return -EINVAL; -} - static struct irq_chip nwl_irq_chip = { .name = "Xilinx MSI", .irq_compose_msi_msg = nwl_compose_msi_msg, - .irq_set_affinity = nwl_msi_set_affinity, }; static int nwl_irq_domain_alloc(struct irq_domain *domain, unsigned int virq, @@ -521,6 +516,60 @@ static int nwl_pcie_init_msi_irq_domain(struct nwl_pcie *pcie) return 0; } +static void nwl_pcie_phy_power_off(struct nwl_pcie *pcie, int i) +{ + int err = phy_power_off(pcie->phy[i]); + + if (err) + dev_err(pcie->dev, "could not power off phy %d (err=%d)\n", i, + err); +} + +static void nwl_pcie_phy_exit(struct nwl_pcie *pcie, int i) +{ + int err = phy_exit(pcie->phy[i]); + + if (err) + dev_err(pcie->dev, "could not exit phy %d (err=%d)\n", i, err); +} + +static int nwl_pcie_phy_enable(struct nwl_pcie *pcie) +{ + int i, ret; + + for (i = 0; i < ARRAY_SIZE(pcie->phy); i++) { + ret = phy_init(pcie->phy[i]); + if (ret) + goto err; + + ret = phy_power_on(pcie->phy[i]); + if (ret) { + nwl_pcie_phy_exit(pcie, i); + goto err; + } + } + + return 0; + +err: + while (i--) { + nwl_pcie_phy_power_off(pcie, i); + nwl_pcie_phy_exit(pcie, i); + } + + return ret; +} + +static void nwl_pcie_phy_disable(struct nwl_pcie *pcie) +{ + int i; + + for (i = ARRAY_SIZE(pcie->phy); i--;) { + nwl_pcie_phy_power_off(pcie, i); + nwl_pcie_phy_exit(pcie, i); + } +} + static int nwl_pcie_init_irq_domain(struct nwl_pcie *pcie) { struct device *dev = pcie->dev; @@ -732,6 +781,7 @@ static int nwl_pcie_parse_dt(struct nwl_pcie *pcie, { struct device *dev = pcie->dev; struct resource *res; + int i; res = platform_get_resource_byname(pdev, IORESOURCE_MEM, "breg"); pcie->breg_base = devm_ioremap_resource(dev, res); @@ -759,6 +809,18 @@ static int nwl_pcie_parse_dt(struct nwl_pcie *pcie, irq_set_chained_handler_and_data(pcie->irq_intx, nwl_pcie_leg_handler, pcie); + + for (i = 0; i < ARRAY_SIZE(pcie->phy); i++) { + pcie->phy[i] = devm_of_phy_get_by_index(dev, dev->of_node, i); + if (PTR_ERR(pcie->phy[i]) == -ENODEV) { + pcie->phy[i] = NULL; + break; + } + + if (IS_ERR(pcie->phy[i])) + return PTR_ERR(pcie->phy[i]); + } + return 0; } @@ -779,6 +841,7 @@ static int nwl_pcie_probe(struct platform_device *pdev) return -ENODEV; pcie = pci_host_bridge_priv(bridge); + platform_set_drvdata(pdev, pcie); pcie->dev = dev; @@ -798,16 +861,22 @@ static int nwl_pcie_probe(struct platform_device *pdev) return err; } + err = nwl_pcie_phy_enable(pcie); + if (err) { + dev_err(dev, "could not enable PHYs\n"); + goto err_clk; + } + err = nwl_pcie_bridge_init(pcie); if (err) { dev_err(dev, "HW Initialization failed\n"); - return err; + goto err_phy; } err = nwl_pcie_init_irq_domain(pcie); if (err) { dev_err(dev, "Failed creating IRQ Domain\n"); - return err; + goto err_phy; } bridge->sysdata = pcie; @@ -817,11 +886,27 @@ static int nwl_pcie_probe(struct platform_device *pdev) err = nwl_pcie_enable_msi(pcie); if (err < 0) { dev_err(dev, "failed to enable MSI support: %d\n", err); - return err; + goto err_phy; } } - return pci_host_probe(bridge); + err = pci_host_probe(bridge); + if (!err) + return 0; + +err_phy: + nwl_pcie_phy_disable(pcie); +err_clk: + clk_disable_unprepare(pcie->clk); + return err; +} + +static void nwl_pcie_remove(struct platform_device *pdev) +{ + struct nwl_pcie *pcie = platform_get_drvdata(pdev); + + nwl_pcie_phy_disable(pcie); + clk_disable_unprepare(pcie->clk); } static struct platform_driver nwl_pcie_driver = { @@ -831,5 +916,6 @@ static struct platform_driver nwl_pcie_driver = { .of_match_table = nwl_pcie_of_match, }, .probe = nwl_pcie_probe, + .remove_new = nwl_pcie_remove, }; builtin_platform_driver(nwl_pcie_driver); diff --git a/drivers/pci/controller/pcie-xilinx.c b/drivers/pci/controller/pcie-xilinx.c index cb6e9f7b0152..0b534f73a942 100644 --- a/drivers/pci/controller/pcie-xilinx.c +++ b/drivers/pci/controller/pcie-xilinx.c @@ -208,11 +208,6 @@ static struct irq_chip xilinx_msi_top_chip = { .irq_ack = xilinx_msi_top_irq_ack, }; -static int xilinx_msi_set_affinity(struct irq_data *d, const struct cpumask *mask, bool force) -{ - return -EINVAL; -} - static void xilinx_compose_msi_msg(struct irq_data *data, struct msi_msg *msg) { struct xilinx_pcie *pcie = irq_data_get_irq_chip_data(data); @@ -225,7 +220,6 @@ static void xilinx_compose_msi_msg(struct irq_data *data, struct msi_msg *msg) static struct irq_chip xilinx_msi_bottom_chip = { .name = "Xilinx MSI", - .irq_set_affinity = xilinx_msi_set_affinity, .irq_compose_msi_msg = xilinx_compose_msi_msg, }; @@ -271,7 +265,8 @@ static const struct irq_domain_ops xilinx_msi_domain_ops = { }; static struct msi_domain_info xilinx_msi_info = { - .flags = (MSI_FLAG_USE_DEF_DOM_OPS | MSI_FLAG_USE_DEF_CHIP_OPS), + .flags = MSI_FLAG_USE_DEF_DOM_OPS | MSI_FLAG_USE_DEF_CHIP_OPS | + MSI_FLAG_NO_AFFINITY, .chip = &xilinx_msi_top_chip, }; diff --git a/drivers/pci/controller/plda/pcie-plda-host.c b/drivers/pci/controller/plda/pcie-plda-host.c index a18923d7cea6..8533dc618d45 100644 --- a/drivers/pci/controller/plda/pcie-plda-host.c +++ b/drivers/pci/controller/plda/pcie-plda-host.c @@ -76,17 +76,10 @@ static void plda_compose_msi_msg(struct irq_data *data, struct msi_msg *msg) (int)data->hwirq, msg->address_hi, msg->address_lo); } -static int plda_msi_set_affinity(struct irq_data *irq_data, - const struct cpumask *mask, bool force) -{ - return -EINVAL; -} - static struct irq_chip plda_msi_bottom_irq_chip = { .name = "PLDA MSI", .irq_ack = plda_msi_bottom_irq_ack, .irq_compose_msi_msg = plda_compose_msi_msg, - .irq_set_affinity = plda_msi_set_affinity, }; static int plda_irq_msi_domain_alloc(struct irq_domain *domain, @@ -146,8 +139,8 @@ static struct irq_chip plda_msi_irq_chip = { }; static struct msi_domain_info plda_msi_domain_info = { - .flags = (MSI_FLAG_USE_DEF_DOM_OPS | MSI_FLAG_USE_DEF_CHIP_OPS | - MSI_FLAG_PCI_MSIX), + .flags = MSI_FLAG_USE_DEF_DOM_OPS | MSI_FLAG_USE_DEF_CHIP_OPS | + MSI_FLAG_NO_AFFINITY | MSI_FLAG_PCI_MSIX, .chip = &plda_msi_irq_chip, }; diff --git a/drivers/pci/controller/vmd.c b/drivers/pci/controller/vmd.c index a726de0af011..264a180403a0 100644 --- a/drivers/pci/controller/vmd.c +++ b/drivers/pci/controller/vmd.c @@ -204,22 +204,11 @@ static void vmd_irq_disable(struct irq_data *data) raw_spin_unlock_irqrestore(&list_lock, flags); } -/* - * XXX: Stubbed until we develop acceptable way to not create conflicts with - * other devices sharing the same vector. - */ -static int vmd_irq_set_affinity(struct irq_data *data, - const struct cpumask *dest, bool force) -{ - return -EINVAL; -} - static struct irq_chip vmd_msi_controller = { .name = "VMD-MSI", .irq_enable = vmd_irq_enable, .irq_disable = vmd_irq_disable, .irq_compose_msi_msg = vmd_compose_msi_msg, - .irq_set_affinity = vmd_irq_set_affinity, }; static irq_hw_number_t vmd_get_hwirq(struct msi_domain_info *info, @@ -326,7 +315,7 @@ static struct msi_domain_ops vmd_msi_domain_ops = { static struct msi_domain_info vmd_msi_domain_info = { .flags = MSI_FLAG_USE_DEF_DOM_OPS | MSI_FLAG_USE_DEF_CHIP_OPS | - MSI_FLAG_PCI_MSIX, + MSI_FLAG_NO_AFFINITY | MSI_FLAG_PCI_MSIX, .ops = &vmd_msi_domain_ops, .chip = &vmd_msi_controller, }; @@ -1053,9 +1042,9 @@ static void vmd_remove(struct pci_dev *dev) static void vmd_shutdown(struct pci_dev *dev) { - struct vmd_dev *vmd = pci_get_drvdata(dev); + struct vmd_dev *vmd = pci_get_drvdata(dev); - vmd_remove_irq_domain(vmd); + vmd_remove_irq_domain(vmd); } #ifdef CONFIG_PM_SLEEP diff --git a/drivers/pci/devres.c b/drivers/pci/devres.c index c7affbbf73ab..b133967faef8 100644 --- a/drivers/pci/devres.c +++ b/drivers/pci/devres.c @@ -730,7 +730,7 @@ EXPORT_SYMBOL(pcim_iounmap); * Mapping and region will get automatically released on driver detach. If * desired, release manually only with pcim_iounmap_region(). */ -static void __iomem *pcim_iomap_region(struct pci_dev *pdev, int bar, +void __iomem *pcim_iomap_region(struct pci_dev *pdev, int bar, const char *name) { int ret; @@ -763,6 +763,7 @@ err_region: return IOMEM_ERR_PTR(ret); } +EXPORT_SYMBOL(pcim_iomap_region); /** * pcim_iounmap_region - Unmap and release a PCI BAR @@ -785,7 +786,7 @@ static void pcim_iounmap_region(struct pci_dev *pdev, int bar) } /** - * pcim_iomap_regions - Request and iomap PCI BARs + * pcim_iomap_regions - Request and iomap PCI BARs (DEPRECATED) * @pdev: PCI device to map IO resources for * @mask: Mask of BARs to request and iomap * @name: Name associated with the requests @@ -793,6 +794,9 @@ static void pcim_iounmap_region(struct pci_dev *pdev, int bar) * Returns: 0 on success, negative error code on failure. * * Request and iomap regions specified by @mask. + * + * This function is DEPRECATED. Do not use it in new code. + * Use pcim_iomap_region() instead. */ int pcim_iomap_regions(struct pci_dev *pdev, int mask, const char *name) { @@ -865,6 +869,7 @@ int pcim_request_region(struct pci_dev *pdev, int bar, const char *name) { return _pcim_request_region(pdev, bar, name, 0); } +EXPORT_SYMBOL(pcim_request_region); /** * pcim_request_region_exclusive - Request a PCI BAR exclusively diff --git a/drivers/pci/endpoint/pci-epc-core.c b/drivers/pci/endpoint/pci-epc-core.c index 84309dfe0c68..17f007109255 100644 --- a/drivers/pci/endpoint/pci-epc-core.c +++ b/drivers/pci/endpoint/pci-epc-core.c @@ -838,6 +838,10 @@ void pci_epc_destroy(struct pci_epc *epc) { pci_ep_cfs_remove_epc_group(epc->group); device_unregister(&epc->dev); + +#ifdef CONFIG_PCI_DOMAINS_GENERIC + pci_bus_release_domain_nr(&epc->dev, epc->domain_nr); +#endif } EXPORT_SYMBOL_GPL(pci_epc_destroy); @@ -900,6 +904,16 @@ __pci_epc_create(struct device *dev, const struct pci_epc_ops *ops, epc->dev.release = pci_epc_release; epc->ops = ops; +#ifdef CONFIG_PCI_DOMAINS_GENERIC + epc->domain_nr = pci_bus_find_domain_nr(NULL, dev); +#else + /* + * TODO: If the architecture doesn't support generic PCI + * domains, then a custom implementation has to be used. + */ + WARN_ONCE(1, "This architecture doesn't support generic PCI domains\n"); +#endif + ret = dev_set_name(&epc->dev, "%s", dev_name(dev)); if (ret) goto put_dev; diff --git a/drivers/pci/hotplug/TODO b/drivers/pci/hotplug/TODO index 9d428b0ea524..92e6e20e8595 100644 --- a/drivers/pci/hotplug/TODO +++ b/drivers/pci/hotplug/TODO @@ -51,11 +51,6 @@ ibmphp: shpchp: -* There is only a single implementation of struct hpc_ops. Can the struct be - removed and its functions invoked directly? This has already been done in - pciehp with commit 82a9e79ef132 ("PCI: pciehp: remove hpc_ops"). Clarify - if there was a specific reason not to apply the same change to shpchp. - * The hardirq handler shpc_isr() queues events on a workqueue. It can be simplified by converting it to threaded IRQ handling. Use pciehp as a template. diff --git a/drivers/pci/hotplug/cpqphp_core.c b/drivers/pci/hotplug/cpqphp_core.c index c94b40e64baf..47a3ed16159a 100644 --- a/drivers/pci/hotplug/cpqphp_core.c +++ b/drivers/pci/hotplug/cpqphp_core.c @@ -328,7 +328,7 @@ get_slot_mapping(struct pci_bus *bus, u8 bus_num, u8 dev_num, u8 *slot) } else { /* Did not get a match on the target PCI device. Check * if the current IRQ table entry is a PCI-to-PCI - * bridge device. If so, and it's secondary bus + * bridge device. If so, and its secondary bus * matches the bus number for the target device, I need * to save the bridge's slot number. If I can not find * an entry for the target device, I will have to diff --git a/drivers/pci/hotplug/cpqphp_pci.c b/drivers/pci/hotplug/cpqphp_pci.c index e9f1fb333a71..718bc6cf12cb 100644 --- a/drivers/pci/hotplug/cpqphp_pci.c +++ b/drivers/pci/hotplug/cpqphp_pci.c @@ -138,7 +138,7 @@ static int PCI_RefinedAccessConfig(struct pci_bus *bus, unsigned int devfn, u8 o if (pci_bus_read_config_dword(bus, devfn, PCI_VENDOR_ID, &vendID) == -1) return -1; - if (vendID == 0xffffffff) + if (PCI_POSSIBLE_ERROR(vendID)) return -1; return pci_bus_read_config_dword(bus, devfn, offset, value); } @@ -253,7 +253,7 @@ static int PCI_GetBusDevHelper(struct controller *ctrl, u8 *bus_num, u8 *dev_num *dev_num = tdevice; ctrl->pci_bus->number = tbus; pci_bus_read_config_dword(ctrl->pci_bus, *dev_num, PCI_VENDOR_ID, &work); - if (!nobridge || (work == 0xffffffff)) + if (!nobridge || PCI_POSSIBLE_ERROR(work)) return 0; dbg("bus_num %d devfn %d\n", *bus_num, *dev_num); diff --git a/drivers/pci/hotplug/s390_pci_hpc.c b/drivers/pci/hotplug/s390_pci_hpc.c index 7333b305f2a5..055518ee354d 100644 --- a/drivers/pci/hotplug/s390_pci_hpc.c +++ b/drivers/pci/hotplug/s390_pci_hpc.c @@ -112,7 +112,7 @@ static int get_power_status(struct hotplug_slot *hotplug_slot, u8 *value) static int get_adapter_status(struct hotplug_slot *hotplug_slot, u8 *value) { - /* if the slot exits it always contains a function */ + /* if the slot exists it always contains a function */ *value = 1; return 0; } diff --git a/drivers/pci/hotplug/shpchp.h b/drivers/pci/hotplug/shpchp.h index 3a97f455336e..f0e2d2d54d71 100644 --- a/drivers/pci/hotplug/shpchp.h +++ b/drivers/pci/hotplug/shpchp.h @@ -72,7 +72,6 @@ struct slot { u8 latch_save; u8 pwr_save; struct controller *ctrl; - const struct hpc_ops *hpc_ops; struct hotplug_slot hotplug_slot; struct list_head slot_list; struct delayed_work work; /* work for button event */ @@ -94,7 +93,6 @@ struct controller { int slot_num_inc; /* 1 or -1 */ struct pci_dev *pci_dev; struct list_head slot_list; - const struct hpc_ops *hpc_ops; wait_queue_head_t queue; /* sleep & wake process */ u8 slot_device_offset; u32 pcix_misc2_reg; /* for amd pogo errata */ @@ -300,24 +298,22 @@ static inline void amd_pogo_errata_restore_misc_reg(struct slot *p_slot) pci_write_config_dword(p_slot->ctrl->pci_dev, PCIX_MISCII_OFFSET, pcix_misc2_temp); } -struct hpc_ops { - int (*power_on_slot)(struct slot *slot); - int (*slot_enable)(struct slot *slot); - int (*slot_disable)(struct slot *slot); - int (*set_bus_speed_mode)(struct slot *slot, enum pci_bus_speed speed); - int (*get_power_status)(struct slot *slot, u8 *status); - int (*get_attention_status)(struct slot *slot, u8 *status); - int (*set_attention_status)(struct slot *slot, u8 status); - int (*get_latch_status)(struct slot *slot, u8 *status); - int (*get_adapter_status)(struct slot *slot, u8 *status); - int (*get_adapter_speed)(struct slot *slot, enum pci_bus_speed *speed); - int (*get_prog_int)(struct slot *slot, u8 *prog_int); - int (*query_power_fault)(struct slot *slot); - void (*green_led_on)(struct slot *slot); - void (*green_led_off)(struct slot *slot); - void (*green_led_blink)(struct slot *slot); - void (*release_ctlr)(struct controller *ctrl); - int (*check_cmd_status)(struct controller *ctrl); -}; +int shpchp_power_on_slot(struct slot *slot); +int shpchp_slot_enable(struct slot *slot); +int shpchp_slot_disable(struct slot *slot); +int shpchp_set_bus_speed_mode(struct slot *slot, enum pci_bus_speed speed); +int shpchp_get_power_status(struct slot *slot, u8 *status); +int shpchp_get_attention_status(struct slot *slot, u8 *status); +int shpchp_set_attention_status(struct slot *slot, u8 status); +int shpchp_get_latch_status(struct slot *slot, u8 *status); +int shpchp_get_adapter_status(struct slot *slot, u8 *status); +int shpchp_get_adapter_speed(struct slot *slot, enum pci_bus_speed *speed); +int shpchp_get_prog_int(struct slot *slot, u8 *prog_int); +int shpchp_query_power_fault(struct slot *slot); +void shpchp_green_led_on(struct slot *slot); +void shpchp_green_led_off(struct slot *slot); +void shpchp_green_led_blink(struct slot *slot); +void shpchp_release_ctlr(struct controller *ctrl); +int shpchp_check_cmd_status(struct controller *ctrl); #endif /* _SHPCHP_H */ diff --git a/drivers/pci/hotplug/shpchp_core.c b/drivers/pci/hotplug/shpchp_core.c index 56c7795ed890..a92e28b72908 100644 --- a/drivers/pci/hotplug/shpchp_core.c +++ b/drivers/pci/hotplug/shpchp_core.c @@ -81,7 +81,6 @@ static int init_slots(struct controller *ctrl) slot->ctrl = ctrl; slot->bus = ctrl->pci_dev->subordinate->number; slot->device = ctrl->slot_device_offset + i; - slot->hpc_ops = ctrl->hpc_ops; slot->number = ctrl->first_slot + (ctrl->slot_num_inc * i); slot->wq = alloc_workqueue("shpchp-%d", 0, 0, slot->number); @@ -150,7 +149,7 @@ static int set_attention_status(struct hotplug_slot *hotplug_slot, u8 status) __func__, slot_name(slot)); slot->attention_save = status; - slot->hpc_ops->set_attention_status(slot, status); + shpchp_set_attention_status(slot, status); return 0; } @@ -183,7 +182,7 @@ static int get_power_status(struct hotplug_slot *hotplug_slot, u8 *value) ctrl_dbg(slot->ctrl, "%s: physical_slot = %s\n", __func__, slot_name(slot)); - retval = slot->hpc_ops->get_power_status(slot, value); + retval = shpchp_get_power_status(slot, value); if (retval < 0) *value = slot->pwr_save; @@ -198,7 +197,7 @@ static int get_attention_status(struct hotplug_slot *hotplug_slot, u8 *value) ctrl_dbg(slot->ctrl, "%s: physical_slot = %s\n", __func__, slot_name(slot)); - retval = slot->hpc_ops->get_attention_status(slot, value); + retval = shpchp_get_attention_status(slot, value); if (retval < 0) *value = slot->attention_save; @@ -213,7 +212,7 @@ static int get_latch_status(struct hotplug_slot *hotplug_slot, u8 *value) ctrl_dbg(slot->ctrl, "%s: physical_slot = %s\n", __func__, slot_name(slot)); - retval = slot->hpc_ops->get_latch_status(slot, value); + retval = shpchp_get_latch_status(slot, value); if (retval < 0) *value = slot->latch_save; @@ -228,7 +227,7 @@ static int get_adapter_status(struct hotplug_slot *hotplug_slot, u8 *value) ctrl_dbg(slot->ctrl, "%s: physical_slot = %s\n", __func__, slot_name(slot)); - retval = slot->hpc_ops->get_adapter_status(slot, value); + retval = shpchp_get_adapter_status(slot, value); if (retval < 0) *value = slot->presence_save; @@ -293,7 +292,7 @@ static int shpc_probe(struct pci_dev *pdev, const struct pci_device_id *ent) err_cleanup_slots: cleanup_slots(ctrl); err_out_release_ctlr: - ctrl->hpc_ops->release_ctlr(ctrl); + shpchp_release_ctlr(ctrl); err_out_free_ctrl: kfree(ctrl); err_out_none: @@ -306,7 +305,7 @@ static void shpc_remove(struct pci_dev *dev) dev->shpc_managed = 0; shpchp_remove_ctrl_files(ctrl); - ctrl->hpc_ops->release_ctlr(ctrl); + shpchp_release_ctlr(ctrl); kfree(ctrl); } diff --git a/drivers/pci/hotplug/shpchp_ctrl.c b/drivers/pci/hotplug/shpchp_ctrl.c index 6a6705e0cf17..e6c6f23bae27 100644 --- a/drivers/pci/hotplug/shpchp_ctrl.c +++ b/drivers/pci/hotplug/shpchp_ctrl.c @@ -51,7 +51,7 @@ u8 shpchp_handle_attention_button(u8 hp_slot, struct controller *ctrl) ctrl_dbg(ctrl, "Attention button interrupt received\n"); p_slot = shpchp_find_slot(ctrl, hp_slot + ctrl->slot_device_offset); - p_slot->hpc_ops->get_adapter_status(p_slot, &(p_slot->presence_save)); + shpchp_get_adapter_status(p_slot, &p_slot->presence_save); /* * Button pressed - See if need to TAKE ACTION!!! @@ -75,8 +75,8 @@ u8 shpchp_handle_switch_change(u8 hp_slot, struct controller *ctrl) ctrl_dbg(ctrl, "Switch interrupt received\n"); p_slot = shpchp_find_slot(ctrl, hp_slot + ctrl->slot_device_offset); - p_slot->hpc_ops->get_adapter_status(p_slot, &(p_slot->presence_save)); - p_slot->hpc_ops->get_latch_status(p_slot, &getstatus); + shpchp_get_adapter_status(p_slot, &p_slot->presence_save); + shpchp_get_latch_status(p_slot, &getstatus); ctrl_dbg(ctrl, "Card present %x Power status %x\n", p_slot->presence_save, p_slot->pwr_save); @@ -116,7 +116,7 @@ u8 shpchp_handle_presence_change(u8 hp_slot, struct controller *ctrl) /* * Save the presence state */ - p_slot->hpc_ops->get_adapter_status(p_slot, &(p_slot->presence_save)); + shpchp_get_adapter_status(p_slot, &p_slot->presence_save); if (p_slot->presence_save) { /* * Card Present @@ -148,7 +148,7 @@ u8 shpchp_handle_power_fault(u8 hp_slot, struct controller *ctrl) p_slot = shpchp_find_slot(ctrl, hp_slot + ctrl->slot_device_offset); - if (!(p_slot->hpc_ops->query_power_fault(p_slot))) { + if (!(shpchp_query_power_fault(p_slot))) { /* * Power fault Cleared */ @@ -181,7 +181,7 @@ static int change_bus_speed(struct controller *ctrl, struct slot *p_slot, int rc = 0; ctrl_dbg(ctrl, "Change speed to %d\n", speed); - rc = p_slot->hpc_ops->set_bus_speed_mode(p_slot, speed); + rc = shpchp_set_bus_speed_mode(p_slot, speed); if (rc) { ctrl_err(ctrl, "%s: Issue of set bus speed mode command failed\n", __func__); @@ -241,14 +241,14 @@ static int board_added(struct slot *p_slot) __func__, p_slot->device, ctrl->slot_device_offset, hp_slot); /* Power on slot without connecting to bus */ - rc = p_slot->hpc_ops->power_on_slot(p_slot); + rc = shpchp_power_on_slot(p_slot); if (rc) { ctrl_err(ctrl, "Failed to power on slot\n"); return -1; } if ((ctrl->pci_dev->vendor == 0x8086) && (ctrl->pci_dev->device == 0x0332)) { - rc = p_slot->hpc_ops->set_bus_speed_mode(p_slot, PCI_SPEED_33MHz); + rc = shpchp_set_bus_speed_mode(p_slot, PCI_SPEED_33MHz); if (rc) { ctrl_err(ctrl, "%s: Issue of set bus speed mode command failed\n", __func__); @@ -256,14 +256,14 @@ static int board_added(struct slot *p_slot) } /* turn on board, blink green LED, turn off Amber LED */ - rc = p_slot->hpc_ops->slot_enable(p_slot); + rc = shpchp_slot_enable(p_slot); if (rc) { ctrl_err(ctrl, "Issue of Slot Enable command failed\n"); return rc; } } - rc = p_slot->hpc_ops->get_adapter_speed(p_slot, &asp); + rc = shpchp_get_adapter_speed(p_slot, &asp); if (rc) { ctrl_err(ctrl, "Can't get adapter speed or bus mode mismatch\n"); return WRONG_BUS_FREQUENCY; @@ -285,7 +285,7 @@ static int board_added(struct slot *p_slot) return rc; /* turn on board, blink green LED, turn off Amber LED */ - rc = p_slot->hpc_ops->slot_enable(p_slot); + rc = shpchp_slot_enable(p_slot); if (rc) { ctrl_err(ctrl, "Issue of Slot Enable command failed\n"); return rc; @@ -313,13 +313,13 @@ static int board_added(struct slot *p_slot) p_slot->is_a_board = 0x01; p_slot->pwr_save = 1; - p_slot->hpc_ops->green_led_on(p_slot); + shpchp_green_led_on(p_slot); return 0; err_exit: /* turn off slot, turn on Amber LED, turn off Green LED */ - rc = p_slot->hpc_ops->slot_disable(p_slot); + rc = shpchp_slot_disable(p_slot); if (rc) { ctrl_err(ctrl, "%s: Issue of Slot Disable command failed\n", __func__); @@ -352,14 +352,14 @@ static int remove_board(struct slot *p_slot) p_slot->status = 0x01; /* turn off slot, turn on Amber LED, turn off Green LED */ - rc = p_slot->hpc_ops->slot_disable(p_slot); + rc = shpchp_slot_disable(p_slot); if (rc) { ctrl_err(ctrl, "%s: Issue of Slot Disable command failed\n", __func__); return rc; } - rc = p_slot->hpc_ops->set_attention_status(p_slot, 0); + rc = shpchp_set_attention_status(p_slot, 0); if (rc) { ctrl_err(ctrl, "Issue of Set Attention command failed\n"); return rc; @@ -401,7 +401,7 @@ static void shpchp_pushbutton_thread(struct work_struct *work) case POWERON_STATE: mutex_unlock(&p_slot->lock); if (shpchp_enable_slot(p_slot)) - p_slot->hpc_ops->green_led_off(p_slot); + shpchp_green_led_off(p_slot); mutex_lock(&p_slot->lock); p_slot->state = STATIC_STATE; break; @@ -446,10 +446,10 @@ void shpchp_queue_pushbutton_work(struct work_struct *work) static void update_slot_info(struct slot *slot) { - slot->hpc_ops->get_power_status(slot, &slot->pwr_save); - slot->hpc_ops->get_attention_status(slot, &slot->attention_save); - slot->hpc_ops->get_latch_status(slot, &slot->latch_save); - slot->hpc_ops->get_adapter_status(slot, &slot->presence_save); + shpchp_get_power_status(slot, &slot->pwr_save); + shpchp_get_attention_status(slot, &slot->attention_save); + shpchp_get_latch_status(slot, &slot->latch_save); + shpchp_get_adapter_status(slot, &slot->presence_save); } /* @@ -462,7 +462,7 @@ static void handle_button_press_event(struct slot *p_slot) switch (p_slot->state) { case STATIC_STATE: - p_slot->hpc_ops->get_power_status(p_slot, &getstatus); + shpchp_get_power_status(p_slot, &getstatus); if (getstatus) { p_slot->state = BLINKINGOFF_STATE; ctrl_info(ctrl, "PCI slot #%s - powering off due to button press\n", @@ -473,8 +473,8 @@ static void handle_button_press_event(struct slot *p_slot) slot_name(p_slot)); } /* blink green LED and turn off amber */ - p_slot->hpc_ops->green_led_blink(p_slot); - p_slot->hpc_ops->set_attention_status(p_slot, 0); + shpchp_green_led_blink(p_slot); + shpchp_set_attention_status(p_slot, 0); queue_delayed_work(p_slot->wq, &p_slot->work, 5*HZ); break; @@ -489,10 +489,10 @@ static void handle_button_press_event(struct slot *p_slot) slot_name(p_slot)); cancel_delayed_work(&p_slot->work); if (p_slot->state == BLINKINGOFF_STATE) - p_slot->hpc_ops->green_led_on(p_slot); + shpchp_green_led_on(p_slot); else - p_slot->hpc_ops->green_led_off(p_slot); - p_slot->hpc_ops->set_attention_status(p_slot, 0); + shpchp_green_led_off(p_slot); + shpchp_set_attention_status(p_slot, 0); ctrl_info(ctrl, "PCI slot #%s - action canceled due to button press\n", slot_name(p_slot)); p_slot->state = STATIC_STATE; @@ -526,8 +526,8 @@ static void interrupt_event_handler(struct work_struct *work) break; case INT_POWER_FAULT: ctrl_dbg(p_slot->ctrl, "%s: Power fault\n", __func__); - p_slot->hpc_ops->set_attention_status(p_slot, 1); - p_slot->hpc_ops->green_led_off(p_slot); + shpchp_set_attention_status(p_slot, 1); + shpchp_green_led_off(p_slot); break; default: update_slot_info(p_slot); @@ -547,17 +547,17 @@ static int shpchp_enable_slot (struct slot *p_slot) /* Check to see if (latch closed, card present, power off) */ mutex_lock(&p_slot->ctrl->crit_sect); - rc = p_slot->hpc_ops->get_adapter_status(p_slot, &getstatus); + rc = shpchp_get_adapter_status(p_slot, &getstatus); if (rc || !getstatus) { ctrl_info(ctrl, "No adapter on slot(%s)\n", slot_name(p_slot)); goto out; } - rc = p_slot->hpc_ops->get_latch_status(p_slot, &getstatus); + rc = shpchp_get_latch_status(p_slot, &getstatus); if (rc || getstatus) { ctrl_info(ctrl, "Latch open on slot(%s)\n", slot_name(p_slot)); goto out; } - rc = p_slot->hpc_ops->get_power_status(p_slot, &getstatus); + rc = shpchp_get_power_status(p_slot, &getstatus); if (rc || getstatus) { ctrl_info(ctrl, "Already enabled on slot(%s)\n", slot_name(p_slot)); @@ -567,10 +567,10 @@ static int shpchp_enable_slot (struct slot *p_slot) p_slot->is_a_board = 1; /* We have to save the presence info for these slots */ - p_slot->hpc_ops->get_adapter_status(p_slot, &(p_slot->presence_save)); - p_slot->hpc_ops->get_power_status(p_slot, &(p_slot->pwr_save)); + shpchp_get_adapter_status(p_slot, &p_slot->presence_save); + shpchp_get_power_status(p_slot, &p_slot->pwr_save); ctrl_dbg(ctrl, "%s: p_slot->pwr_save %x\n", __func__, p_slot->pwr_save); - p_slot->hpc_ops->get_latch_status(p_slot, &getstatus); + shpchp_get_latch_status(p_slot, &getstatus); if ((p_slot->ctrl->pci_dev->vendor == PCI_VENDOR_ID_AMD && p_slot->ctrl->pci_dev->device == PCI_DEVICE_ID_AMD_POGO_7458) @@ -584,9 +584,8 @@ static int shpchp_enable_slot (struct slot *p_slot) retval = board_added(p_slot); if (retval) { - p_slot->hpc_ops->get_adapter_status(p_slot, - &(p_slot->presence_save)); - p_slot->hpc_ops->get_latch_status(p_slot, &getstatus); + shpchp_get_adapter_status(p_slot, &p_slot->presence_save); + shpchp_get_latch_status(p_slot, &getstatus); } update_slot_info(p_slot); @@ -608,17 +607,17 @@ static int shpchp_disable_slot (struct slot *p_slot) /* Check to see if (latch closed, card present, power on) */ mutex_lock(&p_slot->ctrl->crit_sect); - rc = p_slot->hpc_ops->get_adapter_status(p_slot, &getstatus); + rc = shpchp_get_adapter_status(p_slot, &getstatus); if (rc || !getstatus) { ctrl_info(ctrl, "No adapter on slot(%s)\n", slot_name(p_slot)); goto out; } - rc = p_slot->hpc_ops->get_latch_status(p_slot, &getstatus); + rc = shpchp_get_latch_status(p_slot, &getstatus); if (rc || getstatus) { ctrl_info(ctrl, "Latch open on slot(%s)\n", slot_name(p_slot)); goto out; } - rc = p_slot->hpc_ops->get_power_status(p_slot, &getstatus); + rc = shpchp_get_power_status(p_slot, &getstatus); if (rc || !getstatus) { ctrl_info(ctrl, "Already disabled on slot(%s)\n", slot_name(p_slot)); diff --git a/drivers/pci/hotplug/shpchp_hpc.c b/drivers/pci/hotplug/shpchp_hpc.c index 48e4daefc44a..012b9e3fe5b0 100644 --- a/drivers/pci/hotplug/shpchp_hpc.c +++ b/drivers/pci/hotplug/shpchp_hpc.c @@ -167,7 +167,6 @@ static irqreturn_t shpc_isr(int irq, void *dev_id); static void start_int_poll_timer(struct controller *ctrl, int sec); -static int hpc_check_cmd_status(struct controller *ctrl); static inline u8 shpc_readb(struct controller *ctrl, int reg) { @@ -317,7 +316,7 @@ static int shpc_write_cmd(struct slot *slot, u8 t_slot, u8 cmd) if (retval) goto out; - cmd_status = hpc_check_cmd_status(slot->ctrl); + cmd_status = shpchp_check_cmd_status(slot->ctrl); if (cmd_status) { ctrl_err(ctrl, "Failed to issued command 0x%x (error code = %d)\n", cmd, cmd_status); @@ -328,7 +327,7 @@ static int shpc_write_cmd(struct slot *slot, u8 t_slot, u8 cmd) return retval; } -static int hpc_check_cmd_status(struct controller *ctrl) +int shpchp_check_cmd_status(struct controller *ctrl) { int retval = 0; u16 cmd_status = shpc_readw(ctrl, CMD_STATUS) & 0x000F; @@ -357,7 +356,7 @@ static int hpc_check_cmd_status(struct controller *ctrl) } -static int hpc_get_attention_status(struct slot *slot, u8 *status) +int shpchp_get_attention_status(struct slot *slot, u8 *status) { struct controller *ctrl = slot->ctrl; u32 slot_reg = shpc_readl(ctrl, SLOT_REG(slot->hp_slot)); @@ -381,7 +380,7 @@ static int hpc_get_attention_status(struct slot *slot, u8 *status) return 0; } -static int hpc_get_power_status(struct slot *slot, u8 *status) +int shpchp_get_power_status(struct slot *slot, u8 *status) { struct controller *ctrl = slot->ctrl; u32 slot_reg = shpc_readl(ctrl, SLOT_REG(slot->hp_slot)); @@ -406,7 +405,7 @@ static int hpc_get_power_status(struct slot *slot, u8 *status) } -static int hpc_get_latch_status(struct slot *slot, u8 *status) +int shpchp_get_latch_status(struct slot *slot, u8 *status) { struct controller *ctrl = slot->ctrl; u32 slot_reg = shpc_readl(ctrl, SLOT_REG(slot->hp_slot)); @@ -416,7 +415,7 @@ static int hpc_get_latch_status(struct slot *slot, u8 *status) return 0; } -static int hpc_get_adapter_status(struct slot *slot, u8 *status) +int shpchp_get_adapter_status(struct slot *slot, u8 *status) { struct controller *ctrl = slot->ctrl; u32 slot_reg = shpc_readl(ctrl, SLOT_REG(slot->hp_slot)); @@ -427,7 +426,7 @@ static int hpc_get_adapter_status(struct slot *slot, u8 *status) return 0; } -static int hpc_get_prog_int(struct slot *slot, u8 *prog_int) +int shpchp_get_prog_int(struct slot *slot, u8 *prog_int) { struct controller *ctrl = slot->ctrl; @@ -436,7 +435,7 @@ static int hpc_get_prog_int(struct slot *slot, u8 *prog_int) return 0; } -static int hpc_get_adapter_speed(struct slot *slot, enum pci_bus_speed *value) +int shpchp_get_adapter_speed(struct slot *slot, enum pci_bus_speed *value) { int retval = 0; struct controller *ctrl = slot->ctrl; @@ -444,7 +443,7 @@ static int hpc_get_adapter_speed(struct slot *slot, enum pci_bus_speed *value) u8 m66_cap = !!(slot_reg & MHZ66_CAP); u8 pi, pcix_cap; - retval = hpc_get_prog_int(slot, &pi); + retval = shpchp_get_prog_int(slot, &pi); if (retval) return retval; @@ -489,7 +488,7 @@ static int hpc_get_adapter_speed(struct slot *slot, enum pci_bus_speed *value) return retval; } -static int hpc_query_power_fault(struct slot *slot) +int shpchp_query_power_fault(struct slot *slot) { struct controller *ctrl = slot->ctrl; u32 slot_reg = shpc_readl(ctrl, SLOT_REG(slot->hp_slot)); @@ -498,7 +497,7 @@ static int hpc_query_power_fault(struct slot *slot) return !(slot_reg & POWER_FAULT); } -static int hpc_set_attention_status(struct slot *slot, u8 value) +int shpchp_set_attention_status(struct slot *slot, u8 value) { u8 slot_cmd = 0; @@ -520,22 +519,22 @@ static int hpc_set_attention_status(struct slot *slot, u8 value) } -static void hpc_set_green_led_on(struct slot *slot) +void shpchp_green_led_on(struct slot *slot) { shpc_write_cmd(slot, slot->hp_slot, SET_PWR_ON); } -static void hpc_set_green_led_off(struct slot *slot) +void shpchp_green_led_off(struct slot *slot) { shpc_write_cmd(slot, slot->hp_slot, SET_PWR_OFF); } -static void hpc_set_green_led_blink(struct slot *slot) +void shpchp_green_led_blink(struct slot *slot) { shpc_write_cmd(slot, slot->hp_slot, SET_PWR_BLINK); } -static void hpc_release_ctlr(struct controller *ctrl) +void shpchp_release_ctlr(struct controller *ctrl) { int i; u32 slot_reg, serr_int; @@ -575,7 +574,7 @@ static void hpc_release_ctlr(struct controller *ctrl) release_mem_region(ctrl->mmio_base, ctrl->mmio_size); } -static int hpc_power_on_slot(struct slot *slot) +int shpchp_power_on_slot(struct slot *slot) { int retval; @@ -586,7 +585,7 @@ static int hpc_power_on_slot(struct slot *slot) return retval; } -static int hpc_slot_enable(struct slot *slot) +int shpchp_slot_enable(struct slot *slot) { int retval; @@ -599,7 +598,7 @@ static int hpc_slot_enable(struct slot *slot) return retval; } -static int hpc_slot_disable(struct slot *slot) +int shpchp_slot_disable(struct slot *slot) { int retval; @@ -681,7 +680,7 @@ static int shpc_get_cur_bus_speed(struct controller *ctrl) } -static int hpc_set_bus_speed_mode(struct slot *slot, enum pci_bus_speed value) +int shpchp_set_bus_speed_mode(struct slot *slot, enum pci_bus_speed value) { int retval; struct controller *ctrl = slot->ctrl; @@ -871,28 +870,6 @@ static int shpc_get_max_bus_speed(struct controller *ctrl) return retval; } -static const struct hpc_ops shpchp_hpc_ops = { - .power_on_slot = hpc_power_on_slot, - .slot_enable = hpc_slot_enable, - .slot_disable = hpc_slot_disable, - .set_bus_speed_mode = hpc_set_bus_speed_mode, - .set_attention_status = hpc_set_attention_status, - .get_power_status = hpc_get_power_status, - .get_attention_status = hpc_get_attention_status, - .get_latch_status = hpc_get_latch_status, - .get_adapter_status = hpc_get_adapter_status, - - .get_adapter_speed = hpc_get_adapter_speed, - .get_prog_int = hpc_get_prog_int, - - .query_power_fault = hpc_query_power_fault, - .green_led_on = hpc_set_green_led_on, - .green_led_off = hpc_set_green_led_off, - .green_led_blink = hpc_set_green_led_blink, - - .release_ctlr = hpc_release_ctlr, -}; - int shpc_init(struct controller *ctrl, struct pci_dev *pdev) { int rc = -1, num_slots = 0; @@ -978,8 +955,6 @@ int shpc_init(struct controller *ctrl, struct pci_dev *pdev) /* Setup wait queue */ init_waitqueue_head(&ctrl->queue); - ctrl->hpc_ops = &shpchp_hpc_ops; - /* Return PCI Controller Info */ slot_config = shpc_readl(ctrl, SLOT_CONFIG); ctrl->slot_device_offset = (slot_config & FIRST_DEV_NUM) >> 8; diff --git a/drivers/pci/iomap.c b/drivers/pci/iomap.c index a715a4803c95..9fb7cacc15cd 100644 --- a/drivers/pci/iomap.c +++ b/drivers/pci/iomap.c @@ -156,7 +156,7 @@ EXPORT_SYMBOL_GPL(pci_iomap_wc); * the different IOMAP ranges. * * But if the architecture does not use the generic iomap code, and if - * it has _not_ defined it's own private pci_iounmap function, we define + * it has _not_ defined its own private pci_iounmap function, we define * it here. * * NOTE! This default implementation assumes that if the architecture diff --git a/drivers/pci/npem.c b/drivers/pci/npem.c new file mode 100644 index 000000000000..97507e0df769 --- /dev/null +++ b/drivers/pci/npem.c @@ -0,0 +1,595 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * PCIe Enclosure management driver created for LED interfaces based on + * indications. It says *what indications* blink but does not specify *how* + * they blink - it is hardware defined. + * + * The driver name refers to Native PCIe Enclosure Management. It is + * first indication oriented standard with specification. + * + * Native PCIe Enclosure Management (NPEM) + * PCIe Base Specification r6.1 sec 6.28, 7.9.19 + * + * _DSM Definitions for PCIe SSD Status LED + * PCI Firmware Specification, r3.3 sec 4.7 + * + * Two backends are supported to manipulate indications: Direct NPEM register + * access (npem_ops) and indirect access through the ACPI _DSM (dsm_ops). + * _DSM is used if supported, else NPEM. + * + * Copyright (c) 2021-2022 Dell Inc. + * Copyright (c) 2023-2024 Intel Corporation + * Mariusz Tkaczyk + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include "pci.h" + +struct indication { + u32 bit; + const char *name; +}; + +static const struct indication npem_indications[] = { + {PCI_NPEM_IND_OK, "enclosure:ok"}, + {PCI_NPEM_IND_LOCATE, "enclosure:locate"}, + {PCI_NPEM_IND_FAIL, "enclosure:fail"}, + {PCI_NPEM_IND_REBUILD, "enclosure:rebuild"}, + {PCI_NPEM_IND_PFA, "enclosure:pfa"}, + {PCI_NPEM_IND_HOTSPARE, "enclosure:hotspare"}, + {PCI_NPEM_IND_ICA, "enclosure:ica"}, + {PCI_NPEM_IND_IFA, "enclosure:ifa"}, + {PCI_NPEM_IND_IDT, "enclosure:idt"}, + {PCI_NPEM_IND_DISABLED, "enclosure:disabled"}, + {PCI_NPEM_IND_SPEC_0, "enclosure:specific_0"}, + {PCI_NPEM_IND_SPEC_1, "enclosure:specific_1"}, + {PCI_NPEM_IND_SPEC_2, "enclosure:specific_2"}, + {PCI_NPEM_IND_SPEC_3, "enclosure:specific_3"}, + {PCI_NPEM_IND_SPEC_4, "enclosure:specific_4"}, + {PCI_NPEM_IND_SPEC_5, "enclosure:specific_5"}, + {PCI_NPEM_IND_SPEC_6, "enclosure:specific_6"}, + {PCI_NPEM_IND_SPEC_7, "enclosure:specific_7"}, + {0, NULL} +}; + +/* _DSM PCIe SSD LED States correspond to NPEM register values */ +static const struct indication dsm_indications[] = { + {PCI_NPEM_IND_OK, "enclosure:ok"}, + {PCI_NPEM_IND_LOCATE, "enclosure:locate"}, + {PCI_NPEM_IND_FAIL, "enclosure:fail"}, + {PCI_NPEM_IND_REBUILD, "enclosure:rebuild"}, + {PCI_NPEM_IND_PFA, "enclosure:pfa"}, + {PCI_NPEM_IND_HOTSPARE, "enclosure:hotspare"}, + {PCI_NPEM_IND_ICA, "enclosure:ica"}, + {PCI_NPEM_IND_IFA, "enclosure:ifa"}, + {PCI_NPEM_IND_IDT, "enclosure:idt"}, + {PCI_NPEM_IND_DISABLED, "enclosure:disabled"}, + {0, NULL} +}; + +#define for_each_indication(ind, inds) \ + for (ind = inds; ind->bit; ind++) + +/* + * The driver has internal list of supported indications. Ideally, the driver + * should not touch bits that are not defined and for which LED devices are + * not exposed but in reality, it needs to turn them off. + * + * Otherwise, there will be no possibility to turn off indications turned on by + * other utilities or turned on by default and it leads to bad user experience. + * + * Additionally, it excludes NPEM commands like RESET or ENABLE. + */ +static u32 reg_to_indications(u32 caps, const struct indication *inds) +{ + const struct indication *ind; + u32 supported_indications = 0; + + for_each_indication(ind, inds) + supported_indications |= ind->bit; + + return caps & supported_indications; +} + +/** + * struct npem_led - LED details + * @indication: indication details + * @npem: NPEM device + * @name: LED name + * @led: LED device + */ +struct npem_led { + const struct indication *indication; + struct npem *npem; + char name[LED_MAX_NAME_SIZE]; + struct led_classdev led; +}; + +/** + * struct npem_ops - backend specific callbacks + * @get_active_indications: get active indications + * npem: NPEM device + * inds: response buffer + * @set_active_indications: set new indications + * npem: npem device + * inds: bit mask to set + * @inds: supported indications array, set of indications is backend specific + * @name: backend name + */ +struct npem_ops { + int (*get_active_indications)(struct npem *npem, u32 *inds); + int (*set_active_indications)(struct npem *npem, u32 inds); + const struct indication *inds; + const char *name; +}; + +/** + * struct npem - NPEM device properties + * @dev: PCI device this driver is attached to + * @ops: backend specific callbacks + * @lock: serializes concurrent access to NPEM device by multiple LED devices + * @pos: cached offset of NPEM Capability Register in Configuration Space; + * only used if NPEM registers are accessed directly and not through _DSM + * @supported_indications: cached bit mask of supported indications; + * non-indication and reserved bits in the NPEM Capability Register are + * cleared in this bit mask + * @active_indications: cached bit mask of active indications; + * non-indication and reserved bits in the NPEM Control Register are + * cleared in this bit mask + * @active_inds_initialized: whether @active_indications has been initialized; + * On Dell platforms, it is required that IPMI drivers are loaded before + * the GET_STATE_DSM method is invoked: They use an IPMI OpRegion to + * get/set the active LEDs. By initializing @active_indications lazily + * (on first access to an LED), IPMI drivers are given a chance to load. + * If they are not loaded in time, users will see various errors on LED + * access in dmesg. Once they are loaded, the errors go away and LED + * access becomes possible. + * @led_cnt: size of @leds array + * @leds: array containing LED class devices of all supported LEDs + */ +struct npem { + struct pci_dev *dev; + const struct npem_ops *ops; + struct mutex lock; + u16 pos; + u32 supported_indications; + u32 active_indications; + unsigned int active_inds_initialized:1; + int led_cnt; + struct npem_led leds[]; +}; + +static int npem_read_reg(struct npem *npem, u16 reg, u32 *val) +{ + int ret = pci_read_config_dword(npem->dev, npem->pos + reg, val); + + return pcibios_err_to_errno(ret); +} + +static int npem_write_ctrl(struct npem *npem, u32 reg) +{ + int pos = npem->pos + PCI_NPEM_CTRL; + int ret = pci_write_config_dword(npem->dev, pos, reg); + + return pcibios_err_to_errno(ret); +} + +static int npem_get_active_indications(struct npem *npem, u32 *inds) +{ + u32 ctrl; + int ret; + + ret = npem_read_reg(npem, PCI_NPEM_CTRL, &ctrl); + if (ret) + return ret; + + /* If PCI_NPEM_CTRL_ENABLE is not set then no indication should blink */ + if (!(ctrl & PCI_NPEM_CTRL_ENABLE)) { + *inds = 0; + return 0; + } + + *inds = ctrl & npem->supported_indications; + + return 0; +} + +static int npem_set_active_indications(struct npem *npem, u32 inds) +{ + int ctrl, ret, ret_val; + u32 cc_status; + + lockdep_assert_held(&npem->lock); + + /* This bit is always required */ + ctrl = inds | PCI_NPEM_CTRL_ENABLE; + + ret = npem_write_ctrl(npem, ctrl); + if (ret) + return ret; + + /* + * For the case where a NPEM command has not completed immediately, + * it is recommended that software not continuously "spin" on polling + * the status register, but rather poll under interrupt at a reduced + * rate; for example at 10 ms intervals. + * + * PCIe r6.1 sec 6.28 "Implementation Note: Software Polling of NPEM + * Command Completed" + */ + ret = read_poll_timeout(npem_read_reg, ret_val, + ret_val || (cc_status & PCI_NPEM_STATUS_CC), + 10 * USEC_PER_MSEC, USEC_PER_SEC, false, npem, + PCI_NPEM_STATUS, &cc_status); + if (ret) + return ret; + if (ret_val) + return ret_val; + + /* + * All writes to control register, including writes that do not change + * the register value, are NPEM commands and should eventually result + * in a command completion indication in the NPEM Status Register. + * + * PCIe Base Specification r6.1 sec 7.9.19.3 + * + * Register may not be updated, or other conflicting bits may be + * cleared. Spec is not strict here. Read NPEM Control register after + * write to keep cache in-sync. + */ + return npem_get_active_indications(npem, &npem->active_indications); +} + +static const struct npem_ops npem_ops = { + .get_active_indications = npem_get_active_indications, + .set_active_indications = npem_set_active_indications, + .name = "Native PCIe Enclosure Management", + .inds = npem_indications, +}; + +#define DSM_GUID GUID_INIT(0x5d524d9d, 0xfff9, 0x4d4b, 0x8c, 0xb7, 0x74, 0x7e,\ + 0xd5, 0x1e, 0x19, 0x4d) +#define GET_SUPPORTED_STATES_DSM 1 +#define GET_STATE_DSM 2 +#define SET_STATE_DSM 3 + +static const guid_t dsm_guid = DSM_GUID; + +static bool npem_has_dsm(struct pci_dev *pdev) +{ + acpi_handle handle; + + handle = ACPI_HANDLE(&pdev->dev); + if (!handle) + return false; + + return acpi_check_dsm(handle, &dsm_guid, 0x1, + BIT(GET_SUPPORTED_STATES_DSM) | + BIT(GET_STATE_DSM) | BIT(SET_STATE_DSM)); +} + +struct dsm_output { + u16 status; + u8 function_specific_err; + u8 vendor_specific_err; + u32 state; +}; + +/** + * dsm_evaluate() - send DSM PCIe SSD Status LED command + * @pdev: PCI device + * @dsm_func: DSM LED Function + * @output: buffer to copy DSM Response + * @value_to_set: value for SET_STATE_DSM function + * + * To not bother caller with ACPI context, the returned _DSM Output Buffer is + * copied. + */ +static int dsm_evaluate(struct pci_dev *pdev, u64 dsm_func, + struct dsm_output *output, u32 value_to_set) +{ + acpi_handle handle = ACPI_HANDLE(&pdev->dev); + union acpi_object *out_obj, arg3[2]; + union acpi_object *arg3_p = NULL; + + if (dsm_func == SET_STATE_DSM) { + arg3[0].type = ACPI_TYPE_PACKAGE; + arg3[0].package.count = 1; + arg3[0].package.elements = &arg3[1]; + + arg3[1].type = ACPI_TYPE_BUFFER; + arg3[1].buffer.length = 4; + arg3[1].buffer.pointer = (u8 *)&value_to_set; + + arg3_p = arg3; + } + + out_obj = acpi_evaluate_dsm_typed(handle, &dsm_guid, 0x1, dsm_func, + arg3_p, ACPI_TYPE_BUFFER); + if (!out_obj) + return -EIO; + + if (out_obj->buffer.length < sizeof(struct dsm_output)) { + ACPI_FREE(out_obj); + return -EIO; + } + + memcpy(output, out_obj->buffer.pointer, sizeof(struct dsm_output)); + + ACPI_FREE(out_obj); + return 0; +} + +static int dsm_get(struct pci_dev *pdev, u64 dsm_func, u32 *buf) +{ + struct dsm_output output; + int ret = dsm_evaluate(pdev, dsm_func, &output, 0); + + if (ret) + return ret; + + if (output.status != 0) + return -EIO; + + *buf = output.state; + return 0; +} + +static int dsm_get_active_indications(struct npem *npem, u32 *buf) +{ + int ret = dsm_get(npem->dev, GET_STATE_DSM, buf); + + /* Filter out not supported indications in response */ + *buf &= npem->supported_indications; + return ret; +} + +static int dsm_set_active_indications(struct npem *npem, u32 value) +{ + struct dsm_output output; + int ret = dsm_evaluate(npem->dev, SET_STATE_DSM, &output, value); + + if (ret) + return ret; + + switch (output.status) { + case 4: + /* + * Not all bits are set. If this bit is set, the platform + * disregarded some or all of the request state changes. OSPM + * should check the resulting PCIe SSD Status LED States to see + * what, if anything, has changed. + * + * PCI Firmware Specification, r3.3 Table 4-19. + */ + if (output.function_specific_err != 1) + return -EIO; + fallthrough; + case 0: + break; + default: + return -EIO; + } + + npem->active_indications = output.state; + + return 0; +} + +static const struct npem_ops dsm_ops = { + .get_active_indications = dsm_get_active_indications, + .set_active_indications = dsm_set_active_indications, + .name = "_DSM PCIe SSD Status LED Management", + .inds = dsm_indications, +}; + +static int npem_initialize_active_indications(struct npem *npem) +{ + int ret; + + lockdep_assert_held(&npem->lock); + + if (npem->active_inds_initialized) + return 0; + + ret = npem->ops->get_active_indications(npem, + &npem->active_indications); + if (ret) + return ret; + + npem->active_inds_initialized = true; + return 0; +} + +/* + * The status of each indicator is cached on first brightness_ get/set time + * and updated at write time. brightness_get() is only responsible for + * reflecting the last written/cached value. + */ +static enum led_brightness brightness_get(struct led_classdev *led) +{ + struct npem_led *nled = container_of(led, struct npem_led, led); + struct npem *npem = nled->npem; + int ret, val = 0; + + ret = mutex_lock_interruptible(&npem->lock); + if (ret) + return ret; + + ret = npem_initialize_active_indications(npem); + if (ret) + goto out; + + if (npem->active_indications & nled->indication->bit) + val = 1; + +out: + mutex_unlock(&npem->lock); + return val; +} + +static int brightness_set(struct led_classdev *led, + enum led_brightness brightness) +{ + struct npem_led *nled = container_of(led, struct npem_led, led); + struct npem *npem = nled->npem; + u32 indications; + int ret; + + ret = mutex_lock_interruptible(&npem->lock); + if (ret) + return ret; + + ret = npem_initialize_active_indications(npem); + if (ret) + goto out; + + if (brightness == 0) + indications = npem->active_indications & ~(nled->indication->bit); + else + indications = npem->active_indications | nled->indication->bit; + + ret = npem->ops->set_active_indications(npem, indications); + +out: + mutex_unlock(&npem->lock); + return ret; +} + +static void npem_free(struct npem *npem) +{ + struct npem_led *nled; + int cnt; + + if (!npem) + return; + + for (cnt = 0; cnt < npem->led_cnt; cnt++) { + nled = &npem->leds[cnt]; + + if (nled->name[0]) + led_classdev_unregister(&nled->led); + } + + mutex_destroy(&npem->lock); + kfree(npem); +} + +static int pci_npem_set_led_classdev(struct npem *npem, struct npem_led *nled) +{ + struct led_classdev *led = &nled->led; + struct led_init_data init_data = {}; + char *name = nled->name; + int ret; + + init_data.devicename = pci_name(npem->dev); + init_data.default_label = nled->indication->name; + + ret = led_compose_name(&npem->dev->dev, &init_data, name); + if (ret) + return ret; + + led->name = name; + led->brightness_set_blocking = brightness_set; + led->brightness_get = brightness_get; + led->max_brightness = 1; + led->default_trigger = "none"; + led->flags = 0; + + ret = led_classdev_register(&npem->dev->dev, led); + if (ret) + /* Clear the name to indicate that it is not registered. */ + name[0] = 0; + return ret; +} + +static int pci_npem_init(struct pci_dev *dev, const struct npem_ops *ops, + int pos, u32 caps) +{ + u32 supported = reg_to_indications(caps, ops->inds); + int supported_cnt = hweight32(supported); + const struct indication *indication; + struct npem_led *nled; + struct npem *npem; + int led_idx = 0; + int ret; + + npem = kzalloc(struct_size(npem, leds, supported_cnt), GFP_KERNEL); + if (!npem) + return -ENOMEM; + + npem->supported_indications = supported; + npem->led_cnt = supported_cnt; + npem->pos = pos; + npem->dev = dev; + npem->ops = ops; + + mutex_init(&npem->lock); + + for_each_indication(indication, npem_indications) { + if (!(npem->supported_indications & indication->bit)) + continue; + + nled = &npem->leds[led_idx++]; + nled->indication = indication; + nled->npem = npem; + + ret = pci_npem_set_led_classdev(npem, nled); + if (ret) { + npem_free(npem); + return ret; + } + } + + dev->npem = npem; + return 0; +} + +void pci_npem_remove(struct pci_dev *dev) +{ + npem_free(dev->npem); +} + +void pci_npem_create(struct pci_dev *dev) +{ + const struct npem_ops *ops = &npem_ops; + int pos = 0, ret; + u32 cap; + + if (npem_has_dsm(dev)) { + /* + * OS should use the DSM for LED control if it is available + * PCI Firmware Spec r3.3 sec 4.7. + */ + ret = dsm_get(dev, GET_SUPPORTED_STATES_DSM, &cap); + if (ret) + return; + + ops = &dsm_ops; + } else { + pos = pci_find_ext_capability(dev, PCI_EXT_CAP_ID_NPEM); + if (pos == 0) + return; + + if (pci_read_config_dword(dev, pos + PCI_NPEM_CAP, &cap) != 0 || + (cap & PCI_NPEM_CAP_CAPABLE) == 0) + return; + } + + pci_info(dev, "Configuring %s\n", ops->name); + + ret = pci_npem_init(dev, ops, pos, cap); + if (ret) + pci_err(dev, "Failed to register %s, err: %d\n", ops->name, + ret); +} diff --git a/drivers/pci/pci-bridge-emul.c b/drivers/pci/pci-bridge-emul.c index 9334b2dd4764..6658c1edd464 100644 --- a/drivers/pci/pci-bridge-emul.c +++ b/drivers/pci/pci-bridge-emul.c @@ -257,8 +257,8 @@ struct pci_bridge_reg_behavior pcie_cap_regs_behavior[PCI_CAP_PCIE_SIZEOF / 4] = */ .rw = (PCI_EXP_RTCTL_SECEE | PCI_EXP_RTCTL_SENFEE | PCI_EXP_RTCTL_SEFEE | PCI_EXP_RTCTL_PMEIE | - PCI_EXP_RTCTL_CRSSVE), - .ro = PCI_EXP_RTCAP_CRSVIS << 16, + PCI_EXP_RTCTL_RRS_SVE), + .ro = PCI_EXP_RTCAP_RRS_SV << 16, }, [PCI_EXP_RTSTA / 4] = { diff --git a/drivers/pci/pci-driver.c b/drivers/pci/pci-driver.c index f412ef73a6e4..35270172c833 100644 --- a/drivers/pci/pci-driver.c +++ b/drivers/pci/pci-driver.c @@ -1670,7 +1670,7 @@ static void pci_dma_cleanup(struct device *dev) iommu_device_unuse_default_domain(dev); } -struct bus_type pci_bus_type = { +const struct bus_type pci_bus_type = { .name = "pci", .match = pci_bus_match, .uevent = pci_uevent, diff --git a/drivers/pci/pci-sysfs.c b/drivers/pci/pci-sysfs.c index 40cfa716392f..5d0f4db1cab7 100644 --- a/drivers/pci/pci-sysfs.c +++ b/drivers/pci/pci-sysfs.c @@ -31,6 +31,10 @@ #include #include "pci.h" +#ifndef ARCH_PCI_DEV_GROUPS +#define ARCH_PCI_DEV_GROUPS +#endif + static int sysfs_initialized; /* = 0 */ /* show configuration fields */ @@ -1624,6 +1628,7 @@ const struct attribute_group *pci_dev_groups[] = { &pci_dev_acpi_attr_group, #endif &pci_dev_resource_resize_group, + ARCH_PCI_DEV_GROUPS NULL, }; diff --git a/drivers/pci/pci.c b/drivers/pci/pci.c index ffaaca0978cb..7d85c04fbba2 100644 --- a/drivers/pci/pci.c +++ b/drivers/pci/pci.c @@ -1283,7 +1283,9 @@ static int pci_dev_wait(struct pci_dev *dev, char *reset_type, int timeout) { int delay = 1; bool retrain = false; - struct pci_dev *bridge; + struct pci_dev *root, *bridge; + + root = pcie_find_root_port(dev); if (pci_is_pcie(dev)) { bridge = pci_upstream_bridge(dev); @@ -1292,16 +1294,23 @@ static int pci_dev_wait(struct pci_dev *dev, char *reset_type, int timeout) } /* - * After reset, the device should not silently discard config - * requests, but it may still indicate that it needs more time by - * responding to them with CRS completions. The Root Port will - * generally synthesize ~0 (PCI_ERROR_RESPONSE) data to complete - * the read (except when CRS SV is enabled and the read was for the - * Vendor ID; in that case it synthesizes 0x0001 data). + * The caller has already waited long enough after a reset that the + * device should respond to config requests, but it may respond + * with Request Retry Status (RRS) if it needs more time to + * initialize. * - * Wait for the device to return a non-CRS completion. Read the - * Command register instead of Vendor ID so we don't have to - * contend with the CRS SV value. + * If the device is below a Root Port with Configuration RRS + * Software Visibility enabled, reading the Vendor ID returns a + * special data value if the device responded with RRS. Read the + * Vendor ID until we get non-RRS status. + * + * If there's no Root Port or Configuration RRS Software Visibility + * is not enabled, the device may still respond with RRS, but + * hardware may retry the config request. If no retries receive + * Successful Completion, hardware generally synthesizes ~0 + * (PCI_ERROR_RESPONSE) data to complete the read. Reading Vendor + * ID for VFs and non-existent devices also returns ~0, so read the + * Command register until it returns something other than ~0. */ for (;;) { u32 id; @@ -1311,9 +1320,15 @@ static int pci_dev_wait(struct pci_dev *dev, char *reset_type, int timeout) return -ENOTTY; } - pci_read_config_dword(dev, PCI_COMMAND, &id); - if (!PCI_POSSIBLE_ERROR(id)) - break; + if (root && root->config_rrs_sv) { + pci_read_config_dword(dev, PCI_VENDOR_ID, &id); + if (!pci_bus_rrs_vendor_id(id)) + break; + } else { + pci_read_config_dword(dev, PCI_COMMAND, &id); + if (!PCI_POSSIBLE_ERROR(id)) + break; + } if (delay > timeout) { pci_warn(dev, "not ready %dms after %s; giving up\n", @@ -1324,7 +1339,7 @@ static int pci_dev_wait(struct pci_dev *dev, char *reset_type, int timeout) if (delay > PCI_RESET_WAIT) { if (retrain) { retrain = false; - if (pcie_failed_link_retrain(bridge)) { + if (pcie_failed_link_retrain(bridge) == 0) { delay = 1; continue; } @@ -4718,7 +4733,15 @@ int pcie_retrain_link(struct pci_dev *pdev, bool use_lt) pcie_capability_clear_word(pdev, PCI_EXP_LNKCTL, PCI_EXP_LNKCTL_RL); } - return pcie_wait_for_link_status(pdev, use_lt, !use_lt); + rc = pcie_wait_for_link_status(pdev, use_lt, !use_lt); + + /* + * Clear LBMS after a manual retrain so that the bit can be used + * to track link speed or width changes made by hardware itself + * in attempt to correct unreliable link operation. + */ + pcie_capability_write_word(pdev, PCI_EXP_LNKSTA, PCI_EXP_LNKSTA_LBMS); + return rc; } /** @@ -5672,8 +5695,10 @@ static void pci_bus_restore_locked(struct pci_bus *bus) list_for_each_entry(dev, &bus->devices, bus_list) { pci_dev_restore(dev); - if (dev->subordinate) + if (dev->subordinate) { + pci_bridge_wait_for_secondary_bus(dev, "bus reset"); pci_bus_restore_locked(dev->subordinate); + } } } @@ -5707,8 +5732,10 @@ static void pci_slot_restore_locked(struct pci_slot *slot) if (!dev->slot || dev->slot != slot) continue; pci_dev_restore(dev); - if (dev->subordinate) + if (dev->subordinate) { + pci_bridge_wait_for_secondary_bus(dev, "slot reset"); pci_bus_restore_locked(dev->subordinate); + } } } @@ -6802,16 +6829,16 @@ static int of_pci_bus_find_domain_nr(struct device *parent) return ida_alloc(&pci_domain_nr_dynamic_ida, GFP_KERNEL); } -static void of_pci_bus_release_domain_nr(struct pci_bus *bus, struct device *parent) +static void of_pci_bus_release_domain_nr(struct device *parent, int domain_nr) { - if (bus->domain_nr < 0) + if (domain_nr < 0) return; /* Release domain from IDA where it was allocated. */ - if (of_get_pci_domain_nr(parent->of_node) == bus->domain_nr) - ida_free(&pci_domain_nr_static_ida, bus->domain_nr); + if (of_get_pci_domain_nr(parent->of_node) == domain_nr) + ida_free(&pci_domain_nr_static_ida, domain_nr); else - ida_free(&pci_domain_nr_dynamic_ida, bus->domain_nr); + ida_free(&pci_domain_nr_dynamic_ida, domain_nr); } int pci_bus_find_domain_nr(struct pci_bus *bus, struct device *parent) @@ -6820,11 +6847,11 @@ int pci_bus_find_domain_nr(struct pci_bus *bus, struct device *parent) acpi_pci_bus_find_domain_nr(bus); } -void pci_bus_release_domain_nr(struct pci_bus *bus, struct device *parent) +void pci_bus_release_domain_nr(struct device *parent, int domain_nr) { if (!acpi_disabled) return; - of_pci_bus_release_domain_nr(bus, parent); + of_pci_bus_release_domain_nr(parent, domain_nr); } #endif diff --git a/drivers/pci/pci.h b/drivers/pci/pci.h index 79c8398f3938..14d00ce45bfa 100644 --- a/drivers/pci/pci.h +++ b/drivers/pci/pci.h @@ -13,9 +13,24 @@ #define PCIE_LINK_RETRAIN_TIMEOUT_MS 1000 -/* Power stable to PERST# inactive from PCIe card Electromechanical Spec */ +/* + * Power stable to PERST# inactive. + * + * See the "Power Sequencing and Reset Signal Timings" table of the PCI Express + * Card Electromechanical Specification, Revision 5.1, Section 2.9.2, Symbol + * "T_PVPERL". + */ #define PCIE_T_PVPERL_MS 100 +/* + * REFCLK stable before PERST# inactive. + * + * See the "Power Sequencing and Reset Signal Timings" table of the PCI Express + * Card Electromechanical Specification, Revision 5.1, Section 2.9.2, Symbol + * "T_PERST-CLK". + */ +#define PCIE_T_PERST_CLK_US 100 + /* * End of conventional reset (PERST# de-asserted) to first configuration * request (device able to respond with a "Request Retry Status" completion), @@ -124,7 +139,6 @@ void pcie_clear_device_status(struct pci_dev *dev); void pcie_clear_root_pme_status(struct pci_dev *dev); bool pci_check_pme_status(struct pci_dev *dev); void pci_pme_wakeup_bus(struct pci_bus *bus); -int __pci_pme_wakeup(struct pci_dev *dev, void *ign); void pci_pme_restore(struct pci_dev *dev); bool pci_dev_need_resume(struct pci_dev *dev); void pci_dev_adjust_pme(struct pci_dev *dev); @@ -139,6 +153,11 @@ bool pci_bridge_d3_possible(struct pci_dev *dev); void pci_bridge_d3_update(struct pci_dev *dev); int pci_bridge_wait_for_secondary_bus(struct pci_dev *dev, char *reset_type); +static inline bool pci_bus_rrs_vendor_id(u32 l) +{ + return (l & 0xffff) == PCI_VENDOR_ID_PCI_SIG; +} + static inline void pci_wakeup_event(struct pci_dev *dev) { /* Wait 100 ms before the system can be put into a sleep state. */ @@ -169,7 +188,6 @@ static inline bool pcie_downstream_port(const struct pci_dev *dev) } void pci_vpd_init(struct pci_dev *dev); -void pci_vpd_release(struct pci_dev *dev); extern const struct attribute_group pci_dev_vpd_attr_group; /* PCI Virtual Channel */ @@ -290,10 +308,10 @@ void pci_put_host_bridge_device(struct device *dev); int pci_configure_extended_tags(struct pci_dev *dev, void *ign); bool pci_bus_read_dev_vendor_id(struct pci_bus *bus, int devfn, u32 *pl, - int crs_timeout); + int rrs_timeout); bool pci_bus_generic_read_dev_vendor_id(struct pci_bus *bus, int devfn, u32 *pl, - int crs_timeout); -int pci_idt_bus_quirk(struct pci_bus *bus, int devfn, u32 *pl, int crs_timeout); + int rrs_timeout); +int pci_idt_bus_quirk(struct pci_bus *bus, int devfn, u32 *pl, int rrs_timeout); int pci_setup_device(struct pci_dev *dev); int __pci_read_base(struct pci_dev *dev, enum pci_bar_type type, @@ -398,6 +416,14 @@ static inline void pci_doe_destroy(struct pci_dev *pdev) { } static inline void pci_doe_disconnected(struct pci_dev *pdev) { } #endif +#ifdef CONFIG_PCI_NPEM +void pci_npem_create(struct pci_dev *dev); +void pci_npem_remove(struct pci_dev *dev); +#else +static inline void pci_npem_create(struct pci_dev *dev) { } +static inline void pci_npem_remove(struct pci_dev *dev) { } +#endif + /** * pci_dev_set_io_state - Set the new error state if possible. * @@ -606,7 +632,7 @@ void pci_acs_init(struct pci_dev *dev); int pci_dev_specific_acs_enabled(struct pci_dev *dev, u16 acs_flags); int pci_dev_specific_enable_acs(struct pci_dev *dev); int pci_dev_specific_disable_acs_redir(struct pci_dev *dev); -bool pcie_failed_link_retrain(struct pci_dev *dev); +int pcie_failed_link_retrain(struct pci_dev *dev); #else static inline int pci_dev_specific_acs_enabled(struct pci_dev *dev, u16 acs_flags) @@ -621,9 +647,9 @@ static inline int pci_dev_specific_disable_acs_redir(struct pci_dev *dev) { return -ENOTTY; } -static inline bool pcie_failed_link_retrain(struct pci_dev *dev) +static inline int pcie_failed_link_retrain(struct pci_dev *dev) { - return false; + return -ENOTTY; } #endif @@ -887,8 +913,6 @@ static inline pci_power_t mid_pci_get_power_state(struct pci_dev *pdev) #endif int pcim_intx(struct pci_dev *dev, int enable); - -int pcim_request_region(struct pci_dev *pdev, int bar, const char *name); int pcim_request_region_exclusive(struct pci_dev *pdev, int bar, const char *name); void pcim_release_region(struct pci_dev *pdev, int bar); diff --git a/drivers/pci/pcie/aer_inject.c b/drivers/pci/pcie/aer_inject.c index f81b2303bf6a..91acc7b17f68 100644 --- a/drivers/pci/pcie/aer_inject.c +++ b/drivers/pci/pcie/aer_inject.c @@ -430,7 +430,7 @@ static int aer_inject(struct aer_error_inj *einj) else rperr->root_status |= PCI_ERR_ROOT_COR_RCV; rperr->source_id &= 0xffff0000; - rperr->source_id |= (einj->bus << 8) | devfn; + rperr->source_id |= PCI_DEVID(einj->bus, devfn); } if (einj->uncor_status) { if (rperr->root_status & PCI_ERR_ROOT_UNCOR_RCV) @@ -443,7 +443,7 @@ static int aer_inject(struct aer_error_inj *einj) rperr->root_status |= PCI_ERR_ROOT_NONFATAL_RCV; rperr->root_status |= PCI_ERR_ROOT_UNCOR_RCV; rperr->source_id &= 0x0000ffff; - rperr->source_id |= ((einj->bus << 8) | devfn) << 16; + rperr->source_id |= PCI_DEVID(einj->bus, devfn) << 16; } spin_unlock_irqrestore(&inject_lock, flags); diff --git a/drivers/pci/probe.c b/drivers/pci/probe.c index b14b9876c030..4f68414c3086 100644 --- a/drivers/pci/probe.c +++ b/drivers/pci/probe.c @@ -1061,7 +1061,7 @@ unregister: free: #ifdef CONFIG_PCI_DOMAINS_GENERIC - pci_bus_release_domain_nr(bus, parent); + pci_bus_release_domain_nr(parent, bus->domain_nr); #endif kfree(bus); return err; @@ -1203,15 +1203,17 @@ struct pci_bus *pci_add_new_bus(struct pci_bus *parent, struct pci_dev *dev, } EXPORT_SYMBOL(pci_add_new_bus); -static void pci_enable_crs(struct pci_dev *pdev) +static void pci_enable_rrs_sv(struct pci_dev *pdev) { u16 root_cap = 0; - /* Enable CRS Software Visibility if supported */ + /* Enable Configuration RRS Software Visibility if supported */ pcie_capability_read_word(pdev, PCI_EXP_RTCAP, &root_cap); - if (root_cap & PCI_EXP_RTCAP_CRSVIS) + if (root_cap & PCI_EXP_RTCAP_RRS_SV) { pcie_capability_set_word(pdev, PCI_EXP_RTCTL, - PCI_EXP_RTCTL_CRSSVE); + PCI_EXP_RTCTL_RRS_SVE); + pdev->config_rrs_sv = 1; + } } static unsigned int pci_scan_child_bus_extend(struct pci_bus *bus, @@ -1326,7 +1328,7 @@ static int pci_scan_bridge_extend(struct pci_bus *bus, struct pci_dev *dev, pci_write_config_word(dev, PCI_BRIDGE_CONTROL, bctl & ~PCI_BRIDGE_CTL_MASTER_ABORT); - pci_enable_crs(dev); + pci_enable_rrs_sv(dev); if ((secondary || subordinate) && !pcibios_assign_all_busses() && !is_cardbus && !broken) { @@ -2343,28 +2345,23 @@ struct pci_dev *pci_alloc_dev(struct pci_bus *bus) } EXPORT_SYMBOL(pci_alloc_dev); -static bool pci_bus_crs_vendor_id(u32 l) -{ - return (l & 0xffff) == PCI_VENDOR_ID_PCI_SIG; -} - -static bool pci_bus_wait_crs(struct pci_bus *bus, int devfn, u32 *l, +static bool pci_bus_wait_rrs(struct pci_bus *bus, int devfn, u32 *l, int timeout) { int delay = 1; - if (!pci_bus_crs_vendor_id(*l)) - return true; /* not a CRS completion */ + if (!pci_bus_rrs_vendor_id(*l)) + return true; /* not a Configuration RRS completion */ if (!timeout) - return false; /* CRS, but caller doesn't want to wait */ + return false; /* RRS, but caller doesn't want to wait */ /* * We got the reserved Vendor ID that indicates a completion with - * Configuration Request Retry Status (CRS). Retry until we get a + * Configuration Request Retry Status (RRS). Retry until we get a * valid Vendor ID or we time out. */ - while (pci_bus_crs_vendor_id(*l)) { + while (pci_bus_rrs_vendor_id(*l)) { if (delay > timeout) { pr_warn("pci %04x:%02x:%02x.%d: not ready after %dms; giving up\n", pci_domain_nr(bus), bus->number, @@ -2403,8 +2400,8 @@ bool pci_bus_generic_read_dev_vendor_id(struct pci_bus *bus, int devfn, u32 *l, *l == 0x0000ffff || *l == 0xffff0000) return false; - if (pci_bus_crs_vendor_id(*l)) - return pci_bus_wait_crs(bus, devfn, l, timeout); + if (pci_bus_rrs_vendor_id(*l)) + return pci_bus_wait_rrs(bus, devfn, l, timeout); return true; } @@ -2593,6 +2590,8 @@ void pci_device_add(struct pci_dev *dev, struct pci_bus *bus) dev->match_driver = false; ret = device_add(&dev->dev); WARN_ON(ret < 0); + + pci_npem_create(dev); } struct pci_dev *pci_scan_single_device(struct pci_bus *bus, int devfn) diff --git a/drivers/pci/pwrctl/pci-pwrctl-pwrseq.c b/drivers/pci/pwrctl/pci-pwrctl-pwrseq.c index f07758c9edad..a23a4312574b 100644 --- a/drivers/pci/pwrctl/pci-pwrctl-pwrseq.c +++ b/drivers/pci/pwrctl/pci-pwrctl-pwrseq.c @@ -66,6 +66,11 @@ static const struct of_device_id pci_pwrctl_pwrseq_of_match[] = { .compatible = "pci17cb,1101", .data = "wlan", }, + { + /* ATH11K in WCN6855 package. */ + .compatible = "pci17cb,1103", + .data = "wlan", + }, { /* ATH12K in WCN7850 package. */ .compatible = "pci17cb,1107", diff --git a/drivers/pci/quirks.c b/drivers/pci/quirks.c index a2ce4e08edf5..dccb60c1d9cc 100644 --- a/drivers/pci/quirks.c +++ b/drivers/pci/quirks.c @@ -66,7 +66,7 @@ * apply this erratum workaround to any downstream ports as long as they * support Link Active reporting and have the Link Control 2 register. * Restrict the speed to 2.5GT/s then with the Target Link Speed field, - * request a retrain and wait 200ms for the data link to go up. + * request a retrain and check the result. * * If this turns out successful and we know by the Vendor:Device ID it is * safe to do so, then lift the restriction, letting the devices negotiate @@ -74,33 +74,45 @@ * firmware may have already arranged and lift it with ports that already * report their data link being up. * - * Return TRUE if the link has been successfully retrained, otherwise FALSE. + * Otherwise revert the speed to the original setting and request a retrain + * again to remove any residual state, ignoring the result as it's supposed + * to fail anyway. + * + * Return 0 if the link has been successfully retrained. Return an error + * if retraining was not needed or we attempted a retrain and it failed. */ -bool pcie_failed_link_retrain(struct pci_dev *dev) +int pcie_failed_link_retrain(struct pci_dev *dev) { static const struct pci_device_id ids[] = { { PCI_VDEVICE(ASMEDIA, 0x2824) }, /* ASMedia ASM2824 */ {} }; u16 lnksta, lnkctl2; + int ret = -ENOTTY; if (!pci_is_pcie(dev) || !pcie_downstream_port(dev) || !pcie_cap_has_lnkctl2(dev) || !dev->link_active_reporting) - return false; + return ret; pcie_capability_read_word(dev, PCI_EXP_LNKCTL2, &lnkctl2); pcie_capability_read_word(dev, PCI_EXP_LNKSTA, &lnksta); if ((lnksta & (PCI_EXP_LNKSTA_LBMS | PCI_EXP_LNKSTA_DLLLA)) == PCI_EXP_LNKSTA_LBMS) { + u16 oldlnkctl2 = lnkctl2; + pci_info(dev, "broken device, retraining non-functional downstream link at 2.5GT/s\n"); lnkctl2 &= ~PCI_EXP_LNKCTL2_TLS; lnkctl2 |= PCI_EXP_LNKCTL2_TLS_2_5GT; pcie_capability_write_word(dev, PCI_EXP_LNKCTL2, lnkctl2); - if (pcie_retrain_link(dev, false)) { + ret = pcie_retrain_link(dev, false); + if (ret) { pci_info(dev, "retraining failed\n"); - return false; + pcie_capability_write_word(dev, PCI_EXP_LNKCTL2, + oldlnkctl2); + pcie_retrain_link(dev, true); + return ret; } pcie_capability_read_word(dev, PCI_EXP_LNKSTA, &lnksta); @@ -117,13 +129,14 @@ bool pcie_failed_link_retrain(struct pci_dev *dev) lnkctl2 |= lnkcap & PCI_EXP_LNKCAP_SLS; pcie_capability_write_word(dev, PCI_EXP_LNKCTL2, lnkctl2); - if (pcie_retrain_link(dev, false)) { + ret = pcie_retrain_link(dev, false); + if (ret) { pci_info(dev, "retraining failed\n"); - return false; + return ret; } } - return true; + return ret; } static ktime_t fixup_debug_start(struct pci_dev *dev, @@ -3608,6 +3621,8 @@ DECLARE_PCI_FIXUP_FINAL(0x1814, 0x0601, /* Ralink RT2800 802.11n PCI */ quirk_broken_intx_masking); DECLARE_PCI_FIXUP_FINAL(0x1b7c, 0x0004, /* Ceton InfiniTV4 */ quirk_broken_intx_masking); +DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_CREATIVE, PCI_DEVICE_ID_CREATIVE_20K2, + quirk_broken_intx_masking); /* * Realtek RTL8169 PCI Gigabit Ethernet Controller (rev 10) @@ -4246,6 +4261,10 @@ static void quirk_dma_func0_alias(struct pci_dev *dev) DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_RICOH, 0xe832, quirk_dma_func0_alias); DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_RICOH, 0xe476, quirk_dma_func0_alias); +/* Some Glenfly chips use function 0 as the PCIe Requester ID for DMA */ +DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_GLENFLY, 0x3d40, quirk_dma_func0_alias); +DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_GLENFLY, 0x3d41, quirk_dma_func0_alias); + static void quirk_dma_func1_alias(struct pci_dev *dev) { if (PCI_FUNC(dev->devfn) != 1) @@ -5070,6 +5089,8 @@ static const struct pci_dev_acs_enabled { /* QCOM QDF2xxx root ports */ { PCI_VENDOR_ID_QCOM, 0x0400, pci_quirk_qcom_rp_acs }, { PCI_VENDOR_ID_QCOM, 0x0401, pci_quirk_qcom_rp_acs }, + /* QCOM SA8775P root port */ + { PCI_VENDOR_ID_QCOM, 0x0115, pci_quirk_qcom_rp_acs }, /* HXT SD4800 root ports. The ACS design is same as QCOM QDF2xxx */ { PCI_VENDOR_ID_HXT, 0x0401, pci_quirk_qcom_rp_acs }, /* Intel PCH root ports */ diff --git a/drivers/pci/remove.c b/drivers/pci/remove.c index 4770cb87e3f0..e4ce1145aa3e 100644 --- a/drivers/pci/remove.c +++ b/drivers/pci/remove.c @@ -50,6 +50,8 @@ static void pci_destroy_dev(struct pci_dev *dev) if (!dev->dev.kobj.parent) return; + pci_npem_remove(dev); + device_del(&dev->dev); down_write(&pci_bus_sem); @@ -179,7 +181,7 @@ void pci_remove_root_bus(struct pci_bus *bus) #ifdef CONFIG_PCI_DOMAINS_GENERIC /* Release domain_nr if it was dynamically allocated */ if (host_bridge->domain_nr == PCI_DOMAIN_NR_NOT_SET) - pci_bus_release_domain_nr(bus, host_bridge->dev.parent); + pci_bus_release_domain_nr(host_bridge->dev.parent, bus->domain_nr); #endif pci_remove_bus(bus); diff --git a/fs/bcachefs/Kconfig b/fs/bcachefs/Kconfig index 5cdfef3b551a..5bac803ea367 100644 --- a/fs/bcachefs/Kconfig +++ b/fs/bcachefs/Kconfig @@ -87,6 +87,13 @@ config BCACHEFS_SIX_OPTIMISTIC_SPIN is held by another thread, spin for a short while, as long as the thread owning the lock is running. +config BCACHEFS_PATH_TRACEPOINTS + bool "Extra btree_path tracepoints" + depends on BCACHEFS_FS + help + Enable extra tracepoints for debugging btree_path operations; we don't + normally want these enabled because they happen at very high rates. + config MEAN_AND_VARIANCE_UNIT_TEST tristate "mean_and_variance unit tests" if !KUNIT_ALL_TESTS depends on KUNIT diff --git a/fs/bcachefs/Makefile b/fs/bcachefs/Makefile index 0ab533a2b03b..56d20e219f59 100644 --- a/fs/bcachefs/Makefile +++ b/fs/bcachefs/Makefile @@ -69,6 +69,7 @@ bcachefs-y := \ printbuf.o \ quota.o \ rebalance.o \ + rcu_pending.o \ recovery.o \ recovery_passes.o \ reflink.o \ diff --git a/fs/bcachefs/acl.c b/fs/bcachefs/acl.c index 331a17f3f113..87f1be9d4db4 100644 --- a/fs/bcachefs/acl.c +++ b/fs/bcachefs/acl.c @@ -361,7 +361,7 @@ retry: bch2_trans_begin(trans); acl = _acl; - ret = bch2_subvol_is_ro_trans(trans, inode->ei_subvol) ?: + ret = bch2_subvol_is_ro_trans(trans, inode->ei_inum.subvol) ?: bch2_inode_peek(trans, &inode_iter, &inode_u, inode_inum(inode), BTREE_ITER_intent); if (ret) diff --git a/fs/bcachefs/alloc_background.c b/fs/bcachefs/alloc_background.c index dc3a4024aab6..645b5ed4babb 100644 --- a/fs/bcachefs/alloc_background.c +++ b/fs/bcachefs/alloc_background.c @@ -30,6 +30,7 @@ #include #include #include +#include static void bch2_discard_one_bucket_fast(struct bch_dev *, u64); @@ -2183,7 +2184,7 @@ int bch2_dev_freespace_init(struct bch_fs *c, struct bch_dev *ca, * freespace/need_discard/need_gc_gens btrees as needed: */ while (1) { - if (last_updated + HZ * 10 < jiffies) { + if (time_after(jiffies, last_updated + HZ * 10)) { bch_info(ca, "%s: currently at %llu/%llu", __func__, iter.pos.offset, ca->mi.nbuckets); last_updated = jiffies; @@ -2297,6 +2298,36 @@ int bch2_fs_freespace_init(struct bch_fs *c) return 0; } +/* device removal */ + +int bch2_dev_remove_alloc(struct bch_fs *c, struct bch_dev *ca) +{ + struct bpos start = POS(ca->dev_idx, 0); + struct bpos end = POS(ca->dev_idx, U64_MAX); + int ret; + + /* + * We clear the LRU and need_discard btrees first so that we don't race + * with bch2_do_invalidates() and bch2_do_discards() + */ + ret = bch2_dev_remove_stripes(c, ca->dev_idx) ?: + bch2_btree_delete_range(c, BTREE_ID_lru, start, end, + BTREE_TRIGGER_norun, NULL) ?: + bch2_btree_delete_range(c, BTREE_ID_need_discard, start, end, + BTREE_TRIGGER_norun, NULL) ?: + bch2_btree_delete_range(c, BTREE_ID_freespace, start, end, + BTREE_TRIGGER_norun, NULL) ?: + bch2_btree_delete_range(c, BTREE_ID_backpointers, start, end, + BTREE_TRIGGER_norun, NULL) ?: + bch2_btree_delete_range(c, BTREE_ID_bucket_gens, start, end, + BTREE_TRIGGER_norun, NULL) ?: + bch2_btree_delete_range(c, BTREE_ID_alloc, start, end, + BTREE_TRIGGER_norun, NULL) ?: + bch2_dev_usage_remove(c, ca->dev_idx); + bch_err_msg(ca, ret, "removing dev alloc info"); + return ret; +} + /* Bucket IO clocks: */ int bch2_bucket_io_time_reset(struct btree_trans *trans, unsigned dev, @@ -2432,13 +2463,15 @@ static bool bch2_dev_has_open_write_point(struct bch_fs *c, struct bch_dev *ca) /* device goes ro: */ void bch2_dev_allocator_remove(struct bch_fs *c, struct bch_dev *ca) { - unsigned i; + lockdep_assert_held(&c->state_lock); /* First, remove device from allocation groups: */ - for (i = 0; i < ARRAY_SIZE(c->rw_devs); i++) + for (unsigned i = 0; i < ARRAY_SIZE(c->rw_devs); i++) clear_bit(ca->dev_idx, c->rw_devs[i].d); + c->rw_devs_change_count++; + /* * Capacity is calculated based off of devices in allocation groups: */ @@ -2467,11 +2500,13 @@ void bch2_dev_allocator_remove(struct bch_fs *c, struct bch_dev *ca) /* device goes rw: */ void bch2_dev_allocator_add(struct bch_fs *c, struct bch_dev *ca) { - unsigned i; + lockdep_assert_held(&c->state_lock); - for (i = 0; i < ARRAY_SIZE(c->rw_devs); i++) + for (unsigned i = 0; i < ARRAY_SIZE(c->rw_devs); i++) if (ca->mi.data_allowed & (1 << i)) set_bit(ca->dev_idx, c->rw_devs[i].d); + + c->rw_devs_change_count++; } void bch2_dev_allocator_background_exit(struct bch_dev *ca) diff --git a/fs/bcachefs/alloc_background.h b/fs/bcachefs/alloc_background.h index fd790b03fbe1..f8e87c6721b1 100644 --- a/fs/bcachefs/alloc_background.h +++ b/fs/bcachefs/alloc_background.h @@ -16,7 +16,7 @@ enum bch_validate_flags; static inline bool bch2_dev_bucket_exists(struct bch_fs *c, struct bpos pos) { rcu_read_lock(); - struct bch_dev *ca = bch2_dev_rcu(c, pos.inode); + struct bch_dev *ca = bch2_dev_rcu_noerror(c, pos.inode); bool ret = ca && bucket_valid(ca, pos.offset); rcu_read_unlock(); return ret; @@ -338,6 +338,7 @@ static inline const struct bch_backpointer *alloc_v4_backpointers_c(const struct int bch2_dev_freespace_init(struct bch_fs *, struct bch_dev *, u64, u64); int bch2_fs_freespace_init(struct bch_fs *); +int bch2_dev_remove_alloc(struct bch_fs *, struct bch_dev *); void bch2_recalc_capacity(struct bch_fs *); u64 bch2_min_rw_member_capacity(struct bch_fs *); diff --git a/fs/bcachefs/alloc_foreground.c b/fs/bcachefs/alloc_foreground.c index 8563c2d26847..d0e0b56892e3 100644 --- a/fs/bcachefs/alloc_foreground.c +++ b/fs/bcachefs/alloc_foreground.c @@ -600,6 +600,7 @@ static struct open_bucket *bch2_bucket_alloc_trans(struct btree_trans *trans, enum bch_watermark watermark, enum bch_data_type data_type, struct closure *cl, + bool nowait, struct bch_dev_usage *usage) { struct bch_fs *c = trans->c; @@ -609,7 +610,7 @@ static struct open_bucket *bch2_bucket_alloc_trans(struct btree_trans *trans, struct bucket_alloc_state s = { .btree_bitmap = data_type == BCH_DATA_btree, }; - bool waiting = false; + bool waiting = nowait; again: bch2_dev_usage_read_fast(ca, usage); avail = dev_buckets_free(ca, *usage, watermark); @@ -685,7 +686,7 @@ struct open_bucket *bch2_bucket_alloc(struct bch_fs *c, struct bch_dev *ca, bch2_trans_do(c, NULL, NULL, 0, PTR_ERR_OR_ZERO(ob = bch2_bucket_alloc_trans(trans, ca, watermark, - data_type, cl, &usage))); + data_type, cl, false, &usage))); return ob; } @@ -748,7 +749,6 @@ static int add_new_bucket(struct bch_fs *c, unsigned nr_replicas, unsigned *nr_effective, bool *have_cache, - unsigned flags, struct open_bucket *ob) { unsigned durability = ob_dev(c, ob)->mi.durability; @@ -775,7 +775,7 @@ int bch2_bucket_alloc_set_trans(struct btree_trans *trans, unsigned nr_replicas, unsigned *nr_effective, bool *have_cache, - unsigned flags, + enum bch_write_flags flags, enum bch_data_type data_type, enum bch_watermark watermark, struct closure *cl) @@ -801,7 +801,8 @@ int bch2_bucket_alloc_set_trans(struct btree_trans *trans, continue; } - ob = bch2_bucket_alloc_trans(trans, ca, watermark, data_type, cl, &usage); + ob = bch2_bucket_alloc_trans(trans, ca, watermark, data_type, + cl, flags & BCH_WRITE_ALLOC_NOWAIT, &usage); if (!IS_ERR(ob)) bch2_dev_stripe_increment_inlined(ca, stripe, &usage); bch2_dev_put(ca); @@ -815,7 +816,7 @@ int bch2_bucket_alloc_set_trans(struct btree_trans *trans, if (add_new_bucket(c, ptrs, devs_may_alloc, nr_replicas, nr_effective, - have_cache, flags, ob)) { + have_cache, ob)) { ret = 0; break; } @@ -841,7 +842,7 @@ static int bucket_alloc_from_stripe(struct btree_trans *trans, unsigned *nr_effective, bool *have_cache, enum bch_watermark watermark, - unsigned flags, + enum bch_write_flags flags, struct closure *cl) { struct bch_fs *c = trans->c; @@ -883,7 +884,7 @@ got_bucket: ret = add_new_bucket(c, ptrs, devs_may_alloc, nr_replicas, nr_effective, - have_cache, flags, ob); + have_cache, ob); out_put_head: bch2_ec_stripe_head_put(c, h); return ret; @@ -922,7 +923,7 @@ static int bucket_alloc_set_writepoint(struct bch_fs *c, unsigned nr_replicas, unsigned *nr_effective, bool *have_cache, - bool ec, unsigned flags) + bool ec) { struct open_buckets ptrs_skip = { .nr = 0 }; struct open_bucket *ob; @@ -934,7 +935,7 @@ static int bucket_alloc_set_writepoint(struct bch_fs *c, have_cache, ec, ob)) ret = add_new_bucket(c, ptrs, devs_may_alloc, nr_replicas, nr_effective, - have_cache, flags, ob); + have_cache, ob); else ob_push(c, &ptrs_skip, ob); } @@ -950,8 +951,7 @@ static int bucket_alloc_set_partial(struct bch_fs *c, unsigned nr_replicas, unsigned *nr_effective, bool *have_cache, bool ec, - enum bch_watermark watermark, - unsigned flags) + enum bch_watermark watermark) { int i, ret = 0; @@ -983,7 +983,7 @@ static int bucket_alloc_set_partial(struct bch_fs *c, ret = add_new_bucket(c, ptrs, devs_may_alloc, nr_replicas, nr_effective, - have_cache, flags, ob); + have_cache, ob); if (ret) break; } @@ -1003,7 +1003,7 @@ static int __open_bucket_add_buckets(struct btree_trans *trans, unsigned *nr_effective, bool *have_cache, enum bch_watermark watermark, - unsigned flags, + enum bch_write_flags flags, struct closure *_cl) { struct bch_fs *c = trans->c; @@ -1022,18 +1022,15 @@ static int __open_bucket_add_buckets(struct btree_trans *trans, open_bucket_for_each(c, ptrs, ob, i) __clear_bit(ob->dev, devs.d); - if (erasure_code && ec_open_bucket(c, ptrs)) - return 0; - ret = bucket_alloc_set_writepoint(c, ptrs, wp, &devs, nr_replicas, nr_effective, - have_cache, erasure_code, flags); + have_cache, erasure_code); if (ret) return ret; ret = bucket_alloc_set_partial(c, ptrs, wp, &devs, nr_replicas, nr_effective, - have_cache, erasure_code, watermark, flags); + have_cache, erasure_code, watermark); if (ret) return ret; @@ -1074,12 +1071,12 @@ static int open_bucket_add_buckets(struct btree_trans *trans, unsigned *nr_effective, bool *have_cache, enum bch_watermark watermark, - unsigned flags, + enum bch_write_flags flags, struct closure *cl) { int ret; - if (erasure_code) { + if (erasure_code && !ec_open_bucket(trans->c, ptrs)) { ret = __open_bucket_add_buckets(trans, ptrs, wp, devs_have, target, erasure_code, nr_replicas, nr_effective, have_cache, @@ -1376,7 +1373,7 @@ int bch2_alloc_sectors_start_trans(struct btree_trans *trans, unsigned nr_replicas, unsigned nr_replicas_required, enum bch_watermark watermark, - unsigned flags, + enum bch_write_flags flags, struct closure *cl, struct write_point **wp_ret) { @@ -1392,8 +1389,6 @@ int bch2_alloc_sectors_start_trans(struct btree_trans *trans, if (!IS_ENABLED(CONFIG_BCACHEFS_ERASURE_CODING)) erasure_code = false; - BUG_ON(flags & BCH_WRITE_ONLY_SPECIFIED_DEVS); - BUG_ON(!nr_replicas || !nr_replicas_required); retry: ptrs.nr = 0; @@ -1498,11 +1493,12 @@ err: try_decrease_writepoints(trans, write_points_nr)) goto retry; - if (bch2_err_matches(ret, BCH_ERR_open_buckets_empty) || + if (cl && bch2_err_matches(ret, BCH_ERR_open_buckets_empty)) + ret = -BCH_ERR_bucket_alloc_blocked; + + if (cl && !(flags & BCH_WRITE_ALLOC_NOWAIT) && bch2_err_matches(ret, BCH_ERR_freelist_empty)) - return cl - ? -BCH_ERR_bucket_alloc_blocked - : -BCH_ERR_ENOSPC_bucket_alloc; + ret = -BCH_ERR_bucket_alloc_blocked; return ret; } @@ -1733,13 +1729,6 @@ void bch2_dev_alloc_debug_to_text(struct printbuf *out, struct bch_dev *ca) for (unsigned i = 0; i < ARRAY_SIZE(c->open_buckets); i++) nr[c->open_buckets[i].data_type]++; - printbuf_tabstops_reset(out); - printbuf_tabstop_push(out, 12); - printbuf_tabstop_push(out, 16); - printbuf_tabstop_push(out, 16); - printbuf_tabstop_push(out, 16); - printbuf_tabstop_push(out, 16); - bch2_dev_usage_to_text(out, ca, &stats); prt_newline(out); diff --git a/fs/bcachefs/alloc_foreground.h b/fs/bcachefs/alloc_foreground.h index 386d231ceca3..1a16fd5bd4f8 100644 --- a/fs/bcachefs/alloc_foreground.h +++ b/fs/bcachefs/alloc_foreground.h @@ -155,9 +155,10 @@ static inline bool bch2_bucket_is_open_safe(struct bch_fs *c, unsigned dev, u64 return ret; } +enum bch_write_flags; int bch2_bucket_alloc_set_trans(struct btree_trans *, struct open_buckets *, struct dev_stripe_state *, struct bch_devs_mask *, - unsigned, unsigned *, bool *, unsigned, + unsigned, unsigned *, bool *, enum bch_write_flags, enum bch_data_type, enum bch_watermark, struct closure *); @@ -167,7 +168,7 @@ int bch2_alloc_sectors_start_trans(struct btree_trans *, struct bch_devs_list *, unsigned, unsigned, enum bch_watermark, - unsigned, + enum bch_write_flags, struct closure *, struct write_point **); diff --git a/fs/bcachefs/backpointers.c b/fs/bcachefs/backpointers.c index d4da6343efa9..e11989a57ca0 100644 --- a/fs/bcachefs/backpointers.c +++ b/fs/bcachefs/backpointers.c @@ -9,6 +9,7 @@ #include "btree_update_interior.h" #include "btree_write_buffer.h" #include "checksum.h" +#include "disk_accounting.h" #include "error.h" #include @@ -53,7 +54,7 @@ int bch2_backpointer_validate(struct bch_fs *c, struct bkey_s_c k, struct bkey_s_c_backpointer bp = bkey_s_c_to_backpointer(k); rcu_read_lock(); - struct bch_dev *ca = bch2_dev_rcu(c, bp.k->p.inode); + struct bch_dev *ca = bch2_dev_rcu_noerror(c, bp.k->p.inode); if (!ca) { /* these will be caught by fsck */ rcu_read_unlock(); @@ -87,7 +88,7 @@ void bch2_backpointer_to_text(struct printbuf *out, const struct bch_backpointer void bch2_backpointer_k_to_text(struct printbuf *out, struct bch_fs *c, struct bkey_s_c k) { rcu_read_lock(); - struct bch_dev *ca = bch2_dev_rcu(c, k.k->p.inode); + struct bch_dev *ca = bch2_dev_rcu_noerror(c, k.k->p.inode); if (ca) { struct bpos bucket = bp_pos_to_bucket(ca, k.k->p); rcu_read_unlock(); @@ -671,7 +672,7 @@ static int check_extent_to_backpointers(struct btree_trans *trans, continue; rcu_read_lock(); - struct bch_dev *ca = bch2_dev_rcu(c, p.ptr.dev); + struct bch_dev *ca = bch2_dev_rcu_noerror(c, p.ptr.dev); if (ca) bch2_extent_ptr_to_bp(c, ca, btree, level, k, p, entry, &bucket_pos, &bp); rcu_read_unlock(); @@ -750,10 +751,12 @@ static int bch2_get_btree_in_memory_pos(struct btree_trans *trans, s64 mem_may_pin = mem_may_pin_bytes(c); int ret = 0; + bch2_btree_cache_unpin(c); + btree_interior_mask |= btree_leaf_mask; - c->btree_cache.pinned_nodes_leaf_mask = btree_leaf_mask; - c->btree_cache.pinned_nodes_interior_mask = btree_interior_mask; + c->btree_cache.pinned_nodes_mask[0] = btree_leaf_mask; + c->btree_cache.pinned_nodes_mask[1] = btree_interior_mask; c->btree_cache.pinned_nodes_start = start; c->btree_cache.pinned_nodes_end = *end = BBPOS_MAX; @@ -775,6 +778,7 @@ static int bch2_get_btree_in_memory_pos(struct btree_trans *trans, BBPOS(btree, b->key.k.p); break; } + bch2_node_pin(c, b); 0; })); } @@ -782,12 +786,80 @@ static int bch2_get_btree_in_memory_pos(struct btree_trans *trans, return ret; } +struct progress_indicator_state { + unsigned long next_print; + u64 nodes_seen; + u64 nodes_total; + struct btree *last_node; +}; + +static inline void progress_init(struct progress_indicator_state *s, + struct bch_fs *c, + u64 btree_id_mask) +{ + memset(s, 0, sizeof(*s)); + + s->next_print = jiffies + HZ * 10; + + for (unsigned i = 0; i < BTREE_ID_NR; i++) { + if (!(btree_id_mask & BIT_ULL(i))) + continue; + + struct disk_accounting_pos acc = { + .type = BCH_DISK_ACCOUNTING_btree, + .btree.id = i, + }; + + u64 v; + bch2_accounting_mem_read(c, disk_accounting_pos_to_bpos(&acc), &v, 1); + s->nodes_total += div64_ul(v, btree_sectors(c)); + } +} + +static inline bool progress_update_p(struct progress_indicator_state *s) +{ + bool ret = time_after_eq(jiffies, s->next_print); + + if (ret) + s->next_print = jiffies + HZ * 10; + return ret; +} + +static void progress_update_iter(struct btree_trans *trans, + struct progress_indicator_state *s, + struct btree_iter *iter, + const char *msg) +{ + struct bch_fs *c = trans->c; + struct btree *b = path_l(btree_iter_path(trans, iter))->b; + + s->nodes_seen += b != s->last_node; + s->last_node = b; + + if (progress_update_p(s)) { + struct printbuf buf = PRINTBUF; + unsigned percent = s->nodes_total + ? div64_u64(s->nodes_seen * 100, s->nodes_total) + : 0; + + prt_printf(&buf, "%s: %d%%, done %llu/%llu nodes, at ", + msg, percent, s->nodes_seen, s->nodes_total); + bch2_bbpos_to_text(&buf, BBPOS(iter->btree_id, iter->pos)); + + bch_info(c, "%s", buf.buf); + printbuf_exit(&buf); + } +} + static int bch2_check_extents_to_backpointers_pass(struct btree_trans *trans, struct extents_to_bp_state *s) { struct bch_fs *c = trans->c; + struct progress_indicator_state progress; int ret = 0; + progress_init(&progress, trans->c, BIT_ULL(BTREE_ID_extents)|BIT_ULL(BTREE_ID_reflink)); + for (enum btree_id btree_id = 0; btree_id < btree_id_nr_alive(c); btree_id++) { @@ -805,6 +877,7 @@ static int bch2_check_extents_to_backpointers_pass(struct btree_trans *trans, BTREE_ITER_prefetch); ret = for_each_btree_key_continue(trans, iter, 0, k, ({ + progress_update_iter(trans, &progress, &iter, "extents_to_backpointers"); check_extent_to_backpointers(trans, s, btree_id, level, k) ?: bch2_trans_commit(trans, NULL, NULL, BCH_TRANS_COMMIT_no_enospc); })); @@ -865,8 +938,7 @@ int bch2_check_extents_to_backpointers(struct bch_fs *c) bch2_trans_put(trans); bch2_bkey_buf_exit(&s.last_flushed, c); - c->btree_cache.pinned_nodes_leaf_mask = 0; - c->btree_cache.pinned_nodes_interior_mask = 0; + bch2_btree_cache_unpin(c); bch_err_fn(c, ret); return ret; @@ -920,19 +992,24 @@ static int bch2_check_backpointers_to_extents_pass(struct btree_trans *trans, struct bbpos start, struct bbpos end) { + struct bch_fs *c = trans->c; struct bkey_buf last_flushed; + struct progress_indicator_state progress; bch2_bkey_buf_init(&last_flushed); bkey_init(&last_flushed.k->k); + progress_init(&progress, trans->c, BIT_ULL(BTREE_ID_backpointers)); int ret = for_each_btree_key_commit(trans, iter, BTREE_ID_backpointers, POS_MIN, BTREE_ITER_prefetch, k, - NULL, NULL, BCH_TRANS_COMMIT_no_enospc, - check_one_backpointer(trans, start, end, - bkey_s_c_to_backpointer(k), - &last_flushed)); + NULL, NULL, BCH_TRANS_COMMIT_no_enospc, ({ + progress_update_iter(trans, &progress, &iter, "backpointers_to_extents"); + check_one_backpointer(trans, start, end, + bkey_s_c_to_backpointer(k), + &last_flushed); + })); - bch2_bkey_buf_exit(&last_flushed, trans->c); + bch2_bkey_buf_exit(&last_flushed, c); return ret; } @@ -977,8 +1054,7 @@ int bch2_check_backpointers_to_extents(struct bch_fs *c) } bch2_trans_put(trans); - c->btree_cache.pinned_nodes_leaf_mask = 0; - c->btree_cache.pinned_nodes_interior_mask = 0; + bch2_btree_cache_unpin(c); bch_err_fn(c, ret); return ret; diff --git a/fs/bcachefs/backpointers.h b/fs/bcachefs/backpointers.h index 7daecadb764e..3b29fdf519dd 100644 --- a/fs/bcachefs/backpointers.h +++ b/fs/bcachefs/backpointers.h @@ -134,26 +134,35 @@ static inline enum bch_data_type bch2_bkey_ptr_data_type(struct bkey_s_c k, } } +static inline void __bch2_extent_ptr_to_bp(struct bch_fs *c, struct bch_dev *ca, + enum btree_id btree_id, unsigned level, + struct bkey_s_c k, struct extent_ptr_decoded p, + const union bch_extent_entry *entry, + struct bpos *bucket_pos, struct bch_backpointer *bp, + u64 sectors) +{ + u32 bucket_offset; + *bucket_pos = PTR_BUCKET_POS_OFFSET(ca, &p.ptr, &bucket_offset); + *bp = (struct bch_backpointer) { + .btree_id = btree_id, + .level = level, + .data_type = bch2_bkey_ptr_data_type(k, p, entry), + .bucket_offset = ((u64) bucket_offset << MAX_EXTENT_COMPRESS_RATIO_SHIFT) + + p.crc.offset, + .bucket_len = sectors, + .pos = k.k->p, + }; +} + static inline void bch2_extent_ptr_to_bp(struct bch_fs *c, struct bch_dev *ca, enum btree_id btree_id, unsigned level, struct bkey_s_c k, struct extent_ptr_decoded p, const union bch_extent_entry *entry, struct bpos *bucket_pos, struct bch_backpointer *bp) { - enum bch_data_type data_type = bch2_bkey_ptr_data_type(k, p, entry); - s64 sectors = level ? btree_sectors(c) : k.k->size; - u32 bucket_offset; + u64 sectors = ptr_disk_sectors(level ? btree_sectors(c) : k.k->size, p); - *bucket_pos = PTR_BUCKET_POS_OFFSET(ca, &p.ptr, &bucket_offset); - *bp = (struct bch_backpointer) { - .btree_id = btree_id, - .level = level, - .data_type = data_type, - .bucket_offset = ((u64) bucket_offset << MAX_EXTENT_COMPRESS_RATIO_SHIFT) + - p.crc.offset, - .bucket_len = ptr_disk_sectors(sectors, p), - .pos = k.k->p, - }; + __bch2_extent_ptr_to_bp(c, ca, btree_id, level, k, p, entry, bucket_pos, bp, sectors); } int bch2_get_next_backpointer(struct btree_trans *, struct bch_dev *ca, struct bpos, int, diff --git a/fs/bcachefs/bcachefs.h b/fs/bcachefs/bcachefs.h index 0c7086e00d18..c711d4c27a03 100644 --- a/fs/bcachefs/bcachefs.h +++ b/fs/bcachefs/bcachefs.h @@ -542,7 +542,7 @@ struct bch_dev { * gc_gens_lock, for device resize - holding any is sufficient for * access: Or rcu_read_lock(), but only for dev_ptr_stale(): */ - struct bucket_array __rcu *buckets_gc; + GENRADIX(struct bucket) buckets_gc; struct bucket_gens __rcu *bucket_gens; u8 *oldest_gen; unsigned long *buckets_nouse; @@ -871,6 +871,7 @@ struct bch_fs { /* ALLOCATION */ struct bch_devs_mask rw_devs[BCH_DATA_NR]; + unsigned long rw_devs_change_count; u64 capacity; /* sectors */ u64 reserved; /* sectors */ @@ -1023,6 +1024,7 @@ struct bch_fs { /* fs.c */ struct list_head vfs_inodes_list; struct mutex vfs_inodes_lock; + struct rhashtable vfs_inodes_table; /* VFS IO PATH - fs-io.c */ struct bio_set writepage_bioset; @@ -1044,8 +1046,6 @@ struct bch_fs { * for signaling to the toplevel code which pass we want to run now. */ enum bch_recovery_pass curr_recovery_pass; - /* bitmap of explicitly enabled recovery passes: */ - u64 recovery_passes_explicit; /* bitmask of recovery passes that we actually ran */ u64 recovery_passes_complete; /* never rewinds version of curr_recovery_pass */ @@ -1085,7 +1085,6 @@ struct bch_fs { u64 __percpu *counters; unsigned copy_gc_enabled:1; - bool promote_whole_extents; struct bch2_time_stats times[BCH_TIME_STAT_NR]; @@ -1195,12 +1194,15 @@ static inline bool btree_id_cached(const struct bch_fs *c, enum btree_id btree) static inline struct timespec64 bch2_time_to_timespec(const struct bch_fs *c, s64 time) { struct timespec64 t; + s64 sec; s32 rem; time += c->sb.time_base_lo; - t.tv_sec = div_s64_rem(time, c->sb.time_units_per_sec, &rem); - t.tv_nsec = rem * c->sb.nsec_per_time_unit; + sec = div_s64_rem(time, c->sb.time_units_per_sec, &rem); + + set_normalized_timespec64(&t, sec, rem * (s64)c->sb.nsec_per_time_unit); + return t; } diff --git a/fs/bcachefs/bcachefs_format.h b/fs/bcachefs/bcachefs_format.h index 14ce726bf5a3..8c4addddd07e 100644 --- a/fs/bcachefs/bcachefs_format.h +++ b/fs/bcachefs/bcachefs_format.h @@ -795,6 +795,8 @@ LE64_BITMASK(BCH_SB_HAS_ERRORS, struct bch_sb, flags[0], 60, 61); LE64_BITMASK(BCH_SB_HAS_TOPOLOGY_ERRORS,struct bch_sb, flags[0], 61, 62); LE64_BITMASK(BCH_SB_BIG_ENDIAN, struct bch_sb, flags[0], 62, 63); +LE64_BITMASK(BCH_SB_PROMOTE_WHOLE_EXTENTS, + struct bch_sb, flags[0], 63, 64); LE64_BITMASK(BCH_SB_STR_HASH_TYPE, struct bch_sb, flags[1], 0, 4); LE64_BITMASK(BCH_SB_COMPRESSION_TYPE_LO,struct bch_sb, flags[1], 4, 8); diff --git a/fs/bcachefs/bset.c b/fs/bcachefs/bset.c index 575e1d0b6eeb..d1f6092624d8 100644 --- a/fs/bcachefs/bset.c +++ b/fs/bcachefs/bset.c @@ -304,11 +304,6 @@ struct bkey_float { }; #define BKEY_MANTISSA_BITS 16 -static unsigned bkey_float_byte_offset(unsigned idx) -{ - return idx * sizeof(struct bkey_float); -} - struct ro_aux_tree { u8 nothing[0]; struct bkey_float f[]; @@ -328,8 +323,7 @@ static unsigned bset_aux_tree_buf_end(const struct bset_tree *t) return t->aux_data_offset; case BSET_RO_AUX_TREE: return t->aux_data_offset + - DIV_ROUND_UP(t->size * sizeof(struct bkey_float) + - t->size * sizeof(u8), 8); + DIV_ROUND_UP(t->size * sizeof(struct bkey_float), 8); case BSET_RW_AUX_TREE: return t->aux_data_offset + DIV_ROUND_UP(sizeof(struct rw_aux_tree) * t->size, 8); @@ -360,14 +354,6 @@ static struct ro_aux_tree *ro_aux_tree_base(const struct btree *b, return __aux_tree_base(b, t); } -static u8 *ro_aux_tree_prev(const struct btree *b, - const struct bset_tree *t) -{ - EBUG_ON(bset_aux_tree_type(t) != BSET_RO_AUX_TREE); - - return __aux_tree_base(b, t) + bkey_float_byte_offset(t->size); -} - static struct bkey_float *bkey_float(const struct btree *b, const struct bset_tree *t, unsigned idx) @@ -479,15 +465,6 @@ static inline struct bkey_packed *tree_to_bkey(const struct btree *b, bkey_float(b, t, j)->key_offset); } -static struct bkey_packed *tree_to_prev_bkey(const struct btree *b, - const struct bset_tree *t, - unsigned j) -{ - unsigned prev_u64s = ro_aux_tree_prev(b, t)[j]; - - return (void *) ((u64 *) tree_to_bkey(b, t, j)->_data - prev_u64s); -} - static struct rw_aux_tree *rw_aux_tree(const struct btree *b, const struct bset_tree *t) { @@ -585,8 +562,7 @@ static unsigned rw_aux_tree_bsearch(struct btree *b, } static inline unsigned bkey_mantissa(const struct bkey_packed *k, - const struct bkey_float *f, - unsigned idx) + const struct bkey_float *f) { u64 v; @@ -617,7 +593,7 @@ static __always_inline void make_bfloat(struct btree *b, struct bset_tree *t, struct bkey_packed *m = tree_to_bkey(b, t, j); struct bkey_packed *l = is_power_of_2(j) ? min_key - : tree_to_prev_bkey(b, t, j >> ffs(j)); + : tree_to_bkey(b, t, j >> ffs(j)); struct bkey_packed *r = is_power_of_2(j + 1) ? max_key : tree_to_bkey(b, t, j >> (ffz(j) + 1)); @@ -668,7 +644,7 @@ static __always_inline void make_bfloat(struct btree *b, struct bset_tree *t, EBUG_ON(shift < 0 || shift >= BFLOAT_FAILED); f->exponent = shift; - mantissa = bkey_mantissa(m, f, j); + mantissa = bkey_mantissa(m, f); /* * If we've got garbage bits, set them to all 1s - it's legal for the @@ -690,8 +666,7 @@ static unsigned __bset_tree_capacity(struct btree *b, const struct bset_tree *t) static unsigned bset_ro_tree_capacity(struct btree *b, const struct bset_tree *t) { - return __bset_tree_capacity(b, t) / - (sizeof(struct bkey_float) + sizeof(u8)); + return __bset_tree_capacity(b, t) / sizeof(struct bkey_float); } static unsigned bset_rw_tree_capacity(struct btree *b, const struct bset_tree *t) @@ -720,7 +695,7 @@ static noinline void __build_rw_aux_tree(struct btree *b, struct bset_tree *t) static noinline void __build_ro_aux_tree(struct btree *b, struct bset_tree *t) { - struct bkey_packed *prev = NULL, *k = btree_bkey_first(b, t); + struct bkey_packed *k = btree_bkey_first(b, t); struct bkey_i min_key, max_key; unsigned cacheline = 1; @@ -733,12 +708,12 @@ retry: return; } - t->extra = (t->size - rounddown_pow_of_two(t->size - 1)) << 1; + t->extra = eytzinger1_extra(t->size - 1); /* First we figure out where the first key in each cacheline is */ eytzinger1_for_each(j, t->size - 1) { while (bkey_to_cacheline(b, t, k) < cacheline) - prev = k, k = bkey_p_next(k); + k = bkey_p_next(k); if (k >= btree_bkey_last(b, t)) { /* XXX: this path sucks */ @@ -746,17 +721,12 @@ retry: goto retry; } - ro_aux_tree_prev(b, t)[j] = prev->u64s; bkey_float(b, t, j)->key_offset = bkey_to_cacheline_offset(b, t, cacheline++, k); - EBUG_ON(tree_to_prev_bkey(b, t, j) != prev); EBUG_ON(tree_to_bkey(b, t, j) != k); } - while (k != btree_bkey_last(b, t)) - prev = k, k = bkey_p_next(k); - if (!bkey_pack_pos(bkey_to_packed(&min_key), b->data->min_key, b)) { bkey_init(&min_key.k); min_key.k.p = b->data->min_key; @@ -915,66 +885,18 @@ struct bkey_packed *bch2_bkey_prev_filter(struct btree *b, /* Insert */ -static void bch2_bset_fix_lookup_table(struct btree *b, - struct bset_tree *t, - struct bkey_packed *_where, - unsigned clobber_u64s, - unsigned new_u64s) +static void rw_aux_tree_insert_entry(struct btree *b, + struct bset_tree *t, + unsigned idx) { - int shift = new_u64s - clobber_u64s; - unsigned l, j, where = __btree_node_key_to_offset(b, _where); - - EBUG_ON(bset_has_ro_aux_tree(t)); - - if (!bset_has_rw_aux_tree(t)) - return; - - /* returns first entry >= where */ - l = rw_aux_tree_bsearch(b, t, where); - - if (!l) /* never delete first entry */ - l++; - else if (l < t->size && - where < t->end_offset && - rw_aux_tree(b, t)[l].offset == where) - rw_aux_tree_set(b, t, l++, _where); - - /* l now > where */ - - for (j = l; - j < t->size && - rw_aux_tree(b, t)[j].offset < where + clobber_u64s; - j++) - ; - - if (j < t->size && - rw_aux_tree(b, t)[j].offset + shift == - rw_aux_tree(b, t)[l - 1].offset) - j++; - - memmove(&rw_aux_tree(b, t)[l], - &rw_aux_tree(b, t)[j], - (void *) &rw_aux_tree(b, t)[t->size] - - (void *) &rw_aux_tree(b, t)[j]); - t->size -= j - l; - - for (j = l; j < t->size; j++) - rw_aux_tree(b, t)[j].offset += shift; - - EBUG_ON(l < t->size && - rw_aux_tree(b, t)[l].offset == - rw_aux_tree(b, t)[l - 1].offset); + EBUG_ON(!idx || idx > t->size); + struct bkey_packed *start = rw_aux_to_bkey(b, t, idx - 1); + struct bkey_packed *end = idx < t->size + ? rw_aux_to_bkey(b, t, idx) + : btree_bkey_last(b, t); if (t->size < bset_rw_tree_capacity(b, t) && - (l < t->size - ? rw_aux_tree(b, t)[l].offset - : t->end_offset) - - rw_aux_tree(b, t)[l - 1].offset > - L1_CACHE_BYTES / sizeof(u64)) { - struct bkey_packed *start = rw_aux_to_bkey(b, t, l - 1); - struct bkey_packed *end = l < t->size - ? rw_aux_to_bkey(b, t, l) - : btree_bkey_last(b, t); + (void *) end - (void *) start > L1_CACHE_BYTES) { struct bkey_packed *k = start; while (1) { @@ -983,23 +905,78 @@ static void bch2_bset_fix_lookup_table(struct btree *b, break; if ((void *) k - (void *) start >= L1_CACHE_BYTES) { - memmove(&rw_aux_tree(b, t)[l + 1], - &rw_aux_tree(b, t)[l], + memmove(&rw_aux_tree(b, t)[idx + 1], + &rw_aux_tree(b, t)[idx], (void *) &rw_aux_tree(b, t)[t->size] - - (void *) &rw_aux_tree(b, t)[l]); + (void *) &rw_aux_tree(b, t)[idx]); t->size++; - rw_aux_tree_set(b, t, l, k); + rw_aux_tree_set(b, t, idx, k); break; } } } +} +static void bch2_bset_fix_lookup_table(struct btree *b, + struct bset_tree *t, + struct bkey_packed *_where, + unsigned clobber_u64s, + unsigned new_u64s) +{ + int shift = new_u64s - clobber_u64s; + unsigned idx, j, where = __btree_node_key_to_offset(b, _where); + + EBUG_ON(bset_has_ro_aux_tree(t)); + + if (!bset_has_rw_aux_tree(t)) + return; + + if (where > rw_aux_tree(b, t)[t->size - 1].offset) { + rw_aux_tree_insert_entry(b, t, t->size); + goto verify; + } + + /* returns first entry >= where */ + idx = rw_aux_tree_bsearch(b, t, where); + + if (rw_aux_tree(b, t)[idx].offset == where) { + if (!idx) { /* never delete first entry */ + idx++; + } else if (where < t->end_offset) { + rw_aux_tree_set(b, t, idx++, _where); + } else { + EBUG_ON(where != t->end_offset); + rw_aux_tree_insert_entry(b, t, --t->size); + goto verify; + } + } + + EBUG_ON(idx < t->size && rw_aux_tree(b, t)[idx].offset <= where); + if (idx < t->size && + rw_aux_tree(b, t)[idx].offset + shift == + rw_aux_tree(b, t)[idx - 1].offset) { + memmove(&rw_aux_tree(b, t)[idx], + &rw_aux_tree(b, t)[idx + 1], + (void *) &rw_aux_tree(b, t)[t->size] - + (void *) &rw_aux_tree(b, t)[idx + 1]); + t->size -= 1; + } + + for (j = idx; j < t->size; j++) + rw_aux_tree(b, t)[j].offset += shift; + + EBUG_ON(idx < t->size && + rw_aux_tree(b, t)[idx].offset == + rw_aux_tree(b, t)[idx - 1].offset); + + rw_aux_tree_insert_entry(b, t, idx); + +verify: bch2_bset_verify_rw_aux_tree(b, t); bset_aux_tree_verify(b); } void bch2_bset_insert(struct btree *b, - struct btree_node_iter *iter, struct bkey_packed *where, struct bkey_i *insert, unsigned clobber_u64s) @@ -1098,8 +1075,7 @@ static inline void prefetch_four_cachelines(void *p) } static inline bool bkey_mantissa_bits_dropped(const struct btree *b, - const struct bkey_float *f, - unsigned idx) + const struct bkey_float *f) { #if __BYTE_ORDER__ == __ORDER_LITTLE_ENDIAN__ unsigned key_bits_start = b->format.key_u64s * 64 - b->nr_key_bits; @@ -1133,9 +1109,9 @@ static struct bkey_packed *bset_search_tree(const struct btree *b, goto slowpath; l = f->mantissa; - r = bkey_mantissa(packed_search, f, n); + r = bkey_mantissa(packed_search, f); - if (unlikely(l == r) && bkey_mantissa_bits_dropped(b, f, n)) + if (unlikely(l == r) && bkey_mantissa_bits_dropped(b, f)) goto slowpath; n = n * 2 + (l < r); diff --git a/fs/bcachefs/bset.h b/fs/bcachefs/bset.h index 5c6c7a14fa0f..6953d55b72cc 100644 --- a/fs/bcachefs/bset.h +++ b/fs/bcachefs/bset.h @@ -270,8 +270,8 @@ void bch2_bset_init_first(struct btree *, struct bset *); void bch2_bset_init_next(struct btree *, struct btree_node_entry *); void bch2_bset_build_aux_tree(struct btree *, struct bset_tree *, bool); -void bch2_bset_insert(struct btree *, struct btree_node_iter *, - struct bkey_packed *, struct bkey_i *, unsigned); +void bch2_bset_insert(struct btree *, struct bkey_packed *, struct bkey_i *, + unsigned); void bch2_bset_delete(struct btree *, struct bkey_packed *, unsigned); /* Bkey utility code */ diff --git a/fs/bcachefs/btree_cache.c b/fs/bcachefs/btree_cache.c index e52a06d3418c..6e4afb2b5441 100644 --- a/fs/bcachefs/btree_cache.c +++ b/fs/bcachefs/btree_cache.c @@ -15,11 +15,12 @@ #include #include +#include #define BTREE_CACHE_NOT_FREED_INCREMENT(counter) \ do { \ if (shrinker_counter) \ - bc->not_freed_##counter++; \ + bc->not_freed[BCH_BTREE_CACHE_NOT_FREED_##counter]++; \ } while (0) const char * const bch2_btree_node_flags[] = { @@ -31,24 +32,29 @@ const char * const bch2_btree_node_flags[] = { void bch2_recalc_btree_reserve(struct bch_fs *c) { - unsigned i, reserve = 16; + unsigned reserve = 16; if (!c->btree_roots_known[0].b) reserve += 8; - for (i = 0; i < btree_id_nr_alive(c); i++) { + for (unsigned i = 0; i < btree_id_nr_alive(c); i++) { struct btree_root *r = bch2_btree_id_root(c, i); if (r->b) reserve += min_t(unsigned, 1, r->b->c.level) * 8; } - c->btree_cache.reserve = reserve; + c->btree_cache.nr_reserve = reserve; } -static inline unsigned btree_cache_can_free(struct btree_cache *bc) +static inline size_t btree_cache_can_free(struct btree_cache_list *list) { - return max_t(int, 0, bc->used - bc->reserve); + struct btree_cache *bc = container_of(list, struct btree_cache, live[list->idx]); + + size_t can_free = list->nr; + if (!list->idx) + can_free = max_t(ssize_t, 0, can_free - bc->nr_reserve); + return can_free; } static void btree_node_to_freedlist(struct btree_cache *bc, struct btree *b) @@ -63,6 +69,18 @@ static void btree_node_data_free(struct bch_fs *c, struct btree *b) { struct btree_cache *bc = &c->btree_cache; + BUG_ON(btree_node_hashed(b)); + + /* + * This should really be done in slub/vmalloc, but we're using the + * kmalloc_large() path, so we're working around a slub bug by doing + * this here: + */ + if (b->data) + mm_account_reclaimed_pages(btree_buf_bytes(b) / PAGE_SIZE); + if (b->aux_data) + mm_account_reclaimed_pages(btree_aux_data_bytes(b) / PAGE_SIZE); + EBUG_ON(btree_node_write_in_flight(b)); clear_btree_node_just_written(b); @@ -76,7 +94,7 @@ static void btree_node_data_free(struct bch_fs *c, struct btree *b) #endif b->aux_data = NULL; - bc->used--; + bc->nr_freeable--; btree_node_to_freedlist(bc, b); } @@ -102,6 +120,8 @@ static int btree_node_data_alloc(struct bch_fs *c, struct btree *b, gfp_t gfp) { BUG_ON(b->data || b->aux_data); + gfp |= __GFP_ACCOUNT|__GFP_RECLAIMABLE; + b->data = kvmalloc(btree_buf_bytes(b), gfp); if (!b->data) return -BCH_ERR_ENOMEM_btree_node_mem_alloc; @@ -154,7 +174,7 @@ struct btree *__bch2_btree_node_mem_alloc(struct bch_fs *c) bch2_btree_lock_init(&b->c, 0); - bc->used++; + bc->nr_freeable++; list_add(&b->list, &bc->freeable); return b; } @@ -169,10 +189,56 @@ void bch2_btree_node_to_freelist(struct bch_fs *c, struct btree *b) six_unlock_intent(&b->c.lock); } +static inline bool __btree_node_pinned(struct btree_cache *bc, struct btree *b) +{ + struct bbpos pos = BBPOS(b->c.btree_id, b->key.k.p); + + u64 mask = bc->pinned_nodes_mask[!!b->c.level]; + + return ((mask & BIT_ULL(b->c.btree_id)) && + bbpos_cmp(bc->pinned_nodes_start, pos) < 0 && + bbpos_cmp(bc->pinned_nodes_end, pos) >= 0); +} + +void bch2_node_pin(struct bch_fs *c, struct btree *b) +{ + struct btree_cache *bc = &c->btree_cache; + + mutex_lock(&bc->lock); + BUG_ON(!__btree_node_pinned(bc, b)); + if (b != btree_node_root(c, b) && !btree_node_pinned(b)) { + set_btree_node_pinned(b); + list_move(&b->list, &bc->live[1].list); + bc->live[0].nr--; + bc->live[1].nr++; + } + mutex_unlock(&bc->lock); +} + +void bch2_btree_cache_unpin(struct bch_fs *c) +{ + struct btree_cache *bc = &c->btree_cache; + struct btree *b, *n; + + mutex_lock(&bc->lock); + c->btree_cache.pinned_nodes_mask[0] = 0; + c->btree_cache.pinned_nodes_mask[1] = 0; + + list_for_each_entry_safe(b, n, &bc->live[1].list, list) { + clear_btree_node_pinned(b); + list_move(&b->list, &bc->live[0].list); + bc->live[0].nr++; + bc->live[1].nr--; + } + + mutex_unlock(&bc->lock); +} + /* Btree in memory cache - hash table */ void bch2_btree_node_hash_remove(struct btree_cache *bc, struct btree *b) { + lockdep_assert_held(&bc->lock); int ret = rhashtable_remove_fast(&bc->table, &b->hash, bch_btree_cache_params); BUG_ON(ret); @@ -181,7 +247,11 @@ void bch2_btree_node_hash_remove(struct btree_cache *bc, struct btree *b) b->hash_val = 0; if (b->c.btree_id < BTREE_ID_NR) - --bc->used_by_btree[b->c.btree_id]; + --bc->nr_by_btree[b->c.btree_id]; + + bc->live[btree_node_pinned(b)].nr--; + bc->nr_freeable++; + list_move(&b->list, &bc->freeable); } int __bch2_btree_node_hash_insert(struct btree_cache *bc, struct btree *b) @@ -191,23 +261,30 @@ int __bch2_btree_node_hash_insert(struct btree_cache *bc, struct btree *b) int ret = rhashtable_lookup_insert_fast(&bc->table, &b->hash, bch_btree_cache_params); - if (!ret && b->c.btree_id < BTREE_ID_NR) - bc->used_by_btree[b->c.btree_id]++; - return ret; + if (ret) + return ret; + + if (b->c.btree_id < BTREE_ID_NR) + bc->nr_by_btree[b->c.btree_id]++; + + bool p = __btree_node_pinned(bc, b); + mod_bit(BTREE_NODE_pinned, &b->flags, p); + + list_move_tail(&b->list, &bc->live[p].list); + bc->live[p].nr++; + + bc->nr_freeable--; + return 0; } int bch2_btree_node_hash_insert(struct btree_cache *bc, struct btree *b, unsigned level, enum btree_id id) { - int ret; - b->c.level = level; b->c.btree_id = id; mutex_lock(&bc->lock); - ret = __bch2_btree_node_hash_insert(bc, b); - if (!ret) - list_add_tail(&b->list, &bc->live); + int ret = __bch2_btree_node_hash_insert(bc, b); mutex_unlock(&bc->lock); return ret; @@ -261,18 +338,6 @@ static int __btree_node_reclaim(struct bch_fs *c, struct btree *b, bool flush, b int ret = 0; lockdep_assert_held(&bc->lock); - - struct bbpos pos = BBPOS(b->c.btree_id, b->key.k.p); - - u64 mask = b->c.level - ? bc->pinned_nodes_interior_mask - : bc->pinned_nodes_leaf_mask; - - if ((mask & BIT_ULL(b->c.btree_id)) && - bbpos_cmp(bc->pinned_nodes_start, pos) < 0 && - bbpos_cmp(bc->pinned_nodes_end, pos) >= 0) - return -BCH_ERR_ENOMEM_btree_node_reclaim; - wait_on_io: if (b->flags & ((1U << BTREE_NODE_dirty)| (1U << BTREE_NODE_read_in_flight)| @@ -377,8 +442,9 @@ static int btree_node_write_and_reclaim(struct bch_fs *c, struct btree *b) static unsigned long bch2_btree_cache_scan(struct shrinker *shrink, struct shrink_control *sc) { - struct bch_fs *c = shrink->private_data; - struct btree_cache *bc = &c->btree_cache; + struct btree_cache_list *list = shrink->private_data; + struct btree_cache *bc = container_of(list, struct btree_cache, live[list->idx]); + struct bch_fs *c = container_of(bc, struct bch_fs, btree_cache); struct btree *b, *t; unsigned long nr = sc->nr_to_scan; unsigned long can_free = 0; @@ -386,8 +452,7 @@ static unsigned long bch2_btree_cache_scan(struct shrinker *shrink, unsigned long touched = 0; unsigned i, flags; unsigned long ret = SHRINK_STOP; - bool trigger_writes = atomic_read(&bc->dirty) + nr >= - bc->used * 3 / 4; + bool trigger_writes = atomic_long_read(&bc->nr_dirty) + nr >= list->nr * 3 / 4; if (bch2_btree_shrinker_disabled) return SHRINK_STOP; @@ -402,7 +467,7 @@ static unsigned long bch2_btree_cache_scan(struct shrinker *shrink, * succeed, so that inserting keys into the btree can always succeed and * IO can always make forward progress: */ - can_free = btree_cache_can_free(bc); + can_free = btree_cache_can_free(list); nr = min_t(unsigned long, nr, can_free); i = 0; @@ -424,22 +489,24 @@ static unsigned long bch2_btree_cache_scan(struct shrinker *shrink, six_unlock_write(&b->c.lock); six_unlock_intent(&b->c.lock); freed++; - bc->freed++; + bc->nr_freed++; } } restart: - list_for_each_entry_safe(b, t, &bc->live, list) { + list_for_each_entry_safe(b, t, &list->list, list) { touched++; if (btree_node_accessed(b)) { clear_btree_node_accessed(b); - bc->not_freed_access_bit++; + bc->not_freed[BCH_BTREE_CACHE_NOT_FREED_access_bit]++; + --touched;; } else if (!btree_node_reclaim(c, b, true)) { + bch2_btree_node_hash_remove(bc, b); + freed++; btree_node_data_free(c, b); - bc->freed++; + bc->nr_freed++; - bch2_btree_node_hash_remove(bc, b); six_unlock_write(&b->c.lock); six_unlock_intent(&b->c.lock); @@ -450,7 +517,7 @@ restart: !btree_node_will_make_reachable(b) && !btree_node_write_blocked(b) && six_trylock_read(&b->c.lock)) { - list_move(&bc->live, &b->list); + list_move(&list->list, &b->list); mutex_unlock(&bc->lock); __bch2_btree_node_write(c, b, BTREE_WRITE_cache_reclaim); six_unlock_read(&b->c.lock); @@ -464,8 +531,8 @@ restart: break; } out_rotate: - if (&t->list != &bc->live) - list_move_tail(&bc->live, &t->list); + if (&t->list != &list->list) + list_move_tail(&list->list, &t->list); out: mutex_unlock(&bc->lock); out_nounlock: @@ -478,44 +545,45 @@ out_nounlock: static unsigned long bch2_btree_cache_count(struct shrinker *shrink, struct shrink_control *sc) { - struct bch_fs *c = shrink->private_data; - struct btree_cache *bc = &c->btree_cache; + struct btree_cache_list *list = shrink->private_data; if (bch2_btree_shrinker_disabled) return 0; - return btree_cache_can_free(bc); + return btree_cache_can_free(list); } void bch2_fs_btree_cache_exit(struct bch_fs *c) { struct btree_cache *bc = &c->btree_cache; - struct btree *b; - unsigned i, flags; + struct btree *b, *t; + unsigned long flags; - shrinker_free(bc->shrink); + shrinker_free(bc->live[1].shrink); + shrinker_free(bc->live[0].shrink); /* vfree() can allocate memory: */ flags = memalloc_nofs_save(); mutex_lock(&bc->lock); if (c->verify_data) - list_move(&c->verify_data->list, &bc->live); + list_move(&c->verify_data->list, &bc->live[0].list); kvfree(c->verify_ondisk); - for (i = 0; i < btree_id_nr_alive(c); i++) { + for (unsigned i = 0; i < btree_id_nr_alive(c); i++) { struct btree_root *r = bch2_btree_id_root(c, i); if (r->b) - list_add(&r->b->list, &bc->live); + list_add(&r->b->list, &bc->live[0].list); } - list_splice(&bc->freeable, &bc->live); - - while (!list_empty(&bc->live)) { - b = list_first_entry(&bc->live, struct btree, list); + list_for_each_entry_safe(b, t, &bc->live[1].list, list) + bch2_btree_node_hash_remove(bc, b); + list_for_each_entry_safe(b, t, &bc->live[0].list, list) + bch2_btree_node_hash_remove(bc, b); + list_for_each_entry_safe(b, t, &bc->freeable, list) { BUG_ON(btree_node_read_in_flight(b) || btree_node_write_in_flight(b)); @@ -523,12 +591,11 @@ void bch2_fs_btree_cache_exit(struct bch_fs *c) } BUG_ON(!bch2_journal_error(&c->journal) && - atomic_read(&c->btree_cache.dirty)); + atomic_long_read(&c->btree_cache.nr_dirty)); list_splice(&bc->freed_pcpu, &bc->freed_nonpcpu); - while (!list_empty(&bc->freed_nonpcpu)) { - b = list_first_entry(&bc->freed_nonpcpu, struct btree, list); + list_for_each_entry_safe(b, t, &bc->freed_nonpcpu, list) { list_del(&b->list); six_lock_exit(&b->c.lock); kfree(b); @@ -537,6 +604,12 @@ void bch2_fs_btree_cache_exit(struct bch_fs *c) mutex_unlock(&bc->lock); memalloc_nofs_restore(flags); + for (unsigned i = 0; i < ARRAY_SIZE(bc->nr_by_btree); i++) + BUG_ON(bc->nr_by_btree[i]); + BUG_ON(bc->live[0].nr); + BUG_ON(bc->live[1].nr); + BUG_ON(bc->nr_freeable); + if (bc->table_init_done) rhashtable_destroy(&bc->table); } @@ -556,22 +629,32 @@ int bch2_fs_btree_cache_init(struct bch_fs *c) bch2_recalc_btree_reserve(c); - for (i = 0; i < bc->reserve; i++) + for (i = 0; i < bc->nr_reserve; i++) if (!__bch2_btree_node_mem_alloc(c)) goto err; - list_splice_init(&bc->live, &bc->freeable); + list_splice_init(&bc->live[0].list, &bc->freeable); mutex_init(&c->verify_lock); shrink = shrinker_alloc(0, "%s-btree_cache", c->name); if (!shrink) goto err; - bc->shrink = shrink; + bc->live[0].shrink = shrink; shrink->count_objects = bch2_btree_cache_count; shrink->scan_objects = bch2_btree_cache_scan; - shrink->seeks = 4; - shrink->private_data = c; + shrink->seeks = 2; + shrink->private_data = &bc->live[0]; + shrinker_register(shrink); + + shrink = shrinker_alloc(0, "%s-btree_cache-pinned", c->name); + if (!shrink) + goto err; + bc->live[1].shrink = shrink; + shrink->count_objects = bch2_btree_cache_count; + shrink->scan_objects = bch2_btree_cache_scan; + shrink->seeks = 8; + shrink->private_data = &bc->live[1]; shrinker_register(shrink); return 0; @@ -582,7 +665,10 @@ err: void bch2_fs_btree_cache_init_early(struct btree_cache *bc) { mutex_init(&bc->lock); - INIT_LIST_HEAD(&bc->live); + for (unsigned i = 0; i < ARRAY_SIZE(bc->live); i++) { + bc->live[i].idx = i; + INIT_LIST_HEAD(&bc->live[i].list); + } INIT_LIST_HEAD(&bc->freeable); INIT_LIST_HEAD(&bc->freed_pcpu); INIT_LIST_HEAD(&bc->freed_nonpcpu); @@ -644,14 +730,16 @@ static struct btree *btree_node_cannibalize(struct bch_fs *c) struct btree_cache *bc = &c->btree_cache; struct btree *b; - list_for_each_entry_reverse(b, &bc->live, list) - if (!btree_node_reclaim(c, b, false)) - return b; + for (unsigned i = 0; i < ARRAY_SIZE(bc->live); i++) + list_for_each_entry_reverse(b, &bc->live[i].list, list) + if (!btree_node_reclaim(c, b, false)) + return b; while (1) { - list_for_each_entry_reverse(b, &bc->live, list) - if (!btree_node_write_and_reclaim(c, b)) - return b; + for (unsigned i = 0; i < ARRAY_SIZE(bc->live); i++) + list_for_each_entry_reverse(b, &bc->live[i].list, list) + if (!btree_node_write_and_reclaim(c, b)) + return b; /* * Rare case: all nodes were intent-locked. @@ -671,9 +759,7 @@ struct btree *bch2_btree_node_mem_alloc(struct btree_trans *trans, bool pcpu_rea : &bc->freed_nonpcpu; struct btree *b, *b2; u64 start_time = local_clock(); - unsigned flags; - flags = memalloc_nofs_save(); mutex_lock(&bc->lock); /* @@ -725,7 +811,7 @@ got_node: } mutex_lock(&bc->lock); - bc->used++; + bc->nr_freeable++; got_mem: mutex_unlock(&bc->lock); @@ -745,8 +831,6 @@ out: bch2_time_stats_update(&c->times[BCH_TIME_btree_node_mem_alloc], start_time); - memalloc_nofs_restore(flags); - int ret = bch2_trans_relock(trans); if (unlikely(ret)) { bch2_btree_node_to_freelist(c, b); @@ -781,7 +865,6 @@ err: } mutex_unlock(&bc->lock); - memalloc_nofs_restore(flags); return ERR_PTR(-BCH_ERR_ENOMEM_btree_node_mem_alloc); } @@ -1269,8 +1352,8 @@ wait_on_io: BUG_ON(btree_node_dirty(b)); mutex_lock(&bc->lock); - btree_node_data_free(c, b); bch2_btree_node_hash_remove(bc, b); + btree_node_data_free(c, b); mutex_unlock(&bc->lock); out: six_unlock_write(&b->c.lock); @@ -1342,13 +1425,20 @@ void bch2_btree_node_to_text(struct printbuf *out, struct bch_fs *c, const struc } static void prt_btree_cache_line(struct printbuf *out, const struct bch_fs *c, - const char *label, unsigned nr) + const char *label, size_t nr) { prt_printf(out, "%s\t", label); prt_human_readable_u64(out, nr * c->opts.btree_node_size); - prt_printf(out, " (%u)\n", nr); + prt_printf(out, " (%zu)\n", nr); } +static const char * const bch2_btree_cache_not_freed_reasons_strs[] = { +#define x(n) #n, + BCH_BTREE_CACHE_NOT_FREED_REASONS() +#undef x + NULL +}; + void bch2_btree_cache_to_text(struct printbuf *out, const struct btree_cache *bc) { struct bch_fs *c = container_of(bc, struct bch_fs, btree_cache); @@ -1356,24 +1446,21 @@ void bch2_btree_cache_to_text(struct printbuf *out, const struct btree_cache *bc if (!out->nr_tabstops) printbuf_tabstop_push(out, 32); - prt_btree_cache_line(out, c, "total:", bc->used); - prt_btree_cache_line(out, c, "nr dirty:", atomic_read(&bc->dirty)); + prt_btree_cache_line(out, c, "live:", bc->live[0].nr); + prt_btree_cache_line(out, c, "pinned:", bc->live[1].nr); + prt_btree_cache_line(out, c, "freeable:", bc->nr_freeable); + prt_btree_cache_line(out, c, "dirty:", atomic_long_read(&bc->nr_dirty)); prt_printf(out, "cannibalize lock:\t%p\n", bc->alloc_lock); prt_newline(out); - for (unsigned i = 0; i < ARRAY_SIZE(bc->used_by_btree); i++) - prt_btree_cache_line(out, c, bch2_btree_id_str(i), bc->used_by_btree[i]); + for (unsigned i = 0; i < ARRAY_SIZE(bc->nr_by_btree); i++) + prt_btree_cache_line(out, c, bch2_btree_id_str(i), bc->nr_by_btree[i]); prt_newline(out); - prt_printf(out, "freed:\t%u\n", bc->freed); + prt_printf(out, "freed:\t%zu\n", bc->nr_freed); prt_printf(out, "not freed:\n"); - prt_printf(out, " dirty\t%u\n", bc->not_freed_dirty); - prt_printf(out, " write in flight\t%u\n", bc->not_freed_write_in_flight); - prt_printf(out, " read in flight\t%u\n", bc->not_freed_read_in_flight); - prt_printf(out, " lock intent failed\t%u\n", bc->not_freed_lock_intent); - prt_printf(out, " lock write failed\t%u\n", bc->not_freed_lock_write); - prt_printf(out, " access bit\t%u\n", bc->not_freed_access_bit); - prt_printf(out, " no evict failed\t%u\n", bc->not_freed_noevict); - prt_printf(out, " write blocked\t%u\n", bc->not_freed_write_blocked); - prt_printf(out, " will make reachable\t%u\n", bc->not_freed_will_make_reachable); + + for (unsigned i = 0; i < ARRAY_SIZE(bc->not_freed); i++) + prt_printf(out, " %s\t%llu\n", + bch2_btree_cache_not_freed_reasons_strs[i], bc->not_freed[i]); } diff --git a/fs/bcachefs/btree_cache.h b/fs/bcachefs/btree_cache.h index f82064007127..367acd217c6a 100644 --- a/fs/bcachefs/btree_cache.h +++ b/fs/bcachefs/btree_cache.h @@ -19,6 +19,9 @@ int __bch2_btree_node_hash_insert(struct btree_cache *, struct btree *); int bch2_btree_node_hash_insert(struct btree_cache *, struct btree *, unsigned, enum btree_id); +void bch2_node_pin(struct bch_fs *, struct btree *); +void bch2_btree_cache_unpin(struct bch_fs *); + void bch2_btree_node_update_key_early(struct btree_trans *, enum btree_id, unsigned, struct bkey_s_c, struct bkey_i *); diff --git a/fs/bcachefs/btree_gc.c b/fs/bcachefs/btree_gc.c index eb3002c4eae7..b5e0692f03c6 100644 --- a/fs/bcachefs/btree_gc.c +++ b/fs/bcachefs/btree_gc.c @@ -549,9 +549,8 @@ reconstruct_root: six_unlock_read(&b->c.lock); if (ret == DROP_THIS_NODE) { - bch2_btree_node_hash_remove(&c->btree_cache, b); mutex_lock(&c->btree_cache.lock); - list_move(&b->list, &c->btree_cache.freeable); + bch2_btree_node_hash_remove(&c->btree_cache, b); mutex_unlock(&c->btree_cache.lock); r->b = NULL; @@ -753,10 +752,8 @@ static void bch2_gc_free(struct bch_fs *c) genradix_free(&c->reflink_gc_table); genradix_free(&c->gc_stripes); - for_each_member_device(c, ca) { - kvfree(rcu_dereference_protected(ca->buckets_gc, 1)); - ca->buckets_gc = NULL; - } + for_each_member_device(c, ca) + genradix_free(&ca->buckets_gc); } static int bch2_gc_start(struct bch_fs *c) @@ -910,20 +907,12 @@ static int bch2_gc_alloc_start(struct bch_fs *c) int ret = 0; for_each_member_device(c, ca) { - struct bucket_array *buckets = kvmalloc(sizeof(struct bucket_array) + - ca->mi.nbuckets * sizeof(struct bucket), - GFP_KERNEL|__GFP_ZERO); - if (!buckets) { + ret = genradix_prealloc(&ca->buckets_gc, ca->mi.nbuckets, GFP_KERNEL); + if (ret) { bch2_dev_put(ca); ret = -BCH_ERR_ENOMEM_gc_alloc_start; break; } - - buckets->first_bucket = ca->mi.first_bucket; - buckets->nbuckets = ca->mi.nbuckets; - buckets->nbuckets_minus_first = - buckets->nbuckets - buckets->first_bucket; - rcu_assign_pointer(ca->buckets_gc, buckets); } bch_err_fn(c, ret); diff --git a/fs/bcachefs/btree_io.c b/fs/bcachefs/btree_io.c index 56ea9a77cd4a..cb48a9477514 100644 --- a/fs/bcachefs/btree_io.c +++ b/fs/bcachefs/btree_io.c @@ -1666,7 +1666,7 @@ void bch2_btree_node_read(struct btree_trans *trans, struct btree *b, bch2_btree_pos_to_text(&buf, c, b); bch_err_ratelimited(c, "%s", buf.buf); - if (c->recovery_passes_explicit & BIT_ULL(BCH_RECOVERY_PASS_check_topology) && + if (c->opts.recovery_passes & BIT_ULL(BCH_RECOVERY_PASS_check_topology) && c->curr_recovery_pass > BCH_RECOVERY_PASS_check_topology) bch2_fatal_error(c); @@ -1749,10 +1749,8 @@ static int __bch2_btree_root_read(struct btree_trans *trans, enum btree_id id, bch2_btree_node_read(trans, b, true); if (btree_node_read_error(b)) { - bch2_btree_node_hash_remove(&c->btree_cache, b); - mutex_lock(&c->btree_cache.lock); - list_move(&b->list, &c->btree_cache.freeable); + bch2_btree_node_hash_remove(&c->btree_cache, b); mutex_unlock(&c->btree_cache.lock); ret = -BCH_ERR_btree_node_read_error; @@ -2031,7 +2029,7 @@ void __bch2_btree_node_write(struct bch_fs *c, struct btree *b, unsigned flags) do_write: BUG_ON((type == BTREE_WRITE_initial) != (b->written == 0)); - atomic_dec(&c->btree_cache.dirty); + atomic_long_dec(&c->btree_cache.nr_dirty); BUG_ON(btree_node_fake(b)); BUG_ON((b->will_make_reachable != 0) != !b->written); diff --git a/fs/bcachefs/btree_io.h b/fs/bcachefs/btree_io.h index 63d76f5c6403..9b01ca3de907 100644 --- a/fs/bcachefs/btree_io.h +++ b/fs/bcachefs/btree_io.h @@ -18,13 +18,13 @@ struct btree_node_read_all; static inline void set_btree_node_dirty_acct(struct bch_fs *c, struct btree *b) { if (!test_and_set_bit(BTREE_NODE_dirty, &b->flags)) - atomic_inc(&c->btree_cache.dirty); + atomic_long_inc(&c->btree_cache.nr_dirty); } static inline void clear_btree_node_dirty_acct(struct bch_fs *c, struct btree *b) { if (test_and_clear_bit(BTREE_NODE_dirty, &b->flags)) - atomic_dec(&c->btree_cache.dirty); + atomic_long_dec(&c->btree_cache.nr_dirty); } static inline unsigned btree_ptr_sectors_written(struct bkey_s_c k) diff --git a/fs/bcachefs/btree_iter.c b/fs/bcachefs/btree_iter.c index 2e84d22e17bd..bfe9f0c1e1be 100644 --- a/fs/bcachefs/btree_iter.c +++ b/fs/bcachefs/btree_iter.c @@ -1010,9 +1010,9 @@ retry_all: * the same position: */ if (trans->paths[idx].uptodate) { - __btree_path_get(&trans->paths[idx], false); + __btree_path_get(trans, &trans->paths[idx], false); ret = bch2_btree_path_traverse_one(trans, idx, 0, _THIS_IP_); - __btree_path_put(&trans->paths[idx], false); + __btree_path_put(trans, &trans->paths[idx], false); if (bch2_err_matches(ret, BCH_ERR_transaction_restart) || bch2_err_matches(ret, ENOMEM)) @@ -1131,6 +1131,8 @@ int bch2_btree_path_traverse_one(struct btree_trans *trans, if (unlikely(!trans->srcu_held)) bch2_trans_srcu_lock(trans); + trace_btree_path_traverse_start(trans, path); + /* * Ensure we obey path->should_be_locked: if it's set, we can't unlock * and re-traverse the path without a transaction restart: @@ -1194,6 +1196,7 @@ int bch2_btree_path_traverse_one(struct btree_trans *trans, out_uptodate: path->uptodate = BTREE_ITER_UPTODATE; + trace_btree_path_traverse_end(trans, path); out: if (bch2_err_matches(ret, BCH_ERR_transaction_restart) != !!trans->restarted) panic("ret %s (%i) trans->restarted %s (%i)\n", @@ -1225,7 +1228,7 @@ static btree_path_idx_t btree_path_clone(struct btree_trans *trans, btree_path_i { btree_path_idx_t new = btree_path_alloc(trans, src); btree_path_copy(trans, trans->paths + new, trans->paths + src); - __btree_path_get(trans->paths + new, intent); + __btree_path_get(trans, trans->paths + new, intent); #ifdef TRACK_PATH_ALLOCATED trans->paths[new].ip_allocated = ip; #endif @@ -1236,8 +1239,10 @@ __flatten btree_path_idx_t __bch2_btree_path_make_mut(struct btree_trans *trans, btree_path_idx_t path, bool intent, unsigned long ip) { - __btree_path_put(trans->paths + path, intent); + struct btree_path *old = trans->paths + path; + __btree_path_put(trans, trans->paths + path, intent); path = btree_path_clone(trans, path, intent, ip); + trace_btree_path_clone(trans, old, trans->paths + path); trans->paths[path].preserve = false; return path; } @@ -1252,6 +1257,8 @@ __bch2_btree_path_set_pos(struct btree_trans *trans, bch2_trans_verify_not_in_restart(trans); EBUG_ON(!trans->paths[path_idx].ref); + trace_btree_path_set_pos(trans, trans->paths + path_idx, &new_pos); + path_idx = bch2_btree_path_make_mut(trans, path_idx, intent, ip); struct btree_path *path = trans->paths + path_idx; @@ -1361,13 +1368,15 @@ void bch2_path_put(struct btree_trans *trans, btree_path_idx_t path_idx, bool in { struct btree_path *path = trans->paths + path_idx, *dup; - if (!__btree_path_put(path, intent)) + if (!__btree_path_put(trans, path, intent)) return; dup = path->preserve ? have_path_at_pos(trans, path) : have_node_at_pos(trans, path); + trace_btree_path_free(trans, path_idx, dup); + if (!dup && !(!path->preserve && !is_btree_node(path, path->level))) return; @@ -1392,7 +1401,7 @@ void bch2_path_put(struct btree_trans *trans, btree_path_idx_t path_idx, bool in static void bch2_path_put_nokeep(struct btree_trans *trans, btree_path_idx_t path, bool intent) { - if (!__btree_path_put(trans->paths + path, intent)) + if (!__btree_path_put(trans, trans->paths + path, intent)) return; __bch2_path_free(trans, path); @@ -1421,8 +1430,8 @@ void __noreturn bch2_trans_unlocked_error(struct btree_trans *trans) noinline __cold void bch2_trans_updates_to_text(struct printbuf *buf, struct btree_trans *trans) { - prt_printf(buf, "transaction updates for %s journal seq %llu\n", - trans->fn, trans->journal_res.seq); + prt_printf(buf, "%u transaction updates for %s journal seq %llu\n", + trans->nr_updates, trans->fn, trans->journal_res.seq); printbuf_indent_add(buf, 2); trans_for_each_update(trans, i) { @@ -1464,7 +1473,7 @@ static void bch2_btree_path_to_text_short(struct printbuf *out, struct btree_tra { struct btree_path *path = trans->paths + path_idx; - prt_printf(out, "path: idx %2u ref %u:%u %c %c %c btree=%s l=%u pos ", + prt_printf(out, "path: idx %3u ref %u:%u %c %c %c btree=%s l=%u pos ", path_idx, path->ref, path->intent_ref, path->preserve ? 'P' : ' ', path->should_be_locked ? 'S' : ' ', @@ -1716,14 +1725,16 @@ btree_path_idx_t bch2_path_get(struct btree_trans *trans, trans->paths[path_pos].cached == cached && trans->paths[path_pos].btree_id == btree_id && trans->paths[path_pos].level == level) { - __btree_path_get(trans->paths + path_pos, intent); + trace_btree_path_get(trans, trans->paths + path_pos, &pos); + + __btree_path_get(trans, trans->paths + path_pos, intent); path_idx = bch2_btree_path_set_pos(trans, path_pos, pos, intent, ip); path = trans->paths + path_idx; } else { path_idx = btree_path_alloc(trans, path_pos); path = trans->paths + path_idx; - __btree_path_get(path, intent); + __btree_path_get(trans, path, intent); path->pos = pos; path->btree_id = btree_id; path->cached = cached; @@ -1738,6 +1749,8 @@ btree_path_idx_t bch2_path_get(struct btree_trans *trans, path->ip_allocated = ip; #endif trans->paths_sorted = false; + + trace_btree_path_alloc(trans, path); } if (!(flags & BTREE_ITER_nopreserve)) @@ -1857,7 +1870,7 @@ bch2_btree_iter_traverse(struct btree_iter *iter) struct btree_path *path = btree_iter_path(trans, iter); if (btree_path_node(path, path->level)) - btree_path_set_should_be_locked(path); + btree_path_set_should_be_locked(trans, path); return 0; } @@ -1889,7 +1902,7 @@ struct btree *bch2_btree_iter_peek_node(struct btree_iter *iter) iter->path = bch2_btree_path_set_pos(trans, iter->path, b->key.k.p, iter->flags & BTREE_ITER_intent, btree_iter_ip_allocated(iter)); - btree_path_set_should_be_locked(btree_iter_path(trans, iter)); + btree_path_set_should_be_locked(trans, btree_iter_path(trans, iter)); out: bch2_btree_iter_verify_entry_exit(iter); bch2_btree_iter_verify(iter); @@ -1983,7 +1996,7 @@ struct btree *bch2_btree_iter_next_node(struct btree_iter *iter) iter->path = bch2_btree_path_set_pos(trans, iter->path, b->key.k.p, iter->flags & BTREE_ITER_intent, btree_iter_ip_allocated(iter)); - btree_path_set_should_be_locked(btree_iter_path(trans, iter)); + btree_path_set_should_be_locked(trans, btree_iter_path(trans, iter)); EBUG_ON(btree_iter_path(trans, iter)->uptodate); out: bch2_btree_iter_verify_entry_exit(iter); @@ -2155,7 +2168,7 @@ struct bkey_s_c btree_trans_peek_key_cache(struct btree_iter *iter, struct bpos if (unlikely(ret)) return bkey_s_c_err(ret); - btree_path_set_should_be_locked(trans->paths + iter->key_cache_path); + btree_path_set_should_be_locked(trans, trans->paths + iter->key_cache_path); k = bch2_btree_path_peek_slot(trans->paths + iter->key_cache_path, &u); if (k.k && !bkey_err(k)) { @@ -2199,7 +2212,7 @@ static struct bkey_s_c __bch2_btree_iter_peek(struct btree_iter *iter, struct bp goto out; } - btree_path_set_should_be_locked(path); + btree_path_set_should_be_locked(trans, path); k = btree_path_level_peek_all(trans->c, l, &iter->k); @@ -2326,7 +2339,7 @@ struct bkey_s_c bch2_btree_iter_peek_upto(struct btree_iter *iter, struct bpos e * advance, same as on exit for iter->path, but only up * to snapshot */ - __btree_path_get(trans->paths + iter->path, iter->flags & BTREE_ITER_intent); + __btree_path_get(trans, trans->paths + iter->path, iter->flags & BTREE_ITER_intent); iter->update_path = iter->path; iter->update_path = bch2_btree_path_set_pos(trans, @@ -2382,14 +2395,14 @@ struct bkey_s_c bch2_btree_iter_peek_upto(struct btree_iter *iter, struct bpos e iter->flags & BTREE_ITER_intent, btree_iter_ip_allocated(iter)); - btree_path_set_should_be_locked(btree_iter_path(trans, iter)); + btree_path_set_should_be_locked(trans, btree_iter_path(trans, iter)); out_no_locked: if (iter->update_path) { ret = bch2_btree_path_relock(trans, trans->paths + iter->update_path, _THIS_IP_); if (unlikely(ret)) k = bkey_s_c_err(ret); else - btree_path_set_should_be_locked(trans->paths + iter->update_path); + btree_path_set_should_be_locked(trans, trans->paths + iter->update_path); } if (!(iter->flags & BTREE_ITER_all_snapshots)) @@ -2511,6 +2524,7 @@ struct bkey_s_c bch2_btree_iter_peek_prev(struct btree_iter *iter) iter->flags & BTREE_ITER_intent, _THIS_IP_); path = btree_iter_path(trans, iter); + trace_btree_path_save_pos(trans, path, trans->paths + saved_path); saved_k = *k.k; saved_v = k.v; } @@ -2527,7 +2541,7 @@ got_key: continue; } - btree_path_set_should_be_locked(path); + btree_path_set_should_be_locked(trans, path); break; } else if (likely(!bpos_eq(path->l[0].b->data->min_key, POS_MIN))) { /* Advance to previous leaf node: */ @@ -2685,7 +2699,7 @@ struct bkey_s_c bch2_btree_iter_peek_slot(struct btree_iter *iter) } } out: - btree_path_set_should_be_locked(btree_iter_path(trans, iter)); + btree_path_set_should_be_locked(trans, btree_iter_path(trans, iter)); out_no_locked: bch2_btree_iter_verify_entry_exit(iter); bch2_btree_iter_verify(iter); @@ -2712,6 +2726,7 @@ struct bkey_s_c bch2_btree_iter_prev_slot(struct btree_iter *iter) return bch2_btree_iter_peek_slot(iter); } +/* Obsolete, but still used by rust wrapper in -tools */ struct bkey_s_c bch2_btree_iter_peek_and_restart_outlined(struct btree_iter *iter) { struct bkey_s_c k; @@ -2911,9 +2926,9 @@ void bch2_trans_copy_iter(struct btree_iter *dst, struct btree_iter *src) dst->ip_allocated = _RET_IP_; #endif if (src->path) - __btree_path_get(trans->paths + src->path, src->flags & BTREE_ITER_intent); + __btree_path_get(trans, trans->paths + src->path, src->flags & BTREE_ITER_intent); if (src->update_path) - __btree_path_get(trans->paths + src->update_path, src->flags & BTREE_ITER_intent); + __btree_path_get(trans, trans->paths + src->update_path, src->flags & BTREE_ITER_intent); dst->key_cache_path = 0; } @@ -3237,7 +3252,7 @@ void bch2_trans_put(struct btree_trans *trans) bch2_trans_unlock(trans); trans_for_each_update(trans, i) - __btree_path_put(trans->paths + i->path, true); + __btree_path_put(trans, trans->paths + i->path, true); trans->nr_updates = 0; check_btree_paths_leaked(trans); diff --git a/fs/bcachefs/btree_iter.h b/fs/bcachefs/btree_iter.h index 222b7ce8a901..78e63ad7d380 100644 --- a/fs/bcachefs/btree_iter.h +++ b/fs/bcachefs/btree_iter.h @@ -6,6 +6,12 @@ #include "btree_types.h" #include "trace.h" +void bch2_trans_updates_to_text(struct printbuf *, struct btree_trans *); +void bch2_btree_path_to_text(struct printbuf *, struct btree_trans *, btree_path_idx_t); +void bch2_trans_paths_to_text(struct printbuf *, struct btree_trans *); +void bch2_dump_trans_updates(struct btree_trans *); +void bch2_dump_trans_paths_updates(struct btree_trans *); + static inline int __bkey_err(const struct bkey *k) { return PTR_ERR_OR_ZERO(k); @@ -13,16 +19,28 @@ static inline int __bkey_err(const struct bkey *k) #define bkey_err(_k) __bkey_err((_k).k) -static inline void __btree_path_get(struct btree_path *path, bool intent) +static inline void __btree_path_get(struct btree_trans *trans, struct btree_path *path, bool intent) { + unsigned idx = path - trans->paths; + + EBUG_ON(!test_bit(idx, trans->paths_allocated)); + if (unlikely(path->ref == U8_MAX)) { + bch2_dump_trans_paths_updates(trans); + panic("path %u refcount overflow\n", idx); + } + path->ref++; path->intent_ref += intent; + trace_btree_path_get_ll(trans, path); } -static inline bool __btree_path_put(struct btree_path *path, bool intent) +static inline bool __btree_path_put(struct btree_trans *trans, struct btree_path *path, bool intent) { + EBUG_ON(!test_bit(path - trans->paths, trans->paths_allocated)); EBUG_ON(!path->ref); EBUG_ON(!path->intent_ref && intent); + + trace_btree_path_put_ll(trans, path); path->intent_ref -= intent; return --path->ref == 0; } @@ -511,6 +529,12 @@ void bch2_set_btree_iter_dontneed(struct btree_iter *); void *__bch2_trans_kmalloc(struct btree_trans *, size_t); +/** + * bch2_trans_kmalloc - allocate memory for use by the current transaction + * + * Must be called after bch2_trans_begin, which on second and further calls + * frees all memory allocated in this transaction + */ static inline void *bch2_trans_kmalloc(struct btree_trans *trans, size_t size) { size = roundup(size, 8); @@ -814,20 +838,6 @@ transaction_restart: \ struct bkey_s_c bch2_btree_iter_peek_and_restart_outlined(struct btree_iter *); -static inline struct bkey_s_c -__bch2_btree_iter_peek_and_restart(struct btree_trans *trans, - struct btree_iter *iter, unsigned flags) -{ - struct bkey_s_c k; - - while (btree_trans_too_many_iters(trans) || - (k = bch2_btree_iter_peek_type(iter, flags), - bch2_err_matches(bkey_err(k), BCH_ERR_transaction_restart))) - bch2_trans_begin(trans); - - return k; -} - #define for_each_btree_key_upto_norestart(_trans, _iter, _btree_id, \ _start, _end, _flags, _k, _ret) \ for (bch2_trans_iter_init((_trans), &(_iter), (_btree_id), \ @@ -868,7 +878,7 @@ __bch2_btree_iter_peek_and_restart(struct btree_trans *trans, \ if (bch2_err_matches(_ret, ENOMEM)) { \ _gfp = GFP_KERNEL; \ - _ret = drop_locks_do(trans, _do); \ + _ret = drop_locks_do(_trans, _do); \ } \ _ret; \ }) @@ -881,7 +891,7 @@ __bch2_btree_iter_peek_and_restart(struct btree_trans *trans, _ret = 0; \ if (unlikely(!_p)) { \ _gfp = GFP_KERNEL; \ - _ret = drop_locks_do(trans, ((_p = _do), 0)); \ + _ret = drop_locks_do(_trans, ((_p = _do), 0)); \ } \ _p; \ }) @@ -894,12 +904,6 @@ __bch2_btree_iter_peek_and_restart(struct btree_trans *trans, _ret; \ }) -void bch2_trans_updates_to_text(struct printbuf *, struct btree_trans *); -void bch2_btree_path_to_text(struct printbuf *, struct btree_trans *, btree_path_idx_t); -void bch2_trans_paths_to_text(struct printbuf *, struct btree_trans *); -void bch2_dump_trans_updates(struct btree_trans *); -void bch2_dump_trans_paths_updates(struct btree_trans *); - struct btree_trans *__bch2_trans_get(struct bch_fs *, unsigned); void bch2_trans_put(struct btree_trans *); diff --git a/fs/bcachefs/btree_key_cache.c b/fs/bcachefs/btree_key_cache.c index fda7998734cb..244610b1d0b5 100644 --- a/fs/bcachefs/btree_key_cache.c +++ b/fs/bcachefs/btree_key_cache.c @@ -79,134 +79,47 @@ static bool bkey_cached_lock_for_evict(struct bkey_cached *ck) return true; } -static void bkey_cached_evict(struct btree_key_cache *c, +static bool bkey_cached_evict(struct btree_key_cache *c, struct bkey_cached *ck) { - BUG_ON(rhashtable_remove_fast(&c->table, &ck->hash, - bch2_btree_key_cache_params)); - memset(&ck->key, ~0, sizeof(ck->key)); + bool ret = !rhashtable_remove_fast(&c->table, &ck->hash, + bch2_btree_key_cache_params); + if (ret) { + memset(&ck->key, ~0, sizeof(ck->key)); + atomic_long_dec(&c->nr_keys); + } - atomic_long_dec(&c->nr_keys); + return ret; +} + +static void __bkey_cached_free(struct rcu_pending *pending, struct rcu_head *rcu) +{ + struct bch_fs *c = container_of(pending->srcu, struct bch_fs, btree_trans_barrier); + struct bkey_cached *ck = container_of(rcu, struct bkey_cached, rcu); + + this_cpu_dec(*c->btree_key_cache.nr_pending); + kmem_cache_free(bch2_key_cache, ck); } static void bkey_cached_free(struct btree_key_cache *bc, struct bkey_cached *ck) { - struct bch_fs *c = container_of(bc, struct bch_fs, btree_key_cache); - - BUG_ON(test_bit(BKEY_CACHED_DIRTY, &ck->flags)); - - ck->btree_trans_barrier_seq = - start_poll_synchronize_srcu(&c->btree_trans_barrier); - - if (ck->c.lock.readers) { - list_move_tail(&ck->list, &bc->freed_pcpu); - bc->nr_freed_pcpu++; - } else { - list_move_tail(&ck->list, &bc->freed_nonpcpu); - bc->nr_freed_nonpcpu++; - } - atomic_long_inc(&bc->nr_freed); - kfree(ck->k); ck->k = NULL; ck->u64s = 0; six_unlock_write(&ck->c.lock); six_unlock_intent(&ck->c.lock); -} -#ifdef __KERNEL__ -static void __bkey_cached_move_to_freelist_ordered(struct btree_key_cache *bc, - struct bkey_cached *ck) -{ - struct bkey_cached *pos; - - bc->nr_freed_nonpcpu++; - - list_for_each_entry_reverse(pos, &bc->freed_nonpcpu, list) { - if (ULONG_CMP_GE(ck->btree_trans_barrier_seq, - pos->btree_trans_barrier_seq)) { - list_move(&ck->list, &pos->list); - return; - } - } - - list_move(&ck->list, &bc->freed_nonpcpu); -} -#endif - -static void bkey_cached_move_to_freelist(struct btree_key_cache *bc, - struct bkey_cached *ck) -{ - BUG_ON(test_bit(BKEY_CACHED_DIRTY, &ck->flags)); - - if (!ck->c.lock.readers) { -#ifdef __KERNEL__ - struct btree_key_cache_freelist *f; - bool freed = false; - - preempt_disable(); - f = this_cpu_ptr(bc->pcpu_freed); - - if (f->nr < ARRAY_SIZE(f->objs)) { - f->objs[f->nr++] = ck; - freed = true; - } - preempt_enable(); - - if (!freed) { - mutex_lock(&bc->lock); - preempt_disable(); - f = this_cpu_ptr(bc->pcpu_freed); - - while (f->nr > ARRAY_SIZE(f->objs) / 2) { - struct bkey_cached *ck2 = f->objs[--f->nr]; - - __bkey_cached_move_to_freelist_ordered(bc, ck2); - } - preempt_enable(); - - __bkey_cached_move_to_freelist_ordered(bc, ck); - mutex_unlock(&bc->lock); - } -#else - mutex_lock(&bc->lock); - list_move_tail(&ck->list, &bc->freed_nonpcpu); - bc->nr_freed_nonpcpu++; - mutex_unlock(&bc->lock); -#endif - } else { - mutex_lock(&bc->lock); - list_move_tail(&ck->list, &bc->freed_pcpu); - bc->nr_freed_pcpu++; - mutex_unlock(&bc->lock); - } -} - -static void bkey_cached_free_fast(struct btree_key_cache *bc, - struct bkey_cached *ck) -{ - struct bch_fs *c = container_of(bc, struct bch_fs, btree_key_cache); - - ck->btree_trans_barrier_seq = - start_poll_synchronize_srcu(&c->btree_trans_barrier); - - list_del_init(&ck->list); - atomic_long_inc(&bc->nr_freed); - - kfree(ck->k); - ck->k = NULL; - ck->u64s = 0; - - bkey_cached_move_to_freelist(bc, ck); - - six_unlock_write(&ck->c.lock); - six_unlock_intent(&ck->c.lock); + bool pcpu_readers = ck->c.lock.readers != NULL; + rcu_pending_enqueue(&bc->pending[pcpu_readers], &ck->rcu); + this_cpu_inc(*bc->nr_pending); } static struct bkey_cached *__bkey_cached_alloc(unsigned key_u64s, gfp_t gfp) { + gfp |= __GFP_ACCOUNT|__GFP_RECLAIMABLE; + struct bkey_cached *ck = kmem_cache_zalloc(bch2_key_cache, gfp); if (unlikely(!ck)) return NULL; @@ -224,74 +137,14 @@ bkey_cached_alloc(struct btree_trans *trans, struct btree_path *path, unsigned k { struct bch_fs *c = trans->c; struct btree_key_cache *bc = &c->btree_key_cache; - struct bkey_cached *ck = NULL; bool pcpu_readers = btree_uses_pcpu_readers(path->btree_id); int ret; - if (!pcpu_readers) { -#ifdef __KERNEL__ - struct btree_key_cache_freelist *f; - - preempt_disable(); - f = this_cpu_ptr(bc->pcpu_freed); - if (f->nr) - ck = f->objs[--f->nr]; - preempt_enable(); - - if (!ck) { - mutex_lock(&bc->lock); - preempt_disable(); - f = this_cpu_ptr(bc->pcpu_freed); - - while (!list_empty(&bc->freed_nonpcpu) && - f->nr < ARRAY_SIZE(f->objs) / 2) { - ck = list_last_entry(&bc->freed_nonpcpu, struct bkey_cached, list); - list_del_init(&ck->list); - bc->nr_freed_nonpcpu--; - f->objs[f->nr++] = ck; - } - - ck = f->nr ? f->objs[--f->nr] : NULL; - preempt_enable(); - mutex_unlock(&bc->lock); - } -#else - mutex_lock(&bc->lock); - if (!list_empty(&bc->freed_nonpcpu)) { - ck = list_last_entry(&bc->freed_nonpcpu, struct bkey_cached, list); - list_del_init(&ck->list); - bc->nr_freed_nonpcpu--; - } - mutex_unlock(&bc->lock); -#endif - } else { - mutex_lock(&bc->lock); - if (!list_empty(&bc->freed_pcpu)) { - ck = list_last_entry(&bc->freed_pcpu, struct bkey_cached, list); - list_del_init(&ck->list); - bc->nr_freed_pcpu--; - } - mutex_unlock(&bc->lock); - } - - if (ck) { - ret = btree_node_lock_nopath(trans, &ck->c, SIX_LOCK_intent, _THIS_IP_); - if (unlikely(ret)) { - bkey_cached_move_to_freelist(bc, ck); - return ERR_PTR(ret); - } - - btree_path_cached_set(trans, path, ck, BTREE_NODE_INTENT_LOCKED); - - ret = bch2_btree_node_lock_write(trans, path, &ck->c); - if (unlikely(ret)) { - btree_node_unlock(trans, path, 0); - bkey_cached_move_to_freelist(bc, ck); - return ERR_PTR(ret); - } - - return ck; - } + struct bkey_cached *ck = container_of_or_null( + rcu_pending_dequeue(&bc->pending[pcpu_readers]), + struct bkey_cached, rcu); + if (ck) + goto lock; ck = allocate_dropping_locks(trans, ret, __bkey_cached_alloc(key_u64s, _gfp)); @@ -302,15 +155,19 @@ bkey_cached_alloc(struct btree_trans *trans, struct btree_path *path, unsigned k return ERR_PTR(ret); } - if (!ck) - return NULL; + if (ck) { + bch2_btree_lock_init(&ck->c, pcpu_readers ? SIX_LOCK_INIT_PCPU : 0); + ck->c.cached = true; + goto lock; + } - INIT_LIST_HEAD(&ck->list); - bch2_btree_lock_init(&ck->c, pcpu_readers ? SIX_LOCK_INIT_PCPU : 0); - - ck->c.cached = true; - BUG_ON(!six_trylock_intent(&ck->c.lock)); - BUG_ON(!six_trylock_write(&ck->c.lock)); + ck = container_of_or_null(rcu_pending_dequeue_from_all(&bc->pending[pcpu_readers]), + struct bkey_cached, rcu); + if (ck) + goto lock; +lock: + six_lock_intent(&ck->c.lock, NULL, NULL); + six_lock_write(&ck->c.lock, NULL, NULL); return ck; } @@ -322,21 +179,21 @@ bkey_cached_reuse(struct btree_key_cache *c) struct bkey_cached *ck; unsigned i; - mutex_lock(&c->lock); rcu_read_lock(); tbl = rht_dereference_rcu(c->table.tbl, &c->table); for (i = 0; i < tbl->size; i++) rht_for_each_entry_rcu(ck, pos, tbl, i, hash) { if (!test_bit(BKEY_CACHED_DIRTY, &ck->flags) && bkey_cached_lock_for_evict(ck)) { - bkey_cached_evict(c, ck); - goto out; + if (bkey_cached_evict(c, ck)) + goto out; + six_unlock_write(&ck->c.lock); + six_unlock_intent(&ck->c.lock); } } ck = NULL; out: rcu_read_unlock(); - mutex_unlock(&c->lock); return ck; } @@ -415,7 +272,7 @@ static int btree_key_cache_create(struct btree_trans *trans, struct btree_path * path->uptodate = BTREE_ITER_UPTODATE; return 0; err: - bkey_cached_free_fast(bc, ck); + bkey_cached_free(bc, ck); mark_btree_node_locked_noreset(path, 0, BTREE_NODE_UNLOCKED); return ret; @@ -611,8 +468,12 @@ evict: } mark_btree_node_locked_noreset(path, 0, BTREE_NODE_UNLOCKED); - bkey_cached_evict(&c->btree_key_cache, ck); - bkey_cached_free_fast(&c->btree_key_cache, ck); + if (bkey_cached_evict(&c->btree_key_cache, ck)) { + bkey_cached_free(&c->btree_key_cache, ck); + } else { + six_unlock_write(&ck->c.lock); + six_unlock_intent(&ck->c.lock); + } } out: bch2_trans_iter_exit(trans, &b_iter); @@ -722,7 +583,7 @@ void bch2_btree_key_cache_drop(struct btree_trans *trans, } bkey_cached_evict(bc, ck); - bkey_cached_free_fast(bc, ck); + bkey_cached_free(bc, ck); mark_btree_node_locked(trans, path, 0, BTREE_NODE_UNLOCKED); btree_path_set_dirty(path, BTREE_ITER_NEED_TRAVERSE); @@ -735,48 +596,14 @@ static unsigned long bch2_btree_key_cache_scan(struct shrinker *shrink, struct bch_fs *c = shrink->private_data; struct btree_key_cache *bc = &c->btree_key_cache; struct bucket_table *tbl; - struct bkey_cached *ck, *t; + struct bkey_cached *ck; size_t scanned = 0, freed = 0, nr = sc->nr_to_scan; - unsigned start, flags; + unsigned iter, start; int srcu_idx; - mutex_lock(&bc->lock); - bc->requested_to_free += sc->nr_to_scan; - srcu_idx = srcu_read_lock(&c->btree_trans_barrier); - flags = memalloc_nofs_save(); - - /* - * Newest freed entries are at the end of the list - once we hit one - * that's too new to be freed, we can bail out: - */ - list_for_each_entry_safe(ck, t, &bc->freed_nonpcpu, list) { - if (!poll_state_synchronize_srcu(&c->btree_trans_barrier, - ck->btree_trans_barrier_seq)) - break; - - list_del(&ck->list); - six_lock_exit(&ck->c.lock); - kmem_cache_free(bch2_key_cache, ck); - atomic_long_dec(&bc->nr_freed); - bc->nr_freed_nonpcpu--; - bc->freed++; - } - - list_for_each_entry_safe(ck, t, &bc->freed_pcpu, list) { - if (!poll_state_synchronize_srcu(&c->btree_trans_barrier, - ck->btree_trans_barrier_seq)) - break; - - list_del(&ck->list); - six_lock_exit(&ck->c.lock); - kmem_cache_free(bch2_key_cache, ck); - atomic_long_dec(&bc->nr_freed); - bc->nr_freed_pcpu--; - bc->freed++; - } - rcu_read_lock(); + tbl = rht_dereference_rcu(bc->table.tbl, &bc->table); /* @@ -792,17 +619,18 @@ static unsigned long bch2_btree_key_cache_scan(struct shrinker *shrink, return SHRINK_STOP; } - if (bc->shrink_iter >= tbl->size) - bc->shrink_iter = 0; - start = bc->shrink_iter; + iter = bc->shrink_iter; + if (iter >= tbl->size) + iter = 0; + start = iter; do { struct rhash_head *pos, *next; - pos = rht_ptr_rcu(&tbl->buckets[bc->shrink_iter]); + pos = rht_ptr_rcu(&tbl->buckets[iter]); while (!rht_is_a_nulls(pos)) { - next = rht_dereference_bucket_rcu(pos->next, tbl, bc->shrink_iter); + next = rht_dereference_bucket_rcu(pos->next, tbl, iter); ck = container_of(pos, struct bkey_cached, hash); if (test_bit(BKEY_CACHED_DIRTY, &ck->flags)) { @@ -812,29 +640,31 @@ static unsigned long bch2_btree_key_cache_scan(struct shrinker *shrink, bc->skipped_accessed++; } else if (!bkey_cached_lock_for_evict(ck)) { bc->skipped_lock_fail++; - } else { - bkey_cached_evict(bc, ck); + } else if (bkey_cached_evict(bc, ck)) { bkey_cached_free(bc, ck); - bc->moved_to_freelist++; + bc->freed++; freed++; + } else { + six_unlock_write(&ck->c.lock); + six_unlock_intent(&ck->c.lock); } scanned++; if (scanned >= nr) - break; + goto out; pos = next; } - bc->shrink_iter++; - if (bc->shrink_iter >= tbl->size) - bc->shrink_iter = 0; - } while (scanned < nr && bc->shrink_iter != start); + iter++; + if (iter >= tbl->size) + iter = 0; + } while (scanned < nr && iter != start); +out: + bc->shrink_iter = iter; rcu_read_unlock(); - memalloc_nofs_restore(flags); srcu_read_unlock(&c->btree_trans_barrier, srcu_idx); - mutex_unlock(&bc->lock); return freed; } @@ -862,18 +692,13 @@ void bch2_fs_btree_key_cache_exit(struct btree_key_cache *bc) { struct bch_fs *c = container_of(bc, struct bch_fs, btree_key_cache); struct bucket_table *tbl; - struct bkey_cached *ck, *n; + struct bkey_cached *ck; struct rhash_head *pos; LIST_HEAD(items); unsigned i; -#ifdef __KERNEL__ - int cpu; -#endif shrinker_free(bc->shrink); - mutex_lock(&bc->lock); - /* * The loop is needed to guard against racing with rehash: */ @@ -892,44 +717,14 @@ void bch2_fs_btree_key_cache_exit(struct btree_key_cache *bc) for (i = 0; i < tbl->size; i++) while (pos = rht_ptr_rcu(&tbl->buckets[i]), !rht_is_a_nulls(pos)) { ck = container_of(pos, struct bkey_cached, hash); - bkey_cached_evict(bc, ck); - list_add(&ck->list, &items); + BUG_ON(!bkey_cached_evict(bc, ck)); + kfree(ck->k); + kmem_cache_free(bch2_key_cache, ck); } } rcu_read_unlock(); } -#ifdef __KERNEL__ - if (bc->pcpu_freed) { - for_each_possible_cpu(cpu) { - struct btree_key_cache_freelist *f = - per_cpu_ptr(bc->pcpu_freed, cpu); - - for (i = 0; i < f->nr; i++) { - ck = f->objs[i]; - list_add(&ck->list, &items); - } - } - } -#endif - - BUG_ON(list_count_nodes(&bc->freed_pcpu) != bc->nr_freed_pcpu); - BUG_ON(list_count_nodes(&bc->freed_nonpcpu) != bc->nr_freed_nonpcpu); - - list_splice(&bc->freed_pcpu, &items); - list_splice(&bc->freed_nonpcpu, &items); - - mutex_unlock(&bc->lock); - - list_for_each_entry_safe(ck, n, &items, list) { - cond_resched(); - - list_del(&ck->list); - kfree(ck->k); - six_lock_exit(&ck->c.lock); - kmem_cache_free(bch2_key_cache, ck); - } - if (atomic_long_read(&bc->nr_dirty) && !bch2_journal_error(&c->journal) && test_bit(BCH_FS_was_rw, &c->flags)) @@ -943,14 +738,14 @@ void bch2_fs_btree_key_cache_exit(struct btree_key_cache *bc) if (bc->table_init_done) rhashtable_destroy(&bc->table); - free_percpu(bc->pcpu_freed); + rcu_pending_exit(&bc->pending[0]); + rcu_pending_exit(&bc->pending[1]); + + free_percpu(bc->nr_pending); } void bch2_fs_btree_key_cache_init_early(struct btree_key_cache *c) { - mutex_init(&c->lock); - INIT_LIST_HEAD(&c->freed_pcpu); - INIT_LIST_HEAD(&c->freed_nonpcpu); } int bch2_fs_btree_key_cache_init(struct btree_key_cache *bc) @@ -958,11 +753,13 @@ int bch2_fs_btree_key_cache_init(struct btree_key_cache *bc) struct bch_fs *c = container_of(bc, struct bch_fs, btree_key_cache); struct shrinker *shrink; -#ifdef __KERNEL__ - bc->pcpu_freed = alloc_percpu(struct btree_key_cache_freelist); - if (!bc->pcpu_freed) + bc->nr_pending = alloc_percpu(size_t); + if (!bc->nr_pending) + return -BCH_ERR_ENOMEM_fs_btree_cache_init; + + if (rcu_pending_init(&bc->pending[0], &c->btree_trans_barrier, __bkey_cached_free) || + rcu_pending_init(&bc->pending[1], &c->btree_trans_barrier, __bkey_cached_free)) return -BCH_ERR_ENOMEM_fs_btree_cache_init; -#endif if (rhashtable_init(&bc->table, &bch2_btree_key_cache_params)) return -BCH_ERR_ENOMEM_fs_btree_cache_init; @@ -984,45 +781,21 @@ int bch2_fs_btree_key_cache_init(struct btree_key_cache *bc) void bch2_btree_key_cache_to_text(struct printbuf *out, struct btree_key_cache *bc) { - struct bch_fs *c = container_of(bc, struct bch_fs, btree_key_cache); - printbuf_tabstop_push(out, 24); printbuf_tabstop_push(out, 12); - unsigned flags = memalloc_nofs_save(); - mutex_lock(&bc->lock); prt_printf(out, "keys:\t%lu\r\n", atomic_long_read(&bc->nr_keys)); prt_printf(out, "dirty:\t%lu\r\n", atomic_long_read(&bc->nr_dirty)); - prt_printf(out, "freelist:\t%lu\r\n", atomic_long_read(&bc->nr_freed)); - prt_printf(out, "nonpcpu freelist:\t%zu\r\n", bc->nr_freed_nonpcpu); - prt_printf(out, "pcpu freelist:\t%zu\r\n", bc->nr_freed_pcpu); - - prt_printf(out, "\nshrinker:\n"); + prt_printf(out, "table size:\t%u\r\n", bc->table.tbl->size); + prt_newline(out); + prt_printf(out, "shrinker:\n"); prt_printf(out, "requested_to_free:\t%lu\r\n", bc->requested_to_free); prt_printf(out, "freed:\t%lu\r\n", bc->freed); - prt_printf(out, "moved_to_freelist:\t%lu\r\n", bc->moved_to_freelist); prt_printf(out, "skipped_dirty:\t%lu\r\n", bc->skipped_dirty); prt_printf(out, "skipped_accessed:\t%lu\r\n", bc->skipped_accessed); prt_printf(out, "skipped_lock_fail:\t%lu\r\n", bc->skipped_lock_fail); - - prt_printf(out, "srcu seq:\t%lu\r\n", get_state_synchronize_srcu(&c->btree_trans_barrier)); - - struct bkey_cached *ck; - unsigned iter = 0; - list_for_each_entry(ck, &bc->freed_nonpcpu, list) { - prt_printf(out, "freed_nonpcpu:\t%lu\r\n", ck->btree_trans_barrier_seq); - if (++iter > 10) - break; - } - - iter = 0; - list_for_each_entry(ck, &bc->freed_pcpu, list) { - prt_printf(out, "freed_pcpu:\t%lu\r\n", ck->btree_trans_barrier_seq); - if (++iter > 10) - break; - } - mutex_unlock(&bc->lock); - memalloc_flags_restore(flags); + prt_newline(out); + prt_printf(out, "pending:\t%zu\r\n", per_cpu_sum(bc->nr_pending)); } void bch2_btree_key_cache_exit(void) diff --git a/fs/bcachefs/btree_key_cache_types.h b/fs/bcachefs/btree_key_cache_types.h index 237e8bb3ac40..722f1ed10551 100644 --- a/fs/bcachefs/btree_key_cache_types.h +++ b/fs/bcachefs/btree_key_cache_types.h @@ -2,33 +2,25 @@ #ifndef _BCACHEFS_BTREE_KEY_CACHE_TYPES_H #define _BCACHEFS_BTREE_KEY_CACHE_TYPES_H -struct btree_key_cache_freelist { - struct bkey_cached *objs[16]; - unsigned nr; -}; +#include "rcu_pending.h" struct btree_key_cache { - struct mutex lock; struct rhashtable table; bool table_init_done; - struct list_head freed_pcpu; - size_t nr_freed_pcpu; - struct list_head freed_nonpcpu; - size_t nr_freed_nonpcpu; - struct shrinker *shrink; unsigned shrink_iter; - struct btree_key_cache_freelist __percpu *pcpu_freed; - atomic_long_t nr_freed; + /* 0: non pcpu reader locks, 1: pcpu reader locks */ + struct rcu_pending pending[2]; + size_t __percpu *nr_pending; + atomic_long_t nr_keys; atomic_long_t nr_dirty; /* shrinker stats */ unsigned long requested_to_free; unsigned long freed; - unsigned long moved_to_freelist; unsigned long skipped_dirty; unsigned long skipped_accessed; unsigned long skipped_lock_fail; diff --git a/fs/bcachefs/btree_locking.h b/fs/bcachefs/btree_locking.h index 11a64ead8685..7c07f9fa9add 100644 --- a/fs/bcachefs/btree_locking.h +++ b/fs/bcachefs/btree_locking.h @@ -218,16 +218,17 @@ static inline int __btree_node_lock_nopath(struct btree_trans *trans, bool lock_may_not_fail, unsigned long ip) { - int ret; - trans->lock_may_not_fail = lock_may_not_fail; trans->lock_must_abort = false; trans->locking = b; - ret = six_lock_ip_waiter(&b->lock, type, &trans->locking_wait, - bch2_six_check_for_deadlock, trans, ip); + int ret = six_lock_ip_waiter(&b->lock, type, &trans->locking_wait, + bch2_six_check_for_deadlock, trans, ip); WRITE_ONCE(trans->locking, NULL); WRITE_ONCE(trans->locking_wait.start_time, 0); + + if (!ret) + trace_btree_path_lock(trans, _THIS_IP_, b); return ret; } @@ -281,6 +282,7 @@ static inline int btree_node_lock(struct btree_trans *trans, int ret = 0; EBUG_ON(level >= BTREE_MAX_DEPTH); + bch2_trans_verify_not_unlocked(trans); if (likely(six_trylock_type(&b->lock, type)) || btree_node_lock_increment(trans, b, level, (enum btree_node_locked_type) type) || @@ -400,12 +402,13 @@ static inline int bch2_btree_path_upgrade(struct btree_trans *trans, /* misc: */ -static inline void btree_path_set_should_be_locked(struct btree_path *path) +static inline void btree_path_set_should_be_locked(struct btree_trans *trans, struct btree_path *path) { EBUG_ON(!btree_node_locked(path, path->level)); EBUG_ON(path->uptodate); path->should_be_locked = true; + trace_btree_path_should_be_locked(trans, path); } static inline void __btree_path_set_level_up(struct btree_trans *trans, diff --git a/fs/bcachefs/btree_trans_commit.c b/fs/bcachefs/btree_trans_commit.c index a0101d9c5d83..91884da4e30a 100644 --- a/fs/bcachefs/btree_trans_commit.c +++ b/fs/bcachefs/btree_trans_commit.c @@ -214,7 +214,7 @@ bool bch2_btree_bset_insert_key(struct btree_trans *trans, k = bch2_btree_node_iter_bset_pos(node_iter, b, bset_tree_last(b)); overwrite: - bch2_bset_insert(b, node_iter, k, insert, clobber_u64s); + bch2_bset_insert(b, k, insert, clobber_u64s); new_u64s = k->u64s; fix_iter: if (clobber_u64s != new_u64s) diff --git a/fs/bcachefs/btree_types.h b/fs/bcachefs/btree_types.h index b256b2a20a4f..4568a41fefaf 100644 --- a/fs/bcachefs/btree_types.h +++ b/fs/bcachefs/btree_types.h @@ -138,6 +138,31 @@ struct btree { struct list_head list; }; +#define BCH_BTREE_CACHE_NOT_FREED_REASONS() \ + x(lock_intent) \ + x(lock_write) \ + x(dirty) \ + x(read_in_flight) \ + x(write_in_flight) \ + x(noevict) \ + x(write_blocked) \ + x(will_make_reachable) \ + x(access_bit) + +enum bch_btree_cache_not_freed_reasons { +#define x(n) BCH_BTREE_CACHE_NOT_FREED_##n, + BCH_BTREE_CACHE_NOT_FREED_REASONS() +#undef x + BCH_BTREE_CACHE_NOT_FREED_REASONS_NR, +}; + +struct btree_cache_list { + unsigned idx; + struct shrinker *shrink; + struct list_head list; + size_t nr; +}; + struct btree_cache { struct rhashtable table; bool table_init_done; @@ -155,28 +180,19 @@ struct btree_cache { * should never grow past ~2-3 nodes in practice. */ struct mutex lock; - struct list_head live; struct list_head freeable; struct list_head freed_pcpu; struct list_head freed_nonpcpu; + struct btree_cache_list live[2]; - /* Number of elements in live + freeable lists */ - unsigned used; - unsigned reserve; - unsigned freed; - unsigned not_freed_lock_intent; - unsigned not_freed_lock_write; - unsigned not_freed_dirty; - unsigned not_freed_read_in_flight; - unsigned not_freed_write_in_flight; - unsigned not_freed_noevict; - unsigned not_freed_write_blocked; - unsigned not_freed_will_make_reachable; - unsigned not_freed_access_bit; - atomic_t dirty; - struct shrinker *shrink; + size_t nr_freeable; + size_t nr_reserve; + size_t nr_by_btree[BTREE_ID_NR]; + atomic_long_t nr_dirty; - unsigned used_by_btree[BTREE_ID_NR]; + /* shrinker stats */ + size_t nr_freed; + u64 not_freed[BCH_BTREE_CACHE_NOT_FREED_REASONS_NR]; /* * If we need to allocate memory for a new btree node and that @@ -189,8 +205,8 @@ struct btree_cache { struct bbpos pinned_nodes_start; struct bbpos pinned_nodes_end; - u64 pinned_nodes_leaf_mask; - u64 pinned_nodes_interior_mask; + /* btree id mask: 0 for leaves, 1 for interior */ + u64 pinned_nodes_mask[2]; }; struct btree_node_iter { @@ -386,17 +402,16 @@ struct bkey_cached { struct btree_bkey_cached_common c; unsigned long flags; - unsigned long btree_trans_barrier_seq; u16 u64s; struct bkey_cached_key key; struct rhash_head hash; - struct list_head list; struct journal_entry_pin journal; u64 seq; struct bkey_i *k; + struct rcu_head rcu; }; static inline struct bpos btree_node_pos(struct btree_bkey_cached_common *b) @@ -583,7 +598,8 @@ enum btree_write_type { x(dying) \ x(fake) \ x(need_rewrite) \ - x(never_write) + x(never_write) \ + x(pinned) enum btree_flags { /* First bits for btree node write type */ diff --git a/fs/bcachefs/btree_update.c b/fs/bcachefs/btree_update.c index d6f6df10dcc3..514df618548e 100644 --- a/fs/bcachefs/btree_update.c +++ b/fs/bcachefs/btree_update.c @@ -374,7 +374,7 @@ static noinline int flush_new_cached_update(struct btree_trans *trans, i->key_cache_already_flushed = true; i->flags |= BTREE_TRIGGER_norun; - btree_path_set_should_be_locked(btree_path); + btree_path_set_should_be_locked(trans, btree_path); ret = bch2_trans_update_by_path(trans, path_idx, i->k, flags, ip); out: bch2_path_put(trans, path_idx, true); @@ -422,7 +422,9 @@ bch2_trans_update_by_path(struct btree_trans *trans, btree_path_idx_t path_idx, break; } - if (!cmp && i < trans->updates + trans->nr_updates) { + bool overwrite = !cmp && i < trans->updates + trans->nr_updates; + + if (overwrite) { EBUG_ON(i->insert_trigger_run || i->overwrite_trigger_run); bch2_path_put(trans, i->path, true); @@ -449,7 +451,9 @@ bch2_trans_update_by_path(struct btree_trans *trans, btree_path_idx_t path_idx, } } - __btree_path_get(trans->paths + i->path, true); + __btree_path_get(trans, trans->paths + i->path, true); + + trace_update_by_path(trans, path, i, overwrite); /* * If a key is present in the key cache, it must also exist in the @@ -498,7 +502,7 @@ static noinline int bch2_trans_update_get_key_cache(struct btree_trans *trans, return btree_trans_restart(trans, BCH_ERR_transaction_restart_key_cache_raced); } - btree_path_set_should_be_locked(trans->paths + iter->key_cache_path); + btree_path_set_should_be_locked(trans, trans->paths + iter->key_cache_path); } return 0; diff --git a/fs/bcachefs/btree_update_interior.c b/fs/bcachefs/btree_update_interior.c index 8fd112026e7a..190bc1e81756 100644 --- a/fs/bcachefs/btree_update_interior.c +++ b/fs/bcachefs/btree_update_interior.c @@ -16,6 +16,7 @@ #include "clock.h" #include "error.h" #include "extents.h" +#include "io_write.h" #include "journal.h" #include "journal_reclaim.h" #include "keylist.h" @@ -145,7 +146,7 @@ fsck_err: printbuf_exit(&buf); return ret; topology_repair: - if ((c->recovery_passes_explicit & BIT_ULL(BCH_RECOVERY_PASS_check_topology)) && + if ((c->opts.recovery_passes & BIT_ULL(BCH_RECOVERY_PASS_check_topology)) && c->curr_recovery_pass > BCH_RECOVERY_PASS_check_topology) { bch2_inconsistent_error(c); ret = -BCH_ERR_btree_need_topology_repair; @@ -250,8 +251,13 @@ static void bch2_btree_node_free_inmem(struct btree_trans *trans, unsigned i, level = b->c.level; bch2_btree_node_lock_write_nofail(trans, path, &b->c); + + mutex_lock(&c->btree_cache.lock); bch2_btree_node_hash_remove(&c->btree_cache, b); + mutex_unlock(&c->btree_cache.lock); + __btree_node_free(trans, b); + six_unlock_write(&b->c.lock); mark_btree_node_locked_noreset(path, level, BTREE_NODE_INTENT_LOCKED); @@ -283,7 +289,6 @@ static void bch2_btree_node_free_never_used(struct btree_update *as, clear_btree_node_need_write(b); mutex_lock(&c->btree_cache.lock); - list_del_init(&b->list); bch2_btree_node_hash_remove(&c->btree_cache, b); mutex_unlock(&c->btree_cache.lock); @@ -731,6 +736,18 @@ static void btree_update_nodes_written(struct btree_update *as) bch2_fs_fatal_err_on(ret && !bch2_journal_error(&c->journal), c, "%s", bch2_err_str(ret)); err: + /* + * Ensure transaction is unlocked before using btree_node_lock_nopath() + * (the use of which is always suspect, we need to work on removing this + * in the future) + * + * It should be, but bch2_path_get_unlocked_mut() -> bch2_path_get() + * calls bch2_path_upgrade(), before we call path_make_mut(), so we may + * rarely end up with a locked path besides the one we have here: + */ + bch2_trans_unlock(trans); + bch2_trans_begin(trans); + /* * We have to be careful because another thread might be getting ready * to free as->b and calling btree_update_reparent() on us - we'll @@ -750,18 +767,6 @@ err: * we're in journal error state: */ - /* - * Ensure transaction is unlocked before using - * btree_node_lock_nopath() (the use of which is always suspect, - * we need to work on removing this in the future) - * - * It should be, but bch2_path_get_unlocked_mut() -> bch2_path_get() - * calls bch2_path_upgrade(), before we call path_make_mut(), so - * we may rarely end up with a locked path besides the one we - * have here: - */ - bch2_trans_unlock(trans); - bch2_trans_begin(trans); btree_path_idx_t path_idx = bch2_path_get_unlocked_mut(trans, as->btree_id, b->c.level, b->key.k.p); struct btree_path *path = trans->paths + path_idx; @@ -1899,7 +1904,7 @@ static void __btree_increase_depth(struct btree_update *as, struct btree_trans * six_unlock_intent(&n->c.lock); mutex_lock(&c->btree_cache.lock); - list_add_tail(&b->list, &c->btree_cache.live); + list_add_tail(&b->list, &c->btree_cache.live[btree_node_pinned(b)].list); mutex_unlock(&c->btree_cache.lock); bch2_trans_verify_locks(trans); @@ -1981,7 +1986,7 @@ int __bch2_foreground_maybe_merge(struct btree_trans *trans, if (ret) goto err; - btree_path_set_should_be_locked(trans->paths + sib_path); + btree_path_set_should_be_locked(trans, trans->paths + sib_path); m = trans->paths[sib_path].l[level].b; diff --git a/fs/bcachefs/btree_update_interior.h b/fs/bcachefs/btree_update_interior.h index 02c6ecada97c..10f400957f21 100644 --- a/fs/bcachefs/btree_update_interior.h +++ b/fs/bcachefs/btree_update_interior.h @@ -159,6 +159,8 @@ static inline int bch2_foreground_maybe_merge(struct btree_trans *trans, unsigned level, unsigned flags) { + bch2_trans_verify_not_unlocked(trans); + return bch2_foreground_maybe_merge_sibling(trans, path, level, flags, btree_prev_sib) ?: bch2_foreground_maybe_merge_sibling(trans, path, level, flags, diff --git a/fs/bcachefs/buckets.c b/fs/bcachefs/buckets.c index 721bbe1dffc1..546cd01a72e3 100644 --- a/fs/bcachefs/buckets.c +++ b/fs/bcachefs/buckets.c @@ -75,6 +75,15 @@ void bch2_dev_usage_to_text(struct printbuf *out, struct bch_dev *ca, struct bch_dev_usage *usage) { + if (out->nr_tabstops < 5) { + printbuf_tabstops_reset(out); + printbuf_tabstop_push(out, 12); + printbuf_tabstop_push(out, 16); + printbuf_tabstop_push(out, 16); + printbuf_tabstop_push(out, 16); + printbuf_tabstop_push(out, 16); + } + prt_printf(out, "\tbuckets\rsectors\rfragmented\r\n"); for (unsigned i = 0; i < BCH_DATA_NR; i++) { @@ -272,7 +281,7 @@ int bch2_check_fix_ptrs(struct btree_trans *trans, goto err; rcu_read_lock(); - bch2_bkey_drop_ptrs(bkey_i_to_s(new), ptr, !bch2_dev_rcu(c, ptr->dev)); + bch2_bkey_drop_ptrs(bkey_i_to_s(new), ptr, !bch2_dev_exists(c, ptr->dev)); rcu_read_unlock(); if (level) { @@ -477,7 +486,7 @@ out: return ret; err: bch2_dump_trans_updates(trans); - ret = -EIO; + ret = -BCH_ERR_bucket_ref_update; goto out; } @@ -556,22 +565,24 @@ static int bch2_trigger_pointer(struct btree_trans *trans, s64 *sectors, enum btree_iter_update_trigger_flags flags) { + struct bch_fs *c = trans->c; bool insert = !(flags & BTREE_TRIGGER_overwrite); struct printbuf buf = PRINTBUF; int ret = 0; - struct bch_fs *c = trans->c; + u64 abs_sectors = ptr_disk_sectors(level ? btree_sectors(c) : k.k->size, p); + *sectors = insert ? abs_sectors : -abs_sectors; + struct bch_dev *ca = bch2_dev_tryget(c, p.ptr.dev); if (unlikely(!ca)) { if (insert && p.ptr.dev != BCH_SB_MEMBER_INVALID) - ret = -EIO; + ret = -BCH_ERR_trigger_pointer; goto err; } struct bpos bucket; struct bch_backpointer bp; - bch2_extent_ptr_to_bp(trans->c, ca, btree_id, level, k, p, entry, &bucket, &bp); - *sectors = insert ? bp.bucket_len : -((s64) bp.bucket_len); + __bch2_extent_ptr_to_bp(trans->c, ca, btree_id, level, k, p, entry, &bucket, &bp, abs_sectors); if (flags & BTREE_TRIGGER_transactional) { struct bkey_i_alloc_v4 *a = bch2_trans_start_alloc_update(trans, bucket, 0); @@ -593,7 +604,7 @@ static int bch2_trigger_pointer(struct btree_trans *trans, if (bch2_fs_inconsistent_on(!g, c, "reference to invalid bucket on device %u\n %s", p.ptr.dev, (bch2_bkey_val_to_text(&buf, c, k), buf.buf))) { - ret = -EIO; + ret = -BCH_ERR_trigger_pointer; goto err_unlock; } @@ -638,7 +649,7 @@ static int bch2_trigger_stripe_ptr(struct btree_trans *trans, bch2_trans_inconsistent(trans, "stripe pointer doesn't match stripe %llu", (u64) p.ec.idx); - ret = -EIO; + ret = -BCH_ERR_trigger_stripe_pointer; goto err; } @@ -677,7 +688,7 @@ err: (u64) p.ec.idx, buf.buf); printbuf_exit(&buf); bch2_inconsistent_error(c); - return -EIO; + return -BCH_ERR_trigger_stripe_pointer; } m->block_sectors[p.ec.block] += sectors; @@ -741,7 +752,7 @@ static int __trigger_extent(struct btree_trans *trans, return ret; } else if (!p.has_ec) { *replicas_sectors += disk_sectors; - acc_replicas_key.replicas.devs[acc_replicas_key.replicas.nr_devs++] = p.ptr.dev; + replicas_entry_add_dev(&acc_replicas_key.replicas, p.ptr.dev); } else { ret = bch2_trigger_stripe_ptr(trans, k, p, data_type, disk_sectors, flags); if (ret) @@ -957,7 +968,7 @@ static int __bch2_trans_mark_metadata_bucket(struct btree_trans *trans, bch2_data_type_str(a->v.data_type), bch2_data_type_str(type), bch2_data_type_str(type)); - ret = -EIO; + ret = -BCH_ERR_metadata_bucket_inconsistency; goto err; } @@ -1013,7 +1024,7 @@ err: bucket_unlock(g); err_unlock: percpu_up_read(&c->mark_lock); - return -EIO; + return -BCH_ERR_metadata_bucket_inconsistency; } int bch2_trans_mark_metadata_bucket(struct btree_trans *trans, diff --git a/fs/bcachefs/buckets.h b/fs/bcachefs/buckets.h index edbdffd508fc..e2cb7b24b220 100644 --- a/fs/bcachefs/buckets.h +++ b/fs/bcachefs/buckets.h @@ -80,22 +80,9 @@ static inline void bucket_lock(struct bucket *b) TASK_UNINTERRUPTIBLE); } -static inline struct bucket_array *gc_bucket_array(struct bch_dev *ca) -{ - return rcu_dereference_check(ca->buckets_gc, - !ca->fs || - percpu_rwsem_is_held(&ca->fs->mark_lock) || - lockdep_is_held(&ca->fs->state_lock) || - lockdep_is_held(&ca->bucket_lock)); -} - static inline struct bucket *gc_bucket(struct bch_dev *ca, size_t b) { - struct bucket_array *buckets = gc_bucket_array(ca); - - if (b - buckets->first_bucket >= buckets->nbuckets_minus_first) - return NULL; - return buckets->b + b; + return genradix_ptr(&ca->buckets_gc, b); } static inline struct bucket_gens *bucket_gens(struct bch_dev *ca) diff --git a/fs/bcachefs/buckets_types.h b/fs/bcachefs/buckets_types.h index c9698cdf866f..28bd09a253c8 100644 --- a/fs/bcachefs/buckets_types.h +++ b/fs/bcachefs/buckets_types.h @@ -19,14 +19,6 @@ struct bucket { u32 stripe_sectors; } __aligned(sizeof(long)); -struct bucket_array { - struct rcu_head rcu; - u16 first_bucket; - size_t nbuckets; - size_t nbuckets_minus_first; - struct bucket b[]; -}; - struct bucket_gens { struct rcu_head rcu; u16 first_bucket; diff --git a/fs/bcachefs/checksum.c b/fs/bcachefs/checksum.c index e7208bf1974e..ce8fc677bef9 100644 --- a/fs/bcachefs/checksum.c +++ b/fs/bcachefs/checksum.c @@ -100,13 +100,12 @@ static inline int do_encrypt_sg(struct crypto_sync_skcipher *tfm, struct scatterlist *sg, size_t len) { SYNC_SKCIPHER_REQUEST_ON_STACK(req, tfm); - int ret; skcipher_request_set_sync_tfm(req, tfm); skcipher_request_set_callback(req, 0, NULL, NULL); skcipher_request_set_crypt(req, sg, sg, len, nonce.d); - ret = crypto_skcipher_encrypt(req); + int ret = crypto_skcipher_encrypt(req); if (ret) pr_err("got error %i from crypto_skcipher_encrypt()", ret); @@ -118,38 +117,47 @@ static inline int do_encrypt(struct crypto_sync_skcipher *tfm, void *buf, size_t len) { if (!is_vmalloc_addr(buf)) { - struct scatterlist sg; + struct scatterlist sg = {}; - sg_init_table(&sg, 1); - sg_set_page(&sg, - is_vmalloc_addr(buf) - ? vmalloc_to_page(buf) - : virt_to_page(buf), - len, offset_in_page(buf)); + sg_mark_end(&sg); + sg_set_page(&sg, virt_to_page(buf), len, offset_in_page(buf)); return do_encrypt_sg(tfm, nonce, &sg, len); } else { - unsigned pages = buf_pages(buf, len); - struct scatterlist *sg; - size_t orig_len = len; - int ret, i; + DARRAY_PREALLOCATED(struct scatterlist, 4) sgl; + size_t sgl_len = 0; + int ret; - sg = kmalloc_array(pages, sizeof(*sg), GFP_KERNEL); - if (!sg) - return -BCH_ERR_ENOMEM_do_encrypt; + darray_init(&sgl); - sg_init_table(sg, pages); - - for (i = 0; i < pages; i++) { + while (len) { unsigned offset = offset_in_page(buf); - unsigned pg_len = min_t(size_t, len, PAGE_SIZE - offset); + struct scatterlist sg = { + .page_link = (unsigned long) vmalloc_to_page(buf), + .offset = offset, + .length = min(len, PAGE_SIZE - offset), + }; - sg_set_page(sg + i, vmalloc_to_page(buf), pg_len, offset); - buf += pg_len; - len -= pg_len; + if (darray_push(&sgl, sg)) { + sg_mark_end(&darray_last(sgl)); + ret = do_encrypt_sg(tfm, nonce, sgl.data, sgl_len); + if (ret) + goto err; + + nonce = nonce_add(nonce, sgl_len); + sgl_len = 0; + sgl.nr = 0; + BUG_ON(darray_push(&sgl, sg)); + } + + buf += sg.length; + len -= sg.length; + sgl_len += sg.length; } - ret = do_encrypt_sg(tfm, nonce, sg, orig_len); - kfree(sg); + sg_mark_end(&darray_last(sgl)); + ret = do_encrypt_sg(tfm, nonce, sgl.data, sgl_len); +err: + darray_exit(&sgl); return ret; } } @@ -325,39 +333,42 @@ int __bch2_encrypt_bio(struct bch_fs *c, unsigned type, { struct bio_vec bv; struct bvec_iter iter; - struct scatterlist sgl[16], *sg = sgl; - size_t bytes = 0; + DARRAY_PREALLOCATED(struct scatterlist, 4) sgl; + size_t sgl_len = 0; int ret = 0; if (!bch2_csum_type_is_encryption(type)) return 0; - sg_init_table(sgl, ARRAY_SIZE(sgl)); + darray_init(&sgl); bio_for_each_segment(bv, bio, iter) { - if (sg == sgl + ARRAY_SIZE(sgl)) { - sg_mark_end(sg - 1); + struct scatterlist sg = { + .page_link = (unsigned long) bv.bv_page, + .offset = bv.bv_offset, + .length = bv.bv_len, + }; - ret = do_encrypt_sg(c->chacha20, nonce, sgl, bytes); + if (darray_push(&sgl, sg)) { + sg_mark_end(&darray_last(sgl)); + ret = do_encrypt_sg(c->chacha20, nonce, sgl.data, sgl_len); if (ret) - return ret; + goto err; - nonce = nonce_add(nonce, bytes); - bytes = 0; + nonce = nonce_add(nonce, sgl_len); + sgl_len = 0; + sgl.nr = 0; - sg_init_table(sgl, ARRAY_SIZE(sgl)); - sg = sgl; + BUG_ON(darray_push(&sgl, sg)); } - sg_set_page(sg++, bv.bv_page, bv.bv_len, bv.bv_offset); - bytes += bv.bv_len; - } - - if (sg != sgl) { - sg_mark_end(sg - 1); - return do_encrypt_sg(c->chacha20, nonce, sgl, bytes); + sgl_len += sg.length; } + sg_mark_end(&darray_last(sgl)); + ret = do_encrypt_sg(c->chacha20, nonce, sgl.data, sgl_len); +err: + darray_exit(&sgl); return ret; } diff --git a/fs/bcachefs/clock.h b/fs/bcachefs/clock.h index 85c975dfbcfe..82c79c8baf92 100644 --- a/fs/bcachefs/clock.h +++ b/fs/bcachefs/clock.h @@ -20,15 +20,6 @@ static inline void bch2_increment_clock(struct bch_fs *c, u64 sectors, void bch2_io_clock_schedule_timeout(struct io_clock *, u64); -#define bch2_kthread_wait_event_ioclock_timeout(condition, clock, timeout)\ -({ \ - long __ret = timeout; \ - might_sleep(); \ - if (!___wait_cond_timeout(condition)) \ - __ret = __wait_event_timeout(wq, condition, timeout); \ - __ret; \ -}) - void bch2_io_timers_to_text(struct printbuf *, struct io_clock *); void bch2_io_clock_exit(struct io_clock *); diff --git a/fs/bcachefs/darray.c b/fs/bcachefs/darray.c index b7d223f85873..4f06cd8bbbe1 100644 --- a/fs/bcachefs/darray.c +++ b/fs/bcachefs/darray.c @@ -4,12 +4,12 @@ #include #include "darray.h" -int __bch2_darray_resize(darray_char *d, size_t element_size, size_t new_size, gfp_t gfp) +int __bch2_darray_resize_noprof(darray_char *d, size_t element_size, size_t new_size, gfp_t gfp) { if (new_size > d->size) { new_size = roundup_pow_of_two(new_size); - void *data = kvmalloc_array(new_size, element_size, gfp); + void *data = kvmalloc_array_noprof(new_size, element_size, gfp); if (!data) return -ENOMEM; diff --git a/fs/bcachefs/darray.h b/fs/bcachefs/darray.h index 4b340d13caac..8f4c3f0665c4 100644 --- a/fs/bcachefs/darray.h +++ b/fs/bcachefs/darray.h @@ -22,29 +22,23 @@ struct { \ typedef DARRAY(char) darray_char; typedef DARRAY(char *) darray_str; -int __bch2_darray_resize(darray_char *, size_t, size_t, gfp_t); +int __bch2_darray_resize_noprof(darray_char *, size_t, size_t, gfp_t); -static inline int __darray_resize(darray_char *d, size_t element_size, - size_t new_size, gfp_t gfp) -{ - return unlikely(new_size > d->size) - ? __bch2_darray_resize(d, element_size, new_size, gfp) - : 0; -} +#define __bch2_darray_resize(...) alloc_hooks(__bch2_darray_resize_noprof(__VA_ARGS__)) + +#define __darray_resize(_d, _element_size, _new_size, _gfp) \ + (unlikely((_new_size) > (_d)->size) \ + ? __bch2_darray_resize((_d), (_element_size), (_new_size), (_gfp))\ + : 0) #define darray_resize_gfp(_d, _new_size, _gfp) \ - unlikely(__darray_resize((darray_char *) (_d), sizeof((_d)->data[0]), (_new_size), _gfp)) + __darray_resize((darray_char *) (_d), sizeof((_d)->data[0]), (_new_size), _gfp) #define darray_resize(_d, _new_size) \ darray_resize_gfp(_d, _new_size, GFP_KERNEL) -static inline int __darray_make_room(darray_char *d, size_t t_size, size_t more, gfp_t gfp) -{ - return __darray_resize(d, t_size, d->nr + more, gfp); -} - #define darray_make_room_gfp(_d, _more, _gfp) \ - __darray_make_room((darray_char *) (_d), sizeof((_d)->data[0]), (_more), _gfp) + darray_resize_gfp((_d), (_d)->nr + (_more), _gfp) #define darray_make_room(_d, _more) \ darray_make_room_gfp(_d, _more, GFP_KERNEL) diff --git a/fs/bcachefs/data_update.c b/fs/bcachefs/data_update.c index 004894ad4147..757b9884ef55 100644 --- a/fs/bcachefs/data_update.c +++ b/fs/bcachefs/data_update.c @@ -571,7 +571,7 @@ int bch2_extent_drop_ptrs(struct btree_trans *trans, while (data_opts.kill_ptrs) { unsigned i = 0, drop = __fls(data_opts.kill_ptrs); - bch2_bkey_drop_ptrs(bkey_i_to_s(n), ptr, i++ == drop); + bch2_bkey_drop_ptrs_noerror(bkey_i_to_s(n), ptr, i++ == drop); data_opts.kill_ptrs ^= 1U << drop; } diff --git a/fs/bcachefs/dirent.c b/fs/bcachefs/dirent.c index 32bfdf19289a..84dd4a879d98 100644 --- a/fs/bcachefs/dirent.c +++ b/fs/bcachefs/dirent.c @@ -552,62 +552,30 @@ static int bch2_dir_emit(struct dir_context *ctx, struct bkey_s_c_dirent d, subv int bch2_readdir(struct bch_fs *c, subvol_inum inum, struct dir_context *ctx) { - struct btree_trans *trans = bch2_trans_get(c); - struct btree_iter iter; - struct bkey_s_c k; - subvol_inum target; - u32 snapshot; struct bkey_buf sk; - int ret; - bch2_bkey_buf_init(&sk); -retry: - bch2_trans_begin(trans); - ret = bch2_subvolume_get_snapshot(trans, inum.subvol, &snapshot); - if (ret) - goto err; + int ret = bch2_trans_run(c, + for_each_btree_key_in_subvolume_upto(trans, iter, BTREE_ID_dirents, + POS(inum.inum, ctx->pos), + POS(inum.inum, U64_MAX), + inum.subvol, 0, k, ({ + if (k.k->type != KEY_TYPE_dirent) + continue; - for_each_btree_key_upto_norestart(trans, iter, BTREE_ID_dirents, - SPOS(inum.inum, ctx->pos, snapshot), - POS(inum.inum, U64_MAX), 0, k, ret) { - if (k.k->type != KEY_TYPE_dirent) - continue; + /* dir_emit() can fault and block: */ + bch2_bkey_buf_reassemble(&sk, c, k); + struct bkey_s_c_dirent dirent = bkey_i_to_s_c_dirent(sk.k); - /* dir_emit() can fault and block: */ - bch2_bkey_buf_reassemble(&sk, c, k); - struct bkey_s_c_dirent dirent = bkey_i_to_s_c_dirent(sk.k); + subvol_inum target; + int ret2 = bch2_dirent_read_target(trans, inum, dirent, &target); + if (ret2 > 0) + continue; - ret = bch2_dirent_read_target(trans, inum, dirent, &target); - if (ret < 0) - break; - if (ret) - continue; + ret2 ?: drop_locks_do(trans, bch2_dir_emit(ctx, dirent, target)); + }))); - /* - * read_target looks up subvolumes, we can overflow paths if the - * directory has many subvolumes in it - * - * XXX: btree_trans_too_many_iters() is something we'd like to - * get rid of, and there's no good reason to be using it here - * except that we don't yet have a for_each_btree_key() helper - * that does subvolume_get_snapshot(). - */ - ret = drop_locks_do(trans, - bch2_dir_emit(ctx, dirent, target)) ?: - btree_trans_too_many_iters(trans); - if (ret) { - ret = ret < 0 ? ret : 0; - break; - } - } - bch2_trans_iter_exit(trans, &iter); -err: - if (bch2_err_matches(ret, BCH_ERR_transaction_restart)) - goto retry; - - bch2_trans_put(trans); bch2_bkey_buf_exit(&sk, c); - return ret; + return ret < 0 ? ret : 0; } diff --git a/fs/bcachefs/ec.c b/fs/bcachefs/ec.c index 141a4c63142f..1587c6e1866a 100644 --- a/fs/bcachefs/ec.c +++ b/fs/bcachefs/ec.c @@ -18,6 +18,7 @@ #include "ec.h" #include "error.h" #include "io_read.h" +#include "io_write.h" #include "keylist.h" #include "recovery.h" #include "replicas.h" @@ -146,12 +147,18 @@ void bch2_stripe_to_text(struct printbuf *out, struct bch_fs *c, bch2_prt_csum_type(out, s.csum_type); prt_printf(out, " gran %u", 1U << s.csum_granularity_bits); + if (s.disk_label) { + prt_str(out, " label"); + bch2_disk_path_to_text(out, c, s.disk_label - 1); + } + for (unsigned i = 0; i < s.nr_blocks; i++) { const struct bch_extent_ptr *ptr = sp->ptrs + i; if ((void *) ptr >= bkey_val_end(k)) break; + prt_char(out, ' '); bch2_extent_ptr_to_text(out, c, ptr); if (s.csum_type < BCH_CSUM_NR && @@ -192,7 +199,7 @@ static int __mark_stripe_bucket(struct btree_trans *trans, a->dirty_sectors, a->stripe, s.k->p.offset, (bch2_bkey_val_to_text(&buf, c, s.s_c), buf.buf))) { - ret = -EIO; + ret = -BCH_ERR_mark_stripe; goto err; } @@ -203,7 +210,7 @@ static int __mark_stripe_bucket(struct btree_trans *trans, a->dirty_sectors, a->cached_sectors, (bch2_bkey_val_to_text(&buf, c, s.s_c), buf.buf))) { - ret = -EIO; + ret = -BCH_ERR_mark_stripe; goto err; } } else { @@ -213,7 +220,7 @@ static int __mark_stripe_bucket(struct btree_trans *trans, bucket.inode, bucket.offset, a->gen, a->stripe, (bch2_bkey_val_to_text(&buf, c, s.s_c), buf.buf))) { - ret = -EIO; + ret = -BCH_ERR_mark_stripe; goto err; } @@ -223,7 +230,7 @@ static int __mark_stripe_bucket(struct btree_trans *trans, bch2_data_type_str(a->data_type), bch2_data_type_str(data_type), (bch2_bkey_val_to_text(&buf, c, s.s_c), buf.buf))) { - ret = -EIO; + ret = -BCH_ERR_mark_stripe; goto err; } @@ -235,7 +242,7 @@ static int __mark_stripe_bucket(struct btree_trans *trans, a->dirty_sectors, a->cached_sectors, (bch2_bkey_val_to_text(&buf, c, s.s_c), buf.buf))) { - ret = -EIO; + ret = -BCH_ERR_mark_stripe; goto err; } } @@ -273,8 +280,8 @@ static int mark_stripe_bucket(struct btree_trans *trans, struct bch_dev *ca = bch2_dev_tryget(c, ptr->dev); if (unlikely(!ca)) { - if (!(flags & BTREE_TRIGGER_overwrite)) - ret = -EIO; + if (ptr->dev != BCH_SB_MEMBER_INVALID && !(flags & BTREE_TRIGGER_overwrite)) + ret = -BCH_ERR_mark_stripe; goto err; } @@ -293,7 +300,7 @@ static int mark_stripe_bucket(struct btree_trans *trans, if (bch2_fs_inconsistent_on(!g, c, "reference to invalid bucket on device %u\n %s", ptr->dev, (bch2_bkey_val_to_text(&buf, c, s.s_c), buf.buf))) { - ret = -EIO; + ret = -BCH_ERR_mark_stripe; goto err_unlock; } @@ -351,6 +358,19 @@ static int mark_stripe_buckets(struct btree_trans *trans, return 0; } +static inline void stripe_to_mem(struct stripe *m, const struct bch_stripe *s) +{ + m->sectors = le16_to_cpu(s->sectors); + m->algorithm = s->algorithm; + m->nr_blocks = s->nr_blocks; + m->nr_redundant = s->nr_redundant; + m->disk_label = s->disk_label; + m->blocks_nonempty = 0; + + for (unsigned i = 0; i < s->nr_blocks; i++) + m->blocks_nonempty += !!stripe_blockcount_get(s, i); +} + int bch2_trigger_stripe(struct btree_trans *trans, enum btree_id btree, unsigned level, struct bkey_s_c old, struct bkey_s _new, @@ -467,14 +487,7 @@ int bch2_trigger_stripe(struct btree_trans *trans, memset(m, 0, sizeof(*m)); } else { - m->sectors = le16_to_cpu(new_s->sectors); - m->algorithm = new_s->algorithm; - m->nr_blocks = new_s->nr_blocks; - m->nr_redundant = new_s->nr_redundant; - m->blocks_nonempty = 0; - - for (unsigned i = 0; i < new_s->nr_blocks; i++) - m->blocks_nonempty += !!stripe_blockcount_get(new_s, i); + stripe_to_mem(m, new_s); if (!old_s) bch2_stripes_heap_insert(c, m, idx); @@ -816,13 +829,16 @@ err: } /* recovery read path: */ -int bch2_ec_read_extent(struct btree_trans *trans, struct bch_read_bio *rbio) +int bch2_ec_read_extent(struct btree_trans *trans, struct bch_read_bio *rbio, + struct bkey_s_c orig_k) { struct bch_fs *c = trans->c; - struct ec_stripe_buf *buf; + struct ec_stripe_buf *buf = NULL; struct closure cl; struct bch_stripe *v; unsigned i, offset; + const char *msg = NULL; + struct printbuf msgbuf = PRINTBUF; int ret = 0; closure_init_stack(&cl); @@ -835,32 +851,28 @@ int bch2_ec_read_extent(struct btree_trans *trans, struct bch_read_bio *rbio) ret = lockrestart_do(trans, get_stripe_key_trans(trans, rbio->pick.ec.idx, buf)); if (ret) { - bch_err_ratelimited(c, - "error doing reconstruct read: error %i looking up stripe", ret); - kfree(buf); - return -EIO; + msg = "stripe not found"; + goto err; } v = &bkey_i_to_stripe(&buf->key)->v; if (!bch2_ptr_matches_stripe(v, rbio->pick)) { - bch_err_ratelimited(c, - "error doing reconstruct read: pointer doesn't match stripe"); - ret = -EIO; + msg = "pointer doesn't match stripe"; goto err; } offset = rbio->bio.bi_iter.bi_sector - v->ptrs[rbio->pick.ec.block].offset; if (offset + bio_sectors(&rbio->bio) > le16_to_cpu(v->sectors)) { - bch_err_ratelimited(c, - "error doing reconstruct read: read is bigger than stripe"); - ret = -EIO; + msg = "read is bigger than stripe"; goto err; } ret = ec_stripe_buf_init(buf, offset, bio_sectors(&rbio->bio)); - if (ret) + if (ret) { + msg = "-ENOMEM"; goto err; + } for (i = 0; i < v->nr_blocks; i++) ec_block_io(c, buf, REQ_OP_READ, i, &cl); @@ -868,9 +880,7 @@ int bch2_ec_read_extent(struct btree_trans *trans, struct bch_read_bio *rbio) closure_sync(&cl); if (ec_nr_failed(buf) > v->nr_redundant) { - bch_err_ratelimited(c, - "error doing reconstruct read: unable to read enough blocks"); - ret = -EIO; + msg = "unable to read enough blocks"; goto err; } @@ -882,10 +892,17 @@ int bch2_ec_read_extent(struct btree_trans *trans, struct bch_read_bio *rbio) memcpy_to_bio(&rbio->bio, rbio->bio.bi_iter, buf->data[rbio->pick.ec.block] + ((offset - buf->offset) << 9)); -err: +out: ec_stripe_buf_exit(buf); kfree(buf); return ret; +err: + bch2_bkey_val_to_text(&msgbuf, c, orig_k); + bch_err_ratelimited(c, + "error doing reconstruct read: %s\n %s", msg, msgbuf.buf); + printbuf_exit(&msgbuf);; + ret = -BCH_ERR_stripe_reconstruct; + goto out; } /* stripe bucket accounting: */ @@ -1305,7 +1322,7 @@ static int ec_stripe_update_extent(struct btree_trans *trans, bkey_reassemble(n, k); - bch2_bkey_drop_ptrs(bkey_i_to_s(n), ptr, ptr->dev != dev); + bch2_bkey_drop_ptrs_noerror(bkey_i_to_s(n), ptr, ptr->dev != dev); ec_ptr = bch2_bkey_has_device(bkey_i_to_s(n), dev); BUG_ON(!ec_ptr); @@ -1555,10 +1572,12 @@ void bch2_ec_do_stripe_creates(struct bch_fs *c) bch2_write_ref_put(c, BCH_WRITE_REF_stripe_create); } -static void ec_stripe_set_pending(struct bch_fs *c, struct ec_stripe_head *h) +static void ec_stripe_new_set_pending(struct bch_fs *c, struct ec_stripe_head *h) { struct ec_stripe_new *s = h->s; + lockdep_assert_held(&h->lock); + BUG_ON(!s->allocated && !s->err); h->s = NULL; @@ -1571,6 +1590,12 @@ static void ec_stripe_set_pending(struct bch_fs *c, struct ec_stripe_head *h) ec_stripe_new_put(c, s, STRIPE_REF_io); } +static void ec_stripe_new_cancel(struct bch_fs *c, struct ec_stripe_head *h, int err) +{ + h->s->err = err; + ec_stripe_new_set_pending(c, h); +} + void bch2_ec_bucket_cancel(struct bch_fs *c, struct open_bucket *ob) { struct ec_stripe_new *s = ob->ec; @@ -1641,7 +1666,8 @@ static void ec_stripe_key_init(struct bch_fs *c, struct bkey_i *k, unsigned nr_data, unsigned nr_parity, - unsigned stripe_size) + unsigned stripe_size, + unsigned disk_label) { struct bkey_i_stripe *s = bkey_stripe_init(k); unsigned u64s; @@ -1652,7 +1678,7 @@ static void ec_stripe_key_init(struct bch_fs *c, s->v.nr_redundant = nr_parity; s->v.csum_granularity_bits = ilog2(c->opts.encoded_extent_max >> 9); s->v.csum_type = BCH_CSUM_crc32c; - s->v.pad = 0; + s->v.disk_label = disk_label; while ((u64s = stripe_val_u64s(&s->v)) > BKEY_VAL_U64s_MAX) { BUG_ON(1 << s->v.csum_granularity_bits >= @@ -1685,14 +1711,70 @@ static int ec_new_stripe_alloc(struct bch_fs *c, struct ec_stripe_head *h) s->nr_parity = h->redundancy; ec_stripe_key_init(c, &s->new_stripe.key, - s->nr_data, s->nr_parity, h->blocksize); + s->nr_data, s->nr_parity, + h->blocksize, h->disk_label); h->s = s; + h->nr_created++; return 0; } +static void ec_stripe_head_devs_update(struct bch_fs *c, struct ec_stripe_head *h) +{ + struct bch_devs_mask devs = h->devs; + + rcu_read_lock(); + h->devs = target_rw_devs(c, BCH_DATA_user, h->disk_label + ? group_to_target(h->disk_label - 1) + : 0); + unsigned nr_devs = dev_mask_nr(&h->devs); + + for_each_member_device_rcu(c, ca, &h->devs) + if (!ca->mi.durability) + __clear_bit(ca->dev_idx, h->devs.d); + unsigned nr_devs_with_durability = dev_mask_nr(&h->devs); + + h->blocksize = pick_blocksize(c, &h->devs); + + h->nr_active_devs = 0; + for_each_member_device_rcu(c, ca, &h->devs) + if (ca->mi.bucket_size == h->blocksize) + h->nr_active_devs++; + + rcu_read_unlock(); + + /* + * If we only have redundancy + 1 devices, we're better off with just + * replication: + */ + h->insufficient_devs = h->nr_active_devs < h->redundancy + 2; + + if (h->insufficient_devs) { + const char *err; + + if (nr_devs < h->redundancy + 2) + err = NULL; + else if (nr_devs_with_durability < h->redundancy + 2) + err = "cannot use durability=0 devices"; + else + err = "mismatched bucket sizes"; + + if (err) + bch_err(c, "insufficient devices available to create stripe (have %u, need %u): %s", + h->nr_active_devs, h->redundancy + 2, err); + } + + struct bch_devs_mask devs_leaving; + bitmap_andnot(devs_leaving.d, devs.d, h->devs.d, BCH_SB_MEMBERS_MAX); + + if (h->s && !h->s->allocated && dev_mask_nr(&devs_leaving)) + ec_stripe_new_cancel(c, h, -EINTR); + + h->rw_devs_change_count = c->rw_devs_change_count; +} + static struct ec_stripe_head * -ec_new_stripe_head_alloc(struct bch_fs *c, unsigned target, +ec_new_stripe_head_alloc(struct bch_fs *c, unsigned disk_label, unsigned algo, unsigned redundancy, enum bch_watermark watermark) { @@ -1705,34 +1787,11 @@ ec_new_stripe_head_alloc(struct bch_fs *c, unsigned target, mutex_init(&h->lock); BUG_ON(!mutex_trylock(&h->lock)); - h->target = target; + h->disk_label = disk_label; h->algo = algo; h->redundancy = redundancy; h->watermark = watermark; - rcu_read_lock(); - h->devs = target_rw_devs(c, BCH_DATA_user, target); - - for_each_member_device_rcu(c, ca, &h->devs) - if (!ca->mi.durability) - __clear_bit(ca->dev_idx, h->devs.d); - - h->blocksize = pick_blocksize(c, &h->devs); - - for_each_member_device_rcu(c, ca, &h->devs) - if (ca->mi.bucket_size == h->blocksize) - h->nr_active_devs++; - - rcu_read_unlock(); - - /* - * If we only have redundancy + 1 devices, we're better off with just - * replication: - */ - if (h->nr_active_devs < h->redundancy + 2) - bch_err(c, "insufficient devices available to create stripe (have %u, need %u) - mismatched bucket sizes?", - h->nr_active_devs, h->redundancy + 2); - list_add(&h->list, &c->ec_stripe_head_list); return h; } @@ -1743,14 +1802,14 @@ void bch2_ec_stripe_head_put(struct bch_fs *c, struct ec_stripe_head *h) h->s->allocated && bitmap_weight(h->s->blocks_allocated, h->s->nr_data) == h->s->nr_data) - ec_stripe_set_pending(c, h); + ec_stripe_new_set_pending(c, h); mutex_unlock(&h->lock); } static struct ec_stripe_head * __bch2_ec_stripe_head_get(struct btree_trans *trans, - unsigned target, + unsigned disk_label, unsigned algo, unsigned redundancy, enum bch_watermark watermark) @@ -1768,27 +1827,32 @@ __bch2_ec_stripe_head_get(struct btree_trans *trans, if (test_bit(BCH_FS_going_ro, &c->flags)) { h = ERR_PTR(-BCH_ERR_erofs_no_writes); - goto found; + goto err; } list_for_each_entry(h, &c->ec_stripe_head_list, list) - if (h->target == target && + if (h->disk_label == disk_label && h->algo == algo && h->redundancy == redundancy && h->watermark == watermark) { ret = bch2_trans_mutex_lock(trans, &h->lock); - if (ret) + if (ret) { h = ERR_PTR(ret); + goto err; + } goto found; } - h = ec_new_stripe_head_alloc(c, target, algo, redundancy, watermark); + h = ec_new_stripe_head_alloc(c, disk_label, algo, redundancy, watermark); found: - if (!IS_ERR_OR_NULL(h) && - h->nr_active_devs < h->redundancy + 2) { + if (h->rw_devs_change_count != c->rw_devs_change_count) + ec_stripe_head_devs_update(c, h); + + if (h->insufficient_devs) { mutex_unlock(&h->lock); h = NULL; } +err: mutex_unlock(&c->ec_stripe_head_lock); return h; } @@ -1878,7 +1942,6 @@ static int new_stripe_alloc_buckets(struct btree_trans *trans, struct ec_stripe_ return 0; } -/* XXX: doesn't obey target: */ static s64 get_existing_stripe(struct bch_fs *c, struct ec_stripe_head *head) { @@ -1901,7 +1964,8 @@ static s64 get_existing_stripe(struct bch_fs *c, m = genradix_ptr(&c->stripes, stripe_idx); - if (m->algorithm == head->algo && + if (m->disk_label == head->disk_label && + m->algorithm == head->algo && m->nr_redundant == head->redundancy && m->sectors == head->blocksize && m->blocks_nonempty < m->nr_blocks - m->nr_redundant && @@ -2046,9 +2110,19 @@ struct ec_stripe_head *bch2_ec_stripe_head_get(struct btree_trans *trans, struct bch_fs *c = trans->c; struct ec_stripe_head *h; bool waiting = false; + unsigned disk_label = 0; + struct target t = target_decode(target); int ret; - h = __bch2_ec_stripe_head_get(trans, target, algo, redundancy, watermark); + if (t.type == TARGET_GROUP) { + if (t.group > U8_MAX) { + bch_err(c, "cannot create a stripe when disk_label > U8_MAX"); + return NULL; + } + disk_label = t.group + 1; /* 0 == no label */ + } + + h = __bch2_ec_stripe_head_get(trans, disk_label, algo, redundancy, watermark); if (IS_ERR_OR_NULL(h)) return h; @@ -2126,6 +2200,73 @@ err: return ERR_PTR(ret); } +/* device removal */ + +static int bch2_invalidate_stripe_to_dev(struct btree_trans *trans, struct bkey_s_c k_a) +{ + struct bch_alloc_v4 a_convert; + const struct bch_alloc_v4 *a = bch2_alloc_to_v4(k_a, &a_convert); + + if (!a->stripe) + return 0; + + if (a->stripe_sectors) { + bch_err(trans->c, "trying to invalidate device in stripe when bucket has stripe data"); + return -BCH_ERR_invalidate_stripe_to_dev; + } + + struct btree_iter iter; + struct bkey_i_stripe *s = + bch2_bkey_get_mut_typed(trans, &iter, BTREE_ID_stripes, POS(0, a->stripe), + BTREE_ITER_slots, stripe); + int ret = PTR_ERR_OR_ZERO(s); + if (ret) + return ret; + + struct disk_accounting_pos acc = { + .type = BCH_DISK_ACCOUNTING_replicas, + }; + + s64 sectors = 0; + for (unsigned i = 0; i < s->v.nr_blocks; i++) + sectors -= stripe_blockcount_get(&s->v, i); + + bch2_bkey_to_replicas(&acc.replicas, bkey_i_to_s_c(&s->k_i)); + acc.replicas.data_type = BCH_DATA_user; + ret = bch2_disk_accounting_mod(trans, &acc, §ors, 1, false); + if (ret) + goto err; + + struct bkey_ptrs ptrs = bch2_bkey_ptrs(bkey_i_to_s(&s->k_i)); + bkey_for_each_ptr(ptrs, ptr) + if (ptr->dev == k_a.k->p.inode) + ptr->dev = BCH_SB_MEMBER_INVALID; + + sectors = -sectors; + + bch2_bkey_to_replicas(&acc.replicas, bkey_i_to_s_c(&s->k_i)); + acc.replicas.data_type = BCH_DATA_user; + ret = bch2_disk_accounting_mod(trans, &acc, §ors, 1, false); + if (ret) + goto err; +err: + bch2_trans_iter_exit(trans, &iter); + return ret; +} + +int bch2_dev_remove_stripes(struct bch_fs *c, unsigned dev_idx) +{ + return bch2_trans_run(c, + for_each_btree_key_upto_commit(trans, iter, + BTREE_ID_alloc, POS(dev_idx, 0), POS(dev_idx, U64_MAX), + BTREE_ITER_intent, k, + NULL, NULL, 0, ({ + bch2_invalidate_stripe_to_dev(trans, k); + }))); +} + +/* startup/shutdown */ + static void __bch2_ec_stop(struct bch_fs *c, struct bch_dev *ca) { struct ec_stripe_head *h; @@ -2151,8 +2292,7 @@ static void __bch2_ec_stop(struct bch_fs *c, struct bch_dev *ca) } goto unlock; found: - h->s->err = -BCH_ERR_erofs_no_writes; - ec_stripe_set_pending(c, h); + ec_stripe_new_cancel(c, h, -BCH_ERR_erofs_no_writes); unlock: mutex_unlock(&h->lock); } @@ -2197,17 +2337,9 @@ int bch2_stripes_read(struct bch_fs *c) if (ret) break; - const struct bch_stripe *s = bkey_s_c_to_stripe(k).v; - struct stripe *m = genradix_ptr(&c->stripes, k.k->p.offset); - m->sectors = le16_to_cpu(s->sectors); - m->algorithm = s->algorithm; - m->nr_blocks = s->nr_blocks; - m->nr_redundant = s->nr_redundant; - m->blocks_nonempty = 0; - for (unsigned i = 0; i < s->nr_blocks; i++) - m->blocks_nonempty += !!stripe_blockcount_get(s, i); + stripe_to_mem(m, bkey_s_c_to_stripe(k).v); bch2_stripes_heap_insert(c, m, k.k->p.offset); 0; @@ -2252,6 +2384,8 @@ static void bch2_new_stripe_to_text(struct printbuf *out, struct bch_fs *c, for_each_set_bit(i, s->blocks_gotten, v->nr_blocks) prt_printf(out, " %u", s->blocks[i]); prt_newline(out); + bch2_bkey_val_to_text(out, c, bkey_i_to_s_c(&s->new_stripe.key)); + prt_newline(out); } void bch2_new_stripes_to_text(struct printbuf *out, struct bch_fs *c) @@ -2261,9 +2395,10 @@ void bch2_new_stripes_to_text(struct printbuf *out, struct bch_fs *c) mutex_lock(&c->ec_stripe_head_lock); list_for_each_entry(h, &c->ec_stripe_head_list, list) { - prt_printf(out, "target %u algo %u redundancy %u %s:\n", - h->target, h->algo, h->redundancy, - bch2_watermarks[h->watermark]); + prt_printf(out, "disk label %u algo %u redundancy %u %s nr created %llu:\n", + h->disk_label, h->algo, h->redundancy, + bch2_watermarks[h->watermark], + h->nr_created); if (h->s) bch2_new_stripe_to_text(out, c, h->s); diff --git a/fs/bcachefs/ec.h b/fs/bcachefs/ec.h index 9baf3411a8f9..43326370b410 100644 --- a/fs/bcachefs/ec.h +++ b/fs/bcachefs/ec.h @@ -188,10 +188,15 @@ struct ec_stripe_head { struct list_head list; struct mutex lock; - unsigned target; + unsigned disk_label; unsigned algo; unsigned redundancy; enum bch_watermark watermark; + bool insufficient_devs; + + unsigned long rw_devs_change_count; + + u64 nr_created; struct bch_devs_mask devs; unsigned nr_active_devs; @@ -204,7 +209,7 @@ struct ec_stripe_head { struct ec_stripe_new *s; }; -int bch2_ec_read_extent(struct btree_trans *, struct bch_read_bio *); +int bch2_ec_read_extent(struct btree_trans *, struct bch_read_bio *, struct bkey_s_c); void *bch2_writepoint_ec_buf(struct bch_fs *, struct write_point *); @@ -249,6 +254,8 @@ static inline void ec_stripe_new_put(struct bch_fs *c, struct ec_stripe_new *s, } } +int bch2_dev_remove_stripes(struct bch_fs *, unsigned); + void bch2_ec_stop_dev(struct bch_fs *, struct bch_dev *); void bch2_fs_ec_stop(struct bch_fs *); void bch2_fs_ec_flush(struct bch_fs *); diff --git a/fs/bcachefs/ec_format.h b/fs/bcachefs/ec_format.h index 44ce88ba08d7..64ef52e00078 100644 --- a/fs/bcachefs/ec_format.h +++ b/fs/bcachefs/ec_format.h @@ -11,7 +11,14 @@ struct bch_stripe { __u8 csum_granularity_bits; __u8 csum_type; - __u8 pad; + + /* + * XXX: targets should be 16 bits - fix this if we ever do a stripe_v2 + * + * we can manage with this because this only needs to point to a + * disk label, not a target: + */ + __u8 disk_label; struct bch_extent_ptr ptrs[]; } __packed __aligned(8); diff --git a/fs/bcachefs/ec_types.h b/fs/bcachefs/ec_types.h index 1df03dccfc72..8d1e70e830ac 100644 --- a/fs/bcachefs/ec_types.h +++ b/fs/bcachefs/ec_types.h @@ -16,6 +16,7 @@ struct stripe { u8 nr_blocks; u8 nr_redundant; u8 blocks_nonempty; + u8 disk_label; }; struct gc_stripe { diff --git a/fs/bcachefs/errcode.h b/fs/bcachefs/errcode.h index 742dcdd3e5d7..60b7875adada 100644 --- a/fs/bcachefs/errcode.h +++ b/fs/bcachefs/errcode.h @@ -119,8 +119,8 @@ x(EEXIST, EEXIST_str_hash_set) \ x(EEXIST, EEXIST_discard_in_flight_add) \ x(EEXIST, EEXIST_subvolume_create) \ - x(0, open_buckets_empty) \ - x(0, freelist_empty) \ + x(ENOSPC, open_buckets_empty) \ + x(ENOSPC, freelist_empty) \ x(BCH_ERR_freelist_empty, no_buckets_found) \ x(0, transaction_restart) \ x(BCH_ERR_transaction_restart, transaction_restart_fault_inject) \ @@ -244,6 +244,16 @@ x(EIO, btree_node_read_error) \ x(EIO, btree_node_read_validate_error) \ x(EIO, btree_need_topology_repair) \ + x(EIO, bucket_ref_update) \ + x(EIO, trigger_pointer) \ + x(EIO, trigger_stripe_pointer) \ + x(EIO, metadata_bucket_inconsistency) \ + x(EIO, mark_stripe) \ + x(EIO, stripe_reconstruct) \ + x(EIO, key_type_error) \ + x(EIO, no_device_to_read_from) \ + x(EIO, missing_indirect_extent) \ + x(EIO, invalidate_stripe_to_dev) \ x(BCH_ERR_btree_node_read_err, btree_node_read_err_fixable) \ x(BCH_ERR_btree_node_read_err, btree_node_read_err_want_retry) \ x(BCH_ERR_btree_node_read_err, btree_node_read_err_must_retry) \ diff --git a/fs/bcachefs/extents.c b/fs/bcachefs/extents.c index 324303bf4353..cc0d22085aef 100644 --- a/fs/bcachefs/extents.c +++ b/fs/bcachefs/extents.c @@ -115,7 +115,7 @@ int bch2_bkey_pick_read_device(struct bch_fs *c, struct bkey_s_c k, int ret = 0; if (k.k->type == KEY_TYPE_error) - return -EIO; + return -BCH_ERR_key_type_error; rcu_read_lock(); bkey_for_each_ptr_decode(k.k, ptrs, p, entry) { @@ -133,7 +133,7 @@ int bch2_bkey_pick_read_device(struct bch_fs *c, struct bkey_s_c k, * read: */ if (!ret && !p.ptr.cached) - ret = -EIO; + ret = -BCH_ERR_no_device_to_read_from; struct bch_dev *ca = bch2_dev_rcu(c, p.ptr.dev); @@ -146,16 +146,13 @@ int bch2_bkey_pick_read_device(struct bch_fs *c, struct bkey_s_c k, ? f->idx : f->idx + 1; - if (!p.idx && !ca) + if (!p.idx && (!ca || !bch2_dev_is_readable(ca))) p.idx++; if (!p.idx && p.has_ec && bch2_force_reconstruct_read) p.idx++; - if (!p.idx && !bch2_dev_is_readable(ca)) - p.idx++; - - if (p.idx >= (unsigned) p.has_ec + 1) + if (p.idx > (unsigned) p.has_ec) continue; if (ret > 0 && !ptr_better(c, p, *pick)) @@ -821,6 +818,18 @@ void bch2_bkey_drop_ptr_noerror(struct bkey_s k, struct bch_extent_ptr *ptr) void bch2_bkey_drop_ptr(struct bkey_s k, struct bch_extent_ptr *ptr) { + if (k.k->type != KEY_TYPE_stripe) { + struct bkey_ptrs_c ptrs = bch2_bkey_ptrs_c(k.s_c); + const union bch_extent_entry *entry; + struct extent_ptr_decoded p; + + bkey_for_each_ptr_decode(k.k, ptrs, p, entry) + if (p.ptr.dev == ptr->dev && p.has_ec) { + ptr->dev = BCH_SB_MEMBER_INVALID; + return; + } + } + bool have_dirty = bch2_bkey_dirty_devs(k.s_c).nr; bch2_bkey_drop_ptr_noerror(k, ptr); @@ -848,10 +857,7 @@ void bch2_bkey_drop_device(struct bkey_s k, unsigned dev) void bch2_bkey_drop_device_noerror(struct bkey_s k, unsigned dev) { - struct bch_extent_ptr *ptr = bch2_bkey_has_device(k, dev); - - if (ptr) - bch2_bkey_drop_ptr_noerror(k, ptr); + bch2_bkey_drop_ptrs_noerror(k, ptr, ptr->dev == dev); } const struct bch_extent_ptr *bch2_bkey_has_device_c(struct bkey_s_c k, unsigned dev) @@ -1021,7 +1027,7 @@ void bch2_extent_ptr_to_text(struct printbuf *out, struct bch_fs *c, const struc { out->atomic++; rcu_read_lock(); - struct bch_dev *ca = bch2_dev_rcu(c, ptr->dev); + struct bch_dev *ca = bch2_dev_rcu_noerror(c, ptr->dev); if (!ca) { prt_printf(out, "ptr: %u:%llu gen %u%s", ptr->dev, (u64) ptr->offset, ptr->gen, @@ -1125,8 +1131,9 @@ static int extent_ptr_validate(struct bch_fs *c, { int ret = 0; + /* bad pointers are repaired by check_fix_ptrs(): */ rcu_read_lock(); - struct bch_dev *ca = bch2_dev_rcu(c, ptr->dev); + struct bch_dev *ca = bch2_dev_rcu_noerror(c, ptr->dev); if (!ca) { rcu_read_unlock(); return 0; diff --git a/fs/bcachefs/extents.h b/fs/bcachefs/extents.h index 42a7c6d820a0..ed5001dd662e 100644 --- a/fs/bcachefs/extents.h +++ b/fs/bcachefs/extents.h @@ -357,7 +357,7 @@ out: \ __bkey_for_each_ptr_decode(_k, (_p).start, (_p).end, \ _ptr, _entry) -#define bkey_crc_next(_k, _start, _end, _crc, _iter) \ +#define bkey_crc_next(_k, _end, _crc, _iter) \ ({ \ __bkey_extent_entry_for_each_from(_iter, _end, _iter) \ if (extent_entry_is_crc(_iter)) { \ @@ -372,7 +372,7 @@ out: \ #define __bkey_for_each_crc(_k, _start, _end, _crc, _iter) \ for ((_crc) = bch2_extent_crc_unpack(_k, NULL), \ (_iter) = (_start); \ - bkey_crc_next(_k, _start, _end, _crc, _iter); \ + bkey_crc_next(_k, _end, _crc, _iter); \ (_iter) = extent_entry_next(_iter)) #define bkey_for_each_crc(_k, _p, _crc, _iter) \ @@ -611,9 +611,6 @@ unsigned bch2_extent_ptr_desired_durability(struct bch_fs *, struct extent_ptr_d unsigned bch2_extent_ptr_durability(struct bch_fs *, struct extent_ptr_decoded *); unsigned bch2_bkey_durability(struct bch_fs *, struct bkey_s_c); -void bch2_bkey_drop_device(struct bkey_s, unsigned); -void bch2_bkey_drop_device_noerror(struct bkey_s, unsigned); - const struct bch_extent_ptr *bch2_bkey_has_device_c(struct bkey_s_c, unsigned); static inline struct bch_extent_ptr *bch2_bkey_has_device(struct bkey_s k, unsigned dev) @@ -652,6 +649,23 @@ void bch2_extent_ptr_decoded_append(struct bkey_i *, void bch2_bkey_drop_ptr_noerror(struct bkey_s, struct bch_extent_ptr *); void bch2_bkey_drop_ptr(struct bkey_s, struct bch_extent_ptr *); +void bch2_bkey_drop_device_noerror(struct bkey_s, unsigned); +void bch2_bkey_drop_device(struct bkey_s, unsigned); + +#define bch2_bkey_drop_ptrs_noerror(_k, _ptr, _cond) \ +do { \ + __label__ _again; \ + struct bkey_ptrs _ptrs; \ +_again: \ + _ptrs = bch2_bkey_ptrs(_k); \ + \ + bkey_for_each_ptr(_ptrs, _ptr) \ + if (_cond) { \ + bch2_bkey_drop_ptr_noerror(_k, _ptr); \ + goto _again; \ + } \ +} while (0) + #define bch2_bkey_drop_ptrs(_k, _ptr, _cond) \ do { \ __label__ _again; \ diff --git a/fs/bcachefs/fs-common.c b/fs/bcachefs/fs-common.c index 508d029ac53d..7e10a9ddcfd9 100644 --- a/fs/bcachefs/fs-common.c +++ b/fs/bcachefs/fs-common.c @@ -42,7 +42,8 @@ int bch2_create_trans(struct btree_trans *trans, if (ret) goto err; - ret = bch2_inode_peek(trans, &dir_iter, dir_u, dir, BTREE_ITER_intent); + ret = bch2_inode_peek(trans, &dir_iter, dir_u, dir, + BTREE_ITER_intent|BTREE_ITER_with_updates); if (ret) goto err; @@ -163,7 +164,7 @@ int bch2_create_trans(struct btree_trans *trans, name, dir_target, &dir_offset, - STR_HASH_must_create); + STR_HASH_must_create|BTREE_ITER_with_updates); if (ret) goto err; diff --git a/fs/bcachefs/fs-io-buffered.c b/fs/bcachefs/fs-io-buffered.c index ff60c041abe5..48a1ab9a649b 100644 --- a/fs/bcachefs/fs-io-buffered.c +++ b/fs/bcachefs/fs-io-buffered.c @@ -151,7 +151,6 @@ static void bchfs_read(struct btree_trans *trans, struct bkey_buf sk; int flags = BCH_READ_RETRY_IF_STALE| BCH_READ_MAY_PROMOTE; - u32 snapshot; int ret = 0; rbio->c = c; @@ -159,29 +158,23 @@ static void bchfs_read(struct btree_trans *trans, rbio->subvol = inum.subvol; bch2_bkey_buf_init(&sk); -retry: bch2_trans_begin(trans); - iter = (struct btree_iter) { NULL }; - - ret = bch2_subvolume_get_snapshot(trans, inum.subvol, &snapshot); - if (ret) - goto err; - bch2_trans_iter_init(trans, &iter, BTREE_ID_extents, - SPOS(inum.inum, rbio->bio.bi_iter.bi_sector, snapshot), + POS(inum.inum, rbio->bio.bi_iter.bi_sector), BTREE_ITER_slots); while (1) { struct bkey_s_c k; unsigned bytes, sectors, offset_into_extent; enum btree_id data_btree = BTREE_ID_extents; - /* - * read_extent -> io_time_reset may cause a transaction restart - * without returning an error, we need to check for that here: - */ - ret = bch2_trans_relock(trans); + bch2_trans_begin(trans); + + u32 snapshot; + ret = bch2_subvolume_get_snapshot(trans, inum.subvol, &snapshot); if (ret) - break; + goto err; + + bch2_btree_iter_set_snapshot(&iter, snapshot); bch2_btree_iter_set_pos(&iter, POS(inum.inum, rbio->bio.bi_iter.bi_sector)); @@ -189,7 +182,7 @@ retry: k = bch2_btree_iter_peek_slot(&iter); ret = bkey_err(k); if (ret) - break; + goto err; offset_into_extent = iter.pos.offset - bkey_start_offset(k.k); @@ -200,7 +193,7 @@ retry: ret = bch2_read_indirect_extent(trans, &data_btree, &offset_into_extent, &sk); if (ret) - break; + goto err; k = bkey_i_to_s_c(sk.k); @@ -210,7 +203,7 @@ retry: ret = readpage_bio_extend(trans, readpages_iter, &rbio->bio, sectors, extent_partial_reads_expensive(k)); if (ret) - break; + goto err; } bytes = min(sectors, bio_sectors(&rbio->bio)) << 9; @@ -229,17 +222,13 @@ retry: swap(rbio->bio.bi_iter.bi_size, bytes); bio_advance(&rbio->bio, bytes); - - ret = btree_trans_too_many_iters(trans); - if (ret) +err: + if (ret && + !bch2_err_matches(ret, BCH_ERR_transaction_restart)) break; } -err: bch2_trans_iter_exit(trans, &iter); - if (bch2_err_matches(ret, BCH_ERR_transaction_restart)) - goto retry; - if (ret) { bch_err_inum_offset_ratelimited(c, iter.pos.inode, @@ -486,7 +475,7 @@ static void bch2_writepage_io_alloc(struct bch_fs *c, op->nr_replicas = nr_replicas; op->res.nr_replicas = nr_replicas; op->write_point = writepoint_hashed(inode->ei_last_dirtied); - op->subvol = inode->ei_subvol; + op->subvol = inode->ei_inum.subvol; op->pos = POS(inode->v.i_ino, sector); op->end_io = bch2_writepage_io_done; op->devs_need_flush = &inode->ei_devs_need_flush; diff --git a/fs/bcachefs/fs-io-direct.c b/fs/bcachefs/fs-io-direct.c index e246b1e05aa2..ee1c0325f313 100644 --- a/fs/bcachefs/fs-io-direct.c +++ b/fs/bcachefs/fs-io-direct.c @@ -500,7 +500,7 @@ static __always_inline long bch2_dio_write_loop(struct dio_write *dio) dio->op.target = dio->op.opts.foreground_target; dio->op.write_point = writepoint_hashed((unsigned long) current); dio->op.nr_replicas = dio->op.opts.data_replicas; - dio->op.subvol = inode->ei_subvol; + dio->op.subvol = inode->ei_inum.subvol; dio->op.pos = POS(inode->v.i_ino, (u64) req->ki_pos >> 9); dio->op.devs_need_flush = &inode->ei_devs_need_flush; diff --git a/fs/bcachefs/fs-io-pagecache.c b/fs/bcachefs/fs-io-pagecache.c index a9cc5cad9cc9..af3a24546aa3 100644 --- a/fs/bcachefs/fs-io-pagecache.c +++ b/fs/bcachefs/fs-io-pagecache.c @@ -182,18 +182,11 @@ static void __bch2_folio_set(struct folio *folio, int bch2_folio_set(struct bch_fs *c, subvol_inum inum, struct folio **fs, unsigned nr_folios) { - struct btree_trans *trans; - struct btree_iter iter; - struct bkey_s_c k; - struct bch_folio *s; u64 offset = folio_sector(fs[0]); - unsigned folio_idx; - u32 snapshot; bool need_set = false; - int ret; - for (folio_idx = 0; folio_idx < nr_folios; folio_idx++) { - s = bch2_folio_create(fs[folio_idx], GFP_KERNEL); + for (unsigned folio_idx = 0; folio_idx < nr_folios; folio_idx++) { + struct bch_folio *s = bch2_folio_create(fs[folio_idx], GFP_KERNEL); if (!s) return -ENOMEM; @@ -203,53 +196,40 @@ int bch2_folio_set(struct bch_fs *c, subvol_inum inum, if (!need_set) return 0; - folio_idx = 0; - trans = bch2_trans_get(c); -retry: - bch2_trans_begin(trans); + unsigned folio_idx = 0; - ret = bch2_subvolume_get_snapshot(trans, inum.subvol, &snapshot); - if (ret) - goto err; + return bch2_trans_run(c, + for_each_btree_key_in_subvolume_upto(trans, iter, BTREE_ID_extents, + POS(inum.inum, offset), + POS(inum.inum, U64_MAX), + inum.subvol, BTREE_ITER_slots, k, ({ + unsigned nr_ptrs = bch2_bkey_nr_ptrs_fully_allocated(k); + unsigned state = bkey_to_sector_state(k); - for_each_btree_key_norestart(trans, iter, BTREE_ID_extents, - SPOS(inum.inum, offset, snapshot), - BTREE_ITER_slots, k, ret) { - unsigned nr_ptrs = bch2_bkey_nr_ptrs_fully_allocated(k); - unsigned state = bkey_to_sector_state(k); + while (folio_idx < nr_folios) { + struct folio *folio = fs[folio_idx]; + u64 folio_start = folio_sector(folio); + u64 folio_end = folio_end_sector(folio); + unsigned folio_offset = max(bkey_start_offset(k.k), folio_start) - + folio_start; + unsigned folio_len = min(k.k->p.offset, folio_end) - + folio_offset - folio_start; - while (folio_idx < nr_folios) { - struct folio *folio = fs[folio_idx]; - u64 folio_start = folio_sector(folio); - u64 folio_end = folio_end_sector(folio); - unsigned folio_offset = max(bkey_start_offset(k.k), folio_start) - - folio_start; - unsigned folio_len = min(k.k->p.offset, folio_end) - - folio_offset - folio_start; + BUG_ON(k.k->p.offset < folio_start); + BUG_ON(bkey_start_offset(k.k) > folio_end); - BUG_ON(k.k->p.offset < folio_start); - BUG_ON(bkey_start_offset(k.k) > folio_end); + if (!bch2_folio(folio)->uptodate) + __bch2_folio_set(folio, folio_offset, folio_len, nr_ptrs, state); - if (!bch2_folio(folio)->uptodate) - __bch2_folio_set(folio, folio_offset, folio_len, nr_ptrs, state); + if (k.k->p.offset < folio_end) + break; + folio_idx++; + } - if (k.k->p.offset < folio_end) + if (folio_idx == nr_folios) break; - folio_idx++; - } - - if (folio_idx == nr_folios) - break; - } - - offset = iter.pos.offset; - bch2_trans_iter_exit(trans, &iter); -err: - if (bch2_err_matches(ret, BCH_ERR_transaction_restart)) - goto retry; - bch2_trans_put(trans); - - return ret; + 0; + }))); } void bch2_bio_page_state_set(struct bio *bio, struct bkey_s_c k) diff --git a/fs/bcachefs/fs-io-pagecache.h b/fs/bcachefs/fs-io-pagecache.h index fd7d692c087e..fad911cf5068 100644 --- a/fs/bcachefs/fs-io-pagecache.h +++ b/fs/bcachefs/fs-io-pagecache.h @@ -99,9 +99,7 @@ static inline void bch2_folio_release(struct folio *folio) static inline struct bch_folio *__bch2_folio(struct folio *folio) { - return folio_has_private(folio) - ? (struct bch_folio *) folio_get_private(folio) - : NULL; + return folio_get_private(folio); } static inline struct bch_folio *bch2_folio(struct folio *folio) diff --git a/fs/bcachefs/fs-io.c b/fs/bcachefs/fs-io.c index 77b85da30fb2..71d0fa387509 100644 --- a/fs/bcachefs/fs-io.c +++ b/fs/bcachefs/fs-io.c @@ -221,30 +221,11 @@ static inline int range_has_data(struct bch_fs *c, u32 subvol, struct bpos start, struct bpos end) { - struct btree_trans *trans = bch2_trans_get(c); - struct btree_iter iter; - struct bkey_s_c k; - int ret = 0; -retry: - bch2_trans_begin(trans); - - ret = bch2_subvolume_get_snapshot(trans, subvol, &start.snapshot); - if (ret) - goto err; - - for_each_btree_key_upto_norestart(trans, iter, BTREE_ID_extents, start, end, 0, k, ret) - if (bkey_extent_is_data(k.k) && !bkey_extent_is_unwritten(k)) { - ret = 1; - break; - } - start = iter.pos; - bch2_trans_iter_exit(trans, &iter); -err: - if (bch2_err_matches(ret, BCH_ERR_transaction_restart)) - goto retry; - - bch2_trans_put(trans); - return ret; + return bch2_trans_run(c, + for_each_btree_key_in_subvolume_upto(trans, iter, BTREE_ID_extents, start, end, + subvol, 0, k, ({ + bkey_extent_is_data(k.k) && !bkey_extent_is_unwritten(k); + }))); } static int __bch2_truncate_folio(struct bch_inode_info *inode, @@ -267,7 +248,7 @@ static int __bch2_truncate_folio(struct bch_inode_info *inode, * XXX: we're doing two index lookups when we end up reading the * folio */ - ret = range_has_data(c, inode->ei_subvol, + ret = range_has_data(c, inode->ei_inum.subvol, POS(inode->v.i_ino, (index << PAGE_SECTORS_SHIFT)), POS(inode->v.i_ino, (index << PAGE_SECTORS_SHIFT) + PAGE_SECTORS)); if (ret <= 0) @@ -618,7 +599,7 @@ static noinline int __bchfs_fallocate(struct bch_inode_info *inode, int mode, bch2_trans_begin(trans); ret = bch2_subvolume_get_snapshot(trans, - inode->ei_subvol, &snapshot); + inode->ei_inum.subvol, &snapshot); if (ret) goto bkey_err; @@ -813,41 +794,23 @@ static int quota_reserve_range(struct bch_inode_info *inode, u64 start, u64 end) { struct bch_fs *c = inode->v.i_sb->s_fs_info; - struct btree_trans *trans = bch2_trans_get(c); - struct btree_iter iter; - struct bkey_s_c k; - u32 snapshot; u64 sectors = end - start; - u64 pos = start; - int ret; -retry: - bch2_trans_begin(trans); - ret = bch2_subvolume_get_snapshot(trans, inode->ei_subvol, &snapshot); - if (ret) - goto err; + int ret = bch2_trans_run(c, + for_each_btree_key_in_subvolume_upto(trans, iter, + BTREE_ID_extents, + POS(inode->v.i_ino, start), + POS(inode->v.i_ino, end - 1), + inode->ei_inum.subvol, 0, k, ({ + if (bkey_extent_is_allocation(k.k)) { + u64 s = min(end, k.k->p.offset) - + max(start, bkey_start_offset(k.k)); + BUG_ON(s > sectors); + sectors -= s; + } - bch2_trans_iter_init(trans, &iter, BTREE_ID_extents, - SPOS(inode->v.i_ino, pos, snapshot), 0); - - while (!(ret = btree_trans_too_many_iters(trans)) && - (k = bch2_btree_iter_peek_upto(&iter, POS(inode->v.i_ino, end - 1))).k && - !(ret = bkey_err(k))) { - if (bkey_extent_is_allocation(k.k)) { - u64 s = min(end, k.k->p.offset) - - max(start, bkey_start_offset(k.k)); - BUG_ON(s > sectors); - sectors -= s; - } - bch2_btree_iter_advance(&iter); - } - pos = iter.pos.offset; - bch2_trans_iter_exit(trans, &iter); -err: - if (bch2_err_matches(ret, BCH_ERR_transaction_restart)) - goto retry; - - bch2_trans_put(trans); + 0; + }))); return ret ?: bch2_quota_reservation_add(c, inode, res, sectors, true); } @@ -942,42 +905,25 @@ static loff_t bch2_seek_data(struct file *file, u64 offset) { struct bch_inode_info *inode = file_bch_inode(file); struct bch_fs *c = inode->v.i_sb->s_fs_info; - struct btree_trans *trans; - struct btree_iter iter; - struct bkey_s_c k; subvol_inum inum = inode_inum(inode); u64 isize, next_data = MAX_LFS_FILESIZE; - u32 snapshot; - int ret; isize = i_size_read(&inode->v); if (offset >= isize) return -ENXIO; - trans = bch2_trans_get(c); -retry: - bch2_trans_begin(trans); - - ret = bch2_subvolume_get_snapshot(trans, inum.subvol, &snapshot); - if (ret) - goto err; - - for_each_btree_key_upto_norestart(trans, iter, BTREE_ID_extents, - SPOS(inode->v.i_ino, offset >> 9, snapshot), - POS(inode->v.i_ino, U64_MAX), - 0, k, ret) { - if (bkey_extent_is_data(k.k)) { - next_data = max(offset, bkey_start_offset(k.k) << 9); - break; - } else if (k.k->p.offset >> 9 > isize) - break; - } - bch2_trans_iter_exit(trans, &iter); -err: - if (bch2_err_matches(ret, BCH_ERR_transaction_restart)) - goto retry; - - bch2_trans_put(trans); + int ret = bch2_trans_run(c, + for_each_btree_key_in_subvolume_upto(trans, iter, BTREE_ID_extents, + POS(inode->v.i_ino, offset >> 9), + POS(inode->v.i_ino, U64_MAX), + inum.subvol, 0, k, ({ + if (bkey_extent_is_data(k.k)) { + next_data = max(offset, bkey_start_offset(k.k) << 9); + break; + } else if (k.k->p.offset >> 9 > isize) + break; + 0; + }))); if (ret) return ret; @@ -995,50 +941,34 @@ static loff_t bch2_seek_hole(struct file *file, u64 offset) { struct bch_inode_info *inode = file_bch_inode(file); struct bch_fs *c = inode->v.i_sb->s_fs_info; - struct btree_trans *trans; - struct btree_iter iter; - struct bkey_s_c k; subvol_inum inum = inode_inum(inode); u64 isize, next_hole = MAX_LFS_FILESIZE; - u32 snapshot; - int ret; isize = i_size_read(&inode->v); if (offset >= isize) return -ENXIO; - trans = bch2_trans_get(c); -retry: - bch2_trans_begin(trans); - - ret = bch2_subvolume_get_snapshot(trans, inum.subvol, &snapshot); - if (ret) - goto err; - - for_each_btree_key_norestart(trans, iter, BTREE_ID_extents, - SPOS(inode->v.i_ino, offset >> 9, snapshot), - BTREE_ITER_slots, k, ret) { - if (k.k->p.inode != inode->v.i_ino) { - next_hole = bch2_seek_pagecache_hole(&inode->v, - offset, MAX_LFS_FILESIZE, 0, false); - break; - } else if (!bkey_extent_is_data(k.k)) { - next_hole = bch2_seek_pagecache_hole(&inode->v, - max(offset, bkey_start_offset(k.k) << 9), - k.k->p.offset << 9, 0, false); - - if (next_hole < k.k->p.offset << 9) + int ret = bch2_trans_run(c, + for_each_btree_key_in_subvolume_upto(trans, iter, BTREE_ID_extents, + POS(inode->v.i_ino, offset >> 9), + POS(inode->v.i_ino, U64_MAX), + inum.subvol, BTREE_ITER_slots, k, ({ + if (k.k->p.inode != inode->v.i_ino) { + next_hole = bch2_seek_pagecache_hole(&inode->v, + offset, MAX_LFS_FILESIZE, 0, false); break; - } else { - offset = max(offset, bkey_start_offset(k.k) << 9); - } - } - bch2_trans_iter_exit(trans, &iter); -err: - if (bch2_err_matches(ret, BCH_ERR_transaction_restart)) - goto retry; + } else if (!bkey_extent_is_data(k.k)) { + next_hole = bch2_seek_pagecache_hole(&inode->v, + max(offset, bkey_start_offset(k.k) << 9), + k.k->p.offset << 9, 0, false); - bch2_trans_put(trans); + if (next_hole < k.k->p.offset << 9) + break; + } else { + offset = max(offset, bkey_start_offset(k.k) << 9); + } + 0; + }))); if (ret) return ret; diff --git a/fs/bcachefs/fs-ioctl.c b/fs/bcachefs/fs-ioctl.c index 99c7fe987c74..405cf08bda34 100644 --- a/fs/bcachefs/fs-ioctl.c +++ b/fs/bcachefs/fs-ioctl.c @@ -100,7 +100,7 @@ static int bch2_ioc_setflags(struct bch_fs *c, } mutex_lock(&inode->ei_update_lock); - ret = bch2_subvol_is_ro(c, inode->ei_subvol) ?: + ret = bch2_subvol_is_ro(c, inode->ei_inum.subvol) ?: bch2_write_inode(c, inode, bch2_inode_flags_set, &s, ATTR_CTIME); mutex_unlock(&inode->ei_update_lock); @@ -184,7 +184,7 @@ static int bch2_ioc_fssetxattr(struct bch_fs *c, } mutex_lock(&inode->ei_update_lock); - ret = bch2_subvol_is_ro(c, inode->ei_subvol) ?: + ret = bch2_subvol_is_ro(c, inode->ei_inum.subvol) ?: bch2_set_projid(c, inode, fa.fsx_projid) ?: bch2_write_inode(c, inode, fssetxattr_inode_update_fn, &s, ATTR_CTIME); diff --git a/fs/bcachefs/fs.c b/fs/bcachefs/fs.c index 011817afc3ad..4a1bb07a2574 100644 --- a/fs/bcachefs/fs.c +++ b/fs/bcachefs/fs.c @@ -108,7 +108,7 @@ retry: goto retry; bch2_fs_fatal_err_on(bch2_err_matches(ret, ENOENT), c, - "%s: inode %u:%llu not found when updating", + "%s: inode %llu:%llu not found when updating", bch2_err_str(ret), inode_inum(inode).subvol, inode_inum(inode).inum); @@ -152,50 +152,106 @@ int bch2_fs_quota_transfer(struct bch_fs *c, return ret; } -static int bch2_iget5_test(struct inode *vinode, void *p) +static bool subvol_inum_eq(subvol_inum a, subvol_inum b) { - struct bch_inode_info *inode = to_bch_ei(vinode); - subvol_inum *inum = p; - - return inode->ei_subvol == inum->subvol && - inode->ei_inode.bi_inum == inum->inum; + return a.subvol == b.subvol && a.inum == b.inum; } -static int bch2_iget5_set(struct inode *vinode, void *p) +static int bch2_vfs_inode_cmp_fn(struct rhashtable_compare_arg *arg, + const void *obj) { - struct bch_inode_info *inode = to_bch_ei(vinode); - subvol_inum *inum = p; + const struct bch_inode_info *inode = obj; + const subvol_inum *v = arg->key; - inode->v.i_ino = inum->inum; - inode->ei_subvol = inum->subvol; - inode->ei_inode.bi_inum = inum->inum; - return 0; + return !subvol_inum_eq(inode->ei_inum, *v); } -static unsigned bch2_inode_hash(subvol_inum inum) +static const struct rhashtable_params bch2_vfs_inodes_params = { + .head_offset = offsetof(struct bch_inode_info, hash), + .key_offset = offsetof(struct bch_inode_info, ei_inum), + .key_len = sizeof(subvol_inum), + .obj_cmpfn = bch2_vfs_inode_cmp_fn, + .automatic_shrinking = true, +}; + +static void __wait_on_freeing_inode(struct inode *inode) { - return jhash_3words(inum.subvol, inum.inum >> 32, inum.inum, JHASH_INITVAL); + wait_queue_head_t *wq; + DEFINE_WAIT_BIT(wait, &inode->i_state, __I_NEW); + wq = bit_waitqueue(&inode->i_state, __I_NEW); + prepare_to_wait(wq, &wait.wq_entry, TASK_UNINTERRUPTIBLE); + spin_unlock(&inode->i_lock); + schedule(); + finish_wait(wq, &wait.wq_entry); } struct bch_inode_info *__bch2_inode_hash_find(struct bch_fs *c, subvol_inum inum) { - return to_bch_ei(ilookup5_nowait(c->vfs_sb, - bch2_inode_hash(inum), - bch2_iget5_test, - &inum)); + return rhashtable_lookup_fast(&c->vfs_inodes_table, &inum, bch2_vfs_inodes_params); } -static struct bch_inode_info *bch2_inode_insert(struct bch_fs *c, struct bch_inode_info *inode) +static struct bch_inode_info *bch2_inode_hash_find(struct bch_fs *c, struct btree_trans *trans, + subvol_inum inum) { - subvol_inum inum = inode_inum(inode); - struct bch_inode_info *old = to_bch_ei(inode_insert5(&inode->v, - bch2_inode_hash(inum), - bch2_iget5_test, - bch2_iget5_set, - &inum)); - BUG_ON(!old); + struct bch_inode_info *inode; +repeat: + inode = __bch2_inode_hash_find(c, inum); + if (inode) { + spin_lock(&inode->v.i_lock); + if (!test_bit(EI_INODE_HASHED, &inode->ei_flags)) { + spin_unlock(&inode->v.i_lock); + return NULL; + } + if ((inode->v.i_state & (I_FREEING|I_WILL_FREE))) { + if (!trans) { + __wait_on_freeing_inode(&inode->v); + } else { + bch2_trans_unlock(trans); + __wait_on_freeing_inode(&inode->v); + int ret = bch2_trans_relock(trans); + if (ret) + return ERR_PTR(ret); + } + goto repeat; + } + __iget(&inode->v); + spin_unlock(&inode->v.i_lock); + } + + return inode; +} + +static void bch2_inode_hash_remove(struct bch_fs *c, struct bch_inode_info *inode) +{ + spin_lock(&inode->v.i_lock); + bool remove = test_and_clear_bit(EI_INODE_HASHED, &inode->ei_flags); + spin_unlock(&inode->v.i_lock); + + if (remove) { + int ret = rhashtable_remove_fast(&c->vfs_inodes_table, + &inode->hash, bch2_vfs_inodes_params); + BUG_ON(ret); + inode->v.i_hash.pprev = NULL; + } +} + +static struct bch_inode_info *bch2_inode_hash_insert(struct bch_fs *c, + struct btree_trans *trans, + struct bch_inode_info *inode) +{ + struct bch_inode_info *old = inode; + + set_bit(EI_INODE_HASHED, &inode->ei_flags); +retry: + if (unlikely(rhashtable_lookup_insert_fast(&c->vfs_inodes_table, + &inode->hash, + bch2_vfs_inodes_params))) { + old = bch2_inode_hash_find(c, trans, inode->ei_inum); + if (!old) + goto retry; + + clear_bit(EI_INODE_HASHED, &inode->ei_flags); - if (unlikely(old != inode)) { /* * bcachefs doesn't use I_NEW; we have no use for it since we * only insert fully created inodes in the inode hash table. But @@ -209,21 +265,17 @@ static struct bch_inode_info *bch2_inode_insert(struct bch_fs *c, struct bch_ino */ set_nlink(&inode->v, 1); discard_new_inode(&inode->v); - inode = old; + return old; } else { + inode_fake_hash(&inode->v); + + inode_sb_list_add(&inode->v); + mutex_lock(&c->vfs_inodes_lock); list_add(&inode->ei_vfs_inode_list, &c->vfs_inodes_list); mutex_unlock(&c->vfs_inodes_lock); - /* - * Again, I_NEW makes no sense for bcachefs. This is only needed - * for clearing I_NEW, but since the inode was already fully - * created and initialized we didn't actually want - * inode_insert5() to set it for us. - */ - unlock_new_inode(&inode->v); + return inode; } - - return inode; } #define memalloc_flags_do(_flags, _do) \ @@ -241,7 +293,8 @@ static struct inode *bch2_alloc_inode(struct super_block *sb) static struct bch_inode_info *__bch2_new_inode(struct bch_fs *c) { - struct bch_inode_info *inode = kmem_cache_alloc(bch2_inode_cache, GFP_NOFS); + struct bch_inode_info *inode = alloc_inode_sb(c->vfs_sb, + bch2_inode_cache, GFP_NOFS); if (!inode) return NULL; @@ -283,13 +336,24 @@ static struct bch_inode_info *bch2_new_inode(struct btree_trans *trans) return inode; } +static struct bch_inode_info *bch2_inode_hash_init_insert(struct btree_trans *trans, + subvol_inum inum, + struct bch_inode_unpacked *bi, + struct bch_subvolume *subvol) +{ + struct bch_inode_info *inode = bch2_new_inode(trans); + if (IS_ERR(inode)) + return inode; + + bch2_vfs_inode_init(trans, inum, inode, bi, subvol); + + return bch2_inode_hash_insert(trans->c, trans, inode); + +} + struct inode *bch2_vfs_inode_get(struct bch_fs *c, subvol_inum inum) { - struct bch_inode_info *inode = - to_bch_ei(ilookup5_nowait(c->vfs_sb, - bch2_inode_hash(inum), - bch2_iget5_test, - &inum)); + struct bch_inode_info *inode = bch2_inode_hash_find(c, NULL, inum); if (inode) return &inode->v; @@ -300,11 +364,7 @@ struct inode *bch2_vfs_inode_get(struct bch_fs *c, subvol_inum inum) int ret = lockrestart_do(trans, bch2_subvolume_get(trans, inum.subvol, true, 0, &subvol) ?: bch2_inode_find_by_inum_trans(trans, inum, &inode_u)) ?: - PTR_ERR_OR_ZERO(inode = bch2_new_inode(trans)); - if (!ret) { - bch2_vfs_inode_init(trans, inum, inode, &inode_u, &subvol); - inode = bch2_inode_insert(c, inode); - } + PTR_ERR_OR_ZERO(inode = bch2_inode_hash_init_insert(trans, inum, &inode_u, &subvol)); bch2_trans_put(trans); return ret ? ERR_PTR(ret) : &inode->v; @@ -325,6 +385,8 @@ __bch2_create(struct mnt_idmap *idmap, subvol_inum inum; struct bch_subvolume subvol; u64 journal_seq = 0; + kuid_t kuid; + kgid_t kgid; int ret; /* @@ -351,13 +413,15 @@ __bch2_create(struct mnt_idmap *idmap, retry: bch2_trans_begin(trans); - ret = bch2_subvol_is_ro_trans(trans, dir->ei_subvol) ?: + kuid = mapped_fsuid(idmap, i_user_ns(&dir->v)); + kgid = mapped_fsgid(idmap, i_user_ns(&dir->v)); + ret = bch2_subvol_is_ro_trans(trans, dir->ei_inum.subvol) ?: bch2_create_trans(trans, inode_inum(dir), &dir_u, &inode_u, !(flags & BCH_CREATE_TMPFILE) ? &dentry->d_name : NULL, - from_kuid(i_user_ns(&dir->v), current_fsuid()), - from_kgid(i_user_ns(&dir->v), current_fsgid()), + from_kuid(i_user_ns(&dir->v), kuid), + from_kgid(i_user_ns(&dir->v), kgid), mode, rdev, default_acl, acl, snapshot_src, flags) ?: bch2_quota_acct(c, bch_qid(&inode_u), Q_INO, 1, @@ -365,7 +429,7 @@ retry: if (unlikely(ret)) goto err_before_quota; - inum.subvol = inode_u.bi_subvol ?: dir->ei_subvol; + inum.subvol = inode_u.bi_subvol ?: dir->ei_inum.subvol; inum.inum = inode_u.bi_inum; ret = bch2_subvolume_get(trans, inum.subvol, true, @@ -395,8 +459,16 @@ err_before_quota: * we must insert the new inode into the inode cache before calling * bch2_trans_exit() and dropping locks, else we could race with another * thread pulling the inode in and modifying it: + * + * also, calling bch2_inode_hash_insert() without passing in the + * transaction object is sketchy - if we could ever end up in + * __wait_on_freeing_inode(), we'd risk deadlock. + * + * But that shouldn't be possible, since we still have the inode locked + * that we just created, and we _really_ can't take a transaction + * restart here. */ - inode = bch2_inode_insert(c, inode); + inode = bch2_inode_hash_insert(c, NULL, inode); bch2_trans_put(trans); err: posix_acl_release(default_acl); @@ -436,11 +508,7 @@ static struct bch_inode_info *bch2_lookup_trans(struct btree_trans *trans, if (ret) goto err; - struct bch_inode_info *inode = - to_bch_ei(ilookup5_nowait(c->vfs_sb, - bch2_inode_hash(inum), - bch2_iget5_test, - &inum)); + struct bch_inode_info *inode = bch2_inode_hash_find(c, trans, inum); if (inode) goto out; @@ -448,7 +516,7 @@ static struct bch_inode_info *bch2_lookup_trans(struct btree_trans *trans, struct bch_inode_unpacked inode_u; ret = bch2_subvolume_get(trans, inum.subvol, true, 0, &subvol) ?: bch2_inode_find_by_inum_nowarn_trans(trans, inum, &inode_u) ?: - PTR_ERR_OR_ZERO(inode = bch2_new_inode(trans)); + PTR_ERR_OR_ZERO(inode = bch2_inode_hash_init_insert(trans, inum, &inode_u, &subvol)); bch2_fs_inconsistent_on(bch2_err_matches(ret, ENOENT), c, "dirent to missing inode:\n %s", @@ -468,9 +536,6 @@ static struct bch_inode_info *bch2_lookup_trans(struct btree_trans *trans, ret = -ENOENT; goto err; } - - bch2_vfs_inode_init(trans, inum, inode, &inode_u, &subvol); - inode = bch2_inode_insert(c, inode); out: bch2_trans_iter_exit(trans, &dirent_iter); printbuf_exit(&buf); @@ -557,8 +622,8 @@ static int bch2_link(struct dentry *old_dentry, struct inode *vdir, lockdep_assert_held(&inode->v.i_rwsem); - ret = bch2_subvol_is_ro(c, dir->ei_subvol) ?: - bch2_subvol_is_ro(c, inode->ei_subvol) ?: + ret = bch2_subvol_is_ro(c, dir->ei_inum.subvol) ?: + bch2_subvol_is_ro(c, inode->ei_inum.subvol) ?: __bch2_link(c, inode, dir, dentry); if (unlikely(ret)) return bch2_err_class(ret); @@ -614,7 +679,7 @@ static int bch2_unlink(struct inode *vdir, struct dentry *dentry) struct bch_inode_info *dir= to_bch_ei(vdir); struct bch_fs *c = dir->v.i_sb->s_fs_info; - int ret = bch2_subvol_is_ro(c, dir->ei_subvol) ?: + int ret = bch2_subvol_is_ro(c, dir->ei_inum.subvol) ?: __bch2_unlink(vdir, dentry, false); return bch2_err_class(ret); } @@ -671,15 +736,16 @@ static int bch2_rename2(struct mnt_idmap *idmap, struct bch_inode_info *src_inode = to_bch_ei(src_dentry->d_inode); struct bch_inode_info *dst_inode = to_bch_ei(dst_dentry->d_inode); struct bch_inode_unpacked dst_dir_u, src_dir_u; - struct bch_inode_unpacked src_inode_u, dst_inode_u; + struct bch_inode_unpacked src_inode_u, dst_inode_u, *whiteout_inode_u; struct btree_trans *trans; enum bch_rename_mode mode = flags & RENAME_EXCHANGE ? BCH_RENAME_EXCHANGE : dst_dentry->d_inode ? BCH_RENAME_OVERWRITE : BCH_RENAME; + bool whiteout = !!(flags & RENAME_WHITEOUT); int ret; - if (flags & ~(RENAME_NOREPLACE|RENAME_EXCHANGE)) + if (flags & ~(RENAME_NOREPLACE|RENAME_EXCHANGE|RENAME_WHITEOUT)) return -EINVAL; if (mode == BCH_RENAME_OVERWRITE) { @@ -697,8 +763,8 @@ static int bch2_rename2(struct mnt_idmap *idmap, trans = bch2_trans_get(c); - ret = bch2_subvol_is_ro_trans(trans, src_dir->ei_subvol) ?: - bch2_subvol_is_ro_trans(trans, dst_dir->ei_subvol); + ret = bch2_subvol_is_ro_trans(trans, src_dir->ei_inum.subvol) ?: + bch2_subvol_is_ro_trans(trans, dst_dir->ei_inum.subvol); if (ret) goto err; @@ -720,18 +786,48 @@ static int bch2_rename2(struct mnt_idmap *idmap, if (ret) goto err; } +retry: + bch2_trans_begin(trans); - ret = commit_do(trans, NULL, NULL, 0, - bch2_rename_trans(trans, - inode_inum(src_dir), &src_dir_u, - inode_inum(dst_dir), &dst_dir_u, - &src_inode_u, - &dst_inode_u, - &src_dentry->d_name, - &dst_dentry->d_name, - mode)); + ret = bch2_rename_trans(trans, + inode_inum(src_dir), &src_dir_u, + inode_inum(dst_dir), &dst_dir_u, + &src_inode_u, + &dst_inode_u, + &src_dentry->d_name, + &dst_dentry->d_name, + mode); if (unlikely(ret)) + goto err_tx_restart; + + if (whiteout) { + whiteout_inode_u = bch2_trans_kmalloc_nomemzero(trans, sizeof(*whiteout_inode_u)); + ret = PTR_ERR_OR_ZERO(whiteout_inode_u); + if (unlikely(ret)) + goto err_tx_restart; + bch2_inode_init_early(c, whiteout_inode_u); + + ret = bch2_create_trans(trans, + inode_inum(src_dir), &src_dir_u, + whiteout_inode_u, + &src_dentry->d_name, + from_kuid(i_user_ns(&src_dir->v), current_fsuid()), + from_kgid(i_user_ns(&src_dir->v), current_fsgid()), + S_IFCHR|WHITEOUT_MODE, 0, + NULL, NULL, (subvol_inum) { 0 }, 0) ?: + bch2_quota_acct(c, bch_qid(whiteout_inode_u), Q_INO, 1, + KEY_TYPE_QUOTA_PREALLOC); + if (unlikely(ret)) + goto err_tx_restart; + } + + ret = bch2_trans_commit(trans, NULL, NULL, 0); + if (unlikely(ret)) { +err_tx_restart: + if (bch2_err_matches(ret, BCH_ERR_transaction_restart)) + goto retry; goto err; + } BUG_ON(src_inode->v.i_ino != src_inode_u.bi_inum); BUG_ON(dst_inode && @@ -779,11 +875,17 @@ static void bch2_setattr_copy(struct mnt_idmap *idmap, { struct bch_fs *c = inode->v.i_sb->s_fs_info; unsigned int ia_valid = attr->ia_valid; + kuid_t kuid; + kgid_t kgid; - if (ia_valid & ATTR_UID) - bi->bi_uid = from_kuid(i_user_ns(&inode->v), attr->ia_uid); - if (ia_valid & ATTR_GID) - bi->bi_gid = from_kgid(i_user_ns(&inode->v), attr->ia_gid); + if (ia_valid & ATTR_UID) { + kuid = from_vfsuid(idmap, i_user_ns(&inode->v), attr->ia_vfsuid); + bi->bi_uid = from_kuid(i_user_ns(&inode->v), kuid); + } + if (ia_valid & ATTR_GID) { + kgid = from_vfsgid(idmap, i_user_ns(&inode->v), attr->ia_vfsgid); + bi->bi_gid = from_kgid(i_user_ns(&inode->v), kgid); + } if (ia_valid & ATTR_SIZE) bi->bi_size = attr->ia_size; @@ -798,11 +900,11 @@ static void bch2_setattr_copy(struct mnt_idmap *idmap, if (ia_valid & ATTR_MODE) { umode_t mode = attr->ia_mode; kgid_t gid = ia_valid & ATTR_GID - ? attr->ia_gid + ? kgid : inode->v.i_gid; - if (!in_group_p(gid) && - !capable_wrt_inode_uidgid(idmap, &inode->v, CAP_FSETID)) + if (!in_group_or_capable(idmap, &inode->v, + make_vfsgid(idmap, i_user_ns(&inode->v), gid))) mode &= ~S_ISGID; bi->bi_mode = mode; } @@ -818,17 +920,23 @@ int bch2_setattr_nonsize(struct mnt_idmap *idmap, struct btree_iter inode_iter = { NULL }; struct bch_inode_unpacked inode_u; struct posix_acl *acl = NULL; + kuid_t kuid; + kgid_t kgid; int ret; mutex_lock(&inode->ei_update_lock); qid = inode->ei_qid; - if (attr->ia_valid & ATTR_UID) - qid.q[QTYP_USR] = from_kuid(i_user_ns(&inode->v), attr->ia_uid); + if (attr->ia_valid & ATTR_UID) { + kuid = from_vfsuid(idmap, i_user_ns(&inode->v), attr->ia_vfsuid); + qid.q[QTYP_USR] = from_kuid(i_user_ns(&inode->v), kuid); + } - if (attr->ia_valid & ATTR_GID) - qid.q[QTYP_GRP] = from_kgid(i_user_ns(&inode->v), attr->ia_gid); + if (attr->ia_valid & ATTR_GID) { + kgid = from_vfsgid(idmap, i_user_ns(&inode->v), attr->ia_vfsgid); + qid.q[QTYP_GRP] = from_kgid(i_user_ns(&inode->v), kgid); + } ret = bch2_fs_quota_transfer(c, inode, qid, ~0, KEY_TYPE_QUOTA_PREALLOC); @@ -884,13 +992,15 @@ static int bch2_getattr(struct mnt_idmap *idmap, { struct bch_inode_info *inode = to_bch_ei(d_inode(path->dentry)); struct bch_fs *c = inode->v.i_sb->s_fs_info; + vfsuid_t vfsuid = i_uid_into_vfsuid(idmap, &inode->v); + vfsgid_t vfsgid = i_gid_into_vfsgid(idmap, &inode->v); stat->dev = inode->v.i_sb->s_dev; stat->ino = inode->v.i_ino; stat->mode = inode->v.i_mode; stat->nlink = inode->v.i_nlink; - stat->uid = inode->v.i_uid; - stat->gid = inode->v.i_gid; + stat->uid = vfsuid_into_kuid(vfsuid); + stat->gid = vfsgid_into_kgid(vfsgid); stat->rdev = inode->v.i_rdev; stat->size = i_size_read(&inode->v); stat->atime = inode_get_atime(&inode->v); @@ -899,7 +1009,7 @@ static int bch2_getattr(struct mnt_idmap *idmap, stat->blksize = block_bytes(c); stat->blocks = inode->v.i_blocks; - stat->subvol = inode->ei_subvol; + stat->subvol = inode->ei_inum.subvol; stat->result_mask |= STATX_SUBVOL; if ((request_mask & STATX_DIOALIGN) && S_ISREG(inode->v.i_mode)) { @@ -941,7 +1051,7 @@ static int bch2_setattr(struct mnt_idmap *idmap, lockdep_assert_held(&inode->v.i_rwsem); - ret = bch2_subvol_is_ro(c, inode->ei_subvol) ?: + ret = bch2_subvol_is_ro(c, inode->ei_inum.subvol) ?: setattr_prepare(idmap, dentry, iattr); if (ret) return ret; @@ -1034,7 +1144,6 @@ static int bch2_fiemap(struct inode *vinode, struct fiemap_extent_info *info, struct bkey_buf cur, prev; unsigned offset_into_extent, sectors; bool have_extent = false; - u32 snapshot; int ret = 0; ret = fiemap_prep(&ei->v, info, start, &len, FIEMAP_FLAG_SYNC); @@ -1050,21 +1159,30 @@ static int bch2_fiemap(struct inode *vinode, struct fiemap_extent_info *info, bch2_bkey_buf_init(&cur); bch2_bkey_buf_init(&prev); trans = bch2_trans_get(c); -retry: - bch2_trans_begin(trans); - - ret = bch2_subvolume_get_snapshot(trans, ei->ei_subvol, &snapshot); - if (ret) - goto err; bch2_trans_iter_init(trans, &iter, BTREE_ID_extents, - SPOS(ei->v.i_ino, start, snapshot), 0); + POS(ei->v.i_ino, start), 0); - while (!(ret = btree_trans_too_many_iters(trans)) && - (k = bch2_btree_iter_peek_upto(&iter, end)).k && - !(ret = bkey_err(k))) { + while (true) { enum btree_id data_btree = BTREE_ID_extents; + bch2_trans_begin(trans); + + u32 snapshot; + ret = bch2_subvolume_get_snapshot(trans, ei->ei_inum.subvol, &snapshot); + if (ret) + goto err; + + bch2_btree_iter_set_snapshot(&iter, snapshot); + + k = bch2_btree_iter_peek_upto(&iter, end); + ret = bkey_err(k); + if (ret) + goto err; + + if (!k.k) + break; + if (!bkey_extent_is_data(k.k) && k.k->type != KEY_TYPE_reservation) { bch2_btree_iter_advance(&iter); @@ -1108,16 +1226,12 @@ retry: bch2_btree_iter_set_pos(&iter, POS(iter.pos.inode, iter.pos.offset + sectors)); - - ret = bch2_trans_relock(trans); - if (ret) +err: + if (ret && + !bch2_err_matches(ret, BCH_ERR_transaction_restart)) break; } - start = iter.pos.offset; bch2_trans_iter_exit(trans, &iter); -err: - if (bch2_err_matches(ret, BCH_ERR_transaction_restart)) - goto retry; if (!ret && have_extent) { bch2_trans_unlock(trans); @@ -1173,7 +1287,7 @@ static int bch2_open(struct inode *vinode, struct file *file) struct bch_inode_info *inode = to_bch_ei(vinode); struct bch_fs *c = inode->v.i_sb->s_fs_info; - int ret = bch2_subvol_is_ro(c, inode->ei_subvol); + int ret = bch2_subvol_is_ro(c, inode->ei_inum.subvol); if (ret) return ret; } @@ -1305,8 +1419,8 @@ static int bcachefs_fid_valid(int fh_len, int fh_type) static struct bcachefs_fid bch2_inode_to_fid(struct bch_inode_info *inode) { return (struct bcachefs_fid) { - .inum = inode->ei_inode.bi_inum, - .subvol = inode->ei_subvol, + .inum = inode->ei_inum.inum, + .subvol = inode->ei_inum.subvol, .gen = inode->ei_inode.bi_generation, }; } @@ -1391,7 +1505,7 @@ static struct dentry *bch2_get_parent(struct dentry *child) struct bch_fs *c = inode->v.i_sb->s_fs_info; subvol_inum parent_inum = { .subvol = inode->ei_inode.bi_parent_subvol ?: - inode->ei_subvol, + inode->ei_inum.subvol, .inum = inode->ei_inode.bi_dir, }; @@ -1427,7 +1541,7 @@ static int bch2_get_name(struct dentry *parent, char *name, struct dentry *child retry: bch2_trans_begin(trans); - ret = bch2_subvolume_get_snapshot(trans, dir->ei_subvol, &snapshot); + ret = bch2_subvolume_get_snapshot(trans, dir->ei_inum.subvol, &snapshot); if (ret) goto err; @@ -1458,8 +1572,7 @@ retry: if (ret) goto err; - if (target.subvol == inode->ei_subvol && - target.inum == inode->ei_inode.bi_inum) + if (subvol_inum_eq(target, inode->ei_inum)) goto found; } else { /* @@ -1480,8 +1593,7 @@ retry: if (ret) continue; - if (target.subvol == inode->ei_subvol && - target.inum == inode->ei_inode.bi_inum) + if (subvol_inum_eq(target, inode->ei_inum)) goto found; } } @@ -1513,12 +1625,15 @@ static const struct export_operations bch_export_ops = { .get_name = bch2_get_name, }; -static void bch2_vfs_inode_init(struct btree_trans *trans, subvol_inum inum, +static void bch2_vfs_inode_init(struct btree_trans *trans, + subvol_inum inum, struct bch_inode_info *inode, struct bch_inode_unpacked *bi, struct bch_subvolume *subvol) { - bch2_iget5_set(&inode->v, &inum); + inode->v.i_ino = inum.inum; + inode->ei_inum = inum; + inode->ei_inode.bi_inum = inum.inum; bch2_inode_update_after_write(trans, inode, bi, ~0); inode->v.i_blocks = bi->bi_sectors; @@ -1530,7 +1645,6 @@ static void bch2_vfs_inode_init(struct btree_trans *trans, subvol_inum inum, inode->ei_flags = 0; inode->ei_quota_reserved = 0; inode->ei_qid = bch_qid(bi); - inode->ei_subvol = inum.subvol; if (BCH_SUBVOLUME_SNAP(subvol)) set_bit(EI_INODE_SNAPSHOT, &inode->ei_flags); @@ -1597,6 +1711,17 @@ static void bch2_evict_inode(struct inode *vinode) { struct bch_fs *c = vinode->i_sb->s_fs_info; struct bch_inode_info *inode = to_bch_ei(vinode); + bool delete = !inode->v.i_nlink && !is_bad_inode(&inode->v); + + /* + * evict() has waited for outstanding writeback, we'll do no more IO + * through this inode: it's safe to remove from VFS inode hashtable here + * + * Do that now so that other threads aren't blocked from pulling it back + * in, there's no reason for them to be: + */ + if (!delete) + bch2_inode_hash_remove(c, inode); truncate_inode_pages_final(&inode->v.i_data); @@ -1604,12 +1729,18 @@ static void bch2_evict_inode(struct inode *vinode) BUG_ON(!is_bad_inode(&inode->v) && inode->ei_quota_reserved); - if (!inode->v.i_nlink && !is_bad_inode(&inode->v)) { + if (delete) { bch2_quota_acct(c, inode->ei_qid, Q_SPC, -((s64) inode->v.i_blocks), KEY_TYPE_QUOTA_WARN); bch2_quota_acct(c, inode->ei_qid, Q_INO, -1, KEY_TYPE_QUOTA_WARN); bch2_inode_rm(c, inode_inum(inode)); + + /* + * If we are deleting, we need it present in the vfs hash table + * so that fsck can check if unlinked inodes are still open: + */ + bch2_inode_hash_remove(c, inode); } mutex_lock(&c->vfs_inodes_lock); @@ -1639,7 +1770,7 @@ again: mutex_lock(&c->vfs_inodes_lock); list_for_each_entry(inode, &c->vfs_inodes_list, ei_vfs_inode_list) { - if (!snapshot_list_has_id(s, inode->ei_subvol)) + if (!snapshot_list_has_id(s, inode->ei_inum.subvol)) continue; if (!(inode->v.i_state & I_DONTCACHE) && @@ -1801,30 +1932,14 @@ static int bch2_show_devname(struct seq_file *seq, struct dentry *root) static int bch2_show_options(struct seq_file *seq, struct dentry *root) { struct bch_fs *c = root->d_sb->s_fs_info; - enum bch_opt_id i; struct printbuf buf = PRINTBUF; - int ret = 0; - for (i = 0; i < bch2_opts_nr; i++) { - const struct bch_option *opt = &bch2_opt_table[i]; - u64 v = bch2_opt_get_by_id(&c->opts, i); + bch2_opts_to_text(&buf, c->opts, c, c->disk_sb.sb, + OPT_MOUNT, OPT_HIDDEN, OPT_SHOW_MOUNT_STYLE); + printbuf_nul_terminate(&buf); + seq_puts(seq, buf.buf); - if ((opt->flags & OPT_HIDDEN) || - !(opt->flags & OPT_MOUNT)) - continue; - - if (v == bch2_opt_get_by_id(&bch2_opts_default, i)) - continue; - - printbuf_reset(&buf); - bch2_opt_to_text(&buf, c, c->disk_sb.sb, opt, v, - OPT_SHOW_MOUNT_STYLE); - seq_putc(seq, ','); - seq_puts(seq, buf.buf); - } - - if (buf.allocation_failure) - ret = -ENOMEM; + int ret = buf.allocation_failure ? -ENOMEM : 0; printbuf_exit(&buf); return ret; } @@ -2129,12 +2244,23 @@ static int bch2_init_fs_context(struct fs_context *fc) return 0; } +void bch2_fs_vfs_exit(struct bch_fs *c) +{ + if (c->vfs_inodes_table.tbl) + rhashtable_destroy(&c->vfs_inodes_table); +} + +int bch2_fs_vfs_init(struct bch_fs *c) +{ + return rhashtable_init(&c->vfs_inodes_table, &bch2_vfs_inodes_params); +} + static struct file_system_type bcache_fs_type = { .owner = THIS_MODULE, .name = "bcachefs", .init_fs_context = bch2_init_fs_context, .kill_sb = bch2_kill_sb, - .fs_flags = FS_REQUIRES_DEV, + .fs_flags = FS_REQUIRES_DEV | FS_ALLOW_IDMAP, }; MODULE_ALIAS_FS("bcachefs"); @@ -2149,7 +2275,8 @@ int __init bch2_vfs_init(void) { int ret = -ENOMEM; - bch2_inode_cache = KMEM_CACHE(bch_inode_info, SLAB_RECLAIM_ACCOUNT); + bch2_inode_cache = KMEM_CACHE(bch_inode_info, SLAB_RECLAIM_ACCOUNT | + SLAB_ACCOUNT); if (!bch2_inode_cache) goto err; diff --git a/fs/bcachefs/fs.h b/fs/bcachefs/fs.h index 990ec43e0365..da74ecc236e7 100644 --- a/fs/bcachefs/fs.h +++ b/fs/bcachefs/fs.h @@ -13,6 +13,9 @@ struct bch_inode_info { struct inode v; + struct rhash_head hash; + subvol_inum ei_inum; + struct list_head ei_vfs_inode_list; unsigned long ei_flags; @@ -24,8 +27,6 @@ struct bch_inode_info { struct mutex ei_quota_lock; struct bch_qid ei_qid; - u32 ei_subvol; - /* * When we've been doing nocow writes we'll need to issue flushes to the * underlying block devices @@ -50,10 +51,7 @@ struct bch_inode_info { static inline subvol_inum inode_inum(struct bch_inode_info *inode) { - return (subvol_inum) { - .subvol = inode->ei_subvol, - .inum = inode->ei_inode.bi_inum, - }; + return inode->ei_inum; } struct bch_inode_info *__bch2_inode_hash_find(struct bch_fs *, subvol_inum); @@ -69,6 +67,7 @@ struct bch_inode_info *__bch2_inode_hash_find(struct bch_fs *, subvol_inum); * those: */ #define EI_INODE_SNAPSHOT 1 +#define EI_INODE_HASHED 2 #define to_bch_ei(_inode) \ container_of_or_null(_inode, struct bch_inode_info, v) @@ -189,6 +188,9 @@ int __bch2_unlink(struct inode *, struct dentry *, bool); void bch2_evict_subvolume_inodes(struct bch_fs *, snapshot_id_list *); +void bch2_fs_vfs_exit(struct bch_fs *); +int bch2_fs_vfs_init(struct bch_fs *); + void bch2_vfs_exit(void); int bch2_vfs_init(void); @@ -203,6 +205,10 @@ static inline struct bch_inode_info *__bch2_inode_hash_find(struct bch_fs *c, su static inline void bch2_evict_subvolume_inodes(struct bch_fs *c, snapshot_id_list *s) {} + +static inline void bch2_fs_vfs_exit(struct bch_fs *c) {} +static inline int bch2_fs_vfs_init(struct bch_fs *c) { return 0; } + static inline void bch2_vfs_exit(void) {} static inline int bch2_vfs_init(void) { return 0; } diff --git a/fs/bcachefs/inode.c b/fs/bcachefs/inode.c index 2be6be33afa3..6ac0ff7e074b 100644 --- a/fs/bcachefs/inode.c +++ b/fs/bcachefs/inode.c @@ -365,7 +365,7 @@ int bch2_inode_peek(struct btree_trans *trans, subvol_inum inum, unsigned flags) { int ret = bch2_inode_peek_nowarn(trans, iter, inode, inum, flags); - bch_err_msg(trans->c, ret, "looking up inum %u:%llu:", inum.subvol, inum.inum); + bch_err_msg(trans->c, ret, "looking up inum %llu:%llu:", inum.subvol, inum.inum); return ret; } diff --git a/fs/bcachefs/io_read.c b/fs/bcachefs/io_read.c index 7ee3b75480df..b2f50e74bb76 100644 --- a/fs/bcachefs/io_read.c +++ b/fs/bcachefs/io_read.c @@ -286,7 +286,7 @@ static struct promote_op *promote_alloc(struct btree_trans *trans, */ bool promote_full = (failed || *read_full || - READ_ONCE(c->promote_whole_extents)); + READ_ONCE(c->opts.promote_whole_extents)); /* data might have to be decompressed in the write path: */ unsigned sectors = promote_full ? max(pick->crc.compressed_size, pick->crc.live_size) @@ -777,7 +777,7 @@ int __bch2_read_indirect_extent(struct btree_trans *trans, orig_k->k->k.size, reflink_offset); bch2_inconsistent_error(trans->c); - ret = -EIO; + ret = -BCH_ERR_missing_indirect_extent; goto err; } @@ -869,9 +869,15 @@ retry_pick: goto hole; if (pick_ret < 0) { + struct printbuf buf = PRINTBUF; + bch2_bkey_val_to_text(&buf, c, k); + bch_err_inum_offset_ratelimited(c, read_pos.inode, read_pos.offset << 9, - "no device to read from"); + "no device to read from: %s\n %s", + bch2_err_str(pick_ret), + buf.buf); + printbuf_exit(&buf); goto err; } @@ -1086,7 +1092,7 @@ get_bio: trans->notrace_relock_fail = true; } else { /* Attempting reconstruct read: */ - if (bch2_ec_read_extent(trans, rbio)) { + if (bch2_ec_read_extent(trans, rbio, k)) { bch2_rbio_error(rbio, READ_RETRY_AVOID, BLK_STS_IOERR); goto out; } @@ -1214,10 +1220,6 @@ void __bch2_read(struct bch_fs *c, struct bch_read_bio *rbio, swap(bvec_iter.bi_size, bytes); bio_advance_iter(&rbio->bio, &bvec_iter, bytes); - - ret = btree_trans_too_many_iters(trans); - if (ret) - goto err; err: if (ret && !bch2_err_matches(ret, BCH_ERR_transaction_restart) && diff --git a/fs/bcachefs/io_write.c b/fs/bcachefs/io_write.c index 1d4761d15002..d3b5be7fd9bf 100644 --- a/fs/bcachefs/io_write.c +++ b/fs/bcachefs/io_write.c @@ -1447,9 +1447,7 @@ again: op->nr_replicas_required, op->watermark, op->flags, - (op->flags & (BCH_WRITE_ALLOC_NOWAIT| - BCH_WRITE_ONLY_SPECIFIED_DEVS)) - ? NULL : &op->cl, &wp)); + &op->cl, &wp)); if (unlikely(ret)) { if (bch2_err_matches(ret, BCH_ERR_operation_blocked)) break; @@ -1592,6 +1590,9 @@ CLOSURE_CALLBACK(bch2_write) BUG_ON(!op->write_point.v); BUG_ON(bkey_eq(op->pos, POS_MAX)); + if (op->flags & BCH_WRITE_ONLY_SPECIFIED_DEVS) + op->flags |= BCH_WRITE_ALLOC_NOWAIT; + op->nr_replicas_required = min_t(unsigned, op->nr_replicas_required, op->nr_replicas); op->start_time = local_clock(); bch2_keylist_init(&op->insert_keys, op->inline_keys); diff --git a/fs/bcachefs/journal_io.c b/fs/bcachefs/journal_io.c index 7664b68e6a15..30460bce04be 100644 --- a/fs/bcachefs/journal_io.c +++ b/fs/bcachefs/journal_io.c @@ -1353,6 +1353,7 @@ int bch2_journal_read(struct bch_fs *c, genradix_for_each(&c->journal_entries, radix_iter, _i) { struct bch_replicas_padded replicas = { .e.data_type = BCH_DATA_journal, + .e.nr_devs = 0, .e.nr_required = 1, }; @@ -1379,7 +1380,7 @@ int bch2_journal_read(struct bch_fs *c, goto err; darray_for_each(i->ptrs, ptr) - replicas.e.devs[replicas.e.nr_devs++] = ptr->dev; + replicas_entry_add_dev(&replicas.e, ptr->dev); bch2_replicas_entry_sort(&replicas.e); @@ -1950,7 +1951,8 @@ static int bch2_journal_write_pick_flush(struct journal *j, struct journal_buf * if (error || w->noflush || (!w->must_flush && - (jiffies - j->last_flush_write) < msecs_to_jiffies(c->opts.journal_flush_delay) && + time_before(jiffies, j->last_flush_write + + msecs_to_jiffies(c->opts.journal_flush_delay)) && test_bit(JOURNAL_may_skip_flush, &j->flags))) { w->noflush = true; SET_JSET_NO_FLUSH(w->data, true); diff --git a/fs/bcachefs/journal_reclaim.c b/fs/bcachefs/journal_reclaim.c index 70b998d9f19c..ace291f175dd 100644 --- a/fs/bcachefs/journal_reclaim.c +++ b/fs/bcachefs/journal_reclaim.c @@ -641,6 +641,7 @@ static u64 journal_seq_to_flush(struct journal *j) static int __bch2_journal_reclaim(struct journal *j, bool direct, bool kicked) { struct bch_fs *c = container_of(j, struct bch_fs, journal); + struct btree_cache *bc = &c->btree_cache; bool kthread = (current->flags & PF_KTHREAD) != 0; u64 seq_to_flush; size_t min_nr, min_key_cache, nr_flushed; @@ -681,7 +682,8 @@ static int __bch2_journal_reclaim(struct journal *j, bool direct, bool kicked) if (j->watermark != BCH_WATERMARK_stripe) min_nr = 1; - if (atomic_read(&c->btree_cache.dirty) * 2 > c->btree_cache.used) + size_t btree_cache_live = bc->live[0].nr + bc->live[1].nr; + if (atomic_long_read(&bc->nr_dirty) * 2 > btree_cache_live) min_nr = 1; min_key_cache = min(bch2_nr_btree_keys_need_flush(c), (size_t) 128); @@ -689,8 +691,7 @@ static int __bch2_journal_reclaim(struct journal *j, bool direct, bool kicked) trace_and_count(c, journal_reclaim_start, c, direct, kicked, min_nr, min_key_cache, - atomic_read(&c->btree_cache.dirty), - c->btree_cache.used, + atomic_long_read(&bc->nr_dirty), btree_cache_live, atomic_long_read(&c->btree_key_cache.nr_dirty), atomic_long_read(&c->btree_key_cache.nr_keys)); diff --git a/fs/bcachefs/opts.c b/fs/bcachefs/opts.c index e10fc1da71b1..232be8a44051 100644 --- a/fs/bcachefs/opts.c +++ b/fs/bcachefs/opts.c @@ -230,6 +230,8 @@ const struct bch_option bch2_opt_table[] = { #define OPT_STR_NOLIMIT(_choices) .type = BCH_OPT_STR, \ .min = 0, .max = U64_MAX, \ .choices = _choices +#define OPT_BITFIELD(_choices) .type = BCH_OPT_BITFIELD, \ + .choices = _choices #define OPT_FN(_fn) .type = BCH_OPT_FN, .fn = _fn #define x(_name, _bits, _flags, _type, _sb_opt, _default, _hint, _help) \ @@ -376,6 +378,13 @@ int bch2_opt_parse(struct bch_fs *c, *res = ret; break; + case BCH_OPT_BITFIELD: { + s64 v = bch2_read_flag_list(val, opt->choices); + if (v < 0) + return v; + *res = v; + break; + } case BCH_OPT_FN: ret = opt->fn.parse(c, val, res, err); @@ -423,6 +432,9 @@ void bch2_opt_to_text(struct printbuf *out, else prt_str(out, opt->choices[v]); break; + case BCH_OPT_BITFIELD: + prt_bitflags(out, opt->choices, v); + break; case BCH_OPT_FN: opt->fn.to_text(out, c, sb, v); break; @@ -431,6 +443,32 @@ void bch2_opt_to_text(struct printbuf *out, } } +void bch2_opts_to_text(struct printbuf *out, + struct bch_opts opts, + struct bch_fs *c, struct bch_sb *sb, + unsigned show_mask, unsigned hide_mask, + unsigned flags) +{ + bool first = true; + + for (enum bch_opt_id i = 0; i < bch2_opts_nr; i++) { + const struct bch_option *opt = &bch2_opt_table[i]; + + if ((opt->flags & hide_mask) || !(opt->flags & show_mask)) + continue; + + u64 v = bch2_opt_get_by_id(&opts, i); + if (v == bch2_opt_get_by_id(&bch2_opts_default, i)) + continue; + + if (!first) + prt_char(out, ','); + first = false; + + bch2_opt_to_text(out, c, sb, opt, v, flags); + } +} + int bch2_opt_check_may_set(struct bch_fs *c, int id, u64 v) { int ret = 0; @@ -608,10 +646,20 @@ int bch2_opts_from_sb(struct bch_opts *opts, struct bch_sb *sb) return 0; } -void __bch2_opt_set_sb(struct bch_sb *sb, const struct bch_option *opt, u64 v) +struct bch_dev_sb_opt_set { + void (*set_sb)(struct bch_member *, u64); +}; + +static const struct bch_dev_sb_opt_set bch2_dev_sb_opt_setters [] = { +#define x(n, set) [Opt_##n] = { .set_sb = SET_##set }, + BCH_DEV_OPT_SETTERS() +#undef x +}; + +void __bch2_opt_set_sb(struct bch_sb *sb, int dev_idx, + const struct bch_option *opt, u64 v) { - if (opt->set_sb == SET_BCH2_NO_SB_OPT) - return; + enum bch_opt_id id = opt - bch2_opt_table; if (opt->flags & OPT_SB_FIELD_SECTORS) v >>= 9; @@ -619,16 +667,35 @@ void __bch2_opt_set_sb(struct bch_sb *sb, const struct bch_option *opt, u64 v) if (opt->flags & OPT_SB_FIELD_ILOG2) v = ilog2(v); - opt->set_sb(sb, v); + if (opt->flags & OPT_SB_FIELD_ONE_BIAS) + v++; + + if (opt->flags & OPT_FS) { + if (opt->set_sb != SET_BCH2_NO_SB_OPT) + opt->set_sb(sb, v); + } + + if ((opt->flags & OPT_DEVICE) && dev_idx >= 0) { + if (WARN(!bch2_member_exists(sb, dev_idx), + "tried to set device option %s on nonexistent device %i", + opt->attr.name, dev_idx)) + return; + + struct bch_member *m = bch2_members_v2_get_mut(sb, dev_idx); + + const struct bch_dev_sb_opt_set *set = bch2_dev_sb_opt_setters + id; + if (set->set_sb) + set->set_sb(m, v); + else + pr_err("option %s cannot be set via opt_set_sb()", opt->attr.name); + } } -void bch2_opt_set_sb(struct bch_fs *c, const struct bch_option *opt, u64 v) +void bch2_opt_set_sb(struct bch_fs *c, struct bch_dev *ca, + const struct bch_option *opt, u64 v) { - if (opt->set_sb == SET_BCH2_NO_SB_OPT) - return; - mutex_lock(&c->sb_lock); - __bch2_opt_set_sb(c->disk_sb.sb, opt, v); + __bch2_opt_set_sb(c->disk_sb.sb, ca ? ca->dev_idx : -1, opt, v); bch2_write_super(c); mutex_unlock(&c->sb_lock); } diff --git a/fs/bcachefs/opts.h b/fs/bcachefs/opts.h index cda1725702ea..cb2e244a2429 100644 --- a/fs/bcachefs/opts.h +++ b/fs/bcachefs/opts.h @@ -53,23 +53,25 @@ void SET_BCH2_NO_SB_OPT(struct bch_sb *, u64); /* When can be set: */ enum opt_flags { - OPT_FS = (1 << 0), /* Filesystem option */ - OPT_DEVICE = (1 << 1), /* Device option */ - OPT_INODE = (1 << 2), /* Inode option */ - OPT_FORMAT = (1 << 3), /* May be specified at format time */ - OPT_MOUNT = (1 << 4), /* May be specified at mount time */ - OPT_RUNTIME = (1 << 5), /* May be specified at runtime */ - OPT_HUMAN_READABLE = (1 << 6), - OPT_MUST_BE_POW_2 = (1 << 7), /* Must be power of 2 */ - OPT_SB_FIELD_SECTORS = (1 << 8),/* Superblock field is >> 9 of actual value */ - OPT_SB_FIELD_ILOG2 = (1 << 9), /* Superblock field is ilog2 of actual value */ - OPT_HIDDEN = (1 << 10), + OPT_FS = BIT(0), /* Filesystem option */ + OPT_DEVICE = BIT(1), /* Device option */ + OPT_INODE = BIT(2), /* Inode option */ + OPT_FORMAT = BIT(3), /* May be specified at format time */ + OPT_MOUNT = BIT(4), /* May be specified at mount time */ + OPT_RUNTIME = BIT(5), /* May be specified at runtime */ + OPT_HUMAN_READABLE = BIT(6), + OPT_MUST_BE_POW_2 = BIT(7), /* Must be power of 2 */ + OPT_SB_FIELD_SECTORS = BIT(8), /* Superblock field is >> 9 of actual value */ + OPT_SB_FIELD_ILOG2 = BIT(9), /* Superblock field is ilog2 of actual value */ + OPT_SB_FIELD_ONE_BIAS = BIT(10), /* 0 means default value */ + OPT_HIDDEN = BIT(11), }; enum opt_type { BCH_OPT_BOOL, BCH_OPT_UINT, BCH_OPT_STR, + BCH_OPT_BITFIELD, BCH_OPT_FN, }; @@ -263,6 +265,11 @@ enum fsck_err_opts { OPT_BOOL(), \ BCH2_NO_SB_OPT, true, \ NULL, "Enable inline data extents") \ + x(promote_whole_extents, u8, \ + OPT_FS|OPT_MOUNT|OPT_RUNTIME, \ + OPT_BOOL(), \ + BCH_SB_PROMOTE_WHOLE_EXTENTS, true, \ + NULL, "Promote whole extents, instead of just part being read")\ x(acl, u8, \ OPT_FS|OPT_FORMAT|OPT_MOUNT, \ OPT_BOOL(), \ @@ -366,6 +373,16 @@ enum fsck_err_opts { OPT_BOOL(), \ BCH2_NO_SB_OPT, false, \ NULL, "Exit recovery immediately prior to journal replay")\ + x(recovery_passes, u64, \ + OPT_FS|OPT_MOUNT, \ + OPT_BITFIELD(bch2_recovery_passes), \ + BCH2_NO_SB_OPT, 0, \ + NULL, "Recovery passes to run explicitly") \ + x(recovery_passes_exclude, u64, \ + OPT_FS|OPT_MOUNT, \ + OPT_BITFIELD(bch2_recovery_passes), \ + BCH2_NO_SB_OPT, 0, \ + NULL, "Recovery passes to exclude") \ x(recovery_pass_last, u8, \ OPT_FS|OPT_MOUNT, \ OPT_STR_NOLIMIT(bch2_recovery_passes), \ @@ -472,11 +489,16 @@ enum fsck_err_opts { BCH2_NO_SB_OPT, 0, \ "size", "Size of filesystem on device") \ x(durability, u8, \ - OPT_DEVICE, \ + OPT_DEVICE|OPT_SB_FIELD_ONE_BIAS, \ OPT_UINT(0, BCH_REPLICAS_MAX), \ BCH2_NO_SB_OPT, 1, \ "n", "Data written to this device will be considered\n"\ "to have already been replicated n times") \ + x(data_allowed, u8, \ + OPT_DEVICE, \ + OPT_BITFIELD(__bch2_data_types), \ + BCH2_NO_SB_OPT, BIT(BCH_DATA_journal)|BIT(BCH_DATA_btree)|BIT(BCH_DATA_user),\ + "types", "Allowed data types for this device: journal, btree, and/or user")\ x(btree_node_prefetch, u8, \ OPT_FS|OPT_MOUNT|OPT_RUNTIME, \ OPT_BOOL(), \ @@ -484,6 +506,11 @@ enum fsck_err_opts { NULL, "BTREE_ITER_prefetch casuse btree nodes to be\n"\ " prefetched sequentially") +#define BCH_DEV_OPT_SETTERS() \ + x(discard, BCH_MEMBER_DISCARD) \ + x(durability, BCH_MEMBER_DURABILITY) \ + x(data_allowed, BCH_MEMBER_DATA_ALLOWED) + struct bch_opts { #define x(_name, _bits, ...) unsigned _name##_defined:1; BCH_OPTS() @@ -563,8 +590,10 @@ void bch2_opt_set_by_id(struct bch_opts *, enum bch_opt_id, u64); u64 bch2_opt_from_sb(struct bch_sb *, enum bch_opt_id); int bch2_opts_from_sb(struct bch_opts *, struct bch_sb *); -void __bch2_opt_set_sb(struct bch_sb *, const struct bch_option *, u64); -void bch2_opt_set_sb(struct bch_fs *, const struct bch_option *, u64); +void __bch2_opt_set_sb(struct bch_sb *, int, const struct bch_option *, u64); + +struct bch_dev; +void bch2_opt_set_sb(struct bch_fs *, struct bch_dev *, const struct bch_option *, u64); int bch2_opt_lookup(const char *); int bch2_opt_validate(const struct bch_option *, u64, struct printbuf *); @@ -576,6 +605,10 @@ int bch2_opt_parse(struct bch_fs *, const struct bch_option *, void bch2_opt_to_text(struct printbuf *, struct bch_fs *, struct bch_sb *, const struct bch_option *, u64, unsigned); +void bch2_opts_to_text(struct printbuf *, + struct bch_opts, + struct bch_fs *, struct bch_sb *, + unsigned, unsigned, unsigned); int bch2_opt_check_may_set(struct bch_fs *, int, u64); int bch2_opts_check_may_set(struct bch_fs *); diff --git a/fs/bcachefs/rcu_pending.c b/fs/bcachefs/rcu_pending.c new file mode 100644 index 000000000000..40a20192eee8 --- /dev/null +++ b/fs/bcachefs/rcu_pending.c @@ -0,0 +1,650 @@ +// SPDX-License-Identifier: GPL-2.0 +#define pr_fmt(fmt) "%s() " fmt "\n", __func__ + +#include +#include +#include +#include +#include +#include + +#include "rcu_pending.h" +#include "darray.h" +#include "util.h" + +#define static_array_for_each(_a, _i) \ + for (typeof(&(_a)[0]) _i = _a; \ + _i < (_a) + ARRAY_SIZE(_a); \ + _i++) + +enum rcu_pending_special { + RCU_PENDING_KVFREE = 1, + RCU_PENDING_CALL_RCU = 2, +}; + +#define RCU_PENDING_KVFREE_FN ((rcu_pending_process_fn) (ulong) RCU_PENDING_KVFREE) +#define RCU_PENDING_CALL_RCU_FN ((rcu_pending_process_fn) (ulong) RCU_PENDING_CALL_RCU) + +static inline unsigned long __get_state_synchronize_rcu(struct srcu_struct *ssp) +{ + return ssp + ? get_state_synchronize_srcu(ssp) + : get_state_synchronize_rcu(); +} + +static inline unsigned long __start_poll_synchronize_rcu(struct srcu_struct *ssp) +{ + return ssp + ? start_poll_synchronize_srcu(ssp) + : start_poll_synchronize_rcu(); +} + +static inline bool __poll_state_synchronize_rcu(struct srcu_struct *ssp, unsigned long cookie) +{ + return ssp + ? poll_state_synchronize_srcu(ssp, cookie) + : poll_state_synchronize_rcu(cookie); +} + +static inline void __rcu_barrier(struct srcu_struct *ssp) +{ + return ssp + ? srcu_barrier(ssp) + : rcu_barrier(); +} + +static inline void __call_rcu(struct srcu_struct *ssp, struct rcu_head *rhp, + rcu_callback_t func) +{ + if (ssp) + call_srcu(ssp, rhp, func); + else + call_rcu(rhp, func); +} + +struct rcu_pending_seq { + /* + * We're using a radix tree like a vector - we're just pushing elements + * onto the end; we're using a radix tree instead of an actual vector to + * avoid reallocation overhead + */ + GENRADIX(struct rcu_head *) objs; + size_t nr; + struct rcu_head **cursor; + unsigned long seq; +}; + +struct rcu_pending_list { + struct rcu_head *head; + struct rcu_head *tail; + unsigned long seq; +}; + +struct rcu_pending_pcpu { + struct rcu_pending *parent; + spinlock_t lock; + int cpu; + + /* + * We can't bound the number of unprocessed gp sequence numbers, and we + * can't efficiently merge radix trees for expired grace periods, so we + * need darray/vector: + */ + DARRAY_PREALLOCATED(struct rcu_pending_seq, 4) objs; + + /* Third entry is for expired objects: */ + struct rcu_pending_list lists[NUM_ACTIVE_RCU_POLL_OLDSTATE + 1]; + + struct rcu_head cb; + bool cb_armed; + struct work_struct work; +}; + +static bool __rcu_pending_has_pending(struct rcu_pending_pcpu *p) +{ + if (p->objs.nr) + return true; + + static_array_for_each(p->lists, i) + if (i->head) + return true; + + return false; +} + +static void rcu_pending_list_merge(struct rcu_pending_list *l1, + struct rcu_pending_list *l2) +{ +#ifdef __KERNEL__ + if (!l1->head) + l1->head = l2->head; + else + l1->tail->next = l2->head; +#else + if (!l1->head) + l1->head = l2->head; + else + l1->tail->next.next = (void *) l2->head; +#endif + + l1->tail = l2->tail; + l2->head = l2->tail = NULL; +} + +static void rcu_pending_list_add(struct rcu_pending_list *l, + struct rcu_head *n) +{ +#ifdef __KERNEL__ + if (!l->head) + l->head = n; + else + l->tail->next = n; + l->tail = n; + n->next = NULL; +#else + if (!l->head) + l->head = n; + else + l->tail->next.next = (void *) n; + l->tail = n; + n->next.next = NULL; +#endif +} + +static void merge_expired_lists(struct rcu_pending_pcpu *p) +{ + struct rcu_pending_list *expired = &p->lists[NUM_ACTIVE_RCU_POLL_OLDSTATE]; + + for (struct rcu_pending_list *i = p->lists; i < expired; i++) + if (i->head && __poll_state_synchronize_rcu(p->parent->srcu, i->seq)) + rcu_pending_list_merge(expired, i); +} + +#ifndef __KERNEL__ +static inline void kfree_bulk(size_t nr, void ** p) +{ + while (nr--) + kfree(*p); +} + +#define local_irq_save(flags) \ +do { \ + flags = 0; \ +} while (0) +#endif + +static noinline void __process_finished_items(struct rcu_pending *pending, + struct rcu_pending_pcpu *p, + unsigned long flags) +{ + struct rcu_pending_list *expired = &p->lists[NUM_ACTIVE_RCU_POLL_OLDSTATE]; + struct rcu_pending_seq objs = {}; + struct rcu_head *list = NULL; + + if (p->objs.nr && + __poll_state_synchronize_rcu(pending->srcu, p->objs.data[0].seq)) { + objs = p->objs.data[0]; + darray_remove_item(&p->objs, p->objs.data); + } + + merge_expired_lists(p); + + list = expired->head; + expired->head = expired->tail = NULL; + + spin_unlock_irqrestore(&p->lock, flags); + + switch ((ulong) pending->process) { + case RCU_PENDING_KVFREE: + for (size_t i = 0; i < objs.nr; ) { + size_t nr_this_node = min(GENRADIX_NODE_SIZE / sizeof(void *), objs.nr - i); + + kfree_bulk(nr_this_node, (void **) genradix_ptr(&objs.objs, i)); + i += nr_this_node; + } + genradix_free(&objs.objs); + + while (list) { + struct rcu_head *obj = list; +#ifdef __KERNEL__ + list = obj->next; +#else + list = (void *) obj->next.next; +#endif + + /* + * low bit of pointer indicates whether rcu_head needs + * to be freed - kvfree_rcu_mightsleep() + */ + BUILD_BUG_ON(ARCH_SLAB_MINALIGN == 0); + + void *ptr = (void *)(((unsigned long) obj->func) & ~1UL); + bool free_head = ((unsigned long) obj->func) & 1UL; + + kvfree(ptr); + if (free_head) + kfree(obj); + } + + break; + + case RCU_PENDING_CALL_RCU: + for (size_t i = 0; i < objs.nr; i++) { + struct rcu_head *obj = *genradix_ptr(&objs.objs, i); + obj->func(obj); + } + genradix_free(&objs.objs); + + while (list) { + struct rcu_head *obj = list; +#ifdef __KERNEL__ + list = obj->next; +#else + list = (void *) obj->next.next; +#endif + obj->func(obj); + } + break; + + default: + for (size_t i = 0; i < objs.nr; i++) + pending->process(pending, *genradix_ptr(&objs.objs, i)); + genradix_free(&objs.objs); + + while (list) { + struct rcu_head *obj = list; +#ifdef __KERNEL__ + list = obj->next; +#else + list = (void *) obj->next.next; +#endif + pending->process(pending, obj); + } + break; + } +} + +static bool process_finished_items(struct rcu_pending *pending, + struct rcu_pending_pcpu *p, + unsigned long flags) +{ + /* + * XXX: we should grab the gp seq once and avoid multiple function + * calls, this is called from __rcu_pending_enqueue() fastpath in + * may_sleep==true mode + */ + if ((p->objs.nr && __poll_state_synchronize_rcu(pending->srcu, p->objs.data[0].seq)) || + (p->lists[0].head && __poll_state_synchronize_rcu(pending->srcu, p->lists[0].seq)) || + (p->lists[1].head && __poll_state_synchronize_rcu(pending->srcu, p->lists[1].seq)) || + p->lists[2].head) { + __process_finished_items(pending, p, flags); + return true; + } + + return false; +} + +static void rcu_pending_work(struct work_struct *work) +{ + struct rcu_pending_pcpu *p = + container_of(work, struct rcu_pending_pcpu, work); + struct rcu_pending *pending = p->parent; + unsigned long flags; + + do { + spin_lock_irqsave(&p->lock, flags); + } while (process_finished_items(pending, p, flags)); + + spin_unlock_irqrestore(&p->lock, flags); +} + +static void rcu_pending_rcu_cb(struct rcu_head *rcu) +{ + struct rcu_pending_pcpu *p = container_of(rcu, struct rcu_pending_pcpu, cb); + + schedule_work_on(p->cpu, &p->work); + + unsigned long flags; + spin_lock_irqsave(&p->lock, flags); + if (__rcu_pending_has_pending(p)) { + spin_unlock_irqrestore(&p->lock, flags); + __call_rcu(p->parent->srcu, &p->cb, rcu_pending_rcu_cb); + } else { + p->cb_armed = false; + spin_unlock_irqrestore(&p->lock, flags); + } +} + +static __always_inline struct rcu_pending_seq * +get_object_radix(struct rcu_pending_pcpu *p, unsigned long seq) +{ + darray_for_each_reverse(p->objs, objs) + if (objs->seq == seq) + return objs; + + if (darray_push_gfp(&p->objs, ((struct rcu_pending_seq) { .seq = seq }), GFP_ATOMIC)) + return NULL; + + return &darray_last(p->objs); +} + +static noinline bool +rcu_pending_enqueue_list(struct rcu_pending_pcpu *p, unsigned long seq, + struct rcu_head *head, void *ptr, + unsigned long *flags) +{ + if (ptr) { + if (!head) { + /* + * kvfree_rcu_mightsleep(): we weren't passed an + * rcu_head, but we need one: use the low bit of the + * ponter to free to flag that the head needs to be + * freed as well: + */ + ptr = (void *)(((unsigned long) ptr)|1UL); + head = kmalloc(sizeof(*head), __GFP_NOWARN); + if (!head) { + spin_unlock_irqrestore(&p->lock, *flags); + head = kmalloc(sizeof(*head), GFP_KERNEL|__GFP_NOFAIL); + /* + * dropped lock, did GFP_KERNEL allocation, + * check for gp expiration + */ + if (unlikely(__poll_state_synchronize_rcu(p->parent->srcu, seq))) { + kvfree(--ptr); + kfree(head); + spin_lock_irqsave(&p->lock, *flags); + return false; + } + } + } + + head->func = ptr; + } +again: + for (struct rcu_pending_list *i = p->lists; + i < p->lists + NUM_ACTIVE_RCU_POLL_OLDSTATE; i++) { + if (i->seq == seq) { + rcu_pending_list_add(i, head); + return false; + } + } + + for (struct rcu_pending_list *i = p->lists; + i < p->lists + NUM_ACTIVE_RCU_POLL_OLDSTATE; i++) { + if (!i->head) { + i->seq = seq; + rcu_pending_list_add(i, head); + return true; + } + } + + merge_expired_lists(p); + goto again; +} + +/* + * __rcu_pending_enqueue: enqueue a pending RCU item, to be processed (via + * pending->pracess) once grace period elapses. + * + * Attempt to enqueue items onto a radix tree; if memory allocation fails, fall + * back to a linked list. + * + * - If @ptr is NULL, we're enqueuing an item for a generic @pending with a + * process callback + * + * - If @ptr and @head are both not NULL, we're kvfree_rcu() + * + * - If @ptr is not NULL and @head is, we're kvfree_rcu_mightsleep() + * + * - If @may_sleep is true, will do GFP_KERNEL memory allocations and process + * expired items. + */ +static __always_inline void +__rcu_pending_enqueue(struct rcu_pending *pending, struct rcu_head *head, + void *ptr, bool may_sleep) +{ + + struct rcu_pending_pcpu *p; + struct rcu_pending_seq *objs; + struct genradix_node *new_node = NULL; + unsigned long seq, flags; + bool start_gp = false; + + BUG_ON((ptr != NULL) != (pending->process == RCU_PENDING_KVFREE_FN)); + + local_irq_save(flags); + p = this_cpu_ptr(pending->p); + spin_lock(&p->lock); + seq = __get_state_synchronize_rcu(pending->srcu); +restart: + if (may_sleep && + unlikely(process_finished_items(pending, p, flags))) + goto check_expired; + + /* + * In kvfree_rcu() mode, the radix tree is only for slab pointers so + * that we can do kfree_bulk() - vmalloc pointers always use the linked + * list: + */ + if (ptr && unlikely(is_vmalloc_addr(ptr))) + goto list_add; + + objs = get_object_radix(p, seq); + if (unlikely(!objs)) + goto list_add; + + if (unlikely(!objs->cursor)) { + /* + * New radix tree nodes must be added under @p->lock because the + * tree root is in a darray that can be resized (typically, + * genradix supports concurrent unlocked allocation of new + * nodes) - hence preallocation and the retry loop: + */ + objs->cursor = genradix_ptr_alloc_preallocated_inlined(&objs->objs, + objs->nr, &new_node, GFP_ATOMIC|__GFP_NOWARN); + if (unlikely(!objs->cursor)) { + if (may_sleep) { + spin_unlock_irqrestore(&p->lock, flags); + + gfp_t gfp = GFP_KERNEL; + if (!head) + gfp |= __GFP_NOFAIL; + + new_node = genradix_alloc_node(gfp); + if (!new_node) + may_sleep = false; + goto check_expired; + } +list_add: + start_gp = rcu_pending_enqueue_list(p, seq, head, ptr, &flags); + goto start_gp; + } + } + + *objs->cursor++ = ptr ?: head; + /* zero cursor if we hit the end of a radix tree node: */ + if (!(((ulong) objs->cursor) & (GENRADIX_NODE_SIZE - 1))) + objs->cursor = NULL; + start_gp = !objs->nr; + objs->nr++; +start_gp: + if (unlikely(start_gp)) { + /* + * We only have one callback (ideally, we would have one for + * every outstanding graceperiod) - so if our callback is + * already in flight, we may still have to start a grace period + * (since we used get_state() above, not start_poll()) + */ + if (!p->cb_armed) { + p->cb_armed = true; + __call_rcu(pending->srcu, &p->cb, rcu_pending_rcu_cb); + } else { + __start_poll_synchronize_rcu(pending->srcu); + } + } + spin_unlock_irqrestore(&p->lock, flags); +free_node: + if (new_node) + genradix_free_node(new_node); + return; +check_expired: + if (unlikely(__poll_state_synchronize_rcu(pending->srcu, seq))) { + switch ((ulong) pending->process) { + case RCU_PENDING_KVFREE: + kvfree(ptr); + break; + case RCU_PENDING_CALL_RCU: + head->func(head); + break; + default: + pending->process(pending, head); + break; + } + goto free_node; + } + + local_irq_save(flags); + p = this_cpu_ptr(pending->p); + spin_lock(&p->lock); + goto restart; +} + +void rcu_pending_enqueue(struct rcu_pending *pending, struct rcu_head *obj) +{ + __rcu_pending_enqueue(pending, obj, NULL, true); +} + +static struct rcu_head *rcu_pending_pcpu_dequeue(struct rcu_pending_pcpu *p) +{ + struct rcu_head *ret = NULL; + + spin_lock_irq(&p->lock); + darray_for_each(p->objs, objs) + if (objs->nr) { + ret = *genradix_ptr(&objs->objs, --objs->nr); + objs->cursor = NULL; + if (!objs->nr) + genradix_free(&objs->objs); + goto out; + } + + static_array_for_each(p->lists, i) + if (i->head) { + ret = i->head; +#ifdef __KERNEL__ + i->head = ret->next; +#else + i->head = (void *) ret->next.next; +#endif + if (!i->head) + i->tail = NULL; + goto out; + } +out: + spin_unlock_irq(&p->lock); + + return ret; +} + +struct rcu_head *rcu_pending_dequeue(struct rcu_pending *pending) +{ + return rcu_pending_pcpu_dequeue(raw_cpu_ptr(pending->p)); +} + +struct rcu_head *rcu_pending_dequeue_from_all(struct rcu_pending *pending) +{ + struct rcu_head *ret = rcu_pending_dequeue(pending); + + if (ret) + return ret; + + int cpu; + for_each_possible_cpu(cpu) { + ret = rcu_pending_pcpu_dequeue(per_cpu_ptr(pending->p, cpu)); + if (ret) + break; + } + return ret; +} + +static bool rcu_pending_has_pending_or_armed(struct rcu_pending *pending) +{ + int cpu; + for_each_possible_cpu(cpu) { + struct rcu_pending_pcpu *p = per_cpu_ptr(pending->p, cpu); + spin_lock_irq(&p->lock); + if (__rcu_pending_has_pending(p) || p->cb_armed) { + spin_unlock_irq(&p->lock); + return true; + } + spin_unlock_irq(&p->lock); + } + + return false; +} + +void rcu_pending_exit(struct rcu_pending *pending) +{ + int cpu; + + if (!pending->p) + return; + + while (rcu_pending_has_pending_or_armed(pending)) { + __rcu_barrier(pending->srcu); + + for_each_possible_cpu(cpu) { + struct rcu_pending_pcpu *p = per_cpu_ptr(pending->p, cpu); + flush_work(&p->work); + } + } + + for_each_possible_cpu(cpu) { + struct rcu_pending_pcpu *p = per_cpu_ptr(pending->p, cpu); + flush_work(&p->work); + } + + for_each_possible_cpu(cpu) { + struct rcu_pending_pcpu *p = per_cpu_ptr(pending->p, cpu); + + static_array_for_each(p->lists, i) + WARN_ON(i->head); + WARN_ON(p->objs.nr); + darray_exit(&p->objs); + } + free_percpu(pending->p); +} + +/** + * rcu_pending_init: - initialize a rcu_pending + * + * @pending: Object to init + * @srcu: May optionally be used with an srcu_struct; if NULL, uses normal + * RCU flavor + * @process: Callback function invoked on objects once their RCU barriers + * have completed; if NULL, kvfree() is used. + */ +int rcu_pending_init(struct rcu_pending *pending, + struct srcu_struct *srcu, + rcu_pending_process_fn process) +{ + pending->p = alloc_percpu(struct rcu_pending_pcpu); + if (!pending->p) + return -ENOMEM; + + int cpu; + for_each_possible_cpu(cpu) { + struct rcu_pending_pcpu *p = per_cpu_ptr(pending->p, cpu); + p->parent = pending; + p->cpu = cpu; + spin_lock_init(&p->lock); + darray_init(&p->objs); + INIT_WORK(&p->work, rcu_pending_work); + } + + pending->srcu = srcu; + pending->process = process; + + return 0; +} diff --git a/fs/bcachefs/rcu_pending.h b/fs/bcachefs/rcu_pending.h new file mode 100644 index 000000000000..71a2f4ddaade --- /dev/null +++ b/fs/bcachefs/rcu_pending.h @@ -0,0 +1,27 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +#ifndef _LINUX_RCU_PENDING_H +#define _LINUX_RCU_PENDING_H + +#include + +struct rcu_pending; +typedef void (*rcu_pending_process_fn)(struct rcu_pending *, struct rcu_head *); + +struct rcu_pending_pcpu; + +struct rcu_pending { + struct rcu_pending_pcpu __percpu *p; + struct srcu_struct *srcu; + rcu_pending_process_fn process; +}; + +void rcu_pending_enqueue(struct rcu_pending *pending, struct rcu_head *obj); +struct rcu_head *rcu_pending_dequeue(struct rcu_pending *pending); +struct rcu_head *rcu_pending_dequeue_from_all(struct rcu_pending *pending); + +void rcu_pending_exit(struct rcu_pending *pending); +int rcu_pending_init(struct rcu_pending *pending, + struct srcu_struct *srcu, + rcu_pending_process_fn process); + +#endif /* _LINUX_RCU_PENDING_H */ diff --git a/fs/bcachefs/rebalance.c b/fs/bcachefs/rebalance.c index cf81e5128c3a..2d299a37cf07 100644 --- a/fs/bcachefs/rebalance.c +++ b/fs/bcachefs/rebalance.c @@ -13,6 +13,7 @@ #include "errcode.h" #include "error.h" #include "inode.h" +#include "io_write.h" #include "move.h" #include "rebalance.h" #include "subvolume.h" @@ -156,6 +157,7 @@ static struct bkey_s_c next_rebalance_extent(struct btree_trans *trans, data_opts->rewrite_ptrs = bch2_bkey_ptrs_need_rebalance(c, k, r->target, r->compression); data_opts->target = r->target; + data_opts->write_flags |= BCH_WRITE_ONLY_SPECIFIED_DEVS; if (!data_opts->rewrite_ptrs) { /* @@ -263,6 +265,7 @@ static bool rebalance_pred(struct bch_fs *c, void *arg, data_opts->rewrite_ptrs = bch2_bkey_ptrs_need_rebalance(c, k, target, compression); data_opts->target = target; + data_opts->write_flags |= BCH_WRITE_ONLY_SPECIFIED_DEVS; return data_opts->rewrite_ptrs != 0; } diff --git a/fs/bcachefs/recovery.c b/fs/bcachefs/recovery.c index 36de1c6fe8c3..be1e7ca4362f 100644 --- a/fs/bcachefs/recovery.c +++ b/fs/bcachefs/recovery.c @@ -97,7 +97,7 @@ static void bch2_reconstruct_alloc(struct bch_fs *c) bch2_write_super(c); mutex_unlock(&c->sb_lock); - c->recovery_passes_explicit |= bch2_recovery_passes_from_stable(le64_to_cpu(ext->recovery_passes_required[0])); + c->opts.recovery_passes |= bch2_recovery_passes_from_stable(le64_to_cpu(ext->recovery_passes_required[0])); bch2_shoot_down_journal_keys(c, BTREE_ID_alloc, @@ -525,17 +525,17 @@ static int read_btree_roots(struct bch_fs *c) "error reading btree root %s l=%u: %s", bch2_btree_id_str(i), r->level, bch2_err_str(ret))) { if (btree_id_is_alloc(i)) { - c->recovery_passes_explicit |= BIT_ULL(BCH_RECOVERY_PASS_check_allocations); - c->recovery_passes_explicit |= BIT_ULL(BCH_RECOVERY_PASS_check_alloc_info); - c->recovery_passes_explicit |= BIT_ULL(BCH_RECOVERY_PASS_check_lrus); - c->recovery_passes_explicit |= BIT_ULL(BCH_RECOVERY_PASS_check_extents_to_backpointers); - c->recovery_passes_explicit |= BIT_ULL(BCH_RECOVERY_PASS_check_alloc_to_lru_refs); + c->opts.recovery_passes |= BIT_ULL(BCH_RECOVERY_PASS_check_allocations); + c->opts.recovery_passes |= BIT_ULL(BCH_RECOVERY_PASS_check_alloc_info); + c->opts.recovery_passes |= BIT_ULL(BCH_RECOVERY_PASS_check_lrus); + c->opts.recovery_passes |= BIT_ULL(BCH_RECOVERY_PASS_check_extents_to_backpointers); + c->opts.recovery_passes |= BIT_ULL(BCH_RECOVERY_PASS_check_alloc_to_lru_refs); c->sb.compat &= ~(1ULL << BCH_COMPAT_alloc_info); r->error = 0; - } else if (!(c->recovery_passes_explicit & BIT_ULL(BCH_RECOVERY_PASS_scan_for_btree_nodes))) { + } else if (!(c->opts.recovery_passes & BIT_ULL(BCH_RECOVERY_PASS_scan_for_btree_nodes))) { bch_info(c, "will run btree node scan"); - c->recovery_passes_explicit |= BIT_ULL(BCH_RECOVERY_PASS_scan_for_btree_nodes); - c->recovery_passes_explicit |= BIT_ULL(BCH_RECOVERY_PASS_check_topology); + c->opts.recovery_passes |= BIT_ULL(BCH_RECOVERY_PASS_scan_for_btree_nodes); + c->opts.recovery_passes |= BIT_ULL(BCH_RECOVERY_PASS_check_topology); } ret = 0; @@ -706,14 +706,14 @@ int bch2_fs_recovery(struct bch_fs *c) if (check_version_upgrade(c)) write_sb = true; - c->recovery_passes_explicit |= bch2_recovery_passes_from_stable(le64_to_cpu(ext->recovery_passes_required[0])); + c->opts.recovery_passes |= bch2_recovery_passes_from_stable(le64_to_cpu(ext->recovery_passes_required[0])); if (write_sb) bch2_write_super(c); mutex_unlock(&c->sb_lock); if (c->opts.fsck && IS_ENABLED(CONFIG_BCACHEFS_DEBUG)) - c->recovery_passes_explicit |= BIT_ULL(BCH_RECOVERY_PASS_check_topology); + c->opts.recovery_passes |= BIT_ULL(BCH_RECOVERY_PASS_check_topology); if (c->opts.fsck) set_bit(BCH_FS_fsck_running, &c->flags); diff --git a/fs/bcachefs/recovery_passes.c b/fs/bcachefs/recovery_passes.c index 73339a0a3111..735b8adc8f9d 100644 --- a/fs/bcachefs/recovery_passes.c +++ b/fs/bcachefs/recovery_passes.c @@ -40,7 +40,7 @@ static int bch2_set_may_go_rw(struct bch_fs *c) set_bit(BCH_FS_may_go_rw, &c->flags); - if (keys->nr || c->opts.fsck || !c->sb.clean || c->recovery_passes_explicit) + if (keys->nr || c->opts.fsck || !c->sb.clean || c->opts.recovery_passes) return bch2_fs_read_write_early(c); return 0; } @@ -97,14 +97,14 @@ u64 bch2_recovery_passes_from_stable(u64 v) int bch2_run_explicit_recovery_pass(struct bch_fs *c, enum bch_recovery_pass pass) { - if (c->recovery_passes_explicit & BIT_ULL(pass)) + if (c->opts.recovery_passes & BIT_ULL(pass)) return 0; bch_info(c, "running explicit recovery pass %s (%u), currently at %s (%u)", bch2_recovery_passes[pass], pass, bch2_recovery_passes[c->curr_recovery_pass], c->curr_recovery_pass); - c->recovery_passes_explicit |= BIT_ULL(pass); + c->opts.recovery_passes |= BIT_ULL(pass); if (c->curr_recovery_pass >= pass) { c->curr_recovery_pass = pass; @@ -161,7 +161,9 @@ static bool should_run_recovery_pass(struct bch_fs *c, enum bch_recovery_pass pa { struct recovery_pass_fn *p = recovery_pass_fns + pass; - if (c->recovery_passes_explicit & BIT_ULL(pass)) + if (c->opts.recovery_passes_exclude & BIT_ULL(pass)) + return false; + if (c->opts.recovery_passes & BIT_ULL(pass)) return true; if ((p->when & PASS_FSCK) && c->opts.fsck) return true; diff --git a/fs/bcachefs/replicas.c b/fs/bcachefs/replicas.c index 1f34c92a6d11..998c0bd06802 100644 --- a/fs/bcachefs/replicas.c +++ b/fs/bcachefs/replicas.c @@ -123,7 +123,7 @@ static void extent_to_replicas(struct bkey_s_c k, continue; if (!p.has_ec) - r->devs[r->nr_devs++] = p.ptr.dev; + replicas_entry_add_dev(r, p.ptr.dev); else r->nr_required = 0; } @@ -140,7 +140,7 @@ static void stripe_to_replicas(struct bkey_s_c k, for (ptr = s.v->ptrs; ptr < s.v->ptrs + s.v->nr_blocks; ptr++) - r->devs[r->nr_devs++] = ptr->dev; + replicas_entry_add_dev(r, ptr->dev); } void bch2_bkey_to_replicas(struct bch_replicas_entry_v1 *e, @@ -181,7 +181,7 @@ void bch2_devlist_to_replicas(struct bch_replicas_entry_v1 *e, e->nr_required = 1; darray_for_each(devs, i) - e->devs[e->nr_devs++] = *i; + replicas_entry_add_dev(e, *i); bch2_replicas_entry_sort(e); } @@ -795,12 +795,12 @@ bool bch2_have_enough_devs(struct bch_fs *c, struct bch_devs_mask devs, for (unsigned i = 0; i < e->nr_devs; i++) { nr_online += test_bit(e->devs[i], devs.d); - struct bch_dev *ca = bch2_dev_rcu(c, e->devs[i]); + struct bch_dev *ca = bch2_dev_rcu_noerror(c, e->devs[i]); nr_failed += !ca || ca->mi.state == BCH_MEMBER_STATE_failed; } rcu_read_unlock(); - if (nr_failed == e->nr_devs) + if (nr_online + nr_failed == e->nr_devs) continue; if (nr_online < e->nr_required) diff --git a/fs/bcachefs/replicas_format.h b/fs/bcachefs/replicas_format.h index b97208195d06..b7eff904acdb 100644 --- a/fs/bcachefs/replicas_format.h +++ b/fs/bcachefs/replicas_format.h @@ -5,7 +5,7 @@ struct bch_replicas_entry_v0 { __u8 data_type; __u8 nr_devs; - __u8 devs[]; + __u8 devs[] __counted_by(nr_devs); } __packed; struct bch_sb_field_replicas_v0 { @@ -17,7 +17,7 @@ struct bch_replicas_entry_v1 { __u8 data_type; __u8 nr_devs; __u8 nr_required; - __u8 devs[]; + __u8 devs[] __counted_by(nr_devs); } __packed; struct bch_sb_field_replicas { @@ -28,4 +28,9 @@ struct bch_sb_field_replicas { #define replicas_entry_bytes(_i) \ (offsetof(typeof(*(_i)), devs) + (_i)->nr_devs) +#define replicas_entry_add_dev(e, d) ({ \ + (e)->nr_devs++; \ + (e)->devs[(e)->nr_devs - 1] = (d); \ +}) + #endif /* _BCACHEFS_REPLICAS_FORMAT_H */ diff --git a/fs/bcachefs/sb-clean.c b/fs/bcachefs/sb-clean.c index c57d42bb8d1b..025848a9c4c0 100644 --- a/fs/bcachefs/sb-clean.c +++ b/fs/bcachefs/sb-clean.c @@ -155,7 +155,7 @@ struct bch_sb_field_clean *bch2_read_superblock_clean(struct bch_fs *c) SET_BCH_SB_CLEAN(c->disk_sb.sb, false); c->sb.clean = false; mutex_unlock(&c->sb_lock); - return NULL; + return ERR_PTR(-BCH_ERR_invalid_sb_clean); } clean = kmemdup(sb_clean, vstruct_bytes(&sb_clean->field), diff --git a/fs/bcachefs/sb-members.c b/fs/bcachefs/sb-members.c index 4b765422dd77..02bcde3c1b02 100644 --- a/fs/bcachefs/sb-members.c +++ b/fs/bcachefs/sb-members.c @@ -465,3 +465,60 @@ void bch2_dev_btree_bitmap_mark(struct bch_fs *c, struct bkey_s_c k) __bch2_dev_btree_bitmap_mark(mi, ptr->dev, ptr->offset, btree_sectors(c)); } } + +unsigned bch2_sb_nr_devices(const struct bch_sb *sb) +{ + unsigned nr = 0; + + for (unsigned i = 0; i < sb->nr_devices; i++) + nr += bch2_member_exists((struct bch_sb *) sb, i); + return nr; +} + +int bch2_sb_member_alloc(struct bch_fs *c) +{ + unsigned dev_idx = c->sb.nr_devices; + struct bch_sb_field_members_v2 *mi; + unsigned nr_devices; + unsigned u64s; + int best = -1; + u64 best_last_mount = 0; + + if (dev_idx < BCH_SB_MEMBERS_MAX) + goto have_slot; + + for (dev_idx = 0; dev_idx < BCH_SB_MEMBERS_MAX; dev_idx++) { + /* eventually BCH_SB_MEMBERS_MAX will be raised */ + if (dev_idx == BCH_SB_MEMBER_INVALID) + continue; + + struct bch_member m = bch2_sb_member_get(c->disk_sb.sb, dev_idx); + if (bch2_member_alive(&m)) + continue; + + u64 last_mount = le64_to_cpu(m.last_mount); + if (best < 0 || last_mount < best_last_mount) { + best = dev_idx; + best_last_mount = last_mount; + } + } + if (best >= 0) { + dev_idx = best; + goto have_slot; + } + + return -BCH_ERR_ENOSPC_sb_members; +have_slot: + nr_devices = max_t(unsigned, dev_idx + 1, c->sb.nr_devices); + + mi = bch2_sb_field_get(c->disk_sb.sb, members_v2); + u64s = DIV_ROUND_UP(sizeof(struct bch_sb_field_members_v2) + + le16_to_cpu(mi->member_bytes) * nr_devices, sizeof(u64)); + + mi = bch2_sb_field_resize(&c->disk_sb, members_v2, u64s); + if (!mi) + return -BCH_ERR_ENOSPC_sb_members; + + c->disk_sb.sb->nr_devices = nr_devices; + return dev_idx; +} diff --git a/fs/bcachefs/sb-members.h b/fs/bcachefs/sb-members.h index dd93192ec065..762083b564ee 100644 --- a/fs/bcachefs/sb-members.h +++ b/fs/bcachefs/sb-members.h @@ -198,29 +198,37 @@ static inline struct bch_dev *bch2_dev_locked(struct bch_fs *c, unsigned dev) lockdep_is_held(&c->state_lock)); } -static inline struct bch_dev *bch2_dev_rcu(struct bch_fs *c, unsigned dev) +static inline struct bch_dev *bch2_dev_rcu_noerror(struct bch_fs *c, unsigned dev) { return c && dev < c->sb.nr_devices ? rcu_dereference(c->devs[dev]) : NULL; } +void bch2_dev_missing(struct bch_fs *, unsigned); + +static inline struct bch_dev *bch2_dev_rcu(struct bch_fs *c, unsigned dev) +{ + struct bch_dev *ca = bch2_dev_rcu_noerror(c, dev); + if (unlikely(!ca)) + bch2_dev_missing(c, dev); + return ca; +} + static inline struct bch_dev *bch2_dev_tryget_noerror(struct bch_fs *c, unsigned dev) { rcu_read_lock(); - struct bch_dev *ca = bch2_dev_rcu(c, dev); + struct bch_dev *ca = bch2_dev_rcu_noerror(c, dev); if (ca) bch2_dev_get(ca); rcu_read_unlock(); return ca; } -void bch2_dev_missing(struct bch_fs *, unsigned); - static inline struct bch_dev *bch2_dev_tryget(struct bch_fs *c, unsigned dev) { struct bch_dev *ca = bch2_dev_tryget_noerror(c, dev); - if (!ca) + if (unlikely(!ca)) bch2_dev_missing(c, dev); return ca; } @@ -307,6 +315,8 @@ static inline bool bch2_member_exists(struct bch_sb *sb, unsigned dev) return false; } +unsigned bch2_sb_nr_devices(const struct bch_sb *); + static inline struct bch_member_cpu bch2_mi_to_cpu(struct bch_member *mi) { return (struct bch_member_cpu) { @@ -352,4 +362,6 @@ static inline bool bch2_dev_btree_bitmap_marked_sectors(struct bch_dev *ca, u64 bool bch2_dev_btree_bitmap_marked(struct bch_fs *, struct bkey_s_c); void bch2_dev_btree_bitmap_mark(struct bch_fs *, struct bkey_s_c); +int bch2_sb_member_alloc(struct bch_fs *); + #endif /* _BCACHEFS_SB_MEMBERS_H */ diff --git a/fs/bcachefs/str_hash.h b/fs/bcachefs/str_hash.h index c8c266cb5797..215eed4cce6d 100644 --- a/fs/bcachefs/str_hash.h +++ b/fs/bcachefs/str_hash.h @@ -270,7 +270,7 @@ int bch2_hash_set_in_snapshot(struct btree_trans *trans, desc.hash_bkey(info, bkey_i_to_s_c(insert)), snapshot), POS(insert->k.p.inode, U64_MAX), - BTREE_ITER_slots|BTREE_ITER_intent, k, ret) { + BTREE_ITER_slots|BTREE_ITER_intent|flags, k, ret) { if (is_visible_key(desc, inum, k)) { if (!desc.cmp_bkey(k, bkey_i_to_s_c(insert))) goto found; diff --git a/fs/bcachefs/subvolume.h b/fs/bcachefs/subvolume.h index a8299ba2cab2..e62f876541fe 100644 --- a/fs/bcachefs/subvolume.h +++ b/fs/bcachefs/subvolume.h @@ -31,6 +31,51 @@ int bch2_subvolume_get_snapshot(struct btree_trans *, u32, u32 *); int bch2_subvol_is_ro_trans(struct btree_trans *, u32); int bch2_subvol_is_ro(struct bch_fs *, u32); +static inline struct bkey_s_c +bch2_btree_iter_peek_in_subvolume_upto_type(struct btree_iter *iter, struct bpos end, + u32 subvolid, unsigned flags) +{ + u32 snapshot; + int ret = bch2_subvolume_get_snapshot(iter->trans, subvolid, &snapshot); + if (ret) + return bkey_s_c_err(ret); + + bch2_btree_iter_set_snapshot(iter, snapshot); + return bch2_btree_iter_peek_upto_type(iter, end, flags); +} + +#define for_each_btree_key_in_subvolume_upto_continue(_trans, _iter, \ + _end, _subvolid, _flags, _k, _do) \ +({ \ + struct bkey_s_c _k; \ + int _ret3 = 0; \ + \ + do { \ + _ret3 = lockrestart_do(_trans, ({ \ + (_k) = bch2_btree_iter_peek_in_subvolume_upto_type(&(_iter), \ + _end, _subvolid, (_flags)); \ + if (!(_k).k) \ + break; \ + \ + bkey_err(_k) ?: (_do); \ + })); \ + } while (!_ret3 && bch2_btree_iter_advance(&(_iter))); \ + \ + bch2_trans_iter_exit((_trans), &(_iter)); \ + _ret3; \ +}) + +#define for_each_btree_key_in_subvolume_upto(_trans, _iter, _btree_id, \ + _start, _end, _subvolid, _flags, _k, _do) \ +({ \ + struct btree_iter _iter; \ + bch2_trans_iter_init((_trans), &(_iter), (_btree_id), \ + (_start), (_flags)); \ + \ + for_each_btree_key_in_subvolume_upto_continue(_trans, _iter, \ + _end, _subvolid, _flags, _k, _do); \ +}) + int bch2_delete_dead_snapshots(struct bch_fs *); void bch2_delete_dead_snapshots_async(struct bch_fs *); diff --git a/fs/bcachefs/subvolume_types.h b/fs/bcachefs/subvolume_types.h index 9b10c8947828..f2ec4277c2a5 100644 --- a/fs/bcachefs/subvolume_types.h +++ b/fs/bcachefs/subvolume_types.h @@ -30,7 +30,8 @@ struct snapshot_table { }; typedef struct { - u32 subvol; + /* we can't have padding in this struct: */ + u64 subvol; u64 inum; } subvol_inum; diff --git a/fs/bcachefs/super-io.c b/fs/bcachefs/super-io.c index c8c2ccbdfbb5..d86d5dae54c9 100644 --- a/fs/bcachefs/super-io.c +++ b/fs/bcachefs/super-io.c @@ -418,6 +418,9 @@ static int bch2_sb_validate(struct bch_sb_handle *disk_sb, if (le16_to_cpu(sb->version) <= bcachefs_metadata_version_disk_accounting_v2 && !BCH_SB_ALLOCATOR_STUCK_TIMEOUT(sb)) SET_BCH_SB_ALLOCATOR_STUCK_TIMEOUT(sb, 30); + + if (le16_to_cpu(sb->version) <= bcachefs_metadata_version_disk_accounting_v2) + SET_BCH_SB_PROMOTE_WHOLE_EXTENTS(sb, true); } for (opt_id = 0; opt_id < bch2_opts_nr; opt_id++) { @@ -1292,15 +1295,9 @@ void bch2_sb_layout_to_text(struct printbuf *out, struct bch_sb_layout *l) void bch2_sb_to_text(struct printbuf *out, struct bch_sb *sb, bool print_layout, unsigned fields) { - u64 fields_have = 0; - unsigned nr_devices = 0; - if (!out->nr_tabstops) printbuf_tabstop_push(out, 44); - for (int i = 0; i < sb->nr_devices; i++) - nr_devices += bch2_member_exists(sb, i); - prt_printf(out, "External UUID:\t"); pr_uuid(out, sb->user_uuid.b); prt_newline(out); @@ -1356,9 +1353,10 @@ void bch2_sb_to_text(struct printbuf *out, struct bch_sb *sb, prt_newline(out); prt_printf(out, "Clean:\t%llu\n", BCH_SB_CLEAN(sb)); - prt_printf(out, "Devices:\t%u\n", nr_devices); + prt_printf(out, "Devices:\t%u\n", bch2_sb_nr_devices(sb)); prt_printf(out, "Sections:\t"); + u64 fields_have = 0; vstruct_for_each(sb, f) fields_have |= 1 << le32_to_cpu(f->type); prt_bitflags(out, bch2_sb_fields, fields_have); diff --git a/fs/bcachefs/super.c b/fs/bcachefs/super.c index e7fa2de35014..873e4be7e1dc 100644 --- a/fs/bcachefs/super.c +++ b/fs/bcachefs/super.c @@ -370,7 +370,7 @@ void bch2_fs_read_only(struct bch_fs *c) test_bit(BCH_FS_clean_shutdown, &c->flags) && c->recovery_pass_done >= BCH_RECOVERY_PASS_journal_replay) { BUG_ON(c->journal.last_empty_seq != journal_cur_seq(&c->journal)); - BUG_ON(atomic_read(&c->btree_cache.dirty)); + BUG_ON(atomic_long_read(&c->btree_cache.nr_dirty)); BUG_ON(atomic_long_read(&c->btree_key_cache.nr_dirty)); BUG_ON(c->btree_write_buffer.inc.keys.nr); BUG_ON(c->btree_write_buffer.flushing.keys.nr); @@ -543,6 +543,7 @@ static void __bch2_fs_free(struct bch_fs *c) bch2_fs_fs_io_direct_exit(c); bch2_fs_fs_io_buffered_exit(c); bch2_fs_fsio_exit(c); + bch2_fs_vfs_exit(c); bch2_fs_ec_exit(c); bch2_fs_encryption_exit(c); bch2_fs_nocow_locking_exit(c); @@ -810,7 +811,6 @@ static struct bch_fs *bch2_fs_alloc(struct bch_sb *sb, struct bch_opts opts) c->copy_gc_enabled = 1; c->rebalance.enabled = 1; - c->promote_whole_extents = true; c->journal.flush_write_time = &c->times[BCH_TIME_journal_flush_write]; c->journal.noflush_write_time = &c->times[BCH_TIME_journal_noflush_write]; @@ -926,6 +926,7 @@ static struct bch_fs *bch2_fs_alloc(struct bch_sb *sb, struct bch_opts opts) bch2_fs_encryption_init(c) ?: bch2_fs_compress_init(c) ?: bch2_fs_ec_init(c) ?: + bch2_fs_vfs_init(c) ?: bch2_fs_fsio_init(c) ?: bch2_fs_fs_io_buffered_init(c) ?: bch2_fs_fs_io_direct_init(c); @@ -1591,33 +1592,6 @@ int bch2_dev_set_state(struct bch_fs *c, struct bch_dev *ca, /* Device add/removal: */ -static int bch2_dev_remove_alloc(struct bch_fs *c, struct bch_dev *ca) -{ - struct bpos start = POS(ca->dev_idx, 0); - struct bpos end = POS(ca->dev_idx, U64_MAX); - int ret; - - /* - * We clear the LRU and need_discard btrees first so that we don't race - * with bch2_do_invalidates() and bch2_do_discards() - */ - ret = bch2_btree_delete_range(c, BTREE_ID_lru, start, end, - BTREE_TRIGGER_norun, NULL) ?: - bch2_btree_delete_range(c, BTREE_ID_need_discard, start, end, - BTREE_TRIGGER_norun, NULL) ?: - bch2_btree_delete_range(c, BTREE_ID_freespace, start, end, - BTREE_TRIGGER_norun, NULL) ?: - bch2_btree_delete_range(c, BTREE_ID_backpointers, start, end, - BTREE_TRIGGER_norun, NULL) ?: - bch2_btree_delete_range(c, BTREE_ID_alloc, start, end, - BTREE_TRIGGER_norun, NULL) ?: - bch2_btree_delete_range(c, BTREE_ID_bucket_gens, start, end, - BTREE_TRIGGER_norun, NULL) ?: - bch2_dev_usage_remove(c, ca->dev_idx); - bch_err_msg(c, ret, "removing dev alloc info"); - return ret; -} - int bch2_dev_remove(struct bch_fs *c, struct bch_dev *ca, int flags) { struct bch_member *m; @@ -1729,9 +1703,6 @@ int bch2_dev_add(struct bch_fs *c, const char *path) struct bch_opts opts = bch2_opts_empty(); struct bch_sb_handle sb; struct bch_dev *ca = NULL; - struct bch_sb_field_members_v2 *mi; - struct bch_member dev_mi; - unsigned dev_idx, nr_devices, u64s; struct printbuf errbuf = PRINTBUF; struct printbuf label = PRINTBUF; int ret; @@ -1741,7 +1712,7 @@ int bch2_dev_add(struct bch_fs *c, const char *path) if (ret) goto err; - dev_mi = bch2_sb_member_get(sb.sb, sb.sb->dev_idx); + struct bch_member dev_mi = bch2_sb_member_get(sb.sb, sb.sb->dev_idx); if (BCH_MEMBER_GROUP(&dev_mi)) { bch2_disk_path_to_text_sb(&label, sb.sb, BCH_MEMBER_GROUP(&dev_mi) - 1); @@ -1779,55 +1750,19 @@ int bch2_dev_add(struct bch_fs *c, const char *path) goto err_unlock; if (dynamic_fault("bcachefs:add:no_slot")) - goto no_slot; + goto err_unlock; - if (c->sb.nr_devices < BCH_SB_MEMBERS_MAX) { - dev_idx = c->sb.nr_devices; - goto have_slot; - } - - int best = -1; - u64 best_last_mount = 0; - for (dev_idx = 0; dev_idx < BCH_SB_MEMBERS_MAX; dev_idx++) { - struct bch_member m = bch2_sb_member_get(c->disk_sb.sb, dev_idx); - if (bch2_member_alive(&m)) - continue; - - u64 last_mount = le64_to_cpu(m.last_mount); - if (best < 0 || last_mount < best_last_mount) { - best = dev_idx; - best_last_mount = last_mount; - } - } - if (best >= 0) { - dev_idx = best; - goto have_slot; - } -no_slot: - ret = -BCH_ERR_ENOSPC_sb_members; - bch_err_msg(c, ret, "setting up new superblock"); - goto err_unlock; - -have_slot: - nr_devices = max_t(unsigned, dev_idx + 1, c->sb.nr_devices); - - mi = bch2_sb_field_get(c->disk_sb.sb, members_v2); - u64s = DIV_ROUND_UP(sizeof(struct bch_sb_field_members_v2) + - le16_to_cpu(mi->member_bytes) * nr_devices, sizeof(u64)); - - mi = bch2_sb_field_resize(&c->disk_sb, members_v2, u64s); - if (!mi) { - ret = -BCH_ERR_ENOSPC_sb_members; + ret = bch2_sb_member_alloc(c); + if (ret < 0) { bch_err_msg(c, ret, "setting up new superblock"); goto err_unlock; } - struct bch_member *m = bch2_members_v2_get_mut(c->disk_sb.sb, dev_idx); + unsigned dev_idx = ret; /* success: */ - *m = dev_mi; - m->last_mount = cpu_to_le64(ktime_get_real_seconds()); - c->disk_sb.sb->nr_devices = nr_devices; + dev_mi.last_mount = cpu_to_le64(ktime_get_real_seconds()); + *bch2_members_v2_get_mut(c->disk_sb.sb, dev_idx) = dev_mi; ca->disk_sb.sb->dev_idx = dev_idx; bch2_dev_attach(c, ca, dev_idx); diff --git a/fs/bcachefs/sysfs.c b/fs/bcachefs/sysfs.c index 33f2a64c14c9..03e59f86f360 100644 --- a/fs/bcachefs/sysfs.c +++ b/fs/bcachefs/sysfs.c @@ -219,7 +219,6 @@ read_attribute(copy_gc_wait); rw_attribute(rebalance_enabled); sysfs_pd_controller_attribute(rebalance); read_attribute(rebalance_status); -rw_attribute(promote_whole_extents); read_attribute(new_stripes); @@ -234,7 +233,7 @@ write_attribute(perf_test); #define x(_name) \ static struct attribute sysfs_time_stat_##_name = \ - { .name = #_name, .mode = 0444 }; + { .name = #_name, .mode = 0644 }; BCH_TIME_STATS() #undef x @@ -245,14 +244,18 @@ static struct attribute sysfs_state_rw = { static size_t bch2_btree_cache_size(struct bch_fs *c) { + struct btree_cache *bc = &c->btree_cache; size_t ret = 0; struct btree *b; - mutex_lock(&c->btree_cache.lock); - list_for_each_entry(b, &c->btree_cache.live, list) + mutex_lock(&bc->lock); + list_for_each_entry(b, &bc->live[0].list, list) ret += btree_buf_bytes(b); - - mutex_unlock(&c->btree_cache.lock); + list_for_each_entry(b, &bc->live[1].list, list) + ret += btree_buf_bytes(b); + list_for_each_entry(b, &bc->freeable, list) + ret += btree_buf_bytes(b); + mutex_unlock(&bc->lock); return ret; } @@ -288,7 +291,7 @@ static int bch2_compression_stats_to_text(struct printbuf *out, struct bch_fs *c prt_tab_rjust(out); prt_human_readable_u64(out, nr_extents - ? div_u64(sectors_uncompressed << 9, nr_extents) + ? div64_u64(sectors_uncompressed << 9, nr_extents) : 0); prt_tab_rjust(out); prt_newline(out); @@ -347,8 +350,6 @@ SHOW(bch2_fs) if (attr == &sysfs_rebalance_status) bch2_rebalance_status_to_text(out, c); - sysfs_print(promote_whole_extents, c->promote_whole_extents); - /* Debugging: */ if (attr == &sysfs_journal_debug) @@ -436,8 +437,6 @@ STORE(bch2_fs) sysfs_pd_controller_store(rebalance, &c->rebalance.pd); - sysfs_strtoul(promote_whole_extents, c->promote_whole_extents); - /* Debugging: */ if (!test_bit(BCH_FS_started, &c->flags)) @@ -449,11 +448,12 @@ STORE(bch2_fs) return -EROFS; if (attr == &sysfs_trigger_btree_cache_shrink) { + struct btree_cache *bc = &c->btree_cache; struct shrink_control sc; sc.gfp_mask = GFP_KERNEL; sc.nr_to_scan = strtoul_or_return(buf); - c->btree_cache.shrink->scan_objects(c->btree_cache.shrink, &sc); + bc->live[0].shrink->scan_objects(bc->live[0].shrink, &sc); } if (attr == &sysfs_trigger_btree_key_cache_shrink) { @@ -514,7 +514,7 @@ struct attribute *bch2_fs_files[] = { &sysfs_btree_cache_size, &sysfs_btree_write_stats, - &sysfs_promote_whole_extents, + &sysfs_rebalance_status, &sysfs_compression_stats, @@ -614,7 +614,6 @@ struct attribute *bch2_fs_internal_files[] = { &sysfs_copy_gc_wait, &sysfs_rebalance_enabled, - &sysfs_rebalance_status, sysfs_pd_controller_files(rebalance), &sysfs_moving_ctxts, @@ -674,7 +673,7 @@ STORE(bch2_fs_opts_dir) if (ret < 0) goto err; - bch2_opt_set_sb(c, opt, v); + bch2_opt_set_sb(c, NULL, opt, v); bch2_opt_set_by_id(&c->opts, id, v); if (v && @@ -728,6 +727,13 @@ SHOW(bch2_fs_time_stats) STORE(bch2_fs_time_stats) { + struct bch_fs *c = container_of(kobj, struct bch_fs, time_stats); + +#define x(name) \ + if (attr == &sysfs_time_stat_##name) \ + bch2_time_stats_reset(&c->times[BCH_TIME_##name]); + BCH_TIME_STATS() +#undef x return size; } SYSFS_OPS(bch2_fs_time_stats); @@ -821,32 +827,17 @@ STORE(bch2_dev) { struct bch_dev *ca = container_of(kobj, struct bch_dev, kobj); struct bch_fs *c = ca->fs; - struct bch_member *mi; if (attr == &sysfs_discard) { bool v = strtoul_or_return(buf); - mutex_lock(&c->sb_lock); - mi = bch2_members_v2_get_mut(c->disk_sb.sb, ca->dev_idx); - - if (v != BCH_MEMBER_DISCARD(mi)) { - SET_BCH_MEMBER_DISCARD(mi, v); - bch2_write_super(c); - } - mutex_unlock(&c->sb_lock); + bch2_opt_set_sb(c, ca, bch2_opt_table + Opt_discard, v); } if (attr == &sysfs_durability) { u64 v = strtoul_or_return(buf); - mutex_lock(&c->sb_lock); - mi = bch2_members_v2_get_mut(c->disk_sb.sb, ca->dev_idx); - - if (v + 1 != BCH_MEMBER_DURABILITY(mi)) { - SET_BCH_MEMBER_DURABILITY(mi, v + 1); - bch2_write_super(c); - } - mutex_unlock(&c->sb_lock); + bch2_opt_set_sb(c, ca, bch2_opt_table + Opt_durability, v); } if (attr == &sysfs_label) { diff --git a/fs/bcachefs/thread_with_file.c b/fs/bcachefs/thread_with_file.c index 0807ce9b171a..fb3442a7c67f 100644 --- a/fs/bcachefs/thread_with_file.c +++ b/fs/bcachefs/thread_with_file.c @@ -387,7 +387,7 @@ again: seen = buf->buf.nr; char *n = memchr(buf->buf.data, '\n', seen); - if (!n && timeout != MAX_SCHEDULE_TIMEOUT && jiffies >= until) { + if (!n && timeout != MAX_SCHEDULE_TIMEOUT && time_after_eq(jiffies, until)) { spin_unlock(&buf->lock); return -ETIME; } diff --git a/fs/bcachefs/time_stats.c b/fs/bcachefs/time_stats.c index 4508e9dcbee2..3fe82757f93a 100644 --- a/fs/bcachefs/time_stats.c +++ b/fs/bcachefs/time_stats.c @@ -151,6 +151,20 @@ void __bch2_time_stats_update(struct bch2_time_stats *stats, u64 start, u64 end) } } +void bch2_time_stats_reset(struct bch2_time_stats *stats) +{ + spin_lock_irq(&stats->lock); + unsigned offset = offsetof(struct bch2_time_stats, min_duration); + memset((void *) stats + offset, 0, sizeof(*stats) - offset); + + if (stats->buffer) { + int cpu; + for_each_possible_cpu(cpu) + per_cpu_ptr(stats->buffer, cpu)->nr = 0; + } + spin_unlock_irq(&stats->lock); +} + void bch2_time_stats_exit(struct bch2_time_stats *stats) { free_percpu(stats->buffer); diff --git a/fs/bcachefs/time_stats.h b/fs/bcachefs/time_stats.h index 5df61403744b..dc6493f7bbab 100644 --- a/fs/bcachefs/time_stats.h +++ b/fs/bcachefs/time_stats.h @@ -70,6 +70,7 @@ struct time_stat_buffer { struct bch2_time_stats { spinlock_t lock; bool have_quantiles; + struct time_stat_buffer __percpu *buffer; /* all fields are in nanoseconds */ u64 min_duration; u64 max_duration; @@ -87,7 +88,6 @@ struct bch2_time_stats { struct mean_and_variance_weighted duration_stats_weighted; struct mean_and_variance_weighted freq_stats_weighted; - struct time_stat_buffer __percpu *buffer; }; struct bch2_time_stats_quantiles { @@ -142,6 +142,7 @@ static inline bool track_event_change(struct bch2_time_stats *stats, bool v) return false; } +void bch2_time_stats_reset(struct bch2_time_stats *); void bch2_time_stats_exit(struct bch2_time_stats *); void bch2_time_stats_init(struct bch2_time_stats *); diff --git a/fs/bcachefs/trace.h b/fs/bcachefs/trace.h index c62f00322d1e..5597b9d6297f 100644 --- a/fs/bcachefs/trace.h +++ b/fs/bcachefs/trace.h @@ -3,7 +3,6 @@ #define TRACE_SYSTEM bcachefs #if !defined(_TRACE_BCACHEFS_H) || defined(TRACE_HEADER_MULTI_READ) -#define _TRACE_BCACHEFS_H #include @@ -558,6 +557,7 @@ TRACE_EVENT(btree_path_relock_fail, __field(unsigned long, caller_ip ) __field(u8, btree_id ) __field(u8, level ) + __field(u8, path_idx) TRACE_BPOS_entries(pos) __array(char, node, 24 ) __field(u8, self_read_count ) @@ -575,7 +575,8 @@ TRACE_EVENT(btree_path_relock_fail, strscpy(__entry->trans_fn, trans->fn, sizeof(__entry->trans_fn)); __entry->caller_ip = caller_ip; __entry->btree_id = path->btree_id; - __entry->level = path->level; + __entry->level = level; + __entry->path_idx = path - trans->paths; TRACE_BPOS_assign(pos, path->pos); c = bch2_btree_node_lock_counts(trans, NULL, &path->l[level].b->c, level); @@ -588,7 +589,7 @@ TRACE_EVENT(btree_path_relock_fail, c = six_lock_counts(&path->l[level].b->c.lock); __entry->read_count = c.n[SIX_LOCK_read]; __entry->intent_count = c.n[SIX_LOCK_intent]; - scnprintf(__entry->node, sizeof(__entry->node), "%px", b); + scnprintf(__entry->node, sizeof(__entry->node), "%px", &b->c); } __entry->iter_lock_seq = path->l[level].lock_seq; __entry->node_lock_seq = is_btree_node(path, level) @@ -596,9 +597,10 @@ TRACE_EVENT(btree_path_relock_fail, : 0; ), - TP_printk("%s %pS btree %s pos %llu:%llu:%u level %u node %s held %u:%u lock count %u:%u iter seq %u lock seq %u", + TP_printk("%s %pS\nidx %2u btree %s pos %llu:%llu:%u level %u node %s held %u:%u lock count %u:%u iter seq %u lock seq %u", __entry->trans_fn, (void *) __entry->caller_ip, + __entry->path_idx, bch2_btree_id_str(__entry->btree_id), __entry->pos_inode, __entry->pos_offset, @@ -625,6 +627,7 @@ TRACE_EVENT(btree_path_upgrade_fail, __field(unsigned long, caller_ip ) __field(u8, btree_id ) __field(u8, level ) + __field(u8, path_idx) TRACE_BPOS_entries(pos) __field(u8, locked ) __field(u8, self_read_count ) @@ -642,6 +645,7 @@ TRACE_EVENT(btree_path_upgrade_fail, __entry->caller_ip = caller_ip; __entry->btree_id = path->btree_id; __entry->level = level; + __entry->path_idx = path - trans->paths; TRACE_BPOS_assign(pos, path->pos); __entry->locked = btree_node_locked(path, level); @@ -657,9 +661,10 @@ TRACE_EVENT(btree_path_upgrade_fail, : 0; ), - TP_printk("%s %pS btree %s pos %llu:%llu:%u level %u locked %u held %u:%u lock count %u:%u iter seq %u lock seq %u", + TP_printk("%s %pS\nidx %2u btree %s pos %llu:%llu:%u level %u locked %u held %u:%u lock count %u:%u iter seq %u lock seq %u", __entry->trans_fn, (void *) __entry->caller_ip, + __entry->path_idx, bch2_btree_id_str(__entry->btree_id), __entry->pos_inode, __entry->pos_offset, @@ -1438,6 +1443,456 @@ TRACE_EVENT(error_downcast, TP_printk("%s -> %s %s", __entry->bch_err, __entry->std_err, __entry->ip) ); +#ifdef CONFIG_BCACHEFS_PATH_TRACEPOINTS + +TRACE_EVENT(update_by_path, + TP_PROTO(struct btree_trans *trans, struct btree_path *path, + struct btree_insert_entry *i, bool overwrite), + TP_ARGS(trans, path, i, overwrite), + + TP_STRUCT__entry( + __array(char, trans_fn, 32 ) + __field(btree_path_idx_t, path_idx ) + __field(u8, btree_id ) + TRACE_BPOS_entries(pos) + __field(u8, overwrite ) + __field(btree_path_idx_t, update_idx ) + __field(btree_path_idx_t, nr_updates ) + ), + + TP_fast_assign( + strscpy(__entry->trans_fn, trans->fn, sizeof(__entry->trans_fn)); + __entry->path_idx = path - trans->paths; + __entry->btree_id = path->btree_id; + TRACE_BPOS_assign(pos, path->pos); + __entry->overwrite = overwrite; + __entry->update_idx = i - trans->updates; + __entry->nr_updates = trans->nr_updates; + ), + + TP_printk("%s path %3u btree %s pos %llu:%llu:%u overwrite %u update %u/%u", + __entry->trans_fn, + __entry->path_idx, + bch2_btree_id_str(__entry->btree_id), + __entry->pos_inode, + __entry->pos_offset, + __entry->pos_snapshot, + __entry->overwrite, + __entry->update_idx, + __entry->nr_updates) +); + +TRACE_EVENT(btree_path_lock, + TP_PROTO(struct btree_trans *trans, + unsigned long caller_ip, + struct btree_bkey_cached_common *b), + TP_ARGS(trans, caller_ip, b), + + TP_STRUCT__entry( + __array(char, trans_fn, 32 ) + __field(unsigned long, caller_ip ) + __field(u8, btree_id ) + __field(u8, level ) + __array(char, node, 24 ) + __field(u32, lock_seq ) + ), + + TP_fast_assign( + strscpy(__entry->trans_fn, trans->fn, sizeof(__entry->trans_fn)); + __entry->caller_ip = caller_ip; + __entry->btree_id = b->btree_id; + __entry->level = b->level; + + scnprintf(__entry->node, sizeof(__entry->node), "%px", b); + __entry->lock_seq = six_lock_seq(&b->lock); + ), + + TP_printk("%s %pS\nbtree %s level %u node %s lock seq %u", + __entry->trans_fn, + (void *) __entry->caller_ip, + bch2_btree_id_str(__entry->btree_id), + __entry->level, + __entry->node, + __entry->lock_seq) +); + +DECLARE_EVENT_CLASS(btree_path_ev, + TP_PROTO(struct btree_trans *trans, struct btree_path *path), + TP_ARGS(trans, path), + + TP_STRUCT__entry( + __field(u16, idx ) + __field(u8, ref ) + __field(u8, btree_id ) + TRACE_BPOS_entries(pos) + ), + + TP_fast_assign( + __entry->idx = path - trans->paths; + __entry->ref = path->ref; + __entry->btree_id = path->btree_id; + TRACE_BPOS_assign(pos, path->pos); + ), + + TP_printk("path %3u ref %u btree %s pos %llu:%llu:%u", + __entry->idx, __entry->ref, + bch2_btree_id_str(__entry->btree_id), + __entry->pos_inode, + __entry->pos_offset, + __entry->pos_snapshot) +); + +DEFINE_EVENT(btree_path_ev, btree_path_get_ll, + TP_PROTO(struct btree_trans *trans, struct btree_path *path), + TP_ARGS(trans, path) +); + +DEFINE_EVENT(btree_path_ev, btree_path_put_ll, + TP_PROTO(struct btree_trans *trans, struct btree_path *path), + TP_ARGS(trans, path) +); + +DEFINE_EVENT(btree_path_ev, btree_path_should_be_locked, + TP_PROTO(struct btree_trans *trans, struct btree_path *path), + TP_ARGS(trans, path) +); + +TRACE_EVENT(btree_path_alloc, + TP_PROTO(struct btree_trans *trans, struct btree_path *path), + TP_ARGS(trans, path), + + TP_STRUCT__entry( + __field(btree_path_idx_t, idx ) + __field(u8, locks_want ) + __field(u8, btree_id ) + TRACE_BPOS_entries(pos) + ), + + TP_fast_assign( + __entry->idx = path - trans->paths; + __entry->locks_want = path->locks_want; + __entry->btree_id = path->btree_id; + TRACE_BPOS_assign(pos, path->pos); + ), + + TP_printk("path %3u btree %s locks_want %u pos %llu:%llu:%u", + __entry->idx, + bch2_btree_id_str(__entry->btree_id), + __entry->locks_want, + __entry->pos_inode, + __entry->pos_offset, + __entry->pos_snapshot) +); + +TRACE_EVENT(btree_path_get, + TP_PROTO(struct btree_trans *trans, struct btree_path *path, struct bpos *new_pos), + TP_ARGS(trans, path, new_pos), + + TP_STRUCT__entry( + __field(btree_path_idx_t, idx ) + __field(u8, ref ) + __field(u8, preserve ) + __field(u8, locks_want ) + __field(u8, btree_id ) + TRACE_BPOS_entries(old_pos) + TRACE_BPOS_entries(new_pos) + ), + + TP_fast_assign( + __entry->idx = path - trans->paths; + __entry->ref = path->ref; + __entry->preserve = path->preserve; + __entry->locks_want = path->locks_want; + __entry->btree_id = path->btree_id; + TRACE_BPOS_assign(old_pos, path->pos); + TRACE_BPOS_assign(new_pos, *new_pos); + ), + + TP_printk(" path %3u ref %u preserve %u btree %s locks_want %u pos %llu:%llu:%u -> %llu:%llu:%u", + __entry->idx, + __entry->ref, + __entry->preserve, + bch2_btree_id_str(__entry->btree_id), + __entry->locks_want, + __entry->old_pos_inode, + __entry->old_pos_offset, + __entry->old_pos_snapshot, + __entry->new_pos_inode, + __entry->new_pos_offset, + __entry->new_pos_snapshot) +); + +DECLARE_EVENT_CLASS(btree_path_clone, + TP_PROTO(struct btree_trans *trans, struct btree_path *path, struct btree_path *new), + TP_ARGS(trans, path, new), + + TP_STRUCT__entry( + __field(btree_path_idx_t, idx ) + __field(u8, new_idx ) + __field(u8, btree_id ) + __field(u8, ref ) + __field(u8, preserve ) + TRACE_BPOS_entries(pos) + ), + + TP_fast_assign( + __entry->idx = path - trans->paths; + __entry->new_idx = new - trans->paths; + __entry->btree_id = path->btree_id; + __entry->ref = path->ref; + __entry->preserve = path->preserve; + TRACE_BPOS_assign(pos, path->pos); + ), + + TP_printk(" path %3u ref %u preserve %u btree %s %llu:%llu:%u -> %u", + __entry->idx, + __entry->ref, + __entry->preserve, + bch2_btree_id_str(__entry->btree_id), + __entry->pos_inode, + __entry->pos_offset, + __entry->pos_snapshot, + __entry->new_idx) +); + +DEFINE_EVENT(btree_path_clone, btree_path_clone, + TP_PROTO(struct btree_trans *trans, struct btree_path *path, struct btree_path *new), + TP_ARGS(trans, path, new) +); + +DEFINE_EVENT(btree_path_clone, btree_path_save_pos, + TP_PROTO(struct btree_trans *trans, struct btree_path *path, struct btree_path *new), + TP_ARGS(trans, path, new) +); + +DECLARE_EVENT_CLASS(btree_path_traverse, + TP_PROTO(struct btree_trans *trans, + struct btree_path *path), + TP_ARGS(trans, path), + + TP_STRUCT__entry( + __array(char, trans_fn, 32 ) + __field(btree_path_idx_t, idx ) + __field(u8, ref ) + __field(u8, preserve ) + __field(u8, should_be_locked ) + __field(u8, btree_id ) + __field(u8, level ) + TRACE_BPOS_entries(pos) + __field(u8, locks_want ) + __field(u8, nodes_locked ) + __array(char, node0, 24 ) + __array(char, node1, 24 ) + __array(char, node2, 24 ) + __array(char, node3, 24 ) + ), + + TP_fast_assign( + strscpy(__entry->trans_fn, trans->fn, sizeof(__entry->trans_fn)); + + __entry->idx = path - trans->paths; + __entry->ref = path->ref; + __entry->preserve = path->preserve; + __entry->btree_id = path->btree_id; + __entry->level = path->level; + TRACE_BPOS_assign(pos, path->pos); + + __entry->locks_want = path->locks_want; + __entry->nodes_locked = path->nodes_locked; + struct btree *b = path->l[0].b; + if (IS_ERR(b)) + strscpy(__entry->node0, bch2_err_str(PTR_ERR(b)), sizeof(__entry->node0)); + else + scnprintf(__entry->node0, sizeof(__entry->node0), "%px", &b->c); + b = path->l[1].b; + if (IS_ERR(b)) + strscpy(__entry->node1, bch2_err_str(PTR_ERR(b)), sizeof(__entry->node0)); + else + scnprintf(__entry->node1, sizeof(__entry->node0), "%px", &b->c); + b = path->l[2].b; + if (IS_ERR(b)) + strscpy(__entry->node2, bch2_err_str(PTR_ERR(b)), sizeof(__entry->node0)); + else + scnprintf(__entry->node2, sizeof(__entry->node0), "%px", &b->c); + b = path->l[3].b; + if (IS_ERR(b)) + strscpy(__entry->node3, bch2_err_str(PTR_ERR(b)), sizeof(__entry->node0)); + else + scnprintf(__entry->node3, sizeof(__entry->node0), "%px", &b->c); + ), + + TP_printk("%s\npath %3u ref %u preserve %u btree %s %llu:%llu:%u level %u locks_want %u\n" + "locks %u %u %u %u node %s %s %s %s", + __entry->trans_fn, + __entry->idx, + __entry->ref, + __entry->preserve, + bch2_btree_id_str(__entry->btree_id), + __entry->pos_inode, + __entry->pos_offset, + __entry->pos_snapshot, + __entry->level, + __entry->locks_want, + (__entry->nodes_locked >> 6) & 3, + (__entry->nodes_locked >> 4) & 3, + (__entry->nodes_locked >> 2) & 3, + (__entry->nodes_locked >> 0) & 3, + __entry->node3, + __entry->node2, + __entry->node1, + __entry->node0) +); + +DEFINE_EVENT(btree_path_traverse, btree_path_traverse_start, + TP_PROTO(struct btree_trans *trans, + struct btree_path *path), + TP_ARGS(trans, path) +); + +DEFINE_EVENT(btree_path_traverse, btree_path_traverse_end, + TP_PROTO(struct btree_trans *trans, struct btree_path *path), + TP_ARGS(trans, path) +); + +TRACE_EVENT(btree_path_set_pos, + TP_PROTO(struct btree_trans *trans, + struct btree_path *path, + struct bpos *new_pos), + TP_ARGS(trans, path, new_pos), + + TP_STRUCT__entry( + __field(btree_path_idx_t, idx ) + __field(u8, ref ) + __field(u8, preserve ) + __field(u8, btree_id ) + TRACE_BPOS_entries(old_pos) + TRACE_BPOS_entries(new_pos) + __field(u8, locks_want ) + __field(u8, nodes_locked ) + __array(char, node0, 24 ) + __array(char, node1, 24 ) + __array(char, node2, 24 ) + __array(char, node3, 24 ) + ), + + TP_fast_assign( + __entry->idx = path - trans->paths; + __entry->ref = path->ref; + __entry->preserve = path->preserve; + __entry->btree_id = path->btree_id; + TRACE_BPOS_assign(old_pos, path->pos); + TRACE_BPOS_assign(new_pos, *new_pos); + + __entry->nodes_locked = path->nodes_locked; + struct btree *b = path->l[0].b; + if (IS_ERR(b)) + strscpy(__entry->node0, bch2_err_str(PTR_ERR(b)), sizeof(__entry->node0)); + else + scnprintf(__entry->node0, sizeof(__entry->node0), "%px", &b->c); + b = path->l[1].b; + if (IS_ERR(b)) + strscpy(__entry->node1, bch2_err_str(PTR_ERR(b)), sizeof(__entry->node0)); + else + scnprintf(__entry->node1, sizeof(__entry->node0), "%px", &b->c); + b = path->l[2].b; + if (IS_ERR(b)) + strscpy(__entry->node2, bch2_err_str(PTR_ERR(b)), sizeof(__entry->node0)); + else + scnprintf(__entry->node2, sizeof(__entry->node0), "%px", &b->c); + b = path->l[3].b; + if (IS_ERR(b)) + strscpy(__entry->node3, bch2_err_str(PTR_ERR(b)), sizeof(__entry->node0)); + else + scnprintf(__entry->node3, sizeof(__entry->node0), "%px", &b->c); + ), + + TP_printk("\npath %3u ref %u preserve %u btree %s %llu:%llu:%u -> %llu:%llu:%u\n" + "locks %u %u %u %u node %s %s %s %s", + __entry->idx, + __entry->ref, + __entry->preserve, + bch2_btree_id_str(__entry->btree_id), + __entry->old_pos_inode, + __entry->old_pos_offset, + __entry->old_pos_snapshot, + __entry->new_pos_inode, + __entry->new_pos_offset, + __entry->new_pos_snapshot, + (__entry->nodes_locked >> 6) & 3, + (__entry->nodes_locked >> 4) & 3, + (__entry->nodes_locked >> 2) & 3, + (__entry->nodes_locked >> 0) & 3, + __entry->node3, + __entry->node2, + __entry->node1, + __entry->node0) +); + +TRACE_EVENT(btree_path_free, + TP_PROTO(struct btree_trans *trans, btree_path_idx_t path, struct btree_path *dup), + TP_ARGS(trans, path, dup), + + TP_STRUCT__entry( + __field(btree_path_idx_t, idx ) + __field(u8, preserve ) + __field(u8, should_be_locked) + __field(s8, dup ) + __field(u8, dup_locked ) + ), + + TP_fast_assign( + __entry->idx = path; + __entry->preserve = trans->paths[path].preserve; + __entry->should_be_locked = trans->paths[path].should_be_locked; + __entry->dup = dup ? dup - trans->paths : -1; + __entry->dup_locked = dup ? btree_node_locked(dup, dup->level) : 0; + ), + + TP_printk(" path %3u %c %c dup %2i locked %u", __entry->idx, + __entry->preserve ? 'P' : ' ', + __entry->should_be_locked ? 'S' : ' ', + __entry->dup, + __entry->dup_locked) +); + +TRACE_EVENT(btree_path_free_trans_begin, + TP_PROTO(btree_path_idx_t path), + TP_ARGS(path), + + TP_STRUCT__entry( + __field(btree_path_idx_t, idx ) + ), + + TP_fast_assign( + __entry->idx = path; + ), + + TP_printk(" path %3u", __entry->idx) +); + +#else /* CONFIG_BCACHEFS_PATH_TRACEPOINTS */ +#ifndef _TRACE_BCACHEFS_H + +static inline void trace_update_by_path(struct btree_trans *trans, struct btree_path *path, + struct btree_insert_entry *i, bool overwrite) {} +static inline void trace_btree_path_lock(struct btree_trans *trans, unsigned long caller_ip, struct btree_bkey_cached_common *b) {} +static inline void trace_btree_path_get_ll(struct btree_trans *trans, struct btree_path *path) {} +static inline void trace_btree_path_put_ll(struct btree_trans *trans, struct btree_path *path) {} +static inline void trace_btree_path_should_be_locked(struct btree_trans *trans, struct btree_path *path) {} +static inline void trace_btree_path_alloc(struct btree_trans *trans, struct btree_path *path) {} +static inline void trace_btree_path_get(struct btree_trans *trans, struct btree_path *path, struct bpos *new_pos) {} +static inline void trace_btree_path_clone(struct btree_trans *trans, struct btree_path *path, struct btree_path *new) {} +static inline void trace_btree_path_save_pos(struct btree_trans *trans, struct btree_path *path, struct btree_path *new) {} +static inline void trace_btree_path_traverse_start(struct btree_trans *trans, struct btree_path *path) {} +static inline void trace_btree_path_traverse_end(struct btree_trans *trans, struct btree_path *path) {} +static inline void trace_btree_path_set_pos(struct btree_trans *trans, struct btree_path *path, struct bpos *new_pos) {} +static inline void trace_btree_path_free(struct btree_trans *trans, btree_path_idx_t path, struct btree_path *dup) {} +static inline void trace_btree_path_free_trans_begin(btree_path_idx_t path) {} + +#endif +#endif /* CONFIG_BCACHEFS_PATH_TRACEPOINTS */ + +#define _TRACE_BCACHEFS_H #endif /* _TRACE_BCACHEFS_H */ /* This part must be outside protection */ diff --git a/fs/bcachefs/util.c b/fs/bcachefs/util.c index 1b8554460af4..42f565c76181 100644 --- a/fs/bcachefs/util.c +++ b/fs/bcachefs/util.c @@ -64,7 +64,7 @@ static int bch2_pow(u64 n, u64 p, u64 *res) *res = 1; while (p--) { - if (*res > div_u64(U64_MAX, n)) + if (*res > div64_u64(U64_MAX, n)) return -ERANGE; *res *= n; } @@ -140,14 +140,14 @@ static int __bch2_strtou64_h(const char *cp, u64 *res) parse_or_ret(cp, parse_unit_suffix(cp, &b)); - if (v > div_u64(U64_MAX, b)) + if (v > div64_u64(U64_MAX, b)) return -ERANGE; v *= b; - if (f_n > div_u64(U64_MAX, b)) + if (f_n > div64_u64(U64_MAX, b)) return -ERANGE; - f_n = div_u64(f_n * b, f_d); + f_n = div64_u64(f_n * b, f_d); if (v + f_n < v) return -ERANGE; v += f_n; @@ -204,7 +204,7 @@ STRTO_H(strtoll, long long) STRTO_H(strtoull, unsigned long long) STRTO_H(strtou64, u64) -u64 bch2_read_flag_list(char *opt, const char * const list[]) +u64 bch2_read_flag_list(const char *opt, const char * const list[]) { u64 ret = 0; char *p, *s, *d = kstrdup(opt, GFP_KERNEL); @@ -214,7 +214,7 @@ u64 bch2_read_flag_list(char *opt, const char * const list[]) s = strim(d); - while ((p = strsep(&s, ","))) { + while ((p = strsep(&s, ",;"))) { int flag = match_string(list, -1, p); if (flag < 0) { @@ -360,7 +360,7 @@ void bch2_pr_time_units(struct printbuf *out, u64 ns) { const struct time_unit *u = bch2_pick_time_units(ns); - prt_printf(out, "%llu %s", div_u64(ns, u->nsecs), u->name); + prt_printf(out, "%llu %s", div64_u64(ns, u->nsecs), u->name); } static void bch2_pr_time_units_aligned(struct printbuf *out, u64 ns) @@ -477,7 +477,7 @@ void bch2_time_stats_to_text(struct printbuf *out, struct bch2_time_stats *stats bool is_last = eytzinger0_next(i, NR_QUANTILES) == -1; u64 q = max(quantiles->entries[i].m, last_q); - prt_printf(out, "%llu ", div_u64(q, u->nsecs)); + prt_printf(out, "%llu ", div64_u64(q, u->nsecs)); if (is_last) prt_newline(out); last_q = q; diff --git a/fs/bcachefs/util.h b/fs/bcachefs/util.h index 902b7f5406a2..fb02c1c36004 100644 --- a/fs/bcachefs/util.h +++ b/fs/bcachefs/util.h @@ -195,7 +195,7 @@ static inline int bch2_strtoul_h(const char *cp, long *res) bool bch2_is_zero(const void *, size_t); -u64 bch2_read_flag_list(char *, const char * const[]); +u64 bch2_read_flag_list(const char *, const char * const[]); void bch2_prt_u64_base2_nbits(struct printbuf *, u64, unsigned); void bch2_prt_u64_base2(struct printbuf *, u64); diff --git a/fs/bcachefs/xattr.c b/fs/bcachefs/xattr.c index 331f944d73dc..56c8d3fe55a4 100644 --- a/fs/bcachefs/xattr.c +++ b/fs/bcachefs/xattr.c @@ -250,17 +250,27 @@ static int __bch2_xattr_emit(const char *prefix, return 0; } +static inline const char *bch2_xattr_prefix(unsigned type, struct dentry *dentry) +{ + const struct xattr_handler *handler = bch2_xattr_type_to_handler(type); + + if (!xattr_handler_can_list(handler, dentry)) + return NULL; + + return xattr_prefix(handler); +} + static int bch2_xattr_emit(struct dentry *dentry, const struct bch_xattr *xattr, struct xattr_buf *buf) { - const struct xattr_handler *handler = - bch2_xattr_type_to_handler(xattr->x_type); + const char *prefix; - return handler && (!handler->list || handler->list(dentry)) - ? __bch2_xattr_emit(handler->prefix ?: handler->name, - xattr->x_name, xattr->x_name_len, buf) - : 0; + prefix = bch2_xattr_prefix(xattr->x_type, dentry); + if (!prefix) + return 0; + + return __bch2_xattr_emit(prefix, xattr->x_name, xattr->x_name_len, buf); } static int bch2_xattr_list_bcachefs(struct bch_fs *c, @@ -295,54 +305,23 @@ ssize_t bch2_xattr_list(struct dentry *dentry, char *buffer, size_t buffer_size) { struct bch_fs *c = dentry->d_sb->s_fs_info; struct bch_inode_info *inode = to_bch_ei(dentry->d_inode); - struct btree_trans *trans = bch2_trans_get(c); - struct btree_iter iter; - struct bkey_s_c k; struct xattr_buf buf = { .buf = buffer, .len = buffer_size }; u64 offset = 0, inum = inode->ei_inode.bi_inum; - u32 snapshot; - int ret; -retry: - bch2_trans_begin(trans); - iter = (struct btree_iter) { NULL }; - ret = bch2_subvolume_get_snapshot(trans, inode->ei_subvol, &snapshot); - if (ret) - goto err; + int ret = bch2_trans_run(c, + for_each_btree_key_in_subvolume_upto(trans, iter, BTREE_ID_xattrs, + POS(inum, offset), + POS(inum, U64_MAX), + inode->ei_inum.subvol, 0, k, ({ + if (k.k->type != KEY_TYPE_xattr) + continue; - for_each_btree_key_upto_norestart(trans, iter, BTREE_ID_xattrs, - SPOS(inum, offset, snapshot), - POS(inum, U64_MAX), 0, k, ret) { - if (k.k->type != KEY_TYPE_xattr) - continue; + bch2_xattr_emit(dentry, bkey_s_c_to_xattr(k).v, &buf); + }))) ?: + bch2_xattr_list_bcachefs(c, &inode->ei_inode, &buf, false) ?: + bch2_xattr_list_bcachefs(c, &inode->ei_inode, &buf, true); - ret = bch2_xattr_emit(dentry, bkey_s_c_to_xattr(k).v, &buf); - if (ret) - break; - } - - offset = iter.pos.offset; - bch2_trans_iter_exit(trans, &iter); -err: - if (bch2_err_matches(ret, BCH_ERR_transaction_restart)) - goto retry; - - bch2_trans_put(trans); - - if (ret) - goto out; - - ret = bch2_xattr_list_bcachefs(c, &inode->ei_inode, &buf, false); - if (ret) - goto out; - - ret = bch2_xattr_list_bcachefs(c, &inode->ei_inode, &buf, true); - if (ret) - goto out; - - return buf.used; -out: - return bch2_err_class(ret); + return ret ? bch2_err_class(ret) : buf.used; } static int bch2_xattr_get_handler(const struct xattr_handler *handler, @@ -632,10 +611,6 @@ static const struct xattr_handler bch_xattr_bcachefs_effective_handler = { const struct xattr_handler *bch2_xattr_handlers[] = { &bch_xattr_user_handler, -#ifdef CONFIG_BCACHEFS_POSIX_ACL - &nop_posix_acl_access, - &nop_posix_acl_default, -#endif &bch_xattr_trusted_handler, &bch_xattr_security_handler, #ifndef NO_BCACHEFS_FS diff --git a/fs/bcachefs/xattr_format.h b/fs/bcachefs/xattr_format.h index e9f810539552..c7916011ef34 100644 --- a/fs/bcachefs/xattr_format.h +++ b/fs/bcachefs/xattr_format.h @@ -13,7 +13,7 @@ struct bch_xattr { __u8 x_type; __u8 x_name_len; __le16 x_val_len; - __u8 x_name[]; + __u8 x_name[] __counted_by(x_name_len); } __packed __aligned(8); #endif /* _BCACHEFS_XATTR_FORMAT_H */ diff --git a/fs/btrfs/btrfs_inode.h b/fs/btrfs/btrfs_inode.h index 9a4b7c119318..e152fde888fc 100644 --- a/fs/btrfs/btrfs_inode.h +++ b/fs/btrfs/btrfs_inode.h @@ -152,6 +152,7 @@ struct btrfs_inode { * logged_trans), to access/update delalloc_bytes, new_delalloc_bytes, * defrag_bytes, disk_i_size, outstanding_extents, csum_bytes and to * update the VFS' inode number of bytes used. + * Also protects setting struct file::private_data. */ spinlock_t lock; diff --git a/fs/btrfs/ctree.h b/fs/btrfs/ctree.h index 1a44fb9845e3..317a3712270f 100644 --- a/fs/btrfs/ctree.h +++ b/fs/btrfs/ctree.h @@ -463,6 +463,8 @@ struct btrfs_file_private { void *filldir_buf; u64 last_index; struct extent_state *llseek_cached_state; + /* Task that allocated this structure. */ + struct task_struct *owner_task; }; static inline u32 BTRFS_LEAF_DATA_SIZE(const struct btrfs_fs_info *info) diff --git a/fs/btrfs/defrag.c b/fs/btrfs/defrag.c index acf1f39e45d0..b95ef44c326b 100644 --- a/fs/btrfs/defrag.c +++ b/fs/btrfs/defrag.c @@ -213,6 +213,8 @@ void btrfs_cleanup_defrag_inodes(struct btrfs_fs_info *fs_info) &fs_info->defrag_inodes, rb_node) kmem_cache_free(btrfs_inode_defrag_cachep, defrag); + fs_info->defrag_inodes = RB_ROOT; + spin_unlock(&fs_info->defrag_inodes_lock); } diff --git a/fs/btrfs/file.c b/fs/btrfs/file.c index c5e36f58eb07..4fb521d91b06 100644 --- a/fs/btrfs/file.c +++ b/fs/btrfs/file.c @@ -3485,7 +3485,7 @@ static bool find_desired_extent_in_hole(struct btrfs_inode *inode, int whence, static loff_t find_desired_extent(struct file *file, loff_t offset, int whence) { struct btrfs_inode *inode = BTRFS_I(file->f_mapping->host); - struct btrfs_file_private *private = file->private_data; + struct btrfs_file_private *private; struct btrfs_fs_info *fs_info = inode->root->fs_info; struct extent_state *cached_state = NULL; struct extent_state **delalloc_cached_state; @@ -3513,7 +3513,19 @@ static loff_t find_desired_extent(struct file *file, loff_t offset, int whence) inode_get_bytes(&inode->vfs_inode) == i_size) return i_size; - if (!private) { + spin_lock(&inode->lock); + private = file->private_data; + spin_unlock(&inode->lock); + + if (private && private->owner_task != current) { + /* + * Not allocated by us, don't use it as its cached state is used + * by the task that allocated it and we don't want neither to + * mess with it nor get incorrect results because it reflects an + * invalid state for the current task. + */ + private = NULL; + } else if (!private) { private = kzalloc(sizeof(*private), GFP_KERNEL); /* * No worries if memory allocation failed. @@ -3521,7 +3533,23 @@ static loff_t find_desired_extent(struct file *file, loff_t offset, int whence) * lseek SEEK_HOLE/DATA calls to a file when there's delalloc, * so everything will still be correct. */ - file->private_data = private; + if (private) { + bool free = false; + + private->owner_task = current; + + spin_lock(&inode->lock); + if (file->private_data) + free = true; + else + file->private_data = private; + spin_unlock(&inode->lock); + + if (free) { + kfree(private); + private = NULL; + } + } } if (private) diff --git a/fs/btrfs/tree-checker.c b/fs/btrfs/tree-checker.c index 634d69964fe4..7b50263723bc 100644 --- a/fs/btrfs/tree-checker.c +++ b/fs/btrfs/tree-checker.c @@ -1517,7 +1517,7 @@ static int check_extent_item(struct extent_buffer *leaf, dref_objectid > BTRFS_LAST_FREE_OBJECTID)) { extent_err(leaf, slot, "invalid data ref objectid value %llu", - dref_root); + dref_objectid); return -EUCLEAN; } if (unlikely(!IS_ALIGNED(dref_offset, diff --git a/fs/gfs2/aops.c b/fs/gfs2/aops.c index 10d5acd3f742..68fc8af14700 100644 --- a/fs/gfs2/aops.c +++ b/fs/gfs2/aops.c @@ -138,35 +138,6 @@ static int __gfs2_jdata_write_folio(struct folio *folio, return gfs2_write_jdata_folio(folio, wbc); } -/** - * gfs2_jdata_writepage - Write complete page - * @page: Page to write - * @wbc: The writeback control - * - * Returns: errno - * - */ - -static int gfs2_jdata_writepage(struct page *page, struct writeback_control *wbc) -{ - struct folio *folio = page_folio(page); - struct inode *inode = page->mapping->host; - struct gfs2_inode *ip = GFS2_I(inode); - struct gfs2_sbd *sdp = GFS2_SB(inode); - - if (gfs2_assert_withdraw(sdp, ip->i_gl->gl_state == LM_ST_EXCLUSIVE)) - goto out; - if (folio_test_checked(folio) || current->journal_info) - goto out_ignore; - return __gfs2_jdata_write_folio(folio, wbc); - -out_ignore: - folio_redirty_for_writepage(wbc, folio); -out: - folio_unlock(folio); - return 0; -} - /** * gfs2_writepages - Write a bunch of dirty pages back to disk * @mapping: The mapping to write @@ -748,7 +719,6 @@ static const struct address_space_operations gfs2_aops = { }; static const struct address_space_operations gfs2_jdata_aops = { - .writepage = gfs2_jdata_writepage, .writepages = gfs2_jdata_writepages, .read_folio = gfs2_read_folio, .readahead = gfs2_readahead, diff --git a/fs/gfs2/glock.c b/fs/gfs2/glock.c index 12a769077ea0..269c3bc7fced 100644 --- a/fs/gfs2/glock.c +++ b/fs/gfs2/glock.c @@ -1885,14 +1885,16 @@ void gfs2_glock_dq_m(unsigned int num_gh, struct gfs2_holder *ghs) void gfs2_glock_cb(struct gfs2_glock *gl, unsigned int state) { unsigned long delay = 0; - unsigned long holdtime; - unsigned long now = jiffies; gfs2_glock_hold(gl); spin_lock(&gl->gl_lockref.lock); - holdtime = gl->gl_tchange + gl->gl_hold_time; if (!list_empty(&gl->gl_holders) && gl->gl_name.ln_type == LM_TYPE_INODE) { + unsigned long now = jiffies; + unsigned long holdtime; + + holdtime = gl->gl_tchange + gl->gl_hold_time; + if (time_before(now, holdtime)) delay = holdtime - now; if (test_bit(GLF_HAVE_REPLY, &gl->gl_flags)) @@ -2249,6 +2251,7 @@ void gfs2_gl_hash_clear(struct gfs2_sbd *sdp) gfs2_free_dead_glocks(sdp); glock_hash_walk(dump_glock_func, sdp); destroy_workqueue(sdp->sd_glock_wq); + sdp->sd_glock_wq = NULL; } static const char *state2str(unsigned state) diff --git a/fs/gfs2/log.c b/fs/gfs2/log.c index 6ee6013fb825..f9c5089783d2 100644 --- a/fs/gfs2/log.c +++ b/fs/gfs2/log.c @@ -80,15 +80,6 @@ void gfs2_remove_from_ail(struct gfs2_bufdata *bd) brelse(bd->bd_bh); } -static int __gfs2_writepage(struct folio *folio, struct writeback_control *wbc, - void *data) -{ - struct address_space *mapping = data; - int ret = mapping->a_ops->writepage(&folio->page, wbc); - mapping_set_error(mapping, ret); - return ret; -} - /** * gfs2_ail1_start_one - Start I/O on a transaction * @sdp: The superblock @@ -140,7 +131,7 @@ __acquires(&sdp->sd_ail_lock) if (!mapping) continue; spin_unlock(&sdp->sd_ail_lock); - ret = write_cache_pages(mapping, wbc, __gfs2_writepage, mapping); + ret = mapping->a_ops->writepages(mapping, wbc); if (need_resched()) { blk_finish_plug(plug); cond_resched(); @@ -149,6 +140,7 @@ __acquires(&sdp->sd_ail_lock) spin_lock(&sdp->sd_ail_lock); if (ret == -ENODATA) /* if a jdata write into a new hole */ ret = 0; /* ignore it */ + mapping_set_error(mapping, ret); if (ret || wbc->nr_to_write <= 0) break; return -EBUSY; diff --git a/fs/gfs2/meta_io.c b/fs/gfs2/meta_io.c index 2b26e8d529aa..fea3efcc2f93 100644 --- a/fs/gfs2/meta_io.c +++ b/fs/gfs2/meta_io.c @@ -30,9 +30,9 @@ #include "util.h" #include "trace_gfs2.h" -static int gfs2_aspace_writepage(struct page *page, struct writeback_control *wbc) +static void gfs2_aspace_write_folio(struct folio *folio, + struct writeback_control *wbc) { - struct folio *folio = page_folio(page); struct buffer_head *bh, *head; int nr_underway = 0; blk_opf_t write_flags = REQ_META | REQ_PRIO | wbc_to_write_flags(wbc); @@ -66,8 +66,8 @@ static int gfs2_aspace_writepage(struct page *page, struct writeback_control *wb } while ((bh = bh->b_this_page) != head); /* - * The page and its buffers are protected by PageWriteback(), so we can - * drop the bh refcounts early. + * The folio and its buffers are protected from truncation by + * the writeback flag, so we can drop the bh refcounts early. */ BUG_ON(folio_test_writeback(folio)); folio_start_writeback(folio); @@ -84,21 +84,31 @@ static int gfs2_aspace_writepage(struct page *page, struct writeback_control *wb if (nr_underway == 0) folio_end_writeback(folio); +} - return 0; +static int gfs2_aspace_writepages(struct address_space *mapping, + struct writeback_control *wbc) +{ + struct folio *folio = NULL; + int error; + + while ((folio = writeback_iter(mapping, wbc, folio, &error))) + gfs2_aspace_write_folio(folio, wbc); + + return error; } const struct address_space_operations gfs2_meta_aops = { .dirty_folio = block_dirty_folio, .invalidate_folio = block_invalidate_folio, - .writepage = gfs2_aspace_writepage, + .writepages = gfs2_aspace_writepages, .release_folio = gfs2_release_folio, }; const struct address_space_operations gfs2_rgrp_aops = { .dirty_folio = block_dirty_folio, .invalidate_folio = block_invalidate_folio, - .writepage = gfs2_aspace_writepage, + .writepages = gfs2_aspace_writepages, .release_folio = gfs2_release_folio, }; diff --git a/fs/gfs2/ops_fstype.c b/fs/gfs2/ops_fstype.c index ff1f3e3dc65c..e83d293c3614 100644 --- a/fs/gfs2/ops_fstype.c +++ b/fs/gfs2/ops_fstype.c @@ -1307,7 +1307,8 @@ fail_debug: fail_delete_wq: destroy_workqueue(sdp->sd_delete_wq); fail_glock_wq: - destroy_workqueue(sdp->sd_glock_wq); + if (sdp->sd_glock_wq) + destroy_workqueue(sdp->sd_glock_wq); fail_free: free_sbd(sdp); sb->s_fs_info = NULL; diff --git a/fs/inode.c b/fs/inode.c index af78f515403f..471ae4a31549 100644 --- a/fs/inode.c +++ b/fs/inode.c @@ -438,14 +438,6 @@ static void init_once(void *foo) inode_init_once(inode); } -/* - * inode->i_lock must be held - */ -void __iget(struct inode *inode) -{ - atomic_inc(&inode->i_count); -} - /* * get additional reference to inode; caller must already hold one. */ diff --git a/fs/isofs/rock.h b/fs/isofs/rock.h index ee9660e9671c..7755e587f778 100644 --- a/fs/isofs/rock.h +++ b/fs/isofs/rock.h @@ -44,7 +44,7 @@ struct RR_PN_s { struct SL_component { __u8 flags; __u8 len; - __u8 text[]; + __u8 text[] __counted_by(len); } __attribute__ ((packed)); struct RR_SL_s { diff --git a/fs/lockd/host.c b/fs/lockd/host.c index c11516801784..5e6877c37f73 100644 --- a/fs/lockd/host.c +++ b/fs/lockd/host.c @@ -440,7 +440,7 @@ nlm_bind_host(struct nlm_host *host) if ((clnt = host->h_rpcclnt) != NULL) { nlm_rebind_host(host); } else { - unsigned long increment = nlmsvc_timeout; + unsigned long increment = nlm_timeout * HZ; struct rpc_timeout timeparms = { .to_initval = increment, .to_increment = increment, diff --git a/fs/lockd/svc.c b/fs/lockd/svc.c index ab8042a5b895..4ec22c2f2ea3 100644 --- a/fs/lockd/svc.c +++ b/fs/lockd/svc.c @@ -53,7 +53,6 @@ EXPORT_SYMBOL_GPL(nlmsvc_ops); static DEFINE_MUTEX(nlmsvc_mutex); static unsigned int nlmsvc_users; static struct svc_serv *nlmsvc_serv; -unsigned long nlmsvc_timeout; static void nlmsvc_request_retry(struct timer_list *tl) { @@ -68,7 +67,7 @@ unsigned int lockd_net_id; * and also changed through the sysctl interface. -- Jamie Lokier, Aug 2003 */ static unsigned long nlm_grace_period; -static unsigned long nlm_timeout = LOCKD_DFLT_TIMEO; +unsigned long nlm_timeout = LOCKD_DFLT_TIMEO; static int nlm_udpport, nlm_tcpport; /* RLIM_NOFILE defaults to 1024. That seems like a reasonable default here. */ @@ -125,6 +124,8 @@ lockd(void *vrqstp) struct net *net = &init_net; struct lockd_net *ln = net_generic(net, lockd_net_id); + svc_thread_init_status(rqstp, 0); + /* try_to_freeze() is called from svc_recv() */ set_freezable(); @@ -333,10 +334,6 @@ static int lockd_get(void) printk(KERN_WARNING "lockd_up: no pid, %d users??\n", nlmsvc_users); - if (!nlm_timeout) - nlm_timeout = LOCKD_DFLT_TIMEO; - nlmsvc_timeout = nlm_timeout * HZ; - serv = svc_create(&nlmsvc_program, LOCKD_BUFSIZE, lockd); if (!serv) { printk(KERN_WARNING "lockd_up: create service failed\n"); diff --git a/fs/nfs/callback.c b/fs/nfs/callback.c index 8adfcd4c8c1a..6cf92498a5ac 100644 --- a/fs/nfs/callback.c +++ b/fs/nfs/callback.c @@ -76,6 +76,8 @@ nfs4_callback_svc(void *vrqstp) { struct svc_rqst *rqstp = vrqstp; + svc_thread_init_status(rqstp, 0); + set_freezable(); while (!svc_thread_should_stop(rqstp)) diff --git a/fs/nfsd/auth.c b/fs/nfsd/auth.c index e6beaaf4f170..93e33d1ee891 100644 --- a/fs/nfsd/auth.c +++ b/fs/nfsd/auth.c @@ -5,26 +5,26 @@ #include "nfsd.h" #include "auth.h" -int nfsexp_flags(struct svc_rqst *rqstp, struct svc_export *exp) +int nfsexp_flags(struct svc_cred *cred, struct svc_export *exp) { struct exp_flavor_info *f; struct exp_flavor_info *end = exp->ex_flavors + exp->ex_nflavors; for (f = exp->ex_flavors; f < end; f++) { - if (f->pseudoflavor == rqstp->rq_cred.cr_flavor) + if (f->pseudoflavor == cred->cr_flavor) return f->flags; } return exp->ex_flags; } -int nfsd_setuser(struct svc_rqst *rqstp, struct svc_export *exp) +int nfsd_setuser(struct svc_cred *cred, struct svc_export *exp) { struct group_info *rqgi; struct group_info *gi; struct cred *new; int i; - int flags = nfsexp_flags(rqstp, exp); + int flags = nfsexp_flags(cred, exp); /* discard any old override before preparing the new set */ revert_creds(get_cred(current_real_cred())); @@ -32,10 +32,10 @@ int nfsd_setuser(struct svc_rqst *rqstp, struct svc_export *exp) if (!new) return -ENOMEM; - new->fsuid = rqstp->rq_cred.cr_uid; - new->fsgid = rqstp->rq_cred.cr_gid; + new->fsuid = cred->cr_uid; + new->fsgid = cred->cr_gid; - rqgi = rqstp->rq_cred.cr_group_info; + rqgi = cred->cr_group_info; if (flags & NFSEXP_ALLSQUASH) { new->fsuid = exp->ex_anon_uid; diff --git a/fs/nfsd/auth.h b/fs/nfsd/auth.h index dbd66424f600..8c5031bbbcee 100644 --- a/fs/nfsd/auth.h +++ b/fs/nfsd/auth.h @@ -12,6 +12,6 @@ * Set the current process's fsuid/fsgid etc to those of the NFS * client user */ -int nfsd_setuser(struct svc_rqst *, struct svc_export *); +int nfsd_setuser(struct svc_cred *cred, struct svc_export *exp); #endif /* LINUX_NFSD_AUTH_H */ diff --git a/fs/nfsd/blocklayout.c b/fs/nfsd/blocklayout.c index 3c040c81c77d..08a20e5bcf7f 100644 --- a/fs/nfsd/blocklayout.c +++ b/fs/nfsd/blocklayout.c @@ -147,8 +147,7 @@ nfsd4_block_get_device_info_simple(struct super_block *sb, struct pnfs_block_deviceaddr *dev; struct pnfs_block_volume *b; - dev = kzalloc(sizeof(struct pnfs_block_deviceaddr) + - sizeof(struct pnfs_block_volume), GFP_KERNEL); + dev = kzalloc(struct_size(dev, volumes, 1), GFP_KERNEL); if (!dev) return -ENOMEM; gdp->gd_device = dev; @@ -255,8 +254,7 @@ nfsd4_block_get_device_info_scsi(struct super_block *sb, const struct pr_ops *ops; int ret; - dev = kzalloc(sizeof(struct pnfs_block_deviceaddr) + - sizeof(struct pnfs_block_volume), GFP_KERNEL); + dev = kzalloc(struct_size(dev, volumes, 1), GFP_KERNEL); if (!dev) return -ENOMEM; gdp->gd_device = dev; diff --git a/fs/nfsd/blocklayoutxdr.h b/fs/nfsd/blocklayoutxdr.h index b0361e8aa9a7..4e28ac8f1127 100644 --- a/fs/nfsd/blocklayoutxdr.h +++ b/fs/nfsd/blocklayoutxdr.h @@ -47,7 +47,7 @@ struct pnfs_block_volume { struct pnfs_block_deviceaddr { u32 nr_volumes; - struct pnfs_block_volume volumes[]; + struct pnfs_block_volume volumes[] __counted_by(nr_volumes); }; __be32 nfsd4_block_encode_getdeviceinfo(struct xdr_stream *xdr, diff --git a/fs/nfsd/cache.h b/fs/nfsd/cache.h index 66a05fefae98..bb7addef4a31 100644 --- a/fs/nfsd/cache.h +++ b/fs/nfsd/cache.h @@ -10,7 +10,7 @@ #define NFSCACHE_H #include -#include "netns.h" +#include "nfsd.h" /* * Representation of a reply cache entry. diff --git a/fs/nfsd/export.c b/fs/nfsd/export.c index 50b3135d07ac..7bb4f2075ac5 100644 --- a/fs/nfsd/export.c +++ b/fs/nfsd/export.c @@ -1121,7 +1121,7 @@ ok: return 0; denied: - return rqstp->rq_vers < 4 ? nfserr_acces : nfserr_wrongsec; + return nfserr_wrongsec; } /* @@ -1164,19 +1164,35 @@ gss: return gssexp; } +/** + * rqst_exp_find - Find an svc_export in the context of a rqst or similar + * @reqp: The handle to be used to suspend the request if a cache-upcall is needed + * If NULL, missing in-cache information will result in failure. + * @net: The network namespace in which the request exists + * @cl: default auth_domain to use for looking up the export + * @gsscl: an alternate auth_domain defined using deprecated gss/krb5 format. + * @fsid_type: The type of fsid to look for + * @fsidv: The actual fsid to look up in the context of either client. + * + * Perform a lookup for @cl/@fsidv in the given @net for an export. If + * none found and @gsscl specified, repeat the lookup. + * + * Returns an export, or an error pointer. + */ struct svc_export * -rqst_exp_find(struct svc_rqst *rqstp, int fsid_type, u32 *fsidv) +rqst_exp_find(struct cache_req *reqp, struct net *net, + struct auth_domain *cl, struct auth_domain *gsscl, + int fsid_type, u32 *fsidv) { + struct nfsd_net *nn = net_generic(net, nfsd_net_id); struct svc_export *gssexp, *exp = ERR_PTR(-ENOENT); - struct nfsd_net *nn = net_generic(SVC_NET(rqstp), nfsd_net_id); struct cache_detail *cd = nn->svc_export_cache; - if (rqstp->rq_client == NULL) + if (!cl) goto gss; /* First try the auth_unix client: */ - exp = exp_find(cd, rqstp->rq_client, fsid_type, - fsidv, &rqstp->rq_chandle); + exp = exp_find(cd, cl, fsid_type, fsidv, reqp); if (PTR_ERR(exp) == -ENOENT) goto gss; if (IS_ERR(exp)) @@ -1186,10 +1202,9 @@ rqst_exp_find(struct svc_rqst *rqstp, int fsid_type, u32 *fsidv) return exp; gss: /* Otherwise, try falling back on gss client */ - if (rqstp->rq_gssclient == NULL) + if (!gsscl) return exp; - gssexp = exp_find(cd, rqstp->rq_gssclient, fsid_type, fsidv, - &rqstp->rq_chandle); + gssexp = exp_find(cd, gsscl, fsid_type, fsidv, reqp); if (PTR_ERR(gssexp) == -ENOENT) return exp; if (!IS_ERR(exp)) @@ -1220,7 +1235,9 @@ struct svc_export *rqst_find_fsidzero_export(struct svc_rqst *rqstp) mk_fsid(FSID_NUM, fsidv, 0, 0, 0, NULL); - return rqst_exp_find(rqstp, FSID_NUM, fsidv); + return rqst_exp_find(&rqstp->rq_chandle, SVC_NET(rqstp), + rqstp->rq_client, rqstp->rq_gssclient, + FSID_NUM, fsidv); } /* diff --git a/fs/nfsd/export.h b/fs/nfsd/export.h index ca9dc230ae3d..3794ae253a70 100644 --- a/fs/nfsd/export.h +++ b/fs/nfsd/export.h @@ -99,7 +99,8 @@ struct svc_expkey { #define EX_NOHIDE(exp) ((exp)->ex_flags & NFSEXP_NOHIDE) #define EX_WGATHER(exp) ((exp)->ex_flags & NFSEXP_GATHERED_WRITES) -int nfsexp_flags(struct svc_rqst *rqstp, struct svc_export *exp); +struct svc_cred; +int nfsexp_flags(struct svc_cred *cred, struct svc_export *exp); __be32 check_nfsd_access(struct svc_export *exp, struct svc_rqst *rqstp); /* @@ -127,6 +128,8 @@ static inline struct svc_export *exp_get(struct svc_export *exp) cache_get(&exp->h); return exp; } -struct svc_export * rqst_exp_find(struct svc_rqst *, int, u32 *); +struct svc_export *rqst_exp_find(struct cache_req *reqp, struct net *net, + struct auth_domain *cl, struct auth_domain *gsscl, + int fsid_type, u32 *fsidv); #endif /* NFSD_EXPORT_H */ diff --git a/fs/nfsd/filecache.c b/fs/nfsd/filecache.c index f4704f5d4086..24e8f1fbcebb 100644 --- a/fs/nfsd/filecache.c +++ b/fs/nfsd/filecache.c @@ -56,6 +56,7 @@ static DEFINE_PER_CPU(unsigned long, nfsd_file_cache_hits); static DEFINE_PER_CPU(unsigned long, nfsd_file_acquisitions); +static DEFINE_PER_CPU(unsigned long, nfsd_file_allocations); static DEFINE_PER_CPU(unsigned long, nfsd_file_releases); static DEFINE_PER_CPU(unsigned long, nfsd_file_total_age); static DEFINE_PER_CPU(unsigned long, nfsd_file_evictions); @@ -111,7 +112,7 @@ static void nfsd_file_schedule_laundrette(void) { if (test_bit(NFSD_FILE_CACHE_UP, &nfsd_file_flags)) - queue_delayed_work(system_wq, &nfsd_filecache_laundrette, + queue_delayed_work(system_unbound_wq, &nfsd_filecache_laundrette, NFSD_LAUNDRETTE_DELAY); } @@ -151,7 +152,7 @@ nfsd_file_mark_put(struct nfsd_file_mark *nfm) } static struct nfsd_file_mark * -nfsd_file_mark_find_or_create(struct nfsd_file *nf, struct inode *inode) +nfsd_file_mark_find_or_create(struct inode *inode) { int err; struct fsnotify_mark *mark; @@ -215,7 +216,9 @@ nfsd_file_alloc(struct net *net, struct inode *inode, unsigned char need, if (unlikely(!nf)) return NULL; + this_cpu_inc(nfsd_file_allocations); INIT_LIST_HEAD(&nf->nf_lru); + INIT_LIST_HEAD(&nf->nf_gc); nf->nf_birthtime = ktime_get(); nf->nf_file = NULL; nf->nf_cred = get_current_cred(); @@ -393,8 +396,8 @@ nfsd_file_dispose_list(struct list_head *dispose) struct nfsd_file *nf; while (!list_empty(dispose)) { - nf = list_first_entry(dispose, struct nfsd_file, nf_lru); - list_del_init(&nf->nf_lru); + nf = list_first_entry(dispose, struct nfsd_file, nf_gc); + list_del_init(&nf->nf_gc); nfsd_file_free(nf); } } @@ -411,12 +414,12 @@ nfsd_file_dispose_list_delayed(struct list_head *dispose) { while(!list_empty(dispose)) { struct nfsd_file *nf = list_first_entry(dispose, - struct nfsd_file, nf_lru); + struct nfsd_file, nf_gc); struct nfsd_net *nn = net_generic(nf->nf_net, nfsd_net_id); struct nfsd_fcache_disposal *l = nn->fcache_disposal; spin_lock(&l->lock); - list_move_tail(&nf->nf_lru, &l->freeme); + list_move_tail(&nf->nf_gc, &l->freeme); spin_unlock(&l->lock); svc_wake_up(nn->nfsd_serv); } @@ -503,7 +506,8 @@ nfsd_file_lru_cb(struct list_head *item, struct list_lru_one *lru, /* Refcount went to zero. Unhash it and queue it to the dispose list */ nfsd_file_unhash(nf); - list_lru_isolate_move(lru, &nf->nf_lru, head); + list_lru_isolate(lru, &nf->nf_lru); + list_add(&nf->nf_gc, head); this_cpu_inc(nfsd_file_evictions); trace_nfsd_file_gc_disposed(nf); return LRU_REMOVED; @@ -578,7 +582,7 @@ nfsd_file_cond_queue(struct nfsd_file *nf, struct list_head *dispose) /* If refcount goes to 0, then put on the dispose list */ if (refcount_sub_and_test(decrement, &nf->nf_ref)) { - list_add(&nf->nf_lru, dispose); + list_add(&nf->nf_gc, dispose); trace_nfsd_file_closing(nf); } } @@ -654,8 +658,8 @@ nfsd_file_close_inode_sync(struct inode *inode) nfsd_file_queue_for_close(inode, &dispose); while (!list_empty(&dispose)) { - nf = list_first_entry(&dispose, struct nfsd_file, nf_lru); - list_del_init(&nf->nf_lru); + nf = list_first_entry(&dispose, struct nfsd_file, nf_gc); + list_del_init(&nf->nf_gc); nfsd_file_free(nf); } } @@ -909,6 +913,7 @@ nfsd_file_cache_shutdown(void) for_each_possible_cpu(i) { per_cpu(nfsd_file_cache_hits, i) = 0; per_cpu(nfsd_file_acquisitions, i) = 0; + per_cpu(nfsd_file_allocations, i) = 0; per_cpu(nfsd_file_releases, i) = 0; per_cpu(nfsd_file_total_age, i) = 0; per_cpu(nfsd_file_evictions, i) = 0; @@ -1024,7 +1029,7 @@ retry: if (unlikely(nf)) { spin_unlock(&inode->i_lock); rcu_read_unlock(); - nfsd_file_slab_free(&new->nf_rcu); + nfsd_file_free(new); goto wait_for_construction; } nf = new; @@ -1035,8 +1040,6 @@ retry: if (likely(ret == 0)) goto open_file; - if (ret == -EEXIST) - goto retry; trace_nfsd_file_insert_err(rqstp, inode, may_flags, ret); status = nfserr_jukebox; goto construction_err; @@ -1051,6 +1054,7 @@ wait_for_construction: status = nfserr_jukebox; goto construction_err; } + nfsd_file_put(nf); open_retry = false; fh_put(fhp); goto retry; @@ -1074,7 +1078,7 @@ out: open_file: trace_nfsd_file_alloc(nf); - nf->nf_mark = nfsd_file_mark_find_or_create(nf, inode); + nf->nf_mark = nfsd_file_mark_find_or_create(inode); if (nf->nf_mark) { if (file) { get_file(file); @@ -1199,7 +1203,7 @@ nfsd_file_acquire_opened(struct svc_rqst *rqstp, struct svc_fh *fhp, */ int nfsd_file_cache_stats_show(struct seq_file *m, void *v) { - unsigned long releases = 0, evictions = 0; + unsigned long allocations = 0, releases = 0, evictions = 0; unsigned long hits = 0, acquisitions = 0; unsigned int i, count = 0, buckets = 0; unsigned long lru = 0, total_age = 0; @@ -1224,6 +1228,7 @@ int nfsd_file_cache_stats_show(struct seq_file *m, void *v) for_each_possible_cpu(i) { hits += per_cpu(nfsd_file_cache_hits, i); acquisitions += per_cpu(nfsd_file_acquisitions, i); + allocations += per_cpu(nfsd_file_allocations, i); releases += per_cpu(nfsd_file_releases, i); total_age += per_cpu(nfsd_file_total_age, i); evictions += per_cpu(nfsd_file_evictions, i); @@ -1234,6 +1239,7 @@ int nfsd_file_cache_stats_show(struct seq_file *m, void *v) seq_printf(m, "lru entries: %lu\n", lru); seq_printf(m, "cache hits: %lu\n", hits); seq_printf(m, "acquisitions: %lu\n", acquisitions); + seq_printf(m, "allocations: %lu\n", allocations); seq_printf(m, "releases: %lu\n", releases); seq_printf(m, "evictions: %lu\n", evictions); if (releases) diff --git a/fs/nfsd/filecache.h b/fs/nfsd/filecache.h index c61884def906..3fbec24eea6c 100644 --- a/fs/nfsd/filecache.h +++ b/fs/nfsd/filecache.h @@ -44,6 +44,7 @@ struct nfsd_file { struct nfsd_file_mark *nf_mark; struct list_head nf_lru; + struct list_head nf_gc; struct rcu_head nf_rcu; ktime_t nf_birthtime; }; diff --git a/fs/nfsd/netns.h b/fs/nfsd/netns.h index 14ec15656320..37b8bfdcfeea 100644 --- a/fs/nfsd/netns.h +++ b/fs/nfsd/netns.h @@ -148,12 +148,13 @@ struct nfsd_net { u32 s2s_cp_cl_id; struct idr s2s_cp_stateids; spinlock_t s2s_cp_lock; + atomic_t pending_async_copies; /* * Version information */ - bool *nfsd_versions; - bool *nfsd4_minorversions; + bool nfsd_versions[NFSD_MAXVERS + 1]; + bool nfsd4_minorversions[NFSD_SUPPORTED_MINOR_VERSION + 1]; /* * Duplicate reply cache @@ -219,8 +220,6 @@ struct nfsd_net { #define nfsd_netns_ready(nn) ((nn)->sessionid_hashtbl) extern bool nfsd_support_version(int vers); -extern void nfsd_netns_free_versions(struct nfsd_net *nn); - extern unsigned int nfsd_net_id; void nfsd_copy_write_verifier(__be32 verf[2], struct nfsd_net *nn); diff --git a/fs/nfsd/nfs3proc.c b/fs/nfsd/nfs3proc.c index dfcc957e460d..372bdcf5e07a 100644 --- a/fs/nfsd/nfs3proc.c +++ b/fs/nfsd/nfs3proc.c @@ -28,6 +28,29 @@ static int nfs3_ftypes[] = { S_IFIFO, /* NF3FIFO */ }; +static __be32 nfsd3_map_status(__be32 status) +{ + switch (status) { + case nfs_ok: + break; + case nfserr_nofilehandle: + status = nfserr_badhandle; + break; + case nfserr_wrongsec: + case nfserr_file_open: + status = nfserr_acces; + break; + case nfserr_symlink_not_dir: + status = nfserr_notdir; + break; + case nfserr_symlink: + case nfserr_wrong_type: + status = nfserr_inval; + break; + } + return status; +} + /* * NULL call. */ @@ -57,6 +80,7 @@ nfsd3_proc_getattr(struct svc_rqst *rqstp) resp->status = fh_getattr(&resp->fh, &resp->stat); out: + resp->status = nfsd3_map_status(resp->status); return rpc_success; } @@ -80,6 +104,7 @@ nfsd3_proc_setattr(struct svc_rqst *rqstp) if (argp->check_guard) guardtime = &argp->guardtime; resp->status = nfsd_setattr(rqstp, &resp->fh, &attrs, guardtime); + resp->status = nfsd3_map_status(resp->status); return rpc_success; } @@ -103,6 +128,7 @@ nfsd3_proc_lookup(struct svc_rqst *rqstp) resp->status = nfsd_lookup(rqstp, &resp->dirfh, argp->name, argp->len, &resp->fh); + resp->status = nfsd3_map_status(resp->status); return rpc_success; } @@ -122,6 +148,7 @@ nfsd3_proc_access(struct svc_rqst *rqstp) fh_copy(&resp->fh, &argp->fh); resp->access = argp->access; resp->status = nfsd_access(rqstp, &resp->fh, &resp->access, NULL); + resp->status = nfsd3_map_status(resp->status); return rpc_success; } @@ -142,6 +169,7 @@ nfsd3_proc_readlink(struct svc_rqst *rqstp) resp->pages = rqstp->rq_next_page++; resp->status = nfsd_readlink(rqstp, &resp->fh, page_address(*resp->pages), &resp->len); + resp->status = nfsd3_map_status(resp->status); return rpc_success; } @@ -179,6 +207,7 @@ nfsd3_proc_read(struct svc_rqst *rqstp) fh_copy(&resp->fh, &argp->fh); resp->status = nfsd_read(rqstp, &resp->fh, argp->offset, &resp->count, &resp->eof); + resp->status = nfsd3_map_status(resp->status); return rpc_success; } @@ -212,6 +241,7 @@ nfsd3_proc_write(struct svc_rqst *rqstp) rqstp->rq_vec, nvecs, &cnt, resp->committed, resp->verf); resp->count = cnt; + resp->status = nfsd3_map_status(resp->status); return rpc_success; } @@ -359,6 +389,7 @@ nfsd3_proc_create(struct svc_rqst *rqstp) newfhp = fh_init(&resp->fh, NFS3_FHSIZE); resp->status = nfsd3_create_file(rqstp, dirfhp, newfhp, argp); + resp->status = nfsd3_map_status(resp->status); return rpc_success; } @@ -384,6 +415,7 @@ nfsd3_proc_mkdir(struct svc_rqst *rqstp) fh_init(&resp->fh, NFS3_FHSIZE); resp->status = nfsd_create(rqstp, &resp->dirfh, argp->name, argp->len, &attrs, S_IFDIR, 0, &resp->fh); + resp->status = nfsd3_map_status(resp->status); return rpc_success; } @@ -424,6 +456,7 @@ nfsd3_proc_symlink(struct svc_rqst *rqstp) argp->flen, argp->tname, &attrs, &resp->fh); kfree(argp->tname); out: + resp->status = nfsd3_map_status(resp->status); return rpc_success; } @@ -465,6 +498,7 @@ nfsd3_proc_mknod(struct svc_rqst *rqstp) resp->status = nfsd_create(rqstp, &resp->dirfh, argp->name, argp->len, &attrs, type, rdev, &resp->fh); out: + resp->status = nfsd3_map_status(resp->status); return rpc_success; } @@ -486,6 +520,7 @@ nfsd3_proc_remove(struct svc_rqst *rqstp) fh_copy(&resp->fh, &argp->fh); resp->status = nfsd_unlink(rqstp, &resp->fh, -S_IFDIR, argp->name, argp->len); + resp->status = nfsd3_map_status(resp->status); return rpc_success; } @@ -506,6 +541,7 @@ nfsd3_proc_rmdir(struct svc_rqst *rqstp) fh_copy(&resp->fh, &argp->fh); resp->status = nfsd_unlink(rqstp, &resp->fh, S_IFDIR, argp->name, argp->len); + resp->status = nfsd3_map_status(resp->status); return rpc_success; } @@ -528,6 +564,7 @@ nfsd3_proc_rename(struct svc_rqst *rqstp) fh_copy(&resp->tfh, &argp->tfh); resp->status = nfsd_rename(rqstp, &resp->ffh, argp->fname, argp->flen, &resp->tfh, argp->tname, argp->tlen); + resp->status = nfsd3_map_status(resp->status); return rpc_success; } @@ -548,6 +585,7 @@ nfsd3_proc_link(struct svc_rqst *rqstp) fh_copy(&resp->tfh, &argp->tfh); resp->status = nfsd_link(rqstp, &resp->tfh, argp->tname, argp->tlen, &resp->fh); + resp->status = nfsd3_map_status(resp->status); return rpc_success; } @@ -600,6 +638,7 @@ nfsd3_proc_readdir(struct svc_rqst *rqstp) /* Recycle only pages that were part of the reply */ rqstp->rq_next_page = resp->xdr.page_ptr + 1; + resp->status = nfsd3_map_status(resp->status); return rpc_success; } @@ -644,6 +683,7 @@ nfsd3_proc_readdirplus(struct svc_rqst *rqstp) rqstp->rq_next_page = resp->xdr.page_ptr + 1; out: + resp->status = nfsd3_map_status(resp->status); return rpc_success; } @@ -661,6 +701,7 @@ nfsd3_proc_fsstat(struct svc_rqst *rqstp) resp->status = nfsd_statfs(rqstp, &argp->fh, &resp->stats, 0); fh_put(&argp->fh); + resp->status = nfsd3_map_status(resp->status); return rpc_success; } @@ -704,6 +745,7 @@ nfsd3_proc_fsinfo(struct svc_rqst *rqstp) } fh_put(&argp->fh); + resp->status = nfsd3_map_status(resp->status); return rpc_success; } @@ -746,6 +788,7 @@ nfsd3_proc_pathconf(struct svc_rqst *rqstp) } fh_put(&argp->fh); + resp->status = nfsd3_map_status(resp->status); return rpc_success; } @@ -773,6 +816,7 @@ nfsd3_proc_commit(struct svc_rqst *rqstp) argp->count, resp->verf); nfsd_file_put(nf); out: + resp->status = nfsd3_map_status(resp->status); return rpc_success; } diff --git a/fs/nfsd/nfs4callback.c b/fs/nfsd/nfs4callback.c index d756f443fc44..b5b3ab9d719a 100644 --- a/fs/nfsd/nfs4callback.c +++ b/fs/nfsd/nfs4callback.c @@ -1223,6 +1223,7 @@ static void nfsd4_cb_prepare(struct rpc_task *task, void *calldata) * cb_seq_status is only set in decode_cb_sequence4res, * and so will remain 1 if an rpc level failure occurs. */ + trace_nfsd_cb_rpc_prepare(clp); cb->cb_seq_status = 1; cb->cb_status = 0; if (minorversion && !nfsd41_cb_get_slot(cb, task)) @@ -1329,11 +1330,14 @@ static void nfsd4_cb_done(struct rpc_task *task, void *calldata) struct nfsd4_callback *cb = calldata; struct nfs4_client *clp = cb->cb_clp; + trace_nfsd_cb_rpc_done(clp); + if (!nfsd4_cb_sequence_done(task, cb)) return; if (cb->cb_status) { - WARN_ON_ONCE(task->tk_status); + WARN_ONCE(task->tk_status, "cb_status=%d tk_status=%d", + cb->cb_status, task->tk_status); task->tk_status = cb->cb_status; } @@ -1359,6 +1363,8 @@ static void nfsd4_cb_release(void *calldata) { struct nfsd4_callback *cb = calldata; + trace_nfsd_cb_rpc_release(cb->cb_clp); + if (cb->cb_need_restart) nfsd4_queue_cb(cb); else diff --git a/fs/nfsd/nfs4idmap.c b/fs/nfsd/nfs4idmap.c index 7a806ac13e31..8cca1329f348 100644 --- a/fs/nfsd/nfs4idmap.c +++ b/fs/nfsd/nfs4idmap.c @@ -581,6 +581,7 @@ static __be32 idmap_id_to_name(struct xdr_stream *xdr, .id = id, .type = type, }; + __be32 status = nfs_ok; __be32 *p; int ret; struct nfsd_net *nn = net_generic(SVC_NET(rqstp), nfsd_net_id); @@ -593,12 +594,16 @@ static __be32 idmap_id_to_name(struct xdr_stream *xdr, return nfserrno(ret); ret = strlen(item->name); WARN_ON_ONCE(ret > IDMAP_NAMESZ); + p = xdr_reserve_space(xdr, ret + 4); - if (!p) - return nfserr_resource; - p = xdr_encode_opaque(p, item->name, ret); + if (unlikely(!p)) { + status = nfserr_resource; + goto out_put; + } + xdr_encode_opaque(p, item->name, ret); +out_put: cache_put(&item->h, nn->idtoname_cache); - return 0; + return status; } static bool diff --git a/fs/nfsd/nfs4layouts.c b/fs/nfsd/nfs4layouts.c index 4f3072b5979a..fbfddd3c4c94 100644 --- a/fs/nfsd/nfs4layouts.c +++ b/fs/nfsd/nfs4layouts.c @@ -740,6 +740,7 @@ static const struct nfsd4_callback_ops nfsd4_cb_layout_ops = { .prepare = nfsd4_cb_layout_prepare, .done = nfsd4_cb_layout_done, .release = nfsd4_cb_layout_release, + .opcode = OP_CB_LAYOUTRECALL, }; static bool diff --git a/fs/nfsd/nfs4proc.c b/fs/nfsd/nfs4proc.c index 2e39cf2e502a..b5a6bf4f459f 100644 --- a/fs/nfsd/nfs4proc.c +++ b/fs/nfsd/nfs4proc.c @@ -158,7 +158,7 @@ do_open_permission(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfs return fh_verify(rqstp, current_fh, S_IFREG, accmode); } -static __be32 nfsd_check_obj_isreg(struct svc_fh *fh) +static __be32 nfsd_check_obj_isreg(struct svc_fh *fh, u32 minor_version) { umode_t mode = d_inode(fh->fh_dentry)->i_mode; @@ -166,14 +166,15 @@ static __be32 nfsd_check_obj_isreg(struct svc_fh *fh) return nfs_ok; if (S_ISDIR(mode)) return nfserr_isdir; - /* - * Using err_symlink as our catch-all case may look odd; but - * there's no other obvious error for this case in 4.0, and we - * happen to know that it will cause the linux v4 client to do - * the right thing on attempts to open something other than a - * regular file. - */ - return nfserr_symlink; + if (S_ISLNK(mode)) + return nfserr_symlink; + + /* RFC 7530 - 16.16.6 */ + if (minor_version == 0) + return nfserr_symlink; + else + return nfserr_wrong_type; + } static void nfsd4_set_open_owner_reply_cache(struct nfsd4_compound_state *cstate, struct nfsd4_open *open, struct svc_fh *resfh) @@ -466,7 +467,7 @@ do_open_lookup(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate, stru } if (status) goto out; - status = nfsd_check_obj_isreg(*resfh); + status = nfsd_check_obj_isreg(*resfh, cstate->minorversion); if (status) goto out; @@ -751,15 +752,6 @@ nfsd4_access(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate, &access->ac_supported); } -static void gen_boot_verifier(nfs4_verifier *verifier, struct net *net) -{ - __be32 *verf = (__be32 *)verifier->data; - - BUILD_BUG_ON(2*sizeof(*verf) != sizeof(verifier->data)); - - nfsd_copy_write_verifier(verf, net_generic(net, nfsd_net_id)); -} - static __be32 nfsd4_commit(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate, union nfsd4_op_u *u) @@ -1288,6 +1280,7 @@ static void nfs4_put_copy(struct nfsd4_copy *copy) { if (!refcount_dec_and_test(©->refcount)) return; + atomic_dec(©->cp_nn->pending_async_copies); kfree(copy->cp_src); kfree(copy); } @@ -1621,7 +1614,8 @@ static int nfsd4_cb_offload_done(struct nfsd4_callback *cb, static const struct nfsd4_callback_ops nfsd4_cb_offload_ops = { .release = nfsd4_cb_offload_release, - .done = nfsd4_cb_offload_done + .done = nfsd4_cb_offload_done, + .opcode = OP_CB_OFFLOAD, }; static void nfsd4_init_copy_res(struct nfsd4_copy *copy, bool sync) @@ -1630,7 +1624,6 @@ static void nfsd4_init_copy_res(struct nfsd4_copy *copy, bool sync) test_bit(NFSD4_COPY_F_COMMITTED, ©->cp_flags) ? NFS_FILE_SYNC : NFS_UNSTABLE; nfsd4_copy_set_sync(copy, sync); - gen_boot_verifier(©->cp_res.wr_verifier, copy->cp_clp->net); } static ssize_t _nfsd_copy_file_range(struct nfsd4_copy *copy, @@ -1767,7 +1760,7 @@ static int nfsd4_do_async_copy(void *data) { struct nfsd4_copy *copy = (struct nfsd4_copy *)data; - trace_nfsd_copy_do_async(copy); + trace_nfsd_copy_async(copy); if (nfsd4_ssc_is_inter(copy)) { struct file *filp; @@ -1794,6 +1787,7 @@ static int nfsd4_do_async_copy(void *data) do_callback: set_bit(NFSD4_COPY_F_COMPLETED, ©->cp_flags); + trace_nfsd_copy_async_done(copy); nfsd4_send_cb_offload(copy); cleanup_async_copy(copy); return 0; @@ -1803,9 +1797,11 @@ static __be32 nfsd4_copy(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate, union nfsd4_op_u *u) { - struct nfsd4_copy *copy = &u->copy; - __be32 status; + struct nfsd_net *nn = net_generic(SVC_NET(rqstp), nfsd_net_id); struct nfsd4_copy *async_copy = NULL; + struct nfsd4_copy *copy = &u->copy; + struct nfsd42_write_res *result; + __be32 status; /* * Currently, async COPY is not reliable. Force all COPY @@ -1814,6 +1810,9 @@ nfsd4_copy(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate, */ nfsd4_copy_set_sync(copy, true); + result = ©->cp_res; + nfsd_copy_write_verifier((__be32 *)&result->wr_verifier.data, nn); + copy->cp_clp = cstate->clp; if (nfsd4_ssc_is_inter(copy)) { trace_nfsd_copy_inter(copy); @@ -1838,12 +1837,16 @@ nfsd4_copy(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate, memcpy(©->fh, &cstate->current_fh.fh_handle, sizeof(struct knfsd_fh)); if (nfsd4_copy_is_async(copy)) { - struct nfsd_net *nn = net_generic(SVC_NET(rqstp), nfsd_net_id); - - status = nfserrno(-ENOMEM); async_copy = kzalloc(sizeof(struct nfsd4_copy), GFP_KERNEL); if (!async_copy) goto out_err; + async_copy->cp_nn = nn; + /* Arbitrary cap on number of pending async copy operations */ + if (atomic_inc_return(&nn->pending_async_copies) > + (int)rqstp->rq_pool->sp_nrthreads) { + atomic_dec(&nn->pending_async_copies); + goto out_err; + } INIT_LIST_HEAD(&async_copy->copies); refcount_set(&async_copy->refcount, 1); async_copy->cp_src = kmalloc(sizeof(*async_copy->cp_src), GFP_KERNEL); @@ -1851,8 +1854,8 @@ nfsd4_copy(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate, goto out_err; if (!nfs4_init_copy_state(nn, copy)) goto out_err; - memcpy(©->cp_res.cb_stateid, ©->cp_stateid.cs_stid, - sizeof(copy->cp_res.cb_stateid)); + memcpy(&result->cb_stateid, ©->cp_stateid.cs_stid, + sizeof(result->cb_stateid)); dup_copy_fields(copy, async_copy); async_copy->copy_task = kthread_create(nfsd4_do_async_copy, async_copy, "%s", "copy thread"); @@ -1883,7 +1886,7 @@ out_err: } if (async_copy) cleanup_async_copy(async_copy); - status = nfserrno(-ENOMEM); + status = nfserr_jukebox; goto out; } @@ -1942,7 +1945,7 @@ nfsd4_copy_notify(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate, struct nfsd4_copy_notify *cn = &u->copy_notify; __be32 status; struct nfsd_net *nn = net_generic(SVC_NET(rqstp), nfsd_net_id); - struct nfs4_stid *stid; + struct nfs4_stid *stid = NULL; struct nfs4_cpntf_state *cps; struct nfs4_client *clp = cstate->clp; @@ -1951,6 +1954,8 @@ nfsd4_copy_notify(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate, &stid); if (status) return status; + if (!stid) + return nfserr_bad_stateid; cn->cpn_lease_time.tv_sec = nn->nfsd4_lease; cn->cpn_lease_time.tv_nsec = 0; @@ -2231,7 +2236,9 @@ nfsd4_getdeviceinfo(struct svc_rqst *rqstp, return nfserr_noent; } - exp = rqst_exp_find(rqstp, map->fsid_type, map->fsid); + exp = rqst_exp_find(&rqstp->rq_chandle, SVC_NET(rqstp), + rqstp->rq_client, rqstp->rq_gssclient, + map->fsid_type, map->fsid); if (IS_ERR(exp)) { dprintk("%s: could not find device id\n", __func__); return nfserr_noent; diff --git a/fs/nfsd/nfs4recover.c b/fs/nfsd/nfs4recover.c index 67d8673a9391..b7d61eb8afe9 100644 --- a/fs/nfsd/nfs4recover.c +++ b/fs/nfsd/nfs4recover.c @@ -809,6 +809,10 @@ __cld_pipe_inprogress_downcall(const struct cld_msg_v2 __user *cmsg, ci = &cmsg->cm_u.cm_clntinfo; if (get_user(namelen, &ci->cc_name.cn_len)) return -EFAULT; + if (namelen == 0 || namelen > NFS4_OPAQUE_LIMIT) { + dprintk("%s: invalid namelen (%u)", __func__, namelen); + return -EINVAL; + } name.data = memdup_user(&ci->cc_name.cn_id, namelen); if (IS_ERR(name.data)) return PTR_ERR(name.data); @@ -831,6 +835,10 @@ __cld_pipe_inprogress_downcall(const struct cld_msg_v2 __user *cmsg, cnm = &cmsg->cm_u.cm_name; if (get_user(namelen, &cnm->cn_len)) return -EFAULT; + if (namelen == 0 || namelen > NFS4_OPAQUE_LIMIT) { + dprintk("%s: invalid namelen (%u)", __func__, namelen); + return -EINVAL; + } name.data = memdup_user(&cnm->cn_id, namelen); if (IS_ERR(name.data)) return PTR_ERR(name.data); @@ -1895,10 +1903,7 @@ nfsd4_cltrack_upcall_lock(struct nfs4_client *clp) static void nfsd4_cltrack_upcall_unlock(struct nfs4_client *clp) { - smp_mb__before_atomic(); - clear_bit(NFSD4_CLIENT_UPCALL_LOCK, &clp->cl_flags); - smp_mb__after_atomic(); - wake_up_bit(&clp->cl_flags, NFSD4_CLIENT_UPCALL_LOCK); + clear_and_wake_up_bit(NFSD4_CLIENT_UPCALL_LOCK, &clp->cl_flags); } static void diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c index a366fb1c1b9b..ac1859c7cc9d 100644 --- a/fs/nfsd/nfs4state.c +++ b/fs/nfsd/nfs4state.c @@ -400,6 +400,7 @@ static const struct nfsd4_callback_ops nfsd4_cb_notify_lock_ops = { .prepare = nfsd4_cb_notify_lock_prepare, .done = nfsd4_cb_notify_lock_done, .release = nfsd4_cb_notify_lock_release, + .opcode = OP_CB_NOTIFY_LOCK, }; /* @@ -1077,7 +1078,8 @@ static void nfs4_free_deleg(struct nfs4_stid *stid) * When a delegation is recalled, the filehandle is stored in the "new" * filter. * Every 30 seconds we swap the filters and clear the "new" one, - * unless both are empty of course. + * unless both are empty of course. This results in delegations for a + * given filehandle being blocked for between 30 and 60 seconds. * * Each filter is 256 bits. We hash the filehandle to 32bit and use the * low 3 bytes as hash-table indices. @@ -1106,9 +1108,9 @@ static int delegation_blocked(struct knfsd_fh *fh) if (ktime_get_seconds() - bd->swap_time > 30) { bd->entries -= bd->old_entries; bd->old_entries = bd->entries; + bd->new = 1-bd->new; memset(bd->set[bd->new], 0, sizeof(bd->set[0])); - bd->new = 1-bd->new; bd->swap_time = ktime_get_seconds(); } spin_unlock(&blocked_delegations_lock); @@ -1663,9 +1665,7 @@ static void release_openowner(struct nfs4_openowner *oo) { struct nfs4_ol_stateid *stp; struct nfs4_client *clp = oo->oo_owner.so_client; - struct list_head reaplist; - - INIT_LIST_HEAD(&reaplist); + LIST_HEAD(reaplist); spin_lock(&clp->cl_lock); unhash_openowner_locked(oo); @@ -2369,9 +2369,8 @@ __destroy_client(struct nfs4_client *clp) int i; struct nfs4_openowner *oo; struct nfs4_delegation *dp; - struct list_head reaplist; + LIST_HEAD(reaplist); - INIT_LIST_HEAD(&reaplist); spin_lock(&state_lock); while (!list_empty(&clp->cl_delegations)) { dp = list_entry(clp->cl_delegations.next, struct nfs4_delegation, dl_perclnt); @@ -2692,7 +2691,7 @@ static int client_info_show(struct seq_file *m, void *v) clp->cl_nii_time.tv_sec, clp->cl_nii_time.tv_nsec); } seq_printf(m, "callback state: %s\n", cb_state2str(clp->cl_cb_state)); - seq_printf(m, "callback address: %pISpc\n", &clp->cl_cb_conn.cb_addr); + seq_printf(m, "callback address: \"%pISpc\"\n", &clp->cl_cb_conn.cb_addr); seq_printf(m, "admin-revoked states: %d\n", atomic_read(&clp->cl_admin_revoked)); drop_client(clp); @@ -3059,7 +3058,10 @@ nfsd4_cb_getattr_done(struct nfsd4_callback *cb, struct rpc_task *task) { struct nfs4_cb_fattr *ncf = container_of(cb, struct nfs4_cb_fattr, ncf_getattr); + struct nfs4_delegation *dp = + container_of(ncf, struct nfs4_delegation, dl_cb_fattr); + trace_nfsd_cb_getattr_done(&dp->dl_stid.sc_stateid, task); ncf->ncf_cb_status = task->tk_status; switch (task->tk_status) { case -NFS4ERR_DELAY: @@ -3078,19 +3080,20 @@ nfsd4_cb_getattr_release(struct nfsd4_callback *cb) struct nfs4_delegation *dp = container_of(ncf, struct nfs4_delegation, dl_cb_fattr); - clear_bit(CB_GETATTR_BUSY, &ncf->ncf_cb_flags); - wake_up_bit(&ncf->ncf_cb_flags, CB_GETATTR_BUSY); + clear_and_wake_up_bit(CB_GETATTR_BUSY, &ncf->ncf_cb_flags); nfs4_put_stid(&dp->dl_stid); } static const struct nfsd4_callback_ops nfsd4_cb_recall_any_ops = { .done = nfsd4_cb_recall_any_done, .release = nfsd4_cb_recall_any_release, + .opcode = OP_CB_RECALL_ANY, }; static const struct nfsd4_callback_ops nfsd4_cb_getattr_ops = { .done = nfsd4_cb_getattr_done, .release = nfsd4_cb_getattr_release, + .opcode = OP_CB_GETATTR, }; static void nfs4_cb_getattr(struct nfs4_cb_fattr *ncf) @@ -4704,6 +4707,7 @@ void nfsd4_cstate_clear_replay(struct nfsd4_compound_state *cstate) if (so != NULL) { cstate->replay_owner = NULL; atomic_set(&so->so_replay.rp_locked, RP_UNLOCKED); + smp_mb__after_atomic(); wake_up_var(&so->so_replay.rp_locked); nfs4_put_stateowner(so); } @@ -5004,6 +5008,7 @@ move_to_close_lru(struct nfs4_ol_stateid *s, struct net *net) * so tell them to stop waiting. */ atomic_set(&oo->oo_owner.so_replay.rp_locked, RP_UNHASHED); + smp_mb__after_atomic(); wake_up_var(&oo->oo_owner.so_replay.rp_locked); wait_event(close_wq, refcount_read(&s->st_stid.sc_count) == 2); @@ -5218,6 +5223,7 @@ static const struct nfsd4_callback_ops nfsd4_cb_recall_ops = { .prepare = nfsd4_cb_recall_prepare, .done = nfsd4_cb_recall_done, .release = nfsd4_cb_recall_release, + .opcode = OP_CB_RECALL, }; static void nfsd_break_one_deleg(struct nfs4_delegation *dp) @@ -5277,11 +5283,8 @@ static bool nfsd_breaker_owns_lease(struct file_lease *fl) struct svc_rqst *rqst; struct nfs4_client *clp; - if (!i_am_nfsd()) - return false; - rqst = kthread_data(current); - /* Note rq_prog == NFS_ACL_PROGRAM is also possible: */ - if (rqst->rq_prog != NFS_PROGRAM || rqst->rq_vers < 4) + rqst = nfsd_current_rqst(); + if (!nfsd_v4client(rqst)) return false; clp = *(rqst->rq_lease_breaker); return dl->dl_stid.sc_client == clp; @@ -5859,7 +5862,7 @@ nfs4_set_delegation(struct nfsd4_open *open, struct nfs4_ol_stateid *stp, /* * Now that the deleg is set, check again to ensure that nothing - * raced in and changed the mode while we weren't lookng. + * raced in and changed the mode while we weren't looking. */ status = nfsd4_verify_setuid_write(open, fp->fi_deleg_file); if (status) @@ -5912,6 +5915,28 @@ static void nfsd4_open_deleg_none_ext(struct nfsd4_open *open, int status) } } +static bool +nfs4_delegation_stat(struct nfs4_delegation *dp, struct svc_fh *currentfh, + struct kstat *stat) +{ + struct nfsd_file *nf = find_rw_file(dp->dl_stid.sc_file); + struct path path; + int rc; + + if (!nf) + return false; + + path.mnt = currentfh->fh_export->ex_path.mnt; + path.dentry = file_dentry(nf->nf_file); + + rc = vfs_getattr(&path, stat, + (STATX_SIZE | STATX_CTIME | STATX_CHANGE_COOKIE), + AT_STATX_SYNC_AS_STAT); + + nfsd_file_put(nf); + return rc == 0; +} + /* * The Linux NFS server does not offer write delegations to NFSv4.0 * clients in order to avoid conflicts between write delegations and @@ -5947,7 +5972,6 @@ nfs4_open_delegation(struct nfsd4_open *open, struct nfs4_ol_stateid *stp, int cb_up; int status = 0; struct kstat stat; - struct path path; cb_up = nfsd4_cb_channel_good(oo->oo_owner.so_client); open->op_recall = false; @@ -5983,20 +6007,16 @@ nfs4_open_delegation(struct nfsd4_open *open, struct nfs4_ol_stateid *stp, memcpy(&open->op_delegate_stateid, &dp->dl_stid.sc_stateid, sizeof(dp->dl_stid.sc_stateid)); if (open->op_share_access & NFS4_SHARE_ACCESS_WRITE) { - open->op_delegate_type = NFS4_OPEN_DELEGATE_WRITE; - trace_nfsd_deleg_write(&dp->dl_stid.sc_stateid); - path.mnt = currentfh->fh_export->ex_path.mnt; - path.dentry = currentfh->fh_dentry; - if (vfs_getattr(&path, &stat, - (STATX_SIZE | STATX_CTIME | STATX_CHANGE_COOKIE), - AT_STATX_SYNC_AS_STAT)) { + if (!nfs4_delegation_stat(dp, currentfh, &stat)) { nfs4_put_stid(&dp->dl_stid); destroy_delegation(dp); goto out_no_deleg; } + open->op_delegate_type = NFS4_OPEN_DELEGATE_WRITE; dp->dl_cb_fattr.ncf_cur_fsize = stat.size; dp->dl_cb_fattr.ncf_initial_cinfo = nfsd4_change_attribute(&stat, d_inode(currentfh->fh_dentry)); + trace_nfsd_deleg_write(&dp->dl_stid.sc_stateid); } else { open->op_delegate_type = NFS4_OPEN_DELEGATE_READ; trace_nfsd_deleg_read(&dp->dl_stid.sc_stateid); @@ -6271,7 +6291,6 @@ void nfsd4_ssc_init_umount_work(struct nfsd_net *nn) INIT_LIST_HEAD(&nn->nfsd_ssc_mount_list); init_waitqueue_head(&nn->nfsd_ssc_waitq); } -EXPORT_SYMBOL_GPL(nfsd4_ssc_init_umount_work); /* * This is called when nfsd is being shutdown, after all inter_ssc @@ -6619,9 +6638,8 @@ deleg_reaper(struct nfsd_net *nn) { struct list_head *pos, *next; struct nfs4_client *clp; - struct list_head cblist; + LIST_HEAD(cblist); - INIT_LIST_HEAD(&cblist); spin_lock(&nn->client_lock); list_for_each_safe(pos, next, &nn->client_lru) { clp = list_entry(pos, struct nfs4_client, cl_lru); @@ -6647,7 +6665,6 @@ deleg_reaper(struct nfsd_net *nn) cl_ra_cblist); list_del_init(&clp->cl_ra_cblist); clp->cl_ra->ra_keep = 0; - clp->cl_ra->ra_bmval[0] = BIT(RCA4_TYPE_MASK_RDATA_DLG); clp->cl_ra->ra_bmval[0] = BIT(RCA4_TYPE_MASK_RDATA_DLG) | BIT(RCA4_TYPE_MASK_WDATA_DLG); trace_nfsd_cb_recall_any(clp->cl_ra); @@ -6892,7 +6909,8 @@ nfs4_check_file(struct svc_rqst *rqstp, struct svc_fh *fhp, struct nfs4_stid *s, nf = nfs4_find_file(s, flags); if (nf) { - status = nfsd_permission(rqstp, fhp->fh_export, fhp->fh_dentry, + status = nfsd_permission(&rqstp->rq_cred, + fhp->fh_export, fhp->fh_dentry, acc | NFSD_MAY_OWNER_OVERRIDE); if (status) { nfsd_file_put(nf); @@ -7023,11 +7041,7 @@ nfs4_preprocess_stateid_op(struct svc_rqst *rqstp, *nfp = NULL; if (ZERO_STATEID(stateid) || ONE_STATEID(stateid)) { - if (cstid) - status = nfserr_bad_stateid; - else - status = check_special_stateids(net, fhp, stateid, - flags); + status = check_special_stateids(net, fhp, stateid, flags); goto done; } @@ -7481,8 +7495,9 @@ nfsd4_delegreturn(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate, goto put_stateid; trace_nfsd_deleg_return(stateid); - wake_up_var(d_inode(cstate->current_fh.fh_dentry)); destroy_delegation(dp); + smp_mb__after_atomic(); + wake_up_var(d_inode(cstate->current_fh.fh_dentry)); put_stateid: nfs4_put_stid(&dp->dl_stid); out: @@ -8338,7 +8353,7 @@ out: * @cstate: NFSv4 COMPOUND state * @u: RELEASE_LOCKOWNER arguments * - * Check if theree are any locks still held and if not - free the lockowner + * Check if there are any locks still held and if not, free the lockowner * and any lock state that is owned. * * Return values: @@ -8557,6 +8572,7 @@ static int nfs4_state_create_net(struct net *net) spin_lock_init(&nn->client_lock); spin_lock_init(&nn->s2s_cp_lock); idr_init(&nn->s2s_cp_stateids); + atomic_set(&nn->pending_async_copies, 0); spin_lock_init(&nn->blocked_locks_lock); INIT_LIST_HEAD(&nn->blocked_locks_lru); @@ -8836,6 +8852,7 @@ nfsd4_deleg_getattr_conflict(struct svc_rqst *rqstp, struct dentry *dentry, __be32 status; struct nfsd_net *nn = net_generic(SVC_NET(rqstp), nfsd_net_id); struct file_lock_context *ctx; + struct nfs4_delegation *dp = NULL; struct file_lease *fl; struct iattr attrs; struct nfs4_cb_fattr *ncf; @@ -8845,84 +8862,76 @@ nfsd4_deleg_getattr_conflict(struct svc_rqst *rqstp, struct dentry *dentry, ctx = locks_inode_context(inode); if (!ctx) return 0; + +#define NON_NFSD_LEASE ((void *)1) + spin_lock(&ctx->flc_lock); for_each_file_lock(fl, &ctx->flc_lease) { - unsigned char type = fl->c.flc_type; - if (fl->c.flc_flags == FL_LAYOUT) continue; - if (fl->fl_lmops != &nfsd_lease_mng_ops) { - /* - * non-nfs lease, if it's a lease with F_RDLCK then - * we are done; there isn't any write delegation - * on this inode - */ - if (type == F_RDLCK) - break; - - nfsd_stats_wdeleg_getattr_inc(nn); - spin_unlock(&ctx->flc_lock); - - status = nfserrno(nfsd_open_break_lease(inode, NFSD_MAY_READ)); - if (status != nfserr_jukebox || - !nfsd_wait_for_delegreturn(rqstp, inode)) - return status; - return 0; - } - if (type == F_WRLCK) { - struct nfs4_delegation *dp = fl->c.flc_owner; - - if (dp->dl_recall.cb_clp == *(rqstp->rq_lease_breaker)) { - spin_unlock(&ctx->flc_lock); - return 0; - } - nfsd_stats_wdeleg_getattr_inc(nn); - dp = fl->c.flc_owner; - refcount_inc(&dp->dl_stid.sc_count); - ncf = &dp->dl_cb_fattr; - nfs4_cb_getattr(&dp->dl_cb_fattr); - spin_unlock(&ctx->flc_lock); - wait_on_bit_timeout(&ncf->ncf_cb_flags, CB_GETATTR_BUSY, - TASK_INTERRUPTIBLE, NFSD_CB_GETATTR_TIMEOUT); - if (ncf->ncf_cb_status) { - /* Recall delegation only if client didn't respond */ - status = nfserrno(nfsd_open_break_lease(inode, NFSD_MAY_READ)); - if (status != nfserr_jukebox || - !nfsd_wait_for_delegreturn(rqstp, inode)) { - nfs4_put_stid(&dp->dl_stid); - return status; - } - } - if (!ncf->ncf_file_modified && - (ncf->ncf_initial_cinfo != ncf->ncf_cb_change || - ncf->ncf_cur_fsize != ncf->ncf_cb_fsize)) - ncf->ncf_file_modified = true; - if (ncf->ncf_file_modified) { - int err; - - /* - * Per section 10.4.3 of RFC 8881, the server would - * not update the file's metadata with the client's - * modified size - */ - attrs.ia_mtime = attrs.ia_ctime = current_time(inode); - attrs.ia_valid = ATTR_MTIME | ATTR_CTIME | ATTR_DELEG; - inode_lock(inode); - err = notify_change(&nop_mnt_idmap, dentry, &attrs, NULL); - inode_unlock(inode); - if (err) { - nfs4_put_stid(&dp->dl_stid); - return nfserrno(err); - } - ncf->ncf_cur_fsize = ncf->ncf_cb_fsize; - *size = ncf->ncf_cur_fsize; - *modified = true; - } - nfs4_put_stid(&dp->dl_stid); - return 0; + if (fl->c.flc_type == F_WRLCK) { + if (fl->fl_lmops == &nfsd_lease_mng_ops) + dp = fl->c.flc_owner; + else + dp = NON_NFSD_LEASE; } break; } + if (dp == NULL || dp == NON_NFSD_LEASE || + dp->dl_recall.cb_clp == *(rqstp->rq_lease_breaker)) { + spin_unlock(&ctx->flc_lock); + if (dp == NON_NFSD_LEASE) { + status = nfserrno(nfsd_open_break_lease(inode, + NFSD_MAY_READ)); + if (status != nfserr_jukebox || + !nfsd_wait_for_delegreturn(rqstp, inode)) + return status; + } + return 0; + } + + nfsd_stats_wdeleg_getattr_inc(nn); + refcount_inc(&dp->dl_stid.sc_count); + ncf = &dp->dl_cb_fattr; + nfs4_cb_getattr(&dp->dl_cb_fattr); spin_unlock(&ctx->flc_lock); - return 0; + + wait_on_bit_timeout(&ncf->ncf_cb_flags, CB_GETATTR_BUSY, + TASK_INTERRUPTIBLE, NFSD_CB_GETATTR_TIMEOUT); + if (ncf->ncf_cb_status) { + /* Recall delegation only if client didn't respond */ + status = nfserrno(nfsd_open_break_lease(inode, NFSD_MAY_READ)); + if (status != nfserr_jukebox || + !nfsd_wait_for_delegreturn(rqstp, inode)) + goto out_status; + } + if (!ncf->ncf_file_modified && + (ncf->ncf_initial_cinfo != ncf->ncf_cb_change || + ncf->ncf_cur_fsize != ncf->ncf_cb_fsize)) + ncf->ncf_file_modified = true; + if (ncf->ncf_file_modified) { + int err; + + /* + * Per section 10.4.3 of RFC 8881, the server would + * not update the file's metadata with the client's + * modified size + */ + attrs.ia_mtime = attrs.ia_ctime = current_time(inode); + attrs.ia_valid = ATTR_MTIME | ATTR_CTIME | ATTR_DELEG; + inode_lock(inode); + err = notify_change(&nop_mnt_idmap, dentry, &attrs, NULL); + inode_unlock(inode); + if (err) { + status = nfserrno(err); + goto out_status; + } + ncf->ncf_cur_fsize = ncf->ncf_cb_fsize; + *size = ncf->ncf_cur_fsize; + *modified = true; + } + status = 0; +out_status: + nfs4_put_stid(&dp->dl_stid); + return status; } diff --git a/fs/nfsd/nfs4xdr.c b/fs/nfsd/nfs4xdr.c index 97f583777972..f118921250c3 100644 --- a/fs/nfsd/nfs4xdr.c +++ b/fs/nfsd/nfs4xdr.c @@ -1245,14 +1245,6 @@ nfsd4_decode_putfh(struct nfsd4_compoundargs *argp, union nfsd4_op_u *u) return nfs_ok; } -static __be32 -nfsd4_decode_putpubfh(struct nfsd4_compoundargs *argp, union nfsd4_op_u *p) -{ - if (argp->minorversion == 0) - return nfs_ok; - return nfserr_notsupp; -} - static __be32 nfsd4_decode_read(struct nfsd4_compoundargs *argp, union nfsd4_op_u *u) { @@ -2374,7 +2366,7 @@ static const nfsd4_dec nfsd4_dec_ops[] = { [OP_OPEN_CONFIRM] = nfsd4_decode_open_confirm, [OP_OPEN_DOWNGRADE] = nfsd4_decode_open_downgrade, [OP_PUTFH] = nfsd4_decode_putfh, - [OP_PUTPUBFH] = nfsd4_decode_putpubfh, + [OP_PUTPUBFH] = nfsd4_decode_noop, [OP_PUTROOTFH] = nfsd4_decode_noop, [OP_READ] = nfsd4_decode_read, [OP_READDIR] = nfsd4_decode_readdir, @@ -5731,6 +5723,23 @@ __be32 nfsd4_check_resp_size(struct nfsd4_compoundres *resp, u32 respsize) return nfserr_rep_too_big; } +static __be32 nfsd4_map_status(__be32 status, u32 minor) +{ + switch (status) { + case nfs_ok: + break; + case nfserr_wrong_type: + /* RFC 8881 - 15.1.2.9 */ + if (minor == 0) + status = nfserr_inval; + break; + case nfserr_symlink_not_dir: + status = nfserr_symlink; + break; + } + return status; +} + void nfsd4_encode_operation(struct nfsd4_compoundres *resp, struct nfsd4_op *op) { @@ -5798,6 +5807,8 @@ nfsd4_encode_operation(struct nfsd4_compoundres *resp, struct nfsd4_op *op) so->so_replay.rp_buf, len); } status: + op->status = nfsd4_map_status(op->status, + resp->cstate.minorversion); *p = op->status; release: if (opdesc && opdesc->op_release) diff --git a/fs/nfsd/nfsctl.c b/fs/nfsd/nfsctl.c index 34eb2c2cbcde..1c9e5b4bcb0a 100644 --- a/fs/nfsd/nfsctl.c +++ b/fs/nfsd/nfsctl.c @@ -174,6 +174,13 @@ static int export_features_show(struct seq_file *m, void *v) DEFINE_SHOW_ATTRIBUTE(export_features); +static int nfsd_pool_stats_open(struct inode *inode, struct file *file) +{ + struct nfsd_net *nn = net_generic(inode->i_sb->s_fs_info, nfsd_net_id); + + return svc_pool_stats_open(&nn->nfsd_info, file); +} + static const struct file_operations pool_stats_operations = { .open = nfsd_pool_stats_open, .read = seq_read, @@ -1762,7 +1769,7 @@ int nfsd_nl_threads_get_doit(struct sk_buff *skb, struct genl_info *info) struct svc_pool *sp = &nn->nfsd_serv->sv_pools[i]; err = nla_put_u32(skb, NFSD_A_SERVER_THREADS, - atomic_read(&sp->sp_nrthreads)); + sp->sp_nrthreads); if (err) goto err_unlock; } @@ -2224,8 +2231,9 @@ err_free_msg: */ static __net_init int nfsd_net_init(struct net *net) { - int retval; struct nfsd_net *nn = net_generic(net, nfsd_net_id); + int retval; + int i; retval = nfsd_export_init(net); if (retval) @@ -2239,8 +2247,10 @@ static __net_init int nfsd_net_init(struct net *net) goto out_repcache_error; memset(&nn->nfsd_svcstats, 0, sizeof(nn->nfsd_svcstats)); nn->nfsd_svcstats.program = &nfsd_program; - nn->nfsd_versions = NULL; - nn->nfsd4_minorversions = NULL; + for (i = 0; i < sizeof(nn->nfsd_versions); i++) + nn->nfsd_versions[i] = nfsd_support_version(i); + for (i = 0; i < sizeof(nn->nfsd4_minorversions); i++) + nn->nfsd4_minorversions[i] = nfsd_support_version(4); nn->nfsd_info.mutex = &nfsd_mutex; nn->nfsd_serv = NULL; nfsd4_init_leases_net(nn); @@ -2271,7 +2281,6 @@ static __net_exit void nfsd_net_exit(struct net *net) percpu_counter_destroy_many(nn->counter, NFSD_STATS_COUNTERS_NUM); nfsd_idmap_shutdown(net); nfsd_export_shutdown(net); - nfsd_netns_free_versions(nn); } static struct pernet_operations nfsd_net_ops = { diff --git a/fs/nfsd/nfsd.h b/fs/nfsd/nfsd.h index cec8697b1cd6..4ccbf014a2c7 100644 --- a/fs/nfsd/nfsd.h +++ b/fs/nfsd/nfsd.h @@ -23,9 +23,7 @@ #include -#include "netns.h" #include "export.h" -#include "stats.h" #undef ifdebug #ifdef CONFIG_SUNRPC_DEBUG @@ -37,7 +35,14 @@ /* * nfsd version */ +#define NFSD_MINVERS 2 +#define NFSD_MAXVERS 4 #define NFSD_SUPPORTED_MINOR_VERSION 2 +bool nfsd_support_version(int vers); + +#include "netns.h" +#include "stats.h" + /* * Maximum blocksizes supported by daemon under various circumstances. */ @@ -111,11 +116,9 @@ int nfsd_nrthreads(struct net *); int nfsd_nrpools(struct net *); int nfsd_get_nrthreads(int n, int *, struct net *); int nfsd_set_nrthreads(int n, int *, struct net *); -int nfsd_pool_stats_open(struct inode *, struct file *); -int nfsd_pool_stats_release(struct inode *, struct file *); void nfsd_shutdown_threads(struct net *net); -bool i_am_nfsd(void); +struct svc_rqst *nfsd_current_rqst(void); struct nfsdfs_client { struct kref cl_ref; @@ -156,7 +159,7 @@ extern int nfsd_max_blksize; static inline int nfsd_v4client(struct svc_rqst *rq) { - return rq->rq_prog == NFS_PROGRAM && rq->rq_vers == 4; + return rq && rq->rq_prog == NFS_PROGRAM && rq->rq_vers == 4; } static inline struct user_namespace * nfsd_user_namespace(const struct svc_rqst *rqstp) @@ -327,17 +330,36 @@ void nfsd_lockd_shutdown(void); #define nfserr_xattr2big cpu_to_be32(NFS4ERR_XATTR2BIG) #define nfserr_noxattr cpu_to_be32(NFS4ERR_NOXATTR) -/* error codes for internal use */ +/* + * Error codes for internal use. We use enum to choose numbers that are + * not already assigned, then covert to be32 resulting in a number that + * cannot conflict with any existing be32 nfserr value. + */ +enum { + NFSERR_DROPIT = NFS4ERR_FIRST_FREE, /* if a request fails due to kmalloc failure, it gets dropped. * Client should resend eventually */ -#define nfserr_dropit cpu_to_be32(30000) +#define nfserr_dropit cpu_to_be32(NFSERR_DROPIT) + /* end-of-file indicator in readdir */ -#define nfserr_eof cpu_to_be32(30001) + NFSERR_EOF, +#define nfserr_eof cpu_to_be32(NFSERR_EOF) + /* replay detected */ -#define nfserr_replay_me cpu_to_be32(11001) + NFSERR_REPLAY_ME, +#define nfserr_replay_me cpu_to_be32(NFSERR_REPLAY_ME) + /* nfs41 replay detected */ -#define nfserr_replay_cache cpu_to_be32(11002) + NFSERR_REPLAY_CACHE, +#define nfserr_replay_cache cpu_to_be32(NFSERR_REPLAY_CACHE) + +/* symlink found where dir expected - handled differently to + * other symlink found errors by NFSv3. + */ + NFSERR_SYMLINK_NOT_DIR, +#define nfserr_symlink_not_dir cpu_to_be32(NFSERR_SYMLINK_NOT_DIR) +}; /* Check for dir entries '.' and '..' */ #define isdotent(n, l) (l < 3 && n[0] == '.' && (l == 1 || n[1] == '.')) diff --git a/fs/nfsd/nfsfh.c b/fs/nfsd/nfsfh.c index dd4e11a703aa..50d23d56f403 100644 --- a/fs/nfsd/nfsfh.c +++ b/fs/nfsd/nfsfh.c @@ -62,8 +62,7 @@ static int nfsd_acceptable(void *expv, struct dentry *dentry) * the write call). */ static inline __be32 -nfsd_mode_check(struct svc_rqst *rqstp, struct dentry *dentry, - umode_t requested) +nfsd_mode_check(struct dentry *dentry, umode_t requested) { umode_t mode = d_inode(dentry)->i_mode & S_IFMT; @@ -76,17 +75,16 @@ nfsd_mode_check(struct svc_rqst *rqstp, struct dentry *dentry, } return nfs_ok; } - /* - * v4 has an error more specific than err_notdir which we should - * return in preference to err_notdir: - */ - if (rqstp->rq_vers == 4 && mode == S_IFLNK) + if (mode == S_IFLNK) { + if (requested == S_IFDIR) + return nfserr_symlink_not_dir; return nfserr_symlink; + } if (requested == S_IFDIR) return nfserr_notdir; if (mode == S_IFDIR) return nfserr_isdir; - return nfserr_inval; + return nfserr_wrong_type; } static bool nfsd_originating_port_ok(struct svc_rqst *rqstp, int flags) @@ -102,7 +100,7 @@ static bool nfsd_originating_port_ok(struct svc_rqst *rqstp, int flags) static __be32 nfsd_setuser_and_check_port(struct svc_rqst *rqstp, struct svc_export *exp) { - int flags = nfsexp_flags(rqstp, exp); + int flags = nfsexp_flags(&rqstp->rq_cred, exp); /* Check if the request originated from a secure port. */ if (!nfsd_originating_port_ok(rqstp, flags)) { @@ -113,22 +111,14 @@ static __be32 nfsd_setuser_and_check_port(struct svc_rqst *rqstp, } /* Set user creds for this exportpoint */ - return nfserrno(nfsd_setuser(rqstp, exp)); + return nfserrno(nfsd_setuser(&rqstp->rq_cred, exp)); } -static inline __be32 check_pseudo_root(struct svc_rqst *rqstp, - struct dentry *dentry, struct svc_export *exp) +static inline __be32 check_pseudo_root(struct dentry *dentry, + struct svc_export *exp) { if (!(exp->ex_flags & NFSEXP_V4ROOT)) return nfs_ok; - /* - * v2/v3 clients have no need for the V4ROOT export--they use - * the mount protocl instead; also, further V4ROOT checks may be - * in v4-specific code, in which case v2/v3 clients could bypass - * them. - */ - if (!nfsd_v4client(rqstp)) - return nfserr_stale; /* * We're exposing only the directories and symlinks that have to be * traversed on the way to real exports: @@ -162,10 +152,8 @@ static __be32 nfsd_set_fh_dentry(struct svc_rqst *rqstp, struct svc_fh *fhp) int len; __be32 error; - error = nfserr_stale; - if (rqstp->rq_vers > 2) - error = nfserr_badhandle; - if (rqstp->rq_vers == 4 && fh->fh_size == 0) + error = nfserr_badhandle; + if (fh->fh_size == 0) return nfserr_nofilehandle; if (fh->fh_version != 1) @@ -195,7 +183,9 @@ static __be32 nfsd_set_fh_dentry(struct svc_rqst *rqstp, struct svc_fh *fhp) data_left -= len; if (data_left < 0) return error; - exp = rqst_exp_find(rqstp, fh->fh_fsid_type, fh->fh_fsid); + exp = rqst_exp_find(&rqstp->rq_chandle, SVC_NET(rqstp), + rqstp->rq_client, rqstp->rq_gssclient, + fh->fh_fsid_type, fh->fh_fsid); fid = (struct fid *)(fh->fh_fsid + len); error = nfserr_stale; @@ -237,9 +227,7 @@ static __be32 nfsd_set_fh_dentry(struct svc_rqst *rqstp, struct svc_fh *fhp) /* * Look up the dentry using the NFS file handle. */ - error = nfserr_stale; - if (rqstp->rq_vers > 2) - error = nfserr_badhandle; + error = nfserr_badhandle; fileid_type = fh->fh_fileid_type; @@ -282,13 +270,21 @@ static __be32 nfsd_set_fh_dentry(struct svc_rqst *rqstp, struct svc_fh *fhp) case 4: if (dentry->d_sb->s_export_op->flags & EXPORT_OP_NOATOMIC_ATTR) fhp->fh_no_atomic_attr = true; + fhp->fh_64bit_cookies = true; break; case 3: if (dentry->d_sb->s_export_op->flags & EXPORT_OP_NOWCC) fhp->fh_no_wcc = true; + fhp->fh_64bit_cookies = true; + if (exp->ex_flags & NFSEXP_V4ROOT) + goto out; break; case 2: fhp->fh_no_wcc = true; + if (EX_WGATHER(exp)) + fhp->fh_use_wgather = true; + if (exp->ex_flags & NFSEXP_V4ROOT) + goto out; } return 0; @@ -358,7 +354,7 @@ fh_verify(struct svc_rqst *rqstp, struct svc_fh *fhp, umode_t type, int access) * (for example, if different id-squashing options are in * effect on the new filesystem). */ - error = check_pseudo_root(rqstp, dentry, exp); + error = check_pseudo_root(dentry, exp); if (error) goto out; @@ -366,7 +362,7 @@ fh_verify(struct svc_rqst *rqstp, struct svc_fh *fhp, umode_t type, int access) if (error) goto out; - error = nfsd_mode_check(rqstp, dentry, type); + error = nfsd_mode_check(dentry, type); if (error) goto out; @@ -392,7 +388,7 @@ fh_verify(struct svc_rqst *rqstp, struct svc_fh *fhp, umode_t type, int access) skip_pseudoflavor_check: /* Finally, check access permissions. */ - error = nfsd_permission(rqstp, exp, dentry, access); + error = nfsd_permission(&rqstp->rq_cred, exp, dentry, access); out: trace_nfsd_fh_verify_err(rqstp, fhp, type, access, error); if (error == nfserr_stale) diff --git a/fs/nfsd/nfsfh.h b/fs/nfsd/nfsfh.h index 6ebdf7ea27bf..8d46e203d139 100644 --- a/fs/nfsd/nfsfh.h +++ b/fs/nfsd/nfsfh.h @@ -88,6 +88,8 @@ typedef struct svc_fh { * wcc data is not atomic with * operation */ + bool fh_use_wgather; /* NFSv2 wgather option */ + bool fh_64bit_cookies;/* readdir cookie size */ int fh_flags; /* FH flags */ bool fh_post_saved; /* post-op attrs saved */ bool fh_pre_saved; /* pre-op attrs saved */ diff --git a/fs/nfsd/nfsproc.c b/fs/nfsd/nfsproc.c index 36370b957b63..6dda081eb24c 100644 --- a/fs/nfsd/nfsproc.c +++ b/fs/nfsd/nfsproc.c @@ -13,6 +13,31 @@ #define NFSDDBG_FACILITY NFSDDBG_PROC +static __be32 nfsd_map_status(__be32 status) +{ + switch (status) { + case nfs_ok: + break; + case nfserr_nofilehandle: + case nfserr_badhandle: + status = nfserr_stale; + break; + case nfserr_wrongsec: + case nfserr_xdev: + case nfserr_file_open: + status = nfserr_acces; + break; + case nfserr_symlink_not_dir: + status = nfserr_notdir; + break; + case nfserr_symlink: + case nfserr_wrong_type: + status = nfserr_inval; + break; + } + return status; +} + static __be32 nfsd_proc_null(struct svc_rqst *rqstp) { @@ -38,6 +63,7 @@ nfsd_proc_getattr(struct svc_rqst *rqstp) goto out; resp->status = fh_getattr(&resp->fh, &resp->stat); out: + resp->status = nfsd_map_status(resp->status); return rpc_success; } @@ -109,6 +135,7 @@ nfsd_proc_setattr(struct svc_rqst *rqstp) resp->status = fh_getattr(&resp->fh, &resp->stat); out: + resp->status = nfsd_map_status(resp->status); return rpc_success; } @@ -143,6 +170,7 @@ nfsd_proc_lookup(struct svc_rqst *rqstp) resp->status = fh_getattr(&resp->fh, &resp->stat); out: + resp->status = nfsd_map_status(resp->status); return rpc_success; } @@ -164,6 +192,7 @@ nfsd_proc_readlink(struct svc_rqst *rqstp) page_address(resp->page), &resp->len); fh_put(&argp->fh); + resp->status = nfsd_map_status(resp->status); return rpc_success; } @@ -200,6 +229,7 @@ nfsd_proc_read(struct svc_rqst *rqstp) resp->status = fh_getattr(&resp->fh, &resp->stat); else if (resp->status == nfserr_jukebox) set_bit(RQ_DROPME, &rqstp->rq_flags); + resp->status = nfsd_map_status(resp->status); return rpc_success; } @@ -235,6 +265,7 @@ nfsd_proc_write(struct svc_rqst *rqstp) resp->status = fh_getattr(&resp->fh, &resp->stat); else if (resp->status == nfserr_jukebox) set_bit(RQ_DROPME, &rqstp->rq_flags); + resp->status = nfsd_map_status(resp->status); return rpc_success; } @@ -331,10 +362,11 @@ nfsd_proc_create(struct svc_rqst *rqstp) * echo thing > device-special-file-or-pipe * by doing a CREATE with type==0 */ - resp->status = nfsd_permission(rqstp, - newfhp->fh_export, - newfhp->fh_dentry, - NFSD_MAY_WRITE|NFSD_MAY_LOCAL_ACCESS); + resp->status = nfsd_permission( + &rqstp->rq_cred, + newfhp->fh_export, + newfhp->fh_dentry, + NFSD_MAY_WRITE|NFSD_MAY_LOCAL_ACCESS); if (resp->status && resp->status != nfserr_rofs) goto out_unlock; } @@ -403,6 +435,7 @@ done: goto out; resp->status = fh_getattr(&resp->fh, &resp->stat); out: + resp->status = nfsd_map_status(resp->status); return rpc_success; } @@ -419,6 +452,7 @@ nfsd_proc_remove(struct svc_rqst *rqstp) resp->status = nfsd_unlink(rqstp, &argp->fh, -S_IFDIR, argp->name, argp->len); fh_put(&argp->fh); + resp->status = nfsd_map_status(resp->status); return rpc_success; } @@ -437,6 +471,7 @@ nfsd_proc_rename(struct svc_rqst *rqstp) &argp->tfh, argp->tname, argp->tlen); fh_put(&argp->ffh); fh_put(&argp->tfh); + resp->status = nfsd_map_status(resp->status); return rpc_success; } @@ -457,6 +492,7 @@ nfsd_proc_link(struct svc_rqst *rqstp) &argp->ffh); fh_put(&argp->ffh); fh_put(&argp->tfh); + resp->status = nfsd_map_status(resp->status); return rpc_success; } @@ -495,6 +531,7 @@ nfsd_proc_symlink(struct svc_rqst *rqstp) fh_put(&argp->ffh); fh_put(&newfh); out: + resp->status = nfsd_map_status(resp->status); return rpc_success; } @@ -528,6 +565,7 @@ nfsd_proc_mkdir(struct svc_rqst *rqstp) resp->status = fh_getattr(&resp->fh, &resp->stat); out: + resp->status = nfsd_map_status(resp->status); return rpc_success; } @@ -545,6 +583,7 @@ nfsd_proc_rmdir(struct svc_rqst *rqstp) resp->status = nfsd_unlink(rqstp, &argp->fh, S_IFDIR, argp->name, argp->len); fh_put(&argp->fh); + resp->status = nfsd_map_status(resp->status); return rpc_success; } @@ -590,6 +629,7 @@ nfsd_proc_readdir(struct svc_rqst *rqstp) nfssvc_encode_nfscookie(resp, offset); fh_put(&argp->fh); + resp->status = nfsd_map_status(resp->status); return rpc_success; } @@ -607,6 +647,7 @@ nfsd_proc_statfs(struct svc_rqst *rqstp) resp->status = nfsd_statfs(rqstp, &argp->fh, &resp->stats, NFSD_MAY_BYPASS_GSS_ON_ROOT); fh_put(&argp->fh); + resp->status = nfsd_map_status(resp->status); return rpc_success; } diff --git a/fs/nfsd/nfssvc.c b/fs/nfsd/nfssvc.c index 0bc8eaa5e009..defc430f912f 100644 --- a/fs/nfsd/nfssvc.c +++ b/fs/nfsd/nfssvc.c @@ -106,7 +106,7 @@ static struct svc_program nfsd_acl_program = { #endif /* defined(CONFIG_NFSD_V2_ACL) || defined(CONFIG_NFSD_V3_ACL) */ -static const struct svc_version *nfsd_version[] = { +static const struct svc_version *nfsd_version[NFSD_MAXVERS+1] = { #if defined(CONFIG_NFSD_V2) [2] = &nfsd_version2, #endif @@ -116,15 +116,12 @@ static const struct svc_version *nfsd_version[] = { #endif }; -#define NFSD_MINVERS 2 -#define NFSD_NRVERS ARRAY_SIZE(nfsd_version) - struct svc_program nfsd_program = { #if defined(CONFIG_NFSD_V2_ACL) || defined(CONFIG_NFSD_V3_ACL) .pg_next = &nfsd_acl_program, #endif .pg_prog = NFS_PROGRAM, /* program number */ - .pg_nvers = NFSD_NRVERS, /* nr of entries in nfsd_version */ + .pg_nvers = NFSD_MAXVERS+1, /* nr of entries in nfsd_version */ .pg_vers = nfsd_version, /* version table */ .pg_name = "nfsd", /* program name */ .pg_class = "nfsd", /* authentication class */ @@ -135,78 +132,24 @@ struct svc_program nfsd_program = { bool nfsd_support_version(int vers) { - if (vers >= NFSD_MINVERS && vers < NFSD_NRVERS) + if (vers >= NFSD_MINVERS && vers <= NFSD_MAXVERS) return nfsd_version[vers] != NULL; return false; } -static bool * -nfsd_alloc_versions(void) -{ - bool *vers = kmalloc_array(NFSD_NRVERS, sizeof(bool), GFP_KERNEL); - unsigned i; - - if (vers) { - /* All compiled versions are enabled by default */ - for (i = 0; i < NFSD_NRVERS; i++) - vers[i] = nfsd_support_version(i); - } - return vers; -} - -static bool * -nfsd_alloc_minorversions(void) -{ - bool *vers = kmalloc_array(NFSD_SUPPORTED_MINOR_VERSION + 1, - sizeof(bool), GFP_KERNEL); - unsigned i; - - if (vers) { - /* All minor versions are enabled by default */ - for (i = 0; i <= NFSD_SUPPORTED_MINOR_VERSION; i++) - vers[i] = nfsd_support_version(4); - } - return vers; -} - -void -nfsd_netns_free_versions(struct nfsd_net *nn) -{ - kfree(nn->nfsd_versions); - kfree(nn->nfsd4_minorversions); - nn->nfsd_versions = NULL; - nn->nfsd4_minorversions = NULL; -} - -static void -nfsd_netns_init_versions(struct nfsd_net *nn) -{ - if (!nn->nfsd_versions) { - nn->nfsd_versions = nfsd_alloc_versions(); - nn->nfsd4_minorversions = nfsd_alloc_minorversions(); - if (!nn->nfsd_versions || !nn->nfsd4_minorversions) - nfsd_netns_free_versions(nn); - } -} - int nfsd_vers(struct nfsd_net *nn, int vers, enum vers_op change) { - if (vers < NFSD_MINVERS || vers >= NFSD_NRVERS) + if (vers < NFSD_MINVERS || vers > NFSD_MAXVERS) return 0; switch(change) { case NFSD_SET: - if (nn->nfsd_versions) - nn->nfsd_versions[vers] = nfsd_support_version(vers); + nn->nfsd_versions[vers] = nfsd_support_version(vers); break; case NFSD_CLEAR: - nfsd_netns_init_versions(nn); - if (nn->nfsd_versions) - nn->nfsd_versions[vers] = false; + nn->nfsd_versions[vers] = false; break; case NFSD_TEST: - if (nn->nfsd_versions) - return nn->nfsd_versions[vers]; - fallthrough; + return nn->nfsd_versions[vers]; case NFSD_AVAIL: return nfsd_support_version(vers); } @@ -233,23 +176,16 @@ int nfsd_minorversion(struct nfsd_net *nn, u32 minorversion, enum vers_op change switch(change) { case NFSD_SET: - if (nn->nfsd4_minorversions) { - nfsd_vers(nn, 4, NFSD_SET); - nn->nfsd4_minorversions[minorversion] = - nfsd_vers(nn, 4, NFSD_TEST); - } + nfsd_vers(nn, 4, NFSD_SET); + nn->nfsd4_minorversions[minorversion] = + nfsd_vers(nn, 4, NFSD_TEST); break; case NFSD_CLEAR: - nfsd_netns_init_versions(nn); - if (nn->nfsd4_minorversions) { - nn->nfsd4_minorversions[minorversion] = false; - nfsd_adjust_nfsd_versions4(nn); - } + nn->nfsd4_minorversions[minorversion] = false; + nfsd_adjust_nfsd_versions4(nn); break; case NFSD_TEST: - if (nn->nfsd4_minorversions) - return nn->nfsd4_minorversions[minorversion]; - return nfsd_vers(nn, 4, NFSD_TEST); + return nn->nfsd4_minorversions[minorversion]; case NFSD_AVAIL: return minorversion <= NFSD_SUPPORTED_MINOR_VERSION && nfsd_vers(nn, 4, NFSD_AVAIL); @@ -568,11 +504,11 @@ void nfsd_reset_versions(struct nfsd_net *nn) { int i; - for (i = 0; i < NFSD_NRVERS; i++) + for (i = 0; i <= NFSD_MAXVERS; i++) if (nfsd_vers(nn, i, NFSD_TEST)) return; - for (i = 0; i < NFSD_NRVERS; i++) + for (i = 0; i <= NFSD_MAXVERS; i++) if (i != 4) nfsd_vers(nn, i, NFSD_SET); else { @@ -642,9 +578,11 @@ void nfsd_shutdown_threads(struct net *net) mutex_unlock(&nfsd_mutex); } -bool i_am_nfsd(void) +struct svc_rqst *nfsd_current_rqst(void) { - return kthread_func(current) == nfsd; + if (kthread_func(current) == nfsd) + return kthread_data(current); + return NULL; } int nfsd_create_serv(struct net *net) @@ -705,7 +643,7 @@ int nfsd_get_nrthreads(int n, int *nthreads, struct net *net) if (serv) for (i = 0; i < serv->sv_nrpools && i < n; i++) - nthreads[i] = atomic_read(&serv->sv_pools[i].sp_nrthreads); + nthreads[i] = serv->sv_pools[i].sp_nrthreads; return 0; } @@ -905,17 +843,17 @@ nfsd_init_request(struct svc_rqst *rqstp, if (likely(nfsd_vers(nn, rqstp->rq_vers, NFSD_TEST))) return svc_generic_init_request(rqstp, progp, ret); - ret->mismatch.lovers = NFSD_NRVERS; - for (i = NFSD_MINVERS; i < NFSD_NRVERS; i++) { + ret->mismatch.lovers = NFSD_MAXVERS + 1; + for (i = NFSD_MINVERS; i <= NFSD_MAXVERS; i++) { if (nfsd_vers(nn, i, NFSD_TEST)) { ret->mismatch.lovers = i; break; } } - if (ret->mismatch.lovers == NFSD_NRVERS) + if (ret->mismatch.lovers > NFSD_MAXVERS) return rpc_prog_unavail; ret->mismatch.hivers = NFSD_MINVERS; - for (i = NFSD_NRVERS - 1; i >= NFSD_MINVERS; i--) { + for (i = NFSD_MAXVERS; i >= NFSD_MINVERS; i--) { if (nfsd_vers(nn, i, NFSD_TEST)) { ret->mismatch.hivers = i; break; @@ -937,11 +875,9 @@ nfsd(void *vrqstp) /* At this point, the thread shares current->fs * with the init process. We need to create files with the - * umask as defined by the client instead of init's umask. */ - if (unshare_fs_struct() < 0) { - printk("Unable to start nfsd thread: out of memory\n"); - goto out; - } + * umask as defined by the client instead of init's umask. + */ + svc_thread_init_status(rqstp, unshare_fs_struct()); current->fs->umask = 0; @@ -963,7 +899,6 @@ nfsd(void *vrqstp) atomic_dec(&nfsd_th_cnt); -out: /* Release the thread */ svc_exit_thread(rqstp); return 0; @@ -1084,10 +1019,3 @@ bool nfssvc_encode_voidres(struct svc_rqst *rqstp, struct xdr_stream *xdr) { return true; } - -int nfsd_pool_stats_open(struct inode *inode, struct file *file) -{ - struct nfsd_net *nn = net_generic(inode->i_sb->s_fs_info, nfsd_net_id); - - return svc_pool_stats_open(&nn->nfsd_info, file); -} diff --git a/fs/nfsd/state.h b/fs/nfsd/state.h index ec4559ecd193..79c743c01a47 100644 --- a/fs/nfsd/state.h +++ b/fs/nfsd/state.h @@ -79,6 +79,7 @@ struct nfsd4_callback_ops { void (*prepare)(struct nfsd4_callback *); int (*done)(struct nfsd4_callback *, struct rpc_task *); void (*release)(struct nfsd4_callback *); + uint32_t opcode; }; /* diff --git a/fs/nfsd/trace.h b/fs/nfsd/trace.h index 77bbd23aa150..7ab66497e261 100644 --- a/fs/nfsd/trace.h +++ b/fs/nfsd/trace.h @@ -1486,6 +1486,9 @@ DEFINE_NFSD_CB_EVENT(new_state); DEFINE_NFSD_CB_EVENT(probe); DEFINE_NFSD_CB_EVENT(lost); DEFINE_NFSD_CB_EVENT(shutdown); +DEFINE_NFSD_CB_EVENT(rpc_prepare); +DEFINE_NFSD_CB_EVENT(rpc_done); +DEFINE_NFSD_CB_EVENT(rpc_release); TRACE_DEFINE_ENUM(RPC_AUTH_NULL); TRACE_DEFINE_ENUM(RPC_AUTH_UNIX); @@ -1553,6 +1556,19 @@ TRACE_EVENT(nfsd_cb_setup_err, __entry->error) ); +/* Not a real opcode, but there is no 0 operation. */ +#define _CB_NULL 0 + +#define show_nfsd_cb_opcode(val) \ + __print_symbolic(val, \ + { _CB_NULL, "CB_NULL" }, \ + { OP_CB_GETATTR, "CB_GETATTR" }, \ + { OP_CB_RECALL, "CB_RECALL" }, \ + { OP_CB_LAYOUTRECALL, "CB_LAYOUTRECALL" }, \ + { OP_CB_RECALL_ANY, "CB_RECALL_ANY" }, \ + { OP_CB_NOTIFY_LOCK, "CB_NOTIFY_LOCK" }, \ + { OP_CB_OFFLOAD, "CB_OFFLOAD" }) + DECLARE_EVENT_CLASS(nfsd_cb_lifetime_class, TP_PROTO( const struct nfs4_client *clp, @@ -1563,6 +1579,7 @@ DECLARE_EVENT_CLASS(nfsd_cb_lifetime_class, __field(u32, cl_boot) __field(u32, cl_id) __field(const void *, cb) + __field(unsigned long, opcode) __field(bool, need_restart) __sockaddr(addr, clp->cl_cb_conn.cb_addrlen) ), @@ -1570,14 +1587,15 @@ DECLARE_EVENT_CLASS(nfsd_cb_lifetime_class, __entry->cl_boot = clp->cl_clientid.cl_boot; __entry->cl_id = clp->cl_clientid.cl_id; __entry->cb = cb; + __entry->opcode = cb->cb_ops ? cb->cb_ops->opcode : _CB_NULL; __entry->need_restart = cb->cb_need_restart; __assign_sockaddr(addr, &clp->cl_cb_conn.cb_addr, clp->cl_cb_conn.cb_addrlen) ), - TP_printk("addr=%pISpc client %08x:%08x cb=%p%s", - __get_sockaddr(addr), __entry->cl_boot, __entry->cl_id, - __entry->cb, __entry->need_restart ? - " (need restart)" : " (first try)" + TP_printk("addr=%pISpc client %08x:%08x cb=%p%s opcode=%s", + __get_sockaddr(addr), __entry->cl_boot, __entry->cl_id, __entry->cb, + __entry->need_restart ? " (need restart)" : " (first try)", + show_nfsd_cb_opcode(__entry->opcode) ) ); @@ -1830,6 +1848,7 @@ DEFINE_NFSD_CB_DONE_EVENT(nfsd_cb_recall_done); DEFINE_NFSD_CB_DONE_EVENT(nfsd_cb_notify_lock_done); DEFINE_NFSD_CB_DONE_EVENT(nfsd_cb_layout_done); DEFINE_NFSD_CB_DONE_EVENT(nfsd_cb_offload_done); +DEFINE_NFSD_CB_DONE_EVENT(nfsd_cb_getattr_done); TRACE_EVENT(nfsd_cb_recall_any_done, TP_PROTO( @@ -2127,6 +2146,10 @@ DECLARE_EVENT_CLASS(nfsd_copy_class, __field(u32, dst_cl_id) __field(u32, dst_so_id) __field(u32, dst_si_generation) + __field(u32, cb_cl_boot) + __field(u32, cb_cl_id) + __field(u32, cb_so_id) + __field(u32, cb_si_generation) __field(u64, src_cp_pos) __field(u64, dst_cp_pos) __field(u64, cp_count) @@ -2135,6 +2158,7 @@ DECLARE_EVENT_CLASS(nfsd_copy_class, TP_fast_assign( const stateid_t *src_stp = ©->cp_src_stateid; const stateid_t *dst_stp = ©->cp_dst_stateid; + const stateid_t *cb_stp = ©->cp_res.cb_stateid; __entry->intra = test_bit(NFSD4_COPY_F_INTRA, ©->cp_flags); __entry->async = !test_bit(NFSD4_COPY_F_SYNCHRONOUS, ©->cp_flags); @@ -2146,6 +2170,10 @@ DECLARE_EVENT_CLASS(nfsd_copy_class, __entry->dst_cl_id = dst_stp->si_opaque.so_clid.cl_id; __entry->dst_so_id = dst_stp->si_opaque.so_id; __entry->dst_si_generation = dst_stp->si_generation; + __entry->cb_cl_boot = cb_stp->si_opaque.so_clid.cl_boot; + __entry->cb_cl_id = cb_stp->si_opaque.so_clid.cl_id; + __entry->cb_so_id = cb_stp->si_opaque.so_id; + __entry->cb_si_generation = cb_stp->si_generation; __entry->src_cp_pos = copy->cp_src_pos; __entry->dst_cp_pos = copy->cp_dst_pos; __entry->cp_count = copy->cp_count; @@ -2153,14 +2181,17 @@ DECLARE_EVENT_CLASS(nfsd_copy_class, sizeof(struct sockaddr_in6)); ), TP_printk("client=%pISpc intra=%d async=%d " - "src_stateid[si_generation:0x%x cl_boot:0x%x cl_id:0x%x so_id:0x%x] " - "dst_stateid[si_generation:0x%x cl_boot:0x%x cl_id:0x%x so_id:0x%x] " + "src_client %08x:%08x src_stateid %08x:%08x " + "dst_client %08x:%08x dst_stateid %08x:%08x " + "cb_client %08x:%08x cb_stateid %08x:%08x " "cp_src_pos=%llu cp_dst_pos=%llu cp_count=%llu", __get_sockaddr(addr), __entry->intra, __entry->async, - __entry->src_si_generation, __entry->src_cl_boot, - __entry->src_cl_id, __entry->src_so_id, - __entry->dst_si_generation, __entry->dst_cl_boot, - __entry->dst_cl_id, __entry->dst_so_id, + __entry->src_cl_boot, __entry->src_cl_id, + __entry->src_so_id, __entry->src_si_generation, + __entry->dst_cl_boot, __entry->dst_cl_id, + __entry->dst_so_id, __entry->dst_si_generation, + __entry->cb_cl_boot, __entry->cb_cl_id, + __entry->cb_so_id, __entry->cb_si_generation, __entry->src_cp_pos, __entry->dst_cp_pos, __entry->cp_count ) ); @@ -2172,7 +2203,7 @@ DEFINE_EVENT(nfsd_copy_class, nfsd_copy_##name, \ DEFINE_COPY_EVENT(inter); DEFINE_COPY_EVENT(intra); -DEFINE_COPY_EVENT(do_async); +DEFINE_COPY_EVENT(async); TRACE_EVENT(nfsd_copy_done, TP_PROTO( @@ -2193,11 +2224,80 @@ TRACE_EVENT(nfsd_copy_done, __assign_sockaddr(addr, ©->cp_clp->cl_addr, sizeof(struct sockaddr_in6)); ), - TP_printk("addr=%pISpc status=%d intra=%d async=%d ", + TP_printk("addr=%pISpc status=%d intra=%d async=%d", __get_sockaddr(addr), __entry->status, __entry->intra, __entry->async ) ); +TRACE_EVENT(nfsd_copy_async_done, + TP_PROTO( + const struct nfsd4_copy *copy + ), + TP_ARGS(copy), + TP_STRUCT__entry( + __field(int, status) + __field(bool, intra) + __field(bool, async) + __field(u32, src_cl_boot) + __field(u32, src_cl_id) + __field(u32, src_so_id) + __field(u32, src_si_generation) + __field(u32, dst_cl_boot) + __field(u32, dst_cl_id) + __field(u32, dst_so_id) + __field(u32, dst_si_generation) + __field(u32, cb_cl_boot) + __field(u32, cb_cl_id) + __field(u32, cb_so_id) + __field(u32, cb_si_generation) + __field(u64, src_cp_pos) + __field(u64, dst_cp_pos) + __field(u64, cp_count) + __sockaddr(addr, sizeof(struct sockaddr_in6)) + ), + TP_fast_assign( + const stateid_t *src_stp = ©->cp_src_stateid; + const stateid_t *dst_stp = ©->cp_dst_stateid; + const stateid_t *cb_stp = ©->cp_res.cb_stateid; + + __entry->status = be32_to_cpu(copy->nfserr); + __entry->intra = test_bit(NFSD4_COPY_F_INTRA, ©->cp_flags); + __entry->async = !test_bit(NFSD4_COPY_F_SYNCHRONOUS, ©->cp_flags); + __entry->src_cl_boot = src_stp->si_opaque.so_clid.cl_boot; + __entry->src_cl_id = src_stp->si_opaque.so_clid.cl_id; + __entry->src_so_id = src_stp->si_opaque.so_id; + __entry->src_si_generation = src_stp->si_generation; + __entry->dst_cl_boot = dst_stp->si_opaque.so_clid.cl_boot; + __entry->dst_cl_id = dst_stp->si_opaque.so_clid.cl_id; + __entry->dst_so_id = dst_stp->si_opaque.so_id; + __entry->dst_si_generation = dst_stp->si_generation; + __entry->cb_cl_boot = cb_stp->si_opaque.so_clid.cl_boot; + __entry->cb_cl_id = cb_stp->si_opaque.so_clid.cl_id; + __entry->cb_so_id = cb_stp->si_opaque.so_id; + __entry->cb_si_generation = cb_stp->si_generation; + __entry->src_cp_pos = copy->cp_src_pos; + __entry->dst_cp_pos = copy->cp_dst_pos; + __entry->cp_count = copy->cp_count; + __assign_sockaddr(addr, ©->cp_clp->cl_addr, + sizeof(struct sockaddr_in6)); + ), + TP_printk("client=%pISpc status=%d intra=%d async=%d " + "src_client %08x:%08x src_stateid %08x:%08x " + "dst_client %08x:%08x dst_stateid %08x:%08x " + "cb_client %08x:%08x cb_stateid %08x:%08x " + "cp_src_pos=%llu cp_dst_pos=%llu cp_count=%llu", + __get_sockaddr(addr), + __entry->status, __entry->intra, __entry->async, + __entry->src_cl_boot, __entry->src_cl_id, + __entry->src_so_id, __entry->src_si_generation, + __entry->dst_cl_boot, __entry->dst_cl_id, + __entry->dst_so_id, __entry->dst_si_generation, + __entry->cb_cl_boot, __entry->cb_cl_id, + __entry->cb_so_id, __entry->cb_si_generation, + __entry->src_cp_pos, __entry->dst_cp_pos, __entry->cp_count + ) +); + #endif /* _NFSD_TRACE_H */ #undef TRACE_INCLUDE_PATH diff --git a/fs/nfsd/vfs.c b/fs/nfsd/vfs.c index 29b1f3613800..22325b590e17 100644 --- a/fs/nfsd/vfs.c +++ b/fs/nfsd/vfs.c @@ -100,6 +100,7 @@ nfserrno (int errno) { nfserr_io, -EUCLEAN }, { nfserr_perm, -ENOKEY }, { nfserr_no_grace, -ENOGRACE}, + { nfserr_io, -EBADMSG }, }; int i; @@ -421,8 +422,9 @@ nfsd_get_write_access(struct svc_rqst *rqstp, struct svc_fh *fhp, if (iap->ia_size < inode->i_size) { __be32 err; - err = nfsd_permission(rqstp, fhp->fh_export, fhp->fh_dentry, - NFSD_MAY_TRUNC | NFSD_MAY_OWNER_OVERRIDE); + err = nfsd_permission(&rqstp->rq_cred, + fhp->fh_export, fhp->fh_dentry, + NFSD_MAY_TRUNC | NFSD_MAY_OWNER_OVERRIDE); if (err) return err; } @@ -814,7 +816,8 @@ nfsd_access(struct svc_rqst *rqstp, struct svc_fh *fhp, u32 *access, u32 *suppor sresult |= map->access; - err2 = nfsd_permission(rqstp, export, dentry, map->how); + err2 = nfsd_permission(&rqstp->rq_cred, export, + dentry, map->how); switch (err2) { case nfs_ok: result |= map->access; @@ -1160,7 +1163,6 @@ nfsd_vfs_write(struct svc_rqst *rqstp, struct svc_fh *fhp, struct nfsd_file *nf, errseq_t since; __be32 nfserr; int host_err; - int use_wgather; loff_t pos = offset; unsigned long exp_op_flags = 0; unsigned int pflags = current->flags; @@ -1186,12 +1188,11 @@ nfsd_vfs_write(struct svc_rqst *rqstp, struct svc_fh *fhp, struct nfsd_file *nf, } exp = fhp->fh_export; - use_wgather = (rqstp->rq_vers == 2) && EX_WGATHER(exp); if (!EX_ISSYNC(exp)) stable = NFS_UNSTABLE; - if (stable && !use_wgather) + if (stable && !fhp->fh_use_wgather) flags |= RWF_SYNC; iov_iter_kvec(&iter, ITER_SOURCE, vec, vlen, *cnt); @@ -1210,7 +1211,7 @@ nfsd_vfs_write(struct svc_rqst *rqstp, struct svc_fh *fhp, struct nfsd_file *nf, if (host_err < 0) goto out_nfserr; - if (stable && use_wgather) { + if (stable && fhp->fh_use_wgather) { host_err = wait_for_concurrent_writes(file); if (host_err < 0) commit_reset_write_verifier(nn, rqstp, host_err); @@ -1475,7 +1476,8 @@ nfsd_create_locked(struct svc_rqst *rqstp, struct svc_fh *fhp, dirp = d_inode(dentry); dchild = dget(resfhp->fh_dentry); - err = nfsd_permission(rqstp, fhp->fh_export, dentry, NFSD_MAY_CREATE); + err = nfsd_permission(&rqstp->rq_cred, fhp->fh_export, dentry, + NFSD_MAY_CREATE); if (err) goto out; @@ -1767,10 +1769,7 @@ nfsd_link(struct svc_rqst *rqstp, struct svc_fh *ffhp, if (!err) err = nfserrno(commit_metadata(tfhp)); } else { - if (host_err == -EXDEV && rqstp->rq_vers == 2) - err = nfserr_acces; - else - err = nfserrno(host_err); + err = nfserrno(host_err); } dput(dnew); out_drop_write: @@ -1836,7 +1835,7 @@ nfsd_rename(struct svc_rqst *rqstp, struct svc_fh *ffhp, char *fname, int flen, if (!flen || isdotent(fname, flen) || !tlen || isdotent(tname, tlen)) goto out; - err = (rqstp->rq_vers == 2) ? nfserr_acces : nfserr_xdev; + err = nfserr_xdev; if (ffhp->fh_export->ex_path.mnt != tfhp->fh_export->ex_path.mnt) goto out; if (ffhp->fh_export->ex_path.dentry != tfhp->fh_export->ex_path.dentry) @@ -1851,7 +1850,7 @@ retry: trap = lock_rename(tdentry, fdentry); if (IS_ERR(trap)) { - err = (rqstp->rq_vers == 2) ? nfserr_acces : nfserr_xdev; + err = nfserr_xdev; goto out_want_write; } err = fh_fill_pre_attrs(ffhp); @@ -2020,10 +2019,7 @@ out_nfserr: /* name is mounted-on. There is no perfect * error status. */ - if (nfsd_v4client(rqstp)) - err = nfserr_file_open; - else - err = nfserr_acces; + err = nfserr_file_open; } else { err = nfserrno(host_err); } @@ -2178,8 +2174,7 @@ nfsd_readdir(struct svc_rqst *rqstp, struct svc_fh *fhp, loff_t *offsetp, loff_t offset = *offsetp; int may_flags = NFSD_MAY_READ; - /* NFSv2 only supports 32 bit cookies */ - if (rqstp->rq_vers > 2) + if (fhp->fh_64bit_cookies) may_flags |= NFSD_MAY_64BIT_COOKIE; err = nfsd_open(rqstp, fhp, S_IFDIR, may_flags, &file); @@ -2255,9 +2250,9 @@ nfsd_statfs(struct svc_rqst *rqstp, struct svc_fh *fhp, struct kstatfs *stat, in return err; } -static int exp_rdonly(struct svc_rqst *rqstp, struct svc_export *exp) +static int exp_rdonly(struct svc_cred *cred, struct svc_export *exp) { - return nfsexp_flags(rqstp, exp) & NFSEXP_READONLY; + return nfsexp_flags(cred, exp) & NFSEXP_READONLY; } #ifdef CONFIG_NFSD_V4 @@ -2501,8 +2496,8 @@ out_unlock: * Check for a user's access permissions to this inode. */ __be32 -nfsd_permission(struct svc_rqst *rqstp, struct svc_export *exp, - struct dentry *dentry, int acc) +nfsd_permission(struct svc_cred *cred, struct svc_export *exp, + struct dentry *dentry, int acc) { struct inode *inode = d_inode(dentry); int err; @@ -2533,7 +2528,7 @@ nfsd_permission(struct svc_rqst *rqstp, struct svc_export *exp, */ if (!(acc & NFSD_MAY_LOCAL_ACCESS)) if (acc & (NFSD_MAY_WRITE | NFSD_MAY_SATTR | NFSD_MAY_TRUNC)) { - if (exp_rdonly(rqstp, exp) || + if (exp_rdonly(cred, exp) || __mnt_is_readonly(exp->ex_path.mnt)) return nfserr_rofs; if (/* (acc & NFSD_MAY_WRITE) && */ IS_IMMUTABLE(inode)) diff --git a/fs/nfsd/vfs.h b/fs/nfsd/vfs.h index 57cd70062048..01947561d375 100644 --- a/fs/nfsd/vfs.h +++ b/fs/nfsd/vfs.h @@ -153,8 +153,8 @@ __be32 nfsd_readdir(struct svc_rqst *, struct svc_fh *, __be32 nfsd_statfs(struct svc_rqst *, struct svc_fh *, struct kstatfs *, int access); -__be32 nfsd_permission(struct svc_rqst *, struct svc_export *, - struct dentry *, int); +__be32 nfsd_permission(struct svc_cred *cred, struct svc_export *exp, + struct dentry *dentry, int acc); void nfsd_filp_close(struct file *fp); diff --git a/fs/nfsd/xdr4.h b/fs/nfsd/xdr4.h index fbdd42cde1fa..2a21a7662e03 100644 --- a/fs/nfsd/xdr4.h +++ b/fs/nfsd/xdr4.h @@ -713,6 +713,7 @@ struct nfsd4_copy { struct nfsd4_ssc_umount_item *ss_nsui; struct nfs_fh c_fh; nfs4_stateid stateid; + struct nfsd_net *cp_nn; }; static inline void nfsd4_copy_set_sync(struct nfsd4_copy *copy, bool sync) diff --git a/fs/ocfs2/super.c b/fs/ocfs2/super.c index b704983b2112..3d404624bb96 100644 --- a/fs/ocfs2/super.c +++ b/fs/ocfs2/super.c @@ -1571,15 +1571,13 @@ static int __init ocfs2_init(void) ocfs2_set_locking_protocol(); - status = register_quota_format(&ocfs2_quota_format); - if (status < 0) - goto out3; + register_quota_format(&ocfs2_quota_format); + status = register_filesystem(&ocfs2_fs_type); if (!status) return 0; unregister_quota_format(&ocfs2_quota_format); -out3: debugfs_remove(ocfs2_debugfs_root); ocfs2_free_mem_caches(); out2: diff --git a/fs/quota/dquot.c b/fs/quota/dquot.c index 7ae885e6d5d7..b40410cd39af 100644 --- a/fs/quota/dquot.c +++ b/fs/quota/dquot.c @@ -163,13 +163,12 @@ static struct quota_module_name module_names[] = INIT_QUOTA_MODULE_NAMES; /* SLAB cache for dquot structures */ static struct kmem_cache *dquot_cachep; -int register_quota_format(struct quota_format_type *fmt) +void register_quota_format(struct quota_format_type *fmt) { spin_lock(&dq_list_lock); fmt->qf_next = quota_formats; quota_formats = fmt; spin_unlock(&dq_list_lock); - return 0; } EXPORT_SYMBOL(register_quota_format); @@ -1831,7 +1830,6 @@ void dquot_claim_space_nodirty(struct inode *inode, qsize_t number) spin_unlock(&inode->i_lock); mark_all_dquot_dirty(dquots); srcu_read_unlock(&dquot_srcu, index); - return; } EXPORT_SYMBOL(dquot_claim_space_nodirty); @@ -1873,7 +1871,6 @@ void dquot_reclaim_space_nodirty(struct inode *inode, qsize_t number) spin_unlock(&inode->i_lock); mark_all_dquot_dirty(dquots); srcu_read_unlock(&dquot_srcu, index); - return; } EXPORT_SYMBOL(dquot_reclaim_space_nodirty); @@ -2406,7 +2403,7 @@ static int vfs_setup_quota_inode(struct inode *inode, int type) int dquot_load_quota_sb(struct super_block *sb, int type, int format_id, unsigned int flags) { - struct quota_format_type *fmt = find_quota_format(format_id); + struct quota_format_type *fmt; struct quota_info *dqopt = sb_dqopt(sb); int error; @@ -2416,6 +2413,7 @@ int dquot_load_quota_sb(struct super_block *sb, int type, int format_id, if (WARN_ON_ONCE(flags & DQUOT_SUSPENDED)) return -EINVAL; + fmt = find_quota_format(format_id); if (!fmt) return -ESRCH; if (!sb->dq_op || !sb->s_qcop || @@ -2596,7 +2594,8 @@ static int dquot_quota_enable(struct super_block *sb, unsigned int flags) goto out_err; } if (sb_has_quota_limits_enabled(sb, type)) { - ret = -EBUSY; + /* compatible with XFS */ + ret = -EEXIST; goto out_err; } spin_lock(&dq_state_lock); @@ -2610,9 +2609,6 @@ out_err: if (flags & qtype_enforce_flag(type)) dquot_disable(sb, type, DQUOT_LIMITS_ENABLED); } - /* Error code translation for better compatibility with XFS */ - if (ret == -EBUSY) - ret = -EEXIST; return ret; } diff --git a/fs/quota/quota_v1.c b/fs/quota/quota_v1.c index 3f3e8acc05db..6f7f0b4afba9 100644 --- a/fs/quota/quota_v1.c +++ b/fs/quota/quota_v1.c @@ -235,7 +235,8 @@ static struct quota_format_type v1_quota_format = { static int __init init_v1_quota_format(void) { - return register_quota_format(&v1_quota_format); + register_quota_format(&v1_quota_format); + return 0; } static void __exit exit_v1_quota_format(void) diff --git a/fs/quota/quota_v2.c b/fs/quota/quota_v2.c index c48c233f3bef..1fda93dcbc1b 100644 --- a/fs/quota/quota_v2.c +++ b/fs/quota/quota_v2.c @@ -440,12 +440,9 @@ static struct quota_format_type v2r1_quota_format = { static int __init init_v2_quota_format(void) { - int ret; - - ret = register_quota_format(&v2r0_quota_format); - if (ret) - return ret; - return register_quota_format(&v2r1_quota_format); + register_quota_format(&v2r0_quota_format); + register_quota_format(&v2r1_quota_format); + return 0; } static void __exit exit_v2_quota_format(void) diff --git a/include/linux/bcma/bcma_driver_pci.h b/include/linux/bcma/bcma_driver_pci.h index 68da8dba5162..dba41b65ae0d 100644 --- a/include/linux/bcma/bcma_driver_pci.h +++ b/include/linux/bcma/bcma_driver_pci.h @@ -203,7 +203,7 @@ struct pci_dev; #define BCMA_CORE_PCI_MDIO_RXCTRL0 0x840 /* PCIE Root Capability Register bits (Host mode only) */ -#define BCMA_CORE_PCI_RC_CRS_VISIBILITY 0x0001 +#define BCMA_CORE_PCI_RC_RRS_VISIBILITY 0x0001 struct bcma_drv_pci; struct bcma_bus; diff --git a/include/linux/fs.h b/include/linux/fs.h index 6b8df574729c..776298fbfcb4 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -3148,7 +3148,14 @@ static inline bool is_zero_ino(ino_t ino) return (u32)ino == 0; } -extern void __iget(struct inode * inode); +/* + * inode->i_lock must be held + */ +static inline void __iget(struct inode *inode) +{ + atomic_inc(&inode->i_count); +} + extern void iget_failed(struct inode *); extern void clear_inode(struct inode *); extern void __destroy_inode(struct inode *); diff --git a/include/linux/generic-radix-tree.h b/include/linux/generic-radix-tree.h index f3512fddf3d7..5b51c3d582d6 100644 --- a/include/linux/generic-radix-tree.h +++ b/include/linux/generic-radix-tree.h @@ -41,6 +41,7 @@ #include #include #include +#include #include struct genradix_root; @@ -48,10 +49,63 @@ struct genradix_root; #define GENRADIX_NODE_SHIFT 9 #define GENRADIX_NODE_SIZE (1U << GENRADIX_NODE_SHIFT) +#define GENRADIX_ARY (GENRADIX_NODE_SIZE / sizeof(struct genradix_node *)) +#define GENRADIX_ARY_SHIFT ilog2(GENRADIX_ARY) + +/* depth that's needed for a genradix that can address up to ULONG_MAX: */ +#define GENRADIX_MAX_DEPTH \ + DIV_ROUND_UP(BITS_PER_LONG - GENRADIX_NODE_SHIFT, GENRADIX_ARY_SHIFT) + +#define GENRADIX_DEPTH_MASK \ + ((unsigned long) (roundup_pow_of_two(GENRADIX_MAX_DEPTH + 1) - 1)) + +static inline int genradix_depth_shift(unsigned depth) +{ + return GENRADIX_NODE_SHIFT + GENRADIX_ARY_SHIFT * depth; +} + +/* + * Returns size (of data, in bytes) that a tree of a given depth holds: + */ +static inline size_t genradix_depth_size(unsigned depth) +{ + return 1UL << genradix_depth_shift(depth); +} + +static inline unsigned genradix_root_to_depth(struct genradix_root *r) +{ + return (unsigned long) r & GENRADIX_DEPTH_MASK; +} + +static inline struct genradix_node *genradix_root_to_node(struct genradix_root *r) +{ + return (void *) ((unsigned long) r & ~GENRADIX_DEPTH_MASK); +} + struct __genradix { struct genradix_root *root; }; +struct genradix_node { + union { + /* Interior node: */ + struct genradix_node *children[GENRADIX_ARY]; + + /* Leaf: */ + u8 data[GENRADIX_NODE_SIZE]; + }; +}; + +static inline struct genradix_node *genradix_alloc_node(gfp_t gfp_mask) +{ + return kzalloc(GENRADIX_NODE_SIZE, gfp_mask); +} + +static inline void genradix_free_node(struct genradix_node *node) +{ + kfree(node); +} + /* * NOTE: currently, sizeof(_type) must not be larger than GENRADIX_NODE_SIZE: */ @@ -128,6 +182,30 @@ static inline size_t __idx_to_offset(size_t idx, size_t obj_size) #define __genradix_idx_to_offset(_radix, _idx) \ __idx_to_offset(_idx, __genradix_obj_size(_radix)) +static inline void *__genradix_ptr_inlined(struct __genradix *radix, size_t offset) +{ + struct genradix_root *r = READ_ONCE(radix->root); + struct genradix_node *n = genradix_root_to_node(r); + unsigned level = genradix_root_to_depth(r); + unsigned shift = genradix_depth_shift(level); + + if (unlikely(ilog2(offset) >= genradix_depth_shift(level))) + return NULL; + + while (n && shift > GENRADIX_NODE_SHIFT) { + shift -= GENRADIX_ARY_SHIFT; + n = n->children[offset >> shift]; + offset &= (1UL << shift) - 1; + } + + return n ? &n->data[offset] : NULL; +} + +#define genradix_ptr_inlined(_radix, _idx) \ + (__genradix_cast(_radix) \ + __genradix_ptr_inlined(&(_radix)->tree, \ + __genradix_idx_to_offset(_radix, _idx))) + void *__genradix_ptr(struct __genradix *, size_t); /** @@ -142,7 +220,24 @@ void *__genradix_ptr(struct __genradix *, size_t); __genradix_ptr(&(_radix)->tree, \ __genradix_idx_to_offset(_radix, _idx))) -void *__genradix_ptr_alloc(struct __genradix *, size_t, gfp_t); +void *__genradix_ptr_alloc(struct __genradix *, size_t, + struct genradix_node **, gfp_t); + +#define genradix_ptr_alloc_inlined(_radix, _idx, _gfp) \ + (__genradix_cast(_radix) \ + (__genradix_ptr_inlined(&(_radix)->tree, \ + __genradix_idx_to_offset(_radix, _idx)) ?: \ + __genradix_ptr_alloc(&(_radix)->tree, \ + __genradix_idx_to_offset(_radix, _idx), \ + NULL, _gfp))) + +#define genradix_ptr_alloc_preallocated_inlined(_radix, _idx, _new_node, _gfp)\ + (__genradix_cast(_radix) \ + (__genradix_ptr_inlined(&(_radix)->tree, \ + __genradix_idx_to_offset(_radix, _idx)) ?: \ + __genradix_ptr_alloc(&(_radix)->tree, \ + __genradix_idx_to_offset(_radix, _idx), \ + _new_node, _gfp))) /** * genradix_ptr_alloc - get a pointer to a genradix entry, allocating it @@ -157,7 +252,13 @@ void *__genradix_ptr_alloc(struct __genradix *, size_t, gfp_t); (__genradix_cast(_radix) \ __genradix_ptr_alloc(&(_radix)->tree, \ __genradix_idx_to_offset(_radix, _idx), \ - _gfp)) + NULL, _gfp)) + +#define genradix_ptr_alloc_preallocated(_radix, _idx, _new_node, _gfp)\ + (__genradix_cast(_radix) \ + __genradix_ptr_alloc(&(_radix)->tree, \ + __genradix_idx_to_offset(_radix, _idx), \ + _new_node, _gfp)) struct genradix_iter { size_t offset; diff --git a/include/linux/lockd/lockd.h b/include/linux/lockd/lockd.h index 1b95fe31051f..61c4b9c41904 100644 --- a/include/linux/lockd/lockd.h +++ b/include/linux/lockd/lockd.h @@ -200,7 +200,7 @@ extern const struct svc_procedure nlmsvc_procedures[24]; extern const struct svc_procedure nlmsvc_procedures4[24]; #endif extern int nlmsvc_grace_period; -extern unsigned long nlmsvc_timeout; +extern unsigned long nlm_timeout; extern bool nsm_use_hostnames; extern u32 nsm_local_state; diff --git a/include/linux/msi.h b/include/linux/msi.h index 944979763825..b10093c4d00e 100644 --- a/include/linux/msi.h +++ b/include/linux/msi.h @@ -554,6 +554,8 @@ enum { MSI_FLAG_MSIX_CONTIGUOUS = (1 << 19), /* PCI/MSI-X vectors can be dynamically allocated/freed post MSI-X enable */ MSI_FLAG_PCI_MSIX_ALLOC_DYN = (1 << 20), + /* PCI MSIs cannot be steered separately to CPU cores */ + MSI_FLAG_NO_AFFINITY = (1 << 21), }; /** diff --git a/include/linux/nfs4.h b/include/linux/nfs4.h index f9df88091c6d..8d7430d9f218 100644 --- a/include/linux/nfs4.h +++ b/include/linux/nfs4.h @@ -281,15 +281,18 @@ enum nfsstat4 { /* nfs42 */ NFS4ERR_PARTNER_NOTSUPP = 10088, NFS4ERR_PARTNER_NO_AUTH = 10089, - NFS4ERR_UNION_NOTSUPP = 10090, - NFS4ERR_OFFLOAD_DENIED = 10091, - NFS4ERR_WRONG_LFS = 10092, - NFS4ERR_BADLABEL = 10093, - NFS4ERR_OFFLOAD_NO_REQS = 10094, + NFS4ERR_UNION_NOTSUPP = 10090, + NFS4ERR_OFFLOAD_DENIED = 10091, + NFS4ERR_WRONG_LFS = 10092, + NFS4ERR_BADLABEL = 10093, + NFS4ERR_OFFLOAD_NO_REQS = 10094, /* xattr (RFC8276) */ - NFS4ERR_NOXATTR = 10095, - NFS4ERR_XATTR2BIG = 10096, + NFS4ERR_NOXATTR = 10095, + NFS4ERR_XATTR2BIG = 10096, + + /* can be used for internal errors */ + NFS4ERR_FIRST_FREE }; /* error codes for internal client use */ diff --git a/include/linux/pci-epc.h b/include/linux/pci-epc.h index 85bdf2adb760..42ef06136bd1 100644 --- a/include/linux/pci-epc.h +++ b/include/linux/pci-epc.h @@ -128,6 +128,7 @@ struct pci_epc_mem { * @group: configfs group representing the PCI EPC device * @lock: mutex to protect pci_epc ops * @function_num_map: bitmap to manage physical function number + * @domain_nr: PCI domain number of the endpoint controller * @init_complete: flag to indicate whether the EPC initialization is complete * or not */ @@ -145,10 +146,12 @@ struct pci_epc { /* mutex to protect against concurrent access of EP controller */ struct mutex lock; unsigned long function_num_map; + int domain_nr; bool init_complete; }; /** + * enum pci_epc_bar_type - configurability of endpoint BAR * @BAR_PROGRAMMABLE: The BAR mask can be configured by the EPC. * @BAR_FIXED: The BAR mask is fixed by the hardware. * @BAR_RESERVED: The BAR should not be touched by an EPF driver. diff --git a/include/linux/pci.h b/include/linux/pci.h index 4cf89a4b4cbc..573b4c4c2be6 100644 --- a/include/linux/pci.h +++ b/include/linux/pci.h @@ -371,6 +371,7 @@ struct pci_dev { can be generated */ unsigned int pme_poll:1; /* Poll device's PME status bit */ unsigned int pinned:1; /* Whether this dev is pinned */ + unsigned int config_rrs_sv:1; /* Config RRS software visibility */ unsigned int imm_ready:1; /* Supports Immediate Readiness */ unsigned int d1_support:1; /* Low power state D1 is supported */ unsigned int d2_support:1; /* Low power state D2 is supported */ @@ -516,6 +517,9 @@ struct pci_dev { #endif #ifdef CONFIG_PCI_DOE struct xarray doe_mbs; /* Data Object Exchange mailboxes */ +#endif +#ifdef CONFIG_PCI_NPEM + struct npem *npem; /* Native PCIe Enclosure Management */ #endif u16 acs_cap; /* ACS Capability offset */ phys_addr_t rom; /* Physical address if not from BAR */ @@ -1098,7 +1102,7 @@ enum pcie_bus_config_types { extern enum pcie_bus_config_types pcie_bus_config; -extern struct bus_type pci_bus_type; +extern const struct bus_type pci_bus_type; /* Do NOT directly access these two variables, unless you are arch-specific PCI * code, or PCI core code. */ @@ -1884,7 +1888,7 @@ static inline int acpi_pci_bus_find_domain_nr(struct pci_bus *bus) { return 0; } #endif int pci_bus_find_domain_nr(struct pci_bus *bus, struct device *parent); -void pci_bus_release_domain_nr(struct pci_bus *bus, struct device *parent); +void pci_bus_release_domain_nr(struct device *parent, int domain_nr); #endif /* Some architectures require additional setup to direct VGA traffic */ @@ -2290,8 +2294,11 @@ static inline void pci_fixup_device(enum pci_fixup_pass pass, #endif void __iomem *pcim_iomap(struct pci_dev *pdev, int bar, unsigned long maxlen); +void __iomem *pcim_iomap_region(struct pci_dev *pdev, int bar, + const char *name); void pcim_iounmap(struct pci_dev *pdev, void __iomem *addr); void __iomem * const *pcim_iomap_table(struct pci_dev *pdev); +int pcim_request_region(struct pci_dev *pdev, int bar, const char *name); int pcim_iomap_regions(struct pci_dev *pdev, int mask, const char *name); int pcim_iomap_regions_request_all(struct pci_dev *pdev, int mask, const char *name); diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h index 91182aa1d2ec..e4bddb927795 100644 --- a/include/linux/pci_ids.h +++ b/include/linux/pci_ids.h @@ -2662,6 +2662,8 @@ #define PCI_DEVICE_ID_DCI_PCCOM8 0x0002 #define PCI_DEVICE_ID_DCI_PCCOM2 0x0004 +#define PCI_VENDOR_ID_GLENFLY 0x6766 + #define PCI_VENDOR_ID_INTEL 0x8086 #define PCI_DEVICE_ID_INTEL_EESSC 0x0008 #define PCI_DEVICE_ID_INTEL_HDA_CML_LP 0x02c8 diff --git a/include/linux/quota.h b/include/linux/quota.h index 07071e64abf3..89a0d83ddad0 100644 --- a/include/linux/quota.h +++ b/include/linux/quota.h @@ -526,7 +526,7 @@ struct quota_info { const struct quota_format_ops *ops[MAXQUOTAS]; /* Operations for each type */ }; -int register_quota_format(struct quota_format_type *fmt); +void register_quota_format(struct quota_format_type *fmt); void unregister_quota_format(struct quota_format_type *fmt); struct quota_module_name { diff --git a/include/linux/sunrpc/svc.h b/include/linux/sunrpc/svc.h index a7d0406b9ef5..c419a61f60e5 100644 --- a/include/linux/sunrpc/svc.h +++ b/include/linux/sunrpc/svc.h @@ -21,6 +21,7 @@ #include #include #include +#include /* * @@ -33,9 +34,9 @@ * node traffic on multi-node NUMA NFS servers. */ struct svc_pool { - unsigned int sp_id; /* pool id; also node id on NUMA */ + unsigned int sp_id; /* pool id; also node id on NUMA */ struct lwq sp_xprts; /* pending transports */ - atomic_t sp_nrthreads; /* # of threads in pool */ + unsigned int sp_nrthreads; /* # of threads in pool */ struct list_head sp_all_threads; /* all server threads */ struct llist_head sp_idle_threads; /* idle server threads */ @@ -232,6 +233,11 @@ struct svc_rqst { struct net *rq_bc_net; /* pointer to backchannel's * net namespace */ + + int rq_err; /* Thread sets this to inidicate + * initialisation success. + */ + unsigned long bc_to_initval; unsigned int bc_to_retries; void ** rq_lease_breaker; /* The v4 client breaking a lease */ @@ -305,6 +311,31 @@ static inline bool svc_thread_should_stop(struct svc_rqst *rqstp) return test_bit(RQ_VICTIM, &rqstp->rq_flags); } +/** + * svc_thread_init_status - report whether thread has initialised successfully + * @rqstp: the thread in question + * @err: errno code + * + * After performing any initialisation that could fail, and before starting + * normal work, each sunrpc svc_thread must call svc_thread_init_status() + * with an appropriate error, or zero. + * + * If zero is passed, the thread is ready and must continue until + * svc_thread_should_stop() returns true. If a non-zero error is passed + * the call will not return - the thread will exit. + */ +static inline void svc_thread_init_status(struct svc_rqst *rqstp, int err) +{ + rqstp->rq_err = err; + /* memory barrier ensures assignment to error above is visible before + * waitqueue_active() test below completes. + */ + smp_mb(); + wake_up_var(&rqstp->rq_err); + if (err) + kthread_exit(1); +} + struct svc_deferred_req { u32 prot; /* protocol (UDP or TCP) */ struct svc_xprt *xprt; @@ -401,17 +432,13 @@ struct svc_procedure { */ int sunrpc_set_pool_mode(const char *val); int sunrpc_get_pool_mode(char *val, size_t size); -int svc_rpcb_setup(struct svc_serv *serv, struct net *net); void svc_rpcb_cleanup(struct svc_serv *serv, struct net *net); int svc_bind(struct svc_serv *serv, struct net *net); struct svc_serv *svc_create(struct svc_program *, unsigned int, int (*threadfn)(void *data)); -struct svc_rqst *svc_rqst_alloc(struct svc_serv *serv, - struct svc_pool *pool, int node); bool svc_rqst_replace_page(struct svc_rqst *rqstp, struct page *page); void svc_rqst_release_pages(struct svc_rqst *rqstp); -void svc_rqst_free(struct svc_rqst *); void svc_exit_thread(struct svc_rqst *); struct svc_serv * svc_create_pooled(struct svc_program *prog, struct svc_stat *stats, @@ -446,11 +473,6 @@ int svc_generic_rpcbind_set(struct net *net, u32 version, int family, unsigned short proto, unsigned short port); -int svc_rpcbind_set_version(struct net *net, - const struct svc_program *progp, - u32 version, int family, - unsigned short proto, - unsigned short port); #define RPC_MAX_ADDRBUFLEN (63U) diff --git a/include/linux/sunrpc/svc_rdma.h b/include/linux/sunrpc/svc_rdma.h index d33bab33099a..619fc0bd837a 100644 --- a/include/linux/sunrpc/svc_rdma.h +++ b/include/linux/sunrpc/svc_rdma.h @@ -48,6 +48,7 @@ #include #include #include +#include #include #include @@ -76,6 +77,7 @@ struct svcxprt_rdma { struct svc_xprt sc_xprt; /* SVC transport structure */ struct rdma_cm_id *sc_cm_id; /* RDMA connection id */ struct list_head sc_accept_q; /* Conn. waiting accept */ + struct rpcrdma_notification sc_rn; /* removal notification */ int sc_ord; /* RDMA read limit */ int sc_max_send_sges; bool sc_snd_w_inv; /* OK to use Send With Invalidate */ diff --git a/include/linux/sunrpc/svcauth.h b/include/linux/sunrpc/svcauth.h index 61c455f1e1f5..63cf6fb26dcc 100644 --- a/include/linux/sunrpc/svcauth.h +++ b/include/linux/sunrpc/svcauth.h @@ -151,7 +151,6 @@ struct auth_ops { struct svc_xprt; -extern enum svc_auth_status svc_authenticate(struct svc_rqst *rqstp); extern rpc_authflavor_t svc_auth_flavor(struct svc_rqst *rqstp); extern int svc_authorise(struct svc_rqst *rqstp); extern enum svc_auth_status svc_set_client(struct svc_rqst *rqstp); diff --git a/include/linux/sunrpc/svcsock.h b/include/linux/sunrpc/svcsock.h index 7c78ec6356b9..bf45d9e8492a 100644 --- a/include/linux/sunrpc/svcsock.h +++ b/include/linux/sunrpc/svcsock.h @@ -58,8 +58,6 @@ static inline u32 svc_sock_final_rec(struct svc_sock *svsk) */ void svc_recv(struct svc_rqst *rqstp); void svc_send(struct svc_rqst *rqstp); -void svc_drop(struct svc_rqst *); -void svc_sock_update_bufs(struct svc_serv *serv); int svc_addsock(struct svc_serv *serv, struct net *net, const int fd, char *name_return, const size_t len, const struct cred *cred); diff --git a/include/linux/sunrpc/xdrgen/_builtins.h b/include/linux/sunrpc/xdrgen/_builtins.h new file mode 100644 index 000000000000..66ca3ece951a --- /dev/null +++ b/include/linux/sunrpc/xdrgen/_builtins.h @@ -0,0 +1,243 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* + * Copyright (c) 2024 Oracle and/or its affiliates. + * + * This header defines XDR data type primitives specified in + * Section 4 of RFC 4506, used by RPC programs implemented + * in the Linux kernel. + */ + +#ifndef _SUNRPC_XDRGEN__BUILTINS_H_ +#define _SUNRPC_XDRGEN__BUILTINS_H_ + +#include + +static inline bool +xdrgen_decode_void(struct xdr_stream *xdr) +{ + return true; +} + +static inline bool +xdrgen_encode_void(struct xdr_stream *xdr) +{ + return true; +} + +static inline bool +xdrgen_decode_bool(struct xdr_stream *xdr, bool *ptr) +{ + __be32 *p = xdr_inline_decode(xdr, XDR_UNIT); + + if (unlikely(!p)) + return false; + *ptr = (*p != xdr_zero); + return true; +} + +static inline bool +xdrgen_encode_bool(struct xdr_stream *xdr, bool val) +{ + __be32 *p = xdr_reserve_space(xdr, XDR_UNIT); + + if (unlikely(!p)) + return false; + *p = val ? xdr_one : xdr_zero; + return true; +} + +static inline bool +xdrgen_decode_int(struct xdr_stream *xdr, s32 *ptr) +{ + __be32 *p = xdr_inline_decode(xdr, XDR_UNIT); + + if (unlikely(!p)) + return false; + *ptr = be32_to_cpup(p); + return true; +} + +static inline bool +xdrgen_encode_int(struct xdr_stream *xdr, s32 val) +{ + __be32 *p = xdr_reserve_space(xdr, XDR_UNIT); + + if (unlikely(!p)) + return false; + *p = cpu_to_be32(val); + return true; +} + +static inline bool +xdrgen_decode_unsigned_int(struct xdr_stream *xdr, u32 *ptr) +{ + __be32 *p = xdr_inline_decode(xdr, XDR_UNIT); + + if (unlikely(!p)) + return false; + *ptr = be32_to_cpup(p); + return true; +} + +static inline bool +xdrgen_encode_unsigned_int(struct xdr_stream *xdr, u32 val) +{ + __be32 *p = xdr_reserve_space(xdr, XDR_UNIT); + + if (unlikely(!p)) + return false; + *p = cpu_to_be32(val); + return true; +} + +static inline bool +xdrgen_decode_long(struct xdr_stream *xdr, s32 *ptr) +{ + __be32 *p = xdr_inline_decode(xdr, XDR_UNIT); + + if (unlikely(!p)) + return false; + *ptr = be32_to_cpup(p); + return true; +} + +static inline bool +xdrgen_encode_long(struct xdr_stream *xdr, s32 val) +{ + __be32 *p = xdr_reserve_space(xdr, XDR_UNIT); + + if (unlikely(!p)) + return false; + *p = cpu_to_be32(val); + return true; +} + +static inline bool +xdrgen_decode_unsigned_long(struct xdr_stream *xdr, u32 *ptr) +{ + __be32 *p = xdr_inline_decode(xdr, XDR_UNIT); + + if (unlikely(!p)) + return false; + *ptr = be32_to_cpup(p); + return true; +} + +static inline bool +xdrgen_encode_unsigned_long(struct xdr_stream *xdr, u32 val) +{ + __be32 *p = xdr_reserve_space(xdr, XDR_UNIT); + + if (unlikely(!p)) + return false; + *p = cpu_to_be32(val); + return true; +} + +static inline bool +xdrgen_decode_hyper(struct xdr_stream *xdr, s64 *ptr) +{ + __be32 *p = xdr_inline_decode(xdr, XDR_UNIT * 2); + + if (unlikely(!p)) + return false; + *ptr = get_unaligned_be64(p); + return true; +} + +static inline bool +xdrgen_encode_hyper(struct xdr_stream *xdr, s64 val) +{ + __be32 *p = xdr_reserve_space(xdr, XDR_UNIT * 2); + + if (unlikely(!p)) + return false; + put_unaligned_be64(val, p); + return true; +} + +static inline bool +xdrgen_decode_unsigned_hyper(struct xdr_stream *xdr, u64 *ptr) +{ + __be32 *p = xdr_inline_decode(xdr, XDR_UNIT * 2); + + if (unlikely(!p)) + return false; + *ptr = get_unaligned_be64(p); + return true; +} + +static inline bool +xdrgen_encode_unsigned_hyper(struct xdr_stream *xdr, u64 val) +{ + __be32 *p = xdr_reserve_space(xdr, XDR_UNIT * 2); + + if (unlikely(!p)) + return false; + put_unaligned_be64(val, p); + return true; +} + +static inline bool +xdrgen_decode_string(struct xdr_stream *xdr, string *ptr, u32 maxlen) +{ + __be32 *p; + u32 len; + + if (unlikely(xdr_stream_decode_u32(xdr, &len) < 0)) + return false; + if (unlikely(maxlen && len > maxlen)) + return false; + if (len != 0) { + p = xdr_inline_decode(xdr, len); + if (unlikely(!p)) + return false; + ptr->data = (unsigned char *)p; + } + ptr->len = len; + return true; +} + +static inline bool +xdrgen_encode_string(struct xdr_stream *xdr, string val, u32 maxlen) +{ + __be32 *p = xdr_reserve_space(xdr, XDR_UNIT + xdr_align_size(val.len)); + + if (unlikely(!p)) + return false; + xdr_encode_opaque(p, val.data, val.len); + return true; +} + +static inline bool +xdrgen_decode_opaque(struct xdr_stream *xdr, opaque *ptr, u32 maxlen) +{ + __be32 *p; + u32 len; + + if (unlikely(xdr_stream_decode_u32(xdr, &len) < 0)) + return false; + if (unlikely(maxlen && len > maxlen)) + return false; + if (len != 0) { + p = xdr_inline_decode(xdr, len); + if (unlikely(!p)) + return false; + ptr->data = (u8 *)p; + } + ptr->len = len; + return true; +} + +static inline bool +xdrgen_encode_opaque(struct xdr_stream *xdr, opaque val) +{ + __be32 *p = xdr_reserve_space(xdr, XDR_UNIT + xdr_align_size(val.len)); + + if (unlikely(!p)) + return false; + xdr_encode_opaque(p, val.data, val.len); + return true; +} + +#endif /* _SUNRPC_XDRGEN__BUILTINS_H_ */ diff --git a/include/linux/sunrpc/xdrgen/_defs.h b/include/linux/sunrpc/xdrgen/_defs.h new file mode 100644 index 000000000000..be9e62371758 --- /dev/null +++ b/include/linux/sunrpc/xdrgen/_defs.h @@ -0,0 +1,26 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* + * Copyright (c) 2024 Oracle and/or its affiliates. + * + * This header defines XDR data type primitives specified in + * Section 4 of RFC 4506, used by RPC programs implemented + * in the Linux kernel. + */ + +#ifndef _SUNRPC_XDRGEN__DEFS_H_ +#define _SUNRPC_XDRGEN__DEFS_H_ + +#define TRUE (true) +#define FALSE (false) + +typedef struct { + u32 len; + unsigned char *data; +} string; + +typedef struct { + u32 len; + u8 *data; +} opaque; + +#endif /* _SUNRPC_XDRGEN__DEFS_H_ */ diff --git a/include/trace/events/rpcrdma.h b/include/trace/events/rpcrdma.h index a96a985c49b3..e6a72646c507 100644 --- a/include/trace/events/rpcrdma.h +++ b/include/trace/events/rpcrdma.h @@ -2172,6 +2172,29 @@ TRACE_EVENT(svcrdma_qp_error, ) ); +TRACE_EVENT(svcrdma_device_removal, + TP_PROTO( + const struct rdma_cm_id *id + ), + + TP_ARGS(id), + + TP_STRUCT__entry( + __string(name, id->device->name) + __array(unsigned char, addr, sizeof(struct sockaddr_in6)) + ), + + TP_fast_assign( + __assign_str(name); + memcpy(__entry->addr, &id->route.addr.dst_addr, + sizeof(struct sockaddr_in6)); + ), + + TP_printk("device %s to be removed, disconnecting %pISpc\n", + __get_str(name), __entry->addr + ) +); + DECLARE_EVENT_CLASS(svcrdma_sendqueue_class, TP_PROTO( const struct svcxprt_rdma *rdma, diff --git a/include/trace/misc/nfs.h b/include/trace/misc/nfs.h index 7b221d51133a..c82233e950ac 100644 --- a/include/trace/misc/nfs.h +++ b/include/trace/misc/nfs.h @@ -51,6 +51,7 @@ TRACE_DEFINE_ENUM(NFSERR_JUKEBOX); { NFSERR_IO, "IO" }, \ { NFSERR_NXIO, "NXIO" }, \ { ECHILD, "CHILD" }, \ + { ETIMEDOUT, "TIMEDOUT" }, \ { NFSERR_EAGAIN, "AGAIN" }, \ { NFSERR_ACCES, "ACCES" }, \ { NFSERR_EXIST, "EXIST" }, \ diff --git a/include/uapi/linux/pci_regs.h b/include/uapi/linux/pci_regs.h index 94c00996e633..12323b3334a9 100644 --- a/include/uapi/linux/pci_regs.h +++ b/include/uapi/linux/pci_regs.h @@ -634,9 +634,11 @@ #define PCI_EXP_RTCTL_SENFEE 0x0002 /* System Error on Non-Fatal Error */ #define PCI_EXP_RTCTL_SEFEE 0x0004 /* System Error on Fatal Error */ #define PCI_EXP_RTCTL_PMEIE 0x0008 /* PME Interrupt Enable */ -#define PCI_EXP_RTCTL_CRSSVE 0x0010 /* CRS Software Visibility Enable */ +#define PCI_EXP_RTCTL_RRS_SVE 0x0010 /* Config RRS Software Visibility Enable */ +#define PCI_EXP_RTCTL_CRSSVE PCI_EXP_RTCTL_RRS_SVE /* compatibility */ #define PCI_EXP_RTCAP 0x1e /* Root Capabilities */ -#define PCI_EXP_RTCAP_CRSVIS 0x0001 /* CRS Software Visibility capability */ +#define PCI_EXP_RTCAP_RRS_SV 0x0001 /* Config RRS Software Visibility */ +#define PCI_EXP_RTCAP_CRSVIS PCI_EXP_RTCAP_RRS_SV /* compatibility */ #define PCI_EXP_RTSTA 0x20 /* Root Status */ #define PCI_EXP_RTSTA_PME_RQ_ID 0x0000ffff /* PME Requester ID */ #define PCI_EXP_RTSTA_PME 0x00010000 /* PME status */ @@ -740,6 +742,7 @@ #define PCI_EXT_CAP_ID_DVSEC 0x23 /* Designated Vendor-Specific */ #define PCI_EXT_CAP_ID_DLF 0x25 /* Data Link Feature */ #define PCI_EXT_CAP_ID_PL_16GT 0x26 /* Physical Layer 16.0 GT/s */ +#define PCI_EXT_CAP_ID_NPEM 0x29 /* Native PCIe Enclosure Management */ #define PCI_EXT_CAP_ID_PL_32GT 0x2A /* Physical Layer 32.0 GT/s */ #define PCI_EXT_CAP_ID_DOE 0x2E /* Data Object Exchange */ #define PCI_EXT_CAP_ID_MAX PCI_EXT_CAP_ID_DOE @@ -1121,6 +1124,40 @@ #define PCI_PL_16GT_LE_CTRL_USP_TX_PRESET_MASK 0x000000F0 #define PCI_PL_16GT_LE_CTRL_USP_TX_PRESET_SHIFT 4 +/* Native PCIe Enclosure Management */ +#define PCI_NPEM_CAP 0x04 /* NPEM capability register */ +#define PCI_NPEM_CAP_CAPABLE 0x00000001 /* NPEM Capable */ + +#define PCI_NPEM_CTRL 0x08 /* NPEM control register */ +#define PCI_NPEM_CTRL_ENABLE 0x00000001 /* NPEM Enable */ + +/* + * Native PCIe Enclosure Management indication bits and Reset command bit + * are corresponding for capability and control registers. + */ +#define PCI_NPEM_CMD_RESET 0x00000002 /* Reset Command */ +#define PCI_NPEM_IND_OK 0x00000004 /* OK */ +#define PCI_NPEM_IND_LOCATE 0x00000008 /* Locate */ +#define PCI_NPEM_IND_FAIL 0x00000010 /* Fail */ +#define PCI_NPEM_IND_REBUILD 0x00000020 /* Rebuild */ +#define PCI_NPEM_IND_PFA 0x00000040 /* Predicted Failure Analysis */ +#define PCI_NPEM_IND_HOTSPARE 0x00000080 /* Hot Spare */ +#define PCI_NPEM_IND_ICA 0x00000100 /* In Critical Array */ +#define PCI_NPEM_IND_IFA 0x00000200 /* In Failed Array */ +#define PCI_NPEM_IND_IDT 0x00000400 /* Device Type */ +#define PCI_NPEM_IND_DISABLED 0x00000800 /* Disabled */ +#define PCI_NPEM_IND_SPEC_0 0x01000000 +#define PCI_NPEM_IND_SPEC_1 0x02000000 +#define PCI_NPEM_IND_SPEC_2 0x04000000 +#define PCI_NPEM_IND_SPEC_3 0x08000000 +#define PCI_NPEM_IND_SPEC_4 0x10000000 +#define PCI_NPEM_IND_SPEC_5 0x20000000 +#define PCI_NPEM_IND_SPEC_6 0x40000000 +#define PCI_NPEM_IND_SPEC_7 0x80000000 + +#define PCI_NPEM_STATUS 0x0c /* NPEM status register */ +#define PCI_NPEM_STATUS_CC 0x00000001 /* Command Completed */ + /* Data Object Exchange */ #define PCI_DOE_CAP 0x04 /* DOE Capabilities Register */ #define PCI_DOE_CAP_INT_SUP 0x00000001 /* Interrupt Support */ diff --git a/kernel/irq/msi.c b/kernel/irq/msi.c index 1c7e5159064c..3a24d6b5f559 100644 --- a/kernel/irq/msi.c +++ b/kernel/irq/msi.c @@ -832,7 +832,7 @@ static void msi_domain_update_chip_ops(struct msi_domain_info *info) struct irq_chip *chip = info->chip; BUG_ON(!chip || !chip->irq_mask || !chip->irq_unmask); - if (!chip->irq_set_affinity) + if (!chip->irq_set_affinity && !(info->flags & MSI_FLAG_NO_AFFINITY)) chip->irq_set_affinity = msi_domain_set_affinity; } diff --git a/lib/generic-radix-tree.c b/lib/generic-radix-tree.c index fa692c86f069..79e067b51488 100644 --- a/lib/generic-radix-tree.c +++ b/lib/generic-radix-tree.c @@ -5,99 +5,31 @@ #include #include -#define GENRADIX_ARY (GENRADIX_NODE_SIZE / sizeof(struct genradix_node *)) -#define GENRADIX_ARY_SHIFT ilog2(GENRADIX_ARY) - -struct genradix_node { - union { - /* Interior node: */ - struct genradix_node *children[GENRADIX_ARY]; - - /* Leaf: */ - u8 data[GENRADIX_NODE_SIZE]; - }; -}; - -static inline int genradix_depth_shift(unsigned depth) -{ - return GENRADIX_NODE_SHIFT + GENRADIX_ARY_SHIFT * depth; -} - -/* - * Returns size (of data, in bytes) that a tree of a given depth holds: - */ -static inline size_t genradix_depth_size(unsigned depth) -{ - return 1UL << genradix_depth_shift(depth); -} - -/* depth that's needed for a genradix that can address up to ULONG_MAX: */ -#define GENRADIX_MAX_DEPTH \ - DIV_ROUND_UP(BITS_PER_LONG - GENRADIX_NODE_SHIFT, GENRADIX_ARY_SHIFT) - -#define GENRADIX_DEPTH_MASK \ - ((unsigned long) (roundup_pow_of_two(GENRADIX_MAX_DEPTH + 1) - 1)) - -static inline unsigned genradix_root_to_depth(struct genradix_root *r) -{ - return (unsigned long) r & GENRADIX_DEPTH_MASK; -} - -static inline struct genradix_node *genradix_root_to_node(struct genradix_root *r) -{ - return (void *) ((unsigned long) r & ~GENRADIX_DEPTH_MASK); -} - /* * Returns pointer to the specified byte @offset within @radix, or NULL if not * allocated */ void *__genradix_ptr(struct __genradix *radix, size_t offset) { - struct genradix_root *r = READ_ONCE(radix->root); - struct genradix_node *n = genradix_root_to_node(r); - unsigned level = genradix_root_to_depth(r); - - if (ilog2(offset) >= genradix_depth_shift(level)) - return NULL; - - while (1) { - if (!n) - return NULL; - if (!level) - break; - - level--; - - n = n->children[offset >> genradix_depth_shift(level)]; - offset &= genradix_depth_size(level) - 1; - } - - return &n->data[offset]; + return __genradix_ptr_inlined(radix, offset); } EXPORT_SYMBOL(__genradix_ptr); -static inline struct genradix_node *genradix_alloc_node(gfp_t gfp_mask) -{ - return kzalloc(GENRADIX_NODE_SIZE, gfp_mask); -} - -static inline void genradix_free_node(struct genradix_node *node) -{ - kfree(node); -} - /* * Returns pointer to the specified byte @offset within @radix, allocating it if * necessary - newly allocated slots are always zeroed out: */ void *__genradix_ptr_alloc(struct __genradix *radix, size_t offset, + struct genradix_node **preallocated, gfp_t gfp_mask) { struct genradix_root *v = READ_ONCE(radix->root); struct genradix_node *n, *new_node = NULL; unsigned level; + if (preallocated) + swap(new_node, *preallocated); + /* Increase tree depth if necessary: */ while (1) { struct genradix_root *r = v, *new_root; @@ -281,7 +213,7 @@ int __genradix_prealloc(struct __genradix *radix, size_t size, size_t offset; for (offset = 0; offset < size; offset += GENRADIX_NODE_SIZE) - if (!__genradix_ptr_alloc(radix, offset, gfp_mask)) + if (!__genradix_ptr_alloc(radix, offset, NULL, gfp_mask)) return -ENOMEM; return 0; diff --git a/mm/shmem.c b/mm/shmem.c index 6eff8771d9cb..4f11b5506363 100644 --- a/mm/shmem.c +++ b/mm/shmem.c @@ -4961,11 +4961,7 @@ void __init shmem_init(void) shmem_init_inodecache(); #ifdef CONFIG_TMPFS_QUOTA - error = register_quota_format(&shmem_quota_format); - if (error < 0) { - pr_err("Could not register quota format\n"); - goto out3; - } + register_quota_format(&shmem_quota_format); #endif error = register_filesystem(&shmem_fs_type); @@ -5000,7 +4996,6 @@ out1: out2: #ifdef CONFIG_TMPFS_QUOTA unregister_quota_format(&shmem_quota_format); -out3: #endif shmem_destroy_inodecache(); shm_mnt = ERR_PTR(error); diff --git a/net/sunrpc/sunrpc.h b/net/sunrpc/sunrpc.h index d4a362c9e4b3..e3c6e3b63f0b 100644 --- a/net/sunrpc/sunrpc.h +++ b/net/sunrpc/sunrpc.h @@ -36,7 +36,11 @@ static inline int sock_is_loopback(struct sock *sk) return loopback; } +struct svc_serv; +struct svc_rqst; int rpc_clients_notifier_register(void); void rpc_clients_notifier_unregister(void); void auth_domain_cleanup(void); +void svc_sock_update_bufs(struct svc_serv *serv); +enum svc_auth_status svc_authenticate(struct svc_rqst *rqstp); #endif /* _NET_SUNRPC_SUNRPC_H */ diff --git a/net/sunrpc/svc.c b/net/sunrpc/svc.c index 88a59cfa5583..9aff845196ce 100644 --- a/net/sunrpc/svc.c +++ b/net/sunrpc/svc.c @@ -32,6 +32,7 @@ #include #include "fail.h" +#include "sunrpc.h" #define RPCDBG_FACILITY RPCDBG_SVCDSP @@ -417,7 +418,7 @@ struct svc_pool *svc_pool_for_cpu(struct svc_serv *serv) return &serv->sv_pools[pidx % serv->sv_nrpools]; } -int svc_rpcb_setup(struct svc_serv *serv, struct net *net) +static int svc_rpcb_setup(struct svc_serv *serv, struct net *net) { int err; @@ -429,7 +430,6 @@ int svc_rpcb_setup(struct svc_serv *serv, struct net *net) svc_unregister(serv, net); return 0; } -EXPORT_SYMBOL_GPL(svc_rpcb_setup); void svc_rpcb_cleanup(struct svc_serv *serv, struct net *net) { @@ -664,8 +664,21 @@ svc_release_buffer(struct svc_rqst *rqstp) put_page(rqstp->rq_pages[i]); } -struct svc_rqst * -svc_rqst_alloc(struct svc_serv *serv, struct svc_pool *pool, int node) +static void +svc_rqst_free(struct svc_rqst *rqstp) +{ + folio_batch_release(&rqstp->rq_fbatch); + svc_release_buffer(rqstp); + if (rqstp->rq_scratch_page) + put_page(rqstp->rq_scratch_page); + kfree(rqstp->rq_resp); + kfree(rqstp->rq_argp); + kfree(rqstp->rq_auth_data); + kfree_rcu(rqstp, rq_rcu_head); +} + +static struct svc_rqst * +svc_prepare_thread(struct svc_serv *serv, struct svc_pool *pool, int node) { struct svc_rqst *rqstp; @@ -693,27 +706,10 @@ svc_rqst_alloc(struct svc_serv *serv, struct svc_pool *pool, int node) if (!svc_init_buffer(rqstp, serv->sv_max_mesg, node)) goto out_enomem; - return rqstp; -out_enomem: - svc_rqst_free(rqstp); - return NULL; -} -EXPORT_SYMBOL_GPL(svc_rqst_alloc); + rqstp->rq_err = -EAGAIN; /* No error yet */ -static struct svc_rqst * -svc_prepare_thread(struct svc_serv *serv, struct svc_pool *pool, int node) -{ - struct svc_rqst *rqstp; - - rqstp = svc_rqst_alloc(serv, pool, node); - if (!rqstp) - return ERR_PTR(-ENOMEM); - - spin_lock_bh(&serv->sv_lock); serv->sv_nrthreads += 1; - spin_unlock_bh(&serv->sv_lock); - - atomic_inc(&pool->sp_nrthreads); + pool->sp_nrthreads += 1; /* Protected by whatever lock the service uses when calling * svc_set_num_threads() @@ -721,6 +717,10 @@ svc_prepare_thread(struct svc_serv *serv, struct svc_pool *pool, int node) list_add_rcu(&rqstp->rq_all, &pool->sp_all_threads); return rqstp; + +out_enomem: + svc_rqst_free(rqstp); + return NULL; } /** @@ -768,31 +768,22 @@ svc_pool_victim(struct svc_serv *serv, struct svc_pool *target_pool, struct svc_pool *pool; unsigned int i; -retry: pool = target_pool; - if (pool != NULL) { - if (atomic_inc_not_zero(&pool->sp_nrthreads)) - goto found_pool; - return NULL; - } else { + if (!pool) { for (i = 0; i < serv->sv_nrpools; i++) { pool = &serv->sv_pools[--(*state) % serv->sv_nrpools]; - if (atomic_inc_not_zero(&pool->sp_nrthreads)) - goto found_pool; + if (pool->sp_nrthreads) + break; } - return NULL; } -found_pool: - set_bit(SP_VICTIM_REMAINS, &pool->sp_flags); - set_bit(SP_NEED_VICTIM, &pool->sp_flags); - if (!atomic_dec_and_test(&pool->sp_nrthreads)) + if (pool && pool->sp_nrthreads) { + set_bit(SP_VICTIM_REMAINS, &pool->sp_flags); + set_bit(SP_NEED_VICTIM, &pool->sp_flags); return pool; - /* Nothing left in this pool any more */ - clear_bit(SP_NEED_VICTIM, &pool->sp_flags); - clear_bit(SP_VICTIM_REMAINS, &pool->sp_flags); - goto retry; + } + return NULL; } static int @@ -803,6 +794,7 @@ svc_start_kthreads(struct svc_serv *serv, struct svc_pool *pool, int nrservs) struct svc_pool *chosen_pool; unsigned int state = serv->sv_nrthreads-1; int node; + int err; do { nrservs--; @@ -810,8 +802,8 @@ svc_start_kthreads(struct svc_serv *serv, struct svc_pool *pool, int nrservs) node = svc_pool_map_get_node(chosen_pool->sp_id); rqstp = svc_prepare_thread(serv, chosen_pool, node); - if (IS_ERR(rqstp)) - return PTR_ERR(rqstp); + if (!rqstp) + return -ENOMEM; task = kthread_create_on_node(serv->sv_threadfn, rqstp, node, "%s", serv->sv_name); if (IS_ERR(task)) { @@ -825,6 +817,13 @@ svc_start_kthreads(struct svc_serv *serv, struct svc_pool *pool, int nrservs) svc_sock_update_bufs(serv); wake_up_process(task); + + wait_var_event(&rqstp->rq_err, rqstp->rq_err != -EAGAIN); + err = rqstp->rq_err; + if (err) { + svc_exit_thread(rqstp); + return err; + } } while (nrservs > 0); return 0; @@ -871,7 +870,7 @@ svc_set_num_threads(struct svc_serv *serv, struct svc_pool *pool, int nrservs) if (!pool) nrservs -= serv->sv_nrthreads; else - nrservs -= atomic_read(&pool->sp_nrthreads); + nrservs -= pool->sp_nrthreads; if (nrservs > 0) return svc_start_kthreads(serv, pool, nrservs); @@ -933,24 +932,20 @@ void svc_rqst_release_pages(struct svc_rqst *rqstp) } } -/* - * Called from a server thread as it's exiting. Caller must hold the "service - * mutex" for the service. +/** + * svc_exit_thread - finalise the termination of a sunrpc server thread + * @rqstp: the svc_rqst which represents the thread. + * + * When a thread started with svc_new_thread() exits it must call + * svc_exit_thread() as its last act. This must be done with the + * service mutex held. Normally this is held by a DIFFERENT thread, the + * one that is calling svc_set_num_threads() and which will wait for + * SP_VICTIM_REMAINS to be cleared before dropping the mutex. If the + * thread exits for any reason other than svc_thread_should_stop() + * returning %true (which indicated that svc_set_num_threads() is + * waiting for it to exit), then it must take the service mutex itself, + * which can only safely be done using mutex_try_lock(). */ -void -svc_rqst_free(struct svc_rqst *rqstp) -{ - folio_batch_release(&rqstp->rq_fbatch); - svc_release_buffer(rqstp); - if (rqstp->rq_scratch_page) - put_page(rqstp->rq_scratch_page); - kfree(rqstp->rq_resp); - kfree(rqstp->rq_argp); - kfree(rqstp->rq_auth_data); - kfree_rcu(rqstp, rq_rcu_head); -} -EXPORT_SYMBOL_GPL(svc_rqst_free); - void svc_exit_thread(struct svc_rqst *rqstp) { @@ -959,11 +954,8 @@ svc_exit_thread(struct svc_rqst *rqstp) list_del_rcu(&rqstp->rq_all); - atomic_dec(&pool->sp_nrthreads); - - spin_lock_bh(&serv->sv_lock); + pool->sp_nrthreads -= 1; serv->sv_nrthreads -= 1; - spin_unlock_bh(&serv->sv_lock); svc_sock_update_bufs(serv); svc_rqst_free(rqstp); @@ -1098,6 +1090,7 @@ static int __svc_register(struct net *net, const char *progname, return error; } +static int svc_rpcbind_set_version(struct net *net, const struct svc_program *progp, u32 version, int family, @@ -1108,7 +1101,6 @@ int svc_rpcbind_set_version(struct net *net, version, family, proto, port); } -EXPORT_SYMBOL_GPL(svc_rpcbind_set_version); int svc_generic_rpcbind_set(struct net *net, const struct svc_program *progp, @@ -1526,6 +1518,14 @@ err_system_err: goto sendit; } +/* + * Drop request + */ +static void svc_drop(struct svc_rqst *rqstp) +{ + trace_svc_drop(rqstp); +} + /** * svc_process - Execute one RPC transaction * @rqstp: RPC transaction context diff --git a/net/sunrpc/svc_xprt.c b/net/sunrpc/svc_xprt.c index d3735ab3e6d1..53ebc719ff5a 100644 --- a/net/sunrpc/svc_xprt.c +++ b/net/sunrpc/svc_xprt.c @@ -905,15 +905,6 @@ void svc_recv(struct svc_rqst *rqstp) } EXPORT_SYMBOL_GPL(svc_recv); -/* - * Drop request - */ -void svc_drop(struct svc_rqst *rqstp) -{ - trace_svc_drop(rqstp); -} -EXPORT_SYMBOL_GPL(svc_drop); - /** * svc_send - Return reply to client * @rqstp: RPC transaction context diff --git a/net/sunrpc/svcauth.c b/net/sunrpc/svcauth.c index 1619211f0960..93d9e949e265 100644 --- a/net/sunrpc/svcauth.c +++ b/net/sunrpc/svcauth.c @@ -98,7 +98,6 @@ enum svc_auth_status svc_authenticate(struct svc_rqst *rqstp) rqstp->rq_authop = aops; return aops->accept(rqstp); } -EXPORT_SYMBOL_GPL(svc_authenticate); /** * svc_set_client - Assign an appropriate 'auth_domain' as the client diff --git a/net/sunrpc/svcsock.c b/net/sunrpc/svcsock.c index 6b3f01beb294..825ec5357691 100644 --- a/net/sunrpc/svcsock.c +++ b/net/sunrpc/svcsock.c @@ -1378,7 +1378,6 @@ void svc_sock_update_bufs(struct svc_serv *serv) set_bit(XPT_CHNGBUF, &svsk->sk_xprt.xpt_flags); spin_unlock_bh(&serv->sv_lock); } -EXPORT_SYMBOL_GPL(svc_sock_update_bufs); /* * Initialize socket for RPC use and create svc_sock struct diff --git a/net/sunrpc/xprtrdma/svc_rdma_transport.c b/net/sunrpc/xprtrdma/svc_rdma_transport.c index f15750cacacf..c3fbf0779d4a 100644 --- a/net/sunrpc/xprtrdma/svc_rdma_transport.c +++ b/net/sunrpc/xprtrdma/svc_rdma_transport.c @@ -339,7 +339,6 @@ static int svc_rdma_cma_handler(struct rdma_cm_id *cma_id, svc_xprt_enqueue(xprt); break; case RDMA_CM_EVENT_DISCONNECTED: - case RDMA_CM_EVENT_DEVICE_REMOVAL: svc_xprt_deferred_close(xprt); break; default: @@ -370,7 +369,7 @@ static struct svc_xprt *svc_rdma_create(struct svc_serv *serv, listen_id = svc_rdma_create_listen_id(net, sa, cma_xprt); if (IS_ERR(listen_id)) { kfree(cma_xprt); - return (struct svc_xprt *)listen_id; + return ERR_CAST(listen_id); } cma_xprt->sc_cm_id = listen_id; @@ -384,6 +383,16 @@ static struct svc_xprt *svc_rdma_create(struct svc_serv *serv, return &cma_xprt->sc_xprt; } +static void svc_rdma_xprt_done(struct rpcrdma_notification *rn) +{ + struct svcxprt_rdma *rdma = container_of(rn, struct svcxprt_rdma, + sc_rn); + struct rdma_cm_id *id = rdma->sc_cm_id; + + trace_svcrdma_device_removal(id); + svc_xprt_close(&rdma->sc_xprt); +} + /* * This is the xpo_recvfrom function for listening endpoints. Its * purpose is to accept incoming connections. The CMA callback handler @@ -425,6 +434,9 @@ static struct svc_xprt *svc_rdma_accept(struct svc_xprt *xprt) dev = newxprt->sc_cm_id->device; newxprt->sc_port_num = newxprt->sc_cm_id->port_num; + if (rpcrdma_rn_register(dev, &newxprt->sc_rn, svc_rdma_xprt_done)) + goto errout; + newxprt->sc_max_req_size = svcrdma_max_req_size; newxprt->sc_max_requests = svcrdma_max_requests; newxprt->sc_max_bc_requests = svcrdma_max_bc_requests; @@ -580,6 +592,7 @@ static void __svc_rdma_free(struct work_struct *work) { struct svcxprt_rdma *rdma = container_of(work, struct svcxprt_rdma, sc_work); + struct ib_device *device = rdma->sc_cm_id->device; /* This blocks until the Completion Queues are empty */ if (rdma->sc_qp && !IS_ERR(rdma->sc_qp)) @@ -608,6 +621,7 @@ static void __svc_rdma_free(struct work_struct *work) /* Destroy the CM ID */ rdma_destroy_id(rdma->sc_cm_id); + rpcrdma_rn_unregister(device, &rdma->sc_rn); kfree(rdma); } diff --git a/sound/pci/hda/hda_intel.c b/sound/pci/hda/hda_intel.c index edeaf3ee273c..045cd555c291 100644 --- a/sound/pci/hda/hda_intel.c +++ b/sound/pci/hda/hda_intel.c @@ -2688,7 +2688,7 @@ static const struct pci_device_id azx_ids[] = { .driver_data = AZX_DRIVER_ATIHDMI_NS | AZX_DCAPS_PRESET_ATI_HDMI_NS | AZX_DCAPS_PM_RUNTIME }, /* GLENFLY */ - { PCI_DEVICE(0x6766, PCI_ANY_ID), + { PCI_DEVICE(PCI_VENDOR_ID_GLENFLY, PCI_ANY_ID), .class = PCI_CLASS_MULTIMEDIA_HD_AUDIO << 8, .class_mask = 0xffffff, .driver_data = AZX_DRIVER_GFHDMI | AZX_DCAPS_POSFIX_LPIB | diff --git a/tools/net/sunrpc/xdrgen/.gitignore b/tools/net/sunrpc/xdrgen/.gitignore new file mode 100644 index 000000000000..d7366c2f9be8 --- /dev/null +++ b/tools/net/sunrpc/xdrgen/.gitignore @@ -0,0 +1,2 @@ +__pycache__ +generators/__pycache__ diff --git a/tools/net/sunrpc/xdrgen/README b/tools/net/sunrpc/xdrgen/README new file mode 100644 index 000000000000..92f7738ad50c --- /dev/null +++ b/tools/net/sunrpc/xdrgen/README @@ -0,0 +1,244 @@ +xdrgen - Linux Kernel XDR code generator + +Introduction +------------ + +SunRPC programs are typically specified using a language defined by +RFC 4506. In fact, all IETF-published NFS specifications provide a +description of the specified protocol using this language. + +Since the 1990's, user space consumers of SunRPC have had access to +a tool that could read such XDR specifications and then generate C +code that implements the RPC portions of that protocol. This tool is +called rpcgen. + +This RPC-level code is code that handles input directly from the +network, and thus a high degree of memory safety and sanity checking +is needed to help ensure proper levels of security. Bugs in this +code can have significant impact on security and performance. + +However, it is code that is repetitive and tedious to write by hand. + +The C code generated by rpcgen makes extensive use of the facilities +of the user space TI-RPC library and libc. Furthermore, the dialect +of the generated code is very traditional K&R C. + +The Linux kernel's implementation of SunRPC-based protocols hand-roll +their XDR implementation. There are two main reasons for this: + +1. libtirpc (and its predecessors) operate only in user space. The + kernel's RPC implementation and its API are significantly + different than libtirpc. + +2. rpcgen-generated code is believed to be less efficient than code + that is hand-written. + +These days, gcc and its kin are capable of optimizing code better +than human authors. There are only a few instances where writing +XDR code by hand will make a measurable performance different. + +In addition, the current hand-written code in the Linux kernel is +difficult to audit and prove that it implements exactly what is in +the protocol specification. + +In order to accrue the benefits of machine-generated XDR code in the +kernel, a tool is needed that will output C code that works against +the kernel's SunRPC implementation rather than libtirpc. + +Enter xdrgen. + + +Dependencies +------------ + +These dependencies are typically packaged by Linux distributions: + +- python3 +- python3-lark +- python3-jinja2 + +These dependencies are available via PyPi: + +- pip install 'lark[interegular]' + + +XDR Specifications +------------------ + +When adding a new protocol implementation to the kernel, the XDR +specification can be derived by feeding a .txt copy of the RFC to +the script located in tools/net/sunrpc/extract.sh. + + $ extract.sh < rfc0001.txt > new2.x + + +Operation +--------- + +Once a .x file is available, use xdrgen to generate source and +header files containing an implementation of XDR encoding and +decoding functions for the specified protocol. + + $ ./xdrgen definitions new2.x > include/linux/sunrpc/xdrgen/new2.h + $ ./xdrgen declarations new2.x > new2xdr_gen.h + +and + + $ ./xdrgen source new2.x > new2xdr_gen.c + +The files are ready to use for a server-side protocol implementation, +or may be used as a guide for implementing these routines by hand. + +By default, the only comments added to this code are kdoc comments +that appear directly in front of the public per-procedure APIs. For +deeper introspection, specifying the "--annotate" flag will insert +additional comments in the generated code to help readers match the +generated code to specific parts of the XDR specification. + +Because the generated code is targeted for the Linux kernel, it +is tagged with a GPLv2-only license. + +The xdrgen tool can also provide lexical and syntax checking of +an XDR specification: + + $ ./xdrgen lint xdr/new.x + + +How It Works +------------ + +xdrgen does not use machine learning to generate source code. The +translation is entirely deterministic. + +RFC 4506 Section 6 contains a BNF grammar of the XDR specification +language. The grammar has been adapted for use by the Python Lark +module. + +The xdr.ebnf file in this directory contains the grammar used to +parse XDR specifications. xdrgen configures Lark using the grammar +in xdr.ebnf. Lark parses the target XDR specification using this +grammar, creating a parse tree. + +xdrgen then transforms the parse tree into an abstract syntax tree. +This tree is passed to a series of code generators. + +The generators are implemented as Python classes residing in the +generators/ directory. Each generator emits code created from Jinja2 +templates stored in the templates/ directory. + +The source code is generated in the same order in which they appear +in the specification to ensure the generated code compiles. This +conforms with the behavior of rpcgen. + +xdrgen assumes that the generated source code is further compiled by +a compiler that can optimize in a number of ways, including: + + - Unused functions are discarded (ie, not added to the executable) + + - Aggressive function inlining removes unnecessary stack frames + + - Single-arm switch statements are replaced by a single conditional + branch + +And so on. + + +Pragmas +------- + +Pragma directives specify exceptions to the normal generation of +encoding and decoding functions. Currently one directive is +implemented: "public". + +Pragma exclude +------ ------- + + pragma exclude ; + +In some cases, a procedure encoder or decoder function might need +special processing that cannot be automatically generated. The +automatically-generated functions might conflict or interfere with +the hand-rolled function. To avoid editing the generated source code +by hand, a pragma can specify that the procedure's encoder and +decoder functions are not included in the generated header and +source. + +For example: + + pragma exclude NFSPROC3_READDIRPLUS; + +Excludes the decoder function for the READDIRPLUS argument and the +encoder function for the READDIRPLUS result. + +Note that because data item encoder and decoder functions are +defined "static __maybe_unused", subsequent compilation +automatically excludes data item encoder and decoder functions that +are used only by excluded procedure. + +Pragma header +------ ------ + + pragma header ; + +Provide a name to use for the header file. For example: + + pragma header nlm4; + +Adds + + #include "nlm4xdr_gen.h" + +to the generated source file. + +Pragma public +------ ------ + + pragma public ; + +Normally XDR encoder and decoder functions are "static". In case an +implementer wants to call these functions from other source code, +s/he can add a public pragma in the input .x file to indicate a set +of functions that should get a prototype in the generated header, +and the function definitions will not be declared static. + +For example: + + pragma public nfsstat3; + +Adds these prototypes in the generated header: + + bool xdrgen_decode_nfsstat3(struct xdr_stream *xdr, enum nfsstat3 *ptr); + bool xdrgen_encode_nfsstat3(struct xdr_stream *xdr, enum nfsstat3 value); + +And, in the generated source code, both of these functions appear +without the "static __maybe_unused" modifiers. + + +Future Work +----------- + +Finish implementing XDR pointer and list types. + +Generate client-side procedure functions + +Expand the README into a user guide similar to rpcgen(1) + +Add more pragma directives: + + * @pages -- use xdr_read/write_pages() for the specified opaque + field + * @skip -- do not decode, but rather skip, the specified argument + field + +Enable something like a #include to dynamically insert the content +of other specification files + +Properly support line-by-line pass-through via the "%" decorator + +Build a unit test suite for verifying translation of XDR language +into compilable code + +Add a command-line option to insert trace_printk call sites in the +generated source code, for improved (temporary) observability + +Generate kernel Rust code as well as C code diff --git a/tools/net/sunrpc/xdrgen/__init__.py b/tools/net/sunrpc/xdrgen/__init__.py new file mode 100644 index 000000000000..c940e9275252 --- /dev/null +++ b/tools/net/sunrpc/xdrgen/__init__.py @@ -0,0 +1,2 @@ +# SPDX-License-Identifier: GPL-2.0 +# Just to make sphinx-apidoc document this directory diff --git a/tools/net/sunrpc/xdrgen/generators/__init__.py b/tools/net/sunrpc/xdrgen/generators/__init__.py new file mode 100644 index 000000000000..fd2457461274 --- /dev/null +++ b/tools/net/sunrpc/xdrgen/generators/__init__.py @@ -0,0 +1,113 @@ +# SPDX-License-Identifier: GPL-2.0 + +"""Define a base code generator class""" + +import sys +from jinja2 import Environment, FileSystemLoader, Template + +from xdr_ast import _XdrAst, Specification, _RpcProgram, _XdrTypeSpecifier +from xdr_ast import public_apis, pass_by_reference, get_header_name +from xdr_parse import get_xdr_annotate + + +def create_jinja2_environment(language: str, xdr_type: str) -> Environment: + """Open a set of templates based on output language""" + match language: + case "C": + environment = Environment( + loader=FileSystemLoader(sys.path[0] + "/templates/C/" + xdr_type + "/"), + trim_blocks=True, + lstrip_blocks=True, + ) + environment.globals["annotate"] = get_xdr_annotate() + environment.globals["public_apis"] = public_apis + environment.globals["pass_by_reference"] = pass_by_reference + return environment + case _: + raise NotImplementedError("Language not supported") + + +def get_jinja2_template( + environment: Environment, template_type: str, template_name: str +) -> Template: + """Retrieve a Jinja2 template for emitting source code""" + return environment.get_template(template_type + "/" + template_name + ".j2") + + +def find_xdr_program_name(root: Specification) -> str: + """Retrieve the RPC program name from an abstract syntax tree""" + raw_name = get_header_name() + if raw_name != "none": + return raw_name.lower() + for definition in root.definitions: + if isinstance(definition.value, _RpcProgram): + raw_name = definition.value.name + return raw_name.lower().removesuffix("_program").removesuffix("_prog") + return "noprog" + + +def header_guard_infix(filename: str) -> str: + """Extract the header guard infix from the specification filename""" + basename = filename.split("/")[-1] + program = basename.replace(".x", "") + return program.upper() + + +def kernel_c_type(spec: _XdrTypeSpecifier) -> str: + """Return name of C type""" + builtin_native_c_type = { + "bool": "bool", + "int": "s32", + "unsigned_int": "u32", + "long": "s32", + "unsigned_long": "u32", + "hyper": "s64", + "unsigned_hyper": "u64", + } + if spec.type_name in builtin_native_c_type: + return builtin_native_c_type[spec.type_name] + return spec.type_name + + +class Boilerplate: + """Base class to generate boilerplate for source files""" + + def __init__(self, language: str, peer: str): + """Initialize an instance of this class""" + raise NotImplementedError("No language support defined") + + def emit_declaration(self, filename: str, root: Specification) -> None: + """Emit declaration header boilerplate""" + raise NotImplementedError("Header boilerplate generation not supported") + + def emit_definition(self, filename: str, root: Specification) -> None: + """Emit definition header boilerplate""" + raise NotImplementedError("Header boilerplate generation not supported") + + def emit_source(self, filename: str, root: Specification) -> None: + """Emit generic source code for this XDR type""" + raise NotImplementedError("Source boilerplate generation not supported") + + +class SourceGenerator: + """Base class to generate header and source code for XDR types""" + + def __init__(self, language: str, peer: str): + """Initialize an instance of this class""" + raise NotImplementedError("No language support defined") + + def emit_declaration(self, node: _XdrAst) -> None: + """Emit one function declaration for this XDR type""" + raise NotImplementedError("Declaration generation not supported") + + def emit_decoder(self, node: _XdrAst) -> None: + """Emit one decoder function for this XDR type""" + raise NotImplementedError("Decoder generation not supported") + + def emit_definition(self, node: _XdrAst) -> None: + """Emit one definition for this XDR type""" + raise NotImplementedError("Definition generation not supported") + + def emit_encoder(self, node: _XdrAst) -> None: + """Emit one encoder function for this XDR type""" + raise NotImplementedError("Encoder generation not supported") diff --git a/tools/net/sunrpc/xdrgen/generators/constant.py b/tools/net/sunrpc/xdrgen/generators/constant.py new file mode 100644 index 000000000000..f2339caf0953 --- /dev/null +++ b/tools/net/sunrpc/xdrgen/generators/constant.py @@ -0,0 +1,20 @@ +#!/usr/bin/env python3 +# ex: set filetype=python: + +"""Generate code to handle XDR constants""" + +from generators import SourceGenerator, create_jinja2_environment +from xdr_ast import _XdrConstant + +class XdrConstantGenerator(SourceGenerator): + """Generate source code for XDR constants""" + + def __init__(self, language: str, peer: str): + """Initialize an instance of this class""" + self.environment = create_jinja2_environment(language, "constants") + self.peer = peer + + def emit_definition(self, node: _XdrConstant) -> None: + """Emit one definition for a constant""" + template = self.environment.get_template("definition.j2") + print(template.render(name=node.name, value=node.value)) diff --git a/tools/net/sunrpc/xdrgen/generators/enum.py b/tools/net/sunrpc/xdrgen/generators/enum.py new file mode 100644 index 000000000000..855e43f4ae38 --- /dev/null +++ b/tools/net/sunrpc/xdrgen/generators/enum.py @@ -0,0 +1,44 @@ +#!/usr/bin/env python3 +# ex: set filetype=python: + +"""Generate code to handle XDR enum types""" + +from generators import SourceGenerator, create_jinja2_environment +from xdr_ast import _XdrEnum, public_apis + + +class XdrEnumGenerator(SourceGenerator): + """Generate source code for XDR enum types""" + + def __init__(self, language: str, peer: str): + """Initialize an instance of this class""" + self.environment = create_jinja2_environment(language, "enum") + self.peer = peer + + def emit_declaration(self, node: _XdrEnum) -> None: + """Emit one declaration pair for an XDR enum type""" + if node.name in public_apis: + template = self.environment.get_template("declaration/close.j2") + print(template.render(name=node.name)) + + def emit_definition(self, node: _XdrEnum) -> None: + """Emit one definition for an XDR enum type""" + template = self.environment.get_template("definition/open.j2") + print(template.render(name=node.name)) + + template = self.environment.get_template("definition/enumerator.j2") + for enumerator in node.enumerators: + print(template.render(name=enumerator.name, value=enumerator.value)) + + template = self.environment.get_template("definition/close.j2") + print(template.render(name=node.name)) + + def emit_decoder(self, node: _XdrEnum) -> None: + """Emit one decoder function for an XDR enum type""" + template = self.environment.get_template("decoder/enum.j2") + print(template.render(name=node.name)) + + def emit_encoder(self, node: _XdrEnum) -> None: + """Emit one encoder function for an XDR enum type""" + template = self.environment.get_template("encoder/enum.j2") + print(template.render(name=node.name)) diff --git a/tools/net/sunrpc/xdrgen/generators/header_bottom.py b/tools/net/sunrpc/xdrgen/generators/header_bottom.py new file mode 100644 index 000000000000..4b55b282dfc0 --- /dev/null +++ b/tools/net/sunrpc/xdrgen/generators/header_bottom.py @@ -0,0 +1,33 @@ +#!/usr/bin/env python3 +# ex: set filetype=python: + +"""Generate header bottom boilerplate""" + +import os.path +import time + +from generators import Boilerplate, header_guard_infix +from generators import create_jinja2_environment, get_jinja2_template +from xdr_ast import Specification + + +class XdrHeaderBottomGenerator(Boilerplate): + """Generate header boilerplate""" + + def __init__(self, language: str, peer: str): + """Initialize an instance of this class""" + self.environment = create_jinja2_environment(language, "header_bottom") + self.peer = peer + + def emit_declaration(self, filename: str, root: Specification) -> None: + """Emit the bottom header guard""" + template = get_jinja2_template(self.environment, "declaration", "header") + print(template.render(infix=header_guard_infix(filename))) + + def emit_definition(self, filename: str, root: Specification) -> None: + """Emit the bottom header guard""" + template = get_jinja2_template(self.environment, "definition", "header") + print(template.render(infix=header_guard_infix(filename))) + + def emit_source(self, filename: str, root: Specification) -> None: + pass diff --git a/tools/net/sunrpc/xdrgen/generators/header_top.py b/tools/net/sunrpc/xdrgen/generators/header_top.py new file mode 100644 index 000000000000..c6bc21c71f19 --- /dev/null +++ b/tools/net/sunrpc/xdrgen/generators/header_top.py @@ -0,0 +1,45 @@ +#!/usr/bin/env python3 +# ex: set filetype=python: + +"""Generate header top boilerplate""" + +import os.path +import time + +from generators import Boilerplate, header_guard_infix +from generators import create_jinja2_environment, get_jinja2_template +from xdr_ast import Specification + + +class XdrHeaderTopGenerator(Boilerplate): + """Generate header boilerplate""" + + def __init__(self, language: str, peer: str): + """Initialize an instance of this class""" + self.environment = create_jinja2_environment(language, "header_top") + self.peer = peer + + def emit_declaration(self, filename: str, root: Specification) -> None: + """Emit the top header guard""" + template = get_jinja2_template(self.environment, "declaration", "header") + print( + template.render( + infix=header_guard_infix(filename), + filename=filename, + mtime=time.ctime(os.path.getmtime(filename)), + ) + ) + + def emit_definition(self, filename: str, root: Specification) -> None: + """Emit the top header guard""" + template = get_jinja2_template(self.environment, "definition", "header") + print( + template.render( + infix=header_guard_infix(filename), + filename=filename, + mtime=time.ctime(os.path.getmtime(filename)), + ) + ) + + def emit_source(self, filename: str, root: Specification) -> None: + pass diff --git a/tools/net/sunrpc/xdrgen/generators/pointer.py b/tools/net/sunrpc/xdrgen/generators/pointer.py new file mode 100644 index 000000000000..b0b27f1819c8 --- /dev/null +++ b/tools/net/sunrpc/xdrgen/generators/pointer.py @@ -0,0 +1,272 @@ +#!/usr/bin/env python3 +# ex: set filetype=python: + +"""Generate code to handle XDR pointer types""" + +from jinja2 import Environment + +from generators import SourceGenerator, kernel_c_type +from generators import create_jinja2_environment, get_jinja2_template + +from xdr_ast import _XdrBasic, _XdrVariableLengthString +from xdr_ast import _XdrFixedLengthOpaque, _XdrVariableLengthOpaque +from xdr_ast import _XdrFixedLengthArray, _XdrVariableLengthArray +from xdr_ast import _XdrOptionalData, _XdrPointer, _XdrDeclaration +from xdr_ast import public_apis + + +def emit_pointer_declaration(environment: Environment, node: _XdrPointer) -> None: + """Emit a declaration pair for an XDR pointer type""" + if node.name in public_apis: + template = get_jinja2_template(environment, "declaration", "close") + print(template.render(name=node.name)) + + +def emit_pointer_member_definition( + environment: Environment, field: _XdrDeclaration +) -> None: + """Emit a definition for one field in an XDR struct""" + if isinstance(field, _XdrBasic): + template = get_jinja2_template(environment, "definition", field.template) + print( + template.render( + name=field.name, + type=kernel_c_type(field.spec), + classifier=field.spec.c_classifier, + ) + ) + elif isinstance(field, _XdrFixedLengthOpaque): + template = get_jinja2_template(environment, "definition", field.template) + print( + template.render( + name=field.name, + size=field.size, + ) + ) + elif isinstance(field, _XdrVariableLengthOpaque): + template = get_jinja2_template(environment, "definition", field.template) + print(template.render(name=field.name)) + elif isinstance(field, _XdrVariableLengthString): + template = get_jinja2_template(environment, "definition", field.template) + print(template.render(name=field.name)) + elif isinstance(field, _XdrFixedLengthArray): + template = get_jinja2_template(environment, "definition", field.template) + print( + template.render( + name=field.name, + type=kernel_c_type(field.spec), + size=field.size, + ) + ) + elif isinstance(field, _XdrVariableLengthArray): + template = get_jinja2_template(environment, "definition", field.template) + print( + template.render( + name=field.name, + type=kernel_c_type(field.spec), + classifier=field.spec.c_classifier, + ) + ) + elif isinstance(field, _XdrOptionalData): + template = get_jinja2_template(environment, "definition", field.template) + print( + template.render( + name=field.name, + type=kernel_c_type(field.spec), + classifier=field.spec.c_classifier, + ) + ) + + +def emit_pointer_definition(environment: Environment, node: _XdrPointer) -> None: + """Emit a definition for an XDR pointer type""" + template = get_jinja2_template(environment, "definition", "open") + print(template.render(name=node.name)) + + for field in node.fields[0:-1]: + emit_pointer_member_definition(environment, field) + + template = get_jinja2_template(environment, "definition", "close") + print(template.render(name=node.name)) + + +def emit_pointer_member_decoder( + environment: Environment, field: _XdrDeclaration +) -> None: + """Emit a decoder for one field in an XDR pointer""" + if isinstance(field, _XdrBasic): + template = get_jinja2_template(environment, "decoder", field.template) + print( + template.render( + name=field.name, + type=field.spec.type_name, + classifier=field.spec.c_classifier, + ) + ) + elif isinstance(field, _XdrFixedLengthOpaque): + template = get_jinja2_template(environment, "decoder", field.template) + print( + template.render( + name=field.name, + size=field.size, + ) + ) + elif isinstance(field, _XdrVariableLengthOpaque): + template = get_jinja2_template(environment, "decoder", field.template) + print( + template.render( + name=field.name, + maxsize=field.maxsize, + ) + ) + elif isinstance(field, _XdrVariableLengthString): + template = get_jinja2_template(environment, "decoder", field.template) + print( + template.render( + name=field.name, + maxsize=field.maxsize, + ) + ) + elif isinstance(field, _XdrFixedLengthArray): + template = get_jinja2_template(environment, "decoder", field.template) + print( + template.render( + name=field.name, + type=field.spec.type_name, + size=field.size, + classifier=field.spec.c_classifier, + ) + ) + elif isinstance(field, _XdrVariableLengthArray): + template = get_jinja2_template(environment, "decoder", field.template) + print( + template.render( + name=field.name, + type=field.spec.type_name, + maxsize=field.maxsize, + classifier=field.spec.c_classifier, + ) + ) + elif isinstance(field, _XdrOptionalData): + template = get_jinja2_template(environment, "decoder", field.template) + print( + template.render( + name=field.name, + type=field.spec.type_name, + classifier=field.spec.c_classifier, + ) + ) + + +def emit_pointer_decoder(environment: Environment, node: _XdrPointer) -> None: + """Emit one decoder function for an XDR pointer type""" + template = get_jinja2_template(environment, "decoder", "open") + print(template.render(name=node.name)) + + for field in node.fields[0:-1]: + emit_pointer_member_decoder(environment, field) + + template = get_jinja2_template(environment, "decoder", "close") + print(template.render()) + + +def emit_pointer_member_encoder( + environment: Environment, field: _XdrDeclaration +) -> None: + """Emit an encoder for one field in a XDR pointer""" + if isinstance(field, _XdrBasic): + template = get_jinja2_template(environment, "encoder", field.template) + print( + template.render( + name=field.name, + type=field.spec.type_name, + ) + ) + elif isinstance(field, _XdrFixedLengthOpaque): + template = get_jinja2_template(environment, "encoder", field.template) + print( + template.render( + name=field.name, + size=field.size, + ) + ) + elif isinstance(field, _XdrVariableLengthOpaque): + template = get_jinja2_template(environment, "encoder", field.template) + print( + template.render( + name=field.name, + maxsize=field.maxsize, + ) + ) + elif isinstance(field, _XdrVariableLengthString): + template = get_jinja2_template(environment, "encoder", field.template) + print( + template.render( + name=field.name, + maxsize=field.maxsize, + ) + ) + elif isinstance(field, _XdrFixedLengthArray): + template = get_jinja2_template(environment, "encoder", field.template) + print( + template.render( + name=field.name, + type=field.spec.type_name, + size=field.size, + ) + ) + elif isinstance(field, _XdrVariableLengthArray): + template = get_jinja2_template(environment, "encoder", field.template) + print( + template.render( + name=field.name, + type=field.spec.type_name, + maxsize=field.maxsize, + ) + ) + elif isinstance(field, _XdrOptionalData): + template = get_jinja2_template(environment, "encoder", field.template) + print( + template.render( + name=field.name, + type=field.spec.type_name, + classifier=field.spec.c_classifier, + ) + ) + + +def emit_pointer_encoder(environment: Environment, node: _XdrPointer) -> None: + """Emit one encoder function for an XDR pointer type""" + template = get_jinja2_template(environment, "encoder", "open") + print(template.render(name=node.name)) + + for field in node.fields[0:-1]: + emit_pointer_member_encoder(environment, field) + + template = get_jinja2_template(environment, "encoder", "close") + print(template.render()) + + +class XdrPointerGenerator(SourceGenerator): + """Generate source code for XDR pointer""" + + def __init__(self, language: str, peer: str): + """Initialize an instance of this class""" + self.environment = create_jinja2_environment(language, "pointer") + self.peer = peer + + def emit_declaration(self, node: _XdrPointer) -> None: + """Emit one declaration pair for an XDR pointer type""" + emit_pointer_declaration(self.environment, node) + + def emit_definition(self, node: _XdrPointer) -> None: + """Emit one declaration for an XDR pointer type""" + emit_pointer_definition(self.environment, node) + + def emit_decoder(self, node: _XdrPointer) -> None: + """Emit one decoder function for an XDR pointer type""" + emit_pointer_decoder(self.environment, node) + + def emit_encoder(self, node: _XdrPointer) -> None: + """Emit one encoder function for an XDR pointer type""" + emit_pointer_encoder(self.environment, node) diff --git a/tools/net/sunrpc/xdrgen/generators/program.py b/tools/net/sunrpc/xdrgen/generators/program.py new file mode 100644 index 000000000000..ac3cf1694b68 --- /dev/null +++ b/tools/net/sunrpc/xdrgen/generators/program.py @@ -0,0 +1,168 @@ +#!/usr/bin/env python3 +# ex: set filetype=python: + +"""Generate code for an RPC program's procedures""" + +from jinja2 import Environment + +from generators import SourceGenerator, create_jinja2_environment +from xdr_ast import _RpcProgram, _RpcVersion, excluded_apis + + +def emit_version_definitions( + environment: Environment, program: str, version: _RpcVersion +) -> None: + """Emit procedure numbers for each RPC version's procedures""" + template = environment.get_template("definition/open.j2") + print(template.render(program=program.upper())) + + template = environment.get_template("definition/procedure.j2") + for procedure in version.procedures: + if procedure.name not in excluded_apis: + print( + template.render( + name=procedure.name, + value=procedure.number, + ) + ) + + template = environment.get_template("definition/close.j2") + print(template.render()) + + +def emit_version_declarations( + environment: Environment, program: str, version: _RpcVersion +) -> None: + """Emit declarations for each RPC version's procedures""" + arguments = dict.fromkeys([]) + for procedure in version.procedures: + if procedure.name not in excluded_apis: + arguments[procedure.argument.type_name] = None + if len(arguments) > 0: + print("") + template = environment.get_template("declaration/argument.j2") + for argument in arguments: + print(template.render(program=program, argument=argument)) + + results = dict.fromkeys([]) + for procedure in version.procedures: + if procedure.name not in excluded_apis: + results[procedure.result.type_name] = None + if len(results) > 0: + print("") + template = environment.get_template("declaration/result.j2") + for result in results: + print(template.render(program=program, result=result)) + + +def emit_version_argument_decoders( + environment: Environment, program: str, version: _RpcVersion +) -> None: + """Emit server argument decoders for each RPC version's procedures""" + arguments = dict.fromkeys([]) + for procedure in version.procedures: + if procedure.name not in excluded_apis: + arguments[procedure.argument.type_name] = None + + template = environment.get_template("decoder/argument.j2") + for argument in arguments: + print(template.render(program=program, argument=argument)) + + +def emit_version_result_decoders( + environment: Environment, program: str, version: _RpcVersion +) -> None: + """Emit client result decoders for each RPC version's procedures""" + results = dict.fromkeys([]) + for procedure in version.procedures: + if procedure.name not in excluded_apis: + results[procedure.result.type_name] = None + + template = environment.get_template("decoder/result.j2") + for result in results: + print(template.render(program=program, result=result)) + + +def emit_version_argument_encoders( + environment: Environment, program: str, version: _RpcVersion +) -> None: + """Emit client argument encoders for each RPC version's procedures""" + arguments = dict.fromkeys([]) + for procedure in version.procedures: + if procedure.name not in excluded_apis: + arguments[procedure.argument.type_name] = None + + template = environment.get_template("encoder/argument.j2") + for argument in arguments: + print(template.render(program=program, argument=argument)) + + +def emit_version_result_encoders( + environment: Environment, program: str, version: _RpcVersion +) -> None: + """Emit server result encoders for each RPC version's procedures""" + results = dict.fromkeys([]) + for procedure in version.procedures: + if procedure.name not in excluded_apis: + results[procedure.result.type_name] = None + + template = environment.get_template("encoder/result.j2") + for result in results: + print(template.render(program=program, result=result)) + + +class XdrProgramGenerator(SourceGenerator): + """Generate source code for an RPC program's procedures""" + + def __init__(self, language: str, peer: str): + """Initialize an instance of this class""" + self.environment = create_jinja2_environment(language, "program") + self.peer = peer + + def emit_definition(self, node: _RpcProgram) -> None: + """Emit procedure numbers for each of an RPC programs's procedures""" + raw_name = node.name + program = raw_name.lower().removesuffix("_program").removesuffix("_prog") + + for version in node.versions: + emit_version_definitions(self.environment, program, version) + + def emit_declaration(self, node: _RpcProgram) -> None: + """Emit a declaration pair for each of an RPC programs's procedures""" + raw_name = node.name + program = raw_name.lower().removesuffix("_program").removesuffix("_prog") + + for version in node.versions: + emit_version_declarations(self.environment, program, version) + + def emit_decoder(self, node: _RpcProgram) -> None: + """Emit all decoder functions for an RPC program's procedures""" + raw_name = node.name + program = raw_name.lower().removesuffix("_program").removesuffix("_prog") + match self.peer: + case "server": + for version in node.versions: + emit_version_argument_decoders( + self.environment, program, version, + ) + case "client": + for version in node.versions: + emit_version_result_decoders( + self.environment, program, version, + ) + + def emit_encoder(self, node: _RpcProgram) -> None: + """Emit all encoder functions for an RPC program's procedures""" + raw_name = node.name + program = raw_name.lower().removesuffix("_program").removesuffix("_prog") + match self.peer: + case "server": + for version in node.versions: + emit_version_result_encoders( + self.environment, program, version, + ) + case "client": + for version in node.versions: + emit_version_argument_encoders( + self.environment, program, version, + ) diff --git a/tools/net/sunrpc/xdrgen/generators/source_top.py b/tools/net/sunrpc/xdrgen/generators/source_top.py new file mode 100644 index 000000000000..bcf47d93d6f1 --- /dev/null +++ b/tools/net/sunrpc/xdrgen/generators/source_top.py @@ -0,0 +1,32 @@ +#!/usr/bin/env python3 +# ex: set filetype=python: + +"""Generate source code boilerplate""" + +import os.path +import time + +from generators import Boilerplate +from generators import find_xdr_program_name, create_jinja2_environment +from xdr_ast import _RpcProgram, Specification, get_header_name + + +class XdrSourceTopGenerator(Boilerplate): + """Generate source code boilerplate""" + + def __init__(self, language: str, peer: str): + """Initialize an instance of this class""" + self.environment = create_jinja2_environment(language, "source_top") + self.peer = peer + + def emit_source(self, filename: str, root: Specification) -> None: + """Emit the top source boilerplate""" + name = find_xdr_program_name(root) + template = self.environment.get_template(self.peer + ".j2") + print( + template.render( + program=name, + filename=filename, + mtime=time.ctime(os.path.getmtime(filename)), + ) + ) diff --git a/tools/net/sunrpc/xdrgen/generators/struct.py b/tools/net/sunrpc/xdrgen/generators/struct.py new file mode 100644 index 000000000000..b694cd470829 --- /dev/null +++ b/tools/net/sunrpc/xdrgen/generators/struct.py @@ -0,0 +1,272 @@ +#!/usr/bin/env python3 +# ex: set filetype=python: + +"""Generate code to handle XDR struct types""" + +from jinja2 import Environment + +from generators import SourceGenerator, kernel_c_type +from generators import create_jinja2_environment, get_jinja2_template + +from xdr_ast import _XdrBasic, _XdrVariableLengthString +from xdr_ast import _XdrFixedLengthOpaque, _XdrVariableLengthOpaque +from xdr_ast import _XdrFixedLengthArray, _XdrVariableLengthArray +from xdr_ast import _XdrOptionalData, _XdrStruct, _XdrDeclaration +from xdr_ast import public_apis + + +def emit_struct_declaration(environment: Environment, node: _XdrStruct) -> None: + """Emit one declaration pair for an XDR struct type""" + if node.name in public_apis: + template = get_jinja2_template(environment, "declaration", "close") + print(template.render(name=node.name)) + + +def emit_struct_member_definition( + environment: Environment, field: _XdrDeclaration +) -> None: + """Emit a definition for one field in an XDR struct""" + if isinstance(field, _XdrBasic): + template = get_jinja2_template(environment, "definition", field.template) + print( + template.render( + name=field.name, + type=kernel_c_type(field.spec), + classifier=field.spec.c_classifier, + ) + ) + elif isinstance(field, _XdrFixedLengthOpaque): + template = get_jinja2_template(environment, "definition", field.template) + print( + template.render( + name=field.name, + size=field.size, + ) + ) + elif isinstance(field, _XdrVariableLengthOpaque): + template = get_jinja2_template(environment, "definition", field.template) + print(template.render(name=field.name)) + elif isinstance(field, _XdrVariableLengthString): + template = get_jinja2_template(environment, "definition", field.template) + print(template.render(name=field.name)) + elif isinstance(field, _XdrFixedLengthArray): + template = get_jinja2_template(environment, "definition", field.template) + print( + template.render( + name=field.name, + type=kernel_c_type(field.spec), + size=field.size, + ) + ) + elif isinstance(field, _XdrVariableLengthArray): + template = get_jinja2_template(environment, "definition", field.template) + print( + template.render( + name=field.name, + type=kernel_c_type(field.spec), + classifier=field.spec.c_classifier, + ) + ) + elif isinstance(field, _XdrOptionalData): + template = get_jinja2_template(environment, "definition", field.template) + print( + template.render( + name=field.name, + type=kernel_c_type(field.spec), + classifier=field.spec.c_classifier, + ) + ) + + +def emit_struct_definition(environment: Environment, node: _XdrStruct) -> None: + """Emit one definition for an XDR struct type""" + template = get_jinja2_template(environment, "definition", "open") + print(template.render(name=node.name)) + + for field in node.fields: + emit_struct_member_definition(environment, field) + + template = get_jinja2_template(environment, "definition", "close") + print(template.render(name=node.name)) + + +def emit_struct_member_decoder( + environment: Environment, field: _XdrDeclaration +) -> None: + """Emit a decoder for one field in an XDR struct""" + if isinstance(field, _XdrBasic): + template = get_jinja2_template(environment, "decoder", field.template) + print( + template.render( + name=field.name, + type=field.spec.type_name, + classifier=field.spec.c_classifier, + ) + ) + elif isinstance(field, _XdrFixedLengthOpaque): + template = get_jinja2_template(environment, "decoder", field.template) + print( + template.render( + name=field.name, + size=field.size, + ) + ) + elif isinstance(field, _XdrVariableLengthOpaque): + template = get_jinja2_template(environment, "decoder", field.template) + print( + template.render( + name=field.name, + maxsize=field.maxsize, + ) + ) + elif isinstance(field, _XdrVariableLengthString): + template = get_jinja2_template(environment, "decoder", field.template) + print( + template.render( + name=field.name, + maxsize=field.maxsize, + ) + ) + elif isinstance(field, _XdrFixedLengthArray): + template = get_jinja2_template(environment, "decoder", field.template) + print( + template.render( + name=field.name, + type=field.spec.type_name, + size=field.size, + classifier=field.spec.c_classifier, + ) + ) + elif isinstance(field, _XdrVariableLengthArray): + template = get_jinja2_template(environment, "decoder", field.template) + print( + template.render( + name=field.name, + type=field.spec.type_name, + maxsize=field.maxsize, + classifier=field.spec.c_classifier, + ) + ) + elif isinstance(field, _XdrOptionalData): + template = get_jinja2_template(environment, "decoder", field.template) + print( + template.render( + name=field.name, + type=field.spec.type_name, + classifier=field.spec.c_classifier, + ) + ) + + +def emit_struct_decoder(environment: Environment, node: _XdrStruct) -> None: + """Emit one decoder function for an XDR struct type""" + template = get_jinja2_template(environment, "decoder", "open") + print(template.render(name=node.name)) + + for field in node.fields: + emit_struct_member_decoder(environment, field) + + template = get_jinja2_template(environment, "decoder", "close") + print(template.render()) + + +def emit_struct_member_encoder( + environment: Environment, field: _XdrDeclaration +) -> None: + """Emit an encoder for one field in an XDR struct""" + if isinstance(field, _XdrBasic): + template = get_jinja2_template(environment, "encoder", field.template) + print( + template.render( + name=field.name, + type=field.spec.type_name, + ) + ) + elif isinstance(field, _XdrFixedLengthOpaque): + template = get_jinja2_template(environment, "encoder", field.template) + print( + template.render( + name=field.name, + size=field.size, + ) + ) + elif isinstance(field, _XdrVariableLengthOpaque): + template = get_jinja2_template(environment, "encoder", field.template) + print( + template.render( + name=field.name, + maxsize=field.maxsize, + ) + ) + elif isinstance(field, _XdrVariableLengthString): + template = get_jinja2_template(environment, "encoder", field.template) + print( + template.render( + name=field.name, + maxsize=field.maxsize, + ) + ) + elif isinstance(field, _XdrFixedLengthArray): + template = get_jinja2_template(environment, "encoder", field.template) + print( + template.render( + name=field.name, + type=field.spec.type_name, + size=field.size, + ) + ) + elif isinstance(field, _XdrVariableLengthArray): + template = get_jinja2_template(environment, "encoder", field.template) + print( + template.render( + name=field.name, + type=field.spec.type_name, + maxsize=field.maxsize, + ) + ) + elif isinstance(field, _XdrOptionalData): + template = get_jinja2_template(environment, "encoder", field.template) + print( + template.render( + name=field.name, + type=field.spec.type_name, + classifier=field.spec.c_classifier, + ) + ) + + +def emit_struct_encoder(environment: Environment, node: _XdrStruct) -> None: + """Emit one encoder function for an XDR struct type""" + template = get_jinja2_template(environment, "encoder", "open") + print(template.render(name=node.name)) + + for field in node.fields: + emit_struct_member_encoder(environment, field) + + template = get_jinja2_template(environment, "encoder", "close") + print(template.render()) + + +class XdrStructGenerator(SourceGenerator): + """Generate source code for XDR structs""" + + def __init__(self, language: str, peer: str): + """Initialize an instance of this class""" + self.environment = create_jinja2_environment(language, "struct") + self.peer = peer + + def emit_declaration(self, node: _XdrStruct) -> None: + """Emit one declaration pair for an XDR struct type""" + emit_struct_declaration(self.environment, node) + + def emit_definition(self, node: _XdrStruct) -> None: + """Emit one definition for an XDR struct type""" + emit_struct_definition(self.environment, node) + + def emit_decoder(self, node: _XdrStruct) -> None: + """Emit one decoder function for an XDR struct type""" + emit_struct_decoder(self.environment, node) + + def emit_encoder(self, node: _XdrStruct) -> None: + """Emit one encoder function for an XDR struct type""" + emit_struct_encoder(self.environment, node) diff --git a/tools/net/sunrpc/xdrgen/generators/typedef.py b/tools/net/sunrpc/xdrgen/generators/typedef.py new file mode 100644 index 000000000000..85a1b2303333 --- /dev/null +++ b/tools/net/sunrpc/xdrgen/generators/typedef.py @@ -0,0 +1,255 @@ +#!/usr/bin/env python3 +# ex: set filetype=python: + +"""Generate code to handle XDR typedefs""" + +from jinja2 import Environment + +from generators import SourceGenerator, kernel_c_type +from generators import create_jinja2_environment, get_jinja2_template + +from xdr_ast import _XdrBasic, _XdrTypedef, _XdrVariableLengthString +from xdr_ast import _XdrFixedLengthOpaque, _XdrVariableLengthOpaque +from xdr_ast import _XdrFixedLengthArray, _XdrVariableLengthArray +from xdr_ast import _XdrOptionalData, _XdrVoid, _XdrDeclaration +from xdr_ast import public_apis + + +def emit_typedef_declaration(environment: Environment, node: _XdrDeclaration) -> None: + """Emit a declaration pair for one XDR typedef""" + if node.name not in public_apis: + return + if isinstance(node, _XdrBasic): + template = get_jinja2_template(environment, "declaration", node.template) + print( + template.render( + name=node.name, + type=kernel_c_type(node.spec), + classifier=node.spec.c_classifier, + ) + ) + elif isinstance(node, _XdrVariableLengthString): + template = get_jinja2_template(environment, "declaration", node.template) + print(template.render(name=node.name)) + elif isinstance(node, _XdrFixedLengthOpaque): + template = get_jinja2_template(environment, "declaration", node.template) + print(template.render(name=node.name, size=node.size)) + elif isinstance(node, _XdrVariableLengthOpaque): + template = get_jinja2_template(environment, "declaration", node.template) + print(template.render(name=node.name)) + elif isinstance(node, _XdrFixedLengthArray): + template = get_jinja2_template(environment, "declaration", node.template) + print( + template.render( + name=node.name, + type=node.spec.type_name, + size=node.size, + ) + ) + elif isinstance(node, _XdrVariableLengthArray): + template = get_jinja2_template(environment, "declaration", node.template) + print( + template.render( + name=node.name, + type=node.spec.type_name, + classifier=node.spec.c_classifier, + ) + ) + elif isinstance(node, _XdrOptionalData): + raise NotImplementedError(" typedef not yet implemented") + elif isinstance(node, _XdrVoid): + raise NotImplementedError(" typedef not yet implemented") + else: + raise NotImplementedError("typedef: type not recognized") + + +def emit_type_definition(environment: Environment, node: _XdrDeclaration) -> None: + """Emit a definition for one XDR typedef""" + if isinstance(node, _XdrBasic): + template = get_jinja2_template(environment, "definition", node.template) + print( + template.render( + name=node.name, + type=kernel_c_type(node.spec), + classifier=node.spec.c_classifier, + ) + ) + elif isinstance(node, _XdrVariableLengthString): + template = get_jinja2_template(environment, "definition", node.template) + print(template.render(name=node.name)) + elif isinstance(node, _XdrFixedLengthOpaque): + template = get_jinja2_template(environment, "definition", node.template) + print(template.render(name=node.name, size=node.size)) + elif isinstance(node, _XdrVariableLengthOpaque): + template = get_jinja2_template(environment, "definition", node.template) + print(template.render(name=node.name)) + elif isinstance(node, _XdrFixedLengthArray): + template = get_jinja2_template(environment, "definition", node.template) + print( + template.render( + name=node.name, + type=node.spec.type_name, + size=node.size, + ) + ) + elif isinstance(node, _XdrVariableLengthArray): + template = get_jinja2_template(environment, "definition", node.template) + print( + template.render( + name=node.name, + type=node.spec.type_name, + classifier=node.spec.c_classifier, + ) + ) + elif isinstance(node, _XdrOptionalData): + raise NotImplementedError(" typedef not yet implemented") + elif isinstance(node, _XdrVoid): + raise NotImplementedError(" typedef not yet implemented") + else: + raise NotImplementedError("typedef: type not recognized") + + +def emit_typedef_decoder(environment: Environment, node: _XdrDeclaration) -> None: + """Emit a decoder function for one XDR typedef""" + if isinstance(node, _XdrBasic): + template = get_jinja2_template(environment, "decoder", node.template) + print( + template.render( + name=node.name, + type=node.spec.type_name, + ) + ) + elif isinstance(node, _XdrVariableLengthString): + template = get_jinja2_template(environment, "decoder", node.template) + print( + template.render( + name=node.name, + maxsize=node.maxsize, + ) + ) + elif isinstance(node, _XdrFixedLengthOpaque): + template = get_jinja2_template(environment, "decoder", node.template) + print( + template.render( + name=node.name, + size=node.size, + ) + ) + elif isinstance(node, _XdrVariableLengthOpaque): + template = get_jinja2_template(environment, "decoder", node.template) + print( + template.render( + name=node.name, + maxsize=node.maxsize, + ) + ) + elif isinstance(node, _XdrFixedLengthArray): + template = get_jinja2_template(environment, "decoder", node.template) + print( + template.render( + name=node.name, + type=node.spec.type_name, + size=node.size, + classifier=node.spec.c_classifier, + ) + ) + elif isinstance(node, _XdrVariableLengthArray): + template = get_jinja2_template(environment, "decoder", node.template) + print( + template.render( + name=node.name, + type=node.spec.type_name, + maxsize=node.maxsize, + ) + ) + elif isinstance(node, _XdrOptionalData): + raise NotImplementedError(" typedef not yet implemented") + elif isinstance(node, _XdrVoid): + raise NotImplementedError(" typedef not yet implemented") + else: + raise NotImplementedError("typedef: type not recognized") + + +def emit_typedef_encoder(environment: Environment, node: _XdrDeclaration) -> None: + """Emit an encoder function for one XDR typedef""" + if isinstance(node, _XdrBasic): + template = get_jinja2_template(environment, "encoder", node.template) + print( + template.render( + name=node.name, + type=node.spec.type_name, + ) + ) + elif isinstance(node, _XdrVariableLengthString): + template = get_jinja2_template(environment, "encoder", node.template) + print( + template.render( + name=node.name, + maxsize=node.maxsize, + ) + ) + elif isinstance(node, _XdrFixedLengthOpaque): + template = get_jinja2_template(environment, "encoder", node.template) + print( + template.render( + name=node.name, + size=node.size, + ) + ) + elif isinstance(node, _XdrVariableLengthOpaque): + template = get_jinja2_template(environment, "encoder", node.template) + print( + template.render( + name=node.name, + maxsize=node.maxsize, + ) + ) + elif isinstance(node, _XdrFixedLengthArray): + template = get_jinja2_template(environment, "encoder", node.template) + print( + template.render( + name=node.name, + type=node.spec.type_name, + size=node.size, + ) + ) + elif isinstance(node, _XdrVariableLengthArray): + template = get_jinja2_template(environment, "encoder", node.template) + print( + template.render( + name=node.name, + type=node.spec.type_name, + maxsize=node.maxsize, + ) + ) + elif isinstance(node, _XdrOptionalData): + raise NotImplementedError(" typedef not yet implemented") + elif isinstance(node, _XdrVoid): + raise NotImplementedError(" typedef not yet implemented") + else: + raise NotImplementedError("typedef: type not recognized") + + +class XdrTypedefGenerator(SourceGenerator): + """Generate source code for XDR typedefs""" + + def __init__(self, language: str, peer: str): + """Initialize an instance of this class""" + self.environment = create_jinja2_environment(language, "typedef") + self.peer = peer + + def emit_declaration(self, node: _XdrTypedef) -> None: + """Emit one declaration pair for an XDR enum type""" + emit_typedef_declaration(self.environment, node.declaration) + + def emit_definition(self, node: _XdrTypedef) -> None: + """Emit one definition for an XDR typedef""" + emit_type_definition(self.environment, node.declaration) + + def emit_decoder(self, node: _XdrTypedef) -> None: + """Emit one decoder function for an XDR typedef""" + emit_typedef_decoder(self.environment, node.declaration) + + def emit_encoder(self, node: _XdrTypedef) -> None: + """Emit one encoder function for an XDR typedef""" + emit_typedef_encoder(self.environment, node.declaration) diff --git a/tools/net/sunrpc/xdrgen/generators/union.py b/tools/net/sunrpc/xdrgen/generators/union.py new file mode 100644 index 000000000000..7974967bbb9f --- /dev/null +++ b/tools/net/sunrpc/xdrgen/generators/union.py @@ -0,0 +1,243 @@ +#!/usr/bin/env python3 +# ex: set filetype=python: + +"""Generate code to handle XDR unions""" + +from jinja2 import Environment + +from generators import SourceGenerator +from generators import create_jinja2_environment, get_jinja2_template + +from xdr_ast import _XdrBasic, _XdrUnion, _XdrVoid +from xdr_ast import _XdrDeclaration, _XdrCaseSpec, public_apis + + +def emit_union_declaration(environment: Environment, node: _XdrUnion) -> None: + """Emit one declaration pair for an XDR union type""" + if node.name in public_apis: + template = get_jinja2_template(environment, "declaration", "close") + print(template.render(name=node.name)) + + +def emit_union_switch_spec_definition( + environment: Environment, node: _XdrDeclaration +) -> None: + """Emit a definition for an XDR union's discriminant""" + assert isinstance(node, _XdrBasic) + template = get_jinja2_template(environment, "definition", "switch_spec") + print( + template.render( + name=node.name, + type=node.spec.type_name, + classifier=node.spec.c_classifier, + ) + ) + + +def emit_union_case_spec_definition( + environment: Environment, node: _XdrDeclaration +) -> None: + """Emit a definition for an XDR union's case arm""" + if isinstance(node.arm, _XdrVoid): + return + assert isinstance(node.arm, _XdrBasic) + template = get_jinja2_template(environment, "definition", "case_spec") + print( + template.render( + name=node.arm.name, + type=node.arm.spec.type_name, + classifier=node.arm.spec.c_classifier, + ) + ) + + +def emit_union_definition(environment: Environment, node: _XdrUnion) -> None: + """Emit one XDR union definition""" + template = get_jinja2_template(environment, "definition", "open") + print(template.render(name=node.name)) + + emit_union_switch_spec_definition(environment, node.discriminant) + + for case in node.cases: + emit_union_case_spec_definition(environment, case) + + if node.default is not None: + emit_union_case_spec_definition(environment, node.default) + + template = get_jinja2_template(environment, "definition", "close") + print(template.render(name=node.name)) + + +def emit_union_switch_spec_decoder( + environment: Environment, node: _XdrDeclaration +) -> None: + """Emit a decoder for an XDR union's discriminant""" + assert isinstance(node, _XdrBasic) + template = get_jinja2_template(environment, "decoder", "switch_spec") + print(template.render(name=node.name, type=node.spec.type_name)) + + +def emit_union_case_spec_decoder(environment: Environment, node: _XdrCaseSpec) -> None: + """Emit decoder functions for an XDR union's case arm""" + + if isinstance(node.arm, _XdrVoid): + return + + template = get_jinja2_template(environment, "decoder", "case_spec") + for case in node.values: + print(template.render(case=case)) + + assert isinstance(node.arm, _XdrBasic) + template = get_jinja2_template(environment, "decoder", node.arm.template) + print( + template.render( + name=node.arm.name, + type=node.arm.spec.type_name, + classifier=node.arm.spec.c_classifier, + ) + ) + + template = get_jinja2_template(environment, "decoder", "break") + print(template.render()) + + +def emit_union_default_spec_decoder(environment: Environment, node: _XdrUnion) -> None: + """Emit a decoder function for an XDR union's default arm""" + default_case = node.default + + # Avoid a gcc warning about a default case with boolean discriminant + if default_case is None and node.discriminant.spec.type_name == "bool": + return + + template = get_jinja2_template(environment, "decoder", "default_spec") + print(template.render()) + + if default_case is None or isinstance(default_case.arm, _XdrVoid): + template = get_jinja2_template(environment, "decoder", "break") + print(template.render()) + return + + assert isinstance(default_case.arm, _XdrBasic) + template = get_jinja2_template(environment, "decoder", default_case.arm.template) + print( + template.render( + name=default_case.arm.name, + type=default_case.arm.spec.type_name, + classifier=default_case.arm.spec.c_classifier, + ) + ) + + +def emit_union_decoder(environment: Environment, node: _XdrUnion) -> None: + """Emit one XDR union decoder""" + template = get_jinja2_template(environment, "decoder", "open") + print(template.render(name=node.name)) + + emit_union_switch_spec_decoder(environment, node.discriminant) + + for case in node.cases: + emit_union_case_spec_decoder(environment, case) + + emit_union_default_spec_decoder(environment, node) + + template = get_jinja2_template(environment, "decoder", "close") + print(template.render()) + + +def emit_union_switch_spec_encoder( + environment: Environment, node: _XdrDeclaration +) -> None: + """Emit an encoder for an XDR union's discriminant""" + assert isinstance(node, _XdrBasic) + template = get_jinja2_template(environment, "encoder", "switch_spec") + print(template.render(name=node.name, type=node.spec.type_name)) + + +def emit_union_case_spec_encoder(environment: Environment, node: _XdrCaseSpec) -> None: + """Emit encoder functions for an XDR union's case arm""" + + if isinstance(node.arm, _XdrVoid): + return + + template = get_jinja2_template(environment, "encoder", "case_spec") + for case in node.values: + print(template.render(case=case)) + + assert isinstance(node.arm, _XdrBasic) + template = get_jinja2_template(environment, "encoder", node.arm.template) + print( + template.render( + name=node.arm.name, + type=node.arm.spec.type_name, + ) + ) + + template = get_jinja2_template(environment, "encoder", "break") + print(template.render()) + + +def emit_union_default_spec_encoder(environment: Environment, node: _XdrUnion) -> None: + """Emit an encoder function for an XDR union's default arm""" + default_case = node.default + + # Avoid a gcc warning about a default case with boolean discriminant + if default_case is None and node.discriminant.spec.type_name == "bool": + return + + template = get_jinja2_template(environment, "encoder", "default_spec") + print(template.render()) + + if default_case is None or isinstance(default_case.arm, _XdrVoid): + template = get_jinja2_template(environment, "encoder", "break") + print(template.render()) + return + + assert isinstance(default_case.arm, _XdrBasic) + template = get_jinja2_template(environment, "encoder", default_case.arm.template) + print( + template.render( + name=default_case.arm.name, + type=default_case.arm.spec.type_name, + ) + ) + + +def emit_union_encoder(environment, node: _XdrUnion) -> None: + """Emit one XDR union encoder""" + template = get_jinja2_template(environment, "encoder", "open") + print(template.render(name=node.name)) + + emit_union_switch_spec_encoder(environment, node.discriminant) + + for case in node.cases: + emit_union_case_spec_encoder(environment, case) + + emit_union_default_spec_encoder(environment, node) + + template = get_jinja2_template(environment, "encoder", "close") + print(template.render()) + + +class XdrUnionGenerator(SourceGenerator): + """Generate source code for XDR unions""" + + def __init__(self, language: str, peer: str): + """Initialize an instance of this class""" + self.environment = create_jinja2_environment(language, "union") + self.peer = peer + + def emit_declaration(self, node: _XdrUnion) -> None: + """Emit one declaration pair for an XDR union""" + emit_union_declaration(self.environment, node) + + def emit_definition(self, node: _XdrUnion) -> None: + """Emit one definition for an XDR union""" + emit_union_definition(self.environment, node) + + def emit_decoder(self, node: _XdrUnion) -> None: + """Emit one decoder function for an XDR union""" + emit_union_decoder(self.environment, node) + + def emit_encoder(self, node: _XdrUnion) -> None: + """Emit one encoder function for an XDR union""" + emit_union_encoder(self.environment, node) diff --git a/tools/net/sunrpc/xdrgen/grammars/xdr.lark b/tools/net/sunrpc/xdrgen/grammars/xdr.lark new file mode 100644 index 000000000000..f3c4552e548d --- /dev/null +++ b/tools/net/sunrpc/xdrgen/grammars/xdr.lark @@ -0,0 +1,119 @@ +// A Lark grammar for the XDR specification language based on +// https://tools.ietf.org/html/rfc4506 Section 6.3 + +declaration : "opaque" identifier "[" value "]" -> fixed_length_opaque + | "opaque" identifier "<" [ value ] ">" -> variable_length_opaque + | "string" identifier "<" [ value ] ">" -> variable_length_string + | type_specifier identifier "[" value "]" -> fixed_length_array + | type_specifier identifier "<" [ value ] ">" -> variable_length_array + | type_specifier "*" identifier -> optional_data + | type_specifier identifier -> basic + | "void" -> void + +value : decimal_constant + | hexadecimal_constant + | octal_constant + | identifier + +constant : decimal_constant | hexadecimal_constant | octal_constant + +type_specifier : unsigned_hyper + | unsigned_long + | unsigned_int + | hyper + | long + | int + | float + | double + | quadruple + | bool + | enum_type_spec + | struct_type_spec + | union_type_spec + | identifier + +unsigned_hyper : "unsigned" "hyper" +unsigned_long : "unsigned" "long" +unsigned_int : "unsigned" "int" +hyper : "hyper" +long : "long" +int : "int" +float : "float" +double : "double" +quadruple : "quadruple" +bool : "bool" + +enum_type_spec : "enum" enum_body + +enum_body : "{" ( identifier "=" value ) ( "," identifier "=" value )* "}" + +struct_type_spec : "struct" struct_body + +struct_body : "{" ( declaration ";" )+ "}" + +union_type_spec : "union" union_body + +union_body : switch_spec "{" case_spec+ [ default_spec ] "}" + +switch_spec : "switch" "(" declaration ")" + +case_spec : ( "case" value ":" )+ declaration ";" + +default_spec : "default" ":" declaration ";" + +constant_def : "const" identifier "=" value ";" + +type_def : "typedef" declaration ";" -> typedef + | "enum" identifier enum_body ";" -> enum + | "struct" identifier struct_body ";" -> struct + | "union" identifier union_body ";" -> union + +specification : definition* + +definition : constant_def + | type_def + | program_def + | pragma_def + +// +// RPC program definitions not specified in RFC 4506 +// + +program_def : "program" identifier "{" version_def+ "}" "=" constant ";" + +version_def : "version" identifier "{" procedure_def+ "}" "=" constant ";" + +procedure_def : type_specifier identifier "(" type_specifier ")" "=" constant ";" + +pragma_def : "pragma" directive identifier [ identifier ] ";" + +directive : exclude_directive + | header_directive + | pages_directive + | public_directive + | skip_directive + +exclude_directive : "exclude" +header_directive : "header" +pages_directive : "pages" +public_directive : "public" +skip_directive : "skip" + +// +// XDR language primitives +// + +identifier : /([a-z]|[A-Z])(_|[a-z]|[A-Z]|[0-9])*/ + +decimal_constant : /[\+-]?(0|[1-9][0-9]*)/ +hexadecimal_constant : /0x([a-f]|[A-F]|[0-9])+/ +octal_constant : /0[0-7]+/ + +PASSTHRU : "%" | "%" /.+/ +%ignore PASSTHRU + +%import common.C_COMMENT +%ignore C_COMMENT + +%import common.WS +%ignore WS diff --git a/tools/net/sunrpc/xdrgen/subcmds/__init__.py b/tools/net/sunrpc/xdrgen/subcmds/__init__.py new file mode 100644 index 000000000000..c940e9275252 --- /dev/null +++ b/tools/net/sunrpc/xdrgen/subcmds/__init__.py @@ -0,0 +1,2 @@ +# SPDX-License-Identifier: GPL-2.0 +# Just to make sphinx-apidoc document this directory diff --git a/tools/net/sunrpc/xdrgen/subcmds/declarations.py b/tools/net/sunrpc/xdrgen/subcmds/declarations.py new file mode 100644 index 000000000000..c5e8d79986ef --- /dev/null +++ b/tools/net/sunrpc/xdrgen/subcmds/declarations.py @@ -0,0 +1,76 @@ +#!/usr/bin/env python3 +# ex: set filetype=python: + +"""Translate an XDR specification into executable code that +can be compiled for the Linux kernel.""" + +import logging + +from argparse import Namespace +from lark import logger +from lark.exceptions import UnexpectedInput + +from generators.constant import XdrConstantGenerator +from generators.enum import XdrEnumGenerator +from generators.header_bottom import XdrHeaderBottomGenerator +from generators.header_top import XdrHeaderTopGenerator +from generators.pointer import XdrPointerGenerator +from generators.program import XdrProgramGenerator +from generators.typedef import XdrTypedefGenerator +from generators.struct import XdrStructGenerator +from generators.union import XdrUnionGenerator + +from xdr_ast import transform_parse_tree, _RpcProgram, Specification +from xdr_ast import _XdrConstant, _XdrEnum, _XdrPointer +from xdr_ast import _XdrTypedef, _XdrStruct, _XdrUnion +from xdr_parse import xdr_parser, set_xdr_annotate + +logger.setLevel(logging.INFO) + + +def emit_header_declarations( + root: Specification, language: str, peer: str +) -> None: + """Emit header declarations""" + for definition in root.definitions: + if isinstance(definition.value, _XdrEnum): + gen = XdrEnumGenerator(language, peer) + elif isinstance(definition.value, _XdrPointer): + gen = XdrPointerGenerator(language, peer) + elif isinstance(definition.value, _XdrTypedef): + gen = XdrTypedefGenerator(language, peer) + elif isinstance(definition.value, _XdrStruct): + gen = XdrStructGenerator(language, peer) + elif isinstance(definition.value, _XdrUnion): + gen = XdrUnionGenerator(language, peer) + elif isinstance(definition.value, _RpcProgram): + gen = XdrProgramGenerator(language, peer) + else: + continue + gen.emit_declaration(definition.value) + + +def handle_parse_error(e: UnexpectedInput) -> bool: + """Simple parse error reporting, no recovery attempted""" + print(e) + return True + + +def subcmd(args: Namespace) -> int: + """Generate definitions and declarations""" + + set_xdr_annotate(args.annotate) + parser = xdr_parser() + with open(args.filename, encoding="utf-8") as f: + parse_tree = parser.parse(f.read(), on_error=handle_parse_error) + ast = transform_parse_tree(parse_tree) + + gen = XdrHeaderTopGenerator(args.language, args.peer) + gen.emit_declaration(args.filename, ast) + + emit_header_declarations(ast, args.language, args.peer) + + gen = XdrHeaderBottomGenerator(args.language, args.peer) + gen.emit_declaration(args.filename, ast) + + return 0 diff --git a/tools/net/sunrpc/xdrgen/subcmds/definitions.py b/tools/net/sunrpc/xdrgen/subcmds/definitions.py new file mode 100644 index 000000000000..5cd13d53221f --- /dev/null +++ b/tools/net/sunrpc/xdrgen/subcmds/definitions.py @@ -0,0 +1,78 @@ +#!/usr/bin/env python3 +# ex: set filetype=python: + +"""Translate an XDR specification into executable code that +can be compiled for the Linux kernel.""" + +import logging + +from argparse import Namespace +from lark import logger +from lark.exceptions import UnexpectedInput + +from generators.constant import XdrConstantGenerator +from generators.enum import XdrEnumGenerator +from generators.header_bottom import XdrHeaderBottomGenerator +from generators.header_top import XdrHeaderTopGenerator +from generators.pointer import XdrPointerGenerator +from generators.program import XdrProgramGenerator +from generators.typedef import XdrTypedefGenerator +from generators.struct import XdrStructGenerator +from generators.union import XdrUnionGenerator + +from xdr_ast import transform_parse_tree, Specification +from xdr_ast import _RpcProgram, _XdrConstant, _XdrEnum, _XdrPointer +from xdr_ast import _XdrTypedef, _XdrStruct, _XdrUnion +from xdr_parse import xdr_parser, set_xdr_annotate + +logger.setLevel(logging.INFO) + + +def emit_header_definitions( + root: Specification, language: str, peer: str +) -> None: + """Emit header definitions""" + for definition in root.definitions: + if isinstance(definition.value, _XdrConstant): + gen = XdrConstantGenerator(language, peer) + elif isinstance(definition.value, _XdrEnum): + gen = XdrEnumGenerator(language, peer) + elif isinstance(definition.value, _XdrPointer): + gen = XdrPointerGenerator(language, peer) + elif isinstance(definition.value, _RpcProgram): + gen = XdrProgramGenerator(language, peer) + elif isinstance(definition.value, _XdrTypedef): + gen = XdrTypedefGenerator(language, peer) + elif isinstance(definition.value, _XdrStruct): + gen = XdrStructGenerator(language, peer) + elif isinstance(definition.value, _XdrUnion): + gen = XdrUnionGenerator(language, peer) + else: + continue + gen.emit_definition(definition.value) + + +def handle_parse_error(e: UnexpectedInput) -> bool: + """Simple parse error reporting, no recovery attempted""" + print(e) + return True + + +def subcmd(args: Namespace) -> int: + """Generate definitions""" + + set_xdr_annotate(args.annotate) + parser = xdr_parser() + with open(args.filename, encoding="utf-8") as f: + parse_tree = parser.parse(f.read(), on_error=handle_parse_error) + ast = transform_parse_tree(parse_tree) + + gen = XdrHeaderTopGenerator(args.language, args.peer) + gen.emit_definition(args.filename, ast) + + emit_header_definitions(ast, args.language, args.peer) + + gen = XdrHeaderBottomGenerator(args.language, args.peer) + gen.emit_definition(args.filename, ast) + + return 0 diff --git a/tools/net/sunrpc/xdrgen/subcmds/lint.py b/tools/net/sunrpc/xdrgen/subcmds/lint.py new file mode 100644 index 000000000000..36cc43717d30 --- /dev/null +++ b/tools/net/sunrpc/xdrgen/subcmds/lint.py @@ -0,0 +1,33 @@ +#!/usr/bin/env python3 +# ex: set filetype=python: + +"""Translate an XDR specification into executable code that +can be compiled for the Linux kernel.""" + +import logging + +from argparse import Namespace +from lark import logger +from lark.exceptions import UnexpectedInput + +from xdr_parse import xdr_parser +from xdr_ast import transform_parse_tree + +logger.setLevel(logging.DEBUG) + + +def handle_parse_error(e: UnexpectedInput) -> bool: + """Simple parse error reporting, no recovery attempted""" + print(e) + return True + + +def subcmd(args: Namespace) -> int: + """Lexical and syntax check of an XDR specification""" + + parser = xdr_parser() + with open(args.filename, encoding="utf-8") as f: + parse_tree = parser.parse(f.read(), on_error=handle_parse_error) + transform_parse_tree(parse_tree) + + return 0 diff --git a/tools/net/sunrpc/xdrgen/subcmds/source.py b/tools/net/sunrpc/xdrgen/subcmds/source.py new file mode 100644 index 000000000000..00c04ad15b89 --- /dev/null +++ b/tools/net/sunrpc/xdrgen/subcmds/source.py @@ -0,0 +1,118 @@ +#!/usr/bin/env python3 +# ex: set filetype=python: + +"""Translate an XDR specification into executable code that +can be compiled for the Linux kernel.""" + +import logging + +from argparse import Namespace +from lark import logger +from lark.exceptions import UnexpectedInput + +from generators.source_top import XdrSourceTopGenerator +from generators.enum import XdrEnumGenerator +from generators.pointer import XdrPointerGenerator +from generators.program import XdrProgramGenerator +from generators.typedef import XdrTypedefGenerator +from generators.struct import XdrStructGenerator +from generators.union import XdrUnionGenerator + +from xdr_ast import transform_parse_tree, _RpcProgram, Specification +from xdr_ast import _XdrAst, _XdrEnum, _XdrPointer +from xdr_ast import _XdrStruct, _XdrTypedef, _XdrUnion + +from xdr_parse import xdr_parser, set_xdr_annotate + +logger.setLevel(logging.INFO) + + +def emit_source_decoder(node: _XdrAst, language: str, peer: str) -> None: + """Emit one XDR decoder function for a source file""" + if isinstance(node, _XdrEnum): + gen = XdrEnumGenerator(language, peer) + elif isinstance(node, _XdrPointer): + gen = XdrPointerGenerator(language, peer) + elif isinstance(node, _XdrTypedef): + gen = XdrTypedefGenerator(language, peer) + elif isinstance(node, _XdrStruct): + gen = XdrStructGenerator(language, peer) + elif isinstance(node, _XdrUnion): + gen = XdrUnionGenerator(language, peer) + elif isinstance(node, _RpcProgram): + gen = XdrProgramGenerator(language, peer) + else: + return + gen.emit_decoder(node) + + +def emit_source_encoder(node: _XdrAst, language: str, peer: str) -> None: + """Emit one XDR encoder function for a source file""" + if isinstance(node, _XdrEnum): + gen = XdrEnumGenerator(language, peer) + elif isinstance(node, _XdrPointer): + gen = XdrPointerGenerator(language, peer) + elif isinstance(node, _XdrTypedef): + gen = XdrTypedefGenerator(language, peer) + elif isinstance(node, _XdrStruct): + gen = XdrStructGenerator(language, peer) + elif isinstance(node, _XdrUnion): + gen = XdrUnionGenerator(language, peer) + elif isinstance(node, _RpcProgram): + gen = XdrProgramGenerator(language, peer) + else: + return + gen.emit_encoder(node) + + +def generate_server_source(filename: str, root: Specification, language: str) -> None: + """Generate server-side source code""" + + gen = XdrSourceTopGenerator(language, "server") + gen.emit_source(filename, root) + + for definition in root.definitions: + emit_source_decoder(definition.value, language, "server") + for definition in root.definitions: + emit_source_encoder(definition.value, language, "server") + + +def generate_client_source(filename: str, root: Specification, language: str) -> None: + """Generate server-side source code""" + + gen = XdrSourceTopGenerator(language, "client") + gen.emit_source(filename, root) + + # cel: todo: client needs XDR size macros + + for definition in root.definitions: + emit_source_encoder(definition.value, language, "client") + for definition in root.definitions: + emit_source_decoder(definition.value, language, "client") + + # cel: todo: client needs PROC macros + + +def handle_parse_error(e: UnexpectedInput) -> bool: + """Simple parse error reporting, no recovery attempted""" + print(e) + return True + + +def subcmd(args: Namespace) -> int: + """Generate encoder and decoder functions""" + + set_xdr_annotate(args.annotate) + parser = xdr_parser() + with open(args.filename, encoding="utf-8") as f: + parse_tree = parser.parse(f.read(), on_error=handle_parse_error) + ast = transform_parse_tree(parse_tree) + match args.peer: + case "server": + generate_server_source(args.filename, ast, args.language) + case "client": + generate_client_source(args.filename, ast, args.language) + case _: + print("Code generation for", args.peer, "is not yet supported") + + return 0 diff --git a/tools/net/sunrpc/xdrgen/templates/C/constants/definition.j2 b/tools/net/sunrpc/xdrgen/templates/C/constants/definition.j2 new file mode 100644 index 000000000000..d648ca4193f8 --- /dev/null +++ b/tools/net/sunrpc/xdrgen/templates/C/constants/definition.j2 @@ -0,0 +1,3 @@ +{# SPDX-License-Identifier: GPL-2.0 #} + +enum { {{ name }} = {{ value }} }; diff --git a/tools/net/sunrpc/xdrgen/templates/C/enum/declaration/close.j2 b/tools/net/sunrpc/xdrgen/templates/C/enum/declaration/close.j2 new file mode 100644 index 000000000000..ab1e576c9531 --- /dev/null +++ b/tools/net/sunrpc/xdrgen/templates/C/enum/declaration/close.j2 @@ -0,0 +1,4 @@ +{# SPDX-License-Identifier: GPL-2.0 #} + +bool xdrgen_decode_{{ name }}(struct xdr_stream *xdr, enum {{ name }} *ptr); +bool xdrgen_encode_{{ name }}(struct xdr_stream *xdr, enum {{ name }} value); diff --git a/tools/net/sunrpc/xdrgen/templates/C/enum/decoder/enum.j2 b/tools/net/sunrpc/xdrgen/templates/C/enum/decoder/enum.j2 new file mode 100644 index 000000000000..341d829afeda --- /dev/null +++ b/tools/net/sunrpc/xdrgen/templates/C/enum/decoder/enum.j2 @@ -0,0 +1,19 @@ +{# SPDX-License-Identifier: GPL-2.0 #} + +{% if annotate %} +/* enum {{ name }} */ +{% endif %} +{% if name in public_apis %} +bool +{% else %} +static bool __maybe_unused +{% endif %} +xdrgen_decode_{{ name }}(struct xdr_stream *xdr, enum {{ name }} *ptr) +{ + u32 val; + + if (xdr_stream_decode_u32(xdr, &val) < 0) + return false; + *ptr = val; + return true; +} diff --git a/tools/net/sunrpc/xdrgen/templates/C/enum/definition/close.j2 b/tools/net/sunrpc/xdrgen/templates/C/enum/definition/close.j2 new file mode 100644 index 000000000000..9e62344a976a --- /dev/null +++ b/tools/net/sunrpc/xdrgen/templates/C/enum/definition/close.j2 @@ -0,0 +1,2 @@ +{# SPDX-License-Identifier: GPL-2.0 #} +}; diff --git a/tools/net/sunrpc/xdrgen/templates/C/enum/definition/enumerator.j2 b/tools/net/sunrpc/xdrgen/templates/C/enum/definition/enumerator.j2 new file mode 100644 index 000000000000..ff0b893b8b14 --- /dev/null +++ b/tools/net/sunrpc/xdrgen/templates/C/enum/definition/enumerator.j2 @@ -0,0 +1,2 @@ +{# SPDX-License-Identifier: GPL-2.0 #} + {{ name }} = {{ value }}, diff --git a/tools/net/sunrpc/xdrgen/templates/C/enum/definition/open.j2 b/tools/net/sunrpc/xdrgen/templates/C/enum/definition/open.j2 new file mode 100644 index 000000000000..b25335221d48 --- /dev/null +++ b/tools/net/sunrpc/xdrgen/templates/C/enum/definition/open.j2 @@ -0,0 +1,3 @@ +{# SPDX-License-Identifier: GPL-2.0 #} + +enum {{ name }} { diff --git a/tools/net/sunrpc/xdrgen/templates/C/enum/encoder/enum.j2 b/tools/net/sunrpc/xdrgen/templates/C/enum/encoder/enum.j2 new file mode 100644 index 000000000000..bd0a770e50f2 --- /dev/null +++ b/tools/net/sunrpc/xdrgen/templates/C/enum/encoder/enum.j2 @@ -0,0 +1,14 @@ +{# SPDX-License-Identifier: GPL-2.0 #} + +{% if annotate %} +/* enum {{ name }} */ +{% endif %} +{% if name in public_apis %} +bool +{% else %} +static bool __maybe_unused +{% endif %} +xdrgen_encode_{{ name }}(struct xdr_stream *xdr, enum {{ name }} value) +{ + return xdr_stream_encode_u32(xdr, value) == XDR_UNIT; +} diff --git a/tools/net/sunrpc/xdrgen/templates/C/header_bottom/declaration/header.j2 b/tools/net/sunrpc/xdrgen/templates/C/header_bottom/declaration/header.j2 new file mode 100644 index 000000000000..0bb8c6fc0c20 --- /dev/null +++ b/tools/net/sunrpc/xdrgen/templates/C/header_bottom/declaration/header.j2 @@ -0,0 +1,3 @@ +{# SPDX-License-Identifier: GPL-2.0 #} + +#endif /* _LINUX_XDRGEN_{{ infix }}_DECL_H */ diff --git a/tools/net/sunrpc/xdrgen/templates/C/header_bottom/definition/header.j2 b/tools/net/sunrpc/xdrgen/templates/C/header_bottom/definition/header.j2 new file mode 100644 index 000000000000..69069d08dc91 --- /dev/null +++ b/tools/net/sunrpc/xdrgen/templates/C/header_bottom/definition/header.j2 @@ -0,0 +1,3 @@ +{# SPDX-License-Identifier: GPL-2.0 #} + +#endif /* _LINUX_XDRGEN_{{ infix }}_DEF_H */ diff --git a/tools/net/sunrpc/xdrgen/templates/C/header_top/declaration/header.j2 b/tools/net/sunrpc/xdrgen/templates/C/header_top/declaration/header.j2 new file mode 100644 index 000000000000..ebb4e1d32f85 --- /dev/null +++ b/tools/net/sunrpc/xdrgen/templates/C/header_top/declaration/header.j2 @@ -0,0 +1,14 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* Generated by xdrgen. Manual edits will be lost. */ +/* XDR specification file: {{ filename }} */ +/* XDR specification modification time: {{ mtime }} */ + +#ifndef _LINUX_XDRGEN_{{ infix }}_DECL_H +#define _LINUX_XDRGEN_{{ infix }}_DECL_H + +#include + +#include +#include +#include +#include diff --git a/tools/net/sunrpc/xdrgen/templates/C/header_top/definition/header.j2 b/tools/net/sunrpc/xdrgen/templates/C/header_top/definition/header.j2 new file mode 100644 index 000000000000..92f1fd4ba024 --- /dev/null +++ b/tools/net/sunrpc/xdrgen/templates/C/header_top/definition/header.j2 @@ -0,0 +1,10 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* Generated by xdrgen. Manual edits will be lost. */ +/* XDR specification file: {{ filename }} */ +/* XDR specification modification time: {{ mtime }} */ + +#ifndef _LINUX_XDRGEN_{{ infix }}_DEF_H +#define _LINUX_XDRGEN_{{ infix }}_DEF_H + +#include +#include diff --git a/tools/net/sunrpc/xdrgen/templates/C/pointer/declaration/close.j2 b/tools/net/sunrpc/xdrgen/templates/C/pointer/declaration/close.j2 new file mode 100644 index 000000000000..816291184e8c --- /dev/null +++ b/tools/net/sunrpc/xdrgen/templates/C/pointer/declaration/close.j2 @@ -0,0 +1,4 @@ +{# SPDX-License-Identifier: GPL-2.0 #} + +bool xdrgen_decode_{{ name }}(struct xdr_stream *xdr, struct {{ name }} *ptr); +bool xdrgen_encode_{{ name }}(struct xdr_stream *xdr, const struct {{ name }} *value); diff --git a/tools/net/sunrpc/xdrgen/templates/C/pointer/decoder/basic.j2 b/tools/net/sunrpc/xdrgen/templates/C/pointer/decoder/basic.j2 new file mode 100644 index 000000000000..cde4ab53f4be --- /dev/null +++ b/tools/net/sunrpc/xdrgen/templates/C/pointer/decoder/basic.j2 @@ -0,0 +1,6 @@ +{# SPDX-License-Identifier: GPL-2.0 #} +{% if annotate %} + /* member {{ name }} (basic) */ +{% endif %} + if (!xdrgen_decode_{{ type }}(xdr, &ptr->{{ name }})) + return false; diff --git a/tools/net/sunrpc/xdrgen/templates/C/pointer/decoder/close.j2 b/tools/net/sunrpc/xdrgen/templates/C/pointer/decoder/close.j2 new file mode 100644 index 000000000000..5bf010665f84 --- /dev/null +++ b/tools/net/sunrpc/xdrgen/templates/C/pointer/decoder/close.j2 @@ -0,0 +1,3 @@ +{# SPDX-License-Identifier: GPL-2.0 #} + return true; +}; diff --git a/tools/net/sunrpc/xdrgen/templates/C/pointer/decoder/fixed_length_array.j2 b/tools/net/sunrpc/xdrgen/templates/C/pointer/decoder/fixed_length_array.j2 new file mode 100644 index 000000000000..cfd64217ad82 --- /dev/null +++ b/tools/net/sunrpc/xdrgen/templates/C/pointer/decoder/fixed_length_array.j2 @@ -0,0 +1,8 @@ +{# SPDX-License-Identifier: GPL-2.0 #} +{% if annotate %} + /* member {{ name }} (fixed-length array) */ +{% endif %} + for (u32 i = 0; i < {{ size }}; i++) { + if (xdrgen_decode_{{ type }}(xdr, &ptr->{{ name }}.items[i]) < 0) + return false; + } diff --git a/tools/net/sunrpc/xdrgen/templates/C/pointer/decoder/fixed_length_opaque.j2 b/tools/net/sunrpc/xdrgen/templates/C/pointer/decoder/fixed_length_opaque.j2 new file mode 100644 index 000000000000..b4695ece1884 --- /dev/null +++ b/tools/net/sunrpc/xdrgen/templates/C/pointer/decoder/fixed_length_opaque.j2 @@ -0,0 +1,6 @@ +{# SPDX-License-Identifier: GPL-2.0 #} +{% if annotate %} + /* member {{ name }} (fixed-length opaque) */ +{% endif %} + if (xdr_stream_decode_opaque_fixed(xdr, ptr->{{ name }}, {{ size }}) < 0) + return false; diff --git a/tools/net/sunrpc/xdrgen/templates/C/pointer/decoder/open.j2 b/tools/net/sunrpc/xdrgen/templates/C/pointer/decoder/open.j2 new file mode 100644 index 000000000000..c093d9e3c9ad --- /dev/null +++ b/tools/net/sunrpc/xdrgen/templates/C/pointer/decoder/open.j2 @@ -0,0 +1,22 @@ +{# SPDX-License-Identifier: GPL-2.0 #} + +{% if annotate %} +/* pointer {{ name }} */ +{% endif %} +{% if name in public_apis %} +bool +{% else %} +static bool __maybe_unused +{% endif %} +xdrgen_decode_{{ name }}(struct xdr_stream *xdr, struct {{ name }} *ptr) +{ + bool opted; + +{% if annotate %} + /* opted */ +{% endif %} + if (!xdrgen_decode_bool(xdr, &opted)) + return false; + if (!opted) + return true; + diff --git a/tools/net/sunrpc/xdrgen/templates/C/pointer/decoder/optional_data.j2 b/tools/net/sunrpc/xdrgen/templates/C/pointer/decoder/optional_data.j2 new file mode 100644 index 000000000000..b6834299a04b --- /dev/null +++ b/tools/net/sunrpc/xdrgen/templates/C/pointer/decoder/optional_data.j2 @@ -0,0 +1,6 @@ +{# SPDX-License-Identifier: GPL-2.0 #} +{% if annotate %} + /* member {{ name }} (optional data) */ +{% endif %} + if (!xdrgen_decode_{{ type }}(xdr, ptr->{{ name }})) + return false; diff --git a/tools/net/sunrpc/xdrgen/templates/C/pointer/decoder/variable_length_array.j2 b/tools/net/sunrpc/xdrgen/templates/C/pointer/decoder/variable_length_array.j2 new file mode 100644 index 000000000000..2f943909cdf7 --- /dev/null +++ b/tools/net/sunrpc/xdrgen/templates/C/pointer/decoder/variable_length_array.j2 @@ -0,0 +1,13 @@ +{# SPDX-License-Identifier: GPL-2.0 #} +{% if annotate %} + /* member {{ name }} (variable-length array) */ +{% endif %} + if (xdr_stream_decode_u32(xdr, &ptr->{{ name }}.count) < 0) + return false; +{% if maxsize != "0" %} + if (ptr->{{ name }}.count > {{ maxsize }}) + return false; +{% endif %} + for (u32 i = 0; i < ptr->{{ name }}.count; i++) + if (!xdrgen_decode_{{ type }}(xdr, &ptr->{{ name }}.element[i])) + return false; diff --git a/tools/net/sunrpc/xdrgen/templates/C/pointer/decoder/variable_length_opaque.j2 b/tools/net/sunrpc/xdrgen/templates/C/pointer/decoder/variable_length_opaque.j2 new file mode 100644 index 000000000000..9a814de54ae8 --- /dev/null +++ b/tools/net/sunrpc/xdrgen/templates/C/pointer/decoder/variable_length_opaque.j2 @@ -0,0 +1,6 @@ +{# SPDX-License-Identifier: GPL-2.0 #} +{% if annotate %} + /* member {{ name }} (variable-length opaque) */ +{% endif %} + if (!xdrgen_decode_opaque(xdr, (opaque *)ptr, {{ maxsize }})) + return false; diff --git a/tools/net/sunrpc/xdrgen/templates/C/pointer/decoder/variable_length_string.j2 b/tools/net/sunrpc/xdrgen/templates/C/pointer/decoder/variable_length_string.j2 new file mode 100644 index 000000000000..12d20b143b43 --- /dev/null +++ b/tools/net/sunrpc/xdrgen/templates/C/pointer/decoder/variable_length_string.j2 @@ -0,0 +1,6 @@ +{# SPDX-License-Identifier: GPL-2.0 #} +{% if annotate %} + /* member {{ name }} (variable-length string) */ +{% endif %} + if (!xdrgen_decode_string(xdr, (string *)ptr, {{ maxsize }})) + return false; diff --git a/tools/net/sunrpc/xdrgen/templates/C/pointer/definition/basic.j2 b/tools/net/sunrpc/xdrgen/templates/C/pointer/definition/basic.j2 new file mode 100644 index 000000000000..b3430895f311 --- /dev/null +++ b/tools/net/sunrpc/xdrgen/templates/C/pointer/definition/basic.j2 @@ -0,0 +1,5 @@ +{# SPDX-License-Identifier: GPL-2.0 #} +{% if annotate %} + /* (basic) */ +{% endif %} + {{ classifier }}{{ type }} {{ name }}; diff --git a/tools/net/sunrpc/xdrgen/templates/C/pointer/definition/close.j2 b/tools/net/sunrpc/xdrgen/templates/C/pointer/definition/close.j2 new file mode 100644 index 000000000000..9e62344a976a --- /dev/null +++ b/tools/net/sunrpc/xdrgen/templates/C/pointer/definition/close.j2 @@ -0,0 +1,2 @@ +{# SPDX-License-Identifier: GPL-2.0 #} +}; diff --git a/tools/net/sunrpc/xdrgen/templates/C/pointer/definition/fixed_length_array.j2 b/tools/net/sunrpc/xdrgen/templates/C/pointer/definition/fixed_length_array.j2 new file mode 100644 index 000000000000..66be836826a0 --- /dev/null +++ b/tools/net/sunrpc/xdrgen/templates/C/pointer/definition/fixed_length_array.j2 @@ -0,0 +1,5 @@ +{# SPDX-License-Identifier: GPL-2.0 #} +{% if annotate %} + /* (fixed-length array) */ +{% endif %} + {{ type }} {{ name }}[{{ size }}]; diff --git a/tools/net/sunrpc/xdrgen/templates/C/pointer/definition/fixed_length_opaque.j2 b/tools/net/sunrpc/xdrgen/templates/C/pointer/definition/fixed_length_opaque.j2 new file mode 100644 index 000000000000..0daba19aa0f0 --- /dev/null +++ b/tools/net/sunrpc/xdrgen/templates/C/pointer/definition/fixed_length_opaque.j2 @@ -0,0 +1,5 @@ +{# SPDX-License-Identifier: GPL-2.0 #} +{% if annotate %} + /* (fixed-length opaque) */ +{% endif %} + u8 {{ name }}[{{ size }}]; diff --git a/tools/net/sunrpc/xdrgen/templates/C/pointer/definition/open.j2 b/tools/net/sunrpc/xdrgen/templates/C/pointer/definition/open.j2 new file mode 100644 index 000000000000..bc886b818d85 --- /dev/null +++ b/tools/net/sunrpc/xdrgen/templates/C/pointer/definition/open.j2 @@ -0,0 +1,6 @@ +{# SPDX-License-Identifier: GPL-2.0 #} + +{% if annotate %} +/* pointer {{ name }} */ +{% endif %} +struct {{ name }} { diff --git a/tools/net/sunrpc/xdrgen/templates/C/pointer/definition/optional_data.j2 b/tools/net/sunrpc/xdrgen/templates/C/pointer/definition/optional_data.j2 new file mode 100644 index 000000000000..a33341f45e8f --- /dev/null +++ b/tools/net/sunrpc/xdrgen/templates/C/pointer/definition/optional_data.j2 @@ -0,0 +1,5 @@ +{# SPDX-License-Identifier: GPL-2.0 #} +{% if annotate %} + /* (optional data) */ +{% endif %} + {{ classifier }}{{ type }} *{{ name }}; diff --git a/tools/net/sunrpc/xdrgen/templates/C/pointer/definition/variable_length_array.j2 b/tools/net/sunrpc/xdrgen/templates/C/pointer/definition/variable_length_array.j2 new file mode 100644 index 000000000000..5d767f9b3674 --- /dev/null +++ b/tools/net/sunrpc/xdrgen/templates/C/pointer/definition/variable_length_array.j2 @@ -0,0 +1,8 @@ +{# SPDX-License-Identifier: GPL-2.0 #} +{% if annotate %} + /* (variable-length array) */ +{% endif %} + struct { + u32 count; + {{ classifier }}{{ type }} *element; + } {{ name }}; diff --git a/tools/net/sunrpc/xdrgen/templates/C/pointer/definition/variable_length_opaque.j2 b/tools/net/sunrpc/xdrgen/templates/C/pointer/definition/variable_length_opaque.j2 new file mode 100644 index 000000000000..4d0cd84be3db --- /dev/null +++ b/tools/net/sunrpc/xdrgen/templates/C/pointer/definition/variable_length_opaque.j2 @@ -0,0 +1,5 @@ +{# SPDX-License-Identifier: GPL-2.0 #} +{% if annotate %} + /* (variable-length opaque) */ +{% endif %} + opaque {{ name }}; diff --git a/tools/net/sunrpc/xdrgen/templates/C/pointer/definition/variable_length_string.j2 b/tools/net/sunrpc/xdrgen/templates/C/pointer/definition/variable_length_string.j2 new file mode 100644 index 000000000000..2de2feec77db --- /dev/null +++ b/tools/net/sunrpc/xdrgen/templates/C/pointer/definition/variable_length_string.j2 @@ -0,0 +1,5 @@ +{# SPDX-License-Identifier: GPL-2.0 #} +{% if annotate %} + /* (variable-length string) */ +{% endif %} + string {{ name }}; diff --git a/tools/net/sunrpc/xdrgen/templates/C/pointer/encoder/basic.j2 b/tools/net/sunrpc/xdrgen/templates/C/pointer/encoder/basic.j2 new file mode 100644 index 000000000000..a7d3695c5a6a --- /dev/null +++ b/tools/net/sunrpc/xdrgen/templates/C/pointer/encoder/basic.j2 @@ -0,0 +1,10 @@ +{# SPDX-License-Identifier: GPL-2.0 #} +{% if annotate %} + /* member {{ name }} (basic) */ +{% endif %} +{% if type in pass_by_reference %} + if (!xdrgen_encode_{{ type }}(xdr, &value->{{ name }})) +{% else %} + if (!xdrgen_encode_{{ type }}(xdr, value->{{ name }})) +{% endif %} + return false; diff --git a/tools/net/sunrpc/xdrgen/templates/C/pointer/encoder/close.j2 b/tools/net/sunrpc/xdrgen/templates/C/pointer/encoder/close.j2 new file mode 100644 index 000000000000..5bf010665f84 --- /dev/null +++ b/tools/net/sunrpc/xdrgen/templates/C/pointer/encoder/close.j2 @@ -0,0 +1,3 @@ +{# SPDX-License-Identifier: GPL-2.0 #} + return true; +}; diff --git a/tools/net/sunrpc/xdrgen/templates/C/pointer/encoder/fixed_length_array.j2 b/tools/net/sunrpc/xdrgen/templates/C/pointer/encoder/fixed_length_array.j2 new file mode 100644 index 000000000000..b01833a2c7a1 --- /dev/null +++ b/tools/net/sunrpc/xdrgen/templates/C/pointer/encoder/fixed_length_array.j2 @@ -0,0 +1,12 @@ +{# SPDX-License-Identifier: GPL-2.0 #} +{% if annotate %} + /* member {{ name }} (fixed-length array) */ +{% endif %} + for (u32 i = 0; i < {{ size }}; i++) { +{% if type in pass_by_reference %} + if (xdrgen_encode_{{ type }}(xdr, &value->items[i]) < 0) +{% else %} + if (xdrgen_encode_{{ type }}(xdr, value->items[i]) < 0) +{% endif %} + return false; + } diff --git a/tools/net/sunrpc/xdrgen/templates/C/pointer/encoder/fixed_length_opaque.j2 b/tools/net/sunrpc/xdrgen/templates/C/pointer/encoder/fixed_length_opaque.j2 new file mode 100644 index 000000000000..07bc91919898 --- /dev/null +++ b/tools/net/sunrpc/xdrgen/templates/C/pointer/encoder/fixed_length_opaque.j2 @@ -0,0 +1,6 @@ +{# SPDX-License-Identifier: GPL-2.0 #} +{% if annotate %} + /* member {{ name }} (fixed-length opaque) */ +{% endif %} + if (xdr_stream_encode_opaque_fixed(xdr, value->{{ name }}, {{ size }}) < 0) + return false; diff --git a/tools/net/sunrpc/xdrgen/templates/C/pointer/encoder/open.j2 b/tools/net/sunrpc/xdrgen/templates/C/pointer/encoder/open.j2 new file mode 100644 index 000000000000..d67fae200261 --- /dev/null +++ b/tools/net/sunrpc/xdrgen/templates/C/pointer/encoder/open.j2 @@ -0,0 +1,20 @@ +{# SPDX-License-Identifier: GPL-2.0 #} + +{% if annotate %} +/* pointer {{ name }} */ +{% endif %} +{% if name in public_apis %} +bool +{% else %} +static bool __maybe_unused +{% endif %} +xdrgen_encode_{{ name }}(struct xdr_stream *xdr, const struct {{ name }} *value) +{ +{% if annotate %} + /* opted */ +{% endif %} + if (!xdrgen_encode_bool(xdr, value != NULL)) + return false; + if (!value) + return true; + diff --git a/tools/net/sunrpc/xdrgen/templates/C/pointer/encoder/optional_data.j2 b/tools/net/sunrpc/xdrgen/templates/C/pointer/encoder/optional_data.j2 new file mode 100644 index 000000000000..16fb3e09bba1 --- /dev/null +++ b/tools/net/sunrpc/xdrgen/templates/C/pointer/encoder/optional_data.j2 @@ -0,0 +1,6 @@ +{# SPDX-License-Identifier: GPL-2.0 #} +{% if annotate %} + /* member {{ name }} (optional data) */ +{% endif %} + if (!xdrgen_encode_{{ type }}(xdr, value->{{ name }})) + return false; diff --git a/tools/net/sunrpc/xdrgen/templates/C/pointer/encoder/variable_length_array.j2 b/tools/net/sunrpc/xdrgen/templates/C/pointer/encoder/variable_length_array.j2 new file mode 100644 index 000000000000..0ec8660d621a --- /dev/null +++ b/tools/net/sunrpc/xdrgen/templates/C/pointer/encoder/variable_length_array.j2 @@ -0,0 +1,15 @@ +{# SPDX-License-Identifier: GPL-2.0 #} +{% if annotate %} + /* member {{ name }} (variable-length array) */ +{% endif %} + if (value->{{ name }}.count > {{ maxsize }}) + return false; + if (xdr_stream_encode_u32(xdr, value->{{ name }}.count) != XDR_UNIT) + return false; + for (u32 i = 0; i < value->{{ name }}.count; i++) +{% if type in pass_by_reference %} + if (!xdrgen_encode_{{ type }}(xdr, &value->{{ name }}.element[i])) +{% else %} + if (!xdrgen_encode_{{ type }}(xdr, value->{{ name }}.element[i])) +{% endif %} + return false; diff --git a/tools/net/sunrpc/xdrgen/templates/C/pointer/encoder/variable_length_opaque.j2 b/tools/net/sunrpc/xdrgen/templates/C/pointer/encoder/variable_length_opaque.j2 new file mode 100644 index 000000000000..1d477c2d197a --- /dev/null +++ b/tools/net/sunrpc/xdrgen/templates/C/pointer/encoder/variable_length_opaque.j2 @@ -0,0 +1,8 @@ +{# SPDX-License-Identifier: GPL-2.0 #} +{% if annotate %} + /* member {{ name }} (variable-length opaque) */ +{% endif %} + if (value->{{ name }}.len > {{ maxsize }}) + return false; + if (xdr_stream_encode_opaque(xdr, value->{{ name }}.data, value->{{ name }}.len) < 0) + return false; diff --git a/tools/net/sunrpc/xdrgen/templates/C/pointer/encoder/variable_length_string.j2 b/tools/net/sunrpc/xdrgen/templates/C/pointer/encoder/variable_length_string.j2 new file mode 100644 index 000000000000..cf65b71eaef3 --- /dev/null +++ b/tools/net/sunrpc/xdrgen/templates/C/pointer/encoder/variable_length_string.j2 @@ -0,0 +1,8 @@ +{# SPDX-License-Identifier: GPL-2.0 #} +{% if annotate %} + /* member {{ name }} (variable-length string) */ +{% endif %} + if (value->{{ name }}.len > {{ maxsize }}) + return false; + if (xdr_stream_encode_opaque(xdr, value->{{ name }}.data, value->{{ name }}.len) < 0) + return false; diff --git a/tools/net/sunrpc/xdrgen/templates/C/program/declaration/argument.j2 b/tools/net/sunrpc/xdrgen/templates/C/program/declaration/argument.j2 new file mode 100644 index 000000000000..4364fed19162 --- /dev/null +++ b/tools/net/sunrpc/xdrgen/templates/C/program/declaration/argument.j2 @@ -0,0 +1,2 @@ +{# SPDX-License-Identifier: GPL-2.0 #} +bool {{ program }}_svc_decode_{{ argument }}(struct svc_rqst *rqstp, struct xdr_stream *xdr); diff --git a/tools/net/sunrpc/xdrgen/templates/C/program/declaration/result.j2 b/tools/net/sunrpc/xdrgen/templates/C/program/declaration/result.j2 new file mode 100644 index 000000000000..e0ea1e849910 --- /dev/null +++ b/tools/net/sunrpc/xdrgen/templates/C/program/declaration/result.j2 @@ -0,0 +1,2 @@ +{# SPDX-License-Identifier: GPL-2.0 #} +bool {{ program }}_svc_encode_{{ result }}(struct svc_rqst *rqstp, struct xdr_stream *xdr); diff --git a/tools/net/sunrpc/xdrgen/templates/C/program/decoder/argument.j2 b/tools/net/sunrpc/xdrgen/templates/C/program/decoder/argument.j2 new file mode 100644 index 000000000000..0b1709cca0d4 --- /dev/null +++ b/tools/net/sunrpc/xdrgen/templates/C/program/decoder/argument.j2 @@ -0,0 +1,21 @@ +{# SPDX-License-Identifier: GPL-2.0 #} + +/** + * {{ program }}_svc_decode_{{ argument }} - Decode a {{ argument }} argument + * @rqstp: RPC transaction context + * @xdr: source XDR data stream + * + * Return values: + * %true: procedure arguments decoded successfully + * %false: decode failed + */ +bool {{ program }}_svc_decode_{{ argument }}(struct svc_rqst *rqstp, struct xdr_stream *xdr) +{ +{% if argument == 'void' %} + return xdrgen_decode_void(xdr); +{% else %} + struct {{ argument }} *argp = rqstp->rq_argp; + + return xdrgen_decode_{{ argument }}(xdr, argp); +{% endif %} +} diff --git a/tools/net/sunrpc/xdrgen/templates/C/program/decoder/result.j2 b/tools/net/sunrpc/xdrgen/templates/C/program/decoder/result.j2 new file mode 100644 index 000000000000..d304eccb5c40 --- /dev/null +++ b/tools/net/sunrpc/xdrgen/templates/C/program/decoder/result.j2 @@ -0,0 +1,22 @@ +{# SPDX-License-Identifier: GPL-2.0 #} + +{% if annotate %} +/* Decode {{ result }} results */ +{% endif %} +static int {{ program }}_xdr_dec_{{ result }}(struct rpc_rqst *req, + struct xdr_stream *xdr, void *data) +{ +{% if result == 'void' %} + xdrgen_decode_void(xdr); +{% else %} + struct {{ result }} *result = data; + + if (!xdrgen_decode_{{ result }}(xdr, result)) + return -EIO; + if (result->stat != nfs_ok) { + trace_nfs_xdr_status(xdr, (int)result->stat); + return {{ program }}_stat_to_errno(result->stat); + } +{% endif %} + return 0; +} diff --git a/tools/net/sunrpc/xdrgen/templates/C/program/definition/close.j2 b/tools/net/sunrpc/xdrgen/templates/C/program/definition/close.j2 new file mode 100644 index 000000000000..9e62344a976a --- /dev/null +++ b/tools/net/sunrpc/xdrgen/templates/C/program/definition/close.j2 @@ -0,0 +1,2 @@ +{# SPDX-License-Identifier: GPL-2.0 #} +}; diff --git a/tools/net/sunrpc/xdrgen/templates/C/program/definition/open.j2 b/tools/net/sunrpc/xdrgen/templates/C/program/definition/open.j2 new file mode 100644 index 000000000000..f9a6d439f156 --- /dev/null +++ b/tools/net/sunrpc/xdrgen/templates/C/program/definition/open.j2 @@ -0,0 +1,6 @@ +{# SPDX-License-Identifier: GPL-2.0 #} + +{% if annotate %} +/* procedure numbers for {{ program }} */ +{% endif %} +enum { diff --git a/tools/net/sunrpc/xdrgen/templates/C/program/definition/procedure.j2 b/tools/net/sunrpc/xdrgen/templates/C/program/definition/procedure.j2 new file mode 100644 index 000000000000..ff0b893b8b14 --- /dev/null +++ b/tools/net/sunrpc/xdrgen/templates/C/program/definition/procedure.j2 @@ -0,0 +1,2 @@ +{# SPDX-License-Identifier: GPL-2.0 #} + {{ name }} = {{ value }}, diff --git a/tools/net/sunrpc/xdrgen/templates/C/program/encoder/argument.j2 b/tools/net/sunrpc/xdrgen/templates/C/program/encoder/argument.j2 new file mode 100644 index 000000000000..2fbb5bd13aec --- /dev/null +++ b/tools/net/sunrpc/xdrgen/templates/C/program/encoder/argument.j2 @@ -0,0 +1,16 @@ +{# SPDX-License-Identifier: GPL-2.0 #} + +{%if annotate %} +/* Encode {{ argument }} arguments */ +{% endif %} +static void {{ program }}_xdr_enc_{{ argument }}(struct rpc_rqst *req, + struct xdr_stream *xdr, const void *data) +{ +{% if argument == 'void' %} + xdrgen_encode_void(xdr); +{% else %} + const struct {{ argument }} *args = data; + + xdrgen_encode_{{ argument }}(xdr, args); +{% endif %} +} diff --git a/tools/net/sunrpc/xdrgen/templates/C/program/encoder/result.j2 b/tools/net/sunrpc/xdrgen/templates/C/program/encoder/result.j2 new file mode 100644 index 000000000000..6fc61a5d47b7 --- /dev/null +++ b/tools/net/sunrpc/xdrgen/templates/C/program/encoder/result.j2 @@ -0,0 +1,21 @@ +{# SPDX-License-Identifier: GPL-2.0 #} + +/** + * {{ program }}_svc_encode_{{ result }} - Encode a {{ result }} result + * @rqstp: RPC transaction context + * @xdr: target XDR data stream + * + * Return values: + * %true: procedure results encoded successfully + * %false: encode failed + */ +bool {{ program }}_svc_encode_{{ result }}(struct svc_rqst *rqstp, struct xdr_stream *xdr) +{ +{% if result == 'void' %} + return xdrgen_encode_void(xdr); +{% else %} + struct {{ result }} *resp = rqstp->rq_resp; + + return xdrgen_encode_{{ result }}(xdr, resp); +{% endif %} +} diff --git a/tools/net/sunrpc/xdrgen/templates/C/source_top/client.j2 b/tools/net/sunrpc/xdrgen/templates/C/source_top/client.j2 new file mode 100644 index 000000000000..e3a802cbc4d7 --- /dev/null +++ b/tools/net/sunrpc/xdrgen/templates/C/source_top/client.j2 @@ -0,0 +1,8 @@ +// SPDX-License-Identifier: GPL-2.0 +// Generated by xdrgen. Manual edits will be lost. +// XDR specification file: {{ filename }} +// XDR specification modification time: {{ mtime }} + +#include + +#include "{{ program }}xdr_gen.h" diff --git a/tools/net/sunrpc/xdrgen/templates/C/source_top/server.j2 b/tools/net/sunrpc/xdrgen/templates/C/source_top/server.j2 new file mode 100644 index 000000000000..974e1d971e5d --- /dev/null +++ b/tools/net/sunrpc/xdrgen/templates/C/source_top/server.j2 @@ -0,0 +1,8 @@ +// SPDX-License-Identifier: GPL-2.0 +// Generated by xdrgen. Manual edits will be lost. +// XDR specification file: {{ filename }} +// XDR specification modification time: {{ mtime }} + +#include + +#include "{{ program }}xdr_gen.h" diff --git a/tools/net/sunrpc/xdrgen/templates/C/struct/declaration/close.j2 b/tools/net/sunrpc/xdrgen/templates/C/struct/declaration/close.j2 new file mode 100644 index 000000000000..816291184e8c --- /dev/null +++ b/tools/net/sunrpc/xdrgen/templates/C/struct/declaration/close.j2 @@ -0,0 +1,4 @@ +{# SPDX-License-Identifier: GPL-2.0 #} + +bool xdrgen_decode_{{ name }}(struct xdr_stream *xdr, struct {{ name }} *ptr); +bool xdrgen_encode_{{ name }}(struct xdr_stream *xdr, const struct {{ name }} *value); diff --git a/tools/net/sunrpc/xdrgen/templates/C/struct/decoder/basic.j2 b/tools/net/sunrpc/xdrgen/templates/C/struct/decoder/basic.j2 new file mode 100644 index 000000000000..cde4ab53f4be --- /dev/null +++ b/tools/net/sunrpc/xdrgen/templates/C/struct/decoder/basic.j2 @@ -0,0 +1,6 @@ +{# SPDX-License-Identifier: GPL-2.0 #} +{% if annotate %} + /* member {{ name }} (basic) */ +{% endif %} + if (!xdrgen_decode_{{ type }}(xdr, &ptr->{{ name }})) + return false; diff --git a/tools/net/sunrpc/xdrgen/templates/C/struct/decoder/close.j2 b/tools/net/sunrpc/xdrgen/templates/C/struct/decoder/close.j2 new file mode 100644 index 000000000000..5bf010665f84 --- /dev/null +++ b/tools/net/sunrpc/xdrgen/templates/C/struct/decoder/close.j2 @@ -0,0 +1,3 @@ +{# SPDX-License-Identifier: GPL-2.0 #} + return true; +}; diff --git a/tools/net/sunrpc/xdrgen/templates/C/struct/decoder/fixed_length_array.j2 b/tools/net/sunrpc/xdrgen/templates/C/struct/decoder/fixed_length_array.j2 new file mode 100644 index 000000000000..cfd64217ad82 --- /dev/null +++ b/tools/net/sunrpc/xdrgen/templates/C/struct/decoder/fixed_length_array.j2 @@ -0,0 +1,8 @@ +{# SPDX-License-Identifier: GPL-2.0 #} +{% if annotate %} + /* member {{ name }} (fixed-length array) */ +{% endif %} + for (u32 i = 0; i < {{ size }}; i++) { + if (xdrgen_decode_{{ type }}(xdr, &ptr->{{ name }}.items[i]) < 0) + return false; + } diff --git a/tools/net/sunrpc/xdrgen/templates/C/struct/decoder/fixed_length_opaque.j2 b/tools/net/sunrpc/xdrgen/templates/C/struct/decoder/fixed_length_opaque.j2 new file mode 100644 index 000000000000..b4695ece1884 --- /dev/null +++ b/tools/net/sunrpc/xdrgen/templates/C/struct/decoder/fixed_length_opaque.j2 @@ -0,0 +1,6 @@ +{# SPDX-License-Identifier: GPL-2.0 #} +{% if annotate %} + /* member {{ name }} (fixed-length opaque) */ +{% endif %} + if (xdr_stream_decode_opaque_fixed(xdr, ptr->{{ name }}, {{ size }}) < 0) + return false; diff --git a/tools/net/sunrpc/xdrgen/templates/C/struct/decoder/open.j2 b/tools/net/sunrpc/xdrgen/templates/C/struct/decoder/open.j2 new file mode 100644 index 000000000000..289e67259f55 --- /dev/null +++ b/tools/net/sunrpc/xdrgen/templates/C/struct/decoder/open.j2 @@ -0,0 +1,12 @@ +{# SPDX-License-Identifier: GPL-2.0 #} + +{% if annotate %} +/* struct {{ name }} */ +{% endif %} +{% if name in public_apis %} +bool +{% else %} +static bool __maybe_unused +{% endif %} +xdrgen_decode_{{ name }}(struct xdr_stream *xdr, struct {{ name }} *ptr) +{ diff --git a/tools/net/sunrpc/xdrgen/templates/C/struct/decoder/optional_data.j2 b/tools/net/sunrpc/xdrgen/templates/C/struct/decoder/optional_data.j2 new file mode 100644 index 000000000000..b6834299a04b --- /dev/null +++ b/tools/net/sunrpc/xdrgen/templates/C/struct/decoder/optional_data.j2 @@ -0,0 +1,6 @@ +{# SPDX-License-Identifier: GPL-2.0 #} +{% if annotate %} + /* member {{ name }} (optional data) */ +{% endif %} + if (!xdrgen_decode_{{ type }}(xdr, ptr->{{ name }})) + return false; diff --git a/tools/net/sunrpc/xdrgen/templates/C/struct/decoder/variable_length_array.j2 b/tools/net/sunrpc/xdrgen/templates/C/struct/decoder/variable_length_array.j2 new file mode 100644 index 000000000000..2f943909cdf7 --- /dev/null +++ b/tools/net/sunrpc/xdrgen/templates/C/struct/decoder/variable_length_array.j2 @@ -0,0 +1,13 @@ +{# SPDX-License-Identifier: GPL-2.0 #} +{% if annotate %} + /* member {{ name }} (variable-length array) */ +{% endif %} + if (xdr_stream_decode_u32(xdr, &ptr->{{ name }}.count) < 0) + return false; +{% if maxsize != "0" %} + if (ptr->{{ name }}.count > {{ maxsize }}) + return false; +{% endif %} + for (u32 i = 0; i < ptr->{{ name }}.count; i++) + if (!xdrgen_decode_{{ type }}(xdr, &ptr->{{ name }}.element[i])) + return false; diff --git a/tools/net/sunrpc/xdrgen/templates/C/struct/decoder/variable_length_opaque.j2 b/tools/net/sunrpc/xdrgen/templates/C/struct/decoder/variable_length_opaque.j2 new file mode 100644 index 000000000000..9a814de54ae8 --- /dev/null +++ b/tools/net/sunrpc/xdrgen/templates/C/struct/decoder/variable_length_opaque.j2 @@ -0,0 +1,6 @@ +{# SPDX-License-Identifier: GPL-2.0 #} +{% if annotate %} + /* member {{ name }} (variable-length opaque) */ +{% endif %} + if (!xdrgen_decode_opaque(xdr, (opaque *)ptr, {{ maxsize }})) + return false; diff --git a/tools/net/sunrpc/xdrgen/templates/C/struct/decoder/variable_length_string.j2 b/tools/net/sunrpc/xdrgen/templates/C/struct/decoder/variable_length_string.j2 new file mode 100644 index 000000000000..12d20b143b43 --- /dev/null +++ b/tools/net/sunrpc/xdrgen/templates/C/struct/decoder/variable_length_string.j2 @@ -0,0 +1,6 @@ +{# SPDX-License-Identifier: GPL-2.0 #} +{% if annotate %} + /* member {{ name }} (variable-length string) */ +{% endif %} + if (!xdrgen_decode_string(xdr, (string *)ptr, {{ maxsize }})) + return false; diff --git a/tools/net/sunrpc/xdrgen/templates/C/struct/definition/basic.j2 b/tools/net/sunrpc/xdrgen/templates/C/struct/definition/basic.j2 new file mode 100644 index 000000000000..b3430895f311 --- /dev/null +++ b/tools/net/sunrpc/xdrgen/templates/C/struct/definition/basic.j2 @@ -0,0 +1,5 @@ +{# SPDX-License-Identifier: GPL-2.0 #} +{% if annotate %} + /* (basic) */ +{% endif %} + {{ classifier }}{{ type }} {{ name }}; diff --git a/tools/net/sunrpc/xdrgen/templates/C/struct/definition/close.j2 b/tools/net/sunrpc/xdrgen/templates/C/struct/definition/close.j2 new file mode 100644 index 000000000000..9e62344a976a --- /dev/null +++ b/tools/net/sunrpc/xdrgen/templates/C/struct/definition/close.j2 @@ -0,0 +1,2 @@ +{# SPDX-License-Identifier: GPL-2.0 #} +}; diff --git a/tools/net/sunrpc/xdrgen/templates/C/struct/definition/fixed_length_array.j2 b/tools/net/sunrpc/xdrgen/templates/C/struct/definition/fixed_length_array.j2 new file mode 100644 index 000000000000..66be836826a0 --- /dev/null +++ b/tools/net/sunrpc/xdrgen/templates/C/struct/definition/fixed_length_array.j2 @@ -0,0 +1,5 @@ +{# SPDX-License-Identifier: GPL-2.0 #} +{% if annotate %} + /* (fixed-length array) */ +{% endif %} + {{ type }} {{ name }}[{{ size }}]; diff --git a/tools/net/sunrpc/xdrgen/templates/C/struct/definition/fixed_length_opaque.j2 b/tools/net/sunrpc/xdrgen/templates/C/struct/definition/fixed_length_opaque.j2 new file mode 100644 index 000000000000..0daba19aa0f0 --- /dev/null +++ b/tools/net/sunrpc/xdrgen/templates/C/struct/definition/fixed_length_opaque.j2 @@ -0,0 +1,5 @@ +{# SPDX-License-Identifier: GPL-2.0 #} +{% if annotate %} + /* (fixed-length opaque) */ +{% endif %} + u8 {{ name }}[{{ size }}]; diff --git a/tools/net/sunrpc/xdrgen/templates/C/struct/definition/open.j2 b/tools/net/sunrpc/xdrgen/templates/C/struct/definition/open.j2 new file mode 100644 index 000000000000..07cbf5424546 --- /dev/null +++ b/tools/net/sunrpc/xdrgen/templates/C/struct/definition/open.j2 @@ -0,0 +1,6 @@ +{# SPDX-License-Identifier: GPL-2.0 #} + +{% if annotate %} +/* struct {{ name }} */ +{% endif %} +struct {{ name }} { diff --git a/tools/net/sunrpc/xdrgen/templates/C/struct/definition/optional_data.j2 b/tools/net/sunrpc/xdrgen/templates/C/struct/definition/optional_data.j2 new file mode 100644 index 000000000000..a33341f45e8f --- /dev/null +++ b/tools/net/sunrpc/xdrgen/templates/C/struct/definition/optional_data.j2 @@ -0,0 +1,5 @@ +{# SPDX-License-Identifier: GPL-2.0 #} +{% if annotate %} + /* (optional data) */ +{% endif %} + {{ classifier }}{{ type }} *{{ name }}; diff --git a/tools/net/sunrpc/xdrgen/templates/C/struct/definition/variable_length_array.j2 b/tools/net/sunrpc/xdrgen/templates/C/struct/definition/variable_length_array.j2 new file mode 100644 index 000000000000..5d767f9b3674 --- /dev/null +++ b/tools/net/sunrpc/xdrgen/templates/C/struct/definition/variable_length_array.j2 @@ -0,0 +1,8 @@ +{# SPDX-License-Identifier: GPL-2.0 #} +{% if annotate %} + /* (variable-length array) */ +{% endif %} + struct { + u32 count; + {{ classifier }}{{ type }} *element; + } {{ name }}; diff --git a/tools/net/sunrpc/xdrgen/templates/C/struct/definition/variable_length_opaque.j2 b/tools/net/sunrpc/xdrgen/templates/C/struct/definition/variable_length_opaque.j2 new file mode 100644 index 000000000000..4d0cd84be3db --- /dev/null +++ b/tools/net/sunrpc/xdrgen/templates/C/struct/definition/variable_length_opaque.j2 @@ -0,0 +1,5 @@ +{# SPDX-License-Identifier: GPL-2.0 #} +{% if annotate %} + /* (variable-length opaque) */ +{% endif %} + opaque {{ name }}; diff --git a/tools/net/sunrpc/xdrgen/templates/C/struct/definition/variable_length_string.j2 b/tools/net/sunrpc/xdrgen/templates/C/struct/definition/variable_length_string.j2 new file mode 100644 index 000000000000..2de2feec77db --- /dev/null +++ b/tools/net/sunrpc/xdrgen/templates/C/struct/definition/variable_length_string.j2 @@ -0,0 +1,5 @@ +{# SPDX-License-Identifier: GPL-2.0 #} +{% if annotate %} + /* (variable-length string) */ +{% endif %} + string {{ name }}; diff --git a/tools/net/sunrpc/xdrgen/templates/C/struct/encoder/basic.j2 b/tools/net/sunrpc/xdrgen/templates/C/struct/encoder/basic.j2 new file mode 100644 index 000000000000..a7d3695c5a6a --- /dev/null +++ b/tools/net/sunrpc/xdrgen/templates/C/struct/encoder/basic.j2 @@ -0,0 +1,10 @@ +{# SPDX-License-Identifier: GPL-2.0 #} +{% if annotate %} + /* member {{ name }} (basic) */ +{% endif %} +{% if type in pass_by_reference %} + if (!xdrgen_encode_{{ type }}(xdr, &value->{{ name }})) +{% else %} + if (!xdrgen_encode_{{ type }}(xdr, value->{{ name }})) +{% endif %} + return false; diff --git a/tools/net/sunrpc/xdrgen/templates/C/struct/encoder/close.j2 b/tools/net/sunrpc/xdrgen/templates/C/struct/encoder/close.j2 new file mode 100644 index 000000000000..5bf010665f84 --- /dev/null +++ b/tools/net/sunrpc/xdrgen/templates/C/struct/encoder/close.j2 @@ -0,0 +1,3 @@ +{# SPDX-License-Identifier: GPL-2.0 #} + return true; +}; diff --git a/tools/net/sunrpc/xdrgen/templates/C/struct/encoder/fixed_length_array.j2 b/tools/net/sunrpc/xdrgen/templates/C/struct/encoder/fixed_length_array.j2 new file mode 100644 index 000000000000..b01833a2c7a1 --- /dev/null +++ b/tools/net/sunrpc/xdrgen/templates/C/struct/encoder/fixed_length_array.j2 @@ -0,0 +1,12 @@ +{# SPDX-License-Identifier: GPL-2.0 #} +{% if annotate %} + /* member {{ name }} (fixed-length array) */ +{% endif %} + for (u32 i = 0; i < {{ size }}; i++) { +{% if type in pass_by_reference %} + if (xdrgen_encode_{{ type }}(xdr, &value->items[i]) < 0) +{% else %} + if (xdrgen_encode_{{ type }}(xdr, value->items[i]) < 0) +{% endif %} + return false; + } diff --git a/tools/net/sunrpc/xdrgen/templates/C/struct/encoder/fixed_length_opaque.j2 b/tools/net/sunrpc/xdrgen/templates/C/struct/encoder/fixed_length_opaque.j2 new file mode 100644 index 000000000000..07bc91919898 --- /dev/null +++ b/tools/net/sunrpc/xdrgen/templates/C/struct/encoder/fixed_length_opaque.j2 @@ -0,0 +1,6 @@ +{# SPDX-License-Identifier: GPL-2.0 #} +{% if annotate %} + /* member {{ name }} (fixed-length opaque) */ +{% endif %} + if (xdr_stream_encode_opaque_fixed(xdr, value->{{ name }}, {{ size }}) < 0) + return false; diff --git a/tools/net/sunrpc/xdrgen/templates/C/struct/encoder/open.j2 b/tools/net/sunrpc/xdrgen/templates/C/struct/encoder/open.j2 new file mode 100644 index 000000000000..2286a3adf82a --- /dev/null +++ b/tools/net/sunrpc/xdrgen/templates/C/struct/encoder/open.j2 @@ -0,0 +1,12 @@ +{# SPDX-License-Identifier: GPL-2.0 #} + +{% if annotate %} +/* struct {{ name }} */ +{% endif %} +{% if name in public_apis %} +bool +{% else %} +static bool __maybe_unused +{% endif %} +xdrgen_encode_{{ name }}(struct xdr_stream *xdr, const struct {{ name }} *value) +{ diff --git a/tools/net/sunrpc/xdrgen/templates/C/struct/encoder/optional_data.j2 b/tools/net/sunrpc/xdrgen/templates/C/struct/encoder/optional_data.j2 new file mode 100644 index 000000000000..16fb3e09bba1 --- /dev/null +++ b/tools/net/sunrpc/xdrgen/templates/C/struct/encoder/optional_data.j2 @@ -0,0 +1,6 @@ +{# SPDX-License-Identifier: GPL-2.0 #} +{% if annotate %} + /* member {{ name }} (optional data) */ +{% endif %} + if (!xdrgen_encode_{{ type }}(xdr, value->{{ name }})) + return false; diff --git a/tools/net/sunrpc/xdrgen/templates/C/struct/encoder/variable_length_array.j2 b/tools/net/sunrpc/xdrgen/templates/C/struct/encoder/variable_length_array.j2 new file mode 100644 index 000000000000..0ec8660d621a --- /dev/null +++ b/tools/net/sunrpc/xdrgen/templates/C/struct/encoder/variable_length_array.j2 @@ -0,0 +1,15 @@ +{# SPDX-License-Identifier: GPL-2.0 #} +{% if annotate %} + /* member {{ name }} (variable-length array) */ +{% endif %} + if (value->{{ name }}.count > {{ maxsize }}) + return false; + if (xdr_stream_encode_u32(xdr, value->{{ name }}.count) != XDR_UNIT) + return false; + for (u32 i = 0; i < value->{{ name }}.count; i++) +{% if type in pass_by_reference %} + if (!xdrgen_encode_{{ type }}(xdr, &value->{{ name }}.element[i])) +{% else %} + if (!xdrgen_encode_{{ type }}(xdr, value->{{ name }}.element[i])) +{% endif %} + return false; diff --git a/tools/net/sunrpc/xdrgen/templates/C/struct/encoder/variable_length_opaque.j2 b/tools/net/sunrpc/xdrgen/templates/C/struct/encoder/variable_length_opaque.j2 new file mode 100644 index 000000000000..1d477c2d197a --- /dev/null +++ b/tools/net/sunrpc/xdrgen/templates/C/struct/encoder/variable_length_opaque.j2 @@ -0,0 +1,8 @@ +{# SPDX-License-Identifier: GPL-2.0 #} +{% if annotate %} + /* member {{ name }} (variable-length opaque) */ +{% endif %} + if (value->{{ name }}.len > {{ maxsize }}) + return false; + if (xdr_stream_encode_opaque(xdr, value->{{ name }}.data, value->{{ name }}.len) < 0) + return false; diff --git a/tools/net/sunrpc/xdrgen/templates/C/struct/encoder/variable_length_string.j2 b/tools/net/sunrpc/xdrgen/templates/C/struct/encoder/variable_length_string.j2 new file mode 100644 index 000000000000..cf65b71eaef3 --- /dev/null +++ b/tools/net/sunrpc/xdrgen/templates/C/struct/encoder/variable_length_string.j2 @@ -0,0 +1,8 @@ +{# SPDX-License-Identifier: GPL-2.0 #} +{% if annotate %} + /* member {{ name }} (variable-length string) */ +{% endif %} + if (value->{{ name }}.len > {{ maxsize }}) + return false; + if (xdr_stream_encode_opaque(xdr, value->{{ name }}.data, value->{{ name }}.len) < 0) + return false; diff --git a/tools/net/sunrpc/xdrgen/templates/C/typedef/declaration/basic.j2 b/tools/net/sunrpc/xdrgen/templates/C/typedef/declaration/basic.j2 new file mode 100644 index 000000000000..455b10bd90ec --- /dev/null +++ b/tools/net/sunrpc/xdrgen/templates/C/typedef/declaration/basic.j2 @@ -0,0 +1,8 @@ +{# SPDX-License-Identifier: GPL-2.0 #} + +bool xdrgen_decode_{{ name }}(struct xdr_stream *xdr, {{ name }} *ptr); +{% if name in pass_by_reference %} +bool xdrgen_encode_{{ name }}(struct xdr_stream *xdr, const {{ name }} *value); +{%- else -%} +bool xdrgen_encode_{{ name }}(struct xdr_stream *xdr, const {{ name }} value); +{% endif %} diff --git a/tools/net/sunrpc/xdrgen/templates/C/typedef/declaration/fixed_length_array.j2 b/tools/net/sunrpc/xdrgen/templates/C/typedef/declaration/fixed_length_array.j2 new file mode 100644 index 000000000000..3fe3ddd9f359 --- /dev/null +++ b/tools/net/sunrpc/xdrgen/templates/C/typedef/declaration/fixed_length_array.j2 @@ -0,0 +1,4 @@ +{# SPDX-License-Identifier: GPL-2.0 #} + +bool xdrgen_decode_{{ name }}(struct xdr_stream *xdr, {{ classifier }}{{ name }} *ptr); +bool xdrgen_encode_{{ name }}(struct xdr_stream *xdr, const {{ classifier }}{{ name }} value); diff --git a/tools/net/sunrpc/xdrgen/templates/C/typedef/declaration/fixed_length_opaque.j2 b/tools/net/sunrpc/xdrgen/templates/C/typedef/declaration/fixed_length_opaque.j2 new file mode 100644 index 000000000000..3fe3ddd9f359 --- /dev/null +++ b/tools/net/sunrpc/xdrgen/templates/C/typedef/declaration/fixed_length_opaque.j2 @@ -0,0 +1,4 @@ +{# SPDX-License-Identifier: GPL-2.0 #} + +bool xdrgen_decode_{{ name }}(struct xdr_stream *xdr, {{ classifier }}{{ name }} *ptr); +bool xdrgen_encode_{{ name }}(struct xdr_stream *xdr, const {{ classifier }}{{ name }} value); diff --git a/tools/net/sunrpc/xdrgen/templates/C/typedef/declaration/variable_length_array.j2 b/tools/net/sunrpc/xdrgen/templates/C/typedef/declaration/variable_length_array.j2 new file mode 100644 index 000000000000..3fe3ddd9f359 --- /dev/null +++ b/tools/net/sunrpc/xdrgen/templates/C/typedef/declaration/variable_length_array.j2 @@ -0,0 +1,4 @@ +{# SPDX-License-Identifier: GPL-2.0 #} + +bool xdrgen_decode_{{ name }}(struct xdr_stream *xdr, {{ classifier }}{{ name }} *ptr); +bool xdrgen_encode_{{ name }}(struct xdr_stream *xdr, const {{ classifier }}{{ name }} value); diff --git a/tools/net/sunrpc/xdrgen/templates/C/typedef/declaration/variable_length_opaque.j2 b/tools/net/sunrpc/xdrgen/templates/C/typedef/declaration/variable_length_opaque.j2 new file mode 100644 index 000000000000..3fe3ddd9f359 --- /dev/null +++ b/tools/net/sunrpc/xdrgen/templates/C/typedef/declaration/variable_length_opaque.j2 @@ -0,0 +1,4 @@ +{# SPDX-License-Identifier: GPL-2.0 #} + +bool xdrgen_decode_{{ name }}(struct xdr_stream *xdr, {{ classifier }}{{ name }} *ptr); +bool xdrgen_encode_{{ name }}(struct xdr_stream *xdr, const {{ classifier }}{{ name }} value); diff --git a/tools/net/sunrpc/xdrgen/templates/C/typedef/declaration/variable_length_string.j2 b/tools/net/sunrpc/xdrgen/templates/C/typedef/declaration/variable_length_string.j2 new file mode 100644 index 000000000000..3fe3ddd9f359 --- /dev/null +++ b/tools/net/sunrpc/xdrgen/templates/C/typedef/declaration/variable_length_string.j2 @@ -0,0 +1,4 @@ +{# SPDX-License-Identifier: GPL-2.0 #} + +bool xdrgen_decode_{{ name }}(struct xdr_stream *xdr, {{ classifier }}{{ name }} *ptr); +bool xdrgen_encode_{{ name }}(struct xdr_stream *xdr, const {{ classifier }}{{ name }} value); diff --git a/tools/net/sunrpc/xdrgen/templates/C/typedef/decoder/basic.j2 b/tools/net/sunrpc/xdrgen/templates/C/typedef/decoder/basic.j2 new file mode 100644 index 000000000000..da4709403dc9 --- /dev/null +++ b/tools/net/sunrpc/xdrgen/templates/C/typedef/decoder/basic.j2 @@ -0,0 +1,17 @@ +{# SPDX-License-Identifier: GPL-2.0 #} + +{% if annotate %} +/* typedef {{ name }} */ +{% endif %} +{% if name in public_apis %} +bool +{% else %} +static bool __maybe_unused +{% endif %} +xdrgen_decode_{{ name }}(struct xdr_stream *xdr, {{ name }} *ptr) +{ +{% if annotate %} + /* (basic) */ +{% endif %} + return xdrgen_decode_{{ type }}(xdr, ptr); +}; diff --git a/tools/net/sunrpc/xdrgen/templates/C/typedef/decoder/fixed_length_array.j2 b/tools/net/sunrpc/xdrgen/templates/C/typedef/decoder/fixed_length_array.j2 new file mode 100644 index 000000000000..d7c80e472fe3 --- /dev/null +++ b/tools/net/sunrpc/xdrgen/templates/C/typedef/decoder/fixed_length_array.j2 @@ -0,0 +1,25 @@ +{# SPDX-License-Identifier: GPL-2.0 #} + +{% if annotate %} +/* typedef {{ name }} */ +{% endif %} +{% if name in public_apis %} +bool +{% else %} +static bool __maybe_unused +{% endif %} +xdrgen_decode_{{ name }}(struct xdr_stream *xdr, {{ classifier }}{{ name }} *ptr) +{ +{% if annotate %} + /* (fixed-length array) */ +{% endif %} + for (u32 i = 0; i < {{ size }}; i++) { +{%- if classifier == '' %} + if (xdrgen_decode_{{ type }}(xdr, ptr->items[i]) < 0) +{% else %} + if (xdrgen_decode_{{ type }}(xdr, &ptr->items[i]) < 0) +{% endif %} + return false; + } + return true; +}; diff --git a/tools/net/sunrpc/xdrgen/templates/C/typedef/decoder/fixed_length_opaque.j2 b/tools/net/sunrpc/xdrgen/templates/C/typedef/decoder/fixed_length_opaque.j2 new file mode 100644 index 000000000000..8b4ff08c49e5 --- /dev/null +++ b/tools/net/sunrpc/xdrgen/templates/C/typedef/decoder/fixed_length_opaque.j2 @@ -0,0 +1,17 @@ +{# SPDX-License-Identifier: GPL-2.0 #} + +{% if annotate %} +/* typedef {{ name }} */ +{% endif %} +{% if name in public_apis %} +bool +{% else %} +static bool __maybe_unused +{% endif %} +xdrgen_decode_{{ name }}(struct xdr_stream *xdr, {{ classifier }}{{ name }} *ptr) +{ +{% if annotate %} + /* (fixed-length opaque) */ +{% endif %} + return xdr_stream_decode_opaque_fixed(xdr, ptr, {{ size }}) >= 0; +}; diff --git a/tools/net/sunrpc/xdrgen/templates/C/typedef/decoder/variable_length_array.j2 b/tools/net/sunrpc/xdrgen/templates/C/typedef/decoder/variable_length_array.j2 new file mode 100644 index 000000000000..e74ffdd98463 --- /dev/null +++ b/tools/net/sunrpc/xdrgen/templates/C/typedef/decoder/variable_length_array.j2 @@ -0,0 +1,26 @@ +{# SPDX-License-Identifier: GPL-2.0 #} + +{% if annotate %} +/* typedef {{ name }} */ +{% endif %} +{% if name in public_apis %} +bool +{% else %} +static bool __maybe_unused +{% endif %} +xdrgen_decode_{{ name }}(struct xdr_stream *xdr, {{ classifier }}{{ name }} *ptr) +{ +{% if annotate %} + /* (variable-length array) */ +{% endif %} + if (xdr_stream_decode_u32(xdr, &ptr->count) < 0) + return false; +{% if maxsize != "0" %} + if (ptr->count > {{ maxsize }}) + return false; +{% endif %} + for (u32 i = 0; i < ptr->count; i++) + if (!xdrgen_decode_{{ type }}(xdr, &ptr->element[i])) + return false; + return true; +}; diff --git a/tools/net/sunrpc/xdrgen/templates/C/typedef/decoder/variable_length_opaque.j2 b/tools/net/sunrpc/xdrgen/templates/C/typedef/decoder/variable_length_opaque.j2 new file mode 100644 index 000000000000..f28f8b228ad5 --- /dev/null +++ b/tools/net/sunrpc/xdrgen/templates/C/typedef/decoder/variable_length_opaque.j2 @@ -0,0 +1,17 @@ +{# SPDX-License-Identifier: GPL-2.0 #} + +{% if annotate %} +/* typedef {{ name }} */ +{% endif %} +{% if name in public_apis %} +bool +{% else %} +static bool __maybe_unused +{% endif %} +xdrgen_decode_{{ name }}(struct xdr_stream *xdr, {{ classifier }}{{ name }} *ptr) +{ +{% if annotate %} + /* (variable-length opaque) */ +{% endif %} + return xdrgen_decode_opaque(xdr, ptr, {{ maxsize }}); +}; diff --git a/tools/net/sunrpc/xdrgen/templates/C/typedef/decoder/variable_length_string.j2 b/tools/net/sunrpc/xdrgen/templates/C/typedef/decoder/variable_length_string.j2 new file mode 100644 index 000000000000..56c5a17d6a70 --- /dev/null +++ b/tools/net/sunrpc/xdrgen/templates/C/typedef/decoder/variable_length_string.j2 @@ -0,0 +1,17 @@ +{# SPDX-License-Identifier: GPL-2.0 #} + +{% if annotate %} +/* typedef {{ name }} */ +{% endif %} +{% if name in public_apis %} +bool +{% else %} +static bool __maybe_unused +{% endif %} +xdrgen_decode_{{ name }}(struct xdr_stream *xdr, {{ classifier }}{{ name }} *ptr) +{ +{% if annotate %} + /* (variable-length string) */ +{% endif %} + return xdrgen_decode_string(xdr, ptr, {{ maxsize }}); +}; diff --git a/tools/net/sunrpc/xdrgen/templates/C/typedef/definition/basic.j2 b/tools/net/sunrpc/xdrgen/templates/C/typedef/definition/basic.j2 new file mode 100644 index 000000000000..1c5f28135eec --- /dev/null +++ b/tools/net/sunrpc/xdrgen/templates/C/typedef/definition/basic.j2 @@ -0,0 +1,6 @@ +{# SPDX-License-Identifier: GPL-2.0 #} + +{% if annotate %} +/* typedef {{ name }} (basic) */ +{% endif %} +typedef {{ classifier }}{{ type }} {{ name }}; diff --git a/tools/net/sunrpc/xdrgen/templates/C/typedef/definition/fixed_length_array.j2 b/tools/net/sunrpc/xdrgen/templates/C/typedef/definition/fixed_length_array.j2 new file mode 100644 index 000000000000..c3a67c952e77 --- /dev/null +++ b/tools/net/sunrpc/xdrgen/templates/C/typedef/definition/fixed_length_array.j2 @@ -0,0 +1,6 @@ +{# SPDX-License-Identifier: GPL-2.0 #} + +{% if annotate %} +/* typedef {{ name }} (fixed-length array) */ +{% endif %} +typedef {{ type }}{{ name }}[{{ size }}]; diff --git a/tools/net/sunrpc/xdrgen/templates/C/typedef/definition/fixed_length_opaque.j2 b/tools/net/sunrpc/xdrgen/templates/C/typedef/definition/fixed_length_opaque.j2 new file mode 100644 index 000000000000..8788b02fe4f5 --- /dev/null +++ b/tools/net/sunrpc/xdrgen/templates/C/typedef/definition/fixed_length_opaque.j2 @@ -0,0 +1,6 @@ +{# SPDX-License-Identifier: GPL-2.0 #} + +{% if annotate %} +/* typedef {{ name }} (fixed-length opaque) */ +{% endif %} +typedef u8 {{ name }}[{{ size }}]; diff --git a/tools/net/sunrpc/xdrgen/templates/C/typedef/definition/variable_length_array.j2 b/tools/net/sunrpc/xdrgen/templates/C/typedef/definition/variable_length_array.j2 new file mode 100644 index 000000000000..f03393760545 --- /dev/null +++ b/tools/net/sunrpc/xdrgen/templates/C/typedef/definition/variable_length_array.j2 @@ -0,0 +1,9 @@ +{# SPDX-License-Identifier: GPL-2.0 #} + +{% if annotate %} +/* typedef {{ name }} (variable-length array) */ +{% endif %} +typedef struct { + u32 count; + {{ classifier }}{{ type }} *element; +} {{ name }}; diff --git a/tools/net/sunrpc/xdrgen/templates/C/typedef/definition/variable_length_opaque.j2 b/tools/net/sunrpc/xdrgen/templates/C/typedef/definition/variable_length_opaque.j2 new file mode 100644 index 000000000000..162f2610af34 --- /dev/null +++ b/tools/net/sunrpc/xdrgen/templates/C/typedef/definition/variable_length_opaque.j2 @@ -0,0 +1,6 @@ +{# SPDX-License-Identifier: GPL-2.0 #} + +{% if annotate %} +/* typedef {{ name }} (variable-length opaque) */ +{% endif %} +typedef opaque {{ name }}; diff --git a/tools/net/sunrpc/xdrgen/templates/C/typedef/definition/variable_length_string.j2 b/tools/net/sunrpc/xdrgen/templates/C/typedef/definition/variable_length_string.j2 new file mode 100644 index 000000000000..c03c2df8e625 --- /dev/null +++ b/tools/net/sunrpc/xdrgen/templates/C/typedef/definition/variable_length_string.j2 @@ -0,0 +1,6 @@ +{# SPDX-License-Identifier: GPL-2.0 #} + +{% if annotate %} +/* typedef {{ name }} (variable-length string) */ +{% endif %} +typedef string {{ name }}; diff --git a/tools/net/sunrpc/xdrgen/templates/C/typedef/encoder/basic.j2 b/tools/net/sunrpc/xdrgen/templates/C/typedef/encoder/basic.j2 new file mode 100644 index 000000000000..35effe67e4ef --- /dev/null +++ b/tools/net/sunrpc/xdrgen/templates/C/typedef/encoder/basic.j2 @@ -0,0 +1,21 @@ +{# SPDX-License-Identifier: GPL-2.0 #} + +{% if annotate %} +/* typedef {{ name }} */ +{% endif %} +{% if name in public_apis %} +bool +{% else %} +static bool __maybe_unused +{% endif %} +{% if name in pass_by_reference %} +xdrgen_encode_{{ name }}(struct xdr_stream *xdr, const {{ classifier }}{{ name }} *value) +{% else %} +xdrgen_encode_{{ name }}(struct xdr_stream *xdr, const {{ classifier }}{{ name }} value) +{% endif %} +{ +{% if annotate %} + /* (basic) */ +{% endif %} + return xdrgen_encode_{{ type }}(xdr, value); +}; diff --git a/tools/net/sunrpc/xdrgen/templates/C/typedef/encoder/fixed_length_array.j2 b/tools/net/sunrpc/xdrgen/templates/C/typedef/encoder/fixed_length_array.j2 new file mode 100644 index 000000000000..95202ad5ad2d --- /dev/null +++ b/tools/net/sunrpc/xdrgen/templates/C/typedef/encoder/fixed_length_array.j2 @@ -0,0 +1,25 @@ +{# SPDX-License-Identifier: GPL-2.0 #} + +{% if annotate %} +/* typedef {{ name }} */ +{% endif %} +{% if name in public_apis %} +bool +{% else %} +static bool __maybe_unused +{% endif %} +xdrgen_encode_{{ name }}(struct xdr_stream *xdr, const {{ classifier }}{{ name }} value) +{ +{% if annotate %} + /* (fixed-length array) */ +{% endif %} + for (u32 i = 0; i < {{ size }}; i++) { +{% if type in pass_by_reference %} + if (xdrgen_encode_{{ type }}(xdr, &value->items[i]) < 0) +{% else %} + if (xdrgen_encode_{{ type }}(xdr, value->items[i]) < 0) +{% endif %} + return false; + } + return true; +}; diff --git a/tools/net/sunrpc/xdrgen/templates/C/typedef/encoder/fixed_length_opaque.j2 b/tools/net/sunrpc/xdrgen/templates/C/typedef/encoder/fixed_length_opaque.j2 new file mode 100644 index 000000000000..9c66a11b9912 --- /dev/null +++ b/tools/net/sunrpc/xdrgen/templates/C/typedef/encoder/fixed_length_opaque.j2 @@ -0,0 +1,17 @@ +{# SPDX-License-Identifier: GPL-2.0 #} + +{% if annotate %} +/* typedef {{ name }} */ +{% endif %} +{% if name in public_apis %} +bool +{% else %} +static bool __maybe_unused +{% endif %} +xdrgen_encode_{{ name }}(struct xdr_stream *xdr, const {{ classifier }}{{ name }} value) +{ +{% if annotate %} + /* (fixed-length opaque) */ +{% endif %} + return xdr_stream_encode_opaque_fixed(xdr, value, {{ size }}) >= 0; +}; diff --git a/tools/net/sunrpc/xdrgen/templates/C/typedef/encoder/variable_length_array.j2 b/tools/net/sunrpc/xdrgen/templates/C/typedef/encoder/variable_length_array.j2 new file mode 100644 index 000000000000..2d2384f64918 --- /dev/null +++ b/tools/net/sunrpc/xdrgen/templates/C/typedef/encoder/variable_length_array.j2 @@ -0,0 +1,30 @@ +{# SPDX-License-Identifier: GPL-2.0 #} + +{% if annotate %} +/* typedef {{ name }} */ +{% endif %} +{% if name in public_apis %} +bool +{% else %} +static bool __maybe_unused +{% endif %} +xdrgen_encode_{{ name }}(struct xdr_stream *xdr, const {{ classifier }}{{ name }} value) +{ +{% if annotate %} + /* (variable-length array) */ +{% endif %} +{% if maxsize != "0" %} + if (unlikely(value.count > {{ maxsize }})) + return false; +{% endif %} + if (xdr_stream_encode_u32(xdr, value.count) != XDR_UNIT) + return false; + for (u32 i = 0; i < value.count; i++) +{% if type in pass_by_reference %} + if (!xdrgen_encode_{{ type }}(xdr, &value.element[i])) +{% else %} + if (!xdrgen_encode_{{ type }}(xdr, value.element[i])) +{% endif %} + return false; + return true; +}; diff --git a/tools/net/sunrpc/xdrgen/templates/C/typedef/encoder/variable_length_opaque.j2 b/tools/net/sunrpc/xdrgen/templates/C/typedef/encoder/variable_length_opaque.j2 new file mode 100644 index 000000000000..8508f13c95b9 --- /dev/null +++ b/tools/net/sunrpc/xdrgen/templates/C/typedef/encoder/variable_length_opaque.j2 @@ -0,0 +1,17 @@ +{# SPDX-License-Identifier: GPL-2.0 #} + +{% if annotate %} +/* typedef {{ name }} */ +{% endif %} +{% if name in public_apis %} +bool +{% else %} +static bool __maybe_unused +{% endif %} +xdrgen_encode_{{ name }}(struct xdr_stream *xdr, const {{ classifier }}{{ name }} value) +{ +{% if annotate %} + /* (variable-length opaque) */ +{% endif %} + return xdr_stream_encode_opaque(xdr, value.data, value.len) >= 0; +}; diff --git a/tools/net/sunrpc/xdrgen/templates/C/typedef/encoder/variable_length_string.j2 b/tools/net/sunrpc/xdrgen/templates/C/typedef/encoder/variable_length_string.j2 new file mode 100644 index 000000000000..3d490ff180d0 --- /dev/null +++ b/tools/net/sunrpc/xdrgen/templates/C/typedef/encoder/variable_length_string.j2 @@ -0,0 +1,17 @@ +{# SPDX-License-Identifier: GPL-2.0 #} + +{% if annotate %} +/* typedef {{ name }} */ +{% endif %} +{% if name in public_apis %} +bool +{% else %} +static bool __maybe_unused +{% endif %} +xdrgen_encode_{{ name }}(struct xdr_stream *xdr, const {{ classifier }}{{ name }} value) +{ +{% if annotate %} + /* (variable-length string) */ +{% endif %} + return xdr_stream_encode_opaque(xdr, value.data, value.len) >= 0; +}; diff --git a/tools/net/sunrpc/xdrgen/templates/C/union/decoder/basic.j2 b/tools/net/sunrpc/xdrgen/templates/C/union/decoder/basic.j2 new file mode 100644 index 000000000000..4d97cc5395eb --- /dev/null +++ b/tools/net/sunrpc/xdrgen/templates/C/union/decoder/basic.j2 @@ -0,0 +1,6 @@ +{# SPDX-License-Identifier: GPL-2.0 #} +{% if annotate %} + /* member {{ name }} (basic) */ +{% endif %} + if (!xdrgen_decode_{{ type }}(xdr, &ptr->u.{{ name }})) + return false; diff --git a/tools/net/sunrpc/xdrgen/templates/C/union/decoder/break.j2 b/tools/net/sunrpc/xdrgen/templates/C/union/decoder/break.j2 new file mode 100644 index 000000000000..b286d1407029 --- /dev/null +++ b/tools/net/sunrpc/xdrgen/templates/C/union/decoder/break.j2 @@ -0,0 +1,2 @@ +{# SPDX-License-Identifier: GPL-2.0 #} + break; diff --git a/tools/net/sunrpc/xdrgen/templates/C/union/decoder/case_spec.j2 b/tools/net/sunrpc/xdrgen/templates/C/union/decoder/case_spec.j2 new file mode 100644 index 000000000000..5fa2163f0a74 --- /dev/null +++ b/tools/net/sunrpc/xdrgen/templates/C/union/decoder/case_spec.j2 @@ -0,0 +1,2 @@ +{# SPDX-License-Identifier: GPL-2.0 #} + case {{ case }}: diff --git a/tools/net/sunrpc/xdrgen/templates/C/union/decoder/close.j2 b/tools/net/sunrpc/xdrgen/templates/C/union/decoder/close.j2 new file mode 100644 index 000000000000..fdc2dfd1843b --- /dev/null +++ b/tools/net/sunrpc/xdrgen/templates/C/union/decoder/close.j2 @@ -0,0 +1,4 @@ +{# SPDX-License-Identifier: GPL-2.0 #} + } + return true; +}; diff --git a/tools/net/sunrpc/xdrgen/templates/C/union/decoder/default_spec.j2 b/tools/net/sunrpc/xdrgen/templates/C/union/decoder/default_spec.j2 new file mode 100644 index 000000000000..044a002d0589 --- /dev/null +++ b/tools/net/sunrpc/xdrgen/templates/C/union/decoder/default_spec.j2 @@ -0,0 +1,2 @@ +{# SPDX-License-Identifier: GPL-2.0 #} + default: diff --git a/tools/net/sunrpc/xdrgen/templates/C/union/decoder/open.j2 b/tools/net/sunrpc/xdrgen/templates/C/union/decoder/open.j2 new file mode 100644 index 000000000000..eb9941376e49 --- /dev/null +++ b/tools/net/sunrpc/xdrgen/templates/C/union/decoder/open.j2 @@ -0,0 +1,12 @@ +{# SPDX-License-Identifier: GPL-2.0 #} + +{% if annotate %} +/* union {{ name }} */ +{% endif %} +{% if name in public_apis %} +bool +{% else %} +static bool __maybe_unused +{% endif %} +xdrgen_decode_{{ name }}(struct xdr_stream *xdr, struct {{ name }} *ptr) +{ diff --git a/tools/net/sunrpc/xdrgen/templates/C/union/decoder/optional_data.j2 b/tools/net/sunrpc/xdrgen/templates/C/union/decoder/optional_data.j2 new file mode 100644 index 000000000000..e4476f5fd8d3 --- /dev/null +++ b/tools/net/sunrpc/xdrgen/templates/C/union/decoder/optional_data.j2 @@ -0,0 +1,6 @@ +{# SPDX-License-Identifier: GPL-2.0 #} +{% if annotate %} + /* member {{ name }} (optional data) */ +{% endif %} + if (!xdrgen_decode_{{ type }}(xdr, &ptr->u.{{ name }})) + return false; diff --git a/tools/net/sunrpc/xdrgen/templates/C/union/decoder/switch_spec.j2 b/tools/net/sunrpc/xdrgen/templates/C/union/decoder/switch_spec.j2 new file mode 100644 index 000000000000..99b3067ef617 --- /dev/null +++ b/tools/net/sunrpc/xdrgen/templates/C/union/decoder/switch_spec.j2 @@ -0,0 +1,7 @@ +{# SPDX-License-Identifier: GPL-2.0 #} +{% if annotate %} + /* discriminant {{ name }} */ +{% endif %} + if (!xdrgen_decode_{{ type }}(xdr, &ptr->{{ name }})) + return false; + switch (ptr->{{ name }}) { diff --git a/tools/net/sunrpc/xdrgen/templates/C/union/decoder/variable_length_array.j2 b/tools/net/sunrpc/xdrgen/templates/C/union/decoder/variable_length_array.j2 new file mode 100644 index 000000000000..51ad736d2530 --- /dev/null +++ b/tools/net/sunrpc/xdrgen/templates/C/union/decoder/variable_length_array.j2 @@ -0,0 +1,13 @@ +{# SPDX-License-Identifier: GPL-2.0 #} +{% if annotate %} + /* member {{ name }} (variable-length array) */ +{% endif %} + if (xdr_stream_decode_u32(xdr, &count) < 0) + return false; + if (count > {{ maxsize }}) + return false; + for (u32 i = 0; i < count; i++) { + if (xdrgen_decode_{{ type }}(xdr, &ptr->{{ name }}.items[i]) < 0) + return false; + } + ptr->{{ name }}.len = count; diff --git a/tools/net/sunrpc/xdrgen/templates/C/union/decoder/variable_length_opaque.j2 b/tools/net/sunrpc/xdrgen/templates/C/union/decoder/variable_length_opaque.j2 new file mode 100644 index 000000000000..c9d88ed29c78 --- /dev/null +++ b/tools/net/sunrpc/xdrgen/templates/C/union/decoder/variable_length_opaque.j2 @@ -0,0 +1,6 @@ +{# SPDX-License-Identifier: GPL-2.0 #} +{% if annotate %} + /* member {{ name }} (variable-length opaque) */ +{% endif %} + if (!xdrgen_decode_opaque(xdr, (struct opaque *)ptr->u.{{ name }}, {{ maxsize }})) + return false; diff --git a/tools/net/sunrpc/xdrgen/templates/C/union/decoder/variable_length_string.j2 b/tools/net/sunrpc/xdrgen/templates/C/union/decoder/variable_length_string.j2 new file mode 100644 index 000000000000..83b6e5a14e7f --- /dev/null +++ b/tools/net/sunrpc/xdrgen/templates/C/union/decoder/variable_length_string.j2 @@ -0,0 +1,6 @@ +{# SPDX-License-Identifier: GPL-2.0 #} +{% if annotate %} + /* member {{ name }} (variable-length string) */ +{% endif %} + if (!xdrgen_decode_string(xdr, (struct string *)ptr->u.{{ name }}, {{ maxsize }})) + return false; diff --git a/tools/net/sunrpc/xdrgen/templates/C/union/decoder/void.j2 b/tools/net/sunrpc/xdrgen/templates/C/union/decoder/void.j2 new file mode 100644 index 000000000000..65205ce37b36 --- /dev/null +++ b/tools/net/sunrpc/xdrgen/templates/C/union/decoder/void.j2 @@ -0,0 +1,3 @@ +{# SPDX-License-Identifier: GPL-2.0 #} + if (!xdrgen_decode_void(xdr)) + return false; diff --git a/tools/net/sunrpc/xdrgen/templates/C/union/definition/case_spec.j2 b/tools/net/sunrpc/xdrgen/templates/C/union/definition/case_spec.j2 new file mode 100644 index 000000000000..52f8d131b805 --- /dev/null +++ b/tools/net/sunrpc/xdrgen/templates/C/union/definition/case_spec.j2 @@ -0,0 +1,2 @@ +{# SPDX-License-Identifier: GPL-2.0 #} + {{ classifier }}{{ type }} {{ name }}; diff --git a/tools/net/sunrpc/xdrgen/templates/C/union/definition/close.j2 b/tools/net/sunrpc/xdrgen/templates/C/union/definition/close.j2 new file mode 100644 index 000000000000..01d716d0099e --- /dev/null +++ b/tools/net/sunrpc/xdrgen/templates/C/union/definition/close.j2 @@ -0,0 +1,8 @@ +{# SPDX-License-Identifier: GPL-2.0 #} + } u; +}; +{%- if name in public_apis %} + +bool xdrgen_decode_{{ name }}(struct xdr_stream *xdr, struct {{ name }} *ptr); +bool xdrgen_encode_{{ name }}(struct xdr_stream *xdr, const struct {{ name }} *ptr); +{%- endif -%} diff --git a/tools/net/sunrpc/xdrgen/templates/C/union/definition/default_spec.j2 b/tools/net/sunrpc/xdrgen/templates/C/union/definition/default_spec.j2 new file mode 100644 index 000000000000..52f8d131b805 --- /dev/null +++ b/tools/net/sunrpc/xdrgen/templates/C/union/definition/default_spec.j2 @@ -0,0 +1,2 @@ +{# SPDX-License-Identifier: GPL-2.0 #} + {{ classifier }}{{ type }} {{ name }}; diff --git a/tools/net/sunrpc/xdrgen/templates/C/union/definition/open.j2 b/tools/net/sunrpc/xdrgen/templates/C/union/definition/open.j2 new file mode 100644 index 000000000000..20fcfd1fc4e5 --- /dev/null +++ b/tools/net/sunrpc/xdrgen/templates/C/union/definition/open.j2 @@ -0,0 +1,6 @@ +{# SPDX-License-Identifier: GPL-2.0 #} + +{% if annotate %} +/* union {{ name }} */ +{% endif %} +struct {{ name }} { diff --git a/tools/net/sunrpc/xdrgen/templates/C/union/definition/switch_spec.j2 b/tools/net/sunrpc/xdrgen/templates/C/union/definition/switch_spec.j2 new file mode 100644 index 000000000000..3e552732502c --- /dev/null +++ b/tools/net/sunrpc/xdrgen/templates/C/union/definition/switch_spec.j2 @@ -0,0 +1,3 @@ +{# SPDX-License-Identifier: GPL-2.0 #} + {{ classifier }}{{ type }} {{ name }}; + union { diff --git a/tools/net/sunrpc/xdrgen/templates/C/union/encoder/basic.j2 b/tools/net/sunrpc/xdrgen/templates/C/union/encoder/basic.j2 new file mode 100644 index 000000000000..6452d75c6f9a --- /dev/null +++ b/tools/net/sunrpc/xdrgen/templates/C/union/encoder/basic.j2 @@ -0,0 +1,10 @@ +{# SPDX-License-Identifier: GPL-2.0 #} +{% if annotate %} + /* member {{ name }} (basic) */ +{% endif %} +{% if type in pass_by_reference %} + if (!xdrgen_encode_{{ type }}(xdr, &ptr->u.{{ name }})) +{% else %} + if (!xdrgen_encode_{{ type }}(xdr, ptr->u.{{ name }})) +{% endif %} + return false; diff --git a/tools/net/sunrpc/xdrgen/templates/C/union/encoder/break.j2 b/tools/net/sunrpc/xdrgen/templates/C/union/encoder/break.j2 new file mode 100644 index 000000000000..b286d1407029 --- /dev/null +++ b/tools/net/sunrpc/xdrgen/templates/C/union/encoder/break.j2 @@ -0,0 +1,2 @@ +{# SPDX-License-Identifier: GPL-2.0 #} + break; diff --git a/tools/net/sunrpc/xdrgen/templates/C/union/encoder/case_spec.j2 b/tools/net/sunrpc/xdrgen/templates/C/union/encoder/case_spec.j2 new file mode 100644 index 000000000000..5fa2163f0a74 --- /dev/null +++ b/tools/net/sunrpc/xdrgen/templates/C/union/encoder/case_spec.j2 @@ -0,0 +1,2 @@ +{# SPDX-License-Identifier: GPL-2.0 #} + case {{ case }}: diff --git a/tools/net/sunrpc/xdrgen/templates/C/union/encoder/close.j2 b/tools/net/sunrpc/xdrgen/templates/C/union/encoder/close.j2 new file mode 100644 index 000000000000..fdc2dfd1843b --- /dev/null +++ b/tools/net/sunrpc/xdrgen/templates/C/union/encoder/close.j2 @@ -0,0 +1,4 @@ +{# SPDX-License-Identifier: GPL-2.0 #} + } + return true; +}; diff --git a/tools/net/sunrpc/xdrgen/templates/C/union/encoder/default_spec.j2 b/tools/net/sunrpc/xdrgen/templates/C/union/encoder/default_spec.j2 new file mode 100644 index 000000000000..044a002d0589 --- /dev/null +++ b/tools/net/sunrpc/xdrgen/templates/C/union/encoder/default_spec.j2 @@ -0,0 +1,2 @@ +{# SPDX-License-Identifier: GPL-2.0 #} + default: diff --git a/tools/net/sunrpc/xdrgen/templates/C/union/encoder/open.j2 b/tools/net/sunrpc/xdrgen/templates/C/union/encoder/open.j2 new file mode 100644 index 000000000000..e5a206df10c6 --- /dev/null +++ b/tools/net/sunrpc/xdrgen/templates/C/union/encoder/open.j2 @@ -0,0 +1,12 @@ +{# SPDX-License-Identifier: GPL-2.0 #} + +{% if annotate %} +/* union {{ name }} */ +{% endif %} +{% if name in public_apis %} +bool +{% else %} +static bool __maybe_unused +{% endif %} +xdrgen_encode_{{ name }}(struct xdr_stream *xdr, const struct {{ name }} *ptr) +{ diff --git a/tools/net/sunrpc/xdrgen/templates/C/union/encoder/switch_spec.j2 b/tools/net/sunrpc/xdrgen/templates/C/union/encoder/switch_spec.j2 new file mode 100644 index 000000000000..c8c3ecbe038b --- /dev/null +++ b/tools/net/sunrpc/xdrgen/templates/C/union/encoder/switch_spec.j2 @@ -0,0 +1,7 @@ +{# SPDX-License-Identifier: GPL-2.0 #} +{% if annotate %} + /* discriminant {{ name }} */ +{% endif %} + if (!xdrgen_encode_{{ type }}(xdr, ptr->{{ name }})) + return false; + switch (ptr->{{ name }}) { diff --git a/tools/net/sunrpc/xdrgen/templates/C/union/encoder/void.j2 b/tools/net/sunrpc/xdrgen/templates/C/union/encoder/void.j2 new file mode 100644 index 000000000000..84e7c2127d75 --- /dev/null +++ b/tools/net/sunrpc/xdrgen/templates/C/union/encoder/void.j2 @@ -0,0 +1,3 @@ +{# SPDX-License-Identifier: GPL-2.0 #} + if (!xdrgen_encode_void(xdr)) + return false; diff --git a/tools/net/sunrpc/xdrgen/tests/test.x b/tools/net/sunrpc/xdrgen/tests/test.x new file mode 100644 index 000000000000..90c8587f6fe5 --- /dev/null +++ b/tools/net/sunrpc/xdrgen/tests/test.x @@ -0,0 +1,36 @@ +/* Sample XDR specification from RFC 1832 Section 5.5 */ + +const MAXUSERNAME = 32; /* max length of a user name */ +const MAXFILELEN = 65535; /* max length of a file */ +const MAXNAMELEN = 255; /* max length of a file name */ + +/* + * Types of files: + */ +enum filekind { + TEXT = 0, /* ascii data */ + DATA = 1, /* raw data */ + EXEC = 2 /* executable */ +}; + +/* + * File information, per kind of file: + */ +union filetype switch (filekind kind) { +case TEXT: + void; /* no extra information */ +case DATA: + string creator; /* data creator */ +case EXEC: + string interpretor; /* program interpretor */ +}; + +/* + * A complete file: + */ +struct file { + string filename; /* name of file */ + filetype type; /* info about file */ + string owner; /* owner of file */ + opaque data; /* file data */ +}; diff --git a/tools/net/sunrpc/xdrgen/xdr_ast.py b/tools/net/sunrpc/xdrgen/xdr_ast.py new file mode 100644 index 000000000000..dbd3fcf9c957 --- /dev/null +++ b/tools/net/sunrpc/xdrgen/xdr_ast.py @@ -0,0 +1,510 @@ +#!/usr/bin/env python3 +# ex: set filetype=python: + +"""Define and implement the Abstract Syntax Tree for the XDR language.""" + +import sys +from typing import List +from dataclasses import dataclass + +from lark import ast_utils, Transformer +from lark.tree import Meta + +this_module = sys.modules[__name__] + +excluded_apis = [] +header_name = "none" +public_apis = [] +enums = set() +structs = set() +pass_by_reference = set() + + +@dataclass +class _XdrAst(ast_utils.Ast): + """Base class for the XDR abstract syntax tree""" + + +@dataclass +class _XdrIdentifier(_XdrAst): + """Corresponds to 'identifier' in the XDR language grammar""" + + symbol: str + + +@dataclass +class _XdrValue(_XdrAst): + """Corresponds to 'value' in the XDR language grammar""" + + value: str + + +@dataclass +class _XdrConstantValue(_XdrAst): + """Corresponds to 'constant' in the XDR language grammar""" + + value: int + + +@dataclass +class _XdrTypeSpecifier(_XdrAst): + """Corresponds to 'type_specifier' in the XDR language grammar""" + + type_name: str + c_classifier: str + + +@dataclass +class _XdrDefinedType(_XdrTypeSpecifier): + """Corresponds to a type defined by the input specification""" + + +@dataclass +class _XdrBuiltInType(_XdrTypeSpecifier): + """Corresponds to a built-in XDR type""" + + +@dataclass +class _XdrDeclaration(_XdrAst): + """Base class of XDR type declarations""" + + +@dataclass +class _XdrFixedLengthOpaque(_XdrDeclaration): + """A fixed-length opaque declaration""" + + name: str + size: str + template: str = "fixed_length_opaque" + + +@dataclass +class _XdrVariableLengthOpaque(_XdrDeclaration): + """A variable-length opaque declaration""" + + name: str + maxsize: str + template: str = "variable_length_opaque" + + +@dataclass +class _XdrVariableLengthString(_XdrDeclaration): + """A (NUL-terminated) variable-length string declaration""" + + name: str + maxsize: str + template: str = "variable_length_string" + + +@dataclass +class _XdrFixedLengthArray(_XdrDeclaration): + """A fixed-length array declaration""" + + name: str + spec: _XdrTypeSpecifier + size: str + template: str = "fixed_length_array" + + +@dataclass +class _XdrVariableLengthArray(_XdrDeclaration): + """A variable-length array declaration""" + + name: str + spec: _XdrTypeSpecifier + maxsize: str + template: str = "variable_length_array" + + +@dataclass +class _XdrOptionalData(_XdrDeclaration): + """An 'optional_data' declaration""" + + name: str + spec: _XdrTypeSpecifier + template: str = "optional_data" + + +@dataclass +class _XdrBasic(_XdrDeclaration): + """A 'basic' declaration""" + + name: str + spec: _XdrTypeSpecifier + template: str = "basic" + + +@dataclass +class _XdrVoid(_XdrDeclaration): + """A void declaration""" + + template: str = "void" + + +@dataclass +class _XdrConstant(_XdrAst): + """Corresponds to 'constant_def' in the grammar""" + + name: str + value: str + + +@dataclass +class _XdrEnumerator(_XdrAst): + """An 'identifier = value' enumerator""" + + name: str + value: str + + +@dataclass +class _XdrEnum(_XdrAst): + """An XDR enum definition""" + + name: str + minimum: int + maximum: int + enumerators: List[_XdrEnumerator] + + +@dataclass +class _XdrStruct(_XdrAst): + """An XDR struct definition""" + + name: str + fields: List[_XdrDeclaration] + + +@dataclass +class _XdrPointer(_XdrAst): + """An XDR pointer definition""" + + name: str + fields: List[_XdrDeclaration] + + +@dataclass +class _XdrTypedef(_XdrAst): + """An XDR typedef""" + + declaration: _XdrDeclaration + + +@dataclass +class _XdrCaseSpec(_XdrAst): + """One case in an XDR union""" + + values: List[str] + arm: _XdrDeclaration + template: str = "case_spec" + + +@dataclass +class _XdrDefaultSpec(_XdrAst): + """Default case in an XDR union""" + + arm: _XdrDeclaration + template: str = "default_spec" + + +@dataclass +class _XdrUnion(_XdrAst): + """An XDR union""" + + name: str + discriminant: _XdrDeclaration + cases: List[_XdrCaseSpec] + default: _XdrDeclaration + + +@dataclass +class _RpcProcedure(_XdrAst): + """RPC procedure definition""" + + name: str + number: str + argument: _XdrTypeSpecifier + result: _XdrTypeSpecifier + + +@dataclass +class _RpcVersion(_XdrAst): + """RPC version definition""" + + name: str + number: str + procedures: List[_RpcProcedure] + + +@dataclass +class _RpcProgram(_XdrAst): + """RPC program definition""" + + name: str + number: str + versions: List[_RpcVersion] + + +@dataclass +class _Pragma(_XdrAst): + """Empty class for pragma directives""" + + +@dataclass +class Definition(_XdrAst, ast_utils.WithMeta): + """Corresponds to 'definition' in the grammar""" + + meta: Meta + value: _XdrAst + + +@dataclass +class Specification(_XdrAst, ast_utils.AsList): + """Corresponds to 'specification' in the grammar""" + + definitions: List[Definition] + + +class ParseToAst(Transformer): + """Functions that transform productions into AST nodes""" + + def identifier(self, children): + """Instantiate one _XdrIdentifier object""" + return _XdrIdentifier(children[0].value) + + def value(self, children): + """Instantiate one _XdrValue object""" + if isinstance(children[0], _XdrIdentifier): + return _XdrValue(children[0].symbol) + return _XdrValue(children[0].children[0].value) + + def constant(self, children): + """Instantiate one _XdrConstantValue object""" + match children[0].data: + case "decimal_constant": + value = int(children[0].children[0].value, base=10) + case "hexadecimal_constant": + value = int(children[0].children[0].value, base=16) + case "octal_constant": + value = int(children[0].children[0].value, base=8) + return _XdrConstantValue(value) + + def type_specifier(self, children): + """Instantiate one type_specifier object""" + c_classifier = "" + if isinstance(children[0], _XdrIdentifier): + name = children[0].symbol + if name in enums: + c_classifier = "enum " + if name in structs: + c_classifier = "struct " + return _XdrDefinedType( + type_name=name, + c_classifier=c_classifier, + ) + + token = children[0].data + return _XdrBuiltInType( + type_name=token.value, + c_classifier=c_classifier, + ) + + def constant_def(self, children): + """Instantiate one _XdrConstant object""" + name = children[0].symbol + value = children[1].value + return _XdrConstant(name, value) + + # cel: Python can compute a min() and max() for the enumerator values + # so that the generated code can perform proper range checking. + def enum(self, children): + """Instantiate one _XdrEnum object""" + enum_name = children[0].symbol + enums.add(enum_name) + + i = 0 + enumerators = [] + body = children[1] + while i < len(body.children): + name = body.children[i].symbol + value = body.children[i + 1].value + enumerators.append(_XdrEnumerator(name, value)) + i = i + 2 + + return _XdrEnum(enum_name, 0, 0, enumerators) + + def fixed_length_opaque(self, children): + """Instantiate one _XdrFixedLengthOpaque declaration object""" + name = children[0].symbol + size = children[1].value + + return _XdrFixedLengthOpaque(name, size) + + def variable_length_opaque(self, children): + """Instantiate one _XdrVariableLengthOpaque declaration object""" + name = children[0].symbol + if children[1] is not None: + maxsize = children[1].value + else: + maxsize = "0" + + return _XdrVariableLengthOpaque(name, maxsize) + + def variable_length_string(self, children): + """Instantiate one _XdrVariableLengthString declaration object""" + name = children[0].symbol + if children[1] is not None: + maxsize = children[1].value + else: + maxsize = "0" + + return _XdrVariableLengthString(name, maxsize) + + def fixed_length_array(self, children): + """Instantiate one _XdrFixedLengthArray declaration object""" + spec = children[0] + name = children[1].symbol + size = children[2].value + + return _XdrFixedLengthArray(name, spec, size) + + def variable_length_array(self, children): + """Instantiate one _XdrVariableLengthArray declaration object""" + spec = children[0] + name = children[1].symbol + if children[2] is not None: + maxsize = children[2].value + else: + maxsize = "0" + + return _XdrVariableLengthArray(name, spec, maxsize) + + def optional_data(self, children): + """Instantiate one _XdrOptionalData declaration object""" + spec = children[0] + name = children[1].symbol + structs.add(name) + pass_by_reference.add(name) + + return _XdrOptionalData(name, spec) + + def basic(self, children): + """Instantiate one _XdrBasic object""" + spec = children[0] + name = children[1].symbol + + return _XdrBasic(name, spec) + + def void(self, children): + """Instantiate one _XdrVoid declaration object""" + + return _XdrVoid() + + def struct(self, children): + """Instantiate one _XdrStruct object""" + name = children[0].symbol + structs.add(name) + pass_by_reference.add(name) + fields = children[1].children + + last_field = fields[-1] + if ( + isinstance(last_field, _XdrOptionalData) + and name == last_field.spec.type_name + ): + return _XdrPointer(name, fields) + + return _XdrStruct(name, fields) + + def typedef(self, children): + """Instantiate one _XdrTypedef object""" + new_type = children[0] + if isinstance(new_type, _XdrBasic) and isinstance( + new_type.spec, _XdrDefinedType + ): + if new_type.spec.type_name in pass_by_reference: + pass_by_reference.add(new_type.name) + + return _XdrTypedef(new_type) + + def case_spec(self, children): + """Instantiate one _XdrCaseSpec object""" + values = [] + for item in children[0:-1]: + values.append(item.value) + arm = children[-1] + + return _XdrCaseSpec(values, arm) + + def default_spec(self, children): + """Instantiate one _XdrDefaultSpec object""" + arm = children[0] + + return _XdrDefaultSpec(arm) + + def union(self, children): + """Instantiate one _XdrUnion object""" + name = children[0].symbol + structs.add(name) + pass_by_reference.add(name) + + body = children[1] + discriminant = body.children[0].children[0] + cases = body.children[1:-1] + default = body.children[-1] + + return _XdrUnion(name, discriminant, cases, default) + + def procedure_def(self, children): + """Instantiate one _RpcProcedure object""" + result = children[0] + name = children[1].symbol + argument = children[2] + number = children[3].value + + return _RpcProcedure(name, number, argument, result) + + def version_def(self, children): + """Instantiate one _RpcVersion object""" + name = children[0].symbol + number = children[-1].value + procedures = children[1:-1] + + return _RpcVersion(name, number, procedures) + + def program_def(self, children): + """Instantiate one _RpcProgram object""" + name = children[0].symbol + number = children[-1].value + versions = children[1:-1] + + return _RpcProgram(name, number, versions) + + def pragma_def(self, children): + """Instantiate one _Pragma object""" + directive = children[0].children[0].data + match directive: + case "exclude_directive": + excluded_apis.append(children[1].symbol) + case "header_directive": + global header_name + header_name = children[1].symbol + case "public_directive": + public_apis.append(children[1].symbol) + case _: + raise NotImplementedError("Directive not supported") + return _Pragma() + + +transformer = ast_utils.create_transformer(this_module, ParseToAst()) + + +def transform_parse_tree(parse_tree): + """Transform productions into an abstract syntax tree""" + + return transformer.transform(parse_tree) + + +def get_header_name() -> str: + """Return header name set by pragma header directive""" + return header_name diff --git a/tools/net/sunrpc/xdrgen/xdr_parse.py b/tools/net/sunrpc/xdrgen/xdr_parse.py new file mode 100644 index 000000000000..964b44e675df --- /dev/null +++ b/tools/net/sunrpc/xdrgen/xdr_parse.py @@ -0,0 +1,36 @@ +#!/usr/bin/env python3 +# ex: set filetype=python: + +"""Common parsing code for xdrgen""" + +from lark import Lark + + +# Set to True to emit annotation comments in generated source +annotate = False + + +def set_xdr_annotate(set_it: bool) -> None: + """Set 'annotate' if --annotate was specified on the command line""" + global annotate + annotate = set_it + + +def get_xdr_annotate() -> bool: + """Return True if --annotate was specified on the command line""" + return annotate + + +def xdr_parser() -> Lark: + """Return a Lark parser instance configured with the XDR language grammar""" + + return Lark.open( + "grammars/xdr.lark", + rel_to=__file__, + start="specification", + debug=True, + strict=True, + propagate_positions=True, + parser="lalr", + lexer="contextual", + ) diff --git a/tools/net/sunrpc/xdrgen/xdrgen b/tools/net/sunrpc/xdrgen/xdrgen new file mode 100755 index 000000000000..95f303b2861b --- /dev/null +++ b/tools/net/sunrpc/xdrgen/xdrgen @@ -0,0 +1,132 @@ +#!/usr/bin/env python3 +# ex: set filetype=python: + +"""Translate an XDR specification into executable code that +can be compiled for the Linux kernel.""" + +__author__ = "Chuck Lever" +__copyright__ = "Copyright (c) 2024 Oracle and/or its affiliates." +__license__ = "GPL-2.0 only" +__version__ = "0.2" + +import sys +import argparse + +from subcmds import definitions +from subcmds import declarations +from subcmds import lint +from subcmds import source + + +sys.path.insert(1, "@pythondir@") + + +def main() -> int: + """Parse command-line options""" + parser = argparse.ArgumentParser( + formatter_class=argparse.RawDescriptionHelpFormatter, + description="Convert an XDR specification to Linux kernel source code", + epilog="""\ +Copyright (c) 2024 Oracle and/or its affiliates. + +License GPLv2: +This is free software. You are free to change and redistribute it. +There is NO WARRANTY, to the extent permitted by law.""", + ) + parser.add_argument( + "--version", + help="Display the version of this tool", + action="version", + version=__version__, + ) + + subcommands = parser.add_subparsers(title="Subcommands", required=True) + + definitions_parser = subcommands.add_parser( + "definitions", help="Generate XDR definitions" + ) + definitions_parser.add_argument( + "--annotate", + action="store_true", + default=False, + help="Add annotation comments", + ) + definitions_parser.add_argument( + "--language", + action="store_true", + default="C", + help="Output language", + ) + definitions_parser.add_argument( + "--peer", + choices=["server", "client",], + default="server", + help="Generate header code for client or server side", + type=str, + ) + definitions_parser.add_argument("filename", help="File containing an XDR specification") + definitions_parser.set_defaults(func=definitions.subcmd) + + declarations_parser = subcommands.add_parser( + "declarations", help="Generate function declarations" + ) + declarations_parser.add_argument( + "--annotate", + action="store_true", + default=False, + help="Add annotation comments", + ) + declarations_parser.add_argument( + "--language", + action="store_true", + default="C", + help="Output language", + ) + declarations_parser.add_argument( + "--peer", + choices=["server", "client",], + default="server", + help="Generate code for client or server side", + type=str, + ) + declarations_parser.add_argument("filename", help="File containing an XDR specification") + declarations_parser.set_defaults(func=declarations.subcmd) + + linter_parser = subcommands.add_parser("lint", help="Check an XDR specification") + linter_parser.add_argument("filename", help="File containing an XDR specification") + linter_parser.set_defaults(func=lint.subcmd) + + source_parser = subcommands.add_parser( + "source", help="Generate XDR encoder and decoder source code" + ) + source_parser.add_argument( + "--annotate", + action="store_true", + default=False, + help="Add annotation comments", + ) + source_parser.add_argument( + "--language", + action="store_true", + default="C", + help="Output language", + ) + source_parser.add_argument( + "--peer", + choices=["server", "client",], + default="server", + help="Generate code for client or server side", + type=str, + ) + source_parser.add_argument("filename", help="File containing an XDR specification") + source_parser.set_defaults(func=source.subcmd) + + args = parser.parse_args() + return args.func(args) + + +try: + if __name__ == "__main__": + sys.exit(main()) +except (SystemExit, KeyboardInterrupt, BrokenPipeError): + sys.exit(1) diff --git a/tools/pci/Makefile b/tools/pci/Makefile index 57744778b518..62d41f1a1e2c 100644 --- a/tools/pci/Makefile +++ b/tools/pci/Makefile @@ -42,7 +42,7 @@ $(OUTPUT)pcitest: $(PCITEST_IN) clean: rm -f $(ALL_PROGRAMS) rm -rf $(OUTPUT)include/ - find $(or $(OUTPUT),.) -name '*.o' -delete -o -name '\.*.d' -delete + find $(or $(OUTPUT),.) -name '*.o' -delete -o -name '\.*.cmd' -delete -o -name '\.*.d' -delete install: $(ALL_PROGRAMS) install -d -m 755 $(DESTDIR)$(bindir); \ diff --git a/tools/pci/pcitest.c b/tools/pci/pcitest.c index 441b54234635..470258009ddc 100644 --- a/tools/pci/pcitest.c +++ b/tools/pci/pcitest.c @@ -16,8 +16,6 @@ #include -#define BILLION 1E9 - static char *result[] = { "NOT OKAY", "OKAY" }; static char *irq[] = { "LEGACY", "MSI", "MSI-X" };