From c4525b513de399109fc51529fcf7fec04f907be1 Mon Sep 17 00:00:00 2001 From: Choong Yong Liang Date: Thu, 27 Feb 2025 20:15:17 +0800 Subject: [PATCH 01/66] net: phylink: use pl->link_interface in phylink_expects_phy() [ Upstream commit b63263555eaafbf9ab1a82f2020bbee872d83759 ] The phylink_expects_phy() function allows MAC drivers to check if they are expecting a PHY to attach. The checking condition in phylink_expects_phy() aims to achieve the same result as the checking condition in phylink_attach_phy(). However, the checking condition in phylink_expects_phy() uses pl->link_config.interface, while phylink_attach_phy() uses pl->link_interface. Initially, both pl->link_interface and pl->link_config.interface are set to SGMII, and pl->cfg_link_an_mode is set to MLO_AN_INBAND. When the interface switches from SGMII to 2500BASE-X, pl->link_config.interface is updated by phylink_major_config(). At this point, pl->cfg_link_an_mode remains MLO_AN_INBAND, and pl->link_config.interface is set to 2500BASE-X. Subsequently, when the STMMAC interface is taken down administratively and brought back up, it is blocked by phylink_expects_phy(). Since phylink_expects_phy() and phylink_attach_phy() aim to achieve the same result, phylink_expects_phy() should check pl->link_interface, which never changes, instead of pl->link_config.interface, which is updated by phylink_major_config(). Reviewed-by: Russell King (Oracle) Signed-off-by: Choong Yong Liang Link: https://patch.msgid.link/20250227121522.1802832-2-yong.liang.choong@linux.intel.com Signed-off-by: Jakub Kicinski Signed-off-by: Sasha Levin --- drivers/net/phy/phylink.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/phy/phylink.c b/drivers/net/phy/phylink.c index 3e9957b6aa14..b78dfcbec936 100644 --- a/drivers/net/phy/phylink.c +++ b/drivers/net/phy/phylink.c @@ -1811,7 +1811,7 @@ bool phylink_expects_phy(struct phylink *pl) { if (pl->cfg_link_an_mode == MLO_AN_FIXED || (pl->cfg_link_an_mode == MLO_AN_INBAND && - phy_interface_mode_is_8023z(pl->link_config.interface))) + phy_interface_mode_is_8023z(pl->link_interface))) return false; return true; } From 95080412e9304cb5eeadd8a1f54aae9465330c54 Mon Sep 17 00:00:00 2001 From: Ming Lei Date: Wed, 5 Mar 2025 12:31:20 +0800 Subject: [PATCH 02/66] blk-throttle: don't take carryover for prioritized processing of metadata [ Upstream commit a9fc8868b350cbf4ff730a4ea9651319cc669516 ] Commit 29390bb5661d ("blk-throttle: support prioritized processing of metadata") takes bytes/ios carryover for prioritized processing of metadata. Turns out we can support it by charging it directly without trimming slice, and the result is same with carryover. Cc: Tejun Heo Cc: Josef Bacik Cc: Yu Kuai Signed-off-by: Ming Lei Acked-by: Tejun Heo Link: https://lore.kernel.org/r/20250305043123.3938491-3-ming.lei@redhat.com Signed-off-by: Jens Axboe Signed-off-by: Sasha Levin --- block/blk-throttle.c | 15 +++++---------- 1 file changed, 5 insertions(+), 10 deletions(-) diff --git a/block/blk-throttle.c b/block/blk-throttle.c index 2c4192e12efa..6b82fcbd7e77 100644 --- a/block/blk-throttle.c +++ b/block/blk-throttle.c @@ -1593,13 +1593,6 @@ static bool tg_within_limit(struct throtl_grp *tg, struct bio *bio, bool rw) return tg_may_dispatch(tg, bio, NULL); } -static void tg_dispatch_in_debt(struct throtl_grp *tg, struct bio *bio, bool rw) -{ - if (!bio_flagged(bio, BIO_BPS_THROTTLED)) - tg->carryover_bytes[rw] -= throtl_bio_data_size(bio); - tg->carryover_ios[rw]--; -} - bool __blk_throtl_bio(struct bio *bio) { struct request_queue *q = bdev_get_queue(bio->bi_bdev); @@ -1636,10 +1629,12 @@ bool __blk_throtl_bio(struct bio *bio) /* * IOs which may cause priority inversions are * dispatched directly, even if they're over limit. - * Debts are handled by carryover_bytes/ios while - * calculating wait time. + * + * Charge and dispatch directly, and our throttle + * control algorithm is adaptive, and extra IO bytes + * will be throttled for paying the debt */ - tg_dispatch_in_debt(tg, bio, rw); + throtl_charge_bio(tg, bio); } else { /* if above limits, break to queue */ break; From 69bb5b3ae348040e385113efba5bdc76396644d0 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Matti=20Lehtim=C3=A4ki?= Date: Thu, 6 Feb 2025 20:56:48 +0100 Subject: [PATCH 03/66] remoteproc: qcom_wcnss: Handle platforms with only single power domain MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit [ Upstream commit 65991ea8a6d1e68effdc01d95ebe39f1653f7b71 ] Both MSM8974 and MSM8226 have only CX as power domain with MX & PX being handled as regulators. Handle this case by reodering pd_names to have CX first, and handling that the driver core will already attach a single power domain internally. Signed-off-by: Matti Lehtimäki [luca: minor changes] Signed-off-by: Luca Weiss Link: https://lore.kernel.org/r/20250206-wcnss-singlepd-v2-2-9a53ee953dee@lucaweiss.eu [bjorn: Added missing braces to else after multi-statement if] Signed-off-by: Bjorn Andersson Signed-off-by: Sasha Levin --- drivers/remoteproc/qcom_wcnss.c | 33 ++++++++++++++++++++++++++------- 1 file changed, 26 insertions(+), 7 deletions(-) diff --git a/drivers/remoteproc/qcom_wcnss.c b/drivers/remoteproc/qcom_wcnss.c index a7bb9da27029..735d373a9f63 100644 --- a/drivers/remoteproc/qcom_wcnss.c +++ b/drivers/remoteproc/qcom_wcnss.c @@ -117,10 +117,10 @@ static const struct wcnss_data pronto_v1_data = { .pmu_offset = 0x1004, .spare_offset = 0x1088, - .pd_names = { "mx", "cx" }, + .pd_names = { "cx", "mx" }, .vregs = (struct wcnss_vreg_info[]) { - { "vddmx", 950000, 1150000, 0 }, { "vddcx", .super_turbo = true}, + { "vddmx", 950000, 1150000, 0 }, { "vddpx", 1800000, 1800000, 0 }, }, .num_pd_vregs = 2, @@ -131,10 +131,10 @@ static const struct wcnss_data pronto_v2_data = { .pmu_offset = 0x1004, .spare_offset = 0x1088, - .pd_names = { "mx", "cx" }, + .pd_names = { "cx", "mx" }, .vregs = (struct wcnss_vreg_info[]) { - { "vddmx", 1287500, 1287500, 0 }, { "vddcx", .super_turbo = true }, + { "vddmx", 1287500, 1287500, 0 }, { "vddpx", 1800000, 1800000, 0 }, }, .num_pd_vregs = 2, @@ -397,8 +397,17 @@ static irqreturn_t wcnss_stop_ack_interrupt(int irq, void *dev) static int wcnss_init_pds(struct qcom_wcnss *wcnss, const char * const pd_names[WCNSS_MAX_PDS]) { + struct device *dev = wcnss->dev; int i, ret; + /* Handle single power domain */ + if (dev->pm_domain) { + wcnss->pds[0] = dev; + wcnss->num_pds = 1; + pm_runtime_enable(dev); + return 0; + } + for (i = 0; i < WCNSS_MAX_PDS; i++) { if (!pd_names[i]) break; @@ -418,8 +427,15 @@ static int wcnss_init_pds(struct qcom_wcnss *wcnss, static void wcnss_release_pds(struct qcom_wcnss *wcnss) { + struct device *dev = wcnss->dev; int i; + /* Handle single power domain */ + if (wcnss->num_pds == 1 && dev->pm_domain) { + pm_runtime_disable(dev); + return; + } + for (i = 0; i < wcnss->num_pds; i++) dev_pm_domain_detach(wcnss->pds[i], false); } @@ -437,10 +453,13 @@ static int wcnss_init_regulators(struct qcom_wcnss *wcnss, * the regulators for the power domains. For old device trees we need to * reserve extra space to manage them through the regulator interface. */ - if (wcnss->num_pds) - info += num_pd_vregs; - else + if (wcnss->num_pds) { + info += wcnss->num_pds; + /* Handle single power domain case */ + num_vregs += num_pd_vregs - wcnss->num_pds; + } else { num_vregs += num_pd_vregs; + } bulk = devm_kcalloc(wcnss->dev, num_vregs, sizeof(struct regulator_bulk_data), From a23f3910123b518c8cf0fbb89aa9c9e08d23a4ca Mon Sep 17 00:00:00 2001 From: Victor Lu Date: Thu, 13 Feb 2025 18:38:28 -0500 Subject: [PATCH 04/66] drm/amdgpu: Do not program AGP BAR regs under SRIOV in gfxhub_v1_0.c [ Upstream commit 057fef20b8401110a7bc1c2fe9d804a8a0bf0d24 ] SRIOV VF does not have write access to AGP BAR regs. Skip the writes to avoid a dmesg warning. Signed-off-by: Victor Lu Acked-by: Alex Deucher Signed-off-by: Alex Deucher Signed-off-by: Sasha Levin --- drivers/gpu/drm/amd/amdgpu/gfxhub_v1_0.c | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) diff --git a/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_0.c b/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_0.c index 0e3ddea7b8e0..a7bfc9f41d0e 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_0.c @@ -92,12 +92,12 @@ static void gfxhub_v1_0_init_system_aperture_regs(struct amdgpu_device *adev) { uint64_t value; - /* Program the AGP BAR */ - WREG32_SOC15_RLC(GC, 0, mmMC_VM_AGP_BASE, 0); - WREG32_SOC15_RLC(GC, 0, mmMC_VM_AGP_BOT, adev->gmc.agp_start >> 24); - WREG32_SOC15_RLC(GC, 0, mmMC_VM_AGP_TOP, adev->gmc.agp_end >> 24); - if (!amdgpu_sriov_vf(adev) || adev->asic_type <= CHIP_VEGA10) { + /* Program the AGP BAR */ + WREG32_SOC15_RLC(GC, 0, mmMC_VM_AGP_BASE, 0); + WREG32_SOC15_RLC(GC, 0, mmMC_VM_AGP_BOT, adev->gmc.agp_start >> 24); + WREG32_SOC15_RLC(GC, 0, mmMC_VM_AGP_TOP, adev->gmc.agp_end >> 24); + /* Program the system aperture low logical page number. */ WREG32_SOC15_RLC(GC, 0, mmMC_VM_SYSTEM_APERTURE_LOW_ADDR, min(adev->gmc.fb_start, adev->gmc.agp_start) >> 18); From 5bf0fd2bd5efba63f278b5b6fca7a1ddead451be Mon Sep 17 00:00:00 2001 From: Nicholas Kazlauskas Date: Wed, 19 Feb 2025 09:56:53 -0500 Subject: [PATCH 05/66] drm/amd/display: Ensure DMCUB idle before reset on DCN31/DCN35 [ Upstream commit c707ea82c79dbd1d295ec94cc6529a5248c77757 ] [Why] If we soft reset before halt finishes and there are outstanding memory transactions then the memory interface may produce unexpected results, such as out of order transactions when the firmware next runs. These can manifest as random or unexpected load/store violations. [How] Increase the timeout before soft reset to ensure the DMCUB has quiesced. This is effectively 1s maximum based on experimentation. Use the enable bit check on DCN31 like we're doing on DCN35 and reorder the reset writes to follow the HW programming guide. Ensure we're reading SCRATCH7 instead of SCRATCH8 for the HALT code. No current versions of DMCUB firmware use the SCRATCH8 boot bit to dynamically switch where the HALT code goes to maintain backwards compatibility with PSP. Reviewed-by: Dillon Varone Signed-off-by: Nicholas Kazlauskas Signed-off-by: Wayne Lin Tested-by: Daniel Wheeler Signed-off-by: Alex Deucher Signed-off-by: Sasha Levin --- .../gpu/drm/amd/display/dmub/src/dmub_dcn31.c | 17 +++++++++++------ .../gpu/drm/amd/display/dmub/src/dmub_dcn35.c | 4 ++-- 2 files changed, 13 insertions(+), 8 deletions(-) diff --git a/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn31.c b/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn31.c index d9f31b191c69..1a68b5782cac 100644 --- a/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn31.c +++ b/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn31.c @@ -83,8 +83,8 @@ static inline void dmub_dcn31_translate_addr(const union dmub_addr *addr_in, void dmub_dcn31_reset(struct dmub_srv *dmub) { union dmub_gpint_data_register cmd; - const uint32_t timeout = 100; - uint32_t in_reset, scratch, i, pwait_mode; + const uint32_t timeout = 100000; + uint32_t in_reset, is_enabled, scratch, i, pwait_mode; REG_GET(DMCUB_CNTL2, DMCUB_SOFT_RESET, &in_reset); @@ -108,7 +108,7 @@ void dmub_dcn31_reset(struct dmub_srv *dmub) } for (i = 0; i < timeout; ++i) { - scratch = dmub->hw_funcs.get_gpint_response(dmub); + scratch = REG_READ(DMCUB_SCRATCH7); if (scratch == DMUB_GPINT__STOP_FW_RESPONSE) break; @@ -125,9 +125,14 @@ void dmub_dcn31_reset(struct dmub_srv *dmub) /* Force reset in case we timed out, DMCUB is likely hung. */ } - REG_UPDATE(DMCUB_CNTL2, DMCUB_SOFT_RESET, 1); - REG_UPDATE(DMCUB_CNTL, DMCUB_ENABLE, 0); - REG_UPDATE(MMHUBBUB_SOFT_RESET, DMUIF_SOFT_RESET, 1); + REG_GET(DMCUB_CNTL, DMCUB_ENABLE, &is_enabled); + + if (is_enabled) { + REG_UPDATE(DMCUB_CNTL2, DMCUB_SOFT_RESET, 1); + REG_UPDATE(MMHUBBUB_SOFT_RESET, DMUIF_SOFT_RESET, 1); + REG_UPDATE(DMCUB_CNTL, DMCUB_ENABLE, 0); + } + REG_WRITE(DMCUB_INBOX1_RPTR, 0); REG_WRITE(DMCUB_INBOX1_WPTR, 0); REG_WRITE(DMCUB_OUTBOX1_RPTR, 0); diff --git a/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn35.c b/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn35.c index 2ccad79053c5..4581eb479451 100644 --- a/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn35.c +++ b/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn35.c @@ -88,7 +88,7 @@ static inline void dmub_dcn35_translate_addr(const union dmub_addr *addr_in, void dmub_dcn35_reset(struct dmub_srv *dmub) { union dmub_gpint_data_register cmd; - const uint32_t timeout = 100; + const uint32_t timeout = 100000; uint32_t in_reset, is_enabled, scratch, i, pwait_mode; REG_GET(DMCUB_CNTL2, DMCUB_SOFT_RESET, &in_reset); @@ -113,7 +113,7 @@ void dmub_dcn35_reset(struct dmub_srv *dmub) } for (i = 0; i < timeout; ++i) { - scratch = dmub->hw_funcs.get_gpint_response(dmub); + scratch = REG_READ(DMCUB_SCRATCH7); if (scratch == DMUB_GPINT__STOP_FW_RESPONSE) break; From 682c4226f317ec1954755eb4e8e58dd4d3a5cef1 Mon Sep 17 00:00:00 2001 From: George Shen Date: Fri, 14 Feb 2025 22:00:13 -0500 Subject: [PATCH 06/66] drm/amd/display: Skip checking FRL_MODE bit for PCON BW determination [ Upstream commit 0584bbcf0c53c133081100e4f4c9fe41e598d045 ] [Why/How] Certain PCON will clear the FRL_MODE bit despite supporting the link BW indicated in the other bits. Thus, skip checking the FRL_MODE bit when interpreting the hdmi_encoded_link_bw struct. Reviewed-by: Wenjing Liu Signed-off-by: George Shen Signed-off-by: Wayne Lin Tested-by: Daniel Wheeler Signed-off-by: Alex Deucher Signed-off-by: Sasha Levin --- .../dc/link/protocols/link_dp_capability.c | 30 +++++++++---------- 1 file changed, 15 insertions(+), 15 deletions(-) diff --git a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_capability.c b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_capability.c index 885e749cdc6e..23e650e39910 100644 --- a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_capability.c +++ b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_capability.c @@ -250,21 +250,21 @@ static uint32_t intersect_frl_link_bw_support( { uint32_t supported_bw_in_kbps = max_supported_frl_bw_in_kbps; - // HDMI_ENCODED_LINK_BW bits are only valid if HDMI Link Configuration bit is 1 (FRL mode) - if (hdmi_encoded_link_bw.bits.FRL_MODE) { - if (hdmi_encoded_link_bw.bits.BW_48Gbps) - supported_bw_in_kbps = 48000000; - else if (hdmi_encoded_link_bw.bits.BW_40Gbps) - supported_bw_in_kbps = 40000000; - else if (hdmi_encoded_link_bw.bits.BW_32Gbps) - supported_bw_in_kbps = 32000000; - else if (hdmi_encoded_link_bw.bits.BW_24Gbps) - supported_bw_in_kbps = 24000000; - else if (hdmi_encoded_link_bw.bits.BW_18Gbps) - supported_bw_in_kbps = 18000000; - else if (hdmi_encoded_link_bw.bits.BW_9Gbps) - supported_bw_in_kbps = 9000000; - } + /* Skip checking FRL_MODE bit, as certain PCON will clear + * it despite supporting the link BW indicated in the other bits. + */ + if (hdmi_encoded_link_bw.bits.BW_48Gbps) + supported_bw_in_kbps = 48000000; + else if (hdmi_encoded_link_bw.bits.BW_40Gbps) + supported_bw_in_kbps = 40000000; + else if (hdmi_encoded_link_bw.bits.BW_32Gbps) + supported_bw_in_kbps = 32000000; + else if (hdmi_encoded_link_bw.bits.BW_24Gbps) + supported_bw_in_kbps = 24000000; + else if (hdmi_encoded_link_bw.bits.BW_18Gbps) + supported_bw_in_kbps = 18000000; + else if (hdmi_encoded_link_bw.bits.BW_9Gbps) + supported_bw_in_kbps = 9000000; return supported_bw_in_kbps; } From e015cef8b78f6020bb43f898aa4e543a166d3aff Mon Sep 17 00:00:00 2001 From: Dillon Varone Date: Thu, 13 Feb 2025 13:10:41 -0500 Subject: [PATCH 07/66] drm/amd/display: Fix DMUB reset sequence for DCN401 [ Upstream commit 0dfcc2bf269010a6e093793034c048049a40ee93 ] [WHY] It should no longer use DMCUB_SOFT_RESET as it can result in the memory request path becoming desynchronized. [HOW] To ensure robustness in the reset sequence: 1) Extend timeout on the "halt" command sent via gpint, and check for controller to enter "wait" as a stronger guarantee that there are no requests to memory still in flight. 2) Remove usage of DMCUB_SOFT_RESET 3) Rely on PSP to reset the controller safely Reviewed-by: Nicholas Kazlauskas Signed-off-by: Dillon Varone Signed-off-by: Wayne Lin Tested-by: Daniel Wheeler Signed-off-by: Alex Deucher Signed-off-by: Sasha Levin --- .../drm/amd/display/dmub/src/dmub_dcn401.c | 47 ++++++++++++------- .../drm/amd/display/dmub/src/dmub_dcn401.h | 3 +- 2 files changed, 32 insertions(+), 18 deletions(-) diff --git a/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn401.c b/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn401.c index 39a8cb6d7523..e1c4fe1c6e3e 100644 --- a/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn401.c +++ b/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn401.c @@ -63,8 +63,10 @@ static inline void dmub_dcn401_translate_addr(const union dmub_addr *addr_in, void dmub_dcn401_reset(struct dmub_srv *dmub) { union dmub_gpint_data_register cmd; - const uint32_t timeout = 30; - uint32_t in_reset, scratch, i; + const uint32_t timeout_us = 1 * 1000 * 1000; //1s + const uint32_t poll_delay_us = 1; //1us + uint32_t i = 0; + uint32_t in_reset, scratch, pwait_mode; REG_GET(DMCUB_CNTL2, DMCUB_SOFT_RESET, &in_reset); @@ -75,32 +77,35 @@ void dmub_dcn401_reset(struct dmub_srv *dmub) dmub->hw_funcs.set_gpint(dmub, cmd); - /** - * Timeout covers both the ACK and the wait - * for remaining work to finish. - * - * This is mostly bound by the PHY disable sequence. - * Each register check will be greater than 1us, so - * don't bother using udelay. - */ - - for (i = 0; i < timeout; ++i) { + for (i = 0; i < timeout_us; i++) { if (dmub->hw_funcs.is_gpint_acked(dmub, cmd)) break; + + udelay(poll_delay_us); } - for (i = 0; i < timeout; ++i) { + for (; i < timeout_us; i++) { scratch = dmub->hw_funcs.get_gpint_response(dmub); if (scratch == DMUB_GPINT__STOP_FW_RESPONSE) break; + + udelay(poll_delay_us); } - /* Force reset in case we timed out, DMCUB is likely hung. */ + for (; i < timeout_us; i++) { + REG_GET(DMCUB_CNTL, DMCUB_PWAIT_MODE_STATUS, &pwait_mode); + if (pwait_mode & (1 << 0)) + break; + + udelay(poll_delay_us); + } + } + + if (i >= timeout_us) { + /* timeout should never occur */ + BREAK_TO_DEBUGGER(); } - REG_UPDATE(DMCUB_CNTL2, DMCUB_SOFT_RESET, 1); - REG_UPDATE(DMCUB_CNTL, DMCUB_ENABLE, 0); - REG_UPDATE(MMHUBBUB_SOFT_RESET, DMUIF_SOFT_RESET, 1); REG_WRITE(DMCUB_INBOX1_RPTR, 0); REG_WRITE(DMCUB_INBOX1_WPTR, 0); REG_WRITE(DMCUB_OUTBOX1_RPTR, 0); @@ -131,7 +136,10 @@ void dmub_dcn401_backdoor_load(struct dmub_srv *dmub, dmub_dcn401_get_fb_base_offset(dmub, &fb_base, &fb_offset); + /* reset and disable DMCUB and MMHUBBUB DMUIF */ REG_UPDATE(DMCUB_SEC_CNTL, DMCUB_SEC_RESET, 1); + REG_UPDATE(MMHUBBUB_SOFT_RESET, DMUIF_SOFT_RESET, 1); + REG_UPDATE(DMCUB_CNTL, DMCUB_ENABLE, 0); dmub_dcn401_translate_addr(&cw0->offset, fb_base, fb_offset, &offset); @@ -151,6 +159,7 @@ void dmub_dcn401_backdoor_load(struct dmub_srv *dmub, DMCUB_REGION3_CW1_TOP_ADDRESS, cw1->region.top, DMCUB_REGION3_CW1_ENABLE, 1); + /* release DMCUB reset only to prevent premature execution */ REG_UPDATE_2(DMCUB_SEC_CNTL, DMCUB_SEC_RESET, 0, DMCUB_MEM_UNIT_ID, 0x20); } @@ -161,7 +170,10 @@ void dmub_dcn401_backdoor_load_zfb_mode(struct dmub_srv *dmub, { union dmub_addr offset; + /* reset and disable DMCUB and MMHUBBUB DMUIF */ REG_UPDATE(DMCUB_SEC_CNTL, DMCUB_SEC_RESET, 1); + REG_UPDATE(MMHUBBUB_SOFT_RESET, DMUIF_SOFT_RESET, 1); + REG_UPDATE(DMCUB_CNTL, DMCUB_ENABLE, 0); offset = cw0->offset; @@ -181,6 +193,7 @@ void dmub_dcn401_backdoor_load_zfb_mode(struct dmub_srv *dmub, DMCUB_REGION3_CW1_TOP_ADDRESS, cw1->region.top, DMCUB_REGION3_CW1_ENABLE, 1); + /* release DMCUB reset only to prevent premature execution */ REG_UPDATE_2(DMCUB_SEC_CNTL, DMCUB_SEC_RESET, 0, DMCUB_MEM_UNIT_ID, 0x20); } diff --git a/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn401.h b/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn401.h index 4c8843b79695..31f95b27e227 100644 --- a/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn401.h +++ b/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn401.h @@ -169,7 +169,8 @@ struct dmub_srv; DMUB_SF(HOST_INTERRUPT_CSR, HOST_REG_INBOX0_RSP_INT_EN) \ DMUB_SF(HOST_INTERRUPT_CSR, HOST_REG_OUTBOX0_RDY_INT_ACK) \ DMUB_SF(HOST_INTERRUPT_CSR, HOST_REG_OUTBOX0_RDY_INT_STAT) \ - DMUB_SF(HOST_INTERRUPT_CSR, HOST_REG_OUTBOX0_RDY_INT_EN) + DMUB_SF(HOST_INTERRUPT_CSR, HOST_REG_OUTBOX0_RDY_INT_EN) \ + DMUB_SF(DMCUB_CNTL, DMCUB_PWAIT_MODE_STATUS) struct dmub_srv_dcn401_reg_offset { #define DMUB_SR(reg) uint32_t reg; From 1f2b3ea00717f867d8a7e263fc94fd2b9b4f22f1 Mon Sep 17 00:00:00 2001 From: Dillon Varone Date: Wed, 12 Feb 2025 17:06:42 -0500 Subject: [PATCH 08/66] drm/amd/display: Fix p-state type when p-state is unsupported [ Upstream commit a025f424af0407b7561bd5e6217295dde3abbc2e ] [WHY&HOW] P-state type would remain on previously used when unsupported which causes confusion in logging and visual confirm, so set back to zero when unsupported. Reviewed-by: Aric Cyr Signed-off-by: Dillon Varone Signed-off-by: Wayne Lin Tested-by: Daniel Wheeler Signed-off-by: Alex Deucher Signed-off-by: Sasha Levin --- drivers/gpu/drm/amd/display/dc/core/dc_hw_sequencer.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_hw_sequencer.c b/drivers/gpu/drm/amd/display/dc/core/dc_hw_sequencer.c index fdcba960e1e6..d2342a91e7e7 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc_hw_sequencer.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc_hw_sequencer.c @@ -508,6 +508,7 @@ void set_p_state_switch_method( if (!dc->ctx || !dc->ctx->dmub_srv || !pipe_ctx || !vba) return; + pipe_ctx->p_state_type = P_STATE_UNKNOWN; if (vba->DRAMClockChangeSupport[vba->VoltageLevel][vba->maxMpcComb] != dm_dram_clock_change_unsupported) { /* MCLK switching is supported */ From cc714c89ef5a990bc2c3b96171980f06b196e0b7 Mon Sep 17 00:00:00 2001 From: Aric Cyr Date: Thu, 23 Jan 2025 16:39:52 -0500 Subject: [PATCH 09/66] drm/amd/display: Request HW cursor on DCN3.2 with SubVP [ Upstream commit b74f46f3ce1e5f6336645f1e9ff47c56d5dfdef1 ] [why] When SubVP is active the HW cursor size is limited to 64x64, and anything larger will force composition which is bad for gaming on DCN3.2 if the game uses a larger cursor. [how] If HW cursor is requested, typically by a fullscreen game, do not enable SubVP so that up to 256x256 cursor sizes are available for DCN3.2. Reviewed-by: Aric Cyr Signed-off-by: Aric Cyr Signed-off-by: Wayne Lin Tested-by: Daniel Wheeler Signed-off-by: Alex Deucher Signed-off-by: Sasha Levin --- drivers/gpu/drm/amd/display/dc/core/dc.c | 3 ++- drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c | 1 + 2 files changed, 3 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/amd/display/dc/core/dc.c b/drivers/gpu/drm/amd/display/dc/core/dc.c index dd4b131fac6c..fa7db10104c4 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc.c @@ -4744,7 +4744,8 @@ static bool full_update_required(struct dc *dc, stream_update->lut3d_func || stream_update->pending_test_pattern || stream_update->crtc_timing_adjust || - stream_update->scaler_sharpener_update)) + stream_update->scaler_sharpener_update || + stream_update->hw_cursor_req)) return true; if (stream) { diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c b/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c index 6f490d8d7038..56dda686e299 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c @@ -626,6 +626,7 @@ static bool dcn32_assign_subvp_pipe(struct dc *dc, * - Not TMZ surface */ if (pipe->plane_state && !pipe->top_pipe && !pipe->prev_odm_pipe && !dcn32_is_center_timing(pipe) && + !pipe->stream->hw_cursor_req && !(pipe->stream->timing.pix_clk_100hz / 10000 > DCN3_2_MAX_SUBVP_PIXEL_RATE_MHZ) && (!dcn32_is_psr_capable(pipe) || (context->stream_count == 1 && dc->caps.dmub_caps.subvp_psr)) && dc_state_get_pipe_subvp_type(context, pipe) == SUBVP_NONE && From f29c876d72d71b64b0cf0958345f1da0fbd9c383 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Wed, 5 Feb 2025 11:21:28 +0100 Subject: [PATCH 10/66] perf/core: Clean up perf_try_init_event() [ Upstream commit da02f54e81db2f7bf6af9d1d0cfc5b41ec6d0dcb ] Make sure that perf_try_init_event() doesn't leave event->pmu nor event->destroy set on failure. Signed-off-by: Peter Zijlstra (Intel) Signed-off-by: Ingo Molnar Acked-by: Ravi Bangoria Link: https://lore.kernel.org/r/20250205102449.110145835@infradead.org Signed-off-by: Sasha Levin --- kernel/events/core.c | 69 +++++++++++++++++++++++++------------------- 1 file changed, 40 insertions(+), 29 deletions(-) diff --git a/kernel/events/core.c b/kernel/events/core.c index edafe9fc4bdd..19dde12f23b8 100644 --- a/kernel/events/core.c +++ b/kernel/events/core.c @@ -11978,40 +11978,51 @@ static int perf_try_init_event(struct pmu *pmu, struct perf_event *event) if (ctx) perf_event_ctx_unlock(event->group_leader, ctx); - if (!ret) { - if (!(pmu->capabilities & PERF_PMU_CAP_EXTENDED_REGS) && - has_extended_regs(event)) - ret = -EOPNOTSUPP; + if (ret) + goto err_pmu; - if (pmu->capabilities & PERF_PMU_CAP_NO_EXCLUDE && - event_has_any_exclude_flag(event)) - ret = -EINVAL; - - if (pmu->scope != PERF_PMU_SCOPE_NONE && event->cpu >= 0) { - const struct cpumask *cpumask = perf_scope_cpu_topology_cpumask(pmu->scope, event->cpu); - struct cpumask *pmu_cpumask = perf_scope_cpumask(pmu->scope); - int cpu; - - if (pmu_cpumask && cpumask) { - cpu = cpumask_any_and(pmu_cpumask, cpumask); - if (cpu >= nr_cpu_ids) - ret = -ENODEV; - else - event->event_caps |= PERF_EV_CAP_READ_SCOPE; - } else { - ret = -ENODEV; - } - } - - if (ret && event->destroy) - event->destroy(event); + if (!(pmu->capabilities & PERF_PMU_CAP_EXTENDED_REGS) && + has_extended_regs(event)) { + ret = -EOPNOTSUPP; + goto err_destroy; } - if (ret) { - event->pmu = NULL; - module_put(pmu->module); + if (pmu->capabilities & PERF_PMU_CAP_NO_EXCLUDE && + event_has_any_exclude_flag(event)) { + ret = -EINVAL; + goto err_destroy; } + if (pmu->scope != PERF_PMU_SCOPE_NONE && event->cpu >= 0) { + const struct cpumask *cpumask; + struct cpumask *pmu_cpumask; + int cpu; + + cpumask = perf_scope_cpu_topology_cpumask(pmu->scope, event->cpu); + pmu_cpumask = perf_scope_cpumask(pmu->scope); + + ret = -ENODEV; + if (!pmu_cpumask || !cpumask) + goto err_destroy; + + cpu = cpumask_any_and(pmu_cpumask, cpumask); + if (cpu >= nr_cpu_ids) + goto err_destroy; + + event->event_caps |= PERF_EV_CAP_READ_SCOPE; + } + + return 0; + +err_destroy: + if (event->destroy) { + event->destroy(event); + event->destroy = NULL; + } + +err_pmu: + event->pmu = NULL; + module_put(pmu->module); return ret; } From 4731d5328f507ae8fd8a57abbca9119ec7a8d665 Mon Sep 17 00:00:00 2001 From: Hans Verkuil Date: Mon, 24 Feb 2025 14:13:24 +0100 Subject: [PATCH 11/66] media: cx231xx: set device_caps for 417 [ Upstream commit a79efc44b51432490538a55b9753a721f7d3ea42 ] The video_device for the MPEG encoder did not set device_caps. Add this, otherwise the video device can't be registered (you get a WARN_ON instead). Not seen before since currently 417 support is disabled, but I found this while experimenting with it. Signed-off-by: Hans Verkuil Signed-off-by: Mauro Carvalho Chehab Signed-off-by: Sasha Levin --- drivers/media/usb/cx231xx/cx231xx-417.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/media/usb/cx231xx/cx231xx-417.c b/drivers/media/usb/cx231xx/cx231xx-417.c index abb967c8bd35..6cb130eb32d5 100644 --- a/drivers/media/usb/cx231xx/cx231xx-417.c +++ b/drivers/media/usb/cx231xx/cx231xx-417.c @@ -1722,6 +1722,8 @@ static void cx231xx_video_dev_init( vfd->lock = &dev->lock; vfd->release = video_device_release_empty; vfd->ctrl_handler = &dev->mpeg_ctrl_handler.hdl; + vfd->device_caps = V4L2_CAP_READWRITE | V4L2_CAP_STREAMING | + V4L2_CAP_VIDEO_CAPTURE; video_set_drvdata(vfd, dev); if (dev->tuner_type == TUNER_ABSENT) { v4l2_disable_ioctl(vfd, VIDIOC_G_FREQUENCY); From 5aba8ac434d8bcffc673538da6772ec955a81b3f Mon Sep 17 00:00:00 2001 From: Artur Weber Date: Mon, 3 Mar 2025 21:54:47 +0100 Subject: [PATCH 12/66] pinctrl: bcm281xx: Use "unsigned int" instead of bare "unsigned" [ Upstream commit 07b5a2a13f4704c5eae3be7277ec54ffdba45f72 ] Replace uses of bare "unsigned" with "unsigned int" to fix checkpatch warnings. No functional change. Signed-off-by: Artur Weber Link: https://lore.kernel.org/20250303-bcm21664-pinctrl-v3-2-5f8b80e4ab51@gmail.com Signed-off-by: Linus Walleij Signed-off-by: Sasha Levin --- drivers/pinctrl/bcm/pinctrl-bcm281xx.c | 44 +++++++++++++------------- 1 file changed, 22 insertions(+), 22 deletions(-) diff --git a/drivers/pinctrl/bcm/pinctrl-bcm281xx.c b/drivers/pinctrl/bcm/pinctrl-bcm281xx.c index cf6efa9c0364..a039b490cdb8 100644 --- a/drivers/pinctrl/bcm/pinctrl-bcm281xx.c +++ b/drivers/pinctrl/bcm/pinctrl-bcm281xx.c @@ -72,7 +72,7 @@ static enum bcm281xx_pin_type hdmi_pin = BCM281XX_PIN_TYPE_HDMI; struct bcm281xx_pin_function { const char *name; const char * const *groups; - const unsigned ngroups; + const unsigned int ngroups; }; /* @@ -84,10 +84,10 @@ struct bcm281xx_pinctrl_data { /* List of all pins */ const struct pinctrl_pin_desc *pins; - const unsigned npins; + const unsigned int npins; const struct bcm281xx_pin_function *functions; - const unsigned nfunctions; + const unsigned int nfunctions; struct regmap *regmap; }; @@ -941,7 +941,7 @@ static struct bcm281xx_pinctrl_data bcm281xx_pinctrl = { }; static inline enum bcm281xx_pin_type pin_type_get(struct pinctrl_dev *pctldev, - unsigned pin) + unsigned int pin) { struct bcm281xx_pinctrl_data *pdata = pinctrl_dev_get_drvdata(pctldev); @@ -985,7 +985,7 @@ static int bcm281xx_pinctrl_get_groups_count(struct pinctrl_dev *pctldev) } static const char *bcm281xx_pinctrl_get_group_name(struct pinctrl_dev *pctldev, - unsigned group) + unsigned int group) { struct bcm281xx_pinctrl_data *pdata = pinctrl_dev_get_drvdata(pctldev); @@ -993,9 +993,9 @@ static const char *bcm281xx_pinctrl_get_group_name(struct pinctrl_dev *pctldev, } static int bcm281xx_pinctrl_get_group_pins(struct pinctrl_dev *pctldev, - unsigned group, + unsigned int group, const unsigned **pins, - unsigned *num_pins) + unsigned int *num_pins) { struct bcm281xx_pinctrl_data *pdata = pinctrl_dev_get_drvdata(pctldev); @@ -1007,7 +1007,7 @@ static int bcm281xx_pinctrl_get_group_pins(struct pinctrl_dev *pctldev, static void bcm281xx_pinctrl_pin_dbg_show(struct pinctrl_dev *pctldev, struct seq_file *s, - unsigned offset) + unsigned int offset) { seq_printf(s, " %s", dev_name(pctldev->dev)); } @@ -1029,7 +1029,7 @@ static int bcm281xx_pinctrl_get_fcns_count(struct pinctrl_dev *pctldev) } static const char *bcm281xx_pinctrl_get_fcn_name(struct pinctrl_dev *pctldev, - unsigned function) + unsigned int function) { struct bcm281xx_pinctrl_data *pdata = pinctrl_dev_get_drvdata(pctldev); @@ -1037,9 +1037,9 @@ static const char *bcm281xx_pinctrl_get_fcn_name(struct pinctrl_dev *pctldev, } static int bcm281xx_pinctrl_get_fcn_groups(struct pinctrl_dev *pctldev, - unsigned function, + unsigned int function, const char * const **groups, - unsigned * const num_groups) + unsigned int * const num_groups) { struct bcm281xx_pinctrl_data *pdata = pinctrl_dev_get_drvdata(pctldev); @@ -1050,8 +1050,8 @@ static int bcm281xx_pinctrl_get_fcn_groups(struct pinctrl_dev *pctldev, } static int bcm281xx_pinmux_set(struct pinctrl_dev *pctldev, - unsigned function, - unsigned group) + unsigned int function, + unsigned int group) { struct bcm281xx_pinctrl_data *pdata = pinctrl_dev_get_drvdata(pctldev); const struct bcm281xx_pin_function *f = &pdata->functions[function]; @@ -1082,7 +1082,7 @@ static const struct pinmux_ops bcm281xx_pinctrl_pinmux_ops = { }; static int bcm281xx_pinctrl_pin_config_get(struct pinctrl_dev *pctldev, - unsigned pin, + unsigned int pin, unsigned long *config) { return -ENOTSUPP; @@ -1091,9 +1091,9 @@ static int bcm281xx_pinctrl_pin_config_get(struct pinctrl_dev *pctldev, /* Goes through the configs and update register val/mask */ static int bcm281xx_std_pin_update(struct pinctrl_dev *pctldev, - unsigned pin, + unsigned int pin, unsigned long *configs, - unsigned num_configs, + unsigned int num_configs, u32 *val, u32 *mask) { @@ -1207,9 +1207,9 @@ static const u16 bcm281xx_pullup_map[] = { /* Goes through the configs and update register val/mask */ static int bcm281xx_i2c_pin_update(struct pinctrl_dev *pctldev, - unsigned pin, + unsigned int pin, unsigned long *configs, - unsigned num_configs, + unsigned int num_configs, u32 *val, u32 *mask) { @@ -1277,9 +1277,9 @@ static int bcm281xx_i2c_pin_update(struct pinctrl_dev *pctldev, /* Goes through the configs and update register val/mask */ static int bcm281xx_hdmi_pin_update(struct pinctrl_dev *pctldev, - unsigned pin, + unsigned int pin, unsigned long *configs, - unsigned num_configs, + unsigned int num_configs, u32 *val, u32 *mask) { @@ -1321,9 +1321,9 @@ static int bcm281xx_hdmi_pin_update(struct pinctrl_dev *pctldev, } static int bcm281xx_pinctrl_pin_config_set(struct pinctrl_dev *pctldev, - unsigned pin, + unsigned int pin, unsigned long *configs, - unsigned num_configs) + unsigned int num_configs) { struct bcm281xx_pinctrl_data *pdata = pinctrl_dev_get_drvdata(pctldev); enum bcm281xx_pin_type pin_type; From 1a4a834f2af5b51c01ff2303389b9673caa8630e Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Thu, 12 Dec 2024 14:15:07 -0800 Subject: [PATCH 13/66] rcu: Fix get_state_synchronize_rcu_full() GP-start detection MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit [ Upstream commit 85aad7cc417877054c65bd490dc037b087ef21b4 ] The get_state_synchronize_rcu_full() and poll_state_synchronize_rcu_full() functions use the root rcu_node structure's ->gp_seq field to detect the beginnings and ends of grace periods, respectively. This choice is necessary for the poll_state_synchronize_rcu_full() function because (give or take counter wrap), the following sequence is guaranteed not to trigger: get_state_synchronize_rcu_full(&rgos); synchronize_rcu(); WARN_ON_ONCE(!poll_state_synchronize_rcu_full(&rgos)); The RCU callbacks that awaken synchronize_rcu() instances are guaranteed not to be invoked before the root rcu_node structure's ->gp_seq field is updated to indicate the end of the grace period. However, these callbacks might start being invoked immediately thereafter, in particular, before rcu_state.gp_seq has been updated. Therefore, poll_state_synchronize_rcu_full() must refer to the root rcu_node structure's ->gp_seq field. Because this field is updated under this structure's ->lock, any code following a call to poll_state_synchronize_rcu_full() will be fully ordered after the full grace-period computation, as is required by RCU's memory-ordering semantics. By symmetry, the get_state_synchronize_rcu_full() function should also use this same root rcu_node structure's ->gp_seq field. But it turns out that symmetry is profoundly (though extremely infrequently) destructive in this case. To see this, consider the following sequence of events: 1. CPU 0 starts a new grace period, and updates rcu_state.gp_seq accordingly. 2. As its first step of grace-period initialization, CPU 0 examines the current CPU hotplug state and decides that it need not wait for CPU 1, which is currently offline. 3. CPU 1 comes online, and updates its state. But this does not affect the current grace period, but rather the one after that. After all, CPU 1 was offline when the current grace period started, so all pre-existing RCU readers on CPU 1 must have completed or been preempted before it last went offline. The current grace period therefore has nothing it needs to wait for on CPU 1. 4. CPU 1 switches to an rcutorture kthread which is running rcutorture's rcu_torture_reader() function, which starts a new RCU reader. 5. CPU 2 is running rcutorture's rcu_torture_writer() function and collects a new polled grace-period "cookie" using get_state_synchronize_rcu_full(). Because the newly started grace period has not completed initialization, the root rcu_node structure's ->gp_seq field has not yet been updated to indicate that this new grace period has already started. This cookie is therefore set up for the end of the current grace period (rather than the end of the following grace period). 6. CPU 0 finishes grace-period initialization. 7. If CPU 1’s rcutorture reader is preempted, it will be added to the ->blkd_tasks list, but because CPU 1’s ->qsmask bit is not set in CPU 1's leaf rcu_node structure, the ->gp_tasks pointer will not be updated.  Thus, this grace period will not wait on it.  Which is only fair, given that the CPU did not come online until after the grace period officially started. 8. CPUs 0 and 2 then detect the new grace period and then report a quiescent state to the RCU core. 9. Because CPU 1 was offline at the start of the current grace period, CPUs 0 and 2 are the only CPUs that this grace period needs to wait on. So the grace period ends and post-grace-period cleanup starts. In particular, the root rcu_node structure's ->gp_seq field is updated to indicate that this grace period has now ended. 10. CPU 2 continues running rcu_torture_writer() and sees that, from the viewpoint of the root rcu_node structure consulted by the poll_state_synchronize_rcu_full() function, the grace period has ended.  It therefore updates state accordingly. 11. CPU 1 is still running the same RCU reader, which notices this update and thus complains about the too-short grace period. The fix is for the get_state_synchronize_rcu_full() function to use rcu_state.gp_seq instead of the root rcu_node structure's ->gp_seq field. With this change in place, if step 5's cookie indicates that the grace period has not yet started, then any prior code executed by CPU 2 must have happened before CPU 1 came online. This will in turn prevent CPU 1's code in steps 3 and 11 from spanning CPU 2's grace-period wait, thus preventing CPU 1 from being subjected to a too-short grace period. This commit therefore makes this change. Note that there is no change to the poll_state_synchronize_rcu_full() function, which as noted above, must continue to use the root rcu_node structure's ->gp_seq field. This is of course an asymmetry between these two functions, but is an asymmetry that is absolutely required for correct operation. It is a common human tendency to greatly value symmetry, and sometimes symmetry is a wonderful thing. Other times, symmetry results in poor performance. But in this case, symmetry is just plain wrong. Nevertheless, the asymmetry does require an additional adjustment. It is possible for get_state_synchronize_rcu_full() to see a given grace period as having started, but for an immediately following poll_state_synchronize_rcu_full() to see it as having not yet started. Given the current rcu_seq_done_exact() implementation, this will result in a false-positive indication that the grace period is done from poll_state_synchronize_rcu_full(). This is dealt with by making rcu_seq_done_exact() reach back three grace periods rather than just two of them. However, simply changing get_state_synchronize_rcu_full() function to use rcu_state.gp_seq instead of the root rcu_node structure's ->gp_seq field results in a theoretical bug in kernels booted with rcutree.rcu_normal_wake_from_gp=1 due to the following sequence of events: o The rcu_gp_init() function invokes rcu_seq_start() to officially start a new grace period. o A new RCU reader begins, referencing X from some RCU-protected list. The new grace period is not obligated to wait for this reader. o An updater removes X, then calls synchronize_rcu(), which queues a wait element. o The grace period ends, awakening the updater, which frees X while the reader is still referencing it. The reason that this is theoretical is that although the grace period has officially started, none of the CPUs are officially aware of this, and thus will have to assume that the RCU reader pre-dated the start of the grace period. Detailed explanation can be found at [2] and [3]. Except for kernels built with CONFIG_PROVE_RCU=y, which use the polled grace-period APIs, which can and do complain bitterly when this sequence of events occurs. Not only that, there might be some future RCU grace-period mechanism that pulls this sequence of events from theory into practice. This commit therefore also pulls the call to rcu_sr_normal_gp_init() to precede that to rcu_seq_start(). Although this fixes commit 91a967fd6934 ("rcu: Add full-sized polling for get_completed*() and poll_state*()"), it is not clear that it is worth backporting this commit. First, it took me many weeks to convince rcutorture to reproduce this more frequently than once per year. Second, this cannot be reproduced at all without frequent CPU-hotplug operations, as in waiting all of 50 milliseconds from the end of the previous operation until starting the next one. Third, the TREE03.boot settings cause multi-millisecond delays during RCU grace-period initialization, which greatly increase the probability of the above sequence of events. (Don't do this in production workloads!) Fourth, the TREE03 rcutorture scenario was modified to use four-CPU guest OSes, to have a single-rcu_node combining tree, no testing of RCU priority boosting, and no random preemption, and these modifications were necessary to reproduce this issue in a reasonable timeframe. Fifth, extremely heavy use of get_state_synchronize_rcu_full() and/or poll_state_synchronize_rcu_full() is required to reproduce this, and as of v6.12, only kfree_rcu() uses it, and even then not particularly heavily. [boqun: Apply the fix [1], and add the comment before the moved rcu_sr_normal_gp_init(). Additional links are added for explanation.] Signed-off-by: Paul E. McKenney Reviewed-by: Frederic Weisbecker Reviewed-by: Joel Fernandes (Google) Tested-by: Uladzislau Rezki (Sony) Link: https://lore.kernel.org/rcu/d90bd6d9-d15c-4b9b-8a69-95336e74e8f4@paulmck-laptop/ [1] Link: https://lore.kernel.org/rcu/20250303001507.GA3994772@joelnvbox/ [2] Link: https://lore.kernel.org/rcu/Z8bcUsZ9IpRi1QoP@pc636/ [3] Reviewed-by: Joel Fernandes Signed-off-by: Boqun Feng Signed-off-by: Sasha Levin --- kernel/rcu/rcu.h | 2 +- kernel/rcu/tree.c | 15 +++++++++++---- 2 files changed, 12 insertions(+), 5 deletions(-) diff --git a/kernel/rcu/rcu.h b/kernel/rcu/rcu.h index feb3ac1dc5d5..f87c9d6d36fc 100644 --- a/kernel/rcu/rcu.h +++ b/kernel/rcu/rcu.h @@ -162,7 +162,7 @@ static inline bool rcu_seq_done_exact(unsigned long *sp, unsigned long s) { unsigned long cur_s = READ_ONCE(*sp); - return ULONG_CMP_GE(cur_s, s) || ULONG_CMP_LT(cur_s, s - (2 * RCU_SEQ_STATE_MASK + 1)); + return ULONG_CMP_GE(cur_s, s) || ULONG_CMP_LT(cur_s, s - (3 * RCU_SEQ_STATE_MASK + 1)); } /* diff --git a/kernel/rcu/tree.c b/kernel/rcu/tree.c index 8e52c1dd0628..4ed863219521 100644 --- a/kernel/rcu/tree.c +++ b/kernel/rcu/tree.c @@ -1822,10 +1822,14 @@ static noinline_for_stack bool rcu_gp_init(void) /* Advance to a new grace period and initialize state. */ record_gp_stall_check_time(); + /* + * A new wait segment must be started before gp_seq advanced, so + * that previous gp waiters won't observe the new gp_seq. + */ + start_new_poll = rcu_sr_normal_gp_init(); /* Record GP times before starting GP, hence rcu_seq_start(). */ rcu_seq_start(&rcu_state.gp_seq); ASSERT_EXCLUSIVE_WRITER(rcu_state.gp_seq); - start_new_poll = rcu_sr_normal_gp_init(); trace_rcu_grace_period(rcu_state.name, rcu_state.gp_seq, TPS("start")); rcu_poll_gp_seq_start(&rcu_state.gp_seq_polled_snap); raw_spin_unlock_irq_rcu_node(rnp); @@ -4183,14 +4187,17 @@ EXPORT_SYMBOL_GPL(get_state_synchronize_rcu); */ void get_state_synchronize_rcu_full(struct rcu_gp_oldstate *rgosp) { - struct rcu_node *rnp = rcu_get_root(); - /* * Any prior manipulation of RCU-protected data must happen * before the loads from ->gp_seq and ->expedited_sequence. */ smp_mb(); /* ^^^ */ - rgosp->rgos_norm = rcu_seq_snap(&rnp->gp_seq); + + // Yes, rcu_state.gp_seq, not rnp_root->gp_seq, the latter's use + // in poll_state_synchronize_rcu_full() notwithstanding. Use of + // the latter here would result in too-short grace periods due to + // interactions with newly onlined CPUs. + rgosp->rgos_norm = rcu_seq_snap(&rcu_state.gp_seq); rgosp->rgos_exp = rcu_seq_snap(&rcu_state.expedited_sequence); } EXPORT_SYMBOL_GPL(get_state_synchronize_rcu_full); From 94df9fd015acef4556fe16ce06092a8d597c8897 Mon Sep 17 00:00:00 2001 From: Alexander Sverdlin Date: Mon, 3 Mar 2025 08:46:57 +0100 Subject: [PATCH 14/66] net: ethernet: ti: cpsw_new: populate netdev of_node [ Upstream commit 7ff1c88fc89688c27f773ba956f65f0c11367269 ] So that of_find_net_device_by_node() can find CPSW ports and other DSA switches can be stacked downstream. Tested in conjunction with KSZ8873. Reviewed-by: Siddharth Vadapalli Reviewed-by: Andrew Lunn Signed-off-by: Alexander Sverdlin Link: https://patch.msgid.link/20250303074703.1758297-1-alexander.sverdlin@siemens.com Signed-off-by: Jakub Kicinski Signed-off-by: Sasha Levin --- drivers/net/ethernet/ti/cpsw_new.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/net/ethernet/ti/cpsw_new.c b/drivers/net/ethernet/ti/cpsw_new.c index 557cc71b9dd2..0eee1a0527b5 100644 --- a/drivers/net/ethernet/ti/cpsw_new.c +++ b/drivers/net/ethernet/ti/cpsw_new.c @@ -1417,6 +1417,7 @@ static int cpsw_create_ports(struct cpsw_common *cpsw) ndev->netdev_ops = &cpsw_netdev_ops; ndev->ethtool_ops = &cpsw_ethtool_ops; SET_NETDEV_DEV(ndev, dev); + ndev->dev.of_node = slave_data->slave_node; if (!napi_ndev) { /* CPSW Host port CPDMA interface is shared between From ebaed867bfd1ef1b56523b940cc1a437f74ef857 Mon Sep 17 00:00:00 2001 From: Andrei Botila Date: Fri, 28 Feb 2025 17:43:19 +0200 Subject: [PATCH 15/66] net: phy: nxp-c45-tja11xx: add match_phy_device to TJA1103/TJA1104 [ Upstream commit a06a868a0cd96bc51401cdea897313a3f6ad01a0 ] Add .match_phy_device for the existing TJAs to differentiate between TJA1103 and TJA1104. TJA1103 and TJA1104 share the same PHY_ID but TJA1104 has MACsec capabilities while TJA1103 doesn't. Signed-off-by: Andrei Botila Reviewed-by: Andrew Lunn Link: https://patch.msgid.link/20250228154320.2979000-2-andrei.botila@oss.nxp.com Signed-off-by: Jakub Kicinski Signed-off-by: Sasha Levin --- drivers/net/phy/nxp-c45-tja11xx.c | 54 +++++++++++++++++++++++++++++-- 1 file changed, 52 insertions(+), 2 deletions(-) diff --git a/drivers/net/phy/nxp-c45-tja11xx.c b/drivers/net/phy/nxp-c45-tja11xx.c index 9788b820c6be..99a5eee77bec 100644 --- a/drivers/net/phy/nxp-c45-tja11xx.c +++ b/drivers/net/phy/nxp-c45-tja11xx.c @@ -1,6 +1,6 @@ // SPDX-License-Identifier: GPL-2.0 /* NXP C45 PHY driver - * Copyright 2021-2023 NXP + * Copyright 2021-2025 NXP * Author: Radu Pirea */ @@ -18,6 +18,8 @@ #include "nxp-c45-tja11xx.h" +#define PHY_ID_MASK GENMASK(31, 4) +/* Same id: TJA1103, TJA1104 */ #define PHY_ID_TJA_1103 0x001BB010 #define PHY_ID_TJA_1120 0x001BB031 @@ -1930,6 +1932,30 @@ static void tja1120_nmi_handler(struct phy_device *phydev, } } +static int nxp_c45_macsec_ability(struct phy_device *phydev) +{ + bool macsec_ability; + int phy_abilities; + + phy_abilities = phy_read_mmd(phydev, MDIO_MMD_VEND1, + VEND1_PORT_ABILITIES); + macsec_ability = !!(phy_abilities & MACSEC_ABILITY); + + return macsec_ability; +} + +static int tja1103_match_phy_device(struct phy_device *phydev) +{ + return phy_id_compare(phydev->phy_id, PHY_ID_TJA_1103, PHY_ID_MASK) && + !nxp_c45_macsec_ability(phydev); +} + +static int tja1104_match_phy_device(struct phy_device *phydev) +{ + return phy_id_compare(phydev->phy_id, PHY_ID_TJA_1103, PHY_ID_MASK) && + nxp_c45_macsec_ability(phydev); +} + static const struct nxp_c45_regmap tja1120_regmap = { .vend1_ptp_clk_period = 0x1020, .vend1_event_msg_filt = 0x9010, @@ -2000,7 +2026,6 @@ static const struct nxp_c45_phy_data tja1120_phy_data = { static struct phy_driver nxp_c45_driver[] = { { - PHY_ID_MATCH_MODEL(PHY_ID_TJA_1103), .name = "NXP C45 TJA1103", .get_features = nxp_c45_get_features, .driver_data = &tja1103_phy_data, @@ -2022,6 +2047,31 @@ static struct phy_driver nxp_c45_driver[] = { .get_sqi = nxp_c45_get_sqi, .get_sqi_max = nxp_c45_get_sqi_max, .remove = nxp_c45_remove, + .match_phy_device = tja1103_match_phy_device, + }, + { + .name = "NXP C45 TJA1104", + .get_features = nxp_c45_get_features, + .driver_data = &tja1103_phy_data, + .probe = nxp_c45_probe, + .soft_reset = nxp_c45_soft_reset, + .config_aneg = genphy_c45_config_aneg, + .config_init = nxp_c45_config_init, + .config_intr = tja1103_config_intr, + .handle_interrupt = nxp_c45_handle_interrupt, + .read_status = genphy_c45_read_status, + .suspend = genphy_c45_pma_suspend, + .resume = genphy_c45_pma_resume, + .get_sset_count = nxp_c45_get_sset_count, + .get_strings = nxp_c45_get_strings, + .get_stats = nxp_c45_get_stats, + .cable_test_start = nxp_c45_cable_test_start, + .cable_test_get_status = nxp_c45_cable_test_get_status, + .set_loopback = genphy_c45_loopback, + .get_sqi = nxp_c45_get_sqi, + .get_sqi_max = nxp_c45_get_sqi_max, + .remove = nxp_c45_remove, + .match_phy_device = tja1104_match_phy_device, }, { PHY_ID_MATCH_MODEL(PHY_ID_TJA_1120), From c4f025a58eef14983b424b95f7bafdc5602b79a2 Mon Sep 17 00:00:00 2001 From: Jiasheng Jiang Date: Fri, 28 Feb 2025 15:02:10 +0000 Subject: [PATCH 16/66] dpll: Add an assertion to check freq_supported_num [ Upstream commit 39e912a959c19338855b768eaaee2917d7841f71 ] Since the driver is broken in the case that src->freq_supported is not NULL but src->freq_supported_num is 0, add an assertion for it. Signed-off-by: Jiasheng Jiang Reviewed-by: Jiri Pirko Reviewed-by: Vadim Fedorenko Reviewed-by: Arkadiusz Kubalewski Link: https://patch.msgid.link/20250228150210.34404-1-jiashengjiangcool@gmail.com Signed-off-by: Jakub Kicinski Signed-off-by: Sasha Levin --- drivers/dpll/dpll_core.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/drivers/dpll/dpll_core.c b/drivers/dpll/dpll_core.c index 1877201d1aa9..20bdc52f63a5 100644 --- a/drivers/dpll/dpll_core.c +++ b/drivers/dpll/dpll_core.c @@ -443,8 +443,11 @@ static void dpll_pin_prop_free(struct dpll_pin_properties *prop) static int dpll_pin_prop_dup(const struct dpll_pin_properties *src, struct dpll_pin_properties *dst) { + if (WARN_ON(src->freq_supported && !src->freq_supported_num)) + return -EINVAL; + memcpy(dst, src, sizeof(*dst)); - if (src->freq_supported && src->freq_supported_num) { + if (src->freq_supported) { size_t freq_size = src->freq_supported_num * sizeof(*src->freq_supported); dst->freq_supported = kmemdup(src->freq_supported, From af73c8fd7388df67be2bd721510f4f3853101fb4 Mon Sep 17 00:00:00 2001 From: Uday Shankar Date: Fri, 28 Feb 2025 21:31:48 -0700 Subject: [PATCH 17/66] ublk: enforce ublks_max only for unprivileged devices [ Upstream commit 80bdfbb3545b6f16680a72c825063d08a6b44c7a ] Commit 403ebc877832 ("ublk_drv: add module parameter of ublks_max for limiting max allowed ublk dev"), claimed ublks_max was added to prevent a DoS situation with an untrusted user creating too many ublk devices. If that's the case, ublks_max should only restrict the number of unprivileged ublk devices in the system. Enforce the limit only for unprivileged ublk devices, and rename variables accordingly. Leave the external-facing parameter name unchanged, since changing it may break systems which use it (but still update its documentation to reflect its new meaning). As a result of this change, in a system where there are only normal (non-unprivileged) devices, the maximum number of such devices is increased to 1 << MINORBITS, or 1048576. That ought to be enough for anyone, right? Signed-off-by: Uday Shankar Reviewed-by: Ming Lei Link: https://lore.kernel.org/r/20250228-ublks_max-v1-1-04b7379190c0@purestorage.com Signed-off-by: Jens Axboe Signed-off-by: Sasha Levin --- drivers/block/ublk_drv.c | 42 ++++++++++++++++++++++++++-------------- 1 file changed, 27 insertions(+), 15 deletions(-) diff --git a/drivers/block/ublk_drv.c b/drivers/block/ublk_drv.c index 38b9e485e520..5ec5d580ef50 100644 --- a/drivers/block/ublk_drv.c +++ b/drivers/block/ublk_drv.c @@ -484,15 +484,17 @@ static wait_queue_head_t ublk_idr_wq; /* wait until one idr is freed */ static DEFINE_MUTEX(ublk_ctl_mutex); + +#define UBLK_MAX_UBLKS UBLK_MINORS + /* - * Max ublk devices allowed to add + * Max unprivileged ublk devices allowed to add * * It can be extended to one per-user limit in future or even controlled * by cgroup. */ -#define UBLK_MAX_UBLKS UBLK_MINORS -static unsigned int ublks_max = 64; -static unsigned int ublks_added; /* protected by ublk_ctl_mutex */ +static unsigned int unprivileged_ublks_max = 64; +static unsigned int unprivileged_ublks_added; /* protected by ublk_ctl_mutex */ static struct miscdevice ublk_misc; @@ -2203,7 +2205,8 @@ static int ublk_add_chdev(struct ublk_device *ub) if (ret) goto fail; - ublks_added++; + if (ub->dev_info.flags & UBLK_F_UNPRIVILEGED_DEV) + unprivileged_ublks_added++; return 0; fail: put_device(dev); @@ -2241,12 +2244,17 @@ static int ublk_add_tag_set(struct ublk_device *ub) static void ublk_remove(struct ublk_device *ub) { + bool unprivileged; + ublk_stop_dev(ub); cancel_work_sync(&ub->stop_work); cancel_work_sync(&ub->quiesce_work); cdev_device_del(&ub->cdev, &ub->cdev_dev); + unprivileged = ub->dev_info.flags & UBLK_F_UNPRIVILEGED_DEV; ublk_put_device(ub); - ublks_added--; + + if (unprivileged) + unprivileged_ublks_added--; } static struct ublk_device *ublk_get_device_from_id(int idx) @@ -2495,7 +2503,8 @@ static int ublk_ctrl_add_dev(struct io_uring_cmd *cmd) return ret; ret = -EACCES; - if (ublks_added >= ublks_max) + if ((info.flags & UBLK_F_UNPRIVILEGED_DEV) && + unprivileged_ublks_added >= unprivileged_ublks_max) goto out_unlock; ret = -ENOMEM; @@ -3123,23 +3132,26 @@ static void __exit ublk_exit(void) module_init(ublk_init); module_exit(ublk_exit); -static int ublk_set_max_ublks(const char *buf, const struct kernel_param *kp) +static int ublk_set_max_unprivileged_ublks(const char *buf, + const struct kernel_param *kp) { return param_set_uint_minmax(buf, kp, 0, UBLK_MAX_UBLKS); } -static int ublk_get_max_ublks(char *buf, const struct kernel_param *kp) +static int ublk_get_max_unprivileged_ublks(char *buf, + const struct kernel_param *kp) { - return sysfs_emit(buf, "%u\n", ublks_max); + return sysfs_emit(buf, "%u\n", unprivileged_ublks_max); } -static const struct kernel_param_ops ublk_max_ublks_ops = { - .set = ublk_set_max_ublks, - .get = ublk_get_max_ublks, +static const struct kernel_param_ops ublk_max_unprivileged_ublks_ops = { + .set = ublk_set_max_unprivileged_ublks, + .get = ublk_get_max_unprivileged_ublks, }; -module_param_cb(ublks_max, &ublk_max_ublks_ops, &ublks_max, 0644); -MODULE_PARM_DESC(ublks_max, "max number of ublk devices allowed to add(default: 64)"); +module_param_cb(ublks_max, &ublk_max_unprivileged_ublks_ops, + &unprivileged_ublks_max, 0644); +MODULE_PARM_DESC(ublks_max, "max number of unprivileged ublk devices allowed to add(default: 64)"); MODULE_AUTHOR("Ming Lei "); MODULE_DESCRIPTION("Userspace block device"); From 4176d6f2a67b81a2e261bff3833de73aa84862c6 Mon Sep 17 00:00:00 2001 From: Yi Liu Date: Wed, 26 Feb 2025 02:40:12 -0800 Subject: [PATCH 18/66] iommufd: Disallow allocating nested parent domain with fault ID [ Upstream commit 1062d81086156e42878d701b816d2f368b53a77c ] Allocating a domain with a fault ID indicates that the domain is faultable. However, there is a gap for the nested parent domain to support PRI. Some hardware lacks the capability to distinguish whether PRI occurs at stage 1 or stage 2. This limitation may require software-based page table walking to resolve. Since no in-tree IOMMU driver currently supports this functionality, it is disallowed. For more details, refer to the related discussion at [1]. [1] https://lore.kernel.org/linux-iommu/bd1655c6-8b2f-4cfa-adb1-badc00d01811@intel.com/ Link: https://patch.msgid.link/r/20250226104012.82079-1-yi.l.liu@intel.com Suggested-by: Lu Baolu Signed-off-by: Yi Liu Reviewed-by: Kevin Tian Reviewed-by: Lu Baolu Signed-off-by: Jason Gunthorpe Signed-off-by: Sasha Levin --- drivers/iommu/iommufd/hw_pagetable.c | 3 +++ tools/testing/selftests/iommu/iommufd.c | 4 ++++ 2 files changed, 7 insertions(+) diff --git a/drivers/iommu/iommufd/hw_pagetable.c b/drivers/iommu/iommufd/hw_pagetable.c index d06bf6e6c19f..2454627a8b61 100644 --- a/drivers/iommu/iommufd/hw_pagetable.c +++ b/drivers/iommu/iommufd/hw_pagetable.c @@ -122,6 +122,9 @@ iommufd_hwpt_paging_alloc(struct iommufd_ctx *ictx, struct iommufd_ioas *ioas, if ((flags & IOMMU_HWPT_ALLOC_DIRTY_TRACKING) && !device_iommu_capable(idev->dev, IOMMU_CAP_DIRTY_TRACKING)) return ERR_PTR(-EOPNOTSUPP); + if ((flags & IOMMU_HWPT_FAULT_ID_VALID) && + (flags & IOMMU_HWPT_ALLOC_NEST_PARENT)) + return ERR_PTR(-EOPNOTSUPP); hwpt_paging = __iommufd_object_alloc( ictx, hwpt_paging, IOMMUFD_OBJ_HWPT_PAGING, common.obj); diff --git a/tools/testing/selftests/iommu/iommufd.c b/tools/testing/selftests/iommu/iommufd.c index 4927b9add5ad..06f252733660 100644 --- a/tools/testing/selftests/iommu/iommufd.c +++ b/tools/testing/selftests/iommu/iommufd.c @@ -289,6 +289,10 @@ TEST_F(iommufd_ioas, alloc_hwpt_nested) &test_hwpt_id); test_err_hwpt_alloc(EINVAL, self->device_id, self->device_id, 0, &test_hwpt_id); + test_err_hwpt_alloc(EOPNOTSUPP, self->device_id, self->ioas_id, + IOMMU_HWPT_ALLOC_NEST_PARENT | + IOMMU_HWPT_FAULT_ID_VALID, + &test_hwpt_id); test_cmd_hwpt_alloc(self->device_id, self->ioas_id, IOMMU_HWPT_ALLOC_NEST_PARENT, From 8d7e13c31c52690655883dff604238b0760a3644 Mon Sep 17 00:00:00 2001 From: Paul Elder Date: Fri, 28 Feb 2025 18:17:31 +0900 Subject: [PATCH 19/66] media: imx335: Set vblank immediately [ Upstream commit c0aa40f45fef80b4182704d1bc089cbf8ae8bed0 ] When the vblank v4l2 control is set, it does not get written to the hardware unless exposure is also changed. Change the behavior such that the vblank is written immediately when the control is set, as setting the vblank without changing the exposure is a valid use case (such as for changing the frame rate). Signed-off-by: Paul Elder Reviewed-by: Kieran Bingham Signed-off-by: Sakari Ailus Signed-off-by: Hans Verkuil Signed-off-by: Sasha Levin --- drivers/media/i2c/imx335.c | 21 +++++++++++++++------ 1 file changed, 15 insertions(+), 6 deletions(-) diff --git a/drivers/media/i2c/imx335.c b/drivers/media/i2c/imx335.c index fcfd1d851bd4..0beb80b8c458 100644 --- a/drivers/media/i2c/imx335.c +++ b/drivers/media/i2c/imx335.c @@ -559,12 +559,14 @@ static int imx335_set_ctrl(struct v4l2_ctrl *ctrl) imx335->vblank, imx335->vblank + imx335->cur_mode->height); - return __v4l2_ctrl_modify_range(imx335->exp_ctrl, - IMX335_EXPOSURE_MIN, - imx335->vblank + - imx335->cur_mode->height - - IMX335_EXPOSURE_OFFSET, - 1, IMX335_EXPOSURE_DEFAULT); + ret = __v4l2_ctrl_modify_range(imx335->exp_ctrl, + IMX335_EXPOSURE_MIN, + imx335->vblank + + imx335->cur_mode->height - + IMX335_EXPOSURE_OFFSET, + 1, IMX335_EXPOSURE_DEFAULT); + if (ret) + return ret; } /* @@ -575,6 +577,13 @@ static int imx335_set_ctrl(struct v4l2_ctrl *ctrl) return 0; switch (ctrl->id) { + case V4L2_CID_VBLANK: + exposure = imx335->exp_ctrl->val; + analog_gain = imx335->again_ctrl->val; + + ret = imx335_update_exp_gain(imx335, exposure, analog_gain); + + break; case V4L2_CID_EXPOSURE: exposure = ctrl->val; analog_gain = imx335->again_ctrl->val; From f9d15ef2e069d509c3de590270a3fed9b67417b0 Mon Sep 17 00:00:00 2001 From: Peter Seiderer Date: Thu, 27 Feb 2025 14:56:00 +0100 Subject: [PATCH 20/66] net: pktgen: fix mpls maximum labels list parsing [ Upstream commit 2b15a0693f70d1e8119743ee89edbfb1271b3ea8 ] Fix mpls maximum labels list parsing up to MAX_MPLS_LABELS entries (instead of up to MAX_MPLS_LABELS - 1). Addresses the following: $ echo "mpls 00000f00,00000f01,00000f02,00000f03,00000f04,00000f05,00000f06,00000f07,00000f08,00000f09,00000f0a,00000f0b,00000f0c,00000f0d,00000f0e,00000f0f" > /proc/net/pktgen/lo\@0 -bash: echo: write error: Argument list too long Signed-off-by: Peter Seiderer Reviewed-by: Simon Horman Signed-off-by: Paolo Abeni Signed-off-by: Sasha Levin --- net/core/pktgen.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/net/core/pktgen.c b/net/core/pktgen.c index b6db4910359b..4d87da56c56a 100644 --- a/net/core/pktgen.c +++ b/net/core/pktgen.c @@ -898,6 +898,10 @@ static ssize_t get_labels(const char __user *buffer, struct pktgen_dev *pkt_dev) pkt_dev->nr_labels = 0; do { __u32 tmp; + + if (n >= MAX_MPLS_LABELS) + return -E2BIG; + len = hex32_arg(&buffer[i], 8, &tmp); if (len <= 0) return len; @@ -909,8 +913,6 @@ static ssize_t get_labels(const char __user *buffer, struct pktgen_dev *pkt_dev) return -EFAULT; i++; n++; - if (n >= MAX_MPLS_LABELS) - return -E2BIG; } while (c == ','); pkt_dev->nr_labels = n; From 592ba27580364c635b920eb7eae7d060e72b7aa1 Mon Sep 17 00:00:00 2001 From: Saket Kumar Bhaskar Date: Mon, 3 Mar 2025 14:54:51 +0530 Subject: [PATCH 21/66] perf/hw_breakpoint: Return EOPNOTSUPP for unsupported breakpoint type [ Upstream commit 061c991697062f3bf87b72ed553d1d33a0e370dd ] Currently, __reserve_bp_slot() returns -ENOSPC for unsupported breakpoint types on the architecture. For example, powerpc does not support hardware instruction breakpoints. This causes the perf_skip BPF selftest to fail, as neither ENOENT nor EOPNOTSUPP is returned by perf_event_open for unsupported breakpoint types. As a result, the test that should be skipped for this arch is not correctly identified. To resolve this, hw_breakpoint_event_init() should exit early by checking for unsupported breakpoint types using hw_breakpoint_slots_cached() and return the appropriate error (-EOPNOTSUPP). Signed-off-by: Saket Kumar Bhaskar Signed-off-by: Ingo Molnar Cc: Marco Elver Cc: Dmitry Vyukov Cc: Ian Rogers Cc: Frederic Weisbecker Link: https://lore.kernel.org/r/20250303092451.1862862-1-skb99@linux.ibm.com Signed-off-by: Sasha Levin --- kernel/events/hw_breakpoint.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/kernel/events/hw_breakpoint.c b/kernel/events/hw_breakpoint.c index 6c2cb4e4f48d..8f3f624419aa 100644 --- a/kernel/events/hw_breakpoint.c +++ b/kernel/events/hw_breakpoint.c @@ -950,9 +950,10 @@ static int hw_breakpoint_event_init(struct perf_event *bp) return -ENOENT; /* - * no branch sampling for breakpoint events + * Check if breakpoint type is supported before proceeding. + * Also, no branch sampling for breakpoint events. */ - if (has_branch_stack(bp)) + if (!hw_breakpoint_slots_cached(find_slot_idx(bp->attr.bp_type)) || has_branch_stack(bp)) return -EOPNOTSUPP; err = register_perf_hw_breakpoint(bp); From 500d22dbd966214cad9cfaed1d6f9e73689609de Mon Sep 17 00:00:00 2001 From: "Maciej S. Szmigiero" Date: Sun, 16 Feb 2025 22:31:03 +0100 Subject: [PATCH 22/66] ALSA: hda/realtek: Enable PC beep passthrough for HP EliteBook 855 G7 [ Upstream commit aa85822c611aef7cd4dc17d27121d43e21bb82f0 ] PC speaker works well on this platform in BIOS and in Linux until sound card drivers are loaded. Then it stops working. There seems to be a beep generator node at 0x1a in this CODEC (ALC269_TYPE_ALC215) but it seems to be only connected to capture mixers at nodes 0x22 and 0x23. If I unmute the mixer input for 0x1a at node 0x23 and start recording from its "ALC285 Analog" capture device I can clearly hear beeps in that recording. So the beep generator is indeed working properly, however I wasn't able to figure out any way to connect it to speakers. However, the bits in the "Passthrough Control" register (0x36) seems to work at least partially: by zeroing "B" and "h" and setting "S" I can at least make the PIT PC speaker output appear either in this laptop speakers or headphones (depending on whether they are connected or not). There are some caveats, however: * If the CODEC gets runtime-suspended the beeps stop so it needs HDA beep device for keeping it awake during beeping. * If the beep generator node is generating any beep the PC beep passthrough seems to be temporarily inhibited, so the HDA beep device has to be prevented from using the actual beep generator node - but the beep device is still necessary due to the previous point. * In contrast with other platforms here beep amplification has to be disabled otherwise the beeps output are WAY louder than they were on pure BIOS setup. Unless someone (from Realtek probably) knows how to make the beep generator node output appear in speakers / headphones using PC beep passthrough seems to be the only way to make PC speaker beeping actually work on this platform. Signed-off-by: Maciej S. Szmigiero Acked-by: kailang@realtek.com Link: https://patch.msgid.link/7461f695b4daed80f2fc4b1463ead47f04f9ad05.1739741254.git.mail@maciej.szmigiero.name Signed-off-by: Takashi Iwai Signed-off-by: Sasha Levin --- include/sound/hda_codec.h | 1 + sound/pci/hda/hda_beep.c | 15 +++++++++------ sound/pci/hda/patch_realtek.c | 34 +++++++++++++++++++++++++++++++++- 3 files changed, 43 insertions(+), 7 deletions(-) diff --git a/include/sound/hda_codec.h b/include/sound/hda_codec.h index 575e55aa08ca..c1fe6290d04d 100644 --- a/include/sound/hda_codec.h +++ b/include/sound/hda_codec.h @@ -195,6 +195,7 @@ struct hda_codec { /* beep device */ struct hda_beep *beep; unsigned int beep_mode; + bool beep_just_power_on; /* widget capabilities cache */ u32 *wcaps; diff --git a/sound/pci/hda/hda_beep.c b/sound/pci/hda/hda_beep.c index e51d47572557..13a7d92e8d8d 100644 --- a/sound/pci/hda/hda_beep.c +++ b/sound/pci/hda/hda_beep.c @@ -31,8 +31,9 @@ static void generate_tone(struct hda_beep *beep, int tone) beep->power_hook(beep, true); beep->playing = 1; } - snd_hda_codec_write(codec, beep->nid, 0, - AC_VERB_SET_BEEP_CONTROL, tone); + if (!codec->beep_just_power_on) + snd_hda_codec_write(codec, beep->nid, 0, + AC_VERB_SET_BEEP_CONTROL, tone); if (!tone && beep->playing) { beep->playing = 0; if (beep->power_hook) @@ -212,10 +213,12 @@ int snd_hda_attach_beep_device(struct hda_codec *codec, int nid) struct hda_beep *beep; int err; - if (!snd_hda_get_bool_hint(codec, "beep")) - return 0; /* disabled explicitly by hints */ - if (codec->beep_mode == HDA_BEEP_MODE_OFF) - return 0; /* disabled by module option */ + if (!codec->beep_just_power_on) { + if (!snd_hda_get_bool_hint(codec, "beep")) + return 0; /* disabled explicitly by hints */ + if (codec->beep_mode == HDA_BEEP_MODE_OFF) + return 0; /* disabled by module option */ + } beep = kzalloc(sizeof(*beep), GFP_KERNEL); if (beep == NULL) diff --git a/sound/pci/hda/patch_realtek.c b/sound/pci/hda/patch_realtek.c index db72c5fce9d1..0b243adac8cf 100644 --- a/sound/pci/hda/patch_realtek.c +++ b/sound/pci/hda/patch_realtek.c @@ -28,6 +28,7 @@ #include #include "hda_local.h" #include "hda_auto_parser.h" +#include "hda_beep.h" #include "hda_jack.h" #include "hda_generic.h" #include "hda_component.h" @@ -7036,6 +7037,30 @@ static void alc285_fixup_hp_envy_x360(struct hda_codec *codec, } } +static void alc285_fixup_hp_beep(struct hda_codec *codec, + const struct hda_fixup *fix, int action) +{ + if (action == HDA_FIXUP_ACT_PRE_PROBE) { + codec->beep_just_power_on = true; + } else if (action == HDA_FIXUP_ACT_INIT) { +#ifdef CONFIG_SND_HDA_INPUT_BEEP + /* + * Just enable loopback to internal speaker and headphone jack. + * Disable amplification to get about the same beep volume as + * was on pure BIOS setup before loading the driver. + */ + alc_update_coef_idx(codec, 0x36, 0x7070, BIT(13)); + + snd_hda_enable_beep_device(codec, 1); + +#if !IS_ENABLED(CONFIG_INPUT_PCSPKR) + dev_warn_once(hda_codec_dev(codec), + "enable CONFIG_INPUT_PCSPKR to get PC beeps\n"); +#endif +#endif + } +} + /* for hda_fixup_thinkpad_acpi() */ #include "thinkpad_helper.c" @@ -7806,6 +7831,7 @@ enum { ALC285_FIXUP_HP_GPIO_LED, ALC285_FIXUP_HP_MUTE_LED, ALC285_FIXUP_HP_SPECTRE_X360_MUTE_LED, + ALC285_FIXUP_HP_BEEP_MICMUTE_LED, ALC236_FIXUP_HP_MUTE_LED_COEFBIT2, ALC236_FIXUP_HP_GPIO_LED, ALC236_FIXUP_HP_MUTE_LED, @@ -9395,6 +9421,12 @@ static const struct hda_fixup alc269_fixups[] = { .type = HDA_FIXUP_FUNC, .v.func = alc285_fixup_hp_spectre_x360_mute_led, }, + [ALC285_FIXUP_HP_BEEP_MICMUTE_LED] = { + .type = HDA_FIXUP_FUNC, + .v.func = alc285_fixup_hp_beep, + .chained = true, + .chain_id = ALC285_FIXUP_HP_MUTE_LED, + }, [ALC236_FIXUP_HP_MUTE_LED_COEFBIT2] = { .type = HDA_FIXUP_FUNC, .v.func = alc236_fixup_hp_mute_led_coefbit2, @@ -10493,7 +10525,7 @@ static const struct hda_quirk alc269_fixup_tbl[] = { SND_PCI_QUIRK(0x103c, 0x8730, "HP ProBook 445 G7", ALC236_FIXUP_HP_MUTE_LED_MICMUTE_VREF), SND_PCI_QUIRK(0x103c, 0x8735, "HP ProBook 435 G7", ALC236_FIXUP_HP_MUTE_LED_MICMUTE_VREF), SND_PCI_QUIRK(0x103c, 0x8736, "HP", ALC285_FIXUP_HP_GPIO_AMP_INIT), - SND_PCI_QUIRK(0x103c, 0x8760, "HP", ALC285_FIXUP_HP_MUTE_LED), + SND_PCI_QUIRK(0x103c, 0x8760, "HP EliteBook 8{4,5}5 G7", ALC285_FIXUP_HP_BEEP_MICMUTE_LED), SND_PCI_QUIRK(0x103c, 0x876e, "HP ENVY x360 Convertible 13-ay0xxx", ALC245_FIXUP_HP_X360_MUTE_LEDS), SND_PCI_QUIRK(0x103c, 0x877a, "HP", ALC285_FIXUP_HP_MUTE_LED), SND_PCI_QUIRK(0x103c, 0x877d, "HP", ALC236_FIXUP_HP_MUTE_LED), From 3de322a98b365a7398d245d3de06971b3df1617f Mon Sep 17 00:00:00 2001 From: Nicolas Bouchinet Date: Mon, 24 Feb 2025 10:58:19 +0100 Subject: [PATCH 23/66] scsi: logging: Fix scsi_logging_level bounds [ Upstream commit 2cef5b4472c602e6c5a119aca869d9d4050586f3 ] Bound scsi_logging_level sysctl writings between SYSCTL_ZERO and SYSCTL_INT_MAX. The proc_handler has thus been updated to proc_dointvec_minmax. Signed-off-by: Nicolas Bouchinet Link: https://lore.kernel.org/r/20250224095826.16458-5-nicolas.bouchinet@clip-os.org Reviewed-by: Joel Granados Signed-off-by: Martin K. Petersen Signed-off-by: Sasha Levin --- drivers/scsi/scsi_sysctl.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/drivers/scsi/scsi_sysctl.c b/drivers/scsi/scsi_sysctl.c index 093774d77534..daa160459c9b 100644 --- a/drivers/scsi/scsi_sysctl.c +++ b/drivers/scsi/scsi_sysctl.c @@ -17,7 +17,9 @@ static struct ctl_table scsi_table[] = { .data = &scsi_logging_level, .maxlen = sizeof(scsi_logging_level), .mode = 0644, - .proc_handler = proc_dointvec }, + .proc_handler = proc_dointvec_minmax, + .extra1 = SYSCTL_ZERO, + .extra2 = SYSCTL_INT_MAX }, }; static struct ctl_table_header *scsi_table_header; From f33b310eac36a5dfc844bfc7d17f67440a32d170 Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Thu, 27 Feb 2025 20:23:27 -0800 Subject: [PATCH 24/66] ipv4: fib: Move fib_valid_key_len() to rtm_to_fib_config(). [ Upstream commit 254ba7e6032d3fc738050d500b0c1d8197af90ca ] fib_valid_key_len() is called in the beginning of fib_table_insert() or fib_table_delete() to check if the prefix length is valid. fib_table_insert() and fib_table_delete() are called from 3 paths - ip_rt_ioctl() - inet_rtm_newroute() / inet_rtm_delroute() - fib_magic() In the first ioctl() path, rtentry_to_fib_config() checks the prefix length with bad_mask(). Also, fib_magic() always passes the correct prefix: 32 or ifa->ifa_prefixlen, which is already validated. Let's move fib_valid_key_len() to the rtnetlink path, rtm_to_fib_config(). While at it, 2 direct returns in rtm_to_fib_config() are changed to goto to match other places in the same function Signed-off-by: Kuniyuki Iwashima Reviewed-by: Eric Dumazet Reviewed-by: David Ahern Link: https://patch.msgid.link/20250228042328.96624-12-kuniyu@amazon.com Signed-off-by: Jakub Kicinski Signed-off-by: Sasha Levin --- net/ipv4/fib_frontend.c | 18 ++++++++++++++++-- net/ipv4/fib_trie.c | 22 ---------------------- 2 files changed, 16 insertions(+), 24 deletions(-) diff --git a/net/ipv4/fib_frontend.c b/net/ipv4/fib_frontend.c index 793e6781399a..5b7c41333d6f 100644 --- a/net/ipv4/fib_frontend.c +++ b/net/ipv4/fib_frontend.c @@ -829,19 +829,33 @@ static int rtm_to_fib_config(struct net *net, struct sk_buff *skb, } } + if (cfg->fc_dst_len > 32) { + NL_SET_ERR_MSG(extack, "Invalid prefix length"); + err = -EINVAL; + goto errout; + } + + if (cfg->fc_dst_len < 32 && (ntohl(cfg->fc_dst) << cfg->fc_dst_len)) { + NL_SET_ERR_MSG(extack, "Invalid prefix for given prefix length"); + err = -EINVAL; + goto errout; + } + if (cfg->fc_nh_id) { if (cfg->fc_oif || cfg->fc_gw_family || cfg->fc_encap || cfg->fc_mp) { NL_SET_ERR_MSG(extack, "Nexthop specification and nexthop id are mutually exclusive"); - return -EINVAL; + err = -EINVAL; + goto errout; } } if (has_gw && has_via) { NL_SET_ERR_MSG(extack, "Nexthop configuration can not contain both GATEWAY and VIA"); - return -EINVAL; + err = -EINVAL; + goto errout; } if (!cfg->fc_table) diff --git a/net/ipv4/fib_trie.c b/net/ipv4/fib_trie.c index 09e31757e96c..cc86031d2050 100644 --- a/net/ipv4/fib_trie.c +++ b/net/ipv4/fib_trie.c @@ -1193,22 +1193,6 @@ static int fib_insert_alias(struct trie *t, struct key_vector *tp, return 0; } -static bool fib_valid_key_len(u32 key, u8 plen, struct netlink_ext_ack *extack) -{ - if (plen > KEYLENGTH) { - NL_SET_ERR_MSG(extack, "Invalid prefix length"); - return false; - } - - if ((plen < KEYLENGTH) && (key << plen)) { - NL_SET_ERR_MSG(extack, - "Invalid prefix for given prefix length"); - return false; - } - - return true; -} - static void fib_remove_alias(struct trie *t, struct key_vector *tp, struct key_vector *l, struct fib_alias *old); @@ -1229,9 +1213,6 @@ int fib_table_insert(struct net *net, struct fib_table *tb, key = ntohl(cfg->fc_dst); - if (!fib_valid_key_len(key, plen, extack)) - return -EINVAL; - pr_debug("Insert table=%u %08x/%d\n", tb->tb_id, key, plen); fi = fib_create_info(cfg, extack); @@ -1723,9 +1704,6 @@ int fib_table_delete(struct net *net, struct fib_table *tb, key = ntohl(cfg->fc_dst); - if (!fib_valid_key_len(key, plen, extack)) - return -EINVAL; - l = fib_find_node(t, &tp, key); if (!l) return -ESRCH; From f7226dedac039dfb436f3e8999aa493d216801dc Mon Sep 17 00:00:00 2001 From: Andy Yan Date: Mon, 3 Mar 2025 11:44:17 +0800 Subject: [PATCH 25/66] drm/rockchip: vop2: Add uv swap for cluster window [ Upstream commit e7aae9f6d762139f8d2b86db03793ae0ab3dd802 ] The Cluster windows of upcoming VOP on rk3576 also support linear YUV support, we need to set uv swap bit for it. As the VOP2_WIN_UV_SWA register defined on rk3568/rk3588 is 0xffffffff, so this register will not be touched on these two platforms. Signed-off-by: Andy Yan Tested-by: Michael Riesch # on RK3568 Tested-by: Detlev Casanova Signed-off-by: Heiko Stuebner Link: https://patchwork.freedesktop.org/patch/msgid/20250303034436.192400-4-andyshrk@163.com Signed-off-by: Sasha Levin --- drivers/gpu/drm/rockchip/rockchip_drm_vop2.c | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-) diff --git a/drivers/gpu/drm/rockchip/rockchip_drm_vop2.c b/drivers/gpu/drm/rockchip/rockchip_drm_vop2.c index 5880d87fe6b3..2aab2a095678 100644 --- a/drivers/gpu/drm/rockchip/rockchip_drm_vop2.c +++ b/drivers/gpu/drm/rockchip/rockchip_drm_vop2.c @@ -1432,10 +1432,8 @@ static void vop2_plane_atomic_update(struct drm_plane *plane, rb_swap = vop2_win_rb_swap(fb->format->format); vop2_win_write(win, VOP2_WIN_RB_SWAP, rb_swap); - if (!vop2_cluster_window(win)) { - uv_swap = vop2_win_uv_swap(fb->format->format); - vop2_win_write(win, VOP2_WIN_UV_SWAP, uv_swap); - } + uv_swap = vop2_win_uv_swap(fb->format->format); + vop2_win_write(win, VOP2_WIN_UV_SWAP, uv_swap); if (fb->format->is_yuv) { vop2_win_write(win, VOP2_WIN_UV_VIR, DIV_ROUND_UP(fb->pitches[1], 4)); From 564f03a7970c2c3695677d2f4bd3f1b0f23be3c1 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Tue, 25 Feb 2025 07:44:31 -0800 Subject: [PATCH 26/66] block: mark bounce buffering as incompatible with integrity [ Upstream commit 5fd0268a8806d35dcaf89139bfcda92be51b2b2f ] None of the few drivers still using the legacy block layer bounce buffering support integrity metadata. Explicitly mark the features as incompatible and stop creating the slab and mempool for integrity buffers for the bounce bio_set. Signed-off-by: Christoph Hellwig Reviewed-by: Anuj Gupta Reviewed-by: Martin K. Petersen Reviewed-by: Hannes Reinecke Reviewed-by: Johannes Thumshirn Link: https://lore.kernel.org/r/20250225154449.422989-2-hch@lst.de Signed-off-by: Jens Axboe Signed-off-by: Sasha Levin --- block/blk-settings.c | 5 +++++ block/bounce.c | 2 -- 2 files changed, 5 insertions(+), 2 deletions(-) diff --git a/block/blk-settings.c b/block/blk-settings.c index 1e63e3dd5440..7858c92b4483 100644 --- a/block/blk-settings.c +++ b/block/blk-settings.c @@ -124,6 +124,11 @@ static int blk_validate_integrity_limits(struct queue_limits *lim) return 0; } + if (lim->features & BLK_FEAT_BOUNCE_HIGH) { + pr_warn("no bounce buffer support for integrity metadata\n"); + return -EINVAL; + } + if (!IS_ENABLED(CONFIG_BLK_DEV_INTEGRITY)) { pr_warn("integrity support disabled.\n"); return -EINVAL; diff --git a/block/bounce.c b/block/bounce.c index 0d898cd5ec49..09a9616cf209 100644 --- a/block/bounce.c +++ b/block/bounce.c @@ -41,8 +41,6 @@ static void init_bounce_bioset(void) ret = bioset_init(&bounce_bio_set, BIO_POOL_SIZE, 0, BIOSET_NEED_BVECS); BUG_ON(ret); - if (bioset_integrity_create(&bounce_bio_set, BIO_POOL_SIZE)) - BUG_ON(1); ret = bioset_init(&bounce_bio_split, BIO_POOL_SIZE, 0, 0); BUG_ON(ret); From 82209faa87f4aa36140039600714379233af290f Mon Sep 17 00:00:00 2001 From: Caleb Sander Mateos Date: Tue, 25 Feb 2025 14:24:55 -0700 Subject: [PATCH 27/66] ublk: complete command synchronously on error [ Upstream commit 603f9be21c1894e462416e3324962d6c9c2b95f8 ] In case of an error, ublk's ->uring_cmd() functions currently return -EIOCBQUEUED and immediately call io_uring_cmd_done(). -EIOCBQUEUED and io_uring_cmd_done() are intended for asynchronous completions. For synchronous completions, the ->uring_cmd() function can just return the negative return code directly. This skips io_uring_cmd_del_cancelable(), and deferring the completion to task work. So return the error code directly from __ublk_ch_uring_cmd() and ublk_ctrl_uring_cmd(). Update ublk_ch_uring_cmd_cb(), which currently ignores the return value from __ublk_ch_uring_cmd(), to call io_uring_cmd_done() for synchronous completions. Signed-off-by: Caleb Sander Mateos Reviewed-by: Ming Lei Reviewed-by: Keith Busch Link: https://lore.kernel.org/r/20250225212456.2902549-1-csander@purestorage.com Signed-off-by: Jens Axboe Signed-off-by: Sasha Levin --- drivers/block/ublk_drv.c | 11 ++++++----- 1 file changed, 6 insertions(+), 5 deletions(-) diff --git a/drivers/block/ublk_drv.c b/drivers/block/ublk_drv.c index 5ec5d580ef50..a01a547c562f 100644 --- a/drivers/block/ublk_drv.c +++ b/drivers/block/ublk_drv.c @@ -1881,10 +1881,9 @@ static int __ublk_ch_uring_cmd(struct io_uring_cmd *cmd, return -EIOCBQUEUED; out: - io_uring_cmd_done(cmd, ret, 0, issue_flags); pr_devel("%s: complete: cmd op %d, tag %d ret %x io_flags %x\n", __func__, cmd_op, tag, ret, io->flags); - return -EIOCBQUEUED; + return ret; } static inline struct request *__ublk_check_and_get_req(struct ublk_device *ub, @@ -1940,7 +1939,10 @@ static inline int ublk_ch_uring_cmd_local(struct io_uring_cmd *cmd, static void ublk_ch_uring_cmd_cb(struct io_uring_cmd *cmd, unsigned int issue_flags) { - ublk_ch_uring_cmd_local(cmd, issue_flags); + int ret = ublk_ch_uring_cmd_local(cmd, issue_flags); + + if (ret != -EIOCBQUEUED) + io_uring_cmd_done(cmd, ret, 0, issue_flags); } static int ublk_ch_uring_cmd(struct io_uring_cmd *cmd, unsigned int issue_flags) @@ -3065,10 +3067,9 @@ static int ublk_ctrl_uring_cmd(struct io_uring_cmd *cmd, if (ub) ublk_put_device(ub); out: - io_uring_cmd_done(cmd, ret, 0, issue_flags); pr_devel("%s: cmd done ret %d cmd_op %x, dev id %d qid %d\n", __func__, ret, cmd->cmd_op, header->dev_id, header->queue_id); - return -EIOCBQUEUED; + return ret; } static const struct file_operations ublk_ctl_fops = { From b98aad5e5ebf91f644401833b4015790e8880720 Mon Sep 17 00:00:00 2001 From: Ricardo Ribalda Date: Mon, 3 Feb 2025 11:55:51 +0000 Subject: [PATCH 28/66] media: uvcvideo: Add sanity check to uvc_ioctl_xu_ctrl_map [ Upstream commit 990262fdfce24d6055df9711424343d94d829e6a ] Do not process unknown data types. Tested-by: Yunke Cao Reviewed-by: Hans de Goede Signed-off-by: Ricardo Ribalda Link: https://lore.kernel.org/r/20250203-uvc-roi-v17-15-5900a9fed613@chromium.org Signed-off-by: Hans de Goede Signed-off-by: Hans Verkuil Signed-off-by: Sasha Levin --- drivers/media/usb/uvc/uvc_v4l2.c | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/drivers/media/usb/uvc/uvc_v4l2.c b/drivers/media/usb/uvc/uvc_v4l2.c index 7bcd706281da..cb7d9fb589fc 100644 --- a/drivers/media/usb/uvc/uvc_v4l2.c +++ b/drivers/media/usb/uvc/uvc_v4l2.c @@ -106,6 +106,12 @@ static int uvc_ioctl_xu_ctrl_map(struct uvc_video_chain *chain, struct uvc_control_mapping *map; int ret; + if (xmap->data_type > UVC_CTRL_DATA_TYPE_BITMASK) { + uvc_dbg(chain->dev, CONTROL, + "Unsupported UVC data type %u\n", xmap->data_type); + return -EINVAL; + } + map = kzalloc(sizeof(*map), GFP_KERNEL); if (map == NULL) return -ENOMEM; From 2d6231d5ce9b66cdb52c4d1a95d92a0ad5505515 Mon Sep 17 00:00:00 2001 From: Ricardo Ribalda Date: Mon, 3 Feb 2025 11:55:40 +0000 Subject: [PATCH 29/66] media: uvcvideo: Handle uvc menu translation inside uvc_get_le_value [ Upstream commit 9109a0b4cb10fd681e9c6e9a4497a6fec5b91c39 ] map->get() gets a value from an uvc_control in "UVC format" and converts it to a value that can be consumed by v4l2. Instead of using a special get function for V4L2_CTRL_TYPE_MENU, we were converting from uvc_get_le_value in two different places. Move the conversion to uvc_get_le_value(). Reviewed-by: Hans de Goede Tested-by: Yunke Cao Signed-off-by: Ricardo Ribalda Link: https://lore.kernel.org/r/20250203-uvc-roi-v17-4-5900a9fed613@chromium.org Signed-off-by: Hans de Goede Signed-off-by: Hans Verkuil Signed-off-by: Sasha Levin --- drivers/media/usb/uvc/uvc_ctrl.c | 77 +++++++++++++------------------- 1 file changed, 32 insertions(+), 45 deletions(-) diff --git a/drivers/media/usb/uvc/uvc_ctrl.c b/drivers/media/usb/uvc/uvc_ctrl.c index 4837d8df9c03..58d1bc80253e 100644 --- a/drivers/media/usb/uvc/uvc_ctrl.c +++ b/drivers/media/usb/uvc/uvc_ctrl.c @@ -862,6 +862,25 @@ static inline void uvc_clear_bit(u8 *data, int bit) data[bit >> 3] &= ~(1 << (bit & 7)); } +static s32 uvc_menu_to_v4l2_menu(struct uvc_control_mapping *mapping, s32 val) +{ + unsigned int i; + + for (i = 0; BIT(i) <= mapping->menu_mask; ++i) { + u32 menu_value; + + if (!test_bit(i, &mapping->menu_mask)) + continue; + + menu_value = uvc_mapping_get_menu_value(mapping, i); + + if (menu_value == val) + return i; + } + + return val; +} + /* * Extract the bit string specified by mapping->offset and mapping->size * from the little-endian data stored at 'data' and return the result as @@ -896,6 +915,16 @@ static s32 uvc_get_le_value(struct uvc_control_mapping *mapping, if (mapping->data_type == UVC_CTRL_DATA_TYPE_SIGNED) value |= -(value & (1 << (mapping->size - 1))); + /* If it is a menu, convert from uvc to v4l2. */ + if (mapping->v4l2_type != V4L2_CTRL_TYPE_MENU) + return value; + + switch (query) { + case UVC_GET_CUR: + case UVC_GET_DEF: + return uvc_menu_to_v4l2_menu(mapping, value); + } + return value; } @@ -1060,32 +1089,6 @@ static int uvc_ctrl_populate_cache(struct uvc_video_chain *chain, return 0; } -static s32 __uvc_ctrl_get_value(struct uvc_control_mapping *mapping, - const u8 *data) -{ - s32 value = mapping->get(mapping, UVC_GET_CUR, data); - - if (mapping->v4l2_type == V4L2_CTRL_TYPE_MENU) { - unsigned int i; - - for (i = 0; BIT(i) <= mapping->menu_mask; ++i) { - u32 menu_value; - - if (!test_bit(i, &mapping->menu_mask)) - continue; - - menu_value = uvc_mapping_get_menu_value(mapping, i); - - if (menu_value == value) { - value = i; - break; - } - } - } - - return value; -} - static int __uvc_ctrl_load_cur(struct uvc_video_chain *chain, struct uvc_control *ctrl) { @@ -1136,8 +1139,8 @@ static int __uvc_ctrl_get(struct uvc_video_chain *chain, if (ret < 0) return ret; - *value = __uvc_ctrl_get_value(mapping, - uvc_ctrl_data(ctrl, UVC_CTRL_DATA_CURRENT)); + *value = mapping->get(mapping, UVC_GET_CUR, + uvc_ctrl_data(ctrl, UVC_CTRL_DATA_CURRENT)); return 0; } @@ -1287,7 +1290,6 @@ static int __uvc_query_v4l2_ctrl(struct uvc_video_chain *chain, { struct uvc_control_mapping *master_map = NULL; struct uvc_control *master_ctrl = NULL; - unsigned int i; memset(v4l2_ctrl, 0, sizeof(*v4l2_ctrl)); v4l2_ctrl->id = mapping->id; @@ -1330,21 +1332,6 @@ static int __uvc_query_v4l2_ctrl(struct uvc_video_chain *chain, v4l2_ctrl->minimum = ffs(mapping->menu_mask) - 1; v4l2_ctrl->maximum = fls(mapping->menu_mask) - 1; v4l2_ctrl->step = 1; - - for (i = 0; BIT(i) <= mapping->menu_mask; ++i) { - u32 menu_value; - - if (!test_bit(i, &mapping->menu_mask)) - continue; - - menu_value = uvc_mapping_get_menu_value(mapping, i); - - if (menu_value == v4l2_ctrl->default_value) { - v4l2_ctrl->default_value = i; - break; - } - } - return 0; case V4L2_CTRL_TYPE_BOOLEAN: @@ -1627,7 +1614,7 @@ void uvc_ctrl_status_event(struct uvc_video_chain *chain, uvc_ctrl_set_handle(handle, ctrl, NULL); list_for_each_entry(mapping, &ctrl->info.mappings, list) { - s32 value = __uvc_ctrl_get_value(mapping, data); + s32 value = mapping->get(mapping, UVC_GET_CUR, data); /* * handle may be NULL here if the device sends auto-update From 0d232fa3b07690035da54836346898eab441a4ad Mon Sep 17 00:00:00 2001 From: Ahmad Fatoum Date: Tue, 18 Feb 2025 19:26:46 +0100 Subject: [PATCH 30/66] clk: imx8mp: inform CCF of maximum frequency of clocks [ Upstream commit 06a61b5cb6a8638fa8823cd09b17233b29696fa2 ] The IMX8MPCEC datasheet lists maximum frequencies allowed for different modules. Some of these limits are universal, but some depend on whether the SoC is operating in nominal or in overdrive mode. The imx8mp.dtsi currently assumes overdrive mode and configures some clocks in accordance with this. Boards wishing to make use of nominal mode will need to override some of the clock rates manually. As operating the clocks outside of their allowed range can lead to difficult to debug issues, it makes sense to register the maximum rates allowed in the driver, so the CCF can take them into account. Reviewed-by: Peng Fan Signed-off-by: Ahmad Fatoum Link: https://lore.kernel.org/r/20250218-imx8m-clk-v4-6-b7697dc2dcd0@pengutronix.de Signed-off-by: Abel Vesa Signed-off-by: Sasha Levin --- drivers/clk/imx/clk-imx8mp.c | 151 +++++++++++++++++++++++++++++++++++ 1 file changed, 151 insertions(+) diff --git a/drivers/clk/imx/clk-imx8mp.c b/drivers/clk/imx/clk-imx8mp.c index fb18f507f121..fe6dac70f1a1 100644 --- a/drivers/clk/imx/clk-imx8mp.c +++ b/drivers/clk/imx/clk-imx8mp.c @@ -8,6 +8,7 @@ #include #include #include +#include #include #include #include @@ -406,11 +407,151 @@ static const char * const imx8mp_clkout_sels[] = {"audio_pll1_out", "audio_pll2_ static struct clk_hw **hws; static struct clk_hw_onecell_data *clk_hw_data; +struct imx8mp_clock_constraints { + unsigned int clkid; + u32 maxrate; +}; + +/* + * Below tables are taken from IMX8MPCEC Rev. 2.1, 07/2023 + * Table 13. Maximum frequency of modules. + * Probable typos fixed are marked with a comment. + */ +static const struct imx8mp_clock_constraints imx8mp_clock_common_constraints[] = { + { IMX8MP_CLK_A53_DIV, 1000 * HZ_PER_MHZ }, + { IMX8MP_CLK_ENET_AXI, 266666667 }, /* Datasheet claims 266MHz */ + { IMX8MP_CLK_NAND_USDHC_BUS, 266666667 }, /* Datasheet claims 266MHz */ + { IMX8MP_CLK_MEDIA_APB, 200 * HZ_PER_MHZ }, + { IMX8MP_CLK_HDMI_APB, 133333333 }, /* Datasheet claims 133MHz */ + { IMX8MP_CLK_ML_AXI, 800 * HZ_PER_MHZ }, + { IMX8MP_CLK_AHB, 133333333 }, + { IMX8MP_CLK_IPG_ROOT, 66666667 }, + { IMX8MP_CLK_AUDIO_AHB, 400 * HZ_PER_MHZ }, + { IMX8MP_CLK_MEDIA_DISP2_PIX, 170 * HZ_PER_MHZ }, + { IMX8MP_CLK_DRAM_ALT, 666666667 }, + { IMX8MP_CLK_DRAM_APB, 200 * HZ_PER_MHZ }, + { IMX8MP_CLK_CAN1, 80 * HZ_PER_MHZ }, + { IMX8MP_CLK_CAN2, 80 * HZ_PER_MHZ }, + { IMX8MP_CLK_PCIE_AUX, 10 * HZ_PER_MHZ }, + { IMX8MP_CLK_I2C5, 66666667 }, /* Datasheet claims 66MHz */ + { IMX8MP_CLK_I2C6, 66666667 }, /* Datasheet claims 66MHz */ + { IMX8MP_CLK_SAI1, 66666667 }, /* Datasheet claims 66MHz */ + { IMX8MP_CLK_SAI2, 66666667 }, /* Datasheet claims 66MHz */ + { IMX8MP_CLK_SAI3, 66666667 }, /* Datasheet claims 66MHz */ + { IMX8MP_CLK_SAI5, 66666667 }, /* Datasheet claims 66MHz */ + { IMX8MP_CLK_SAI6, 66666667 }, /* Datasheet claims 66MHz */ + { IMX8MP_CLK_ENET_QOS, 125 * HZ_PER_MHZ }, + { IMX8MP_CLK_ENET_QOS_TIMER, 200 * HZ_PER_MHZ }, + { IMX8MP_CLK_ENET_REF, 125 * HZ_PER_MHZ }, + { IMX8MP_CLK_ENET_TIMER, 125 * HZ_PER_MHZ }, + { IMX8MP_CLK_ENET_PHY_REF, 125 * HZ_PER_MHZ }, + { IMX8MP_CLK_NAND, 500 * HZ_PER_MHZ }, + { IMX8MP_CLK_QSPI, 400 * HZ_PER_MHZ }, + { IMX8MP_CLK_USDHC1, 400 * HZ_PER_MHZ }, + { IMX8MP_CLK_USDHC2, 400 * HZ_PER_MHZ }, + { IMX8MP_CLK_I2C1, 66666667 }, /* Datasheet claims 66MHz */ + { IMX8MP_CLK_I2C2, 66666667 }, /* Datasheet claims 66MHz */ + { IMX8MP_CLK_I2C3, 66666667 }, /* Datasheet claims 66MHz */ + { IMX8MP_CLK_I2C4, 66666667 }, /* Datasheet claims 66MHz */ + { IMX8MP_CLK_UART1, 80 * HZ_PER_MHZ }, + { IMX8MP_CLK_UART2, 80 * HZ_PER_MHZ }, + { IMX8MP_CLK_UART3, 80 * HZ_PER_MHZ }, + { IMX8MP_CLK_UART4, 80 * HZ_PER_MHZ }, + { IMX8MP_CLK_ECSPI1, 80 * HZ_PER_MHZ }, + { IMX8MP_CLK_ECSPI2, 80 * HZ_PER_MHZ }, + { IMX8MP_CLK_PWM1, 66666667 }, /* Datasheet claims 66MHz */ + { IMX8MP_CLK_PWM2, 66666667 }, /* Datasheet claims 66MHz */ + { IMX8MP_CLK_PWM3, 66666667 }, /* Datasheet claims 66MHz */ + { IMX8MP_CLK_PWM4, 66666667 }, /* Datasheet claims 66MHz */ + { IMX8MP_CLK_GPT1, 100 * HZ_PER_MHZ }, + { IMX8MP_CLK_GPT2, 100 * HZ_PER_MHZ }, + { IMX8MP_CLK_GPT3, 100 * HZ_PER_MHZ }, + { IMX8MP_CLK_GPT4, 100 * HZ_PER_MHZ }, + { IMX8MP_CLK_GPT5, 100 * HZ_PER_MHZ }, + { IMX8MP_CLK_GPT6, 100 * HZ_PER_MHZ }, + { IMX8MP_CLK_WDOG, 66666667 }, /* Datasheet claims 66MHz */ + { IMX8MP_CLK_IPP_DO_CLKO1, 200 * HZ_PER_MHZ }, + { IMX8MP_CLK_IPP_DO_CLKO2, 200 * HZ_PER_MHZ }, + { IMX8MP_CLK_HDMI_REF_266M, 266 * HZ_PER_MHZ }, + { IMX8MP_CLK_USDHC3, 400 * HZ_PER_MHZ }, + { IMX8MP_CLK_MEDIA_MIPI_PHY1_REF, 300 * HZ_PER_MHZ }, + { IMX8MP_CLK_MEDIA_DISP1_PIX, 250 * HZ_PER_MHZ }, + { IMX8MP_CLK_MEDIA_CAM2_PIX, 277 * HZ_PER_MHZ }, + { IMX8MP_CLK_MEDIA_LDB, 595 * HZ_PER_MHZ }, + { IMX8MP_CLK_MEDIA_MIPI_TEST_BYTE, 200 * HZ_PER_MHZ }, + { IMX8MP_CLK_ECSPI3, 80 * HZ_PER_MHZ }, + { IMX8MP_CLK_PDM, 200 * HZ_PER_MHZ }, + { IMX8MP_CLK_SAI7, 66666667 }, /* Datasheet claims 66MHz */ + { IMX8MP_CLK_MAIN_AXI, 400 * HZ_PER_MHZ }, + { /* Sentinel */ } +}; + +static const struct imx8mp_clock_constraints imx8mp_clock_nominal_constraints[] = { + { IMX8MP_CLK_M7_CORE, 600 * HZ_PER_MHZ }, + { IMX8MP_CLK_ML_CORE, 800 * HZ_PER_MHZ }, + { IMX8MP_CLK_GPU3D_CORE, 800 * HZ_PER_MHZ }, + { IMX8MP_CLK_GPU3D_SHADER_CORE, 800 * HZ_PER_MHZ }, + { IMX8MP_CLK_GPU2D_CORE, 800 * HZ_PER_MHZ }, + { IMX8MP_CLK_AUDIO_AXI_SRC, 600 * HZ_PER_MHZ }, + { IMX8MP_CLK_HSIO_AXI, 400 * HZ_PER_MHZ }, + { IMX8MP_CLK_MEDIA_ISP, 400 * HZ_PER_MHZ }, + { IMX8MP_CLK_VPU_BUS, 600 * HZ_PER_MHZ }, + { IMX8MP_CLK_MEDIA_AXI, 400 * HZ_PER_MHZ }, + { IMX8MP_CLK_HDMI_AXI, 400 * HZ_PER_MHZ }, + { IMX8MP_CLK_GPU_AXI, 600 * HZ_PER_MHZ }, + { IMX8MP_CLK_GPU_AHB, 300 * HZ_PER_MHZ }, + { IMX8MP_CLK_NOC, 800 * HZ_PER_MHZ }, + { IMX8MP_CLK_NOC_IO, 600 * HZ_PER_MHZ }, + { IMX8MP_CLK_ML_AHB, 300 * HZ_PER_MHZ }, + { IMX8MP_CLK_VPU_G1, 600 * HZ_PER_MHZ }, + { IMX8MP_CLK_VPU_G2, 500 * HZ_PER_MHZ }, + { IMX8MP_CLK_MEDIA_CAM1_PIX, 400 * HZ_PER_MHZ }, + { IMX8MP_CLK_VPU_VC8000E, 400 * HZ_PER_MHZ }, /* Datasheet claims 500MHz */ + { IMX8MP_CLK_DRAM_CORE, 800 * HZ_PER_MHZ }, + { IMX8MP_CLK_GIC, 400 * HZ_PER_MHZ }, + { /* Sentinel */ } +}; + +static const struct imx8mp_clock_constraints imx8mp_clock_overdrive_constraints[] = { + { IMX8MP_CLK_M7_CORE, 800 * HZ_PER_MHZ}, + { IMX8MP_CLK_ML_CORE, 1000 * HZ_PER_MHZ }, + { IMX8MP_CLK_GPU3D_CORE, 1000 * HZ_PER_MHZ }, + { IMX8MP_CLK_GPU3D_SHADER_CORE, 1000 * HZ_PER_MHZ }, + { IMX8MP_CLK_GPU2D_CORE, 1000 * HZ_PER_MHZ }, + { IMX8MP_CLK_AUDIO_AXI_SRC, 800 * HZ_PER_MHZ }, + { IMX8MP_CLK_HSIO_AXI, 500 * HZ_PER_MHZ }, + { IMX8MP_CLK_MEDIA_ISP, 500 * HZ_PER_MHZ }, + { IMX8MP_CLK_VPU_BUS, 800 * HZ_PER_MHZ }, + { IMX8MP_CLK_MEDIA_AXI, 500 * HZ_PER_MHZ }, + { IMX8MP_CLK_HDMI_AXI, 500 * HZ_PER_MHZ }, + { IMX8MP_CLK_GPU_AXI, 800 * HZ_PER_MHZ }, + { IMX8MP_CLK_GPU_AHB, 400 * HZ_PER_MHZ }, + { IMX8MP_CLK_NOC, 1000 * HZ_PER_MHZ }, + { IMX8MP_CLK_NOC_IO, 800 * HZ_PER_MHZ }, + { IMX8MP_CLK_ML_AHB, 400 * HZ_PER_MHZ }, + { IMX8MP_CLK_VPU_G1, 800 * HZ_PER_MHZ }, + { IMX8MP_CLK_VPU_G2, 700 * HZ_PER_MHZ }, + { IMX8MP_CLK_MEDIA_CAM1_PIX, 500 * HZ_PER_MHZ }, + { IMX8MP_CLK_VPU_VC8000E, 500 * HZ_PER_MHZ }, /* Datasheet claims 400MHz */ + { IMX8MP_CLK_DRAM_CORE, 1000 * HZ_PER_MHZ }, + { IMX8MP_CLK_GIC, 500 * HZ_PER_MHZ }, + { /* Sentinel */ } +}; + +static void imx8mp_clocks_apply_constraints(const struct imx8mp_clock_constraints constraints[]) +{ + const struct imx8mp_clock_constraints *constr; + + for (constr = constraints; constr->clkid; constr++) + clk_hw_set_rate_range(hws[constr->clkid], 0, constr->maxrate); +} + static int imx8mp_clocks_probe(struct platform_device *pdev) { struct device *dev = &pdev->dev; struct device_node *np; void __iomem *anatop_base, *ccm_base; + const char *opmode; int err; np = of_find_compatible_node(NULL, NULL, "fsl,imx8mp-anatop"); @@ -715,6 +856,16 @@ static int imx8mp_clocks_probe(struct platform_device *pdev) imx_check_clk_hws(hws, IMX8MP_CLK_END); + imx8mp_clocks_apply_constraints(imx8mp_clock_common_constraints); + + err = of_property_read_string(np, "fsl,operating-mode", &opmode); + if (!err) { + if (!strcmp(opmode, "nominal")) + imx8mp_clocks_apply_constraints(imx8mp_clock_nominal_constraints); + else if (!strcmp(opmode, "overdrive")) + imx8mp_clocks_apply_constraints(imx8mp_clock_overdrive_constraints); + } + err = of_clk_add_hw_provider(np, of_clk_hw_onecell_get, clk_hw_data); if (err < 0) { dev_err(dev, "failed to register hws for i.MX8MP\n"); From e7e30a4a37d1e5fd0bae01a51ef738c17606f9f8 Mon Sep 17 00:00:00 2001 From: Breno Leitao Date: Thu, 31 Oct 2024 04:06:17 -0700 Subject: [PATCH 31/66] x86/bugs: Make spectre user default depend on MITIGATION_SPECTRE_V2 [ Upstream commit 98fdaeb296f51ef08e727a7cc72e5b5c864c4f4d ] Change the default value of spectre v2 in user mode to respect the CONFIG_MITIGATION_SPECTRE_V2 config option. Currently, user mode spectre v2 is set to auto (SPECTRE_V2_USER_CMD_AUTO) by default, even if CONFIG_MITIGATION_SPECTRE_V2 is disabled. Set the spectre_v2 value to auto (SPECTRE_V2_USER_CMD_AUTO) if the Spectre v2 config (CONFIG_MITIGATION_SPECTRE_V2) is enabled, otherwise set the value to none (SPECTRE_V2_USER_CMD_NONE). Important to say the command line argument "spectre_v2_user" overwrites the default value in both cases. When CONFIG_MITIGATION_SPECTRE_V2 is not set, users have the flexibility to opt-in for specific mitigations independently. In this scenario, setting spectre_v2= will not enable spectre_v2_user=, and command line options spectre_v2_user and spectre_v2 are independent when CONFIG_MITIGATION_SPECTRE_V2=n. Signed-off-by: Breno Leitao Signed-off-by: Ingo Molnar Reviewed-by: Pawan Gupta Acked-by: Josh Poimboeuf Cc: Peter Zijlstra Cc: David Kaplan Link: https://lore.kernel.org/r/20241031-x86_bugs_last_v2-v2-2-b7ff1dab840e@debian.org Signed-off-by: Sasha Levin --- Documentation/admin-guide/kernel-parameters.txt | 2 ++ arch/x86/kernel/cpu/bugs.c | 10 +++++++--- 2 files changed, 9 insertions(+), 3 deletions(-) diff --git a/Documentation/admin-guide/kernel-parameters.txt b/Documentation/admin-guide/kernel-parameters.txt index e691f75c97e7..b5cb36148554 100644 --- a/Documentation/admin-guide/kernel-parameters.txt +++ b/Documentation/admin-guide/kernel-parameters.txt @@ -6261,6 +6261,8 @@ Selecting 'on' will also enable the mitigation against user space to user space task attacks. + Selecting specific mitigation does not force enable + user mitigations. Selecting 'off' will disable both the kernel and the user space protections. diff --git a/arch/x86/kernel/cpu/bugs.c b/arch/x86/kernel/cpu/bugs.c index c683abd640fd..0e9ab0b9a494 100644 --- a/arch/x86/kernel/cpu/bugs.c +++ b/arch/x86/kernel/cpu/bugs.c @@ -1442,9 +1442,13 @@ static __ro_after_init enum spectre_v2_mitigation_cmd spectre_v2_cmd; static enum spectre_v2_user_cmd __init spectre_v2_parse_user_cmdline(void) { + enum spectre_v2_user_cmd mode; char arg[20]; int ret, i; + mode = IS_ENABLED(CONFIG_MITIGATION_SPECTRE_V2) ? + SPECTRE_V2_USER_CMD_AUTO : SPECTRE_V2_USER_CMD_NONE; + switch (spectre_v2_cmd) { case SPECTRE_V2_CMD_NONE: return SPECTRE_V2_USER_CMD_NONE; @@ -1457,7 +1461,7 @@ spectre_v2_parse_user_cmdline(void) ret = cmdline_find_option(boot_command_line, "spectre_v2_user", arg, sizeof(arg)); if (ret < 0) - return SPECTRE_V2_USER_CMD_AUTO; + return mode; for (i = 0; i < ARRAY_SIZE(v2_user_options); i++) { if (match_option(arg, ret, v2_user_options[i].option)) { @@ -1467,8 +1471,8 @@ spectre_v2_parse_user_cmdline(void) } } - pr_err("Unknown user space protection option (%s). Switching to AUTO select\n", arg); - return SPECTRE_V2_USER_CMD_AUTO; + pr_err("Unknown user space protection option (%s). Switching to default\n", arg); + return mode; } static inline bool spectre_v2_in_ibrs_mode(enum spectre_v2_mitigation mode) From cd62e9d42fe76265d545a0c340e1d7307ba45808 Mon Sep 17 00:00:00 2001 From: Alexander Stein Date: Mon, 10 Feb 2025 15:59:30 +0100 Subject: [PATCH 32/66] hwmon: (gpio-fan) Add missing mutex locks [ Upstream commit 9fee7d19bab635f89223cc40dfd2c8797fdc4988 ] set_fan_speed() is expected to be called with fan_data->lock being locked. Add locking for proper synchronization. Signed-off-by: Alexander Stein Link: https://lore.kernel.org/r/20250210145934.761280-3-alexander.stein@ew.tq-group.com Signed-off-by: Guenter Roeck Signed-off-by: Sasha Levin --- drivers/hwmon/gpio-fan.c | 16 +++++++++++++++- 1 file changed, 15 insertions(+), 1 deletion(-) diff --git a/drivers/hwmon/gpio-fan.c b/drivers/hwmon/gpio-fan.c index d92c536be9af..b779240328d5 100644 --- a/drivers/hwmon/gpio-fan.c +++ b/drivers/hwmon/gpio-fan.c @@ -393,7 +393,12 @@ static int gpio_fan_set_cur_state(struct thermal_cooling_device *cdev, if (state >= fan_data->num_speed) return -EINVAL; + mutex_lock(&fan_data->lock); + set_fan_speed(fan_data, state); + + mutex_unlock(&fan_data->lock); + return 0; } @@ -489,7 +494,11 @@ MODULE_DEVICE_TABLE(of, of_gpio_fan_match); static void gpio_fan_stop(void *data) { + struct gpio_fan_data *fan_data = data; + + mutex_lock(&fan_data->lock); set_fan_speed(data, 0); + mutex_unlock(&fan_data->lock); } static int gpio_fan_probe(struct platform_device *pdev) @@ -562,7 +571,9 @@ static int gpio_fan_suspend(struct device *dev) if (fan_data->gpios) { fan_data->resume_speed = fan_data->speed_index; + mutex_lock(&fan_data->lock); set_fan_speed(fan_data, 0); + mutex_unlock(&fan_data->lock); } return 0; @@ -572,8 +583,11 @@ static int gpio_fan_resume(struct device *dev) { struct gpio_fan_data *fan_data = dev_get_drvdata(dev); - if (fan_data->gpios) + if (fan_data->gpios) { + mutex_lock(&fan_data->lock); set_fan_speed(fan_data, fan_data->resume_speed); + mutex_unlock(&fan_data->lock); + } return 0; } From 89800836bf32f5b0c14734c64d804fe73e34744f Mon Sep 17 00:00:00 2001 From: Li Bin Date: Thu, 27 Feb 2025 08:51:56 -0700 Subject: [PATCH 33/66] ARM: at91: pm: fix at91_suspend_finish for ZQ calibration [ Upstream commit bc4722c3598d0e2c2dbf9609a3d3198993093e2b ] For sama7g5 and sama7d65 backup mode, we encountered a "ZQ calibrate error" during recalibrating the impedance in BootStrap. We found that the impedance value saved in at91_suspend_finish() before the DDR entered self-refresh mode did not match the resistor values. The ZDATA field in the DDR3PHY_ZQ0CR0 register uses a modified gray code to select the different impedance setting. But these gray code are incorrect, a workaournd from design team fixed the bug in the calibration logic. The ZDATA contains four independent impedance elements, but the algorithm combined the four elements into one. The elements were fixed using properly shifted offsets. Signed-off-by: Li Bin [nicolas.ferre@microchip.com: fix indentation and combine 2 patches] Signed-off-by: Nicolas Ferre Tested-by: Ryan Wanner Tested-by: Durai Manickam KR Tested-by: Andrei Simion Signed-off-by: Ryan Wanner Link: https://lore.kernel.org/r/28b33f9bcd0ca60ceba032969fe054d38f2b9577.1740671156.git.Ryan.Wanner@microchip.com Signed-off-by: Claudiu Beznea Signed-off-by: Sasha Levin --- arch/arm/mach-at91/pm.c | 21 +++++++++++---------- 1 file changed, 11 insertions(+), 10 deletions(-) diff --git a/arch/arm/mach-at91/pm.c b/arch/arm/mach-at91/pm.c index 05a1547642b6..6c3e6aa22606 100644 --- a/arch/arm/mach-at91/pm.c +++ b/arch/arm/mach-at91/pm.c @@ -545,11 +545,12 @@ extern u32 at91_pm_suspend_in_sram_sz; static int at91_suspend_finish(unsigned long val) { - unsigned char modified_gray_code[] = { - 0x00, 0x01, 0x02, 0x03, 0x06, 0x07, 0x04, 0x05, 0x0c, 0x0d, - 0x0e, 0x0f, 0x0a, 0x0b, 0x08, 0x09, 0x18, 0x19, 0x1a, 0x1b, - 0x1e, 0x1f, 0x1c, 0x1d, 0x14, 0x15, 0x16, 0x17, 0x12, 0x13, - 0x10, 0x11, + /* SYNOPSYS workaround to fix a bug in the calibration logic */ + unsigned char modified_fix_code[] = { + 0x00, 0x01, 0x01, 0x06, 0x07, 0x0c, 0x06, 0x07, 0x0b, 0x18, + 0x0a, 0x0b, 0x0c, 0x0d, 0x0d, 0x0a, 0x13, 0x13, 0x12, 0x13, + 0x14, 0x15, 0x15, 0x12, 0x18, 0x19, 0x19, 0x1e, 0x1f, 0x14, + 0x1e, 0x1f, }; unsigned int tmp, index; int i; @@ -560,25 +561,25 @@ static int at91_suspend_finish(unsigned long val) * restore the ZQ0SR0 with the value saved here. But the * calibration is buggy and restoring some values from ZQ0SR0 * is forbidden and risky thus we need to provide processed - * values for these (modified gray code values). + * values for these. */ tmp = readl(soc_pm.data.ramc_phy + DDR3PHY_ZQ0SR0); /* Store pull-down output impedance select. */ index = (tmp >> DDR3PHY_ZQ0SR0_PDO_OFF) & 0x1f; - soc_pm.bu->ddr_phy_calibration[0] = modified_gray_code[index]; + soc_pm.bu->ddr_phy_calibration[0] = modified_fix_code[index] << DDR3PHY_ZQ0SR0_PDO_OFF; /* Store pull-up output impedance select. */ index = (tmp >> DDR3PHY_ZQ0SR0_PUO_OFF) & 0x1f; - soc_pm.bu->ddr_phy_calibration[0] |= modified_gray_code[index]; + soc_pm.bu->ddr_phy_calibration[0] |= modified_fix_code[index] << DDR3PHY_ZQ0SR0_PUO_OFF; /* Store pull-down on-die termination impedance select. */ index = (tmp >> DDR3PHY_ZQ0SR0_PDODT_OFF) & 0x1f; - soc_pm.bu->ddr_phy_calibration[0] |= modified_gray_code[index]; + soc_pm.bu->ddr_phy_calibration[0] |= modified_fix_code[index] << DDR3PHY_ZQ0SR0_PDODT_OFF; /* Store pull-up on-die termination impedance select. */ index = (tmp >> DDR3PHY_ZQ0SRO_PUODT_OFF) & 0x1f; - soc_pm.bu->ddr_phy_calibration[0] |= modified_gray_code[index]; + soc_pm.bu->ddr_phy_calibration[0] |= modified_fix_code[index] << DDR3PHY_ZQ0SRO_PUODT_OFF; /* * The 1st 8 words of memory might get corrupted in the process From 7857d8977e9864fef9a61d98b3814dd71c64443f Mon Sep 17 00:00:00 2001 From: AngeloGioacchino Del Regno Date: Mon, 17 Feb 2025 16:47:58 +0100 Subject: [PATCH 34/66] drm/mediatek: mtk_dpi: Add checks for reg_h_fre_con existence [ Upstream commit 8c9da7cd0bbcc90ab444454fecf535320456a312 ] In preparation for adding support for newer DPI instances which do support direct-pin but do not have any H_FRE_CON register, like the one found in MT8195 and MT8188, add a branch to check if the reg_h_fre_con variable was declared in the mtk_dpi_conf structure for the probed SoC DPI version. As a note, this is useful specifically only for cases in which the support_direct_pin variable is true, so mt8195-dpintf is not affected by any issue. Reviewed-by: CK Hu Signed-off-by: AngeloGioacchino Del Regno Link: https://patchwork.kernel.org/project/dri-devel/patch/20250217154836.108895-6-angelogioacchino.delregno@collabora.com/ Signed-off-by: Chun-Kuang Hu Signed-off-by: Sasha Levin --- drivers/gpu/drm/mediatek/mtk_dpi.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/drm/mediatek/mtk_dpi.c b/drivers/gpu/drm/mediatek/mtk_dpi.c index 9c11d3158324..20a50180d4d4 100644 --- a/drivers/gpu/drm/mediatek/mtk_dpi.c +++ b/drivers/gpu/drm/mediatek/mtk_dpi.c @@ -410,12 +410,13 @@ static void mtk_dpi_config_swap_input(struct mtk_dpi *dpi, bool enable) static void mtk_dpi_config_2n_h_fre(struct mtk_dpi *dpi) { - mtk_dpi_mask(dpi, dpi->conf->reg_h_fre_con, H_FRE_2N, H_FRE_2N); + if (dpi->conf->reg_h_fre_con) + mtk_dpi_mask(dpi, dpi->conf->reg_h_fre_con, H_FRE_2N, H_FRE_2N); } static void mtk_dpi_config_disable_edge(struct mtk_dpi *dpi) { - if (dpi->conf->edge_sel_en) + if (dpi->conf->edge_sel_en && dpi->conf->reg_h_fre_con) mtk_dpi_mask(dpi, dpi->conf->reg_h_fre_con, 0, EDGE_SEL_EN); } From b38fbf98645be29e53f02380fa679237f667fded Mon Sep 17 00:00:00 2001 From: Kuhanh Murugasen Krishnan Date: Thu, 13 Feb 2025 06:12:49 +0800 Subject: [PATCH 35/66] fpga: altera-cvp: Increase credit timeout [ Upstream commit 0f05886a40fdc55016ba4d9ae0a9c41f8312f15b ] Increase the timeout for SDM (Secure device manager) data credits from 20ms to 40ms. Internal stress tests running at 500 loops failed with the current timeout of 20ms. At the start of a FPGA configuration, the CVP host driver reads the transmit credits from SDM. It then sends bitstream FPGA data to SDM based on the total credits. Each credit allows the CVP host driver to send 4kBytes of data. There are situations whereby, the SDM did not respond in time during testing. Signed-off-by: Ang Tien Sung Signed-off-by: Kuhanh Murugasen Krishnan Acked-by: Xu Yilun Link: https://lore.kernel.org/r/20250212221249.2715929-1-tien.sung.ang@intel.com Signed-off-by: Xu Yilun Signed-off-by: Sasha Levin --- drivers/fpga/altera-cvp.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/fpga/altera-cvp.c b/drivers/fpga/altera-cvp.c index 6b0914432445..5af0bd33890c 100644 --- a/drivers/fpga/altera-cvp.c +++ b/drivers/fpga/altera-cvp.c @@ -52,7 +52,7 @@ /* V2 Defines */ #define VSE_CVP_TX_CREDITS 0x49 /* 8bit */ -#define V2_CREDIT_TIMEOUT_US 20000 +#define V2_CREDIT_TIMEOUT_US 40000 #define V2_CHECK_CREDIT_US 10 #define V2_POLL_TIMEOUT_US 1000000 #define V2_USER_TIMEOUT_US 500000 From e359d62886cb1d7224e331575d6ab07f97b7fb14 Mon Sep 17 00:00:00 2001 From: "Rob Herring (Arm)" Date: Tue, 18 Feb 2025 14:39:56 -0600 Subject: [PATCH 36/66] perf: arm_pmuv3: Call kvm_vcpu_pmu_resync_el0() before enabling counters [ Upstream commit 04bd15c4cbc3f7bd2399d1baab958c5e738dbfc9 ] Counting events related to setup of the PMU is not desired, but kvm_vcpu_pmu_resync_el0() is called just after the PMU counters have been enabled. Move the call to before enabling the counters. Signed-off-by: Rob Herring (Arm) Reviewed-by: Anshuman Khandual Tested-by: James Clark Link: https://lore.kernel.org/r/20250218-arm-brbe-v19-v20-1-4e9922fc2e8e@kernel.org Signed-off-by: Will Deacon Signed-off-by: Sasha Levin --- drivers/perf/arm_pmuv3.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/perf/arm_pmuv3.c b/drivers/perf/arm_pmuv3.c index 0afe02f879b4..d9b56f0f9094 100644 --- a/drivers/perf/arm_pmuv3.c +++ b/drivers/perf/arm_pmuv3.c @@ -816,10 +816,10 @@ static void armv8pmu_start(struct arm_pmu *cpu_pmu) else armv8pmu_disable_user_access(); + kvm_vcpu_pmu_resync_el0(); + /* Enable all counters */ armv8pmu_pmcr_write(armv8pmu_pmcr_read() | ARMV8_PMU_PMCR_E); - - kvm_vcpu_pmu_resync_el0(); } static void armv8pmu_stop(struct arm_pmu *cpu_pmu) From 92c6687ba951f311dc36c0bb7c62dd9b5fb9e244 Mon Sep 17 00:00:00 2001 From: Janne Grunau Date: Wed, 26 Feb 2025 19:00:05 +0000 Subject: [PATCH 37/66] soc: apple: rtkit: Use high prio work queue [ Upstream commit 22af2fac88fa5dbc310bfe7d0b66d4de3ac47305 ] rtkit messages as communication with the DCP firmware for framebuffer swaps or input events are time critical so use WQ_HIGHPRI to prevent user space CPU load to increase latency. With kwin_wayland 6's explicit sync mode user space load was able to delay the IOMFB rtkit communication enough to miss vsync for surface swaps. Minimal test scenario is constantly resizing a glxgears Xwayland window. Signed-off-by: Janne Grunau Reviewed-by: Alyssa Rosenzweig Link: https://lore.kernel.org/r/20250226-apple-soc-misc-v2-3-c3ec37f9021b@svenpeter.dev Signed-off-by: Sven Peter Signed-off-by: Sasha Levin --- drivers/soc/apple/rtkit.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/soc/apple/rtkit.c b/drivers/soc/apple/rtkit.c index e6d940292c9f..00d59a81db88 100644 --- a/drivers/soc/apple/rtkit.c +++ b/drivers/soc/apple/rtkit.c @@ -667,7 +667,7 @@ struct apple_rtkit *apple_rtkit_init(struct device *dev, void *cookie, rtk->mbox->rx = apple_rtkit_rx; rtk->mbox->cookie = rtk; - rtk->wq = alloc_ordered_workqueue("rtkit-%s", WQ_MEM_RECLAIM, + rtk->wq = alloc_ordered_workqueue("rtkit-%s", WQ_HIGHPRI | WQ_MEM_RECLAIM, dev_name(rtk->dev)); if (!rtk->wq) { ret = -ENOMEM; From 1c2c538bbd60288f5de1a70de1789d8a00294336 Mon Sep 17 00:00:00 2001 From: Hector Martin Date: Wed, 26 Feb 2025 19:00:04 +0000 Subject: [PATCH 38/66] soc: apple: rtkit: Implement OSLog buffers properly [ Upstream commit a06398687065e0c334dc5fc4d2778b5b87292e43 ] Apparently nobody can figure out where the old logic came from, but it seems like it has never been actually used on any supported firmware to this day. OSLog buffers were apparently never requested. But starting with 13.3, we actually need this implemented properly for MTP (and later AOP) to work, so let's actually do that. Signed-off-by: Hector Martin Reviewed-by: Alyssa Rosenzweig Link: https://lore.kernel.org/r/20250226-apple-soc-misc-v2-2-c3ec37f9021b@svenpeter.dev Signed-off-by: Sven Peter Signed-off-by: Sasha Levin --- drivers/soc/apple/rtkit-internal.h | 1 + drivers/soc/apple/rtkit.c | 56 ++++++++++++++++++------------ 2 files changed, 35 insertions(+), 22 deletions(-) diff --git a/drivers/soc/apple/rtkit-internal.h b/drivers/soc/apple/rtkit-internal.h index 27c9fa745fd5..b8d5244678f0 100644 --- a/drivers/soc/apple/rtkit-internal.h +++ b/drivers/soc/apple/rtkit-internal.h @@ -44,6 +44,7 @@ struct apple_rtkit { struct apple_rtkit_shmem ioreport_buffer; struct apple_rtkit_shmem crashlog_buffer; + struct apple_rtkit_shmem oslog_buffer; struct apple_rtkit_shmem syslog_buffer; char *syslog_msg_buffer; diff --git a/drivers/soc/apple/rtkit.c b/drivers/soc/apple/rtkit.c index 00d59a81db88..45ccbe2cbcd6 100644 --- a/drivers/soc/apple/rtkit.c +++ b/drivers/soc/apple/rtkit.c @@ -66,8 +66,9 @@ enum { #define APPLE_RTKIT_SYSLOG_MSG_SIZE GENMASK_ULL(31, 24) #define APPLE_RTKIT_OSLOG_TYPE GENMASK_ULL(63, 56) -#define APPLE_RTKIT_OSLOG_INIT 1 -#define APPLE_RTKIT_OSLOG_ACK 3 +#define APPLE_RTKIT_OSLOG_BUFFER_REQUEST 1 +#define APPLE_RTKIT_OSLOG_SIZE GENMASK_ULL(55, 36) +#define APPLE_RTKIT_OSLOG_IOVA GENMASK_ULL(35, 0) #define APPLE_RTKIT_MIN_SUPPORTED_VERSION 11 #define APPLE_RTKIT_MAX_SUPPORTED_VERSION 12 @@ -251,15 +252,21 @@ static int apple_rtkit_common_rx_get_buffer(struct apple_rtkit *rtk, struct apple_rtkit_shmem *buffer, u8 ep, u64 msg) { - size_t n_4kpages = FIELD_GET(APPLE_RTKIT_BUFFER_REQUEST_SIZE, msg); u64 reply; int err; + /* The different size vs. IOVA shifts look odd but are indeed correct this way */ + if (ep == APPLE_RTKIT_EP_OSLOG) { + buffer->size = FIELD_GET(APPLE_RTKIT_OSLOG_SIZE, msg); + buffer->iova = FIELD_GET(APPLE_RTKIT_OSLOG_IOVA, msg) << 12; + } else { + buffer->size = FIELD_GET(APPLE_RTKIT_BUFFER_REQUEST_SIZE, msg) << 12; + buffer->iova = FIELD_GET(APPLE_RTKIT_BUFFER_REQUEST_IOVA, msg); + } + buffer->buffer = NULL; buffer->iomem = NULL; buffer->is_mapped = false; - buffer->iova = FIELD_GET(APPLE_RTKIT_BUFFER_REQUEST_IOVA, msg); - buffer->size = n_4kpages << 12; dev_dbg(rtk->dev, "RTKit: buffer request for 0x%zx bytes at %pad\n", buffer->size, &buffer->iova); @@ -284,11 +291,21 @@ static int apple_rtkit_common_rx_get_buffer(struct apple_rtkit *rtk, } if (!buffer->is_mapped) { - reply = FIELD_PREP(APPLE_RTKIT_SYSLOG_TYPE, - APPLE_RTKIT_BUFFER_REQUEST); - reply |= FIELD_PREP(APPLE_RTKIT_BUFFER_REQUEST_SIZE, n_4kpages); - reply |= FIELD_PREP(APPLE_RTKIT_BUFFER_REQUEST_IOVA, - buffer->iova); + /* oslog uses different fields and needs a shifted IOVA instead of size */ + if (ep == APPLE_RTKIT_EP_OSLOG) { + reply = FIELD_PREP(APPLE_RTKIT_OSLOG_TYPE, + APPLE_RTKIT_OSLOG_BUFFER_REQUEST); + reply |= FIELD_PREP(APPLE_RTKIT_OSLOG_SIZE, buffer->size); + reply |= FIELD_PREP(APPLE_RTKIT_OSLOG_IOVA, + buffer->iova >> 12); + } else { + reply = FIELD_PREP(APPLE_RTKIT_SYSLOG_TYPE, + APPLE_RTKIT_BUFFER_REQUEST); + reply |= FIELD_PREP(APPLE_RTKIT_BUFFER_REQUEST_SIZE, + buffer->size >> 12); + reply |= FIELD_PREP(APPLE_RTKIT_BUFFER_REQUEST_IOVA, + buffer->iova); + } apple_rtkit_send_message(rtk, ep, reply, NULL, false); } @@ -482,25 +499,18 @@ static void apple_rtkit_syslog_rx(struct apple_rtkit *rtk, u64 msg) } } -static void apple_rtkit_oslog_rx_init(struct apple_rtkit *rtk, u64 msg) -{ - u64 ack; - - dev_dbg(rtk->dev, "RTKit: oslog init: msg: 0x%llx\n", msg); - ack = FIELD_PREP(APPLE_RTKIT_OSLOG_TYPE, APPLE_RTKIT_OSLOG_ACK); - apple_rtkit_send_message(rtk, APPLE_RTKIT_EP_OSLOG, ack, NULL, false); -} - static void apple_rtkit_oslog_rx(struct apple_rtkit *rtk, u64 msg) { u8 type = FIELD_GET(APPLE_RTKIT_OSLOG_TYPE, msg); switch (type) { - case APPLE_RTKIT_OSLOG_INIT: - apple_rtkit_oslog_rx_init(rtk, msg); + case APPLE_RTKIT_OSLOG_BUFFER_REQUEST: + apple_rtkit_common_rx_get_buffer(rtk, &rtk->oslog_buffer, + APPLE_RTKIT_EP_OSLOG, msg); break; default: - dev_warn(rtk->dev, "RTKit: Unknown oslog message: %llx\n", msg); + dev_warn(rtk->dev, "RTKit: Unknown oslog message: %llx\n", + msg); } } @@ -710,6 +720,7 @@ int apple_rtkit_reinit(struct apple_rtkit *rtk) apple_rtkit_free_buffer(rtk, &rtk->ioreport_buffer); apple_rtkit_free_buffer(rtk, &rtk->crashlog_buffer); + apple_rtkit_free_buffer(rtk, &rtk->oslog_buffer); apple_rtkit_free_buffer(rtk, &rtk->syslog_buffer); kfree(rtk->syslog_msg_buffer); @@ -890,6 +901,7 @@ void apple_rtkit_free(struct apple_rtkit *rtk) apple_rtkit_free_buffer(rtk, &rtk->ioreport_buffer); apple_rtkit_free_buffer(rtk, &rtk->crashlog_buffer); + apple_rtkit_free_buffer(rtk, &rtk->oslog_buffer); apple_rtkit_free_buffer(rtk, &rtk->syslog_buffer); kfree(rtk->syslog_msg_buffer); From 7add9c10ca284b9ff599fa6a1faae3c357425db7 Mon Sep 17 00:00:00 2001 From: Vinith Kumar R Date: Fri, 22 Nov 2024 23:04:32 +0530 Subject: [PATCH 39/66] wifi: ath12k: Report proper tx completion status to mac80211 [ Upstream commit d2d9c9b8de725e1006d3aa3d18678a732f5d3584 ] Currently Tx completion for few exception packets are received from firmware and the tx status updated to mac80211. The tx status values of HAL_WBM_REL_HTT_TX_COMP_STATUS_DROP and HAL_WBM_REL_HTT_TX_COMP_STATUS_TTL are considered as tx failure and reported as tx failure to mac80211. But these failure status is due to internal firmware tx drop and these packets were not tried to transmit in the air. In case of mesh this invalid tx status report might trigger mpath broken issue due to increase in mpath fail average. So do not report these tx status as tx failure instead free the skb by calling ieee80211_free_txskb(), and that will be accounted as dropped frame. Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.3.1-00173-QCAHKSWPL_SILICONZ-1 Signed-off-by: Vinith Kumar R Signed-off-by: Tamizh Chelvam Raja Acked-by: Jeff Johnson Link: https://patch.msgid.link/20241122173432.2064858-1-quic_tamizhr@quicinc.com Signed-off-by: Jeff Johnson Signed-off-by: Sasha Levin --- drivers/net/wireless/ath/ath12k/dp_tx.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/net/wireless/ath/ath12k/dp_tx.c b/drivers/net/wireless/ath/ath12k/dp_tx.c index 44406e0b4a34..ad21fbfbcbe2 100644 --- a/drivers/net/wireless/ath/ath12k/dp_tx.c +++ b/drivers/net/wireless/ath/ath12k/dp_tx.c @@ -557,13 +557,13 @@ ath12k_dp_tx_process_htt_tx_complete(struct ath12k_base *ab, switch (wbm_status) { case HAL_WBM_REL_HTT_TX_COMP_STATUS_OK: - case HAL_WBM_REL_HTT_TX_COMP_STATUS_DROP: - case HAL_WBM_REL_HTT_TX_COMP_STATUS_TTL: ts.acked = (wbm_status == HAL_WBM_REL_HTT_TX_COMP_STATUS_OK); ts.ack_rssi = le32_get_bits(status_desc->info2, HTT_TX_WBM_COMP_INFO2_ACK_RSSI); ath12k_dp_tx_htt_tx_complete_buf(ab, msdu, tx_ring, &ts); break; + case HAL_WBM_REL_HTT_TX_COMP_STATUS_DROP: + case HAL_WBM_REL_HTT_TX_COMP_STATUS_TTL: case HAL_WBM_REL_HTT_TX_COMP_STATUS_REINJ: case HAL_WBM_REL_HTT_TX_COMP_STATUS_INSPECT: ath12k_dp_tx_free_txbuf(ab, msdu, mac_id, tx_ring); From 3ffaa2e999380477774e76680ff4cef247451168 Mon Sep 17 00:00:00 2001 From: Stanimir Varbanov Date: Mon, 24 Feb 2025 10:35:58 +0200 Subject: [PATCH 40/66] PCI: brcmstb: Expand inbound window size up to 64GB MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit [ Upstream commit 25a98c727015638baffcfa236e3f37b70cedcf87 ] The BCM2712 memory map can support up to 64GB of system memory, thus expand the inbound window size in calculation helper function. The change is safe for the currently supported SoCs that have smaller inbound window sizes. Signed-off-by: Stanimir Varbanov Reviewed-by: Florian Fainelli Reviewed-by: Jim Quinlan Tested-by: Ivan T. Ivanov Link: https://lore.kernel.org/r/20250224083559.47645-7-svarbanov@suse.de [kwilczynski: commit log] Signed-off-by: Krzysztof Wilczyński Signed-off-by: Sasha Levin --- drivers/pci/controller/pcie-brcmstb.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/pci/controller/pcie-brcmstb.c b/drivers/pci/controller/pcie-brcmstb.c index 792d24cea574..32ffb0c14c3c 100644 --- a/drivers/pci/controller/pcie-brcmstb.c +++ b/drivers/pci/controller/pcie-brcmstb.c @@ -309,8 +309,8 @@ static int brcm_pcie_encode_ibar_size(u64 size) if (log2_in >= 12 && log2_in <= 15) /* Covers 4KB to 32KB (inclusive) */ return (log2_in - 12) + 0x1c; - else if (log2_in >= 16 && log2_in <= 35) - /* Covers 64KB to 32GB, (inclusive) */ + else if (log2_in >= 16 && log2_in <= 36) + /* Covers 64KB to 64GB, (inclusive) */ return log2_in - 15; /* Something is awry so disable */ return 0; From 12153e3948c596737853c3ec4ff7e4e3f4a9f9a6 Mon Sep 17 00:00:00 2001 From: Stanimir Varbanov Date: Mon, 24 Feb 2025 10:35:56 +0200 Subject: [PATCH 41/66] PCI: brcmstb: Add a softdep to MIP MSI-X driver MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit [ Upstream commit 2294059118c550464dd8906286324d90c33b152b ] Then the brcmstb PCIe driver and MIP MSI-X interrupt controller drivers are built as modules there could be a race in probing. To avoid this, add a softdep to MIP driver to guarantee that MIP driver will be load first. Signed-off-by: Stanimir Varbanov Reviewed-by: Florian Fainelli Tested-by: Ivan T. Ivanov Link: https://lore.kernel.org/r/20250224083559.47645-5-svarbanov@suse.de [kwilczynski: commit log] Signed-off-by: Krzysztof Wilczyński Signed-off-by: Sasha Levin --- drivers/pci/controller/pcie-brcmstb.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/pci/controller/pcie-brcmstb.c b/drivers/pci/controller/pcie-brcmstb.c index 32ffb0c14c3c..81f085cebf62 100644 --- a/drivers/pci/controller/pcie-brcmstb.c +++ b/drivers/pci/controller/pcie-brcmstb.c @@ -1947,3 +1947,4 @@ module_platform_driver(brcm_pcie_driver); MODULE_LICENSE("GPL"); MODULE_DESCRIPTION("Broadcom STB PCIe RC driver"); MODULE_AUTHOR("Broadcom"); +MODULE_SOFTDEP("pre: irq_bcm2712_mip"); From 3a3efeef64364c2a028cf0d03d68c831813a97fd Mon Sep 17 00:00:00 2001 From: Viresh Kumar Date: Fri, 17 Jan 2025 15:35:52 +0530 Subject: [PATCH 42/66] firmware: arm_ffa: Set dma_mask for ffa devices [ Upstream commit cc0aac7ca17e0ea3ca84b552fc79f3e86fd07f53 ] Set dma_mask for FFA devices, otherwise DMA allocation using the device pointer lead to following warning: WARNING: CPU: 1 PID: 1 at kernel/dma/mapping.c:597 dma_alloc_attrs+0xe0/0x124 Signed-off-by: Viresh Kumar Message-Id: Signed-off-by: Sudeep Holla Signed-off-by: Sasha Levin --- drivers/firmware/arm_ffa/bus.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/firmware/arm_ffa/bus.c b/drivers/firmware/arm_ffa/bus.c index dfda5ffc14db..dea3eb741d95 100644 --- a/drivers/firmware/arm_ffa/bus.c +++ b/drivers/firmware/arm_ffa/bus.c @@ -212,6 +212,7 @@ ffa_device_register(const struct ffa_partition_info *part_info, dev = &ffa_dev->dev; dev->bus = &ffa_bus_type; dev->release = ffa_release_device; + dev->dma_mask = &dev->coherent_dma_mask; dev_set_name(&ffa_dev->dev, "arm-ffa-%d", id); ffa_dev->id = id; From df888ad55f8f8aaf22bbeafcee6bf7c27374bba3 Mon Sep 17 00:00:00 2001 From: Satyanarayana K V P Date: Mon, 24 Feb 2025 15:58:07 +0530 Subject: [PATCH 43/66] drm/xe/vf: Retry sending MMIO request to GUC on timeout error MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit [ Upstream commit ba757a65d2a28d46a8ccf50538f4f05036983f1b ] Add support to allow retrying the sending of MMIO requests from the VF to the GUC in the event of an error. During the suspend/resume process, VFs begin resuming only after the PF has resumed. Although the PF resumes, the GUC reset and provisioning occur later in a separate worker process. When there are a large number of VFs, some may attempt to resume before the PF has completed its provisioning. Therefore, if a MMIO request from a VF fails during this period, we will retry sending the request up to GUC_RESET_VF_STATE_RETRY_MAX times, which is set to a maximum of 10 attempts. Signed-off-by: Satyanarayana K V P Cc: Michał Wajdeczko Cc: Michał Winiarski Cc: Piotr Piórkowski Reviewed-by: Piotr Piorkowski Signed-off-by: Michal Wajdeczko Link: https://patchwork.freedesktop.org/patch/msgid/20250224102807.11065-3-satyanarayana.k.v.p@intel.com Signed-off-by: Sasha Levin --- drivers/gpu/drm/xe/xe_gt_sriov_vf.c | 9 ++++++++- 1 file changed, 8 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c index f982d6f9f218..7ddbfeaf494a 100644 --- a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c +++ b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c @@ -46,12 +46,19 @@ static int guc_action_vf_reset(struct xe_guc *guc) return ret > 0 ? -EPROTO : ret; } +#define GUC_RESET_VF_STATE_RETRY_MAX 10 static int vf_reset_guc_state(struct xe_gt *gt) { + unsigned int retry = GUC_RESET_VF_STATE_RETRY_MAX; struct xe_guc *guc = >->uc.guc; int err; - err = guc_action_vf_reset(guc); + do { + err = guc_action_vf_reset(guc); + if (!err || err != -ETIMEDOUT) + break; + } while (--retry); + if (unlikely(err)) xe_gt_sriov_err(gt, "Failed to reset GuC state (%pe)\n", ERR_PTR(err)); return err; From 1f512005f497dab8fb62218a3256ce1547ad9d1a Mon Sep 17 00:00:00 2001 From: Satyanarayana K V P Date: Mon, 24 Feb 2025 15:58:06 +0530 Subject: [PATCH 44/66] drm/xe/pf: Create a link between PF and VF devices MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit [ Upstream commit 8c0aff7d92e2be25717669eb65a81a89740a24f2 ] When both PF and VF devices are enabled on the host, they resume simultaneously during system resume. However, the PF must finish provisioning the VF before any VFs can successfully resume. Establish a parent-child device link between the PF and VF devices to ensure the correct order of resumption. V4 -> V5: - Added missing break in the error condition. V3 -> V4: - Made xe_pci_pf_get_vf_dev() as a static function and updated input parameter types. - Updated xe_sriov_warn() to xe_sriov_abort() when VF device cannot be found. V2 -> V3: - Added function documentation for xe_pci_pf_get_vf_dev(). - Added assertion if not called from PF. V1 -> V2: - Added a helper function to get VF pci_dev. - Updated xe_sriov_notice() to xe_sriov_warn() if vf pci_dev is not found. Signed-off-by: Satyanarayana K V P Cc: Michał Wajdeczko Cc: Michał Winiarski Cc: Piotr Piórkowski Reviewed-by: Piotr Piorkowski Signed-off-by: Michal Wajdeczko Link: https://patchwork.freedesktop.org/patch/msgid/20250224102807.11065-2-satyanarayana.k.v.p@intel.com Signed-off-by: Sasha Levin --- drivers/gpu/drm/xe/xe_pci_sriov.c | 51 +++++++++++++++++++++++++++++++ 1 file changed, 51 insertions(+) diff --git a/drivers/gpu/drm/xe/xe_pci_sriov.c b/drivers/gpu/drm/xe/xe_pci_sriov.c index aaceee748287..09ee8a06fe2e 100644 --- a/drivers/gpu/drm/xe/xe_pci_sriov.c +++ b/drivers/gpu/drm/xe/xe_pci_sriov.c @@ -62,6 +62,55 @@ static void pf_reset_vfs(struct xe_device *xe, unsigned int num_vfs) xe_gt_sriov_pf_control_trigger_flr(gt, n); } +static struct pci_dev *xe_pci_pf_get_vf_dev(struct xe_device *xe, unsigned int vf_id) +{ + struct pci_dev *pdev = to_pci_dev(xe->drm.dev); + + xe_assert(xe, IS_SRIOV_PF(xe)); + + /* caller must use pci_dev_put() */ + return pci_get_domain_bus_and_slot(pci_domain_nr(pdev->bus), + pdev->bus->number, + pci_iov_virtfn_devfn(pdev, vf_id)); +} + +static void pf_link_vfs(struct xe_device *xe, int num_vfs) +{ + struct pci_dev *pdev_pf = to_pci_dev(xe->drm.dev); + struct device_link *link; + struct pci_dev *pdev_vf; + unsigned int n; + + /* + * When both PF and VF devices are enabled on the host, during system + * resume they are resuming in parallel. + * + * But PF has to complete the provision of VF first to allow any VFs to + * successfully resume. + * + * Create a parent-child device link between PF and VF devices that will + * enforce correct resume order. + */ + for (n = 1; n <= num_vfs; n++) { + pdev_vf = xe_pci_pf_get_vf_dev(xe, n - 1); + + /* unlikely, something weird is happening, abort */ + if (!pdev_vf) { + xe_sriov_err(xe, "Cannot find VF%u device, aborting link%s creation!\n", + n, str_plural(num_vfs)); + break; + } + + link = device_link_add(&pdev_vf->dev, &pdev_pf->dev, + DL_FLAG_AUTOREMOVE_CONSUMER); + /* unlikely and harmless, continue with other VFs */ + if (!link) + xe_sriov_notice(xe, "Failed linking VF%u\n", n); + + pci_dev_put(pdev_vf); + } +} + static int pf_enable_vfs(struct xe_device *xe, int num_vfs) { struct pci_dev *pdev = to_pci_dev(xe->drm.dev); @@ -92,6 +141,8 @@ static int pf_enable_vfs(struct xe_device *xe, int num_vfs) if (err < 0) goto failed; + pf_link_vfs(xe, num_vfs); + xe_sriov_info(xe, "Enabled %u of %u VF%s\n", num_vfs, total_vfs, str_plural(total_vfs)); return num_vfs; From 11e721443c5647a2b1c005a598438890d8dcfd3b Mon Sep 17 00:00:00 2001 From: Moshe Shemesh Date: Wed, 26 Feb 2025 14:25:40 +0200 Subject: [PATCH 45/66] net/mlx5: Avoid report two health errors on same syndrome [ Upstream commit b5d7b2f04ebcff740f44ef4d295b3401aeb029f4 ] In case health counter has not increased for few polling intervals, miss counter will reach max misses threshold and health report will be triggered for FW health reporter. In case syndrome found on same health poll another health report will be triggered. Avoid two health reports on same syndrome by marking this syndrome as already known. Signed-off-by: Moshe Shemesh Reviewed-by: Shahar Shitrit Signed-off-by: Tariq Toukan Reviewed-by: Kalesh AP Signed-off-by: David S. Miller Signed-off-by: Sasha Levin --- drivers/net/ethernet/mellanox/mlx5/core/health.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/health.c b/drivers/net/ethernet/mellanox/mlx5/core/health.c index a6329ca2d9bf..52c8035547be 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/health.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/health.c @@ -799,6 +799,7 @@ static void poll_health(struct timer_list *t) health->prev = count; if (health->miss_counter == MAX_MISSES) { mlx5_core_err(dev, "device's health compromised - reached miss count\n"); + health->synd = ioread8(&h->synd); print_health_info(dev); queue_work(health->wq, &health->report_work); } From baaf3084c2cb92933c7ec29d584bccd1e9183a2a Mon Sep 17 00:00:00 2001 From: Kevin Krakauer Date: Wed, 26 Feb 2025 11:27:23 -0800 Subject: [PATCH 46/66] selftests/net: have `gro.sh -t` return a correct exit code [ Upstream commit 784e6abd99f24024a8998b5916795f0bec9d2fd9 ] Modify gro.sh to return a useful exit code when the -t flag is used. It formerly returned 0 no matter what. Tested: Ran `gro.sh -t large` and verified that test failures return 1. Signed-off-by: Kevin Krakauer Reviewed-by: Willem de Bruijn Link: https://patch.msgid.link/20250226192725.621969-2-krakauer@google.com Signed-off-by: Jakub Kicinski Signed-off-by: Sasha Levin --- tools/testing/selftests/net/gro.sh | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/tools/testing/selftests/net/gro.sh b/tools/testing/selftests/net/gro.sh index 02c21ff4ca81..aabd6e5480b8 100755 --- a/tools/testing/selftests/net/gro.sh +++ b/tools/testing/selftests/net/gro.sh @@ -100,5 +100,6 @@ trap cleanup EXIT if [[ "${test}" == "all" ]]; then run_all_tests else - run_test "${proto}" "${test}" + exit_code=$(run_test "${proto}" "${test}") + exit $exit_code fi; From 1dd943dfb56f86f65f624cad9fb1834a3cb717fe Mon Sep 17 00:00:00 2001 From: Inochi Amaoto Date: Tue, 11 Feb 2025 13:17:49 +0800 Subject: [PATCH 47/66] pinctrl: sophgo: avoid to modify untouched bit when setting cv1800 pinconf [ Upstream commit ef1a5121ae3da02372fcb66d9632ed3d47ad5637 ] When setting pinconf configuration for cv1800 SoC, the driver just writes the value. It may zero some bits of the pinconf register and cause some unexpected error. Add a mask to avoid this. Signed-off-by: Inochi Amaoto Link: https://lore.kernel.org/20250211051801.470800-2-inochiama@gmail.com Signed-off-by: Linus Walleij Signed-off-by: Sasha Levin --- drivers/pinctrl/sophgo/pinctrl-cv18xx.c | 33 +++++++++++++++++-------- 1 file changed, 23 insertions(+), 10 deletions(-) diff --git a/drivers/pinctrl/sophgo/pinctrl-cv18xx.c b/drivers/pinctrl/sophgo/pinctrl-cv18xx.c index 57f2674e75d6..84b4850771ce 100644 --- a/drivers/pinctrl/sophgo/pinctrl-cv18xx.c +++ b/drivers/pinctrl/sophgo/pinctrl-cv18xx.c @@ -574,10 +574,10 @@ static int cv1800_pinconf_compute_config(struct cv1800_pinctrl *pctrl, struct cv1800_pin *pin, unsigned long *configs, unsigned int num_configs, - u32 *value) + u32 *value, u32 *mask) { int i; - u32 v = 0; + u32 v = 0, m = 0; enum cv1800_pin_io_type type; int ret; @@ -596,10 +596,12 @@ static int cv1800_pinconf_compute_config(struct cv1800_pinctrl *pctrl, case PIN_CONFIG_BIAS_PULL_DOWN: v &= ~PIN_IO_PULLDOWN; v |= FIELD_PREP(PIN_IO_PULLDOWN, arg); + m |= PIN_IO_PULLDOWN; break; case PIN_CONFIG_BIAS_PULL_UP: v &= ~PIN_IO_PULLUP; v |= FIELD_PREP(PIN_IO_PULLUP, arg); + m |= PIN_IO_PULLUP; break; case PIN_CONFIG_DRIVE_STRENGTH_UA: ret = cv1800_pinctrl_oc2reg(pctrl, pin, arg); @@ -607,6 +609,7 @@ static int cv1800_pinconf_compute_config(struct cv1800_pinctrl *pctrl, return ret; v &= ~PIN_IO_DRIVE; v |= FIELD_PREP(PIN_IO_DRIVE, ret); + m |= PIN_IO_DRIVE; break; case PIN_CONFIG_INPUT_SCHMITT_UV: ret = cv1800_pinctrl_schmitt2reg(pctrl, pin, arg); @@ -614,6 +617,7 @@ static int cv1800_pinconf_compute_config(struct cv1800_pinctrl *pctrl, return ret; v &= ~PIN_IO_SCHMITT; v |= FIELD_PREP(PIN_IO_SCHMITT, ret); + m |= PIN_IO_SCHMITT; break; case PIN_CONFIG_POWER_SOURCE: /* Ignore power source as it is always fixed */ @@ -621,10 +625,12 @@ static int cv1800_pinconf_compute_config(struct cv1800_pinctrl *pctrl, case PIN_CONFIG_SLEW_RATE: v &= ~PIN_IO_OUT_FAST_SLEW; v |= FIELD_PREP(PIN_IO_OUT_FAST_SLEW, arg); + m |= PIN_IO_OUT_FAST_SLEW; break; case PIN_CONFIG_BIAS_BUS_HOLD: v &= ~PIN_IO_BUS_HOLD; v |= FIELD_PREP(PIN_IO_BUS_HOLD, arg); + m |= PIN_IO_BUS_HOLD; break; default: return -ENOTSUPP; @@ -632,17 +638,19 @@ static int cv1800_pinconf_compute_config(struct cv1800_pinctrl *pctrl, } *value = v; + *mask = m; return 0; } static int cv1800_pin_set_config(struct cv1800_pinctrl *pctrl, unsigned int pin_id, - u32 value) + u32 value, u32 mask) { struct cv1800_pin *pin = cv1800_get_pin(pctrl, pin_id); unsigned long flags; void __iomem *addr; + u32 reg; if (!pin) return -EINVAL; @@ -650,7 +658,10 @@ static int cv1800_pin_set_config(struct cv1800_pinctrl *pctrl, addr = cv1800_pinctrl_get_component_addr(pctrl, &pin->conf); raw_spin_lock_irqsave(&pctrl->lock, flags); - writel(value, addr); + reg = readl(addr); + reg &= ~mask; + reg |= value; + writel(reg, addr); raw_spin_unlock_irqrestore(&pctrl->lock, flags); return 0; @@ -662,16 +673,17 @@ static int cv1800_pconf_set(struct pinctrl_dev *pctldev, { struct cv1800_pinctrl *pctrl = pinctrl_dev_get_drvdata(pctldev); struct cv1800_pin *pin = cv1800_get_pin(pctrl, pin_id); - u32 value; + u32 value, mask; if (!pin) return -ENODEV; if (cv1800_pinconf_compute_config(pctrl, pin, - configs, num_configs, &value)) + configs, num_configs, + &value, &mask)) return -ENOTSUPP; - return cv1800_pin_set_config(pctrl, pin_id, value); + return cv1800_pin_set_config(pctrl, pin_id, value, mask); } static int cv1800_pconf_group_set(struct pinctrl_dev *pctldev, @@ -682,7 +694,7 @@ static int cv1800_pconf_group_set(struct pinctrl_dev *pctldev, struct cv1800_pinctrl *pctrl = pinctrl_dev_get_drvdata(pctldev); const struct group_desc *group; const struct cv1800_pin_mux_config *pinmuxs; - u32 value; + u32 value, mask; int i; group = pinctrl_generic_get_group(pctldev, gsel); @@ -692,11 +704,12 @@ static int cv1800_pconf_group_set(struct pinctrl_dev *pctldev, pinmuxs = group->data; if (cv1800_pinconf_compute_config(pctrl, pinmuxs[0].pin, - configs, num_configs, &value)) + configs, num_configs, + &value, &mask)) return -ENOTSUPP; for (i = 0; i < group->grp.npins; i++) - cv1800_pin_set_config(pctrl, group->grp.pins[i], value); + cv1800_pin_set_config(pctrl, group->grp.pins[i], value, mask); return 0; } From 0d8562e358176106d9a0a3c8932cf706119eccde Mon Sep 17 00:00:00 2001 From: Philip Yang Date: Mon, 17 Feb 2025 20:08:29 -0500 Subject: [PATCH 48/66] drm/amdkfd: KFD release_work possible circular locking [ Upstream commit 1b9366c601039d60546794c63fbb83ce8e53b978 ] If waiting for gpu reset done in KFD release_work, thers is WARNING: possible circular locking dependency detected #2 kfd_create_process kfd_process_mutex flush kfd release work #1 kfd release work wait for amdgpu reset work #0 amdgpu_device_gpu_reset kgd2kfd_pre_reset kfd_process_mutex Possible unsafe locking scenario: CPU0 CPU1 ---- ---- lock((work_completion)(&p->release_work)); lock((wq_completion)kfd_process_wq); lock((work_completion)(&p->release_work)); lock((wq_completion)amdgpu-reset-dev); To fix this, KFD create process move flush release work outside kfd_process_mutex. Signed-off-by: Philip Yang Reviewed-by: Felix Kuehling Signed-off-by: Alex Deucher Signed-off-by: Sasha Levin --- drivers/gpu/drm/amd/amdkfd/kfd_process.c | 16 ++++++++-------- 1 file changed, 8 insertions(+), 8 deletions(-) diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_process.c b/drivers/gpu/drm/amd/amdkfd/kfd_process.c index 0ec8b457494b..45923da7709f 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_process.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_process.c @@ -842,6 +842,14 @@ struct kfd_process *kfd_create_process(struct task_struct *thread) return ERR_PTR(-EINVAL); } + /* If the process just called exec(3), it is possible that the + * cleanup of the kfd_process (following the release of the mm + * of the old process image) is still in the cleanup work queue. + * Make sure to drain any job before trying to recreate any + * resource for this process. + */ + flush_workqueue(kfd_process_wq); + /* * take kfd processes mutex before starting of process creation * so there won't be a case where two threads of the same process @@ -860,14 +868,6 @@ struct kfd_process *kfd_create_process(struct task_struct *thread) if (process) { pr_debug("Process already found\n"); } else { - /* If the process just called exec(3), it is possible that the - * cleanup of the kfd_process (following the release of the mm - * of the old process image) is still in the cleanup work queue. - * Make sure to drain any job before trying to recreate any - * resource for this process. - */ - flush_workqueue(kfd_process_wq); - process = create_process(thread); if (IS_ERR(process)) goto out; From b8fcb1cdbfc0c876a443d34ae3fe5154d6642715 Mon Sep 17 00:00:00 2001 From: Daniel Gomez Date: Mon, 24 Feb 2025 07:23:13 +0100 Subject: [PATCH 49/66] drm/xe: xe_gen_wa_oob: replace program_invocation_short_name [ Upstream commit 89eb42b5539f6ae6a0cabcb39e5b6fcc83c106a1 ] program_invocation_short_name may not be available in other systems. Instead, replace it with the argv[0] to pass the executable name. Fixes build error when program_invocation_short_name is not available: drivers/gpu/drm/xe/xe_gen_wa_oob.c:34:3: error: use of undeclared identifier 'program_invocation_short_name' 34 | program_invocation_short_name); | ^ 1 error generated. Suggested-by: Masahiro Yamada Signed-off-by: Daniel Gomez Reviewed-by: Lucas De Marchi Link: https://patchwork.freedesktop.org/patch/msgid/20250224-macos-build-support-xe-v3-1-d2c9ed3a27cc@samsung.com Signed-off-by: Lucas De Marchi Signed-off-by: Sasha Levin --- drivers/gpu/drm/xe/xe_gen_wa_oob.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers/gpu/drm/xe/xe_gen_wa_oob.c b/drivers/gpu/drm/xe/xe_gen_wa_oob.c index 904cf47925aa..ed9183599e31 100644 --- a/drivers/gpu/drm/xe/xe_gen_wa_oob.c +++ b/drivers/gpu/drm/xe/xe_gen_wa_oob.c @@ -28,10 +28,10 @@ "\n" \ "#endif\n" -static void print_usage(FILE *f) +static void print_usage(FILE *f, const char *progname) { fprintf(f, "usage: %s \n", - program_invocation_short_name); + progname); } static void print_parse_error(const char *err_msg, const char *line, @@ -144,7 +144,7 @@ int main(int argc, const char *argv[]) if (argc < 3) { fprintf(stderr, "ERROR: wrong arguments\n"); - print_usage(stderr); + print_usage(stderr, argv[0]); return 1; } From 562b4f70bf88add47ef146e02bf972b9a8ab370d Mon Sep 17 00:00:00 2001 From: Yuanjun Gong Date: Sun, 23 Feb 2025 20:14:59 +0800 Subject: [PATCH 50/66] leds: pwm-multicolor: Add check for fwnode_property_read_u32 [ Upstream commit 6d91124e7edc109f114b1afe6d00d85d0d0ac174 ] Add a check to the return value of fwnode_property_read_u32() in case it fails. Signed-off-by: Yuanjun Gong Link: https://lore.kernel.org/r/20250223121459.2889484-1-ruc_gongyuanjun@163.com Signed-off-by: Lee Jones Signed-off-by: Sasha Levin --- drivers/leds/rgb/leds-pwm-multicolor.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/drivers/leds/rgb/leds-pwm-multicolor.c b/drivers/leds/rgb/leds-pwm-multicolor.c index e1a81e0109e8..c0aa34b1d0e2 100644 --- a/drivers/leds/rgb/leds-pwm-multicolor.c +++ b/drivers/leds/rgb/leds-pwm-multicolor.c @@ -135,8 +135,11 @@ static int led_pwm_mc_probe(struct platform_device *pdev) /* init the multicolor's LED class device */ cdev = &priv->mc_cdev.led_cdev; - fwnode_property_read_u32(mcnode, "max-brightness", + ret = fwnode_property_read_u32(mcnode, "max-brightness", &cdev->max_brightness); + if (ret) + goto release_mcnode; + cdev->flags = LED_CORE_SUSPENDRESUME; cdev->brightness_set_blocking = led_pwm_mc_set; From 459b3f7cf0dcfa7f4fa1b8ccab8a176e96ebcf48 Mon Sep 17 00:00:00 2001 From: Eric Woudstra Date: Tue, 25 Feb 2025 21:15:09 +0100 Subject: [PATCH 51/66] net: ethernet: mtk_ppe_offload: Allow QinQ, double ETH_P_8021Q only [ Upstream commit 7fe0353606d77a32c4c7f2814833dd1c043ebdd2 ] mtk_foe_entry_set_vlan() in mtk_ppe.c already supports double vlan tagging, but mtk_flow_offload_replace() in mtk_ppe_offload.c only allows for 1 vlan tag, optionally in combination with pppoe and dsa tags. However, mtk_foe_entry_set_vlan() only allows for setting the vlan id. The protocol cannot be set, it is always ETH_P_8021Q, for inner and outer tag. This patch adds QinQ support to mtk_flow_offload_replace(), only in the case that both inner and outer tags are ETH_P_8021Q. Only PPPoE-in-Q (as before) and Q-in-Q are allowed. A combination of PPPoE and Q-in-Q is not allowed. Signed-off-by: Eric Woudstra Link: https://patch.msgid.link/20250225201509.20843-1-ericwouds@gmail.com Signed-off-by: Paolo Abeni Signed-off-by: Sasha Levin --- .../net/ethernet/mediatek/mtk_ppe_offload.c | 22 +++++++++---------- 1 file changed, 11 insertions(+), 11 deletions(-) diff --git a/drivers/net/ethernet/mediatek/mtk_ppe_offload.c b/drivers/net/ethernet/mediatek/mtk_ppe_offload.c index f20bb390df3a..c855fb799ce1 100644 --- a/drivers/net/ethernet/mediatek/mtk_ppe_offload.c +++ b/drivers/net/ethernet/mediatek/mtk_ppe_offload.c @@ -34,8 +34,10 @@ struct mtk_flow_data { u16 vlan_in; struct { - u16 id; - __be16 proto; + struct { + u16 id; + __be16 proto; + } vlans[2]; u8 num; } vlan; struct { @@ -349,18 +351,19 @@ mtk_flow_offload_replace(struct mtk_eth *eth, struct flow_cls_offload *f, case FLOW_ACTION_CSUM: break; case FLOW_ACTION_VLAN_PUSH: - if (data.vlan.num == 1 || + if (data.vlan.num + data.pppoe.num == 2 || act->vlan.proto != htons(ETH_P_8021Q)) return -EOPNOTSUPP; - data.vlan.id = act->vlan.vid; - data.vlan.proto = act->vlan.proto; + data.vlan.vlans[data.vlan.num].id = act->vlan.vid; + data.vlan.vlans[data.vlan.num].proto = act->vlan.proto; data.vlan.num++; break; case FLOW_ACTION_VLAN_POP: break; case FLOW_ACTION_PPPOE_PUSH: - if (data.pppoe.num == 1) + if (data.pppoe.num == 1 || + data.vlan.num == 2) return -EOPNOTSUPP; data.pppoe.sid = act->pppoe.sid; @@ -450,12 +453,9 @@ mtk_flow_offload_replace(struct mtk_eth *eth, struct flow_cls_offload *f, if (offload_type == MTK_PPE_PKT_TYPE_BRIDGE) foe.bridge.vlan = data.vlan_in; - if (data.vlan.num == 1) { - if (data.vlan.proto != htons(ETH_P_8021Q)) - return -EOPNOTSUPP; + for (i = 0; i < data.vlan.num; i++) + mtk_foe_entry_set_vlan(eth, &foe, data.vlan.vlans[i].id); - mtk_foe_entry_set_vlan(eth, &foe, data.vlan.id); - } if (data.pppoe.num == 1) mtk_foe_entry_set_pppoe(eth, &foe, data.pppoe.sid); From f872f7aaa4cac5f4c1dcae4a0489374735ff1048 Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Tue, 25 Feb 2025 17:33:33 +0100 Subject: [PATCH 52/66] net: xgene-v2: remove incorrect ACPI_PTR annotation [ Upstream commit 01358e8fe922f716c05d7864ac2213b2440026e7 ] Building with W=1 shows a warning about xge_acpi_match being unused when CONFIG_ACPI is disabled: drivers/net/ethernet/apm/xgene-v2/main.c:723:36: error: unused variable 'xge_acpi_match' [-Werror,-Wunused-const-variable] Signed-off-by: Arnd Bergmann Link: https://patch.msgid.link/20250225163341.4168238-2-arnd@kernel.org Signed-off-by: Paolo Abeni Signed-off-by: Sasha Levin --- drivers/net/ethernet/apm/xgene-v2/main.c | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) diff --git a/drivers/net/ethernet/apm/xgene-v2/main.c b/drivers/net/ethernet/apm/xgene-v2/main.c index 9e90c2381491..68335935cea7 100644 --- a/drivers/net/ethernet/apm/xgene-v2/main.c +++ b/drivers/net/ethernet/apm/xgene-v2/main.c @@ -9,8 +9,6 @@ #include "main.h" -static const struct acpi_device_id xge_acpi_match[]; - static int xge_get_resources(struct xge_pdata *pdata) { struct platform_device *pdev; @@ -731,7 +729,7 @@ MODULE_DEVICE_TABLE(acpi, xge_acpi_match); static struct platform_driver xge_driver = { .driver = { .name = "xgene-enet-v2", - .acpi_match_table = ACPI_PTR(xge_acpi_match), + .acpi_match_table = xge_acpi_match, }, .probe = xge_probe, .remove_new = xge_remove, From e39fd41117bd94b2e992f857cd77cf07972566e6 Mon Sep 17 00:00:00 2001 From: Hangbin Liu Date: Tue, 25 Feb 2025 03:39:14 +0000 Subject: [PATCH 53/66] bonding: report duplicate MAC address in all situations MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit [ Upstream commit 28d68d396a1cd21591e8c6d74afbde33a7ea107e ] Normally, a bond uses the MAC address of the first added slave as the bond’s MAC address. And the bond will set active slave’s MAC address to bond’s address if fail_over_mac is set to none (0) or follow (2). When the first slave is removed, the bond will still use the removed slave’s MAC address, which can lead to a duplicate MAC address and potentially cause issues with the switch. To avoid confusion, let's warn the user in all situations, including when fail_over_mac is set to 2 or not in active-backup mode. Signed-off-by: Hangbin Liu Reviewed-by: Nikolay Aleksandrov Link: https://patch.msgid.link/20250225033914.18617-1-liuhangbin@gmail.com Signed-off-by: Jakub Kicinski Signed-off-by: Sasha Levin --- drivers/net/bonding/bond_main.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c index 4d73abae503d..4d2e30f4ee25 100644 --- a/drivers/net/bonding/bond_main.c +++ b/drivers/net/bonding/bond_main.c @@ -2542,7 +2542,7 @@ static int __bond_release_one(struct net_device *bond_dev, RCU_INIT_POINTER(bond->current_arp_slave, NULL); - if (!all && (!bond->params.fail_over_mac || + if (!all && (bond->params.fail_over_mac != BOND_FOM_ACTIVE || BOND_MODE(bond) != BOND_MODE_ACTIVEBACKUP)) { if (ether_addr_equal_64bits(bond_dev->dev_addr, slave->perm_hwaddr) && bond_has_slaves(bond)) From 54ef1a89e287ee81382358889a1e8b9637879af4 Mon Sep 17 00:00:00 2001 From: Ramasamy Kaliappan Date: Fri, 7 Feb 2025 11:30:05 +0530 Subject: [PATCH 54/66] wifi: ath12k: Improve BSS discovery with hidden SSID in 6 GHz band [ Upstream commit 27d38bdfd416f4db70e09c3bef3b030c86fd235a ] Currently, sometimes, the station is unable to identify the configured AP SSID in its scan results when the AP is not broadcasting its name publicly and has a hidden SSID. Currently, channel dwell time for an ath12k station is 30 ms. Sometimes, station can send broadcast probe request to AP close to the end of dwell time. In some of these cases, before AP sends a response to the received probe request, the dwell time on the station side would come to an end. So, the station will move to scan next channel and will not be able to acknowledge the unicast probe response. Resolve this issue by increasing station's channel dwell time to 70 ms, so that the it remains on the same channel for a longer period. This would increase the station's chance of receiving probe response from the AP. The station will then send a response acknowledgment back to the AP, thus leading to successful scan and BSS discovery. With an increased dwell time, scan would take longer than it takes now. But, this fix is an improvement for hidden SSID scan issue. Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.4.1-00199-QCAHKSWPL_SILICONZ-1 Signed-off-by: Ramasamy Kaliappan Signed-off-by: Roopni Devanathan Reviewed-by: Vasanthakumar Thiagarajan Link: https://patch.msgid.link/20250207060005.153835-1-quic_rdevanat@quicinc.com Signed-off-by: Jeff Johnson Signed-off-by: Sasha Levin --- drivers/net/wireless/ath/ath12k/wmi.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/net/wireless/ath/ath12k/wmi.c b/drivers/net/wireless/ath/ath12k/wmi.c index a6ba97949440..30836a09d550 100644 --- a/drivers/net/wireless/ath/ath12k/wmi.c +++ b/drivers/net/wireless/ath/ath12k/wmi.c @@ -2206,8 +2206,8 @@ void ath12k_wmi_start_scan_init(struct ath12k *ar, arg->dwell_time_active = 50; arg->dwell_time_active_2g = 0; arg->dwell_time_passive = 150; - arg->dwell_time_active_6g = 40; - arg->dwell_time_passive_6g = 30; + arg->dwell_time_active_6g = 70; + arg->dwell_time_passive_6g = 70; arg->min_rest_time = 50; arg->max_rest_time = 500; arg->repeat_probe_time = 0; From 06100e642f4b83ec758744e5f992d7a2039ec98f Mon Sep 17 00:00:00 2001 From: Andrew Davis Date: Thu, 23 Jan 2025 12:17:26 -0600 Subject: [PATCH 55/66] soc: ti: k3-socinfo: Do not use syscon helper to build regmap [ Upstream commit a5caf03188e44388e8c618dcbe5fffad1a249385 ] The syscon helper device_node_to_regmap() is used to fetch a regmap registered to a device node. It also currently creates this regmap if the node did not already have a regmap associated with it. This should only be used on "syscon" nodes. This driver is not such a device and instead uses device_node_to_regmap() on its own node as a hacky way to create a regmap for itself. This will not work going forward and so we should create our regmap the normal way by defining our regmap_config, fetching our memory resource, then using the normal regmap_init_mmio() function. Signed-off-by: Andrew Davis Link: https://lore.kernel.org/r/20250123181726.597144-1-afd@ti.com Signed-off-by: Nishanth Menon Signed-off-by: Sasha Levin --- drivers/soc/ti/k3-socinfo.c | 13 ++++++++++++- 1 file changed, 12 insertions(+), 1 deletion(-) diff --git a/drivers/soc/ti/k3-socinfo.c b/drivers/soc/ti/k3-socinfo.c index 4fb0f0a24828..704039eb3c07 100644 --- a/drivers/soc/ti/k3-socinfo.c +++ b/drivers/soc/ti/k3-socinfo.c @@ -105,6 +105,12 @@ err_unknown_variant: return -ENODEV; } +static const struct regmap_config k3_chipinfo_regmap_cfg = { + .reg_bits = 32, + .val_bits = 32, + .reg_stride = 4, +}; + static int k3_chipinfo_probe(struct platform_device *pdev) { struct device_node *node = pdev->dev.of_node; @@ -112,13 +118,18 @@ static int k3_chipinfo_probe(struct platform_device *pdev) struct device *dev = &pdev->dev; struct soc_device *soc_dev; struct regmap *regmap; + void __iomem *base; u32 partno_id; u32 variant; u32 jtag_id; u32 mfg; int ret; - regmap = device_node_to_regmap(node); + base = devm_platform_ioremap_resource(pdev, 0); + if (IS_ERR(base)) + return PTR_ERR(base); + + regmap = regmap_init_mmio(dev, base, &k3_chipinfo_regmap_cfg); if (IS_ERR(regmap)) return PTR_ERR(regmap); From 673dde8d3c3ec97d10312caced2143194873de6f Mon Sep 17 00:00:00 2001 From: Amery Hung Date: Tue, 25 Feb 2025 15:35:44 -0800 Subject: [PATCH 56/66] bpf: Search and add kfuncs in struct_ops prologue and epilogue [ Upstream commit d519594ee2445d7cd1ad51f4db4cee58f8213400 ] Currently, add_kfunc_call() is only invoked once before the main verification loop. Therefore, the verifier could not find the bpf_kfunc_btf_tab of a new kfunc call which is not seen in user defined struct_ops operators but introduced in gen_prologue or gen_epilogue during do_misc_fixup(). Fix this by searching kfuncs in the patching instruction buffer and add them to prog->aux->kfunc_tab. Signed-off-by: Amery Hung Acked-by: Eduard Zingerman Acked-by: Martin KaFai Lau Link: https://lore.kernel.org/r/20250225233545.285481-1-ameryhung@gmail.com Signed-off-by: Alexei Starovoitov Signed-off-by: Sasha Levin --- kernel/bpf/verifier.c | 25 ++++++++++++++++++++++++- 1 file changed, 24 insertions(+), 1 deletion(-) diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c index 8656208aa4bb..294fbafbeba7 100644 --- a/kernel/bpf/verifier.c +++ b/kernel/bpf/verifier.c @@ -2987,6 +2987,21 @@ bpf_jit_find_kfunc_model(const struct bpf_prog *prog, return res ? &res->func_model : NULL; } +static int add_kfunc_in_insns(struct bpf_verifier_env *env, + struct bpf_insn *insn, int cnt) +{ + int i, ret; + + for (i = 0; i < cnt; i++, insn++) { + if (bpf_pseudo_kfunc_call(insn)) { + ret = add_kfunc_call(env, insn->imm, insn->off); + if (ret < 0) + return ret; + } + } + return 0; +} + static int add_subprog_and_kfunc(struct bpf_verifier_env *env) { struct bpf_subprog_info *subprog = env->subprog_info; @@ -19768,7 +19783,7 @@ static int convert_ctx_accesses(struct bpf_verifier_env *env) { struct bpf_subprog_info *subprogs = env->subprog_info; const struct bpf_verifier_ops *ops = env->ops; - int i, cnt, size, ctx_field_size, delta = 0, epilogue_cnt = 0; + int i, cnt, size, ctx_field_size, ret, delta = 0, epilogue_cnt = 0; const int insn_cnt = env->prog->len; struct bpf_insn *epilogue_buf = env->epilogue_buf; struct bpf_insn *insn_buf = env->insn_buf; @@ -19797,6 +19812,10 @@ static int convert_ctx_accesses(struct bpf_verifier_env *env) return -ENOMEM; env->prog = new_prog; delta += cnt - 1; + + ret = add_kfunc_in_insns(env, epilogue_buf, epilogue_cnt - 1); + if (ret < 0) + return ret; } } @@ -19817,6 +19836,10 @@ static int convert_ctx_accesses(struct bpf_verifier_env *env) env->prog = new_prog; delta += cnt - 1; + + ret = add_kfunc_in_insns(env, insn_buf, cnt - 1); + if (ret < 0) + return ret; } } From de48b82dcf583563227ea84606ffd1d4eab44404 Mon Sep 17 00:00:00 2001 From: Hariprasad Kelam Date: Mon, 24 Feb 2025 09:26:03 +0530 Subject: [PATCH 57/66] Octeontx2-af: RPM: Register driver with PCI subsys IDs [ Upstream commit fc9167192f29485be5621e2e9c8208b717b65753 ] Although the PCI device ID and Vendor ID for the RPM (MAC) block have remained the same across Octeon CN10K and the next-generation CN20K silicon, Hardware architecture has changed (NIX mapped RPMs and RFOE Mapped RPMs). Add PCI Subsystem IDs to the device table to ensure that this driver can be probed from NIX mapped RPM devices only. Signed-off-by: Hariprasad Kelam Link: https://patch.msgid.link/20250224035603.1220913-1-hkelam@marvell.com Signed-off-by: Jakub Kicinski Signed-off-by: Sasha Levin --- drivers/net/ethernet/marvell/octeontx2/af/cgx.c | 14 ++++++++++++-- drivers/net/ethernet/marvell/octeontx2/af/rvu.h | 2 ++ 2 files changed, 14 insertions(+), 2 deletions(-) diff --git a/drivers/net/ethernet/marvell/octeontx2/af/cgx.c b/drivers/net/ethernet/marvell/octeontx2/af/cgx.c index e43c4608d3ba..971993586fb4 100644 --- a/drivers/net/ethernet/marvell/octeontx2/af/cgx.c +++ b/drivers/net/ethernet/marvell/octeontx2/af/cgx.c @@ -66,8 +66,18 @@ static int cgx_fwi_link_change(struct cgx *cgx, int lmac_id, bool en); /* Supported devices */ static const struct pci_device_id cgx_id_table[] = { { PCI_DEVICE(PCI_VENDOR_ID_CAVIUM, PCI_DEVID_OCTEONTX2_CGX) }, - { PCI_DEVICE(PCI_VENDOR_ID_CAVIUM, PCI_DEVID_CN10K_RPM) }, - { PCI_DEVICE(PCI_VENDOR_ID_CAVIUM, PCI_DEVID_CN10KB_RPM) }, + { PCI_DEVICE_SUB(PCI_VENDOR_ID_CAVIUM, PCI_DEVID_CN10K_RPM, + PCI_ANY_ID, PCI_SUBSYS_DEVID_CN10K_A) }, + { PCI_DEVICE_SUB(PCI_VENDOR_ID_CAVIUM, PCI_DEVID_CN10K_RPM, + PCI_ANY_ID, PCI_SUBSYS_DEVID_CNF10K_A) }, + { PCI_DEVICE_SUB(PCI_VENDOR_ID_CAVIUM, PCI_DEVID_CN10K_RPM, + PCI_ANY_ID, PCI_SUBSYS_DEVID_CNF10K_B) }, + { PCI_DEVICE_SUB(PCI_VENDOR_ID_CAVIUM, PCI_DEVID_CN10KB_RPM, + PCI_ANY_ID, PCI_SUBSYS_DEVID_CN10K_B) }, + { PCI_DEVICE_SUB(PCI_VENDOR_ID_CAVIUM, PCI_DEVID_CN10KB_RPM, + PCI_ANY_ID, PCI_SUBSYS_DEVID_CN20KA) }, + { PCI_DEVICE_SUB(PCI_VENDOR_ID_CAVIUM, PCI_DEVID_CN10KB_RPM, + PCI_ANY_ID, PCI_SUBSYS_DEVID_CNF20KA) }, { 0, } /* end of table */ }; diff --git a/drivers/net/ethernet/marvell/octeontx2/af/rvu.h b/drivers/net/ethernet/marvell/octeontx2/af/rvu.h index 8555edbb1c8f..f94bf04788e9 100644 --- a/drivers/net/ethernet/marvell/octeontx2/af/rvu.h +++ b/drivers/net/ethernet/marvell/octeontx2/af/rvu.h @@ -30,6 +30,8 @@ #define PCI_SUBSYS_DEVID_CNF10K_A 0xBA00 #define PCI_SUBSYS_DEVID_CNF10K_B 0xBC00 #define PCI_SUBSYS_DEVID_CN10K_B 0xBD00 +#define PCI_SUBSYS_DEVID_CN20KA 0xC220 +#define PCI_SUBSYS_DEVID_CNF20KA 0xC320 /* PCI BAR nos */ #define PCI_AF_REG_BAR_NUM 0 From 5cab1de0cea2840a8a7d061ae13413fb77ff96c0 Mon Sep 17 00:00:00 2001 From: Nir Lichtman Date: Fri, 10 Jan 2025 12:05:00 +0000 Subject: [PATCH 58/66] x86/build: Fix broken copy command in genimage.sh when making isoimage [ Upstream commit e451630226bd09dc730eedb4e32cab1cc7155ae8 ] Problem: Currently when running the "make isoimage" command there is an error related to wrong parameters passed to the cp command: "cp: missing destination file operand after 'arch/x86/boot/isoimage/'" This is caused because FDINITRDS is an empty array. Solution: Check if FDINITRDS is empty before executing the "cp" command, similar to how it is done in the case of hdimage. Signed-off-by: Nir Lichtman Signed-off-by: Ingo Molnar Cc: "H. Peter Anvin" Cc: Ard Biesheuvel Cc: Masahiro Yamada Cc: Michal Marek Link: https://lore.kernel.org/r/20250110120500.GA923218@lichtman.org Signed-off-by: Sasha Levin --- arch/x86/boot/genimage.sh | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/arch/x86/boot/genimage.sh b/arch/x86/boot/genimage.sh index c9299aeb7333..3882ead513f7 100644 --- a/arch/x86/boot/genimage.sh +++ b/arch/x86/boot/genimage.sh @@ -22,6 +22,7 @@ # This script requires: # bash # syslinux +# genisoimage # mtools (for fdimage* and hdimage) # edk2/OVMF (for hdimage) # @@ -251,7 +252,9 @@ geniso() { cp "$isolinux" "$ldlinux" "$tmp_dir" cp "$FBZIMAGE" "$tmp_dir"/linux echo default linux "$KCMDLINE" > "$tmp_dir"/isolinux.cfg - cp "${FDINITRDS[@]}" "$tmp_dir"/ + if [ ${#FDINITRDS[@]} -gt 0 ]; then + cp "${FDINITRDS[@]}" "$tmp_dir"/ + fi genisoimage -J -r -appid 'LINUX_BOOT' -input-charset=utf-8 \ -quiet -o "$FIMAGE" -b isolinux.bin \ -c boot.cat -no-emul-boot -boot-load-size 4 \ From 54dd746ed8a30175444076c9b4fdd15cf86e0842 Mon Sep 17 00:00:00 2001 From: Yihan Zhu Date: Wed, 12 Feb 2025 15:17:56 -0500 Subject: [PATCH 59/66] drm/amd/display: handle max_downscale_src_width fail check [ Upstream commit 02a940da2ccc0cc0299811379580852b405a0ea2 ] [WHY] If max_downscale_src_width check fails, we exit early from TAP calculation and left a NULL value to the scaling data structure to cause the zero divide in the DML validation. [HOW] Call set default TAP calculation before early exit in get_optimal_number_of_taps due to max downscale limit exceed. Reviewed-by: Samson Tam Signed-off-by: Yihan Zhu Signed-off-by: Zaeem Mohamed Tested-by: Daniel Wheeler Signed-off-by: Alex Deucher Signed-off-by: Sasha Levin --- drivers/gpu/drm/amd/display/dc/dpp/dcn30/dcn30_dpp.c | 11 ++++++----- 1 file changed, 6 insertions(+), 5 deletions(-) diff --git a/drivers/gpu/drm/amd/display/dc/dpp/dcn30/dcn30_dpp.c b/drivers/gpu/drm/amd/display/dc/dpp/dcn30/dcn30_dpp.c index 40acebd13e46..abf439e743f2 100644 --- a/drivers/gpu/drm/amd/display/dc/dpp/dcn30/dcn30_dpp.c +++ b/drivers/gpu/drm/amd/display/dc/dpp/dcn30/dcn30_dpp.c @@ -425,11 +425,6 @@ bool dpp3_get_optimal_number_of_taps( int min_taps_y, min_taps_c; enum lb_memory_config lb_config; - if (scl_data->viewport.width > scl_data->h_active && - dpp->ctx->dc->debug.max_downscale_src_width != 0 && - scl_data->viewport.width > dpp->ctx->dc->debug.max_downscale_src_width) - return false; - /* * Set default taps if none are provided * From programming guide: taps = min{ ceil(2*H_RATIO,1), 8} for downscaling @@ -467,6 +462,12 @@ bool dpp3_get_optimal_number_of_taps( else scl_data->taps.h_taps_c = in_taps->h_taps_c; + // Avoid null data in the scl data with this early return, proceed non-adaptive calcualtion first + if (scl_data->viewport.width > scl_data->h_active && + dpp->ctx->dc->debug.max_downscale_src_width != 0 && + scl_data->viewport.width > dpp->ctx->dc->debug.max_downscale_src_width) + return false; + /*Ensure we can support the requested number of vtaps*/ min_taps_y = dc_fixpt_ceil(scl_data->ratios.vert); min_taps_c = dc_fixpt_ceil(scl_data->ratios.vert_c); From 990e4fb37bcd49161bbc5331a24b73616aefafbf Mon Sep 17 00:00:00 2001 From: Charlene Liu Date: Thu, 13 Feb 2025 12:37:10 -0500 Subject: [PATCH 60/66] drm/amd/display: fix dcn4x init failed [ Upstream commit 23ef388a84c72b0614a6c10f866ffeac7e807719 ] [why] failed due to cmdtable not created. switch atombios cmdtable as default. Reviewed-by: Alvin Lee Signed-off-by: Charlene Liu Signed-off-by: Zaeem Mohamed Tested-by: Daniel Wheeler Signed-off-by: Alex Deucher Signed-off-by: Sasha Levin --- drivers/gpu/drm/amd/display/dc/bios/command_table2.c | 9 --------- .../gpu/drm/amd/display/dc/bios/command_table_helper2.c | 3 +-- 2 files changed, 1 insertion(+), 11 deletions(-) diff --git a/drivers/gpu/drm/amd/display/dc/bios/command_table2.c b/drivers/gpu/drm/amd/display/dc/bios/command_table2.c index 7d18f372ce7a..6bc59b7ef007 100644 --- a/drivers/gpu/drm/amd/display/dc/bios/command_table2.c +++ b/drivers/gpu/drm/amd/display/dc/bios/command_table2.c @@ -101,7 +101,6 @@ static void init_dig_encoder_control(struct bios_parser *bp) bp->cmd_tbl.dig_encoder_control = encoder_control_digx_v1_5; break; default: - dm_output_to_console("Don't have dig_encoder_control for v%d\n", version); bp->cmd_tbl.dig_encoder_control = encoder_control_fallback; break; } @@ -238,7 +237,6 @@ static void init_transmitter_control(struct bios_parser *bp) bp->cmd_tbl.transmitter_control = transmitter_control_v1_7; break; default: - dm_output_to_console("Don't have transmitter_control for v%d\n", crev); bp->cmd_tbl.transmitter_control = transmitter_control_fallback; break; } @@ -408,8 +406,6 @@ static void init_set_pixel_clock(struct bios_parser *bp) bp->cmd_tbl.set_pixel_clock = set_pixel_clock_v7; break; default: - dm_output_to_console("Don't have set_pixel_clock for v%d\n", - BIOS_CMD_TABLE_PARA_REVISION(setpixelclock)); bp->cmd_tbl.set_pixel_clock = set_pixel_clock_fallback; break; } @@ -554,7 +550,6 @@ static void init_set_crtc_timing(struct bios_parser *bp) set_crtc_using_dtd_timing_v3; break; default: - dm_output_to_console("Don't have set_crtc_timing for v%d\n", dtd_version); bp->cmd_tbl.set_crtc_timing = NULL; break; } @@ -671,8 +666,6 @@ static void init_enable_crtc(struct bios_parser *bp) bp->cmd_tbl.enable_crtc = enable_crtc_v1; break; default: - dm_output_to_console("Don't have enable_crtc for v%d\n", - BIOS_CMD_TABLE_PARA_REVISION(enablecrtc)); bp->cmd_tbl.enable_crtc = NULL; break; } @@ -864,8 +857,6 @@ static void init_set_dce_clock(struct bios_parser *bp) bp->cmd_tbl.set_dce_clock = set_dce_clock_v2_1; break; default: - dm_output_to_console("Don't have set_dce_clock for v%d\n", - BIOS_CMD_TABLE_PARA_REVISION(setdceclock)); bp->cmd_tbl.set_dce_clock = NULL; break; } diff --git a/drivers/gpu/drm/amd/display/dc/bios/command_table_helper2.c b/drivers/gpu/drm/amd/display/dc/bios/command_table_helper2.c index 73458e295103..df8139bda142 100644 --- a/drivers/gpu/drm/amd/display/dc/bios/command_table_helper2.c +++ b/drivers/gpu/drm/amd/display/dc/bios/command_table_helper2.c @@ -87,8 +87,7 @@ bool dal_bios_parser_init_cmd_tbl_helper2( return true; default: - /* Unsupported DCE */ - BREAK_TO_DEBUGGER(); + *h = dal_cmd_tbl_helper_dce112_get_table2(); return false; } } From 6e5935b755b4a81ccebde4eed0ebd37f5875ddb3 Mon Sep 17 00:00:00 2001 From: "Assadian, Navid" Date: Thu, 19 Dec 2024 17:19:09 -0500 Subject: [PATCH 61/66] drm/amd/display: Fix mismatch type comparison [ Upstream commit 26873260d394b1e33cdd720154aedf0af95327f9 ] The mismatch type comparison/assignment may cause data loss. Since the values are always non-negative, it is safe to use unsigned variables to resolve the mismatch. Signed-off-by: Navid Assadian Reviewed-by: Joshua Aberback Tested-by: Daniel Wheeler Signed-off-by: Alex Deucher Signed-off-by: Sasha Levin --- drivers/gpu/drm/amd/display/dc/spl/dc_spl.c | 4 ++-- drivers/gpu/drm/amd/display/dc/spl/dc_spl_types.h | 2 +- 2 files changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers/gpu/drm/amd/display/dc/spl/dc_spl.c b/drivers/gpu/drm/amd/display/dc/spl/dc_spl.c index 014e8a296f0c..54c7d6aecf51 100644 --- a/drivers/gpu/drm/amd/display/dc/spl/dc_spl.c +++ b/drivers/gpu/drm/amd/display/dc/spl/dc_spl.c @@ -875,8 +875,8 @@ static bool spl_get_optimal_number_of_taps( bool *enable_isharp) { int num_part_y, num_part_c; - int max_taps_y, max_taps_c; - int min_taps_y, min_taps_c; + unsigned int max_taps_y, max_taps_c; + unsigned int min_taps_y, min_taps_c; enum lb_memory_config lb_config; bool skip_easf = false; diff --git a/drivers/gpu/drm/amd/display/dc/spl/dc_spl_types.h b/drivers/gpu/drm/amd/display/dc/spl/dc_spl_types.h index 2a74ff5fdfdb..a2c28949ec47 100644 --- a/drivers/gpu/drm/amd/display/dc/spl/dc_spl_types.h +++ b/drivers/gpu/drm/amd/display/dc/spl/dc_spl_types.h @@ -479,7 +479,7 @@ struct spl_sharpness_range { }; struct adaptive_sharpness { bool enable; - int sharpness_level; + unsigned int sharpness_level; struct spl_sharpness_range sharpness_range; }; enum linear_light_scaling { // convert it in translation logic From 4a21b57d83c553bd96577be4b9fb5eeaf1f3ec5c Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?N=C3=ADcolas=20F=2E=20R=2E=20A=2E=20Prado?= Date: Tue, 25 Feb 2025 11:33:49 -0300 Subject: [PATCH 62/66] ASoC: mediatek: mt8188: Treat DMIC_GAINx_CUR as non-volatile MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit [ Upstream commit 7d87bde21c73731ddaf15e572020f80999c38ee3 ] The DMIC_GAINx_CUR registers contain the current (as in present) gain of each DMIC. During capture, this gain will ramp up until a target value is reached, and therefore the register is volatile since it is updated automatically by hardware. However, after capture the register's value returns to the value that was written to it. So reading these registers returns the current gain, and writing configures the initial gain for every capture. >From an audio configuration perspective, reading the instantaneous gain is not really useful. Instead, reading back the initial gain that was configured is the desired behavior. For that reason, consider the DMIC_GAINx_CUR registers as non-volatile, so the regmap's cache can be used to retrieve the values, rather than requiring pm runtime resuming the device. Signed-off-by: Nícolas F. R. A. Prado Reviewed-by: AngeloGioacchino Del Regno Link: https://patch.msgid.link/20250225-genio700-dmic-v2-3-3076f5b50ef7@collabora.com Signed-off-by: Mark Brown Signed-off-by: Sasha Levin --- sound/soc/mediatek/mt8188/mt8188-afe-pcm.c | 4 ---- 1 file changed, 4 deletions(-) diff --git a/sound/soc/mediatek/mt8188/mt8188-afe-pcm.c b/sound/soc/mediatek/mt8188/mt8188-afe-pcm.c index 73e5c63aeec8..d36520c6272d 100644 --- a/sound/soc/mediatek/mt8188/mt8188-afe-pcm.c +++ b/sound/soc/mediatek/mt8188/mt8188-afe-pcm.c @@ -2855,10 +2855,6 @@ static bool mt8188_is_volatile_reg(struct device *dev, unsigned int reg) case AFE_DMIC3_SRC_DEBUG_MON0: case AFE_DMIC3_UL_SRC_MON0: case AFE_DMIC3_UL_SRC_MON1: - case DMIC_GAIN1_CUR: - case DMIC_GAIN2_CUR: - case DMIC_GAIN3_CUR: - case DMIC_GAIN4_CUR: case ETDM_IN1_MONITOR: case ETDM_IN2_MONITOR: case ETDM_OUT1_MONITOR: From 7bbfaa24c02f46ad06ac44cc7f4d3b194ce9a2fb Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?N=C3=ADcolas=20F=2E=20R=2E=20A=2E=20Prado?= Date: Tue, 25 Feb 2025 11:33:48 -0300 Subject: [PATCH 63/66] ASoC: mediatek: mt8188: Add reference for dmic clocks MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit [ Upstream commit bf1800073f4d55f08191b034c86b95881e99b6fd ] Add the names for the dmic clocks, aud_afe_dmic* and aud_dmic_hires*, so they can be acquired and enabled by the platform driver. Signed-off-by: Nícolas F. R. A. Prado Reviewed-by: AngeloGioacchino Del Regno Link: https://patch.msgid.link/20250225-genio700-dmic-v2-2-3076f5b50ef7@collabora.com Signed-off-by: Mark Brown Signed-off-by: Sasha Levin --- sound/soc/mediatek/mt8188/mt8188-afe-clk.c | 8 ++++++++ sound/soc/mediatek/mt8188/mt8188-afe-clk.h | 8 ++++++++ 2 files changed, 16 insertions(+) diff --git a/sound/soc/mediatek/mt8188/mt8188-afe-clk.c b/sound/soc/mediatek/mt8188/mt8188-afe-clk.c index e69c1bb2cb23..7f411b857782 100644 --- a/sound/soc/mediatek/mt8188/mt8188-afe-clk.c +++ b/sound/soc/mediatek/mt8188/mt8188-afe-clk.c @@ -58,7 +58,15 @@ static const char *aud_clks[MT8188_CLK_NUM] = { [MT8188_CLK_AUD_ADC] = "aud_adc", [MT8188_CLK_AUD_DAC_HIRES] = "aud_dac_hires", [MT8188_CLK_AUD_A1SYS_HP] = "aud_a1sys_hp", + [MT8188_CLK_AUD_AFE_DMIC1] = "aud_afe_dmic1", + [MT8188_CLK_AUD_AFE_DMIC2] = "aud_afe_dmic2", + [MT8188_CLK_AUD_AFE_DMIC3] = "aud_afe_dmic3", + [MT8188_CLK_AUD_AFE_DMIC4] = "aud_afe_dmic4", [MT8188_CLK_AUD_ADC_HIRES] = "aud_adc_hires", + [MT8188_CLK_AUD_DMIC_HIRES1] = "aud_dmic_hires1", + [MT8188_CLK_AUD_DMIC_HIRES2] = "aud_dmic_hires2", + [MT8188_CLK_AUD_DMIC_HIRES3] = "aud_dmic_hires3", + [MT8188_CLK_AUD_DMIC_HIRES4] = "aud_dmic_hires4", [MT8188_CLK_AUD_I2SIN] = "aud_i2sin", [MT8188_CLK_AUD_TDM_IN] = "aud_tdm_in", [MT8188_CLK_AUD_I2S_OUT] = "aud_i2s_out", diff --git a/sound/soc/mediatek/mt8188/mt8188-afe-clk.h b/sound/soc/mediatek/mt8188/mt8188-afe-clk.h index ec53c171c170..c6c78d684f3e 100644 --- a/sound/soc/mediatek/mt8188/mt8188-afe-clk.h +++ b/sound/soc/mediatek/mt8188/mt8188-afe-clk.h @@ -54,7 +54,15 @@ enum { MT8188_CLK_AUD_ADC, MT8188_CLK_AUD_DAC_HIRES, MT8188_CLK_AUD_A1SYS_HP, + MT8188_CLK_AUD_AFE_DMIC1, + MT8188_CLK_AUD_AFE_DMIC2, + MT8188_CLK_AUD_AFE_DMIC3, + MT8188_CLK_AUD_AFE_DMIC4, MT8188_CLK_AUD_ADC_HIRES, + MT8188_CLK_AUD_DMIC_HIRES1, + MT8188_CLK_AUD_DMIC_HIRES2, + MT8188_CLK_AUD_DMIC_HIRES3, + MT8188_CLK_AUD_DMIC_HIRES4, MT8188_CLK_AUD_I2SIN, MT8188_CLK_AUD_TDM_IN, MT8188_CLK_AUD_I2S_OUT, From 7eb29d704d2760454d39c2a6c63460d7ff79892d Mon Sep 17 00:00:00 2001 From: Waiman Long Date: Thu, 6 Feb 2025 14:18:44 -0500 Subject: [PATCH 64/66] x86/nmi: Add an emergency handler in nmi_desc & use it in nmi_shootdown_cpus() [ Upstream commit fe37c699ae3eed6e02ee55fbf5cb9ceb7fcfd76c ] Depending on the type of panics, it was found that the __register_nmi_handler() function can be called in NMI context from nmi_shootdown_cpus() leading to a lockdep splat: WARNING: inconsistent lock state inconsistent {INITIAL USE} -> {IN-NMI} usage. lock(&nmi_desc[0].lock); lock(&nmi_desc[0].lock); Call Trace: _raw_spin_lock_irqsave __register_nmi_handler nmi_shootdown_cpus kdump_nmi_shootdown_cpus native_machine_crash_shutdown __crash_kexec In this particular case, the following panic message was printed before: Kernel panic - not syncing: Fatal hardware error! This message seemed to be given out from __ghes_panic() running in NMI context. The __register_nmi_handler() function which takes the nmi_desc lock with irq disabled shouldn't be called from NMI context as this can lead to deadlock. The nmi_shootdown_cpus() function can only be invoked once. After the first invocation, all other CPUs should be stuck in the newly added crash_nmi_callback() and cannot respond to a second NMI. Fix it by adding a new emergency NMI handler to the nmi_desc structure and provide a new set_emergency_nmi_handler() helper to set crash_nmi_callback() in any context. The new emergency handler will preempt other handlers in the linked list. That will eliminate the need to take any lock and serve the panic in NMI use case. Signed-off-by: Waiman Long Signed-off-by: Ingo Molnar Acked-by: Rik van Riel Cc: Thomas Gleixner Link: https://lore.kernel.org/r/20250206191844.131700-1-longman@redhat.com Signed-off-by: Sasha Levin --- arch/x86/include/asm/nmi.h | 2 ++ arch/x86/kernel/nmi.c | 42 ++++++++++++++++++++++++++++++++++++++ arch/x86/kernel/reboot.c | 10 +++------ 3 files changed, 47 insertions(+), 7 deletions(-) diff --git a/arch/x86/include/asm/nmi.h b/arch/x86/include/asm/nmi.h index 41a0ebb699ec..f677382093f3 100644 --- a/arch/x86/include/asm/nmi.h +++ b/arch/x86/include/asm/nmi.h @@ -56,6 +56,8 @@ int __register_nmi_handler(unsigned int, struct nmiaction *); void unregister_nmi_handler(unsigned int, const char *); +void set_emergency_nmi_handler(unsigned int type, nmi_handler_t handler); + void stop_nmi(void); void restart_nmi(void); void local_touch_nmi(void); diff --git a/arch/x86/kernel/nmi.c b/arch/x86/kernel/nmi.c index ed163c8c8604..9a95d00f1423 100644 --- a/arch/x86/kernel/nmi.c +++ b/arch/x86/kernel/nmi.c @@ -40,8 +40,12 @@ #define CREATE_TRACE_POINTS #include +/* + * An emergency handler can be set in any context including NMI + */ struct nmi_desc { raw_spinlock_t lock; + nmi_handler_t emerg_handler; struct list_head head; }; @@ -132,9 +136,22 @@ static void nmi_check_duration(struct nmiaction *action, u64 duration) static int nmi_handle(unsigned int type, struct pt_regs *regs) { struct nmi_desc *desc = nmi_to_desc(type); + nmi_handler_t ehandler; struct nmiaction *a; int handled=0; + /* + * Call the emergency handler, if set + * + * In the case of crash_nmi_callback() emergency handler, it will + * return in the case of the crashing CPU to enable it to complete + * other necessary crashing actions ASAP. Other handlers in the + * linked list won't need to be run. + */ + ehandler = desc->emerg_handler; + if (ehandler) + return ehandler(type, regs); + rcu_read_lock(); /* @@ -224,6 +241,31 @@ void unregister_nmi_handler(unsigned int type, const char *name) } EXPORT_SYMBOL_GPL(unregister_nmi_handler); +/** + * set_emergency_nmi_handler - Set emergency handler + * @type: NMI type + * @handler: the emergency handler to be stored + * + * Set an emergency NMI handler which, if set, will preempt all the other + * handlers in the linked list. If a NULL handler is passed in, it will clear + * it. It is expected that concurrent calls to this function will not happen + * or the system is screwed beyond repair. + */ +void set_emergency_nmi_handler(unsigned int type, nmi_handler_t handler) +{ + struct nmi_desc *desc = nmi_to_desc(type); + + if (WARN_ON_ONCE(desc->emerg_handler == handler)) + return; + desc->emerg_handler = handler; + + /* + * Ensure the emergency handler is visible to other CPUs before + * function return + */ + smp_wmb(); +} + static void pci_serr_error(unsigned char reason, struct pt_regs *regs) { diff --git a/arch/x86/kernel/reboot.c b/arch/x86/kernel/reboot.c index dc1dd3f3e67f..9aaac1f9f45b 100644 --- a/arch/x86/kernel/reboot.c +++ b/arch/x86/kernel/reboot.c @@ -926,15 +926,11 @@ void nmi_shootdown_cpus(nmi_shootdown_cb callback) shootdown_callback = callback; atomic_set(&waiting_for_crash_ipi, num_online_cpus() - 1); - /* Would it be better to replace the trap vector here? */ - if (register_nmi_handler(NMI_LOCAL, crash_nmi_callback, - NMI_FLAG_FIRST, "crash")) - return; /* Return what? */ + /* - * Ensure the new callback function is set before sending - * out the NMI + * Set emergency handler to preempt other handlers. */ - wmb(); + set_emergency_nmi_handler(NMI_LOCAL, crash_nmi_callback); apic_send_IPI_allbutself(NMI_VECTOR); From 6398dd09d50b432e4d01fa8cfa8c3ae854fdf0f9 Mon Sep 17 00:00:00 2001 From: Mike Christie Date: Tue, 3 Dec 2024 13:15:11 -0600 Subject: [PATCH 65/66] vhost-scsi: Return queue full for page alloc failures during copy [ Upstream commit 891b99eab0f89dbe08d216f4ab71acbeaf7a3102 ] This has us return queue full if we can't allocate a page during the copy operation so the initiator can retry. Signed-off-by: Mike Christie Message-Id: <20241203191705.19431-5-michael.christie@oracle.com> Signed-off-by: Michael S. Tsirkin Acked-by: Stefan Hajnoczi Signed-off-by: Sasha Levin --- drivers/vhost/scsi.c | 15 +++++++++------ 1 file changed, 9 insertions(+), 6 deletions(-) diff --git a/drivers/vhost/scsi.c b/drivers/vhost/scsi.c index ecad2f53b763..38d243d914d0 100644 --- a/drivers/vhost/scsi.c +++ b/drivers/vhost/scsi.c @@ -762,7 +762,7 @@ vhost_scsi_copy_iov_to_sgl(struct vhost_scsi_cmd *cmd, struct iov_iter *iter, size_t len = iov_iter_count(iter); unsigned int nbytes = 0; struct page *page; - int i; + int i, ret; if (cmd->tvc_data_direction == DMA_FROM_DEVICE) { cmd->saved_iter_addr = dup_iter(&cmd->saved_iter, iter, @@ -775,6 +775,7 @@ vhost_scsi_copy_iov_to_sgl(struct vhost_scsi_cmd *cmd, struct iov_iter *iter, page = alloc_page(GFP_KERNEL); if (!page) { i--; + ret = -ENOMEM; goto err; } @@ -782,8 +783,10 @@ vhost_scsi_copy_iov_to_sgl(struct vhost_scsi_cmd *cmd, struct iov_iter *iter, sg_set_page(&sg[i], page, nbytes, 0); if (cmd->tvc_data_direction == DMA_TO_DEVICE && - copy_page_from_iter(page, 0, nbytes, iter) != nbytes) + copy_page_from_iter(page, 0, nbytes, iter) != nbytes) { + ret = -EFAULT; goto err; + } len -= nbytes; } @@ -798,7 +801,7 @@ err: for (; i >= 0; i--) __free_page(sg_page(&sg[i])); kfree(cmd->saved_iter_addr); - return -ENOMEM; + return ret; } static int @@ -1282,9 +1285,9 @@ vhost_scsi_handle_vq(struct vhost_scsi *vs, struct vhost_virtqueue *vq) " %d\n", cmd, exp_data_len, prot_bytes, data_direction); if (data_direction != DMA_NONE) { - if (unlikely(vhost_scsi_mapal(cmd, prot_bytes, - &prot_iter, exp_data_len, - &data_iter))) { + ret = vhost_scsi_mapal(cmd, prot_bytes, &prot_iter, + exp_data_len, &data_iter); + if (unlikely(ret)) { vq_err(vq, "Failed to map iov to sgl\n"); vhost_scsi_release_cmd_res(&cmd->tvc_se_cmd); goto err; From 0d508cefcd24a3ac1b934313d647ebd68bf9b946 Mon Sep 17 00:00:00 2001 From: Konstantin Shkolnyy Date: Tue, 4 Feb 2025 11:31:27 -0600 Subject: [PATCH 66/66] vdpa/mlx5: Fix mlx5_vdpa_get_config() endianness on big-endian machines MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit [ Upstream commit 439252e167ac45a5d46f573aac1da7d8f3e051ad ] mlx5_vdpa_dev_add() doesn’t initialize mvdev->actual_features. It’s initialized later by mlx5_vdpa_set_driver_features(). However, mlx5_vdpa_get_config() depends on the VIRTIO_F_VERSION_1 flag in actual_features, to return data with correct endianness. When it’s called before mlx5_vdpa_set_driver_features(), the data are incorrectly returned as big-endian on big-endian machines, while QEMU then interprets them as little-endian. The fix is to initialize this VIRTIO_F_VERSION_1 as early as possible, especially considering that mlx5_vdpa_dev_add() insists on this flag to always be set anyway. Signed-off-by: Konstantin Shkolnyy Message-Id: <20250204173127.166673-1-kshk@linux.ibm.com> Signed-off-by: Michael S. Tsirkin Reviewed-by: Dragos Tatulea Acked-by: Jason Wang Signed-off-by: Sasha Levin --- drivers/vdpa/mlx5/net/mlx5_vnet.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drivers/vdpa/mlx5/net/mlx5_vnet.c b/drivers/vdpa/mlx5/net/mlx5_vnet.c index 5f581e71e201..76aedac37a78 100644 --- a/drivers/vdpa/mlx5/net/mlx5_vnet.c +++ b/drivers/vdpa/mlx5/net/mlx5_vnet.c @@ -3884,6 +3884,9 @@ static int mlx5_vdpa_dev_add(struct vdpa_mgmt_dev *v_mdev, const char *name, ndev->mvdev.max_vqs = max_vqs; mvdev = &ndev->mvdev; mvdev->mdev = mdev; + /* cpu_to_mlx5vdpa16() below depends on this flag */ + mvdev->actual_features = + (device_features & BIT_ULL(VIRTIO_F_VERSION_1)); ndev->vqs = kcalloc(max_vqs, sizeof(*ndev->vqs), GFP_KERNEL); ndev->event_cbs = kcalloc(max_vqs + 1, sizeof(*ndev->event_cbs), GFP_KERNEL);