Add hooks for file folio reclaim. the vendor hook keep reclaimed file folios
and help direct && kswapd reclaim.
Fix Conflicts:
drivers/android/vendor_hooks.c
include/trace/hooks/mm.h
include/trace/hooks/vmscan.h
mm/vmscan.c
remove the change of mm/memcontrol.c because
https://android-review.googlesource.com/c/kernel/common/+/3603374
Bug: 407947362
Bug: 449066551
Bug: 451083029
Change-Id: I4364ae2d269e4b4b88c12823ebf66abaa4650585
Signed-off-by: Hailong Liu <hailong.liu@oppo.com>
(cherry picked from commit 43bf81d175f5480a5537205f92b632ec8e5f1787)
(cherry picked from commit 6bae5d457e35a9e3ebefdf45e8e9e8f957fd6c84)
Along with I746f4fbc20df5cf394d7644ff2cd6f25916c9790,
it also needs to adjust check_flags.
Bug: 416087944
Bug: 436973821
Change-Id: I2871ab305963da00fcc8b0a7755bd8db7859e4c5
Signed-off-by: Minchan Kim <minchan@google.com>
(cherry picked from commit 8f5494d90cac148e269da2937456243ed2d5906f)
Signed-off-by: Richard Chang <richardycc@google.com>
The following hooks are added for memory allocation tuning:
android_vh_mm_customize_alloc_anon_thp - When allocating anonymous folios, this
hook will set different orders based on the importance of the processes. In
particular, some critical processes will allocate folios directly in this hook
to speed up memory allocation.
android_vh_mm_customize_ac - In __alloc_pages_noprof, this hook provides the
ability to modify alloc_context and alloc_flags to accommodate processes of
different importance.
android_vh_mm_customize_rmqueue - In rmqueue, this hook provides the ability to
modify alloc_flags and migratetype, which will be used by vendors to customize
alloc_flags and migratetype to speed up memory allocation.
android_vh_mm_customize_suitable_zone - For some specific gfp_mask and order
requests, this hook is used to skip zones that should not be allocated, and for
some critical processes, the wmark limit is skipped and jumps to try_this_zone
for allocation to prevent it from being stuck.
android_vh_mm_customize_wmark_ok - Vendors use this hook to customize the return
value of wmark_ok based on different payloads.
android_vh_mm_customize_zone_max_order - This hook is used to limit the maximum
order allowed for special zone. It reduces a lot of unnecessary repeated merges
and splits (because large orders are rarely allocated).
android_vh_mm_customize_zone_pageset - This hook allows vendors to customize
the high and batch of pagesets.
Bug: 431730501
Bug: 431672372
Change-Id: I65d6b2cd63688291e6360597bc2a4157fa228df7
Signed-off-by: Pengfei Li <pengfei.kernel@vivo.corp-partner.google.com>
Added a new function zone_pageset_high_and_batch_update, which is like
zone_pcp_update but with new high_min, high_max and batch values provided
directly by the caller (rather than returned by zone_highsize). This gives
vendors an interface to directly tune the pageset to suit varying loads.
Bug: 431672372
Change-Id: Icd1e297f4cfb2309b36caa7f7c61b1c7c7cd9f24
Signed-off-by: Pengfei Li <pengfei.kernel@vivo.corp-partner.google.com>
Added early param "pcp_thp_order", which specifies the order of pcp used by THP.
This allows vendors to customize the order different from HPAGE_PMD_ORDER for
THP to use.
Bug: 431672372
Change-Id: Id2c8f3b03b74added54827ffbc33fd4e4bdf619e
Signed-off-by: Pengfei Li <pengfei.kernel@vivo.corp-partner.google.com>
A vendor hook to bypass free_one_page() via __free_pages_ok() was added
via https://android-review.googlesource.com/c/kernel/common/+/3458032 for
vendor modules to hook pages on certain conditions to use later instead
of freeing pages directly.
In kernel v6.12, for same purpose, free_unref_folios() and free_unref_page()
path also needs same vendor hook to bypass free_one_page().
Bug: 410432482
Change-Id: I1863eb497eee796903b483a3a9f8808ab372b475
Signed-off-by: Sooyong Suk <s.suk@samsung.corp-partner.google.com>
Steps on the way to 6.12.34
Resolves merge conflicts in:
kernel/sched/core.c
net/netfilter/xt_mark.c
Change-Id: I6df5e27c2a5bfa8b077b1f2814ad98b2a3dc0877
Signed-off-by: Greg Kroah-Hartman <gregkh@google.com>
GKI (arm64) relevant 137 out of 624 changes, affecting 192 files +1647/-1035
a4f865ecdb nvmem: core: fix bit offsets of more than one byte [1 file, +17/-7]
4327479e55 nvmem: core: verify cell's raw_len [1 file, +12/-0]
410f8b72e0 nvmem: core: update raw_len if the bit reading is required [1 file, +3/-1]
7aea1517fb scsi: ufs: Introduce quirk to extend PA_HIBERN8TIME for UFS devices [2 files, +35/-0]
b730cb1096 virtio_ring: Fix data race by tagging event_triggered as racy for KCSAN [1 file, +1/-1]
2998813177 dma/mapping.c: dev_dbg support for dma_addressing_limited [1 file, +10/-1]
3eec42a17a dma-mapping: avoid potential unused data compilation warning [1 file, +8/-4]
97edaa0ec6 cgroup: Fix compilation issue due to cgroup_mutex not being exported [1 file, +1/-1]
f93675793b vhost_task: fix vhost_task_create() documentation [1 file, +1/-1]
e22034cbee dma-mapping: Fix warning reported for missing prototype [1 file, +8/-8]
4f5553a08f fs/buffer: split locking for pagecache lookups [1 file, +25/-16]
e138fc2316 fs/buffer: introduce sleeping flavors for pagecache lookups [2 files, +17/-0]
a49a4a87ce fs/buffer: use sleeping version of __find_get_block() [1 file, +9/-2]
f1c5aa614b fs/jbd2: use sleeping version of __find_get_block() [1 file, +9/-6]
9ece099e95 fs/ext4: use sleeping version of sb_find_get_block() [1 file, +2/-1]
64f505b08e block: fix race between set_blocksize and read paths [4 files, +43/-1]
218c838d03 io_uring: don't duplicate flushing in io_req_post_cqe [1 file, +8/-3]
8014d3e56e bpf: fix possible endless loop in BPF map iteration [1 file, +1/-1]
d40ca27602 fuse: Return EPERM rather than ENOSYS from link() [1 file, +2/-0]
bab0bd1389 exfat: call bh_read in get_block only when necessary [1 file, +77/-82]
01677e7ee1 io_uring/msg: initialise msg request opcode [1 file, +1/-0]
e506751b7d arm64: Add support for HIP09 Spectre-BHB mitigation [2 files, +3/-0]
4f427ca9ed tracing: Mark binary printing functions with __printf() attribute [4 files, +18/-21]
15787ab82a mailbox: use error ret code of of_parse_phandle_with_args() [1 file, +4/-3]
f48ee562c0 Bluetooth: Disable SCO support if READ_VOICE_SETTING is unsupported/broken [1 file, +3/-0]
44b79041c4 dql: Fix dql->limit value when reset. [1 file, +1/-1]
ac30595154 lockdep: Fix wait context check on softirq for PREEMPT_RT [1 file, +18/-0]
e63b634806 PCI: dwc: ep: Ensure proper iteration over outbound map windows [1 file, +1/-1]
37ac2434aa ext4: on a remount, only log the ro or r/w state when it has changed [1 file, +4/-3]
1d1e1efad1 libnvdimm/labels: Fix divide error in nd_label_data_init() [1 file, +2/-1]
123bcd8f42 pidfs: improve multi-threaded exec and premature thread-group leader exit polling [3 files, +9/-9]
8f82cf305e cgroup/rstat: avoid disabling irqs for O(num_cpu) [1 file, +5/-7]
a5a507fa5f blk-cgroup: improve policy registration error handling [1 file, +12/-10]
94c3cbc69a ext4: reorder capability check last [1 file, +2/-2]
e658f2d94a bpf: Return prog btf_id without capable check [1 file, +2/-2]
e2520cc19b PCI: dwc: Use resource start as ioremap() input in dw_pcie_pme_turn_off() [1 file, +1/-1]
50452704ec jbd2: do not try to recover wiped journal [1 file, +6/-5]
dab35f4921 tcp: reorganize tcp_in_ack_event() and tcp_count_delivered() [1 file, +32/-24]
555c0b713c bpf: Allow pre-ordering for bpf cgroup progs [5 files, +30/-9]
572ed3fb99 kconfig: do not clear SYMBOL_VALID when reading include/config/auto.conf [1 file, +12/-7]
174dedce64 dm: restrict dm device size to 2^63-512 bytes [1 file, +4/-0]
2f5f326214 ext4: reject the 'data_err=abort' option in nojournal mode [1 file, +12/-0]
d0dc233fe2 posix-timers: Add cond_resched() to posix_timer_add() search loop [1 file, +1/-0]
ae22452d15 posix-timers: Ensure that timer initialization is fully visible [1 file, +14/-7]
3fb9ee05ec timer_list: Don't use %pK through printk() [1 file, +2/-2]
21153e0974 netfilter: conntrack: Bound nf_conntrack sysctl writes [1 file, +9/-3]
236a87e9d2 PNP: Expand length of fixup id string [1 file, +1/-1]
6215143ad3 arm64/mm: Check pmd_table() in pmd_trans_huge() [1 file, +12/-12]
8ad58a7eba arm64/mm: Check PUD_TYPE_TABLE in pud_bad() [1 file, +2/-1]
28306c58da mmc: sdhci: Disable SD card clock before changing parameters [1 file, +7/-2]
3a75fe58a1 usb: xhci: Don't change the status of stalled TDs on failed Stop EP [1 file, +11/-1]
101a3b9920 printk: Check CON_SUSPEND when unblanking a console [1 file, +12/-2]
faba68a86a wifi: cfg80211: allow IR in 20 MHz configurations [5 files, +46/-25]
c1502fc84d ipv6: save dontfrag in cork [2 files, +6/-4]
75ae2a3553 badblocks: Fix a nonsense WARN_ON() which checks whether a u64 variable < 0 [1 file, +3/-2]
7caad075ac crypto: lzo - Fix compression buffer overrun [6 files, +106/-28]
73d01bcbf2 tcp: bring back NUMA dispersion in inet_ehash_locks_alloc() [1 file, +26/-11]
1c17190880 usb: xhci: set page size to the xHCI-supported size [2 files, +22/-20]
93f581d763 drm/gem: Test for imported GEM buffers with helper [2 files, +16/-2]
c4525b513d net: phylink: use pl->link_interface in phylink_expects_phy() [1 file, +1/-1]
f29c876d72 perf/core: Clean up perf_try_init_event() [1 file, +38/-27]
af73c8fd73 ublk: enforce ublks_max only for unprivileged devices [1 file, +27/-15]
592ba27580 perf/hw_breakpoint: Return EOPNOTSUPP for unsupported breakpoint type [1 file, +3/-2]
3de322a98b scsi: logging: Fix scsi_logging_level bounds [1 file, +3/-1]
f33b310eac ipv4: fib: Move fib_valid_key_len() to rtm_to_fib_config(). [2 files, +16/-24]
564f03a797 block: mark bounce buffering as incompatible with integrity [2 files, +5/-2]
82209faa87 ublk: complete command synchronously on error [1 file, +6/-5]
b98aad5e5e media: uvcvideo: Add sanity check to uvc_ioctl_xu_ctrl_map [1 file, +6/-0]
2d6231d5ce media: uvcvideo: Handle uvc menu translation inside uvc_get_le_value [1 file, +32/-45]
e359d62886 perf: arm_pmuv3: Call kvm_vcpu_pmu_resync_el0() before enabling counters [1 file, +2/-2]
673dde8d3c bpf: Search and add kfuncs in struct_ops prologue and epilogue [1 file, +24/-1]
083383aba0 cpuidle: menu: Avoid discarding useful information [1 file, +12/-1]
20a53c3689 loop: check in LO_FLAGS_DIRECT_IO in loop_default_blocksize [1 file, +1/-1]
b55a97d1bd dm: fix unconditional IO throttle caused by REQ_PREFLUSH [1 file, +6/-2]
9f27b38771 crypto: ahash - Set default reqsize from ahash_alg [2 files, +7/-0]
897c98fb32 crypto: skcipher - Zap type in crypto_alloc_sync_skcipher [1 file, +1/-0]
4d9fa2ebc0 net: ipv6: Init tunnel link-netns before registering dev [4 files, +9/-7]
53f42776e4 genirq/msi: Store the IOMMU IOVA directly in msi_desc instead of iommu_cookie [2 files, +25/-36]
2b129e89b8 bpf: don't do clean_live_states when state->loop_entry->branches > 0 [1 file, +4/-0]
46ba5757a7 bpf: copy_verifier_state() should copy 'loop_entry' field [1 file, +3/-0]
82b54455b6 PCI: Fix old_size lower bound in calculate_iosize() too [1 file, +2/-4]
dc5f5c9d2b hrtimers: Replace hrtimer_clock_to_base_table with switch-case [1 file, +12/-17]
000dd6e344 ASoC: ops: Enforce platform maximum on initial value [1 file, +28/-1]
c4260bf83b ASoC: soc-dai: check return value at snd_soc_dai_set_tdm_slot() [1 file, +5/-3]
5b1b4cb46d pinctrl: devicetree: do not goto err when probing hogs in pinctrl_dt_to_map [1 file, +8/-2]
69689d1138 media: v4l: Memset argument to 0 before calling get_mbus_config pad op [2 files, +5/-1]
e6e31b0182 sched: Reduce the default slice to avoid tasks getting an extra tick [1 file, +3/-3]
ef31dc41cf phy: core: don't require set_mode() callback for phy_get_mode() to work [1 file, +4/-3]
06daedb443 xfrm: prevent high SEQ input in non-ESN mode [1 file, +12/-0]
9f2911868a ip: fib_rules: Fetch net from fib_rule in fib[46]_rule_configure(). [2 files, +4/-4]
7fea5a9140 r8152: add vendor/device ID pair for Dell Alienware AW1022z [2 files, +2/-0]
16ddd67bb5 pstore: Change kmsg_bytes storage size to u32 [3 files, +9/-8]
73733c2fdb ext4: don't write back data before punch hole in nojournal mode [1 file, +5/-13]
1d15319323 f2fs: introduce f2fs_base_attr for global sysfs entries [1 file, +52/-22]
ded26f9e4c ipv4: ip_gre: Fix set but not used warning in ipgre_err() if IPv4-only [1 file, +10/-6]
76e56dbe50 net: flush_backlog() small changes [1 file, +8/-4]
58cdd1ee65 bridge: mdb: Allow replace of a host-joined group [2 files, +2/-2]
fcabb69674 rcu: handle unstable rdp in rcu_read_unlock_strict() [2 files, +11/-2]
d402437cde rcu: fix header guard for rcu_all_qs() [1 file, +1/-1]
887e39ac47 perf: Avoid the read if the count is already updated [3 files, +24/-18]
c80b2d159c bpf: Use kallsyms to find the function name of a struct_ops's stub function [1 file, +44/-54]
46f1c2b508 firmware: arm_scmi: Relax duplicate name constraint across protocol ids [1 file, +6/-13]
1351052877 drm/atomic: clarify the rules around drm_atomic_state->allow_modeset [1 file, +21/-2]
9fddd1f154 drm: Add valid clones check [1 file, +28/-0]
ff214b079d nvme-pci: add quirks for device 126f:1001 [1 file, +3/-0]
6d196cae4b nvme-pci: add quirks for WDC Blue SN550 15b7:5009 [1 file, +3/-0]
6a09b6bad0 ALSA: usb-audio: Fix duplicated name in MIDI substream names [1 file, +12/-4]
ad3e83a6c8 io_uring/fdinfo: annotate racy sq/cq head/tail reads [1 file, +2/-2]
7f7c8c03fe btrfs: correct the order of prelim_ref arguments in btrfs__prelim_ref [1 file, +1/-1]
8cafd7266f __legitimize_mnt(): check for MNT_SYNC_UMOUNT should be under mount_lock [1 file, +1/-5]
28756f22de espintcp: fix skb leaks [3 files, +9/-3]
9cbca30102 espintcp: remove encap socket caching to avoid reference leak [4 files, +8/-94]
b1a687eb15 xfrm: Fix UDP GRO handling for some corner cases [2 files, +20/-16]
447c8f0c06 kernel/fork: only call untrack_pfn_clear() on VMAs duplicated for fork() [1 file, +5/-4]
252f78a931 xfrm: Sanitize marks before insert [2 files, +6/-0]
7207effe47 driver core: Split devres APIs to device/devres.h [2 files, +125/-118]
1e8b7e96f7 Bluetooth: L2CAP: Fix not checking l2cap_chan security level [1 file, +8/-7]
cd7f022296 loop: don't require ->write_iter for writable files in loop_configure [1 file, +0/-3]
873ebaf3c1 io_uring: fix overflow resched cqe reordering [1 file, +1/-0]
689a205cd9 net/tipc: fix slab-use-after-free Read in tipc_aead_encrypt_done [1 file, +5/-0]
adb05149a9 can: slcan: allow reception of short error messages [1 file, +20/-6]
cc55dd28c2 can: bcm: add locking for bcm_op runtime updates [1 file, +45/-21]
63567ecd99 can: bcm: add missing rcu read protection for procfs content [1 file, +9/-4]
bf85e49aaf ALSA: pcm: Fix race of buffer access at PCM OSS layer [3 files, +14/-2]
e78908caf1 pmdomain: core: Fix error checking in genpd_dev_pm_attach_by_id() [1 file, +1/-1]
dc9bdfb9b0 drm/edid: fixed the bug that hdr metadata was not reset [1 file, +1/-0]
cb9a1019a6 Input: xpad - add more controllers [1 file, +3/-0]
9b8263cae6 highmem: add folio_test_partial_kmap() [2 files, +12/-5]
314bf771cb memcg: always call cond_resched() after fn() [1 file, +2/-4]
9da33ce114 mm/page_alloc.c: avoid infinite retries caused by cpuset race [1 file, +8/-0]
9f9517f156 mm: mmap: map MAP_STACK to VM_NOHUGEPAGE only if THP is enabled [1 file, +2/-0]
94efb0d656 mm: vmalloc: actually use the in-place vrealloc region [1 file, +1/-0]
483ac74183 mm: vmalloc: only zero-init on vrealloc shrink [1 file, +7/-5]
1d45e0170c spi: use container_of_cont() for to_spi_device() [1 file, +1/-4]
d28b0305f7 err.h: move IOMEM_ERR_PTR() to err.h [2 files, +3/-2]
80eb73778d bpf: abort verification if env->cur_state->loop_entry != NULL [1 file, +4/-2]
85fb1edd05 drm/gem: Internally test import_attach for imported objects [1 file, +1/-2]
Changes in 6.12.31
drm/amd/display: Configure DTBCLK_P with OPTC only for dcn401
drm/amd/display: Do not enable replay when vtotal update is pending.
drm/amd/display: Correct timing_adjust_pending flag setting.
drm/amd/display: Defer BW-optimization-blocked DRR adjustments
i2c: designware: Use temporary variable for struct device
i2c: designware: Fix an error handling path in i2c_dw_pci_probe()
phy: renesas: rcar-gen3-usb2: Move IRQ request in probe
phy: renesas: rcar-gen3-usb2: Lock around hardware registers and driver data
phy: renesas: rcar-gen3-usb2: Assert PLL reset on PHY power off
cpufreq: Add SM8650 to cpufreq-dt-platdev blocklist
nvmem: rockchip-otp: Move read-offset into variant-data
nvmem: rockchip-otp: add rk3576 variant data
nvmem: core: fix bit offsets of more than one byte
nvmem: core: verify cell's raw_len
nvmem: core: update raw_len if the bit reading is required
nvmem: qfprom: switch to 4-byte aligned reads
scsi: target: iscsi: Fix timeout on deleted connection
scsi: ufs: Introduce quirk to extend PA_HIBERN8TIME for UFS devices
virtio_ring: Fix data race by tagging event_triggered as racy for KCSAN
dma/mapping.c: dev_dbg support for dma_addressing_limited
intel_th: avoid using deprecated page->mapping, index fields
mei: vsc: Use struct vsc_tp_packet as vsc-tp tx_buf and rx_buf type
dma-mapping: avoid potential unused data compilation warning
cgroup: Fix compilation issue due to cgroup_mutex not being exported
vhost_task: fix vhost_task_create() documentation
vhost-scsi: protect vq->log_used with vq->mutex
scsi: mpi3mr: Add level check to control event logging
net: enetc: refactor bulk flipping of RX buffers to separate function
dma-mapping: Fix warning reported for missing prototype
ima: process_measurement() needlessly takes inode_lock() on MAY_READ
fs/buffer: split locking for pagecache lookups
fs/buffer: introduce sleeping flavors for pagecache lookups
fs/buffer: use sleeping version of __find_get_block()
fs/ocfs2: use sleeping version of __find_get_block()
fs/jbd2: use sleeping version of __find_get_block()
fs/ext4: use sleeping version of sb_find_get_block()
drm/amd/display: Enable urgent latency adjustment on DCN35
drm/amdgpu: Allow P2P access through XGMI
selftests/bpf: Mitigate sockmap_ktls disconnect_after_delete failure
block: fix race between set_blocksize and read paths
io_uring: don't duplicate flushing in io_req_post_cqe
bpf: fix possible endless loop in BPF map iteration
samples/bpf: Fix compilation failure for samples/bpf on LoongArch Fedora
kconfig: merge_config: use an empty file as initfile
x86/fred: Fix system hang during S4 resume with FRED enabled
s390/vfio-ap: Fix no AP queue sharing allowed message written to kernel log
cifs: Add fallback for SMB2 CREATE without FILE_READ_ATTRIBUTES
cifs: Fix querying and creating MF symlinks over SMB1
cifs: Fix negotiate retry functionality
smb: client: Store original IO parameters and prevent zero IO sizes
fuse: Return EPERM rather than ENOSYS from link()
exfat: call bh_read in get_block only when necessary
io_uring/msg: initialise msg request opcode
NFSv4: Check for delegation validity in nfs_start_delegation_return_locked()
NFS: Don't allow waiting for exiting tasks
SUNRPC: Don't allow waiting for exiting tasks
arm64: Add support for HIP09 Spectre-BHB mitigation
iommufd: Extend IOMMU_GET_HW_INFO to report PASID capability
tracing: Mark binary printing functions with __printf() attribute
ACPI: PNP: Add Intel OC Watchdog IDs to non-PNP device list
tpm: Convert warn to dbg in tpm2_start_auth_session()
mailbox: pcc: Use acpi_os_ioremap() instead of ioremap()
mailbox: use error ret code of of_parse_phandle_with_args()
riscv: Allow NOMMU kernels to access all of RAM
fbdev: fsl-diu-fb: add missing device_remove_file()
fbcon: Use correct erase colour for clearing in fbcon
fbdev: core: tileblit: Implement missing margin clearing for tileblit
cifs: Set default Netbios RFC1001 server name to hostname in UNC
cifs: add validation check for the fields in smb_aces
cifs: Fix establishing NetBIOS session for SMB2+ connection
NFSv4: Treat ENETUNREACH errors as fatal for state recovery
SUNRPC: rpc_clnt_set_transport() must not change the autobind setting
SUNRPC: rpcbind should never reset the port to the value '0'
spi-rockchip: Fix register out of bounds access
ASoC: codecs: wsa884x: Correct VI sense channel mask
ASoC: codecs: wsa883x: Correct VI sense channel mask
mctp: Fix incorrect tx flow invalidation condition in mctp-i2c
net: tn40xx: add pci-id of the aqr105-based Tehuti TN4010 cards
net: tn40xx: create swnode for mdio and aqr105 phy and add to mdiobus
thermal/drivers/mediatek/lvts: Start sensor interrupts disabled
thermal/drivers/qoriq: Power down TMU on system suspend
Bluetooth: btmtksdio: Prevent enabling interrupts after IRQ handler removal
Bluetooth: Disable SCO support if READ_VOICE_SETTING is unsupported/broken
dql: Fix dql->limit value when reset.
lockdep: Fix wait context check on softirq for PREEMPT_RT
objtool: Properly disable uaccess validation
PCI: dwc: ep: Ensure proper iteration over outbound map windows
r8169: disable RTL8126 ZRX-DC timeout
tools/build: Don't pass test log files to linker
pNFS/flexfiles: Report ENETDOWN as a connection error
drm/amdgpu/discovery: check ip_discovery fw file available
drm/amdkfd: set precise mem ops caps to disabled for gfx 11 and 12
PCI: vmd: Disable MSI remapping bypass under Xen
xen/pci: Do not register devices with segments >= 0x10000
ext4: on a remount, only log the ro or r/w state when it has changed
libnvdimm/labels: Fix divide error in nd_label_data_init()
pidfs: improve multi-threaded exec and premature thread-group leader exit polling
staging: vchiq_arm: Create keep-alive thread during probe
mmc: host: Wait for Vdd to settle on card power off
drm/amdgpu: Skip pcie_replay_count sysfs creation for VF
cgroup/rstat: avoid disabling irqs for O(num_cpu)
wifi: mt76: only mark tx-status-failed frames as ACKed on mt76x0/2
wifi: mt76: mt7996: fix SER reset trigger on WED reset
wifi: mt76: mt7996: revise TXS size
wifi: mt76: mt7925: load the appropriate CLC data based on hardware type
wifi: mt76: mt7925: fix fails to enter low power mode in suspend state
x86/headers: Replace __ASSEMBLY__ with __ASSEMBLER__ in UAPI headers
x86/stackprotector/64: Only export __ref_stack_chk_guard on CONFIG_SMP
x86/smpboot: Fix INIT delay assignment for extended Intel Families
x86/microcode: Update the Intel processor flag scan check
x86/mm: Check return value from memblock_phys_alloc_range()
i2c: qup: Vote for interconnect bandwidth to DRAM
i2c: pxa: fix call balance of i2c->clk handling routines
btrfs: make btrfs_discard_workfn() block_group ref explicit
btrfs: avoid linker error in btrfs_find_create_tree_block()
btrfs: run btrfs_error_commit_super() early
btrfs: fix non-empty delayed iputs list on unmount due to async workers
btrfs: get zone unusable bytes while holding lock at btrfs_reclaim_bgs_work()
btrfs: send: return -ENAMETOOLONG when attempting a path that is too long
blk-cgroup: improve policy registration error handling
drm/amdgpu: release xcp_mgr on exit
drm/amd/display: Guard against setting dispclk low for dcn31x
drm/amdgpu: adjust drm_firmware_drivers_only() handling
i3c: master: svc: Fix missing STOP for master request
s390/tlb: Use mm_has_pgste() instead of mm_alloc_pgste()
dlm: make tcp still work in multi-link env
clocksource/drivers/timer-riscv: Stop stimecmp when cpu hotplug
um: Store full CSGSFS and SS register from mcontext
um: Update min_low_pfn to match changes in uml_reserved
wifi: mwifiex: Fix HT40 bandwidth issue.
bnxt_en: Query FW parameters when the CAPS_CHANGE bit is set
riscv: Call secondary mmu notifier when flushing the tlb
ext4: reorder capability check last
hypfs_create_cpu_files(): add missing check for hypfs_mkdir() failure
scsi: st: Tighten the page format heuristics with MODE SELECT
scsi: st: ERASE does not change tape location
vfio/pci: Handle INTx IRQ_NOTCONNECTED
bpf: Return prog btf_id without capable check
PCI: dwc: Use resource start as ioremap() input in dw_pcie_pme_turn_off()
jbd2: do not try to recover wiped journal
tcp: reorganize tcp_in_ack_event() and tcp_count_delivered()
rtc: rv3032: fix EERD location
objtool: Fix error handling inconsistencies in check()
thunderbolt: Do not add non-active NVM if NVM upgrade is disabled for retimer
erofs: initialize decompression early
spi: spi-mux: Fix coverity issue, unchecked return value
ASoC: pcm6240: Drop bogus code handling IRQ as GPIO
ASoC: mediatek: mt6359: Add stub for mt6359_accdet_enable_jack_detect
bpf: Allow pre-ordering for bpf cgroup progs
kbuild: fix argument parsing in scripts/config
kconfig: do not clear SYMBOL_VALID when reading include/config/auto.conf
crypto: octeontx2 - suppress auth failure screaming due to negative tests
dm: restrict dm device size to 2^63-512 bytes
net/smc: use the correct ndev to find pnetid by pnetid table
xen: Add support for XenServer 6.1 platform device
pinctrl-tegra: Restore SFSEL bit when freeing pins
mfd: tps65219: Remove TPS65219_REG_TI_DEV_ID check
drm/amdgpu/gfx12: don't read registers in mqd init
drm/amdgpu/gfx11: don't read registers in mqd init
drm/amdgpu: Update SRIOV video codec caps
ASoC: sun4i-codec: support hp-det-gpios property
clk: qcom: lpassaudiocc-sc7280: Add support for LPASS resets for QCM6490
ext4: reject the 'data_err=abort' option in nojournal mode
ext4: do not convert the unwritten extents if data writeback fails
RDMA/uverbs: Propagate errors from rdma_lookup_get_uobject()
posix-timers: Add cond_resched() to posix_timer_add() search loop
posix-timers: Ensure that timer initialization is fully visible
net: stmmac: dwmac-rk: Validate GRF and peripheral GRF during probe
net: hsr: Fix PRP duplicate detection
timer_list: Don't use %pK through printk()
wifi: rtw89: set force HE TB mode when connecting to 11ax AP
netfilter: conntrack: Bound nf_conntrack sysctl writes
PNP: Expand length of fixup id string
phy: rockchip: usbdp: Only verify link rates/lanes/voltage when the corresponding set flags are set
arm64/mm: Check pmd_table() in pmd_trans_huge()
arm64/mm: Check PUD_TYPE_TABLE in pud_bad()
mmc: dw_mmc: add exynos7870 DW MMC support
mmc: sdhci: Disable SD card clock before changing parameters
usb: xhci: Don't change the status of stalled TDs on failed Stop EP
wifi: iwlwifi: mvm: fix setting the TK when associated
hwmon: (dell-smm) Increment the number of fans
iommu: Keep dev->iommu state consistent
printk: Check CON_SUSPEND when unblanking a console
wifi: iwlwifi: don't warn when if there is a FW error
wifi: iwlwifi: w/a FW SMPS mode selection
wifi: iwlwifi: fix debug actions order
wifi: iwlwifi: mark Br device not integrated
wifi: iwlwifi: fix the ECKV UEFI variable name
wifi: mac80211: fix warning on disconnect during failed ML reconf
wifi: mac80211_hwsim: Fix MLD address translation
wifi: cfg80211: allow IR in 20 MHz configurations
ipv6: save dontfrag in cork
drm/amd/display: remove minimum Dispclk and apply oem panel timing.
drm/amd/display: calculate the remain segments for all pipes
drm/amd/display: not abort link train when bw is low
drm/amd/display: Fix incorrect DPCD configs while Replay/PSR switch
gfs2: Check for empty queue in run_queue
auxdisplay: charlcd: Partially revert "Move hwidth and bwidth to struct hd44780_common"
ASoC: qcom: sm8250: explicitly set format in sm8250_be_hw_params_fixup()
badblocks: Fix a nonsense WARN_ON() which checks whether a u64 variable < 0
coresight-etb10: change etb_drvdata spinlock's type to raw_spinlock_t
iommu/amd/pgtbl_v2: Improve error handling
cpufreq: tegra186: Share policy per cluster
watchdog: aspeed: Update bootstatus handling
PCI: endpoint: pci-epf-test: Fix double free that causes kernel to oops
misc: pci_endpoint_test: Give disabled BARs a distinct error code
crypto: lzo - Fix compression buffer overrun
crypto: mxs-dcp - Only set OTP_KEY bit for OTP key
drm/amdkfd: Set per-process flags only once for gfx9/10/11/12
drm/amdkfd: Set per-process flags only once cik/vi
drm/amdgpu: Fix missing drain retry fault the last entry
arm64: tegra: p2597: Fix gpio for vdd-1v8-dis regulator
arm64: tegra: Resize aperture for the IGX PCIe C5 slot
powerpc/prom_init: Fixup missing #size-cells on PowerBook6,7
ALSA: seq: Improve data consistency at polling
tcp: bring back NUMA dispersion in inet_ehash_locks_alloc()
rtc: ds1307: stop disabling alarms on probe
ieee802154: ca8210: Use proper setters and getters for bitwise types
drm/xe: Nuke VM's mapping upon close
drm/xe: Retry BO allocation
soc: samsung: include linux/array_size.h where needed
ARM: tegra: Switch DSI-B clock parent to PLLD on Tegra114
media: c8sectpfe: Call of_node_put(i2c_bus) only once in c8sectpfe_probe()
usb: xhci: set page size to the xHCI-supported size
dm cache: prevent BUG_ON by blocking retries on failed device resumes
soc: mediatek: mtk-mutex: Add DPI1 SOF/EOF to MT8188 mutex tables
orangefs: Do not truncate file size
drm/gem: Test for imported GEM buffers with helper
net: phylink: use pl->link_interface in phylink_expects_phy()
blk-throttle: don't take carryover for prioritized processing of metadata
remoteproc: qcom_wcnss: Handle platforms with only single power domain
drm/amdgpu: Do not program AGP BAR regs under SRIOV in gfxhub_v1_0.c
drm/amd/display: Ensure DMCUB idle before reset on DCN31/DCN35
drm/amd/display: Skip checking FRL_MODE bit for PCON BW determination
drm/amd/display: Fix DMUB reset sequence for DCN401
drm/amd/display: Fix p-state type when p-state is unsupported
drm/amd/display: Request HW cursor on DCN3.2 with SubVP
perf/core: Clean up perf_try_init_event()
media: cx231xx: set device_caps for 417
pinctrl: bcm281xx: Use "unsigned int" instead of bare "unsigned"
rcu: Fix get_state_synchronize_rcu_full() GP-start detection
net: ethernet: ti: cpsw_new: populate netdev of_node
net: phy: nxp-c45-tja11xx: add match_phy_device to TJA1103/TJA1104
dpll: Add an assertion to check freq_supported_num
ublk: enforce ublks_max only for unprivileged devices
iommufd: Disallow allocating nested parent domain with fault ID
media: imx335: Set vblank immediately
net: pktgen: fix mpls maximum labels list parsing
perf/hw_breakpoint: Return EOPNOTSUPP for unsupported breakpoint type
ALSA: hda/realtek: Enable PC beep passthrough for HP EliteBook 855 G7
scsi: logging: Fix scsi_logging_level bounds
ipv4: fib: Move fib_valid_key_len() to rtm_to_fib_config().
drm/rockchip: vop2: Add uv swap for cluster window
block: mark bounce buffering as incompatible with integrity
ublk: complete command synchronously on error
media: uvcvideo: Add sanity check to uvc_ioctl_xu_ctrl_map
media: uvcvideo: Handle uvc menu translation inside uvc_get_le_value
clk: imx8mp: inform CCF of maximum frequency of clocks
x86/bugs: Make spectre user default depend on MITIGATION_SPECTRE_V2
hwmon: (gpio-fan) Add missing mutex locks
ARM: at91: pm: fix at91_suspend_finish for ZQ calibration
drm/mediatek: mtk_dpi: Add checks for reg_h_fre_con existence
fpga: altera-cvp: Increase credit timeout
perf: arm_pmuv3: Call kvm_vcpu_pmu_resync_el0() before enabling counters
soc: apple: rtkit: Use high prio work queue
soc: apple: rtkit: Implement OSLog buffers properly
wifi: ath12k: Report proper tx completion status to mac80211
PCI: brcmstb: Expand inbound window size up to 64GB
PCI: brcmstb: Add a softdep to MIP MSI-X driver
firmware: arm_ffa: Set dma_mask for ffa devices
drm/xe/vf: Retry sending MMIO request to GUC on timeout error
drm/xe/pf: Create a link between PF and VF devices
net/mlx5: Avoid report two health errors on same syndrome
selftests/net: have `gro.sh -t` return a correct exit code
pinctrl: sophgo: avoid to modify untouched bit when setting cv1800 pinconf
drm/amdkfd: KFD release_work possible circular locking
drm/xe: xe_gen_wa_oob: replace program_invocation_short_name
leds: pwm-multicolor: Add check for fwnode_property_read_u32
net: ethernet: mtk_ppe_offload: Allow QinQ, double ETH_P_8021Q only
net: xgene-v2: remove incorrect ACPI_PTR annotation
bonding: report duplicate MAC address in all situations
wifi: ath12k: Improve BSS discovery with hidden SSID in 6 GHz band
soc: ti: k3-socinfo: Do not use syscon helper to build regmap
bpf: Search and add kfuncs in struct_ops prologue and epilogue
Octeontx2-af: RPM: Register driver with PCI subsys IDs
x86/build: Fix broken copy command in genimage.sh when making isoimage
drm/amd/display: handle max_downscale_src_width fail check
drm/amd/display: fix dcn4x init failed
drm/amd/display: Fix mismatch type comparison
ASoC: mediatek: mt8188: Treat DMIC_GAINx_CUR as non-volatile
ASoC: mediatek: mt8188: Add reference for dmic clocks
x86/nmi: Add an emergency handler in nmi_desc & use it in nmi_shootdown_cpus()
vhost-scsi: Return queue full for page alloc failures during copy
vdpa/mlx5: Fix mlx5_vdpa_get_config() endianness on big-endian machines
cpuidle: menu: Avoid discarding useful information
media: adv7180: Disable test-pattern control on adv7180
media: tc358746: improve calculation of the D-PHY timing registers
net/mlx5e: Add correct match to check IPSec syndromes for switchdev mode
scsi: mpi3mr: Update timestamp only for supervisor IOCs
loop: check in LO_FLAGS_DIRECT_IO in loop_default_blocksize
libbpf: Fix out-of-bound read
dm: fix unconditional IO throttle caused by REQ_PREFLUSH
scsi: scsi_debug: First fixes for tapes
net/mlx5: Change POOL_NEXT_SIZE define value and make it global
x86/kaslr: Reduce KASLR entropy on most x86 systems
crypto: ahash - Set default reqsize from ahash_alg
crypto: skcipher - Zap type in crypto_alloc_sync_skcipher
net: ipv6: Init tunnel link-netns before registering dev
drm/xe/oa: Ensure that polled read returns latest data
MIPS: Use arch specific syscall name match function
drm/amdgpu: remove all KFD fences from the BO on release
x86/locking: Use ALT_OUTPUT_SP() for percpu_{,try_}cmpxchg{64,128}_op()
genirq/msi: Store the IOMMU IOVA directly in msi_desc instead of iommu_cookie
MIPS: pm-cps: Use per-CPU variables as per-CPU, not per-core
clocksource: mips-gic-timer: Enable counter when CPUs start
PCI: epf-mhi: Update device ID for SA8775P
scsi: mpt3sas: Send a diag reset if target reset fails
wifi: rtw88: Fix rtw_init_vht_cap() for RTL8814AU
wifi: rtw88: Fix rtw_init_ht_cap() for RTL8814AU
wifi: rtw88: Fix rtw_desc_to_mcsrate() to handle MCS16-31
wifi: rtw89: fw: propagate error code from rtw89_h2c_tx()
wifi: rtw89: fw: get sb_sel_ver via get_unaligned_le32()
wifi: rtw89: fw: add blacklist to avoid obsolete secure firmware
wifi: rtw89: 8922a: fix incorrect STA-ID in EHT MU PPDU
net: pktgen: fix access outside of user given buffer in pktgen_thread_write()
power: supply: axp20x_battery: Update temp sensor for AXP717 from device tree
EDAC/ie31200: work around false positive build warning
i3c: master: svc: Flush FIFO before sending Dynamic Address Assignment(DAA)
mfd: axp20x: AXP717: Add AXP717_TS_PIN_CFG to writeable regs
eeprom: ee1004: Check chip before probing
irqchip/riscv-imsic: Separate next and previous pointers in IMSIC vector
drm/amd/pm: Fetch current power limit from PMFW
drm/amd/display: Add support for disconnected eDP streams
drm/amd/display: Guard against setting dispclk low when active
drm/amd/display: Fix BT2020 YCbCr limited/full range input
drm/amd/display: Read LTTPR ALPM caps during link cap retrieval
Revert "drm/amd/display: Request HW cursor on DCN3.2 with SubVP"
drm/amd/display: Don't treat wb connector as physical in create_validate_stream_for_sink
serial: mctrl_gpio: split disable_ms into sync and no_sync APIs
RDMA/core: Fix best page size finding when it can cross SG entries
pmdomain: imx: gpcv2: use proper helper for property detection
can: c_can: Use of_property_present() to test existence of DT property
bpf: don't do clean_live_states when state->loop_entry->branches > 0
bpf: copy_verifier_state() should copy 'loop_entry' field
eth: mlx4: don't try to complete XDP frames in netpoll
PCI: Fix old_size lower bound in calculate_iosize() too
ACPI: HED: Always initialize before evged
vxlan: Join / leave MC group after remote changes
hrtimers: Replace hrtimer_clock_to_base_table with switch-case
irqchip/riscv-imsic: Set irq_set_affinity() for IMSIC base
media: test-drivers: vivid: don't call schedule in loop
net/mlx5: Modify LSB bitmask in temperature event to include only the first bit
net/mlx5: Apply rate-limiting to high temperature warning
firmware: arm_ffa: Reject higher major version as incompatible
firmware: arm_ffa: Handle the presence of host partition in the partition info
firmware: xilinx: Dont send linux address to get fpga config get status
ASoC: ops: Enforce platform maximum on initial value
ASoC: tas2764: Add reg defaults for TAS2764_INT_CLK_CFG
ASoC: tas2764: Mark SW_RESET as volatile
ASoC: tas2764: Power up/down amp on mute ops
ASoC: soc-dai: check return value at snd_soc_dai_set_tdm_slot()
pinctrl: devicetree: do not goto err when probing hogs in pinctrl_dt_to_map
smack: recognize ipv4 CIPSO w/o categories
smack: Revert "smackfs: Added check catlen"
kunit: tool: Use qboot on QEMU x86_64
media: i2c: imx219: Correct the minimum vblanking value
media: v4l: Memset argument to 0 before calling get_mbus_config pad op
net/mlx4_core: Avoid impossible mlx4_db_alloc() order value
drm/xe: Stop ignoring errors from xe_ttm_stolen_mgr_init()
drm/xe: Fix xe_tile_init_noalloc() error propagation
clk: qcom: ipq5018: allow it to be bulid on arm32
clk: qcom: clk-alpha-pll: Do not use random stack value for recalc rate
drm/xe/debugfs: fixed the return value of wedged_mode_set
drm/xe/debugfs: Add missing xe_pm_runtime_put in wedge_mode_set
x86/ibt: Handle FineIBT in handle_cfi_failure()
x86/traps: Cleanup and robustify decode_bug()
sched: Reduce the default slice to avoid tasks getting an extra tick
serial: sh-sci: Update the suspend/resume support
pinctrl: renesas: rzg2l: Add suspend/resume support for pull up/down
phy: phy-rockchip-samsung-hdptx: Swap the definitions of LCPLL_REF and ROPLL_REF
phy: core: don't require set_mode() callback for phy_get_mode() to work
phy: exynos5-usbdrd: fix EDS distribution tuning (gs101)
soundwire: amd: change the soundwire wake enable/disable sequence
soundwire: cadence_master: set frame shape and divider based on actual clk freq
net: stmmac: dwmac-loongson: Set correct {tx,rx}_fifo_size
drm/amdgpu/mes11: fix set_hw_resources_1 calculation
drm/amdkfd: fix missing L2 cache info in topology
drm/amdgpu: Set snoop bit for SDMA for MI series
drm/amd/display: pass calculated dram_speed_mts to dml2
drm/amd/display: Don't try AUX transactions on disconnected link
drm/amdgpu: reset psp->cmd to NULL after releasing the buffer
drm/amd/pm: Skip P2S load for SMU v13.0.12
drm/amd/display: Support multiple options during psr entry.
Revert "drm/amd/display: Exit idle optimizations before attempt to access PHY"
drm/amd/display: Update CR AUX RD interval interpretation
drm/amd/display: Initial psr_version with correct setting
drm/amd/display: Increase block_sequence array size
drm/amd/display: Use Nominal vBlank If Provided Instead Of Capping It
drm/amd/display: Populate register address for dentist for dcn401
drm/amdgpu: Use active umc info from discovery
drm/amdgpu: enlarge the VBIOS binary size limit
drm/amd/display/dm: drop hw_support check in amdgpu_dm_i2c_xfer()
scsi: target: spc: Fix loop traversal in spc_rsoc_get_descr()
net/mlx5: XDP, Enable TX side XDP multi-buffer support
net/mlx5: Extend Ethtool loopback selftest to support non-linear SKB
net/mlx5e: set the tx_queue_len for pfifo_fast
net/mlx5e: reduce rep rxq depth to 256 for ECPF
net/mlx5e: reduce the max log mpwrq sz for ECPF and reps
drm/v3d: Add clock handling
xfrm: prevent high SEQ input in non-ESN mode
wifi: ath12k: fix the ampdu id fetch in the HAL_RX_MPDU_START TLV
mptcp: pm: userspace: flags: clearer msg if no remote addr
wifi: iwlwifi: use correct IMR dump variable
wifi: iwlwifi: don't warn during reprobe
wifi: mac80211: don't unconditionally call drv_mgd_complete_tx()
wifi: mac80211: remove misplaced drv_mgd_complete_tx() call
wifi: mac80211: set ieee80211_prep_tx_info::link_id upon Auth Rx
net: fec: Refactor MAC reset to function
powerpc/pseries/iommu: memory notifier incorrectly adds TCEs for pmemory
powerpc/pseries/iommu: create DDW for devices with DMA mask less than 64-bits
arch/powerpc/perf: Check the instruction type before creating sample with perf_mem_data_src
ip: fib_rules: Fetch net from fib_rule in fib[46]_rule_configure().
r8152: add vendor/device ID pair for Dell Alienware AW1022z
iio: adc: ad7944: don't use storagebits for sizing
pstore: Change kmsg_bytes storage size to u32
leds: trigger: netdev: Configure LED blink interval for HW offload
ext4: don't write back data before punch hole in nojournal mode
ext4: remove writable userspace mappings before truncating page cache
wifi: rtw88: Fix download_firmware_validate() for RTL8814AU
wifi: rtw88: Fix __rtw_download_firmware() for RTL8814AU
wifi: rtw89: coex: Assign value over than 0 to avoid firmware timer hang
wifi: rtw89: fw: validate multi-firmware header before getting its size
wifi: rtw89: fw: validate multi-firmware header before accessing
wifi: rtw89: call power_on ahead before selecting firmware
clk: qcom: camcc-sm8250: Use clk_rcg2_shared_ops for some RCGs
net: page_pool: avoid false positive warning if NAPI was never added
tools/power turbostat: Clustered Uncore MHz counters should honor show/hide options
hwmon: (xgene-hwmon) use appropriate type for the latency value
f2fs: introduce f2fs_base_attr for global sysfs entries
media: qcom: camss: csid: Only add TPG v4l2 ctrl if TPG hardware is available
media: qcom: camss: Add default case in vfe_src_pad_code
drm/rockchip: vop2: Improve display modes handling on RK3588 HDMI0
eth: fbnic: set IFF_UNICAST_FLT to avoid enabling promiscuous mode when adding unicast addrs
tools: ynl-gen: don't output external constants
net/mlx5e: Avoid WARN_ON when configuring MQPRIO with HTB offload enabled
cpufreq: amd-pstate: Remove unnecessary driver_lock in set_boost
vxlan: Annotate FDB data races
ipv4: ip_gre: Fix set but not used warning in ipgre_err() if IPv4-only
r8169: don't scan PHY addresses > 0
net: flush_backlog() small changes
bridge: mdb: Allow replace of a host-joined group
ice: init flow director before RDMA
ice: treat dyn_allowed only as suggestion
rcu: handle quiescent states for PREEMPT_RCU=n, PREEMPT_COUNT=y
rcu: handle unstable rdp in rcu_read_unlock_strict()
rcu: fix header guard for rcu_all_qs()
perf: Avoid the read if the count is already updated
ice: count combined queues using Rx/Tx count
drm/xe/relay: Don't use GFP_KERNEL for new transactions
net/mana: fix warning in the writer of client oob
scsi: lpfc: Handle duplicate D_IDs in ndlp search-by D_ID routine
scsi: lpfc: Ignore ndlp rport mismatch in dev_loss_tmo callbk
scsi: lpfc: Free phba irq in lpfc_sli4_enable_msi() when pci_irq_vector() fails
scsi: st: Restore some drive settings after reset
wifi: ath12k: Avoid napi_sync() before napi_enable()
HID: usbkbd: Fix the bit shift number for LED_KANA
arm64: zynqmp: add clock-output-names property in clock nodes
ASoC: codecs: pcm3168a: Allow for 24-bit in provider mode
ASoC: rt722-sdca: Add some missing readable registers
irqchip/riscv-aplic: Add support for hart indexes
dm vdo indexer: prevent unterminated string warning
dm vdo: use a short static string for thread name prefix
drm/ast: Find VBIOS mode from regular display size
bpf: Use kallsyms to find the function name of a struct_ops's stub function
bpftool: Fix readlink usage in get_fd_type
firmware: arm_scmi: Relax duplicate name constraint across protocol ids
perf/amd/ibs: Fix perf_ibs_op.cnt_mask for CurCnt
perf/amd/ibs: Fix ->config to sample period calculation for OP PMU
clk: renesas: rzg2l-cpg: Refactor Runtime PM clock validation
wifi: rtl8xxxu: retry firmware download on error
wifi: rtw88: Don't use static local variable in rtw8822b_set_tx_power_index_by_rate
wifi: rtw89: add wiphy_lock() to work that isn't held wiphy_lock() yet
spi: zynqmp-gqspi: Always acknowledge interrupts
regulator: ad5398: Add device tree support
wifi: ath12k: fix ath12k_hal_tx_cmd_ext_desc_setup() info1 override
accel/qaic: Mask out SR-IOV PCI resources
drm/xe/pf: Reset GuC VF config when unprovisioning critical resource
wifi: ath9k: return by of_get_mac_address
wifi: ath12k: Fetch regdb.bin file from board-2.bin
wifi: ath12k: Fix end offset bit definition in monitor ring descriptor
drm: bridge: adv7511: fill stream capabilities
drm/nouveau: fix the broken marco GSP_MSG_MAX_SIZE
wifi: ath11k: Use dma_alloc_noncoherent for rx_tid buffer allocation
drm/xe: Move suballocator init to after display init
drm/xe: Do not attempt to bootstrap VF in execlists mode
wifi: rtw89: coex: Separated Wi-Fi connecting event from Wi-Fi scan event
drm/xe/sa: Always call drm_suballoc_manager_fini()
drm/xe: Reject BO eviction if BO is bound to current VM
drm/atomic: clarify the rules around drm_atomic_state->allow_modeset
drm/buddy: fix issue that force_merge cannot free all roots
drm/panel-edp: Add Starry 116KHD024006
drm: Add valid clones check
ASoC: imx-card: Adjust over allocation of memory in imx_card_parse_of()
book3s64/radix: Fix compile errors when CONFIG_ARCH_WANT_OPTIMIZE_DAX_VMEMMAP=n
pinctrl: meson: define the pull up/down resistor value as 60 kOhm
smb: server: smb2pdu: check return value of xa_store()
platform/x86/intel: hid: Add Pantherlake support
platform/x86: asus-wmi: Disable OOBE state after resume from hibernation
platform/x86: ideapad-laptop: add support for some new buttons
ASoC: cs42l43: Disable headphone clamps during type detection
ASoC: Intel: bytcr_rt5640: Add DMI quirk for Acer Aspire SW3-013
ALSA: hda/realtek: Add quirk for HP Spectre x360 15-df1xxx
nvme-pci: add quirks for device 126f:1001
nvme-pci: add quirks for WDC Blue SN550 15b7:5009
ALSA: usb-audio: Fix duplicated name in MIDI substream names
nvmet-tcp: don't restore null sk_state_change
io_uring/fdinfo: annotate racy sq/cq head/tail reads
cifs: Fix and improve cifs_query_path_info() and cifs_query_file_info()
cifs: Fix changing times and read-only attr over SMB1 smb_set_file_info() function
ASoC: intel/sdw_utils: Add volume limit to cs42l43 speakers
btrfs: compression: adjust cb->compressed_folios allocation type
btrfs: correct the order of prelim_ref arguments in btrfs__prelim_ref
btrfs: handle empty eb->folios in num_extent_folios()
btrfs: avoid NULL pointer dereference if no valid csum tree
tools: ynl-gen: validate 0 len strings from kernel
block: only update request sector if needed
wifi: iwlwifi: add support for Killer on MTL
x86/Kconfig: make CFI_AUTO_DEFAULT depend on !RUST or Rust >= 1.88
xenbus: Allow PVH dom0 a non-local xenstore
drm/amd/display: Call FP Protect Before Mode Programming/Mode Support
__legitimize_mnt(): check for MNT_SYNC_UMOUNT should be under mount_lock
soundwire: bus: Fix race on the creation of the IRQ domain
espintcp: fix skb leaks
espintcp: remove encap socket caching to avoid reference leak
xfrm: Fix UDP GRO handling for some corner cases
dmaengine: idxd: Fix allowing write() from different address spaces
x86/sev: Fix operator precedence in GHCB_MSR_VMPL_REQ_LEVEL macro
kernel/fork: only call untrack_pfn_clear() on VMAs duplicated for fork()
remoteproc: qcom_wcnss: Fix on platforms without fallback regulators
clk: sunxi-ng: d1: Add missing divider for MMC mod clocks
xfrm: Sanitize marks before insert
dmaengine: idxd: Fix ->poll() return value
dmaengine: fsl-edma: Fix return code for unhandled interrupts
driver core: Split devres APIs to device/devres.h
devres: Introduce devm_kmemdup_array()
ASoC: SOF: Intel: hda: Fix UAF when reloading module
irqchip/riscv-imsic: Start local sync timer on correct CPU
perf/x86/intel: Fix segfault with PEBS-via-PT with sample_freq
Bluetooth: L2CAP: Fix not checking l2cap_chan security level
Bluetooth: btusb: use skb_pull to avoid unsafe access in QCA dump handling
ptp: ocp: Limit signal/freq counts in summary output functions
bridge: netfilter: Fix forwarding of fragmented packets
ice: fix vf->num_mac count with port representors
ice: Fix LACP bonds without SRIOV environment
idpf: fix null-ptr-deref in idpf_features_check
loop: don't require ->write_iter for writable files in loop_configure
pinctrl: qcom: switch to devm_register_sys_off_handler()
net: dwmac-sun8i: Use parsed internal PHY address instead of 1
net: lan743x: Restore SGMII CTRL register on resume
io_uring: fix overflow resched cqe reordering
idpf: fix idpf_vport_splitq_napi_poll()
sch_hfsc: Fix qlen accounting bug when using peek in hfsc_enqueue()
octeontx2-pf: Add AF_XDP non-zero copy support
net/tipc: fix slab-use-after-free Read in tipc_aead_encrypt_done
octeontx2-af: Set LMT_ENA bit for APR table entries
octeontx2-af: Fix APR entry mapping based on APR_LMT_CFG
clk: s2mps11: initialise clk_hw_onecell_data::num before accessing ::hws[] in probe()
crypto: algif_hash - fix double free in hash_accept
padata: do not leak refcount in reorder_work
can: slcan: allow reception of short error messages
can: bcm: add locking for bcm_op runtime updates
can: bcm: add missing rcu read protection for procfs content
ASoC: SOF: ipc4-control: Use SOF_CTRL_CMD_BINARY as numid for bytes_ext
ASoC: SOF: Intel: hda-bus: Use PIO mode on ACE2+ platforms
ASoc: SOF: topology: connect DAI to a single DAI link
ASoC: SOF: ipc4-pcm: Delay reporting is only supported for playback direction
ALSA: pcm: Fix race of buffer access at PCM OSS layer
ALSA: hda/realtek: Add quirk for Lenovo Yoga Pro 7 14ASP10
llc: fix data loss when reading from a socket in llc_ui_recvmsg()
can: kvaser_pciefd: Continue parsing DMA buf after dropped RX
can: kvaser_pciefd: Fix echo_skb race
net: dsa: microchip: linearize skb for tail-tagging switches
vmxnet3: update MTU after device quiesce
pmdomain: renesas: rcar: Remove obsolete nullify checks
pmdomain: core: Fix error checking in genpd_dev_pm_attach_by_id()
platform/x86: dell-wmi-sysman: Avoid buffer overflow in current_password_store()
thermal: intel: x86_pkg_temp_thermal: Fix bogus trip temperature
drm/edid: fixed the bug that hdr metadata was not reset
smb: client: Fix use-after-free in cifs_fill_dirent
arm64: dts: marvell: uDPU: define pinctrl state for alarm LEDs
smb: client: Reset all search buffer pointers when releasing buffer
Revert "drm/amd: Keep display off while going into S4"
Input: xpad - add more controllers
highmem: add folio_test_partial_kmap()
memcg: always call cond_resched() after fn()
mm/page_alloc.c: avoid infinite retries caused by cpuset race
mm: mmap: map MAP_STACK to VM_NOHUGEPAGE only if THP is enabled
mm: vmalloc: actually use the in-place vrealloc region
mm: vmalloc: only zero-init on vrealloc shrink
nilfs2: fix deadlock warnings caused by lock dependency in init_nilfs()
Bluetooth: btmtksdio: Check function enabled before doing close
Bluetooth: btmtksdio: Do close if SDIO card removed without close
Revert "arm64: dts: allwinner: h6: Use RSB for AXP805 PMIC connection"
ksmbd: fix stream write failure
platform/x86: think-lmi: Fix attribute name usage for non-compliant items
spi: use container_of_cont() for to_spi_device()
spi: spi-fsl-dspi: restrict register range for regmap access
spi: spi-fsl-dspi: Halt the module after a new message transfer
spi: spi-fsl-dspi: Reset SR flags before sending a new message
err.h: move IOMEM_ERR_PTR() to err.h
gcc-15: make 'unterminated string initialization' just a warning
gcc-15: disable '-Wunterminated-string-initialization' entirely for now
Fix mis-uses of 'cc-option' for warning disablement
kbuild: Properly disable -Wunterminated-string-initialization for clang
drm/amd/display: Exit idle optimizations before accessing PHY
bpf: abort verification if env->cur_state->loop_entry != NULL
serial: sh-sci: Save and restore more registers
drm/amdkfd: Correct F8_MODE for gfx950
watchdog: aspeed: fix 64-bit division
pinctrl: tegra: Fix off by one in tegra_pinctrl_get_group()
i3c: master: svc: Fix implicit fallthrough in svc_i3c_master_ibi_work()
x86/mm/init: Handle the special case of device private pages in add_pages(), to not increase max_pfn and trigger dma_addressing_limited() bounce buffers bounce buffers
drm/gem: Internally test import_attach for imported objects
Linux 6.12.31
Change-Id: I017795966fb764f9320a6a0df1571d19e5e631fe
Signed-off-by: Greg Kroah-Hartman <gregkh@google.com>
Add vendor hook to reset unused bits in page flags to zero when
releasing a page. This reserves space for OEMs to define
vendor-specific features.
Bug: 420361650
Bug: 426115925
Change-Id: Ife2b96de64eea670d257f1a8d9fd07d16c0ddc83
Signed-off-by: Yang Yang <yang.yang@vivo.com>
(cherry picked from commit 8fb77f6f9d20be56cd76b579b2830eb287e2e8bb)
Signed-off-by: ying zuxin <11154159@vivo.com>
find_suitable_fallback() is not as efficient as it could be, and
somewhat difficult to follow.
1. should_try_claim_block() is a loop invariant. There is no point in
checking fallback areas if the caller is interested in claimable
blocks but the order and the migratetype don't allow for that.
2. __rmqueue_steal() doesn't care about claimability, so it shouldn't
have to run those tests.
Different callers want different things from this helper:
1. __compact_finished() scans orders up until it finds a claimable block
2. __rmqueue_claim() scans orders down as long as blocks are claimable
3. __rmqueue_steal() doesn't care about claimability at all
Move should_try_claim_block() out of the loop. Only test it for the two
callers who care in the first place. Distinguish "no blocks" from "order
+ mt are not claimable" in the return value; __rmqueue_claim() can stop
once order becomes unclaimable, __compact_finished() can keep advancing
until order becomes claimable.
Before:
Performance counter stats for './run case-lru-file-mmap-read' (5 runs):
85,294.85 msec task-clock # 5.644 CPUs utilized ( +- 0.32% )
15,968 context-switches # 187.209 /sec ( +- 3.81% )
153 cpu-migrations # 1.794 /sec ( +- 3.29% )
801,808 page-faults # 9.400 K/sec ( +- 0.10% )
733,358,331,786 instructions # 1.87 insn per cycle ( +- 0.20% ) (64.94%)
392,622,904,199 cycles # 4.603 GHz ( +- 0.31% ) (64.84%)
148,563,488,531 branches # 1.742 G/sec ( +- 0.18% ) (63.86%)
152,143,228 branch-misses # 0.10% of all branches ( +- 1.19% ) (62.82%)
15.1128 +- 0.0637 seconds time elapsed ( +- 0.42% )
After:
Performance counter stats for './run case-lru-file-mmap-read' (5 runs):
84,380.21 msec task-clock # 5.664 CPUs utilized ( +- 0.21% )
16,656 context-switches # 197.392 /sec ( +- 3.27% )
151 cpu-migrations # 1.790 /sec ( +- 3.28% )
801,703 page-faults # 9.501 K/sec ( +- 0.09% )
731,914,183,060 instructions # 1.88 insn per cycle ( +- 0.38% ) (64.90%)
388,673,535,116 cycles # 4.606 GHz ( +- 0.24% ) (65.06%)
148,251,482,143 branches # 1.757 G/sec ( +- 0.37% ) (63.92%)
149,766,550 branch-misses # 0.10% of all branches ( +- 1.22% ) (62.88%)
14.8968 +- 0.0486 seconds time elapsed ( +- 0.33% )
Link:
https://lkml.kernel.org/r/20250407180154.63348-2-hannes@cmpxchg.org
Signed-off-by: Johannes Weiner <hannes@cmpxchg.org> Reviewed-by: Brendan
Jackman <jackmanb@google.com> Tested-by: Shivank Garg <shivankg@amd.com>
Reviewed-by: Vlastimil Babka <vbabka@suse.cz> Cc: Carlos Song
<carlos.song@nxp.com> Cc: Mel Gorman <mgorman@techsingularity.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Bug: 420836317
Change-Id: I2886de9da0fd99047cf5c675cd2ae7c386267770
(cherry picked from commit ee414bd97b3fa0a4f74e40004e3b4191326bd46c
https://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm.git/?h=mm-everything)
Signed-off-by: yipeng xiang <yipengxiang@honor.corp-partner.google.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@google.com>
There's lots of text here but it's a little hard to follow, this is an
attempt to break it up and align its structure more closely with the code.
Reword the top-level function comment to just explain what question the
function answers from the point of view of the caller.
Break up the internal logic into different sections that can have their
own commentary describing why that part of the rationale is present.
Note the page_group_by_mobility_disabled logic is not explained in the
commentary, that is outside the scope of this patch...
Link: https://lkml.kernel.org/r/20250228-clarify-steal-v4-2-cb2ef1a4e610@google.com
Signed-off-by: Brendan Jackman <jackmanb@google.com>
Reviewed-by: Vlastimil Babka <vbabka@suse.cz>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Mel Gorman <mgorman@techsingularity.net>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: Yosry Ahmed <yosry.ahmed@linux.dev>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Bug: 420836317
(cherry picked from commit a14efee04796dd3f614eaf5348ca1ac099c21349)
Change-Id: I6c7f908a4e9f025726dadab210c2d59004fe1946
Signed-off-by: yipeng xiang <yipengxiang@honor.corp-partner.google.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@google.com>
Patch series "mm/page_alloc: Some clarifications for migratetype
fallback", v4.
A couple of patches to try and make the code easier to follow.
This patch (of 2):
This code is rather confusing because:
1. "Steal" is sometimes used to refer to the general concept of
allocating from a from a block of a fallback migratetype
(steal_suitable_fallback()) but sometimes it refers specifically to
converting a whole block's migratetype (can_steal_fallback()).
2. can_steal_fallback() sounds as though it's answering the question "am
I functionally permitted to allocate from that other type" but in
fact it is encoding a heuristic preference.
3. The same piece of data has different names in different places:
can_steal vs whole_block. This reinforces point 2 because it looks
like the different names reflect a shift in intent from "am I
allowed to steal" to "do I want to steal", but no such shift exists.
Fix 1. by avoiding the term "steal" in ambiguous contexts. Start using
the term "claim" to refer to the special case of stealing the entire
block.
Fix 2. by using "should" instead of "can", and also rename its
parameters and add some commentary to make it more explicit what they
mean.
Fix 3. by adopting the new "claim" terminology universally for this
set of variables.
Link: https://lkml.kernel.org/r/20250228-clarify-steal-v4-0-cb2ef1a4e610@google.com
Link: https://lkml.kernel.org/r/20250228-clarify-steal-v4-1-cb2ef1a4e610@google.com
Signed-off-by: Brendan Jackman <jackmanb@google.com>
Reviewed-by: Vlastimil Babka <vbabka@suse.cz>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Mel Gorman <mgorman@techsingularity.net>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: Yosry Ahmed <yosry.ahmed@linux.dev>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Bug: 420836317
(cherry picked from commit e47f1f56dd82cc6d91f5c4d914a534aa03cd12ca)
Change-Id: I8f1b57aebf308f378f50cd1381f31d249362078e
Signed-off-by: yipeng xiang <yipengxiang@honor.corp-partner.google.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@google.com>
The way the fallback rules are spread out makes them hard to follow. Move
the functions next to each other at least.
Link: https://lkml.kernel.org/r/20250225001023.1494422-4-hannes@cmpxchg.org
Signed-off-by: Johannes Weiner <hannes@cmpxchg.org>
Reviewed-by: Brendan Jackman <jackmanb@google.com>
Reviewed-by: Vlastimil Babka <vbabka@suse.cz>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Bug: 420836317
(cherry picked from commit a4138a2702a4428317ecdb115934554df4b788b4)
[
1. In the original patch of the find_suitable_fallback function, replace MIGRATE_PCPTYPES with MIGRATE_FALLBACKS.;
2. Keep the hook function in the reserve_highatomic_pageblock and unreserve_highatomic_pageblock functions.
]
Change-Id: I069e8dd7f8b009c686daef4459f9f1452b3f4c2c
Signed-off-by: yipeng xiang <yipengxiang@honor.corp-partner.google.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@google.com>
GKI (arm64) relevant 18 out of 143 changes, affecting 32 files +213/-83
10d1496f85 fs/xattr.c: fix simple_xattr_list to always include security.* xattrs [1 file, +24/-0]
bc4c54cbb4 binfmt_elf: Move brk for static PIE even if ASLR disabled [1 file, +47/-24]
f0d70d8dca cgroup/cpuset: Extend kthread_is_per_cpu() check to all PF_NO_SETAFFINITY tasks [1 file, +4/-2]
517c11fe4f tracing: probes: Fix a possible race in trace_probe_log APIs [5 files, +27/-3]
94e7272b63 HID: uclogic: Add NULL check in uclogic_input_configured() [1 file, +4/-3]
28826a89fd Bluetooth: MGMT: Fix MGMT_OP_ADD_DEVICE invalid device flags [1 file, +6/-3]
d1365ca80b net_sched: Flush gso_skb list too during ->change() [7 files, +21/-6]
ddfa034da3 nvme-pci: make nvme_pci_npages_prp() __always_inline [1 file, +1/-1]
a3c147040b nvme-pci: acquire cq_poll_lock in nvme_poll_irqdisable [1 file, +2/-0]
c88f4ff535 ALSA: usb-audio: Add sample rate quirk for Audioengine D1 [1 file, +2/-0]
93152dac0b ALSA: usb-audio: Add sample rate quirk for Microdia JP001 USB Camera [1 file, +2/-0]
fe1bebd0ed dma-buf: insert memory barrier before updating num_fences [1 file, +3/-2]
7d353da580 ftrace: Fix preemption accounting for stacktrace trigger command [1 file, +1/-1]
bffc3038a2 scsi: sd_zbc: block: Respect bio vector limits for REPORT ZONES buffer [3 files, +7/-2]
20d6e621be ring-buffer: Fix persistent buffer when commit page is the reader page [1 file, +5/-3]
fe0756daad mm: userfaultfd: correct dirty flags set for both present and swap pte [1 file, +10/-2]
74953f93f4 mm/page_alloc: fix race condition in unaccepted memory handling [1 file, +0/-23]
5924b32446 usb: typec: ucsi: displayport: Fix deadlock [3 files, +47/-8]
Changes in 6.12.30
arm64: dts: rockchip: Assign RT5616 MCLK rate on rk3588-friendlyelec-cm3588
fs/xattr.c: fix simple_xattr_list to always include security.* xattrs
drivers/platform/x86/amd: pmf: Check for invalid sideloaded Smart PC Policies
drivers/platform/x86/amd: pmf: Check for invalid Smart PC Policies
riscv: dts: sophgo: fix DMA data-width configuration for CV18xx
binfmt_elf: Move brk for static PIE even if ASLR disabled
platform/x86/amd/pmc: Declare quirk_spurious_8042 for MECHREVO Wujie 14XA (GX4HRXL)
platform/x86: asus-wmi: Fix wlan_ctrl_by_user detection
arm64: dts: imx8mp-var-som: Fix LDO5 shutdown causing SD card timeout
cgroup/cpuset: Extend kthread_is_per_cpu() check to all PF_NO_SETAFFINITY tasks
tracing: probes: Fix a possible race in trace_probe_log APIs
tpm: tis: Double the timeout B to 4s
uio_hv_generic: Fix sysfs creation path for ring buffer
KVM: Add member to struct kvm_gfn_range to indicate private/shared
KVM: x86/mmu: Prevent installing hugepages when mem attributes are changing
iio: adc: ad7266: Fix potential timestamp alignment issue.
iio: chemical: pms7003: use aligned_s64 for timestamp
iio: pressure: mprls0025pa: use aligned_s64 for timestamp
drm/amd: Add Suspend/Hibernate notification callback support
Revert "drm/amd: Stop evicting resources on APUs in suspend"
xhci: dbc: Improve performance by removing delay in transfer event polling.
xhci: dbc: Avoid event polling busyloop if pending rx transfers are inactive.
iio: adc: ad7768-1: Fix insufficient alignment of timestamp.
iio: chemical: sps30: use aligned_s64 for timestamp
virtio_ring: add a func argument 'recycle_done' to virtqueue_reset()
virtio_net: ensure netdev_tx_reset_queue is called on bind xsk for tx
RDMA/rxe: Fix slab-use-after-free Read in rxe_queue_cleanup bug
HID: thrustmaster: fix memory leak in thrustmaster_interrupts()
HID: uclogic: Add NULL check in uclogic_input_configured()
nfs: handle failure of nfs_get_lock_context in unlock path
spi: loopback-test: Do not split 1024-byte hexdumps
RDMA/core: Fix "KASAN: slab-use-after-free Read in ib_register_device" problem
Bluetooth: MGMT: Fix MGMT_OP_ADD_DEVICE invalid device flags
net_sched: Flush gso_skb list too during ->change()
tools/net/ynl: ethtool: fix crash when Hardware Clock info is missing
mctp: no longer rely on net->dev_index_head[]
net: mctp: Don't access ifa_index when missing
selftests: ncdevmem: Redirect all non-payload output to stderr
selftests: ncdevmem: Separate out dmabuf provider
selftests: ncdevmem: Unify error handling
selftests: ncdevmem: Make client_ip optional
selftests: ncdevmem: Switch to AF_INET6
tests/ncdevmem: Fix double-free of queue array
net: mctp: Ensure keys maintain only one ref to corresponding dev
ALSA: seq: Fix delivery of UMP events to group ports
ALSA: ump: Fix a typo of snd_ump_stream_msg_device_info
net: cadence: macb: Fix a possible deadlock in macb_halt_tx.
net: dsa: sja1105: discard incoming frames in BR_STATE_LISTENING
nvme-pci: make nvme_pci_npages_prp() __always_inline
nvme-pci: acquire cq_poll_lock in nvme_poll_irqdisable
ALSA: sh: SND_AICA should depend on SH_DMA_API
net: dsa: b53: prevent standalone from trying to forward to other ports
vsock/test: Fix occasional failure in SIOCOUTQ tests
net/mlx5e: Disable MACsec offload for uplink representor profile
qlcnic: fix memory leak in qlcnic_sriov_channel_cfg_cmd()
regulator: max20086: fix invalid memory access
drm/xe: Save CTX_TIMESTAMP mmio value instead of LRC value
netlink: specs: tc: fix a couple of attribute names
netlink: specs: tc: all actions are indexed arrays
octeontx2-pf: macsec: Fix incorrect max transmit size in TX secy
net: ethernet: mtk_eth_soc: fix typo for declaration MT7988 ESW capability
octeontx2-af: Fix CGX Receive counters
octeontx2-pf: Do not reallocate all ntuple filters
wifi: mac80211: Set n_channels after allocating struct cfg80211_scan_request
mlxsw: spectrum_router: Fix use-after-free when deleting GRE net devices
net/tls: fix kernel panic when alloc_page failed
tsnep: fix timestamping with a stacked DSA driver
NFSv4/pnfs: Reset the layout state after a layoutreturn
dmaengine: Revert "dmaengine: dmatest: Fix dmatest waiting less when interrupted"
sched_ext: bpf_iter_scx_dsq_new() should always initialize iterator
udf: Make sure i_lenExtents is uptodate on inode eviction
HID: bpf: abort dispatch if device destroyed
LoongArch: Prevent cond_resched() occurring within kernel-fpu
LoongArch: Move __arch_cpu_idle() to .cpuidle.text section
LoongArch: Save and restore CSR.CNTC for hibernation
LoongArch: Fix MAX_REG_OFFSET calculation
LoongArch: uprobes: Remove user_{en,dis}able_single_step()
LoongArch: uprobes: Remove redundant code about resume_era
btrfs: fix discard worker infinite loop after disabling discard
btrfs: fix folio leak in submit_one_async_extent()
btrfs: add back warning for mount option commit values exceeding 300
Revert "drm/amd/display: Hardware cursor changes color when switched to software cursor"
drm/amdgpu: fix incorrect MALL size for GFX1151
drm/amdgpu: csa unmap use uninterruptible lock
drm/amd/display: Correct the reply value when AUX write incomplete
drm/amd/display: Avoid flooding unnecessary info messages
MAINTAINERS: Update Alexey Makhalov's email address
gpio: pca953x: fix IRQ storm on system wake up
ACPI: PPTT: Fix processor subtable walk
ALSA: es1968: Add error handling for snd_pcm_hw_constraint_pow2()
ALSA: usb-audio: Add sample rate quirk for Audioengine D1
ALSA: usb-audio: Add sample rate quirk for Microdia JP001 USB Camera
dma-buf: insert memory barrier before updating num_fences
hv_netvsc: Use vmbus_sendpacket_mpb_desc() to send VMBus messages
hv_netvsc: Preserve contiguous PFN grouping in the page buffer array
hv_netvsc: Remove rmsg_pgcnt
arm64: dts: amlogic: dreambox: fix missing clkc_audio node
arm64: dts: rockchip: Remove overdrive-mode OPPs from RK3588J SoC dtsi
Drivers: hv: Allow vmbus_sendpacket_mpb_desc() to create multiple ranges
Drivers: hv: vmbus: Remove vmbus_sendpacket_pagebuffer()
kbuild: Disable -Wdefault-const-init-unsafe
ftrace: Fix preemption accounting for stacktrace trigger command
ftrace: Fix preemption accounting for stacktrace filter command
tracing: samples: Initialize trace_array_printk() with the correct function
phy: tegra: xusb: Use a bitmask for UTMI pad power state tracking
phy: Fix error handling in tegra_xusb_port_init
phy: renesas: rcar-gen3-usb2: Fix role detection on unbind/bind
phy: renesas: rcar-gen3-usb2: Set timing registers only once
scsi: sd_zbc: block: Respect bio vector limits for REPORT ZONES buffer
smb: client: fix memory leak during error handling for POSIX mkdir
spi: tegra114: Use value to check for invalid delays
tpm: Mask TPM RC in tpm2_start_auth_session()
wifi: mt76: disable napi on driver removal
ring-buffer: Fix persistent buffer when commit page is the reader page
net: qede: Initialize qede_ll_ops with designated initializer
mm: userfaultfd: correct dirty flags set for both present and swap pte
dmaengine: ti: k3-udma: Add missing locking
dmaengine: ti: k3-udma: Use cap_mask directly from dma_device structure instead of a local copy
dmaengine: idxd: fix memory leak in error handling path of idxd_setup_wqs
dmaengine: idxd: fix memory leak in error handling path of idxd_setup_engines
dmaengine: idxd: fix memory leak in error handling path of idxd_setup_groups
dmaengine: idxd: Add missing cleanup for early error out in idxd_setup_internals
dmaengine: idxd: Add missing cleanups in cleanup internals
dmaengine: idxd: Add missing idxd cleanup to fix memory leak in remove call
dmaengine: idxd: fix memory leak in error handling path of idxd_alloc
dmaengine: idxd: fix memory leak in error handling path of idxd_pci_probe
dmaengine: idxd: Refactor remove call with idxd_cleanup() helper
CIFS: New mount option for cifs.upcall namespace resolution
drm/xe/gsc: do not flush the GSC worker from the reset path
mm/page_alloc: fix race condition in unaccepted memory handling
accel/ivpu: Rename ivpu_log_level to fw_log_level
accel/ivpu: Reset fw log on cold boot
accel/ivpu: Refactor functions in ivpu_fw_log.c
accel/ivpu: Fix fw log printing
iio: light: opt3001: fix deadlock due to concurrent flag access
Bluetooth: btnxpuart: Fix kernel panic during FW release
drm/fbdev-dma: Support struct drm_driver.fbdev_probe
drm/panel-mipi-dbi: Run DRM default client setup
drm/tiny: panel-mipi-dbi: Use drm_client_setup_with_fourcc()
usb: typec: ucsi: displayport: Fix deadlock
phy: tegra: xusb: remove a stray unlock
drm/amdgpu: fix pm notifier handling
Linux 6.12.30
Change-Id: I4fefed85c02f1ed826b7ee014700b80c10300bb5
Signed-off-by: Greg Kroah-Hartman <gregkh@google.com>
GKI (arm64) relevant 44 out of 185 changes, affecting 54 files +634/-365
b32411f045 dm: add missing unlock on in dm_keyslot_evict() [1 file, +2/-1]
61e0fc3312 fs/erofs/fileio: call erofs_onlinefolio_split() after bio_add_folio() [1 file, +2/-2]
14ee85b748 firmware: arm_scmi: Fix timeout checks on polling path [1 file, +8/-5]
98cd7ed927 sch_htb: make htb_deactivate() idempotent [1 file, +6/-9]
35be4c0cdf gre: Fix again IPv6 link-local address generation. [1 file, +9/-6]
c33927f385 can: gw: fix RCU/BH usage in cgw_create_job() [1 file, +90/-59]
4555c4a13a wifi: mac80211: fix the type of status_code for negotiated TID to Link Mapping [2 files, +7/-7]
64385c0d02 erofs: ensure the extra temporary copy is valid for shortened bvecs [1 file, +14/-17]
b37e54259c bpf: Scrub packet on bpf_redirect_peer [1 file, +1/-0]
bb8f86f40e net: export a helper for adding up queue stats [2 files, +56/-19]
302a0cd0bb Input: xpad - fix Share button on Xbox One controllers [1 file, +20/-15]
bf239d3835 Input: xpad - add support for 8BitDo Ultimate 2 Wireless Controller [1 file, +1/-0]
38bb0170d6 Input: xpad - fix two controller table values [1 file, +2/-2]
2910019b04 mm: vmalloc: support more granular vrealloc() sizing [2 files, +25/-7]
6166c3cf40 mm/huge_memory: fix dereferencing invalid pmd migration entry [1 file, +8/-3]
b543a5a73b mm/userfaultfd: fix uninitialized output field for -EAGAIN race [1 file, +22/-6]
7f37e31483 io_uring: ensure deferred completions are flushed for multishot [1 file, +8/-0]
abbc99e898 arm64: cpufeature: Move arm64_use_ng_mappings to the .data section to prevent wrong idmap generation [1 file, +8/-1]
d66a22f6a4 memblock: Accept allocated memory before use in memblock_double_array() [1 file, +8/-1]
d63851049f module: ensure that kobject_put() is safe for module type kobjects [1 file, +3/-1]
75f23e49ad usb: gadget: f_ecm: Add get_status callback [1 file, +7/-0]
d1c8fa4c6e usb: gadget: Use get_status callback to set remote wakeup capability [1 file, +5/-7]
3366a19948 usb: typec: tcpm: delay SNK_TRY_WAIT_DEBOUNCE to SRC_TRYWAIT transition [1 file, +1/-1]
5ad298d6d4 usb: typec: ucsi: displayport: Fix NULL pointer access [1 file, +2/-0]
afe8849597 types: Complement the aligned types with signed 64-bit one [2 files, +3/-1]
02a77b3020 loop: Use bdev limit helpers for configuring discard [1 file, +4/-4]
722f6dece7 loop: Simplify discard granularity calc [1 file, +1/-2]
0558ce095b loop: Fix ABBA locking race [1 file, +15/-15]
5e1470b276 loop: refactor queue limits updates [1 file, +20/-16]
a781ffe410 loop: factor out a loop_assign_backing_file helper [1 file, +10/-10]
184b147b9f loop: Add sanity check for read/write_iter [1 file, +23/-0]
19fa2a4830 nvme: unblock ctrl state transition for firmware update [1 file, +2/-1]
3edac2949e io_uring/sqpoll: Increase task_work submission batch size [1 file, +1/-1]
cd010271a9 do_umount(): add missing barrier before refcount checks in sync case [1 file, +2/-1]
2482f7705b io_uring: always arm linked timeouts prior to issue [1 file, +15/-35]
564d25b1a6 mm: page_alloc: don't steal single pages from biggest buddy [1 file, +34/-46]
16bae58f73 mm: page_alloc: speed up fallbacks in rmqueue_bulk() [1 file, +80/-33]
86b37810fa sched/eevdf: Fix se->slice being set to U64_MAX and resulting crash [1 file, +1/-3]
2a3915e861 arm64: insn: Add support for encoding DSB [2 files, +38/-23]
ec5bca57af arm64: proton-pack: Expose whether the platform is mitigated by firmware [2 files, +6/-0]
f2aebb8ec6 arm64: proton-pack: Expose whether the branchy loop k value [2 files, +6/-0]
38c345fd54 arm64: bpf: Add BHB mitigation to the epilogue for cBPF programs [3 files, +52/-5]
e5f5100f1c arm64: bpf: Only mitigate cBPF programs loaded by unprivileged users [1 file, +3/-0]
2176530849 arm64: proton-pack: Add new CPUs 'k' values for branch mitigation [2 files, +3/-0]
Changes in 6.12.29
dm: add missing unlock on in dm_keyslot_evict()
fs/erofs/fileio: call erofs_onlinefolio_split() after bio_add_folio()
Revert "btrfs: canonicalize the device path before adding it"
arm64: dts: imx8mm-verdin: Link reg_usdhc2_vqmmc to usdhc2
firmware: arm_scmi: Fix timeout checks on polling path
can: mcan: m_can_class_unregister(): fix order of unregistration calls
s390/pci: Fix missing check for zpci_create_device() error return
wifi: cfg80211: fix out-of-bounds access during multi-link element defragmentation
vfio/pci: Align huge faults to order
s390/pci: Fix duplicate pci_dev_put() in disable_slot() when PF has child VFs
can: mcp251xfd: mcp251xfd_remove(): fix order of unregistration calls
can: rockchip_canfd: rkcanfd_remove(): fix order of unregistration calls
ksmbd: prevent rename with empty string
ksmbd: prevent out-of-bounds stream writes by validating *pos
ksmbd: Fix UAF in __close_file_table_ids
openvswitch: Fix unsafe attribute parsing in output_userspace()
ksmbd: fix memory leak in parse_lease_state()
s390/entry: Fix last breaking event handling in case of stack corruption
sch_htb: make htb_deactivate() idempotent
virtio_net: xsk: bind/unbind xsk for tx
virtio-net: free xsk_buffs on error in virtnet_xsk_pool_enable()
gre: Fix again IPv6 link-local address generation.
net: ethernet: mtk_eth_soc: reset all TX queues on DMA free
net: ethernet: mtk_eth_soc: do not reset PSE when setting FE
can: m_can: m_can_class_allocate_dev(): initialize spin lock on device probe
can: mcp251xfd: fix TDC setting for low data bit rates
can: gw: fix RCU/BH usage in cgw_create_job()
wifi: mac80211: fix the type of status_code for negotiated TID to Link Mapping
ice: Initial support for E825C hardware in ice_adapter
ice: use DSN instead of PCI BDF for ice_adapter index
erofs: ensure the extra temporary copy is valid for shortened bvecs
ipvs: fix uninit-value for saddr in do_output_route4
netfilter: ipset: fix region locking in hash types
bpf: Scrub packet on bpf_redirect_peer
net: dsa: b53: allow leaky reserved multicast
net: dsa: b53: keep CPU port always tagged again
net: dsa: b53: fix clearing PVID of a port
net: dsa: b53: fix flushing old pvid VLAN on pvid change
net: dsa: b53: fix VLAN ID for untagged vlan on bridge leave
net: dsa: b53: always rejoin default untagged VLAN on bridge leave
net: dsa: b53: do not allow to configure VLAN 0
net: dsa: b53: do not program vlans when vlan filtering is off
net: dsa: b53: fix toggling vlan_filtering
net: dsa: b53: fix learning on VLAN unaware bridges
net: dsa: b53: do not set learning and unicast/multicast on up
fbnic: Fix initialization of mailbox descriptor rings
fbnic: Gate AXI read/write enabling on FW mailbox
fbnic: Actually flush_tx instead of stalling out
fbnic: Improve responsiveness of fbnic_mbx_poll_tx_ready
fbnic: Pull fbnic_fw_xmit_cap_msg use out of interrupt context
fbnic: Do not allow mailbox to toggle to ready outside fbnic_mbx_poll_tx_ready
net: export a helper for adding up queue stats
virtio-net: fix total qstat values
Input: cyttsp5 - ensure minimum reset pulse width
Input: cyttsp5 - fix power control issue on wakeup
Input: mtk-pmic-keys - fix possible null pointer dereference
Input: xpad - fix Share button on Xbox One controllers
Input: xpad - add support for 8BitDo Ultimate 2 Wireless Controller
Input: xpad - fix two controller table values
Input: synaptics - enable InterTouch on Dynabook Portege X30-D
Input: synaptics - enable InterTouch on Dynabook Portege X30L-G
Input: synaptics - enable InterTouch on Dell Precision M3800
Input: synaptics - enable SMBus for HP Elitebook 850 G1
Input: synaptics - enable InterTouch on TUXEDO InfinityBook Pro 14 v5
rust: clean Rust 1.88.0's `unnecessary_transmutes` lint
objtool/rust: add one more `noreturn` Rust function for Rust 1.87.0
rust: clean Rust 1.88.0's warning about `clippy::disallowed_macros` configuration
staging: iio: adc: ad7816: Correct conditional logic for store mode
staging: bcm2835-camera: Initialise dev in v4l2_dev
staging: axis-fifo: Remove hardware resets for user errors
staging: axis-fifo: Correct handling of tx_fifo_depth for size validation
x86/mm: Eliminate window where TLB flushes may be inadvertently skipped
mm: fix folio_pte_batch() on XEN PV
mm: vmalloc: support more granular vrealloc() sizing
mm/huge_memory: fix dereferencing invalid pmd migration entry
mm/userfaultfd: fix uninitialized output field for -EAGAIN race
selftests/mm: compaction_test: support platform with huge mount of memory
selftests/mm: fix a build failure on powerpc
KVM: SVM: Forcibly leave SMM mode on SHUTDOWN interception
drm/amd/display: Shift DMUB AUX reply command if necessary
io_uring: ensure deferred completions are flushed for multishot
iio: adc: ad7606: fix serial register access
iio: adc: rockchip: Fix clock initialization sequence
iio: adis16201: Correct inclinometer channel resolution
iio: imu: inv_mpu6050: align buffer for timestamp
iio: imu: st_lsm6dsx: fix possible lockup in st_lsm6dsx_read_fifo
iio: imu: st_lsm6dsx: fix possible lockup in st_lsm6dsx_read_tagged_fifo
drm/v3d: Add job to pending list if the reset was skipped
drm/xe: Add page queue multiplier
drm/amdgpu/vcn: using separate VCN1_AON_SOC offset
drm/amd/display: Fix invalid context error in dml helper
drm/amd/display: more liberal vmin/vmax update for freesync
drm/amd/display: Fix the checking condition in dmub aux handling
drm/amd/display: Remove incorrect checking in dmub aux handler
drm/amd/display: Fix wrong handling for AUX_DEFER case
drm/amd/display: Copy AUX read reply data whenever length > 0
drm/amdgpu/hdp4: use memcfg register to post the write for HDP flush
drm/amdgpu/hdp5.2: use memcfg register to post the write for HDP flush
drm/amdgpu/hdp5: use memcfg register to post the write for HDP flush
drm/amdgpu/hdp6: use memcfg register to post the write for HDP flush
drm/amdgpu/hdp7: use memcfg register to post the write for HDP flush
usb: uhci-platform: Make the clock really optional
smb: client: Avoid race in open_cached_dir with lease breaks
xen: swiotlb: Use swiotlb bouncing if kmalloc allocation demands it
xenbus: Use kref to track req lifetime
accel/ivpu: Increase state dump msg timeout
arm64: cpufeature: Move arm64_use_ng_mappings to the .data section to prevent wrong idmap generation
clocksource/i8253: Use raw_spinlock_irqsave() in clockevent_i8253_disable()
memblock: Accept allocated memory before use in memblock_double_array()
module: ensure that kobject_put() is safe for module type kobjects
x86/microcode: Consolidate the loader enablement checking
ocfs2: fix the issue with discontiguous allocation in the global_bitmap
ocfs2: switch osb->disable_recovery to enum
ocfs2: implement handshaking with ocfs2 recovery thread
ocfs2: stop quota recovery before disabling quotas
usb: dwc3: gadget: Make gadget_wakeup asynchronous
usb: cdnsp: Fix issue with resuming from L1
usb: cdnsp: fix L1 resume issue for RTL_REVISION_NEW_LPM version
usb: gadget: f_ecm: Add get_status callback
usb: gadget: tegra-xudc: ACK ST_RC after clearing CTRL_RUN
usb: gadget: Use get_status callback to set remote wakeup capability
usb: host: tegra: Prevent host controller crash when OTG port is used
usb: misc: onboard_usb_dev: fix support for Cypress HX3 hubs
usb: typec: tcpm: delay SNK_TRY_WAIT_DEBOUNCE to SRC_TRYWAIT transition
usb: typec: ucsi: displayport: Fix NULL pointer access
USB: usbtmc: use interruptible sleep in usbtmc_read
usb: usbtmc: Fix erroneous get_stb ioctl error returns
usb: usbtmc: Fix erroneous wait_srq ioctl return
usb: usbtmc: Fix erroneous generic_read ioctl return
iio: accel: adxl367: fix setting odr for activity time update
iio: temp: maxim-thermocouple: Fix potential lack of DMA safe buffer.
types: Complement the aligned types with signed 64-bit one
iio: accel: adxl355: Make timestamp 64-bit aligned using aligned_s64
iio: adc: dln2: Use aligned_s64 for timestamp
MIPS: Fix MAX_REG_OFFSET
riscv: misaligned: Add handling for ZCB instructions
loop: Use bdev limit helpers for configuring discard
loop: Simplify discard granularity calc
loop: Fix ABBA locking race
loop: refactor queue limits updates
loop: factor out a loop_assign_backing_file helper
loop: Add sanity check for read/write_iter
drm/panel: simple: Update timings for AUO G101EVN010
nvme: unblock ctrl state transition for firmware update
riscv: misaligned: factorize trap handling
riscv: misaligned: enable IRQs while handling misaligned accesses
drm/xe/tests/mocs: Update xe_force_wake_get() return handling
drm/xe/tests/mocs: Hold XE_FORCEWAKE_ALL for LNCF regs
io_uring/sqpoll: Increase task_work submission batch size
do_umount(): add missing barrier before refcount checks in sync case
Revert "um: work around sched_yield not yielding in time-travel mode"
rust: allow Rust 1.87.0's `clippy::ptr_eq` lint
rust: clean Rust 1.88.0's `clippy::uninlined_format_args` lint
io_uring: always arm linked timeouts prior to issue
Bluetooth: btmtk: Remove resetting mt7921 before downloading the fw
Bluetooth: btmtk: Remove the resetting step before downloading the fw
mm: page_alloc: don't steal single pages from biggest buddy
mm: page_alloc: speed up fallbacks in rmqueue_bulk()
sched/eevdf: Fix se->slice being set to U64_MAX and resulting crash
arm64: insn: Add support for encoding DSB
arm64: proton-pack: Expose whether the platform is mitigated by firmware
arm64: proton-pack: Expose whether the branchy loop k value
arm64: bpf: Add BHB mitigation to the epilogue for cBPF programs
arm64: bpf: Only mitigate cBPF programs loaded by unprivileged users
arm64: proton-pack: Add new CPUs 'k' values for branch mitigation
x86/bpf: Call branch history clearing sequence on exit
x86/bpf: Add IBHF call at end of classic BPF
x86/bhi: Do not set BHI_DIS_S in 32-bit mode
x86/speculation: Simplify and make CALL_NOSPEC consistent
x86/speculation: Add a conditional CS prefix to CALL_NOSPEC
x86/speculation: Remove the extra #ifdef around CALL_NOSPEC
Documentation: x86/bugs/its: Add ITS documentation
x86/its: Enumerate Indirect Target Selection (ITS) bug
x86/its: Add support for ITS-safe indirect thunk
x86/its: Add support for ITS-safe return thunk
x86/its: Enable Indirect Target Selection mitigation
x86/its: Add "vmexit" option to skip mitigation on some CPUs
x86/its: Add support for RSB stuffing mitigation
x86/its: Align RETs in BHB clear sequence to avoid thunking
x86/ibt: Keep IBT disabled during alternative patching
x86/its: Use dynamic thunks for indirect branches
selftest/x86/bugs: Add selftests for ITS
x86/its: Fix build errors when CONFIG_MODULES=n
x86/its: FineIBT-paranoid vs ITS
Linux 6.12.29
Change-Id: I00ff9cc212474331d43028ec90a190dcd1dfa697
Signed-off-by: Greg Kroah-Hartman <gregkh@google.com>
commit e05741fb10c38d70bbd7ec12b23c197b6355d519 upstream.
__alloc_pages_slowpath has no change detection for ac->nodemask in the
part of retry path, while cpuset can modify it in parallel. For some
processes that set mempolicy as MPOL_BIND, this results ac->nodemask
changes, and then the should_reclaim_retry will judge based on the latest
nodemask and jump to retry, while the get_page_from_freelist only
traverses the zonelist from ac->preferred_zoneref, which selected by a
expired nodemask and may cause infinite retries in some cases
cpu 64:
__alloc_pages_slowpath {
/* ..... */
retry:
/* ac->nodemask = 0x1, ac->preferred->zone->nid = 1 */
if (alloc_flags & ALLOC_KSWAPD)
wake_all_kswapds(order, gfp_mask, ac);
/* cpu 1:
cpuset_write_resmask
update_nodemask
update_nodemasks_hier
update_tasks_nodemask
mpol_rebind_task
mpol_rebind_policy
mpol_rebind_nodemask
// mempolicy->nodes has been modified,
// which ac->nodemask point to
*/
/* ac->nodemask = 0x3, ac->preferred->zone->nid = 1 */
if (should_reclaim_retry(gfp_mask, order, ac, alloc_flags,
did_some_progress > 0, &no_progress_loops))
goto retry;
}
Simultaneously starting multiple cpuset01 from LTP can quickly reproduce
this issue on a multi node server when the maximum memory pressure is
reached and the swap is enabled
Link: https://lkml.kernel.org/r/20250416082405.20988-1-zhangtianyang@loongson.cn
Fixes: c33d6c06f6 ("mm, page_alloc: avoid looking up the first zone in a zonelist twice")
Signed-off-by: Tianyang Zhang <zhangtianyang@loongson.cn>
Reviewed-by: Suren Baghdasaryan <surenb@google.com>
Reviewed-by: Vlastimil Babka <vbabka@suse.cz>
Cc: Michal Hocko <mhocko@suse.com>
Cc: Brendan Jackman <jackmanb@google.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Zi Yan <ziy@nvidia.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
the pointer is not initialized in __alloc_pages_noprof.
And it will lead to panic when CONFIG_INIT_STACK_ALL_ZERO=n.
Because in the function flow, the pointer of the page is not
initialized, it becomes non-null, and the vendorhook may not
be executed. This leads to the non-null check being passed,
causing the out process to be skipped and the program continuing
with the subsequent steps.
Bug: 420376550
Change-Id: Ibd2eb2bd8a24e431ce90781bed3b72694c06bb42
Signed-off-by: Dezhen Wang <dezhen.wang@amlogic.com>
commit fefc075182275057ce607effaa3daa9e6e3bdc73 upstream.
The page allocator tracks the number of zones that have unaccepted memory
using static_branch_enc/dec() and uses that static branch in hot paths to
determine if it needs to deal with unaccepted memory.
Borislav and Thomas pointed out that the tracking is racy: operations on
static_branch are not serialized against adding/removing unaccepted pages
to/from the zone.
Sanity checks inside static_branch machinery detects it:
WARNING: CPU: 0 PID: 10 at kernel/jump_label.c:276 __static_key_slow_dec_cpuslocked+0x8e/0xa0
The comment around the WARN() explains the problem:
/*
* Warn about the '-1' case though; since that means a
* decrement is concurrent with a first (0->1) increment. IOW
* people are trying to disable something that wasn't yet fully
* enabled. This suggests an ordering problem on the user side.
*/
The effect of this static_branch optimization is only visible on
microbenchmark.
Instead of adding more complexity around it, remove it altogether.
Link: https://lkml.kernel.org/r/20250506133207.1009676-1-kirill.shutemov@linux.intel.com
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Fixes: dcdfdd40fa ("mm: Add support for unaccepted memory")
Link: https://lore.kernel.org/all/20250506092445.GBaBnVXXyvnazly6iF@fat_crate.local
Reported-by: Borislav Petkov <bp@alien8.de>
Tested-by: Borislav Petkov (AMD) <bp@alien8.de>
Reported-by: Thomas Gleixner <tglx@linutronix.de>
Cc: Vlastimil Babka <vbabka@suse.cz>
Cc: Suren Baghdasaryan <surenb@google.com>
Cc: Michal Hocko <mhocko@suse.com>
Cc: Brendan Jackman <jackmanb@google.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: <stable@vger.kernel.org> [6.5+]
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 90abee6d7895d5eef18c91d870d8168be4e76e9d upstream.
The test robot identified c2f6ea38fc1b ("mm: page_alloc: don't steal
single pages from biggest buddy") as the root cause of a 56.4% regression
in vm-scalability::lru-file-mmap-read.
Carlos reports an earlier patch, c0cd6f557b ("mm: page_alloc: fix
freelist movement during block conversion"), as the root cause for a
regression in worst-case zone->lock+irqoff hold times.
Both of these patches modify the page allocator's fallback path to be less
greedy in an effort to stave off fragmentation. The flip side of this is
that fallbacks are also less productive each time around, which means the
fallback search can run much more frequently.
Carlos' traces point to rmqueue_bulk() specifically, which tries to refill
the percpu cache by allocating a large batch of pages in a loop. It
highlights how once the native freelists are exhausted, the fallback code
first scans orders top-down for whole blocks to claim, then falls back to
a bottom-up search for the smallest buddy to steal. For the next batch
page, it goes through the same thing again.
This can be made more efficient. Since rmqueue_bulk() holds the
zone->lock over the entire batch, the freelists are not subject to outside
changes; when the search for a block to claim has already failed, there is
no point in trying again for the next page.
Modify __rmqueue() to remember the last successful fallback mode, and
restart directly from there on the next rmqueue_bulk() iteration.
Oliver confirms that this improves beyond the regression that the test
robot reported against c2f6ea38fc1b:
commit:
f3b92176f4 ("tools/selftests: add guard region test for /proc/$pid/pagemap")
c2f6ea38fc ("mm: page_alloc: don't steal single pages from biggest buddy")
acc4d5ff0b ("Merge tag 'net-6.15-rc0' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net")
2c847f27c3 ("mm: page_alloc: speed up fallbacks in rmqueue_bulk()") <--- your patch
f3b92176f4f7100f c2f6ea38fc1b640aa7a2e155cc1 acc4d5ff0b61eb1715c498b6536 2c847f27c37da65a93d23c237c5
---------------- --------------------------- --------------------------- ---------------------------
%stddev %change %stddev %change %stddev %change %stddev
\ | \ | \ | \
25525364 ± 3% -56.4% 11135467 -57.8% 10779336 +31.6% 33581409 vm-scalability.throughput
Carlos confirms that worst-case times are almost fully recovered
compared to before the earlier culprit patch:
2dd482ba62 (before freelist hygiene): 1ms
c0cd6f557b (after freelist hygiene): 90ms
next-20250319 (steal smallest buddy): 280ms
this patch : 8ms
[jackmanb@google.com: comment updates]
Link: https://lkml.kernel.org/r/D92AC0P9594X.3BML64MUKTF8Z@google.com
[hannes@cmpxchg.org: reset rmqueue_mode in rmqueue_buddy() error loop, per Yunsheng Lin]
Link: https://lkml.kernel.org/r/20250409140023.GA2313@cmpxchg.org
Link: https://lkml.kernel.org/r/20250407180154.63348-1-hannes@cmpxchg.org
Fixes: c0cd6f557b ("mm: page_alloc: fix freelist movement during block conversion")
Fixes: c2f6ea38fc1b ("mm: page_alloc: don't steal single pages from biggest buddy")
Signed-off-by: Johannes Weiner <hannes@cmpxchg.org>
Signed-off-by: Brendan Jackman <jackmanb@google.com>
Reported-by: kernel test robot <oliver.sang@intel.com>
Reported-by: Carlos Song <carlos.song@nxp.com>
Tested-by: Carlos Song <carlos.song@nxp.com>
Tested-by: kernel test robot <oliver.sang@intel.com>
Closes: https://lore.kernel.org/oe-lkp/202503271547.fc08b188-lkp@intel.com
Reviewed-by: Brendan Jackman <jackmanb@google.com>
Tested-by: Shivank Garg <shivankg@amd.com>
Acked-by: Zi Yan <ziy@nvidia.com>
Reviewed-by: Vlastimil Babka <vbabka@suse.cz>
Cc: <stable@vger.kernel.org> [6.10+]
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Johannes Weiner <hannes@cmpxchg.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit c2f6ea38fc1b640aa7a2e155cc1c0410ff91afa2 upstream.
The fallback code searches for the biggest buddy first in an attempt to
steal the whole block and encourage type grouping down the line.
The approach used to be this:
- Non-movable requests will split the largest buddy and steal the
remainder. This splits up contiguity, but it allows subsequent
requests of this type to fall back into adjacent space.
- Movable requests go and look for the smallest buddy instead. The
thinking is that movable requests can be compacted, so grouping is
less important than retaining contiguity.
c0cd6f557b ("mm: page_alloc: fix freelist movement during block
conversion") enforces freelist type hygiene, which restricts stealing to
either claiming the whole block or just taking the requested chunk; no
additional pages or buddy remainders can be stolen any more.
The patch mishandled when to switch to finding the smallest buddy in that
new reality. As a result, it may steal the exact request size, but from
the biggest buddy. This causes fracturing for no good reason.
Fix this by committing to the new behavior: either steal the whole block,
or fall back to the smallest buddy.
Remove single-page stealing from steal_suitable_fallback(). Rename it to
try_to_steal_block() to make the intentions clear. If this fails, always
fall back to the smallest buddy.
The following is from 4 runs of mmtest's thpchallenge. "Pollute" is
single page fallback, "steal" is conversion of a partially used block.
The numbers for free block conversions (omitted) are comparable.
vanilla patched
@pollute[unmovable from reclaimable]: 27 106
@pollute[unmovable from movable]: 82 46
@pollute[reclaimable from unmovable]: 256 83
@pollute[reclaimable from movable]: 46 8
@pollute[movable from unmovable]: 4841 868
@pollute[movable from reclaimable]: 5278 12568
@steal[unmovable from reclaimable]: 11 12
@steal[unmovable from movable]: 113 49
@steal[reclaimable from unmovable]: 19 34
@steal[reclaimable from movable]: 47 21
@steal[movable from unmovable]: 250 183
@steal[movable from reclaimable]: 81 93
The allocator appears to do a better job at keeping stealing and polluting
to the first fallback preference. As a result, the numbers for "from
movable" - the least preferred fallback option, and most detrimental to
compactability - are down across the board.
Link: https://lkml.kernel.org/r/20250225001023.1494422-2-hannes@cmpxchg.org
Fixes: c0cd6f557b ("mm: page_alloc: fix freelist movement during block conversion")
Signed-off-by: Johannes Weiner <hannes@cmpxchg.org>
Suggested-by: Vlastimil Babka <vbabka@suse.cz>
Reviewed-by: Brendan Jackman <jackmanb@google.com>
Reviewed-by: Vlastimil Babka <vbabka@suse.cz>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Johannes Weiner <hannes@cmpxchg.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
This patch add a vendor hook to override batch size passed
in rmqueue_bulk, which is used to mitigate issues from long
local IRQ off.
Bug: 416197949
Change-Id: I2ff642ed2afa1a5c3f5fe1111446aa6e768f0b64
Signed-off-by: Chinwen Chang <chinwen.chang@mediatek.com>
Reverse migration is used to do the balancing the occupancy of memory
zones in a node in the system whose imabalance may be caused by
migration of pages to other zones by an operation, eg: hotremove and
then hotadding the same memory. In this case there is a lot of free
memory in newly hotadd memory which can be filled up by the previous
migrated pages(as part of offline/hotremove) thus may free up some
pressure in other zones of the node.
Upstream discussion: https://lore.kernel.org/all/ee78c83d-da9b-f6d1-4f66-934b7782acfb@codeaurora.org/
Port to ACK6.6: Fold in following simple fixes from ACK6.1:
commit d0652b3a8559 ("ANDROID: inline isolate_and_split_free_page")
commit 8a98feb47ec0 ("ANDROID: mm: compaction: fix isolate_and_split_free_page() redefinition")
Port to ACK6.12: Make below changes to patch:
1. Move isolate_and_split_free_page() into mm/page_alloc to be able to
make use of split_free_pages() [1]. The @list argument is also
now an array of lists based on page order on account of switching to the
split_free_pages API.
2. isolate_lru_page is also deprecated. Hence isolate_anon_lru_page() is
now using folio_isolate_lru() which is the folio equivalent.
[1] https://lore.kernel.org/all/20240814035451.773331-1-yuzhao@google.com/T/#u
Bug: 201263307
Signed-off-by: Charan Teja Reddy <quic_charante@quicinc.com>
Signed-off-by: Sukadev Bhattiprolu <quic_sukadev@quicinc.com>
Change-Id: Ib3137dab0db66ecf6858c4077dcadb9dfd0c6b1c
This vendor hook enables or disables updating the LMKD zone watermark level.
Bug: 396115949
Test: build
Change-Id: I0089a0586821120e47c46e08bcfea11a1602d516
Signed-off-by: Martin Liu <liumartin@google.com>
To monitor the efficiency of memory relciaming in
__alloc_pages_slowpath, add vendor hooks in each
stages of __alloc_pages_slowpath including
__alloc_pages_may_oom and __alloc_pages_direct_reclaim.
android_vh_mm_direct_reclaim_enter()
android_vh_mm_direct_reclaim_exit(unsigned long, int)
android_vh_mm_may_oom_exit(struct oom_control *, unsigned long)
Bug: 301044280
Bug: 330661608
Bug: 411525596
Change-Id: Ic5b5f1c2ad31b16e7339f539fcf54659e9acaba7
Signed-off-by: John Hsu <john.hsu@mediatek.com>
Signed-off-by: Chinwen Chang <chinwen.chang@mediatek.com>
GKI (arm64) relevant 48 out of 271 changes, affecting 92 files +576/-223
5b414ed3bb Revert "of: reserved-memory: Fix using wrong number of cells to get property 'alignment'" [1 file, +2/-2]
48a934fc47 Revert "mm/page_alloc.c: don't show protection in zone's ->lowmem_reserve[] for empty zone" [1 file, +1/-2]
88310caff6 Bluetooth: Add check for mgmt_alloc_skb() in mgmt_remote_name() [1 file, +2/-0]
7841180342 Bluetooth: Add check for mgmt_alloc_skb() in mgmt_device_connected() [1 file, +3/-0]
2d448dbd47 userfaultfd: do not block on locking a large folio with raised refcount [1 file, +16/-1]
f57e89c1cb block: fix conversion of GPT partition name to 7-bit [1 file, +1/-1]
9426f38372 mm/page_alloc: fix uninitialized variable [1 file, +1/-0]
79636d2981 mm: abort vma_modify() on merge out of memory failure [1 file, +8/-4]
605f53f13b mm: don't skip arch_sync_kernel_mappings() in error paths [2 files, +6/-4]
9ed33c7bac mm: fix finish_fault() handling for large folios [1 file, +10/-5]
576a2f4c43 hwpoison, memory_hotplug: lock folio before unmap hwpoisoned folio [1 file, +4/-1]
2e66d69941 mm: memory-hotplug: check folio ref count first in do_migrate_range [1 file, +7/-13]
3c63fb6ef7 nvme-pci: use sgls for all user requests if possible [2 files, +13/-4]
9dedafd86e nvme-ioctl: fix leaked requests on mapping error [1 file, +8/-4]
084819b0d8 net: gso: fix ownership in __udp_gso_segment [1 file, +6/-2]
1688acf477 perf/core: Fix pmus_lock vs. pmus_srcu ordering [1 file, +2/-2]
a899adf706 HID: hid-steam: Fix use-after-free when detaching device [1 file, +1/-1]
8aa8a40c76 ppp: Fix KMSAN uninit-value warning with bpf [1 file, +19/-9]
b71cd95764 ethtool: linkstate: migrate linkstate functions to support multi-PHY setups [1 file, +15/-8]
9c1d09cdbc net: ethtool: plumb PHY stats to PHY drivers [7 files, +167/-2]
639c703529 net: ethtool: netlink: Allow NULL nlattrs when getting a phy_device [9 files, +19/-18]
30e8aee778 vlan: enforce underlying device type [1 file, +2/-1]
5d609f0d2f exfat: fix just enough dentries but allocate a new cluster to dir [1 file, +1/-1]
c897b8ec46 exfat: fix soft lockup in exfat_clear_bitmap [3 files, +16/-7]
611015122d exfat: short-circuit zero-byte writes in exfat_file_write_iter [1 file, +1/-1]
2b484789e9 net-timestamp: support TCP GSO case for a few missing flags [1 file, +7/-4]
b08e290324 ublk: set_params: properly check if parameters can be applied [1 file, +5/-2]
b5741e4b9e sched/fair: Fix potential memory corruption in child_cfs_rq_on_list [1 file, +4/-2]
39c2b2767e xhci: Restrict USB4 tunnel detection for USB3 devices to Intel hosts [1 file, +8/-0]
4ea3319f3e usb: hub: lack of clearing xHC resources [1 file, +33/-0]
0cab185c73 usb: quirks: Add DELAY_INIT and NO_LPM for Prolific Mass Storage Card Reader [1 file, +4/-0]
079a3e52f3 usb: typec: ucsi: Fix NULL pointer access [1 file, +7/-6]
840afbea3f usb: gadget: u_ether: Set is_suspend flag if remote wakeup fails [1 file, +2/-2]
ced69d88eb usb: dwc3: Set SUSPENDENABLE soon after phy init [3 files, +45/-30]
35db1f1829 usb: dwc3: gadget: Prevent irq storm when TH re-executes [2 files, +13/-13]
b387312527 usb: typec: ucsi: increase timeout for PPM reset operations [1 file, +1/-1]
4bf6c57a89 usb: gadget: Set self-powered based on MaxPower and bmAttributes [1 file, +11/-5]
dcd7ffdefb usb: gadget: Fix setting self-powered state on suspend [1 file, +2/-1]
395011ee82 usb: gadget: Check bmAttributes only if configuration is valid [1 file, +1/-1]
012b98cdb5 acpi: typec: ucsi: Introduce a ->poll_cci method [7 files, +25/-12]
d7015bb3c5 xhci: pci: Fix indentation in the PCI device ID definitions [1 file, +4/-4]
ea39f99864 usb: xhci: Enable the TRB overfetch quirk on VIA VL805 [3 files, +10/-5]
4e8df56636 char: misc: deallocate static minor in error path [1 file, +1/-1]
b50e18791f drivers: core: fix device leak in __fw_devlink_relax_cycles() [1 file, +1/-0]
a684bad77e mm: hugetlb: Add huge page size param to huge_ptep_get_and_clear() [16 files, +46/-28]
6ad9643aa5 fs/netfs/read_pgpriv2: skip folio queues without `marks3` [1 file, +3/-2]
5bc6e5b10f fs/netfs/read_collect: fix crash due to uninitialized `prev` variable [1 file, +11/-10]
86b7ebddab uprobes: Fix race in uprobe_free_utask [1 file, +1/-1]
Changes in 6.12.19
x86/amd_nb: Use rdmsr_safe() in amd_get_mmconfig_range()
rust: block: fix formatting in GenDisk doc
drm/i915/dsi: convert to struct intel_display
drm/i915/dsi: Use TRANS_DDI_FUNC_CTL's own port width macro
gpio: vf610: use generic device_get_match_data()
gpio: vf610: add locking to gpio direction functions
cifs: Remove symlink member from cifs_open_info_data union
smb311: failure to open files of length 1040 when mounting with SMB3.1.1 POSIX extensions
btrfs: fix data overwriting bug during buffered write when block size < page size
x86/microcode/AMD: Add some forgotten models to the SHA check
loongarch: Use ASM_REACHABLE
rust: workqueue: remove unneeded ``#[allow(clippy::new_ret_no_self)]`
rust: sort global Rust flags
rust: types: avoid repetition in `{As,From}Bytes` impls
rust: enable `clippy::undocumented_unsafe_blocks` lint
rust: enable `clippy::unnecessary_safety_comment` lint
rust: enable `clippy::unnecessary_safety_doc` lint
rust: enable `clippy::ignored_unit_patterns` lint
rust: enable `rustdoc::unescaped_backticks` lint
rust: init: remove unneeded `#[allow(clippy::disallowed_names)]`
rust: sync: remove unneeded `#[allow(clippy::non_send_fields_in_send_ty)]`
rust: introduce `.clippy.toml`
rust: replace `clippy::dbg_macro` with `disallowed_macros`
rust: provide proper code documentation titles
rust: enable Clippy's `check-private-items`
Documentation: rust: add coding guidelines on lints
rust: start using the `#[expect(...)]` attribute
Documentation: rust: discuss `#[expect(...)]` in the guidelines
rust: error: make conversion functions public
rust: error: optimize error type to use nonzero
rust: alloc: add `Allocator` trait
rust: alloc: separate `aligned_size` from `krealloc_aligned`
rust: alloc: rename `KernelAllocator` to `Kmalloc`
rust: alloc: implement `ReallocFunc`
rust: alloc: make `allocator` module public
rust: alloc: implement `Allocator` for `Kmalloc`
rust: alloc: add module `allocator_test`
rust: alloc: implement `Vmalloc` allocator
rust: alloc: implement `KVmalloc` allocator
rust: alloc: add __GFP_NOWARN to `Flags`
rust: alloc: implement kernel `Box`
rust: treewide: switch to our kernel `Box` type
rust: alloc: remove extension of std's `Box`
rust: alloc: add `Box` to prelude
rust: alloc: introduce `ArrayLayout`
rust: alloc: implement kernel `Vec` type
rust: alloc: implement `IntoIterator` for `Vec`
rust: alloc: implement `collect` for `IntoIter`
rust: treewide: switch to the kernel `Vec` type
rust: alloc: remove `VecExt` extension
rust: alloc: add `Vec` to prelude
rust: error: use `core::alloc::LayoutError`
rust: error: check for config `test` in `Error::name`
rust: alloc: implement `contains` for `Flags`
rust: alloc: implement `Cmalloc` in module allocator_test
rust: str: test: replace `alloc::format`
rust: alloc: update module comment of alloc.rs
kbuild: rust: remove the `alloc` crate and `GlobalAlloc`
MAINTAINERS: add entry for the Rust `alloc` module
drm/panic: avoid reimplementing Iterator::find
drm/panic: remove unnecessary borrow in alignment_pattern
drm/panic: prefer eliding lifetimes
drm/panic: remove redundant field when assigning value
drm/panic: correctly indent continuation of line in list item
drm/panic: allow verbose boolean for clarity
drm/panic: allow verbose version check
rust: kbuild: expand rusttest target for macros
rust: fix size_t in bindgen prototypes of C builtins
rust: map `__kernel_size_t` and friends also to usize/isize
rust: use custom FFI integer types
rust: alloc: Fix `ArrayLayout` allocations
Revert "of: reserved-memory: Fix using wrong number of cells to get property 'alignment'"
tracing: tprobe-events: Fix a memory leak when tprobe with $retval
tracing: tprobe-events: Reject invalid tracepoint name
stmmac: loongson: Pass correct arg to PCI function
LoongArch: Convert unreachable() to BUG()
LoongArch: Use polling play_dead() when resuming from hibernation
LoongArch: Set max_pfn with the PFN of the last page
LoongArch: KVM: Add interrupt checking for AVEC
LoongArch: KVM: Reload guest CSR registers after sleep
LoongArch: KVM: Fix GPA size issue about VM
HID: appleir: Fix potential NULL dereference at raw event handle
ksmbd: fix type confusion via race condition when using ipc_msg_send_request
ksmbd: fix out-of-bounds in parse_sec_desc()
ksmbd: fix use-after-free in smb2_lock
ksmbd: fix bug on trap in smb2_lock
gpio: rcar: Use raw_spinlock to protect register access
gpio: aggregator: protect driver attr handlers against module unload
ALSA: seq: Avoid module auto-load handling at event delivery
ALSA: hda: intel: Add Dell ALC3271 to power_save denylist
ALSA: hda/realtek - add supported Mic Mute LED for Lenovo platform
ALSA: hda/realtek: update ALC222 depop optimize
btrfs: fix a leaked chunk map issue in read_one_chunk()
hwmon: (peci/dimmtemp) Do not provide fake thresholds data
drm/amd/display: Fix null check for pipe_ctx->plane_state in resource_build_scaling_params
drm/amdkfd: Fix NULL Pointer Dereference in KFD queue
drm/amd/pm: always allow ih interrupt from fw
drm/imagination: avoid deadlock on fence release
drm/imagination: Hold drm_gem_gpuva lock for unmap
drm/imagination: only init job done fences once
drm/radeon: Fix rs400_gpu_init for ATI mobility radeon Xpress 200M
Revert "mm/page_alloc.c: don't show protection in zone's ->lowmem_reserve[] for empty zone"
Revert "selftests/mm: remove local __NR_* definitions"
platform/x86: thinkpad_acpi: Add battery quirk for ThinkPad X131e
x86/boot: Sanitize boot params before parsing command line
x86/cacheinfo: Validate CPUID leaf 0x2 EDX output
x86/cpu: Validate CPUID leaf 0x2 EDX output
x86/cpu: Properly parse CPUID leaf 0x2 TLB descriptor 0x63
drm/xe: Add staging tree for VM binds
drm/xe/hmm: Style- and include fixes
drm/xe/hmm: Don't dereference struct page pointers without notifier lock
drm/xe/vm: Fix a misplaced #endif
drm/xe/vm: Validate userptr during gpu vma prefetching
mptcp: fix 'scheduling while atomic' in mptcp_pm_nl_append_new_local_addr
drm/xe: Fix GT "for each engine" workarounds
drm/xe: Fix fault mode invalidation with unbind
drm/xe/userptr: properly setup pfn_flags_mask
drm/xe/userptr: Unmap userptrs in the mmu notifier
Bluetooth: Add check for mgmt_alloc_skb() in mgmt_remote_name()
Bluetooth: Add check for mgmt_alloc_skb() in mgmt_device_connected()
wifi: cfg80211: regulatory: improve invalid hints checking
wifi: nl80211: reject cooked mode if it is set along with other flags
selftests/damon/damos_quota_goal: handle minimum quota that cannot be further reduced
selftests/damon/damos_quota: make real expectation of quota exceeds
selftests/damon/damon_nr_regions: set ops update for merge results check to 100ms
selftests/damon/damon_nr_regions: sort collected regiosn before checking with min/max boundaries
rapidio: add check for rio_add_net() in rio_scan_alloc_net()
rapidio: fix an API misues when rio_add_net() fails
dma: kmsan: export kmsan_handle_dma() for modules
s390/traps: Fix test_monitor_call() inline assembly
NFS: fix nfs_release_folio() to not deadlock via kcompactd writeback
userfaultfd: do not block on locking a large folio with raised refcount
block: fix conversion of GPT partition name to 7-bit
mm/page_alloc: fix uninitialized variable
mm: abort vma_modify() on merge out of memory failure
mm: memory-failure: update ttu flag inside unmap_poisoned_folio
mm: don't skip arch_sync_kernel_mappings() in error paths
mm: fix finish_fault() handling for large folios
hwpoison, memory_hotplug: lock folio before unmap hwpoisoned folio
mm: memory-hotplug: check folio ref count first in do_migrate_range
wifi: iwlwifi: mvm: clean up ROC on failure
wifi: iwlwifi: mvm: don't try to talk to a dead firmware
wifi: iwlwifi: limit printed string from FW file
wifi: iwlwifi: Free pages allocated when failing to build A-MSDU
wifi: iwlwifi: Fix A-MSDU TSO preparation
HID: google: fix unused variable warning under !CONFIG_ACPI
HID: intel-ish-hid: Fix use-after-free issue in hid_ishtp_cl_remove()
HID: intel-ish-hid: Fix use-after-free issue in ishtp_hid_remove()
coredump: Only sort VMAs when core_sort_vma sysctl is set
nvme-pci: add support for sgl metadata
nvme-pci: use sgls for all user requests if possible
nvme-ioctl: fix leaked requests on mapping error
wifi: mac80211: Support parsing EPCS ML element
wifi: mac80211: fix MLE non-inheritance parsing
wifi: mac80211: fix vendor-specific inheritance
drm/fbdev-helper: Move color-mode lookup into 4CC format helper
drm/fbdev: Add memory-agnostic fbdev client
drm: Add client-agnostic setup helper
drm/fbdev-ttm: Support struct drm_driver.fbdev_probe
drm/nouveau: Run DRM default client setup
drm/nouveau: select FW caching
bluetooth: btusb: Initialize .owner field of force_poll_sync_fops
nvme-tcp: add basic support for the C2HTermReq PDU
nvme-tcp: fix potential memory corruption in nvme_tcp_recv_pdu()
nvmet-tcp: Fix a possible sporadic response drops in weakly ordered arch
ALSA: hda/realtek: Remove (revert) duplicate Ally X config
net: gso: fix ownership in __udp_gso_segment
caif_virtio: fix wrong pointer check in cfv_probe()
perf/core: Fix pmus_lock vs. pmus_srcu ordering
hwmon: (pmbus) Initialise page count in pmbus_identify()
hwmon: (ntc_thermistor) Fix the ncpXXxh103 sensor table
hwmon: (ad7314) Validate leading zero bits and return error
tracing: probe-events: Remove unused MAX_ARG_BUF_LEN macro
drm/imagination: Fix timestamps in firmware traces
ALSA: usx2y: validate nrpacks module parameter on probe
llc: do not use skb_get() before dev_queue_xmit()
hwmon: fix a NULL vs IS_ERR_OR_NULL() check in xgene_hwmon_probe()
drm/sched: Fix preprocessor guard
be2net: fix sleeping while atomic bugs in be_ndo_bridge_getlink
net: hns3: make sure ptp clock is unregister and freed if hclge_ptp_get_cycle returns an error
drm/i915/color: Extract intel_color_modeset()
drm/i915: Plumb 'dsb' all way to the plane hooks
drm/xe: Remove double pageflip
HID: hid-steam: Fix use-after-free when detaching device
net: ipa: Fix v4.7 resource group names
net: ipa: Fix QSB data for v4.7
net: ipa: Enable checksum for IPA_ENDPOINT_AP_MODEM_{RX,TX} for v4.7
ppp: Fix KMSAN uninit-value warning with bpf
ethtool: linkstate: migrate linkstate functions to support multi-PHY setups
net: ethtool: plumb PHY stats to PHY drivers
net: ethtool: netlink: Allow NULL nlattrs when getting a phy_device
vlan: enforce underlying device type
x86/sgx: Fix size overflows in sgx_encl_create()
exfat: fix just enough dentries but allocate a new cluster to dir
exfat: fix soft lockup in exfat_clear_bitmap
exfat: short-circuit zero-byte writes in exfat_file_write_iter
net-timestamp: support TCP GSO case for a few missing flags
ublk: set_params: properly check if parameters can be applied
sched/fair: Fix potential memory corruption in child_cfs_rq_on_list
nvme-tcp: fix signedness bug in nvme_tcp_init_connection()
net: dsa: mt7530: Fix traffic flooding for MMIO devices
mctp i3c: handle NULL header address
net: ipv6: fix dst ref loop in ila lwtunnel
net: ipv6: fix missing dst ref drop in ila lwtunnel
gpio: rcar: Fix missing of_node_put() call
Revert "drivers/card_reader/rtsx_usb: Restore interrupt based detection"
usb: renesas_usbhs: Call clk_put()
xhci: Restrict USB4 tunnel detection for USB3 devices to Intel hosts
usb: renesas_usbhs: Use devm_usb_get_phy()
usb: hub: lack of clearing xHC resources
usb: quirks: Add DELAY_INIT and NO_LPM for Prolific Mass Storage Card Reader
usb: typec: ucsi: Fix NULL pointer access
usb: renesas_usbhs: Flush the notify_hotplug_work
usb: gadget: u_ether: Set is_suspend flag if remote wakeup fails
usb: atm: cxacru: fix a flaw in existing endpoint checks
usb: dwc3: Set SUSPENDENABLE soon after phy init
usb: dwc3: gadget: Prevent irq storm when TH re-executes
usb: typec: ucsi: increase timeout for PPM reset operations
usb: typec: tcpci_rt1711h: Unmask alert interrupts to fix functionality
usb: gadget: Set self-powered based on MaxPower and bmAttributes
usb: gadget: Fix setting self-powered state on suspend
usb: gadget: Check bmAttributes only if configuration is valid
kbuild: userprogs: use correct lld when linking through clang
acpi: typec: ucsi: Introduce a ->poll_cci method
rust: finish using custom FFI integer types
rust: map `long` to `isize` and `char` to `u8`
xhci: pci: Fix indentation in the PCI device ID definitions
usb: xhci: Enable the TRB overfetch quirk on VIA VL805
KVM: SVM: Set RFLAGS.IF=1 in C code, to get VMRUN out of the STI shadow
KVM: SVM: Save host DR masks on CPUs with DebugSwap
KVM: SVM: Drop DEBUGCTL[5:2] from guest's effective value
KVM: SVM: Suppress DEBUGCTL.BTF on AMD
KVM: x86: Snapshot the host's DEBUGCTL in common x86
KVM: SVM: Manually context switch DEBUGCTL if LBR virtualization is disabled
KVM: x86: Snapshot the host's DEBUGCTL after disabling IRQs
KVM: x86: Explicitly zero EAX and EBX when PERFMON_V2 isn't supported by KVM
cdx: Fix possible UAF error in driver_override_show()
mei: me: add panther lake P DID
mei: vsc: Use "wakeuphostint" when getting the host wakeup GPIO
intel_th: pci: Add Arrow Lake support
intel_th: pci: Add Panther Lake-H support
intel_th: pci: Add Panther Lake-P/U support
char: misc: deallocate static minor in error path
drivers: core: fix device leak in __fw_devlink_relax_cycles()
slimbus: messaging: Free transaction ID in delayed interrupt scenario
bus: mhi: host: pci_generic: Use pci_try_reset_function() to avoid deadlock
eeprom: digsy_mtc: Make GPIO lookup table match the device
drivers: virt: acrn: hsm: Use kzalloc to avoid info leak in pmcmd_ioctl
iio: filter: admv8818: Force initialization of SDO
iio: light: apds9306: fix max_scale_nano values
iio: dac: ad3552r: clear reset status flag
iio: adc: ad7192: fix channel select
iio: adc: at91-sama5d2_adc: fix sama7g5 realbits value
mm: hugetlb: Add huge page size param to huge_ptep_get_and_clear()
arm64: hugetlb: Fix huge_ptep_get_and_clear() for non-present ptes
fs/netfs/read_pgpriv2: skip folio queues without `marks3`
fs/netfs/read_collect: fix crash due to uninitialized `prev` variable
kbuild: hdrcheck: fix cross build with clang
ALSA: hda: realtek: fix incorrect IS_REACHABLE() usage
nvme-tcp: Fix a C2HTermReq error message
docs: rust: remove spurious item in `expect` list
Revert "KVM: e500: always restore irqs"
Revert "KVM: PPC: e500: Use __kvm_faultin_pfn() to handle page faults"
Revert "KVM: PPC: e500: Mark "struct page" pfn accessed before dropping mmu_lock"
Revert "KVM: PPC: e500: Mark "struct page" dirty in kvmppc_e500_shadow_map()"
KVM: e500: always restore irqs
uprobes: Fix race in uprobe_free_utask
selftests/bpf: Clean up open-coded gettid syscall invocations
x86/mm: Don't disable PCID when INVLPG has been fixed by microcode
wifi: iwlwifi: pcie: Fix TSO preparation
Linux 6.12.19
Change-Id: Ia0c2b2c6a95b53a66e21505ed6ba756c6b0a2388
Signed-off-by: Greg Kroah-Hartman <gregkh@google.com>
During zone initialization adjust a more suitable watermark for Customize, which can optimize kswapd load and system memory status.
Bug: 364225965
Change-Id: Ib6d94ec244f504a1c474657833a5f7558ea5d00b
Signed-off-by: Justin Jiang <justinjiang@vivo.corp-partner.google.com>
Signed-off-by: wengle <wengle@oppo.com>
Previously, I've added a vendor hook not to use certain type of
pageblocks as migration target with below commit.
https://android-review.googlesource.com/c/kernel/common/+/3475721
For this, export get_pfnblock_flags_mask() function for vendor module
to get migratetype of pageblock.
Bug: 410432482
Change-Id: Ia2d065927ab7288cd7367c41bdd9fb116ba995b0
Signed-off-by: Sooyong Suk <s.suk@samsung.corp-partner.google.com>
When allocating memory, we can adjust the water level to open a fast
allocation channel for some important foreground processes, while
allowing less important background processes to wait.
Bug: 405889864
Change-Id: I71d709c8c782aa614d2c9e1d3150ce14a0106acb
Signed-off-by: Qianfeng Rong <rongqianfeng@vivo.corp-partner.google.com>
add vendor hook for supporting customize alloc_gfp at alloc_page_slowpath,
give the ability to customize alloc_gfp according to customer specific requirement.
Bug: 337192903
Bug: 395197442
Change-Id: I42fd97fa76a47b9fde554434f607285bee8915c0
Signed-off-by: wei li <liwei1234@oppo.com>
Signed-off-by: wengle <wengle@oppo.com>
Add a vendor hook for costly order page counting
and other vendor specific functions.
Bug: 174521902
Bug: 172987241
Signed-off-by: Chiawei Wang <chiaweiwang@google.com>
Change-Id: I89206727a462548cc3500b695d85c83ff003eec7
Signed-off-by: Richard Chang <richardycc@google.com>
(cherry picked from commit 369de3780428a17e9afece2f5747f03619d589b6)
Signed-off-by: liangjlee <liangjlee@google.com>
(cherry picked from commit 86a45bb2dc315fd0b1d0b513b0f2e1be82d25f70)
Signed-off-by: Dmitry Skiba <dskiba@google.com>
This commit introduces a new trace event,
`mm_calculate_totalreserve_pages`, which reports the new reserve value at
the exact time when it takes effect.
The `totalreserve_pages` value represents the total amount of memory
reserved across all zones and nodes in the system. This reserved memory
is crucial for ensuring that critical kernel operations have access to
sufficient memory, even under memory pressure.
By tracing the `totalreserve_pages` value, developers can gain insights
that how the total reserved memory changes over time.
Link: https://lkml.kernel.org/r/20250308034606.2036033-4-liumartin@google.com
Signed-off-by: Martin Liu <liumartin@google.com>
Acked-by: David Rientjes <rientjes@google.com>
Cc: "Masami Hiramatsu (Google)" <mhiramat@kernel.org>
Cc: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Bug: 396115949
(cherry picked from commit 15766485e4a51bec2dcce304c089a95550720033
https://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm.git mm-unstable)
Change-Id: Iced6ea39ad8a36a50bf4393814b6bca2f64ac3b0
Signed-off-by: Martin Liu <liumartin@google.com>
This commit introduces the `mm_setup_per_zone_lowmem_reserve` trace
event,which provides detailed insights into the kernel's per-zone lowmem
reserve configuration.
The trace event provides precise timestamps, allowing developers to
1. Correlate lowmem reserve changes with specific kernel events and
able to diagnose unexpected kswapd or direct reclaim behavior triggered
by dynamic changes in lowmem reserve.
2. Know memory allocation failures that occur due to insufficient
lowmem reserve, by precisely correlating allocation attempts with
reserve adjustments.
Link: https://lkml.kernel.org/r/20250308034606.2036033-3-liumartin@google.com
Signed-off-by: Martin Liu <liumartin@google.com>
Acked-by: David Rientjes <rientjes@google.com>
Cc: "Masami Hiramatsu (Google)" <mhiramat@kernel.org>
Cc: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Bug: 396115949
(cherry picked from commit a293aba4a584709889f77a0ad0c45746aecf1b9f
https://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm.git mm-unstable)
Change-Id: I271fc260ec60645230681bf0afbcd10d84453c88
Signed-off-by: Martin Liu <liumartin@google.com>
Patch series "Add tracepoints for lowmem reserves, watermarks and
totalreserve_pages", v2.
This patchset introduces tracepoints to track changes in the lowmem
reserves, watermarks and totalreserve_pages. This helps to track
the exact timing of such changes and understand their relation to
reclaim activities.
The tracepoints added are:
mm_setup_per_zone_lowmem_reserve
mm_setup_per_zone_wmarks
mm_calculate_totalreserve_pagesi
This patch (of 3):
This commit introduces the `mm_setup_per_zone_wmarks` trace event,
which provides detailed insights into the kernel's per-zone watermark
configuration, offering precise timing and the ability to correlate
watermark changes with specific kernel events.
While `/proc/zoneinfo` provides some information about zone watermarks,
this trace event offers:
1. The ability to link watermark changes to specific kernel events and
logic.
2. The ability to capture rapid or short-lived changes in watermarks
that may be missed by user-space polling
3. Diagnosing unexpected kswapd activity or excessive direct reclaim
triggered by rapidly changing watermarks.
Link: https://lkml.kernel.org/r/20250308034606.2036033-1-liumartin@google.com
Link: https://lkml.kernel.org/r/20250308034606.2036033-2-liumartin@google.com
Signed-off-by: Martin Liu <liumartin@google.com>
Acked-by: David Rientjes <rientjes@google.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Martin Liu <liumartin@google.com>
Cc: "Masami Hiramatsu (Google)" <mhiramat@kernel.org>
Cc: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Bug: 396115949
(cherry picked from commit 8c02048d1c6126527f15752a5e0849dc49cefeeb
https://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm.git mm-unstable)
Change-Id: I7e326e78542abb6fa5f3ccbe5d61a59f42d7cf2f
Signed-off-by: Martin Liu <liumartin@google.com>
Provide a vendor hook to allow drain_all_pages to be skipped
during direct reclaim in some cases to avoid delays caused by
it in cases when the benefits of draining pcp lists are known
to be small.
Bug: 220811627
Bug: 234405962
Bug: 395197442
Signed-off-by: Liujie Xie <xieliujie@oppo.com>
Change-Id: I0805241f81e0a94afcf62c98e97cff125d4061e2
Signed-off-by: wengle <wengle@oppo.com>
We add these hooks to avoid key threads blocked in memory allocation
path.
-android_vh_free_unref_page_bypass ----We create a memory pool for the
key threads. This hook determines whether a page should be free to the
pool or to buddy freelist. It works with a existing hook
`android_vh_alloc_pages_reclaim_bypass`, which takes pages out of the
pool.
-android_vh_kvmalloc_node_use_vmalloc ----For key threads, we perfer
not to run into direct reclaim. So we clear __GFP_DIRECT_RECLAIM flag.
For threads which are not that important, we perfer use vmalloc.
-android_vh_should_alloc_pages_retry ----Before key threads run into
direct reclaim, we want to retry with a lower watermark.
-android_vh_unreserve_highatomic_bypass ----We want to keep more
highatomic pages when unreserve them to avoid highatomic allocation
failures.
-android_vh_rmqueue_bulk_bypass ----We found sometimes when key threads
run into rmqueue_bulk, it took several milliseconds spinning at
zone->lock or filling per-cpu pages. We use this hook to take pages from
the mempool mentioned above, rather than grab zone->lock and fill a
batch of pages to per-cpu.
Bug: 288216516
Bug: 395197442
Change-Id: I1656032d6819ca627723341987b6094775bc345f
Signed-off-by: Oven <liyangouwen1@oppo.com>
Signed-off-by: huzhanyuan <huzhanyuan@oppo.com>
Signed-off-by: wengle <wengle@oppo.com>
commit eae116d1f0449ade3269ca47a67432622f5c6438 upstream.
Commit 96a5c186ef ("mm/page_alloc.c: don't show protection in zone's
->lowmem_reserve[] for empty zone") removes the protection of lower zones
from allocations targeting memory-less high zones. This had an unintended
impact on the pattern of reclaims because it makes the high-zone-targeted
allocation more likely to succeed in lower zones, which adds pressure to
said zones. I.e, the following corresponding checks in
zone_watermark_ok/zone_watermark_fast are less likely to trigger:
if (free_pages <= min + z->lowmem_reserve[highest_zoneidx])
return false;
As a result, we are observing an increase in reclaim and kswapd scans, due
to the increased pressure. This was initially observed as increased
latency in filesystem operations when benchmarking with fio on a machine
with some memory-less zones, but it has since been associated with
increased contention in locks related to memory reclaim. By reverting
this patch, the original performance was recovered on that machine.
The original commit was introduced as a clarification of the
/proc/zoneinfo output, so it doesn't seem there are usecases depending on
it, making the revert a simple solution.
For reference, I collected vmstat with and without this patch on a freshly
booted system running intensive randread io from an nvme for 5 minutes. I
got:
rpm-6.12.0-slfo.1.2 -> pgscan_kswapd 5629543865
Patched -> pgscan_kswapd 33580844
33M scans is similar to what we had in kernels predating this patch.
These numbers is fairly representative of the workload on this machine, as
measured in several runs. So we are talking about a 2-order of magnitude
increase.
Link: https://lkml.kernel.org/r/20250226032258.234099-1-krisman@suse.de
Fixes: 96a5c186ef ("mm/page_alloc.c: don't show protection in zone's ->lowmem_reserve[] for empty zone")
Signed-off-by: Gabriel Krisman Bertazi <krisman@suse.de>
Reviewed-by: Vlastimil Babka <vbabka@suse.cz>
Acked-by: Michal Hocko <mhocko@suse.com>
Acked-by: Mel Gorman <mgorman@suse.de>
Cc: Baoquan He <bhe@redhat.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Some memory regions are already being tracked by previous patches. But
there are many memory allocations from memblock and frees to memblock
during the boot time.
This patch tracks the memblock size used for the common kernel. To to
this, tracking memblock size is disabled for some memory handling logics
like early param, device tree, and default cma size.
For precise kernel size, this patch counts not actually freed size to
buddy at boot time, and does not count freed size from ramdisk and init
section.
Additionally this patch does one important thing. This patch blocks
memblock_add_range of memblock_remove_range not to update memsize if
free pages were already released to the buddy allocator.
This is an example. The kernel size is newly added by this patch.
.kernel : 135137 KB
.unusable : 788073 KB
.reusable : 294912 KB
Bug: 340432773
Signed-off-by: Jaewon Kim <jaewon31.kim@samsung.com>
Link: https://lore.kernel.org/linux-mm/20240521023957.2587005-8-jaewon31.kim@samsung.com/
Change-Id: Ice0c2aa568385490a3468f7567b4b8c083e043ec
When restrict_cma_redirect boot parameter was introduced, its default
was set to follow the upstream behavior, which does not restrict any
movable allocation from using CMA. However this poses an issue when
partners upgrade from previous Android kernels and expect the same
behavior, which is to restrict CMA usage to movable allocations with
__GFP_CMA.
Change the default value of restrict_cma_redirect boot parameter to
keep backward compatibility with earlier ACK versions. Partners who
need upstream behavior will need to set restrict_cma_redirect=false
explicitly.
Bug: 399727765
Change-Id: Ia88008578557e38da54a455bc4ce3dc6f86fe52e
Signed-off-by: Suren Baghdasaryan <surenb@google.com>
For CMA allocation, it's really critical to migrate a page but
sometimes it fails. One of the reasons is some driver holds a
page refcount for a long time so VM couldn't migrate the page
at that time.
The concern here is there is no way to find the who hold the
refcount of the page effectively. This patch introduces feature
to keep tracking page's pinner. All get_page sites are vulnerable
to pin a page for a long time but the cost to keep track it would
be significat since get_page is the most frequent kernel operation.
Furthermore, the page could be not user page but kernel page which
is not related to the page migration failure.
Thus, this patch keeps tracks of only migration failed pages to
reduce runtime cost. Once page migration fails in CMA allocation
path, those pages are marked as "migration failure" and every
put_page operation against those pages, callstack of the put
are recorded into page_pinner buffer. Later, admin can see
what pages were failed and who released the refcount since the
failure. It really helps effectively to find out longtime refcount
holder to prevent the page migration.
note: page_pinner doesn't guarantee attributing/unattributing are
atomic if they happen at the same time. It's just best effort so
false-positive could happen.
Fix build issue during rebase, replace page_mapping() with
folio_mapping() and page_mapcount() with folio_mapcount().
Bug: 183414571
Bug: 240196534
Signed-off-by: Minchan Kim <minchan@kernel.org>
Signed-off-by: Minchan Kim <minchan@google.com>
Change-Id: I603d0c0122734c377db6b1eb95848a6f734173a0
(cherry picked from commit e6e6e1273db431fe03ae6b0bf48738970f6ccaea)
Signed-off-by: Georgi Djakov <quic_c_gdjako@quicinc.com>
Add vendor hook inside of get_page_from_freelist() to check
and modify the watermark in some special situations.
Additional page flag bit will be set for future identification.
Separately, a vendor hook inside of page_add_new_anon_rmap()
is added to set the referenced bit in some situations, e.g.
if the special bit in the page flag mentioned before is set,
we will give this page one more chance before it gets reclaimed.
Bug: 279793368
Bug: 336900791
Change-Id: I363853a050a87201f6f368ccc580485dddd6c6b6
Signed-off-by: Dezhi Huang <huangdezhi@hihonor.com>
Signed-off-by: huzhanyuan <huzhanyuan@oppo.com>
(cherry picked from commit 49fa9c5a7e11922caa132a0d014b46b7a72f10d3)
Hailong Liu
Minchan Kim
Pengfei Li
Sooyong Suk
Greg Kroah-Hartman
Yang Yang
Toke Høiland-Jørgensen
Johannes Weiner
Brendan Jackman
gaoxiang17
Tianyang Zhang
akash.tyagi
Dezhen Wang
Kirill A. Shutemov
Chinwen Chang
Charan Teja Reddy
Martin Liu
JohnHsu
Justin Jiang
Qianfeng Rong
wei li
Chiawei Wang
xiaofeng
Liujie Xie
Oven
Hao Zhang
Gabriel Krisman Bertazi
Jaewon Kim
Suren Baghdasaryan
Dezhi Huang