April 2025 - Linux-stable-mirror

by Greg Kroah-Hartman

This is the start of the stable review cycle for the 6.12.11 release. There are 122 patches in this series, all will be posted as a response to this one. If anyone has any issues with these being applied, please let me know. Responses should be made by Thu, 23 Jan 2025 17:45:02 +0000. Anything received after that time might be too late. The whole patch series can be found in one patch at: https://www.kernel.org/pub/linux/kernel/v6.x/stable-review/patch-6.12.11-rc… or in the git tree and branch at: git://git.kernel.org/pub/scm/linux/kernel/git/stable/linux-stable-rc.git linux-6.12.y and the diffstat can be found below. thanks, greg k-h ------------- Pseudo-Shortlog of commits: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> Linux 6.12.11-rc1 Ryan Lee <ryan.lee(a)canonical.com> apparmor: allocate xmatch for nullpdb inside aa_alloc_null Wayne Lin <Wayne.Lin(a)amd.com> drm/amd/display: Validate mdoe under MST LCT=1 case as well Nicholas Susanto <Nicholas.Susanto(a)amd.com> Revert "drm/amd/display: Enable urgent latency adjustments for DCN35" Leo Li <sunpeng.li(a)amd.com> drm/amd/display: Do not wait for PSR disable on vbl enable Tom Chung <chiahsuan.chung(a)amd.com> drm/amd/display: Disable replay and psr while VRR is enabled Tom Chung <chiahsuan.chung(a)amd.com> drm/amd/display: Fix PSR-SU not support but still call the amdgpu_dm_psr_enable Christian König <christian.koenig(a)amd.com> drm/amdgpu: always sync the GFX pipe on ctx switch Kenneth Feng <kenneth.feng(a)amd.com> drm/amdgpu: disable gfxoff with the compute workload on gfx12 Gui Chengming <Jack.Gui(a)amd.com> drm/amdgpu: fix fw attestation for MP0_14_0_{2/3} Alex Deucher <alexander.deucher(a)amd.com> drm/amdgpu/smu13: update powersave optimizations Ashutosh Dixit <ashutosh.dixit(a)intel.com> drm/xe/oa: Add missing VISACTL mux registers Matthew Brost <matthew.brost(a)intel.com> drm/xe: Mark ComputeCS read mode as UC on iGPU Ville Syrjälä <ville.syrjala(a)linux.intel.com> drm/i915/fb: Relax clear color alignment to 64 bytes Xin Li (Intel) <xin(a)zytor.com> x86/fred: Fix the FRED RSP0 MSR out of sync with its per-CPU cache Frederic Weisbecker <frederic(a)kernel.org> timers/migration: Enforce group initialization visibility to tree walkers Frederic Weisbecker <frederic(a)kernel.org> timers/migration: Fix another race between hotplug and idle entry/exit Koichiro Den <koichiro.den(a)canonical.com> hrtimers: Handle CPU state correctly on hotplug Tomas Krcka <krckatom(a)amazon.de> irqchip/gic-v3-its: Don't enable interrupts in its_irq_set_vcpu_affinity() Yogesh Lal <quic_ylal(a)quicinc.com> irqchip/gic-v3: Handle CPU_PM_ENTER_FAILED correctly Joe Hattori <joe(a)pf.is.s.u-tokyo.ac.jp> irqchip: Plug a OF node reference leak in platform_irqchip_probe() Steven Rostedt <rostedt(a)goodmis.org> tracing: gfp: Fix the GFP enum values shown for user space tracing tools Donet Tom <donettom(a)linux.ibm.com> mm: vmscan : pgdemote vmstat is not getting updated when MGLRU is enabled. Ryan Roberts <ryan.roberts(a)arm.com> mm: clear uffd-wp PTE/PMD state on mremap() Leo Li <sunpeng.li(a)amd.com> drm/amd/display: Do not elevate mem_type change to full update Ryan Roberts <ryan.roberts(a)arm.com> selftests/mm: set allocated memory to non-zero content in cow test Guo Weikang <guoweikang.kernel(a)gmail.com> mm/kmemleak: fix percpu memory leak detection failure Xiaolei Wang <xiaolei.wang(a)windriver.com> pmdomain: imx8mp-blk-ctrl: add missing loop break condition Suren Baghdasaryan <surenb(a)google.com> tools: fix atomic_set() definition to set the value correctly Sean Anderson <sean.anderson(a)linux.dev> gpio: xilinx: Convert gpio_lock to raw spinlock Rik van Riel <riel(a)surriel.com> fs/proc: fix softlockup in __read_vmcore (part 2) Marco Nelissen <marco.nelissen(a)gmail.com> filemap: avoid truncating 64-bit offset to 32 bits Paul Fertser <fercerpav(a)gmail.com> net/ncsi: fix locking in Get MAC Address handling Takashi Iwai <tiwai(a)suse.de> drm/nouveau/disp: Fix missing backlight control on Macbook 5,1 Dave Airlie <airlied(a)redhat.com> nouveau/fence: handle cross device fences properly Stefano Garzarella <sgarzare(a)redhat.com> vsock: prevent null-ptr-deref in vsock_*[has_data|has_space] Stefano Garzarella <sgarzare(a)redhat.com> vsock: reset socket state when de-assigning the transport Stefano Garzarella <sgarzare(a)redhat.com> vsock/virtio: cancel close work in the destructor Stefano Garzarella <sgarzare(a)redhat.com> vsock/virtio: discard packets if the transport changes Stefano Garzarella <sgarzare(a)redhat.com> vsock/bpf: return early if transport is not assigned Heiner Kallweit <hkallweit1(a)gmail.com> net: ethernet: xgbe: re-add aneg to supported features in PHY quirks Paolo Abeni <pabeni(a)redhat.com> selftests: mptcp: avoid spurious errors on disconnect Paolo Abeni <pabeni(a)redhat.com> mptcp: fix spurious wake-up on under memory pressure Paolo Abeni <pabeni(a)redhat.com> mptcp: be sure to send ack when mptcp-level window re-opens Tomi Valkeinen <tomi.valkeinen+renesas(a)ideasonboard.com> i2c: atr: Fix client detach Kairui Song <kasong(a)tencent.com> zram: fix potential UAF of zram table Luke D. Jones <luke(a)ljones.dev> ALSA: hda/realtek: fixup ASUS H7606W Luke D. Jones <luke(a)ljones.dev> ALSA: hda/realtek: fixup ASUS GA605W Stefan Binding <sbinding(a)opensource.cirrus.com> ALSA: hda/realtek: Add support for Ayaneo System using CS35L41 HDA Juergen Gross <jgross(a)suse.com> x86/asm: Make serialize() always_inline Peter Zijlstra <peterz(a)infradead.org> sched/fair: Fix update_cfs_group() vs DELAY_DEQUEUE Peter Zijlstra <peterz(a)infradead.org> sched/fair: Fix EEVDF entity placement bug causing scheduling lag Luis Chamberlain <mcgrof(a)kernel.org> nvmet: propagate npwg topology Tejun Heo <tj(a)kernel.org> sched_ext: Fix dsq_local_on selftest Hongguang Gao <hongguang.gao(a)broadcom.com> RDMA/bnxt_re: Fix to export port num to ib_query_qp David Vernet <void(a)manifault.com> scx: Fix maximal BPF selftest prog Ihor Solodrai <ihor.solodrai(a)pm.me> selftests/sched_ext: fix build after renames in sched_ext API Oleg Nesterov <oleg(a)redhat.com> poll_wait: add mb() to fix theoretical race between waitqueue_active() and .poll() Lizhi Xu <lizhi.xu(a)windriver.com> afs: Fix merge preference rule failure condition Marco Nelissen <marco.nelissen(a)gmail.com> iomap: avoid avoid truncating 64-bit offset to 32 bits Henry Huang <henry.hj(a)antgroup.com> sched_ext: keep running prev when prev->scx.slice != 0 Hans de Goede <hdegoede(a)redhat.com> ACPI: resource: acpi_dev_irq_override(): Check DMI match last Srinivas Pandruvada <srinivas.pandruvada(a)linux.intel.com> platform/x86: ISST: Add Clearwater Forest to support list Srinivas Pandruvada <srinivas.pandruvada(a)linux.intel.com> platform/x86/intel: power-domains: Add Clearwater Forest support Jakub Kicinski <kuba(a)kernel.org> selftests: tc-testing: reduce rshift value Koichiro Den <koichiro.den(a)canonical.com> gpio: sim: lock up configfs that an instantiated device depends on Koichiro Den <koichiro.den(a)canonical.com> gpio: virtuser: lock up configfs that an instantiated device depends on Manivannan Sadhasivam <manivannan.sadhasivam(a)linaro.org> scsi: ufs: core: Honor runtime/system PM levels if set by host controller drivers Max Kellermann <max.kellermann(a)ionos.com> cachefiles: Parse the "secctx" immediately David Howells <dhowells(a)redhat.com> netfs: Fix non-contiguous donation between completed reads David Howells <dhowells(a)redhat.com> kheaders: Ignore silly-rename files Zhang Kunbo <zhangkunbo(a)huawei.com> fs: fix missing declaration of init_files Brahmajit Das <brahmajit.xyz(a)gmail.com> fs/qnx6: Fix building with GCC 15 Leo Stone <leocstone(a)gmail.com> hfs: Sanity check the root record Lizhi Xu <lizhi.xu(a)windriver.com> mac802154: check local interfaces before deleting sdata list Paulo Alcantara <pc(a)manguebit.com> smb: client: fix double free of TCP_Server_Info::hostname David Lechner <dlechner(a)baylibre.com> hwmon: (ltc2991) Fix mixed signed/unsigned in DIV_ROUND_CLOSEST Wolfram Sang <wsa+renesas(a)sang-engineering.com> i2c: testunit: on errors, repeat NACK until STOP Wolfram Sang <wsa+renesas(a)sang-engineering.com> i2c: rcar: fix NACK handling when being a target Wolfram Sang <wsa+renesas(a)sang-engineering.com> i2c: mux: demux-pinctrl: check initial mux selection, too Pratyush Yadav <pratyush(a)kernel.org> Revert "mtd: spi-nor: core: replace dummy buswidth from addr to data" David Lechner <dlechner(a)baylibre.com> hwmon: (tmp513) Fix division of negative numbers Chenyuan Yang <chenyuan0y(a)gmail.com> platform/x86: lenovo-yoga-tab2-pro-1380-fastcharger: fix serdev race Chenyuan Yang <chenyuan0y(a)gmail.com> platform/x86: dell-uart-backlight: fix serdev race Joe Hattori <joe(a)pf.is.s.u-tokyo.ac.jp> i2c: core: fix reference leak in i2c_register_adapter() MD Danish Anwar <danishanwar(a)ti.com> soc: ti: pruss: Fix pruss APIs Claudiu Beznea <claudiu.beznea.uj(a)bp.renesas.com> reset: rzg2l-usbphy-ctrl: Assign proper of node to the allocated device Maíra Canal <mcanal(a)igalia.com> drm/v3d: Ensure job pointer is set to NULL after job completion Ian Forbes <ian.forbes(a)broadcom.com> drm/vmwgfx: Add new keep_resv BO param Ian Forbes <ian.forbes(a)broadcom.com> drm/vmwgfx: Unreserve BO on error Yu-Chun Lin <eleanor15x(a)gmail.com> drm/tests: helpers: Fix compiler warning Jakub Kicinski <kuba(a)kernel.org> netdev: avoid CFI problems with sock priv helpers Leon Romanovsky <leon(a)kernel.org> net/mlx5e: Always start IPsec sequence number from 1 Leon Romanovsky <leon(a)kernel.org> net/mlx5e: Rely on reqid in IPsec tunnel mode Leon Romanovsky <leon(a)kernel.org> net/mlx5e: Fix inversion dependency warning while enabling IPsec tunnel Mark Zhang <markzhang(a)nvidia.com> net/mlx5: Clear port select structure when fail to create Chris Mi <cmi(a)nvidia.com> net/mlx5: SF, Fix add port error handling Yishai Hadas <yishaih(a)nvidia.com> net/mlx5: Fix a lockdep warning as part of the write combining test Patrisious Haddad <phaddad(a)nvidia.com> net/mlx5: Fix RDMA TX steering prio Pavel Begunkov <asml.silence(a)gmail.com> net: make page_pool_ref_netmem work with net iovs Kevin Groeneveld <kgroeneveld(a)lenbrook.com> net: fec: handle page_pool_dev_alloc_pages error Sean Anderson <sean.anderson(a)linux.dev> net: xilinx: axienet: Fix IRQ coalescing packet count overflow Dan Carpenter <dan.carpenter(a)linaro.org> nfp: bpf: prevent integer overflow in nfp_bpf_event_output() Viresh Kumar <viresh.kumar(a)linaro.org> cpufreq: Move endif to the end of Kconfig file Kuniyuki Iwashima <kuniyu(a)amazon.com> pfcp: Destroy device along with udp socket's netns dismantle. Kuniyuki Iwashima <kuniyu(a)amazon.com> gtp: Destroy device along with udp socket's netns dismantle. Kuniyuki Iwashima <kuniyu(a)amazon.com> gtp: Use for_each_netdev_rcu() in gtp_genl_dump_pdp(). Qu Wenruo <wqu(a)suse.com> btrfs: add the missing error handling inside get_canonical_dev_path Rafael J. Wysocki <rafael.j.wysocki(a)intel.com> cpuidle: teo: Update documentation after previous changes Karol Kolacinski <karol.kolacinski(a)intel.com> ice: Add correct PHY lane assignment Sergey Temerkhanov <sergey.temerkhanov(a)intel.com> ice: Use ice_adapter for PTP shared data instead of auxdev Sergey Temerkhanov <sergey.temerkhanov(a)intel.com> ice: Add ice_get_ctrl_ptp() wrapper to simplify the code Sergey Temerkhanov <sergey.temerkhanov(a)intel.com> ice: Introduce ice_get_phy_model() wrapper Karol Kolacinski <karol.kolacinski(a)intel.com> ice: Fix ETH56G FC-FEC Rx offset value Karol Kolacinski <karol.kolacinski(a)intel.com> ice: Fix quad registers read on E825 Karol Kolacinski <karol.kolacinski(a)intel.com> ice: Fix E825 initialization Artem Chernyshev <artem.chernyshev(a)red-soft.ru> pktgen: Avoid out-of-bounds access in get_imix_entries Ilya Maximets <i.maximets(a)ovn.org> openvswitch: fix lockup on tx to unregistering netdev with carrier Paul Barker <paul.barker.ct(a)bp.renesas.com> net: ravb: Fix max TX frame size for RZ/V2M Jakub Kicinski <kuba(a)kernel.org> eth: bnxt: always recalculate features after XDP clearing, fix null-deref Michal Luczaj <mhal(a)rbox.co> bpf: Fix bpf_sk_select_reuseport() memory leak Sudheer Kumar Doredla <s-doredla(a)ti.com> net: ethernet: ti: cpsw_ale: Fix cpsw_ale_get_field() Ard Biesheuvel <ardb(a)kernel.org> efi/zboot: Limit compression options to GZIP and ZSTD ------------- Diffstat: Makefile | 4 +- arch/x86/include/asm/special_insns.h | 2 +- arch/x86/kernel/fred.c | 8 +- drivers/acpi/resource.c | 6 +- drivers/block/zram/zram_drv.c | 1 + drivers/cpufreq/Kconfig | 4 +- drivers/cpuidle/governors/teo.c | 91 +++---- drivers/firmware/efi/Kconfig | 4 - drivers/firmware/efi/libstub/Makefile.zboot | 18 +- drivers/gpio/gpio-sim.c | 48 +++- drivers/gpio/gpio-virtuser.c | 49 +++- drivers/gpio/gpio-xilinx.c | 32 +-- drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c | 5 +- drivers/gpu/drm/amd/amdgpu/amdgpu_fw_attestation.c | 4 + drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c | 4 +- drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c | 41 ++- .../gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crc.c | 25 +- .../gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crtc.c | 4 +- .../gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crtc.h | 2 +- .../drm/amd/display/amdgpu_dm/amdgpu_dm_debugfs.c | 2 +- .../amd/display/amdgpu_dm/amdgpu_dm_mst_types.c | 14 +- .../gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_psr.c | 35 ++- .../gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_psr.h | 3 +- .../gpu/drm/amd/display/dc/dml/dcn35/dcn35_fpu.c | 4 +- .../gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_0_ppt.c | 11 +- drivers/gpu/drm/i915/display/intel_fb.c | 2 +- drivers/gpu/drm/nouveau/nouveau_fence.c | 6 +- drivers/gpu/drm/nouveau/nvkm/engine/disp/mcp77.c | 1 + drivers/gpu/drm/tests/drm_kunit_helpers.c | 3 +- drivers/gpu/drm/v3d/v3d_irq.c | 4 + drivers/gpu/drm/vmwgfx/vmwgfx_bo.c | 3 +- drivers/gpu/drm/vmwgfx/vmwgfx_bo.h | 3 +- drivers/gpu/drm/vmwgfx/vmwgfx_drv.c | 7 +- drivers/gpu/drm/vmwgfx/vmwgfx_gem.c | 1 + drivers/gpu/drm/vmwgfx/vmwgfx_kms.c | 20 +- drivers/gpu/drm/vmwgfx/vmwgfx_shader.c | 7 +- drivers/gpu/drm/vmwgfx/vmwgfx_ttm_buffer.c | 5 +- drivers/gpu/drm/xe/xe_hw_engine.c | 2 +- drivers/gpu/drm/xe/xe_oa.c | 1 + drivers/hwmon/ltc2991.c | 2 +- drivers/hwmon/tmp513.c | 7 +- drivers/i2c/busses/i2c-rcar.c | 20 +- drivers/i2c/i2c-atr.c | 2 +- drivers/i2c/i2c-core-base.c | 1 + drivers/i2c/i2c-slave-testunit.c | 19 +- drivers/i2c/muxes/i2c-demux-pinctrl.c | 4 +- drivers/infiniband/hw/bnxt_re/ib_verbs.c | 1 + drivers/infiniband/hw/bnxt_re/ib_verbs.h | 4 + drivers/infiniband/hw/bnxt_re/qplib_fp.c | 1 + drivers/infiniband/hw/bnxt_re/qplib_fp.h | 1 + drivers/irqchip/irq-gic-v3-its.c | 2 +- drivers/irqchip/irq-gic-v3.c | 2 +- drivers/irqchip/irqchip.c | 4 +- drivers/mtd/spi-nor/core.c | 2 +- drivers/net/ethernet/amd/xgbe/xgbe-phy-v2.c | 19 +- drivers/net/ethernet/broadcom/bnxt/bnxt.c | 25 +- drivers/net/ethernet/broadcom/bnxt/bnxt.h | 2 +- drivers/net/ethernet/broadcom/bnxt/bnxt_xdp.c | 7 - drivers/net/ethernet/freescale/fec_main.c | 19 +- drivers/net/ethernet/intel/ice/ice.h | 5 + drivers/net/ethernet/intel/ice/ice_adapter.c | 6 + drivers/net/ethernet/intel/ice/ice_adapter.h | 22 +- drivers/net/ethernet/intel/ice/ice_adminq_cmd.h | 1 + drivers/net/ethernet/intel/ice/ice_common.c | 51 ++++ drivers/net/ethernet/intel/ice/ice_common.h | 1 + drivers/net/ethernet/intel/ice/ice_main.c | 6 +- drivers/net/ethernet/intel/ice/ice_ptp.c | 165 +++++++----- drivers/net/ethernet/intel/ice/ice_ptp.h | 9 +- drivers/net/ethernet/intel/ice/ice_ptp_consts.h | 2 +- drivers/net/ethernet/intel/ice/ice_ptp_hw.c | 285 +++++++++++---------- drivers/net/ethernet/intel/ice/ice_ptp_hw.h | 5 + drivers/net/ethernet/intel/ice/ice_type.h | 2 - .../ethernet/mellanox/mlx5/core/en_accel/ipsec.c | 22 +- .../mellanox/mlx5/core/en_accel/ipsec_fs.c | 12 +- .../mellanox/mlx5/core/en_accel/ipsec_offload.c | 11 +- drivers/net/ethernet/mellanox/mlx5/core/fs_core.c | 1 + .../net/ethernet/mellanox/mlx5/core/lag/port_sel.c | 4 +- .../net/ethernet/mellanox/mlx5/core/sf/devlink.c | 1 + drivers/net/ethernet/mellanox/mlx5/core/wc.c | 24 +- drivers/net/ethernet/netronome/nfp/bpf/offload.c | 3 +- drivers/net/ethernet/renesas/ravb_main.c | 1 + drivers/net/ethernet/ti/cpsw_ale.c | 14 +- drivers/net/ethernet/xilinx/xilinx_axienet_main.c | 6 + drivers/net/gtp.c | 26 +- drivers/net/pfcp.c | 15 +- drivers/nvme/target/io-cmd-bdev.c | 2 +- drivers/platform/x86/dell/dell-uart-backlight.c | 5 +- .../x86/intel/speed_select_if/isst_if_common.c | 1 + drivers/platform/x86/intel/tpmi_power_domains.c | 1 + .../x86/lenovo-yoga-tab2-pro-1380-fastcharger.c | 5 +- drivers/pmdomain/imx/imx8mp-blk-ctrl.c | 2 +- drivers/reset/reset-rzg2l-usbphy-ctrl.c | 1 + drivers/ufs/core/ufshcd.c | 9 +- fs/afs/addr_prefs.c | 6 +- fs/btrfs/volumes.c | 4 + fs/cachefiles/daemon.c | 14 +- fs/cachefiles/internal.h | 3 +- fs/cachefiles/security.c | 6 +- fs/file.c | 1 + fs/hfs/super.c | 4 +- fs/iomap/buffered-io.c | 2 +- fs/netfs/read_collect.c | 9 +- fs/proc/vmcore.c | 2 + fs/qnx6/inode.c | 11 +- fs/smb/client/connect.c | 3 +- include/linux/hrtimer.h | 1 + include/linux/poll.h | 10 +- include/linux/pruss_driver.h | 12 +- include/linux/userfaultfd_k.h | 12 + include/net/page_pool/helpers.h | 2 +- include/trace/events/mmflags.h | 63 +++++ kernel/cpu.c | 2 +- kernel/gen_kheaders.sh | 1 + kernel/sched/ext.c | 11 +- kernel/sched/fair.c | 151 ++--------- kernel/time/hrtimer.c | 11 +- kernel/time/timer_migration.c | 43 +++- mm/filemap.c | 2 +- mm/huge_memory.c | 12 + mm/hugetlb.c | 14 +- mm/kmemleak.c | 2 +- mm/mremap.c | 32 ++- mm/vmscan.c | 3 + net/core/filter.c | 30 ++- net/core/netdev-genl-gen.c | 14 +- net/core/pktgen.c | 6 +- net/mac802154/iface.c | 4 + net/mptcp/options.c | 6 +- net/mptcp/protocol.h | 9 +- net/ncsi/internal.h | 2 + net/ncsi/ncsi-manage.c | 16 +- net/ncsi/ncsi-rsp.c | 19 +- net/openvswitch/actions.c | 4 +- net/vmw_vsock/af_vsock.c | 18 ++ net/vmw_vsock/virtio_transport_common.c | 38 ++- net/vmw_vsock/vsock_bpf.c | 9 + security/apparmor/policy.c | 1 + sound/pci/hda/patch_realtek.c | 3 + tools/net/ynl/ynl-gen-c.py | 16 +- tools/testing/selftests/mm/cow.c | 8 +- tools/testing/selftests/net/mptcp/mptcp_connect.c | 43 +++- .../selftests/sched_ext/ddsp_bogus_dsq_fail.bpf.c | 2 +- .../selftests/sched_ext/ddsp_vtimelocal_fail.bpf.c | 4 +- .../testing/selftests/sched_ext/dsp_local_on.bpf.c | 7 +- tools/testing/selftests/sched_ext/dsp_local_on.c | 5 +- .../selftests/sched_ext/enq_select_cpu_fails.bpf.c | 2 +- tools/testing/selftests/sched_ext/exit.bpf.c | 4 +- tools/testing/selftests/sched_ext/maximal.bpf.c | 8 +- .../selftests/sched_ext/select_cpu_dfl.bpf.c | 2 +- .../sched_ext/select_cpu_dfl_nodispatch.bpf.c | 2 +- .../selftests/sched_ext/select_cpu_dispatch.bpf.c | 2 +- .../sched_ext/select_cpu_dispatch_bad_dsq.bpf.c | 2 +- .../sched_ext/select_cpu_dispatch_dbl_dsp.bpf.c | 4 +- .../selftests/sched_ext/select_cpu_vtime.bpf.c | 8 +- .../tc-testing/tc-tests/filters/flow.json | 4 +- tools/testing/shared/linux/maple_tree.h | 2 +- tools/testing/vma/linux/atomic.h | 2 +- 157 files changed, 1345 insertions(+), 766 deletions(-)

45 minutes

18
145
0 0

IPC drop down on AMD epyc 7702P

by Jean-Baptiste Roquefere

Hi, We (Ateme, a video encoding company) may have found an unwanted behavior in the scheduler since 5.10 (commit 16b0a7a1a0af), then 5.16 (commit c5b0a7eefc70), then 5.19 (commit not found yet), then maybe some other commits from 5.19 to 6.12, with a consequence of IPC decrease. Problem still appears on lasts 6.12, 6.13 and 6.14 We have reverted both 16b0a7a1a0af and c5b0a7eefc70 commits that reduce our performances (see fair.patch attached, applicable on 6.12.17). Performances increase but still doesnt reach our reference on 5.4.152. Instead of trying to find every single commits from 5.18 to 6.12 that could decrease our performance, I chosed to bench 5.4.152 versus 6.12.17 with and without fair.patch. The problem appeared clear : a lot of CPU migrations go out of CCX, then L3 miss, then IPC decrease. Context of our bench: video decoder which work at a regulated speed, 1 process, 21 main threads, everyone of them creates 10 threads, 8 of them have a fine granularity, meaning they go to sleep quite often, giving the scheduler a lot of opportunities to act). Hardware is an AMD Epyc 7702P, 128 cores, grouped by shared LLC 4 cores +4 hyperthreaded cores. NUMA topology is set by the BIOS to 1 node per socket. Every pthread are created with default attributes. I use AMDuProf (-C -A system -a -m ipc,l1,l2,l3,memory) for CPU Utilization (%), CPU effective freq, IPC, L2 access (pti), L2 miss (pti), L3 miss (absolute) and Mem (GB/s, and perf (stat -d -d -d -a) for Context switches, CPU migrations and Real time (s). We noted that upgrade 5.4.152 to 6.12.17 without any special preempt configuration : Two fold increase in CPU migration 30% memory bandwidth increase 20% L3 cache misses increase 10% IPC decrease With the attached fair.patch applied to 6.12.17 (reminder : this patch reverts one commit appeared in 5.10 and another in 5.16) we managed to reduce CPU migrations and increase IPC but not as much as we had on 5.4.152. Our goal is to keep kernel "clean" without any patch (we don't want to apply and maintain fair.patch) then for the rest of my email we will consider stock kernel 6.12.17. I've reduced the "sub threads count" to stays below 128 threads. Then still 21 main threads and instead of 10 worker per main thread I set 5 workers (4 of them with fine granularity) giving 105 pthreads -> everything goes fine in 6.12.17, no extra CPU migration, no extra memory bandwidth... But as soon as we increase worker threads count (10 instead of 5) the problem appears. We know our decoder may have too many threads but that's out of our scope, it has been designed like that some years ago and moving from "lot of small threads to few of big thread" is for now not possible. We have a work around : we group threads using pthread affinities. Every main thread (and by inheritance of affinities every worker threads) on a single CCX so we reduce the L3 miss for them, then decrease memory bandwidth, then finally increasing IPC. With that solution, we go above our original performances, for both kernels, and they perform at the same level. However, it is impractical to productize as such. I've tried many kernel build configurations (CONFIG_PREMPT_*, CONFIG_SCHEDULER_*, tuning of fair.c:sysctl_sched_migration_cost) on 6.12.17, 6.12.21 (longterm), 6.13.9 (mainline), and 6.14.0 Nothing changes. Q: Is there anyway to tune the kernel so we can get our performance back without using the pthread affinities work around ? Feel free to ask an archive containing binaries and payload. I first posted on https://bugzilla.kernel.org/show_bug.cgi?id=220000 but one told me the best way to get answers where these mailing lists Regards, Jean-Baptiste Roquefere, Ateme Attached bench.tar.gz : * bench/fair.patch * bench/bench.ods with 2 sheets : - regulated : decoder speed is regulated to keep real time constant - no regul : decoder speed is not regulated and uses from 1 to 76 main threads with 10 worker per main thread * bench/regulated.csv : bench.ods:regulated exported in csv format * bench/not-regulated : bench.ods:no regul exported in csv format

5 hours, 7 minutes

5
14
0 0

[PATCH] wifi: ath12k: fix ring-buffer corruption

by Johan Hovold

Users of the Lenovo ThinkPad X13s have reported that Wi-Fi sometimes breaks and the log fills up with errors like: ath11k_pci 0006:01:00.0: HTC Rx: insufficient length, got 1484, expected 1492 ath11k_pci 0006:01:00.0: HTC Rx: insufficient length, got 1460, expected 1484 which based on a quick look at the ath11k driver seemed to indicate some kind of ring-buffer corruption. Miaoqing Pan tracked it down to the host seeing the updated destination ring head pointer before the updated descriptor, and the error handling for that in turn leaves the ring buffer in an inconsistent state. While this has not yet been observed with ath12k, the ring-buffer implementation is very similar to the ath11k one and it suffers from the same bugs. Add the missing memory barrier to make sure that the descriptor is read after the head pointer to address the root cause of the corruption while fixing up the error handling in case there are ever any (ordering) bugs on the device side. Note that the READ_ONCE() are only needed to avoid compiler mischief in case the ring-buffer helpers are ever inlined. Tested-on: WCN7850 hw2.0 WLAN.HMT.1.0.c5-00481-QCAHMTSWPL_V1.0_V2.0_SILICONZ-3 Fixes: d889913205cf ("wifi: ath12k: driver for Qualcomm Wi-Fi 7 devices") Cc: stable(a)vger.kernel.org # 6.3 Link: https://bugzilla.kernel.org/show_bug.cgi?id=218623 Link: https://lore.kernel.org/20250310010217.3845141-3-quic_miaoqing@quicinc.com Cc: Miaoqing Pan <quic_miaoqing(a)quicinc.com> Signed-off-by: Johan Hovold <johan+linaro(a)kernel.org> --- drivers/net/wireless/ath/ath12k/ce.c | 11 +++++------ drivers/net/wireless/ath/ath12k/hal.c | 4 ++-- 2 files changed, 7 insertions(+), 8 deletions(-) diff --git a/drivers/net/wireless/ath/ath12k/ce.c b/drivers/net/wireless/ath/ath12k/ce.c index be0d669d31fc..740586fe49d1 100644 --- a/drivers/net/wireless/ath/ath12k/ce.c +++ b/drivers/net/wireless/ath/ath12k/ce.c @@ -343,11 +343,10 @@ static int ath12k_ce_completed_recv_next(struct ath12k_ce_pipe *pipe, goto err; } + /* Make sure descriptor is read after the head pointer. */ + dma_rmb(); + *nbytes = ath12k_hal_ce_dst_status_get_length(desc); - if (*nbytes == 0) { - ret = -EIO; - goto err; - } *skb = pipe->dest_ring->skb[sw_index]; pipe->dest_ring->skb[sw_index] = NULL; @@ -380,8 +379,8 @@ static void ath12k_ce_recv_process_cb(struct ath12k_ce_pipe *pipe) dma_unmap_single(ab->dev, ATH12K_SKB_RXCB(skb)->paddr, max_nbytes, DMA_FROM_DEVICE); - if (unlikely(max_nbytes < nbytes)) { - ath12k_warn(ab, "rxed more than expected (nbytes %d, max %d)", + if (unlikely(max_nbytes < nbytes || nbytes == 0)) { + ath12k_warn(ab, "unexpected rx length (nbytes %d, max %d)", nbytes, max_nbytes); dev_kfree_skb_any(skb); continue; diff --git a/drivers/net/wireless/ath/ath12k/hal.c b/drivers/net/wireless/ath/ath12k/hal.c index cd59ff8e6c7b..91d5126ca149 100644 --- a/drivers/net/wireless/ath/ath12k/hal.c +++ b/drivers/net/wireless/ath/ath12k/hal.c @@ -1962,7 +1962,7 @@ u32 ath12k_hal_ce_dst_status_get_length(struct hal_ce_srng_dst_status_desc *desc { u32 len; - len = le32_get_bits(desc->flags, HAL_CE_DST_STATUS_DESC_FLAGS_LEN); + len = le32_get_bits(READ_ONCE(desc->flags), HAL_CE_DST_STATUS_DESC_FLAGS_LEN); desc->flags &= ~cpu_to_le32(HAL_CE_DST_STATUS_DESC_FLAGS_LEN); return len; @@ -2132,7 +2132,7 @@ void ath12k_hal_srng_access_begin(struct ath12k_base *ab, struct hal_srng *srng) srng->u.src_ring.cached_tp = *(volatile u32 *)srng->u.src_ring.tp_addr; else - srng->u.dst_ring.cached_hp = *srng->u.dst_ring.hp_addr; + srng->u.dst_ring.cached_hp = READ_ONCE(*srng->u.dst_ring.hp_addr); } /* Update cached ring head/tail pointers to HW. ath12k_hal_srng_access_begin() -- 2.48.1

1 day, 2 hours

4
3
0 0

[PATCH v2 1/2] drm/i915/gem: Allow EXEC_CAPTURE on recoverable contexts on DG1

by Ville Syrjala

From: Ville Syrjälä <ville.syrjala(a)linux.intel.com> The intel-media-driver is currently broken on DG1 because it uses EXEC_CAPTURE with recovarable contexts. Relax the check to allow that. I've also submitted a fix for the intel-media-driver: https://github.com/intel/media-driver/pull/1920 Cc: stable(a)vger.kernel.org Cc: Matthew Auld <matthew.auld(a)intel.com> Cc: Thomas Hellström <thomas.hellstrom(a)linux.intel.com> Testcase: igt/gem_exec_capture/capture-invisible Fixes: 71b1669ea9bd ("drm/i915/uapi: tweak error capture on recoverable contexts") Reviewed-by: Andi Shyti <andi.shyti(a)linux.intel.com> Signed-off-by: Ville Syrjälä <ville.syrjala(a)linux.intel.com> --- drivers/gpu/drm/i915/gem/i915_gem_execbuffer.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/gpu/drm/i915/gem/i915_gem_execbuffer.c b/drivers/gpu/drm/i915/gem/i915_gem_execbuffer.c index ca7e9216934a..ea9d5063ce78 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_execbuffer.c +++ b/drivers/gpu/drm/i915/gem/i915_gem_execbuffer.c @@ -2013,7 +2013,7 @@ static int eb_capture_stage(struct i915_execbuffer *eb) continue; if (i915_gem_context_is_recoverable(eb->gem_context) && - (IS_DGFX(eb->i915) || GRAPHICS_VER_FULL(eb->i915) > IP_VER(12, 0))) + GRAPHICS_VER_FULL(eb->i915) > IP_VER(12, 10)) return -EINVAL; for_each_batch_create_order(eb, j) { -- 2.49.0

1 day, 6 hours

4
4
0 0

[PATCH v3] usb: hub: lack of clearing xHC resources

by Pawel Laszczak

The xHC resources allocated for USB devices are not released in correct order after resuming in case when while suspend device was reconnected. This issue has been detected during the fallowing scenario: - connect hub HS to root port - connect LS/FS device to hub port - wait for enumeration to finish - force host to suspend - reconnect hub attached to root port - wake host For this scenario during enumeration of USB LS/FS device the Cadence xHC reports completion error code for xHC commands because the xHC resources used for devices has not been properly released. XHCI specification doesn't mention that device can be reset in any order so, we should not treat this issue as Cadence xHC controller bug. Similar as during disconnecting in this case the device resources should be cleared starting form the last usb device in tree toward the root hub. To fix this issue usbcore driver should call hcd->driver->reset_device for all USB devices connected to hub which was reconnected while suspending. Fixes: 3d82904559f4 ("usb: cdnsp: cdns3 Add main part of Cadence USBSSP DRD Driver") cc: <stable(a)vger.kernel.org> Signed-off-by: Pawel Laszczak <pawell(a)cadence.com> --- Changelog: v3: - Changed patch title - Corrected typo - Moved hub_hc_release_resources above mutex_lock(hcd->address0_mutex) v2: - Replaced disconnection procedure with releasing only the xHC resources drivers/usb/core/hub.c | 33 +++++++++++++++++++++++++++++++++ 1 file changed, 33 insertions(+) diff --git a/drivers/usb/core/hub.c b/drivers/usb/core/hub.c index a76bb50b6202..dcba4281ea48 100644 --- a/drivers/usb/core/hub.c +++ b/drivers/usb/core/hub.c @@ -6065,6 +6065,36 @@ void usb_hub_cleanup(void) usb_deregister(&hub_driver); } /* usb_hub_cleanup() */ +/** + * hub_hc_release_resources - clear resources used by host controller + * @udev: pointer to device being released + * + * Context: task context, might sleep + * + * Function releases the host controller resources in correct order before + * making any operation on resuming usb device. The host controller resources + * allocated for devices in tree should be released starting from the last + * usb device in tree toward the root hub. This function is used only during + * resuming device when usb device require reinitialization – that is, when + * flag udev->reset_resume is set. + * + * This call is synchronous, and may not be used in an interrupt context. + */ +static void hub_hc_release_resources(struct usb_device *udev) +{ + struct usb_hub *hub = usb_hub_to_struct_hub(udev); + struct usb_hcd *hcd = bus_to_hcd(udev->bus); + int i; + + /* Release up resources for all children before this device */ + for (i = 0; i < udev->maxchild; i++) + if (hub->ports[i]->child) + hub_hc_release_resources(hub->ports[i]->child); + + if (hcd->driver->reset_device) + hcd->driver->reset_device(hcd, udev); +} + /** * usb_reset_and_verify_device - perform a USB port reset to reinitialize a device * @udev: device to reset (not in SUSPENDED or NOTATTACHED state) @@ -6129,6 +6159,9 @@ static int usb_reset_and_verify_device(struct usb_device *udev) bos = udev->bos; udev->bos = NULL; + if (udev->reset_resume) + hub_hc_release_resources(udev); + mutex_lock(hcd->address0_mutex); for (i = 0; i < PORT_INIT_TRIES; ++i) { -- 2.43.0

1 day, 7 hours

4
5
0 0

FAILED: patch "[PATCH] hrtimers: Force migrate away hrtimers queued after" failed to apply to 6.6-stable tree

by gregkh＠linuxfoundation.org

The patch below does not apply to the 6.6-stable tree. If someone wants it applied there, or to any other stable or longterm tree, then please email the backport, including the original git commit id to <stable(a)vger.kernel.org>. To reproduce the conflict and resubmit, you may use the following commands: git fetch https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/ linux-6.6.y git checkout FETCH_HEAD git cherry-pick -x 53dac345395c0d2493cbc2f4c85fe38aef5b63f5 # <resolve conflicts, build, test, etc.> git commit -s git send-email --to '<stable(a)vger.kernel.org>' --in-reply-to '2025021052-avenging-aflutter-192c@gregkh' --subject-prefix 'PATCH 6.6.y' HEAD^.. Possible dependencies: thanks, greg k-h ------------------ original commit in Linus's tree ------------------ From 53dac345395c0d2493cbc2f4c85fe38aef5b63f5 Mon Sep 17 00:00:00 2001 From: Frederic Weisbecker <frederic(a)kernel.org> Date: Sat, 18 Jan 2025 00:24:33 +0100 Subject: [PATCH] hrtimers: Force migrate away hrtimers queued after CPUHP_AP_HRTIMERS_DYING hrtimers are migrated away from the dying CPU to any online target at the CPUHP_AP_HRTIMERS_DYING stage in order not to delay bandwidth timers handling tasks involved in the CPU hotplug forward progress. However wakeups can still be performed by the outgoing CPU after CPUHP_AP_HRTIMERS_DYING. Those can result again in bandwidth timers being armed. Depending on several considerations (crystal ball power management based election, earliest timer already enqueued, timer migration enabled or not), the target may eventually be the current CPU even if offline. If that happens, the timer is eventually ignored. The most notable example is RCU which had to deal with each and every of those wake-ups by deferring them to an online CPU, along with related workarounds: _ e787644caf76 (rcu: Defer RCU kthreads wakeup when CPU is dying) _ 9139f93209d1 (rcu/nocb: Fix RT throttling hrtimer armed from offline CPU) _ f7345ccc62a4 (rcu/nocb: Fix rcuog wake-up from offline softirq) The problem isn't confined to RCU though as the stop machine kthread (which runs CPUHP_AP_HRTIMERS_DYING) reports its completion at the end of its work through cpu_stop_signal_done() and performs a wake up that eventually arms the deadline server timer: WARNING: CPU: 94 PID: 588 at kernel/time/hrtimer.c:1086 hrtimer_start_range_ns+0x289/0x2d0 CPU: 94 UID: 0 PID: 588 Comm: migration/94 Not tainted Stopper: multi_cpu_stop+0x0/0x120 <- stop_machine_cpuslocked+0x66/0xc0 RIP: 0010:hrtimer_start_range_ns+0x289/0x2d0 Call Trace: <TASK> start_dl_timer enqueue_dl_entity dl_server_start enqueue_task_fair enqueue_task ttwu_do_activate try_to_wake_up complete cpu_stopper_thread Instead of providing yet another bandaid to work around the situation, fix it in the hrtimers infrastructure instead: always migrate away a timer to an online target whenever it is enqueued from an offline CPU. This will also allow to revert all the above RCU disgraceful hacks. Fixes: 5c0930ccaad5 ("hrtimers: Push pending hrtimers away from outgoing CPU earlier") Reported-by: Vlad Poenaru <vlad.wing(a)gmail.com> Reported-by: Usama Arif <usamaarif642(a)gmail.com> Signed-off-by: Frederic Weisbecker <frederic(a)kernel.org> Signed-off-by: Paul E. McKenney <paulmck(a)kernel.org> Signed-off-by: Thomas Gleixner <tglx(a)linutronix.de> Cc: stable(a)vger.kernel.org Tested-by: Paul E. McKenney <paulmck(a)kernel.org> Link: https://lore.kernel.org/all/20250117232433.24027-1-frederic@kernel.org Closes: 20241213203739.1519801-1-usamaarif642(a)gmail.com diff --git a/include/linux/hrtimer_defs.h b/include/linux/hrtimer_defs.h index c3b4b7ed7c16..84a5045f80f3 100644 --- a/include/linux/hrtimer_defs.h +++ b/include/linux/hrtimer_defs.h @@ -125,6 +125,7 @@ struct hrtimer_cpu_base { ktime_t softirq_expires_next; struct hrtimer *softirq_next_timer; struct hrtimer_clock_base clock_base[HRTIMER_MAX_CLOCK_BASES]; + call_single_data_t csd; } ____cacheline_aligned; diff --git a/kernel/time/hrtimer.c b/kernel/time/hrtimer.c index 4fb81f8c6f1c..deb1aa32814e 100644 --- a/kernel/time/hrtimer.c +++ b/kernel/time/hrtimer.c @@ -58,6 +58,8 @@ #define HRTIMER_ACTIVE_SOFT (HRTIMER_ACTIVE_HARD << MASK_SHIFT) #define HRTIMER_ACTIVE_ALL (HRTIMER_ACTIVE_SOFT | HRTIMER_ACTIVE_HARD) +static void retrigger_next_event(void *arg); + /* * The timer bases: * @@ -111,7 +113,8 @@ DEFINE_PER_CPU(struct hrtimer_cpu_base, hrtimer_bases) = .clockid = CLOCK_TAI, .get_time = &ktime_get_clocktai, }, - } + }, + .csd = CSD_INIT(retrigger_next_event, NULL) }; static const int hrtimer_clock_to_base_table[MAX_CLOCKS] = { @@ -124,6 +127,14 @@ static const int hrtimer_clock_to_base_table[MAX_CLOCKS] = { [CLOCK_TAI] = HRTIMER_BASE_TAI, }; +static inline bool hrtimer_base_is_online(struct hrtimer_cpu_base *base) +{ + if (!IS_ENABLED(CONFIG_HOTPLUG_CPU)) + return true; + else + return likely(base->online); +} + /* * Functions and macros which are different for UP/SMP systems are kept in a * single place @@ -178,27 +189,54 @@ struct hrtimer_clock_base *lock_hrtimer_base(const struct hrtimer *timer, } /* - * We do not migrate the timer when it is expiring before the next - * event on the target cpu. When high resolution is enabled, we cannot - * reprogram the target cpu hardware and we would cause it to fire - * late. To keep it simple, we handle the high resolution enabled and - * disabled case similar. + * Check if the elected target is suitable considering its next + * event and the hotplug state of the current CPU. + * + * If the elected target is remote and its next event is after the timer + * to queue, then a remote reprogram is necessary. However there is no + * guarantee the IPI handling the operation would arrive in time to meet + * the high resolution deadline. In this case the local CPU becomes a + * preferred target, unless it is offline. + * + * High and low resolution modes are handled the same way for simplicity. * * Called with cpu_base->lock of target cpu held. */ -static int -hrtimer_check_target(struct hrtimer *timer, struct hrtimer_clock_base *new_base) +static bool hrtimer_suitable_target(struct hrtimer *timer, struct hrtimer_clock_base *new_base, + struct hrtimer_cpu_base *new_cpu_base, + struct hrtimer_cpu_base *this_cpu_base) { ktime_t expires; + /* + * The local CPU clockevent can be reprogrammed. Also get_target_base() + * guarantees it is online. + */ + if (new_cpu_base == this_cpu_base) + return true; + + /* + * The offline local CPU can't be the default target if the + * next remote target event is after this timer. Keep the + * elected new base. An IPI will we issued to reprogram + * it as a last resort. + */ + if (!hrtimer_base_is_online(this_cpu_base)) + return true; + expires = ktime_sub(hrtimer_get_expires(timer), new_base->offset); - return expires < new_base->cpu_base->expires_next; + + return expires >= new_base->cpu_base->expires_next; } -static inline -struct hrtimer_cpu_base *get_target_base(struct hrtimer_cpu_base *base, - int pinned) +static inline struct hrtimer_cpu_base *get_target_base(struct hrtimer_cpu_base *base, int pinned) { + if (!hrtimer_base_is_online(base)) { + int cpu = cpumask_any_and(cpu_online_mask, housekeeping_cpumask(HK_TYPE_TIMER)); + + return &per_cpu(hrtimer_bases, cpu); + } + #if defined(CONFIG_SMP) && defined(CONFIG_NO_HZ_COMMON) if (static_branch_likely(&timers_migration_enabled) && !pinned) return &per_cpu(hrtimer_bases, get_nohz_timer_target()); @@ -249,8 +287,8 @@ switch_hrtimer_base(struct hrtimer *timer, struct hrtimer_clock_base *base, raw_spin_unlock(&base->cpu_base->lock); raw_spin_lock(&new_base->cpu_base->lock); - if (new_cpu_base != this_cpu_base && - hrtimer_check_target(timer, new_base)) { + if (!hrtimer_suitable_target(timer, new_base, new_cpu_base, + this_cpu_base)) { raw_spin_unlock(&new_base->cpu_base->lock); raw_spin_lock(&base->cpu_base->lock); new_cpu_base = this_cpu_base; @@ -259,8 +297,7 @@ switch_hrtimer_base(struct hrtimer *timer, struct hrtimer_clock_base *base, } WRITE_ONCE(timer->base, new_base); } else { - if (new_cpu_base != this_cpu_base && - hrtimer_check_target(timer, new_base)) { + if (!hrtimer_suitable_target(timer, new_base, new_cpu_base, this_cpu_base)) { new_cpu_base = this_cpu_base; goto again; } @@ -706,8 +743,6 @@ static inline int hrtimer_is_hres_enabled(void) return hrtimer_hres_enabled; } -static void retrigger_next_event(void *arg); - /* * Switch to high resolution mode */ @@ -1195,6 +1230,7 @@ static int __hrtimer_start_range_ns(struct hrtimer *timer, ktime_t tim, u64 delta_ns, const enum hrtimer_mode mode, struct hrtimer_clock_base *base) { + struct hrtimer_cpu_base *this_cpu_base = this_cpu_ptr(&hrtimer_bases); struct hrtimer_clock_base *new_base; bool force_local, first; @@ -1206,9 +1242,15 @@ static int __hrtimer_start_range_ns(struct hrtimer *timer, ktime_t tim, * and enforce reprogramming after it is queued no matter whether * it is the new first expiring timer again or not. */ - force_local = base->cpu_base == this_cpu_ptr(&hrtimer_bases); + force_local = base->cpu_base == this_cpu_base; force_local &= base->cpu_base->next_timer == timer; + /* + * Don't force local queuing if this enqueue happens on a unplugged + * CPU after hrtimer_cpu_dying() has been invoked. + */ + force_local &= this_cpu_base->online; + /* * Remove an active timer from the queue. In case it is not queued * on the current CPU, make sure that remove_hrtimer() updates the @@ -1238,8 +1280,27 @@ static int __hrtimer_start_range_ns(struct hrtimer *timer, ktime_t tim, } first = enqueue_hrtimer(timer, new_base, mode); - if (!force_local) - return first; + if (!force_local) { + /* + * If the current CPU base is online, then the timer is + * never queued on a remote CPU if it would be the first + * expiring timer there. + */ + if (hrtimer_base_is_online(this_cpu_base)) + return first; + + /* + * Timer was enqueued remote because the current base is + * already offline. If the timer is the first to expire, + * kick the remote CPU to reprogram the clock event. + */ + if (first) { + struct hrtimer_cpu_base *new_cpu_base = new_base->cpu_base; + + smp_call_function_single_async(new_cpu_base->cpu, &new_cpu_base->csd); + } + return 0; + } /* * Timer was forced to stay on the current CPU to avoid

1 day, 15 hours

3
2
0 0

FAILED: patch "[PATCH] hrtimers: Force migrate away hrtimers queued after" failed to apply to 6.1-stable tree

by gregkh＠linuxfoundation.org

The patch below does not apply to the 6.1-stable tree. If someone wants it applied there, or to any other stable or longterm tree, then please email the backport, including the original git commit id to <stable(a)vger.kernel.org>. To reproduce the conflict and resubmit, you may use the following commands: git fetch https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/ linux-6.1.y git checkout FETCH_HEAD git cherry-pick -x 53dac345395c0d2493cbc2f4c85fe38aef5b63f5 # <resolve conflicts, build, test, etc.> git commit -s git send-email --to '<stable(a)vger.kernel.org>' --in-reply-to '2025021053-unranked-silt-0282@gregkh' --subject-prefix 'PATCH 6.1.y' HEAD^.. Possible dependencies: thanks, greg k-h ------------------ original commit in Linus's tree ------------------ From 53dac345395c0d2493cbc2f4c85fe38aef5b63f5 Mon Sep 17 00:00:00 2001 From: Frederic Weisbecker <frederic(a)kernel.org> Date: Sat, 18 Jan 2025 00:24:33 +0100 Subject: [PATCH] hrtimers: Force migrate away hrtimers queued after CPUHP_AP_HRTIMERS_DYING hrtimers are migrated away from the dying CPU to any online target at the CPUHP_AP_HRTIMERS_DYING stage in order not to delay bandwidth timers handling tasks involved in the CPU hotplug forward progress. However wakeups can still be performed by the outgoing CPU after CPUHP_AP_HRTIMERS_DYING. Those can result again in bandwidth timers being armed. Depending on several considerations (crystal ball power management based election, earliest timer already enqueued, timer migration enabled or not), the target may eventually be the current CPU even if offline. If that happens, the timer is eventually ignored. The most notable example is RCU which had to deal with each and every of those wake-ups by deferring them to an online CPU, along with related workarounds: _ e787644caf76 (rcu: Defer RCU kthreads wakeup when CPU is dying) _ 9139f93209d1 (rcu/nocb: Fix RT throttling hrtimer armed from offline CPU) _ f7345ccc62a4 (rcu/nocb: Fix rcuog wake-up from offline softirq) The problem isn't confined to RCU though as the stop machine kthread (which runs CPUHP_AP_HRTIMERS_DYING) reports its completion at the end of its work through cpu_stop_signal_done() and performs a wake up that eventually arms the deadline server timer: WARNING: CPU: 94 PID: 588 at kernel/time/hrtimer.c:1086 hrtimer_start_range_ns+0x289/0x2d0 CPU: 94 UID: 0 PID: 588 Comm: migration/94 Not tainted Stopper: multi_cpu_stop+0x0/0x120 <- stop_machine_cpuslocked+0x66/0xc0 RIP: 0010:hrtimer_start_range_ns+0x289/0x2d0 Call Trace: <TASK> start_dl_timer enqueue_dl_entity dl_server_start enqueue_task_fair enqueue_task ttwu_do_activate try_to_wake_up complete cpu_stopper_thread Instead of providing yet another bandaid to work around the situation, fix it in the hrtimers infrastructure instead: always migrate away a timer to an online target whenever it is enqueued from an offline CPU. This will also allow to revert all the above RCU disgraceful hacks. Fixes: 5c0930ccaad5 ("hrtimers: Push pending hrtimers away from outgoing CPU earlier") Reported-by: Vlad Poenaru <vlad.wing(a)gmail.com> Reported-by: Usama Arif <usamaarif642(a)gmail.com> Signed-off-by: Frederic Weisbecker <frederic(a)kernel.org> Signed-off-by: Paul E. McKenney <paulmck(a)kernel.org> Signed-off-by: Thomas Gleixner <tglx(a)linutronix.de> Cc: stable(a)vger.kernel.org Tested-by: Paul E. McKenney <paulmck(a)kernel.org> Link: https://lore.kernel.org/all/20250117232433.24027-1-frederic@kernel.org Closes: 20241213203739.1519801-1-usamaarif642(a)gmail.com diff --git a/include/linux/hrtimer_defs.h b/include/linux/hrtimer_defs.h index c3b4b7ed7c16..84a5045f80f3 100644 --- a/include/linux/hrtimer_defs.h +++ b/include/linux/hrtimer_defs.h @@ -125,6 +125,7 @@ struct hrtimer_cpu_base { ktime_t softirq_expires_next; struct hrtimer *softirq_next_timer; struct hrtimer_clock_base clock_base[HRTIMER_MAX_CLOCK_BASES]; + call_single_data_t csd; } ____cacheline_aligned; diff --git a/kernel/time/hrtimer.c b/kernel/time/hrtimer.c index 4fb81f8c6f1c..deb1aa32814e 100644 --- a/kernel/time/hrtimer.c +++ b/kernel/time/hrtimer.c @@ -58,6 +58,8 @@ #define HRTIMER_ACTIVE_SOFT (HRTIMER_ACTIVE_HARD << MASK_SHIFT) #define HRTIMER_ACTIVE_ALL (HRTIMER_ACTIVE_SOFT | HRTIMER_ACTIVE_HARD) +static void retrigger_next_event(void *arg); + /* * The timer bases: * @@ -111,7 +113,8 @@ DEFINE_PER_CPU(struct hrtimer_cpu_base, hrtimer_bases) = .clockid = CLOCK_TAI, .get_time = &ktime_get_clocktai, }, - } + }, + .csd = CSD_INIT(retrigger_next_event, NULL) }; static const int hrtimer_clock_to_base_table[MAX_CLOCKS] = { @@ -124,6 +127,14 @@ static const int hrtimer_clock_to_base_table[MAX_CLOCKS] = { [CLOCK_TAI] = HRTIMER_BASE_TAI, }; +static inline bool hrtimer_base_is_online(struct hrtimer_cpu_base *base) +{ + if (!IS_ENABLED(CONFIG_HOTPLUG_CPU)) + return true; + else + return likely(base->online); +} + /* * Functions and macros which are different for UP/SMP systems are kept in a * single place @@ -178,27 +189,54 @@ struct hrtimer_clock_base *lock_hrtimer_base(const struct hrtimer *timer, } /* - * We do not migrate the timer when it is expiring before the next - * event on the target cpu. When high resolution is enabled, we cannot - * reprogram the target cpu hardware and we would cause it to fire - * late. To keep it simple, we handle the high resolution enabled and - * disabled case similar. + * Check if the elected target is suitable considering its next + * event and the hotplug state of the current CPU. + * + * If the elected target is remote and its next event is after the timer + * to queue, then a remote reprogram is necessary. However there is no + * guarantee the IPI handling the operation would arrive in time to meet + * the high resolution deadline. In this case the local CPU becomes a + * preferred target, unless it is offline. + * + * High and low resolution modes are handled the same way for simplicity. * * Called with cpu_base->lock of target cpu held. */ -static int -hrtimer_check_target(struct hrtimer *timer, struct hrtimer_clock_base *new_base) +static bool hrtimer_suitable_target(struct hrtimer *timer, struct hrtimer_clock_base *new_base, + struct hrtimer_cpu_base *new_cpu_base, + struct hrtimer_cpu_base *this_cpu_base) { ktime_t expires; + /* + * The local CPU clockevent can be reprogrammed. Also get_target_base() + * guarantees it is online. + */ + if (new_cpu_base == this_cpu_base) + return true; + + /* + * The offline local CPU can't be the default target if the + * next remote target event is after this timer. Keep the + * elected new base. An IPI will we issued to reprogram + * it as a last resort. + */ + if (!hrtimer_base_is_online(this_cpu_base)) + return true; + expires = ktime_sub(hrtimer_get_expires(timer), new_base->offset); - return expires < new_base->cpu_base->expires_next; + + return expires >= new_base->cpu_base->expires_next; } -static inline -struct hrtimer_cpu_base *get_target_base(struct hrtimer_cpu_base *base, - int pinned) +static inline struct hrtimer_cpu_base *get_target_base(struct hrtimer_cpu_base *base, int pinned) { + if (!hrtimer_base_is_online(base)) { + int cpu = cpumask_any_and(cpu_online_mask, housekeeping_cpumask(HK_TYPE_TIMER)); + + return &per_cpu(hrtimer_bases, cpu); + } + #if defined(CONFIG_SMP) && defined(CONFIG_NO_HZ_COMMON) if (static_branch_likely(&timers_migration_enabled) && !pinned) return &per_cpu(hrtimer_bases, get_nohz_timer_target()); @@ -249,8 +287,8 @@ switch_hrtimer_base(struct hrtimer *timer, struct hrtimer_clock_base *base, raw_spin_unlock(&base->cpu_base->lock); raw_spin_lock(&new_base->cpu_base->lock); - if (new_cpu_base != this_cpu_base && - hrtimer_check_target(timer, new_base)) { + if (!hrtimer_suitable_target(timer, new_base, new_cpu_base, + this_cpu_base)) { raw_spin_unlock(&new_base->cpu_base->lock); raw_spin_lock(&base->cpu_base->lock); new_cpu_base = this_cpu_base; @@ -259,8 +297,7 @@ switch_hrtimer_base(struct hrtimer *timer, struct hrtimer_clock_base *base, } WRITE_ONCE(timer->base, new_base); } else { - if (new_cpu_base != this_cpu_base && - hrtimer_check_target(timer, new_base)) { + if (!hrtimer_suitable_target(timer, new_base, new_cpu_base, this_cpu_base)) { new_cpu_base = this_cpu_base; goto again; } @@ -706,8 +743,6 @@ static inline int hrtimer_is_hres_enabled(void) return hrtimer_hres_enabled; } -static void retrigger_next_event(void *arg); - /* * Switch to high resolution mode */ @@ -1195,6 +1230,7 @@ static int __hrtimer_start_range_ns(struct hrtimer *timer, ktime_t tim, u64 delta_ns, const enum hrtimer_mode mode, struct hrtimer_clock_base *base) { + struct hrtimer_cpu_base *this_cpu_base = this_cpu_ptr(&hrtimer_bases); struct hrtimer_clock_base *new_base; bool force_local, first; @@ -1206,9 +1242,15 @@ static int __hrtimer_start_range_ns(struct hrtimer *timer, ktime_t tim, * and enforce reprogramming after it is queued no matter whether * it is the new first expiring timer again or not. */ - force_local = base->cpu_base == this_cpu_ptr(&hrtimer_bases); + force_local = base->cpu_base == this_cpu_base; force_local &= base->cpu_base->next_timer == timer; + /* + * Don't force local queuing if this enqueue happens on a unplugged + * CPU after hrtimer_cpu_dying() has been invoked. + */ + force_local &= this_cpu_base->online; + /* * Remove an active timer from the queue. In case it is not queued * on the current CPU, make sure that remove_hrtimer() updates the @@ -1238,8 +1280,27 @@ static int __hrtimer_start_range_ns(struct hrtimer *timer, ktime_t tim, } first = enqueue_hrtimer(timer, new_base, mode); - if (!force_local) - return first; + if (!force_local) { + /* + * If the current CPU base is online, then the timer is + * never queued on a remote CPU if it would be the first + * expiring timer there. + */ + if (hrtimer_base_is_online(this_cpu_base)) + return first; + + /* + * Timer was enqueued remote because the current base is + * already offline. If the timer is the first to expire, + * kick the remote CPU to reprogram the clock event. + */ + if (first) { + struct hrtimer_cpu_base *new_cpu_base = new_base->cpu_base; + + smp_call_function_single_async(new_cpu_base->cpu, &new_cpu_base->csd); + } + return 0; + } /* * Timer was forced to stay on the current CPU to avoid

1 day, 15 hours

5
6
0 0

FAILED: patch "[PATCH] btrfs: check folio mapping after unlock in" failed to apply to 6.1-stable tree

by gregkh＠linuxfoundation.org

The patch below does not apply to the 6.1-stable tree. If someone wants it applied there, or to any other stable or longterm tree, then please email the backport, including the original git commit id to <stable(a)vger.kernel.org>. To reproduce the conflict and resubmit, you may use the following commands: git fetch https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/ linux-6.1.y git checkout FETCH_HEAD git cherry-pick -x 3e74859ee35edc33a022c3f3971df066ea0ca6b9 # <resolve conflicts, build, test, etc.> git commit -s git send-email --to '<stable(a)vger.kernel.org>' --in-reply-to '2024123045-parka-sublet-a95d@gregkh' --subject-prefix 'PATCH 6.1.y' HEAD^.. Possible dependencies: thanks, greg k-h ------------------ original commit in Linus's tree ------------------ From 3e74859ee35edc33a022c3f3971df066ea0ca6b9 Mon Sep 17 00:00:00 2001 From: Boris Burkov <boris(a)bur.io> Date: Fri, 13 Dec 2024 12:22:32 -0800 Subject: [PATCH] btrfs: check folio mapping after unlock in relocate_one_folio() When we call btrfs_read_folio() to bring a folio uptodate, we unlock the folio. The result of that is that a different thread can modify the mapping (like remove it with invalidate) before we call folio_lock(). This results in an invalid page and we need to try again. In particular, if we are relocating concurrently with aborting a transaction, this can result in a crash like the following: BUG: kernel NULL pointer dereference, address: 0000000000000000 PGD 0 P4D 0 Oops: 0000 [#1] SMP CPU: 76 PID: 1411631 Comm: kworker/u322:5 Workqueue: events_unbound btrfs_reclaim_bgs_work RIP: 0010:set_page_extent_mapped+0x20/0xb0 RSP: 0018:ffffc900516a7be8 EFLAGS: 00010246 RAX: ffffea009e851d08 RBX: ffffea009e0b1880 RCX: 0000000000000000 RDX: 0000000000000000 RSI: ffffc900516a7b90 RDI: ffffea009e0b1880 RBP: 0000000003573000 R08: 0000000000000001 R09: ffff88c07fd2f3f0 R10: 0000000000000000 R11: 0000194754b575be R12: 0000000003572000 R13: 0000000003572fff R14: 0000000000100cca R15: 0000000005582fff FS: 0000000000000000(0000) GS:ffff88c07fd00000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 0000000000000000 CR3: 000000407d00f002 CR4: 00000000007706f0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 PKRU: 55555554 Call Trace: <TASK> ? __die+0x78/0xc0 ? page_fault_oops+0x2a8/0x3a0 ? __switch_to+0x133/0x530 ? wq_worker_running+0xa/0x40 ? exc_page_fault+0x63/0x130 ? asm_exc_page_fault+0x22/0x30 ? set_page_extent_mapped+0x20/0xb0 relocate_file_extent_cluster+0x1a7/0x940 relocate_data_extent+0xaf/0x120 relocate_block_group+0x20f/0x480 btrfs_relocate_block_group+0x152/0x320 btrfs_relocate_chunk+0x3d/0x120 btrfs_reclaim_bgs_work+0x2ae/0x4e0 process_scheduled_works+0x184/0x370 worker_thread+0xc6/0x3e0 ? blk_add_timer+0xb0/0xb0 kthread+0xae/0xe0 ? flush_tlb_kernel_range+0x90/0x90 ret_from_fork+0x2f/0x40 ? flush_tlb_kernel_range+0x90/0x90 ret_from_fork_asm+0x11/0x20 </TASK> This occurs because cleanup_one_transaction() calls destroy_delalloc_inodes() which calls invalidate_inode_pages2() which takes the folio_lock before setting mapping to NULL. We fail to check this, and subsequently call set_extent_mapping(), which assumes that mapping != NULL (in fact it asserts that in debug mode) Note that the "fixes" patch here is not the one that introduced the race (the very first iteration of this code from 2009) but a more recent change that made this particular crash happen in practice. Fixes: e7f1326cc24e ("btrfs: set page extent mapped after read_folio in relocate_one_page") CC: stable(a)vger.kernel.org # 6.1+ Reviewed-by: Qu Wenruo <wqu(a)suse.com> Signed-off-by: Boris Burkov <boris(a)bur.io> Signed-off-by: David Sterba <dsterba(a)suse.com> diff --git a/fs/btrfs/relocation.c b/fs/btrfs/relocation.c index bf267bdfa8f8..db8b42f674b7 100644 --- a/fs/btrfs/relocation.c +++ b/fs/btrfs/relocation.c @@ -2902,6 +2902,7 @@ static int relocate_one_folio(struct reloc_control *rc, const bool use_rst = btrfs_need_stripe_tree_update(fs_info, rc->block_group->flags); ASSERT(index <= last_index); +again: folio = filemap_lock_folio(inode->i_mapping, index); if (IS_ERR(folio)) { @@ -2937,6 +2938,11 @@ static int relocate_one_folio(struct reloc_control *rc, ret = -EIO; goto release_folio; } + if (folio->mapping != inode->i_mapping) { + folio_unlock(folio); + folio_put(folio); + goto again; + } } /*

1 day, 15 hours

4
5
0 0

FAILED: patch "[PATCH] btrfs: check folio mapping after unlock in" failed to apply to 6.6-stable tree

by gregkh＠linuxfoundation.org

The patch below does not apply to the 6.6-stable tree. If someone wants it applied there, or to any other stable or longterm tree, then please email the backport, including the original git commit id to <stable(a)vger.kernel.org>. To reproduce the conflict and resubmit, you may use the following commands: git fetch https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/ linux-6.6.y git checkout FETCH_HEAD git cherry-pick -x 3e74859ee35edc33a022c3f3971df066ea0ca6b9 # <resolve conflicts, build, test, etc.> git commit -s git send-email --to '<stable(a)vger.kernel.org>' --in-reply-to '2024123042-limelight-doily-8703@gregkh' --subject-prefix 'PATCH 6.6.y' HEAD^.. Possible dependencies: thanks, greg k-h ------------------ original commit in Linus's tree ------------------ From 3e74859ee35edc33a022c3f3971df066ea0ca6b9 Mon Sep 17 00:00:00 2001 From: Boris Burkov <boris(a)bur.io> Date: Fri, 13 Dec 2024 12:22:32 -0800 Subject: [PATCH] btrfs: check folio mapping after unlock in relocate_one_folio() When we call btrfs_read_folio() to bring a folio uptodate, we unlock the folio. The result of that is that a different thread can modify the mapping (like remove it with invalidate) before we call folio_lock(). This results in an invalid page and we need to try again. In particular, if we are relocating concurrently with aborting a transaction, this can result in a crash like the following: BUG: kernel NULL pointer dereference, address: 0000000000000000 PGD 0 P4D 0 Oops: 0000 [#1] SMP CPU: 76 PID: 1411631 Comm: kworker/u322:5 Workqueue: events_unbound btrfs_reclaim_bgs_work RIP: 0010:set_page_extent_mapped+0x20/0xb0 RSP: 0018:ffffc900516a7be8 EFLAGS: 00010246 RAX: ffffea009e851d08 RBX: ffffea009e0b1880 RCX: 0000000000000000 RDX: 0000000000000000 RSI: ffffc900516a7b90 RDI: ffffea009e0b1880 RBP: 0000000003573000 R08: 0000000000000001 R09: ffff88c07fd2f3f0 R10: 0000000000000000 R11: 0000194754b575be R12: 0000000003572000 R13: 0000000003572fff R14: 0000000000100cca R15: 0000000005582fff FS: 0000000000000000(0000) GS:ffff88c07fd00000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 0000000000000000 CR3: 000000407d00f002 CR4: 00000000007706f0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 PKRU: 55555554 Call Trace: <TASK> ? __die+0x78/0xc0 ? page_fault_oops+0x2a8/0x3a0 ? __switch_to+0x133/0x530 ? wq_worker_running+0xa/0x40 ? exc_page_fault+0x63/0x130 ? asm_exc_page_fault+0x22/0x30 ? set_page_extent_mapped+0x20/0xb0 relocate_file_extent_cluster+0x1a7/0x940 relocate_data_extent+0xaf/0x120 relocate_block_group+0x20f/0x480 btrfs_relocate_block_group+0x152/0x320 btrfs_relocate_chunk+0x3d/0x120 btrfs_reclaim_bgs_work+0x2ae/0x4e0 process_scheduled_works+0x184/0x370 worker_thread+0xc6/0x3e0 ? blk_add_timer+0xb0/0xb0 kthread+0xae/0xe0 ? flush_tlb_kernel_range+0x90/0x90 ret_from_fork+0x2f/0x40 ? flush_tlb_kernel_range+0x90/0x90 ret_from_fork_asm+0x11/0x20 </TASK> This occurs because cleanup_one_transaction() calls destroy_delalloc_inodes() which calls invalidate_inode_pages2() which takes the folio_lock before setting mapping to NULL. We fail to check this, and subsequently call set_extent_mapping(), which assumes that mapping != NULL (in fact it asserts that in debug mode) Note that the "fixes" patch here is not the one that introduced the race (the very first iteration of this code from 2009) but a more recent change that made this particular crash happen in practice. Fixes: e7f1326cc24e ("btrfs: set page extent mapped after read_folio in relocate_one_page") CC: stable(a)vger.kernel.org # 6.1+ Reviewed-by: Qu Wenruo <wqu(a)suse.com> Signed-off-by: Boris Burkov <boris(a)bur.io> Signed-off-by: David Sterba <dsterba(a)suse.com> diff --git a/fs/btrfs/relocation.c b/fs/btrfs/relocation.c index bf267bdfa8f8..db8b42f674b7 100644 --- a/fs/btrfs/relocation.c +++ b/fs/btrfs/relocation.c @@ -2902,6 +2902,7 @@ static int relocate_one_folio(struct reloc_control *rc, const bool use_rst = btrfs_need_stripe_tree_update(fs_info, rc->block_group->flags); ASSERT(index <= last_index); +again: folio = filemap_lock_folio(inode->i_mapping, index); if (IS_ERR(folio)) { @@ -2937,6 +2938,11 @@ static int relocate_one_folio(struct reloc_control *rc, ret = -EIO; goto release_folio; } + if (folio->mapping != inode->i_mapping) { + folio_unlock(folio); + folio_put(folio); + goto again; + } } /*

1 day, 15 hours

3
2
0 0

[PATCH] ext4: inline: fix len overflow in ext4_prepare_inline_data

by Thadeu Lima de Souza Cascardo

When running the following code on an ext4 filesystem with inline_data feature enabled, it will lead to the bug below. fd = open("file1", O_RDWR | O_CREAT | O_TRUNC, 0666); ftruncate(fd, 30); pwrite(fd, "a", 1, (1UL << 40) + 5UL); That happens because write_begin will succeed as when ext4_generic_write_inline_data calls ext4_prepare_inline_data, pos + len will be truncated, leading to ext4_prepare_inline_data parameter to be 6 instead of 0x10000000006. Then, later when write_end is called, we hit: BUG_ON(pos + len > EXT4_I(inode)->i_inline_size); at ext4_write_inline_data. Fix it by using a loff_t type for the len parameter in ext4_prepare_inline_data instead of an unsigned int. [ 44.545164] ------------[ cut here ]------------ [ 44.545530] kernel BUG at fs/ext4/inline.c:240! [ 44.545834] Oops: invalid opcode: 0000 [#1] SMP NOPTI [ 44.546172] CPU: 3 UID: 0 PID: 343 Comm: test Not tainted 6.15.0-rc2-00003-g9080916f4863 #45 PREEMPT(full) 112853fcebfdb93254270a7959841d2c6aa2c8bb [ 44.546523] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.16.3-debian-1.16.3-2 04/01/2014 [ 44.546523] RIP: 0010:ext4_write_inline_data+0xfe/0x100 [ 44.546523] Code: 3c 0e 48 83 c7 48 48 89 de 5b 41 5c 41 5d 41 5e 41 5f 5d e9 e4 fa 43 01 5b 41 5c 41 5d 41 5e 41 5f 5d c3 cc cc cc cc cc 0f 0b <0f> 0b 0f 1f 44 00 00 55 41 57 41 56 41 55 41 54 53 48 83 ec 20 49 [ 44.546523] RSP: 0018:ffffb342008b79a8 EFLAGS: 00010216 [ 44.546523] RAX: 0000000000000001 RBX: ffff9329c579c000 RCX: 0000010000000006 [ 44.546523] RDX: 000000000000003c RSI: ffffb342008b79f0 RDI: ffff9329c158e738 [ 44.546523] RBP: 0000000000000001 R08: 0000000000000001 R09: 0000000000000000 [ 44.546523] R10: 00007ffffffff000 R11: ffffffff9bd0d910 R12: 0000006210000000 [ 44.546523] R13: fffffc7e4015e700 R14: 0000010000000005 R15: ffff9329c158e738 [ 44.546523] FS: 00007f4299934740(0000) GS:ffff932a60179000(0000) knlGS:0000000000000000 [ 44.546523] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ 44.546523] CR2: 00007f4299a1ec90 CR3: 0000000002886002 CR4: 0000000000770eb0 [ 44.546523] PKRU: 55555554 [ 44.546523] Call Trace: [ 44.546523] <TASK> [ 44.546523] ext4_write_inline_data_end+0x126/0x2d0 [ 44.546523] generic_perform_write+0x17e/0x270 [ 44.546523] ext4_buffered_write_iter+0xc8/0x170 [ 44.546523] vfs_write+0x2be/0x3e0 [ 44.546523] __x64_sys_pwrite64+0x6d/0xc0 [ 44.546523] do_syscall_64+0x6a/0xf0 [ 44.546523] ? __wake_up+0x89/0xb0 [ 44.546523] ? xas_find+0x72/0x1c0 [ 44.546523] ? next_uptodate_folio+0x317/0x330 [ 44.546523] ? set_pte_range+0x1a6/0x270 [ 44.546523] ? filemap_map_pages+0x6ee/0x840 [ 44.546523] ? ext4_setattr+0x2fa/0x750 [ 44.546523] ? do_pte_missing+0x128/0xf70 [ 44.546523] ? security_inode_post_setattr+0x3e/0xd0 [ 44.546523] ? ___pte_offset_map+0x19/0x100 [ 44.546523] ? handle_mm_fault+0x721/0xa10 [ 44.546523] ? do_user_addr_fault+0x197/0x730 [ 44.546523] ? do_syscall_64+0x76/0xf0 [ 44.546523] ? arch_exit_to_user_mode_prepare+0x1e/0x60 [ 44.546523] ? irqentry_exit_to_user_mode+0x79/0x90 [ 44.546523] entry_SYSCALL_64_after_hwframe+0x55/0x5d [ 44.546523] RIP: 0033:0x7f42999c6687 [ 44.546523] Code: 48 89 fa 4c 89 df e8 58 b3 00 00 8b 93 08 03 00 00 59 5e 48 83 f8 fc 74 1a 5b c3 0f 1f 84 00 00 00 00 00 48 8b 44 24 10 0f 05 <5b> c3 0f 1f 80 00 00 00 00 83 e2 39 83 fa 08 75 de e8 23 ff ff ff [ 44.546523] RSP: 002b:00007ffeae4a7930 EFLAGS: 00000202 ORIG_RAX: 0000000000000012 [ 44.546523] RAX: ffffffffffffffda RBX: 00007f4299934740 RCX: 00007f42999c6687 [ 44.546523] RDX: 0000000000000001 RSI: 000055ea6149200f RDI: 0000000000000003 [ 44.546523] RBP: 00007ffeae4a79a0 R08: 0000000000000000 R09: 0000000000000000 [ 44.546523] R10: 0000010000000005 R11: 0000000000000202 R12: 0000000000000000 [ 44.546523] R13: 00007ffeae4a7ac8 R14: 00007f4299b86000 R15: 000055ea61493dd8 [ 44.546523] </TASK> [ 44.546523] Modules linked in: [ 44.568501] ---[ end trace 0000000000000000 ]--- [ 44.568889] RIP: 0010:ext4_write_inline_data+0xfe/0x100 [ 44.569328] Code: 3c 0e 48 83 c7 48 48 89 de 5b 41 5c 41 5d 41 5e 41 5f 5d e9 e4 fa 43 01 5b 41 5c 41 5d 41 5e 41 5f 5d c3 cc cc cc cc cc 0f 0b <0f> 0b 0f 1f 44 00 00 55 41 57 41 56 41 55 41 54 53 48 83 ec 20 49 [ 44.570931] RSP: 0018:ffffb342008b79a8 EFLAGS: 00010216 [ 44.571356] RAX: 0000000000000001 RBX: ffff9329c579c000 RCX: 0000010000000006 [ 44.571959] RDX: 000000000000003c RSI: ffffb342008b79f0 RDI: ffff9329c158e738 [ 44.572571] RBP: 0000000000000001 R08: 0000000000000001 R09: 0000000000000000 [ 44.573148] R10: 00007ffffffff000 R11: ffffffff9bd0d910 R12: 0000006210000000 [ 44.573748] R13: fffffc7e4015e700 R14: 0000010000000005 R15: ffff9329c158e738 [ 44.574335] FS: 00007f4299934740(0000) GS:ffff932a60179000(0000) knlGS:0000000000000000 [ 44.575027] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ 44.575520] CR2: 00007f4299a1ec90 CR3: 0000000002886002 CR4: 0000000000770eb0 [ 44.576112] PKRU: 55555554 [ 44.576338] Kernel panic - not syncing: Fatal exception [ 44.576517] Kernel Offset: 0x1a600000 from 0xffffffff81000000 (relocation range: 0xffffffff80000000-0xffffffffbfffffff) Reported-by: syzbot+fe2a25dae02a207717a0(a)syzkaller.appspotmail.com Closes: https://syzkaller.appspot.com/bug?extid=fe2a25dae02a207717a0 Fixes: f19d5870cbf7 ("ext4: add normal write support for inline data") Signed-off-by: Thadeu Lima de Souza Cascardo <cascardo(a)igalia.com> Cc: stable(a)vger.kernel.org --- fs/ext4/inline.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/fs/ext4/inline.c b/fs/ext4/inline.c index 2c9b762925c72f2ff5a402b02500370bc1eb0eb1..e5e6bf0d338b965a885fb99581f9ed5e51c5257c 100644 --- a/fs/ext4/inline.c +++ b/fs/ext4/inline.c @@ -397,7 +397,7 @@ static int ext4_update_inline_data(handle_t *handle, struct inode *inode, } static int ext4_prepare_inline_data(handle_t *handle, struct inode *inode, - unsigned int len) + loff_t len) { int ret, size, no_expand; struct ext4_inode_info *ei = EXT4_I(inode); --- base-commit: 8ffd015db85fea3e15a77027fda6c02ced4d2444 change-id: 20250415-ext4-prepare-inline-overflow-8db0e747cb16 Best regards, -- Thadeu Lima de Souza Cascardo <cascardo(a)igalia.com>

3 days, 2 hours

4
3
0 0

2025

2024

2023

2022

2021

2020

2019

2018

2017

Linux-stable-mirror April 2025