This project is mirrored from https://git.kernel.org/pub/scm/linux/kernel/git/rt/linux-rt-devel.git.
Pull mirroring updated .
- Jan 23, 2025
-
-
Greg Kroah-Hartman authored
Link: https://lore.kernel.org/r/20250121174523.429119852@linuxfoundation.org Tested-by:
Florian Fainelli <florian.fainelli@broadcom.com> Tested-by:
Peter Schneider <pschneider1968@googlemail.com> Tested-by:
Shuah Khan <skhan@linuxfoundation.org> Tested-by:
SeongJae Park <sj@kernel.org> Tested-by:
Ron Economos <re@w6rz.net> Tested-by:
Linux Kernel Functional Testing <lkft@linaro.org> Tested-by:
Jon Hunter <jonathanh@nvidia.com> Tested-by:
kernelci.org bot <bot@kernelci.org> Tested-by:
Mark Brown <broonie@kernel.org> Signed-off-by:
Greg Kroah-Hartman <gregkh@linuxfoundation.org>
-
Ryan Lee authored
commit 17d0d04f upstream. attach->xmatch was not set when allocating a null profile, which is used in complain mode to allocate a learning profile. This was causing downstream failures in find_attach, which expected a valid xmatch but did not find one under a certain sequence of profile transitions in complain mode. This patch ensures the xmatch is set up properly for null profiles. Signed-off-by:
Ryan Lee <ryan.lee@canonical.com> Signed-off-by:
John Johansen <john.johansen@canonical.com> Cc: Paul Kramme <kramme@digitalmanufaktur.com> Signed-off-by:
-
Wayne Lin authored
commit b5cd418f upstream. [Why & How] Currently in dm_dp_mst_is_port_support_mode(), when valdidating mode under dsc decoding at the last DP link config, we only validate the case when there is an UFP. However, if the MSTB LCT=1, there is no UFP. Under this case, use root_link_bw_in_kbps as the available bw to compare. Link: https://gitlab.freedesktop.org/drm/amd/-/issues/3720 Fixes: fa57924c ("drm/amd/display: Refactor function dm_dp_mst_is_port_support_mode()") Cc: Mario Limonciello <mario.limonciello@amd.com> Cc: Alex Deucher <alexander.deucher@amd.com> Reviewed-by:
Jerry Zuo <jerry.zuo@amd.com> Signed-off-by:
Wayne Lin <Wayne.Lin@amd.com> Signed-off-by:
Tom Chung <chiahsuan.chung@amd.com> Tested-by:
Daniel Wheeler <daniel.wheeler@amd.com> Signed-off-by:
Alex Deucher <alexander.deucher@amd.com> (cherry picked from commit a04d9534) Cc: stable@vger.kernel.org Signed-off-by:
-
Nicholas Susanto authored
commit 3412860c upstream. Revert commit 284f141f ("drm/amd/display: Enable urgent latency adjustments for DCN35") [Why & How] Urgent latency increase caused 2.8K OLED monitor caused it to block this panel support P0. Reverting this change does not reintroduce the netflix corruption issue which it fixed. Fixes: 284f141f ("drm/amd/display: Enable urgent latency adjustments for DCN35") Reviewed-by:
Charlene Liu <charlene.liu@amd.com> Signed-off-by:
Nicholas Susanto <Nicholas.Susanto@amd.com> Signed-off-by:
-
Leo Li authored
commit ff2e4d87 upstream. [Why] Outside of a modeset/link configuration change, we should not have to wait for the panel to exit PSR. Depending on the panel and it's state, it may take multiple frames for it to exit PSR. Therefore, waiting in all scenarios may cause perceived stuttering, especially in combination with faster vblank shutdown. [How] PSR1 disable is hooked up to the vblank enable event, and vice versa. In case of vblank enable, do not wait for panel to exit PSR, but still wait in all other cases. We also avoid a call to unnecessarily change power_opts on disable - this ends up sending another command to dmcub fw. When testing against IGT, some crc tests like kms_plane_alpha_blend and amd_hotplug were failing due to CRC timeouts. This was found to be caused by the early return before HW has fully exited PSR1. Fix this by first making sure we grab a vblank reference, then waiting for panel to exit PSR1, before programming hw for CRC generation. Fixes: 58a261bf ("drm/amd/display: use a more lax vblank enable policy for older ASICs") Closes: https://gitlab.freedesktop.org/drm/amd/-/issues/3743 Reviewed-by:
Leo Li <sunpeng.li@amd.com> Signed-off-by:
-
Tom Chung authored
commit 67edb81d upstream. [Why] Replay and PSR will cause some video corruption while VRR is enabled. [How] 1. Disable the Replay and PSR while VRR is enabled. 2. Change the amdgpu_dm_crtc_vrr_active() parameter to const. Because the function will only read data from dm_crtc_state. Reviewed-by:
Roman Li <roman.li@amd.com> Tested-by:
-
Tom Chung authored
commit b0a3e840 upstream. [Why] The enum DC_PSR_VERSION_SU_1 of psr_version is 1 and DC_PSR_VERSION_UNSUPPORTED is 0xFFFFFFFF. The original code may has chance trigger the amdgpu_dm_psr_enable() while psr version is set to DC_PSR_VERSION_UNSUPPORTED. [How] Modify the condition to psr->psr_version == DC_PSR_VERSION_SU_1 Reviewed-by:
-
Christian König authored
commit af04b320 upstream. That is needed to enforce isolation between contexts. Signed-off-by:
Christian König <christian.koenig@amd.com> Reviewed-by:
-
Kenneth Feng authored
commit 90505894 upstream. Disable gfxoff with the compute workload on gfx12. This is a workaround for the opencl test failure. Signed-off-by:
Kenneth Feng <kenneth.feng@amd.com> Acked-by:
-
Gui Chengming authored
commit bd275e6c upstream. FW attestation was disabled on MP0_14_0_{2/3}. V2: Move check into is_fw_attestation_support func. (Frank) Remove DRM_WARN log info. (Alex) Fix format. (Christian) Signed-off-by:
Gui Chengming <Jack.Gui@amd.com> Reviewed-by:
Frank.Min <Frank.Min@amd.com> Reviewed-by:
-
Alex Deucher authored
commit 11510e67 upstream. Only apply when compute profile is selected. This is the only supported configuration. Selecting other profiles can lead to performane degradations. Reviewed-by:
-
Ashutosh Dixit authored
commit 79a21fc9 upstream. Add missing VISACTL mux registers required for some OA config's (e.g. RenderPipeCtrl). Fixes: cdf02fe1 ("drm/xe/oa/uapi: Add/remove OA config perf ops") Cc: stable@vger.kernel.org Signed-off-by:
Ashutosh Dixit <ashutosh.dixit@intel.com> Reviewed-by:
Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20250111021539.2920346-1-ashutosh.dixit@intel.com (cherry picked from commit c26f22dac3449d8a687237cdfc59a6445eb8f75a) Signed-off-by:
Thomas Hellström <thomas.hellstrom@linux.intel.com> Signed-off-by:
-
Matthew Brost authored
commit b1231ff7 upstream. RING_CMD_CCTL read index should be UC on iGPU parts due to L3 caching structure. Having this as WB blocks ULLS from being enabled. Change to UC to unblock ULLS on iGPU. v2: - Drop internal communications commnet, bspec is updated Cc: Balasubramani Vivekanandan <balasubramani.vivekanandan@intel.com> Cc: Michal Mrozek <michal.mrozek@intel.com> Cc: Paulo Zanoni <paulo.r.zanoni@intel.com> Cc: José Roberto de Souza <jose.souza@intel.com> Cc: stable@vger.kernel.org Fixes: 328e089b ("drm/xe: Leverage ComputeCS read L3 caching") Signed-off-by:
Matthew Brost <matthew.brost@intel.com> Acked-by:
Michal Mrozek <michal.mrozek@intel.com> Reviewed-by:
Stuart Summers <stuart.summers@intel.com> Reviewed-by:
Matt Roper <matthew.d.roper@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20250114002507.114087-1-matthew.brost@intel.com (cherry picked from commit 758debf35b9cda5450e40996991a6e4b222899bd) Signed-off-by:
-
Ville Syrjälä authored
commit 1a5401ec upstream. Mesa changed its clear color alignment from 4k to 64 bytes without informing the kernel side about the change. This is now likely to cause framebuffer creation to fail. The only thing we do with the clear color buffer in i915 is: 1. map a single page 2. read out bytes 16-23 from said page 3. unmap the page So the only requirement we really have is that those 8 bytes are all contained within one page. Thus we can deal with the Mesa regression by reducing the alignment requiment from 4k to the same 64 bytes in the kernel. We could even go as low as 32 bytes, but IIRC 64 bytes is the hardware requirement on the 3D engine side so matching that seems sensible. Note that the Mesa alignment chages were partially undone so the regression itself was already fixed on userspace side. Cc: stable@vger.kernel.org Cc: Sagar Ghuge <sagar.ghuge@intel.com> Cc: Nanley Chery <nanley.g.chery@intel.com> Reported-by:
Xi Ruoyao <xry111@xry111.site> Closes: https://gitlab.freedesktop.org/drm/i915/kernel/-/issues/13057 Closes: https://lore.kernel.org/all/45a5bba8de009347262d86a4acb27169d9ae0d9f.camel@xry111.site/ Link: https://gitlab.freedesktop.org/mesa/mesa/-/commit/17f97a69c13832a6c1b0b3aad45b06f07d4b852f Link: https://gitlab.freedesktop.org/mesa/mesa/-/commit/888f63cf1baf34bc95e847a30a041dc7798edddb Signed-off-by:
Ville Syrjälä <ville.syrjala@linux.intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20241129065014.8363-2-ville.syrjala@linux.intel.com Tested-by:
José Roberto de Souza <jose.souza@intel.com> (cherry picked from commit ed3a892e5e3d6b3f6eeb76db7c92a968aeb52f3d) Signed-off-by:
Tvrtko Ursulin <tursulin@ursulin.net> Signed-off-by:
-
Xin Li (Intel) authored
commit de31b3cd upstream. The FRED RSP0 MSR is only used for delivering events when running userspace. Linux leverages this property to reduce expensive MSR writes and optimize context switches. The kernel only writes the MSR when about to run userspace *and* when the MSR has actually changed since the last time userspace ran. This optimization is implemented by maintaining a per-CPU cache of FRED RSP0 and then checking that against the value for the top of current task stack before running userspace. However cpu_init_fred_exceptions() writes the MSR without updating the per-CPU cache. This means that the kernel might return to userspace with MSR_IA32_FRED_RSP0==0 when it needed to point to the top of current task stack. This would induce a double fault (#DF), which is bad. A context switch after cpu_init_fred_exceptions() can paper over the issue since it updates the cached value. That evidently happens most of the time explaining how this bug got through. Fix the bug through resynchronizing the FRED RSP0 MSR with its per-CPU cache in cpu_init_fred_exceptions(). Fixes: fe85ee39 ("x86/entry: Set FRED RSP0 on return to userspace instead of context switch") Signed-off-by:
Xin Li (Intel) <xin@zytor.com> Signed-off-by:
Dave Hansen <dave.hansen@linux.intel.com> Acked-by:
-
Frederic Weisbecker authored
commit de3ced72 upstream. Commit 2522c84db513 ("timers/migration: Fix another race between hotplug and idle entry/exit") fixed yet another race between idle exit and CPU hotplug up leading to a wrong "0" value migrator assigned to the top level. However there is yet another situation that remains unhandled: [GRP0:0] migrator = TMIGR_NONE active = NONE groupmask = 1 / \ \ 0 1 2..7 idle idle idle 0) The system is fully idle. [GRP0:0] migrator = CPU 0 active = CPU 0 groupmask = 1 / \ \ 0 1 2..7 active idle idle 1) CPU 0 is activating. It has done the cmpxchg on the top's ->migr_state but it hasn't yet returned to __walk_groups(). [GRP0:0] migrator = CPU 0 active = CPU 0, CPU 1 groupmask = 1 / \ \ 0 1 2..7 active active idle 2) CPU 1 is activating. CPU 0 stays the migrator (still stuck in __walk_groups(), delayed by #VMEXIT for example). [GRP1:0] migrator = TMIGR_NONE active = NONE groupmask = 1 / \ [GRP0:0] [GRP0:1] migrator = CPU 0 migrator = TMIGR_NONE active = CPU 0, CPU1 active = NONE groupmask = 1 groupmask = 2 / \ \ 0 1 2..7 8 active active idle !online 3) CPU 8 is preparing to boot. CPUHP_TMIGR_PREPARE is being ran by CPU 1 which has created the GRP0:1 and the new top GRP1:0 connected to GRP0:1 and GRP0:0. CPU 1 hasn't yet propagated its activation up to GRP1:0. [GRP1:0] migrator = GRP0:0 active = GRP0:0 groupmask = 1 / \ [GRP0:0] [GRP0:1] migrator = CPU 0 migrator = TMIGR_NONE active = CPU 0, CPU1 active = NONE groupmask = 1 groupmask = 2 / \ \ 0 1 2..7 8 active active idle !online 4) CPU 0 finally resumed after its #VMEXIT. It's in __walk_groups() returning from tmigr_cpu_active(). The new top GRP1:0 is visible and fetched and the pre-initialized groupmask of GRP0:0 is also visible. As a result tmigr_active_up() is called to GRP1:0 with GRP0:0 as active and migrator. CPU 0 is returning to __walk_groups() but suffers again a #VMEXIT. [GRP1:0] migrator = GRP0:0 active = GRP0:0 groupmask = 1 / \ [GRP0:0] [GRP0:1] migrator = CPU 0 migrator = TMIGR_NONE active = CPU 0, CPU1 active = NONE groupmask = 1 groupmask = 2 / \ \ 0 1 2..7 8 active active idle !online 5) CPU 1 propagates its activation of GRP0:0 to GRP1:0. This has no effect since CPU 0 did it already. [GRP1:0] migrator = GRP0:0 active = GRP0:0, GRP0:1 groupmask = 1 / \ [GRP0:0] [GRP0:1] migrator = CPU 0 migrator = CPU 8 active = CPU 0, CPU1 active = CPU 8 groupmask = 1 groupmask = 2 / \ \ \ 0 1 2..7 8 active active idle active 6) CPU 1 links CPU 8 to its group. CPU 8 boots and goes through CPUHP_AP_TMIGR_ONLINE which propagates activation. [GRP2:0] migrator = TMIGR_NONE active = NONE groupmask = 1 / \ [GRP1:0] [GRP1:1] migrator = GRP0:0 migrator = TMIGR_NONE active = GRP0:0, GRP0:1 active = NONE groupmask = 1 groupmask = 2 / \ [GRP0:0] [GRP0:1] [GRP0:2] migrator = CPU 0 migrator = CPU 8 migrator = TMIGR_NONE active = CPU 0, CPU1 active = CPU 8 active = NONE groupmask = 1 groupmask = 2 groupmask = 0 / \ \ \ 0 1 2..7 8 64 active active idle active !online 7) CPU 64 is booting. CPUHP_TMIGR_PREPARE is being ran by CPU 1 which has created the GRP1:1, GRP0:2 and the new top GRP2:0 connected to GRP1:1 and GRP1:0. CPU 1 hasn't yet propagated its activation up to GRP2:0. [GRP2:0] migrator = 0 (!!!) active = NONE groupmask = 1 / \ [GRP1:0] [GRP1:1] migrator = GRP0:0 migrator = TMIGR_NONE active = GRP0:0, GRP0:1 active = NONE groupmask = 1 groupmask = 2 / \ [GRP0:0] [GRP0:1] [GRP0:2] migrator = CPU 0 migrator = CPU 8 migrator = TMIGR_NONE active = CPU 0, CPU1 active = CPU 8 active = NONE groupmask = 1 groupmask = 2 groupmask = 0 / \ \ \ 0 1 2..7 8 64 active active idle active !online 8) CPU 0 finally resumed after its #VMEXIT. It's in __walk_groups() returning from tmigr_cpu_active(). The new top GRP2:0 is visible and fetched but the pre-initialized groupmask of GRP1:0 is not because no ordering made its initialization visible. As a result tmigr_active_up() may be called to GRP2:0 with a "0" child's groumask. Leaving the timers ignored for ever when the system is fully idle. The race is highly theoretical and perhaps impossible in practice but the groupmask of the child is not the only concern here as the whole initialization of the child is not guaranteed to be visible to any tree walker racing against hotplug (idle entry/exit, remote handling, etc...). Although the current code layout seem to be resilient to such hazards, this doesn't tell much about the future. Fix this with enforcing address dependency between group initialization and the write/read to the group's parent's pointer. Fortunately that doesn't involve any barrier addition in the fast paths. Fixes: 10a0e6f3 ("timers/migration: Move hierarchy setup into cpuhotplug prepare callback") Signed-off-by:
Frederic Weisbecker <frederic@kernel.org> Signed-off-by:
Thomas Gleixner <tglx@linutronix.de> Cc: stable@vger.kernel.org Link: https://lore.kernel.org/all/20250114231507.21672-3-frederic@kernel.org Signed-off-by:
-
Frederic Weisbecker authored
commit b729cc1e upstream. Commit 10a0e6f3 ("timers/migration: Move hierarchy setup into cpuhotplug prepare callback") fixed a race between idle exit and CPU hotplug up leading to a wrong "0" value migrator assigned to the top level. However there is still a situation that remains unhandled: [GRP0:0] migrator = TMIGR_NONE active = NONE groupmask = 0 / \ \ 0 1 2..7 idle idle idle 0) The system is fully idle. [GRP0:0] migrator = CPU 0 active = CPU 0 groupmask = 0 / \ \ 0 1 2..7 active idle idle 1) CPU 0 is activating. It has done the cmpxchg on the top's ->migr_state but it hasn't yet returned to __walk_groups(). [GRP0:0] migrator = CPU 0 active = CPU 0, CPU 1 groupmask = 0 / \ \ 0 1 2..7 active active idle 2) CPU 1 is activating. CPU 0 stays the migrator (still stuck in __walk_groups(), delayed by #VMEXIT for example). [GRP1:0] migrator = TMIGR_NONE active = NONE groupmask = 0 / \ [GRP0:0] [GRP0:1] migrator = CPU 0 migrator = TMIGR_NONE active = CPU 0, CPU1 active = NONE groupmask = 2 groupmask = 1 / \ \ 0 1 2..7 8 active active idle !online 3) CPU 8 is preparing to boot. CPUHP_TMIGR_PREPARE is being ran by CPU 1 which has created the GRP0:1 and the new top GRP1:0 connected to GRP0:1 and GRP0:0. The groupmask of GRP0:0 is now 2. CPU 1 hasn't yet propagated its activation up to GRP1:0. [GRP1:0] migrator = 0 (!!!) active = NONE groupmask = 0 / \ [GRP0:0] [GRP0:1] migrator = CPU 0 migrator = TMIGR_NONE active = CPU 0, CPU1 active = NONE groupmask = 2 groupmask = 1 / \ \ 0 1 2..7 8 active active idle !online 4) CPU 0 finally resumed after its #VMEXIT. It's in __walk_groups() returning from tmigr_cpu_active(). The new top GRP1:0 is visible and fetched but the freshly updated groupmask of GRP0:0 may not be visible due to lack of ordering! As a result tmigr_active_up() is called to GRP0:0 with a child's groupmask of "0". This buggy "0" groupmask then becomes the migrator for GRP1:0 forever. As a result, timers on a fully idle system get ignored. One possible fix would be to define TMIGR_NONE as "0" so that such a race would have no effect. And after all TMIGR_NONE doesn't need to be anything else. However this would leave an uncomfortable state machine where gears happen not to break by chance but are vulnerable to future modifications. Keep TMIGR_NONE as is instead and pre-initialize to "1" the groupmask of any newly created top level. This groupmask is guaranteed to be visible upon fetching the corresponding group for the 1st time: _ By the upcoming CPU thanks to CPU hotplug synchronization between the control CPU (BP) and the booting one (AP). _ By the control CPU since the groupmask and parent pointers are initialized locally. _ By all CPUs belonging to the same group than the control CPU because they must wait for it to ever become idle before needing to walk to the new top. The cmpcxhg() on ->migr_state then makes sure its groupmask is visible. With this pre-initialization, it is guaranteed that if a future top level is linked to an old one, it is walked through with a valid groupmask. Fixes: 10a0e6f3 ("timers/migration: Move hierarchy setup into cpuhotplug prepare callback") Signed-off-by:
-
Koichiro Den authored
commit 2f8dea16 upstream. Consider a scenario where a CPU transitions from CPUHP_ONLINE to halfway through a CPU hotunplug down to CPUHP_HRTIMERS_PREPARE, and then back to CPUHP_ONLINE: Since hrtimers_prepare_cpu() does not run, cpu_base.hres_active remains set to 1 throughout. However, during a CPU unplug operation, the tick and the clockevents are shut down at CPUHP_AP_TICK_DYING. On return to the online state, for instance CFS incorrectly assumes that the hrtick is already active, and the chance of the clockevent device to transition to oneshot mode is also lost forever for the CPU, unless it goes back to a lower state than CPUHP_HRTIMERS_PREPARE once. This round-trip reveals another issue; cpu_base.online is not set to 1 after the transition, which appears as a WARN_ON_ONCE in enqueue_hrtimer(). Aside of that, the bulk of the per CPU state is not reset either, which means there are dangling pointers in the worst case. Address this by adding a corresponding startup() callback, which resets the stale per CPU state and sets the online flag. [ tglx: Make the new callback unconditionally available, remove the online modification in the prepare() callback and clear the remaining state in the starting callback instead of the prepare callback ] Fixes: 5c0930cc ("hrtimers: Push pending hrtimers away from outgoing CPU earlier") Signed-off-by:
Koichiro Den <koichiro.den@canonical.com> Signed-off-by:
-
Tomas Krcka authored
commit 35cb2c6c upstream. The following call-chain leads to enabling interrupts in a nested interrupt disabled section: irq_set_vcpu_affinity() irq_get_desc_lock() raw_spin_lock_irqsave() <--- Disable interrupts its_irq_set_vcpu_affinity() guard(raw_spinlock_irq) <--- Enables interrupts when leaving the guard() irq_put_desc_unlock() <--- Warns because interrupts are enabled This was broken in commit b97e8a2f, which replaced the original raw_spin_[un]lock() pair with guard(raw_spinlock_irq). Fix the issue by using guard(raw_spinlock). [ tglx: Massaged change log ] Fixes: b97e8a2f ("irqchip/gic-v3-its: Fix potential race condition in its_vlpi_prop_update()") Signed-off-by:
Tomas Krcka <krckatom@amazon.de> Signed-off-by:
Marc Zyngier <maz@kernel.org> Cc: stable@vger.kernel.org Link: https://lore.kernel.org/all/20241230150825.62894-1-krckatom@amazon.de Signed-off-by:
-
Yogesh Lal authored
commit 0d62a49a upstream. When a CPU attempts to enter low power mode, it disables the redistributor and Group 1 interrupts and reinitializes the system registers upon wakeup. If the transition into low power mode fails, then the CPU_PM framework invokes the PM notifier callback with CPU_PM_ENTER_FAILED to allow the drivers to undo the state changes. The GIC V3 driver ignores CPU_PM_ENTER_FAILED, which leaves the GIC in disabled state. Handle CPU_PM_ENTER_FAILED in the same way as CPU_PM_EXIT to restore normal operation. [ tglx: Massage change log, add Fixes tag ] Fixes: 3708d52f ("irqchip: gic-v3: Implement CPU PM notifier") Signed-off-by:
Yogesh Lal <quic_ylal@quicinc.com> Signed-off-by:
-
Joe Hattori authored
commit 9322d191 upstream. platform_irqchip_probe() leaks a OF node when irq_init_cb() fails. Fix it by declaring par_np with the __free(device_node) cleanup construct. This bug was found by an experimental static analysis tool that I am developing. Fixes: f8410e62 ("irqchip: Add IRQCHIP_PLATFORM_DRIVER_BEGIN/END and IRQCHIP_MATCH helper macros") Signed-off-by:
Joe Hattori <joe@pf.is.s.u-tokyo.ac.jp> Signed-off-by:
-
Steven Rostedt authored
commit 60295b94 upstream. Tracing tools like perf and trace-cmd read the /sys/kernel/tracing/events/*/*/format files to know how to parse the data and also how to print it. For the "print fmt" portion of that file, if anything uses an enum that is not exported to the tracing system, user space will not be able to parse it. The GFP flags use to be defines, and defines get translated in the print fmt sections. But now they are converted to use enums, which is not. The mm_page_alloc trace event format use to have: print fmt: "page=%p pfn=0x%lx order=%d migratetype=%d gfp_flags=%s", REC->pfn != -1UL ? (((struct page *)vmemmap_base) + (REC->pfn)) : ((void *)0), REC->pfn != -1UL ? REC->pfn : 0, REC->order, REC->migratetype, (REC->gfp_flags) ? __print_flags(REC->gfp_flags, "|", {( unsigned long)(((((((( gfp_t)(0x400u|0x800u)) | (( gfp_t)0x40u) | (( gfp_t)0x80u) | (( gfp_t)0x100000u)) | (( gfp_t)0x02u)) | (( gfp_t)0x08u) | (( gfp_t)0)) | (( gfp_t)0x40000u) | (( gfp_t)0x80000u) | (( gfp_t)0x2000u)) & ~(( gfp_t)(0x400u|0x800u))) | (( gfp_t)0x400u)), "GFP_TRANSHUGE"}, {( unsigned long)((((((( gfp_t)(0x400u|0x800u)) | (( gfp_t)0x40u) | (( gfp_t)0x80u) | (( gfp_t)0x100000u)) | (( gfp_t)0x02u)) | (( gfp_t)0x08u) | (( gfp_t)0)) ... Where the GFP values are shown and not their names. But after the GFP flags were converted to use enums, it has: print fmt: "page=%p pfn=0x%lx order=%d migratetype=%d gfp_flags=%s", REC->pfn != -1UL ? (vmemmap + (REC->pfn)) : ((void *)0), REC->pfn != -1UL ? REC->pfn : 0, REC->order, REC->migratetype, (REC->gfp_flags) ? __print_flags(REC->gfp_flags, "|", {( unsigned long)(((((((( gfp_t)(((((1UL))) << (___GFP_DIRECT_RECLAIM_BIT))|((((1UL))) << (___GFP_KSWAPD_RECLAIM_BIT)))) | (( gfp_t)((((1UL))) << (___GFP_IO_BIT))) | (( gfp_t)((((1UL))) << (___GFP_FS_BIT))) | (( gfp_t)((((1UL))) << (___GFP_HARDWALL_BIT)))) | (( gfp_t)((((1UL))) << (___GFP_HIGHMEM_BIT)))) | (( gfp_t)((((1UL))) << (___GFP_MOVABLE_BIT))) | (( gfp_t)0)) | (( gfp_t)((((1UL))) << (___GFP_COMP_BIT))) ... Where the enums names like ___GFP_KSWAPD_RECLAIM_BIT are shown and not their values. User space has no way to convert these names to their values and the output will fail to parse. What is shown is now: mm_page_alloc: page=0xffffffff981685f3 pfn=0x1d1ac1 order=0 migratetype=1 gfp_flags=0x140cca The TRACE_DEFINE_ENUM() macro was created to handle enums in the print fmt files. This causes them to be replaced at boot up with the numbers, so that user space tooling can parse it. By using this macro, the output is back to the human readable: mm_page_alloc: page=0xffffffff981685f3 pfn=0x122233 order=0 migratetype=1 gfp_flags=GFP_HIGHUSER_MOVABLE|__GFP_COMP Cc: stable@vger.kernel.org Cc: Masami Hiramatsu <mhiramat@kernel.org> Cc: Mark Rutland <mark.rutland@arm.com> Cc: Mathieu Desnoyers <mathieu.desnoyers@efficios.com> Cc: Andrew Morton <akpm@linux-foundation.org> Cc: Veronika Molnarova <vmolnaro@redhat.com> Cc: Suren Baghdasaryan <surenb@google.com> Cc: Linus Torvalds <torvalds@linux-foundation.org> Link: https://lore.kernel.org/20250116214438.749504792@goodmis.org Reported-by:
Michael Petlan <mpetlan@redhat.com> Closes: https://lore.kernel.org/all/87be5f7c-1a0-dad-daa0-54e342efaea7@redhat.com/ Fixes: 772dd034 ("mm: enumerate all gfp flags") Signed-off-by:
Steven Rostedt (Google) <rostedt@goodmis.org> Signed-off-by:
-
Donet Tom authored
commit bd3d56ff upstream. When MGLRU is enabled, the pgdemote_kswapd, pgdemote_direct, and pgdemote_khugepaged stats in vmstat are not being updated. Commit f77f0c75 ("mm,memcg: provide per-cgroup counters for NUMA balancing operations") moved the pgdemote vmstat update from demote_folio_list() to shrink_inactive_list(), which is in the normal LRU path. As a result, the pgdemote stats are updated correctly for the normal LRU but not for MGLRU. To address this, we have added the pgdemote stat update in the evict_folios() function, which is in the MGLRU path. With this patch, the pgdemote stats will now be updated correctly when MGLRU is enabled. Without this patch vmstat output when MGLRU is enabled ====================================================== pgdemote_kswapd 0 pgdemote_direct 0 pgdemote_khugepaged 0 With this patch vmstat output when MGLRU is enabled =================================================== pgdemote_kswapd 43234 pgdemote_direct 4691 pgdemote_khugepaged 0 Link: https://lkml.kernel.org/r/20250109060540.451261-1-donettom@linux.ibm.com Fixes: f77f0c75 ("mm,memcg: provide per-cgroup counters for NUMA balancing operations") Signed-off-by:
Donet Tom <donettom@linux.ibm.com> Acked-by:
Yu Zhao <yuzhao@google.com> Tested-by:
Li Zhijian <lizhijian@fujitsu.com> Reviewed-by:
Andrew Morton <akpm@linux-foundation.org> Signed-off-by:
-
Ryan Roberts authored
commit 0cef0bb8 upstream. When mremap()ing a memory region previously registered with userfaultfd as write-protected but without UFFD_FEATURE_EVENT_REMAP, an inconsistency in flag clearing leads to a mismatch between the vma flags (which have uffd-wp cleared) and the pte/pmd flags (which do not have uffd-wp cleared). This mismatch causes a subsequent mprotect(PROT_WRITE) to trigger a warning in page_table_check_pte_flags() due to setting the pte to writable while uffd-wp is still set. Fix this by always explicitly clearing the uffd-wp pte/pmd flags on any such mremap() so that the values are consistent with the existing clearing of VM_UFFD_WP. Be careful to clear the logical flag regardless of its physical form; a PTE bit, a swap PTE bit, or a PTE marker. Cover PTE, huge PMD and hugetlb paths. Link: https://lkml.kernel.org/r/20250107144755.1871363-2-ryan.roberts@arm.com Co-developed-by:
Mikołaj Lenczewski <miko.lenczewski@arm.com> Signed-off-by:
Ryan Roberts <ryan.roberts@arm.com> Closes: https://lore.kernel.org/linux-mm/810b44a8-d2ae-4107-b665-5a42eae2d948@arm.com/ Fixes: 63b2d417 ("userfaultfd: wp: add the writeprotect API to userfaultfd ioctl") Cc: David Hildenbrand <david@redhat.com> Cc: Jann Horn <jannh@google.com> Cc: Liam R. Howlett <Liam.Howlett@Oracle.com> Cc: Lorenzo Stoakes <lorenzo.stoakes@oracle.com> Cc: Mark Rutland <mark.rutland@arm.com> Cc: Muchun Song <muchun.song@linux.dev> Cc: Peter Xu <peterx@redhat.com> Cc: Shuah Khan <shuah@kernel.org> Cc: Vlastimil Babka <vbabka@suse.cz> Cc: <stable@vger.kernel.org> Signed-off-by:
-
Leo Li authored
commit 35ca53b7 upstream. [Why] There should not be any need to revalidate bandwidth on memory placement change, since the fb is expected to be pinned to DCN-accessable memory before scanout. For APU it's DRAM, and DGPU, it's VRAM. However, async flips + memory type change needs to be rejected. [How] Do not set lock_and_validation_needed on mem_type change. Instead, reject an async_flip request if the crtc's buffer(s) changed mem_type. This may fix stuttering/corruption experienced with PSR SU and PSR1 panels, if the compositor allocates fbs in both VRAM carveout and GTT and flips between them. Fixes: a7c0cad0 ("drm/amd/display: ensure async flips are only accepted for fast updates") Reviewed-by:
-
Ryan Roberts authored
commit a32bf5bb upstream. After commit b1f20206 ("mm: remap unused subpages to shared zeropage when splitting isolated thp"), cow test cases involving swapping out THPs via madvise(MADV_PAGEOUT) started to be skipped due to the subsequent check via pagemap determining that the memory was not actually swapped out. Logs similar to this were emitted: ... # [RUN] Basic COW after fork() ... with swapped-out, PTE-mapped THP (16 kB) ok 2 # SKIP MADV_PAGEOUT did not work, is swap enabled? # [RUN] Basic COW after fork() ... with single PTE of swapped-out THP (16 kB) ok 3 # SKIP MADV_PAGEOUT did not work, is swap enabled? # [RUN] Basic COW after fork() ... with swapped-out, PTE-mapped THP (32 kB) ok 4 # SKIP MADV_PAGEOUT did not work, is swap enabled? ... The commit in question introduces the behaviour of scanning THPs and if their content is predominantly zero, it splits them and replaces the pages which are wholly zero with the zero page. These cow test cases were getting caught up in this. So let's avoid that by filling the contents of all allocated memory with a non-zero value. With this in place, the tests are passing again. Link: https://lkml.kernel.org/r/20250107142555.1870101-1-ryan.roberts@arm.com Fixes: b1f20206 ("mm: remap unused subpages to shared zeropage when splitting isolated thp") Signed-off-by:
David Hildenbrand <david@redhat.com> Cc: Usama Arif <usamaarif642@gmail.com> Cc: Yu Zhao <yuzhao@google.com> Cc: <stable@vger.kernel.org> Signed-off-by:
-
Guo Weikang authored
commit 76d5d4c5 upstream. kmemleak_alloc_percpu gives an incorrect min_count parameter, causing percpu memory to be considered a gray object. Link: https://lkml.kernel.org/r/20241227092311.3572500-1-guoweikang.kernel@gmail.com Fixes: 8c868592 ("mm/kmemleak: use IS_ERR_PCPU() for pointer in the percpu address space") Signed-off-by:
Guo Weikang <guoweikang.kernel@gmail.com> Acked-by:
Uros Bizjak <ubizjak@gmail.com> Acked-by:
Catalin Marinas <catalin.marinas@arm.com> Cc: Guo Weikang <guoweikang.kernel@gmail.com> Cc: <stable@vger.kernel.org> Signed-off-by:
-
Xiaolei Wang authored
commit 726efa92 upstream. Currently imx8mp_blk_ctrl_remove() will continue the for loop until an out-of-bounds exception occurs. pstate: 60000005 (nZCv daif -PAN -UAO -TCO -DIT -SSBS BTYPE=--) pc : dev_pm_domain_detach+0x8/0x48 lr : imx8mp_blk_ctrl_shutdown+0x58/0x90 sp : ffffffc084f8bbf0 x29: ffffffc084f8bbf0 x28: ffffff80daf32ac0 x27: 0000000000000000 x26: ffffffc081658d78 x25: 0000000000000001 x24: ffffffc08201b028 x23: ffffff80d0db9490 x22: ffffffc082340a78 x21: 00000000000005b0 x20: ffffff80d19bc180 x19: 000000000000000a x18: ffffffffffffffff x17: ffffffc080a39e08 x16: ffffffc080a39c98 x15: 4f435f464f006c72 x14: 0000000000000004 x13: ffffff80d0172110 x12: 0000000000000000 x11: ffffff80d0537740 x10: ffffff80d05376c0 x9 : ffffffc0808ed2d8 x8 : ffffffc084f8bab0 x7 : 0000000000000000 x6 : 0000000000000000 x5 : ffffff80d19b9420 x4 : fffffffe03466e60 x3 : 0000000080800077 x2 : 0000000000000000 x1 : 0000000000000001 x0 : 0000000000000000 Call trace: dev_pm_domain_detach+0x8/0x48 platform_shutdown+0x2c/0x48 device_shutdown+0x158/0x268 kernel_restart_prepare+0x40/0x58 kernel_kexec+0x58/0xe8 __do_sys_reboot+0x198/0x258 __arm64_sys_reboot+0x2c/0x40 invoke_syscall+0x5c/0x138 el0_svc_common.constprop.0+0x48/0xf0 do_el0_svc+0x24/0x38 el0_svc+0x38/0xc8 el0t_64_sync_handler+0x120/0x130 el0t_64_sync+0x190/0x198 Code: 8128c2d0 ffffffc0 aa1e03e9 d503201f Fixes: 556f5cf9 ("soc: imx: add i.MX8MP HSIO blk-ctrl") Cc: stable@vger.kernel.org Signed-off-by:
Xiaolei Wang <xiaolei.wang@windriver.com> Reviewed-by:
Lucas Stach <l.stach@pengutronix.de> Reviewed-by:
Fabio Estevam <festevam@gmail.com> Reviewed-by:
Frank Li <Frank.Li@nxp.com> Link: https://lore.kernel.org/r/20250115014118.4086729-1-xiaolei.wang@windriver.com Signed-off-by:
Ulf Hansson <ulf.hansson@linaro.org> Signed-off-by:
-
Suren Baghdasaryan authored
commit 4bbb6df6 upstream. Currently vma test is failing because of the new vma_assert_attached() assertion. The check is failing because previous refcount_set() inside vma_mark_attached() is a NoOp. Fix the definition of atomic_set() to correctly set the value of the atomic. Link: https://lkml.kernel.org/r/20241227222220.1726384-1-surenb@google.com Fixes: 9325b8b5 ("tools: add skeleton code for userland testing of VMA logic") Signed-off-by:
Suren Baghdasaryan <surenb@google.com> Reviewed-by:
Lorenzo Stoakes <lorenzo.stoakes@oracle.com> Cc: Jann Horn <jannh@google.com> Cc: Liam R. Howlett <Liam.Howlett@Oracle.com> Cc: Vlastimil Babka <vbabka@suse.cz> Cc: <stable@vger.kernel.org> Signed-off-by:
-
Sean Anderson authored
commit 9860370c upstream. irq_chip functions may be called in raw spinlock context. Therefore, we must also use a raw spinlock for our own internal locking. This fixes the following lockdep splat: [ 5.349336] ============================= [ 5.353349] [ BUG: Invalid wait context ] [ 5.357361] 6.13.0-rc5+ #69 Tainted: G W [ 5.363031] ----------------------------- [ 5.367045] kworker/u17:1/44 is trying to lock: [ 5.371587] ffffff88018b02c0 (&chip->gpio_lock){....}-{3:3}, at: xgpio_irq_unmask (drivers/gpio/gpio-xilinx.c:433 (discriminator 8)) [ 5.380079] other info that might help us debug this: [ 5.385138] context-{5:5} [ 5.387762] 5 locks held by kworker/u17:1/44: [ 5.392123] #0: ffffff8800014958 ((wq_completion)events_unbound){+.+.}-{0:0}, at: process_one_work (kernel/workqueue.c:3204) [ 5.402260] #1: ffffffc082fcbdd8 (deferred_probe_work){+.+.}-{0:0}, at: process_one_work (kernel/workqueue.c:3205) [ 5.411528] #2: ffffff880172c900 (&dev->mutex){....}-{4:4}, at: __device_attach (drivers/base/dd.c:1006) [ 5.419929] #3: ffffff88039c8268 (request_class#2){+.+.}-{4:4}, at: __setup_irq (kernel/irq/internals.h:156 kernel/irq/manage.c:1596) [ 5.428331] #4: ffffff88039c80c8 (lock_class#2){....}-{2:2}, at: __setup_irq (kernel/irq/manage.c:1614) [ 5.436472] stack backtrace: [ 5.439359] CPU: 2 UID: 0 PID: 44 Comm: kworker/u17:1 Tainted: G W 6.13.0-rc5+ #69 [ 5.448690] Tainted: [W]=WARN [ 5.451656] Hardware name: xlnx,zynqmp (DT) [ 5.455845] Workqueue: events_unbound deferred_probe_work_func [ 5.461699] Call trace: [ 5.464147] show_stack+0x18/0x24 C [ 5.467821] dump_stack_lvl (lib/dump_stack.c:123) [ 5.471501] dump_stack (lib/dump_stack.c:130) [ 5.474824] __lock_acquire (kernel/locking/lockdep.c:4828 kernel/locking/lockdep.c:4898 kernel/locking/lockdep.c:5176) [ 5.478758] lock_acquire (arch/arm64/include/asm/percpu.h:40 kernel/locking/lockdep.c:467 kernel/locking/lockdep.c:5851 kernel/locking/lockdep.c:5814) [ 5.482429] _raw_spin_lock_irqsave (include/linux/spinlock_api_smp.h:111 kernel/locking/spinlock.c:162) [ 5.486797] xgpio_irq_unmask (drivers/gpio/gpio-xilinx.c:433 (discriminator 8)) [ 5.490737] irq_enable (kernel/irq/internals.h:236 kernel/irq/chip.c:170 kernel/irq/chip.c:439 kernel/irq/chip.c:432 kernel/irq/chip.c:345) [ 5.494060] __irq_startup (kernel/irq/internals.h:241 kernel/irq/chip.c:180 kernel/irq/chip.c:250) [ 5.497645] irq_startup (kernel/irq/chip.c:270) [ 5.501143] __setup_irq (kernel/irq/manage.c:1807) [ 5.504728] request_threaded_irq (kernel/irq/manage.c:2208) Fixes: a32c7cae ("gpio: gpio-xilinx: Add interrupt support") Signed-off-by:
Sean Anderson <sean.anderson@linux.dev> Cc: stable@vger.kernel.org Link: https://lore.kernel.org/r/20250110163354.2012654-1-sean.anderson@linux.dev Signed-off-by:
Bartosz Golaszewski <bartosz.golaszewski@linaro.org> Signed-off-by:
-
Rik van Riel authored
commit cbc5dde0 upstream. Since commit 5cbcb62d ("fs/proc: fix softlockup in __read_vmcore") the number of softlockups in __read_vmcore at kdump time have gone down, but they still happen sometimes. In a memory constrained environment like the kdump image, a softlockup is not just a harmless message, but it can interfere with things like RCU freeing memory, causing the crashdump to get stuck. The second loop in __read_vmcore has a lot more opportunities for natural sleep points, like scheduling out while waiting for a data write to happen, but apparently that is not always enough. Add a cond_resched() to the second loop in __read_vmcore to (hopefully) get rid of the softlockups. Link: https://lkml.kernel.org/r/20250110102821.2a37581b@fangorn Fixes: 5cbcb62d ("fs/proc: fix softlockup in __read_vmcore") Signed-off-by:
Rik van Riel <riel@surriel.com> Reported-by:
Breno Leitao <leitao@debian.org> Cc: Baoquan He <bhe@redhat.com> Cc: Dave Young <dyoung@redhat.com> Cc: Vivek Goyal <vgoyal@redhat.com> Cc: <stable@vger.kernel.org> Signed-off-by:
-
Marco Nelissen authored
commit f505e6c9 upstream. On 32-bit kernels, folio_seek_hole_data() was inadvertently truncating a 64-bit value to 32 bits, leading to a possible infinite loop when writing to an xfs filesystem. Link: https://lkml.kernel.org/r/20250102190540.1356838-1-marco.nelissen@gmail.com Fixes: 54fa39ac ("iomap: use mapping_seek_hole_data") Signed-off-by:
Marco Nelissen <marco.nelissen@gmail.com> Cc: Matthew Wilcox (Oracle) <willy@infradead.org> Cc: <stable@vger.kernel.org> Signed-off-by:
-
Paul Fertser authored
commit 9e2bbab9 upstream. Obtaining RTNL lock in a response handler is not allowed since it runs in an atomic softirq context. Postpone setting the MAC address by adding a dedicated step to the configuration FSM. Fixes: 79007134 ("net/ncsi: change from ndo_set_mac_address to dev_set_mac_address") Cc: stable@vger.kernel.org Link: https://lore.kernel.org/20241129-potin-revert-ncsi-set-mac-addr-v1-1-94ea2cb596af@gmail.com Signed-off-by:
Paul Fertser <fercerpav@gmail.com> Tested-by:
Potin Lai <potin.lai.pt@gmail.com> Link: https://patch.msgid.link/20250109145054.30925-1-fercerpav@gmail.com Signed-off-by:
Jakub Kicinski <kuba@kernel.org> Signed-off-by:
-
Takashi Iwai authored
commit 35243fc7 upstream. Macbook 5,1 with MCP79 lost its backlight control since the recent change for supporting GSP-RM; it rewrote the whole nv50 backlight control code and each display engine is supposed to have an entry for IOR bl callback, but it didn't cover mcp77. This patch adds the missing bl entry initialization for mcp77 display engine to recover the backlight control. Fixes: 2274ce7e ("drm/nouveau/disp: add output backlight control methods") Cc: stable@vger.kernel.org Link: https://bugzilla.suse.com/show_bug.cgi?id=1223838 Signed-off-by:
Takashi Iwai <tiwai@suse.de> Signed-off-by:
Danilo Krummrich <dakr@kernel.org> Link: https://patchwork.freedesktop.org/patch/msgid/20250102114944.11499-1-tiwai@suse.de Signed-off-by:
-
Dave Airlie authored
commit 1f9910b4 upstream. The fence sync logic doesn't handle a fence sync across devices as it tries to write to a channel offset from one device into the fence bo from a different device, which won't work so well. This patch fixes that to avoid using the sync path in the case where the fences come from different nouveau drm devices. This works fine on a single device as the fence bo is shared across the devices, and mapped into each channels vma space, the channel offsets are therefore okay to pass between sides, so one channel can sync on the seqnos from the other by using the offset into it's vma. Signed-off-by:
Dave Airlie <airlied@redhat.com> Cc: stable@vger.kernel.org Reviewed-by:
Ben Skeggs <bskeggs@nvidia.com> [ Fix compilation issue; remove version log from commit messsage. - Danilo ] Signed-off-by:
-
Stefano Garzarella authored
commit 91751e24 upstream. Recent reports have shown how we sometimes call vsock_*_has_data() when a vsock socket has been de-assigned from a transport (see attached links), but we shouldn't. Previous commits should have solved the real problems, but we may have more in the future, so to avoid null-ptr-deref, we can return 0 (no space, no data available) but with a warning. This way the code should continue to run in a nearly consistent state and have a warning that allows us to debug future problems. Fixes: c0cfa2d8 ("vsock: add multi-transports support") Cc: stable@vger.kernel.org Link: https://lore.kernel.org/netdev/Z2K%2FI4nlHdfMRTZC@v4bel-B760M-AORUS-ELITE-AX/ Link: https://lore.kernel.org/netdev/5ca20d4c-1017-49c2-9516-f6f75fd331e9@rbox.co/ Link: https://lore.kernel.org/netdev/677f84a8.050a0220.25a300.01b3.GAE@google.com/ Co-developed-by:
Hyunwoo Kim <v4bel@theori.io> Signed-off-by:
Wongi Lee <qwerty@theori.io> Signed-off-by:
Stefano Garzarella <sgarzare@redhat.com> Reviewed-by:
Luigi Leonardi <leonardi@redhat.com> Reviewed-by:
Paolo Abeni <pabeni@redhat.com> Signed-off-by:
-
Stefano Garzarella authored
commit a24009bc upstream. Transport's release() and destruct() are called when de-assigning the vsock transport. These callbacks can touch some socket state like sock flags, sk_state, and peer_shutdown. Since we are reassigning the socket to a new transport during vsock_connect(), let's reset these fields to have a clean state with the new transport. Fixes: c0cfa2d8 ("vsock: add multi-transports support") Cc: stable@vger.kernel.org Signed-off-by:
-
Stefano Garzarella authored
commit df137da9 upstream. During virtio_transport_release() we can schedule a delayed work to perform the closing of the socket before destruction. The destructor is called either when the socket is really destroyed (reference counter to zero), or it can also be called when we are de-assigning the transport. In the former case, we are sure the delayed work has completed, because it holds a reference until it completes, so the destructor will definitely be called after the delayed work is finished. But in the latter case, the destructor is called by AF_VSOCK core, just after the release(), so there may still be delayed work scheduled. Refactor the code, moving the code to delete the close work already in the do_close() to a new function. Invoke it during destruction to make sure we don't leave any pending work. Fixes: c0cfa2d8 ("vsock: add multi-transports support") Cc: stable@vger.kernel.org Reported-by:
-
Stefano Garzarella authored
commit 2cb7c756 upstream. If the socket has been de-assigned or assigned to another transport, we must discard any packets received because they are not expected and would cause issues when we access vsk->transport. A possible scenario is described by Hyunwoo Kim in the attached link, where after a first connect() interrupted by a signal, and a second connect() failed, we can find `vsk->transport` at NULL, leading to a NULL pointer dereference. Fixes: c0cfa2d8 ("vsock: add multi-transports support") Cc: stable@vger.kernel.org Reported-by:
-
Stefano Garzarella authored
commit f6abafcd upstream. Some of the core functions can only be called if the transport has been assigned. As Michal reported, a socket might have the transport at NULL, for example after a failed connect(), causing the following trace: BUG: kernel NULL pointer dereference, address: 00000000000000a0 #PF: supervisor read access in kernel mode #PF: error_code(0x0000) - not-present page PGD 12faf8067 P4D 12faf8067 PUD 113670067 PMD 0 Oops: Oops: 0000 [#1] PREEMPT SMP NOPTI CPU: 15 UID: 0 PID: 1198 Comm: a.out Not tainted 6.13.0-rc2+ RIP: 0010:vsock_connectible_has_data+0x1f/0x40 Call Trace: vsock_bpf_recvmsg+0xca/0x5e0 sock_recvmsg+0xb9/0xc0 __sys_recvfrom+0xb3/0x130 __x64_sys_recvfrom+0x20/0x30 do_syscall_64+0x93/0x180 entry_SYSCALL_64_after_hwframe+0x76/0x7e So we need to check the `vsk->transport` in vsock_bpf_recvmsg(), especially for connected sockets (stream/seqpacket) as we already do in __vsock_connectible_recvmsg(). Fixes: 634f1a71 ("vsock: support sockmap") Cc: stable@vger.kernel.org Reported-by:
Michal Luczaj <mhal@rbox.co> Closes: https://lore.kernel.org/netdev/5ca20d4c-1017-49c2-9516-f6f75fd331e9@rbox.co/ Tested-by:
<syzbot+3affdbfc986ecd9200fd@syzkaller.appspotmail.com> Closes: https://lore.kernel.org/netdev/677f84a8.050a0220.25a300.01b3.GAE@google.com/ Tested-by:
Michael S. Tsirkin <mst@redhat.com> Reviewed-by:
-
