linux/kernel/events/core.c, branch v6.17

perf: Fix the POLL_HUP delivery breakage

2025-09-03T08:10:59Z

The event_limit can be set by the PERF_EVENT_IOC_REFRESH to limit the number of events. When the event_limit reaches 0, the POLL_HUP signal should be sent. But it's missed. The corresponding counter should be stopped when the event_limit reaches 0. It was implemented in the ARCH-specific code. However, since the commit 9734e25fbf5a ("perf: Fix the throttle logic for a group"), all the ARCH-specific code has been moved to the generic code. The code to handle the event_limit was lost. Add the event->pmu->stop(event, 0); back. Fixes: 9734e25fbf5a ("perf: Fix the throttle logic for a group") Closes: https://lore.kernel.org/lkml/aICYAqM5EQUlTqtX@li-2b55cdcc-350b-11b2-a85c-a78bff51fc11.ibm.com/ Reported-by: Sumanth Korikkar Signed-off-by: Kan Liang Signed-off-by: Peter Zijlstra (Intel) Tested-by: Sumanth Korikkar Link: https://lkml.kernel.org/r/20250811182644.1305952-1-kan.liang@linux.intel.com

perf: Avoid undefined behavior from stopping/starting inactive events

2025-08-15T11:12:56Z

Calling pmu->start()/stop() on perf events in PERF_EVENT_STATE_OFF can leave event->hw.idx at -1. When PMU drivers later attempt to use this negative index as a shift exponent in bitwise operations, it leads to UBSAN shift-out-of-bounds reports. The issue is a logical flaw in how event groups handle throttling when some members are intentionally disabled. Based on the analysis and the reproducer provided by Mark Rutland (this issue on both arm64 and x86-64). The scenario unfolds as follows: 1. A group leader event is configured with a very aggressive sampling period (e.g., sample_period = 1). This causes frequent interrupts and triggers the throttling mechanism. 2. A child event in the same group is created in a disabled state (.disabled = 1). This event remains in PERF_EVENT_STATE_OFF. Since it hasn't been scheduled onto the PMU, its event->hw.idx remains initialized at -1. 3. When throttling occurs, perf_event_throttle_group() and later perf_event_unthrottle_group() iterate through all siblings, including the disabled child event. 4. perf_event_throttle()/unthrottle() are called on this inactive child event, which then call event->pmu->start()/stop(). 5. The PMU driver receives the event with hw.idx == -1 and attempts to use it as a shift exponent. e.g., in macros like PMCNTENSET(idx), leading to the UBSAN report. The throttling mechanism attempts to start/stop events that are not actively scheduled on the hardware. Move the state check into perf_event_throttle()/perf_event_unthrottle() so that inactive events are skipped entirely. This ensures only active events with a valid hw.idx are processed, preventing undefined behavior and silencing UBSAN warnings. The corrected check ensures true before proceeding with PMU operations. The problem can be reproduced with the syzkaller reproducer: Fixes: 9734e25fbf5a ("perf: Fix the throttle logic for a group") Signed-off-by: Yunseong Kim Signed-off-by: Peter Zijlstra (Intel) Reviewed-by: Kan Liang Link: https://lore.kernel.org/r/20250812181046.292382-2-ysk@kzalloc.com

perf/core: Prevent VMA split of buffer mappings

2025-08-05T19:55:29Z

The perf mmap code is careful about mmap()'ing the user page with the ringbuffer and additionally the auxiliary buffer, when the event supports it. Once the first mapping is established, subsequent mapping have to use the same offset and the same size in both cases. The reference counting for the ringbuffer and the auxiliary buffer depends on this being correct. Though perf does not prevent that a related mapping is split via mmap(2), munmap(2) or mremap(2). A split of a VMA results in perf_mmap_open() calls, which take reference counts, but then the subsequent perf_mmap_close() calls are not longer fulfilling the offset and size checks. This leads to reference count leaks. As perf already has the requirement for subsequent mappings to match the initial mapping, the obvious consequence is that VMA splits, caused by resizing of a mapping or partial unmapping, have to be prevented. Implement the vm_operations_struct::may_split() callback and return unconditionally -EINVAL. That ensures that the mapping offsets and sizes cannot be changed after the fact. Remapping to a different fixed address with the same size is still possible as it takes the references for the new mapping and drops those of the old mapping. Fixes: 45bfb2e50471 ("perf: Add AUX area to ring buffer for raw data streams") Reported-by: zdi-disclosures@trendmicro.com # ZDI-CAN-27504 Signed-off-by: Thomas Gleixner Reviewed-by: Lorenzo Stoakes Acked-by: Arnaldo Carvalho de Melo Acked-by: Vlastimil Babka Cc: stable@vger.kernel.org

perf/core: Handle buffer mapping fail correctly in perf_mmap()

2025-08-05T19:55:29Z

After successful allocation of a buffer or a successful attachment to an existing buffer perf_mmap() tries to map the buffer read only into the page table. If that fails, the already set up page table entries are zapped, but the other perf specific side effects of that failure are not handled. The calling code just cleans up the VMA and does not invoke perf_mmap_close(). This leaks reference counts, corrupts user->vm accounting and also results in an unbalanced invocation of event::event_mapped(). Cure this by moving the event::event_mapped() invocation before the map_range() call so that on map_range() failure perf_mmap_close() can be invoked without causing an unbalanced event::event_unmapped() call. perf_mmap_close() undoes the reference counts and eventually frees buffers. Fixes: b709eb872e19 ("perf: map pages in advance") Signed-off-by: Thomas Gleixner Reviewed-by: Lorenzo Stoakes Cc: stable@vger.kernel.org

perf/core: Exit early on perf_mmap() fail

2025-08-05T19:55:29Z

When perf_mmap() fails to allocate a buffer, it still invokes the event_mapped() callback of the related event. On X86 this might increase the perf_rdpmc_allowed reference counter. But nothing undoes this as perf_mmap_close() is never called in this case, which causes another reference count leak. Return early on failure to prevent that. Fixes: 1e0fb9ec679c ("perf: Add pmu callbacks to track event mapping and unmapping") Signed-off-by: Thomas Gleixner Reviewed-by: Lorenzo Stoakes Cc: stable@vger.kernel.org

perf/core: Don't leak AUX buffer refcount on allocation failure

2025-08-05T19:55:29Z

Failure of the AUX buffer allocation leaks the reference count. Set the reference count to 1 only when the allocation succeeds. Fixes: 45bfb2e50471 ("perf: Add AUX area to ring buffer for raw data streams") Signed-off-by: Thomas Gleixner Reviewed-by: Lorenzo Stoakes Cc: stable@vger.kernel.org

perf/core: Preserve AUX buffer allocation failure result

2025-08-05T19:55:28Z

A recent overhaul sets the return value to 0 unconditionally after the allocations, which causes reference count leaks and corrupts the user->vm accounting. Preserve the AUX buffer allocation failure return value, so that the subsequent code works correctly. Fixes: 0983593f32c4 ("perf/core: Lift event->mmap_mutex in perf_mmap()") Signed-off-by: Thomas Gleixner Reviewed-by: Lorenzo Stoakes Cc: stable@vger.kernel.org

perf/core: Fix WARN in perf_sigtrap()

2025-07-09T11:40:17Z

Since exit_task_work() runs after perf_event_exit_task_context() updated ctx->task to TASK_TOMBSTONE, perf_sigtrap() from perf_pending_task() might observe event->ctx->task == TASK_TOMBSTONE. Swap the early exit tests in order not to hit WARN_ON_ONCE(). Closes: https://syzkaller.appspot.com/bug?extid=2fe61cb2a86066be6985 Reported-by: syzbot Signed-off-by: Tetsuo Handa Signed-off-by: Peter Zijlstra (Intel) Link: https://lkml.kernel.org/r/b1c224bd-97f9-462c-a3e3-125d5e19c983@I-love.SAKURA.ne.jp

perf: Revert to requiring CAP_SYS_ADMIN for uprobes

2025-07-03T08:33:55Z

Jann reports that uprobes can be used destructively when used in the middle of an instruction. The kernel only verifies there is a valid instruction at the requested offset, but due to variable instruction length cannot determine if this is an instruction as seen by the intended execution stream. Additionally, Mark Rutland notes that on architectures that mix data in the text segment (like arm64), a similar things can be done if the data word is 'mistaken' for an instruction. As such, require CAP_SYS_ADMIN for uprobes. Fixes: c9e0924e5c2b ("perf/core: open access to probes for CAP_PERFMON privileged process") Reported-by: Jann Horn Signed-off-by: Peter Zijlstra (Intel) Link: https://lkml.kernel.org/r/CAG48ez1n4520sq0XrWYDHKiKxE_+WCfAK+qt9qkY4ZiBGmL-5g@mail.gmail.com

perf/core: Fix the WARN_ON_ONCE is out of lock protected region

2025-06-30T07:32:49Z

commit 3172fb986666 ("perf/core: Fix WARN in perf_cgroup_switch()") try to fix a concurrency problem between perf_cgroup_switch and perf_cgroup_event_disable. But it does not to move the WARN_ON_ONCE into lock-protected region, so the warning is still be triggered. Fixes: 3172fb986666 ("perf/core: Fix WARN in perf_cgroup_switch()") Signed-off-by: Luo Gengkun Signed-off-by: Peter Zijlstra (Intel) Link: https://lkml.kernel.org/r/20250626135403.2454105-1-luogengkun@huaweicloud.com