linux/kernel/events/core.c, branch v6.14

perf/core: Fix perf_pmu_register() vs. perf_init_event()

2025-03-01T18:38:42Z

There is a fairly obvious race between perf_init_event() doing idr_find() and perf_pmu_register() doing idr_alloc() with an incompletely initialized PMU pointer. Avoid by doing idr_alloc() on a NULL pointer to register the id, and swizzling the real struct pmu pointer at the end using idr_replace(). Also making sure to not set struct pmu members after publishing the struct pmu, duh. [ introduce idr_cmpxchg() in order to better handle the idr_replace() error case -- if it were to return an unexpected pointer, it will already have replaced the value and there is no going back. ] Signed-off-by: Peter Zijlstra (Intel) Signed-off-by: Ingo Molnar Link: https://lore.kernel.org/r/20241104135517.858805880@infradead.org

perf/core: Fix pmus_lock vs. pmus_srcu ordering

2025-03-01T18:38:42Z

Commit a63fbed776c7 ("perf/tracing/cpuhotplug: Fix locking order") placed pmus_lock inside pmus_srcu, this makes perf_pmu_unregister() trip lockdep. Move the locking about such that only pmu_idr and pmus (list) are modified while holding pmus_lock. This avoids doing synchronize_srcu() while holding pmus_lock and all is well again. Fixes: a63fbed776c7 ("perf/tracing/cpuhotplug: Fix locking order") Signed-off-by: Peter Zijlstra (Intel) Signed-off-by: Ingo Molnar Link: https://lore.kernel.org/r/20241104135517.679556858@infradead.org

perf/core: Fix low freq setting via IOC_PERIOD

2025-02-25T13:54:14Z

A low attr::freq value cannot be set via IOC_PERIOD on some platforms. The perf_event_check_period() introduced in: 81ec3f3c4c4d ("perf/x86: Add check_period PMU callback") was intended to check the period, rather than the frequency. A low frequency may be mistakenly rejected by limit_period(). Fix it. Fixes: 81ec3f3c4c4d ("perf/x86: Add check_period PMU callback") Signed-off-by: Kan Liang Signed-off-by: Ingo Molnar Reviewed-by: Ravi Bangoria Cc: Peter Zijlstra Cc: stable@vger.kernel.org Link: https://lore.kernel.org/r/20250117151913.3043942-2-kan.liang@linux.intel.com Closes: https://lore.kernel.org/lkml/20250115154949.3147-1-ravi.bangoria@amd.com/

perf/core: Order the PMU list to fix warning about unordered pmu_ctx_list

2025-02-24T18:22:37Z

Syskaller triggers a warning due to prev_epc->pmu != next_epc->pmu in perf_event_swap_task_ctx_data(). vmcore shows that two lists have the same perf_event_pmu_context, but not in the same order. The problem is that the order of pmu_ctx_list for the parent is impacted by the time when an event/PMU is added. While the order for a child is impacted by the event order in the pinned_groups and flexible_groups. So the order of pmu_ctx_list in the parent and child may be different. To fix this problem, insert the perf_event_pmu_context to its proper place after iteration of the pmu_ctx_list. The follow testcase can trigger above warning: # perf record -e cycles --call-graph lbr -- taskset -c 3 ./a.out & # perf stat -e cpu-clock,cs -p xxx // xxx is the pid of a.out test.c void main() { int count = 0; pid_t pid; printf("%d running\n", getpid()); sleep(30); printf("running\n"); pid = fork(); if (pid == -1) { printf("fork error\n"); return; } if (pid == 0) { while (1) { count++; } } else { while (1) { count++; } } } The testcase first opens an LBR event, so it will allocate task_ctx_data, and then open tracepoint and software events, so the parent context will have 3 different perf_event_pmu_contexts. On inheritance, child ctx will insert the perf_event_pmu_context in another order and the warning will trigger. [ mingo: Tidied up the changelog. ] Fixes: bd2756811766 ("perf: Rewrite core context handling") Signed-off-by: Luo Gengkun Signed-off-by: Ingo Molnar Reviewed-by: Kan Liang Link: https://lore.kernel.org/r/20250122073356.1824736-1-luogengkun@huaweicloud.com

perf/core: Add RCU read lock protection to perf_iterate_ctx()

2025-02-24T18:17:04Z

The perf_iterate_ctx() function performs RCU list traversal but currently lacks RCU read lock protection. This causes lockdep warnings when running perf probe with unshare(1) under CONFIG_PROVE_RCU_LIST=y: WARNING: suspicious RCU usage kernel/events/core.c:8168 RCU-list traversed in non-reader section!! Call Trace: lockdep_rcu_suspicious ? perf_event_addr_filters_apply perf_iterate_ctx perf_event_exec begin_new_exec ? load_elf_phdrs load_elf_binary ? lock_acquire ? find_held_lock ? bprm_execve bprm_execve do_execveat_common.isra.0 __x64_sys_execve do_syscall_64 entry_SYSCALL_64_after_hwframe This protection was previously present but was removed in commit bd2756811766 ("perf: Rewrite core context handling"). Add back the necessary rcu_read_lock()/rcu_read_unlock() pair around perf_iterate_ctx() call in perf_event_exec(). [ mingo: Use scoped_guard() as suggested by Peter ] Fixes: bd2756811766 ("perf: Rewrite core context handling") Signed-off-by: Breno Leitao Signed-off-by: Ingo Molnar Acked-by: Peter Zijlstra (Intel) Cc: stable@vger.kernel.org Link: https://lore.kernel.org/r/20250117-fix_perf_rcu-v1-1-13cb9210fc6a@debian.org

perf: map pages in advance

2025-01-10T17:16:50Z

We are adjusting struct page to make it smaller, removing unneeded fields which correctly belong to struct folio. Two of those fields are page->index and page->mapping. Perf is currently making use of both of these. This is unnecessary. This patch eliminates this. Perf establishes its own internally controlled memory-mapped pages using vm_ops hooks. The first page in the mapping is the read/write user control page, and the rest of the mapping consists of read-only pages. The VMA is backed by kernel memory either from the buddy allocator or vmalloc depending on configuration. It is intended to be mapped read/write, but because it has a page_mkwrite() hook, vma_wants_writenotify() indicates that it should be mapped read-only. When a write fault occurs, the provided page_mkwrite() hook, perf_mmap_fault() (doing double duty handing faults as well) uses the vmf->pgoff field to determine if this is the first page, allowing for the desired read/write first page, read-only rest mapping. For this to work the implementation has to carefully work around faulting logic. When a page is write-faulted, the fault() hook is called first, then its page_mkwrite() hook is called (to allow for dirty tracking in file systems). On fault we set the folio's mapping in perf_mmap_fault(), this is because when do_page_mkwrite() is subsequently invoked, it treats a missing mapping as an indicator that the fault should be retried. We also set the folio's index so, given the folio is being treated as faux user memory, it correctly references its offset within the VMA. This explains why the mapping and index fields are used - but it's not necessary. We preallocate pages when perf_mmap() is called for the first time via rb_alloc(), and further allocate auxiliary pages via rb_aux_alloc() as needed if the mapping requires it. This allocation is done in the f_ops->mmap() hook provided in perf_mmap(), and so we can instead simply map all the memory right away here - there's no point in handling (read) page faults when we don't demand page nor need to be notified about them (perf does not). This patch therefore changes this logic to map everything when the mmap() hook is called, establishing a PFN map. It implements vm_ops->pfn_mkwrite() to provide the required read/write vs. read-only behaviour, which does not require the previously implemented workarounds. While it is not ideal to use a VM_PFNMAP here, doing anything else will result in the page_mkwrite() hook need to be provided, which requires the same page->mapping hack this patch seeks to undo. It will also result in the pages being treated as folios and placed on the rmap, which really does not make sense for these mappings. Semantically it makes sense to establish this as some kind of special mapping, as the pages are managed by perf and are not strictly user pages, but currently the only means by which we can do so functionally while maintaining the required R/W and R/O behaviour is a PFN map. There should be no change to actual functionality as a result of this change. Signed-off-by: Lorenzo Stoakes Signed-off-by: Peter Zijlstra (Intel) Link: https://lkml.kernel.org/r/20250103153151.124163-1-lorenzo.stoakes@oracle.com

perf/core: Export perf_exclude_event()

2024-12-09T14:50:31Z

While at it, rename the same function in s390 cpum_sf PMU. Signed-off-by: Namhyung Kim Signed-off-by: Ingo Molnar Tested-by: Ravi Bangoria Reviewed-by: Ravi Bangoria Acked-by: Thomas Richter Link: https://lore.kernel.org/r/20241203180441.1634709-2-namhyung@kernel.org

Merge tag 'v6.13-rc1' into perf/core, to refresh the branch

2024-12-02T10:52:59Z

Signed-off-by: Ingo Molnar

Merge tag 'mm-nonmm-stable-2024-11-24-02-05' of git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm

2024-11-26T00:09:48Z

Pull non-MM updates from Andrew Morton: - The series "resource: A couple of cleanups" from Andy Shevchenko performs some cleanups in the resource management code - The series "Improve the copy of task comm" from Yafang Shao addresses possible race-induced overflows in the management of task_struct.comm[] - The series "Remove unnecessary header includes from {tools/}lib/list_sort.c" from Kuan-Wei Chiu adds some cleanups and a small fix to the list_sort library code and to its selftest - The series "Enhance min heap API with non-inline functions and optimizations" also from Kuan-Wei Chiu optimizes and cleans up the min_heap library code - The series "nilfs2: Finish folio conversion" from Ryusuke Konishi finishes off nilfs2's folioification - The series "add detect count for hung tasks" from Lance Yang adds more userspace visibility into the hung-task detector's activity - Apart from that, singelton patches in many places - please see the individual changelogs for details * tag 'mm-nonmm-stable-2024-11-24-02-05' of git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm: (71 commits) gdb: lx-symbols: do not error out on monolithic build kernel/reboot: replace sprintf() with sysfs_emit() lib: util_macros_kunit: add kunit test for util_macros.h util_macros.h: fix/rework find_closest() macros Improve consistency of '#error' directive messages ocfs2: fix uninitialized value in ocfs2_file_read_iter() hung_task: add docs for hung_task_detect_count hung_task: add detect count for hung tasks dma-buf: use atomic64_inc_return() in dma_buf_getfile() fs/proc/kcore.c: fix coccinelle reported ERROR instances resource: avoid unnecessary resource tree walking in __region_intersects() ocfs2: remove unused errmsg function and table ocfs2: cluster: fix a typo lib/scatterlist: use sg_phys() helper checkpatch: always parse orig_commit in fixes tag nilfs2: convert metadata aops from writepage to writepages nilfs2: convert nilfs_recovery_copy_block() to take a folio nilfs2: convert nilfs_page_count_clean_buffers() to take a folio nilfs2: remove nilfs_writepage nilfs2: convert checkpoint file to be folio-based ...

Merge tag 'perf-core-2024-11-18' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip

2024-11-19T21:34:06Z

Pull performance events updates from Ingo Molnar: "Uprobes: - Add BPF session support (Jiri Olsa) - Switch to RCU Tasks Trace flavor for better performance (Andrii Nakryiko) - Massively increase uretprobe SMP scalability by SRCU-protecting the uretprobe lifetime (Andrii Nakryiko) - Kill xol_area->slot_count (Oleg Nesterov) Core facilities: - Implement targeted high-frequency profiling by adding the ability for an event to "pause" or "resume" AUX area tracing (Adrian Hunter) VM profiling/sampling: - Correct perf sampling with guest VMs (Colton Lewis) New hardware support: - x86/intel: Add PMU support for Intel ArrowLake-H CPUs (Dapeng Mi) Misc fixes and enhancements: - x86/intel/pt: Fix buffer full but size is 0 case (Adrian Hunter) - x86/amd: Warn only on new bits set (Breno Leitao) - x86/amd/uncore: Avoid a false positive warning about snprintf truncation in amd_uncore_umc_ctx_init (Jean Delvare) - uprobes: Re-order struct uprobe_task to save some space (Christophe JAILLET) - x86/rapl: Move the pmu allocation out of CPU hotplug (Kan Liang) - x86/rapl: Clean up cpumask and hotplug (Kan Liang) - uprobes: Deuglify xol_get_insn_slot/xol_free_insn_slot paths (Oleg Nesterov)" * tag 'perf-core-2024-11-18' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (32 commits) perf/core: Correct perf sampling with guest VMs perf/x86: Refactor misc flag assignments perf/powerpc: Use perf_arch_instruction_pointer() perf/core: Hoist perf_instruction_pointer() and perf_misc_flags() perf/arm: Drop unused functions uprobes: Re-order struct uprobe_task to save some space perf/x86/amd/uncore: Avoid a false positive warning about snprintf truncation in amd_uncore_umc_ctx_init perf/x86/intel: Do not enable large PEBS for events with aux actions or aux sampling perf/x86/intel/pt: Add support for pause / resume perf/core: Add aux_pause, aux_resume, aux_start_paused perf/x86/intel/pt: Fix buffer full but size is 0 case uprobes: SRCU-protect uretprobe lifetime (with timeout) uprobes: allow put_uprobe() from non-sleepable softirq context perf/x86/rapl: Clean up cpumask and hotplug perf/x86/rapl: Move the pmu allocation out of CPU hotplug uprobe: Add support for session consumer uprobe: Add data pointer to consumer handlers perf/x86/amd: Warn only on new bits set uprobes: fold xol_take_insn_slot() into xol_get_insn_slot() uprobes: kill xol_area->slot_count ...