From 8d7861ac507d23024c7d74b6cb59a9cca248bcb7 Mon Sep 17 00:00:00 2001 From: Nam Cao Date: Fri, 11 Apr 2025 09:37:17 +0200 Subject: [PATCH 01/16] rv: Fix out-of-bound memory access in rv_is_container_monitor() When rv_is_container_monitor() is called on the last monitor in rv_monitors_list, KASAN yells: BUG: KASAN: global-out-of-bounds in rv_is_container_monitor+0x101/0x110 Read of size 8 at addr ffffffff97c7c798 by task setup/221 The buggy address belongs to the variable: rv_monitors_list+0x18/0x40 This is due to list_next_entry() is called on the last entry in the list. It wraps around to the first list_head, and the first list_head is not embedded in struct rv_monitor_def. Fix it by checking if the monitor is last in the list. Cc: stable@vger.kernel.org Cc: Gabriele Monaco Fixes: cb85c660fcd4 ("rv: Add option for nested monitors and include sched") Link: https://lore.kernel.org/e85b5eeb7228bfc23b8d7d4ab5411472c54ae91b.1744355018.git.namcao@linutronix.de Signed-off-by: Nam Cao Signed-off-by: Steven Rostedt (Google) --- kernel/trace/rv/rv.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) diff --git a/kernel/trace/rv/rv.c b/kernel/trace/rv/rv.c index 968c5c3b0246..e4077500a91d 100644 --- a/kernel/trace/rv/rv.c +++ b/kernel/trace/rv/rv.c @@ -225,7 +225,12 @@ bool rv_is_nested_monitor(struct rv_monitor_def *mdef) */ bool rv_is_container_monitor(struct rv_monitor_def *mdef) { - struct rv_monitor_def *next = list_next_entry(mdef, list); + struct rv_monitor_def *next; + + if (list_is_last(&mdef->list, &rv_monitors_list)) + return false; + + next = list_next_entry(mdef, list); return next->parent == mdef->monitor || !mdef->monitor->enable; } -- 2.51.0 From ce7e8a65aa1b7e8a6833403b314fa8f2cf133119 Mon Sep 17 00:00:00 2001 From: Tom Vierjahn Date: Mon, 24 Mar 2025 23:09:30 +0100 Subject: [PATCH 02/16] Documentation: ext4: Add fields to ext4_super_block documentation Documentation and implementation of the ext4 super block have slightly diverged: Padding has been removed in order to make room for new fields that are still missing in the documentation. Add the new fields s_encryption_level, s_first_error_errorcode, s_last_error_errorcode to the documentation of the ext4 super block. Fixes: f542fbe8d5e8 ("ext4 crypto: reserve codepoints used by the ext4 encryption feature") Fixes: 878520ac45f9 ("ext4: save the error code which triggered an ext4_error() in the superblock") Signed-off-by: Tom Vierjahn Reviewed-by: Ojaswin Mujoo Link: https://patch.msgid.link/20250324221004.5268-1-tom.vierjahn@acm.org Signed-off-by: Theodore Ts'o --- Documentation/filesystems/ext4/super.rst | 20 ++++++++++++++------ 1 file changed, 14 insertions(+), 6 deletions(-) diff --git a/Documentation/filesystems/ext4/super.rst b/Documentation/filesystems/ext4/super.rst index a1eb4a11a1d0..1b240661bfa3 100644 --- a/Documentation/filesystems/ext4/super.rst +++ b/Documentation/filesystems/ext4/super.rst @@ -328,9 +328,13 @@ The ext4 superblock is laid out as follows in - s_checksum_type - Metadata checksum algorithm type. The only valid value is 1 (crc32c). * - 0x176 - - __le16 - - s_reserved_pad - - + - \_\_u8 + - s\_encryption\_level + - Versioning level for encryption. + * - 0x177 + - \_\_u8 + - s\_reserved\_pad + - Padding to next 32bits. * - 0x178 - __le64 - s_kbytes_written @@ -466,9 +470,13 @@ The ext4 superblock is laid out as follows in - s_last_error_time_hi - Upper 8 bits of the s_last_error_time field. * - 0x27A - - __u8 - - s_pad[2] - - Zero padding. + - \_\_u8 + - s\_first\_error\_errcode + - + * - 0x27B + - \_\_u8 + - s\_last\_error\_errcode + - * - 0x27C - __le16 - s_encoding -- 2.51.0 From 7e50bbb134aba1df0854f171b596b3a42d35605a Mon Sep 17 00:00:00 2001 From: "Gustavo A. R. Silva" Date: Wed, 26 Mar 2025 16:55:51 -0600 Subject: [PATCH 03/16] ext4: avoid -Wflex-array-member-not-at-end warning -Wflex-array-member-not-at-end was introduced in GCC-14, and we are getting ready to enable it, globally. Use the `DEFINE_RAW_FLEX()` helper for an on-stack definition of a flexible structure where the size of the flexible-array member is known at compile-time, and refactor the rest of the code, accordingly. So, with these changes, fix the following warning: fs/ext4/mballoc.c:3041:40: warning: structure containing a flexible array member is not at the end of another structure [-Wflex-array-member-not-at-end] Signed-off-by: Gustavo A. R. Silva Reviewed-by: Kees Cook Link: https://patch.msgid.link/Z-SF97N3AxcIMlSi@kspp Signed-off-by: Theodore Ts'o --- fs/ext4/mballoc.c | 18 ++++++++---------- 1 file changed, 8 insertions(+), 10 deletions(-) diff --git a/fs/ext4/mballoc.c b/fs/ext4/mballoc.c index 0d523e9fb3d5..f88424c28194 100644 --- a/fs/ext4/mballoc.c +++ b/fs/ext4/mballoc.c @@ -3037,10 +3037,8 @@ static int ext4_mb_seq_groups_show(struct seq_file *seq, void *v) unsigned char blocksize_bits = min_t(unsigned char, sb->s_blocksize_bits, EXT4_MAX_BLOCK_LOG_SIZE); - struct sg { - struct ext4_group_info info; - ext4_grpblk_t counters[EXT4_MAX_BLOCK_LOG_SIZE + 2]; - } sg; + DEFINE_RAW_FLEX(struct ext4_group_info, sg, bb_counters, + EXT4_MAX_BLOCK_LOG_SIZE + 2); group--; if (group == 0) @@ -3048,7 +3046,7 @@ static int ext4_mb_seq_groups_show(struct seq_file *seq, void *v) " 2^0 2^1 2^2 2^3 2^4 2^5 2^6 " " 2^7 2^8 2^9 2^10 2^11 2^12 2^13 ]\n"); - i = (blocksize_bits + 2) * sizeof(sg.info.bb_counters[0]) + + i = (blocksize_bits + 2) * sizeof(sg->bb_counters[0]) + sizeof(struct ext4_group_info); grinfo = ext4_get_group_info(sb, group); @@ -3068,14 +3066,14 @@ static int ext4_mb_seq_groups_show(struct seq_file *seq, void *v) * We care only about free space counters in the group info and * these are safe to access even after the buddy has been unloaded */ - memcpy(&sg, grinfo, i); - seq_printf(seq, "#%-5u: %-5u %-5u %-5u [", group, sg.info.bb_free, - sg.info.bb_fragments, sg.info.bb_first_free); + memcpy(sg, grinfo, i); + seq_printf(seq, "#%-5u: %-5u %-5u %-5u [", group, sg->bb_free, + sg->bb_fragments, sg->bb_first_free); for (i = 0; i <= 13; i++) seq_printf(seq, " %-5u", i <= blocksize_bits + 1 ? - sg.info.bb_counters[i] : 0); + sg->bb_counters[i] : 0); seq_puts(seq, " ]"); - if (EXT4_MB_GRP_BBITMAP_CORRUPT(&sg.info)) + if (EXT4_MB_GRP_BBITMAP_CORRUPT(sg)) seq_puts(seq, " Block bitmap corrupted!"); seq_putc(seq, '\n'); return 0; -- 2.51.0 From ccad447a3d331a239477c281533bacb585b54a98 Mon Sep 17 00:00:00 2001 From: Ojaswin Mujoo Date: Fri, 28 Mar 2025 11:54:52 +0530 Subject: [PATCH 04/16] ext4: make block validity check resistent to sb bh corruption Block validity checks need to be skipped in case they are called for journal blocks since they are part of system's protected zone. Currently, this is done by checking inode->ino against sbi->s_es->s_journal_inum, which is a direct read from the ext4 sb buffer head. If someone modifies this underneath us then the s_journal_inum field might get corrupted. To prevent against this, change the check to directly compare the inode with journal->j_inode. **Slight change in behavior**: During journal init path, check_block_validity etc might be called for journal inode when sbi->s_journal is not set yet. In this case we now proceed with ext4_inode_block_valid() instead of returning early. Since systems zones have not been set yet, it is okay to proceed so we can perform basic checks on the blocks. Suggested-by: Baokun Li Reviewed-by: Baokun Li Reviewed-by: Jan Kara Reviewed-by: Zhang Yi Signed-off-by: Ojaswin Mujoo Link: https://patch.msgid.link/0c06bc9ebfcd6ccfed84a36e79147bf45ff5adc1.1743142920.git.ojaswin@linux.ibm.com Signed-off-by: Theodore Ts'o --- fs/ext4/block_validity.c | 5 ++--- fs/ext4/inode.c | 7 ++++--- 2 files changed, 6 insertions(+), 6 deletions(-) diff --git a/fs/ext4/block_validity.c b/fs/ext4/block_validity.c index 87ee3a17bd29..e8c5525afc67 100644 --- a/fs/ext4/block_validity.c +++ b/fs/ext4/block_validity.c @@ -351,10 +351,9 @@ int ext4_check_blockref(const char *function, unsigned int line, { __le32 *bref = p; unsigned int blk; + journal_t *journal = EXT4_SB(inode->i_sb)->s_journal; - if (ext4_has_feature_journal(inode->i_sb) && - (inode->i_ino == - le32_to_cpu(EXT4_SB(inode->i_sb)->s_es->s_journal_inum))) + if (journal && inode == journal->j_inode) return 0; while (bref < p+max) { diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index 5c57a34cd82c..f386de8c12f6 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c @@ -384,10 +384,11 @@ static int __check_block_validity(struct inode *inode, const char *func, unsigned int line, struct ext4_map_blocks *map) { - if (ext4_has_feature_journal(inode->i_sb) && - (inode->i_ino == - le32_to_cpu(EXT4_SB(inode->i_sb)->s_es->s_journal_inum))) + journal_t *journal = EXT4_SB(inode->i_sb)->s_journal; + + if (journal && inode == journal->j_inode) return 0; + if (!ext4_inode_block_valid(inode, map->m_pblk, map->m_len)) { ext4_error_inode(inode, func, line, map->m_pblk, "lblock %lu mapped to illegal pblock %llu " -- 2.51.0 From 94824ac9a8aaf2fb3c54b4bdde842db80ffa555d Mon Sep 17 00:00:00 2001 From: Artem Sadovnikov Date: Fri, 4 Apr 2025 08:28:05 +0000 Subject: [PATCH 05/16] ext4: fix off-by-one error in do_split Syzkaller detected a use-after-free issue in ext4_insert_dentry that was caused by out-of-bounds access due to incorrect splitting in do_split. BUG: KASAN: use-after-free in ext4_insert_dentry+0x36a/0x6d0 fs/ext4/namei.c:2109 Write of size 251 at addr ffff888074572f14 by task syz-executor335/5847 CPU: 0 UID: 0 PID: 5847 Comm: syz-executor335 Not tainted 6.12.0-rc6-syzkaller-00318-ga9cda7c0ffed #0 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 10/30/2024 Call Trace: __dump_stack lib/dump_stack.c:94 [inline] dump_stack_lvl+0x241/0x360 lib/dump_stack.c:120 print_address_description mm/kasan/report.c:377 [inline] print_report+0x169/0x550 mm/kasan/report.c:488 kasan_report+0x143/0x180 mm/kasan/report.c:601 kasan_check_range+0x282/0x290 mm/kasan/generic.c:189 __asan_memcpy+0x40/0x70 mm/kasan/shadow.c:106 ext4_insert_dentry+0x36a/0x6d0 fs/ext4/namei.c:2109 add_dirent_to_buf+0x3d9/0x750 fs/ext4/namei.c:2154 make_indexed_dir+0xf98/0x1600 fs/ext4/namei.c:2351 ext4_add_entry+0x222a/0x25d0 fs/ext4/namei.c:2455 ext4_add_nondir+0x8d/0x290 fs/ext4/namei.c:2796 ext4_symlink+0x920/0xb50 fs/ext4/namei.c:3431 vfs_symlink+0x137/0x2e0 fs/namei.c:4615 do_symlinkat+0x222/0x3a0 fs/namei.c:4641 __do_sys_symlink fs/namei.c:4662 [inline] __se_sys_symlink fs/namei.c:4660 [inline] __x64_sys_symlink+0x7a/0x90 fs/namei.c:4660 do_syscall_x64 arch/x86/entry/common.c:52 [inline] do_syscall_64+0xf3/0x230 arch/x86/entry/common.c:83 entry_SYSCALL_64_after_hwframe+0x77/0x7f The following loop is located right above 'if' statement. for (i = count-1; i >= 0; i--) { /* is more than half of this entry in 2nd half of the block? */ if (size + map[i].size/2 > blocksize/2) break; size += map[i].size; move++; } 'i' in this case could go down to -1, in which case sum of active entries wouldn't exceed half the block size, but previous behaviour would also do split in half if sum would exceed at the very last block, which in case of having too many long name files in a single block could lead to out-of-bounds access and following use-after-free. Found by Linux Verification Center (linuxtesting.org) with Syzkaller. Cc: stable@vger.kernel.org Fixes: 5872331b3d91 ("ext4: fix potential negative array index in do_split()") Signed-off-by: Artem Sadovnikov Reviewed-by: Jan Kara Link: https://patch.msgid.link/20250404082804.2567-3-a.sadovnikov@ispras.ru Signed-off-by: Theodore Ts'o --- fs/ext4/namei.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/fs/ext4/namei.c b/fs/ext4/namei.c index 43f96ef4aa01..dda1791e9e1a 100644 --- a/fs/ext4/namei.c +++ b/fs/ext4/namei.c @@ -1971,7 +1971,7 @@ static struct ext4_dir_entry_2 *do_split(handle_t *handle, struct inode *dir, * split it in half by count; each resulting block will have at least * half the space free. */ - if (i > 0) + if (i >= 0) split = count - move; else split = count/2; -- 2.51.0 From 8ffd015db85fea3e15a77027fda6c02ced4d2444 Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Sun, 13 Apr 2025 11:54:49 -0700 Subject: [PATCH 06/16] Linux 6.15-rc2 --- Makefile | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/Makefile b/Makefile index f42418556507..c91afd55099e 100644 --- a/Makefile +++ b/Makefile @@ -2,7 +2,7 @@ VERSION = 6 PATCHLEVEL = 15 SUBLEVEL = 0 -EXTRAVERSION = -rc1 +EXTRAVERSION = -rc2 NAME = Baby Opossum Posse # *DOCUMENTATION* -- 2.51.0 From 5c3627b6f0595f1ec27e6f5df903bd072e9b9136 Mon Sep 17 00:00:00 2001 From: Thorsten Blum Date: Sun, 13 Apr 2025 12:41:09 +0200 Subject: [PATCH 07/16] perf/x86/intel/bts: Replace offsetof() with struct_size() Use struct_size() to calculate the number of bytes to allocate for a new bts_buffer. Compared to offsetof(), struct_size() provides additional compile-time checks (e.g., __must_be_array()). Signed-off-by: Thorsten Blum Signed-off-by: Ingo Molnar Cc: Peter Zijlstra Link: https://lore.kernel.org/r/20250413104108.49142-2-thorsten.blum@linux.dev --- arch/x86/events/intel/bts.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/x86/events/intel/bts.c b/arch/x86/events/intel/bts.c index da03f53bfa18..16bc89c8023b 100644 --- a/arch/x86/events/intel/bts.c +++ b/arch/x86/events/intel/bts.c @@ -101,7 +101,7 @@ bts_buffer_setup_aux(struct perf_event *event, void **pages, if (overwrite && nr_buf > 1) return NULL; - bb = kzalloc_node(offsetof(struct bts_buffer, buf[nr_buf]), GFP_KERNEL, node); + bb = kzalloc_node(struct_size(bb, buf, nr_buf), GFP_KERNEL, node); if (!bb) return NULL; -- 2.51.0 From 96a720db59ab330c8562b2437153faa45dac705f Mon Sep 17 00:00:00 2001 From: Kan Liang Date: Wed, 16 Apr 2025 07:24:24 -0700 Subject: [PATCH 08/16] perf/x86/intel/uncore: Fix the scale of IIO free running counters on SNR There was a mistake in the SNR uncore spec. The counter increments for every 32 bytes of data sent from the IO agent to the SOC, not 4 bytes which was documented in the spec. The event list has been updated: "EventName": "UNC_IIO_BANDWIDTH_IN.PART0_FREERUN", "BriefDescription": "Free running counter that increments for every 32 bytes of data sent from the IO agent to the SOC", Update the scale of the IIO bandwidth in free running counters as well. Fixes: 210cc5f9db7a ("perf/x86/intel/uncore: Add uncore support for Snow Ridge server") Signed-off-by: Kan Liang Signed-off-by: Ingo Molnar Acked-by: Peter Zijlstra Cc: stable@vger.kernel.org Link: https://lore.kernel.org/r/20250416142426.3933977-1-kan.liang@linux.intel.com --- arch/x86/events/intel/uncore_snbep.c | 16 ++++++++-------- 1 file changed, 8 insertions(+), 8 deletions(-) diff --git a/arch/x86/events/intel/uncore_snbep.c b/arch/x86/events/intel/uncore_snbep.c index 60973c209c0e..35da2c486e8d 100644 --- a/arch/x86/events/intel/uncore_snbep.c +++ b/arch/x86/events/intel/uncore_snbep.c @@ -4891,28 +4891,28 @@ static struct uncore_event_desc snr_uncore_iio_freerunning_events[] = { INTEL_UNCORE_EVENT_DESC(ioclk, "event=0xff,umask=0x10"), /* Free-Running IIO BANDWIDTH IN Counters */ INTEL_UNCORE_EVENT_DESC(bw_in_port0, "event=0xff,umask=0x20"), - INTEL_UNCORE_EVENT_DESC(bw_in_port0.scale, "3.814697266e-6"), + INTEL_UNCORE_EVENT_DESC(bw_in_port0.scale, "3.0517578125e-5"), INTEL_UNCORE_EVENT_DESC(bw_in_port0.unit, "MiB"), INTEL_UNCORE_EVENT_DESC(bw_in_port1, "event=0xff,umask=0x21"), - INTEL_UNCORE_EVENT_DESC(bw_in_port1.scale, "3.814697266e-6"), + INTEL_UNCORE_EVENT_DESC(bw_in_port1.scale, "3.0517578125e-5"), INTEL_UNCORE_EVENT_DESC(bw_in_port1.unit, "MiB"), INTEL_UNCORE_EVENT_DESC(bw_in_port2, "event=0xff,umask=0x22"), - INTEL_UNCORE_EVENT_DESC(bw_in_port2.scale, "3.814697266e-6"), + INTEL_UNCORE_EVENT_DESC(bw_in_port2.scale, "3.0517578125e-5"), INTEL_UNCORE_EVENT_DESC(bw_in_port2.unit, "MiB"), INTEL_UNCORE_EVENT_DESC(bw_in_port3, "event=0xff,umask=0x23"), - INTEL_UNCORE_EVENT_DESC(bw_in_port3.scale, "3.814697266e-6"), + INTEL_UNCORE_EVENT_DESC(bw_in_port3.scale, "3.0517578125e-5"), INTEL_UNCORE_EVENT_DESC(bw_in_port3.unit, "MiB"), INTEL_UNCORE_EVENT_DESC(bw_in_port4, "event=0xff,umask=0x24"), - INTEL_UNCORE_EVENT_DESC(bw_in_port4.scale, "3.814697266e-6"), + INTEL_UNCORE_EVENT_DESC(bw_in_port4.scale, "3.0517578125e-5"), INTEL_UNCORE_EVENT_DESC(bw_in_port4.unit, "MiB"), INTEL_UNCORE_EVENT_DESC(bw_in_port5, "event=0xff,umask=0x25"), - INTEL_UNCORE_EVENT_DESC(bw_in_port5.scale, "3.814697266e-6"), + INTEL_UNCORE_EVENT_DESC(bw_in_port5.scale, "3.0517578125e-5"), INTEL_UNCORE_EVENT_DESC(bw_in_port5.unit, "MiB"), INTEL_UNCORE_EVENT_DESC(bw_in_port6, "event=0xff,umask=0x26"), - INTEL_UNCORE_EVENT_DESC(bw_in_port6.scale, "3.814697266e-6"), + INTEL_UNCORE_EVENT_DESC(bw_in_port6.scale, "3.0517578125e-5"), INTEL_UNCORE_EVENT_DESC(bw_in_port6.unit, "MiB"), INTEL_UNCORE_EVENT_DESC(bw_in_port7, "event=0xff,umask=0x27"), - INTEL_UNCORE_EVENT_DESC(bw_in_port7.scale, "3.814697266e-6"), + INTEL_UNCORE_EVENT_DESC(bw_in_port7.scale, "3.0517578125e-5"), INTEL_UNCORE_EVENT_DESC(bw_in_port7.unit, "MiB"), { /* end: all zeroes */ }, }; -- 2.51.0 From 32c7f1150225694d95a51110a93be25db03bb5db Mon Sep 17 00:00:00 2001 From: Kan Liang Date: Wed, 16 Apr 2025 07:24:25 -0700 Subject: [PATCH 09/16] perf/x86/intel/uncore: Fix the scale of IIO free running counters on ICX There was a mistake in the ICX uncore spec too. The counter increments for every 32 bytes rather than 4 bytes. The same as SNR, there are 1 ioclk and 8 IIO bandwidth in free running counters. Reuse the snr_uncore_iio_freerunning_events(). Fixes: 2b3b76b5ec67 ("perf/x86/intel/uncore: Add Ice Lake server uncore support") Reported-by: Tang Jun Signed-off-by: Kan Liang Signed-off-by: Ingo Molnar Acked-by: Peter Zijlstra Cc: stable@vger.kernel.org Link: https://lore.kernel.org/r/20250416142426.3933977-2-kan.liang@linux.intel.com --- arch/x86/events/intel/uncore_snbep.c | 33 +--------------------------- 1 file changed, 1 insertion(+), 32 deletions(-) diff --git a/arch/x86/events/intel/uncore_snbep.c b/arch/x86/events/intel/uncore_snbep.c index 35da2c486e8d..fb08911a1cf6 100644 --- a/arch/x86/events/intel/uncore_snbep.c +++ b/arch/x86/events/intel/uncore_snbep.c @@ -5485,37 +5485,6 @@ static struct freerunning_counters icx_iio_freerunning[] = { [ICX_IIO_MSR_BW_IN] = { 0xaa0, 0x1, 0x10, 8, 48, icx_iio_bw_freerunning_box_offsets }, }; -static struct uncore_event_desc icx_uncore_iio_freerunning_events[] = { - /* Free-Running IIO CLOCKS Counter */ - INTEL_UNCORE_EVENT_DESC(ioclk, "event=0xff,umask=0x10"), - /* Free-Running IIO BANDWIDTH IN Counters */ - INTEL_UNCORE_EVENT_DESC(bw_in_port0, "event=0xff,umask=0x20"), - INTEL_UNCORE_EVENT_DESC(bw_in_port0.scale, "3.814697266e-6"), - INTEL_UNCORE_EVENT_DESC(bw_in_port0.unit, "MiB"), - INTEL_UNCORE_EVENT_DESC(bw_in_port1, "event=0xff,umask=0x21"), - INTEL_UNCORE_EVENT_DESC(bw_in_port1.scale, "3.814697266e-6"), - INTEL_UNCORE_EVENT_DESC(bw_in_port1.unit, "MiB"), - INTEL_UNCORE_EVENT_DESC(bw_in_port2, "event=0xff,umask=0x22"), - INTEL_UNCORE_EVENT_DESC(bw_in_port2.scale, "3.814697266e-6"), - INTEL_UNCORE_EVENT_DESC(bw_in_port2.unit, "MiB"), - INTEL_UNCORE_EVENT_DESC(bw_in_port3, "event=0xff,umask=0x23"), - INTEL_UNCORE_EVENT_DESC(bw_in_port3.scale, "3.814697266e-6"), - INTEL_UNCORE_EVENT_DESC(bw_in_port3.unit, "MiB"), - INTEL_UNCORE_EVENT_DESC(bw_in_port4, "event=0xff,umask=0x24"), - INTEL_UNCORE_EVENT_DESC(bw_in_port4.scale, "3.814697266e-6"), - INTEL_UNCORE_EVENT_DESC(bw_in_port4.unit, "MiB"), - INTEL_UNCORE_EVENT_DESC(bw_in_port5, "event=0xff,umask=0x25"), - INTEL_UNCORE_EVENT_DESC(bw_in_port5.scale, "3.814697266e-6"), - INTEL_UNCORE_EVENT_DESC(bw_in_port5.unit, "MiB"), - INTEL_UNCORE_EVENT_DESC(bw_in_port6, "event=0xff,umask=0x26"), - INTEL_UNCORE_EVENT_DESC(bw_in_port6.scale, "3.814697266e-6"), - INTEL_UNCORE_EVENT_DESC(bw_in_port6.unit, "MiB"), - INTEL_UNCORE_EVENT_DESC(bw_in_port7, "event=0xff,umask=0x27"), - INTEL_UNCORE_EVENT_DESC(bw_in_port7.scale, "3.814697266e-6"), - INTEL_UNCORE_EVENT_DESC(bw_in_port7.unit, "MiB"), - { /* end: all zeroes */ }, -}; - static struct intel_uncore_type icx_uncore_iio_free_running = { .name = "iio_free_running", .num_counters = 9, @@ -5523,7 +5492,7 @@ static struct intel_uncore_type icx_uncore_iio_free_running = { .num_freerunning_types = ICX_IIO_FREERUNNING_TYPE_MAX, .freerunning = icx_iio_freerunning, .ops = &skx_uncore_iio_freerunning_ops, - .event_descs = icx_uncore_iio_freerunning_events, + .event_descs = snr_uncore_iio_freerunning_events, .format_group = &skx_uncore_iio_freerunning_format_group, }; -- 2.51.0 From 506f981ab40f0b03a11a640cfd77f48b09aff330 Mon Sep 17 00:00:00 2001 From: Kan Liang Date: Wed, 16 Apr 2025 07:24:26 -0700 Subject: [PATCH 10/16] perf/x86/intel/uncore: Fix the scale of IIO free running counters on SPR The scale of IIO bandwidth in free running counters is inherited from the ICX. The counter increments for every 32 bytes rather than 4 bytes. The IIO bandwidth out free running counters don't increment with a consistent size. The increment depends on the requested size. It's impossible to find a fixed increment. Remove it from the event_descs. Fixes: 0378c93a92e2 ("perf/x86/intel/uncore: Support IIO free-running counters on Sapphire Rapids server") Reported-by: Tang Jun Signed-off-by: Kan Liang Signed-off-by: Ingo Molnar Acked-by: Peter Zijlstra Cc: stable@vger.kernel.org Link: https://lore.kernel.org/r/20250416142426.3933977-3-kan.liang@linux.intel.com --- arch/x86/events/intel/uncore_snbep.c | 58 +--------------------------- 1 file changed, 1 insertion(+), 57 deletions(-) diff --git a/arch/x86/events/intel/uncore_snbep.c b/arch/x86/events/intel/uncore_snbep.c index fb08911a1cf6..76d96df1475a 100644 --- a/arch/x86/events/intel/uncore_snbep.c +++ b/arch/x86/events/intel/uncore_snbep.c @@ -6289,69 +6289,13 @@ static struct freerunning_counters spr_iio_freerunning[] = { [SPR_IIO_MSR_BW_OUT] = { 0x3808, 0x1, 0x10, 8, 48 }, }; -static struct uncore_event_desc spr_uncore_iio_freerunning_events[] = { - /* Free-Running IIO CLOCKS Counter */ - INTEL_UNCORE_EVENT_DESC(ioclk, "event=0xff,umask=0x10"), - /* Free-Running IIO BANDWIDTH IN Counters */ - INTEL_UNCORE_EVENT_DESC(bw_in_port0, "event=0xff,umask=0x20"), - INTEL_UNCORE_EVENT_DESC(bw_in_port0.scale, "3.814697266e-6"), - INTEL_UNCORE_EVENT_DESC(bw_in_port0.unit, "MiB"), - INTEL_UNCORE_EVENT_DESC(bw_in_port1, "event=0xff,umask=0x21"), - INTEL_UNCORE_EVENT_DESC(bw_in_port1.scale, "3.814697266e-6"), - INTEL_UNCORE_EVENT_DESC(bw_in_port1.unit, "MiB"), - INTEL_UNCORE_EVENT_DESC(bw_in_port2, "event=0xff,umask=0x22"), - INTEL_UNCORE_EVENT_DESC(bw_in_port2.scale, "3.814697266e-6"), - INTEL_UNCORE_EVENT_DESC(bw_in_port2.unit, "MiB"), - INTEL_UNCORE_EVENT_DESC(bw_in_port3, "event=0xff,umask=0x23"), - INTEL_UNCORE_EVENT_DESC(bw_in_port3.scale, "3.814697266e-6"), - INTEL_UNCORE_EVENT_DESC(bw_in_port3.unit, "MiB"), - INTEL_UNCORE_EVENT_DESC(bw_in_port4, "event=0xff,umask=0x24"), - INTEL_UNCORE_EVENT_DESC(bw_in_port4.scale, "3.814697266e-6"), - INTEL_UNCORE_EVENT_DESC(bw_in_port4.unit, "MiB"), - INTEL_UNCORE_EVENT_DESC(bw_in_port5, "event=0xff,umask=0x25"), - INTEL_UNCORE_EVENT_DESC(bw_in_port5.scale, "3.814697266e-6"), - INTEL_UNCORE_EVENT_DESC(bw_in_port5.unit, "MiB"), - INTEL_UNCORE_EVENT_DESC(bw_in_port6, "event=0xff,umask=0x26"), - INTEL_UNCORE_EVENT_DESC(bw_in_port6.scale, "3.814697266e-6"), - INTEL_UNCORE_EVENT_DESC(bw_in_port6.unit, "MiB"), - INTEL_UNCORE_EVENT_DESC(bw_in_port7, "event=0xff,umask=0x27"), - INTEL_UNCORE_EVENT_DESC(bw_in_port7.scale, "3.814697266e-6"), - INTEL_UNCORE_EVENT_DESC(bw_in_port7.unit, "MiB"), - /* Free-Running IIO BANDWIDTH OUT Counters */ - INTEL_UNCORE_EVENT_DESC(bw_out_port0, "event=0xff,umask=0x30"), - INTEL_UNCORE_EVENT_DESC(bw_out_port0.scale, "3.814697266e-6"), - INTEL_UNCORE_EVENT_DESC(bw_out_port0.unit, "MiB"), - INTEL_UNCORE_EVENT_DESC(bw_out_port1, "event=0xff,umask=0x31"), - INTEL_UNCORE_EVENT_DESC(bw_out_port1.scale, "3.814697266e-6"), - INTEL_UNCORE_EVENT_DESC(bw_out_port1.unit, "MiB"), - INTEL_UNCORE_EVENT_DESC(bw_out_port2, "event=0xff,umask=0x32"), - INTEL_UNCORE_EVENT_DESC(bw_out_port2.scale, "3.814697266e-6"), - INTEL_UNCORE_EVENT_DESC(bw_out_port2.unit, "MiB"), - INTEL_UNCORE_EVENT_DESC(bw_out_port3, "event=0xff,umask=0x33"), - INTEL_UNCORE_EVENT_DESC(bw_out_port3.scale, "3.814697266e-6"), - INTEL_UNCORE_EVENT_DESC(bw_out_port3.unit, "MiB"), - INTEL_UNCORE_EVENT_DESC(bw_out_port4, "event=0xff,umask=0x34"), - INTEL_UNCORE_EVENT_DESC(bw_out_port4.scale, "3.814697266e-6"), - INTEL_UNCORE_EVENT_DESC(bw_out_port4.unit, "MiB"), - INTEL_UNCORE_EVENT_DESC(bw_out_port5, "event=0xff,umask=0x35"), - INTEL_UNCORE_EVENT_DESC(bw_out_port5.scale, "3.814697266e-6"), - INTEL_UNCORE_EVENT_DESC(bw_out_port5.unit, "MiB"), - INTEL_UNCORE_EVENT_DESC(bw_out_port6, "event=0xff,umask=0x36"), - INTEL_UNCORE_EVENT_DESC(bw_out_port6.scale, "3.814697266e-6"), - INTEL_UNCORE_EVENT_DESC(bw_out_port6.unit, "MiB"), - INTEL_UNCORE_EVENT_DESC(bw_out_port7, "event=0xff,umask=0x37"), - INTEL_UNCORE_EVENT_DESC(bw_out_port7.scale, "3.814697266e-6"), - INTEL_UNCORE_EVENT_DESC(bw_out_port7.unit, "MiB"), - { /* end: all zeroes */ }, -}; - static struct intel_uncore_type spr_uncore_iio_free_running = { .name = "iio_free_running", .num_counters = 17, .num_freerunning_types = SPR_IIO_FREERUNNING_TYPE_MAX, .freerunning = spr_iio_freerunning, .ops = &skx_uncore_iio_freerunning_ops, - .event_descs = spr_uncore_iio_freerunning_events, + .event_descs = snr_uncore_iio_freerunning_events, .format_group = &skx_uncore_iio_freerunning_format_group, }; -- 2.51.0 From a5f5e1238f4ff919816f69e77d2537a48911767b Mon Sep 17 00:00:00 2001 From: Dapeng Mi Date: Tue, 15 Apr 2025 10:41:34 +0000 Subject: [PATCH 11/16] perf/x86/intel: Don't clear perf metrics overflow bit unconditionally The below code would always unconditionally clear other status bits like perf metrics overflow bit once PEBS buffer overflows: status &= intel_ctrl | GLOBAL_STATUS_TRACE_TOPAPMI; This is incorrect. Perf metrics overflow bit should be cleared only when fixed counter 3 in PEBS counter group. Otherwise perf metrics overflow could be missed to handle. Closes: https://lore.kernel.org/all/20250225110012.GK31462@noisy.programming.kicks-ass.net/ Fixes: 7b2c05a15d29 ("perf/x86/intel: Generic support for hardware TopDown metrics") Signed-off-by: Dapeng Mi Signed-off-by: Peter Zijlstra (Intel) Signed-off-by: Ingo Molnar Reviewed-by: Kan Liang Cc: stable@vger.kernel.org Link: https://lore.kernel.org/r/20250415104135.318169-1-dapeng1.mi@linux.intel.com --- arch/x86/events/intel/core.c | 13 +++++++++++-- 1 file changed, 11 insertions(+), 2 deletions(-) diff --git a/arch/x86/events/intel/core.c b/arch/x86/events/intel/core.c index 09d2d66c9f21..2b70a3adde2f 100644 --- a/arch/x86/events/intel/core.c +++ b/arch/x86/events/intel/core.c @@ -3049,7 +3049,6 @@ static int handle_pmi_common(struct pt_regs *regs, u64 status) struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events); int bit; int handled = 0; - u64 intel_ctrl = hybrid(cpuc->pmu, intel_ctrl); inc_irq_stat(apic_perf_irqs); @@ -3093,7 +3092,6 @@ static int handle_pmi_common(struct pt_regs *regs, u64 status) handled++; x86_pmu_handle_guest_pebs(regs, &data); static_call(x86_pmu_drain_pebs)(regs, &data); - status &= intel_ctrl | GLOBAL_STATUS_TRACE_TOPAPMI; /* * PMI throttle may be triggered, which stops the PEBS event. @@ -3104,6 +3102,15 @@ static int handle_pmi_common(struct pt_regs *regs, u64 status) */ if (pebs_enabled != cpuc->pebs_enabled) wrmsrl(MSR_IA32_PEBS_ENABLE, cpuc->pebs_enabled); + + /* + * Above PEBS handler (PEBS counters snapshotting) has updated fixed + * counter 3 and perf metrics counts if they are in counter group, + * unnecessary to update again. + */ + if (cpuc->events[INTEL_PMC_IDX_FIXED_SLOTS] && + is_pebs_counter_event_group(cpuc->events[INTEL_PMC_IDX_FIXED_SLOTS])) + status &= ~GLOBAL_STATUS_PERF_METRICS_OVF_BIT; } /* @@ -3123,6 +3130,8 @@ static int handle_pmi_common(struct pt_regs *regs, u64 status) static_call(intel_pmu_update_topdown_event)(NULL, NULL); } + status &= hybrid(cpuc->pmu, intel_ctrl); + /* * Checkpointed counters can lead to 'spurious' PMIs because the * rollback caused by the PMI will have cleared the overflow status -- 2.51.0 From 71dcc11c2cd9e434c34a63154ecadca21c135ddd Mon Sep 17 00:00:00 2001 From: Dapeng Mi Date: Tue, 15 Apr 2025 10:41:35 +0000 Subject: [PATCH 12/16] perf/x86/intel: Allow to update user space GPRs from PEBS records Currently when a user samples user space GPRs (--user-regs option) with PEBS, the user space GPRs actually always come from software PMI instead of from PEBS hardware. This leads to the sampled GPRs to possibly be inaccurate for single PEBS record case because of the skid between counter overflow and GPRs sampling on PMI. For the large PEBS case, it is even worse. If user sets the exclude_kernel attribute, large PEBS would be used to sample user space GPRs, but since PEBS GPRs group is not really enabled, it leads to all samples in the large PEBS record to share the same piece of user space GPRs, like this reproducer shows: $ perf record -e branches:pu --user-regs=ip,ax -c 100000 ./foo $ perf report -D | grep "AX" .... AX 0x000000003a0d4ead .... AX 0x000000003a0d4ead .... AX 0x000000003a0d4ead .... AX 0x000000003a0d4ead .... AX 0x000000003a0d4ead .... AX 0x000000003a0d4ead .... AX 0x000000003a0d4ead .... AX 0x000000003a0d4ead .... AX 0x000000003a0d4ead .... AX 0x000000003a0d4ead .... AX 0x000000003a0d4ead So enable GPRs group for user space GPRs sampling and prioritize reading GPRs from PEBS. If the PEBS sampled GPRs is not user space GPRs (single PEBS record case), perf_sample_regs_user() modifies them to user space GPRs. [ mingo: Clarified the changelog. ] Fixes: c22497f5838c ("perf/x86/intel: Support adaptive PEBS v4") Signed-off-by: Dapeng Mi Signed-off-by: Peter Zijlstra (Intel) Signed-off-by: Ingo Molnar Cc: stable@vger.kernel.org Link: https://lore.kernel.org/r/20250415104135.318169-2-dapeng1.mi@linux.intel.com --- arch/x86/events/intel/ds.c | 8 +++++--- 1 file changed, 5 insertions(+), 3 deletions(-) diff --git a/arch/x86/events/intel/ds.c b/arch/x86/events/intel/ds.c index 1f7e1a692a7a..18c3ab579b8b 100644 --- a/arch/x86/events/intel/ds.c +++ b/arch/x86/events/intel/ds.c @@ -1399,8 +1399,10 @@ static u64 pebs_update_adaptive_cfg(struct perf_event *event) * + precise_ip < 2 for the non event IP * + For RTM TSX weight we need GPRs for the abort code. */ - gprs = (sample_type & PERF_SAMPLE_REGS_INTR) && - (attr->sample_regs_intr & PEBS_GP_REGS); + gprs = ((sample_type & PERF_SAMPLE_REGS_INTR) && + (attr->sample_regs_intr & PEBS_GP_REGS)) || + ((sample_type & PERF_SAMPLE_REGS_USER) && + (attr->sample_regs_user & PEBS_GP_REGS)); tsx_weight = (sample_type & PERF_SAMPLE_WEIGHT_TYPE) && ((attr->config & INTEL_ARCH_EVENT_MASK) == @@ -2123,7 +2125,7 @@ static void setup_pebs_adaptive_sample_data(struct perf_event *event, regs->flags &= ~PERF_EFLAGS_EXACT; } - if (sample_type & PERF_SAMPLE_REGS_INTR) + if (sample_type & (PERF_SAMPLE_REGS_INTR | PERF_SAMPLE_REGS_USER)) adaptive_pebs_save_regs(regs, gprs); } -- 2.51.0 From 7950de14ff5fd8da355d872b887ee8b7b5a1f327 Mon Sep 17 00:00:00 2001 From: Kan Liang Date: Tue, 15 Apr 2025 11:44:07 +0000 Subject: [PATCH 13/16] perf/x86/intel: Add Panther Lake support From PMU's perspective, Panther Lake is similar to the previous generation Lunar Lake. Both are hybrid platforms, with e-core and p-core. The key differences are the ARCH PEBS feature and several new events. The ARCH PEBS is supported in the following patches. The new events will be supported later in perf tool. Share the code path with the Lunar Lake. Only update the name. Signed-off-by: Kan Liang Signed-off-by: Peter Zijlstra (Intel) Signed-off-by: Ingo Molnar Link: https://lkml.kernel.org/r/20250415114428.341182-2-dapeng1.mi@linux.intel.com --- arch/x86/events/intel/core.c | 11 +++++++++-- 1 file changed, 9 insertions(+), 2 deletions(-) diff --git a/arch/x86/events/intel/core.c b/arch/x86/events/intel/core.c index 2b70a3adde2f..00dfe487bd00 100644 --- a/arch/x86/events/intel/core.c +++ b/arch/x86/events/intel/core.c @@ -7314,8 +7314,17 @@ __init int intel_pmu_init(void) name = "meteorlake_hybrid"; break; + case INTEL_PANTHERLAKE_L: + pr_cont("Pantherlake Hybrid events, "); + name = "pantherlake_hybrid"; + goto lnl_common; + case INTEL_LUNARLAKE_M: case INTEL_ARROWLAKE: + pr_cont("Lunarlake Hybrid events, "); + name = "lunarlake_hybrid"; + + lnl_common: intel_pmu_init_hybrid(hybrid_big_small); x86_pmu.pebs_latency_data = lnl_latency_data; @@ -7337,8 +7346,6 @@ __init int intel_pmu_init(void) intel_pmu_init_skt(&pmu->pmu); intel_pmu_pebs_data_source_lnl(); - pr_cont("Lunarlake Hybrid events, "); - name = "lunarlake_hybrid"; break; case INTEL_ARROWLAKE_H: -- 2.51.0 From f6938a562a6249000de211a710807ebf0b8fdf26 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Wed, 16 Apr 2025 20:50:27 +0200 Subject: [PATCH 14/16] perf/core: Fix perf-stat / read() In the zeal to adjust all event->state checks to include the new REVOKED state, one adjustment was made in error. Notably it resulted in read() on the perf filedesc to stop working for any state lower than ERROR, specifically EXIT. This leads to problems with (among others) perf-stat, which wants to read the counts after a program has finished execution. Fixes: da916e96e2de ("perf: Make perf_pmu_unregister() useable") Reported-by: "Mi, Dapeng" Reported-by: James Clark Tested-by: James Clark Signed-off-by: Peter Zijlstra (Intel) Signed-off-by: Ingo Molnar Link: https://lkml.kernel.org/r/77036114-8723-4af9-a068-1d535f4e2e81@linaro.org Link: https://lore.kernel.org/r/20250417080725.GH38216@noisy.programming.kicks-ass.net --- kernel/events/core.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/kernel/events/core.c b/kernel/events/core.c index 2eb9cd5d86a1..e4d7a0c4b308 100644 --- a/kernel/events/core.c +++ b/kernel/events/core.c @@ -6021,7 +6021,7 @@ __perf_read(struct perf_event *event, char __user *buf, size_t count) * error state (i.e. because it was pinned but it couldn't be * scheduled on to the CPU at some point). */ - if (event->state <= PERF_EVENT_STATE_ERROR) + if (event->state == PERF_EVENT_STATE_ERROR) return 0; if (count < event->read_size) -- 2.51.0 From 2839f393c69456bc356738e521b2e70b82977f46 Mon Sep 17 00:00:00 2001 From: Frederic Weisbecker Date: Wed, 9 Apr 2025 15:01:12 +0200 Subject: [PATCH 15/16] perf/core: Fix put_ctx() ordering So there are three situations: * If perf_event_free_task() has removed all the children from the parent list before perf_event_release_kernel() got a chance to even iterate them, then it's all good as there is no get_ctx() pending. * If perf_event_release_kernel() iterates a child event, but it gets freed meanwhile by perf_event_free_task() while the mutexes are temporarily unlocked, it's all good because while locking again the ctx mutex, perf_event_release_kernel() observes TASK_TOMBSTONE. * But if perf_event_release_kernel() frees the child event before perf_event_free_task() got a chance, we may face this scenario: perf_event_release_kernel() perf_event_free_task() -------------------------- ------------------------ mutex_lock(&event->child_mutex) get_ctx(child->ctx) mutex_unlock(&event->child_mutex) mutex_lock(ctx->mutex) mutex_lock(&event->child_mutex) perf_remove_from_context(child) mutex_unlock(&event->child_mutex) mutex_unlock(ctx->mutex) // This lock acquires ctx->refcount == 2 // visibility mutex_lock(ctx->mutex) ctx->task = TASK_TOMBSTONE mutex_unlock(ctx->mutex) wait_var_event() // enters prepare_to_wait() since // ctx->refcount == 2 // is guaranteed to be seen set_current_state(TASK_INTERRUPTIBLE) smp_mb() if (ctx->refcount != 1) schedule() put_ctx() // NOT fully ordered! Only RELEASE semantics refcount_dec_and_test() atomic_fetch_sub_release() // So TASK_TOMBSTONE is not guaranteed to be seen if (ctx->task == TASK_TOMBSTONE) wake_up_var() Basically it's a broken store buffer: perf_event_release_kernel() perf_event_free_task() -------------------------- ------------------------ ctx->task = TASK_TOMBSTONE smp_store_release(&ctx->refcount, ctx->refcount - 1) smp_mb() READ_ONCE(ctx->refcount) READ_ONCE(ctx->task) So we need a smp_mb__after_atomic() before looking at ctx->task. Fixes: 59f3aa4a3ee2 ("perf: Simplify perf_event_free_task() wait") Signed-off-by: Frederic Weisbecker Signed-off-by: Peter Zijlstra (Intel) Signed-off-by: Ingo Molnar Link: https://lkml.kernel.org/r/Z_ZvmEhjkAhplCBE@localhost.localdomain --- kernel/events/core.c | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-) diff --git a/kernel/events/core.c b/kernel/events/core.c index e4d7a0c4b308..1a19df9d54fd 100644 --- a/kernel/events/core.c +++ b/kernel/events/core.c @@ -1271,9 +1271,10 @@ static void put_ctx(struct perf_event_context *ctx) if (ctx->task && ctx->task != TASK_TOMBSTONE) put_task_struct(ctx->task); call_rcu(&ctx->rcu_head, free_ctx); - } else if (ctx->task == TASK_TOMBSTONE) { - smp_mb(); /* pairs with wait_var_event() */ - wake_up_var(&ctx->refcount); + } else { + smp_mb__after_atomic(); /* pairs with wait_var_event() */ + if (ctx->task == TASK_TOMBSTONE) + wake_up_var(&ctx->refcount); } } -- 2.51.0 From 162c9e3faf58eef653c74d0c774e6583d9225467 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Tue, 15 Apr 2025 12:12:52 +0200 Subject: [PATCH 16/16] perf/core: Fix event->parent life-time issue Due to an oversight in merging: da916e96e2de ("perf: Make perf_pmu_unregister() useable") on top of: 56799bc03565 ("perf: Fix hang while freeing sigtrap event") .. it is now possible to hit put_event(EVENT_TOMBSTONE), which makes the computer sad. This also means that for the event->parent == EVENT_TOMBSTONE, the put_event() matching inherit_event() has gone missing. Previously this was done in perf_event_release_kernel() after calling perf_remove_from_context(), but with it delegated to put_event(), this case is now entirely missed, leading to leaks. Fixes: da916e96e2de ("perf: Make perf_pmu_unregister() useable") Reported-by: kernel test robot Tested-by: kernel test robot Tested-by: James Clark Tested-by: Venkat Rao Bagalkote Signed-off-by: Peter Zijlstra (Intel) Signed-off-by: Ingo Molnar Closes: https://lore.kernel.org/oe-lkp/202504131701.941039cd-lkp@intel.com Link: https://lkml.kernel.org/r/20250415131446.GN5600@noisy.programming.kicks-ass.net --- kernel/events/core.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/kernel/events/core.c b/kernel/events/core.c index 1a19df9d54fd..43d87de3d4f2 100644 --- a/kernel/events/core.c +++ b/kernel/events/core.c @@ -2343,6 +2343,7 @@ static void perf_child_detach(struct perf_event *event) * not being a child event. See for example unaccount_event(). */ event->parent = EVENT_TOMBSTONE; + put_event(parent_event); } static bool is_orphaned_event(struct perf_event *event) @@ -5688,7 +5689,7 @@ static void put_event(struct perf_event *event) _free_event(event); /* Matches the refcount bump in inherit_event() */ - if (parent) + if (parent && parent != EVENT_TOMBSTONE) put_event(parent); } -- 2.51.0