kernel/linux.git/tools/perf/util, branch v6.6.127

perf tools: Fix split kallsyms DSO counting

2026-01-11T14:21:35+00:00

[ Upstream commit ad0b9c4865b98dc37f4d606d26b1c19808796805 ] It's counted twice as it's increased after calling maps__insert(). I guess we want to increase it only after it's added properly. Reviewed-by: Ian Rogers Fixes: 2e538c4a1847291cf ("perf tools: Improve kernel/modules symbol lookup") Signed-off-by: Namhyung Kim Signed-off-by: Sasha Levin

perf arm_spe: Fix memset subclass in operation

2026-01-11T14:21:26+00:00

[ Upstream commit 33e1fffea492b7158a168914dc0da6aedf78d08e ] The operation subclass is extracted from bits [7..1] of the payload. Since bit [0] is not parsed, there is no chance to match the memset type (0x25). As a result, the memset payload is never parsed successfully. Instead of extracting a unified bit field, change to extract the specific bits for each operation subclass. Fixes: 34fb60400e32 ("perf arm-spe: Add raw decoding for SPEv1.3 MTE and MOPS load/store") Signed-off-by: Leo Yan Reviewed-by: Ian Rogers Reviewed-by: James Clark Signed-off-by: Namhyung Kim Signed-off-by: Sasha Levin

perf arm-spe: Extend branch operations

2026-01-11T14:21:26+00:00

[ Upstream commit 64d86c03e1441742216b6332bdfabfb6ede31662 ] In Arm ARM (ARM DDI 0487, L.a), the section "D18.2.7 Operation Type packet", the branch subclass is extended for Call Return (CR), Guarded control stack data access (GCS). This commit adds support CR and GCS operations. The IND (indirect) operation is defined only in bit [1], its macro is updated accordingly. Move the COND (Conditional) macro into the same group with other operations for better maintenance. Reviewed-by: Ian Rogers Reviewed-by: James Clark Signed-off-by: Leo Yan Link: https://lore.kernel.org/r/20250304111240.3378214-8-leo.yan@arm.com Signed-off-by: Namhyung Kim Stable-dep-of: 33e1fffea492 ("perf arm_spe: Fix memset subclass in operation") Signed-off-by: Sasha Levin

perf lock contention: Load kernel map before lookup

2026-01-11T14:21:19+00:00

[ Upstream commit 553d18c98a896094b99a01765b9698b204183d49 ] On some machines, it caused troubles when it tried to find kernel symbols. I think it's because kernel modules and kallsyms are messed up during load and split. Basically we want to make sure the kernel map is loaded and the code has it in the lock_contention_read(). But recently we added more lookups in the lock_contention_prepare() which is called before _read(). Also the kernel map (kallsyms) may not be the first one in the group like on ARM. Let's use machine__kernel_map() rather than just loading the first map. Reviewed-by: Ian Rogers Fixes: 688d2e8de231c54e ("perf lock contention: Add -l/--lock-addr option") Signed-off-by: Namhyung Kim Signed-off-by: Sasha Levin

perf maps: Add maps__load_first()

2026-01-11T14:21:18+00:00

[ Upstream commit e77b0236cd0cd1572c6a9b25097b207eab799e74 ] Avoid bpf_lock_contention_read touching the internal maps data structure by adding a helper function. As access is done directly on the map in maps, hold the read lock to stop it being removed. Signed-off-by: Ian Rogers Cc: Adrian Hunter Cc: Alexander Shishkin Cc: Andi Kleen Cc: Athira Rajeev Cc: Changbin Du Cc: Colin Ian King Cc: Dmitrii Dolgov <9erthalion6@gmail.com> Cc: German Gomez Cc: Guilherme Amadio Cc: Huacai Chen Cc: Ingo Molnar Cc: James Clark Cc: Jiri Olsa Cc: K Prateek Nayak Cc: Kajol Jain Cc: Kan Liang Cc: Leo Yan Cc: Li Dong Cc: Liam Howlett Cc: Mark Rutland Cc: Masami Hiramatsu Cc: Miguel Ojeda Cc: Ming Wang Cc: Namhyung Kim Cc: Nick Terrell Cc: Paolo Bonzini Cc: Peter Zijlstra Cc: Ravi Bangoria Cc: Sandipan Das Cc: Sean Christopherson Cc: Steinar H. Gunderson Cc: Vincent Whitchurch Cc: Wenyu Liu Cc: Yang Jihong Link: https://lore.kernel.org/r/20231207011722.1220634-20-irogers@google.com Signed-off-by: Arnaldo Carvalho de Melo Stable-dep-of: 553d18c98a89 ("perf lock contention: Load kernel map before lookup") Signed-off-by: Sasha Levin

Revert "perf dso: Add missed dsoput to dsoload_kcore"

2025-11-24T09:30:10+00:00

This reverts commit e5de9ea7796e79f3cd082624f788cc3442bff2a8. The patch introduced `map__zput(new_node->map)` in the kcore load path, causing a segmentation fault when running `perf c2c report`. The issue arises because `maps__merge_in` directly modifies and inserts the caller's `new_map`, causing it to be freed prematurely while still referenced by kmaps. Later branchs (6.12, 6.15, 6.16) are not affected because they use a different merge approach with a lazily sorted array, which avoids modifying the original `new_map`. Fixes: e5de9ea7796e ("perf dso: Add missed dso__put to dso__load_kcore") Signed-off-by: jingxian.li Signed-off-by: Greg Kroah-Hartman

perf python: split Clang options when invoking Popen

2025-10-19T14:30:43+00:00

[ Upstream commit c6a43bc3e8f6102a47da0d2e53428d08f00172fb ] When passing a list to subprocess.Popen, each element maps to one argv token. Current code bundles multiple Clang flags into a single element, something like: cmd = ['clang', '--target=x86_64-linux-gnu -fintegrated-as -Wno-cast-function-type-mismatch', 'test-hello.c'] So Clang only sees one long, invalid option instead of separate flags, as a result, the script cannot capture any log via PIPE. Fix this by using shlex.split() to separate the string so each option becomes its own argv element. The fixed list will be: cmd = ['clang', '--target=x86_64-linux-gnu', '-fintegrated-as', '-Wno-cast-function-type-mismatch', 'test-hello.c'] Fixes: 09e6f9f98370 ("perf python: Fix splitting CC into compiler and options") Signed-off-by: Leo Yan Reviewed-by: Ian Rogers Link: https://lore.kernel.org/r/20251006-perf_build_android_ndk-v3-2-4305590795b2@arm.com Cc: Palmer Dabbelt Cc: Albert Ou Cc: Alexandre Ghiti Cc: Nick Desaulniers Cc: Justin Stitt Cc: Bill Wendling Cc: Adrian Hunter Cc: Arnaldo Carvalho de Melo Cc: Jiri Olsa Cc: Namhyung Kim Cc: Nathan Chancellor Cc: James Clark Cc: linux-riscv@lists.infradead.org Cc: llvm@lists.linux.dev Cc: Paul Walmsley Cc: linux-kernel@vger.kernel.org Cc: linux-perf-users@vger.kernel.org Signed-off-by: Arnaldo Carvalho de Melo Signed-off-by: Sasha Levin

perf evsel: Ensure the fallback message is always written to

2025-10-19T14:30:41+00:00

[ Upstream commit 24937ee839e4bbc097acde73eeed67812bad2d99 ] The fallback message is unconditionally printed in places like record__open(). If no fallback is attempted this can lead to printing uninitialized data, crashes, etc. Fixes: c0a54341c0e89333 ("perf evsel: Introduce event fallback method") Signed-off-by: Ian Rogers Cc: Adrian Hunter Cc: Alexander Shishkin Cc: Howard Chu Cc: Ingo Molnar Cc: James Clark Cc: Jiri Olsa Cc: Kan Liang Cc: Mark Rutland Cc: Namhyung Kim Cc: Peter Zijlstra Signed-off-by: Arnaldo Carvalho de Melo Signed-off-by: Sasha Levin

perf tools: Add fallback for exclude_guest

2025-10-19T14:30:41+00:00

[ Upstream commit bb6e7cb11d97ce1957894d30d13bfad3e8bfefe9 ] Commit 7b100989b4f6bce70 ("perf evlist: Remove __evlist__add_default") changed to parse "cycles:P" event instead of creating a new cycles event for perf record. But it also changed the way how modifiers are handled so it doesn't set the exclude_guest bit by default. It seems Apple M1 PMU requires exclude_guest set and returns EOPNOTSUPP if not. Let's add a fallback so that it can work with default events. Also update perf stat hybrid tests to handle possible u or H modifiers. Reviewed-by: Ian Rogers Reviewed-by: James Clark Reviewed-by: Ravi Bangoria Acked-by: Kan Liang Cc: James Clark Cc: Atish Patra Cc: Mingwei Zhang Cc: Kajol Jain Cc: Thomas Richter Cc: Palmer Dabbelt Link: https://lore.kernel.org/r/20241016062359.264929-2-namhyung@kernel.org Fixes: 7b100989b4f6bce70 ("perf evlist: Remove __evlist__add_default") Signed-off-by: Namhyung Kim Stable-dep-of: 24937ee839e4 ("perf evsel: Ensure the fallback message is always written to") Signed-off-by: Sasha Levin

perf session: Fix handling when buffer exceeds 2 GiB

2025-10-19T14:30:41+00:00

[ Upstream commit c17dda8013495d8132c976cbf349be9949d0fbd1 ] If a user specifies an AUX buffer larger than 2 GiB, the returned size may exceed 0x80000000. Since the err variable is defined as a signed 32-bit integer, such a value overflows and becomes negative. As a result, the perf record command reports an error: 0x146e8 [0x30]: failed to process type: 71 [Unknown error 183711232] Change the type of the err variable to a signed 64-bit integer to accommodate large buffer sizes correctly. Fixes: d5652d865ea734a1 ("perf session: Add ability to skip 4GiB or more") Reported-by: Tamas Zsoldos Signed-off-by: Leo Yan Acked-by: Namhyung Kim Cc: Adrian Hunter Cc: Ian Rogers Cc: Jiri Olsa Link: https://lore.kernel.org/r/20250808-perf_fix_big_buffer_size-v1-1-45f45444a9a4@arm.com Signed-off-by: Arnaldo Carvalho de Melo Signed-off-by: Sasha Levin

kernel/linux.git/tools/perf/util, branch v6.6.127

perf tools: Fix split kallsyms DSO counting

perf arm_spe: Fix memset subclass in operation

perf arm-spe: Extend branch operations

perf lock contention: Load kernel map before lookup

perf maps: Add maps__load_first()

Revert "perf dso: Add missed dso__put to dso__load_kcore"

perf python: split Clang options when invoking Popen

perf evsel: Ensure the fallback message is always written to

perf tools: Add fallback for exclude_guest

perf session: Fix handling when buffer exceeds 2 GiB

Revert "perf dso: Add missed dsoput to dsoload_kcore"