kernel/linux.git/tools/perf/Documentation, branch v6.18.21

perf docs: Document building with Clang

2025-10-06T19:49:25+00:00

Add example commands for building perf with Clang. Since recent Android NDK releases use Clang as the default compiler, a separate Android specific document is no longer needed; point to the general build documentation instead. Signed-off-by: Leo Yan Reviewed-by: Ian Rogers Link: https://lore.kernel.org/r/20251006-perf_build_android_ndk-v3-9-4305590795b2@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

perf check: Add libLLVM feature

2025-10-06T18:24:13+00:00

Advertise when perf is built with the HAVE_LIBLLVM_SUPPORT option. Committer testing: $ perf -vv | grep LLVM libLLVM: [ on ] # HAVE_LIBLLVM_SUPPORT $ And the form to use in scripts, notably the tools/perf/tests/shell/ 'perf test' ones: $ perf check feature libllvm libLLVM: [ on ] # HAVE_LIBLLVM_SUPPORT $ perf check -q feature libllvm && echo LLVM is present LLVM is present $ perf check -q feature liballvm && echo ALLVM is present $ Signed-off-by: Ian Rogers Tested-by: Arnaldo Carvalho de Melo Cc: Adrian Hunter Cc: Alexander Shishkin Cc: Alexandre Ghiti Cc: Andi Kleen Cc: Athira Rajeev Cc: Bill Wendling Cc: Charlie Jenkins Cc: Collin Funk Cc: Dmitriy Vyukov Cc: Dr. David Alan Gilbert Cc: Eric Biggers Cc: Haibo Xu Cc: Ingo Molnar Cc: James Clark Cc: Jiri Olsa Cc: Justin Stitt Cc: Li Huafei Cc: Masami Hiramatsu Cc: Namhyung Kim Cc: Nathan Chancellor Cc: Nick Desaulniers Cc: Peter Zijlstra Cc: Song Liu Cc: Stephen Brennan Signed-off-by: Arnaldo Carvalho de Melo

perf record: Add ratio-to-prev term

2025-10-03T19:49:51+00:00

Provide ratio-to-prev term which allows the user to set the event sample period of two events corresponding to a desired ratio. If using on an Intel x86 platform with Auto Counter Reload support, also set corresponding event's config2 attribute with a bitmask which counters to reset and which counters to sample if the desired ratio is met or exceeded. On other platforms, only the sample period is affected by the ratio-to-prev term. Reviewed-by: Ian Rogers Signed-off-by: Thomas Falcon Cc: Adrian Hunter Cc: Alexander Shishkin Cc: Andi Kleen Cc: Dapeng Mi Cc: Ingo Molnar Cc: Jiri Olsa Cc: Kan Liang Cc: Mark Rutland Cc: Namhyung Kim Cc: Peter Zijlstra Signed-off-by: Arnaldo Carvalho de Melo

perf tools: Fix duplicated words in documentation and comments

2025-10-01T12:44:02+00:00

- "the the" - "in in" - "a a" Reviewed-by: Ian Rogers Signed-off-by: Markus Heidelberg Cc: Adrian Hunter Cc: Alexander Shishkin Cc: Ingo Molnar Cc: Jiri Olsa Cc: Kan Liang Cc: Mark Rutland Cc: Namhyung Kim Cc: Peter Zijlstra Signed-off-by: Arnaldo Carvalho de Melo

perf bench mem: Add mmap() workloads

2025-09-19T15:43:59+00:00

Add two mmap() workloads: one that eagerly populates a region and another that demand faults it in. The intent is to probe the memory subsytem performance incurred by mmap(). $ perf bench mem mmap -s 4gb -p 4kb -l 10 -f populate # Running 'mem/mmap' benchmark: # function 'populate' (Eagerly populated map()) # Copying 4gb bytes ... 1.811691 GB/sec $ perf bench mem mmap -s 4gb -p 2mb -l 10 -f populate # Running 'mem/mmap' benchmark: # function 'populate' (Eagerly populated mmap()) # Copying 4gb bytes ... 12.272017 GB/sec $ perf bench mem mmap -s 4gb -p 1gb -l 10 -f populate # Running 'mem/mmap' benchmark: # function 'populate' (Eagerly populated mmap()) # Copying 4gb bytes ... 17.085927 GB/sec Signed-off-by: Ankur Arora Tested-by: Arnaldo Carvalho de Melo Cc: Andrew Morton Cc: Andy Lutomirski Cc: Boris Ostrovsky Cc: Borislav Petkov Cc: Dave Hansen Cc: David Hildenbrand Cc: H. Peter Anvin Cc: Ingo Molnar Cc: Konrad Rzeszutek Wilk Cc: Mateusz Guzik Cc: Matthew Wilcox Cc: Namhyung Kim Cc: Peter Zijlstra Cc: Raghavendra K T Cc: Thomas Gleixner Signed-off-by: Arnaldo Carvalho de Melo

perf bench mem: Allow chunking on a memory region

2025-09-19T15:43:38+00:00

There can be a significant gap in memset/memcpy performance depending on the size of the region being operated on. With chunk-size=4kb: $ echo madvise > /sys/kernel/mm/transparent_hugepage/enabled $ perf bench mem memset -p 4kb -k 4kb -s 4gb -l 10 -f x86-64-stosq # Running 'mem/memset' benchmark: # function 'x86-64-stosq' (movsq-based memset() in arch/x86/lib/memset_64.S) # Copying 4gb bytes ... 13.011655 GB/sec With chunk-size=1gb: $ echo madvise > /sys/kernel/mm/transparent_hugepage/enabled $ perf bench mem memset -p 4kb -k 1gb -s 4gb -l 10 -f x86-64-stosq # Running 'mem/memset' benchmark: # function 'x86-64-stosq' (movsq-based memset() in arch/x86/lib/memset_64.S) # Copying 4gb bytes ... 21.936355 GB/sec So, allow the user to specify the chunk-size. The default value is identical to the total size of the region, which preserves current behaviour. Reviewed-by: Namhyung Kim Signed-off-by: Ankur Arora Tested-by: Arnaldo Carvalho de Melo Cc: Andrew Morton Cc: Andy Lutomirski Cc: Boris Ostrovsky Cc: Borislav Petkov Cc: Dave Hansen Cc: David Hildenbrand Cc: H. Peter Anvin Cc: Ingo Molnar Cc: Konrad Rzeszutek Wilk Cc: Mateusz Guzik Cc: Matthew Wilcox Cc: Peter Zijlstra Cc: Raghavendra K T Cc: Thomas Gleixner Signed-off-by: Arnaldo Carvalho de Melo

perf bench mem: Allow mapping of hugepages

2025-09-19T15:43:26+00:00

Page sizes that can be selected: 4KB, 2MB, 1GB. Both the reservation and node from which hugepages are allocated from are expected to be addressed by the user. An example of page-size selection: $ perf bench mem memset -s 4gb -p 2mb # Running 'mem/memset' benchmark: # function 'default' (Default memset() provided by glibc) # Copying 4gb bytes ... 14.919194 GB/sec # function 'x86-64-unrolled' (unrolled memset() in arch/x86/lib/memset_64.S) # Copying 4gb bytes ... 11.514503 GB/sec # function 'x86-64-stosq' (movsq-based memset() in arch/x86/lib/memset_64.S) # Copying 4gb bytes ... 12.600568 GB/sec Signed-off-by: Ankur Arora Tested-by: Arnaldo Carvalho de Melo Cc: Andrew Morton Cc: Andy Lutomirski Cc: Boris Ostrovsky Cc: Borislav Petkov Cc: Dave Hansen Cc: David Hildenbrand Cc: H. Peter Anvin Cc: Ingo Molnar Cc: Konrad Rzeszutek Wilk Cc: Mateusz Guzik Cc: Matthew Wilcox Cc: Namhyung Kim Cc: Peter Zijlstra Cc: Raghavendra K T Cc: Thomas Gleixner Signed-off-by: Arnaldo Carvalho de Melo

perf tools: Update header documentation on BPF_PROG_INFO

2025-09-19T15:14:30+00:00

Update the perf.data file format description on header section HEADER_BPF_PROG_INFO. The information is taken from process_bpf_prog_info() and write_bpf_prog_info() from file util/header.c. Reviewed-by: Jan Polensky Signed-off-by: Thomas Richter Acked-by: Namhyung Kim Acked-by: Sumanth Korikkar Cc: Alexander Gordeev Cc: Heiko Carstens Cc: Vasily Gorbik Signed-off-by: Arnaldo Carvalho de Melo

perf trace: Add --max-summary option

2025-09-19T15:14:29+00:00

The --max-summary option is to limit the number of output lines for syscall summary stats. The max applies to each entries like thread and cgroups. For total summary, it will just print up to the given number. For example, $ sudo perf trace -as --max-summary 3 sleep 0.1 ThreadPoolServi (1011651), 114 events, 14.8% syscall calls errors total min avg max stddev (msec) (msec) (msec) (msec) (%) --------------- -------- ------ -------- --------- --------- --------- ------ epoll_wait 38 0 95.589 0.000 2.515 11.153 28.98% futex 9 0 0.040 0.002 0.004 0.014 28.63% read 10 0 0.037 0.003 0.004 0.005 4.67% sleep (1050529), 250 events, 32.4% syscall calls errors total min avg max stddev (msec) (msec) (msec) (msec) (%) --------------- -------- ------ -------- --------- --------- --------- ------ clock_nanosleep 1 0 100.156 100.156 100.156 100.156 0.00% execve 4 3 1.020 0.005 0.255 0.989 95.93% openat 36 17 0.416 0.003 0.012 0.029 10.58% ... And this is for per-cgroup summary using BPF. $ sudo perf trace -as --max-summary 3 --summary-mode=cgroup --bpf-summary sleep 0.1 cgroup /user.slice/user-657345.slice/user@657345.service/session.slice/org.gnome.Shell@x11.service, 12 events syscall calls errors total min avg max stddev (msec) (msec) (msec) (msec) (%) --------------- -------- ------ -------- --------- --------- --------- ------ recvmsg 8 7 0.016 0.001 0.002 0.006 39.73% ppoll 1 0 0.014 0.014 0.014 0.014 0.00% write 2 0 0.010 0.002 0.005 0.008 61.02% cgroup /user.slice/user-657345.slice/session-4.scope, 73 events syscall calls errors total min avg max stddev (msec) (msec) (msec) (msec) (%) --------------- -------- ------ -------- --------- --------- --------- ------ epoll_wait 8 0 13.461 0.010 1.683 12.235 89.66% ioctl 20 0 0.204 0.001 0.010 0.113 54.01% writev 11 0 0.164 0.004 0.015 0.042 20.34% Reviewed-by: Howard Chu Signed-off-by: Namhyung Kim Tested-by: Arnaldo Carvalho de Melo Cc: Adrian Hunter Cc: Ian Rogers Cc: Ingo Molnar Cc: Jiri Olsa Cc: Kan Liang Cc: Peter Zijlstra Signed-off-by: Arnaldo Carvalho de Melo

perf parse-events: Add 'X' modifier to exclude an event from being regrouped

2025-09-12T18:53:32+00:00

The function parse_events__sort_events_and_fix_groups is needed to fix uncore events like: ``` $ perf stat -e '{data_read,data_write}' ... ``` so that the multiple uncore PMUs have a group each of data_read and data_write events. The same function will perform architecture sorting and group fixing, in particular for Intel topdown/perf-metric events. Grouping multiple perf metric events together causes perf_event_open to fail as the group can only support one. This means command lines like: ``` $ perf stat -e 'slots,slots' ... ``` fail as the slots events are forced into a group together to try to satisfy the perf-metric event constraints. As the user may know better than parse_events__sort_events_and_fix_groups add a 'X' modifier to skip its regrouping behavior. This allows the following to succeed rather than fail on the second slots event being opened: ``` $ perf stat -e 'slots,slots:X' -a sleep 1 Performance counter stats for 'system wide': 6,834,154,071 cpu_core/slots/ (50.13%) 5,548,629,453 cpu_core/slots/X (49.87%) 1.002634606 seconds time elapsed ``` Closes: https://lore.kernel.org/lkml/20250822082233.1850417-1-dapeng1.mi@linux.intel.com/ Reported-by: Dapeng Mi Reported-by: Xudong Hao Reviewed-by: Dapeng Mi Signed-off-by: Ian Rogers Tested-by: Dapeng Mi Cc: Adrian Hunter Cc: Alexander Shishkin Cc: Andi Kleen Cc: Greg Kroah-Hartman Cc: Howard Chu Cc: Ingo Molnar Cc: James Clark Cc: Jiri Olsa Cc: Kan Liang Cc: Mark Rutland Cc: Namhyung Kim Cc: Peter Zijlstra Cc: Thomas Falcon Cc: Yoshihiro Furudera Signed-off-by: Arnaldo Carvalho de Melo