kernel/linux.git/tools/perf/builtin-trace.c, branch v7.1

perf trace: Add beautifier script for fsmount flags

2026-05-14T21:48:32+00:00

And move the existing one to fsmount_attr.sh to be more precise. Now the fsmount_flags[] is generated from the mount.h like below. The ilog2() + 1 is an existing pattern to handle bit flags. $ cat tools/perf/trace/beauty/generated/fsmount_arrays.c static const char *fsmount_flags[] = { [ilog2(0x00000001) + 1] = "CLOEXEC", [ilog2(0x00000002) + 1] = "NAMESPACE", }; It was found by Sashiko during the review. Reviewed-by: Ian Rogers Signed-off-by: Namhyung Kim

perf tools: Use calloc() where applicable

2026-04-09T02:21:05+00:00

Instead of using zalloc(nr_entries * sizeof_entry) that is what calloc() does. In some places where linux/zalloc.h isn't needed, remove it, add when needed and was getting it indirectly. Signed-off-by: Arnaldo Carvalho de Melo Signed-off-by: Namhyung Kim

perf trace: Fix potential u64 underflow in duration calculation

2026-04-03T01:43:35+00:00

Although it happens very rarely, in case of out-of-order events (i.e. due to CPU migration when a syscall is executed), the calculation of event duration might underflow and thus a bogus value is printed: 2.804 ( 0.001 ms): :49553/49553 rt_sigaction(sig: QUIT, act: 0x7fff403ed6e0, oact: 0x7fff403ed780, sigsetsize: 8) = 0 2.807 ( 0.001 ms): :49553/49553 rt_sigaction(sig: CHLD, act: 0x7fff403ed6e0, oact: 0x7fff403ed780, sigsetsize: 8) = 0 2.815 (18446744073709.438 ms): :49553/49553 execve(filename: 0xbb173a30, argv: 0x55aabb171930, envp: 0x55aabb171120) = 0 2.815 ( 0.534 ms): pwd/49553 ... [continued]: execve()) = 0 Check for possible underflow first and in case of a bogus value, do not print it. 2.804 ( 0.001 ms): :49553/49553 rt_sigaction(sig: QUIT, act: 0x7fff403ed6e0, oact: 0x7fff403ed780, sigsetsize: 8) = 0 2.807 ( 0.001 ms): :49553/49553 rt_sigaction(sig: CHLD, act: 0x7fff403ed6e0, oact: 0x7fff403ed780, sigsetsize: 8) = 0 2.815 ( ): :49553/49553 execve(filename: 0xbb173a30, argv: 0x55aabb171930, envp: 0x55aabb171120) = 0 2.815 ( 0.534 ms): pwd/49553 ... [continued]: execve()) = 0 Signed-off-by: Michael Petlan Reviewed-by: Ian Rogers Signed-off-by: Namhyung Kim

perf trace: Skip unnecessary synthesis for summary-only mode

2026-04-02T19:51:09+00:00

It needs to synthesize task info for the comm name. The mmap information is only needed for callchain symbolization which is not used by the summary mode. Also total or cgroup summary mode don't require the task info. Let's skip the processing if possible. Reviewed-by: Ian Rogers Signed-off-by: Namhyung Kim

perf callchain: Refactor callchain option parsing

2026-03-19T21:42:46+00:00

record_opts__parse_callchain is shared by builtin-record and builtin-trace, it is declared in callchain.h. Move the declaration to callchain.c for consistency with the header. In other cases make the option callback a small static stub that then calls into callchain.c. Make the no argument '-g' callchain option just a short-cut for '--call-graph fp' so that there is consistency in how the arguments are handled. This requires the const char* string to be strdup-ed in __parse_callchain_report_opt. For consistency also make parse_callchain_record use strdup and remove some unnecessary casts. Also, be more explicit about the '-g' behavior if there is a .perfconfig file setting. Signed-off-by: Ian Rogers Tested-by: Thomas Richter Signed-off-by: Namhyung Kim

perf bpf_map: Remove unused code

2026-03-13T21:26:28+00:00

bpf_map__fprintf is unused so delete it, the header file declaring it and the now unused static helper functions. Signed-off-by: Ian Rogers Signed-off-by: Namhyung Kim

perf trace: Avoid an ERR_PTR in syscall_stats

2026-03-03T01:13:19+00:00

hashmap__new may return an ERR_PTR and previously this would be assigned to syscall_stats meaning all use of syscall_stats needs to test for NULL (uninitialized) or an ERR_PTR. Given the only reason hashmap__new can fail is ENOMEM, just use NULL to indicate the allocation failure and avoid the code having to test for NULL and IS_ERR. Fixes: 96f202eab813 (perf trace: Fix IS_ERR() vs NULL check bug) Signed-off-by: Ian Rogers Signed-off-by: Namhyung Kim

perf trace: Fix IS_ERR() vs NULL check bug

2026-02-26T18:48:14+00:00

The alloc_syscall_stats() function always returns an error pointer (ERR_PTR) on failure. So replace NULL check with IS_ERR() check after calling delete_syscall_stats() function. Fixes: ef2da619b132c6f74 ("perf trace: Convert syscall_stats to hashmap") Signed-off-by: wangguangju Reviewed-by: Howard Chu Acked-by: Ian Rogers Signed-off-by: Namhyung Kim

perf thread: Add optional e_flags output argument to thread__e_machine

2026-01-26T21:21:20+00:00

The e_flags are needed to accurately compute complete perf register information for CSKY. Add the ability to read and have this value associated with a thread. This change doesn't wire up the use of the e_flags except in disasm where use already exists but just wasn't set up yet. Signed-off-by: Ian Rogers Cc: Aditya Bodkhe Cc: Adrian Hunter Cc: Alexander Shishkin Cc: Athira Rajeev Cc: Chun-Tse Shao Cc: Guo Ren Cc: Howard Chu Cc: Ingo Molnar Cc: James Clark Cc: Jiri Olsa Cc: Namhyung Kim Cc: Peter Zijlstra Cc: Sergei Trofimovich Cc: Shimin Guo Cc: Stephen Brennan Cc: Swapnil Sapkal Cc: Tianyou Li Signed-off-by: Arnaldo Carvalho de Melo

perf trace: Deal with compiler const checks

2026-01-20T23:49:59+00:00

The strchr() function these days return const/non-const based on the arg it receives, and sometimes we need to use casts when we're dealing with variables that are used in code that needs to safely change the returned value and sometimes not (as it points to really const areas). Tweak one such case. Reviewed-by: Ian Rogers Signed-off-by: Arnaldo Carvalho de Melo