kernel/linux.git/tools/perf/util/header.c, branch v7.0-rc7

perf header: Add e_machine/e_flags to the header

2026-02-03T21:01:27+00:00

Add 64-bits of feature data to record the ELF machine and flags. This allows readers to initialize based on the data. For example, `perf kvm stat` wants to initialize based on the kind of data to be read, but at initialization time there are no threads to base this data upon and using the host means cross platform support won't work. The values in the perf_env also act as a cache for these within the session. Signed-off-by: Ian Rogers Cc: Aditya Bodkhe Cc: Adrian Hunter Cc: Albert Ou Cc: Alexander Shishkin Cc: Alexandre Ghiti Cc: Andi Kleen Cc: Andrew Jones Cc: Anubhav Shelat Cc: Anup Patel Cc: Athira Rajeev Cc: Blake Jones Cc: Chun-Tse Shao Cc: Dapeng Mi Cc: Dmitriy Vyukov Cc: Howard Chu Cc: Ingo Molnar Cc: James Clark Cc: Jiri Olsa Cc: John Garry Cc: Leo Yan Cc: Namhyung Kim Cc: Palmer Dabbelt Cc: Paul Walmsley Cc: Peter Zijlstra Cc: Quan Zhou Cc: Shimin Guo Cc: Swapnil Sapkal Cc: Thomas Falcon Cc: Will Deacon Cc: Yunseong Kim Signed-off-by: Arnaldo Carvalho de Melo

perf header: Replace hardcoded max cpus by MAX_NR_CPUS

2026-01-28T18:18:11+00:00

cpumask and cpulist from cpu-domain header have hardcoded max_cpus value of 1024. Current systems have more cpus than this value. Replace it with MAX_NR_CPUS. Also define a macro to represent domain name length. Fixes: d40c68a49f69c9bd ("perf header: Support CPU DOMAIN relation info") Reported-by: Shrikanth Hegde Reviewed-by: Shrikanth Hegde Signed-off-by: Swapnil Sapkal Cc: Adrian Hunter Cc: Alexander Shishkin Cc: Anubhav Shelat Cc: Chen Yu Cc: Gautham Shenoy Cc: Ian Rogers Cc: Ingo Molnar Cc: James Clark Cc: Jiri Olsa Cc: Mark Rutland Cc: Namhyung Kim Cc: Peter Zijlstra Cc: Ravi Bangoria Cc: Thomas Falcon Signed-off-by: Arnaldo Carvalho de Melo

perf header: Fix memory leaks in process_cpu_domain_info()

2026-01-23T19:58:38+00:00

do_read_string() returns a string in allocated memory, for some reason there was unused memory allocations and unnecessary strdups. Remove these and make the "perf annotate basic tests" leak sanitizer clean. Fixes: d40c68a49f69c9bd ("perf header: Support CPU DOMAIN relation info") Reviewed-by: James Clark Signed-off-by: Ian Rogers Cc: Aditya Bodkhe Cc: Adrian Hunter Cc: Albert Ou Cc: Alexander Shishkin Cc: Alexandre Ghiti Cc: Athira Rajeev Cc: Bill Wendling Cc: Dr. David Alan Gilbert Cc: Guo Ren Cc: Howard Chu Cc: Ingo Molnar Cc: Jiri Olsa Cc: John Garry Cc: Julia Lawall Cc: Justin Stitt Cc: Krzysztof Łopatowski Cc: Leo Yan Cc: linux-arm-kernel@lists.infradead.org Cc: linux-csky@vger.kernel.org Cc: linux-riscv@lists.infradead.org Cc: Namhyung Kim Cc: Nathan Chancellor Cc: Nick Desaulniers Cc: Palmer Dabbelt Cc: Paul Walmsley Cc: Peter Zijlstra Cc: Sergei Trofimovich Cc: Shimin Guo Cc: Suchit Karunakaran Cc: Swapnil Sapkal Cc: Thomas Falcon Cc: Tianyou Li Cc: Will Deacon Cc: Zecheng Li Signed-off-by: Arnaldo Carvalho de Melo

perf sched stats: Add support for live mode

2026-01-22T15:44:03+00:00

The live mode works similar to simple `perf stat` command, by profiling the target and printing results on the terminal as soon as the target finishes. Example usage: # perf sched stats -- true Description ---------------------------------------------------------------------------------------------------- DESC -> Description of the field COUNT -> Value of the field PCT_CHANGE -> Percent change with corresponding base value AVG_JIFFIES -> Avg time in jiffies between two consecutive occurrence of event ---------------------------------------------------------------------------------------------------- Time elapsed (in jiffies) : 1 ---------------------------------------------------------------------------------------------------- CPU: ---------------------------------------------------------------------------------------------------- DESC COUNT PCT_CHANGE ---------------------------------------------------------------------------------------------------- yld_count : 0 array_exp : 0 sched_count : 0 sched_goidle : 0 ( 0.00% ) ttwu_count : 0 ttwu_local : 0 ( 0.00% ) rq_cpu_time : 27875 run_delay : 0 ( 0.00% ) pcount : 0 ---------------------------------------------------------------------------------------------------- CPU: | DOMAIN: SMT ---------------------------------------------------------------------------------------------------- DESC COUNT AVG_JIFFIES ----------------------------------------- ------------------------------------------ busy_lb_count : 0 $ 0.00 $ busy_lb_balanced : 0 $ 0.00 $ busy_lb_failed : 0 $ 0.00 $ busy_lb_imbalance_load : 0 busy_lb_imbalance_util : 0 busy_lb_imbalance_task : 0 busy_lb_imbalance_misfit : 0 busy_lb_gained : 0 busy_lb_hot_gained : 0 busy_lb_nobusyq : 0 $ 0.00 $ busy_lb_nobusyg : 0 $ 0.00 $ *busy_lb_success_count : 0 *busy_lb_avg_pulled : 0.00 ... and so on. Output will show similar data for all the cpus in the system. Co-developed-by: Ravi Bangoria Signed-off-by: Ravi Bangoria Signed-off-by: Swapnil Sapkal Tested-by: Chen Yu Tested-by: James Clark Acked-by: Ian Rogers Acked-by: Peter Zijlstra Cc: Adrian Hunter Cc: Alexander Shishkin Cc: Andi Kleen Cc: Anubhav Shelat Cc: Ben Gainey Cc: Blake Jones Cc: Chun-Tse Shao Cc: David Vernet Cc: Dmitriy Vyukov Cc: Dr. David Alan Gilbert Cc: Gautham Shenoy Cc: Graham Woodward Cc: Ingo Molnar Cc: James Clark Cc: Jiri Olsa Cc: Juri Lelli Cc: K Prateek Nayak Cc: Kan Liang Cc: Leo Yan Cc: Madadi Vineeth Reddy Cc: Mark Rutland Cc: Namhyung Kim Cc: Sandipan Das Cc: Santosh Shukla Cc: Shrikanth Hegde Cc: Steven Rostedt (VMware) Cc: Tejun Heo Cc: Thomas Falcon Cc: Tim Chen Cc: Vincent Guittot Cc: Yang Jihong Cc: Yujie Liu Cc: Zhongqiu Han [ Avoid potentially using 'sv' uninitialized by calling free_cpu_domain_info() only when build_cpu_domain_map() is called ] Signed-off-by: Arnaldo Carvalho de Melo

perf header: Support CPU DOMAIN relation info

2026-01-22T00:55:09+00:00

The '/proc/schedstat' file gives info about load balancing statistics within a given domain. It also contains the cpu_mask giving information about the sibling cpus and domain names after schedstat version 17. Storing this information in perf header will help tools like `perf sched stats` for better analysis. Signed-off-by: Swapnil Sapkal Tested-by: Chen Yu Acked-by: Ian Rogers Acked-by: Namhyung Kim Acked-by: Peter Zijlstra Cc: Adrian Hunter Cc: Alexander Shishkin Cc: Andi Kleen Cc: Anubhav Shelat Cc: Ben Gainey Cc: Blake Jones Cc: Chun-Tse Shao Cc: David Vernet Cc: Dmitriy Vyukov Cc: Dr. David Alan Gilbert Cc: Gautham Shenoy Cc: Graham Woodward Cc: Ingo Molnar Cc: James Clark Cc: Jiri Olsa Cc: Juri Lelli Cc: K Prateek Nayak Cc: Kan Liang Cc: Leo Yan Cc: Madadi Vineeth Reddy Cc: Mark Rutland Cc: Ravi Bangoria Cc: Sandipan Das Cc: Santosh Shukla Cc: Shrikanth Hegde Cc: Steven Rostedt (VMware) Cc: Tejun Heo Cc: Thomas Falcon Cc: Tim Chen Cc: Vincent Guittot Cc: Yang Jihong Cc: Yujie Liu Cc: Zhongqiu Han Signed-off-by: Arnaldo Carvalho de Melo

Merge tag 'perf-tools-for-v6.19-2025-12-06' of git://git.kernel.org/pub/scm/linux/kernel/git/perf/perf-tools

2025-12-07T15:07:02+00:00

Pull perf tools updates from Namhyung Kim: "Perf event/metric description: Unify all event and metric descriptions in JSON format. Now event parsing and handling is greatly simplified by that. From users point of view, perf list will provide richer information about hardware events like the following. $ perf list hw List of pre-defined events (to be used in -e or -M): legacy hardware: branch-instructions [Retired branch instructions [This event is an alias of branches]. Unit: cpu] branch-misses [Mispredicted branch instructions. Unit: cpu] branches [Retired branch instructions [This event is an alias of branch-instructions]. Unit: cpu] bus-cycles [Bus cycles,which can be different from total cycles. Unit: cpu] cache-misses [Cache misses. Usually this indicates Last Level Cache misses; this is intended to be used in conjunction with the PERF_COUNT_HW_CACHE_REFERENCES event to calculate cache miss rates. Unit: cpu] cache-references [Cache accesses. Usually this indicates Last Level Cache accesses but this may vary depending on your CPU. This may include prefetches and coherency messages; again this depends on the design of your CPU. Unit: cpu] cpu-cycles [Total cycles. Be wary of what happens during CPU frequency scaling [This event is an alias of cycles]. Unit: cpu] cycles [Total cycles. Be wary of what happens during CPU frequency scaling [This event is an alias of cpu-cycles]. Unit: cpu] instructions [Retired instructions. Be careful,these can be affected by various issues,most notably hardware interrupt counts. Unit: cpu] ref-cycles [Total cycles; not affected by CPU frequency scaling. Unit: cpu] But most notable changes would be in the perf stat. On the right side, the default metrics are better named and aligned. :) $ perf stat -- perf test -w noploop Performance counter stats for 'perf test -w noploop': 11 context-switches # 10.8 cs/sec cs_per_second 0 cpu-migrations # 0.0 migrations/sec migrations_per_second 3,612 page-faults # 3532.5 faults/sec page_faults_per_second 1,022.51 msec task-clock # 1.0 CPUs CPUs_utilized 110,466 branch-misses # 0.0 % branch_miss_rate (88.66%) 6,934,452,104 branches # 6781.8 M/sec branch_frequency (88.66%) 4,657,032,590 cpu-cycles # 4.6 GHz cycles_frequency (88.65%) 27,755,874,218 instructions # 6.0 instructions insn_per_cycle (89.03%) TopdownL1 # 0.3 % tma_backend_bound # 9.3 % tma_bad_speculation (89.05%) # 9.7 % tma_frontend_bound (77.86%) # 80.7 % tma_retiring (88.81%) 1.025318171 seconds time elapsed 1.013248000 seconds user 0.012014000 seconds sys Deferred unwinding support: With the kernel support (commit c69993ecdd4d: "perf: Support deferred user unwind"), perf can use deferred callchains for userspace stack trace with frame pointers like below: $ perf record --call-graph fp,defer ... This will be transparent to users when it comes to other commands like perf report and perf script. They will merge the deferred callchains to the previous samples as if they were collected together. ARM SPE updates - Extensive enhancements to support various kinds of memory operations including GCS, MTE allocation tags, memcpy/memset, register access, and SIMD operations. - Add inverted data source filter (inv_data_src_filter) support to exclude certain data sources. - Improve documentation. Vendor event updates: - Intel: Updated event files for Sierra Forest, Panther Lake, Meteor Lake, Lunar Lake, Granite Rapids, and others. - Arm64: Added metrics for i.MX94 DDR PMU and Cortex-A720AE definitions. - RISC-V: Added JSON support for T-HEAD C920V2. Misc: - Improve pointer tracking in data type profiling. It'd give better output when the variable is using container_of() to convert type. - Annotation support for perf c2c report in TUI. Press 'a' key to enter annotation view from cacheline browser window. This will show which instruction is causing the cacheline contention. - Lots of fixes and test coverage improvements!" * tag 'perf-tools-for-v6.19-2025-12-06' of git://git.kernel.org/pub/scm/linux/kernel/git/perf/perf-tools: (214 commits) libperf: Use 'extern' in LIBPERF_API visibility macro perf stat: Improve handling of termination by signal perf tests stat: Add test for error for an offline CPU perf stat: When no events, don't report an error if there is none perf tests stat: Add "--null" coverage perf cpumap: Add "any" CPU handling to cpu_map__snprint_mask libperf cpumap: Fix perf_cpu_map__max for an empty/NULL map perf stat: Allow no events to open if this is a "--null" run perf test kvm: Add some basic perf kvm test coverage perf tests evlist: Add basic evlist test perf tests script dlfilter: Add a dlfilter test perf tests kallsyms: Add basic kallsyms test perf tests timechart: Add a perf timechart test perf tests top: Add basic perf top coverage test perf tests buildid: Add purge and remove testing perf tests c2c: Add a basic c2c perf c2c: Clean up some defensive gets and make asan clean perf jitdump: Fix missed dso__put perf mem-events: Don't leak online CPU map perf hist: In init, ensure mem_info is put on error paths ...

perf header: Switch "cpu" for find_core_pmu in caps feature writing

2025-11-20T01:49:45+00:00

Writing currently fails on non-x86 and hybrid CPUs. Switch to the more regular find_core_pmu that is normally used in this case. Tested on hybrid alderlake system. Signed-off-by: Ian Rogers Signed-off-by: Namhyung Kim

perf header: Write bpf_prog (infos|btfs)_cnt to data file

2025-11-13T20:16:33+00:00

With commit f0d0f978f3f5830a ("perf header: Don't write empty BPF/BTF info"), the write_bpf_( prog_info() | btf() ) functions exit without writing anything if env->bpf_prog.(infos| btfs)_cnt is zero. process_bpf_( prog_info() | btf() ), however, still expect a "count" value to exist in the data file. If btf information is empty, for example, process_bpf_btf will read garbage or some other data as the number of btf nodes in the data file. As a result, the data file will not be processed correctly. Instead, write the count to the data file and exit if it is zero. Fixes: f0d0f978f3f5830a ("perf header: Don't write empty BPF/BTF info") Reviewed-by: Ian Rogers Signed-off-by: Thomas Falcon Acked-by: Namhyung Kim Cc: Adrian Hunter Cc: Alexander Shishkin Cc: Ingo Molnar Cc: Jiri Olsa Cc: Mark Rutland Cc: Peter Zijlstra Signed-off-by: Arnaldo Carvalho de Melo

perf tool: Add the perf_tool argument to all callbacks

2025-11-07T21:25:05+00:00

Getting context for what a tool is doing, such as the perf_inject instance, using container_of the tool is a common pattern in the code. This isn't possible event_op2, event_op3 and event_op4 callbacks as the tool isn't passed. Add the argument and then fix function signatures to match. As tools maybe reading a tool from somewhere else, change that code to use the passed in tool. Signed-off-by: Ian Rogers Signed-off-by: Namhyung Kim

perf record: Make sure to update build-ID cache

2025-11-07T01:29:16+00:00

Recent change on enabling --buildid-mmap by default brought an issue with build-id handling. With build-ID in MMAP2 records, we don't need to save the build-ID table in the header of a perf data file. But the actual file contents still need to be cached in the debug directory for annotation etc. Split the build-ID header processing and caching and make sure perf record to save hit DSOs in the build-ID cache by moving perf_session__cache_build_ids() to the end of the record__ finish_output(). Reviewed-by: Ian Rogers Signed-off-by: Namhyung Kim