<feed xmlns='http://www.w3.org/2005/Atom'>
<title>linux-toradex.git/tools/perf, branch master</title>
<subtitle>Linux kernel for Apalis and Colibri modules</subtitle>
<link rel='alternate' type='text/html' href='https://git.toradex.cn/cgit/linux-toradex.git/'/>
<entry>
<title>Merge tag 'perf-tools-for-v7.4-1-2026-10-30' of git://git.kernel.org/pub/scm/linux/kernel/git/perf/perf-tools-next</title>
<updated>2026-10-31T04:54:38+00:00</updated>
<author>
<name>Linus Torvalds</name>
<email>torvalds@linux-foundation.org</email>
</author>
<published>2026-10-31T04:54:38+00:00</published>
<link rel='alternate' type='text/html' href='https://git.toradex.cn/cgit/linux-toradex.git/commit/?id=dd384658c738d15783a786d3af2851e02490f171'/>
<id>dd384658c738d15783a786d3af2851e02490f171</id>
<content type='text'>
Pull perf tool updates from Arnaldo Carvalho de Melo:

 - Support instruction tracking on arm64 for data type profiling,
   resolving stack, global, per-CPU and 'current' variable accesses, so
   that 'perf annotate --data-type' resolves types on arm64 the way it
   does on x86.

 - Port the Python scripts to the perf module and standalone execution,
   and improve the module: lazy sample callchains and event copies, GIL
   handling, better session/thread lookup, new treport, flamegraph and
   sqlite/postgres exporters.

 - Reenable and fix the attribute expectation test suite: per-PMU bit
   masks, other core PMUs' events, dummy tracking events, group checks
   and spelled out skip reasons.

 - perf trace: filter the target's tasks in BPF, open the BPF output
   event on every CPU, add an option to disable syscall augmentation and
   raise TRACE_AUG_MAX_BUF.

 - perf jitdump: harden record parsing, validating unwinding sizes and
   symbol names, byte-swapping debug entries via safe accessors and
   fixing frees on error paths.

 - perf annotate-data fixes: bound the member nesting recursion, resolve
   type DIEs in the debug file they came from, adjust the offset for
   flex-arrays and show the sample count in the data-type browser.

 - perf symbol and debug info: try linked debug files for DSOs, avoid
   repeated failing searches, set binary/symtab types for split
   kallsyms, JIT maps and the build-id cache, fix libbfd symbol loading.

 - perf config: serialize config file access with a mutex, add the
   DEFINE_MUTEX() and DO_ONCE() primitives and a robust process shared
   mutex, add the core.hybrid-merge option.

 - perf report: add --progress and --no-progress options.

 - perf mem: add a false sharing workload to 'perf test', request
   PERF_SAMPLE_CPU by default, use the IBS swfilt filter when available
   and decode L0 in the memory level.

 - perf test: take turns using the perf mlock budget in the mmaping
   tests, add the 'needs_mlock' suite flag, scope the probe based tests
   to their PID and make more tests non-exclusive.

 - perf vendor events intel: update events and metrics to TMA 5.2,
   refresh the event tables and fix umask encodings.

 - perf build: run shellcheck, mypy and pylint in parallel, make pylint
   opt-out with NO_PYLINT=1, fix clean and install dependencies.

 - Add a script and a 'make sync-headers' target to sync the tools/
   copies of kernel headers with the kernel sources, plus the resulting
   syncs of the cpufeatures, KVM, fs, mount, fcntl, socket, rtnetlink,
   if_addr and sound headers.

 - Fix cs-etm HW_ID mapping synthesis for raw trace, queue teardown
   leaks and duplicate v0 global mappings.

 - Add a 'perf bench atomic' benchmark for atomic operations.

 - Filter events before creating lock_stat in 'perf lock contention'.

 - Add a 'perf test' case for events re-armed with
   PERF_EVENT_IOC_REFRESH.

 - Fix memory leaks on 'perf c2c' output fields and add function view
   stdio coverage to its test.

 - Fix a NULL dereference resolving callchains in 'perf machine'.

 - Optimize the libdw fallback when DWARF is missing and try the symbol
   file first when checking for debuginfo.

 - Don't call eprintf() in pr_debugN() when not verbose.

 - Don't delete cumulative caller entries before the stat_acc decays in
   'perf report'.

 - Fix ring buffer cursor handling: keep the consumed cursor in forward
   mode and use the head selected by read_init.

 - Fix 'perf test mmap-basic' user rdpmc detection and only test
   userspace reading on supported platforms.

 - Build fixes: include sys/types.h for ssize_t, errno.h before kernel
   headers and fix jevents.py type annotations for mypy 2.x.

* tag 'perf-tools-for-v7.4-1-2026-10-30' of git://git.kernel.org/pub/scm/linux/kernel/git/perf/perf-tools-next: (425 commits)
  perf machine: Fix NULL deref in thread__resolve_callchain_sample()
  perf pmu-events: Fix jevents.py type annotations for mypy 2.x
  perf parse-events: Include errno.h before kernel headers
  perf print_insn: Include sys/types.h for ssize_t
  perf tests mmap-basic: Fix user rdpmc detection logic
  perf tests mmap-basic: Only test userspace reading on supported platforms
  perf mmap: Preserve the consumed cursor in forward mode
  perf mmap: Use the head selected by read_init
  perf hist: Do not delete cumulative caller entries before stat_acc decays
  perf debug: Avoid calling eprintf() in pr_debugN() when verbose &lt; n
  perf libdw: Optimize fallback logic for missing DWARF
  tools headers: Sync KVM headers with the kernel sources
  tools headers: Sync UAPI sound/asound.h with the kernel sources
  tools headers: Sync UAPI linux/mount.h with the kernel sources
  tools headers: Sync UAPI linux/fs.h with the kernel sources
  tools headers: Sync UAPI linux/fcntl.h with the kernel sources
  tools headers: Sync linux/socket.h with the kernel sources
  tools arch x86: Sync the cpufeatures.h copy with the kernel sources
  tools headers UAPI: Sync linux/rtnetlink.h with the kernel sources
  tools headers UAPI: Sync linux/if_addr.h with the kernel sources
  ...
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
Pull perf tool updates from Arnaldo Carvalho de Melo:

 - Support instruction tracking on arm64 for data type profiling,
   resolving stack, global, per-CPU and 'current' variable accesses, so
   that 'perf annotate --data-type' resolves types on arm64 the way it
   does on x86.

 - Port the Python scripts to the perf module and standalone execution,
   and improve the module: lazy sample callchains and event copies, GIL
   handling, better session/thread lookup, new treport, flamegraph and
   sqlite/postgres exporters.

 - Reenable and fix the attribute expectation test suite: per-PMU bit
   masks, other core PMUs' events, dummy tracking events, group checks
   and spelled out skip reasons.

 - perf trace: filter the target's tasks in BPF, open the BPF output
   event on every CPU, add an option to disable syscall augmentation and
   raise TRACE_AUG_MAX_BUF.

 - perf jitdump: harden record parsing, validating unwinding sizes and
   symbol names, byte-swapping debug entries via safe accessors and
   fixing frees on error paths.

 - perf annotate-data fixes: bound the member nesting recursion, resolve
   type DIEs in the debug file they came from, adjust the offset for
   flex-arrays and show the sample count in the data-type browser.

 - perf symbol and debug info: try linked debug files for DSOs, avoid
   repeated failing searches, set binary/symtab types for split
   kallsyms, JIT maps and the build-id cache, fix libbfd symbol loading.

 - perf config: serialize config file access with a mutex, add the
   DEFINE_MUTEX() and DO_ONCE() primitives and a robust process shared
   mutex, add the core.hybrid-merge option.

 - perf report: add --progress and --no-progress options.

 - perf mem: add a false sharing workload to 'perf test', request
   PERF_SAMPLE_CPU by default, use the IBS swfilt filter when available
   and decode L0 in the memory level.

 - perf test: take turns using the perf mlock budget in the mmaping
   tests, add the 'needs_mlock' suite flag, scope the probe based tests
   to their PID and make more tests non-exclusive.

 - perf vendor events intel: update events and metrics to TMA 5.2,
   refresh the event tables and fix umask encodings.

 - perf build: run shellcheck, mypy and pylint in parallel, make pylint
   opt-out with NO_PYLINT=1, fix clean and install dependencies.

 - Add a script and a 'make sync-headers' target to sync the tools/
   copies of kernel headers with the kernel sources, plus the resulting
   syncs of the cpufeatures, KVM, fs, mount, fcntl, socket, rtnetlink,
   if_addr and sound headers.

 - Fix cs-etm HW_ID mapping synthesis for raw trace, queue teardown
   leaks and duplicate v0 global mappings.

 - Add a 'perf bench atomic' benchmark for atomic operations.

 - Filter events before creating lock_stat in 'perf lock contention'.

 - Add a 'perf test' case for events re-armed with
   PERF_EVENT_IOC_REFRESH.

 - Fix memory leaks on 'perf c2c' output fields and add function view
   stdio coverage to its test.

 - Fix a NULL dereference resolving callchains in 'perf machine'.

 - Optimize the libdw fallback when DWARF is missing and try the symbol
   file first when checking for debuginfo.

 - Don't call eprintf() in pr_debugN() when not verbose.

 - Don't delete cumulative caller entries before the stat_acc decays in
   'perf report'.

 - Fix ring buffer cursor handling: keep the consumed cursor in forward
   mode and use the head selected by read_init.

 - Fix 'perf test mmap-basic' user rdpmc detection and only test
   userspace reading on supported platforms.

 - Build fixes: include sys/types.h for ssize_t, errno.h before kernel
   headers and fix jevents.py type annotations for mypy 2.x.

* tag 'perf-tools-for-v7.4-1-2026-10-30' of git://git.kernel.org/pub/scm/linux/kernel/git/perf/perf-tools-next: (425 commits)
  perf machine: Fix NULL deref in thread__resolve_callchain_sample()
  perf pmu-events: Fix jevents.py type annotations for mypy 2.x
  perf parse-events: Include errno.h before kernel headers
  perf print_insn: Include sys/types.h for ssize_t
  perf tests mmap-basic: Fix user rdpmc detection logic
  perf tests mmap-basic: Only test userspace reading on supported platforms
  perf mmap: Preserve the consumed cursor in forward mode
  perf mmap: Use the head selected by read_init
  perf hist: Do not delete cumulative caller entries before stat_acc decays
  perf debug: Avoid calling eprintf() in pr_debugN() when verbose &lt; n
  perf libdw: Optimize fallback logic for missing DWARF
  tools headers: Sync KVM headers with the kernel sources
  tools headers: Sync UAPI sound/asound.h with the kernel sources
  tools headers: Sync UAPI linux/mount.h with the kernel sources
  tools headers: Sync UAPI linux/fs.h with the kernel sources
  tools headers: Sync UAPI linux/fcntl.h with the kernel sources
  tools headers: Sync linux/socket.h with the kernel sources
  tools arch x86: Sync the cpufeatures.h copy with the kernel sources
  tools headers UAPI: Sync linux/rtnetlink.h with the kernel sources
  tools headers UAPI: Sync linux/if_addr.h with the kernel sources
  ...
</pre>
</div>
</content>
</entry>
<entry>
<title>perf machine: Fix NULL deref in thread__resolve_callchain_sample()</title>
<updated>2026-10-27T20:45:34+00:00</updated>
<author>
<name>lvyating</name>
<email>lv.yating@zte.com.cn</email>
</author>
<published>2026-10-27T11:47:32+00:00</published>
<link rel='alternate' type='text/html' href='https://git.toradex.cn/cgit/linux-toradex.git/commit/?id=4626d370ae89b8af4d46a45372b269f926ec662d'/>
<id>4626d370ae89b8af4d46a45372b269f926ec662d</id>
<content type='text'>
When recording with -j stack, resolve_lbr_callchain_sample() can
dereference a NULL 'chain' pointer while resolving the branch
stack.

$ perf record -j stack -e branches:u uname
perf: Segmentation fault

Fix it by checking 'chain' for NULL before dereferencing it.

Fixes: 29be60c93d2d9300 ("perf build-id: Mark DSO in sample callchains")
Reviewed-by: Ian Rogers &lt;irogers@google.com&gt;
Reviewed-by: Shengming Hu &lt;hu.shengming@zte.com.cn&gt;
Signed-off-by: lvyating &lt;lv.yating@zte.com.cn&gt;
Tested-by: Ian Rogers &lt;irogers@google.com&gt;
Cc: Adrian Hunter &lt;adrian.hunter@intel.com&gt;
Cc: Alexander Shishkin &lt;alexander.shishkin@linux.intel.com&gt;
Cc: Dai Hualiang &lt;dai.hualiang@zte.com.cn&gt;
Cc: he.yilin@zte.com.cn
Cc: Ian Rogers &lt;irogers@google.com&gt;
Cc: Ingo Molnar &lt;mingo@redhat.com&gt;
Cc: James Clark &lt;james.clark@linaro.org&gt;
Cc: Jiri Olsa &lt;jolsa@kernel.org&gt;
Cc: Mark Rutland &lt;mark.rutland@arm.com&gt;
Cc: Namhyung Kim &lt;namhyung@kernel.org&gt;
Cc: Peter Zijlstra &lt;peterz@infradead.org&gt;
Cc: Qingtao Liu &lt;liu.qingtao2@zte.com.cn&gt;
Cc: Shen Lin &lt;shen.lin1@zte.com.cn&gt;
Cc: stable@vger.kernel.org
Cc: ye.zhihao1@zte.com.cn
Cc: zhang.cheng35@zte.com.cn
Signed-off-by: Arnaldo Carvalho de Melo &lt;acme@redhat.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
When recording with -j stack, resolve_lbr_callchain_sample() can
dereference a NULL 'chain' pointer while resolving the branch
stack.

$ perf record -j stack -e branches:u uname
perf: Segmentation fault

Fix it by checking 'chain' for NULL before dereferencing it.

Fixes: 29be60c93d2d9300 ("perf build-id: Mark DSO in sample callchains")
Reviewed-by: Ian Rogers &lt;irogers@google.com&gt;
Reviewed-by: Shengming Hu &lt;hu.shengming@zte.com.cn&gt;
Signed-off-by: lvyating &lt;lv.yating@zte.com.cn&gt;
Tested-by: Ian Rogers &lt;irogers@google.com&gt;
Cc: Adrian Hunter &lt;adrian.hunter@intel.com&gt;
Cc: Alexander Shishkin &lt;alexander.shishkin@linux.intel.com&gt;
Cc: Dai Hualiang &lt;dai.hualiang@zte.com.cn&gt;
Cc: he.yilin@zte.com.cn
Cc: Ian Rogers &lt;irogers@google.com&gt;
Cc: Ingo Molnar &lt;mingo@redhat.com&gt;
Cc: James Clark &lt;james.clark@linaro.org&gt;
Cc: Jiri Olsa &lt;jolsa@kernel.org&gt;
Cc: Mark Rutland &lt;mark.rutland@arm.com&gt;
Cc: Namhyung Kim &lt;namhyung@kernel.org&gt;
Cc: Peter Zijlstra &lt;peterz@infradead.org&gt;
Cc: Qingtao Liu &lt;liu.qingtao2@zte.com.cn&gt;
Cc: Shen Lin &lt;shen.lin1@zte.com.cn&gt;
Cc: stable@vger.kernel.org
Cc: ye.zhihao1@zte.com.cn
Cc: zhang.cheng35@zte.com.cn
Signed-off-by: Arnaldo Carvalho de Melo &lt;acme@redhat.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>perf pmu-events: Fix jevents.py type annotations for mypy 2.x</title>
<updated>2026-10-27T20:40:37+00:00</updated>
<author>
<name>Lalit Maganti</name>
<email>lalitmaganti@gmail.com</email>
</author>
<published>2026-10-18T06:53:45+00:00</published>
<link rel='alternate' type='text/html' href='https://git.toradex.cn/cgit/linux-toradex.git/commit/?id=aa0fec5853eea1eca4f0bd9fd26e0a9d25e1c109'/>
<id>aa0fec5853eea1eca4f0bd9fd26e0a9d25e1c109</id>
<content type='text'>
Since the build runs mypy whenever it is installed, the perf build fails
with mypy 2.4 on 7 errors in jevents.py, e.g.:

  pmu-events/jevents.py:542: error: Item "None" of "str | None" has no
  attribute "replace"  [union-attr]
  pmu-events/jevents.py:563: error: Item "None" of "BigCString | None"
  has no attribute "offsets"  [union-attr]

Both come from wrong annotations rather than from code that can see
None:
 - unit_to_pmu() is annotated as returning Optional[str] but always
   returns a string, so JsonEvent.pmu is typed as optional where it is
   used as a str.
 - _bcs is set to None before the BigCString class is defined and only
   assigned an instance after it, so mypy types it as BigCString | None.

Correct the return type and declare _bcs with its type instead of the
None placeholder. The generated pmu-events.c and pmu-events-string.c
are unchanged, and mypy 1.9 still passes.

Fixes: b3e9c396d0acab27 ("perf build: Make mypy build test opt-out (NO_MYPY=1)")
Signed-off-by: Lalit Maganti &lt;lalitmaganti@gmail.com&gt;
Cc: Adrian Hunter &lt;adrian.hunter@intel.com&gt;
Cc: Ian Rogers &lt;irogers@google.com&gt;
Cc: James Clark &lt;james.clark@linaro.org&gt;
Cc: Namhyung Kim &lt;namhyung@kernel.org&gt;
Assisted-by: LLM
Signed-off-by: Arnaldo Carvalho de Melo &lt;acme@redhat.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
Since the build runs mypy whenever it is installed, the perf build fails
with mypy 2.4 on 7 errors in jevents.py, e.g.:

  pmu-events/jevents.py:542: error: Item "None" of "str | None" has no
  attribute "replace"  [union-attr]
  pmu-events/jevents.py:563: error: Item "None" of "BigCString | None"
  has no attribute "offsets"  [union-attr]

Both come from wrong annotations rather than from code that can see
None:
 - unit_to_pmu() is annotated as returning Optional[str] but always
   returns a string, so JsonEvent.pmu is typed as optional where it is
   used as a str.
 - _bcs is set to None before the BigCString class is defined and only
   assigned an instance after it, so mypy types it as BigCString | None.

Correct the return type and declare _bcs with its type instead of the
None placeholder. The generated pmu-events.c and pmu-events-string.c
are unchanged, and mypy 1.9 still passes.

Fixes: b3e9c396d0acab27 ("perf build: Make mypy build test opt-out (NO_MYPY=1)")
Signed-off-by: Lalit Maganti &lt;lalitmaganti@gmail.com&gt;
Cc: Adrian Hunter &lt;adrian.hunter@intel.com&gt;
Cc: Ian Rogers &lt;irogers@google.com&gt;
Cc: James Clark &lt;james.clark@linaro.org&gt;
Cc: Namhyung Kim &lt;namhyung@kernel.org&gt;
Assisted-by: LLM
Signed-off-by: Arnaldo Carvalho de Melo &lt;acme@redhat.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>perf parse-events: Include errno.h before kernel headers</title>
<updated>2026-10-27T20:40:37+00:00</updated>
<author>
<name>Florian Fainelli</name>
<email>florian.fainelli@broadcom.com</email>
</author>
<published>2026-10-26T17:48:10+00:00</published>
<link rel='alternate' type='text/html' href='https://git.toradex.cn/cgit/linux-toradex.git/commit/?id=022b0ac5d3e87f87cb7ba42d04d0bc3ddfdd6be4'/>
<id>022b0ac5d3e87f87cb7ba42d04d0bc3ddfdd6be4</id>
<content type='text'>
Building perf against uClibc fails in config_term_common():

  util/parse-events.c:954:30: error: 'ERANGE' undeclared

The kernel headers included before errno.h pull in asm/errno.h. uClibc's
bits/errno.h undefines EDOM, EILSEQ and ERANGE before including
linux/errno.h, but the kernel header include guards prevent these macros
from being defined again.

Include errno.h before the kernel headers so ERANGE remains defined.

Fixes: 6b9c0261b3fb298b ("perf record: Add ratio-to-prev term")
Reviewed-by: James Clark &lt;james.clark@linaro.org&gt;
Signed-off-by: Florian Fainelli &lt;florian.fainelli@broadcom.com&gt;
Cc: Adrian Hunter &lt;adrian.hunter@intel.com&gt;
Cc: Alexander Shishkin &lt;alexander.shishkin@linux.intel.com&gt;
Cc: Faisal Bukhari &lt;faisalbukhari523@gmail.com&gt;
Cc: Ian Rogers &lt;irogers@google.com&gt;
Cc: Ingo Molnar &lt;mingo@redhat.com&gt;
Cc: Jiri Olsa &lt;jolsa@kernel.org&gt;
Cc: Mark Rutland &lt;mark.rutland@arm.com&gt;
Cc: Namhyung Kim &lt;namhyung@kernel.org&gt;
Cc: Peter Zijlstra &lt;peterz@infradead.org&gt;
Cc: Qinxin Xia &lt;xiaqinxin@huawei.com&gt;
Cc: Thomas Falcon &lt;thomas.falcon@intel.com&gt;
Cc: Thomas Richter &lt;tmricht@linux.ibm.com&gt;
Signed-off-by: Arnaldo Carvalho de Melo &lt;acme@redhat.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
Building perf against uClibc fails in config_term_common():

  util/parse-events.c:954:30: error: 'ERANGE' undeclared

The kernel headers included before errno.h pull in asm/errno.h. uClibc's
bits/errno.h undefines EDOM, EILSEQ and ERANGE before including
linux/errno.h, but the kernel header include guards prevent these macros
from being defined again.

Include errno.h before the kernel headers so ERANGE remains defined.

Fixes: 6b9c0261b3fb298b ("perf record: Add ratio-to-prev term")
Reviewed-by: James Clark &lt;james.clark@linaro.org&gt;
Signed-off-by: Florian Fainelli &lt;florian.fainelli@broadcom.com&gt;
Cc: Adrian Hunter &lt;adrian.hunter@intel.com&gt;
Cc: Alexander Shishkin &lt;alexander.shishkin@linux.intel.com&gt;
Cc: Faisal Bukhari &lt;faisalbukhari523@gmail.com&gt;
Cc: Ian Rogers &lt;irogers@google.com&gt;
Cc: Ingo Molnar &lt;mingo@redhat.com&gt;
Cc: Jiri Olsa &lt;jolsa@kernel.org&gt;
Cc: Mark Rutland &lt;mark.rutland@arm.com&gt;
Cc: Namhyung Kim &lt;namhyung@kernel.org&gt;
Cc: Peter Zijlstra &lt;peterz@infradead.org&gt;
Cc: Qinxin Xia &lt;xiaqinxin@huawei.com&gt;
Cc: Thomas Falcon &lt;thomas.falcon@intel.com&gt;
Cc: Thomas Richter &lt;tmricht@linux.ibm.com&gt;
Signed-off-by: Arnaldo Carvalho de Melo &lt;acme@redhat.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>perf print_insn: Include sys/types.h for ssize_t</title>
<updated>2026-10-27T20:40:37+00:00</updated>
<author>
<name>Florian Fainelli</name>
<email>florian.fainelli@broadcom.com</email>
</author>
<published>2026-10-26T17:45:31+00:00</published>
<link rel='alternate' type='text/html' href='https://git.toradex.cn/cgit/linux-toradex.git/commit/?id=e0e4470aa53c0022418290fb551c21da4bd50209'/>
<id>e0e4470aa53c0022418290fb551c21da4bd50209</id>
<content type='text'>
print_insn.h declares fprintf_insn_asm() with an ssize_t return type but
does not include the header defining that type. This causes builds where
stdio.h does not provide ssize_t to fail when compiling util/print_insn.c:

  util/print_insn.h:21:1: error: unknown type name 'ssize_t'

Include sys/types.h so the header provides its own dependency for ssize_t.

Fixes: 218c200f677d8af4 ("perf script: Consolidate capstone print functions")
Reviewed-by: James Clark &lt;james.clark@linaro.org&gt;
Signed-off-by: Florian Fainelli &lt;florian.fainelli@broadcom.com&gt;
Cc: Adrian Hunter &lt;adrian.hunter@intel.com&gt;
Cc: Alexander Shishkin &lt;alexander.shishkin@linux.intel.com&gt;
Cc: Andi Kleen &lt;ak@linux.intel.com&gt;
Cc: Ian Rogers &lt;irogers@google.com&gt;
Cc: Ingo Molnar &lt;mingo@redhat.com&gt;
Cc: Jiri Olsa &lt;jolsa@kernel.org&gt;
Cc: Mark Rutland &lt;mark.rutland@arm.com&gt;
Cc: Namhyung Kim &lt;namhyung@kernel.org&gt;
Cc: Peter Zijlstra &lt;peterz@infradead.org&gt;
Signed-off-by: Arnaldo Carvalho de Melo &lt;acme@redhat.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
print_insn.h declares fprintf_insn_asm() with an ssize_t return type but
does not include the header defining that type. This causes builds where
stdio.h does not provide ssize_t to fail when compiling util/print_insn.c:

  util/print_insn.h:21:1: error: unknown type name 'ssize_t'

Include sys/types.h so the header provides its own dependency for ssize_t.

Fixes: 218c200f677d8af4 ("perf script: Consolidate capstone print functions")
Reviewed-by: James Clark &lt;james.clark@linaro.org&gt;
Signed-off-by: Florian Fainelli &lt;florian.fainelli@broadcom.com&gt;
Cc: Adrian Hunter &lt;adrian.hunter@intel.com&gt;
Cc: Alexander Shishkin &lt;alexander.shishkin@linux.intel.com&gt;
Cc: Andi Kleen &lt;ak@linux.intel.com&gt;
Cc: Ian Rogers &lt;irogers@google.com&gt;
Cc: Ingo Molnar &lt;mingo@redhat.com&gt;
Cc: Jiri Olsa &lt;jolsa@kernel.org&gt;
Cc: Mark Rutland &lt;mark.rutland@arm.com&gt;
Cc: Namhyung Kim &lt;namhyung@kernel.org&gt;
Cc: Peter Zijlstra &lt;peterz@infradead.org&gt;
Signed-off-by: Arnaldo Carvalho de Melo &lt;acme@redhat.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>perf tests mmap-basic: Fix user rdpmc detection logic</title>
<updated>2026-10-27T20:40:24+00:00</updated>
<author>
<name>James Clark</name>
<email>james.clark@linaro.org</email>
</author>
<published>2026-10-07T10:52:13+00:00</published>
<link rel='alternate' type='text/html' href='https://git.toradex.cn/cgit/linux-toradex.git/commit/?id=fe7b73c293f1a5b0b7d8a4748b4c076fa7db852c'/>
<id>fe7b73c293f1a5b0b7d8a4748b4c076fa7db852c</id>
<content type='text'>
RISC-V and Arm control userspace counter access through
/proc/sys/kernel/perf_user_access. Add that as a fallback to
set_user_read() so the test can exercise both enabled and disabled
states on those platforms.

Document disabled, enabled and legacy states in user_read_state and fail
on values outside the known 0-2 range. Preserve and restore any value in
this range, rather than just 0 or 1.

Skip when the control files exist but can't be written to
(USER_READ_ERR_PERM). We can't rely on always being able to write to
them when an exclude_kernel=0 event can be opened because opening events
might succeed for non-root users when perf_event_paranoid=-1.

This test isn't run on unsupported platforms since the previous commit,
so we can simplify the following things:

 * Test that cap_user_rdpmc is always equal to the requested state
   rather than checking it only when disabled. As long as we stop
   setting the cap in the attr unconditionally on Arm, this is ok.

 * Remove the USER_READ_UNKNOWN/rdpmc_supported fallback in the checks.
   Now the expected state is always the one requested. USER_READ_UNKNOWN
   only controls whether to skip restoration of the state on unsupported
   platforms, so call it USER_READ_ERR_NO_FILE.

Signed-off-by: James Clark &lt;james.clark@linaro.org&gt;
Cc: Adrian Hunter &lt;adrian.hunter@intel.com&gt;
Cc: Albert Ou &lt;aou@eecs.berkeley.edu&gt;
Cc: Alexander Shishkin &lt;alexander.shishkin@linux.intel.com&gt;
Cc: Alexandre Ghiti &lt;alex@ghiti.fr&gt;
Cc: Ian Rogers &lt;irogers@google.com&gt;
Cc: Ingo Molnar &lt;mingo@redhat.com&gt;
Cc: Jiri Olsa &lt;jolsa@kernel.org&gt;
Cc: Leo Yan &lt;leo.yan@arm.com&gt;
Cc: Mark Rutland &lt;mark.rutland@arm.com&gt;
Cc: Michael Petlan &lt;mpetlan@redhat.com&gt;
Cc: Namhyung Kim &lt;namhyung@kernel.org&gt;
Cc: Palmer Dabbelt &lt;palmer@dabbelt.com&gt;
Cc: Paul Walmsley &lt;pjw@kernel.org&gt;
Cc: Peter Zijlstra &lt;peterz@infradead.org&gt;
Assisted-by: Codex:GPT-6.1-Sol
[Re-write: [Re-write to fix bugs in set_user_read() and simplify tests]
Signed-off-by: Qiao Zhao &lt;qzhao@redhat.com&gt;
Signed-off-by: Arnaldo Carvalho de Melo &lt;acme@redhat.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
RISC-V and Arm control userspace counter access through
/proc/sys/kernel/perf_user_access. Add that as a fallback to
set_user_read() so the test can exercise both enabled and disabled
states on those platforms.

Document disabled, enabled and legacy states in user_read_state and fail
on values outside the known 0-2 range. Preserve and restore any value in
this range, rather than just 0 or 1.

Skip when the control files exist but can't be written to
(USER_READ_ERR_PERM). We can't rely on always being able to write to
them when an exclude_kernel=0 event can be opened because opening events
might succeed for non-root users when perf_event_paranoid=-1.

This test isn't run on unsupported platforms since the previous commit,
so we can simplify the following things:

 * Test that cap_user_rdpmc is always equal to the requested state
   rather than checking it only when disabled. As long as we stop
   setting the cap in the attr unconditionally on Arm, this is ok.

 * Remove the USER_READ_UNKNOWN/rdpmc_supported fallback in the checks.
   Now the expected state is always the one requested. USER_READ_UNKNOWN
   only controls whether to skip restoration of the state on unsupported
   platforms, so call it USER_READ_ERR_NO_FILE.

Signed-off-by: James Clark &lt;james.clark@linaro.org&gt;
Cc: Adrian Hunter &lt;adrian.hunter@intel.com&gt;
Cc: Albert Ou &lt;aou@eecs.berkeley.edu&gt;
Cc: Alexander Shishkin &lt;alexander.shishkin@linux.intel.com&gt;
Cc: Alexandre Ghiti &lt;alex@ghiti.fr&gt;
Cc: Ian Rogers &lt;irogers@google.com&gt;
Cc: Ingo Molnar &lt;mingo@redhat.com&gt;
Cc: Jiri Olsa &lt;jolsa@kernel.org&gt;
Cc: Leo Yan &lt;leo.yan@arm.com&gt;
Cc: Mark Rutland &lt;mark.rutland@arm.com&gt;
Cc: Michael Petlan &lt;mpetlan@redhat.com&gt;
Cc: Namhyung Kim &lt;namhyung@kernel.org&gt;
Cc: Palmer Dabbelt &lt;palmer@dabbelt.com&gt;
Cc: Paul Walmsley &lt;pjw@kernel.org&gt;
Cc: Peter Zijlstra &lt;peterz@infradead.org&gt;
Assisted-by: Codex:GPT-6.1-Sol
[Re-write: [Re-write to fix bugs in set_user_read() and simplify tests]
Signed-off-by: Qiao Zhao &lt;qzhao@redhat.com&gt;
Signed-off-by: Arnaldo Carvalho de Melo &lt;acme@redhat.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>perf tests mmap-basic: Only test userspace reading on supported platforms</title>
<updated>2026-10-27T20:03:20+00:00</updated>
<author>
<name>James Clark</name>
<email>james.clark@linaro.org</email>
</author>
<published>2026-10-07T10:52:12+00:00</published>
<link rel='alternate' type='text/html' href='https://git.toradex.cn/cgit/linux-toradex.git/commit/?id=e793b6a4be3c7810c92eeeb21bfd0f03c22fcbcb'/>
<id>e793b6a4be3c7810c92eeeb21bfd0f03c22fcbcb</id>
<content type='text'>
mmap_user_read_instr() and mmap_user_read_instr_disabled() are the same
test for platforms that libperf doesn't support userspace counter
reading for. Only run the 'enabled' test for platforms that do support
it in order to simplify it in the next commit and not run the same test
twice.

The test can then assert stronger guarantees and have fewer edge cases
to handle as we know it only runs in places where 'rdmpc_supported' will
be true.

"unsupported" hasn't been a relevant skip reason since
commit 588d22b40480 ("perf test: Expand user space event reading (rdpmc)
tests") added the UNKNOWN fallback for unsupported platforms. Remove it
and leave only "permissions".

Signed-off-by: James Clark &lt;james.clark@linaro.org&gt;
Cc: Adrian Hunter &lt;adrian.hunter@intel.com&gt;
Cc: Albert Ou &lt;aou@eecs.berkeley.edu&gt;
Cc: Alexander Shishkin &lt;alexander.shishkin@linux.intel.com&gt;
Cc: Alexandre Ghiti &lt;alex@ghiti.fr&gt;
Cc: Ian Rogers &lt;irogers@google.com&gt;
Cc: Ingo Molnar &lt;mingo@redhat.com&gt;
Cc: Jiri Olsa &lt;jolsa@kernel.org&gt;
Cc: Leo Yan &lt;leo.yan@arm.com&gt;
Cc: Mark Rutland &lt;mark.rutland@arm.com&gt;
Cc: Michael Petlan &lt;mpetlan@redhat.com&gt;
Cc: Namhyung Kim &lt;namhyung@kernel.org&gt;
Cc: Palmer Dabbelt &lt;palmer@dabbelt.com&gt;
Cc: Paul Walmsley &lt;pjw@kernel.org&gt;
Cc: Peter Zijlstra &lt;peterz@infradead.org&gt;
Cc: Qiao Zhao &lt;qzhao@redhat.com&gt;
Signed-off-by: Arnaldo Carvalho de Melo &lt;acme@redhat.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
mmap_user_read_instr() and mmap_user_read_instr_disabled() are the same
test for platforms that libperf doesn't support userspace counter
reading for. Only run the 'enabled' test for platforms that do support
it in order to simplify it in the next commit and not run the same test
twice.

The test can then assert stronger guarantees and have fewer edge cases
to handle as we know it only runs in places where 'rdmpc_supported' will
be true.

"unsupported" hasn't been a relevant skip reason since
commit 588d22b40480 ("perf test: Expand user space event reading (rdpmc)
tests") added the UNKNOWN fallback for unsupported platforms. Remove it
and leave only "permissions".

Signed-off-by: James Clark &lt;james.clark@linaro.org&gt;
Cc: Adrian Hunter &lt;adrian.hunter@intel.com&gt;
Cc: Albert Ou &lt;aou@eecs.berkeley.edu&gt;
Cc: Alexander Shishkin &lt;alexander.shishkin@linux.intel.com&gt;
Cc: Alexandre Ghiti &lt;alex@ghiti.fr&gt;
Cc: Ian Rogers &lt;irogers@google.com&gt;
Cc: Ingo Molnar &lt;mingo@redhat.com&gt;
Cc: Jiri Olsa &lt;jolsa@kernel.org&gt;
Cc: Leo Yan &lt;leo.yan@arm.com&gt;
Cc: Mark Rutland &lt;mark.rutland@arm.com&gt;
Cc: Michael Petlan &lt;mpetlan@redhat.com&gt;
Cc: Namhyung Kim &lt;namhyung@kernel.org&gt;
Cc: Palmer Dabbelt &lt;palmer@dabbelt.com&gt;
Cc: Paul Walmsley &lt;pjw@kernel.org&gt;
Cc: Peter Zijlstra &lt;peterz@infradead.org&gt;
Cc: Qiao Zhao &lt;qzhao@redhat.com&gt;
Signed-off-by: Arnaldo Carvalho de Melo &lt;acme@redhat.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>perf mmap: Use the head selected by read_init</title>
<updated>2026-10-27T20:00:15+00:00</updated>
<author>
<name>Paul Moses</name>
<email>p@1g4.org</email>
</author>
<published>2026-10-20T17:20:58+00:00</published>
<link rel='alternate' type='text/html' href='https://git.toradex.cn/cgit/linux-toradex.git/commit/?id=022c64992af554cfd8c564bff86100fb5ed98678'/>
<id>022c64992af554cfd8c564bff86100fb5ed98678</id>
<content type='text'>
perf_mmap__push() reads the producer head before calling
perf_mmap__read_init(), which reads it again to select the range to
copy.

If the producer advances between those reads, push() copies through the
newer head but stores the earlier head in prev. The next drain starts at
that earlier position and can write the same records to perf.data again.

Use the head selected by read_init(): end for forward buffers and start
for overwrite buffers. Save it before the copy loop advances start so
prev matches the range that was copied.

Fixes: 8872481bd04850b1 ("perf mmap: Introduce perf_mmap__read_init()")
Signed-off-by: Paul Moses &lt;p@1g4.org&gt;
Cc: Adrian Hunter &lt;adrian.hunter@intel.com&gt;
Cc: Alexander Shishkin &lt;alexander.shishkin@linux.intel.com&gt;
Cc: Ian Rogers &lt;irogers@google.com&gt;
Cc: Ingo Molnar &lt;mingo@redhat.com&gt;
Cc: James Clark &lt;james.clark@linaro.org&gt;
Cc: Jiri Olsa &lt;jolsa@kernel.org&gt;
Cc: Kan Liang &lt;kan.liang@intel.com&gt;
Cc: Kan Liang &lt;kan.liang@linux.intel.com&gt;
Cc: Mark Rutland &lt;mark.rutland@arm.com&gt;
Cc: Namhyung Kim &lt;namhyung@kernel.org&gt;
Cc: Peter Zijlstra &lt;peterz@infradead.org&gt;
Signed-off-by: Arnaldo Carvalho de Melo &lt;acme@redhat.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
perf_mmap__push() reads the producer head before calling
perf_mmap__read_init(), which reads it again to select the range to
copy.

If the producer advances between those reads, push() copies through the
newer head but stores the earlier head in prev. The next drain starts at
that earlier position and can write the same records to perf.data again.

Use the head selected by read_init(): end for forward buffers and start
for overwrite buffers. Save it before the copy loop advances start so
prev matches the range that was copied.

Fixes: 8872481bd04850b1 ("perf mmap: Introduce perf_mmap__read_init()")
Signed-off-by: Paul Moses &lt;p@1g4.org&gt;
Cc: Adrian Hunter &lt;adrian.hunter@intel.com&gt;
Cc: Alexander Shishkin &lt;alexander.shishkin@linux.intel.com&gt;
Cc: Ian Rogers &lt;irogers@google.com&gt;
Cc: Ingo Molnar &lt;mingo@redhat.com&gt;
Cc: James Clark &lt;james.clark@linaro.org&gt;
Cc: Jiri Olsa &lt;jolsa@kernel.org&gt;
Cc: Kan Liang &lt;kan.liang@intel.com&gt;
Cc: Kan Liang &lt;kan.liang@linux.intel.com&gt;
Cc: Mark Rutland &lt;mark.rutland@arm.com&gt;
Cc: Namhyung Kim &lt;namhyung@kernel.org&gt;
Cc: Peter Zijlstra &lt;peterz@infradead.org&gt;
Signed-off-by: Arnaldo Carvalho de Melo &lt;acme@redhat.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>perf hist: Do not delete cumulative caller entries before stat_acc decays</title>
<updated>2026-10-27T19:59:44+00:00</updated>
<author>
<name>Ian Rogers</name>
<email>irogers@google.com</email>
</author>
<published>2026-10-26T06:50:45+00:00</published>
<link rel='alternate' type='text/html' href='https://git.toradex.cn/cgit/linux-toradex.git/commit/?id=98892841bdd8332c6a34654c9c45c0a9e860e2b2'/>
<id>98892841bdd8332c6a34654c9c45c0a9e860e2b2</id>
<content type='text'>
When symbol_conf.cumulate_callchain is enabled (the default in
'perf top -g'), callchain parent/caller entries have he-&gt;stat.period == 0
and accumulate their period in he-&gt;stat_acc-&gt;period.

hists__decay_entry() decays both he-&gt;stat.period and he-&gt;stat_acc-&gt;period,
but only checks if he-&gt;stat.period == 0 as the return condition for
deleting the entry in hists__decay_entries(). As a result, every callchain
caller entry with 0 self-period is unconditionally deleted from the
histogram on the first decay pass, destroying its callchain tree while it
still has non-zero cumulative period.

Only mark an entry for deletion in hists__decay_entry() when both
he-&gt;stat.period == 0 and (when cumulate_callchain is active)
he-&gt;stat_acc-&gt;period == 0.

Signed-off-by: Ian Rogers &lt;irogers@google.com&gt;
Cc: Adrian Hunter &lt;adrian.hunter@intel.com&gt;
Cc: Alessio Podda &lt;aleph.pi.gh@gmail.com&gt;
Cc: Alireza Haghdoost &lt;haghdoost@uber.com&gt;
Cc: Ingo Molnar &lt;mingo@redhat.com&gt;
Cc: James Clark &lt;james.clark@linaro.org&gt;
Cc: Jiri Olsa &lt;jolsa@kernel.org&gt;
Cc: Michael Liang &lt;mliang@purestorage.com&gt;
Cc: Namhyung Kim &lt;namhyung@kernel.org&gt;
Cc: Peter Zijlstra &lt;peterz@infradead.org&gt;
Cc: Shimin Guo &lt;shimin.guo@skydio.com&gt;
Cc: Tanushree Shah &lt;tshah@linux.ibm.com&gt;
Signed-off-by: Arnaldo Carvalho de Melo &lt;acme@redhat.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
When symbol_conf.cumulate_callchain is enabled (the default in
'perf top -g'), callchain parent/caller entries have he-&gt;stat.period == 0
and accumulate their period in he-&gt;stat_acc-&gt;period.

hists__decay_entry() decays both he-&gt;stat.period and he-&gt;stat_acc-&gt;period,
but only checks if he-&gt;stat.period == 0 as the return condition for
deleting the entry in hists__decay_entries(). As a result, every callchain
caller entry with 0 self-period is unconditionally deleted from the
histogram on the first decay pass, destroying its callchain tree while it
still has non-zero cumulative period.

Only mark an entry for deletion in hists__decay_entry() when both
he-&gt;stat.period == 0 and (when cumulate_callchain is active)
he-&gt;stat_acc-&gt;period == 0.

Signed-off-by: Ian Rogers &lt;irogers@google.com&gt;
Cc: Adrian Hunter &lt;adrian.hunter@intel.com&gt;
Cc: Alessio Podda &lt;aleph.pi.gh@gmail.com&gt;
Cc: Alireza Haghdoost &lt;haghdoost@uber.com&gt;
Cc: Ingo Molnar &lt;mingo@redhat.com&gt;
Cc: James Clark &lt;james.clark@linaro.org&gt;
Cc: Jiri Olsa &lt;jolsa@kernel.org&gt;
Cc: Michael Liang &lt;mliang@purestorage.com&gt;
Cc: Namhyung Kim &lt;namhyung@kernel.org&gt;
Cc: Peter Zijlstra &lt;peterz@infradead.org&gt;
Cc: Shimin Guo &lt;shimin.guo@skydio.com&gt;
Cc: Tanushree Shah &lt;tshah@linux.ibm.com&gt;
Signed-off-by: Arnaldo Carvalho de Melo &lt;acme@redhat.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>perf debug: Avoid calling eprintf() in pr_debugN() when verbose &lt; n</title>
<updated>2026-10-27T19:59:18+00:00</updated>
<author>
<name>Ian Rogers</name>
<email>irogers@google.com</email>
</author>
<published>2026-10-26T06:50:44+00:00</published>
<link rel='alternate' type='text/html' href='https://git.toradex.cn/cgit/linux-toradex.git/commit/?id=2fb45894e80d246ac6a9f0343343d581921cc3f2'/>
<id>2fb45894e80d246ac6a9f0343343d581921cc3f2</id>
<content type='text'>
pr_debugN() unconditionally evaluated its arguments and called eprintf()
-&gt; veprintf() before checking if verbose &gt;= n. In hot paths such as
dso__load_sym_internal(), two pr_debug4() calls per symbol accounted for
~15% of total CPU time in 'perf top' even at the default verbose == 0.

Guard eprintf() in pr_debugN() with unlikely(verbose &gt;= (n)) and define
pr_debug() in terms of pr_debugN(1, ...).

In test__parsing_callback() (tools/perf/tests/pmu-events.c), move the
test_generic_metric(mexp, 0) call out of the pr_debug() argument list so
the metric evaluation continues to run during 'perf test' when verbose
== 0.

Signed-off-by: Ian Rogers &lt;irogers@google.com&gt;
Cc: Adrian Hunter &lt;adrian.hunter@intel.com&gt;
Cc: Alessio Podda &lt;aleph.pi.gh@gmail.com&gt;
Cc: Alireza Haghdoost &lt;haghdoost@uber.com&gt;
Cc: Ingo Molnar &lt;mingo@redhat.com&gt;
Cc: James Clark &lt;james.clark@linaro.org&gt;
Cc: Jiri Olsa &lt;jolsa@kernel.org&gt;
Cc: Michael Liang &lt;mliang@purestorage.com&gt;
Cc: Namhyung Kim &lt;namhyung@kernel.org&gt;
Cc: Peter Zijlstra &lt;peterz@infradead.org&gt;
Cc: Shimin Guo &lt;shimin.guo@skydio.com&gt;
Cc: Tanushree Shah &lt;tshah@linux.ibm.com&gt;
Signed-off-by: Arnaldo Carvalho de Melo &lt;acme@redhat.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
pr_debugN() unconditionally evaluated its arguments and called eprintf()
-&gt; veprintf() before checking if verbose &gt;= n. In hot paths such as
dso__load_sym_internal(), two pr_debug4() calls per symbol accounted for
~15% of total CPU time in 'perf top' even at the default verbose == 0.

Guard eprintf() in pr_debugN() with unlikely(verbose &gt;= (n)) and define
pr_debug() in terms of pr_debugN(1, ...).

In test__parsing_callback() (tools/perf/tests/pmu-events.c), move the
test_generic_metric(mexp, 0) call out of the pr_debug() argument list so
the metric evaluation continues to run during 'perf test' when verbose
== 0.

Signed-off-by: Ian Rogers &lt;irogers@google.com&gt;
Cc: Adrian Hunter &lt;adrian.hunter@intel.com&gt;
Cc: Alessio Podda &lt;aleph.pi.gh@gmail.com&gt;
Cc: Alireza Haghdoost &lt;haghdoost@uber.com&gt;
Cc: Ingo Molnar &lt;mingo@redhat.com&gt;
Cc: James Clark &lt;james.clark@linaro.org&gt;
Cc: Jiri Olsa &lt;jolsa@kernel.org&gt;
Cc: Michael Liang &lt;mliang@purestorage.com&gt;
Cc: Namhyung Kim &lt;namhyung@kernel.org&gt;
Cc: Peter Zijlstra &lt;peterz@infradead.org&gt;
Cc: Shimin Guo &lt;shimin.guo@skydio.com&gt;
Cc: Tanushree Shah &lt;tshah@linux.ibm.com&gt;
Signed-off-by: Arnaldo Carvalho de Melo &lt;acme@redhat.com&gt;
</pre>
</div>
</content>
</entry>
</feed>
