In 754187ad73b73bcb ("perf build: Remove NO_AUXTRACE build option")
sys/types.h was removed, which broke the build in all Alpine Linux
releases, as musl libc has pid_t defined via sys/types.h, add it back.
Fixes: 754187ad73b73bcb ("perf build: Remove NO_AUXTRACE build option") Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> Reviewed-by: Ian Rogers <irogers@google.com> Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Namhyung Kim [Tue, 2 Dec 2025 23:57:18 +0000 (15:57 -0800)]
perf test: Add kallsyms split test
Create a fake root directory for /proc/{version,modules,kallsyms} in
/tmp for testing. The kallsyms has a bad symbol in the module and it
causes the main map splitted. The test ensures it only has two maps -
kernel and the module and it finds the initial map after the module
without creating the split maps like [kernel].0 and so on.
$ perf test -vv "split kallsyms"
69: split kallsyms:
--- start ---
test child forked, pid 1016196
try to create fake root directory
create kernel maps from the fake root directory
maps__set_modules_path_dir: cannot open /tmp/perf-test.Zrv6Sy/lib/modules/X.Y.Z dir
Problems setting modules path maps, continuing anyway...
Failed to open /tmp/perf-test.Zrv6Sy/proc/kcore. Note /proc/kcore requires CAP_SYS_RAWIO capability to access.
Using /tmp/perf-test.Zrv6Sy/proc/kallsyms for symbols
kernel map loaded - check symbol and map
---- end(0) ----
69: split kallsyms : Ok
Reviewed-by: Ian Rogers <irogers@google.com> Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Namhyung Kim [Tue, 2 Dec 2025 23:57:17 +0000 (15:57 -0800)]
perf tools: Use machine->root_dir to find /proc/kallsyms
This is for test functions to find the kallsyms correctly. It can find
the machine from the kernel maps and use its root_dir. This is helpful
to setup fake /proc directory for testing.
Reviewed-by: Ian Rogers <irogers@google.com> Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Namhyung Kim [Tue, 2 Dec 2025 23:57:16 +0000 (15:57 -0800)]
perf tools: Fallback to initial kernel map properly
In maps__split_kallsyms(), it assumes new kernel map when it finds a
symbol without module after any module and the initial kernel map has
some symbols. Because it expects modules are out of the kernel map so
modules should not have symbols in the kernel map.
For example, the following memory map shows symbols and maps. Any
symbols in the module 1 area will go to the module 1. The main kernel
map starts at 0xffffffffbc200000. But if any symbol has a module
between the symbols in that area, next symbols after 0xffffffffbd008000
will generate new kernel maps like [kernel].1.
This is very fragile when the module has a symbol that falls into the
main kernel map for some reason. My system has a livepatch module with
such symbols. And it created a lot of new kernel maps after those
symbols. But the symbol may have broken addresses and the later symbols
can still be found in the initial kernel map.
Let's check the symbol address in the initial map and use it if found.
Reviewed-by: Ian Rogers <irogers@google.com> Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Namhyung Kim [Tue, 2 Dec 2025 23:57:14 +0000 (15:57 -0800)]
perf tools: Mark split kallsyms DSOs as loaded
The maps__split_kallsyms() will split symbols to module DSOs if it comes
from a module. It also handled some unusual kernel symbols after modules
by creating new kernel maps like "[kernel].0".
But they are pseudo DSOs to have those unexpected symbols. They should
not be considered as unloaded kernel DSOs. Otherwise the dso__load()
for them will end up calling dso__load_kallsyms() and then
maps__split_kallsyms() again and again.
Reviewed-by: Ian Rogers <irogers@google.com> Fixes: 2e538c4a1847291cf ("perf tools: Improve kernel/modules symbol lookup") Signed-off-by: Namhyung Kim <namhyung@kernel.org>
It's possible that some kernel samples don't have matching deferred
callchain records when the profiling session was ended before the
threads came back to userspace. Let's flush the samples before
finish the session.
Reviewed-by: Ian Rogers <irogers@google.com> Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Namhyung Kim [Thu, 20 Nov 2025 23:48:03 +0000 (15:48 -0800)]
perf tools: Merge deferred user callchains
Save samples with deferred callchains in a separate list and deliver
them after merging the user callchains. If users don't want to merge
they can set tool->merge_deferred_callchains to false to prevent the
behavior.
With previous result, now perf script will show the merged callchains.
Namhyung Kim [Thu, 20 Nov 2025 23:48:00 +0000 (15:48 -0800)]
perf tools: Minimal DEFERRED_CALLCHAIN support
Add a new event type for deferred callchains and a new callback for the
struct perf_tool. For now it doesn't actually handle the deferred
callchains but it just marks the sample if it has the PERF_CONTEXT_
USER_DEFFERED in the callchain array.
At least, perf report can dump the raw data with this change. Actually
this requires the next commit to enable attr.defer_callchain, but if you
already have a data file, it'll show the following result.
Ian Rogers [Tue, 2 Dec 2025 17:50:01 +0000 (09:50 -0800)]
perf jevents: Allow metric groups not to be named
It can be convenient to have unnamed metric groups for the sake of
organizing other metrics and metric groups. An unspecified name
shouldn't contribute to the MetricGroup json value, so don't record
it.
Signed-off-by: Ian Rogers <irogers@google.com> Tested-by: Thomas Falcon <thomas.falcon@intel.com> Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Ian Rogers [Tue, 2 Dec 2025 17:49:59 +0000 (09:49 -0800)]
perf jevents: Update metric constraint support
Previous metric constraints were binary, either none or don't group
when the NMI watchdog is present. Update to match the definitions in
'enum metric_event_groups' in pmu-events.h.
Signed-off-by: Ian Rogers <irogers@google.com> Tested-by: Thomas Falcon <thomas.falcon@intel.com> Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Ian Rogers [Tue, 2 Dec 2025 17:49:56 +0000 (09:49 -0800)]
perf python: Correct copying of metric_leader in an evsel
Ensure the metric_leader is copied and set up correctly. In
compute_metric determine the correct metric_leader event to match the
requested CPU. Fixes the handling of metrics particularly on hybrid
machines.
Signed-off-by: Ian Rogers <irogers@google.com> Tested-by: Thomas Falcon <thomas.falcon@intel.com> Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Namhyung Kim [Tue, 25 Nov 2025 08:07:47 +0000 (00:07 -0800)]
perf test: Add python JIT dump test
Add a test case for the python interpreter like below so that we can
make sure it won't break again. To validate the effect of build-ID
generation, it adds and removes the JIT'ed DSOs to/from the build-ID
cache for the test.
$ perf test -vv jitdump
84: python profiling with jitdump:
--- start ---
test child forked, pid 214316
Run python with -Xperf_jit
[ perf record: Woken up 5 times to write data ]
[ perf record: Captured and wrote 1.180 MB /tmp/__perf_test.perf.data.XbqZNm (140 samples) ]
Generate JIT-ed DSOs using perf inject
Add JIT-ed DSOs to the build-ID cache
Check the symbol containing the script name
Found 108 matching lines
Remove JIT-ed DSOs from the build-ID cache
---- end(0) ----
84: python profiling with jitdump : Ok
Namhyung Kim [Tue, 25 Nov 2025 08:07:46 +0000 (00:07 -0800)]
perf jitdump: Add sym/str-tables to build-ID generation
It was reported that python backtrace with JIT dump was broken after the
change to built-in SHA-1 implementation. It seems python generates the
same JIT code for each function. They will become separate DSOs but the
contents are the same. Only difference is in the symbol name.
But this caused a problem that every JIT'ed DSOs will have the same
build-ID which makes perf confused. And it resulted in no python
symbols (from JIT) in the output.
Looking back at the original code before the conversion, it used the
load_addr as well as the code section to distinguish each DSO. But it'd
be better to use contents of symtab and strtab instead as it aligns with
some linker behaviors.
This patch adds a buffer to save all the contents in a single place for
SHA-1 calculation. Probably we need to add sha1_update() or similar to
update the existing hash value with different contents and use it here.
But it's out of scope for this change and I'd like something that can be
backported to the stable trees easily.
Reviewed-by: Ian Rogers <irogers@google.com> Cc: Eric Biggers <ebiggers@kernel.org> Cc: Pablo Galindo <pablogsal@gmail.com> Cc: Fangrui Song <maskray@sourceware.org> Link: https://github.com/python/cpython/issues/139544 Fixes: e3f612c1d8f3945b ("perf genelf: Remove libcrypto dependency and use built-in sha1()") Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Ian Rogers [Mon, 1 Dec 2025 23:11:36 +0000 (15:11 -0800)]
perf test: Fix hybrid testing of event fallback test
The mem-loads-aux event exists on hybrid systems but the "cpu" PMU
does not. This causes an event parsing error which erroneously makes
the test look like it is failing. Avoid naming the PMU to avoid
this. Rather than cleaning up perf.data in the directory the test is
run, explicitly send the 'perf record' output to /dev/null and avoid
any cleanup scripts.
Fixes: fc9c17b22352 ("perf test: Add a perf event fallback test") Signed-off-by: Ian Rogers <irogers@google.com> Reviewed-by: Dapeng Mi <dapeng1.mi@linux.intel.com> Signed-off-by: Namhyung Kim <namhyung@kernel.org>
We have a different way to handle those arguments and this change
confuses perf trace then make some tests failing. Fix it by skipping
the new fields that have "__data_loc char[]" type.
Maybe we can switch to this instead of the BPF augmentation later.
Reviewed-by: Howard Chu <howardchu95@gmail.com> Tested-by: Thomas Richter <tmricht@linux.ibm.com> Tested-by: Steven Rostedt (Google) <rostedt@goodmis.org> Cc: Steven Rostedt <rostedt@goodmis.org> Cc: Howard Chu <howardchu95@gmail.com> Reported-by: Thomas Richter <tmricht@linux.ibm.com> Signed-off-by: Namhyung Kim <namhyung@kernel.org>
James Clark [Mon, 24 Nov 2025 10:59:08 +0000 (10:59 +0000)]
perf tools: Don't read build-ids from non-regular files
Simplify the build ID reading code by removing the non-blocking option.
Having to pass the correct option to this function was fragile and a
mistake would result in a hang, see the linked fix. Furthermore,
compressed files are always opened blocking anyway, ignoring the
non-blocking option.
We also don't expect to read build IDs from non-regular files. The only
hits to this function that are non-regular are devices that won't be elf
files with build IDs, for example "/dev/dri/renderD129".
Now instead of opening these as non-blocking and failing to read, we
skip them. Even if something like a pipe or character device did have a
build ID, I don't think it would have worked because you need to call
read() in a loop, check for -EAGAIN and handle timeouts to make
non-blocking reads work.
James Clark [Tue, 11 Nov 2025 11:37:59 +0000 (11:37 +0000)]
perf docs: arm-spe: Document new SPE filtering features
FEAT_SPE_EFT and FEAT_SPE_FDS etc have new user facing format attributes
so document them. Also document existing 'event_filter' bits that were
missing from the doc and the fact that latency values are stored in the
weight field.
Reviewed-by: Leo Yan <leo.yan@arm.com> Tested-by: Leo Yan <leo.yan@arm.com> Reviewed-by: Ian Rogers <irogers@google.com> Signed-off-by: James Clark <james.clark@linaro.org> Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org> Signed-off-by: Namhyung Kim <namhyung@kernel.org>
James Clark [Tue, 11 Nov 2025 11:37:58 +0000 (11:37 +0000)]
perf tools: Add support for perf_event_attr::config4
perf_event_attr has gained a new field, config4, so add support for it
extending the existing configN support.
Reviewed-by: Leo Yan <leo.yan@arm.com> Reviewed-by: Ian Rogers <irogers@google.com> Tested-by: Leo Yan <leo.yan@arm.com> Signed-off-by: James Clark <james.clark@linaro.org> Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org> Signed-off-by: Namhyung Kim <namhyung@kernel.org>
James Clark [Tue, 11 Nov 2025 11:37:57 +0000 (11:37 +0000)]
tools headers UAPI: Sync linux/perf_event.h with the kernel sources
To pickup config4 changes.
Tested-by: Leo Yan <leo.yan@arm.com> Reviewed-by: Ian Rogers <irogers@google.com> Signed-off-by: James Clark <james.clark@linaro.org> Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org> Signed-off-by: Namhyung Kim <namhyung@kernel.org>
perf: replace strcpy() with strncpy() in util/jitdump.c
Usage of strcpy() can lead to buffer overflows. Therefore, it has been
replaced with strncpy(). The output file path is provided as a parameter
and might be restricted by command-line by default. But this defensive
patch will prevent any potential overflow, making the code more robust
against future changes in input handling.
Testing:
- ran perf test from tools/perf and did not observe any regression with
the earlier code
Signed-off-by: Hrishikesh Suresh <hrishikesh123s@gmail.com> Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Namhyung Kim [Thu, 20 Nov 2025 00:47:25 +0000 (16:47 -0800)]
perf list: Share print state with JSON output
The JSON print state has only one different field (need_sep). Let's
add the default print state to the json state and use it. Then we can
use the 'ps' variable to update the state properly.
This is a preparation for the next commit.
Reviewed-by: Ian Rogers <irogers@google.com> Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Namhyung Kim [Thu, 20 Nov 2025 00:47:24 +0000 (16:47 -0800)]
perf list: Print matching PMU events for --unit
When --unit option is used, pmu_glob is set to the argument. It should
match with event PMU and display the matching ones only. But it also
shows raw events and metrics after that.
$ perf list --unit tool
List of pre-defined events (to be used in -e or -M):
tool:
core_wide
[1 if not SMT,if SMT are events being gathered on all SMT threads 1 otherwise 0. Unit: tool]
duration_time
[Wall clock interval time in nanoseconds. Unit: tool]
has_pmem
[1 if persistent memory installed otherwise 0. Unit: tool]
num_cores
[Number of cores. A core consists of 1 or more thread,with each thread being associated with a logical Linux CPU. Unit: tool]
num_cpus
[Number of logical Linux CPUs. There may be multiple such CPUs on a core. Unit: tool]
...
rNNN [Raw event descriptor]
cpu/event=0..255,pc,edge,.../modifier [Raw event descriptor]
[(see 'man perf-list' or 'man perf-record' on how to encode it)]
breakpoint//modifier [Raw event descriptor]
cstate_core/event=0..0xffffffffffffffff/modifier [Raw event descriptor]
cstate_pkg/event=0..0xffffffffffffffff/modifier [Raw event descriptor]
drm_i915//modifier [Raw event descriptor]
hwmon_acpitz//modifier [Raw event descriptor]
hwmon_ac//modifier [Raw event descriptor]
hwmon_bat0//modifier [Raw event descriptor]
hwmon_coretemp//modifier [Raw event descriptor]
...
Metric Groups:
Backend: [Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet]
tma_core_bound
[This metric represents fraction of slots where Core non-memory issues were of a bottleneck]
tma_info_core_ilp
[Instruction-Level-Parallelism (average number of uops executed when there is execution) per thread (logical-processor)]
tma_info_memory_l2mpki
[L2 cache true misses per kilo instruction for retired demand loads]
...
This change makes it print the tool PMU events only.
Reviewed-by: Ian Rogers <irogers@google.com> Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Ian Rogers [Wed, 19 Nov 2025 19:30:47 +0000 (11:30 -0800)]
perf test all metrics: Fully ignore Default metric failures
Determine if a metric is default from `perf list --raw-dump $m` eg:
```
$ perf list --raw-dump l1_prefetch_miss_rate
Default4 l1_prefetch_miss_rate
```
If a metric has "not supported" or "no supported events" then ignore
these failures for default metrics. Tidy up the skip/fail messages in
the output to make them easier to spot/read.
```
$ perf list -vv "all metrics"
...
Testing llc_miss_rate
[Ignored llc_miss_rate] failed but as a Default metric this can be expected
Error: No supported events found. The LLC-loads event is not supported.
...
```
Reported-by: Thomas Richter <tmricht@linux.ibm.com> Closes: https://lore.kernel.org/linux-perf-users/20251119104751.51960-1-tmricht@linux.ibm.com/ Reported-by: Namhyung Kim <namhyung@kernel.org> Reported-by: James Clark <james.clark@linaro.org> Closes: https://lore.kernel.org/lkml/aRi9xnwdLh3Dir9f@google.com/ Signed-off-by: Ian Rogers <irogers@google.com> Reviewed-by: James Clark <james.clark@linaro.org> Tested-by: Thomas Richter <tmricht@linux.ibm.com> Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Ian Rogers [Fri, 14 Nov 2025 22:05:47 +0000 (14:05 -0800)]
perf evsel: Skip store_evsel_ids for non-perf-event PMUs
The IDs are associated with perf events and not applicable to non-perf
event PMUs. The failure to generate the ids was causing perf stat
record to fail.
Some events weren't counted. Try disabling the NMI watchdog:
echo 0 > /proc/sys/kernel/nmi_watchdog
perf stat ...
echo 1 > /proc/sys/kernel/nmi_watchdog
```
Reported-by: James Clark <james.clark@linaro.org> Closes: https://lore.kernel.org/lkml/ca0f0cd3-7335-48f9-8737-2f70a75b019a@linaro.org/ Signed-off-by: Ian Rogers <irogers@google.com> Tested-by: Thomas Falcon <thomas.falcon@intel.com> Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Ian Rogers [Fri, 14 Nov 2025 22:05:45 +0000 (14:05 -0800)]
perf header: Switch "cpu" for find_core_pmu in caps feature writing
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 <irogers@google.com> Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Ian Rogers [Wed, 19 Nov 2025 05:05:54 +0000 (21:05 -0800)]
perf maps: Avoid RC_CHK use after free
The case of __maps__fixup_overlap_and_insert where the "new" maps
covers existing mappings can create a use-after-free with reference
count checking enabled. The issue is that "pos" holds a map pointer
from maps_by_address that is put from maps_by_address but then used to
look for a map in maps_by_name (the compared map is now a
use-after-free). The issue stems from using maps__remove which redoes
some of the searches already done by __maps__fixup_overlap_and_insert,
so optimize the code (by avoiding repeated searches) and avoid the
use-after-free by inlining the appropriate removal code.
Reported-by: kernel test robot <oliver.sang@intel.com> Closes: https://lore.kernel.org/oe-lkp/202511141407.f9edcfa6-lkp@intel.com Signed-off-by: Ian Rogers <irogers@google.com> Reviewed-by: James Clark <james.clark@linaro.org> Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Ian Rogers [Tue, 18 Nov 2025 21:13:24 +0000 (13:13 -0800)]
perf stat: Read tool events last
When reading a metric like memory bandwidth on multiple sockets, the
additional sockets will be on CPUS > 0. Because of the affinity
reading, the counters are read on CPU 0 along with the time, then the
later sockets are read. This can lead to the later sockets having a
bandwidth larger than is possible for the period of time. To avoid
this move the reading of tool events to occur after all other events
are read.
Signed-off-by: Ian Rogers <irogers@google.com> Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Leo Yan [Wed, 12 Nov 2025 18:24:43 +0000 (18:24 +0000)]
perf arm_spe: Synthesize memory samples for SIMD operations
Synthesize memory samples for SIMD operations (including Advanced SIMD,
SVE, and SME). To provide complete information, also generate data
source entries for SIMD operations.
Since memory operations are not limited to load and store, set
PERF_MEM_OP_STORE if the operation does not fall into these cases.
Signed-off-by: Leo Yan <leo.yan@arm.com> Reviewed-by: Ian Rogers <irogers@google.com> Reviewed-by: James Clark <james.clark@linaro.org> Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Leo Yan [Wed, 12 Nov 2025 18:24:42 +0000 (18:24 +0000)]
perf arm_spe: Expose SIMD information in other operations
The other operations contain SME data processing, ASE (Advanced SIMD)
and floating-point operations. Expose these info in the records.
Signed-off-by: Leo Yan <leo.yan@arm.com> Reviewed-by: Ian Rogers <irogers@google.com> Reviewed-by: James Clark <james.clark@linaro.org> Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Leo Yan [Wed, 12 Nov 2025 18:24:41 +0000 (18:24 +0000)]
perf arm_spe: Report GCS in record
Report GCS related info in records.
Signed-off-by: Leo Yan <leo.yan@arm.com> Reviewed-by: Ian Rogers <irogers@google.com> Reviewed-by: James Clark <james.clark@linaro.org> Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Leo Yan [Wed, 12 Nov 2025 18:24:40 +0000 (18:24 +0000)]
perf arm_spe: Report memset and memcpy in records
Expose memset and memcpy related info in records.
Signed-off-by: Leo Yan <leo.yan@arm.com> Reviewed-by: Ian Rogers <irogers@google.com> Reviewed-by: James Clark <james.clark@linaro.org> Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Leo Yan [Wed, 12 Nov 2025 18:24:39 +0000 (18:24 +0000)]
perf arm_spe: Report associated info for SVE / SME operations
SVE / SME operations can be predicated or Gather load / scatter store,
save the relevant info into record.
Signed-off-by: Leo Yan <leo.yan@arm.com> Reviewed-by: Ian Rogers <irogers@google.com> Reviewed-by: James Clark <james.clark@linaro.org> Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Leo Yan [Wed, 12 Nov 2025 18:24:38 +0000 (18:24 +0000)]
perf arm_spe: Report extended memory operations in records
Extended memory operations include atomic (AT), acquire/release (AR),
and exclusive (EXCL) operations. Save the relevant information
in the records.
Signed-off-by: Leo Yan <leo.yan@arm.com> Reviewed-by: Ian Rogers <irogers@google.com> Reviewed-by: James Clark <james.clark@linaro.org> Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Leo Yan [Wed, 12 Nov 2025 18:24:37 +0000 (18:24 +0000)]
perf arm_spe: Report MTE allocation tag in record
Save MTE tag info in memory record.
Signed-off-by: Leo Yan <leo.yan@arm.com> Reviewed-by: Ian Rogers <irogers@google.com> Reviewed-by: James Clark <james.clark@linaro.org> Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Leo Yan [Wed, 12 Nov 2025 18:24:36 +0000 (18:24 +0000)]
perf arm_spe: Report register access in record
Record register access info for load / store operations.
Signed-off-by: Leo Yan <leo.yan@arm.com> Reviewed-by: Ian Rogers <irogers@google.com> Reviewed-by: James Clark <james.clark@linaro.org> Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Leo Yan [Wed, 12 Nov 2025 18:24:35 +0000 (18:24 +0000)]
perf arm_spe: Introduce data processing macro for SVE operations
Introduce the ARM_SPE_OP_DP (data processing) macro as associated
information for SVE operations. For SVE register access, only
ARM_SPE_OP_SVE is set; for SVE data processing, both ARM_SPE_OP_SVE and
ARM_SPE_OP_DP are set together.
Signed-off-by: Leo Yan <leo.yan@arm.com> Reviewed-by: Ian Rogers <irogers@google.com> Reviewed-by: James Clark <james.clark@linaro.org> Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Leo Yan [Wed, 12 Nov 2025 18:24:34 +0000 (18:24 +0000)]
perf arm_spe: Consolidate operation types
Consolidate operation types in a way:
(a) Extract the second-level types into separate enums.
(b) The second-level types for memory and SIMD operations are classified
by modules. E.g., an operation may relate to general register,
SIMD/FP, SVE, etc.
(c) The associated information tells details. E.g., an operation is
load or store, whether it is atomic operation, etc.
Start the enum items for the second-level types from 8 to accommodate
more entries within a 32-bit integer.
Signed-off-by: Leo Yan <leo.yan@arm.com> Reviewed-by: Ian Rogers <irogers@google.com> Reviewed-by: James Clark <james.clark@linaro.org> Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Leo Yan [Wed, 12 Nov 2025 18:24:33 +0000 (18:24 +0000)]
perf arm_spe: Remove unused operation types
Remove unused SVE operation types. These operations will be reintroduced
in subsequent refactoring, but with a different format.
Signed-off-by: Leo Yan <leo.yan@arm.com> Reviewed-by: Ian Rogers <irogers@google.com> Reviewed-by: James Clark <james.clark@linaro.org> Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Signed-off-by: Leo Yan <leo.yan@arm.com> Reviewed-by: Ian Rogers <irogers@google.com> Reviewed-by: James Clark <james.clark@linaro.org> Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Leo Yan [Wed, 12 Nov 2025 18:24:31 +0000 (18:24 +0000)]
perf arm_spe: Decode ASE and FP fields in other operation
Add a check for other operation, which prevents any incorrectly
classifying. Parse the ASE and FP fields.
After:
. 0000002f: 48 06 OTHER ASE FP INSN-OTHER
. 00000031: b2 08 80 48 01 08 00 ff ff VA 0xffff000801488008
. 0000003a: 9a 00 00 LAT 0 XLAT
. 0000003d: 42 16 EV RETIRED L1D-ACCESS TLB-ACCESS
Signed-off-by: Leo Yan <leo.yan@arm.com> Reviewed-by: Ian Rogers <irogers@google.com> Reviewed-by: James Clark <james.clark@linaro.org> Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Rename the macro to SPE_OP_PKT_OTHER_SUBCLASS_SVE to unify naming.
Signed-off-by: Leo Yan <leo.yan@arm.com> Reviewed-by: Ian Rogers <irogers@google.com> Reviewed-by: James Clark <james.clark@linaro.org> Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Signed-off-by: Leo Yan <leo.yan@arm.com> Reviewed-by: Ian Rogers <irogers@google.com> Reviewed-by: James Clark <james.clark@linaro.org> Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Leo Yan [Wed, 12 Nov 2025 18:24:28 +0000 (18:24 +0000)]
perf arm_spe: Unify operation naming
Rename extended subclass and SVE/SME register access subclass, so that
the naming can be consistent cross all sub classes.
Add an log "SVE-SME-REG" for the SVE/SME register access, this is easier
for parsing.
Signed-off-by: Leo Yan <leo.yan@arm.com> Reviewed-by: Ian Rogers <irogers@google.com> Reviewed-by: James Clark <james.clark@linaro.org> Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Leo Yan [Wed, 12 Nov 2025 18:24:27 +0000 (18:24 +0000)]
perf arm_spe: Fix memset subclass in operation
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 <leo.yan@arm.com> Reviewed-by: Ian Rogers <irogers@google.com> Reviewed-by: James Clark <james.clark@linaro.org> Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Ian Rogers [Thu, 13 Nov 2025 18:05:13 +0000 (10:05 -0800)]
perf tool_pmu: More accurately set the cpus for tool events
The user and system time events can record on different CPUs, but for
all other events a single CPU map of just CPU 0 makes sense. In
parse-events detect a tool PMU and then pass the perf_event_attr so
that the tool_pmu can return CPUs specific for the event. This avoids
a CPU map of all online CPUs being used for events like
duration_time. Avoiding this avoids the evlist CPUs containing CPUs
for which duration_time just gives 0. Minimizing the evlist CPUs can
remove unnecessary sched_setaffinity syscalls that delay metric
calculations.
Signed-off-by: Ian Rogers <irogers@google.com> Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Ian Rogers [Thu, 13 Nov 2025 18:05:11 +0000 (10:05 -0800)]
perf stat: Reduce scope of ru_stats
The ru_stats are used to capture user and system time stats when a
process exits. These are then applied to user and system time tool
events if their reads fail due to the process terminating. Reduce the
scope now the metric code no longer reads these values.
Signed-off-by: Ian Rogers <irogers@google.com> Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Ian Rogers [Thu, 13 Nov 2025 18:05:10 +0000 (10:05 -0800)]
perf stat-shadow: Read tool events directly
When reading time values for metrics don't use the globals updated in
builtin-stat, just read the events as regular events. The only
exception is for time events where nanoseconds need converting to
seconds as metrics assume time metrics are in seconds.
Signed-off-by: Ian Rogers <irogers@google.com> Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Ian Rogers [Thu, 13 Nov 2025 18:05:09 +0000 (10:05 -0800)]
perf tool_pmu: Use old_count when computing count values for time events
When running in interval mode every third count of a time event isn't
showing properly:
```
$ perf stat -e duration_time -a -I 1000
1.001082862 1,002,290,425 duration_time
2.004264262 1,003,183,516 duration_time
3.007381401 <not counted> duration_time
4.011160141 1,003,705,631 duration_time
5.014515385 1,003,290,110 duration_time
6.018539680 <not counted> duration_time
7.022065321 1,003,591,720 duration_time
```
The regression came in with a different fix, found through bisection,
commit 68cb1567439f ("perf tool_pmu: Fix aggregation on
duration_time"). The issue is caused by the enabled and running time
of the event matching the old_count's and creating a delta of 0, which
is indicative of an error.
Fixes: 68cb1567439f ("perf tool_pmu: Fix aggregation on duration_time") Signed-off-by: Ian Rogers <irogers@google.com> Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Ian Rogers [Thu, 13 Nov 2025 18:05:07 +0000 (10:05 -0800)]
libperf cpumap: Reduce allocations and sorting in intersect
On hybrid platforms the CPU maps are often disjoint. Rather than copy
CPUs and trim, compute the number of common CPUs, if none early exit,
otherwise copy in an sorted order. This avoids memory allocation in
the disjoint case and avoids a second malloc and useless sort in the
previous trim cases.
Signed-off-by: Ian Rogers <irogers@google.com> Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Ian Rogers [Wed, 12 Nov 2025 19:53:11 +0000 (11:53 -0800)]
perf stat: Display metric-only for 0 counters
0 counters may occur in hypervisor settings but metric-only output is
always expected. This resolves an issue in the "perf stat STD output
linter" test.
Signed-off-by: Ian Rogers <irogers@google.com> Tested-by: Thomas Falcon <thomas.falcon@intel.com> Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Ian Rogers [Wed, 12 Nov 2025 19:53:08 +0000 (11:53 -0800)]
perf test: Be tolerant of missing json metric none value
print_metric_only_json and print_metric_end in stat-display.c may
create a metric value of "none" which fails validation as isfloat. Add
a helper to properly validate metric numeric values.
Signed-off-by: Ian Rogers <irogers@google.com> Signed-off-by: Namhyung Kim <namhyung@kernel.org>
builtin-script.c:347:36: error: unused function 'evsel_script' [-Werror,-Wunused-function] Fixes: 3622990efaab ("perf script: Change metric format to use json metrics") Signed-off-by: James Clark <james.clark@linaro.org> Reviewed-by: Ian Rogers <irogers@google.com> Signed-off-by: Namhyung Kim <namhyung@kernel.org>
The instructions event is now provided in json meaning the has_event
test always succeeds. Switch to using non-legacy event names in the
affected metrics.
Reported-by: Thomas Richter <tmricht@linux.ibm.com> Closes: https://lore.kernel.org/linux-perf-users/3e80f453-f015-4f4f-93d3-8df6bb6b3c95@linux.ibm.com/ Fixes: 0012e0fa221b ("perf jevents: Add legacy-hardware and legacy-cache json") Signed-off-by: Ian Rogers <irogers@google.com> Reviewed-by: Thomas Richter <tmricht@linux.ibm.com> Tested-by: Thomas Richter <tmricht@linux.ibm.com> Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Ian Rogers [Mon, 10 Nov 2025 01:31:51 +0000 (17:31 -0800)]
perf build: Remove NO_AUXTRACE build option
The NO_AUXTRACE build option was used when the __get_cpuid feature
test failed or if it was provided on the command line. The option no
longer avoids a dependency on a library and so having the option is
just adding complexity to the code base. Remove the option
CONFIG_AUXTRACE from Build files and HAVE_AUXTRACE_SUPPORT by assuming
it is always defined.
Signed-off-by: Ian Rogers <irogers@google.com> Reviewed-by: James Clark <james.clark@linaro.org> Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Ian Rogers [Mon, 10 Nov 2025 01:31:50 +0000 (17:31 -0800)]
tool build: Remove __get_cpuid feature test
This feature test is no longer used so remove.
The function tested by the feature test is used in:
tools/power/x86/x86_energy_perf_policy/x86_energy_perf_policy.c
however, the Makefile just assumes the presence of the function and
doesn't perform a build feature test for it.
Signed-off-by: Ian Rogers <irogers@google.com> Reviewed-by: James Clark <james.clark@linaro.org> Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Ian Rogers [Mon, 10 Nov 2025 01:31:48 +0000 (17:31 -0800)]
perf intel-pt: Use the perf provided "cpuid.h"
Rather than having a feature test and include of <cpuid.h> for the
__get_cpuid function, use the cpuid function provided by
tools/perf/arch/x86/util/cpuid.h.
Signed-off-by: Ian Rogers <irogers@google.com> Reviewed-by: James Clark <james.clark@linaro.org> Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Zide Chen [Wed, 12 Nov 2025 16:48:23 +0000 (08:48 -0800)]
perf test: Add a perf event fallback test
This adds test cases to verify the precise ip fallback logic:
- If the system supports precise ip, for an event given with the maximum
precision level, it should be able to decrease precise_ip to find a
supported level.
- The same fallback behavior should also work in more complex scenarios,
such as event groups or when PEBS is involved
Additional fallback tests, such as those covering missing feature cases,
can be added in the future.
Suggested-by: Ian Rogers <irogers@google.com> Reviewed-by: Dapeng Mi <dapeng1.mi@linux.intel.com> Signed-off-by: Zide Chen <zide.chen@intel.com> Reviewed-by: Ian Rogers <irogers!@google.com> Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Namhyung Kim [Thu, 6 Nov 2025 07:28:34 +0000 (23:28 -0800)]
perf stat: Align metric output without events
One of my concern in the perf stat output was the alignment in the
metrics and shadow stats. I think it missed to calculate the basic
output length using COUNTS_LEN and EVNAME_LEN but missed to add the
unit length like "msec" and surround 2 spaces. I'm not sure why it's
not printed below though.
But anyway, now it shows correctly aligned metric output.
Ian Rogers [Tue, 11 Nov 2025 21:22:06 +0000 (13:22 -0800)]
perf tool_pmu: Make core_wide and target_cpu json events
For the sake of better documentation, add core_wide and target_cpu to
the tool.json. When the values of system_wide and
user_requested_cpu_list are unknown, use the values from the global
stat_config.
Example output showing how '-a' modifies the values in `perf stat`:
```
$ perf stat -e core_wide,target_cpu true
$ perf list
...
tool:
core_wide
[1 if not SMT,if SMT are events being gathered on all SMT threads 1 otherwise 0. Unit: tool]
...
target_cpu
[1 if CPUs being analyzed,0 if threads/processes. Unit: tool]
...
```
Signed-off-by: Ian Rogers <irogers@google.com> Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Ian Rogers [Tue, 11 Nov 2025 21:22:05 +0000 (13:22 -0800)]
perf test stat csv: Update test expectations and events
Explicitly use a metric rather than implicitly expecting '-e
instructions,cycles' to produce a metric. Use a metric with software
events to make it more compatible.
Signed-off-by: Ian Rogers <irogers@google.com> Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Ian Rogers [Tue, 11 Nov 2025 21:22:04 +0000 (13:22 -0800)]
perf test stat: Update test expectations and events
test_stat_record_report and test_stat_record_script used default
output which triggers a bug when sending metrics. As this isn't
relevant to the test switch to using named software events.
Update the match in test_hybrid as the cycles event is now cpu-cycles
to workaround potential ARM issues.
Signed-off-by: Ian Rogers <irogers@google.com> Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Ian Rogers [Tue, 11 Nov 2025 21:21:59 +0000 (13:21 -0800)]
perf test stat+json: Improve metric-only testing
When testing metric-only, pass a metric to perf rather than expecting
a hard coded metric value to be generated.
Remove keys that were really metric-only units and instead don't
expect metric only to have a matching json key as it encodes metrics
as {"metric_name", "metric_value"}.
Signed-off-by: Ian Rogers <irogers@google.com> Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Ian Rogers [Tue, 11 Nov 2025 21:21:56 +0000 (13:21 -0800)]
perf stat: Fix default metricgroup display on hybrid
The logic to skip output of a default metric line was firing on
Alderlake and not displaying 'TopdownL1 (cpu_atom)'. Remove the
need_full_name check as it is equivalent to the different PMU test in
the cases we care about, merge the 'if's and flip the evsel of the PMU
test. The 'if' is now basically saying, if the output matches the last
printed output then skip the output.
Ian Rogers [Tue, 11 Nov 2025 21:21:55 +0000 (13:21 -0800)]
perf stat: Remove hard coded shadow metrics
Now that the metrics are encoded in common json the hard coded
printing means the metrics are shown twice. Remove the hard coded
version.
This means that when specifying events, and those events correspond to
a hard coded metric, the metric will no longer be displayed. The
metric will be displayed if the metric is requested. Due to the adhoc
printing in the previous approach it was often found frustrating, the
new approach avoids this.
The default perf stat output on an alderlake now looks like:
```
$ perf stat -a -- sleep 1
Ian Rogers [Tue, 11 Nov 2025 21:21:54 +0000 (13:21 -0800)]
perf script: Change metric format to use json metrics
The metric format option isn't properly supported. This change
improves that by making the sample events update the counts of an
evsel, where the shadow metric code expects to read the values. To
support printing metrics, metrics need to be found. This is done on
the first attempt to print a metric. Every metric is parsed and then
the evsels in the metric's evlist compared to those in perf script
using the perf_event_attr type and config. If the metric matches then
it is added for printing. As an event in the perf script's evlist may
have >1 metric id, or different leader for aggregation, the first
metric matched will be displayed in those cases.
An example use is:
```
$ perf record -a -e '{instructions,cpu-cycles}:S' -a -- sleep 1
$ perf script -F period,metric
...
867817
metric: 0.30 insn per cycle
125394
metric: 0.04 insn per cycle
313516
metric: 0.11 insn per cycle
metric: 1.00 insn per cycle
```
Signed-off-by: Ian Rogers <irogers@google.com> Signed-off-by: Namhyung Kim <namhyung@kernel.org>
projects / thirdparty / kernel / linux.git / log
