Leo Yan [Tue, 4 Mar 2025 11:12:40 +0000 (11:12 +0000)]
perf arm-spe: Support previous branch target (PBT) address
When FEAT_SPE_PBT is implemented, the previous branch target address
(named as PBT) before the sampled operation, will be recorded.
This commit first introduces a 'prev_br_tgt' field in the record for
saving the PBT address in the decoder.
If the current operation is a branch instruction, by combining with PBT,
it can create a chain with two consecutive branches. As the branch
stack stores branches in descending order, meaning a newer branch is
stored in a lower entry in the stack. Arm SPE stores the latest branch
in the first entry of branch stack, and the previous branch coming from
PBT is stored into the second entry.
Otherwise, if current operation is not a branch, the last branch will be
saved for PBT only. PBT lacks associated information such as branch
source address, branch type, and events. The branch entry fills zeros
for the corresponding fields and only set its target address.
Leo Yan [Tue, 4 Mar 2025 11:12:39 +0000 (11:12 +0000)]
perf arm-spe: Add branch stack
Although Arm SPE cannot generate continuous branch records, this commit
creates a branch stack with only one branch entry. A single branch info
can be used for performance optimization.
A branch stack structure is dynamically allocated in the decode queue.
The branch stack and stack flags are synthesized based on branch types
and associated events.
Leo Yan [Tue, 4 Mar 2025 11:12:38 +0000 (11:12 +0000)]
perf arm-spe: Set sample flags with supplement info
Based on the supplement information in the record, this commit sets the
sample flags for conditional branch, function call, return. It also
sets events in flags, such as mispredict, not taken, and in transaction.
Leo Yan [Tue, 4 Mar 2025 11:12:35 +0000 (11:12 +0000)]
perf arm-spe: Extend branch operations
In Arm ARM (ARM DDI 0487, L.a), the section "D18.2.7 Operation Type
packet", the branch subclass is extended for Call Return (CR), Guarded
control stack data access (GCS).
This commit adds support CR and GCS operations. The IND (indirect)
operation is defined only in bit [1], its macro is updated accordingly.
Move the COND (Conditional) macro into the same group with other
operations for better maintenance.
Leo Yan [Tue, 4 Mar 2025 11:12:34 +0000 (11:12 +0000)]
perf arm-spe: Fix load-store operation checking
The ARM_SPE_OP_LD and ARM_SPE_OP_ST operations are secondary operation
type, they are overlapping with other second level's operation types
belonging to SVE and branch operations. As a result, a non load-store
operation can be parsed for data source and memory sample.
To fix the issue, this commit introduces a is_ldst_op() macro for
checking LDST operation, and apply the checking when synthesize data
source and memory samples.
Fixes: a89dbc9b988f ("perf arm-spe: Set sample's data source field") Signed-off-by: Leo Yan <leo.yan@arm.com> Reviewed-by: James Clark <james.clark@linaro.org> Link: https://lore.kernel.org/r/20250304111240.3378214-7-leo.yan@arm.com Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Leo Yan [Tue, 4 Mar 2025 11:12:31 +0000 (11:12 +0000)]
perf script: Separate events from branch types
Branch types and events are two different things. A branch type can be
a conditional branch, an indirect branch, a procedure call, a return, or
an exception taken, etc. The extra event information is provided for
what happens during a branch, e.g. if a branch is mispredicted or not
taken (specific to conditional branches).
To deliver information about branches, this commit separates events from
branch types. It parses branch types first, then appends event strings
embraced by the '/' character. If multiple events occur, the events is
separated with a comma (,).
Also add a minor improvement by adding char 'm' in char array for branch
mispredict event.
Leo Yan [Tue, 4 Mar 2025 11:12:30 +0000 (11:12 +0000)]
perf script: Refactor sample_flags_to_name() function
When generating a string for sample flags, the sample_flags_to_name()
function lacks the ability to parse the trace start bit or trace end bit.
Therefore, the function is invoked multiple times after clearing its
unsupported bits.
This commit improves the sample_flags_to_name() function to parse sample
flags in one go for three kinds of information:
- The prefix info for trace start, trace end, etc.
- Branch types.
- Extra info for transaction and interrupt related info.
As a result, the code is simplified to call the sample_flags_to_name()
only once. No expectation for any changes in the perf script output.
Leo Yan [Tue, 4 Mar 2025 11:12:29 +0000 (11:12 +0000)]
perf script: Make printing flags reliable
Add a check for the generated string of flags. Print out the raw number
if the string generation fails.
Use the SAMPLE_FLAGS_STR_ALIGNED_SIZE macro to replace the value '21'.
Reviewed-by: Ian Rogers <irogers@google.com> Reviewed-by: James Clark <james.clark@linaro.org> Signed-off-by: Leo Yan <leo.yan@arm.com> Reviewed-by: Adrian Hunter <adrian.hunter@intel.com> Link: https://lore.kernel.org/r/20250304111240.3378214-2-leo.yan@arm.com Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Legacy hybrid events have attr.type == PERF_TYPE_HARDWARE, so they look
like plain legacy events if we only look at attr.type. But legacy events
should still be uniquified if they were opened on a non-legacy PMU. Fix
it by checking if the evsel is hybrid and forcing needs_uniquify
before looking at the attr.type.
This restores PMU names on hybrid systems and also changes "perf stat
metrics (shadow stat) test" from a FAIL back to a SKIP (on hybrid). The
test was gated on "cycles" appearing alone which doesn't happen on
here.
Before:
$ perf stat -- true
...
<not counted> instructions:u (0.00%)
162,536 instructions:u # 0.58 insn per cycle
...
After:
$ perf stat -- true
...
<not counted> cpu_atom/instructions/u (0.00%)
162,541 cpu_core/instructions/u # 0.62 insn per cycle
...
Fixes: 357b965deba9 ("perf stat: Changes to event name uniquification") Suggested-by: Ian Rogers <irogers@google.com> Signed-off-by: James Clark <james.clark@linaro.org> Tested-by: Thomas Falcon <thomas.falcon@intel.com> Link: https://lore.kernel.org/r/20250226145526.632380-1-james.clark@linaro.org Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Namhyung Kim [Wed, 26 Feb 2025 20:30:39 +0000 (12:30 -0800)]
perf tools: Skip BPF sideband event for userspace profiling
The BPF sideband information is tracked using a separate thread and
evlist. But it's only useful for profiling kernel and we can skip it
when users profile their application only.
It seems it already fails to open the sideband event in that case.
Let's remove the noise in the verbose output anyway.
Luca Ceresoli [Fri, 24 Jan 2025 13:06:08 +0000 (14:06 +0100)]
perf build: Fix in-tree build due to symbolic link
Building perf in-tree is broken after commit 890a1961c812 ("perf tools:
Create source symlink in perf object dir") which added a 'source' symlink
in the output dir pointing to the source dir.
With in-tree builds, the added 'SOURCE = ...' line is executed multiple
times (I observed 2 during the build plus 2 during installation). This is a
minor inefficiency, in theory not harmful because symlink creation is
assumed to be idempotent. But it is not.
1. ln -sf $(srctree)/tools/perf $(OUTPUT)/source
-> creates /absolute/path/to/linux/tools/perf/source
link to /absolute/path/to/linux/tools/perf
=> OK, that's what was intended
2. ln -sf $(srctree)/tools/perf $(OUTPUT)/source # same command as 1
-> creates /absolute/path/to/linux/tools/perf/perf
link to /absolute/path/to/linux/tools/perf
=> Not what was intended, not idempotent
3. Now the build _should_ create the 'perf' executable, but it fails
The reason is the tricky 'ln' command line. At the first invocation 'ln'
uses the 1st form:
ln [OPTION]... [-T] TARGET LINK_NAME
and creates a link to TARGET *called LINK_NAME*.
At the second invocation $(OUTPUT)/source exists, so 'ln' uses the 3rd
form:
ln [OPTION]... TARGET... DIRECTORY
and creates a link to TARGET *called TARGET* inside DIRECTORY.
Fix by adding -n/--no-dereference to "treat LINK_NAME as a normal file
if it is a symbolic link to a directory", as the manpage says.
Closes: https://lore.kernel.org/all/20241125182506.38af9907@booty/ Fixes: 890a1961c812 ("perf tools: Create source symlink in perf object dir") Signed-off-by: Luca Ceresoli <luca.ceresoli@bootlin.com> Reviewed-by: Charlie Jenkins <charlie@rivosinc.com> Tested-by: Charlie Jenkins <charlie@rivosinc.com> Link: https://lore.kernel.org/r/20250124-perf-fix-intree-build-v1-1-485dd7a855e4@bootlin.com Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Ian Rogers [Tue, 25 Feb 2025 19:36:00 +0000 (11:36 -0800)]
tools/x86: Fix linux/unaligned.h include path in lib/insn.c
tools/arch/x86/include/linux doesn't exist but building is working by
virtue of a -I. Building using bazel this fails. Use angle brackets to
include unaligned.h so there isn't an invalid relative include.
Fixes: 5f60d5f6bbc1 ("move asm/unaligned.h to linux/unaligned.h") Signed-off-by: Ian Rogers <irogers@google.com> Acked-by: Josh Poimboeuf <jpoimboe@kernel.org> Cc: Al Viro <viro@zeniv.linux.org.uk> Link: https://lore.kernel.org/r/20250225193600.90037-1-irogers@google.com Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Leo Yan [Thu, 27 Feb 2025 08:55:44 +0000 (08:55 +0000)]
perf arm-spe: Report error if set frequency
When users set the parameter '-F' to specify frequency for Arm SPE, the
tool reports error:
perf record -F 1000 -e arm_spe_0// -- sleep 1
Error:
Invalid event (arm_spe_0//) in per-thread mode, enable system wide with '-a'.
The output logs are confused and it does not give the correct reminding.
Arm SPE does not support frequency setting given it adopts a statistical
based approach.
Alternatively, Arm SPE supports setting period. This commit adds a
for frequency setting. It reports error and reminds users to set period
instead.
After:
perf record -F 1000 -e arm_spe_0// -- sleep 1
Arm SPE: Frequency is not supported. Set period with -c option or PMU parameter (-e arm_spe_0/period=NUM/).
Chun-Tse Shao [Thu, 27 Feb 2025 00:28:56 +0000 (16:28 -0800)]
perf lock: Report owner stack in usermode
This patch parses `owner_lock_stat` into a RB tree, enabling ordered
reporting of owner lock statistics with stack traces. It also updates
the documentation for the `-o` option in contention mode, decouples `-o`
from `-t`, and issues a warning to inform users about the new behavior
of `-ov`.
Example output:
$ sudo ~/linux/tools/perf/perf lock con -abvo -Y mutex-spin -E3 perf bench sched pipe
...
contended total wait max wait avg wait type caller
171 1.55 ms 20.26 us 9.06 us mutex pipe_read+0x57
0xffffffffac6318e7 pipe_read+0x57
0xffffffffac623862 vfs_read+0x332
0xffffffffac62434b ksys_read+0xbb
0xfffffffface604b2 do_syscall_64+0x82
0xffffffffad00012f entry_SYSCALL_64_after_hwframe+0x76
36 193.71 us 15.27 us 5.38 us mutex pipe_write+0x50
0xffffffffac631ee0 pipe_write+0x50
0xffffffffac6241db vfs_write+0x3bb
0xffffffffac6244ab ksys_write+0xbb
0xfffffffface604b2 do_syscall_64+0x82
0xffffffffad00012f entry_SYSCALL_64_after_hwframe+0x76
4 51.22 us 16.47 us 12.80 us mutex do_epoll_wait+0x24d
0xffffffffac691f0d do_epoll_wait+0x24d
0xffffffffac69249b do_epoll_pwait.part.0+0xb
0xffffffffac693ba5 __x64_sys_epoll_pwait+0x95
0xfffffffface604b2 do_syscall_64+0x82
0xffffffffad00012f entry_SYSCALL_64_after_hwframe+0x76
=== owner stack trace ===
3 31.24 us 15.27 us 10.41 us mutex pipe_read+0x348
0xffffffffac631bd8 pipe_read+0x348
0xffffffffac623862 vfs_read+0x332
0xffffffffac62434b ksys_read+0xbb
0xfffffffface604b2 do_syscall_64+0x82
0xffffffffad00012f entry_SYSCALL_64_after_hwframe+0x76
...
Chun-Tse Shao [Thu, 27 Feb 2025 00:28:54 +0000 (16:28 -0800)]
perf lock: Retrieve owner callstack in bpf program
This implements per-callstack aggregation of lock owners in addition to
per-thread. The owner callstack is captured using `bpf_get_task_stack()`
at `contention_begin()` and it also adds a custom stackid function for the
owner stacks to be compared easily.
The owner info is kept in a hash map using lock addr as a key to handle
multiple waiters for the same lock. At `contention_end()`, it updates the
owner lock stat based on the info that was saved at `contention_begin()`.
If there are more waiters, it'd update the owner pid to itself as
`contention_end()` means it gets the lock now. But it also needs to check
the return value of the lock function in case task was killed by a signal
or something.
Chun-Tse Shao [Thu, 27 Feb 2025 00:28:53 +0000 (16:28 -0800)]
perf lock: Add bpf maps for owner stack tracing
Add a struct and few bpf maps in order to tracing owner stack.
`struct owner_tracing_data`: Contains owner's pid, stack id, timestamp for
when the owner acquires lock, and the count of lock waiters.
`stack_buf`: Percpu buffer for retrieving owner stacktrace.
`owner_stacks`: For tracing owner stacktrace to customized owner stack id.
`owner_data`: For tracing lock_address to `struct owner_tracing_data` in
bpf program.
`owner_stat`: For reporting owner stacktrace in usermode.
Ian Rogers [Mon, 10 Feb 2025 19:12:31 +0000 (11:12 -0800)]
perf cpumap: Reduce cpu size from int to int16_t
Fewer than 32k logical CPUs are currently supported by perf. A cpumap
is indexed by an integer (see perf_cpu_map__cpu) yielding a perf_cpu
that wraps a 4-byte int for the logical CPU - the wrapping is done
deliberately to avoid confusing a logical CPU with an index into a
cpumap. Using a 4-byte int within the perf_cpu is larger than required
so this patch reduces it to the 2-byte int16_t. For a cpumap
containing 16 entries this will reduce the array size from 64 to 32
bytes. For very large servers with lots of logical CPUs the size
savings will be greater.
Backtrace pointed to :
?? ()
perf_session.process_user_event ()
reader.read_event ()
perf_session.process_events ()
cmd_trace ()
run_builtin ()
handle_internal_command ()
main ()
Further debug pointed that, segmentation fault happens when
trying to access id_index. Code snippet:
case PERF_RECORD_ID_INDEX:
err = tool->id_index(session, event);
Since 'commit 15d4a6f41d72 ("perf tool: Remove
perf_tool__fill_defaults()")', perf_tool__fill_defaults is
removed. All tools are initialized using perf_tool__init()
prior to use. But in builtin-trace, perf_tool__init is not
used and hence the defaults are not initialized. Use
perf_tool__init() in perf trace to handle the initialization.
James Clark [Wed, 26 Feb 2025 10:41:01 +0000 (10:41 +0000)]
perf pmu: Don't double count common sysfs and json events
After pmu_add_cpu_aliases() is called, perf_pmu__num_events() returns an
incorrect value that double counts common events and doesn't match the
actual count of events in the alias list. This is because after
'cpu_aliases_added == true', the number of events returned is
'sysfs_aliases + cpu_json_aliases'. But when adding 'case
EVENT_SRC_SYSFS' events, 'sysfs_aliases' and 'cpu_json_aliases' are both
incremented together, failing to account that these ones overlap and
only add a single item to the list. Fix it by adding another counter for
overlapping events which doesn't influence 'cpu_json_aliases'.
There doesn't seem to be a current issue because it's used in perf list
before pmu_add_cpu_aliases() so the correct value is returned. Other
uses in tests may also miss it for other reasons like only looking at
uncore events. However it's marked as a fixes commit in case any new fix
with new uses of perf_pmu__num_events() is backported.
Fixes: d9c5f5f94c2d ("perf pmu: Count sys and cpuid JSON events separately") Reviewed-by: Ian Rogers <irogers@google.com> Signed-off-by: James Clark <james.clark@linaro.org> Link: https://lore.kernel.org/r/20250226104111.564443-3-james.clark@linaro.org Signed-off-by: Namhyung Kim <namhyung@kernel.org>
James Clark [Wed, 26 Feb 2025 10:41:00 +0000 (10:41 +0000)]
perf pmu: Dynamically allocate tool PMU
perf_pmus__destroy() treats all PMUs as allocated and free's them so we
can't have any static PMUs that are added to the PMU lists. Fix it by
allocating the tool PMU in the same way as the others. Current users of
the tool PMU already use find_pmu() and not perf_pmus__tool_pmu(), so
rename the function to add 'new' to avoid it being misused in the
future.
perf_pmus__fake_pmu() can remain as static as it's not added to the
PMU lists.
Fixes the following error:
$ perf bench internals pmu-scan
# Running 'internals/pmu-scan' benchmark:
Computing performance of sysfs PMU event scan for 100 times
munmap_chunk(): invalid pointer
Aborted (core dumped)
Fixes: 240505b2d0ad ("perf tool_pmu: Factor tool events into their own PMU") Reviewed-by: Ian Rogers <irogers@google.com> Signed-off-by: James Clark <james.clark@linaro.org> Link: https://lore.kernel.org/r/20250226104111.564443-2-james.clark@linaro.org Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Code in synthesize_probe_trace_arg() access a null value and results in
segfault. Data structure which is null:
struct probe_trace_arg arg->value
We are hitting a case where arg->value is null in probe point:
"vfs_fstatat $params". This is happening since 'commit e896474fe485
("getname_maybe_null() - the third variant of pathname copy-in")'
Before the commit, probe point for vfs_fstatat was getting added only
for one location:
With this change, vfs_fstatat code is inlined for other locations in the
code:
Probe point found: __do_sys_lstat64+48
Probe point found: __do_sys_stat64+48
Probe point found: __do_sys_newlstat+48
Probe point found: __do_sys_newstat+48
Probe point found: vfs_fstatat+0
When trying to find matching dwarf information entry (DIE)
from the debuginfo, the code incorrectly picks DIE which is
not referring to vfs_fstatat. Snippet from dwarf entry in vmlinux
debuginfo file.
The main abstract die is:
<1><4214883>: Abbrev Number: 147 (DW_TAG_subprogram)
<4214885> DW_AT_external : 1
<4214885> DW_AT_name : (indirect string, offset: 0x17b9f3): vfs_fstatat
While collecting variables/parameters for a probe point, the function
copy_variables_cb() also looks at dwarf debug entries based on the
instruction address. Snippet
if (dwarf_haspc(die_mem, vf->pf->addr))
return DIE_FIND_CB_CONTINUE;
else
return DIE_FIND_CB_SIBLING;
But incase of inlined function instance for vfs_fstatat, there are two
entries which has the instruction address entry point as same.
Instance 1: which is for vfs_fstatat and DW_AT_abstract_origin points to
0x4214883 (reference above for main abstract die)
But the copy_variables_cb() continues to add parameters from second
instance also based on the dwarf_haspc() check. This results in
formal parameters for getname also appended to params. But while
filling in the args->value for these parameters, since these args
are not part of dwarf with offset "42131fa". Hence value will be
null. This incorrect args results in segfault when value field is
accessed.
Save the dwarf dieoffset of the actual DW_TAG_subprogram as part of
"struct probe_finder". In copy_variables_cb(), include check to make
sure the DW_AT_abstract_origin points to the correct entry if the
dwarf_haspc() matches the instruction address.
Gabriele Monaco [Fri, 7 Feb 2025 08:04:45 +0000 (09:04 +0100)]
perf ftrace latency: allow to hide empty buckets
Especially while using several buckets, it isn't uncommon to have some
of them empty and reading the histogram may be a bit more complex:
# perf ftrace latency -a -T mutex_lock --bucket-range 5 --max-latency 200
# DURATION | COUNT | GRAPH |
0 - 5 us | 14816 | ###################################### |
5 - 10 us | 1228 | ### |
10 - 15 us | 438 | # |
15 - 20 us | 106 | |
20 - 25 us | 21 | |
25 - 30 us | 11 | |
30 - 35 us | 1 | |
35 - 40 us | 2 | |
40 - 45 us | 4 | |
45 - 50 us | 0 | |
50 - 55 us | 1 | |
55 - 60 us | 0 | |
60 - 65 us | 1 | |
65 - 70 us | 1 | |
70 - 75 us | 1 | |
75 - 80 us | 2 | |
80 - 85 us | 0 | |
85 - 90 us | 1 | |
90 - 95 us | 0 | |
95 - 100 us | 1 | |
100 - 105 us | 0 | |
105 - 110 us | 0 | |
110 - 115 us | 0 | |
115 - 120 us | 0 | |
120 - 125 us | 1 | |
125 - 130 us | 0 | |
130 - 135 us | 0 | |
135 - 140 us | 1 | |
140 - 145 us | 0 | |
145 - 150 us | 0 | |
150 - 155 us | 0 | |
155 - 160 us | 0 | |
160 - 165 us | 0 | |
165 - 170 us | 0 | |
170 - 175 us | 0 | |
175 - 180 us | 0 | |
180 - 185 us | 0 | |
185 - 190 us | 0 | |
190 - 195 us | 0 | |
195 - 200 us | 0 | |
200 - ... us | 2 | |
Allow the optional flag --hide-empty to remove buckets with no element
and produce a more compact graph. This feature could be misleading since
there is no clear indication for missing buckets, for this reason it's
disabled by default.
# perf ftrace latency -a -T mutex_lock --bucket-range 5 --max-latency --hide-empty 200
# DURATION | COUNT | GRAPH |
0 - 5 us | 14816 | ###################################### |
5 - 10 us | 1228 | ### |
10 - 15 us | 438 | # |
15 - 20 us | 106 | |
20 - 25 us | 21 | |
25 - 30 us | 11 | |
30 - 35 us | 1 | |
35 - 40 us | 2 | |
40 - 45 us | 4 | |
50 - 55 us | 1 | |
60 - 65 us | 1 | |
65 - 70 us | 1 | |
70 - 75 us | 1 | |
75 - 80 us | 2 | |
85 - 90 us | 1 | |
95 - 100 us | 1 | |
120 - 125 us | 1 | |
135 - 140 us | 1 | |
200 - ... us | 2 | |
Gabriele Monaco [Fri, 7 Feb 2025 08:04:44 +0000 (09:04 +0100)]
perf ftrace latency: variable histogram buckets
The max-latency value can make the histogram smaller, but not larger, we
have a maximum of 22 buckets and specifying a max-latency that would
require more buckets has no effect.
Dynamically allocate the buckets and compute the bucket number from the
max latency as (max-min) / range + 2
If the maximum is not specified, we still set the bucket number to 22
and compute the maximum accordingly.
Fail if the maximum is smaller than min+range, this way we make sure we
always have 3 buckets: those below min, those above max and one in the
middle.
Since max-latency is not available in log2 mode, always use 22 buckets.
Namhyung Kim [Sun, 26 Jan 2025 21:02:42 +0000 (13:02 -0800)]
perf annotate-data: Handle direct use of stack pointer without fbreg
Sometimes compiler generates code to use the stack pointer register
without frame pointer. As we know RSP is the stack register on x86,
let's treat it as same as fbreg. But the offset would be opposite
direction so update the debug message accordingly.
Thomas Falcon [Thu, 20 Feb 2025 04:59:42 +0000 (22:59 -0600)]
perf report: Fix sample number stats for branch entry mode
Currently, stats->nr_samples is incremented per entry in the branch stack
instead of per sample taken. As a result, statistics of samples taken
during perf record in --branch-filter or --branch-any mode does not
seem correct. Instead call hists__inc_nr_samples() for each sample taken
instead of for each entry in the branch stack.
Ian Rogers [Sat, 22 Feb 2025 06:10:13 +0000 (22:10 -0800)]
perf machine: Reuse module path buffer
Rather than copying the path and appending the directory entry in a
fresh path buffer, append to the path at the end of where it is for
the recursion level. This saves a PATH_MAX buffer per recursion level
and some unnecessary copying.
Ian Rogers [Sat, 22 Feb 2025 06:10:09 +0000 (22:10 -0800)]
perf header: Switch mem topology to io_dir__readdir
Switch memory_node__read and build_mem_topology from opendir/readdir
to io_dir__readdir, with smaller stack allocations. Reduces peak
memory consumption of perf record by 10kb.
Ian Rogers [Sat, 22 Feb 2025 06:10:06 +0000 (22:10 -0800)]
tools lib api: Add io_dir an allocation free readdir alternative
glibc's opendir allocates a minimum of 32kb, when called recursively
for a directory tree the memory consumption can add up - nearly 300kb
during perf start-up when processing modules. Add a stack allocated
variant of readdir sized a little more than 1kb.
As getdents64 may be missing from libc, add support using syscall. As
the system call number maybe missing, add #defines for those.
Note, an earlier version of this patch had a feature test for
getdents64 but there were problems on certains distros where
getdents64 would be #define renamed to getdents breaking the code. The
syscall use was made uncondtional to work around this. There is
context in:
https://lore.kernel.org/lkml/20231207050433.1426834-1-irogers@google.com/
Ian Rogers [Thu, 9 Jan 2025 17:54:01 +0000 (09:54 -0800)]
perf parse-events: Tidy name token matching
Prior to commit 70c90e4a6b2f ("perf parse-events: Avoid scanning PMUs
before parsing") names (generally event names) excluded hyphen (minus)
symbols as the formation of legacy names with hyphens was handled in
the yacc code. That commit allowed hyphens supposedly making
name_minus unnecessary. However, changing name_minus to name has
issues in the term config tokens as then name ends up having priority
over numbers and name allows matching numbers since commit 5ceb57990bf4 ("perf parse: Allow tracepoint names to start with digits
"). It is also permissable for a name to match with a colon (':') in
it when its in a config term list. To address this rename name_minus
to term_name, make the pattern match name's except for the colon, add
number matching into the config term region with a higher priority
than name matching. This addresses an inconsistency and allows greater
matching for names inside of term lists, for example, they may start
with a number.
Rename name_tag to quoted_name and update comments and helper
functions to avoid str detecting quoted strings which was already done
by the lexer.
Krzysztof Łopatowski [Thu, 6 Feb 2025 11:33:15 +0000 (12:33 +0100)]
perf tools: Improve startup time by reducing unnecessary stat() calls
When testing perf trace on NixOS, I noticed significant startup delays:
- `ls`: ~2ms
- `strace ls`: ~10ms
- `perf trace ls`: ~550ms
Profiling showed that 51% of the time is spent reading files,
26% in loading BPF programs, and 11% in `newfstatat`.
This patch optimizes module path exploration by avoiding `stat()` calls
unless necessary. For filesystems that do not implement `d_type`
(DT_UNKNOWN), it falls back to the old behavior.
See `readdir(3)` for details.
This reduces `perf trace ls` time to ~500ms.
A more thorough startup optimization based on command parameters would
be ideal, but that is a larger effort.
Dmitry Vyukov [Wed, 8 Jan 2025 06:36:23 +0000 (07:36 +0100)]
perf report: Fix input reload/switch with symbol sort key
Currently the code checks that there is no "ipc" in the sort order
and add an ipc string. This will always error out on the second pass
after input reload/switch, since the sort order already contains "ipc".
Do the ipc check/fixup only on the first pass.
Namhyung Kim [Tue, 11 Feb 2025 06:07:45 +0000 (22:07 -0800)]
perf report: Support switching data w/ and w/o callchains
The symbol_conf.use_callchain should be reset when switching to new data
file, otherwise report__setup_sample_type() will show an error message
that it enabled callchains but no callchain data. The function also
will turn on the callchains if the data has PERF_SAMPLE_CALLCHAIN so I
think it's ok to reset symbol_conf.use_callchain here.
Namhyung Kim [Tue, 11 Feb 2025 06:07:44 +0000 (22:07 -0800)]
perf report: Switch data file correctly in TUI
The 's' key is to switch to a new data file and load the data in the
same window. The switch_data_file() will show a popup menu to select
which data file user wants and update the 'input_name' global variable.
But in the cmd_report(), it didn't update the data.path using the new
'input_name' and keep usng the old file. This is fairly an old bug and
I assume people don't use this feature much. :)
Link: https://lore.kernel.org/r/20250211060745.294289-1-namhyung@kernel.org Closes: https://lore.kernel.org/linux-perf-users/89e678bc-f0af-4929-a8a6-a2666f1294a4@linaro.org Fixes: f5fc14124c5cefdd ("perf tools: Add data object to handle perf data file") Reported-by: James Clark <james.clark@linaro.org> Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Greg Kroah-Hartman [Wed, 19 Feb 2025 13:40:56 +0000 (14:40 +0100)]
perf tools: Fix up some comments and code to properly use the event_source bus
In sysfs, the perf events are all located in
/sys/bus/event_source/devices/ but some places ended up hard-coding the
location to be at the root of /sys/devices/ which could be very risky as
you do not exactly know what type of device you are accessing in sysfs
at that location.
So fix this all up by properly pointing everything at the bus device
list instead of the root of the sysfs devices/ tree.
Cc: stable <stable@kernel.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Reviewed-by: Kan Liang <kan.liang@linux.intel.com> Link: https://lore.kernel.org/r/2025021955-implant-excavator-179d@gregkh Signed-off-by: Namhyung Kim <namhyung@kernel.org>
James Clark [Wed, 19 Feb 2025 15:16:21 +0000 (15:16 +0000)]
perf list: Also append PMU name in verbose mode
When listing in verbose mode, the long description is used but the PMU
name isn't appended. There doesn't seem to be a reason to exclude it
when asking for more information, so use the same print block for both
long and short descriptions.
Before:
$ perf list -v
...
inst_retired
[Instruction architecturally executed]
Yangyu Chen [Thu, 13 Feb 2025 08:44:09 +0000 (16:44 +0800)]
perf vendor events arm64: Fix incorrect CPU_CYCLE in metrics expr
Some existing metrics for Neoverse N3 and V3 expressions use CPU_CYCLE
to represent the number of cycles, but this is incorrect. The correct
event to use is CPU_CYCLES.
I encountered this issue while working on a patch to add pmu events for
Cortex A720 and A520 by reusing the existing patch for Neoverse N3 and
V3 by James Clark [1] and my check script [2] reported this issue.
Namhyung Kim [Wed, 19 Feb 2025 00:05:58 +0000 (16:05 -0800)]
perf script: Fix hangup in offline flamegraph report
A recent change in the flamegraph script fixed an issue with live mode
but it created another for offline mode. It needs to pass "-" to -i
option to read from stdin in the live mode. Actually there's a logic
to pass the option in the perf script code, but the script was written
with "-- $@" which prevented the option to go to the perf script. So
the previous commit added the hard-coded "-i -" to the report command.
But it's a problem for the offline mode which expects input from a file
and now it's stuck on reading from stdin. Let's remove the "-i - --"
part and let it pass the options properly to perf script.
Closes: https://lore.kernel.org/linux-perf-users/c41e4b04-e1fd-45ab-80b0-ec2ac6e94310@linux.ibm.com Fixes: 23e0a63c6dd3f69c ("perf script: force stdin for flamegraph in live mode") Reported-by: Thomas Richter <tmricht@linux.ibm.com> Tested-by: Thomas Richter <tmricht@linux.ibm.com> Cc: Anubhav Shelat <ashelat@redhat.com> Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Dmitry Vyukov [Thu, 13 Feb 2025 09:08:19 +0000 (10:08 +0100)]
perf report: Add --latency flag
Add record/report --latency flag that allows to capture and show
latency-centric profiles rather than the default CPU-consumption-centric
profiles. For latency profiles record captures context switch events,
and report shows Latency as the first column.
Dmitry Vyukov [Thu, 13 Feb 2025 09:08:18 +0000 (10:08 +0100)]
perf report: Add latency output field
Latency output field is similar to overhead, but represents overhead for
latency rather than CPU consumption. It's re-scaled from overhead by dividing
weight by the current parallelism level at the time of the sample.
It effectively models profiling with 1 sample taken per unit of wall-clock
time rather than unit of CPU time.
Dmitry Vyukov [Thu, 13 Feb 2025 09:08:16 +0000 (10:08 +0100)]
perf report: Switch filtered from u8 to u16
We already have all u8 bits taken, adding one more filter leads to unpleasant
failure mode, where code compiles w/o warnings, but the last filters silently
don't work. Add a typedef and switch to u16.
Dmitry Vyukov [Thu, 13 Feb 2025 09:08:14 +0000 (10:08 +0100)]
perf report: Add machine parallelism
Add calculation of the current parallelism level (number of threads actively
running on CPUs). The parallelism level can be shown in reports on its own,
and to calculate latency overheads.
Namhyung Kim [Fri, 14 Feb 2025 19:16:41 +0000 (11:16 -0800)]
perf tools: Fix compile error on sample->user_regs
It's recently changed to allocate dynamically but misses to update some
arch-dependent codes to use perf_sample__user_regs().
Fixes: dc6d2bc2d893a878 ("perf sample: Make user_regs and intr_regs optional") Reported-by: Stephen Rothwell <sfr@canb.auug.org.au> Reviewed-by: Ian Rogers <irogers@google.com> Link: https://lore.kernel.org/r/20250214191641.756664-1-namhyung@kernel.org Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Leo Yan [Fri, 14 Feb 2025 11:10:25 +0000 (11:10 +0000)]
perf tools: Fix compilation error on arm64
Since the commit dc6d2bc2d893 ("perf sample: Make user_regs and
intr_regs optional"), the building for Arm64 reports error:
arch/arm64/util/unwind-libdw.c: In function ‘libdw__arch_set_initial_registers’:
arch/arm64/util/unwind-libdw.c:11:32: error: initialization of ‘struct regs_dump *’ from incompatible pointer type ‘struct regs_dump **’ [-Werror=incompatible-pointer-types]
11 | struct regs_dump *user_regs = &ui->sample->user_regs;
| ^
cc1: all warnings being treated as errors
make[6]: *** [/home/niayan01/linux/tools/build/Makefile.build:85: arch/arm64/util/unwind-libdw.o] Error 1
make[5]: *** [/home/niayan01/linux/tools/build/Makefile.build:138: util] Error 2
arch/arm64/tests/dwarf-unwind.c: In function ‘test__arch_unwind_sample’:
arch/arm64/tests/dwarf-unwind.c:48:27: error: initialization of ‘struct regs_dump *’ from incompatible pointer type ‘struct regs_dump **’ [-Werror=incompatible-pointer-types]
48 | struct regs_dump *regs = &sample->user_regs;
| ^
To fix the issue, use the helper perf_sample__user_regs() to retrieve
the user_regs.
Fixes: dc6d2bc2d893 ("perf sample: Make user_regs and intr_regs optional") Signed-off-by: Leo Yan <leo.yan@arm.com> Reviewed-by: James Clark <james.clark@linaro.org> Link: https://lore.kernel.org/r/20250214111025.14478-1-leo.yan@arm.com Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Ian Rogers [Mon, 6 Jan 2025 21:54:42 +0000 (13:54 -0800)]
tools build: Fix a number of Wconversion warnings
There's some expressed interest in having the compiler flag
-Wconversion detect at build time certain kinds of potential problems:
https://lore.kernel.org/lkml/20250103182532.GB781381@e132581.arm.com/
As feature detection passes -Wconversion from CFLAGS when set, the
feature detection compile tests need to not fail because of
-Wconversion as the failure will be interpretted as a missing
feature. Switch various types to avoid the -Wconversion issue, the
exact meaning of the code is unimportant as it is typically looking
for header file definitions.
Ian Rogers [Mon, 13 Jan 2025 19:43:45 +0000 (11:43 -0800)]
perf sample: Make user_regs and intr_regs optional
The struct dump_regs contains 512 bytes of cache_regs, meaning the two
values in perf_sample contribute 1088 bytes of its total 1384 bytes
size. Initializing this much memory has a cost reported by Tavian
Barnes <tavianator@tavianator.com> as about 2.5% when running `perf
script --itrace=i0`:
https://lore.kernel.org/lkml/d841b97b3ad2ca8bcab07e4293375fb7c32dfce7.1736618095.git.tavianator@tavianator.com/
Adrian Hunter <adrian.hunter@intel.com> replied that the zero
initialization was necessary and couldn't simply be removed.
This patch aims to strike a middle ground of still zeroing the
perf_sample, but removing 79% of its size by make user_regs and
intr_regs optional pointers to zalloc-ed memory. To support the
allocation accessors are created for user_regs and intr_regs. To
support correct cleanup perf_sample__init and perf_sample__exit
functions are created and added throughout the code base.
Ian Rogers [Tue, 11 Feb 2025 21:30:31 +0000 (13:30 -0800)]
perf test stat_all_metrics: Ensure missing events fail test
Issue reported by Thomas Falcon and diagnosed by Kan Liang here:
https://lore.kernel.org/lkml/d44036481022c27d83ce0faf8c7f77042baedb34.camel@intel.com/
Metrics with missing events can be erroneously skipped if they contain
FP, AMX or PMM events.
Update uncore IIO events umask with the change:
https://github.com/intel/perfmon/commit/d78e8a166537c9ceab4f2e901dc96c53667a2174
which should address an issue originally raised by Michael Petlan: Reported-by: Michael Petlan <mpetlan@redhat.com> Closes: https://lore.kernel.org/all/alpine.LRH.2.20.2401300733310.11354@Diego/ Co-developed-by: Caleb Biggers <caleb.biggers@intel.com> Signed-off-by: Caleb Biggers <caleb.biggers@intel.com> Signed-off-by: Ian Rogers <irogers@google.com> Acked-by: Kan Liang <kan.liang@linux.intel.com> Tested-by: Thomas Falcon <thomas.falcon@intel.com> Link: https://lore.kernel.org/r/20250211213031.114209-21-irogers@google.com Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Update uncore IIO events umask with the change:
https://github.com/intel/perfmon/commit/d78e8a166537c9ceab4f2e901dc96c53667a2174
which should address an issue originally raised by Michael Petlan: Reported-by: Michael Petlan <mpetlan@redhat.com> Closes: https://lore.kernel.org/all/alpine.LRH.2.20.2401300733310.11354@Diego/ Co-developed-by: Caleb Biggers <caleb.biggers@intel.com> Signed-off-by: Caleb Biggers <caleb.biggers@intel.com> Acked-by: Kan Liang <kan.liang@linux.intel.com> Signed-off-by: Ian Rogers <irogers@google.com> Tested-by: Thomas Falcon <thomas.falcon@intel.com> Link: https://lore.kernel.org/r/20250211213031.114209-20-irogers@google.com Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Update uncore IIO events umask with the change:
https://github.com/intel/perfmon/commit/d78e8a166537c9ceab4f2e901dc96c53667a2174
which should address an issue originally raised by Michael Petlan: Reported-by: Michael Petlan <mpetlan@redhat.com> Closes: https://lore.kernel.org/all/alpine.LRH.2.20.2401300733310.11354@Diego/ Co-developed-by: Caleb Biggers <caleb.biggers@intel.com> Signed-off-by: Caleb Biggers <caleb.biggers@intel.com> Signed-off-by: Ian Rogers <irogers@google.com> Acked-by: Kan Liang <kan.liang@linux.intel.com> Tested-by: Thomas Falcon <thomas.falcon@intel.com> Link: https://lore.kernel.org/r/20250211213031.114209-12-irogers@google.com Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Update uncore IIO events umask with the change:
https://github.com/intel/perfmon/commit/d78e8a166537c9ceab4f2e901dc96c53667a2174
which should address an issue originally raised by Michael Petlan: Reported-by: Michael Petlan <mpetlan@redhat.com> Closes: https://lore.kernel.org/all/alpine.LRH.2.20.2401300733310.11354@Diego/ Co-developed-by: Caleb Biggers <caleb.biggers@intel.com> Signed-off-by: Caleb Biggers <caleb.biggers@intel.com> Acked-by: Kan Liang <kan.liang@linux.intel.com> Signed-off-by: Ian Rogers <irogers@google.com> Tested-by: Thomas Falcon <thomas.falcon@intel.com> Link: https://lore.kernel.org/r/20250211213031.114209-11-irogers@google.com Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Update uncore IIO events umask with the change:
https://github.com/intel/perfmon/commit/d78e8a166537c9ceab4f2e901dc96c53667a2174
which should address an issue originally raised by Michael Petlan: Reported-by: Michael Petlan <mpetlan@redhat.com> Closes: https://lore.kernel.org/all/alpine.LRH.2.20.2401300733310.11354@Diego/ Co-developed-by: Caleb Biggers <caleb.biggers@intel.com> Signed-off-by: Caleb Biggers <caleb.biggers@intel.com> Acked-by: Kan Liang <kan.liang@linux.intel.com> Signed-off-by: Ian Rogers <irogers@google.com> Tested-by: Thomas Falcon <thomas.falcon@intel.com> Link: https://lore.kernel.org/r/20250211213031.114209-10-irogers@google.com Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Namhyung Kim [Wed, 12 Feb 2025 22:14:45 +0000 (14:14 -0800)]
perf tools: Use symfs when opening debuginfo by path
I found that it failed to load a binary using --symfs option. Say I
have a binary in /home/user/prog/xxx and a perf data file with it. If I
move them to a different machine and use --symfs, it tries to find the
binary in some locations under symfs using dso__read_binary_type_filename(),
but not the last one.
Namhyung Kim [Wed, 5 Feb 2025 20:54:43 +0000 (12:54 -0800)]
perf trace: Add --summary-mode option
The --summary-mode option will select how to show the syscall summary at
the end. By default, it'll show the summary for each thread and it's
the same as if --summary-mode=thread is passed.
The other option is to show total summary, which is --summary-mode=total.
I'd like to have this instead of a separate option like --total-summary
because we may want to add a new summary mode (by cgroup) later.
Namhyung Kim [Wed, 5 Feb 2025 20:54:41 +0000 (12:54 -0800)]
perf trace: Convert syscall_stats to hashmap
It was using a RBtree-based int-list as a hash and a custom resort
logic for that. As we have hashmap, let's convert to it and add a
custom sort function for the hashmap entries using an array. It
should be faster and more light-weighted. It's also to prepare
supporting system-wide syscall stats.
No functional changes intended.
Reviewed-by: Howard Chu <howardchu95@gmail.com> Acked-by: Arnaldo Carvalho de Melo <acme@redhat.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Link: https://lore.kernel.org/r/20250205205443.1986408-3-namhyung@kernel.org Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Namhyung Kim [Wed, 5 Feb 2025 20:54:40 +0000 (12:54 -0800)]
perf trace: Allocate syscall stats only if summary is on
The syscall stats are used only when summary is requested. Let's avoid
unnecessary operations. While at it, let's pass 'trace' pointer
directly instead of passing 'output' file pointer and 'summary' option
in the 'trace' separately.
Reviewed-by: Howard Chu <howardchu95@gmail.com> Acked-by: Arnaldo Carvalho de Melo <acme@redhat.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Link: https://lore.kernel.org/r/20250205205443.1986408-2-namhyung@kernel.org Signed-off-by: Namhyung Kim <namhyung@kernel.org>
James Clark [Wed, 12 Feb 2025 16:38:56 +0000 (16:38 +0000)]
perf tests: Fix Tool PMU test segfault
tool_pmu__event_to_str() now handles skipped events by returning NULL,
so it's wrong to re-check for a skip on the resulting string. Calling
tool_pmu__skip_event() with a NULL string results in a segfault so
remove the unnecessary skip to fix it:
$ perf test -vv "parsing with PMU name"
12.2: Parsing with PMU name:
...
---- unexpected signal (11) ----
12.2: Parsing with PMU name : FAILED!
Fixes: ee8aef2d2321 ("perf tools: Add skip check in tool_pmu__event_to_str()") Signed-off-by: James Clark <james.clark@linaro.org> Reported-by: Athira Rajeev <atrajeev@linux.vnet.ibm.com> Acked-by: Kan Liang <kan.liang@linux.intel.com> Tested-by: Ian Rogers <irogers@google.com> Link: https://lore.kernel.org/r/20250212163859.1489916-1-james.clark@linaro.org Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Kan Liang [Fri, 7 Feb 2025 15:28:44 +0000 (07:28 -0800)]
perf tools: Add skip check in tool_pmu__event_to_str()
Some topdown related metrics may fail on hybrid machines.
$ perf stat -M tma_frontend_bound
Cannot resolve IDs for tma_frontend_bound:
cpu_atom@TOPDOWN_FE_BOUND.ALL@ / (8 * cpu_atom@CPU_CLK_UNHALTED.CORE@)
In the find_tool_events(), the tool_pmu__event_to_str() is used to
compare the tool_events. It only checks the event name, no PMU or arch.
So the tool_events[TOOL_PMU__EVENT_SLOTS] is set to true, because the
p-core Topdown metrics has "slots" event.
The tool_events is shared. So when parsing the e-core metrics, the
"slots" is automatically added.
The "slots" event as a tool event should only be available on arm64. It
has a different meaning on X86. The tool_pmu__skip_event() intends
handle the case. Apply it for tool_pmu__event_to_str() as well.
There is a lack of sanity check in the expr__get_id(). Add the check.
Closes: https://lore.kernel.org/lkml/608077bc-4139-4a97-8dc4-7997177d95c4@linux.intel.com/ Fixes: 069057239a67 ("perf tool_pmu: Move expr literals to tool_pmu") Signed-off-by: Kan Liang <kan.liang@linux.intel.com> Reviewed-by: Ian Rogers <irogers@google.com> Cc: thomas.falcon@intel.com Link: https://lore.kernel.org/r/20250207152844.302167-1-kan.liang@linux.intel.com Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Ian Rogers [Sat, 1 Feb 2025 07:43:20 +0000 (23:43 -0800)]
perf stat: Changes to event name uniquification
The existing logic would disable uniquification on an evlist or enable
it per evsel, this is unfortunate as uniquification is most needed
when events have the same name and so the whole evlist must be
considered. Change the initial disable uniquify on an evlist
processing to also set a needs_uniquify flag, for cases like the
matching event names. This must be done as an initial pass as
uniquification of an event name will change the behavior of the
check. Keep the per counter uniquification but now only uniquify event
names when the needs_uniquify flag is set.
Before this change a hwmon like temp1 wouldn't be uniquified and
afterwards it will (ie the PMU is added to the temp1 event's name).
Ian Rogers [Sat, 1 Feb 2025 07:43:19 +0000 (23:43 -0800)]
perf stat: Don't merge counters purely on name
Counter merging was added in commit 942c5593393d ("perf stat: Add
perf_stat_merge_counters()"), however, it merges events with the same
name on different PMUs regardless of whether the different PMUs are
actually of the same type (ie they differ only in the suffix on the
PMU). For hwmon events there may be a temp1 event on every PMU, but
the PMU names are all unique and don't differ just by a suffix. The
merging is over eager and will merge all the hwmon counters together
meaning an aggregated and very large temp1 value is shown. The same
would be true for say cache events and memory controller events where
the PMUs differ but the event names are the same.
Fix the problem by correctly saying two PMUs alias when they differ
only by suffix.
Note, there is an overlap with evsel's merged_stat with aggregation
and the evsel's metric_leader where aggregation happens for metrics.
Fixes: 942c5593393d ("perf stat: Add perf_stat_merge_counters()") Signed-off-by: Ian Rogers <irogers@google.com> Reviewed-by: Kan Liang <kan.liang@linux.intel.com> Link: https://lore.kernel.org/r/20250201074320.746259-5-irogers@google.com Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Ian Rogers [Sat, 1 Feb 2025 07:43:18 +0000 (23:43 -0800)]
perf pmu: Rename name matching for no suffix or wildcard variants
Wildcard PMU naming will match a name like pmu_1 to a PMU name like
pmu_10 but not to a PMU name like pmu_2 as the suffix forms part of
the match. No suffix matching will match pmu_10 to either pmu_1 or
pmu_2. Add or rename matching functions on PMU to make it clearer what
kind of matching is being performed.
Ian Rogers [Sat, 1 Feb 2025 07:43:17 +0000 (23:43 -0800)]
perf pmus: Restructure pmu_read_sysfs to scan fewer PMUs
Rather than scanning core or all PMUs, allow pmu_read_sysfs to read
some combination of core, other, hwmon and tool PMUs. The PMUs that
should be read and are already read are held as bitmaps. It is known
that a "hwmon_" prefix is necessary for a hwmon PMU's name, similarly
with "tool", so only scan those PMUs in situations the PMU name or the
PMU's type number make sense to.
The number of openat system calls reduces from 276 to 98 for a hwmon
event. The number of openats for regular perf events isn't changed.
Ian Rogers [Sat, 1 Feb 2025 07:43:16 +0000 (23:43 -0800)]
perf evsel: Reduce scanning core PMUs in is_hybrid
evsel__is_hybrid returns true if there are multiple core PMUs and the
evsel is for a core PMU. Determining the number of core PMUs can
require loading/scanning PMUs. There's no point doing the scanning if
evsel for the is_hybrid test isn't core so reorder the tests to reduce
PMU scanning.