Pull arm64 updates from Will Deacon:
"We've got a little less than normal thanks to the holidays in
December, but there's the usual summary below. The highlight is
probably the 52-bit physical addressing (LPA2) clean-up from Ard.
Confidential Computing:
- Register a platform device when running in CCA realm mode to enable
automatic loading of dependent modules
CPU Features:
- Update a bunch of system register definitions to pick up new field
encodings from the architectural documentation
- Add hwcaps and selftests for the new (2024) dpISA extensions
Documentation:
- Update EL3 (firmware) requirements for booting Linux on modern
arm64 designs
- Remove stale information about the kernel virtual memory map
Miscellaneous:
- Minor cleanups and typo fixes
Memory management:
- Fix vmemmap_check_pmd() to look at the PMD type bits
- LPA2 (52-bit physical addressing) cleanups and minor fixes
- Adjust physical address space depending upon whether or not LPA2 is
enabled
Perf and PMUs:
- Add port filtering support for NVIDIA's NVLINK-C2C Coresight PMU
- Extend AXI filtering support for the DDR PMU on NXP IMX SoCs
- Fix Designware PCIe PMU event numbering
- Add generic branch events for the Apple M1 CPU PMU
- Add support for Marvell Odyssey DDR and LLC-TAD PMUs
- Cleanups to the Hisilicon DDRC and Uncore PMU code
- Advertise discard mode for the SPE PMU
- Add the perf users mailing list to our MAINTAINERS entry"
* tag 'arm64-upstream' of git://git.kernel.org/pub/scm/linux/kernel/git/arm64/linux: (64 commits)
Documentation: arm64: Remove stale and redundant virtual memory diagrams
perf docs: arm_spe: Document new discard mode
perf: arm_spe: Add format option for discard mode
MAINTAINERS: Add perf list for drivers/perf/
arm64: Remove duplicate included header
drivers/perf: apple_m1: Map generic branch events
arm64: rsi: Add automatic arm-cca-guest module loading
kselftest/arm64: Add 2024 dpISA extensions to hwcap test
KVM: arm64: Allow control of dpISA extensions in ID_AA64ISAR3_EL1
arm64/hwcap: Describe 2024 dpISA extensions to userspace
arm64/sysreg: Update ID_AA64SMFR0_EL1 to DDI0601 2024-12
arm64: Filter out SVE hwcaps when FEAT_SVE isn't implemented
drivers/perf: hisi: Set correct IRQ affinity for PMUs with no association
arm64/sme: Move storage of reg_smidr to __cpuinfo_store_cpu()
arm64: mm: Test for pmd_sect() in vmemmap_check_pmd()
arm64/mm: Replace open encodings with PXD_TABLE_BIT
arm64/mm: Rename pte_mkpresent() as pte_mkvalid()
arm64/sysreg: Update ID_AA64ISAR2_EL1 to DDI0601 2024-09
arm64/sysreg: Update ID_AA64ZFR0_EL1 to DDI0601 2024-09
arm64/sysreg: Update ID_AA64FPFR0_EL1 to DDI0601 2024-09
...
Since the linked fixes: commit, err is returned uninitialized due to the
removal of "return 0". Initialize err to fix it.
This fixes the following intermittent test failure on release builds:
$ perf test "testsuite_probe"
...
-- [ FAIL ] -- perf_probe :: test_invalid_options :: mutually exclusive options :: -L foo -V bar (output regexp parsing)
Regexp not found: \"Error: switch .+ cannot be used with switch .+\"
...
Fixes: 080e47b2a2 ("perf probe: Introduce quotation marks support")
Tested-by: Namhyung Kim <namhyung@kernel.org>
Reviewed-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Acked-by: Masami Hiramatsu (Google) <mhiramat@kernel.org>
Signed-off-by: James Clark <james.clark@linaro.org>
Link: https://lore.kernel.org/r/20241211085525.519458-2-james.clark@linaro.org
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
The temp directory is made and a known fake hwmon PMU created within
it. Prior to this fix the events were being incorrectly written to the
temp directory rather than the fake PMU directory. This didn't impact
the test as the directory fd matched the wrong location, but it
doesn't mirror what a hwmon PMU would actually look like.
Signed-off-by: Ian Rogers <irogers@google.com>
Tested-by: Arnaldo Carvalho de Melo <acme@kernel.org>
Link: https://lore.kernel.org/r/20241206042306.1055913-2-irogers@google.com
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
The hwmon PMU test will make a temp directory, open the directory with
O_DIRECTORY then fill it with contents. As the open is before the
filling the contents the later fdopendir may reflect the initial empty
state, meaning no events are seen. Change to re-open the directory,
rather than dup the fd, so the latest contents are seen.
Minor tweaks/additions to debug messages.
Signed-off-by: Ian Rogers <irogers@google.com>
Tested-by: Arnaldo Carvalho de Melo <acme@kernel.org>
Link: https://lore.kernel.org/r/20241206042306.1055913-1-irogers@google.com
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
The comparison function cmp_profile_data() violates the C standard's
requirements for qsort() comparison functions, which mandate symmetry
and transitivity:
* Symmetry: If x < y, then y > x.
* Transitivity: If x < y and y < z, then x < z.
When v1 and v2 are equal, the function incorrectly returns 1, breaking
symmetry and transitivity. This causes undefined behavior, which can
lead to memory corruption in certain versions of glibc [1].
Fix the issue by returning 0 when v1 and v2 are equal, ensuring
compliance with the C standard and preventing undefined behavior.
Link: https://www.qualys.com/2024/01/30/qsort.txt [1]
Fixes: 0f223813ed ("perf ftrace: Add 'profile' command")
Fixes: 74ae366c37 ("perf ftrace profile: Add -s/--sort option")
Cc: stable@vger.kernel.org
Signed-off-by: Kuan-Wei Chiu <visitorckw@gmail.com>
Reviewed-by: Namhyung Kim <namhyung@kernel.org>
Reviewed-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: jserv@ccns.ncku.edu.tw
Cc: chuang@cs.nycu.edu.tw
Link: https://lore.kernel.org/r/20241209134226.1939163-1-visitorckw@gmail.com
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
It should only have generic flags in the array but the recent header
sync brought a new flags to fcntl.h and caused a build error. Let's
update the shell script to exclude flags specific to name_to_handle_at().
CC trace/beauty/fs_at_flags.o
In file included from trace/beauty/fs_at_flags.c:21:
tools/perf/trace/beauty/generated/fs_at_flags_array.c:13:30: error: initialized field overwritten [-Werror=override-init]
13 | [ilog2(0x002) + 1] = "HANDLE_CONNECTABLE",
| ^~~~~~~~~~~~~~~~~~~~
tools/perf/trace/beauty/generated/fs_at_flags_array.c:13:30: note: (near initialization for ‘fs_at_flags[2]’)
Reviewed-by: James Clark <james.clark@linaro.org>
Link: https://lore.kernel.org/r/20241203035349.1901262-12-namhyung@kernel.org
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
To pick up the changes in this cset:
6140be90ec ("fs/xattr: add *at family syscalls")
This addresses these perf build warnings:
Warning: Kernel ABI header differences:
diff -u tools/include/uapi/asm-generic/unistd.h include/uapi/asm-generic/unistd.h
diff -u tools/perf/arch/x86/entry/syscalls/syscall_32.tbl arch/x86/entry/syscalls/syscall_32.tbl
diff -u tools/perf/arch/x86/entry/syscalls/syscall_64.tbl arch/x86/entry/syscalls/syscall_64.tbl
diff -u tools/perf/arch/powerpc/entry/syscalls/syscall.tbl arch/powerpc/kernel/syscalls/syscall.tbl
diff -u tools/perf/arch/s390/entry/syscalls/syscall.tbl arch/s390/kernel/syscalls/syscall.tbl
diff -u tools/perf/arch/mips/entry/syscalls/syscall_n64.tbl arch/mips/kernel/syscalls/syscall_n64.tbl
The arm64 changes are not included as it requires more changes in the
tools. It'll be worked for the later cycle.
Please see tools/include/uapi/README for further details.
Reviewed-by: James Clark <james.clark@linaro.org>
Cc: Arnd Bergmann <arnd@arndb.de>
Cc: Christian Brauner <brauner@kernel.org>
CC: x86@kernel.org
CC: linux-mips@vger.kernel.org
CC: linuxppc-dev@lists.ozlabs.org
CC: linux-s390@vger.kernel.org
Link: https://lore.kernel.org/r/20241203035349.1901262-7-namhyung@kernel.org
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
This signal handler loops over all tests on ctrl-C, but it's active
while the test list is being constructed. process.pid is 0, then -1,
then finally set to the child pid on fork. If the Ctrl-C is received
during this point a kill(-1, SIGINT) can be sent which affects all
processes.
Make sure the child has forked first before forwarding the signal. This
can be reproduced with ctrl-C immediately after launching perf test
which terminates the ssh connection.
Fixes: 553d5efeb3 ("perf test: Add a signal handler to kill forked child processes")
Signed-off-by: James Clark <james.clark@linaro.org>
Reviewed-by: Ian Rogers <irogers@google.com>
Link: https://lore.kernel.org/r/20241129151948.3199732-1-james.clark@linaro.org
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
The build-id events written at the end of the record session are broken
due to unexpected data. The write_buildid() writes the fixed length
event first and then variable length filename.
But a recent change made it write more data in the padding area
accidentally. So readers of the event see zero-filled data for the
next entry and treat it incorrectly. This resulted in wrong kernel
symbols because the kernel DSO loaded a random vmlinux image in the
path as it didn't have a valid build-id.
Fixes: ae39ba1655 ("perf inject: Fix build ID injection")
Reported-by: Linus Torvalds <torvalds@linux-foundation.org>
Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Reviewed-by: Ian Rogers <irogers@google.com>
Link: https://lore.kernel.org/r/Z0aRFFW9xMh3mqKB@google.com
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Pull perf tools updates from Namhyung Kim:
"perf record:
- Enable leader sampling for inherited task events. It was supported
only for system-wide events but the kernel started to support such
a setup since v6.12.
This is to reduce the number of PMU interrupts. The samples of the
leader event will contain counts of other events and no samples
will be generated for the other member events.
$ perf record -e '{cycles,instructions}:S' ${MYPROG}
perf report:
- Fix --branch-history option to display more branch-related
information like prediction, abort and cycles which is available
on Intel machines.
$ perf record -bg -- perf test -w brstack
$ perf report --branch-history
...
#
# Overhead Source:Line Symbol Shared Object Predicted Abort Cycles IPC [IPC Coverage]
# ........ ........................ .............. .................... ......... ..... ...... ....................
#
8.17% copy_page_64.S:19 [k] copy_page [kernel.kallsyms] 50.0% 0 5 - -
|
---xas_load xarray.h:171
|
|--5.68%--xas_load xarray.c:245 (cycles:1)
| xas_load xarray.c:242
| xas_load xarray.h:1260 (cycles:1)
| xas_descend xarray.c:146
| xas_load xarray.c:244 (cycles:2)
| xas_load xarray.c:245
| xas_descend xarray.c:218 (cycles:10)
...
perf stat:
- Add HWMON PMU support.
The HWMON provides various system information like CPU/GPU
temperature, fan speed and so on. Expose them as PMU events so that
users can see the values using perf stat commands.
$ perf stat -e temp_cpu,fan1 true
Performance counter stats for 'true':
60.00 'C temp_cpu
0 rpm fan1
0.000745382 seconds time elapsed
0.000883000 seconds user
0.000000000 seconds sys
- Display metric threshold in JSON output.
Some metrics define thresholds to classify value ranges. It used to
be in a different color but it won't work for JSON.
Add "metric-threshold" field to the JSON that can be one of "good",
"less good", "nearly bad" and "bad".
# perf stat -a -M TopdownL1 -j true
{"counter-value" : "18693525.000000", "unit" : "", "event" : "TOPDOWN.SLOTS", "event-runtime" : 5552708, "pcnt-running" : 100.00, "metric-value" : "43.226002", "metric-unit" : "% tma_backend_bound", "metric-threshold" : "bad"}
{"metric-value" : "29.212267", "metric-unit" : "% tma_frontend_bound", "metric-threshold" : "bad"}
{"metric-value" : "7.138972", "metric-unit" : "% tma_bad_speculation", "metric-threshold" : "good"}
{"metric-value" : "20.422759", "metric-unit" : "% tma_retiring", "metric-threshold" : "good"}
{"counter-value" : "3817732.000000", "unit" : "", "event" : "topdown-retiring", "event-runtime" : 5552708, "pcnt-running" : 100.00, }
{"counter-value" : "5472824.000000", "unit" : "", "event" : "topdown-fe-bound", "event-runtime" : 5552708, "pcnt-running" : 100.00, }
{"counter-value" : "7984780.000000", "unit" : "", "event" : "topdown-be-bound", "event-runtime" : 5552708, "pcnt-running" : 100.00, }
{"counter-value" : "1418181.000000", "unit" : "", "event" : "topdown-bad-spec", "event-runtime" : 5552708, "pcnt-running" : 100.00, }
...
perf sched:
- Add -P/--pre-migrations option for 'timehist' sub-command to track
time a task waited on a run-queue before migrating to a different
CPU.
$ perf sched timehist -P
time cpu task name wait time sch delay run time pre-mig time
[tid/pid] (msec) (msec) (msec) (msec)
--------------- ------ ------------------------------ --------- --------- --------- ---------
585940.535527 [0000] perf[584885] 0.000 0.000 0.000 0.000
585940.535535 [0000] migration/0[20] 0.000 0.002 0.008 0.000
585940.535559 [0001] perf[584885] 0.000 0.000 0.000 0.000
585940.535563 [0001] migration/1[25] 0.000 0.001 0.004 0.000
585940.535678 [0002] perf[584885] 0.000 0.000 0.000 0.000
585940.535686 [0002] migration/2[31] 0.000 0.002 0.008 0.000
585940.535905 [0001] <idle> 0.000 0.000 0.342 0.000
585940.535938 [0003] perf[584885] 0.000 0.000 0.000 0.000
585940.537048 [0001] sleep[584886] 0.000 0.019 1.142 0.001
585940.537749 [0002] <idle> 0.000 0.000 2.062 0.000
...
Build:
- Make libunwind opt-in (LIBUNWIND=1) rather than opt-out.
The perf tools are generally built with libelf and libdw which has
unwinder functionality. The libunwind support predates it and no
need to have duplicate unwinders by default.
- Rename NO_DWARF=1 build option to NO_LIBDW=1 in order to clarify
it's using libdw for handling DWARF information.
Internals:
- Do not set exclude_guest bit in the perf_event_attr by default.
This was causing a trouble in AMD IBS PMU as it doesn't support the
bit. The bit will be set when it's needed later by the fallback
logic. Also update the missing feature detection logic to make sure
not clear supported bits unnecessarily.
- Run perf test in parallel by default and mark flaky tests
"exclusive" to run them serially at the end. Some test numbers are
changed but the test can complete in less than half the time.
JSON vendor events:
- Add AMD Zen 5 events and metrics.
- Add i.MX91 and i.MX95 DDR metrics
- Fix HiSilicon HIP08 Topdown metric name.
- Support compat events on PowerPC"
* tag 'perf-tools-for-v6.13-2024-11-24' of git://git.kernel.org/pub/scm/linux/kernel/git/perf/perf-tools: (232 commits)
perf tests: Fix hwmon parsing with PMU name test
perf hwmon_pmu: Ensure hwmon key union is zeroed before use
perf tests hwmon_pmu: Remove double evlist__delete()
perf/test: fix perf ftrace test on s390
perf bpf-filter: Return -ENOMEM directly when pfi allocation fails
perf test: Correct hwmon test PMU detection
perf: Remove unused del_perf_probe_events()
perf pmu: Move pmu_metrics_table__find and remove ARM override
perf jevents: Add map_for_cpu()
perf header: Pass a perf_cpu rather than a PMU to get_cpuid_str
perf header: Avoid transitive PMU includes
perf arm64 header: Use cpu argument in get_cpuid
perf header: Refactor get_cpuid to take a CPU for ARM
perf header: Move is_cpu_online to numa bench
perf jevents: fix breakage when do perf stat on system metric
perf test: Add missing __exit calls in tool/hwmon tests
perf tests: Make leader sampling test work without branch event
perf util: Remove kernel version deadcode
perf test shell trace_exit_race: Use --no-comm to avoid cases where COMM isn't resolved
perf test shell trace_exit_race: Show what went wrong in verbose mode
...
Incorrectly the hwmon with PMU name test didn't pass "true". Fix and
address issue with hwmon_pmu__config_terms needing to load events - a
load bearing assert fired. Also fix missing list deletion when putting
the hwmon test PMU and lower some debug warnings to make the hwmon PMU
less spammy in verbose mode.
Signed-off-by: Ian Rogers <irogers@google.com>
Tested-by: Namhyung Kim <namhyung@kernel.org>
Link: https://lore.kernel.org/r/20241121000955.536930-1-irogers@google.com
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
On s390 the perf test case ftrace sometimes fails as follows:
# ./perf test ftrace
79: perf ftrace tests : FAILED!
#
The failure depends on the kernel .config file. Some configurations
always work fine, some do not. The ftrace profile test mostly fails,
because the ring buffer was not large enough, and some lines
(especially the interesting ones with nanosleep in it) where dropped.
To achieve success for all tested kernel configurations, enlarge
the buffer to store the traces completely without wrapping.
The default buffer size is too small for all kernel configurations.
Set the buffer size of for the ftrace profile test to 16 MB.
Output after:
# ./perf test ftrace
79: perf ftrace tests : Ok
#
Signed-off-by: Thomas Richter <tmricht@linux.ibm.com>
Acked-by: Namhyung Kim <namhyung@kernel.org>
Cc: agordeev@linux.ibm.com
Cc: gor@linux.ibm.com
Cc: hca@linux.ibm.com
Cc: sumanthk@linux.ibm.com
Link: https://lore.kernel.org/r/20241119064856.641446-1-tmricht@linux.ibm.com
Suggested-by: Sven Schnelle <svens@linux.ibm.com>
Suggested-by: Namhyung Kim <namhyung@kernel.org>
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
The purpose of this test is to test for races in the exit of 'perf
trace' missing the last events, it was failing when the COMM wasn't
resolved either because we missed some PERF_RECORD_COMM or somehow
raced on getting it from procfs.
Add --no-comm to the 'perf trace' command line so that we get a
consistent, pid only output, which allows the test to achieve its goal.
This is the output from
'perf trace --no-comm -e syscalls:sys_enter_exit_group':
0.000 21953 syscalls:sys_enter_exit_group()
0.000 21955 syscalls:sys_enter_exit_group()
0.000 21957 syscalls:sys_enter_exit_group()
0.000 21959 syscalls:sys_enter_exit_group()
0.000 21961 syscalls:sys_enter_exit_group()
0.000 21963 syscalls:sys_enter_exit_group()
0.000 21965 syscalls:sys_enter_exit_group()
0.000 21967 syscalls:sys_enter_exit_group()
0.000 21969 syscalls:sys_enter_exit_group()
0.000 21971 syscalls:sys_enter_exit_group()
Now it passes:
root@number:~# perf test "trace exit race"
110: perf trace exit race : Ok
root@number:~#
root@number:~# perf test -v "trace exit race"
110: perf trace exit race : Ok
root@number:~#
If we artificially make it run just 9 times instead of the 10 it runs,
i.e. by manually doing:
trace_shutdown_race() {
for _ in $(seq 9); do
that 9 is $iter, 10 in the patch, we get:
root@number:~# vim ~acme/libexec/perf-core/tests/shell/trace_exit_race.sh
root@number:~# perf test -v "trace exit race"
--- start ---
test child forked, pid 24629
Missing output, expected 10 but only got 9
---- end(-1) ----
110: perf trace exit race : FAILED!
root@number:~#
I.e. 9 'perf trace' calls produced the expected output, the inverse grep
didn't show anything, so the patch provided by Howard for the previous
patch kicks in and shows a more informative message.
Tested-by: Howard Chu <howardchu95@gmail.com>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Benjamin Peterson <benjamin@engflow.com>
Cc: Ian Rogers <irogers@google.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Kan Liang <kan.liang@linux.intel.com>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: https://lore.kernel.org/lkml/ZzdknoHqrJbojb6P@x1
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
If it fails we need to check what was the reason, what were the lines
that didn't match the expected format, so:
root@number:~# perf test -v "trace exit race"
--- start ---
test child forked, pid 2028724
Lines not matching the expected regexp: ' +[0-9]+\.[0-9]+ +true/[0-9]+ syscalls:sys_enter_exit_group\(\)$':
0.000 :2028750/2028750 syscalls:sys_enter_exit_group()
---- end(-1) ----
110: perf trace exit race : FAILED!
root@number:~#
In this case we're not resolving the process COMM for some reason and
fallback to printing just the pid/tid, this will be fixed in a followup
patch.
Howard Chu spotted a problem with single code surrounding a regexp, that
made the test always fail, but since there were some failures when I
tested (COMM not being resolved in some of the results) the end inverse
grep would show some lines and thus didn't notice the single quote
problem.
He also provided a patch to test if less than the number of expected
matches took place but all of them with the expected output, in which
case the inverse grep wouldn't show anything, confusing the tester.
Reviewed-by: Howard Chu <howardchu95@gmail.com>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Benjamin Peterson <benjamin@engflow.com>
Cc: Ian Rogers <irogers@google.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Kan Liang <kan.liang@linux.intel.com>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: https://lore.kernel.org/lkml/ZzdknoHqrJbojb6P@x1
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
If a perf trace event selector specifies a maximum number of events to output
(i.e., "/nr=N/" syntax), the event printing handler, trace__event_handler,
disables the event selector after the maximum number events are
printed.
Furthermore, trace__event_handler checked if the event selector was
disabled before doing any work. This avoided exceeding the maximum
number of events to print if more events were in the buffer before the
selector was disabled.
However, the event selector can be disabled for reasons other than
exceeding the maximum number of events. In particular, when the traced
subprocess exits, the main loop disables all event selectors. This meant
the last events of a traced subprocess might be lost to the printing
handler's short-circuiting logic.
This nondeterministic problem could be seen by running the following many times:
$ perf trace -e syscalls:sys_enter_exit_group true
trace__event_handler should simply check for exceeding the maximum number of
events to print rather than the state of the event selector.
Fixes: a9c5e6c1e9 ("perf trace: Introduce per-event maximum number of events property")
Signed-off-by: Benjamin Peterson <benjamin@engflow.com>
Tested-by: Howard Chu <howardchu95@gmail.com>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Ian Rogers <irogers@google.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Kan Liang <kan.liang@linux.intel.com>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: https://lore.kernel.org/r/20241107232128.108981-1-benjamin@engflow.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
In non-C languages, it is possible to have ':' in the function names.
It is possible to escape it with backslashes, but if there are too many
backslashes, it is annoying.
This introduce quotation marks (`"` or `'`) support.
For example, without quotes, we have to pass it as below
$ perf probe -x cro3 -L "cro3\:\:cmd\:\:servo\:\:run_show"
<run_show@/work/cro3/src/cmd/servo.rs:0>
0 fn run_show(args: &ArgsShow) -> Result<()> {
1 let list = ServoList::discover()?;
2 let s = list.find_by_serial(&args.servo)?;
3 if args.json {
4 println!("{s}");
With quotes, we can more naturally write the function name as below;
$ perf probe -x cro3 -L \"cro3::cmd::servo::run_show\"
<run_show@/work/cro3/src/cmd/servo.rs:0>
0 fn run_show(args: &ArgsShow) -> Result<()> {
1 let list = ServoList::discover()?;
2 let s = list.find_by_serial(&args.servo)?;
3 if args.json {
4 println!("{s}");
Signed-off-by: Masami Hiramatsu <mhiramat@kernel.org>
Cc: Alexander Lobakin <aleksander.lobakin@intel.com>
Cc: Dima Kogan <dima@secretsauce.net>
Cc: Ian Rogers <irogers@google.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Przemek Kitszel <przemyslaw.kitszel@intel.com>
Link: https://lore.kernel.org/r/173099116941.2431889.11609129616090100386.stgit@mhiramat.roam.corp.google.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
strpbrk_esq() and strdup_esq() are new variants for strpbrk() and
strdup() which handles escaped characters and quoted strings.
- strpbrk_esq() searches specified set of characters but ignores the
escaped characters and quoted strings.
e.g. strpbrk_esq("'quote\d' \queue quiz", "qd") returns "quiz".
- strdup_esq() duplicates string but removes backslash and quotes which
is used for quotation. It also keeps the string (including backslash)
in the quoted part.
e.g. strdup_esq("'quote\d' \queue quiz") returns "quote\d queue quiz".
The (single, double) quotes in the quoted part should be escaped by
backslash. In this case, strdup_esq() removes that backslash.
The same quotes must be paired. If you use double quotation, you need
to use the double quotation to close the quoted part.
Signed-off-by: Masami Hiramatsu <mhiramat@kernel.org>
Cc: Alexander Lobakin <aleksander.lobakin@intel.com>
Cc: Dima Kogan <dima@secretsauce.net>
Cc: Ian Rogers <irogers@google.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Przemek Kitszel <przemyslaw.kitszel@intel.com>
Link: https://lore.kernel.org/r/173099116045.2431889.15772916605719019533.stgit@mhiramat.roam.corp.google.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
The perf tools annotation code used for a long time parsing the output
of binutils's objdump (or its reimplementations, like llvm's) to then
parse and augment it with samples, allow navigation, etc.
More recently disassemblers from the capstone and llvm (libraries, not
parsing the output of tools using those libraries to mimic binutils's
objdump output) were introduced.
So when all those methods are available, there is a static preference
for a series of attempts of disassembling a binary, with the 'llvm,
capstone, objdump' sequence being hard coded.
This patch allows users to change that sequence, specifying via a 'perf
config' 'annotate.disassemblers' entry which and in what order
disassemblers should be attempted.
As alluded to in the comments in the source code of this series, this
flexibility is useful for users and developers alike, elliminating the
requirement to rebuild the tool with some specific set of libraries to
see how the output of disassembling would be for one of these methods.
root@x1:~# rm -f ~/.perfconfig
root@x1:~# perf annotate -v --stdio2 update_load_avg
<SNIP>
symbol__disassemble:
filename=/usr/lib/debug/lib/modules/6.11.4-201.fc40.x86_64/vmlinux,
sym=update_load_avg, start=0xffffffffb6148fe0, en>
annotating [0x6ff7170]
/usr/lib/debug/lib/modules/6.11.4-201.fc40.x86_64/vmlinux :
[0x7407ca0] update_load_avg
Disassembled with llvm
annotate.disassemblers=llvm,capstone,objdump
Samples: 66 of event 'cpu_atom/cycles/P', 10000 Hz,
Event count (approx.): 5185444, [percent: local period]
update_load_avg()
/usr/lib/debug/lib/modules/6.11.4-201.fc40.x86_64/vmlinux
Percent 0xffffffff81148fe0 <update_load_avg>:
1.61 pushq %r15
pushq %r14
1.00 pushq %r13
movl %edx,%r13d
1.90 pushq %r12
pushq %rbp
movq %rsi,%rbp
pushq %rbx
movq %rdi,%rbx
subq $0x18,%rsp
15.14 movl 0x1a4(%rdi),%eax
root@x1:~# perf config annotate.disassemblers=capstone
root@x1:~# cat ~/.perfconfig
# this file is auto-generated.
[annotate]
disassemblers = capstone
root@x1:~#
root@x1:~# perf annotate -v --stdio2 update_load_avg
<SNIP>
Disassembled with capstone
annotate.disassemblers=capstone
Samples: 66 of event 'cpu_atom/cycles/P', 10000 Hz,
Event count (approx.): 5185444, [percent: local period]
update_load_avg()
/usr/lib/debug/lib/modules/6.11.4-201.fc40.x86_64/vmlinux
Percent 0xffffffff81148fe0 <update_load_avg>:
1.61 pushq %r15
pushq %r14
1.00 pushq %r13
movl %edx,%r13d
1.90 pushq %r12
pushq %rbp
movq %rsi,%rbp
pushq %rbx
movq %rdi,%rbx
subq $0x18,%rsp
15.14 movl 0x1a4(%rdi),%eax
root@x1:~# perf config annotate.disassemblers=objdump,capstone
root@x1:~# perf config annotate.disassemblers
annotate.disassemblers=objdump,capstone
root@x1:~# cat ~/.perfconfig
# this file is auto-generated.
[annotate]
disassemblers = objdump,capstone
root@x1:~# perf annotate -v --stdio2 update_load_avg
Executing: objdump --start-address=0xffffffff81148fe0 \
--stop-address=0xffffffff811497aa \
-d --no-show-raw-insn -S -C "$1"
Disassembled with objdump
annotate.disassemblers=objdump,capstone
Samples: 66 of event 'cpu_atom/cycles/P', 10000 Hz,
Event count (approx.): 5185444, [percent: local period]
update_load_avg()
/usr/lib/debug/lib/modules/6.11.4-201.fc40.x86_64/vmlinux
Percent
Disassembly of section .text:
ffffffff81148fe0 <update_load_avg>:
#define DO_ATTACH 0x4
ffffffff81148fe0 <update_load_avg>:
#define DO_ATTACH 0x4
#define DO_DETACH 0x8
/* Update task and its cfs_rq load average */
static inline void update_load_avg(struct cfs_rq *cfs_rq,
struct sched_entity *se,
int flags)
{
1.61 push %r15
push %r14
1.00 push %r13
mov %edx,%r13d
1.90 push %r12
push %rbp
mov %rsi,%rbp
push %rbx
mov %rdi,%rbx
sub $0x18,%rsp
}
/* rq->task_clock normalized against any time
this cfs_rq has spent throttled */
static inline u64 cfs_rq_clock_pelt(struct cfs_rq *cfs_rq)
{
if (unlikely(cfs_rq->throttle_count))
15.14 mov 0x1a4(%rdi),%eax
root@x1:~#
After adding a way to select the disassembler from the command line a
'perf test' comparing the output of the various diassemblers should be
introduced, to test these codebases.
Acked-by: Ian Rogers <irogers@google.com>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Athira Rajeev <atrajeev@linux.vnet.ibm.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Kan Liang <kan.liang@linux.intel.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Steinar H. Gunderson <sesse@google.com>
Link: https://lore.kernel.org/r/20241111151734.1018476-4-acme@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
