| Age | Commit message (Collapse) | Author | Files | Lines |
|---|
|
Adding available control commands in separate paragraph, so it's more
readable and easier to add new commands.
Signed-off-by: Jiri Olsa <[email protected]>
Cc: Alexander Shishkin <[email protected]>
Cc: Alexei Budankov <[email protected]>
Cc: Ian Rogers <[email protected]>
Cc: Mark Rutland <[email protected]>
Cc: Michael Petlan <[email protected]>
Cc: Namhyung Kim <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Stephane Eranian <[email protected]>
Link: http://lore.kernel.org/lkml/[email protected]
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
|
|
in the SYNOPSIS section.
Committer testing:
With the previously documented example:
$ perf config --user report sort-order=srcline
The config variable does not contain a section name: report
$
With the fixed example line:
$ perf config --user report.sort-order=srcline
$ perf config --user report.sort-order
report.sort-order=srcline
$
Signed-off-by: Nick Thompson <[email protected]>
Cc: Adrian Hunter <[email protected]>
Cc: Ian Rogers <[email protected]>
Cc: Jiri Olsa <[email protected]>
Cc: Namhyung Kim <[email protected]>
Link: https://lore.kernel.org/linux-perf-users/[email protected]/
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
|
|
To fix this:
$ perf test -v 27
27: Sample parsing :
--- start ---
test child forked, pid 586013
sample format has changed, some new PERF_SAMPLE_ bit was introduced - test needs updating
test child finished with -1
---- end ----
Sample parsing: FAILED!
$
This patchset is still not completely merged, so when adding the
PERF_SAMPLE_CODE_PAGE_SIZE to 'struct perf_sample' we need to add the
bits added in this patch for 'perf_sample.data_page_size'.
Fixes: 251cc77b8176de37 ("tools headers UAPI: Update tools's copy of linux/perf_event.h")
Cc: Adrian Hunter <[email protected]>
Cc: Ian Rogers <[email protected]>
Cc: Jiri Olsa <[email protected]>
Cc: Kan Liang <[email protected]>
Cc: Namhyung Kim <[email protected]>
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
|
|
Adding support to decompress file before reading build id.
Adding filename__read_build_id and change its current versions to
read_build_id.
Shutting down stderr output of perf list in the shell test:
82: Check open filename arg using perf trace + vfs_getname : Ok
because with decompression code in the place we the
filename__read_build_id function is more verbose in case
of error and the test did not account for that.
Signed-off-by: Jiri Olsa <[email protected]>
Cc: Adrian Hunter <[email protected]>
Cc: Alexander Shishkin <[email protected]>
Cc: Alexei Budankov <[email protected]>
Cc: Andi Kleen <[email protected]>
Cc: Ian Rogers <[email protected]>
Cc: Mark Rutland <[email protected]>
Cc: Michael Petlan <[email protected]>
Cc: Namhyung Kim <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Song Liu <[email protected]>
Cc: Stephane Eranian <[email protected]>
Link: http://lore.kernel.org/lkml/[email protected]
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
|
|
Allow to set debug output file via new debug_set_file function.
It's called during perf startup in perf_debug_setup to set stderr file
as default and any perf command can set it later to different file.
It will be used in perf daemon command to get verbose output into log
file.
Signed-off-by: Jiri Olsa <[email protected]>
Cc: Alexander Shishkin <[email protected]>
Cc: Alexei Budankov <[email protected]>
Cc: Ian Rogers <[email protected]>
Cc: Mark Rutland <[email protected]>
Cc: Michael Petlan <[email protected]>
Cc: Namhyung Kim <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Stephane Eranian <[email protected]>
Link: http://lore.kernel.org/lkml/[email protected]
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
|
|
Adding interface to enable/disable single event in the evlist based on
its name. It will be used later in new control enable/disable interface.
Keeping the evlist::enabled true when one or more events are enabled so
the toggle can work properly and toggle evlist to disabled state.
Signed-off-by: Jiri Olsa <[email protected]>
Acked-by: Namhyung Kim <[email protected]>
Acked-by: Alexei Budankov <[email protected]>
Cc: Alexander Shishkin <[email protected]>
Cc: Ian Rogers <[email protected]>
Cc: Ingo Molnar <[email protected]>
Cc: Mark Rutland <[email protected]>
Cc: Michael Petlan <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Stephane Eranian <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
|
|
Likewise, perf evlist command should print event attributes by reading
PERF_RECORD_HEADER_ATTR records.
Before:
$ perf record -o- true | ./perf evlist -i-
(prints nothing)
After:
$ perf record -o- true | ./perf evlist -i-
cycles:pppH
Committer testing:
Verbose mode also works as expected:
$ perf record -o- true | perf evlist -i-
cycles:uhH
$ perf record -o- true | perf evlist -vi-
cycles:uhH: size: 120, { sample_period, sample_freq }: 4000, sample_type: IP|TID|TIME|ID|PERIOD, read_format: ID, disabled: 1, inherit: 1, exclude_kernel: 1, mmap: 1, comm: 1, freq: 1, enable_on_exec: 1, task: 1, sample_id_all: 1, exclude_guest: 1, mmap2: 1, comm_exec: 1, ksymbol: 1, bpf_event: 1
$
Signed-off-by: Namhyung Kim <[email protected]>
Acked-by: Jiri Olsa <[email protected]>
Tested-by: Arnaldo Carvalho de Melo <[email protected]>
Cc: Alexander Shishkin <[email protected]>
Cc: Ian Rogers <[email protected]>
Cc: Ingo Molnar <[email protected]>
Cc: Mark Rutland <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Stephane Eranian <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
|
|
The --header-only checks file header and prints the feature data. But
as pipe mode doesn't have it in the header it prints almost nothing.
Change it to process first few records until it founds HEADER_FEATURE.
Before:
$ perf record -o- true | perf report -i- --header-only
# ========
# captured on : Thu Dec 10 14:34:59 2020
# header version : 1
# data offset : 0
# data size : 0
# feat offset : 0
# ========
#
After:
$ perf record -o- true | perf report -i- --header-only
# ========
# captured on : Thu Dec 10 14:49:11 2020
# header version : 1
# data offset : 0
# data size : 0
# feat offset : 0
# ========
#
# hostname : balhae
# os release : 5.7.17-1xxx
# perf version : 5.10.rc6.gdb0ea13cc741
# arch : x86_64
# nrcpus online : 8
# nrcpus avail : 8
# cpudesc : Intel(R) Core(TM) i7-8665U CPU @ 1.90GHz
# cpuid : GenuineIntel,6,142,12
# total memory : 16158916 kB
# cmdline : perf record -o- true
# event : name = cycles, , id = { 81, 82, 83, 84, 85, 86, 87, 88 }, size = 120, ...
# CPU_TOPOLOGY info available, use -I to display
# NUMA_TOPOLOGY info available, use -I to display
# pmu mappings: intel_pt = 9, intel_bts = 8, software = 1, power = 20, uprobe = 7, ...
# time of first sample : 0.000000
# time of last sample : 0.000000
# sample duration : 0.000 ms
# MEM_TOPOLOGY info available, use -I to display
# cpu pmu capabilities: branches=32, max_precise=3, pmu_name=skylake
Signed-off-by: Namhyung Kim <[email protected]>
Tested-by: Arnaldo Carvalho de Melo <[email protected]>
Acked-by: Jiri Olsa <[email protected]>
Cc: Alexander Shishkin <[email protected]>
Cc: Ian Rogers <[email protected]>
Cc: Ingo Molnar <[email protected]>
Cc: Mark Rutland <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Stephane Eranian <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
|
|
Add JSON metrics for imx8mm DDR Perf.
Signed-off-by: Joakim Zhang <[email protected]>
Acked-by: Kajol Jain <[email protected]>
Cc: Alexander Shishkin <[email protected]>
Cc: Andi Kleen <[email protected]>
Cc: Ian Rogers <[email protected]>
Cc: Jiri Olsa <[email protected]>
Cc: John Garry <[email protected]>
Cc: Kan Liang <[email protected]>
Cc: Kim Phillips <[email protected]>
Cc: Leo Yan <[email protected]>
Cc: Mark Rutland <[email protected]>
Cc: Mathieu Poirier <[email protected]>
Cc: Namhyung Kim <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Shaokun Zhang <[email protected]>
Cc: Will Deacon <[email protected]>
Cc: [email protected]
Cc: [email protected]
Signed-off-by: John Garry <[email protected]>
Link: http://lore.kernel.org/lkml/[email protected]
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
|
|
Currently adding metrics for core- or uncore-based events matched by CPUID
is supported.
Extend this for system events.
Signed-off-by: John Garry <[email protected]>
Acked-by: Kajol Jain <[email protected]>
Cc: Alexander Shishkin <[email protected]>
Cc: Andi Kleen <[email protected]>
Cc: Ian Rogers <[email protected]>
Cc: Jiri Olsa <[email protected]>
Cc: Joakim Zhang <[email protected]>
Cc: Kan Liang <[email protected]>
Cc: Kim Phillips <[email protected]>
Cc: Leo Yan <[email protected]>
Cc: Mark Rutland <[email protected]>
Cc: Mathieu Poirier <[email protected]>
Cc: Namhyung Kim <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Shaokun Zhang <[email protected]>
Cc: Will Deacon <[email protected]>
Cc: [email protected]
Cc: [email protected]
Link: http://lore.kernel.org/lkml/[email protected]
[ Reorder 'struct metricgroup_add_iter_data' field to avoid alignment holes ]
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
|
|
Currently printing metricgroups for core- or uncore-based events matched
by CPUID is supported.
Extend this for system events.
Signed-off-by: John Garry <[email protected]>
Acked-by: Kajol Jain <[email protected]>
Cc: Alexander Shishkin <[email protected]>
Cc: Andi Kleen <[email protected]>
Cc: Ian Rogers <[email protected]>
Cc: Jiri Olsa <[email protected]>
Cc: Joakim Zhang <[email protected]>
Cc: Kan Liang <[email protected]>
Cc: Kim Phillips <[email protected]>
Cc: Leo Yan <[email protected]>
Cc: Mark Rutland <[email protected]>
Cc: Mathieu Poirier <[email protected]>
Cc: Namhyung Kim <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Shaokun Zhang <[email protected]>
Cc: Will Deacon <[email protected]>
Cc: [email protected]
Cc: [email protected]
Link: http://lore.kernel.org/lkml/[email protected]
[ Reorder 'struct metricgroup_print_sys_idata' field to avoid alignment holes ]
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
|
|
To aid supporting system event metric groups, break up the function
metricgroup__print() into a part which iterates metrics and a part which
actually "prints" the metric.
No functional change intended.
Signed-off-by: John Garry <[email protected]>
Acked-by: Kajol Jain <[email protected]>
Cc: Alexander Shishkin <[email protected]>
Cc: Andi Kleen <[email protected]>
Cc: Ian Rogers <[email protected]>
Cc: Jiri Olsa <[email protected]>
Cc: Joakim Zhang <[email protected]>
Cc: Kan Liang <[email protected]>
Cc: Kim Phillips <[email protected]>
Cc: Leo Yan <[email protected]>
Cc: Mark Rutland <[email protected]>
Cc: Mathieu Poirier <[email protected]>
Cc: Namhyung Kim <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Shaokun Zhang <[email protected]>
Cc: Will Deacon <[email protected]>
Cc: [email protected]
Cc: [email protected]
Link: http://lore.kernel.org/lkml/[email protected]
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
|
|
Support for metric expressions using aliases which cover multiple PMUs
is broken. Consider the following test metric expression:
"MetricExpr": "UNC_CBO_XSNP_RESPONSE.MISS_XCORE * UNC_CBO_XSNP_RESPONSE.MISS_EVICTION"
When used on my broadwell, "perf stat" gives:
unc_cbo_xsnp_response.miss_eviction -> uncore_cbox_1/umask=0x81,event=0x22/
unc_cbo_xsnp_response.miss_eviction -> uncore_cbox_0/umask=0x81,event=0x22/
unc_cbo_xsnp_response.miss_xcore -> uncore_cbox_1/umask=0x41,event=0x22/
unc_cbo_xsnp_response.miss_xcore -> uncore_cbox_0/umask=0x41,event=0x22/
Control descriptor is not initialized
unc_cbo_xsnp_response.miss_eviction: 3645925 1000850523 1000850523
unc_cbo_xsnp_response.miss_xcore: 106850 1000850523 1000850523
Performance counter stats for 'system wide':
3,645,925 unc_cbo_xsnp_response.miss_eviction # 389567086250.00 test_metric_inc
106,850 unc_cbo_xsnp_response.miss_xcore
1.000883096 seconds time elapsed
Notice that only the results from one PMU are included. Fix the logic of
find_evsel_group() to enable events which apply to multiple PMUs, by
checking if the event pmu_name matches that of the metric event.
With that, "perf stat" now gives:
unc_cbo_xsnp_response.miss_eviction -> uncore_cbox_1/umask=0x81,event=0x22/
unc_cbo_xsnp_response.miss_eviction -> uncore_cbox_0/umask=0x81,event=0x22/
unc_cbo_xsnp_response.miss_xcore -> uncore_cbox_1/umask=0x41,event=0x22/
unc_cbo_xsnp_response.miss_xcore -> uncore_cbox_0/umask=0x41,event=0x22/
Control descriptor is not initialized
unc_cbo_xsnp_response.miss_eviction: 4237983 1000904100 1000904100
unc_cbo_xsnp_response.miss_xcore: 218643 1000904100 1000904100
unc_cbo_xsnp_response.miss_eviction: 4254148 1000902629 1000902629
unc_cbo_xsnp_response.miss_xcore: 213352 1000902629 1000902629
Performance counter stats for 'system wide':
4,237,983 unc_cbo_xsnp_response.miss_eviction # 3668558131345.00 test_metric_inc
218,643 unc_cbo_xsnp_response.miss_xcore
4,254,148 unc_cbo_xsnp_response.miss_eviction
213,352 unc_cbo_xsnp_response.miss_xcore
1.000938151 seconds time elapsed
Signed-off-by: John Garry <[email protected]>
Acked-by: Kajol Jain <[email protected]>
Cc: Alexander Shishkin <[email protected]>
Cc: Andi Kleen <[email protected]>
Cc: Ian Rogers <[email protected]>
Cc: Jiri Olsa <[email protected]>
Cc: Joakim Zhang <[email protected]>
Cc: Kan Liang <[email protected]>
Cc: Kim Phillips <[email protected]>
Cc: Leo Yan <[email protected]>
Cc: Mark Rutland <[email protected]>
Cc: Mathieu Poirier <[email protected]>
Cc: Namhyung Kim <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Shaokun Zhang <[email protected]>
Cc: Will Deacon <[email protected]>
Cc: [email protected]
Cc: [email protected]
Link: http://lore.kernel.org/lkml/[email protected]
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
|
|
Function find_evsel_group() expects events to be ordered such that they
are grouped after their leader.
Modify evlist__splice_list_tail() to guarantee this (ordering).
[Should prob also change the function name]
Signed-off-by: John Garry <[email protected]>
Acked-by: Kajol Jain <[email protected]>
Cc: Alexander Shishkin <[email protected]>
Cc: Andi Kleen <[email protected]>
Cc: Ian Rogers <[email protected]>
Cc: Jiri Olsa <[email protected]>
Cc: Joakim Zhang <[email protected]>
Cc: Kan Liang <[email protected]>
Cc: Kim Phillips <[email protected]>
Cc: Leo Yan <[email protected]>
Cc: Mark Rutland <[email protected]>
Cc: Mathieu Poirier <[email protected]>
Cc: Namhyung Kim <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Shaokun Zhang <[email protected]>
Cc: Will Deacon <[email protected]>
Cc: [email protected]
Cc: [email protected]
Link: http://lore.kernel.org/lkml/[email protected]
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
|
|
Add pmu_add_sys_aliases() to add system PMU events aliases.
For adding system PMU events, iterate through all the events for all SoC
event tables in pmu_sys_event_tables[].
Matches must satisfy both:
- PMU identifier matches event "compat" value
- event "Unit" member must match, same as uncore event aliases matched by
CPUID
Signed-off-by: John Garry <[email protected]>
Acked-by: Kajol Jain <[email protected]>
Cc: Alexander Shishkin <[email protected]>
Cc: Andi Kleen <[email protected]>
Cc: Ian Rogers <[email protected]>
Cc: Jiri Olsa <[email protected]>
Cc: Joakim Zhang <[email protected]>
Cc: Kan Liang <[email protected]>
Cc: Kim Phillips <[email protected]>
Cc: Leo Yan <[email protected]>
Cc: Mark Rutland <[email protected]>
Cc: Mathieu Poirier <[email protected]>
Cc: Namhyung Kim <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Shaokun Zhang <[email protected]>
Cc: Will Deacon <[email protected]>
Cc: [email protected]
Cc: [email protected]
Link: http://lore.kernel.org/lkml/[email protected]
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
|
|
Add a function to read the PMU id sysfs entry. This is only done for uncore
PMUs where this would possibly be relevant.
Signed-off-by: John Garry <[email protected]>
Acked-by: Kajol Jain <[email protected]>
Cc: Alexander Shishkin <[email protected]>
Cc: Andi Kleen <[email protected]>
Cc: Ian Rogers <[email protected]>
Cc: Jiri Olsa <[email protected]>
Cc: Joakim Zhang <[email protected]>
Cc: Kan Liang <[email protected]>
Cc: Kim Phillips <[email protected]>
Cc: Leo Yan <[email protected]>
Cc: Mark Rutland <[email protected]>
Cc: Mathieu Poirier <[email protected]>
Cc: Namhyung Kim <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Shaokun Zhang <[email protected]>
Cc: Will Deacon <[email protected]>
Cc: [email protected]
Cc: [email protected]
Link: http://lore.kernel.org/lkml/[email protected]
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
|
|
Process the JSONs to find support for "system" events, which are not
tied to a specific CPUID.
A "COMPAT" property is now used to match against the namespace ID from
the kernel PMU driver.
The generated pmu-events.c will now have 2 tables:
a. CPU events, as before.
b. New pmu_sys_event_tables[] table, which will have events matched to
specific SoCs.
It will look like this:
struct pmu_event pme_hisilicon_hip09_sys[] = {
{
.name = "cycles",
.compat = "0x00030736",
.event = "event=0",
.desc = "Clock cycles",
.topic = "smmu v3 pmcg",
.long_desc = "Clock cycles",
},
{
.name = "smmuv3_pmcg.l1_tlb",
.compat = "0x00030736",
.event = "event=0x8a",
.desc = "SMMUv3 PMCG l1_tlb. Unit: smmuv3_pmcg ",
.topic = "smmu v3 pmcg",
.long_desc = "SMMUv3 PMCG l1_tlb",
.pmu = "smmuv3_pmcg",
},
...
};
struct pmu_event pme_arm_cortex_a53[] = {
{
.name = "ext_mem_req",
.event = "event=0xc0",
.desc = "External memory request",
.topic = "memory",
},
{
.name = "ext_mem_req_nc",
.event = "event=0xc1",
.desc = "Non-cacheable external memory request",
.topic = "memory",
},
...
};
struct pmu_event pme_hisilicon_hip09_cpu[] = {
{
.name = "l2d_cache_refill_wr",
.event = "event=0x53",
.desc = "L2D cache refill, write",
.topic = "core imp def",
.long_desc = "Attributable Level 2 data cache refill, write",
},
...
};
struct pmu_events_map pmu_events_map[] = {
{
.cpuid = "0x00000000410fd030",
.version = "v1",
.type = "core",
.table = pme_arm_cortex_a53
},
{
.cpuid = "0x00000000480fd010",
.version = "v1",
.type = "core",
.table = pme_hisilicon_hip09_cpu
},
{
.table = 0
},
};
struct pmu_event pme_hisilicon_hip09_cpu[] = {
{
.name = "uncore_hisi_l3c.rd_cpipe",
.event = "event=0",
.desc = "Total read accesses. Unit: hisi_sccl,l3c ",
.topic = "uncore l3c",
.long_desc = "Total read accesses",
.pmu = "hisi_sccl,l3c",
},
{
.name = "uncore_hisi_l3c.wr_cpipe",
.event = "event=0x1",
.desc = "Total write accesses. Unit: hisi_sccl,l3c ",
.topic = "uncore l3c",
.long_desc = "Total write accesses",
.pmu = "hisi_sccl,l3c",
},
...
};
struct pmu_sys_events pmu_sys_event_tables[] = {
{
.table = pme_hisilicon_hip09_sys,
},
...
};
Committer notes:
Added the fix for architectures without PMU events, provided by John
after I reported the build failing in such systems.
Link: https://lore.kernel.org/lkml/[email protected]/
Signed-off-by: John Garry <[email protected]>
Acked-by: Kajol Jain <[email protected]>
Cc: Alexander Shishkin <[email protected]>
Cc: Andi Kleen <[email protected]>
Cc: Ian Rogers <[email protected]>
Cc: Jiri Olsa <[email protected]>
Cc: Joakim Zhang <[email protected]>
Cc: Kan Liang <[email protected]>
Cc: Kim Phillips <[email protected]>
Cc: Leo Yan <[email protected]>
Cc: Mark Rutland <[email protected]>
Cc: Mathieu Poirier <[email protected]>
Cc: Namhyung Kim <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Shaokun Zhang <[email protected]>
Cc: Will Deacon <[email protected]>
Cc: [email protected]
Cc: [email protected]
Link: http://lore.kernel.org/lkml/[email protected]
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
|
|
Currently only upto a level 2 directory is supported, in form
vendor/platform.
Add support for a further level, to support vendor/platform
sub-directories in future, which will be vendor/platform/cpu and
vendor/platform/sys.
Signed-off-by: John Garry <[email protected]>
Acked-by: Kajol Jain <[email protected]>
Cc: Alexander Shishkin <[email protected]>
Cc: Andi Kleen <[email protected]>
Cc: Ian Rogers <[email protected]>
Cc: Jiri Olsa <[email protected]>
Cc: Joakim Zhang <[email protected]>
Cc: Kan Liang <[email protected]>
Cc: Kim Phillips <[email protected]>
Cc: Leo Yan <[email protected]>
Cc: Mark Rutland <[email protected]>
Cc: Mathieu Poirier <[email protected]>
Cc: Namhyung Kim <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Shaokun Zhang <[email protected]>
Cc: Will Deacon <[email protected]>
Cc: [email protected]
Cc: [email protected]
Link: http://lore.kernel.org/lkml/[email protected]
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
|
|
sample_type bit
Before we had this unhelpful message:
$ perf record --data-page-size sleep 1
Error:
The sys_perf_event_open() syscall returned with 22 (Invalid argument) for event (cycles:u).
/bin/dmesg | grep -i perf may provide additional information.
$
Add support to the perf_missing_features variable to remember what
caused evsel__open() to fail and then use that information in
evsel__open_strerror().
$ perf record --data-page-size sleep 1
Error:
Asking for the data page size isn't supported by this kernel.
$
Cc: Kan Liang <[email protected]>
Cc: Namhyung Kim <[email protected]>
Cc: Andi Kleen <[email protected]>
Cc: Jiri Olsa <[email protected]>
Cc: Mark Rutland <[email protected]>
Cc: Michael Ellerman <[email protected]>
Cc: Stephane Eranian <[email protected]>
Cc: Will Deacon <[email protected]>
Link: http://lore.kernel.org/lkml/[email protected]
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
|
|
Support new sample type PERF_SAMPLE_DATA_PAGE_SIZE for page size.
Add new option --data-page-size to record sample data page size.
Signed-off-by: Kan Liang <[email protected]>
Acked-by: Namhyung Kim <[email protected]>
Cc: Andi Kleen <[email protected]>
Cc: Jiri Olsa <[email protected]>
Cc: Mark Rutland <[email protected]>
Cc: Michael Ellerman <[email protected]>
Cc: Stephane Eranian <[email protected]>
Cc: Will Deacon <[email protected]>
Link: http://lore.kernel.org/lkml/[email protected]
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
|
|
To get the changes in:
commit 8d97e71811aa ("perf/core: Add PERF_SAMPLE_DATA_PAGE_SIZE")
commit 995f088efebe ("perf/core: Add support for PERF_SAMPLE_CODE_PAGE_SIZE")
This silences this perf tools build warning:
Warning: Kernel ABI header at 'tools/include/uapi/linux/perf_event.h'
differs from latest version at 'include/uapi/linux/perf_event.h'
diff -u tools/include/uapi/linux/perf_event.h
include/uapi/linux/perf_event.h
Signed-off-by: Kan Liang <[email protected]>
Acked-by: Namhyung Kim <[email protected]>
Cc: Andi Kleen <[email protected]>
Cc: Jiri Olsa <[email protected]>
Cc: Mark Rutland <[email protected]>
Cc: Michael Ellerman <[email protected]>
Cc: Stephane Eranian <[email protected]>
Cc: Will Deacon <[email protected]>
Link: http://lore.kernel.org/lkml/[email protected]
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
|
|
elfutils needs to be provided main binary and separate debug info file
respectively. Providing separate debug info file instead of the main
binary is not sufficient.
One needs to try both supplied filename and its possible cache by its
build-id depending on the use case.
Signed-off-by: Jan Kratochvil <[email protected]>
Tested-by: Jiri Olsa <[email protected]>
Cc: Adrian Hunter <[email protected]>
Cc: David Ahern <[email protected]>
Cc: Ian Rogers <[email protected]>
Cc: Namhyung Kim <[email protected]>
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
|
|
When using 'perf record's option '-I' or '--user-regs=' along with
argument '?' to list available register names, memory of variable 'os'
allocated by strdup() needs to be released before __parse_regs()
returns, otherwise memory leak will occur.
Fixes: bcc84ec65ad1 ("perf record: Add ability to name registers to record")
Signed-off-by: Zheng Zengkai <[email protected]>
Acked-by: Jiri Olsa <[email protected]>
Cc: Alexander Shishkin <[email protected]>
Cc: Li Bin <[email protected]>
Cc: Mark Rutland <[email protected]>
Cc: Namhyung Kim <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
|
|
We're missing -lcap in test-all.bin target, so in case it's the only
library missing (if more are missing test-all.bin fails anyway), we will
falsely claim that we detected it and fail build, like:
$ make
...
Auto-detecting system features:
... dwarf: [ on ]
... dwarf_getlocations: [ on ]
... glibc: [ on ]
... libbfd: [ on ]
... libbfd-buildid: [ on ]
... libcap: [ on ]
... libelf: [ on ]
... libnuma: [ on ]
... numa_num_possible_cpus: [ on ]
... libperl: [ on ]
... libpython: [ on ]
... libcrypto: [ on ]
... libunwind: [ on ]
... libdw-dwarf-unwind: [ on ]
... zlib: [ on ]
... lzma: [ on ]
... get_cpuid: [ on ]
... bpf: [ on ]
... libaio: [ on ]
... libzstd: [ on ]
... disassembler-four-args: [ on ]
...
CC builtin-ftrace.o
In file included from builtin-ftrace.c:29:
util/cap.h:11:10: fatal error: sys/capability.h: No such file or directory
11 | #include <sys/capability.h>
| ^~~~~~~~~~~~~~~~~~
compilation terminated.
Fixes: 74d5f3d06f707eb5 ("tools build: Add capability-related feature detection")
Signed-off-by: Jiri Olsa <[email protected]>
Cc: Alexander Shishkin <[email protected]>
Cc: Ian Rogers <[email protected]>
Cc: Igor Lubashev <[email protected]>
Cc: Mark Rutland <[email protected]>
Cc: Michael Petlan <[email protected]>
Cc: Namhyung Kim <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Stephane Eranian <[email protected]>
Link: http://lore.kernel.org/lkml/[email protected]
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
|
|
Commit 991fcb77f490 ("drm/edid: Fix uninitialized variable in
drm_cvt_modes()") just replaced one warning with another.
The original warning about a possibly uninitialized variable was due to
the compiler not being smart enough to see that the case statement
actually enumerated all possible cases. And the initial fix was just to
add a "default" case that had a single "unreachable()", just to tell the
compiler that that situation cannot happen.
However, that doesn't actually fix the fundamental reason for the
problem: the compiler still doesn't see that the existing case
statements enumerate all possibilities, so the compiler will still
generate code to jump to that unreachable case statement. It just won't
complain about an uninitialized variable any more.
So now the compiler generates code to our inline asm marker that we told
it would not fall through, and end end result is basically random. We
have created a bridge to nowhere.
And then, depending on the random details of just exactly what the
compiler ends up doing, 'objtool' might end up complaining about the
conditional branches (for conditions that cannot happen, and that thus
will never be taken - but if the compiler was not smart enough to figure
that out, we can't expect objtool to do so) going off in the weeds.
So depending on how the compiler has laid out the result, you might see
something like this:
drivers/gpu/drm/drm_edid.o: warning: objtool: do_cvt_mode() falls through to next function drm_mode_detailed.isra.0()
and now you have a truly inscrutable warning that makes no sense at all
unless you start looking at whatever random code the compiler happened
to generate for our bare "unreachable()" statement.
IOW, don't use "unreachable()" unless you have an _active_ operation
that generates code that actually makes it obvious that something is not
reachable (ie an UD instruction or similar).
Solve the "compiler isn't smart enough" problem by just marking one of
the cases as "default", so that even when the compiler doesn't otherwise
see that we've enumerated all cases, the compiler will feel happy and
safe about there always being a valid case that initializes the 'width'
variable.
This also generates better code, since now the compiler doesn't generate
comparisons for five different possibilities (the four real ones and the
one that can't happen), but just for the three real ones and "the rest"
(which is that last one).
A smart enough compiler that sees that we cover all the cases won't care.
Cc: Lyude Paul <[email protected]>
Cc: Ilia Mirkin <[email protected]>
Cc: Josh Poimboeuf <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
|
|
Linux does not have a driver for / does not use the "Intel Baytrail
Mailbox Device" (ACIP HID INT33BD). Add it to the acpi_ignore_dep_ids
list, so that we do not defer probing ACPI devices which depend on
another ACPI device with this HID.
Specifically this makes us not defer the probing of the GPO1 ACPI
device / GPIO controller on the Acer Switch 10E SW3-016. On this
tablet model the _HID method of the ACPI node for the UART attached
Bluetooth, reads GPIOs to detect the installed wifi chip and updates
the reported _HID for the Bluetooth's ACPI node accordingly.
For the Bluetooth's ACPI node to report the correct _HID the GPO1 device
must be probed and attached during the first scan pass. Adding the
"INT33BD" HID to the acpi_ignore_dep_ids list makes this all work.
Signed-off-by: Hans de Goede <[email protected]>
Signed-off-by: Rafael J. Wysocki <[email protected]>
|
|
If there are no devices whose enumeration has been deferred after
the first pass in acpi_bus_scan(), the second pass is not necssary,
so avoid it with the help of a new static variable.
Signed-off-by: Rafael J. Wysocki <[email protected]>
Reviewed-by: Hans de Goede <[email protected]>
Tested-by: Hans de Goede <[email protected]>
Reviewed-by: Mika Westerberg <[email protected]>
|
|
In some cases ACPI control methods used during device enumeration
(such as _HID or _STA) may rely on Operation Region handlers
supplied by the drivers of other devices [1]:
An example of this is the Acer Switch 10E SW3-016 model. The _HID
method of the ACPI node for the UART attached Bluetooth, reads
GPIOs to detect the installed wifi chip and update the _HID for the
Bluetooth's ACPI node accordingly. The current ACPI scan code calls
_HID before the GPIO controller's OpRegions are available, leading
to the wrong _HID being used and Bluetooth not working.
In principle, in those cases there should be a _DEP control method
under the device object with OpRegion enumeration dependencies, so
deferring the enumeration of devices with _DEP returning a non-empty
list of suppliers of OpRegions depended on by the given device
(modulo some known exceptions that don't really supply any OpRegions
and are listed by _DEP for other reasons irrelevant for Linux) should
at least address the first-order dependencies by allowing the OpRegion
suppliers to be enumerated before their consumers.
Implement the above idea by modifying acpi_bus_scan() to enumerate
devices in the given scope of the ACPI namespace in two passes,
where the first pass covers the devices without "significant" lists
of dependencies coming from _DEP only and the second pass covers
all of the devices that were not enumerated in the first pass.
Take _DEP into account only for device objects with _HID, mostly in
order to avoid deferring the creation of ACPI device objects that
represent PCI devices and must be present during the enumeration
of the PCI bus (which takes place during the processing of the ACPI
device object that represents the host bridge), so that they can
be properly associated with the corresponding PCI devices.
Link: https://lore.kernel.org/linux-acpi/[email protected]/ # [1]
Reported-by: Hans de Goede <[email protected]>
Signed-off-by: Rafael J. Wysocki <[email protected]>
Reviewed-by: Hans de Goede <[email protected]>
Tested-by: Hans de Goede <[email protected]>
Reviewed-by: Mika Westerberg <[email protected]>
|
|
Evaluate _DEP before calling acpi_add_single_object() from
acpi_bus_check_add() and do that only for ACPI_BUS_TYPE_DEVICE
objects.
While at it, rename acpi_device_dep_initialize() to
acpi_scan_check_dep(), fix up a memory allocation statement in
that function, consistently treat memory allocation failures in
there as intermittent errors and make some related janitorial
changes in it.
This change will help to avoid calling acpi_add_single_object() if
there are unmet _DEP dependencies in the future, as that may cause
some control methods, potentially depending on the presence of
operation regions supplied by other devices, to be evaluated
prematurely.
Signed-off-by: Rafael J. Wysocki <[email protected]>
Reviewed-by: Hans de Goede <[email protected]>
Tested-by: Hans de Goede <[email protected]>
Reviewed-by: Mika Westerberg <[email protected]>
|
|
We don't need to take uring_lock for SQPOLL|IOPOLL to do
io_cqring_overflow_flush() when cq_overflow_list is empty, remove it
from the hot path.
Signed-off-by: Pavel Begunkov <[email protected]>
Signed-off-by: Jens Axboe <[email protected]>
|
|
There is only one user of it and the name is misleading, get rid of it
by inlining. By the way make overflow_flush's return value deduction
simpler.
Signed-off-by: Pavel Begunkov <[email protected]>
Signed-off-by: Jens Axboe <[email protected]>
|
|
Add a helper which calculates number of events in CQ. Handcoded version
of it in io_cqring_overflow_flush() is not the clearest thing, so it
makes it slightly more readable.
Signed-off-by: Pavel Begunkov <[email protected]>
Signed-off-by: Jens Axboe <[email protected]>
|
|
list_empty_careful() is not racy only if some conditions are met, i.e.
no re-adds after del_init. io_cqring_overflow_flush() does list_move(),
so it's actually racy.
Remove those checks, we have ->cq_check_overflow for the fast path.
Signed-off-by: Pavel Begunkov <[email protected]>
Signed-off-by: Jens Axboe <[email protected]>
|
|
io_uring_cancel_task_requests() doesn't imply that the ring is going
away, it may continue to work well after that. The problem is that it
sets ->cq_overflow_flushed effectively disabling the CQ overflow feature
Split setting cq_overflow_flushed from flush, and do the first one only
on exit. It's ok in terms of cancellations because there is a
io_uring->in_idle check in __io_cqring_fill_event().
It also fixes a race with setting ->cq_overflow_flushed in
io_uring_cancel_task_requests, whuch's is not atomic and a part of a
bitmask with other flags. Though, the only other flag that's not set
during init is drain_next, so it's not as bad for sane architectures.
Signed-off-by: Pavel Begunkov <[email protected]>
Fixes: 0f2122045b946 ("io_uring: don't rely on weak ->files references")
Signed-off-by: Jens Axboe <[email protected]>
|
|
It will be helpful to trace the iocg's whole state, including active and
idle state. And we can easily expand the original iocost_iocg_activate
trace event to support a state trace class, including active and idle
state tracing.
Acked-by: Tejun Heo <[email protected]>
Signed-off-by: Baolin Wang <[email protected]>
Signed-off-by: Jens Axboe <[email protected]>
|
|
The creation path of the NBD device respects max_part and only scans for
partitions if max_part is not 0. However, some other code paths ignore
max_part, and unconditionally scan for partitions. Add a check for
max_part on each partition scan.
Signed-off-by: Josh Triplett <[email protected]>
Reviewed-by: Josef Bacik <[email protected]>
Signed-off-by: Jens Axboe <[email protected]>
|
|
The kernel doesn't seem to have clear rules around this, but various
spots are using the mmap_sem to serialize access to modifying the
locked_vm count. Play it safe and lock the mm for write when accounting
or unaccounting locked memory.
Signed-off-by: Jens Axboe <[email protected]>
|
|
Add ACPI IDs for thermal drivers for Alder Lake support.
Signed-off-by: Srinivas Pandruvada <[email protected]>
Signed-off-by: Daniel Lezcano <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
|
|
Commit d3817a647059 ("pwm: sun4i: Remove redundant needs_delay") changed
the logic of an else branch so that the PWM_EN and PWM_CLK_GATING bits
are now cleared if the PWM is to be disabled, whereas previously the
condition was always false, and hence the branch never got executed.
This code is reported causing backlight issues on boards based on the
Allwinner A20 SoC. Fix this by removing the else branch, which restores
the behaviour prior to the offending commit.
Note that the PWM_EN and PWM_CLK_GATING bits still get cleared later in
sun4i_pwm_apply() if the PWM is to be disabled.
Fixes: d3817a647059 ("pwm: sun4i: Remove redundant needs_delay")
Reported-by: Taras Galchenko <[email protected]>
Suggested-by: Taras Galchenko <[email protected]>
Tested-by: Taras Galchenko <[email protected]>
Reviewed-by: Uwe Kleine-König <[email protected]>
Signed-off-by: Thierry Reding <[email protected]>
|
|
It is good practice to set the driver data before registering a device
with a subsystem because the subsystem or the driver core may call back
into the driver implementation. This is not currently an issue, but to
prevent future changes from causing this to break unexpectedly, make
sure that the driver data is set before the PWM chip registration.
Signed-off-by: Thierry Reding <[email protected]>
|
|
This is only defined with CONFIG_PWM unset and was introduced together
with pwmchip_add_with_polarity() (which is only defined with CONFIG_PWM
enabled). I guess the series that introduced pwmchip_add_with_polarity()
had a different concept in earlier revisions and the !CONFIG_PWM part
was just not updated accordingly.
Given that there is no implementation for pwmchip_add_with_polarity()
without CONFIG_PWM, just drop pwmchip_add_inversed() instead of renaming
it to pwmchip_add_with_polarity().
Signed-off-by: Uwe Kleine-König <[email protected]>
Acked-by: Lee Jones <[email protected]>
Signed-off-by: Thierry Reding <[email protected]>
|
|
The second parameter of do_div is an u32 and NSEC_PER_SEC * prescale
overflows this for bigger periods. Assuming the usual pwm input clk rate
of 66 MHz this happens starting at requested period > 606060 ns.
Splitting the division into two operations doesn't loose any precision.
It doesn't need to be feared that c / NSEC_PER_SEC doesn't fit into the
unsigned long variable "duty_cycles" because in this case the assignment
above to period_cycles would already have been overflowing as
period >= duty_cycle and then the calculation is moot anyhow.
Fixes: aef1a3799b5c ("pwm: imx27: Fix rounding behavior")
Signed-off-by: Uwe Kleine-König <[email protected]>
Tested-by: Johannes Pointner <[email protected]>
Signed-off-by: Thierry Reding <[email protected]>
|
|
Use the newer .apply function of pwm_ops instead of .config, .enable,
.disable and .set_polarity. This guarantees atomic changes of the pwm
controller configuration. It also reduces the size of the driver.
Since now period is a 64 bit value, add an extra check to reject periods
that exceed the possible max value for the 32 bit register.
This has been tested on a Raspberry PI 4.
Signed-off-by: Lino Sanfilippo <[email protected]>
Reviewed-by: Uwe Kleine-König <[email protected]>
Signed-off-by: Thierry Reding <[email protected]>
|
|
The driver used this construct:
#define KMB_PWM_LEADIN_MASK GENMASK(30, 0)
static inline void keembay_pwm_update_bits(struct keembay_pwm *priv, u32 mask,
u32 val, u32 offset)
{
u32 buff = readl(priv->base + offset);
buff = u32_replace_bits(buff, val, mask);
writel(buff, priv->base + offset);
}
...
keembay_pwm_update_bits(priv, KMB_PWM_LEADIN_MASK, 0,
KMB_PWM_LEADIN_OFFSET(pwm->hwpwm));
With CONFIG_CC_OPTIMIZE_FOR_SIZE the compiler (here: gcc 10.2.0) this
triggers:
In file included from /home/uwe/gsrc/linux/drivers/pwm/pwm-keembay.c:16:
In function ‘field_multiplier’,
inlined from ‘keembay_pwm_update_bits’ at /home/uwe/gsrc/linux/include/linux/bitfield.h:124:17:
/home/uwe/gsrc/linux/include/linux/bitfield.h:119:3: error: call to ‘__bad_mask’ declared with attribute error: bad bitfield mask
119 | __bad_mask();
| ^~~~~~~~~~~~
In function ‘field_multiplier’,
inlined from ‘keembay_pwm_update_bits’ at /home/uwe/gsrc/linux/include/linux/bitfield.h:154:1:
/home/uwe/gsrc/linux/include/linux/bitfield.h:119:3: error: call to ‘__bad_mask’ declared with attribute error: bad bitfield mask
119 | __bad_mask();
| ^~~~~~~~~~~~
The compiler doesn't seem to be able to notice that with field being
0x3ffffff the expression
if ((field | (field - 1)) & ((field | (field - 1)) + 1))
__bad_mask();
can be optimized away.
So use __always_inline and document the problem in a comment to fix
this.
Reported-by: kernel test robot <[email protected]>
Signed-off-by: Uwe Kleine-König <[email protected]>
Tested-by: Vijayakannan Ayyathurai <[email protected]>
Signed-off-by: Thierry Reding <[email protected]>
|
|
Permission bits are easier readable in octal than with using the
symbolic names.
Fixes the following warning generated by checkpatch:
WARNING: Symbolic permissions 'S_IRUGO' are not preferred. Consider using octal permissions '0444'.
#1341: FILE: drivers/pwm/core.c:1341:
+ debugfs_create_file("pwm", S_IFREG | S_IRUGO, NULL, NULL,
Signed-off-by: Soham Biswas <[email protected]>
Acked-by: Uwe Kleine-König <[email protected]>
Signed-off-by: Thierry Reding <[email protected]>
|
|
All used ACPI functions have dummy implementations, and there is no hard
dependency on x86.
Signed-off-by: Uwe Kleine-König <[email protected]>
Signed-off-by: Thierry Reding <[email protected]>
|
|
Drivers making use of IO remapping must depend on HAS_IOMEM.
Reported-by: kernel test robot <[email protected]>
Signed-off-by: Uwe Kleine-König <[email protected]>
Signed-off-by: Thierry Reding <[email protected]>
|
|
Instead of using a mix of -EOPNOTSUPP and -ENOTSUPP, use the more
standard -EINVAL to signal that the specified polarity value was
invalid.
Acked-by: Uwe Kleine-König <[email protected]>
Acked-by: Lee Jones <[email protected]>
Signed-off-by: Thierry Reding <[email protected]>
|
|
A single blank line is enough to separate logical code blocks.
Acked-by: Uwe Kleine-König <[email protected]>
Acked-by: Lee Jones <[email protected]>
Signed-off-by: Thierry Reding <[email protected]>
|
|
Using gotos for conditional code complicates this code significantly.
Convert the code to simple conditional blocks to increase readability.
Suggested-by: Uwe Kleine-König <[email protected]>
Acked-by: Uwe Kleine-König <[email protected]>
Acked-by: Lee Jones <[email protected]>
Signed-off-by: Thierry Reding <[email protected]>
|