| Age | Commit message (Collapse) | Author |
|---|
|
Cross-merge networking fixes after downstream PR.
Conflicts:
drivers/net/ethernet/ti/icssg/icssg_prueth.c
net/mac80211/chan.c
89884459a0b9 ("wifi: mac80211: fix idle calculation with multi-link")
87f5500285fb ("wifi: mac80211: simplify ieee80211_assign_link_chanctx()")
https://lore.kernel.org/all/20240422105623.7b1fbda2@canb.auug.org.au/
net/unix/garbage.c
1971d13ffa84 ("af_unix: Suppress false-positive lockdep splat for spin_lock() in __unix_gc().")
4090fa373f0e ("af_unix: Replace garbage collection algorithm.")
drivers/net/ethernet/ti/icssg/icssg_prueth.c
drivers/net/ethernet/ti/icssg/icssg_common.c
4dcd0e83ea1d ("net: ti: icssg-prueth: Fix signedness bug in prueth_init_rx_chns()")
e2dc7bfd677f ("net: ti: icssg-prueth: Move common functions into a separate file")
No adjacent changes.
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
I got a report for a failure in BPF verifier on a recent kernel with
perf lock contention command. It checks task->sighand->siglock without
checking if sighand is NULL or not. Let's add one.
; if (&curr->sighand->siglock == (void *)lock)
265: (79) r1 = *(u64 *)(r0 +2624) ; frame1: R0_w=trusted_ptr_task_struct(off=0,imm=0)
; R1_w=rcu_ptr_or_null_sighand_struct(off=0,imm=0)
266: (b7) r2 = 0 ; frame1: R2_w=0
267: (0f) r1 += r2
R1 pointer arithmetic on rcu_ptr_or_null_ prohibited, null-check it first
processed 164 insns (limit 1000000) max_states_per_insn 1 total_states 15 peak_states 15 mark_read 5
-- END PROG LOAD LOG --
libbpf: prog 'contention_end': failed to load: -13
libbpf: failed to load object 'lock_contention_bpf'
libbpf: failed to load BPF skeleton 'lock_contention_bpf': -13
Failed to load lock-contention BPF skeleton
lock contention BPF setup failed
lock contention did not detect any lock contention
Fixes: 1811e82767dcc ("perf lock contention: Track and show siglock with address")
Reviewed-by: Ian Rogers <irogers@google.com>
Acked-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Song Liu <song@kernel.org>
Cc: bpf@vger.kernel.org
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Link: https://lore.kernel.org/r/20240409225542.1870999-1-namhyung@kernel.org
|
|
The symbol__annotate2() initializes some data structures needed by TUI.
It has a logic to prevent calling it multiple times by checking if it
has the annotated source. But data type profiling uses a different
code (symbol__annotate) to allocate the annotated lines in advance.
So TUI missed to call symbol__annotate2() when it shows the annotation
browser.
Make symbol__annotate() reentrant and handle that situation properly.
This fixes a crash in the annotation browser started by perf report in
TUI like below.
$ perf report -s type,sym --tui
# and press 'a' key and then move down
Fixes: 81e57deec325 ("perf report: Support data type profiling")
Reviewed-by: Ian Rogers <irogers@google.com>
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Link: https://lore.kernel.org/r/20240405211800.1412920-2-namhyung@kernel.org
|
|
Avoid open-coding that simple expression each time by moving
BYTES_TO_BITS() from the probes code to <linux/bitops.h> to export
it to the rest of the kernel.
Simplify the macro while at it. `BITS_PER_LONG / sizeof(long)` always
equals to %BITS_PER_BYTE, regardless of the target architecture.
Do the same for the tools ecosystem as well (incl. its version of
bitops.h). The previous implementation had its implicit type of long,
while the new one is int, so adjust the format literal accordingly in
the perf code.
Suggested-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Reviewed-by: Przemek Kitszel <przemyslaw.kitszel@intel.com>
Acked-by: Yury Norov <yury.norov@gmail.com>
Signed-off-by: Alexander Lobakin <aleksander.lobakin@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
Reviewed-by: Ian Rogers <irogers@google.com>
Reviewed-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Andi Kleen <ak@linux.intel.com>
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Link: https://lore.kernel.org/r/20240304230815.1440583-5-namhyung@kernel.org
|
|
It's not used anymore and the code is coverted to use a hash map. Now
sym_hist has a static size, so no need to have sizeof_sym_hist in the
struct annotated_source.
Reviewed-by: Ian Rogers <irogers@google.com>
Reviewed-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Andi Kleen <ak@linux.intel.com>
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Link: https://lore.kernel.org/r/20240304230815.1440583-4-namhyung@kernel.org
|
|
Use annotated_source.samples hashmap instead of addr array in the
struct sym_hist.
Reviewed-by: Ian Rogers <irogers@google.com>
Reviewed-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Andi Kleen <ak@linux.intel.com>
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Link: https://lore.kernel.org/r/20240304230815.1440583-3-namhyung@kernel.org
|
|
Now symbol histogram uses an array to save per-offset sample counts.
But it wastes a lot of memory if the symbol has a few samples only.
Add a hashmap to save values only for actual samples.
For now, it has duplicate histogram (one in the existing array and
another in the new hash map). Once it can convert to use the hash
in all places, we can get rid of the array later.
Reviewed-by: Ian Rogers <irogers@google.com>
Reviewed-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Andi Kleen <ak@linux.intel.com>
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Link: https://lore.kernel.org/r/20240304230815.1440583-2-namhyung@kernel.org
|
|
The threads data structure is an array of hashmaps, previously
rbtrees. The two levels allows for a fixed outer array where access is
guarded by rw_semaphores. Commit 91e467bc568f ("perf machine: Use
hashtable for machine threads") sized the outer table at 256 entries
to avoid future scalability problems, however, this means the threads
struct is sized at 30,720 bytes. As the hashmaps allow O(1) access for
the common find/insert/remove operations, lower the number of entries
to 8. This reduces the size overhead to 960 bytes.
Signed-off-by: Ian Rogers <irogers@google.com>
Acked-by: Namhyung Kim <namhyung@kernel.org>
Cc: Yang Jihong <yangjihong1@huawei.com>
Cc: Oliver Upton <oliver.upton@linux.dev>
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Link: https://lore.kernel.org/r/20240301053646.1449657-8-irogers@google.com
|
|
The rbtree provides a sorting on entries but this is unused. Switch to
using hashmap for O(1) rather than O(log n) find/insert/remove
complexity.
Signed-off-by: Ian Rogers <irogers@google.com>
Acked-by: Namhyung Kim <namhyung@kernel.org>
Cc: Yang Jihong <yangjihong1@huawei.com>
Cc: Oliver Upton <oliver.upton@linux.dev>
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Link: https://lore.kernel.org/r/20240301053646.1449657-7-irogers@google.com
|
|
Move threads out of machine and into its own file.
Signed-off-by: Ian Rogers <irogers@google.com>
Acked-by: Namhyung Kim <namhyung@kernel.org>
Cc: Yang Jihong <yangjihong1@huawei.com>
Cc: Oliver Upton <oliver.upton@linux.dev>
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Link: https://lore.kernel.org/r/20240301053646.1449657-6-irogers@google.com
|
|
Move thread_rb_node into the machine.c file. This hides the
implementation of threads from the rest of the code allowing for it to
be refactored.
Locking discipline is tightened up in this change. As the lock is now
encapsulated in threads, the findnew function requires holding it (as
it already did in machine). Rather than do conditionals with locks
based on whether the thread should be created (which could potentially
be error prone with a read lock match with a write unlock), have a
separate threads__find that won't create the thread and only holds the
read lock. This effectively duplicates the findnew logic, with the
existing findnew logic only operating under a write lock assuming
creation is necessary as a previous find failed. The creation may
still fail with the write lock due to another thread. The duplication
is removed in a later next patch that delegates the implementation to
hashtable.
Signed-off-by: Ian Rogers <irogers@google.com>
Acked-by: Namhyung Kim <namhyung@kernel.org>
Cc: Yang Jihong <yangjihong1@huawei.com>
Cc: Oliver Upton <oliver.upton@linux.dev>
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Link: https://lore.kernel.org/r/20240301053646.1449657-5-irogers@google.com
|
|
Avoid exposing the threads data structure by switching to the callback
machine__for_each_thread approach. machine__fprintf is only used in
tests and verbose >3 output so don't turn to list and sort. Add
machine__threads_nr to be refactored later.
Note, all existing *_fprintf routines ignore fprintf errors.
Signed-off-by: Ian Rogers <irogers@google.com>
Acked-by: Namhyung Kim <namhyung@kernel.org>
Cc: Yang Jihong <yangjihong1@huawei.com>
Cc: Oliver Upton <oliver.upton@linux.dev>
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Link: https://lore.kernel.org/r/20240301053646.1449657-4-irogers@google.com
|
|
Commit 91e467bc568f ("perf machine: Use hashtable for machine
threads") made the iteration of thread tids unordered. The perf trace
--summary output sorts and prints each hash bucket, rather than all
threads globally. Change this behavior by turn all threads into a
list, sort the list by number of trace events then by tids, finally
print the list. This also allows the rbtree in threads to be not
accessed outside of machine.
Signed-off-by: Ian Rogers <irogers@google.com>
Acked-by: Namhyung Kim <namhyung@kernel.org>
Cc: Yang Jihong <yangjihong1@huawei.com>
Cc: Oliver Upton <oliver.upton@linux.dev>
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Link: https://lore.kernel.org/r/20240301053646.1449657-3-irogers@google.com
|
|
Commit 91e467bc568f ("perf machine: Use hashtable for machine
threads") made the iteration of thread tids unordered. The perf report
--tasks output now shows child threads in an order determined by the
hashing. For example, in this snippet tid 3 appears after tid 256 even
though they have the same ppid 2:
```
$ perf report --tasks
% pid tid ppid comm
0 0 -1 |swapper
2 2 0 | kthreadd
256 256 2 | kworker/12:1H-k
693761 693761 2 | kworker/10:1-mm
1301762 1301762 2 | kworker/1:1-mm_
1302530 1302530 2 | kworker/u32:0-k
3 3 2 | rcu_gp
...
```
The output is easier to read if threads appear numerically
increasing. To allow for this, read all threads into a list then sort
with a comparator that orders by the child task's of the first common
parent. The list creation and deletion are created as utilities on
machine. The indentation is possible by counting the number of
parents a child has.
With this change the output for the same data file is now like:
```
$ perf report --tasks
% pid tid ppid comm
0 0 -1 |swapper
1 1 0 | systemd
823 823 1 | systemd-journal
853 853 1 | systemd-udevd
3230 3230 1 | systemd-timesyn
3236 3236 1 | auditd
3239 3239 3236 | audisp-syslog
3321 3321 1 | accounts-daemon
...
```
Signed-off-by: Ian Rogers <irogers@google.com>
Acked-by: Namhyung Kim <namhyung@kernel.org>
Cc: Yang Jihong <yangjihong1@huawei.com>
Cc: Oliver Upton <oliver.upton@linux.dev>
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Link: https://lore.kernel.org/r/20240301053646.1449657-2-irogers@google.com
|
|
The find will get the map, ensure puts are done on all paths.
Signed-off-by: Ian Rogers <irogers@google.com>
Acked-by: Namhyung Kim <namhyung@kernel.org>
Cc: Masami Hiramatsu <mhiramat@kernel.org>
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Link: https://lore.kernel.org/r/20240229062048.558799-1-irogers@google.com
|
|
Currently it accounts the contention using delta between timestamps in
lock:contention_begin and lock:contention_end tracepoints. But it means
the lock should see the both events during the monitoring period.
Actually there are 4 cases that happen with the monitoring:
monitoring period
/ \
| |
1: B------+-----------------------+--------E
2: B----+-------------E |
3: | B-----------+----E
4: | B-------------E |
| |
t0 t1
where B and E mean contention BEGIN and END, respectively. So it only
accounts the case 4 for now. It seems there's no way to handle the case
1. The case 2 might be handled if it saved the timestamp (t0), but it
lacks the information from the B notably the flags which shows the lock
types. Also it could be a nested lock which it currently ignores. So
I think we should ignore the case 2.
However we can handle the case 3 if we save the timestamp (t1) at the
end of the period. And then it can iterate the map entries in the
userspace and update the lock stat accordinly.
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Reviewed-by: Ian Rogers <irogers@google.com>
Reviwed-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Song Liu <song@kernel.org>
Cc: bpf@vger.kernel.org
Link: https://lore.kernel.org/r/20240228053335.312776-1-namhyung@kernel.org
|
|
A metric may have no events, for example, the transaction metrics on
x86 are dependent on there being TSX events. Fix a segv where an evsel
of NULL is dereferenced for a metric leader value.
Fixes: a59fb796a36b ("perf metrics: Compute unmerged uncore metrics individually")
Signed-off-by: Ian Rogers <irogers@google.com>
Reviewed-by: Kan Liang <kan.liang@linux.intel.com>
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Link: https://lore.kernel.org/r/20240224011420.3066322-2-irogers@google.com
|
|
The metric match function fails for cases like looking for "metric" in
the string "all;foo_metric;metric" as the "metric" in "foo_metric"
matches but isn't preceeded by a ';'. Fix this by matching the first
list item and recursively matching on failure the next item after a
semicolon.
Signed-off-by: Ian Rogers <irogers@google.com>
Reviewed-by: Kan Liang <kan.liang@linux.intel.com>
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Link: https://lore.kernel.org/r/20240224011420.3066322-1-irogers@google.com
|
|
The commit in Fixes has reordered some code, but missed an error handling
path.
'goto err' now, in order to avoid a memory leak in case of error.
Fixes: f63a536f03a2 ("perf pmu: Merge JSON events with sysfs at load time")
Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
Reviewed-by: Ian Rogers <irogers@google.com>
Cc: kernel-janitors@vger.kernel.org
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Link: https://lore.kernel.org/r/9538b2b634894c33168dfe9d848d4df31fd4d801.1693085544.git.christophe.jaillet@wanadoo.fr
|
|
Arm64 doesn't have Model in /proc/cpuinfo and, thus, cpu_desc doesn't get
assigned.
Running
$ perf data convert --to-json perf.data.json
ends up calling output_json_string() with NULL pointer, which causes a
segmentation fault.
Signed-off-by: Ilkka Koskinen <ilkka@os.amperecomputing.com>
Acked-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: James Clark <james.clark@arm.com>
Cc: Evgeny Pistun <kotborealis@awooo.ru>
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Link: https://lore.kernel.org/r/20240223220458.15282-1-ilkka@os.amperecomputing.com
|
|
Currently the perf tool doesn't detect support for extended event types
on Apple M1/M2 systems, and will not auto-expand plain PERF_EVENT_TYPE
hardware events into per-PMU events. This is due to the detection of
extended event types not handling mandatory filters required by the
M1/M2 PMU driver.
PMU drivers and the core perf_events code can require that
perf_event_attr::exclude_* filters are configured in a specific way and
may reject certain configurations of filters, for example:
(a) Many PMUs lack support for any event filtering, and require all
perf_event_attr::exclude_* bits to be clear. This includes Alpha's
CPU PMU, and ARM CPU PMUs prior to the introduction of PMUv2 in
ARMv7,
(b) When /proc/sys/kernel/perf_event_paranoid >= 2, the perf core
requires that perf_event_attr::exclude_kernel is set.
(c) The Apple M1/M2 PMU requires that perf_event_attr::exclude_guest is
set as the hardware PMU does not count while a guest is running (but
might be extended in future to do so).
In is_event_supported(), we try to account for cases (a) and (b), first
attempting to open an event without any filters, and if this fails,
retrying with perf_event_attr::exclude_kernel set. We do not account for
case (c), or any other filters that drivers could theoretically require
to be set.
Thus is_event_supported() will fail to detect support for any events
targeting an Apple M1/M2 PMU, even where events would be supported with
perf_event_attr:::exclude_guest set.
Since commit:
82fe2e45cdb00de4 ("perf pmus: Check if we can encode the PMU number in perf_event_attr.type")
... we use is_event_supported() to detect support for extended types,
with the PMU ID encoded into the perf_event_attr::type. As above, on an
Apple M1/M2 system this will always fail to detect that the event is
supported, and consequently we fail to detect support for extended types
even when these are supported, as they have been since commit:
5c816728651ae425 ("arm_pmu: Add PERF_PMU_CAP_EXTENDED_HW_TYPE capability")
Due to this, the perf tool will not automatically expand plain
PERF_TYPE_HARDWARE events into per-PMU events, even when all the
necessary kernel support is present.
This patch updates is_event_supported() to additionally try opening
events with perf_event_attr::exclude_guest set, allowing support for
events to be detected on Apple M1/M2 systems. I believe that this is
sufficient for all contemporary CPU PMU drivers, though in future it may
be necessary to check for other combinations of filter bits.
I've deliberately changed the check to not expect a specific error code
for missing filters, as today ;the kernel may return a number of
different error codes for missing filters (e.g. -EACCESS, -EINVAL, or
-EOPNOTSUPP) depending on why and where the filter configuration is
rejected, and retrying for any error is more robust.
Note that this does not remove the need for commit:
a24d9d9dc096fc0d ("perf parse-events: Make legacy events lower priority than sysfs/JSON")
... which is still necessary so that named-pmu/event/ events work on
kernels without extended type support, even if the event name happens to
be the same as a PERF_EVENT_TYPE_HARDWARE event (e.g. as is the case for
the M1/M2 PMU's 'cycles' and 'instructions' events).
Fixes: 82fe2e45cdb00de4 ("perf pmus: Check if we can encode the PMU number in perf_event_attr.type")
Signed-off-by: Mark Rutland <mark.rutland@arm.com>
Tested-by: Ian Rogers <irogers@google.com>
Tested-by: James Clark <james.clark@arm.com>
Tested-by: Marc Zyngier <maz@kernel.org>
Cc: Hector Martin <marcan@marcan.st>
Cc: James Clark <james.clark@arm.com>
Cc: John Garry <john.g.garry@oracle.com>
Cc: Leo Yan <leo.yan@linux.dev>
Cc: Mike Leach <mike.leach@linaro.org>
Cc: Suzuki K Poulose <suzuki.poulose@arm.com>
Cc: Thomas Richter <tmricht@linux.ibm.com>
Cc: Will Deacon <will@kernel.org>
Cc: linux-arm-kernel@lists.infradead.org
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Link: https://lore.kernel.org/r/20240126145605.1005472-1-mark.rutland@arm.com
|
|
scandirat is used during the printing of tracepoint events but may be
missing from certain libcs. Add a compatibility implementation that
uses the symlink of an fd in /proc as a path for the reliably present
scandir.
Signed-off-by: Ian Rogers <irogers@google.com>
Cc: James Clark <james.clark@arm.com>
Cc: Justin Stitt <justinstitt@google.com>
Cc: Bill Wendling <morbo@google.com>
Cc: Nick Desaulniers <ndesaulniers@google.com>
Cc: Yang Jihong <yangjihong1@huawei.com>
Cc: Nathan Chancellor <nathan@kernel.org>
Cc: Kan Liang <kan.liang@linux.intel.com>
Cc: Athira Jajeev <atrajeev@linux.vnet.ibm.com>
Cc: llvm@lists.linux.dev
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Link: https://lore.kernel.org/r/20240221034155.1500118-3-irogers@google.com
|
|
Scanning /proc is inherently racy. Scanning /proc/pid/task within that
is also racy as the pid can terminate. Rather than failing in
__thread_map__new_all_cpus, skip pids for such failures.
Signed-off-by: Ian Rogers <irogers@google.com>
Cc: James Clark <james.clark@arm.com>
Cc: Justin Stitt <justinstitt@google.com>
Cc: Bill Wendling <morbo@google.com>
Cc: Nick Desaulniers <ndesaulniers@google.com>
Cc: Yang Jihong <yangjihong1@huawei.com>
Cc: Nathan Chancellor <nathan@kernel.org>
Cc: Kan Liang <kan.liang@linux.intel.com>
Cc: Athira Jajeev <atrajeev@linux.vnet.ibm.com>
Cc: llvm@lists.linux.dev
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Link: https://lore.kernel.org/r/20240221034155.1500118-2-irogers@google.com
|
|
Aggregation index was being computed using the evsel's cpumap which
may have a different (typically the same or fewer) entries.
Before:
```
$ perf stat --metric-only -A -M memory_bandwidth_total -a sleep 1
Performance counter stats for 'system wide':
MB/s memory_bandwidth_total MB/s memory_bandwidth_total MB/s memory_bandwidth_total MB/s memory_bandwidth_total MB/s memory_bandwidth_total MB/s memory_bandwidth_total
CPU0 12.8 0.0 12.9 12.7 0.0 12.6
CPU1
1.007806367 seconds time elapsed
```
After:
```
$ perf stat --metric-only -A -M memory_bandwidth_total -a sleep 1
Performance counter stats for 'system wide':
MB/s memory_bandwidth_total MB/s memory_bandwidth_total MB/s memory_bandwidth_total MB/s memory_bandwidth_total MB/s memory_bandwidth_total MB/s memory_bandwidth_total
CPU0 15.4 0.0 15.3 15.0 0.0 14.9
CPU18 0.0 0.0 13.5 5.2 0.0 11.9
1.007858736 seconds time elapsed
```
Signed-off-by: Ian Rogers <irogers@google.com> |
Acked-by: Namhyung Kim <namhyung@kernel.org>
Cc: K Prateek Nayak <kprateek.nayak@amd.com>
Cc: Stephane Eranian <eranian@google.com>
Cc: Kaige Ye <ye@kaige.org>
Cc: Kajol Jain <kjain@linux.ibm.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Kan Liang <kan.liang@linux.intel.com>
Cc: John Garry <john.g.garry@oracle.com>
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Link: https://lore.kernel.org/r/20240221070754.4163916-3-irogers@google.com
|
|
When merging counts from multiple uncore PMUs the metric is only
computed for the metric leader. When merging/aggregation is disabled,
prior to this patch just the leader's metric would be computed. Fix
this by computing the metric for each PMU.
On a SkylakeX:
Before:
```
$ perf stat -A -M memory_bandwidth_total -a sleep 1
Performance counter stats for 'system wide':
CPU0 82,217 UNC_M_CAS_COUNT.RD [uncore_imc_0] # 9.2 MB/s memory_bandwidth_total
CPU18 0 UNC_M_CAS_COUNT.RD [uncore_imc_0] # 0.0 MB/s memory_bandwidth_total
CPU0 61,395 UNC_M_CAS_COUNT.WR [uncore_imc_0]
CPU18 0 UNC_M_CAS_COUNT.WR [uncore_imc_0]
CPU0 0 UNC_M_CAS_COUNT.RD [uncore_imc_1]
CPU18 0 UNC_M_CAS_COUNT.RD [uncore_imc_1]
CPU0 0 UNC_M_CAS_COUNT.WR [uncore_imc_1]
CPU18 0 UNC_M_CAS_COUNT.WR [uncore_imc_1]
CPU0 81,570 UNC_M_CAS_COUNT.RD [uncore_imc_2]
CPU18 113,886 UNC_M_CAS_COUNT.RD [uncore_imc_2]
CPU0 62,330 UNC_M_CAS_COUNT.WR [uncore_imc_2]
CPU18 66,942 UNC_M_CAS_COUNT.WR [uncore_imc_2]
CPU0 75,489 UNC_M_CAS_COUNT.RD [uncore_imc_3]
CPU18 27,958 UNC_M_CAS_COUNT.RD [uncore_imc_3]
CPU0 55,864 UNC_M_CAS_COUNT.WR [uncore_imc_3]
CPU18 38,727 UNC_M_CAS_COUNT.WR [uncore_imc_3]
CPU0 0 UNC_M_CAS_COUNT.RD [uncore_imc_4]
CPU18 0 UNC_M_CAS_COUNT.RD [uncore_imc_4]
CPU0 0 UNC_M_CAS_COUNT.WR [uncore_imc_4]
CPU18 0 UNC_M_CAS_COUNT.WR [uncore_imc_4]
CPU0 75,423 UNC_M_CAS_COUNT.RD [uncore_imc_5]
CPU18 104,527 UNC_M_CAS_COUNT.RD [uncore_imc_5]
CPU0 57,596 UNC_M_CAS_COUNT.WR [uncore_imc_5]
CPU18 56,777 UNC_M_CAS_COUNT.WR [uncore_imc_5]
CPU0 1,003,440,851 ns duration_time
1.003440851 seconds time elapsed
```
After:
```
$ perf stat -A -M memory_bandwidth_total -a sleep 1
Performance counter stats for 'system wide':
CPU0 88,968 UNC_M_CAS_COUNT.RD [uncore_imc_0] # 9.5 MB/s memory_bandwidth_total
CPU18 0 UNC_M_CAS_COUNT.RD [uncore_imc_0] # 0.0 MB/s memory_bandwidth_total
CPU0 59,498 UNC_M_CAS_COUNT.WR [uncore_imc_0]
CPU18 0 UNC_M_CAS_COUNT.WR [uncore_imc_0]
CPU0 0 UNC_M_CAS_COUNT.RD [uncore_imc_1] # 0.0 MB/s memory_bandwidth_total
CPU18 0 UNC_M_CAS_COUNT.RD [uncore_imc_1] # 0.0 MB/s memory_bandwidth_total
CPU0 0 UNC_M_CAS_COUNT.WR [uncore_imc_1]
CPU18 0 UNC_M_CAS_COUNT.WR [uncore_imc_1]
CPU0 88,635 UNC_M_CAS_COUNT.RD [uncore_imc_2] # 9.5 MB/s memory_bandwidth_total
CPU18 117,975 UNC_M_CAS_COUNT.RD [uncore_imc_2] # 11.5 MB/s memory_bandwidth_total
CPU0 60,829 UNC_M_CAS_COUNT.WR [uncore_imc_2]
CPU18 62,105 UNC_M_CAS_COUNT.WR [uncore_imc_2]
CPU0 82,238 UNC_M_CAS_COUNT.RD [uncore_imc_3] # 8.7 MB/s memory_bandwidth_total
CPU18 22,906 UNC_M_CAS_COUNT.RD [uncore_imc_3] # 3.6 MB/s memory_bandwidth_total
CPU0 53,959 UNC_M_CAS_COUNT.WR [uncore_imc_3]
CPU18 32,990 UNC_M_CAS_COUNT.WR [uncore_imc_3]
CPU0 0 UNC_M_CAS_COUNT.RD [uncore_imc_4] # 0.0 MB/s memory_bandwidth_total
CPU18 0 UNC_M_CAS_COUNT.RD [uncore_imc_4] # 0.0 MB/s memory_bandwidth_total
CPU0 0 UNC_M_CAS_COUNT.WR [uncore_imc_4]
CPU18 0 UNC_M_CAS_COUNT.WR [uncore_imc_4]
CPU0 83,595 UNC_M_CAS_COUNT.RD [uncore_imc_5] # 8.9 MB/s memory_bandwidth_total
CPU18 110,151 UNC_M_CAS_COUNT.RD [uncore_imc_5] # 10.5 MB/s memory_bandwidth_total
CPU0 56,540 UNC_M_CAS_COUNT.WR [uncore_imc_5]
CPU18 53,816 UNC_M_CAS_COUNT.WR [uncore_imc_5]
CPU0 1,003,353,416 ns duration_time
```
Signed-off-by: Ian Rogers <irogers@google.com> |
Acked-by: Namhyung Kim <namhyung@kernel.org>
Cc: K Prateek Nayak <kprateek.nayak@amd.com>
Cc: Stephane Eranian <eranian@google.com>
Cc: Kaige Ye <ye@kaige.org>
Cc: Kajol Jain <kjain@linux.ibm.com>
Cc: Kan Liang <kan.liang@linux.intel.com>
Cc: John Garry <john.g.garry@oracle.com>
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Link: https://lore.kernel.org/r/20240221070754.4163916-2-irogers@google.com
|
|
Pass metric_expr and evsel rather than specific variables from the
struct, thereby reducing the number of arguments. This will enable
later fixes.
To reduce the size of the diff, local variables are added to match the
previous parameter names. This isn't done in the case of "name" as
evsel->name is more intention revealing. A whitespace issue is also
addressed.
Signed-off-by: Ian Rogers <irogers@google.com>
Acked-by: Namhyung Kim <namhyung@kernel.org>
Cc: K Prateek Nayak <kprateek.nayak@amd.com>
Cc: Stephane Eranian <eranian@google.com>
Cc: Kaige Ye <ye@kaige.org>
Cc: Kajol Jain <kjain@linux.ibm.com>
Cc: Kan Liang <kan.liang@linux.intel.com>
Cc: John Garry <john.g.garry@oracle.com>
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Link: https://lore.kernel.org/r/20240221070754.4163916-1-irogers@google.com
|
|
Currently, the instructions of samples are shown as raw hex strings
which are hard to read. x86 has a special option '--xed' to disassemble
the hex string via intel XED tool.
Here we use capstone as our disassembler engine to give more friendly
instructions. We select libcapstone because capstone can provide more
insn details. Perf will fallback to raw instructions if libcapstone is
not available.
The advantages compared to XED tool:
* Support arm, arm64, x86-32, x86_64 (more could be supported),
xed only for x86_64.
* Immediate address operands are shown as symbol+offs.
Signed-off-by: Changbin Du <changbin.du@huawei.com>
Reviewed-by: Adrian Hunter <adrian.hunter@intel.com>
Cc: changbin.du@gmail.com
Cc: Thomas Richter <tmricht@linux.ibm.com>
Cc: Andi Kleen <ak@linux.intel.com>
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Link: https://lore.kernel.org/r/20240217074046.4100789-3-changbin.du@huawei.com
|
|
I got a strange error on ARM to fail on processing FINISHED_ROUND
record. It turned out that it was failing in symbol__alloc_hist()
because the symbol size is too big.
When a sample is captured on a specific BPF program, it failed. I've
added a debug code and found the end address of the symbol is from
the next module which is placed far way.
ffff800008795778-ffff80000879d6d8: bpf_prog_1bac53b8aac4bc58_netcg_sock [bpf]
ffff80000879d6d8-ffff80000ad656b4: bpf_prog_76867454b5944e15_netcg_getsockopt [bpf]
ffff80000ad656b4-ffffd69b7af74048: bpf_prog_1d50286d2eb1be85_hn_egress [bpf] <---------- here
ffffd69b7af74048-ffffd69b7af74048: $x.5 [sha3_generic]
ffffd69b7af74048-ffffd69b7af740b8: crypto_sha3_init [sha3_generic]
ffffd69b7af740b8-ffffd69b7af741e0: crypto_sha3_update [sha3_generic]
The logic in symbols__fixup_end() just uses curr->start to update the
prev->end. But in this case, it won't work as it's too different.
I think ARM has a different kernel memory layout for modules and BPF
than on x86. Actually there's a logic to handle kernel and module
boundary. Let's do the same for symbols between different modules.
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Reviewed-by: Leo Yan <leo.yan@linux.dev>
Cc: Will Deacon <will@kernel.org>
Cc: Mike Leach <mike.leach@linaro.org>
Cc: John Garry <john.g.garry@oracle.com>
Link: https://lore.kernel.org/r/20240212233322.1855161-1-namhyung@kernel.org
|
|
If we instead decide to generate vmlinux.h from BTF info, it will be
there:
$ pahole timespec64
struct timespec64 {
time64_t tv_sec; /* 0 8 */
long int tv_nsec; /* 8 8 */
/* size: 16, cachelines: 1, members: 2 */
/* last cacheline: 16 bytes */
};
$
pahole manages to find it from /sys/kernel/btf/vmlinux, that is
generated from the kernel types.
With this linux/bpf.h doesn't need to be included, as its already in the
minimalistic tools/perf/util/bpf_skel/vmlinux/vmlinux.h file or what we
need comes when generating a vmlinux.h file from BTF info, i.e. when
using GEN_VMLINUX_H=1, as noticed by Namyung in a build break before
removing linux/bpf.h.
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Link: https://lore.kernel.org/r/Zc_fp6CgDClPhS_O@x1
|
|
Every architecture can provide a register list for sampling. If an
architecture doesn't support register sampling, it won't define the data
structure 'sample_reg_masks'. Consequently, any code using this
structure must be protected by the macro 'HAVE_PERF_REGS_SUPPORT'.
This patch defines a weak function, arch__sample_reg_masks(), which will
be replaced by an architecture-defined function for returning the
architecture's register list. With this refactoring, the function always
exists, the condition checking for 'HAVE_PERF_REGS_SUPPORT' is not
needed anymore, so remove it.
Signed-off-by: Leo Yan <leo.yan@linux.dev>
Reviewed-by: Ian Rogers <irogers@google.com>
Cc: James Clark <james.clark@arm.com>
Cc: Palmer Dabbelt <palmer@dabbelt.com>
Cc: Albert Ou <aou@eecs.berkeley.edu>
Cc: Huacai Chen <chenhuacai@kernel.org>
Cc: Guo Ren <guoren@kernel.org>
Cc: Will Deacon <will@kernel.org>
Cc: Mike Leach <mike.leach@linaro.org>
Cc: Kan Liang <kan.liang@linux.intel.com>
Cc: Ming Wang <wangming01@loongson.cn>
Cc: John Garry <john.g.garry@oracle.com>
Cc: Paul Walmsley <paul.walmsley@sifive.com>
Cc: linux-csky@vger.kernel.org
Cc: linux-riscv@lists.infradead.org
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Link: https://lore.kernel.org/r/20240214113947.240957-4-leo.yan@linux.dev
|
|
Currently, the macro HAVE_PERF_REGS_SUPPORT is used as a switch to turn
on or turn off the code of perf registers. If any architecture cannot
support perf register, it disables the perf register parsing, for both
the native parsing and cross parsing for other architectures.
To support both the native parsing and cross parsing, the tool should
always build the perf regs functions. Thus, this patch removes
HAVE_PERF_REGS_SUPPORT from the perf regs files.
Signed-off-by: Leo Yan <leo.yan@linux.dev>
Reviewed-by: Ian Rogers <irogers@google.com>
Cc: James Clark <james.clark@arm.com>
Cc: Palmer Dabbelt <palmer@dabbelt.com>
Cc: Albert Ou <aou@eecs.berkeley.edu>
Cc: Huacai Chen <chenhuacai@kernel.org>
Cc: Guo Ren <guoren@kernel.org>
Cc: Will Deacon <will@kernel.org>
Cc: Mike Leach <mike.leach@linaro.org>
Cc: Kan Liang <kan.liang@linux.intel.com>
Cc: Ming Wang <wangming01@loongson.cn>
Cc: John Garry <john.g.garry@oracle.com>
Cc: Paul Walmsley <paul.walmsley@sifive.com>
Cc: linux-csky@vger.kernel.org
Cc: linux-riscv@lists.infradead.org
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Link: https://lore.kernel.org/r/20240214113947.240957-3-leo.yan@linux.dev
|
|
Counts were switched from the scaled saved value form to the
aggregated count to avoid double accounting. When this happened the
removing of scaling for a count should have been removed, however, it
wasn't and this wasn't observed as it normally doesn't matter because
a counter's scale is 1. A problem was observed with RAPL events that
are scaled.
Fixes: 37cc8ad77cf8 ("perf metric: Directly use counts rather than saved_value")
Signed-off-by: Ian Rogers <irogers@google.com>
Reviewed-by: Kan Liang <kan.liang@linux.intel.com>
Cc: K Prateek Nayak <kprateek.nayak@amd.com>
Cc: James Clark <james.clark@arm.com>
Cc: Kaige Ye <ye@kaige.org>
Cc: John Garry <john.g.garry@oracle.com>
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Link: https://lore.kernel.org/r/20240209204947.3873294-5-irogers@google.com
|
|
Cycles is recognized as part of a hard coded metric in stat-shadow.c,
it may call print_metric_only with a NULL fmt string leading to a
segfault. Handle the NULL fmt explicitly.
Fixes: 088519f318be ("perf stat: Move the display functions to stat-display.c")
Signed-off-by: Ian Rogers <irogers@google.com>
Reviewed-by: Kan Liang <kan.liang@linux.intel.com>
Cc: K Prateek Nayak <kprateek.nayak@amd.com>
Cc: James Clark <james.clark@arm.com>
Cc: Kaige Ye <ye@kaige.org>
Cc: John Garry <john.g.garry@oracle.com>
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Link: https://lore.kernel.org/r/20240209204947.3873294-4-irogers@google.com
|
|
Events in metrics cannot use '/' as a separator, it would be
recognized as a divide, so they use '@'. The '@' is recognized in the
metricgroups code and changed to '/', do the same in the has_event
function so that the parsing is only tried without the @s.
Fixes: 4a4a9bf9075f ("perf expr: Add has_event function")
Signed-off-by: Ian Rogers <irogers@google.com>
Reviewed-by: Kan Liang <kan.liang@linux.intel.com>
Cc: K Prateek Nayak <kprateek.nayak@amd.com>
Cc: James Clark <james.clark@arm.com>
Cc: Kaige Ye <ye@kaige.org>
Cc: John Garry <john.g.garry@oracle.com>
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Link: https://lore.kernel.org/r/20240209204947.3873294-3-irogers@google.com
|
|
Currently only floating point numbers can be parsed, add a special
case for NaN.
Signed-off-by: Ian Rogers <irogers@google.com>
Reviewed-by: Kan Liang <kan.liang@linux.intel.com>
Cc: K Prateek Nayak <kprateek.nayak@amd.com>
Cc: James Clark <james.clark@arm.com>
Cc: Kaige Ye <ye@kaige.org>
Cc: John Garry <john.g.garry@oracle.com>
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Link: https://lore.kernel.org/r/20240209204947.3873294-2-irogers@google.com
|
|
After this change maps__nr_maps is only used by tests, existing users
are migrated to maps__empty. Compute maps__empty under the read lock.
Signed-off-by: Ian Rogers <irogers@google.com>
Acked-by: Namhyung Kim <namhyung@kernel.org>
Cc: K Prateek Nayak <kprateek.nayak@amd.com>
Cc: James Clark <james.clark@arm.com>
Cc: Vincent Whitchurch <vincent.whitchurch@axis.com>
Cc: Alexey Dobriyan <adobriyan@gmail.com>
Cc: Colin Ian King <colin.i.king@gmail.com>
Cc: Changbin Du <changbin.du@huawei.com>
Cc: Masami Hiramatsu <mhiramat@kernel.org>
Cc: Song Liu <song@kernel.org>
Cc: Leo Yan <leo.yan@linux.dev>
Cc: Athira Rajeev <atrajeev@linux.vnet.ibm.com>
Cc: Liam Howlett <liam.howlett@oracle.com>
Cc: Artem Savkov <asavkov@redhat.com>
Cc: bpf@vger.kernel.org
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Link: https://lore.kernel.org/r/20240210031746.4057262-7-irogers@google.com
|
|
Move the struct into the C file. Add maps__equal to work around
exposing the struct for reference count checking. Add accessors for
the unwind_libunwind_ops. Move maps_list_node to its only use in
symbol.c.
Signed-off-by: Ian Rogers <irogers@google.com>
Acked-by: Namhyung Kim <namhyung@kernel.org>
Cc: K Prateek Nayak <kprateek.nayak@amd.com>
Cc: James Clark <james.clark@arm.com>
Cc: Vincent Whitchurch <vincent.whitchurch@axis.com>
Cc: Alexey Dobriyan <adobriyan@gmail.com>
Cc: Colin Ian King <colin.i.king@gmail.com>
Cc: Changbin Du <changbin.du@huawei.com>
Cc: Masami Hiramatsu <mhiramat@kernel.org>
Cc: Song Liu <song@kernel.org>
Cc: Leo Yan <leo.yan@linux.dev>
Cc: Athira Rajeev <atrajeev@linux.vnet.ibm.com>
Cc: Liam Howlett <liam.howlett@oracle.com>
Cc: Artem Savkov <asavkov@redhat.com>
Cc: bpf@vger.kernel.org
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Link: https://lore.kernel.org/r/20240210031746.4057262-6-irogers@google.com
|
|
Finding a map is done under a lock, returning the map without a
reference count means it can be removed without notice and causing
uses after free. Grab a reference count to the map within the lock
region and return this. Fix up locations that need a map__put
following this.
Signed-off-by: Ian Rogers <irogers@google.com>
Acked-by: Namhyung Kim <namhyung@kernel.org>
Cc: K Prateek Nayak <kprateek.nayak@amd.com>
Cc: James Clark <james.clark@arm.com>
Cc: Vincent Whitchurch <vincent.whitchurch@axis.com>
Cc: Alexey Dobriyan <adobriyan@gmail.com>
Cc: Colin Ian King <colin.i.king@gmail.com>
Cc: Changbin Du <changbin.du@huawei.com>
Cc: Masami Hiramatsu <mhiramat@kernel.org>
Cc: Song Liu <song@kernel.org>
Cc: Leo Yan <leo.yan@linux.dev>
Cc: Athira Rajeev <atrajeev@linux.vnet.ibm.com>
Cc: Liam Howlett <liam.howlett@oracle.com>
Cc: Artem Savkov <asavkov@redhat.com>
Cc: bpf@vger.kernel.org
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Link: https://lore.kernel.org/r/20240210031746.4057262-5-irogers@google.com
|
|
Finding a map is done under a lock, returning the map without a
reference count means it can be removed without notice and causing
uses after free. Grab a reference count to the map within the lock
region and return this. Fix up locations that need a map__put
following this. Also fix some reference counted pointer comparisons.
Signed-off-by: Ian Rogers <irogers@google.com>
Acked-by: Namhyung Kim <namhyung@kernel.org>
Cc: K Prateek Nayak <kprateek.nayak@amd.com>
Cc: James Clark <james.clark@arm.com>
Cc: Vincent Whitchurch <vincent.whitchurch@axis.com>
Cc: Alexey Dobriyan <adobriyan@gmail.com>
Cc: Colin Ian King <colin.i.king@gmail.com>
Cc: Changbin Du <changbin.du@huawei.com>
Cc: Masami Hiramatsu <mhiramat@kernel.org>
Cc: Song Liu <song@kernel.org>
Cc: Leo Yan <leo.yan@linux.dev>
Cc: Athira Rajeev <atrajeev@linux.vnet.ibm.com>
Cc: Liam Howlett <liam.howlett@oracle.com>
Cc: Artem Savkov <asavkov@redhat.com>
Cc: bpf@vger.kernel.org
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Link: https://lore.kernel.org/r/20240210031746.4057262-4-irogers@google.com
|
|
Finding a map is done under a lock, returning the map without a
reference count means it can be removed without notice and causing
uses after free. Grab a reference count to the map within the lock
region and return this. Fix up locations that need a map__put
following this.
Signed-off-by: Ian Rogers <irogers@google.com>
Acked-by: Namhyung Kim <namhyung@kernel.org>
Cc: K Prateek Nayak <kprateek.nayak@amd.com>
Cc: James Clark <james.clark@arm.com>
Cc: Vincent Whitchurch <vincent.whitchurch@axis.com>
Cc: Alexey Dobriyan <adobriyan@gmail.com>
Cc: Colin Ian King <colin.i.king@gmail.com>
Cc: Changbin Du <changbin.du@huawei.com>
Cc: Masami Hiramatsu <mhiramat@kernel.org>
Cc: Song Liu <song@kernel.org>
Cc: Leo Yan <leo.yan@linux.dev>
Cc: Athira Rajeev <atrajeev@linux.vnet.ibm.com>
Cc: Liam Howlett <liam.howlett@oracle.com>
Cc: Artem Savkov <asavkov@redhat.com>
Cc: bpf@vger.kernel.org
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Link: https://lore.kernel.org/r/20240210031746.4057262-3-irogers@google.com
|
|
Maps is a collection of maps primarily sorted by the starting address
of the map. Prior to this change the maps were held in an rbtree
requiring 4 pointers per node. Prior to reference count checking, the
rbnode was embedded in the map so 3 pointers per node were
necessary. This change switches the rbtree to an array lazily sorted
by address, much as the array sorting nodes by name. 1 pointer is
needed per node, but to avoid excessive resizing the backing array may
be twice the number of used elements. Meaning the memory overhead is
roughly half that of the rbtree. For a perf record with
"--no-bpf-event -g -a" of true, the memory overhead of perf inject is
reduce fom 3.3MB to 3MB, so 10% or 300KB is saved.
Map inserts always happen at the end of the array. The code tracks
whether the insertion violates the sorting property. O(log n) rb-tree
complexity is switched to O(1).
Remove slides the array, so O(log n) rb-tree complexity is degraded to
O(n).
A find may need to sort the array using qsort which is O(n*log n), but
in general the maps should be sorted and so average performance should
be O(log n) as with the rbtree.
An rbtree node consumes a cache line, but with the array 4 nodes fit
on a cache line. Iteration is simplified to scanning an array rather
than pointer chasing.
Overall it is expected the performance after the change should be
comparable to before, but with half of the memory consumed.
To avoid a list and repeated logic around splitting maps,
maps__merge_in is rewritten in terms of
maps__fixup_overlap_and_insert. maps_merge_in splits the given mapping
inserting remaining gaps. maps__fixup_overlap_and_insert splits the
existing mappings, then adds the incoming mapping. By adding the new
mapping first, then re-inserting the existing mappings the splitting
behavior matches.
Signed-off-by: Ian Rogers <irogers@google.com>
Acked-by: Namhyung Kim <namhyung@kernel.org>
Cc: K Prateek Nayak <kprateek.nayak@amd.com>
Cc: James Clark <james.clark@arm.com>
Cc: Vincent Whitchurch <vincent.whitchurch@axis.com>
Cc: Alexey Dobriyan <adobriyan@gmail.com>
Cc: Colin Ian King <colin.i.king@gmail.com>
Cc: Changbin Du <changbin.du@huawei.com>
Cc: Masami Hiramatsu <mhiramat@kernel.org>
Cc: Song Liu <song@kernel.org>
Cc: Leo Yan <leo.yan@linux.dev>
Cc: Athira Rajeev <atrajeev@linux.vnet.ibm.com>
Cc: Liam Howlett <liam.howlett@oracle.com>
Cc: Artem Savkov <asavkov@redhat.com>
Cc: bpf@vger.kernel.org
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Link: https://lore.kernel.org/r/20240210031746.4057262-2-irogers@google.com
|
|
To get some fixes in the perf test and JSON metrics into the development
branch.
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
|
|
The child_process for addr2line sets in and out to -1 so that pipes
get created. It is the caller's responsibility to close the pipes,
finish_command doesn't do it. Add the missed closes.
Fixes: b3801e791231 ("perf srcline: Simplify addr2line subprocess")
Signed-off-by: Ian Rogers <irogers@google.com>
Cc: Ravi Bangoria <ravi.bangoria@amd.com>
Cc: James Clark <james.clark@arm.com>
Cc: Nick Desaulniers <ndesaulniers@google.com>
Cc: Nathan Chancellor <nathan@kernel.org>
Cc: Kan Liang <kan.liang@linux.intel.com>
Cc: John Garry <john.g.garry@oracle.com>
Cc: Tom Rix <trix@redhat.com>
Cc: llvm@lists.linux.dev
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Link: https://lore.kernel.org/r/20240201001504.1348511-8-irogers@google.com
|
|
Some platforms have 'cluster' topology and CPUs in the cluster will
share resources like L3 Cache Tag (for HiSilicon Kunpeng SoC) or L2
cache (for Intel Jacobsville). Currently parsing and building cluster
topology have been supported since [1].
perf stat has already supported aggregation for other topologies like
die or socket, etc. It'll be useful to aggregate per-cluster to find
problems like L3T bandwidth contention.
This patch add support for "--per-cluster" option for per-cluster
aggregation. Also update the docs and related test. The output will
be like:
[root@localhost tmp]# perf stat -a -e LLC-load --per-cluster -- sleep 5
Performance counter stats for 'system wide':
S56-D0-CLS158 4 1,321,521,570 LLC-load
S56-D0-CLS594 4 794,211,453 LLC-load
S56-D0-CLS1030 4 41,623 LLC-load
S56-D0-CLS1466 4 41,646 LLC-load
S56-D0-CLS1902 4 16,863 LLC-load
S56-D0-CLS2338 4 15,721 LLC-load
S56-D0-CLS2774 4 22,671 LLC-load
[...]
On a legacy system without cluster or cluster support, the output will
be look like:
[root@localhost perf]# perf stat -a -e cycles --per-cluster -- sleep 1
Performance counter stats for 'system wide':
S56-D0-CLS0 64 18,011,485 cycles
S7182-D0-CLS0 64 16,548,835 cycles
Note that this patch doesn't mix the cluster information in the outputs
of --per-core to avoid breaking any tools/scripts using it.
Note that perf recently supports "--per-cache" aggregation, but it's not
the same with the cluster although cluster CPUs may share some cache
resources. For example on my machine all clusters within a die share the
same L3 cache:
$ cat /sys/devices/system/cpu/cpu0/cache/index3/shared_cpu_list
0-31
$ cat /sys/devices/system/cpu/cpu0/topology/cluster_cpus_list
0-3
[1] commit c5e22feffdd7 ("topology: Represent clusters of CPUs within a die")
Tested-by: Jie Zhan <zhanjie9@hisilicon.com>
Reviewed-by: Tim Chen <tim.c.chen@linux.intel.com>
Reviewed-by: Ian Rogers <irogers@google.com>
Signed-off-by: Yicong Yang <yangyicong@hisilicon.com>
Cc: james.clark@arm.com
Cc: 21cnbao@gmail.com
Cc: prime.zeng@hisilicon.com
Cc: Jonathan.Cameron@huawei.com
Cc: fanghao11@huawei.com
Cc: linuxarm@huawei.com
Cc: tim.c.chen@intel.com
Cc: linux-arm-kernel@lists.infradead.org
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Link: https://lore.kernel.org/r/20240208024026.2691-1-yangyicong@huawei.com
|
|
When it converts sample IP to or from objdump-capable one, there's a
comment saying that kernel modules have DSO_SPACE__USER. But commit
02213cec64bb ("perf maps: Mark module DSOs with kernel type") changed
it and makes the comment confusing. Let's get rid of it.
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Acked-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Reviewed-by: Ian Rogers <irogers@google.com>
Link: https://lore.kernel.org/r/20240208181025.1329645-1-namhyung@kernel.org
|
|
slist needs to be freed in both error path and normal path in
thread_map__new_by_tid_str().
Fixes: b52956c961be3a04 ("perf tools: Allow multiple threads or processes in record, stat, top")
Reviewed-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: Yang Jihong <yangjihong1@huawei.com>
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Link: https://lore.kernel.org/r/20240206083228.172607-6-yangjihong1@huawei.com
|
|
Currently perf does not record module section addresses except for
the .text section. In general that means perf cannot get module section
mappings correct (except for .text) when loading symbols from a kernel
module file. (Note using --kcore does not have this issue)
Improve that situation slightly by identifying executable sections that
use the same mapping as the .text section. That happens when an
executable section comes directly after the .text section, both in memory
and on file, something that can be determined by following the same layout
rules used by the kernel, refer kernel layout_sections(). Note whether
that happens is somewhat arbitrary, so this is not a final solution.
Example from tracing a virtual machine process:
Before:
$ perf script | grep unknown
CPU 0/KVM 1718 203.511270: 318341 cpu-cycles:P: ffffffffc13e8a70 [unknown] (/lib/modules/6.7.2-local/kernel/arch/x86/kvm/kvm-intel.ko)
$ perf script -vvv 2>&1 >/dev/null | grep kvm.intel | grep 'noinstr.text\|ffff'
Map: 0-7e0 41430 [kvm_intel].noinstr.text
Map: ffffffffc13a7000-ffffffffc1421000 a0 /lib/modules/6.7.2-local/kernel/arch/x86/kvm/kvm-intel.ko
After:
$ perf script | grep 203.511270
CPU 0/KVM 1718 203.511270: 318341 cpu-cycles:P: ffffffffc13e8a70 vmx_vmexit+0x0 (/lib/modules/6.7.2-local/kernel/arch/x86/kvm/kvm-intel.ko)
$ perf script -vvv 2>&1 >/dev/null | grep kvm.intel | grep 'noinstr.text\|ffff'
Map: ffffffffc13a7000-ffffffffc1421000 a0 /lib/modules/6.7.2-local/kernel/arch/x86/kvm/kvm-intel.ko
Reported-by: Like Xu <like.xu.linux@gmail.com>
Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Acked-by: Namhyung Kim <namhyung@kernel.org>
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Link: https://lore.kernel.org/r/20240208085326.13432-3-adrian.hunter@intel.com
|
|
Dump kmaps if using 'perf --debug kmaps' or verbose > 2 (e.g. -vvv) for
tools 'perf script' and 'perf report' if there is no browser.
Example:
$ perf --debug kmaps script 2>&1 >/dev/null | grep kvm.intel
build id event received for /lib/modules/6.7.2-local/kernel/arch/x86/kvm/kvm-intel.ko: 0691d75e10e72ebbbd45a44c59f6d00a5604badf [20]
Map: 0-3a3 4f5d8 [kvm_intel].modinfo
Map: 0-5240 5f280 [kvm_intel]__versions
Map: 0-30 64 [kvm_intel].note.Linux
Map: 0-14 644c0 [kvm_intel].orc_header
Map: 0-5297 43680 [kvm_intel].rodata
Map: 0-5bee 3b837 [kvm_intel].text.unlikely
Map: 0-7e0 41430 [kvm_intel].noinstr.text
Map: 0-2080 713c0 [kvm_intel].bss
Map: 0-26 705c8 [kvm_intel].data..read_mostly
Map: 0-5888 6a4c0 [kvm_intel].data
Map: 0-22 70220 [kvm_intel].data.once
Map: 0-40 705f0 [kvm_intel].data..percpu
Map: 0-1685 41d20 [kvm_intel].init.text
Map: 0-4b8 6fd60 [kvm_intel].init.data
Map: 0-380 70248 [kvm_intel]__dyndbg
Map: 0-8 70218 [kvm_intel].exit.data
Map: 0-438 4f980 [kvm_intel]__param
Map: 0-5f5 4ca0f [kvm_intel].rodata.str1.1
Map: 0-3657 493b8 [kvm_intel].rodata.str1.8
Map: 0-e0 70640 [kvm_intel].data..ro_after_init
Map: 0-500 70ec0 [kvm_intel].gnu.linkonce.this_module
Map: ffffffffc13a7000-ffffffffc1421000 a0 /lib/modules/6.7.2-local/kernel/arch/x86/kvm/kvm-intel.ko
The example above shows how the module section mappings are all wrong
except for the main .text mapping at 0xffffffffc13a7000.
Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Acked-by: Namhyung Kim <namhyung@kernel.org>
Cc: Like Xu <like.xu.linux@gmail.com>
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Link: https://lore.kernel.org/r/20240208085326.13432-2-adrian.hunter@intel.com
|
|
perf script exposes the evsel_name to python scripts as part of the data
passed to the sample or tracepoint handler function, and it passes the id and
stream_id to the throttled/unthrottled handler functions. This makes matching
throttle events and samples difficult.
To make this possible, this change exposes the sample id and stream_id values
to the script.
Signed-off-by: Ben Gainey <ben.gainey@arm.com>
Reviewed-by: Adrian Hunter <adrian.hunter@intel.com>
Cc: will@kernel.org
Cc: linux-arm-kernel@lists.infradead.org
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Link: https://lore.kernel.org/r/20240123103137.1890779-2-ben.gainey@arm.com
|
