Command 'perf ftrace -v -- ls' fails in s390 (at least 5.12.0rc6).
The root cause is a missing pointer dereference which causes an
array element address to be used as PID.
Fix this by extracting the PID.
Output before:
# ./perf ftrace -v -- ls
function_graph tracer is used
write '-263732416' to tracing/set_ftrace_pid failed: Invalid argument
failed to set ftrace pid
#
Output after:
./perf ftrace -v -- ls
function_graph tracer is used
# tracer: function_graph
#
# CPU DURATION FUNCTION CALLS
# | | | | | | |
4) | rcu_read_lock_sched_held() {
4) 0.552 us | rcu_lockdep_current_cpu_online();
4) 6.124 us | }
Reported-by: Alexander Schmidt <alexschm@de.ibm.com>
Signed-off-by: Thomas Richter <tmricht@linux.ibm.com>
Acked-by: Namhyung Kim <namhyung@kernel.org>
Cc: Heiko Carstens <hca@linux.ibm.com>
Cc: Sumanth Korikkar <sumanthk@linux.ibm.com>
Cc: Sven Schnelle <svens@linux.ibm.com>
Cc: Vasily Gorbik <gor@linux.ibm.com>
Link: http://lore.kernel.org/lkml/20210421120400.2126433-1-tmricht@linux.ibm.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
To use in automated tests inside containers from a tarball generated
by 'make perf-tar-src-pkg*', where testing building from a tarball
is obviously not needed, so add a 'build-test-tarball' for that case.
And don't build with gtk2 as this complicates things for cross builds
where we don't always have all the libraries a full perf build requires
available for the target arch, ditto for static builds.
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
This functionality is based on recently introduced sysfs attributes for
Intel® Xeon® Scalable processor family (code name Skylake-SP):
Commit bb42b3d397 ("perf/x86/intel/uncore: Expose an Uncore unit to IIO PMON mapping")
Mode is intended to provide four I/O performance metrics in MB per each
PCIe root port:
- Inbound Read: I/O devices below root port read from the host memory
- Inbound Write: I/O devices below root port write to the host memory
- Outbound Read: CPU reads from I/O devices below root port
- Outbound Write: CPU writes to I/O devices below root port
Each metric requiries only one uncore event which increments at every 4B
transfer in corresponding direction. The formulas to compute metrics
are generic:
#EventCount * 4B / (1024 * 1024)
Acked-by: Namhyung Kim <namhyung@kernel.org>
Signed-off-by: Alexander Antonov <alexander.antonov@linux.intel.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Alexey V Bayduraev <alexey.v.bayduraev@linux.intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Arnaldo Carvalho de Melo <acme@kernel.org>
Cc: Ian Rogers <irogers@google.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: https://lore.kernel.org/r/20210419094147.15909-4-alexander.antonov@linux.intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
- keep the ZC code, drop the code related to reinit
net/bridge/netfilter/ebtables.c
- fix build after move to net_generic
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Turns out, the default setting of attr.aux_watermark to half of the total
buffer size is not very useful, especially with smaller buffers. The
problem is that, after half of the buffer is filled up, the kernel updates
->aux_head and sets up the next "transaction", while observing that
->aux_tail is still zero (as userspace haven't had the chance to update
it), meaning that the trace will have to stop at the end of this second
"transaction". This means, for example, that the second PERF_RECORD_AUX in
every trace comes with TRUNCATED flag set.
Setting attr.aux_watermark to quarter of the buffer gives enough space for
the ->aux_tail update to be observed and prevents the data loss.
The obligatory before/after showcase:
> # perf_before record -e intel_pt//u -m,8 uname
> Linux
> [ perf record: Woken up 6 times to write data ]
> Warning:
> AUX data lost 4 times out of 10!
>
> [ perf record: Captured and wrote 0.099 MB perf.data ]
> # perf record -e intel_pt//u -m,8 uname
> Linux
> [ perf record: Woken up 4 times to write data ]
> [ perf record: Captured and wrote 0.039 MB perf.data ]
The effect is still visible with large workloads and large buffers,
although less pronounced.
Signed-off-by: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/20210414154955.49603-3-alexander.shishkin@linux.intel.com
Before:
# perf record -a cycles,instructions,cache-misses
Workload failed: No such file or directory
#
After:
# perf record -a cycles,instructions,cache-misses
Failed to collect 'cycles' for the 'cycles,instructions,cache-misses' workload: No such file or directory
#
Helps disambiguating other error scenarios:
# perf record -a -e cycles,instructions,cache-misses bla
Failed to collect 'cycles,instructions,cache-misses' for the 'bla' workload: No such file or directory
# perf record -a cycles,instructions,cache-misses sleep 1
Failed to collect 'cycles' for the 'cycles,instructions,cache-misses' workload: No such file or directory
#
When all goes well we're back to the usual:
# perf record -a -e cycles,instructions,cache-misses sleep 1
[ perf record: Woken up 3 times to write data ]
[ perf record: Captured and wrote 3.151 MB perf.data (21242 samples) ]
#
Acked-by: Ian Rogers <irogers@google.com>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Link: http://lore.kernel.org/lkml/20210414131628.2064862-3-acme@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
In hist__find_annotations(), since different 'struct hist_entry' entries
may point to same symbol, we free notes->src to signal already processed
this symbol in stdio mode; when annotate, entry will skipped if
notes->src is NULL to avoid repeated output.
However, there is a problem, for example, run the following command:
# perf record -e branch-misses -e branch-instructions -a sleep 1
perf.data file contains different types of sample event.
If the same IP sample event exists in branch-misses and branch-instructions,
this event uses the same symbol. When annotate branch-misses events, notes->src
corresponding to this event is set to null, as a result, when annotate
branch-instructions events, this event is skipped and no annotate is output.
Solution of this patch is to remove zfree in hists__find_annotations and
change sort order to "dso,symbol" to avoid duplicate output when different
processes correspond to the same symbol.
Signed-off-by: Yang Jihong <yangjihong1@huawei.com>
Acked-by: Namhyung Kim <namhyung@kernel.org>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Gustavo A. R. Silva <gustavoars@kernel.org>
Cc: Jin Yao <yao.jin@linux.intel.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Martin Liška <mliska@suse.cz>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: zhangjinhao2@huawei.com
Link: http://lore.kernel.org/lkml/20210319123527.173883-1-yangjihong1@huawei.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Conflicts:
MAINTAINERS
- keep Chandrasekar
drivers/net/ethernet/mellanox/mlx5/core/en_main.c
- simple fix + trust the code re-added to param.c in -next is fine
include/linux/bpf.h
- trivial
include/linux/ethtool.h
- trivial, fix kdoc while at it
include/linux/skmsg.h
- move to relevant place in tcp.c, comment re-wrapped
net/core/skmsg.c
- add the sk = sk // sk = NULL around calls
net/tipc/crypto.c
- trivial
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Add PMU events for AMD Zen3 processors as documented in the AMD Processor
Programming Reference for Family 19h and Model 01h [1].
Below are the events which are new on Zen3:
PMCx041 ls_mab_alloc.{all_allocations|hardware_prefetcher_allocations|load_store_allocations}
PMCx043 ls_dmnd_fills_from_sys.ext_cache_local
PMCx044 ls_any_fills_from_sys.{mem_io_remote|ext_cache_remote|mem_io_local|ext_cache_local|int_cache|lcl_l2}
PMCx047 ls_misal_loads.{ma4k|ma64}
PMCx059 ls_sw_pf_dc_fills.ext_cache_local
PMCx05a ls_hw_pf_dc_fills.ext_cache_local
PMCx05f ls_alloc_mab_count
PMCx085 bp_l1_tlb_miss_l2_tlb_miss.coalesced_4k
PMCx0ab de_dis_cops_from_decoder.disp_op_type.{any_integer_dispatch|any_fp_dispatch}
PMCx0cc ex_ret_ind_brch_instr
PMCx18e ic_tag_hit_miss.{all_instruction_cache_accesses|instruction_cache_miss|instruction_cache_hit}
PMCx1c7 ex_ret_msprd_brnch_instr_dir_msmtch
PMCx28f op_cache_hit_miss.{all_op_cache_accesses|op_cache_miss|op_cache_hit}
Section 2.1.17.2 "Performance Measurement" of "PPR for AMD Family 19h,
Model 01h, Revision B1 Processors - 55898 Rev 0.35 - Feb 5, 2021." lists
new metrics. Add them.
Preserve the events for Zen3 if they are measurable and non-zero as taken
from Zen2 directory even if the PPR of Zen3 [1] omits them. Those events
are the following:
PMCx000 fpu_pipe_assignment.{total|total0|total1|total2|total3}
PMCx004 fp_num_mov_elim_scal_op.{optimized|opt_potential|sse_mov_ops_elim|sse_mov_ops}
PMCx02D ls_rdtsc
PMCx040 ls_dc_accesses
PMCx046 ls_tablewalker.{iside|ic_type1|ic_type0|dside|dc_type1|dc_type0}
PMCx061 l2_request_g2.{group1|ls_rd_sized|ls_rd_sized_nc|ic_rd_sized|ic_rd_sized_nc|smc_inval|bus_lock_originator|bus_locks_responses}
PMCx062 l2_latency.l2_cycles_waiting_on_fills
PMCx063 l2_wcb_req.{wcb_write|wcb_close|zero_byte_store|cl_zero}
PMCx06d l2_fill_pending.l2_fill_busy
PMCx080 ic_fw32
PMCx081 ic_fw32_miss
PMCx086 bp_snp_re_sync
PMCx087 ic_fetch_stall.{ic_stall_any|ic_stall_dq_empty|ic_stall_back_pressure}
PMCx08a bp_l1_btb_correct
PMCx08c ic_cache_inval.{l2_invalidating_probe|fill_invalidated}
PMCx099 bp_tlb_rel
PMCx0a9 de_dis_uop_queue_empty_di0
PMCx0c7 ex_ret_brn_resync
PMCx28a ic_oc_mode_switch.{oc_ic_mode_switch|ic_oc_mode_switch}
L3PMCx01 l3_request_g1.caching_l3_cache_accesses
L3PMCx06 l3_comb_clstr_state.{other_l3_miss_typs|request_miss}
[1] Processor Programming Reference (PPR) for AMD Family 19h, Model 01h,
Revision B1 Processors - 55898 Rev 0.35 - Feb 5, 2021.
[2] Processor Programming Reference (PPR) for AMD Family 17h Model 71h,
Revision B0 Processors, 56176 Rev 3.06 - Jul 17, 2019.
[3] Processor Programming Reference (PPR) for AMD Family 17h Models
01h,08h, Revision B2 Processors, 54945 Rev 3.03 - Jun 14, 2019.
All of the PPRs can be found at:
https://bugzilla.kernel.org/show_bug.cgi?id=206537
Reviewed-by: Robert Richter <rrichter@amd.com>
Signed-off-by: Smita Koralahalli <Smita.KoralahalliChannabasappa@amd.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Ian Rogers <irogers@google.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Kim Phillips <kim.phillips@amd.com>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Martin Liška <mliska@suse.cz>
Cc: Michael Petlan <mpetlan@redhat.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Vijay Thakkar <vijaythakkar@me.com>
Cc: linux-perf-users@vger.kernel.org
Link: https://lore.kernel.org/r/20210406215944.113332-5-Smita.KoralahalliChannabasappa@amd.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Commit 08ed77e414 ("perf vendor events amd: Add recommended events")
added the hits event "L2 Cache Hits from L2 HWPF" with the same metric
expression as the accesses event "L2 Cache Accesses from L2 HWPF":
$ perf list --details
...
l2_cache_accesses_from_l2_hwpf
[L2 Cache Accesses from L2 HWPF]
[l2_pf_hit_l2 + l2_pf_miss_l2_hit_l3 + l2_pf_miss_l2_l3]
l2_cache_hits_from_l2_hwpf
[L2 Cache Hits from L2 HWPF]
[l2_pf_hit_l2 + l2_pf_miss_l2_hit_l3 + l2_pf_miss_l2_l3]
...
This was wrong and led to counting hits the same as accesses. Section
2.1.15.2 "Performance Measurement" of "PPR for AMD Family 17h Model 31h
B0 - 55803 Rev 0.54 - Sep 12, 2019", documents the hits event with
EventCode 0x70 which is the same as l2_pf_hit_l2.
Fix this, and massage the description for l2_pf_hit_l2 as the hits event
is now the duplicate of l2_pf_hit_l2. AMD recommends using the recommended
event over other events if the duplicate exists and maintain both for
consistency. Hence, l2_cache_hits_from_l2_hwpf should override
l2_pf_hit_l2.
Before:
# perf stat -M l2_cache_accesses_from_l2_hwpf,l2_cache_hits_from_l2_hwpf sleep 1
Performance counter stats for 'sleep 1':
1,436 l2_pf_miss_l2_l3 # 11114.00 l2_cache_accesses_from_l2_hwpf
# 11114.00 l2_cache_hits_from_l2_hwpf
4,482 l2_pf_hit_l2
5,196 l2_pf_miss_l2_hit_l3
1.001765339 seconds time elapsed
After:
# perf stat -M l2_cache_accesses_from_l2_hwpf sleep 1
Performance counter stats for 'sleep 1':
1,477 l2_pf_miss_l2_l3 # 10442.00 l2_cache_accesses_from_l2_hwpf
3,978 l2_pf_hit_l2
4,987 l2_pf_miss_l2_hit_l3
1.001491186 seconds time elapsed
# perf stat -e l2_cache_hits_from_l2_hwpf sleep 1
Performance counter stats for 'sleep 1':
3,983 l2_cache_hits_from_l2_hwpf
1.001329970 seconds time elapsed
Note the difference in performance counter values for the accesses
versus the hits after the fix, and the hits event now counting the same
as l2_pf_hit_l2.
Fixes: 08ed77e414 ("perf vendor events amd: Add recommended events")
Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=206537
Reviewed-by: Robert Richter <rrichter@amd.com>
Signed-off-by: Smita Koralahalli <Smita.KoralahalliChannabasappa@amd.com>
Tested-by: Arnaldo Carvalho de Melo <acme@kernel.org> # On a 3900X
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Ian Rogers <irogers@google.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Kim Phillips <kim.phillips@amd.com>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Martin Liška <mliska@suse.cz>
Cc: Michael Petlan <mpetlan@redhat.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Vijay Thakkar <vijaythakkar@me.com>
Cc: linux-perf-users@vger.kernel.org
Link: https://lore.kernel.org/r/20210406215944.113332-2-Smita.KoralahalliChannabasappa@amd.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Since commit 14d3d54052 ("perf session: Try to read pipe data from
file") 'perf inject' has started printing "PERFILE2h" when not processing
pipes.
The commit exposed perf to the possiblity that the input is not a pipe
but the 'repipe' parameter gets used. That causes the printing because
perf inject sets 'repipe' to true always.
The 'repipe' parameter of perf_session__new() is used by 2 functions:
- perf_file_header__read_pipe()
- trace_report()
In both cases, the functions copy data to STDOUT_FILENO when 'repipe' is
true.
Fix by setting 'repipe' to true only if the output is a pipe.
Fixes: e558a5bd8b ("perf inject: Work with files")
Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Acked-by: Jiri Olsa <jolsa@redhat.com>
Cc: Andrew Vagin <avagin@openvz.org>
Link: http://lore.kernel.org/lkml/20210401103605.9000-1-adrian.hunter@intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
In particular we want to have this upstream commit:
b908297047: ("bpf: Use NOP_ATOMIC5 instead of emit_nops(&prog, 5) for BPF_TRAMP_F_CALL_ORIG")
... before merging in x86/cpu changes and the removal of the NOP optimizations, and
applying PeterZ's !retpoline objtool series.
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Some OCaml developers reported that this bit of information is sometimes
useful for disambiguating functions for which the OCaml compiler assigns
the same name, e.g. nested or inlined functions.
Signed-off-by: Fabian Hemmer <copy@copy.sh>
Link: http://lore.kernel.org/lkml/20210226075223.p3s5oz4jbxwnqjtv@nyu
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
When executing the daemon test on Arm64 and x86 with Debian (Buster)
distro, both skip the test case with the log:
# ./perf test -v 76
76: daemon operations :
--- start ---
test child forked, pid 11687
test daemon list
trap: SIGINT: bad trap
./tests/shell/daemon.sh: 173: local: cpu-clock: bad variable name
test child finished with -2
---- end ----
daemon operations: Skip
So the error happens for the variable expansion when use local variable
in the shell script. Since Debian Buster uses dash but not bash as
non-interactive shell, when execute the daemon testing, it hits a known
issue for dash which was reported [1].
To resolve this issue, one option is to add double quotes for all local
variables assignment, so need to change the code from:
local line=`perf daemon --config ${config} -x: | head -2 | tail -1`
... to:
local line="`perf daemon --config ${config} -x: | head -2 | tail -1`"
But the testing script has bunch of local variables, this leads to big
changes for whole script.
On the other hand, the testing script asks to use the "local" feature
which is bash-specific, so this patch explicitly uses "#!/bin/bash" to
ensure running the script with bash.
After:
# ./perf test -v 76
76: daemon operations :
--- start ---
test child forked, pid 11329
test daemon list
test daemon reconfig
test daemon stop
test daemon signal
signal 12 sent to session 'test [11596]'
signal 12 sent to session 'test [11596]'
test daemon ping
test daemon lock
test child finished with 0
---- end ----
daemon operations: Ok
[1] https://bugs.launchpad.net/ubuntu/+source/dash/+bug/139097
Fixes: 2291bb915b ("perf tests: Add daemon 'list' command test")
Signed-off-by: Leo Yan <leo.yan@linaro.org>
Acked-by: Namhyung Kim <namhyung@kernel.org>
Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Ian Rogers <irogers@google.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lore.kernel.org/lkml/20210320104554.529213-1-leo.yan@linaro.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
The pipeline stage cycles details can be recorded on powerpc from the
contents of Performance Monitor Unit (PMU) registers. On ISA v3.1
platform, sampling registers exposes the cycles spent in different
pipeline stages. Patch adds perf tools support to present two of the
cycle counter information along with memory latency (weight).
Re-use the field 'ins_lat' for storing the first pipeline stage cycle.
This is stored in 'var2_w' field of 'perf_sample_weight'.
Add a new field 'p_stage_cyc' to store the second pipeline stage cycle
which is stored in 'var3_w' field of perf_sample_weight.
Add new sort function 'Pipeline Stage Cycle' and include this in
default_mem_sort_order[]. This new sort function may be used to denote
some other pipeline stage in another architecture. So add this to list
of sort entries that can have dynamic header string.
Signed-off-by: Athira Rajeev <atrajeev@linux.vnet.ibm.com>
Reviewed-by: Madhavan Srinivasan <maddy@linux.ibm.com>
Acked-by: Jiri Olsa <jolsa@redhat.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Kajol Jain <kjain@linux.ibm.com>
Cc: Kan Liang <kan.liang@linux.intel.com>
Cc: Michael Ellerman <mpe@ellerman.id.au>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Ravi Bangoria <ravi.bangoria@linux.ibm.com>
Link: https://lore.kernel.org/r/1616425047-1666-5-git-send-email-atrajeev@linux.vnet.ibm.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Currently the header string for different columns in perf report is
fixed. Some fields of perf sample could have different meaning for
different architectures than the meaning conveyed by the header string.
An example is the new field 'var2_w' of perf_sample_weight structure.
This is presently captured as 'Local INSTR Latency' in perf mem report.
But this could be used to denote a different latency cycle in another
architecture.
Introduce a weak function arch_perf_header_entry() to set the arch
specific header string for the fields which can contain dynamic header.
If the architecture do not have this function, fall back to the default
header string value.
Signed-off-by: Athira Rajeev <atrajeev@linux.vnet.ibm.com>
Reviewed-by: Madhavan Srinivasan <maddy@linux.ibm.com>
Acked-by: Jiri Olsa <jolsa@redhat.com>
Cc: Kajol Jain <kjain@linux.ibm.com>
Cc: Kan Liang <kan.liang@linux.intel.com>
Cc: Michael Ellerman <mpe@ellerman.id.au>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Ravi Bangoria <ravi.bangoria@linux.ibm.com>
Link: https://lore.kernel.org/r/1616425047-1666-3-git-send-email-atrajeev@linux.vnet.ibm.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
