The low level index of raw branch records for the most recent branch can
be recorded in a sample with PERF_SAMPLE_BRANCH_HW_INDEX
branch_sample_type. Extend struct branch_stack to support it.
However, if the PERF_SAMPLE_BRANCH_HW_INDEX is not applied, only nr and
entries[] will be output by kernel. The pointer of entries[] could be
wrong, since the output format is different with new struct
branch_stack. Add a variable no_hw_idx in struct perf_sample to
indicate whether the hw_idx is output. Add get_branch_entry() to return
corresponding pointer of entries[0].
To make dummy branch sample consistent as new branch sample, add hw_idx
in struct dummy_branch_stack for cs-etm and intel-pt.
Apply the new struct branch_stack for synthetic events as well.
Extend test case sample-parsing to support new struct branch_stack.
Committer notes:
Renamed get_branch_entries() to perf_sample__branch_entries() to have
proper namespacing and pave the way for this to be moved to libperf,
eventually.
Add 'static' to that inline as it is in a header.
Add 'hw_idx' to 'struct dummy_branch_stack' in cs-etm.c to fix the build
on arm64.
Signed-off-by: Kan Liang <kan.liang@linux.intel.com>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Alexey Budankov <alexey.budankov@linux.intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Mathieu Poirier <mathieu.poirier@linaro.org>
Cc: Michael Ellerman <mpe@ellerman.id.au>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Pavel Gerasimov <pavel.gerasimov@intel.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Ravi Bangoria <ravi.bangoria@linux.ibm.com>
Cc: Stephane Eranian <eranian@google.com>
Cc: Vitaly Slobodskoy <vitaly.slobodskoy@intel.com>
Link: http://lore.kernel.org/lkml/20200228163011.19358-2-kan.liang@linux.intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
brstackinsn must be allowed to be set by the user when AUX area data has
been captured because, in that case, the branch stack might be
synthesized on the fly. This fixes the following error:
Before:
$ perf record -e '{intel_pt//,cpu/mem_inst_retired.all_loads,aux-sample-size=8192/pp}:u' grep -rqs jhgjhg /boot
[ perf record: Woken up 19 times to write data ]
[ perf record: Captured and wrote 2.274 MB perf.data ]
$ perf script -F +brstackinsn --xed --itrace=i1usl100 | head
Display of branch stack assembler requested, but non all-branch filter set
Hint: run 'perf record -b ...'
After:
$ perf record -e '{intel_pt//,cpu/mem_inst_retired.all_loads,aux-sample-size=8192/pp}:u' grep -rqs jhgjhg /boot
[ perf record: Woken up 19 times to write data ]
[ perf record: Captured and wrote 2.274 MB perf.data ]
$ perf script -F +brstackinsn --xed --itrace=i1usl100 | head
grep 13759 [002] 8091.310257: 1862 instructions:uH: 5641d58069eb bmexec+0x86b (/bin/grep)
bmexec+2485:
00005641d5806b35 jnz 0x5641d5806bd0 # MISPRED
00005641d5806bd0 movzxb (%r13,%rdx,1), %eax
00005641d5806bd6 add %rdi, %rax
00005641d5806bd9 movzxb -0x1(%rax), %edx
00005641d5806bdd cmp %rax, %r14
00005641d5806be0 jnb 0x5641d58069c0 # MISPRED
mismatch of LBR data and executable
00005641d58069c0 movzxb (%r13,%rdx,1), %edi
Fixes: 48d02a1d5c ("perf script: Add 'brstackinsn' for branch stacks")
Reported-by: Andi Kleen <ak@linux.intel.com>
Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Link: http://lore.kernel.org/lkml/20191127095322.15417-1-adrian.hunter@intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
My earlier patch to just enable --reltime with --time was a little too
optimistic. The --time parsing would accept absolute time, which is
very confusing to the user.
Support relative time in --time parsing too. This only works with recent
perf record that records the first sample time. Otherwise we error out.
Fixes: 3714437d3f ("perf script: Allow --time with --reltime")
Signed-off-by: Andi Kleen <ak@linux.intel.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Link: http://lore.kernel.org/lkml/20191011182140.8353-1-andi@firstfloor.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
The original --reltime patch forbid --time with --reltime.
But it turns out --time doesn't really care about --reltime, because the
relative time is only used at final output, while the time filtering
always works earlier on absolute time.
So just remove the check and allow combining the two options.
Fixes: 90b10f47c0 ("perf script: Support relative time")
Signed-off-by: Andi Kleen <ak@linux.intel.com>
Acked-by: Jiri Olsa <jolsa@kernel.org>
Link: http://lore.kernel.org/lkml/20191002164642.1719-1-andi@firstfloor.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
When the LBR data and the instructions in a binary do not match the loop
printing instructions could get confused and print a long stream of
bogus <bad> instructions.
The problem was that if the instruction decoder cannot decode an
instruction it ilen wasn't initialized, so the loop going through the
basic block would continue with the previous value.
Harden the code to avoid such problems:
- Make sure ilen is always freshly initialized and is 0 for bad
instructions.
- Do not overrun the code buffer while printing instructions
- Print a warning message if the final jump is not on an instruction
boundary.
Signed-off-by: Andi Kleen <ak@linux.intel.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Link: http://lore.kernel.org/lkml/20190927233546.11533-1-andi@firstfloor.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Move the perf_event_attr struct fron 'struct evsel' to 'struct perf_evsel'.
Committer notes:
Fixed up these:
tools/perf/arch/arm/util/auxtrace.c
tools/perf/arch/arm/util/cs-etm.c
tools/perf/arch/arm64/util/arm-spe.c
tools/perf/arch/s390/util/auxtrace.c
tools/perf/util/cs-etm.c
Also
cc1: warnings being treated as errors
tests/sample-parsing.c: In function 'do_test':
tests/sample-parsing.c:162: error: missing initializer
tests/sample-parsing.c:162: error: (near initialization for 'evsel.core.cpus')
struct evsel evsel = {
.needs_swap = false,
- .core.attr = {
- .sample_type = sample_type,
- .read_format = read_format,
+ .core = {
+ . attr = {
+ .sample_type = sample_type,
+ .read_format = read_format,
+ },
[perfbuilder@a70e4eeb5549 /]$ gcc --version |& head -1
gcc (GCC) 4.4.7
Also we don't need to include perf_event.h in
tools/perf/lib/include/perf/evsel.h, forward declaring 'struct
perf_event_attr' is enough. And this even fixes the build in some
systems where things are used somewhere down the include path from
perf_event.h without defining __always_inline.
Signed-off-by: Jiri Olsa <jolsa@kernel.org>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Alexey Budankov <alexey.budankov@linux.intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Michael Petlan <mpetlan@redhat.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/20190721112506.12306-43-jolsa@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
In pipe mode, session->header.env.arch is not populated until the events
are processed. Therefore, the following command crashes:
perf record -o - | perf script
(gdb) bt
It fails when we try to compare env.arch against uts.machine:
if (!strcmp(uts.machine, session->header.env.arch) ||
(!strcmp(uts.machine, "x86_64") &&
!strcmp(session->header.env.arch, "i386")))
native_arch = true;
In pipe mode, it is tricky to find env.arch at this stage. To keep it
simple, let's just assume native_arch is always true for pipe mode.
Reported-by: David Carrillo Cisneros <davidca@fb.com>
Signed-off-by: Song Liu <songliubraving@fb.com>
Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: kernel-team@fb.com
Cc: stable@vger.kernel.org #v5.1+
Fixes: 3ab481a1cf ("perf script: Support insn output for normal samples")
Link: http://lkml.kernel.org/r/20190621014438.810342-1-songliubraving@fb.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
The 'perf kvm' command set up things so that we can record, report, top,
etc, but not 'script', so make 'perf script' be able to process samples
by allowing to pass guest kallsyms, vmlinux, modules, etc, and if at
least one of those is provided, set perf_guest to true so that guest
samples get properly resolved.
Testing it:
# perf kvm --guest --guestkallsyms /wb/rhel6.kallsyms --guestmodules /wb/rhel6.modules record -e cycles:Gk
^C[ perf record: Woken up 7 times to write data ]
[ perf record: Captured and wrote 3.602 MB perf.data.guest (10492 samples) ]
#
# perf evlist -i perf.data.guest
cycles:Gk
# perf evlist -v -i perf.data.guest
cycles:Gk: size: 112, { sample_period, sample_freq }: 4000, sample_type: IP|TID|TIME|CPU|PERIOD, read_format: ID, disabled: 1, inherit: 1, exclude_user: 1, exclude_hv: 1, mmap: 1, comm: 1, freq: 1, task: 1, sample_id_all: 1, exclude_host: 1, mmap2: 1, comm_exec: 1, ksymbol: 1, bpf_event: 1
#
# perf kvm --guestkallsyms /wb/rhel6.kallsyms --guestmodules /wb/rhel6.modules report --stdio -s sym | head -30
# To display the perf.data header info, please use --header/--header-only options.
#
#
# Total Lost Samples: 0
#
# Samples: 10K of event 'cycles:Gk'
# Event count (approx.): 2434201408
#
# Overhead Symbol
# ........ ..............................................
#
11.93% [g] avtab_search_node
3.95% [g] sidtab_context_to_sid
2.41% [g] n_tty_write
2.20% [g] _spin_unlock_irqrestore
1.37% [g] _aesni_dec4
1.33% [g] kmem_cache_alloc
1.07% [g] native_write_cr0
0.99% [g] kfree
0.95% [g] _spin_lock
0.91% [g] __memset
0.87% [g] schedule
0.83% [g] _spin_lock_irqsave
0.76% [g] __kmalloc
0.67% [g] avc_has_perm_noaudit
0.66% [g] kmem_cache_free
0.65% [g] glue_xts_crypt_128bit
0.59% [g] __d_lookup
0.59% [g] __audit_syscall_exit
0.56% [g] __memcpy
#
Then, when trying to use perf script to generate a python script and
then process the events after adding a python hook for non-tracepoint
events:
# perf script -i perf.data.guest -g python
generated Python script: perf-script.py
# vim perf-script.py
# tail -2 perf-script.py
def process_event(param_dict):
print(param_dict["symbol"])
#
# perf script -i perf.data.guest -s perf-script.py | head
in trace_begin
vmx_vmexit
vmx_vmexit
vmx_vmexit
vmx_vmexit
vmx_vmexit
vmx_vmexit
vmx_vmexit
vmx_vmexit
vmx_vmexit
231
#
We'd see just the vmx_vmexit, i.e. the samples from the guest don't show
up.
After this patch:
# perf script --guestkallsyms /wb/rhel6.kallsyms --guestmodules /wb/rhel6.modules -i perf.data.guest -s perf-script.py 2> /dev/null | head -30
in trace_begin
apic_timer_interrupt
apic_timer_interrupt
apic_timer_interrupt
apic_timer_interrupt
apic_timer_interrupt
save_args
do_timer
drain_array
inode_permission
avc_has_perm_noaudit
run_timer_softirq
apic_timer_interrupt
apic_timer_interrupt
apic_timer_interrupt
apic_timer_interrupt
apic_timer_interrupt
kvm_guest_apic_eoi_write
run_posix_cpu_timers
_spin_lock
handle_pte_fault
rcu_irq_enter
delay_tsc
delay_tsc
native_read_tsc
apic_timer_interrupt
sys_open
internal_add_timer
list_del
rcu_exit_nohz
#
Jiri Olsa noticed we need to set 'perf_guest' to true if we want to
process guest samples and I made it be set if one of the guest files
settings get set via the command line options added in this patch, that
match those present in the 'perf kvm' command.
We probably want to have 'perf record', 'perf report' etc to notice that
there are guest samples and do the right thing, which is to look for
files with some suffix that make it be associated with the guest used to
collect the samples, i.e. if a vmlinux file is passed, we can get the
build-id from it, if not some other identifier or simply looking for
"kallsyms.guest", for instance, in the current directory.
Reported-by: Mariano Pache <npache@redhat.com>
Tested-by: Mariano Pache <npache@redhat.com>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Alexander Yarygin <yarygin@linux.vnet.ibm.com>
Cc: Ali Raza <alirazabhutta.10@gmail.com>
Cc: Christian Borntraeger <borntraeger@de.ibm.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Joe Mario <jmario@redhat.com>
Cc: Larry Woodman <lwoodman@redhat.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Orran Krieger <okrieger@redhat.com>
Cc: Ramkumar Ramachandra <artagnon@gmail.com>
Cc: Yunlong Song <yunlong.song@huawei.com>
Link: https://lkml.kernel.org/n/tip-d54gj64rerlxcqsrod05biwn@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
We got the sane_ctype.h headers from git and kept using it so far, but
since that code originally came from the kernel sources to the git
sources, perhaps its better to just use the one in the kernel, so that
we can leverage tools/perf/check_headers.sh to be notified when our copy
gets out of sync, i.e. when fixes or goodies are added to the code we've
copied.
This will help with things like tools/lib/string.c where we want to have
more things in common with the kernel, such as strim(), skip_spaces(),
etc so as to go on removing the things that we have in tools/perf/util/
and instead using the code in the kernel, indirectly and removing things
like EXPORT_SYMBOL(), etc, getting notified when fixes and improvements
are made to the original code.
Hopefully this also should help with reducing the difference of code
hosted in tools/ to the one in the kernel proper.
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Link: https://lkml.kernel.org/n/tip-7k9868l713wqtgo01xxygn12@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Add field 'ipc' to display instructions-per-cycle.
Example:
perf record -e intel_pt/cyc/u ls
perf script --insn-trace --xed -F+ipc,-dso,-cpu,-tid
ls 2670177.697113434: 7f0dfdbcd090 _start+0x0 mov %rsp, %rdi IPC: 0.00 (1/877)
ls 2670177.697113434: 7f0dfdbcd093 _start+0x3 callq 0x7f0dfdbce030
ls 2670177.697113434: 7f0dfdbce030 _dl_start+0x0 pushq %rbp
ls 2670177.697113434: 7f0dfdbce031 _dl_start+0x1 mov %rsp, %rbp
ls 2670177.697113434: 7f0dfdbce034 _dl_start+0x4 pushq %r15
ls 2670177.697113434: 7f0dfdbce036 _dl_start+0x6 pushq %r14
ls 2670177.697113434: 7f0dfdbce038 _dl_start+0x8 pushq %r13
ls 2670177.697113434: 7f0dfdbce03a _dl_start+0xa pushq %r12
ls 2670177.697113434: 7f0dfdbce03c _dl_start+0xc mov %rdi, %r12
ls 2670177.697113434: 7f0dfdbce03f _dl_start+0xf pushq %rbx
ls 2670177.697113434: 7f0dfdbce040 _dl_start+0x10 sub $0x38, %rsp
ls 2670177.697113434: 7f0dfdbce044 _dl_start+0x14 rdtsc
ls 2670177.697113434: 7f0dfdbce046 _dl_start+0x16 mov %eax, %eax
ls 2670177.697113434: 7f0dfdbce048 _dl_start+0x18 shl $0x20, %rdx
ls 2670177.697113434: 7f0dfdbce04c _dl_start+0x1c or %rax, %rdx
ls 2670177.697114471: 7f0dfdbce04f _dl_start+0x1f movq 0x27e22(%rip), %rax IPC: 0.00 (15/1685)
ls 2670177.697116177: 7f0dfdbce056 _dl_start+0x26 movq %rdx, 0x27683(%rip) IPC: 0.00 (1/881)
Note, the IPC values are low due to page faults at the beginning of
execution. The additional cycles are due to the time to enter the
kernel, not the actual kernel page fault handler.
Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Link: http://lkml.kernel.org/r/20190520113728.14389-9-adrian.hunter@intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
