From: Leo Yan <leo.yan@linaro.org>
To: Jiri Olsa <jolsa@redhat.com>
Cc: Arnaldo Carvalho de Melo <arnaldo.melo@gmail.com>,
Jiri Olsa <jolsa@kernel.org>,
Jesper Dangaard Brouer <brouer@redhat.com>,
Alexei Starovoitov <ast@kernel.org>,
Daniel Borkmann <daniel@iogearbox.net>,
lkml <linux-kernel@vger.kernel.org>,
Ingo Molnar <mingo@kernel.org>,
Namhyung Kim <namhyung@kernel.org>,
Alexander Shishkin <alexander.shishkin@linux.intel.com>,
Peter Zijlstra <a.p.zijlstra@chello.nl>,
Stanislav Fomichev <sdf@fomichev.me>,
Song Liu <songliubraving@fb.com>,
Adrian Hunter <adrian.hunter@intel.com>,
Andi Kleen <ak@linux.intel.com>
Subject: Re: [PATCHv3 00/12] perf tools: Display eBPF code in intel_pt trace
Date: Sun, 2 Jun 2019 14:42:11 +0800 [thread overview]
Message-ID: <20190602064211.GB9580@leoy-ThinkPad-X240s> (raw)
In-Reply-To: <20190531091944.GA24672@krava>
On Fri, May 31, 2019 at 11:19:44AM +0200, Jiri Olsa wrote:
> On Thu, May 30, 2019 at 10:05:10PM +0800, Leo Yan wrote:
> > Hi Arnaldo,
> >
> > On Thu, May 30, 2019 at 10:36:45AM -0300, Arnaldo Carvalho de Melo wrote:
> >
> > [...]
> >
> > > One other way of testing this:
> > >
> > > I used perf trace's use of BPF, using:
> > >
> > > [root@quaco ~]# cat ~/.perfconfig
> > > [llvm]
> > > dump-obj = true
> > > clang-opt = -g
> > > [trace]
> > > add_events = /home/acme/git/perf/tools/perf/examples/bpf/augmented_raw_syscalls.c
> > > show_zeros = yes
> > > show_duration = no
> > > no_inherit = yes
> > > show_timestamp = no
> > > show_arg_names = no
> > > args_alignment = 40
> > > show_prefix = yes
> > >
> > > For arm64 this needs fixing, tools/perf/examples/bpf/augmented_raw_syscalls.c
> > > (its in the kernel sources) is still hard coded for x86_64 syscall numbers :-\
> >
> > Thanks a lot for sharing this, I will test with this method and let you
> > and Jiri know the result in tomorrow.
>
> it's always battle of having too much data but caturing everything,
> versus having reasonable data size and being lucky to hit the trace ;-)
>
> with sampling on high freq and 1 second trace in another terminal
> I got the trace below:
Thanks a lot for confirmation, Jiri.
I tries multiple times with different frequency for sampleip command,
but still no luck to capture eBPF object info in perf data. Will dig
into it and will keep you posted for this.
Thanks,
Leo Yan
> ---
> terminal 1:
> # sampleip -F 1000 20
>
> terminal2:
> # perf-with-kcore record pt -e intel_pt//k -m100M,100M -C 1 -- sleep 1
> # perf-with-kcore script pt --call-trace
> ...
> swapper 0 [001] 85820.051207146: ([kernel.kallsyms] ) __perf_event_overflow
> swapper 0 [001] 85820.051207146: ([kernel.kallsyms] ) __perf_event_account_interrupt
> swapper 0 [001] 85820.051207146: ([kernel.kallsyms] ) __x86_indirect_thunk_rax
> swapper 0 [001] 85820.051207146: ([kernel.kallsyms] ) __x86_indirect_thunk_rax
> swapper 0 [001] 85820.051207146: ([kernel.kallsyms] ) __x86_indirect_thunk_rax
> swapper 0 [001] 85820.051207146: ([kernel.kallsyms] ) __x86_indirect_thunk_rax
> swapper 0 [001] 85820.051207467: (bpf_prog_19578a12836c4115 ) __htab_map_lookup_elem
> swapper 0 [001] 85820.051207788: ([kernel.kallsyms] ) memcmp
> ...
> # perf-with-kcore script pt --insn-trace --xed
> ...
> swapper 0 [001] 85820.051207467: ffffffff90c00c40 __x86_indirect_thunk_rax+0x10 ([kernel.kallsyms]) retq
> swapper 0 [001] 85820.051207467: ffffffffc0557710 bpf_prog_19578a12836c4115+0x0 (bpf_prog_19578a12836c4115) pushq %rbp
> swapper 0 [001] 85820.051207467: ffffffffc0557711 bpf_prog_19578a12836c4115+0x1 (bpf_prog_19578a12836c4115) mov %rsp, %rbp
> swapper 0 [001] 85820.051207467: ffffffffc0557714 bpf_prog_19578a12836c4115+0x4 (bpf_prog_19578a12836c4115) sub $0x38, %rsp
> swapper 0 [001] 85820.051207467: ffffffffc055771b bpf_prog_19578a12836c4115+0xb (bpf_prog_19578a12836c4115) sub $0x28, %rbp
> swapper 0 [001] 85820.051207467: ffffffffc055771f bpf_prog_19578a12836c4115+0xf (bpf_prog_19578a12836c4115) movq %rbx, (%rbp)
> swapper 0 [001] 85820.051207467: ffffffffc0557723 bpf_prog_19578a12836c4115+0x13 (bpf_prog_19578a12836c4115) movq %r13, 0x8(%rbp)
> swapper 0 [001] 85820.051207467: ffffffffc0557727 bpf_prog_19578a12836c4115+0x17 (bpf_prog_19578a12836c4115) movq %r14, 0x10(%rbp)
> swapper 0 [001] 85820.051207467: ffffffffc055772b bpf_prog_19578a12836c4115+0x1b (bpf_prog_19578a12836c4115) movq %r15, 0x18(%rbp)
> swapper 0 [001] 85820.051207467: ffffffffc055772f bpf_prog_19578a12836c4115+0x1f (bpf_prog_19578a12836c4115) xor %eax, %eax
> swapper 0 [001] 85820.051207467: ffffffffc0557731 bpf_prog_19578a12836c4115+0x21 (bpf_prog_19578a12836c4115) movq %rax, 0x20(%rbp)
> swapper 0 [001] 85820.051207467: ffffffffc0557735 bpf_prog_19578a12836c4115+0x25 (bpf_prog_19578a12836c4115) mov $0x1, %esi
> swapper 0 [001] 85820.051207467: ffffffffc055773a bpf_prog_19578a12836c4115+0x2a (bpf_prog_19578a12836c4115) movl %esi, -0xc(%rbp)
> swapper 0 [001] 85820.051207467: ffffffffc055773d bpf_prog_19578a12836c4115+0x2d (bpf_prog_19578a12836c4115) movq (%rdi), %rdi
> swapper 0 [001] 85820.051207467: ffffffffc0557741 bpf_prog_19578a12836c4115+0x31 (bpf_prog_19578a12836c4115) movq 0x80(%rdi), %rdi
> swapper 0 [001] 85820.051207467: ffffffffc0557748 bpf_prog_19578a12836c4115+0x38 (bpf_prog_19578a12836c4115) movq %rdi, -0x8(%rbp)
> swapper 0 [001] 85820.051207467: ffffffffc055774c bpf_prog_19578a12836c4115+0x3c (bpf_prog_19578a12836c4115) mov %rbp, %rsi
> swapper 0 [001] 85820.051207467: ffffffffc055774f bpf_prog_19578a12836c4115+0x3f (bpf_prog_19578a12836c4115) add $0xfffffffffffffff8, %rsi
> swapper 0 [001] 85820.051207467: ffffffffc0557753 bpf_prog_19578a12836c4115+0x43 (bpf_prog_19578a12836c4115) mov $0xffff9ac0b2926000, %rdi
> swapper 0 [001] 85820.051207467: ffffffffc055775d bpf_prog_19578a12836c4115+0x4d (bpf_prog_19578a12836c4115) callq 0xffffffff901fde70
> swapper 0 [001] 85820.051207467: ffffffff901fde70 __htab_map_lookup_elem+0x0 ([kernel.kallsyms]) nopl %eax, (%rax,%rax,1)
> swapper 0 [001] 85820.051207467: ffffffff901fde75 __htab_map_lookup_elem+0x5 ([kernel.kallsyms]) pushq %rbx
> swapper 0 [001] 85820.051207467: ffffffff901fde76 __htab_map_lookup_elem+0x6 ([kernel.kallsyms]) movl 0x1c(%rdi), %r9d
> swapper 0 [001] 85820.051207467: ffffffff901fde7a __htab_map_lookup_elem+0xa ([kernel.kallsyms]) mov %rsi, %rdx
> swapper 0 [001] 85820.051207467: ffffffff901fde7d __htab_map_lookup_elem+0xd ([kernel.kallsyms]) movl 0x214(%rdi), %ecx
> swapper 0 [001] 85820.051207467: ffffffff901fde83 __htab_map_lookup_elem+0x13 ([kernel.kallsyms]) mov %rdx, %r10
> swapper 0 [001] 85820.051207467: ffffffff901fde86 __htab_map_lookup_elem+0x16 ([kernel.kallsyms]) mov %r9d, %r8d
> swapper 0 [001] 85820.051207467: ffffffff901fde89 __htab_map_lookup_elem+0x19 ([kernel.kallsyms]) add %r9d, %ecx
> swapper 0 [001] 85820.051207467: ffffffff901fde8c __htab_map_lookup_elem+0x1c ([kernel.kallsyms]) leal -0x21524111(%rcx), %ebx
> swapper 0 [001] 85820.051207467: ffffffff901fde92 __htab_map_lookup_elem+0x22 ([kernel.kallsyms]) mov %ebx, %esi
> ...
next prev parent reply other threads:[~2019-06-02 7:08 UTC|newest]
Thread overview: 50+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-05-08 13:19 [PATCHv3 00/12] perf tools: Display eBPF code in intel_pt trace Jiri Olsa
2019-05-08 13:19 ` [PATCH 01/12] perf tools: Separate generic code in dso__data_file_size Jiri Olsa
2019-05-13 19:47 ` Arnaldo Carvalho de Melo
2019-05-13 20:00 ` Jiri Olsa
2019-05-23 3:10 ` Namhyung Kim
2019-05-23 7:49 ` Jiri Olsa
2019-05-30 8:08 ` [tip:perf/core] perf dso: Separate generic code in dso__data_file_size() tip-bot for Jiri Olsa
2019-05-08 13:20 ` [PATCH 02/12] perf tools: Separate generic code in dso_cache__read Jiri Olsa
2019-05-30 8:09 ` [tip:perf/core] perf dso: " tip-bot for Jiri Olsa
2019-05-08 13:20 ` [PATCH 03/12] perf tools: Simplify dso_cache__read function Jiri Olsa
2019-05-30 8:09 ` [tip:perf/core] perf dso: " tip-bot for Jiri Olsa
2019-05-08 13:20 ` [PATCH 04/12] perf tools: Add bpf dso read and size hooks Jiri Olsa
2019-05-30 8:10 ` [tip:perf/core] perf dso: Add BPF DSO " tip-bot for Jiri Olsa
2019-05-08 13:20 ` [PATCH 05/12] perf tools: Read also the end of the kernel Jiri Olsa
2019-05-24 18:15 ` Arnaldo Carvalho de Melo
2019-05-24 18:17 ` Arnaldo Carvalho de Melo
2019-05-24 18:46 ` Arnaldo Carvalho de Melo
2019-05-26 13:09 ` Jiri Olsa
2019-05-24 23:21 ` Jiri Olsa
2019-05-28 21:29 ` [tip:perf/urgent] perf machine: " tip-bot for Jiri Olsa
2019-05-08 13:20 ` [PATCH 06/12] perf tools: Keep zero in pgoff bpf map Jiri Olsa
2019-05-30 7:55 ` [tip:perf/core] perf machine: Keep zero in pgoff BPF map tip-bot for Jiri Olsa
2019-05-08 13:20 ` [PATCH 07/12] perf script: Pad dso name for --call-trace Jiri Olsa
2019-05-30 8:11 ` [tip:perf/core] perf script: Pad DSO " tip-bot for Jiri Olsa
2019-05-08 13:20 ` [PATCH 08/12] perf tools: Preserve eBPF maps when loading kcore Jiri Olsa
2019-05-22 16:06 ` Arnaldo Carvalho de Melo
2019-05-22 21:22 ` Jiri Olsa
2019-05-30 7:56 ` [tip:perf/core] " tip-bot for Jiri Olsa
2019-05-08 13:20 ` [PATCH 09/12] perf tests: Add map_groups__merge_in test Jiri Olsa
2019-05-30 8:11 ` [tip:perf/core] " tip-bot for Jiri Olsa
2019-05-08 13:20 ` [PATCH 10/12] perf script: Add --show-bpf-events to show eBPF related events Jiri Olsa
2019-05-30 8:12 ` [tip:perf/core] " tip-bot for Jiri Olsa
2019-05-08 13:20 ` [PATCH 11/12] perf script: Remove superfluous bpf event titles Jiri Olsa
2019-05-30 8:13 ` [tip:perf/core] perf script: Remove superfluous BPF " tip-bot for Jiri Olsa
2019-05-08 13:20 ` [PATCH 12/12] perf script: Add --show-all-events option Jiri Olsa
2019-05-22 16:18 ` Arnaldo Carvalho de Melo
2019-05-22 21:30 ` Jiri Olsa
2019-05-08 21:49 ` [PATCHv3 00/12] perf tools: Display eBPF code in intel_pt trace Song Liu
2019-05-08 23:41 ` Arnaldo Carvalho de Melo
2019-05-24 0:27 ` Arnaldo Carvalho de Melo
2019-05-24 10:06 ` Jiri Olsa
2019-05-30 10:54 ` Leo Yan
2019-05-30 12:07 ` Jiri Olsa
2019-05-30 12:57 ` Leo Yan
2019-05-30 13:27 ` Jiri Olsa
2019-05-30 13:36 ` Arnaldo Carvalho de Melo
2019-05-30 14:05 ` Leo Yan
2019-05-31 9:19 ` Jiri Olsa
2019-06-02 6:42 ` Leo Yan [this message]
2019-05-31 9:09 ` Leo Yan
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20190602064211.GB9580@leoy-ThinkPad-X240s \
--to=leo.yan@linaro.org \
--cc=a.p.zijlstra@chello.nl \
--cc=adrian.hunter@intel.com \
--cc=ak@linux.intel.com \
--cc=alexander.shishkin@linux.intel.com \
--cc=arnaldo.melo@gmail.com \
--cc=ast@kernel.org \
--cc=brouer@redhat.com \
--cc=daniel@iogearbox.net \
--cc=jolsa@kernel.org \
--cc=jolsa@redhat.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@kernel.org \
--cc=namhyung@kernel.org \
--cc=sdf@fomichev.me \
--cc=songliubraving@fb.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).