Not able to see the profiling results properly for a profile generated from gperftools on mac OS High sierra

オープン
#324 コメント 16 件 リアクション 0 件 担当者 0 名 GitHub で見る

メンテナーはふだん 1 日以内に返信

まだ誰も着手していません。

評価

難易度
4/5
見積もり時間
3〜5日
初心者へのやさしさ
25/100
issue の種類
バグ
明瞭さ
おおむね明確
活発さ
停滞
技術スタック
go, macos

調査の方向性

まず pprof --text の呼び出し、添付された Makefile.txt、sbmpart_prof.txt を確認し、gperftools のプロファイルとバイナリを、この Go pprof の実行が受け付けるものと比較してください。macOS High Sierra でコマンドシーケンスを再現し、報告されたパースエラーとシンボル化エラーを使用して、必要なサポートまたは診断を定義してください。添付されたプロファイルが報告されたエラーなしで解析されるか、サポートされる制限と必要な呼び出し方法が文書化されれば完了です。

索引モデルが issue の本文から書いたものです。

説明

Priority: p3 type: question

Please answer these questions before submitting your issue. Thanks!

What version of pprof are you using?

Not able to find the version. How do I find one? I used go get github.com/google/prof to download the pprof.

If you are using pprof via go tool pprof, what's your go env output?
If you run pprof from GitHub, what's the Git revision?

git log -1 produces the below output

commit 20ceacaff62f6bf3e7c08128ea03d6faa84c880c (HEAD -> master, origin/master, origin/HEAD)
Author: Alexey Alexandrov <[email protected]>
Date:   Wed Feb 14 09:16:22 2018 -0800

    Enable MacOS binutils test. (#322)

What operating system and processor architecture are you using?

macOS High Sierra version 10.13.1 (2.9 GHz Intel Core i5)

What did you do?

pprof --text sbmpart sbmpart_prof.out
where sbmpart is the binary executable and sbmpart_prof.out is the profile output when the sbmpart binary is generated with -lprofiler flag for google gperftools.

If possible, provide a recipe for reproducing the error.
Attaching a profile you are trying to analyze is good.

I did the following steps:

make debug
CPUPROFILE=sbmpart_prof.out ./sbmpart argv1 argv2 argv3
pprof --text sbmpart sbmpart_prof.out

Note that the pprof referred to in step 3 is not from gperftools (I first install gperftools and then remove its pprof and then use go pprof)

I am also attaching the makefile and the profile generated from gperftools.

Makefile.txt
sbmpart_prof.txt

What did you expect to see?

I expect to see the CPU profiling results for all the functions without any error.

What did you see instead?

I see some errors. see below:
I have the below issues:

  1. how to fix the unrecognized profile format
  2. how to fix the unrecognized binary errors.
sbmpart: parsing profile: unrecognized profile format
Fetched 1 source profiles out of 2
Local symbolization failed for libdyld.dylib: unrecognized binary: /usr/lib/system/libdyld.dylib
Local symbolization failed for libsystem_platform.dylib: unrecognized binary: /usr/lib/system/libsystem_platform.dylib
Local symbolization failed for libsystem_c.dylib: unrecognized binary: /usr/lib/system/libsystem_c.dylib
Local symbolization failed for libsystem_m.dylib: unrecognized binary: /usr/lib/system/libsystem_m.dylib
Local symbolization failed for libsystem_kernel.dylib: unrecognized binary: /usr/lib/system/libsystem_kernel.dylib
Local symbolization failed for libsystem_malloc.dylib: unrecognized binary: /usr/lib/system/libsystem_malloc.dylib
Local symbolization failed for libsystem_kernel.dylib: unrecognized binary: /usr/lib/system/libsystem_kernel.dylib
Local symbolization failed for sbmpart: unrecognized binary: /Users/AkhilaRasamrtaMurthih/work/GraphChallenge/stochastic-block-partition/./sbmpart
Local symbolization failed for libsystem_c.dylib: unrecognized binary: /usr/lib/system/libsystem_c.dylib
Some binary filenames not available. Symbolization may be incomplete.
Try setting PPROF_BINARY_PATH to the search path for local binaries.
File: libobjc.A.dylib
Type: cpu
Showing nodes accounting for 19.24s, 99.12% of 19.41s total
Dropped 6 nodes (cum <= 0.10s)
      flat  flat%   sum%        cum   cum%
     7.81s 40.24% 40.24%      7.82s 40.29%  [libsystem_kernel.dylib]
     4.42s 22.77% 63.01%      4.61s 23.75%  sbm_delta_entropy
     1.64s  8.45% 71.46%      2.12s 10.92%  [libsystem_platform.dylib]
     1.61s  8.29% 79.75%      1.66s  8.55%  sbm_met_has
     1.36s  7.01% 86.76%      1.70s  8.76%  sbm_dyn_csr_add
     0.70s  3.61% 90.37%      5.86s 30.19%  sbm_dyn_csr_flush_adj_lists
     0.59s  3.04% 93.41%      0.68s  3.50%  [libsystem_malloc.dylib]
     0.39s  2.01% 95.41%      0.39s  2.01%  [libsystem_m.dylib]
     0.23s  1.18% 96.60%      0.36s  1.85%  sbm_adj_list_node_init
     0.21s  1.08% 97.68%      0.22s  1.13%  sbm_propose_new_blk
     0.14s  0.72% 98.40%      1.61s  8.29%  [libsystem_c.dylib]
     0.06s  0.31% 98.71%      0.12s  0.62%  monotonic_seconds
     0.02s   0.1% 98.82%      0.34s  1.75%  sbm_adj_list_add_node
     0.02s   0.1% 98.92%      1.45s  7.47%  sbm_compute_iblk_ec
     0.02s   0.1% 99.02%      0.14s  0.72%  sbm_read_tsv_append
     0.01s 0.052% 99.07%     10.04s 51.73%  sbm_partition
     0.01s 0.052% 99.12%      5.98s 30.81%  sbm_update_iblk_ec
         0     0% 99.12%     10.20s 52.55%  <unknown>
         0     0% 99.12%     10.18s 52.45%  [libdyld.dylib]
         0     0% 99.12%      1.20s  6.18%  [libprofiler.0.dylib]
         0     0% 99.12%     10.18s 52.45%  main
         0     0% 99.12%     10.18s 52.45%  sbm_stream_partition
主要言語
Go
スター
9.3k
フォーク
686
平均マージ
4日 17時間
マージ済み PR(30日)
14

環境構築

はじめの一歩

  1. issue を最後まで読み、次にプロジェクトのコントリビューションガイドを読みます。
  2. 着手することを issue にコメントします — 二人が同じ作業をするのを防げます。
  3. リポジトリをフォークし、ブランチを切って変更します。
  4. issue 番号を参照したプルリクエストを送ります。

google/pprof のほかの issue

google/pprof の issue をすべて見る

似ている issue

Go の issue をもっと見る

新しい issue をメールで受け取る

初心者向けの GitHub issue を短くまとめたダイジェスト。