[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: Elisp native profiler
From: |
Eli Zaretskii |
Subject: |
Re: Elisp native profiler |
Date: |
Sat, 29 Sep 2012 17:24:32 +0200 |
> From: Stefan Monnier <address@hidden>
> Cc: address@hidden
> Date: Wed, 26 Sep 2012 13:49:57 -0400
>
> I typically do:
>
> emacs -Q
> M-x profiler-start RET
> C-x C-f src/regexp.c RET
> page-down until I get to the end of the buffer, then page-up to the
> beginning
> M-x profiler-report RET
>
> Verifying that the resulting profile is accurate is rather difficult,
> of course, so as long as it looks credible, I assume it's more or
> less correct.
Is the cpu profile below (obtained with the above recipe) reasonable?
Why do functions like scroll-up-command and jit-lock-function appear
twice or three times, once under call-interactively, the other time by
themselves or under scroll-up?
(Should we rename "Automatic GC" like we did with redisplay?)
- call-interactively 83 47%
- scroll-up-command 43 24%
- scroll-up 43 24%
- jit-lock-function 40 22%
- jit-lock-fontify-now 40 22%
- byte-code 40 22%
- run-hook-with-args 40 22%
- font-lock-fontify-region 40 22%
- c-font-lock-fontify-region 40 22%
- font-lock-default-fontify-region 38 21%
- font-lock-fontify-keywords-region 37 21%
- c-font-lock-declarations 28 16%
- c-find-decl-spots 28 16%
- #<compiled 0xd66b09> 25 14%
+ c-determine-limit 16 9%
looking-at 2 1%
+ c-parse-state 2 1%
+ c-forward-decl-or-cast-1 1 0%
+ c-font-lock-declarators 1 0%
+ c-cheap-inside-bracelist-p 1 0%
+ c-backward-sws 1 0%
looking-at 2 1%
+ c-forward-sws 1 0%
re-search-forward 3 1%
+ c-font-lock-enclosing-decls 2 1%
+ c-font-lock-enum-tail 1 0%
+ #<compiled 0xd6756d> 1 0%
+ c-font-lock-complex-decl-prepare 1 0%
+ #<compiled 0xd67523> 1 0%
+ font-lock-fontify-syntactically-region 1 0%
- mapc 2 1%
+ #<compiled 0xd36357> 2 1%
- list 40 22%
- read-extended-command 40 22%
- completing-read 40 22%
- completing-read-default 40 22%
read-from-minibuffer 40 22%
- scroll-up 20 11%
- jit-lock-function 20 11%
- jit-lock-fontify-now 20 11%
- byte-code 20 11%
+ run-hook-with-args 20 11%
Automatic GC 17 9%
+ redisplay_internal (C function) 16 9%
- scroll-up-command 14 8%
+ scroll-up 14 8%
+ jit-lock-function 14 8%
+ byte-code 6 3%
+ jit-lock-fontify-now 4 2%
+ font-lock-fontify-region 1 0%
- Re: Elisp native profiler, (continued)
- Re: Elisp native profiler, Eli Zaretskii, 2012/09/26
- Re: Elisp native profiler, Stefan Monnier, 2012/09/26
- Re: Elisp native profiler, Eli Zaretskii, 2012/09/26
- Re: Elisp native profiler, Eli Zaretskii, 2012/09/26
- Re: Elisp native profiler, Stefan Monnier, 2012/09/26
- Re: Elisp native profiler, Eli Zaretskii, 2012/09/26
- Re: Elisp native profiler, Stefan Monnier, 2012/09/26
- Re: Elisp native profiler, Tomohiro Matsuyama, 2012/09/26
- Re: Elisp native profiler, Stefan Monnier, 2012/09/27
- Re: Elisp native profiler, Paul Eggert, 2012/09/30
- Re: Elisp native profiler,
Eli Zaretskii <=
- Re: Elisp native profiler, Stefan Monnier, 2012/09/29
- Re: Elisp native profiler, Eli Zaretskii, 2012/09/29
- Re: Elisp native profiler, Stefan Monnier, 2012/09/29
- Re: Elisp native profiler, Eli Zaretskii, 2012/09/30
Re: Elisp native profiler, Juanma Barranquero, 2012/09/25