[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: Elisp native profiler
From: |
Paul Eggert |
Subject: |
Re: Elisp native profiler |
Date: |
Tue, 02 Oct 2012 22:11:46 -0700 |
User-agent: |
Mozilla/5.0 (X11; Linux i686; rv:15.0) Gecko/20120912 Thunderbird/15.0.1 |
On 10/02/2012 06:03 AM, Stefan Monnier wrote:
> What is an "overrun" exactly? Is it when the signal gets blocked
> because we're still in the signal handler?
Close. It's when a signal gets discarded because the
same signal is already pending; this can occur when
the signal is being blocked, and the signal can be
blocked because we're in the signal handler.
It's the performance figures, not the overruns, that
were the real eye-opener. Going with a 1 ms sampling
interval made Emacs 3x slower. Going with 10 ms made
Emacs run nearly at full speed. That's a strong argument
for going with 10 ms.
- Re: Elisp native profiler, Eli Zaretskii, 2012/10/01
- Re: Elisp native profiler, Tomohiro Matsuyama, 2012/10/01
- Re: Elisp native profiler, Paul Eggert, 2012/10/01
- Re: Elisp native profiler, Eli Zaretskii, 2012/10/01
- Re: Elisp native profiler, Stefan Monnier, 2012/10/01
- Re: Elisp native profiler, Paul Eggert, 2012/10/01
- Message not available
- Re: Elisp native profiler, Paul Eggert, 2012/10/02
- Re: Elisp native profiler, Stefan Monnier, 2012/10/02
- Re: Elisp native profiler,
Paul Eggert <=
- Re: Elisp native profiler, Stefan Monnier, 2012/10/03
- Re: Elisp native profiler, Andreas Schwab, 2012/10/03
- Re: Elisp native profiler, Stefan Monnier, 2012/10/03
- Re: Elisp native profiler, Eli Zaretskii, 2012/10/03
- Re: Elisp native profiler, Paul Eggert, 2012/10/01