GNU bug report logs - #66288
29.1; Performance regression using pipe for subprocess

Previous Next

Full log

Message #110 received at 66288 <at> debbugs.gnu.org (full text, mbox):

From: Dmitry Gutov <dmitry <at> gutov.dev>
To: Eli Zaretskii <eliz <at> gnu.org>
Cc: gerd.moellmann <at> gmail.com, 66288 <at> debbugs.gnu.org, gregory <at> heytings.org,
 eggert <at> cs.ucla.edu, cph <at> chris-hanson.org
Subject: Re: bug#66288: 29.1; Performance regression using pipe for subprocess
Date: Thu, 5 Oct 2023 13:48:55 +0300

On 05/10/2023 08:50, Eli Zaretskii wrote:
>> Date: Thu, 5 Oct 2023 01:54:43 +0300
>> Cc:gregory <at> heytings.org,gerd.moellmann <at> gmail.com,66288 <at> debbugs.gnu.org
>> From: Dmitry Gutov<dmitry <at> gutov.dev>
>>
>> On 05/10/2023 01:49, Paul Eggert wrote:
>>>> +      fcntl (inchannel, F_SETPIPE_SZ, readmax);
>>> This call can fail if you aren't root and you exceed the system limit in
>>> /proc/sys/fs/pipe-max-size. So I suggest that if this fnctl fails with
>>> EPERM, trying it again after clipping to that limit.
>> Perhaps we'd rather fail loudly, so the user is aware that their
>> customized value cannot be applied.
> Why not silently?  The clip_to_bounds call can already silently change
> the original request anyway.

clip_to_bounds clips to 4294967296 or higher, if I'm not mistaken. And 
my current system-wide limit is 1048576.

I usually like to know when my applied configuration is not used (or 
use-able). Anyway, it's not a hard requirement (not for emacs-29 anyway).

Previous Next

GNU bug tracking system
Copyright (C) 1999 Darren O. Benham, 1997,2003 nCipher Corporation Ltd, 1994-97 Ian Jackson.