Re: [MERGE] Use 1 second main loop timeout by default

On Thu, 2008-03-20 at 01:31 +0100, Henrik Nordstrom wrote:
> At the sprint the issue about our current 10ms main loop timeout came
> up, and it was suggested the problem most likely have been fixed in
> HEAD. And even if it hasn't been fixed it's something which should be
> fixed rather than plastered over by spinning around on a short timeout
> when there is no activity..

bb:approve

  I need to go over that code once again, especially if 1 second change
breaks things again. I am for your change to HEAD because it might
expose the unresolved bug and prompt me to polish that code
further :-).

The change should _not_ be backported to v3.0 though.

Once this is committed, we could close related "Squid runs too hot" bug
reports, with a risk of having to reopen them.

BTW, is there any reason for a 1 second timeout other than to give Squid
an opportunity to notice signals (on some platforms some signals do not
interrupt poll/select)? If that is the only reason, should we increase
the timeout further? If that is not the only reason, the knowledge of
other implicit dependencies may help me to polish the code further.

Thank you,

Alex.

> This merge request backs out the change to revert the change..
>
> Alex commit reversing the change said:
>
> revno: 8332
> committer: rousskov
> branch nick: HEAD
> timestamp: Tue 2007-07-24 01:55:21 +0000
> message:
> Reversed bug #2011 fix because it may slow down ICAP, BodyPipe, and other code
> using zero-delay events to implement "asynchronous" calls.
>
> The code should probably be rewritten (a) to avoid any waiting/blocking when
> there are ready events and (b) to allow waiting longer when there are no ready
> events.
>
> Regards
> Henrik
Received on Wed Mar 19 2008 - 22:39:47 MDT