* inline-unit-growth=50 a bad idea
@ 2004-11-28 18:33 Richard Guenther
2004-11-28 18:58 ` Steven Bosscher
0 siblings, 1 reply; 2+ messages in thread
From: Richard Guenther @ 2004-11-28 18:33 UTC (permalink / raw)
To: gcc
Hi!
Compared to 3.4 we get a 340% performance regression because of the
different inline-unit-growth default (50 vs. 150). See PR18704.
Why was the limit set so low? This may cause compile time performance
comparisons to be apples-to-oranges comparisons.
At least fix the docs, which still say 150 is the default.
Thanks,
Richard.
^ permalink raw reply [flat|nested] 2+ messages in thread
* Re: inline-unit-growth=50 a bad idea
2004-11-28 18:33 inline-unit-growth=50 a bad idea Richard Guenther
@ 2004-11-28 18:58 ` Steven Bosscher
0 siblings, 0 replies; 2+ messages in thread
From: Steven Bosscher @ 2004-11-28 18:58 UTC (permalink / raw)
To: gcc; +Cc: Richard Guenther, gcc
On Sunday 28 November 2004 19:18, Richard Guenther wrote:
> Hi!
>
> Compared to 3.4 we get a 340% performance regression because of the
> different inline-unit-growth default (50 vs. 150). See PR18704.
> Why was the limit set so low? This may cause compile time performance
> comparisons to be apples-to-oranges comparisons.
Well news flash: it's always comparing apples and oranges.
But you could try setting it back to 150 and report what happens ;-)
> At least fix the docs, which still say 150 is the default.
Does cvs blame tell you when this happened?
Gr.
Steven
^ permalink raw reply [flat|nested] 2+ messages in thread
end of thread, other threads:[~2004-11-28 18:22 UTC | newest]
Thread overview: 2+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2004-11-28 18:33 inline-unit-growth=50 a bad idea Richard Guenther
2004-11-28 18:58 ` Steven Bosscher
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).