lkml.org 
[lkml]   [1998]   [Aug]   [24]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
SubjectRe: pre egcs-1.1 testing and Linux 2.1.x
Date
From
Linus wrote:

>Umm.. Did you not follow the discussion where the _same_ bug could be
>triggered even without regparm?

Yes. But that just makes it even more important to understand what *does*
cause the real bug. Simply turning off optimisations all over the place
until the bug seems to go away isn't the right way to attack it.

>The regparm thing has been singled out because it makes it easier to show
>the bugs, not because of any technical reasons. And that's why it's so
>wrong to disable it, and hiding the real problem.

Precisely, and my objection to Andrea's patch was that it just disabled more
optimisations, again to hide the bug.

Andrea wrote:

>What I think is that the optimization code I commented out is buggy and
>the bug is triggered by the regparm attribute.

I think it's more subtle than that. The bug is not in the optimisations per
se, it's in the register allocator and just happens to be provoked more often
by higher optimisation (particularly when regparm is in use, because that
increases register pressure significantly near function calls).

p.



-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.altern.org/andrebalsa/doc/lkml-faq.html

\
 
 \ /
  Last update: 2005-03-22 13:44    [W:0.134 / U:0.156 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site