lkml.org 
[lkml]   [2018]   [Feb]   [8]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
From
Date
SubjectRe: [linus:master] BUILD REGRESSION a2e5790d841658485d642196dbb0927303d6c22f
On Thu, Feb 8, 2018 at 1:13 AM, Peter Zijlstra <peterz@infradead.org> wrote:
>
> _The_ problem is that new binutils cannot sanely decode any function
> that has a WARN in (this very much includes perf annotate):

Ugh.

Is there any reason why we don't just use UD2 and avoid this whole issue?

Do we even *do* anything different for UD0 vs UD2? I see that
'is_valid_bugaddr()' seems to accept either, I'm not seeing anything
that cares.

Yes, yes, we currently use UD0 for warnings, and UD2 for BUG(), but
the fact is, we don't actually seem to care, because the *real*
differentiator is BUGFLAG_WARNING.

No?

So all the UD0 games seem a bit pointless. Let's not use it at all
because it's so badly documented.

Linus

\
 
 \ /
  Last update: 2018-02-08 18:29    [W:0.089 / U:6.020 seconds]
©2003-2018 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site