lkml.org 
[lkml]   [2014]   [Mar]   [1]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: [PATCH 2/2] perf tools: Fix strict alias issue for find_first_bit

* David Ahern <dsahern@gmail.com> wrote:

> On 2/28/14, 2:29 PM, Peter Zijlstra wrote:
> >Hurm; didn't I suggest using -fno-strict-aliasing just like the kernel
> >does? Because the C aliasing rules are bonghits heavy?
>
> you, and Ingo in 2009 -- 65014ab3

Yeah, so that's certainly true for the kernel, but for user-space the
aggressive optimizations that come with the aliasing rules were pretty
good, last I checked.

So it would be nice to check the code generation and performance
impact of -fno-strict-aliasing on perf (if any). If the impact is
restricted to an odd few annotations for weird, low-level methods
like find_bit(), then we might be able to live with it.

The aliasing warnings can also find real bugs and uncleanlinesses.

So I'm really of two minds regarding this.

Thanks,

Ingo


\
 
 \ /
  Last update: 2014-03-01 10:41    [W:0.038 / U:0.600 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site