This is the mail archive of the
mailing list for the GCC project.
Re: [Google/4-8] Support for user-guided feedback-directed library optimization
- From: Andi Kleen <andi at firstfloor dot org>
- To: Teresa Johnson <tejohnson at google dot com>
- Cc: Andi Kleen <andi at firstfloor dot org>, "gcc-patches at gcc dot gnu dot org" <gcc-patches at gcc dot gnu dot org>, David Li <davidxl at google dot com>, Jan Hubicka <hubicka at ucw dot cz>
- Date: Sat, 10 May 2014 23:59:57 +0200
- Subject: Re: [Google/4-8] Support for user-guided feedback-directed library optimization
- Authentication-results: sourceware.org; auth=none
- References: <CAAe5K+U7wTXua5=96kqA47UWpPhuY-X3WW9NefpvXOUCffVQ7g at mail dot gmail dot com> <874n0yic04 dot fsf at tassilo dot jf dot intel dot com> <CAAe5K+WVvBp7ZEB9Y69FHFYbSsW11JEPmy471Vby+vpojQSWoQ at mail dot gmail dot com>
On Fri, May 09, 2014 at 08:11:40PM -0700, Teresa Johnson wrote:
> Thanks for catching that, I will fix it.
BTW I first misunderstood the goal of you patch.
(probably because there is no documentation ... something that should
also be fixed)
I originally thought it was a way to let user code use the value profiling
libgcov functions to profile arbitary values, and then later use the
most common value for some optimization.
But instead it seems to be only an (somewhat over complicated) IO
mechanism to let user programs write arbitrary values to the gcov file and
pass it back.
I think I would find the first more useful.