This is the mail archive of the
mailing list for the GCC project.
Re: Loop peeling
- From: Tejas Belagod <tejas dot belagod at arm dot com>
- To: Richard Biener <richard dot guenther at gmail dot com>, Evandro Menezes <e dot menezes at samsung dot com>, Jan Hubicka <hubicka at ucw dot cz>
- Cc: GCC Development <gcc at gcc dot gnu dot org>
- Date: Wed, 29 Oct 2014 11:53:05 +0000
- Subject: Re: Loop peeling
- Authentication-results: sourceware.org; auth=none
- References: <033101cff2c7$96bff550$c43fdff0$ at samsung dot com> <CAFiYyc23OhPmV3DJa9z62DB5jwR1JWKowqQDKRyCbrBiKai0xA at mail dot gmail dot com>
On 29/10/14 09:32, Richard Biener wrote:
On Tue, Oct 28, 2014 at 4:55 PM, Evandro Menezes <email@example.com> wrote:
While doing some benchmark flag mining on AArch64, I noticed that
-fpeel-loops was a mined option often. As a matter of fact, when using it
always, even without FDO, it seemed to raise most benchmarks and to leave
almost all of the rest flat, with a barely noticeable cost in code-size. It
seems to me that it might be safe enough to be implied perhaps at -O3. Is
there any reason why this never came into being?
Loop peeling is done by default on AArch64 unless, IIRC,
-fvect-cost-model=cheap is specified which switches it off. There was a
general thread on loop peeling around the same time last year
(https://gcc.gnu.org/ml/gcc/2013-11/msg00307.html) where Richard
suggested that peeling vs. non-peeling should be factored into the
vector cost model and is a more generic improvement.
Not sure, but peeling is/was very stupid (peeling 8 times unconditionally
or not at all). At least without FDO (and with -fprofile-use it is enabled).
Similar case for -funroll-loops.
For GCC 5 peeling now moved to GIMPLE, so maybe things changed
for that (but I'd doubt that). Honza?