This is the mail archive of the
mailing list for the GCC project.
RE: [RFC] [PATCH, i386] Adjust unroll factor for bdver3 and bdver4
- From: "Gopalasubramanian, Ganesh" <Ganesh dot Gopalasubramanian at amd dot com>
- To: "H.J. Lu" <hjl dot tools at gmail dot com>
- Cc: "gcc-patches at gcc dot gnu dot org" <gcc-patches at gcc dot gnu dot org>, "Uros Bizjak (ubizjak at gmail dot com)" <ubizjak at gmail dot com>, "Richard Guenther <richard dot guenther at gmail dot com> (richard dot guenther at gmail dot com)" <richard dot guenther at gmail dot com>, "borntraeger at de dot ibm dot com" <borntraeger at de dot ibm dot com>, "Jakub Jelinek (jakub at redhat dot com)" <jakub at redhat dot com>
- Date: Thu, 21 Nov 2013 05:05:08 +0000
- Subject: RE: [RFC] [PATCH, i386] Adjust unroll factor for bdver3 and bdver4
- Authentication-results: sourceware.org; auth=none
- References: <EB4625145972F94C9680D8CADD6516155E73BF13 at SATLEXDAG02 dot amd dot com> <CAMe9rOp1vduYxXmDh4GuEJXTX3vv=qaPGjGk6ecGs8yzZ74dvA at mail dot gmail dot com>
> I suggest you add this to x86-tune.def and enable it for
> bdver3 and bdver4.
The macro TARGET_LOOP_UNROLL_ADJUST is not new. It is already available and is used by target s390.
Since it is not an "x86 only" feature I didn't add that in x86-tune.def.
From: H.J. Lu [mailto:firstname.lastname@example.org]
Sent: Thursday, November 21, 2013 12:02 AM
To: Gopalasubramanian, Ganesh
Cc: email@example.com; Uros Bizjak (firstname.lastname@example.org); Richard Guenther <email@example.com> (firstname.lastname@example.org); email@example.com; Jakub Jelinek (firstname.lastname@example.org)
Subject: Re: [RFC] [PATCH, i386] Adjust unroll factor for bdver3 and bdver4
On Wed, Nov 20, 2013 at 10:26 AM, Gopalasubramanian, Ganesh <Ganesh.Gopalasubramanian@amd.com> wrote:
> Steamroller processors contain a loop predictor and a loop buffer, which may make unrolling small loops less important.
> When unrolling small loops for steamroller, making the unrolled loop fit in the loop buffer should be a priority.
> This patch uses a heuristic approach (number of memory references) to decide the unrolling factor for small loops.
> This patch has some noise in SPEC 2006 results.
> Bootstrapping passes.
> I would like to know your comments before committing.
I suggest you add this to x86-tune.def and enable it for
bdver3 and bdver4.