From: jiawei@iscas.ac.cn
To: "Richard Biener" <rguenther@suse.de>
Cc: gcc-patches@gcc.gnu.org, jakub@redhat.com, palmer@rivosinc.com,
kito.cheng@gmail.com, jim.wilson.gcc@gmail.com,
wuwei2016@iscas.ac.cn
Subject: Re: Re: [PATCH] testsuite: Add extra RISC-V options so that -fprefetch-loop-arrays works
Date: Thu, 28 Jul 2022 17:28:00 +0800 (GMT+08:00) [thread overview]
Message-ID: <5fd99d6f.4496.182442264ac.Coremail.jiawei@iscas.ac.cn> (raw)
In-Reply-To: <nycvar.YFH.7.77.849.2207280744540.6583@jbgna.fhfr.qr>
> -----原始邮件-----
> 发件人: "Richard Biener" <rguenther@suse.de>
> 发送时间: 2022-07-28 15:45:27 (星期四)
> 收件人: jiawei <jiawei@iscas.ac.cn>
> 抄送: gcc-patches@gcc.gnu.org, jakub@redhat.com, palmer@rivosinc.com, kito.cheng@gmail.com, jim.wilson.gcc@gmail.com, wuwei2016@iscas.ac.cn
> 主题: Re: [PATCH] testsuite: Add extra RISC-V options so that -fprefetch-loop-arrays works
>
> On Thu, 28 Jul 2022, jiawei wrote:
>
> > This patch adds the additional options on RISC-V target.
> > "-fprefetch-loop-arrays" option needs enable prefetch instruction,
> > for RISC-V that contained in "zicbop" extension.
> > Use "-march" with "zicbop" will enable this feature.
>
> OK.
>
> Note -fprefetch-loop-arrays is just required to trigger an ICE,
> do you see a diagnostic when prefetching is not supported? Maybe
> simply adding -w is better then.
Yes, without -march support it report warning info:
============================================================================================
cc1: warning: '-fprefetch-loop-arrays' not supported for this target (try '-march' switches)
============================================================================================
after add -w the warning ignored.
>
> > gcc/testsuite/ChangeLog:
> >
> > * gcc.dg/pr106397.c: New dg-additional-options for RISC-V.
> >
> > ---
> > gcc/testsuite/gcc.dg/pr106397.c | 2 ++
> > 1 file changed, 2 insertions(+)
> >
> > diff --git a/gcc/testsuite/gcc.dg/pr106397.c b/gcc/testsuite/gcc.dg/pr106397.c
> > index 2bc17f8cf80..19274fa8771 100644
> > --- a/gcc/testsuite/gcc.dg/pr106397.c
> > +++ b/gcc/testsuite/gcc.dg/pr106397.c
> > @@ -1,6 +1,8 @@
> > /* { dg-do compile } */
> > /* { dg-options "-O3 -fprefetch-loop-arrays --param l2-cache-size=0 --param prefetch-latency=3 -fprefetch-loop-arrays" } */
> > /* { dg-additional-options "-march=i686 -msse" { target { { i?86-*-* x86_64-*-* } && ia32 } } } */
> > +/* { dg-additional-options "-march=rv64gc_zicbop" { target { riscv64-*-* } } */
> > +/* { dg-additional-options "-march=rv32gc_zicbop" { target { riscv32-*-* } } */
> >
> > int
> > bar (void)
> >
>
> --
> Richard Biener <rguenther@suse.de>
> SUSE Software Solutions Germany GmbH, Frankenstrasse 146, 90461 Nuernberg,
> Germany; GF: Ivo Totev, Andrew Myers, Andrew McDonald, Boudien Moerman;
> HRB 36809 (AG Nuernberg)
</rguenther@suse.de></jiawei@iscas.ac.cn></rguenther@suse.de>
prev parent reply other threads:[~2022-07-28 9:28 UTC|newest]
Thread overview: 3+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-07-28 3:25 jiawei
2022-07-28 7:45 ` Richard Biener
2022-07-28 9:28 ` jiawei [this message]
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=5fd99d6f.4496.182442264ac.Coremail.jiawei@iscas.ac.cn \
--to=jiawei@iscas.ac.cn \
--cc=gcc-patches@gcc.gnu.org \
--cc=jakub@redhat.com \
--cc=jim.wilson.gcc@gmail.com \
--cc=kito.cheng@gmail.com \
--cc=palmer@rivosinc.com \
--cc=rguenther@suse.de \
--cc=wuwei2016@iscas.ac.cn \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).