From: Richard Biener <rguenther@suse.de>
To: "Kewen.Lin" <linkw@linux.ibm.com>
Cc: "juzhe.zhong@rivai.ai" <juzhe.zhong@rivai.ai>,
gcc-patches <gcc-patches@gcc.gnu.org>,
"richard.sandiford" <richard.sandiford@arm.com>
Subject: Re: decremnt IV patch create fails on PowerPC
Date: Tue, 30 May 2023 10:00:54 +0000 (UTC) [thread overview]
Message-ID: <nycvar.YFH.7.77.849.2305300956480.4723@jbgna.fhfr.qr> (raw)
In-Reply-To: <e481dea4-c278-fa84-6d68-b715864df669@linux.ibm.com>
On Tue, 30 May 2023, Kewen.Lin wrote:
> on 2023/5/30 17:26, juzhe.zhong@rivai.ai wrote:
> > Ok.
> >
> > It seems that for this conditions:
> >
> > + /* If we're vectorizing a loop that uses length "controls" and
> > + can iterate more than once, we apply decrementing IV approach
> > + in loop control. */
> > + if (LOOP_VINFO_CAN_USE_PARTIAL_VECTORS_P (loop_vinfo)
> > + && !LOOP_VINFO_LENS (loop_vinfo).is_empty ()
> > + && LOOP_VINFO_PARTIAL_LOAD_STORE_BIAS (loop_vinfo) == 0
> > + && !(LOOP_VINFO_NITERS_KNOWN_P (loop_vinfo)
> > + && known_le (LOOP_VINFO_INT_NITERS (loop_vinfo),
> > + LOOP_VINFO_VECT_FACTOR (loop_vinfo))))
> > + LOOP_VINFO_USING_DECREMENTING_IV_P (loop_vinfo) = true;
> >
> >
> > I should add direct_supportted_p (SELECT_VL...) to this is that right?
>
> I guess no, with this condition any targets without SELECT_VL are unable
> to leverage the new decrement scheme for lengths, as your reply in PR109971
> you didn't meant to disable it. IIUC, what Richi suggested is to introduce
> one new IV just like the previous one which has non-variable step, then it's
> SCEV-ed and some analysis based on it can do a good job.
No, I said the current scheme does sth along
do {
remain -= MIN (vf, remain);
} while (remain != 0);
and I suggest to instead do
do {
old_remain = remain;
len = MIN (vf, remain);
remain -= vf;
} while (old_remain >= vf);
basically since only the last iteration will have len < vf we can
ignore that remain -= vf will underflow there if we appropriately
rewrite the exit test to use the pre-decrement value.
> Since this is mainly for targets without SELECT_VL capability, I can follow
> up this if you don't mind.
>
> BR,
> Kewen
>
--
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)
next prev parent reply other threads:[~2023-05-30 10:00 UTC|newest]
Thread overview: 15+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-05-25 22:45 钟居哲
2023-05-26 6:46 ` Richard Biener
2023-05-26 7:46 ` juzhe.zhong
2023-05-30 9:22 ` Richard Biener
2023-05-30 9:26 ` juzhe.zhong
2023-05-30 9:50 ` Richard Biener
2023-05-30 9:55 ` juzhe.zhong
2023-05-30 11:30 ` juzhe.zhong
2023-05-30 9:51 ` Kewen.Lin
2023-05-30 10:00 ` Richard Biener [this message]
2023-05-30 10:05 ` juzhe.zhong
2023-05-30 10:12 ` Richard Sandiford
2023-05-30 10:43 ` Richard Biener
2023-05-30 11:29 ` Richard Sandiford
2023-05-30 11:37 ` Richard Biener
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=nycvar.YFH.7.77.849.2305300956480.4723@jbgna.fhfr.qr \
--to=rguenther@suse.de \
--cc=gcc-patches@gcc.gnu.org \
--cc=juzhe.zhong@rivai.ai \
--cc=linkw@linux.ibm.com \
--cc=richard.sandiford@arm.com \
/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).