From: Richard Biener <richard.guenther@gmail.com>
To: Richard Biener <richard.guenther@gmail.com>,
"Kewen.Lin" <linkw@linux.ibm.com>,
GCC Patches <gcc-patches@gcc.gnu.org>,
Segher Boessenkool <segher@kernel.crashing.org>,
Bill Schmidt <wschmidt@linux.ibm.com>,
Richard Sandiford <richard.sandiford@arm.com>
Subject: Re: [RFC/PATCH] vect: Recog mul_highpart pattern
Date: Tue, 13 Jul 2021 11:44:35 +0200 [thread overview]
Message-ID: <CAFiYyc0Fp9bcyx8nhpo0AwHhyK3bq89ta6yGNXZWqCvna3ugWA@mail.gmail.com> (raw)
In-Reply-To: <mptzguqd014.fsf@arm.com>
On Tue, Jul 13, 2021 at 11:40 AM Richard Sandiford
<richard.sandiford@arm.com> wrote:
>
> Richard Biener <richard.guenther@gmail.com> writes:
> > On Tue, Jul 13, 2021 at 10:53 AM Kewen.Lin <linkw@linux.ibm.com> wrote:
> >>
> >> Hi,
> >>
> >> When I added the support for Power10 newly introduced multiply
> >> highpart instrutions, I noticed that currently vectorizer
> >> doesn't try to vectorize multiply highpart pattern, I hope
> >> this isn't intentional?
> >>
> >> This patch is to extend the existing pattern mulhs handlings
> >> to cover multiply highpart. Another alternative seems to
> >> recog mul_highpart operation in a general place applied for
> >> scalar code when the target supports the optab for the scalar
> >> operation, it's based on the assumption that one target which
> >> supports vector version of multiply highpart should have the
> >> scalar version. I noticed that the function can_mult_highpart_p
> >> can check/handle mult_highpart well even without mul_highpart
> >> optab support, I think to recog this pattern in vectorizer
> >> is better. Is it on the right track?
> >
> > I think it's on the right track, using IFN_LAST is a bit awkward
> > in case yet another case pops up so maybe you can use
> > a code_helper instance instead which unifies tree_code,
> > builtin_code and internal_fn?
> >
> > I also notice that can_mult_highpart_p will return true if
> > only vec_widen_[us]mult_{even,odd,hi,lo} are available,
> > but then the result might be less optimal (or even not
> > handled later)?
> >
> > That is, what about adding optab internal functions
> > for [us]mul_highpart instead, much like the existing
> > ones for MULH{R,}S?
>
> Yeah, that's be my preference too FWIW. All uses of MULT_HIGHPART_EXPR
> already have to be guarded by can_mult_highpart_p, so replacing it with
> a directly-mapped ifn seems like a natural fit. (Then can_mult_highpart_p
> should be replaced with a direct_internal_fn_supported_p query.)
But note can_mult_highpart_t covers use via vec_widen_[us]mult_{even,odd,hi,lo}
but I think this specific pattern should key on [us]mul_highpart only?
Because vec_widen_* implies a higher VF (or else we might miss vectorizing?)?
Richard.
> Thanks,
> Richard
next prev parent reply other threads:[~2021-07-13 9:44 UTC|newest]
Thread overview: 31+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-07-13 8:52 Kewen.Lin
2021-07-13 8:58 ` [PATCH] rs6000: Support [u]mul<mode>3_highpart for vector Kewen.Lin
2021-07-13 22:07 ` Segher Boessenkool
2021-07-14 2:12 ` Kewen.Lin
2021-07-14 18:38 ` Segher Boessenkool
2021-07-13 9:35 ` [RFC/PATCH] vect: Recog mul_highpart pattern Richard Biener
2021-07-13 9:40 ` Richard Sandiford
2021-07-13 9:44 ` Richard Biener [this message]
2021-07-13 10:11 ` Richard Sandiford
2021-07-13 10:25 ` Kewen.Lin
2021-07-13 12:42 ` Richard Biener
2021-07-13 14:59 ` Kewen.Lin
2021-07-14 6:38 ` Richard Biener
2021-07-14 7:45 ` Kewen.Lin
2021-07-14 8:38 ` Richard Sandiford
2021-07-14 10:02 ` Kewen.Lin
2021-07-14 11:32 ` Richard Sandiford
2021-07-14 19:32 ` Segher Boessenkool
2021-07-15 1:40 ` Kewen.Lin
2021-07-15 23:08 ` Segher Boessenkool
2021-07-15 1:37 ` Kewen.Lin
2021-07-15 7:06 ` [PATCH v3] " Kewen.Lin
2021-07-15 7:17 ` Uros Bizjak
2021-07-15 8:04 ` Kewen.Lin
2021-07-15 8:23 ` Uros Bizjak
2021-07-15 8:49 ` Kewen.Lin
2021-07-15 9:41 ` Uros Bizjak
2021-07-15 8:40 ` Kewen.Lin
2021-07-15 11:58 ` Richard Biener
2021-07-16 5:33 ` [PATCH v4] " Kewen.Lin
2021-07-19 10:35 ` 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=CAFiYyc0Fp9bcyx8nhpo0AwHhyK3bq89ta6yGNXZWqCvna3ugWA@mail.gmail.com \
--to=richard.guenther@gmail.com \
--cc=gcc-patches@gcc.gnu.org \
--cc=linkw@linux.ibm.com \
--cc=richard.sandiford@arm.com \
--cc=segher@kernel.crashing.org \
--cc=wschmidt@linux.ibm.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).