From: Alan Lawrence <alan.lawrence@arm.com>
To: gcc-patches@gcc.gnu.org,
Bill Schmidt <wschmidt@linux.vnet.ibm.com>,
Ramana Radhakrishnan <Ramana.Radhakrishnan@arm.com>,
Richard Sandiford <Richard.Sandiford@arm.com>,
Alan Hayward <Alan.Hayward@arm.com>
Subject: Re: [PATCH] vectorizing conditional expressions (PR tree-optimization/65947)
Date: Mon, 14 Sep 2015 09:50:00 -0000 [thread overview]
Message-ID: <55F69799.8010901@arm.com> (raw)
In-Reply-To: <1441977591.2795.11.camel@gnopaine>
On 11/09/15 14:19, Bill Schmidt wrote:
>
> A secondary concern for powerpc is that REDUC_MAX_EXPR produces a scalar
> that has to be broadcast back to a vector, and the best way to implement
> it for us already has the max value in all positions of a vector. But
> that is something we should be able to fix with simplify-rtx in the back
> end.
Reading this thread again, this bit stands out as unaddressed. Yes PowerPC can
"fix" this with simplify-rtx, but the vector cost model will not take this into
account - it will think that the broadcast-back-to-a-vector requires an extra
operation after the reduction, whereas in fact it will not.
Does that suggest we should have a new entry in vect_cost_for_stmt for
vec_to_scalar-and-back-to-vector (that defaults to vec_to_scalar+scalar_to_vec,
but on some architectures e.g. PowerPC would be the same as vec_to_scalar)?
(I agree that if that's the limit of how "different" conditional reductions may
be between architectures, then we should not have a vec_cost_for_stmt for a
whole conditional reduction.)
Cheers, Alan
next prev parent reply other threads:[~2015-09-14 9:47 UTC|newest]
Thread overview: 26+ messages / expand[flat|nested] mbox.gz Atom feed top
2015-09-10 15:11 Alan Hayward
2015-09-10 22:34 ` Bill Schmidt
2015-09-11 9:19 ` Alan Hayward
2015-09-11 13:23 ` Bill Schmidt
2015-09-11 13:55 ` Ramana Radhakrishnan
2015-09-11 14:41 ` Richard Sandiford
2015-09-11 15:14 ` Bill Schmidt
2015-09-11 15:30 ` Richard Sandiford
2015-09-11 15:50 ` Bill Schmidt
2015-09-11 16:54 ` Ramana Radhakrishnan
2015-09-15 11:47 ` Richard Biener
2015-09-14 9:50 ` Alan Lawrence [this message]
2015-09-14 14:20 ` Bill Schmidt
2015-09-15 12:10 ` Richard Biener
2015-09-15 15:41 ` Alan Hayward
2015-09-18 12:22 ` Richard Biener
2015-09-18 13:36 ` Alan Lawrence
2015-09-18 14:14 ` Alan Hayward
[not found] ` <D221D55E.8386%alan.hayward@arm.com>
2015-09-23 16:07 ` Alan Hayward
2015-09-30 12:49 ` Richard Biener
2015-10-01 15:22 ` Alan Hayward
2015-09-15 12:12 ` Richard Biener
2015-10-19 8:34 Alan Hayward
2015-10-21 10:46 ` Richard Biener
2015-10-22 14:18 ` Alan Lawrence
2015-10-22 14:23 ` Alan Hayward
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=55F69799.8010901@arm.com \
--to=alan.lawrence@arm.com \
--cc=Alan.Hayward@arm.com \
--cc=Ramana.Radhakrishnan@arm.com \
--cc=Richard.Sandiford@arm.com \
--cc=gcc-patches@gcc.gnu.org \
--cc=wschmidt@linux.vnet.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).