public inbox for gcc-patches@gcc.gnu.org
 help / color / mirror / Atom feed
From: Richard Sandiford <richard.sandiford@arm.com>
To: Richard Biener <rguenther@suse.de>
Cc: gcc-patches@gcc.gnu.org,  linkw@linux.ibm.com
Subject: Re: [PATCH] Add emulated gather capability to the vectorizer
Date: Mon, 02 Aug 2021 09:05:20 +0100	[thread overview]
Message-ID: <mptfsvsff0f.fsf@arm.com> (raw)
In-Reply-To: <nycvar.YFH.7.76.2108020937420.11781@zhemvz.fhfr.qr> (Richard Biener's message of "Mon, 2 Aug 2021 09:44:09 +0200 (CEST)")

Richard Biener <rguenther@suse.de> writes:
> On Fri, 30 Jul 2021, Richard Sandiford wrote:
>> > @@ -9456,6 +9499,51 @@ vectorizable_load (vec_info *vinfo,
>> >  			data_ref = NULL_TREE;
>> >  			break;
>> >  		      }
>> > +		    else if (memory_access_type == VMAT_GATHER_SCATTER)
>> > +		      {
>> > +			/* Emulated gather-scatter.  */
>> > +			gcc_assert (!final_mask);
>> 
>> For this to be safe, we need to make sure that
>> check_load_store_for_partial_vectors clears
>> LOOP_VINFO_CAN_USE_PARTIAL_VECTORS_P.  Is that already guaranteed?
>> (Maybe yes, I didn't look closely. :-))
>
> I've made sure to fail earlier:
>
> @@ -8692,6 +8725,15 @@ vectorizable_load (vec_info *vinfo,
>                              "unsupported access type for masked 
> load.\n");
>           return false;
>         }
> +      else if (memory_access_type == VMAT_GATHER_SCATTER
> +              && gs_info.ifn == IFN_LAST
> +              && !gs_info.decl)
> +       {
> +         if (dump_enabled_p ())
> +           dump_printf_loc (MSG_MISSED_OPTIMIZATION, vect_location,
> +                            "unsupported masked emulated gather.\n");
> +         return false;
> +       }
>      }
>
> at least I don't see a way to emulate masked gathers ;)

Yeah, that handles masked loads created by ifcvt.  But we also need to
protect against an unconditional load being converted into a predicated
load for partial loop vectorisation.  I think we still need a check in
check_load_store_for_partial_vectors to handle that.  The check wouldn't
prevent vectorisation, it would just prevent using fully-predicated loops.

Thanks,
Richard

  reply	other threads:[~2021-08-02  8:05 UTC|newest]

Thread overview: 13+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-07-30 11:34 Richard Biener
2021-07-30 13:42 ` Richard Sandiford
2021-08-02  7:44   ` Richard Biener
2021-08-02  8:05     ` Richard Sandiford [this message]
2021-08-02  8:15       ` Richard Biener
2021-08-02  8:20         ` Richard Sandiford
2021-07-30 14:04 ` Tamar Christina
2021-07-30 14:04 ` Kewen.Lin
2021-08-02  6:07   ` Kewen.Lin
2021-08-02  7:09     ` Richard Biener
2021-08-02  8:35       ` Kewen.Lin
2021-08-02  9:11         ` Richard Biener
2021-08-02  9:20           ` Kewen.Lin

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=mptfsvsff0f.fsf@arm.com \
    --to=richard.sandiford@arm.com \
    --cc=gcc-patches@gcc.gnu.org \
    --cc=linkw@linux.ibm.com \
    --cc=rguenther@suse.de \
    /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).