public inbox for gcc-patches@gcc.gnu.org
 help / color / mirror / Atom feed
From: Jakub Jelinek <jakub@redhat.com>
To: Tobias Burnus <tobias@codesourcery.com>
Cc: gcc-patches <gcc-patches@gcc.gnu.org>, fortran <fortran@gcc.gnu.org>
Subject: Re: [Patch] Fortran/openmp: Fix '!$omp end'
Date: Fri, 12 Nov 2021 13:02:34 +0100	[thread overview]
Message-ID: <20211112120234.GB2710@tucnak> (raw)
In-Reply-To: <f8ea4e1b-5129-d166-11a5-4e6ff5c8b3c3@codesourcery.com>

On Fri, Nov 12, 2021 at 12:01:27PM +0100, Tobias Burnus wrote:
> With the attached patch, the following combined/composite
> directives accept 'nowait' at 'end':

I've filed https://github.com/OpenMP/spec/issues/3184 because
I think OpenMP 5.2 got it wrong (and 5.1 got it wrong for
the end directives in many other cases too).

I believe the general rule should be:
1) non-combined/composite constructs allow nowait when
   they mention the clause (critical, for/do, scope, sections, single,
   workshare)
2) for simd, do simd allow it
3) anything combined with target allows it

So:
> 
> "end critical"
> "end do simd"
> "end do"
> "end scope"
> "end sections"
> "end single"
> "end target parallel"               (newly permits nowait)
> "end target simd"                   (newly permits nowait)
> "end target teams distribute simd"  (newly permits nowait)
> "end target teams distribute"       (newly permits nowait)
> "end target teams loop"             (newly permits nowait)
> "end target teams"                  (newly permits nowait)
> "end target"
> "end workshare"

is ok, but:

> and the following don't
> 
> "end target parallel do simd"
> "end target parallel do"
> "end target parallel loop"  (was completely missing before)
> "end target teams distribute parallel do simd"
> "end target teams distribute parallel do"

The above 5 should allow it too.
As per e.g. 5.2 [341:23],
"The effect of the nowait clause is as if it is applied to the
outermost leaf construct that permits it."
so even for the above 5 the clause splitting should put
nowait on target and not the others and be done with it.

	Jakub


  reply	other threads:[~2021-11-12 12:02 UTC|newest]

Thread overview: 6+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-11-11 17:11 Tobias Burnus
2021-11-11 18:01 ` Jakub Jelinek
2021-11-12 11:01   ` Tobias Burnus
2021-11-12 12:02     ` Jakub Jelinek [this message]
2021-11-12 15:56       ` Tobias Burnus
2021-11-12 16:07         ` Jakub Jelinek

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=20211112120234.GB2710@tucnak \
    --to=jakub@redhat.com \
    --cc=fortran@gcc.gnu.org \
    --cc=gcc-patches@gcc.gnu.org \
    --cc=tobias@codesourcery.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).