public inbox for gcc-bugs@sourceware.org
help / color / mirror / Atom feed
From: "burnus at gcc dot gnu dot org" <gcc-bugzilla@gcc.gnu.org>
To: gcc-bugs@gcc.gnu.org
Subject: [Bug fortran/43665]  New: Optimization of libgfortran calls: function annotations for noclobber/noescape arguments
Date: Tue, 06 Apr 2010 13:30:00 -0000	[thread overview]
Message-ID: <bug-43665-13404@http.gcc.gnu.org/bugzilla/> (raw)

After the support goes into the middle end, cf.
http://gcc.gnu.org/ml/fortran/2010-04/msg00012.html and
http://gcc.gnu.org/ml/gcc-patches/2009-11/msg01257.html , one should use the
the new function call argument/return value attributes.

Note: One needs to be careful about this in terms of multi-image coarrays (SYNC
ALL, array single-side access) and true asynchronous I/O (WRITE ... WAIT) as
otherwise the middle end optimizes too much. (Telling this the middle end via a
specifically tailored attribute might be better, though.)

+ /* Call argument flags.  */
+ 
+ /* Nonzero if the argument is not dereferenced recursively, thus only
+    directly reachable memory is read or written.  */
+ #define EAF_DIRECT            (1 << 0)
+ /* Nonzero if memory reached by the argument is not clobbered.  */
+ #define EAF_NOCLOBBER         (1 << 1)
+ /* Nonzero if the argument does not escape.  */
+ #define EAF_NOESCAPE          (1 << 2)
+ /* Nonzero if the argument is not used by the function.  */
+ #define EAF_UNUSED            (1 << 3)
+
+ /* Call return flags.  */
+ 
+ /* Mask for the argument number that is returned.  Lower two bits of
+    the return flags, encodes argument slots zero to three.  */
+ #define ERF_RETURN_ARG_MASK   (3)
+ /* Nonzero if the return value is equal to the argument number
+    flags & ERF_RETURN_ARG_MASK.  */
+ #define ERF_RETURNS_ARG               (1 << 2)
+ /* Nonzero if the return value does not alias with anything.  Functions
+    with the malloc attribute have this set on their return value.  */
+ #define ERF_NOALIAS           (1 << 3)


-- 
           Summary: Optimization of libgfortran calls: function annotations
                    for noclobber/noescape arguments
           Product: gcc
           Version: 4.5.0
            Status: UNCONFIRMED
          Keywords: missed-optimization
          Severity: normal
          Priority: P3
         Component: fortran
        AssignedTo: unassigned at gcc dot gnu dot org
        ReportedBy: burnus at gcc dot gnu dot org


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=43665


             reply	other threads:[~2010-04-06 13:30 UTC|newest]

Thread overview: 27+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2010-04-06 13:30 burnus at gcc dot gnu dot org [this message]
2010-04-06 13:51 ` [Bug fortran/43665] " burnus at gcc dot gnu dot org
2010-05-10 10:11 ` burnus at gcc dot gnu dot org
2010-05-13 10:31 ` dfranke at gcc dot gnu dot org
2010-07-13 13:21 ` burnus at gcc dot gnu dot org
2010-07-13 17:26 ` burnus at gcc dot gnu dot org
2010-07-13 17:32 ` [Bug fortran/43665] INTENT(IN) etc. optimization of " burnus at gcc dot gnu dot org
2010-07-20 15:06 ` burnus at gcc dot gnu dot org
2010-07-20 15:26 ` burnus at gcc dot gnu dot org
2010-07-20 15:58 ` burnus at gcc dot gnu dot org
2010-07-20 16:01 ` rguenth at gcc dot gnu dot org
2010-07-20 16:12 ` burnus at gcc dot gnu dot org
2010-07-21  8:09 ` rguenther at suse dot de
2010-07-21  8:27 ` jamborm at gcc dot gnu dot org
2010-07-22 15:36 ` burnus at gcc dot gnu dot org
2010-09-06 14:14 ` jamborm at gcc dot gnu dot org
2010-09-06 18:26 ` jamborm at gcc dot gnu dot org
2010-09-06 18:44 ` burnus at gcc dot gnu dot org
2010-09-07 17:01 ` jamborm at gcc dot gnu dot org
2010-09-08  6:25 ` burnus at gcc dot gnu dot org
2010-09-09  8:43 ` burnus at gcc dot gnu dot org
2010-09-10 12:09 ` burnus at gcc dot gnu dot org
2010-09-10 12:12 ` burnus at gcc dot gnu dot org
2010-09-11 15:13 ` dominiq at lps dot ens dot fr
2010-09-12  9:32 ` burnus at gcc dot gnu dot org
2010-09-12 10:14 ` dominiq at lps dot ens dot fr
2010-09-16 21:30 ` burnus at gcc dot gnu dot org

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=bug-43665-13404@http.gcc.gnu.org/bugzilla/ \
    --to=gcc-bugzilla@gcc.gnu.org \
    --cc=gcc-bugs@gcc.gnu.org \
    /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).