public inbox for gcc-bugs@sourceware.org
help / color / mirror / Atom feed
From: "hubicka at gcc dot gnu.org" <gcc-bugzilla@gcc.gnu.org>
To: gcc-bugs@gcc.gnu.org
Subject: [Bug tree-optimization/55334] mgrid regression (ipa-cp disables vectorization)
Date: Thu, 15 Nov 2012 11:42:00 -0000	[thread overview]
Message-ID: <bug-55334-4-9mtrNdy1vt@http.gcc.gnu.org/bugzilla/> (raw)
In-Reply-To: <bug-55334-4@http.gcc.gnu.org/bugzilla/>


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=55334

Jan Hubicka <hubicka at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |NEW
   Last reconfirmed|                            |2012-11-15
     Ever Confirmed|0                           |1

--- Comment #3 from Jan Hubicka <hubicka at gcc dot gnu.org> 2012-11-15 11:42:21 UTC ---
OK, it is not base_address, but base_object
Creating dr for MEM[(real(kind=8)[4] *)&x + 58071104B][0]
analyze_innermost: success.
        base_address: &x
        offset from base address: 0
        constant offset from base address: 58071104
        step: 0
        aligned to: 128
        base_object: MEM[(real(kind=8)[4] *)&x + 58071104B]
        Access function 0: 0

I do not see why we need to keep the base object to be offsetted. 
While it is first time I see the code, i think it is trying to look for
MEM_REFs and remember the fact that after putting it to 0 the offset may go
across the type boundry, since we changed the type via MEM_REF.  For
non-envolving mem-refs this however is not a problem.
I am testing the following that simply removes the constant from MEM_REF 

Index: tree-data-ref.c
===================================================================
--- tree-data-ref.c    (revision 193503)
+++ tree-data-ref.c    (working copy)
@@ -922,6 +922,16 @@ dr_analyze_indices (struct data_referenc
       DR_UNCONSTRAINED_BASE (dr) = true;
       VEC_safe_push (tree, heap, access_fns, access_fn);
     }
+      else
+    {
+      if (!integer_zerop (TREE_OPERAND (ref, 1)))
+        ref = fold_build2_loc (EXPR_LOCATION (ref),
+                   MEM_REF, TREE_TYPE (ref),
+                   op,
+                   build_int_cst (TREE_TYPE (TREE_OPERAND (ref,
+                                 1)),
+                          0));
+    }
     }
   else if (DECL_P (ref))
     {


  parent reply	other threads:[~2012-11-15 11:42 UTC|newest]

Thread overview: 52+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2012-11-15  2:27 [Bug tree-optimization/55334] New: " hubicka at gcc dot gnu.org
2012-11-15  7:11 ` [Bug tree-optimization/55334] " vincenzo.innocente at cern dot ch
2012-11-15 11:07 ` hubicka at gcc dot gnu.org
2012-11-15 11:42 ` hubicka at gcc dot gnu.org [this message]
2012-11-16 10:39 ` hubicka at gcc dot gnu.org
2012-11-16 10:56 ` vincenzo.innocente at cern dot ch
2012-11-16 12:06 ` jakub at gcc dot gnu.org
2012-11-16 13:20 ` hubicka at gcc dot gnu.org
2012-11-16 13:22 ` hubicka at gcc dot gnu.org
2012-11-16 13:29 ` jakub at gcc dot gnu.org
2012-11-16 13:46 ` vincenzo.innocente at cern dot ch
2012-11-16 13:56 ` jakub at gcc dot gnu.org
2012-11-16 13:59 ` vincenzo.innocente at cern dot ch
2012-11-16 14:17 ` jakub at gcc dot gnu.org
2012-11-16 14:45 ` [Bug tree-optimization/55334] [4.8 Regression] " jakub at gcc dot gnu.org
2012-12-07 10:12 ` rguenth at gcc dot gnu.org
2012-12-10 15:34 ` rguenth at gcc dot gnu.org
2012-12-21 12:26 ` rguenth at gcc dot gnu.org
2012-12-21 13:49 ` hubicka at ucw dot cz
2012-12-21 14:02 ` rguenth at gcc dot gnu.org
2012-12-21 14:09 ` rguenth at gcc dot gnu.org
2012-12-21 14:10 ` jakub at gcc dot gnu.org
2012-12-21 14:12 ` rguenth at gcc dot gnu.org
2012-12-21 14:22 ` hubicka at ucw dot cz
2012-12-21 14:27 ` rguenth at gcc dot gnu.org
2013-02-12 20:11 ` jamborm at gcc dot gnu.org
2013-02-13  9:09 ` rguenth at gcc dot gnu.org
2013-02-19 12:55 ` jamborm at gcc dot gnu.org
2013-02-19 13:50 ` rguenther at suse dot de
2013-02-20 11:13 ` jamborm at gcc dot gnu.org
2013-02-20 12:57 ` jamborm at gcc dot gnu.org
2013-02-20 14:08 ` [Bug tree-optimization/55334] [4.8/4.9 " rguenth at gcc dot gnu.org
2013-03-28 12:36 ` jamborm at gcc dot gnu.org
2013-07-10 23:54 ` chris_s_jones at yahoo dot com
2013-07-24 15:47 ` jamborm at gcc dot gnu.org
2013-07-25 16:33 ` hubicka at ucw dot cz
2014-04-22 11:38 ` [Bug tree-optimization/55334] [4.8/4.9/4.10 " jakub at gcc dot gnu.org
2014-07-16 13:30 ` jakub at gcc dot gnu.org
2014-09-02 15:45 ` [Bug tree-optimization/55334] [4.8/4.9/5 " vp at gcc dot gnu.org
2014-09-03 10:04 ` jamborm at gcc dot gnu.org
2014-09-03 10:25 ` rguenther at suse dot de
2014-09-03 14:11 ` rguenth at gcc dot gnu.org
2014-09-05 12:41 ` rguenth at gcc dot gnu.org
2014-10-30 10:43 ` jakub at gcc dot gnu.org
2014-11-24  9:25 ` rguenth at gcc dot gnu.org
2014-11-24  9:25 ` [Bug tree-optimization/55334] [4.8/4.9 " rguenth at gcc dot gnu.org
2014-11-24  9:25 ` [Bug tree-optimization/55334] [4.8/4.9/5 " rguenth at gcc dot gnu.org
2015-02-16 11:26 ` [Bug tree-optimization/55334] [4.8/4.9 " andrew.n.senkevich at gmail dot com
2015-02-16 11:29 ` jakub at gcc dot gnu.org
2015-02-16 12:17 ` rguenther at suse dot de
2015-06-26 20:01 ` [Bug tree-optimization/55334] [4.9 " jakub at gcc dot gnu.org
2015-06-26 20:31 ` jakub at gcc dot gnu.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-55334-4-9mtrNdy1vt@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).