Patchwork [i386] : Fix PR51109, symbol size in scheduler state machine is reduced

login
register
mail settings
Submitter Jakub Jelinek
Date Oct. 10, 2012, 11 a.m.
Message ID <20121010110030.GX26735@tucnak.redhat.com>
Download mbox | patch
Permalink /patch/190612/
State New
Headers show

Comments

Jakub Jelinek - Oct. 10, 2012, 11 a.m.
On Wed, Oct 10, 2012 at 12:57:09PM +0200, Paolo Carlini wrote:
> On 10/10/2012 12:53 PM, Jakub Jelinek wrote:
> >Yeah, clearly a different version of the patch has been posted
> >vs. what has been checked in.  The difference is removal of the
> >(define_cpu_unit "bdver1-mult" "bdver1_mult")
> >line (present in the posted patch, not in the checked in patch).
> >
> >Also, the ChangeLog entry is wrong:
> >	PR 51109
> >	* gcc/config/i386/bdver1.md (bdver1_int): Automaton has been
> >	split to reduce state transitions.
> >- should start with:
> >	PR target/51109
> >	* config/i386/bdver1.md
> >and really document the changes it has done, or just not contain
> >(bdver1_int) and document the file changes.
> >
> >And Copyright year hasn't been updated (but this wasn't first
> >commit in 2012, others made the same mistake, no changes in 2011
> >btw).
> Shall we (I) just revert the change for now?

I have removed the extra line as obvious in SVN, to allow my
bootstraps to continue.


	Jakub
Paolo Carlini - Oct. 10, 2012, 11:02 a.m.
On 10/10/2012 01:00 PM, Jakub Jelinek wrote:
> I have removed the extra line as obvious in SVN, to allow my
> bootstraps to continue.
Thanks!

Paolo.
Gopalasubramanian, Ganesh - Oct. 10, 2012, 11:14 a.m.
That was obvious. Sorry for the wrong commit.

Thanks Jakub.

-Ganesh

-----Original Message-----
From: Paolo Carlini [mailto:paolo.carlini@oracle.com] 
Sent: Wednesday, October 10, 2012 4:33 PM
To: Jakub Jelinek
Cc: Gopalasubramanian, Ganesh; Uros Bizjak; gcc-patches@gcc.gnu.org; vekumar@gcc.gnu.org
Subject: Re: [PATCH, i386]: Fix PR51109, symbol size in scheduler state machine is reduced

On 10/10/2012 01:00 PM, Jakub Jelinek wrote:
> I have removed the extra line as obvious in SVN, to allow my 
> bootstraps to continue.
Thanks!

Paolo.

Patch

--- gcc/ChangeLog	(revision 192304)
+++ gcc/ChangeLog	(revision 192305)
@@ -1,5 +1,8 @@ 
 2012-10-10  Jakub Jelinek  <jakub@redhat.com>
 
+	PR target/51109
+	* config/i386/bdver1.md (bdver1-mult): Remove.
+
 	PR middle-end/54879
 	* combine.c (count_rtxs): Use RTX_BIN_ARITH resp. RTX_COMM_ARITH
 	instead of '2' resp. 'c' for GET_RTX_CLASS comparisons.
@@ -17,7 +20,7 @@ 
 2012-10-10  Ganesh Gopalasubramanian  <Ganesh.Gopalasubramanian@amd.com>
 
 	PR target/51109
-	* gcc/config/i386/bdver1.md (bdver1_int): Automaton has been
+	* config/i386/bdver1.md (bdver1_int): Automaton has been
 	split to reduce state transitions.
 
 2012-10-10  Richard Biener  <rguenther@suse.de>
--- gcc/config/i386/bdver1.md	(revision 192304)
+++ gcc/config/i386/bdver1.md	(revision 192305)
@@ -1,4 +1,4 @@ 
-;; Copyright (C) 2010, Free Software Foundation, Inc.
+;; Copyright (C) 2010, 2012 Free Software Foundation, Inc.
 ;;
 ;; This file is part of GCC.
 ;;
@@ -79,8 +79,6 @@  (define_cpu_unit "bdver1-agu0" "bdver1_a
 (define_cpu_unit "bdver1-agu1" "bdver1_agu")
 (define_reservation "bdver1-agu" "(bdver1-agu0 | bdver1-agu1)")
 
-(define_cpu_unit "bdver1-mult" "bdver1_mult")
-
 (define_cpu_unit "bdver1-load0" "bdver1_load")
 (define_cpu_unit "bdver1-load1" "bdver1_load")
 (define_reservation "bdver1-load" "bdver1-agu,