http://gcc.gnu.org/bugzilla/show_bug.cgi?id=54146
--- Comment #62 from Steven Bosscher 2012-10-15
21:32:04 UTC ---
Author: steven
Date: Mon Oct 15 21:31:57 2012
New Revision: 192476
URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=192476
Log:
Backport from trunk (r190222):
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=54146
--- Comment #61 from Jan Hubicka 2012-09-10 16:04:38
UTC ---
> > I think it would be good to deprecate the flatten attribute...
>
> It still can be useful I think, if only for creating testcases
> with arbitrary large functions ;)
I think it is
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=54146
--- Comment #60 from Steven Bosscher 2012-09-04
10:50:47 UTC ---
Another improvement that can be done for the test case, is to assign
pseudo-register numbers such that the density of the DF_LR and DF_LIVE bitmaps
increases.
For the density I use
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=54146
--- Comment #59 from Steven Bosscher 2012-09-02
22:54:34 UTC ---
FWIW Martin: SRA blows up this test case's register pressure. Compiling with
SRA enabled takes ~900s, but with -fno-tree-sra compile time almost halves.
There are extremely long liv
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=54146
--- Comment #58 from stevenb.gcc at gmail dot com 2012-08-21 13:56:27 UTC ---
FWIW, I think all patches addressing parts of this bug are candidates
for back-porting to release branches. They are all almost trivial.
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=54146
--- Comment #57 from Richard Guenther 2012-08-21
13:34:28 UTC ---
Author: rguenth
Date: Tue Aug 21 13:34:19 2012
New Revision: 190562
URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=190562
Log:
2012-08-21 Richard Guenther
Backport
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=54146
--- Comment #56 from rguenther at suse dot de
2012-08-21 07:55:14 UTC ---
On Mon, 20 Aug 2012, steven at gcc dot gnu.org wrote:
> http://gcc.gnu.org/bugzilla/show_bug.cgi?id=54146
>
> Steven Bosscher changed:
>
>What|Removed
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=54146
Steven Bosscher changed:
What|Removed |Added
Status|NEW |RESOLVED
Resolution|
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=54146
--- Comment #54 from Steven Bosscher 2012-08-17
09:42:15 UTC ---
Author: steven
Date: Fri Aug 17 09:42:06 2012
New Revision: 190475
URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=190475
Log:
PR middle-end/54146
* tree-ssa-loop-im.
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=54146
Steven Bosscher changed:
What|Removed |Added
Attachment #28020|0 |1
is obsolete|
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=54146
--- Comment #52 from Richard Guenther 2012-08-16
14:27:59 UTC ---
Author: rguenth
Date: Thu Aug 16 14:27:51 2012
New Revision: 190445
URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=190445
Log:
2012-08-16 Richard Guenther
PR middle
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=54146
--- Comment #51 from rguenther at suse dot de
2012-08-16 14:06:06 UTC ---
On Thu, 16 Aug 2012, stevenb.gcc at gmail dot com wrote:
> http://gcc.gnu.org/bugzilla/show_bug.cgi?id=54146
>
> --- Comment #50 from stevenb.gcc at gmail dot com com> 2
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=54146
--- Comment #50 from stevenb.gcc at gmail dot com 2012-08-16 13:55:40 UTC ---
On Thu, Aug 16, 2012 at 2:10 PM, rguenth at gcc dot gnu.org
wrote:
> bitmap stats are confusing because they show leaks for bitmaps we free
> by releasing their obstack
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=54146
Richard Guenther changed:
What|Removed |Added
Keywords||memory-hog
--- Comment #49 from Richar
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=54146
--- Comment #48 from Steven Bosscher 2012-08-16
10:52:20 UTC ---
Author: steven
Date: Thu Aug 16 10:52:14 2012
New Revision: 190442
URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=190442
Log:
PR middle-end/54146
* tree-flow.h (comp
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=54146
--- Comment #47 from Steven Bosscher 2012-08-15
15:07:05 UTC ---
(In reply to comment #46)
> Created attachment 28020 [details]
> Faster rewrite_into_loop_closed_ssa
After this patch, IRA is the only major bottle-neck left, although there are
st
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=54146
Steven Bosscher changed:
What|Removed |Added
Attachment #27930|0 |1
is obsolete|
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=54146
--- Comment #45 from Marc Glisse 2012-08-14
13:20:55 UTC ---
(In reply to comment #11)
> > Marc, do you know where the use of the
> > flatten attribute comes from in your code?
> Comes from the Eigen library, I'll talk to them about it and see if
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=54146
--- Comment #44 from Richard Guenther 2012-08-14
12:38:41 UTC ---
Author: rguenth
Date: Tue Aug 14 12:38:32 2012
New Revision: 190382
URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=190382
Log:
2012-08-14 Richard Guenther
PR tree-o
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=54146
--- Comment #43 from Jan Hubicka 2012-08-10
07:52:31 UTC ---
Author: hubicka
Date: Fri Aug 10 07:52:23 2012
New Revision: 190283
URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=190283
Log:
PR middle-end/54146
* ipa-inline-transfor
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=54146
--- Comment #42 from Steven Bosscher 2012-08-10
06:32:37 UTC ---
(In reply to comment #40)
> Quite an achivement that Steven managed to
> chase out all the other cases.
Thanks for the compliment :-)
I'm still working on the rewrite_into_loop_cl
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=54146
--- Comment #41 from Jan Hubicka 2012-08-10
05:18:52 UTC ---
Created attachment 27979
--> http://gcc.gnu.org/bugzilla/attachment.cgi?id=27979
Path for inliner slowness
Hi,
this is patch I am testing. After some consideration I do not like the
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=54146
--- Comment #40 from Jan Hubicka 2012-08-10
04:34:54 UTC ---
OK,
my simple ^C profiling shows:
#14 0x0091f17f in estimate_edge_size_and_time (e=0x7fffb7192d68,
size=, time=0x7fffc04110b0, prob=1) at
../../gcc/ipa-inline-analysis.c:218
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=54146
--- Comment #39 from Jan Hubicka 2012-08-10 03:53:46
UTC ---
> Martin, can you look at comment #14 and the patch? I think what we want to
> do in flatten_function is before
>
> inline_call (e, true, NULL, NULL);
>
> reset the edge predic
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=54146
--- Comment #38 from Jan Hubicka 2012-08-10 03:50:26
UTC ---
> I don't understand how inline_merge_summary is supposed to work, so I'm going
> to leave that one for Richi and Honza.
Well, it produces inline summary for function that is result of
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=54146
--- Comment #37 from Jan Hubicka 2012-08-10 03:45:31
UTC ---
>
> I suppose it's the old issue that we update fibheap keys along each
> inlining decision - and with flatten there are just very many ... Honza?
Well, I killed most of updating for
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=54146
--- Comment #36 from Steven Bosscher 2012-08-08
17:39:49 UTC ---
Author: steven
Date: Wed Aug 8 17:39:46 2012
New Revision: 190235
URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=190235
Log:
PR middle-end/54146
* gimpify.c (gimpli
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=54146
--- Comment #35 from Richard Guenther 2012-08-08
11:49:21 UTC ---
(In reply to comment #34)
> (In reply to comment #33)
> > I think you should simply move compute_global_livein to its single use
> > and make it static.
>
> Yes, and I need to add
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=54146
--- Comment #34 from Steven Bosscher 2012-08-08
10:10:46 UTC ---
(In reply to comment #33)
> I think you should simply move compute_global_livein to its single use
> and make it static.
Yes, and I need to add the same smarts there as in find_use
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=54146
--- Comment #33 from Richard Guenther 2012-08-08
08:09:25 UTC ---
(In reply to comment #29)
> Created attachment 27957 [details]
> Do not traverse sibling loops
>
> The idea here is to note that for a nested loop we know for sure that the loop
>
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=54146
Steven Bosscher changed:
What|Removed |Added
Component|tree-optimization |middle-end
--- Comment #32 from Steven
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=54146
Steven Bosscher changed:
What|Removed |Added
Component|tree-optimization |middle-end
--- Comment #31 from Steven
32 matches
Mail list logo
