> -Original Message-
> From: pins...@gmail.com [mailto:pins...@gmail.com] On Behalf Of Andrew
> Pinski
> Sent: Wednesday, July 04, 2012 2:58 PM
> To: Terry Guo
> Cc: Richard Guenther; gcc-patches@gcc.gnu.org; tob...@grosser.es;
> seb...@gmail.com; Michael Matz; Diego Novillo; Joey Ye
> Su
> -Original Message-
> From: Terry Guo [mailto:terry@arm.com]
> Sent: Wednesday, July 04, 2012 15:12
> To: 'Andrew Pinski'; tob...@grosser.es
> Cc: gcc-patches@gcc.gnu.org; Joey Ye
> Subject: RE: [PATCH] Move Graphite from using PPL over to ISL
>
>
>
> > -Original Message-
Hello Mikael,
On 07/03/2012 09:56 PM, Mikael Morin wrote:
The patch below (which is the one posted on bugzilla) disables
reduction handling if subscript is true, so that the reductions are
handled from the outer call instead of from the subscript call.
I have finally convinced myself that it is
On Wed, Jul 4, 2012 at 8:42 AM, Steven Bosscher wrote:
> Hello,
>
> The attached patch splits out the code generation bits from
> expand_case to two new functions, emit_case_decision_tree and
> emit_case_dispatch_table.
>
> I also added some bits of new code to help me better understand how
> the
Hello,
The attached patch simplifies the SH zero_extend expanders a little.
According to CSiBE it also improves code size a little bit here and
there.
Tested against rev 189081 with
make -k check RUNTESTFLAGS="--target_board=sh-sim
\{-m2/-ml,-m2/-mb,-m2a/-mb,-m2a-single/-mb,-m4/-ml,-m4/-mb,
-m4-s
On Wed, Jul 4, 2012 at 12:11 AM, Terry Guo wrote:
>
>
>> -Original Message-
>> From: pins...@gmail.com [mailto:pins...@gmail.com] On Behalf Of Andrew
>> Pinski
>> Sent: Wednesday, July 04, 2012 2:58 PM
>> To: Terry Guo
>> Cc: Richard Guenther; gcc-patches@gcc.gnu.org; tob...@grosser.es;
>>
> > Hi Tobi and Andrew,
> >
> > Thanks for your timely answers. I just saw Sebastian's comments:
> >
> > Yes, having GCC only depend on ISL and CLooG-ISL (and not depend
> > anymore on PPL) is our plan for 4.7.
> >
> > from http://gcc.gnu.org/ml/gcc-patches/2011-08/msg01161.html.
>
> Yes but thos
On Jul 2, 2012, at 1:43 PM, Richard Guenther wrote:
> On Thu, 28 Jun 2012, Diego Novillo wrote:
>
>> On 12-06-27 11:06 , Richard Guenther wrote:
>>
>>> 2012-06-27 Richard Guenther
>>> Michael Matz
>>> Tobias Grosser
>>> Sebastian Pop
>>>
>>> config/
>>> * cloog.m4: S
On Wed, 4 Jul 2012, Tristan Gingold wrote:
>
> On Jul 2, 2012, at 1:43 PM, Richard Guenther wrote:
>
> > On Thu, 28 Jun 2012, Diego Novillo wrote:
> >
> >> On 12-06-27 11:06 , Richard Guenther wrote:
> >>
> >>> 2012-06-27 Richard Guenther
> >>> Michael Matz
> >>> Tobias Grosser
> >>>
Mikael,
There is a typo in your test gfortran.dg/inline_sum_4.f90:
"arr(1, :, :, =" should be "arr(1, :, :, :) =".
Thanks for the patch.
Dominique
On Jul 4, 2012, at 10:58 AM, Richard Guenther wrote:
> On Wed, 4 Jul 2012, Tristan Gingold wrote:
>
>>
>> On Jul 2, 2012, at 1:43 PM, Richard Guenther wrote:
>>
>>> On Thu, 28 Jun 2012, Diego Novillo wrote:
>>>
On 12-06-27 11:06 , Richard Guenther wrote:
> 2012-06-27 Richard G
On Wed, 4 Jul 2012, Tristan Gingold wrote:
>
> On Jul 4, 2012, at 10:58 AM, Richard Guenther wrote:
>
> > On Wed, 4 Jul 2012, Tristan Gingold wrote:
> >
> >>
> >> On Jul 2, 2012, at 1:43 PM, Richard Guenther wrote:
> >>
> >>> On Thu, 28 Jun 2012, Diego Novillo wrote:
> >>>
> On 12-06-27
The previous patch adjusting free_lang_data_in_type missed to adjust
find_decls_types_r. We might not be visiting all types we use in the
end due to this.
LTO bootstrapped and tested on x86_64-unknown-linux-gnu, applying to
trunk and branch.
Richard.
2012-07-04 Richard Guenther
* t
Tejas Belagod wrote:
Ulrich Weigand wrote:
Tejas Belagod wrote:
Therefore strict_memory_address_addr_space_P () thinks that
(mem:OI (reg sp)) is a valid target address and lets it pass as
a subreg and does not narrow the subreg into a narrower memref.
find_reloads_toplev () should have infact
On Jul 4, 2012, at 12:32 PM, Richard Guenther wrote:
> On Wed, 4 Jul 2012, Tristan Gingold wrote:
>
>>
>> On Jul 4, 2012, at 10:58 AM, Richard Guenther wrote:
>>
>>> On Wed, 4 Jul 2012, Tristan Gingold wrote:
>>>
On Jul 2, 2012, at 1:43 PM, Richard Guenther wrote:
> On Th
On Wed, 4 Jul 2012, Tristan Gingold wrote:
>
> On Jul 4, 2012, at 12:32 PM, Richard Guenther wrote:
>
> > On Wed, 4 Jul 2012, Tristan Gingold wrote:
> >
> >>
> >> On Jul 4, 2012, at 10:58 AM, Richard Guenther wrote:
> >>
> >>> On Wed, 4 Jul 2012, Tristan Gingold wrote:
> >>>
>
> O
On Jul 4, 2012, at 12:59 PM, Richard Guenther wrote:
> On Wed, 4 Jul 2012, Tristan Gingold wrote:
>
>>
>> On Jul 4, 2012, at 12:32 PM, Richard Guenther wrote:
>>
>>> On Wed, 4 Jul 2012, Tristan Gingold wrote:
>>>
On Jul 4, 2012, at 10:58 AM, Richard Guenther wrote:
> On W
On Wed, 4 Jul 2012, Tristan Gingold wrote:
>
> On Jul 4, 2012, at 12:59 PM, Richard Guenther wrote:
>
> > On Wed, 4 Jul 2012, Tristan Gingold wrote:
> >
> >>
> >> On Jul 4, 2012, at 12:32 PM, Richard Guenther wrote:
> >>
> >>> On Wed, 4 Jul 2012, Tristan Gingold wrote:
> >>>
>
> O
On Wed, 2012-07-04 at 10:49 +0200, Richard Guenther wrote:
> On Tue, 3 Jul 2012, William J. Schmidt wrote:
>
> > Hi Richard,
> >
> > Here's a revision incorporating changes addressing your comments. As
> > before it passes bootstrap and regression testing on powerpc64-linux-gnu
> > and compiles
This fixes PR53849 - we were calling add_referenced_var on a global
which is not permitted. The following fixes that and also simplifies
the code, making it faster (cfun push/pop can be quite expensive).
Bootstrapped and tested on x86_64-unknown-linux-gnu, applied.
Richard.
2012-07-04 Richard
This fixes a missed dead store elimination case. We were not
properly disregarding a loop PHI use we have already visited.
Bootstrapped and tested on x86_64-unknown-linux-gnu, applied.
Richard.
2012-07-04 Richard Guenther
PR tree-optimization/53844
* tree-ssa-dse.c (dse_pos
On Jul 4, 2012, at 1:07 PM, Richard Guenther wrote:
> On Wed, 4 Jul 2012, Tristan Gingold wrote:
>
>>
>> On Jul 4, 2012, at 12:59 PM, Richard Guenther wrote:
>>
>>> On Wed, 4 Jul 2012, Tristan Gingold wrote:
>>>
On Jul 4, 2012, at 12:32 PM, Richard Guenther wrote:
> On We
In initializer constant folding we use error_mark_node as a magic
return value which indicates a not present but kown zero initializer.
LTO on the other hand uses error_mark_node to mark something as
having a non-empty initializer but not present in the current ltrans
unit. Those uses conflict -
Hello!
Several otherwise harmless "may be used uninitialized" warnings break
LTO profiled bootstrap. Attached patch fixes all these places,
enabling profiled LTO bootstrap to continue.
2012-07-04 Uros Bizjak
* expmed.c (expand_mult): Initialize coeff and is_neg.
java/ChangeLog:
201
Hello!
Attached patch partially reverts r187375 [1]. Apparently, the patch
added some new functionality to symtab_remove_unreachable_nodes,
although the ChangeLog said:
Log:
...
(symtab_remove_unreachable_nodes): Cleanup.
It looks to me that since we changed something, we also ha
On Tue, Jul 3, 2012 at 11:07 PM, Sharad Singhai wrote:
> Apologies for the spam. Attempting to resend the patch after shrinking it.
>
> I have updated the attached patch to use a new dump message
> classification system for the vectorizer. It currently uses four
> classes, viz, MSG_OPTIMIZED_LOCAT
On Wed, 4 Jul 2012, Uros Bizjak wrote:
> Hello!
>
> Attached patch partially reverts r187375 [1]. Apparently, the patch
> added some new functionality to symtab_remove_unreachable_nodes,
> although the ChangeLog said:
>
> Log:
> ...
> (symtab_remove_unreachable_nodes): Cleanup.
>
>
On 06/29/2012 06:31 PM, Ramana Radhakrishnan wrote:
Ok with this comment?
+;; Split symbol_refs at the later stage (after cprop), instead of
generating
+;; movt/movw pair directly at expand. Otherwise corresponding high_sum
+;; and lo_sum would be merged back into memory load at cprop. Howeve
Comment?
David
On Mon, Jul 2, 2012 at 4:30 PM, Xinliang David Li wrote:
> I extended the patch a little so that the option can be used to set
> multiple stack reuse levels: -fstack-reuse=[all|name_vars|none]
>
> all: enable stack reuse for all local vars (named vars and compiler
> generated temp
On Mon, Jul 2, 2012 at 12:04 PM, Jason Merrill wrote:
> While a C++ enumeration is being defined, the enumerators have the types of
> their initializers, rather than the enumeration type itself. As a result,
> if two enumerators are initialized from two other enumerations, and then the
> two are
The last merge was with unstable track. So I merged with trunk again at
189234.
On 06/05/2012 10:34 AM, Sandra Loosemore wrote:
2012-06-05 Sandra Loosemore
gcc/
* tree-ssa-loop-ivopts.c (comp_cost): Make complexity field signed.
Update comments to indicate this is for addressing mode complexity.
(new_cost): Make signedness of parameters mat
Hello,
starting with 4.7, if multiple __builtin_unreachable statements occur in
a single function, they are no longer optimized as they used to be.
For example,
int foo(int a)
{
if (a <= 0)
__builtin_unreachable();
if (a > 2)
__builtin_unreachable();
return a > 0;
}
On 31/01/12 22:07, Tom de Vries wrote:
> On 31/01/12 22:05, Tom de Vries wrote:
>> Richard,
>>
>
> Sorry, with patch this time.
>
>> this patch fixes PR52009.
>>
>> Consider this test-case:
>> ...
>> int z;
>>
>> void
>> foo (int y)
>> {
>> if (y == 6)
>> z = 5;
>> else
>> z = 5;
>> }
On Wed, Jul 4, 2012 at 10:02 AM, Ulrich Weigand wrote:
> Any suggestions how to fix this? Should tail merging detect
> __builtin_unreachable and not merge such block? Or else, should
> the CFG optimizer be extended (how?) to handle unreachable blocks
> with multiple predecessors better?
This bu
On Wed, Jul 4, 2012 at 7:02 PM, Ulrich Weigand wrote:
> Any suggestions how to fix this? Should tail merging detect
> __builtin_unreachable and not merge such block?
That seems to be the most straight-forward thing to do. I don't think
there are any other passes that do this kind of code merging
Hello.
Ever since the ISL patch went in my builds have failed.
I am building with local copies of all the libraries, so I have added
gmp, mpfr, mpc, isl and cloog from ftp://gcc.gnu.org/pub/gcc/infrastructure
to my source directory.
The command sequence I use is
svn co svn://gcc.gnu.org/svn/gcc
On 07/03/2012 04:37 PM, Lawrence Crowl wrote:
On 7/1/12, Jason Merrill wrote:
As discussed, I would say that RTTI is currently not permitted
but could be added later.
But isn't "could be added later" always true? Other folks have
objected to such wording on the grounds that it adds no inform
The following patch fixes one testsuite regression on ppc64. LRA
removed two insns setting asm reg variable assigned to eliminable hard
regnos. It should be not done.
The patch was successfully bootstrapped on x86-64, ppc64, ia64.
Committed as rev. 189264.
2012-07-04 Vladimir Makarov
retrofit_lang_decl now wants to look at the DECL_CONTEXT of an
enumerator, but we were calling it before setting the context. The
comment before the call to build_lang_decl_loc mentions
DECL_CLASS_CONTEXT, but that macro doesn't use DECL_LANG_SPECIFIC
anymore, so I decided to check whether we
Oleg Endo wrote:
> The attached patch simplifies the SH zero_extend expanders a little.
> According to CSiBE it also improves code size a little bit here and
> there.
This patch is OK.
Regards,
kaz
This fixes a deadlock in condition_variable_any, which I should have
fixed as part of PR 50862.
PR libstdc++/53830
* include/std/condition_variable (condition_variable_any::wait):
Move _Unlock type to class scope.
(condition_variable_any::wait_until): Reuse it.
Using "Index" as a node name in Texinfo results in a file named
"Index.html" when you generate HTML output. On systems with
case-insensitive filenames, like Windows, this conflicts with the
top-level "index.html" that contains the table of contents. In
particular, see:
http://gcc.gnu.org/on
Hi Richard,
picking up a yet uncommitted part of the MIPS16 changes, see below:
On 2012/2/3 11:28 PM, Richard Sandiford wrote:
> Chung-Lin Tang writes:
>>> (2) is interesting if there is also a way to build those MIPS16 libraries
>>> out of the box. I'd like such a mechanism to be added at the s
44 matches
Mail list logo
