> On 02/28/2014 03:56 PM, Jan Hubicka wrote:
> >I think we can safely test here DECL_ARTIFICIAL && (DECL_EXTERNAL ||
> >DECL_COMDAT). If the dtor is going to be output anyway, we are safe to use
> >it.
>
> We already skipped DECL_EXTERNAL decls, and artificial members are
> always DECL_COMDAT, b
On Fri, Feb 28, 2014 at 7:23 PM, H.J. Lu wrote:
> On Fri, Feb 28, 2014 at 9:42 AM, H.J. Lu wrote:
>> On Fri, Feb 28, 2014 at 9:25 AM, H.J. Lu wrote:
>>> On Fri, Feb 28, 2014 at 8:11 AM, H.J. Lu wrote:
On Fri, Feb 28, 2014 at 2:09 AM, Richard Biener
wrote:
> On Fri, Feb 28, 2014 a
Hi
> On 28/feb/2014, at 23:48, Marc Glisse wrote:
>
> Hello,
>
> this is a stage 1 patch, and I'll ping it then, but if you have comments
> now...
>
> Passes bootstrap+testsuite on x86_64-linux-gnu.
>
> 2014-02-28 Marc Glisse
>
>PR tree-optimization/57742
> gcc/
>* tree-ssa-forwp
Hi H.J,
Sorry that I will be out of office next week, and don't have chance to
reproduce it until back. BTW, does x32 refer to x86 32 bit?
Thanks,
bin
On Sat, Mar 1, 2014 at 2:23 AM, H.J. Lu wrote:
> On Fri, Feb 28, 2014 at 9:42 AM, H.J. Lu wrote:
>> On Fri, Feb 28, 2014 at 9:25 AM, H.J. Lu w
Hi,
I've just committed the attached patch that fixes PR 60071 as rev
208242. Originally tested by Kaz:
http://gcc.gnu.org/ml/gcc-testresults/2014-02/msg01866.html
Tested the new test case with:
make check-gcc RUNTESTFLAGS="compile.exp=pr60071.c --target_board=sh-sim
\{-m2/-ml,-m2/-mb,-m2a/-mb,-m
There is some uncertainty in the PR discussion about what semantics
vector && should have, so this patch just avoids the ICE for now.
Tested x86_64-pc-linux-gnu, applying to trunk and 4.8.
commit cbf13b08495010d24a784170f76684e5109bd92b
Author: Jason Merrill
Date: Fri Feb 28 23:56:22 2014 -05
On 03/01/2014 03:52 AM, Jan Hubicka wrote:
a hidden function. We don't do that for user-defined virtual
functions because the user controls whether or not they are defined
in the header, and we don't devirtualize if no definition is
available, but implicitly-declared functions are different beca
Hello,
again, a stage 1 patch that I will ping then, but early comments are
welcome.
PR 59100 was asking to transform n?rotate(x,n):x to rotate(x,n) (because
it can be hard to write a strictly valid rotate in plain C). The operation
is really:
(x != neutral) ? x op y : y
where neutral is su
There seems to be a sufficiently large consensus that volatile asms should not
be treated as full optimization barriers by the compiler. This patch is a
first step towards implementing this conclusion and aims only at addressing
the code quality regressions introduced by
http://gcc.gnu.org/vi
On Sat, 1 Mar 2014, Paolo Carlini wrote:
Hi
On 28/feb/2014, at 23:48, Marc Glisse wrote:
Hello,
this is a stage 1 patch, and I'll ping it then, but if you have comments now...
Passes bootstrap+testsuite on x86_64-linux-gnu.
2014-02-28 Marc Glisse
PR tree-optimization/57742
gcc/
Hi Mikael, hi all,
2014-02-22 16:38 GMT+01:00 Mikael Morin :
> Le 19/02/2014 16:51, Janus Weil a écrit :
>> The patch was not applying cleanly any more, so here is a re-diffed
>> version for current trunk. It works nicely on the included test case
>> as well as the one provided by Walter Spector i
Eric Botcazou writes:
> There seems to be a sufficiently large consensus that volatile asms should
> not
> be treated as full optimization barriers by the compiler. This patch is a
> first step towards implementing this conclusion and aims only at addressing
> the code quality regressions int
Hi all,
this is to notify about the patch I've submitted:
http://gcc.gnu.org/ml/gcc-bugs/2014-02/msg02863.html
Cheers,
Wojtek
> On 03/01/2014 03:52 AM, Jan Hubicka wrote:
> >>a hidden function. We don't do that for user-defined virtual
> >>functions because the user controls whether or not they are defined
> >>in the header, and we don't devirtualize if no definition is
> >>available, but implicitly-declared functions ar
... in fact, we can also imagine the below clean-up, for another
DECL_DELETED_FN use and likewise for DECL_DECLARED_CONSTEXPR_P (which
immediately applies STRIP_TEMPLATE to its argument) uses. Or for Stage
1, maybe?
Thanks,
Paolo.
//
2014-03-01 Paolo Carlini
* meth
Hi,
On 02/28/2014 04:50 PM, Jason Merrill wrote:
OK, thanks.
Applied. I have just noticed (sorry) that get_AT_ref (thus get_AT) isn't
trivial at all, thus I propose to apply the below. Is it Ok with you?
Thanks,
Paolo.
//
2014-03-01 Paolo Carlini
* dwarf2out.c
On Sat, Mar 1, 2014 at 7:13 AM, Kirill Yukhin wrote:
> Hello Uroš,
> On 28 Feb 13:55, Uros Bizjak wrote:
>> On Fri, Feb 28, 2014 at 1:14 PM, Kirill Yukhin
>> wrote:
>> > Hello,
>> > This is relatively obvious patch which eliminates comparision
>> > of inifinities for exp2 AVX-512 test and proper
Hi Lin,
On Fri, 28 Feb 2014 19:14:11, lin zuojian wrote:
>
> 于 2014年02月28日 15:58, lin zuojian 写道:
>> Hi Bernd,
>> I agree you with the mode problem.
>>
>> And I have not change the stack alignment.What I change is the virtual
>> register base's alignment.
I tried your patch on this test case:
gc
PR c++/60377
* parser.c (cp_parser_parameter_declaration_clause): Unwind generic
function scope on parse error in function parameter list.
PR c++/60377
* g++.dg/cpp1y/pr60377.C: New testcase.
---
gcc/cp/parser.c | 7 ++-
gcc/testsui
We shouldn't complain if a loop condition isn't a valid constant
expression; it isn't supposed to be.
Tested x86_64-pc-linux-gnu, applying to trunk.
commit 5266ff07b6077df5d7bf6d5748856a43e48ad6ee
Author: Jason Merrill
Date: Sat Mar 1 14:22:02 2014 -0500
PR c++/60379
* semantics.c
OK.
Jason
OK.
Jason
OK.
Jason
Hi all,
The attached patch fixes this by actually implementing it. I cleaned up some of
the code by getting rid of the tmp_delim variables and adding a "mode" to
write_character which is used to ignore delimiters when writing out variable
names and other namelist parts.
I will prepare a test cas
Hello,
I have just regression tested and committed a patch fixing PR 60341 by
adding two expression type checks before union accesses (it's the same
as the one of the PR).
I plan to backport tomorrow (4.8 and 4.7).
Mikael
Index: gcc/testsuite/gfortran.dg/str_comp_optimize_1.f90
This patch fixes lto/55113 for powerpc.
Combining -fshort-double with -flto is now working fine.
I attach patch and testcase (unsure if testcase is in the right place).
Tested with target powerpc-abispe.
2014-03-01 Paulo Matos
* c-family/c.opt: Add LTO FE support for fshort-double
Built and tested on x86-64-linux.
This is just a test case.
2014-03-01 Edward Smith-Rowland <3dw...@verizon.net>
PR c++/50025
* g++.dg/cpp0x/pr50025.C: New.
Index: g++.dg/cpp0x/pr50025.C
===
--- g++.dg/cpp0x/pr50
Hi,
The following patch fixes two issues: the first issue is PR c++/25940 and the
second is related to PR c++/13699.
The first issue is that the C++ frontend fails to reject duplicate definitions
of functions declared to have C language linkage. This results in the compiler
emitting ASM that def
On Sat, 1 Mar 2014, Patrick Palka wrote:
+ error_at (input_location,
+ "redefinition of %q+#D with C language linkage",
+ x);
+ inform (input_location,
+ "%q+#D previou
On Sat, Mar 1, 2014 at 7:53 PM, Marc Glisse wrote:
> On Sat, 1 Mar 2014, Patrick Palka wrote:
>
>> + error_at (input_location,
>> + "redefinition of %q+#D with C language
>> linkage",
>> + x);
>> +
Hi Bernd,
set_mem_align is not working like magic.
set_mem_align just set the alignment of a memory rtx.And You must aware
that you do so because you are sure this rtx IS aligned like this.
For arm machines, the base the virtual registers are aligned to 8
bytes.You can't just set_mem_align to get
Hi Bernd,
You may send a patch too.Your idea will be more clear.
--
Regards
lin zuojian
On Sun, Mar 02, 2014 at 10:24:52AM +0800, lin zuojian wrote:
> Hi Bernd,
>
> set_mem_align is not working like magic.
>
> set_mem_align just set the alignment of a memory rtx.And You must aware
> that you do
Hi,
my mail is not achived by
http://gcc.gnu.org/ml/gcc-patches/2014-02/. What's happening?
--
Regards
lin zuojian
Hi,
I forgot now is Mar.I thought it's Feb.Sorry.
--
Regards
lin zuojian
On Sun, Mar 02, 2014 at 10:57:15AM +0800, lin zuojian wrote:
> Hi,
> my mail is not achived by
> http://gcc.gnu.org/ml/gcc-patches/2014-02/. What's happening?
>
> --
> Regards
> lin zuojian
On Sat, Mar 1, 2014 at 9:57 PM, lin zuojian wrote:
> Hi,
> my mail is not achived by
> http://gcc.gnu.org/ml/gcc-patches/2014-02/. What's happening?
That's last month's archive.
Yeah, I have realized that.
On Sat, Mar 01, 2014 at 10:01:41PM -0500, Patrick Palka wrote:
> On Sat, Mar 1, 2014 at 9:57 PM, lin zuojian wrote:
> > Hi,
> > my mail is not achived by
> > http://gcc.gnu.org/ml/gcc-patches/2014-02/. What's happening?
>
> That's last month's archive.
Pre-remark: I currently get Stage 2/Stage 3 miscompares. As this is
unlikely to be caused by my patch, it properly means that Honza's patch
bitrotted even though it still applies.
Dear all,
attached is an updated patch set for this PR.
Background: gfortran uses internally __builtin_expect. H
37 matches
Mail list logo
