On May 10, 2012, at 10:41 PM, Vladimir Makarov wrote:
> On 05/10/2012 09:10 AM, Tristan Gingold wrote:
>> Hi,
>>
>> I am getting a segfault in ira-color.c:2945 on the trunk:
>>
>> Program received signal SIGSEGV, Segmentation fault.
>> 0x00a79f37 in move_spill_restore () at ../../src/gc
On Thu, 10 May 2012, William J. Schmidt wrote:
> On Thu, 2012-05-10 at 18:49 +0200, Jakub Jelinek wrote:
> > On Thu, May 10, 2012 at 11:44:27AM -0500, William J. Schmidt wrote:
> > > Backporting this patch to 4.7 fixes a problem building Fedora 17.
> > > Bootstrapped and regression tested on power
On 05/11/2012 04:48 AM, Gabriel Dos Reis wrote:
On Thu, May 10, 2012 at 6:40 PM, Paolo Carlini wrote:
Hi,
an ICE on invalid (per Daniel's analysis): when r is NULL_TREE the next
DECL_CONTEXT (r) can only crash. Plus a garbled error message because
pp_cxx_simple_type_specifier doesn't handle BO
I have committed the attached patch as obvious. libgfortran (since GCC
4.5) allocates the memory rank()-times instead of only once in eoshift2.
For the test case of the PR, gfortran leaked 6 MB per call!
Build and regtested on x86-64-linux and comitted as Rev. 187395 to the
4.8 trunk. I intent
Hi,
On 05/11/2012 06:41 AM, Jason Merrill wrote:
On 05/10/2012 05:31 PM, Paolo Carlini wrote:
Let's see if we can do something *now* ;) My concrete proposal would be:
TYPE_PTRMEM_P rename to TYPE_PTRDATAMEM_P (consistent with
TYPE_PTRMEMFUNC_P)
TYPE_PTR_TO_MEMBER_P rename to TYPE_PTRMEM_P
and
On Fri, May 11, 2012 at 4:51 AM, H.J. Lu wrote:
> This patch adds RTM support to -march=native. Tested on Linux/x86-64.
> OK for trunk?
>
> 2012-05-10 H.J. Lu
>
> * config/i386/driver-i386.c (host_detect_local_cpu): Support
> RTM.
OK.
Thanks,
Uros.
On Fri, May 11, 2012 at 3:07 AM, Paolo Carlini wrote:
> Thanks Gaby. Then I guess I'm going to commit the variant with NULL_TREE, I
> like it a tad better because after all formally these functions return
> trees. And consistently I change another instance only a few lines above.
OK. Thanks;
--
On Thu, May 10, 2012 at 6:28 PM, Xinliang David Li wrote:
> I like your suggestion and support the end goal you have. I don't
> like the -fopt-info behavior to interfere with regular -fdump-xxx
> options either.
>
> I think we should stage the changes in multiple steps as originally
> planned. Is
Hi,
I previously noticed from testcases that gcc now does not sink loads from
memory
in tree-ssa-sink.c. After discussing I worked out a patch to support this in
gcc.
Please refer to http://gcc.gnu.org/ml/gcc/2012-04/msg00404.html for some
info.
I think it is a trivial optimization, but it might
... for concreteness, this is something which actually boots and passes
testing on x86_64-linux for the affected languages. I like it ;)
Thanks,
Paolo.
Index: doc/generic.texi
===
--- doc/generic.texi(r
It turns out that g++.dg/debug/dwarf2/nested-3.C was still failing on
Solaris/SPARC: with both as and gas, ! is used as a comment character.
This patch accounts for that.
Tested with the appropritate runtest invocation on sparc-sun-solaris2.11
with both as and gas, installed on mainline.
On Fri, May 11, 2012 at 10:53 AM, Bin Cheng wrote:
> Hi,
> I previously noticed from testcases that gcc now does not sink loads from
> memory
> in tree-ssa-sink.c. After discussing I worked out a patch to support this in
> gcc.
> Please refer to http://gcc.gnu.org/ml/gcc/2012-04/msg00404.html for
Il 11/05/2012 07:13, Matthias Klose ha scritto:
> ok, I did clarify it in the existing documentation of MULTIARCH_DIRNAME in
> fragments.texi, detailing the search order for the files. Should the search
> order be mentioned in some user documentation as well? if yes, where?
Thanks! I don't think
Hi Uros,
>>> there, an alternative patch might be to remove it from the port now
>>> instead of the patch.
>>
>> Right, that's why I was asking for review rather than just installing on
>> my own.
>
> I'd rather see that we remove semicolon in this case, but please note
> that xchg with memory ope
On Thu, May 10, 2012 at 1:19 PM, Jan Hubicka wrote:
> Hi,
> after some thought, the changes into omp-low are not as obviously harmless as
> I
> originally tought. So i decided to handle this by separate patch. This patch
> simply makes cgraph to not release bodies of artificial functions that p
This fixes the dependency of vectorization of strided loads on
gather support. For that to work we need to lift the restriction
in data-ref analysis that requries a constant DR_STEP. Fortunately
fallout is small.
As a side-effect the data-reference for strided loads now makes
sense (rather than
On Fri, May 11, 2012 at 1:54 AM, Uros Bizjak wrote:
> Hello!
>
> Recently testsuite/gcc.c-torture/execute/ieee/pr50310.c started to ICE
> when compiled with -O3 -mieee on alphaev68-pc-linux-gnu:
>
> $ ~/gcc-build-alpha/gcc/cc1 -O3 -mieee -quiet pr50310.c
> pr50310.c: In function ‘foo’:
> pr50310.c
This makes some more pass structs internal to passes.c. It also
makes the phinodes cache global (which it is already - sort of,
it gets cleaned after each early optimization but is kept after
IPA opts).
Bootstrapped on x86_64-unknown-linux-gnu.
Richard.
2012-05-11 Richard Guenther
PING #2!
On Wed, May 2, 2012 at 10:22 PM, Janne Blomqvist
wrote:
> PING
>
> On Thu, Apr 26, 2012 at 12:08 AM, Janne Blomqvist
> wrote:
>> Hi,
>>
>> currently when -frange-check is enabled, we check for overflow when
>> doing a formatted read of an integer value. This check, however, is
>> agains
Ping?
Teresa
On Fri, May 4, 2012 at 3:41 PM, Teresa Johnson wrote:
>
> On David's suggestion, I have removed the changes that rename niter_desc
> to
> loop_desc from this patch to focus the patch on the unrolling changes. I
> can
> submit a cleanup patch to do the renaming as soon as this one goe
Hello,
We have been using these for a few years now. I have just verified
that they do allow exception propagation from signal for Ada on aix 5.3
with a recent mainline.
Posting here in case that is of interest, even though it was only
exercised on aix 5.2 and 5.3, for 32bit code generation (test
On May 10, 2012, at 14:00 , Richard Guenther wrote:
> Of course the mere existence
> of DECL_OFFSET_ALIGN complicates matters for no good reasons (well,
> at least I did not find a good use of it until now ...).
I remember an old discussion about it ... hmm ...
http://gcc.gnu.org/ml/gcc/2006
This removes the renaming variant of find_referenced_vars_in. I'm
working towards removing the mark_symbols_for_renaming hammer.
Bootstrapped and tested on x86_64-unknown-linux-gnu, applied.
Richard.
2012-05-11 Richard Guenther
* tree-flow.h (referenced_var_check_and_insert): Remov
Dear all,
the following patch fixes
character(len=:), allocatable :: str
str = repeat('X', n)
Before the patch, the (re)allocation happened before the RHS was
evaluated. Thus, the string length wasn't known! (Work around: Add
parentheses around the function call.)
Note that the patch a
On Fri, 11 May 2012, Olivier Hainque wrote:
>
> On May 10, 2012, at 14:00 , Richard Guenther wrote:
> > Of course the mere existence
> > of DECL_OFFSET_ALIGN complicates matters for no good reasons (well,
> > at least I did not find a good use of it until now ...).
>
> I remember an old discussi
On 05/11/2012 08:17 AM, Richard Earnshaw wrote:
On 10/05/12 20:58, Vladimir Makarov wrote:
The following patch is for PR53125. The PR is described on
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=53125.
The patch improves the compilation speed by 35% for the case.
The patch was successfully boo
On 05/11/2012 05:40 AM, Uros Bizjak wrote:
2012-05-11 Uros Bizjak
* config/alpha/alpha.c (alpha_emit_conditional_branch): Handle
ORDERED and UNORDERED conditions.
Ok.
r~
>
>
>> You also don't mention how this patch was tested.
>
> I used the testsuite I developed some time ago to test all the Neon
> builtins, which I posted last year on the qemu mailing-list. With the
> current GCCs, this bug is the only remaining one I could detect.
>
Fair enough.
>
>> Alterna
On 10 May 2012 16:05, Joseph S. Myers wrote:
> On Wed, 9 May 2012, Manuel López-Ibáñez wrote:
>
>> 2012-05-09 Manuel López-Ibáñez
>>
>> PR 53063
>> gcc/
>> * doc/options.texi (EnabledBy): Document
>> * opts.c: Include opts.h and options.h before tm.h.
>> (finish_options)
On Fri, May 11, 2012 at 1:49 AM, Richard Guenther
wrote:
> On Thu, May 10, 2012 at 6:28 PM, Xinliang David Li wrote:
>> I like your suggestion and support the end goal you have. I don't
>> like the -fopt-info behavior to interfere with regular -fdump-xxx
>> options either.
>>
>> I think we shoul
On Mon, Apr 16, 2012 at 10:47 AM, H.J. Lu wrote:
> On Tue, Apr 3, 2012 at 7:49 AM, H.J. Lu wrote:
>> On Thu, Mar 29, 2012 at 7:34 AM, H.J. Lu wrote:
>>> On Sat, Mar 3, 2012 at 9:54 AM, H.J. Lu wrote:
Hi,
This patch backports x32 support to libtool:
http://git.savannah.g
On Sun, Apr 29, 2012 at 10:27 AM, H.J. Lu wrote:
> Hi,
>
> This patch adds x32 support to boehm-gc. The same patch has been
> sent to the boehm-gc mailing list. Tested on Linux/x32 and Linux/x86-64.
> OK for trunk?
>
> Thanks.
>
>
> H.J.
> - Forwarded message from "H.J. Lu" -
> Date: Mon
Currently avr_assemble_integer emits an assembler warning to
make gas complain about missing feature
http://sourceware.org/PR13503
if a 3-byte address must be assembled.
As PR13503 is implemented now, avr-gcc can use this feature.
It's only needed for the new __memx address space.
Ok to instal
On Fri, 11 May 2012, Manuel López-Ibáñez wrote:
> Great! Now we have EnabledBy for common options. Now, what should we
> do with language specific settings? One idea could be to have
> something like:
>
> LangEnabledBy(Fortran Ada, Wall)
>
> and then auto-generate something like:
>
> ada_handle
On 11 May 2012 19:04, Joseph S. Myers wrote:
> On Fri, 11 May 2012, Manuel López-Ibáñez wrote:
>
>> Great! Now we have EnabledBy for common options. Now, what should we
>> do with language specific settings? One idea could be to have
>> something like:
>>
>> LangEnabledBy(Fortran Ada, Wall)
>>
>>
On 11 May 2012 19:09, Manuel López-Ibáñez wrote:
> On 11 May 2012 19:04, Joseph S. Myers wrote:
>> On Fri, 11 May 2012, Manuel López-Ibáñez wrote:
>>
>>> Great! Now we have EnabledBy for common options. Now, what should we
>>> do with language specific settings? One idea could be to have
>>> some
On 18 April 2012 at 18:57, Bernhard Reutner-Fischer wrote:
On Tue, Apr 17, 2012 at 12:47:48AM +0200, Tobias Burnus wrote:
Approved but not yet committed:
Bernhard:
- [PATCH] gfortran testsuite: implicitly cleanup-modules, part 2
http://gcc.gnu.org/ml/fortran/2012-04/msg00065.html
Before actua
On Fri, May 11, 2012 at 6:55 PM, H.J. Lu wrote:
>> This patch adds x32 support to boehm-gc. The same patch has been
>> sent to the boehm-gc mailing list. Tested on Linux/x32 and Linux/x86-64.
>> OK for trunk?
>
> This patch has been checked into upstream:
>
> https://github.com/ivmai/bdwgc/commi
To be more specific, does the following match what your envisioned?
1) when multiple streams are specified for dumping, the information
will be dumped to all streams IF the new dumping interfaces are used
(see below). For legacy code, the default dump file will still be used
and the user specified
Hi,
This patch uses + in constraint and match_dup in xbegin_1. OK for
trunk?
Andi, can you provide a run-time testcase patch with proper cpuid check?
Thanks.
H.J.
---
2012-05-11 Andrew Pinski
H.J. Lu
PR target/53315
* config/i386/i386.md (xbegin_1): Use + in
Reviewers: carrot, xur,
Message:
Hi Rong and Carrot, could you take a look at this patch, thanks!
The modification is in upstream (with a minor modification).
Background -
There is the need to agree on the runtime linker path for the various
ARM
hard-float ports that distros are working on rig
On 11.05.2012 12:51, Paolo Bonzini wrote:
> Il 11/05/2012 07:13, Matthias Klose ha scritto:
>> ok, I did clarify it in the existing documentation of MULTIARCH_DIRNAME in
>> fragments.texi, detailing the search order for the files. Should the search
>> order be mentioned in some user documentation a
>>> 2012-05-10 Vladimir Makarov
>>>
>>> PR rtl-optimization/53125
>>> * ira.c (ira): Call find_moveable_pseudos and
>>> move_unallocated_pseudos if only ira_conflicts_p is true.
And the attached patch fixes the reginfo slowness.
The reginfo pass was doing:
for each
Now with patch
On Fri, May 11, 2012 at 8:42 PM, Steven Bosscher wrote:
2012-05-10 Vladimir Makarov
PR rtl-optimization/53125
* ira.c (ira): Call find_moveable_pseudos and
move_unallocated_pseudos if only ira_conflicts_p is true.
>
> And the att
Please add ChangeLog and the trunk patch revision.
http://codereview.appspot.com/6206055/
Hi Tobias,
Hello Thomas,
below a very timely review - your patch is not even a month old
and was never pinged, besides, you have chosen an unlucky day.
(In other words: Sorry for the slow review.)
As a matter of fact, I had become 3/4 convinced that I had already
committed this patch, which i
On Fri, 11 May 2012, Manuel López-Ibáñez wrote:
> >> What cases do we have where a language-independent option enables another
> >> language-independent option only for some front ends? That's the only
> >> case that should need a language-dependent generated function here.
> >
> > Wall enables W
Attached patch applied to trunk.
2012-05-11 François Dumont
PR libstdc++/53263
* include/debug/safe_iterator.h (__gnu_debug::__base): Move...
* include/debug/functions.h: ... Here. Add debug function
overloads to perform checks on normal iterators when possible.
* include/d
Hi!
Please look at the modified patch in the attachment. ChangeLog remains the same.
Tested in android environment(x86_64-*-linux-android), also
bootstrapped on x86_64-unknown-linux-gnu.
I also started regtesting on linux. Is it ok after successfull regtesting?
Thanks,
Igor
On Mon, May 7, 2012
Ping?
On Fri, Apr 27, 2012 at 4:42 PM, Igor Zamyatin wrote:
> On Wed, Apr 25, 2012 at 6:41 PM, Richard Guenther
> wrote:
>> On Wed, Apr 25, 2012 at 4:32 PM, Igor Zamyatin wrote:
>>> On Wed, Apr 25, 2012 at 1:14 PM, Richard Guenther
>>> wrote:
On Wed, Apr 25, 2012 at 10:56 AM, Igor Zamyati
On 2012/05/11 18:48:40, carrot wrote:
Please add ChangeLog and the trunk patch revision.
Thanks.
Trunk patch revision is 186859.
ChangeLog entry added.
Patch started.
diff --git a/gcc/ChangeLog b/gcc/ChangeLog
index 2f85855..588fa04 100644
--- a/gcc/ChangeLog
+++ b/gcc/ChangeLog
@@ -1,
http://codereview.appspot.com/6206055/diff/4001/gcc/ChangeLog
File gcc/ChangeLog (right):
http://codereview.appspot.com/6206055/diff/4001/gcc/ChangeLog#newcode11
gcc/ChangeLog:11:
It seems you also merged in another patch r187012, please mention it.
http://codereview.appspot.com/6206055/diff/40
On 05/10/2012 04:28 PM, Jason Merrill wrote:
Looks good.
Thanks Jason. The below is the idea fully implemented.
The call.c bits are exactly in the form I had in mind a couple of days ago.
The parser.c bits, the ones I actually preliminarily posted, are now a
bit different: I noticed that in t
On 03/02/2012 03:29 PM, Eric Botcazou wrote:
I notice that D.7 seems to suggest that if the nested function is not
inlinable and shared between all instances of the containing function
that we put a normal (non-abstract/concrete) instance of the nested
function inside the abstract function for
On 05/11/2012 06:03 AM, Paolo Carlini wrote:
if (TYPE_PTR_P (type)
&& !TYPE_PTRFN_P (type)
- && !TYPE_PTR_TO_MEMBER_P (type))
+ && !TYPE_PTRMEM_P (type))
The check for !pointer to member is no longer necessary, since they
don't use POINTER_TYPE.
OK with that ch
I find the name "sentry" confusing; I don't see how it applies. Perhaps
"current" instead?
Otherwise, the patch is OK.
Jason
56 matches
Mail list logo
