Hi,
this patch fixes the last failed fortran test case on arm-linux-gnueabihf:
select_type_4.f90, which did always fail at -O2 and above.
It was caused by a strict aliasing violation, when passing a value of the type
"class(x),pointer" to a formal procedure parameter of the type
"class(x),target
Hi,
while looking into -ftime-report, I noticed that ggc can take up to 10% of WPA
memory
while it does almost nothing: it is run just after streaming that explicitly
frees memory that becomes unreachable. The first GGC run usually saves at
most 1% of memory and then it is never run again.
I bel
This patch adds initial set of tests and dg infrastructure for
LeakSanitizer runtime.
Are you sure we need testsuite for something that doesn't have a GCC code
generation counterpart at all?
That's a valid point. We want this in GCC because
(1) support for cross-compilation and cross-testing i
Hi,
this one liner should save some WPA streaming memory by throwing out the
inline info that is no longer needed after partitioning.
Bootstrapped/regtested x86_64-linux, tested with Firefox, will commit
it tomorrow.
* lto/lto.c: Include ipa-inline.h
(do_whole_program_analysis): Fr
Hi,
while looking into firefox profiles, I noticed that we miss devirtualizations
to comdat symbols, because we manage to get different profile_id in each
unit. This is easily fixed by the following patch that makes profiled_id
to by crc32 of the symbol name in this case.
Bootstrapped/regtested x
The following patch fixes
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=60769
The patch was bootstrapped and tested on x86/x86-64.
Committed as rev. 209285.
2014-04-10 Vladimir Makarov
PR rtl-optimization/60769
* lra-constraints.c (simplify_operand_subreg): Force reload
Hans-Peter Nilsson: Did you have any follow up here? Or did this
response (below) address all your concerns?
I have made some minor corrections since this post:
- "make dvi" now builds my text without errors (can't say this for the
rest of extend.texi).
- All (instead of nearly all) the asm-r
Looks good to me.
David
On Thu, Apr 10, 2014 at 2:04 PM, Teresa Johnson wrote:
> This patch fixes an issue where self-recursive calls were missed
> during ipa inlining in LIPO compiles, since the resolved nodes were
> not checked. When a self-recursive call was inlined incorrectly,
> ipa_inline
This patch fixes an issue where self-recursive calls were missed
during ipa inlining in LIPO compiles, since the resolved nodes were
not checked. When a self-recursive call was inlined incorrectly,
ipa_inline redirected the edge to the resolved node, leading to
a cycle in the cgraph that wasn't exp
Richard Sandiford writes:
> Robert Suchanek writes:
>> I'm not particularly happy with this either. This was an attempt to fix an
>> ICE for
>> the following RTL (gcc.dg/torture/asm-subreg-1.c compiled with -mips32r2
>> -mips16 -O1):
>>
>> (insn 9 8 0 2 (asm_operands/v ("") ("") 0 [
>>
Hi,
On Thu, 10 Apr 2014 17:15:16, Tobias Burnus wrote:
>
> Hi Bernd,
>
> On Thu, Apr 10, 2014 at 11:54:52AM +0200, Tobias Burnus wrote:
>> No re-reading the orginal patch, I think it is okay. Thanks!
>> http://gcc.gnu.org/ml/gcc-patches/2014-04/msg00376.html
>
>
> Thanks for the commit. However, c
On Thu, Apr 10, 2014 at 07:55:26PM +, Zamyatin, Igor wrote:
> So below is the patch for 4.9 trunk.
Ok, thanks.
>
> Thanks,
> Igor
>
> gcc/c/Changelog:
>
> 2014-04-10 Igor Zamyatin
>
> PR middle-end/60469
> * c-array-notation.c (fix_builtin_array_notation_fn): Use
> cr
Patch ping. This is not a regression as I can reproduce it on GCC 4.8.*
but it is an ICE on legal C code. Any chance of having it approved for
4.9?
Steve Ellcey
sell...@mips.com
On Thu, 2014-03-20 at 09:48 -0700, Steve Ellcey wrote:
> This patch fixes pr60556, a GCC ICE. The problem is in con
> -Original Message-
> From: Jakub Jelinek [mailto:ja...@redhat.com]
> Sent: Thursday, April 10, 2014 10:21 PM
> To: Zamyatin, Igor
> Cc: GCC Patches (gcc-patches@gcc.gnu.org); Iyer, Balaji V
> Subject: Re: [PATCH, PR60469, Cilk+] Fix ICE of using Cilk_spawn and Array
> Notation together
>
On 04/10/2014 09:10 AM, Jakub Jelinek wrote:
> 2014-04-10 Jakub Jelinek
>
> PR rtl-optimization/60663
> * cse.c (cse_insn): Set src_volatile on ASM_OPERANDS in
> PARALLEL.
>
> * gcc.target/arm/pr60663.c: New test.
Ok if it passes.
But you're right that ARM backend nee
Tobias Burnus wrote:
Missing is adding the intrinsic in resolve.c – and converting it into
code in trans-intrinsic.c. I have a draft patch for it, but I still
need to fix something and clean up the patch.
I have now also committed a patch, which moves the existing code higher
up in the file;
Hi!
Honza has acked this patch on IRC, so I've committed it to trunk after
bootstrap/regtest on x86_64-linux and i686-linux.
2014-04-10 Jan Hubicka
Jakub Jelinek
PR lto/60567
* ipa.c (function_and_variable_visibility): Copy forced_by_abi flag from
decl_no
DJ Delorie wrote:
>This is for the host libiberty only, and only when gcc is configured
>a certain way. The intent is to have libiberty that is going to be
>linked into all the build and host tools instrumented, so that we
>actually catch bugs in libiberty or bugs in host/build tools calling
>li
On Thu, Apr 10, 2014 at 06:16:22PM +, Zamyatin, Igor wrote:
> Fixed patch is below.
>
> Just out of curiosity - why do you think this patch is not for 4.9?
The one with just create_tmp_var (integer_type_node, NULL); is
for 4.9.
For the iterator var changes, I guess one needs to analyze all th
Fixed patch is below.
Just out of curiosity - why do you think this patch is not for 4.9?
Thanks,
Igor
gcc/c/Changelog:
2014-04-10 Igor Zamyatin
PR middle-end/60469
* c-array-notation.c (fix_builtin_array_notation_fn): Use
create_tmp_var instead build_decl for creat
> However it would be nice to be assured that the gcc change is ok in
> principle first.
The DWARF bits are fine with me.
-cary
On Thu, Apr 10, 2014 at 05:19:33PM +, Zamyatin, Igor wrote:
> @@ -282,8 +283,7 @@ fix_builtin_array_notation_fn (tree an_builtin_fn, tree
> *new_var)
>
>for (ii = 0; ii < rank; ii++)
> {
> - an_loop_info[ii].var = build_decl (location, VAR_DECL, NULL_TREE,
> -
"Zamyatin, Igor" writes:
> diff --git a/gcc/testsuite/c-c++-common/cilk-plus/CK/pr60469.c
> b/gcc/testsuite/c-c++-common/cilk-plus/CK/pr60469.c
> new file mode 100644
> index 000..61e8014
> --- /dev/null
> +++ b/gcc/testsuite/c-c++-common/cilk-plus/CK/pr60469.c
> @@ -0,0 +1,15 @@
> +/* middl
Hi!
This patches fixes the ICE when array notation is used along with Cilk_spawn.
We need to expand AN's loop using other routine for creating temporary
variables.
Bootstrapped/regtested on x86_64.
Also it introduces no new failures and fixes couple of ICE for CilkPlus
Conformance suite which
This patch adds the support for coindexed arrays in expressions (or the
RHS of assignments) to the single-image implementation of the library.
Additionally, it adds the required function definitions to the compiler.
Missing is adding the intrinsic in resolve.c – and converting it into
code in
On Tue, Apr 01, 2014 at 11:41:12AM +0800, Zhenqiang Chen wrote:
> Ping?
>
> Bootstrap and no make check regression on X86-64.
>
> Bootstrap on ARM. In ARM regression test, some new PASS and FAIL of
> debug info check for gcc.dg/guality/pr36728-1.c and
> gcc.dg/guality/pr36728-2.c since register a
On 04/10/2014 09:19 AM, Jakub Jelinek wrote:
Thus, I've bootstrapped/regtested this version on x86_64-linux and
i686-linux, is this ok to everybody for now?
OK.
Jason
> But ubsan is a new feature in 4.9, and it is a bootstrap failure
> with that feature.
I will leave it up to the release manager to decide if they want this
non-regression patch applied before the branch, then.
> This is for the host libiberty only, and only when gcc is configured
> a certain w
> "Mark" == Mark Wielaard writes:
Mark> Add a new lang-hook that provides the underlying base type of an
Mark> ENUMERAL_TYPE. Including implementations for C and C++. Use this
Mark> enum_underlying_base_type lang-hook in dwarf2out.c to add a DW_AT_type
Mark> base type reference to a DW_TAG_en
On April 10, 2014 3:19:42 PM CEST, Jakub Jelinek wrote:
>On Wed, Apr 09, 2014 at 02:10:04PM -0400, Jason Merrill wrote:
>> On 04/09/2014 04:21 AM, Richard Biener wrote:
>> >>The names of the in-charge and not-in-charge constructor clones are
>> >>complete_ctor_identifier and base_ctor_identifier (
Hi Bernd,
On Thu, Apr 10, 2014 at 11:54:52AM +0200, Tobias Burnus wrote:
> No re-reading the orginal patch, I think it is okay. Thanks!
> http://gcc.gnu.org/ml/gcc-patches/2014-04/msg00376.html
Thanks for the commit. However, can you also fix the ChangeLog?
You used gcc/ChangeLog but the proper
Hi Bernd,
Post stations are not that 90's,and they charge.It took me $30 to
post the file to USA.It's so inconvenient and expensive that I can't
send a scaned version.
--
Regards
lin zuojian
On Wed, Apr 09, 2014 at 06:47:56PM +0800, lin zuojian wrote:
> Hi Bernd,
> I am asking th
On Thu, Apr 10, 2014 at 02:23:16PM +, Zamyatin, Igor wrote:
> 2014-04-10 Igor Zamyatin
>
> PR c++/60189
> * parser.c (cp_parser_postfix_expression): Make sure only
> semicolon can go after Cilk_sync.
>
> gcc/testsuite/ChangeLog:
>
> 2014-04-10 Igor Zamyatin
>
> PR c++/60189
> * c-c++-
Resending with correct Changelog
Is it OK?
Thanks,
Igor
gcc/cp/ChangeLog:
2014-04-10 Igor Zamyatin
PR c++/60189
* parser.c (cp_parser_postfix_expression): Make sure only
semicolon can go after Cilk_sync.
gcc/testsuite/ChangeLog:
2014-04-10 Igor Zamyatin
PR c++/60189
* c-c++-common/cil
On Thu, Apr 10, 2014 at 02:20:42PM +, Zamyatin, Igor wrote:
> Resending with correct Changelog.
>
> Is it OK?
>
> Thanks,
> Igor
>
> gcc/c-family/ChangeLog:
>
> 2014-04-10 Igor Zamyatin
>
> PR middle-end/60467
> * cilk.c (cilk_set_spawn_marker): Remove FUNCTION_DECL
>
Resending with correct Changelog.
Is it OK?
Thanks,
Igor
gcc/c-family/ChangeLog:
2014-04-10 Igor Zamyatin
PR middle-end/60467
* cilk.c (cilk_set_spawn_marker): Remove FUNCTION_DECL
as possible argument for Cilk_spawn.
gcc/testsuite/ChangeLog:
2014-04-10 Igor Zamyatin
Hi!
This patch filters out another incorrect usage of Cilk_spawn keyword.
Bootstrapped/regtested on x86_64. Ok for trunk?
Thanks,
Igor
gcc/ChangeLog:
2014-04-10 Igor Zamyatin
PR middle-end/60467
* c-family/cilk.c (cilk_set_spawn_marker): Remove FUNCTION_DECL
as possible
"Zamyatin, Igor" writes:
> gcc/ChangeLog:
>
> 2014-04-10 Igor Zamyatin
>
> PR middle-end/60467
> * c-family/cilk.c (cilk_set_spawn_marker): Remove FUNCTION_DECL
> as possible argument for Cilk_spawn.
Both c-family (and cp in your other patch) have their own ChangeLog
files.
Hi!
This fixes ICE on inappropriate usage of Cilk_sync keyword.
Bootstrapped/regtested on x86_64. Ok for trunk?
Thanks,
Igor
gcc/ChangeLog:
2014-04-10 Igor Zamyatin
PR c++/60189
* cp/parser.c (cp_parser_postfix_expression): Make sure only
semicolon can go after Cilk_syn
On 04/10/2014 09:27 AM, Jakub Jelinek wrote:
The following patch copies that flag in the C++ FE, another alternative
is to do that in cgraph_add_thunk function (apparently only called by
use_thunk, thus practically the same spot, or it can be done in
function_and_variable_visibility.
I think Ho
On Thu, Apr 10, 2014 at 03:23:43PM +0200, Rainer Orth wrote:
> > I see failures from last night on aarch64-none-elf and arm-none-eabi
> > (both bare-metal) configurations even after moving up to dejagnu
> > 1.5.1. If this can't be fixed easily should we consider reverting this
> > patch in the inte
On 09.04.2014 15:12, Ramana Radhakrishnan wrote:
On 04/09/14 09:07, Alexey Merzlyakov wrote:
On 04.04.2014 14:44, Alexey Merzlyakov wrote:
Hi all,
Here is a patch, that fixes infinite backtraces in __cxa_end_cleanup().
The Bugzilla entry for
this:http://gcc.gnu.org/bugzilla/show_bug.cgi?id=607
Hi!
As Honza has determined, the problem on this testcase is that
forced_by_abi flag wasn't propagated from thunk's target to the thunk
(in this case the thunk's target was forced_by_abi = true, while the
thunk mistakenly didn't have that flag set).
The following patch copies that flag in the C++
Ramana Radhakrishnan writes:
> On Tue, Apr 8, 2014 at 6:28 PM, Steve Ellcey wrote:
>> On Tue, 2014-04-08 at 10:10 +0200, Dominique Dhumieres wrote:
>>> > richi asked for a testcase for 60731, and since we didn't already
>>> > have support for tests using dlopen, I had to add it.
>>> > Does this
On Wed, Apr 09, 2014 at 02:10:04PM -0400, Jason Merrill wrote:
> On 04/09/2014 04:21 AM, Richard Biener wrote:
> >>The names of the in-charge and not-in-charge constructor clones are
> >>complete_ctor_identifier and base_ctor_identifier (and dtor for
> >>destructors); you could check for those.
> >
On Tue, Apr 8, 2014 at 6:28 PM, Steve Ellcey wrote:
> On Tue, 2014-04-08 at 10:10 +0200, Dominique Dhumieres wrote:
>> > richi asked for a testcase for 60731, and since we didn't already
>> > have support for tests using dlopen, I had to add it.
>> > Does this approach make sense?
>>
>> r209187 ca
On 04/09/2014 05:27 PM, Marc Glisse wrote:
The doc for the "format" attribute says clearly:
"Since non-static C++ methods have an implicit this argument, the
arguments of such methods should be counted from two, not one, when
giving values for string-index and first-to-check."
Ah. That seems
Hi all,
This is the 4.8 version of the patch posted at:
http://gcc.gnu.org/ml/gcc-patches/2014-04/msg00315.html
TARGET_CRYPTO was not defined in 4.8 therefore that hunk is removed.
Ok for the 4.8 branch?
Thanks,
Kyrill
2014-04-10 Kyrylo Tkachov
* config/aarch64/aarch64.h (TARGET_SIMD):
Hi,
the htm-builtins-compile-1.c fails on 31 bit due to compile warnings.
Fixed with the attached patch.
Bye,
-Andreas-
2014-04-10 Andreas Krebbel
* gcc.target/s390/htm-builtins-compile-1.c: Replace long long with
long.
diff --git a/gcc/testsuite/gcc.target/s390/htm-built
I would like to ping the patch:
http://gcc.gnu.org/ml/gcc-patches/2014-03/msg01673.html
It's a syntax issue: It simply makes no sense to have a @menu item
which points to @nodes which are at the same or higher level
than the section in which the @menu is.
Newer makeinfo rightfully complain abo
Hi Bernd,
On Thu, Apr 10, 2014 at 11:20:47AM +0200, Bernd Edlinger wrote:
> I thought about that, too. But it is likely that the pattern for array of
> target class
> "__class_%s_%d_%d" will eventually clash with the non-array target class
> "__class_%s"
which proves that I am too sleep depriev
Hi Tobias,
On Thu, 10 Apr 2014 10:50:24, Tobias Burnus wrote:
>
> Bernd Edlinger wrote:
>> this patch fixes a recently discovered name-clash in gfc_build_class_symbol.
>>
>> Fortunately it is quite easy to fix: just make sure that the class names of
>> target
>> classes end with "_t", and target
Andreas Schwab writes:
> Jakub Jelinek writes:
>
>> I'd say so, perhaps with the exception of Solaris where I'd leave that to
>> the target maintainer to decide.
>
> Done.
I'd leave them in the solaris2.10 baselines: there are no non-TLS
toolchains there, and I'd rather know if TLS support some
Bernd Edlinger wrote:
> this patch fixes a recently discovered name-clash in gfc_build_class_symbol.
>
> Fortunately it is quite easy to fix: just make sure that the class names of
> target
> classes end with "_t", and target array classes end with "[0-9]t".
> This trick makes all names unique aga
On Thu, Apr 10, 2014 at 10:42:28AM +0200, Andreas Schwab wrote:
> Jakub Jelinek writes:
>
> > I'd say so, perhaps with the exception of Solaris where I'd leave that to
> > the target maintainer to decide.
>
> Done.
>
> > The other alternative is to have the TLS symbols included everywhere and a
Jakub Jelinek writes:
> I'd say so, perhaps with the exception of Solaris where I'd leave that to
> the target maintainer to decide.
Done.
> The other alternative is to have the TLS symbols included everywhere and add
> some magic to the abi checking scripts/proglet which would detect the
> emu
4.9 changes for ARM / AArch64. Sorry it's taken me a while to get this
out but better late than never :)
Ok ?
Ramana
--
Ramana Radhakrishnan
Principal Engineer
ARM Ltd.Index: htdocs/gcc-4.9/changes.html
===
RCS file: /cvs/gcc/wwwdo
On April 10, 2014 10:02:29 AM CEST, Jakub Jelinek wrote:
>Hi!
>
>This backport didn't apply cleanly, because 4.8 doesn't have
>build_all_ones_cst function.
>
>So, either we can apply something like the patch below
>(bootstrapped/regtested with the other backports), or as another
>alternative I see
On Thu, Apr 10, 2014 at 10:15:59AM +0200, Andreas Schwab wrote:
> Jakub Jelinek writes:
>
> > On Thu, Apr 10, 2014 at 09:52:35AM +0200, Andreas Schwab wrote:
> >> Tested on ia64-suse-linux and installed as obvious.
> >
> > Please don't add the TLS symbols in there, they aren't included
> > in any
Jakub Jelinek writes:
> On Thu, Apr 10, 2014 at 09:52:35AM +0200, Andreas Schwab wrote:
>> Tested on ia64-suse-linux and installed as obvious.
>
> Please don't add the TLS symbols in there, they aren't included
> in any of the linux baselines
Actually, there are some:
libstdc++-v3/config/abi/po
Hi,
This fixes the tail of PR60655 where there were still issues without
-fdata-sections despite the fix in const_ok_for_output_1. For now, given
the pressure to get 4.9 out of the door, it's probably best to handle
this in the backend and reject constants that cannot be handled by GAS.
For s
Hi!
This backport didn't apply cleanly, because 4.8 doesn't have
build_all_ones_cst function.
So, either we can apply something like the patch below
(bootstrapped/regtested with the other backports), or as another
alternative I see backport the addition of build_minus_one_cst
and build_all_ones_c
Hi,
the hoist-register-pressure testcases currently fail on S/390 since
the rtl hoist pass requires that the expression to be hoisted can be
assigned without clobbering cc. We do not have a 32 bit add which
does not clobber cc.
On 64 bit we might use load address if the operands are DImode. On
Hi!
I've committed several backports patches from trunk to 4.8 branch after
bootstrapping/regtesting them on x86_64-linux and i686-linux.
Jakub
2014-04-10 Jakub Jelinek
Backport from mainline
2014-03-06 Jakub Jelinek
Meador Inge
PR tar
On Thu, Apr 10, 2014 at 09:52:35AM +0200, Andreas Schwab wrote:
> Tested on ia64-suse-linux and installed as obvious.
Please don't add the TLS symbols in there, they aren't included
in any of the linux baselines so that there aren't failures in
--disable-tls configurations or when using too old bi
65 matches
Mail list logo
