On Tue, 22 Dec 2015, Martin Sebor wrote:
The attached patch adds handling of dependent arguments to
attribute aligned and attribute vector_size, fixing c++/58109
and 69022 - attribute vector_size ignored with dependent bytes.
In the longer term, Gaby used to suggest that internally we should
On 12/22/2015 10:26 AM, David Malcolm wrote:
Currently, trunk emits this for a bad unary * in C:
bad-dereference.c:10:10: error: invalid type argument of unary ‘*’ (have ‘int’)
return *some_f.x;
^
The following patch fixes the reported location to highlight the
expression that wa
On 12/11/2015 02:11 AM, Ajit Kumar Agarwal wrote:
Mibench/EEMBC benchmarks (Target Microblaze)
Automotive_qsort1(4.03%), Office_ispell(4.29%), Office_stringsearch1(3.5%).
Telecom_adpcm_d( 1.37%), ospfv2_lite(1.35%).
I'm having a real tough time reproducing any of these results. In fact,
I'm
The attached patch rejects invocations of atomic fetch_op intrinsics
on objects of _Bool type as discussed in the context of PR c/68908.
Tested on x86_64.
Martin
gcc/testsuite/ChangeLog:
2015-12-22 Martin Sebor
PR c/68966
* gcc.dg/atomic-fetch-bool.c: New test.
* gcc.dg/sync-fetch-bool.c:
The attached patch adds handling of dependent arguments to
attribute aligned and attribute vector_size, fixing c++/58109
and 69022 - attribute vector_size ignored with dependent bytes.
Tested on x86_64.
Martin
gcc/testsuite/ChangeLog:
2015-12-22 Martin Sebor
PR c++/58109
PR c++/69022
* g+
On 22/12/15 22:58 +0100, Daniel Krügler wrote:
2015-12-22 22:42 GMT+01:00 Jonathan Wakely :
On 21/12/15 12:45 +0100, Daniel Krügler wrote:
2015-12-14 21:48 GMT+01:00 Daniel Krügler :
This is a reimplementation of __is_swappable and
__is_nothrow_swappable according to
http://www.open-std.org
On Tue, 2015-12-22 at 17:36 +, Alan Lawrence wrote:
> On 22/12/15 16:05, Bill Schmidt wrote:
> > On Tue, 2015-12-22 at 15:54 +, Alan Lawrence wrote:
> >> On 21/12/15 13:13, Alan Lawrence wrote:
> >>> This is a respin of previous patch series:
> >>> https://gcc.gnu.org/ml/gcc-patches/2015-10
Jan Hubicka writes:
>> > Hi,
>> > the alias-2.c testcase fails on targets with anchors. The reason is that
>> > the variable itself is anchored while the alias is not and they point to
>> > the
>> > same location. I folllowed the docs of SYMBOL_REF claiming that
>> > SYMBOL_REF_DECL if the sym
Dominik Vogt writes:
> On Fri, Dec 18, 2015 at 09:55:54AM +, Richard Sandiford wrote:
>> Dominik Vogt writes:
>> > The attached patch fixes the handling of LABEL_REF in genrecog and
>> > genpreds.
>> >
>> > The current code assumes that X can have only a mode than PRED (X,
>> > MODE) if X is
2015-12-22 22:42 GMT+01:00 Jonathan Wakely :
> On 21/12/15 12:45 +0100, Daniel Krügler wrote:
>>
>> 2015-12-14 21:48 GMT+01:00 Daniel Krügler :
>>>
>>> This is a reimplementation of __is_swappable and
>>> __is_nothrow_swappable according to
>>>
>>> http://www.open-std.org/jtc1/sc22/wg21/docs/papers
This is something Richi pointed out. I didn't think it was a big deal
and didn't want to unnecessarily complicate the code like I did for jump
threading to avoid a little bit of block copying.
As it turns out avoiding the copying in path splitting is trivial, so
trivial we actually get to r
On 21/12/15 12:45 +0100, Daniel Krügler wrote:
2015-12-14 21:48 GMT+01:00 Daniel Krügler :
This is a reimplementation of __is_swappable and
__is_nothrow_swappable according to
http://www.open-std.org/jtc1/sc22/wg21/docs/papers/2015/n4511.html
and also adds a missing usage of __is_nothrow_swapp
An earlier patch of mine to prevent confusing array length deduction
with a PMF constant was over-zealous; we should allow deduction from an
already-deduced array type.
Tested x86_64-pc-linux-gnu, applying to trunk, 5 and 4.9.
commit 90ff39cf3d376c2359e6ecf9dc187907682a5f2a
Author: Jason Merril
A template can declare a class template or a variable template, but not
both at once. We were failing to diagnose this.
Tested x86_64-pc-linux-gnu, applying to trunk.
commit 316f3a84bf6f512e09b3d3aa72a1600f4cba0a75
Author: Jason Merrill
Date: Mon Dec 21 13:46:09 2015 -0500
PR c++/67257
We can't assume that a RECORD_TYPE has TYPE_LANG_SPECIFIC anymore.
Tested x86_64-pc-linux-gnu, applying to trunk and 5.
commit dafb66bbcf0fd6ef503e3573b2b315230d054c0e
Author: Jason Merrill
Date: Mon Dec 21 12:31:22 2015 -0500
PR c++/67339
* parser.c (cp_parser_elaborated_type_speci
On December 22, 2015 4:24:40 PM GMT+01:00, Jakub Jelinek
wrote:
>Hi!
>
>As the testcase shows, we have some regressions in constexpr folding in
>GCC
>5+.
>One issue is that fold_binary_loc attempts to swap operands of
>comparison
>operators only if one of them is constant, most of the optimizatio
The must-not-throw wrapper for protect_cleanup_actions gets in
the way of being able to access __builtin_eh_pointer without
confusion as the identit of the exception to which we are
referring (b_eh_p has no usable argument up to this point).
Since EH_ELSE never comes from user derived code, let's
The problem in this PR is that when we duplicate blocks, we fail
to duplicate eh regions.
Most of the places that we want to do block duplication, that's fine.
We've already identified that we're operating on a SESE region, or
suchlike, and so EH doesn't apply -- if EH were present it wouldn't
be
Hi all,
the commit for PR68617 broke boostrap on armv6*-*-freebsd*.
We still have unaligned_access = 0 on armv6 here on FreeBSD.
The commit from the above PR overrides my SUBTARGET_OVERRIDE_OPTIONS I
called in arm_option_override. And it sets the unaligned_access to 1.
The attached patch fix
On Tue, Dec 22, 2015 at 2:30 PM, Peter Bergner wrote:
> On Tue, 2015-12-22 at 13:36 -0500, David Edelsohn wrote:
>> On Tue, Dec 22, 2015 at 12:13 PM, Peter Bergner wrote:
>> Ultimately rs6000_file_start ".machine" directive support should be
>> strengthened to handle -mcpu=.
>
> I'm not sure the
On Tue, 2015-12-22 at 13:36 -0500, David Edelsohn wrote:
> On Tue, Dec 22, 2015 at 12:13 PM, Peter Bergner wrote:
> AIX does not support PPC64LE, so the aix53.h, aix61.h and aix71.h
> changes should not be applied.
Heh, I should know better. Thanks for catching that.
> Ultimately rs6000_file_s
Hi all,
The attached patch fixes a problem you get if you build dynamic binaries
for a 32-bit powerpc target on a 64-bit powerpc host.
At the time I did this port I didn't fully understand all the scenarios
you might run into.
The issue is this, on all FreeBSD archs the interpreter is ld-el
On 12/22/2015 04:24 AM, Richard Earnshaw (lists) wrote:
The bigger problem here is that branch costs are a property of a
specific CPU, not the target architecture. So deciding whether or not
we should skip this test (and perhaps others like it) is impossible
given that we can't know what the d
On Tue, Dec 22, 2015 at 12:13 PM, Peter Bergner wrote:
> Currently, -mcpu=powerpc64le correctly sets the TARGET_* flags for an
> LE compile, meaning it mimics a -mcpu=power8 compile, but it doesn't
> pass the correct -mpower8/-mpwr8 option to the assembler, so we die
> with lots of assembler error
Currently, -mcpu=powerpc64le correctly sets the TARGET_* flags for an
LE compile, meaning it mimics a -mcpu=power8 compile, but it doesn't
pass the correct -mpower8/-mpwr8 option to the assembler, so we die
with lots of assembler errors on POWER8 instructions. This patch
fixes things so we pass th
On 22/12/15 16:05, Bill Schmidt wrote:
On Tue, 2015-12-22 at 15:54 +, Alan Lawrence wrote:
On 21/12/15 13:13, Alan Lawrence wrote:
This is a respin of previous patch series:
https://gcc.gnu.org/ml/gcc-patches/2015-10/msg03271.html
Minus three of the smaller patches having already been commi
On Tue, Dec 22, 2015 at 12:26:29PM -0500, David Malcolm wrote:
> Currently, trunk emits this for a bad unary * in C:
>
> bad-dereference.c:10:10: error: invalid type argument of unary ‘*’ (have
> ‘int’)
>return *some_f.x;
> ^
>
> The following patch fixes the reported location to h
Dear Dominique,
This will have to wait until after the 28th. For reasons that I cannot
determine, I don't seem able to tunnel through to gnu central. Given
the imminent arrival of guests for the holiday, I do not feel
motivated to investigate further :-)
Happy holidays to one and all!
Paul
On 2
We are emitting a bogus warning for the code
do foo (0); while (flagA);
^--- NEXT
^ BODY
^--- GUARD
In general I don't think we can get any useful indentation warning out
of a do-while statement, so this patch makes it so that we just skip
them.
Is
The attached patch eliminates the boehm-gc and associated libjava
test suite failures on darwin15 due to the recompilation of the system
libunwind.dylib with Apple clang 7.0. The new optimizations in Apple
Clang 7.0 introduced by...
http://llvm.org/viewvc/llvm-project?view=revision&revision=226
Hi Kyrill,
Thanks for the reply, I think this latest patch addresses your
comments. I have added a test for the scalar forms of fmax/fmin.
Is this ok now?
Tested:
arm-none-eabi: no regressions
Cheers,
Dave.
ChangeLog:
2015-12-22 David Sherwood
gcc/
* config/arm/iterators.md (
Currently, trunk emits this for a bad unary * in C:
bad-dereference.c:10:10: error: invalid type argument of unary ‘*’ (have ‘int’)
return *some_f.x;
^
The following patch fixes the reported location to highlight the
expression that was attempted to be dereferenced:
bad-dereference.
On Tue, 2015-12-22 at 16:00 +, Alan Lawrence wrote:
> On 21/12/15 15:33, Bill Schmidt wrote:
> >>
> >> Not on a stage1 compiler - check_p8vector_hw_available itself requires
> >> being
> >> able to run executables - I'll check on gcc112. However, both look like
> >> they're
> >> really about
On Tue, 2015-12-22 at 15:54 +, Alan Lawrence wrote:
> On 21/12/15 13:13, Alan Lawrence wrote:
> > This is a respin of previous patch series:
> > https://gcc.gnu.org/ml/gcc-patches/2015-10/msg03271.html
> > Minus three of the smaller patches having already been committed; with the
> > updated ve
On 21/12/15 15:33, Bill Schmidt wrote:
Not on a stage1 compiler - check_p8vector_hw_available itself requires being
able to run executables - I'll check on gcc112. However, both look like they're
really about the host (ability to execute an asm instruction), not the target
(/ability for gcc to o
On Thu, Dec 17, 2015 at 9:33 PM, H.J. Lu wrote:
> On Thu, Dec 17, 2015 at 8:49 AM, H.J. Lu wrote:
>> Since Pmode is 64-bit with -maddress-mode=long for x32, indirect call
>> via GOT slot doesn't need zero_extend. This patch limits *call_got_x32
>> and *call_value_got_x32 patterns to 32-bit Pmode
On 21/12/15 13:13, Alan Lawrence wrote:
This is a respin of previous patch series:
https://gcc.gnu.org/ml/gcc-patches/2015-10/msg03271.html
Minus three of the smaller patches having already been committed; with the
updated version of the main patch to SRA; and the patches to DOM reworked
to avoid
Hello Uroš,
I (hopefully fixed all of inputs, thanks!
Updated patch for i386.md in the bottom,
rest patch is the same.
Bootstrap in progress. New tests pass.
Is it ok for trunk if bootstrap will pass?
On 20 Dec 11:56, Uros Bizjak wrote:
> > +(define_expand "rdpkru"
> > + [(set (match_operand:S
On Mon, 21 Dec 2015, Marek Polacek wrote:
> Uuugh, sorry about that. Fixed in this version and I've added a run-time
> test to verify this issue. In addition to that, I've also added a new test
> that checks the size of lhs_type -- so that we know we can use some _N variant
> of the atomic fet
On 22/12/15 15:15, Alan Lawrence wrote:
On 16/12/15 09:31, Kyrill Tkachov wrote:
Hi David,
On 16/12/15 08:53, David Sherwood wrote:
Hi,
Here is the last patch of the fmin/fmax change, which adds the optabs
to the arm backend.
Tested:
arm-none-eabi: no regressions
Good to go?
David Sherwoo
Hi!
As the testcase shows, we have some regressions in constexpr folding in GCC
5+.
One issue is that fold_binary_loc attempts to swap operands of comparison
operators only if one of them is constant, most of the optimizations in
there are written for constant in arg1 and therefore ok, but the two
On 21/12/15 11:58, Bilyan Borisov wrote:
This patch implements the vmaxnmQ_ST and vminnmQ_ST intrinsincs. It also
implements the __ARM_FEATURE_NUMERIC_MAXMIN macro, which is defined when
__ARM_ARCH >= 8, and which enables the intrinsincs.
Tested on arm-none-eabi, armeb-none-eabi, arm-none-linux-
On 16/12/15 09:31, Kyrill Tkachov wrote:
Hi David,
On 16/12/15 08:53, David Sherwood wrote:
Hi,
Here is the last patch of the fmin/fmax change, which adds the optabs
to the arm backend.
Tested:
arm-none-eabi: no regressions
Good to go?
David Sherwood.
ChangeLog:
2015-12-08 David Sherwood
On Tue, 22 Dec 2015, Marek Polacek wrote:
> This patch adds a warning (enabled by default) to warn about accessing
> elements
> of atomic structures or unions, which is undefined behavior according to the C
> standard. I didn't make this a translation-time error because it's unclear if
> it shou
This patch adds a warning (enabled by default) to warn about accessing elements
of atomic structures or unions, which is undefined behavior according to the C
standard. I didn't make this a translation-time error because it's unclear if
it should be so.
Bootstrapped/regtested on x86_64-linux and
Jan (Beulich) ran into this, and indeed I could not find it
documented. So I added it. ;-)
Applied.
Gerald
--- svnwrite.html.orig 2015-04-06 20:15:18.0 +0800
+++ svnwrite.html 2015-12-22 17:07:33.636074381 +0800
@@ -420,6 +420,10 @@
ssh username@gcc.gnu.org email mynewaddr...@ex
On Thu, 10 Dec 2015, Tobias Burnus wrote:
> PS: Talking about the release notes, my feeling is that both the wiki and
> the release notes miss some changes, but I have to admit that I am really
> out of sync. It currently only lists Submodules at the Wiki,
>https://gcc.gnu.org/wiki/GFortran/New
Hi Andre,
On Tue, 17 Nov 2015, Andre Vieira wrote:
> This series is aimed at backporting algorithmic optimizations and a
> change to a test it affects from trunk to the embedded-5-branch.
I do not see such a branch documented in https://gcc.gnu.org/svn.html ?
Mind adding it there? (If you don'
I just found a related patch again that we have in NetBSD's gcc that
allows fixing __FILE__ references.
See
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=47047
No progress since 2010 in getting this included though :(
Thomas
On Thu, Dec 10, 2015 at 12:36:18PM -0500, Daniel Kahn Gillmor wrote:
>
On 11/12/15 21:43 +, Jonathan Wakely wrote:
Fix std::invoke support for reference_wrappers
PR libstdc++/59768
* include/std/functional (_Unwrap, __invfwd): Define.
(__invoke_impl): Remove reference_wrapper overloads and use __invfwd.
* include/std/type_trai
Hello,
This is part of a set of changes we have had in-house for a while
to improve the support of shared libraries on VxWorks.
Today, the port links with libc_internal ("a superset of libgcc.a
that we want to use" says comment) for RTPs, static or dynamic.
Shared RTPs on VxWorks are linked with
On 21/12/15 19:38, Jeff Law wrote:
> On 12/18/2015 02:55 AM, James Greenhalgh wrote:
>> This is a multi-part message in MIME format.
>> --2.2.0.1.gd394abb.dirty
>> Content-Type: text/plain; charset=UTF-8; format=fixed
>> Content-Transfer-Encoding: 8bit
>>
>>
>> Hi,
>>
>> PR68232 is a te
On Fri, Dec 18, 2015 at 09:55:54AM +, Richard Sandiford wrote:
> Dominik Vogt writes:
> > The attached patch fixes the handling of LABEL_REF in genrecog and
> > genpreds.
> >
> > The current code assumes that X can have only a mode than PRED (X,
> > MODE) if X is CONST_INT, CONST_DOUBLE or CON
On 17/12/15 14:49, Segher Boessenkool wrote:
Hi Kyrill,
On Tue, Dec 15, 2015 at 05:07:41PM +, Kyrill Tkachov wrote:
As part of the war on conditional compilation here's an #if check on
WORD_REGISTER_OPERATIONS that
seems to have been missed out.
Bootstrapped and tested on arm, aarch64, x8
Hi Guys,
The patch below is a proposed fix for PR 66655. The issue I believe
is not that the ming32 definition of bind_local_p is wrong, but rather
that G++ thinks that it cannot make the decl weak even though
bind_local_p says that it should. The answer is to define
MAKE_DECL_ONE_ONLY
Hi Guys,
The patch below fixes PR 68770 - a warning from valgrind about an
uninitialised value being used in the default_secondary_reload. The
problem turns out to the in copy_costs which creates its own secondary
reload information structure, but it does not initialise all of the
field
Hi Richard,
The patch below is a fix for this problem, although I am not sure if
it is the correct fix. The patch selects the corresponding SFmode
variant of the DFmode vector type when flag_short_doubles is enabled.
Maybe a better patch would be to disallow these builtins when using an
Hi Jeff,
PR 68913 notes that the test gcc.dg/lto/pr61886_0.c test fails on
targets whose C library does not provide a __fread_chk function.
Since the purpose of the test is to show that GCC will correctly
discard the invocation of __fread_chk_warn, we do not need to actually
link
Ping.
On 2015/11/24 6:27 PM, Chung-Lin Tang wrote:
> Hi, this patch reworks some of the way that asynchronous copyouts are
> implemented for OpenACC in libgomp.
>
> Before this patch, we had a somewhat confusing way of implementing this
> by having two refcounts for each mapping: refcount and asy
Hi!
The following testcases ICE because tsubst_decl seems to assume
templates of local VAR_DECLs don't have DECL_LANG_SPECIFIC set for them,
but for GNU TLS vars they do, yet nothing clears DECL_TEMPLATE_INFO
in the DEC_LANG_SPECIFIC copy. The stale DECL_TEMPLATE_INFO then
causes mangling to give
60 matches
Mail list logo
