This is a regression present on the mainline: the compiler aborts on the
renaming of a type conversion of the dereference of an access value, if the
renaming is declared at library level and used in a subprogram.
Fixed thusly, tested on x86_64-suse-linux, applied on the mainline.
2016-03-07 E
On 29/02/16 04:26, Tom de Vries wrote:
On 22-02-16 11:57, Jakub Jelinek wrote:
On Mon, Feb 22, 2016 at 11:54:46AM +0100, Tom de Vries wrote:
Following up on your suggestion to implement this during
gimplification, I
wrote attached patch.
I'll put it through some openacc testing and add testcas
On Mon, Mar 7, 2016 at 12:10 AM, H.J. Lu wrote:
> On Sun, Mar 6, 2016 at 1:39 PM, Uros Bizjak wrote:
>> Hello!
>>
>> Attached patch prevents red-zone with -fPIC for leaf functions on
>> i686. The idea is to mark when set_got is expanded as a call to a pc
>> thunk, and use this flag to prevent red
On Sun, Mar 06, 2016 at 10:39:19PM +0100, Uros Bizjak wrote:
> Attached patch prevents red-zone with -fPIC for leaf functions on
> i686. The idea is to mark when set_got is expanded as a call to a pc
> thunk, and use this flag to prevent red-zone in the current function.
>
> Please note that using
On Mon, Mar 7, 2016 at 9:34 AM, Jakub Jelinek wrote:
> On Sun, Mar 06, 2016 at 10:39:19PM +0100, Uros Bizjak wrote:
>> Attached patch prevents red-zone with -fPIC for leaf functions on
>> i686. The idea is to mark when set_got is expanded as a call to a pc
>> thunk, and use this flag to prevent re
On 7 March 2016 at 05:49, Ramana Radhakrishnan
wrote:
> On Tue, Jan 26, 2016 at 2:42 PM, Christophe Lyon
> wrote:
>> Hi,
>>
>> This is a followup to PR63304.
>>
>> As discussed in bugzilla, this patch disables pcrelative_literal_loads
>> when -mfix-cortex-a53-843419 (or its default configure opti
This isn't a regression but the problem only affects ASIS mode and the fix
cannot do any harm in regular mode. It brings the support of bitpacked array
types on par with that of other array types.
Tested on x86_64-suse-linux, applied on the mainline.
2016-03-07 Eric Botcazou
* gcc
Hi,
While preparing the cleanup of neon-testgen.ml, I'm adding the missing
tests to gcc.target/aarch64/advsimd-intrinsics.
All the *_p64 and *_p128 are currently missing, and I am wondering
what's the best option. I can think of:
1- Update existing tests using #ifdef __ARM_FEATURE_CRYPTO
2- Updat
On 06/03/16 13:58, Christophe Lyon wrote:
Hi,
In commit r233654, Christian introduced a new test: pragma_cpp_fma.
Unfortunately, this test fails when gcc is configured --with-fpu >=
neonvfpv4: __ARM_FEATURE_FMA is still defined after the last
pop_options.
To address this, I propose to simply
On Fri, Mar 4, 2016 at 6:06 PM, Richard Biener
wrote:
>
> On March 4, 2016 5:35:13 PM GMT+01:00, "Bin.Cheng"
> wrote:
> >On Fri, Mar 4, 2016 at 11:57 AM, Richard Biener
> > wrote:
> >> On Fri, Mar 4, 2016 at 12:07 PM, Bin Cheng wrote:
> >>> Hi,
> >>> A address canonicalization interface was int
There have been several PRs that all boil down to the fact that SSE
insns segfault since one of their operands isn't sufficiently aligned.
Initially, it seemed that only Solaris 9/x86 wass affected, and the issue
was fixed on the 4.9 branch (the last supported on Solaris 9) by
enabling -mstackreali
[ was: Re: [PATCH 3/3] Fix ubsan tests by disabling of an optimization. ]
On 10/07/15 22:11, Jeff Law wrote:
On 07/10/2015 02:19 AM, Richard Biener wrote:
But the warning on the "bogus" line will still be warranted, so user
goes and
fixes it.
But when the user gets the "bogus" line, he may lo
On Mon, Mar 07, 2016 at 10:49:14AM +0100, Tom de Vries wrote:
> This patch implements that.
>
> Should the asan internal function (ASAN_CHECK) be handled in the same way,
> as suggested here ( https://gcc.gnu.org/ml/gcc-patches/2015-07/msg00776.html
> )?
>
> OK for stage4 trunk if bootstrap and r
Dear All,
I had promised to get the 5-branch up to date in respect of deferred
character patches after then had been in place on trunk for "a few
weeks". Well, I got pulled away by PR69423 and have only now come back
to the earlier patch.
The attached patch corresponds to trunk revisions 232450 a
On 04/03/16 13:27, Richard Biener wrote:
> I think to make it work with LTO you need to mark it 'Optimization'.
> Also it's about
> arrays so maybe
>
> 'Assume common declarations may be overridden with ones with a larger
> trailing array'
>
> also if we document it here we should eventually docume
On 07/03/16 11:01, Jakub Jelinek wrote:
On Mon, Mar 07, 2016 at 10:49:14AM +0100, Tom de Vries wrote:
This patch implements that.
Should the asan internal function (ASAN_CHECK) be handled in the same way,
as suggested here ( https://gcc.gnu.org/ml/gcc-patches/2015-07/msg00776.html
)?
OK for st
On Mon, Mar 07, 2016 at 12:26:10PM +0100, Tom de Vries wrote:
> OK for stage4 trunk if bootstrap and reg-test succeeds?
Ok, with a minor nits.
> 2016-03-07 Tom de Vries
>
> PR tree-optimization/70116
> * tree-ssa-tail-merge.c (merge_stmts_p): New function, factored out
> of
On 2016.01.19 at 13:11 +0100, Jan Hubicka wrote:
> according to Trevor, the assumption about THIS pointer being non-NULL breaks
> several bigger C++ packages (definitly including Firefox, but I believe
> kdevelop was mentioned, too). This patch makes the feature to be controlable
> by a dedicated
The old gcc.dg/vect/O3-pr36098.c now FAILs on target that support
general permutation. It was added for a miscompile but only tested
that we don't vectorize - now we do, but correctly.
I believe we've recently got a duplicate but proper runtime testcase.
Tested on x86_64-unknown-linux-gnu with
On Mon, Mar 7, 2016 at 10:27 AM, Bin.Cheng wrote:
> On Fri, Mar 4, 2016 at 6:06 PM, Richard Biener
> wrote:
>>
>> On March 4, 2016 5:35:13 PM GMT+01:00, "Bin.Cheng"
>> wrote:
>> >On Fri, Mar 4, 2016 at 11:57 AM, Richard Biener
>> > wrote:
>> >> On Fri, Mar 4, 2016 at 12:07 PM, Bin Cheng wrote:
On Mon, Mar 7, 2016 at 12:35 PM, Markus Trippelsdorf
wrote:
> On 2016.01.19 at 13:11 +0100, Jan Hubicka wrote:
>> according to Trevor, the assumption about THIS pointer being non-NULL breaks
>> several bigger C++ packages (definitly including Firefox, but I believe
>> kdevelop was mentioned, too).
On 07/03/16 12:32, Jakub Jelinek wrote:
On Mon, Mar 07, 2016 at 12:26:10PM +0100, Tom de Vries wrote:
OK for stage4 trunk if bootstrap and reg-test succeeds?
Ok, with a minor nits.
2016-03-07 Tom de Vries
PR tree-optimization/70116
* tree-ssa-tail-merge.c (merge_stmts_p):
On 03/06/2016 07:24 PM, Matthias Klose wrote:
On 05.03.2016 19:28, Richard Henderson wrote:
When I went to apply my symbol versioning patch to upstream,
I discovered that upstream had already bumped their soname to
6.4.0, beyond the bump that I'd applied to gcc to 5.0.0.
So I bumped upstream to
On Mon, Mar 07, 2016 at 12:57:20PM +0100, Richard Biener wrote:
> > Honza, can you please repost the patch. Richard said on IRC that he may
> > reconsider his rejection after all.
> >
> > I've tested the patch on my Gentoo test machine and it fixes segfaults
> > in LLVM, QT, Chromium, Kdevelop.
> >
On Mon, Mar 7, 2016 at 1:03 PM, Jakub Jelinek wrote:
> On Mon, Mar 07, 2016 at 12:57:20PM +0100, Richard Biener wrote:
>> > Honza, can you please repost the patch. Richard said on IRC that he may
>> > reconsider his rejection after all.
>> >
>> > I've tested the patch on my Gentoo test machine and
On 2016.03.07 at 13:03 +0100, Jakub Jelinek wrote:
> On Mon, Mar 07, 2016 at 12:57:20PM +0100, Richard Biener wrote:
> > (no strong opinion about that - but for example the recent -flifetime-dse
> > strengthening fallout is similar)
>
> -flifetime-dse change is different, we default to -flifetime-
On Mon, Mar 7, 2016 at 1:14 PM, Markus Trippelsdorf
wrote:
> On 2016.03.07 at 13:03 +0100, Jakub Jelinek wrote:
>> On Mon, Mar 07, 2016 at 12:57:20PM +0100, Richard Biener wrote:
>> > (no strong opinion about that - but for example the recent -flifetime-dse
>> > strengthening fallout is similar)
>
On 03/05/2016 06:27 AM, Jeff Law wrote:
PR rtl-optimization/69941
* postreload.c (reload_combine_recognize_pattern): Ensure all uses of
the reg share its mode.
testsuite/
PR rtl-optimization/69941
* gcc.dg/torture/pr69941.c: New test.
OK.
For branches as well?
Bernd
On 7 March 2016 at 05:43, Ramana Radhakrishnan
wrote:
> On Wed, Feb 17, 2016 at 10:20 AM, Christophe Lyon
> wrote:
>> On 17 February 2016 at 11:05, Kyrill Tkachov
>> wrote:
>>>
>>> On 17/02/16 10:03, Christophe Lyon wrote:
On 15 February 2016 at 12:32, Kyrill Tkachov
wrote:
>
Hi Kyrill,
> This is missing a second hunk from the patch you attached in the PR that I
> think is necessary
> for this to work (setting to x_flag_omit_frame_pointer)...
Doh! Silly me - there was a snafu restoring the patch after I had reverted it
in order to
check that the pre- and post- patc
On Fri, Mar 04, 2016 at 07:19:56PM +0100, Jakub Jelinek wrote:
> > --- gcc/c-family/c-common.c
> > +++ gcc/c-family/c-common.c
> > @@ -11443,6 +11443,10 @@ resolve_overloaded_builtin (location_t loc, tree
> > function,
> > && orig_code != BUILT_IN_ATOMIC_STORE_N)
> > result = sync_re
On Fri, Mar 04, 2016 at 07:17:46PM +0100, Uros Bizjak wrote:
> Hello!
>
> > This is not a regression but I thought I'd post this anyway. Martin
> > reported
> > that we generate -Wunused-value warnings on the attached testcase, which
> > arguable doesn't make sense. Setting TREE_USED suppresses
Ping.
https://gcc.gnu.org/ml/gcc-patches/2016-02/msg01626.html
Thanks,
Kyrill
On 24/02/16 11:23, Kyrill Tkachov wrote:
Hi all,
This is the GCC 5 backport of
https://gcc.gnu.org/ml/gcc-patches/2016-02/msg01338.html.
The differences are that TARGET_HAVE_LPAE has to be defined in arm.h in a
diff
Ping.
https://gcc.gnu.org/ml/gcc-patches/2016-02/msg01627.html
Thanks,
Kyrill
On 24/02/16 11:23, Kyrill Tkachov wrote:
Hi all,
This is the GCC 4.9 backport of
https://gcc.gnu.org/ml/gcc-patches/2016-02/msg01338.html.
The differences are that TARGET_HAVE_LPAE has to be defined in arm.h in a
di
On 03/06/2016 09:30 AM, Patrick Palka wrote:
On Sun, Mar 6, 2016 at 1:42 AM, Jason Merrill wrote:
On 02/08/2016 12:19 AM, Patrick Palka wrote:
Here, we are calling template_class_depth on a FIELD_DECL corresponding
to a lambda that is used inside variable template. template_class_depth
howev
This fixes the PR by removing an incomplete duplicate implementation.
Bootstrapped and tested on x86_64-unknown-linux-gnu, applied.
Richard
2016-03-07 Richard Biener
PR tree-optimization/70115
* tree-ssa-loop-ivcanon.c (propagate_into_all_uses): Remove.
(propagate_co
On 24/02/16 14:37, Tom de Vries wrote:
On 17/02/16 16:48, Tom de Vries wrote:
On 17/02/16 13:30, Jakub Jelinek wrote:
On Wed, Feb 17, 2016 at 01:02:17PM +0100, Tom de Vries wrote:
Mark offload symbols as global in lto
I'm really not familiar with that part of LTO, so I'm CCing Honza and
Rich
On Fri, Mar 04, 2016 at 06:39:54PM +0100, Bernd Schmidt wrote:
> I've managed to reproduce this, and I think your analysis of the problem is
> correct. So the patch is probably ok from the point of you of "will it
> work". I can probably be convinced to approve it as-is, but I wonder if
> you'd be
On Sat, 2016-03-05 at 23:46 +0100, Bernhard Reutner-Fischer wrote:
[...]
> diff --git a/gcc/fortran/misc.c b/gcc/fortran/misc.c
> index 405bae0..72ed311 100644
> --- a/gcc/fortran/misc.c
> +++ b/gcc/fortran/misc.c
[...]
> @@ -274,3 +275,41 @@ get_c_kind(const char *c_kind_name,teropKind_tki
> nds
Ping.
https://gcc.gnu.org/ml/gcc-patches/2016-03/msg00046.html
Thanks,
Kyrill
On 01/03/16 16:17, Kyrill Tkachov wrote:
Hi all,
For GCC 6 we want to deprecate architecture revisions prior to ARMv4T.
This patch implements this by documenting the deprecation in invoke.texi and
adding
a warning w
Hi,
when testing the experimental hsa branch, where dynamic parallelism is
not disabled and get_hsa_kernel_dispatch_offset is executed quite a
bit more frequently, I have come across hsa_kernel_dispatch_type being
freed by gcc even though it is marked with a GTY flag. The reason is
that the file
On 03/07/2016 05:37 AM, Bernd Schmidt wrote:
On 03/05/2016 06:27 AM, Jeff Law wrote:
PR rtl-optimization/69941
* postreload.c (reload_combine_recognize_pattern): Ensure all
uses of
the reg share its mode.
testsuite/
PR rtl-optimization/69941
* gcc.dg/torture/pr69941.c: New t
This is an updated patch to address BZ69740. When the patch was
originally committed, there was a flurry of regressions related to the
loop optimizers.
Essentially the loop optimizers would change the CFG and we'd
conservatively set LOOPS_NEED_FIXUP. At the end of the given loop
optimizer
Hi,
in order to consolidate things, I have decided to re-post all "hsa
testsuite" patches under this thread. With the patches applied, we do
no not get any spurious failures because of hsa warnings or libgomp
testcases failing because they are run on the host fallback.
Moreover, the first patch a
Hi,
just like with the compiler gomp testsuite, we need to add -Wno-hsa to
options when compiling libgomp testcases in order not to have "excess
errors" failures when HSA is enabled. There are quite many of such
testcases on the trunk because I have disabled the dynamic parallelism
way of executi
Hi,
the patch below adds a DejaGNU effective target predicate (is that the
correct dejagnu term?) offload_hsa so that selected tests can be run
only if the hsa offloading is enabled. I hope it is fairly standard
stuff. Additionally, it adds one C/C++ and one Fortran testsuite to
check that gridi
Hi,
we would like a place to have some HSA-specific tests, which would
only run not only when HSA is enabled at configuration time but also
when HSA hardware is present and used for offloading.
I have proposed the first version of this patch as
https://gcc.gnu.org/ml/gcc-patches/2016-02/msg01817.
Hi,
this patch avoids run-time failures in libgomp testsuite that
curtrently happen when HSA offloading is actually used. All of these
tests require the offload_device effective target which the patch
changes to offload_device_nonshared_as one.
For some tests, such as libgomp.c/examples-4/device
Hi,
as Jakub requested, this patch deals with HSA "excess errors" in the
gomp compiler testsuite by passing -Wno-hsa to all of them. After
discussing this in the thread about similar libgomp tests[1] (which
are however handled differently), Jakub expressed preference for
passing the option in def
On Mon, Feb 29, 2016 at 04:55:44PM +0100, Martin Jambor wrote:
> the patch below adds a DejaGNU effective target predicate (is that the
> correct dejagnu term?) offload_hsa so that selected tests can be run
> only if the hsa offloading is enabled. I hope it is fairly standard
> stuff. Additionall
On Mon, Feb 29, 2016 at 05:02:34PM +0100, Martin Jambor wrote:
>
> [1] https://gcc.gnu.org/ml/gcc-patches/2016-03/msg00381.html
>
>
> 2016-03-04 Martin Jambor
>
> * c-c++-common/gomp/clauses-1.c: Remove dg-options.
> * c-c++-common/gomp/if-1.c: Likewise.
> * c-c++-common/go
On Mon, Feb 29, 2016 at 05:22:36PM +0100, Martin Jambor wrote:
> 2016-03-04 Martin Jambor
>
> * testsuite/lib/libgomp.exp (libgomp_init): Append -Wno-hsa to
> ALWAYS_CFLAGS.
Ok.
Jakub
On Mon, Feb 29, 2016 at 05:30:05PM +0100, Martin Jambor wrote:
> Tested both with and without HSA (enabled or present). OK for trunk?
Ok. These tests are from the Examples 4.0.x document, so we don't want to
diverge too much on them.
> 2016-02-12 Martin Jambor
>
> libgomp/
> * testsui
On Mon, Feb 29, 2016 at 05:31:53PM +0100, Martin Jambor wrote:
> Is the patch OK for trunk?
Ok, thanks.
> 2016-03-03 Martin Jambor
>
> * testsuite/lib/libgomp.exp
> (check_effective_target_hsa_offloading_selected_nocache): New.
> (check_effective_target_hsa_offloading_selecte
Bernd Schmidt writes:
> On 03/02/2016 10:53 PM, Vladimir Makarov wrote:
>>> 2. update_costs_from_allocno records a cost update not just for the
>>> initial allocno, but for each of the visited ones. I can sort of see
>>> an argument for doing that (let's say if you assign an allocno in the
>>> mid
Hi!
What the STV pass does is clearly wrong, it meant to grab a sequence
of instructions out of a sequence and emit it somewhere else, but
calls end_sequence too late, so the sequence bookkepping vars get
corrupted.
Fixed thusly, bootstrapped/regtested on x86_64-linux and i686-linux, ok for
trunk
On 03/06/2016 10:18 AM, Paul Richard Thomas wrote:
> Dear All,
>
> These are two rather trivial modifications to permit, 'module' to
> appear at any position in the list of prefixes in the procedure
> declaration and to allow module procedures to appear within a module
> contains section. I was ra
Hi,
As discussed in LKML:
http://lists.infradead.org/pipermail/linux-arm-kernel/2015-July/355996.html,
the cost of changing a cache line
from shared to exclusive state can be significant on aarch64 cores,
especially when this is triggered by an exclusive store, since it may
result i
On Mon, Mar 7, 2016 at 7:27 PM, Yangfei (Felix) wrote:
> Hi,
>
> As discussed in LKML:
> http://lists.infradead.org/pipermail/linux-arm-kernel/2015-July/355996.html,
> the cost of changing a cache line
> from shared to exclusive state can be significant on aarch64 cores,
> especially wh
> On Mon, Mar 7, 2016 at 7:27 PM, Yangfei (Felix) wrote:
> > Hi,
> >
> > As discussed in LKML:
> http://lists.infradead.org/pipermail/linux-arm-kernel/2015-July/355996.html,
> the
> cost of changing a cache line
> > from shared to exclusive state can be significant on aarch64 cores,
> esp
On Mon, Mar 7, 2016 at 8:12 PM, Yangfei (Felix) wrote:
>> On Mon, Mar 7, 2016 at 7:27 PM, Yangfei (Felix)
>> wrote:
>> > Hi,
>> >
>> > As discussed in LKML:
>> http://lists.infradead.org/pipermail/linux-arm-kernel/2015-July/355996.html,
>> the
>> cost of changing a cache line
>> > from
On Tue, Mar 8, 2016 at 12:41 AM, Jakub Jelinek wrote:
> Hi!
>
> What the STV pass does is clearly wrong, it meant to grab a sequence
> of instructions out of a sequence and emit it somewhere else, but
> calls end_sequence too late, so the sequence bookkepping vars get
> corrupted.
>
> Fixed thusly
62 matches
Mail list logo
