The following patch fixes PR66917 - the vectorizer assumes element
alignment even for unaligned accesses, which in case of "packed"
ones is not a valid assumption. This breaks testcases on arm
at least which can do unaligned (but not "completely" unaligned)
loads and stores.
Bootstrapped and tes
On Fri, Jul 31, 2015 at 1:17 PM, Yuri Rumyantsev wrote:
> Hi Richard,
>
> I learned your updated patch for 23825 and it is more general in
> comparison with my.
> I'd like to propose you a compromise - let's consider my patch only
> for force-vectorize outer loop only to allow outer-loop
> vecctor
On Fri, Jul 31, 2015 at 8:27 PM, Jeff Law wrote:
> On 07/31/2015 12:18 PM, Michael Collison wrote:
>>
>> Hi Jeff,
>>
>> Yes I will create a test case. I'm not quite sure what to check for even
>> in the machine dependent test case. It's quite possible for the
>> instructions that are generated to
Hi Michael,
It looks like this patch introduces regressions on armeb in:
gcc.dg/vect/vect-reduc-7.c
gcc.dg/vect/vect-reduc-8.c
gcc.dg/vect/vect-reduc-9.c
See
http://people.linaro.org/~christophe.lyon/cross-validation/gcc/trunk/226476/report-build-info.html
for a bit more details.
Can you have a
On Thu, Jul 30, 2015 at 8:41 PM, H.J. Lu wrote:
> On Tue, Jul 21, 2015 at 02:45:39PM -0700, H.J. Lu wrote:
>> When __builtin_frame_address is used to retrieve the address of the
>> function stack frame, the frame pointer is always kept, which wastes one
>> register and 2 instructions. For x86-32,
On 07/25/2015 09:44 PM, Pierre-Marie de Rodat wrote:
This change makes GCC materialize subprogram renamings in Ada as
imported declarations (DW_TAG_imported_declarations). For instance,
procedure Foo renames Bar;
will output:
DW_TAG_imported_declaration:
DW_AT_name: foo
On 07/20/2015 09:39 AM, Pierre-Marie de Rodat wrote:
This patch fixes the static link description in DWARF to comply with the
specification. In order to do so, it appends a field to all FRAME
objects to hold the frame base address (DW_AT_frame_base) so that the
nested subrograms can directly refe
On Jul 26, 2015, at 11:50 AM, Andi Kleen wrote:
> I've been compiling gcc with tcmalloc to do a similar speedup. It would be
> interesting to compare that to your patch.
I repeated the test with TCMalloc and jemalloc. TCMalloc shows nice results,
though it required some tweaks: this allocator has
On 07/20/2015 02:45 PM, Pierre-Marie de Rodat wrote:
On PowerPC targets with -mlongcall, most subprogram calls are turned
into indirect calls: the call target is read from a register even though
it is compile-time known. This makes it difficult for machine code
static analysis engines to recover
Ian Lance Taylor writes:
> This patch from Chris Manghane fixes the Go frontend error reporting
> for invalid builtin calls, by not losing track of whether the call is
> erroneous. This fixes https://golang.org/issue/11561. Bootstrapped
> and ran Go testsuite on x86_64-unknown-linux-gnu. Commi
On Sun, Aug 2, 2015 at 9:40 PM, Patrick Palka wrote:
> On Sun, Aug 2, 2015 at 2:44 PM, Bernhard Reutner-Fischer
> wrote:
>> On August 2, 2015 8:18:51 PM GMT+02:00, Patrick Palka
>> wrote:
>>
>> No comment on the patch itself, but please
>> s/visted/visited/
>> s/VISIED_PHIS/VISITED_PHIS/
>>
>>
On 24/07/15 09:36, Kyrill Tkachov wrote:
On 21/07/15 17:52, James Greenhalgh wrote:
On Thu, Jul 16, 2015 at 04:21:05PM +0100, Kyrill Tkachov wrote:
Hi all,
This patch implements target pragmas for aarch64.
The pragmas accepted are the same as for target attributes (as required).
In addition p
2015-08-03 5:49 GMT+02:00 Jason Merrill :
> On 07/31/2015 05:54 PM, Kai Tietz wrote:
>>
>> The "STRIP_NOPS-requirement in 'reduced_constant_expression_p'" I could
>> remove, but for one case in constexpr. Without folding we don't do
>> type-sinking/raising.
>
>
> Right.
>
>> So binary/unary operat
On Fri, Jul 24, 2015 at 09:21:46AM +0100, Kyrill Tkachov wrote:
>
> Thanks for the review, here's an updated version.
> In the end, I chose to retain the use alloca (other patches in the series
> are reworked to use it too).
>
> How's this?
A nit or two from code you were moving or that got caug
Since we allow arbitrary IDs in captures the capture array is dense
and thus we don't need to clear it before initializing. We can
also emit more compact source by simply using an initializer.
- tree captures[3] ATTRIBUTE_UNUSED = {};
- captures[0] = o20;
- captures
Could you probably review the patch, please?
2015-07-28 18:42 GMT+03:00 Maxim Blumental :
> Applied the idea with python script alternative. Review, please.
>
> 2015-07-24 17:18 GMT+03:00 David Malcolm :
>> On Fri, 2015-07-24 at 10:01 +0200, Jakub Jelinek wrote:
>>> #!/usr/bin/python
>>> import s
Hi!
On Tue, 28 Jul 2015 10:20:39 +0200, Tom de Vries wrote:
> this patch adds a test-case with a double reduction in an oacc kernels
> region.
>
> In order to get it in the proper shape for parloops to deal with, I
> needed to repeat the pass_lim/pass_copy_prop sequence.
>
> Bootstrapped and
Le 29/07/2015 17:08, Paul Richard Thomas a écrit :
Dear All,
On 24 July 2015 at 10:08, Damian Rouson wrote:
I love this idea and had similar thoughts as well.
:D
Sent from my iPhone
On Jul 24, 2015, at 1:06 AM, Paul Richard Thomas
wrote:
Dear Mikael,
It had crossed my mind also that a
On 28/07/15 11:21, Szabolcs Nagy wrote:
On 24/07/15 12:27, Kyrill Tkachov wrote:
On 24/07/15 12:10, Szabolcs Nagy wrote:
(-a)*b should not be compiled to vnmul a,b with -frounding-math.
Added a new -(a*b) pattern for vnmul and the old one is only
used if !flag_rounding_math. Updated the costs
On Fri, Jul 24, 2015 at 11:43:32AM +0100, Kyrill Tkachov wrote:
>
> On 21/07/15 16:37, James Greenhalgh wrote:
> > On Thu, Jul 16, 2015 at 04:20:59PM +0100, Kyrill Tkachov wrote:
> >> Hi all,
> >>
> >> This patch implements target attribute support via the
> >> TARGET_OPTION_VALID_ATTRIBUTE_P hoo
On Thu, Jul 23, 2015 at 11:17:20AM +0100, Kyrill Tkachov wrote:
> Thanks, I've implemented the suggestions.
> Re-bootstrapped and tested on aarch64.
> How's this?
This is good. OK for trunk.
Thanks,
James
Hi!
On Wed, 29 Jul 2015 19:13:01 -0400, Nathan Sidwell wrote:
> gcc/
> * gimplify.c (oacc_default_clause): Outer scope searching moved to
> omp_notice_variable.
> (omp_notice_variable): For OpenACC search enclosing scopes before
> applying default.
Committed to gomp
On Mon, Aug 3, 2015 at 11:10 AM, Andreas Schwab wrote:
> Ian Lance Taylor writes:
>
>> This patch from Chris Manghane fixes the Go frontend error reporting
>> for invalid builtin calls, by not losing track of whether the call is
>> erroneous. This fixes https://golang.org/issue/11561. Bootstrap
Hi!
On Sat, 1 Aug 2015 20:20:49 -0400, Nathan Sidwell wrote:
> this is the gomp4 committed version.
> --- liboffloadmic/plugin/libgomp-plugin-intelmic.cpp (revision 226462)
> +++ liboffloadmic/plugin/libgomp-plugin-intelmic.cpp (working copy)
> @@ -327,12 +327,26 @@ offload_image (const void *
On Mon, Aug 03, 2015 at 10:36:17AM +0100, Kyrill Tkachov wrote:
> And here is a rebased version to resolve a conflict after Alan's patches went
> in.
>
OK with the nits below fixed.
> 2015-08-03 Kyrylo Tkachov
>
> * config.gcc (aarch64*-*-*): Specify c_target_objs and cxx_target_objs.
On Fri, Jul 24, 2015 at 09:38:34AM +0100, Kyrill Tkachov wrote:
> Thanks, here's an updated version.
>
> 2015-07-24 Kyrylo Tkachov
>
> * config/aarch64/aarch64.c (aarch64_option_valid_attribute_p):
> Initialize simd builtins if TARGET_SIMD.
> * config/aarch64/aarch64-builtins.c
On Fri, Jul 24, 2015 at 09:40:28AM +0100, Kyrill Tkachov wrote:
>
> On 21/07/15 18:14, James Greenhalgh wrote:
> > On Thu, Jul 16, 2015 at 04:21:15PM +0100, Kyrill Tkachov wrote:
> >> Hi all,
> >>
> >> These are the tests for target attributes and pragmas.
> >> I've tried to test for the inlining
Hi!
On Sun, 2 Aug 2015 21:23:30 -0400, Nathan Sidwell wrote:
> I've committed this to gomp4 branch. It expands the acc_on_device builtin
> earlier in the new oacc_xform pass. This will allow more optimization
> earlier on.
Thanks!
> The existing expansion point now only needs to deal with t
Richard Biener wrote:
> On Wed, Jul 29, 2015 at 3:57 PM, Ulrich Weigand wrote:
> > Hello,
> >
> > this patch is a workaround for the problem discussed here:
> > https://gcc.gnu.org/ml/gcc-patches/2015-07/msg01597.html
> >
> > The problem is that the new pool allocator code relies on C++ aliasing
>
I've committed this to gomp4. The ptx backend can now examine the openacc
attribute to determine launch dimensions and figure out whether vector or worker
single neutering is needed.
nathan
2015-08-03 Nathan Sidwell
* config/nvptx/nvptx.c (nvptx_reorg): Check get_oacc_fn_attrib for
launc
On 08/03/15 07:37, Thomas Schwinge wrote:
Hi!
On Sun, 2 Aug 2015 21:23:30 -0400, Nathan Sidwell wrote:
I've committed this to gomp4 branch. It expands the acc_on_device builtin
earlier in the new oacc_xform pass. This will allow more optimization earlier
on.
Thanks!
The existing expansi
Hi!
On Sun, 2 Aug 2015 21:23:30 -0400, Nathan Sidwell wrote:
> I've committed this to gomp4 branch. It expands the acc_on_device builtin
> earlier in the new oacc_xform pass. This will allow more optimization
> earlier on.
As far as I remember, the Fortran front end doesn't know about/doesn'
Hello!
> 2015-07-30 Kyrylo Tkachov
>
> * ifcvt.c (noce_try_store_flag_constants): Make logic of the case
> when diff == STORE_FLAG_VALUE or diff == -STORE_FLAG_VALUE more
> explicit. Prefer to add the flag whenever possible.
> (noce_process_if_block): Try noce_try_store_flag_co
Dear Mikael,
Thanks for your green light!
I have been mulling over the trans-decl part of the patch and having
been wondering if it is necessary. Without optimization, private
entities can be linked to. Given the discussion concerning the
combination of submodules and private entities, I wonder i
Hi!
On Mon, 3 Aug 2015 08:00:36 -0400, Nathan Sidwell
wrote:
> On 08/03/15 07:37, Thomas Schwinge wrote:
> > On Sun, 2 Aug 2015 21:23:30 -0400, Nathan Sidwell wrote:
> >> I've committed this to gomp4 branch. It expands the acc_on_device builtin
> >> earlier in the new oacc_xform pass. This wi
Catherine,
Inline-memcpy-2.c updated to not run with -Os.
Patch rebased off current gcc sources.
Thanks,
Simon
diff --git a/gcc/config/mips/mips.c b/gcc/config/mips/mips.c
index 1733457..627e078 100644
--- a/gcc/config/mips/mips.c
+++ b/gcc/config/mips/mips.c
@@ -7520,12 +7520,22 @@ mips_block
On Fri, Jul 24, 2015 at 11:55:33AM +0100, Kyrill Tkachov wrote:
> Hi all,
>
> This patch implements an aarch64-specific expansion of the signed modulo by a
> power of 2.
> The proposed sequence makes use of the conditional negate instruction CSNEG.
> For a power of N, x % N can be calculated with
On 03/08/15 13:33, Uros Bizjak wrote:
Hello!
2015-07-30 Kyrylo Tkachov
* ifcvt.c (noce_try_store_flag_constants): Make logic of the case
when diff == STORE_FLAG_VALUE or diff == -STORE_FLAG_VALUE more
explicit. Prefer to add the flag whenever possible.
(noce_process_if
On 08/02/15 23:44, Jason Merrill wrote:
It seems to me that DR 1658 ignores vbases of abstract classes for determining
whether a destructor is deleted, but says nothing about exception
specifications.
DR 1351 specifically ignores vbases of abstract classes for determining the
exception specifi
On 08/03/15 07:15, Thomas Schwinge wrote:
Hi!
On Sat, 1 Aug 2015 20:20:49 -0400, Nathan Sidwell wrote:
this is the gomp4 committed version.
--- liboffloadmic/plugin/libgomp-plugin-intelmic.cpp(revision 226462)
+++ liboffloadmic/plugin/libgomp-plugin-intelmic.cpp(working copy)
@@ -32
This changes for lowering for most cases that can end up being tail-merged
to not substitute the operator into the result expression but refer to it
using a local variable of the name of the for iterator.
Boostrapped and tested on x86_64-unknown-linux-gnu, applied to trunk.
This drops the number
On Mon, Aug 3, 2015 at 3:02 PM, Kyrill Tkachov wrote:
>
> On 03/08/15 13:33, Uros Bizjak wrote:
>>
>> Hello!
>>
>>> 2015-07-30 Kyrylo Tkachov
>>>
>>> * ifcvt.c (noce_try_store_flag_constants): Make logic of the case
>>> when diff == STORE_FLAG_VALUE or diff == -STORE_FLAG_VALUE more
>
On Mon, 27 Jul 2015, Richard Biener wrote:
> On Mon, 27 Jul 2015, Richard Biener wrote:
>
> > On Mon, 27 Jul 2015, Richard Biener wrote:
> >
> > >
> > > I noticed that the code we generate for a simple gimple_assign_rhs1 (stmt)
> > > is quite bad as we have two checking pieces. The implementat
On 03/08/15 14:37, Uros Bizjak wrote:
On Mon, Aug 3, 2015 at 3:02 PM, Kyrill Tkachov wrote:
On 03/08/15 13:33, Uros Bizjak wrote:
Hello!
2015-07-30 Kyrylo Tkachov
* ifcvt.c (noce_try_store_flag_constants): Make logic of the case
when diff == STORE_FLAG_VALUE or diff == -STOR
This patch improves diagnostics for invalid bit-fields.
Bootstrapped/regtested on x86_64-linux, applying to trunk.
2015-08-03 Marek Polacek
PR c/67088
* c-decl.c (check_bitfield_type_and_width): Add location parameter.
Use it.
(grokdeclarator): Pass LOC down to
On Mon, Aug 3, 2015 at 3:02 PM, Kyrill Tkachov wrote:
>
> On 03/08/15 13:33, Uros Bizjak wrote:
>>
>> Hello!
>>
>>> 2015-07-30 Kyrylo Tkachov
>>>
>>> * ifcvt.c (noce_try_store_flag_constants): Make logic of the case
>>> when diff == STORE_FLAG_VALUE or diff == -STORE_FLAG_VALUE more
>
On 03/08/15 14:43, Kyrill Tkachov wrote:
On 03/08/15 14:37, Uros Bizjak wrote:
On Mon, Aug 3, 2015 at 3:02 PM, Kyrill Tkachov wrote:
On 03/08/15 13:33, Uros Bizjak wrote:
Hello!
2015-07-30 Kyrylo Tkachov
* ifcvt.c (noce_try_store_flag_constants): Make logic of the case
wh
On Mon, Aug 3, 2015 at 3:43 PM, Kyrill Tkachov wrote:
>
> On 03/08/15 14:37, Uros Bizjak wrote:
>>
>> On Mon, Aug 3, 2015 at 3:02 PM, Kyrill Tkachov
>> wrote:
>>>
>>> On 03/08/15 13:33, Uros Bizjak wrote:
Hello!
> 2015-07-30 Kyrylo Tkachov
>
> * ifcvt.c (noce_t
I've committed this patch to gomp4. The existing implementation of firstprivate
presumes the existence of memory at the CTA level. This patch does away with
that, treating firstprivate as thread-private variables initialized from the
host.
During development there was some fallout from decla
I've committed this to gomp4. I was reusing a size_t variable in a loop where
unsigned would do, and fprintf got upset.
nathan
2015-08-03 Nathan Sidwell
* config/nvptx/mkoffload.c (process): Avoid printf formatting
warning.
Index: config/nvptx/mkoffload.c
===
On 08/03/2015 05:55 AM, Jan-Benedict Glaw wrote:
On Sun, 2015-08-02 17:15:27 -0600, Martin Sebor wrote:
OK for the trunk. Sorry for the delay.
Thank you. Committed in revision 226480.
...und breaks native builds. When doing builds using config-list.mk, I
first build a GCC for the build mac
On Mon, Aug 3, 2015 at 7:04 AM, Uros Bizjak wrote:
> On Mon, Aug 3, 2015 at 3:02 PM, Kyrill Tkachov wrote:
>>
>> On 03/08/15 13:33, Uros Bizjak wrote:
>>>
>>> Hello!
>>>
2015-07-30 Kyrylo Tkachov
* ifcvt.c (noce_try_store_flag_constants): Make logic of the case
wh
On Fri, Oct 17, 2014 at 7:48 AM, Richard Sandiford
wrote:
> This is the main patch, to add new preferred_for_size and
> preferred_for_speed attributes that can be used to selectively disable
> alternatives when optimising for size or speed. As explained in the
> docs, the new attributes are just
The two files *-match.o files always finish building last, so if we
start building them as soon as possible (instead of pretty late) the
total build time will be less on a parallel build.
Bootstrapped and tested on powerpc64-linux. Is this okay for trunk?
Segher
2014-080-3 Segher Boessenkool
Hello.
Following patch has been installed to HSA branch.
Thanks,
Martin
>From 64f7f75a8b16dee2071fef7f546fbf6b06f82c4f Mon Sep 17 00:00:00 2001
From: mliska
Date: Fri, 31 Jul 2015 17:53:57 +0200
Subject: [PATCH 1/3] HSA: introduce more pure kernel information emission.
libgomp/ChangeLog:
2015-
Hello.
Following patch has been installed to HSA branch.
Thanks,
Martin
>From 125ee1ed4123527d4ce841631d4930ac96bec281 Mon Sep 17 00:00:00 2001
From: mliska
Date: Mon, 3 Aug 2015 13:00:42 +0200
Subject: [PATCH 2/3] HSA: fix kernel initialization for dependent kernels
libgomp/ChangeLog:
2015-08
Hello.
Following patch has been installed to HSA branch.
Thanks,
Martin
>From 163c6a4bb917ae76dad265dca5762a3bd87c947a Mon Sep 17 00:00:00 2001
From: mliska
Date: Mon, 3 Aug 2015 14:35:26 +0200
Subject: [PATCH 3/3] HSA: calculate properly maximum OMP data size which is
necessary.
libgomp/Chang
On Sun, 2015-08-02 at 12:18 +0200, Andreas Schwab wrote:
> Bill Schmidt writes:
>
> > * gcc.target/powerpc/vec-cmp-sel.c: New test.
>
> FAIL: gcc.target/powerpc/vec-cmp-sel.c (test for excess errors)
> Excess errors:
> /daten/gcc/gcc-20150801/gcc/testsuite/gcc.target/powerpc/vec-cmp-sel.c:14
PING.
Thanks,
Martin
On 07/20/2015 09:51 AM, Martin Liška wrote:
> On 07/16/2015 01:03 PM, Martin Liška wrote:
>> On 07/09/2015 06:24 PM, Jeff Law wrote:
>>> On 07/09/2015 07:56 AM, mliska wrote:
gcc/ChangeLog:
2015-07-09 Martin Liska
* dbgcnt.def: Add new debug co
On 03/08/15 11:52, James Greenhalgh wrote:
On Fri, Jul 24, 2015 at 11:43:32AM +0100, Kyrill Tkachov wrote:
On 21/07/15 16:37, James Greenhalgh wrote:
On Thu, Jul 16, 2015 at 04:20:59PM +0100, Kyrill Tkachov wrote:
Hi all,
This patch implements target attribute support via the
TARGET_OPTION_
On 07/16/2015 04:05 PM, Martin Liška wrote:
> On 07/10/2015 03:31 PM, Martin Jambor wrote:
>> Hi,
>>
>> thanks for working on this and sorry for a tad late review:
>>
>> On Thu, Jul 09, 2015 at 11:13:52AM +0200, Martin Liska wrote:
>>> gcc/ChangeLog:
>>>
>>> 2015-07-03 Martin Liska
>>>
>>> *
On August 3, 2015 5:03:05 PM GMT+02:00, Segher Boessenkool
wrote:
>The two files *-match.o files always finish building last, so if we
>start building them as soon as possible (instead of pretty late) the
>total build time will be less on a parallel build.
>
>Bootstrapped and tested on powerpc64-
On 08/03/2015 08:48 AM, Martin Sebor wrote:
On 08/03/2015 05:55 AM, Jan-Benedict Glaw wrote:
On Sun, 2015-08-02 17:15:27 -0600, Martin Sebor wrote:
OK for the trunk. Sorry for the delay.
Thank you. Committed in revision 226480.
...und breaks native builds. When doing builds using config-l
On 03/08/15 15:15, Uros Bizjak wrote:
On Mon, Aug 3, 2015 at 3:43 PM, Kyrill Tkachov wrote:
On 03/08/15 14:37, Uros Bizjak wrote:
On Mon, Aug 3, 2015 at 3:02 PM, Kyrill Tkachov
wrote:
On 03/08/15 13:33, Uros Bizjak wrote:
Hello!
2015-07-30 Kyrylo Tkachov
* ifcvt.c (noce_try_st
On 08/03/2015 05:42 AM, Kai Tietz wrote:
2015-08-03 5:49 GMT+02:00 Jason Merrill :
On 07/31/2015 05:54 PM, Kai Tietz wrote:
The "STRIP_NOPS-requirement in 'reduced_constant_expression_p'" I could
remove, but for one case in constexpr. Without folding we don't do
type-sinking/raising.
Right
Le 03/08/2015 14:36, Paul Richard Thomas a écrit :
Dear Mikael,
Thanks for your green light!
I have been mulling over the trans-decl part of the patch and having
been wondering if it is necessary.
You mean marking entities as public? Or setting the hidden visibility
attribute? Or both?
I th
On Mon, Aug 3, 2015 at 5:37 PM, Kyrill Tkachov wrote:
>> Looking at the x86 movcc expansion code (ix86_expand_int_movcc) I
>> don't think this is a good idea. In the expander, there is already
>> quite some target-dependent code that goes great length to utilize sbb
>> insn as much as possible, b
2015-08-03 17:04 GMT+03:00 Uros Bizjak :
> On Mon, Aug 3, 2015 at 3:02 PM, Kyrill Tkachov wrote:
>>
>> On 03/08/15 13:33, Uros Bizjak wrote:
>>>
>>> Hello!
>>>
2015-07-30 Kyrylo Tkachov
* ifcvt.c (noce_try_store_flag_constants): Make logic of the case
when diff ==
On 03/08/15 16:45, Uros Bizjak wrote:
On Mon, Aug 3, 2015 at 5:37 PM, Kyrill Tkachov wrote:
Looking at the x86 movcc expansion code (ix86_expand_int_movcc) I
don't think this is a good idea. In the expander, there is already
quite some target-dependent code that goes great length to utilize s
We can't use _N as an identifier, as per
https://gcc.gnu.org/onlinedocs/libstdc++/manual/source_code_style.html
Tested ppc64le-linux, committed to trunk.
commit ae65b96cfb69c64d636f86d9f9154da2364e0440
Author: Jonathan Wakely
Date: Mon Aug 3 16:29:44 2015 +0100
PR libstdc++/67078
*
Hi,
Ping for the patch submitted here
https://gcc.gnu.org/ml/gcc-patches/2015-07/msg02251.html
Thanks,
Pawel Kupidura
On 07/20/2015 08:19 AM, James Greenhalgh wrote:
I think we either want to defer the unordered_remove until we're done
processing all the vector elements, or make sure to look at element 'i'
again after we've moved something new in to it.
Correct. Two loops had this mistake -- while others got i
On Mon, Jul 27, 2015 at 02:22:41PM +0100, Pawel Kupidura wrote:
> diff --git a/gcc/ChangeLog b/gcc/ChangeLog
> index 10df325..ffafc3f 100644
> --- a/gcc/ChangeLog
> +++ b/gcc/ChangeLog
> @@ -1,3 +1,7 @@
> +2015-07-27 Pawel Kupidura
Two spaces between your name and your email address, like so:
2
On 08/03/2015 09:50 AM, Ilya Enkovich wrote:
The original code looks better, tree height is just 2 and therefore it
can be executed in 2 cycles. New code has more dependencies and tree
height becomes 5. It is always hard to say for all x86 targets but as
a generic code the original version is b
On Thu, Jul 30, 2015 at 04:08:47PM +0200, Mikael Morin wrote:
> Le 29/07/2015 18:45, Steve Kargl a écrit :
> >
> > This builds and passes regression testing on x86_64-*-freebsd.
> > OP found the problem by running the sanatizers. I don't
> > know how to build gcc with this as a build option. I'll
On 07/26/2015 07:05 PM, Kugan wrote:
I thought that when !gimple_assign_cast_p (stmt), RHS will always
boolean. I have now added this check in the attached patch.
Thanks.
I also noticed that in maybe_optimize_range_tests, GIMPLE_COND can
have non compatible types when new_op is updated
(bool
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=67077
is fixed by
https://gcc.gnu.org/viewcvs/gcc?view=revision&revision=226089
I checked in this patch to add a testcase for PR tree-optimization/67077.
Index: ChangeLog
===
--- ChangeL
On 03/08/15 16:53, Kyrill Tkachov wrote:
On 03/08/15 16:45, Uros Bizjak wrote:
On Mon, Aug 3, 2015 at 5:37 PM, Kyrill Tkachov wrote:
Looking at the x86 movcc expansion code (ix86_expand_int_movcc) I
don't think this is a good idea. In the expander, there is already
quite some target-dependen
Bill Schmidt writes:
> Index: gcc/testsuite/gcc.target/powerpc/vec-cmp-sel.c
> ===
> --- gcc/testsuite/gcc.target/powerpc/vec-cmp-sel.c(revision 226505)
> +++ gcc/testsuite/gcc.target/powerpc/vec-cmp-sel.c(working copy)
> @@
On Mon, Aug 3, 2015 at 7:20 PM, Kyrill Tkachov wrote:
> Looking at the x86 movcc expansion code (ix86_expand_int_movcc) I
> don't think this is a good idea. In the expander, there is already
> quite some target-dependent code that goes great length to utilize sbb
> insn as much as
On 07/16/2015 05:03 AM, Martin Liška wrote:
So a general question. We're passing in STRICT to several routines, which is
fine. But then we're also checking M_TAIL_MERGE_MODE. What's the difference
between the two? Can they be unified?
Hello.
I would say that STRICT is a bit generic mecha
On 07/20/2015 01:52 AM, Martin Liška wrote:
Due to changes dump file, it will be necessary to amend scanning of dump files.
Those changes are fine and can be installed once the prerequisites have
been approved/installed.
Thanks,
Jeff
The 64-bit register indirect call does not use register %r1. So, %r1 shouldn't
be clobbered in
the call pattern. Normally, this doesn't matter as register %r1 is call
clobbered. However,
reload can fail when the register used for the indirect call needs to be
reloaded from a static
location.
On 07/31/2015 03:45 PM, David Malcolm wrote:
In r223092 (aka dd4d567f4b6b498242097c41d63666bdae320ac1) I moved the
state of timevar.c from being global data into a "class timer" with
a global instance "g_timer".
This followup patch generalizes the timing within toplev so that an
external "timer"
On Mon, Aug 3, 2015 at 2:10 AM, Andreas Schwab wrote:
> ../../../libgo/runtime/mprof.goc: In function 'runtime_Stack':
> ../../../libgo/runtime/mprof.goc:408:5: error: calling
> '__builtin_frame_address' with a nonzero argument is unsafe
> [-Werror=frame-address]
> sp = runtime_getcallersp(&b
On 08/02/2015 05:03 AM, Kumar, Venkataramanan wrote:
Hi Jakub,
Thank you for reviewing the patch.
I have incorporated your comments in the attached patch.
Note Jakub is on PTO for the next 3 weeks.
vectorize_mults_via_shift.diff.txt
diff --git a/gcc/testsuite/gcc.dg/vect/vect-mult-patt
This patch from Andrew Wilkins kills off extra sleep processes when
running the libgo testsuite. Otherwise, the sleep process would
continue after the test completes. This doesn't affect the testsuite
but the extra sleep processes are not desirable. Committed to
mainline.
Ian
Index: gcc/go/gofr
On 07/30/2015 04:19 PM, Evgeny Stupachenko wrote:
Hi All,
The patch enables new attribute 'ctarget',
The attribute force compiler to create clones of a function with the attribute.
For example:
__attribute__((ctarget("avx","arch=slm","arch=core-avx2","default")))
So presumably we're allowing bo
PING: https://gcc.gnu.org/ml/gcc-patches/2015-07/msg01860.html
Actually, the xfailed test was because the function folded to nothing
and the offending code was removed without warning. Fixed in the
attached version. Same changelog.
On 22 July 2015 at 17:52, Manuel López-Ibáñez wrote:
> I took
PING: https://gcc.gnu.org/ml/gcc-patches/2015-07/msg02581.html
Thanks,
Manuel.
On 30 July 2015 at 17:35, Manuel López-Ibáñez wrote:
> When I fixed PR59304, I forgot that a command-line warning can be also
> an error if -Werror was enabled. This introduced a regression since
> anything enabled i
This patch by Chris Manghane changes the Go frontend to not make type
descriptor for named types if we have seen errors during the
compilation. This avoids a compiler crash, and fixes
https://golang.org/issue/11560 . Bootstrapped and ran Go testsuite on
x86_64-unknown-linux-gnu. Committed to mai
On Fri, Jul 31, 2015 at 12:43:20AM +, Joseph Myers wrote:
> On Wed, 29 Jul 2015, Michael Meissner wrote:
>
> > #6Add support for using different names for the 64/128-bit integer
> > conversion to IBM extended double, to allow a future version to
> > switch the default for wha
On 07/28/2015 06:00 AM, Kugan wrote:
On 27/07/15 05:38, Andreas Schwab wrote:
Kugan writes:
* cse.c (cse_insn): Fix missing check for STRICT_LOW_PART and minor
clean up.
This breaks
gcc.target/m68k/tls-ie-xgot.c scan-assembler jsr __m68k_read_tp
gcc.target/m68k/tls-ie.c s
On 06/29/2015 07:32 AM, Alexander Basov wrote:
I've updated patch with attributes lookup.
is it OK?
-- Alexander 2015-06-26 9:33 GMT+03:00 Alexander Basov :
>2015-06-25 21:47 GMT+03:00 Jeff Law:
>>On 06/03/2015 02:15 PM, Alexander Basov wrote:
>>>
>>>Hello Jeff,
>>>please find updated patch a
This patch fixes an oversight in the tabort. pattern (only generated via
the __builtin_tabort (cause) builtin), which currently allows using r0.
The problem is that if r0 is used, the "cause" argument value passed in
by the user will be ignored, since using r0 means the value zero and not
the conte
The Go frontend used to erroneously permit the same function to be
declared multiple times. This patch from Chris Manghane fixes that.
This fixes https://golang.org/issue/11573. Bootstrapped and ran Go
testsuite on x86_64-unknown-linux-gnu. Committed to mainline.
Ian
Index: gcc/go/gofrontend/ME
On Mon, Aug 3, 2015 at 3:39 PM, Peter Bergner wrote:
> This patch fixes an oversight in the tabort. pattern (only generated via
> the __builtin_tabort (cause) builtin), which currently allows using r0.
> The problem is that if r0 is used, the "cause" argument value passed in
> by the user will be
On 07/28/2015 01:36 PM, Richard Sandiford wrote:
> Index: gcc/target-insns.def
> ===
> --- gcc/target-insns.def 2015-07-28 20:56:29.721512028 +0100
> +++ gcc/target-insns.def 2015-07-28 20:56:29.713512127 +0100
> @@ -34,6 +34
I've committed this to gomp4 branch. It replaces the regular builtins
__builtin_GOACC_nid/__builtin_GOACC_id with internal functions IFN_OACC_DIM_SIZE
and IFN_OACC_DIM_POS -- moving further away from the PTX-specific naming of id &
nid. These functions should never turn into library calls or b
On 06/19/2015 07:21 AM, Ilya Enkovich wrote:
Hi,
This patch tries to improve 64bit integer computations on 32bit
target. This is achieved by an additional i386 target pass which
searches for all conversion candidates and tries to transform them
into vector mode when profitable.
Presumably you'r
1 - 100 of 117 matches
Mail list logo
