On 10/14/2015 10:15 PM, Renato Golin wrote:
On 14 October 2015 at 20:00, Andrew Pinski wrote:
Then until that happens I think we should disable asan and tsan for
AARCH64 for GCC.
I can't comment on that, but we'll continue running the tests on our
side on both 39 and 42 VMA configurations, to
On Wed, Oct 14, 2015 at 5:17 PM, H.J. Lu wrote:
> On Wed, Oct 14, 2015 at 8:15 AM, H.J. Lu wrote:
>> On Wed, Oct 14, 2015 at 8:08 AM, Yulia Koval wrote:
>>> Hi,
>>>
>>> This patch fixes the issue:
>>> https://gcc.gnu.org/bugzilla/show_bug.cgi?id=67963
>>>
>>> gcc/config/i386/i386.c (ix86_optio
On 14/10/15 20:17, Andrew Pinski wrote:
> On Wed, Oct 14, 2015 at 12:15 PM, Renato Golin
> wrote:
>> On 14 October 2015 at 20:00, Andrew Pinski wrote:
>>> Then until that happens I think we should disable asan and tsan for
>>> AARCH64 for GCC.
>>
>> I can't comment on that, but we'll continue
On Sun, Oct 11, 2015 at 12:47:19PM -0700, Aldy Hernandez wrote:
> I'm investigating balanced binary tree options for the multiple priorities
> variant of the task scheduler. In looking at the splay tree adaption in
> libgomp, I noticed that it requires preexisting typedefs and other
> definitions
On Wed, Oct 14, 2015 at 5:53 PM, Jeff Law wrote:
> On 10/14/2015 09:43 AM, Jan Hubicka wrote:
>>
>> I think he asked for trivial forward threads though due to repeated
>> tests.
>> I hacked FRE to do this (I think), but maybe some trivial cleanup
>> opportunities
>> are sti
On Wed, Oct 14, 2015 at 6:29 PM, Jan Hubicka wrote:
> Hi,
> this patch adds VIEW_CONVERT_EXPR which is another code omitted in
> operand_equal_p. During bootstrap there are about 1000 matches.
>
> Bootstrapped/regtested x86_64-linux, OK?
Eric, does that look ok WRT TYPE_ALIGN_OK? (that is, did
On 15 October 2015 at 08:29, Yury Gribov wrote:
> Do you have any estimation for when full AArch64 support is ready in LLVM?
> If it's still months away, I wonder if we may want to enable at least
> current (partial) support for non-Thunder users.
Hi Yury,
Unfortunately, no. Basic support is the
On Wed, Oct 14, 2015 at 6:50 PM, Richard Sandiford
wrote:
> This patch fixes some fallout from my patch to move the sqrt and cbrt
> folding rules to match.pd. The rules included canonicalisations like:
>
>sqrt(sqrt(x))->pow(x,1/4)
>
> which in the original code was only ever done at the gener
On Wed, Oct 14, 2015 at 6:21 PM, H.J. Lu wrote:
> By default, there is no visibility on builtin functions. When there is
> explicitly declared visibility on the C library function which a builtin
> function fall back on, we should honor the explicit visibility on the
> the C library function.
>
>
On Wed, Oct 14, 2015 at 9:15 PM, Bernd Schmidt wrote:
> On 10/14/2015 07:43 PM, Jeff Law wrote:
>>
>> Obviously some pessimization relative to current code is necessary to
>> fix some of the problems WRT thread safety and avoiding things like
>> introducing faults in code which did not previously
On 14/10/15 16:24, Andreas Schwab wrote:
Jiong Wang writes:
diff --git a/gcc/config.gcc b/gcc/config.gcc
index 5818663..215ad9a 100644
--- a/gcc/config.gcc
+++ b/gcc/config.gcc
@@ -3544,7 +3544,7 @@ case "${target}" in
eval "val=\$with_$which"
base_val=`echo $
Jiong Wang writes:
> I am surprised "-" doesn't need escape as it serve as range operator.
It's taken literally because it occurs at the end (it can also be put at
the start to have the same effect). There is no escape syntax inside
bracked expressions, only the position matters.
Andreas.
--
On Thu, Oct 15, 2015 at 09:57:17AM +0100, Jiong Wang wrote:
> On 14/10/15 16:24, Andreas Schwab wrote:
> >Jiong Wang writes:
> >
> >>diff --git a/gcc/config.gcc b/gcc/config.gcc
> >>index 5818663..215ad9a 100644
> >>--- a/gcc/config.gcc
> >>+++ b/gcc/config.gcc
> >>@@ -3544,7 +3544,7 @@ case "${ta
On Wed, Oct 14, 2015 at 5:59 PM, Eric Botcazou wrote:
> Hi,
>
> this is the regression of ACATS c37213k at -O2 with an upcoming change in
> the front-end of the Ada compiler:
>
> eric@polaris:~/gnat/gnat-head/native> gcc/gnat1 -quiet c37213k.adb -I
> /home/eric/gnat/bugs/support -O2
> +===
Hi!
Committed to gomp-4_0-branch in r228836 (that's the GCC trunk revision
before Jakub's big "Merge from gomp-4_1-branch to trunk"):
commit ff970ce56583aec594e741a2941505db91edb4ec
Merge: 47c1a67 e2c514f
Author: tschwinge
Date: Thu Oct 15 08:18:41 2015 +
svn merge -r 228315:228776 sv
> On Oct 15, 2015, at 1:42 AM, Renato Golin wrote:
>
>> On 15 October 2015 at 08:29, Yury Gribov wrote:
>> Do you have any estimation for when full AArch64 support is ready in LLVM?
>> If it's still months away, I wonder if we may want to enable at least
>> current (partial) support for non-Th
Hi,
This patch removes redundant memset and memcpy calls from libmpx. Bootstrapped
and tested w/ MPX on x86_64-unknown-linux-gnu. Applied to trunk.
Thanks,
Ilya
--
libmpx/
2015-10-15 Ilya Enkovich
PR other/66887
* mpxrt/mpxrt.c (read_mpx_status_sig): Remove useless code.
On Wed, Oct 14, 2015 at 08:54:11PM +0200, Marc Glisse wrote:
> On Wed, 14 Oct 2015, Marek Polacek wrote:
>
> >Evidently, the X - (X / Y) * Y -> X % Y pattern can't change the
> >signedness of X from signed to unsigned, otherwise we'd generate
> >wrong code. (But unsigned -> signed should be fine.
Hi all,
I'd like to turn on the scheduling-for-autoprefetching heuristic from
haifa-sched for aarch64.
This will have the effect of sorting sequences of loads and stores in ascending
order of offsets from a common base.
However, there is a limitation with the current code that I'd like to remov
On 15 October 2015 at 10:21, wrote:
> So in summary just enable 48 bit va support in the upstream kernel right now
> and not needed to test on thunderx. So please enable 48 bit va in the kernel.
> It is supported on a kernel that supports juno, apm and amd processors.
Hi Andrew,
I'm sorry but
On 10/15/2015 04:13 AM, Trevor Saunders wrote:
On Mon, Oct 12, 2015 at 12:19:08PM +0200, Bernd Schmidt wrote:
If we go to the trouble of changing this, could we convert macros to target
hooks instead while we're there? REVERSE_CONDITION, SETUP_FRAME_ADDRESSES
and FRAME_ADDR_RTX all seem to be us
On Thu, 15 Oct 2015, Marek Polacek wrote:
On Wed, Oct 14, 2015 at 08:54:11PM +0200, Marc Glisse wrote:
On Wed, 14 Oct 2015, Marek Polacek wrote:
Evidently, the X - (X / Y) * Y -> X % Y pattern can't change the
signedness of X from signed to unsigned, otherwise we'd generate
wrong code. (But
On 13/10/15 10:12 +0100, Szabolcs Nagy wrote:
Dejagnu tweak:
Check the compiler output in libstdc++-dg-test using
${tool}_check_unsupported_p and mark the output
unsupported accordingly to avoid "relocation truncated"
failures cluttering the test results on aarch64-none-elf
with -mcmodel=tiny. (g
On 10/15/2015 11:40 AM, Kyrill Tkachov wrote:
The code that analyzes the offsets of the loads/stores doesn't try to
handle load/store-multiple insns.
These appear rather frequently in memory streaming workloads on aarch64
in the form of load-pair/store-pair instructions
i.e. ldp/stp. In RTL, the
Hi Mikael, hi all,
I have checked that the patch (my initial one for pr59678) does compile
and test fine. Given that the patch lives in trunk-6 for quite some
time now, without any major complaints, I approve to commit to gcc-5.
Given the patch was reviewed by Paul already, I don't see any reason
On 14/10/15 10:30, Jakub Jelinek wrote:
On Tue, Oct 13, 2015 at 02:16:23PM +0300, Maxim Ostapenko wrote:
This patch introduces required compiler changes. Now, we don't version
asan_init, we have a special __asan_version_mismatch_check_v[n] symbol for
this.
For this, I just have to wonder what i
On Thu, Oct 15, 2015 at 1:44 AM, Richard Biener
wrote:
> On Wed, Oct 14, 2015 at 6:21 PM, H.J. Lu wrote:
>> By default, there is no visibility on builtin functions. When there is
>> explicitly declared visibility on the C library function which a builtin
>> function fall back on, we should honor
On Thu, 15 Oct 2015, Marc Glisse wrote:
> On Thu, 15 Oct 2015, Marek Polacek wrote:
>
> > On Wed, Oct 14, 2015 at 08:54:11PM +0200, Marc Glisse wrote:
> > > On Wed, 14 Oct 2015, Marek Polacek wrote:
> > >
> > > > Evidently, the X - (X / Y) * Y -> X % Y pattern can't change the
> > > > signedness
On Fri, Oct 09, 2015 at 08:26:25PM +0300, Ilya Verbin wrote:
> On Fri, Oct 09, 2015 at 09:55:07 +0200, Jakub Jelinek wrote:
> > -GOMP_4.1 {
> > +GOMP_4.5 {
> >global:
> > GOMP_target_41;
> > GOMP_target_data_41;
>
> Should we rename it to GOMP_target*_45, or do you know some more mnemo
> Eric, does that look ok WRT TYPE_ALIGN_OK? (that is, did we decide
> TYPE_ALIGN_OK is no longer needed?)
I'm not sure we need to care about TYPE_ALIGN_OK here so no objection by me.
--
Eric Botcazou
Get rid of some of the ancient way of building stmts and SSA names.
Bootstrapped & tested on x86_64-unknown-linux-gnu, applied.
Richard.
2015-10-15 Richard Biener
* tree-vectorizer.h (vect_get_new_ssa_name): Declare.
* tree-vect-data-refs.c (vect_get_new_ssa_name): New helpe
On Wed, Oct 14, 2015 at 5:08 PM, Yulia Koval wrote:
> Hi,
>
> This patch fixes the issue:
> https://gcc.gnu.org/bugzilla/show_bug.cgi?id=67963
>
> gcc/config/i386/i386.c (ix86_option_override_internal) Disable
> 80387 mask if lakemont target is set.
>
> diff --git a/gcc/config/i386/i386.c b/
Hi,
this one seems simple, matter of handling the relevant tree codes.
Tested x86_64-linux.
Thanks,
Paolo.
/
/cp
2015-10-15 Paolo Carlini
PR c++/67926
* constexpr.c (potential_constant_expression_1): Handle
UNARY_LEFT_FOLD_EXPR, UNARY_RIGHT_FOLD
On Thu, 15 Oct 2015, Richard Biener wrote:
>
> Get rid of some of the ancient way of building stmts and SSA names.
>
> Bootstrapped & tested on x86_64-unknown-linux-gnu, applied.
Committed an old patch, fixed now.
2015-10-15 Richard Biener
* tree-vect-stmts.c (vect_init_vector): Re
On Wed, Oct 14, 2015 at 06:18:03PM -0600, Martin Sebor wrote:
> On 10/14/2015 03:42 PM, Arkadiusz Drabczyk wrote:
> >On Wed, Oct 14, 2015 at 08:36:43AM -0600, Martin Sebor wrote:
> >>On 10/13/2015 04:47 PM, Arkadiusz Drabczyk wrote:
> >>>* gcc/doc/extend.texi: documentation says that functions decl
On Thu, Oct 15, 2015 at 8:11 AM, Hurugalawadi, Naveen
wrote:
> Hi,
>
> Thanks for all the suggestions.
> Please find attached the modified patch as per your suggestions.
>
> I had missed a mail as pointed by Marc Glisse. Now I have implemented
> everything suggested.
> Please review the patch and
This patch makes sure that, for every simplification that uses
fold_strip_sign_ops, there are associated match.pd rules for the
leaf sign ops, i.e. abs, negate and copysign. A follow-on patch
will add a pass to handle more complex cases.
Tested on x86_64-linux-gnu, aarch64-linux-gnu and arm-linux
This patch adds a pass that collects information that is common to all
uses of an SSA name X and back-propagates that information up the statements
that generate X. The general idea is to use the information to simplify
instructions (rather than a pure DCE) so I've simply called it
tree-ssa-backpr
builtins-20.c had:
if (cos((y*=2, -fabs(tan(x/-y != cos((y*=2,tan(x/y
link_error ();
which is undefined behaviour. The test expected that y had the same
value in x/y and x/-y, but gimplification actually implements the
"obvious" interpretation, multiplying y by 2, using it for one
Stripping unnecessary sign ops at the gimple level means that we're
no longer able to optimise:
if (cos(y<10 ? -fabs(x) : tan(x<20 ? -x : -fabs(y)))
!= cos(y<10 ? x : tan(x<20 ? x : y)))
link_error ();
because we're currently not able to fold away the equality in:
int
f1 (double x, d
Richi pointed out that SPEC 2006 447.dealII ICEs with a SIGSEGV
because of my recent changes in reassoc. This patch fixes it.
Bootstrapped/regtested on x86_64-linux, applying to trunk.
2015-10-15 Marek Polacek
* tree-ssa-reassoc.c (attempt_builtin_copysign): Call
gimple_call_
This patch deletes fold_strip_sign_ops in favour of the tree-ssa-backprop.c
pass.
Tested on x86_64-linux-gnu, aarch64-linux-gnu and arm-linux-gnueabi.
OK to install?
Thanks,
Richard
gcc/
* fold-const.h (fold_strip_sign_ops): Delete.
* fold-const.c (fold_strip_sign_ops): Likewise
On 9 October 2015 at 18:11, James Greenhalgh wrote:
> On Thu, Oct 08, 2015 at 01:29:34PM +0100, Richard Biener wrote:
>> > Thanks again for the comments Richard!
>> >
>> > A new algorithmic optimisation:
>> >
>> > ((X inner_op C0) outer_op C1)
>> > With X being a tree where value_range has reasone
On Thu, Oct 15, 2015 at 12:18:12PM +0200, Andre Vehreschild wrote:
>
> Regards,
> Andre
>
> PS: Note, I don't have reviewer status.
>
Given your contributions to gfortran and the fact that
you probably understand portions of the (de)allocation
code better than anyone, I suspect that your
Hi all,
This is a backport patch to loosen restrictions on core registers for
DImode values in Thumb2.
It fixes PR67383. In this particular case, reload tries to spill a hard
register, and use next register together as a pair to reload a DImode
pseudo register. However, the spilled register
Hi!
CCing various people, because I'd like to have something that won't work on
XeonPhi only.
On Fri, Oct 02, 2015 at 10:28:01PM +0300, Ilya Verbin wrote:
> On Tue, Sep 08, 2015 at 11:20:14 +0200, Jakub Jelinek wrote:
> > nowait support for #pragma omp target is not implemented yet, supposedly we
Hello,
On 14 Oct 13:40, Joseph Myers wrote:
> On Wed, 14 Oct 2015, Kirill Yukhin wrote:
>
> > Is it ok for trunk?
>
> This patch has no documentation. Documentation for new attributes must be
> added to extend.texi.
Fixed. Extra entry to gcc/Changelog:
* doc/extend.texi (simd): Document
On Thu, Oct 15, 2015 at 05:33:32PM +0300, Kirill Yukhin wrote:
> --- a/gcc/doc/extend.texi
> +++ b/gcc/doc/extend.texi
> @@ -3066,6 +3066,20 @@ This function attribute make a stack protection of the
> function if
> flags @option{fstack-protector} or @option{fstack-protector-strong}
> or @option{
Hi Jakub,
On 15 Oct 16:39, Jakub Jelinek wrote:
> On Thu, Oct 15, 2015 at 05:33:32PM +0300, Kirill Yukhin wrote:
> > --- a/gcc/doc/extend.texi
> > +++ b/gcc/doc/extend.texi
> > @@ -3066,6 +3066,20 @@ This function attribute make a stack protection of
> > the function if
> > flags @option{fstack-p
On 10/15/2015 12:56 AM, Jakub Jelinek wrote:
On Sun, Oct 11, 2015 at 12:47:19PM -0700, Aldy Hernandez wrote:
I'm investigating balanced binary tree options for the multiple priorities
variant of the task scheduler. In looking at the splay tree adaption in
libgomp, I noticed that it requires pre
> In other words,
> consider youself a reviewer for patches in an area
> of the compiler that you are comfortable.
Seconded.
FX
On 15/10/15 11:16, Bernd Schmidt wrote:
On 10/15/2015 11:40 AM, Kyrill Tkachov wrote:
The code that analyzes the offsets of the loads/stores doesn't try to
handle load/store-multiple insns.
These appear rather frequently in memory streaming workloads on aarch64
in the form of load-pair/store-pa
On Thu, 15 Oct 2015, Jakub Jelinek wrote:
> Looking at Cuda, for async target region kernels we'd probably use
> a non-default stream and enqueue the async kernel in there. I see
> we can e.g. cudaEventRecord into the stream and then either cudaEventQuery
> to busy poll the event, or cudaEventSync
On Thu, 15 Oct 2015, Richard Sandiford wrote:
This patch makes sure that, for every simplification that uses
fold_strip_sign_ops, there are associated match.pd rules for the
leaf sign ops, i.e. abs, negate and copysign. A follow-on patch
will add a pass to handle more complex cases.
Tested on
On 10/13/2015 11:32 AM, Jeff Law wrote:
On 10/13/2015 02:21 AM, Nikolai Bozhenov wrote:
2015-10-13 Nikolai Bozhenov
* gcc/rtl.h (print_insn): fix prototype
Installed on the trunk after bootstrap & regression test.
jeff
Sorry, a little late to the party.. but why is print_insn even in
> Le 15 oct. 2015 à 16:59, FX a écrit :
>
>> In other words,
>> consider youself a reviewer for patches in an area
>> of the compiler that you are comfortable.
>
> Seconded.
>
> FX
Agreed,
Dominique
On Thu, Oct 15, 2015 at 16:01:56 +0200, Jakub Jelinek wrote:
> On Fri, Oct 02, 2015 at 10:28:01PM +0300, Ilya Verbin wrote:
> > Here is my WIP patch. target.c part is obviously incorrect, but it
> > demonstrates
> > a possible libgomp <-> plugin interface for running a target task function
> > as
On 10/15/2015 07:28 PM, Andrew MacLeod wrote:
On 10/13/2015 11:32 AM, Jeff Law wrote:
On 10/13/2015 02:21 AM, Nikolai Bozhenov wrote:
2015-10-13 Nikolai Bozhenov
* gcc/rtl.h (print_insn): fix prototype
Installed on the trunk after bootstrap & regression test.
jeff
Sorry, a little lat
On Thu, Oct 15, 2015 at 1:45 PM, Uros Bizjak wrote:
> On Wed, Oct 14, 2015 at 5:08 PM, Yulia Koval wrote:
>> Hi,
>>
>> This patch fixes the issue:
>> https://gcc.gnu.org/bugzilla/show_bug.cgi?id=67963
>>
>> gcc/config/i386/i386.c (ix86_option_override_internal) Disable
>> 80387 mask if lake
> > * fold-const.c (operand_equal_p): Handle VIEW_CONVERT_EXPR.
> > Index: fold-const.c
> > ===
> > --- fold-const.c(revision 228735)
> > +++ fold-const.c(working copy)
> > @@ -2962,6 +2968,12 @@ operand_equal_p
On Thu, Oct 15, 2015 at 9:57 AM, Uros Bizjak wrote:
> On Thu, Oct 15, 2015 at 1:45 PM, Uros Bizjak wrote:
>> On Wed, Oct 14, 2015 at 5:08 PM, Yulia Koval wrote:
>>> Hi,
>>>
>>> This patch fixes the issue:
>>> https://gcc.gnu.org/bugzilla/show_bug.cgi?id=67963
>>>
>>> gcc/config/i386/i386.c (ix
I've committed this to gomp4 branch.
It's the next in the series moving partioning decisions into the target
compiler. This patch moves the updating of the IF_GOACC_LOOP internal
function's mask and chunking parameters. After reconstructing the OpenACC
loops, we scan the block(s) justy after
On Thu, Oct 15, 2015 at 07:18:53PM +0300, Alexander Monakov wrote:
> On Thu, 15 Oct 2015, Jakub Jelinek wrote:
> > Looking at Cuda, for async target region kernels we'd probably use
> > a non-default stream and enqueue the async kernel in there. I see
> > we can e.g. cudaEventRecord into the strea
A powerpc toolchain built with (or without) --enable-secureplt
currently creates a binary that uses bss plt if
(1) any of the linked PIC objects have bss plt relocs
(2) or all the linked objects are non-PIC or have no relocs,
because this is the binutils linker behaviour.
This patch passes --se
Hi!
This patch implements some new restrictions, clarifications etc.
on the declare target construct. Committed to gomp-4_5-branch.
2015-10-15 Jakub Jelinek
gcc/c/
* c-parser.c (c_parser_omp_declare_target): Call c_finish_omp_clauses
in the parenthesized extended-list syntax
Me too!
Paul
On 15 October 2015 at 18:31, Dominique d'Humières wrote:
>
>> Le 15 oct. 2015 à 16:59, FX a écrit :
>>
>>> In other words,
>>> consider youself a reviewer for patches in an area
>>> of the compiler that you are comfortable.
>>
>> Seconded.
>>
>> FX
>
> Agreed,
>
> Dominique
>
--
On Thu, 15 Oct 2015, Jakub Jelinek wrote:
> > - this functionality doesn't currently work through CUDA MPS
> > ("multi-process
> > server", for funneling CUDA calls from different processes through a
> > single "server" process, avoiding context-switch overhead on the device,
> > som
On Sun, Oct 11, 2015 at 9:07 AM, Alan Modra wrote:
> On Sat, Oct 10, 2015 at 11:25:38PM +0200, Andreas Schwab wrote:
>> "Lynn A. Boger" writes:
>>
>> > Index: gcc/config/rs6000/sysv4.h
>> > ===
>> > --- gcc/config/rs6000/sysv4.h
On Thu, Oct 15, 2015 at 07:48:18PM +0300, Nikolai Bozhenov wrote:
>
> On 10/15/2015 07:28 PM, Andrew MacLeod wrote:
> >On 10/13/2015 11:32 AM, Jeff Law wrote:
> >>On 10/13/2015 02:21 AM, Nikolai Bozhenov wrote:
> >>>2015-10-13 Nikolai Bozhenov
> >>>
> >>> * gcc/rtl.h (print_insn): fix prototy
On 06/05/15 12:25, Szabolcs Nagy wrote:
On 29/04/15 14:51, Szabolcs Nagy wrote:
On 29/04/15 14:17, Michael Eager wrote:
On 04/27/2015 07:35 AM, Szabolcs Nagy wrote:
On 20/04/15 19:54, Szabolcs Nagy wrote:
Set up dynamic linker name for microblaze.
Patch v2.
(undef MUSL_DYNAMIC_LINKER that
On 10/15/2015 12:00 PM, Szabolcs Nagy wrote:
On 06/05/15 12:25, Szabolcs Nagy wrote:
On 29/04/15 14:51, Szabolcs Nagy wrote:
On 29/04/15 14:17, Michael Eager wrote:
On 04/27/2015 07:35 AM, Szabolcs Nagy wrote:
On 20/04/15 19:54, Szabolcs Nagy wrote:
Set up dynamic linker name for microblaze.
On 15/10/15 20:04, Michael Eager wrote:
On 10/15/2015 12:00 PM, Szabolcs Nagy wrote:
Ping,
patch still works.
https://gcc.gnu.org/ml/gcc-patches/2015-05/msg00437.html
gcc/Changelog:
2015-05-06 Gregor Richards
Szabolcs Nagy
* config/microblaze/linux.h (MUSL_DYNAMIC_LINKER): D
On Thu, Oct 15, 2015 at 6:59 PM, H.J. Lu wrote:
> On Thu, Oct 15, 2015 at 9:57 AM, Uros Bizjak wrote:
>> On Thu, Oct 15, 2015 at 1:45 PM, Uros Bizjak wrote:
>>> On Wed, Oct 14, 2015 at 5:08 PM, Yulia Koval wrote:
Hi,
This patch fixes the issue:
https://gcc.gnu.org/bugzilla/s
On Thu, Oct 15, 2015 at 12:16 PM, Uros Bizjak wrote:
> On Thu, Oct 15, 2015 at 6:59 PM, H.J. Lu wrote:
>> On Thu, Oct 15, 2015 at 9:57 AM, Uros Bizjak wrote:
>>> On Thu, Oct 15, 2015 at 1:45 PM, Uros Bizjak wrote:
On Wed, Oct 14, 2015 at 5:08 PM, Yulia Koval wrote:
> Hi,
>
> T
On Oct 15, 2015, at 6:18 AM, Richard Sandiford
wrote:
> Stripping unnecessary sign ops at the gimple level means that we're
> no longer able to optimise:
>
> if (cos(y<10 ? -fabs(x) : tan(x<20 ? -x : -fabs(y)))
> != cos(y<10 ? x : tan(x<20 ? x : y)))
>link_error ();
>
> because we're
On Thu, Oct 15, 2015 at 9:21 PM, H.J. Lu wrote:
> On Thu, Oct 15, 2015 at 12:16 PM, Uros Bizjak wrote:
>> On Thu, Oct 15, 2015 at 6:59 PM, H.J. Lu wrote:
>>> On Thu, Oct 15, 2015 at 9:57 AM, Uros Bizjak wrote:
On Thu, Oct 15, 2015 at 1:45 PM, Uros Bizjak wrote:
> On Wed, Oct 14, 2015
Marc Glisse writes:
> On Thu, 15 Oct 2015, Richard Sandiford wrote:
>
>> This patch makes sure that, for every simplification that uses
>> fold_strip_sign_ops, there are associated match.pd rules for the
>> leaf sign ops, i.e. abs, negate and copysign. A follow-on patch
>> will add a pass to hand
> I'm not sure we need to care about TYPE_ALIGN_OK here so no objection by me.
Actually I have one: can we please fix the multiple Ada breakages caused by
the previous controversial change from Jan before going farther in the series?
The build is still broken on IA-64 and I'm still seeing ICEs on
Mike Stump writes:
> On Oct 15, 2015, at 6:18 AM, Richard Sandiford
> wrote:
>> Stripping unnecessary sign ops at the gimple level means that we're
>> no longer able to optimise:
>>
>> if (cos(y<10 ? -fabs(x) : tan(x<20 ? -x : -fabs(y)))
>> != cos(y<10 ? x : tan(x<20 ? x : y)))
>>link_
Alan said he did this, which was a bootstrap and regression test of
all the combinations:
/I built four configurations, powerpc-linux 32-bit only,
powerpc64le-linux 64-bit only, biarch powerpc-linux with 32-bit default,
and biarch powerpc64-linux with 64-bit default/
I also did verify that on
On Oct 15, 2015, at 12:47 PM, Richard Sandiford
wrote:
> I can see that argument if people are only taking work items from
> the PR database. But it's possible (likely even) that people will
> independently find a problem like this and just fix it, if the missed
> optimisation happens to be impo
Mike Stump writes:
> On Oct 15, 2015, at 12:47 PM, Richard Sandiford
> wrote:
>> I can see that argument if people are only taking work items from
>> the PR database. But it's possible (likely even) that people will
>> independently find a problem like this and just fix it, if the missed
>> opti
On Oct 15, 2015, at 1:38 PM, Richard Sandiford
wrote:
> Mike Stump writes:
>> On Oct 15, 2015, at 12:47 PM, Richard Sandiford
>> wrote:
>>> I can see that argument if people are only taking work items from
>>> the PR database. But it's possible (likely even) that people will
>>> independently
I've committed this to gomp4 branch. It fixes the routine-7 regression I
caused when reworking the reduction machinery.
nathan
2015-10-15 Nathan Sidwell
* omp-low.c (lower_oacc_reductions): Check outer context is a
target before lookup.
Index: gcc/omp-low.c
=
On Wed, 14 Oct 2015, Martin Sebor wrote:
> On 10/14/2015 01:40 PM, Joseph Myers wrote:
> > On Wed, 14 Oct 2015, Martin Sebor wrote:
> >
> > > IMO, printing the aliased option's help text makes using the output
> > > easier for users who find the undocumented option first, in that
> > > they don't
The patch is at the following link:
https://gcc.gnu.org/ml/gcc-patches/2015-10/msg00915.html
Martin
On 10/08/2015 08:55 PM, Martin Sebor wrote:
Gcc attempts to diagnose invalid offsetof expressions whose member
designator is an array element with an out-of-bounds index. The
logic in the func
This seventh revision of the patch:
* Cleans up style issues.
* Correct bug in condition.
* Improved testing code.
Ok for check in.
Benedikt Huber (1):
2015-10-15 Benedikt Huber
Philipp Tomsich
gcc/ChangeLog | 20
gcc/config
* config/aarch64/aarch64-builtins.c: Builtins for rsqrt and rsqrtf.
* config/aarch64/aarch64-protos.h: Declare.
* config/aarch64/aarch64-simd.md: Matching expressions for frsqrte and
frsqrts.
* config/aarch64/aarch64-tuning-flags.def: Added recip_sqrt.
I've applied this to move the execute_oacc_device_lower function later in the
file. I'll shortly be changing it to explicitly call a default oacc handler,
and reordering makes the diff confusing (diff choses to make this diff confusing
enough).
nathan
2015-10-15 Nathan Sidwell
* omp-low.
On 10/15/2015 04:51 PM, Kugan wrote:
+generation of subreg in RTL, that intern results in removal of
s/intern/in turn/
r~
OK.
Jason
> > I'm not sure we need to care about TYPE_ALIGN_OK here so no objection by me.
>
> Actually I have one: can we please fix the multiple Ada breakages caused by
> the previous controversial change from Jan before going farther in the series?
> The build is still broken on IA-64 and I'm still seei
Hi,
as Richard noticed in my port of the code to operand_equal_p, the checking of
CONSTURCTOR in ipa-icf-gimple is incomplete missing the index checks.
It is also unnecesary since non-empty ctors does not happen as gimple
operands. This patch thus removes the unnecesary code.
Bootstrapped/regtest
On Thu, Oct 15, 2015 at 1:50 PM, Szabolcs Nagy wrote:
> A powerpc toolchain built with (or without) --enable-secureplt
> currently creates a binary that uses bss plt if
>
> (1) any of the linked PIC objects have bss plt relocs
> (2) or all the linked objects are non-PIC or have no relocs,
>
> beca
On 10/15/2015 11:27 AM, Kyrill Tkachov wrote:
On 15/10/15 11:16, Bernd Schmidt wrote:
On 10/15/2015 11:40 AM, Kyrill Tkachov wrote:
The code that analyzes the offsets of the loads/stores doesn't try to
handle load/store-multiple insns.
These appear rather frequently in memory streaming workloa
On Thu, Oct 15, 2015 at 11:47:10AM +0200, Bernd Schmidt wrote:
> On 10/15/2015 04:13 AM, Trevor Saunders wrote:
> >On Mon, Oct 12, 2015 at 12:19:08PM +0200, Bernd Schmidt wrote:
> >>If we go to the trouble of changing this, could we convert macros to target
> >>hooks instead while we're there? REVE
> Jan Hubicka writes:
>
> > Does the patch in https://gcc.gnu.org/ml/gcc-patches/2015-10/msg00902.html
> > help?
>
> No, it doesn't.
>
Andreas,
I am sorry for getting late to this. I hoped that the alternative patch by
Alexandre would fix this.
I still don't know how to reproduce without IA-6
On Thu, Oct 15, 2015 at 9:30 PM, Uros Bizjak wrote:
Do we support -O2 -march=lakemont with
__attribute__((target("arch=silvermont")))
>>>
>>> Hm, no.
>>>
>>
>> Do we issue an error or silently ignore
>> __attribute__((target("arch=silvermont")))?
>> If we don't support it, should w
98 matches
Mail list logo
