Some of the SVE patches extend SLP to predicated operations created by
ifcvt. However, ifcvt currently forces the mask into a temporary:
mask = ifc_temp_var (TREE_TYPE (mask), mask, &gsi);
and at the moment SLP doesn't handle simple assignments like:
SSA_NAME = SSA_NAME
SS
Prakhar Bahuguna writes:
> On 31/05/2017 14:11:43, Richard Sandiford wrote:
>> Prakhar Bahuguna writes:
>> > On 31/05/2017 09:19:40, Richard Sandiford wrote:
>> >> const_ints are supposed to be stored in sign-extended form, so a 32-bit
>> >> integer with the MSB set should be 0x8000|x
On Wed, May 31, 2017 at 06:33:37PM -0400, Michael Meissner wrote:
> Here is the updated version of the target_clone attribute support.
>
> The changes include:
>
> 1) Change the order of the CLONE priority list from default being 0 to ISA 3.0
> being the highest (and eliminating the enum and cast
On 05/27/2017 06:44 PM, Martin Sebor wrote:
+ /* True if the class is trivial and has a trivial non-deleted copy
+ assignment, copy ctor, and default ctor, respectively. The last
+ one isn't used to issue warnings but only to decide what suitable
+ alternatives to offer as replaceme
On 31 May 2017 at 21:03, Nicolas Koenig wrote:
> Hello Dominique,
>
> attached is the next try, this time without stupidities (I hope). Both test
> cases you posted don't ICE anymore.
>
> Ok for trunk?
Please check contrib/check_GNU_style.sh /tmp/p8.diff
and let me point you to contrib/vimrc
Fur
Here is the updated version of the target_clone attribute support.
The changes include:
1) Change the order of the CLONE priority list from default being 0 to ISA 3.0
being the highest (and eliminating the enum and casts).
2) I tried to fix the various spacing issues. I found one place further
On Wed, May 31, 2017 at 1:39 PM, wrote:
> From: Vladimir Mezentsev
>
> Tested on sparc64-linux-gnu. 7243 tests passed
>
> 2017-05-31 Vladimir Mezentsev
>
> * libgo/go/runtime/lfstack_64bit.go (lfstackPack, lfstackUnpack): Handle
> sparc64-linux
> * libgo/mkrsysinfo.sh (epollevent structure):
Hi!
On Wed, May 31, 2017 at 12:35:57PM -0500, Steven Munroe wrote:
> Index: gcc/testsuite/gcc.target/powerpc/bmi2-pext64-1a.c
> ===
> --- gcc/testsuite/gcc.target/powerpc/bmi2-pext64-1a.c (revision 248468)
> +++ gcc/testsuite/gcc.targ
The canonical RTL for "nor" is (and (not ()) (not ())), and that is
indeed what we use in boolccv2df3_internal1. So, the splitter for
*vector_uneq should use that form, not (not (ior () ())), which
does not match any pattern.
Tested on powerpc64-linux {-m32,-m64}, and tested the pr50310-2.c
testc
On Tue, May 30, 2017 at 2:14 AM, Volker Reichelt
wrote:
> On 24 May, Jason Merrill wrote:
>> On Mon, May 15, 2017 at 3:58 PM, Martin Sebor wrote:
So how about the following then? I stayed with the catch part and added
a parameter to the warning to let the user decide on the warnings she
On 05/30/2017 05:06 AM, Pierre-Marie de Rodat wrote:
Hello,
In Ada, the Character type is supposed to be unsigned. However,
depending on the sign of C char types, GNAT can materialize it as a
signed type for code generation purposes. When this is the case, GNAT
also attach a debug type to it s
On 05/25/2017 05:29 AM, Ville Voutilainen wrote:
Tested on Linux-x64, running full suite on Linux-ppc64. It seems fitting
to put the test into the library tests, we don't have separate tests
on the front-end side for __is_constructible, so I think adding such
would be a separate job.
2017-05-25
On 31/05/2017 12:34, Jonathan Wakely wrote:
Well in general the is_nothrow_default_constructible trait also tells
you if the type is default-constructible at all, but the form above
won't compile if it isn't default-constructible. In this specific case
it doesn't matter, because that constructor
Hi,
Add support for early expansion of vector shifts. Including
vec_sl (shift left), vec_sr (shift right), vec_sra (shift
right algebraic), vec_rl (rotate left).
Part of this includes adding the vector shift right instructions to
the list of those instructions having an unsigned second argument.
Hi,
Add support for early expansion of vector eqv built-ins.
Bootstraps currently running.
OK for trunk?
Thanks,
-Will
[gcc]
2017-05-26 Will Schmidt
* config/rs6000/rs6000.c (rs6000_gimple_fold_builtin): Add handling
for early expansion of vec_eqv.
Hi,
(resending with folks on CC, apologies to anyone having deja-vu)
Add support for early expansion of vec_min, vec_max built-ins.
Bootstraps currently running.
OK for trunk?
Thanks,
-Will
[gcc]
2017-05-26 Will Schmidt
* config/rs6000/rs6000.c (rs6000_gimple_fold_builtin): Add h
Hi,
Add support for early expansion of vector absolute built-ins.
[V2] Per reviews and feedback, skip the early folding for
integral types based on a check against TYPE_OVERFLOW_WRAPS(arg0).
Added test variants to exercise the -fwrapv option during
this folding.
OK for trunk? (bootstraps runn
On May 31, 2017, at 12:33 AM, jay pokarna wrote:
>
>Could you tell the way as to how can I measure the time taken
> by my algorithm and compare it with the inbuilt functions ?
No, that's beyond our charter. We review patches for gcc. I'd recommend
google, it has answers to most questi
Hello Dominique,
attached is the next try, this time without stupidities (I hope). Both
test cases you posted don't ICE anymore.
Ok for trunk?
Nicolas
Regression tested for x86_64-pc-linux-gnu.
Changelog (still the same):
2017-05-27 Nicolas Koenig
PR fortran/35339
* fron
OK.
On Wed, May 31, 2017 at 8:04 AM, Paolo Carlini wrote:
> Hi,
>
> this one appears to be a rather simple case of missing diagnostic: in
> convert_to_void we aren't calling maybe_warn_nodiscard when we strip an
> INDIRECT_REF wrapping a CALL_EXPR thus we don't issue the diagnostic that we
> norm
Hello!
Attached patch allows direct XMM->GR zero extensions for 32bit
targets. This insn will be split after reload to a direct
XMM->lowpart(GR) move and 0->highpart(GR) zeroing.
2017-05-31 Uros Bizjak
* config/i386/i386.md (*zero_extendsidi2): Enable alternative (?r, *Yj)
also for 32
On Tue, May 30, 2017 at 4:29 PM, Peryt, Sebastian
wrote:
> Hi,
>
> This patch adds missing intrinsics for MAX[SD,SS] and MIN[SD,SS] listed below:
> - _mm_mask_max_sd,
> - _mm_maskz_max_sd,
> - _mm_mask_max_ss,
> - _mm_maskz_max_ss,
>
> - _mm_mask_min_sd,
> - _mm_maskz_min_sd,
> - _mm_mask_min_ss,
On Fri, May 26, 2017 at 12:20:05PM -0500, Will Schmidt wrote:
> This test has been flaky on both AIX and in older linux based
> environments. Notably, the number of xxlor instructions
> generated varies depending on the platform and the specified
> bit-size (32/64), with older environments generati
Bill Seurer pointed out that building the BMI tests on a power8 but with
gcc built --with-cpu=power6 fails with link errors. The intrinsics
_pdep_u64/32 and _pext_u64/32 are guarded with #ifdef _ARCH_PWR7 as the
implementation uses bpermd and popcntd instructions introduced with
power7 (PowerISA-2.
On Wed, May 31, 2017 at 01:16:36PM -0400, Nathan Sidwell wrote:
> On 05/31/2017 01:05 PM, Jakub Jelinek wrote:
>
> > That is weird, that sounds like a bug somewhere? Which testcase is it on?
>
> I think decomp25 & decomp7 & 8 (sorry, can't recall full name).
> I discovered them because my first
On 05/31/2017 01:05 PM, Jakub Jelinek wrote:
That is weird, that sounds like a bug somewhere? Which testcase is it on?
I think decomp25 & decomp7 & 8 (sorry, can't recall full name).
I discovered them because my first attempt had an assert that the
incoming base was the same as the stored on
In convert_nontype_argument to reference type we were inappropriately
checking value_dependent_expression_p on an expression that might be a
VAR_DECL and might be a TEMPLATE_PARM_INDEX of reference type. It's
inappropriate in the former case because we don't care about the value
of the object, onl
On Wed, May 31, 2017 at 12:54:24PM -0400, Nathan Sidwell wrote:
> This patch reworks how decl_decomposition is marked. With the new
> lang_decl_decomp struct, the selector grew another used but -- which
> affected me on the modules branch. however, we can use that use selector
> value in place of
This patch reworks how decl_decomposition is marked. With the new
lang_decl_decomp struct, the selector grew another used but -- which
affected me on the modules branch. however, we can use that use
selector value in place of the decompostion_p bitfield.
This patch makes that change, but it
On May 31, 2017 6:28:26 PM GMT+02:00, Jakub Jelinek wrote:
>On Wed, May 31, 2017 at 05:36:12PM +0200, Richard Biener wrote:
>> On May 31, 2017 5:10:04 PM GMT+02:00, Jakub Jelinek
> wrote:
>> >On Wed, May 31, 2017 at 10:20:51AM -0400, Aldy Hernandez wrote:
>> >> The biggest number of SSA_NAMEs I sa
I've committed this new testcase, from the modules branch. Something
that got fixed in the name-lookup change.
nathan
--
Nathan Sidwell
2017-05-31 Nathan Sidwell
* g++.dg/lookup/lambda1.C New.
Index: testsuite/g++.dg/lookup/lambda1.C
* Claudiu Zissulescu [2017-05-19 12:31:01
+0200]:
> From: Claudiu Zissulescu
>
> If the stack pointer is needed, emit a special barrier that will prevent
> the scheduler from moving stores to the frame before the stack adjustment.
>
> 2017-01-03 Claudiu Zissulescu
>
> * config/arc/a
On Wed, May 31, 2017 at 05:36:12PM +0200, Richard Biener wrote:
> On May 31, 2017 5:10:04 PM GMT+02:00, Jakub Jelinek wrote:
> >On Wed, May 31, 2017 at 10:20:51AM -0400, Aldy Hernandez wrote:
> >> The biggest number of SSA_NAMEs I saw was actually 472,225. Of
> >these,
> >> 357,032 were non-point
* Claudiu Zissulescu [2017-05-19 12:31:02
+0200]:
> arc_can_eliminate is using arc_frmae_pointer_required() which is wrong
> as the frame_pointer_needed can be set on different conditions. Fix it
> by calling arc_frame_pointer_needed().
>
> gcc/
> 2017-01-09 Claudiu Zissulescu
>
> * c
> Le 31 mai 2017 à 17:40, Dominique d'Humières a écrit :
>
> If I am not mistaken, compiling the following code with the patch applied
simpler test
print *,(huge(0),i=1,6)
! print*,(i,i=1,6)
! print*,(i,i=1,6,1)
end
>
> gives an ICE.
>
> TIA
>
> Dominique
If I am not mistaken, compiling the following code with the patch applied
program test_ivs
use iso_varying_string
implicit none
type(varying_string),dimension(:,:),allocatable :: array2d
type(varying_string) :: extra
integer :: i,j
allocate(array2d(2,3))
extra = "four"
array2d(
* Claudiu Zissulescu [2017-05-19 12:31:00
+0200]:
> gcc/
> 2016-12-20 Claudiu Zissulescu
>
> * config/arc/arc.md (commutative_binary_comparison): Remove 'I'
> constraint. It is not valid for the pattern.
> (noncommutative_binary_comparison): Likewise.
Looks good, thanks,
A
On 31/05/2017 14:11:43, Richard Sandiford wrote:
> Prakhar Bahuguna writes:
> > On 31/05/2017 09:19:40, Richard Sandiford wrote:
> >> const_ints are supposed to be stored in sign-extended form, so a 32-bit
> >> integer with the MSB set should be 0x8000|x instead of
> >> 0x8000|x.
On May 31, 2017 5:10:04 PM GMT+02:00, Jakub Jelinek wrote:
>On Wed, May 31, 2017 at 10:20:51AM -0400, Aldy Hernandez wrote:
>> The biggest number of SSA_NAMEs I saw was actually 472,225. Of
>these,
>> 357,032 were non-pointers, so could conceivably have range
>information. In
>> reality, 77,398 h
* Claudiu Zissulescu [2017-05-19 12:30:59
+0200]:
> From: Claudiu Zissulescu
>
> 2016-12-17 Claudiu Zissulescu
>
> * config/arc/simdext.md (movv2hi_insn): Change predicate to avoid
> scaled addresses.
Seems reasonable.
Thanks,
Andrew
> ---
> gcc/config/arc/simdext.md | 2 +
On Wed, May 31, 2017 at 12:33 PM, Senkevich, Andrew
wrote:
> Hi,
>
> attached patches are for renaming __builtin_ia32_kmov16 to
> __builtin_ia32_kmovw in GCC 5.* and 6.* branches since it was renamed in
> master.
> Bootstrapped and regtested on x86_64-linux-gnu.
>
> gcc/
> * config/i386/
On Wed, May 31, 2017 at 10:20:51AM -0400, Aldy Hernandez wrote:
> The biggest number of SSA_NAMEs I saw was actually 472,225. Of these,
> 357,032 were non-pointers, so could conceivably have range information. In
> reality, 77,398 had range information, so 16% of all pointer and non-pointer
> SSA_
* Claudiu Zissulescu [2017-05-19 12:30:58
+0200]:
> gcc/
> 2016-12-12 Claudiu Zissulescu
>
> * config/arc/arc.c (arc_conditional_register_usage): Allow r30 to
> be used by the reg-alloc.
Looks good, thanks,
Andrew
> ---
> gcc/config/arc/arc.c | 9 -
> gcc/config/arc/
* Claudiu Zissulescu [2017-05-19 12:30:57
+0200]:
> gcc/
> 2017-04-10 Claudiu Zissulescu
>
> * config/arc/arc.md (mulsi3): Avoid use of hard registers before
> reg-alloc when having mul64 or mul32x16 instructions.
> (mulsidi3): Likewise.
> (umulsidi3): Likewise.
>
On 31 May 2017 at 16:27, Robin Dapp wrote:
>> Since this commit (r248678), I've noticed regressions on some arm targets.
>> Executed from: gcc.dg/tree-ssa/tree-ssa.exp
>> gcc.dg/tree-ssa/gen-vect-26.c scan-tree-dump-times vect "Alignment
>> of access forced using peeling" 1
>> gcc.dg/tre
> Since this commit (r248678), I've noticed regressions on some arm targets.
> Executed from: gcc.dg/tree-ssa/tree-ssa.exp
> gcc.dg/tree-ssa/gen-vect-26.c scan-tree-dump-times vect "Alignment
> of access forced using peeling" 1
> gcc.dg/tree-ssa/gen-vect-26.c scan-tree-dump-times vect
> "
On 05/23/2017 08:11 AM, Jakub Jelinek wrote:
On Tue, May 23, 2017 at 06:48:15AM -0400, Aldy Hernandez wrote:
[ughh, one more time, but CCing the list.]
Sorry, for the delayed response. I was fighting with Firefox + LTO to
gather some data :).
I'm worried a lot here about compile time memo
On 05/31/2017 03:31 PM, Richard Biener wrote:
> On Wed, May 31, 2017 at 2:28 PM, Martin Liška wrote:
>> On 05/31/2017 02:04 PM, Richard Biener wrote:
>>> On Wed, May 31, 2017 at 1:51 PM, Jakub Jelinek wrote:
On Wed, May 31, 2017 at 01:46:00PM +0200, Richard Biener wrote:
> Just wanting t
..adding missing patch
>From 3021b695a8111e1552176529ab3342cdd2ae3a43 Mon Sep 17 00:00:00 2001
From: marxin
Date: Wed, 3 May 2017 11:42:41 +0200
Subject: [PATCH] Doxygen: add default location for filters and output folder.
contrib/ChangeLog:
2017-05-03 Martin Liska
* gcc.doxy: Add default l
Hi.
This is patch which removes legacy perl scripts and set default values
for both OUTPUT_DIRECTORY and INPUT_FILTER.
Ready for trunk?
Thanks,
Martin
So I've come back to PR66313 and found a solution to the tailrecursion
missed optimization when fixing the factoring folding to use an unsigned
type when we're not sure of overflow.
The folding part is identical to my last try from 2015, the tailrecursion
part makes us handle intermittent stmts t
Hello.
After discussion with Richi, he approved to install patches separately
to current perl scripts. I'm attaching these patches and I will send patch
that will remove the legacy Perl scripts. The patch will be subject for
normal review process.
Thanks
Martin
>From 818d9da7892bcdb70df6fb456f7ea
On Wed, May 31, 2017 at 3:02 PM, Richard Biener
wrote:
> On Wed, May 31, 2017 at 3:56 PM, Will Schmidt
> wrote:
>> On Tue, 2017-05-30 at 09:00 +0200, Richard Biener wrote:
>>> On Mon, May 29, 2017 at 2:21 PM, Segher Boessenkool
>>> wrote:
>>> > On Mon, May 29, 2017 at 01:35:22PM +0200, Richard
On 31.05.2017 11:00, Jakub Jelinek wrote:
On Wed, May 31, 2017 at 10:48:07AM +0200, Georg-Johann Lay wrote:
because divmod in not a single_set:
(gdb) p seq
$10 = (const rtx_insn *) 0x7730d500
(gdb) pr
warning: Expression is not an assignment (and might have no effect)
(insn 14 13 0 (parallel
On Wed, May 31, 2017 at 3:56 PM, Will Schmidt wrote:
> On Tue, 2017-05-30 at 09:00 +0200, Richard Biener wrote:
>> On Mon, May 29, 2017 at 2:21 PM, Segher Boessenkool
>> wrote:
>> > On Mon, May 29, 2017 at 01:35:22PM +0200, Richard Biener wrote:
>> >> >> What's the documented behavior for vec_abs
On Tue, 2017-05-30 at 09:00 +0200, Richard Biener wrote:
> On Mon, May 29, 2017 at 2:21 PM, Segher Boessenkool
> wrote:
> > On Mon, May 29, 2017 at 01:35:22PM +0200, Richard Biener wrote:
> >> >> What's the documented behavior for vec_abs with respect to an
> >> >argument
> >> >> of value INT_MIN?
Hi,
On 23 May 2017 at 17:59, Robin Dapp wrote:
> gcc/ChangeLog:
>
> 2017-05-23 Robin Dapp
>
> * tree-vect-data-refs.c (vect_get_data_access_cost):
> Workaround for SLP handling.
> (vect_enhance_data_refs_alignment):
> Compute costs for doing no peeling at all, c
On Wed, May 31, 2017 at 3:31 PM, Martin Liška wrote:
> Hi.
>
> One has to set stage2_werror_flags in action-if-{not,}-given
> in order to properly respect the configure option.
>
> Ready to be installed?
Ok.
Richard.
> Martin
On Wed, May 31, 2017 at 2:43 PM, Bin.Cheng wrote:
> On Fri, May 26, 2017 at 12:49 PM, Richard Biener
> wrote:
>> On Thu, May 25, 2017 at 8:00 PM, Bin Cheng wrote:
>>> Hi,
>>> I believe this tests has been wrongly modified previously. It is to test
>>> that the exit check on
>>> pointer shouldn
On Wed, May 31, 2017 at 2:28 PM, Martin Liška wrote:
> On 05/31/2017 02:04 PM, Richard Biener wrote:
>> On Wed, May 31, 2017 at 1:51 PM, Jakub Jelinek wrote:
>>> On Wed, May 31, 2017 at 01:46:00PM +0200, Richard Biener wrote:
Just wanting to add that "ab-"using options/variables to implement
Hi.
One has to set stage2_werror_flags in action-if-{not,}-given
in order to properly respect the configure option.
Ready to be installed?
Martin
>From 77244d330010b2a17ca81fc866e2904e2f3fece0 Mon Sep 17 00:00:00 2001
From: marxin
Date: Wed, 31 May 2017 15:27:05 +0200
Subject: [PATCH] Fix config
Ping.
(Sorry for the very aggressive ping; this fixes 764 testsuite failures
on powerpc-linux).
Segher
On Sun, May 28, 2017 at 12:31:12PM +, Segher Boessenkool wrote:
> __atomic_add_fetch adds a value to some memory, and returns the result.
> If there is no direct support for this, expand_
Prakhar Bahuguna writes:
> On 31/05/2017 09:19:40, Richard Sandiford wrote:
>> const_ints are supposed to be stored in sign-extended form, so a 32-bit
>> integer with the MSB set should be 0x8000|x instead of
>> 0x8000|x. It's a bug if you have one where that isn't true.
>>
>> In
On 31/05/2017 09:19:40, Richard Sandiford wrote:
> const_ints are supposed to be stored in sign-extended form, so a 32-bit
> integer with the MSB set should be 0x8000|x instead of
> 0x8000|x. It's a bug if you have one where that isn't true.
>
> In the patch it looks like this cou
vector does addressof(*ptr) where ptr is returned by
allocate(n), but if n==0 that pointer might not be dereferencable.
While testing the fix I also found some bugs in the
__gnu_test::PointerBase helper that needed correcting.
PR libstdc++/80893
* include/bits/stl_bvector.h (vect
On Fri, May 26, 2017 at 12:49 PM, Richard Biener
wrote:
> On Thu, May 25, 2017 at 8:00 PM, Bin Cheng wrote:
>> Hi,
>> I believe this tests has been wrongly modified previously. It is to test
>> that the exit check on
>> pointer shouldn't be replaced by integer IV. Somehow GCC starts replacing
On 04/28/2017 02:03 PM, Martin Liška wrote:
> You were not brave enough to port remaining pattern in
> contrib/filter_knr2ansi.pl,
> right :) ?
Well. It shows the script just screws up many places and I bet we don't have
many KNR2 declarations
(if any). I'm attaching diff how it transforms input
On 05/31/2017 02:04 PM, Richard Biener wrote:
> On Wed, May 31, 2017 at 1:51 PM, Jakub Jelinek wrote:
>> On Wed, May 31, 2017 at 01:46:00PM +0200, Richard Biener wrote:
>>> Just wanting to add that "ab-"using options/variables to implement
>>> what are really
>>> function attributes doesn't look v
On 05/31/2017 02:06 PM, Richard Biener wrote:
> On Wed, May 31, 2017 at 2:01 PM, Jakub Jelinek wrote:
>> On Wed, May 31, 2017 at 01:57:48PM +0200, Martin Liška wrote:
>>> On 05/31/2017 01:51 PM, Jakub Jelinek wrote:
On Wed, May 31, 2017 at 01:46:00PM +0200, Richard Biener wrote:
> Just wa
Approved by Ilya in the PR.
Bootstrapped / tested on x86_64-unknown-linux-gnu, applied.
Richard.
2017-05-31 Richard Biener
PR target/80880
* config/i386/i386.c (ix86_expand_builtin): Remove assert
for arg being an SSA name when expanding IX86_BUILTIN_BNDRET.
On Wed, May 31, 2017 at 2:01 PM, Jakub Jelinek wrote:
> On Wed, May 31, 2017 at 01:57:48PM +0200, Martin Liška wrote:
>> On 05/31/2017 01:51 PM, Jakub Jelinek wrote:
>> > On Wed, May 31, 2017 at 01:46:00PM +0200, Richard Biener wrote:
>> >> Just wanting to add that "ab-"using options/variables to
On Wed, May 31, 2017 at 1:51 PM, Jakub Jelinek wrote:
> On Wed, May 31, 2017 at 01:46:00PM +0200, Richard Biener wrote:
>> Just wanting to add that "ab-"using options/variables to implement
>> what are really
>> function attributes doesn't look very clean. Unless the plan is to get rid
>> of
>>
Hi,
this one appears to be a rather simple case of missing diagnostic: in
convert_to_void we aren't calling maybe_warn_nodiscard when we strip an
INDIRECT_REF wrapping a CALL_EXPR thus we don't issue the diagnostic
that we normally provide for plain CALL_EXPRs (eg, for a func returning
a plai
On Wed, May 31, 2017 at 01:57:48PM +0200, Martin Liška wrote:
> On 05/31/2017 01:51 PM, Jakub Jelinek wrote:
> > On Wed, May 31, 2017 at 01:46:00PM +0200, Richard Biener wrote:
> >> Just wanting to add that "ab-"using options/variables to implement
> >> what are really
> >> function attributes does
The Solaris linker recently gained sysroot support. The following patch
enables that, although there isn't much to do:
* Until recently, ld --help output went to stderr, not being caught by
gcc/configure's tests which only checked stdout. However, older ld
versions still differ here and libt
On 05/31/2017 01:51 PM, Jakub Jelinek wrote:
> On Wed, May 31, 2017 at 01:46:00PM +0200, Richard Biener wrote:
>> Just wanting to add that "ab-"using options/variables to implement
>> what are really
>> function attributes doesn't look very clean. Unless the plan is to get rid
>> of
>> function a
On 05/31/2017 01:46 PM, Richard Biener wrote:
> On Wed, May 31, 2017 at 1:33 PM, Jakub Jelinek wrote:
>> On Wed, May 31, 2017 at 01:24:47PM +0200, Martin Liška wrote:
>>> On 05/31/2017 10:35 AM, Jakub Jelinek wrote:
On Wed, May 31, 2017 at 10:04:53AM +0200, Martin Liška wrote:
> diff --gi
On Wed, May 31, 2017 at 01:46:00PM +0200, Richard Biener wrote:
> Just wanting to add that "ab-"using options/variables to implement
> what are really
> function attributes doesn't look very clean. Unless the plan is to get rid of
> function attributes in favor of per-function options.
Function a
On Wed, May 31, 2017 at 1:33 PM, Jakub Jelinek wrote:
> On Wed, May 31, 2017 at 01:24:47PM +0200, Martin Liška wrote:
>> On 05/31/2017 10:35 AM, Jakub Jelinek wrote:
>> > On Wed, May 31, 2017 at 10:04:53AM +0200, Martin Liška wrote:
>> >> diff --git a/gcc/common.opt b/gcc/common.opt
>> >> index 13
Hello.
Installing as obvious as it only touches comment.
Martin
>From 4b0eebe5accdc7aa0782acccdd61a151c0a48378 Mon Sep 17 00:00:00 2001
From: marxin
Date: Wed, 31 May 2017 13:35:41 +0200
Subject: [PATCH] Fix typo in a comment in cpuid.h (PR target/79155).
gcc/ChangeLog:
2017-05-31 Martin Lisk
On Wed, May 31, 2017 at 01:24:47PM +0200, Martin Liška wrote:
> On 05/31/2017 10:35 AM, Jakub Jelinek wrote:
> > On Wed, May 31, 2017 at 10:04:53AM +0200, Martin Liška wrote:
> >> diff --git a/gcc/common.opt b/gcc/common.opt
> >> index 13305558d2d..5e9942d5100 100644
> >> --- a/gcc/common.opt
> >>
On 05/31/2017 10:31 AM, Alexander Monakov wrote:
> On Wed, 31 May 2017, Martin Liška wrote:
>> I added to common.opt:
>> Common RejectNegative Joined UInteger Var(flag_no_sanitize_fn) PerFunction
>> No sanitize flags for a function
>
> This needs a period at the end ("for a function.").
Ah, I see
On 05/31/2017 10:35 AM, Jakub Jelinek wrote:
> On Wed, May 31, 2017 at 10:04:53AM +0200, Martin Liška wrote:
>> diff --git a/gcc/common.opt b/gcc/common.opt
>> index 13305558d2d..5e9942d5100 100644
>> --- a/gcc/common.opt
>> +++ b/gcc/common.opt
>> @@ -222,9 +222,13 @@ bool flag_opts_finished
>> V
On Mon, 29 May 2017, Yuri Gribov wrote:
> This no longer fixes the PR but still works in some cases as
> demonstrated by the test. So I decided to keep it.
As Richard noticed you don't need widest_ints but can use wide_ints.
Please use == 0 instead of ! on wide-ints as well.
+(for cmp (le gt)
+
On Mon, 29 May 2017, Yuri Gribov wrote:
> This improve VRP handling for bitfield comparisons added by previous patch.
>
> -I
+is_masked_range_test (tree name, tree valt, enum tree_code cond_code,
bool is_else_edge,
+ tree *new_name,
long line
+ wide_int mask = maskt,
+
On 28/05/17 22:13 +0200, François Dumont wrote:
Sure but like freedom which stop where start others' freedom so does
those requirements :-). Because the Standard says that an allocator
will be value-init when there is no default-init it makes usage of the
C++11 default constructor more complica
On 29/05/17 22:55 +0200, François Dumont wrote:
Hi
It wasn't such a big deal to restore value-init of the allocator.
So here is the updated patch.
I used:
_Bvector_impl() _GLIBCXX_NOEXCEPT_IF( noexcept(_Bit_alloc_type()) )
rather than using is_nothrow_default_constructible. Any
On 31 May 11:38, Kirill Yukhin wrote:
> Hello Julia,
> On 26 May 09:13, Koval, Julia wrote:
> > Hi,
> > This patch fixes these PR's. Ok for trunk?
> >
> > gcc/
> > * config/i386/subst.md (round): Fix round pattern.
> > * config/i386/i386.c (ix86_erase_embedded_rounding):
> > Fix erasin
Hi,
attached patches are for renaming __builtin_ia32_kmov16 to __builtin_ia32_kmovw
in GCC 5.* and 6.* branches since it was renamed in master.
Bootstrapped and regtested on x86_64-linux-gnu.
gcc/
* config/i386/i386.c (__builtin_ia32_kmovw): Renamed from
__builtin_ia32_kmov16 sin
We have decided to apply the following patch to ARM/embedded-7-branch and
ARM/embedded-6-branch to enable Purecode support for ARMv8-M Baseline targets.
ChangeLog:
2017-05-31 Prakhar Bahuguna
Backport from mainline
2017-05-04 Prakhar Bahuguna
Andre Simoe
On 30.05.2017 16:32, Mike Stump wrote:
> On May 28, 2017, at 2:16 PM, Iain Buclaw wrote:
>>
>> This patch adds D language support to the GCC test suite.
>
> Ok. If you could ensure that gcc without D retains all it's goodness and
> that gcc with D works on 2 different systems, that will help en
On 05/31/2017 09:34 AM, Richard Biener wrote:
Actually for the bigger picture I'd refactor rest_of_decl_compilation, not
calling it from the frontends but rely on finalize_decl/function. The missing
part would then be calling the dwarf hook which should eventually be done
at some of the places t
On Wed, May 31, 2017 at 10:48:07AM +0200, Georg-Johann Lay wrote:
> > > because divmod in not a single_set:
> > > (gdb) p seq
> > > $10 = (const rtx_insn *) 0x7730d500
> > > (gdb) pr
> > > warning: Expression is not an assignment (and might have no effect)
> > > (insn 14 13 0 (parallel [
> > >
On 31.05.2017 10:15, Jakub Jelinek wrote:
On Wed, May 31, 2017 at 10:06:34AM +0200, Georg-Johann Lay wrote:
Hi, this causes a performance degradation for avr.
When optimizing for speed, and with a known denominatior, then v6 uses
s/umulMM3_highpart insn to avoid division because no div instruct
Hello Julia,
On 26 May 09:13, Koval, Julia wrote:
> Hi,
> This patch fixes these PR's. Ok for trunk?
>
> gcc/
> * config/i386/subst.md (round): Fix round pattern.
> * config/i386/i386.c (ix86_erase_embedded_rounding):
> Fix erasing rounding for the fixed pattern.
>
> Thanks,
> J
On Wed, May 31, 2017 at 10:04:53AM +0200, Martin Liška wrote:
> diff --git a/gcc/common.opt b/gcc/common.opt
> index 13305558d2d..5e9942d5100 100644
> --- a/gcc/common.opt
> +++ b/gcc/common.opt
> @@ -222,9 +222,13 @@ bool flag_opts_finished
> Variable
> unsigned int flag_sanitize
>
> +###
> +C
On Wed, 31 May 2017, Martin Liška wrote:
> I added to common.opt:
> Common RejectNegative Joined UInteger Var(flag_no_sanitize_fn) PerFunction
> No sanitize flags for a function
This needs a period at the end ("for a function.").
> FAIL: compiler driver --help=optimizers option(s): "^ +-.*[^:.]$"
I have added myself to the Write After Approval section of the MAINTAINERS
list.
ChangeLog:
2017-05-31 Prakhar Bahuguna
* MAINTAINERS: Add self to Write After Approval
--
Prakhar Bahuguna
Prakhar Bahuguna writes:
> This patch tackles the issue reported in PR71607. This patch takes a different
> approach for disabling the creation of literal pools. Instead of disabling the
> patterns that would normally transform the rtl into actual literal pools, it
> disables the creation of this
On Wed, May 31, 2017 at 10:06:34AM +0200, Georg-Johann Lay wrote:
> Hi, this causes a performance degradation for avr.
>
> When optimizing for speed, and with a known denominatior, then v6 uses
> s/umulMM3_highpart insn to avoid division because no div instruction is
> available.
>
> unsigned sca
1 - 100 of 108 matches
Mail list logo
