On Thu, Oct 26, 2017 at 11:48 PM, Jakub Jelinek wrote:
> Hi!
>
> The following testcase is miscompiled, because due to STV we end up
> with a load from V2DFmode MEM for which get_pool_entry returns
> a V1TImode vector. maybe_get_pool_constant doesn't have code
> to handle mode differences between
Martin Sebor writes:
> On 10/26/2017 11:52 AM, Richard Sandiford wrote:
>> Martin Sebor writes:
>>> For offset_int the default precision is 128-bits. Making that
>>> the default also for wide_int should be unsurprising.
>>
>> I think it'd be surprising. offset_int should always be used in
>> pr
On Thu, Oct 26, 2017 at 9:37 PM, Eric Botcazou wrote:
>> Can you figure what oldest GCC release supports the C++11/14 POD handling
>> that would be required?
>
> GCC needs to be buildable by other compilers than itself though.
There's always the possibility of building GCC 4.8 with the other comp
On 10/26/2017 06:30 PM, Sandra Loosemore wrote:
On 10/26/2017 01:21 AM, Martin Liška wrote:
On 10/20/2017 06:03 AM, Sandra Loosemore wrote:
On 10/19/2017 12:26 PM, Eric Gallager wrote:
On 10/19/17, Martin Liška wrote:
Hi.
As discussed in the PR, we should be more precise in our documentatio
On Thu, Oct 26, 2017 at 9:43 PM, Jakub Jelinek wrote:
> On Thu, Oct 26, 2017 at 02:43:55PM +0200, Richard Biener wrote:
>> On Thu, Oct 26, 2017 at 2:18 PM, Richard Sandiford
>> wrote:
>> > Richard Biener writes:
>> >> On Mon, Oct 23, 2017 at 1:22 PM, Richard Sandiford
>> >> wrote:
>> >>> This p
On Fri, Oct 27, 2017 at 10:35:56AM +0200, Richard Biener wrote:
> > I think it is too early for that, we aren't LLVM or Rust that don't really
> > care about what build requirements they impose on users.
>
> That's true, which is why I asked. For me requiring sth newer than GCC 4.8
> would be a b
On Fri, Oct 27, 2017 at 12:12 AM, Jim Wilson wrote:
> On Thu, 2017-10-26 at 11:38 +0200, Richard Biener wrote:
>> You can eventually keep the option, marking it as Ignore (like we do
>> for options we remove but "keep" for backward compatibility). The
>> diagnostic (as warning, given the option w
On Thu, Oct 26, 2017 at 04:19:35PM +0100, James Greenhalgh wrote:
> On Tue, Jul 25, 2017 at 02:58:04PM +0100, Wilco Dijkstra wrote:
> > This patch makes some changes to the frame layout in order to simplify
> > stack probing. We want to use the save of LR as a probe in any non-leaf
> > function.
> There's always the possibility of building GCC 4.8 with the other compiler
> and then GCC 9+ (?) with GCC 4.8.
What an user-friendly solution...
> What's the list of other compilers people routinely use? I see various
> comments on other compilers in install.texi but those are already saying
>
Hi,
today I noticed once more that the function pointlessly still returns an
int instead of a bool. Unless somebody complaints, I'm going to apply
the below. Tested x86_64-linux.
Thanks, Paolo.
2017-10-27 Paolo Carlini
* pt.c (invalid_nontype_parm_type_p): R
This is my attempt at the empty class ABI change. To recap quickly, the C++
compiler has used a different calling convention for passing empty classes,
because C++ says they have size 1, while the GCC C extension gives them size 0.
But this difference doesn't mean that they need to be passed diffe
On 10/27/2017 09:35 AM, Richard Biener wrote:
> On Thu, Oct 26, 2017 at 9:43 PM, Jakub Jelinek wrote:
>> On Thu, Oct 26, 2017 at 02:43:55PM +0200, Richard Biener wrote:
>>> Can you figure what oldest GCC release supports the C++11/14 POD handling
>>> that would be required?
>>
>> I think it is to
Hi all,
as noted on IRC is one of the error message in check.c co_reduce misleading.
The attached patch fixes this. The intention of the error message is to tell
that the type of the argument bound to parameter A is of wrong type and not
that an unspecific argument has a wrong type.
If no one obj
On Fri, 27 Oct 2017, Marek Polacek wrote:
> This is my attempt at the empty class ABI change. To recap quickly, the C++
> compiler has used a different calling convention for passing empty classes,
> because C++ says they have size 1, while the GCC C extension gives them size
> 0.
> But this dif
On Fri, Oct 27, 2017 at 12:31:46PM +0200, Richard Biener wrote:
> I fear it doesn't work at all with LTO (you'll always get the old ABI
> if I read the patch correctly). This is because the function
> computing the size looks at flag_abi_version which isn't saved
> per function / TU.
>
> Similarl
Hi,
while looking for x86 tuning issues I noticed PR81614 about partial register
stalls on core. We currently support two schemes for our of order CPUs -
partial register dependencies where registers are renamed always as whole
and thus it is important to always write complete register at the begg
On Fri, 27 Oct 2017, Jakub Jelinek wrote:
> On Fri, Oct 27, 2017 at 12:31:46PM +0200, Richard Biener wrote:
> > I fear it doesn't work at all with LTO (you'll always get the old ABI
> > if I read the patch correctly). This is because the function
> > computing the size looks at flag_abi_version w
Hello.
This is small improvement that can catch a virtual call after a lifetime
scope of an object.
Patch can bootstrap on ppc64le-redhat-linux and survives regression tests.
Ready to be installed?
Martin
gcc/cp/ChangeLog:
2017-10-27 Martin Liska
* decl.c (begin_destructor_body):
On 25 October 2017 at 20:44, Jan Hubicka wrote:
>> On 24 October 2017 at 16:26, Jan Hubicka wrote:
>> >> 2017-10-13 Prathamesh Kulkarni
>> >>
>> >> * cgraph.h (set_malloc_flag): Declare.
>> >> * cgraph.c (set_malloc_flag_1): New function.
>> >> (set_malloc_flag): Likewise.
>>
On Fri, Oct 27, 2017 at 12:47:12PM +0200, Martin Liška wrote:
> Hello.
>
> This is small improvement that can catch a virtual call after a lifetime
> scope of an object.
>
>
> Patch can bootstrap on ppc64le-redhat-linux and survives regression tests.
>
> Ready to be installed?
The decl.c chang
Applied the attached patch to changes.html.
Thanks,
Prathamesh
* changes.html: Document change that ipa-pure-const is extended to
propagate malloc attribute.
Index: changes.html
===
RCS file: /cvs/gcc/wwwdocs/htdocs/gcc-8/ch
On 10/27/2017 12:52 PM, Jakub Jelinek wrote:
The decl.c change seems to be only incremental change from a not publicly
posted patch rather than the full diff against trunk.
Sorry for that. Sending full patch.
Martin
>From df0cc0c2da18b150b1f0fbef418450a223470d7f Mon Sep 17 00:00:00 2001
From:
On Fri, Oct 27, 2017 at 01:16:08PM +0200, Martin Liška wrote:
> On 10/27/2017 12:52 PM, Jakub Jelinek wrote:
> > The decl.c change seems to be only incremental change from a not publicly
> > posted patch rather than the full diff against trunk.
>
> Sorry for that. Sending full patch.
Thanks.
> -
I happened to notice that install.texi still contains references to the
Tru64 UNIX and IRIX ports I've removed in GCC 4.8. I believe it's time
now to get rid of those completely.
Tested with make doc/gccinstall.info and doc/gccinstall.pdf. Ok for
mainline? This falls under my prior maintainersh
There's one Solaris 12 reference in install.texi that also needs
adapting for the Solaris 12 -> 11.4 change. However, instead of simply
updating it I chose to update the relevant sections for current binutils
versions.
Tested with make doc/gccinstall.info doc/gccinstall.pdf and visually
inspectin
This enum doesn't do anything for POSIX systems, so this is a pretty
uninteresting change.
* include/bits/fs_path.h (path::format): Define new enumeration type.
(path(string_type&&), path(const Source&))
(path(InputIterator, InputIterator))
(path(const Source&, con
Prathamesh
> > OK, thanks!
> Thanks, committed as r254140 after following validation:
> 1] Bootstrap+test with --enable-languages=all,ada,go on
> x86_64-unknown-linux-gnu and ppc64le-linux-gnu.
> 2] LTO bootstrap+test on x86_64-unknown-linux-gnu and ppc64le-linux-gnu
> 3] Cross tested on arm*-*-* a
> On Thu, Oct 26, 2017 at 2:55 PM, Jan Hubicka wrote:
> >> I think the limit should be on the number of generated copies and not
> >> the overall size of the structure... If the struct were composed of
> >> 32 individual chars we wouldn't want to emit 32 loads and 32 stores...
> >>
> >> I wonder
> On 25 October 2017 at 20:44, Jan Hubicka wrote:
> >> On 24 October 2017 at 16:26, Jan Hubicka wrote:
> >> >> 2017-10-13 Prathamesh Kulkarni
> >> >>
> >> >> * cgraph.h (set_malloc_flag): Declare.
> >> >> * cgraph.c (set_malloc_flag_1): New function.
> >> >> (set_malloc_flag)
This implements part of LWG 2676. I haven't added the new members
taking wide character strings, because they're only needed on Windows,
where the Filesystem library doesn't work yet. I'll send a follow-up
patch about those overloads.
I've implemented these new functions as templates, which means
From: Siddhesh Poyarekar
This patch adds an mcpu option for the Qualcomm saphira server part.
Tested on aarch64 and did not find any regressions resulting from this
patch.
Siddhesh
2017-10-27 Siddhesh Poyarekar
Jim Wilson
gcc/
* config/aarch64/a
On Fri, 27 Oct 2017, Jan Hubicka wrote:
> > On 25 October 2017 at 20:44, Jan Hubicka wrote:
> > >> On 24 October 2017 at 16:26, Jan Hubicka wrote:
> > >> >> 2017-10-13 Prathamesh Kulkarni
> > >> >>
> > >> >> * cgraph.h (set_malloc_flag): Declare.
> > >> >> * cgraph.c (set_malloc_f
On 27/10/17 13:43 +0100, Jonathan Wakely wrote:
This implements part of LWG 2676. I haven't added the new members
taking wide character strings, because they're only needed on Windows,
where the Filesystem library doesn't work yet. I'll send a follow-up
patch about those overloads.
This patch s
Andrew Pinski wrote:
> I think this goes against what most folks are used to. I know you are
> saying most folks are used to a compiler defaulting to optimizations
> on but I don't think that is true. In fact GCC has been this way
> since day one.
Well it may depend which part of the industry y
On 10/25/2017 02:19 PM, Markus Trippelsdorf wrote:
On 2017.08.30 at 11:45 +0200, Martin Liška wrote:
diff --git a/Makefile.in b/Makefile.in
index 78db0982ba2..16b76906ad0 100644
--- a/Makefile.in
+++ b/Makefile.in
@@ -529,13 +529,14 @@ STAGE1_CONFIGURE_FLAGS = --disable-intermodule
$(STAGE1_CHE
Hello.
So eventually it looks needed fix is much simpler, one just need to rename a
folder.
Patch can bootstrap on ppc64le-redhat-linux and survives regression tests.
I'm going to install it.
Martin
>From 028b5cc3865bdc77c185172de84627e140030303 Mon Sep 17 00:00:00 2001
From: marxin
Date: Fri
Hello.
It's improvement that I consider still useful even though we're not going to use
it for profiled bootstrap.
Patch can bootstrap on ppc64le-redhat-linux and survives regression tests.
Ready for trunk?
Thanks,
Martin
>From 1a32e0b41b291bef3d58126754834f6c41148ace Mon Sep 17 00:00:00 2001
F
Hi,
rather embarrasingly, I found out that there is a missing condition to
make sure that HSA grid size is zero when the OpenMP loop bounds
should preclude the loop from executing at all. I do not know whether
I lost is somewhere when preparing patches for trunk or whether I
forgot about it from
This series of patches are the AArch64 changes needed before SVE
support goes in. It's based on top of:
https://gcc.gnu.org/ml/gcc-patches/2017-09/msg01163.html
and Jeff's latest stach-clash protection changes.
Series tested on aarch64-linux-gnu.
Richard
This patch replaces switch statements that call specific generator
functions with code that constructs the rtl pattern directly.
This seemed to scale better to SVE and also seems less error-prone.
As a side-effect, the patch fixes the REV handling for diff==1,
vmode==E_V4HFmode and adds missing su
This patch simply moves code around, in order to make the later
patches easier to read, and to avoid forward declarations.
It doesn't add the missing function comments because the interfaces
will change in a later patch.
2017-10-26 Richard Sandiford
Alan Hayward
David
The port had aarch64_add_offset and aarch64_add_constant routines
that did similar things. This patch replaces them with an expanded
version of aarch64_add_offset that takes separate source and
destination registers. The new routine also takes a poly_int64 offset
instead of a HOST_WIDE_INT offset
The SVE port uses the public constraints "Upl" and "Upa" to mean
"low predicate register" and "any predicate register" respectively.
"Upl" was already used as an internal-only constraint by the
addition patterns, so this patch renames it to "Uaa" ("two adds
needed").
2017-10-27 Richard Sandiford
This patch reworks aarch64_simd_valid_immediate so that
it's easier to add SVE support. The main changes are:
- make simd_immediate_info easier to construct
- replace the while (1) { ... break; } blocks with checks that use
the full 64-bit value of the constant
- treat floating-point modes as i
Later patches turn the number of vector units into a poly_int.
We deliberately don't support applying GEN_INT to those (except
in target code that doesn't disguish between poly_ints and normal
constants); gen_int_mode needs to be used instead.
This patch therefore replaces instances of:
GEN_INT
This patch passes the number of units to aarch64_reverse_mask,
which avoids a to_constant () once GET_MODE_NUNITS is variable.
2017-10-26 Richard Sandiford
Alan Hayward
David Sherwood
gcc/
* config/aarch64/aarch64-protos.h (aarch64_reverse_mask): Take
This patch passes the number of units to aarch64_simd_vect_par_cnst_half,
which avoids a to_constant () once GET_MODE_NUNITS is variable.
2017-10-27 Richard Sandiford
Alan Hayward
David Sherwood
gcc/
* config/aarch64/aarch64-protos.h (aarch64_simd_vect_par_
This patch passes the number of units to aarch64_expand_vec_perm
and aarch64_expand_vec_perm_const, which avoids a to_constant ()
once GET_MODE_NUNITS is variable.
2017-10-27 Richard Sandiford
Alan Hayward
David Sherwood
gcc/
* config/aarch64/aarch64-protos
This patch switches the AArch64 port to use 2 poly_int coefficients
and updates code as necessary to keep it compiling.
One potentially-significant change is to
aarch64_hard_regno_caller_save_mode. The old implementation
was written in a pretty conservative way: it changed the default
behaviour f
aarch64_rtx_costs uses the number of registers in a mode as the basis
of SET costs. This patch makes it get the number of registers from
aarch64_hard_regno_nregs rather than repeating the calcalation inline.
Handling SVE modes in aarch64_hard_regno_nregs is then enough to get
the correct SET cost
This patch records the integer value of the address offset in
aarch64_address_info, so that it doesn't need to be re-extracted
from the rtx. The SVE port will make more use of this. The patch
also uses poly_int64 routines to manipulate the offset, rather than
just handling CONST_INTs.
2017-10-2
On Fri, Oct 27, 2017 at 03:19:05PM +0200, Martin Jambor wrote:
> 2017-10-10 Martin Jambor
>
> * omp-grid.c (grid_attempt_target_gridification): Also insert a
> condition whether loop should be executed at all.
Ok, thanks.
> --- a/gcc/omp-grid.c
> +++ b/gcc/omp-grid.c
> @@ -1315,6
On 10/27/2017 01:26 PM, Jakub Jelinek wrote:
On Fri, Oct 27, 2017 at 01:16:08PM +0200, Martin Liška wrote:
On 10/27/2017 12:52 PM, Jakub Jelinek wrote:
The decl.c change seems to be only incremental change from a not publicly
posted patch rather than the full diff against trunk.
Sorry for tha
On Fri, Oct 27, 2017 at 03:48:41PM +0200, Martin Liška wrote:
> --- a/gcc/cp/decl.c
> +++ b/gcc/cp/decl.c
> @@ -14639,8 +14639,12 @@ implicit_default_ctor_p (tree fn)
> /* Clobber the contents of *this to let the back end know that the object
> storage is dead when we enter the constructor or
[ was: Re: [RFC PATCH] Coalesce host to device transfers in libgomp ]
On 10/25/2017 01:38 PM, Jakub Jelinek wrote:
And we don't really have the async target implemented yet for NVPTX:(,
guess that should be the highest priority after this optimization.
Hi,
how about this approach:
1 - Move asy
On Fri, Oct 27, 2017 at 12:19:02PM +0200, Andre Vehreschild wrote:
> Hi all,
>
> as noted on IRC is one of the error message in check.c co_reduce misleading.
> The attached patch fixes this. The intention of the error message is to tell
> that the type of the argument bound to parameter A is of wr
On 2017.10.27 at 15:03 +0200, Martin Liška wrote:
> > And BTW would it make sense to add -gtoggle to stage2 in bootstrap-lto?
>
> Why do you want to have it there? Am I right that we do not do a stage
> comparison with LTO bootstrap?
The idea was to trigger -g -flto at least during one stage, jus
GCC currently defaults to -fmath-errno. This generates code assuming math
functions set errno and the application checks errno. Very few applications
test errno and various systems and math libraries no longer set errno since it
is optional. GCC generates much faster code for simple math functio
On 10/26/2017 01:50 PM, Jakub Jelinek wrote:
> On Thu, Oct 26, 2017 at 05:12:40PM +, Wilco Dijkstra wrote:
>> GCC's default optimization level is -O0. Unfortunately unlike other
>> compilers,
>> GCC generates extremely inefficient code with -O0. It is almost unusable for
>> low-level debuggi
On 10/27/2017 02:35 AM, Richard Biener wrote:
> On Thu, Oct 26, 2017 at 9:43 PM, Jakub Jelinek wrote:
>> On Thu, Oct 26, 2017 at 02:43:55PM +0200, Richard Biener wrote:
>>> On Thu, Oct 26, 2017 at 2:18 PM, Richard Sandiford
>>> wrote:
Richard Biener writes:
> On Mon, Oct 23, 2017 at 1:2
Committed.
On Thu, 26 Oct 2017 09:45:07 PDT (-0700), Palmer Dabbelt wrote:
> The documentation for the "-mabi" argument on RISC-V was incorrect. We
> chose to treat this as a documentation bug rather than a code bug, and
> to make the documentation match what GCC currently does. In the
> process
On 10/23/2017 05:21 AM, Richard Sandiford wrote:
> This patch adds a way of treating certain kinds of CONST as unique,
> so that pointer equality is equivalent to value equality. For now it
> is restricted to VEC_DUPLICATE and VEC_SERIES, although the code to
> generate them remains in the else ar
Hi,
V3. :-)
[PATCH, rs6000] (v2) Gimple folding for vec_madd()
Add support for gimple folding of the vec_madd() (vector multiply-add)
intrinsics.
Renamed the define_insn of altivec_vmladduhm to fmav8hi4, Refreshed the
caller of gen_altivec_vmladduhm to call gen_fmav8hi, and updated the
rs60
Jeff Law writes:
> On 10/23/2017 05:21 AM, Richard Sandiford wrote:
>> This patch adds a way of treating certain kinds of CONST as unique,
>> so that pointer equality is equivalent to value equality. For now it
>> is restricted to VEC_DUPLICATE and VEC_SERIES, although the code to
>> generate the
Hi,
Update the vec-perm testcase to use 'long long' rather than 'long'. This was a
missed typo
from when i initially committed the test.
Credit given to Carl for noticing this one.
OK for trunk?
Thanks,
-Will
diff --git a/gcc/testsuite/gcc.target/powerpc/fold-vec-perm-longlong.c
On 10/26/2017 12:20 PM, Andre Vehreschild wrote:
> Hi Paul,
>
> Without having tested the patch, it looks reasonable to me. So ok from my
> side.
>
> - Andre
>
Seconded, thanks.
Jerry
Jeff Law writes:
> On 10/23/2017 05:16 AM, Richard Sandiford wrote:
>> This patch adds helper functions for generating constant and
>> non-constant vector duplicates. These routines help with SVE because
>> it is then easier to use:
>>
>>(const:M (vec_duplicate:M X))
>>
>> for a broadcast o
Jeff Law writes:
> On 10/23/2017 05:18 AM, Richard Sandiford wrote:
>> This patch allows (const ...) wrappers to be used for rtx vector
>> constants, as an alternative to const_vector. This is useful
>> for SVE, where the number of elements isn't known until runtime.
> Right. It's constant, but
On 10/26/2017 12:11 PM, Richard Biener wrote:
> On October 26, 2017 6:50:15 PM GMT+02:00, Jeff Law wrote:
[ Big snip ]
>>> Both patches look ok to me though it would be nice to
>>> do the actual composition with a comment that the
>>> lattices might be moved here (if all workers became
>>> membe
On 10/27/2017 05:48 AM, Rainer Orth wrote:
> I happened to notice that install.texi still contains references to the
> Tru64 UNIX and IRIX ports I've removed in GCC 4.8. I believe it's time
> now to get rid of those completely.
>
> Tested with make doc/gccinstall.info and doc/gccinstall.pdf. Ok
On 10/27/2017 04:38 AM, Jakub Jelinek wrote:
> On Fri, Oct 27, 2017 at 12:31:46PM +0200, Richard Biener wrote:
>> I fear it doesn't work at all with LTO (you'll always get the old ABI
>> if I read the patch correctly). This is because the function
>> computing the size looks at flag_abi_version wh
No existing glibc version defines math_errhandling based on
__NO_MATH_ERRNO__. I'd expect such a change to come with a glibc patch,
and indeed a GCC execution test of the value of math_errhandling to make
sure the compiler's behavior isn't contradicting what's declared by the
runtime libraries
On 10/27/2017 05:48 AM, Rainer Orth wrote:
I happened to notice that install.texi still contains references to the
Tru64 UNIX and IRIX ports I've removed in GCC 4.8. I believe it's time
now to get rid of those completely.
Tested with make doc/gccinstall.info and doc/gccinstall.pdf. Ok for
main
Hi Wilco,
Sorry for the delay.
On 16/10/17 12:30, Wilco Dijkstra wrote:
ping
From: Wilco Dijkstra
Sent: 17 January 2017 19:23
To: GCC Patches
Cc: nd; Kyrill Tkachov; Richard Earnshaw
Subject: [PATCH][ARM] Remove DImode expansions for 1-bit shifts
A left shift of 1 can always be do
Segher:
I still have issues with the .
> > +(define_mode_attr VSX_XXBR [(V16QI "q")
> > + (V8HI "h")
> > + (V4SI "w")
> > + (V4SF "w")
> > + (V2DF "d")
> > + (V2DI "d")
> > + (V1TI "q")]
Hi Wilco,
On 16/10/17 12:29, Wilco Dijkstra wrote:
ping
Kyrill Tkachov wrote:
On 14/12/16 16:37, Wilco Dijkstra wrote:
Merge the movdi_vfp_cortexa8 pattern into movdi_vfp and remove it to avoid
unnecessary duplication and repeating bugs like PR78439 due to changes being
applied only to
On Fri, Oct 27, 2017 at 11:51 AM, Will Schmidt
wrote:
> Hi,
> V3. :-)
>
> [PATCH, rs6000] (v2) Gimple folding for vec_madd()
>
> Add support for gimple folding of the vec_madd() (vector multiply-add)
> intrinsics.
> Renamed the define_insn of altivec_vmladduhm to fmav8hi4, Refreshed the
> caller
On 10/26/2017 05:03 PM, Jeff Law wrote:
On 10/18/2017 10:59 AM, Egeyar Bagcioglu wrote:
Hello,
Test case "guality.exp=nrv-1.c" fails on aarch64. Optimizations reorder
the instructions and cause the value of a variable to be checked before
its first assignment. The following patch is moving the
Prereq for eventually embedding range analysis into the sprintf warning
pass. The only thing that changed since the original from a few days
ago was the addition of FINAL OVERRIDE to the before_dom_children
override function.
Re-bootstrapped and regression tested on x86.
Installing on the trunk
On Fri, Oct 27, 2017 at 12:51:56PM -0400, David Edelsohn wrote:
> On Fri, Oct 27, 2017 at 11:51 AM, Will Schmidt
> wrote:
> > [PATCH, rs6000] (v2) Gimple folding for vec_madd()
> >
> > Add support for gimple folding of the vec_madd() (vector multiply-add)
> > intrinsics.
> > Renamed the define_ins
Also add tests for experimental::file_status.
* include/bits/fs_dir.h (file_status): Make default constructor
non-explicit (LWG 2787).
* testsuite/27_io/filesystem/file_status/1.cc: New test.
* testsuite/experimental/filesystem/file_status/1.cc: New test.
Tested p
These issues haven't been resolved yet, but the discussion is correct
that these operations allocate memory, so the proposed resolutions are
correct.
* include/bits/fs_dir.h (directory_iterator): Remove noexcept from
constructors and increment member (LWG 3013).
(recursive
I think this header was going to be called math_specfun.h but was
renamed at the last minute, without re-ordering it in the Makefile.
* include/Makefile.am: Put headers in alphabetical order.
* include/Makefile.in: Regenerate.
Tested powerpc64le-linux, committed to trunk.
commit
We can use auto return types and if constexpr to do this without
including .
* include/bits/node_handle.h (_Node_insert_return::get): Avoid
use of std::tie and std::get.
Tested powerpc64le-linux, committed to trunk.
commit 397d3b68eb53ff6b229ac777a05dff4ae842a19b
Author: Jonatha
On Fri, 2017-10-27 at 10:55 -0600, Jeff Law wrote:
> Prereq for eventually embedding range analysis into the sprintf
> warning
> pass. The only thing that changed since the original from a few days
> ago was the addition of FINAL OVERRIDE to the before_dom_children
> override function.
>
> Re-boo
On Fri, Oct 27, 2017 at 9:52 AM, Jakub Jelinek wrote:
> On Fri, Oct 27, 2017 at 03:48:41PM +0200, Martin Liška wrote:
>> --- a/gcc/cp/decl.c
>> +++ b/gcc/cp/decl.c
>> @@ -14639,8 +14639,12 @@ implicit_default_ctor_p (tree fn)
>> /* Clobber the contents of *this to let the back end know that the o
On Fri, Oct 27, 2017 at 02:10:10PM -0400, Jason Merrill wrote:
> On Fri, Oct 27, 2017 at 9:52 AM, Jakub Jelinek wrote:
> > On Fri, Oct 27, 2017 at 03:48:41PM +0200, Martin Liška wrote:
> >> --- a/gcc/cp/decl.c
> >> +++ b/gcc/cp/decl.c
> >> @@ -14639,8 +14639,12 @@ implicit_default_ctor_p (tree fn)
Hello!
As discussed in the PR, different modes of a FP compare RTX are not
strong enough to survive through RTL optimization passes. Attached
testcase was miscompiled due to combine changing the mode of FP
compare through SELECT_CC_MODE.
The solution, implemented in the attached patch, is to drop
On 10/27/2017 02:18 PM, Jakub Jelinek wrote:
On Fri, Oct 27, 2017 at 02:10:10PM -0400, Jason Merrill wrote:
If the point is to clear the vptr, why are you also clearing the rest
of the object?
Can there be multiple vptr pointers in the object or is there just one?
Even if there can be multip
On Fri, Oct 27, 2017 at 02:30:39PM -0400, Nathan Sidwell wrote:
> On 10/27/2017 02:18 PM, Jakub Jelinek wrote:
> > On Fri, Oct 27, 2017 at 02:10:10PM -0400, Jason Merrill wrote:
>
> > > If the point is to clear the vptr, why are you also clearing the rest
> > > of the object?
> >
> > Can there be
Hi!
The following patch attempts to improve store merging, for the time being
it still only optimizes constant stores to adjacent memory.
The biggest improvement is handling bitfields, it uses the get_bit_range
helper to find the bounds of what can be modified when modifying the bitfield
and inst
On 10/27/2017 02:34 PM, Jakub Jelinek wrote:
But when singly inheriting a polymorphic base and thus mapped to the same
vptr all but the last dtor will not be in charge, right?
Correct.
So, if using build_clobber_this for this, instead of clobbering what we
clobber we'd just clear the single
Hi!
I've backported following 5 patches to 7.x.
The PR81715 one, because it could be risky, is enabled only
for -fsanitize=kernel-address which it has been reported for
and where it causes major issues for the kernel during sanitization.
Bootstrapped/regtested on x86_64-linux and i686-linux, comm
On 10/27/2017 12:03 PM, David Malcolm wrote:
> On Fri, 2017-10-27 at 10:55 -0600, Jeff Law wrote:
>> Prereq for eventually embedding range analysis into the sprintf
>> warning
>> pass. The only thing that changed since the original from a few days
>> ago was the addition of FINAL OVERRIDE to the b
When I originally wrote this test I wasn't wasn't aware of the
--with-cpu configure option, so this change explicitly disables avx to
make sure we choose the sse implementation, even when --with-cpu
specifies an arch that has avx support.
OK for head?
gcc/testsuite/ChangeLog:
gcc.target/
On Fri, Oct 27, 2017 at 11:17 PM, Daniel Santos wrote:
> When I originally wrote this test I wasn't wasn't aware of the
> --with-cpu configure option, so this change explicitly disables avx to
> make sure we choose the sse implementation, even when --with-cpu
> specifies an arch that has avx suppo
C17, a bug-fix version of the C11 standard with DR resolutions
integrated, will soon go to ballot. This patch adds corresponding
options -std=c17, -std=gnu17 (new default version, replacing
-std=gnu11 as the default), -std=iso9899:2017. As a bug-fix version
of the standard, there is no need for f
Hello world,
this is a draft patch which interchanges the indices for FORALL and
DO CONCURRENT loops for cases like PR 82471, where code like
DO CONCURRENT( K=1:N, J=1:M, I=1:L)
C(I,J,K) = A(I,J,K) + B(I,J,K)
END DO
led to very poor code because of stride issues. Currently,
Graphite i
Hi Thomas,
In general, I like the idea. I have some minor suggestions below.
On Sat, Oct 28, 2017 at 12:03:58AM +0200, Thomas Koenig wrote:
> +/* Callback function to determine if an expression is the
> + corresponding variable. */
> +
> +static int
static bool
> +has_var (gfc_expr **e, i
The power9 (running PowerPC ISA 3.0) has a round to integer instruction
(XSRQPI) that does various flavors of round an IEEE 128-bit floating point to
integeral values. This patch adds the support to the machine independent
portion of the compiler, and adds the necessary support for ceilf128,
round
1 - 100 of 102 matches
Mail list logo
