On Mon, 29 Sep 2025 at 08:00 +0200, Luc Grosheintz wrote:
This commit adds the right padded layout as described in N5014, with
LWG4372 (dynamic padding value) and LWG4314 (move in operator()).
PR libstdc++/110352
libstdc++-v3/ChangeLog:
* include/std/mdspan (_RightPaddedIndices
On 10/8/25 12:46, Tomasz Kaminski wrote:
I have just landed the first 3 patches.
Thanks, and sorry about not seeing too many cases of line break
after the return type.
On Fri, Oct 3, 2025 at 2:44 PM Luc Grosheintz
wrote:
On 10/3/25 12:09 PM, Tomasz Kaminski wrote:
On Thu, Oct 2, 202
On Wed, Oct 8, 2025 at 10:37 AM Jonathan Wakely wrote:
> On Fri, 03 Oct 2025 at 12:48 +0200, Tomasz Kamiński wrote:
> >From: Luc Grosheintz
> >
> >The improvement is that in __index_type_cast, we don't need to check at
> >runtime if we know that _IndexType is smaller than _OIndexType.
> >
> >The
On Wed, 08 Oct 2025 at 21:21 +0100, Mike Crowe wrote:
On Wednesday 08 October 2025 at 12:27:06 +0100, Jonathan Wakely wrote:
A few months ago I was testing the patch below, which replaces all the
repeated code that creates timespec objects with a call to a single
function. That function avoids o
Hi,
I am fairly new myself, but I think I can answer your questions.
On 10/8/25 20:03, Yuxuan Chen wrote:
1. For the legal requirements, most likely I will need to use the
Developer Certificate of Origin. Does that mean I will just need to do
`git commit --signoff`?
As far as I am aware, yes,
GCC gives a -Wignored-attributes warning when a class template is
instantiated with a type that has an aligned(n) attribute. Specifically,
cris-elf uses 'typedef int __attribute_((__aligned(4))) _Atomic_word;'
and so compiling libstdc++ headers gives:
warning: ignoring attributes on template argume
libstdc++-v3/ChangeLog:
* include/bits/simd_mask_reductions.h: New file.
Signed-off-by: Matthias Kretz
---
.../include/bits/simd_mask_reductions.h | 94 +++
1 file changed, 94 insertions(+)
create mode 100644 libstdc++-v3/include/bits/simd_mask_reductions.h
--
On Wed, Oct 8, 2025, 14:53 Harald Anlauf wrote:
> Am 08.10.25 um 10:43 schrieb NightStrike:
> > On Tue, Oct 7, 2025, 16:25 Jerry D wrote:
> >
> >> On 10/7/25 1:13 PM, Harald Anlauf wrote:
> >>> Dear All,
> >>>
> >>> the attached patch addresses a rather old (> 14 years) issue.
> >>> We generated
The cselim pass is enabled only for targets that have conditional move
instructions. Since pru-unknown-elf doesn't have such instructions, the
pass is not executed, and the test fails with:
gcc.dg/tree-ssa/cselim-2.c: dump file does not exist
UNRESOLVED: gcc.dg/tree-ssa/cselim-2.c scan-tree-d
On 10/8/25 09:47, Shreya Munnangi wrote:
> In pr120811, we have cases where GCC is emitting an extra |addi| instruction
> instead of using the 12-bit signed-immediate of |ld|.
> |addi t1, t1, 1 ld t1, 0(t1) |
>
> This problem occurs when |fp -> sp+offset| elimination results in an
> out-of-range
Am 08.10.25 um 10:43 schrieb NightStrike:
On Tue, Oct 7, 2025, 16:25 Jerry D wrote:
On 10/7/25 1:13 PM, Harald Anlauf wrote:
Dear All,
the attached patch addresses a rather old (> 14 years) issue.
We generated warnings for standard conforming code, where a symbol
was given a bind(c) attribut
This patch adds functions for variant name mangling and context selector
merging that are shared by the C and C++ front ends.
The OpenMP specification says that name mangling is supposed to encode
the context selector for the variant, but also provides for no way to
reference these functions direc
> On 8 Oct 2025, at 16:04, Jason Merrill wrote:
>
> On 10/8/25 3:08 PM, Iain Sandoe wrote:
Otherwise, I’m looking for guidance on what’s needed to land this.
>>> IIRC the paper talks about a call to an unknown function having the desired
>>> semantics. I’ll go with that, making it not t
On Wed, 8 Oct 2025 at 15:01, Tomasz Kaminski wrote:
>
>
>
> On Wed, Oct 8, 2025 at 3:57 PM Jonathan Wakely wrote:
>>
>> On Wed, 08 Oct 2025 at 15:49 +0200, Tomasz Kaminski wrote:
>> >On Wed, Oct 8, 2025 at 2:11 PM Jonathan Wakely wrote:
>> >> diff --git a/libstdc++-v3/include/bits/version.tpl
>>
On Wed, 8 Oct 2025, Jakub Jelinek wrote:
> This patch is mostly script generated, with only
> contrib/unicode/README
> gcc/c-c++-common/cpp/named-universal-char-escape-1.c
> libcpp/makeuname2c.cc
These are OK.
--
Joseph S. Myers
[email protected]
>> Otherwise, I’m looking for guidance on what’s needed to land this.
> IIRC the paper talks about a call to an unknown function having the desired
> semantics. I’ll go with that, making it not throwing, const but looping and
> leaf. We can make it stronger if necessary, weakening it after p
From: Alfie Richards
Adds tests that check the aarch64 version features are supported, that they
have the correct priority ordering, and that the generated resolver is correct.
gcc/testsuite/ChangeLog:
* gcc.target/aarch64/fmv_priority1.c: New test.
* gcc.target/aarch64/fmv_prio
On 10/6/25 12:21 AM, H.J. Lu wrote:
On Sun, Oct 5, 2025 at 10:59 PM Jason Merrill wrote:
On 9/23/25 8:02 AM, H.J. Lu wrote:
On Tue, Sep 23, 2025 at 2:22 PM Florian Weimer wrote:
* H. J. Lu:
It seems to me this goes in the wrong direction: guard_decl doesn't have
proper location informati
From: Alfie Richards
The version data-structure already stores the versions in a sorted order so
sorting here is unnecessary.
gcc/ChangeLog:
* config/aarch64/aarch64.cc (dispatch_function_versions): Remove
unnecessary sorting and data structure.
---
gcc/config/aarch64/aarch64.c
> On Oct 7, 2025, at 23:29, Kees Cook wrote:
>
> On Tue, Oct 07, 2025 at 09:02:21PM +, Qing Zhao wrote:
>> Hi, Kees,
>>
>> I took a look at the patch today.
>> Overall, I think the patch is good and should work well.
>
> Thanks!
>
>> My major questions are:
>>
>> 1. Are the five functi
From: Alfie Richards
This fixes the versioning rules for aarch64.
Previously this would prioritize the version string with more extensions
specified regardless of the extension.
The ACLE rules are that any two version strings should be ordered by the
highest priority feature that the versions d
The value should use divide instead of modulo, as given 1st of month
being weekday X (Mon, Tue, ...), 01 is always X[1], 08 is X[2], e.t.c.
This values is currently not observable, as there is no user-accessible
format specifier that will print it, however it may be exposed in future.
libstdc++-v
Hi,
Following is version 4 of the patch proposed for master to fix PR104116.
Bootstrapped/regtested on powerpc64le and x86_64 on linux. Kindly review.
There are some issues I found for ceil and round division, which I
believe is existing.
For CEIL_DIV_EXPR, ROUND_DIV_EXPR in the unsigned case,
This implements basic_vec for all but complex value types.
libstdc++-v3/ChangeLog:
* include/bits/simd_vec.h: New file.
Signed-off-by: Matthias Kretz
---
libstdc++-v3/include/bits/simd_vec.h | 2130 ++
1 file changed, 2130 insertions(+)
create mode 100644 lib
On Tue, 2025-10-07 at 22:19 +0200, Arsen Arsenović wrote:
> Jonathan Wakely writes:
>
> > Storing both .pdf and .pdf.gz files seems silly, when compression
> > makes very little difference for PDF files.
> >
> > Storing both .ps and .ps.gz seems silly when the .ps files are quite
> > large (and
libstdc++-v3/ChangeLog:
* include/bits/simd_details.h: New file.
* include/bits/simd_x86.h: New file.
* include/bits/vec_ops.h: New file.
Signed-off-by: Matthias Kretz
---
libstdc++-v3/include/bits/simd_details.h | 1443 ++
libstdc++-v3/include/bits
libstdc++-v3/ChangeLog:
* include/bits/simd_iterator.h: New file.
Signed-off-by: Matthias Kretz
---
libstdc++-v3/include/bits/simd_iterator.h | 153 ++
1 file changed, 153 insertions(+)
create mode 100644 libstdc++-v3/include/bits/simd_iterator.h
--
Hi Jerry and Harald,
I committed the patch, less the whitespace changes, as r16-4282.
I'll deal with the whitespace at a later date.
Cheers and thanks
Paul
On Tue, 7 Oct 2025 at 18:12, Harald Anlauf wrote:
>
> Am 07.10.25 um 18:56 schrieb Jerry D:
> > On 10/7/25 2:19 AM, Paul Richard Thomas w
On Wednesday 08 October 2025 at 12:05:24 +0100, Jonathan Wakely wrote:
> Thanks for the patches, Mike, I've finally managed to do the first set
> of reviews ...
Thanks for the review.
[snip suggested changes]
> I can push this if you're happy with those changes.
I'm happy with those changes and
From: Luc Grosheintz
This commit adds the right padded layout as described in N5014, with
LWG4372 (dynamic padding value) and LWG4314 (move in operator()).
PR libstdc++/110352
libstdc++-v3/ChangeLog:
* include/std/mdspan (_RightPaddedIndices): Traits for right
padded la
On Wed, Oct 08, 2025 at 10:13:42AM +0100, Yury Khrustalev wrote:
> On Wed, Oct 08, 2025 at 10:49:21AM +0200, Jakub Jelinek wrote:
> > On Wed, Oct 08, 2025 at 09:35:06AM +0100, Yury Khrustalev wrote:
> > > I think we have 2 things here:
> > >
> > > 1. Detect situation when we work with something li
From: Luc Grosheintz
This commit adds a new layout layout_left_padded as standardized in
N5014. It adds a purely internal feature testing macro padded_layouts
and registers layout_left_padded in the std module.
This commit implements LWG4372, because without it's not possible
to properly test pa
On Thu, 2 Oct 2025 at 10:29, Luc Grosheintz wrote:
>
> Add the class and updates the std module.
>
> PR libstdc++/110352
>
> libstdc++-v3/ChangeLog:
>
> * include/std/mdspan (full_extent_t): New class.
> * src/c++23/std.cc.in (full_extent_t): Add.
OK for trunk, thanks.
>
Status
==
Stage 1 of the development of GCC 16 will end with Stage 3 starting
at Monday, Nov 17th, followed by Stage 4 starting Jan 12th.
There's not much to say about the statistics below, the P3 regressions
have not been triaged yet.
Quality Data
Priority # Change
On Tue, 30 Sep 2025 at 10:14 +0200, Tomasz Kamiński wrote:
From: Luc Grosheintz
This commit adds a new layout layout_left_padded as standardized in
N5014. It adds a purely internal feature testing macro padded_layouts
and registers layout_left_padded in the std module.
This commit implements L
On Tue, Oct 07, 2025 at 08:37:52AM +0800, H.J. Lu wrote:
> The new test failed on Linux/x86-64 with -m32:
Ran into the same.
The test FAILs on ilp32 targets with
pr121987.c:5:21: warning: unsigned conversion from 'long long int' to 'long
unsigned int' changes value from '100' to '1410065
On Tue, Oct 7, 2025 at 5:10 AM Patrick Palka wrote:
> Tested on x86_64-pc-linux-gnu, does this look OK for trunk? I wonder
> about backports, for either the whole paper or just the common_reference
> change in for avoiding dangling references.
>
> -- >8 --
>
> This patch implements this paper as
On Tue, Oct 7, 2025 at 5:09 AM Patrick Palka wrote:
> ... and in passing use requires-clauses instead of void_t based SFINAE.
> This is a non-functional change that'll simplify implementing the
> P2655R3 change to common_reference.
> libstdc++-v3/ChangeLog:
>
> * include/std/type_traits
38 matches
Mail list logo
