Hi,
This extension defines vector load instructions to move sign-extended or
zero-extended INT4 data into 8-bit vector register elements.
gcc/ChangeLog:
* config/riscv/andes-vector-builtins-bases.cc
(nds_nibbleload): New class.
* config/riscv/andes-vector-builtins-bases.h
libstdc++-v3/ChangeLog:
* include/std/mdspan (default_accessor): New class.
* src/c++23/std.cc.in: Register default_accessor.
* testsuite/23_containers/mdspan/accessors/default.cc: New test.
* testsuite/23_containers/mdspan/accessors/default_neg.cc: New test.
Signe
Previously the prerequisite that the arguments passed to operator() are
a multi-dimensional index (of extents()) was not checked.
This commit adds the __glibcxx_asserts and the required tests.
libstdc++-v3/ChangeLog:
* include/std/mdspan: Check prerequisites of
layout_*::operator
This is the second iteration of this patch series and the first also
sent to gcc-patches.
The individual patches can be restructured as desired.
The final patch is a proposal to strengthen some exception guarantees to
make mdspan nothrow movable. The standard doesn't require this; but I
felt it m
This fourth iteration adds several additional tests to confirm the
pointer convertiblity constraint of the ctor.
Luc Grosheintz (1):
libstdc++: Implement default_accessor from mdspan.
libstdc++-v3/include/std/mdspan | 31
libstdc++-v3/src/c++23/std.cc.in |
Hi Jeff,
> You've added a new type (nds_vfpmad). You'll need to udpate all the
> existing DFAs to handle insns with that type. Otherwise someone that
> asks for code generation for your design, but a different tuning model
> will get aborts.
> It's generally considered OK to add them to a dummy
The class IntLike is used for testing extents with user-defined classes
that convert to int. This commit places the class into a separate header
file. This allows it to be reused across different parts of the mdspan
related testsuite.
libstdc++-v3/ChangeLog:
* testsuite/23_containers/mdsp
On Fri, Jun 27, 2025 at 11:23 AM Jonathan Wakely wrote:
> On Fri, 27 Jun 2025 at 10:10, Tomasz Kaminski wrote:
> >
> >
> >
> > On Fri, Jun 27, 2025 at 10:31 AM Tomasz Kamiński
> wrote:
> >>
> >> This patch adjust all internal std::format call inside of
> __formatter_chrono,
> >> to use runtime
The following amends the SLP addsub pattern to also match blends
of .FMA/.FMS and form .FMADDSUB even when -ffp-contract=off.
Bootstrap and regtest ongoing on x86_64-unknown-linux-gnu.
Richard.
PR tree-optimization/120808
* tree-vect-slp-patterns.cc (vect_match_expression_p):
On Fri, 27 Jun 2025 at 11:01, Tomasz Kaminski wrote:
>
>
>
> On Fri, Jun 27, 2025 at 11:23 AM Jonathan Wakely wrote:
>>
>> On Fri, 27 Jun 2025 at 10:10, Tomasz Kaminski wrote:
>> >
>> >
>> >
>> > On Fri, Jun 27, 2025 at 10:31 AM Tomasz Kamiński
>> > wrote:
>> >>
>> >> This patch adjust all int
On 2025-06-27 07:40, Siddhesh Poyarekar wrote:
MEM_REF cast of a subobject to its containing object has negative
offsets, which objsz sees as an invalid access. Support this use case
by peeking into the structure to validate that the containing object
indeed contains a type of the subobject at t
On Fri, Jun 27, 2025 at 11:23 AM Jonathan Wakely wrote:
> On Fri, 27 Jun 2025 at 10:10, Tomasz Kaminski wrote:
> >
> >
> >
> > On Fri, Jun 27, 2025 at 10:31 AM Tomasz Kamiński
> wrote:
> >>
> >> This patch adjust all internal std::format call inside of
> __formatter_chrono,
> >> to use runtime
> Am 27.06.2025 um 13:40 schrieb Siddhesh Poyarekar :
>
> MEM_REF cast of a subobject to its containing object has negative
> offsets, which objsz sees as an invalid access. Support this use case
> by peeking into the structure to validate that the containing object
> indeed contains a type o
On Fri, Jun 27, 2025 at 02:00:26PM +0200, Richard Biener wrote:
> > + gcc_assert (RECORD_OR_UNION_TYPE_P (container));
> > +
> > + for (tree t = TYPE_FIELDS (container); t; t = DECL_CHAIN (t))
> > +{
> > + if (TREE_CODE (t) != FIELD_DECL)
> > +continue;
> > +
> > + tree byte_off
On 27/06/25 13:01 +0200, Tomasz Kamiński wrote:
The r16-1709-g4b3cefed1a08344495fedec4982d85168bd8173f caused `-Woverflow`
in empty_spec.cc file. This warning is not cause by any issue in shipping
code, and results in taking to much shorcut when implementing a test-only
custom representation type
On Fri, 27 Jun 2025, Richard Biener wrote:
> > I am testing partial fixes for these issues.
>
> Can you check again after r16-1731-g08bdb6b4a32f1f?
Certainly. There's no more fmaddsub-related testsuite failures, and only
three tests need adjustment on x86.
* gcc.target/i386/pr116979.c
Here
I found that isn't included in this patch after I committed that...I try to
fix that soon but CI caught me :P
https://gcc.gnu.org/git/?p=gcc.git;a=commit;h=7c67f7f8d4c8aadbe8efd733c29d13bfcbb0f50f
On Thu, Jun 26, 2025 at 10:05 PM Robin Dapp wrote:
> Hi Kito,
>
> > This patch adds a comment to t
Hi,
A status update on this patch: (Actually a good news!)
> On Jun 10, 2025, at 11:32, Qing Zhao wrote:
>
>
>
>>>
>>> It's difficult to do any meaningful pruning I think. Consider
>>>
>>> if (i == -1)
>>> tem = a[i];
>>>
>>> when we transform this to
>>>
>>> if (i == -1)
>>> t
I take this patch back. It seems to be incomplete.
- Andre
On Fri, 27 Jun 2025 14:45:36 +0200
Andre Vehreschild wrote:
> Hi all,
>
> this patch fixes a reject valid when the coranks of two operands do not match
> and no coindex is given. I.e. when only an implicit this_image co-ref is used.
>
> Hi Honza,
> > So merging the profiles will also lead to inconsistencies making the
> > .part variant to seem more hot than it is...
>
> I am looking into this and will post the patch as a follow up patch.
Thanks. Note that now with merging being done recursively to inline
instances while offli
On 26/06/25 23:12 -0400, Patrick Palka wrote:
On Thu, 26 Jun 2025, Patrick Palka wrote:
PR libstdc++/100795
libstdc++-v3/ChangeLog:
* include/bits/ranges_algo.h (__sample_fn::operator()):
Reimplement the forward_iterator branch directly.
* testsuite/25_algorith
OK.
Hmm, I'm still seeing test failures in the CI. Could you check if those are
valid?
--
Regards
Robin
A right shift of 31 will become 0 or 1, this can be checked for
treg_set_expr_not_const01 to avoid matching addc_t_r as this
can expand to a 3 insn sequence instead.
This improves tests 023 to 026 from gcc.target/sh/pr54236-2.c, e.g.:
test_023:
shllr5
mov #0,r1
mov r4,r0
rts
addcr1,
On 26/06/25 22:25 -0400, Patrick Palka wrote:
PR libstdc++/100795
OK for trunk.
libstdc++-v3/ChangeLog:
* include/bits/ranges_algo.h (shuffle_fn::operator()):
Reimplement directly.
* testsuite/25_algorithms/shuffle/constrained.cc (test02):
---
libstdc++-v3/in
On Fri, 2025-06-27 at 10:51 -0300, Raphael Moreira Zinsly wrote:
> A right shift of 31 will become 0 or 1, this can be checked for
> treg_set_expr_not_const01 to avoid matching addc_t_r as this
> can expand to a 3 insn sequence instead.
> This improves tests 023 to 026 from gcc.target/sh/pr54236-
Is there anyway we can retrigger the test somewhere ? If no I can send a v3
series with the commit reordered and see.
I don't think there's a way other than re-submitting. But if you're sure you
tested properly and the CI is mistaken we can go ahead. I just wanted to make
sure as with the s
From: Pan Li
This patch would like to introduce the combine of vec_dup + vssubu.vv
into vssubu.vx on the cost value of GR2VR. The late-combine will take
place if the cost of GR2VR is zero, or reject the combine if non-zero
like 1, 2, 15 in test. There will be two cases for the combine:
Case 0:
On Fri, Jun 27, 2025 at 08:11:29AM +0200, Uros Bizjak wrote:
> On Fri, Jun 27, 2025 at 7:27 AM Andi Kleen wrote:
> >
> > Uros Bizjak writes:
> >
> > > Introduce crc_revsi4 expanders to generate CRC32 instruction when
> > > using
> > > __builtin_rev_crc32_data* builtins with 0x1EDC6F41 poylnomial
On Fri, 27 Jun 2025, Richard Biener wrote:
> > I think such a warning should be based on an attribute on the time_t type
> > that means "warn for implicit truncation of this type" (I'm less clear on
> > why warnings for implicit widening conversions *to* time_t are supposed to
> > be useful), r
Hi all,
For targets that have expensive shifts this may not get a better
sequence right now, specially for AVR and MSP430 according to
our tests.
Before I start looking for a fix on those targets I want to know
if someone has any advise or other concerns with this transformation.
Thanks,
-- >8
s390 missed constant vector permutation cases based on the vector pack
instruction or changing the size of the vector elements during vector
merge. This enables some more patterns that do not need to load a
constant vector for permutation.
Bootstrapped and regtested on s390. Okay for trunk?
gcc
The test added by r16-1671-ge7ff8e8d77df74 passed despite using
regular expressions that would never match real assembly language
output from the compiler. Because the regular expressions were not
expected to match, and didn't, this was not noticeable; however,
it also made that part of the test us
The test added by r16-1671-ge7ff8e8d77df74 passed despite using
regular expressions that would never match real assembly language
output from the compiler.
Modified the test added by r16-1671-ge7ff8e8d77df74 to use scan-assembler-times
instead of scan-assembler-not (because we expect to see the ol
> From:Kito Cheng
> Send Time:2025 Jun. 9 (Mon.) 17:33
> To:yunzezhu
> CC:"gcc-patches"
> Subject:Re: [PATCH] [RFC] RISC-V: Add extra check to help choosing multilib
> with equivalent arch.
>
>
> Oh, yeah, I got your point, I was just misreading, the march is
> rv32imac rather than rv32imafc, t
Le 23/06/2025 à 09:34, Richard Biener a écrit :
On Sun, Jun 22, 2025 at 10:23 PM Mikael Morin wrote:
From: Mikael Morin
See the description in the ChangeLog entry below.
The testcases are best effort; for some operators the fortran frontend
generates a temporary variable, so the simplificat
On 26/06/25 22:25 -0400, Patrick Palka wrote:
ranges::push_heap, ranges::pop_heap, ranges::make_heap and
ranges::sort_heap are currently defined in terms of the corresponding
STL-style algorithms, but this is incorrect because the STL-style
algorithms rely on the legacy iterator system, and so mi
On 26/06/25 22:25 -0400, Patrick Palka wrote:
As with the previous patch, this patch reimplements ranges::sort
directly instead of incorrectly forwarding to std::sort. In addition to
the compatibility changes listed in the previous patch we also:
- use ranges::iter_swap instead of std::iter_sw
On 26/06/25 22:25 -0400, Patrick Palka wrote:
As with the previous patch, this patch reimplements ranges::inplace_merge
directly instead of incorrectly forwarding to std::inplace_merge. In
addition to the compatibility changes listed in the previous patch we
also:
- explicitly cast the differe
MEM_REF cast of a subobject to its containing object has negative
offsets, which objsz sees as an invalid access. Support this use case
by peeking into the structure to validate that the containing object
indeed contains a type of the subobject at that offset and if present,
adjust the wholesize f
On 2025-06-27 08:05, Jakub Jelinek wrote:
On Fri, Jun 27, 2025 at 02:00:26PM +0200, Richard Biener wrote:
+ gcc_assert (RECORD_OR_UNION_TYPE_P (container));
+
+ for (tree t = TYPE_FIELDS (container); t; t = DECL_CHAIN (t))
+{
+ if (TREE_CODE (t) != FIELD_DECL)
+continue;
+
+
Hi all,
this patch fixes a reject valid when the coranks of two operands do not match
and no coindex is given. I.e. when only an implicit this_image co-ref is used.
Regtests ok on x86_64-pc-linux-gnu / F41. Ok for mainline?
Regards,
Andre
--
Andre Vehreschild * Email: vehre ad gmx dot d
> -Original Message-
> From: Cui, Lili
> Sent: Friday, June 27, 2025 5:04 PM
> To: H.J. Lu
> Cc: ubiz...@gmail.com; gcc-patches@gcc.gnu.org; Liu, Hongtao
> ; richard.guent...@gmail.com; Michael Matz
> ; Sam James ; kenjin4...@gmail.com
> Subject: RE: [PATCH V3] x86: Enable separate shrin
Hi!
On Wed, Jun 25, 2025 at 02:50:14PM -0400, Michael Meissner wrote:
> This is patch #1 of 3 that adds the support that can be used in developing GCC
> support for potential future PowerPC processors. With all 3 patches, the
> tuning
> for the 'future' processor is the same as power10 and power
On 2025-06-27 08:39, Siddhesh Poyarekar wrote:
MEM_REF cast of a subobject to its containing object has negative
offsets, which objsz sees as an invalid access. Support this use case
by peeking into the structure to validate that the containing object
indeed contains a type of the subobject at t
On Fri, Jun 27, 2025, 11:06 AM Raphael Moreira Zinsly <
rzin...@ventanamicro.com> wrote:
> Hi all,
>
> For targets that have expensive shifts this may not get a better
> sequence right now, specially for AVR and MSP430 according to
> our tests.
> Before I start looking for a fix on those targets I
On Fri, Jun 27, 2025, 11:06 AM Raphael Moreira Zinsly <
rzin...@ventanamicro.com> wrote:
> Hi all,
>
> For targets that have expensive shifts this may not get a better
> sequence right now, specially for AVR and MSP430 according to
> our tests.
> Before I start looking for a fix on those targets I
Tested x86_64-pc-linux-gnu, applying to trunk.
-- 8< --
With Jakub's constexpr virtual base patch,
23_containers/vector/bool/cmp_c++20.cc failed the assert I add to
fixed_type_or_null, meaning that it returned the wrong value. Let's fix the
result as well as adding the assert, and fix cp_parser_
On Fri, Jun 27, 2025 at 01:38:51PM +, Qing Zhao wrote:
> The good news is: With the above simple heuristic and a simple back tracing
> of the CFG, all the
> current testing cases for the following PRs passed without any issue:
>
> PR109071
> PR88771
> PR85788
> PR108770
> PR106762
> PR11
On 6/27/25 12:30 PM, Andrew Pinski wrote:
On Fri, Jun 27, 2025, 11:06 AM Raphael Moreira Zinsly
mailto:rzin...@ventanamicro.com>> wrote:
Hi all,
For targets that have expensive shifts this may not get a better
sequence right now, specially for AVR and MSP430 according to
On 6/27/25 1:48 PM, Harald Anlauf wrote:
Dear all,
my original patch caused a regression on previously working code where
an imported interface was *not* renamed-on-use, as the related new logic
did not expect a local_name to be an empty string.
Funnily, there was no previously existing test in
Am 27.06.25 um 22:57 schrieb Jerry D:
On 6/27/25 1:48 PM, Harald Anlauf wrote:
Dear all,
my original patch caused a regression on previously working code where
an imported interface was *not* renamed-on-use, as the related new logic
did not expect a local_name to be an empty string.
Funnily, t
On Fri, Jun 27, 2025 at 05:17:55PM -0400, Jason Merrill wrote:
> I've pushed those changes.
Great.
> > /* For a non-pointer simple base reference, express it as a
> > COMPONENT_REF
> >without taking its address (and so causing lambda capture, 91933).
> > */
> > - if (code == PLUS_
Hi!
My changes for "Module Declarations Shouldn’t be Macros" paper broke
the following testcase. The backup handling intentionally tries to
drop CPP_PRAGMA_EOL token if things go wrong, which is desirable for the
case where we haven't committed to the module preprocessing directive
(i.e. changed
Hi!
The following testcase is miscompiled since the introduction of UBSan,
cp_build_array_ref COND_EXPR handling replaces
(cond ? a : b)[idx] with cond ? a[idx] : b[idx], but if there are
SAVE_EXPRs inside of idx, they will be evaluated just in one of the
branches and the other uses uninitialized
On 6/27/25 5:17 PM, Jason Merrill wrote:
On 6/26/25 1:30 PM, Jakub Jelinek wrote:
On Thu, Jun 26, 2025 at 01:33:08PM +0200, Jakub Jelinek wrote:
I get some regressions (which I didn't get with the earlier patch, but
it isn't obvious by what it has been caused):
It ICEs were caused by the cano
On 6/26/25 1:30 PM, Jakub Jelinek wrote:
On Thu, Jun 26, 2025 at 01:33:08PM +0200, Jakub Jelinek wrote:
I get some regressions (which I didn't get with the earlier patch, but
it isn't obvious by what it has been caused):
It ICEs were caused by the canonicalize_obj_off change and indeed
The IC
On 6/27/25 5:03 AM, Nathaniel Shead wrote:
On Wed, Jun 25, 2025 at 01:05:39PM -0400, Jason Merrill wrote:
On 5/21/25 10:14 PM, Nathaniel Shead wrote:
This patch isn't currently necessary with how I've currently done the
follow-up patches, but is needed for avoiding any potential issues in
the f
Dear all,
my original patch caused a regression on previously working code where
an imported interface was *not* renamed-on-use, as the related new logic
did not expect a local_name to be an empty string.
Funnily, there was no previously existing test in the testsuite...
The attached fixes this
The cherry-picked patch fixes sanitizer builds with musl libc. Which
should fix both ppc and riscv issues using the sanitizer with musl.
Pushed to the trunk.
Jeff
commit 5111ea055f5699ad47605890fb45af5c3db1de4f
Author: Jeff Law
Date: Fri Jun 27 15:11:41 2025 -0600
[sanitizer_common]
On 6/27/25 3:28 AM, Nathaniel Shead wrote:
On Wed, Jun 25, 2025 at 11:52:14AM -0400, Jason Merrill wrote:
On 6/25/25 9:02 AM, Nathaniel Shead wrote:
On Tue, Jun 24, 2025 at 12:10:09PM -0400, Patrick Palka wrote:
On Tue, 24 Jun 2025, Jason Merrill wrote:
On 6/23/25 5:41 PM, Nathaniel Shead wr
On 6/26/25 4:12 PM, Marek Polacek wrote:
On Wed, Jun 25, 2025 at 03:13:25PM -0400, Jason Merrill wrote:
On 6/25/25 1:28 PM, Marek Polacek wrote:
@@ -24604,7 +24604,7 @@ resolve_nondeduced_context (tree orig_expr,
tsubst_flags_t complain)
}
if (good == 1)
{
- ma
On 6/27/25 5:58 PM, Jakub Jelinek wrote:
Hi!
The following testcase is miscompiled since the introduction of UBSan,
cp_build_array_ref COND_EXPR handling replaces
(cond ? a : b)[idx] with cond ? a[idx] : b[idx], but if there are
SAVE_EXPRs inside of idx, they will be evaluated just in one of the
Add "void debug (tree)" to support:
(gdb) call debug (expr)
unit-size
user align:256 warn_if_not_align:0 symtab:0 alias-set -1 canonical-type
0x7fffe99cebd0
fields
XF x.c:2:15
size
unit-size
align:128 warn_if_not_align:0
On 6/25/25 10:22 AM, Iain Sandoe wrote:
On 25 Jun 2025, at 15:17, Martin Jambor wrote:
Hi,
when building GCC with clang, it warns that the private member suffix
in class cp_coroutine_transform (defined in gcc/cp/coroutines.h) is
not used which indeed looks like it is the case. This patch t
Since a backend may ignore user type alignment for arguments passed on
stack, update alignment for arguments passed on stack when copying MEM's
memory attributes.
gcc/
PR target/120839
* emit-rtl.cc (set_mem_attrs): Update alignment for argument on
stack.
gcc/testsuite/
PR target/120839
* gcc.t
On Sat, Jun 28, 2025 at 12:03 PM Jeff Law wrote:
>
>
>
> On 6/27/25 5:34 PM, H.J. Lu wrote:
> > Add "void debug (tree)" to support:
> >
> > (gdb) call debug (expr)
> >> type > size
> > unit-size
> > user align:256 warn_if_not_align:0 symtab:0 alias-set -1
>
Martin Jambor writes:
> Hi,
>
> When GCC is built with clang, it emits warnings that several member
> functions of various ranger classes override a virtual function of an
> ancestor but are not marked with the override keyword. After
> inspecting the cases, I found that all these classes had ot
The CI reported 2 failures but seems unrelated and passed locally, thus will
commit this before the end of next Monday(6/30) if no more concerns.
> FAIL: gcc.dg/vect/pr115669.c -flto -ffat-lto-objects (test for excess errors)
Executing on host:
/home/lipan/gnu-toolchain/build/build-riscv64-unkn
On Sat, Jun 28, 2025 at 12:24 AM Cui, Lili wrote:
>
>
>
> > -Original Message-
> > From: Cui, Lili
> > Sent: Friday, June 27, 2025 5:04 PM
> > To: H.J. Lu
> > Cc: ubiz...@gmail.com; gcc-patches@gcc.gnu.org; Liu, Hongtao
> > ; richard.guent...@gmail.com; Michael Matz
> > ; Sam James ; kenj
Hi Jakub,
> signbit is documented to be a macro, so please don't declare
> int signbit (double);
> function in the testcase and instead of signbit use __builtin_signbit.
This is indeed my negligence. Done.
If everything looks good, could you please help me merge this patch? Thank you!
Yuao
0
HI,
I have tested your patch on exchange2 and noticed multiple problems:
1) with LTO the translation from dwarf names to symbol names is disabled
since we free lang data sooner. I moved the offline pass upstream which
however also may make us miss clones intorduced betwen free lang dat
On 6/27/25 5:34 PM, H.J. Lu wrote:
Add "void debug (tree)" to support:
(gdb) call debug (expr)
unit-size
user align:256 warn_if_not_align:0 symtab:0 alias-set -1 canonical-type
0x7fffe99cebd0
fields
XF x.c:2:15
size
unit-
101 - 172 of 172 matches
Mail list logo
