From: Pan Li
Add test case for both the run and asm check of mul based SAT_MUL.
gcc/testsuite/ChangeLog:
* gcc.target/riscv/sat/sat_u_mul-6-u16-from-u32.c: New test.
* gcc.target/riscv/sat/sat_u_mul-6-u16-from-u64.rv32.c: New test.
* gcc.target/riscv/sat/sat_u_mul-6-u16-
From: Pan Li
This patch would like to try to match the the unsigned
SAT_MUL form 5, aka below:
#define DEF_SAT_U_MUL_FMT_5(NT, WT) \
NT __attribute__((noinline))\
sat_u_mul_##NT##_from_##WT##_fmt_5 (NT a, NT b) \
{
On Thu, 18 Sep 2025, Jakub Jelinek wrote:
> On Thu, Sep 18, 2025 at 06:54:06PM +0200, Jason Merrill wrote:
> > > There are some regressions caused by the removal of {CLOBBER(bob)}
> > > clobbers from the start of certain constructors, e.g. one testcase has
> > > struct A
> > > {
> > >int f,g;
On Fri, Sep 19, 2025 at 12:08 AM Peter Damianov wrote:
>
> POSIX says that sin and cos should set errno to EDOM when infinity is passed
> to
> them. Make sure this is accounted for in builtins.def.
>
> When sin/cos are called with values that set errno (like INFINITY), GCC was
> incorrectly optim
On Thu, Sep 18, 2025 at 10:19 PM Robin Dapp wrote:
>
> > But the vector type we perform the permutation on should be unchanged (it's
> > not the punned type but the original type we pun the loaded vector back to)?
>
> Yeah, I was trying to re-use what we have but I see now that just passing a
> di
> -Original Message-
> From: Yury Khrustalev
> Sent: Wednesday, September 17, 2025 4:39 PM
> To: gcc-patches@gcc.gnu.org
> Cc: Tamar Christina ; Srinath Parvathaneni
> ; Tejas Belagod ;
> ja...@redhat.com
> Subject: [PATCH] dwarf: Save bit stride information array type entry
> [PR121964]
On Mon, 8 Sept 2025 at 17:53, Richard Biener wrote:
> Usually sth like contracts aids in formal program verification. But … too
> many C++ papers to read. Still C++ sometimes feels like going the „coming
> from X, now also available in C++“, aka chasing for feature completeness as
> its own
> -Original Message-
> From: Richard Biener
> Sent: Thursday, September 11, 2025 10:58 AM
> To: Tamar Christina
> Cc: gcc-patches@gcc.gnu.org; nd
> Subject: Re: [PATCH v4 3/3]middle-end: Use addhn for compression instead of
> inclusive OR when reducing comparison values
>
> On Tue, 9 Se
On Thu, Sep 18, 2025 at 10:49 PM Andrew Pinski
wrote:
>
> On Tue, Sep 16, 2025 at 11:50 PM Richard Biener
> wrote:
> >
> > On Wed, Sep 17, 2025 at 4:55 AM Andrew Pinski
> > wrote:
> > >
> > > Since now optimize_aggr_zeroprop and optimize_agr_copyprop work by
> > > forward walk to prop
> > > the
On Tue, Sep 16, 2025 at 11:50 PM Richard Biener
wrote:
>
> On Wed, Sep 17, 2025 at 4:55 AM Andrew Pinski
> wrote:
> >
> > Since now optimize_aggr_zeroprop and optimize_agr_copyprop work by forward
> > walk to prop
> > the zero/aggregate and does not change the statement at hand, there is no
> >
Hi Alexander,
> -Original Message-
> From: Alexandre Oliva
> Sent: Saturday, September 13, 2025 8:02 AM
> To: gcc-patches@gcc.gnu.org
> Cc: Vladimir Makarov ; Rainer Orth Bielefeld.DE>; Mike Stump ; Richard Earnshaw
> ; Tamar Christina ;
> Kyrylo Tkachov
> Subject: [PATCH 1v1/2] [aarch6
> On 28 Aug 2025, at 09:19, Jennifer Schmitz wrote:
>
> External email: Use caution opening links or attachments
>
>
>> On 25 Aug 2025, at 15:40, Kyrylo Tkachov wrote:
>>
>> Hi Jennifer,
>>
>>> On 25 Aug 2025, at 12:56, Jennifer Schmitz wrote:
>>>
>>> When op2 in SVE2 saturating add intri
On 9/9/25 4:36 AM, Patrick Palka wrote:
Bootstrapped and regttested on x86_64-pc-linux-gnu, does this look
OK for trunk?
OK.
-- >8 --
Apparently an explicitly defined operator<=> isn't required to return
std::foo_ordering, so build_min_non_dep_op_overload needs to be able
to handle a (x <=>
This didn't show up in my inbox for some reason, so sorry for awkward
reply:
Can you do this? { dg-additional-options "-fPIE" { target pie } }
It looks good to me if that works fine.
On Fri, Sep 12, 2025 at 02:01:57PM +, Qing Zhao wrote:
>
> > On Sep 12, 2025, at 03:32, Kees Cook wrote:
> >
> > On Thu, Sep 11, 2025 at 03:04:01PM +, Qing Zhao wrote:
> >>
> >>
> >>> On Sep 10, 2025, at 23:05, Kees Cook wrote:
> >>>
> >>> On Tue, Sep 09, 2025 at 06:49:22PM +, Qi
On Tue, Sep 9, 2025 at 5:03 PM Jonathan Wakely wrote:
> On Tue, 9 Sept 2025 at 15:57, Tomasz Kaminski wrote:
> >
> >
> >
> > On Tue, Sep 9, 2025 at 4:41 PM Jonathan Wakely
> wrote:
> >>
> >> On Mon, 8 Sept 2025 at 12:41, Tomasz Kamiński
> wrote:
> >> >
> >> > I have double checked that impleme
On Sat, Sep 06, 2025 at 03:53:33PM -0700, Andrew Pinski wrote:
> On Sat, Sep 6, 2025 at 3:46 PM Jakub Jelinek wrote:
> >
> > On Sat, Sep 06, 2025 at 03:27:53PM -0700, Andrew Pinski wrote:
> > > This new pass will ICE if the target does not define the
> > > macro_fusion_pair_p
> > > pass. The pass
This patch doesn't work. But the first one is okay, despite that it pessimizes
sin and cos calls by skipping the optimization pass.
Sep 18, 2025 11:04:16 PM Peter Damianov :
> In my previous patch, I disabled pass_cse_sincos for -fmath-errno. This was
> because it was causing ICEs when trying to
Bootstrapped and tested with check-target-libstdc++-v3 on
x86_64-pc-linux-gnu.
Could someone help review and commit?
For applying, please use the attached patch file since my email client has
not preserved
the tabs.
Thanks,
-Ben
-- 8< --
In order to emplace a value in the middle of a deque, a
Extending the To-list to try to get this reviewed.
This patch is implementing what Richard Earnshaw suggested in
https://gcc.gnu.org/pipermail/gcc-patches/2025-August/692224.html.
Kind regards,
Torbjörn
On 2025-09-01 20:14, Torbjorn SVENSSON wrote:
On 2025-08-27 17:07, Torbjörn SVENSSON wro
From: Viljar Indus
When frontend is operating in GNATprove mode (where expander is disabled), it
should check ghost policy for assignment statements just like it does for other
statements. This is because we want ghost policy errors to be reported not just
by GNAT, but also by GNATprove.
Additio
On Fri, 12 Sept 2025 at 18:39, Patrick Palka wrote:
>
> Tested on x86_64-pc-linux-gnu, does this look OK for trunk?
> Patch generated with -w due to otherwise noisy whitespace changes.
>
> -- >8 --
>
> The ranges::shuffle optimization, copied from std::shuffle, has a
> two-at-a-time PRNG optimizat
POSIX says that sin and cos should set errno to EDOM when infinity is passed to
them. Make sure this is accounted for in builtins.def.
When sin/cos are called with values that set errno (like INFINITY), GCC was
incorrectly optimizing the code by assuming these functions never modify errno.
This ca
On Thu, 18 Sep 2025, David Faust wrote:
> So, I think I would change the above to (tests/documentation accordingly):
>
> if (TREE_CODE (*node) == FUNCTION_TYPE || TREE_CODE (*node) == METHOD_TYPE)
>{
> warning (OPT_Wattributes,
> "%qE attribute does not apply to functions"
> On Sep 18, 2025, at 17:42, Jason Merrill wrote:
>
> On 9/18/25 10:52 PM, Qing Zhao wrote:
>>> On Sep 17, 2025, at 12:08, Jakub Jelinek wrote:
>>>
>>> Hi!
>>>
>>> -ftrivial-auto-var-init= also calls .DEFERRED_INIT for temporaries
>>> if they have VOID_TYPE_P TARGET_EXPR_INITIAL (otherwise
> But the vector type we perform the permutation on should be unchanged (it's
> not the punned type but the original type we pun the loaded vector back to)?
Yeah, I was trying to re-use what we have but I see now that just passing a
different vectype to vect_transform_slp_perm_load doesn't work i
Tested x86_64-pc-linux-gnu, applying to trunk.
-- 8< --
It's now possible to see an active member of a union that has not actually
been initialized, so let's clarify the diagnostic.
gcc/cp/ChangeLog:
* constexpr.cc (cxx_eval_component_reference): Clarify diagnostic.
gcc/testsuite/Chang
Add further tests for things defined in C2Y that correspond to how GCC
already behaves. Some of these are for previous undefined behavior
where what is violated has become a constraint but previously was a
"shall" outside Constraints. Others (c2y-anon-init-1.c and
c2y-incomplete-2.c) reflect sema
In my previous patch, I disabled pass_cse_sincos for -fmath-errno. This was
because it was causing ICEs when trying to replace sin/cos calls, and more
importantly, the pass was transforming sin+cos to cexpi which doesn't set
errno like sin/cos do. The optimization is valid when either:
1. We don't
No functional change intended.
Successfully bootstrapped & regrtested on x86_64-pc-linux-gnu.
Pushed to trunk as r16-3957-g60d748800835fb.
gcc/ChangeLog:
* diagnostics/buffering.cc: Drop include of "diagnostic.h".
* diagnostics/buffering.h: Likewise.
* diagnostics/context.
Tested x86_64-pc-linux-gnu, applying to trunk.
-- 8< --
r16-3022 changed placement new to clobber the object, and improved constexpr
handling to do more with clobbers. But it occurred to me that in a lot of
cases we don't need to introduce a constructor_elt to represent an
uninitialized member o
Tested x86_64-pc-linux-gnu, applying to trunk.
-- 8< --
Since Jakub's P2795R5 work removes the initial clobber from constructors,
let's add a clobber for new of all non-empty classes.
gcc/cp/ChangeLog:
* init.cc (build_new_1): Clobber classes.
gcc/testsuite/ChangeLog:
* g++.dg
Tested x86_64-pc-linux-gnu, applying to trunk.
-- 8< --
With new (nothrow) we need to check whether the allocation returned null,
and the resulting COND_EXPR was confusing -Wmismatched-new-delete-2 at -O0.
Let's help it out by appending the address if it's the same on both arms.
gcc/cp/ChangeLog
On 9/18/25 10:52 PM, Qing Zhao wrote:
On Sep 17, 2025, at 12:08, Jakub Jelinek wrote:
Hi!
-ftrivial-auto-var-init= also calls .DEFERRED_INIT for temporaries
if they have VOID_TYPE_P TARGET_EXPR_INITIAL (otherwise they are expanded
as INIT_EXPR of the TARGET_EXPR_SLOT and TARGET_EXPR_INITIAL
On Thu, 18 Sept 2025 at 14:59, Nathan Myers wrote:
>
> Changes in v11:
> * Fix "#if 202306L" cases in synopsis test.
> * Add noexcept specifications in synopsis test.
I assume this passes tests now with C++26 being used?
OK for trunk if so.
It helps to state the testing that was done, as it say
On Thu, Sep 18, 2025 at 10:23 AM Srinath Parvathaneni
wrote:
>
> v1 -> v2 changes:
>
> * Fixed the wrong conditional check.
> * Fixed the typos in the testcases.
> * Added support for inline assembly checking.
> * Added new tests for inline assembly sysreg checking.
> ---
> Hi All,
> On Sep 17, 2025, at 12:08, Jakub Jelinek wrote:
>
> Hi!
>
> -ftrivial-auto-var-init= also calls .DEFERRED_INIT for temporaries
> if they have VOID_TYPE_P TARGET_EXPR_INITIAL (otherwise they are expanded
> as INIT_EXPR of the TARGET_EXPR_SLOT and TARGET_EXPR_INITIAL and so
> everything is ini
On Mon, 18 Aug 2025 at 19:30, Christophe Lyon
wrote:
>
> We get lots of error messages when compiling arm_neon.h under
> e.g. -mcpu=cortex-m55, because Neon builtins are enabled only when
> !TARGET_HAVE_MVE. This has been the case since MVE support was
> introduced.
>
> This patch uses an approac
On 9/15/25 12:51 PM, Alexandre Oliva wrote:
An aarch64 toolchain built with --enable-default-pie fails
gcc.target/aarch64/sme/nonlocal_goto_[123].c because the register
allocator ends up resetting equivalences, so it concludes it needs to
preserve a rematerializable function invariant across a
Successfully bootstrapped & regrtested on x86_64-pc-linux-gnu.
Pushed to trunk as r16-3959-g423d7b24e7cff2.
gcc/ChangeLog:
* diagnostics/client-data-hooks.h (client_data_hooks::dump): New.
* diagnostics/context.cc (context::dump): Dump
m_client_data_hooks.
(client_d
Modernization; no functional change intended.
Successfully bootstrapped & regrtested on x86_64-pc-linux-gnu.
Pushed to trunk as r16-3958-g21fe45f111cf53.
gcc/ChangeLog:
* diagnostics/paths-output.cc: Update for conversion of
location_aspect to enum class.
* diagnostics/sou
> On Sep 18, 2025, at 15:39, Kees Cook wrote:
>
> On Wed, Sep 17, 2025 at 02:09:54PM -0700, Kees Cook wrote:
>> On Wed, Sep 17, 2025 at 01:42:32PM +, Qing Zhao wrote:
On Sep 13, 2025, at 19:23, Kees Cook wrote:
+- Keep indirect calls from being merged (see earlier example) by
>>
On Mon, 15 Sep 2025, Eczbek wrote:
> From d70ac2b0a7b524829d22b3bb6bacbe1ef381c7fb Mon Sep 17 00:00:00 2001
> From: Eczbek
> Date: Mon, 15 Sep 2025 19:37:20 -0400
> Subject: [PATCH] c++: Fix lambdas with variadic parameters and static
> specifier [PR119048]
>
> PR c++/119048
>
> gcc/cp/C
On Wed, Sep 17, 2025 at 02:09:54PM -0700, Kees Cook wrote:
> On Wed, Sep 17, 2025 at 01:42:32PM +, Qing Zhao wrote:
> > > On Sep 13, 2025, at 19:23, Kees Cook wrote:
> > > +- Keep indirect calls from being merged (see earlier example) by
> > > + checking the KCFI insn's typeid for equality.
>
Hi Joseph,
(sorry for the long delay)
On 9/2/25 11:14, Joseph Myers wrote:
> On Fri, 22 Aug 2025, David Faust wrote:
>
>> +{
>> + /* Treat btf_type_tag applied to a function type as applying to the
>> + return type instead. Otherwise with GNU syntax there is no way to
>> + appl
On Thu, Sep 18, 2025 at 06:48:03PM +, Qing Zhao wrote:
>
>
> > On Sep 18, 2025, at 14:20, Kees Cook wrote:
> >
> > +- External functions that are address-taken have a weak
> > __kcfi_typeid_$func
> > + symbol added with the typeid value available so that the typeid can
>
Bootstrapped and regtested on x86_64-pc-linux-gnu, does this look
OK for trunk and eventually backports?
-- >8 --
Here the normal form of the two immediately-declared D<, V>
constraints is the same, so we rightfully share the normal form between
them. However we first normalize the constraint fr
Hi,
This patch adds gather/scatter handling for grouped access. The idea is
to e.g. replace an access (for uint8_t elements) like
arr[0]
arr[1]
arr[2]
arr[3]
arr[0 + step]
arr[1 + step]
...
by a gather load of uint32_t
arr[0]
arr[0 + step * 1]
arr[0 + step * 2]
...
where the offset ve
On Tue, Sep 9, 2025 at 3:31 AM wrote:
>
> From: Pan Li
>
> The widening-mul will insert a cast for the widen-mul, the
> function build_and_insert_cast is design to take care of it.
>
> In some case the optimized gimple has some unnecessary cast,
> for example as below code.
>
> #define SAT_U_MU
If the results include several configurations (schedule of
variations), do not report summary lines as duplicates. Indeed with
several configurations, it's likely that the results contain the same
# of expected passesX
The patch just keeps lines starting with test state prefix
Hi,
Following is version 2 of the patch proposed for master aiming to fix
PR104116. This has been bootstrapped and regtested on powerpc64le with
regression failures.
Kindly review.
Just had one question.
If I have to implement similar thing for {CEIL, ROUND}_{MOD,DIV} can I
create a separate patc
> On Sep 18, 2025, at 14:20, Kees Cook wrote:
>
> +- External functions that are address-taken have a weak
> __kcfi_typeid_$func
> + symbol added with the typeid value available so that the typeid can be
> + referenced from assembly linkages, etc, where the typeid values cann
Am Donnerstag, dem 18.09.2025 um 11:09 -0700 schrieb Kees Cook:
> On Thu, Sep 18, 2025 at 09:20:52AM +0200, Martin Uecker wrote:
> > Am Mittwoch, dem 17.09.2025 um 17:56 + schrieb Qing Zhao:
> > > Hi,
> > >
> > > > On Sep 13, 2025, at 19:23, Kees Cook wrote:
> > > >
> > > > To support the K
On Thu, Sep 18, 2025 at 04:59:56PM +, Qing Zhao wrote:
>
>
> > On Sep 17, 2025, at 17:09, Kees Cook wrote:
> >
> > On Wed, Sep 17, 2025 at 01:42:32PM +, Qing Zhao wrote:
> >> This version of the middle-end change is much simpler and cleaner-:).
> >
> > Thanks! I think it's getter close
Hi gcc-patches mailing list,
cl...@gcc.gnu.org has requested that the following forgejo pull request
be published on the mailing list.
Created on: 2025-09-03 07:45:35+00:00
Latest update: 2025-09-18 13:25:21+00:00
Changes: 1 changed files, 47 additions, 0 deletions
Head revision: clyon/gcc-TEST re
On 9/1/2025 4:51 PM, via Sourceware Forge wrote:
From: Christophe Lyon
Add a sample workflow for Forgejo, as an example of integrated CI.
To keep it lightweight, we run only two small checks on each patch of
the series:
- contrib/check_GNU_style.py
which catches common mistakes (spaces
On Mon, Aug 25, 2025 at 5:59 AM H.J. Lu wrote:
>
> On Fri, Jun 27, 2025 at 05:53:57AM +0800, H.J. Lu wrote:
> > Here is the v3 patch. The difference from v2 is to use
> >
> > if (MEM_P (src)
> > && MEM_EXPR (src)
> > && (TREE_CODE (get_base_address (MEM_EXPR
On Mon, Aug 25, 2025 at 6:06 AM H.J. Lu wrote:
>
> On Thu, Jun 26, 2025 at 3:32 PM H.J. Lu wrote:
> >
> > On Thu, May 8, 2025 at 11:11 AM H.J. Lu wrote:
> > >
> > > Conditional and unconditional branch targets can be either a label or
> > > a symbol. For conditional jump:
> > >
> > > (jump_insn
On Thu, Sep 18, 2025 at 09:20:52AM +0200, Martin Uecker wrote:
> Am Mittwoch, dem 17.09.2025 um 17:56 + schrieb Qing Zhao:
> > Hi,
> >
> > > On Sep 13, 2025, at 19:23, Kees Cook wrote:
> > >
> > > To support the KCFI typeid and future type-based allocators,
>
> What I find problematic thou
On Thu, 11 Sep 2025, Tamar Christina wrote:
> > -Original Message-
> > From: Richard Biener
> > Sent: Thursday, September 11, 2025 12:56 PM
> > To: Tamar Christina
> > Cc: Robin Dapp ; GCC Patches > patc...@gcc.gnu.org>; rdsandif...@googlemail.com
> > Subject: Re: New optabs and IFN req
> Does this fix https://gcc.gnu.org/bugzilla/show_bug.cgi?id=115866 ?
Ah yes, thanks for reminding.
I will wait for review comments first, and add the PR reference in my
next version then.
Thanks,
Pengfei
From: Viljar Indus
gcc/ada/ChangeLog:
* ghost.adb (Check_Ghost_Policy): Update coding style.
Tested on x86_64-pc-linux-gnu, committed on master.
---
gcc/ada/ghost.adb | 11 +--
1 file changed, 5 insertions(+), 6 deletions(-)
diff --git a/gcc/ada/ghost.adb b/gcc/ada/ghost.adb
On 9/10/25 4:48 AM, Anton Blanchard wrote:
Hi Robin,
diff --git a/gcc/config/riscv/tt-ascalon-d8.md
b/gcc/config/riscv/tt-ascalon-d8.md
index a57c0b31a81..25b99b6129e 100644
--- a/gcc/config/riscv/tt-ascalon-d8.md
+++ b/gcc/config/riscv/tt-ascalon-d8.md
@@ -285,38 +285,38 @@ (define_insn_re
v1 -> v2 changes:
* Fixed the wrong conditional check.
* Fixed the typos in the testcases.
* Added support for inline assembly checking.
* Added new tests for inline assembly sysreg checking.
---
Hi All,
In the current Binutils we have disabled the feature gating for sysreg
by def
On Wed, Sep 17, 2025 at 10:01 AM Thomas Koenig wrote:
>
> Hi Steve,
>
> > On Tue, Sep 16, 2025 at 10:10:16PM +0200, Thomas Koenig wrote:
> >> Hello world,
> >>
> >> the attached patch implements a new option, -fexternal-blas64,
> >> so people can use 64-bit libraries for external BLAS, like Intel
On Thu, Sep 18, 2025 at 06:54:06PM +0200, Jason Merrill wrote:
> > There are some regressions caused by the removal of {CLOBBER(bob)}
> > clobbers from the start of certain constructors, e.g. one testcase has
> > struct A
> > {
> >int f,g;
> >A() { f = g; // { dg-warning "g. is used uniniti
Am 18.09.25 um 19:02 schrieb H.J. Lu:
FAIL: gfortran.dg/matmul_blas_3.f90 -O scan-tree-dump-not original
"integer\\(kind=4\\)"
Hm... would
! { dg-require-effective-target lp64 }
be the right way to fix this?
Best regards
Thomas
> > Hi,
> >
> > Fix ICE with AutoFDO by adding initialization check
> > before accessing IPA counts to avoid issues with uninitialized profile
> > counts in self-recursive clone processing.
>
> + n->make_profile_local ();
> + if (n->count.quality () == AFDO)
> +
> On Sep 17, 2025, at 17:09, Kees Cook wrote:
>
> On Wed, Sep 17, 2025 at 01:42:32PM +, Qing Zhao wrote:
>> This version of the middle-end change is much simpler and cleaner-:).
>
> Thanks! I think it's getter closer (hopefully). :)
>
>>> On Sep 13, 2025, at 19:23, Kees Cook wrote:
>>> +
On 9/17/25 6:08 PM, Jakub Jelinek wrote:
Hi!
Here is my current version of the C++26 P2795R5 support.
This now includes redirecting of forward/backward gotos
which cross vacuous initializations for -std=c++26 and
-ftrivial-auto-var-init={zero,pattern} by adding an artificial
if (0) { lab1: v1 =
Hi Paul!
Am 18.09.25 um 11:10 schrieb Paul Richard Thomas:
pdt_33.f03 needed a bit of a white space tidy up and the missed
Can you elaborate?
I don't see anything wrong with it, and it's the kind of programming
style of the reporter of the related PR.
On 17/09/25 2:10 pm, Avinash Jayakar wrote:
> Hi,
>
> Following is version 5 of the patch proposed for master to fix PR119702.
> This patch has been bootstrapped and regtested on powerpc64le-linux.
> Kindly review the patch.
>
> Thanks and regards,
> Avinash Jayakar
>
> Changes from v4:
>
On 9/18/25 06:33, Zhongyao Chen wrote:
From: Zhongyao Chen
The vectorizer's compute_estimated_lmul function could previously
return a bad value when the estimated lmul was larger than RVV_M8.
This is corrected to return RVV_M8, preventing a register spill.
The patch includes a new regressio
On 9/18/25 2:10 AM, Paul Richard Thomas wrote:
Hi All,
The recent change of PDT constructors to allow "structure-constructor
is type-name [ ( type-param-spec-list ) ]( [ component-spec-list ] )"
missed the possibility that the (type-param-spec-list) is optional, if
default values are available.
On Thu, Sep 18, 2025 at 5:34 AM Pengfei Li wrote:
>
> This patch adds vectorization support to early-break loops with gswitch
> statements. Such gswitches may come from original switch-case constructs
> in the source or the iftoswitch pass which rewrites if conditions with a
> chain of comparisons
On Tue, 16 Sep 2025 at 14:02 -0400, Nathan Myers wrote:
Changes in v14:
* Remove dependence on __STRICT_ANSI__, which is turned on by
"-std=c++26".
Changes in v13:
* Spell __STRICT_ANSI__ correctly.
Changes in v12:
* use __builtin_popcountg() which works on all unsigned integer types.
Changes
Hi!
The following testcase ICEs in functions called from
lower_omp_regimplify_operands_p, because maybe_lookup_decl returns
NULL for this (on the outer taskloop context) when regimplifying the
taskloop pre body. If it isn't found in current context, we should
look in outer ones.
Tested on x86_64
Hi Tobias,
On Tue, Sep 16, 2025 at 5:11 PM Tobias Burnus wrote:
> PS: Already with the current code, we may run into the issue of passing
> an actual argument like '(cond ? "abc" : "cdfg")' to 'class(*)' – and I
> am not sure whether we handle this correctly or not.
That is a great test case, an
Changes in v11:
* Fix "#if 202306L" cases in synopsis test.
* Add noexcept specifications in synopsis test.
Changes in v10:
* Merge external updates.
* Use long literals "202306L" in #if guards.
* Remove redundant #include .
* Test constexpr not_fn .
* Fix return type declarations in synop
From: Christophe Lyon
Add a sample workflow for Forgejo, as an example of integrated CI.
To keep it lightweight, we run only two small checks on each patch of
the series:
- contrib/check_GNU_style.py
which catches common mistakes (spaces vs tab, missing spaces, ...)
but has some false posit
From: Zhongyao Chen
The vectorizer's compute_estimated_lmul function could previously
return a bad value when the estimated lmul was larger than RVV_M8.
This is corrected to return RVV_M8, preventing a register spill.
The patch includes a new regression test for PR target/121910, based
on the x2
This patch adds vectorization support to early-break loops with gswitch
statements. Such gswitches may come from original switch-case constructs
in the source or the iftoswitch pass which rewrites if conditions with a
chain of comparisons, like below, to gswitch statements.
if (a[i] == c1
On Thu, Sep 18, 2025 at 1:21 PM Robin Dapp wrote:
>
> Hi,
>
> This patch adds an explicit variant of vect_transform_slp_perm_load that
> just does the analysis part of vect_transform_slp_perm_load.
>
> I find it slightly clearer to indicate "analysis" in the
> function name already rather than hav
Hi,
This patch adds an explicit variant of vect_transform_slp_perm_load that
just does the analysis part of vect_transform_slp_perm_load.
I find it slightly clearer to indicate "analysis" in the
function name already rather than having to pass "analyze_only = true"
and set two other params to NUL
The following restricts the number of locations we register a predicate
as valid which avoids the expensive linear search for cases like
if (a)
A;
if (a)
B;
if (a)
C;
...
where we register a != 0 as true for locations A, B, C ... in an
unlimited way. The patch simply c
Hi All,
The following patch has been bootstrapped and regtested on powerpc64le-linux.
PowerPC vector shift left instructions (vslb, vslh, vslw, vsld) use modulo
semantics for the shift amount. Shifts by (element_bit_width - 1) can be
optimized by replacing the shift amount splat with a vector of
Sorry for the long delay.
SP_DERIVED_VALUE_P still deals with stack-pointer related situations, and since
there have been optimizations and fixes targeting stack access specifically,
it does not seem correct to simply replace it. (Actually there were regressions
when I tried.)
As for the debug i
Hi All,
The recent change of PDT constructors to allow "structure-constructor
is type-name [ ( type-param-spec-list ) ]( [ component-spec-list ] )"
missed the possibility that the (type-param-spec-list) is optional, if
default values are available. This patch permits this by looking for a
left han
On Sat, 13 Sep 2025, Alejandro Colomar wrote:
> @@ -675,7 +675,7 @@ declaration}, and it serves the purpose of making the
> name @code{len}
> known when the declaration of @code{data} is parsed.
>
> You can write any number of such parameter forward declarations in the
> -parameter list. The
While working on a different patch, Sandra found out that
due to tree sharing, the user condition in a context
selector was only evaluated once - which is bad if it is
dynamic (here: global variable) that can be changed.
The included testcase is a slightly enhanced version of
hers, attached to th
Am Mittwoch, dem 17.09.2025 um 17:56 + schrieb Qing Zhao:
> Hi,
>
> > On Sep 13, 2025, at 19:23, Kees Cook wrote:
> >
> > To support the KCFI typeid and future type-based allocators,
>
> Could you please explain a little bit more on the “future type-based
> allocators”?
People work on al
91 matches
Mail list logo
