On Thu, Sep 11, 2025 at 6:06 PM Robin Dapp wrote:
>
> > Hmm, so the existing "punning" code for VMAT_STRIDED_SLP does
> >
> > tree vtype
> > = vector_vector_composition_type (vectype, const_nunits / n,
> > &ptype);
> >
Thank you
On Thu, Sep 11, 2025 at 2:21 AM H.J. Lu wrote:
> On Mon, Sep 8, 2025 at 7:47 AM H.J. Lu wrote:
> >
> > -mno-direct-extern-access is used to disable direct access to external
> > symbol from executable with and without PIE for x86. Require PIE and
> > pass -fPIE to disable direct acce
Kwok Cheung Yeung wrote on July 9, 2025:
Subject: [PATCH 05/11] openmp, fortran: Add support for iterators in OpenMP
'target update' constructs (Fortran)
This adds Fortran support for iterators in 'to' and 'from' clauses in the
'target update' OpenMP directive.
...
@@ -1418,16 +1419,67 @@ gf
> FAIL: gcc.target/i386/pr116896.c scan-assembler-times \tjp\t 2
> FAIL: g++.target/i386/pr116896-1.C -std=gnu++20 scan-assembler-times \tjp\t
> 1
> FAIL: g++.target/i386/pr116896-1.C -std=gnu++23 scan-assembler-times \tjp\t
> 1
> FAIL: g++.target/i386/pr116896-1.C -std=gnu++26 scan-assembl
On Wed, Sep 10, 2025 at 10:54 AM Jonathan Wakely wrote:
> On Wed, 10 Sept 2025 at 09:52, Jonathan Wakely wrote:
> >
> > On Thu, 4 Sept 2025 at 14:39, Tomasz Kaminski
> wrote:
> > >> + if constexpr (is_same_v<_Out, _Iter_for_t<_CharT>>)
> > >
> > > This would be better if we use _Drop_iter
On Linux/x86_64,
9ff5cadac4579fe714dc8de3a39eaef8534c64f7 is the first bad commit
commit 9ff5cadac4579fe714dc8de3a39eaef8534c64f7
Author: Robin Dapp
Date: Tue Sep 9 16:15:48 2025 +0200
ifcvt: Clarify if_info.original_cost.
caused
FAIL: gcc.target/i386/pr116896.c scan-assembler-times \tjp
Does this fix seem reasonable, or is there something I've missed?
My change to g++.dg/lto/pr101396_0.C also causes it to fail link with
some flags on my machine, with:
FAIL: g++.dg/lto/pr101396 cp_lto_pr101396_0.o-cp_lto_pr101396_1.o link, -O0
-flto -flto-partition=none -fuse-linker-plugin
FAIL:
gcc/ada/ChangeLog:
* env.c (__gnat_clearenv): Adjust comment.
* libgnarl/a-intnam__bsd.ads: Fix copyright date.
Tested on x86_64-pc-linux-gnu, committed on master.
---
gcc/ada/env.c | 3 ++-
gcc/ada/libgnarl/a-intnam__bsd.ads | 2 +-
2 files changed, 3 inser
> -Original Message-
> From: Robin Dapp
> Sent: Thursday, September 11, 2025 1:03 PM
> To: Richard Biener ; Tamar Christina
>
> Cc: rdapp@gmail.com; gcc-patches@gcc.gnu.org;
> rdsandif...@googlemail.com
> Subject: Re: New optabs and IFN required for early break [bikeshed]
>
> > So AV
This patch implements _Escaping_sink that stores characters in a local (stack)
buffer. When the buffer is full, the range of characters is escaped and written
to the underlying sink.
To support above, the __write_escaped_unicode_part function are defined.
It takes __str and __prev_esc by reference
On Tue, Sep 09, 2025 at 08:49:51PM +0200, Harald Anlauf wrote:
> Hi Jerry!
>
> Am 08.09.25 um 23:56 schrieb Jerry D:
> > On 9/8/25 12:59 PM, Harald Anlauf wrote:
> > > Dear all,
> > >
> > > this is the second (and hopefully final) patch to fix this PR for good.
> > > This makes the GNU intrinsics
From: Ronan Desplanques
The comment this patch changes was made incorrect by the possibility of
completing an incomplete view with a private type declaration in Ada
2012. To avoid any possible confusion, this patch brings the comment up
to date.
gcc/ada/ChangeLog:
* sem_ch3.adb (Find_Ty
On Mon, 8 Sep 2025, H.J. Lu wrote:
> Don't check tls_model attribute when optimizing TLS access since
>
> 1. -ftls-model=initial-exec also can specify tls_model without tls_model
> attribute.
There's a difference: -ftls-model is usable only for upgrading the model;
if you pass -ftls-model=glob
On Thu, Sep 11, 2025 at 09:29:35AM +0200, Peter Zijlstra wrote:
> On Wed, Sep 10, 2025 at 08:05:11PM -0700, Kees Cook wrote:
>
> > > > +/* Check if a function needs KCFI preamble generation.
> > > > + ALL functions get preambles when -fsanitize=kcfi is enabled,
> > > > regardless
> > > > + of
On Fri, 5 Sep 2025, Richard Biener wrote:
> The PR reports
>
> vectorizer.h:276:3: runtime error: load of value 32695, which is not a valid
> value for type 'internal_fn'
>
> which I believe is from
>
> slp_node->data = new vect_load_store_data (std::move (ls));
>
> where 'ls' can be pa
SLP may take a broadcast as kind of vec_perm, the patch checks the
permutation index to exclude those false positive.
> > > so the vectorizer costs sth withy count == 0? I'll see to fix that,
> > > but this also
> > > means the code should have used m_num_avx256_vec_perm[where] += count.
Changed.
On Mon, 8 Sept 2025 at 14:38, Tomasz Kaminski wrote:
>
>
>
> On Mon, Sep 8, 2025 at 2:45 PM Jonathan Wakely wrote:
>>
>> This adds checks when incrementing the shared count and weak count and
>> will trap if they would be
>> be incremented past its maximum. The maximum value is the value at which
> 70a27555d900caf71932a2c794d9a77d06048b6a..b677a5308fb3498314cbbfc1
> 865eae6bf1c815b9 100644
> > --- a/gcc/doc/md.texi
> > +++ b/gcc/doc/md.texi
> > @@ -6087,6 +6087,25 @@ vectors with N signed/unsigned elements of size
> S@. Find the absolute
> > difference between operands 1 and 2 and widen t
Hi Suwa-san,
On Thu, Sep 11, 2025 at 1:48 PM Takayuki 'January June' Suwa
wrote:
>
> If the Floating-Point Coprocessor option is configured and there are no
> floating-point coprocessor instructions in the function, the unused
> floating-point registers (f0 thru f15) can be good targets for spill
> -Original Message-
> From: Wilco Dijkstra
> Sent: Thursday, September 4, 2025 4:32 PM
> To: Kyrylo Tkachov ; Alex Coplan
> ; Tamar Christina ; Andrew
> Pinski ; Alice Carlotti
> Cc: GCC Patches
> Subject: [PATCH] AArch64: Add isnan expander [PR 66462]
>
>
> Add an expander for isn
Hi!
On 2025/09/12 14:14, Max Filippov wrote:
Hi Suwa-san,
On Thu, Sep 11, 2025 at 1:48 PM Takayuki 'January June' Suwa
wrote:
If the Floating-Point Coprocessor option is configured and there are no
floating-point coprocessor instructions in the function, the unused
floating-point registers (
On Linux/x86_64,
6456da6bab8a2c43e7899afda991589065d96595 is the first bad commit
commit 6456da6bab8a2c43e7899afda991589065d96595
Author: Jonathan Wakely
Date: Thu Jul 17 22:02:45 2025 +0100
libstdc++: Make atomicity helpers use unsigned arithmetic [PR121148]
caused
FAIL: g++.dg/analyzer
Ok，I will try to implement it in QEMU and send the second version of
the patch.
在 2025/9/11 21:18, Jeff Law 写道:
On 9/11/25 1:28 AM, Kito Cheng wrote:
This Patch implements the Quad-precision Float extension for RISC-V,
based on
Kito's work five years ago in
https://github.com/riscvarchive/
From: Viljar Indus
gcc/ada/ChangeLog:
* ghost.adb (Mark_Ghost_Declaration_Or_Body): Mark ghost
entity explicitly as ignored or checked.
Tested on x86_64-pc-linux-gnu, committed on master.
---
gcc/ada/ghost.adb | 37 +
1 file changed, 33 inse
From: Eric Botcazou
This is a follow-up to a recent change, where a warning was implemented
for huge library-level objects. However it is not given if the objects
are imported, although an indirection is also added for them under the
hood to match the export side.
gcc/ada/ChangeLog:
*
On Sat, 6 Sep 2025, Martin Uecker wrote:
>
> This is the revised version adding the test to C_DECL_REGISTER.
> This meant I also had to change one place where it is set on
> function declarators before it becomes an error.
>
> Bootstrapped and regression tested on x86_64.
>
> Martin
>
>
>
From: Viljar Indus
Declarations should take the ghost policy from the region if
they do not have an explicit ghost aspect/pragam themselves.
gcc/ada/ChangeLog:
* ghost.adb (Mark_And_Set_Ghost_Declaration): apply the
ghost policy and level from the declaration only if the declara
Whenever we use operator+ or similar operators on random access
iterators we need to be careful to use the iterator's difference_type
rather than some other integer type. It's not guaranteed that an
expression with an arbitrary integer type, such as `it + 1u`, has the
same effects as `it + iter_dif
On Wed, 10 Sep 2025, Nathaniel Shead wrote:
> Does this fix seem reasonable, or is there something I've missed?
>
> My change to g++.dg/lto/pr101396_0.C also causes it to fail link with
> some flags on my machine, with:
>
> FAIL: g++.dg/lto/pr101396 cp_lto_pr101396_0.o-cp_lto_pr101396_1.o link,
Excerpts from Iain Sandoe's message of September 6, 2025 9:27 pm:
> Tested on x86_64-darwin21 (has __fork) and x86_64-darwin10 (does not).
> OK for trunk?
> thanks
> Iain
>
Looks OK.
Thanks,
Iain
On 9/9/25 15:54, Tomasz Kaminski wrote:
On Mon, Sep 8, 2025 at 10:14 PM Luc Grosheintz
wrote:
I decided to deviate once (described in detail in the commit message).
Without this deviation writing tests is very clumsy and sometimes not
possible, e.g. when mixing uint8_t and dynamic padding v
On 9/11/25 17:05, Luc Grosheintz wrote:
Static checks already cover this.
but what if `_M_exents is completely dynamic, e.g.
std::extents{128, 4}
I see, we can constexpr if the check away for fully
static extents.
On 9/10/25 2:41 AM, Robin Dapp wrote:
> So IMHO we shouldn't complicate it and the patch is OK. The change is
> reasonable and reflects what the hardware does (one could argue it's even
> better than the mode) - latency depends on the element size.
Ok, given the bootstrap/regtest came back clea
After
commit 8cad8f94b450be9b73d07bdeef7fa1778d3f2b96
Author: H.J. Lu
Date: Fri Sep 5 15:40:51 2025 -0700
c: Update TLS model after processing a TLS variable
GCC will upgrade local-dynamic TLS model to local-exec without -fPIC.
Compile TLS LD tests with -fPIC to keep local-dynamic TLS mod
The length returned by vect_get_loop_len is REALLEN + BIAS, but was
assumed to be REALLEN - BIAS. If BIAS is -1, this leads to wrong
code.
Bootstrapped and regtested on s390. Ok for trunk?
gcc/ChangeLog:
* tree-vect-loop.cc (vectorizable_live_operation_1): Fix
load/store bias h
Hi all,
attached is a patch to update the Fortran section of the GCC 16
release notes. I am sure that I missed some fine print and the
most important changes are also some bug fixes, which are difficult
to list. → https://gcc.gnu.org/gcc-16/changes.html
Comments?
Tobias
PS: I am aware of addit
In libstdc++ the prefix _S is used for static members only. In
there's several type aliases that also used the prefix _S. They now use
double underscore instead.
libstdc++-v3/ChangeLog:
* include/std/mdspan (_ExtentsStorage::__base): New name for
_S_base.
(_ExtentsStorage
Bootstrapped and regtested on x86_64-pc-linux-gnu, OK for trunk?
-- >8 --
A static member template doesn't always have a DECL_INITIAL, as in the
below testcase, and so checking its TREE_CODE performs a null-pointer
dereference. I don't think we need to specially detect this case as
traits we car
This email is to discuss the optabs and IFNs that I will need to finish off the
early break optimizations in GCC 16 and to minimize respins on the final
patches.
Optabs only, These will be handled through expand rewriting them if the targets
supports them:
-- cbranch
At the moment we model vect
LGTM, I want to do that for a while but I have not found enough time
yet, thanks :)
On Mon, Sep 8, 2025 at 7:27 PM Jiawei wrote:
>
> Moving RISC-V Profiles definations into 'riscv-profiles.def'. Add comments for
> 'riscv_profiles'.
>
> gcc/ChangeLog:
>
> * common/config/riscv/riscv-common
On Thu, Sep 11, 2025 at 11:08:54AM +0200, Richard Biener wrote:
> On Thu, 11 Sep 2025, Richard Biener wrote:
>
> > On Wed, 10 Sep 2025, Nathaniel Shead wrote:
> >
> > > Does this fix seem reasonable, or is there something I've missed?
> > >
> > > My change to g++.dg/lto/pr101396_0.C also causes
- Am 10. Jul 2025 um 7:53 schrieb Sebastian Huber
sebastian.hu...@embedded-brains.de:
> There are targets, which only offer 32-bit atomic operations (for
> example 32-bit RISC-V). For these targets, split the 64-bit atomic
> bitwise-or operation into two parts.
Could someone please have a l
在 2025-9-12 08:50, Jonathan Yong 写道:
On 9/10/25 3:29 AM, Peter Damianov wrote:
UTF-8 characters in diagnostic output (such as the warning emoji ⚠️
used by fanalyzer) display as mojibake on Windows unless the utf8
code page is being used
This patch adds UTF-8 to UTF-16 conversion when outputting
Whenever we use operator+ or similar operators on random access
iterators we need to be careful to use the iterator's difference_type
rather than some other integer type. It's not guaranteed that an
expression with an arbitrary integer type, such as `it + 1u`, has the
same effects as `it + iter_dif
Le 11/09/2025 à 20:28, Harald Anlauf a écrit :
Dear all,
here's a - once found - seemingly simple and obvious fix for a memory
corruption happening when intrinsic assignment is used to set a scalar
allocatable polymorphic component of a derived type when the latter
is instanciated as an array of
On 9/10/25 3:29 AM, Peter Damianov wrote:
UTF-8 characters in diagnostic output (such as the warning emoji ⚠️
used by fanalyzer) display as mojibake on Windows unless the utf8
code page is being used
This patch adds UTF-8 to UTF-16 conversion when outputting to a console
on Windows.
gcc/ChangeL
From: Bob Duff
Previous change, "Make pp and friends more robust (base type only)"
introduced a bug in Root_Type. Etype (T) can, in fact, be Empty
(but only in case of errors.) This patch fixes it.
gcc/ada/ChangeLog:
* einfo-utils.adb (Root_Type): Deal with missing Etype.
(Root_
I am writing this follow-up email to specify that I executed the tests
contained in this patch on aarch64-arm64-linux-gnu
Best regards,
Matteo
> On Sep 9, 2025, at 10:39 PM, Matteo Nicoli wrote:
>
> Dear reviewers,
>
> I attached the patch for this Bugzilla issue:
> https://gcc.gnu.org/bugz
The test part is still OK of course.
--
Regards
Robin
Using std::move(*it) is incorrect for iterators that use proxy refs, we
should use ranges::iter_move(it) instead.
libstdc++-v3/ChangeLog:
PR libstdc++/121913
* include/bits/ranges_algo.h (__rotate_fn::operator()): Use
ranges::iter_move(it) instead of std::move(*it).
If the Floating-Point Coprocessor option is configured and there are no
floating-point coprocessor instructions in the function, the unused
floating-point registers (f0 thru f15) can be good targets for spilling
address (integer) registers.
For reference, the ISA manual says that [RW]FR machine i
From: Javier Miranda
The compilation of files b3a10041.ads and b3a10042.adb crash when
the compiler is built with assertions enabled.
gcc/ada/ChangeLog:
* freeze.adb (Freeze_Entity): Protect call to Associated_Storage_Pool
since it cannot be used when the Etype is not set.
On Tue, Sep 2, 2025 at 1:39 PM Joseph Myers wrote:
>
> On Wed, 28 May 2025, Trevor Gross wrote:
>
> > Documentation for `__cmpsf2` and similar functions currently indicate a
> > return type of `int`. This is not correct however; the `libgcc`
> > functions return `CMPtype`, the size of which is det
Hi Jerry!
Am 11.09.25 um 21:30 schrieb Jerry D:
On 9/11/25 11:28 AM, Harald Anlauf wrote:
Dear all,
here's a - once found - seemingly simple and obvious fix for a memory
corruption happening when intrinsic assignment is used to set a scalar
allocatable polymorphic component of a derived type w
On Thu, Sep 11, 2025 at 1:09 PM Trevor Gross wrote:
> (Returning i128 indirectly and f128 in xmm0 would also match the SysV
> return ABI.)
Slight clarification here the SysV ABI says that i128 can be
represented as a 16-aligned struct:
typedef struct {
long low, high;
} __int128;
Dear all,
here's a - once found - seemingly simple and obvious fix for a memory
corruption happening when intrinsic assignment is used to set a scalar
allocatable polymorphic component of a derived type when the latter
is instanciated as an array of rank > 0. Just get the dimension
attribute rig
On Wed Sep 10, 2025 at 2:41 AM EDT, LIU Hao wrote:
> 在 2025-9-10 13:41, Trevor Gross 写道:
>> Just following up here, how would you like me to proceed LH?
>>
>
> I agree that `_Float16` should be passed in a vector register, because the
> hardware expects it, like in
> `vcvtph2ps xmm0, xmm1` and t
From: Johannes Kliemann
The QNX Certified Products Defect Notification from February 2025
mentions a potential memory leak when pthread_create is interrupted by a
signal. It recommends to disable signals for this function call.
gcc/ada/ChangeLog:
* adaint.c: Add functions to disable and
Hmm, so the existing "punning" code for VMAT_STRIDED_SLP does
tree vtype
= vector_vector_composition_type (vectype, const_nunits / n,
&ptype);
if (vtype != NULL_TREE)
{
On Thu, 11 Sep 2025, Alejandro Colomar wrote:
> Warn about this:
>
> void f(int x; int x; int x);
>
> This would allow eventually adding another semicolon in function
> prototypes to separate a different feature.
I don't think that's a plausible future addition. The small visual
differe
The handling of ISO week-calendar year specifiers (%G, %g) and ISO week
number (%V) was merged into a single _M_g_G_V function, as the latter
requires ISO year value, computed by the former.
The values for %U and %W, which are based on the number of days since the
first Sunday and Monday of the ye
From: Zhongyao Chen
users can now write code like the following to adapt to the
current RISC-V profile selected at compile time:
```c
#ifdef __riscv_rva23u64
// Code specific to the rva23u64 profile
#endif
```
Changes since v1:
add testcases
gcc/
* common/config/riscv/r
Seems like I also implement that on qemu many years ago:
https://github.com/kito-cheng/riscv-qemu/tree/q-ext-dev
Feel free to take it
On Thu, Sep 11, 2025 at 9:54 PM Liao Shihua wrote:
>
> Ok，I will try to implement it in QEMU and send the second version of
> the patch.
>
> 在 2025/9/11 21:18,
On Thu, 11 Sep 2025, Jonathan Wakely wrote:
> Whenever we use operator+ or similar operators on random access
> iterators we need to be careful to use the iterator's difference_type
> rather than some other integer type. It's not guaranteed that an
> expression with an arbitrary integer type, such
Whenever we use operator+ or similar operators on random access
iterators we need to be careful to use the iterator's difference_type
rather than some other integer type. It's not guaranteed that an
expression with an arbitrary integer type, such as `it + 1u`, has the
same effects as `it + iter_dif
Ah great! Does it just take a mask? could you point me to some docs?
You mean just a mask and not a length? No, (almost) all our insns are
length-masked. Specs are here:
https://github.com/riscvarchive/riscv-v-spec/blob/master/v-spec.adoc#165-vector-compress-instruction
--
Regards
Robin
On 9/11/25 14:41, Tomasz Kaminski wrote:
On Thu, Sep 11, 2025 at 1:50 PM Luc Grosheintz
wrote:
Changes since v1:
- Use cmp_* when comparing two integers of differnt types.
- Remove the FTM submdspan and replace it with an unofficial
purely internal FTM called padded_layouts.
-
libstdc++-v3/ChangeLog:
* include/std/syncstream: Remove trailing whitespace.
---
Pushed to trunk.
diff --git a/libstdc++-v3/include/std/syncstream
b/libstdc++-v3/include/std/syncstream
index 92fbbdc760e7..1e1759714c69 100644
--- a/libstdc++-v3/include/std/syncstream
+++ b/libstdc++-
On 9/11/25 1:28 AM, Kito Cheng wrote:
This Patch implements the Quad-precision Float extension for RISC-V, based on
Kito's work five years ago in
https://github.com/riscvarchive/riscv-gcc/tree/q-ext
Thanks for bringing this back, I almost forgot this still has not merged yet.
Note that I do
On Thu, Sep 11, 2025 at 2:18 PM Jonathan Wakely wrote:
> These _S_noexcept() functions are only used in noexcept-specifiers and
> never need to be called at runtime. They can be immediate functions,
> i.e. consteval.
>
> libstdc++-v3/ChangeLog:
>
> * include/bits/iterator_concepts.h (_Ite
These _S_noexcept() functions are only used in noexcept-specifiers and
never need to be called at runtime. They can be immediate functions,
i.e. consteval.
libstdc++-v3/ChangeLog:
* include/bits/iterator_concepts.h (_IterMove::_S_noexcept)
(_IterSwap::_S_noexcept): Change constexp
Most of the basis operations for ranges such as ranges::begin and
ranges::next are trivial one-line function bodies, so can be made
always_inline to reduce the abstraction penalty for -O0 code.
Now that we no longer need to support the -fconcepts-ts grammar, we can
also move some [[nodiscard]] att
On Fri, 5 Sept 2025 at 13:54, Richard Biener wrote:
>
> The PR reports
>
> vectorizer.h:276:3: runtime error: load of value 32695, which is not a valid
> value for type 'internal_fn'
>
> which I believe is from
>
> slp_node->data = new vect_load_store_data (std::move (ls));
>
> where 'ls' c
So AVX512 has vcompressp{d,s} and vexpandp{d,s} (but nothing for smaller
integer element types). Those could be used for this but they have
a vector result (and element zero would be the first active).
But don't you possibly want the last inactive as well, dependent on
whether this is a peeled/n
On Thu, Sep 11, 2025 at 1:48 PM Luc Grosheintz
wrote:
> For padded layouts we want to check that the product of the
> padded stride with the remaining extents is representable.
>
> Creating a second overload, allows passing in subspans of the
> static extents and retains the ergonomics for the co
On Thu, 11 Sep 2025, Tamar Christina wrote:
> This email is to discuss the optabs and IFNs that I will need to finish off
> the
> early break optimizations in GCC 16 and to minimize respins on the final
> patches.
>
> Optabs only, These will be handled through expand rewriting them if the
> ta
On Thu, Sep 11, 2025 at 1:42 PM Luc Grosheintz
wrote:
> libstdc++-v3/ChangeLog:
>
> * testsuite/23_containers/mdspan/layouts/mapping.cc
> (test_stride_1d): Fix test.
>
> Signed-off-by: Luc Grosheintz
> ---
LGTM.
> libstdc++-v3/testsuite/23_containers/mdspan/layouts/mapping.cc
libstdc++-v3/ChangeLog:
* testsuite/23_containers/mdspan/layouts/mapping.cc
(test_stride_1d): Fix test.
Signed-off-by: Luc Grosheintz
---
libstdc++-v3/testsuite/23_containers/mdspan/layouts/mapping.cc | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/libstdc++-
Changes since v1:
- Use cmp_* when comparing two integers of differnt types.
- Remove the FTM submdspan and replace it with an unofficial
purely internal FTM called padded_layouts.
- Make __get_static_stride consteval.
- Make __is_*_padded_layout only available inside
__glibcxx_padde
> -Original Message-
> From: Wilco Dijkstra
> Sent: Wednesday, September 10, 2025 5:25 PM
> To: Tamar Christina ; GCC Patches patc...@gcc.gnu.org>
> Cc: Kyrylo Tkachov ; Alex Coplan
> ; Andrew Pinski ; Alice Carlotti
>
> Subject: Re: [PATCH] AArch64: Add isfinite expander [PR 66462]
>
>
On Tue, Sep 9, 2025 at 10:39 PM Matteo Nicoli
wrote:
>
> Dear Richard,
>
> You are right, there was a typo in my patch. I removed the duplicate */ and
> updated the description of the test in the commit description.
Thanks. I have bootstrapped and tested this on x86_64-unknown-linux-gnu
(please
On Wed, 10 Sep 2025, Avinash Jayakar wrote:
> Hi,
>
> The following patch implements the vectorization logic for FLOOR_MOD_EXPR and
> FLOOR_DIV_EXPR. According to the logic mentioned in the PR, we have
> For signed operands,
> r = x %[fl] y;
On Mon, Sep 8, 2025 at 7:58 PM Edwin Lu wrote:
>
> This patch tries to add support for a variant of SAT_TRUNC where
> negative numbers are clipped to 0 instead of NARROW_TYPE_MAX_VALUE.
> This form is seen in x264, aka
>
> UT clip (T a)
> {
> return a & (UT)(-1) ? (-a) >> 31 : a;
> }
>
> Where s
On Mon, Sep 8, 2025 at 3:19 PM Robin Dapp wrote:
>
> 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 o
Hi Avinash,
On 08/09/25 2:54 pm, Avinash Jayakar wrote:
> Hi,
>
> This is the third version of the patch proposed for master aiming to fix
> PR119702. Requesting review of this patch.
>
> The following sequence of assembly in powerpc64le
> vspltisw 0,1
> vsld 2,2,0
> is replaced by
The handling of ISO week-calendar year specifiers (%G, %g) and ISO week
number (%V) was merged into a single _M_g_G_V function, as the latter
requires ISO year value, computed by the former.
The values for %U and %W, which are based on the number of days since the
first Sunday and Monday of the ye
Hi,
This patch adjusts vect_gather_scatter_fn_p to always check an offset
type with swapped signedness (vs. the original offset argument).
If the target supports the gather/scatter with the new offset type the
offset is converted to it before emitting the gather/scatter. In the
same way the cost
BACKLOG_MAX represents the number of outstanding connections in the
socket's listen queue.
gcc/ada/ChangeLog:
* libgnat/g-socket.adb (Listen_Socket): Change default value.
* libgnat/g-socket.ads (Listen_Socket): Likewise.
* s-oscons-tmplt.c (BACKLOG_MAX): New.
Tested on x
Hi,
Here's v3, this time a proper patch, tested for regressions, and with
tests and ChangeLog entries.
It only contains a trivial fix in the tests since v2. See a reply to v2
for the results of regression testing. You can check that the commit
hash of those tests corresponds to the commit hash
From: Piotr Trojanek
Fix grammar in comment.
gcc/ada/ChangeLog:
* einfo.ads (Ghost_Assertion_Level): Fix comment.
Tested on x86_64-pc-linux-gnu, committed on master.
---
gcc/ada/einfo.ads | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/gcc/ada/einfo.ads b/gcc/
From: Steve Baird
In the case of a call to a subprogram that has an out (or in-out) parameter
that is passed by copy, the caller performs copy-back after the call returns.
If the actual parameter is a view conversion to a subtype that has an enabled
predicate, then the predicate check performed a
From: Viljar Indus
It is OK to define a checked ghost type with an iterable aspect
that has ignored Iterable functions.
gcc/ada/ChangeLog:
* ghost.adb (Check_Ghost_Policy): Avoid triggering a ghost
policy error if the policy is referenced within the Iterable
aspect.
Tes
From: Eric Botcazou
The implementation is essentially mirrored from the one for signed types.
gcc/ada/ChangeLog:
* gcc-interface/gigi.h (standard_datatypes): Add ADT_uns_mulv64_decl
and ADT_uns_mulv128_decl.
(uns_mulv64_decl): New macro.
(uns_mulv128_decl): Likew
From: Bob Duff
Recent changes "Fix regression in Root_Type" and
"Crash on b3a1004 with assertions enabled" are partially
redundant; they are addressing the same bug.
This patch adjusts the former in the case of Root_Type.
But we leave Root_Type_If_Set alone; debugging printouts
should survive bug
From: Tonu Naks
gcc/ada/ChangeLog:
* libgnat/s-crtl.ads: define unsigned
* libgnat/s-crtl__mingw.adb (read, write): change arg type
Tested on x86_64-pc-linux-gnu, committed on master.
---
gcc/ada/libgnat/s-crtl.ads| 3 +++
gcc/ada/libgnat/s-crtl__mingw.adb | 8
From: Denis Mazzucato
This patch improves the warning message when the actual file name of a child
package with a single letter parent isn't the expected one because it may
collide with a predefined unit name. The warning explain why in this specific
case the expected name is not the standard one
From: Bob Duff
Prior to this fix, if pp(N) tried to print a "base type only" field, and
Base_Type(N) was not yet set, it would raise an exception, which was
confusing. This patch makes it simply ignore such fields. Similarly
for Impl_Base_Type_Only and Root_Type_Only fields.
We do this by having
From: Javier Miranda
A compiler built with assertions enabled crashes processing
a null aggregate of multidimensional type.
gcc/ada/ChangeLog:
* sem_aggr.adb (Report_Null_Array_Constraint_Error): Adjust code
for reporting the error on enumeration types.
(Resolve_Null_Arr
From: Denis Mazzucato
Unfolding of static expressions is needed when evaluating static bounds, even in
the presence of strict analysis. Otherwise, we may wrongly identify static
predicates as dynamic ones, and thus require unnecessary "others" default case.
gcc/ada/ChangeLog:
* sem_attr
From: Ronan Desplanques
The meaning of the return value of Find_Type_Name depends greatly on
whether the declaration it's passed is a completion. This patch adds a
description of this to the documentation comment of Find_Type_Name.
gcc/ada/ChangeLog:
* sem_ch3.ads (Find_Type_Name): Impr
1 - 100 of 114 matches
Mail list logo
