A number of folks have had their fingers in this code and it's going to
take a few submissions to do everything we want to do.
This patch is primarily concerned with avoiding signaling that fusion
can occur in cases where it obviously should not be signaling fusion.
Every DEC based fusion I'm
On Thu, 3 Jul 2025, 00:56 Nathan Myers, wrote:
> Changes in V2:
> * Generalize private member _M_check_initial_position for use with
> both string and string_view arguments.
> * Remove unnecessary #if guards for version and hostedness.
> * Remove redundant "std::" qualifications in new code.
>
On Wed, Jul 2, 2025 at 9:12 AM H.J. Lu wrote:
>
> While working on
>
> https://gcc.gnu.org/bugzilla/show_bug.cgi?id=120881
>
> I tried to use check-function-bodies to verify that
>
> 1: call mcount
>
> generated by "-pg" is placed at the function entry. Add "^[0-9]+:" to
> check-function-bodies t
On Thu, Jul 03, 2025 at 12:56:24AM +0800, Yuao Ma wrote:
>
> This patch adds the required function for Fortran trigonometric functions to
> work with glibc versions prior to 2.26. It's based on glibc source commit
> 632d895f3e5d98162f77b9c3c1da4ec19968b671.
>
> I've built it successfully on my en
The rules for conditional branches were spread throughout `aarch64.md`.
Group them together so it is easier to understand how `cbranch4`
is lowered to RTL.
gcc/ChangeLog:
* config/aarch64/aarch64.md (condjump): Move.
(*compare_condjump): Likewise.
(aarch64_cb1): Likewise.
Move the rules for CBZ/TBZ to be above the rules for
CBB/CBH/CB. We want them to have higher priority
because they can express larger displacements.
gcc/ChangeLog:
* config/aarch64/aarch64.md (aarch64_cbz1): Move
above rules for CBB/CBH/CB.
(*aarch64_tbz1): Likewise.
gcc/
On 7/2/25 9:40 AM, Jerry D wrote:
On 7/2/25 3:14 AM, Andre Vehreschild wrote:
Hi all,
I successfully created a big mess with the previous patch. First of all by
applying an outdated one and secondly by adding the conformance checks for
coranks in a3f1cdd8ed46f9816b31ab162ae4dac547d34ebc. Checki
Give the `define_insn` rules used in lowering `cbranch4` to RTL
more descriptive and consistent names: from now on, each rule is named
after the AArch64 instruction that it generates. Also add comments to
document each rule.
gcc/ChangeLog:
* config/aarch64/aarch64.md (condjump): Rename to
The rules for conditional branches were spread throughout `aarch64.md`.
Group them together so it is easier to understand how `cbranch4`
is lowered to RTL.
gcc/ChangeLog:
* config/aarch64/aarch64.md (condjump): Move.
(*compare_condjump): Likewise.
(aarch64_cb1): Likewise.
Add the `+cmpbr` option to enable the FEAT_CMPBR architectural
extension.
gcc/ChangeLog:
* config/aarch64/aarch64-option-extensions.def (cmpbr): New
option.
* config/aarch64/aarch64.h (TARGET_CMPBR): New macro.
* doc/invoke.texi (cmpbr): New option.
---
gcc/config
Move the rules for CBZ/TBZ to be above the rules for
CBB/CBH/CB. We want them to have higher priority
because they can express larger displacements.
gcc/ChangeLog:
* config/aarch64/aarch64.md (aarch64_cbz1): Move
above rules for CBB/CBH/CB.
(*aarch64_tbz1): Likewise.
gcc/
Add rules for lowering `cbranch4` to CBB/CBH/CB when
CMPBR extension is enabled.
gcc/ChangeLog:
* config/aarch64/aarch64-protos.h (aarch64_cb_rhs): New function.
* config/aarch64/aarch64.cc (aarch64_cb_rhs): Likewise.
* config/aarch64/aarch64.md (cbranch4): Rename to ...
Commit the test file `cmpbr.c` before rules for generating the new
instructions are added, so that the changes in codegen are more obvious
in the next commit.
gcc/testsuite/ChangeLog:
* lib/target-supports.exp: Add `cmpbr` to the list of extensions.
* gcc.target/aarch64/cmpbr.c: N
Extract the hardcoded values for the minimum PC-relative displacements
into named constants and document them.
gcc/ChangeLog:
* config/aarch64/aarch64.md (BRANCH_LEN_P_128MiB): New constant.
(BRANCH_LEN_N_128MiB): Likewise.
(BRANCH_LEN_P_1MiB): Likewise.
(BRANCH_LE
The `far_branch` attribute only ever takes the values 0 or 1, so make it
a `no/yes` valued string attribute instead.
gcc/ChangeLog:
* config/aarch64/aarch64.md (far_branch): Replace 0/1 with
no/yes.
(aarch64_bcond): Handle rename.
(aarch64_cbz1): Likewise.
Make the formatting of the RTL templates in the rules for branch
instructions more consistent with each other.
gcc/ChangeLog:
* config/aarch64/aarch64.md (cbranch4): Reformat.
(cbranchcc4): Likewise.
(condjump): Likewise.
(*compare_condjump): Likewise.
(aar
This patch series adds support for the CMPBR extension. It includes the
new `+cmpbr` option and rules to generate the new instructions when
lowering conditional branches.
Changelog:
* v9:
- Mark the non-far branches unlikely, so that the branch is consistently
generated as:
```asm
bra
On Thu, Jul 3, 2025 at 6:32 AM H.J. Lu wrote:
>
> *tls_local_dynamic_64_ uses RDI as the __tls_get_addr argument.
> Add RDI clobber to tls_local_dynamic_64 patterns to show it.
>
> PR target/120908
> * config/i386/i386.cc (legitimize_tls_address): Pass RDI to
> gen_tls_local_dynamic_64.
> * config
On Wed, Jul 2, 2025 at 11:32 PM 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 o
On Wed, Jul 2, 2025 at 7:17 PM David Faust wrote:
>
>
>
> On 7/2/25 00:35, Richard Biener wrote:
> > On Tue, Jul 1, 2025 at 11:20 PM David Faust wrote:
> >>
> >>
> >>
> >> On 7/1/25 01:02, Richard Biener wrote:
> >>> On Mon, Jun 30, 2025 at 9:12 PM David Faust
> >>> wrote:
>
>
>
*tls_global_dynamic_64_ uses RDI as the __tls_get_addr argument.
Add RDI clobber to tls_global_dynamic_64 patterns to show it.
PR target/120908
* config/i386/i386.cc (legitimize_tls_address): Pass RDI to
gen_tls_global_dynamic_64.
* config/i386/i386.md (*tls_global_dynamic_64_): Add RDI
clobber an
On Wed, Jul 02, 2025 at 11:06:38AM +0200, Jakub Jelinek wrote:
> On Tue, Jul 01, 2025 at 02:50:40PM -0500, Segher Boessenkool wrote:
> > No tests become good tests without effort. And tests that are not good
> > tests require constant maintenance!
>
> Here are two patches, either just the first o
[ Whoops, forgot to push the send button... ]
I've gone back and forth of these problems multiple times. We have two
passes, ext-dce and combine which eliminate extensions using totally
different mechanisms.
ext-dce looks for cases where the state of upper bits in an object
aren't observabl
On Wed, Jul 02, 2025 at 01:32:37PM +, Cui, Lili wrote:
> > > + /* Don't mess with the following registers. */ if
> > > + (frame_pointer_needed)
> > > +bitmap_clear_bit (components, HARD_FRAME_POINTER_REGNUM);
> >
> > What is that about? Isn't that one of the bigger possible wins?
>
>
On Wed, 11 Jun 2025, Jonathan Wakely wrote:
> As suggested by Jason, this makes all __normal_iterator operators into
> friends so they can be found by ADL and don't need to be separately
> exported in module std.
>
> The operator<=> comparing two iterators of the same type is removed
> entirely
On 7/2/25 3:14 AM, Andre Vehreschild wrote:
Hi all,
I successfully created a big mess with the previous patch. First of all by
applying an outdated one and secondly by adding the conformance checks for
coranks in a3f1cdd8ed46f9816b31ab162ae4dac547d34ebc. Checking the standard even
using AI (haha
Hi all,
This patch adds the required function for Fortran trigonometric
functions to work with glibc versions prior to 2.26. It's based on glibc
source commit 632d895f3e5d98162f77b9c3c1da4ec19968b671.
I've built it successfully on my end. Documentation is also included.
Please take a look wh
On Jul 02 2025, Stefan Schulze Frielinghaus wrote:
> I'm pretty new to tcl and didn't do extensive testing but for my few
> experiments it worked so far. I guess `string match` uses globbing so
> something like "* -f(no-)?stack-protector* *" doesn't work which is why
> I used two matches.
You ca
Hi all,
I successfully created a big mess with the previous patch. First of all by
applying an outdated one and secondly by adding the conformance checks for
coranks in a3f1cdd8ed46f9816b31ab162ae4dac547d34ebc. Checking the standard even
using AI (haha) to figure if coranks of an expression have r
On Wed, 2 Jul 2025 at 14:45, Mateusz Zych wrote:
>
> > Oh actually the radix members should be of type int, not bool. I can fix
> > that.
>
> Yes - thank you very much Jonathan for catching that!
> It was my honest oversight - I am so used to using auto, that I have
> accidentally copied the wro
On Mon, 9 Jun 2025, Jeremy Rifkin wrote:
> Hi,
> This fixes PR c/82134 which concerns gcc emitting an incorrect unused
> result diagnostic for empty types. This diagnostic is emitted from
> tree-cfg.cc because of a couple code paths which attempt to avoid
> copying empty types, resulting in GIMPLE
On Wed, 2 Jul 2025, Jonathan Wakely wrote:
> On Wed, 2 Jul 2025 at 15:29, Patrick Palka wrote:
> >
> >
> > On Wed, 11 Jun 2025, Jonathan Wakely wrote:
> >
> > > As suggested by Jason, this makes all __normal_iterator operators into
> > > friends so they can be found by ADL and don't need to be se
On 7/2/25 00:35, Richard Biener wrote:
> On Tue, Jul 1, 2025 at 11:20 PM David Faust wrote:
>>
>>
>>
>> On 7/1/25 01:02, Richard Biener wrote:
>>> On Mon, Jun 30, 2025 at 9:12 PM David Faust wrote:
On 6/30/25 06:11, Richard Biener wrote:
>> +static void
>> +gen_btf_
On Wed, 2 Jul 2025 at 18:03, Patrick Palka wrote:
>
> On Wed, 2 Jul 2025, Jonathan Wakely wrote:
>
> > On Wed, 2 Jul 2025 at 15:29, Patrick Palka wrote:
> > >
> > >
> > > On Wed, 11 Jun 2025, Jonathan Wakely wrote:
> > >
> > > > As suggested by Jason, this makes all __normal_iterator operators in
> Oh actually the radix members should be of type int, not bool. I can fix
that.
Yes - thank you very much Jonathan for catching that!
It was my honest oversight - I am so used to using auto, that I have
accidentally copied the wrong type.
Also, thank you for adding tests - I should have added the
> Hi Jose,
> Apologies for the late reply, I haven't been feeling well for the past
> few days.
No worries!
>> > This patch adds the vmtest-tool subdirectory under contrib which tests
>> > BPF programs under a live kernel using a QEMU VM. It automatically
>> > builds the specified kernel vers
OK, then I’ll prepare appropriate patch with tests and send it when I’m
done implementing it.
Thanks, Mateusz Zych
On Wed, 2 Jul 2025 at 16:59, Jonathan Wakely wrote:
> On Wed, 2 Jul 2025 at 14:45, Mateusz Zych wrote:
> >
> > > Oh actually the radix members should be of type int, not bool. I c
> -Original Message-
> From: Segher Boessenkool
> Sent: Monday, June 30, 2025 6:23 AM
> To: Cui, Lili
> Cc: ubiz...@gmail.com; gcc-patches@gcc.gnu.org; Liu, Hongtao
> ; richard.guent...@gmail.com; Michael Matz
>
> Subject: Re: [PATCH V3] x86: Enable separate shrink wrapping
>
> Hi!
>
On Wed, 2 Jul 2025 at 15:29, Patrick Palka wrote:
>
>
> On Wed, 11 Jun 2025, Jonathan Wakely wrote:
>
> > As suggested by Jason, this makes all __normal_iterator operators into
> > friends so they can be found by ADL and don't need to be separately
> > exported in module std.
> >
> > The operator<
On Wed, 2 Jul 2025 at 17:15, Mateusz Zych wrote:
>
> OK, then I’ll prepare appropriate patch with tests and send it when I’m done
> implementing it.
That would be great, thanks. I won't push the initial patch, we can
wait for you to prepare the complete fix.
Please note that for a more significa
Hi Jose,
Apologies for the late reply, I haven't been feeling well for the past
few days.
> This patch adds the vmtest-tool subdirectory under contrib which tests
> BPF programs under a live kernel using a QEMU VM. It automatically
> builds the specified kernel version with eBPF support enabled
> On 02/07/25 07:26, Kugan Vivekanandarajah wrote:
> >
> >
> > >
> > > Given the latest few patches that you have committed, is this patch
> > > necessary
> > > anymore? I have not fully understood the new logic as I was on holiday
> > > last
> > > week, but it looks like the propagation is oc
On Wed, Jul 2, 2025 at 2:43 PM H.J. Lu wrote:
>
> *tls_global_dynamic_64_ uses RDI as the __tls_get_addr argument.
> Add RDI clobber to tls_global_dynamic_64 patterns to show it.
>
> PR target/120908
> * config/i386/i386.cc (legitimize_tls_address): Pass RDI to
> gen_tls_global_dynamic_64.
> * con
On Wed, Jul 02, 2025 at 04:36:38AM -0700, Damian Rouson wrote:
> git branch
> gir checkout
> git add
> git commit
> git rebase
> git push
>
> It’s time to move beyond emailing patches! (Please.)
I don't use git other than 'git clone', 'git reset --hard',
and 'git diff'. If gfortran development
On Tue, 1 Jul 2025 at 23:36, Nathan Myers wrote:
>
> Add a bitset constructor from string_view, with other arguments
> matching the constructor from string. Test in ways that exercise
> code paths not checked in existing tests for other constructors.
> Fix existing tests that would fail to detect
On 7/2/25 2:16 AM, Robin Dapp wrote:
CI-testing was failed:
https://github.com/ewlu/gcc-precommit-ci/
issues/3585#issuecomment-3022157670
for sat_u_add-5-u32.c and vect-reduc-sad-1.c. These failures are
compile issues
appeared due to afdo-crossmodule-1b.c file. For some reason, in both
case
On 01/07/2025 11:02, Richard Sandiford wrote:
Karl Meakin writes:
@@ -763,6 +784,68 @@ (define_expand "cbranchcc4"
""
)
+;; Emit a `CB (register)` or `CB (immediate)` instruction.
+;; The immediate range depends on the comparison code.
+;; Comparisons against immediates outside this
Wilco Dijkstra writes:
> Since all Armv9 cores support shifted LDRH/STRH, use the correct cost of zero
> for these.
>
> Passes regress, OK for commit?
>
> gcc:
> * config/aarch64/tuning_models/generic_armv9_a.h
> (generic_armv9_a_addrcost_table): Use zero cost for himode.
OK if th
I'm not sure? I'd prefer some refactoring to make this more obvious
(and the split between the two functions doesn't help ...).
If you're sure it's all covered then ignore this comment, I can do
the refactoring as followup. It just wasn't obvious to me.
Ah, I think I misread your original com
CI-testing was failed:
https://github.com/ewlu/gcc-precommit-ci/issues/3585#issuecomment-3022157670
for sat_u_add-5-u32.c and vect-reduc-sad-1.c. These failures are compile issues
appeared due to afdo-crossmodule-1b.c file. For some reason, in both cases
the following snippets are being inserted i
Ping for this patch.
Thanks,
Alfie
On 20/06/2025 14:23, Alfie Richards wrote:
Thanks for the pointer Joseph.
This update adds tests to gcc/testsuite/g++.dg/warn/Wformat-gcc_diag-1.C
as this seems to be where similar tests are done (eg, %D for tree).
I couldn't find any tests for the actual ou
On Tue, Jul 01, 2025 at 02:50:40PM -0500, Segher Boessenkool wrote:
> No tests become good tests without effort. And tests that are not good
> tests require constant maintenance!
Here are two patches, either just the first one or both can be used
and both were tested on powerpc64le-linux.
The fi
On Wed, 2 Jul 2025 at 10:33, Tomasz Kaminski wrote:
>
>
>
> On Wed, Jul 2, 2025 at 11:27 AM Jonathan Wakely wrote:
>>
>> On 01/07/25 16:54 +0200, Tomasz Kamiński wrote:
>> >This patch lifts locale initialization from locale-specific handling methods
>> >into _M_format_to function, and pass the lo
On Tue, 1 Jul 2025 at 23:34, Nathan Myers wrote:
>
> Make range view constructors explicit, per P2711. Technically, this
> is a breaking change, but it is unlikely to break any production
> code, as reliance on non-explicit construction is unidiomatic..
>
> libstdc++-v3/ChangeLog
> PR libs
On Tue, Jul 01, 2025 at 06:33:12PM +0200, Jakub Jelinek wrote:
> On Tue, Jul 01, 2025 at 03:47:53PM +0200, Stefan Schulze Frielinghaus wrote:
> > In the past years I have started to use more and more function body
> > checks whenever gcc emits optimal code for a function. With that I
> > wanted to
On 02/07/25 03:36 +0300, Mateusz Zych wrote:
Hello libstdc++ Team!
I have recently found a bug in libstdc++, that is,
the std::numeric_limits<> template specializations for integer-class types
are missing some of static data members,
which results in compilation errors of valid C++ code:
- Co
On Wed, Jul 02, 2025 at 11:43:13AM +0200, Stefan Schulze Frielinghaus wrote:
> On Tue, Jul 01, 2025 at 06:33:12PM +0200, Jakub Jelinek wrote:
> > On Tue, Jul 01, 2025 at 03:47:53PM +0200, Stefan Schulze Frielinghaus wrote:
> > > In the past years I have started to use more and more function body
>
To do that, you may need to extract a debug/throws checks from this
functions, i.e. having something like:
if (_M_check_ok(__t, __conv))
{};
else (__use_locale_fmt && ...)
else switch()
Where _M_check_ok() would have a switch that checks if value is ok() based
on specifier, and throw, print mes
On Wed, 2 Jul 2025 at 10:50, Jonathan Wakely wrote:
>
> On 02/07/25 03:36 +0300, Mateusz Zych wrote:
> >Hello libstdc++ Team!
> >
> >I have recently found a bug in libstdc++, that is,
> >the std::numeric_limits<> template specializations for integer-class types
> >are missing some of static data m
On 02/07/25 11:52 +0100, Jonathan Wakely wrote:
On Wed, 2 Jul 2025 at 10:50, Jonathan Wakely wrote:
On 02/07/25 03:36 +0300, Mateusz Zych wrote:
>Hello libstdc++ Team!
>
>I have recently found a bug in libstdc++, that is,
>the std::numeric_limits<> template specializations for integer-class ty
git branch
gir checkout
git add
git commit
git rebase
git push
It’s time to move beyond emailing patches! (Please.)
Damian
On Wed, Jul 2, 2025 at 03:17 Andre Vehreschild wrote:
> Hi all,
>
> I successfully created a big mess with the previous patch. First of all by
> applying an outdated one
On 01/07/25 16:54 +0200, Tomasz Kamiński wrote:
This patch lifts locale initialization from locale-specific handling methods
into _M_format_to function, and pass the locale by const reference.
To avoid unnecessary computation of locale::classic(), we use _Optional_locale,
and emplace into it only
On Wed, Jul 2, 2025 at 9:13 AM XU Kailiang wrote:
>
> C++ formatting locale could have a custom time_put that performs
> differently from the C locale, so do not use __timepunct directly.
>
> libstdc++-v3/ChangeLog:
>
> PR libstdc++/117214
> * include/bits/chrono_io.h (__formatter
On Wed, 2 Jul 2025 at 10:30, Tomasz Kaminski wrote:
>
>
>
> On Wed, Jul 2, 2025 at 9:13 AM XU Kailiang wrote:
>>
>>
>> C++ formatting locale could have a custom time_put that performs
>> differently from the C locale, so do not use __timepunct directly.
>>
>> libstdc++-v3/ChangeLog:
>>
>>
On Wed, Jul 2, 2025 at 11:27 AM Jonathan Wakely wrote:
> On 01/07/25 16:54 +0200, Tomasz Kamiński wrote:
> >This patch lifts locale initialization from locale-specific handling
> methods
> >into _M_format_to function, and pass the locale by const reference.
> >To avoid unnecessary computation of
On Wed, 2 Jul 2025, Robin Dapp wrote:
> > The else (get_group_load_store_type) can end up returning
> > VMAT_GATHER_SCATTER and thus require the above checking as well.
>
> Isn't this already covered by
>
> if (*memory_access_type == VMAT_ELEMENTWISE
> || (*memory_access_type == VMAT_GATHE
On Tue, Jun 10, 2025 at 11:40 PM David Faust wrote:
>
> Add two new c-family attributes, "btf_type_tag" and "btf_decl_tag"
> along with a simple shared handler for them.
>
> gcc/c-family/
> * c-attribs.cc (c_common_attribute_table): Add btf_decl_tag and
> btf_type_tag attributes.
>
On Tue, 1 Jul 2025, Alex Coplan wrote:
> This extends the documentation of the vec_mask_load_lanes optab to
> explicitly state that the mode of the else operand is n, i.e. the mode
> of a single subvector.
>
> OK to install?
OK.
> Thanks,
> Alex
>
> gcc/ChangeLog:
>
> * doc/md.texi (Sta
The else (get_group_load_store_type) can end up returning
VMAT_GATHER_SCATTER and thus require the above checking as well.
Isn't this already covered by
if (*memory_access_type == VMAT_ELEMENTWISE
|| (*memory_access_type == VMAT_GATHER_SCATTER
&& GATHER_SCATTER_LEGACY_P (*gs_inf
> On 1 Jul 2025, at 18:37, Alex Coplan wrote:
>
> The "else operand" to maskload should always be a const_vector, never a
> const_int.
>
> This was just an issue I noticed while looking through the code, I don't
> have a testcase which shows a concrete problem due to this.
>
> Testing of tha
On Tue, Jul 01, 2025 at 09:28:34AM +0200, Robin Dapp wrote:
> > It corrects the shift type of interleaved stepped patterns for const vector
> > expanding in LRA. The shift instruction was initially LSHIFTRT, and it seems
> > still should be the same type for both LRA and other cases.
>
> This is O
Hi Jason,
Thanks for the feedback, see below an updated patch.
Again reg-tested on Aarch64 and x86.
Thanks,
Alfie
-- >8 --
Add logic for the case of two FMV annotated functions with identical
signature other than the return type.
Previously this was ignored, this changes the behavior to emit
When doing --param vect-partial-vector-usage=1 we'd continue querying
the target whether it wants more vector epilogues, but when it comes
back with a suggestion we then might iterate endlessly. Do not
even ask the target when we decided for the last epilogue to be
one with partial vectors.
Boots
Make the formatting of the RTL templates in the rules for branch
instructions more consistent with each other.
gcc/ChangeLog:
* config/aarch64/aarch64.md (cbranch4): Reformat.
(cbranchcc4): Likewise.
(condjump): Likewise.
(*compare_condjump): Likewise.
(aar
Tested x86_64-pc-linux-gnu, applying to trunk.
-- 8< --
In r15-7532 for PR118856 I introduced a TARGET_EXPR with a
TARGET_EXPR_INITIAL of void_node to express that no initialization is done.
And indeed evaluating that doesn't store a value for the TARGET_EXPR_SLOT
variable.
But then at the end o
*tls_local_dynamic_64_ uses RDI as the __tls_get_addr argument.
Add RDI clobber to tls_local_dynamic_64 patterns to show it.
PR target/120908
* config/i386/i386.cc (legitimize_tls_address): Pass RDI to
gen_tls_local_dynamic_64.
* config/i386/i386.md (*tls_local_dynamic_64_): Add RDI
clobber and us
Hi All,
The following patch has been bootstrapped and regtested on powerpc64le-linux.
Changes from V3:
* Replaced named operands with positional operands in inline assembly for
better readability.
* Considered using _ADDR[0] and _ADDR[1] to make memory reads more explicit to
the compiler.
* Cle
Indentation are updated accordingly and no regress found.
gcc/ChangeLog:
*config/riscv/riscv-cores.def(RISCV_CORE): Updated the supported march.
*config/riscv/riscv-ext-mips.def(DEFINE_RISCV_EXT):
New file added for mips conditional mov extension.
*config/riscv/ris
On Tue, Jul 1, 2025 at 5:17 PM Qing Zhao wrote:
>
>
>
> > On Jul 1, 2025, at 03:14, Richard Biener wrote:
> >
> > On Mon, Jun 30, 2025 at 10:37 PM Qing Zhao wrote:
> >>
> >> Hi, David,
> >>
> >> Thank you for the info.
> >>
> >> Yes, this does sound like a general issue in this area.
> >>
> >> I
C++ formatting locale could have a custom time_put that performs
differently from the C locale, so do not use __timepunct directly.
libstdc++-v3/ChangeLog:
PR libstdc++/117214
* include/bits/chrono_io.h (__formatter_chrono::_M_a_A,
__formatter_chrono::_M_b_B, __formatter
The register_operand predicate can match subreg, then we'd have a subreg
of subreg and it's invalid. Use lowpart_subreg to avoid the nested
subreg.
gcc/ChangeLog:
* config/loongarch/loongarch.md (crc_combine): Avoid nested
subreg.
gcc/testsuite/ChangeLog:
* gcc.c-tortu
On Tue, Jul 1, 2025 at 11:20 PM David Faust wrote:
>
>
>
> On 7/1/25 01:02, Richard Biener wrote:
> > On Mon, Jun 30, 2025 at 9:12 PM David Faust wrote:
> >>
> >>
> >>
> >> On 6/30/25 06:11, Richard Biener wrote:
> +static void
> +gen_btf_decl_tag_dies (tree t, dw_die_ref target, dw_die
The vectorizer tracks alignment of datarefs with dr_aligned
and dr_unaligned_supported but that's aligned with respect to
the target alignment which can be less aligned than the mode
used for the access. The following fixes this discrepancy
for vectorizing loads and stores. The issue is visible f
This defines the testsuite assertion macro VERIFY so that it allows
un-parenthesized expressions containing commas. This matches how assert
is defined in C++26, following the approval of P2264R7.
The primary motivation is to allow expressions that the preprocessor
splits into multiple arguments, e
Commit the test file `cmpbr.c` before rules for generating the new
instructions are added, so that the changes in codegen are more obvious
in the next commit.
gcc/testsuite/ChangeLog:
* lib/target-supports.exp: Add `cmpbr` to the list of extensions.
* gcc.target/aarch64/cmpbr.c: N
Extract the hardcoded values for the minimum PC-relative displacements
into named constants and document them.
gcc/ChangeLog:
* config/aarch64/aarch64.md (BRANCH_LEN_P_128MiB): New constant.
(BRANCH_LEN_N_128MiB): Likewise.
(BRANCH_LEN_P_1MiB): Likewise.
(BRANCH_LE
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
Add rules for lowering `cbranch4` to CBB/CBH/CB when
CMPBR extension is enabled.
gcc/ChangeLog:
* config/aarch64/aarch64-protos.h (aarch64_cb_rhs): New function.
* config/aarch64/aarch64.cc (aarch64_cb_rhs): Likewise.
* config/aarch64/aarch64.md (cbranch4): Rename to ...
Karl Meakin writes:
> On 01/07/2025 11:02, Richard Sandiford wrote:
>> Karl Meakin writes:
>>> @@ -763,6 +784,68 @@ (define_expand "cbranchcc4"
>>> ""
>>> )
>>>
>>> +;; Emit a `CB (register)` or `CB (immediate)` instruction.
>>> +;; The immediate range depends on the comparison code.
>>>
This patch series adds support for the CMPBR extension. It includes the
new `+cmpbr` option and rules to generate the new instructions when
lowering conditional branches.
Changelog:
* v8:
- Support far branches for the `CBB` and `CBH` instructions, and add tests
for them.
- Mark the branch in
The `far_branch` attribute only ever takes the values 0 or 1, so make it
a `no/yes` valued string attribute instead.
gcc/ChangeLog:
* config/aarch64/aarch64.md (far_branch): Replace 0/1 with
no/yes.
(aarch64_bcond): Handle rename.
(aarch64_cbz1): Likewise.
Give the `define_insn` rules used in lowering `cbranch4` to RTL
more descriptive and consistent names: from now on, each rule is named
after the AArch64 instruction that it generates. Also add comments to
document each rule.
gcc/ChangeLog:
* config/aarch64/aarch64.md (condjump): Rename to
Add the `+cmpbr` option to enable the FEAT_CMPBR architectural
extension.
gcc/ChangeLog:
* config/aarch64/aarch64-option-extensions.def (cmpbr): New
option.
* config/aarch64/aarch64.h (TARGET_CMPBR): New macro.
* doc/invoke.texi (cmpbr): New option.
---
gcc/config
Changes in V2:
* Generalize private member _M_check_initial_position for use with
both string and string_view arguments.
* Remove unnecessary #if guards for version and hostedness.
* Remove redundant "std::" qualifications in new code.
* Improve Doxygen source readability.
* Clarify commit messag
Bootstrapped and regtested on x86_64-pc-linux-gnu, does this look OK
for trunk?
-- >8 --
Here the flag -fno-delete-null-pointer-checks causes the trivial address
comparison in
inline int a, b;
static_assert(&a != &b);
to be rejected as non-constant because with the flag we can't assume
such
95 matches
Mail list logo
