Pushed to trunk.
libstdc++-v3/ChangeLog:
* acinclude.m4 (GLIBCXX_ENABLE_FULLY_DYNAMIC_STRING): Improve
comments.
* configure: Regenerate.
---
libstdc++-v3/acinclude.m4 | 17 +++--
libstdc++-v3/configure| 4 ++--
2 files changed, 13 insertions(+), 8
Tested x86_64-linux, pushed to trunk.
Instead of hardcoded preprocessor conditionals with explicit target
checks, just rely on the fact that __BYTE_ORDER__ is always defined by
GCC.
libstdc++-v3/ChangeLog:
PR libstdc++/104080
* src/c++17/fast_float/LOCAL_PATCHES: Update.
> Unfortunately it's not OK, these are the most annoying/delicate dependencies,
> so
> we'd rather not reintroduce them. I'll instead update prerequisites to
> document that GCC 5.1 or later is required to build GNAT.
Now pushed on master:
Update prerequisites for GNAT
*
Hi!
On 2022-01-05T12:06:12+0100, Martin Liška wrote:
> The patch is about small refactoring of the plugin.
Yay!
> include/ChangeLog:
>
> * plugin-api.h (enum ld_plugin_tag): Do not set implicit enum
> values.
I did wonder whether the original idea there would've been to fixate
On 12/16/21 12:46, Martin Liška wrote:
On 11/8/21 13:26, Martin Liška wrote:
On 11/5/21 17:08, Martin Liška wrote:
On 8/23/21 10:54, Martin Liška wrote:
On 8/16/21 13:13, Martin Liška wrote:
I'm going to apply the following 3 tested patches.
Martin
One more patch I've just tested.
Martin
Sorry, I had not seen that this was entirely within my amdgcn remit
On 12/01/2022 09:43, Marcel Vollweiler wrote:
Hi,
Currently omp_get_device_num does not work on gcn targets with more than
one offload device. The reason is that GOMP_DEVICE_NUM_VAR is static in
icv-device.c and thus
Pushed to trunk.
libstdc++-v3/ChangeLog:
* doc/xml/manual/status_cxx2017.xml: Update C++17 status.
* doc/xml/manual/status_cxx2020.xml: Use 12.1 instead of 12 for
upcoming release.
* doc/html/manual/status.html: Regenerate.
---
From: Andrew Pinski
The Linux kernel started to fail compile when the jump threader was improved
(r12-2591-g2e96b5f14e4025691). This failure was due to the IPA splitting code
decided now to split off the basic block which contained two functions,
one of those functions included the error
Thomas,
> OK to push (after more testing) the attached
> 'Revert parts of "[Ada] Reduce runtime dependencies on stage1"', for the
> reason detailed therein? Should the comment before 'GNAT1_C_OBJS' be
> re-instated/adjusted, too? Would appreciate a careful review, as I don't
> really know what
As pointed out by pahole.
Bootstrapped and tested on x86_64-unknown-linux-gnu, pushed.
2022-01-18 Richard Biener
* cgraph.h (struct cgraph_simd_clone_arg): Re-arrange fields to
avoid padding.
* function.h (struct function): Likewise.
---
gcc/cgraph.h | 6 +++---
Hi!
Maybe I'm just totally confused -- as so often ;-) -- but things seem
strange here:
On 2022-01-12T10:43:05+0100, Marcel Vollweiler wrote:
> Currently omp_get_device_num does not work on gcn targets with more than
> one offload device. The reason is that GOMP_DEVICE_NUM_VAR
I understand the
On Tue, Jan 18, 2022 at 01:16:14PM +0100, Richard Biener wrote:
> As pointed out by pahole.
>
> Bootstrapped & tested on x86_64-unknown-linux-gnu.
>
> OK?
>
> Thanks,
> Richard.
>
> 2022-01-18 Richard Biener
>
> libcpp/
> * include/line-map.h (class line_maps): Re-arrange fields
>
On 18/01/2022 12:25, Thomas Schwinge wrote:
Hi!
Maybe I'm just totally confused -- as so often ;-) -- but things seem
strange here:
On 2022-01-12T10:43:05+0100, Marcel Vollweiler wrote:
Currently omp_get_device_num does not work on gcn targets with more than
one offload device. The reason is
On Mon, 17 Jan 2022, Jiufu Guo wrote:
> Richard Biener writes:
>
> > On Fri, 14 Jan 2022, Jiufu Guo wrote:
> >
> >> Richard Biener writes:
> >>
> >> > On Thu, 13 Jan 2022, guojiufu wrote:
> >> >
> >> >> On 2022-01-03 22:30, Richard Biener wrote:
> >> >> > On Wed, 22 Dec 2021, Jiufu Guo wrote:
On Tue, 18 Jan 2022, Richard Biener wrote:
> On Mon, 17 Jan 2022, Jiufu Guo wrote:
>
> > Richard Biener writes:
> >
> > > On Fri, 14 Jan 2022, Jiufu Guo wrote:
> > >
> > >> Richard Biener writes:
> > >>
> > >> > On Thu, 13 Jan 2022, guojiufu wrote:
> > >> >
> > >> >> On 2022-01-03 22:30,
In g:526e1639aa76b0a8496b0dc3a3ff2c450229544e I'd added support
for finding more consecutive MEMs. However, the check was too
eager, in that it matched MEM_REFs with the same base address
even if that base address was an arbitrary SSA name. This can
give wrong results if a MEM_REF from one loop
On Tue, Jan 18, 2022 at 11:17:41AM +0100, Jakub Jelinek via Gcc-patches wrote:
> Anyway, the following has been successfully bootstrapped/regtested on
> x86_64-linux and i686-linux, ok for trunk?
Actually, I missed one regression (thought it is caused by PR104025 patch
but it is this one):
+FAIL:
For some CPUs, the assembler machine directive cannot be determined by ISA
flags.
gcc/
PR 104090/target
* config/rs6000/rs6000.cc (rs6000_machine_from_flags): Use also
rtems_cpu.
---
gcc/config/rs6000/rs6000.cc | 28
1 file changed, 28
As pointed out by pahole.
Bootstrapped & tested on x86_64-unknown-linux-gnu.
OK?
Thanks,
Richard.
2022-01-18 Richard Biener
libcpp/
* include/line-map.h (class line_maps): Re-arrange fields
to minimize padding.
(class rich_location): Likewise.
* line-map.cc
The following change avoids doing IPA inlining of small functions
into functions compiled with -Og - those functions will see almost no
followup scalar cleanups so that the benefit anticipated by the
inliner will not be realized and instead the late diagnostic code
will be confused by dead code
On Tue, Jan 18, 2022 at 1:22 PM Richard Sandiford via Gcc-patches
wrote:
>
> In g:526e1639aa76b0a8496b0dc3a3ff2c450229544e I'd added support
> for finding more consecutive MEMs. However, the check was too
> eager, in that it matched MEM_REFs with the same base address
> even if that base address
In this PR the waccess pass was fed:
D.10779 ={v} {CLOBBER};
VIEW_CONVERT_EXPR(D.10779) = .MASK_LOAD_LANES (addr_5(D), 64B,
_2);
_7 = D.10779.__val[0];
However, the tracking of m_clobbers only looked at gassigns,
so it missed that the clobber on the first line was overwritten
by the call
> With -Og we are not prepared to do cleanup after IPA optimizations
> and dead code exposed by those confuses late diagnostic passes.
> This is a first patch removing unwanted IPA optimizations, namely
> both late modref and pure-const analysis.
>
> Bootstrap and regtest running on
On 1/18/22 09:36, Eric Botcazou wrote:
Martin,
The renaming patches have been just installed and I've built a few target
compilers so far. I'll be online in ~10 hours from now so I can address
potential issues caused by the patch.
Who approved it though?
Release managers.
The renaming of
On Mon, Jan 17, 2022 at 7:20 PM Sandra Loosemore
wrote:
>
> On 1/17/22 1:29 AM, Richard Biener wrote:
> > On Mon, Jan 17, 2022 at 4:15 AM Sandra Loosemore
> > wrote:
> >>
> >> On 1/8/22 9:24 PM, Sandra Loosemore wrote:
> >>> This patch fixes the ICE I reported in PR103163. We were initializing
From: LiaoShihua
When I built riscv-gcc with newlib, it will be terminated with message "fatal
error: endian.h: No such file or directory".
So, fixed it in fast_float.h.
libstdc++-v3\ChangeLog:
* src/c++17/fast_float/fast_float.h (defined):fast_float.h will include
with using newlib
Tested x86_64-linux, pushed to trunk.
Reported upstream as https://github.com/fastfloat/fast_float/pull/122
libstdc++-v3/ChangeLog:
PR libstdc++/104080
* src/c++17/fast_float/LOCAL_PATCHES: UPDATE.
* src/c++17/fast_float/fast_float.h (round): Use explicit
On Tue, 18 Jan 2022 at 09:05, LiaoShihua wrote:
>
> When I built riscv-gcc with newlib, it will be terminated with message
> "fatal error: endian.h: No such file or directory".
> So, fixed it in fast_float.h.
>
> libstdc++-v3\ChangeLog:
>
N.B. you need to use a forward slash here, not
On 1/12/22 10:03, Martin Liška wrote:
Ready to be installed?
I'm going to install the patch as it's pre-approved by Jeff.
Martin
Hi!
The following testcase is miscompiled. We see the constructor is immediate,
in build_over_call we trigger:
if (obj_arg && is_dummy_object (obj_arg))
{
call = build_cplus_new (DECL_CONTEXT (fndecl), call, complain);
obj_arg = NULL_TREE;
On Thu, Jan 13, 2022 at 04:18:33PM -0500, Jason Merrill wrote:
> > Note, address_compare has some special cases, e.g. it assumes that
> > static vars are never adjacent to automatic vars, which is the case
> > for the usual layout where automatic vars are on the stack and after
> > .rodata/.data
With -Og we are not prepared to do cleanup after IPA optimizations
and dead code exposed by those confuses late diagnostic passes.
This is a first patch removing unwanted IPA optimizations, namely
both late modref and pure-const analysis.
Bootstrap and regtest running on x86_64-unknown-linux-gnu,
On 17/01/2022 21:41, Martin Liška wrote:
On 1/13/22 12:01, Martin Liška wrote:
Hello.
Based on the discussion with release managers, the change is going to
happen
after stage4 begins.
Martin
Hi.
The renaming patches have been just installed and I've built a few
target compilers so
On 1/18/22 14:10, Richard Earnshaw wrote:
Is that worth adding to contrib/gcc-git-customization.sh ?
Yes, can you please do that?
Thanks,
Martin
"Kewen.Lin" writes:
> on 2022/1/18 锟斤拷锟斤拷11:06, Kewen.Lin via Gcc-patches wrote:
>> Hi,
>>
>> As discussed in PR104015, the test case slp-perm-9.c can be
>> fragile when vectorizer tries to use different vectorisation
>> strategies.
>>
>> As Richard suggested, this patch tries to make the check
This adds a missing guard for a pointer to the DSE modref query,
otherwise we can end up invoking APIs desired to only work on
pointers on non-pointers when there are mismatches between declared
and actual arguments of functions in the program.
Bootstrapped and tested on x86_64-unknown-linux-gnu,
> The following change avoids doing IPA inlining of small functions
> into functions compiled with -Og - those functions will see almost no
> followup scalar cleanups so that the benefit anticipated by the
> inliner will not be realized and instead the late diagnostic code
> will be confused by
On Mon, Jan 17, 2022 at 8:56 PM Andrew MacLeod via Gcc-patches
wrote:
>
> As mentioned in the PR, this demonstrates the potentially quadratic
> performance behaviour of adding transitive relations over a series of
> cascading calculations.
>
> As the lookup in a basic block is also linear in
LGTM.
I've just installed that, the Ada testsuite should work now.
Martin
On Tue, Jan 18, 2022 at 7:26 AM liuhongt wrote:
>
> For testcase in PR, the patch supports QI:4 -> HI:16 pack with
> multi steps(first pack QI:4 -> QI:8 through vec_pack_sbool_trunc_qi,
> then pack QI:8 -> HI:16 through vec_pack_trunc_hi).
> Similar for QI:2 -> HI:16 which is test4 in
Since we order DRs after DR_INIT we know the difference will be
positive and thus can avoid signed overflow issues by using
unsigned arithmetic to produce the known unsigned result.
Bootstrapped and tested on x86_64-unknown-linux-gnu, pushed.
2022-01-18 Richard Biener
PR
on 2022/1/18 上午11:06, Kewen.Lin via Gcc-patches wrote:
> Hi,
>
> As discussed in PR104015, the test case slp-perm-9.c can be
> fragile when vectorizer tries to use different vectorisation
> strategies.
>
> As Richard suggested, this patch tries to make the check not
> sensitive on the re-trying
On 1/18/22 08:59, Thomas Schwinge wrote:
Hi!
On 2022-01-10T17:14:00+0100, Martin Liška wrote:
The script is capable of checking if MAINTAINER names are sorted
alphabetically.
Irrespective of the pre-existing issue that a concept of a first and a
last name doesn't exist in all cultures, and
On Mon, Jan 17, 2022 at 09:08:22PM -0800, Andrew Pinski wrote:
> > Perhaps even
> > /* Check builtins that prevent splitting. */
> > if (gimple_code (stmt) == GIMPLE_CALL)
> > if (tree decl = gimple_call_fndecl (stmt))
> > {
> > if (fndecl_built_in_p
On 1/18/22 09:53, Richard Biener wrote:
Technically all release managers are also global reviewers, but I
agree the speciality
of the Ada setup (esp. the runtime being in gcc/) could have been anticipated.
Yeah, I thought that building Ada FE is enough for testing effort, sorry.
I would like
The patch adds detection of non-standard 'unicode' python module.
Martin
gcc/testsuite/ChangeLog:
* gcc.src/maintainers.exp: Use
check_effective_target_python3_module for checking of unicode
module.
* lib/target-supports.exp: Add
Martin,
> The renaming patches have been just installed and I've built a few target
> compilers so far. I'll be online in ~10 hours from now so I can address
> potential issues caused by the patch.
Who approved it though? The renaming of the C files in the ada/ directory is
wrong and should be
On Tue, Jan 18, 2022 at 10:08:20AM +0100, Martin Liška wrote:
> On 1/18/22 09:53, Richard Biener wrote:
> > Technically all release managers are also global reviewers, but I
> > agree the speciality
> > of the Ada setup (esp. the runtime being in gcc/) could have been
> > anticipated.
>
> Yeah,
Pushed as obvious.
Martin
gcc/testsuite/ChangeLog:
* gcc.misc-tests/options.exp: Include renamed file.
---
gcc/testsuite/gcc.misc-tests/options.exp | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/gcc/testsuite/gcc.misc-tests/options.exp
> Release managers.
They certainly have authority on the timing, but not on the contents.
> Working on that right now, sorry..
OK, thanks in advance.
--
Eric Botcazou
RISC-V glibc intend to removed its own `sigcontext.h`[1] and use the linux
kernel's one, however the struct layout is slightly different between those two
version, fortunately they have identical layout, so we just need a
magical way to detect which one we are used.
libgcc/ChangeLog:
*
On Mon, 17 Jan 2022, Martin Sebor wrote:
> On 1/17/22 07:32, Richard Biener via Gcc-patches wrote:
> > The warning control falls into the C++ trap of using a reference
> > to old hashtable contents for a put operation which can end up
> > re-allocating that before reading from the old freed
Hi!
On 2022-01-10T09:41:08+, Pierre-Marie de Rodat via Gcc-cvs
wrote:
> https://gcc.gnu.org/g:a6eae6a9bbf83ada3aab7f5a2a74e7024889d237
>
> commit r12-6398-ga6eae6a9bbf83ada3aab7f5a2a74e7024889d237
> Author: Arnaud Charlet
> Date: Mon Nov 29 10:15:26 2021 +
>
> [Ada] Reduce
On Sat, Jan 15, 2022 at 09:24:05AM -0500, Patrick Palka wrote:
> On Fri, 14 Jan 2022, Marek Polacek via Gcc-patches wrote:
>
> > This is a "canonical types differ for identical types" ICE, which started
> > with r11-4682. It's a bit tricky to explain. Consider:
> >
> > template struct S {
>
Pushed as pre-approved by Jeff. The patch fixes -Wformat-diag warnings.
Martin
gcc/ChangeLog:
* config/nvptx/nvptx.cc (nvptx_goacc_validate_dims_1): Wrap
keyword.
* config/nvptx/nvptx.md: Remove trailing dot.
libgomp/ChangeLog:
*
On 1/18/22 01:36, Richard Biener wrote:
On Mon, 17 Jan 2022, Martin Sebor wrote:
On 1/17/22 07:32, Richard Biener via Gcc-patches wrote:
The warning control falls into the C++ trap of using a reference
to old hashtable contents for a put operation which can end up
re-allocating that before
Thanks Martin!
On Wed, Jan 19, 2022 at 12:23 AM Martin Liška wrote:
>
> Pushed as pre-approved by Jeff. The patch fixes -Wformat-diag warnings.
>
> Martin
>
> gcc/ChangeLog:
>
> * common/config/riscv/riscv-common.cc (riscv_subset_list::add):
> Wrap keywords with quotes and remove
Tested powerpc64le-linux, pushed to trunk.
libstdc++-v3/ChangeLog:
* include/bits/stl_function.h (ptr_fun): Fix suggestion for
non-deprecated alternative.
---
libstdc++-v3/include/bits/stl_function.h | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git
On Tue, Jan 18, 2022 at 05:21:30PM +0100, Martin Liška wrote:
> I'm going to install the following -Wformat-diag that is based on Segher's
> wording.
Thanks!
(But please fix the format=flawed:
> Content-Type: text/plain; charset=UTF-8; format=flowed
)
Segher
On Tue, 18 Jan 2022, Jan Hubicka wrote:
> > The following change avoids doing IPA inlining of small functions
> > into functions compiled with -Og - those functions will see almost no
> > followup scalar cleanups so that the benefit anticipated by the
> > inliner will not be realized and instead
On Mon, Jan 17, 2022 at 01:48:48PM -0500, Jason Merrill wrote:
> On 1/14/22 19:22, Marek Polacek wrote:
> > This is a "canonical types differ for identical types" ICE, which started
> > with r11-4682. It's a bit tricky to explain. Consider:
> >
> >template struct S {
> > S bar()
On 1/16/22 01:31, Martin Sebor wrote:
Hello.
(I think your message in unrelated to this email thread).
Martin, I've looked into removing the -Wno-error for this warning
for just a subset of targets. It seems doable with some hardcoding
in configure.ac but if you're planning to do the cleanup
Hi Martin!
On 2022-01-18T17:22:53+0100, Martin Liška wrote:
> Pushed as pre-approved by Jeff. The patch fixes -Wformat-diag warnings.
> --- a/gcc/config/nvptx/nvptx.cc
> +++ b/gcc/config/nvptx/nvptx.cc
> @@ -5894,7 +5894,7 @@ nvptx_goacc_validate_dims_1 (tree decl, int dims[], int
> fn_level,
From: Tom Honermann
Tested powerpc64le-linux, pushed to trunk.
This patch completes implementation of the C++20 proposal P0482R6 [1] by
adding declarations of std::c8rtomb() and std::mbrtoc8() in if
provided by the C library in .
This patch addresses feedback provided in response to a
Hi,
This patch considers the environment variable syntax extension for
device-specific variants of environment variables from OpenMP 5.1 (see
OpenMP 5.1 specification, p. 75 and p. 639). An environment variable
(e.g. OMP_NUM_TEAMS) can have different suffixes:
_DEV (e.g. OMP_NUM_TEAMS_DEV):
Hello.
The patch fixes -Wformat-diag in various targets.
I'm going to install the patch based on Jeff's pre-approval of similar patches.
Martin
gcc/ChangeLog:
* collect2.cc (scan_libraries): Fix -Wformat-diag issues.
* config/aarch64/aarch64-builtins.cc
I'm going to install the following -Wformat-diag that is based on Segher's
wording.
Pushed to master.
Martin
gcc/ChangeLog:
* config/rs6000/rs6000-call.cc (rs6000_invalid_builtin): Change
wording of an error message.
gcc/testsuite/ChangeLog:
*
Tested powerpc64le-linux, pushed to trunk.
This adds the C++23 header, as proposed by P0943R6, for
compatibility with C code.
There are still some ABI differences between atomic_xxx in C and C++
std::atomic_xxx in C++, so this only provides source compatibility, not
binary compatibility.
LGTM, thanks :)
On Wed, Jan 19, 2022 at 12:59 AM Maciej W. Rozycki wrote:
>
> Avoid undefined arithmetic involving a pointer to a heap allocation that
> has been freed and move a problematic calculation ahead of the following
> call to `free' in `riscv_subset_list::parse_multiletter_ext',
On Mon, Jan 03, 2022 at 10:15:26PM +0800, Chung-Lin Tang wrote:
> This issue was triggered after the patch extending syntax for component access
> in map clauses
> (https://gcc.gnu.org/git/gitweb.cgi?p=gcc.git;h=0ab29cf0bb68960c)
>
> In gimplify_scan_omp_clauses, the case for handling indirect
On Tue, Jan 18, 2022 at 09:08:03AM -0500, Jason Merrill wrote:
> > which makes call a TARGET_EXPR with the dtor in TARGET_EXPR_CLEANUP,
> > but then call cxx_constant_value on it. In cxx_eval_outermost_constant_expr
> > it triggers the:
> >else if (TREE_CODE (t) != CONSTRUCTOR)
> >
Also, I put the extern in gcc/system.h, but I'm not sure it's the right
place.
Where should I put it?
Le lundi 17 janvier 2022 à 19:46 -0500, Antoni Boucher via Jit a
écrit :
> I missed the comment about the new define, so here's the updated
> patch.
>
> Le lundi 17 janvier 2022 à 19:24 -0500,
Pushed as pre-approved by Jeff. The patch fixes -Wformat-diag warnings.
Martin
gcc/ChangeLog:
* config/v850/v850-c.cc (pop_data_area): Fix -Wformat-diag
warning.
(ghs_pragma_section): Likewise.
(ghs_pragma_interrupt): Likewise.
(ghs_pragma_starttda):
Pushed as pre-approved by Jeff. The patch fixes -Wformat-diag warnings.
Martin
gcc/ChangeLog:
* config/nds32/nds32-intrinsic.cc (nds32_expand_builtin_impl):
Fix warnings.
* config/nds32/nds32-intrinsic.md: Likewise.
* config/nds32/nds32-isr.cc
Tested powerpc64le-linux, pushed to trunk.
We should not assume that std::iter_value_t etc. are defined
unconditionally for C++20 mode.
libstdc++-v3/ChangeLog:
PR libstdc++/104098
* include/bits/stl_iterator.h (reverse_iterator): Check
__cpp_lib_concepts instead of
Pushed as pre-approved by Jeff. The patch fixes -Wformat-diag warnings.
Martin
gcc/ChangeLog:
* config/vms/vms-c.cc (vms_pragma_nostandard): Fix -Wformat-diag
warning.
(vms_pragma_standard): Likewise.
(vms_pragma_extern_prefix): Likewise.
---
Pushed as pre-approved by Jeff. The patch fixes -Wformat-diag warnings.
Martin
gcc/ChangeLog:
* config/xtensa/xtensa.cc (print_operand): Fix warnings.
(print_operand_address): Likewise.
(xtensa_multibss_section_type_flags): Likewise.
---
gcc/config/xtensa/xtensa.cc | 6
Tested powerpc64le-linux, pushed to trunk.
The deprecated non-standard std::pair constructors that allow
constructing std::pair from an rvalue and
a literal zero where not sufficiently constrained. They were viable when
constructing std::pair, and that case
should work fine using the standard
On 1/18/22 17:36, Thomas Schwinge wrote:
Similar here: '%'.
Grüße
Thomas
I see, thank you for the review.
Can you please test for me the following patch candidate?
Thanks,
MartinFrom de5cafe27131c002e6634dd7417577452eb1aac8 Mon Sep 17 00:00:00 2001
From: Martin Liska
Date: Tue, 18 Jan
On 1/18/22 09:26, Jakub Jelinek wrote:
On Tue, Jan 18, 2022 at 09:08:03AM -0500, Jason Merrill wrote:
which makes call a TARGET_EXPR with the dtor in TARGET_EXPR_CLEANUP,
but then call cxx_constant_value on it. In cxx_eval_outermost_constant_expr
it triggers the:
else if (TREE_CODE (t)
On 1/18/22 05:10, Jakub Jelinek wrote:
Hi!
The following testcase is miscompiled. We see the constructor is immediate,
in build_over_call we trigger:
if (obj_arg && is_dummy_object (obj_arg))
{
call = build_cplus_new (DECL_CONTEXT (fndecl), call,
Hi Thomas,
Am 18.01.2022 um 13:25 schrieb Thomas Schwinge:
Hi!
Maybe I'm just totally confused -- as so often ;-) -- but things seem
strange here:
On 2022-01-12T10:43:05+0100, Marcel Vollweiler wrote:
Currently omp_get_device_num does not work on gcn targets with more than
one offload
On 1/18/22 05:17, Jakub Jelinek wrote:
On Thu, Jan 13, 2022 at 04:18:33PM -0500, Jason Merrill wrote:
Note, address_compare has some special cases, e.g. it assumes that
static vars are never adjacent to automatic vars, which is the case
for the usual layout where automatic vars are on the stack
On Tue, Jan 18, 2022 at 11:25:38AM -0500, Jason Merrill wrote:
> > Can you please comment on
> > https://gcc.gnu.org/bugzilla/show_bug.cgi?id=86369#c1
> > then?
>
> Done.
Thanks.
> About the rest of the patch, I thought I had seen richi comment on IRC (but
> can't find the comment now) that
Avoid undefined arithmetic involving a pointer to a heap allocation that
has been freed and move a problematic calculation ahead of the following
call to `free' in `riscv_subset_list::parse_multiletter_ext', removing a
compilation error:
.../gcc/common/config/riscv/riscv-common.cc: In member
We're incorrectly rejecting the below testcase during template argument
coercion because invalid_nontype_parm_type_p returns true for
DEPENDENT_OPERATOR_TYPE.
This patch fixes this by partially rewriting invalid_nontype_parm_type_p
in terms of WILDCARD_TYPE_P, of which DEPENDENT_OPERATOR_TYPE is
Hi,
This patch considers the environment variable syntax extension for
device-specific variants of environment variables from OpenMP 5.1 (see
OpenMP 5.1 specification, p. 75 and p. 639). An environment variable
(e.g. OMP_NUM_TEAMS) can have different suffixes:
_DEV (e.g. OMP_NUM_TEAMS_DEV):
Pushed as pre-approved by Jeff. The patch fixes -Wformat-diag warnings.
Martin
gcc/ChangeLog:
* common/config/riscv/riscv-common.cc (riscv_subset_list::add):
Wrap keywords with quotes and remove trailing dots.
(riscv_subset_list::parsing_subset_version): Likewise.
On Tue, 18 Jan 2022 08:31:12 PST (-0800), gcc-patches@gcc.gnu.org wrote:
Thanks Martin!
Yep. Seeing this go by, though, I think there's some English issues
with the original messages. I'd write it more like this, but I'm never
100% sure on these things:
diff --git
On 1/18/22 11:40, Jakub Jelinek wrote:
On Tue, Jan 18, 2022 at 11:25:38AM -0500, Jason Merrill wrote:
Can you please comment on https://gcc.gnu.org/bugzilla/show_bug.cgi?id=86369#c1
then?
Done.
Thanks.
About the rest of the patch, I thought I had seen richi comment on IRC (but
can't find
Aligne __EH_FRAME_BEGIN__ to pointer size since gcc/unwind-dw2-fde.h has
/* The first few fields of a CIE. The CIE_id field is 0 for a CIE,
to distinguish it from a valid FDE. FDEs are aligned to an addressing
unit boundary, but the fields within are unaligned. */
struct dwarf_cie
{
The code generated by -mcmodel=medany is defined to be
position-independent, but is not guaranteed to function correctly when
linked into position-independent executables or libraries. See the
recent discussion at the psABI specification [1] for more details.
It would be better to reject these
On Wed, 19 Jan 2022, Kito Cheng wrote:
> LGTM, thanks :)
>
> > Avoid undefined arithmetic involving a pointer to a heap allocation that
> > has been freed and move a problematic calculation ahead of the following
> > call to `free' in `riscv_subset_list::parse_multiletter_ext', removing a
> >
Tested powerpc64le-linux, pushed to trunk.
The new deleted constructors added by P2166R1 are a breaking change,
making previously valid code ill-formed in C++23. As a result, they
should only be defined for C++23 and not for C++11 and up.
libstdc++-v3/ChangeLog:
PR libstdc++/104099
BPF CO-RE relocations contain offsets to strings buffered in the BTF
string table. These BTF-specific strings are stored in memory in the
CTF auxilliary strtab, which at output time is concatenated onto the end
of the standard strtab.
Previously, these string offsets were computed at the time the
On 1/18/22 11:08, Patrick Palka wrote:
We're incorrectly rejecting the below testcase during template argument
coercion because invalid_nontype_parm_type_p returns true for
DEPENDENT_OPERATOR_TYPE.
This patch fixes this by partially rewriting invalid_nontype_parm_type_p
in terms of
Successfully bootstrapped & regrtested on x86_64-pc-linux-gnu.
Pushed to trunk as r12-6700-g2aefe248aa4160205c44808166393a42031d2dea.
gcc/analyzer/ChangeLog:
PR analyzer/104089
* region-model-manager.cc
(region_model_manager::get_or_create_constant_svalue): Assert that
For this testcase, the cleanup that is supposed to happen if initialization
throws was wrongly being run on the normal control path as well. This turns
out to be because the EH-only handling in gimple_push_cleanup didn't apply
to conditional cleanups such as we have for nothrow new, since we
This patch introduced new AIX testsuite failures.
PR libstdc++/104101
Thanks, David
On 1/18/22 06:37, Richard Sandiford wrote:
In this PR the waccess pass was fed:
D.10779 ={v} {CLOBBER};
VIEW_CONVERT_EXPR(D.10779) = .MASK_LOAD_LANES (addr_5(D),
64B, _2);
_7 = D.10779.__val[0];
However, the tracking of m_clobbers only looked at gassigns,
so it missed that the
1 - 100 of 165 matches
Mail list logo
