Hi,
In gimple_expand_vec_cond_expr:
icode = get_vcond_icode (mode, cmp_op_mode, unsignedp);
if (icode == CODE_FOR_nothing)
{
if (tcode == LT_EXPR
&& op0a == op0)
{
/* A VEC_COND_EXPR condition could be folded from EQ_EXPR/NE_EXPR
into a consta
Hi!
The following testcase ICEs on x86_64-linux with -m32 due to a bug in
apply_identity_attributes. The function is being smart and attempts not
to duplicate the chain unnecessarily, if either there are no attributes
that affect type identity or there is possibly empty set of attributes
that do
> On 04/10/2021 18:48 Iain Sandoe wrote:
>
> For targets that don’t support HAVE_LD_STATIC_DYNAMIC it would be useful
> to push the option back out, so that they can use that to substitute a static
> version
> of the library using %:replace-outfile(-lx libx+.a%s) [ see darwin.h
> for
>
On Tue, 5 Oct 2021, Qing Zhao wrote:
> Hi,
>
> This is the patch to fix this issue based on our discussion.
>
> I have tested it on aarch64 with bootstrap and regtests. X86 bootstrap was
> done, regtests is ongoing.
>
> Okay for trunk?
>
> Thanks.
>
> Qing
>
> ==
> From
When "for" or "some" keyword is missing from a quantified expression, it
is interpreted as an iterated component association by GNAT, leading to
misleading error messages such as:
badquant.ads:4:07: error: expected type "Standard.Boolean"
badquant.ads:4:07: error: found a composite type
Recognize
Nearly complete functional specification of this unit based on the Ada
RM requirements in Ada RM A.4.2. The small differences are noted in
comments. Also add comments from the Ada RM text.
GNATprove proves both absence of runtime errors and that the code
correctly implements the specified contrac
Complete functional specification of this unit based on the Ada RM
requirements in Ada RM A.3.2. This includes also obsolete subprograms
moved to Ada.Characters.Conversions in Ada 2005.
GNATprove proves both absence of runtime errors and that the code
correctly implements the specified contracts.
A previous change improves an error message to "missing ALL or SOME in
quantified expression", but this was effective only -gnat2022 mode.
This change allows the better error message to be printed in older
language modes.
It also treats the relevant features identically, except for the "...is
an A
Just like postconditions, contract cases on formal containers use
functional containers which might leak memory. Disable them.
Tested on x86_64-pc-linux-gnu, committed on trunk
gcc/ada/
* libgnat/a-cfdlli.ads: Use pragma Assertion_Policy to disable
contract cases at execution.
SPARK emitted spurious checks on Unbounded_String variables that were
not initialized. This patch adds a default initial condition to the type
Unbounded_String so that they disappear.
Tested on x86_64-pc-linux-gnu, committed on trunk
gcc/ada/
* libgnat/a-strunb.ads, libgnat/a-strunb__sha
When creating the entity of a derived subprogram we copy all kinds of
properties from the parent, but we failed to copy the ghost status of
the entity.
Tested on x86_64-pc-linux-gnu, committed on trunk
gcc/ada/
* sem_ch3.adb (Derive_Subprogram): Copy ghost status from parent
to d
When building a derived class-wide pre- or postcondition we are mapping
references to inherited formals and to the inherited subprogram itself.
We now do the same for references to the inherited operator. Such
references might appear as prefixes of the Result attribute, e.g.
"="'Result.
Tested on
This set of changes implements the GNAT front end's support for the new
aspects Storage_Model_Type and Designated_Storage_Model. It also defines
a set of functions intended to provide an interface for back ends to
call to retrieve operations and other entities associated with types
that specify the
Contrary to record aggregates, array aggregate cannot mix named and
positional associations. Make it clear in the error message.
Tested on x86_64-pc-linux-gnu, committed on trunk
gcc/ada/
* sem_aggr.adb (Resolve_Array_Aggregate): Improve error message.diff --git a/gcc/ada/sem_aggr.adb b/
See RM A.3.6 and AI12-0260.
Tested on x86_64-pc-linux-gnu, committed on trunk
gcc/ada/
* libgnat/a-zchhan.ads, libgnat/a-zchhan.adb
(Character_Set_Version, Is_Basic, To_Basic): New.
* libgnat/a-zchuni.ads, libgnat/a-zchuni.adb (Is_Basic,
To_Basic): New.diff --git
As part of an improvement to the precision of
Ada.Directories.Modification_Time, we use the type `struct timespec`
when we can (if _POSIX_C_SOURCE is greater than 200809L). For QNX, we
need to include the system header `sys/time.h` to get the type
declaration, `time.h` does not have it.
Tested on
The support for the attribute itself was removed a long time ago.
Tested on x86_64-pc-linux-gnu, committed on trunk
gcc/ada/
* exp_attr.adb (Expand_Fpt_Attribute): Likewise.
* snames.ads-tmpl (Name_Unaligned_Valid): Delete.diff --git a/gcc/ada/exp_attr.adb b/gcc/ada/exp_attr.adb
To get access to the errno variable, we need to include errno.h.
Tested on x86_64-pc-linux-gnu, committed on trunk
gcc/ada/
* init.c (QNX): Add #include errno.h.diff --git a/gcc/ada/init.c b/gcc/ada/init.c
--- a/gcc/ada/init.c
+++ b/gcc/ada/init.c
@@ -2551,6 +2551,7 @@ __gnat_install_han
This marks the local renaming generated for private components of protected
types as coming from source, so that the components are displayed when the
'info locals' command is used in GDB.
Tested on x86_64-pc-linux-gnu, committed on trunk
gcc/ada/
* exp_ch9.adb (Install_Private_Data_Decl
This is the GNAT part of the patch that adds the strub attribute for
function and variable types, command-line options, passes and
adjustments to implement it, documentation, and tests.
Besides documentation, the bulk of the patch adds strub(callable) to
subprograms that the compiler may call impl
Avoid calling End_Label on the Empty node.
Tested on x86_64-pc-linux-gnu, committed on trunk
gcc/ada/
* gcc-interface/trans.c (set_end_locus_from_node): Check that
Handled_Statement_Sequence is not Empty before calling
End_Label, because the Empty node has no End_Label, a
Starting with VxWorks 21.03, a call to fopen() can now set errno to
ENODEV if a prefix of the path does not match any known device. This
led the runtime to raise the wrong exception type when trying to a file
for which the parent directory did not exist and caused the acats
testsuite to fail. This
This is necessary when expression functions are really too large.
Tested on x86_64-pc-linux-gnu, committed on trunk
gcc/ada/
* gcc-interface/trans.c (Subprogram_Body_to_gnu): Do not set the
DECL_DISREGARD_INLINE_LIMITS flag if -gnatd.8 is specified.diff --git a/gcc/ada/gcc-interf
Richard Biener via Gcc-patches writes:
> On Mon, 4 Oct 2021, Qing Zhao wrote:
>
>>
>>
>> > On Oct 4, 2021, at 12:19 PM, Richard Biener wrote:
>> >
>> > On October 4, 2021 7:00:10 PM GMT+02:00, Qing Zhao
>> > wrote:
>> >> I have several questions on this fix:
>> >>
>> >> 1. This fix avoided
Hi!
This patch implements C++23 P2334R1, which is easy because Joseph has done
all the hard work for C2X already.
Unlike the C N2645 paper, the C++ P2334R1 contains one important addition
(but not in the normative text):
"While this is a new preprocessor feature and cannot be treated as a defect
r
On Tue, 5 Oct 2021, Richard Sandiford wrote:
> Richard Biener via Gcc-patches writes:
> > On Mon, 4 Oct 2021, Qing Zhao wrote:
> >
> >>
> >>
> >> > On Oct 4, 2021, at 12:19 PM, Richard Biener wrote:
> >> >
> >> > On October 4, 2021 7:00:10 PM GMT+02:00, Qing Zhao
> >> > wrote:
> >> >> I hav
Before this commit, in case these pragmas were used inside adc file,
gnat1 would fail with a `constraint_error` as it tries to get the
context of the pragma.
This commit induces a regression on dubious uses of these pragmas as
configuration pragmas in ads/adb files.
Review documentation on config
On 01/10/21 20:43 +0100, Jonathan Wakely wrote:
This adds a feature that was recently added to the C++23 working draft.
Signed-off-by: Jonathan Wakely
libstdc++-v3/ChangeLog:
* include/bits/stl_queue.h
(__cpp_lib_adaptor_iterator_pair_constructor): Define for C++23, as
To avoid needing to export a new symbol from the library (for now) the
new member function uses __attribute__((always_inline)).
libstdc++-v3/ChangeLog:
* include/std/ostream (operator<<(const volatile void*)):
Add new overload, as per P1147R1.
* testsuite/27_io/basic_ostre
libstdc++-v3/ChangeLog:
* include/bits/std_function.h (_Any_data::_M_access): Add
noexcept.
(_Function_base::_Base_manager::_M_get_pointer): Likewise.
(_Function_base::_Base_manager::_M_not_empty_function):
Likewise.
Tested powerpc64le-linux. Committed to t
libstdc++-v3/ChangeLog:
PR c++/102535
* testsuite/20_util/is_trivially_constructible/value.cc: Adjust
expected value for C++20.
Tested powerpc64le-linux. Committed to trunk.
commit 7646847df71e57edca5ec5b8c3c3dc4550dcb49d
Author: Jonathan Wakely
Date: Tue Oct 5 09:32:1
Am Di., 5. Okt. 2021 um 10:55 Uhr schrieb Jonathan Wakely via
Libstdc++ :
>
> To avoid needing to export a new symbol from the library (for now) the
> new member function uses __attribute__((always_inline)).
>
> libstdc++-v3/ChangeLog:
>
> * include/std/ostream (operator<<(const volatile vo
On Mon, Oct 4, 2021 at 1:32 PM Martin Liška wrote:
>
> On 10/4/21 13:16, Richard Biener wrote:
> > I meant in merge_one_data do not check ->stamp or ->checksum but instead
> > rely
> > on the counter merging code to detect mismatches (there's read_mismatch and
> > read_error). There's multiple t
On Mon, 4 Oct 2021, H.J. Lu wrote:
> commit adedd5c173388ae505470df152b9cb3947339566
> Author: Jakub Jelinek
> Date: Tue May 3 13:37:25 2016 +0200
>
> re PR target/49244 (__sync or __atomic builtins will not emit 'lock
> bts/btr/btc')
>
> optimized bit test on atomic builtin return with
The following makes sure to amend more function names with the
associated DECL_UID when dumping with TDF_UID, in particular
function names printed as part of calls and in the function header.
That allows one to more easily follow the call flow of PR102528
where coroutines cause three clones of the
Hi,
we have been using an Ada compiler for the Aarch64 architecture configured
with SJLJ exceptions as for the other architectures for some time, and have
not run into any problems up to now so the setting looks obsolete.
OK for the mainline?
2021-10-05 Eric Botcazou
* config/aarc
This moves calls to various option processing stages to one place,
toplev::main.
Bootstrap and regtest ongoing on x86_64-unknown-linux-gnu.
2021-10-05 Richard Biener
* toplev.c (no_backend): Remove global var.
(process_options): Pass in no_backend, move post_options
la
On Mon, Oct 4, 2021 at 6:29 PM Michael Matz wrote:
>
> Hello,
>
> On Mon, 4 Oct 2021, Jeff Law wrote:
>
> > And just in case it got lost. Here's the analysis on 960218-1 on visium:
> >
> > We've got this going into ethread:
> >
> > int f (int x)
> > {
> > int D.1420;
> > int a;
> >
> > ;; b
On Mon, Oct 4, 2021 at 8:28 PM Jason Merrill via Gcc-patches
wrote:
>
> When r12-4038 introduced the global auto_vec save_opt_decoded_options,
> it broke compilers configured with --enable-gather-detailed-mem-stats,
> due to the memory descriptors getting discarded before the auto_vec was
> destro
On Tue, Oct 5, 2021 at 1:27 PM Richard Biener
wrote:
>
> On Mon, Oct 4, 2021 at 8:28 PM Jason Merrill via Gcc-patches
> wrote:
> >
> > When r12-4038 introduced the global auto_vec save_opt_decoded_options,
> > it broke compilers configured with --enable-gather-detailed-mem-stats,
> > due to the m
On Tue, Oct 5, 2021 at 9:11 AM Prathamesh Kulkarni via Gcc-patches
wrote:
>
> Hi,
> In gimple_expand_vec_cond_expr:
>
> icode = get_vcond_icode (mode, cmp_op_mode, unsignedp);
> if (icode == CODE_FOR_nothing)
> {
> if (tcode == LT_EXPR
> && op0a == op0)
> {
>
On Mon, Oct 4, 2021 at 1:13 PM Martin Liška wrote:
>
> On 9/22/21 11:59, Richard Biener wrote:
> > On Thu, Sep 16, 2021 at 3:12 PM Martin Liška wrote:
> >>
> >> This patch comes up with asm_out_state and a new global variable casm.
> >>
> >> Tested on all cross compilers.
> >> Patch can bootstrap
Adapt testcase to better catch the current message
and expected follow-up messages + use the proper
dg-*. — Result: XFAIL changed to PASS.
Committed as r12-4185.
Probably some other testcases could be checked as well.
Maybe we also find some real & unfixed bugs that way.
Tobias
---
Hi!
The following patch implements C++23 P2242R3 - Non-literal variables
(and labels and gotos) in constexpr functions.
I think it is mostly straightforward, don't diagnose certain
statements/declarations just because of their presence in
constexpr/consteval functions, but (except for the non-lite
Hello,
On Tue, 5 Oct 2021, Richard Biener wrote:
> > First notice that this doesn't have an empty latch block to start with
> > (in fact, there is no separate latch block at all), so the loop
> > optimizers haven't been initialized for simple latches at this point.
> > Still, I would say that
Hi All,
This turns an inversion of the sign bit + arithmetic right shift into a
comparison with 0.
i.e.
void fun1(int32_t *x, int n)
{
for (int i = 0; i < (n & -16); i++)
x[i] = (-x[i]) >> 31;
}
now generates:
.L3:
ldr q0, [x0]
cmgtv0.4s, v0.4s, #0
str
The following makes more stmts consumable with the GIMPLE FE
when dumping with -gimple. In particular addresses in GIMPLE
operand position require wrapping with _Literal.
The TDF_ flag space is now exhausted and I've removed overlaps
and re-ordered things as to how it is supposed to work.
Bootst
The gimple FE is too restrictive in what it accepts as
literals, the following makes it also accept &a[10] for example.
Bootstrapped on x86_64-unknown-linux-gnu, testing in progress.
Richard.
2021-10-05 Richard Biener
PR c/102605
gcc/c/
* gimple-parser.c (c_parser_gimple_post
Hi All,
Here's a new version of the patch handling both scalar and vector modes
and non-uniform constant vectors.
Bootstrapped Regtested on aarch64-none-linux-gnu,
x86_64-pc-linux-gnu and no regressions.
In order to not break IVopts and CSE I have added a
requirement for the scalar version to be
On 05/10/2021 13:50, Tamar Christina via Gcc-patches wrote:
Hi All,
This turns an inversion of the sign bit + arithmetic right shift into a
comparison with 0.
i.e.
void fun1(int32_t *x, int n)
{
for (int i = 0; i < (n & -16); i++)
x[i] = (-x[i]) >> 31;
}
Notwithstanding that I
> -Original Message-
> From: Richard Earnshaw
> Sent: Tuesday, October 5, 2021 1:56 PM
> To: Tamar Christina ; gcc-patches@gcc.gnu.org
> Cc: nd ; rguent...@suse.de
> Subject: Re: [PATCH]middle-end convert negate + right shift into compare
> greater.
>
>
>
> On 05/10/2021 13:50, Tamar
On Mon, Oct 4, 2021 at 6:29 PM Michael Matz wrote:
>
> Hello,
>
> On Mon, 4 Oct 2021, Jeff Law wrote:
>
> > And just in case it got lost. Here's the analysis on 960218-1 on visium:
> >
> > We've got this going into ethread:
> >
> > int f (int x)
> > {
> > int D.1420;
> > int a;
> >
> > ;; b
On 05/10/2021 14:30, Tamar Christina wrote:
-Original Message-
From: Richard Earnshaw
Sent: Tuesday, October 5, 2021 1:56 PM
To: Tamar Christina ; gcc-patches@gcc.gnu.org
Cc: nd ; rguent...@suse.de
Subject: Re: [PATCH]middle-end convert negate + right shift into compare
greater.
In g:62acc72a957b5614 I'd stopped the unroller from using
an epilogue loop in cases where the iteration count was
known to be a multiple of the unroll factor. The epilogue
and non-epilogue cases still shared this (preexisting) code
to update the edge frequencies:
basic_block exit_bb = single_pr
> -Original Message-
> From: Richard Earnshaw
> Sent: Tuesday, October 5, 2021 2:34 PM
> To: Tamar Christina ; gcc-patches@gcc.gnu.org
> Cc: nd ; rguent...@suse.de
> Subject: Re: [PATCH]middle-end convert negate + right shift into compare
> greater.
>
>
>
> On 05/10/2021 14:30, Tamar Ch
On 05/10/2021 14:49, Tamar Christina wrote:
-Original Message-
From: Richard Earnshaw
Sent: Tuesday, October 5, 2021 2:34 PM
To: Tamar Christina ; gcc-patches@gcc.gnu.org
Cc: nd ; rguent...@suse.de
Subject: Re: [PATCH]middle-end convert negate + right shift into compare
greater.
On
> -Original Message-
> From: Richard Earnshaw
> Sent: Tuesday, October 5, 2021 2:52 PM
> To: Tamar Christina ; gcc-patches@gcc.gnu.org
> Cc: nd ; rguent...@suse.de
> Subject: Re: [PATCH]middle-end convert negate + right shift into compare
> greater.
>
>
>
> On 05/10/2021 14:49, Tamar
> On Oct 5, 2021, at 3:19 AM, Richard Biener wrote:
>
> On Tue, 5 Oct 2021, Qing Zhao wrote:
>
>> Hi,
>>
>> This is the patch to fix this issue based on our discussion.
>>
>> I have tested it on aarch64 with bootstrap and regtests. X86 bootstrap was
>> done, regtests is ongoing.
>>
>> Oka
From: Petr Mikhalicin
Autotools tests for libstdc++ check only declaration of required
symbols, but some symbols may not be defined.
This patch adds trying to link "C99 TR1" tests, not only compile them.
Signed-off-by: Petr Mikhalicin
---
libstdc++-v3/acinclude.m4 | 16
1 fil
Hi,
On Thu, 2021-09-30 21:27:23 +0200, Jan-Benedict Glaw wrote:
> gcc/ChangeLog:
>
> * common/config/avr/avr-common.c (avr_handle_option): Mark
> argument as ATTRIBUTE_UNUSED.
>
> diff --git a/gcc/common/config/avr/avr-common.c
> b/gcc/common/config/avr/avr-common.c
> index 6486659
Hi,
On Thu, 2021-09-30 21:34:51 +0200, Jan-Benedict Glaw wrote:
> gcc/ChangeLog:
>
> * config/lm32/uclinux-elf.h (LINK_GCC_C_SEQUENCE_SPEC):
> Undefine before redefinition.
>
> diff --git a/gcc/config/lm32/uclinux-elf.h b/gcc/config/lm32/uclinux-elf.h
> index 370df4c55dd..5b638fa5db
On 10/5/2021 8:46 AM, Jan-Benedict Glaw wrote:
Hi,
On Thu, 2021-09-30 21:27:23 +0200, Jan-Benedict Glaw wrote:
gcc/ChangeLog:
* common/config/avr/avr-common.c (avr_handle_option): Mark
argument as ATTRIBUTE_UNUSED.
diff --git a/gcc/common/config/avr/avr-common.c
b/gcc/com
On 10/5/2021 8:46 AM, Jan-Benedict Glaw wrote:
Hi,
On Thu, 2021-09-30 21:34:51 +0200, Jan-Benedict Glaw wrote:
gcc/ChangeLog:
* config/lm32/uclinux-elf.h (LINK_GCC_C_SEQUENCE_SPEC):
Undefine before redefinition.
diff --git a/gcc/config/lm32/uclinux-elf.h b/gcc/config/lm32/
On 10/5/2021 5:22 AM, Richard Biener wrote:
On Mon, Oct 4, 2021 at 6:29 PM Michael Matz wrote:
Hello,
On Mon, 4 Oct 2021, Jeff Law wrote:
And just in case it got lost. Here's the analysis on 960218-1 on visium:
We've got this going into ethread:
int f (int x)
{
int D.1420;
int a;
libstdc++-v3/ChangeLog:
* include/bits/utility.h (__is_in_place_type_v): Define
variable template to detect in_place_type_t specializations.
(__is_in_place_type): Replace class template with alias
template using __is_in_place_type_v.
* include/std/any (any(T
This was omitted from the commit that added these comparisons.
libstdc++-v3/ChangeLog:
* testsuite/20_util/integer_comparisons/greater.cc: New test.
Tested powerpc64le-linux. Committed to trunk.
commit 824e0855732c601e0866d0e8a9264a85f758213e
Author: Jonathan Wakely
Date: Tue Oct 5 1
On 10/4/2021 10:52 AM, Luís Ferreira wrote:
On Thu, 2021-09-23 at 09:50 -0600, Jeff Law wrote:
On 9/23/2021 4:16 AM, ibuclaw--- via Gcc-patches wrote:
On 22/09/2021 03:10 Luís Ferreira wrote:
Currently a stack/heap overflow may happen if a crafted mangle is
maliciously used to cause
Hi Jeff,
On Tue, 2021-10-05 08:53:00 -0600, Jeff Law wrote:
> On 10/5/2021 8:46 AM, Jan-Benedict Glaw wrote:
> > On Thu, 2021-09-30 21:27:23 +0200, Jan-Benedict Glaw
> > wrote:
> > > gcc/ChangeLog:
> > >
> > > * common/config/avr/avr-common.c (avr_handle_option): Mark
> > > argument as ATT
On 10/5/2021 9:03 AM, Jan-Benedict Glaw wrote:
Hi Jeff,
On Tue, 2021-10-05 08:53:00 -0600, Jeff Law wrote:
On 10/5/2021 8:46 AM, Jan-Benedict Glaw wrote:
On Thu, 2021-09-30 21:27:23 +0200, Jan-Benedict Glaw wrote:
gcc/ChangeLog:
* common/config/avr/avr-common.c (avr_handle_optio
On Tue, 5 Oct 2021 at 10:29, Daniel Krügler wrote:
>
> Am Di., 5. Okt. 2021 um 10:55 Uhr schrieb Jonathan Wakely via
> Libstdc++ :
> >
> > To avoid needing to export a new symbol from the library (for now) the
> > new member function uses __attribute__((always_inline)).
> >
> > libstdc++-v3/ChangeL
On 10/5/2021 7:33 AM, Aldy Hernandez wrote:
On Mon, Oct 4, 2021 at 6:29 PM Michael Matz wrote:
Hello,
On Mon, 4 Oct 2021, Jeff Law wrote:
And just in case it got lost. Here's the analysis on 960218-1 on visium:
We've got this going into ethread:
int f (int x)
{
int D.1420;
int a;
On Tue, Oct 05, 2021 at 10:35:12AM +0200, Jakub Jelinek wrote:
> Hi!
>
> This patch implements C++23 P2334R1, which is easy because Joseph has done
> all the hard work for C2X already.
> Unlike the C N2645 paper, the C++ P2334R1 contains one important addition
> (but not in the normative text):
>
More minor fallout from Andrew's work. Committed to the trunk.
Jeff
commit f9e6f20f280daf74763cf3fe21f3c211d3d04620
Author: Jeff Law
Date: Tue Oct 5 11:15:02 2021 -0400
Fix s390 test to have pointer type for computed goto
gcc/testsuite
* gcc.target/s390/pr80725.c: En
> On Oct 5, 2021, at 1:25 AM, Richard Biener wrote:
>
> On Mon, 4 Oct 2021, Qing Zhao wrote:
>
>>
>>
>>> On Oct 4, 2021, at 12:19 PM, Richard Biener wrote:
>>>
>>> On October 4, 2021 7:00:10 PM GMT+02:00, Qing Zhao
>>> wrote:
I have several questions on this fix:
1. This
This is the updated patch, I will push it after done with testing.
Let me know if you see any issue.
Thanks.
Qing.
>From 1f07c20ecdc9a015369c8bb472ebbd3bcaabdf17 Mon Sep 17 00:00:00 2001
From: qing zhao
Date: Tue, 5 Oct 2021 08:10:02 -0700
Subject: [PATCH] middle-
The recently approved P2251R1 paper requires these types to be trivially
copyable. They always have been in libstdc++, but add tests to check it.
libstdc++-v3/ChangeLog:
*
testsuite/21_strings/basic_string_view/requirements/trivially_copyable.cc:
New test.
* testsuite/23_
Hi,
the first issue is that the !gotoff_operand path of legitimize_pic_address in
large PIC model does not make use of REG when it is available, which breaks
for thunks because new pseudo-registers can no longer be created. And the
second issue is that the system compiler (LLVM) generates @GOT
On 10/5/2021 7:33 AM, Aldy Hernandez wrote:
OK?
BTW Jeff, this fixes all the regressions you mention except:
1. pr68911.c
The path not being threaded here is 7->10->11->12. It crosses loops
multiple times, so I believe the restriction code is correct.
7, 10, 12 are in loop1.
11 is in loo
On Sat, 25 Sep 2021, Hu Jialun wrote:
> Hello,
>
> Sorry for bumping it again but I guess it was getting overlooked.
>
> I am very junior with mailing list open source contributions so please feel
> free to point out if I have inadvertantly done something in an incorrect way.
>
> The archive of the
On Tue, Oct 5, 2021 at 6:08 PM Jeff Law wrote:
> OK. I think there may still be some concern if the latch is marked as a
> joiner.I think we should always reject those before the loop optimizers
> run as the joiner's clone would introduce a second latch. I think that can
> be handled in
On Tue, Oct 5, 2021 at 3:07 AM Richard Biener wrote:
>
> On Mon, 4 Oct 2021, H.J. Lu wrote:
>
> > commit adedd5c173388ae505470df152b9cb3947339566
> > Author: Jakub Jelinek
> > Date: Tue May 3 13:37:25 2016 +0200
> >
> > re PR target/49244 (__sync or __atomic builtins will not emit 'lock
>
On 9/28/21 7:50 AM, Richard Biener wrote:
On Wed, Sep 15, 2021 at 10:46 AM Martin Liška wrote:
Hello.
The patch extends the loop unswitching pass so that gswitch
statements are supported. The pass now uses ranger which marks
switch edges that are known to be unreachable in a versioned loop.
P
This patch fixes a bug on the D demangler by parsing and skip anonymous symbols
correctly, according the ABI specification. Furthermore, it also includes tests
to cover anonymous symbols.
The spec specifies [1] that a symbol name can be anonymous and multiple
anonymous symbols are allowed.
[1]: h
Pushed. Jakub already added the feature test macros.
commit 78f03a4a633950743a416f5b9e7f721db7892090
Author: Marek Polacek
Date: Tue Oct 5 12:57:19 2021 -0400
cxx-status: Add papers from the October 2021 WG21 plenary
diff --git a/htdocs/projects/cxx-status.html b/htdocs/projects/cxx-stat
On Tue, 5 Oct 2021, Jakub Jelinek via Gcc-patches wrote:
> One is in the patch below, ignores that sentence and only implements it
> for -std=c++23/-std=gnu++23 like it is only implemented for -std=c23.
> Another option would be to implement it also in the older GNU modes but
> not in the C/CXX mo
On Tue, 2021-10-05 at 09:00 -0600, Jeff Law wrote:
>
>
> On 10/4/2021 10:52 AM, Luís Ferreira wrote:
> > On Thu, 2021-09-23 at 09:50 -0600, Jeff Law wrote:
> > >
> > > On 9/23/2021 4:16 AM, ibuclaw--- via Gcc-patches wrote:
> > > > > On 22/09/2021 03:10 Luís Ferreira
> > > > > wrote:
> > > > >
On Tue, 2021-10-05 at 18:13 +0100, Luís Ferreira wrote:
> This patch fixes a bug on the D demangler by parsing and skip
> anonymous symbols
> correctly, according the ABI specification. Furthermore, it also
> includes tests
> to cover anonymous symbols.
>
> The spec specifies [1] that a symbol nam
This patch to the nvptx backend changes the backend's STORE_FLAG_VALUE
from -1 to 1, by using BImode predicates and selp instructions, instead
of set instructions (almost always followed by integer negation).
Historically, it was reasonable (through rare) for backends to use -1
for representing t
On Tue, 2021-10-05 at 18:13 +0100, Luís Ferreira wrote:
> This patch fixes a bug on the D demangler by parsing and skip anonymous
> symbols
> correctly, according the ABI specification. Furthermore, it also
> includes tests
> to cover anonymous symbols.
>
> The spec specifies [1] that a symbol nam
Played around with the warning in the 'omp_lib' module (needs tweaking
as for the current version, no warning is done). Turned out that already
use omp_lib
outputs a warning even when not used.
That's fixed by the attached patch - even if the location is not perfect.
OK for GCC 12 + GCC 11 bac
On 10/5/21 10:32, Qing Zhao wrote:
On Oct 5, 2021, at 3:19 AM, Richard Biener wrote:
On Tue, 5 Oct 2021, Qing Zhao wrote:
Hi,
This is the patch to fix this issue based on our discussion.
I have tested it on aarch64 with bootstrap and regtests. X86 bootstrap was
done, regtests is ongoing
Without dwarf2 unwind tables available _Unwind_Backtrace() is not
able to return the full backtrace.
This patch adds a fallback function on powerpc to get the backtrace
by doing a backchain, this code was originally at glibc.
libgcc/ChangeLog:
* config/rs6000/linux-unwind.h (struct rt_sig
On 10/5/21 07:27, Richard Biener wrote:
On Mon, Oct 4, 2021 at 8:28 PM Jason Merrill via Gcc-patches
wrote:
When r12-4038 introduced the global auto_vec save_opt_decoded_options,
it broke compilers configured with --enable-gather-detailed-mem-stats,
due to the memory descriptors getting discar
On 10/5/21 03:50, Jakub Jelinek wrote:
Hi!
The following testcase ICEs on x86_64-linux with -m32 due to a bug in
apply_identity_attributes. The function is being smart and attempts not
to duplicate the chain unnecessarily, if either there are no attributes
that affect type identity or there is
On 10/1/21 13:29, Patrick Palka wrote:
On Fri, 1 Oct 2021, Jason Merrill wrote:
On 10/1/21 10:26, Patrick Palka wrote:
On Fri, 1 Oct 2021, Jason Merrill wrote:
On 10/1/21 09:46, Patrick Palka wrote:
Here during partial ordering of the two partial specializations we end
up in unify with parm
On 10/1/21 16:14, Marek Polacek wrote:
On Fri, Oct 01, 2021 at 09:30:41AM -0400, Jason Merrill wrote:
On 9/30/21 17:56, Marek Polacek wrote:
On Thu, Sep 30, 2021 at 03:34:24PM -0400, Jason Merrill wrote:
On 9/30/21 10:50, Marek Polacek wrote:
This patch addresses one of my leftovers from GCC
Hi Tobias,
Am 05.10.21 um 20:03 schrieb Tobias Burnus:
Played around with the warning in the 'omp_lib' module (needs tweaking
as for the current version, no warning is done). Turned out that already
use omp_lib
outputs a warning even when not used.
that must have been a non-trivial example;
On Mon, 4 Oct 2021, Patrick Palka wrote:
> When passing a function template as the argument to a function NTTP
> inside a template, we resolve it to the right specialization ahead of
> time via resolve_address_of_overloaded_function, though the call to
> mark_used within defers odr-using it until
On 10/1/21 15:52, Nick Huang wrote:
...the subject line for the commit should be the first line of the
commit message, followed by a blank line, followed by the description of
the patch; without the subject line, git format-patch thought your whole
description was the subject of the patch.
oh,
This fixes a bootstrap fail because saw_static_libcxx was unused for
targets without support for -Bstatic/dynamic.
The fix applied pushes the -static-libstdc++ back onto the command
line, which allows a target to substitute a static version of the
c++ standard library using specs.
tested on x86_6
1 - 100 of 117 matches
Mail list logo
