Adjust BLKmode argument size for parameter alignment for sibcall check.
gcc/
PR middle-end/117098
* calls.cc (store_one_arg): Adjust BLKmode argument size for
alignment padding for sibcall check.
gcc/testsuite/
PR middle-end/117098
* gcc.dg/sibcall-12.c: New test.
OK for master?
H.J.
From 8b
This is a better patch to fix PR 117096 (phiopt vs clobbers).
The only time we remove clobbers of local variables is during
remove_unused_locals. Since DSE might remove all of the stores
to a local variable, it makes sense to also try to remove unused local
variables afterwards.
This shows up more
On 11 October 2024 16:42:44 CEST, David Malcolm wrote:
>On Fri, 2024-10-11 at 15:34 +0100, Paul Richard Thomas wrote:
>> Hi Tobias,
>>
>> Good catch! It looks 'obvious' to me too :-)
yeah, thanks for the explicit cleanup
>> > * I want to move support range-based locations, which is also a
>>
I noticed that several diagnostic strings were not tagged as translatable.
I fixed them by adding _ or G_ as prefix ( →gcc/ABOUT-GCC-NLS) and moved a single-use string to the message to make
it more readable. One error message did not quit fit the pattern, hence,
I modified it slightly, and a fe
Dear friend，
This is Anne, Office Manager of Wellsucceed Embroidery from Dongguan, China, sorry to occupy you a few seconds.
We are a professional underwear manufacturer with more than 10 years experience in this field. Many big brands are our clients. All kinds of styles can be branded here wit
Am Samstag, dem 12.10.2024 um 18:44 +0200 schrieb Richard Biener:
>
> > Am 12.10.2024 um 16:43 schrieb Martin Uecker :
> >
> >
> > There is code which should not fail at run-time. For this,
> > it is helpful to get a warning when a compiler inserts traps
> > (e.g. sanitizers, hardbools, __buil
Hi Jerry,
the attached patch implements the constants for UNSIGNED for
ISO_FORTRAN_ENV and ISO_C_BINDING. With this, the implementation
of UNSIGNED for gfortran should be complete, modulo bugs, of course.
OK for trunk?
Looks good to go Thomas.
Committed, thanks for the review!
Actually, t
> Am 12.10.2024 um 16:43 schrieb Martin Uecker :
>
>
> There is code which should not fail at run-time. For this,
> it is helpful to get a warning when a compiler inserts traps
> (e.g. sanitizers, hardbools, __builtin_trap(), etc.).
>
> Having a warning for this also has many other use case
On 10/12/24 9:04 AM, Thomas Koenig wrote:
Hello world,
the attached patch implements the constants for UNSIGNED for
ISO_FORTRAN_ENV and ISO_C_BINDING. With this, the implementation
of UNSIGNED for gfortran should be complete, modulo bugs, of course.
OK for trunk?
Looks good to go Thomas.
Th
Anyone feeling like reviewing this patch? (Mainly the Fortran side?!?) —
Or should I declare it as OpenMP/(OpenACC) patch and just commit it?
→ https://gcc.gnu.org/pipermail/gcc-patches/2024-October/664985.html
Tobias
Tobias Burnus write:
I forgot to update the subject line. To make it easier
Hello world,
the attached patch implements the constants for UNSIGNED for
ISO_FORTRAN_ENV and ISO_C_BINDING. With this, the implementation
of UNSIGNED for gfortran should be complete, modulo bugs, of course.
OK for trunk?
Best regards
Thomas
gcc/fortran/ChangeLog:
* dump-par
Added.
Thanks,
Feng
From: Richard Biener
Sent: Saturday, October 12, 2024 8:12 PM
To: Feng Xue OS
Cc: gcc-patches@gcc.gnu.org
Subject: Re: [PATCH] vect: Fix inconsistency in fully-masked lane-reducing op
generation [PR116985]
On Sat, Oct 12, 2024 at 9:1
This is a minor patch from Jivan from roughly a year ago. The basic
idea here is similar to what we do when extending values for the sake of
comparisons. Specifically if the value is already known to be properly
extended, then an extension is just a copy.
The original idea was to use a simil
For vector types we have to make sure the comparison result is a vector
type and the resulting compare operation is supported. As the resulting
compare is never an equality compare I didn't bother to check for the
cbranch case.
Bootstrapped and tested on x86_64-unknown-linux-gnu, pushed
There is code which should not fail at run-time. For this,
it is helpful to get a warning when a compiler inserts traps
(e.g. sanitizers, hardbools, __builtin_trap(), etc.).
Having a warning for this also has many other use cases, e.g.
one can use it with some sanitizer to rule out that some
pi
On 10/10/24 7:59 PM, Tsung Chun Lin wrote:
Hi Jeff,
Thanks for reviewing and sorry for the testsuite without any update in
my patch to cause regression failures..
I will re-submit the patches with the updated testsuite.
No worries, it happens to all of us at some time or another. I should
Now committed as r15-4291-g34b77d1b9ac53c – see attachment.
Except that I have excluded the diagnostic-location changes;
here, using ranges makes more sense. (I have patch, that works,
but I need to check that it won't cause corner-case issues.)
Fortran-part of the ChangeLog (see attachment for
On 10/11/24 4:28 AM, Richard Sandiford wrote:
Kyrylo Tkachov writes:
Hi all,
In the testcase from patch [2/2] we want to match a vector rotate operation from
an IOR of left and right shifts by immediate. simplify-rtx has code for just
that but it looks like it's prepared to do handle only
On Sat, Oct 12, 2024 at 9:12 AM Feng Xue OS wrote:
>
> To align vectorized def/use when lane-reducing op is present in loop
> reduction,
> we may need to insert extra trivial pass-through copies, which would cause
> mismatch between lane-reducing vector copy and loop mask index. This could be
> f
On Sat, Oct 12, 2024 at 6:27 AM wrote:
>
> From: Pan Li
>
> This patch would like to support the form 3 of the vector signed
> integer SAT_SUB. Aka below example:
>
> Form 3:
> #define DEF_VEC_SAT_S_SUB_FMT_3(T, UT, MIN, MAX) \
> void __attribute__((noinline))
On Sat, Oct 12, 2024 at 3:42 AM Andrew Pinski wrote:
>
> Clobbers in a condition don't cause any improvements and are getting
> in the way of doing match and simplify with phiopt. Need to ignore/skip
> over them in when seeing if there is only one statement that can moved.
> Also since clobbers ha
On Fri, Oct 11, 2024 at 8:37 PM Victor Do Nascimento
wrote:
>
> The recent refactoring of the dot_prod optab to convert-type exposed a
> limitation in how `find_widening_optab_handler_and_mode' is currently
> implemented, owing to the fact that, while the function expects the
>
> GET_MODE_CLASS
Hi,
"James K. Lowden" writes:
> Hello,
>
> I just joined the list to begin contributing patches for the COBOL
> front end we've been touting for the last 4 years. It's my first
> attempt. Please tell me if you'd like to see something different.
>
> What follows mimics to some degree the out
> Am 12.10.2024 um 12:00 schrieb Jakub Jelinek :
>
> Hi!
>
> Seems our buildbot is unhappy about my latest commit to link genmatch with
> libcommon.a in order to support gcc_diag diagnostics in libcpp.
>
> We have in gcc/configure.ac:
> if test x$enable_host_shared = xyes; then
> PICFLAG=-f
Hi!
Seems our buildbot is unhappy about my latest commit to link genmatch with
libcommon.a in order to support gcc_diag diagnostics in libcpp.
We have in gcc/configure.ac:
if test x$enable_host_shared = xyes; then
PICFLAG=-fPIC
elif test x$enable_host_pie = xyes; then
PICFLAG=-fPIE
elif test
On Sat, Oct 12, 2024 at 12:21 AM H.J. Lu wrote:
>
> Before
>
> 1089d083117 Simplify (B * v + C) * D -> BD* v + CD when B,C,D are all
> INTEGER_CS
> T.
>
> the loop was
>
> .L2:
> movl (%rdi,%rdx), %eax
> addl $12345, %eax
> imull $-1564285888, %eax, %eax
> leal -333519936(%rax), %eax
> movl %eax,
On Sat, Oct 12, 2024 at 12:23 AM H.J. Lu wrote:
>
> Use word_mode integer with func so that 64-bit integer is used with
> x32.
>
> * gcc.target/i386/pr115749.c (uword): New.
> (func): Replace unsigned long with uword.
OK.
Thanks,
Uros.
On Sat, Oct 12, 2024 at 12:22 AM H.J. Lu wrote:
>
> Since long is 32-bit for x32, use long long for 64-bit integer.
>
> * gcc.target/i386/pr55583.c: Use long long for 64-bit integer.
OK.
Thanks,
Uros.
On Sat, Oct 12, 2024 at 12:19 AM H.J. Lu wrote:
>
> Since x32 uses (%edx), instead of (%rdx), also scan (%edx).
>
> * gcc.target/i386/invariant-ternlog-1.c: Also scan (%edx).
OK.
Thanks,
Uros.
PR middle-end/112600
gcc/testsuite/ChangeLog:
* gcc.target/i386/pr112600-4a.c: New test.
* gcc.target/i386/pr112600-4b.c: New test.
Tested on x86_64-linux-gnu {,-m32}.
Uros.
diff --git a/gcc/testsuite/gcc.target/i386/pr112600-4a.c
b/gcc/testsuite/gcc.target/i386/pr112600-4a.c
new f
Hi Tobias,
This unifies the two locus to location_t conversion functions, preparing
for some changes I want to do later.
In principle, I had the patch this morning; however, the assert is now
exercised more often than before - and it triggered rather unexpected
when running the testsuite.
To align vectorized def/use when lane-reducing op is present in loop reduction,
we may need to insert extra trivial pass-through copies, which would cause
mismatch between lane-reducing vector copy and loop mask index. This could be
fixed by computing the right index around a new counter on effecti
32 matches
Mail list logo
