Richard Biener writes:
>> gcc/testsuite/
>> PR tree-optimization/114440
>> * gcc.dg/vect/vect-reduc-chain-1.c
>> * gcc.dg/vect/vect-reduc-chain-2.c
>> * gcc.dg/vect/vect-reduc-chain-3.c
>> * gcc.dg/vect/vect-reduc-chain-dot-slp-1.c
>> * gcc.dg/vect/
pan2...@intel.com writes:
> From: Pan Li
>
> The direct_internal_fn_supported_p has no restrictions for the type
> modes. For example the bitfield like below will be recog as .SAT_TRUNC.
>
> struct e
> {
> unsigned pre : 12;
> unsigned a : 4;
> };
>
> __attribute__((noipa))
> void bug (e * v,
Hi!
First, note this is now GCC PR115989
"[15 regression] libgomp.oacc-fortran/privatized-ref-2.f90 fails after
r15-2135-gc3aa339ea50f05".
Otherwise:
On 2024-07-19T06:54:46+0100, Paul Richard Thomas
wrote:
> Thanks for doing that test. Here is what the error looks like on 14-branch:
> libgomp
Previously we optimized expressions of the form a < 0 ? -1 : 0 to
(signed)a >> 31 during vcond expanding. Since r15-1741-g2ccdd0f22312a1
this is done in match.pd. The implementation in the back end as well as
in match.pd are basically the same but still distinct. For the tests in
vcond-shift.c t
Hi Haochen,
Thanks for doing that test. Here is what the error looks like on 14-branch:
libgomp.oacc-fortran/privatized-ref-2.f90:36:22:
36 | A = [(3*j, j=1, 10)]
| ^
Warning: ‘a.offset’ is used uninitialized [-Wuninitialized]
libgomp.oacc-fortran/privatized-ref-2.f
Attempting to resend with a different client, plain text enabled...
Hi Jason,
I’ve managed to address your review comments, and rewrite the patch to
include both -Winvalid-noreturn and -Winvalid-noreturn= but have trouble
figuring out how to format invoke.texi and where to add the documentation
f
There are still some cases which can't utilize vx or vf after
last_combine pass.
1. integer comparison when imm isn't in range of [-16, 15]
2. float imm is 0.0
3. DI or DF mode under RV32
This patch fix above mentioned issues.
Tested on RV32 and RV64.
Signed-off-by: demin.han
gcc/ChangeLog:
On Thu, Jul 18, 2024 at 10:31 PM Stefan Schulze Frielinghaus
wrote:
>
> Previously we optimized expressions of the form a < 0 ? -1 : 0 to
> (signed)a >> 31 during vcond expanding. Since r15-1741-g2ccdd0f22312a1
> this is done in match.pd. The implementation in the back end as well as
> in match.
On Thu, Jul 18, 2024 at 11:58:10PM -0700, Andrew Pinski wrote:
> On Thu, Jul 18, 2024 at 10:31 PM Stefan Schulze Frielinghaus
> wrote:
> >
> > Previously we optimized expressions of the form a < 0 ? -1 : 0 to
> > (signed)a >> 31 during vcond expanding. Since r15-1741-g2ccdd0f22312a1
> > this is d
Currently, some binops of vector vs double scalar under RV32 can't
translated to vf but vfmv+vxx.vv.
The cause is that vec_duplicate is also expanded to broadcast for double mode
under RV32. last-combine can't process expanded broadcast.
gcc/ChangeLog:
* config/riscv/vector.md: Add !FLOA
On Thu, Jul 18, 2024 at 09:53:05AM -0500, Peter Bergner wrote:
> On 7/18/24 8:23 AM, Segher Boessenkool wrote:
> > Everything in gcc.target/powerpc is only run for target powerpc*-*-*
> > anyway, so make this just
> >
> > /* { dg-do compile } */
> >
> > please. (Or nothing, it is the default aft
Hello Richard:
All comments are addressed.
Common infrastructure using generic code for pair mem fusion of different
targets.
rs6000 target specific code implement virtual functions defined by generic code.
Target specific code are added in rs6000-mem-fusion.cc.
Bootstrapped and regtested on p
Hello Richard:
On 18/07/24 4:44 pm, Richard Sandiford wrote:
> Ajit Agarwal writes:
>> [...]
+// Set subreg for OO mode pair to generate sequential registers given
+// insn_info pairs I1, I2 and LOAD_P is true iff load insn and false
+// if store insn.
+void
+rs6000_pair_
On Fri, Jun 21, 2024 at 05:30:02PM +0200, Tobias Burnus wrote:
> gcc/ChangeLog:
>
> * config/gcn/mkoffload.cc (read_file): Remove.
> (process_obj): Generate C file that uses #embed.
> (main): Update call to it; remove no longer needed file I/O.
> + fprintf (cfile,
> +"s
> + TEST_COND_IMM_FLOAT (T, >, 0.0, _gt)
> \
> + TEST_COND_IMM_FLOAT (T, <, 0.0, _lt)
> \
> + TEST_COND_IMM_FLOAT (T, >=, 0.0, _ge)
> \
> + TEST_COND_IMM_FLOAT (T, <=, 0.0, _le)
From: Pan Li
The direct_internal_fn_supported_p has no restrictions for the type
modes. For example the bitfield like below will be recog as .SAT_TRUNC.
struct e
{
unsigned pre : 12;
unsigned a : 4;
};
__attribute__((noipa))
void bug (e * v, unsigned def, unsigned use) {
e & defE = *v;
Thanks Richard S for comments and suggestions, updated in v2.
Pan
-Original Message-
From: Richard Sandiford
Sent: Friday, July 19, 2024 3:46 PM
To: Li, Pan2
Cc: gcc-patches@gcc.gnu.org; juzhe.zh...@rivai.ai; kito.ch...@gmail.com;
richard.guent...@gmail.com; tamar.christ...@arm.com; j
> -Original Message-
> From: Li, Pan2
> Sent: 2024年7月19日 18:33
> To: Demin Han ; gcc-patches@gcc.gnu.org
> Cc: juzhe.zh...@rivai.ai; kito.ch...@gmail.com; jeffreya...@gmail.com;
> rdapp@gmail.com
> Subject: RE: [PATCH v2] RISC-V: More support of vx and vf for autovec
> comparison
>
>
Hi Paul,
thanks for the review.
> While I realise that this is not your doing, should we not
> check DECL_LANG_SPECIFIC ()) before touching GFC_DECL_SAVED_DESCRIPTOR?
I like that idea. I have added it. But what should we do when
DECL_LANG_SPECIFIC is not set? I have chosen to add a gcc_unreachab
Hi Harald,
you right, that solution looks pretty obvious to me. Ok by me. Thanks for the
patch and the opportunity to give back a review.
Keep up the good work!
Regards,
Andre
On Thu, 18 Jul 2024 21:27:13 +0200
Harald Anlauf wrote:
> Dear all,
>
> here's a quite obvious fix for an ICE
Fixed and pushed!
Thanks,
Cupertino
David Faust writes:
> On 7/15/24 08:33, Cupertino Miranda wrote:
>> Both xchg and cmpxchg instructions, in the pseudo-C dialect, do not
>> expect their memory address operand to be surrounded by parentheses.
>> For example, it should be output as "w0 =cmpxch
Ok for trunk and releases/gcc-14?
--
On Cortex-A7 with -mfloat-abi=hard and -mfpu=neon, the following
instructions are generated for the test case:
vldrd16, .L3
vst1.32 {d16}, [r0]
For Cortex-A7 with -mfloat-abi=soft, it's instead:
movsr2, #1
movsr3,
On 11/07/2024 19:31, Richard Sandiford wrote:
> These tests used to generate:
>
> bl swap
> ldr r2, [sp, #4]
> mov r0, r2 @ __fp16
>
> but g:9d20529d94b23275885f380d155fe8671ab5353a means that we can
> load directly into r0:
>
> bl swap
>
On Thu, 18 Jul 2024, Jason Merrill wrote:
> On 7/18/24 12:45 PM, Patrick Palka wrote:
> > Bootstrapped and regtested on x86_64-pc-linux-gnu, does thi look
> > OK for trunk/14?
> >
> > -- >8 --
> >
> > As a followup of r15-2047-g7954bb4fcb6fa8, we also need to consider
> > dependent attributes wh
On 7/18/24 9:13 PM, Xiao Zeng wrote:
2024-07-18 01:53 Jeff Law wrote:
On 7/17/24 2:01 AM, Xiao Zeng wrote:
libgcc/ChangeLog:
* Makefile.in: Support __divbc3 and __mulbc3.
* libgcc2.c (if): Support BC mode for __bf16.
(defined): Ditto.
(MTYPE): Ditto.
(CTYPE): Ditto.
(AMTYPE): Ditto.
(M
On Thu, Jul 18, 2024 at 04:18:56PM -0700, Andi Kleen wrote:
> > > > > + set_musttail_on_return (retval, xloc, musttail_p);
> > > > > +
> > > > >if (retval)
> > > > > {
> > > > >tree semantic_type = NULL_TREE;
> > > >
> > > > Is it deliberate that set_musttail_on_return is called
The following addresses the bad hash function of cselib which uses
integer plus for merging. This causes a huge number of collisions
for the testcase in the PR and thus very large compile-time.
The following rewrites it to use inchash, eliding duplicate mixing
of RTX code and mode in some cases a
On 19/07/2024 14:07, Torbjörn SVENSSON wrote:
> Ok for trunk and releases/gcc-14?
>
> --
>
> On Cortex-A7 with -mfloat-abi=hard and -mfpu=neon, the following
> instructions are generated for the test case:
>
> vldrd16, .L3
> vst1.32 {d16}, [r0]
>
> For Cortex-A7 with -mfloat
On 12/07/2024 09:02, Torbjörn SVENSSON wrote:
> Ok for trunk and releases/gcc-14?
>
> --
>
> Overriding the -mpfu and -mfloat-abi might be incompatible with selected
> multilib. As a result, verify that the current multilib is compatible
> with the effective target without changing the -mfpu or -
On Fri, Jul 5, 2024 at 12:18 PM Patrick Palka wrote:
>
> Bootstrapped and regtested on x86_64-pc-linux-gnu, does this look OK
> for trunk/14 and perhaps 13?
>
> Alternatively we can set current_template_parms from weakly_subsumes
> instead, who has only one caller anyway.
Ping.
>
> -- >8 --
>
>
This patch series is only a refactoring of the existing implementation of PAuth
and returned-address signing. The existing behavior is preserved.
1. aarch64: store signing key and signing method in DWARF _Unwind_FrameState
_Unwind_FrameState already contains several CIE and FDE information (see
This patch is only a refactoring of the existing implementation
of PAuth and returned-address signing. The existing behavior is
preserved.
_Unwind_FrameState already contains several CIE and FDE information
(see the attributes below the comment "The information we care
about from the CIE/FDE" in l
This patch provides a new handler MD_ARCH_FRAME_STATE_T to hide an
architecture-specific structure containing CIE and FDE data related
to DWARF architecture extensions.
Hiding the architecture-specific attributes behind a handler has the
following benefits:
1. isolating those data from the generic
On Fri, Jul 5, 2024 at 1:50 PM Patrick Palka wrote:
>
> Bootstrapped and regtested on x86_64-pc-linux-gnu, does this look
> OK for trunk/14?
>
> -- >8 --
>
> During the alias CTAD transformation, if substitution failed for some
> guide we should just discard the guide silently. We currently do
>
libgcc/ChangeLog:
* configure: Regenerate.
---
libgcc/configure | 2 ++
1 file changed, 2 insertions(+)
diff --git a/libgcc/configure b/libgcc/configure
index a69d314374a..15a0be23644 100755
--- a/libgcc/configure
+++ b/libgcc/configure
@@ -587,6 +587,7 @@ ac_includes_default='/* none */
On 7/18/24 3:12 PM, Georg-Johann Lay wrote:
This new builtin provides a faster way to compute
expressions like 1 << x or ~(1 << x) that are sometimes
used as masks for setting bits in the I/O region, and when
x is not known at compile-time.
The open coded C expression takes 5 + 4 * x cycles
On Sat, Jul 6, 2024 at 8:22 PM Patrick Palka wrote:
>
> Bootstrapped and regtested on x86_64-pc-linux-gnu, does this
> look OK for trunk/14?
Ping.
>
> -- >8 --
>
> The code path for rejecting an object-less call to a non-static
> member function should also consider xobj member functions so
> th
On 11/07/2024 10:36, Christophe Lyon wrote:
> This patch fixes a bug where the mode iterator for mve_vdup
> should be MVE_VLD_ST instead of MVE_vecs: V2DI and V2DF (thus vdup.64)
> are not supported by MVE.
>
> 2024-07-02 Jolen Li
> Christophe Lyon
>
> gcc/
> * config/a
Hi,
Jakub Jelinek wrote:
+ "#if defined(__STDC_EMBED_FOUND__) && __has_embed (\"%s\") ==
__STDC_EMBED_FOUND__\n"
If this was an attempt to deal gracefully with no #embed support, then
the above would be wrong and should have been
#if defined(__STDC_EMBED_FOUND__) && defined(__has_embe
Seems to be fixed by r15-521-g6ad7ca1bb90573.
PR c++/109464
gcc/testsuite/ChangeLog:
* g++.dg/template/explicit-instantiation8.C: New test.
---
.../g++.dg/template/explicit-instantiation8.C | 24 +++
1 file changed, 24 insertions(+)
create mode 100644 gcc/testsu
On 2024-07-19 16:36, Richard Earnshaw (lists) wrote:
On 19/07/2024 14:07, Torbjörn SVENSSON wrote:
Ok for trunk and releases/gcc-14?
--
On Cortex-A7 with -mfloat-abi=hard and -mfpu=neon, the following
instructions are generated for the test case:
vldrd16, .L3
vst1.32
On 11/07/2024 10:36, Christophe Lyon wrote:
> This patch makes the non-predicated vdupq_n MVE intrinsics use
> vec_duplicate rather than an unspec. This enables the compiler to
> generate better code sequences (for instance using vmov when
> possible).
>
> The patch renames the existing mve_vdup
On 19/07/2024 16:10, Torbjorn SVENSSON wrote:
>
>
> On 2024-07-19 16:36, Richard Earnshaw (lists) wrote:
>> On 19/07/2024 14:07, Torbjörn SVENSSON wrote:
>>> Ok for trunk and releases/gcc-14?
>>>
>>> --
>>>
>>> On Cortex-A7 with -mfloat-abi=hard and -mfpu=neon, the following
>>> instructions are
On 7/19/24 2:29 AM, Julian Waters wrote:
Attempting to resend with a different client, plain text enabled...
Still corrupted by word wrap and quoted-printable encoding,
unfortunately. Probably the easiest solution is to send the patch as an
attachment rather than pasted into the message.
I
On 7/6/24 8:22 PM, Patrick Palka wrote:
Bootstrapped and regtested on x86_64-pc-linux-gnu, does this
look OK for trunk/14?
OK.
-- >8 --
The code path for rejecting an object-less call to a non-static
member function should also consider xobj member functions so
that we properly reject the be
On 7/19/24 10:55 AM, Patrick Palka wrote:
On Fri, Jul 5, 2024 at 1:50 PM Patrick Palka wrote:
Bootstrapped and regtested on x86_64-pc-linux-gnu, does this look
OK for trunk/14?
-- >8 --
During the alias CTAD transformation, if substitution failed for some
guide we should just discard the gui
Hi Andre,
The patch looks fine to me. Please add the original testcase as
pr88624.f90, since it can be a compile only. The addition to
coarray/dummy_1.f90 is fine as well but I think that it is good to address
the reporter's problem directly.
Thanks
Paul
On Wed, 17 Jul 2024 at 14:10, Andre Veh
On Fri, 19 Jul 2024, Jason Merrill wrote:
> On 7/19/24 10:55 AM, Patrick Palka wrote:
> > On Fri, Jul 5, 2024 at 1:50 PM Patrick Palka wrote:
> > >
> > > Bootstrapped and regtested on x86_64-pc-linux-gnu, does this look
> > > OK for trunk/14?
> > >
> > > -- >8 --
> > >
> > > During the alias C
Hi All,
Ping!
I understand now why this works. The scope of the block is merged and so
all the previous declarations that would otherwise disappear are added,
even by the empty statement.
Regards
Paul
On Mon, 15 Jul 2024 at 17:10, Paul Richard Thomas <
paul.richard.tho...@gmail.com> wrote:
>
Thanks for the patch and chasing down the magic.
Path is ok to commit.
--
steve
On Fri, Jul 19, 2024 at 05:32:26PM +0100, Paul Richard Thomas wrote:
> Hi All,
>
> Ping!
>
> I understand now why this works. The scope of the block is merged and so
> all the previous declarations that would othe
Min and max could be optimized if both operands are defined by
(same) variable restricted by an and(&). For signed types,
optimization can be done when both constant have same sign bit.
The patch also adds optimization for specific case of min/max(a, a&CST).
This patch adds match pattern for:
max
On Thu, Jul 18, 2024 at 08:08:44AM -0500, Segher Boessenkool wrote:
> Hi!
>
> [ I reviewed this together with Ke Wen. All blame should go to me, all
> praise to him. ]
>
> On Wed, Jul 10, 2024 at 01:34:26PM -0400, Michael Meissner wrote:
> > [This is a repost of the June 4th patch]
>
> It is
In g:9d20529d94b23275885f380d155fe8671ab5353a, I'd extended
insn_propagation to handle simple cases of hard-reg mode punning.
The punned "to" value was created using simplify_subreg rather
than simplify_gen_subreg, on the basis that hard-coded subregs
aren't generally useful after RA (where hard-re
Previously we built vector boolean constants using 1 for true
elements and 0 for false elements. This matches the predicates
produced by SVE's PTRUE instruction, but leads to a miscompilation
on AVX512, where all bits of a boolean element should be set.
One option for RTL would be to make this ta
> Am 19.07.2024 um 19:44 schrieb Richard Sandiford :
>
> Previously we built vector boolean constants using 1 for true
> elements and 0 for false elements. This matches the predicates
> produced by SVE's PTRUE instruction, but leads to a miscompilation
> on AVX512, where all bits of a boolean
On 7/19/24 12:34 PM, Michael Meissner wrote:
> On Thu, Jul 18, 2024 at 08:08:44AM -0500, Segher Boessenkool wrote:
>>> --- a/gcc/config/rs6000/ppc-auxv.h
>>> +++ b/gcc/config/rs6000/ppc-auxv.h
>>> @@ -47,9 +47,8 @@
>>> #define PPC_PLATFORM_PPC47612
>>> #define PPC_PLATFORM_POWER8
On 7/19/24 2:55 AM, demin.han wrote:
Currently, some binops of vector vs double scalar under RV32 can't
translated to vf but vfmv+vxx.vv.
The cause is that vec_duplicate is also expanded to broadcast for double mode
under RV32. last-combine can't process expanded broadcast.
gcc/ChangeLog:
The inner loop in build_option_suggestions uses OPTION to take the
address of OPTB and use it across iterations, which is undefined
behaviour since OPTB is defined within the loop. Pull it outside the
loop to make this defined.
It's a trivial fix, so posting early. OK to commit once build and te
On Fri, Jul 19, 2024 at 12:52:06PM -0400, Siddhesh Poyarekar wrote:
> The inner loop in build_option_suggestions uses OPTION to take the
> address of OPTB and use it across iterations, which is undefined
> behaviour since OPTB is defined within the loop. Pull it outside the
> loop to make this def
GCC developers:
The following patch adds the int128 varients to the existing overloaded
built-ins vec_sld, vec_sldb, vec_sldw, vec_sll, vec_slo, vec_srdb,
vec_srl, vec_sro. These varients were requested by Steve Munroe.
The patch has been tested on a Power 10 system with no regressions.
Ple
Upstream commit
https://github.com/llvm/llvm-project/commit/26800a2c7e7996dc773b4e990dd5cca41c45e1a9
of LLVM added a #undef _TIME_BITS in
libsanitizer/sanitizer_common/sanitizer_platform_limits_posix.cpp to
fix the build on 32-bit Linux platforms that have enabled 64-bit
time_t using _TIME_BITS=64.
> -(match_operand:V_VLSF 3 "register_operand")]))]
> +(match_operand:V_VLSF 3 "nonmemory_operand")]))]
Even though the integer compares have nonmemory operand here their respective
insn patterns don't (but constrain properly).
I guess what's happening with register operand and a c
> I have a test.
> The backend can't see -0.0 and It becomes 0.0 when translate to gimple.
I don't think it should except when specifying -ffast-math or similar.
But we don't have a shortcut to load a negative zero, just the positive
one.
--
Regards
Robin
When registering an equivalence, the oracle creates an initial
equivalence for an SSA_NAME with itself in the def block. This
primarily prevents dominator queries from searching past the definition
block.
If the definition stmt for the SSA_NAME has not been placed in the CFG
yet, we should
Hi!
On 2021-01-13T15:48:42-0800, Julian Brown wrote:
> For AMD GCN, the instructions available for loading/storing vectors are
> always scatter/gather operations (i.e. there are separate addresses for
> each vector lane), so the current heuristic to avoid gather/scatter
> operations with too many
"Kewen.Lin" writes:
> Hi Sam,
Hi Kewen,
>
> on 2024/7/19 11:28, Sam James wrote:
>> 'dg-run' is not a valid dejagnu directive, 'dg-do run' is needed here
>> for the test to be executed.
>>
>> 2024-07-18 Sam James
>>
>> PR target/108699
>> * gcc.target/powerpc/pr108699.c: Fix 'dg-
Originally added in r0-44646-g204250d2fcd084 and r0-44627-gfd350d241fecf6 whic
moved -fno-common from all builds to just checking builds.
Since r10-4867-g6271dd984d7f92, GCC defaults to -fno-common. There's no need
to pass it specially for checking builds.
We could keep it for older bootstrap com
On Fri, Jul 19, 2024 at 5:23 PM Sam James wrote:
>
> Originally added in r0-44646-g204250d2fcd084 and r0-44627-gfd350d241fecf6 whic
> moved -fno-common from all builds to just checking builds.
>
> Since r10-4867-g6271dd984d7f92, GCC defaults to -fno-common. There's no need
> to pass it specially f
Pushed to r15-2167.
在 2024/7/13 下午5:04, Lulu Cheng 写道:
gcc/ChangeLog:
* config/loongarch/loongarch-protos.h
(loongarch_split_128bit_move): Delete.
(loongarch_split_128bit_move_p): Delete.
(loongarch_split_256bit_move): Delete.
(loongarch_split_256bit_move
Some distributions like Gentoo make -Wformat and -Wformat-security
enabled by default. Pass -Wno-format to the test to avoid a spurious
fail in such environments.
gcc/testsuite/
PR rtl-optimization/115929
* gcc.dg/torture/pr115929-1.c: Pass -Wno-format.
---
Richard, is this OK? I c
On Sat, 2024-07-20 at 06:52 +0100, Sam James wrote:
> Some distributions like Gentoo make -Wformat and -Wformat-security
> enabled by default. Pass -Wno-format to the test to avoid a spurious
> fail in such environments.
>
> gcc/testsuite/
> PR rtl-optimization/115929
> * gcc.dg/tortur
> Am 20.07.2024 um 02:31 schrieb Andrew Pinski :
>
> On Fri, Jul 19, 2024 at 5:23 PM Sam James wrote:
>>
>> Originally added in r0-44646-g204250d2fcd084 and r0-44627-gfd350d241fecf6
>> whic
>> moved -fno-common from all builds to just checking builds.
>>
>> Since r10-4867-g6271dd984d7f92,
Xi Ruoyao writes:
> On Sat, 2024-07-20 at 06:52 +0100, Sam James wrote:
>> Some distributions like Gentoo make -Wformat and -Wformat-security
>> enabled by default. Pass -Wno-format to the test to avoid a spurious
>> fail in such environments.
>>
>> gcc/testsuite/
>> PR rtl-optimization/115
I'm pinging this early since I would like to make sure that it gets into
14.2 RC which is about to be done on Tuesday 23rd July.
On Tue, Jul 16, 2024 at 04:50:29PM +0200, Stefan Schulze Frielinghaus wrote:
> Code attribute bhfgq is missing a mapping for TF. This results in
> unresolved iterators
74 matches
Mail list logo
