I pushed the attached patch.
I kept the operator names... too late, there were already operator names in
the stdx::simd implemenation anyway. ;)
- Matthias
On Monday, 22 May 2023 22:51:49 CEST Jonathan Wakely wrote:
> On Mon, 22 May 2023 at 21:27, Matthias Kretz wrote:
> > On Monday, 22 May 20
Hi Juzhe,
in general I find the revised structure quite logical and it is definitely
an improvement. Some abstraction are still a bit leaky but we can always
refactor "on the fly". Some comments on the general parts, skipping
over the later details.
> bool m_has_dest_p;
Why does a store not
From: Javier Miranda
The compiler crashes processing a compilation unit has limited-with
context clauses, and the profile of some dispatching primitive
references a type visible through a limited-with clause, and
the dispatching primitive has class-wide preconditions.
gcc/ada/
* sem_ch1
From: Piotr Trojanek
Spotted while reviewing a patch with a similar typo.
gcc/ada/
* libgnat/s-mmap.adb (Mapped_Region_Record): Fix typo in comment.
Tested on x86_64-pc-linux-gnu, committed on master.
---
gcc/ada/libgnat/s-mmap.adb | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-
From: Eric Botcazou
This just merges two conditional blocks depending on the same condition.
gcc/ada/
* frontend.adb (Frontend): Merge two conditional blocks and adjust.
Tested on x86_64-pc-linux-gnu, committed on master.
---
gcc/ada/frontend.adb | 21 -
1 file ch
From: Eric Botcazou
This is most notably the addition of addresses in Interfaces.C.Pointers and
System.Bitfield_Utils. There is also a change to System.Stream_Attributes,
which was representing a thin pointer as a record, which is not problematic
per se, but is in the end, because the expanded c
From: Ronan Desplanques
gcc/ada/
* sem_ch7.adb: Remove duplicate comment.
Tested on x86_64-pc-linux-gnu, committed on master.
---
gcc/ada/sem_ch7.adb | 4
1 file changed, 4 deletions(-)
diff --git a/gcc/ada/sem_ch7.adb b/gcc/ada/sem_ch7.adb
index 5021d0ee04f..2610a7bba4d 100644
From: Piotr Trojanek
Expressions with parentheses are notoriously problematic to
pretty-print. In GNATprove we had a post-processing fix for them,
but it is better to have this fix in the GNAT frontend repository
and apply it for CodePeer as well.
gcc/ada/
* pprint.adb (Expression_Image
From: Piotr Trojanek
This patch propagates the apparently cleanest solutions between various
variants of the runtime units for interrupt handler registration.
In particular, the unnecessary default expressions for list cells with
interrupt handler addresses are removed, the list is changed from
From: Arnaud Charlet
Improve -gnatyx to check additional complete conditions,
and introduce a new switch -gnatyz to check for unnecessary
parentheses according to operator precedence rules.
Enable -gnatyz as part of -gnatyg.
gcc/ada/
* par-ch5.adb, style.ads, styleg.adb, styleg.ads
From: Piotr Trojanek
The expression pretty-printer that is used for SPARK counterexamples was
essentially duplicating the logic of First_Node/Last_Node and
First_Sloc/Last_Sloc routines. Now it simply reuses those routines.
gcc/ada/
* errout.adb
(Paren_Required): New subsidiary
From: Piotr Trojanek
Add some defensive code to get pretty-printed CodePeer outputs for ACATS
back to shape. At least some of this code appears to be redundant and
perhaps unnecessary, but we can this up later.
Expression pretty-printer should not be called with
N_Defining_Identifier nodes at al
From: Piotr Trojanek
When looking for the end of an numeric literal we consumed '+' and '-'
characters, because they might appear in the exponent part. This was too
aggressive when they separated the number from the subsequent operand,
like in "123+456". Now we skip past numeric literals by stric
From: Eric Botcazou
The implementation of task attributes in the runtime defines an atomic clone
of System.Address, which is awkward for targets where addresses and pointers
have a specific representation, so this change replaces that with a pragma
Atomic_Components on the Attribute_Array type.
From: Ronan Desplanques
There are multiple possible interpretations of "LSP". For example,
"Language Server Protocol". This patch clarifies that the
occurrences of "LSP" in GNAT's source code refer to the Liskov
Substitution Principle.
gcc/ada/
* einfo.ads: Mention full name of LSP.
Te
From: Piotr Trojanek
Warning about check failing at run time is likely spurious for mutually
recursive subprograms with multiple variant clauses. These will be
non-trivial to detect, so we simply suppress the warning altogether for
all subprogram variants.
gcc/ada/
* exp_prag.adb (Expan
From: Javier Miranda
The compiler reports spurious errors processing the class-wide
preconditions of a dispatching primitive of a private type T, when
the class-wide precondition invokes another dispatching primitive
of T that has the same name as a record component of T.
gcc/ada/
* sem
From: Raphael Amiard
gcc/ada/
* scng.adb (Scan): Replace occurrences of All_Extensions_Allowed
by Core_Extensions_Allowed.
Tested on x86_64-pc-linux-gnu, committed on master.
---
gcc/ada/scng.adb | 12 ++--
1 file changed, 6 insertions(+), 6 deletions(-)
diff --git a/
From: Eric Botcazou
The problem is that the special function created by the compiler to check
the predicate does not inherit the public status of the type, because it
is generated as part of the freezing of the quantified expression, which
occurs from within a couple of intermediate internal scop
From: Piotr Trojanek
When expanding of inequality operators for GNAT we were adding extra
parens, supposedly to "fix Sprint output" (source print); for GNATprove
we didn't, as these extra parens were leading to wrong columns for check
messages.
Adding these extra parens couldn't be a right, beca
From: Yannick Moy
Avoid spurious alarm by CodePeer analysis by adding default value
for a variable initialization.
gcc/ada/
* sem_util.adb (Check_Node): Add default init on local Id.
Tested on x86_64-pc-linux-gnu, committed on master.
---
gcc/ada/sem_util.adb | 3 ++-
1 file changed,
From: Piotr Trojanek
Some of the calls to Error_Msg_N controlled by the flag
Warn_On_Redundant_Constructs missed the "?r?" tag in their message
string. This caused a misleading "[enabled by default]" label to appear
next to the error message.
Spotted while adding a warning about duplicated choic
From: Piotr Trojanek
When detecting duplicate choices in exception handlers we had
inconsistent pairs of First/Next_Non_Pragma and First_Non_Pragma/Next.
This was harmless, because exception choices don't allow pragmas at all,
e.g.:
when Program_Error | Constraint_Error | ...; -- pragma not
From: Piotr Trojanek
Extra actual parameters for accessibility checks are confusing for the
expression pretty-printer that is used by CodePeer. It seems that nodes
created for the accessibility checks should use the Sloc of the source
expression of accessibility checks, not the target. However, t
From: Yannick Moy
Nightly runs of GNATprove fail on proof of the assertion following
the loop. Add a loop invariant to facilitate that proof.
gcc/ada/
* libgnat/i-c.adb (To_Ada): Add loop invariant.
Tested on x86_64-pc-linux-gnu, committed on master.
---
gcc/ada/libgnat/i-c.adb | 1 +
From: Gary Dismukes
The compiler blows up (such as with a Storage_Error or Assert_Failure)
on a call to a limited build-in-place function occurring in the return
for a function with a limited class-wide result. Such a function
should include extra formals for a task master and activation chain
(b
From: Eric Botcazou
It turns out that skipping compiler-generated block scopes is problematic
when computing the public status of a subprogram, because this subprogram
may end up being nested in the elaboration procedure of a package spec or
body, in which case it may not be public.
This replace
From: Vadim Godunko
Holder object is constant and protected from modification by tampering
rules.
gcc/ada/
* libgnat/a-coinho__shared.adb (Constant_Reference): Remove call
of Detach
(Query_Element): Likewise.
Tested on x86_64-pc-linux-gnu, committed on master.
---
gcc
From: Eric Botcazou
This happens when the limited record is initialized with a function call
because of a couple of issues: incorrect tree sharing when building the
predicate check and too late freezing for a compiler-generated subtype.
It turns out that building the predicate check manually is
From: Piotr Trojanek
Add new aspect Exceptional_Cases, which is intended for SPARK and
describes in which cases an exception will be raised, and optionally
supply a postcondition that shall be verified in this case.
The implementation is heavily modeled after Subprogram_Variant, which in
turn wa
From: Eric Botcazou
The resolution must be identical inside and outside the System hierarchy.
gcc/ada/
* sem_res.adb (Resolve_Intrinsic_Operator): Always perform the same
resolution for the special mod operator of System.Storage_Elements.
Tested on x86_64-pc-linux-gnu, committe
From: Piotr Trojanek
We pretty-print numeric literals that do not come from source by relying
on their Sloc. This generally works well, but sporadically the Sloc is
set wrongly. We might want to trace and fix such occurrences, but for
now it is simpler to replace an otherwise reasonable assertion
From: Eric Botcazou
Offset calculations should use the operator of System.Storage_Elements.
gcc/ada/
* libgnat/s-dwalin.adb (Enable_Cache): Use the subtract operator of
System.Storage_Elements to compute the offset.
(Symbolic_Address): Likewise.
Tested on x86_64-pc-linu
> OK.
Thanks!
> Can we handle non-integer bitfields by recursing with a temporary buffer to
> encode it byte-aligned and then apply shifting and masking to get it in
> place? Or is that not worth it?
Certainly doable, something along these lines is implemented in varasm.c to
output these non-in
From: Eric Botcazou
All the subprograms declared in the unit have convention Intrinsic and
their current implementation makes some implicit assumptions that are
not valid universally, so it is replaced by a direct expansion.
This is mostly straightforward because Resolve_Intrinsic_Operator alrea
From: Claire Dross
Introduce new ghost helper functions to facilitate proof.
gcc/ada/
* libgnat/s-valueu.adb (Scan_Raw_Unsigned): Use new helpers.
* libgnat/s-vauspe.ads (Raw_Unsigned_Starts_As_Based_Ghost,
Raw_Unsigned_Is_Based_Ghost): New ghost helper functions.
From: Ronan Desplanques
gcc/ada/
* sem_disp.adb: Fix reference to Ada issue in comment.
Tested on x86_64-pc-linux-gnu, committed on master.
---
gcc/ada/sem_disp.adb | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/gcc/ada/sem_disp.adb b/gcc/ada/sem_disp.adb
index b0
From: Eric Botcazou
The problem is that, unlike for protected subprograms, the expansion of
cleanups for protected entries is not delayed when they contain package
instances with a body, so the cleanups are generated twice and this may
yield two finalizers if the secondary stack is used in the en
From: Steve Baird
gcc/ada/
* sem_util.adb
(Is_Variable): Correctly return False for a selected component
name of the form Some_Object.Some_Discriminant, even if
Some_Object is a variable. We don't want to allow such a name as
an actual parameter in a call
From: Patrick Bernardi
gcc/ada/
* libgnat/s-stchop.adb (Stack_Check): Remove redundant parentheses.
Tested on x86_64-pc-linux-gnu, committed on master.
---
gcc/ada/libgnat/s-stchop.adb | 5 ++---
1 file changed, 2 insertions(+), 3 deletions(-)
diff --git a/gcc/ada/libgnat/s-stchop.ad
From: Eric Botcazou
This is most notably the addition of addresses in Expand_Interface_Thunk.
There is also a small change to Expand_Dispatching_Call, which was directly
accessing a class-wide interface object as a tag, thus giving rise later to
unchecked conversions between either the root or th
On 5/22/23 13:25, Costas Argyris wrote:
Currently on Windows, when CreateProcess is called with a command-line
that exceeds the 32k Windows limit, we get a very bad error:
"CreateProcess: No such file or directory"
This patch detects the case where this would happen and writes the
long command-
From: Eric Botcazou
This operator is special because the left operand is of Address type while
the right operand and the result are of Storage_Offset type (but we raise
Constraint_Error if the right operand is not positive) and it needs to be
resolved to the type of the left operand to implement
On Tue, May 23, 2023 at 8:24 AM Martin Uecker wrote:
>
> Am Dienstag, dem 23.05.2023 um 08:13 +0200 schrieb Richard Biener:
> > On Mon, May 22, 2023 at 7:24 PM Martin Uecker via Gcc-patches
> > wrote:
> > >
> > >
> > >
> > > This version contains the middle-end changes for PR109450
> > > and test
Hi, Robin.
>> Why does a store not have a destination (as commented below)?
OK, V2 patch will have more comments.
>> m_all_unmasked_p or m_fully_unmasked_p?
OK.
>> Apart from the insn-centric name, couldn't we also decide this
>> based on the context later? In the vector-builtins.cc we have
>>
> ARM uses rtx operands[] in many places and I personally prefer this way since
> it will make codes much cleaner.
> I dislike the way making the function argument with multiple operand ,like
> this:
> void func(rtx dest, rtx src1, rtx src2, )
> If we are doing this, we will need to add helper
From: Juzhe-Zhong
This patch is to refactor the framework of RVV auto-vectorization.
Since we find out are keep adding helpers && wrappers when implementing
auto-vectorization.
It will make the RVV auto-vectorizaiton very messy.
After double check my downstream RVV GCC, assemble all auto-vector
Yeah. I know.
Like ARM does everywhere:
(define_expand "vcond"
[(set (match_operand:SVE_ALL 0 "register_operand")
(if_then_else:SVE_ALL
(match_operator 3 "comparison_operator"
[(match_operand:SVE_I 4 "register_operand")
(match_operand:SVE_I 5 "nonmemory_operand")])
(match_
Am Dienstag, dem 23.05.2023 um 10:18 +0200 schrieb Richard Biener:
> On Tue, May 23, 2023 at 8:24 AM Martin Uecker
> wrote:
> >
> > Am Dienstag, dem 23.05.2023 um 08:13 +0200 schrieb Richard Biener:
> > > On Mon, May 22, 2023 at 7:24 PM Martin Uecker via Gcc-patches
> > > wrote:
> > > >
> > > >
On 10.01.23 16:38, Sebastian Huber wrote:
On 19/12/2022 17:02, Sebastian Huber wrote:
Build libatomic for all targets. Use gthr.h to provide a default
implementation. If the thread model is "single", then this
implementation will
not work if for example atomic operations are used for thread/
The gcc.dg/analyzer/data-model-4.c and
gcc.dg/analyzer/torture/conftest-1.c fail with recent glibc headers
and succeed with older headers.
The new error message is:
warning: use of possibly-NULL 'f' where non-null expected [CWE-690]
[-Wanalyzer-possible-null-argument]
Like similar previous fixes
Hi Juzhe,
thanks, IMHO it's clearer with the changes now. There are still
things that could be improved but it is surely an improvement over
what we currently have. Therefore I'd vote to go ahead so we can
continue with more expanders and changes.
Still, we should be prepared for more refactori
Oh， Thanks.
Let's wait for Kito's final approved.
juzhe.zh...@rivai.ai
From: Robin Dapp
Date: 2023-05-23 17:44
To: juzhe.zhong; gcc-patches
CC: rdapp.gcc; kito.cheng; kito.cheng; palmer; palmer; jeffreyalaw
Subject: Re: [PATCH V2] RISC-V: Refactor the framework of RVV auto-vectorization
Hi Juz
The following fixes code hoisting to properly consider ANTIC_OUT instead
of ANTIC_IN. That's a bit expensive to re-compute but since we no
longer iterate we're doing this only once per BB which should be
acceptable. This avoids missing hoistings to the end of blocks where
something in the block c
Lgtm, we can always improve later, I am not intend to block things too :)
juzhe.zh...@rivai.ai 於 2023年5月23日 週二 17:46 寫道：
> Oh， Thanks.
> Let's wait for Kito's final approved.
>
>
>
> juzhe.zh...@rivai.ai
>
> From: Robin Dapp
> Date: 2023-05-23 17:44
> To: juzhe.zhong; gcc-patches
> CC: rdapp.gcc
Hi all,
In this PR we ICE because the substituted pattern for mla "lost" its predicate
and constraint for operand 0
because the define_subst template:
[(set (match_operand: 0)
(vec_concat:
(match_dup 1)
(match_operand:VDZ 2 "aarch64_simd_or_scalar_imm_zero")))])
Uses
From: Juzhe-Zhong
I just notice the warning:
../../../riscv-gcc/gcc/config/riscv/vector.md:618:1: warning: source missing a
mode?
gcc/ChangeLog:
* config/riscv/vector.md: Add mode.
---
gcc/config/riscv/vector.md | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/gcc/
Committed, thanks Kito.
Pan
-Original Message-
From: Gcc-patches On Behalf
Of Kito Cheng via Gcc-patches
Sent: Tuesday, May 23, 2023 5:57 PM
To: 钟居哲
Cc: Robin Dapp ; gcc-patches ;
Kito.cheng ; palmer ; palmer
; jeffreyalaw
Subject: Re: Re: [PATCH V2] RISC-V: Refactor the framework o
At -O2, and so with SLP vectorisation enabled:
struct complx_t { float re, im; };
complx_t add(complx_t a, complx_t b) {
return {a.re + b.re, a.im + b.im};
}
generates:
fmovw3, s1
fmovx0, d0
fmovx1, d2
fmovw2, s3
bfi x
In a follow-up patch, I wanted to use an int iterator to iterate
over various possible values of a const_int. But one problem
with int iterators was that there was no way of referring to the
current value of the iterator. This is unlike modes and codes,
which provide automatic "mode", "MODE", "co
On Tue, May 23, 2023 at 12:38 PM Richard Sandiford via Gcc-patches
wrote:
>
> At -O2, and so with SLP vectorisation enabled:
>
> struct complx_t { float re, im; };
> complx_t add(complx_t a, complx_t b) {
> return {a.re + b.re, a.im + b.im};
> }
>
> generates:
>
> fmov
This dumps ANTIC_OUT before pruning clobbered mems from it as part
of the ANTIC_IN compute.
Bootstrapped and tested on x86_64-unknown-linux-gnu, pushed.
* tree-ssa-pre.cc (compute_antic_aux): Dump the correct
ANTIC_OUT.
---
gcc/tree-ssa-pre.cc | 7 ---
1 file changed, 4 inser
Richard Biener writes:
> On Tue, May 23, 2023 at 12:38 PM Richard Sandiford via Gcc-patches
> wrote:
>>
>> At -O2, and so with SLP vectorisation enabled:
>>
>> struct complx_t { float re, im; };
>> complx_t add(complx_t a, complx_t b) {
>> return {a.re + b.re, a.im + b.im};
>> }
Hi Alex,
> -Original Message-
> From: Alex Coplan
> Sent: Thursday, May 11, 2023 12:15 PM
> To: gcc-patches@gcc.gnu.org
> Cc: ni...@redhat.com; Richard Earnshaw ;
> Ramana Radhakrishnan ; Kyrylo Tkachov
>
> Subject: [PATCH] arm: Fix ICE due to infinite splitting [PR109800]
>
> Hi,
>
>
On Tue, 23 May 2023, Richard Biener wrote:
> > Ah, no, I deliberately decided against that, because that way we would go
> > via gimplify_arg, which would emit all side effects in *pre_p. That seems
> > wrong if arguments had side-effects that should go in *post_p.
>
> Ah, true - that warrants a
"juzhe.zh...@rivai.ai" writes:
> Yeah. I know.
> Like ARM does everywhere:
> (define_expand "vcond"
> [(set (match_operand:SVE_ALL 0 "register_operand")
> (if_then_else:SVE_ALL
> (match_operator 3 "comparison_operator"
> [(match_operand:SVE_I 4 "register_operand")
>(match_op
Jeff Law via Gcc-patches writes:
> On 5/17/23 03:03, Jin Ma wrote:
>> For example:
>> (define_insn "mov_lowpart_sidi2"
>>[(set (match_operand:SI0 "register_operand" "=r")
>> (subreg:SI (match_operand:DI 1 "register_operand" " r") 0))]
>>"TARGET_64BIT"
>>"mov\t%0,%1
Bootstrap on X86 passed.
Ok for trunk?
Thanks.
juzhe.zh...@rivai.ai
From: juzhe.zhong
Date: 2023-05-22 16:38
To: gcc-patches
CC: richard.sandiford; rguenther; Ju-Zhe Zhong
Subject: [PATCH V12] VECT: Add decrement IV iteration loop control by variable
amount support
From: Ju-Zhe Zhong
gcc/C
PR target/104327 not only affects s390 but also avr:
The avr backend pre-sets some options depending on optimization level.
The inliner then thinks that always_inline functions are not eligible
for inlining and terminates with an error.
Proposing the following patch that implements TARGET_CAN_INL
[Andrew, do you remotely remember what if anything this did? It came
from a wholesale merge from our long forgotten branch, so there's no
history on the specifics of it. Not important, I'm just curious. It
was probably me high on something.]
This patch removes a buggy special case in irange::in
BTW, we should probably backport this to god knows how many branches.
Aldy
On Tue, May 23, 2023 at 2:58 PM Aldy Hernandez wrote:
>
> [Andrew, do you remotely remember what if anything this did? It came
> from a wholesale merge from our long forgotten branch, so there's no
> history on the speci
On 5/23/23 04:09, juzhe.zh...@rivai.ai wrote:
From: Juzhe-Zhong
I just notice the warning:
../../../riscv-gcc/gcc/config/riscv/vector.md:618:1: warning: source missing a
mode?
gcc/ChangeLog:
* config/riscv/vector.md: Add mode.
While I'm a big fan of the gen* warnings, I do wish t
From: Juzhe-Zhong
This patch enable RVV auto-vectorization including floating-point
unorder and order comparison.
The testcases are leveraged from Richard.
So include Richard as co-author.
Co-Authored-By: Richard Sandiford
gcc/ChangeLog:
* config/riscv/autovec.md (@vcond_mask_): New
> +(define_expand "vec_cmp"
> + [(set (match_operand: 0 "register_operand")
> + (match_operator: 1 "comparison_operator"
> + [(match_operand:VI 2 "register_operand")
> +(match_operand:VI 3 "register_operand")]))]
> + "TARGET_VECTOR"
> + {
> +riscv_vector::expand_vec_cmp (op
Committed, thanks Jeff.
Pan
-Original Message-
From: Gcc-patches On Behalf
Of Jeff Law via Gcc-patches
Sent: Tuesday, May 23, 2023 9:43 PM
To: juzhe.zh...@rivai.ai; gcc-patches@gcc.gnu.org
Cc: kito.ch...@gmail.com; kito.ch...@sifive.com; pal...@dabbelt.com;
pal...@rivosinc.com; rdapp..
> > + if (reject_unsigned && TYPE_UNSIGNED (abs_type))
> > +return false;
> > + if (!ANY_INTEGRAL_TYPE_P (abs_type) || TYPE_OVERFLOW_WRAPS (abs_type))
> > +return false;
>
> Could you explain the reject_unsigned behaviour? I'd have expected
> TYPE_OVERFLOW_WRAPS (abs_type) to reject the
Glibc defines int32_t as 'int' while newlib defines it as 'long int'.
Although these correspond to the same size, g++ complains when using the
On 23/05/2023 15:41, Christophe Lyon wrote:
Glibc defines int32_t as 'int' while newlib defines it as 'long int'.
Although these correspond to the same size, g++ complains when using the
From: oluade01
This adds a recognition pattern for the non-widening
absolute difference (ABD).
gcc/ChangeLog:
* doc/md.texi (sabd, uabd): Document them.
* internal-fn.def (ABD): Use new optab.
* optabs.def (sabd_optab, uabd_optab): New optabs,
* tree-vect-pattern
From: Juzhe-Zhong
This patch enable RVV auto-vectorization including floating-point
unorder and order comparison.
The testcases are leveraged from Richard.
So include Richard as co-author.
Co-Authored-By: Richard Sandiford
gcc/ChangeLog:
* config/riscv/autovec.md (@vcond_mask_): New
Hi, Robin.
>> Don't you want to use your shiny new operand passing style here as
>> with the other expanders?
H, I do this just following ARM code style.
You can see I do pass rtx[] for expand_vcond and pass rtx,rtx,rtx for
expand_vec_cmp.
Well, I just follow ARM SVE implementation (You can c
This is a small adjustment to the work done for PR108752 and
better reflects the cost of the generated sequence.
Bootstrapped and tested on x86_64-unknown-linux-gnu, pushed.
PR tree-optimization/108752
* tree-vect-stmts.cc (vectorizable_operation): For bit
operations with
The x86 backend looks at the SLP node passed to the add_stmt_cost
hook when costing vec_construct, looking for elements that require
a move from a GPR to a vector register and cost that. But since
vect_prologue_cost_for_slp decomposes the cost for an external
SLP node into individual pieces this c
The following also accounts for a GPR->XMM move cost for splat
operations and properly guards eliding the cost when moving from
memory only for SSE4.1 or HImode or larger operands. This
doesn't fix the PR fully yet.
Bootstrapped and tested on x86_64-unknown-linux-gnu, OK?
Thanks,
Richard.
On Tue, May 23, 2023 at 5:18 PM Richard Biener wrote:
>
> The following also accounts for a GPR->XMM move cost for splat
> operations and properly guards eliding the cost when moving from
> memory only for SSE4.1 or HImode or larger operands. This
> doesn't fix the PR fully yet.
>
> Bootstrapped
Add V8QImode and V4QImode vector shift patterns that call into
ix86_expand_vecop_qihi_partial. Generate special sequences
for constant count operands.
The patch regresses g++.dg/pr91838.C - as explained in PR91838, the
test returns different results, depending on whether V8QImode shift
pattern is
Richard Biener writes:
> The x86 backend looks at the SLP node passed to the add_stmt_cost
> hook when costing vec_construct, looking for elements that require
> a move from a GPR to a vector register and cost that. But since
> vect_prologue_cost_for_slp decomposes the cost for an external
> SLP
Applied this patchlet that implements proper cost computation of
(set (zero_extract (...) ...))
kind patterns that do single-bit (inverted) bit insertions.
Johann
--
Improve cost computation for single-bit bit insertions.
Some miscomputation of rtx_costs lead to sub-optimal code for
single-
On 5/22/23 17:58, Jan-Benedict Glaw wrote:
Hi!
One of the supplied argument strings is unneccesarily long (c-sky, using
basically the same code, fixed it to a shorter length) and this fixes overflow
warnings, as GCC fails to deduce that the full 256 bytes for load_op[] are
not used at all.
PR middle-end/109840 is a regression introduced by my recent patch to
fold popcount(bswap(x)) as popcount(x). When the bswap and the popcount
have the same precision, everything works fine, but this optimization also
allowed a zero-extension between the two. The oversight is that we need
to be s
Since the type_traits header is a C++11 header file, using can be used instead
of typedef. This patch provides more readability, especially for long type
names.
libstdc++-v3/ChangeLog:
* include/std/type_traits: Use using instead of typedef
---
libstdc++-v3/include/std/type_traits | 158
In the case where the target supports extension instructions,
it is preferable to use that instead of doing the same in other ways.
For the following case
void foo (unsigned long a, unsigned long* ptr) {
ptr[0] = a & 0xUL;
ptr[1] &= 0xUL;
}
GCC generates
foo:
li
On Mon, May 22, 2023 at 10:48 PM Takayuki 'January June' Suwa
wrote:
>
> On 2023/05/23 11:27, Max Filippov wrote:
> > Hi Suwa-san,
>
> Hi!
>
> > This change introduces a bunch of test failures on big endian configuration.
> > I believe that's because the starting bit position for zero_extract is
On Mon, May 22, 2023 at 12:06 AM Takayuki 'January June' Suwa
wrote:
>
> By making use of the 'addsub_operator' added in the last patch.
>
> gcc/ChangeLog:
>
> * config/xtensa/xtensa.md (*addsubx): Rename from '*addx',
> and change to also accept '*subx' pattern.
> (*subx):
>>> Don't you want to use your shiny new operand passing style here as
>>> with the other expanders?
> H, I do this just following ARM code style.
> You can see I do pass rtx[] for expand_vcond and pass rtx,rtx,rtx for
> expand_vec_cmp.
> Well, I just follow ARM SVE implementation (You can che
Signed-off-by: Matthias Kretz
libstdc++-v3/ChangeLog:
PR libstdc++/109261
* include/experimental/bits/simd_neon.h (_S_reduce): Add
constexpr and make NEON implementation conditional on
not __builtin_is_constant_evaluated.
---
.../include/experimental/bits/simd_n
On 5/23/23 12:24 AM, Kewen.Lin wrote:
> on 2023/5/23 01:31, Carl Love wrote:
>> The builtins were requested for use in GLibC. As of version 2.31 they
>> were added as inline asm. They requested a builtin so the asm could be
>> removed.
>
> So IMHO we also want the similar support for mffscrn, tha
While trying to understand how to use the ! operand for match
patterns, I noticed that the debug dumps would print out applying
a pattern but nothing when it was rejected in the end. This was confusing
me.
This adds that by emitting a dump for the failed case.
Note the patch is little more complex
From: Juzhe-Zhong
This simple patch fixes the magic number, replaced by enum to make code
more reasonable.
Ok for trunk ?
gcc/ChangeLog:
* config/riscv/riscv-v.cc (expand_vec_series): Fix magic number.
(expand_const_vector): Ditto.
(legitimize_move): Ditto.
(exp
Ok. Let's wait for Kito's more comments.
Thanks.
juzhe.zh...@rivai.ai
From: Robin Dapp
Date: 2023-05-24 05:07
To: 钟居哲; gcc-patches
CC: rdapp.gcc; kito.cheng; kito.cheng; palmer; palmer; Jeff Law;
richard.sandiford
Subject: Re: [PATCH V2] RISC-V: Add RVV comparison autovectorization
>>> Don't y
1 - 100 of 149 matches
Mail list logo
