On Thu, Jun 4, 2020 at 2:43 PM Richard Biener wrote:
>
> On Thu, 4 Jun 2020, Hongtao Liu wrote:
>
> > Hi Richard:
> > Could you help review this patch.
> > uros said he wouldn't review patches related to x86 vector ISA anymore.
>
> I can't spot anything wrong but I know nothing about what a
>
On Thu, Jun 4, 2020 at 8:09 AM Hongtao Liu via Gcc-patches
wrote:
>
> This patch to is fix uppercase of mode in trunc2, it
> should be lowercase for standard pattern name.
>
> Bootstrap is ok, regression test on i386/x86-64 backend is ok.
OK.
Richard.
> gcc/ChangeLog:
> * config/i386/
On Thu, Jun 4, 2020 at 5:34 AM Jiufu Guo wrote:
>
> Jiufu Guo writes:
>
> Hi,
>
> Patch is updated a little according to comments.
> Please see if this is ok to commit.
OK with a proper ChangeLog after bootstrap / testing.
It's also OK to backport this to the GCC 10 branch if powerpc
folks want
On Thu, 4 Jun 2020, Hao Liu OS wrote:
> Hi Jakub,
>
> I've updated the incorrect ChangLog.
>
> gcc/:
>
> PR tree-optimization/89430
> * tree-ssa-phiopt.c
> (struct name_to_bb): Rename to ref_to_bb; add a new field exp;
> remove ssa_name_ver, store, offset fields
Hi!
The usual way to write it would be:
On Thu, Jun 04, 2020 at 06:17:13AM +, Hao Liu OS wrote:
> gcc/:
>
> PR tree-optimization/89430
> * tree-ssa-phiopt.c
> (struct name_to_bb): Rename to ref_to_bb; add a new field exp;
> remove ssa_name_ver, store, offset fi
On Thu, 4 Jun 2020, Hongtao Liu wrote:
> Hi Richard:
> Could you help review this patch.
> uros said he wouldn't review patches related to x86 vector ISA anymore.
I can't spot anything wrong but I know nothing about what a
define_subst is. You say "Bootstrap is ok, regression test on i386/x8
Hi Jakub,
I've updated the incorrect ChangLog.
gcc/:
PR tree-optimization/89430
* tree-ssa-phiopt.c
(struct name_to_bb): Rename to ref_to_bb; add a new field exp;
remove ssa_name_ver, store, offset fields.
(struct ssa_names_hasher): Rename to refs_hasher;
This patch to is fix uppercase of mode in trunc2, it
should be lowercase for standard pattern name.
Bootstrap is ok, regression test on i386/x86-64 backend is ok.
gcc/ChangeLog:
* config/i386/sse.md (pmov_dst_3_lower): New mode attribute.
(trunc2): Refine from
trunc2.
g
On Thu, Jun 04, 2020 at 04:47:43AM +, Hao Liu OS wrote:
> The patch is refactored a little according to the last comment. Do you have
> more comments? If no, I will commit it later.
>
> Tested on X86_64 and AArch64.
>
> gcc/:
>
> PR tree-optimization/89430
> * tree-ssa-phiop
Hi All,
The patch is refactored a little according to the last comment. Do you have
more comments? If no, I will commit it later.
Tested on X86_64 and AArch64.
gcc/:
PR tree-optimization/89430
* tree-ssa-phiopt.c (cond_store_replacement): Extend non-trap checking
to sup
Hello,
I accidentally wrote 'free(loc)' instead of 'free (loc)'. Please see the
fixed patch attached below (contrib/check_GNU_style.sh says it's OK
now):
From 0445fba96ee9030feb00ebec893f8dfed153b12d Mon Sep 17 00:00:00 2001
From: Asher Gordon
Date: Wed, 3 Jun 2020 17:20:08 -0400
Subject: [PATCH]
guojiufu writes:
> From: Jiufu Guo
>
> --- a/gcc/config/rs6000/rs6000.c
> +++ b/gcc/config/rs6000/rs6000.c
> @@ -4567,7 +4567,12 @@ rs6000_option_override_internal (bool global_init_p)
> unroll_only_small_loops = 0;
> if (!global_options_set.x_flag_rename_registers)
>
Jiufu Guo writes:
Hi,
Patch is updated a little according to comments.
Please see if this is ok to commit.
diff --git a/gcc/common.opt b/gcc/common.opt
index 4464049fc1f..570e2aa53c8 100644
--- a/gcc/common.opt
+++ b/gcc/common.opt
@@ -2856,6 +2856,10 @@ funroll-all-loops
Common Report Var(fla
Hi Richard:
Could you help review this patch.
uros said he wouldn't review patches related to x86 vector ISA anymore.
On Wed, Jun 3, 2020 at 10:26 AM Hongtao Liu wrote:
>
> Hi:
> When dest is memory, zero-masking is not valid, only merging-masking
> is available,
>
> Bootstrap is ok, regr
Bootstrapped and regtested on x86_64-redhat-linux, ppc64le-redhat-linux
and s390x-redhat-linux.
Linux Kernel (specifically, drivers/leds/trigger/ledtrig-cpu.c) build
with GCC 10 fails on s390 with "impossible constraint".
The problem is that jump threading makes __builtin_constant_p lie when
an
On Thu, 4 Jun 2020 at 03:05, Ville Voutilainen
wrote:
> > > "noexcept" is a red herring, what matters is defaulted vs user-provided.
> > > In one case, we end up zero-initializing the whole buffer, and not in the
> > > other.
> >
> > Yes, I just came to that conclusion. This is value-init, so the
On Thu, 4 Jun 2020 at 02:20, Ville Voutilainen
wrote:
>
> On Thu, 4 Jun 2020 at 02:13, Marc Glisse wrote:
> >
> > On Thu, 4 Jun 2020, Ville Voutilainen wrote:
> >
> > > On Thu, 4 Jun 2020 at 01:52, Marc Glisse wrote:
> > >>
> > >> Hello,
> > >>
> > >> is there any drawback to the attached patch?
On Wed, 3 Jun 2020, Martin Liška wrote:
> Sorry for the breakage. Can you please paste full build output for the
> problematic .o file?
>
> I bet it's a C file compilation, where we should use:
>
> __sync_val_compare_and_swap (counter, 0, (intptr_t)node);
>
> Can you please test it?
c++ -std=c
Hi!
On Mon, Jun 01, 2020 at 09:14:47AM -0700, Carl Love wrote:
> * config/rs6000/altivec.h: Add define for vec_replace_elt and
> vec_replace_unaligned.
* config/rs6000/altivec.h (vec_replace_elt, vec_replace_unaligned):
New defines.
> * config/rs6000/vsx.md
On Thu, 4 Jun 2020 at 02:13, Marc Glisse wrote:
>
> On Thu, 4 Jun 2020, Ville Voutilainen wrote:
>
> > On Thu, 4 Jun 2020 at 01:52, Marc Glisse wrote:
> >>
> >> Hello,
> >>
> >> is there any drawback to the attached patch? It changes the code generated
> >> for
> >
> > I don't get it. The noexce
On Mon, 1 Jun 2020, Mark Wielaard wrote:
> On Sun, May 24, 2020 at 02:30:13AM +0200, Mark Wielaard wrote:
> > This adds a flag to c_parser so we know when we were trying to
> > construct a string literal. If there is a parse error and we were
> > constructing a string literal, and the next token i
Hi!
On Mon, Jun 01, 2020 at 09:14:40AM -0700, Carl Love wrote:
> * config/rs6000/altivec.h: Add define vec_insertl, vec_inserth.
* config/rs6000/altivec.h (vec_insertl, vec_inserth): New defines.
> * config/rs6000/rs6000-builtin.def (BU_FUTURE_V_3): Add definition for
>
On Sun, 31 May 2020, Mark Wielaard wrote:
> On Sun, May 24, 2020 at 11:46:34PM +0200, Mark Wielaard wrote:
> > The C++ parser already tracks function call parens matching, but the C
> > parser doesn't. This adds the same functionality to the C parser and adds
> > a testcase showing the C++ and C p
On Thu, 4 Jun 2020, Ville Voutilainen wrote:
On Thu, 4 Jun 2020 at 01:52, Marc Glisse wrote:
Hello,
is there any drawback to the attached patch? It changes the code generated for
I don't get it. The noexceptness of the defaulted default constructor
should be a computation
of the noexceptne
On Thu, 4 Jun 2020 at 01:52, Marc Glisse wrote:
>
> Hello,
>
> is there any drawback to the attached patch? It changes the code generated for
I don't get it. The noexceptness of the defaulted default constructor
should be a computation
of the noexceptness of the subobjects, and that should boil d
Asher Gordon writes:
> Also note that this patch does not implement a -Wno-designated-init
> flag, but that shouldn't be too hard to implement.
Sorry, I meant -Wno-universal-initializer, not -Wno-designated-init.
Asher
--
Vimes grunted. "Where there are policemen there's crime, sergeant,
rem
On Fri, 2020-01-31 at 15:50 -0300, Vitor Guidi wrote:
> I took the required steps. The patch is attached to this email, I hope
> I got it right this time. I did not forward the patch to gcc-patches
> the first time, sorry for the inconvenience.
>
> Thank you for your attention,
>
> Vitor.
>
> in
Hi carl,
On Mon, Jun 01, 2020 at 09:14:33AM -0700, Carl Love wrote:
> * config/rs6000/altivec.md: Move UNSPEC_EXTRACTL, UNSPEC_EXTRACTR
> declarations to gcc/config/rs6000/vsx.md.
> (define_expand): Move vextractl and vextractr to
> gcc/config/rs6000/vsx.md.
> (defi
Closes #95379.
gcc/ChangeLog:
* doc/extend.texi: Document { 0 } as a special case for the
designated_init attribute.
gcc/c/ChangeLog:
* c-typeck.c (struct location_list): New type.
(struct initializer_stack): Add positional_init_locs for
-Wdesignated-init
Hello,
is there any drawback to the attached patch? It changes the code generated for
std::optional>f(){return{};}
from
movq$0, (%rdi)
movq%rdi, %r8
leaq8(%rdi), %rdi
xorl%eax, %eax
movq$0, 4084(%rdi)
movq%r8, %rcx
On Mon, 2020-05-18 at 12:02 +0200, Andreas Schwab wrote:
> When TARGET_VTABLE_USES_DESCRIPTORS is defined then function pointers in
> the vtable are output by ASM_OUTPUT_FDESC. The only current user of
> this is ia64, but its implementation of ASM_OUTPUT_FDESC lacks a call to
> assemble_external.
This passed regstrap and was approved offline by Segher, posting
the final form (minus my debug code, oops).
The same problem also arises for plfs where prefixed_load_p()
doesn't recognize it so we get just lfs in the asm output
with an @pcrel address.
PR target/95347
* config/rs6
Hi Carl,
On Mon, Jun 01, 2020 at 09:14:25AM -0700, Carl Love wrote:
> Based on previous IBM internal reviews of the patch set, the desire is
> for all of the vector insert and extract support to be in vsx.md as
> there is a longer term plan to re-work this support for PPC.
Very long time ago, it
On Wed, 2020-06-03 at 21:56 +0200, Romain Naour wrote:
> Hi Jeff,
>
> Le 03/06/2020 à 20:33, Jeff Law a écrit :
> > On Thu, 2020-05-21 at 17:35 +0200, Romain Naour via Gcc-patches wrote:
> > > As reported by several Buildroot users [1][2][3], the gcc build
> > > may fail while running selftests ma
Hi Jeff,
Le 03/06/2020 à 20:33, Jeff Law a écrit :
> On Thu, 2020-05-21 at 17:35 +0200, Romain Naour via Gcc-patches wrote:
>> As reported by several Buildroot users [1][2][3], the gcc build
>> may fail while running selftests makefile target.
>>
>> The problem only occurs when ccache is used with
David Edelsohn wrote:
On Wed, Jun 3, 2020 at 3:14 PM Iain Sandoe wrote:
libstdc++-v3/
* Makefile.am (tmake_file): Build and install AIX-style FAT libraries.
* Makefile.in: Regenerate.
* configure.ac (tmake_file): Substitute.
* configure: Regenerate.
* configure
On Wed, Jun 3, 2020 at 3:14 PM Iain Sandoe wrote:
>
> Hi David,
>
> thanks for working on this!
>
> David Edelsohn wrote:
>
> > [I'll start by repeating what I wrote about a similar libgcc change to
> > provide background and context.]
> >
> > When AIX added 64 bit support, it implemented what Ap
On 6/2/20 5:45 PM, Patrick Palka wrote:
When checking that a constrained partial specialization is more
constrained than the primary template, we pass only the innermost level
of generic template arguments to strictly_subsumes. This leads to us
doing a nonsensical substitution from normalize_con
Hi David,
thanks for working on this!
David Edelsohn wrote:
[I'll start by repeating what I wrote about a similar libgcc change to
provide background and context.]
When AIX added 64 bit support, it implemented what Apple MacOS Darwin
calls "FAT" libraries for its equivalent functionality --
On 6/2/20 3:43 PM, Patrick Palka wrote:
When determining the most specialized partial specialization of a
primary template that is nested inside a class template, we first
tsubst the outer template arguments into the TEMPLATE_DECL of each
partial specialization, and then check for satisfaction of
On Wed, Jun 03, 2020 at 02:09:11PM -0500, Segher Boessenkool wrote:
> Yeah, but that is specific to the vectoriser, not something that the
> RTL expander should have to deal with. A big impedance mismatch there.
It isn't specific to vectorizer, many named patterns aren't allowed to FAIL
and many
Hi,
it turns out that half of the global decl stream of cc1 LTO build consits
TREE_LISTS, identifiers and integer cosntats representing TYPE_VALUES of enums.
Those are streamed only to produce ODR warning and used otherwise, so this
patch moves the info to a separate section that is represented and
On Wed, Jun 03, 2020 at 08:38:04PM +0200, Richard Biener wrote:
> On June 3, 2020 8:23:14 PM GMT+02:00, Segher Boessenkool
> wrote:
> >On Wed, Jun 03, 2020 at 07:23:47PM +0200, Richard Biener wrote:
> >> >> mask = vec_cmp of the comparison
> >> >> true_masked = true_op & mask;
> >> >> false
On 6/1/20 6:50 PM, Jakub Jelinek wrote:
On Fri, May 29, 2020 at 01:26:32PM -0400, Jason Merrill via Gcc-patches wrote:
This is a diagnostic quality regression, moving the error message away from
the line where the actual problem is.
Maybe use error_at (loc, ...)?
That works fine, bootstrapped
On Wed, Jun 3, 2020 at 2:38 PM Richard Biener
wrote:
>
> On June 3, 2020 8:23:14 PM GMT+02:00, Segher Boessenkool
> wrote:
> >On Wed, Jun 03, 2020 at 07:23:47PM +0200, Richard Biener wrote:
> >> >> mask = vec_cmp of the comparison
> >> >> true_masked = true_op & mask;
> >> >> false_masked
On June 3, 2020 8:23:14 PM GMT+02:00, Segher Boessenkool
wrote:
>On Wed, Jun 03, 2020 at 07:23:47PM +0200, Richard Biener wrote:
>> >> mask = vec_cmp of the comparison
>> >> true_masked = true_op & mask;
>> >> false_masked = false_op & ~mask;
>> >> result = true_masked | false_masked;
>>
On Thu, 2020-05-21 at 17:35 +0200, Romain Naour via Gcc-patches wrote:
> As reported by several Buildroot users [1][2][3], the gcc build
> may fail while running selftests makefile target.
>
> The problem only occurs when ccache is used with gcc 9 and 10,
> probably due to a race condition.
>
> W
Hi Martin,
Okay, let's try this out. Okay for trunk. Thanks for the work!
On Tue, Jun 02, 2020 at 05:00:56PM +0200, Martin Liška wrote:
> >From 22db04d058c9bbd140041e7aa2caf1613767095a Mon Sep 17 00:00:00 2001
> From: Martin Liska
> Date: Tue, 2 Jun 2020 15:29:37 +0200
> Subject: [PATCH] rs600
On Wed, Jun 03, 2020 at 07:23:47PM +0200, Richard Biener wrote:
> >> mask = vec_cmp of the comparison
> >> true_masked = true_op & mask;
> >> false_masked = false_op & ~mask;
> >> result = true_masked | false_masked;
> >>
> >> but I believe this would be dead code never triggered.
> >
> >B
In a non-template, grokdeclarator notices when we build a pointer to a
variably-modified type, and inserts a TYPE_DECL to make sure that type gets
gimplified. But in a template we can't always recognize a variably-modified
type, so we need to deal with it at instantiation time.
Tested x86_64-pc-l
Hi Jose,
Proposed patch to PRs 66833, 67938 and 95214 ICE(s) on using assumed
rank character array in different situations.
Reviewed and committed with some trival changes:
It is better not to use STOP codes > 255, so I just
counted them up.
Some changes to the ChangeLog: Mentioned all PRs t
On June 3, 2020 7:01:39 PM GMT+02:00, Segher Boessenkool
wrote:
>Hi!
>
>On Wed, Jun 03, 2020 at 04:46:12PM +0200, Richard Biener wrote:
>> On Wed, Jun 3, 2020 at 4:17 PM David Edelsohn
>wrote:
>> > On Wed, Jun 3, 2020 at 9:41 AM Richard Sandiford
>> > wrote:
>> > > Well, it seems unfortunate to
Hi!
On Wed, Jun 03, 2020 at 04:46:12PM +0200, Richard Biener wrote:
> On Wed, Jun 3, 2020 at 4:17 PM David Edelsohn wrote:
> > On Wed, Jun 3, 2020 at 9:41 AM Richard Sandiford
> > wrote:
> > > Well, it seems unfortunate to have to do that.
> > >
> > > I think Martin's powerpc patch is the correc
David Malcolm writes:
> Thanks for the patch.
>
> Is this entrypoint only needed for the ahead-of-time use case? If the
> client code is instead going to do an in-memory compilation, then I
> believe it can simply build the buffer of data in memory and expose it
> to the jitted code via gcc_jit_
Hi Julian!
On 2020-06-03T14:36:14+0200, I wrote:
> On 2020-05-22T15:16:05-0700, Julian Brown wrote:
>> This patch adjusts the semantics of dynamic reference counts, as described
>> in the parent email.
>
> Thanks!
>
> A few questions, but no need to send an updated patch.
>
>> --- a/libgomp/oacc-
Successfully bootstrapped & regrtested on x86_64-pc-linux-gnu.
Pushed to master as r11-877-g6d36cc21b69d952967a4df6653954f572a101796.
gcc/jit/ChangeLog:
PR jit/95306
* jit-builtins.c (builtins_manager::make_primitive_type):
Implement BT_CONST_VOLATILE_PTR.
gcc/testsuite/Ch
On Wed, 2020-06-03 at 12:11 +0200, Andrea Corallo wrote:
> Hi all,
>
> I'd like to submit this patch to extend libgccjit to allow for
> storing
> binary blobs (equivalent to initialized char arrays).
>
> The use-case is when libgccjit is used as a back-end for dynamic
> programming languages. In
On Wed, Jun 3, 2020 at 4:17 PM David Edelsohn wrote:
>
> On Wed, Jun 3, 2020 at 9:41 AM Richard Sandiford
> wrote:
> >
> > Richard Biener writes:
> > > On Tue, Jun 2, 2020 at 5:00 PM Martin Liška wrote:
> > >>
> > >> On 6/2/20 1:09 PM, Richard Biener wrote:
> > >> > So please be constructive.
On Wed, 3 Jun 2020, Feng Xue OS via Gcc-patches wrote:
Ah, looking at the PR, you decided to perform the operation as unsigned
because that has fewer NOP conversions, which, in that particular testcase
where the offsets are originally unsigned, means we simplify better. But I
would expect it to
On Wed, Jun 3, 2020 at 9:41 AM Richard Sandiford
wrote:
>
> Richard Biener writes:
> > On Tue, Jun 2, 2020 at 5:00 PM Martin Liška wrote:
> >>
> >> On 6/2/20 1:09 PM, Richard Biener wrote:
> >> > So please be constructive. Like, provide a testcase that ICEs
> >> > with the FAILs replaced by gcc
This fixes a place where I missed to use SLP_TREE_REPRESENTATIVE
after its introduction.
2020-06-03 Richard Biener
PR tree-optimization/95495
* tree-vect-slp.c (vect_slp_analyze_node_operations): Use
SLP_TREE_REPRESENTATIVE in the shift assertion.
* gcc.dg/vect
I am testing the following patch on x86_64-unknown-linux-gnu. Test
coverage for autovectorization of scatters seems rare.
Richard.
This makes sure to get a truth type for scatter masks even when they
are invariant.
2020-06-03 Richard Biener
PR tree-optimization/95487
* tr
Richard Biener writes:
> On Tue, Jun 2, 2020 at 5:00 PM Martin Liška wrote:
>>
>> On 6/2/20 1:09 PM, Richard Biener wrote:
>> > So please be constructive. Like, provide a testcase that ICEs
>> > with the FAILs replaced by gcc_unreachable (). Martin, may I suggest
>> > to do this replacement and
Hi,
On Thu, 14 May 2020 at 17:08, Christophe Lyon
wrote:
>
> On Fri, 1 May 2020 at 12:57, Kyrylo Tkachov wrote:
> >
> >
> >
> > > -Original Message-
> > > From: Christophe Lyon
> > > Sent: 30 April 2020 09:51
> > > To: Kyrylo Tkachov
> > > Cc: gcc-patches@gcc.gnu.org
> > > Subject: Re
On Wed, 3 Jun 2020, Richard Biener wrote:
>
> The following patch avoids keeping the inherited MEM_ATTRs when
> set_mem_attributes_minus_bitpos is called with a variable ARRAY_REF.
> The inherited ones may not reflect the correct offset.
>
> We try to preserve the MEM_EXPR from the base object e
Ping?
https://gcc.gnu.org/pipermail/gcc-patches/2020-May/545747.html
On Wed, 27 May 2020 at 13:52, Christophe Lyon
wrote:
>
> Ping?
>
> On Thu, 14 May 2020 at 16:57, Christophe Lyon
> wrote:
> >
> > The interrupt attribute does not guarantee that the FP registers are
> > saved, which can result
Hi Julian!
On 2020-05-22T15:16:05-0700, Julian Brown wrote:
> This patch adjusts the semantics of dynamic reference counts, as described
> in the parent email.
Thanks!
A few questions, but no need to send an updated patch.
> --- a/libgomp/oacc-mem.c
> +++ b/libgomp/oacc-mem.c
> @@ -1018,13 +1
Hi All!
Proposed patch to Bug 94022 - Array slices of assumed-size arrays.
Patch tested only on x86_64-pc-linux-gnu.
Make sure that when passing array sections of assumed-size arrays to
procedures expecting an assumed-rank array the upper bound of the last
dimension of the array section does
The following patch avoids keeping the inherited MEM_ATTRs when
set_mem_attributes_minus_bitpos is called with a variable ARRAY_REF.
The inherited ones may not reflect the correct offset.
We try to preserve the MEM_EXPR from the base object expansion
but we have to make sure to also preserve the
This changes the loop over all stmts to a loop over all costed stmts.
Bootstrapped / tested on x86_64-unknown-linux-gnu, applied.
2020-06-03 Richard Biener
* tree-vect-slp.c (vect_bb_vectorization_profitable_p): Loop over
the cost vector to unset the visited flag on stmts.
---
"Kewen.Lin" writes:
> on 2020/6/3 下午5:27, Richard Biener wrote:
>> On Wed, 3 Jun 2020, Kewen.Lin wrote:
>>
>>> on 2020/6/3 下午3:07, Richard Biener wrote:
On Wed, 3 Jun 2020, Kewen.Lin wrote:
> Hi Richi,
>
>
> snip ...
>
>>
>> I'd just mention there are other targets that
on 2020/6/3 下午5:27, Richard Biener wrote:
> On Wed, 3 Jun 2020, Kewen.Lin wrote:
>
>> on 2020/6/3 下午3:07, Richard Biener wrote:
>>> On Wed, 3 Jun 2020, Kewen.Lin wrote:
>>>
Hi Richi,
snip ...
>
> I'd just mention there are other targets that have the choice between
> the ab
Hi Richard,
Thanks for reviewing the patch and sharing your comments.
>> pattern should be guarded with !TYPE_OVERFLOW_SANITIZED
>> and/or !TYPE_OVERFLOW_TRAPS
Done.
>> can you add some cases with unsigned types and types
>> smaller than int (thus with integral promotion applied)?
Done.
Please
Hi all,
I'd like to submit this patch to extend libgccjit to allow for storing
binary blobs (equivalent to initialized char arrays).
The use-case is when libgccjit is used as a back-end for dynamic
programming languages. In this case is often necessary to store
serialized objects into the compil
Andrea Corallo writes:
> Andrea Corallo writes:
>
>> Hi all,
>>
>> I'd like to submit this patch introducing the following 64bit builtins
>> variants as FPCR and FPSR registers getter/setter:
>>
>> unsigned long long __builtin_aarch64_get_fpcr64 ()
>> void __builtin_aarch64_set_fpcr64 (unsigned
This patch fixes an issue whereby the compiler generates spurious
ineffective with_clause warnings under -gnatwr when the only use of the
with'ed package occurs in a use_type clause in the context area of the
body of the target package and such use_type clause has multiple
prefixes.
Tested on x86_
This changes the strategy for dealing with the limitations of the
Has_Private_View mechanism, which is responsible for updating the views
in instantiations of a type first declared as private; this is necessary
because the views of a type may be different at the declaration point of
a generic and a
This was there for big integers after the previous code changes but not
(yet) for big reals, now done.
Tested on x86_64-pc-linux-gnu, committed on trunk
2020-06-03 Arnaud Charlet
gcc/ada/
* libgnat/a-nbnbin.ads: Minor reformatting.
* libgnat/a-nbnbre.ads, libgnat/a-nbnbre.adb
This changes a few places in the System.Atomic_Operations packages to
using static expressions, which guarantees that contant folding is done
in the front-end instead of the code generator.
No functional changes.
Tested on x86_64-pc-linux-gnu, committed on trunk
2020-06-03 Eric Botcazou
gcc/
This further tweaks the expanded code generated by the front-end
for attributes returning Universal_Integer, in particular removes
hardcoded references to it and fixes a couple of type mismatches
in the process.
The only observable change is that 'Val is now expanded by the
front-end for integer t
This improves the expanded code generated by the front-end for the Val,
Pred and Succ attributes when they are applied to an enumeration type
with a non-standard but contiguous representation. There is no need to
call the Rep_To_Pos routine to generate the required constraint checks
in this case,
This implements the documented semantics of the pragma for operators,
that is to say, introduces an implicit unchecked conversion from the
integer value to type System.Address, thus making the pragma work
on 64-bit targets as well.
Tested on x86_64-pc-linux-gnu, committed on trunk
2020-06-03 Eri
Routines like Next_Index and Next_Formal are implemented both as
procedures and functions. The procedure variant seems meant to be used
when iterating, e.g.:
Next_Formal (Formal);
because it is more readable than the corresponding functions:
Formal := Next_Formal (Formal);
(and it is inli
We introduce new functions to detect a list of statements with no side
effects as well as a loop with no side effect so that we can then mark
them for removal.
Tested on x86_64-pc-linux-gnu, committed on trunk
2020-06-03 Arnaud Charlet
gcc/ada/
* sem_util.ads, sem_util.adb (Side_Effe
This is part of a larger project to add a tree checker in GNAT to check
that the GNAT tree out of expansion is properly formed before passing it
to the code generator.
Tested on x86_64-pc-linux-gnu, committed on trunk
2020-06-03 Arnaud Charlet
gcc/ada/
* frontend.adb (Frontend): Call
This gets rid of one more case where the front-end would create a
temporary in Universal_Integer type, which is unnecessary.
The reason is that Universal_Integer must be a type as large as the
largest supported integer type and, therefore, can be much larger than
what is really needed here.
No fu
SPARK_Mode aspect/pragma used to signal parts of the code in SPARK is
now allowed to be Off inside generic instantiations for parts of the
code that should not be considered in SPARK.
Tested on x86_64-pc-linux-gnu, committed on trunk
2020-06-03 Yannick Moy
gcc/ada/
* rtsfind.adb (Loa
This extends the constant folding done in the procedure to literals of
enumeration types. This is just a matter of calling Expr_Rep_Value in
lieu of Expr_Value on the operand, since the result is the bit pattern.
No functional changes.
Tested on x86_64-pc-linux-gnu, committed on trunk
2020-06-0
This reverts a series of 3 old changes made without real explanation to
the second main case of Check_Private_View, which appear to be largely
obsolete by now and have made it quite hard to follow, and also changes
it to using In_Open_Scopes as the other cases.
No functional changes.
Tested on x8
This patch fixes an issue whereby the compiler incorrectly omits static
and dynamic accessibility checks on explicitly aliased parameters within
functions.
Tested on x86_64-pc-linux-gnu, committed on trunk
2020-06-03 Justin Squirek
gcc/ada/
* libgnat/a-cborse.adb, libgnat/a-cihase.ad
This extends the exception made for attribute references in the code
generating range checks to the simple expressions containing a single
attribute reference, thus avoiding to create a temporary whose type is
Universal_Integer when the attribute returns Universal_Integer, which is
the common case.
Fix the comment about the form of a 'term'. Allow a sequence of elmts,
in {...}, but not nested {...}.
Tested on x86_64-pc-linux-gnu, committed on trunk
2020-06-03 Bob Duff
gcc/ada/
* libgnat/s-regexp.ads: Fix comment--- gcc/ada/libgnat/s-regexp.ads
+++ gcc/ada/libgnat/s-regexp.ads
@
This fixes a couple of places to using the standard idiom for choosing
an appropriately-sized integer type. No practical change since the old
and the new type are effectively identical.
Tested on x86_64-pc-linux-gnu, committed on trunk
2020-06-03 Eric Botcazou
gcc/ada/
* exp_attr.ad
The debugger doesn't allow stepping into a subprogram when the
subprogram has a class-wide precondition (or postcondition). The
compiler generates wrappers for such subprograms and creates a new
"class-wide clone" of the subprogram, and the Needs_Debug_Info flag
isn't set on the clone subprogram's
For expansion of a Loop_Entry attribute, a function body is generated in
association with the condition of a generated while loop, and the
entities within the function need to have their scopes reset to the
function Entity_Id, for proper handling of up-level-reference processing
in the GNAT-LLVM co
ASIS is now maintained on a separate branch, so newer compilers no
longer need to generate ASIS trees. First remove processing of -gnatt,
then in a second stage, we'll remove all associated code.
Tested on x86_64-pc-linux-gnu, committed on trunk
2020-06-03 Arnaud Charlet
gcc/ada/
* s
This patch removes completely the processing associated with -gnatt.
Tested on x86_64-pc-linux-gnu, committed on trunk
2020-06-03 Arnaud Charlet
gcc/ada/
* aspects.adb, aspects.ads, atree.adb, atree.ads, elists.adb,
elists.ads, fname.adb, fname.ads, gnat1drv.adb, lib.adb,
The current implementation is suboptimal for 32-bit or smaller types
because it does the computation in Universal_Integer and can generate
checks. That's unnecessary in the common case, i.e. for integer types
or enumeration types with contiguous representation.
No functional changes.
Tested on x
On Wed, 3 Jun 2020, Kewen.Lin wrote:
> on 2020/6/3 下午3:07, Richard Biener wrote:
> > On Wed, 3 Jun 2020, Kewen.Lin wrote:
> >
> >> Hi Richi,
> >>
> >> on 2020/6/2 下午7:38, Richard Biener wrote:
> >>> On Thu, 28 May 2020, Kewen.Lin wrote:
> >>>
> Hi,
>
> This is one repost and you ca
On 6/1/20 1:40 PM, Martin Liška wrote:
Adding Honza as Uros recommended him for a review.
@Honza asked me personally to remind him this patch ;)
Martin
1 - 100 of 111 matches
Mail list logo
