This is the first in a series of patches to fix the fallout from recent
combiner changes in how shift-add style insns are canonicalized.
This patch effectively just adds a new shift-add insn to the PA port.
The old shift-add insn stays for now, but will be removed in a follow-up
once I'm con
I noticed that -Wc++14-compat was warning about headers that had been
updated to include a declaration of a global sized operator delete.
This was intended to catch problematic placement deletes, but now I
think that C++14 headers that just don't bother to guard the declaration
with a C++14 #if
register_specialization complains about a specialization in a different
namespace from the main template; we shouldn't give that error about an
instantiation of a partial specialization. And we were about to
SET_DECL_IMPLICIT_INSTANTIATION anyway, so let's just do that before we
call register_
On 05/20/2015 03:18 PM, Jason Merrill wrote:
On 05/08/2015 09:14 PM, Aldy Hernandez wrote:
+ if (!flag_syntax_only)
+c_parse_final_cleanups ();
The condition is a significant change of behavior for the C++ front end;
doing final instantiation and such even with -fsyntax-only was a
deliber
Update for the c-family bits removing the flag_syntax_only part Jason
requested.
And BTW, ping for you C front-end maintainers (unless Jason is reviewing
the C bits, in which case the rest of you can sit back and look pretty).
Aldy
gcc/c-family/
* c-common.h (c_parse_final_cleanups): New pr
On 05/20/2015 05:01 PM, Jan Hubicka wrote:
commit 8824b5ecba26cef065e47b34609c72677c3c36fc
Author: Aldy Hernandez
Date: Wed May 20 16:31:14 2015 -0400
Set DECL_IGNORED_P on temporary arrays created in the switch
conversion pass.
diff --git a/gcc/tree-switch-conversion.c b/gcc/tree
As described in PR target/63810, this addresses several problems with
the validation and encoding of deployment target version strings for the
__ENVIRONMENT_MAC_OS_X_VERSION_MIN_REQUIRED__ macro. There are
currently four testcases exercising inputs to -mmacosx-version-min
(gcc/testsuite/gcc.dg/da
On May 20, 2015, Jason Merrill wrote:
> I want to explicitly pass -std=c++98 to the compiler used in building
> stage 1. Does this seem like the right way to do that?
It certainly works, but I'm a bit concerned about its overriding the
earlier choices of STAGE1_CXXFLAGS, that use CXXFLAGS or ST
Hi,
this patch extends statistics from tree-ssa-alias to also cover TBAA oracle.
This is useful to keep track of aliasing effectivity. For example the hack
in alias.c putting globbing all pointers to one costs about 20% of all
answers on firefox. I.e. from 15500978 disambiguations/23744267 querrie
On Wed, 20 May 2015, Aldy Hernandez wrote:
> Update for the c-family bits removing the flag_syntax_only part Jason
> requested.
>
> And BTW, ping for you C front-end maintainers (unless Jason is reviewing the C
> bits, in which case the rest of you can sit back and look pretty).
The C front-end
On 05/20/2015 06:16 PM, Joseph Myers wrote:
On Wed, 20 May 2015, Aldy Hernandez wrote:
Update for the c-family bits removing the flag_syntax_only part Jason
requested.
And BTW, ping for you C front-end maintainers (unless Jason is reviewing the C
bits, in which case the rest of you can sit bac
Perhaps I should've sent this to the java-patches list.
PING.
gcc/java/
* class.c: Remove pending_static_fields.
(add_field): Remove use of pending_static_fields.
(java_write_globals): Remove.
* decl.c (java_mark_class_local): Remove use of
pending_static_fields.
* java-tree.h: Remove pen
> From: "H.J. Lu"
> Date: Tue, 19 May 2015 21:20:46 +0200
> This patch affects your target. Can you verify if it is OK on
> your target?
If you mean philosophically, it's ok when the general bits and
direction are ok, but for the practical part of checking
e.g. for typos through compilation, pl
GCC bootstraps with this patch.
On Wed, May 20, 2015 at 12:54 PM, Ian Lance Taylor wrote:
> This is OK if GCC bootstraps.
>
> Thanks.
>
> Ian
>
> On Wed, May 20, 2015 at 11:25 AM, Yunlian Jiang wrote:
>> I have the following change to make libiberty compile with _GNU_SOURCE
>> defined
>> and re
Now that I know a bit more from Jakub & Yuri's comments, I don't think
we should be turning that flag on for all the targets in the testsuite.
I was primarily trying to avoid someone else having to go through the
same analysis and reach the same conclusion for another port. But I'm
less concern
On 05/20/2015 07:32 AM, Jakub Jelinek wrote:
> For OpenMP/OpenACC, there is still lots of diagnostics emitted during
> gimplification and at the start of the omp lowering phases, so for
> diagnostics purposes you can consider them as part of a common layer for all
> the 3 FEs.
Thanks, that makes
On Wed, May 20, 2015 at 3:53 PM, Hans-Peter Nilsson
wrote:
>> From: "H.J. Lu"
>> Date: Tue, 19 May 2015 21:20:46 +0200
>
>> This patch affects your target. Can you verify if it is OK on
>> your target?
>
> If you mean philosophically, it's ok when the general bits and
> direction are ok, but for
OK.
Jason
Older assemblers don't understand .abiversion, so I'm committing the
following as obvious to fix a problem Michael Meissner found when
building gcc for powerpc64-linux.
PR libgcc/66225
* config/rs6000/morestack.S: Remove ".abiversion 1".
Index: libgcc/config/rs6000/morestack.S
===
> From: gcc-patches-ow...@gcc.gnu.org [mailto:gcc-patches-
> ow...@gcc.gnu.org] On Behalf Of Thomas Preud'homme
> > From: Steven Bosscher [mailto:stevenb@gmail.com]
> > Sent: Tuesday, May 19, 2015 7:21 PM
> >
> > Not OK.
> > This will break in move_invariants() when it looks at REGNO (inv->reg)
From: Trevor Saunders
Hi,
Another week and more patches to remove conditional compilation ;-)
each patch individually bootstrapped + regtested on x86_64-unknown-linux-gnu,
and run through config-list.mk with a couple more patches I'll commit when I
finish writing ChangeLogs. Committing to trun
From: Trevor Saunders
gcc/c-family/ChangeLog:
2015-05-20 Trevor Saunders
* c-cppbuiltin.c (c_cpp_builtins): Use if instead of #if with
STACK_GROWS_DOWNWARD.
gcc/ChangeLog:
2015-05-20 Trevor Saunders
* *.c: Use if instead of preprocessor checks with
STACK
From: Trevor Saunders
gcc/ChangeLog:
2015-05-20 Trevor Saunders
* defaults.h: Add default for STACK_PUSH_CODE.
* expr.c: Don't redefine STACK_PUSH_CODE.
* recog.c: Likewise.
---
gcc/ChangeLog | 6 ++
gcc/defaults.h | 8
gcc/expr.c | 8
gcc/
From: Trevor Saunders
gcc/c-family/ChangeLog:
2015-05-20 Trevor Saunders
* c-cppbuiltin.c (c_cpp_builtins): Check the value of
STACK_GROWS_DOWNWARD rather than if it is defined.
gcc/ChangeLog:
2015-05-20 Trevor Saunders
* *.c: Check the value of STACK_GROWS_DOWN
From: Trevor Saunders
gcc/ChangeLog:
2015-05-20 Trevor Saunders
* *.c: Remove comparison of ARG_FRAME_POINTER_REGNUM and
FRAME_POINTER_REGNUM with the preprocessor.
---
gcc/ChangeLog | 5 +
gcc/combine.c | 18 +++---
gcc/df-problems.c | 5 ++---
gcc
From: Trevor Saunders
gcc/ChangeLog:
2015-05-20 Trevor Saunders
* final.c (final_scan_insn): Don't check HAVE_peephole with the
preprocessor.
* output.h: Likewise.
* genconfig.c (main): Alwways define HAVE_peephole.
* genpeep.c: Don't emit checks of HA
From: Trevor Saunders
gcc/ChangeLog:
2015-05-20 Trevor Saunders
* genconfig.c (main): Always define HAVE_conditional_move.
* *.c: Don't check if HAVE_conditional_move is defined.
---
gcc/ChangeLog | 4
gcc/combine.c | 2 +-
gcc/expmed.c | 4 ++--
gcc
From: Trevor Saunders
gcc/ChangeLog:
2015-05-20 Trevor Saunders
* *.c, *.h: DOn't check HAVE_conditional_move with the preprocessor.
---
gcc/ChangeLog | 4
gcc/combine.c | 5 ++---
gcc/expmed.c | 13 +
gcc/expr.c| 10 ++
gcc/optabs.c | 30 +++
On targets such as ARM, some arches are compatible with options needed
to enable compilation with vectorization, but the specific hardware (or
simulator or BSP) available for execution tests may not implement or
enable those features. The vect.exp test harness already includes some
magic to de
ARM testing shares the AArch64 advsimd-intrinsics execution tests. On
ARM, though, the NEON support being tested is optional -- some arches
are compatible with the NEON compilation options but hardware available
for testing might or might not be able to execute those instructions.
In arm-none-
find_split_point will tend to favor splitting complex insns in such a
way as to encourage multiply-add insns. It does this by splitting an
unrecognizable insn at the (plus (mult)).
Now that many MULTs are canonicalized as ASHIFT, that code to prefer the
multiply-add is no longer triggering
On 05/20/2015 08:09 PM, tbsaunde+...@tbsaunde.org wrote:
From: Trevor Saunders
gcc/ChangeLog:
2015-05-20 Trevor Saunders
* final.c (final_scan_insn): Don't check HAVE_peephole with the
preprocessor.
* output.h: Likewise.
* genconfig.c (main): Alwways define
This is another patch aimed at fixing bugs relating to trying to execute
NEON code on a target that doesn't support it revealed by my
arm-none-eabi testing on a gazillion different multilibs. Inspired by
what vect.exp does and my other patch in this group to fix
advsimd-intrinsics.exp, I've ha
On May 21, 2015 12:13:19 AM GMT+02:00, Jan Hubicka wrote:
>Hi,
>this patch extends statistics from tree-ssa-alias to also cover TBAA
>oracle.
>This is useful to keep track of aliasing effectivity. For example the
>hack
>in alias.c putting globbing all pointers to one costs about 20% of all
>answe
Hi,
On 8 May 2015 at 00:07, Joseph Myers wrote:
> On Mon, 20 Apr 2015, Pavel Kopyl wrote:
>
>> Hi all,
>>
>>
>> To build a GCC-4.9.2 ARM cross-compiler for my setting I need to configure it
>> with "--with-sysroot=/ --with-gxx-include-dir=/usr/include/c++/4.9.2".
>> But I found that gcc driver r
Hi Julian!
On Thu, 7 May 2015 16:56:11 +0100, Julian Brown wrote:
> On Tue, 5 May 2015 16:09:18 +0200
> Thomas Schwinge wrote:
> > On Tue, 5 May 2015 08:43:48 -0400, John David Anglin
> > wrote:
> > > On 2015-05-05 5:43 AM, Thomas Schwinge wrote:
> > > >> FAIL: libgomp.oacc-c/../libgomp.oacc-c-
101 - 136 of 136 matches
Mail list logo
