This patch to the Go frontend fixes the fieldtrack info emitted for an
unexported type. Bootstrapped on x86_64-unknown-linux-gnu. Committed
to mainline.
Ian
diff -r 0a58667fb9cd go/expressions.cc
--- a/go/expressions.cc Thu Nov 29 09:34:24 2012 -0800
+++ b/go/expressions.cc Thu Nov 29 22:55:58
Hi,
I committed the patch fixing size regression of regrename on Thumb2 to ARM
Embedded-4_7 branch as r193980.
Thanks.
gcc/ChangeLog.arm
2012-11-30 Bin Cheng
* config/arm/arm-protos.h (tune_params): Add
preferred_renaming_class.
* config/arm/arm.c (arm_slowmul_tune, ar
Hi,
I committed the patch disabling "-fira-hoist-pressure" on Thumb2 to ARM
Embedded-4_7 branch as r193979.
Thanks.
gcc/ChangeLog.arm
2012-11-30 Bin Cheng
* config/arm/arm.c (arm_option_override): Disable option
-fira-hoist-pressure on Thumb2.
gcc/testsuite/ChangeLog.arm
2012
Hi,
Big code size regression is introduced by register renaming pass on thumb2
instruction set. The root cause comprises two aspects:
1. The pass renames LO_REGS into high registers, changing 2 bytes
instructions into 4 bytes. This is because function find_best_rename_reg
does not forbid renaming L
I have merged trunk revision 193973 to the gccgo branch.
Ian
This patch is pending for a long time.
Any progress?
Cheers,
Yi-Hsiu, Hsu
-Original Message-
From: Yi-Hsiu Hsu
Sent: Tuesday, June 26, 2012 1:51 PM
To: 'Chung-Lin Tang'
Cc: Ramana Radhakrishnan; gcc-patches@gcc.gnu.org
Subject: RE: [PATCH, ARM] New CPU support for Marvell PJ4 cores
Hi
Am 29.11.2012 23:51, schrieb Janus Weil:
one thing that I do not like about your patch is the modification of
"gfc_find_derived_vtab": You create two versions of it, one of which
creates the vtab if it does not exist, while the other version does
not do this. [...] can you explain to me why thi
Diego Novillo writes:
> On Sat, Nov 24, 2012 at 5:58 PM, Doug Evans wrote:
> > Hi.
> > This seventh patch adds new options --manifest_subdir, --manifest_name.
> > Useful when using validate_failures.py with a different tool, instead of
> > gcc.
> >
> > Ok to check in?
> >
> > 2012-11-24
Hello,
This patch rewrites sched-vis.c, and graph.c, to use the pretty-print
machinery. This makes it easier to add GIMPLE cfg dumping also,
because all tree/gimple dumping is done via pretty-print. It also
removes the ugly (and unsafe?) static print buffers that sched-vis.c
and its users had.
Bo
Hi Tobias,
> effectively, this patch doesn't do anything. Except, it updates the –
> deactivated – finalization wrapper.
>
>
> Note: This patch does not include any code to actually call the finalization
> wrapper. Nor is the modified code ever called in gfortran. However, that
> patch paves the r
Hello!
> When changing reg_known_value to a VEC, Steven sneaked in a small
> semantic change. That change has the advantage of improving compilation
> time substantially on some testcases (including PR55489), but it is a
> bit heavy-handed: it also makes set_known_reg_value a no-op, while
> get_k
On Thu, Nov 29, 2012 at 3:46 PM, Doug Evans wrote:
> 2012-11-29 Doug Evans
>
> * testsuite-management/validate_failures.py: Rename variable
> manifest_name to manifest_path everywhere.
OK.
Diego.
A different fix is required for 4.6, because the __atomic
functions don't exist yet.
r~
PR libgcc/48076
* emutls.c (__emutls_get_address): Add memory barrier before
referencing emutls_key.
diff --git a/gcc/emutls.c b/gcc/emutls.c
index b7ee3bd..fcf6eb5 100644
--- a/gcc/e
On Thu, 29 Nov 2012, Christophe Lyon wrote:
> 2012-11-28 Christophe Lyon
>
> gcc/
> * config/arm/arm-protos.h (tune_params): Add
> prefer_neon_for_64bits field.
> * config/arm/arm.c (prefer_neon_for_64bits): New variable.
> (arm_slowmul_tune): Default prefer_neon_for_64bits
Diego Novillo writes:
> On Sat, Nov 24, 2012 at 5:50 PM, Doug Evans wrote:
> > Hi.
> > This fourth patch renames option --manifest to --manifest_path.
> > I have a later patch that adds a --manifest_name option, making
> > "--manifest"
> > too confusing/ambiguous.
> >
> > Ok to check in?
On Thu, Nov 29, 2012 at 11:06 AM, Jakub Jelinek wrote:
> On Thu, Nov 29, 2012 at 07:24:38PM +0100, Paolo Carlini wrote:
>> On 11/29/2012 06:36 PM, Tobias Burnus wrote:
>> >H.J. Lu wrote:
>> >>This patch adds --with-build-config=bootstrap-asan support. Tested on
>> >>Linux/x86-64. OK to install?
> Of course, I also verified that the darwin10 and sparc issues that
> spawned this long thread are fixed.
Confirmed: bootstrapped on top of r193939, aermod.f90 is now correctly
compiled.
Thanks,
Dominique
My earlier semi-fix for this issue failed to update
LAMBDA_EXPR_THIS_CAPTURE to point to the proxy variable, so uses ended
up with an unattached FIELD_DECL. Fixed thus.
Tested x86_64-pc-linux-gnu, applying to trunk and 4.7.
commit 73e140fd81aa10994184bb8edfe9cfafdc1499a9
Author: Jason Merrill
On Thu, Nov 29, 2012 at 11:06 AM, Jakub Jelinek wrote:
> On Thu, Nov 29, 2012 at 07:24:38PM +0100, Paolo Carlini wrote:
>> On 11/29/2012 06:36 PM, Tobias Burnus wrote:
>> >H.J. Lu wrote:
>> >>This patch adds --with-build-config=bootstrap-asan support. Tested on
>> >>Linux/x86-64. OK to install?
On Thu, Nov 29, 2012 at 11:03 AM, H.J. Lu wrote:
> On Thu, Nov 29, 2012 at 10:32 AM, H.J. Lu wrote:
>> On Thu, Nov 29, 2012 at 10:24 AM, Paolo Carlini
>> wrote:
>>> On 11/29/2012 06:36 PM, Tobias Burnus wrote:
H.J. Lu wrote:
>
> This patch adds --with-build-config=bootstrap-asa
On Thu, Nov 29, 2012 at 07:24:38PM +0100, Paolo Carlini wrote:
> On 11/29/2012 06:36 PM, Tobias Burnus wrote:
> >H.J. Lu wrote:
> >>This patch adds --with-build-config=bootstrap-asan support. Tested on
> >>Linux/x86-64. OK to install?
> >
> >I think that patch has broken bootstrap for me. If I do
On 2012-11-28 23:22, Michael Zolotukhin wrote:
> + debug_rtx (desc);
... oh, I see. You wanted dump_rtx here, so you can output to stdout.
r~
On Thu, Nov 29, 2012 at 10:32 AM, H.J. Lu wrote:
> On Thu, Nov 29, 2012 at 10:24 AM, Paolo Carlini
> wrote:
>> On 11/29/2012 06:36 PM, Tobias Burnus wrote:
>>>
>>> H.J. Lu wrote:
This patch adds --with-build-config=bootstrap-asan support. Tested on
Linux/x86-64. OK to install?
>>
The following patch fixes
gcc.gnu.org/bugzilla/show_bug.cgi?id=55456
The patch was successfully bootstrapped and tested on x86/x86-64.
Committed as rev. 193948.
2012-11-28 Vladimir Makarov
PR middle-end/55456
* lra-int.h (lra_new_regno_start): New external.
*
On Thu, Nov 29, 2012 at 10:40 AM, H.J. Lu wrote:
> On Thu, Nov 29, 2012 at 10:34 AM, Andrew Pinski wrote:
>> On Thu, Nov 29, 2012 at 10:30 AM, H.J. Lu wrote:
>>> Hi,
>>>
>>> Since libsanitizer is used for bootstrap and compiled with raw_cxx,
>>> we need to use explicit -I for libstdc++-v3 header
On Thu, Nov 29, 2012 at 10:34 AM, Andrew Pinski wrote:
> On Thu, Nov 29, 2012 at 10:30 AM, H.J. Lu wrote:
>> Hi,
>>
>> Since libsanitizer is used for bootstrap and compiled with raw_cxx,
>> we need to use explicit -I for libstdc++-v3 header files in
>> libsanitizer. Otherwise, we will get
>>
>>
On 2012-11-28 17:05, Aldy Hernandez wrote:
> Author: Aldy Hernandez
> Date: Wed Nov 28 18:17:33 2012 -0600
>
> PR middle-end/55401
> * trans-mem.c (get_tm_region_blocks): Exclude uninstrumented
> blocks from vector if requested.
> (collect_bb2reg): Pass new argument to
>
On 2012-11-28 23:22, Michael Zolotukhin wrote:
> +.PHONY: mddump
> +mddump: $(BUILD_RTL) $(MD_DEPS) build/genmddump$(build_exeext)
> + $(RUN_GEN) build/genmddump$(build_exeext) $(md_file) 2> tmp-mddump.md
Why in the world are you dumping to stderr instead of stdout?
r~
On Thu, Nov 29, 2012 at 10:30 AM, H.J. Lu wrote:
> Hi,
>
> Since libsanitizer is used for bootstrap and compiled with raw_cxx,
> we need to use explicit -I for libstdc++-v3 header files in
> libsanitizer. Otherwise, we will get
>
> libtool: compile: unrecognized option `-D_GNU_SOURCE'
> libtool:
On 11/16/2012 01:21 PM, Caroline Tice wrote:
-start_objects (int method_type, int initp)
+start_objects (int method_type, int initp, const char *extra_name)
I don't think we want to mess with start_objects and such here. Can't
you just use
DECL_STATIC_CONSTRUCTOR (decl) = 1;
SET
On Thu, Nov 29, 2012 at 10:24 AM, Paolo Carlini
wrote:
> On 11/29/2012 06:36 PM, Tobias Burnus wrote:
>>
>> H.J. Lu wrote:
>>>
>>> This patch adds --with-build-config=bootstrap-asan support. Tested on
>>> Linux/x86-64. OK to install?
>>
>>
>> I think that patch has broken bootstrap for me. If I
On Thu, Nov 29, 2012 at 10:27 AM, Kai Tietz wrote:
> Fine. Tested patch using XNEWVEC/XRESIZEVEC for this. Ok for apply?
This is OK with a ChangeLog entry.
Thanks.
Ian
> Index: prefix.c
> ===
> --- prefix.c(Revision 193939)
On 2012-11-29 00:58, Michael Zolotukhin wrote:
> 2012-11-29 Michael Zolotukhin
>
> * gensupport.c (maybe_eval_c_test): Remove not-null check for expr.
> * read-rtl.c (apply_iterators): Initialize condition with "" instead
> of NULL.
Ok.
r~
Hi,
Since libsanitizer is used for bootstrap and compiled with raw_cxx,
we need to use explicit -I for libstdc++-v3 header files in
libsanitizer. Otherwise, we will get
libtool: compile: unrecognized option `-D_GNU_SOURCE'
libtool: compile: Try `libtool --help' for more information.
This patch
Fine. Tested patch using XNEWVEC/XRESIZEVEC for this. Ok for apply?
Kai
Index: prefix.c
===
--- prefix.c(Revision 193939)
+++ prefix.c(Arbeitskopie)
@@ -157,12 +157,12 @@ lookup_key (char *key)
}
size = 32;
- dst
On 11/29/2012 06:36 PM, Tobias Burnus wrote:
H.J. Lu wrote:
This patch adds --with-build-config=bootstrap-asan support. Tested on
Linux/x86-64. OK to install?
I think that patch has broken bootstrap for me. If I do a normal
bootstrap, Stage1 fails with:
libtool: compile: unrecognized opti
On Thu, Nov 29, 2012 at 4:09 AM, Kai Tietz wrote:
>
> ChangeLog
>
> 2012-11-29 Kai Tietz
>
> PR target/53912
> * prefix.c (lookup_key): Explicit cast return-type of xmalloc/xrealloc
> to char *.
>
> Tested for i686-w64-mingw32, x86_64-w64-mingw32, and
> x86_64-unknown-gnu-l
This patch implements support for somewhat experimental field tracking.
If a struct field has a tag go:"track", the compiler will track
references to the field. A new function runtime.Fieldtrack will add the
names of all tracked fields to a map. The idea is that linker garbage
collection will rem
H.J. Lu wrote:
Please try this: http://gcc.gnu.org/ml/gcc-patches/2012-11/msg01530.html
I just did locally "git log -2 -p|patch -p1 -R" - and wait until you
sort that out with the patch reviewers, which seemingly didn't like that
patch.
BTW: I am not sure whether it is possible with the cu
PR 55073 is a case where scheduling appears to mess up the order of
instructions to the extent that they no-longer give the correct results.
However, looking at the patterns, I think they are ill-defined. IIRC
tied operands should tie a source to a destination, rather than a
destination to a so
Hi,
All plugin tests failed when GCC is configured with
--with-build-config=bootstrap-asan, since linker can't find
libasan at link-time. -fsanitize=address isn't needed to test
plugin support. This patch filters out -fsanitize=address from
PLUGINCFLAGS. OK to install?
Thanks.
H.J.
---
gcc/
> Yikes, sorry, it wasn't clear to me what PROP_loops really does. Anyway,
> I think I have a better fix now. The problem is just that when removing
> BB 4 (which was a header), we have to zap ->header and ->latch. We
> already have code for this:
>
> if (current_loops != NULL
> && e->s
On Thu, Nov 29, 2012 at 9:36 AM, Tobias Burnus wrote:
> H.J. Lu wrote:
>>
>> This patch adds --with-build-config=bootstrap-asan support. Tested on
>> Linux/x86-64. OK to install?
>
>
> I think that patch has broken bootstrap for me. If I do a normal bootstrap,
> Stage1 fails with:
>
> libtool: c
Hi,
This patch puts a breakpoint on __asan_report_error if CFLAGS contains
-fsanitize=address, similar to fancy_abort and internal_error. OK to
install?
H.J.
--
2012-11-24 H.J. Lu
* configure.ac: Append gdbasan.in to .gdbinit if CFLAGS contains
-fsanitize=address.
*
Hi,
When GCC is configured with --with-build-config="bootstrap-asan", all
-flto tests will fail since -fsanitize=address is used to compile host
libiberty, which is used to create liblto_plugin.so, and linker isn't
compiled with -fsanitize=address. This patch filters out
-fsanitize=address from C
Hi,
When GCC is configured with --with-build-config="bootstrap-asan", all
-flto tests will fail since -fsanitize=address is used to compile
liblto_plugin.so and linker isn't compiled with -fsanitize=address.
This patch filters out -fsanitize=address from CFLAGS if we aren't
in a combined tree with
H.J. Lu wrote:
This patch adds --with-build-config=bootstrap-asan support. Tested on
Linux/x86-64. OK to install?
I think that patch has broken bootstrap for me. If I do a normal
bootstrap, Stage1 fails with:
libtool: compile: unrecognized option `-D_GNU_SOURCE'
libtool: compile: Try `libt
Hi,
I have been working on a patch to avoid using Neon for 64 bits bitops
when it's too expensive to move data between core and Neon registers.
Benchmarking and validation look OK on the 4.7 branch (compiler
configured for thumb and hard FP)
- validation on cortex-a9 board OK
- bencharking shows
On 2012-11-29 06:29, Eric Botcazou wrote:
> PR middle-end/55321
> * calls.c (emit_library_call_value_1): Mark as no-nonlocal if no-throw.
Ok.
r~
> If you can confirm you have a copyright assignment on file I can take care
> of checking it in.
Yes, I am covered by the copyright assignement RT 211150 between
STMicroelectronics and FSF.
Thanks,
Yvan
On 11/29/2012 10:42 AM, Yvan Roux wrote:
Hi,
on ARMv7, the code generated for the __atomic_load builtins in the
__ATOMIC_ACQUIRE memory model, puts a memory barrier before the load, whereas
the semantic of the acquire memory model implies a barrier after.
The issue seems to be in expand_atomic_
On Thu, Nov 29, 2012 at 04:50:19PM +0100, Steven Bosscher wrote:
> > 2012-11-29 Marek Polacek <>
> >
> > PR middle-end/54838
> > * cprop.c (bypass_block): Set header and latch to NULL when
> > BB has more than one latch edge.
> > (n_latches): New variable.
>
> You
>
> I assume you mean use the bitpack streaming functionality used other
> places in lto-cgraph.c, and not change the way it is done in gcov-io.c
> where this was cloned from? I will change the lto-cgraph.c code to use
> the bitpacking.
Yes, we don't have the facility on gcov-io, so we need to do
On Thu, Nov 29, 2012 at 8:17 AM, Jan Hubicka wrote:
>> This patch ensures that the histograms from the profile summary are streamed
>> through the LTO files so that the working set can be computed for use in
>> downstream optimizations.
>>
>> Bootstrapped and tested on x86_64-unknown-linux-gnu. Ok
> This patch ensures that the histograms from the profile summary are streamed
> through the LTO files so that the working set can be computed for use in
> downstream optimizations.
>
> Bootstrapped and tested on x86_64-unknown-linux-gnu. Ok for trunk?
>
> Thanks,
> Teresa
>
> 2012-11-28 Teresa
On Thu, Nov 22, 2012 at 12:37:47PM -0800, H.J. Lu wrote:
> 2012-11-21 H.J. Lu
>
> * Makefile.def (target_modules): Add bootstrap=true and
> raw_cxx=true to libsanitizer.
> * configure.ac (bootstrap_target_libs): Add libsanitizer.
> * Makefile.in: Regenerated.
> * c
On Thu, Nov 29, 2012 at 12:43 AM, Richard Biener
wrote:
> On Thu, Nov 29, 2012 at 5:18 AM, H.J. Lu wrote:
>> From: "H.J. Lu"
>> To: gcc-patches@gcc.gnu.org
>> Cc: Joseph Myers , Paolo Bonzini
>> Bcc:
>> Subject: [PATCH] Support -fuse-ld=bfd and -fuse-ld=gold
>> Reply-To:
>>
>> Hi,
>>
>> Binuti
Am 29.11.2012 10:10, schrieb Markus Trippelsdorf:
> Additionally, what is the rationale for this patch? IOW why isn't the
> following
> enough?
>
> x4 ~ # ln -f /usr/x86_64-pc-linux-gnu/binutils-bin/git/ld.gold
> /usr/x86_64-pc-linux-gnu/binutils-bin/git/ld
> x4 ~ # ld -v
> GNU gold (GNU Binuti
On Thu, Nov 29, 2012 at 4:38 PM, Marek Polacek wrote:
> On Thu, Nov 29, 2012 at 09:34:31AM +0100, Richard Biener wrote:
>> Definitely not - that means to not preserve loops until after cprop. The
>> goal
>> is to preserve loops everywhere!
>
> Yikes, sorry, it wasn't clear to me what PROP_loops r
Am 29.11.2012 09:43, schrieb Richard Biener:
> Do we need to consider GNU ld and gold coming from different binutils versions
> and thus do we need two sets of linker feature tests at configure
> time?
both GNU ld and gold are built from the same sources. So it is likely that they
come from the sa
Hi,
on ARMv7, the code generated for the __atomic_load builtins in the
__ATOMIC_ACQUIRE memory model, puts a memory barrier before the load, whereas
the semantic of the acquire memory model implies a barrier after.
The issue seems to be in expand_atomic_load which puts a memory fence before
the l
On Thu, Nov 29, 2012 at 04:38:52PM +0100, Marek Polacek wrote:
> 2012-11-29 Marek Polacek
>
> PR middle-end/54838
> * cprop.c (bypass_block): Set header and latch to NULL when
> BB has more than one latch edge.
> (n_latches): New variable.
Of course here should be (n_la
On Thu, Nov 29, 2012 at 09:34:31AM +0100, Richard Biener wrote:
> Definitely not - that means to not preserve loops until after cprop. The goal
> is to preserve loops everywhere!
Yikes, sorry, it wasn't clear to me what PROP_loops really does. Anyway,
I think I have a better fix now. The proble
> I agree that this kind of MEM is less than ideal, but I thought:
>
> set_mem_attributes_minus_bitpos (to_rtx, to, 0, bitpos);
>
> said that the attributes of TO_RTX will to be TO _once a pending offset-and-
> narrowing operation has been applied_. So we have:
>
> /* If we modifi
Please find the updated patch that improves the comment added to the
test in the generic part of the testsuite.
Thanks,
Yufeng
On 11/26/12 09:55, Marcus Shawcroft wrote:
On 21/11/12 14:31, Yufeng Zhang wrote:
Hi,
This patch updates the AArch64 port to mangle __va_list as it is in
namespace s
OK.
Jason
OK.
Jason
> "Thomas" == Thomas Quinot writes:
Thomas> As for the size impact on line_map_ordinary, indeed that's an additional
Thomas> int here, but I'm not sure we allocate so many objects of that kind that
Thomas> the increase is significant.
Could you please measure it?
My guess is that it is worse
On 28/11/12 11:07, James Greenhalgh wrote:
Hi,
The backport for this patch was not entirely clean,
__builtin_thread_pointer was made a front end builtin on the
4.8 branch, but remains a back-end builtin on aarch64-4.7.
Otherwise, there are no differences between this patch and the
patch which
On 29/11/12 14:58, Greta Yorsh wrote:
-Original Message-
From: Richard Earnshaw
Sent: 29 November 2012 10:12
To: Greta Yorsh
Cc: GCC Patches; Ramana Radhakrishnan; ni...@redhat.com;
p...@codesourcery.com
Subject: Re: [PATCH,ARM] Subdivide alu into alu_reg and simple_alu_imm
; ???
On 26/11/12 16:50, Sofiane Naci wrote:
Hi,
Constant building in the AArch64 backend spits out assembly code, which
affects scheduling of the generated code.
This patch rewrites the code to use RTL patterns.
A full aarch64-none-elf regression run shows no issues.
Thanks
Sofiane
-
ChangeLo
> -Original Message-
> From: Richard Earnshaw
> Sent: 29 November 2012 10:12
> To: Greta Yorsh
> Cc: GCC Patches; Ramana Radhakrishnan; ni...@redhat.com;
> p...@codesourcery.com
> Subject: Re: [PATCH,ARM] Subdivide alu into alu_reg and simple_alu_imm
>
>
> ; ??? Check Thumb-2 split leng
On Thu, Nov 29, 2012 at 12:53 AM, Steven Bosscher wrote:
> Attached is a different fix. It splits DF_REF_IN_NOTE in two: One flag
> for each kind of note. This allows the dead note removal code to
> distinguish the source note for the EQ_USES. I needed to remove one
> flag to keep the df_ref_flags
PING
On 20 November 2012 20:34, Matthew Gretton-Dann
wrote:
> All,
>
> This patch backports Matthais Klose's arm*-*-linux-gnueabihf triplet
> support patch of 2012-10-15 to 4.7.
>
> The backport was not clean as 4.8 has obsoleted various arm*-*-*
> triplets which are valid in 4.7.
>
> I have test
On 24 November 2012 00:27, Ramana Radhakrishnan
wrote:
> On Wed, Nov 21, 2012 at 7:59 PM, Matthew Gretton-Dann
> wrote:
[snip]
>> The fix is to decrease the pool_range of all insns by 2 when generating
>> Thumb code. There is no need to change neg_pool_range values as rounding
>> down here will
On Thu, Nov 29, 2012 at 1:33 PM, Martin Jambor wrote:
> Hi,
>
> On Thu, Nov 29, 2012 at 12:54:20PM +0100, Richard Biener wrote:
>> On Thu, Nov 29, 2012 at 11:06 AM, Martin Jambor wrote:
>> > Hi,
>> >
>> > thanks for the review. When writing a reply I realized I indeed made
>> > a mistake or two
> Well, there's a call at the tree level for __sync_synchronize as well.
>
> So the question is: why does __s_s ICE but memcmp doesn't? I suppose
> it's the mere rarity with which HAVE_cmpmemsi is defined, and expansion
> of gen_cmpmemsi fails.
>
> There are plenty of other calls to emit_library_
Hi all,
This patch adds the intrinsics support for the vrnd intrinsics that are
implemented by the vrint instructions.
The .ml scripts contain the new information and should used to regenerate
the arm_neon.h header file, tests and documentation.
In particular:
* config/arm/arm_neon.h should be rege
Hi all,
This patch adds support for the vrint builtins. It also gathers the unspec
definitions in the various .md files in one file: unspecs.md.
A new iterator is defined that iterates over some new unspecs, in a similar
way to the scalar vrint implementation.
No regressions on arm-none-eabi.
Ok
Hi all,
This patch adds an effective target check to the testsuite for ARMv8 NEON
support. A corresponding add_options procedure is added.
This is used by the AArch32 NEON intrinsics tests. An entry in the
documentation is added as well.
Ok for trunk?
Thanks,
Kyrill
gcc/ChangeLog
2012-11-29 Ky
Hi all,
This set of patches adds support for the AArch32 vrint NEON builtins. It
also adds related NEON intrinsics. It is organised in 3 patches:
* The first patch adds support for the builtin versions of the instructions.
* The second is a small testsuite patch that adds the effective target chec
On Thu, 29 Nov 2012, Jakub Jelinek wrote:
> Hi!
>
> As discussed earlier on mailing list and on IRC with Richard,
> pt_solution_includes can be used for testing of whether var
> might escape current function (for -O0 unfortunately returns
> true even for !may_be_aliased automatic vars, so I'm tes
On Thu, Nov 29, 2012 at 4:24 PM, Jakub Jelinek wrote:
> Hi!
>
> As discussed earlier on mailing list and on IRC with Richard,
> pt_solution_includes can be used for testing of whether var
> might escape current function (for -O0 unfortunately returns
> true even for !may_be_aliased automatic vars,
On Thu, Nov 29, 2012 at 1:10 AM, Markus Trippelsdorf
wrote:
> On 2012.11.29 at 09:43 +0100, Richard Biener wrote:
>> On Thu, Nov 29, 2012 at 5:18 AM, H.J. Lu wrote:
>> > From: "H.J. Lu"
>> > To: gcc-patches@gcc.gnu.org
>> > Cc: Joseph Myers , Paolo Bonzini
>> >
>> > Bcc:
>> > Subject: [PATCH]
When working on PR55124 I noticed we do some useless work in PRE.
This cleans it up and makes it more obvious what we do in compute_avail.
Bootstrapped and tested on x86_64-unknown-linux-gnu, applied to trunk.
Richard.
2012-11-29 Richard Biener
* tree-ssa-pre.c (get_expr_value_id):
Updated variant using HOST_WIDE_INT_PRINT.
Tested for i686-w64-mingw32 and x86_64-w64-mingw32. Ok for apply?
Kai
Index: tree-dump.c
===
--- tree-dump.c (Revision 193925)
+++ tree-dump.c (Arbeitskopie)
@@ -177,7 +177,8 @@ void
dump
Hello,
this patch fixes several ICEs when using constexpr SIMD vectors. Support
for subscripting is still missing though, which is why I am not adding
some static_asserts to the testcase. I don't use build_vector_from_ctor
because it doesn't check for non-constant elements and doesn't handle
Ok, here is the tested adjusted version using (u)intptr_t and hwi. Ok
for apply?
Kai
Index: cp/class.c
===
--- cp/class.c (Revision 193925)
+++ cp/class.c (Arbeitskopie)
@@ -7815,9 +7815,9 @@ dump_class_hierarchy_r (FILE *stream,
On Thu, Nov 29, 2012 at 1:14 PM, Tom de Vries wrote:
> On 29/11/12 11:26, Richard Biener wrote:
>> I'm continuing trying to move value-ids back to PRE land. Your patch
>> would be nice in a form that verifies the order is indeed topological,
>> maybe you can re-work it in a way that does this in
On 11/29/2012 12:45 PM, Kai Tietz wrote:
> 2012/11/29 Pedro Alves :
>> On 11/29/2012 12:03 PM, Kai Tietz wrote:
>>> Hello,
>>>
>>> this trivial patch fixes a bootstrap issue on LLP64 hosts.
>>
>> But it's trivial at the expense of truncating the pointer.
>> Looking around, I'd think that something
2012/11/29 Pedro Alves :
> On 11/29/2012 12:03 PM, Kai Tietz wrote:
>> Hello,
>>
>> this trivial patch fixes a bootstrap issue on LLP64 hosts.
>
> But it's trivial at the expense of truncating the pointer.
> Looking around, I'd think that something like casting to hwi (after
> uintptr_t),
> and th
Hi,
On Thu, Nov 29, 2012 at 12:54:20PM +0100, Richard Biener wrote:
> On Thu, Nov 29, 2012 at 11:06 AM, Martin Jambor wrote:
> > Hi,
> >
> > thanks for the review. When writing a reply I realized I indeed made
> > a mistake or two in the part concerning prev_base and the code was not
> > what it
On 11/29/2012 12:03 PM, Kai Tietz wrote:
> Hello,
>
> this trivial patch fixes a bootstrap issue on LLP64 hosts.
But it's trivial at the expense of truncating the pointer.
Looking around, I'd think that something like casting to hwi (after uintptr_t),
and then use HOST_WIDE_INT_PRINT_HEX as forma
Hi!
As discussed earlier on mailing list and on IRC with Richard,
pt_solution_includes can be used for testing of whether var
might escape current function (for -O0 unfortunately returns
true even for !may_be_aliased automatic vars, so I'm testing
for that too), and there are IMHO tons of complete
On 29/11/12 11:26, Richard Biener wrote:
> I'm continuing trying to move value-ids back to PRE land. Your patch
> would be nice in a form that verifies the order is indeed topological,
> maybe you can re-work it in a way that does this in
> sorted_array_from_bitmap_set in a ENABLE_CHECKING piece?
Hello,
this trivial patch fixes a bootstrap issue on LLP64 hosts.
ChangeLog
2012-11-29 Kai Tietz
PR target/53912
* tree-dump.c (dump_pointer): Cast from pointer via intptr_t.
Tested for i686-w64-mingw32, x86_64-w64-mingw32, and
x86_64-unknown-gnu-linux. Ok for apply?
Regards,
Hello,
this trivial patch fixes a bootstrap issue on LLP64 hosts.
ChangeLog
2012-11-29 Kai Tietz
PR target/53912
* stmt.c (compute_cases_per_edge): Cast from pointer via intptr_t.
(expand_case): Likewise.
Tested for i686-w64-mingw32, x86_64-w64-mingw32, and
x86_64-unkno
Hello,
this trivial patch fixes a bootstrap issue on LLP64 hosts.
ChangeLog
2012-11-29 Kai Tietz
PR target/53912
* print-tree.c (print_node): Cast from pointer via uintptr_t.
Tested for i686-w64-mingw32, x86_64-w64-mingw32, and
x86_64-unknown-gnu-linux. Ok for apply?
Regards,
Hello,
this trivial patch fixes a bootstrap issue on LLP64 hosts.
ChangeLog
2012-11-29 Kai Tietz
PR target/53912
* prefix.c (lookup_key): Explicit cast return-type of xmalloc/xrealloc
to char *.
Tested for i686-w64-mingw32, x86_64-w64-mingw32, and
x86_64-unknown-gnu-lin
Hello,
this trivial patch fixes a bootstrap issue on LLP64 hosts.
ChangeLog
2012-11-29 Kai Tietz
PR target/53912
* pointer-set.c (hash1): Cast from pointer via uintptr_t.
Tested for i686-w64-mingw32, x86_64-w64-mingw32, and
x86_64-unknown-gnu-linux. Ok for apply?
Regards,
Kai
1 - 100 of 122 matches
Mail list logo
