2015-03-24 17:06 GMT+03:00 Jakub Jelinek :
> On Tue, Mar 24, 2015 at 12:22:27PM +0300, Ilya Enkovich wrote:
>> 2015-03-24 11:33 GMT+03:00 Jakub Jelinek :
>> > On Thu, Mar 19, 2015 at 11:29:44AM +0300, Ilya Enkovich wrote:
>> >> + /* We might propagate instrumented function pointer into
>> >> +
On Wed, Mar 25, 2015 at 11:05:17AM +0300, Ilya Enkovich wrote:
> > The question is what you want to do in the LTO case for the different cases,
> > in particular a TU compiled with -fcheck-pointer-bounds and LTO link without
> > that, or TU compiled without -fcheck-pointer-bounds and LTO link with
Hi Tom!
On Sat, 21 Mar 2015 23:30:51 +0100, Tom de Vries wrote:
> On 20-03-15 12:38, Tom de Vries wrote:
> > On 19-03-15 12:05, Tom de Vries wrote:
> >> On 18-03-15 18:22, Tom de Vries wrote:
> >>> this patch fixes PR65460.
> >>>
> >>> The patch marks offloaded functions as parallelized, which me
We appear to have erroneously set 'extra_parts' in nios2-linux libgcc,
to include the crti.o/crtn.o files intended for nios2 EABI. This still
largely worked, which is why we haven't noticed it till now, expect some
features like gprof profiling wasn't properly set up.
This patch removes the extra_
2015-03-24 17:40 GMT+03:00 Richard Biener :
> On Tue, Mar 24, 2015 at 3:06 PM, Jakub Jelinek wrote:
>> On Tue, Mar 24, 2015 at 12:22:27PM +0300, Ilya Enkovich wrote:
>>
>> The question is what you want to do in the LTO case for the different cases,
>> in particular a TU compiled with -fcheck-point
2015-03-25 11:16 GMT+03:00 Jakub Jelinek :
> On Wed, Mar 25, 2015 at 11:05:17AM +0300, Ilya Enkovich wrote:
>> > The question is what you want to do in the LTO case for the different
>> > cases,
>> > in particular a TU compiled with -fcheck-pointer-bounds and LTO link
>> > without
>> > that, or T
On Tue, 24 Mar 2015, Jakub Jelinek wrote:
> On Tue, Mar 24, 2015 at 04:19:46PM +0100, Jakub Jelinek wrote:
> > Bootstrapped/regtested on x86_64-linux and i686-linux, ok for trunk?
>
> Also tested with
> ../configure --with-build-config=bootstrap-lto
> --enable-languages=c,c++,fortran,objc,obj-c+
On Wed, Mar 25, 2015 at 9:50 AM, Ilya Enkovich wrote:
> 2015-03-24 17:40 GMT+03:00 Richard Biener :
>> On Tue, Mar 24, 2015 at 3:06 PM, Jakub Jelinek wrote:
>>> On Tue, Mar 24, 2015 at 12:22:27PM +0300, Ilya Enkovich wrote:
>>>
>>> The question is what you want to do in the LTO case for the diffe
Hi Andre,
> Le 24 mars 2015 à 18:06, Andre Vehreschild a écrit :
>
> Hi all,
>
> I have worked on the comments Mikael gave me. I am now checking for
> class_pointer in the way he pointed out.
>
> Furthermore did I *join the two parts* of the patch into this one, because
> keeping both in sync
On Wed, Mar 25, 2015 at 10:38:56AM +0100, Richard Biener wrote:
> --- gcc/passes.c(revision 221633)
> +++ gcc/passes.c(working copy)
> @@ -156,7 +156,8 @@ void
> pass_manager::execute_early_local_passes ()
> {
>execute_pass_list (cfun, pass_build_ssa_passes_1->sub);
> - execu
2015-03-25 12:50 GMT+03:00 Jakub Jelinek :
> On Wed, Mar 25, 2015 at 10:38:56AM +0100, Richard Biener wrote:
>> --- gcc/passes.c(revision 221633)
>> +++ gcc/passes.c(working copy)
>> @@ -156,7 +156,8 @@ void
>> pass_manager::execute_early_local_passes ()
>> {
>>execute_pass_li
On Wed, Mar 25, 2015 at 01:06:46PM +0300, Ilya Enkovich wrote:
> > There is still the wasteful pass_fixup_cfg at the start of:
> > PUSH_INSERT_PASSES_WITHIN (pass_local_optimization_passes)
> > NEXT_PASS (pass_fixup_cfg);
> > which wasn't there before chkp. Perhaps this should be a different
> >
On Wed, Mar 25, 2015 at 10:50 AM, Jakub Jelinek wrote:
> On Wed, Mar 25, 2015 at 10:38:56AM +0100, Richard Biener wrote:
>> --- gcc/passes.c(revision 221633)
>> +++ gcc/passes.c(working copy)
>> @@ -156,7 +156,8 @@ void
>> pass_manager::execute_early_local_passes ()
>> {
>>ex
Le 24/03/2015 23:39, Mikael Morin a écrit :
> The patch I propose here adds a flag to remember the function has been
> called, and skip it the second time.
> I considered reusing the existing 'resolved' field, but I had to
> slightly change its semantics to prevent regressing somewhere, and I was
>
On Wed, Mar 25, 2015 at 11:11 AM, Jakub Jelinek wrote:
> On Wed, Mar 25, 2015 at 01:06:46PM +0300, Ilya Enkovich wrote:
>> > There is still the wasteful pass_fixup_cfg at the start of:
>> > PUSH_INSERT_PASSES_WITHIN (pass_local_optimization_passes)
>> > NEXT_PASS (pass_fixup_cfg);
>> > which was
2015-03-25 13:15 GMT+03:00 Richard Biener :
> On Wed, Mar 25, 2015 at 10:50 AM, Jakub Jelinek wrote:
>> On Wed, Mar 25, 2015 at 10:38:56AM +0100, Richard Biener wrote:
>>> --- gcc/passes.c(revision 221633)
>>> +++ gcc/passes.c(working copy)
>>> @@ -156,7 +156,8 @@ void
>>> pass_ma
On 03/25/2015 12:37 AM, Jan Hubicka wrote:
On Tue, Mar 24, 2015 at 10:54:25PM +0100, Martin Liška wrote:
--- a/gcc/symbol-summary.h
+++ b/gcc/symbol-summary.h
@@ -81,6 +81,12 @@ public:
m_symtab_insertion_hook = NULL;
m_symtab_removal_hook = NULL;
m_symtab_duplication_hook = NU
Hello.
Following patch correctly creates symlink that now points to a wrong location.
Ready for trunk?
Thanks,
Martin
>From 5681b55f531f579ba75aad21f5628f86fba4bc8a Mon Sep 17 00:00:00 2001
From: mliska
Date: Wed, 25 Mar 2015 10:09:21 +0100
Subject: [PATCH] Fix vimrc file link creation.
Change
On Wed, Mar 25, 2015 at 11:57:08AM +0100, Martin Liška wrote:
> Following patch correctly creates symlink that now points to a wrong location.
Only if $(srcdir) is a relative path I'd say.
> Ready for trunk?
In any case, LGTM.
> 2015-03-25 Martin Liska
> Yury Gribov
>
> *
Avoids a fixup-cfg and cgraph edge rebuild.
Bootstrapped and tested on x86_64-unknown-linux-gnu, applied.
Richard.
2015-03-25 Richard Biener
* passes.c (pass_manager::execute_early_local_passes): Guard
execution of pass_chkp_instrumentation_passes with
flag_check_poi
Hi,
ChangeLog
2015-03-25 Kai Tietz
PR libgomp/64972
* oacc-parallel.c (GOACC_parallel): Use PRIu64 if available.
(GOACC_data_start): Likewise.
* target.c (gomp_map_vars): Likewise.
Tested for i686-w64-mingw32. Fix got preapproved by Jakub, so I will
commit this soon, if ther
Hi,
This patch fixes PR target/65508 by proper copy of static chain for
instrumented calls. Bootstrapped and tested on x86_64-unknown-linux-gnu. OK
for trunk or wait for stage 1?
Thanks,
Ilya
--
gcc/
2015-03-25 Ilya Enkovich
PR target/65508
* tree-chkp.c (chkp_add_bounds_
On Wed, Mar 25, 2015 at 1:35 PM, Ilya Enkovich wrote:
> Hi,
>
> This patch fixes PR target/65508 by proper copy of static chain for
> instrumented calls. Bootstrapped and tested on x86_64-unknown-linux-gnu. OK
> for trunk or wait for stage 1?
Ok for trunk.
Richard.
> Thanks,
> Ilya
> --
> g
Committed.
Richard.
2015-03-25 Richard Biener
PR tree-optimization/62630
* gcc.dg/graphite/vect-pr43423.c: XFAIL.
Index: gcc/testsuite/gcc.dg/graphite/vect-pr43423.c
===
--- gcc/testsuite/gcc.dg/graphite/vect-pr
Hi all,
please find attached a fix for the recently introduced regression when
allocating arrays with an intrinsic function for source=. The patch addresses
this issue by using gfc_conv_expr_descriptor () for intrinsic functions.
Bootstraps and regtests ok on x86_64-linux-gnu/F20.
Ok for trunk?
The following fixes an ICE on invalid code by checking that DECL_SIZE is
not null before feeding it to integer_zerop.
Bootstrapped/regtested on x86_64-linux, ok for trunk?
2015-03-25 Marek Polacek
PR c++/61670
* class.c (remove_zero_width_bit_fields): Check for null DECL_SIZE.
On 03/19/15 13:54, Sebastian Pop wrote:
Richard Biener wrote:
>please instead fixup after copy_bbs in duplicate_seme_region.
>
Thanks for the review.
Attached patch that does not modify copy_bbs.
Fixes make check in hmmer and make check RUNTESTFLAGS=tree-ssa.exp
Full bootstrap and regtest in p
On 03/18/15 19:40, Honggyu Kim wrote:
Hi,
I have modified the test-case to check parameter passing bug based on the
comments from Kyrill Tkachov, Christophe Lyon, and Segher Boessenkool
as follows:
1. move from "gcc.target/arm" to "gcc.dg"
2. change "dg-do compile" to "dg-do run"
Please let
> -Original Message-
> From: Jeff Law [mailto:l...@redhat.com]
> Sent: 25 March 2015 12:27
> To: Honggyu Kim; gcc-patches@gcc.gnu.org
> Cc: Kyrylo Tkachov; seg...@kernel.crashing.org; christophe.l...@st.com
> Subject: Re: [PATCH v2] New testcase to check parameter passing bug
>
> On 03/1
On Wed, Mar 25, 2015 at 01:35:02PM +0100, Kai Tietz wrote:
> ChangeLog
>
> 2015-03-25 Kai Tietz
>
> PR libgomp/64972
> * oacc-parallel.c (GOACC_parallel): Use PRIu64 if available.
> (GOACC_data_start): Likewise.
> * target.c (gomp_map_vars): Likewise.
>
> Tested for i686-w64-m
On Tue, Mar 24, 2015 at 05:43:09PM +0100, Uros Bizjak wrote:
> Attached patch introduces bootstrap-lto-noplugin bootstrap
> configuration for hosts that do not support linker plugin (e.g. CentOS
> 5.11 with binutils 2.17). Also, the patch adds some additional
> documentation to bootstrap-lto option
When cross-testing, the -DITERATIONS=1000 flag replaced the -pthread required
for linux targets, so the test failed to build. I've pushed the following test
fix as r221666:
Index: libstdc++-v3/testsuite/21_strings/basic_string/pthread33394.cc
On 25/03/15 15:49 +, Alan Lawrence wrote:
When cross-testing, the -DITERATIONS=1000 flag replaced the -pthread
required for linux targets, so the test failed to build. I've pushed
the following test fix as r221666:
Ah yes, of course it does! Thanks for the fix.
Index: libstdc++-v3/testsu
On 18/02/15 12:15 +, Jonathan Wakely wrote:
On 12/02/15 13:23 -0800, Richard Henderson wrote:
When we fixed PR54005, making sure that atomic_is_lock_free returns the same
value for all objects of a given type, we probably should have changed the
interface so that we would pass size and align
Hi,
The POWER-specific little-endian swap optimization pass has been burning
in on mainline since last August. Since then there have been a few
improvements and bug fixes, but the code is very stable. I've had some
recent requests to get this code backported to 4.9, as it provides
important perf
Hi Dominique, hi all,
you are absolutely right, Dominique: I missed the part of pr60322_base_*.
But this time it is there and furthermore does solve the allocate( mold=e) and
the loc(e) issue.
Paul: I have simplified your patch by only checking whether the
arg_expr.ts.type == BT_CLASS. All tes
On Tue, Mar 24, 2015 at 05:46:57PM +, Alan Lawrence wrote:
> Hmmm. This is not the right fix: the tests Richard fixed, were failing because
> of lack of constant propagation and DCE at compile-time, which then didn't
> eliminate the call to link_error. The AArch64 test is failing because this
On 03/25/2015 09:22 AM, Jonathan Wakely wrote:
> private:
> - _Tp _M_i;
> + // Align 1/2/4/8/16-byte types the same as integer types of that size.
> + // This matches the alignment effects of the C11 _Atomic qualifier.
> + static constexpr int _S_alignment
> + = sizeof(
Hi!
As discussed in the PR, fixing this issue for real (make sure we at least
until the objsz pass don't lose information on which field's address if any
has been taken) is probably too dangerous at this point, so this patch
just adds a simple workaround:
another objsz pass instance run early befo
On 03/25/2015 09:22 AM, Jonathan Wakely wrote:
> +static_assert( alignof(std::atomic) > alignof(int),
> + "std::atomic not suitably aligned" );
This is only true if int64_t has alignment larger than int32_t,
which is unfortunately not always the case.
r~
On 25/03/15 11:36 -0700, Richard Henderson wrote:
On 03/25/2015 09:22 AM, Jonathan Wakely wrote:
private:
- _Tp _M_i;
+ // Align 1/2/4/8/16-byte types the same as integer types of that size.
+ // This matches the alignment effects of the C11 _Atomic qualifier.
+ static c
As discussed in the PR, the abi_tag on an anonymous namespace is useless,
but we shouldn't ICE if the user attempts to do that.
Bootstrapped/regtested on x86_64-linux, ok for trunk?
2015-03-25 Marek Polacek
PR c++/65558
* name-lookup.c (handle_namespace_attrs): Ignore abi_tag
On 03/25/2015 11:49 AM, Jonathan Wakely wrote:
> On 25/03/15 11:36 -0700, Richard Henderson wrote:
>> On 03/25/2015 09:22 AM, Jonathan Wakely wrote:
> On 25/03/15 11:39 -0700, Richard Henderson wrote:
>> On 03/25/2015 09:22 AM, Jonathan Wakely wrote:
>>> +static_assert( alignof(std::atomic) > align
On Wed, Mar 25, 2015 at 3:23 PM, Jakub Jelinek wrote:
>> Attached patch introduces bootstrap-lto-noplugin bootstrap
>> configuration for hosts that do not support linker plugin (e.g. CentOS
>> 5.11 with binutils 2.17). Also, the patch adds some additional
>> documentation to bootstrap-lto option.
OK.
Jason
OK.
Jason
On 03/24/2015 02:00 PM, Aldy Hernandez wrote:
I found that for locally scoped functions we were not emitting early
dwarf.
Why weren't they being emitted as part of their enclosing function?
They should be.
Jason
On 03/25/2015 12:37 PM, Jason Merrill wrote:
On 03/24/2015 02:00 PM, Aldy Hernandez wrote:
I found that for locally scoped functions we were not emitting early
dwarf.
Why weren't they being emitted as part of their enclosing function? They
should be.
Jason
Hmm, you're right. Sorry for bei
PR 65570 points out that the recent patch to the go tool breaks the
use of cgo (and obviously also points out that we need better testing
for go and cgo). The problem is that the go tool treats the
runtime/cgo package specially. Although gccgo doesn't use that
package, the go tool needs to know t
On Wed, Mar 25, 2015 at 12:42 PM, Bill Schmidt
wrote:
> Hi,
>
> The POWER-specific little-endian swap optimization pass has been burning
> in on mainline since last August. Since then there have been a few
> improvements and bug fixes, but the code is very stable. I've had some
> recent requests
Hello!
For some reason gcc.target/i386/sse-13.c lost its #include
. Attached patch fixes this issue and adjusts
corresponding #defines. The patch also removes extra #includes from
sse-23.c.
2015-03-25 Uros Bizjak
* gcc.target/i386/sse-13.c: Include x86intrin.h and adjust #defines.
*
On Wed, 2015-03-25 at 17:56 -0400, David Edelsohn wrote:
> On Wed, Mar 25, 2015 at 12:42 PM, Bill Schmidt
> wrote:
> > Hi,
> >
> > The POWER-specific little-endian swap optimization pass has been burning
> > in on mainline since last August. Since then there have been a few
> > improvements and b
Hi,
linemap is optimized for situation where parser enters positions into it in
source order.
LTO does not work this way - it attach locations to trees and reads them more
or less
randomly. This results in large memory use of linemaps, slow lookups (that are
critical
for WPA stremaing) and as i
The attached patch adds tests to lib/target-supports.exp
to avoid unnecessarily invoking the compiler on non-ARM
targets to check for the support for a number of ARM
vectorization features.
Okay to commit to trunk?
Martin
2015-03-23 Martin Sebor
* lib/target-supports.exp (check_effective_ta
On Wed, Mar 25, 2015 at 05:04:32PM -0600, Martin Sebor wrote:
> The attached patch adds tests to lib/target-supports.exp
> to avoid unnecessarily invoking the compiler on non-ARM
> targets to check for the support for a number of ARM
> vectorization features.
>
> Okay to commit to trunk?
>
> Mart
Jeff Law wrote:
> > PR tree-optimization/65177
> > * tree-ssa-threadupdate.c (verify_seme): Renamed verify_jump_thread.
> > (bb_in_bbs): New.
> > (duplicate_seme_region): Renamed duplicate_thread_path. Redirect all
> > edges not adjacent on the path to the original code.
> OK f
Pat Haugen runs a spec regression tester on various PowerPC boxes, and he
noticed that my fix for PR 65240 (the bug involving floating point constants
and -ffast-math under VSX) caused a regression in building the dealII benchmark
on power6x. I looked into it, and discovered I had missed extenddft
This patch from Dominik Vogt fixes some s390 failures in libgo. Ran
Go testsuite on x86_64-unknown-linux-gnu. Committed to mainline.
Ian
diff -r bdce421e579e libgo/go/runtime/chan_test.go
--- a/libgo/go/runtime/chan_test.go Wed Mar 25 14:16:52 2015 -0700
+++ b/libgo/go/runtime/chan_test.go
Hi,
I read linemap_line_start and I think I noticed few issues with respect
to overflows and lines being added randomly.
1) line_delta is computed as to_line SOURCE_LINE (map, set->highest_line)
I think the last inserted line is not very releavnt. What we care about is
the base of the last
> Hello!
>
> Attached patch introduces bootstrap-lto-noplugin bootstrap
> configuration for hosts that do not support linker plugin (e.g. CentOS
> 5.11 with binutils 2.17). Also, the patch adds some additional
> documentation to bootstrap-lto option.
>
> config/ChangeLog:
>
> 2015-03-24 Uros Bi
On 03/25/2015 05:05 PM, Aldy Hernandez wrote:
Or we could cheat and just remove them as mainline does, but only when
reusing a declaration (as in the attached patch).
This seems right to me.
Jason
Committed as obvious and simple.
revision 221682.
Regards,
Jerry
2015-03-25 Jerry DeLisle
PR libgfortran/65541
* io/write.c (nml_write_obj): Convert '+' to '%' before emitting
object names in namelists.
Index: io/write.c
=
On Wed, Mar 25, 2015 at 8:09 PM, Michael Meissner
wrote:
> Pat Haugen runs a spec regression tester on various PowerPC boxes, and he
> noticed that my fix for PR 65240 (the bug involving floating point constants
> and -ffast-math under VSX) caused a regression in building the dealII
> benchmark
>
Jan Hubicka writes:
>
> Bootstrapped/regtested x86_64-linux, the patch saves about 1GB of locators
> for chromium
> and 400MB for firefox LTO.
Great. On my LTO builds linemap was always high up in the profiles too.
-Andi
--
a...@linux.intel.com -- Speaking for myself only
> Jan Hubicka writes:
> >
> > Bootstrapped/regtested x86_64-linux, the patch saves about 1GB of locators
> > for chromium
> > and 400MB for firefox LTO.
>
> Great. On my LTO builds linemap was always high up in the profiles too.
Yep, these was always high. I am re-running some profiles now. I
Hi,
this patch (as suggested by Richard) adds very simple discovery of
DECL_NOTHORW to build_ssa passes.
The reason is that in 4.9 we did build_ssa in parallel with early optimization
that does nothrow discovery as part of local pure const. Bounds checking
patches broke the pass queue into multip
On 03/19/2015 09:01 AM, Marat Zakirov wrote:
On 03/04/2015 11:07 AM, Andrew Pinski wrote:
On Wed, Mar 4, 2015 at 12:00 AM, Marat Zakirov
wrote:
Hi all!
Here is the patch which forces ASan to work on memory access without
proper
alignment. it's useful because some programs like linux kerne
67 matches
Mail list logo
