Just committed the following:
* MAINTAINERS (Write After Approval): Add myself.
Index: MAINTAINERS
===
--- MAINTAINERS (revision 181466)
+++ MAINTAINERS (working copy)
@@ -419,6 +419,7 @@
Marc Lehmann
On Wed, Oct 12, 2011 at 7:01 PM, Diego Novillo wrote:
>
> On 11-10-10 17:47 , Sandeep Soni wrote:
>>
>> Hi,
>> The following patch is a basic attempt to build a symbol table that
>> stores the names of all the declarations made in the input file.
>>
>> Index: gcc/gimple/parser.c
>> ===
Latest results for 4.6.x.
This replaces the previous unapplied patches for the buildstat page.
-tgc
Testresults for 4.6.2:
arm-unknown-rtemseabi4.11 (2)
hppa2.0w-hp-hpux11.11
hppa64-hp-hpux11.11
i386-pc-solaris2.8
i686-apple-darwin9 (2)
i686-pc-linux-gnu (2)
powerpc-apple-darwin8.11.0
po
To raise awareness, a track at http://gcc.gnu.org/bugzilla/show_bug.cgi?id=51200
- Joey
> > >
> > > The current x86 memset/memcpy expansion is broken. It miscompiles
> > > many programs, including GCC itself. Should it be reverted for now?
> >
> > There was problem in the new code doing loopy epilogues.
> > I am currently testing the following patch that shold fix the problem.
> >
On Thu, Nov 17, 2011 at 12:24:08PM -1000, Richard Henderson wrote:
> On 11/17/2011 10:25 AM, Richard Henderson wrote:
> > so that we don't write back into the subreg.
>
> Like so. Tested on ppc64-linux and committed.
Thanks! This, along with my mutex.h rewrite plus adding a
memory_barrier insn
On Thu, Nov 17, 2011 at 02:56:31PM -1000, Richard Henderson wrote:
> On 11/17/2011 01:56 AM, Alan Modra wrote:
> > This is part of work in progress getting locking back into shape on
> > powerpc. (If we ever were in shape, which is doubtful.) Using the
> > ia64 version means we have a needless sy
Ping
On Thu, Nov 10, 2011 at 11:57 AM, Joey Ye wrote:
> Trunk gcc mis-handles following volatile bitfield case on ARM target:
>
> $ cat a.c
> extern void check(int);
> typedef struct {
> volatile unsigned short a:8, b:8;
> } BitStruct;
> BitStruct bits = {1, 2};
> int main ()
> {
> check(bits.a
On 11/17/2011 01:56 AM, Alan Modra wrote:
> This is part of work in progress getting locking back into shape on
> powerpc. (If we ever were in shape, which is doubtful.) Using the
> ia64 version means we have a needless sync at the start of
> gomp_mutex_lock (courtesy of __sync_val_compare_and_sw
* testsuite/30_threads/thread/native_handle/typesizes.cc: Do not run
on darwin.
This fixes a FAIL on darwin:
http://gcc.gnu.org/ml/gcc-testresults/2011-11/msg01606.html
It passes (by design) on GNU/Linux but there's no reason to assume or
require that this test passes elsewhere.
I
Hi,
On 11/17/2011 09:52 PM, Paolo Carlini wrote:
On 11/17/2011 09:37 PM, Jason Merrill wrote:
On 11/16/2011 10:00 AM, Paolo Carlini wrote:
this is an ICE on valid, 4.6/4.7 Regression, which manifests itself as
tree codes unhandled by cxx_eval_constant_expression.
Hmm, those codes shouldn't m
>From f63ec02c0a720174489fe450b3cc43eb00fd4bdd Mon Sep 17 00:00:00 2001
From: Harshit Chopra
Date: Thu, 3 Nov 2011 17:29:23 -0700
Subject: [PATCH] Mechanism to provide efficient function instrumentation
(x86-64)
Summary: This patch aims at providing an efficient way to instrument a binary
at fu
I think I was a bit too hasty on this subject.
I am bootstrapping and testing the (IMHO better) patch below on
x86_64-unknown-linux-gnu against trunk.
Sorry for the noise.
From: Dodji Seketeli
Date: Thu, 17 Nov 2011 19:07:58 +0100
Subject: [PATCH] PR c++/51191 - ICE on alias of alias template i
Simply ignoring the possibility of reference counting seems to match up with
corresponding code in eh_throw.cc __gxx_exception_cleanup?
Can this possibly be correct?
r~
diff --git a/libstdc++-v3/libsupc++/eh_tm.cc b/libstdc++-v3/libsupc++/eh_tm.cc
index f5815a1..a646102 100644
--- a/libstdc++-v
On 11/17/2011 10:25 AM, Richard Henderson wrote:
> On 11/17/2011 12:37 AM, Alan Modra wrote:
>> - oldval = convert_modes (SImode, mode, oldval, 1);
>> + oldval = gen_reg_rtx (SImode);
>> + convert_move (oldval, orig, 1);
>>oldval = expand_simple_binop (SImode, ASHIFT, oldval,
Hi Richard,
When you have time to review the patch in detail, you might prefer to
look at this one instead. I've been working on the stage2 code in the
meanwhile, and ended up factoring a bunch of the stage1 code for reuse.
I applied the same factorization to the stage1 code so that eventually
it
Greetings,
While investigating http://gcc.gnu.org/bugzilla/show_bug.cgi?id=39976, I
found that tree-outof-ssa.c:insert_backedge_copies tests whether edges
are back edges, but the back edges haven't been marked to facilitate
this. (This was the reason the failures started happening in r149206,
alt
> >
> > The current x86 memset/memcpy expansion is broken. It miscompiles
> > many programs, including GCC itself. Should it be reverted for now?
>
> There was problem in the new code doing loopy epilogues.
> I am currently testing the following patch that shold fix the problem.
> We could eithe
The zero_or_v7_operand hack could alternately be handled by
forcing MO_SC for v7 in sparc_options_override.
---
gcc/config/sparc/predicates.md |4
gcc/config/sparc/sparc.c | 14 +-
gcc/config/sparc/sync.md | 15 +++
3 files changed, 28 insertions(+)
---
gcc/config/sparc/sync.md |1 +
1 files changed, 1 insertions(+), 0 deletions(-)
diff --git a/gcc/config/sparc/sync.md b/gcc/config/sparc/sync.md
index 782aa60..821d59e 100644
--- a/gcc/config/sparc/sync.md
+++ b/gcc/config/sparc/sync.md
@@ -229,6 +229,7 @@
FAIL;
if (mode != QImode
... of course, I immediately found two problems by inspection.
(1) Accidentally dropped the barrier for sync_test_and_set when
used on post-v7. Given that we have only SImode atomic_exchange,
and we do have QImode sync_test_and_set, we will probably try to
use LDSTUB for the c++ atomic_bool_test_
> Eric, this is just something I noticed while trying to fix the
> vec_init problems last week.
>
> I'm confident that the issue is real, however I can't point to any
> real bugs that are caused by this.
>
> Therefore I'm reluctant to commit this change.
>
> What do you think?
I'd install it, this
Hello,
In the example of the patch below the two U parameters of template C
and D compare equal, so the two A types also compare equal.
When we lookup up A inside struct D, lookup_template_class_1
actually yields the A that is inside struct C.
This is a problem because when substituting the argum
>
> The current x86 memset/memcpy expansion is broken. It miscompiles
> many programs, including GCC itself. Should it be reverted for now?
There was problem in the new code doing loopy epilogues.
I am currently testing the following patch that shold fix the problem.
We could either revert now a
> Dave and Eric, can you please double-check everything and also test
> on Solaris (i.e. RMO).
Yes, I will, but note that Solaris is TSO. The only RMO case was Linux AFAIK.
--
Eric Botcazou
Hello,
In the example of the patch below, we crash when trying to print the
alias of alias template instantiation 'Alias'. This is because
alias_template_specialization_p carelessly looks at
CLASSTYPE_TEMPLATE_INFO first, while TYPE_TEMPLATE_INFO is NULL in
this case.
Fixed thus, bootstrapped an
Let's give a more helpful message in this case.
Tested x86_64-pc-linux-gnu, applying to trunk.
commit 303f975019af8e6048120b1d27a5f46cbf0db27d
Author: Jason Merrill
Date: Thu Nov 17 15:19:20 2011 -0500
PR c++/51186
* decl.c (grokdeclarator): Improve C++98 trailing return diagnostic.
On 11/17/2011 09:37 PM, Jason Merrill wrote:
On 11/16/2011 10:00 AM, Paolo Carlini wrote:
this is an ICE on valid, 4.6/4.7 Regression, which manifests itself as
tree codes unhandled by cxx_eval_constant_expression.
Hmm, those codes shouldn't make it this far; we should go through
fold_non_dep
---
gcc/config/sparc/sparc-opts.h | 10
gcc/config/sparc/sparc.c | 48 -
gcc/config/sparc/sparc.opt| 22 ++
3 files changed, 79 insertions(+), 1 deletions(-)
diff --git a/gcc/config/sparc/sparc-opts.h b/gcc/config/spa
Aid the backend with mem_noofs_operand and a constraint to match.
---
gcc/config/sparc/constraints.md |5 +++-
gcc/config/sparc/predicates.md |4 +++
gcc/config/sparc/sparc-protos.h |2 +-
gcc/config/sparc/sparc.c| 56 ---
gcc/config/sparc
---
gcc/config/sparc/sync.md | 91 +++---
1 files changed, 30 insertions(+), 61 deletions(-)
diff --git a/gcc/config/sparc/sync.md b/gcc/config/sparc/sync.md
index d38a828..782aa60 100644
--- a/gcc/config/sparc/sync.md
+++ b/gcc/config/sparc/sync.md
@@ -1
---
gcc/config/sparc/predicates.md |5 +++
gcc/config/sparc/sparc.md |1 +
gcc/config/sparc/sync.md | 59
3 files changed, 65 insertions(+), 0 deletions(-)
diff --git a/gcc/config/sparc/predicates.md b/gcc/config/sparc/predicates.md
in
---
gcc/config/sparc/sparc-protos.h |2 +
gcc/config/sparc/sparc.c| 45 +++
gcc/config/sparc/sync.md| 55 --
3 files changed, 93 insertions(+), 9 deletions(-)
diff --git a/gcc/config/sparc/sparc-protos.h b/gcc
The first four patches simply do the conversion, a piece at a time,
assuming the RMO for all cpus.
The fifth patch adds the ability to explicitly set the memory model
for the program, and to adjust the barriers emitted based on that
memory model. If we agree on the spelling of that option (3 m's
On 11/16/2011 10:00 AM, Paolo Carlini wrote:
this is an ICE on valid, 4.6/4.7 Regression, which manifests itself as
tree codes unhandled by cxx_eval_constant_expression.
Hmm, those codes shouldn't make it this far; we should go through
fold_non_dependent_expr before we get to cxx_eval_constant
> the following is reg-strapping on x86-64-darwin10,
Thanks a lot!
> OK for trunk (and 4.6) if it succeeds (with a ChangeLog, of course) ?
Yes, modulo:
> +static void
> +__gnat_error_handler (int sig, siginfo_t *si, void *ucontext
> ATTRIBUTE_UNUSED)
Remove ATTRIBUTE_UNUSED and...
> +#if defi
On 11/17/2011 02:33 PM, Mike Stump wrote:
The patch also sets timeout to 10 seconds, due to huge amount of
produced log and improves fail messages a bit.
10 might be a bit low... (slow targets, simulators, complex remote target
setup) but we can go with that for now until someone has a nee
Hello!
Currently, bootstrap crashes with --enable-stage1-checking=all with:
../../../libgcc/libgcc2.c:553:3: internal compiler error: tree check: expected
tree that contains ‘typed’ structure, have ‘block’ in fold_checksum_tree, at
fold-const.c:14160
The problem is in fold_checksum_tree function
On 11/17/2011 02:29 PM, Steve Ellcey wrote:
Testing on IA64 HP-UX shows that this fixes the bootstrap failure and
it found no regressions. Presumably it does slow down the compilation
slightly since the point of fprint_w was to be faster then fprintf.
If someone (Dimitrios?) wants to fix fprint
Coincidentally, IBM's XL compiler is encountering a similar issue and
proposing a similar solution of having the stack pointer update
conflict with all memory accesses. I think all these corner cases
confirm that it is impractical for the compiler to track all accesses
against SP and the only expe
On Nov 17, 2011, at 1:54 AM, Uros Bizjak wrote:
> The problem was, that dejagnu didn't kill gdb in case of timeout (i.e.
> broken LL/SC sequences for alpha).
> The patch also sets timeout to 10 seconds, due to huge amount of
> produced log and improves fail messages a bit.
10 might be a bit low..
The checkin r181279 which speeds up the printing of integer constants
caused PR 51144 and broke the IA64 and s390 bootstraps.
The problem is with fprint_w which takes a HOST_WIDE_INT value argument
and uses sprint_ul_rev to process it. But sprint_ul_rev takes a long
argument so if the argument t
On Tue, 2011-11-15 at 08:53 -1000, Richard Henderson wrote:
> This is relatively straight-forward, given that most of the
> language actually matches up with the opcodes. ;-)
>
> As mentioned in the patch itself, this is based on the data
> presented in
>
>http://www.cl.cam.ac.uk/~pes20/cpp
On 12 Nov 2011, at 07:44, Eric Botcazou wrote:
This has been filed as radar #10302855, but we need a work-around
until that is resolved (possibly forever on older systems).
OK for trunk?
(what opinion about 4.6?)
Did you apply it to the 4.6 branch? I think that this would be
appropriate.
> Here comes the 2nd version patch modified according to your comments. Is it
> ok?
Yes, modulo formatting nits (double space after period, operator on the next
line when the entire expression doesn't fit on a single line).
> Also could you please commit it if ok because I have no write access?
On 17/11/2011, at 9:58 PM, Tom de Vries wrote:
> On 11/15/2011 10:07 PM, Maxim Kuvyrkov wrote:
>> On 30/10/2011, at 8:17 AM, Tom de Vries wrote:
>>
>>> Richard,
>>>
>>> I have a tentative fix for PR50764.
>>
>> Richard,
>>
>> Tom's patch is good (with the comments below addressed), and I would
> I understand your concern and the patch indeed will have an impact on the
> generated code of probably several back-ends. But with entering stage 3 the
> backend maintainers anyway are supposed to have a closer look at
> regressions.
Yes, that's why risking to introduce more of them isn't the be
On Mon, Nov 14, 2011 at 3:48 PM, H.J. Lu wrote:
> On Mon, Nov 14, 2011 at 9:03 AM, Jan Hubicka wrote:
>> Hi,
>> this is hopefully final variant of patch. The epilogue code was broken in
>> some
>> scenarios for memset, but should work safely now. I also fixed the tables
>> for
>> core/buldozer
As things stand, Darwin's 'as' cannot produce dwarf debug.
Unfortunately, if "-g" is given to -x assembler-with-cpp ... this
results in 'as' being called with --stabs.
So, it can be the case that an exe ends up with some stabs debug in it
- usually with a source file of "{standard input}"
At the Batavia meeting last year, the C++ committee changed the
conditions for implicitly declaring move constructors/assignment ops;
now they are suppressed by having any of (copy ctor/op=, move ctor/op=,
dtor). I needed to make some adjustments to the libstdc++ testsuite as
a result; bkoz, d
On 11/17/2011 04:50 AM, David Edelsohn wrote:
> On Thu, Nov 17, 2011 at 6:07 AM, Alan Modra wrote:
>> Lacking this pattern means the builtin __sync_synchronize() on powerpc
>> is just an asm with a memory clobber (see builtins.c), which is hardly
>> a "full memory barrier" as extend.texi says it s
My patch for 51029 fixed some virtual calls under
fold_non_dependent_expr, but clearly not all. This patch goes farther.
Tested x86_64-pc-linux-gnu, applying to trunk.
commit 9c6ea6fd08c5c99ba4455b88bb8d0b0b6a71d171
Author: Jason Merrill
Date: Tue Nov 15 14:27:04 2011 -0500
PR c++/5113
Hi!
Please find attached the patch updating NEG_POOL_RANGE from 1008 to
1020 -(8 + ).
Tested with check-gcc with no regression.
The test-case failing for patch 'STRD generation instead of PUSH in A15
ARM prologue' (http://gcc.gnu.org/ml/gcc-patches/2011-11/msg01158.html)
passes with this fix.
g
mmary ===
# of expected passes108
# of unsupported tests 22
/Volumes/ScratchCS/gcc-4-7-trunk-build/gcc/xgcc version 4.7.0
2017 (experimental) [trunk revision 181436] (GCC)
(I haven't finished the m64 gdb port yet, so can't try that at the
moment).
:-)
Iain
On Thu, Nov 17, 2011 at 6:07 AM, Alan Modra wrote:
> Lacking this pattern means the builtin __sync_synchronize() on powerpc
> is just an asm with a memory clobber (see builtins.c), which is hardly
> a "full memory barrier" as extend.texi says it should be. This patch
> fixes multiple libgomp test
On Thu, Nov 17, 2011 at 2:22 PM, Iain Sandoe
wrote:
>
> On 17 Nov 2011, at 10:49, Alan Modra wrote:
>
>> On Thu, Nov 17, 2011 at 10:54:23AM +0100, Uros Bizjak wrote:
>>>
>>> The solution for this problem is to search for "simulate_thread_done"
>>> in gdb log.
>>
>> Thanks! I like this for another
On Tue, 15 Nov 2011, Andrey Belevantsev wrote:
> > diff --git a/gcc/sel-sched.c b/gcc/sel-sched.c
> > index f11faca..91fb0fe 100644
> > --- a/gcc/sel-sched.c
> > +++ b/gcc/sel-sched.c
> > @@ -5234,6 +5234,7 @@ move_exprs_to_boundary (bnd_t bnd, expr_t expr_vliw,
> >
> > b = move_op (BND_TO (b
On 17 Nov 2011, at 09:25, Andreas Krebbel wrote:
On 11/17/2011 03:44 AM, David Edelsohn wrote:
Andreas,
This patch seems to have introduced a failure for all of the
gcc.dg-struct-layout tests on AIX.
gcc.dg-struct-layout-1/t001_test.h:8:1: internal compiler error: in
int_mode_for_mode, at st
On 17 Nov 2011, at 10:49, Alan Modra wrote:
On Thu, Nov 17, 2011 at 10:54:23AM +0100, Uros Bizjak wrote:
The solution for this problem is to search for "simulate_thread_done"
in gdb log.
Thanks! I like this for another reason too: I've seen these tests
fail due to gdb never reporting simul
On 11/16/2011 06:53 AM, Jeff Law wrote:
> I'd ask the meta question, is there a compelling reason to push this
> patch into the tree now? My obvious concern is that this change
> potentially effects every target and twiddles one of GCC's most
> sensitive areas. Even if the patch is sound it's the
Andrew Pinski writes:
> On Mon, Nov 14, 2011 at 3:41 PM, Andrew Pinski
> wrote:
>> Since Linux 2.6.36, both the heap and mmap are randomized on MIPS like
>> other architectures. This breaks PCH because currently
>> TRY_EMPTY_VM_SPACE is not defined for MIPS.
>> This patch fixes the issue and allo
Hello Joseph,
Thanks for your comments. When start to work on Multilib for ARM backend, I
do run into some limitations of GCC Multilib mechanism. The most important
one is just as what you said in thread
http://gcc.gnu.org/ml/gcc/2010-01/msg00063.html, e.g. the limitation of
MULTILIB_MATCHES. So I
This is part of work in progress getting locking back into shape on
powerpc. (If we ever were in shape, which is doubtful.) Using the
ia64 version means we have a needless sync at the start of
gomp_mutex_lock (courtesy of __sync_val_compare_and_swap) and a
needless isync at the end of gomp_mutex_
The new version of the patch as suggested by Tristan is OK to commit,
thanks.
Arno
Lacking this pattern means the builtin __sync_synchronize() on powerpc
is just an asm with a memory clobber (see builtins.c), which is hardly
a "full memory barrier" as extend.texi says it should be. This patch
fixes multiple libgomp testsuite failures. Bootstrapped and
regression tested powerpc-
On Thu, Nov 17, 2011 at 10:54:23AM +0100, Uros Bizjak wrote:
> The solution for this problem is to search for "simulate_thread_done"
> in gdb log.
Thanks! I like this for another reason too: I've seen these tests
fail due to gdb never reporting simulate_thread_main on any
instruction in that fun
The testcase c-c++-common/tm/omp.c fails on arm-none-eabi:
Executing: arm-none-eabi-gcc
/work/local-checkouts/gcc-fsf/gcc/testsuite/c-c++-common/tm/omp.c -fgnu-tm
-fopenmp -S -o omp.s
arm-none-eabi-gcc: error: unrecognized command line option '-pthread'
FAIL: c-c++-common/tm/omp.c
The attached
* doc/xml/manual/status_cxx2011.xml: Status of piecewise construction
and ios_base::failure.
* doc/xml/manual/backwards_compatibility.xml: List headers in
alphabetical order.
committed to trunk
Index: doc/xml/manual/backwards_compatibility.xml
==
Hello!
The problem was, that dejagnu didn't kill gdb in case of timeout (i.e.
broken LL/SC sequences for alpha). The cause for this was in the way
remote_expect was called. The script found "simulate_thread_main" to
assume that gdb works OK, but unfortunately, this overrode the
"timeout" default c
On 11/17/2011 03:44 AM, David Edelsohn wrote:
> Andreas,
>
> This patch seems to have introduced a failure for all of the
> gcc.dg-struct-layout tests on AIX.
>
> gcc.dg-struct-layout-1/t001_test.h:8:1: internal compiler error: in
> int_mode_for_mode, at stor-layout.c:424
>
> After your change,
On 11/15/2011 10:07 PM, Maxim Kuvyrkov wrote:
> On 30/10/2011, at 8:17 AM, Tom de Vries wrote:
>
>> Richard,
>>
>> I have a tentative fix for PR50764.
>
> Richard,
>
> Tom's patch is good (with the comments below addressed), and I would
> appreciate you validating my review with your formal app
71 matches
Mail list logo