This is great to see that there is some activity around the trans-mem
branch.
+@subsection [New] Transactional dynamic memory management
Remark: I think this part of documentation may also mention new and delete
operators.
Patrick Marlier.
On Thu, 26 May 2011, Torvald Riegel wrote:
Patch
On Thu, May 26, 2011 at 11:56 PM, Andrew Pinski wrote:
> The problem here is that vec.c accesses VEC__base via a
> vec_prefix but VEC__base's first field is not a vec_prefix.
> This patch fixes the problem by changing how VEC_*_base is done by
> using vec_prefix as the real base.
>
> OK? Bootstrap
On Fri, May 27, 2011 at 12:12 AM, Janis Johnson
wrote:
> On 05/25/2011 05:15 AM, Jan Hubicka wrote:
>> Am Wed 25 May 2011 11:04:06 AM CEST schrieb Richard Guenther
>> :
>>
>>> On Wed, May 25, 2011 at 10:38 AM, Rainer Orth
>>> wrote:
Janis Johnson writes:
> Archived test results for
On 05/27/2011 02:32 AM, Jack Howarth wrote:
> This patch breaks the bootstrap on all darwin targets...
>
> http://gcc.gnu.org/regtest/HEAD/native-lastbuild.txt.gzip
>
> make[3]: *** No rule to make target
> `/Users/regress/tbox/svn-gcc/libgcc/../gcc/libgcc-std.ver', needed by
> `libgcc.map'. S
When regrename is enabled, it makes new copies of its replacement
registers, which means that pointer equality tests can't be used
afterwards to check whether two regs are equal.
sched_insn_conditions_mutex_p has one such test, and it causes
pessimized code in some situations on C6X.
The following
On 20 May 2011 17:32, Jason Merrill wrote:
> G++ has had a long-standing bug with unqualified name resolution in
> templates: if we didn't find any declaration when looking up a name in the
> template definition, we would do an additional unqualified lookup at the
> point of instantiation. This le
Hello,
Recent changes to dwarf information adds call_arg_location notes just
after the call instruction. They need to follow the call instruction
strictly. But, this means that they could occur in the middle of a VLIW
instruction. Our assembler currently can not handle labels in the middle
of
Hi,
On 05/26/2011 02:30 PM, Paolo Carlini wrote:
We want to do that only when processing a template, because otherwise we
get a duplicate diagnostic, see, eg, auto9.C
Hmm, where's the error coming from in the non-template case? From
cp_build_c_cast? In that case always giving the error in
bu
Hi,
here is updated patch incorporating the feedback. I now use variant of uleb
format,
just do 4 bit chunks instead of 8bit. It works better in my tests and is also
used
by LLVM so it must be cool ;)
I also fixed off by one error in filename streaming. While debugging this I
noticed that bitpa
Hi,
That only shows up if you configure with --enable-libstcxx-time
I'll finish reg-testing it and submit it when I get home this evening.
thanks Jon. Looks like we have also to fix parallel-mode (49187), I hope
there isn't too much to reshuffle, if you spot something...
Thanks again,
Paolo.
On 05/27/2011 11:05 AM, Paolo Carlini wrote:
Hi,
That only shows up if you configure with --enable-libstcxx-time
I'll finish reg-testing it and submit it when I get home this evening.
thanks Jon. Looks like we have also to fix parallel-mode (49187), I
hope there isn't too much to reshuffle, if
On Fri, May 27, 2011 at 10:54 AM, Jan Hubicka wrote:
> Hi,
> here is updated patch incorporating the feedback. I now use variant of uleb
> format,
> just do 4 bit chunks instead of 8bit. It works better in my tests and is also
> used
> by LLVM so it must be cool ;)
>
> I also fixed off by one e
> On Fri, May 27, 2011 at 10:54 AM, Jan Hubicka wrote:
> > Hi,
> > here is updated patch incorporating the feedback. I now use variant of
> > uleb format,
> > just do 4 bit chunks instead of 8bit. It works better in my tests and is
> > also used
> > by LLVM so it must be cool ;)
> >
> > I also
On Fri, May 27, 2011 at 12:02 AM, Xinliang David Li wrote:
> The latest version that only exports 2 functions from passes.c.
Ok with ...
@@ -637,4 +637,8 @@ extern bool first_pass_instance;
/* Declare for plugins. */
extern void do_per_function_toporder (void (*) (void *), void *);
+extern v
On Thu, May 26, 2011 at 7:50 PM, William J. Schmidt
wrote:
>
> On Thu, 2011-05-26 at 11:22 +0200, Richard Guenther wrote:
>
>
>
>
>> There are several extra pre-conditions in the original code in
>> builtins.c as well as commens for reasonings (yes, there seem
>> to be duplicates of several of th
This fixes PR49177 by re-introducing the removed folding in a
correct way.
Bootstrapped and tested on x86_64-unknown-linux-gnu, applied to trunk.
Richard.
2011-05-27 Richard Guenther
PR middle-end/49177
* fold-const.c (fold_unary_loc): Fold (T)(A CMP B) to
A CMP B ?
On Thu, 26 May 2011, Hans-Peter Nilsson wrote:
> Good, a way forward (well, for me) even though the cure is
> suspiciously incidental. Thanks for your patience.
JFTR, setting CRIS_CC0_REGNUM to fixed just moves the assert to
trig at the same place in the v10 multilib, where
CRIS_MOF_REGNUM is ava
Hi,
On Thu, 26 May 2011, Martin Jambor wrote:
> I assume DSE does not remove the stores as that would defeat the purpose
> of the patch.
Right. (The volatileness currently prevents the removal).
> If after optimizations such as SRA, these special stores are the only
> statements accessing th
Rainer has verified the following patch fixes the problem he observed in
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=49170. OK to commit?
2011-05-27 Bill Schmidt
PR tree-optimization/49170
* tree-ssa-math-opts.c (execute_cse_sincos): Add checks for
sincos or cexp.
Uros Bizjak writes:
> Please find attached the patch that introduces %p to output only the
> raw symbol name. @plt is added "manually", since it doesn't really
> follow -fpic flag.
Thanks, that almost did it:
* HAVE_AS_IX86_{TLSGDPLT, TLSLDMPLT} are always defined as 0/1, so we
need if, not #
Hi,
objc.dg/torture/forward-1.m is also expected to fail on x32. Checking
it in.
H.J.
---
commit cabf76cce0f4efcfdd3fb033255830c1336a0c64
Author: H.J. Lu
Date: Fri May 27 04:59:36 2011 -0700
Require ia32 instead of ilp32.
diff --git a/gcc/testsuite/ChangeLog.x32 b/gcc/testsuite/ChangeLo
Fix testsuite error introduced in r174121.
r174121 introduces a new warn_self_assign options variable that conflicts
with a static function in gcc/testsuite/g{cc,++}.dg/plugin/selfassign.c.
Import a small portion of r172992 to avoid the problem.
Okay for google/integration?
gcc/testsuite/ChangeL
On Fri, May 27, 2011 at 08:35, Simon Baldwin wrote:
> Fix testsuite error introduced in r174121.
>
> r174121 introduces a new warn_self_assign options variable that conflicts
> with a static function in gcc/testsuite/g{cc,++}.dg/plugin/selfassign.c.
> Import a small portion of r172992 to avoid the
On Fri, May 27, 2011 at 2:25 PM, Rainer Orth
wrote:
>> Please find attached the patch that introduces %p to output only the
>> raw symbol name. @plt is added "manually", since it doesn't really
>> follow -fpic flag.
>
> Thanks, that almost did it:
>
> * HAVE_AS_IX86_{TLSGDPLT, TLSLDMPLT} are alwa
This fixes PR49189 by re-installing the non-folding of
conversions of comparison results.
Bootstrapped and tested on x86_64-unknown-linux-gnu, applied.
Richard.
2011-05-27 Richard Guenther
PR middle-end/49189
* fold-const.c (fold_unary_loc): Do not re-fold folding conversion
On Fri, May 27, 2011 at 2:06 PM, William J. Schmidt
wrote:
> Rainer has verified the following patch fixes the problem he observed in
> http://gcc.gnu.org/bugzilla/show_bug.cgi?id=49170. OK to commit?
Ok.
Thanks,
Richard.
>
> 2011-05-27 Bill Schmidt
>
> PR tree-optimization/49170
>
Hi,
When combine sees:
(insn 5 2 7 2 (parallel [
(set (reg/f:DI 61)
(plus:DI (reg/f:DI 20 frame)
(const_int -64 [0xffc0])))
(clobber (reg:CC 17 flags))
]) x.i:12 252 {*adddi_1}
(expr_list:REG_UNUSED (reg:CC 17 fl
http://codereview.appspot.com/4526083/diff/1/gcc/cp/cp-objcp-common.c
File gcc/cp/cp-objcp-common.c (right):
http://codereview.appspot.com/4526083/diff/1/gcc/cp/cp-objcp-common.c#newcode103
gcc/cp/cp-objcp-common.c:103:
case TEMPLATE_INFO: return sizeof (struct
tree_template_info);
On 05/26/2011 11:10 PM, Tobias Burnus wrote:
Attached is a small coarray fix, solving two issues:
- For -fcoarray=single, nonallocatable coarrays are nonpointer - but currently
they get set the "restrict" qualifier. (With -fcoarray=lib, nonallocatble
coarrays are always pointers.)
Fixed by n
Hi,
When reload gets:
(insn 588 587 589 28 (set (mem:DF (zero_extend:DI (plus:SI (subreg:SI
(reg/v/f:DI 182 [ b ]) 0)
(const_int 8 [0x8]))) [4 MEM[base: b_96(D), index:
D.15020_278, step: 8, offset: 0B]+0 S8 A64])
(reg:DF 340 [ D.14980 ])) spooles.c:291 106
{*movdf_in
[PATCH ARM] cleanup: local label prefix
http://gcc.gnu.org/ml/gcc-patches/2011-05/msg01433.html
OK to commit?
Thanks,
Greta
On Fri, 2011-05-27 at 12:24 +0200, Richard Guenther wrote:
> Now, the tree_expr_nonnegative_p check will always return false
> because arg0 is a gimple value. We would need to implement
> either some propagation engine to derive properties for floating
> point entities or mimic what tree_expr_
OK, but please combine those tests into one file.
Jason
Hi,
On Thu, 26 May 2011, Steven Bosscher wrote:
> > on IRC we discussed about this, here's the RFC patch. It bootstraps
> > and causes some minor regressions most probably due to some missing
> > sprinkled checks for the special clobber insns and sometimes due to
> > having to adjust some reg
The test contains undefined signed integer overflow and fails with
a patch I have in testing. Fixed by adding -fwrapv.
Richard.
2011-05-27 Richard Guenther
* gcc.c-torture/execute/920612-1.x: Add -fwrapv.
Index: gcc/testsuite/gcc.c-torture/execute/920612-1.x
===
On Fri, May 27, 2011 at 03:59:47PM +0200, Michael Matz wrote:
> On Thu, 26 May 2011, Steven Bosscher wrote:
> > > on IRC we discussed about this, here's the RFC patch. It bootstraps
> > > and causes some minor regressions most probably due to some missing
> > > sprinkled checks for the special c
On 27 May 2011 14:35, Greta Yorsh wrote:
> [PATCH ARM] cleanup: local label prefix
> http://gcc.gnu.org/ml/gcc-patches/2011-05/msg01433.html
>
> OK to commit?
OK.
For future please remember that Changelog entries require 2 spaces
between the end of the last name and the beginning of the email
ad
This patch fixes the lack of combining comparisons that appear
on the RHS of assignments (as opposed to in gimple-conds or
in cond-exprs on the RHS of assignments).
Bootstrapped on x86_64-unknown-linux-gnu, testing in progress.
Richard.
2011-05-27 Richard Guenther
* tree-ssa-forwpro
Another one.
Richard.
2011-05-27 Richard Guenther
* gcc.c-torture/execute/920711-1.x: Add -fwrapv.
Index: gcc/testsuite/gcc.c-torture/execute/920711-1.x
===
--- gcc/testsuite/gcc.c-torture/execute/920711-1.x (revis
On 05/25/2011 12:29 AM, Vladimir Makarov wrote:
> So I need to finish review of the 4th patch (I hope to finish it
> tomorrow). All other patches is ok.
Thanks, Vlad. I'm trickling them in now.
>> http://gcc.gnu.org/ml/gcc-patches/2011-05/msg00755.html
>>
> Ok. Only changelog entry for deps_anal
On 05/27/2011 03:57 PM, Jason Merrill wrote:
OK, but please combine those tests into one file.
Thanks. As applied.
Paolo.
//
/cp
2011-05-27 Paolo Carlini
PR c++/42056
* typeck2.c (build_functional_cast): Complain early for invalid uses
of 'auto' and set
Georg-Johann Lay schrieb:
> Georg-Johann Lay schrieb:
>> Georg-Johann Lay schrieb:
>>> Weddington, Eric schrieb:
> -Original Message-
> From: Georg-Johann Lay [mailto:a...@gjlay.de]
> Sent: Friday, April 01, 2011 4:01 PM
> To: Georg-Johann Lay
> Cc: gcc-patches@gcc.gnu.o
> 2011-05-27 Richard Guenther
>
> PR middle-end/49189
> * fold-const.c (fold_unary_loc): Do not re-fold folding conversions
> of comparisons.
>
> * gnat.dg/bit_packed_array5.adb: New testcase.
> * gnat.dg/bit_packed_array5.ads: Likewise.
Thanks a lot! I managed to
Hi,
On Fri, 27 May 2011, Jakub Jelinek wrote:
> On Fri, May 27, 2011 at 03:59:47PM +0200, Michael Matz wrote:
> > On Thu, 26 May 2011, Steven Bosscher wrote:
> > > > on IRC we discussed about this, here's the RFC patch. It bootstraps
> > > > and causes some minor regressions most probably due t
This patch adds support for the "nonnull" attribute for Objective-C methods
(clang has it too).
The implementation follows the existing framework and is very similar to the
"format" attribute one.
Testcases included.
Ok to commit ?
Thanks
PS: There is a small unsatisfactory issue with the ge
Uros Bizjak writes:
>> * HAVE_AS_IX86_{TLSGDPLT, TLSLDMPLT} are always defined as 0/1, so we
>> need if, not #ifdef.
>
> There is no need to always define it. Please simplify your
> configure.ac check, see i.e. check for sahf mnemonic in configure.ac.
Ok, done.
> Please change the check in con
On 05/17/2011 06:34 AM, Paolo Bonzini wrote:
> On 05/12/2011 05:40 PM, Bernd Schmidt wrote:
>> + if (targetm.calls.function_arg_round_to_arg_boundary (passed_mode,
>> type))
>> +round_boundary = boundary;
>> + else
>> +round_boundary = PARM_BOUNDARY;
>
> Why add an if, instead of making
This patch removes the (maybe) dependency of libobjc from target libiberty from
the makefiles.
I don't know of any reasons why libobjc would need target libiberty; I also
looked around
and couldn't spot anything. Most other target libraries don't depend on
libiberty; I'm
not sure why libobjc wo
The helper functions used to implement double-precision arithmetic on
ARM processors that only support single-precision arithmetic in hardware
should use the soft-float ABI (i.e. passing and returning floating-point
arguments in core registers), even when -mfloat-abi=hard is in effect.
This patch t
Hi all,
This patch adds support for the ARM SMLALTB, SMLALTT, SMLATB, and SMLATT
instructions.
These instructions do HImode -> DI/SImode widening
multiply-and-accumulate with one or both operands taken from the top 16
bits of the source register.
Note that this patch won't do much until my
Hi!
This patch adds a few peephole2s to help optimize if (!--*x) etc.
This is something the combiner doesn't and can't easily handle, because in
reg0 = mem1
reg0 {+,-,&,|,^}= x
mem1 = reg0
cc = compare (reg0, 0)
reg0 is used also by both the store and compare and there is no dependence
in between
Hi,
parallel-mode should be back to usable. Note, I didn't try to clean-up
the tangle of includes we have in parallel-mode files, a lot can be
improved, for sure. I also added gates for the *_debug.cc pb_ds
testfiles, otherwise all spuriously fail in check-parallel, etc.
Tested x86_64-linux
Hi!
As the testcase shows (it is eh/cond5.C just with s/true :/1 :/g),
c_common_truthvalue_conversion needs to be aware of void type arms
too, otherwise it errors on valid source.
Bootstrapped/regtested on x86_64-linux and i686-linux, ok for trunk/4.6?
2011-05-27 Jakub Jelinek
PR c++
On May 27, 2011, at 12:13 PM, Joseph S. Myers wrote:
> ...Full tests are still running for this version of the patch, but OK to
> commit subject to results of testing (bootstrap on
> x86_64-unknown-linux-gnu, building cc1 and xgcc for crosses to:
> alpha-linux-gnu arm-eabi avr-elf bfin-elf cris-e
OK.
Jason
2011/5/27 Georg-Johann Lay :
> Georg-Johann Lay schrieb:
>> Georg-Johann Lay schrieb:
>>> Georg-Johann Lay schrieb:
Weddington, Eric schrieb:
>> -Original Message-
>> From: Georg-Johann Lay [mailto:a...@gjlay.de]
>> Sent: Friday, April 01, 2011 4:01 PM
>> To: Georg-Joha
This is a revised version of
http://gcc.gnu.org/ml/gcc-patches/2011-05/msg00098.html
to move the Solaris startup files to toplevel libgcc. While I can
commit most if it myself, I need approval/testing in a few areas:
* Non-Solaris build changes:
** libgcc/Makefile.in now has cpu_type,
On 5/26/11, Nathan Froyd wrote:
> On 05/26/2011 10:24 PM, Lawrence Crowl wrote:
>> Index: gcc/cp/cp-objcp-common.c
>> ===
>> --- gcc/cp/cp-objcp-common.c (revision 174301)
>> +++ gcc/cp/cp-objcp-common.c (working copy)
>> @@ -99,6 +99
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
Updated based on some comments from Bernd; specifically the other use of
delete_dead_insn has been removed.
WRT the assembly differences on MIPS Bernd referred to; what ultimately
caused this problem were two dead insns that had been previously
elimin
The LTO streamer supports shared tree pointers by using a cache of
pickled tree nodes. The first time a tree is streamed, it's pickled
and added to the cache. Subsequent stream operations use the pickle
reference.
This patch adds the same support for all the data structures pickled
in lang-spec
Hello,
I'm checking in this patch after Gerald Pfeifer's off-list approval to add
Zdenek Sojka to the list of contributors. I am sure Zdenek's bug reports are
useful and help make GCC a better and more robust compiler. Thank you!
2011-05-27 Alexander Monakov
* doc/contrib.texi: Upd
On 5/27/11, dnovi...@google.com wrote:
>
> http://codereview.appspot.com/4526083/diff/1/gcc/cp/cp-objcp-common.c
> File gcc/cp/cp-objcp-common.c (right):
>
> http://codereview.appspot.com/4526083/diff/1/gcc/cp/cp-objcp-common.c#newcode103
> gcc/cp/cp-objcp-common.c:103:
> case TEMPLATE_INFO:
As $SUBJECT suggests.
-Nathan
* MAINTAINERS (Write After Approval): Update my email address.
Index: MAINTAINERS
===
--- MAINTAINERS (revision 174345)
+++ MAINTAINERS (working copy)
@@ -352,7 +352,7 @@ Li Feng
> * Non-Solaris SPARC changes:
>
> After I had moved sparc/sol2-c[in].asm to libgcc, I noticed that
> despite the name a few non-Solaris targets uses those files, too:
>
> sparc-*-elf*, sparc-*-rtems*, sparc64-*-elf*, sparc64-*-rtems*
Yes, I tried to untangle that, but the RTEMS folks compla
On Fri, May 27, 2011 at 10:54 AM, Alexander Monakov wrote:
> Hello,
>
> I'm checking in this patch after Gerald Pfeifer's off-list approval to add
> Zdenek Sojka to the list of contributors. I am sure Zdenek's bug reports are
> useful and help make GCC a better and more robust compiler. Thank yo
This is the test I was referring to in
http://gcc.gnu.org/ml/gcc-patches/2011-05/msg02184.html
The test, taken from the C testsuite, creates hundreds of thousands of
shared bindings. Without sharing support, the streamer was trying to
instantiate new objects, and ICEing eventually when the stack
Approved by Jerry on IRC and committed as Rev. 174348.
Tobias
Tobias Burnus wrote:
This patch adds the rank intrinsic of the TR 29113 draft. It is a bit
boring until gfortran supports DIMENSION(..) ;-)
Build and regtested on x86-64-linux.
OK for the trunk?
Tobias
PS: The current draft of TR
On Fri, May 27, 2011 at 09:54:50PM +0400, Alexander Monakov wrote:
> I'm checking in this patch after Gerald Pfeifer's off-list approval to add
> Zdenek Sojka to the list of contributors. I am sure Zdenek's bug reports are
> useful and help make GCC a better and more robust compiler. Thank you!
While working on a C6X scheduling patch, I found myself wondering - what
would be involved in making the register renamer operate on extended
basic blocks rather than simple bbs? Somewhat surprisingly, the answer
turns out to be "not much". After the last rewrite, all the conflict
tests are based o
This patch contains the actual changes for renaming across extended
basic blocks (where ebb here is a tree where each basic block except the
root has exactly one predecessor).
All we really have to do is reconstruct some information when starting
to scan a different bb; the open_chains list can be
On 05/26/2011 03:04 PM, Nathan Froyd wrote:
Thanks, this is looking pretty close. A few more issues, mostly to do
with wording of the diagnostics:
@@ -14084,7 +14350,7 @@ fn_type_unification (tree fn,
sarg = tree_cons (NULL_TREE, TREE_TYPE (substed), sarg);
for (i = 0;
On 05/26/2011 02:37 PM, Aldy Hernandez wrote:
The narrowest integer mode containing the bit field is still 32, so we
access the bitfield with an SI instruction as expected.
OK, then:
struct A
{
int i: 4;
int j: 4;
int k: 8;
int l: 8;
int m: 8;
};
now the narrowest mode containing 'j
Pretty straightforward, we just didn't handle the compound assignment
operators before.
Tested x86_64-pc-linux-gnu, applying to trunk.
commit 5aba1b056466d0713a6479a364a96504bcfc93b9
Author: Jason Merrill
Date: Fri May 27 13:48:39 2011 -0400
PR c++/47132
* mangle.c (write_expressio
Like 48935, we don't want to call constructor_name_p on non-classes.
Tested x86_64-pc-linux-gnu, applying to trunk.
commit 6e1aed3b584b9bf7bd3127d12516dd5ac5da7ba4
Author: Jason Merrill
Date: Fri May 27 12:52:37 2011 -0400
PR c++/47277
* parser.c (cp_parser_unqualified_id): Don't ch
Here, deciding whether the nested lambda was dependent meant checking
the enclosing function, which meant checking the function's type, which
meant checking its return type, the nested lambda. I've broken this
cycle by looking not at whether the enclosing function is a dependent
expression, bu
We already handle reference refs properly elsewhere, but needed to
handle it here as well.
Tested x86_64-pc-linux-gnu, applying to trunk.
commit 1e9af8c4d3c615f04f456306587959cdd6c49c26
Author: Jason Merrill
Date: Fri May 27 10:54:40 2011 -0400
PR c++/48284
* error.c (dump_expr) [C
Here, the problem was that when we went to generate the
partially-instantiated function type in the mangler, we weren't in the
proper scope so lookup of an identifier failed. Fixed by doing
push_access_scope like we do in other situations.
Tested x86_64-pc-linux-gnu, applying to trunk.
commit
On May 27, 2011, at 8:45 AM, Nicola Pero wrote:
> This patch adds support for the "nonnull" attribute for Objective-C methods
> (clang has it too).
> Ok to commit ?
Ok.
On 05/27/2011 05:54 PM, Bernd Schmidt wrote:
Like this? Untested so far beyond making sure it builds.
Yes.
Paolo
Hello Alexander,
thank you for this! I am happy I can help GCC development at least this way.
Zdenek
> Původní zpráva
> Od: Alexander Monakov
> Předmět: [committed] Mention Zdenek Sojka in contributors.html
> Datum: 27.5.2011 20:25:32
>
More C tests that used to fail. There is roughly one test per class
of failure that I fixed with previous patches.
Committed to branch.
Diego.
* g++.dg/pph/c1attr-warn-unused-result.cc: New.
* g++.dg/pph/c1attr-warn-unused-result.h: New.
* g++.dg/pph/c1builtin-integra
On 27 May 2011 09:30, Jonathan Wakely wrote:
>
> I think this piece is also needed due to the fix for 29131:
>
> Index: include/std/thread
> ===
> --- include/std/thread (revision 174307)
> +++ include/std/thread (working copy)
> @@
Building gcc-4.6 arm android toolchain fails because of an
incompatible function definition between libiberty and bionic.
Thanking Joseph, I have learned that "there's no such thing as a
target libiberty" and we should rip all the target-libiberty rules
out. I don't know if someone is working on
This patch to the Go frontend uses the backend interface to export data.
This moves one gcc-specific function from the Go frontend proper into
the gcc interface. With this change there are three remaining files in
the Go frontend proper which have gcc-specific code: expressions.cc,
gogo-tree.cc, a
"Joseph S. Myers" writes:
> Ian, what is the right way to handle the gofrontend change in this
> patch (to update code for the move of a hook from targetm to
> targetm_common)? It's small, mechanical and I think below the
> threshold of significance for copyright.
I agree, but to avoid any conf
Remove case TREE_BINFO in cp_tree_size().
Index: gcc/cp/ChangeLog.pph
2011-05-27 Lawrence Crowl
* cp-objcp-common.c (cp_tree_size): Remove case TREE_BINFO.
Index: gcc/cp/cp-objcp-common.c
===
--- gcc/cp/cp-objcp-common
Hi Sofiane,
I find your following patch has been approved by Richard in Oct last
year, but it is not trunk.
Is there any problem with it?
http://gcc.gnu.org/ml/gcc-patches/2010-10/msg00266.html
If you don't mind, I can help to commit the patch.
Thanks,
Jing
2011-05-28 Jonathan Wakely
* include/std/future: Use noexcept.
* src/future.cc: Likewise.
Tested x86_84-linux, committed to trunk
Index: include/std/future
===
--- include/std/future (revision 174358)
+++ include/
This updates the std::launch enum to match the FDIS.
2011-05-28 Jonathan Wakely
* include/std/future (launch): Update enumerators and define
operators required for bitmask type. Remove trailing whitespace.
* src/future.cc: Remove trailing whitespace.
* testsuite
This patch uses glibc's get_nprocs() or the non-standard but widely
available _SC_NPROCESSORS_ONLN sysconf variable to determine the
number of processors.
2011-05-28 Jonathan Wakely
* acinclude.m4: Define GLIBCXX_CHECK_GET_NPROCS and
GLIBCXX_CHECK_SC_NPROCESSORS_ONLN.
*
The problem with this testcase was that we were trying to generate the
mangled name of an uninstantiated lambda; we shouldn't ever try to
mangle an uninstantiated template. So I changed
start_preparsed_function to avoid that. After that was fixed, I found
that I also needed to modify maybe_ad
Hi,
This patch uses __sync_bool_compare_and_swap and removes
sysdep/x86-64/locks.h. OK for trunk if there is no regression?
Thanks.
H.J.
---
2011-05-27 H.J. Lu
PR libgcj/49193
* configure.host (sysdeps_dir): Set to i386 for x86_64.
* sysdep/i386/locks.h (compare_and
On 05/27/2011 04:22 PM, Nathan Froyd wrote:
+ /* The PARM is not one we're trying to unify. Just check
+to see if it matches ARG. */
+ int result = !(TREE_CODE (arg) == TREE_CODE (parm)
+&& cp_tree_equal (parm, arg));
+ if (result)
+
On Thu, 26 May 2011, Vladimir Makarov wrote:
> On 05/26/2011 04:47 AM, Hans-Peter Nilsson wrote:
> > On Wed, 25 May 2011, Vladimir Makarov wrote:
> > It sounds like you're saying that "the narrowest register
> > classes must be formed of registers for which there are trivial
> > instructions to mov
On Sat, 28 May 2011, Hans-Peter Nilsson wrote:
> +You must define the narrowest register class for a register so that
> +class either has no subclasses, or that for some mode, the move cost
> +between registers within the class are cheaper than moving a register
Bah: "is cheaper" is the grammatica
Thanks for Richard for debugging this!
Setting TREE_TYPE in trans-decl.c is nonsense and leads to types of the
form "_Complex _Complex int". The fix is rather obvious.
Build and regtested on x86-64-linux.
OK for the trunk?
Tobias
2011-05-28 Tobias Burnus
PR fortran/18918
* trans-
96 matches
Mail list logo
