Hi,
no real reason to add this tiny bit of C++11, and the related export
ungliness, at this late stage in the 4.8 release process. I would say,
let's just do it in the next ABI.
Committed.
Thanks,
Paolo.
/
2013-02-11 Paolo Carlini paolo.carl...@oracle.com
On 05/02/13 18:18, Christophe Lyon wrote:
Hi,
Following the discussion about disable peeling [1] a few weeks ago,
it turned out that the vectorizer cost model needed some
implementation for ARM.
The attached patch implements arm_builtin_vectorization_cost and
arm_add_stmt_cost, providing
On Wed, Jan 23, 2013 at 04:24:01PM +0400, Evgeniy Stepanov wrote:
What if glibc adds a scanf hook (like it has already printf hooks), apps
could then register their own stuff and the above would then break. It
really should be very conservative, and should be checked e.g. with all
On Mon, Feb 11, 2013 at 3:38 PM, Jakub Jelinek ja...@redhat.com wrote:
On Wed, Jan 23, 2013 at 04:24:01PM +0400, Evgeniy Stepanov wrote:
What if glibc adds a scanf hook (like it has already printf hooks), apps
could then register their own stuff and the above would then break. It
really
This is another -Warray-bounds false-positive triggered by more
aggressive loop unrolling. Fortunately VRP has everything available
to compute correct bounds - if it were not for VRP itself pessimizing
itself (it runs twice after all) by folding predicates from
a C to a != C (and similar
This fixes PR56264 - the machinery to communicate changed-blocks from
fix_loop_structure to cfgcleanups call to rewrite_into_loop_closed_ssa
doesn't work (at least) in the face of newly discovered loops as that
exposes new definition sites and not only new use sites (as the code
assumes).
Fixed
On 02/10/13 09:32, Uros Bizjak wrote:
Hello!
Attached patch prevents out-of-bounds offset in the call to
simplify_subreg, where the subreg offset is taken from the offset of a
memory access. The problem triggers on 4.6 branch and is latent on 4.7
and mainline.
2013-02-10 Uros Bizjak
On Mon, Feb 11, 2013 at 12:38:00PM +0100, Jakub Jelinek wrote:
On Wed, Jan 23, 2013 at 04:24:01PM +0400, Evgeniy Stepanov wrote:
What if glibc adds a scanf hook (like it has already printf hooks), apps
could then register their own stuff and the above would then break. It
really
We now optimize it better via tail-merging reassoc before VRP2
can transform the 2nd test to !=. The following restores the
input to VRP to match the original testcase expectations.
Committed.
Richard.
2013-02-11 Richard Biener rguent...@suse.de
* gcc.dg/tree-ssa/vrp17.c: Disable
Ping! (From the silence I infer that this is rather 4.9 material?)
2013/1/19 Janus Weil ja...@gcc.gnu.org:
Hi all,
here is a simple patch for a rejects-valid problem with deferred
type-bound procedures. This is the patch from comment 5 in the PR. A
simpler (but equivalent) patch can be
Hi,
It seems that glibc 2.17 changes the abbreviated names of weekdays for
ru_RU locales by removing an extraneous ., as described in:
http://sourceware.org/bugzilla/show_bug.cgi?id=10873
An earlier patch (circa glibc 2.14) changed (IIUC!) archaic/unusual
three-letter abbreviations to
Richard,
Thanks for your comments.
Here a new version with the changes you suggested.
Christophe
On 11 February 2013 11:57, Richard Earnshaw rearn...@arm.com wrote:
On 05/02/13 18:18, Christophe Lyon wrote:
Hi,
Following the discussion about disable peeling [1] a few weeks ago,
it
This fixes the compile-time sink in find_uses_to_rename, that we
scan the whole function when nothing is to do (well, appearantly).
-O3 bootstrap and regtest on x86_64-unknown-linux-gnu in progress,
scheduled for stage1.
Richard.
2013-02-11 Richard Biener rguent...@suse.de
*
Another missing maybe_instantiate_noexcept.
Tested x86_64-pc-linux-gnu, applying to trunk and 4.7.
commit 1a9ad88eb3e501aed25ca6542776aacf431c7adc
Author: Jason Merrill ja...@redhat.com
Date: Sat Feb 9 15:58:29 2013 -0500
PR c++/56268
* semantics.c
Iain Sandoe discovered that on intel darwin9, the asan testsuite suffers
hundreds of
failures due to the absence of dispatch calls (Grand Central Dispatch) prior to
darwin10.
The attached patch disables building libsanitizer on darwin8 and darwin9 until
upstream
decides to support the
On 11 Feb 2013, at 15:55, Jack Howarth wrote:
Iain Sandoe discovered that on intel darwin9, the asan testsuite suffers
hundreds of
failures due to the absence of dispatch calls (Grand Central Dispatch) prior
to darwin10.
The attached patch disables building libsanitizer on darwin8 and
For the record, ASan never claimed darwin9 support and isn't going to
support darwin9 in the future, so it should be fine to disable
building ASan on this platform.
On Mon, Feb 11, 2013 at 7:55 PM, Jack Howarth howa...@bromo.med.uc.edu wrote:
Iain Sandoe discovered that on intel darwin9, the
On Feb 9, 2013, at 3:19 PM, Ian Lance Taylor i...@google.com wrote:
If the libgo testsuite is using DejaGNU, there is no need to frob the
log file to make it look like DejaGNU output. Bootstrapped and ran Go
testsuite on x86_64-unknown-linux-gnu (without DejaGNU). Committed to
mainline.
Ping
On Wed, Feb 6, 2013 at 12:10 AM, Kai Tietz ktiet...@googlemail.com wrote:
Hi,
this patch fixes an issue about recursice LN_S for mingw-host. The
issue was already addressed by autotools, but an upgrade of version
isn't suitable right now.
For further information see the bug-report PR
Hi,
On 02/11/2013 04:33 PM, Julian Brown wrote:
Hi,
It seems that glibc 2.17 changes the abbreviated names of weekdays for
ru_RU locales by removing an extraneous ., as described in:
http://sourceware.org/bugzilla/show_bug.cgi?id=10873
An earlier patch (circa glibc 2.14) changed (IIUC!)
On Mon, Feb 11, 2013 at 9:18 AM, Paolo Carlini paolo.carl...@oracle.com wrote:
Hi,
On 02/11/2013 04:33 PM, Julian Brown wrote:
Hi,
It seems that glibc 2.17 changes the abbreviated names of weekdays for
ru_RU locales by removing an extraneous ., as described in:
On Feb 6, 2013, at 05:10, Kai Tietz ktiet...@googlemail.com wrote:
this patch fixes an issue about recursice LN_S for mingw-host. The
issue was already addressed by autotools, but an upgrade of version
isn't suitable right now.
For further information see the bug-report PR 52122.
While I was going over the C++11 status page, I noticed that quick_exit
is implemented by glibc, it just needed to be added to the libstdc++
cstdlib header.
Does this configury/feature macro handling look right? Is this enough
testcase? Is this small/safe enough to go in for 4.8?
commit
Does this configury/feature macro handling look right?
yes.
Is this enough testcase?
yes.
Is this small/safe enough to go in for 4.8?
yes.
commit 884a2a7815a95bade9d23f01b4c64a16808c7f05
Author: Jason Merrill ja...@redhat.com
Date: Mon Feb 11 11:23:30 2013 -0500
*
Hi,
On 02/11/2013 06:33 PM, Jason Merrill wrote:
While I was going over the C++11 status page, I noticed that
quick_exit is implemented by glibc, it just needed to be added to the
libstdc++ cstdlib header.
Does this configury/feature macro handling look right? Is this enough
testcase? Is
Hi,
This patch updates GCC 4.8 changes.html to mention
-maddress-mode=[short|long] for x32. OK to install?
H.J.
---
Index: changes.html
===
RCS file: /cvs/gcc/wwwdocs/htdocs/gcc-4.8/changes.html,v
retrieving revision 1.95
diff -u
Hello Everyone,
Attached, please find a patch that implements Array Notations for C++. This
patch affects the C++ compiler and should build (or be applied) on top of the
Array Notation for C patch (2 of n:
http://gcc.gnu.org/ml/gcc-patches/2013-01/msg01213.html). To my best knowledge,
I
.. also, I think we should have #if __cplusplus = 201103L protecting
these functions too, exactly like all the other C++11 bits. Eg, I don't
think we should unconditionally, ie in C++03 mode too, declare
at_quick_exit and quick_exit in namespace std.
I can tweak things for the two issues I
Hi,
Is this alright to commit?
Thanks
Sri
On Thu, Feb 7, 2013 at 4:10 PM, Sriraman Tallam tmsri...@google.com wrote:
On Thu, Feb 7, 2013 at 6:29 AM, Jason Merrill ja...@redhat.com wrote:
On 02/06/2013 08:39 PM, Sriraman Tallam wrote:
+// Test to check if an error is generated when virtual
.. also, I think we should have #if __cplusplus = 201103L protecting
these functions too, exactly like all the other C++11 bits. Eg, I
don't think we should unconditionally, ie in C++03 mode too, declare
at_quick_exit and quick_exit in namespace std.
Yeah.
I can tweak things for the two
On 02/11/2013 07:54 PM, Benjamin De Kosnik wrote:
I can tweak things for the two issues I noticed.
Thanks.
It seems like this is still 4.8 material to me.
Great. I'll take care of that later today. I'm also going to make sure
the new testcase doesn't spuriously fail if the functions aren't
This patch from Rainer Orth in PR 56171 fixes some Solaris portability
problems in libgo.
* Check for nanosleep in -lrt.
* Pass $(LIBS) when testing so that the tests use -lrt as needed.
* Make sure that AF_LOCAL is defined.
Bootstrapped and ran Go testsuite on x86_64-unknown-linux-gnu.
On 02/11/2013 12:28 PM, H.J. Lu wrote:
On Mon, Feb 11, 2013 at 9:18 AM, Paolo Carlini paolo.carl...@oracle.com
wrote:
On 02/11/2013 04:33 PM, Julian Brown wrote:
Hi,
It seems that glibc 2.17 changes the abbreviated names of weekdays for
ru_RU locales by removing an extraneous ., as
On 11 February 2013 17:18, Paolo Carlini wrote:
I think it's Ok, yes. Thanks. However, I would appreciate if somebody with a
glibc 2.17 system at hand could double check. Maybe HJ?
Although I'm sure Carlos is right that they're close enough to not
need checking, I've got a Rawhide setup with
This patch adds optimized support for (short) (un)signed _Sat _Accum
multiplication (HA, UHA, SA, USA).
The 32-bit multiply uses 64 = 32 * 32 widening multiplication. As a spin-off,
the widening mul is available as [u]mulsidi3 default pattern.
Okay for trunk?
Johann
gcc/
PR
Hi,
This patch was approved by Ian:
http://sourceware.org/ml/binutils/2013-02/msg00136.html
I have synced plugin-api.h
* plugin-api.h (enum ld_plugin_level): Assign integers
explicitly for all values.
Index: include/plugin-api.h
Hi!
As discussed in this PR, MEM1 op= MEM2 is usually better expanded
as temp = MEM2; MEM1 op= temp; if target supports that, even when it
means we can't add a REG_EQUIV note in that case (it would be
self-referential).
Fixed thusly, bootstrapped/regtested on x86_64-linux and i686-linux, ok
for
On 02/11/13 12:49, Jakub Jelinek wrote:
Hi!
As discussed in this PR, MEM1 op= MEM2 is usually better expanded
as temp = MEM2; MEM1 op= temp; if target supports that, even when it
means we can't add a REG_EQUIV note in that case (it would be
self-referential).
Fixed thusly,
On 11 February 2013 19:15, Jonathan Wakely wrote:
On 11 February 2013 17:18, Paolo Carlini wrote:
I think it's Ok, yes. Thanks. However, I would appreciate if somebody with a
glibc 2.17 system at hand could double check. Maybe HJ?
Although I'm sure Carlos is right that they're close enough
That's crystal clear. I think I can recognize one or two words from my
original proposal like 'The' or 'in' :-)
François
On 02/11/2013 01:24 AM, Jonathan Wakely wrote:
On 7 February 2013 21:01, François Dumont wrote:
Thanks for taking care of it. Here is another version, I think clearer.
On Feb 11, 2013, at 7:55 AM, Jack Howarth howa...@bromo.med.uc.edu wrote:
Iain Sandoe discovered that on intel darwin9, the asan testsuite suffers
hundreds of
failures due to the absence of dispatch calls (Grand Central Dispatch) prior
to darwin10.
The attached patch disables building
... this is what I committed.
Thanks,
Paolo.
/
2013-02-11 Paolo Carlini paolo.carl...@oracle.com
* include/c_std/cstdlib (at_quick_exit, quick_exit): Do not declare.
* include/c_global/cstdlib (at_quick_exit, quick_exit): Declare only
in C++11 mode
The problem here is that -ffast-math is overridden when switching
optimization options on a per function basis with
__attribute__((optimize(O))).
The x86 ceilf* instructions depend on unsafe math optimizations, but the
optabs are created at the beginning of the compilation. When fast math
Since commit r195676[1], it looks like
libstdc++-v3/src/c++11/hashtable_c++0x.cc is missing an explicit
instantiation for std::lower_bound.
it's missing an implicit instantiation of std::lower_bound.
This leads to
The following patch fixes the assignment of the insn type attribute
for loads and stores. Specifically, it will now appropriately assign
update form and update-indexed form type attributes which allows for
better instruction scheduling by honoring group formation restrictions
and (to a lesser
OK, thanks.
Jason
On Tue, Feb 12, 2013 at 3:16 AM, Mike Stump mikest...@comcast.net wrote:
On Feb 11, 2013, at 7:55 AM, Jack Howarth howa...@bromo.med.uc.edu wrote:
Iain Sandoe discovered that on intel darwin9, the asan testsuite suffers
hundreds of
failures due to the absence of dispatch calls (Grand Central
47 matches
Mail list logo
