On Sun, 11 Nov 2012, Jason Merrill wrote:
On 11/11/2012 06:55 AM, Jonathan Wakely wrote:
For std::list I assume we want the attribute on the type itself, not
just member functions, since its size changes.
When we are changing the whole type, I assume the point of using this
attribute instead
On 11/12/2012 12:05 AM, Jason Merrill wrote:
OK.
Jason
Committed.
2012-11-12 Ed Smith-Rowland <3dw...@verizon.net>
* parser.c (cp_parser_objc_class_ivars):
Index declspecs.locations by ds_typedef rather than ds_thread.
Index: parser.c
==
> You can see from the changes of sse.md that this is functionally a no-op
> change.
Sseshuf replaces sselog.
So, do you mean it should be added with sselog instead of sseadd?
Adding it with sseadd (instead of sselog) influences the latency information.
Regards
Ganesh
-Original Message-
OK.
Jason
On 11/09/2012 08:33 PM, Sriraman Tallam wrote:
+ /* Skip calling decls_match for versioned functions. */
+ if (DECL_FUNCTION_VERSIONED (fn)
+ && DECL_FUNCTION_VERSIONED (TREE_PURPOSE (match)))
+ continue;
+ if (!decls_match (fn, TREE_PURPOSE (match)
On 11/11/2012 08:01 AM, Florian Weimer wrote:
I'm not sure if this sufficiently far-reaching. It seems that this
doesn't allow me to implement a virtual function which takes a
std::string parameter in new-ABI-mode when the base class is also used
in old-ABI-mode.
Ah, yes; since virtual functio
On 11/11/2012 06:55 AM, Jonathan Wakely wrote:
For std::list I assume we want the attribute on the type itself, not
just member functions, since its size changes.
Yes. Note that I believe for the list and string changes we want to
change the default ABI for C++98 and C++11, not just C++11. A
The problem exposed in PR55030 (repeatable on x86_64-linux with -m32 at
r192676) is that the fake-frame-pointer "frame" is replaced with the
actual-frame-pointer "bp" in cse1, around the critical insn in
__builtin_setjmp_receiver that restores their defined offset. The
patch in PR55030/r192676 rem
> -Original Message-
> From: Jeff Law [mailto:l...@redhat.com]
> Sent: Friday, November 09, 2012 9:58 PM
> To: Bin Cheng
> Cc: gcc-patches@gcc.gnu.org
> Subject: Re: [PATCH Version 2][RFA]Improving register pressure directed
hoist
>
> On 11/07/2012 02:05 PM, Bin Cheng wrote:
>
> > 2012-
There are four in-tree target architectures that already use %|.
I think
it would be better if you made these new escapes target-specific.
Escaped curly braces cannot be target-specific since
do_assembler_dialects() in final.c ignores any % and considers '{' and
'}' to be alternative delimete
Hello,
The documentation of the standard name pattern sincos is wrong about the
the operands for sine and cosine. Operand 0 is cosine and operand 1 is
sine, not the other way around.
The attached patch corrects this.
Tested with 'make info dvi pdf'.
Committed as obvious as rev 193424.
Cheers,
O
Hello,
When refactoring the SH fsca patterns to use the sincos standard name
pattern I got the order of the sin and cos operands wrong.
This obvious patch fixes it.
Tested with 'make all-gcc' and
make -k check-gcc RUNTESTFLAGS="sh.exp=sh4a-fprun* --target_board=sh-sim
\{-m4a/-ml,-m4a/-mb}"
Comm
Hello,
The sh4a-fprun.c test case was bogus. It did not really use the fsca
instruction to do any calculations, because the things it tests could be
evaluated at compile time and folded away. The test case was supposed
to be failing since rev 188149, as I introduced a bug in the commit for
PR 53
On Sun, Nov 11, 2012 at 1:55 PM, Eric Botcazou wrote:
>> It fixes the problem. Can you check it in?
>
> Done.
>
> --
> Eric Botcazou
Now middle-end is fixed. I checked in this patch to remove "!" from riF->o
alternative. No regressions on Linux/x32.
Thanks.
--
H.J.
---
diff --git a/gcc/Chan
On Sun, 4 Nov 2012, Gerald Pfeifer wrote:
> Looking at that page, I noticed a lot of broken navigation (broken
> images,...). This removes that navigation and also address the worst
> in terms of broken markup.
>
> Still bin/preprocess excludes this file from the regular treatment
> that our gene
From: Eric Botcazou
Date: Sun, 11 Nov 2012 23:28:38 +0100
>> Eric and Rainer, I think that functionally this patch is fully ready
>> to go into the tree except for the Solaris aspects which I do not have
>> the means to work on. Have either of you made any progress in this
>> area?
>
> Rainer,
> I comitted it this afternoon, hope everything is fine now.
Yes, thanks.
--
Eric Botcazou
> > MThis patch looks OK. i will re-test and commit it on the afernoon.
>
> Ping. The ICE shows up when building Python.
I comitted it this afternoon, hope everything is fine now.
Honza
>
> --
> Eric Botcazou
> Eric and Rainer, I think that functionally this patch is fully ready
> to go into the tree except for the Solaris aspects which I do not have
> the means to work on. Have either of you made any progress in this
> area?
Rainer, could you post an excerpt of the man page of a recent 'as' supportin
This fixes a pasto in the --with-cpu=niagara4 support on Solaris, as well as
applies a minor tweak to config/sparc/sparc.h.
Tested on SPARC/Solaris, applied on the mainline and 4.7 branch.
2012-11-11 Eric Botcazou
* config/sparc/sparc.h (AS_NIAGARA3_FLAG): Tweak.
* config/sp
> It fixes the problem. Can you check it in?
Done.
--
Eric Botcazou
On Sun, Nov 11, 2012 at 7:01 AM, Eric Botcazou wrote:
>> This patch also handles SIGN_EXTEND. Tested on Linux/x32. OK to
>> install?
>
> I'd cautious here, that's uncharted territory and the SIGN_EXTEND case isn't
> covered by your testing.
>
>> 2012-11-10 H.J. Lu
>>
>> PR middle-end/55
On Fri, Nov 9, 2012 at 9:23 AM, Uros Bizjak wrote:
> On Fri, Nov 9, 2012 at 10:17 AM, H.J. Lu wrote:
>
>> Since x32 runs in 64-bit mode, for address -0x4300(%rax), hardware
>> sign-extends displacement from 32-bits to 64-bits and adds it to %rax.
>> But x32 wants 32-bit -0x4300, not 64-bi
On Fri, Nov 9, 2012 at 4:39 AM, Gopalasubramanian, Ganesh
wrote:
> Changes done with respect to the review comments.
> Conditionally setting "sseshuf" type attribute has been removed.
> Instead new attribute is added and is included for other attribute
> calculations.
>
> The patch is attached a
On Sun, Nov 11, 2012 at 7:36 PM, Uros Bizjak wrote:
> Regarding vzeroupper insertion pass - we will use gcc pass manager to
> insert a target-dependant pass directly after reload ...
... like attached patch. The patch inserts vzeroupper pass directly
after reload, so spills from 256bit registers
On 11 November 2012 19:26, Andreas Schwab wrote:
> Whitespace is apparently significant for , you get something
> like
>
>
>
> in the HTML output.
Huh, thanks for noticing it. Assuming it passes 'make
doc-xml-validate-docbook' the patch is OK.
On 11 November 2012 19:18, Andreas Schwab wrote:
> /usr/share/xml/docbook/stylesheet/nwalsh5/current is the place used by
> openSUSE.
>
> Andreas.
>
> * acinclude.m4 (GLIBCXX_CONFIGURE_DOCBOOK): Also check for
> /usr/share/xml/docbook/stylesheet/nwalsh5/current.
> * configur
Per the GCC Coding Conventions, "builtin" is not a word, and the
preferred terminology already used in most places is "built-in
function". I've checked in this patch to tidy up a bunch of incorrect
uses of "builtin" in extend.texi.
-Sandra
2012-11-11 Sandra Loosemore
gcc/
Whitespace is apparently significant for , you get something
like
in the HTML output.
Andreas.
* doc/xml/chapter.txml: Remove whitespace inside .
* doc/xml/class.txml: Likewise.
* doc/xml/manual/abi.xml: Likewise.
* doc/xml/manual/algorithms.xml: Likewise.
Tested on hppa2.0w-hp-hpux11.11 and hppa64-hp-hpux11.11. Committed to trunk.
Dave
--
J. David Anglin dave.ang...@nrc-cnrc.gc.ca
National Research Council of Canada (613) 990-0752 (FAX: 952-6602)
2012-11-11 John David Anglin
* gcc.dg/tort
/usr/share/xml/docbook/stylesheet/nwalsh5/current is the place used by
openSUSE.
Andreas.
* acinclude.m4 (GLIBCXX_CONFIGURE_DOCBOOK): Also check for
/usr/share/xml/docbook/stylesheet/nwalsh5/current.
* configure: Regenerate.
diff --git a/libstdc++-v3/acinclude.m4 b/libstd
Hello!
Attached patch uses for_each_rtx function to find AVX256 registers in
the RTX. The patch also implements better MODE_AFTER handling: we
already switched to DIRTY state if AVX256 is referenced through
MODE_NEEDED. We just force CLEAN state after vzeroupper/vzero and for
call instructions wit
On 2012.11.07 at 15:13 +0100, Martin Jambor wrote:
> Hi,
>
> On Tue, Nov 06, 2012 at 04:22:41PM +0100, Jan Hubicka wrote:
> > >
> > > 2012-11-05 Martin Jambor
> > >
> > > PR tree-optimization/53787
This caused:
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=55253
--
Markus
Hi,
the cunroll-1.c testcase has bogus count on number of unrollings needed. This
is my mistake:
I originally introudced the testcase for cunroll pass where the number is
different due
to loop header copying.
Regtested x86_64-linux, comitted.
Honza
Index: ChangeLog
=
> MThis patch looks OK. i will re-test and commit it on the afernoon.
Ping. The ICE shows up when building Python.
--
Eric Botcazou
Hi,
this should be last of inline hint patches for 4.8. It makes inliner to preffer
inlining
of functions where array index will become constant. This helps especially
accesses to
Fortran arrays with array descriptors.
It helps some of Fotran benchmarks, most importantly probably to fotigue, whe
> Bootstrapped on x86_64-unknown-linux-gnu, testing in progress. OK if it
> passes?
>
> Ciao!
> Steven
>
> gcc/
> PR middle-end/55263
> * postreload.c (rest_of_handle_postreload): With non-call
> exceptions, if edges are purged call cleanup_cfg to remove unreachable
> blocks.
>
> This patch also handles SIGN_EXTEND. Tested on Linux/x32. OK to
> install?
I'd cautious here, that's uncharted territory and the SIGN_EXTEND case isn't
covered by your testing.
> 2012-11-10 H.J. Lu
>
> PR middle-end/55247
> PR middle-end/55259
> * emit-rtl.c (adjust_add
On 11/06/2012 05:03 AM, Jason Merrill wrote:
+The @code{abi_tag} attribute can be applied to a function or class
+declaration. It modifies the mangled name of the function or class to
+incorporate the tag name, in order to distinguish the function or
+class from an earlier version with a differe
Hi Steven,
On Sat, Nov 10, 2012 at 3:00 PM, Thomas Koenig wrote:
I wrote:
after the dicsussion on c.l.f, it become clear that passing a DO loop
variable to an INTENT(OUT) or INTENT(INOUT) dummy argument is an error.
The attached patch throws an error for both cases.
But should we really iss
Hi,
On 11/11/2012 12:55 PM, Jonathan Wakely wrote:
On 11 November 2012 02:27, Jason Merrill wrote:
Is this patch OK for trunk? Does someone on the library team want to look
at the other cases?
The patch looks good to me. Thanks for implementing the attribute.
I'll start looking at some of th
On 11 November 2012 02:27, Jason Merrill wrote:
>
> Is this patch OK for trunk? Does someone on the library team want to look
> at the other cases?
The patch looks good to me. Thanks for implementing the attribute.
I'll start looking at some of the other cases.
For std::list I assume we want th
On 11 November 2012 09:58, Gerald Pfeifer wrote:
> On Mon, 5 Nov 2012, Jonathan Wakely wrote:
>> I'd use http://pubs.opengroup.org/onlinepubs/9699919799/functions/select.html
>
> Thanks, Jonathan. I committed the patch below.
Great, thanks.
>> Are you sure the other links haven't already been fi
On Mon, 5 Nov 2012, Jonathan Wakely wrote:
> I'd use http://pubs.opengroup.org/onlinepubs/9699919799/functions/select.html
Thanks, Jonathan. I committed the patch below.
> Are you sure the other links haven't already been fixed in the
> libstdc++ sources?
>
> The copies on the website are usual
Hello guys!
This is a PING. Could you pls have a look?
Thanks, K
>
> Is it OK?
>
> Thanks, K
On Sat, 10 Nov 2012, Jason Merrill wrote:
As mentioned in http://gcc.gnu.org/wiki/Cxx11AbiCompatibility, C++11 changes
the return type of complex::real and imag, leading to a binary
incompatibility between C++98 and C++11 code if the functions are used
without inlining. This patch adds an ABI
46 matches
Mail list logo
