Installed now, after an exchange with Igor and monitoring the mirror
for ten days.
Thanks, Igor!
Gerald
On Mon, 9 Apr 2012, Gerald Pfeifer wrote:
> I was just going to add your server by means of the patch below,
> alas all attempts to rach the server time out. What's going on?
>
> Gerald
>
>
On 16 April 2012 21:28, Jonathan Wakely wrote:
> I have a patch to add the checks to
And here it is, only checking in debug mode because noone objected to
that suggestion.
* include/debug/forward_list (forward_list::splice_after): Check
allocators are equal.
* src/c++11/d
* include/bits/ptr_traits.h (pointer_traits::rebind): Make public.
* testsuite/20_util/pointer_traits/requirements/typedefs.cc: Check
rebind works.
Tested x86_64linux, committed to trunk and will commit to 4.7 soon.
commit 6141cdceb14025ef258b8809301558f5962bf7ab
Author: Jo
The allocator_traits wrapper is missing difference_type, noticed while
making vstring allocator-aware.
* include/ext/alloc_traits.h (__alloc_traits::difference_type):
Define.
Tested x86_64-linux, committed to trunk.
commit 7a3e74660df7df20bebb7676cd9142841637ba40
Author: Jonathan Wa
Thomas: Please write branch changes not in libgfortran/ChangeLog but in
libgfortran/ChangeLog.fortran-dev; that avoids merge problems when
updating from the trunk.
The branch (rev. 186674) now matches the trunk (Rev. 185178->186672).
Unfortunately, there are still about a dozen failures.
Tob
As described by Linus here:
http://lkml.indiana.edu/hypermail/linux/kernel/0611.3/1020.html
Wshadow warns whenever any declaration shadows a global function
declaration. This is almost always noise, since most (always?) of the
time one cannot mistakenly replace a function by another variable. The
On Fri, Apr 20, 2012 at 12:18, Tobias Burnus wrote:
> Dear all,
>
> some compilers support using "q" to indicate quad precision, e.g. "4.0q0".
> Since GCC 4.7, gfortran supports this vendor extension in the source code.
> However, READing the floating-point number "4.0q0" was not supported.
>
> Th
On Thu, Apr 19, 2012 at 5:47 AM, Richard Guenther wrote:
>
> This fixes PR53031 and reverts back to using the number of latch
> iterations in VRP when computing ranges for induction variables.
> Now after ira_allocno_object_iter_cond is fixed this no longer
> breaks bootstrap.
>
> Bootstrapped on
On Sat, Apr 21, 2012 at 3:43 PM, H.J. Lu wrote:
> On Thu, Apr 19, 2012 at 5:47 AM, Richard Guenther wrote:
>>
>> This fixes PR53031 and reverts back to using the number of latch
>> iterations in VRP when computing ranges for induction variables.
>> Now after ira_allocno_object_iter_cond is fixed
On Sun, Apr 22, 2012 at 10:14 AM, H.J. Lu wrote:
> On Thu, Apr 19, 2012 at 5:47 AM, Richard Guenther wrote:
>>
>> This fixes PR53031 and reverts back to using the number of latch
>> iterations in VRP when computing ranges for induction variables.
>> Now after ira_allocno_object_iter_cond is fixed
On Sun, Apr 22, 2012 at 10:50 AM, Manuel López-Ibáñez
wrote:
> This patch makes Wpedantic the canonical form of -pedantic. This makes
> -Wno-pedantic, -Werror=pedantic, #pragma diagnostics and other parts
> of the diagnostic machinery that expect warning options to start with
> -W, work with -Wped
On Sun, Apr 22, 2012 at 11:00 AM, Manuel López-Ibáñez
wrote:
> As described by Linus here:
> Interestingly, the C++ FE does not warn for this case, but it is not
> very clear to me where this decision is taken.
C++'s type system does not make it very likely to get into the
troubles that -Wshadow
Hi,
tested x86_64-linux, committed mainline and branch.
Thanks,
Paolo.
///
2012-04-22 Paolo Carlini
PR libstdc++/53067
* include/bits/hashtable_policy.h: Change inheritances to public.
* testsuite/23_containers/unordered_map/requirements/53067.cc:
Greetings,
The patch below is needed for google/integration branch:
we want to be able build libgomp.a with -fPIC, be able to link it into a
shared library, and be able to dlopen that library without running out of
static TLS space (-ftls-model=initial-exec precludes that last part).
Google ref b
On Jun 4, 2011, Alexandre Oliva wrote:
> On Feb 15, 2011, Alexandre Oliva wrote:
>> VTA only tracks locations of gimple regs, while addressable variables
>> still use the old var tracking strategy. This means addressable
>> variables, during var-tracking, got locations that were not based on
>
On 22 April 2012 19:25, Gabriel Dos Reis wrote:
> On Sun, Apr 22, 2012 at 10:50 AM, Manuel López-Ibáñez
> wrote:
>> This patch makes Wpedantic the canonical form of -pedantic. This makes
>> -Wno-pedantic, -Werror=pedantic, #pragma diagnostics and other parts
>> of the diagnostic machinery that ex
Okay for google/integration.
http://codereview.appspot.com/6107046/
This patch to gcc and libgo changes the Sizeof values in the syscall
package from var to const. This matches the master Go library, and
makes the names easier to use because the consts are untyped.
Bootstrapped and ran Go testsuite on x86_64-unknown-linux-gnu.
Committed to mainline and 4.7 branch.
On 04/22/2012 02:42 PM, Manuel López-Ibáñez wrote:
Which seems to suggest that we add an option name for each pedwarn
enabled by default. Is this also what you suggest?
I agree with this, and I think that's also what Gaby was suggesting by
"finer-grained".
Jason
When we aren't printing the expression, please print the type instead;
in a template it might not be clear what the type of the expression
we're complaining about works out to be.
Jason
On 22 April 2012 21:15, Jason Merrill wrote:
> On 04/22/2012 02:42 PM, Manuel López-Ibáńez wrote:
>>
>> Which seems to suggest that we add an option name for each pedwarn
>> enabled by default. Is this also what you suggest?
>
>
> I agree with this, and I think that's also what Gaby was suggesting
On 22 April 2012 21:20, Jason Merrill wrote:
> When we aren't printing the expression, please print the type instead; in a
> template it might not be clear what the type of the expression we're
> complaining about works out to be.
I can do that. Should I also change it in the C FE?
Cheers,
Manu
This patch to libgo defines more structs in the syscall package,
corresponding to structs defined in the master library. Bootstrapped
and ran Go testsuite on x86_64-unknown-linux-gnu. Committed to trunk
and 4.7 branch.
Ian
diff -r c7f1cc64c45f -r 8d768cdf12ce libgo/mksysinfo.sh
--- a/libgo/mksy
This patch to libgo provides more fixes to the syscall package to make
it more like the master Go library. This is based on running the
procedure described at
http://code.google.com/p/go/issues/detail?id=1790#c12 . Bootstrapped
and ran Go testsuite on x86_64-unknown-linux-gnu. Committed to mainl
On Sun, Apr 22, 2012 at 2:15 PM, Jason Merrill wrote:
> On 04/22/2012 02:42 PM, Manuel López-Ibáńez wrote:
>>
>> Which seems to suggest that we add an option name for each pedwarn
>> enabled by default. Is this also what you suggest?
>
>
> I agree with this, and I think that's also what Gaby was s
On Sun, Apr 22, 2012 at 2:38 PM, Manuel López-Ibáñez
wrote:
> For example, -Wmain is enabled by default but also by -Wall and
> -pedantic. However, -Werror=all does not enable -Werror=main. Is this
> a bug or the desired behaviour?
this particular one is a bug.
On 22 April 2012 22:09, Gabriel Dos Reis wrote:
> On Sun, Apr 22, 2012 at 2:15 PM, Jason Merrill wrote:
>> On 04/22/2012 02:42 PM, Manuel López-Ibáńez wrote:
>>>
>>> Which seems to suggest that we add an option name for each pedwarn
>>> enabled by default. Is this also what you suggest?
>>
>>
>>
On Sun, Apr 22, 2012 at 3:23 PM, Manuel López-Ibáñez
wrote:
> On 22 April 2012 22:09, Gabriel Dos Reis
> wrote:
>> On Sun, Apr 22, 2012 at 2:15 PM, Jason Merrill wrote:
>>> On 04/22/2012 02:42 PM, Manuel López-Ibáńez wrote:
Which seems to suggest that we add an option name for each pe
Change the dynamic linker path for ARM hard float executables.
Matches the path discussed and agreed on last week[1]. Carlos will
follow up with the matching patch to GLIBC[2]. I'm happy to if he's
busy.
OK for trunk?
-- Michael
[1] http://sourceware.org/ml/libc-ports/2012-04/msg00060.html
[2]
Hi,
when building Mozilla we ICE in expr_invariant_in_loop_p walking ARRAY_REF,
because one of the
fields is NULL (that is allowed)
Bootstrapped/regtested x86_64-linux, OK?
Honza
* tree-ssa-loop-ivopts.c (expr_invariant_in_loop_p): Bail out at NULL
tree refs.
Index: tree-ssa-loop-ivopt
On Tue, Apr 17, 2012 at 11:12 AM, Alan Modra wrote:
> This fixes a lot of confusion in rs6000_frame_related call arguments.
> At the time rs6000_frame_related first appeared, the prologue only
> used sp_reg_rtx (r1) or frame_ptr_rtx (r12) as frame_reg_rtx to access
> register save slots. If r12 w
On Tue, Apr 17, 2012 at 11:13 AM, Alan Modra wrote:
> This continues the prologue and epilogue cleanup. Not many user
> visible changes here, except for:
> - a bugfix to the LR save RTL emitted by rs6000_emit_savres_rtx which
> may affect SPE,
> - a bugfix for SPE code emitted when using a stati
On Sun, Apr 22, 2012 at 4:04 PM, Jan Hubicka wrote:
> Hi,
> when building Mozilla we ICE in expr_invariant_in_loop_p walking ARRAY_REF,
> because one of the
> fields is NULL (that is allowed)
>
> Bootstrapped/regtested x86_64-linux, OK?
>
> Honza
>
> * tree-ssa-loop-ivopts.c (expr_invarian
Add new option, -Wreserved-user-defined-literal.
This option, which is enabled by default, causes the preprocessor to warn
when a string or character literal is followed by a ud-suffix which does
not begin with an underscore. According to [lex.ext]p10, this is
ill-formed.
Also modifies the prepr
Could you try to get this into mainline instead of just the google
branches? In http://gcc.gnu.org/PR52538, Jonathan sounded like he'd
consider accepting it.
On Sun, Apr 22, 2012 at 7:54 PM, Ollie Wild wrote:
> Add new option, -Wreserved-user-defined-literal.
>
> This option, which is enabled by
Let's let the discussion _start_ before assuming it'll be protracted.
;) I don't think it'll kill us to run the next round of testing in
C++98 mode. If the patch doesn't look close to acceptance in a couple
days, I think it'll make sense to put the then-current version into
the google branches whil
Okay, I'll send out a trunk patch for review now, too.
Ollie
On Sun, Apr 22, 2012 at 10:29 PM, Jeffrey Yasskin wrote:
>
> Let's let the discussion _start_ before assuming it'll be protracted.
> ;) I don't think it'll kill us to run the next round of testing in
> C++98 mode. If the patch doesn't
Thanks!
On Sun, Apr 22, 2012 at 8:32 PM, Ollie Wild wrote:
> Okay, I'll send out a trunk patch for review now, too.
>
> Ollie
>
> On Sun, Apr 22, 2012 at 10:29 PM, Jeffrey Yasskin wrote:
>>
>> Let's let the discussion _start_ before assuming it'll be protracted.
>> ;) I don't think it'll kill us
Add new option, -Wreserved-user-defined-literal.
This option, which is enabled by default, causes the preprocessor to warn
when a string or character literal is followed by a ud-suffix which does
not begin with an underscore. According to [lex.ext]p10, this is
ill-formed.
Also modifies the prepr
On Sun, Apr 22, 2012 at 10:50 AM, Paul Pluzhnikov
wrote:
> Greetings,
>
> The patch below is needed for google/integration branch:
> we want to be able build libgomp.a with -fPIC, be able to link it into a
> shared library, and be able to dlopen that library without running out of
> static TLS spa
On Sun, Apr 22, 2012 at 10:09:46PM -0700, Andrew Pinski wrote:
> On Sun, Apr 22, 2012 at 10:50 AM, Paul Pluzhnikov
> wrote:
> > Greetings,
> >
> > The patch below is needed for google/integration branch:
> > we want to be able build libgomp.a with -fPIC, be able to link it into a
> > shared librar
On Sun, Apr 22, 2012 at 10:09 PM, Andrew Pinski wrote:
> IIRC the main reason is because the slow down from not using
> initial-exec model for GOMP is a lot.
Given a choice between "slows down a lot" and "doesn't work at all",
we prefer the former ;-)
I see that we are not alone:
http://old.nab
42 matches
Mail list logo
