Jakub Jelinek wrpte:
On Fri, May 09, 2014 at 09:35:03PM +0200, Tobias Burnus wrote:
Fortran - as also C and C++ - does not provide much support for
aligned memory. [Well, C++ has now alignas and align(ment_)of.] But
contrary to C/C++, using library functions for dynamic allocation is
more diff
On 9.5.2014 23:14, David Malcolm wrote:
> This patch is 2.2MB in size, so I've uploaded it to:
> http://dmalcolm.fedorapeople.org/gcc/large-patches/6fb783b39f914574a1889aa51d06c08cf55678b4-0002-Autogenerated-part-of-introduction-of-GCC_OPTION-mac.patch
>
> (even just the ChangeLog would be 160KB)
Thanks for catching that, I will fix it.
Teresa
On Fri, May 9, 2014 at 7:49 PM, Andi Kleen wrote:
> Teresa Johnson writes:
>>
>> Passes regression tests. Ok for google branches?
>
> +{
> + char parameter[1000];
> + sprintf (parameter, "%s=%ld",
> + curr_parm->macro_na
Teresa Johnson writes:
>
> Passes regression tests. Ok for google branches?
+{
+ char parameter[1000];
+ sprintf (parameter, "%s=%ld",
I don't think we want cp_parser_class_name to find enums; better I think
to change cp_parser_qualifying_entity to use something other than
cp_parser_type_name to look for enums.
Jason
Georg-Johann Lay writes:
> This patch adds GCC Runtime Library Exception to files that go into
> libgcc because libgcc2.c includes tm.h and libgcc_tm.h.
>
> Most of these files contain much code, some used by libgcc, some
> not. Some potential users of (lib)gcc have objections that missing RLE
>
On Fri, May 09, 2014 at 05:14:45PM -0400, David Malcolm wrote:
> GCC's code is full of references to options like:
>
> static bool
> gate_vrp (void)
> {
> return flag_tree_vrp != 0;
> }
>
> where "flag_tree_vrp" is actually an autogenerated macro to:
>
> global_options.x_flag_tree_
I guess the subject says it all. Tested on Linux x86-64.
diff --git a/gcc/cp/lambda.c b/gcc/cp/lambda.c
index 3ce9ebb..bb6014b 100644
--- a/gcc/cp/lambda.c
+++ b/gcc/cp/lambda.c
@@ -367,10 +367,7 @@ build_capture_proxy (tree member)
object = TREE_OPERAND (object, 0);
/* Remove the __ inse
Richard/Vlad,
Richard Sandiford writes:
> Kyrill Tkachov writes:
> > On 03/05/14 20:21, Richard Sandiford wrote:
...snip...
> > Hi all,
> > This caused some testsuite failures on arm:
> > FAIL: gcc.target/arm/vfp-ldmdbd.c scan-assembler fldmdbd
> > FAIL: gcc.target/arm/vfp-ldmdbs.c scan-assemb
On Fri, May 9, 2014 at 12:20 PM, DJ Delorie wrote:
>
> The key to the msp430 change is that I reviewed every pattern that
> used the predicates, and only changed the ones where the pattern was
> known to be volatile-safe. For the RL78, the devirtualizer may make a
> pattern non-volatile-safe, and
This patch is 2.2MB in size, so I've uploaded it to:
http://dmalcolm.fedorapeople.org/gcc/large-patches/6fb783b39f914574a1889aa51d06c08cf55678b4-0002-Autogenerated-part-of-introduction-of-GCC_OPTION-mac.patch
(even just the ChangeLog would be 160KB)
GCC's code is full of references to options like:
static bool
gate_vrp (void)
{
return flag_tree_vrp != 0;
}
where "flag_tree_vrp" is actually an autogenerated macro to:
global_options.x_flag_tree_vrp
This is deeply confusing to a newbie (and indeed still to me after ~two
years of
This is the hand-written half of the patch series.
gcc/ada/gcc-interface/misc.c requires some special care: it "undeffed"
some options macros in order to support explicit variables for 5
options, whilst using other options via macros. I ported this
file by hand, disabling the refactoring script f
On Fri, May 9, 2014 at 7:35 AM, Andrew Burgess wrote:
>if ((AUTO_DEMANGLING || GNU_DEMANGLING))
> {
> success = gnu_special (work, &mangled, &decl);
> + if (!success)
> +{
> + delete_work_stuff (work);
> + string_delete (&de
The attached patch adds support for user-guided feedback-directed
library optimization, for google/gcc-4_8 initially (to be ported to
google/gcc-4_9).
Cc'ing Honza for any comments since I would eventually like to send
this and follow-on work to trunk.
Patch to add support for user-guided feedbac
On 04/25/14 05:39, Prathamesh Kulkarni wrote:
On Thu, Apr 24, 2014 at 9:18 PM, Diego Novillo wrote:
Please remember to send patches to gcc-patches.
I'm wondering if you couldn't just use std::string here. No need to change
one complicated allocation scheme with another.
Since we were using c
On 05/09/14 01:27, Marek Polacek wrote:
Another piece of puzzle. error_init didn't have a location parameter,
so initialization errors had wrong column info. This patch adds
location parameter to error_init and to a few of other functions.
The testcases show what has better column info now.
Not
On Fri, May 09, 2014 at 09:35:03PM +0200, Tobias Burnus wrote:
> >This patch attempts to implement first part of Fortran OpenMP 4.0 support,
> >The testsuite coverage could certainly be improved, volunteers for that
> >and/or bugreports will be certainly appreciated. Don't know how to actually
> >
On 04/25/14 09:03, Patrick Palka wrote:
I forgot the ChangeLog entry:
2014-04-25 Patrick Palka
* doc/invoke.texi: Fix typo.
* tree-vrp.c: Fix typos.
* gimple.c (infer_nonnull_range): Reorder operands of an &&
condition.
This is fine. Please install if yo
Thanks for the patch, and for pinging me directly.
It looks like you don't have a copyright assignment on file; this and
your other patch are small enough not to need one, but if you're going
to continue working on GCC it would be good to get on file so that it
doesn't hold up future patches.
On 04/17/14 07:58, Christian Bruel wrote:
Hello,
He is a new version of the patch. It hookizes the mode-setting and
mode-toggling macros. Split in 2 parts.
Successfully bootstrapped/regtested on ix86 and SH4/SH4a.
I was able to do a limited build on Epiphany, if someone could give it a
try on
Jakub noticed I got the exports wrong for std::regex_error.
Tested x86_64-linux, committed to trunk. Not planning to fix this on
the 4.9 branch.
commit 33024aacc31cc7a1fdaba934952150fdc3db7e3f
Author: Jonathan Wakely
Date: Fri May 9 13:51:39 2014 +0100
* config/abi/pre/gnu.ver (GLIBCXX_
On 04/17/14 10:43, Kyrill Tkachov wrote:
Hi all,
While looking at the build logs I noticed a warning while building
tree-ssa-loop-ivcanon.c about a potential use of an uninitialised variable.
This patchlet fixes that warning by initialising it to 0.
Tested arm-none-eabi.
Ok for trunk?
2014-04
Hi!
Jakub Jelinek wrote:
This patch attempts to implement first part of Fortran OpenMP 4.0 support,
The testsuite coverage could certainly be improved, volunteers for that
and/or bugreports will be certainly appreciated. Don't know how to actually
test aligned clauses, is there any way (except
On 04/29/14 08:14, Prathamesh Kulkarni wrote:
On Tue, Apr 29, 2014 at 7:37 PM, Prathamesh Kulkarni
wrote:
On Tue, Apr 29, 2014 at 6:49 PM, Manuel López-Ibáñez
wrote:
This hunk in your patch doesn't seem related:
Oops, Sorry. Removed that, and re-tested.
@@ -509,6 +509,9 @@ Warn about missi
Peter Bergner writes:
> Pat noticed that the updated pack03.c DFP test case is FAILing on
> older Power hardware without DFP support (ie, power5 and earlier).
> The problem is that the dg-require-effective-target dfprt use
> only protects against use on system without a DFP runtime, not
> a syste
The key to the msp430 change is that I reviewed every pattern that
used the predicates, and only changed the ones where the pattern was
known to be volatile-safe. For the RL78, the devirtualizer may make a
pattern non-volatile-safe, and many patterns are macros which are not
always volatile-safe.
On 04/21/14 10:57, David Malcolm wrote:
gcc/
* coretypes.h (gimple_try): New typedef.
(const_gimple_try): New typedef.
* gimple.h (gimple_statement_base::as_a_gimple_try): New.
* gimple-low.c (gimple_try_catch_may_fallthru): Require a
gimple_try rather th
On Fri, May 9, 2014 at 2:43 PM, Peter Bergner wrote:
> Pat noticed that the updated pack03.c DFP test case is FAILing on
> older Power hardware without DFP support (ie, power5 and earlier).
> The problem is that the dg-require-effective-target dfprt use
> only protects against use on system withou
On 04/21/14 10:57, David Malcolm wrote:
gcc/
* asan.c (insert_if_then_before_iter): Require a gimple cond
rathern than a plain gimple.
(instrument_mem_region_access): Strengthen local from gimple to
gimple_cond.
* cfgloopmanip.c (create_empty_if_region_on_
On 04/21/14 10:56, David Malcolm wrote:
gcc/
* coretypes.h (gimple_call): New typedef.
(const_gimple_call): New typedef.
* gimple.h (gimple_statement_base::as_a_gimple_call): New.
(gimple_statement_base::dyn_cast_gimple_call): New.
(gimple_call_reset_alias
http://gcc.gnu.org/ml/gcc-patches/2014-04/msg01108.html
Original Message
Subject: [RFC][PATCH] RL78 - Add predicates to reduce code bloat when
accessing volatile memory.
Date: Fri, 18 Apr 2014 21:15:01 +0200
From: Richard Hulme
To: GCC Patches
Hi,
This patch adds predica
On 05/09/14 12:47, David Malcolm wrote:
Jeff: thanks. Note that I'm currently working on a grand renaming, as
per the subthread here:
http://gcc.gnu.org/ml/gcc-patches/2014-05/msg00346.html
so that all of these will be "gphi *" rather than "gimple_phi", with
analogous change to the other class
On 04/21/14 10:56, David Malcolm wrote:
gcc/
* gimple-iterator.h (struct gimple_phi_iterator): New subclass of
gimple_stmt_iterator with identical layout, but adding...
(gimple_phi_iterator::phi): ...new method, equivalent to
gsi_stmt (), but casting the underlying
On Fri, 2014-05-09 at 12:34 -0600, Jeff Law wrote:
> On 04/21/14 10:56, David Malcolm wrote:
> > gcc/
> > * tree-ssa-loop-ivopts.c (determine_biv_step): Require a gimple_phi.
> > (find_bivs): Convert local "phi" into a gimple_phi.
> > (mark_bivs): Likewise.
> OK when prerequisites have
On 04/21/14 10:56, David Malcolm wrote:
gcc/
* coretypes.h (gimple_return): New typedef.
(const_gimple_return): New typedef.
* gimple.h (gimple_statement_return): New subclass of
gimple_statement_with_memory_ops, adding the invariant that
stmt->code == GIM
Pat noticed that the updated pack03.c DFP test case is FAILing on
older Power hardware without DFP support (ie, power5 and earlier).
The problem is that the dg-require-effective-target dfprt use
only protects against use on system without a DFP runtime, not
a system without dfp hardware instruction
On 04/21/14 10:56, David Malcolm wrote:
gcc/
* tree-predcom.c (find_looparound_phi): Return a gimple_phi rather
than just a gimple.
(insert_looparound_copy): Require a gimple_phi rather than just a
gimple.
(add_looparound_copies): Convert local "phi" to be
On 04/21/14 10:56, David Malcolm wrote:
gcc/
* tree-ssa-loop-ivopts.c (determine_biv_step): Require a gimple_phi.
(find_bivs): Convert local "phi" into a gimple_phi.
(mark_bivs): Likewise.
OK when prerequisites have gone in.
Actually that's true for #17 & #18 as well.
j
On 04/21/14 10:56, David Malcolm wrote:
gcc/
* tree-ssa-loop-manip.c (add_exit_phi): Convert local "phi" to be a
gimple_phi.
(split_loop_exit_edge): Likewise for "phi" and "new_phi".
OK when prerequisites have gone in.
jeff
OK.
Jason
On 04/21/14 10:56, David Malcolm wrote:
gcc/
* tree-ssa-loop-niter.c (chain_of_csts_start): Return a gimple_phi
rather than a gimple.
(get_base_for): Likewise; convert local "phi" to be a gimple_phi.
(loop_niter_by_eval): Convert local "phi" to be a gimple_phi.
OK
On 04/21/14 10:56, David Malcolm wrote:
gcc/
* tree-ssa-phiprop.c (phiprop_insert_phi): Strengthen types of
parameter "phi" and local "new_phi" from gimple to gimple_phi.
OK when prerequisites have gone in.
jeff
On 04/21/14 10:56, David Malcolm wrote:
gcc/
* tree-parloops.c (struct reduction_info): Strengthen field "new_phi"
from gimple to gimple_phi.
(create_phi_for_local_result): Convert local "new_phi" to gimple_phi.
(loop_has_vector_phi_nodes): Require a gimple_phi rat
On 04/21/14 10:56, David Malcolm wrote:
gcc/
* tree-ssa-propagate.h (typedef ssa_prop_visit_phi_fn): Strengthen
type of parameter from gimple to gimple_phi.
* tree-complex.c (complex_visit_phi): Update signature of callback
implementation accordingly.
* tr
On 04/21/14 10:56, David Malcolm wrote:
gcc/
* coretypes.h (gimple_phi): New typedef.
(const_gimple_phi): New typedef.
* gimple.h (gimple_statement_base::as_a_gimple_phi): New.
(gimple_statement_base::dyn_cast_gimple_phi): New.
* gdbhooks.py (build_pretty
> Well, it wasn't a hard requirement, it is just that the library has
> to use a more complicated way to get the precision (use (unsigned
> TYPE)(-1) to get the unsigned max and compute the precision from
> that, probably).
We could define macros for the precision too, and we already know max
and
CWG DR 5 specifies that in the two-step copy-initialization process, the
temporary object of the target type which is created and then copied
into the target object has cv-unqualified type; this allows volatile
objects to be copy-initialized with the usual copy-constructor signature.
Tested x8
While I was working on reference binding diagnostics, I noticed a latent
bug with conversion10.C that I tracked down to a problem with the fix
for c++/22434: we don't want to do a bad conversion early if we will
find a better one later through overload resolution. So the first patch
fixes that
On Wed, May 7, 2014 at 12:06 PM, Kai Tietz wrote:
>
> this patch adds for Windows targets the define
> _GTHREAD_USE_MUTEX_INIT_FUNC, which is necessary as pthread-emulation
> for those targets are just handling pthread_mutext_init,
> othread_mutex_destroy proper.
>
> ChangeLog libgcc
>
> 2014-05-0
On May 9, 2014 5:53:06 PM CEST, Teresa Johnson wrote:
>Fixed a place where the wrong dump file was being used, leading to
>an inconsistency and seg fault when dump_file was non-NULL but
>cgraph_dump_file was NULL.
>
>Bootstrapped and tested on x86-64-unknown-linux-gnu. Ok for trunk?
OK.
Thanks,
Hi,
noticed only today that convert_nontype_argument calls
convert_nontype_argument_function which has error calls. Tested
x86_64-linux.
Thanks,
Paolo.
//
2014-05-09 Paolo Carlini
* pt.c (convert_nontype_argument_function): Add tsubst_flags_t
parameter
On Fri, 9 May 2014, Marek Polacek wrote:
> Another piece of puzzle. error_init didn't have a location parameter,
> so initialization errors had wrong column info. This patch adds
> location parameter to error_init and to a few of other functions.
> The testcases show what has better column info
On Wed, 7 May 2014 09:41:27 -0600
Tom Tromey wrote:
> Tom> The usual approach is some appropriate text somewhere on the GCC
> Tom> wiki (though I suppose a note in the mail archives would do in a
> Tom> pinch) along with a URL in a comment in the appropriate file
> Tom> (dwarf2.h or dwarf2.def).
On Fri, May 9, 2014 at 9:27 AM, Richard Earnshaw wrote:
> It was pointed out that AArch64 wasn't listed in backends.html.
>
> This is an attempt to rectify that. Hopefully I've got all the bits the
> right way around.
You missed:
r Port can switch between ILP32 and LP64 at runtime.
It was pointed out that AArch64 wasn't listed in backends.html.
This is an attempt to rectify that. Hopefully I've got all the bits the
right way around.
R.Index: htdocs/backends.html
===
RCS file: /cvs/gcc/wwwdocs/htdocs/backends.h
On Fri, May 09, 2014 at 11:03:35AM -0500, James Norris wrote:
> Removed three occurrences of a shadow variable.
>
> Bootstrapped and tested on x86-64-unknown-linux-gnu.
>
> 2014-05-09 James Norris
>
>* omp-low.c (expand_parallel_call): Remove shadow variable.
>(expand_omp_taskreg): Li
Hi!
Removed three occurrences of a shadow variable.
Bootstrapped and tested on x86-64-unknown-linux-gnu.
2014-05-09 James Norris
* omp-low.c (expand_parallel_call): Remove shadow variable.
(expand_omp_taskreg): Likewise.
diff --git a/gcc/omp-low.c b/gcc/omp-low.c
index 453f580..5949
I discovered that the support for the documented -fdump-* options
"optimized", "missed", "note" and "optall" was missing. Added that and
fixed a minor typo in the documentation.
Bootstrapped and tested on x86-64-unknown-linux-gnu. Ok for trunk?
Thanks,
Teresa
2014-05-09 Teresa Johnson
Fixed a place where the wrong dump file was being used, leading to
an inconsistency and seg fault when dump_file was non-NULL but
cgraph_dump_file was NULL.
Bootstrapped and tested on x86-64-unknown-linux-gnu. Ok for trunk?
Thanks,
Teresa
2014-05-09 Teresa Johnson
* cgraphunit.c (ana
On Tue, Apr 29, 2014 at 6:16 AM, Trevor Saunders wrote:
>> >+# In any case if we got the diff on stdin then write the ChangeLog to
>> >stdout.
>>
>> Hm, this is breaks semantics: you only dump CL instead of CL+diff just
>> because diff comes from stdin. Perhaps we could append contents of
>> @di
On Mon, Apr 28, 2014 at 10:11 PM, wrote:
> 2014-04-28 Trevor Saunders
>
> * mklog: if in a git checkout try to get name and email from git.
> ---
> contrib/mklog | 14 ++
> 1 file changed, 14 insertions(+)
>
> diff --git a/contrib/mklog b/contrib/mklog
> index fb489b0..5f5
> This fixes a regression introduced with 4.8, ...
This caused pr61126.
Dominique
Fixes issue:
https://sourceware.org/bugzilla/show_bug.cgi?id=16817
A call to gnu_special within internal_cplus_demangle could cause memory
resources to be allocated, even if the demangle eventually fails. The
following call into demangle_prefix will then be passed some partially
initialised sta
[ping]
On 11 January 2014 19:19, Jonathan Wakely wrote:
> On 31 October 2013 16:46, Tom Tromey wrote:
>>> "Matthias" == Matthias Klose writes:
>>
>> Matthias> Starting with gdb 7.6, gdb can be linked with both Python 2.x
>> Matthias> and Python 3.x. Therefore the pretty printers should be
>
On 04/21/14 10:56, David Malcolm wrote:
gcc/
* coretypes.h (gimple_debug): New typedef.
(const_gimple_debug): New typedef.
* gimple.h (struct gimple_statement_debug): New subclass of
gimple_statement_with_ops, adding the invariant that
stmt->code == GIMPLE
On 04/21/14 10:56, David Malcolm wrote:
gcc/
* coretypes.h (gimple_label): New typedef.
(const_gimple_label): New typedef.
* gimple.h (struct gimple_statement_label): New subclass of
gimple_statement_with_ops, adding the invariant that
stmt->code == GIMPLE
On 04/21/14 10:56, David Malcolm wrote:
gcc/
* coretypes.h (gimple_assign): New typedef.
(const_gimple_assign): New typedef.
* gimple.h (struct gimple_statement_assign): New subclass of
gimple_statement_with_memory_ops, adding the invariant that
stmt->code
On 04/21/14 10:56, David Malcolm wrote:
gcc/
* coretypes.h (gimple_cond): New typedef.
(const_gimple_cond): Likewise.
* gimple.h (struct gimple_statement_cond): New subclass of
gimple_statement_with_ops, adding the invariant that
stmt->code == GIMPLE_COND.
This fixes a small bug in the conversion of the fortran front-end conversion to
wide-int.
PR fortran/61109
* trans-array.c (gfc_conv_array_initializer): Fix wide-int
conversion bug.
Index: trans-array.c
===
-
On 05/09/14 04:36, Richard Biener wrote:
On Thu, May 8, 2014 at 9:45 PM, H.J. Lu wrote:
On Thu, May 8, 2014 at 12:28 PM, Jeff Law wrote:
On 05/08/14 02:17, Ilya Enkovich wrote:
Right. Richi explicitly wanted the entire set approved before staging in
any of the bits.
I thought it would
This patch adds GCC Runtime Library Exception to files that go into libgcc
because libgcc2.c includes tm.h and libgcc_tm.h.
Most of these files contain much code, some used by libgcc, some not. Some
potential users of (lib)gcc have objections that missing RLE might "infect"
their target code.
2014-05-09 Kenneth Zadeck
PR middle-end/6
* fold-const.c (fold_binary_loc): Changed width of mask.
committed as revision 210274.
kenny
Index: gcc/fold-const.c
===
--- gcc/fold-const.c (revision 210253)
+++ gcc/fold-
This branch existed solely to facilitate merging of the port into GCC
mainline. It's been there for a while now, so this branch is now
defunct. Marked accordingly.
R.
Index: htdocs/svn.html
===
RCS file: /cvs/gcc/wwwdocs/htdocs/svn.
http://gcc.gnu.org/ml/gcc-cvs/2014-05/msg00311.html
Hi, applied this obvious patch to trunk in order to fix build warnings like
these:
gcc/config/avr/avr-fixed.md:434:59: warning: narrowing conversion of ‘-1’ from
‘int’ to ‘const unsigned int’ inside { } is ill-formed in C++11 [-Wnarrowing]
Hi, I just applied the following backport to 4.9:
http://gcc.gnu.org/ml/gcc-cvs/2014-05/msg00310.html
Thanks.
2014-04-16 Pitchumani Sivanupandi
Backport from 2014-04-16 trunk r209446
2014-04-16 Pitchumani Sivanupandi
* config/avr/avr-mcus.def: Correct typo for
> -Original Message-
> From: Richard Sandiford [mailto:rdsandif...@googlemail.com]
> Sent: 09 May 2014 12:07
> To: Andrew Bennett
> Cc: gcc-patches@gcc.gnu.org; Matthew Fortune; Saeed Ghazanfar; Rich Fuhler
> Subject: Re: [PATCH] Add support for MIPS r3 and r5
>
> Andrew Bennett writes:
>
Andrew Bennett writes:
>> > @@ -141,7 +151,8 @@ along with GCC; see the file COPYING3. If not see
>> >"%{EL:-m elf32lmip} \
>> > %{EB:-m elf32bmip} \
>> > %(endian_spec) \
>> > - %{G*} %{mips1} %{mips2} %{mips3} %{mips4} %{mips32} %{mips32r2}
>> %{mips64} \
>> > + %{G*} %{mips1} %
2014-05-08 17:04 GMT+04:00 Georg-Johann Lay :
> Some instructions like INC, DEC, NEG currently set cc0 to set_zn which is
> not the whole story because they also set the V flag. This leads to a wrong
> branch instruction in the remainder as cc0 is used. Fix is to do same as
> clobber cc0. For th
On Fri, May 9, 2014 at 8:49 AM, Jeff Law wrote:
> On 05/02/14 15:56, David Malcolm wrote:
>>
>> gcc/
>> * coretypes.h (gimple): Drop typedef.
>> (const_gimple): Likewise.
>> (gimple_seq): Convert from being a "gimple" to a "gimple_stmt *".
>
> So instead, we just want to co
On Fri, May 9, 2014 at 8:18 AM, Jeff Law wrote:
> On 05/07/14 14:52, Richard Sandiford wrote:
>>
>> I noticed for_each_rtx showing up in profiles and thought I'd have a go
>> at using worklist-based iterators instead. So far I have three:
>>
>>FOR_EACH_SUBRTX: iterates over const_rtx subrtxes
> -Original Message-
> From: Richard Sandiford [mailto:rdsandif...@googlemail.com]
> Sent: 08 May 2014 20:28
> To: Andrew Bennett
> Cc: gcc-patches@gcc.gnu.org; Matthew Fortune; Saeed Ghazanfar; Rich Fuhler
> Subject: Re: [PATCH] Add support for MIPS r3 and r5
>
> Andrew Bennett writes:
>
On Thu, May 8, 2014 at 9:45 PM, H.J. Lu wrote:
> On Thu, May 8, 2014 at 12:28 PM, Jeff Law wrote:
>> On 05/08/14 02:17, Ilya Enkovich wrote:
Right. Richi explicitly wanted the entire set approved before staging in
any of the bits.
>>>
>>>
>>> I thought it would be useful to h
Sorry, took longer than expected as I got distracted by some other patch.
I merged the whole patchset in a single patch as I was told the current setup
is actually more difficult to read.
Here are the updated ChangeLogs:
*** gcc/ChangeLog ***
2014-05-09 Thomas Preud'homme
PR tree-opt
Eric Botcazou writes:
>> Sure, that'd be fine by me. I'm not sure whether the backport has
>> been approved yet though.
>
> At least it looks fine to me...
OK, thanks. Richard also gave an RM's OK on IRC so I've now applied it.
> [I still don't grasp this "wait-a-little-before-backporting" pol
On 08/05/14 19:31, Richard Sandiford wrote:
> "Joseph S. Myers" writes:
>> On Thu, 8 May 2014, Ramana Radhakrishnan wrote:
>>
>>> Ramana Radhakrishnan
>>>
>>>* wide-int.cc (UTItype): Define.
>>>(UDWtype): Define for appropriate W_TYPE_SIZE.
>>
>> This breaks builds for 32-bit h
On 2014-04-28 18:24, Mike Stump wrote:
On Apr 28, 2014, at 12:32 AM, Sebastian
Huber wrote:
>On 2014-04-28 04:23, Mike Stump wrote:
>>On Apr 27, 2014, at 10:45 AM, Sebastian
Huber wrote:
>>>2014-04-27 Sebastian Huber
>>>
>>>* testsuite/lib/libffi.exp (load_gcc_lib): Register loaded li
On Fri, May 09, 2014 at 10:16:14AM +0200, Eric Botcazou wrote:
> > Sure, that'd be fine by me. I'm not sure whether the backport has
> > been approved yet though.
>
> At least it looks fine to me...
>
> [I still don't grasp this "wait-a-little-before-backporting" policy, it
> always
> leads to
+#if GCC_VERSION >= 3000 && (W_TYPE_SIZE == 32 || defined (__SIZEOF_INT128__))
W_TYPE_SIZE == 32 is always false and on 32-bit hosts,
__SIZEOF_INT128__ won't be defined.
Right, but we won't try to use TImode if in future we do revert to using
32-bit types for 32-bit hosts.
For now, I've re
Ping
On 14.04.2014 23:53, Momchil Velikov wrote:
Hello,
During overload resolution of function calls in the body of a lambda it
is possible to use an implicit/dummy 'this', which differs in const-ness
from the actual 'this' that would be captured, resulting in choosing a
non-const member functi
Hi,
this makes sure we always generate a static chain for nested functions at -O0
and force it onto the stack in this case so that DW_AT_static_link is always
meaningful (but not necessarily fully correct, see the audit trail).
FWIW this has been done for years in AdaCore's compilers.
Tested o
> Sure, that'd be fine by me. I'm not sure whether the backport has
> been approved yet though.
At least it looks fine to me...
[I still don't grasp this "wait-a-little-before-backporting" policy, it always
leads to forgotten patches that need to be applied just days ahead of releases
instead
On Thu, 8 May 2014, DJ Delorie wrote:
Assuming that the formula sizeof(type)*char_bit==precision works for all
It doesn't. THe MSP430 has __int20 for example.
Well, it wasn't a hard requirement, it is just that the library has to
use a more complicated way to get the precision (use (unsign
Boostrapped and tested on x86_64-unknown-linux-gnu, applied.
Richard.
2014-05-09 Richard Biener
* Makefile.in (GTFILES): Remove tree-ssa-propagate.c.
* tree-ssa-propagate.c: Do not include gt-tree-ssa-propagate.h.
(interesting_ssa_edges, varying_ssa_edges): Move out o
On 9 May 2014 14:00, Jeff Law wrote:
> On 05/08/14 02:06, Zhenqiang Chen wrote:
>>
>> Hi,
>>
>> Similar issue was discussed in thread
>> http://gcc.gnu.org/ml/gcc-patches/2013-04/msg01145.html. The patches
>> are close to Jeff's suggestion: "sink just the moves out of the
>> incoming argument regi
Another piece of puzzle. error_init didn't have a location parameter,
so initialization errors had wrong column info. This patch adds
location parameter to error_init and to a few of other functions.
The testcases show what has better column info now.
Note what I wrote in
http://gcc.gnu.org/bugzi
Eric Botcazou writes:
>> It has been two weeks since Richard commited this to trunk. Perhaps it's
>> ok to backport to 4.8 branch now?
>
> Richard, can you do that before the 4.8.3 release? Thanks in advance.
Sure, that'd be fine by me. I'm not sure whether the backport has
been approved yet th
On 5/9/2014 07:26, Jeff Law wrote:
> On 05/03/14 01:11, John Marino wrote:
>
> In config.gcc:
>
> +no | gnat | single)
> + # Let these non-posix thread selections fall through if requested
> Support for "gnat" as a thread model was removed in 2011. So I think
> you need to remove that c
98 matches
Mail list logo
