This fixes minor issues in the recently added tests for 71947.
First in two tests we use an x86 opcode. We don't actually assemble the
tests, so it really doesn't matter, but just to make it clear the test
is not x86 specific, the opcode was changed to "xyzzy" :-)
One test is dependent on b
On 10/21/2016 04:34 AM, Prathamesh Kulkarni wrote:
On 20 October 2016 at 15:02, Richard Biener wrote:
On Wed, 19 Oct 2016, Jeff Law wrote:
On 10/15/2016 11:59 PM, Prathamesh Kulkarni wrote:
Hi,
After approval from Bernd Schmidt, I committed the patch to remove
optab functions for
sdivmod_opt
This patch to libgo fixes the expected alignment of int64 types on
32-bit SPARC. Without this most calls to select fail at runtime, as
the compiler and the library disagree about the expected size of the
hselect struct. Bootstrapped and ran Go testsuite on
x86_pc-linux-gnu, bootstrapped and ran a
On Fri, Oct 21, 2016 at 02:38:11PM -0600, Tom Tromey wrote:
> > "Jakub" == Jakub Jelinek writes:
>
> Jakub> Also, as this effectively requires the latest unreleased GDB under the
> Jakub> default options for something that has been working previously, I
> wonder
> Jakub> if it e.g. for some
On Oct 21, 2016, at 12:47 PM, Martin Sebor wrote:
>
> The latest patch works as expected for me, both with an operand
> and with stdin. But since I'm not empowered to approve it one
> of the others reviewers will need to give it their blessing.
Seems fine from a test suite perspective, but not
On Fri, Oct 21, 2016 at 4:16 AM, Rainer Orth
wrote:
>
>> This patch to libgo rewrites the interface code from C to Go.
>>
>> I started to copy the Go 1.7 interface code, but the gc and gccgo
>> representations of interfaces are too different. So instead I rewrote
>> the gccgo interface code from
As noted in the BZ, jump threading is isolating a path which contains a
b_c_p call. The result of the isolation is that on the original path,
b_c_p continues to return 0 (not-constant), but on the isolated path it
returns 1 (because the feeding PHI argument is constant).
That in turn causes
Hi,
I've been meaning for some time now to do a better job handling strength
reduction candidates where the stride of the candidate and its basis
involves a cast (usually widening) from another type. The two PRs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=71490 and
https://gcc.gnu.org/bugzilla/s
On 10/20/16 06:42, Jeff Law wrote:
>> On 6/4/16 21:25, cheng...@emindsoft.com.cn wrote:
>>> From: Chen Gang
>>>
>>> r10 may also be as parameter stack pointer for the nested function, so
>>> need save it before call mcount.
>>>
>>> Also clean up code: use '!' instead of "== 0" for checking
>>> sta
The quoting in the diagnostic should be %<&&%>, not '&&'.
--
Joseph S. Myers
jos...@codesourcery.com
Hi,
this implements support for signed overflow arithmetic on PowerPC. It's an
implementation for Power ISA v2.0x, i.e. it doesn't take account the new OV32
flag introduced in v3.0. It doesn't implement unsigned overflow arithmetic
because my understanding is that the generic support already
> Début du message réexpédié :
>
> De: Dominique d'Humières
> Objet: Rép : [Patch, fortran] PR69834 - Collision in derived type hashes
> Date: 22 octobre 2016 à 01:04:21 UTC+2
> À: Paul Richard Thomas
> Cc: Andre Vehreschild , fort...@gcc.gnu.org, gcc-patches List
>
>
> Dear Paul,
>
> If I
Bug 78039 complains that the fix for c++/71912 recently backported
to the GCC 6 branch causes GCC 6 to reject Glibc tests that expect
to be able to define structs with multiple flexible array members,
despite it violating the C standard(*).
The rejected code is unsafe and was intended to be rejec
All,
The attached patch fixes PR fortran/78033. This was a REAL pain
to fix because Fortran overloads REAL as an intrinsic type and
an intrinsic subprogram.
gfc_match_type_spec() in match.c is used to match Fortran 2003
type-specs in things like array constructors and TYPE IS statements.
At some
David Malcolm writes:
>> gcc/go/ChangeLog:
>> > * go-lang.c (go_langhook_type_for_mode): Remove redundant cast
>> > from result of GET_MODE_CLASS. Minor formatting fixes.
This is OK.
Thanks.
Ian
On Fri, Oct 21, 2016 at 9:28 AM, James Greenhalgh
wrote:
> On Fri, Oct 21, 2016 at 04:57:22PM +0100, Richard Earnshaw (lists) wrote:
>> On 21/10/16 14:59, James Greenhalgh wrote:
>> > On Sat, Oct 15, 2016 at 07:38:40PM -0700, Andrew Pinski wrote:
>> >> On Wed, Nov 25, 2015 at 11:59 AM, Andrew Pins
On 10/21/2016 04:37 PM, Joseph Myers wrote:
The quoting in the diagnostic should be %<&&%>, not '&&'.
Presumably same for '*' (i.e., %<*%>).
But I would actually suggest a somewhat more formal phrasing than
"better use xxx here" such as "suggest %<&&%> instead" or something
akin to what's alre
On 10/22/16 04:17, Martin Sebor wrote:
> On 10/21/2016 04:37 PM, Joseph Myers wrote:
>> The quoting in the diagnostic should be %<&&%>, not '&&'.
>
> Presumably same for '*' (i.e., %<*%>).
>
> But I would actually suggest a somewhat more formal phrasing than
> "better use xxx here" such as "suggest
Hi Steve,
Thanks for persevering with this. The patch looks good to me. If it
has regtested OK, please feel free to commit.
Cheers
Paul
On 22 October 2016 at 02:22, Steve Kargl
wrote:
> All,
>
> The attached patch fixes PR fortran/78033. This was a REAL pain
> to fix because Fortran overloads
101 - 119 of 119 matches
Mail list logo
