Hello!
... to make a couple of trivial cleanups.
2012-08-04 Uros Bizjak
* config/i386/i386.h (QI_REGNO_P): New define.
(ANY_QI_REGNO_P): Ditto.
(GENERAL_REGNO_P): Use IN_RANGE macro.
(QI_REG_P): Use QI_REGNO_P.
(ANY_QI_REG_P): Use GENERAL_REGNO_P and QI
Dear Mikael,
This looks to be "obvious" and is certainly OK for trunk.
Thanks for the patch.
Paul
On 3 August 2012 16:18, Mikael Morin wrote:
> Hello,
>
> here is the fix for the regression I have introduced with my assumed
> rank bounds patch.
>
> Will test and commit as obvious.
> Mikael
>
>
On Sat, Aug 4, 2012 at 1:38 AM, Cary Coutant wrote:
> diff --git a/gcc/final.c b/gcc/final.c
> index 095d608..d22130f 100644
> --- a/gcc/final.c
> +++ b/gcc/final.c
> @@ -1863,11 +1863,12 @@ final (rtx first, FILE *file, int optimize_p)
>start_to_bb = XCNEWVEC (basic_block, bb_map_size);
>
On 08/03/2012 05:19 PM, Ollie Wild wrote:
On Fri, Aug 3, 2012 at 2:39 AM, Paolo Carlini wrote:
Ok, but, can you also double check and in case fix unordered_map too?
Looks like we have the same issue, right?
Indeed, we do. I'll send a separate patch for the unordered_map problem.
But apparentl
.. note anyway, that only the new testcase was failing, no regressions
on pre existing testcases. Thus, it may well be that only the testcase
had issues, whereas the code changes themselves (likewise those for
unordered_map) are fine as they are, no changes needed elsewhere, neithe
in the specs
On Fri, Aug 3, 2012 at 10:37 PM, Uros Bizjak wrote:
>> Without this, on the new testcase we hit the assert in
>> inline_secondary_memory_needed. The comment before the function states:
>>
>> The macro can't work reliably when one of the CLASSES is class
>> containing registers from multiple u
Hi,
as discussed on the audit trail, I'm changing c_sizeof_or_alignof_type
to unconditionally pedwarn in C++ mode. I have to also tweak an existing
testcase, which was exactly relying on the warning to be suppressed by
-Wno-pointer-arith.
Booted and tested x86_64-linux.
Thanks,
Paolo.
On Sat, 4 Aug 2012, Paolo Carlini wrote:
.. note anyway, that only the new testcase was failing, no regressions on pre
existing testcases.
What I am seeing is a different testcase (with the same name but in a
different directory) failing, because:
typedef std::pair V;
static_assert(std:
Sandra Loosemore writes:
> The existing scheduler bypass information for madd on the 74k uses some
> bits copied from the 24k, and is not quite correct. This patch is based
> on one originally sent to us by MIPS and has been present in our local
> source base for years. I've confirmed that we
Sandra Loosemore writes:
> This patch changes the 24k/74k scheduling descriptions to use the existing
> mips_store_data_bypass_p predicate instead of treating cprestore as a special
> case. OK for mainline?
Nice cleanup, thanks.
> 2012-08-02 Sandra Loosemore
> Maxim Kuvyrkov
>
Sandra Loosemore writes:
> This is another patch that has been present in our local source base for some
> years now. It originally came from MIPS; I've verified that we have legal
> permission to contribute it to the FSF.
>
> The 74k.md parts of this patch depend on the not-yet-reviewed "74k ma
On 03/08/12 19:10, Richard Henderson wrote:
> On 2012-08-02 09:24, Richard Earnshaw wrote:
>> +/* Extz only supports SImode, but we can coerce the operands
>> + into that mode. */
>> +emit_constant_insn (cond,
>> +gen_extzv_t2 (gen_lowpart (mod
Sandra Loosemore writes:
> For all supported MIPS ABIs other than EABI, compiling with
> -fpic/-fPIC/-fpie/-fPIE implicitly requires abicalls support. On Linux
> targets, -mabicalls is the default, but bare-metal targets like
> mips-sde-elf default to -mno-abicalls and naively compiling with -
Sandra Loosemore writes:
> 2012-08-03 Sandra Loosemore
> Catherine Moore
>
> gcc/testsuite/
> * gcc.target/mips/clear-cache-1.c: Test for alternate cache
> flush function names too.
> * gcc.target/mips/clear-cache-1.c: Likewise.
OK, thanks.
Richard
Sandra Loosemore writes:
> 2012-08-03 Catherine Moore
> Sandra Loosemore
>
> gcc/
> * config/mips/xlr.md (ir_xlr_alu_clz): New insn_reservation.
> (ir_xlr_alu): Remove clz.
> * config/mips/mips-cpus.def (xlr): Set PTF_AVOID_BRANCHLIKELY.
OK, thanks.
Richard
On 08/04/2012 03:26 PM, Marc Glisse wrote:
On Sat, 4 Aug 2012, Paolo Carlini wrote:
.. note anyway, that only the new testcase was failing, no
regressions on pre existing testcases.
What I am seeing is a different testcase (with the same name but in a
different directory) failing, because:
On Sat, 4 Aug 2012, Paolo Carlini wrote:
I'm not sure to understand which specific testcase you are discussing, but
for sure we don't want regressions. I agree that we should assume a priori
that the standard is right, but correcting the make_pair should not lead to
failures elsewhere (unless
On 08/04/2012 05:16 PM, Marc Glisse wrote:
On Sat, 4 Aug 2012, Paolo Carlini wrote:
I'm not sure to understand which specific testcase you are
discussing, but for sure we don't want regressions. I agree that we
should assume a priori that the standard is right, but correcting the
make_pair sh
On Sat, 4 Aug 2012, Paolo Carlini wrote:
On 08/04/2012 05:16 PM, Marc Glisse wrote:
On Sat, 4 Aug 2012, Paolo Carlini wrote:
I'm not sure to understand which specific testcase you are discussing, but
for sure we don't want regressions. I agree that we should assume a priori
that the standard
On 08/04/2012 05:27 PM, Marc Glisse wrote:
On Sat, 4 Aug 2012, Paolo Carlini wrote:
On 08/04/2012 05:16 PM, Marc Glisse wrote:
On Sat, 4 Aug 2012, Paolo Carlini wrote:
I'm not sure to understand which specific testcase you are
discussing, but for sure we don't want regressions. I agree that
On Fri, 3 Aug 2012, Ian Lance Taylor wrote:
2012-08-04 Dimitrios Apostolou
* libiberty.h
(XOBDELETE,XOBGROW,XOBGROWVEC,XOBSHRINK,XOBSHRINKVEC): New
type-safe macros for obstack allocation.
(XOBFINISH): Renamed argument to PT since it is a pointer to T.
+/* Ty
On Fri, Aug 3, 2012 at 10:09 AM, Thomas Koenig wrote:
> Hi,
>
>> OK for trunk?
>
>
> I will be on holiday starting tomorrow, so I would like to commit
> this today, if possible (with a mistake in the ChangeLog noted
> by Uros fixed. Thanks!)
>
> If not, somebody please commit this (or another fix
On Fri, Aug 3, 2012 at 7:18 AM, Mikael Morin wrote:
> Hello,
>
> here is the fix for the regression I have introduced with my assumed
> rank bounds patch.
>
> Will test and commit as obvious.
> Mikael
>
>
> 2012-08-02 Mikael Morin
>
> PR fortran/54166
> * trans-array.c (set_loop
On Sat, 4 Aug 2012, venkataramanan.ku...@amd.com wrote:
> Index: gcc/doc/extend.texi
> ===
> +@item btver1
> +AMD family 14h cpu.
CPU...
> @item amdfam15h
> AMD family 15h CPU.
...like you already had here. :-)
> +@item btver2
>
On 08/04/2012 07:48 AM, Richard Sandiford wrote:
Sandra Loosemore writes:
The existing scheduler bypass information for madd on the 74k uses some
bits copied from the 24k, and is not quite correct. This patch is based
on one originally sent to us by MIPS and has been present in our local
sourc
On Wed, 1 Aug 2012, Richard Sandiford wrote:
>> 2012-08-01 Catherine Moore
>> Sandra Loosemore
>>
>> gcc/
>> * config/mips/mips-cpus.def (34kn): New.
>> * config/mips/mips.h (MIPS_ARCH_FLOAT_SPEC): Add 34kn.
>> (BASE_DRIVER_SELF_SPECS): Do not imply -mdsp for the 3
On Mon, 23 Apr 2012, Gerald Pfeifer wrote:
> This fixes two more occurrences per Rainer's earlier observation.
And this should be the last occurrence in our tree.
Applied.
Gerald
Index: htdocs/gcc-4.5/changes.html
===
RCS file: /cv
Applied. (Of course I realized I missed one after declaring
victory just a few minutes ago. ;-)
Gerald
Index: htdocs/projects/cxx0x.html
===
RCS file: /cvs/gcc/wwwdocs/htdocs/projects/cxx0x.html,v
retrieving revision 1.56
diff -u -3
On 08/04/2012 03:01 PM, Gerald Pfeifer wrote:
On Wed, 1 Aug 2012, Richard Sandiford wrote:
2012-08-01 Catherine Moore
Sandra Loosemore
gcc/
* config/mips/mips-cpus.def (34kn): New.
* config/mips/mips.h (MIPS_ARCH_FLOAT_SPEC): Add 34kn.
(BASE_DRIVER_SELF_
Hi,
Marc noticed that this issue, where we are using an available templated
conversion function for a cast to void, apparently can be rather easily
attacked by moving the convert_to_void case a bit earlier in
build_static_cast_1, that is before the
perform_direct_initialization_if_possible ca
On 08/04/2012 08:09 AM, Richard Sandiford wrote:
Sandra Loosemore writes:
For all supported MIPS ABIs other than EABI, compiling with
-fpic/-fPIC/-fpie/-fPIE implicitly requires abicalls support. On Linux
targets, -mabicalls is the default, but bare-metal targets like
mips-sde-elf default to -
On Sat, Aug 4, 2012 at 9:40 AM, Dimitrios Apostolou wrote:
> On Fri, 3 Aug 2012, Ian Lance Taylor wrote:
>
>>> 2012-08-04 Dimitrios Apostolou
>>>
>>> * libiberty.h
>>> (XOBDELETE,XOBGROW,XOBGROWVEC,XOBSHRINK,XOBSHRINKVEC): New
>>> type-safe macros for obstack allocation.
>
32 matches
Mail list logo
