On Mon, Mar 18, 2013 at 06:18:58PM +0100, Richard Biener wrote:
> I was asking for the ifunc selector to be
> Overridable by ld_preload or a similar mechanism at dynamic load time.
Please don't. Calling an ifunc resolver function in another library
is just asking for trouble with current glibc.
On 16-Feb-13 23:21, Maxim Kuvyrkov wrote:
On 14/02/2013, at 10:18 AM, rbmj wrote:
Here's the updated, (trivial) patch.
Thanks. I'll apply this once 4.8 branches and trunk is back into development
mode.
Since GCC 4.9 has branched now are you still willing to commit (maybe
after the outage
Interesting idea about lazy IFUNC relocation.
David
On Mon, Mar 18, 2013 at 2:02 AM, Richard Biener
wrote:
> On Fri, Mar 15, 2013 at 10:55 PM, Sriraman Tallam wrote:
>> Hi,
>>
>>This patch is meant for google/gcc-4_7 but I want this to be
>> considered for trunk when it opens again. This pa
On Mon, Mar 18, 2013 at 10:18 AM, Richard Biener
wrote:
> "H.J. Lu" wrote:
>>We can pass environment variables to IFUNC selector. Maybe we can
>>enable it for debug build.
Enabling this for just debug builds would not cover my use case.
If the environment variable is used at loader initializ
+cc libc-alpha
On Mon, Mar 18, 2013 at 9:05 AM, Xinliang David Li wrote:
> Interesting idea about lazy IFUNC relocation.
> On Mon, Mar 18, 2013 at 2:02 AM, Richard Biener
> wrote:
>> On Fri, Mar 15, 2013 at 10:55 PM, Sriraman Tallam
>> wrote:
>>>This patch is meant for google/gcc-4_7 bu
On Mon, Mar 18, 2013 at 10:02 AM, Paul Pluzhnikov
wrote:
> +cc libc-alpha
>
> On Mon, Mar 18, 2013 at 9:05 AM, Xinliang David Li wrote:
>> Interesting idea about lazy IFUNC relocation.
>
>> On Mon, Mar 18, 2013 at 2:02 AM, Richard Biener
>> wrote:
>
>>> On Fri, Mar 15, 2013 at 10:55 PM, Sriraman
"H.J. Lu" wrote:
>On Mon, Mar 18, 2013 at 10:02 AM, Paul Pluzhnikov
> wrote:
>> +cc libc-alpha
>>
>> On Mon, Mar 18, 2013 at 9:05 AM, Xinliang David Li
> wrote:
>>> Interesting idea about lazy IFUNC relocation.
>>
>>> On Mon, Mar 18, 2013 at 2:02 AM, Richard Biener
>>> wrote:
>>
On Fri, Mar
On 16 March 2013 13:29, Andi Kleen wrote:
>
> With inline __attribute__((always_inline)) these functions
> get inlined even with -O0.
>
> I hardcoded the attribute in the header for now, assuming
> that all compilers that support libstdc++ have attribute
> always_inline too. If not it would need to
On 18 March 2013 16:28, Jonathan Wakely wrote:
> On 16 March 2013 13:29, Andi Kleen wrote:
>>
>> With inline __attribute__((always_inline)) these functions
>> get inlined even with -O0.
>>
>> I hardcoded the attribute in the header for now, assuming
>> that all compilers that support libstdc++ have
On Mon, Mar 18, 2013 at 04:28:13PM +, Jonathan Wakely wrote:
> On 16 March 2013 13:29, Andi Kleen wrote:
> >
> > With inline __attribute__((always_inline)) these functions
> > get inlined even with -O0.
> >
> > I hardcoded the attribute in the header for now, assuming
> > that all compilers tha
Hi,
ix86_function_arg sets cfun->machine->callee_pass_avx256_p from the
current argument. It clears callee_pass_avx256_p when ix86_function_arg
is called to generate a library call to passs an argument. This patch
adds callee_pass_avx256_p and callee_return_avx256_p to ix86_args to store
the AVX
Ping.
On 03/05/2013 12:15 PM, Meador Inge wrote:
> Hi All,
>
> This patch fixes a minor annoyance that causes backtraces to disappear
> inside of aeabi_ldivmod and aeabi_uldivmod due to the lack of appropriate
> DWARF information. I fixed the problem by adding the necessary cfi_*
> macros in the
> 2013-03-13 Bill Schmidt
> Steven Bosscher
>
> PR rtl-optimization/56605
> * loop-iv.c (implies_p): Handle equal RTXs and subregs.
>
> gcc/testsuite:
>
> 2013-03-13 Bill Schmidt wschm...@linux.vnet.ibm.com>
>
> PR rtl-optimization/56605
> * gcc.target/po
Changelog
2013-03-18 David Holsgrove
* gcc/config/microblaze/microblaze.md (jump):
Account for jumps to SYMBOL_REFs.
Signed-off-by: David Holsgrove
0007-Patch-microblaze-Extend-jump-insn-to-accept-bri-to-S.patch
Description: 0007-Patch-microblaze-Extend-jump-insn-to-accept-bri-to-S.pat
This patch, long on my TODO list, speeds up PTA by removing the
costly hashtable lookup when looking for related fields of
a sub-variable. Instead we now keep a pointer to the first
field. For space-savings I changed head/next to be the variable
info ID instead. This speeds up a fortran testcas
Changelog
2013-03-18 David Holsgrove
* gcc/config/microblaze/microblaze.h: Define SIZE_TYPE
and PTRDIFF_TYPE.
Signed-off-by: David Holsgrove
0006-Patch-microblaze-Add-SIZE_TYPE-and-PTRDIFF_TYPE-to-m.patch
Description: 0006-Patch-microblaze-Add-SIZE_TYPE-and-PTRDIFF_TYPE-to-m.patch
MicroBlaze doesn't have restrictions that would force us to
reload regs via memory. Don't define SECONDARY_MEMORY_NEEDED.
Fixes an ICE when compiling OpenSSL for linux.
Changelog
2013-03-18 Edgar E. Iglesias
* gcc/config/microblaze/microblaze.h: Remove SECONDARY_MEMORY_NEEDED
definition.
Changelog
2013-03-18 David Holsgrove
* gcc/config/microblaze/microblaze.c: Add microblaze_asm_output_mi_thunk
and define TARGET_ASM_OUTPUT_MI_THUNK and TARGET_ASM_CAN_OUTPUT_MI_THUNK
Signed-off-by: David Holsgrove
0003-Patch-microblaze-Add-TARGET_ASM_OUTPUT_MI_THUNK-to-s.patch
Descript
Add sync_compare_and_swapsi and sync_test_and_setsi
implementations for MicroBlaze.
Changelog
2013-03-18 David Holsgrove
* gcc/config/microblaze/sync.md: New file.
* gcc/config/microblaze/microblaze.md: Add UNSPEC_SYNC_CAS,
UNSPEC_SYNC_XCHG and include sync.md.
* gcc/config/microblaze/mi
Add DWARF exception handling support for MicroBlaze.
Changelog
2013-03-18 Edgar E. Iglesias
David Holsgrove
* common/config/microblaze/microblaze-common.c: Remove
TARGET_EXCEPT_UNWIND_INFO definition.
* config/microblaze/microblaze-protos.h: Add microblaze_eh_return prototype
Changelog
2013-03-18 David Holsgrove
* gcc/config/microblaze/microblaze.c (microblaze_expand_prologue):
Add check for flag_stack_usage to handle -fstack-usage support
Signed-off-by: David Holsgrove
0004-Patch-microblaze-Add-fstack-usage-support.patch
Description: 0004-Patch-microblaze-
On Mon, Mar 18, 2013 at 1:00 PM, Jakub Jelinek wrote:
> Hi!
>
> On the attached testcase we ICE, because cselim uses
>if (!is_gimple_reg_type (TREE_TYPE (lhs))
>|| !operand_equal_p (lhs, gimple_assign_lhs (else_assign), 0))
> to guard an optimization, and we have two MEM_REFs where ope
Hi!
On the attached testcase we ICE, because cselim uses
if (!is_gimple_reg_type (TREE_TYPE (lhs))
|| !operand_equal_p (lhs, gimple_assign_lhs (else_assign), 0))
to guard an optimization, and we have two MEM_REFs where operand_equal_p
is true, but they don't have compatible types (the fi
Hi all,
Given code:
#define MAX(a, b) (a > b ? a : b)
void foo (int ilast, float* w, float* w2)
{
int i;
for (i = 0; i < ilast; ++i)
{
w[i] = MAX (0.0f, w2[i]);
}
}
compiled with
-O1 -funsafe-math-optimizations -ftree-vectorize -mfpu=neon -mfloat-abi=hard
on
arm-none-eabi will cause
This is a collection of changes that cleanup PTA.
Bootstrap and regtest pending on x86_64-unknown-linux-gnu.
Richard.
2013-03-18 Richard Biener
* tree-ssa-structalias.c (find): Use gcc_checking_assert.
(unite): Likewise.
(merge_node_constraints): Likewise.
(b
The following is a collection of TLC sitting in my local tree,
mostly resulting in less obscure IL after vectorization.
Bootstrapped on x86_64-unknown-linux-gnu, testing in progress.
Richard.
2013-03-18 Richard Biener
* tree-vect-loop-manip.c (vect_create_cond_for_alias_checks):
This moves VRP after late DOM. This is because VRP has a hard
time dealing with non-copyproped (and not CSEd) IL and conveniently
DOM provides both. I noticed this when working on PR56273 where
we miss quite some VRP opportunities because of this.
I cannot think of a good reason to have the curr
On Mon, Mar 18, 2013 at 12:06 PM, Marc Glisse wrote:
> On Mon, 18 Mar 2013, Richard Biener wrote:
>
>> On Mon, Mar 18, 2013 at 11:27 AM, Marc Glisse
>> wrote:
>>>
>>> On Mon, 18 Mar 2013, Richard Biener wrote:
You mean that the VEC_COND_EXPRs can never be used as an lvalue in
the C
On Mon, 18 Mar 2013, Richard Biener wrote:
On Mon, Mar 18, 2013 at 11:27 AM, Marc Glisse wrote:
On Mon, 18 Mar 2013, Richard Biener wrote:
You mean that the VEC_COND_EXPRs can never be used as an lvalue in
the C++ frontend?
Yes, as I mention in the ChangeLog. Not just the C++ front-end, it
On Mon, Mar 18, 2013 at 11:27 AM, Marc Glisse wrote:
> On Mon, 18 Mar 2013, Richard Biener wrote:
>
>>> 2013-03-17 Marc Glisse
>>>
>>> gcc/
>>> * fold-const.c (fold_cond_expr_with_comparison): Use
>>> build_zero_cst.
>>> VEC_COND_EXPR cannot be lvalues.
>>> (fold_ternary
Hi Guys,
I am applying this small patch to remove a spurious backslash escape
at the end of a line in stormy16.c.
Cheers
Nick
gcc/ChangeLog
2013-03-18 Nick Clifton
* config/stormy16/stormy16.c (xstormy16_expand_prologue): Remove
spurious backslash.
Index: gcc/config/st
Hi Guys
I am applying this patch as obvious. It adds a line missing from the
comment describing the mn10300_get_live_callee_saved_regs function.
Cheers
Nick
gcc/ChangeLog
2013-03-18 Nick Clifton
* config/mn10300/mn10300.c (mn10300_get_live_callee_saved_regs):
Add missi
This fixes an issue that shows up in PR56210.
Bootstrapped and tested on x86_64-unknown-linux-gnu, applied to trunk.
Richard.
2013-03-18 Richard Biener
PR tree-optimization/56210
* tree-ssa-structalias.c (find_func_aliases_for_builtin_call):
Handle string / character
This fixes PR56483 by properly testing for boolean values
during expansion of conditionals.
Bootstrapped and tested on x86_64-unknown-linux-gnu, applied to trunk.
Richard.
2013-03-18 Richard Biener
PR middle-end/56483
* cfgexpand.c (expand_gimple_cond): Inline gimple_cond_si
On Mon, 18 Mar 2013, Richard Biener wrote:
2013-03-17 Marc Glisse
gcc/
* fold-const.c (fold_cond_expr_with_comparison): Use build_zero_cst.
VEC_COND_EXPR cannot be lvalues.
(fold_ternary_loc) : Call
fold_cond_expr_with_comparison.
gcc/cp/
* call.c (build_cond
On Wed, Feb 27, 2013 at 4:49 PM, Richard Biener wrote:
>
> This splits data reference group analysis away from data dependence
> checking and splits the latter into loop and a BB vectorization
> functions. This allows us to perform the now no longer quadratic
> but O(n * log (n)) data reference g
Hi,
On 03/17/2013 06:45 PM, Jonathan Wakely wrote:
On 17 March 2013 17:14, Paolo Carlini wrote:
I guess we could at least work around the problem by going back to
_M_get_Tp_allocator().construct in _M_create_node (or, better, the
allocator_traits<>::construct equivalent, per the recent fix for
On Sun, Mar 17, 2013 at 4:46 PM, Marc Glisse wrote:
> Hello,
>
> this patch adds a bit of folding to VEC_COND_EXPR so it is possible to
> generate ABS_EXPR and MAX_EXPR for vectors without relying on the
> vectorizer. I would have preferred to merge the COND_EXPR and VEC_COND_EXPR
> cases, but the
On Fri, Mar 15, 2013 at 10:55 PM, Sriraman Tallam wrote:
> Hi,
>
>This patch is meant for google/gcc-4_7 but I want this to be
> considered for trunk when it opens again. This patch makes it easy to
> test for code coverage of multiversioned functions. Here is a
> motivating example:
>
> __att
On Wed, Jan 16, 2013 at 4:57 PM, Richard Biener wrote:
>
> This fixes PR3713 by properly propagating ->has_constants in SCCVN.
> With that we are able to simplify (unsigned) Bar & 1 properly.
> Only copyprop later turns the call into a direct one though,
> so I'm testing the important fact - that
On Mon, Feb 11, 2013 at 4:45 PM, Richard Biener wrote:
>
> This fixes the compile-time sink in find_uses_to_rename, that we
> scan the whole function when nothing is to do (well, appearantly).
>
> -O3 bootstrap and regtest on x86_64-unknown-linux-gnu in progress,
> scheduled for stage1.
Committed
On Fri, 8 Feb 2013, Richard Biener wrote:
> On Fri, 1 Feb 2013, Richard Biener wrote:
>
> > On Fri, 1 Feb 2013, Jakub Jelinek wrote:
> >
> > > On Fri, Feb 01, 2013 at 10:00:00AM +0100, Richard Biener wrote:
> > > >
> > > > This reduces compile-time of the testcase in PR56113 (with n = 4)
>
On Fri, 15 Mar 2013, Richard Biener wrote:
> On Thu, 14 Mar 2013, Richard Biener wrote:
>
> >
> > This extracts pieces from the already posted patch series that are
> > most worthwhile and applicable for backporting to both 4.8 and 4.7.
> > It also re-implements the limiting of the maximum numbe
43 matches
Mail list logo
