[gomp4, committed] Add goacc/kernels-acc-on-device.c

2015-10-10 Thread Tom de Vries
Hi, this patch adds a new kernels region test-case. Committed to gomp-4_0-branch as trivial. Thanks, - Tom Add goacc/kernels-acc-on-device.c 2015-10-10 Tom de Vries * c-c++-common/goacc/kernels-acc-on-device.c: New test. --- .../c-c++-common/goacc/kernels-acc-on-device.c | 39 +++

Re: [PING][PR67476] Add param parloops-schedule

2015-10-10 Thread Tom de Vries
On 06/10/15 11:30, Bernd Schmidt wrote: On 10/04/2015 05:36 PM, Tom de Vries wrote: I'll try to give a bit of context: The omp-expand machinery is used in two contexts: 1. when omp annotations are added to the source. In that case, omp-expand is used in non-ssa gimple context. 2. when parlo

[PATCH, 1/5] Handle simple latch in expand_omp_for_generic

2015-10-10 Thread Tom de Vries
On 10/10/15 13:06, Tom de Vries wrote: OK, I'll repost with the patch split up, as follows: 1Handle simple latch in expand_omp_for_generic 2Add missing phis in expand_omp_for_generic 3Handle original loop tree in expand_omp_for_generic 4Support DEFPARAMENU

[PATCH, 2/5] Add missing phis in expand_omp_for_generic

2015-10-10 Thread Tom de Vries
On 10/10/15 13:06, Tom de Vries wrote: OK, I'll repost with the patch split up, as follows: 1Handle simple latch in expand_omp_for_generic 2Add missing phis in expand_omp_for_generic 3Handle original loop tree in expand_omp_for_generic 4Support DEFPARAMENU

[PATCH, 3/5] Handle original loop tree in expand_omp_for_generic

2015-10-10 Thread Tom de Vries
On 10/10/15 13:06, Tom de Vries wrote: OK, I'll repost with the patch split up, as follows: 1Handle simple latch in expand_omp_for_generic 2Add missing phis in expand_omp_for_generic 3Handle original loop tree in expand_omp_for_generic 4Support DEFPARAMENU

[PATCH, 4/5] Support DEFPARAMENUM in params.def

2015-10-10 Thread Tom de Vries
On 10/10/15 13:06, Tom de Vries wrote: OK, I'll repost with the patch split up, as follows: 1Handle simple latch in expand_omp_for_generic 2Add missing phis in expand_omp_for_generic 3Handle original loop tree in expand_omp_for_generic 4Support DEFPARAMENU

[PATCH, 5/5] Add param parloops-schedule

2015-10-10 Thread Tom de Vries
On 10/10/15 13:06, Tom de Vries wrote: OK, I'll repost with the patch split up, as follows: 1Handle simple latch in expand_omp_for_generic 2Add missing phis in expand_omp_for_generic 3Handle original loop tree in expand_omp_for_generic 4Support DEFPARAMENU

Re: [PR67891] drop is_gimple_reg test from set_parm_rtl

2015-10-10 Thread Alexandre Oliva
On Oct 9, 2015, Richard Biener wrote: > Ok. Note that I think emit_block_move shouldn't mess with the addressable > flag. I have successfully tested a patch that stops it from doing so, reverting https://gcc.gnu.org/bugzilla/show_bug.cgi?id=49429#c11 but according to bugs 49429 and 49454, it

Re: [PATCH] PR66870 PowerPC64 Enable gold linker with split stack

2015-10-10 Thread David Edelsohn
On Fri, Oct 9, 2015 at 3:51 PM, Lynn A. Boger wrote: > Here's a new one. > > Tried all the variations, verified that split stack is still enabled and > uses gold linker for 64 bit targets > when using a 64 bit default compiler, and does not give the split stack > error for 32 bit default > compile

[PATCH] remove an unneeded as_a ()

2015-10-10 Thread tbsaunde+gcc
From: Trevor Saunders Hi, obvious clean up, bootstrapped + regtested x86_64-linux-gnu, committed. Trev gcc/ChangeLog: 2015-10-10 Trevor Saunders * function.c (stack_protect_epilogue): Remove as_a () call that isn't needed. --- gcc/ChangeLog | 5 + gcc/function.c | 6

[gomp4] remove bogus tests

2015-10-10 Thread Nathan Sidwell
I've committed this to gomp4 branch. Both these tests are trying an 'acc loop' outside of an offload region. That's an error. Of course, a diagnostic should be emitted, but the remainder of the tests are apparently checking valid constructs. So I chose to simply remove the erroneous case

Re: [PATCH] Fix PR67783, quadraticness in IPA inline analysis

2015-10-10 Thread Jan Hubicka
> > Ah, sorry. I wrote you a reply but apparently did not send. Yes, the patch > looks > resonable - it is a heuristics after all. Lets watch if the change make any > difference > on polyhedron and other benchmarks. It seems there was regression on fatigue/fatigue2 http://gcc.opensuse.org/c+

Re: Do not compare types in operands_equal_p if OEP_ADDRESS_OF is set

2015-10-10 Thread Jan Hubicka
Hi, this is final version of patch I commited. I applied the changes suggested and droped the redundant OEP_CONSTANT_ADDRESS_OF & updated documentation that the flag is supposed to be additive to OEP_ADDRESS_OF I also disabled the sanity check about NOP_EXPR inside ADDR_EXPR until we resolve the

Re: Fix more of C/fortran canonical type issues

2015-10-10 Thread Jan Hubicka
Hi, this is a variant of patch I commited (adding the suggested predicate) Honza * tree.c (type_with_interoperable_signedness): New. (gimple_canonical_types_compatible_p): Use it. * tree.h (type_with_interoperable_signedness): Declare * lto.c (hash_canonical_type)

Re: Use OEP_ADDRESS_OF in emit-rtl.c

2015-10-10 Thread Jan Hubicka
> > Hmm, I see. I wonder how complex the expressions are and if we can't simply > > compare AO properties of MEM_REF at toplevel and then dispatch to > > operand_equal_p (..., OEP_ADDRESS_OF) > > which would make more sense to me. > > I can't see how we can do that comparison conservatively witho

Re: [PATCH] PR66870 PowerPC64 Enable gold linker with split stack

2015-10-10 Thread Matthias Klose
On 10.10.2015 16:00, David Edelsohn wrote: On Fri, Oct 9, 2015 at 3:51 PM, Lynn A. Boger wrote: Here's a new one. Tried all the variations, verified that split stack is still enabled and uses gold linker for 64 bit targets when using a 64 bit default compiler, and does not give the split stack

Re: [PATCH] PR66870 PowerPC64 Enable gold linker with split stack

2015-10-10 Thread Andreas Schwab
"Lynn A. Boger" writes: > Index: gcc/config/rs6000/sysv4.h > === > --- gcc/config/rs6000/sysv4.h (revision 228653) > +++ gcc/config/rs6000/sysv4.h (working copy) > @@ -940,13 +940,15 @@ ncrtn.o%s" > #undef TARGET_ASAN_SHADOW_OFFSET

[ping] Re: [patch] fix gotools cross build

2015-10-10 Thread Matthias Klose
still needed, both on the trunk and the branch. On 06.05.2015 14:34, Matthias Klose wrote: Yes, it's documented that there is still some work to do for cross builds, however a cross build for gotools currently fails. The toplevel make always passes the GOC variable in the environment, overwriti

[PATCH 0/6] more ifdef removal

2015-10-10 Thread tbsaunde+gcc
From: Trevor Saunders Hi, more of the same sort of ifdef removal adding defaults. patches individually bootstrapped + regtested on x86_64-linux-gnu, committing as preapproved. Trev Trevor Saunders (6): always define INITIAL_FRAME_ADDRESS_RTX always define SETUP_FRAME_ADDRESSES always d

[PATCH 2/6] always define SETUP_FRAME_ADDRESSES

2015-10-10 Thread tbsaunde+gcc
From: Trevor Saunders gcc/ChangeLog: 2015-10-10 Trevor Saunders * defaults.h (SETUP_FRAME_ADDRESSES): New default definition. * builtins.c (expand_builtin_return_addr): Adjust. * doc/tm.texi: Likewise. * doc/tm.texi.in: Likewise. * except.c (expand_bui

[PATCH 4/6] always define FRAME_ADDR_RTX

2015-10-10 Thread tbsaunde+gcc
From: Trevor Saunders gcc/ChangeLog: 2015-10-10 Trevor Saunders * defaults.h (FRAME_ADDR_RTX): New default definition. * builtins.c (expand_builtin_return_addr): Adjust. --- gcc/builtins.c | 4 gcc/defaults.h | 4 2 files changed, 4 insertions(+), 4 deletions(-) d

[PATCH 1/6] always define INITIAL_FRAME_ADDRESS_RTX

2015-10-10 Thread tbsaunde+gcc
From: Trevor Saunders gcc/ChangeLog: 2015-10-10 Trevor Saunders * builtins.c (expand_builtin_return_addr): Adjust. * defaults.h (INITIAL_FRAME_ADDRESS_RTX): New default definition. --- gcc/builtins.c | 41 +++-- gcc/defaults.h | 4 2

[PATCH 3/6] always define DYNAMIC_CHAIN_ADDRESS

2015-10-10 Thread tbsaunde+gcc
From: Trevor Saunders gcc/ChangeLog: 2015-10-10 Trevor Saunders * defaults.h (DYNAMIC_CHAIN_ADDRESS): New default definition. * builtins.c (expand_builtin_return_addr): Adjust. --- gcc/builtins.c | 2 -- gcc/defaults.h | 4 2 files changed, 4 insertions(+), 2 deletions(

[PATCH 5/6] remove unneeded #if for HARD_FRAME_POINTER_IS_ARG_POINTER

2015-10-10 Thread tbsaunde+gcc
From: Trevor Saunders gcc/ChangeLog: 2015-10-10 Trevor Saunders * builtins.c (expand_builtin_setjmp_receiver): Don't use #if to check HARD_FRAME_POINTER_IS_ARG_POINTER. --- gcc/builtins.c | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) diff --git a/gcc/builtins.c b

[PATCH 6/6] always define REVERSE_CONDITION

2015-10-10 Thread tbsaunde+gcc
From: Trevor Saunders gcc/ChangeLog: 2015-10-10 Trevor Saunders * defaults.h (REVERSE_CONDITION): New default definition. * jump.c (reversed_comparison_code_parts): Adjust. --- gcc/defaults.h | 4 gcc/jump.c | 8 +--- 2 files changed, 5 insertions(+), 7 deletion

Re: [3/7] Optimize ZEXT_EXPR with tree-vrp

2015-10-10 Thread Kugan
On 09/10/15 21:29, Richard Biener wrote: > + unsigned int prec = tree_to_uhwi (vr1.min); > > this should use unsigned HOST_WIDE_INT > > + wide_int sign_bit = wi::shwi (1ULL << (prec - 1), > + TYPE_PRECISION (TREE_TYPE (vr0.min))); > > use wi::one (TY