Re: [PATCH] libffi: define FFI_SIZEOF_JAVA_RAW for aarch64 ILP32

James Greenhalgh writes: > On Tue, Mar 15, 2016 at 03:18:43PM +0100, Andreas Schwab wrote: >> Like x32, aarch64 ILP32 needs to define FFI_SIZEOF_JAVA_RAW. This fixes >> the java interpreter. > > Should this go through upstream libffi first? I can't remember the right >

Re: Please include ada-hurd.diff upstream (try2)

On Wed, 2016-03-16 at 11:53 +0100, Eric Botcazou wrote: > > > > Sorry for not supplying a ChangeLog yet, I've been busy lately with other > > things. Since there is another new file s-osinte-gnu.adb in the latest > > patch, that might be added too. > AFAICS there is only s-osinte-gnu.ads in the

[PATCH, PR70185] Only finalize dot files that have been initialized

Hi, Atm, using fdump-tree-all-graph produces invalid dot files: ... $ rm *.c.* ; gcc test.c -O2 -S -fdump-tree-all-graph $ for f in *.dot; do dot -Tpdf $f -o dot.pdf; done Warning: test.c.006t.omplower.dot: syntax error in line 1 near '}' Warning: test.c.007t.lower.dot: syntax error in line 1

Re: Please include ada-hurd.diff upstream (try2)

> Sorry for not supplying a ChangeLog yet, I've been busy lately with other > things. Since there is another new file s-osinte-gnu.adb in the latest > patch, that might be added too. AFAICS there is only s-osinte-gnu.ads in the posted patch. -- Eric Botcazou

Re: [PATCH] boehm-gc: add supprt for aarch64 ILP32

On Tue, Mar 15, 2016 at 03:46:00PM +0100, Andreas Schwab wrote: > * include/private/gcconfig.h [AARCH64] (ALIGNMENT, CPP_WORDSZ): > Define for __ILP32__. OK. Thanks, James > --- > boehm-gc/include/private/gcconfig.h | 9 +++-- > 1 file changed, 7 insertions(+), 2 deletions(-) >

Re: [PATCH] libffi: define FFI_SIZEOF_JAVA_RAW for aarch64 ILP32

On Tue, Mar 15, 2016 at 03:18:43PM +0100, Andreas Schwab wrote: > Like x32, aarch64 ILP32 needs to define FFI_SIZEOF_JAVA_RAW. This fixes > the java interpreter. Should this go through upstream libffi first? I can't remember the right order for these. Anyway, this is OK in whatever order is

Re: Please include ada-hurd.diff upstream (try2)

On Wed, 2016-03-16 at 11:08 +0100, Eric Botcazou wrote: > > > > Maybe the time has come now for a proper inclusion of Ada/Gnat support in > > gcc-* for GNU/Hurd. (Yes, I commit to maintain Ada in gcc for GNU/Hurd in > > the foreseeable future, according to my abilities). > > > > Attached is an

Re: [PATCH, PR68715] Add missing single_pred_p test in scop_detection::merge_sese

On 16/03/16 09:53, Richard Biener wrote: Hmm, it looks like for all what this function does this effectively pessimizes scop merging and it would be easier to split 'exit' in case its destination is unsuitable (not trivially empty). Agreed. The /* For now we just want to bail out when

Re: Please include ada-hurd.diff upstream (try2)

> Maybe the time has come now for a proper inclusion of Ada/Gnat support in > gcc-* for GNU/Hurd. (Yes, I commit to maintain Ada in gcc for GNU/Hurd in > the foreseeable future, according to my abilities). > > Attached is an updated patch, now living e.g. in GNU/Debian versions of > gcc-5 and

Re: [PATCH PR69042/01]Add IV candidate for use with constant offset stripped in base.

On Wed, Mar 16, 2016 at 10:48 AM, Bin Cheng wrote: > Hi, > When I tried to decrease # of IV candidates, I removed code that adds IV > candidates for use with constant offset stripped in use->base. This is kind > of too aggressive and triggers PR69042. So here is a patch

[PATCH PR69489/01]Improve tree ifcvt by storing/tracking DR against its innermost loop bahavior if possible

Hi, One issue revealed in tree ifcvt is the pass stores/tracks DRs against its memory references in IR. This causes difficulty in identifying same memory references appearing in different forms. Given below example: void foo (int a[], int b[]) { int i; for (i = 0; i < 100; i++) {

[PATCH PR69042/01]Add IV candidate for use with constant offset stripped in base.

Hi, When I tried to decrease # of IV candidates, I removed code that adds IV candidates for use with constant offset stripped in use->base. This is kind of too aggressive and triggers PR69042. So here is a patch adding back the missing candidates. Honestly, this patch doesn't truly fix the

Re: PING^1: [PATCH] Add TYPE_EMPTY_RECORD for C++ empty class

On March 16, 2016 3:17:20 AM GMT+01:00, "H.J. Lu" wrote: >> Where is the current definition of empty types you're proposing for >use in >> GCC? Is the behavior of this case clear from that definition? > >https://gcc.gnu.org/ml/gcc/2016-03/msg00071.html > >Jason's patch

Re: [PATCH GCC][OBVIOUS]Fix dump info by reporting malformed loop nest.

On Wed, Mar 16, 2016 at 10:28 AM, Bin Cheng wrote: > Hi, > This is an obvious change to dump message in vect_analyze_loop_2. > Apparently, the wrong message is copies/pasted from another place, the code > has nothing to do with function calls or data references. We should

Re: [PATCH, PR70187] Safely use nodes[0] in possible_polymorphic_call_targets

On Wed, Mar 16, 2016 at 9:13 AM, Tom de Vries wrote: > Hi, > > this patch fixes lto PR70187, a 6 regression. > > We run into an ICE in in possible_polymorphic_call_targets when accessing > nodes[0]->decl because nodes == vNULL: > ... > if

Re: [PATCH] Fix 70199

On Tue, Mar 15, 2016 at 8:42 PM, Richard Henderson wrote: > On 03/15/2016 07:13 AM, Richard Biener wrote: >> >> On Tue, Mar 15, 2016 at 4:44 AM, Richard Henderson wrote: >>> >>> The problem here is that >>> >>>void* labels[] = { >>> &, &, & >>>};

[PATCH GCC][OBVIOUS]Fix dump info by reporting malformed loop nest.

Hi, This is an obvious change to dump message in vect_analyze_loop_2. Apparently, the wrong message is copies/pasted from another place, the code has nothing to do with function calls or data references. We should report that loop cannot be vectorized because of malformed loop nest. Build

Re: [PATCH, PR68715] Add missing single_pred_p test in scop_detection::merge_sese

On Wed, 16 Mar 2016, Tom de Vries wrote: > Hi, > > this patch fixes graphite PR68715, a 6 regression. > > In scop_detection::merge_sese, we check if the exit bb of the merged sese > region is dominated by the entry bb: > ... > if (... > || !dominated_by_p (CDI_DOMINATORS, get_exit_bb

Re: [PATCH, PR68809] Fix same_close_phi_node

On Wed, 16 Mar 2016, Tom de Vries wrote: > Hi, > > this patch fixes graphite PR68809, a 6 regression. > > The problem we run into is that we return true in same_close_phi_node for two > single-argument phis, one with an int 0 argument, and one with a char 0 > argument: > ... > _1 = PHI

[PING] [PATCH, libstdc++] Add missing free-standing headers to install rule

Ping... Hi, when the free-standing libstdc++-headers are installed, the C++ header file does not always compile, because it includes and this includes under certain conditions (__cplusplus >= 201103L && ATOMIC_INT_LOCK_FREE > 1) the header file but that fails to compile because it needs

[PATCH, PR70187] Safely use nodes[0] in possible_polymorphic_call_targets

Hi, this patch fixes lto PR70187, a 6 regression. We run into an ICE in in possible_polymorphic_call_targets when accessing nodes[0]->decl because nodes == vNULL: ... if (!outer_type->all_derivations_known) { if (!speculative && final_warning_records && TREE_CODE

[PATCH, PR68715] Add missing single_pred_p test in scop_detection::merge_sese

Hi, this patch fixes graphite PR68715, a 6 regression. In scop_detection::merge_sese, we check if the exit bb of the merged sese region is dominated by the entry bb: ... if (... || !dominated_by_p (CDI_DOMINATORS, get_exit_bb (combined), get_entry_bb

[PATCH, PR68809] Fix same_close_phi_node

Hi, this patch fixes graphite PR68809, a 6 regression. The problem we run into is that we return true in same_close_phi_node for two single-argument phis, one with an int 0 argument, and one with a char 0 argument: ... _1 = PHI <(int)0> _2 = PHI <(char)0> ... The result types of the two