Re: [PATCH] PR70117, ppc long double isinf

On April 8, 2016 5:03:04 AM GMT+02:00, Alan Modra wrote: >On Thu, Apr 07, 2016 at 10:43:31AM -0400, David Edelsohn wrote: >> Yes, I have been following this entertaining thread. > >How to waste lots of time over one bit. Floating point is like that. >:-) > >I see the bug was

Re: openacc reference reductions

IZE is zero. */ #define OMP_CLAUSE_MAP_MAYBE_ZERO_LENGTH_ARRAY_SECTION(NODE) \ TREE_PROTECTED (OMP_CLAUSE_SUBCODE_CHECK (NODE, OMP_CLAUSE_MAP)) +/* Nonzero if this map clause is for an ACC parallel reduction variable. */ +#define OMP_CLAUSE_MAP_IN_REDUCTION(NODE) \ + TREE_PRIVATE (OMP_CLAUSE_SUBCODE_CHECK (NODE, OMP_CLAUSE_MAP)) #define OMP_CLAUSE_PROC_BIND_KIND(NODE) \ (OMP_CLAUSE_SUBCODE_CHECK (NODE, OMP_CLAUSE_PROC_BIND)->omp_clause.subcode.proc_bind_kind) pr70533-20160407-full.diff.gz Description: application/gzip

Re: [PATCH] PR70117, ppc long double isinf

On Thu, Apr 07, 2016 at 10:43:31AM -0400, David Edelsohn wrote: > Yes, I have been following this entertaining thread. How to waste lots of time over one bit. Floating point is like that. :-) I see the bug was opened against 5.3, so OK to commit there after a few days and maybe 4.9 too,

Re: [PATCH] 69517 - [5/6 regression] SEGV on a VLA with excess initializer elements

I've spent a ton of time trying to implement the suggested changes (far too much in large part because of my own mistakes) but I don't think they will work. I'll try to clean up what I have and post it for review. I wanted to respond to this how in case you have some suggestions or concerns

Re: [Patch] Avoid deadlock in guality tests.

On 04/07/2016 02:41 PM, Yvan Roux wrote: > Here is the new patch which makes GDB pass all signals except SIGTRAP. > validated on native armv8l target without regessions. ok for trunk ? LGTM, FWIW. Thanks, Pedro Alves

[PATCH], Re-fix PR 70381 (disable -mfloat128 by default) and add workaround for PR 70589

After applying the fix for PR 70381 to not enable -mfloat128 by default, I discovered the IEEE 128-bit floating point emulation routines in libgcc are no longer being built. The reason for this is the configuration test involved compiling this program: #pragma GCC target ("vsx,float128")

[PATCH] Don't add REG_EQUAL notes in fwprop for paradoxical subregs (PR rtl-optimization/70574)

Hi! The following testcase is miscompiled, because we have: (set (reg:SI ...) (subreg:SI (reg:QI (...)) 0)) instruction and the fwprop attempts to propagate (const_int -1) into the reg:QI use in there, but gives up because costs don't say it is beneficial and adds instead REG_EQUAL (const_int -1)

Re: [C PATCH] PR43651: add warning for duplicate qualifier

New options need documenting in invoke.texi. -- Joseph S. Myers jos...@codesourcery.com

[committed] Fix lto/simd_function_0.c testcase (PR testsuite/70581)

Hi! This fixes this testcase when using assembler that doesn't support AVX512F, which is now required for simd clones (previously it has been just avx2 and thus avx2 effective target has been sufficient). Regtested on x86_64-linux, committed to trunk. 2016-04-07 Jakub Jelinek

Re: [committed] OpenMP declare simd ABI changes on x86_64/i686

On Wed, Apr 06, 2016 at 02:54:31PM +0200, Jakub Jelinek wrote: > Bootstrapped/regtested on x86_64-linux and i686-linux, committed to trunk. > I'll try to coordinate with Intel about 3) as well as the default alignment > if aligned clause is used on declare simd without any explicit alignment.

Re: [PATCH, rs6000] Add support for int versions of vec_adde

On 04/05/16 21:27, David Edelsohn wrote: On Tue, Apr 5, 2016 at 3:36 PM, Bill Seurer wrote: This patch adds support for the signed and unsigned int versions of the vec_adde altivec builtins from the Power Architecture 64-Bit ELF V2 ABI OpenPOWER ABI for Linux

Re: [Patch] Fix PR 60040

On 04/07/2016 01:52 PM, Senthil Kumar Selvaraj wrote: The below patch fixes PR 60040 by not halting with a hard error on a spill failure, if reload knows that it has to run again anyway. Some additional information as to how this situation creates a spill failure would be useful. It's

Re: [Patch AArch64 1/3] Enable CRC by default for armv8.1-a

On Thu, Apr 07, 2016 at 05:23:59PM +0200, Christophe Lyon wrote: > On 6 April 2016 at 12:10, James Greenhalgh wrote: > > > > Hi, > > > > This change reflects binutils support for CRC, where it is always enabled > > for armv8.1-a. > > > > Does v8.1 always enable CRC?

Re: [gomp4] Also test -O0 for OpenACC C, C++ offloading test cases

Hi! On Fri, 1 Apr 2016 10:55:49 +0200, I wrote: > On Thu, 24 Mar 2016 22:31:29 +0100, I wrote: > > On Wed, 23 Mar 2016 19:57:50 +0100, Bernd Schmidt > > wrote: > > > Ok with [...]. > > > > Thanks for the review; committed in r234471: > > > Also test -O0 for OpenACC C,

Re: [PATCH, testsuite/ARM] Skip pr70496.c for cortex-m devices

On 07/04/16 17:13, Thomas Preudhomme wrote: On Wednesday 06 April 2016 12:09:25 Kyrill Tkachov wrote: Hi Thomas, On 06/04/16 12:03, Thomas Preudhomme wrote: Hi, Testcase in gcc.target/arm/pr70496.c uses an .arm directive so assumes the target has an ARM execution state. This patch adds a

Re: [PATCH, testsuite/ARM] Skip pr70496.c for cortex-m devices

On Wednesday 06 April 2016 12:09:25 Kyrill Tkachov wrote: > Hi Thomas, > > On 06/04/16 12:03, Thomas Preudhomme wrote: > > Hi, > > > > Testcase in gcc.target/arm/pr70496.c uses an .arm directive so assumes the > > target has an ARM execution state. This patch adds a dg-skip-if directive > > to

Re: Scan for parallelization of the oacc kernels test-cases in gfortran.dg/goacc

Hi! On Tue, 5 Apr 2016 11:12:44 +0200, Tom de Vries wrote: > On 18/03/16 13:37, Thomas Schwinge wrote: > > On Wed, 9 Mar 2016 10:17:28 +0100, Tom de Vries > > wrote: > >> [Should have cited > >>

Re: [Patch AArch64 1/3] Enable CRC by default for armv8.1-a

On 6 April 2016 at 12:10, James Greenhalgh wrote: > > Hi, > > This change reflects binutils support for CRC, where it is always enabled > for armv8.1-a. > Does v8.1 always enable CRC? If not, then don't you want to change the binutils default instead? Christophe. >

Re: [C++ PATCH] PR 70501, ICE in verify ctor sanity

On 04/06/16 07:49, Jason Merrill wrote: Sure, but that also seems unnecessary; vector rvalues don't have object identity the way class and array rvalues do. I attach 2 patches. 70501-2.patch fixes the ICE by treating VECTOR_TYPEs thesame as PMFs in cxx_eval_bare_aggregate).

Re: Do not give realistic estimates for loop with array accesses

On 30/03/16 14:36, Richard Biener wrote: On Wed, 30 Mar 2016, Jan Hubicka wrote: > > > >You are only changing one place in this file. > >You are right. I am attaching the updated patch which I am re-testing now. > > > >The vectorizer already checks this (albeit indirectly): > > > >

[PATCH][ARM] PR target/70566 Check that condition register is dead in tst-imm -> lsls-imm Thumb2 peepholes

Hi all, In this wrong-code PR we have a Thumb2 peephole transforming: tstr3, #2 bne.L3 beq.L6 into: lslsr3, r3, #30 // LSLS is shorter than TST in Thumb2 bmi.L3 beq.L6 that is, the branch following the extract+compare has its condition properly

Re: [PATCH] PR70117, ppc long double isinf

On Thu, Apr 7, 2016 at 10:17 AM, Alan Modra wrote: > On Thu, Apr 07, 2016 at 11:32:58AM +0200, Richard Biener wrote: >> That's good to know. I think the patch is OK but please seek approval from >> a ppc maintainer as well > > There's only one of those. David? Thread starts

Re: [patch] libstdc++/70503 Ensure std::thread helpers have internal linkage

On 07/04/16 15:42 +0200, Jakub Jelinek wrote: Or perhaps instead change the test, so that instead of testing for undefined symbols it provides some definition of those 2 symbols and makes sure libstdc++.a(thread.o) is linked in too (such as using std::thread::detach or similar somewhere). Then

Re: [PATCH] PR70117, ppc long double isinf

On Thu, Apr 07, 2016 at 11:32:58AM +0200, Richard Biener wrote: > That's good to know. I think the patch is OK but please seek approval from a > ppc maintainer as well There's only one of those. David? Thread starts here https://gcc.gnu.org/ml/gcc-patches/2016-04/msg00213.html -- Alan Modra

Re: [PATCH] PR47040 - Make error message for empty array constructor more helpful/correct

The latter is obvious as this "fixes" neither a regression nor documentation. For the former, see Fortran 95, section 4.5. -- steve On Thu, Apr 07, 2016 at 07:51:14AM +0200, Dominique d'Humières wrote: > Could you please elaborate. > > Dominique > > > Le 7 avr. 2016 à 07:48, Steve Kargl

Re: [lto-partition.c] move assignment to refs_node, last_visited_node outside if-else

On 7 April 2016 at 19:11, Marek Polacek wrote: > On Thu, Apr 07, 2016 at 07:05:09PM +0530, Prathamesh Kulkarni wrote: >> Hi, >> This is a silly patch that moves the following assignments outside >> if-else conditionals, >> refs_node = node; >> last_visited_node++; >> resulting

Re: [lto-partition.c] move assignment to refs_node, last_visited_node outside if-else

On Thu, Apr 07, 2016 at 07:05:09PM +0530, Prathamesh Kulkarni wrote: > Hi, > This is a silly patch that moves the following assignments outside > if-else conditionals, > refs_node = node; > last_visited_node++; > resulting in empty else block. > OK for trunk ? > > Thanks, > Prathamesh > diff

Re: [PATCH][ARM] Add deprecation warning on pre-v4t architecture revisions

Ping. Thanks, Kyrill On 31/03/16 13:28, Kyrill Tkachov wrote: Ping. Thanks, Kyrill On 23/03/16 10:11, Kyrill Tkachov wrote: Ping. Thanks, Kyrill On 07/03/16 15:40, Kyrill Tkachov wrote: Ping. https://gcc.gnu.org/ml/gcc-patches/2016-03/msg00046.html Thanks, Kyrill On 01/03/16 16:17,

Re: [patch] libstdc++/70503 Ensure std::thread helpers have internal linkage

On Thu, Apr 07, 2016 at 02:34:51PM +0100, Jonathan Wakely wrote: > On 07/04/16 15:27 +0200, Tom de Vries wrote: > >FYI, this testcase fails for me as follows: > >... > >/usr/bin/ld: Dwarf Error: found dwarf version '4', this reader only > >handles version 2 and 3 information.^M > >/tmp/ccMBMHB5.o:

Re: [Patch] Avoid deadlock in guality tests.

Hi, On 6 April 2016 at 17:29, Yvan Roux wrote: > On 6 April 2016 at 17:24, Pedro Alves wrote: >> On 04/06/2016 04:13 PM, Yvan Roux wrote: >>> On 6 April 2016 at 17:09, Pedro Alves wrote: On 04/06/2016 03:53 PM, Yvan Roux wrote:

Re: [patch] libstdc++/70503 Ensure std::thread helpers have internal linkage

On 07/04/16 15:27 +0200, Tom de Vries wrote: Hi, FYI, this testcase fails for me as follows: ... /usr/bin/ld: Dwarf Error: found dwarf version '4', this reader only handles version 2 and 3 information.^M /tmp/ccMBMHB5.o: In function `main':^M 70503.cc:(.text.startup.main+0x5): undefined

[lto-partition.c] move assignment to refs_node, last_visited_node outside if-else

Hi, This is a silly patch that moves the following assignments outside if-else conditionals, refs_node = node; last_visited_node++; resulting in empty else block. OK for trunk ? Thanks, Prathamesh diff --git a/gcc/lto/lto-partition.c b/gcc/lto/lto-partition.c index 9eb63c2..dca59af 100644 ---

Re: [patch] libstdc++/70503 Ensure std::thread helpers have internal linkage

Hi, FYI, this testcase fails for me as follows: ... /usr/bin/ld: Dwarf Error: found dwarf version '4', this reader only handles version 2 and 3 information.^M /tmp/ccMBMHB5.o: In function `main':^M 70503.cc:(.text.startup.main+0x5): undefined reference to `execute_native_thread_routine'^M

Re: [PATCH] Avoid needless unsharing during constexpr evaluation (PR c++/70452)

OK. Jason

Re: [PATCH][ARM][RFC] PR target/65578 Fix gcc.dg/torture/stackalign/builtin-apply-4.c for single-precision fpus

Hi Ramana, On 23/03/16 12:09, Ramana Radhakrishnan wrote: On Tue, Feb 9, 2016 at 5:21 PM, Kyrill Tkachov wrote: Hi all, In this wrong-code PR the builtin-apply-4.c test fails with -flto but only when targeting an fpu with only single-precision capabilities. bar

Re: [Patch AArch64 3/3] Fix up for pr70133

Hi all, On 06/04/16 11:10, James Greenhalgh wrote: Hi, Having updated the way we parse and output extension strings, now we just need to wire up the native detection to use these new features. In doing some cleanup and rename I ended up fixing 8-spaces to tabs in about half the file. I've

[Patch] Fix PR 60040

Hi, The below patch fixes PR 60040 by not halting with a hard error on a spill failure, if reload knows that it has to run again anyway. It also fixes two reload related ICEs on trunk (gcc.c-torture/compile/920625-1.c and gcc.dg/tree-ssa/pr70232.c) for the AVR target. I've slighly

Re: [gcc-5] Fix handling of subarrays with update directive

Hi! On Thu, 7 Apr 2016 13:29:58 +0200, Jakub Jelinek wrote: > On Thu, Apr 07, 2016 at 01:28:48PM +0200, Thomas Schwinge wrote: > > On Wed, 23 Mar 2016 14:10:31 +0100, Jakub Jelinek wrote: > > > On Wed, Mar 23, 2016 at 08:05:19AM -0500, James Norris wrote: > >

Re: [gcc-5] Fix handling of subarrays with update directive

On Thu, Apr 07, 2016 at 01:28:48PM +0200, Thomas Schwinge wrote: > Hi Jakub! > > On Wed, 23 Mar 2016 14:10:31 +0100, Jakub Jelinek wrote: > > On Wed, Mar 23, 2016 at 08:05:19AM -0500, James Norris wrote: > > > On 03/23/2016 05:24 AM, Jakub Jelinek wrote: > > > 2016-03-23 James

[gcc-5] Fix handling of subarrays with update directive

Hi Jakub! On Wed, 23 Mar 2016 14:10:31 +0100, Jakub Jelinek wrote: > On Wed, Mar 23, 2016 at 08:05:19AM -0500, James Norris wrote: > > On 03/23/2016 05:24 AM, Jakub Jelinek wrote: > > 2016-03-23 James Norris > > Daichi Fukuoka

Re: [PATCH] Fix PR31096

On Thu, 7 Apr 2016, Hurugalawadi, Naveen wrote: +/* Fold A * 10 == B * 10 into A == B. */ +(for cmp (eq ne) + (simplify + (cmp (mult:c @0 @1) (mult:c @2 @1)) + (if (TYPE_OVERFLOW_UNDEFINED (type) type is the return type of the comparison. The relevant type here is TREE_TYPE (@0). Maybe add

Re: [PATCH] Fix PR31096

Hi, Thanks for the review, views and comments on the issue. >> -1 is an integer constant, so that's still invalid. It is also invalid for >> unsigned. The :s are useless since the output is a single insn. The patch is modified as per your review comments. Currently the following conditions had

Re: openacc reference reductions

On Wed, Apr 06, 2016 at 01:21:30PM -0700, Cesar Philippidis wrote: > That's a good idea. I went ahead and combined this patch with the data > map reduction fix for PR70289 that I posted on Monday, > , because I'm > already scanning for

Re: Add some C++17 items to gcc-6/changes.html

On 06/04/16 10:02 -0400, Jason Merrill wrote: On 04/04/2016 06:22 AM, Jonathan Wakely wrote: I plan to commit this to wwwdocs CVS. Have I missed anything that should be listed? I'd mention that TM requires the -fgnu-tm flag. Done, the attached patch is what I committed. Index:

Re: [PATCH] PR70117, ppc long double isinf

On April 7, 2016 10:03:54 AM GMT+02:00, Alan Modra wrote: >On Wed, Apr 06, 2016 at 06:49:19PM +0930, Alan Modra wrote: >> On Wed, Apr 06, 2016 at 10:46:48AM +0200, Richard Biener wrote: >> > Can you add a testcase or two for the isnormal () case? >> >> Sure. I'll adapt the

Re: [RFA 1/2]: Don't ignore target_header_dir when deciding inhibit_libc

On 17/03/16 16:33, Andre Vieira (lists) wrote: > On 23/10/15 12:31, Bernd Schmidt wrote: >> On 10/12/2015 11:58 AM, Ulrich Weigand wrote: >>> >>> Index: gcc/configure.ac >>> === >>> --- gcc/configure.ac(revision 228530) >>> +++

Re: [PATCH] PR70117, ppc long double isinf

On Wed, Apr 06, 2016 at 06:49:19PM +0930, Alan Modra wrote: > On Wed, Apr 06, 2016 at 10:46:48AM +0200, Richard Biener wrote: > > Can you add a testcase or two for the isnormal () case? > > Sure. I'll adapt the testcase I was using to verify the output, Revised testcase - target fixed, compiled