[PATCH][match-and-simplify] Fix last commit

Committed. Richard. 2014-09-10 Richard Biener * match-conversions.pd: Fix unsigned type used for shortened multiplication. Index: gcc/match-conversions.pd === --- gcc/match-conversions.pd(revision 215057) ++

RE: [Patch ARM v2] Fix PR target/56846

> -Original Message- > From: Jonathan Wakely [mailto:jwak...@redhat.com] > Sent: Tuesday, September 09, 2014 6:40 PM > To: Tony Wang > Cc: Ramana Radhakrishnan; gcc-patches; libstd...@gcc.gnu.org > Subject: Re: [Patch ARM v2] Fix PR target/56846 > > On 09/09/14 17:47 +0800, Tony Wang wrote

Re: [PATCH] -fsanitize=nonnull-attribute and -fsanitize=returns-nonnull-attribute support

On Tue, 9 Sep 2014, Jakub Jelinek wrote: > Hi! > > On Fri, Jun 27, 2014 at 09:13:07AM +0200, Jakub Jelinek wrote: > > The patch adds two new (trivial handlers) to libubsan, as it is maintained > > in llvm's compiler-rt, will talk to them if they are interested in those > > and what exact wording

Re: [PATCH] -fsanitize=nonnull-attribute and -fsanitize=returns-nonnull-attribute support

On Tue, 9 Sep 2014, Jakub Jelinek wrote: > On Fri, Jun 27, 2014 at 09:13:07AM +0200, Jakub Jelinek wrote: > > The patch adds two new (trivial handlers) to libubsan, as it is maintained > > in llvm's compiler-rt, will talk to them if they are interested in those > > and what exact wording and form

[PATCHv2] Vimrc config with GNU formatting

Hi all, This is a second version of patch which adds a Vim config (.local.vimrc) to root folder to allow automatic setup of GNU formatting for C/C++/Java/Lex GCC files. I've updated the code with comments from Richard and Bernhard (which fixed formatting of lonely closing bracket). The patc

Re: [PATCH ARM]memset inlining patch for arm

On Sat, Sep 6, 2014 at 3:33 AM, Evandro Menezes wrote: > Bin, > > This is perhaps a plus for Aarch64 as well. Is there any plan to add a > 64-bit version of this patch or should a bug be open for this? Hi Evandro, Yes, AARCH64 may want this too. I think Ramana/Marcus should have the answer/plan

Re: [PATCH ARM]memset inlining patch for arm

On Tue, Sep 9, 2014 at 6:49 PM, Richard Earnshaw wrote: > On 04/09/14 07:08, Bin Cheng wrote: >> @@ -1872,7 +1892,9 @@ const struct tune_params arm_cortex_a53_tune = >>{true, true}, /* Prefer non short >> circuit. */ >>&arm_default_vec_cost,

Re: [PATCHv2] Vimrc config with GNU formatting

> I personally prefer 2) Sorry, I meant 1) of course that is enable vimrc config by default. -Y

Re: [PATCH][ARM][1/7] Convert FP mnemonics to UAL | mov patterns

Hi, On 9 September 2014 13:02, Ramana Radhakrishnan wrote: > On Tue, Aug 19, 2014 at 4:22 PM, Kyrill Tkachov > wrote: >> Hi all, >> >> In this patch the move patterns are updated. >> For the fconst case where the constant is encoded in a decimal >> representation before going into the immediat

[PATCH] S/390: Implement SD<->TD conversions

Hi, on S/390 we have DFP conversions for SD<->DD and DD<->TD but miss SD<->TD. The attached patch expands the missing patterns to SD->DD->TD and TD->DD->SD to avoid library calls being emitted. No regressions on s390 and s390x. Bye, -Andreas- 2014-09-10 Andreas Krebbel * config/s39

[PATCH] S/390: PR62662 Fix r14 restore optimization

Hi, this fixes a problem with the prologue optimization in machine dependent reorg. For more details please see PR62662. No regressions on s390 and s390x. -Andreas- 2014-09-10 Andreas Krebbel PR target/62662 * config/s390/s390.c (s390_emit_epilogue): When doing the return

Re: [PATCH ARM]memset inlining patch for arm

On Wed, Sep 10, 2014 at 9:16 AM, Bin.Cheng wrote: > On Tue, Sep 9, 2014 at 6:49 PM, Richard Earnshaw wrote: >> On 04/09/14 07:08, Bin Cheng wrote: >>> @@ -1872,7 +1892,9 @@ const struct tune_params arm_cortex_a53_tune = >>>{true, true}, /* Prefer non short

Re: [PATCH][ARM][1/7] Convert FP mnemonics to UAL | mov patterns

On 10/09/14 09:40, Christophe Lyon wrote: Hi, Hi Christophe, On 9 September 2014 13:02, Ramana Radhakrishnan wrote: On Tue, Aug 19, 2014 at 4:22 PM, Kyrill Tkachov wrote: Hi all, In this patch the move patterns are updated. For the fconst case where the constant is encoded in a decimal

DBL_DENORM_MIN should never be 0

Hello, according to the C++ standard, numeric_limits::denorm_min should return min (not 0) when there are no denormals. Tested with bootstrap+testsuite on x86_64-linux-gnu. I also tested a basic make all-gcc for vax (only target without denormals apparently) and the macro did change as expec

[C++, OpenMP] Crash when OpenMP target's array section handling is used with templates

Hi! Are the following issues known? template void f(T A, T B) { extern int *v; T a = 2; T b = 4; #pragma omp target map(to: v[a:b]) v[a] = 0; #pragma omp target map(to: v[A:B]) v[a] = 0; } void g() { f(1, 5);

Re: [C++ Patch] Add default arguments to cp_parser_assignment_expression and cp_parser_constant_expression

Hi, On 08/19/2014 02:18 PM, Paolo Carlini wrote: Hi, two more functions. Tested x86_64-linux. I'm pinging this. https://gcc.gnu.org/ml/gcc-patches/2014-08/msg01853.html Frankly, the cp_parser_constant_expression bits seem rather straightforward to me. As regards the cp_parser_assignment

Re: [PATCH ARM]memset inlining patch for arm

On 10/09/14 09:16, Bin.Cheng wrote: On Tue, Sep 9, 2014 at 6:49 PM, Richard Earnshaw wrote: On 04/09/14 07:08, Bin Cheng wrote: @@ -1872,7 +1892,9 @@ const struct tune_params arm_cortex_a53_tune = {true, true}, /* Prefer non short circuit. */ &arm

Re: [PATCH][ARM][1/7] Convert FP mnemonics to UAL | mov patterns

On Wed, Sep 10, 2014 at 4:57 PM, Kyrill Tkachov wrote: > > On 10/09/14 09:40, Christophe Lyon wrote: >> >> Hi, > > Hi Christophe, > >> >> On 9 September 2014 13:02, Ramana Radhakrishnan >> wrote: >>> >>> On Tue, Aug 19, 2014 at 4:22 PM, Kyrill Tkachov >>> wrote: Hi all, In th

Re: [C++, OpenMP] Crash when OpenMP target's array section handling is used with templates

On Wed, Sep 10, 2014 at 12:12:03PM +0200, Thomas Schwinge wrote: > Are the following issues known? No, please file a PR. Supposedly either we should call cp_omp_mappable_type only for non-type-dependent expressions, or not at all if processing_template_decl and only do that once finish_omp_clause

[OpenMP] Spurious »set but not used« warnings when actually used in OpenMP target's array section's lower-bound and length

Hi! This is similar to what has previously been addressed in , : int f(int A, int B) { int r = 0; extern int *v; int a = 2; in

RE: [PATCH][MIPS] Fix ICE in bitmap routines with LRA and inline assembly language

> The patch is ok to commit. I think I made a typo as other analogous > code contains '|='. > > Thanks for fixing this, Robert. The LRA part of this is committed as r215119. Eric: Is the test OK to commit for MIPS? I suggest removing the dg-skip-if as the test is for successful compilation, not

Re: [PATCH 4/5] recog_memoized works on an rtx_insn *

Tested on m68k-suse-linux, installed as obvious. Andreas. * coretypes.h (struct _dont_use_rtx_insn_here_, rtx_insn) [USED_FOR_TARGET]: Define. diff --git a/gcc/coretypes.h b/gcc/coretypes.h index 9951f10..c850ff4 100644 --- a/gcc/coretypes.h +++ b/gcc/coretypes.h @@ -199,11 +199,

[PATCH] doc/install.texi: add documentation for --disable-libsanitizer

This patch adds an item for --disable-libsanitizer to the configuration documentation. The option is especially useful during bisection, because it reduces build time a lot. Ok for trunk and 4.9 branch? Thanks. 2014-09-10 Markus Trippelsdorf * doc/install.texi (Options specification

[PATCH][match-and-simplify] More flexible 'for', polish match-builtins.pd

This completes match-builtins.pd by handling all builtin variants in all present patterns. That requires a little more flexibility in the 'for' handling to support for example /* Optimize sqrt(expN(x)) = expN(x*0.5). */ (for expfn (BUILT_IN_EXP10F BUILT_IN_EXP10 BUILT_IN_EXP10L B

[wwwdocs] Mention Cilk Plus support

Hi! Following change mentions that now all Cilk Plus features added to GCC Index: htdocs/index.html === RCS file: /cvs/gcc/wwwdocs/htdocs/index.html,v retrieving revision 1.936 diff -p -r1.936 index.html *** htdocs/index.html   14 Au

[PING][PATCH] Asan optimization for aligned accesses.

On 09/02/2014 07:09 PM, Marat Zakirov wrote: Hi all! Here's a simple optimization patch for Asan. It stores alignment information into ASAN_CHECK which is then extracted by sanopt to reduce number of "and 0x7" instructions for sufficiently aligned accesses. I checked it on linux kernel by com

[PATCH i386 AVX512] [34/n] AVX-512. Extend vpermvar insn patterns.

Hello, patch in the bottom extends PERMVAR based patterns. Hello, Bootstrapped. AVX-512* tests on top of patch-set all pass under simulator. Is it ok for trunk? gcc/ * config/i386/sse.md (define_mode_attr avx2_avx512): Rename from avx2_avx512bw. (define_mode_iterator VI4

Re: [PATCH, c++ testsuite]: Improve g++.dg/abi/mangle33.C scan-asm pattern

[ originally sent off list ] >> I filed: >> >> http://core.tcl.tk/tcl/tktview/267b7e2334ee2e9de34c4b00d6e72e2f1997085f > Slightly more than a year, but this is your wake-up call. > > The referred tcl bug is fixed with 8.6.1. This version is able to > finish gcc testsuite without issues. Thank

Re: PR debug/60655, debug loc expressions

On Tue, 9 Sep 2014, Alan Modra wrote: > This cures PR60655 on PowerPC by passing the horrible debug_loc > expressions we have through simplify_rtx. Not only do we get > reg10 + &.LANCHOR0 + const(0x14f - &.LANCHOR0) and > reg10 + &modulus + const(~&d_data), > the two expressions that

[PATCH i386 AVX512] [35/n] Add vperm[it]2 insns support.

Hello, Patch in the bottom extends perm[t|i] patterns. Bootstrapped. AVX-512* tests on top of patch-set all pass under simulator. Is it ok for trunk? gcc/ * config/i386/sse.md (define_expand "_vpermi2var3_maskz" with VI48F mode iterator): Rename from "avx512f_vper

Re: PR debug/60655, debug loc expressions

On Wed, Sep 10, 2014 at 01:43:22PM +0100, Maciej W. Rozycki wrote: > Thanks for your work on this issue, I have tested your change with my > usual powerpc-gnu-linux multilibs with the old and new result for > gcc.c-torture/compile/pr60655-2.c noted on the right: > > -mcpu=603e

[PATCH i386 AVX512] [36/n] Extend gather insn patterns.

Hello, Patch in the bottom extends gather instructions support. Bootstrapped. AVX-512* tests on top of patch-set all pass under simulator. Is it ok for trunk? gcc/ * config/i386/sse.md (define_expand "_gathersi"): Rename from "avx512f_gathersi". (define_insn "*avx

Re: [PATCH libstdc++ v5] - Add xmethods for std::vector and std::unique_ptr

On 04/09/14 05:56 -0700, Siva Chandra wrote: On Wed, Sep 3, 2014 at 3:47 PM, Siva Chandra wrote: On Wed, Sep 3, 2014 at 3:35 PM, Jonathan Wakely wrote: I was waiting to see which version of the patch actually works, so that users can use the xmethods. There's no point committing the patch if

Re: [PATCH] RE: gcc parallel make check

On Tue, Sep 09, 2014 at 03:14:08PM +, VandeVondele Joost wrote: > Attached is a further revision of the patch, now dealing with check-c++. > Roughly 50% speedup here at '-j32' (18m vs 12m). For my setup > (--enable-languages=c,c++,fortran) I have now improved all targets called in > 'make -

Re: PR debug/60655, debug loc expressions

On Wed, 10 Sep 2014, Alan Modra wrote: > I always build powerpc64-linux with --enable-targets=powerpc-linux and > regression test with RUNTESTFLAGS=--target_board=unix/'{-m32,-m64}', > so my "bootstrapped and regression tested powerpc64-linux" claim > includes a -m32 regression test too. Not quit

Re: [PATCH libstdc++ v5] - Add xmethods for std::vector and std::unique_ptr

On Wed, Sep 10, 2014 at 6:18 AM, Jonathan Wakely wrote: > I've committed the v5 patch to trunk now. Thanks a lot. > It would be nice to address Tom's comment at some future time, so we > only have one entry point in the hook.in file. I will do it before I add more xmethods. Thanks, Siva Chandr

RE: [PATCH] RE: gcc parallel make check

Thanks for testing. The vect-args.c I explained earlier, and is indeed due to i386.exp hardcoding those. The libstdc++ double counts didn't appear in my testing, but I'll have a look. Note that these patterns are handwritten, so error prone. The long tests in libstdc++ come from (in timing o

Re: [wwwdocs] Mention Cilk Plus support

Hi Igor, On Wed, 10 Sep 2014, Zamyatin, Igor wrote: > + Complete support for http://cilk.org";>Cilk Plus > features was added to GCC > + [2014-09-02] > + Contributed by Jakub Jelinek, Iyer Balaji and Igor Zamyatin. can you please make this "Cilk Plus support" or "Full Cilk Plus support"

Re: [PATCH] RE: gcc parallel make check

On Wed, Sep 10, 2014 at 01:57:01PM +, VandeVondele Joost wrote: > Thanks for testing. > > The vect-args.c I explained earlier, and is indeed due to i386.exp hardcoding > those. IMHO the best fix for that is following, use the same predicate whether to run the vect-args.c tests or not as is

C++ PATCH to 4.9 for c++/61659 (devirt referencing undefined comdats)

My change to always set DECL_COMDAT on vague-linkage decls has caused various problems on the trunk, so I'm not comfortable applying it to 4.9. So this patch does a much more limited version of the same thing: only set it on vague-linkage functions that might not be defined, and don't set it u

Re: [Patch ARM v2] Fix PR target/56846

On 10/09/14 15:41 +0800, Tony Wang wrote: -Original Message- From: Jonathan Wakely [mailto:jwak...@redhat.com] Sent: Tuesday, September 09, 2014 6:40 PM To: Tony Wang Cc: Ramana Radhakrishnan; gcc-patches; libstd...@gcc.gnu.org Subject: Re: [Patch ARM v2] Fix PR target/56846 On 09/09/14

[PATCH, committed] insn_extract takes an rtx_insn

The attached trivial patch strengthens the param of the generated "insn_extract" function, from rtx to rtx_insn *, and thus falls under the pre-approval granted by Jeff here: https://gcc.gnu.org/ml/gcc-patches/2014-08/msg01310.html Bootstrapped on x86_64-unknown-linux-gnu (Fedora 20), and has be

[PATCH, committed] error_for_asm and warning_for_asm take const rtx_insn *

The attached patch strengthens various diagnostic-reporting rtx from rtx to rtx_insn *, and thus falls under the pre-approval granted by Jeff here: https://gcc.gnu.org/ml/gcc-patches/2014-08/msg01310.html Bootstrapped on x86_64-unknown-linux-gnu (Fedora 20), and has been rebuilt as part of a con

[PATCH] Fix up libffi linux64*.S for ppc32-linux

Hi! I've noticed that on 4.8 branch libgo recently (in the last few months) started being linked with GNU_STACK 0x00 0x 0x 0x0 0x0 RWE 0x10 i.e. requiring executable stack on powerpc-linux (32-bit). The problem is that we link into libffi linux64.o and linux64_c

Re: [PATCH, 4.9, PR 61654] Handle newly truly expanded artificial_thunks

On Wed, Sep 03, 2014 at 10:45:34AM +0200, Martin Jambor wrote: > Hi, > > I did not think it was possible, but it can happen that when > duplicate_thunk_for_node creates a duplicate of a thunk which > previously expand_thunk left alone to be expanded into assembly by the > back end, the newly creat

Re: [C++, OpenMP] Crash when OpenMP target's array section handling is used with templates

Hi! On Wed, 10 Sep 2014 12:23:04 +0200, Jakub Jelinek wrote: > On Wed, Sep 10, 2014 at 12:12:03PM +0200, Thomas Schwinge wrote: > > Are the following issues known? > > No, please file a PR. Will do tomorrow. > Supposedly either we should call cp_omp_mappable_type > only for non-type-dependent

Re: [PATCH ARM] Fix PR target/63209

With gcc regression test, no regressions are found. David On Tue, Sep 9, 2014 at 11:45 AM, Xinliang David Li wrote: > Richard, thanks for the review. The revised patch is attached. Is this > one OK (after testing is done)? > > David > > > 2014-09-08 Xinliang David Li > > PR target/632

Re: PR debug/60655, debug loc expressions

2014-04-10 Ramana Radhakrishnan PR debug/60655 * config/arm/arm.c (TARGET_CONST_NOT_OK_FOR_DEBUG_P): Define (arm_const_not_ok_for_debug_p): Reject MINUS with SYM_REF's ameliorating the cases where it can be. can I suppose be reverted too. That was always s

Re: [ARM][tests] Make input and output arrays 128-bit aligned in vectorisation tests

On 09/09/14 16:14, Kyrill Tkachov wrote: Hi all, As Christophe mentioned at https://gcc.gnu.org/ml/gcc-patches/2014-09/msg00202.html These tests fail on big-endian. The reason is that the input is not aligned to 128 bit forcing the use of a movmisalign which we don't support on big-endian. A

Re: [Patch ARM v2] Fix PR target/56846

On 10/09/14 15:38, Jonathan Wakely wrote: On 10/09/14 15:41 +0800, Tony Wang wrote: -Original Message- From: Jonathan Wakely [mailto:jwak...@redhat.com] Sent: Tuesday, September 09, 2014 6:40 PM To: Tony Wang Cc: Ramana Radhakrishnan; gcc-patches; libstd...@gcc.gnu.org Subject: Re: [Pa

Re: [patch,avr] Make jump tables work with bootloader

2014-09-08 21:12 GMT+04:00 Georg Lay : > avr-gcc puts jump tables in section .progmem.gcc_sw_table and assumes that > .text starts at 0, i.e. LPM is on order to read table entries. > > This is no more the case if .text starts ata higher address like typically > used with boot loaders. > > This patc

[Patch, MIPS] Add .note.GNU-stack section

Someone noticed that the MIPS GCC compiler was not putting out the .note.GNU-stack section. This simple patch fixes that problem by calling the standard file_end_indicate_exec_stack function. Tested on mips-mti-linux-gnu, OK to checkin? Steve Ellcey sell...@mips.com 2014-09-10 Steve Ellcey

Re: [Patch, MIPS] Add .note.GNU-stack section

> On Sep 10, 2014, at 9:24 AM, "Steve Ellcey " wrote: > > Someone noticed that the MIPS GCC compiler was not putting out the > .note.GNU-stack section. This simple patch fixes that problem by > calling the standard file_end_indicate_exec_stack function. > > Tested on mips-mti-linux-gnu, OK to

Re: [Ping v2][PATCH] Add patch for debugging compiler ICEs.

On Wed, 10 Sep 2014, Jakub Jelinek wrote: > On Tue, Sep 09, 2014 at 10:51:23PM +, Joseph S. Myers wrote: > > On Thu, 28 Aug 2014, Maxim Ostapenko wrote: > > > > > diff --git a/gcc/diagnostic.c b/gcc/diagnostic.c > > > index 0cc7593..67b8c5b 100644 > > > --- a/gcc/diagnostic.c > > > +++ b/gcc/

Re: DBL_DENORM_MIN should never be 0

On Wed, 10 Sep 2014, Marc Glisse wrote: > Hello, > > according to the C++ standard, numeric_limits::denorm_min should return min > (not 0) when there are no denormals. > > Tested with bootstrap+testsuite on x86_64-linux-gnu. I also tested a basic > make all-gcc for vax (only target without denor

Re: [PATCH i386 AVX512] [34/n] AVX-512. Extend vpermvar insn patterns.

On Wed, Sep 10, 2014 at 2:37 PM, Kirill Yukhin wrote: > Hello, > patch in the bottom extends PERMVAR based patterns. > > Hello, > > Bootstrapped. > AVX-512* tests on top of patch-set all pass > under simulator. > > Is it ok for trunk? > > gcc/ > * config/i386/sse.md > (define_mode_

Re: [PATCH i386 AVX512] [35/n] Add vperm[it]2 insns support.

On Wed, Sep 10, 2014 at 3:05 PM, Kirill Yukhin wrote: > Hello, > Patch in the bottom extends perm[t|i] patterns. > > Bootstrapped. > AVX-512* tests on top of patch-set all pass > under simulator. > > Is it ok for trunk? > > gcc/ > * config/i386/sse.md > (define_expand "_vpermi2var3

[GOOGLE] Fix gcda build info support

While porting recent support for a build info section in the gcda from google/4_8 to 4_9 and doing manual testing, I discovered that it does not interact well with the COMDAT fixup handling. This patch fixes the issue, and adds a test case that exposes the problem without the fix. Here is the goog

Re: PR debug/60655, debug loc expressions

On Wed, 10 Sep 2014, Ramana Radhakrishnan wrote: > > 2014-04-10 Ramana Radhakrishnan > > > > PR debug/60655 > > * config/arm/arm.c (TARGET_CONST_NOT_OK_FOR_DEBUG_P): Define > > (arm_const_not_ok_for_debug_p): Reject MINUS with SYM_REF's > > ameliorating the cases where it can b

Another C++ PATCH for c++/61659 (DECL_COMDAT and devirt)

After * decl.c (maybe_commonize_var): Don't use DECL_COMDAT to trigger comdat_linkage. targets without weak symbol support started getting link errors bootstrapping GCC. This ought to fix that. Tested x86_64-pc-linux-gnu, applying to trunk. commit e3859ec10649526a24b439301db7

Re: [PATCH i386 AVX512] [36/n] Extend gather insn patterns.

On Wed, Sep 10, 2014 at 3:12 PM, Kirill Yukhin wrote: > Hello, > Patch in the bottom extends gather instructions support. > > Bootstrapped. > AVX-512* tests on top of patch-set all pass > under simulator. > > Is it ok for trunk? > > gcc/ > * config/i386/sse.md > (define_expand "_ga

RFA: PATCH to ipa-devirt for c++/58678

The failure with -flto is due to make_decl_local clearing DECL_COMDAT, so the check in ipa_devirt allows devirtualization to an implicitly declared destructor. It's not clear to me why make_decl_local needs to clear DECL_COMDAT, but it's simpler to just remove that check from ipa_devirt. Tes

[C++ Patch] PR 61489

Hi, the bug (Jon, in fact) argues that probably we shouldn't -Wmissing-field-initializers warn for cases like obj11 below (or warn5.C). I noticed that in practice this is also the specific behavior which clang++ implements for their warning with the same name. Tested x86_64-linux. Thanks, P

Re: [PATCH] gcc parallel make check

On Sep 10, 2014, at 7:16 AM, Jakub Jelinek wrote: > IMHO the best fix for that is following, use the same predicate whether to > run the vect-args.c tests or not as is used for all other tests. > > Ok for trunk? Looks suspiciously looks familiar. Ok. Thanks.

Re: [PATCH] gcc parallel make check

On Sep 9, 2014, at 8:14 AM, VandeVondele Joost wrote: > Attached is a further revision of the patch, now dealing with check-c++. So when last I played in this area, I wanted a command line tool that would bin-pack from the command line. I would then grab the seconds per for each .exp, and bin

Re: [PATCH ARM] Fix PR target/63209

On 09/09/14 19:45, Xinliang David Li wrote: > Richard, thanks for the review. The revised patch is attached. Is this > one OK (after testing is done)? > > David > > OK, but ... > 2014-09-08 Xinliang David Li > > PR target/63209 > * config/arm/arm.md (movcond_addsi): Handle

Re: [PATCH] rs6000: Use xori for HTM builtins and vector compares

On Tue, 2014-09-09 at 19:28 -0400, David Edelsohn wrote: > On Tue, Sep 9, 2014 at 3:29 PM, Segher Boessenkool > > 2014-09-09 Segher Boessenkool > > > > * config/rs6000/htm.md (tabort, tabortdc, tabortdci, tabortwc, > > tabortwci, tbegin, tcheck, tend, trechkpt, treclaim, tsr): Us

[jit] Merger from trunk into dmalcolm/jit

I've merged svn trunk r215090 (2014-09-09) into the dmalcolm/jit git branch [1], bringing in all changes since r203980 (2013-10-23); i.e. about 10.5 months of changes from trunk. Various manual fixes were necessary to enable the code to compile, and for the test suite to run successfully. With the

[jit] Update the TODO

Committed to branch dmalcolm/jit: gcc/jit/ChangeLog.jit: * TODO.rst (Test suite): Multithreaded test is done. --- gcc/jit/ChangeLog.jit | 4 gcc/jit/TODO.rst | 4 2 files changed, 4 insertions(+), 4 deletions(-) diff --git a/gcc/jit/ChangeLog.jit b/gcc/jit/ChangeLog.jit i

Re: [PATCHv2] Vimrc config with GNU formatting

On Wed, Sep 10, 2014 at 12:09:25PM +0400, Yury Gribov wrote: > The patch caused a lively debate with Segher who wanted .local.vimrc to > not be enabled > by default. No, that is not what I said. I am saying it is very anti-social to make people's editor behave differently from what they are used

Re: [PATCH] rs6000: Use xori for HTM builtins and vector compares

On Wed, Sep 10, 2014 at 01:38:13PM -0500, Peter Bergner wrote: > On Tue, 2014-09-09 at 19:28 -0400, David Edelsohn wrote: > > On Tue, Sep 9, 2014 at 3:29 PM, Segher Boessenkool > > > 2014-09-09 Segher Boessenkool > > > > > > * config/rs6000/htm.md (tabort, tabortdc, tabortdci, tabortwc,

Committed: update simtest-howto.html

(Thanks to people CC'ed and others forgotten; I hope I incorporated at least some of everyone's suggestions.) First, as noted, the instructions are outdated due to repos merging, splitting, moving and switching, with fallout such as it now seemed odd as-is with one minor component randomly being n

Re: [PATCH ARM] Fix PR target/63209

Fixed the formatting and committed it to trunk: r215136. Will backport it to gcc-4_9 branch. thanks, David On Wed, Sep 10, 2014 at 11:24 AM, Richard Earnshaw wrote: > On 09/09/14 19:45, Xinliang David Li wrote: >> Richard, thanks for the review. The revised patch is attached. Is this >> one OK

Re: Committed: update simtest-howto.html

Hi, On Wed, 2014-09-10 at 15:46 -0400, Hans-Peter Nilsson wrote: > (Thanks to people CC'ed and others forgotten; I hope I > incorporated at least some of everyone's suggestions.) > > [...] > Also, the idea of using a *combined* tree here, has been challenged, > so I toned down the wording from th

[PATCH], Fix constraints on VSX Fma, Fix, and Reduce options

In doing work on improving power8 fusion support, I noticed that in several of the patterns (vector fused multiply-add, optimization of float (fix (x)), and vector reduction), I used the "ws" constraint which is the constraint for scalar double precision floating point (currently FLOAT_REGS) in cas

Re: [PATCH] rs6000: Use xori for HTM builtins and vector compares

On Wed, 2014-09-10 at 14:29 -0500, Segher Boessenkool wrote: > Huh, interesting. I assumed 1-(0_or_1) and (0_or_1)^1 would look the > same to combine, but no. > > With subfic, combine optimises it all to a branch on cr0. With xori, > for some reason combine has a much easier job, and it optimise

RE: [PATCH] RE: gcc parallel make check

> You mean enhancing the script to split across arbitrarily long prefixes? > That would be great. I've now a script that does something like that: ~/test$ find /data/vjoost/gnu/gcc_trunk/gcc/gcc/testsuite/gfortran.dg/ -maxdepth 1 -type f -printf "%f\n" | ./generate_patterns.py 500 foo All 3947

Re: [PATCH] gcc parallel make check

On Wed, 2014-09-10 at 11:19 -0700, Mike Stump wrote: > On Sep 9, 2014, at 8:14 AM, VandeVondele Joost > wrote: > > Attached is a further revision of the patch, now dealing with > check-c++. > > So when last I played in this area, I wanted a command line tool that > would bin-pack from the command

Re: [PATCH], Fix constraints on VSX Fma, Fix, and Reduce options

On Wed, Sep 10, 2014 at 4:16 PM, Michael Meissner wrote: > In doing work on improving power8 fusion support, I noticed that in several of > the patterns (vector fused multiply-add, optimization of float (fix (x)), and > vector reduction), I used the "ws" constraint which is the constraint for > sc

Re: [PATCH], Fix constraints on VSX Fma, Fix, and Reduce options

On Wed, Sep 10, 2014 at 04:42:06PM -0400, David Edelsohn wrote: > Needs a ChangeLog. Whoops, I forgot to include it: 2014-09-10 Michael Meissner * config/rs6000/vsx.md (vsx_fmav4sf4): Use correct constraints for V2DF, V4SF, DF, and DI modes. (vsx_fmav2df2): Likewise.

[Bug libstdc++/62313] Data race in debug iterators

Hi Here is a proposal to fix this data race issue. I finally generalized bitset approach to fix it by inheriting from the normal iterator first and then the _Safe_iterator_base type. None of the libstdc++ iterator types are final so it is fine. Surprisingly, despite inheritance being

Re: Committed: update simtest-howto.html

On Wed, 10 Sep 2014, Oleg Endo wrote: > On Wed, 2014-09-10 at 15:46 -0400, Hans-Peter Nilsson wrote: > > (Thanks to people CC'ed and others forgotten; I hope I > > incorporated at least some of everyone's suggestions.) > > > > [...] > > Also, the idea of using a *combined* tree here, has been chall

Re: [PATCH] gcc parallel make check

On Sep 10, 2014, at 1:38 PM, David Malcolm wrote: > Perhaps this is a silly question, but has anyone tried going the whole > way and not having buckets, going to an extremely fine-grained approach No, we fear the overhead, but do not know what it is.

Re: [PATCH] gcc parallel make check

On Wed, 10 Sep 2014, David Malcolm wrote: > (A) test discovery; write out a fine-grained Makefile in which *every* > testcase is its own make target (to the extreme limit of > parallelizability e.g. on the per-input-file level) The DejaGnu design doesn't allow test discovery in general (as the se

Re: [PATCH] gcc parallel make check

On Wed, Sep 10, 2014 at 04:38:32PM -0400, David Malcolm wrote: > > So when last I played in this area, I wanted a command line tool that > > would bin-pack from the command line. I would then grab the seconds > > per for each .exp, and bin pack to the fixed N, where N was the core > > count or rel

PR 63186 (fnsplit and forced labels)

Hi, this patch fixes the testcasein PR63186. Here a forced label is used and fnsplit decides to split its use from definition that does not work. There is code that is supposed to prevent this. It however works on non-ssa declarations in general and it has an optimization - it allows the non-ssa d

Re: [PATCH] gcc parallel make check

On Wed, Sep 10, 2014 at 11:08:22PM +0200, Jakub Jelinek wrote: > Perhaps better approach might be if we have some way how to synchronize among > multiple expect processes and spawn only as many expects (of course, per > check target) as there are CPUs. E.g. if mkdir is atomic on all > hosts/filesy

Re: [PATCH] gcc parallel make check

On Sep 10, 2014, at 2:23 PM, Jakub Jelinek wrote: > Seems file mkdir in tcl doesn't error on pre-existing directory, shell mkdir will. :-)

Re: [GOOGLE] Fix gcda build info support

Can you share the buildinfo reader code with the merger by defininig some hooks for different callbacks? David On Wed, Sep 10, 2014 at 10:24 AM, Teresa Johnson wrote: > While porting recent support for a build info section in the gcda from > google/4_8 to 4_9 and doing manual testing, I discover

Re: RFA: PATCH to ipa-devirt for c++/58678

> The failure with -flto is due to make_decl_local clearing > DECL_COMDAT, so the check in ipa_devirt allows devirtualization to > an implicitly declared destructor. It's not clear to me why > make_decl_local needs to clear DECL_COMDAT, but it's simpler to just > remove that check from ipa_devirt.

Re: RFA: PATCH to ipa-devirt for c++/58678

Hi and forgot to add, I can implement the change if it seems to make sense to you ;) Honza

Re: [PATCH] Fix up libffi linux64*.S for ppc32-linux

On Wed, Sep 10, 2014 at 04:48:33PM +0200, Jakub Jelinek wrote: > * src/powerpc/linux64.S: Emit .note.GNU-stack even when > POWERPC64 is not defined. > * src/powerpc/linux64_closure.S: Likewise. Also test _CALL_ELF == 2. Looks reasononable to me, and oops, I think I introduced th

[PATCH] Introduce LABEL_REF_LABEL

The following patch adds a macro: /* Get the label that a LABEL_REF references. */ #define LABEL_REF_LABEL(LABREF) XCEXP (LABREF, 0, LABEL_REF) and uses it in place of XEXP (foo, 0) for "foo" known to be a LABEL_REF, throughout the "gcc" directory, in the hope of (A) improving the clarity of

Re: RFA: PATCH to ipa-devirt for c++/58678

On 09/10/2014 06:47 PM, Jan Hubicka wrote: Again - I think main problem is that we provide a middle end a function body that it is not allowed to use. We do not really have a concept of function definitions that we may not use and worse yet, we are preventing that just in one special case - i.e.

[Ping] Port of VTV for Cygwin and MinGW

Ping for https://gcc.gnu.org/ml/gcc-patches/2014-08/msg02559.html Also added Caroline Tice, as libvtv maintainer, to cc and attached virtual_func_test_min_UAF.cpp, which I forgot in the original mail. Patrick On 28.08.2014 13:03, Patrick Wollgast wrote: > This patch contains a port of VTV -fvtab

Re: [PATCHv2] Vimrc config with GNU formatting

Segher Boessenkool kernel.crashing.org> writes: > I am saying it is very anti-social to make > people's editor behave differently from what they are used to. > ... > The Emacs dir-locals file simply > configures some settings for files with certain major modes in that dir. > For example, ours say

Re: [middle-end/PATCH 0/2] Fix AARCH64 ILP32 ld.so miscompiling

Ping. On Fri, Aug 8, 2014 at 8:51 PM, Andrew Pinski wrote: > The problem is that HJL's patch in 2011 to try and fix his miscompiling of x32 > ld.so was in fact incorrect and causes problems with AARCH64 ILP32. > The testcase which is being miscompiled after HJL's patch (which he was also > trying

Re: RFA: PATCH to ipa-devirt for c++/58678

> On 09/10/2014 06:47 PM, Jan Hubicka wrote: > >Again - I think main problem is that we provide a middle end a function body > >that it is not allowed to use. We do not really have a concept of function > >definitions that we may not use and worse yet, we are preventing that just in > >one special

Re: DBL_DENORM_MIN should never be 0

On Wed, 10 Sep 2014, Joseph S. Myers wrote: On Wed, 10 Sep 2014, Marc Glisse wrote: Hello, according to the C++ standard, numeric_limits::denorm_min should return min (not 0) when there are no denormals. Tested with bootstrap+testsuite on x86_64-linux-gnu. I also tested a basic make all-gcc

RE: [PATCH] RE: gcc parallel make check

Jakub, > First of all, the -j2 testing shows more tests tested in gcc and libstdc++: > >-# of expected passes 10133 >+# of expected passes 10152 > >+PASS: 23_containers/set/modifiers/erase/abi_tag.cc (test for excess errors) >[...] > >Not sure where the bug is, could be e.g. in

  1   2   >