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

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 Supplement (16 July 2015 Version 1.1). There are > many of the builtins

Re: openacc reference reductions

On 02/22/2016 07:34 AM, Cesar Philippidis wrote: > Ping. This patch still needs a review. Ping. I've attached a rebased version of this patch. The omp-low.c bits haven't changed, but the test cases have since Thomas has been merging some of them from trunk. I separated the omp-low.c changes becaus

Re: [committed] Fix linux blk-merge boot problem on hppa

On 2016-03-17, at 7:38 PM, John David Anglin wrote: > The attach patch fixes a problem causing block/blk-merge.c in the linux > kernel to be miscompiled. As a result, > block segments were not properly split and boot failed since linux 4.3. > > The problem was found by a regression search. The

Re: a patch for PR68695

On Tue, Apr 05, 2016 at 10:48:58AM +0100, Kyrill Tkachov wrote: > So for the test gcc.dg/pr10474.c on arm with -marm -O3 before this patch we > perform shrink-wrapping: > cmpr0, #0 > bxeqlr > push{r4, lr} > movr4, r0 > ... > > And after the patch we don't: >

Re: [AArch64] Emit square root using the Newton series

On 04/05/16 13:37, Wilco Dijkstra wrote: I can't get any of these to work... Not only do I get a large number of collisions and duplicated code between these patches, when I try to resolve them, all I get is crashes whenever I try to use sqrt (even rsqrt stopped working). Do you have a patchset

Re: [PATCH] C++/70512 Fix may_alias canonical type mismatch

OK. Jason

[PATCH] C++/70512 Fix may_alias canonical type mismatch

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=70512 this fixes the canonical type mismatch of 70512. As surmised we don't correct the ptr-to and ref-to types created during the structure parse when late-applying the may_alias attribute. This patch adjusts fixup_attribute_variants to walk th

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

On 04/05/16 12:40, Jason Merrill wrote: It's not clear to me that we really need a TARGET_EXPR for vector values. Since one element of a vector can't refer to another, we don't need the ctx->ctor handling. Perhaps we should handle vectors like we do PMF types in cxx_eval_bare_aggregate? That

Re: Do not optimize some polymorphic calls with -fsanitize=undefined

Hi, this is patch I commited to address the feedback. Regtested on x86_64-linux Honza * ipa-devirt.c (maybe_record_node): Fix comment; use SANITIZE_UNREACHABLE instead of SANITIZE_UNDEFINED. * g++.dg/ipa/devirt-51.C: Use -fsanitize=unreachable. Index: ipa-devirt.c ==

Re: [PATCH] Limit some match.pd conversion folding to GIMPLE (PR c++/70336)

On April 5, 2016 6:47:20 PM GMT+02:00, Jakub Jelinek wrote: >Hi! > >As the testcase shows, the folding which previously has been done >only in tree-ssa-forwprop.c breaks some cases with explicit casts >in -Wconversion, so this patch limits that folding to GIMPLE only. >Unfortunately that breaks a

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

On 04/04/2016 01:26 PM, Nathan Sidwell wrote: https://gcc.gnu.org/bugzilla/show_bug.cgi?id=70501 This fixes 70501. The cause is an omission in typeck when converting a scalar operand to a vector. We use build_vector_from_val, which can return a CONSTRUCTOR. We fail to wrap that CONSTRUCTOR in

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

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 Supplement (16 July 2015 Version 1.1). There are many of the builtins that are missing and this is the first of a series of patche

Re: [PR c/68966] Restore atomic builtins usage in libstdc++-v3

On 05/04/16 19:08 +0100, Jonathan Wakely wrote: On 05/04/16 12:01 +0100, Jonathan Wakely wrote: Well I guess it's mine, and this is a fairly serious regression (is it tracked in Bugzilla anywhere?) so the patch is OK for trunk. This is now https://gcc.gnu.org/bugzilla/show_bug.cgi?id=70554 I

Re: [PATCH, cpp] Fix pr61817 and 69391

On 04/05/2016 11:03 AM, Manuel López-Ibáñez wrote: > Why use -ftrack-macro-expansion=0? That's the point of the PR -- we were producing totally bogus results. r~

Re: [PR c/68966] Restore atomic builtins usage in libstdc++-v3

On 05/04/16 12:01 +0100, Jonathan Wakely wrote: Well I guess it's mine, and this is a fairly serious regression (is it tracked in Bugzilla anywhere?) so the patch is OK for trunk. This is now https://gcc.gnu.org/bugzilla/show_bug.cgi?id=70554

Re: [PATCH, cpp] Fix pr61817 and 69391

On 05/04/16 17:22, Richard Henderson wrote: These two related PRs are all about remembering where a macro is expanded. Worse, we've got two competing goals -- the real location of the expansion, for __LINE__, and the virtual location of the expansion, for diagnostics. There seems to be no way to

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

On 05/04/16 14:37 +0100, Jonathan Wakely wrote: Ensure std::thread helpers have internal linkage PR libstdc++/70503 * src/c++11/thread.cc (execute_native_thread_routine, execute_native_thread_routine_compat): Give internal linkage. * testsuite/30_threads/thread

Re: [PATCH] Fix PR70509 (wrong code with extract from a v64qi)

On Mon, Apr 04, 2016 at 11:16:48PM +0200, Zdenek Sojka wrote: > gcc/Changelog: > > 2016-04-04  Zdenek Sojka  > >     PR tree-optimization/70509 >     * tree-ssa-forwprop.c (simplify_bitfield_ref): Use bitsize_int () instead > of the vector base type for index. > > > gcc/testsuite/Changelog: >

Re: [PATCH] Fix REE with vector modes (PR rtl-optimization/70542)

On 04/05/2016 10:56 AM, Jakub Jelinek wrote: Hi! As mentioned in PR64286 already, unlike integral modes, vector modes don't have the property that lowpart subreg of sign/zero extended value contains the original non-extended bits, so if we touch some definition, we really have to change all uses

[PATCH] Fix REE with vector modes (PR rtl-optimization/70542)

Hi! As mentioned in PR64286 already, unlike integral modes, vector modes don't have the property that lowpart subreg of sign/zero extended value contains the original non-extended bits, so if we touch some definition, we really have to change all uses. Except that we really don't have infrastruct

[PATCH] Limit some match.pd conversion folding to GIMPLE (PR c++/70336)

Hi! As the testcase shows, the folding which previously has been done only in tree-ssa-forwprop.c breaks some cases with explicit casts in -Wconversion, so this patch limits that folding to GIMPLE only. Unfortunately that breaks a few cases in builtin-isinf_sign-1.c test at -O0, but IMNSHO expecti

[PATCH, i386]: Fix PR70510, invalid %-code with -mavx512bw -masm=intel when emitting vpbroatcast

2016-04-05 Uros Bizjak PR target/70510 * config/i386/sse.md (iptr): Add V64QI, V32HI, V16SI and V8DI modes. testsuite/ChangeLog: 2016-04-05 Uros Bizjak PR target/70510 * gcc.target/i386/pr70510.c: New test. Bootstrapped and regression tested on x86_64-linux-gnu {,-m32}. C

Re: [wwwdocs] [1/3] projects/cxx-status.html -- introduce global CSS for tables

On Sat, 2 Apr 2016, Gerald Pfeifer wrote: > Index: gcc.css > === : > +/* C++ status tables. */ > +table.cxxstatus th, td { border: 1px solid gray; } Intuition is not always a good guide. The above applied this style to all elements.

[PATCH, cpp] Fix pr61817 and 69391

These two related PRs are all about remembering where a macro is expanded. Worse, we've got two competing goals -- the real location of the expansion, for __LINE__, and the virtual location of the expansion, for diagnostics. There seems to be no way to unify the two competing goals. If we simp

[wwwdocs] [5/3] projects/cxx-status.html -- introduce global CSS for tables

Somehow four cases of inline styles had remained, which did not actually make a difference anymore, but still can be simplified. Applied. Gerald Remove four final, and superflous cases of style="text-align:center;". Index: projects/cxx-status.html ===

Re: [PING][PATCH] Remove incorrect warning for parallel firstprivate clause

On Tue, Apr 05, 2016 at 12:17:16PM +0200, Tom de Vries wrote: > On 24/03/16 18:02, Tom de Vries wrote: > >Remove incorrect warning for parallel firstprivate clause > > > >2016-03-24 Tom de Vries > > > > * omp-low.c (lower_omp_target): Set TREE_NO_WARNING for oacc > > firstprivate clause.

Re: [wwwdocs] [1/3] projects/cxx-status.html -- introduce global CSS for tables

On Tue, 5 Apr 2016, Jason Merrill wrote: >> href="http://www.open-std.org/jtc1/sc22/wg21/docs/papers/2006/n2118.html";>N2118 >> GCC >> 4.3 >> __cpp_rvalue_references >= 200610 >> + >> > > This doesn't seem to have been needed; now when I look at the page I see > an e

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

This makes the execute_native_thread_routine* helpers 'static' rather than using an unnamed namespace, because the latter doesn't give them static linkage due to them being extern "C". Tested x86_64-linux, committed to trunk. Backports to gcc-5 and gcc-4.9 to follow soon. commit 84538059d9d8608c

Re: [PATCH] Fix PR c++/70452 (regression in C++ parsing performance)

OK, thanks. Jason

Re: [wwwdocs] [1/3] projects/cxx-status.html -- introduce global CSS for tables

Thanks a lot! One thing: On 04/02/2016 06:19 AM, Gerald Pfeifer wrote: @@ -390,6 +389,7 @@ http://www.open-std.org/jtc1/sc22/wg21/docs/papers/2006/n2118.html";>N2118 GCC 4.3 __cpp_rvalue_references >= 200610 + Rvalue references for *this

Re: [RFC] introduce --param max-lto-partition for having an upper bound on partition size

On Tue, 5 Apr 2016, Prathamesh Kulkarni wrote: > On 5 April 2016 at 16:58, Richard Biener wrote: > > On Tue, 5 Apr 2016, Prathamesh Kulkarni wrote: > > > >> On 4 April 2016 at 19:44, Jan Hubicka wrote: > >> > > >> >> diff --git a/gcc/lto/lto-partition.c b/gcc/lto/lto-partition.c > >> >> index 9e

Re: [RFC] introduce --param max-lto-partition for having an upper bound on partition size

On 5 April 2016 at 16:58, Richard Biener wrote: > On Tue, 5 Apr 2016, Prathamesh Kulkarni wrote: > >> On 4 April 2016 at 19:44, Jan Hubicka wrote: >> > >> >> diff --git a/gcc/lto/lto-partition.c b/gcc/lto/lto-partition.c >> >> index 9eb63c2..bc0c612 100644 >> >> --- a/gcc/lto/lto-partition.c >> >

[PATCH] Fix PR70526

SRA uses bogus alias sets for access replacements if a alias-punning MEM_REF is based on a decl. The following patch fixes this. Bootstrapped on x86_64-unknown-linux-gnu, testing in progress. The testcase is a GCC 6 regression but the issue is older. Richard. 2016-04-05 Richard Biener

Re: [RFC] introduce --param max-lto-partition for having an upper bound on partition size

On Tue, 5 Apr 2016, Prathamesh Kulkarni wrote: > On 4 April 2016 at 19:44, Jan Hubicka wrote: > > > >> diff --git a/gcc/lto/lto-partition.c b/gcc/lto/lto-partition.c > >> index 9eb63c2..bc0c612 100644 > >> --- a/gcc/lto/lto-partition.c > >> +++ b/gcc/lto/lto-partition.c > >> @@ -511,9 +511,20 @@

Re: [v3 PATCH] PR libstdc++/70437

On 05/04/16 14:07 +0300, Ville Voutilainen wrote: On 5 April 2016 at 13:53, Jonathan Wakely wrote: I wonder if we want an __is_samey trait that checks if two decayed types are the same. If such checks become more common, then yes. For now, perhaps not. We already do it in packaged_task, fun

Re: [PATCH] Fix detection of setrlimit in libstdc++ testsuite

This patch fixes an obscure cross-testing problem that crashed (OOMed) our boards at Linaro. Several tests in libstdc++ (e.g., [1]) limit themselves to some reasonable amount of RAM and then try to allocate 32 gigs. Unfortunately, the configure test that checks presence of setrlimit is rather

Re: [PATCH][GCC 4.9][PR sanitizer/70474] Fix ASan failures on new Darwin platforms (OS X 10.11+, iOS 9.0+).

On Tue, Apr 05, 2016 at 02:05:12PM +0300, Maxim Ostapenko wrote: > this is adapted patch from > https://gcc.gnu.org/ml/gcc-patches/2016-01/msg00803.html that fixes exactly > the same issue (ASan fails on newer Darwin platforms) for GCC 4.9.4. > Regtested and bootstrapped on x86_64-unknown-linux-gnu

Re: [RFC] introduce --param max-lto-partition for having an upper bound on partition size

On 4 April 2016 at 19:44, Jan Hubicka wrote: > >> diff --git a/gcc/lto/lto-partition.c b/gcc/lto/lto-partition.c >> index 9eb63c2..bc0c612 100644 >> --- a/gcc/lto/lto-partition.c >> +++ b/gcc/lto/lto-partition.c >> @@ -511,9 +511,20 @@ lto_balanced_map (int n_lto_partitions) >>varpool_order.qs

Re: [v3 PATCH] PR libstdc++/70437

On 5 April 2016 at 13:53, Jonathan Wakely wrote: > I wonder if we want an __is_samey trait that checks if two decayed > types are the same. If such checks become more common, then yes. For now, perhaps not. > More seriously, a comment might be useful to explain that although > these "concepts" r

[PATCH][GCC 4.9][PR sanitizer/70474] Fix ASan failures on new Darwin platforms (OS X 10.11+, iOS 9.0+).

Hi, this is adapted patch from https://gcc.gnu.org/ml/gcc-patches/2016-01/msg00803.html that fixes exactly the same issue (ASan fails on newer Darwin platforms) for GCC 4.9.4. Regtested and bootstrapped on x86_64-unknown-linux-gnu by myself and on x86_64-apple-darwin15.2.0 by Dominique, OK for

Re: [PR c/68966] Restore atomic builtins usage in libstdc++-v3

On 21/03/16 17:01 +0100, Thomas Schwinge wrote: Hi! On Mon, 21 Mar 2016 15:01:49 +, Jonathan Wakely wrote: On 21/03/16 13:08 +0100, Thomas Schwinge wrote: >Per my (admittedly, not in-depth) reading of libstdc++-v3 source code, >the _GLIBCXX_ATOMIC_BUILTINS conditional is only used in combi

Re: [v3 PATCH] PR libstdc++/70437

On 04/04/16 21:45 +0300, Ville Voutilainen wrote: And yes, -ENOPATCH. On 4 April 2016 at 21:42, Ville Voutilainen wrote: Tested on Linux-PPC64. 2016-04-04 Ville Voutilainen PR libstdc++/70437 * include/bits/stl_pair.h (_ConstructiblePair, _ImplicitlyConvertiblePair, _MoveCon

[PING][PATCH] Remove incorrect warning for parallel firstprivate clause

On 24/03/16 18:02, Tom de Vries wrote: Hi, This patch fixes an incorrect warning for the oacc firstprivate clause. Consider this test-case: ... void foo (void) { int i; #pragma acc parallel { i = 1; } } ... When compiling with -fopenacc -Wuninitialized, we get an 'is used unini

[PING][PATCH] Remove incorrect warning for kernels copy clause

On 24/03/16 17:59, Tom de Vries wrote: Hi, This patch fixes an incorrect warning for the oacc copy clause. Consider this test-case: ... void foo (void) { int i; #pragma acc kernels { i = 1; } } ... When compiling with -fopenacc -Wuninitialized, we get an 'is used uninitialized'

Re: [PATCH] Fix PR70484, RTL DSE using wrong dependence check

On Tue, Apr 05, 2016 at 11:48:29AM +0200, Richard Biener wrote: > Like this? Yeah, thanks. > 2016-04-05 Richard Biener > > * dse.c (struct store_info): Remove alias_set member. > (struct read_info_type): Likewise. > (clear_alias_group, clear_alias_mode_table, clear_alias_mod

Re: a patch for PR68695

Hi all, On 01/04/16 21:43, Jakub Jelinek wrote: On Fri, Apr 01, 2016 at 04:26:41PM -0400, Vladimir Makarov wrote: I've noticed that after this patch, 2 tests regress (PASS -> FAIL) on arm: gcc.dg/ira-shrinkwrap-prep-2.c scan-rtl-dump pro_and_epilogue "Performing shrink-wrapping" gcc.dg/pr

Re: [PATCH] Fix PR70484, RTL DSE using wrong dependence check

On Tue, 5 Apr 2016, Jakub Jelinek wrote: > On Tue, Apr 05, 2016 at 11:05:44AM +0200, Richard Biener wrote: > > True. By simple constant propagation I can remove a lot of code. > > > > I'm going to bootstrap / test the following - is this ok for trunk > > now (I'm going to write a better changelo

Re: [PING 5, PATCH] PR/68089: C++-11: Ingore "alignas(0)".

On Mon, Jan 04, 2016 at 12:33:21PM +0100, Dominik Vogt wrote: > On Fri, Jan 01, 2016 at 05:53:08PM -0700, Martin Sebor wrote: > > On 12/31/2015 04:50 AM, Dominik Vogt wrote: > > >The attached patch fixes C++-11 handling of "alignas(0)" which > > >should be ignored but currently generates an error m

Re: [RFC] PR70117, ppc long double isinf

On Tue, Apr 5, 2016 at 10:33 AM, Alan Modra wrote: > This patch fixes the incompatibility between GNUlib's 107 bit > precision LDBL_MAX for IBM extended precision and gcc's 106 bit > LDBL_MAX used to test for Inf, by just testing the high double for inf > and nan. This agrees with the ABI which h

Re: [PATCH] Fix PR70484, RTL DSE using wrong dependence check

On Tue, Apr 05, 2016 at 11:05:44AM +0200, Richard Biener wrote: > True. By simple constant propagation I can remove a lot of code. > > I'm going to bootstrap / test the following - is this ok for trunk > now (I'm going to write a better changelog). LGTM with better Changelog, though I have small

Re: [PATCH] Fix PR31096

On Tue, 5 Apr 2016, Hurugalawadi, Naveen wrote: Hi, Looks like you are turning x*-1 < y*-1 into x Please find attached the modified patch that works on integer constant values. Please review the patch and let me know if this is okay? -1 is an integer constant, so that's still invalid. It i

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

On 18/03/16 13:37, Thomas Schwinge wrote: Hi! On Wed, 9 Mar 2016 10:17:28 +0100, Tom de Vries wrote: [Should have cited instead of the C/C++ tests] Retested on current trunk. Committed, minus the kernels-p

[PING 3] [PATCH] Fix PR 31531: A microoptimization of isnegative of signed integer

Hi, Please review the patch at the following link and let me know if there should be any modifications in it:- https://gcc.gnu.org/ml/gcc-patches/2016-02/msg01035.html Thanks, Naveen

Re: [PATCH] Fix PR31096

Hi, >> Looks like you are turning x*-1 < y*-1 into xdiff --git a/gcc/match.pd b/gcc/match.pd index c0ed305..e073e9f 100644 --- a/gcc/match.pd +++ b/gcc/match.pd @@ -894,7 +894,11 @@ DEFINE_INT_AND_FLOAT_ROUND_FN (RINT) && tree_nop_conversion_p (type, TREE_TYPE (@1))) (convert (bit_and (b

Re: [PATCH] Fix PR70484, RTL DSE using wrong dependence check

On Mon, 4 Apr 2016, Jakub Jelinek wrote: > On Mon, Apr 04, 2016 at 11:24:41AM +0200, Richard Biener wrote: > > On Fri, 1 Apr 2016, Bernd Schmidt wrote: > > > > > On 04/01/2016 11:08 AM, Richard Biener wrote: > > > > { > > > > ! if (canon_true_dependence (s_info->mem, > > > > !

[RFC] PR70117, ppc long double isinf

This patch fixes the incompatibility between GNUlib's 107 bit precision LDBL_MAX for IBM extended precision and gcc's 106 bit LDBL_MAX used to test for Inf, by just testing the high double for inf and nan. This agrees with the ABI which has stated for many years that IBM extended precision "does n

[PATCH: RL78] Optimize libgcc routines using clrw and clrb

Hi, Please find below a patch that optimizes libgcc routines for the RL78 target. This is similar to my earlier patch submitted here, https://gcc.gnu.org/ml/gcc-patches/2016-02/msg00415.html The patch optimizes the loading of immediate value in the case of 0x00, by using the clrw or clrb instruc

[PATCH] Fix PR70499

The following patch fixes a SSA rewrite issue in gimple_regimplify_operands (some bigger rewrite is IMHO necessary here or in the caller in this case, the inliner, but that's not appropriate now or for branches). Bootstrapped and tested on x86_64-unknown-linux-gnu, applied to trunk. Richard. 2

Re: [PATCH] Fix up AVX512 andnot (PR target/70525)

On Mon, Apr 4, 2016 at 7:44 PM, Jakub Jelinek wrote: > Hi! > > This patch fixes various *andnot3* issues. There are two issues on > the ISA side that makes stuff harder for andnot: there are no VPANDNB and > VPANDNW instructions, and while there used to be just VPANDN instruction > in AVX/AVX2, t

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

On Mon, Apr 4, 2016 at 4:14 PM, Bin.Cheng wrote: > On Mon, Apr 4, 2016 at 2:07 PM, Richard Biener > wrote: >> On Thu, Mar 31, 2016 at 6:43 PM, Bin.Cheng wrote: >>> On Tue, Mar 29, 2016 at 9:37 AM, Richard Biener >>> wrote: On Mon, Mar 28, 2016 at 9:57 PM, Bin.Cheng wrote: > Sorry, Sho

[PATCH] Remove XFAIL from gcc.dg/tree-ssa/20030814-6.c

This removes the XFAIL by fixing the testcase - it cannot be ever optimized in the desired way because the write to type.alias_set _does_ kill the previous load. But that's because the testcase bogously simplified the tree inheritance. Fixed. Also disabled FRE so this really tests DOM. (all th