Re: [PATCH] Selftest framework (v7)

On Mon, Jun 06, 2016 at 11:57:49PM +0200, Jakub Jelinek wrote: > On Mon, Jun 06, 2016 at 05:53:50PM -0400, Trevor Saunders wrote: > > > > As far as I can > > > > tell this just involves moving the start of namespace selftest > > > > upwards a > > > > bit in the files where we have tests. > > >

[committed] Generate indirect long calls to non-local functions on 64-bit hppa

The attached patch generates indirect long calls to non-local functions on 64-bit hppa. This improves opportunities for optimization and scheduling. Tested on hppa64-hp-hpux11.11 with no observed regressions. Dave -- John David Anglin dave.ang...@bell.net 2016-06-06 John David Anglin

[committed] Remove R1 clobbers from 64-bit indirect call value patters and splitters

The instruction sequence generated for 64-bit indirect calls on hppa does not clobber register %r1, so the clobbers for this register can be removed. Tested on hppa64-hp-hpux11.11 with no observed regressions. Dave -- John David Anglin dave.ang...@bell.net 2016-06-06 John David Anglin

[PATCH,rs6000] Add built-in function support for new Power9 vector absolute difference unsigned instructions

This patch adds built-in function support for the ISA 3.0 vabsub, vabsduh, and vabsduw instructions. I have bootstrapped and tested on powerpc64le-unkonwn-linux-gnu with no regressions. Is this ok for the trunk? I have also tested against the gcc-6 branch without regressions. Is this ok

[PATCH, i386]: Insert CLD insns using mode-switching pass

Hello! Attached patch inserts CLD instruction at optimal location using mode-switching pass, so there is no unnecessary access to flags reg in paths that don't use calls or string instructions. The patch handles insertions for interrupt handler functions and also insertions for legacy

Re: [Diagnostic Patch] Clean-up diagnostic facilities in diagnostic.c

Hi David, On 06/06/2016 22:26, David Malcolm wrote: Thanks, looks like a nice simplification. I see the new prototypes are wrapped in #ifdef ATTRIBUTE_GCC_DIAG. Isn't that redundant? Looking at diagnostic-core.h, isn't it always defined? (perhaps to the empty string) Also the new functions

Re: [PATCH] Selftest framework (v7)

On Mon, Jun 06, 2016 at 05:53:50PM -0400, Trevor Saunders wrote: > > > As far as I can > > > tell this just involves moving the start of namespace selftest > > > upwards a > > > bit in the files where we have tests. > > > > Yes, and it does seem cleaner to have all of the selftest code start >

Re: [PATCH] Selftest framework (v7)

> > As far as I can > > tell this just involves moving the start of namespace selftest > > upwards a > > bit in the files where we have tests. > > Yes, and it does seem cleaner to have all of the selftest code start > like this: > > #if CHECKING_P What are we gaining by ifdefing this? I

Re: C PATCH to improve location for abstract declarators (PR c/71362)

On Fri, 3 Jun 2016, Marek Polacek wrote: > This fixes an imprecise location info with abstract declarators. The problem > was that when we build_id_declarator, the default location was input_location > and we never attempted to use a more precise location. This patch does it. > >

Re: [PING] [PR other/70945] Handle function_glibc_finite_math in offloading

On Fri, 3 Jun 2016, Jakub Jelinek wrote: > On Fri, Jun 03, 2016 at 04:44:15PM +0200, Thomas Schwinge wrote: > > Hi! > > > > Ping. > > I think it would be better to just add this support to newlib. That suggestion doesn't really make sense to me. Why should newlib be expected to follow the

[Fortran, Patch] First patch for coarray FAILED IMAGES (TS 18508)

Dear all, please find in attachment the first patch (of n) for the FAILED IMAGES capability defined in the coarray TS 18508. The patch adds support for three new intrinsic functions defined in the TS for simulating a failure (fail image), checking an image status (image_status) and getting the

[PATCH] Add ggc-tests.c

Jeff approved an earlier version of this (as unittests/test-ggc.c): https://gcc.gnu.org/ml/gcc-patches/2015-10/msg03306.html > Not terribly happy with that counter to used to create a big list > to detect recursion. But I'm not offhand sure how to avoid without > exposing more of the ggc system

[PATCH] Add selftest for pretty-print.c

This adds another set of test cases to -fself-test, this time for the basic functionality within pretty-print.c. Successfully bootstrapped on x86_64-pc-linux-gnu. OK for trunk? gcc/ChangeLog: * pretty-print.c: Include "selftest.h". (pp_format): Fix comment.

[PATCH] spellcheck.c: add test_find_closest_string

This adds another test case to -fself-test. Successfully bootstrapped on x86_64-pc-linux-gnu. OK for trunk? gcc/ChangeLog: * spellcheck.c (selftest::test_find_closest_string): New function. (spellcheck_c_tests): Call the above. --- gcc/spellcheck.c | 23 +++

Re: [Diagnostic Patch] Clean-up diagnostic facilities in diagnostic.c

On Mon, 2016-06-06 at 10:55 +0200, Paolo Carlini wrote: > Hi, > > yesterday I had the idea of this small clean-up: move the work done > by > emit_diagnostic to a new non-variadic diagnostic_impl and use it to > implement the former and all the various inform, warning, permerror, > etc >

Re: [PATCH] integer overflow checking builtins in constant expressions

Hi! On Mon, Jun 06, 2016 at 02:36:17PM +0200, Jakub Jelinek wrote: > 2016-06-06 Martin Sebor > Jakub Jelinek > > PR c++/70507 > PR c/68120 > * builtins.def (BUILT_IN_ADD_OVERFLOW_P, BUILT_IN_SUB_OVERFLOW_P, >

Re: [C++ PATCH] Avoid exponential compile time in cp_fold_function (PR c++/70847, PR c++/71330, PR c++/71393)

OK. Jason

Re: [PING] [PATCH] Fix asm X constraint (PR inline-asm/59155)

On Mon, Jun 06, 2016 at 09:27:56PM +0200, Marc Glisse wrote: > The last one would miss floating point registers (no 2 platforms use the > same letter for those, hence my quest for something more generic). > > The goal of the experiment is described in PR59159 (for which "+X" is > unlikely to be

Re: [PING] [PATCH] Fix asm X constraint (PR inline-asm/59155)

On Mon, 6 Jun 2016, Jakub Jelinek wrote: On Mon, Jun 06, 2016 at 12:04:04PM -0600, Jeff Law wrote: On 06/06/2016 12:01 PM, Jakub Jelinek wrote: On Mon, Jun 06, 2016 at 11:54:04AM -0600, Jeff Law wrote: As for recog.c, I can not approve this as I am not a maintainer of it. I only can say that

Re: [PATCH, i386] Add native support for VIA C7, Eden and Nano CPUs

On Mon, 2016-06-06 at 17:27 +, Joseph Myers wrote: > This patch is missing the invoke.texi changes to document all the new > CPU  > names. Hi, correct, please consider adding the following patch to fix this. Regards. --- diff --git a/gcc/doc/invoke.texi b/gcc/doc/invoke.texi index

[PATCH] Support i386 TLS code sequences without PLT

On Mon, Jun 6, 2016 at 11:04 AM, H.J. Lu wrote: > On Mon, Jun 6, 2016 at 8:01 AM, Carlos O'Donell wrote: >> On 06/03/2016 05:21 PM, H.J. Lu wrote: >>> We can generate x86-64 TLS code sequences for general and local dynamic >>> models without PLT, which

Re: [PING] [PATCH] Fix asm X constraint (PR inline-asm/59155)

On Mon, Jun 06, 2016 at 12:04:04PM -0600, Jeff Law wrote: > On 06/06/2016 12:01 PM, Jakub Jelinek wrote: > >On Mon, Jun 06, 2016 at 11:54:04AM -0600, Jeff Law wrote: > >>>As for recog.c, I can not approve this as I am not a maintainer of it. > >>>I only can say that the code looks questionable to

Re: [PING] [PATCH] Fix asm X constraint (PR inline-asm/59155)

On 06/06/2016 12:01 PM, Jakub Jelinek wrote: On Mon, Jun 06, 2016 at 11:54:04AM -0600, Jeff Law wrote: As for recog.c, I can not approve this as I am not a maintainer of it. I only can say that the code looks questionable to me. I think the question on the recog part is a matter of how we

Re: [PING] [PATCH] Fix asm X constraint (PR inline-asm/59155)

On Mon, Jun 06, 2016 at 11:54:04AM -0600, Jeff Law wrote: > >As for recog.c, I can not approve this as I am not a maintainer of it. > >I only can say that the code looks questionable to me. > I think the question on the recog part is a matter of how we choose to > interpret what the "X" constraint

Re: [PATCH][vectorizer] Remove blank debug lines after dump_gimple_stmt

On 06/06/2016 06:46 AM, Alan Hayward wrote: Lots of code calls dump_gimple_stmt then print a newline, however dump_gimple_stmt will print a newline itself. This makes the vectorizer debug file messy. I think the confusion is because dump_generic_expr does NOT print a newline. This patch removes

Re: [PING] [PATCH] Fix asm X constraint (PR inline-asm/59155)

On 06/06/2016 11:04 AM, Vladimir Makarov wrote: On 06/06/2016 09:32 AM, Bernd Edlinger wrote: Ping... see https://gcc.gnu.org/ml/gcc-patches/2016-05/msg02010.html Thank you for working on the PR and sorry for the delay with LRA part of review. Change in lra-constraints.c is ok for me with

Re: [PATCH] Fix SLP wrong-code with VECTOR_BOOLEAN_TYPE_P (PR tree-optimization/71259)

On Mon, Jun 06, 2016 at 10:05:57AM +0200, Richard Biener wrote: > So this ends up generating { a ? -1 : 0, b ? -1 : 0, ... }. That Yes, that is already what we do now for loop vectorization. > might be less optimal than doing { a, b, ... } ? { -1, -1 ... } : { 0, 0, > .. } Well, it would need

Re: [PATCH, i386] Add native support for VIA C7, Eden and Nano CPUs

This patch is missing the invoke.texi changes to document all the new CPU names. -- Joseph S. Myers jos...@codesourcery.com

[Committed] Selftest framework (v8)

On Mon, 2016-06-06 at 16:40 +0200, Bernd Schmidt wrote: > On 06/06/2016 04:17 PM, David Malcolm wrote: > > I'm testing a revised patch now, incorporating the above, and > > renaming > > s-selftests (plural) to s-selftest (singular) etc within > > gcc/Makefile.in as requested by Bernhard elsewhere

Re: [PATCH][RTL ifcvt] Print name of noce trasform that succeeded in dump file

On 06/06/2016 06:28 PM, Kyrill Tkachov wrote: This patch adds the name of the transform that succeeded in if-conversion and prints it to the dump file so that we can pinpoint the extact noce_try* function that triggered. Ok. Bernd

Re: [PING] [PATCH] Fix asm X constraint (PR inline-asm/59155)

On 06/06/2016 09:32 AM, Bernd Edlinger wrote: Ping... see https://gcc.gnu.org/ml/gcc-patches/2016-05/msg02010.html Thank you for working on the PR and sorry for the delay with LRA part of review. Change in lra-constraints.c is ok for me with the following change. Instead of just -

Re: Unreviewed patches

On Mon, 6 Jun 2016, Rainer Orth wrote: > The following patches have remained unreviewed for a week: > > [gotools, libcc1] Update copyright dates > https://gcc.gnu.org/ml/gcc-patches/2016-05/msg02307.html > > Richard already approved the update-copyright.py changes, but the actual >

[PATCH v2] Update documentation for ARM architecture

* use lexicographical ordering, as "gcc -march=foo" does * correct usage of @samp vs @option, add @samp where appropriate * add armv6k, armv6z, arm6zk -march option values * remove -march=ep9312, it is only valid for -mcpu * add armv6s-m and document it, as it is no official ARM name.

[PATCH][RTL ifcvt] Print name of noce trasform that succeeded in dump file

Hi all, When debugging the noce if-conversion passes one of the most frustrating and time-consuming things I have to do is find which of the dozen or so transforms triggered. You'd think going through the cascade of if-gotos in noce_process_if_block in gdb would work, but this tends to be

Re: [PATCH][AArch64] Add missing fcsel in Cortex-A57 scheduler

On 02/06/16 17:09, Wilco Dijkstra wrote: The Cortex-A57 scheduler is missing fcsel, so add it. OK for commit? ChangeLog: 2016-06-02 Wilco Dijkstra * config/arm/cortex-a57.md (cortex_a57_fp_cpys): Add fcsel. Ok from an arm perspective too. Thanks, Kyrill ---

[PATCH][ARM] Add initial support for Cortex-A73

Hi all, This patch adds initial support for the Cortex-A73 processor through the cortex-a73, cortex-a73.cortex-a35 and cortex-a73.cortex-a53 arguments to -mcpu and -mtune. The Cortex-A73 is an ARMv8-A processor. Bootstrapped and tested on arm-none-linux-gnueabihf with an appropriately patched

[PATCH][AArch64] Add initial support for Cortex-A73

Hi all, This patch adds initial support for the Cortex-A73 processor through the cortex-a73, cortex-a73.cortex-a35 and cortex-a73.cortex-a53 arguments to -mcpu and -mtune. The Cortex-A73 is an ARMv8-A processor and the initial tuning is based on the Cortex-A57 tuning (though not an exact

Re: [v3 PATCH] Support allocators in tuples of zero size.

On 05/06/16 21:15 +0300, Ville Voutilainen wrote: { public: void swap(tuple&) noexcept { /* no-op */ } + // We need the default since we're going to define no-op + // allocator constructors. + tuple() = default; + // No-op allocator constructors. + template

[PATCH][AArch64] Model CSEL instruction in Cortex-A57 scheduling model

Hi all, This small patch adds handling of the CSEL-type instructions to the Cortex-A57 scheduling model. It is treated the same as simple ALU instructions. With this patch I didn't see any overall differences in SPEC2006. Bootstrapped and tested on arm-none-linux-gnueabihf and

[PATCH GCC]Remove duplciated alias check in vectorizer

Hi, GCC now generates duplicated alias check in vectorizer when versioning loops. In current implementation, DR_OFFSET and DR_INIT are added together too early when creating structure dr_with_seg_len. This has two disadvantages: A) structure dr_with_seg_len_pair_t is only canonicalized

[PATCH] libstdc++/71320 Add or remove file permissions correctly

This adds the missing functionality to filesystem::permissions(). PR libstdc++/71320 * src/filesystem/ops.cc (permissions(const path&, perms, error_code&)): Add or remove permissions according to perms argument. *

Re: C PATCH for comptypes handling of TYPE_REF_CAN_ALIAS_ALL

On Mon, 6 Jun 2016, Marek Polacek wrote: > > I don't see how this test is supposed to verify properties of the > > composite type. I'd expect you to need to verify that something does not > > get optimized away, that would get optimized away in the absence of > > may_alias. > > Well, were it

Re: [PATCH v2] gcov: Runtime configurable destination output

I applied this patch. Aaron's patch, AFAICT, would repeatedly fopen the error file. nathan 2016-06-05 Aaron Conole Nathan Sidwell PR libgcc/71400 * libgcov-driver-system.c (__gcov_error_file): Disable if IN_GCOV_TOOL. (get_gcov_error_file):

Re: C PATCH for comptypes handling of TYPE_REF_CAN_ALIAS_ALL

On Mon, Jun 06, 2016 at 03:06:18PM +, Joseph Myers wrote: > On Tue, 31 May 2016, Marek Polacek wrote: > > > > diff --git gcc/testsuite/c-c++-common/attr-may-alias-1.c > > > gcc/testsuite/c-c++-common/attr-may-alias-1.c > > > index e69de29..978b9a5 100644 > > > ---

Re: [PATCH 4/4] C: add fixit hint to misspelled field names

On Tue, 31 May 2016, David Malcolm wrote: > Ping: > https://gcc.gnu.org/ml/gcc-patches/2016-04/msg01834.html OK. What about field names in designated initializers (both C99-style and old-style)? -- Joseph S. Myers jos...@codesourcery.com

Re: [PATCH] Support x86-64 TLS code sequences without PLT

On 06/03/2016 05:21 PM, H.J. Lu wrote: > We can generate x86-64 TLS code sequences for general and local dynamic > models without PLT, which uses indirect call via GOT: > > call *__tls_get_addr@GOTPCREL(%rip) > > instead of direct call: > > call __tls_get_addr[@PLT] What are the actual pros

Re: C PATCH for comptypes handling of TYPE_REF_CAN_ALIAS_ALL

On Tue, 31 May 2016, Marek Polacek wrote: > > diff --git gcc/testsuite/c-c++-common/attr-may-alias-1.c > > gcc/testsuite/c-c++-common/attr-may-alias-1.c > > index e69de29..978b9a5 100644 > > --- gcc/testsuite/c-c++-common/attr-may-alias-1.c > > +++ gcc/testsuite/c-c++-common/attr-may-alias-1.c >

Re: [PATCH][1/3][ARM] Keep ctz expressions together until after reload

On Thu, May 26, 2016 at 10:52 AM, Kyrill Tkachov wrote: > Hi all, > > On arm we don't have a dedicated instruction that corresponds to a CTZ rtx > but we synthesise it > with an RBIT instruction followed by a CLZ. This is currently done at expand > time. > However,

Re: [PATCH] Selftest framework (v7)

On 06/06/2016 04:17 PM, David Malcolm wrote: I'm testing a revised patch now, incorporating the above, and renaming s-selftests (plural) to s-selftest (singular) etc within gcc/Makefile.in as requested by Bernhard elsewhere in this thread. I assume that change is OK? Sure. Bernd

Re: [PATCH][1/3][ARM] Keep ctz expressions together until after reload

Ping. https://gcc.gnu.org/ml/gcc-patches/2016-05/msg02078.html Patches 1 and 3 have been approved. Thanks, Kyrill On 26/05/16 10:52, Kyrill Tkachov wrote: Hi all, On arm we don't have a dedicated instruction that corresponds to a CTZ rtx but we synthesise it with an RBIT instruction followed

Re: [PATCH][3/3][RTL ifcvt] PR middle-end/37780: Conditional expression with __builtin_clz() should be optimized out

On 05/26/2016 11:53 AM, Kyrill Tkachov wrote: 2016-05-26 Kyrylo Tkachov PR middle-end/37780 * ifcvt.c (noce_try_ifelse_collapse): New function. Declare prototype. (noce_process_if_block): Call noce_try_ifelse_collapse. * simplify-rtx.c

Re: [PATCH][3/3][RTL ifcvt] PR middle-end/37780: Conditional expression with __builtin_clz() should be optimized out

Ping. https://gcc.gnu.org/ml/gcc-patches/2016-05/msg02080.html Thanks, Kyrill On 26/05/16 10:53, Kyrill Tkachov wrote: Hi all, In this PR we want to optimise: int foo (int i) { return (i == 0) ? N : __builtin_clz (i); } on targets where CLZ is defined at zero to the constant 'N'. This is

Re: [PATCH] Selftest framework (v7)

On Sun, 2016-06-05 at 13:37 +0200, Bernd Schmidt wrote: > On 06/03/2016 09:12 PM, David Malcolm wrote: > > It's not clear to me if these approvals still hold. > > I was willing to go with it; I had a look through some of these > patches > and didn't spot anything untoward. To make it clear, this

[v2][AArch64, 5/6] Reimplement fabd intrinsics & merge rtl patterns

These intrinsics were implemented before "fabd_3" introduces. Meanwhile the patterns "fabd_3" and "*fabd_scalar3" can be merged into a single "fabd3" using VALLF. This patch migrate the implementation to builtins backed by this pattern. gcc/ 2016-06-01 Jiong Wang

[v2][AArch64, 6/6] Reimplement vpadd intrinsics & extend rtl patterns to all modes

These intrinsics was implemented by inline assembly using "faddp" instruction. There was a pattern "aarch64_addpv4sf" which supportsV4SF mode only while we can extend this pattern to support VDQF mode, then we can reimplement these intrinsics through builtlins. gcc/ 2016-06-06 Jiong

[v2][AArch64, 4/6] Reimplement frsqrts intrinsics

Similar as [3/6], these intrinsics were implemented before the instruction pattern "aarch64_rsqrts" added, that these intrinsics were implemented through inline assembly. This mirgrate the implementation to builtin. gcc/ 2016-06-06 Jiong Wang *

[v2][AArch64, 3/6] Reimplement frsqrte intrinsics

These intrinsics were implemented before the instruction pattern "aarch64_rsqrte" added, that these intrinsics were implemented through inline assembly. This mirgrate the implementation to builtin. gcc/ 2016-06-06 Jiong Wang * config/aarch64/aarch64-builtins.def

[v2][AArch64, 2/6] Reimplement vector fixed-point intrinsics

Based on top of [1/6], this patch reimplement vector intrinsics for conversion between floating-point and fixed-point. gcc/ 2016-06-06 Jiong Wang * config/aarch64/aarch64-builtins.def (scvtf): Register vector modes. (ucvtf): Likewise. (fcvtzs):

[v2][AArch64, 1/6] Reimplement scalar fixed-point intrinsics

On 27/05/16 17:52, Jiong Wang wrote: On 27/05/16 14:03, James Greenhalgh wrote: On Tue, May 24, 2016 at 09:23:36AM +0100, Jiong Wang wrote: * config/aarch64/aarch64-simd-builtins.def: Rename to aarch64-builtins.def. Why? We already have some number of intrinsics in here

[PING] [PATCH] Fix asm X constraint (PR inline-asm/59155)

Ping... see https://gcc.gnu.org/ml/gcc-patches/2016-05/msg02010.html Thanks Bernd. On 05/25/16 14:58, Bernd Edlinger wrote: > Hi! > > This restricts the X constraint in asm statements, which > can be easily folded by combine in something completely > invalid. > > It is necessary to allow

Re: loop-ch tweek

Hi, does this look better? Honza * gimple.c: Include builtins.h (gimple_inexpensive_call_p): New function. * gimple.h (gimple_inexpensive_call_p): Declare. * tree-ssa-loop-ch.c (should_duplicate_loop_header_p): Use it. * tree-ssa-loop-ivcanon.c

[PATCH][vectorizer] Remove blank debug lines after dump_gimple_stmt

Lots of code calls dump_gimple_stmt then print a newline, however dump_gimple_stmt will print a newline itself. This makes the vectorizer debug file messy. I think the confusion is because dump_generic_expr does NOT print a newline. This patch removes all prints of a newline direcly after a

Re: [PATCH] integer overflow checking builtins in constant expressions

On Fri, Jun 03, 2016 at 02:09:44PM -0600, Martin Sebor wrote: > I see. I've made the change in the latest update to the patch > but I wasn't able to create a test case to verify it. Maybe > that's because this is constexpr the COMPLEX_EXPR doesn't make > it far enough to trigger a problem. If

Re: loop-ch tweek

> On Mon, 6 Jun 2016, Jan Hubicka wrote: > > > > On Sun, 5 Jun 2016, Jan Hubicka wrote: > > > > > > > Hi, > > > > both loop-ch and loop-ivcanon want to trottle down the heuristics on > > > > paths > > > > containing call. Testing for presence of GIMPLE_CALL is wrong for > > > > internal > > >

Re: [middle-end][PATCH] Update alignment_for_piecewise_move

On 06/02/2016 03:37 PM, H.J. Lu wrote: Are you planning to submit your patch before July? If not, I will resubmit mine and work out all the issues. It may take a long time to review and I have patches to enable SSE, AVX, AVX512 f memset and memcpy, which depend on it. I'd like to see them

Re: [PATCH v1] Support for SPARC M7 and VIS 4.0

> This patch adds support for -mcpu=niagara7, corresponding to the SPARC > M7 CPU as documented in the Oracle SPARC Architecture 2015 and the M7 > Processor Supplement. The patch also includes intrinsics support for > all the VIS 4.0 instructions. > > This patch has been

Re: [PATCH PR68030/PR69710][RFC]Introduce a simple local CSE interface and use it in vectorizer

On Fri, May 27, 2016 at 12:56 PM, Richard Biener wrote: > On Fri, May 27, 2016 at 1:11 PM, Bin.Cheng wrote: >> On Fri, May 27, 2016 at 11:45 AM, Richard Biener >> wrote: >>> On Wed, May 25, 2016 at 1:22 PM, Bin Cheng

Re: Remove word_mode hack for split bitfields

On 05/26/2016 04:36 PM, Richard Sandiford wrote: This patch is effectively reverting a change from 1994. The reason I think it's a hack is that store_bit_field_1 is creating a subreg reference to one word of a field even though it has already proven that the field spills into the following

Re: loop-ch tweek

On Mon, 6 Jun 2016, Jan Hubicka wrote: > > On Sun, 5 Jun 2016, Jan Hubicka wrote: > > > > > Hi, > > > both loop-ch and loop-ivcanon want to trottle down the heuristics on paths > > > containing call. Testing for presence of GIMPLE_CALL is wrong for internal > > > call and cheap builtins that are

backward threading heuristics tweek

Hi, while looking into profile mismatches introduced by the backward threading pass I noticed that the heuristics seems quite simplistics. First it should be profile sensitive and disallow duplication when optimizing cold paths. Second it should use estimate_num_insns because gimple statement

Re: tuple code simplification patch

On 02/06/16 23:00 +0200, François Dumont wrote: Hi I was trying to play with tuple implementation and was annoyed by repetition of _Head type when instantiating _Head_base so I thought about this patch. How do you like it ? I still need to run tests, will do before commit, ok ?

Re: [patch] doc/sourcebuild.texi (Directives): Remove extra closing braces.

On 29/05/16 23:05 +0200, Gerald Pfeifer wrote: On Sat, 16 Jan 2016, Jonathan Wakely wrote: This removes stray closing braces in the docs for dg-error, dg-warning etc. OK for trunk? Yes. Sorry for the delay. I expected someone else to pick this up for review/approval, but now noticed that

Re: loop-ch tweek

> On Sun, 5 Jun 2016, Jan Hubicka wrote: > > > Hi, > > both loop-ch and loop-ivcanon want to trottle down the heuristics on paths > > containing call. Testing for presence of GIMPLE_CALL is wrong for internal > > call and cheap builtins that are expanded inline. > > > > Bootstrapped/regtested

[Ada] Make Gigi_Equivalent_Type idempotent

This fixes a small regression in ASIS mode introduced by the new elaboration model for subprograms. Tested on x86_64-suse-linux, applied on the mainline. 2016-06-06 Eric Botcazou * gcc-interface/decl.c (Gigi_Equivalent_Type): Make sure equivalent

Re: [PING^2] Re: Updated autofdo bootstrap and testing patches

On 05/30/2016 12:17 AM, Andi Kleen wrote: Andi Kleen writes: Ping^2! I think patches #1 and #5 had unaddressed comments from Jan. I think you should ping build system maintainers directly for #4. Bernd

[Ada] Add support for noinline and noclone attributes

They are automatically set by the middle-end when the interrupt attribute is. Tested on x86_64-suse-linux, applied on the mainline and 6 branch. 2016-06-06 Eric Botcazou * gcc-interface/utils.c (gnat_internal_attribute_table): Add support for noinline

[Ada] Fix double elaboration of qualified expression in allocator

This is a regression present on the mainline and 6 branch: under some peculiar circumstances, the qualified expression of an allocator can be elaborated twice by gigi. Tested on x86_64-suse-linux, applied on the mainline and 6 branch. 2016-06-06 Eric Botcazou

[Ada] Fix single-stepping regression with tagged types

This ensures that we don't import an artificial location from the spec when an object of a tagged type is elaborated. Tested on x86_64-suse-linux, applied on the mainline. 2016-06-06 Eric Botcazou * gcc-interface/trans.c (gnat_to_gnu): Rework special code

[Ada] Small adjustments in handling of subprograms

No functional change. Tested on x86_64-suse-linux, applied on the mainline. 2016-06-06 Eric Botcazou * gcc-interface/decl.c (gnat_to_gnu_entity) : Remove useless 'else' statements and tidy up. : Fully deal with the declaration here. :

Re: [PING] [PATCH] Make basic asm implicitly clobber memory

On 06/05/2016 06:02 PM, Bernd Edlinger wrote: I think we all agreed on the general direction of this patch. The patch is basically unchanged from previous version, except one line in doc/extend.texi has been updated. So I would like to ask if it is OK for trunk. Are there any users of

[Ada] Remove deferred freezing of subprograms in Freeze_Entity

This removes the specific mechanism present in Freeze_Entity to defer the freezing of functions returning an incomplete type coming from a limited context. It was invented to cope with the old elaboration model for subprograms in gigi, which didn't really implement AI05-151 and AI05-019. The new

[Ada] Follow-up work for AI05-0151 (incomplete types in profiles)

This is a follow up to the initial implementation done in https://gcc.gnu.org/ml/gcc-patches/2016-04/msg01710.html It completes the transition to the new elaboration model, makes it a bit more efficient by reusing already elaborated entities as much as possibke and fixes an issue with

Re: [PATCH][2/3] Vectorize inductions that are live after the loop

On Mon, Jun 06, 2016 at 10:03:12AM +0100, Alan Hayward wrote: > > On 03/06/2016 18:45, "Jakub Jelinek" wrote: > > >On Thu, Jun 02, 2016 at 05:11:15PM +0100, Alan Hayward wrote: > >>* gcc.dg/vect/vect-live-1.c: New test. > >>* gcc.dg/vect/vect-live-2.c: New test. > >>

Re: [PATCH][2/3] Vectorize inductions that are live after the loop

On 03/06/2016 18:45, "Jakub Jelinek" wrote: >On Thu, Jun 02, 2016 at 05:11:15PM +0100, Alan Hayward wrote: >> * gcc.dg/vect/vect-live-1.c: New test. >> * gcc.dg/vect/vect-live-2.c: New test. >> * gcc.dg/vect/vect-live-5.c: New test. >> *

Re: [PR71408] - Fix wrong code at -Os and above

On Sun, Jun 5, 2016 at 12:54 PM, kugan wrote: > Hi All, > > For the testcase in PR71408 zero_one_operation seems still broken. In > handling NEGATE_EXPR, as part of undistribute_ops_list, in > zero_one_operation, we are doing propagate_op_to_single_use (op,

[Diagnostic Patch] Clean-up diagnostic facilities in diagnostic.c

Hi, yesterday I had the idea of this small clean-up: move the work done by emit_diagnostic to a new non-variadic diagnostic_impl and use it to implement the former and all the various inform, warning, permerror, etc (lately we have the *_at_rich_loc variants too). Something similar can be

Re: [PR71281] ICE on gcc trunk on knl, wsm, ivb and bdw targets

On Sat, Jun 4, 2016 at 4:25 AM, kugan wrote: > Hi, > > PR71281 happens when we use factored out negate stmt in other > reassociations. Since we don't set the uid for this stmt, we hit the > gcc_assert (in reassoc_stmt_dominates_stmt_p) which checks for uid being

[Ada] Fix minor glitches related to subprogram declarations

This patch fixes a few minor glitches that yield small irregularities in the expanded code handed down to gigi, for example the declaration of subprograms before that of the type of their parameters, types that are neither regular nor Itypes, or return types with circularities. No functional

Unreviewed patches

The following patches have remained unreviewed for a week: [gotools, libcc1] Update copyright dates https://gcc.gnu.org/ml/gcc-patches/2016-05/msg02307.html Richard already approved the update-copyright.py changes, but the actual effects on gotools and libcc1 require either

Re: [PING] [PATCH] Make basic asm implicitly clobber memory

On Sun, 5 Jun 2016, Bernd Edlinger wrote: > Ping... > > I think we all agreed on the general direction of this patch. > > The patch is basically unchanged from previous version, > except one line in doc/extend.texi has been updated. > > So I would like to ask if it is OK for trunk. The

Re: [PATCH] Fix SLP wrong-code with VECTOR_BOOLEAN_TYPE_P (PR tree-optimization/71259)

On Fri, 3 Jun 2016, Jakub Jelinek wrote: > On Tue, Jan 12, 2016 at 05:21:37PM +0300, Ilya Enkovich wrote: > > > --- gcc/tree-vect-slp.c.jj 2016-01-08 21:45:57.0 +0100 > > > +++ gcc/tree-vect-slp.c 2016-01-11 12:07:19.633366712 +0100 > > > @@ -2999,12 +2999,9 @@

Re: loop-ch tweek

On Sun, 5 Jun 2016, Jan Hubicka wrote: > Hi, > both loop-ch and loop-ivcanon want to trottle down the heuristics on paths > containing call. Testing for presence of GIMPLE_CALL is wrong for internal > call and cheap builtins that are expanded inline. > > Bootstrapped/regtested x86_64-linux, OK?

[PATCH] Fix PR71398

Bootstrapped and tested on x86_64-unknown-linux-gnu, applied. Richard. 2016-06-06 Richard Biener PR tree-optimization/71398 * tree-ssa-loop-ivcanon.c (unloop_loops): First unloop, then remove edges. * gcc.dg/torture/pr71398.c: New