Fix typo breaking bootstrap

Hi, while doing a last minute cleanup, I managed to introduce a typo into the patch, which effectively disables inlining. Sorry for that. Comitted as obvious, Honza PR bootstrap/83062 * ipa-inline.c (can_inline_edge_p): Fix typo in previous patch. Index: ipa-inline.c

Re: [RFA][PATCH] 8/n Pull evrp range analyzer into its own file

On Sat, Nov 18, 2017 at 9:00 AM, Jeff Law wrote: > > > > This is just a straight cut-n-paste pulling the evrp_range_analyzer > methods out of gimple-ssa-evrp.c into gimple-ssa-evrp-analyze.[ch]. > > Bootstrapped and regression tested on x86_64. > > > OK for the trunk? Hum.

Re: Fix PR81360

> I believe this caused hundreds of new FAILs on both x86_64 and i686-linux. > Seems after this change we no longer inline always_inline functions into > -O0 callers, which is wrong. That's worse than this, inlining is totally broken... -- Eric Botcazou

Re: [RFA][PATCH] 11/n More VRP related cleanups

On Sat, Nov 18, 2017 at 9:49 AM, Jeff Law wrote: > > Not as extensive as the last set. Just getting a couple data members > privatized as well as one method within vr_values. > > I'd really like to just get values_propagated out of vr_values and bury > it in tree-vrp.c, but

Re: [RFA][PATCH] 10/n Various class definition cleanups

On Sat, Nov 18, 2017 at 9:39 AM, Jeff Law wrote: > > A batch of random cleanups in the various vrp related bits. > > The goal here is to start banging some of the classes into shape. > Included in this patch: > > > For evrp_folder: > Accepts vr_values in its ctor to attach. >

Re: [RFA][PATCH] 9/n Finish moving BB_VISITED handling into range analyzer

On Sat, Nov 18, 2017 at 9:05 AM, Jeff Law wrote: > > I didn't want to rebase my patchset yet again to fix this minor goof. I > moved initialization of BB_VISITED into the range analyzer, but forgot > to move setting it as we visit each block into the analyzer. > > This fixes

Re: Fix PR81360

> > I believe this caused hundreds of new FAILs on both x86_64 and i686-linux. > > Seems after this change we no longer inline always_inline functions into > > -O0 callers, which is wrong. > > That's worse than this, inlining is totally broken... It was a typo in the conditional (I cleaned it up

Re: [RFC][PATCH 12/n Embed range analysis in DOM

On Sat, Nov 18, 2017 at 10:31 AM, Jeff Law wrote: > And a WIP. I can justify continuing work on this during stage3 for > pr78496. But I thought it was worth giving folks a looksie. > > The goal here is to make tree-vrp's threading obsolete and remove it and > at the same time

Re: [RFC][PATCH] Change default to -fcommon

On Sun, Nov 19, 2017 at 2:48 AM, Sandra Loosemore wrote: > On 11/17/2017 12:12 PM, Wilco Dijkstra wrote: >> >> GCC currently defaults to -fcommon. This is an optional C feature dating >> back to early C implementations. On many targets this means global >> variable >>

Re: [patch] Do not report non-executed blocks in Ada coverage

> Is this still needed? I know Martin L did another teak for non-executed > blocks in Ada... BUt I don't know enough to determine if it was > sufficient to address this problem. Yes, Martin's patch fixed the regression in Ada so mine is obsolete. -- Eric Botcazou

Re: [committed][PATCH] Fix bogus propagation in DOM

On Sun, Nov 19, 2017 at 9:16 PM, Jeff Law wrote: > On my local branch gcc.dg/torture/pr56349.c fails by sending GCC into an > infinite loop trying to simplify a self-referring statement. ie > something like > > x_1 = x_1 + 10; > > That, of course, shouldn't be happening in SSA

Re: VRP: x+1 and -x cannot be INT_MIN

On Sun, Nov 19, 2017 at 11:41 AM, Marc Glisse wrote: > On Mon, 13 Nov 2017, Richard Biener wrote: > >> On Sat, Nov 11, 2017 at 11:03 PM, Marc Glisse >> wrote: >>> >>> Hello, >>> >>> with undefined overflow, just because we know nothing about one of the

Re: Fix PR81360

> It was a typo in the conditional (I cleaned it up just before comitting > because I noticed I can merge it with earlier conditional and accidentally > dropped !). It should be fixed now. OK, thanks, unfortunately the bug made it into the latest source snapshot. -- Eric Botcazou

RE: [arc] Remove semicolon after do while (0) in FUNCTION_PROFILER

Thanks Tom, Claudiu From: Tom de Vries [tom_devr...@mentor.com] Sent: Sunday, November 19, 2017 1:27 PM To: GCC Patches Cc: Claudiu Zissulescu Subject: [arc] Remove semicolon after do while (0) in FUNCTION_PROFILER Hi, this patch removes a semicolon after

Re: [PATCH] Fix bug in simplify_ternary_operation

On 11/20/2017 03:52 AM, Jeff Law wrote: Sorry, it's taken so long to get back to this patch... On 09/01/2017 02:51 AM, Tom de Vries wrote: On 08/31/2017 11:44 PM, Jeff Law wrote: On 08/28/2017 12:26 PM, Tom de Vries wrote: Hi, I think I found a bug in r17465: ...     * cse.c

[PATCH] RTEMS: Enable some libstdc++ features

libstdc++/ * acinclude.m4 (GLIBCXX_ENABLE_LIBSTDCXX_TIME): Add rtems*. (GLIBCXX_ENABLE_FILESYSTEM_TS): Likewise. * configure: Regenerate. --- libstdc++-v3/acinclude.m4 | 5 - libstdc++-v3/configure| 5 - 2 files changed, 8 insertions(+), 2 deletions(-) diff

Re: [patch, fortran] Implement maxloc and minloc for character

On Sun, Nov 19, 2017 at 11:11 PM, Thomas Koenig wrote: > There is one question regarding the ABI. Apparently, the string length > is passed as an int even on a 64-bit system. I verified that this > is indeed the case by doing the actual work on a >

Re: [patch, libgfortran] Bug 78549 - [7/8 Regression] Very slow formatted internal file output

On Mon, Nov 20, 2017 at 12:56 AM, Jerry DeLisle wrote: > Hi all, > > After the usual considerable digging around I determined that for the test > case > doing 100 writes to a string that we were allocating 100 newunit > numbers > and the associated unit structures

Re: [PATCH] Use bswap framework in store-merging (PR tree-optimization/78821)

On Thu, 16 Nov 2017, Jakub Jelinek wrote: > Hi! > > This patch uses the bswap pass framework inside of the store merging > pass to handle adjacent stores which produce together a 16/32/64 bit > store of bswapped value (loaded or from SSA_NAME) or identity (usually > only from SSA_NAME, the code

Re: Make ivopts handle calls to internal functions

On Fri, Nov 17, 2017 at 3:03 PM, Richard Sandiford wrote: > ivopts previously treated pointer arguments to internal functions > like IFN_MASK_LOAD and IFN_MASK_STORE as normal gimple values. > This patch makes it treat them as addresses instead. This makes > a

Re: Add support for in-order addition reduction using SVE FADDA

On Fri, Nov 17, 2017 at 5:53 PM, Richard Sandiford wrote: > This patch adds support for in-order floating-point addition reductions, > which are suitable even in strict IEEE mode. > > Previously vect_is_simple_reduction would reject any cases that forbid >

Re: [patch] Add support for #pragma GCC unroll

> Looking at the middle-end changes. The change to ANNOTATE_EXPR to three > operands is approved also for existing frontends (just in case you > don't get review). > Found the missing possibility of an argument limiting myself... I see. ;-) > Your changes to the RTL unrolling pass as far as I

Re: PING Fwd: [patch] implement generic debug() for vectors and hash sets

Minor oversight... debug_vec_tree() no longer exist. I forgot to remove the prototype. Also, gdbinit.in has a macro that uses it, but this is no longer necessary as we can print tree vectors generically with "print debug(xxx)". Applied as obvious. Aldy On Sun, Nov 19, 2017 at 5:45 PM, Jeff

Re: [PATCH GCC]Support load in CT_STORE_STORE chain if dominated by store in the same loop iteration

On Fri, Nov 17, 2017 at 2:49 PM, Bin Cheng wrote: > Hi, > I previously introduced CT_STORE_STORE chains in predcom. This patch further > supports load > reference in CT_STORE_STORE chain if the load is dominated by a store > reference in the same > loop iteration. So

Re: [patch] Add support for #pragma GCC unroll

On 20 November 2017 at 12:26, Eric Botcazou wrote: >> If anybody finds the time to push the corresponding Fortran changes then I'd >> be grateful. I won't have time for this until end of stage 1... >> >> https://gcc.gnu.org/ml/fortran/2015-02/msg00014.html > > OK, I'm going

Re: Adjust empty class parameter passing ABI (PR c++/60336)

On Thu, Nov 16, 2017 at 02:20:59PM -0500, Jason Merrill wrote: > On Thu, Nov 16, 2017 at 12:41 PM, Marek Polacek wrote: > > On Tue, Nov 14, 2017 at 07:34:54AM +0100, Richard Biener wrote: > >> On November 14, 2017 6:21:41 AM GMT+01:00, Jason Merrill > >>

Re: [PATCH libstdc++/66689] comp_ellint_3 and ellint_3 return garbage values

On 18/11/17 11:49 -0500, Ed Smith-Rowland wrote: On 11/17/2017 03:54 PM, Jonathan Wakely wrote: Hmm, you're probably right. I'd be tempted to though. I had an idea.  What about a macro _GLIBCXX_ELLINT_3_POS_NU or something that: 1. would allow users to detect which convention is on by

Re: [RFA][PATCH] 8/n Pull evrp range analyzer into its own file

On November 20, 2017 4:46:08 PM GMT+01:00, Jeff Law wrote: >On 11/20/2017 03:05 AM, Richard Biener wrote: >> On Sat, Nov 18, 2017 at 9:00 AM, Jeff Law wrote: >>> >>> >>> >>> This is just a straight cut-n-paste pulling the evrp_range_analyzer >>> methods out of

Re: [Patch][aarch64] Use IFUNCs to enable LSE instructions in libatomic on aarch64

On Mon, Nov 20, 2017 at 05:39:25PM +, Steve Ellcey wrote: > Re-ping with a CC to the Aarch64 maintainers. If I'm completely honest with myself, I don't know enough about this area to review the patch. Szabolcs' OK holds a lot of weight with me, but I'd like to understand more of the

Re: [RFC][PATCH] Change default to -fcommon

Richard Biener wrote: > A target specific default might be a good idea if we decide to revert. > > Note I proposed this change a few times already, but the fear was always > we'll break too much legacy code. It will definitely break some code, but new warnings with -Werror might too... > Note

Re: [RFA][PATCH] 8/n Pull evrp range analyzer into its own file

On 11/20/2017 03:05 AM, Richard Biener wrote: > On Sat, Nov 18, 2017 at 9:00 AM, Jeff Law wrote: >> >> >> >> This is just a straight cut-n-paste pulling the evrp_range_analyzer >> methods out of gimple-ssa-evrp.c into gimple-ssa-evrp-analyze.[ch]. >> >> Bootstrapped and

Re: [PATCH] Add _Float/_FloatX rounding built-ins & improve gimple optimization of _Float/_FloatX built-in functions

On Mon, Nov 20, 2017 at 11:32:08AM -0600, Segher Boessenkool wrote: > Hi! > > On Fri, Nov 17, 2017 at 07:35:05PM -0500, Michael Meissner wrote: > > Here is the fixed patch. It fixes the btrunc2 insn to use the correct > > XSRPQI variant for truncf128. I added the float128-hw11.c test as a

Re: [patch, fortran] Implement maxloc and minloc for character

Am 20.11.2017 um 09:30 schrieb Janne Blomqvist: On Sun, Nov 19, 2017 at 11:11 PM, Thomas Koenig wrote: There is one question regarding the ABI. Apparently, the string length is passed as an int even on a 64-bit system. I verified that this is indeed the case by doing the

Re: [PATCH libstdc++/66689] comp_ellint_3 and ellint_3 return garbage values

This patch has introduced new regressions on at least PowerPC and AArch64. FAIL: ext/special_functions/hyperg/check_value.cc execution test FAIL: tr1/5_numerical_facilities/special_functions/17_hyperg/check_value.cc execution test Thanks, David

Re: std::advance istreambuf_iterator overload

On 20/11/17 07:40 +0100, François Dumont wrote: Here is the latest version I plan to commit after validation. I like the simplified logic for the while-loop. I'd like to keep the change to the money_get/get/[char/wchar_t]/9.cc tests. At the moment those tests passes only because there is no

Re: [RFC][PATCH] Change default to -fcommon

Hi, On Mon, 20 Nov 2017, Wilco Dijkstra wrote: > > Note you have to make sure GFortran still works! So I think the patch > > should be changed to make the default behavior be frontend dependent > > or have a fortran/ adjustment that fixes things up for the fortran > > dialects that need it.

Re: [RFC][PATCH 12/n Embed range analysis in DOM

On 11/20/2017 03:13 AM, Richard Biener wrote: > On Sat, Nov 18, 2017 at 10:31 AM, Jeff Law wrote: >> And a WIP. I can justify continuing work on this during stage3 for >> pr78496. But I thought it was worth giving folks a looksie. >> >> The goal here is to make tree-vrp's

Re: [Patch][aarch64] Use IFUNCs to enable LSE instructions in libatomic on aarch64

Re-ping with a CC to the Aarch64 maintainers. Steve Ellcey sell...@cavium.com On Tue, 2017-10-24 at 11:11 -0700, Steve Ellcey wrote: > Ping. > > Steve Ellcey > sell...@cavium.com > > On Tue, 2017-10-03 at 11:57 -0700, Steve Ellcey wrote: > > > > On Mon, 2017-10-02 at 15:38 +0100, Szabolcs

Re: Adjust empty class parameter passing ABI (PR c++/60336)

On November 20, 2017 4:51:04 PM GMT+01:00, Marek Polacek wrote: >On Thu, Nov 16, 2017 at 02:20:59PM -0500, Jason Merrill wrote: >> On Thu, Nov 16, 2017 at 12:41 PM, Marek Polacek >wrote: >> > On Tue, Nov 14, 2017 at 07:34:54AM +0100, Richard Biener wrote:

Re: Patch ping

On 11/20/2017 02:55 AM, Jakub Jelinek wrote: http://gcc.gnu.org/ml/gcc-patches/2017-11/msg00851.html C++2A P0428R2 - familiar template syntax for generic lambdas Are there existing testcases checking this is permitted w/o warning under the appropriate circumstances? That seems to be

Re: [libstdc++] Expose Airy functions.

On 18/11/17 20:33 -0500, Ed Smith-Rowland wrote: Here is the final patch fir libstdc++ Airy functions... OK for trunk, thanks.

Re: [PATCH] Add _Float/_FloatX rounding built-ins & improve gimple optimization of _Float/_FloatX built-in functions

Hi! On Fri, Nov 17, 2017 at 07:35:05PM -0500, Michael Meissner wrote: > Here is the fixed patch. It fixes the btrunc2 insn to use the correct > XSRPQI variant for truncf128. I added the float128-hw11.c test as a runtime > test to make sure round, trunc, ceil, and floor return the correct

Re: [committed][PATCH] Fix bogus propagation in DOM

On 11/20/2017 03:25 AM, Richard Biener wrote: > On Sun, Nov 19, 2017 at 9:16 PM, Jeff Law wrote: >> On my local branch gcc.dg/torture/pr56349.c fails by sending GCC into an >> infinite loop trying to simplify a self-referring statement. ie >> something like >> >> x_1 = x_1 + 10;

RFC: SAFE_MACRO_STMT

Hi, when writing macros that are intended to be used as if they were single statements, there are a couple of common problems: 1. Ending in semicolon ... #define foo stmt; ... This works fine if we use it like: ... void bar (void) { foo; } ... but we get an "else without a previous if"

Re: Patch ping

On 11/20/2017 02:55 AM, Jakub Jelinek wrote: Hi! http://gcc.gnu.org/ml/gcc-patches/2017-11/msg01026.html C++2A P0329R4: Designated Initialization OK, thanks -- Nathan Sidwell

Re: [PATCH libstdc++/66689] comp_ellint_3 and ellint_3 return garbage values

On 20/11/17 21:07 +, Jonathan Wakely wrote: On 20/11/17 21:01 +, Jonathan Wakely wrote: On 20/11/17 21:43 +0100, Christophe Lyon wrote: On 20 November 2017 at 17:02, David Edelsohn wrote: This patch has introduced new regressions on at least PowerPC and AArch64.

Re: [PATCH][aarch64] Fix pr81356 - copy empty string with wrz, not a ldrb/strb

On Mon, 2017-11-20 at 15:24 +0100, Christophe Lyon wrote: >  > As a result of this patch, we now have: > XPASS: gfortran.dg/pr45636.f90   -O   scan-tree-dump-times forwprop2 > "memset" 0 > instead of: > XFAIL: gfortran.dg/pr45636.f90   -O   scan-tree-dump-times forwprop2 > "memset" 0 (found 2

Re: [PATCH, rs6000] Testcase updates for power9 codegen

Hi! On Fri, Nov 17, 2017 at 11:30:27AM -0600, Will Schmidt wrote: > [testsuite] > * fold-vec-neg-int.c: Specify -mcpu=power8 Dot at end of sentence (quite a few times), please. > -/* { dg-final { scan-assembler-times "vspltisw|vxor" 1 } } */ > +//# Power9 added xxspltib instruction. > +/*

Re: [RFC][PATCH] Change default to -fcommon

Hi, On Mon, 20 Nov 2017, Sandra Loosemore wrote: > > I have a hard time imaging that, so can you give details? FWIW I've > > personally always considered using common symbols nicer. > > The optimization case I'm familiar with is for targets that support -G > and GP-relative addressing modes

Small C++ PATCH to avoid duplicate visibility warning

We've been warning about a base with different visibility while walking the fields list as well as the base list. Tested x86_64-pc-linux-gnu, applying to trunk. commit de7f9f640df9dc1e181284009ba48ac281bef825 Author: Jason Merrill Date: Fri Nov 17 17:13:40 2017 -0500

Re: [PATCH] avoid assuming strncat/strncpy don't change length of source string (PR 83075)

On 11/20/2017 12:39 PM, Jakub Jelinek wrote: On Mon, Nov 20, 2017 at 12:15:09PM -0700, Martin Sebor wrote: PR tree-optimization/83075 * tree-ssa-strlen.c (handle_builtin_stxncpy): Avoid assuming strncat/strncpy don't change length of source string.

Re: [PATCH] Avoid static initialization in the strlen pass

On Mon, Nov 20, 2017 at 02:25:35PM -0700, Martin Sebor wrote: > On 11/20/2017 02:14 PM, Jakub Jelinek wrote: > > Hi! > > > > All the hash_maps in tree-ssa-strlen.c except for the newly added one > > were pointers to hash maps, which were constructed either lazily or during > > the pass. But

[PATCH, rs6000] (v2) fold-vec-ld* testcase updates for power9

Hi, Update the scan-assembler stanza to include those instructions generated for a power9 target. 2017-11-20 Will Schmidt [testsuite] * fold-vec-ld-char.c: Add lxv insn to expected output. * fold-vec-ld-double.c: Add lxv

[PATCH, rs6000] (v2) fold-vec-splat* testcase updates for power9

Hi, Update the scan-assembler stanza to include those instructions generated for a power9 target. 2017-11-20 Will Schmidt [testsuite] * fold-vec-splat-8.c : Add vspltisb to expected output. * fold-vec-splats-int.c : Add

Re: [PATCH] avoid assuming strncat/strncpy don't change length of source string (PR 83075)

On Mon, Nov 20, 2017 at 02:38:10PM -0700, Martin Sebor wrote: > > I think it is better to have a dg-do run testcase and not scan for abort. > > We can be able to optimize at some point n0 to 3 and n1 to 6 and optimize > > away the abort. In the testcase I've added to the PR there was a separate >

Re: [RFC][PATCH] Change default to -fcommon

Hi, On Mon, 20 Nov 2017, Richard Biener wrote: > Also we cannot raise alignment of commons and thus vectorization is > pessimized (all vectorizer testcases use - fno-common). That would be a simple oversight then. That's one of the nice things with common symbols, they contain their own

[PATCH, rs6000] (v2) testcase updates for fold-vec-abs-* for power9 codegen

Hi, Add additional scan-assembler entries for those tests with simple variations in codegen. (for power9, versus power8, etc). 2017-11-20 Will Schmidt [testsuite] * fold-vec-abs-char-fwrapv.c: Add xxspltib insn to expected

Re: Patch ping

On 11/19/2017 11:55 PM, Jakub Jelinek wrote: I'd like to ping the following patches: http://gcc.gnu.org/ml/gcc-patches/2017-10/msg01895.html PR debug/82718 Fix DWARF5 .debug_loclist handling with hot/cold partitioning I already responded to this one.

Re: [patches] Re: [PATCH] RISC-V: Implement __umulsidi3, umul_ppmm and __muluw3

On Sun, 19 Nov 2017 23:31:56 PST (-0800), ja...@redhat.com wrote: On Sun, Nov 19, 2017 at 08:53:00PM -0800, Jim Wilson wrote: > 2017-11-24 Kito Cheng > > * longlong.h [__riscv] (__umulsidi3): Define. > [__riscv] (umul_ppmm) Likewise. > [__riscv]

Re: [RFC][PATCH] Change default to -fcommon

On 11/20/2017 10:34 AM, Michael Matz wrote: What's your rationale for changing this? In your initial mail you said: "On many targets this means global variable accesses having an unnecessary codesize and performance penalty in C code (the same source generates better code when built as C++)."

Re: Increase precision of static profiles

On 19 November 2017 at 11:01, Tom de Vries wrote: > On 11/17/2017 08:53 PM, Jan Hubicka wrote: >> >> Hi, >> this patch makes static profile to be in range 0...2^30 rather than >> 0...1. This is safe now as profile-counts are taking care of >> possible overflow when

Re: [PATCH libstdc++/66689] comp_ellint_3 and ellint_3 return garbage values

On 20 November 2017 at 17:02, David Edelsohn wrote: > This patch has introduced new regressions on at least PowerPC and AArch64. > > FAIL: ext/special_functions/hyperg/check_value.cc execution test > FAIL: tr1/5_numerical_facilities/special_functions/17_hyperg/check_value.cc >

Re: [14/nn] Add helpers for shift count modes

Richard Biener writes: > On Thu, Oct 26, 2017 at 2:06 PM, Richard Biener > wrote: >> On Mon, Oct 23, 2017 at 1:25 PM, Richard Sandiford >> wrote: >>> This patch adds a stub helper routine to provide the mode

Re: [PATCH libstdc++/66689] comp_ellint_3 and ellint_3 return garbage values

On 20/11/17 21:43 +0100, Christophe Lyon wrote: On 20 November 2017 at 17:02, David Edelsohn wrote: This patch has introduced new regressions on at least PowerPC and AArch64. FAIL: ext/special_functions/hyperg/check_value.cc execution test FAIL:

[PATCH] Fix ICEs from expand_mul_overflow (PR target/82981)

Hi! Apparently ARM can do the widening SImode multiply only using a libcall, so the recently changed expand_mul_overflow ignores it at first, then sees possibility to use the HImode code, but uses expand_simple_binop with OPTAB_DIRECT which requires actual HImode optabs, which ARM doesn't have,

[PATCH] Fix ICE with __RTL and -g (PR debug/82933)

Hi! The dwarf2out.c code relies on the assembly_start debug hook being invoked before any RTL is processed from final.c (and needs it to be done just once). Normally it is called from cgraphunit.c, but when __RTL is seen with a starting pass, run_rtl_passes is called already from the FE. While

[PATCH, i386]: Add bswaphi2 insn pattern

Hello! Attached patch introduces bswaphi2 named insn pattern that results in movbe instruction. Without the patch, the following testcase:

Re: [patch] Add support for #pragma GCC unroll

On 20 November 2017 at 15:45, Steve Kargl wrote: > On Mon, Nov 20, 2017 at 12:57:47PM +0100, Bernhard Reutner-Fischer wrote: >> On 20 November 2017 at 12:26, Eric Botcazou wrote: >> >> If anybody finds the time to push the corresponding

Re: [PATCH, rs6000] Testcase updates for power9 codegen

On Mon, 2017-11-20 at 11:23 -0600, Segher Boessenkool wrote: > Hi! > > On Fri, Nov 17, 2017 at 11:30:27AM -0600, Will Schmidt wrote: > > [testsuite] > > * fold-vec-neg-int.c: Specify -mcpu=power8 > > Dot at end of sentence (quite a few times), please. > > > -/* { dg-final {

[committed] i386.c diagnostics fix

Hi! While working on the just posted patch, I've noticed another spot in i386.c that uses incorrect capitalization in diagnostics. Fixed thusly, bootstrapped/regtested on x86_64-linux and i686-linux, committed to trunk as obvious. 2017-11-20 Jakub Jelinek *

Re: [PATCH 2/3] [ARM] Refactor costs calculation for MEM.

On 15 September 2017 at 18:01, Kyrill Tkachov wrote: > > On 15/09/17 16:38, Charles Baylis wrote: >> >> On 13 September 2017 at 10:02, Kyrill Tkachov >> wrote: >>> >>> Hi Charles, >>> >>> On 12/09/17 09:34, charles.bay...@linaro.org

Re: [PATCH 1/3] [ARM] Add bus_width_bits to tune_params

On 15 September 2017 at 18:01, Kyrill Tkachov wrote: > From what I can tell Ramana and Richard preferred to encode this attribute > as > a tuning struct property rather than an inline conditional based on > arm_arch7. > I agree that if we want to use that

[PATCH] avoid assuming strncat/strncpy don't change length of source string (PR 83075)

Unlike strcat and strcpy, it is valid for strncat and strncpy to result in modifying the source string by overwriting its terminating nul. The attached patch removes an assumption to the contrary introduced with the -Wstringop-truncation enhancement. Martin PR tree-optimization/83075 - Invalid

Re: [RFC][PATCH] Change default to -fcommon

On November 20, 2017 9:02:55 PM GMT+01:00, Sandra Loosemore wrote: >On 11/20/2017 10:34 AM, Michael Matz wrote: >> What's your rationale for changing this? In your initial mail you >said: >> >> "On many targets this means global variable accesses having an >unnecessary

[PATCH] Fix load_gsi computation in store-merging (PR tree-optimization/83047)

Hi! This is something the bswap pass has been already doing, but not the new load handling code in store-merging. If all the loads have the same vuse, it still doesn't mean they are necessarily in the same basic block. If they are in multiple bbs, previously we've chosen randomly (well, from the

[C/C++ PATCH] Fix ICE in get_atomic_generic_size (PR c++/83059)

Hi! Not all INTEGER_CSTs satisfy tree_fits_uhwi_p, and those that don't ICE in tree_to_uhwi. But, what the memmodel_base function actually does is mask only the 16 low bits, the upper bits are reserved for target dependent flags and -Winvalid-memory-model already warns about those during

[PATCH] PR 53796 Improve INQUIRE(RECL=...) handling

The current F2018 draft (N2137) specifies behavior of the RECL= specifier in the INQUIRE statement, where it previously was left as undefined. Namely: - If the unit is not connected, RECL= should be given the value -1. - If the unit is connected with stream access, RECL= should be given the

[PATCH] Avoid static initialization in the strlen pass

Hi! All the hash_maps in tree-ssa-strlen.c except for the newly added one were pointers to hash maps, which were constructed either lazily or during the pass. But strlen_to_stridx is now constructed at the compiler start, which is something I'd prefer to avoid, it affects even -O0 that way and

Re: [Patch][aarch64] Use IFUNCs to enable LSE instructions in libatomic on aarch64

On Mon, 2017-11-20 at 18:27 +, James Greenhalgh wrote: > > If you have the time, would you mind giving me a quick run-down of what > design decisions went in to this patch, and why they are the right thing > to do? Sorry to offload that, but it will be the most efficient route > to a review.

Re: [PATCH] avoid assuming strncat/strncpy don't change length of source string (PR 83075)

On Mon, Nov 20, 2017 at 12:15:09PM -0700, Martin Sebor wrote: > PR tree-optimization/83075 > * tree-ssa-strlen.c (handle_builtin_stxncpy): Avoid assuming > strncat/strncpy don't change length of source string. > > gcc/testsuite/ChangeLog: > > PR tree-optimization/83075 >

Re: [patch] Add support for #pragma GCC unroll

On 11/17/2017 03:23 AM, Eric Botcazou wrote: Index: doc/extend.texi === --- doc/extend.texi (revision 254797) +++ doc/extend.texi (working copy) @@ -22376,6 +22376,18 @@ void ignore_vec_dep (int *a, int k, int @} @end

Re: [PATCH libstdc++/66689] comp_ellint_3 and ellint_3 return garbage values

On 20/11/17 21:01 +, Jonathan Wakely wrote: On 20/11/17 21:43 +0100, Christophe Lyon wrote: On 20 November 2017 at 17:02, David Edelsohn wrote: This patch has introduced new regressions on at least PowerPC and AArch64. FAIL: ext/special_functions/hyperg/check_value.cc

Re: [C/C++ PATCH] Fix ICE in get_atomic_generic_size (PR c++/83059)

On 11/20/2017 04:03 PM, Jakub Jelinek wrote: Hi! Not all INTEGER_CSTs satisfy tree_fits_uhwi_p, and those that don't ICE in tree_to_uhwi. But, what the memmodel_base function actually does is mask only the 16 low bits, the upper bits are reserved for target dependent flags and

Re: [PATCH] Avoid static initialization in the strlen pass

On 11/20/2017 02:14 PM, Jakub Jelinek wrote: Hi! All the hash_maps in tree-ssa-strlen.c except for the newly added one were pointers to hash maps, which were constructed either lazily or during the pass. But strlen_to_stridx is now constructed at the compiler start, which is something I'd

[Patch][aarch64][committed] Fix test broken by patch for PR target/81356

My patch for PR target/81356 caused the test gfortran.dg/pr45636.f90 to generate an XPASS instead of an XFAIL.  This change seems correct since we no longer define TARGET_USE_BY_PIECES_INFRASTRUCTURE_P so I am going to check this patch in as obvious in order to make the test a PASS instead of

Re: [PATCH] Fix ICE with __RTL and -g (PR debug/82933)

On 11/20/2017 02:10 PM, Jakub Jelinek wrote: > Hi! > > The dwarf2out.c code relies on the assembly_start debug hook being > invoked before any RTL is processed from final.c (and needs it to be > done just once). Normally it is called from cgraphunit.c, but when > __RTL is seen with a starting

Re: [014/nnn] poly_int: indirect_refs_may_alias_p

On 11/20/2017 06:00 AM, Richard Sandiford wrote: > Jeff Law writes: >> On 10/23/2017 11:05 AM, Richard Sandiford wrote: >>> This patch makes indirect_refs_may_alias_p use ranges_may_overlap_p >>> rather than ranges_overlap_p. Unlike the former, the latter can handle >>> negative

Re: [PATCH] detect nonstring arguments to string functions (PR 82945)

On 11/19/2017 03:26 PM, Martin Sebor wrote: >>> I have seen it in the early IL but this late I couldn't come >>> up with a test case where such a loop would be necessary. >>> The COMPONENT_REF always refers to the member array.  If you >>> can show me an example to test with I'll add the handling

Re: [PING][patch] PR81794: have "would be stringified in traditional C" warning in libcpp/macro.c be controlled by -Wtraditional

On Wed, 2017-10-25 at 14:45 -0400, Eric Gallager wrote: > On Sat, Sep 30, 2017 at 8:05 PM, Eric Gallager > wrote: > > On Fri, Sep 29, 2017 at 11:15 AM, David Malcolm > m> wrote: > > > On Sun, 2017-09-17 at 20:00 -0400, Eric Gallager wrote: > > > >

Re: Patch ping

On 11/20/2017 02:58 PM, Jim Wilson wrote: The dwarf2out.c patch looks good to me, though the testcase does not fail on unpatched mainline anymore.  I had to go back to the 2017-10-22 snapshot to see the failure.  There was a followup from Mark Wielaard mentioning that elfutils still fails on

Re: [PATCH 1/7]: SVE: Add CLOBBER_HIGH expression

On 11/20/2017 08:04 AM, Alan Hayward wrote: >>> >>> Yes, that was an early alternative option for the patch. >>> >>> With that it would effect every operation that uses SVE registers. A simple >>> add of two registers now has 4 inputs and two outputs. It would get in the >>> way when debugging any

Re: [PATCH] Don't split call from its call arg location

On 11/20/2017 04:04 AM, Claudiu Zissulescu wrote: > When a port is using hardware loops (like ARC) makes use of > reorg_loops to find and analyze loops that end in loop_end > instructions. The very same function can be set to reorder the cfg > such that the loop end occurs after the loop start.

Re: [PATCH] Fix ICEs from expand_mul_overflow (PR target/82981)

On 11/20/2017 02:07 PM, Jakub Jelinek wrote: > Hi! > > Apparently ARM can do the widening SImode multiply only using a libcall, > so the recently changed expand_mul_overflow ignores it at first, then > sees possibility to use the HImode code, but uses expand_simple_binop > with OPTAB_DIRECT which

libgo patch committed: Fix Makefile bug setting LD_LIBRARY_PATH

This patch by Than McIntosh fixes a small bug in the libgo Makefile recipe that constructs the directory from which to pick up libgcc_s.so; the gccgo invocation with -print-libgcc-file-name was missing the flags, which meant that for -m32 builds we'd see the 64-bit libgcc dir. Bootstrapped and

[PING] Plugin support on Windows/MinGW

Hi, I would like to ping this patch: https://gcc.gnu.org/ml/gcc-patches/2017-11/msg01040.html The changes are fairly conservative: they do not touch much of the existing module implementation and plugin support on MinGW is disabled by default (there are also fixes for a couple of bugs as a

Go patch committed: report error for ++/-- applied to a non-numeric type

This patch to the Go frontend reports an error when ++ or -- is used with a non-numeric type. This avoids a later compiler crash. This fixes GCC PR 83071. Bootstrapped and ran Go testsuite on x86_64-pc-linux-gnu. Committed to mainline. Ian Index: gcc/go/gofrontend/MERGE

Re: [105/nnn] poly_int: expand_assignment

On 10/23/2017 11:42 AM, Richard Sandiford wrote: > This patch makes the CONCAT handing in expand_assignment cope with > polynomial mode sizes. The mode of the CONCAT must be complex, > so we can base the tests on the sizes of the real and imaginary > components. > > > 2017-10-23 Richard

Re: [107/nnn] poly_int: GET_MODE_SIZE

On 10/23/2017 11:43 AM, Richard Sandiford wrote: > This patch changes GET_MODE_SIZE from unsigned short to poly_uint16. > The non-mechanical parts were handled by previous patches. > > > 2017-10-23 Richard Sandiford > Alan Hayward

Re: [106/nnn] poly_int: GET_MODE_BITSIZE

On 10/23/2017 11:43 AM, Richard Sandiford wrote: > This patch changes GET_MODE_BITSIZE from an unsigned short > to a poly_uint16. > > > 2017-10-23 Richard Sandiford > Alan Hayward > David Sherwood

Re: [102/nnn] poly_int: vect_permute_load/store_chain

On 10/23/2017 11:41 AM, Richard Sandiford wrote: > The GET_MODE_NUNITS patch made vect_grouped_store_supported and > vect_grouped_load_supported check for a constant number of elements, > so vect_permute_store_chain and vect_permute_load_chain can assert > for that. This patch adds commentary to

  1   2   >