Re: [Patch] [OpenMP, Fortran] Add structure/derived-type element mapping

Hi Julian, Tobias! On 2020-06-24T19:32:09+0200, Tobias Burnus wrote: > (OpenMP 5 extends this a lot, but this is about OpenMP 4.5. > It touches code which is also used by OpenACC's attach/detach.) > > @OpenACC/Julian: I think the character attach/detach for > deferred-length strings does not work

Re: [PATCH] rs6000: Define movsf_from_si2 to extract high part SF element from DImode[PR89310]

Hi David, On 2020/7/14 22:17, David Edelsohn wrote: > Unfortunately this patch is eliciting a number of new testsuite > failures, all like > > error: unrecognizable insn: > (insn 44 43 45 5 (parallel [ > (set (reg:SI 199) > (unspec:SI [ > (re

Re: [PATCH] libgomp: Add OMPD Address Space Information functions.

Hello Jakub, I remember now why I had gompd_env_info as a global variable. It was meant to be in libgomp.so.1. It was mistakenly placed in libgompd.h; it was supposed to be in libgomp.h. I am having trouble connecting the dots and need help. To expose a variable, such as the gompd_env_info, so th

Re: [PATCH rs6000]: Refine RTL unroll hook for small loops

Hi Jiu Fu, On Mon, Jul 13, 2020 at 07:50:28PM +0800, guojiufu wrote: > For very small loops (< 6 insns), it would be fine to unroll 4 > times to run fast with less latency and better cache usage. > - /* TODO: This is hardcoded to 10 right now. It can be refined, for > - example we may

Re: [Patch] libgomp: Add Fortran routine support for allocators

On Tue, Jul 14, 2020 at 11:42:15PM +0200, Tobias Burnus wrote: > + fortran@, which I forgot for the initial patch. > > On 7/14/20 11:43 AM, Jakub Jelinek wrote: > > > > + type omp_alloctrait > > > +integer (omp_alloctrait_key_kind) key > > > +integer (omp_alloctrait_val_kind)

Re: RFC: make combine do as advertised (cheaper-than)?

Hi! On Mon, Jul 06, 2020 at 04:01:54AM +0200, Hans-Peter Nilsson via Gcc-patches wrote: > Most comments, including the second sentence in the head comment > of combine_validate_cost, the main decision-maker of the combine > pass, refer to the function as returning true if the new > insns(s) *chea

Re: RFC: make combine do as advertised (cheaper-than)?

Hi! On Mon, Jul 06, 2020 at 10:48:25AM +0100, Richard Sandiford wrote: > Originally combine always produced shorter sequences, so by the (Shorter in # insns, not # bytes). > above reasoning, combining for == was correct. These days we allow > N-to-N replacements too, which are obviously a good

Re: [Patch] libgomp: Add Fortran routine support for allocators

+ fortran@, which I forgot for the initial patch. On 7/14/20 11:43 AM, Jakub Jelinek wrote: + type omp_alloctrait +integer (omp_alloctrait_key_kind) key +integer (omp_alloctrait_val_kind) value + end type omp_alloctrait I know this is a problem in the standard, but wo

Re: RFA: Fix combine.c combining a move and a non-move into two non-moves, PR93372

Hi! On Mon, Jul 13, 2020 at 07:25:37AM +0200, Hans-Peter Nilsson wrote: > > > > > TL;DR: fixing a misdetection of what is a "simple move". > > > > > > > > That is not a very correct characterisation of what this does :-) > > > > > > That's apparently where we completely disagree. :-) > > > > We

[committed] c++: Add new test [PR59978]

Fixed in r224162. That came without a test so adding this one. Previously, we issued a bogus "too few arguments to function" error. Tested x86_64-pc-linux-gnu, applying to trunk. gcc/testsuite/ChangeLog: PR c++/59978 * g++.dg/cpp0x/vt-59978.C: New test. --- gcc/testsuite/g++.dg

[PATCH] PR fortran/89574 - [8/9/10/11 Regression] ICE in conv_function_val, at fortran/trans-expr.c:3792

As Fortran allows to rename a symbol on use, we need to be careful to check whether a name we are looking for is a global one or belongs to a different module. Check for this condition. Regtested on x86_64-pc-linux-gnu. OK for master / backports after some time for a proper burn-in? Thanks, Har

Re: [PATCH] libgccjit: Handle truncation and extension for casts [PR 95498]

On Sun, 2020-07-12 at 20:30 -0400, Antoni Boucher via Jit wrote: > Hello. > > As mentioned in bug 95498, some conversions do not work. After > investigation, it turns out that it's caused by multiple casts on an > expression where it should do a truncation/extension. Thanks for investigating th

[PATCH] x86: Replace __glibc_unlikely with __builtin_expect

On Mon, Jul 13, 2020 at 9:44 AM Jeff Law wrote: > > On Sun, 2020-05-31 at 16:10 -0700, H.J. Lu via Gcc-patches wrote: > > cmpstrnsi expander may pass the actual string length directly to cmpstrnqi > > patterns. For cmpstrnsi, one of the strings must be a constant and > > expand_builtin_strncmp re

Re: RFA: Fix combine.c combining a move and a non-move into two non-moves, PR93372

On Mon, Jul 13, 2020 at 07:29:00AM +0200, Hans-Peter Nilsson wrote: > > From: Segher Boessenkool > > Date: Tue, 7 Jul 2020 22:50:43 +0200 > > > I'll make a simpler patch. Thanks! > > You're welcome. So, you'll take care of the updated patch > yourself? Yes. Delayed a bit, I had a lot to do f

Re: [PATCH 0/6 ver 4] ] Permute Class Operations

Hi! On Wed, Jul 08, 2020 at 12:59:19PM -0700, Carl Love wrote: > [PATCH 4/6] rs6000, Add vector shift double builtin support > * config/rs6000/altivec.h (vec_sldb, vec_srdb): New defines. > * config/rs6000/altivec.md (UNSPEC_SLDB, UNSPEC_SRDB): New. > (SLDB_LR): New attribute. >

[PATCH] libgomp: Add OMPD Address Space Information functions.

This patch adds Address Space Information function implementations as defined in section 5.5.4 of OpenMP API Specification 5.0. It also defines a struct that stores various information used by OMPD. This patch addressed all feedbacks. 2020-07-14 Tony Sim libgomp/ChangeLog: * Makefile

[PATCH 3/4] libstdc++: Add floating-point std::to_chars implementation

This implements the floating-point std::to_chars overloads for float, double and long double. We use the Ryu library to compute the shortest round-trippable fixed and scientific forms of a number for float, double and long double. We also use Ryu for performing fixed and scientific formatting of

[PATCH 2/4] libstdc++: Apply modifications to our local copy of Ryu

This performs the following modifications to our local copy of Ryu in order to make it more easily usable for our std::to_chars implementation: * Remove all #includes * Remove copy_special_str routines * Adjust the exponent formatting to match printf * Remove some functions we're not going

Re: [PATCH v3] c++: Make convert_like complain about bad ck_ref_bind again [PR95789]

On 7/13/20 4:48 PM, Marek Polacek via Gcc-patches wrote: Bootstrapped/regtested on x86_64-pc-linux-gnu, ok for trunk/10? -- >8 -- convert_like issues errors about bad_p conversions at the beginning of the function, but in the ck_ref_bind case, it only issues them after we've called convert_like

c++: Parser initialization cleanup

The handling of PCH is a little trick, because we have to deal with it before allocating memory. I found the layering somewhat confusing. This patch reorganizes that, so that the stopping of PCH is done in exactly one place, and the ordering of lexer creation relative to that is much clearer.

[PATCH v2] sparc/sparc64: use crtendS.o for default-pie executables [PR96190]

From: Sergei Trofimovich In --enable-default-pie mode compiler should switch from using crtend.o to crtendS.o. On sparc it is especially important because crtend.o contains PIC-unfriendly code. We use GNU_USER_TARGET_ENDFILE_SPEC as a baseline spec to get crtendS.o instead of crtend.o in !no-pie

Re: [PATCH] rs6000: clean up testsuite power10_hw check

Hi! On Mon, Jul 13, 2020 at 04:27:42PM -0500, Aaron Sawdey via Gcc-patches wrote: > Because the check for power10_hw is not called > check_effective_target_power10_hw, it needs to be looked > for by is-effective-target-keyword. Ah, so *that* is the reason we have this... So probably we should ju

[PATCH, rs6000, gcc-8 ] Improve handling of built-in initialization. [PR95952]

Hi, We've got a scenario with a combination of old hardware, gcc-8 and binutils where gcc will ICE during it's selftest. This ICE was exposed when the builtin processing for better #pragma support was added, where we no longer skip builtin initialization based on the current mask. Per the

Re: [PATCH] Add TARGET_LOWER_LOCAL_DECL_ALIGNMENT [PR95237]

On Sat, Jul 4, 2020 at 9:11 AM Richard Biener wrote: > > On July 3, 2020 11:16:46 PM GMT+02:00, Jason Merrill wrote: > >On 6/29/20 5:00 AM, Richard Biener wrote: > >> On Fri, Jun 26, 2020 at 10:11 PM H.J. Lu wrote: > >>> > >>> On Thu, Jun 25, 2020 at 1:10 AM Richard Biener > >>> wrote: > >

c++: tree dumper

A couple of C++ nodes were unknown to the tree dumper. gcc/cp/ * ptree.c (cxx_print_type): Add TYPEOF_TYPE and BASES. pushed -- Nathan Sidwell diff --git i/gcc/cp/ptree.c w/gcc/cp/ptree.c index 7d329049549..224cf14edb5 100644 --- i/gcc/cp/ptree.c +++ w/gcc/cp/ptree.c @@

c++: refactor a few class.c fns

Storing CLASSTYPE_AS_BASE in a local variable makes some code clearer (and textually no longer). For some reason we store a DECL in a variable called 'value', which is confusing. gcc/cp/ * class.c (build_base_field_1): Cache CLASSTYPE_AS_BASE. (build_self_re

Re: [PATCH] libgomp: Add OMPD functions in 5.5.6 and related data types.

Hello Jakub, On Tue, Jul 14, 2020 at 6:07 AM Jakub Jelinek wrote: > On Sat, Jul 11, 2020 at 06:05:36PM -0400, y2s1982 wrote: > > 2020-07-11 Tony Sim > > > > libgomp/ChangeLog: > > > > * libgompd.h (ompd_thread_handle_t): Add. > > (ompd_parallel_handle_t): Add. > > (ompd_tas

[Patch][Middle-end]Add -fzero-call-used-regs=[skip|used-gpr|all-gpr|used|all]

Hi, Gcc team, This patch is a follow-up on the previous patch and corresponding discussion: https://gcc.gnu.org/pipermail/gcc-patches/2020-May/545101.html From the previous round of discussion, the major issues raised were: A. sho

[Patch, committed] [OpenMP, Fortran] Fix goacc/finalize-1.f tree dump-scanning for -m32 (was: [Patch] [OpenMP, Fortran] Add structure/derived-type element mapping)

As pointed out by Jakub, this fails with -m32 as the dump-scanning assumed ptrdiff node = integer(kind=8). Committed as obvious (and as pre-approved) to mainline + OG10 after testing. Tobias - Mentor Graphics (Deutschland) GmbH, Arnulfstraße 201, 80634 München / Germany Register

Re: [PATCH] rs6000: Define movsf_from_si2 to extract high part SF element from DImode[PR89310]

Unfortunately this patch is eliciting a number of new testsuite failures, all like error: unrecognizable insn: (insn 44 43 45 5 (parallel [ (set (reg:SI 199) (unspec:SI [ (reg:SF 202) ] UNSPEC_SI_FROM_SF)) (clobber

Re: [PATCH] x86: Provide expanders for truncdisi2 and friends.

On Mon, Jul 13, 2020 at 4:50 PM Roger Sayle wrote: > > > Hi Richard, > > > It seems to be improving TARGET_TRULY_NOOP_TRUNCATION documentation might > > be useful here. > > This is an excellent suggestion. How about the following/attached: > > 2020-07-13 Roger Sayle > > gcc/ChangeLog: >

Re: [PATCH] builtins: Avoid useless char/short -> int promotions before atomics [PR96176]

On Tue, 14 Jul 2020, Jakub Jelinek wrote: > On Tue, Jul 14, 2020 at 02:33:41PM +0200, Richard Biener wrote: > > > As mentioned in the PR, we generate a useless movzbl insn before lock > > > cmpxchg. > > > The problem is that the builtin for the char/short cases has the arguments > > > promoted to

Re: [PATCH] builtins: Avoid useless char/short -> int promotions before atomics [PR96176]

Hi! On Tue, Jul 14, 2020 at 03:26:50PM +0200, Jakub Jelinek wrote: > On Tue, Jul 14, 2020 at 02:33:41PM +0200, Richard Biener wrote: > > > As mentioned in the PR, we generate a useless movzbl insn before lock > > > cmpxchg. > > > The problem is that the builtin for the char/short cases has the ar

Re: [PATCH] expr: Unbreak build of mesa [PR96194]

On July 14, 2020 3:37:35 PM GMT+02:00, Jakub Jelinek wrote: >Hi! > >On Tue, Jul 07, 2020 at 09:04:40AM +0200, Richard Biener via >Gcc-patches wrote: >> > The store to the whole of each volatile object was picked apart >> > like there had been an individual assignment to each of the >> > fields. R

Re: [PATCH] diagnostics: Add options to control the column units [PR49973] [PR86904]

On Mon, 2020-07-13 at 17:07 -0400, Lewis Hyatt wrote: > On Mon, Jul 13, 2020 at 03:04:20PM -0400, David Malcolm wrote: [...] > > OK for trunk with those nits fixed. > > > > Dave > > > > > > Thanks again for your time! I will address the above and then push in > a day or two. Excellent - than

Re: [PATCH] libgomp: Add OMPD Address Space Information functions.

Hello Jakub, Thank you for the detailed information. I will make those changes right away. Cheers, Tony Sim On Tue, Jul 14, 2020 at 5:57 AM Jakub Jelinek wrote: > On Thu, Jul 09, 2020 at 07:01:00PM -0400, y2s1982 via Gcc-patches wrote: > > --- a/libgomp/libgompd.h > > +++ b/libgomp/libgompd.h

[PATCH] expr: Unbreak build of mesa [PR96194]

Hi! On Tue, Jul 07, 2020 at 09:04:40AM +0200, Richard Biener via Gcc-patches wrote: > > The store to the whole of each volatile object was picked apart > > like there had been an individual assignment to each of the > > fields. Reads were added as part of that; see PR for details. > > The reads f

Re: [PATCH] builtins: Avoid useless char/short -> int promotions before atomics [PR96176]

On Tue, Jul 14, 2020 at 02:33:41PM +0200, Richard Biener wrote: > > As mentioned in the PR, we generate a useless movzbl insn before lock > > cmpxchg. > > The problem is that the builtin for the char/short cases has the arguments > > promoted to int and combine gives up, because the instructions h

c++: comments & formatting

unsurprisingly, most of the reformatting I discovered is in the C++ FE. I found some bad formatting and misleading or incomplete comments during my spelunking around the c++FE. May as well clean up trunk and record what I noted. gcc/cp/ * cp-tree.h: Correct some tree l

core: comment & formatting

One of hash_map's getters returns a pointer to T or null, The comment said it returned T or null. Also an unaligned pair of comments. gcc/ * hash-map.h (hash_map::get): Note it is a pointer to value. * incpath.h (incpath_kind): Align comments. pushing to tru

core: tree-node comments and robustify

As we've moved to 64-bit systems, the padding information has become conditionally inaccurate. I also hit cases where invalid tree codes did not get flagged as invalid. gcc/ * tree-core.h (tree_decl_with_vis, tree_function_decl): Note additional padding on 6

doc: fix a couple of typos

gty calls gt_clear*e*_cache not gt_clear_cache. I know not why it is named so, but at least document it correctly. invoke.texi had a duplicate opindex. gcc/ * doc/gty.texi: Fic gt_cleare_cache name. * doc/invoke.texi: Remove duplicate opindex Wabi-tag. pus

Re: Patch RFA: In C++ demangler don't treat lambda as substitution candidate

On 7/10/20 12:33 PM, Ian Lance Taylor wrote: As discussed in PR 96143, the C++ frontend and the demangler disagree as to whether a lambda is a substitution candidate. According to the ABI, the C++ frontend is behaving correctly. This patch changes the demangler accordingly. This caused the dem

Re: [PATCH] c++: Improve checking of decls with trailing return type [PR95820]

On 6/24/20 7:27 PM, Marek Polacek wrote: This is an ICE-on-invalid but I've been seeing it when reducing various testcases, so it's more important for me than usually. splice_late_return_type now checks that if we've seen a late return type, the function return type was auto. That's a fair assu

Re: [PATCH] sparc/sparc64: use PIE_SPEC to select crtendS.o [PR96190]

> In --enable-default-pie mode compiler should switch from > using crtend.o to crtendS.o. On sparc it is especially visible > because crtend.o contains PIC-unfriendly code. Let's use GNU_USER_TARGET_ENDFILE_SPEC in ENDFILE_SPEC since we presumably use GNU_USER_TARGET_STARTFILE_SPEC for STARTFILE_

Re: [PATCH] libsanitizer: Fix GetPcSpBp determination of SP on 32-bit Solaris/x86

On Tue, Jul 14, 2020 at 02:32:57PM +0200, Rainer Orth wrote: > The latest Solaris 11.4/x86 update uncovered a libsanitizer bug that > caused one test to FAIL for 32-bit: > > +FAIL: c-c++-common/asan/null-deref-1.c -O0 output pattern test > +FAIL: c-c++-common/asan/null-deref-1.c -O1 output p

Re: [PATCH][RFC] vector creation from two parts of two vectors produces TBL rather than ins (PR93720)

Hi, please take a look at updated patch with all comments addressed (attached). Thanks, Dmitrij On Sat, Jul 11, 2020 at 09:52:40AM +0100, Richard Sandiford wrote: ... > > For this point, I meant that we should remove the first loop too. I.e.: > ... > > is now redundant with the later: > ...

[PATCH] libsanitizer: Fix GetPcSpBp determination of SP on 32-bit Solaris/x86

The latest Solaris 11.4/x86 update uncovered a libsanitizer bug that caused one test to FAIL for 32-bit: +FAIL: c-c++-common/asan/null-deref-1.c -O0 output pattern test +FAIL: c-c++-common/asan/null-deref-1.c -O1 output pattern test +FAIL: c-c++-common/asan/null-deref-1.c -O2 output patte

Re: [PATCH] builtins: Avoid useless char/short -> int promotions before atomics [PR96176]

On Tue, 14 Jul 2020, Jakub Jelinek wrote: > Hi! > > As mentioned in the PR, we generate a useless movzbl insn before lock cmpxchg. > The problem is that the builtin for the char/short cases has the arguments > promoted to int and combine gives up, because the instructions have > MEM_VOLATILE_P ar

Re: [PATCH 8/9] [OpenACC] Fix standalone attach for Fortran assumed-shape array pointers

Hi Julian, Tobias! On 2020-06-16T15:39:44-0700, Julian Brown wrote: > As mentioned in the blurb for the previous patch, an "attach" operation > for a Fortran pointer with an array descriptor must copy that array > descriptor to the target. Heh, I see -- I don't think I had read the OpenACC stand

Re: [PATCH] libgomp: Fix hang when profiling OpenACC programs with CUDA 9.0 nvprof

Hi Kwok! On 2020-07-13T16:29:14+0100, Kwok Cheung Yeung wrote: > When the version of nvprof in CUDA 9.0 is run on an OpenACC program, [...] the > program deadlocks. > I have added a testcase that sets up the situation presented by nvprof. Thanks. I have extended this one a little bit, to add s

Re: [PATCH] libgomp: Add OMPD functions in 5.5.6 and related data types.

On Sat, Jul 11, 2020 at 06:05:36PM -0400, y2s1982 wrote: > 2020-07-11 Tony Sim > > libgomp/ChangeLog: > > * libgompd.h (ompd_thread_handle_t): Add. > (ompd_parallel_handle_t): Add. > (ompd_task_handle_t): Add. > * ompd-parallel.c: New file. So you add a new file, but d

Re: [PATCH] libgccjit: Add new gcc_jit_context_new_blob entry point

Andrea Corallo writes: > Andrea Corallo writes: > >>> It occurred to me that the entrypoint is combining two things: >>> - creating a global char[] >>> - creating an initializer for that global >>> >>> which got me wondering if we should instead have a way to add >>> initializers for globals. >>

Re: [PATCH] libgomp: Add OMPD Address Space Information functions.

On Thu, Jul 09, 2020 at 07:01:00PM -0400, y2s1982 via Gcc-patches wrote: > --- a/libgomp/libgompd.h > +++ b/libgomp/libgompd.h > @@ -47,4 +47,19 @@ typedef struct _ompd_aspace_handle { >ompd_size_t ref_count; > } ompd_address_space_handle_t; > > +struct gompd_env > +{ > + /* TODO: when the

Re: [Patch] libgomp: Add Fortran routine support for allocators

On Tue, Jul 14, 2020 at 09:15:21AM +0200, Tobias Burnus wrote: > libgomp/ChangeLog: > > * allocator.c: Add ialias for omp_init_allocator and > omp_destroy_allocator. > * configure.ac: Set INTPTR_T_KIND. > * configure: Regenerate. > * Makefile.in: Regenerate. > *

Re: [Patch][Fortran, OpenMP] Fix allocatable-components check (PR67311)

On Thu, Jun 25, 2020 at 04:06:24PM +0200, Tobias Burnus wrote: > [Fortran, OpenMP] Fix allocatable-components check (PR67311) > > gcc/fortran/ChangeLog: > > PR fortran/67311 > * trans-openmp.c (gfc_has_alloc_comps): Return false also for > pointers to arrays. > > libgomp/Change

Re: [Patch] [OpenMP, Fortran] Add structure/derived-type element mapping

On Wed, Jun 24, 2020 at 07:32:09PM +0200, Tobias Burnus wrote: > Comments, remarks, suggestions? > Otherwise: OK for the trunk? LGTM, thanks. > [OpenMP, Fortran] Add structure/derived-type element mapping > > gcc/fortran/ChangeLog: > > * openmp.c (gfc_match_omp_clauses): Match also derive

[PATCH] driver: Fix several memory leaks

Updating the subject since this is really just a driver change (and therefore needs a review from those who can approve patches there). Thanks, Alex -Original Message- From: Jit On Behalf Of Alex Coplan Sent: 09 July 2020 21:13 To: gcc-patches@gcc.gnu.org; j...@gcc.gnu.org Cc: nd Subjec

[PATCH] c++: Use error_at rather than warning_at for missing return in constexpr functions [PR96182]

Hi! For C++11 we already emit an error if a constexpr function doesn't contain a return statement, because in C++11 that is the only thing it needs to contain, but for C++14 we would normally issue a -Wreturn-type warning. As mentioned by Jonathan, such constexpr functions are invalid, no diagnos

[PATCH] builtins: Avoid useless char/short -> int promotions before atomics [PR96176]

Hi! As mentioned in the PR, we generate a useless movzbl insn before lock cmpxchg. The problem is that the builtin for the char/short cases has the arguments promoted to int and combine gives up, because the instructions have MEM_VOLATILE_P arguments and recog in that case doesn't recognize anythi

[PATCH] target: fix default value checking of x_str_align_functions in aarch64.c

Hi, This patch deal with the -falign-X=0 options. According to man pages, if zero is specified, a machine-dependent default value should be used. But in fact, zero was used in internal process, it is inconsistent. Tested on aarch64-linux cross compiler, Is that OK? BTW, the similar problems exis

[Patch] libgomp: Add Fortran routine support for allocators

This patch adds the Fortran part to Jakub's generic + C/C++ commit "openmp: Add basic library allocator support." https://gcc.gnu.org/g:800bcc8c00f3ce940aa174845bb61faca9e85d36 OK for the trunk? (Will then later be also backported to OG10) Tobias - Mentor Graphics (Deutschland)

[PATCH] sparc/sparc64: use PIE_SPEC to select crtendS.o [PR96190]

From: Sergei Trofimovich In --enable-default-pie mode compiler should switch from using crtend.o to crtendS.o. On sparc it is especially visible because crtend.o contains PIC-unfriendly code. gcc: 2020-07-14 Sergei Trofimovich PR driver/96190 * config/sparc/linux.h: Use PIE_