Re: Coarray on trunk
Hi Jerry, thanks for the report. I am positive that I merged the correct patch, but I'll check tomorrow when it's a little bit cooler. At the moment it's so incredibly hot, one can't concentrate. Sorry for the mess, Andre Andre Vehreschild * ve...@gmx.de Am 1. Juli 2025 19:33:41 schrieb Jerry D : Hi all, With latest trunk this morning after Andre's commits I am still getting failures with Toon's random_weather.f90. Was there another patch I missed? Jerry $ gfc -fcoarray=single random-weather.f90 f951: internal compiler error: in gfc_ref_this_image, at fortran/expr.cc:5948 0x21dce1f internal_error(char const*, ...) ../../trunk/gcc/diagnostic-global-context.cc:517 0x647e93 fancy_abort(char const*, int, char const*) ../../trunk/gcc/diagnostic.cc:1810 0x40369d gfc_ref_this_image(gfc_ref*) ../../trunk/gcc/fortran/expr.cc:5948 0x40369d gfc_ref_this_image(gfc_ref*) ../../trunk/gcc/fortran/expr.cc:5944 0x730068 resolve_operator ../../trunk/gcc/fortran/resolve.cc:4831 0x730068 gfc_resolve_expr(gfc_expr*) ../../trunk/gcc/fortran/resolve.cc:7860 0x72e449 gfc_resolve_expr(gfc_expr*) ../../trunk/gcc/fortran/resolve.cc:7834 0x72e449 resolve_operator ../../trunk/gcc/fortran/resolve.cc:4356 0x72e449 gfc_resolve_expr(gfc_expr*) ../../trunk/gcc/fortran/resolve.cc:7860 0x734503 gfc_resolve_expr(gfc_expr*) ../../trunk/gcc/fortran/resolve.cc:7834
Re: Coarray on trunk
Hi Jerry, I just had a look into pr120843 and I have successfully merged the old version of the patch. This what you are hitting now. I am very sorry. The new 'else if' near the coarray error message should be something like: else if (!op1.ref || !gfc_ref_this_image (op1.ref) || and the same for op2. I will fix it tomorrow. Sorry again for all the bother. - Andre Andre Vehreschild * ve...@gmx.de Am 1. Juli 2025 19:33:41 schrieb Jerry D : Hi all, With latest trunk this morning after Andre's commits I am still getting failures with Toon's random_weather.f90. Was there another patch I missed? Jerry $ gfc -fcoarray=single random-weather.f90 f951: internal compiler error: in gfc_ref_this_image, at fortran/expr.cc:5948 0x21dce1f internal_error(char const*, ...) ../../trunk/gcc/diagnostic-global-context.cc:517 0x647e93 fancy_abort(char const*, int, char const*) ../../trunk/gcc/diagnostic.cc:1810 0x40369d gfc_ref_this_image(gfc_ref*) ../../trunk/gcc/fortran/expr.cc:5948 0x40369d gfc_ref_this_image(gfc_ref*) ../../trunk/gcc/fortran/expr.cc:5944 0x730068 resolve_operator ../../trunk/gcc/fortran/resolve.cc:4831 0x730068 gfc_resolve_expr(gfc_expr*) ../../trunk/gcc/fortran/resolve.cc:7860 0x72e449 gfc_resolve_expr(gfc_expr*) ../../trunk/gcc/fortran/resolve.cc:7834 0x72e449 resolve_operator ../../trunk/gcc/fortran/resolve.cc:4356 0x72e449 gfc_resolve_expr(gfc_expr*) ../../trunk/gcc/fortran/resolve.cc:7860 0x734503 gfc_resolve_expr(gfc_expr*) ../../trunk/gcc/fortran/resolve.cc:7834
[PATCH] Fortran: fix minor issues with coarrays
Dear all, the attached patch fixes the following minor issues found by running f951 under valgrind for the just added new testcases coindexed_6.f90 and coindexed_7.f90: - minor front-end memleaks with non-freed strings and lost GMP variables (these are simple and obvious fixes) - an inconsistency between pure/elemental functions being either non-intrinsic or intrinsic. Checking for the latter was likely missed from the beginning. No new testcase. Regtested on x86_64-pc-linux-gnu. OK for mainline? Thanks, Harald From 1630d24e3e1d3ab5ae6450da0da555f4eed6339c Mon Sep 17 00:00:00 2001 From: Harald Anlauf Date: Tue, 1 Jul 2025 21:41:53 +0200 Subject: [PATCH] Fortran: fix minor issues with coarrays gcc/fortran/ChangeLog: * coarray.cc (check_add_new_component): Treat pure and elemental intrinsic functions the same as non-intrinsic ones. (create_caf_add_data_parameter_type): Fix front-end memleaks. * trans-intrinsic.cc (conv_caf_func_index): Likewise. --- gcc/fortran/coarray.cc | 7 +-- gcc/fortran/trans-intrinsic.cc | 2 +- 2 files changed, 6 insertions(+), 3 deletions(-) diff --git a/gcc/fortran/coarray.cc b/gcc/fortran/coarray.cc index 6914697c78b..ef8fd4e42d0 100644 --- a/gcc/fortran/coarray.cc +++ b/gcc/fortran/coarray.cc @@ -697,7 +697,10 @@ check_add_new_component (gfc_symbol *type, gfc_expr *e, gfc_symbol *add_data) break; case EXPR_FUNCTION: if (!e->symtree->n.sym->attr.pure - && !e->symtree->n.sym->attr.elemental) + && !e->symtree->n.sym->attr.elemental + && !(e->value.function.isym + && (e->value.function.isym->pure + || e->value.function.isym->elemental))) /* Treat non-pure/non-elemental functions. */ check_add_new_comp_handle_array (e, type, add_data); else @@ -743,7 +746,6 @@ create_caf_add_data_parameter_type (gfc_expr *expr, gfc_namespace *ns, add_data->as->lower[0] = gfc_get_constant_expr (BT_INTEGER, gfc_default_integer_kind, &expr->where); - mpz_init (add_data->as->lower[0]->value.integer); mpz_set_si (add_data->as->lower[0]->value.integer, 1); for (gfc_ref *ref = expr->ref; ref; ref = ref->next) @@ -763,6 +765,7 @@ create_caf_add_data_parameter_type (gfc_expr *expr, gfc_namespace *ns, type->declared_at = expr->where; gfc_set_sym_referenced (type); gfc_commit_symbol (type); + free (name); return type; } diff --git a/gcc/fortran/trans-intrinsic.cc b/gcc/fortran/trans-intrinsic.cc index fce5ee28de8..f1bfd3eee51 100644 --- a/gcc/fortran/trans-intrinsic.cc +++ b/gcc/fortran/trans-intrinsic.cc @@ -1052,7 +1052,7 @@ conv_caf_func_index (stmtblock_t *block, gfc_namespace *ns, const char *pat, index_st->n.sym->value = gfc_get_constant_expr (BT_INTEGER, gfc_default_integer_kind, &gfc_current_locus); - mpz_init_set_si (index_st->n.sym->value->value.integer, -1); + mpz_set_si (index_st->n.sym->value->value.integer, -1); index_st->n.sym->ts.type = BT_INTEGER; index_st->n.sym->ts.kind = gfc_default_integer_kind; gfc_set_sym_referenced (index_st->n.sym); -- 2.43.0
Re: [PATCH] Fortran: fix minor issues with coarrays
On 7/1/25 12:51 PM, Harald Anlauf wrote: Dear all, the attached patch fixes the following minor issues found by running f951 under valgrind for the just added new testcases coindexed_6.f90 and coindexed_7.f90: - minor front-end memleaks with non-freed strings and lost GMP variables (these are simple and obvious fixes) - an inconsistency between pure/elemental functions being either non-intrinsic or intrinsic. Checking for the latter was likely missed from the beginning. No new testcase. Regtested on x86_64-pc-linux-gnu. OK for mainline? Thanks, Harald Yes, OK, straight-forward. Thanks, Jerry
Coarray on trunk
Hi all, With latest trunk this morning after Andre's commits I am still getting failures with Toon's random_weather.f90. Was there another patch I missed? Jerry $ gfc -fcoarray=single random-weather.f90 f951: internal compiler error: in gfc_ref_this_image, at fortran/expr.cc:5948 0x21dce1f internal_error(char const*, ...) ../../trunk/gcc/diagnostic-global-context.cc:517 0x647e93 fancy_abort(char const*, int, char const*) ../../trunk/gcc/diagnostic.cc:1810 0x40369d gfc_ref_this_image(gfc_ref*) ../../trunk/gcc/fortran/expr.cc:5948 0x40369d gfc_ref_this_image(gfc_ref*) ../../trunk/gcc/fortran/expr.cc:5944 0x730068 resolve_operator ../../trunk/gcc/fortran/resolve.cc:4831 0x730068 gfc_resolve_expr(gfc_expr*) ../../trunk/gcc/fortran/resolve.cc:7860 0x72e449 gfc_resolve_expr(gfc_expr*) ../../trunk/gcc/fortran/resolve.cc:7834 0x72e449 resolve_operator ../../trunk/gcc/fortran/resolve.cc:4356 0x72e449 gfc_resolve_expr(gfc_expr*) ../../trunk/gcc/fortran/resolve.cc:7860 0x734503 gfc_resolve_expr(gfc_expr*) ../../trunk/gcc/fortran/resolve.cc:7834
Re: [PATCH] Fortran: fix minor issues with coarrays
Am 01.07.25 um 21:54 schrieb Jerry D: On 7/1/25 12:51 PM, Harald Anlauf wrote: Dear all, the attached patch fixes the following minor issues found by running f951 under valgrind for the just added new testcases coindexed_6.f90 and coindexed_7.f90: - minor front-end memleaks with non-freed strings and lost GMP variables (these are simple and obvious fixes) - an inconsistency between pure/elemental functions being either non-intrinsic or intrinsic. Checking for the latter was likely missed from the beginning. No new testcase. Regtested on x86_64-pc-linux-gnu. OK for mainline? Thanks, Harald Yes, OK, straight-forward. Thanks, Jerry Thanks, Jerry! Pushed as r16-1903-gfc029f5d34168c. Hope it doesn't interfere with Andre's work... Harald
Re: [PATCH] gcc: middle-end opt for trigonometric pi-based functions builtins
On Sat, Jun 28, 2025 at 02:19:59AM +, Yuao Ma wrote: > > signbit is documented to be a macro, so please don't declare > > int signbit (double); > > function in the testcase and instead of signbit use __builtin_signbit. > > This is indeed my negligence. Done. > > If everything looks good, could you please help me merge this patch? Thank > you! Committed in r16-1839. I don't have mpfr 4.2.0 yet, so didn't test myself with it. Jakub
Re: [Fortran, Patch, PR120843, v2] Fix reject valid, because of inconformable coranks
Hi Harald, thanks for the review. Committed as gcc-16-1885-g1b0930e9046. Will backport to gcc-15 in about a week. Thanks again. Regards, Andre On Mon, 30 Jun 2025 22:31:08 +0200 Harald Anlauf wrote: > Am 30.06.25 um 15:25 schrieb Andre Vehreschild: > > Hi all, > > > > here now the version of the patch that seems to be more complete. > > > > Regtests ok on x86_64-pc-linux-gnu / F41. Ok for mainline and later > > backport to gcc-15? > > This looks good to me. OK for both. > > Thanks for the patch! > > Harald > > > Regards, > > Andre > > > > On Fri, 27 Jun 2025 15:44:20 +0200 > > Andre Vehreschild wrote: > > > >> I take this patch back. It seems to be incomplete. > >> > >> - Andre > >> > >> On Fri, 27 Jun 2025 14:45:36 +0200 > >> Andre Vehreschild wrote: > >> > >>> Hi all, > >>> > >>> this patch fixes a reject valid when the coranks of two operands do not > >>> match and no coindex is given. I.e. when only an implicit this_image > >>> co-ref is used. > >>> > >>> Regtests ok on x86_64-pc-linux-gnu / F41. Ok for mainline? > >>> > >>> Regards, > >>> Andre > >> > >> > > > > > -- Andre Vehreschild * Email: vehre ad gmx dot de
Re: [Fortran, Patch, PR120847 (was: PR120846), v1] Fix ICE when a function is called more than once in a coarray expression.
Hi Harald, thanks for review. Committed as gcc-16-1891-gee31ab9b195. Sorry for the typo in the email title. I should have proof read that before sending. I will set me a reminder to backport in a week to gcc-15. Thanks again, Andre On Mon, 30 Jun 2025 21:44:03 +0200 Harald Anlauf wrote: > Am 30.06.25 um 15:29 schrieb Andre Vehreschild: > > Hi all, > > > > attached patch fixes an ICE when in an expression with a coindex a function > > was used more than once. E.g. coarray(mod( something ), mod( something else > > ))[i] ICE'd because a component for aliasing the second mod() could not be > > created. > > > > Regtests ok on x86_64-pc-linux-gnu / F41. Ok for mainline and gcc-15 later > > on? > > > > Regards, > > Andre > > Yes, this is OK for both. > > (Be careful about PR numbers, they are correct in the patch; > the subject is off-by-one.) > > Thanks for the patch! > > Harald > -- Andre Vehreschild * Email: vehre ad gmx dot de
