diff options
Diffstat (limited to 'gcc')
468 files changed, 11822 insertions, 1809 deletions
diff --git a/gcc/BASE-VER b/gcc/BASE-VER index 2bbd2b4..946789e 100644 --- a/gcc/BASE-VER +++ b/gcc/BASE-VER @@ -1 +1 @@ -15.0.1 +16.0.0 diff --git a/gcc/ChangeLog b/gcc/ChangeLog index a3e0333..4a2a79f 100644 --- a/gcc/ChangeLog +++ b/gcc/ChangeLog @@ -1,3 +1,713 @@ +2025-04-19 Jeff Law <jlaw@ventanamicro.com> + + PR target/119865 + * config/riscv/riscv.cc (parse_features_for_version): Do not + explicitly free the architecture string. + +2025-04-19 Jeff Law <jlaw@ventanamicro.com> + + PR target/118410 + * config/riscv/bitmanip.md (logical with constant argument): New + splitter for cases where synthesizing ~C is cheaper than synthesizing + the original constant C. + +2025-04-19 Jan Hubicka <hubicka@ucw.cz> + + * config/i386/i386.cc (vec_fp_conversion_cost): New function. + (ix86_rtx_costs): Use it for SSE/AVX FP conversoins. + (ix86_builtin_vectorization_cost): Fix indentation; + and use vec_fp_conversion_cost in vec_promote_demote. + (fp_conversion_stmt_cost): New function. + (ix86_vector_costs::add_stmt_cost): Use it to cost NOP_EXPR + and vec_promote_demote. + * config/i386/i386.h (struct processor_costs): + * config/i386/x86-tune-costs.h (struct processor_costs): + +2025-04-19 Andrew Pinski <quic_apinski@quicinc.com> + + PR rtl-optimization/111949 + * combine.cc (find_split_point): Add a split point + for `(and (not X) Y)` if not in the outer set already. + +2025-04-19 Jiaxun Yang <jiaxun.yang@flygoat.com> + + PR target/111814 + * config/sh/sh-modes.def (RESET_FLOAT_FORMAT): Use mips format. + (FLOAT_MODE): Use mips mode. + +2025-04-19 Maciej W. Rozycki <macro@orcam.me.uk> + + * config/alpha/alpha.cc + (alpha_get_mem_rtx_alignment_and_offset): Recurse into + COMPONENT_REF nodes. + +2025-04-18 Jeff Law <jlaw@ventanamicro.com> + + * config/riscv/bitmanip.md (*bext<mode>_mask_pos): New pattern + for extracting a single bit at masked bit position. + +2025-04-18 Andrew Pinski <quic_apinski@quicinc.com> + + PR tree-optimization/87901 + * tree-ssa-dse.cc (maybe_trim_constructor_store): Add was_integer_cst argument. + Check for was_integer_cst instead of `{}` when was_integer_cst is true. + (maybe_trim_partially_dead_store): Handle INTEGER_CST stores of 0 as stores of `{}`. + Udpate call to maybe_trim_constructor_store for CONSTRUCTOR. + +2025-04-18 Andrew Pinski <quic_apinski@quicinc.com> + + PR tree-optimization/87901 + * tree-ssa-dse.cc (maybe_trim_constructor_store): Strip over useless type + conversions after taking the address of the MEM_REF. + +2025-04-18 Andrew Pinski <quic_apinski@quicinc.com> + + PR tree-optimization/118902 + * fold-const.cc (tree_swap_operands_p): Place invariants in the first operand + if not used with constants. + +2025-04-18 Andrew Pinski <quic_apinski@quicinc.com> + + PR tree-optimization/118947 + * gimple-fold.cc (optimize_memcpy_to_memset): Walk back until we get a + statement that may clobber the read. + +2025-04-18 Andrew Pinski <quic_apinski@quicinc.com> + + PR tree-optimization/78408 + PR tree-optimization/118947 + * gimple-fold.cc (optimize_memcpy_to_memset): Handle STRING_CST case too. + +2025-04-18 Richard Braun <rbraun@sceen.net> + + * config/c6x/c6x.h (ASM_PREFERRED_EH_DATA_FORMAT): Remove the + DW_EH_PE_indirect flag. + +2025-04-18 Richard Biener <rguenther@suse.de> + + PR tree-optimization/119858 + * tree-vect-loop.cc (vectorizable_live_operation): Convert + pointer offset to sizetype. + +2025-04-18 Hakan Candar <hakancandar@protonmail.com> + + * config.gcc: Recognize riscv*-*-gnu* targets. + * config/riscv/gnu.h: New file. + +2025-04-18 Alexey Merzlyakov <alexey.merzlyakov@samsung.com> + + PR middle-end/108016 + PR middle-end/108016 + * config/riscv/riscv.md (addv<mode>4, uaddv<mode>4, subv<mode>4, + usubv<mode>4): Tunes for unnecessary sext.w elimination. + +2025-04-18 kelefth <konstantinos.eleftheriou@vrull.eu> + + PR rtl-optimization/119160 + * avoid-store-forwarding.cc (process_store_forwarding): + Zero-extend the value stored in the base register, in case + of load-elimination, only when the mode of the destination + is wider. + +2025-04-18 kelefth <konstantinos.eleftheriou@vrull.eu> + + * doc/cfg.texi: Update the exception handling section for the + REG_EH_REGION notes to make it clear that the note is attached + to the instruction throwing the exception. + +2025-04-17 翁愷邑 <kaiweng9487@gmail.com> + + * config/riscv/riscv-target-attr.cc + (riscv_target_attr_parser::update_settings): + Do not manually free any arch string. + +2025-04-17 Eric Botcazou <ebotcazou@gcc.gnu.org> + + * tree.def (BOOLEAN_TYPE): Add more details. + +2025-04-17 Sam James <sam@gentoo.org> + + * doc/invoke.texi: Use "compatible types" term. Rephrase to be + more precise (and correct). + +2025-04-17 Tamar Christina <tamar.christina@arm.com> + + PR tree-optimization/119351 + * tree-vect-stmts.cc (vectorizable_early_exit): Mask both operands of + the gcond for partial masking support. + +2025-04-17 Jakub Jelinek <jakub@redhat.com> + + PR target/119834 + * config/s390/s390.md (define_split after *cpymem_short): Use + (clobber (match_scratch N)) instead of (clobber (scratch)). Use + (match_dup 4) and operands[4] instead of (match_dup 3) and operands[3] + in the last of those. + (define_split after *clrmem_short): Use (clobber (match_scratch N)) + instead of (clobber (scratch)). + (define_split after *cmpmem_short): Likewise. + +2025-04-17 Thomas Schwinge <tschwinge@baylibre.com> + + * config/nvptx/nvptx.cc (TARGET_ASM_NEED_VAR_DECL_BEFORE_USE): + Don't '#define'. + +2025-04-17 Hans-Peter Nilsson <hp@axis.com> + + * combine.cc: Correct comments about combine_validate_cost. + +2025-04-16 Sandra Loosemore <sloosemore@baylibre.com> + + PR c/88382 + * doc/extend.texi (Syntax Extensions): Adjust menu. + (Raw String Literals): New section. + +2025-04-16 Keith Packard <keithp@keithp.com> + + * config/rx/rx.md (cmpstrnsi): Allow constant length. For + static length 0, just store 0 into the output register. + For dynamic zero, set C/Z appropriately. + (rxcmpstrn): No longer set C/Z. + +2025-04-16 Eric Botcazou <ebotcazou@gcc.gnu.org> + + * tree-ssa-phiopt.cc (factor_out_conditional_operation): Do not + bypass the int_fits_type_p test for boolean types whose precision + is not 1. + +2025-04-16 Sandra Loosemore <sloosemore@baylibre.com> + + * common.opt.urls: Regenerated. + +2025-04-16 Ard Biesheuvel <ardb@kernel.org> + + PR target/119386 + * config/i386/i386-options.cc: Permit -mnop-mcount when + using -fpic with PLTs. + +2025-04-16 Ard Biesheuvel <ardb@kernel.org> + + PR target/119386 + * config/i386/i386.cc (x86_print_call_or_nop): Add @PLT suffix + where appropriate. + (x86_function_profiler): Fall through to x86_print_call_or_nop() + for PIC codegen when flag_plt is set. + +2025-04-16 Sandra Loosemore <sloosemore@baylibre.com> + + PR driver/90465 + * doc/invoke.texi (Overall Options): Add a @cindex for -Q in + connection with --help=. + (Developer Options): Point at --help= documentation for the + other use of -Q. + +2025-04-16 Thomas Schwinge <tschwinge@baylibre.com> + + PR target/97106 + * config/nvptx/nvptx.cc (nvptx_asm_output_def_from_decls) + [ACCEL_COMPILER]: Make sure to emit C++ constructor, destructor + aliases. + +2025-04-16 Jan Hubicka <hubicka@ucw.cz> + + PR tree-optimization/119614 + * ipa-prop.cc (ipa_write_return_summaries): New function. + (ipa_record_return_value_range_1): Break out from .... + (ipa_record_return_value_range): ... here. + (ipa_read_return_summaries): New function. + (ipa_prop_read_section): Read return summaries. + (read_ipcp_transformation_info): Read return summaries. + (ipcp_write_transformation_summaries): Write return summaries; + do not stream stray 0. + +2025-04-16 Tamar Christina <tamar.christina@arm.com> + + PR tree-optimization/119351 + * tree-vectorizer.h (LOOP_VINFO_MASK_NITERS_PFA_OFFSET, + LOOP_VINFO_NON_LINEAR_IV): New. + (class _loop_vec_info): Add mask_skip_niters_pfa_offset and + nonlinear_iv. + * tree-vect-loop.cc (_loop_vec_info::_loop_vec_info): Initialize them. + (vect_analyze_scalar_cycles_1): Record non-linear inductions. + (vectorizable_induction): If early break and PFA using masking create a + new phi which tracks where the scalar code needs to start... + (vectorizable_live_operation): ...and generate the adjustments here. + (vect_use_loop_mask_for_alignment_p): Reject non-linear inductions and + early break needing peeling. + +2025-04-16 Jakub Jelinek <jakub@redhat.com> + + PR middle-end/119808 + * gimple-lower-bitint.cc (gimple_lower_bitint): Don't set + m_single_use_names bits for SSA_NAMEs which have single use but + their SSA_NAME_DEF_STMT is a copy from another SSA_NAME which doesn't + have a single use, or single use which is such a copy etc. + +2025-04-16 Jesse Huang <jesse.huang@sifive.com> + + * config/riscv/riscv.cc (riscv_file_end): Fix .p2align value. + +2025-04-16 Kito Cheng <kito.cheng@sifive.com> + + * config/riscv/riscv.h (JUMP_TABLES_IN_TEXT_SECTION): Check if + large code model. + +2025-04-16 Tejas Belagod <tejas.belagod@arm.com> + + * config/aarch64/aarch64-sve.md (vec_extract<vpred><Vel>): Fix operand + order to gen_vcond_mask_*. + +2025-04-16 Alice Carlotti <alice.carlotti@arm.com> + + * config/aarch64/aarch64.cc + (aarch64_valid_sysreg_name_p): Remove feature check. + (aarch64_retrieve_sysreg): Ditto. + +2025-04-15 Sandra Loosemore <sloosemore@baylibre.com> + + PR tree-optimization/71094 + * doc/invoke.texi (Optimize Options): Document that -fivopts is + enabled at -O1 and higher. Add blurb about -O0 causing GCC to + completely ignore most optimization options. + +2025-04-15 Iain Sandoe <iain@sandoe.co.uk> + + * configure: Regenerate. + * configure.ac: Recognise PROJECT:ld-mmmm.nn.aa as an identifier + for Darwin's static linker. + +2025-04-15 Iain Sandoe <iainsandoe@mini-05-seq.local> + + PR target/116827 + * ginclude/stddef.h: Undefine __PTRDIFF_T and __SIZE_T for module- + enabled c++ on Darwin/macOS platforms. + +2025-04-15 Kyrylo Tkachov <ktkachov@nvidia.com> + + * common.opt.urls: Regenerate. + +2025-04-15 Jan Hubicka <hubicka@ucw.cz> + + * config/i386/x86-tune-sched.cc (ix86_issue_rate): Set + to 4 for znver5. + +2025-04-15 Jan Hubicka <hubicka@ucw.cz> + + PR target/119298 + * config/i386/x86-tune-costs.h (znver5_cost): Set ADDSS cost to 3. + +2025-04-15 Vineet Gupta <vineetg@rivosinc.com> + + PR target/119533 + * config/riscv/riscv-vsetvl.cc (invalid_opt_bb_p): Check for + EDGE_ABNOMAL. + (pre_vsetvl::compute_lcm_local_properties): Initialize kill + bitmap. + Debug dump skipped edge. + +2025-04-15 Robin Dapp <rdapp@ventanamicro.com> + + PR target/119547 + * config/riscv/riscv-vsetvl.cc (pre_vsetvl::earliest_fuse_vsetvl_info): + Do not perform lift if block is not transparent. + +2025-04-15 Kyrylo Tkachov <ktkachov@nvidia.com> + + * Makefile.in (OBJS): Add ipa-locality-cloning.o. + * cgraph.h (set_new_clone_decl_and_node_flags): Declare prototype. + * cgraphclones.cc (set_new_clone_decl_and_node_flags): Remove static + qualifier. + * common.opt (fipa-reorder-for-locality): New flag. + (LTO_PARTITION_DEFAULT): Declare. + (flto-partition): Change default to LTO_PARTITION_DFEAULT. + * doc/invoke.texi: Document -fipa-reorder-for-locality. + * flag-types.h (enum lto_locality_cloning_model): Declare. + (lto_partitioning_model): Add LTO_PARTITION_DEFAULT. + * lto-cgraph.cc (lto_set_symtab_encoder_in_partition): Add dumping of + node and index. + * opts.cc (validate_ipa_reorder_locality_lto_partition): Define. + (finish_options): Handle LTO_PARTITION_DEFAULT. + * params.opt (lto_locality_cloning_model): New enum. + (lto-partition-locality-cloning): New param. + (lto-partition-locality-frequency-cutoff): Likewise. + (lto-partition-locality-size-cutoff): Likewise. + (lto-max-locality-partition): Likewise. + * passes.def: Register pass_ipa_locality_cloning. + * timevar.def (TV_IPA_LC): New timevar. + * tree-pass.h (make_pass_ipa_locality_cloning): Declare. + * ipa-locality-cloning.cc: New file. + * ipa-locality-cloning.h: New file. + +2025-04-15 Martin Jambor <mjambor@suse.cz> + Jakub Jelinek <jakub@redhat.com> + + PR ipa/119803 + * ipa-cp.cc (ipcp_bits_lattice::meet_with_1): Move m_value adjustmed + according to m_mask below the adjustment of the latter according to + cap_mask. Optimize the calculation of cap_mask a bit. + (ipcp_bits_lattice::meet_with): Optimize the calculation of cap_mask a + bit. + +2025-04-15 Jakub Jelinek <jakub@redhat.com> + + * ipa-cp.cc (ipcp_print_widest_int): Print values with all ones in + bits 128+ with "0xf..f" prefix instead of "all ones folled by ". + Simplify wide_int check for -1 or all ones above least significant + 128 bits. + +2025-04-15 Jakub Jelinek <jakub@redhat.com> + + PR sanitizer/119801 + * sanitizer.def (BUILT_IN_TSAN_FUNC_EXIT): Use BT_FN_VOID rather + than BT_FN_VOID_PTR. + * tree-tailcall.cc: Include attribs.h and asan.h. + (struct tailcall): Add has_tsan_func_exit member. + (empty_eh_cleanup): Add eh_has_tsan_func_exit argument, set what + it points to to 1 if there is exactly one __tsan_func_exit call + and ignore that call otherwise. Adjust recursive call. + (find_tail_calls): Add RETRY_TSAN_FUNC_EXIT argument, pass it + to recursive calls. When seeing __tsan_func_exit call with + RETRY_TSAN_FUNC_EXIT 0, set it to -1. If RETRY_TSAN_FUNC_EXIT + is 1, initially ignore __tsan_func_exit calls. Adjust + empty_eh_cleanup caller. When looking through stmts after the call, + ignore exactly one __tsan_func_exit call but remember it in + t->has_tsan_func_exit. Diagnose if EH cleanups didn't have + __tsan_func_exit and normal path did or vice versa. + (optimize_tail_call): Emit __tsan_func_exit before the tail call + or tail recursion. + (tree_optimize_tail_calls_1): Adjust find_tail_calls callers. If + find_tail_calls changes retry_tsan_func_exit to -1, set it to 1 + and call it again with otherwise the same arguments. + +2025-04-15 Sandra Loosemore <sloosemore@baylibre.com> + + PR ipa/113203 + * doc/extend.texi (Common Function Attributes): Explain how to + use always_inline in programs that have multiple translation + units, and that LTO inlining additionally needs optimization + enabled. + +2025-04-15 liuhongt <hongtao.liu@intel.com> + + PR target/108134 + * doc/extend.texi: Remove documents from r11-344-g0fec3f62b9bfc0. + +2025-04-15 Sandra Loosemore <sloosemore@baylibre.com> + + PR target/42683 + * doc/invoke.texi (x86 Options): Clarify that -march=pentiumpro + doesn't include MMX. + +2025-04-14 Thomas Schwinge <tschwinge@baylibre.com> + + PR target/118794 + * config/gcn/gcn.opt (-mfake-exceptions): Support. + * config/nvptx/nvptx.opt (-mfake-exceptions): Likewise. + * config/gcn/gcn.md (define_expand "exception_receiver"): Use it. + * config/nvptx/nvptx.md (define_expand "exception_receiver"): + Likewise. + * config/gcn/mkoffload.cc (main): Set it. + * config/nvptx/mkoffload.cc (main): Likewise. + * config/nvptx/nvptx.cc (nvptx_assemble_integer) + <in_section == exception_section>: Special handling for + 'SYMBOL_REF's. + * except.cc (expand_dw2_landing_pad_for_region): Don't generate + bogus code for (default) + '#define EH_RETURN_DATA_REGNO(N) INVALID_REGNUM'. + +2025-04-14 Jakub Jelinek <jakub@redhat.com> + + PR rtl-optimization/119785 + * expmed.cc (init_expmed): Always pass QImode rather than mode to + set_src_cost passed to set_zero_cost. + +2025-04-14 Jakub Jelinek <jakub@redhat.com> + + PR tree-optimization/119718 + * tree-pretty-print.cc (dump_generic_node) <case CALL_EXPR>: Dump + also CALL_EXPR_MUST_TAIL_CALL flag. + * calls.cc (maybe_complain_about_tail_call): Emit error about + CALL_EXPR_MUST_TAIL_CALL only after emitting dump message, not before + it. + +2025-04-14 Arthur Cohen <arthur.cohen@embecosm.com> + + * doc/install.texi: Add requirements for building gccrs. + +2025-04-14 H.J. Lu <hjl.tools@gmail.com> + + PR target/119784 + * config/i386/i386.cc (ix86_using_red_zone): Don't use red-zone + with 32 GPRs and no caller-saved registers. + +2025-04-14 Martin Jambor <mjambor@suse.cz> + + PR ipa/118097 + * ipa-cp.cc (ipa_get_jf_arith_result): Require res_operand for + anything except NOP_EXPR or ADDR_EXPR, document it and remove the code + trying to deduce it. + (ipa_value_from_jfunc): Use the stored and streamed type of arithmetic + pass-through functions. + (ipa_agg_value_from_jfunc): Use the stored and streamed type of + arithmetic pass-through functions, convert to the type used to store + the value if necessary. + (get_val_across_arith_op): New parameter op_type, pass it to + ipa_get_jf_arith_result. + (propagate_vals_across_arith_jfunc): New parameter op_type, pass it to + get_val_across_arith_op. + (propagate_vals_across_pass_through): Use the stored and streamed type + of arithmetic pass-through functions. + (propagate_aggregate_lattice): Likewise. + (push_agg_values_for_index_from_edge): Use the stored and streamed + type of arithmetic pass-through functions, convert to the type used to + store the value if necessary. + +2025-04-14 Martin Jambor <mjambor@suse.cz> + + PR ipa/118785 + * ipa-cp.cc (ipa_vr_intersect_with_arith_jfunc): Use the stored + and streamed type of arithmetic pass-through functions. + +2025-04-14 Martin Jambor <mjambor@suse.cz> + + * ipa-cp.cc (ipcp_print_widest_int): Also add a truncated form of + dumping of widest ints which only have zeros in the lowest 128 bits. + Update the comment. + (ipcp_bits_lattice::print): Also dump the mask using + ipcp_print_widest_int. + (ipcp_store_vr_results): Likewise. + +2025-04-14 Martin Jambor <mjambor@suse.cz> + + PR ipa/119318 + * ipa-cp.cc (ipcp_bits_lattice::meet_with_1): Set all mask bits + not covered by precision to one. + (ipcp_bits_lattice::meet_with): Likewise. + (propagate_bits_across_jump_function): Use the stored operation + type to perform meet with other lattices. + +2025-04-14 Martin Jambor <mjambor@suse.cz> + + PR ipa/118097 + PR ipa/118785 + PR ipa/119318 + * lto-streamer.h (lto_variably_modified_type_p): Declare. + * ipa-prop.h (ipa_pass_through_data): New field op_type. + (ipa_get_jf_pass_through_op_type): New function. + * ipa-prop.cc: Include lto-streamer.h. + (ipa_dump_jump_function): Dump also pass-through + operation types, if any. Dump pass-through operands only if not NULL. + (ipa_set_jf_simple_pass_through): Set op_type accordingly. + (compute_complex_assign_jump_func): Set op_type of arithmetic + pass-through jump_functions. + (analyze_agg_content_value): Update lhs when walking assighment + copies. Set op_type of aggregate arithmetic pass-through + jump_functions. + (update_jump_functions_after_inlining): Also transfer the operation + type from the source arithmentic pass-through jump function to the + destination jump function. + (ipa_write_jump_function): Stream also the op_type when necessary. + (ipa_read_jump_function): Likewise. + (ipa_agg_pass_through_jf_equivalent_p): Also compare operation types. + * lto-streamer-out.cc (lto_variably_modified_type_p): Make public. + +2025-04-14 Richard Biener <rguenther@suse.de> + + PR tree-optimization/119757 + * tree-vect-slp.cc (vect_build_slp_tree_1): Record and compare + whether a stmt uses a maks. + +2025-04-14 Richard Biener <rguenther@suse.de> + + PR tree-optimization/119778 + * tree-inline.cc (copy_edges_for_bb): Mark calls that are + source of abnormal edges as altering control-flow. + +2025-04-14 Gaius Mulley <gaiusmod2@gmail.com> + + PR modula2/119779 + * doc/gm2.texi (Interface to assembly language): Use eax + rather than rax in both examples. + +2025-04-14 Jakub Jelinek <jakub@redhat.com> + + PR driver/119727 + * configure.ac (HOST_HAS_PERSONALITY_ADDR_NO_RANDOMIZE): New check. + * gcc.cc: Include sys/personality.h if + HOST_HAS_PERSONALITY_ADDR_NO_RANDOMIZE is defined. + (try_generate_repro): Call + personality (personality (0xffffffffU) | ADDR_NO_RANDOMIZE) + if HOST_HAS_PERSONALITY_ADDR_NO_RANDOMIZE is defined. + * config.in: Regenerate. + * configure: Regenerate. + +2025-04-13 Stefan Schulze Frielinghaus <stefansf@gcc.gnu.org> + + * config/s390/s390.cc: Add z17 scheduler description. + * config/s390/s390.h: Ditto. + * config/s390/s390.md: Ditto. + * config/s390/9175.md: New file. + +2025-04-13 Stefan Schulze Frielinghaus <stefansf@gcc.gnu.org> + + * common/config/s390/s390-common.cc: Rename arch15 to z17. + * config.gcc: Add z17. + * config/s390/driver-native.cc: Detect z17 machine. + * config/s390/s390-builtins.def (B_VXE3): Rename arch15 to z17. + * config/s390/s390-c.cc (s390_resolve_overloaded_builtin): Ditto. + * config/s390/s390-opts.h (enum processor_type): Ditto. + * config/s390/s390.cc: Ditto. + * config/s390/s390.h: Ditto. + * config/s390/s390.md: Ditto. + * config/s390/s390.opt: Add z17. + * doc/invoke.texi: Ditto. + +2025-04-12 Sandra Loosemore <sloosemore@baylibre.com> + + PR target/97585 + * doc/invoke.texi (x86 Options): Document list of extensions + supported by -march=x86_64, according to the declaration of + PTA_X86_64_BASELINE in config/i386/i386.h. + +2025-04-12 Jakub Jelinek <jakub@redhat.com> + + PR driver/119727 + * gcc.cc (files_equal_p): Rewritten using fopen/fgets/fclose instead + of open/fstat/read/close. At the start of lines, ignore lowercase + hexadecimal addresses followed by space. + +2025-04-12 Jakub Jelinek <jakub@redhat.com> + + PR tree-optimization/119722 + * gimple-lower-bitint.h (build_bitint_stmt_ssa_conflicts): Add + CLEAR argument. + * gimple-lower-bitint.cc (build_bitint_stmt_ssa_conflicts): Add + CLEAR argument. Call clear on gimple_assign_copy_p rhs1 if lhs + is large/huge bitint unless lhs is not in names. + * tree-ssa-coalesce.cc (build_ssa_conflict_graph): Adjust + build_bitint_stmt_ssa_conflicts caller. Move gimple_assign_copy_p + handling to after the build_bitint_stmt_ssa_conflicts call. + +2025-04-12 Jakub Jelinek <jakub@redhat.com> + + PR tree-optimization/119718 + * tree-tailcall.cc (maybe_error_musttail): Dump the GIMPLE at the + end of the Cannot tail-call line rather than on the line before it. + * calls.cc (maybe_complain_about_tail_call): Dump the GENERIC + at the end of the ;; Cannot tail-call line rather than on the + line before it. + +2025-04-12 Jakub Jelinek <jakub@redhat.com> + + PR tree-optimization/119718 + * tree-tailcall.cc (maybe_error_musttail): Only dump into dump_file + if dump_flags & TDF_DETAILS. Use "Cannot tail-call: " prefix instead + of "Cannot convert: ". + (find_tail_calls, tree_optimize_tail_calls_1): Formatting fixes + for maybe_error_musttail calls. + * calls.cc (maybe_complain_about_tail_call): Emit also a message + into dump_file when dump_flags & TDF_DETAILS for CALL_EXPR_TAILCALL + calls. + (initialize_argument_information): Formatting fix for + maybe_complain_about_tail_call calls. + (can_implement_as_sibling_call_p, expand_call): Likewise. + +2025-04-11 Sandra Loosemore <sloosemore@baylibre.com> + + PR c++/106618 + * doc/invoke.texi (Option Summary): Remove -fargs-in-order, add + -fstrong-eval-order. + (C++ Dialect Options): Explicitly document that -fstrong-eval-order + takes an optional argument and what the choices are. Generalize + references to C++17. + +2025-04-11 Sandra Loosemore <sloosemore@baylibre.com> + + PR middle-end/105548 + * doc/invoke.texi (Optimize Options): Delete misleading sentence + about conversions. + +2025-04-11 Jakub Jelinek <jakub@redhat.com> + + PR tree-optimization/119707 + * gimple-lower-bitint.cc (bitint_large_huge::handle_cast): Only use + m_data[save_data_cnt] instead of m_data[save_data_cnt + 1] if + idx is odd and equal to low + 1. Remember tree_to_uhwi (idx) in + a temporary instead of calling the function multiple times. + +2025-04-11 Sandra Loosemore <sloosemore@baylibre.com> + + PR tree-optimization/87909 + * common.opt.urls: Regenerate. + * doc/invoke.texi (Option Summary): Add -ftree-cselim. + (Optimize Options): Likewise. + +2025-04-11 Sandra Loosemore <sloosemore@baylibre.com> + + PR middle-end/14708 + * doc/invoke.texi (Optimize Options): List -fexcess-precision + before -ffloat-store, moving some background discussion to the + former from the latter. Recommend using -fexcess-precision=standard + instead of -ffloat-store. + +2025-04-10 Iain Sandoe <iain@sandoe.co.uk> + + * config/darwin.h (LINK_SPEC): Add support for + -static-libgcobol. + +2025-04-10 Richard Biener <rguenther@suse.de> + + PR middle-end/119706 + * gimple-expr.cc (is_gimple_mem_ref_addr): Also allow + POLY_INT_CST. + +2025-04-10 Richard Sandiford <richard.sandiford@arm.com> + + PR tree-optimization/119399 + * tree-data-ref.cc (create_waw_or_war_checks): Use a MINUS_EXPR + on two converted pointers, rather than converting a POINTER_DIFF_EXPR + on the pointers. + +2025-04-10 Kito Cheng <kito.cheng@sifive.com> + + * config/riscv/multilib-generator: Remove the compact code model + and check large code model for RV32. + +2025-04-09 Jakub Jelinek <jakub@redhat.com> + + * pretty-print.cc (pretty_printer::format): Use %.Ns instead of + %Ns in function comment. + +2025-04-09 Jakub Jelinek <jakub@redhat.com> + + PR target/119664 + * config/h8300/jumpcall.md (bit test and jump define_insn_and_split): + Use HSI iterator rather than QHSI. + +2025-04-09 Jakub Jelinek <jakub@redhat.com> + + * config/riscv/riscv-vector-builtins.cc (verify_type_context): + Diagnose RVV types for a given OpenMP context. + +2025-04-09 Richard Biener <rguenther@suse.de> + + PR rtl-optimization/119689 + PR rtl-optimization/115568 + * lra-remat.cc (create_cands): Use prev_nonnote_nondebug_insn + to check whether insn2 is directly before insn. + +2025-04-09 Robin Dapp <rdapp@ventanamicro.com> + + PR middle-end/116595 + * expr.cc (categorize_ctor_elements_1): Use + constant_lower_bound. + +2025-04-09 Yang Yujie <yangyujie@loongson.cn> + + * config/loongarch/genopts/gen-evolution.awk: remove + usage of "asort". + * config/loongarch/genopts/genstr.sh: replace sed with awk. + 2025-04-08 Sandra Loosemore <sloosemore@baylibre.com> PR c++/90468 diff --git a/gcc/DATESTAMP b/gcc/DATESTAMP index cb38c2c..2aaf995 100644 --- a/gcc/DATESTAMP +++ b/gcc/DATESTAMP @@ -1 +1 @@ -20250409 +20250421 diff --git a/gcc/Makefile.in b/gcc/Makefile.in index ebfcd8a..55b4cd7 100644 --- a/gcc/Makefile.in +++ b/gcc/Makefile.in @@ -1555,6 +1555,7 @@ OBJS = \ incpath.o \ init-regs.o \ internal-fn.o \ + ipa-locality-cloning.o \ ipa-cp.o \ ipa-sra.o \ ipa-devirt.o \ @@ -3026,6 +3027,7 @@ GTFILES = $(CPPLIB_H) $(srcdir)/input.h $(srcdir)/coretypes.h \ $(srcdir)/ipa-param-manipulation.h $(srcdir)/ipa-sra.cc \ $(srcdir)/ipa-modref.h $(srcdir)/ipa-modref.cc \ $(srcdir)/ipa-modref-tree.h \ + $(srcdir)/ipa-locality-cloning.cc \ $(srcdir)/signop.h \ $(srcdir)/diagnostic-spec.h $(srcdir)/diagnostic-spec.cc \ $(srcdir)/dwarf2out.h \ diff --git a/gcc/ada/ChangeLog b/gcc/ada/ChangeLog index f7a9025..650de8b 100644 --- a/gcc/ada/ChangeLog +++ b/gcc/ada/ChangeLog @@ -1,3 +1,13 @@ +2025-04-17 Jakub Jelinek <jakub@redhat.com> + + * gnatvsn.ads: Bump Library_Version to 16. + +2025-04-12 Eric Botcazou <ebotcazou@adacore.com> + + PR ada/119643 + * sem_ch8.adb (Inherit_Renamed_Profile): Add guard against the + peculiarities of Natural and Positive. + 2025-04-07 Eric Botcazou <ebotcazou@adacore.com> * gcc-interface/utils.cc (convert) <POINTER_TYPE>: Use fold_convert diff --git a/gcc/ada/gnatvsn.ads b/gcc/ada/gnatvsn.ads index 176f88a..ad11c30 100644 --- a/gcc/ada/gnatvsn.ads +++ b/gcc/ada/gnatvsn.ads @@ -32,7 +32,7 @@ package Gnatvsn is -- Static string identifying this version, that can be used as an argument -- to e.g. pragma Ident. - Library_Version : constant String := "15"; + Library_Version : constant String := "16"; -- Library version. It needs to be updated whenever the major version -- number is changed. -- diff --git a/gcc/ada/sem_ch8.adb b/gcc/ada/sem_ch8.adb index d4ab44f..0a9ef41 100644 --- a/gcc/ada/sem_ch8.adb +++ b/gcc/ada/sem_ch8.adb @@ -9314,11 +9314,12 @@ package body Sem_Ch8 is -- If the new type is a renaming of the old one, as is the case -- for actuals in instances, retain its name, to simplify later - -- disambiguation. + -- disambiguation. Beware of Natural and Positive, see Cstand. if Nkind (Parent (New_T)) = N_Subtype_Declaration and then Is_Entity_Name (Subtype_Indication (Parent (New_T))) and then Entity (Subtype_Indication (Parent (New_T))) = Old_T + and then Scope (New_T) /= Standard_Standard then null; else diff --git a/gcc/avoid-store-forwarding.cc b/gcc/avoid-store-forwarding.cc index 34a7bba..ded8d7e 100644 --- a/gcc/avoid-store-forwarding.cc +++ b/gcc/avoid-store-forwarding.cc @@ -238,10 +238,15 @@ process_store_forwarding (vec<store_fwd_info> &stores, rtx_insn *load_insn, { start_sequence (); - rtx ext0 = gen_rtx_ZERO_EXTEND (GET_MODE (dest), it->mov_reg); - if (ext0) + machine_mode dest_mode = GET_MODE (dest); + rtx base_reg = it->mov_reg; + if (known_gt (GET_MODE_BITSIZE (dest_mode), + GET_MODE_BITSIZE (GET_MODE (it->mov_reg)))) + base_reg = gen_rtx_ZERO_EXTEND (dest_mode, it->mov_reg); + + if (base_reg) { - rtx_insn *move0 = emit_move_insn (dest, ext0); + rtx_insn *move0 = emit_move_insn (dest, base_reg); if (recog_memoized (move0) >= 0) { insns = get_insns (); diff --git a/gcc/c/ChangeLog b/gcc/c/ChangeLog index dec9f16..f75a0f6 100644 --- a/gcc/c/ChangeLog +++ b/gcc/c/ChangeLog @@ -1,3 +1,9 @@ +2025-04-15 Qing Zhao <qing.zhao@oracle.com> + + PR c/119717 + * c-typeck.cc (build_access_with_size_for_counted_by): Fully fold the + parameters for call to .ACCESS_WITH_SIZE. + 2025-04-08 Martin Uecker <uecker@tugraz.at> PR c/119612 diff --git a/gcc/c/c-typeck.cc b/gcc/c/c-typeck.cc index 3870e8a..55d896e 100644 --- a/gcc/c/c-typeck.cc +++ b/gcc/c/c-typeck.cc @@ -3013,12 +3013,16 @@ build_access_with_size_for_counted_by (location_t loc, tree ref, gcc_assert (c_flexible_array_member_type_p (TREE_TYPE (ref))); /* The result type of the call is a pointer to the flexible array type. */ tree result_type = c_build_pointer_type (TREE_TYPE (ref)); + tree first_param + = c_fully_fold (array_to_pointer_conversion (loc, ref), false, NULL); + tree second_param + = c_fully_fold (counted_by_ref, false, NULL); tree call = build_call_expr_internal_loc (loc, IFN_ACCESS_WITH_SIZE, result_type, 6, - array_to_pointer_conversion (loc, ref), - counted_by_ref, + first_param, + second_param, build_int_cst (integer_type_node, 1), build_int_cst (counted_by_type, 0), build_int_cst (integer_type_node, -1), diff --git a/gcc/calls.cc b/gcc/calls.cc index b3dccd8..076e046 100644 --- a/gcc/calls.cc +++ b/gcc/calls.cc @@ -1273,11 +1273,19 @@ void maybe_complain_about_tail_call (tree call_expr, const char *reason) { gcc_assert (TREE_CODE (call_expr) == CALL_EXPR); - if (!CALL_EXPR_MUST_TAIL_CALL (call_expr)) - return; - - error_at (EXPR_LOCATION (call_expr), "cannot tail-call: %s", reason); - CALL_EXPR_MUST_TAIL_CALL (call_expr) = 0; + if (CALL_EXPR_TAILCALL (call_expr) + && dump_file + && (dump_flags & TDF_DETAILS)) + { + fprintf (dump_file, ";; Cannot tail-call: %s: ", reason); + print_generic_expr (dump_file, call_expr, TDF_SLIM); + fprintf (dump_file, "\n"); + } + if (CALL_EXPR_MUST_TAIL_CALL (call_expr)) + { + error_at (EXPR_LOCATION (call_expr), "cannot tail-call: %s", reason); + CALL_EXPR_MUST_TAIL_CALL (call_expr) = 0; + } } /* Fill in ARGS_SIZE and ARGS array based on the parameters found in @@ -1447,10 +1455,10 @@ initialize_argument_information (int num_actuals ATTRIBUTE_UNUSED, if (!call_from_thunk_p && DECL_P (base) && !TREE_STATIC (base)) { *may_tailcall = false; - maybe_complain_about_tail_call (exp, - _("a callee-copied argument is" - " stored in the current" - " function's frame")); + maybe_complain_about_tail_call (exp, _("a callee-copied " + "argument is stored " + "in the current " + "function's frame")); } args[i].tree_value = build_fold_addr_expr_loc (loc, @@ -2534,10 +2542,9 @@ can_implement_as_sibling_call_p (tree exp, if (!targetm.have_sibcall_epilogue () && !targetm.emit_epilogue_for_sibcall) { - maybe_complain_about_tail_call - (exp, - _("machine description does not have" - " a sibcall_epilogue instruction pattern")); + maybe_complain_about_tail_call (exp, _("machine description does not " + "have a sibcall_epilogue " + "instruction pattern")); return false; } @@ -2555,9 +2562,8 @@ can_implement_as_sibling_call_p (tree exp, into a sibcall. */ if (!targetm.function_ok_for_sibcall (fndecl, exp)) { - maybe_complain_about_tail_call (exp, - _("target is not able to optimize the" - " call into a sibling call")); + maybe_complain_about_tail_call (exp, _("target is not able to optimize " + "the call into a sibling call")); return false; } @@ -2606,9 +2612,8 @@ can_implement_as_sibling_call_p (tree exp, if (maybe_gt (args_size.constant, crtl->args.size - crtl->args.pretend_args_size)) { - maybe_complain_about_tail_call (exp, - _("callee required more stack slots" - " than the caller")); + maybe_complain_about_tail_call (exp, _("callee required more stack " + "slots than the caller")); return false; } @@ -2621,9 +2626,8 @@ can_implement_as_sibling_call_p (tree exp, (current_function_decl), crtl->args.size))) { - maybe_complain_about_tail_call (exp, - _("inconsistent number of" - " popped arguments")); + maybe_complain_about_tail_call (exp, _("inconsistent number of" + " popped arguments")); return false; } @@ -2685,7 +2689,7 @@ expand_call (tree exp, rtx target, int ignore) so this shouldn't really happen unless the the musttail pass gave up walking before finding the call. */ if (!try_tail_call) - maybe_complain_about_tail_call (exp, _("other reasons")); + maybe_complain_about_tail_call (exp, _("other reasons")); int pass; /* Register in which non-BLKmode value will be returned, @@ -3092,8 +3096,9 @@ expand_call (tree exp, rtx target, int ignore) if (MEM_P (*iter)) { try_tail_call = 0; - maybe_complain_about_tail_call (exp, - _("hidden string length argument passed on stack")); + maybe_complain_about_tail_call (exp, _("hidden string length " + "argument passed on " + "stack")); break; } } @@ -3140,10 +3145,9 @@ expand_call (tree exp, rtx target, int ignore) || partial_subreg_p (caller_mode, callee_mode))))) { try_tail_call = 0; - maybe_complain_about_tail_call (exp, - _("caller and callee disagree in" - " promotion of function" - " return value")); + maybe_complain_about_tail_call (exp, _("caller and callee disagree " + "in promotion of function " + "return value")); } } diff --git a/gcc/cgraph.h b/gcc/cgraph.h index 065fcc7..abde770 100644 --- a/gcc/cgraph.h +++ b/gcc/cgraph.h @@ -2627,6 +2627,7 @@ void tree_function_versioning (tree, tree, vec<ipa_replace_map *, va_gc> *, void dump_callgraph_transformation (const cgraph_node *original, const cgraph_node *clone, const char *suffix); +void set_new_clone_decl_and_node_flags (cgraph_node *new_node); /* In cgraphbuild.cc */ int compute_call_stmt_bb_frequency (tree, basic_block bb); void record_references_in_initializer (tree, bool); diff --git a/gcc/cgraphclones.cc b/gcc/cgraphclones.cc index 5332a43..e6223fa 100644 --- a/gcc/cgraphclones.cc +++ b/gcc/cgraphclones.cc @@ -158,7 +158,7 @@ cgraph_edge::clone (cgraph_node *n, gcall *call_stmt, unsigned stmt_uid, /* Set flags of NEW_NODE and its decl. NEW_NODE is a newly created private clone or its thunk. */ -static void +void set_new_clone_decl_and_node_flags (cgraph_node *new_node) { DECL_EXTERNAL (new_node->decl) = 0; diff --git a/gcc/cobol/ChangeLog b/gcc/cobol/ChangeLog index e386591..27c31c1 100644 --- a/gcc/cobol/ChangeLog +++ b/gcc/cobol/ChangeLog @@ -1,3 +1,82 @@ +2025-04-16 Bob Dubner <rdubner@symas.com> + + PR cobol/119759 + * LICENSE: Deleted. + +2025-04-15 Richard Biener <rguenther@suse.de> + + PR cobol/119302 + * Make-lang.in (GCOBOLIO_INSTALL_NAME): Define. + Use $(GCOBOLIO_INSTALL_NAME) for gcobol.3 manpage source + upon install. + +2025-04-14 Jakub Jelinek <jakub@redhat.com> + + PR cobol/119776 + * lang.opt (fmax-errors): Remove. + * lang.opt.urls: Regenerate. + * cobol1.cc (cobol_langhook_handle_option) <case OPT_fmax_errors>: + Remove. + * gcobol.1: Document -fmax-errors=nerror rather than + -fmax-errors nerror. + +2025-04-14 Jakub Jelinek <jakub@redhat.com> + + PR cobol/119777 + * lang.opt (include): Remove Var(cobol_include). + * cobol1.cc (cobol_langhook_handle_option) <case OPT_include>: Use + arg instead of cobol_include. + +2025-04-14 Jakub Jelinek <jakub@redhat.com> + + PR cobol/119777 + * lang.opt (fsyntax-only): Remove. + * lang.opt.urls: Regenerate. + +2025-04-13 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE> + Simon Sobisch <simonsobisch@gnu.org> + + PR cobol/119217 + * parse.y: Rename OVERFLOW to OVERFLOW_kw. + Specify type name in %token directive. + * scan.l: Likewise. + * token_names.h: Regenerate. + +2025-04-13 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE> + + PR cobol/119217 + * util.cc (class timespec_t): Rename to cbl_timespec. + +2025-04-13 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE> + + * genapi.cc: Include target.h. + (section_label): Use ASM_COMMENT_START. + (paragraph_label): Likewise. + (parser_perform): Likewise. + (internal_perform_through): Likewise. + (hijack_for_development): Likewise. + +2025-04-12 Bob Dubner <rdubner@symas.com> + + PR cobol/119694 + * cbldiag.h: Eliminate getenv() calls. + * cdf.y: Likewise. + * cobol1.cc: Likewise. + * except.cc: Likewise. + * genapi.cc: Likewise. + * lexio.cc: Likewise. + * parse.y: Likewise. + * scan_ante.h: Likewise. + * show_parse.h: Likewise. + * symbols.cc: Likewise. + * symfind.cc: Likewise. + * util.cc: Likewise. + +2025-04-09 Bob Dubner <rdubner@symas.com> + + PR cobol/119682 + * genapi.cc: (cobol_compare): Change the call to __gg__compare(). + 2025-04-08 Jakub Jelinek <jakub@redhat.com> PR cobol/119364 diff --git a/gcc/cobol/LICENSE b/gcc/cobol/LICENSE deleted file mode 100644 index aa5ba60..0000000 --- a/gcc/cobol/LICENSE +++ /dev/null @@ -1,29 +0,0 @@ -######################################################################### -# -# Copyright (c) 2021-2025 Symas Corporation -# -# Redistribution and use in source and binary forms, with or without -# modification, are permitted provided that the following conditions are -# met: -# -# * Redistributions of source code must retain the above copyright -# notice, this list of conditions and the following disclaimer. -# * Redistributions in binary form must reproduce the above -# copyright notice, this list of conditions and the following disclaimer -# in the documentation and/or other materials provided with the -# distribution. -# * Neither the name of the Symas Corporation nor the names of its -# contributors may be used to endorse or promote products derived from -# this software without specific prior written permission. -# -# THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS -# "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT -# LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR -# A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT -# OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, -# SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT -# LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, -# DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY -# THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT -# (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE -# OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. diff --git a/gcc/cobol/Make-lang.in b/gcc/cobol/Make-lang.in index 422ebe2..9b74dd3 100644 --- a/gcc/cobol/Make-lang.in +++ b/gcc/cobol/Make-lang.in @@ -35,6 +35,7 @@ # - define the names for selecting the language in LANGUAGES. GCOBOL_INSTALL_NAME := $(shell echo gcobol|sed '$(program_transform_name)') +GCOBOLIO_INSTALL_NAME := $(shell echo gcobol-io|sed '$(program_transform_name)') GCOBOL_TARGET_INSTALL_NAME := $(target_noncanonical)-$(shell echo gcobol|sed '$(program_transform_name)') GCOBC_INSTALL_NAME := $(shell echo gcobc|sed '$(program_transform_name)') @@ -293,7 +294,7 @@ cobol.install-common: installdirs cobol.install-man: installdirs $(INSTALL_DATA) $(srcdir)/cobol/gcobol.1 $(DESTDIR)$(man1dir)/$(GCOBOL_INSTALL_NAME)$(man1ext) - $(INSTALL_DATA) $(srcdir)/cobol/gcobol.3 $(DESTDIR)$(man3dir)/ + $(INSTALL_DATA) $(srcdir)/cobol/gcobol.3 $(DESTDIR)$(man3dir)/$(GCOBOLIO_INSTALL_NAME)$(man3ext) cobol.install-info: @@ -342,8 +343,8 @@ cobol.uninstall: rm -rf $(DESTDIR)$(bindir)/$(GCOBOL_INSTALL_NAME)$(exeext) \ $(DESTDIR)$(bindir)/$(GCOBC_INSTALL_NAME) \ $(DESTDIR)$(datadir)/gcobol/ \ - $(DESTDIR)$(man1dir)/$(GCOBOL_INSTALL_NAME).1 \ - $(DESTDIR)$(man3dir)/gcobol.3 + $(DESTDIR)$(man1dir)/$(GCOBOL_INSTALL_NAME)$(man1ext) \ + $(DESTDIR)$(man3dir)/$(GCOBOLIO_INSTALL_NAME)$(man3ext) cobol.man: cobol.srcman: diff --git a/gcc/cobol/cbldiag.h b/gcc/cobol/cbldiag.h index ed754f1..d7ee98f 100644 --- a/gcc/cobol/cbldiag.h +++ b/gcc/cobol/cbldiag.h @@ -33,6 +33,12 @@ #else #define _CBLDIAG_H +#if 0 +#define gcobol_getenv(x) getenv(x) +#else +#define gcobol_getenv(x) ((char *)nullptr) +#endif + const char * cobol_filename(); /* @@ -101,7 +107,7 @@ template <typename LOC> static void location_dump( const char func[], int line, const char tag[], const LOC& loc) { extern int yy_flex_debug; - if( yy_flex_debug && getenv("update_location") ) + if( yy_flex_debug && gcobol_getenv("update_location") ) fprintf(stderr, "%s:%d: %s location (%d,%d) to (%d,%d)\n", func, line, tag, loc.first_line, loc.first_column, loc.last_line, loc.last_column); diff --git a/gcc/cobol/cdf.y b/gcc/cobol/cdf.y index 6392f89..e06ccf3 100644 --- a/gcc/cobol/cdf.y +++ b/gcc/cobol/cdf.y @@ -226,7 +226,7 @@ apply_cdf_turn( exception_turns_t& turns ) { turns.location, elem.first, files); } - if( getenv("SHOW_PARSE") ) enabled_exceptions.dump(); + if( getenv("GCOBOL_SHOW") ) enabled_exceptions.dump(); return true; } %} diff --git a/gcc/cobol/cobol1.cc b/gcc/cobol/cobol1.cc index 1e690ff..3bd21c7 100644 --- a/gcc/cobol/cobol1.cc +++ b/gcc/cobol/cobol1.cc @@ -294,7 +294,7 @@ cobol_langhook_init_options_struct (struct gcc_options *opts) { cobol_set_debugging( false, false, false ); - copybook_directory_add( getenv("GCOB_COPYBOOK") ); + copybook_directory_add( getenv("GCOBOL_COPYBOOK") ); } static unsigned int @@ -385,10 +385,6 @@ cobol_langhook_handle_option (size_t scode, return true; } - case OPT_fmax_errors: - flag_max_errors = atoi(arg); - return true; - case OPT_ffixed_form: cobol_set_indicator_column(-7); return true; @@ -413,8 +409,8 @@ cobol_langhook_handle_option (size_t scode, } return true; case OPT_include: - if( ! include_file_add(cobol_include) ) { - cbl_errx( "could not include %s", cobol_include); + if( ! include_file_add(arg) ) { + cbl_errx( "could not include %s", arg); } return true; diff --git a/gcc/cobol/except.cc b/gcc/cobol/except.cc index 1485a33..7a6a922 100644 --- a/gcc/cobol/except.cc +++ b/gcc/cobol/except.cc @@ -312,11 +312,11 @@ file_status_t current_file_handled_status(); void declarative_runtime_match( cbl_field_t *declaratives, cbl_label_t *lave ) { - if( getenv("SHOW_PARSE") ) + if( getenv("GCOBOL_SHOW") ) { fprintf(stderr, "( %d ) %s: \n", cobol_location().first_line, __func__); } - if( getenv("TRACE1") ) + if( getenv("GCOBOL_TRACE") ) { gg_printf(">>>>>>( %d )(%s) declaratives:%s lave:%s\n", build_int_cst_type(INT, cobol_location().first_line), diff --git a/gcc/cobol/gcobol.1 b/gcc/cobol/gcobol.1 index 64c017c..4377c14 100644 --- a/gcc/cobol/gcobol.1 +++ b/gcc/cobol/gcobol.1 @@ -224,7 +224,7 @@ had appeared. Not all exception conditions are implemented. Any that are not produce a warning message. . -.It Fl fmax-errors Ar nerror +.It Fl fmax-errors Ns Li = Ns Ar nerror .Ar nerror represents the number of error messages produced. Without this option, .Nm diff --git a/gcc/cobol/genapi.cc b/gcc/cobol/genapi.cc index fdf76aa..c8911f9 100644 --- a/gcc/cobol/genapi.cc +++ b/gcc/cobol/genapi.cc @@ -34,6 +34,7 @@ #include "tree-iterator.h" #include "stringpool.h" #include "diagnostic-core.h" +#include "target.h" #include "../../libgcobol/ec.h" #include "../../libgcobol/common-defs.h" @@ -75,7 +76,7 @@ static int pseudo_label = 1; static bool suppress_cobol_entry_point = false; static char ach_cobol_entry_point[256] = ""; -bool bSHOW_PARSE = getenv("SHOW_PARSE"); +bool bSHOW_PARSE = getenv("GCOBOL_SHOW"); bool show_parse_sol = true; int show_parse_indent = 0; @@ -198,7 +199,7 @@ trace1_init() trace_handle = gg_define_variable(INT, "trace_handle", vs_static); trace_indent = gg_define_variable(INT, "trace_indent", vs_static); - bTRACE1 = getenv("TRACE1") ? getenv("TRACE1") : gv_trace_switch; + bTRACE1 = getenv("GCOBOL_TRACE") ? getenv("GCOBOL_TRACE") : gv_trace_switch; if( bTRACE1 && strcmp(bTRACE1, "0") != 0 ) { @@ -2357,7 +2358,8 @@ section_label(struct cbl_proc_t *procedure) cbl_label_t *label = procedure->label; // The _initialize_program section isn't relevant. - char *psz = xasprintf("# SECTION %s in %s (%ld)", + char *psz = xasprintf("%s SECTION %s in %s (%ld)", + ASM_COMMENT_START, label->name, current_function->our_unmangled_name, deconflictor); @@ -2408,7 +2410,8 @@ paragraph_label(struct cbl_proc_t *procedure) char *psz1 = xasprintf( - "# PARAGRAPH %s of %s in %s (%ld)", + "%s PARAGRAPH %s of %s in %s (%ld)", + ASM_COMMENT_START, para_name ? para_name: "" , section_name ? section_name: "(null)" , current_function->our_unmangled_name ? current_function->our_unmangled_name: "" , @@ -3006,7 +3009,8 @@ parser_perform(cbl_label_t *label, bool suppress_nexting) para_name = label->name; sect_name = section_label->name; sprintf(ach, - "# PERFORM %s of %s of %s (%ld)", + "%s PERFORM %s of %s of %s (%ld)", + ASM_COMMENT_START, para_name, sect_name, program_name, @@ -3018,7 +3022,8 @@ parser_perform(cbl_label_t *label, bool suppress_nexting) { sect_name = label->name; sprintf(ach, - "# PERFORM %s of %s (%ld)", + "%s PERFORM %s of %s (%ld)", + ASM_COMMENT_START, sect_name, program_name, deconflictor); @@ -3170,8 +3175,8 @@ internal_perform_through( cbl_label_t *proc_1, pseudo_return_push(proc2, return_addr); // Create the code that will launch the first procedure - gg_insert_into_assembler("# PERFORM %s THROUGH %s", - proc_1->name, proc_2->name); + gg_insert_into_assembler("%s PERFORM %s THROUGH %s", + ASM_COMMENT_START, proc_1->name, proc_2->name); if( !suppress_nexting ) { @@ -6632,22 +6637,6 @@ parser_division(cbl_division_t division, } gg_assign(base, gg_cast(UCHAR_P, parameter)); - IF( gg_call_expr( CHAR_P, - "getenv", - gg_string_literal("PARAMETERS_ON_ENTRY"), - NULL_TREE), - ne_op, - gg_cast(CHAR_P, null_pointer_node)); - { - gg_printf("parameter_on_entry: %s(): %d %p\n", - gg_string_literal(current_function->our_unmangled_name), - build_int_cst_type(INT, i+1), - base, - NULL_TREE); - } - ELSE - ENDIF - if( args[i].refer.field->attr & any_length_e ) { // gg_printf("side channel: Length of \"%s\" is %ld\n", @@ -13622,7 +13611,7 @@ hijack_for_development(const char *funcname) // Assume that funcname is lowercase with no hyphens enter_program_common(funcname, funcname); parser_display_literal("You have been hijacked by a program named \"dubner\""); - gg_insert_into_assembler("# HIJACKED DUBNER CODE START"); + gg_insert_into_assembler("%s HIJACKED DUBNER CODE START", ASM_COMMENT_START); for(int i=0; i<10; i++) { @@ -13635,7 +13624,7 @@ hijack_for_development(const char *funcname) NULL_TREE); } - gg_insert_into_assembler("# HIJACKED DUBNER CODE END"); + gg_insert_into_assembler("%s HIJACKED DUBNER CODE END", ASM_COMMENT_START); gg_return(0); } @@ -15871,38 +15860,6 @@ psa_global(cbl_field_t *new_var) sprintf(ach, "__gg__%s", mname); free(mname); - if( getenv("SHOW_GLOBAL_VARIABLES") ) - { - char ach_type[32]; - strcpy(ach_type, cbl_field_type_str(new_var->type)); - - fprintf(stderr, "struct cblc_field_t %s = {\n", ach); - fprintf(stderr, " .data = NULL ,\n" ); - fprintf(stderr, " .capacity = %d ,\n", new_var->data.capacity ); - fprintf(stderr, " .offset = %ld ,\n" , new_var->offset ); - fprintf(stderr, " .name = \"%s\" ,\n" , new_var->name ); - fprintf(stderr, " .picture = \"%s\" ,\n" , new_var->data.picture ? new_var->data.picture : "" ); - if( new_var->data.initial || new_var->type == FldPointer ) - { - fprintf(stderr, " .initial = \"%s\" ,\n" , new_var->data.picture ? new_var->data.picture : "" ); - } - else - { - fprintf(stderr, " .initial = NULL ,\n" ); - } - fprintf(stderr, " .parent = NULL,\n" ); - fprintf(stderr, " .depending_on = NULL ,\n" ); - fprintf(stderr, " .depends_on = NULL ,\n" ); - fprintf(stderr, " .occurs_lower = 0 ,\n" ); - fprintf(stderr, " .occurs_upper = 0 ,\n" ); - fprintf(stderr, " .attr = 0x%lx ,\n" , new_var->attr ); - fprintf(stderr, " .type = %s ,\n" , ach_type); - fprintf(stderr, " .level = %d ,\n" , new_var->level ); - fprintf(stderr, " .digits = %d ,\n" , new_var->data.digits ); - fprintf(stderr, " .rdigits = %d ,\n" , new_var->data.rdigits ); - fprintf(stderr, " };\n"); - } - if( strcmp(new_var->name, "_VERY_TRUE") == 0 ) { new_var->var_decl_node = boolean_true_node; diff --git a/gcc/cobol/lang.opt b/gcc/cobol/lang.opt index 42c4020..59278a1 100644 --- a/gcc/cobol/lang.opt +++ b/gcc/cobol/lang.opt @@ -77,10 +77,6 @@ ffixed-form Cobol RejectNegative Assume that the source file is fixed form. -fsyntax-only -Cobol RejectNegative -; Documented in c.opt - ffree-form Cobol RejectNegative Assume that the source file is free form. @@ -93,10 +89,6 @@ finternal-ebcdic Cobol Var(cobol_ebcdic, 1) Init(0) -finternal-ebcdic Internal processing is in EBCDIC Code Page 1140 -fmax-errors -Cobol Joined Separate -; Documented in C - fstatic-call Cobol Var(cobol_static_call, 1) Init(1) Enable/disable static linkage for CALL literals @@ -118,7 +110,7 @@ Cobol Joined Separate ; Documented in C include -Cobol Joined Separate Var(cobol_include) +Cobol Joined Separate ; Documented in C isysroot diff --git a/gcc/cobol/lang.opt.urls b/gcc/cobol/lang.opt.urls index 6a5dc1c..69f5297 100644 --- a/gcc/cobol/lang.opt.urls +++ b/gcc/cobol/lang.opt.urls @@ -13,15 +13,9 @@ UrlSuffix(gcc/Directory-Options.html#index-I) LangUrlSuffix_D(gdc/Directory-Opti ffixed-form LangUrlSuffix_Fortran(gfortran/Fortran-Dialect-Options.html#index-ffixed-form) -fsyntax-only -UrlSuffix(gcc/Warning-Options.html#index-fsyntax-only) LangUrlSuffix_D(gdc/Warnings.html#index-fno-syntax-only) LangUrlSuffix_Fortran(gfortran/Error-and-Warning-Options.html#index-fsyntax-only) - ffree-form LangUrlSuffix_Fortran(gfortran/Fortran-Dialect-Options.html#index-ffree-form) -fmax-errors -UrlSuffix(gcc/Warning-Options.html#index-fmax-errors) LangUrlSuffix_D(gdc/Warnings.html#index-fmax-errors) - iprefix UrlSuffix(gcc/Directory-Options.html#index-iprefix) LangUrlSuffix_D(gdc/Directory-Options.html#index-iprefix) LangUrlSuffix_Fortran(gfortran/Preprocessing-Options.html#index-iprefix) diff --git a/gcc/cobol/lexio.cc b/gcc/cobol/lexio.cc index 82bacf2..afe3725 100644 --- a/gcc/cobol/lexio.cc +++ b/gcc/cobol/lexio.cc @@ -380,7 +380,9 @@ struct buffer_t : public bytespan_t { dbgmsg("flex input buffer: '%.*s'\n[xelf]", int(pos - data), data); } void dump() const { +#ifdef GETENV_OK if( getenv("lexer_input") ) show(); +#endif } }; @@ -457,11 +459,11 @@ update_yylloc( const csub_match& stmt, const csub_match& term ) { class dump_loc_on_exit { public: dump_loc_on_exit() { - if( getenv( "update_yylloc" ) ) + if( gcobol_getenv( "update_yylloc" ) ) location_dump( "update_yylloc", __LINE__, "begin", yylloc); } ~dump_loc_on_exit() { - if( getenv( "update_yylloc" ) ) + if( gcobol_getenv( "update_yylloc" ) ) location_dump( "update_yylloc", __LINE__, "end ", yylloc); } } dloe; diff --git a/gcc/cobol/parse.y b/gcc/cobol/parse.y index 61ffa7c..55c26fe 100644 --- a/gcc/cobol/parse.y +++ b/gcc/cobol/parse.y @@ -338,7 +338,7 @@ %token <number> INVALID %token <number> NUMBER NEGATIVE %token <numstr> NUMSTR "numeric literal" -%token <number> OVERFLOW +%token <number> OVERFLOW_kw "OVERFLOW" %token <computational> COMPUTATIONAL %token <boolean> PERFORM BACKWARD @@ -997,7 +997,7 @@ DELETE DISPLAY DIVIDE EVALUATE END EOP EXIT FILLER_kw GOBACK GOTO INITIALIZE INSPECT - MERGE MOVE MULTIPLY OPEN OVERFLOW PARAGRAPH PERFORM + MERGE MOVE MULTIPLY OPEN OVERFLOW_kw PARAGRAPH PERFORM READ RELEASE RETURN REWRITE SEARCH SET SELECT SORT SORT_MERGE STRING_kw STOP SUBTRACT START @@ -3306,13 +3306,6 @@ level_name: LEVEL ctx_name data_descr: data_descr1 { $$ = current_field($1); // make available for occurs, etc. - char *env = getenv("symbols_update"); - if( env && env[0] == 'P' ) { - dbgmsg("parse.y:%d: %-15s %s (%s)", __LINE__, - cbl_field_type_str($$->type) + 3, - field_str($$), - cbl_field_type_str($$->usage) + 3); - } } | error { static cbl_field_t none = {}; $$ = &none; } ; @@ -9495,7 +9488,7 @@ call_except: EXCEPTION std::swap($$.on_error, $$.not_error); } } - | OVERFLOW + | OVERFLOW_kw { $$.not_error = NULL; $$.on_error = label_add(LblArith, @@ -9503,7 +9496,7 @@ call_except: EXCEPTION if( !$$.on_error ) YYERROR; parser_call_exception( $$.on_error ); - assert( $1 == OVERFLOW || $1 == NOT ); + assert( $1 == OVERFLOW_kw || $1 == NOT ); if( $1 == NOT ) { std::swap($$.on_error, $$.not_error); } @@ -9758,7 +9751,7 @@ on_overflows: on_overflow[over] statements %prec ADD } ; -on_overflow: OVERFLOW +on_overflow: OVERFLOW_kw { $$.not_error = NULL; $$.on_error = label_add(LblString, @@ -9766,7 +9759,7 @@ on_overflow: OVERFLOW if( !$$.on_error ) YYERROR; parser_string_overflow( $$.on_error ); - assert( $1 == OVERFLOW || $1 == NOT ); + assert( $1 == OVERFLOW_kw || $1 == NOT ); if( $1 == NOT ) { std::swap($$.on_error, $$.not_error); } @@ -11078,23 +11071,6 @@ void ast_call( const YYLTYPE& loc, cbl_refer_t name, cbl_refer_t returning, parser_symbol_add(name.field); } - if( getenv("ast_call") ) { - dbgmsg("%s: calling %s returning %s with %zu args:", __func__, - name_of(name.field), - (returning.field)? returning.field->name : "[none]", - narg); - for( size_t i=0; i < narg; i++ ) { - const char *crv = "?"; - switch(args[i].crv) { - case by_default_e: crv = "def"; break; - case by_reference_e: crv = "ref"; break; - case by_content_e: crv = "con"; break; - case by_value_e: crv = "val"; break; - } - dbgmsg("%s: %4zu: %s @%p %s", __func__, - i, crv, args[i].refer.field, args[i].refer.field->name); - } - } parser_call( name, returning, narg, args, except, not_except, is_function ); } @@ -11403,11 +11379,6 @@ label_add( const YYLTYPE& loc, assert( !(p->type == LblSection && p->parent > 0) ); - if( getenv(__func__) ) { - yywarn("%s: added label %3zu %10s for '%s' of %zu", __func__, - symbol_elem_of(p) - symbols_begin(), p->type_str()+3, p->name, p->parent); - } - return p; } @@ -11468,20 +11439,12 @@ paragraph_reference( const char name[], size_t section ) strcpy(label.name, name); if( label.type == LblNone ) assert(label.parent == 0); - const symbol_elem_t *last = symbols_end(); - p = symbol_label_add(PROGRAM, &label); assert(p); const char *sect_name = section? cbl_label_of(symbol_at(section))->name : NULL; procedure_reference_add(sect_name, p->name, yylineno, current.program_section()); - if( getenv(__func__) ) { - yywarn("%s: %s label %3zu %10s for '%s' of %zu", __func__, - symbols_end() == last? "added" : "found", - symbol_index(symbol_elem_of(p)), p->type_str()+3, p->name, p->parent); - } - return p; } @@ -11675,10 +11638,6 @@ ast_add( arith_t *arith ) { pC = use_any(arith->tgts, C); pA = use_any(arith->A, A); - if( getenv(__func__) ) { - dbgmsg("%s:%d: %-12s C{%zu %p} A{%zu %p}", __func__, __LINE__, - arith->format_str(), nC, pC, nA, pA ); - } parser_add( nC, pC, nA, pA, arith->format, arith->on_error, arith->not_error ); ec_type_t handled = arith->on_error || arith->not_error ? ec_size_e : ec_none_e; @@ -11780,9 +11739,6 @@ stringify( refer_collection_t *inputs, } assert( inputs->lists.back().marker ); std::copy( inputs->lists.begin(), inputs->lists.end(), sources.begin() ); - if( yydebug && getenv(__func__) ) { - std::for_each(sources.begin(), sources.end(), stringify_src_t::dump); - } parser_string( into, pointer, sources.size(), sources.data(), on_error, not_error ); } @@ -12227,9 +12183,6 @@ initialize_one( cbl_num_result_t target, bool with_filler, } else { parser_move(tgt, src, current_rounded_mode()); } - if( getenv(__func__) ) { - yywarn("%s:%-5s: %s", __func__, keyword_str(token), field_str(tgt.field)); - } return true; } @@ -12246,10 +12199,6 @@ initialize_one( cbl_num_result_t target, bool with_filler, parser_initialize(tgt); } } - - if( getenv(__func__) ) { - yywarn("%s: value: %s", __func__, field_str(tgt.field)); - } } // apply REPLACING, possibly overwriting VALUE @@ -12262,75 +12211,15 @@ initialize_one( cbl_num_result_t target, bool with_filler, if( r != replacements.end() ) { parser_move( tgt, *r->second ); - if( getenv(__func__) ) { - cbl_field_t *from = r->second->field; - char from_str[128]; // copy static buffer from field_str - strcpy( from_str, field_str(from) ); - yywarn("%s: move: %-18s %s \n\t from %-18s %s", __func__, - cbl_field_type_str(tgt.field->type) + 3, field_str(tgt.field), - cbl_field_type_str(from->type) + 3, from_str); - } return true; } return true; - } typedef std::pair<cbl_field_t*,cbl_field_t*> field_span_t; typedef std::pair<size_t, size_t> cbl_bytespan_t; -static void -dump_spans( size_t isym, - const cbl_field_t *table, - const std::list<field_span_t>& spans, - size_t nrange, - const cbl_bytespan_t ranges[], - size_t depth, - const std::list<cbl_subtable_t>& subtables ) -{ - int i=0; - assert( nrange == 0 || nrange == spans.size() ); - - if( isym != field_index(table) ) { - dbgmsg("%s:%d: isym %zu is not #%zu %02u %s", __func__, __LINE__, - isym, field_index(table), table->level, table->name); - } - dbgmsg( "%s: [%zu] #%zu %s has %zu spans and %zu subtables", - __func__, depth, isym, table->name, nrange, subtables.size() ); - for( auto span : spans ) { - unsigned int last_level = 0; - const char *last_name = "<none>"; - if( span.second ) { - last_level = span.second->level; - last_name = span.second->name; - } - - char at_subtable[64] = {}; - size_t offset = nrange? ranges[i].first : 0; - auto p = std::find_if(subtables.begin(), subtables.end(), - [offset]( const cbl_subtable_t& tbl ) { - return tbl.offset == offset; - }); - if( p != subtables.end() ) { - sprintf(at_subtable, "(subtable #%zu)", p->isym); - } - dbgmsg("\t %02u %-20s to %02u %-20s: %3zu-%zu %s", - span.first->level, span.first->name, - last_level, last_name, - nrange? ranges[i].first : 1, - nrange? ranges[i].second : 0, - at_subtable); - i++; - } - if( ! subtables.empty() ) { - dbgmsg("\ttable #%zu has %zu subtables", isym, subtables.size()); - for( auto tbl : subtables ) { - dbgmsg("\t #%zu @ %4zu", tbl.isym, tbl.offset); - } - } -} - /* * After the 1st record is initialized, copy it to the others. */ @@ -12339,9 +12228,6 @@ initialize_table( cbl_num_result_t target, size_t nspan, const cbl_bytespan_t spans[], const std::list<cbl_subtable_t>& subtables ) { - if( getenv("initialize_statement") ) { - dbgmsg("%s:%d: %s ", __func__, __LINE__, target.refer.str()); - } assert( target.refer.nsubscript == dimensions(target.refer.field) ); const cbl_refer_t& src( target.refer ); size_t n( src.field->occurs.ntimes()); @@ -12391,12 +12277,6 @@ initialize_statement( const cbl_num_result_t& target, bool with_filler, const category_map_t& replacements, size_t depth = 0 ) { - if( getenv(__func__) ) { - dbgmsg("%s:%d: %2zu: %s (%s%zuR)", - __func__, __LINE__, depth, target.refer.str(), - with_filler? "F" : "", - replacements.size()); - } const cbl_refer_t& tgt( target.refer ); assert(dimensions(tgt.field) == tgt.nsubscript || 0 < depth); assert(!is_literal(tgt.field)); @@ -12480,10 +12360,6 @@ initialize_statement( const cbl_num_result_t& target, bool with_filler, return std::make_pair(first, second); } ); } - if( getenv("initialize_statement") ) { - dump_spans( field_index(output.refer.field), output.refer.field, - field_spans, ranges.size(), ranges.data(), depth, subtables ); - } return initialize_table( output, nrange, ranges.data(), subtables ); } } @@ -12550,18 +12426,6 @@ static void initialize_statement( std::list<cbl_num_result_t>& tgts, bool with_filler, data_category_t value_category, const category_map_t& replacements) { - if( yydebug && getenv(__func__) ) { - yywarn( "%s: %zu targets, %s filler", - __func__, tgts.size(), with_filler? "with" : "no"); - for( auto tgt : tgts ) { - fprintf( stderr, "%28s: %s\n", __func__, name_of(tgt.refer.field) ); - } - for( const auto& elem : replacements ) { - fprintf( stderr, "%28s: %s <-%s\n", __func__, - data_category_str(elem.first), - name_of(elem.second->field) ); - } - } bool is_refmod = std::any_of( tgts.begin(), tgts.end(), []( const auto& tgt ) { diff --git a/gcc/cobol/scan.l b/gcc/cobol/scan.l index 2cb7d30..e30634d 100644 --- a/gcc/cobol/scan.l +++ b/gcc/cobol/scan.l @@ -1543,9 +1543,9 @@ USE({SPC}FOR)? { return USE; } NOT{SPC}(ON{SPC})?EXCEPTION { yylval.number = NOT; return EXCEPTION; } - (ON{SPC})?OVERFLOW { yylval.number = OVERFLOW; return OVERFLOW; } + (ON{SPC})?OVERFLOW { yylval.number = OVERFLOW_kw; return OVERFLOW_kw; } NOT{SPC}(ON{SPC})?OVERFLOW { - yylval.number = NOT; return OVERFLOW; } + yylval.number = NOT; return OVERFLOW_kw; } (AT{SPC})?END/[[:space:]] { yylval.number = END; return END; } @@ -2312,7 +2312,7 @@ BASIS { yy_push_state(basis); return BASIS; } ORGANIZATION { return ORGANIZATION; } OTHER { return OTHER; } OUTPUT { return OUTPUT; } - OVERFLOW { return OVERFLOW; } + OVERFLOW { return OVERFLOW_kw; } OVERRIDE { return OVERRIDE; } PACKED-DECIMAL { return PACKED_DECIMAL; } PAGE { return PAGE; } diff --git a/gcc/cobol/scan_ante.h b/gcc/cobol/scan_ante.h index b9bbd30..cfeacfc 100644 --- a/gcc/cobol/scan_ante.h +++ b/gcc/cobol/scan_ante.h @@ -694,9 +694,6 @@ tmpstring_append( int len ) { const char *extant = tmpstring == NULL ? "" : tmpstring; char *s = xasprintf("%s%.*s", extant, len, yytext); free(tmpstring); - if( yy_flex_debug && getenv(__func__) ) { - yywarn("%s: value is now '%s'", __func__, s); - } return tmpstring = s; } diff --git a/gcc/cobol/show_parse.h b/gcc/cobol/show_parse.h index ad26584..9b1abb4 100644 --- a/gcc/cobol/show_parse.h +++ b/gcc/cobol/show_parse.h @@ -42,9 +42,6 @@ // SHOW_PARSE must be followed by a bracketed set of instructions, no semicolon -// This construction isn't really necessary; getenv() apparently runs pretty -// fast. But using makes compiling a large number of programs just perceptably -// quicker. So, I am using it; it's cheap. extern bool bSHOW_PARSE; extern bool show_parse_sol; extern int show_parse_indent; diff --git a/gcc/cobol/symbols.cc b/gcc/cobol/symbols.cc index 4067459..ddb8e68 100644 --- a/gcc/cobol/symbols.cc +++ b/gcc/cobol/symbols.cc @@ -369,12 +369,6 @@ special_pair_cmp( const cbl_special_name_t& key, const cbl_special_name_t& elem ) { const bool matched = key.id == elem.id || 0 == strcasecmp(key.name, elem.name); - if( getenv(__func__) ) { - dbgmsg("%s:%d: key: id=%2d, %s", __func__, __LINE__, key.id, key.name); - dbgmsg("%s:%d: elem: id=%2d, %s => %s", __func__, __LINE__, - elem.id, elem.name, matched? "match" : "no match"); - } - return matched; } @@ -893,13 +887,6 @@ update_block_offsets( struct symbol_elem_t *block) uint32_t offset = cbl_field_of(block)->offset; const uint32_t block_level = cbl_field_of(block)->level; - if( getenv(__func__) ) { - cbl_field_t *field = cbl_field_of(block); - dbgmsg( "%s: offset is %3zu for %2u %-30s #%3zu P%zu", - __func__, field->offset, field->level, field->name, - symbol_index(block), field->parent ); - } - struct symbol_elem_t *e = block; for( ++e; e < symbols_end(); e++ ) { if( e->type != SymField ) { @@ -929,12 +916,6 @@ update_block_offsets( struct symbol_elem_t *block) offset += field_memsize(field); } - if( getenv(__func__) ) { - dbgmsg( "%s: offset is %3zu for %2u %-30s #%3zu P%zu", - __func__, field->offset, field->level, field->name, - symbol_index(e), field->parent ); - } - if( field->type == FldGroup ) { e = update_block_offsets(e) - 1; } @@ -1051,7 +1032,6 @@ symbol_find_odo_debug( cbl_field_t * field ) { // Return OCCURS DEPENDING ON table subordinate to field, if any. struct cbl_field_t * symbol_find_odo( cbl_field_t * field ) { - if( getenv(__func__) ) return symbol_find_odo_debug(field); size_t bog = field_index(field), eog = end_of_group(bog); auto e = std::find_if( symbol_at(bog), symbol_at_impl(eog, true), has_odo ); return e == symbol_at_impl(eog, true)? NULL : cbl_field_of(e); @@ -1288,10 +1268,6 @@ static struct symbol_elem_t * // Print accumulating details for one group to debug log. bool details = false; - if( yydebug ) { - const auto details_for = getenv("symbols_update"); - details = details_for && 0 == strcasecmp(details_for, group->name); - } // At end of group, members is a list of all immediate children, any // of which might have been redefined and so acquired a memsize. @@ -1363,23 +1339,6 @@ verify_block( const struct symbol_elem_t *block, if( e->type != SymField ) { continue; } - const struct cbl_field_t *field = cbl_field_of(e); - - if( getenv(__func__) ) { - if( e == block ) { - static const char ds[] = "--------------------------------"; - dbgmsg( "%17s %-3s %-3s %-18s %-3s %3s %-16s C/D/R = init\n" - "%.25s %-.3s %-.3s %-.18s %-.3s %.3s %-.16s %-.7s %-.16s", - "", "ndx", "off", "type", "par", "lvl", "name", - ds, ds, ds, ds, ds, ds, ds, ds, ds ); - } - dbgmsg( "%s:%d: %3zu %3zu %-18s %3zu %02d %-16s %2u/%u/%d = '%s'", - __func__, __LINE__, e - symbols.elems, field->offset, - cbl_field_type_str(field->type), - field->parent, field->level, field->name, - field->data.capacity, field->data.digits, field->data.rdigits, - field->data.initial? field->data.initial : "(none)" ); - } } } @@ -1694,6 +1653,9 @@ operator<<( std::ostream& os, const cbl_occurs_bounds_t& bound ) { return os << bound.lower << ',' << bound.upper; } +#pragma GCC diagnostic push +#pragma GCC diagnostic ignored "-Wunused-function" +// Keep this debugging function around for when it is needed static std::ostream& operator<<( std::ostream& os, const cbl_field_data_t& field ) { return os << field.memsize << ',' @@ -1717,16 +1679,7 @@ operator<<( std::ostream& os, const cbl_field_t& field ) { << ',' << field.line << ',' << field.data; } - -static void -write_field_csv( size_t isym, const cbl_field_t *field ) { - static std::ofstream os( getenv("GCOBOL_DATA") ); - assert(os.is_open()); - - if( symbols.first_program < isym) { - os << isym << "," << *field << std::endl; - } -} +#pragma GCC diagnostic pop static std::map<size_t, std::set<size_t>> same_record_areas; size_t parse_error_count(); @@ -1747,11 +1700,6 @@ symbols_update( size_t first, bool parsed_ok ) { struct symbol_elem_t *p, *pend; std::list<cbl_field_t*> shared_record_areas; - if( getenv(__func__) ) { - fprintf(stderr, "Initial"); - symbols_dump(std::max(first, symbols.first_program), true); - } - for( p = symbols_begin(first); p < symbols_end(); p++ ) { if( p->type == SymAlphabet ) continue; // Alphabets already processed. @@ -1796,10 +1744,6 @@ symbols_update( size_t first, bool parsed_ok ) { // no special processing for other levels } - if( getenv("GCOBOL_DATA") ) { - write_field_csv( p - symbols_begin(), field ); - } - // Update ODO field in situ. if( is_table(field) ) { size_t& odo = field->occurs.depending_on; @@ -1869,11 +1813,6 @@ symbols_update( size_t first, bool parsed_ok ) { assert( !(field->data.memsize > 0 && symbol_explicitly_redefines(field)) ); } - if( getenv(__func__) ) { - fprintf(stderr, "Pre"); - symbols_dump(std::max(first, symbols.first_program), true); - } - // A shared record area has no 01 child because that child redefines its parent. for( auto sharer : shared_record_areas ) { auto redefined = cbl_field_of(symbol_at(sharer->parent)); @@ -2391,8 +2330,6 @@ symbol_table_init(void) { symbols.registers.return_code = symbol_index(symbol_field(0,0, "RETURN-CODE")); symbols.registers.very_true = symbol_index(symbol_field(0,0, "_VERY_TRUE")); symbols.registers.very_false = symbol_index(symbol_field(0,0, "_VERY_FALSE")); - - if( getenv(__func__) ) symbols_dump(0, true); } /* @@ -2589,26 +2526,6 @@ symbol_field_add( size_t program, struct cbl_field_t *field ) } } - char *s; - if( (s = getenv(__func__)) != NULL ) { - if( s[0] == 'D' ) { - for( struct symbol_elem_t *e = symbols_begin(); e < symbols_end(); e++ ) { - fprintf(stderr, "%zu: %s ", e - symbols.elems, symbol_type_str(e->type)); - if( e->type == SymField ) { - fprintf(stderr, "%s = %s", - cbl_field_of(e)->name, cbl_field_of(e)->data.initial); - } - fprintf(stderr, "\n"); - } - } - - dbgmsg( "%s:%d: %3zu %-18s %02d %-16s %u/%u/%d = '%s'", __func__, __LINE__, - field->offset, - cbl_field_type_str(field->type), field->level, field->name, - field->data.capacity, field->data.digits, field->data.rdigits, - field->data.initial? field->data.initial : "(none)" ); - } - if( is_forward(field) ) { auto *e = symbol_field( program, field->parent, field->name ); if( e ) { @@ -3120,12 +3037,6 @@ symbol_file_record_sizes( struct cbl_file_t *file ) { output.min = cbl_field_of(&*p.first)->data.capacity; output.max = cbl_field_of(&*p.second)->data.capacity; - if( yydebug && getenv(__func__) ) { - dbgmsg("%s: %s: min '%s' %zu, max '%s' %zu", __func__, file->name, - cbl_field_of(&*p.first)->name, output.min, - cbl_field_of(&*p.second)->name, output.max); - } - assert(output.min > 0 && "min record size is 0"); assert(output.min <= output.max); @@ -3304,10 +3215,6 @@ new_temporary_impl( enum cbl_field_type_t type ) snprintf(f->name, sizeof(f->name), "_literal%d",++nliteral); } else { snprintf(f->name, sizeof(f->name), "_stack%d",++nstack); - - if( getenv("symbol_temporaries_free") ) { - dbgmsg("%s: %s, %s", __func__, f->name, 3 + cbl_field_type_str(f->type)); - } } return f; @@ -3400,14 +3307,6 @@ temporaries_t::dump() const { } temporaries_t::~temporaries_t() { - if( getenv( "symbol_temporaries_free" ) ) { - dbgmsg("%s: %zu literals", __func__, literals.size()); - for( const auto& elem : literals ) { - const literal_an& key(elem.first); - fprintf(stderr, "%c '%s'\n", key.is_quoted? 'Q' : ' ', key.value.c_str()); - } - dump(); - } } cbl_field_t * @@ -3451,7 +3350,6 @@ temporaries_t::acquire( cbl_field_type_t type ) { void symbol_temporaries_free() { - if( getenv(__func__) ) temporaries.dump(); for( auto& elem : temporaries.used ) { const cbl_field_type_t& type(elem.first); temporaries_t::fieldset_t& used(elem.second); @@ -3599,9 +3497,6 @@ cbl_field_t::internalize() { yywarn("failed iconv_open tocode = '%s' fromcode = %s", tocode, fromcode); } - // Sat Mar 16 11:45:08 2024: require temporary environment for testing - if( getenv( "INTERNALIZE_NO") ) return data.initial; - bool using_assumed = fromcode == os_locale.assumed; if( fromcode == tocode || has_attr(hex_encoded_e) ) { @@ -3649,16 +3544,6 @@ cbl_field_t::internalize() { if( 0 != memcmp(data.initial, output.data(), out - output.data()) ) { assert(out <= output.data() + data.capacity); - if( getenv(__func__) ) { - const char *eoi = data.initial + data.capacity, *p; - char nullitude[64] = "no null"; - if( (p = std::find(data.initial, eoi, '\0')) != eoi ) { - sprintf(nullitude, "NUL @ %zu", p - data.initial); - } - dbgmsg("%s:%d: before: %-15s %-20s: '%.*s'{%u}, %s", __func__, __LINE__, - 3 + cbl_field_type_str(type), name, - data.capacity, data.initial, data.capacity, nullitude); - } dbgmsg("%s: converted '%.*s' to %s", __func__, data.capacity, data.initial, tocode); @@ -3677,18 +3562,6 @@ cbl_field_t::internalize() { free(const_cast<char*>(data.initial)); data.initial = mem; - - if( getenv(__func__) ) { - const char *eoi = data.initial + data.capacity, *p; - char nullitude[64] = "no null"; - if( (p = std::find(data.initial, eoi, '\0')) != eoi ) { - sprintf(nullitude, "NUL @ %zu", p - data.initial); - } - dbgmsg("%s:%d: after: %-15s %-20s: '%.*s'{%u}, %s", __func__, __LINE__, - "", name, - data.capacity, data.initial, data.capacity, nullitude); - } - } return data.initial; @@ -3808,37 +3681,14 @@ common_callables_update( const size_t iprog ) { cbl_label_t * symbol_label_add( size_t program, cbl_label_t *input ) { - if( getenv(__func__) ) { - const cbl_label_t *L = input; - dbgmsg( "%s:%d: %-5s #%3zu %-9s '%s' of '%s' at line %d", __func__, __LINE__, - "input", - size_t(0), - L->type_str()+3, - L->name, - L->parent? cbl_label_of(symbol_at(L->parent))->name : "", - L->line ); - } - cbl_label_t *label = symbol_label(program, input->type, input->parent, input->name); if( label && label->type == LblNone ) { - const char *verb = "set"; label->type = input->type; label->parent = input->parent; label->line = input->line; - if( getenv(__func__) ) { - const cbl_label_t *L = label; - dbgmsg( "%s:%d: %-5s #%3zu %-9s '%s' of '%s' at line %d", - __func__, __LINE__, - verb, - symbol_elem_of(L) - symbols_begin(), - L->type_str()+3, - L->name, - L->parent? cbl_label_of(symbol_at(L->parent))->name : "", - L->line ); - } return label; } @@ -3864,15 +3714,6 @@ symbol_label_add( size_t program, cbl_label_t *input ) // restore munged line number unless symbol_add returned an existing label if( e->elem.label.line < 0 ) e->elem.label.line = -e->elem.label.line; - if( getenv(__func__) ) { - const cbl_label_t *L = cbl_label_of(e); - dbgmsg( "%s:%d: added #%3zu %-9s '%s' of '%s' at line %d", __func__, __LINE__, - e - symbols_begin(), - L->type_str()+3, - L->name, - L->parent? cbl_label_of(symbol_at(L->parent))->name : "", - L->line ); - } symbols.labelmap_add(e); return cbl_label_of(e); } @@ -3965,11 +3806,6 @@ symbol_special_add( size_t program, struct cbl_special_name_t *special ) struct symbol_elem_t *e = symbol_special(program, special->name); if( e ) { - cbl_special_name_t *s = cbl_special_name_of(e); - if( getenv(__func__) ) { - dbgmsg("%s:%d matches %s %d (%s)", __func__, __LINE__, - special->name, int(s->id), s->name); - } return e; } assert(e == NULL); @@ -3980,11 +3816,6 @@ symbol_special_add( size_t program, struct cbl_special_name_t *special ) cbl_errx( "%s:%d: could not add '%s'", __func__, __LINE__, special->name); } - if( getenv(__func__) ) { - dbgmsg( "%s:%d: added special '%s'", __func__, __LINE__, - e->elem.special.name); - } - elem_key_t key(program, cbl_special_name_of(e)->name); symbols.specials[key] = symbol_index(e); diff --git a/gcc/cobol/symfind.cc b/gcc/cobol/symfind.cc index 8995715..8c5f4af 100644 --- a/gcc/cobol/symfind.cc +++ b/gcc/cobol/symfind.cc @@ -200,12 +200,6 @@ field_structure( symbol_elem_t& sym ) { static const symbol_map_t::value_type none( symbol_map_t::key_type( 0, "", 0 ), std::vector<size_t>() ); - if( getenv(__func__) && sym.type == SymField ) { - const auto& field = *cbl_field_of(&sym); - dbgmsg("%s: #%zu %s: '%s' is_data_field: %s", __func__, - symbol_index(&sym), cbl_field_type_str(field.type), field.name, - is_data_field(sym)? "yes" : "no" ); - } if( !is_data_field(sym) ) return none; cbl_field_t *field = cbl_field_of(&sym); @@ -233,12 +227,6 @@ field_structure( symbol_elem_t& sym ) { } } - if( getenv(__func__) && yydebug ) { - dbgmsg( "%s:%d: '%s' has %zu ancestors", __func__, __LINE__, - elem.first.c_str(), elem.second.size() ); - dump_symbol_map_value(__func__, elem); - } - return elem; } @@ -270,12 +258,6 @@ build_symbol_map() { if( yydebug ) { dbgmsg( "%s:%d: %zu of %zu symbols inserted into %zu in symbol_map", __func__, __LINE__, nsym, end, symbol_map.size() ); - - if( getenv(__func__) ) { - for( const auto& elem : symbol_map ) { - dump_symbol_map_value1(elem); - } - } } } @@ -291,9 +273,6 @@ public: is_name( const char *name ) : name(name) {} bool operator()( symbol_map_t::value_type& elem ) { const bool tf = elem.first == name; - if( tf && getenv("is_name") ) { - dump_key( "matched", elem.first ); - } return tf; } protected: @@ -587,12 +566,6 @@ symbol_elem_t * symbol_find_of( size_t program, std::list<const char *> names, size_t group ) { symbol_map_t input = symbol_match(program, names); - if( getenv(__func__) && input.size() != 1 ) { - dbgmsg( "%s:%d: '%s' has %zu candidates for group %zu", - __func__, __LINE__, names.back(), input.size(), group ); - std::for_each( input.begin(), input.end(), dump_symbol_map_value1 ); - } - symbol_map_t items; std::copy_if( input.begin(), input.end(), std::inserter(items, items.begin()), in_group(group) ); diff --git a/gcc/cobol/token_names.h b/gcc/cobol/token_names.h index a082078..d1e3b5d 100644 --- a/gcc/cobol/token_names.h +++ b/gcc/cobol/token_names.h @@ -49,7 +49,7 @@ tokens = { { "number", NUMBER }, // 302 { "negative", NEGATIVE }, // 303 { "numstr", NUMSTR }, // 304 - { "overflow", OVERFLOW }, // 305 + { "overflow", OVERFLOW_kw }, // 305 { "computational", COMPUTATIONAL }, // 306 { "perform", PERFORM }, // 307 { "backward", BACKWARD }, // 308 diff --git a/gcc/cobol/util.cc b/gcc/cobol/util.cc index 1c0810b..f28fddf 100644 --- a/gcc/cobol/util.cc +++ b/gcc/cobol/util.cc @@ -1172,12 +1172,6 @@ valid_move( const struct cbl_field_t *tgt, const struct cbl_field_t *src ) } } - if( yydebug && getenv(__func__) ) { - dbgmsg("%s:%d: ok to move %s to %s (0x%02x)", __func__, __LINE__, - cbl_field_type_str(src->type), cbl_field_type_str(tgt->type), - retval); - } - return retval; } @@ -1443,15 +1437,6 @@ locally_unique( size_t program, const procdef_t& key, const procref_t& ref ) { const char *section_name = ref.has_section()? ref.section() : key.section(); procref_base_t full_ref(section_name, ref.paragraph()); - if( getenv(__func__) ) { - dbgmsg("%s: %zu for ref %s of '%s' (line %d) " - "in %s of '%s' (as %s of '%s')", __func__, - procedures.count(full_ref), - ref.paragraph(), ref.section(), ref.line_number(), - key.paragraph(), key.section(), - full_ref.paragraph(), full_ref.section() ); - } - return 1 == procedures.count(full_ref); } @@ -1473,9 +1458,6 @@ procedure_definition_add( size_t program, const cbl_label_t *procedure ) { } procdef_t key( section_name, paragraph_name, isym ); - if( getenv(__func__) ) { - dbgmsg("%s: #%3zu %s of %s", __func__, isym, paragraph_name, section_name); - } current_procedure = programs[program].insert( make_pair(key, procedures_t::mapped_type()) ); } @@ -1485,9 +1467,6 @@ void procedure_reference_add( const char *section, const char *paragraph, int line, size_t context ) { - if( getenv(__func__) ) { - dbgmsg("%s: line %3d %s of %s", __func__, line, paragraph, section); - } current_procedure->second.push_back( procref_t(section, paragraph, line, context) ); } @@ -1518,7 +1497,7 @@ ambiguous_reference( size_t program ) { ambiguous = find_if_not( proc.second.begin(), proc.second.end(), is_unique(program, proc.first) ); if( proc.second.end() != ambiguous ) { - if( yydebug || getenv("symbol_label_add")) { + if( yydebug ) { dbgmsg("%s: %s of '%s' has %zu potential matches", __func__, ambiguous->paragraph(), ambiguous->section(), procedures.count(*ambiguous)); @@ -1842,10 +1821,6 @@ bool cobol_filename( const char *name, ino_t inode ) { input_filename_vestige = name; bool pushed = input_filenames.push( input_file_t(name, inode, 1, lines) ); input_filenames.top().lineno = yylineno = 1; - if( getenv(__func__) ) { - dbgmsg(" saving %s with lineno as %d", - input_filenames.top().name, input_filenames.top().lineno); - } return pushed; } @@ -1854,9 +1829,6 @@ cobol_lineno_save() { if( input_filenames.empty() ) return NULL; auto& input( input_filenames.top() ); input.lineno = yylineno; - if( getenv(__func__) ) { - dbgmsg(" setting %s with lineno as %d", input.name, input.lineno); - } return input.name; } @@ -1880,9 +1852,6 @@ cobol_filename_restore() { input.lines = linemap_add(line_table, LC_LEAVE, sysp, NULL, 0); yylineno = input.lineno; - if( getenv("cobol_filename") ) { - dbgmsg("restoring %s with lineno to %d", input.name, input.lineno); - } return input.name; } @@ -2118,8 +2087,6 @@ cobol_fileline_set( const char line[] ) { input_file_t input_file( filename, ino_t(0), fileline ); // constructor sets inode - if( getenv(__func__) ) return filename; // ignore #line directive - if( input_filenames.empty() ) { input_file.lines = linemap_add(line_table, LC_ENTER, sysp, filename, 1); input_filenames.push(input_file); @@ -2132,20 +2099,20 @@ cobol_fileline_set( const char line[] ) { return file.name; } -class timespec_t { +class cbl_timespec { struct timespec now; public: - timespec_t() { + cbl_timespec() { clock_gettime(CLOCK_MONOTONIC, &now); } double ns() const { return now.tv_sec * 1000000000 + now.tv_nsec; } - friend double operator-( const timespec_t& now, const timespec_t& then ); + friend double operator-( const cbl_timespec& now, const cbl_timespec& then ); }; double -operator-( const timespec_t& then, const timespec_t& now ) { +operator-( const cbl_timespec& then, const cbl_timespec& now ) { return (now.ns() - then.ns()) / 1000000000; } @@ -2158,11 +2125,11 @@ parse_file( const char filename[] ) parser_enter_file(filename); - timespec_t start; + cbl_timespec start; int erc = yyparse(); - timespec_t finish; + cbl_timespec finish; double dt = finish - start; parser_leave_file(); diff --git a/gcc/combine.cc b/gcc/combine.cc index 5f08518..873c2bd 100644 --- a/gcc/combine.cc +++ b/gcc/combine.cc @@ -815,7 +815,7 @@ do_SUBST_LINK (struct insn_link **into, struct insn_link *newval) #define SUBST_LINK(oldval, newval) do_SUBST_LINK (&oldval, newval) /* Subroutine of try_combine. Determine whether the replacement patterns - NEWPAT, NEWI2PAT and NEWOTHERPAT are cheaper according to insn_cost + NEWPAT, NEWI2PAT and NEWOTHERPAT are more expensive according to insn_cost than the original sequence I0, I1, I2, I3 and undobuf.other_insn. Note that I0, I1 and/or NEWI2PAT may be NULL_RTX. Similarly, NEWOTHERPAT and undobuf.other_insn may also both be NULL_RTX. Return false if the cost @@ -4129,8 +4129,8 @@ try_combine (rtx_insn *i3, rtx_insn *i2, rtx_insn *i1, rtx_insn *i0, } } - /* Only allow this combination if insn_cost reports that the - replacement instructions are cheaper than the originals. */ + /* Reject this combination if insn_cost reports that the replacement + instructions are more expensive than the originals. */ if (!combine_validate_cost (i0, i1, i2, i3, newpat, newi2pat, other_pat)) { undo_all (); @@ -5280,6 +5280,12 @@ find_split_point (rtx *loc, rtx_insn *insn, bool set_src) SUBST (XEXP (x, 0), XEXP (x, 1)); SUBST (XEXP (x, 1), tem); } + /* Many targets have a `(and (not X) Y)` and/or `(ior (not X) Y)` instructions. + Split at that insns. However if this is + the SET_SRC, we likely do not have such an instruction and it's + worthless to try this split. */ + if (!set_src && GET_CODE (XEXP (x, 0)) == NOT) + return loc; break; case PLUS: diff --git a/gcc/common.opt b/gcc/common.opt index 2c8fdde..88d987e 100644 --- a/gcc/common.opt +++ b/gcc/common.opt @@ -2116,6 +2116,10 @@ fipa-modref Common Var(flag_ipa_modref) Optimization Perform interprocedural modref analysis. +fipa-reorder-for-locality +Common Var(flag_ipa_reorder_for_locality) Init(0) Optimization +Perform reordering and cloning of functions to maximize locality. + fipa-profile Common Var(flag_ipa_profile) Init(0) Optimization Perform interprocedural profile propagation. @@ -2275,6 +2279,9 @@ Enum Name(lto_partition_model) Type(enum lto_partition_model) UnknownError(unknown LTO partitioning model %qs) EnumValue +Enum(lto_partition_model) String(default) Value(LTO_PARTITION_DEFAULT) + +EnumValue Enum(lto_partition_model) String(none) Value(LTO_PARTITION_NONE) EnumValue @@ -2293,7 +2300,7 @@ EnumValue Enum(lto_partition_model) String(cache) Value(LTO_PARTITION_CACHE) flto-partition= -Common Joined RejectNegative Enum(lto_partition_model) Var(flag_lto_partition) Init(LTO_PARTITION_BALANCED) +Common Joined RejectNegative Enum(lto_partition_model) Var(flag_lto_partition) Init(LTO_PARTITION_DEFAULT) Specify the algorithm to partition symbols and vars at linktime. ; The initial value of -1 comes from Z_DEFAULT_COMPRESSION in zlib.h. diff --git a/gcc/common.opt.urls b/gcc/common.opt.urls index 860ebd0..0077511 100644 --- a/gcc/common.opt.urls +++ b/gcc/common.opt.urls @@ -31,8 +31,9 @@ UrlSuffix(gcc/Optimize-Options.html#index-Og) Oz UrlSuffix(gcc/Optimize-Options.html#index-Oz) -Q -UrlSuffix(gcc/Developer-Options.html#index-Q) +; skipping UrlSuffix for 'Q' due to multiple URLs: +; duplicate: 'gcc/Developer-Options.html#index-Q-1' +; duplicate: 'gcc/Overall-Options.html#index-Q' Qn UrlSuffix(gcc/System-V-Options.html#index-Qn) @@ -868,6 +869,9 @@ UrlSuffix(gcc/Optimize-Options.html#index-fipa-bit-cp) fipa-modref UrlSuffix(gcc/Optimize-Options.html#index-fipa-modref) +fipa-reorder-for-locality +UrlSuffix(gcc/Optimize-Options.html#index-fipa-reorder-for-locality) + fipa-profile UrlSuffix(gcc/Optimize-Options.html#index-fipa-profile) @@ -1459,6 +1463,9 @@ UrlSuffix(gcc/Optimize-Options.html#index-ftree-coalesce-vars) ftree-copy-prop UrlSuffix(gcc/Optimize-Options.html#index-ftree-copy-prop) +ftree-cselim +UrlSuffix(gcc/Optimize-Options.html#index-ftree-cselim) + ftree-switch-conversion UrlSuffix(gcc/Optimize-Options.html#index-ftree-switch-conversion) diff --git a/gcc/common/config/s390/s390-common.cc b/gcc/common/config/s390/s390-common.cc index 4b0691d..8a147d7 100644 --- a/gcc/common/config/s390/s390-common.cc +++ b/gcc/common/config/s390/s390-common.cc @@ -54,10 +54,10 @@ EXPORTED_CONST int processor_flags_table[] = | PF_EXTIMM | PF_DFP | PF_Z10 | PF_Z196 | PF_ZEC12 | PF_TX | PF_Z13 | PF_VX | PF_VXE | PF_Z14 | PF_VXE2 | PF_Z15 | PF_NNPA | PF_Z16, - /* arch15 */ PF_IEEE_FLOAT | PF_ZARCH | PF_LONG_DISPLACEMENT + /* z17 */ PF_IEEE_FLOAT | PF_ZARCH | PF_LONG_DISPLACEMENT | PF_EXTIMM | PF_DFP | PF_Z10 | PF_Z196 | PF_ZEC12 | PF_TX | PF_Z13 | PF_VX | PF_VXE | PF_Z14 | PF_VXE2 | PF_Z15 - | PF_NNPA | PF_Z16 | PF_VXE3 | PF_ARCH15 + | PF_NNPA | PF_Z16 | PF_VXE3 | PF_Z17 }; /* Change optimizations to be performed, depending on the diff --git a/gcc/config.gcc b/gcc/config.gcc index f7f2002..d98df88 100644 --- a/gcc/config.gcc +++ b/gcc/config.gcc @@ -2540,6 +2540,20 @@ riscv*-*-linux*) gcc_cv_initfini_array=yes with_tls=${with_tls:-trad} ;; +riscv*-*-gnu*) + tm_file="elfos.h gnu-user.h gnu.h glibc-stdint.h ${tm_file} riscv/gnu.h" + tmake_file="${tmake_file} riscv/t-riscv" + gnu_ld=yes + gas=yes + case $target in + riscv32be-*|riscv64be-*) + tm_defines="${tm_defines} TARGET_BIG_ENDIAN_DEFAULT=1" + ;; + esac + # Force .init_array support. The configure script cannot always + # automatically detect that GAS supports it, yet we require it. + gcc_cv_initfini_array=yes + ;; riscv*-*-elf* | riscv*-*-rtems*) tm_file="elfos.h newlib-stdint.h ${tm_file} riscv/elf.h" case ${target} in @@ -5756,7 +5770,7 @@ case "${target}" in for which in arch tune; do eval "val=\$with_$which" case ${val} in - "" | native | z900 | z990 | z9-109 | z9-ec | z10 | z196 | zEC12 | z13 | z14 | z15 | z16 | arch5 | arch6 | arch7 | arch8 | arch9 | arch10 | arch11 | arch12 | arch13 | arch14 | arch15 ) + "" | native | z900 | z990 | z9-109 | z9-ec | z10 | z196 | zEC12 | z13 | z14 | z15 | z16 | z17 | arch5 | arch6 | arch7 | arch8 | arch9 | arch10 | arch11 | arch12 | arch13 | arch14 | arch15 ) # OK ;; *) diff --git a/gcc/config.in b/gcc/config.in index 7c89cab..a79c51a 100644 --- a/gcc/config.in +++ b/gcc/config.in @@ -2320,6 +2320,13 @@ #endif +/* Define if personality and ADDR_NO_RANDOMIZE are declared in + sys/personality.h. */ +#ifndef USED_FOR_TARGET +#undef HOST_HAS_PERSONALITY_ADDR_NO_RANDOMIZE +#endif + + /* Define which stat syscall is able to handle 64bit indodes. */ #ifndef USED_FOR_TARGET #undef HOST_STAT_FOR_64BIT_INODES diff --git a/gcc/config/aarch64/aarch64-sve.md b/gcc/config/aarch64/aarch64-sve.md index 3dbd659..d4af370 100644 --- a/gcc/config/aarch64/aarch64-sve.md +++ b/gcc/config/aarch64/aarch64-sve.md @@ -3133,9 +3133,9 @@ "TARGET_SVE" { rtx tmp = gen_reg_rtx (<MODE>mode); - emit_insn (gen_vcond_mask_<mode><vpred> (tmp, operands[1], - CONST1_RTX (<MODE>mode), - CONST0_RTX (<MODE>mode))); + emit_insn (gen_vcond_mask_<mode><vpred> (tmp, CONST1_RTX (<MODE>mode), + CONST0_RTX (<MODE>mode), + operands[1])); emit_insn (gen_vec_extract<mode><Vel> (operands[0], tmp, operands[2])); DONE; } diff --git a/gcc/config/aarch64/aarch64.cc b/gcc/config/aarch64/aarch64.cc index 4e80114..433ec97 100644 --- a/gcc/config/aarch64/aarch64.cc +++ b/gcc/config/aarch64/aarch64.cc @@ -31073,8 +31073,6 @@ aarch64_valid_sysreg_name_p (const char *regname) const sysreg_t *sysreg = aarch64_lookup_sysreg_map (regname); if (sysreg == NULL) return aarch64_is_implem_def_reg (regname); - if (sysreg->arch_reqs) - return bool (aarch64_isa_flags & sysreg->arch_reqs); return true; } @@ -31098,8 +31096,6 @@ aarch64_retrieve_sysreg (const char *regname, bool write_p, bool is128op) if ((write_p && (sysreg->properties & F_REG_READ)) || (!write_p && (sysreg->properties & F_REG_WRITE))) return NULL; - if ((~aarch64_isa_flags & sysreg->arch_reqs) != 0) - return NULL; return sysreg->encoding; } diff --git a/gcc/config/alpha/alpha.cc b/gcc/config/alpha/alpha.cc index ba470d9..14e7da5 100644 --- a/gcc/config/alpha/alpha.cc +++ b/gcc/config/alpha/alpha.cc @@ -4291,14 +4291,10 @@ alpha_get_mem_rtx_alignment_and_offset (rtx expr, int &a, HOST_WIDE_INT &o) tree mem = MEM_EXPR (expr); if (mem != NULL_TREE) - switch (TREE_CODE (mem)) - { - case MEM_REF: - tree_offset = mem_ref_offset (mem).force_shwi (); - tree_align = get_object_alignment (get_base_address (mem)); - break; + { + HOST_WIDE_INT comp_offset = 0; - case COMPONENT_REF: + for (; TREE_CODE (mem) == COMPONENT_REF; mem = TREE_OPERAND (mem, 0)) { tree byte_offset = component_ref_field_offset (mem); tree bit_offset = DECL_FIELD_BIT_OFFSET (TREE_OPERAND (mem, 1)); @@ -4307,14 +4303,15 @@ alpha_get_mem_rtx_alignment_and_offset (rtx expr, int &a, HOST_WIDE_INT &o) || !poly_int_tree_p (byte_offset, &offset) || !tree_fits_shwi_p (bit_offset)) break; - tree_offset = offset + tree_to_shwi (bit_offset) / BITS_PER_UNIT; + comp_offset += offset + tree_to_shwi (bit_offset) / BITS_PER_UNIT; } - tree_align = get_object_alignment (get_base_address (mem)); - break; - default: - break; - } + if (TREE_CODE (mem) == MEM_REF) + { + tree_offset = comp_offset + mem_ref_offset (mem).force_shwi (); + tree_align = get_object_alignment (get_base_address (mem)); + } + } if (reg_align > mem_align) { diff --git a/gcc/config/c6x/c6x.h b/gcc/config/c6x/c6x.h index e7da250..50bad27 100644 --- a/gcc/config/c6x/c6x.h +++ b/gcc/config/c6x/c6x.h @@ -444,11 +444,9 @@ struct GTY(()) machine_function #define TARG_VEC_PERMUTE_COST 1 #endif -/* ttype entries (the only interesting data references used) are - sb-relative got-indirect (aka .ehtype). */ +/* .ehtype ttype entries are sb-relative. */ #define ASM_PREFERRED_EH_DATA_FORMAT(code, data) \ - (((code) == 0 && (data) == 1) ? (DW_EH_PE_datarel | DW_EH_PE_indirect) \ - : DW_EH_PE_absptr) + (((code) == 0 && (data) == 1) ? DW_EH_PE_datarel : DW_EH_PE_absptr) /* This should be the same as the definition in elfos.h, plus the call to output special unwinding directives. */ diff --git a/gcc/config/darwin.h b/gcc/config/darwin.h index 8c164fd..9b9a3fe 100644 --- a/gcc/config/darwin.h +++ b/gcc/config/darwin.h @@ -504,6 +504,7 @@ extern GTY(()) int darwin_ms_struct; %{static|static-libgcc|static-libgfortran:%:replace-outfile(-lgfortran libgfortran.a%s)}\ %{static|static-libgcc|static-libquadmath:%:replace-outfile(-lquadmath libquadmath.a%s)}\ %{static|static-libgcc|static-libphobos:%:replace-outfile(-lgphobos libgphobos.a%s)}\ + %{static|static-libgcc|static-libgcobol:%:replace-outfile(-lgcobol libgcobol.a%s)}\ %{static|static-libgcc|static-libstdc++|static-libgfortran:%:replace-outfile(-lgomp libgomp.a%s)}\ %{static|static-libgcc|static-libstdc++:%:replace-outfile(-lstdc++ libstdc++.a%s)}\ %{static|static-libgm2:%:replace-outfile(-lm2pim libm2pim.a%s)}\ diff --git a/gcc/config/gcn/gcn.md b/gcc/config/gcn/gcn.md index 695656f..e0fb735 100644 --- a/gcc/config/gcn/gcn.md +++ b/gcc/config/gcn/gcn.md @@ -1018,7 +1018,9 @@ [(const_int 0)] "" { - sorry ("exception handling not supported"); + if (!fake_exceptions) + sorry ("exception handling not supported"); + DONE; }) ;; }}} diff --git a/gcc/config/gcn/gcn.opt b/gcc/config/gcn/gcn.opt index 142b439..99d6aeb 100644 --- a/gcc/config/gcn/gcn.opt +++ b/gcc/config/gcn/gcn.opt @@ -101,3 +101,11 @@ Enum(gcn_preferred_vectorization_factor) String(32) Value(32) EnumValue Enum(gcn_preferred_vectorization_factor) String(64) Value(64) + +mfake-exceptions +Target Var(fake_exceptions) Init(0) Undocumented +; With '-mfake-exceptions' enabled, the user-visible behavior in presence of +; exception handling constructs changes such that the compile-time +; 'sorry, unimplemented: exception handling not supported' is skipped, code +; generation proceeds, and instead, exception handling constructs 'abort' at +; run time. (..., or don't, if they're in dead code.) diff --git a/gcc/config/gcn/mkoffload.cc b/gcc/config/gcn/mkoffload.cc index f5b89c9..b284ff4 100644 --- a/gcc/config/gcn/mkoffload.cc +++ b/gcc/config/gcn/mkoffload.cc @@ -1160,6 +1160,9 @@ main (int argc, char **argv) obstack_ptr_grow (&cc_argv_obstack, "-xlto"); if (fopenmp) obstack_ptr_grow (&cc_argv_obstack, "-mgomp"); + /* The host code may contain exception handling constructs. + Handle these as good as we can. */ + obstack_ptr_grow (&cc_argv_obstack, "-mfake-exceptions"); for (int ix = 1; ix != argc; ix++) { diff --git a/gcc/config/i386/i386-expand.cc b/gcc/config/i386/i386-expand.cc index cdfd94d..36f71eb 100644 --- a/gcc/config/i386/i386-expand.cc +++ b/gcc/config/i386/i386-expand.cc @@ -19256,8 +19256,6 @@ ix86_emit_swdivsf (rtx res, rtx a, rtx b, machine_mode mode) e1 = gen_reg_rtx (mode); x1 = gen_reg_rtx (mode); - /* a / b = a * ((rcp(b) + rcp(b)) - (b * rcp(b) * rcp (b))) */ - b = force_reg (mode, b); /* x0 = rcp(b) estimate */ @@ -19270,20 +19268,42 @@ ix86_emit_swdivsf (rtx res, rtx a, rtx b, machine_mode mode) emit_insn (gen_rtx_SET (x0, gen_rtx_UNSPEC (mode, gen_rtvec (1, b), UNSPEC_RCP))); - /* e0 = x0 * b */ - emit_insn (gen_rtx_SET (e0, gen_rtx_MULT (mode, x0, b))); + unsigned vector_size = GET_MODE_SIZE (mode); + + /* (a - (rcp(b) * a * b)) * rcp(b) + rcp(b) * a + N-R step with 2 fma implementation. */ + if (TARGET_FMA + || (TARGET_AVX512F && vector_size == 64) + || (TARGET_AVX512VL && (vector_size == 32 || vector_size == 16))) + { + /* e0 = x0 * a */ + emit_insn (gen_rtx_SET (e0, gen_rtx_MULT (mode, x0, a))); + /* e1 = e0 * b - a */ + emit_insn (gen_rtx_SET (e1, gen_rtx_FMA (mode, e0, b, + gen_rtx_NEG (mode, a)))); + /* res = - e1 * x0 + e0 */ + emit_insn (gen_rtx_SET (res, gen_rtx_FMA (mode, + gen_rtx_NEG (mode, e1), + x0, e0))); + } + else + /* a / b = a * ((rcp(b) + rcp(b)) - (b * rcp(b) * rcp (b))) */ + { + /* e0 = x0 * b */ + emit_insn (gen_rtx_SET (e0, gen_rtx_MULT (mode, x0, b))); - /* e0 = x0 * e0 */ - emit_insn (gen_rtx_SET (e0, gen_rtx_MULT (mode, x0, e0))); + /* e1 = x0 + x0 */ + emit_insn (gen_rtx_SET (e1, gen_rtx_PLUS (mode, x0, x0))); - /* e1 = x0 + x0 */ - emit_insn (gen_rtx_SET (e1, gen_rtx_PLUS (mode, x0, x0))); + /* e0 = x0 * e0 */ + emit_insn (gen_rtx_SET (e0, gen_rtx_MULT (mode, x0, e0))); - /* x1 = e1 - e0 */ - emit_insn (gen_rtx_SET (x1, gen_rtx_MINUS (mode, e1, e0))); + /* x1 = e1 - e0 */ + emit_insn (gen_rtx_SET (x1, gen_rtx_MINUS (mode, e1, e0))); - /* res = a * x1 */ - emit_insn (gen_rtx_SET (res, gen_rtx_MULT (mode, a, x1))); + /* res = a * x1 */ + emit_insn (gen_rtx_SET (res, gen_rtx_MULT (mode, a, x1))); + } } /* Output code to perform a Newton-Rhapson approximation of a diff --git a/gcc/config/i386/i386-options.cc b/gcc/config/i386/i386-options.cc index a9fac01..964449f 100644 --- a/gcc/config/i386/i386-options.cc +++ b/gcc/config/i386/i386-options.cc @@ -2828,8 +2828,8 @@ ix86_option_override_internal (bool main_args_p, if (flag_nop_mcount) error ("%<-mnop-mcount%> is not compatible with this target"); #endif - if (flag_nop_mcount && flag_pic) - error ("%<-mnop-mcount%> is not implemented for %<-fPIC%>"); + if (flag_nop_mcount && flag_pic && !flag_plt) + error ("%<-mnop-mcount%> is not implemented for %<-fno-plt%>"); /* Accept -msseregparm only if at least SSE support is enabled. */ if (TARGET_SSEREGPARM_P (opts->x_target_flags) diff --git a/gcc/config/i386/i386.cc b/gcc/config/i386/i386.cc index 4f8380c4..d15f91d 100644 --- a/gcc/config/i386/i386.cc +++ b/gcc/config/i386/i386.cc @@ -100,6 +100,7 @@ along with GCC; see the file COPYING3. If not see #include "i386-features.h" #include "function-abi.h" #include "rtl-error.h" +#include "gimple-pretty-print.h" /* This file should be included last. */ #include "target-def.h" @@ -458,6 +459,9 @@ int ix86_arch_specified; indirect thunk pushes the return address onto stack, destroying red-zone. + NB: Don't use red-zone for functions with no_caller_saved_registers + and 32 GPRs since 128-byte red-zone is too small for 31 GPRs. + TODO: If we can reserve the first 2 WORDs, for PUSH and, another for CALL, in red-zone, we can allow local indirect jumps with indirect thunk. */ @@ -467,6 +471,9 @@ ix86_using_red_zone (void) { return (TARGET_RED_ZONE && !TARGET_64BIT_MS_ABI + && (!TARGET_APX_EGPR + || (cfun->machine->call_saved_registers + != TYPE_NO_CALLER_SAVED_REGISTERS)) && (!cfun->machine->has_local_indirect_jump || cfun->machine->indirect_branch_type == indirect_branch_keep)); } @@ -21810,6 +21817,25 @@ ix86_insn_cost (rtx_insn *insn, bool speed) return insn_cost + pattern_cost (PATTERN (insn), speed); } +/* Return cost of SSE/AVX FP->FP conversion (extensions and truncates). */ + +static int +vec_fp_conversion_cost (const struct processor_costs *cost, int size) +{ + if (size < 128) + return cost->cvtss2sd; + else if (size < 256) + { + if (TARGET_SSE_SPLIT_REGS) + return cost->cvtss2sd * size / 64; + return cost->cvtss2sd; + } + if (size < 512) + return cost->vcvtps2pd256; + else + return cost->vcvtps2pd512; +} + /* Compute a (partial) cost for rtx X. Return true if the complete cost has been computed, and false if subexpressions should be scanned. In either case, *TOTAL contains the cost result. */ @@ -22473,17 +22499,18 @@ ix86_rtx_costs (rtx x, machine_mode mode, int outer_code_i, int opno, return false; case FLOAT_EXTEND: + /* x87 represents all values extended to 80bit. */ if (!SSE_FLOAT_MODE_SSEMATH_OR_HFBF_P (mode)) *total = 0; else - *total = ix86_vec_cost (mode, cost->addss); + *total = vec_fp_conversion_cost (cost, GET_MODE_BITSIZE (mode)); return false; case FLOAT_TRUNCATE: if (!SSE_FLOAT_MODE_SSEMATH_OR_HFBF_P (mode)) *total = cost->fadd; else - *total = ix86_vec_cost (mode, cost->addss); + *total = vec_fp_conversion_cost (cost, GET_MODE_BITSIZE (mode)); return false; case ABS: @@ -23158,6 +23185,12 @@ x86_print_call_or_nop (FILE *file, const char *target) if (flag_nop_mcount || !strcmp (target, "nop")) /* 5 byte nop: nopl 0(%[re]ax,%[re]ax,1) */ fprintf (file, "1:" ASM_BYTE "0x0f, 0x1f, 0x44, 0x00, 0x00\n"); + else if (!TARGET_PECOFF && flag_pic) + { + gcc_assert (flag_plt); + + fprintf (file, "1:\tcall\t%s@PLT\n", target); + } else fprintf (file, "1:\tcall\t%s\n", target); } @@ -23321,7 +23354,7 @@ x86_function_profiler (FILE *file, int labelno ATTRIBUTE_UNUSED) break; case CM_SMALL_PIC: case CM_MEDIUM_PIC: - if (!ix86_direct_extern_access) + if (!flag_plt) { if (ASSEMBLER_DIALECT == ASM_INTEL) fprintf (file, "1:\tcall\t[QWORD PTR %s@GOTPCREL[rip]]\n", @@ -23352,7 +23385,9 @@ x86_function_profiler (FILE *file, int labelno ATTRIBUTE_UNUSED) "\tleal\t%sP%d@GOTOFF(%%ebx), %%" PROFILE_COUNT_REGISTER "\n", LPREFIX, labelno); #endif - if (ASSEMBLER_DIALECT == ASM_INTEL) + if (flag_plt) + x86_print_call_or_nop (file, mcount_name); + else if (ASSEMBLER_DIALECT == ASM_INTEL) fprintf (file, "1:\tcall\t[DWORD PTR %s@GOT[ebx]]\n", mcount_name); else fprintf (file, "1:\tcall\t*%s@GOT(%%ebx)\n", mcount_name); @@ -24669,7 +24704,7 @@ ix86_builtin_vectorization_cost (enum vect_cost_for_stmt type_of_cost, switch (type_of_cost) { case scalar_stmt: - return fp ? ix86_cost->addss : COSTS_N_INSNS (1); + return fp ? ix86_cost->addss : COSTS_N_INSNS (1); case scalar_load: /* load/store costs are relative to register move which is 2. Recompute @@ -24740,7 +24775,11 @@ ix86_builtin_vectorization_cost (enum vect_cost_for_stmt type_of_cost, return ix86_cost->cond_not_taken_branch_cost; case vec_perm: + return ix86_vec_cost (mode, ix86_cost->sse_op); + case vec_promote_demote: + if (fp) + return vec_fp_conversion_cost (ix86_tune_cost, mode); return ix86_vec_cost (mode, ix86_cost->sse_op); case vec_construct: @@ -25328,6 +25367,9 @@ ix86_vector_costs::add_stmt_cost (int count, vect_cost_for_stmt kind, (TREE_TYPE (gimple_assign_lhs (stmt_info->stmt)), TREE_TYPE (gimple_assign_rhs1 (stmt_info->stmt)))) stmt_cost = 0; + else if (fp) + stmt_cost = vec_fp_conversion_cost + (ix86_tune_cost, GET_MODE_BITSIZE (mode)); break; case BIT_IOR_EXPR: @@ -25369,6 +25411,29 @@ ix86_vector_costs::add_stmt_cost (int count, vect_cost_for_stmt kind, break; } + if (kind == vec_promote_demote + && fp && FLOAT_TYPE_P (TREE_TYPE (gimple_assign_rhs1 (stmt_info->stmt)))) + { + int outer_size + = tree_to_uhwi + (TYPE_SIZE + (TREE_TYPE (gimple_assign_lhs (stmt_info->stmt)))); + int inner_size + = tree_to_uhwi + (TYPE_SIZE + (TREE_TYPE (gimple_assign_rhs1 (stmt_info->stmt)))); + int stmt_cost = vec_fp_conversion_cost + (ix86_tune_cost, GET_MODE_BITSIZE (mode)); + /* VEC_PACK_TRUNC_EXPR: If inner size is greater than outer size we will end + up doing two conversions and packing them. */ + if (inner_size > outer_size) + { + int n = inner_size / outer_size; + stmt_cost = stmt_cost * n + + (n - 1) * ix86_vec_cost (mode, ix86_cost->sse_op); + } + } + /* If we do elementwise loads into a vector then we are bound by latency and execution resources for the many scalar loads (AGU and load ports). Try to account for this by scaling the diff --git a/gcc/config/i386/i386.h b/gcc/config/i386/i386.h index 8507243..18aa42d 100644 --- a/gcc/config/i386/i386.h +++ b/gcc/config/i386/i386.h @@ -207,6 +207,12 @@ struct processor_costs { const int divsd; /* cost of DIVSD instructions. */ const int sqrtss; /* cost of SQRTSS instructions. */ const int sqrtsd; /* cost of SQRTSD instructions. */ + const int cvtss2sd; /* cost SSE FP conversions, + such as CVTSS2SD. */ + const int vcvtps2pd256; /* cost 256bit packed FP conversions, + such as VCVTPD2PS with larger reg in ymm. */ + const int vcvtps2pd512; /* cost 512bit packed FP conversions, + such as VCVTPD2PS with larger reg in zmm. */ const int reassoc_int, reassoc_fp, reassoc_vec_int, reassoc_vec_fp; /* Specify reassociation width for integer, fp, vector integer and vector fp diff --git a/gcc/config/i386/x86-tune-costs.h b/gcc/config/i386/x86-tune-costs.h index 7c8cb73..cddcf61 100644 --- a/gcc/config/i386/x86-tune-costs.h +++ b/gcc/config/i386/x86-tune-costs.h @@ -121,16 +121,19 @@ struct processor_costs ix86_size_cost = {/* costs for tuning for size */ COSTS_N_BYTES (2), /* cost of FCHS instruction. */ COSTS_N_BYTES (2), /* cost of FSQRT instruction. */ - COSTS_N_BYTES (2), /* cost of cheap SSE instruction. */ - COSTS_N_BYTES (2), /* cost of ADDSS/SD SUBSS/SD insns. */ - COSTS_N_BYTES (2), /* cost of MULSS instruction. */ - COSTS_N_BYTES (2), /* cost of MULSD instruction. */ - COSTS_N_BYTES (2), /* cost of FMA SS instruction. */ - COSTS_N_BYTES (2), /* cost of FMA SD instruction. */ - COSTS_N_BYTES (2), /* cost of DIVSS instruction. */ - COSTS_N_BYTES (2), /* cost of DIVSD instruction. */ - COSTS_N_BYTES (2), /* cost of SQRTSS instruction. */ - COSTS_N_BYTES (2), /* cost of SQRTSD instruction. */ + COSTS_N_BYTES (4), /* cost of cheap SSE instruction. */ + COSTS_N_BYTES (4), /* cost of ADDSS/SD SUBSS/SD insns. */ + COSTS_N_BYTES (4), /* cost of MULSS instruction. */ + COSTS_N_BYTES (4), /* cost of MULSD instruction. */ + COSTS_N_BYTES (4), /* cost of FMA SS instruction. */ + COSTS_N_BYTES (4), /* cost of FMA SD instruction. */ + COSTS_N_BYTES (4), /* cost of DIVSS instruction. */ + COSTS_N_BYTES (4), /* cost of DIVSD instruction. */ + COSTS_N_BYTES (4), /* cost of SQRTSS instruction. */ + COSTS_N_BYTES (4), /* cost of SQRTSD instruction. */ + COSTS_N_BYTES (4), /* cost of CVTSS2SD etc. */ + COSTS_N_BYTES (4), /* cost of 256bit VCVTPS2PD etc. */ + COSTS_N_BYTES (6), /* cost of 512bit VCVTPS2PD etc. */ 1, 1, 1, 1, /* reassoc int, fp, vec_int, vec_fp. */ ix86_size_memcpy, ix86_size_memset, @@ -243,6 +246,9 @@ struct processor_costs i386_cost = { /* 386 specific costs */ COSTS_N_INSNS (88), /* cost of DIVSD instruction. */ COSTS_N_INSNS (122), /* cost of SQRTSS instruction. */ COSTS_N_INSNS (122), /* cost of SQRTSD instruction. */ + COSTS_N_INSNS (27), /* cost of CVTSS2SD etc. */ + COSTS_N_INSNS (54), /* cost of 256bit VCVTPS2PD etc. */ + COSTS_N_INSNS (108), /* cost of 512bit VCVTPS2PD etc. */ 1, 1, 1, 1, /* reassoc int, fp, vec_int, vec_fp. */ i386_memcpy, i386_memset, @@ -356,6 +362,9 @@ struct processor_costs i486_cost = { /* 486 specific costs */ COSTS_N_INSNS (74), /* cost of DIVSD instruction. */ COSTS_N_INSNS (83), /* cost of SQRTSS instruction. */ COSTS_N_INSNS (83), /* cost of SQRTSD instruction. */ + COSTS_N_INSNS (8), /* cost of CVTSS2SD etc. */ + COSTS_N_INSNS (16), /* cost of 256bit VCVTPS2PD etc. */ + COSTS_N_INSNS (32), /* cost of 512bit VCVTPS2PD etc. */ 1, 1, 1, 1, /* reassoc int, fp, vec_int, vec_fp. */ i486_memcpy, i486_memset, @@ -467,6 +476,9 @@ struct processor_costs pentium_cost = { COSTS_N_INSNS (39), /* cost of DIVSD instruction. */ COSTS_N_INSNS (70), /* cost of SQRTSS instruction. */ COSTS_N_INSNS (70), /* cost of SQRTSD instruction. */ + COSTS_N_INSNS (3), /* cost of CVTSS2SD etc. */ + COSTS_N_INSNS (6), /* cost of 256bit VCVTPS2PD etc. */ + COSTS_N_INSNS (12), /* cost of 512bit VCVTPS2PD etc. */ 1, 1, 1, 1, /* reassoc int, fp, vec_int, vec_fp. */ pentium_memcpy, pentium_memset, @@ -571,6 +583,9 @@ struct processor_costs lakemont_cost = { COSTS_N_INSNS (60), /* cost of DIVSD instruction. */ COSTS_N_INSNS (31), /* cost of SQRTSS instruction. */ COSTS_N_INSNS (63), /* cost of SQRTSD instruction. */ + COSTS_N_INSNS (5), /* cost of CVTSS2SD etc. */ + COSTS_N_INSNS (10), /* cost of 256bit VCVTPS2PD etc. */ + COSTS_N_INSNS (20), /* cost of 512bit VCVTPS2PD etc. */ 1, 1, 1, 1, /* reassoc int, fp, vec_int, vec_fp. */ pentium_memcpy, pentium_memset, @@ -690,6 +705,9 @@ struct processor_costs pentiumpro_cost = { COSTS_N_INSNS (18), /* cost of DIVSD instruction. */ COSTS_N_INSNS (31), /* cost of SQRTSS instruction. */ COSTS_N_INSNS (31), /* cost of SQRTSD instruction. */ + COSTS_N_INSNS (3), /* cost of CVTSS2SD etc. */ + COSTS_N_INSNS (6), /* cost of 256bit VCVTPS2PD etc. */ + COSTS_N_INSNS (12), /* cost of 512bit VCVTPS2PD etc. */ 1, 1, 1, 1, /* reassoc int, fp, vec_int, vec_fp. */ pentiumpro_memcpy, pentiumpro_memset, @@ -800,6 +818,9 @@ struct processor_costs geode_cost = { COSTS_N_INSNS (47), /* cost of DIVSD instruction. */ COSTS_N_INSNS (54), /* cost of SQRTSS instruction. */ COSTS_N_INSNS (54), /* cost of SQRTSD instruction. */ + COSTS_N_INSNS (6), /* cost of CVTSS2SD etc. */ + COSTS_N_INSNS (12), /* cost of 256bit VCVTPS2PD etc. */ + COSTS_N_INSNS (24), /* cost of 512bit VCVTPS2PD etc. */ 1, 1, 1, 1, /* reassoc int, fp, vec_int, vec_fp. */ geode_memcpy, geode_memset, @@ -913,6 +934,9 @@ struct processor_costs k6_cost = { COSTS_N_INSNS (56), /* cost of DIVSD instruction. */ COSTS_N_INSNS (56), /* cost of SQRTSS instruction. */ COSTS_N_INSNS (56), /* cost of SQRTSD instruction. */ + COSTS_N_INSNS (2), /* cost of CVTSS2SD etc. */ + COSTS_N_INSNS (4), /* cost of 256bit VCVTPS2PD etc. */ + COSTS_N_INSNS (8), /* cost of 512bit VCVTPS2PD etc. */ 1, 1, 1, 1, /* reassoc int, fp, vec_int, vec_fp. */ k6_memcpy, k6_memset, @@ -1027,6 +1051,9 @@ struct processor_costs athlon_cost = { COSTS_N_INSNS (24), /* cost of DIVSD instruction. */ COSTS_N_INSNS (19), /* cost of SQRTSS instruction. */ COSTS_N_INSNS (19), /* cost of SQRTSD instruction. */ + COSTS_N_INSNS (4), /* cost of CVTSS2SD etc. */ + COSTS_N_INSNS (8), /* cost of 256bit VCVTPS2PD etc. */ + COSTS_N_INSNS (16), /* cost of 512bit VCVTPS2PD etc. */ 1, 1, 1, 1, /* reassoc int, fp, vec_int, vec_fp. */ athlon_memcpy, athlon_memset, @@ -1150,6 +1177,9 @@ struct processor_costs k8_cost = { COSTS_N_INSNS (20), /* cost of DIVSD instruction. */ COSTS_N_INSNS (19), /* cost of SQRTSS instruction. */ COSTS_N_INSNS (27), /* cost of SQRTSD instruction. */ + COSTS_N_INSNS (4), /* cost of CVTSS2SD etc. */ + COSTS_N_INSNS (8), /* cost of 256bit VCVTPS2PD etc. */ + COSTS_N_INSNS (16), /* cost of 512bit VCVTPS2PD etc. */ 1, 1, 1, 1, /* reassoc int, fp, vec_int, vec_fp. */ k8_memcpy, k8_memset, @@ -1281,6 +1311,9 @@ struct processor_costs amdfam10_cost = { COSTS_N_INSNS (20), /* cost of DIVSD instruction. */ COSTS_N_INSNS (19), /* cost of SQRTSS instruction. */ COSTS_N_INSNS (27), /* cost of SQRTSD instruction. */ + COSTS_N_INSNS (4), /* cost of CVTSS2SD etc. */ + COSTS_N_INSNS (8), /* cost of 256bit VCVTPS2PD etc. */ + COSTS_N_INSNS (16), /* cost of 512bit VCVTPS2PD etc. */ 1, 1, 1, 1, /* reassoc int, fp, vec_int, vec_fp. */ amdfam10_memcpy, amdfam10_memset, @@ -1405,6 +1438,9 @@ const struct processor_costs bdver_cost = { COSTS_N_INSNS (27), /* cost of DIVSD instruction. */ COSTS_N_INSNS (15), /* cost of SQRTSS instruction. */ COSTS_N_INSNS (26), /* cost of SQRTSD instruction. */ + COSTS_N_INSNS (4), /* cost of CVTSS2SD etc. */ + COSTS_N_INSNS (7), /* cost of 256bit VCVTPS2PD etc. */ + COSTS_N_INSNS (14), /* cost of 512bit VCVTPS2PD etc. */ 1, 2, 1, 1, /* reassoc int, fp, vec_int, vec_fp. */ bdver_memcpy, bdver_memset, @@ -1553,6 +1589,10 @@ struct processor_costs znver1_cost = { COSTS_N_INSNS (13), /* cost of DIVSD instruction. */ COSTS_N_INSNS (10), /* cost of SQRTSS instruction. */ COSTS_N_INSNS (15), /* cost of SQRTSD instruction. */ + COSTS_N_INSNS (3), /* cost of CVTSS2SD etc. */ + /* Real latency is 4, but for split regs multiply cost of half op by 2. */ + COSTS_N_INSNS (6), /* cost of 256bit VCVTPS2PD etc. */ + COSTS_N_INSNS (12), /* cost of 512bit VCVTPS2PD etc. */ /* Zen can execute 4 integer operations per cycle. FP operations take 3 cycles and it can execute 2 integer additions and 2 multiplications thus reassociation may make sense up to with of 6. SPEC2k6 bencharks suggests @@ -1712,6 +1752,9 @@ struct processor_costs znver2_cost = { COSTS_N_INSNS (13), /* cost of DIVSD instruction. */ COSTS_N_INSNS (10), /* cost of SQRTSS instruction. */ COSTS_N_INSNS (15), /* cost of SQRTSD instruction. */ + COSTS_N_INSNS (3), /* cost of CVTSS2SD etc. */ + COSTS_N_INSNS (5), /* cost of 256bit VCVTPS2PD etc. */ + COSTS_N_INSNS (10), /* cost of 512bit VCVTPS2PD etc. */ /* Zen can execute 4 integer operations per cycle. FP operations take 3 cycles and it can execute 2 integer additions and 2 multiplications thus reassociation may make sense up to with of 6. @@ -1847,6 +1890,9 @@ struct processor_costs znver3_cost = { COSTS_N_INSNS (13), /* cost of DIVSD instruction. */ COSTS_N_INSNS (10), /* cost of SQRTSS instruction. */ COSTS_N_INSNS (15), /* cost of SQRTSD instruction. */ + COSTS_N_INSNS (3), /* cost of CVTSS2SD etc. */ + COSTS_N_INSNS (5), /* cost of 256bit VCVTPS2PD etc. */ + COSTS_N_INSNS (10), /* cost of 512bit VCVTPS2PD etc. */ /* Zen can execute 4 integer operations per cycle. FP operations take 3 cycles and it can execute 2 integer additions and 2 multiplications thus reassociation may make sense up to with of 6. @@ -1984,6 +2030,10 @@ struct processor_costs znver4_cost = { COSTS_N_INSNS (13), /* cost of DIVSD instruction. */ COSTS_N_INSNS (15), /* cost of SQRTSS instruction. */ COSTS_N_INSNS (21), /* cost of SQRTSD instruction. */ + COSTS_N_INSNS (3), /* cost of CVTSS2SD etc. */ + COSTS_N_INSNS (5), /* cost of 256bit VCVTPS2PD etc. */ + /* Real latency is 6, but for split regs multiply cost of half op by 2. */ + COSTS_N_INSNS (10), /* cost of 512bit VCVTPS2PD etc. */ /* Zen can execute 4 integer operations per cycle. FP operations take 3 cycles and it can execute 2 integer additions and 2 multiplications thus reassociation may make sense up to with of 6. @@ -2120,7 +2170,7 @@ struct processor_costs znver5_cost = { COSTS_N_INSNS (1), /* cost of cheap SSE instruction. */ /* ADDSS has throughput 2 and latency 2 (in some cases when source is another addition). */ - COSTS_N_INSNS (2), /* cost of ADDSS/SD SUBSS/SD insns. */ + COSTS_N_INSNS (3), /* cost of ADDSS/SD SUBSS/SD insns. */ /* MULSS has throughput 2 and latency 3. */ COSTS_N_INSNS (3), /* cost of MULSS instruction. */ COSTS_N_INSNS (3), /* cost of MULSD instruction. */ @@ -2135,6 +2185,9 @@ struct processor_costs znver5_cost = { COSTS_N_INSNS (14), /* cost of SQRTSS instruction. */ /* DIVSD has throughtput 0.13 and latency 20. */ COSTS_N_INSNS (20), /* cost of SQRTSD instruction. */ + COSTS_N_INSNS (3), /* cost of CVTSS2SD etc. */ + COSTS_N_INSNS (5), /* cost of 256bit VCVTPS2PD etc. */ + COSTS_N_INSNS (5), /* cost of 512bit VCVTPS2PD etc. */ /* Zen5 can execute: - integer ops: 6 per cycle, at most 3 multiplications. latency 1 for additions, 3 for multiplications (pipelined) @@ -2274,6 +2327,9 @@ struct processor_costs skylake_cost = { COSTS_N_INSNS (14), /* cost of DIVSD instruction. */ COSTS_N_INSNS (12), /* cost of SQRTSS instruction. */ COSTS_N_INSNS (18), /* cost of SQRTSD instruction. */ + COSTS_N_INSNS (2), /* cost of CVTSS2SD etc. */ + COSTS_N_INSNS (2), /* cost of 256bit VCVTPS2PD etc. */ + COSTS_N_INSNS (4), /* cost of 512bit VCVTPS2PD etc. */ 1, 4, 2, 2, /* reassoc int, fp, vec_int, vec_fp. */ skylake_memcpy, skylake_memset, @@ -2403,6 +2459,9 @@ struct processor_costs icelake_cost = { COSTS_N_INSNS (14), /* cost of DIVSD instruction. */ COSTS_N_INSNS (12), /* cost of SQRTSS instruction. */ COSTS_N_INSNS (18), /* cost of SQRTSD instruction. */ + COSTS_N_INSNS (2), /* cost of CVTSS2SD etc. */ + COSTS_N_INSNS (2), /* cost of 256bit VCVTPS2PD etc. */ + COSTS_N_INSNS (2), /* cost of 512bit VCVTPS2PD etc. */ 1, 4, 2, 2, /* reassoc int, fp, vec_int, vec_fp. */ icelake_memcpy, icelake_memset, @@ -2526,6 +2585,9 @@ struct processor_costs alderlake_cost = { COSTS_N_INSNS (17), /* cost of DIVSD instruction. */ COSTS_N_INSNS (14), /* cost of SQRTSS instruction. */ COSTS_N_INSNS (18), /* cost of SQRTSD instruction. */ + COSTS_N_INSNS (2), /* cost of CVTSS2SD etc. */ + COSTS_N_INSNS (2), /* cost of 256bit VCVTPS2PD etc. */ + COSTS_N_INSNS (2), /* cost of 512bit VCVTPS2PD etc. */ 1, 4, 3, 3, /* reassoc int, fp, vec_int, vec_fp. */ alderlake_memcpy, alderlake_memset, @@ -2642,6 +2704,9 @@ const struct processor_costs btver1_cost = { COSTS_N_INSNS (17), /* cost of DIVSD instruction. */ COSTS_N_INSNS (14), /* cost of SQRTSS instruction. */ COSTS_N_INSNS (48), /* cost of SQRTSD instruction. */ + COSTS_N_INSNS (4), /* cost of CVTSS2SD etc. */ + COSTS_N_INSNS (7), /* cost of 256bit VCVTPS2PD etc. */ + COSTS_N_INSNS (14), /* cost of 512bit VCVTPS2PD etc. */ 1, 1, 1, 1, /* reassoc int, fp, vec_int, vec_fp. */ btver1_memcpy, btver1_memset, @@ -2755,6 +2820,9 @@ const struct processor_costs btver2_cost = { COSTS_N_INSNS (19), /* cost of DIVSD instruction. */ COSTS_N_INSNS (16), /* cost of SQRTSS instruction. */ COSTS_N_INSNS (21), /* cost of SQRTSD instruction. */ + COSTS_N_INSNS (4), /* cost of CVTSS2SD etc. */ + COSTS_N_INSNS (7), /* cost of 256bit VCVTPS2PD etc. */ + COSTS_N_INSNS (14), /* cost of 512bit VCVTPS2PD etc. */ 1, 1, 1, 1, /* reassoc int, fp, vec_int, vec_fp. */ btver2_memcpy, btver2_memset, @@ -2867,6 +2935,9 @@ struct processor_costs pentium4_cost = { COSTS_N_INSNS (38), /* cost of DIVSD instruction. */ COSTS_N_INSNS (23), /* cost of SQRTSS instruction. */ COSTS_N_INSNS (38), /* cost of SQRTSD instruction. */ + COSTS_N_INSNS (10), /* cost of CVTSS2SD etc. */ + COSTS_N_INSNS (20), /* cost of 256bit VCVTPS2PD etc. */ + COSTS_N_INSNS (40), /* cost of 512bit VCVTPS2PD etc. */ 1, 1, 1, 1, /* reassoc int, fp, vec_int, vec_fp. */ pentium4_memcpy, pentium4_memset, @@ -2982,6 +3053,9 @@ struct processor_costs nocona_cost = { COSTS_N_INSNS (40), /* cost of DIVSD instruction. */ COSTS_N_INSNS (32), /* cost of SQRTSS instruction. */ COSTS_N_INSNS (41), /* cost of SQRTSD instruction. */ + COSTS_N_INSNS (10), /* cost of CVTSS2SD etc. */ + COSTS_N_INSNS (20), /* cost of 256bit VCVTPS2PD etc. */ + COSTS_N_INSNS (40), /* cost of 512bit VCVTPS2PD etc. */ 1, 1, 1, 1, /* reassoc int, fp, vec_int, vec_fp. */ nocona_memcpy, nocona_memset, @@ -3095,6 +3169,9 @@ struct processor_costs atom_cost = { COSTS_N_INSNS (60), /* cost of DIVSD instruction. */ COSTS_N_INSNS (31), /* cost of SQRTSS instruction. */ COSTS_N_INSNS (63), /* cost of SQRTSD instruction. */ + COSTS_N_INSNS (6), /* cost of CVTSS2SD etc. */ + COSTS_N_INSNS (12), /* cost of 256bit VCVTPS2PD etc. */ + COSTS_N_INSNS (24), /* cost of 512bit VCVTPS2PD etc. */ 2, 2, 2, 2, /* reassoc int, fp, vec_int, vec_fp. */ atom_memcpy, atom_memset, @@ -3208,6 +3285,9 @@ struct processor_costs slm_cost = { COSTS_N_INSNS (69), /* cost of DIVSD instruction. */ COSTS_N_INSNS (20), /* cost of SQRTSS instruction. */ COSTS_N_INSNS (35), /* cost of SQRTSD instruction. */ + COSTS_N_INSNS (3), /* cost of CVTSS2SD etc. */ + COSTS_N_INSNS (6), /* cost of 256bit VCVTPS2PD etc. */ + COSTS_N_INSNS (12), /* cost of 512bit VCVTPS2PD etc. */ 1, 2, 1, 1, /* reassoc int, fp, vec_int, vec_fp. */ slm_memcpy, slm_memset, @@ -3335,6 +3415,9 @@ struct processor_costs tremont_cost = { COSTS_N_INSNS (17), /* cost of DIVSD instruction. */ COSTS_N_INSNS (14), /* cost of SQRTSS instruction. */ COSTS_N_INSNS (18), /* cost of SQRTSD instruction. */ + COSTS_N_INSNS (3), /* cost of CVTSS2SD etc. */ + COSTS_N_INSNS (6), /* cost of 256bit VCVTPS2PD etc. */ + COSTS_N_INSNS (12), /* cost of 512bit VCVTPS2PD etc. */ 1, 4, 3, 3, /* reassoc int, fp, vec_int, vec_fp. */ tremont_memcpy, tremont_memset, @@ -3448,6 +3531,9 @@ struct processor_costs intel_cost = { COSTS_N_INSNS (20), /* cost of DIVSD instruction. */ COSTS_N_INSNS (40), /* cost of SQRTSS instruction. */ COSTS_N_INSNS (40), /* cost of SQRTSD instruction. */ + COSTS_N_INSNS (8), /* cost of CVTSS2SD etc. */ + COSTS_N_INSNS (16), /* cost of 256bit VCVTPS2PD etc. */ + COSTS_N_INSNS (32), /* cost of 512bit VCVTPS2PD etc. */ 1, 4, 1, 1, /* reassoc int, fp, vec_int, vec_fp. */ intel_memcpy, intel_memset, @@ -3566,6 +3652,9 @@ struct processor_costs lujiazui_cost = { COSTS_N_INSNS (17), /* cost of DIVSD instruction. */ COSTS_N_INSNS (32), /* cost of SQRTSS instruction. */ COSTS_N_INSNS (60), /* cost of SQRTSD instruction. */ + COSTS_N_INSNS (3), /* cost of CVTSS2SD etc. */ + COSTS_N_INSNS (6), /* cost of 256bit VCVTPS2PD etc. */ + COSTS_N_INSNS (12), /* cost of 512bit VCVTPS2PD etc. */ 1, 4, 3, 3, /* reassoc int, fp, vec_int, vec_fp. */ lujiazui_memcpy, lujiazui_memset, @@ -3682,6 +3771,9 @@ struct processor_costs yongfeng_cost = { COSTS_N_INSNS (14), /* cost of DIVSD instruction. */ COSTS_N_INSNS (20), /* cost of SQRTSS instruction. */ COSTS_N_INSNS (35), /* cost of SQRTSD instruction. */ + COSTS_N_INSNS (3), /* cost of CVTSS2SD etc. */ + COSTS_N_INSNS (6), /* cost of 256bit VCVTPS2PD etc. */ + COSTS_N_INSNS (12), /* cost of 512bit VCVTPS2PD etc. */ 4, 4, 4, 4, /* reassoc int, fp, vec_int, vec_fp. */ yongfeng_memcpy, yongfeng_memset, @@ -3798,6 +3890,9 @@ struct processor_costs shijidadao_cost = { COSTS_N_INSNS (14), /* cost of DIVSD instruction. */ COSTS_N_INSNS (11), /* cost of SQRTSS instruction. */ COSTS_N_INSNS (18), /* cost of SQRTSD instruction. */ + COSTS_N_INSNS (3), /* cost of CVTSS2SD etc. */ + COSTS_N_INSNS (6), /* cost of 256bit VCVTPS2PD etc. */ + COSTS_N_INSNS (12), /* cost of 512bit VCVTPS2PD etc. */ 4, 4, 4, 4, /* reassoc int, fp, vec_int, vec_fp. */ shijidadao_memcpy, shijidadao_memset, @@ -3922,6 +4017,9 @@ struct processor_costs generic_cost = { COSTS_N_INSNS (17), /* cost of DIVSD instruction. */ COSTS_N_INSNS (14), /* cost of SQRTSS instruction. */ COSTS_N_INSNS (18), /* cost of SQRTSD instruction. */ + COSTS_N_INSNS (3), /* cost of CVTSS2SD etc. */ + COSTS_N_INSNS (4), /* cost of 256bit VCVTPS2PD etc. */ + COSTS_N_INSNS (5), /* cost of 512bit VCVTPS2PD etc. */ 1, 4, 3, 3, /* reassoc int, fp, vec_int, vec_fp. */ generic_memcpy, generic_memset, @@ -4051,6 +4149,9 @@ struct processor_costs core_cost = { COSTS_N_INSNS (32), /* cost of DIVSD instruction. */ COSTS_N_INSNS (30), /* cost of SQRTSS instruction. */ COSTS_N_INSNS (58), /* cost of SQRTSD instruction. */ + COSTS_N_INSNS (2), /* cost of CVTSS2SD etc. */ + COSTS_N_INSNS (2), /* cost of 256bit VCVTPS2PD etc. */ + COSTS_N_INSNS (2), /* cost of 512bit VCVTPS2PD etc. */ 1, 4, 2, 2, /* reassoc int, fp, vec_int, vec_fp. */ core_memcpy, core_memset, diff --git a/gcc/config/i386/x86-tune-sched.cc b/gcc/config/i386/x86-tune-sched.cc index 685a83c..15d3d91 100644 --- a/gcc/config/i386/x86-tune-sched.cc +++ b/gcc/config/i386/x86-tune-sched.cc @@ -81,6 +81,14 @@ ix86_issue_rate (void) case PROCESSOR_YONGFENG: case PROCESSOR_SHIJIDADAO: case PROCESSOR_GENERIC: + /* For znver5 decoder can handle 4 or 8 instructions per cycle, + op cache 12 instruction/cycle, dispatch 8 instructions + integer rename 8 instructions and Fp 6 instructions. + + The scheduler, without understanding out of order nature of the CPU + is not going to be able to use more than 4 instructions since that + is limits of the decoders. */ + case PROCESSOR_ZNVER5: return 4; case PROCESSOR_ICELAKE_CLIENT: @@ -91,13 +99,6 @@ ix86_issue_rate (void) return 5; case PROCESSOR_SAPPHIRERAPIDS: - /* For znver5 decoder can handle 4 or 8 instructions per cycle, - op cache 12 instruction/cycle, dispatch 8 instructions - integer rename 8 instructions and Fp 6 instructions. - - The scheduler, without understanding out of order nature of the CPU - is unlikely going to be able to fill all of these. */ - case PROCESSOR_ZNVER5: return 6; default: diff --git a/gcc/config/mips/mips.cc b/gcc/config/mips/mips.cc index 24a28dc..0d3d026 100644 --- a/gcc/config/mips/mips.cc +++ b/gcc/config/mips/mips.cc @@ -20678,6 +20678,9 @@ mips_option_override (void) "-mcompact-branches=never"); } + if (is_micromips && TARGET_MSA) + error ("unsupported combination: %s", "-mmicromips -mmsa"); + /* Require explicit relocs for MIPS R6 onwards. This enables simplification of the compact branch and jump support through the backend. */ if (!TARGET_EXPLICIT_RELOCS && mips_isa_rev >= 6) diff --git a/gcc/config/nvptx/mkoffload.cc b/gcc/config/nvptx/mkoffload.cc index bdfe7f5..e7ec0ef 100644 --- a/gcc/config/nvptx/mkoffload.cc +++ b/gcc/config/nvptx/mkoffload.cc @@ -778,6 +778,9 @@ main (int argc, char **argv) } if (fopenmp) obstack_ptr_grow (&argv_obstack, "-mgomp"); + /* The host code may contain exception handling constructs. + Handle these as good as we can. */ + obstack_ptr_grow (&argv_obstack, "-mfake-exceptions"); for (int ix = 1; ix != argc; ix++) { diff --git a/gcc/config/nvptx/nvptx.cc b/gcc/config/nvptx/nvptx.cc index 87364bf..f893971 100644 --- a/gcc/config/nvptx/nvptx.cc +++ b/gcc/config/nvptx/nvptx.cc @@ -2359,7 +2359,25 @@ nvptx_assemble_integer (rtx x, unsigned int size, int ARG_UNUSED (aligned_p)) { gcc_checking_assert (!init_frag.active); /* Just use the default machinery; it's not getting used, anyway. */ - return default_assemble_integer (x, size, aligned_p); + bool ok = default_assemble_integer (x, size, aligned_p); + /* ..., but a few cases need special handling. */ + switch (GET_CODE (x)) + { + case SYMBOL_REF: + /* The default machinery won't work: we don't define the necessary + operations; don't use them outside of this. */ + gcc_checking_assert (!ok); + { + /* Just emit something; it's not getting used, anyway. */ + const char *op = "\t.symbol_ref\t"; + ok = (assemble_integer_with_op (op, x), true); + } + break; + + default: + break; + } + return ok; } gcc_checking_assert (init_frag.active); @@ -7771,6 +7789,18 @@ nvptx_asm_output_def_from_decls (FILE *stream, tree name, #endif cgraph_node *cnode = cgraph_node::get (name); +#ifdef ACCEL_COMPILER + /* For nvptx offloading, make sure to emit C++ constructor, destructor aliases [PR97106] + + For some reason (yet to be analyzed), they're not 'cnode->referred_to_p ()'. + (..., or that's not the right approach at all; + <https://inbox.sourceware.org/87v7rx8lbx.fsf@euler.schwinge.ddns.net> + "Re: [committed][nvptx] Use .alias directive for mptx >= 6.3"). */ + if (DECL_CXX_CONSTRUCTOR_P (name) + || DECL_CXX_DESTRUCTOR_P (name)) + ; + else +#endif if (!cnode->referred_to_p ()) /* Prevent "Internal error: reference to deleted section". */ return; @@ -7875,8 +7905,6 @@ nvptx_asm_output_def_from_decls (FILE *stream, tree name, #define TARGET_ASM_DECLARE_CONSTANT_NAME nvptx_asm_declare_constant_name #undef TARGET_USE_BLOCKS_FOR_CONSTANT_P #define TARGET_USE_BLOCKS_FOR_CONSTANT_P hook_bool_mode_const_rtx_true -#undef TARGET_ASM_NEED_VAR_DECL_BEFORE_USE -#define TARGET_ASM_NEED_VAR_DECL_BEFORE_USE true #undef TARGET_MACHINE_DEPENDENT_REORG #define TARGET_MACHINE_DEPENDENT_REORG nvptx_reorg diff --git a/gcc/config/nvptx/nvptx.md b/gcc/config/nvptx/nvptx.md index 3201247..7c3bd69 100644 --- a/gcc/config/nvptx/nvptx.md +++ b/gcc/config/nvptx/nvptx.md @@ -1644,7 +1644,9 @@ [(const_int 0)] "" { - sorry ("exception handling not supported"); + if (!fake_exceptions) + sorry ("exception handling not supported"); + DONE; }) (define_expand "nonlocal_goto" diff --git a/gcc/config/nvptx/nvptx.opt b/gcc/config/nvptx/nvptx.opt index 9be81ae..ce9fbc7 100644 --- a/gcc/config/nvptx/nvptx.opt +++ b/gcc/config/nvptx/nvptx.opt @@ -168,6 +168,14 @@ Target Var(nvptx_alias) Init(0) Undocumented mexperimental Target Var(nvptx_experimental) Init(0) Undocumented +mfake-exceptions +Target Var(fake_exceptions) Init(0) Undocumented +; With '-mfake-exceptions' enabled, the user-visible behavior in presence of +; exception handling constructs changes such that the compile-time +; 'sorry, unimplemented: exception handling not supported' is skipped, code +; generation proceeds, and instead, exception handling constructs 'abort' at +; run time. (..., or don't, if they're in dead code.) + mfake-ptx-alloca Target Var(nvptx_fake_ptx_alloca) Init(0) Undocumented ; With '-mfake-ptx-alloca' enabled, the user-visible behavior changes only diff --git a/gcc/config/riscv/bitmanip.md b/gcc/config/riscv/bitmanip.md index 5ed5e18..d0919ec 100644 --- a/gcc/config/riscv/bitmanip.md +++ b/gcc/config/riscv/bitmanip.md @@ -908,6 +908,24 @@ "bext\t%0,%1,%2" [(set_attr "type" "bitmanip")]) +;; We do not define SHIFT_COUNT_TRUNCATED, so we have to have variants +;; that mask/extend the count if we want to eliminate those ops +;; +;; We could (in theory) use GPR for the various modes, but I haven't +;; seen those cases appear in practice. Without a testcase I've +;; elected to keep the modes X which is easy to reason about. +(define_insn "*bext<mode>_mask_pos" + [(set (match_operand:X 0 "register_operand" "=r") + (zero_extract:X (match_operand:X 1 "register_operand" "r") + (const_int 1) + (and:X + (match_operand:X 2 "register_operand" "r") + (match_operand 3 "const_int_operand"))))] + "(TARGET_ZBS + && INTVAL (operands[3]) + 1 == GET_MODE_BITSIZE (<MODE>mode))" + "bext\t%0,%1,%2" + [(set_attr "type" "bitmanip")]) + ;; This is a bext followed by a seqz. Normally this would be a 3->2 split ;; But the and-not pattern with a constant operand is a define_insn_and_split, ;; so this looks like a 2->2 split, which combine rejects. So implement it @@ -1245,3 +1263,41 @@ expand_crc_using_clmul (<SUBX:MODE>mode, <SUBX1:MODE>mode, operands); DONE; }) + +;; If we have an XOR/IOR with a constant operand (C) and the we can +;; synthesize ~C more efficiently than C, then synthesize ~C and use +;; xnor/orn instead. +;; +;; The same can be done for AND, but mvconst_internal's issues get in +;; the way. That's future work. +(define_split + [(set (match_operand:X 0 "register_operand") + (any_or:X (match_operand:X 1 "register_operand") + (match_operand:X 2 "const_int_operand"))) + (clobber (match_operand:X 3 "register_operand"))] + "TARGET_ZBB + && (riscv_const_insns (operands[2], true) + > riscv_const_insns (GEN_INT (~INTVAL (operands[2])), true))" + [(const_int 0)] +{ + /* Get the inverted constant into the temporary register. */ + riscv_emit_move (operands[3], GEN_INT (~INTVAL (operands[2]))); + + /* For xnor, the NOT operation is in a different position. So + we have to customize the split code we generate a bit. + + It is expected that AND will be handled like IOR in the future. */ + if (<CODE> == XOR) + { + rtx x = gen_rtx_XOR (<X:MODE>mode, operands[1], operands[3]); + x = gen_rtx_NOT (<X:MODE>mode, x); + emit_insn (gen_rtx_SET (operands[0], x)); + } + else + { + rtx x = gen_rtx_NOT (<X:MODE>mode, operands[3]); + x = gen_rtx_IOR (<X:MODE>mode, x, operands[1]); + emit_insn (gen_rtx_SET (operands[0], x)); + } + DONE; +}) diff --git a/gcc/config/riscv/gnu.h b/gcc/config/riscv/gnu.h new file mode 100644 index 0000000..047399b --- /dev/null +++ b/gcc/config/riscv/gnu.h @@ -0,0 +1,59 @@ +/* Definitions for RISC-V GNU/Hurd systems with ELF format. + Copyright (C) 1998-2025 Free Software Foundation, Inc. + +This file is part of GCC. + +GCC is free software; you can redistribute it and/or modify +it under the terms of the GNU General Public License as published by +the Free Software Foundation; either version 3, or (at your option) +any later version. + +GCC is distributed in the hope that it will be useful, +but WITHOUT ANY WARRANTY; without even the implied warranty of +MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the +GNU General Public License for more details. + +You should have received a copy of the GNU General Public License +along with GCC; see the file COPYING3. If not see +<http://www.gnu.org/licenses/>. */ + +#define TARGET_OS_CPP_BUILTINS() \ + do { \ + GNU_USER_TARGET_OS_CPP_BUILTINS(); \ + } while (0) + +#define GNU_USER_DYNAMIC_LINKER "/lib/ld-riscv" XLEN_SPEC "-" ABI_SPEC ".so.1" + +#define ICACHE_FLUSH_FUNC "__riscv_flush_icache" + +#define CPP_SPEC "%{pthread:-D_REENTRANT}" + +#define LD_EMUL_SUFFIX \ + "%{mabi=lp64d:}" \ + "%{mabi=lp64f:_lp64f}" \ + "%{mabi=lp64:_lp64}" \ + "%{mabi=ilp32d:}" \ + "%{mabi=ilp32f:_ilp32f}" \ + "%{mabi=ilp32:_ilp32}" + +#define LINK_SPEC "\ +-melf" XLEN_SPEC DEFAULT_ENDIAN_SPEC "riscv" LD_EMUL_SUFFIX " \ +%{mno-relax:--no-relax} \ +-X \ +%{mbig-endian:-EB} \ +%{mlittle-endian:-EL} \ +%{shared} \ + %{!shared: \ + %{!static: \ + %{!static-pie: \ + %{rdynamic:-export-dynamic} \ + -dynamic-linker " GNU_USER_DYNAMIC_LINKER "}} \ + %{static:-static} %{static-pie:-static -pie --no-dynamic-linker -z text}}" + +#define STARTFILE_PREFIX_SPEC \ + "/lib" XLEN_SPEC "/" ABI_SPEC "/ " \ + "/usr/lib" XLEN_SPEC "/" ABI_SPEC "/ " \ + "/lib/ " \ + "/usr/lib/ " + +#define RISCV_USE_CUSTOMISED_MULTI_LIB select_by_abi diff --git a/gcc/config/riscv/multilib-generator b/gcc/config/riscv/multilib-generator index 4828016..6ad1cf0 100755 --- a/gcc/config/riscv/multilib-generator +++ b/gcc/config/riscv/multilib-generator @@ -159,8 +159,8 @@ for cmodel in cmodels: "e.g. rv32imafd-ilp32--" % cfg) sys.exit(1) - # Compact code model only support rv64. - if cmodel == "compact" and arch.startswith("rv32"): + # Large code model only support rv64. + if cmodel == "large" and arch.startswith("rv32"): continue arch = arch_canonicalize (arch, args.misa_spec) diff --git a/gcc/config/riscv/riscv-target-attr.cc b/gcc/config/riscv/riscv-target-attr.cc index 1d96865..8ad3025 100644 --- a/gcc/config/riscv/riscv-target-attr.cc +++ b/gcc/config/riscv/riscv-target-attr.cc @@ -257,11 +257,7 @@ riscv_target_attr_parser::update_settings (struct gcc_options *opts) const { std::string local_arch = m_subset_list->to_string (true); const char* local_arch_str = local_arch.c_str (); - struct cl_target_option *default_opts - = TREE_TARGET_OPTION (target_option_default_node); - if (opts->x_riscv_arch_string != default_opts->x_riscv_arch_string) - free (CONST_CAST (void *, (const void *) opts->x_riscv_arch_string)); - opts->x_riscv_arch_string = xstrdup (local_arch_str); + opts->x_riscv_arch_string = ggc_strdup (local_arch_str); riscv_set_arch_by_subset_list (m_subset_list, opts); } diff --git a/gcc/config/riscv/riscv-vsetvl.cc b/gcc/config/riscv/riscv-vsetvl.cc index 0ac2538..a8c9256 100644 --- a/gcc/config/riscv/riscv-vsetvl.cc +++ b/gcc/config/riscv/riscv-vsetvl.cc @@ -685,7 +685,7 @@ invalid_opt_bb_p (basic_block cfg_bb) /* We only do LCM optimizations on blocks that are post dominated by EXIT block, that is, we don't do LCM optimizations on infinite loop. */ FOR_EACH_EDGE (e, ei, cfg_bb->succs) - if (e->flags & EDGE_FAKE) + if ((e->flags & EDGE_FAKE) || (e->flags & EDGE_ABNORMAL)) return true; return false; @@ -2698,6 +2698,7 @@ pre_vsetvl::compute_lcm_local_properties () m_avout = sbitmap_vector_alloc (last_basic_block_for_fn (cfun), num_exprs); bitmap_vector_clear (m_avloc, last_basic_block_for_fn (cfun)); + bitmap_vector_clear (m_kill, last_basic_block_for_fn (cfun)); bitmap_vector_clear (m_antloc, last_basic_block_for_fn (cfun)); bitmap_vector_ones (m_transp, last_basic_block_for_fn (cfun)); @@ -2749,6 +2750,10 @@ pre_vsetvl::compute_lcm_local_properties () if (invalid_opt_bb_p (bb->cfg_bb ())) { + if (dump_file && (dump_flags & TDF_DETAILS)) + fprintf (dump_file, "\n --- skipping bb %u due to weird edge", + bb->index ()); + bitmap_clear (m_antloc[bb_index]); bitmap_clear (m_transp[bb_index]); } @@ -3022,6 +3027,18 @@ pre_vsetvl::earliest_fuse_vsetvl_info (int iter) continue; } + /* We cannot lift a vsetvl into the source block if the block is + not transparent WRT to it. + This is too restrictive for blocks where a register's use only + feeds into vsetvls and no regular insns. One example is the + test rvv/vsetvl/avl_single-68.c which is currently XFAILed for + that reason. + In order to support this case we'd need to check the vsetvl's + AVL operand's uses in the source block and make sure they are + only used in other vsetvls. */ + if (!bitmap_bit_p (m_transp[eg->src->index], expr_index)) + continue; + if (dump_file && (dump_flags & TDF_DETAILS)) { fprintf (dump_file, diff --git a/gcc/config/riscv/riscv.cc b/gcc/config/riscv/riscv.cc index 38f3ae7..bad59e2 100644 --- a/gcc/config/riscv/riscv.cc +++ b/gcc/config/riscv/riscv.cc @@ -10382,7 +10382,7 @@ riscv_file_end () fprintf (asm_out_file, "1:\n"); /* pr_type. */ - fprintf (asm_out_file, "\t.p2align\t3\n"); + fprintf (asm_out_file, "\t.p2align\t%u\n", p2align); fprintf (asm_out_file, "2:\n"); fprintf (asm_out_file, "\t.long\t0xc0000000\n"); /* pr_datasz. */ @@ -13136,9 +13136,6 @@ parse_features_for_version (tree decl, DECL_SOURCE_LOCATION (decl)); gcc_assert (parse_res); - if (arch_string != default_opts->x_riscv_arch_string) - free (CONST_CAST (void *, (const void *) arch_string)); - cl_target_option_restore (&global_options, &global_options_set, &cur_target); } diff --git a/gcc/config/riscv/riscv.h b/gcc/config/riscv/riscv.h index 2bcabd0..2759a4c 100644 --- a/gcc/config/riscv/riscv.h +++ b/gcc/config/riscv/riscv.h @@ -888,7 +888,7 @@ extern enum riscv_cc get_riscv_cc (const rtx use); #define ASM_OUTPUT_OPCODE(STREAM, PTR) \ (PTR) = riscv_asm_output_opcode(STREAM, PTR) -#define JUMP_TABLES_IN_TEXT_SECTION 0 +#define JUMP_TABLES_IN_TEXT_SECTION (riscv_cmodel == CM_LARGE) #define CASE_VECTOR_MODE SImode #define CASE_VECTOR_PC_RELATIVE (riscv_cmodel != CM_MEDLOW) diff --git a/gcc/config/riscv/riscv.md b/gcc/config/riscv/riscv.md index 26a247c..eec9687 100644 --- a/gcc/config/riscv/riscv.md +++ b/gcc/config/riscv/riscv.md @@ -789,7 +789,7 @@ rtx t5 = gen_reg_rtx (DImode); rtx t6 = gen_reg_rtx (DImode); - riscv_emit_binary (PLUS, operands[0], operands[1], operands[2]); + emit_insn (gen_addsi3_extended (t6, operands[1], operands[2])); if (GET_CODE (operands[1]) != CONST_INT) emit_insn (gen_extend_insn (t4, operands[1], DImode, SImode, 0)); else @@ -799,7 +799,10 @@ else t5 = operands[2]; emit_insn (gen_adddi3 (t3, t4, t5)); - emit_insn (gen_extend_insn (t6, operands[0], DImode, SImode, 0)); + rtx t7 = gen_lowpart (SImode, t6); + SUBREG_PROMOTED_VAR_P (t7) = 1; + SUBREG_PROMOTED_SET (t7, SRP_SIGNED); + emit_move_insn (operands[0], t7); riscv_expand_conditional_branch (operands[3], NE, t6, t3); } @@ -835,8 +838,11 @@ emit_insn (gen_extend_insn (t3, operands[1], DImode, SImode, 0)); else t3 = operands[1]; - riscv_emit_binary (PLUS, operands[0], operands[1], operands[2]); - emit_insn (gen_extend_insn (t4, operands[0], DImode, SImode, 0)); + emit_insn (gen_addsi3_extended (t4, operands[1], operands[2])); + rtx t5 = gen_lowpart (SImode, t4); + SUBREG_PROMOTED_VAR_P (t5) = 1; + SUBREG_PROMOTED_SET (t5, SRP_SIGNED); + emit_move_insn (operands[0], t5); riscv_expand_conditional_branch (operands[3], LTU, t4, t3); } @@ -966,7 +972,7 @@ rtx t5 = gen_reg_rtx (DImode); rtx t6 = gen_reg_rtx (DImode); - riscv_emit_binary (MINUS, operands[0], operands[1], operands[2]); + emit_insn (gen_subsi3_extended (t6, operands[1], operands[2])); if (GET_CODE (operands[1]) != CONST_INT) emit_insn (gen_extend_insn (t4, operands[1], DImode, SImode, 0)); else @@ -976,7 +982,10 @@ else t5 = operands[2]; emit_insn (gen_subdi3 (t3, t4, t5)); - emit_insn (gen_extend_insn (t6, operands[0], DImode, SImode, 0)); + rtx t7 = gen_lowpart (SImode, t6); + SUBREG_PROMOTED_VAR_P (t7) = 1; + SUBREG_PROMOTED_SET (t7, SRP_SIGNED); + emit_move_insn (operands[0], t7); riscv_expand_conditional_branch (operands[3], NE, t6, t3); } @@ -1015,8 +1024,11 @@ emit_insn (gen_extend_insn (t3, operands[1], DImode, SImode, 0)); else t3 = operands[1]; - riscv_emit_binary (MINUS, operands[0], operands[1], operands[2]); - emit_insn (gen_extend_insn (t4, operands[0], DImode, SImode, 0)); + emit_insn (gen_subsi3_extended (t4, operands[1], operands[2])); + rtx t5 = gen_lowpart (SImode, t4); + SUBREG_PROMOTED_VAR_P (t5) = 1; + SUBREG_PROMOTED_SET (t5, SRP_SIGNED); + emit_move_insn (operands[0], t5); riscv_expand_conditional_branch (operands[3], LTU, t3, t4); } diff --git a/gcc/config/rx/rx.md b/gcc/config/rx/rx.md index edb2c96..a3d966e 100644 --- a/gcc/config/rx/rx.md +++ b/gcc/config/rx/rx.md @@ -2541,10 +2541,17 @@ (unspec_volatile:SI [(match_operand:BLK 1 "memory_operand") ;; String1 (match_operand:BLK 2 "memory_operand")] ;; String2 UNSPEC_CMPSTRN)) - (use (match_operand:SI 3 "register_operand")) ;; Max Length + (use (match_operand:SI 3 "nonmemory_operand")) ;; Max Length (match_operand:SI 4 "immediate_operand")] ;; Known Align "rx_allow_string_insns" { + bool const_len = CONST_INT_P (operands[3]); + if (const_len && operands[3] == CONST0_RTX (SImode)) + { + emit_move_insn (operands[0], CONST0_RTX (SImode)); + DONE; + } + rtx str1 = gen_rtx_REG (SImode, 1); rtx str2 = gen_rtx_REG (SImode, 2); rtx len = gen_rtx_REG (SImode, 3); @@ -2553,6 +2560,13 @@ emit_move_insn (str2, force_operand (XEXP (operands[2], 0), NULL_RTX)); emit_move_insn (len, operands[3]); + /* Set flags in case len is zero */ + if (!const_len) + { + emit_insn (gen_setpsw (GEN_INT ('C'))); + emit_insn (gen_setpsw (GEN_INT ('Z'))); + } + emit_insn (gen_rx_cmpstrn (operands[0], operands[1], operands[2])); DONE; } @@ -2590,9 +2604,7 @@ (clobber (reg:SI 3)) (clobber (reg:CC CC_REG))] "rx_allow_string_insns" - "setpsw z ; Set flags in case len is zero - setpsw c - scmpu ; Perform the string comparison + "scmpu ; Perform the string comparison mov #-1, %0 ; Set up -1 result (which cannot be created ; by the SC insn) bnc ?+ ; If Carry is not set skip over diff --git a/gcc/config/s390/9175.md b/gcc/config/s390/9175.md new file mode 100644 index 0000000..d0ac0e1 --- /dev/null +++ b/gcc/config/s390/9175.md @@ -0,0 +1,316 @@ +;; Scheduling description for z17. +;; Copyright (C) 2025 Free Software Foundation, Inc. + +;; This file is part of GCC. + +;; GCC is free software; you can redistribute it and/or modify it under +;; the terms of the GNU General Public License as published by the Free +;; Software Foundation; either version 3, or (at your option) any later +;; version. + +;; GCC is distributed in the hope that it will be useful, but WITHOUT ANY +;; WARRANTY; without even the implied warranty of MERCHANTABILITY or +;; FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License +;; for more details. + +;; You should have received a copy of the GNU General Public License +;; along with GCC; see the file COPYING3. If not see +;; <http://www.gnu.org/licenses/>. + +(define_attr "z17_unit_fpd" "" + (cond [(eq_attr "mnemonic" "ddb,ddbr,deb,debr,dxbr,sqdb,sqdbr,sqeb,sqebr,\ +sqxbr,vdf,vdg,vdlf,vdlg,vdlq,vdq,vfddb,vfdsb,vfsqdb,vfsqsb,vrf,vrg,vrlf,vrlg,\ +vrlq,vrq,wfddb,wfdsb,wfdxb,wfsqdb,wfsqxb") + (const_int 1)] (const_int 0))) + +(define_attr "z17_unit_fxa" "" + (cond [(eq_attr "mnemonic" "a,afi,ag,agf,agfi,agfr,agh,aghi,aghik,agr,agrk,ah,\ +ahi,ahik,ahy,al,alc,alcg,alcgr,alcr,alfi,alg,algf,algfi,algfr,alghsik,algr,\ +algrk,alhsik,alr,alrk,aly,ar,ark,ay,bdepg,bextg,clzg,ctzg,etnd,flogr,ic,icm,\ +icmh,icmy,icy,iihf,iilf,ipm,la,larl,lay,lb,lbr,lcgfr,lcgr,lcr,lgb,lgbr,lgf,\ +lgfi,lgfr,lgfrl,lgh,lghi,lghr,lghrl,lgr,lh,lhi,lhr,lhrl,lhy,llcr,llgcr,llgfr,\ +llghr,llgtr,llhr,llihf,llihh,llihl,llilf,llilh,llill,llxab,llxaf,llxag,llxah,\ +llxaq,lngfr,lngr,lnr,loc,locg,locghi,locgr,lochi,locr,lpgfr,lpgr,lpr,lr,lrv,\ +lrvg,lrvgr,lrvh,lrvr,lt,ltg,ltgf,ltgfr,ltgr,ltr,lxab,lxaf,lxag,lxah,lxaq,m,mfy,\ +mg,mgh,mghi,mgrk,mh,mhi,mhy,ml,mlg,mlgr,mlr,mr,ms,msc,msfi,msg,msgc,msgf,msgfi,\ +msgfr,msgr,msgrkc,msr,msrkc,msy,n,ncgrk,ncrk,ng,ngr,ngrk,nihf,nihh,nihl,nilf,\ +nilh,nill,nngrk,nnrk,nogrk,nork,nr,nrk,nxgrk,nxrk,ny,o,ocgrk,ocrk,og,ogr,ogrk,\ +oihf,oihh,oihl,oilf,oilh,oill,or,ork,oy,pfpo,popcnt,risbg,risbgn,rll,rllg,\ +rnsbg,rosbg,rxsbg,s,selgr,selr,sg,sgf,sgfr,sgh,sgr,sgrk,sh,shy,sl,slb,slbg,\ +slbgr,slbr,sldl,slfi,slg,slgf,slgfi,slgfr,slgr,slgrk,sll,sllg,sllk,slr,slrk,\ +sly,sr,sra,srag,srak,srda,srdl,srk,srl,srlg,srlk,sy,x,xg,xgr,xgrk,xihf,xilf,xr,\ +xrk,xy") + (const_int 1)] (const_int 0))) + +(define_attr "z17_unit_fxb" "" + (cond [(eq_attr "mnemonic" "agsi,algsi,alsi,asi,b,bc,bcr,bi,br,c,cfi,cg,cgf,\ +cgfi,cgfr,cgfrl,cgh,cghi,cghrl,cghsi,cgit,cgr,cgrl,cgrt,ch,chi,chrl,chsi,chy,\ +cit,cl,clfhsi,clfi,clfit,clg,clgf,clgfi,clgfr,clgfrl,clghrl,clghsi,clgit,clgr,\ +clgrl,clgrt,clgt,clhhsi,clhrl,cli,cliy,clm,clmy,clr,clrl,clrt,clt,cly,cr,crl,\ +crt,cy,laa,laag,lan,lang,lao,laog,lat,lax,laxg,lcdfr,ldgr,ldr,lgat,lgdr,lndfr,\ +lpdfr,lxr,lzdr,lzer,lzxr,mvghi,mvhhi,mvhi,mvi,mviy,ni,niy,nop,nopr,ntstg,oi,\ +oiy,ppa,st,stc,stcy,std,stdy,ste,stey,stg,stgrl,sth,sthrl,sthy,stoc,stocg,strl,\ +strv,strvg,strvh,sty,tend,tm,tmh,tmhh,tmhl,tml,tmlh,tmll,tmy,vlgvb,vlgvf,vlgvg,\ +vlgvh,vlr,vlvgb,vlvgf,vlvgg,vlvgh,vlvgp,vscef,vsceg,vst,vstbrf,vstbrg,vstbrh,\ +vstbrq,vstebrf,vstebrg,vstef,vsteg,vsterf,vsterg,vsterh,vstl,vstrl,vstrlr,xi,\ +xiy") + (const_int 1)] (const_int 0))) + +(define_attr "z17_unit_fxd" "" + (cond [(eq_attr "mnemonic" "dlgr,dlr,dr,dsgfr,dsgr") + (const_int 1)] (const_int 0))) + +(define_attr "z17_unit_lsu" "" + (cond [(eq_attr "mnemonic" "clc,ear,l,lam,lcbb,ld,lde,ldy,lg,lgrl,llc,llgc,\ +llgf,llgfrl,llgh,llghrl,llgt,llh,llhrl,lm,lmg,lmy,lpq,lrl,ly,mvcrl,sar,sfpc,\ +tabort,vl,vlbb,vlbrf,vlbrg,vlbrh,vlbrq,vlbrrepf,vlbrrepg,vlbrreph,vlerf,vlerg,\ +vlerh,vll,vllebrzf,vllebrzg,vllebrzh,vllezb,vllezf,vllezg,vllezh,vllezlf,\ +vlrepb,vlrepf,vlrepg,vlreph,vlrl,vlrlr") + (const_int 1)] (const_int 0))) + +(define_attr "z17_unit_vfu" "" + (cond [(eq_attr "mnemonic" "adb,adbr,adtr,aeb,aebr,axbr,axtr,brcl,cdb,cdbr,\ +cdtr,ceb,cebr,cpsdr,cxbr,cxtr,ddtr,dxtr,fidbr,fidbra,fidtr,fiebr,fiebra,fixbr,\ +fixbra,fixtr,j,jg,kdb,kdbr,kdtr,keb,kebr,kxbr,kxtr,lcdbr,lcebr,lcxbr,ldeb,\ +ldebr,ldetr,le,ledbr,ledtr,ler,ley,lndbr,lnebr,lnxbr,lpdbr,lpebr,lpxbr,ltdbr,\ +ltdtr,ltebr,ltxbr,ltxtr,lxdb,lxdbr,lxdtr,lxeb,lxebr,madb,madbr,maeb,maebr,mdb,\ +mdbr,mdtr,meeb,meebr,msdb,msdbr,mseb,msebr,mxbr,mxtr,sdb,sdbr,sdtr,seb,sebr,\ +sxbr,sxtr,tcdb,tceb,tcxb,tdcdt,tdcet,tdcxt,vab,vaccb,vacccq,vaccf,vaccg,vacch,\ +vaccq,vacq,vaf,vag,vah,vaq,vavgb,vavgf,vavgg,vavgh,vavglb,vavglf,vavglg,vavglh,\ +vavglq,vavgq,vblendb,vblendf,vblendg,vblendh,vblendq,vbperm,vcdgb,vcdlgb,vcefb,\ +vcelfb,vceqb,vceqbs,vceqf,vceqfs,vceqg,vceqgs,vceqh,vceqhs,vceqq,vceqqs,vcfeb,\ +vcfn,vcgdb,vchb,vchbs,vchf,vchfs,vchg,vchgs,vchh,vchhs,vchlb,vchlbs,vchlf,\ +vchlfs,vchlg,vchlgs,vchlh,vchlhs,vchlq,vchlqs,vchq,vchqs,vcksm,vclfeb,vclfnh,\ +vclfnl,vclgdb,vclzb,vclzf,vclzg,vclzh,vclzq,vcnf,vcrnf,vctzb,vctzf,vctzg,vctzh,\ +vctzq,verimb,verimf,verimg,verimh,verllb,verllf,verllg,verllh,verllvb,verllvf,\ +verllvg,verllvh,veslb,veslf,veslg,veslh,veslvb,veslvf,veslvg,veslvh,vesrab,\ +vesraf,vesrag,vesrah,vesravb,vesravf,vesravg,vesravh,vesrlb,vesrlf,vesrlg,\ +vesrlh,vesrlvb,vesrlvf,vesrlvg,vesrlvh,veval,vfadb,vfasb,vfcedb,vfcedbs,vfcesb,\ +vfcesbs,vfchdb,vfchdbs,vfchedb,vfchedbs,vfchesb,vfchesbs,vfchsb,vfchsbs,vfeeb,\ +vfeef,vfeeh,vfeezbs,vfeezfs,vfeezhs,vfeneb,vfenef,vfeneh,vfenezb,vfenezf,\ +vfenezh,vfidb,vfisb,vfkedb,vfkesb,vfkhdb,vfkhedb,vfkhesb,vfkhsb,vflcdb,vflcsb,\ +vflndb,vflnsb,vflpdb,vflpsb,vfmadb,vfmasb,vfmaxdb,vfmaxsb,vfmdb,vfmindb,\ +vfminsb,vfmsb,vfmsdb,vfmssb,vfnmadb,vfnmasb,vfnmsdb,vfnmssb,vfsdb,vfssb,\ +vftcidb,vftcisb,vgbm,vgemb,vgemf,vgemg,vgemh,vgemq,vgfmab,vgfmaf,vgfmag,vgfmah,\ +vgfmb,vgfmf,vgfmg,vgfmh,vgm,vistrb,vistrbs,vistrf,vistrfs,vistrh,vistrhs,vlcb,\ +vlcf,vlcg,vlch,vldeb,vleb,vlebrf,vlebrg,vlebrh,vledb,vlef,vleg,vleh,vleib,\ +vleif,vleig,vleih,vlpb,vlpf,vlpg,vlph,vlpq,vmaeb,vmaef,vmaeg,vmaeh,vmahb,vmahf,\ +vmahg,vmahh,vmahq,vmalb,vmaleb,vmalef,vmaleg,vmaleh,vmalf,vmalg,vmalhb,vmalhf,\ +vmalhg,vmalhh,vmalhq,vmalhw,vmalob,vmalof,vmalog,vmaloh,vmalq,vmaob,vmaof,\ +vmaog,vmaoh,vmeb,vmef,vmeg,vmeh,vmhb,vmhf,vmhg,vmhh,vmhq,vmlb,vmleb,vmlef,\ +vmleg,vmleh,vmlf,vmlg,vmlhb,vmlhf,vmlhg,vmlhh,vmlhq,vmlhw,vmlob,vmlof,vmlog,\ +vmloh,vmlq,vmnb,vmnf,vmng,vmnh,vmnlb,vmnlf,vmnlg,vmnlh,vmnlq,vmnq,vmob,vmof,\ +vmog,vmoh,vmrhb,vmrhf,vmrhg,vmrhh,vmrlb,vmrlf,vmrlg,vmrlh,vmslg,vmxb,vmxf,vmxg,\ +vmxh,vmxlb,vmxlf,vmxlg,vmxlh,vmxlq,vmxq,vn,vnc,vnn,vno,vnot,vnx,vo,voc,vone,\ +vpdi,vperm,vpkf,vpkg,vpkh,vpklsf,vpklsfs,vpklsg,vpklsgs,vpklsh,vpklshs,vpksf,\ +vpksfs,vpksg,vpksgs,vpksh,vpkshs,vpopct,vpopctb,vpopctf,vpopctg,vpopcth,vrepb,\ +vrepf,vrepg,vreph,vrepi,vrepib,vrepif,vrepig,vrepih,vsb,vsbcbiq,vsbiq,vscbib,\ +vscbif,vscbig,vscbih,vscbiq,vsegb,vsegf,vsegh,vsel,vsf,vsg,vsh,vsl,vslb,vsld,\ +vsldb,vsq,vsra,vsrab,vsrd,vsrl,vsrlb,vsumb,vsumgf,vsumgh,vsumh,vsumqf,vsumqg,\ +vtm,vuphb,vuphf,vuphg,vuphh,vuplb,vuplf,vuplg,vuplhb,vuplhf,vuplhg,vuplhh,\ +vuplhw,vupllb,vupllf,vupllg,vupllh,vx,vzero,wcdgb,wcdlgb,wcefb,wcelfb,wcfeb,\ +wcgdb,wclfeb,wclgdb,wfadb,wfasb,wfaxb,wfcdb,wfcedb,wfcesb,wfcexb,wfcexbs,\ +wfchdb,wfchedb,wfchesb,wfchexb,wfchexbs,wfchsb,wfchxb,wfchxbs,wfcsb,wfcxb,\ +wfidb,wfisb,wfixb,wfkdb,wfkedb,wfkesb,wfkexb,wfkhdb,wfkhedb,wfkhesb,wfkhexb,\ +wfkhsb,wfkhxb,wfksb,wfkxb,wflcdb,wflcsb,wflcxb,wflld,wflndb,wflnsb,wflnxb,\ +wflpdb,wflpsb,wflpxb,wflrx,wfmadb,wfmasb,wfmaxb,wfmaxxb,wfmdb,wfminxb,wfmsb,\ +wfmsdb,wfmssb,wfmsxb,wfmxb,wfnmaxb,wfnmsxb,wfsdb,wfssb,wfsxb,wftcixb,wldeb,\ +wledb") + (const_int 1)] (const_int 0))) + +(define_attr "z17_cracked" "" + (cond [(eq_attr "mnemonic" "bas,basr,bras,brasl,cdfbr,cdftr,cdgbr,cdgtr,\ +cdlfbr,cdlftr,cdlgbr,cdlgtr,cefbr,cegbr,celfbr,celgbr,cfdbr,cfebr,cfxbr,cgdbr,\ +cgdtr,cgebr,cgxbr,cgxtr,chhsi,clfdbr,clfdtr,clfebr,clfxbr,clfxtr,clgdbr,clgdtr,\ +clgebr,clgxbr,clgxtr,cs,csg,csy,efpc,ex,exrl,lcgfr,lngfr,lpgfr,lpq,lxr,lzxr,\ +mvc,nc,oc,rnsbg,rosbg,rxsbg,stpq,vgef,vgeg,vscef,vsceg,vsteb,vstebrh,vsteh,xc") + (const_int 1)] (const_int 0))) + +(define_attr "z17_expanded" "" + (cond [(eq_attr "mnemonic" "cds,cdsg,cdsy,cxfbr,cxftr,cxgbr,cxgtr,cxlfbr,\ +cxlftr,cxlgbr,cxlgtr,d,dl,dlg,dsg,dsgf,lam,lm,lmg,lmy,sldl,srda,srdl,stam,stm,\ +stmg,stmy,tbegin,tbeginc") + (const_int 1)] (const_int 0))) + +(define_attr "z17_groupalone" "" + (cond [(eq_attr "mnemonic" "alc,alcg,alcgr,alcr,axbr,axtr,clc,cxbr,cxfbr,\ +cxftr,cxgbr,cxgtr,cxlfbr,cxlftr,cxlgbr,cxlgtr,cxtr,d,dl,dlg,dlgr,dlr,dr,dsg,\ +dsgf,dsgfr,dsgr,dxbr,dxtr,ex,exrl,fixbr,fixbra,fixtr,flogr,kxbr,kxtr,lcxbr,\ +lnxbr,lpxbr,ltxbr,ltxtr,lxdb,lxdbr,lxdtr,lxeb,lxebr,m,madb,maeb,maebr,mfy,mg,\ +mgrk,ml,mlg,mlgr,mlr,mr,msdb,mseb,msebr,mvc,mvcrl,mxbr,mxtr,nc,oc,ppa,sfpc,slb,\ +slbg,slbgr,slbr,sqxbr,sxbr,sxtr,tabort,tbegin,tbeginc,tcxb,tdcxt,tend,xc") + (const_int 1)] (const_int 0))) + +(define_attr "z17_endgroup" "" + (cond [(eq_attr "mnemonic" "bas,basr,bcr,br,bras,brasl,cdsg,clfebr,cs,csg,csy,\ +efpc,ex,exrl,ipm,lam,lpq,lxr,nopr,sldl,srda,srdl,stam,stm,stmg,stmy,tbegin,\ +tbeginc") + (const_int 1)] (const_int 0))) + +(define_attr "z17_groupoftwo" "" + (cond [(eq_attr "mnemonic" "cdfbr,cdftr,cdgbr,cdgtr,cdlfbr,cdlftr,cdlgbr,\ +cdlgtr,cefbr,cegbr,celfbr,celgbr,cfdbr,cfebr,cfxbr,cgdbr,cgdtr,cgebr,cgxbr,\ +cgxtr,chhsi,clfdbr,clfdtr,clfxbr,clfxtr,clgdbr,clgdtr,clgebr,clgxbr,clgxtr,\ +lcgfr,lngfr,lpgfr,lzxr,vacccq,vacq,vblendb,vblendf,vblendg,vblendh,vblendq,\ +veval,vfmadb,vfmasb,vfmsdb,vfmssb,vfnmadb,vfnmasb,vfnmsdb,vfnmssb,vgef,vgeg,\ +vgfmab,vgfmaf,vgfmag,vgfmah,vmaeb,vmaef,vmaeg,vmaeh,vmahb,vmahf,vmahg,vmahh,\ +vmahq,vmalb,vmaleb,vmalef,vmaleg,vmaleh,vmalf,vmalg,vmalhb,vmalhf,vmalhg,\ +vmalhh,vmalhq,vmalhw,vmalob,vmalof,vmalog,vmaloh,vmalq,vmaob,vmaof,vmaog,vmaoh,\ +vmslg,vperm,vsbcbiq,vsbiq,vscef,vsceg,vsel,vsteb,vstebrh,vsteh,wfmadb,wfmasb,\ +wfmaxb,wfmsdb,wfmssb,wfmsxb,wfnmaxb,wfnmsxb") + (const_int 1)] (const_int 0))) + +(define_insn_reservation "z17_0" 0 + (and (eq_attr "cpu" "z17") + (eq_attr "mnemonic" "a,afi,ag,agfi,aghi,aghik,agr,agrk,ahi,ahik,al,alfi,alg,\ +algf,algfi,algfr,alghsik,algr,algrk,alhsik,alr,alrk,aly,ar,ark,ay,b,bc,bcr,bi,\ +br,brcl,c,cfi,cg,cgfi,cghi,cghsi,cgit,cgr,cgrl,cgrt,chi,chsi,cit,cl,clfhsi,\ +clfi,clfit,clg,clgf,clgfi,clgfr,clgfrl,clghrl,clghsi,clgit,clgr,clgrl,clgrt,\ +clgt,clhhsi,clhrl,cli,cliy,clr,clrl,clrt,clt,cly,cr,crl,crt,cy,etnd,ic,icm,\ +icmh,icmy,icy,iihf,iilf,j,jg,la,larl,lat,lay,lb,lbr,lcdfr,lcgr,lcr,ldgr,ldr,\ +lgat,lgb,lgbr,lgf,lgfi,lgfr,lgfrl,lgh,lghi,lghr,lghrl,lgr,lh,lhi,lhr,lhrl,lhy,\ +llcr,llgcr,llgfr,llghr,llgtr,llhr,llihf,llihh,llihl,llilf,llilh,llill,lndfr,\ +lngr,lnr,lpdfr,lpgr,lpr,lr,lrv,lrvg,lrvgr,lrvh,lrvr,lt,ltg,ltgf,ltgfr,ltgr,ltr,\ +lzdr,lzer,n,ncgrk,ncrk,ng,ngr,ngrk,nihf,nihh,nihl,nilf,nilh,nill,nngrk,nnrk,\ +nogrk,nop,nopr,nork,nr,nrk,nxgrk,nxrk,ny,o,ocgrk,ocrk,og,ogr,ogrk,oihf,oihh,\ +oihl,oilf,oilh,oill,or,ork,oy,pfpo,risbg,risbgn,rll,rllg,s,sg,sgr,sgrk,sl,sldl,\ +slfi,slg,slgf,slgfi,slgfr,slgr,slgrk,sll,sllg,sllk,slr,slrk,sly,sr,sra,srag,\ +srak,srda,srdl,srk,srl,srlg,srlk,sy,tm,tmh,tmhh,tmhl,tml,tmlh,tmll,tmy,vlr,\ +vlvgb,vlvgf,vlvgg,vlvgh,x,xg,xgr,xgrk,xihf,xilf,xr,xrk,xy")) "nothing") + +(define_insn_reservation "z17_1" 1 + (and (eq_attr "cpu" "z17") + (eq_attr "mnemonic" "agf,agfr,agh,agsi,ah,ahy,algsi,alsi,asi,cgf,cgfr,cgfrl,\ +cgh,cghrl,ch,chrl,chy,clm,clmy,cpsdr,laa,laag,lan,lang,lao,laog,lax,laxg,le,\ +ler,ley,llxab,llxaf,llxag,llxah,llxaq,loc,locg,locghi,locgr,lochi,locr,lxab,\ +lxaf,lxag,lxah,lxaq,mvghi,mvhhi,mvhi,mvi,mviy,ni,niy,ntstg,oi,oiy,selgr,selr,\ +sgf,sgfr,sgh,sh,shy,st,stc,stcy,stg,stgrl,sth,sthrl,sthy,stoc,stocg,strl,strv,\ +strvg,strvh,sty,vab,vaccb,vacccq,vaccf,vaccg,vacch,vaccq,vacq,vaf,vag,vah,vaq,\ +vavgb,vavgf,vavgg,vavgh,vavglb,vavglf,vavglg,vavglh,vavglq,vavgq,vblendb,\ +vblendf,vblendg,vblendh,vblendq,vbperm,vceqb,vceqbs,vceqf,vceqfs,vceqg,vceqgs,\ +vceqh,vceqhs,vceqq,vceqqs,vcfn,vchb,vchbs,vchf,vchfs,vchg,vchgs,vchh,vchhs,\ +vchlb,vchlbs,vchlf,vchlfs,vchlg,vchlgs,vchlh,vchlhs,vchlq,vchlqs,vchq,vchqs,\ +vclfnh,vclfnl,vclzb,vclzf,vclzg,vclzh,vclzq,vcnf,vcrnf,vctzb,vctzf,vctzg,vctzh,\ +vctzq,verimb,verimf,verimg,verimh,verllb,verllf,verllg,verllh,verllvb,verllvf,\ +verllvg,verllvh,veslb,veslf,veslg,veslh,veslvb,veslvf,veslvg,veslvh,vesrab,\ +vesraf,vesrag,vesrah,vesravb,vesravf,vesravg,vesravh,vesrlb,vesrlf,vesrlg,\ +vesrlh,vesrlvb,vesrlvf,vesrlvg,vesrlvh,veval,vfcedb,vfcedbs,vfcesb,vfcesbs,\ +vfchdb,vfchdbs,vfchedb,vfchedbs,vfchesb,vfchesbs,vfchsb,vfchsbs,vfkedb,vfkesb,\ +vfkhdb,vfkhedb,vfkhesb,vfkhsb,vflcdb,vflcsb,vflndb,vflnsb,vflpdb,vflpsb,\ +vfmaxdb,vfmaxsb,vfmindb,vfminsb,vgbm,vgemb,vgemf,vgemg,vgemh,vgemq,vgm,vlcb,\ +vlcf,vlcg,vlch,vleb,vlebrf,vlebrg,vlebrh,vlef,vleg,vleh,vleib,vleif,vleig,\ +vleih,vlpb,vlpf,vlpg,vlph,vlpq,vmnb,vmnf,vmng,vmnh,vmnlb,vmnlf,vmnlg,vmnlh,\ +vmnlq,vmnq,vmrhb,vmrhf,vmrhg,vmrhh,vmrlb,vmrlf,vmrlg,vmrlh,vmxb,vmxf,vmxg,vmxh,\ +vmxlb,vmxlf,vmxlg,vmxlh,vmxlq,vmxq,vn,vnc,vnn,vno,vnot,vnx,vo,voc,vone,vpdi,\ +vperm,vpkf,vpkg,vpkh,vpklsf,vpklsfs,vpklsg,vpklsgs,vpklsh,vpklshs,vpksf,vpksfs,\ +vpksg,vpksgs,vpksh,vpkshs,vpopct,vpopctb,vpopctf,vpopctg,vpopcth,vrepb,vrepf,\ +vrepg,vreph,vrepi,vrepib,vrepif,vrepig,vrepih,vsb,vsbcbiq,vsbiq,vscbib,vscbif,\ +vscbig,vscbih,vscbiq,vsegb,vsegf,vsegh,vsel,vsf,vsg,vsh,vsl,vslb,vsld,vsldb,\ +vsq,vsra,vsrab,vsrd,vsrl,vsrlb,vuphb,vuphf,vuphg,vuphh,vuplb,vuplf,vuplg,\ +vuplhb,vuplhf,vuplhg,vuplhh,vuplhw,vupllb,vupllf,vupllg,vupllh,vx,vzero,wfcedb,\ +wfcesb,wfcexb,wfcexbs,wfchdb,wfchedb,wfchesb,wfchexb,wfchexbs,wfchsb,wfchxb,\ +wfchxbs,wfkedb,wfkesb,wfkexb,wfkhdb,wfkhedb,wfkhesb,wfkhexb,wfkhsb,wfkhxb,\ +wflcdb,wflcsb,wflcxb,wflndb,wflnsb,wflnxb,wflpdb,wflpsb,wflpxb,wfmaxxb,wfminxb,\ +xi,xiy")) "nothing") + +(define_insn_reservation "z17_2" 2 + (and (eq_attr "cpu" "z17") + (eq_attr "mnemonic" "cdb,cdbr,ceb,cebr,clzg,ctzg,ear,ipm,kdb,kdbr,keb,kebr,l,\ +lcbb,lcdbr,lcebr,ld,lde,ldy,lg,lgdr,lgrl,llc,llgc,llgf,llgfrl,llgh,llghrl,llgt,\ +llh,llhrl,lm,lmg,lmy,lndbr,lnebr,lpdbr,lpebr,lrl,ltdbr,ltebr,ly,popcnt,sar,\ +tcdb,tceb,vfeeb,vfeef,vfeeh,vfeezbs,vfeezfs,vfeezhs,vfeneb,vfenef,vfeneh,\ +vfenezb,vfenezf,vfenezh,vftcidb,vftcisb,vistrb,vistrbs,vistrf,vistrfs,vistrh,\ +vistrhs,vlbrrepf,vlbrrepg,vlbrreph,vlgvb,vlgvf,vlgvg,vlgvh,vllebrzf,vllebrzg,\ +vllebrzh,vllezb,vllezf,vllezg,vllezh,vllezlf,vlrepb,vlrepf,vlrepg,vlreph,vlrl,\ +vlvgp,wfcdb,wfcsb,wfcxb,wfkdb,wfksb,wfkxb,wftcixb")) "nothing") + +(define_insn_reservation "z17_3" 3 + (and (eq_attr "cpu" "z17") + (eq_attr "mnemonic" "bdepg,bextg,cds,cdsy,mgh,mghi,mh,mhi,mhy,ms,msc,msfi,msg,\ +msgc,msgf,msgfi,msgfr,msgr,msgrkc,msr,msrkc,msy,std,stdy,ste,stey,vcksm,vgfmab,\ +vgfmaf,vgfmag,vgfmah,vgfmb,vgfmf,vgfmg,vgfmh,vl,vlbb,vlbrf,vlbrg,vlbrh,vlbrq,\ +vlerf,vlerg,vlerh,vll,vlrlr,vmaeb,vmaef,vmaeg,vmaeh,vmahb,vmahf,vmahg,vmahh,\ +vmahq,vmalb,vmaleb,vmalef,vmaleg,vmaleh,vmalf,vmalg,vmalhb,vmalhf,vmalhg,\ +vmalhh,vmalhq,vmalhw,vmalob,vmalof,vmalog,vmaloh,vmalq,vmaob,vmaof,vmaog,vmaoh,\ +vmeb,vmef,vmeg,vmeh,vmhb,vmhf,vmhg,vmhh,vmhq,vmlb,vmleb,vmlef,vmleg,vmleh,vmlf,\ +vmlg,vmlhb,vmlhf,vmlhg,vmlhh,vmlhq,vmlhw,vmlob,vmlof,vmlog,vmloh,vmlq,vmob,\ +vmof,vmog,vmoh,vsumb,vsumgf,vsumgh,vsumh,vsumqf,vsumqg,vtm")) "nothing") + +(define_insn_reservation "z17_4" 4 + (and (eq_attr "cpu" "z17") + (eq_attr "mnemonic" "bas,basr,bras,brasl,chhsi,clc,ex,exrl,lam,lcgfr,lngfr,\ +lpgfr,lxr,lzxr,mvcrl,ppa,rnsbg,rosbg,rxsbg,tabort,tend,vst,vstbrf,vstbrg,\ +vstbrh,vstbrq,vstebrf,vstebrg,vstef,vsteg,vsterf,vsterg,vsterh,vstl,vstrl,\ +vstrlr")) "nothing") + +(define_insn_reservation "z17_5" 5 + (and (eq_attr "cpu" "z17") + (eq_attr "mnemonic" "adb,adbr,aeb,aebr,alc,alcg,alcgr,alcr,cs,csg,csy,fidbr,\ +fidbra,fiebr,fiebra,ldeb,ldebr,ledbr,madbr,mdb,mdbr,meeb,meebr,msdbr,sdb,sdbr,\ +seb,sebr,slb,slbg,slbgr,slbr,stm,stmg,stmy,vcdgb,vcdlgb,vcefb,vcelfb,vcfeb,\ +vcgdb,vclfeb,vclgdb,vldeb,vledb,vmslg,wcdgb,wcdlgb,wcefb,wcelfb,wcfeb,wcgdb,\ +wclfeb,wclgdb,wflld,wflrx,wldeb,wledb")) "nothing") + +(define_insn_reservation "z17_6" 6 + (and (eq_attr "cpu" "z17") + (eq_attr "mnemonic" "sfpc")) "nothing") + +(define_insn_reservation "z17_7" 7 + (and (eq_attr "cpu" "z17") + (eq_attr "mnemonic" "adtr,cdtr,fidtr,kdtr,ldetr,ltdtr,sdtr,tdcdt,tdcet,vfadb,\ +vfasb,vfidb,vfisb,vfsdb,vfssb,vgef,vgeg,wfadb,wfasb,wfaxb,wfidb,wfisb,wfixb,\ +wfsdb,wfssb,wfsxb")) "nothing") + +(define_insn_reservation "z17_8" 8 + (and (eq_attr "cpu" "z17") + (eq_attr "mnemonic" "cdgtr,cdlgtr,cdsg,cxgtr,cxlgtr,flogr,lpq,m,mfy,mg,mgrk,\ +ml,mlg,mlgr,mlr,mr,stpq,vsteb,vstebrh,vsteh")) "nothing") + +(define_insn_reservation "z17_9" 9 + (and (eq_attr "cpu" "z17") + (eq_attr "mnemonic" "cdfbr,cdgbr,cdlfbr,cdlgbr,cefbr,cegbr,celfbr,celgbr,madb,\ +maeb,maebr,msdb,mseb,msebr,stam")) "nothing") + +(define_insn_reservation "z17_10" 10 + (and (eq_attr "cpu" "z17") + (eq_attr "mnemonic" "cgdtr,cgxtr,clfdtr,clfxtr,clgdtr,clgxtr,d,dl,dlg,dsg,\ +dsgf,efpc,lxdb,lxdbr,lxeb,lxebr,vscef,vsceg")) "nothing") + +(define_insn_reservation "z17_11" 11 + (and (eq_attr "cpu" "z17") + (eq_attr "mnemonic" "cfdbr,cfebr,cgdbr,cgebr,clfdbr,clfebr,clgdbr,clgebr")) "nothing") + +(define_insn_reservation "z17_12" 12 + (and (eq_attr "cpu" "z17") + (eq_attr "mnemonic" "cxbr,cxtr,kxbr,kxtr,tbegin,tbeginc,tcxb,tdcxt")) "nothing") + +(define_insn_reservation "z17_13" 13 + (and (eq_attr "cpu" "z17") + (eq_attr "mnemonic" "axbr,axtr,cxfbr,cxgbr,cxlfbr,cxlgbr,fixbr,fixbra,fixtr,\ +lcxbr,lnxbr,lpxbr,ltxbr,ltxtr,lxdtr,sxbr,sxtr")) "nothing") + +(define_insn_reservation "z17_14" 14 + (and (eq_attr "cpu" "z17") + (eq_attr "mnemonic" "cfxbr,cgxbr,clfxbr,clgxbr,ledtr")) "nothing") + +(define_insn_reservation "z17_15" 15 + (and (eq_attr "cpu" "z17") + (eq_attr "mnemonic" "nc,oc")) "nothing") + +(define_insn_reservation "z17_16" 16 + (and (eq_attr "cpu" "z17") + (eq_attr "mnemonic" "cdftr,cdlftr,cxftr,cxlftr")) "nothing") + +(define_insn_reservation "z17_18" 18 + (and (eq_attr "cpu" "z17") + (eq_attr "mnemonic" "xc")) "nothing") + +(define_insn_reservation "z17_20" 20 + (and (eq_attr "cpu" "z17") + (eq_attr "mnemonic" "ddb,ddbr,ddtr,deb,debr,dlgr,dlr,dr,dsgfr,dsgr,dxbr,dxtr,\ +mdtr,mvc,mxbr,mxtr,sqdb,sqdbr,sqeb,sqebr,sqxbr,vdf,vdg,vdlf,vdlg,vdlq,vdq,\ +vfddb,vfdsb,vfmadb,vfmasb,vfmdb,vfmsb,vfmsdb,vfmssb,vfnmadb,vfnmasb,vfnmsdb,\ +vfnmssb,vfsqdb,vfsqsb,vrf,vrg,vrlf,vrlg,vrlq,vrq,wfddb,wfdsb,wfdxb,wfmadb,\ +wfmasb,wfmaxb,wfmdb,wfmsb,wfmsdb,wfmssb,wfmsxb,wfmxb,wfnmaxb,wfnmsxb,wfsqdb,\ +wfsqxb")) "nothing") + diff --git a/gcc/config/s390/driver-native.cc b/gcc/config/s390/driver-native.cc index 49e8fa0..7a7ceea 100644 --- a/gcc/config/s390/driver-native.cc +++ b/gcc/config/s390/driver-native.cc @@ -127,6 +127,10 @@ s390_host_detect_local_cpu (int argc, const char **argv) case 0x3932: cpu = "arch14"; break; + case 0x9175: + case 0x9176: + cpu = "arch15"; + break; default: cpu = "arch15"; break; diff --git a/gcc/config/s390/s390-builtins.def b/gcc/config/s390/s390-builtins.def index d9af9b1..cee2326 100644 --- a/gcc/config/s390/s390-builtins.def +++ b/gcc/config/s390/s390-builtins.def @@ -300,8 +300,8 @@ #define B_VXE2 (1 << 4) /* Builtins requiring the z15 vector extensions. */ #define B_DEP (1 << 5) /* Builtin has been deprecated and a warning should be issued. */ #define B_NNPA (1 << 6) /* Builtins requiring the NNPA Facility. */ -#define B_VXE3 (1 << 7) /* Builtins requiring the arch15 vector extensions. */ -#define B_ARCH15 (1 << 8) /* Builtins requiring arch15. */ +#define B_VXE3 (1 << 7) /* Builtins requiring the z17 vector extensions. */ +#define B_Z17 (1 << 8) /* Builtins requiring z17. */ /* B_DEF defines a standard (not overloaded) builtin B_DEF (<builtin name>, <RTL expander name>, <function attributes>, <builtin flags>, <operand flags, see above>, <fntype>) @@ -3318,8 +3318,8 @@ B_DEF (s390_vcnf, vcnf_v8hi, 0, /* arch 15 builtins */ -B_DEF (s390_bdepg, bdepg, 0, B_ARCH15, 0, BT_FN_ULONG_ULONG_ULONG) -B_DEF (s390_bextg, bextg, 0, B_ARCH15, 0, BT_FN_ULONG_ULONG_ULONG) +B_DEF (s390_bdepg, bdepg, 0, B_Z17, 0, BT_FN_ULONG_ULONG_ULONG) +B_DEF (s390_bextg, bextg, 0, B_Z17, 0, BT_FN_ULONG_ULONG_ULONG) OB_DEF (s390_vec_blend, s390_vec_blend_s8, s390_vec_blend_dbl, B_VXE3, BT_FN_OV4SI_OV4SI_OV4SI_OV4SI) OB_DEF_VAR (s390_vec_blend_s8, s390_vblendb, 0, 0, BT_OV_V16QI_V16QI_V16QI_V16QI) diff --git a/gcc/config/s390/s390-c.cc b/gcc/config/s390/s390-c.cc index 311d74a..a01c44c 100644 --- a/gcc/config/s390/s390-c.cc +++ b/gcc/config/s390/s390-c.cc @@ -962,7 +962,7 @@ s390_resolve_overloaded_builtin (location_t loc, tree ob_fndecl, if (!TARGET_VXE3 && (ob_flags & B_VXE3)) { - error_at (loc, "%qF requires arch15 or higher", ob_fndecl); + error_at (loc, "%qF requires z17 or higher", ob_fndecl); return error_mark_node; } @@ -1056,7 +1056,7 @@ s390_resolve_overloaded_builtin (location_t loc, tree ob_fndecl, if (!TARGET_VXE3 && bflags_overloaded_builtin_var[last_match_index] & B_VXE3) { - error_at (loc, "%qs matching variant requires arch15 or higher", + error_at (loc, "%qs matching variant requires z17 or higher", IDENTIFIER_POINTER (DECL_NAME (ob_fndecl))); return error_mark_node; } diff --git a/gcc/config/s390/s390-opts.h b/gcc/config/s390/s390-opts.h index 437d3b9..9cacb2c 100644 --- a/gcc/config/s390/s390-opts.h +++ b/gcc/config/s390/s390-opts.h @@ -39,7 +39,7 @@ enum processor_type PROCESSOR_3906_Z14, PROCESSOR_8561_Z15, PROCESSOR_3931_Z16, - PROCESSOR_ARCH15, + PROCESSOR_9175_Z17, PROCESSOR_NATIVE, PROCESSOR_max }; diff --git a/gcc/config/s390/s390.cc b/gcc/config/s390/s390.cc index 0ff3fd5..d82b16e 100644 --- a/gcc/config/s390/s390.cc +++ b/gcc/config/s390/s390.cc @@ -342,7 +342,7 @@ const struct s390_processor processor_table[] = { "z14", "arch12", PROCESSOR_3906_Z14, &zEC12_cost, 12 }, { "z15", "arch13", PROCESSOR_8561_Z15, &zEC12_cost, 13 }, { "z16", "arch14", PROCESSOR_3931_Z16, &zEC12_cost, 14 }, - { "arch15", "arch15", PROCESSOR_ARCH15, &zEC12_cost, 15 }, + { "z17", "arch15", PROCESSOR_9175_Z17, &zEC12_cost, 15 }, { "native", "", PROCESSOR_NATIVE, NULL, 0 } }; @@ -916,7 +916,7 @@ s390_expand_builtin (tree exp, rtx target, rtx subtarget ATTRIBUTE_UNUSED, if ((bflags & B_VXE3) && !TARGET_VXE3) { - error ("Builtin %qF requires arch15 or higher", fndecl); + error ("Builtin %qF requires z17 or higher", fndecl); return const0_rtx; } } @@ -9204,7 +9204,7 @@ s390_issue_rate (void) case PROCESSOR_3906_Z14: case PROCESSOR_8561_Z15: case PROCESSOR_3931_Z16: - case PROCESSOR_ARCH15: + case PROCESSOR_9175_Z17: default: return 1; } @@ -15632,7 +15632,6 @@ s390_get_sched_attrmask (rtx_insn *insn) mask |= S390_SCHED_ATTR_MASK_GROUPOFTWO; break; case PROCESSOR_3931_Z16: - case PROCESSOR_ARCH15: if (get_attr_z16_cracked (insn)) mask |= S390_SCHED_ATTR_MASK_CRACKED; if (get_attr_z16_expanded (insn)) @@ -15644,6 +15643,18 @@ s390_get_sched_attrmask (rtx_insn *insn) if (get_attr_z16_groupoftwo (insn)) mask |= S390_SCHED_ATTR_MASK_GROUPOFTWO; break; + case PROCESSOR_9175_Z17: + if (get_attr_z17_cracked (insn)) + mask |= S390_SCHED_ATTR_MASK_CRACKED; + if (get_attr_z17_expanded (insn)) + mask |= S390_SCHED_ATTR_MASK_EXPANDED; + if (get_attr_z17_endgroup (insn)) + mask |= S390_SCHED_ATTR_MASK_ENDGROUP; + if (get_attr_z17_groupalone (insn)) + mask |= S390_SCHED_ATTR_MASK_GROUPALONE; + if (get_attr_z17_groupoftwo (insn)) + mask |= S390_SCHED_ATTR_MASK_GROUPOFTWO; + break; default: gcc_unreachable (); } @@ -15691,7 +15702,6 @@ s390_get_unit_mask (rtx_insn *insn, int *units) mask |= 1 << 3; break; case PROCESSOR_3931_Z16: - case PROCESSOR_ARCH15: *units = 4; if (get_attr_z16_unit_lsu (insn)) mask |= 1 << 0; @@ -15702,6 +15712,17 @@ s390_get_unit_mask (rtx_insn *insn, int *units) if (get_attr_z16_unit_vfu (insn)) mask |= 1 << 3; break; + case PROCESSOR_9175_Z17: + *units = 4; + if (get_attr_z17_unit_lsu (insn)) + mask |= 1 << 0; + if (get_attr_z17_unit_fxa (insn)) + mask |= 1 << 1; + if (get_attr_z17_unit_fxb (insn)) + mask |= 1 << 2; + if (get_attr_z17_unit_vfu (insn)) + mask |= 1 << 3; + break; default: gcc_unreachable (); } @@ -15715,7 +15736,8 @@ s390_is_fpd (rtx_insn *insn) return false; return get_attr_z13_unit_fpd (insn) || get_attr_z14_unit_fpd (insn) - || get_attr_z15_unit_fpd (insn) || get_attr_z16_unit_fpd (insn); + || get_attr_z15_unit_fpd (insn) || get_attr_z16_unit_fpd (insn) + || get_attr_z17_unit_fpd (insn); } static bool @@ -15725,7 +15747,8 @@ s390_is_fxd (rtx_insn *insn) return false; return get_attr_z13_unit_fxd (insn) || get_attr_z14_unit_fxd (insn) - || get_attr_z15_unit_fxd (insn) || get_attr_z16_unit_fxd (insn); + || get_attr_z15_unit_fxd (insn) || get_attr_z16_unit_fxd (insn) + || get_attr_z17_unit_fxd (insn); } /* Returns TRUE if INSN is a long-running instruction. */ diff --git a/gcc/config/s390/s390.h b/gcc/config/s390/s390.h index 6f7195d..8b04bc9 100644 --- a/gcc/config/s390/s390.h +++ b/gcc/config/s390/s390.h @@ -45,12 +45,12 @@ enum processor_flags PF_NNPA = 32768, PF_Z16 = 65536, PF_VXE3 = 131072, - PF_ARCH15 = 262144 + PF_Z17 = 262144 }; /* This is necessary to avoid a warning about comparing different enum types. */ -#define s390_tune_attr ((enum attr_cpu)(s390_tune > PROCESSOR_3931_Z16 ? PROCESSOR_3931_Z16 : s390_tune )) +#define s390_tune_attr ((enum attr_cpu)(s390_tune > PROCESSOR_9175_Z17 ? PROCESSOR_9175_Z17 : s390_tune )) /* These flags indicate that the generated code should run on a cpu providing the respective hardware facility regardless of the @@ -124,10 +124,10 @@ enum processor_flags (s390_arch_flags & PF_VXE3) #define TARGET_CPU_VXE3_P(opts) \ (opts->x_s390_arch_flags & PF_VXE3) -#define TARGET_CPU_ARCH15 \ - (s390_arch_flags & PF_ARCH15) -#define TARGET_CPU_ARCH15_P(opts) \ - (opts->x_s390_arch_flags & PF_ARCH15) +#define TARGET_CPU_Z17 \ + (s390_arch_flags & PF_Z17) +#define TARGET_CPU_Z17_P(opts) \ + (opts->x_s390_arch_flags & PF_Z17) #define TARGET_HARD_FLOAT_P(opts) (!TARGET_SOFT_FLOAT_P(opts)) @@ -198,9 +198,9 @@ enum processor_flags (TARGET_VX && TARGET_CPU_VXE3) #define TARGET_VXE3_P(opts) \ (TARGET_VX_P (opts) && TARGET_CPU_VXE3_P (opts)) -#define TARGET_ARCH15 (TARGET_ZARCH && TARGET_CPU_ARCH15) -#define TARGET_ARCH15_P(opts) \ - (TARGET_ZARCH_P (opts->x_target_flags) && TARGET_CPU_ARCH15_P (opts)) +#define TARGET_Z17 (TARGET_ZARCH && TARGET_CPU_Z17) +#define TARGET_Z17_P(opts) \ + (TARGET_ZARCH_P (opts->x_target_flags) && TARGET_CPU_Z17_P (opts)) #if defined(HAVE_AS_VECTOR_LOADSTORE_ALIGNMENT_HINTS_ON_Z13) #define TARGET_VECTOR_LOADSTORE_ALIGNMENT_HINTS TARGET_Z13 diff --git a/gcc/config/s390/s390.md b/gcc/config/s390/s390.md index 9d49580..05b9da6 100644 --- a/gcc/config/s390/s390.md +++ b/gcc/config/s390/s390.md @@ -599,11 +599,11 @@ ;; Processor type. This attribute must exactly match the processor_type ;; enumeration in s390.h. -(define_attr "cpu" "z900,z990,z9_109,z9_ec,z10,z196,zEC12,z13,z14,z15,z16" +(define_attr "cpu" "z900,z990,z9_109,z9_ec,z10,z196,zEC12,z13,z14,z15,z16,z17" (const (symbol_ref "s390_tune_attr"))) (define_attr "cpu_facility" - "standard,ieee,zarch,cpu_zarch,longdisp,extimm,dfp,z10,z196,zEC12,vx,z13,z14,vxe,z15,vxe2,z16,nnpa,vxe3,arch15" + "standard,ieee,zarch,cpu_zarch,longdisp,extimm,dfp,z10,z196,zEC12,vx,z13,z14,vxe,z15,vxe2,z16,nnpa,vxe3,z17" (const_string "standard")) (define_attr "enabled" "" @@ -681,8 +681,8 @@ (match_test "TARGET_VXE3")) (const_int 1) - (and (eq_attr "cpu_facility" "arch15") - (match_test "TARGET_ARCH15")) + (and (eq_attr "cpu_facility" "z17") + (match_test "TARGET_Z17")) (const_int 1) ] (const_int 0))) @@ -725,6 +725,9 @@ ;; Pipeline description for z16 (include "3931.md") +;; Pipeline description for z17 +(include "9175.md") + ;; Predicates (include "predicates.md") @@ -2056,7 +2059,7 @@ [(set (match_operand:DI 0 "register_operand" "=d") (ashift:DI (sign_extend:DI (match_operand:SI 1 "register_operand" "a")) (const_int LXAMODEITER)))] - "TARGET_ARCH15 && TARGET_64BIT" + "TARGET_Z17 && TARGET_64BIT" "lxa<lxamode>\t%0,0(%1,0)" [(set_attr "op_type" "RXY")]) @@ -2066,7 +2069,7 @@ (ashift:DI (sign_extend:DI (plus:SI (match_operand:SI 1 "register_operand" "a") (match_operand:SI 2 "const_int_operand"))) (const_int LXAMODEITER)))] - "TARGET_ARCH15 && TARGET_64BIT && INTVAL (operands[2]) >= -0x80000 && INTVAL (operands[2]) <= 0x7FFFF" + "TARGET_Z17 && TARGET_64BIT && INTVAL (operands[2]) >= -0x80000 && INTVAL (operands[2]) <= 0x7FFFF" "lxa<lxamode>\t%0,%2(%1,0)" [(set_attr "op_type" "RXY")]) @@ -2076,7 +2079,7 @@ (plus:DI (ashift:DI (sign_extend:DI (match_operand:SI 1 "register_operand" "a")) (const_int LXAMODEITER)) (match_operand:DI 2 "register_operand" "a")))] - "TARGET_ARCH15 && TARGET_64BIT" + "TARGET_Z17 && TARGET_64BIT" "lxa<lxamode>\t%0,0(%1,%2)" [(set_attr "op_type" "RXY")]) @@ -2087,7 +2090,7 @@ (match_operand:SI 2 "const_int_operand"))) (const_int LXAMODEITER)) (match_operand:DI 3 "register_operand" "a")))] - "TARGET_ARCH15 && TARGET_64BIT && INTVAL (operands[2]) >= -0x80000 && INTVAL (operands[2]) <= 0x7FFFF" + "TARGET_Z17 && TARGET_64BIT && INTVAL (operands[2]) >= -0x80000 && INTVAL (operands[2]) <= 0x7FFFF" "lxa<lxamode>\t%0,%2(%1,%3)" [(set_attr "op_type" "RXY")]) @@ -2096,7 +2099,7 @@ (plus:DI (sign_extend:DI (plus:SI (match_operand:SI 1 "register_operand" "a") (match_operand:SI 2 "const_int_operand"))) (match_operand:DI 3 "register_operand" "a")))] - "TARGET_ARCH15 && TARGET_64BIT && INTVAL (operands[2]) >= -0x80000 && INTVAL (operands[2]) <= 0x7FFFF" + "TARGET_Z17 && TARGET_64BIT && INTVAL (operands[2]) >= -0x80000 && INTVAL (operands[2]) <= 0x7FFFF" "lxab\t%0,%2(%1,%3)" [(set_attr "op_type" "RXY")]) @@ -2113,7 +2116,7 @@ 0) (const_int LXAMODEITER)) (const_int <LLXAMASK>)))] - "TARGET_ARCH15 && TARGET_64BIT && INTVAL (operands[2]) >= -0x80000 && INTVAL (operands[2]) <= 0x7FFFF" + "TARGET_Z17 && TARGET_64BIT && INTVAL (operands[2]) >= -0x80000 && INTVAL (operands[2]) <= 0x7FFFF" "llxa<lxamode>\t%0,%2(%1,0)" [(set_attr "op_type" "RXY")]) @@ -2124,7 +2127,7 @@ (const_int LXAMODEITER)) (const_int <LLXAMASK>)) (match_operand:DI 2 "register_operand" "a")))] - "TARGET_ARCH15 && TARGET_64BIT" + "TARGET_Z17 && TARGET_64BIT" "llxa<lxamode>\t%0,0(%1,%2)" [(set_attr "op_type" "RXY")]) @@ -2137,7 +2140,7 @@ (const_int LXAMODEITER)) (const_int <LLXAMASK>)) (match_operand:DI 3 "register_operand" "a")))] - "TARGET_ARCH15 && TARGET_64BIT && INTVAL (operands[2]) >= -0x80000 && INTVAL (operands[2]) <= 0x7FFFF" + "TARGET_Z17 && TARGET_64BIT && INTVAL (operands[2]) >= -0x80000 && INTVAL (operands[2]) <= 0x7FFFF" "llxa<lxamode>\t%0,%2(%1,%3)" [(set_attr "op_type" "RXY")]) @@ -2146,7 +2149,7 @@ (plus:DI (zero_extend:DI (plus:SI (match_operand:SI 1 "register_operand" "a") (match_operand:SI 2 "const_int_operand"))) (match_operand:DI 3 "register_operand" "a")))] - "TARGET_ARCH15 && TARGET_64BIT && INTVAL (operands[2]) >= -0x80000 && INTVAL (operands[2]) <= 0x7FFFF" + "TARGET_Z17 && TARGET_64BIT && INTVAL (operands[2]) >= -0x80000 && INTVAL (operands[2]) <= 0x7FFFF" "llxab\t%0,%2(%1,%3)" [(set_attr "op_type" "RXY")]) @@ -3594,7 +3597,7 @@ (match_operand:BLK 1 "memory_operand" "")) (use (match_operand 2 "const_int_operand" "")) (use (match_operand 3 "immediate_operand" "")) - (clobber (scratch))] + (clobber (match_scratch 4))] "reload_completed" [(parallel [(set (match_dup 0) (match_dup 1)) @@ -3606,7 +3609,7 @@ (match_operand:BLK 1 "memory_operand" "")) (use (match_operand 2 "register_operand" "")) (use (match_operand 3 "memory_operand" "")) - (clobber (scratch))] + (clobber (match_scratch 4))] "reload_completed" [(parallel [(unspec [(match_dup 2) (match_dup 3) @@ -3620,14 +3623,14 @@ (match_operand:BLK 1 "memory_operand" "")) (use (match_operand 2 "register_operand" "")) (use (const:BLK (unspec:BLK [(const_int 0)] UNSPEC_INSN))) - (clobber (scratch))] + (clobber (match_scratch 3))] "TARGET_Z10 && reload_completed" [(parallel [(unspec [(match_dup 2) (const_int 0) - (label_ref (match_dup 3))] UNSPEC_EXECUTE) + (label_ref (match_dup 4))] UNSPEC_EXECUTE) (set (match_dup 0) (match_dup 1)) (use (const_int 1))])] - "operands[3] = gen_label_rtx ();") + "operands[4] = gen_label_rtx ();") (define_split [(set (match_operand:BLK 0 "memory_operand" "") @@ -3849,7 +3852,7 @@ (const_int 0)) (use (match_operand 1 "const_int_operand" "")) (use (match_operand 2 "immediate_operand" "")) - (clobber (scratch)) + (clobber (match_scratch 3)) (clobber (reg:CC CC_REGNUM))] "reload_completed" [(parallel @@ -3863,7 +3866,7 @@ (const_int 0)) (use (match_operand 1 "register_operand" "")) (use (match_operand 2 "memory_operand" "")) - (clobber (scratch)) + (clobber (match_scratch 3)) (clobber (reg:CC CC_REGNUM))] "reload_completed" [(parallel @@ -3879,7 +3882,7 @@ (const_int 0)) (use (match_operand 1 "register_operand" "")) (use (const:BLK (unspec:BLK [(const_int 0)] UNSPEC_INSN))) - (clobber (scratch)) + (clobber (match_scratch 2)) (clobber (reg:CC CC_REGNUM))] "TARGET_Z10 && reload_completed" [(parallel @@ -4044,7 +4047,7 @@ (match_operand:BLK 1 "memory_operand" ""))) (use (match_operand 2 "const_int_operand" "")) (use (match_operand 3 "immediate_operand" "")) - (clobber (scratch))] + (clobber (match_scratch 4))] "reload_completed" [(parallel [(set (reg:CCU CC_REGNUM) (compare:CCU (match_dup 0) (match_dup 1))) @@ -4057,7 +4060,7 @@ (match_operand:BLK 1 "memory_operand" ""))) (use (match_operand 2 "register_operand" "")) (use (match_operand 3 "memory_operand" "")) - (clobber (scratch))] + (clobber (match_scratch 4))] "reload_completed" [(parallel [(unspec [(match_dup 2) (match_dup 3) @@ -4072,7 +4075,7 @@ (match_operand:BLK 1 "memory_operand" ""))) (use (match_operand 2 "register_operand" "")) (use (const:BLK (unspec:BLK [(const_int 0)] UNSPEC_INSN))) - (clobber (scratch))] + (clobber (match_scratch 3))] "TARGET_Z10 && reload_completed" [(parallel [(unspec [(match_dup 2) (const_int 0) @@ -4940,7 +4943,7 @@ (unspec:DI [(match_operand:DI 1 "register_operand" "d") (match_operand:DI 2 "register_operand" "d")] UNSPEC_BDEPG))] - "TARGET_ARCH15 && TARGET_64BIT" + "TARGET_Z17 && TARGET_64BIT" "bdepg\t%0,%1,%2" [(set_attr "op_type" "RRF")]) @@ -4953,7 +4956,7 @@ (unspec:DI [(match_operand:DI 1 "register_operand" "d") (match_operand:DI 2 "register_operand" "d")] UNSPEC_BEXTG))] - "TARGET_ARCH15 && TARGET_64BIT" + "TARGET_Z17 && TARGET_64BIT" "bextg\t%0,%1,%2" [(set_attr "op_type" "RRF")]) @@ -9580,7 +9583,7 @@ (clz:DI (match_operand:DI 1 "register_operand" "d")))] "TARGET_EXTIMM && TARGET_ZARCH" { - if (!(TARGET_ARCH15 && TARGET_64BIT)) + if (!(TARGET_Z17 && TARGET_64BIT)) { rtx_insn *insn; rtx clz_equal; @@ -9601,7 +9604,7 @@ (define_insn "*clzg" [(set (match_operand:DI 0 "register_operand" "=d") (clz:DI (match_operand:DI 1 "register_operand" "d")))] - "TARGET_ARCH15 && TARGET_64BIT" + "TARGET_Z17 && TARGET_64BIT" "clzg\t%0,%1" [(set_attr "op_type" "RRE")]) @@ -9630,7 +9633,7 @@ (define_insn "ctzdi2" [(set (match_operand:DI 0 "register_operand" "=d") (ctz:DI (match_operand:DI 1 "register_operand" "d")))] - "TARGET_ARCH15 && TARGET_64BIT" + "TARGET_Z17 && TARGET_64BIT" "ctzg\t%0,%1" [(set_attr "op_type" "RRE")]) diff --git a/gcc/config/s390/s390.opt b/gcc/config/s390/s390.opt index f064597..6753a93 100644 --- a/gcc/config/s390/s390.opt +++ b/gcc/config/s390/s390.opt @@ -122,7 +122,10 @@ EnumValue Enum(processor_type) String(z16) Value(PROCESSOR_3931_Z16) EnumValue -Enum(processor_type) String(arch15) Value(PROCESSOR_ARCH15) +Enum(processor_type) String(arch15) Value(PROCESSOR_9175_Z17) + +EnumValue +Enum(processor_type) String(z17) Value(PROCESSOR_9175_Z17) EnumValue Enum(processor_type) String(native) Value(PROCESSOR_NATIVE) DriverOnly diff --git a/gcc/config/sh/sh-modes.def b/gcc/config/sh/sh-modes.def index 80650b4..e31ae69 100644 --- a/gcc/config/sh/sh-modes.def +++ b/gcc/config/sh/sh-modes.def @@ -17,6 +17,12 @@ You should have received a copy of the GNU General Public License along with GCC; see the file COPYING3. If not see <http://www.gnu.org/licenses/>. */ +/* SH has the same reversed quiet bit as MIPS. */ +RESET_FLOAT_FORMAT (SF, mips_single_format); +RESET_FLOAT_FORMAT (DF, mips_double_format); +/* TFmode: IEEE quad floating point (software). */ +FLOAT_MODE (TF, 16, mips_quad_format); + /* Vector modes. */ VECTOR_MODE (INT, QI, 2); /* V2QI */ VECTOR_MODES (INT, 4); /* V4QI V2HI */ diff --git a/gcc/configure b/gcc/configure index ab6bec1..1696595 100755 --- a/gcc/configure +++ b/gcc/configure @@ -3948,7 +3948,7 @@ if test x"${DEFAULT_LINKER+set}" = x"set"; then as_fn_error $? "cannot execute: $DEFAULT_LINKER: check --with-ld or env. var. DEFAULT_LINKER" "$LINENO" 5 elif $DEFAULT_LINKER -v < /dev/null 2>&1 | grep GNU > /dev/null; then gnu_ld_flag=yes - elif $DEFAULT_LINKER -v < /dev/null 2>&1 | grep ld64- > /dev/null; then + elif $DEFAULT_LINKER -v < /dev/null 2>&1 | grep 'PROJECT:ld\(64\)*-' > /dev/null; then ld64_flag=yes fi @@ -12694,6 +12694,42 @@ $as_echo "#define HOST_HAS_O_NONBLOCK 1" >>confdefs.h fi +# Check if personality and ADDR_NO_RANDOMIZE are declared +# in sys/personality.h +{ $as_echo "$as_me:${as_lineno-$LINENO}: checking for personality ADDR_NO_RANDOMIZE" >&5 +$as_echo_n "checking for personality ADDR_NO_RANDOMIZE... " >&6; } +if ${ac_cv_personality_addr_no_randomize+:} false; then : + $as_echo_n "(cached) " >&6 +else + +cat confdefs.h - <<_ACEOF >conftest.$ac_ext +/* end confdefs.h. */ + +#include <sys/personality.h> +int +main () +{ + +personality (personality (0xffffffffU) | ADDR_NO_RANDOMIZE); + ; + return 0; +} +_ACEOF +if ac_fn_cxx_try_compile "$LINENO"; then : + ac_cv_personality_addr_no_randomize=yes +else + ac_cv_personality_addr_no_randomize=no +fi +rm -f core conftest.err conftest.$ac_objext conftest.$ac_ext +fi +{ $as_echo "$as_me:${as_lineno-$LINENO}: result: $ac_cv_personality_addr_no_randomize" >&5 +$as_echo "$ac_cv_personality_addr_no_randomize" >&6; } +if test $ac_cv_personality_addr_no_randomize = yes; then + +$as_echo "#define HOST_HAS_PERSONALITY_ADDR_NO_RANDOMIZE 1" >>confdefs.h + +fi + # C++ Modules would like some networking features to provide the mapping # server. You can still use modules without them though. @@ -21484,7 +21520,7 @@ else lt_dlunknown=0; lt_dlno_uscore=1; lt_dlneed_uscore=2 lt_status=$lt_dlunknown cat > conftest.$ac_ext <<_LT_EOF -#line 21487 "configure" +#line 21523 "configure" #include "confdefs.h" #if HAVE_DLFCN_H @@ -21590,7 +21626,7 @@ else lt_dlunknown=0; lt_dlno_uscore=1; lt_dlneed_uscore=2 lt_status=$lt_dlunknown cat > conftest.$ac_ext <<_LT_EOF -#line 21593 "configure" +#line 21629 "configure" #include "confdefs.h" #if HAVE_DLFCN_H @@ -32694,8 +32730,9 @@ $as_echo "$gcc_cv_ld64_major" >&6; } { $as_echo "$as_me:${as_lineno-$LINENO}: checking linker version" >&5 $as_echo_n "checking linker version... " >&6; } if test x"${gcc_cv_ld64_version}" = x; then - gcc_cv_ld64_version=`$gcc_cv_ld -v 2>&1 | $EGREP 'ld64|dyld' \ - | sed -e 's/.*ld64-//' -e 's/.*dyld-//'| awk '{print $1}'` + gcc_cv_ld64_version=`$gcc_cv_ld -v 2>&1 | $EGREP 'ld64|dyld|PROJECT:ld' \ + | sed -e 's/.*ld64-//' -e 's/.*dyld-//' -e 's/.*PROJECT:ld-//' \ + | awk '{print $1}'` fi { $as_echo "$as_me:${as_lineno-$LINENO}: result: $gcc_cv_ld64_version" >&5 $as_echo "$gcc_cv_ld64_version" >&6; } diff --git a/gcc/configure.ac b/gcc/configure.ac index fca0579..9f67e62 100644 --- a/gcc/configure.ac +++ b/gcc/configure.ac @@ -358,7 +358,7 @@ if test x"${DEFAULT_LINKER+set}" = x"set"; then AC_MSG_ERROR([cannot execute: $DEFAULT_LINKER: check --with-ld or env. var. DEFAULT_LINKER]) elif $DEFAULT_LINKER -v < /dev/null 2>&1 | grep GNU > /dev/null; then gnu_ld_flag=yes - elif $DEFAULT_LINKER -v < /dev/null 2>&1 | grep ld64- > /dev/null; then + elif $DEFAULT_LINKER -v < /dev/null 2>&1 | grep 'PROJECT:ld\(64\)*-' > /dev/null; then ld64_flag=yes fi AC_DEFINE_UNQUOTED(DEFAULT_LINKER,"$DEFAULT_LINKER", @@ -1781,6 +1781,21 @@ if test $ac_cv_have_decl_O_NONBLOCK = yes; then [Define if O_NONBLOCK supported by fcntl.]) fi +# Check if personality and ADDR_NO_RANDOMIZE are declared +# in sys/personality.h +AC_CACHE_CHECK(for personality ADDR_NO_RANDOMIZE, + ac_cv_personality_addr_no_randomize, [ +AC_COMPILE_IFELSE([AC_LANG_PROGRAM([[ +#include <sys/personality.h>]], [[ +personality (personality (0xffffffffU) | ADDR_NO_RANDOMIZE);]])], +[ac_cv_personality_addr_no_randomize=yes], +[ac_cv_personality_addr_no_randomize=no])]) +if test $ac_cv_personality_addr_no_randomize = yes; then + AC_DEFINE(HOST_HAS_PERSONALITY_ADDR_NO_RANDOMIZE, 1, + [Define if personality and ADDR_NO_RANDOMIZE are declared in +sys/personality.h.]) +fi + # C++ Modules would like some networking features to provide the mapping # server. You can still use modules without them though. @@ -6403,8 +6418,9 @@ if test x"$ld64_flag" = x"yes"; then # If the version was not specified, try to find it. AC_MSG_CHECKING(linker version) if test x"${gcc_cv_ld64_version}" = x; then - gcc_cv_ld64_version=`$gcc_cv_ld -v 2>&1 | $EGREP 'ld64|dyld' \ - | sed -e 's/.*ld64-//' -e 's/.*dyld-//'| awk '{print $1}'` + gcc_cv_ld64_version=`$gcc_cv_ld -v 2>&1 | $EGREP 'ld64|dyld|PROJECT:ld' \ + | sed -e 's/.*ld64-//' -e 's/.*dyld-//' -e 's/.*PROJECT:ld-//' \ + | awk '{print $1}'` fi AC_MSG_RESULT($gcc_cv_ld64_version) diff --git a/gcc/cp/ChangeLog b/gcc/cp/ChangeLog index d266ef9..6975efb 100644 --- a/gcc/cp/ChangeLog +++ b/gcc/cp/ChangeLog @@ -1,3 +1,183 @@ +2025-04-19 Jason Merrill <jason@redhat.com> + + * coroutines.cc (coro_build_expr_stmt) + (coro_build_cvt_void_expr_stmt): Remove. + (build_actor_fn): Use finish_expr_stmt. + * semantics.cc (finish_expr_stmt): Avoid wrapping statement in + EXPR_STMT. + (finish_stmt_expr_expr): Add comment. + +2025-04-17 Jason Merrill <jason@redhat.com> + + * constexpr.cc (is_valid_constexpr_fn): Improve diagnostic. + +2025-04-17 Jason Merrill <jason@redhat.com> + + * constexpr.cc (cxx_eval_outermost_constant_expr): Give both + expression and allocation location in allocated storage diagnostics. + +2025-04-17 Jason Merrill <jason@redhat.com> + + * name-lookup.cc (name_lookup::preserve_state): Fix reserve call. + * rtti.cc (get_tinfo_desc): Use vec_safe_grow_cleared. + +2025-04-17 Jason Merrill <jason@redhat.com> + + * semantics.cc (finish_type_pack_element): Add more info + to diagnostics. + +2025-04-17 Jason Merrill <jason@redhat.com> + + * decl.cc (cp_make_fname_decl): Prevent silent failure. + +2025-04-17 Jason Merrill <jason@redhat.com> + + * lex.cc (unqualified_name_lookup_error): Handle 'requires' better. + +2025-04-17 Jason Merrill <jason@redhat.com> + + PR c++/113360 + * cp-tree.h (struct language_function): Add erroneous bit. + * constexpr.cc (explain_invalid_constexpr_fn): Return if set. + (cxx_eval_call_expression): Quiet if set. + * parser.cc (cp_parser_function_definition_after_declarator) + * pt.cc (instantiate_body): Set it. + +2025-04-16 Jason Merrill <jason@redhat.com> + + PR c++/114772 + PR c++/101180 + * pt.cc (apply_late_template_attributes): Also override + target_option_current_node. + +2025-04-16 Jason Merrill <jason@redhat.com> + + PR c++/116954 + * contracts.cc (remove_contract_attributes): Preserve flags + on the attribute list. + +2025-04-15 Nathaniel Shead <nathanieloshead@gmail.com> + + PR c++/119755 + * lambda.cc (prune_lambda_captures): Remove pruned capture from + function's BLOCK_VARS and BIND_EXPR_VARS. + +2025-04-15 Jason Merrill <jason@redhat.com> + + PR c++/111075 + * constexpr.cc (cxx_eval_call_expression): Allow trivial + call from a thunk. + +2025-04-15 Patrick Palka <ppalka@redhat.com> + + PR c++/119807 + PR c++/112288 + * pt.cc (tsubst_friend_function): Skip remapping an + existing specialization if it doesn't match the shape of + the new friend definition. + +2025-04-15 Jason Merrill <jason@redhat.com> + + PR c++/113835 + * constexpr.cc (cxx_eval_outermost_constant_expr): Bail out early + for std::vector(N). + +2025-04-14 Patrick Palka <ppalka@redhat.com> + + PR c++/99214 + * constraint.cc (satisfy_declaration_constraints): Pass the + original ARGS to push_tinst_level. + +2025-04-13 Patrick Palka <ppalka@redhat.com> + + PR c++/115639 + * constexpr.cc (struct constexpr_call): Add NSDMIs to each + field. Replace 'result' data member with 3-element 'results' + array and a 'result' accessor function. Remove + 'manifestly_const_eval' data member. + (constexpr_call_hasher::equal): Adjust after constexpr_call + layout change. + (cxx_eval_call_expression): Likewise. Define some local + variables closer to their first use. Use unknown_type_node + instead of NULL_TREE as the "in progress" result. After + successully evaluating a call with mce_unknown, also cache the + result in the corresponding mce_true and mce_false slots. + +2025-04-13 Nathaniel Shead <nathanieloshead@gmail.com> + + * module.cc (trees_in::is_matching_decl): Don't check for + mismatches when importing a DECL_MAYBE_DELETED function over one + that's already finished. + +2025-04-13 Nathaniel Shead <nathanieloshead@gmail.com> + + * module.cc (trees_in::is_matching_decl): Add custom errors for + different kinds of mismatches. + +2025-04-12 Patrick Palka <ppalka@redhat.com> + + PR c++/116416 + * constexpr.cc (maybe_constant_init_1): Generalize type of + of manifestly_const_eval parameter from bool to mce_value. + (maybe_constant_init): Define 3-parameter version taking a + manifestly_const_eval instead of bool parameter. + (cxx_constant_init): Adjust. + * cp-gimplify.cc (cp_fold_r) <case TARGET_EXPR>: Pass mce_false + to maybe_constant_init during prvalue folding if ff_mce_false is + set. + * cp-tree.h (maybe_constant_init): Declare new overload. + +2025-04-11 Jason Merrill <jason@redhat.com> + + PR c++/114970 + * cp-gimplify.cc (cp_build_init_expr_for_ctor): Suppress warnings on + return_this COMPOUND_EXPR. + +2025-04-10 Jason Merrill <jason@redhat.com> + + PR c++/119345 + * pt.cc (add_extra_args): Also register a specialization + of the captured variable. + +2025-04-10 Patrick Palka <ppalka@redhat.com> + + PR c++/119687 + * pt.cc (alias_ctad_tweaks): Use lkp_range / lkp_iterator + instead of ovl_iterator. + +2025-04-10 Jakub Jelinek <jakub@redhat.com> + + PR translation/119684 + * error.cc (cp_print_error_function): Use G_ instead of _ for + pp_printf arguments. + (function_category): Use G_ instead of _. + (print_instantiation_full_context): Use G_ instead of _ in pp_verbatim + arguments. + (print_location): Likewise. + (print_instantiation_partial_context): Likewise. + (maybe_print_constexpr_context): Likewise. + (print_constrained_decl_info): Use G_() around pp_verbatim argument. + (print_concept_check_info): Likewise. + (print_constraint_context_head): Likewise. + (print_requires_expression_info): Likewise. Merge separate pp_verbatim + "in requirements " and "with " into one with conditional messages. + +2025-04-10 Jason Merrill <jason@redhat.com> + + PR c++/119175 + * mangle.cc (decl_mangling_context): Look through lambda type. + +2025-04-09 Patrick Palka <ppalka@redhat.com> + + PR c++/119574 + * pt.cc (add_extra_args): Remove checking assert. + +2025-04-09 Jason Merrill <jason@redhat.com> + + PR c++/118698 + * constraint.cc (struct norm_info): Add tf_partial. + * pt.cc (any_template_parm_r): Handle LAMBDA_EXPR_EXTRA_ARGS. + 2025-04-08 Jason Merrill <jason@redhat.com> PR c++/117530 diff --git a/gcc/cp/constexpr.cc b/gcc/cp/constexpr.cc index 497f64f..8a11e62 100644 --- a/gcc/cp/constexpr.cc +++ b/gcc/cp/constexpr.cc @@ -307,7 +307,14 @@ is_valid_constexpr_fn (tree fun, bool complain) { ret = false; if (complain) - error ("%q#T has virtual base classes", DECL_CONTEXT (fun)); + { + if (DECL_CONSTRUCTOR_P (fun)) + error ("%<constexpr%> constructor in %q#T that has " + "virtual base classes", DECL_CONTEXT (fun)); + else + error ("%<constexpr%> destructor in %q#T that has " + "virtual base classes", DECL_CONTEXT (fun)); + } } return ret; @@ -1048,6 +1055,12 @@ explain_invalid_constexpr_fn (tree fun) { static hash_set<tree> *diagnosed; tree body; + + /* Don't try to explain a function we already complained about. */ + if (function *f = DECL_STRUCT_FUNCTION (fun)) + if (f->language->erroneous) + return; + /* In C++23, a function marked 'constexpr' may not actually be a constant expression. We haven't diagnosed the problem yet: -Winvalid-constexpr wasn't enabled. The function was called, so diagnose why it cannot be @@ -1119,20 +1132,22 @@ explain_invalid_constexpr_fn (tree fun) struct GTY((for_user)) constexpr_call { /* Description of the constexpr function definition. */ - constexpr_fundef *fundef; + constexpr_fundef *fundef = nullptr; /* Parameter bindings environment. A TREE_VEC of arguments. */ - tree bindings; - /* Result of the call. - NULL means the call is being evaluated. + tree bindings = NULL_TREE; + /* Result of the call, indexed by the value of + constexpr_ctx::manifestly_const_eval. + unknown_type_node means the call is being evaluated. error_mark_node means that the evaluation was erroneous or otherwise uncacheable (e.g. because it depends on the caller). Otherwise, the actual value of the call. */ - tree result; + tree results[3] = { NULL_TREE, NULL_TREE, NULL_TREE }; /* The hash of this call; we remember it here to avoid having to recalculate it when expanding the hash table. */ - hashval_t hash; - /* The value of constexpr_ctx::manifestly_const_eval. */ - enum mce_value manifestly_const_eval; + hashval_t hash = 0; + + /* The result slot corresponding to the given mce_value. */ + tree& result (mce_value mce) { return results[1 + int(mce)]; } }; struct constexpr_call_hasher : ggc_ptr_hash<constexpr_call> @@ -1427,8 +1442,6 @@ constexpr_call_hasher::equal (constexpr_call *lhs, constexpr_call *rhs) return true; if (lhs->hash != rhs->hash) return false; - if (lhs->manifestly_const_eval != rhs->manifestly_const_eval) - return false; if (!constexpr_fundef_hasher::equal (lhs->fundef, rhs->fundef)) return false; return cp_tree_equal (lhs->bindings, rhs->bindings); @@ -2855,9 +2868,6 @@ cxx_eval_call_expression (const constexpr_ctx *ctx, tree t, { location_t loc = cp_expr_loc_or_input_loc (t); tree fun = get_function_named_in_call (t); - constexpr_call new_call - = { NULL, NULL, NULL, 0, ctx->manifestly_const_eval }; - int depth_ok; if (fun == NULL_TREE) return cxx_eval_internal_function (ctx, t, lval, @@ -2946,12 +2956,11 @@ cxx_eval_call_expression (const constexpr_ctx *ctx, tree t, gcc_assert (arg0); if (new_op_p) { - /* FIXME: We should not get here; the VERIFY_CONSTANT above - should have already caught it. But currently a conversion - from pointer type to arithmetic type is only considered - non-constant for CONVERT_EXPRs, not NOP_EXPRs. */ if (!tree_fits_uhwi_p (arg0)) { + /* We should not get here; the VERIFY_CONSTANT above + should have already caught it. */ + gcc_checking_assert (false); if (!ctx->quiet) error_at (loc, "cannot allocate array: size not constant"); *non_constant_p = true; @@ -3082,6 +3091,7 @@ cxx_eval_call_expression (const constexpr_ctx *ctx, tree t, } constexpr_ctx new_ctx = *ctx; + ctx = &new_ctx; if (DECL_CONSTRUCTOR_P (fun) && !ctx->object && TREE_CODE (t) == AGGR_INIT_EXPR) { @@ -3091,27 +3101,26 @@ cxx_eval_call_expression (const constexpr_ctx *ctx, tree t, tree ctor = new_ctx.ctor = build_constructor (DECL_CONTEXT (fun), NULL); CONSTRUCTOR_NO_CLEARING (ctor) = true; ctx->global->put_value (new_ctx.object, ctor); - ctx = &new_ctx; } /* An immediate invocation is manifestly constant evaluated including the arguments of the call, so use mce_true even for the argument evaluation. */ if (DECL_IMMEDIATE_FUNCTION_P (fun)) - { - new_ctx.manifestly_const_eval = mce_true; - new_call.manifestly_const_eval = mce_true; - ctx = &new_ctx; - } + new_ctx.manifestly_const_eval = mce_true; /* We used to shortcut trivial constructor/op= here, but nowadays we can only get a trivial function here with -fno-elide-constructors. */ gcc_checking_assert (!trivial_fn_p (fun) || !flag_elide_constructors + /* Or it's a call from maybe_thunk_body (111075). */ + || (TREE_CODE (t) == CALL_EXPR ? CALL_FROM_THUNK_P (t) + : AGGR_INIT_FROM_THUNK_P (t)) /* We don't elide constructors when processing a noexcept-expression. */ || cp_noexcept_operand); bool non_constant_args = false; + constexpr_call new_call; new_call.bindings = cxx_bind_parameters_in_call (ctx, t, fun, non_constant_p, overflow_p, &non_constant_args); @@ -3185,7 +3194,12 @@ cxx_eval_call_expression (const constexpr_ctx *ctx, tree t, } } - depth_ok = push_cx_call_context (t); + /* Don't complain about problems evaluating an ill-formed function. */ + if (function *f = DECL_STRUCT_FUNCTION (fun)) + if (f->language->erroneous) + new_ctx.quiet = true; + + int depth_ok = push_cx_call_context (t); /* Remember the object we are constructing or destructing. */ tree new_obj = NULL_TREE; @@ -3227,8 +3241,6 @@ cxx_eval_call_expression (const constexpr_ctx *ctx, tree t, new_call.hash = constexpr_fundef_hasher::hash (new_call.fundef); new_call.hash = iterative_hash_template_arg (new_call.bindings, new_call.hash); - new_call.hash - = iterative_hash_object (ctx->manifestly_const_eval, new_call.hash); /* If we have seen this call before, we are done. */ maybe_initialize_constexpr_call_table (); @@ -3246,22 +3258,23 @@ cxx_eval_call_expression (const constexpr_ctx *ctx, tree t, the slot can move during evaluation of the body. */ *slot = entry = ggc_alloc<constexpr_call> (); *entry = new_call; + entry->result (ctx->manifestly_const_eval) = unknown_type_node; fb.preserve (); } } - /* Calls that are in progress have their result set to NULL, so that we - can detect circular dependencies. Now that we only cache up to - constexpr_cache_depth this won't catch circular dependencies that + /* Calls that are in progress have their result set to unknown_type_node, + so that we can detect circular dependencies. Now that we only cache + up to constexpr_cache_depth this won't catch circular dependencies that start deeper, but they'll hit the recursion or ops limit. */ - else if (entry->result == NULL) + else if (entry->result (ctx->manifestly_const_eval) == unknown_type_node) { if (!ctx->quiet) error ("call has circular dependency"); *non_constant_p = true; - entry->result = result = error_mark_node; + entry->result (ctx->manifestly_const_eval) = result = error_mark_node; } else - result = entry->result; + result = entry->result (ctx->manifestly_const_eval); } if (!depth_ok) @@ -3482,7 +3495,22 @@ cxx_eval_call_expression (const constexpr_ctx *ctx, tree t, else if (!result) result = void_node; if (entry) - entry->result = cacheable ? result : error_mark_node; + { + entry->result (ctx->manifestly_const_eval) + = cacheable ? result : error_mark_node; + + if (result != error_mark_node + && ctx->manifestly_const_eval == mce_unknown) + { + /* Evaluation succeeded and was independent of whether we're in a + manifestly constant-evaluated context, so we can also reuse + this result when evaluating this call with a fixed context. */ + if (!entry->result (mce_true)) + entry->result (mce_true) = entry->result (mce_unknown); + if (!entry->result (mce_false)) + entry->result (mce_false) = entry->result (mce_unknown); + } + } } /* The result of a constexpr function must be completely initialized. @@ -8450,7 +8478,7 @@ cxx_eval_constant_expression (const constexpr_ctx *ctx, tree t, if (TREE_CODE (t) == CONVERT_EXPR && ARITHMETIC_TYPE_P (type) && INDIRECT_TYPE_P (TREE_TYPE (op)) - && ctx->manifestly_const_eval == mce_true) + && ctx->strict) { if (!ctx->quiet) error_at (loc, @@ -9116,6 +9144,15 @@ cxx_eval_outermost_constant_expr (tree t, bool allow_non_constant, tree fndecl = cp_get_callee_fndecl_nofold (x); if (fndecl && DECL_IMMEDIATE_FUNCTION_P (fndecl)) is_consteval = true; + /* Don't try to evaluate a std::vector constructor taking an integer, it + will fail in the 'if (heap_var)' block below after doing all the work + (c++/113835). This will need adjustment if P3554 is accepted. Note + that evaluation of e.g. the vector default constructor can succeed, so + we don't shortcut all vector constructors. */ + if (fndecl && DECL_CONSTRUCTOR_P (fndecl) && allow_non_constant + && is_std_class (type, "vector") && call_expr_nargs (x) > 1 + && TREE_CODE (TREE_TYPE (get_nth_callarg (x, 1))) == INTEGER_TYPE) + return t; } if (AGGREGATE_TYPE_P (type) || VECTOR_TYPE_P (type)) { @@ -9190,11 +9227,6 @@ cxx_eval_outermost_constant_expr (tree t, bool allow_non_constant, if (r == void_node && !constexpr_dtor && ctx.ctor) r = ctx.ctor; - if (!constexpr_dtor) - verify_constant (r, allow_non_constant, &non_constant_p, &overflow_p); - else - DECL_INITIALIZED_BY_CONSTANT_EXPRESSION_P (object) = true; - unsigned int i; tree cleanup; /* Evaluate the cleanups. */ @@ -9213,15 +9245,6 @@ cxx_eval_outermost_constant_expr (tree t, bool allow_non_constant, non_constant_p = true; } - if (TREE_CODE (r) == CONSTRUCTOR && CONSTRUCTOR_NO_CLEARING (r)) - { - if (!allow_non_constant) - error ("%qE is not a constant expression because it refers to " - "an incompletely initialized variable", t); - TREE_CONSTANT (r) = false; - non_constant_p = true; - } - if (!non_constant_p && cxx_dialect >= cxx20 && !global_ctx.heap_vars.is_empty ()) { @@ -9231,9 +9254,11 @@ cxx_eval_outermost_constant_expr (tree t, bool allow_non_constant, if (heap_var) { if (!allow_non_constant && !non_constant_p) - error_at (DECL_SOURCE_LOCATION (heap_var), - "%qE is not a constant expression because it refers to " - "a result of %<operator new%>", t); + { + error ("%qE is not a constant expression because it refers to " + "a result of %<operator new%>", t); + inform (DECL_SOURCE_LOCATION (heap_var), "allocated here"); + } r = t; non_constant_p = true; } @@ -9242,9 +9267,11 @@ cxx_eval_outermost_constant_expr (tree t, bool allow_non_constant, if (DECL_NAME (heap_var) != heap_deleted_identifier) { if (!allow_non_constant && !non_constant_p) - error_at (DECL_SOURCE_LOCATION (heap_var), - "%qE is not a constant expression because allocated " - "storage has not been deallocated", t); + { + error ("%qE is not a constant expression because allocated " + "storage has not been deallocated", t); + inform (DECL_SOURCE_LOCATION (heap_var), "allocated here"); + } r = t; non_constant_p = true; } @@ -9274,6 +9301,21 @@ cxx_eval_outermost_constant_expr (tree t, bool allow_non_constant, non_constant_p = true; } + if (!non_constant_p && !constexpr_dtor) + verify_constant (r, allow_non_constant, &non_constant_p, &overflow_p); + + /* After verify_constant because reduced_constant_expression_p can unset + CONSTRUCTOR_NO_CLEARING. */ + if (!non_constant_p + && TREE_CODE (r) == CONSTRUCTOR && CONSTRUCTOR_NO_CLEARING (r)) + { + if (!allow_non_constant) + error ("%qE is not a constant expression because it refers to " + "an incompletely initialized variable", t); + TREE_CONSTANT (r) = false; + non_constant_p = true; + } + if (non_constant_p) /* If we saw something bad, go back to our argument. The wrapping below is only for the cases of TREE_CONSTANT argument or overflow. */ @@ -9290,13 +9332,17 @@ cxx_eval_outermost_constant_expr (tree t, bool allow_non_constant, if (non_constant_p && !allow_non_constant) return error_mark_node; - else if (constexpr_dtor) - return r; else if (non_constant_p && TREE_CONSTANT (r)) r = mark_non_constant (r); else if (non_constant_p) return t; + if (constexpr_dtor) + { + DECL_INITIALIZED_BY_CONSTANT_EXPRESSION_P (object) = true; + return r; + } + /* Check we are not trying to return the wrong type. */ if (!same_type_ignoring_top_level_qualifiers_p (type, TREE_TYPE (r))) { @@ -9448,6 +9494,9 @@ fold_simple (tree t) tree fold_to_constant (tree t) { + if (processing_template_decl) + return t; + tree r = fold (t); if (CONSTANT_CLASS_P (r) && !TREE_OVERFLOW (r)) return r; @@ -9679,7 +9728,7 @@ fold_non_dependent_init (tree t, static tree maybe_constant_init_1 (tree t, tree decl, bool allow_non_constant, - bool manifestly_const_eval) + mce_value manifestly_const_eval) { if (!t) return t; @@ -9705,17 +9754,27 @@ maybe_constant_init_1 (tree t, tree decl, bool allow_non_constant, { /* [basic.start.static] allows constant-initialization of variables with static or thread storage duration even if it isn't required, but we - shouldn't bend the rules the same way for automatic variables. */ + shouldn't bend the rules the same way for automatic variables. + + But still enforce the requirements of constexpr/constinit. + [dcl.constinit] "If a variable declared with the constinit specifier + has dynamic initialization, the program is ill-formed, even if the + implementation would perform that initialization as a static + initialization." */ bool is_static = (decl && DECL_P (decl) && (TREE_STATIC (decl) || DECL_EXTERNAL (decl))); + bool strict = (!is_static + || (decl && DECL_P (decl) + && (DECL_DECLARED_CONSTEXPR_P (decl) + || DECL_DECLARED_CONSTINIT_P (decl)))); if (is_static) - manifestly_const_eval = true; + manifestly_const_eval = mce_true; - if (cp_unevaluated_operand && !manifestly_const_eval) + if (cp_unevaluated_operand && manifestly_const_eval != mce_true) return fold_to_constant (t); - t = cxx_eval_outermost_constant_expr (t, allow_non_constant, !is_static, - mce_value (manifestly_const_eval), + t = cxx_eval_outermost_constant_expr (t, allow_non_constant, strict, + manifestly_const_eval, false, decl); } if (TREE_CODE (t) == TARGET_EXPR) @@ -9732,6 +9791,12 @@ maybe_constant_init_1 (tree t, tree decl, bool allow_non_constant, tree maybe_constant_init (tree t, tree decl, bool manifestly_const_eval) { + return maybe_constant_init_1 (t, decl, true, mce_value (manifestly_const_eval)); +} + +tree +maybe_constant_init (tree t, tree decl, mce_value manifestly_const_eval) +{ return maybe_constant_init_1 (t, decl, true, manifestly_const_eval); } @@ -9740,7 +9805,7 @@ maybe_constant_init (tree t, tree decl, bool manifestly_const_eval) tree cxx_constant_init (tree t, tree decl) { - return maybe_constant_init_1 (t, decl, false, true); + return maybe_constant_init_1 (t, decl, false, mce_true); } #if 0 diff --git a/gcc/cp/constraint.cc b/gcc/cp/constraint.cc index 2f1678c..44fb086 100644 --- a/gcc/cp/constraint.cc +++ b/gcc/cp/constraint.cc @@ -2704,6 +2704,8 @@ satisfy_declaration_constraints (tree t, sat_info info) static tree satisfy_declaration_constraints (tree t, tree args, sat_info info) { + tree orig_args = args; + /* Update the declaration for diagnostics. */ info.in_decl = t; @@ -2732,7 +2734,7 @@ satisfy_declaration_constraints (tree t, tree args, sat_info info) tree result = boolean_true_node; if (tree norm = get_normalized_constraints_from_decl (t, info.noisy ())) { - if (!push_tinst_level (t, args)) + if (!push_tinst_level (t, orig_args)) return result; tree pattern = DECL_TEMPLATE_RESULT (t); push_to_top_level (); diff --git a/gcc/cp/contracts.cc b/gcc/cp/contracts.cc index f2b126c..3ca2102 100644 --- a/gcc/cp/contracts.cc +++ b/gcc/cp/contracts.cc @@ -863,7 +863,11 @@ remove_contract_attributes (tree fndecl) tree list = NULL_TREE; for (tree p = DECL_ATTRIBUTES (fndecl); p; p = TREE_CHAIN (p)) if (!cxx_contract_attribute_p (p)) - list = tree_cons (TREE_PURPOSE (p), TREE_VALUE (p), list); + { + tree nl = copy_node (p); + TREE_CHAIN (nl) = list; + list = nl; + } DECL_ATTRIBUTES (fndecl) = nreverse (list); } diff --git a/gcc/cp/coroutines.cc b/gcc/cp/coroutines.cc index b92d09f..743da06 100644 --- a/gcc/cp/coroutines.cc +++ b/gcc/cp/coroutines.cc @@ -1852,21 +1852,6 @@ coro_build_frame_access_expr (tree coro_ref, tree member_id, bool preserve_ref, return expr; } -/* Helpers to build EXPR_STMT and void-cast EXPR_STMT, common ops. */ - -static tree -coro_build_expr_stmt (tree expr, location_t loc) -{ - return maybe_cleanup_point_expr_void (build_stmt (loc, EXPR_STMT, expr)); -} - -static tree -coro_build_cvt_void_expr_stmt (tree expr, location_t loc) -{ - tree t = build1 (CONVERT_EXPR, void_type_node, expr); - return coro_build_expr_stmt (t, loc); -} - /* Helpers to build an artificial var, with location LOC, NAME and TYPE, in CTX, and with initializer INIT. */ @@ -2582,8 +2567,7 @@ build_actor_fn (location_t loc, tree coro_frame_type, tree actor, tree fnbody, tree hfa = build_new_method_call (ash, hfa_m, &args, NULL_TREE, LOOKUP_NORMAL, NULL, tf_warning_or_error); r = cp_build_init_expr (ash, hfa); - r = coro_build_cvt_void_expr_stmt (r, loc); - add_stmt (r); + finish_expr_stmt (r); release_tree_vector (args); /* Now we know the real promise, and enough about the frame layout to @@ -2678,8 +2662,7 @@ build_actor_fn (location_t loc, tree coro_frame_type, tree actor, tree fnbody, we must tail call them. However, some targets do not support indirect tail calls to arbitrary callees. See PR94359. */ CALL_EXPR_TAILCALL (resume) = true; - resume = coro_build_cvt_void_expr_stmt (resume, loc); - add_stmt (resume); + finish_expr_stmt (resume); r = build_stmt (loc, RETURN_EXPR, NULL); gcc_checking_assert (maybe_cleanup_point_expr_void (r) == r); diff --git a/gcc/cp/cp-gimplify.cc b/gcc/cp/cp-gimplify.cc index 550cea29..d2423fd 100644 --- a/gcc/cp/cp-gimplify.cc +++ b/gcc/cp/cp-gimplify.cc @@ -1199,8 +1199,11 @@ cp_build_init_expr_for_ctor (tree call, tree init) tree s = build_fold_indirect_ref_loc (loc, a); init = cp_build_init_expr (s, init); if (return_this) - init = build2_loc (loc, COMPOUND_EXPR, TREE_TYPE (call), init, - fold_convert_loc (loc, TREE_TYPE (call), a)); + { + init = build2_loc (loc, COMPOUND_EXPR, TREE_TYPE (call), init, + fold_convert_loc (loc, TREE_TYPE (call), a)); + suppress_warning (init); + } return init; } @@ -1480,7 +1483,9 @@ cp_fold_r (tree *stmt_p, int *walk_subtrees, void *data_) *walk_subtrees = 0; if (!flag_no_inline) { - tree folded = maybe_constant_init (init, TARGET_EXPR_SLOT (stmt)); + tree folded = maybe_constant_init (init, TARGET_EXPR_SLOT (stmt), + (data->flags & ff_mce_false + ? mce_false : mce_unknown)); if (folded != init && TREE_CONSTANT (folded)) init = folded; } diff --git a/gcc/cp/cp-tree.h b/gcc/cp/cp-tree.h index 927f51b..7798efb 100644 --- a/gcc/cp/cp-tree.h +++ b/gcc/cp/cp-tree.h @@ -2206,6 +2206,8 @@ struct GTY(()) language_function { BOOL_BITFIELD invalid_constexpr : 1; BOOL_BITFIELD throwing_cleanup : 1; + /* True if we gave any errors in this function. */ + BOOL_BITFIELD erroneous : 1; hash_table<named_label_hash> *x_named_labels; @@ -8837,6 +8839,7 @@ extern void cxx_constant_dtor (tree, tree); extern tree cxx_constant_init (tree, tree = NULL_TREE); extern tree maybe_constant_value (tree, tree = NULL_TREE, mce_value = mce_unknown); extern tree maybe_constant_init (tree, tree = NULL_TREE, bool = false); +extern tree maybe_constant_init (tree, tree, mce_value); extern tree fold_non_dependent_expr (tree, tsubst_flags_t = tf_warning_or_error, bool = false, tree = NULL_TREE); diff --git a/gcc/cp/decl.cc b/gcc/cp/decl.cc index 4e97093..84398e5 100644 --- a/gcc/cp/decl.cc +++ b/gcc/cp/decl.cc @@ -5339,6 +5339,8 @@ cp_make_fname_decl (location_t loc, tree id, int type_dep) decl = pushdecl_outermost_localscope (decl); if (decl != error_mark_node) add_decl_expr (decl); + else + gcc_assert (seen_error ()); } else { diff --git a/gcc/cp/error.cc b/gcc/cp/error.cc index ec7527e..499eb1b 100644 --- a/gcc/cp/error.cc +++ b/gcc/cp/error.cc @@ -3788,18 +3788,18 @@ cp_print_error_function (diagnostic_text_output_format &text_output, { if (text_output.show_column_p () && s.column != 0) pp_printf (pp, - _(" inlined from %qD at %r%s:%d:%d%R"), + G_(" inlined from %qD at %r%s:%d:%d%R"), fndecl, "locus", s.file, s.line, s.column); else pp_printf (pp, - _(" inlined from %qD at %r%s:%d%R"), + G_(" inlined from %qD at %r%s:%d%R"), fndecl, "locus", s.file, s.line); } else - pp_printf (pp, _(" inlined from %qD"), + pp_printf (pp, G_(" inlined from %qD"), fndecl); } } @@ -3825,22 +3825,22 @@ function_category (tree fn) && DECL_FUNCTION_MEMBER_P (fn)) { if (DECL_STATIC_FUNCTION_P (fn)) - return _("In static member function %qD"); + return G_("In static member function %qD"); else if (DECL_COPY_CONSTRUCTOR_P (fn)) - return _("In copy constructor %qD"); + return G_("In copy constructor %qD"); else if (DECL_CONSTRUCTOR_P (fn)) - return _("In constructor %qD"); + return G_("In constructor %qD"); else if (DECL_DESTRUCTOR_P (fn)) - return _("In destructor %qD"); + return G_("In destructor %qD"); else if (LAMBDA_FUNCTION_P (fn)) - return _("In lambda function"); + return G_("In lambda function"); else if (DECL_XOBJ_MEMBER_FUNCTION_P (fn)) - return _("In explicit object member function %qD"); + return G_("In explicit object member function %qD"); else - return _("In member function %qD"); + return G_("In member function %qD"); } else - return _("In function %qD"); + return G_("In function %qD"); } /* Disable warnings about missing quoting in GCC diagnostics for @@ -3867,8 +3867,8 @@ print_instantiation_full_context (diagnostic_text_output_format &text_output) char *indent = text_output.build_indent_prefix (true); pp_verbatim (text_output.get_printer (), p->list_p () - ? _("%s%s%sIn substitution of %qS:\n") - : _("%s%s%sIn instantiation of %q#D:\n"), + ? G_("%s%s%sIn substitution of %qS:\n") + : G_("%s%s%sIn instantiation of %q#D:\n"), indent, show_file ? LOCATION_FILE (location) : "", show_file ? ": " : "", @@ -3888,10 +3888,10 @@ print_location (diagnostic_text_output_format &text_output, expanded_location xloc = expand_location (loc); pretty_printer *const pp = text_output.get_printer (); if (text_output.show_column_p ()) - pp_verbatim (pp, _("%r%s:%d:%d:%R "), + pp_verbatim (pp, G_("%r%s:%d:%d:%R "), "locus", xloc.file, xloc.line, xloc.column); else - pp_verbatim (pp, _("%r%s:%d:%R "), + pp_verbatim (pp, G_("%r%s:%d:%R "), "locus", xloc.file, xloc.line); } @@ -3984,22 +3984,22 @@ print_instantiation_partial_context_line (diagnostic_text_output_format &text_ou if (t->list_p ()) pp_verbatim (pp, recursive_p - ? _("recursively required by substitution of %qS\n") - : _("required by substitution of %qS\n"), + ? G_("recursively required by substitution of %qS\n") + : G_("required by substitution of %qS\n"), t->get_node ()); else pp_verbatim (pp, recursive_p - ? _("recursively required from %q#D\n") - : _("required from %q#D\n"), + ? G_("recursively required from %q#D\n") + : G_("required from %q#D\n"), t->get_node ()); } else { pp_verbatim (pp, recursive_p - ? _("recursively required from here\n") - : _("required from here\n")); + ? G_("recursively required from here\n") + : G_("required from here\n")); } } @@ -4049,8 +4049,8 @@ print_instantiation_partial_context (diagnostic_text_output_format &text_output, { auto_context_line sentinel (text_output, loc); pp_verbatim (text_output.get_printer (), - _("[ skipping %d instantiation contexts," - " use -ftemplate-backtrace-limit=0 to disable ]\n"), + G_("[ skipping %d instantiation contexts," + " use -ftemplate-backtrace-limit=0 to disable ]\n"), skip); do { loc = t->locus; @@ -4101,7 +4101,7 @@ maybe_print_constexpr_context (diagnostic_text_output_format &text_output) pretty_printer *const pp = text_output.get_printer (); auto_context_line sentinel (text_output, EXPR_LOCATION (t)); pp_verbatim (pp, - _("in %<constexpr%> expansion of %qs"), + G_("in %<constexpr%> expansion of %qs"), s); pp_newline (pp); } @@ -4114,7 +4114,7 @@ print_constrained_decl_info (diagnostic_text_output_format &text_output, { auto_context_line sentinel (text_output, DECL_SOURCE_LOCATION (decl)); pretty_printer *const pp = text_output.get_printer (); - pp_verbatim (pp, "required by the constraints of %q#D\n", decl); + pp_verbatim (pp, G_("required by the constraints of %q#D\n"), decl); } static void @@ -4129,7 +4129,7 @@ print_concept_check_info (diagnostic_text_output_format &text_output, cxx_pretty_printer *const pp = (cxx_pretty_printer *)text_output.get_printer (); - pp_verbatim (pp, "required for the satisfaction of %qE", expr); + pp_verbatim (pp, G_("required for the satisfaction of %qE"), expr); if (map && map != error_mark_node) { tree subst_map = tsubst_parameter_mapping (map, args, tf_none, NULL_TREE); @@ -4151,7 +4151,7 @@ print_constraint_context_head (diagnostic_text_output_format &text_output, { auto_context_line sentinel (text_output, input_location); pretty_printer *const pp = text_output.get_printer (); - pp_verbatim (pp, "required for constraint satisfaction\n"); + pp_verbatim (pp, G_("required for constraint satisfaction\n")); return NULL_TREE; } if (DECL_P (src)) @@ -4180,11 +4180,10 @@ print_requires_expression_info (diagnostic_text_output_format &text_output, auto_context_line sentinel (text_output, cp_expr_loc_or_input_loc (expr)); cxx_pretty_printer *const pp = static_cast <cxx_pretty_printer *> (text_output.get_printer ()); - pp_verbatim (pp, "in requirements "); tree parms = TREE_OPERAND (expr, 0); - if (parms) - pp_verbatim (pp, "with "); + pp_verbatim (pp, parms ? G_("in requirements with ") + : G_("in requirements ")); while (parms) { pp_verbatim (pp, "%q#D", parms); diff --git a/gcc/cp/init.cc b/gcc/cp/init.cc index e589e45..062a493 100644 --- a/gcc/cp/init.cc +++ b/gcc/cp/init.cc @@ -3405,7 +3405,7 @@ build_new_1 (vec<tree, va_gc> **placement, tree type, tree nelts, errval = throw_bad_array_new_length (); if (outer_nelts_check != NULL_TREE) size = build3 (COND_EXPR, sizetype, outer_nelts_check, size, errval); - size = cp_fully_fold (size); + size = fold_to_constant (size); /* Create the argument list. */ vec_safe_insert (*placement, 0, size); /* Do name-lookup to find the appropriate operator. */ @@ -3462,7 +3462,7 @@ build_new_1 (vec<tree, va_gc> **placement, tree type, tree nelts, outer_nelts_check = NULL_TREE; } - size = cp_fully_fold (size); + size = fold_to_constant (size); /* If size is zero e.g. due to type having zero size, try to preserve outer_nelts for constant expression evaluation purposes. */ diff --git a/gcc/cp/lambda.cc b/gcc/cp/lambda.cc index f0a54b6..b2e0ecd 100644 --- a/gcc/cp/lambda.cc +++ b/gcc/cp/lambda.cc @@ -1858,6 +1858,13 @@ prune_lambda_captures (tree body) cp_walk_tree_without_duplicates (&body, mark_const_cap_r, &const_vars); + tree bind_expr = expr_single (DECL_SAVED_TREE (lambda_function (lam))); + if (bind_expr && TREE_CODE (bind_expr) == MUST_NOT_THROW_EXPR) + bind_expr = expr_single (TREE_OPERAND (bind_expr, 0)); + /* FIXME: We don't currently handle noexcept lambda captures correctly, + so bind_expr may not be set; see PR c++/119764. */ + gcc_assert (!bind_expr || TREE_CODE (bind_expr) == BIND_EXPR); + tree *fieldp = &TYPE_FIELDS (LAMBDA_EXPR_CLOSURE (lam)); for (tree *capp = &LAMBDA_EXPR_CAPTURE_LIST (lam); *capp; ) { @@ -1879,6 +1886,23 @@ prune_lambda_captures (tree body) fieldp = &DECL_CHAIN (*fieldp); *fieldp = DECL_CHAIN (*fieldp); + /* And out of the bindings for the function. */ + tree *blockp = &BLOCK_VARS (current_binding_level->blocks); + while (*blockp != DECL_EXPR_DECL (**use)) + blockp = &DECL_CHAIN (*blockp); + *blockp = DECL_CHAIN (*blockp); + + /* And maybe out of the vars declared in the containing + BIND_EXPR, if it's listed there. */ + if (bind_expr) + { + tree *bindp = &BIND_EXPR_VARS (bind_expr); + while (*bindp && *bindp != DECL_EXPR_DECL (**use)) + bindp = &DECL_CHAIN (*bindp); + if (*bindp) + *bindp = DECL_CHAIN (*bindp); + } + /* And remove the capture proxy declaration. */ **use = void_node; continue; diff --git a/gcc/cp/lex.cc b/gcc/cp/lex.cc index c12b084..12af81e 100644 --- a/gcc/cp/lex.cc +++ b/gcc/cp/lex.cc @@ -749,6 +749,9 @@ unqualified_name_lookup_error (tree name, location_t loc) if (IDENTIFIER_ANY_OP_P (name)) error_at (loc, "%qD not defined", name); + else if (!flag_concepts && name == ridpointers[(int)RID_REQUIRES]) + error_at (loc, "%<requires%> only available with %<-std=c++20%> or " + "%<-fconcepts%>"); else { if (!objc_diagnose_private_ivar (name)) diff --git a/gcc/cp/mangle.cc b/gcc/cp/mangle.cc index 02129c6..3d5e96b 100644 --- a/gcc/cp/mangle.cc +++ b/gcc/cp/mangle.cc @@ -1048,6 +1048,12 @@ decl_mangling_context (tree decl) tree extra = LAMBDA_TYPE_EXTRA_SCOPE (TREE_TYPE (decl)); if (extra) return extra; + tcontext = CP_DECL_CONTEXT (decl); + if (LAMBDA_TYPE_P (tcontext)) + /* Lambda type context means this lambda appears between the + lambda-introducer and the open brace of another lambda (c++/119175). + That isn't a real scope; look further into the enclosing scope. */ + return decl_mangling_context (TYPE_NAME (tcontext)); } else if (template_type_parameter_p (decl)) /* template type parms have no mangling context. */ diff --git a/gcc/cp/module.cc b/gcc/cp/module.cc index 37fab5b..5ff5c46 100644 --- a/gcc/cp/module.cc +++ b/gcc/cp/module.cc @@ -12090,6 +12090,8 @@ trees_in::is_matching_decl (tree existing, tree decl, bool is_typedef) gcc_checking_assert (TREE_CODE (e_inner) == TREE_CODE (d_inner)); } + // FIXME: do more precise errors at point of mismatch + const char *mismatch_msg = nullptr; if (TREE_CODE (d_inner) == FUNCTION_DECL) { tree e_ret = fndecl_declared_return_type (existing); @@ -12099,13 +12101,20 @@ trees_in::is_matching_decl (tree existing, tree decl, bool is_typedef) && LAMBDA_TYPE_P (DECL_CONTEXT (d_inner))) /* This has a recursive type that will compare different. */; else if (!same_type_p (d_ret, e_ret)) - goto mismatch; + { + mismatch_msg = G_("conflicting type for imported declaration %#qD"); + goto mismatch; + } tree e_type = TREE_TYPE (e_inner); tree d_type = TREE_TYPE (d_inner); if (DECL_EXTERN_C_P (d_inner) != DECL_EXTERN_C_P (e_inner)) - goto mismatch; + { + mismatch_msg = G_("conflicting language linkage for imported " + "declaration %#qD"); + goto mismatch; + } for (tree e_args = TYPE_ARG_TYPES (e_type), d_args = TYPE_ARG_TYPES (d_type); @@ -12113,10 +12122,18 @@ trees_in::is_matching_decl (tree existing, tree decl, bool is_typedef) e_args = TREE_CHAIN (e_args), d_args = TREE_CHAIN (d_args)) { if (!(e_args && d_args)) - goto mismatch; + { + mismatch_msg = G_("conflicting argument list for imported " + "declaration %#qD"); + goto mismatch; + } if (!same_type_p (TREE_VALUE (d_args), TREE_VALUE (e_args))) - goto mismatch; + { + mismatch_msg = G_("conflicting argument types for imported " + "declaration %#qD"); + goto mismatch; + } } /* If EXISTING has an undeduced or uninstantiated exception @@ -12147,9 +12164,14 @@ trees_in::is_matching_decl (tree existing, tree decl, bool is_typedef) } } } - else if (!DEFERRED_NOEXCEPT_SPEC_P (d_spec) + else if (!DECL_MAYBE_DELETED (d_inner) + && !DEFERRED_NOEXCEPT_SPEC_P (d_spec) && !comp_except_specs (d_spec, e_spec, ce_type)) - goto mismatch; + { + mismatch_msg = G_("conflicting %<noexcept%> specifier for " + "imported declaration %#qD"); + goto mismatch; + } /* Similarly if EXISTING has an undeduced return type, but DECL's is already deduced. */ @@ -12163,16 +12185,26 @@ trees_in::is_matching_decl (tree existing, tree decl, bool is_typedef) } else if (type_uses_auto (d_ret) && !same_type_p (TREE_TYPE (d_type), TREE_TYPE (e_type))) - goto mismatch; + { + mismatch_msg = G_("conflicting deduced return type for " + "imported declaration %#qD"); + goto mismatch; + } /* Similarly if EXISTING has undeduced constexpr, but DECL's is already deduced. */ if (DECL_MAYBE_DELETED (e_inner) && !DECL_MAYBE_DELETED (d_inner) && DECL_DECLARED_CONSTEXPR_P (d_inner)) DECL_DECLARED_CONSTEXPR_P (e_inner) = true; + else if (!DECL_MAYBE_DELETED (e_inner) && DECL_MAYBE_DELETED (d_inner)) + /* Nothing to do. */; else if (DECL_DECLARED_CONSTEXPR_P (e_inner) != DECL_DECLARED_CONSTEXPR_P (d_inner)) - goto mismatch; + { + mismatch_msg = G_("conflicting %<constexpr%> for imported " + "declaration %#qD"); + goto mismatch; + } /* Don't synthesize a defaulted function if we're importing one we've already determined. */ @@ -12184,13 +12216,17 @@ trees_in::is_matching_decl (tree existing, tree decl, bool is_typedef) if (!DECL_ORIGINAL_TYPE (e_inner) || !same_type_p (DECL_ORIGINAL_TYPE (d_inner), DECL_ORIGINAL_TYPE (e_inner))) - goto mismatch; + { + mismatch_msg = G_("conflicting imported declaration %q#D"); + goto mismatch; + } } /* Using cp_tree_equal because we can meet TYPE_ARGUMENT_PACKs here. I suspect the entities that directly do that are things that shouldn't go to duplicate_decls (FIELD_DECLs etc). */ else if (!cp_tree_equal (TREE_TYPE (decl), TREE_TYPE (existing))) { + mismatch_msg = G_("conflicting type for imported declaration %#qD"); mismatch: if (DECL_IS_UNDECLARED_BUILTIN (existing)) /* Just like duplicate_decls, presum the user knows what @@ -12203,11 +12239,9 @@ trees_in::is_matching_decl (tree existing, tree decl, bool is_typedef) equality isn't feasible in general for local entities. */; else { - // FIXME:QOI Might be template specialization from a module, - // not necessarily global module + gcc_checking_assert (mismatch_msg); auto_diagnostic_group d; - error_at (DECL_SOURCE_LOCATION (decl), - "conflicting global module declaration %#qD", decl); + error_at (DECL_SOURCE_LOCATION (decl), mismatch_msg, decl); inform (DECL_SOURCE_LOCATION (existing), "existing declaration %#qD", existing); return false; diff --git a/gcc/cp/name-lookup.cc b/gcc/cp/name-lookup.cc index 1cd982e..498126a 100644 --- a/gcc/cp/name-lookup.cc +++ b/gcc/cp/name-lookup.cc @@ -583,7 +583,7 @@ name_lookup::preserve_state () if (previous) { unsigned length = vec_safe_length (previous->scopes); - vec_safe_reserve (previous->scopes, length * 2); + vec_safe_reserve (previous->scopes, length); for (unsigned ix = length; ix--;) { tree decl = (*previous->scopes)[ix]; diff --git a/gcc/cp/parser.cc b/gcc/cp/parser.cc index 812a7c5..3628cfe 100644 --- a/gcc/cp/parser.cc +++ b/gcc/cp/parser.cc @@ -33634,6 +33634,8 @@ cp_parser_function_definition_after_declarator (cp_parser* parser, = parser->num_template_parameter_lists; parser->num_template_parameter_lists = 0; + int errs = errorcount + sorrycount; + /* If the next token is `try', `__transaction_atomic', or `__transaction_relaxed`, then we are looking at either function-try-block or function-transaction-block. Note that all of these include the @@ -33653,6 +33655,9 @@ cp_parser_function_definition_after_declarator (cp_parser* parser, fn = finish_function (inline_p); check_module_decl_linkage (fn); + if ((errorcount + sorrycount) > errs) + DECL_STRUCT_FUNCTION (fn)->language->erroneous = true; + if (modules_p () && !inline_p && TYPE_P (DECL_CONTEXT (fn)) diff --git a/gcc/cp/pt.cc b/gcc/cp/pt.cc index 63c2ec0..a71705f 100644 --- a/gcc/cp/pt.cc +++ b/gcc/cp/pt.cc @@ -11772,6 +11772,10 @@ tsubst_friend_function (tree decl, tree args) elt.args = DECL_TI_ARGS (spec); elt.spec = NULL_TREE; + if (TMPL_ARGS_HAVE_MULTIPLE_LEVELS (DECL_TI_ARGS (spec)) + && !is_specialization_of_friend (spec, new_template)) + continue; + decl_specializations->remove_elt (&elt); tree& spec_args = DECL_TI_ARGS (spec); @@ -12425,6 +12429,8 @@ apply_late_template_attributes (tree *decl_p, tree attributes, int attr_flags, auto o4 = make_temp_override (scope_chain->omp_declare_target_attribute, NULL); auto o5 = make_temp_override (scope_chain->omp_begin_assumes, NULL); + auto o6 = make_temp_override (target_option_current_node, + target_option_default_node); cplus_decl_attributes (decl_p, late_attrs, attr_flags); @@ -13748,6 +13754,8 @@ add_extra_args (tree extra, tree args, tsubst_flags_t complain, tree in_decl) inst = local; /* else inst is already a full instantiation of the pack. */ register_local_specialization (inst, gen); + if (is_normal_capture_proxy (gen)) + register_local_specialization (inst, DECL_CAPTURED_VARIABLE (gen)); } gcc_assert (!TREE_PURPOSE (extra)); extra = TREE_VALUE (extra); @@ -27750,6 +27758,11 @@ instantiate_body (tree pattern, tree args, tree d, bool nested_p) if (DECL_OMP_DECLARE_REDUCTION_P (code_pattern)) cp_check_omp_declare_reduction (d); + + if (int errs = errorcount + sorrycount) + if (errs > current_tinst_level->errors) + if (function *f = DECL_STRUCT_FUNCTION (d)) + f->language->erroneous = true; } /* We're not deferring instantiation any more. */ @@ -30936,9 +30949,8 @@ alias_ctad_tweaks (tree tmpl, tree uguides) tree aguides = NULL_TREE; tree atparms = INNERMOST_TEMPLATE_PARMS (fullatparms); unsigned natparms = TREE_VEC_LENGTH (atparms); - for (ovl_iterator iter (uguides); iter; ++iter) + for (tree f : lkp_range (uguides)) { - tree f = *iter; tree in_decl = f; location_t loc = DECL_SOURCE_LOCATION (f); tree ret = TREE_TYPE (TREE_TYPE (f)); diff --git a/gcc/cp/rtti.cc b/gcc/cp/rtti.cc index 3539962..18bc479 100644 --- a/gcc/cp/rtti.cc +++ b/gcc/cp/rtti.cc @@ -1318,18 +1318,9 @@ get_pseudo_ti_index (tree type) static tinfo_s * get_tinfo_desc (unsigned ix) { - unsigned len = tinfo_descs->length (); - - if (len <= ix) - { - /* too short, extend. */ - len = ix + 1 - len; - vec_safe_reserve (tinfo_descs, len); - tinfo_s elt; - elt.type = elt.vtable = elt.name = NULL_TREE; - while (len--) - tinfo_descs->quick_push (elt); - } + if (tinfo_descs->length () <= ix) + /* too short, extend. */ + vec_safe_grow_cleared (tinfo_descs, ix + 1); tinfo_s *res = &(*tinfo_descs)[ix]; diff --git a/gcc/cp/semantics.cc b/gcc/cp/semantics.cc index a10ef34..1aa35d3 100644 --- a/gcc/cp/semantics.cc +++ b/gcc/cp/semantics.cc @@ -1180,10 +1180,13 @@ finish_expr_stmt (tree expr) expr = error_mark_node; /* Simplification of inner statement expressions, compound exprs, - etc can result in us already having an EXPR_STMT. */ + etc can result in us already having an EXPR_STMT or other statement + tree. Don't wrap them in EXPR_STMT. */ if (TREE_CODE (expr) != CLEANUP_POINT_EXPR) { - if (TREE_CODE (expr) != EXPR_STMT) + if (TREE_CODE (expr) != EXPR_STMT + && !STATEMENT_CLASS_P (expr) + && TREE_CODE (expr) != STATEMENT_LIST) expr = build_stmt (loc, EXPR_STMT, expr); expr = maybe_cleanup_point_expr_void (expr); } @@ -3082,6 +3085,7 @@ finish_stmt_expr_expr (tree expr, tree stmt_expr) } else if (processing_template_decl) { + /* Not finish_expr_stmt because we don't want convert_to_void. */ expr = build_stmt (input_location, EXPR_STMT, expr); expr = add_stmt (expr); /* Mark the last statement so that we can recognize it as such at @@ -5088,22 +5092,32 @@ static tree finish_type_pack_element (tree idx, tree types, tsubst_flags_t complain) { idx = maybe_constant_value (idx, NULL_TREE, mce_true); - if (TREE_CODE (idx) != INTEGER_CST || !INTEGRAL_TYPE_P (TREE_TYPE (idx))) + if (!INTEGRAL_TYPE_P (TREE_TYPE (idx))) { if (complain & tf_error) - error ("pack index is not an integral constant"); + error ("pack index has non-integral type %qT", TREE_TYPE (idx)); + return error_mark_node; + } + if (TREE_CODE (idx) != INTEGER_CST) + { + if (complain & tf_error) + { + error ("pack index is not an integral constant"); + cxx_constant_value (idx); + } return error_mark_node; } if (tree_int_cst_sgn (idx) < 0) { if (complain & tf_error) - error ("pack index is negative"); + error ("pack index %qE is negative", idx); return error_mark_node; } if (wi::to_widest (idx) >= TREE_VEC_LENGTH (types)) { if (complain & tf_error) - error ("pack index is out of range"); + error ("pack index %qE is out of range for pack of length %qd", + idx, TREE_VEC_LENGTH (types)); return error_mark_node; } return TREE_VEC_ELT (types, tree_to_shwi (idx)); diff --git a/gcc/d/ChangeLog b/gcc/d/ChangeLog index 23b8a81..f816c70 100644 --- a/gcc/d/ChangeLog +++ b/gcc/d/ChangeLog @@ -1,3 +1,78 @@ +2025-04-17 Iain Buclaw <ibuclaw@gdcproject.org> + + * dmd/MERGE: Merge upstream dmd 956e73d64e. + +2025-04-15 Iain Buclaw <ibuclaw@gdcproject.org> + + PR d/119826 + * types.cc (TypeVisitor::visit (TypeEnum *)): Propagate flags of main + enum types to all forward-referenced variants. + +2025-04-15 Iain Buclaw <ibuclaw@gdcproject.org> + + PR d/119799 + * decl.cc (DeclVisitor::visit (VarDeclaration *)): Check front-end + type size before building the VAR_DECL. Allow C symbols to have a + size of `0'. + +2025-04-15 Iain Buclaw <ibuclaw@gdcproject.org> + + PR d/119817 + * imports.cc (ImportVisitor::visit (OverloadSet *)): Don't push + NULL_TREE to vector of import symbols. + +2025-04-12 Iain Buclaw <ibuclaw@gdcproject.org> + + PR d/109023 + * d-compiler.cc: Include dmd/errors.h. + (Compiler::onImport): Implement. + * d-lang.cc (d_handle_option): Handle -finclude-imports. + (d_parse_file): Run semantic on included imports. + * gdc.texi: Document -finclude-imports. + * lang.opt: Add finclude-imports. + * lang.opt.urls: Regenerate. + +2025-04-12 Iain Buclaw <ibuclaw@gdcproject.org> + + PR d/119758 + * d-lang.cc (d_parse_file): Use endswith in test for -fonly= argument. + * d-spec.cc (lang_specific_driver): Rework -fonly= and pass all input + files to the front-end compiler when the option is seen. + +2025-04-11 Iain Buclaw <ibuclaw@gdcproject.org> + + * dmd/MERGE: Merge upstream dmd 1b34fea478. + +2025-04-09 Iain Buclaw <ibuclaw@gdcproject.org> + + PR d/118309 + * modules.cc: Include debug.h + (d_finish_compilation): Call debug_hooks->type_decl on all TYPE_DECLs. + * types.cc: Remove toplev.h include. + (finish_aggregate_type): Don't call rest_of_type_compilation or + rest_of_decl_compilation on type. + (TypeVisitor::visit (TypeEnum *)): Likewise. + +2025-04-09 Iain Buclaw <ibuclaw@gdcproject.org> + + PR d/117832 + * d-tree.h (build_padded_constructor): New prototype. + * d-codegen.cc (build_padded_constructor): New function. + (d_array_value): Call it. + (build_memset_call): Likewise. + (build_struct_literal): Likewise. + (underlying_complex_expr): Likewise. + (build_array_from_val): Likewise. + (build_array_from_exprs): Likewise. + (d_build_call): Likewise. + (get_frame_for_symbol): Likewise. + * d-convert.cc (convert_for_rvalue): Likewise. + (convert_for_assignment): Likewise. + * decl.cc (class DeclVisitor): Likewise. + * expr.cc (class ExprVisitor): Likewise. + * modules.cc (layout_moduleinfo): Likewise. + * typeinfo.cc (class TypeInfoVisitor): Likewise. + 2025-04-08 Iain Buclaw <ibuclaw@gdcproject.org> * dmd/MERGE: Merge upstream dmd 51816cd01d. diff --git a/gcc/d/d-compiler.cc b/gcc/d/d-compiler.cc index 160539d..e18f5d3 100644 --- a/gcc/d/d-compiler.cc +++ b/gcc/d/d-compiler.cc @@ -20,6 +20,7 @@ along with GCC; see the file COPYING3. If not see #include "coretypes.h" #include "dmd/compiler.h" +#include "dmd/errors.h" #include "dmd/expression.h" #include "dmd/identifier.h" #include "dmd/module.h" @@ -164,7 +165,39 @@ Compiler::onParseModule (Module *m) driver intends on compiling the import. */ bool -Compiler::onImport (Module *) +Compiler::onImport (Module *m) { - return false; + if (!includeImports) + return false; + + if (m->filetype != FileType::d && m->filetype != FileType::c) + return false; + + /* All imports modules are included except those in the runtime library. */ + ModuleDeclaration *md = m->md; + if (md && md->id) + { + if (md->packages.length >= 1) + { + if (!strcmp (md->packages.ptr[0]->toChars (), "core") + || !strcmp (md->packages.ptr[0]->toChars (), "std") + || !strcmp (md->packages.ptr[0]->toChars (), "gcc") + || !strcmp (md->packages.ptr[0]->toChars (), "etc")) + return false; + } + else if (!strcmp (md->id->toChars (), "object")) + return false; + } + else if (m->ident) + { + if (!strcmp (m->ident->toChars (), "object")) + return false; + } + + /* This import will be compiled. */ + if (global.params.v.verbose) + message ("compileimport (%s)", m->srcfile.toChars ()); + + compiledImports.push (m); + return true; } diff --git a/gcc/d/d-lang.cc b/gcc/d/d-lang.cc index b3786be..ec2ea59 100644 --- a/gcc/d/d-lang.cc +++ b/gcc/d/d-lang.cc @@ -523,6 +523,10 @@ d_handle_option (size_t scode, const char *arg, HOST_WIDE_INT value, global.params.ignoreUnsupportedPragmas = value; break; + case OPT_finclude_imports: + includeImports = true; + break; + case OPT_finvariants: global.params.useInvariants = value ? CHECKENABLEon : CHECKENABLEoff; break; @@ -1085,9 +1089,9 @@ d_parse_file (void) /* Buffer for contents of .ddoc files. */ OutBuffer ddocbuf; - /* In this mode, the first file name is supposed to be a duplicate - of one of the input files. */ - if (d_option.fonly && strcmp (d_option.fonly, main_input_filename) != 0) + /* In this mode, the main input file is supposed to be the same as the one + given by -fonly=. */ + if (d_option.fonly && !endswith (main_input_filename, d_option.fonly)) error ("%<-fonly=%> argument is different from first input file name"); for (size_t i = 0; i < num_in_fnames; i++) @@ -1309,6 +1313,21 @@ d_parse_file (void) dmd::semantic3 (m, NULL); } + if (includeImports) + { + for (size_t i = 0; i < compiledImports.length; i++) + { + Module *m = compiledImports[i]; + gcc_assert (m->isRoot ()); + + if (global.params.v.verbose) + message ("semantic3 %s", m->toChars ()); + + dmd::semantic3 (m, NULL); + modules.push (m); + } + } + Module::runDeferredSemantic3 (); /* Check again, incase semantic3 pass loaded any more modules. */ diff --git a/gcc/d/d-spec.cc b/gcc/d/d-spec.cc index 7f4a779..c788048 100644 --- a/gcc/d/d-spec.cc +++ b/gcc/d/d-spec.cc @@ -104,8 +104,8 @@ lang_specific_driver (cl_decoded_option **in_decoded_options, /* The total number of arguments with the new stuff. */ unsigned int num_args = 1; - /* "-fonly" if it appears on the command line. */ - const char *only_source_option = 0; + /* "-fonly=" if it appears on the command line. */ + const char *only_source_arg = 0; /* Whether the -o option was used. */ bool saw_opt_o = false; @@ -280,13 +280,13 @@ lang_specific_driver (cl_decoded_option **in_decoded_options, case OPT_fonly_: args[i] |= SKIPOPT; - only_source_option = decoded_options[i].orig_option_with_args_text; + only_source_arg = arg; if (arg != NULL) { - const char *suffix = strrchr (only_source_option, '.'); + const char *suffix = strrchr (only_source_arg, '.'); if (suffix == NULL || strcmp (suffix, ".d") != 0) - only_source_option = concat (only_source_option, ".d", NULL); + only_source_arg = concat (only_source_arg, ".d", NULL); } break; @@ -335,48 +335,52 @@ lang_specific_driver (cl_decoded_option **in_decoded_options, + (phobos_library != PHOBOS_NOLINK) * 4 + 2; new_decoded_options = XNEWVEC (cl_decoded_option, num_args); - i = 0; j = 0; /* Copy the 0th argument, i.e., the name of the program itself. */ - new_decoded_options[j++] = decoded_options[i++]; + new_decoded_options[j++] = decoded_options[0]; /* NOTE: We start at 1 now, not 0. */ - while (i < argc) + for (i = 1; i < argc; i++) { if (args[i] & SKIPOPT) - { - ++i; - continue; - } - - new_decoded_options[j] = decoded_options[i]; + continue; if (!saw_libcxx && (args[i] & WITHLIBCXX)) { - --j; saw_libcxx = &decoded_options[i]; + continue; } - if (args[i] & DSOURCE) + if (only_source_arg && (args[i] & DSOURCE)) { - if (only_source_option) - --j; + if (!endswith (decoded_options[i].arg, only_source_arg)) + continue; } - i++; + new_decoded_options[j] = decoded_options[i]; j++; } - if (only_source_option) + if (only_source_arg) { - const char *only_source_arg = only_source_option + 7; + /* Generate -fonly= option, then copy D input sources that were initially + skipped in first pass over all decoded_options. */ generate_option (OPT_fonly_, only_source_arg, 1, CL_DRIVER, &new_decoded_options[j]); j++; - generate_option_input_file (only_source_arg, - &new_decoded_options[j++]); + for (i = 1; i < argc; i++) + { + if (!(args[i] & DSOURCE)) + continue; + + if (endswith (decoded_options[i].arg, only_source_arg)) + continue; + + new_decoded_options[j] = decoded_options[i]; + j++; + } } /* If no reason to link against libphobos library, then don't add it. */ diff --git a/gcc/d/decl.cc b/gcc/d/decl.cc index 136f78b..9ddf7cf 100644 --- a/gcc/d/decl.cc +++ b/gcc/d/decl.cc @@ -791,6 +791,12 @@ public: } else if (d->isDataseg ()) { + /* When the front-end type size is invalid, an error has already been + given for the declaration or type. */ + dinteger_t size = dmd::size (d->type, d->loc); + if (size == SIZE_INVALID) + return; + tree decl = get_symbol_decl (d); /* Only need to build the VAR_DECL for extern declarations. */ @@ -804,9 +810,7 @@ public: return; /* How big a symbol can be should depend on back-end. */ - tree size = build_integer_cst (dmd::size (d->type, d->loc), - build_ctype (Type::tsize_t)); - if (!valid_constant_size_p (size)) + if (!valid_constant_size_p (build_integer_cst (size, size_type_node))) { error_at (make_location_t (d->loc), "size is too large"); return; @@ -835,8 +839,9 @@ public: } /* Frontend should have already caught this. */ - gcc_assert (!integer_zerop (size) - || d->type->toBasetype ()->isTypeSArray ()); + gcc_assert ((size != 0 && size != SIZE_INVALID) + || d->type->toBasetype ()->isTypeSArray () + || d->isCsymbol ()); d_finish_decl (decl); diff --git a/gcc/d/dmd/MERGE b/gcc/d/dmd/MERGE index a05a50e..58d19b4 100644 --- a/gcc/d/dmd/MERGE +++ b/gcc/d/dmd/MERGE @@ -1,4 +1,4 @@ -51816cd01deee5cc1d7d2c6e1e24788ec655b73e +956e73d64e532a68213970316c2590c572ec03f3 The first line of this file holds the git revision number of the last merge done from the dlang/dmd repository. diff --git a/gcc/d/dmd/expressionsem.d b/gcc/d/dmd/expressionsem.d index 19111e3..b02f6ea 100644 --- a/gcc/d/dmd/expressionsem.d +++ b/gcc/d/dmd/expressionsem.d @@ -6978,10 +6978,10 @@ private extern (C++) final class ExpressionSemanticVisitor : Visitor while (1) { AttribDeclaration ad = s.isAttribDeclaration(); - if (!ad) - break; - if (ad.decl && ad.decl.length == 1) + if (ad && ad.decl && ad.decl.length == 1) s = (*ad.decl)[0]; + else + break; } //printf("inserting '%s' %p into sc = %p\n", s.toChars(), s, sc); diff --git a/gcc/d/dmd/globals.h b/gcc/d/dmd/globals.h index 59952a2..62a575e 100644 --- a/gcc/d/dmd/globals.h +++ b/gcc/d/dmd/globals.h @@ -421,6 +421,7 @@ struct SourceLoc uint32_t line; uint32_t column; uint32_t fileOffset; + DString fileContent; }; struct Loc diff --git a/gcc/d/dmd/lexer.d b/gcc/d/dmd/lexer.d index 63313ac..ed9f7f1 100644 --- a/gcc/d/dmd/lexer.d +++ b/gcc/d/dmd/lexer.d @@ -132,7 +132,7 @@ class Lexer // debug printf("Lexer::Lexer(%p)\n", base); // debug printf("lexer.filename = %s\n", filename); token = Token.init; - this.baseLoc = newBaseLoc(filename, endoffset); + this.baseLoc = newBaseLoc(filename, base[0 .. endoffset]); this.linnum = 1; this.base = base; this.end = base + endoffset; @@ -224,7 +224,7 @@ class Lexer inTokenStringConstant = 0; lastDocLine = 0; - baseLoc = newBaseLoc("#defines", slice.length); + baseLoc = newBaseLoc("#defines", slice); scanloc = baseLoc.getLoc(0); } diff --git a/gcc/d/dmd/location.d b/gcc/d/dmd/location.d index 54b3fb6..393ffb8 100644 --- a/gcc/d/dmd/location.d +++ b/gcc/d/dmd/location.d @@ -64,7 +64,7 @@ nothrow: extern (C++) static Loc singleFilename(const char* filename) { Loc result; - locFileTable ~= new BaseLoc(filename.toDString, locIndex, 0, [0]); + locFileTable ~= new BaseLoc(filename.toDString, null, locIndex, 0, [0]); result.index = locIndex++; return result; } @@ -235,16 +235,20 @@ struct SourceLoc uint column; /// column number (starts at 1) uint fileOffset; /// byte index into file + /// Index `fileOffset` into this to to obtain source code context of this location + const(char)[] fileContent; + // aliases for backwards compatibility alias linnum = line; alias charnum = column; - this(const(char)[] filename, uint line, uint column, uint fileOffset = 0) nothrow @nogc pure @safe + this(const(char)[] filename, uint line, uint column, uint fileOffset = 0, const(char)[] fileContent = null) nothrow @nogc pure @safe { this.filename = filename; this.line = line; this.column = column; this.fileOffset = fileOffset; + this.fileContent = fileContent; } this(Loc loc) nothrow @nogc @trusted @@ -300,15 +304,15 @@ private size_t fileTableIndex(uint index) nothrow @nogc * Create a new source location map for a file * Params: * filename = source file name - * size = space to reserve for locations, equal to the file size in bytes + * fileContent = content of source file * Returns: new BaseLoc */ -BaseLoc* newBaseLoc(const(char)* filename, size_t size) nothrow +BaseLoc* newBaseLoc(const(char)* filename, const(char)[] fileContent) nothrow { - locFileTable ~= new BaseLoc(filename.toDString, locIndex, 1, [0]); + locFileTable ~= new BaseLoc(filename.toDString, fileContent, locIndex, 1, [0]); // Careful: the endloc of a FuncDeclaration can // point to 1 past the very last byte in the file, so account for that - locIndex += size + 1; + locIndex += fileContent.length + 1; return locFileTable[$ - 1]; } @@ -354,6 +358,7 @@ struct BaseLoc @safe nothrow: const(char)[] filename; /// Source file name + const(char)[] fileContents; /// Source file contents uint startIndex; /// Subtract this from Loc.index to get file offset int startLine = 1; /// Line number at index 0 uint[] lines; /// For each line, the file offset at which it starts. At index 0 there's always a 0 entry. @@ -384,11 +389,11 @@ struct BaseLoc { auto fname = filename.toDString; if (substitutions.length == 0) - substitutions ~= BaseLoc(this.filename, 0, 0); + substitutions ~= BaseLoc(this.filename, null, 0, 0); if (fname.length == 0) fname = substitutions[$ - 1].filename; - substitutions ~= BaseLoc(fname, offset, cast(int) (line - lines.length + startLine - 2)); + substitutions ~= BaseLoc(fname, null, offset, cast(int) (line - lines.length + startLine - 2)); } /// Returns: `loc` modified by substitutions from #file / #line directives @@ -408,7 +413,7 @@ struct BaseLoc private SourceLoc getSourceLoc(uint offset) @nogc { const i = getLineIndex(offset); - const sl = SourceLoc(filename, cast(int) (i + startLine), cast(int) (1 + offset - lines[i]), offset); + const sl = SourceLoc(filename, cast(int) (i + startLine), cast(int) (1 + offset - lines[i]), offset, fileContents); return substitute(sl); } diff --git a/gcc/d/dmd/typesem.d b/gcc/d/dmd/typesem.d index 3bc0489..d4c7a58 100644 --- a/gcc/d/dmd/typesem.d +++ b/gcc/d/dmd/typesem.d @@ -3266,9 +3266,19 @@ Type merge(Type type) case Tsarray: // prevents generating the mangle if the array dim is not yet known - if (!type.isTypeSArray().dim.isIntegerExp()) - return type; - goto default; + if (auto ie = type.isTypeSArray().dim.isIntegerExp()) + { + // After TypeSemantic, the length is always converted to size_t, but the parser + // usually generates regular integer types (e.g. in cast(const ubyte[2])) which + // it may try to merge, which then leads to failing implicit conversions as 2LU != 2 + // according to Expression.equals. Only merge array types with size_t lengths for now. + // https://github.com/dlang/dmd/issues/21179 + if (ie.type != Type.tsize_t) + return type; + + goto default; + } + return type; case Tenum: break; diff --git a/gcc/d/gdc.texi b/gcc/d/gdc.texi index 2cb0c4a62..3a8bea0 100644 --- a/gcc/d/gdc.texi +++ b/gcc/d/gdc.texi @@ -277,6 +277,12 @@ Sets @code{__traits(getTargetInfo, "cppStd")} to @code{202002}. Sets @code{__traits(getTargetInfo, "cppStd")} to @code{202302}. @end table +@opindex finclude-imports +@item -finclude-imports +Include imported modules in the compilation, as if they were given on the +command line. When this option is enabled, all imported modules are compiled +except those that are part of libphobos. + @opindex finvariants @opindex fno-invariants @item -fno-invariants diff --git a/gcc/d/imports.cc b/gcc/d/imports.cc index 776caaf..16e4df6 100644 --- a/gcc/d/imports.cc +++ b/gcc/d/imports.cc @@ -182,7 +182,11 @@ public: vec_alloc (tset, d->a.length); for (size_t i = 0; i < d->a.length; i++) - vec_safe_push (tset, build_import_decl (d->a[i])); + { + tree overload = build_import_decl (d->a[i]); + if (overload != NULL_TREE) + vec_safe_push (tset, overload); + } this->result_ = build_tree_list_vec (tset); tset->truncate (0); diff --git a/gcc/d/lang.opt b/gcc/d/lang.opt index 50c6f2f..298ff58 100644 --- a/gcc/d/lang.opt +++ b/gcc/d/lang.opt @@ -327,6 +327,10 @@ fignore-unknown-pragmas D Ignore unsupported pragmas. +finclude-imports +D RejectNegative +Include imported modules in the compilation. + finvariants D Var(flag_invariants) Generate code for class invariant contracts. diff --git a/gcc/d/lang.opt.urls b/gcc/d/lang.opt.urls index fa311d4..b4886bf 100644 --- a/gcc/d/lang.opt.urls +++ b/gcc/d/lang.opt.urls @@ -155,6 +155,9 @@ LangUrlSuffix_D(gdc/Runtime-Options.html#index-fextern-std) fignore-unknown-pragmas LangUrlSuffix_D(gdc/Warnings.html#index-fignore-unknown-pragmas) +finclude-imports +LangUrlSuffix_D(gdc/Runtime-Options.html#index-finclude-imports) + finvariants LangUrlSuffix_D(gdc/Runtime-Options.html#index-finvariants) diff --git a/gcc/d/types.cc b/gcc/d/types.cc index e43fa88..1c74840 100644 --- a/gcc/d/types.cc +++ b/gcc/d/types.cc @@ -1179,6 +1179,26 @@ public: layout_type (t->ctype); + /* Fix up all forward-referenced variants of this enum type. */ + for (tree v = TYPE_MAIN_VARIANT (t->ctype); v; + v = TYPE_NEXT_VARIANT (v)) + { + if (v == t->ctype) + continue; + + TYPE_VALUES (v) = TYPE_VALUES (t->ctype); + TYPE_LANG_SPECIFIC (v) = TYPE_LANG_SPECIFIC (t->ctype); + TYPE_MIN_VALUE (v) = TYPE_MIN_VALUE (t->ctype); + TYPE_MAX_VALUE (v) = TYPE_MAX_VALUE (t->ctype); + TYPE_UNSIGNED (v) = TYPE_UNSIGNED (t->ctype); + TYPE_SIZE (v) = TYPE_SIZE (t->ctype); + TYPE_SIZE_UNIT (v) = TYPE_SIZE_UNIT (t->ctype); + SET_TYPE_MODE (v, TYPE_MODE (t->ctype)); + TYPE_PRECISION (v) = TYPE_PRECISION (t->ctype); + SET_TYPE_ALIGN (v, TYPE_ALIGN (t->ctype)); + TYPE_USER_ALIGN (v) = TYPE_USER_ALIGN (t->ctype); + } + /* Complete forward-referenced fields of this enum type. */ finish_incomplete_fields (t->ctype); } diff --git a/gcc/doc/cfg.texi b/gcc/doc/cfg.texi index b8c6427..bbd6694 100644 --- a/gcc/doc/cfg.texi +++ b/gcc/doc/cfg.texi @@ -297,10 +297,12 @@ edge. The opposite conversion is difficult, but should not happen anyway. The edges can be eliminated via @code{purge_dead_edges} call. @findex REG_EH_REGION, EDGE_ABNORMAL_CALL -In the RTL representation, the destination of an exception edge is -specified by @code{REG_EH_REGION} note attached to the insn. -In case of a trapping call the @code{EDGE_ABNORMAL_CALL} flag is set -too. In the @code{GIMPLE} representation, this extra flag is not set. +In the RTL representation, a @code{REG_EH_REGION} note is attached to +an instruction that can throw an exception. The destination of the +exception edge originating at such an instruction is specified by the +value of the @code{REG_EH_REGION} note. In case of a trapping call +the @code{EDGE_ABNORMAL_CALL} flag is set too. In the @code{GIMPLE} +representation, this extra flag is not set. @findex may_trap_p, tree_could_trap_p In the RTL representation, the predicate @code{may_trap_p} may be used diff --git a/gcc/doc/extend.texi b/gcc/doc/extend.texi index ae3357f..0978c4c 100644 --- a/gcc/doc/extend.texi +++ b/gcc/doc/extend.texi @@ -1933,6 +1933,13 @@ Note that if such a function is called indirectly the compiler may or may not inline it depending on optimization level and a failure to inline an indirect call may or may not be diagnosed. +If you need to use the inlined function in multiple translation units, +you should put the @code{always_inline} attribute on a function +definition in a header file that is included in all translation units +where the function is used. Link-time optimization can inline +functions across translation units, but only if an optimization level +that normally enables inlining is additionally specified. + @cindex @code{artificial} function attribute @item artificial This attribute is useful for small inline wrappers that if possible @@ -12299,15 +12306,6 @@ for the @samp{att} and @samp{intel} dialects of assembler: @item @code{%3} @tab @code{$.L3} @tab @code{OFFSET FLAT:.L3} -@item @code{%4} -@tab @code{$8} -@tab @code{8} -@item @code{%5} -@tab @code{%xmm0} -@tab @code{xmm0} -@item @code{%7} -@tab @code{$0} -@tab @code{0} @end multitable The table below shows the list of supported modifiers and their effects. @@ -12324,32 +12322,17 @@ The table below shows the list of supported modifiers and their effects. @tab @code{%b0} @tab @code{%al} @tab @code{al} -@item @code{B} -@tab print the opcode suffix of b. -@tab @code{%B0} -@tab @code{b} -@tab @item @code{c} @tab Require a constant operand and print the constant expression with no punctuation. @tab @code{%c1} @tab @code{2} @tab @code{2} -@item @code{d} -@tab print duplicated register operand for AVX instruction. -@tab @code{%d5} -@tab @code{%xmm0, %xmm0} -@tab @code{xmm0, xmm0} @item @code{E} @tab Print the address in Double Integer (DImode) mode (8 bytes) when the target is 64-bit. Otherwise mode is unspecified (VOIDmode). @tab @code{%E1} @tab @code{%(rax)} @tab @code{[rax]} -@item @code{g} -@tab Print the V16SFmode name of the register. -@tab @code{%g0} -@tab @code{%zmm0} -@tab @code{zmm0} @item @code{h} @tab Print the QImode name for a ``high'' register. @tab @code{%h0} @@ -12371,16 +12354,6 @@ high 8 bytes of SSE values. For a memref in (%rax), it generates @tab @code{%l3} @tab @code{.L3} @tab @code{.L3} -@item @code{L} -@tab print the opcode suffix of l. -@tab @code{%L0} -@tab @code{l} -@tab -@item @code{N} -@tab print maskz. -@tab @code{%N7} -@tab @code{@{z@}} -@tab @code{@{z@}} @item @code{p} @tab Print raw symbol name (without syntax-specific prefixes). @tab @code{%p2} @@ -12396,76 +12369,20 @@ issue the bare constant. See @code{p} above. @tab @code{%q0} @tab @code{%rax} @tab @code{rax} -@item @code{Q} -@tab print the opcode suffix of q. -@tab @code{%Q0} -@tab @code{q} -@tab -@item @code{R} -@tab print embedded rounding and sae. -@tab @code{%R4} -@tab @code{@{rn-sae@}, } -@tab @code{, @{rn-sae@}} -@item @code{r} -@tab print only sae. -@tab @code{%r4} -@tab @code{@{sae@}, } -@tab @code{, @{sae@}} -@item @code{s} -@tab print a shift double count, followed by the assemblers argument -delimiterprint the opcode suffix of s. -@tab @code{%s1} -@tab @code{$2, } -@tab @code{2, } -@item @code{S} -@tab print the opcode suffix of s. -@tab @code{%S0} -@tab @code{s} -@tab -@item @code{t} -@tab print the V8SFmode name of the register. -@tab @code{%t5} -@tab @code{%ymm0} -@tab @code{ymm0} -@item @code{T} -@tab print the opcode suffix of t. -@tab @code{%T0} -@tab @code{t} -@tab -@item @code{V} -@tab print naked full integer register name without %. -@tab @code{%V0} -@tab @code{eax} -@tab @code{eax} @item @code{w} @tab Print the HImode name of the register. @tab @code{%w0} @tab @code{%ax} @tab @code{ax} -@item @code{W} -@tab print the opcode suffix of w. -@tab @code{%W0} -@tab @code{w} -@tab -@item @code{x} -@tab print the V4SFmode name of the register. -@tab @code{%x5} -@tab @code{%xmm0} -@tab @code{xmm0} -@item @code{y} -@tab print "st(0)" instead of "st" as a register. -@tab @code{%y6} -@tab @code{%st(0)} -@tab @code{st(0)} @item @code{z} @tab Print the opcode suffix for the size of the current integer operand (one of @code{b}/@code{w}/@code{l}/@code{q}). @tab @code{%z0} @tab @code{l} @tab -@item @code{Z} -@tab Like @code{z}, with special suffixes for x87 instructions. @end multitable +@code{V} is a special modifier which prints the name of the full integer +register without @code{%}. @anchor{x86floatingpointasmoperands} @subsubsection x86 Floating-Point @code{asm} Operands @@ -13061,6 +12978,7 @@ C and/or C++ standards, while others remain specific to GNU C. * Binary constants:: Binary constants using the @samp{0b} prefix. * Dollar Signs:: Dollar sign is allowed in identifiers. * Character Escapes:: @samp{\e} stands for the character @key{ESC}. +* Raw String Literals:: C++ raw string literals are supported in C. * Alternate Keywords:: @code{__const__}, @code{__asm__}, etc., for header files. * Function Names:: Printable strings which are the name of the current function. @@ -14082,6 +14000,25 @@ machines, typically because the target assembler does not allow them. You can use the sequence @samp{\e} in a string or character constant to stand for the ASCII character @key{ESC}. +@node Raw String Literals +@subsection Raw String Literals +@cindex raw string literals +@cindex string literals, raw + +The C++11 standard added syntax for raw string literals prefixed +with @samp{R}. This syntax allows you to use an arbitrary delimiter +sequence instead of escaping special characters within the string. +For example, these string constants are all equivalent: + +@smallexample +const char *s1 = "\\"; +const char *s2 = R"(\)"; +const char *s3 = R"foo(\)foo"; +@end smallexample + +As an extension, GCC also accepts raw string literals in C with +@option{-std=gnu99} or later. + @node Alternate Keywords @subsection Alternate Keywords @cindex alternate keywords diff --git a/gcc/doc/gm2.texi b/gcc/doc/gm2.texi index 8baee24..cb52e8c 100644 --- a/gcc/doc/gm2.texi +++ b/gcc/doc/gm2.texi @@ -2699,10 +2699,10 @@ PROCEDURE Example (foo, bar: CARDINAL) : CARDINAL ; VAR myout: CARDINAL ; BEGIN - ASM VOLATILE ("movq %1,%%rax; addq %2,%%rax; movq %%rax,%0" + ASM VOLATILE ("movl %1,%%eax; addl %2,%%eax; movl %%eax,%0" : "=rm" (myout) (* outputs *) : "rm" (foo), "rm" (bar) (* inputs *) - : "rax") ; (* we trash *) + : "eax") ; (* we trash *) RETURN( myout ) END Example ; @end example @@ -2720,10 +2720,10 @@ VAR myout: CARDINAL ; BEGIN ASM VOLATILE ( - "movq %[left],%%rax; addq %[right],%%rax; movq %%rax,%[output]" + "movl %[left],%%eax; addl %[right],%%eax; movl %%eax,%[output]" : [output] "=rm" (myout) (* outputs *) : [left] "rm" (foo), [right] "rm" (bar) (* inputs *) - : "rax") ; (* we trash *) + : "eax") ; (* we trash *) RETURN( myout ) END Example ; @end example diff --git a/gcc/doc/install.texi b/gcc/doc/install.texi index b5509ff..1af0082 100644 --- a/gcc/doc/install.texi +++ b/gcc/doc/install.texi @@ -350,6 +350,12 @@ documentation including the target @code{SYSTEM} definition module. If Python3 is unavailable Modula-2 documentation will include a target independent version of the SYSTEM modules. +@item @anchor{gccrs-prerequisite}gccrs + +The official Rust compiler and Rust build system (cargo) are required for +building various parts of the gccrs frontend, until gccrs can compile them +by itself. The minimum supported Rust version is 1.49. + @item A ``working'' POSIX compatible shell, or GNU bash Necessary when running @command{configure} because some diff --git a/gcc/doc/invoke.texi b/gcc/doc/invoke.texi index a040d65..020442a 100644 --- a/gcc/doc/invoke.texi +++ b/gcc/doc/invoke.texi @@ -211,8 +211,7 @@ in the following sections. @item C++ Language Options @xref{C++ Dialect Options,,Options Controlling C++ Dialect}. @gccoptlist{-fabi-version=@var{n} -fno-access-control --faligned-new=@var{n} -fargs-in-order=@var{n} --fno-assume-sane-operators-new-delete +-faligned-new=@var{n} -fno-assume-sane-operators-new-delete -fchar8_t -fcheck-new -fconcepts -fconstexpr-depth=@var{n} -fconstexpr-cache-depth=@var{n} -fconstexpr-loop-limit=@var{n} -fconstexpr-ops-limit=@var{n} @@ -235,6 +234,7 @@ in the following sections. -fno-optional-diags -fno-pretty-templates -frange-for-ext-temps -fno-rtti -fsized-deallocation +-fstrong-eval-order@r{[}=@var{kind}@r{]} -ftemplate-backtrace-limit=@var{n} -ftemplate-depth=@var{n} -fno-threadsafe-statics -fuse-cxa-atexit @@ -593,7 +593,7 @@ Objective-C and Objective-C++ Dialects}. -finline-functions -finline-functions-called-once -finline-limit=@var{n} -finline-small-functions -fipa-modref -fipa-cp -fipa-cp-clone -fipa-bit-cp -fipa-vrp -fipa-pta -fipa-profile -fipa-pure-const --fipa-reference -fipa-reference-addressable +-fipa-reference -fipa-reference-addressable -fipa-reorder-for-locality -fipa-stack-alignment -fipa-icf -fira-algorithm=@var{algorithm} -flate-combine-instructions -flifetime-dse -flive-patching=@var{level} -fira-region=@var{region} -fira-hoist-pressure @@ -642,8 +642,8 @@ Objective-C and Objective-C++ Dialects}. -fsplit-wide-types -fsplit-wide-types-early -fssa-backprop -fssa-phiopt -fstdarg-opt -fstore-merging -fstrict-aliasing -fipa-strict-aliasing -fthread-jumps -ftracer -ftree-bit-ccp --ftree-builtin-call-dce -ftree-ccp -ftree-ch --ftree-coalesce-vars -ftree-copy-prop -ftree-dce -ftree-dominator-opts +-ftree-builtin-call-dce -ftree-ccp -ftree-ch -ftree-coalesce-vars +-ftree-copy-prop -ftree-cselim -ftree-dce -ftree-dominator-opts -ftree-dse -ftree-forwprop -ftree-fre -fcode-hoisting -ftree-loop-if-convert -ftree-loop-im -ftree-phiprop -ftree-loop-distribution -ftree-loop-distribute-patterns @@ -2199,6 +2199,7 @@ those that have already been displayed. If @option{--help} is also specified anywhere on the command line then this takes precedence over any @option{--help=} option. +@opindex Q If the @option{-Q} option appears on the command line before the @option{--help=} option, then the descriptive text displayed by @option{--help=} is changed. Instead of describing the displayed @@ -3581,12 +3582,19 @@ type. @opindex fstrong-eval-order @item -fstrong-eval-order +@itemx -fstrong-eval-order=@var{kind} Evaluate member access, array subscripting, and shift expressions in left-to-right order, and evaluate assignment in right-to-left order, -as adopted for C++17. Enabled by default with @option{-std=c++17}. +as adopted for C++17. @option{-fstrong-eval-order} is equivalent to +@option{-fstrong-eval-order=all}, +and is enabled by default with @option{-std=c++17} or later. + @option{-fstrong-eval-order=some} enables just the ordering of member -access and shift expressions, and is the default without -@option{-std=c++17}. +access and shift expressions, and is the default for C++ dialects prior to +C++17. + +@option{-fstrong-eval-order=none} is equivalent to +@option{-fno-strong-eval-order}. @opindex ftemplate-backtrace-limit @item -ftemplate-backtrace-limit=@var{n} @@ -12739,6 +12747,7 @@ complexity than at @option{-O}. -fipa-pure-const -fipa-reference -fipa-reference-addressable +-fivopts -fmerge-constants -fmove-loop-invariants -fmove-loop-stores @@ -12847,6 +12856,13 @@ by @option{-O2} and also turns on the following optimization flags: Reduce compilation time and make debugging produce the expected results. This is the default. +At @option{-O0}, GCC completely disables most optimization passes; +they are not run even if you explicitly enable them on the command +line, or are listed by @option{-Q --help=optimizers} as being enabled by +default. Many optimizations performed by GCC depend on code analysis +or canonicalization passes that are enabled by @option{-O}, and it would +not be useful to run individual optimization passes in isolation. + @opindex Os @item -Os Optimize for size. @option{-Os} enables all @option{-O2} optimizations @@ -13864,6 +13880,21 @@ Enabled by default at @option{-O1} and higher. Discover read-only, write-only and non-addressable static variables. Enabled by default at @option{-O1} and higher. +@opindex fipa-reorder-for-locality +@item -fipa-reorder-for-locality +Group call chains close together in the binary layout to improve code +locality and minimize jump distances between frequently called functions. +Unlike @option{-freorder-functions} this pass considers the call +chains between functions and groups them together, rather than grouping all +hot/normal/cold/never-executed functions into separate sections. +Unlike @option{-fprofile-reorder-functions} it aims to improve code locality +throughout the runtime of the program rather than focusing on program startup. +This option is incompatible with an explicit +@option{-flto-partition=} option since it enforces a custom partitioning +scheme. +If using this option it is recommended to also use profile feedback, but this +option is not enabled by default otherwise. + @opindex fipa-stack-alignment @item -fipa-stack-alignment Reduce stack alignment on call sites if possible. @@ -14072,6 +14103,12 @@ in this pass can be limited using @option{max-tail-merge-comparisons} parameter and @option{max-tail-merge-iterations} parameter. +@opindex ftree-cselim +@item -ftree-cselim +Perform conditional store elimination on trees. This flag is enabled by +default at @option{-O1} and higher on targets that have conditional +move instructions. + @opindex ftree-dce @item -ftree-dce Perform dead code elimination (DCE) on trees. This flag is enabled by @@ -14278,6 +14315,7 @@ Enabled by default at @option{-O1} and higher. @item -fivopts Perform induction variable optimizations (strength reduction, induction variable merging and induction variable elimination) on trees. +Enabled by default at @option{-O1} and higher. @opindex ftree-parallelize-loops @item -ftree-parallelize-loops=n @@ -14593,11 +14631,13 @@ Enabled for x86 at levels @option{-O2}, @option{-O3}, @option{-Os}. @opindex freorder-functions @item -freorder-functions Reorder functions in the object file in order to -improve code locality. This is implemented by using special -subsections @code{.text.hot} for most frequently executed functions and -@code{.text.unlikely} for unlikely executed functions. Reordering is done by -the linker so object file format must support named sections and linker must -place them in a reasonable way. +improve code locality. Unlike @option{-fipa-reorder-for-locality} this option +prioritises grouping all functions within a category +(hot/normal/cold/never-executed) together. +This is implemented by using special subsections @code{.text.hot} for most +frequently executed functions and @code{.text.unlikely} for unlikely executed +functions. Reordering is done by the linker so object file format must support +named sections and linker must place them in a reasonable way. This option isn't effective unless you either provide profile feedback (see @option{-fprofile-arcs} for details) or manually annotate functions with @@ -14609,12 +14649,14 @@ Enabled at levels @option{-O2}, @option{-O3}, @option{-Os}. @item -fstrict-aliasing Allow the compiler to assume the strictest aliasing rules applicable to the language being compiled. For C (and C++), this activates -optimizations based on the type of expressions. In particular, an -object of one type is assumed never to reside at the same address as an -object of a different type, unless the types are almost the same. For -example, an @code{unsigned int} can alias an @code{int}, but not a -@code{void*} or a @code{double}. A character type may alias any other -type. +optimizations based on the type of expressions. In particular, accessing +an object of one type via an expression of a different type is not allowed, +unless the types are @dfn{compatible types}, differ only in signedness or +qualifiers, or the expression has a character type. Accessing scalar +objects via a corresponding vector type is also allowed. + +For example, an @code{unsigned int} can alias an @code{int}, but not a +@code{void*} or a @code{double}. A character type may alias any other type. @anchor{Type-punning}Pay special attention to code like this: @smallexample @@ -15318,32 +15360,24 @@ arithmetic. These options trade off between speed and correctness. All must be specifically enabled. @table @gcctabopt -@opindex ffloat-store -@item -ffloat-store -Do not store floating-point variables in registers, and inhibit other -options that might change whether a floating-point value is taken from a -register or memory. - -@cindex floating-point precision -This option prevents undesirable excess precision on machines such as -the 68000 where the floating registers (of the 68881) keep more -precision than a @code{double} is supposed to have. Similarly for the -x86 architecture. For most programs, the excess precision does only -good, but a few programs rely on the precise definition of IEEE floating -point. Use @option{-ffloat-store} for such programs, after modifying -them to store all pertinent intermediate computations into variables. - @opindex fexcess-precision @item -fexcess-precision=@var{style} -This option allows further control over excess precision on machines +This option allows control over excess precision on machines where floating-point operations occur in a format with more precision or -range than the IEEE standard and interchange floating-point types. By -default, @option{-fexcess-precision=fast} is in effect; this means that +range than the IEEE standard and interchange floating-point types. +An example of such a target is x87 floating point on x86 processors, +which uses an 80-bit representation internally instead of the 64-bit +IEEE format. For most programs, the excess precision is harmless, +but some programs may rely on the +requirements of the C or C++ language standards for handling IEEE values. + +By default, @option{-fexcess-precision=fast} is in effect; this means that operations may be carried out in a wider precision than the types specified in the source if that would result in faster code, and it is unpredictable when rounding to the types specified in the source code takes place. When compiling C or C++, if @option{-fexcess-precision=standard} is specified -then excess precision follows the rules specified in ISO C99 or C++; in particular, +then excess precision follows the rules specified in ISO C99 or C++; +in particular, both casts and assignments cause values to be rounded to their semantic types (whereas @option{-ffloat-store} only affects assignments). This option is enabled by default for C or C++ if a strict @@ -15361,6 +15395,18 @@ or @option{-mfpmath=sse+387} is specified; in the former case, IEEE semantics apply without excess precision, and in the latter, rounding is unpredictable. +@opindex ffloat-store +@item -ffloat-store +Do not store floating-point variables in registers, and inhibit other +options that might change whether a floating-point value is taken from a +register or memory. This option has generally been subsumed by +@option{-fexcess-precision=standard}, which is more general. If you do use +@option{-ffloat-store}, you may need to modify your program to explicitly +store intermediate computations in temporary variables since +@option{-ffloat-store} handles rounding to IEEE format +only on assignments and not casts as @option{-fexcess-precision=standard} +does. + @opindex ffast-math @item -ffast-math Sets the options @option{-fno-math-errno}, @option{-funsafe-math-optimizations}, @@ -15495,9 +15541,8 @@ default state for @code{FENV_ACCESS}. @opindex frounding-math @item -frounding-math Disable transformations and optimizations that assume default floating-point -rounding behavior. This is round-to-zero for all floating point -to integer conversions, and round-to-nearest for all other arithmetic -truncations. This option should be specified for programs that change +rounding behavior (round-to-nearest). +This option should be specified for programs that change the FP rounding mode dynamically, or that may be executed with a non-default rounding mode. This option disables constant folding of floating-point expressions at compile time (which may be affected by @@ -15619,7 +15664,8 @@ Enabled by @option{-fprofile-generate}, @option{-fprofile-use}, and @item -fprofile-reorder-functions Function reordering based on profile instrumentation collects first time of execution of a function and orders these functions -in ascending order. +in ascending order, aiming to optimize program startup through more +efficient loading of text segments. Enabled with @option{-fprofile-use}. @@ -21268,8 +21314,13 @@ Toggle @option{-fvar-tracking-assignments}, in the same way that @opindex Q @item -Q -Makes the compiler print out each function name as it is compiled, and -print some statistics about each pass when it finishes. +When used on the command line prior to @option{--help=}, @option{-Q} +acts as a modifier to the help output. @xref{Overall Options}, +for details about @option{--help=}. + +Otherwise, this option makes the compiler print out each function name +as it is compiled, and print some statistics about each pass when it +finishes. @opindex ftime-report @item -ftime-report @@ -33328,7 +33379,7 @@ system representing a certain processor type. Possible values for @samp{z9-109}, @samp{z9-ec}/@samp{arch7}, @samp{z10}/@samp{arch8}, @samp{z196}/@samp{arch9}, @samp{zEC12}, @samp{z13}/@samp{arch11}, @samp{z14}/@samp{arch12}, @samp{z15}/@samp{arch13}, -@samp{z16}/@samp{arch14}, and @samp{native}. +@samp{z16}/@samp{arch14}, @samp{z17}/@samp{arch15}, and @samp{native}. The default is @option{-march=z900}. @@ -34822,7 +34873,8 @@ produces code optimized for the local machine under the constraints of the selected instruction set. @item x86-64 -A generic CPU with 64-bit extensions. +A generic CPU with 64-bit extensions, MMX, SSE, SSE2, and FXSR instruction set +support. @item x86-64-v2 @itemx x86-64-v3 @@ -34855,7 +34907,7 @@ Intel Lakemont MCU, based on Intel Pentium CPU. Intel Pentium MMX CPU, based on Pentium core with MMX instruction set support. @item pentiumpro -Intel Pentium Pro CPU@. +Intel Pentium Pro CPU with no MMX support. @item i686 When used with @option{-march}, the Pentium Pro diff --git a/gcc/except.cc b/gcc/except.cc index d5eb927..0fe1e09 100644 --- a/gcc/except.cc +++ b/gcc/except.cc @@ -970,12 +970,26 @@ expand_dw2_landing_pad_for_region (eh_region region) { /* Nothing */ } if (region->exc_ptr_reg) - emit_move_insn (region->exc_ptr_reg, - gen_rtx_REG (ptr_mode, EH_RETURN_DATA_REGNO (0))); + { + rtx exc_ptr_reg; + if (EH_RETURN_DATA_REGNO (0) != INVALID_REGNUM) + exc_ptr_reg = gen_rtx_REG (ptr_mode, EH_RETURN_DATA_REGNO (0)); + else + /* The target must be doing something special. Submit a dummy. */ + exc_ptr_reg = constm1_rtx; + emit_move_insn (region->exc_ptr_reg, exc_ptr_reg); + } if (region->filter_reg) - emit_move_insn (region->filter_reg, - gen_rtx_REG (targetm.eh_return_filter_mode (), - EH_RETURN_DATA_REGNO (1))); + { + rtx filter_reg; + if (EH_RETURN_DATA_REGNO (1) != INVALID_REGNUM) + filter_reg = gen_rtx_REG (targetm.eh_return_filter_mode (), + EH_RETURN_DATA_REGNO (1)); + else + /* The target must be doing something special. Submit a dummy. */ + filter_reg = constm1_rtx; + emit_move_insn (region->filter_reg, filter_reg); + } } /* Expand the extra code needed at landing pads for dwarf2 unwinding. */ @@ -2935,7 +2949,14 @@ switch_to_exception_section (const char * ARG_UNUSED (fnname)) { section *s; - if (exception_section) + if (exception_section + /* Don't use the cached section for comdat if it will be different. */ +#ifdef HAVE_LD_EH_GC_SECTIONS + && !(targetm_common.have_named_sections + && DECL_COMDAT_GROUP (current_function_decl) + && HAVE_COMDAT_GROUP) +#endif + ) s = exception_section; else { diff --git a/gcc/expmed.cc b/gcc/expmed.cc index df09cbc..8cf10d9 100644 --- a/gcc/expmed.cc +++ b/gcc/expmed.cc @@ -285,7 +285,7 @@ init_expmed (void) for (speed = 0; speed < 2; speed++) { crtl->maybe_hot_insn_p = speed; - set_zero_cost (speed, set_src_cost (const0_rtx, mode, speed)); + set_zero_cost (speed, set_src_cost (const0_rtx, QImode, speed)); for (mode = MIN_MODE_INT; mode <= MAX_MODE_INT; mode = (machine_mode)(mode + 1)) diff --git a/gcc/flag-types.h b/gcc/flag-types.h index 0127698..db57376 100644 --- a/gcc/flag-types.h +++ b/gcc/flag-types.h @@ -404,7 +404,15 @@ enum lto_partition_model { LTO_PARTITION_BALANCED = 2, LTO_PARTITION_1TO1 = 3, LTO_PARTITION_MAX = 4, - LTO_PARTITION_CACHE = 5 + LTO_PARTITION_CACHE = 5, + LTO_PARTITION_DEFAULT= 6 +}; + +/* flag_lto_locality_cloning initialization values. */ +enum lto_locality_cloning_model { + LTO_LOCALITY_NO_CLONING = 0, + LTO_LOCALITY_NON_INTERPOSABLE_CLONING = 1, + LTO_LOCALITY_MAXIMAL_CLONING = 2, }; /* flag_lto_linker_output initialization values. */ diff --git a/gcc/fold-const.cc b/gcc/fold-const.cc index 1275ef7..c9471ea 100644 --- a/gcc/fold-const.cc +++ b/gcc/fold-const.cc @@ -7246,6 +7246,12 @@ tree_swap_operands_p (const_tree arg0, const_tree arg1) if (TREE_CONSTANT (arg0)) return true; + /* Put invariant address in arg1. */ + if (is_gimple_invariant_address (arg1)) + return false; + if (is_gimple_invariant_address (arg0)) + return true; + /* It is preferable to swap two SSA_NAME to ensure a canonical form for commutative and comparison operators. Ensuring a canonical form allows the optimizers to find additional redundancies without diff --git a/gcc/fortran/ChangeLog b/gcc/fortran/ChangeLog index c9c07b35..56325a9 100644 --- a/gcc/fortran/ChangeLog +++ b/gcc/fortran/ChangeLog @@ -1,3 +1,94 @@ +2025-04-19 Steven G. Kargl <kargl@gcc.gnu.org> + + PR fortran/119836 + * resolve.cc (check_pure_function): Fix checking for + an impure subprogram within a DO CONCURRENT construct. + (pure_subroutine): Ditto. + +2025-04-16 Harald Anlauf <anlauf@gmx.de> + + PR fortran/106948 + * resolve.cc (gfc_pure_function): If a function has been resolved, + but esym is not yet set, look at its attributes to see whether it + is pure or elemental. + +2025-04-15 Tobias Burnus <tburnus@baylibre.com> + + * f95-lang.cc (LANG_HOOKS_OMP_DEEP_MAPPING, + LANG_HOOKS_OMP_DEEP_MAPPING_P, LANG_HOOKS_OMP_DEEP_MAPPING_CNT): + Define. + * openmp.cc (gfc_match_omp_clause_reduction): Fix location setting. + (resolve_omp_clauses): Permit allocatable components, reject + them and polymorphic variables in PRIVATE/FIRSTPRIVATE. + * trans-decl.cc (add_clause): Set clause location. + * trans-openmp.cc (gfc_has_alloc_comps): Add ptr_ok and + shallow_alloc_only Boolean arguments. + (gfc_omp_replace_alloc_by_to_mapping): New. + (gfc_omp_private_outer_ref, gfc_walk_alloc_comps, + gfc_omp_clause_default_ctor, gfc_omp_clause_copy_ctor, + gfc_omp_clause_assign_op, gfc_omp_clause_dtor): Update call to it. + (gfc_omp_finish_clause): Minor cleanups, improve location data, + handle allocatable components. + (gfc_omp_deep_mapping_map, gfc_omp_deep_mapping_item, + gfc_omp_deep_mapping_comps, gfc_omp_gen_simple_loop, + gfc_omp_get_array_size, gfc_omp_elmental_loop, + gfc_omp_deep_map_kind_p, gfc_omp_deep_mapping_int_p, + gfc_omp_deep_mapping_p, gfc_omp_deep_mapping_do, + gfc_omp_deep_mapping_cnt, gfc_omp_deep_mapping): New. + (gfc_trans_omp_array_section): Save array descriptor in case + deep-mapping lang hook will need it. + (gfc_trans_omp_clauses): Likewise; use better clause location data. + * trans.h (gfc_omp_deep_mapping_p, gfc_omp_deep_mapping_cnt, + gfc_omp_deep_mapping): Add function prototypes. + +2025-04-13 Thomas Koenig <tkoenig@gcc.gnu.org> + + PR fortran/119669 + * interface.cc (compare_parameter): Error when mismatch between + formal argument as subroutine and function. If the dummy + argument is a known function, set its typespec. + +2025-04-12 Thomas Schwinge <tschwinge@baylibre.com> + + PR fortran/101602 + * trans-stmt.cc (gfc_trans_concurrent_locality_spec): Fix + 'static_assert'. + +2025-04-09 Harald Anlauf <anlauf@gmx.de> + + PR fortran/119656 + * interface.cc (gfc_compare_actual_formal): Fix front-end memleak + when searching for matching interfaces. + * trans-expr.cc (gfc_conv_procedure_call): If there is a formal + dummy corresponding to an absent argument, use its type, and only + fall back to inferred type otherwise. + +2025-04-09 Paul Thomas <pault@gcc.gnu.org> + and Harald Anlauf <anlauf@gcc.gnu.org> + + PR fortran/119460 + * iresolve.cc (generate_reduce_op_wrapper): Increase the size + of 'tname'. Change intent of 'a' and 'b' to intent_in. + * trans-decl.cc (add_argument_checking): Do not test artificial + formal symbols. + * trans-expr.cc (gfc_conv_procedure_call): Remove reduce_scalar + and the blocks triggered by it. + * trans-intrinsic.cc (gfc_conv_intrinsic_function): Set the + result of non-character, scalar reduce to be allocatable. + +2025-04-09 Tobias Burnus <tburnus@baylibre.com> + + PR fortran/101602 + * resolve.cc (resolve_locality_spec): Remove 'sorry, unimplemented'. + * trans-stmt.cc (struct symbol_and_tree_t): New. + (gfc_trans_concurrent_locality_spec): New. + (gfc_trans_forall_1): Call it; update to handle local and local_init. + * trans-decl.cc (gfc_start_saved_local_decls, + gfc_stop_saved_local_decls): New; moved code from ... + (gfc_process_block_locals): ... here. Call it. + * trans.h (gfc_start_saved_local_decls, + gfc_stop_saved_local_decls): Declare. + 2025-04-02 Sandra Loosemore <sloosemore@baylibre.com> PR middle-end/118965 diff --git a/gcc/fortran/f95-lang.cc b/gcc/fortran/f95-lang.cc index 124d62f..1f09553 100644 --- a/gcc/fortran/f95-lang.cc +++ b/gcc/fortran/f95-lang.cc @@ -148,6 +148,9 @@ gfc_get_sarif_source_language (const char *) #undef LANG_HOOKS_OMP_CLAUSE_LINEAR_CTOR #undef LANG_HOOKS_OMP_CLAUSE_DTOR #undef LANG_HOOKS_OMP_FINISH_CLAUSE +#undef LANG_HOOKS_OMP_DEEP_MAPPING +#undef LANG_HOOKS_OMP_DEEP_MAPPING_P +#undef LANG_HOOKS_OMP_DEEP_MAPPING_CNT #undef LANG_HOOKS_OMP_ALLOCATABLE_P #undef LANG_HOOKS_OMP_SCALAR_TARGET_P #undef LANG_HOOKS_OMP_SCALAR_P @@ -188,6 +191,9 @@ gfc_get_sarif_source_language (const char *) #define LANG_HOOKS_OMP_CLAUSE_LINEAR_CTOR gfc_omp_clause_linear_ctor #define LANG_HOOKS_OMP_CLAUSE_DTOR gfc_omp_clause_dtor #define LANG_HOOKS_OMP_FINISH_CLAUSE gfc_omp_finish_clause +#define LANG_HOOKS_OMP_DEEP_MAPPING gfc_omp_deep_mapping +#define LANG_HOOKS_OMP_DEEP_MAPPING_P gfc_omp_deep_mapping_p +#define LANG_HOOKS_OMP_DEEP_MAPPING_CNT gfc_omp_deep_mapping_cnt #define LANG_HOOKS_OMP_ALLOCATABLE_P gfc_omp_allocatable_p #define LANG_HOOKS_OMP_SCALAR_P gfc_omp_scalar_p #define LANG_HOOKS_OMP_SCALAR_TARGET_P gfc_omp_scalar_target_p diff --git a/gcc/fortran/interface.cc b/gcc/fortran/interface.cc index c702239..1e552a3 100644 --- a/gcc/fortran/interface.cc +++ b/gcc/fortran/interface.cc @@ -2534,16 +2534,33 @@ compare_parameter (gfc_symbol *formal, gfc_expr *actual, gfc_find_symbol (actual_name, gsym->ns, 0, &global_asym); if (global_asym != NULL) { - gcc_assert (formal->attr.function); - if (!gfc_compare_types (&global_asym->ts, &formal->ts)) + if (formal->attr.subroutine) { - gfc_error ("Type mismatch at %L passing global " - "function %qs declared at %L (%s/%s)", - &actual->where, actual_name, &gsym->where, - gfc_typename (&global_asym->ts), - gfc_dummy_typename (&formal->ts)); + gfc_error ("Mismatch between subroutine and " + "function at %L", &actual->where); return false; } + else if (formal->attr.function) + { + if (!gfc_compare_types (&global_asym->ts, + &formal->ts)) + { + gfc_error ("Type mismatch at %L passing global " + "function %qs declared at %L (%s/%s)", + &actual->where, actual_name, + &gsym->where, + gfc_typename (&global_asym->ts), + gfc_dummy_typename (&formal->ts)); + return false; + } + } + else + { + /* The global symbol is a function. Set the formal + argument acordingly. */ + formal->attr.function = 1; + formal->ts = global_asym->ts; + } } } } diff --git a/gcc/fortran/openmp.cc b/gcc/fortran/openmp.cc index ded80b7..df82940 100644 --- a/gcc/fortran/openmp.cc +++ b/gcc/fortran/openmp.cc @@ -1588,7 +1588,7 @@ gfc_match_omp_clause_reduction (char pc, gfc_omp_clauses *c, bool openacc, { gfc_omp_namelist *p = gfc_get_omp_namelist (), **tl; p->sym = n->sym; - p->where = p->where; + p->where = n->where; p->u.map.op = OMP_MAP_ALWAYS_TOFROM; tl = &c->lists[OMP_LIST_MAP]; @@ -9681,22 +9681,6 @@ resolve_omp_clauses (gfc_code *code, gfc_omp_clauses *omp_clauses, && n->sym->as->type == AS_ASSUMED_SIZE) gfc_error ("Assumed size array %qs in %s clause at %L", n->sym->name, name, &n->where); - if (!openacc - && list == OMP_LIST_MAP - && n->sym->ts.type == BT_DERIVED - && n->sym->ts.u.derived->attr.alloc_comp) - gfc_error ("List item %qs with allocatable components is not " - "permitted in map clause at %L", n->sym->name, - &n->where); - if (!openacc - && (list == OMP_LIST_MAP - || list == OMP_LIST_FROM - || list == OMP_LIST_TO) - && ((n->expr && n->expr->ts.type == BT_CLASS) - || (!n->expr && n->sym->ts.type == BT_CLASS))) - gfc_warning (OPT_Wopenmp, - "Mapping polymorphic list item at %L is " - "unspecified behavior", &n->where); if (list == OMP_LIST_MAP && !openacc) switch (code->op) { @@ -10008,9 +9992,11 @@ resolve_omp_clauses (gfc_code *code, gfc_omp_clauses *omp_clauses, n->sym->name, name, &n->where); if (!openacc - && list == OMP_LIST_FIRSTPRIVATE - && ((n->expr && n->expr->ts.type == BT_CLASS) - || (!n->expr && n->sym->ts.type == BT_CLASS))) + && (list == OMP_LIST_PRIVATE + || list == OMP_LIST_FIRSTPRIVATE) + && ((n->sym->ts.type == BT_DERIVED + && n->sym->ts.u.derived->attr.alloc_comp) + || n->sym->ts.type == BT_CLASS)) switch (code->op) { case EXEC_OMP_TARGET: @@ -10025,9 +10011,19 @@ resolve_omp_clauses (gfc_code *code, gfc_omp_clauses *omp_clauses, case EXEC_OMP_TARGET_TEAMS_DISTRIBUTE_PARALLEL_DO_SIMD: case EXEC_OMP_TARGET_TEAMS_DISTRIBUTE_SIMD: case EXEC_OMP_TARGET_TEAMS_LOOP: - gfc_warning (OPT_Wopenmp, - "FIRSTPRIVATE with polymorphic list item at " - "%L is unspecified behavior", &n->where); + if (n->sym->ts.type == BT_DERIVED + && n->sym->ts.u.derived->attr.alloc_comp) + gfc_error ("Sorry, list item %qs at %L with allocatable" + " components is not yet supported in %s " + "clause", n->sym->name, &n->where, + list == OMP_LIST_PRIVATE ? "PRIVATE" + : "FIRSTPRIVATE"); + else + gfc_error ("Polymorphic list item %qs at %L in %s " + "clause has unspecified behavior and " + "unsupported", n->sym->name, &n->where, + list == OMP_LIST_PRIVATE ? "PRIVATE" + : "FIRSTPRIVATE"); break; default: break; diff --git a/gcc/fortran/resolve.cc b/gcc/fortran/resolve.cc index cdf043b..f03708e 100644 --- a/gcc/fortran/resolve.cc +++ b/gcc/fortran/resolve.cc @@ -3190,6 +3190,13 @@ gfc_pure_function (gfc_expr *e, const char **name) || e->value.function.isym->elemental; *name = e->value.function.isym->name; } + else if (e->symtree && e->symtree->n.sym && e->symtree->n.sym->attr.dummy) + { + /* The function has been resolved, but esym is not yet set. + This can happen with functions as dummy argument. */ + pure = e->symtree->n.sym->attr.pure; + *name = e->symtree->n.sym->name; + } else { /* Implicit functions are not pure. */ @@ -3253,14 +3260,30 @@ static bool check_pure_function (gfc_expr *e) gfc_do_concurrent_flag = 0 when the check for an impure function occurs. Check the stack to see if the source code has a nested BLOCK construct. */ + for (stack = cs_base; stack; stack = stack->prev) { - if (stack->current->op == EXEC_BLOCK) saw_block = true; + if (!saw_block && stack->current->op == EXEC_BLOCK) + { + saw_block = true; + continue; + } + if (saw_block && stack->current->op == EXEC_DO_CONCURRENT) { - gfc_error ("Reference to impure function at %L inside a " - "DO CONCURRENT", &e->where); - return false; + bool is_pure; + is_pure = (e->value.function.isym + && (e->value.function.isym->pure + || e->value.function.isym->elemental)) + || (e->value.function.esym + && (e->value.function.esym->attr.pure + || e->value.function.esym->attr.elemental)); + if (!is_pure) + { + gfc_error ("Reference to impure function at %L inside a " + "DO CONCURRENT", &e->where); + return false; + } } } @@ -3656,16 +3679,29 @@ pure_subroutine (gfc_symbol *sym, const char *name, locus *loc) /* A BLOCK construct within a DO CONCURRENT construct leads to gfc_do_concurrent_flag = 0 when the check for an impure subroutine - occurs. Check the stack to see if the source code has a nested - BLOCK construct. */ + occurs. Walk up the stack to see if the source code has a nested + construct. */ + for (stack = cs_base; stack; stack = stack->prev) { - if (stack->current->op == EXEC_BLOCK) saw_block = true; + if (stack->current->op == EXEC_BLOCK) + { + saw_block = true; + continue; + } + if (saw_block && stack->current->op == EXEC_DO_CONCURRENT) { - gfc_error ("Subroutine call at %L in a DO CONCURRENT block " - "is not PURE", loc); - return false; + + bool is_pure = true; + is_pure = sym->attr.pure || sym->attr.elemental; + + if (!is_pure) + { + gfc_error ("Subroutine call at %L in a DO CONCURRENT block " + "is not PURE", loc); + return false; + } } } diff --git a/gcc/fortran/trans-decl.cc b/gcc/fortran/trans-decl.cc index aea132d..ddc4960 100644 --- a/gcc/fortran/trans-decl.cc +++ b/gcc/fortran/trans-decl.cc @@ -6920,6 +6920,7 @@ add_clause (gfc_symbol *sym, gfc_omp_map_op map_op) n = gfc_get_omp_namelist (); n->sym = sym; + n->where = sym->declared_at; n->u.map.op = map_op; if (!module_oacc_clauses) diff --git a/gcc/fortran/trans-openmp.cc b/gcc/fortran/trans-openmp.cc index 03d9432..0b8150f 100644 --- a/gcc/fortran/trans-openmp.cc +++ b/gcc/fortran/trans-openmp.cc @@ -25,6 +25,10 @@ along with GCC; see the file COPYING3. If not see #include "options.h" #include "tree.h" #include "gfortran.h" +#include "basic-block.h" +#include "tree-ssa.h" +#include "function.h" +#include "gimple.h" #include "gimple-expr.h" #include "trans.h" #include "stringpool.h" @@ -41,6 +45,8 @@ along with GCC; see the file COPYING3. If not see #include "omp-low.h" #include "memmodel.h" /* For MEMMODEL_ enums. */ #include "dependency.h" +#include "gimple-iterator.h" /* For gsi_iterator_update. */ +#include "gimplify-me.h" /* For force_gimple_operand. */ #undef GCC_DIAG_STYLE #define GCC_DIAG_STYLE __gcc_tdiag__ @@ -375,22 +381,28 @@ gfc_omp_report_decl (tree decl) return decl; } -/* Return true if TYPE has any allocatable components. */ +/* Return true if TYPE has any allocatable components; + if ptr_ok, the decl itself is permitted to have the POINTER attribute. + if shallow_alloc_only, returns only true if any of the fields is an + allocatable; called with true by gfc_omp_replace_alloc_by_to_mapping. */ static bool -gfc_has_alloc_comps (tree type, tree decl) +gfc_has_alloc_comps (tree type, tree decl, bool ptr_ok, + bool shallow_alloc_only=false) { tree field, ftype; if (POINTER_TYPE_P (type)) { - if (GFC_DECL_GET_SCALAR_ALLOCATABLE (decl)) + if (GFC_DECL_GET_SCALAR_ALLOCATABLE (decl) + || (ptr_ok && GFC_DECL_GET_SCALAR_POINTER (decl))) type = TREE_TYPE (type); else if (GFC_DECL_GET_SCALAR_POINTER (decl)) return false; } - if (GFC_DESCRIPTOR_TYPE_P (type) + if (!ptr_ok + && GFC_DESCRIPTOR_TYPE_P (type) && (GFC_TYPE_ARRAY_AKIND (type) == GFC_ARRAY_POINTER || GFC_TYPE_ARRAY_AKIND (type) == GFC_ARRAY_POINTER_CONT)) return false; @@ -409,12 +421,51 @@ gfc_has_alloc_comps (tree type, tree decl) if (GFC_DESCRIPTOR_TYPE_P (ftype) && GFC_TYPE_ARRAY_AKIND (ftype) == GFC_ARRAY_ALLOCATABLE) return true; - if (gfc_has_alloc_comps (ftype, field)) + if (!shallow_alloc_only + && gfc_has_alloc_comps (ftype, field, false)) return true; } return false; } +/* gfc_omp_replace_alloc_by_to_mapping is used with gfc_omp_deep_mapping... to + handle the following: + + For map(alloc: dt), the array descriptors of allocatable components should + be mapped as 'to'; this could be done by (A) adding 'map(to: dt%alloc_comp)' + for each component (and avoiding to increment the reference count). + Or (B) by just mapping all of 'dt' as 'to'. + + If 'dt' contains several allocatable components and not much other data, + (A) is more efficient. If 'dt' contains a large const-size array, (A) will + copy it to the device instead of only 'alloc'ating it. + + IMPLEMENTATION CHOICE: We do (A). It avoids the ref-count issue and it is + expected that, for real-world code, derived types with allocatable + components only have few other components and either no const-size arrays. + This copying is done irrespectively whether the allocatables are allocated. + + If users wanted to save memory, they have to use 'map(alloc:dt%comp)' as + also with 'map(alloc:dt)' all components get copied. + + For the copy to the device, only allocatable arrays are relevant as their + the bounds are required; the pointer is set separately (GOMP_MAP_ATTACH) + and the only setting required for scalars. However, when later copying out + of the device, an unallocated allocatable must remain unallocated/NULL on + the host; to achieve this we also must have it set to NULL on the device + to avoid issues with uninitialized memory being copied back for the pointer + address. If we could set the pointer to NULL, gfc_has_alloc_comps's + shallow_alloc_only could be restricted to return true only for arrays. + + We only need to return true if there are allocatable-array components. */ + +static bool +gfc_omp_replace_alloc_by_to_mapping (tree type, tree decl, bool ptr_ok) +{ + return gfc_has_alloc_comps (type, decl, ptr_ok, true); +} + + /* Return true if TYPE is polymorphic but not with pointer attribute. */ static bool @@ -487,7 +538,7 @@ gfc_omp_private_outer_ref (tree decl) if (GFC_DECL_GET_SCALAR_ALLOCATABLE (decl)) return true; - if (gfc_has_alloc_comps (type, decl)) + if (gfc_has_alloc_comps (type, decl, false)) return true; return false; @@ -627,7 +678,7 @@ gfc_walk_alloc_comps (tree decl, tree dest, tree var, { tree ftype = TREE_TYPE (field); tree declf, destf = NULL_TREE; - bool has_alloc_comps = gfc_has_alloc_comps (ftype, field); + bool has_alloc_comps = gfc_has_alloc_comps (ftype, field, false); if ((!GFC_DESCRIPTOR_TYPE_P (ftype) || GFC_TYPE_ARRAY_AKIND (ftype) != GFC_ARRAY_ALLOCATABLE) && !GFC_DECL_GET_SCALAR_ALLOCATABLE (field) @@ -751,7 +802,7 @@ gfc_omp_clause_default_ctor (tree clause, tree decl, tree outer) && (!GFC_DECL_GET_SCALAR_ALLOCATABLE (OMP_CLAUSE_DECL (clause)) || !POINTER_TYPE_P (type))) { - if (gfc_has_alloc_comps (type, OMP_CLAUSE_DECL (clause))) + if (gfc_has_alloc_comps (type, OMP_CLAUSE_DECL (clause), false)) { gcc_assert (outer); gfc_start_block (&block); @@ -804,7 +855,7 @@ gfc_omp_clause_default_ctor (tree clause, tree decl, tree outer) else gfc_add_modify (&cond_block, unshare_expr (decl), fold_convert (TREE_TYPE (decl), ptr)); - if (gfc_has_alloc_comps (type, OMP_CLAUSE_DECL (clause))) + if (gfc_has_alloc_comps (type, OMP_CLAUSE_DECL (clause), false)) { tree tem = gfc_walk_alloc_comps (outer, decl, OMP_CLAUSE_DECL (clause), @@ -945,7 +996,7 @@ gfc_omp_clause_copy_ctor (tree clause, tree dest, tree src) && (!GFC_DECL_GET_SCALAR_ALLOCATABLE (OMP_CLAUSE_DECL (clause)) || !POINTER_TYPE_P (type))) { - if (gfc_has_alloc_comps (type, OMP_CLAUSE_DECL (clause))) + if (gfc_has_alloc_comps (type, OMP_CLAUSE_DECL (clause), false)) { gfc_start_block (&block); gfc_add_modify (&block, dest, src); @@ -1004,7 +1055,7 @@ gfc_omp_clause_copy_ctor (tree clause, tree dest, tree src) builtin_decl_explicit (BUILT_IN_MEMCPY), 3, ptr, srcptr, size); gfc_add_expr_to_block (&cond_block, fold_convert (void_type_node, call)); - if (gfc_has_alloc_comps (type, OMP_CLAUSE_DECL (clause))) + if (gfc_has_alloc_comps (type, OMP_CLAUSE_DECL (clause), false)) { tree tem = gfc_walk_alloc_comps (src, dest, OMP_CLAUSE_DECL (clause), @@ -1049,7 +1100,7 @@ gfc_omp_clause_assign_op (tree clause, tree dest, tree src) && (!GFC_DECL_GET_SCALAR_ALLOCATABLE (OMP_CLAUSE_DECL (clause)) || !POINTER_TYPE_P (type))) { - if (gfc_has_alloc_comps (type, OMP_CLAUSE_DECL (clause))) + if (gfc_has_alloc_comps (type, OMP_CLAUSE_DECL (clause), false)) { gfc_start_block (&block); /* First dealloc any allocatable components in DEST. */ @@ -1071,7 +1122,7 @@ gfc_omp_clause_assign_op (tree clause, tree dest, tree src) gfc_start_block (&block); - if (gfc_has_alloc_comps (type, OMP_CLAUSE_DECL (clause))) + if (gfc_has_alloc_comps (type, OMP_CLAUSE_DECL (clause), false)) { then_b = gfc_walk_alloc_comps (dest, NULL_TREE, OMP_CLAUSE_DECL (clause), WALK_ALLOC_COMPS_DTOR); @@ -1186,7 +1237,7 @@ gfc_omp_clause_assign_op (tree clause, tree dest, tree src) builtin_decl_explicit (BUILT_IN_MEMCPY), 3, ptr, srcptr, size); gfc_add_expr_to_block (&cond_block, fold_convert (void_type_node, call)); - if (gfc_has_alloc_comps (type, OMP_CLAUSE_DECL (clause))) + if (gfc_has_alloc_comps (type, OMP_CLAUSE_DECL (clause), false)) { tree tem = gfc_walk_alloc_comps (src, dest, OMP_CLAUSE_DECL (clause), @@ -1438,7 +1489,7 @@ gfc_omp_clause_dtor (tree clause, tree decl) && (!GFC_DECL_GET_SCALAR_ALLOCATABLE (OMP_CLAUSE_DECL (clause)) || !POINTER_TYPE_P (type))) { - if (gfc_has_alloc_comps (type, OMP_CLAUSE_DECL (clause))) + if (gfc_has_alloc_comps (type, OMP_CLAUSE_DECL (clause), false)) return gfc_walk_alloc_comps (decl, NULL_TREE, OMP_CLAUSE_DECL (clause), WALK_ALLOC_COMPS_DTOR); @@ -1458,7 +1509,7 @@ gfc_omp_clause_dtor (tree clause, tree decl) tem = gfc_call_free (decl); tem = gfc_omp_unshare_expr (tem); - if (gfc_has_alloc_comps (type, OMP_CLAUSE_DECL (clause))) + if (gfc_has_alloc_comps (type, OMP_CLAUSE_DECL (clause), false)) { stmtblock_t block; tree then_b; @@ -1538,6 +1589,7 @@ gfc_omp_finish_clause (tree c, gimple_seq *pre_p, bool openacc) return; tree decl = OMP_CLAUSE_DECL (c); + location_t loc = OMP_CLAUSE_LOCATION (c); /* Assumed-size arrays can't be mapped implicitly, they have to be mapped explicitly using array sections. */ @@ -1553,13 +1605,9 @@ gfc_omp_finish_clause (tree c, gimple_seq *pre_p, bool openacc) return; } - if (!openacc && GFC_CLASS_TYPE_P (TREE_TYPE (decl))) - warning_at (OMP_CLAUSE_LOCATION (c), OPT_Wopenmp, - "Implicit mapping of polymorphic variable %qD is " - "unspecified behavior", decl); - tree c2 = NULL_TREE, c3 = NULL_TREE, c4 = NULL_TREE; tree present = gfc_omp_check_optional_argument (decl, true); + tree orig_decl = NULL_TREE; if (POINTER_TYPE_P (TREE_TYPE (decl))) { if (!gfc_omp_privatize_by_reference (decl) @@ -1568,7 +1616,7 @@ gfc_omp_finish_clause (tree c, gimple_seq *pre_p, bool openacc) && !GFC_DECL_CRAY_POINTEE (decl) && !GFC_DESCRIPTOR_TYPE_P (TREE_TYPE (TREE_TYPE (decl)))) return; - tree orig_decl = decl; + orig_decl = decl; c4 = build_omp_clause (OMP_CLAUSE_LOCATION (c), OMP_CLAUSE_MAP); OMP_CLAUSE_SET_MAP_KIND (c4, GOMP_MAP_POINTER); @@ -1579,16 +1627,16 @@ gfc_omp_finish_clause (tree c, gimple_seq *pre_p, bool openacc) && (GFC_DECL_GET_SCALAR_POINTER (orig_decl) || GFC_DECL_GET_SCALAR_ALLOCATABLE (orig_decl))) { - c2 = build_omp_clause (input_location, OMP_CLAUSE_MAP); + c2 = build_omp_clause (loc, OMP_CLAUSE_MAP); OMP_CLAUSE_SET_MAP_KIND (c2, GOMP_MAP_POINTER); - OMP_CLAUSE_DECL (c2) = decl; + OMP_CLAUSE_DECL (c2) = unshare_expr (decl); OMP_CLAUSE_SIZE (c2) = size_int (0); stmtblock_t block; gfc_start_block (&block); - tree ptr = decl; - ptr = gfc_build_cond_assign_expr (&block, present, decl, - null_pointer_node); + tree ptr = gfc_build_cond_assign_expr (&block, present, + unshare_expr (decl), + null_pointer_node); gimplify_and_add (gfc_finish_block (&block), pre_p); ptr = build_fold_indirect_ref (ptr); OMP_CLAUSE_DECL (c) = ptr; @@ -1605,10 +1653,10 @@ gfc_omp_finish_clause (tree c, gimple_seq *pre_p, bool openacc) { c3 = build_omp_clause (OMP_CLAUSE_LOCATION (c), OMP_CLAUSE_MAP); OMP_CLAUSE_SET_MAP_KIND (c3, GOMP_MAP_POINTER); - OMP_CLAUSE_DECL (c3) = unshare_expr (decl); + OMP_CLAUSE_DECL (c3) = decl; OMP_CLAUSE_SIZE (c3) = size_int (0); decl = build_fold_indirect_ref (decl); - OMP_CLAUSE_DECL (c) = decl; + OMP_CLAUSE_DECL (c) = unshare_expr (decl); } } if (GFC_DESCRIPTOR_TYPE_P (TREE_TYPE (decl))) @@ -1634,7 +1682,7 @@ gfc_omp_finish_clause (tree c, gimple_seq *pre_p, bool openacc) gcc_assert (POINTER_TYPE_P (TREE_TYPE (ptr))); ptr = build_fold_indirect_ref (ptr); OMP_CLAUSE_DECL (c) = ptr; - c2 = build_omp_clause (input_location, OMP_CLAUSE_MAP); + c2 = build_omp_clause (loc, OMP_CLAUSE_MAP); OMP_CLAUSE_SET_MAP_KIND (c2, GOMP_MAP_TO_PSET); if (present) { @@ -1651,7 +1699,7 @@ gfc_omp_finish_clause (tree c, gimple_seq *pre_p, bool openacc) : GOMP_MAP_POINTER); if (present) { - ptr = gfc_conv_descriptor_data_get (decl); + ptr = gfc_conv_descriptor_data_get (unshare_expr (decl)); ptr = gfc_build_addr_expr (NULL, ptr); ptr = gfc_build_cond_assign_expr (&block, present, ptr, null_pointer_node); @@ -1664,6 +1712,17 @@ gfc_omp_finish_clause (tree c, gimple_seq *pre_p, bool openacc) tree size = create_tmp_var (gfc_array_index_type); tree elemsz = TYPE_SIZE_UNIT (gfc_get_element_type (type)); elemsz = fold_convert (gfc_array_index_type, elemsz); + + if (orig_decl == NULL_TREE) + orig_decl = decl; + if (!openacc + && gfc_has_alloc_comps (type, orig_decl, true)) + { + /* Save array descriptor for use in gfc_omp_deep_mapping{,_p,_cnt}; + force evaluate to ensure that it is not gimplified + is a decl. */ + gfc_allocate_lang_decl (size); + GFC_DECL_SAVED_DESCRIPTOR (size) = orig_decl; + } enum gfc_array_kind akind = GFC_TYPE_ARRAY_AKIND (type); if (akind == GFC_ARRAY_ALLOCATABLE || akind == GFC_ARRAY_POINTER @@ -1692,14 +1751,14 @@ gfc_omp_finish_clause (tree c, gimple_seq *pre_p, bool openacc) else_b = gfc_finish_block (&cond_block); tem = gfc_conv_descriptor_data_get (unshare_expr (decl)); tem = fold_convert (pvoid_type_node, tem); - cond = fold_build2_loc (input_location, NE_EXPR, + cond = fold_build2_loc (loc, NE_EXPR, boolean_type_node, tem, null_pointer_node); if (present) { - cond = fold_build2_loc (input_location, TRUTH_ANDIF_EXPR, + cond = fold_build2_loc (loc, TRUTH_ANDIF_EXPR, boolean_type_node, present, cond); } - gfc_add_expr_to_block (&block, build3_loc (input_location, COND_EXPR, + gfc_add_expr_to_block (&block, build3_loc (loc, COND_EXPR, void_type_node, cond, then_b, else_b)); } @@ -1739,11 +1798,30 @@ gfc_omp_finish_clause (tree c, gimple_seq *pre_p, bool openacc) tree stmt = gfc_finish_block (&block); gimplify_and_add (stmt, pre_p); } + else + { + if (OMP_CLAUSE_SIZE (c) == NULL_TREE) + OMP_CLAUSE_SIZE (c) + = DECL_P (decl) ? DECL_SIZE_UNIT (decl) + : TYPE_SIZE_UNIT (TREE_TYPE (decl)); + + tree type = TREE_TYPE (decl); + if (POINTER_TYPE_P (type) && POINTER_TYPE_P (TREE_TYPE (type))) + type = TREE_TYPE (type); + if (!openacc + && orig_decl != NULL_TREE + && gfc_has_alloc_comps (type, orig_decl, true)) + { + /* Save array descriptor for use in gfc_omp_deep_mapping{,_p,_cnt}; + force evaluate to ensure that it is not gimplified + is a decl. */ + tree size = create_tmp_var (TREE_TYPE (OMP_CLAUSE_SIZE (c))); + gfc_allocate_lang_decl (size); + GFC_DECL_SAVED_DESCRIPTOR (size) = orig_decl; + gimplify_assign (size, OMP_CLAUSE_SIZE (c), pre_p); + OMP_CLAUSE_SIZE (c) = size; + } + } tree last = c; - if (OMP_CLAUSE_SIZE (c) == NULL_TREE) - OMP_CLAUSE_SIZE (c) - = DECL_P (decl) ? DECL_SIZE_UNIT (decl) - : TYPE_SIZE_UNIT (TREE_TYPE (decl)); if (gimplify_expr (&OMP_CLAUSE_SIZE (c), pre_p, NULL, is_gimple_val, fb_rvalue) == GS_ERROR) OMP_CLAUSE_SIZE (c) = size_int (0); @@ -1767,6 +1845,715 @@ gfc_omp_finish_clause (tree c, gimple_seq *pre_p, bool openacc) } +/* map(<flag>: data [len: <size>]) + map(attach: &data [bias: <bias>]) + offset += 2; offset_data += 2 */ +static void +gfc_omp_deep_mapping_map (tree data, tree size, unsigned HOST_WIDE_INT tkind, + location_t loc, tree data_array, tree sizes_array, + tree kinds_array, tree offset_data, tree offset, + gimple_seq *seq, const gimple *ctx) +{ + tree one = build_int_cst (size_type_node, 1); + + STRIP_NOPS (data); + if (!POINTER_TYPE_P (TREE_TYPE (data))) + { + gcc_assert (TREE_CODE (data) == INDIRECT_REF); + data = TREE_OPERAND (data, 0); + } + + /* data_array[offset_data] = data; */ + tree tmp = build4 (ARRAY_REF, TREE_TYPE (TREE_TYPE (data_array)), + unshare_expr (data_array), offset_data, + NULL_TREE, NULL_TREE); + gimplify_assign (tmp, data, seq); + + /* offset_data++ */ + tmp = build2_loc (loc, PLUS_EXPR, size_type_node, offset_data, one); + gimplify_assign (offset_data, tmp, seq); + + /* data_array[offset_data] = &data; */ + tmp = build4 (ARRAY_REF, TREE_TYPE (TREE_TYPE (data_array)), + unshare_expr (data_array), + offset_data, NULL_TREE, NULL_TREE); + gimplify_assign (tmp, build_fold_addr_expr (data), seq); + + /* offset_data++ */ + tmp = build2_loc (loc, PLUS_EXPR, size_type_node, offset_data, one); + gimplify_assign (offset_data, tmp, seq); + + /* sizes_array[offset] = size */ + tmp = build2_loc (loc, MULT_EXPR, size_type_node, + TYPE_SIZE_UNIT (size_type_node), offset); + tmp = build2_loc (loc, POINTER_PLUS_EXPR, TREE_TYPE (sizes_array), + sizes_array, tmp); + gimple_seq seq2 = NULL; + tmp = force_gimple_operand (tmp, &seq2, true, NULL_TREE); + gimple_seq_add_seq (seq, seq2); + tmp = build_fold_indirect_ref_loc (loc, tmp); + gimplify_assign (tmp, size, seq); + + /* FIXME: tkind |= talign << talign_shift; */ + /* kinds_array[offset] = tkind. */ + tmp = build2_loc (loc, MULT_EXPR, size_type_node, + TYPE_SIZE_UNIT (short_unsigned_type_node), offset); + tmp = build2_loc (loc, POINTER_PLUS_EXPR, TREE_TYPE (kinds_array), + kinds_array, tmp); + seq2 = NULL; + tmp = force_gimple_operand (tmp, &seq2, true, NULL_TREE); + gimple_seq_add_seq (seq, seq2); + tmp = build_fold_indirect_ref_loc (loc, tmp); + gimplify_assign (tmp, build_int_cst (short_unsigned_type_node, tkind), seq); + + /* offset++ */ + tmp = build2_loc (loc, PLUS_EXPR, size_type_node, offset, one); + gimplify_assign (offset, tmp, seq); + + /* sizes_array[offset] = bias (= 0). */ + tmp = build2_loc (loc, MULT_EXPR, size_type_node, + TYPE_SIZE_UNIT (size_type_node), offset); + tmp = build2_loc (loc, POINTER_PLUS_EXPR, TREE_TYPE (sizes_array), + sizes_array, tmp); + seq2 = NULL; + tmp = force_gimple_operand (tmp, &seq2, true, NULL_TREE); + gimple_seq_add_seq (seq, seq2); + tmp = build_fold_indirect_ref_loc (loc, tmp); + gimplify_assign (tmp, build_zero_cst (size_type_node), seq); + + gcc_assert (gimple_code (ctx) == GIMPLE_OMP_TARGET); + tkind = (gimple_omp_target_kind (ctx) == GF_OMP_TARGET_KIND_EXIT_DATA + ? GOMP_MAP_DETACH : GOMP_MAP_ATTACH); + + /* kinds_array[offset] = tkind. */ + tmp = build2_loc (loc, MULT_EXPR, size_type_node, + TYPE_SIZE_UNIT (short_unsigned_type_node), offset); + tmp = build2_loc (loc, POINTER_PLUS_EXPR, TREE_TYPE (kinds_array), + kinds_array, tmp); + seq2 = NULL; + tmp = force_gimple_operand (tmp, &seq2, true, NULL_TREE); + gimple_seq_add_seq (seq, seq2); + tmp = build_fold_indirect_ref_loc (loc, tmp); + gimplify_assign (tmp, build_int_cst (short_unsigned_type_node, tkind), seq); + + /* offset++ */ + tmp = build2_loc (loc, PLUS_EXPR, size_type_node, offset, one); + gimplify_assign (offset, tmp, seq); +} + +static void gfc_omp_deep_mapping_item (bool, bool, bool, location_t, tree, + tree *, unsigned HOST_WIDE_INT, tree, + tree, tree, tree, tree, tree, + gimple_seq *, const gimple *, bool *); + +/* Map allocatable components. */ +static void +gfc_omp_deep_mapping_comps (bool is_cnt, location_t loc, tree decl, + tree *token, unsigned HOST_WIDE_INT tkind, + tree data_array, tree sizes_array, tree kinds_array, + tree offset_data, tree offset, tree num, + gimple_seq *seq, const gimple *ctx, + bool *poly_warned) +{ + tree type = TREE_TYPE (decl); + if (TREE_CODE (type) != RECORD_TYPE) + return; + for (tree field = TYPE_FIELDS (type); field; field = DECL_CHAIN (field)) + { + type = TREE_TYPE (field); + if (gfc_is_polymorphic_nonptr (type) + || GFC_DECL_GET_SCALAR_ALLOCATABLE (field) + || (GFC_DESCRIPTOR_TYPE_P (type) + && GFC_TYPE_ARRAY_AKIND (type) == GFC_ARRAY_ALLOCATABLE)) + { + tree tmp = fold_build3_loc (loc, COMPONENT_REF, TREE_TYPE (field), + decl, field, NULL_TREE); + gfc_omp_deep_mapping_item (is_cnt, true, true, loc, tmp, token, + tkind, data_array, sizes_array, + kinds_array, offset_data, offset, num, + seq, ctx, poly_warned); + } + else if (GFC_DECL_GET_SCALAR_POINTER (field) + || GFC_DESCRIPTOR_TYPE_P (type)) + continue; + else if (gfc_has_alloc_comps (TREE_TYPE (field), field, false)) + { + tree tmp = fold_build3_loc (loc, COMPONENT_REF, TREE_TYPE (field), + decl, field, NULL_TREE); + if (TREE_CODE (TREE_TYPE (tmp)) == ARRAY_TYPE) + gfc_omp_deep_mapping_item (is_cnt, false, false, loc, tmp, + token, tkind, data_array, sizes_array, + kinds_array, offset_data, offset, num, + seq, ctx, poly_warned); + else + gfc_omp_deep_mapping_comps (is_cnt, loc, tmp, token, tkind, + data_array, sizes_array, kinds_array, + offset_data, offset, num, seq, ctx, + poly_warned); + } + } +} + +static void +gfc_omp_gen_simple_loop (tree var, tree begin, tree end, enum tree_code cond, + tree step, location_t loc, gimple_seq *seq1, + gimple_seq *seq2) +{ + tree tmp; + + /* var = begin. */ + gimplify_assign (var, begin, seq1); + + /* Loop: for (var = begin; var <cond> end; var += step). */ + tree label_loop = create_artificial_label (loc); + tree label_cond = create_artificial_label (loc); + + gimplify_and_add (fold_build1_loc (loc, GOTO_EXPR, void_type_node, + label_cond), seq1); + gimple_seq_add_stmt (seq1, gimple_build_label (label_loop)); + + /* Everything above is seq1; place loop body here. */ + + /* End of loop body -> put into seq2. */ + tmp = fold_build2_loc (loc, PLUS_EXPR, TREE_TYPE (var), var, step); + gimplify_assign (var, tmp, seq2); + gimple_seq_add_stmt (seq2, gimple_build_label (label_cond)); + tmp = fold_build2_loc (loc, cond, boolean_type_node, var, end); + tmp = build3_v (COND_EXPR, tmp, build1_v (GOTO_EXPR, label_loop), + build_empty_stmt (loc)); + gimplify_and_add (tmp, seq2); +} + +/* Return size variable with the size of an array. */ +static tree +gfc_omp_get_array_size (location_t loc, tree desc, gimple_seq *seq) +{ + tree tmp; + gimple_seq seq1 = NULL, seq2 = NULL; + tree size = build_decl (loc, VAR_DECL, create_tmp_var_name ("size"), + size_type_node); + tree extent = build_decl (loc, VAR_DECL, create_tmp_var_name ("extent"), + gfc_array_index_type); + tree idx = build_decl (loc, VAR_DECL, create_tmp_var_name ("idx"), + signed_char_type_node); + + tree begin = build_zero_cst (signed_char_type_node); + tree end; + if (GFC_TYPE_ARRAY_AKIND (TREE_TYPE (desc)) == GFC_ARRAY_ASSUMED_SHAPE_CONT + || GFC_TYPE_ARRAY_AKIND (TREE_TYPE (desc)) == GFC_ARRAY_ASSUMED_SHAPE) + end = gfc_conv_descriptor_rank (desc); + else + end = build_int_cst (signed_char_type_node, + GFC_TYPE_ARRAY_RANK (TREE_TYPE (desc))); + tree step = build_int_cst (signed_char_type_node, 1); + + /* size = 0 + for (idx = 0; idx < rank; idx++) + extent = gfc->dim[i].ubound - gfc->dim[i].lbound + 1 + if (extent < 0) extent = 0 + size *= extent. */ + gimplify_assign (size, build_int_cst (size_type_node, 1), seq); + + gfc_omp_gen_simple_loop (idx, begin, end, LT_EXPR, step, loc, &seq1, &seq2); + gimple_seq_add_seq (seq, seq1); + + tmp = fold_build2_loc (loc, MINUS_EXPR, gfc_array_index_type, + gfc_conv_descriptor_ubound_get (desc, idx), + gfc_conv_descriptor_lbound_get (desc, idx)); + tmp = fold_build2_loc (loc, PLUS_EXPR, gfc_array_index_type, + tmp, gfc_index_one_node); + gimplify_assign (extent, tmp, seq); + tmp = fold_build2_loc (loc, LT_EXPR, boolean_type_node, + extent, gfc_index_zero_node); + tmp = build3_v (COND_EXPR, tmp, + fold_build2_loc (loc, MODIFY_EXPR, + gfc_array_index_type, + extent, gfc_index_zero_node), + build_empty_stmt (loc)); + gimplify_and_add (tmp, seq); + /* size *= extent. */ + gimplify_assign (size, fold_build2_loc (loc, MULT_EXPR, size_type_node, size, + fold_convert (size_type_node, + extent)), seq); + gimple_seq_add_seq (seq, seq2); + return size; +} + +/* Generate loop to access every array element; takes addr of first element + (decl's data comp); returns loop code in seq1 + seq2 + and the pointer to the element as return value. */ +static tree +gfc_omp_elmental_loop (location_t loc, tree decl, tree size, tree elem_len, + gimple_seq *seq1, gimple_seq *seq2) +{ + tree idx = build_decl (loc, VAR_DECL, create_tmp_var_name ("idx"), + size_type_node); + tree begin = build_zero_cst (size_type_node); + tree end = size; + tree step = build_int_cst (size_type_node, 1); + tree ptr; + + gfc_omp_gen_simple_loop (idx, begin, end, LT_EXPR, step, loc, seq1, seq2); + + tree type = TREE_TYPE (decl); + if (POINTER_TYPE_P (type)) + { + type = TREE_TYPE (type); + gcc_assert (TREE_CODE (type) == ARRAY_TYPE); + decl = fold_convert (build_pointer_type (TREE_TYPE (type)), decl); + } + else + { + gcc_assert (TREE_CODE (type) == ARRAY_TYPE); + decl = build_fold_addr_expr_loc (loc, decl); + } + decl = fold_convert (build_pointer_type (TREE_TYPE (type)), decl); + tree tmp = build2_loc (loc, MULT_EXPR, size_type_node, idx, + fold_convert (size_type_node, elem_len)); + ptr = build2_loc (loc, POINTER_PLUS_EXPR, TREE_TYPE (decl), decl, tmp); + gimple_seq seq3 = NULL; + ptr = force_gimple_operand (ptr, &seq3, true, NULL_TREE); + gimple_seq_add_seq (seq1, seq3); + + return ptr; +} + + +/* If do_copy, copy data pointer and vptr (if applicable) as well. + Otherwise, only handle allocatable components. + do_copy == false can happen only with nonpolymorphic arguments + to a copy clause. + if (is_cnt) token ... offset is ignored and num is used, otherwise + num is NULL_TREE and unused. */ + +static void +gfc_omp_deep_mapping_item (bool is_cnt, bool do_copy, bool do_alloc_check, + location_t loc, tree decl, tree *token, + unsigned HOST_WIDE_INT tkind, tree data_array, + tree sizes_array, tree kinds_array, tree offset_data, + tree offset, tree num, gimple_seq *seq, + const gimple *ctx, bool *poly_warned) +{ + tree tmp; + tree type = TREE_TYPE (decl); + if (POINTER_TYPE_P (type)) + type = TREE_TYPE (type); + tree end_label = NULL_TREE; + tree size = NULL_TREE, elem_len = NULL_TREE; + + bool poly = gfc_is_polymorphic_nonptr (type); + if (poly && is_cnt && !*poly_warned) + { + if (gfc_is_unlimited_polymorphic_nonptr (type)) + error_at (loc, + "Mapping of unlimited polymorphic list item %qD is " + "unspecified behavior and unsupported", decl); + + else + warning_at (loc, OPT_Wopenmp, + "Mapping of polymorphic list item %qD is " + "unspecified behavior", decl); + *poly_warned = true; + } + if (do_alloc_check) + { + tree then_label = create_artificial_label (loc); + end_label = create_artificial_label (loc); + tmp = decl; + if (TREE_CODE (TREE_TYPE (tmp)) == REFERENCE_TYPE + || (POINTER_TYPE_P (TREE_TYPE (tmp)) + && (POINTER_TYPE_P (TREE_TYPE (TREE_TYPE (tmp))) + || GFC_DESCRIPTOR_TYPE_P (TREE_TYPE (TREE_TYPE (tmp)))))) + tmp = build_fold_indirect_ref_loc (loc, tmp); + if (poly) + tmp = gfc_class_data_get (tmp); + if (GFC_DESCRIPTOR_TYPE_P (TREE_TYPE (tmp))) + tmp = gfc_conv_descriptor_data_get (tmp); + gimple_seq seq2 = NULL; + tmp = force_gimple_operand (tmp, &seq2, true, NULL_TREE); + gimple_seq_add_seq (seq, seq2); + + gimple_seq_add_stmt (seq, + gimple_build_cond (NE_EXPR, tmp, null_pointer_node, + then_label, end_label)); + gimple_seq_add_stmt (seq, gimple_build_label (then_label)); + } + tree class_decl = decl; + if (poly) + { + decl = gfc_class_data_get (decl); + type = TREE_TYPE (decl); + } + if (POINTER_TYPE_P (TREE_TYPE (decl))) + { + decl = build_fold_indirect_ref (decl); + type = TREE_TYPE (decl); + } + + if (is_cnt && do_copy) + { + tree tmp = fold_build2_loc (loc, PLUS_EXPR, size_type_node, + num, build_int_cst (size_type_node, 1)); + gimplify_assign (num, tmp, seq); + } + else if (do_copy) + { + /* copy data pointer */ + tree bytesize; + if (GFC_DESCRIPTOR_TYPE_P (TREE_TYPE (decl))) + { + /* TODO: Optimization: Shouldn't this be an expr. const, except for + deferred-length strings. (Cf. also below). */ + elem_len = (poly ? gfc_class_vtab_size_get (class_decl) + : gfc_conv_descriptor_elem_len (decl)); + tmp = (POINTER_TYPE_P (TREE_TYPE (decl)) + ? build_fold_indirect_ref (decl) : decl); + size = gfc_omp_get_array_size (loc, tmp, seq); + bytesize = fold_build2_loc (loc, MULT_EXPR, size_type_node, + fold_convert (size_type_node, size), + fold_convert (size_type_node, elem_len)); + tmp = gfc_conv_descriptor_data_get (decl); + } + else if (poly) + { + tmp = decl; + bytesize = fold_convert (size_type_node, + gfc_class_vtab_size_get (class_decl)); + } + else + { + tmp = decl; + bytesize = TYPE_SIZE_UNIT (TREE_TYPE (decl)); + } + unsigned HOST_WIDE_INT tkind2 = tkind; + if (!is_cnt + && (tkind == GOMP_MAP_ALLOC + || (tkind == GOMP_MAP_FROM + && (gimple_omp_target_kind (ctx) + != GF_OMP_TARGET_KIND_EXIT_DATA))) + && gfc_omp_replace_alloc_by_to_mapping (TREE_TYPE (decl), decl, true)) + tkind2 = tkind == GOMP_MAP_ALLOC ? GOMP_MAP_TO : GOMP_MAP_TOFROM; + + gfc_omp_deep_mapping_map (tmp, bytesize, tkind2, loc, data_array, + sizes_array, kinds_array, offset_data, + offset, seq, ctx); + } + + tmp = decl; + if (POINTER_TYPE_P (TREE_TYPE (decl))) + while (TREE_CODE (tmp) == COMPONENT_REF || TREE_CODE (tmp) == ARRAY_REF) + tmp = TREE_OPERAND (tmp, TREE_CODE (tmp) == COMPONENT_REF ? 1 : 0); + if (poly || gfc_has_alloc_comps (type, tmp, true)) + { + gimple_seq seq2 = NULL; + if (GFC_DESCRIPTOR_TYPE_P (TREE_TYPE (decl))) + { + if (elem_len == NULL_TREE) + { + elem_len = gfc_conv_descriptor_elem_len (decl); + size = fold_convert (size_type_node, + gfc_omp_get_array_size (loc, decl, seq)); + } + decl = gfc_conv_descriptor_data_get (decl); + decl = gfc_omp_elmental_loop (loc, decl, size, elem_len, seq, &seq2); + decl = build_fold_indirect_ref_loc (loc, decl); + } + else if (TREE_CODE (TREE_TYPE (tmp)) == ARRAY_TYPE) + { + type = TREE_TYPE (tmp); + /* FIXME: PR95868 - for var%str of deferred length, elem_len == 0; + len is stored as var%_str_length, but not in GFC_DECL_STRING_LEN + nor in TYPE_SIZE_UNIT as expression. */ + elem_len = TYPE_SIZE_UNIT (TREE_TYPE (type)); + size = fold_convert (size_type_node, GFC_TYPE_ARRAY_SIZE (type)); + decl = gfc_omp_elmental_loop (loc, decl, size, elem_len, seq, &seq2); + decl = build_fold_indirect_ref_loc (loc, decl); + } + else if (POINTER_TYPE_P (TREE_TYPE (decl))) + decl = build_fold_indirect_ref (decl); + + gfc_omp_deep_mapping_comps (is_cnt, loc, decl, token, tkind, + data_array, sizes_array, kinds_array, + offset_data, offset, num, seq, ctx, + poly_warned); + gimple_seq_add_seq (seq, seq2); + } + if (end_label) + gimple_seq_add_stmt (seq, gimple_build_label (end_label)); +} + + +/* Which map types to check/handle for deep mapping. */ +static bool +gfc_omp_deep_map_kind_p (tree clause) +{ + switch (OMP_CLAUSE_CODE (clause)) + { + case OMP_CLAUSE_MAP: + break; + case OMP_CLAUSE_FIRSTPRIVATE: + case OMP_CLAUSE_TO: + case OMP_CLAUSE_FROM: + return true; + default: + gcc_unreachable (); + } + + switch (OMP_CLAUSE_MAP_KIND (clause)) + { + case GOMP_MAP_TO: + case GOMP_MAP_FROM: + case GOMP_MAP_TOFROM: + case GOMP_MAP_ALWAYS_TO: + case GOMP_MAP_ALWAYS_FROM: + case GOMP_MAP_ALWAYS_TOFROM: + case GOMP_MAP_ALWAYS_PRESENT_FROM: + case GOMP_MAP_ALWAYS_PRESENT_TO: + case GOMP_MAP_ALWAYS_PRESENT_TOFROM: + case GOMP_MAP_FIRSTPRIVATE: + case GOMP_MAP_ALLOC: + return true; + case GOMP_MAP_POINTER: + case GOMP_MAP_TO_PSET: + case GOMP_MAP_FORCE_PRESENT: + case GOMP_MAP_DELETE: + case GOMP_MAP_FORCE_DEVICEPTR: + case GOMP_MAP_DEVICE_RESIDENT: + case GOMP_MAP_LINK: + case GOMP_MAP_IF_PRESENT: + case GOMP_MAP_PRESENT_ALLOC: + case GOMP_MAP_PRESENT_FROM: + case GOMP_MAP_PRESENT_TO: + case GOMP_MAP_PRESENT_TOFROM: + case GOMP_MAP_FIRSTPRIVATE_INT: + case GOMP_MAP_USE_DEVICE_PTR: + case GOMP_MAP_ZERO_LEN_ARRAY_SECTION: + case GOMP_MAP_FORCE_ALLOC: + case GOMP_MAP_FORCE_TO: + case GOMP_MAP_FORCE_FROM: + case GOMP_MAP_FORCE_TOFROM: + case GOMP_MAP_USE_DEVICE_PTR_IF_PRESENT: + case GOMP_MAP_STRUCT: + case GOMP_MAP_STRUCT_UNORD: + case GOMP_MAP_ALWAYS_POINTER: + case GOMP_MAP_POINTER_TO_ZERO_LENGTH_ARRAY_SECTION: + case GOMP_MAP_DELETE_ZERO_LEN_ARRAY_SECTION: + case GOMP_MAP_RELEASE: + case GOMP_MAP_ATTACH: + case GOMP_MAP_DETACH: + case GOMP_MAP_FORCE_DETACH: + case GOMP_MAP_ATTACH_ZERO_LENGTH_ARRAY_SECTION: + case GOMP_MAP_FIRSTPRIVATE_POINTER: + case GOMP_MAP_FIRSTPRIVATE_REFERENCE: + case GOMP_MAP_ATTACH_DETACH: + break; + default: + gcc_unreachable (); + } + return false; +} + +/* Three OpenMP deep-mapping lang hooks: gfc_omp_deep_mapping{_p,_cnt,}. */ + +/* Common check for gfc_omp_deep_mapping_p and gfc_omp_deep_mapping_do. */ + +static tree +gfc_omp_deep_mapping_int_p (const gimple *ctx, tree clause) +{ + if (is_gimple_omp_oacc (ctx) || !gfc_omp_deep_map_kind_p (clause)) + return NULL_TREE; + tree decl = OMP_CLAUSE_DECL (clause); + if (OMP_CLAUSE_SIZE (clause) != NULL_TREE + && DECL_P (OMP_CLAUSE_SIZE (clause)) + && DECL_LANG_SPECIFIC (OMP_CLAUSE_SIZE (clause)) + && GFC_DECL_SAVED_DESCRIPTOR (OMP_CLAUSE_SIZE (clause))) + /* Saved decl. */ + decl = GFC_DECL_SAVED_DESCRIPTOR (OMP_CLAUSE_SIZE (clause)); + else if (TREE_CODE (decl) == MEM_REF || TREE_CODE (decl) == INDIRECT_REF) + /* The following can happen for, e.g., class(t) :: var(..) */ + decl = TREE_OPERAND (decl, 0); + if (TREE_CODE (decl) == INDIRECT_REF) + /* The following can happen for, e.g., class(t) :: var(..) */ + decl = TREE_OPERAND (decl, 0); + if (DECL_P (decl) + && DECL_LANG_SPECIFIC (decl) + && GFC_DECL_SAVED_DESCRIPTOR (decl)) + decl = GFC_DECL_SAVED_DESCRIPTOR (decl); + /* Handle map(to: var.desc) map([to/from/tofrom:] var.desc.data) + to get proper map kind by skipping to the next item. */ + tree tmp = OMP_CLAUSE_CHAIN (clause); + if (tmp != NULL_TREE + && OMP_CLAUSE_CODE (tmp) == OMP_CLAUSE_CODE (clause) + && OMP_CLAUSE_SIZE (tmp) != NULL_TREE + && DECL_P (OMP_CLAUSE_SIZE (tmp)) + && DECL_LANG_SPECIFIC (OMP_CLAUSE_SIZE (tmp)) + && GFC_DECL_SAVED_DESCRIPTOR (OMP_CLAUSE_SIZE (tmp)) == decl) + return NULL_TREE; + if (DECL_P (decl) + && DECL_LANG_SPECIFIC (decl) + && GFC_DECL_SAVED_DESCRIPTOR (decl)) + decl = GFC_DECL_SAVED_DESCRIPTOR (decl); + tree type = TREE_TYPE (decl); + if (POINTER_TYPE_P (type)) + type = TREE_TYPE (type); + if (POINTER_TYPE_P (type)) + type = TREE_TYPE (type); + tmp = decl; + while (TREE_CODE (tmp) == COMPONENT_REF || TREE_CODE (tmp) == ARRAY_REF) + tmp = TREE_OPERAND (tmp, TREE_CODE (tmp) == COMPONENT_REF ? 1 : 0); + if (!gfc_is_polymorphic_nonptr (type) + && !gfc_has_alloc_comps (type, tmp, true)) + return NULL_TREE; + return decl; +} + +/* Return true if there is deep mapping, even if the number of mapping is known + at compile time. */ +bool +gfc_omp_deep_mapping_p (const gimple *ctx, tree clause) +{ + tree decl = gfc_omp_deep_mapping_int_p (ctx, clause); + if (decl == NULL_TREE) + return false; + return true; +} + +/* Handle gfc_omp_deep_mapping{,_cnt} */ +static tree +gfc_omp_deep_mapping_do (bool is_cnt, const gimple *ctx, tree clause, + unsigned HOST_WIDE_INT tkind, tree data, tree sizes, + tree kinds, tree offset_data, tree offset, + gimple_seq *seq) +{ + tree num = NULL_TREE; + location_t loc = OMP_CLAUSE_LOCATION (clause); + tree decl = gfc_omp_deep_mapping_int_p (ctx, clause); + bool poly_warned = false; + if (decl == NULL_TREE) + return NULL_TREE; + /* Handle: map(alloc:dt%cmp [len: ptr_size]) map(tofrom: D.0123...), + where GFC_DECL_SAVED_DESCRIPTOR(D.0123) is the same (here: dt%cmp). */ + if (OMP_CLAUSE_CODE (clause) == OMP_CLAUSE_MAP + && (OMP_CLAUSE_MAP_KIND (clause) == GOMP_MAP_ALLOC + || OMP_CLAUSE_MAP_KIND (clause) == GOMP_MAP_PRESENT_ALLOC)) + { + tree c = clause; + while ((c = OMP_CLAUSE_CHAIN (c)) != NULL_TREE) + { + if (!gfc_omp_deep_map_kind_p (c)) + continue; + tree d = gfc_omp_deep_mapping_int_p (ctx, c); + if (d != NULL_TREE && operand_equal_p (decl, d, 0)) + return NULL_TREE; + } + } + tree type = TREE_TYPE (decl); + if (POINTER_TYPE_P (type)) + type = TREE_TYPE (type); + if (POINTER_TYPE_P (type)) + type = TREE_TYPE (type); + bool poly = gfc_is_polymorphic_nonptr (type); + + if (is_cnt) + { + num = build_decl (loc, VAR_DECL, + create_tmp_var_name ("n_deepmap"), size_type_node); + tree tmp = fold_build2_loc (loc, MODIFY_EXPR, size_type_node, num, + build_int_cst (size_type_node, 0)); + gimple_add_tmp_var (num); + gimplify_and_add (tmp, seq); + } + else + gcc_assert (short_unsigned_type_node == TREE_TYPE (TREE_TYPE (kinds))); + + bool do_copy = poly; + bool do_alloc_check = false; + tree token = NULL_TREE; + tree tmp = decl; + if (poly) + { + tmp = TYPE_FIELDS (type); + type = TREE_TYPE (tmp); + } + else + while (TREE_CODE (tmp) == COMPONENT_REF || TREE_CODE (tmp) == ARRAY_REF) + tmp = TREE_OPERAND (tmp, TREE_CODE (tmp) == COMPONENT_REF ? 1 : 0); + /* If the clause argument is nonallocatable, skip is-allocate check. */ + if (GFC_DECL_GET_SCALAR_ALLOCATABLE (tmp) + || GFC_DECL_GET_SCALAR_POINTER (tmp) + || (GFC_DESCRIPTOR_TYPE_P (type) + && (GFC_TYPE_ARRAY_AKIND (type) == GFC_ARRAY_ALLOCATABLE + || GFC_TYPE_ARRAY_AKIND (type) == GFC_ARRAY_POINTER + || GFC_TYPE_ARRAY_AKIND (type) == GFC_ARRAY_POINTER_CONT))) + do_alloc_check = true; + + if (!is_cnt + && OMP_CLAUSE_CODE (clause) == OMP_CLAUSE_MAP + && (tkind == GOMP_MAP_ALLOC + || (tkind == GOMP_MAP_FROM + && (gimple_omp_target_kind (ctx) + != GF_OMP_TARGET_KIND_EXIT_DATA))) + && (poly || gfc_omp_replace_alloc_by_to_mapping (type, tmp, true))) + OMP_CLAUSE_SET_MAP_KIND (clause, tkind == GOMP_MAP_ALLOC ? GOMP_MAP_TO + : GOMP_MAP_TOFROM); + + /* TODO: For map(a(:)), we know it is present & allocated. */ + + tree present = (DECL_P (decl) ? gfc_omp_check_optional_argument (decl, true) + : NULL_TREE); + if (POINTER_TYPE_P (TREE_TYPE (decl)) + && POINTER_TYPE_P (TREE_TYPE (TREE_TYPE (decl)))) + decl = build_fold_indirect_ref (decl); + if (present) + { + tree then_label = create_artificial_label (loc); + tree end_label = create_artificial_label (loc); + gimple_seq seq2 = NULL; + tmp = force_gimple_operand (present, &seq2, true, NULL_TREE); + gimple_seq_add_seq (seq, seq2); + gimple_seq_add_stmt (seq, + gimple_build_cond_from_tree (present, + then_label, end_label)); + gimple_seq_add_stmt (seq, gimple_build_label (then_label)); + gfc_omp_deep_mapping_item (is_cnt, do_copy, do_alloc_check, loc, decl, + &token, tkind, data, sizes, kinds, + offset_data, offset, num, seq, ctx, + &poly_warned); + gimple_seq_add_stmt (seq, gimple_build_label (end_label)); + } + else + gfc_omp_deep_mapping_item (is_cnt, do_copy, do_alloc_check, loc, decl, + &token, tkind, data, sizes, kinds, offset_data, + offset, num, seq, ctx, &poly_warned); + /* Multiply by 2 as there are two mappings: data + pointer assign. */ + if (is_cnt) + gimplify_assign (num, + fold_build2_loc (loc, MULT_EXPR, + size_type_node, num, + build_int_cst (size_type_node, 2)), seq); + return num; +} + +/* Return tree with a variable which contains the count of deep-mappyings + (value depends, e.g., on allocation status) */ +tree +gfc_omp_deep_mapping_cnt (const gimple *ctx, tree clause, gimple_seq *seq) +{ + return gfc_omp_deep_mapping_do (true, ctx, clause, 0, NULL_TREE, NULL_TREE, + NULL_TREE, NULL_TREE, NULL_TREE, seq); +} + +/* Does the actual deep mapping. */ +void +gfc_omp_deep_mapping (const gimple *ctx, tree clause, + unsigned HOST_WIDE_INT tkind, tree data, + tree sizes, tree kinds, tree offset_data, tree offset, + gimple_seq *seq) +{ + (void) gfc_omp_deep_mapping_do (false, ctx, clause, tkind, data, sizes, kinds, + offset_data, offset, seq); +} + /* Return true if DECL is a scalar variable (for the purpose of implicit firstprivatization/mapping). Only if 'ptr_alloc_ok.' is true, allocatables and pointers are permitted. */ @@ -2478,6 +3265,18 @@ gfc_trans_omp_array_section (stmtblock_t *block, gfc_exec_op op, elemsz = fold_convert (gfc_array_index_type, elemsz); OMP_CLAUSE_SIZE (node) = fold_build2 (MULT_EXPR, gfc_array_index_type, OMP_CLAUSE_SIZE (node), elemsz); + if (n->expr->ts.type == BT_DERIVED + && n->expr->ts.u.derived->attr.alloc_comp) + { + /* Save array descriptor for use in gfc_omp_deep_mapping{,_p,_cnt}; + force evaluate to ensure that it is not gimplified + is a decl. */ + tree tmp = OMP_CLAUSE_SIZE (node); + tree var = gfc_create_var (TREE_TYPE (tmp), NULL); + gfc_add_modify_loc (input_location, block, var, tmp); + OMP_CLAUSE_SIZE (node) = var; + gfc_allocate_lang_decl (var); + GFC_DECL_SAVED_DESCRIPTOR (var) = se.expr; + } } gcc_assert (se.post.head == NULL_TREE); gcc_assert (POINTER_TYPE_P (TREE_TYPE (ptr))); @@ -3213,8 +4012,9 @@ gfc_trans_omp_clauses (stmtblock_t *block, gfc_omp_clauses *clauses, if (!n->sym->attr.referenced) continue; + location_t map_loc = gfc_get_location (&n->where); bool always_modifier = false; - tree node = build_omp_clause (input_location, OMP_CLAUSE_MAP); + tree node = build_omp_clause (map_loc, OMP_CLAUSE_MAP); tree node2 = NULL_TREE; tree node3 = NULL_TREE; tree node4 = NULL_TREE; @@ -3361,7 +4161,7 @@ gfc_trans_omp_clauses (stmtblock_t *block, gfc_omp_clauses *clauses, && n->u.map.op != OMP_MAP_RELEASE) { gcc_assert (n->sym->ts.u.cl->backend_decl); - node5 = build_omp_clause (input_location, OMP_CLAUSE_MAP); + node5 = build_omp_clause (map_loc, OMP_CLAUSE_MAP); OMP_CLAUSE_SET_MAP_KIND (node5, GOMP_MAP_ALWAYS_TO); OMP_CLAUSE_DECL (node5) = n->sym->ts.u.cl->backend_decl; OMP_CLAUSE_SIZE (node5) @@ -3378,7 +4178,7 @@ gfc_trans_omp_clauses (stmtblock_t *block, gfc_omp_clauses *clauses, ptr = build_fold_indirect_ref (ptr); OMP_CLAUSE_DECL (node) = ptr; OMP_CLAUSE_SIZE (node) = gfc_class_vtab_size_get (decl); - node2 = build_omp_clause (input_location, OMP_CLAUSE_MAP); + node2 = build_omp_clause (map_loc, OMP_CLAUSE_MAP); OMP_CLAUSE_SET_MAP_KIND (node2, GOMP_MAP_ATTACH_DETACH); OMP_CLAUSE_DECL (node2) = gfc_class_data_get (decl); OMP_CLAUSE_SIZE (node2) = size_int (0); @@ -3434,8 +4234,7 @@ gfc_trans_omp_clauses (stmtblock_t *block, gfc_omp_clauses *clauses, size = TYPE_SIZE_UNIT (TREE_TYPE (decl)); else size = size_int (0); - node4 = build_omp_clause (input_location, - OMP_CLAUSE_MAP); + node4 = build_omp_clause (map_loc, OMP_CLAUSE_MAP); OMP_CLAUSE_SET_MAP_KIND (node4, gmk); OMP_CLAUSE_DECL (node4) = decl; OMP_CLAUSE_SIZE (node4) = size; @@ -3459,8 +4258,7 @@ gfc_trans_omp_clauses (stmtblock_t *block, gfc_omp_clauses *clauses, size = TYPE_SIZE_UNIT (TREE_TYPE (decl)); else size = size_int (0); - node3 = build_omp_clause (input_location, - OMP_CLAUSE_MAP); + node3 = build_omp_clause (map_loc, OMP_CLAUSE_MAP); OMP_CLAUSE_SET_MAP_KIND (node3, gmk); OMP_CLAUSE_DECL (node3) = decl; OMP_CLAUSE_SIZE (node3) = size; @@ -3477,7 +4275,7 @@ gfc_trans_omp_clauses (stmtblock_t *block, gfc_omp_clauses *clauses, gcc_assert (POINTER_TYPE_P (TREE_TYPE (ptr))); ptr = build_fold_indirect_ref (ptr); OMP_CLAUSE_DECL (node) = ptr; - node2 = build_omp_clause (input_location, OMP_CLAUSE_MAP); + node2 = build_omp_clause (map_loc, OMP_CLAUSE_MAP); OMP_CLAUSE_DECL (node2) = decl; OMP_CLAUSE_SIZE (node2) = TYPE_SIZE_UNIT (type); if (n->u.map.op == OMP_MAP_DELETE) @@ -3493,8 +4291,7 @@ gfc_trans_omp_clauses (stmtblock_t *block, gfc_omp_clauses *clauses, && n->u.map.op != OMP_MAP_DELETE && n->u.map.op != OMP_MAP_RELEASE) { - node3 = build_omp_clause (input_location, - OMP_CLAUSE_MAP); + node3 = build_omp_clause (map_loc, OMP_CLAUSE_MAP); if (present) { ptr = gfc_conv_descriptor_data_get (decl); @@ -3634,10 +4431,10 @@ gfc_trans_omp_clauses (stmtblock_t *block, gfc_omp_clauses *clauses, { /* A single indirectref is handled by the middle end. */ gcc_assert (!POINTER_TYPE_P (TREE_TYPE (decl))); - decl = TREE_OPERAND (decl, 0); - decl = gfc_build_cond_assign_expr (block, present, decl, + tree tmp = TREE_OPERAND (decl, 0); + tmp = gfc_build_cond_assign_expr (block, present, tmp, null_pointer_node); - OMP_CLAUSE_DECL (node) = build_fold_indirect_ref (decl); + OMP_CLAUSE_DECL (node) = build_fold_indirect_ref (tmp); } else OMP_CLAUSE_DECL (node) = decl; @@ -3672,6 +4469,33 @@ gfc_trans_omp_clauses (stmtblock_t *block, gfc_omp_clauses *clauses, size = gfc_evaluate_now (size, block); OMP_CLAUSE_SIZE (node) = size; } + if ((TREE_CODE (decl) != PARM_DECL + || DECL_ARTIFICIAL (OMP_CLAUSE_DECL (node))) + && n->sym->ts.type == BT_DERIVED + && n->sym->ts.u.derived->attr.alloc_comp) + { + /* Save array descriptor for use in + gfc_omp_deep_mapping{,_p,_cnt}; force evaluate + to ensure that it is not gimplified + is a decl. */ + tree tmp = OMP_CLAUSE_SIZE (node); + if (tmp == NULL_TREE) + tmp = DECL_P (decl) ? DECL_SIZE_UNIT (decl) + : TYPE_SIZE_UNIT (TREE_TYPE (decl)); + tree var = gfc_create_var (TREE_TYPE (tmp), NULL); + gfc_add_modify_loc (input_location, block, var, tmp); + OMP_CLAUSE_SIZE (node) = var; + gfc_allocate_lang_decl (var); + if (TREE_CODE (decl) == INDIRECT_REF) + decl = TREE_OPERAND (decl, 0); + if (TREE_CODE (decl) == INDIRECT_REF) + decl = TREE_OPERAND (decl, 0); + if (DECL_LANG_SPECIFIC (decl) + && GFC_DECL_SAVED_DESCRIPTOR (decl)) + GFC_DECL_SAVED_DESCRIPTOR (var) + = GFC_DECL_SAVED_DESCRIPTOR (decl); + else + GFC_DECL_SAVED_DESCRIPTOR (var) = decl; + } } else if (n->expr && n->expr->expr_type == EXPR_VARIABLE @@ -3727,8 +4551,7 @@ gfc_trans_omp_clauses (stmtblock_t *block, gfc_omp_clauses *clauses, goto finalize_map_clause; } - node2 = build_omp_clause (input_location, - OMP_CLAUSE_MAP); + node2 = build_omp_clause (map_loc, OMP_CLAUSE_MAP); OMP_CLAUSE_SET_MAP_KIND (node2, GOMP_MAP_ATTACH_DETACH); OMP_CLAUSE_DECL (node2) = POINTER_TYPE_P (TREE_TYPE (se.expr)) @@ -3754,13 +4577,37 @@ gfc_trans_omp_clauses (stmtblock_t *block, gfc_omp_clauses *clauses, kind = GOMP_MAP_RELEASE; else kind = GOMP_MAP_TO; - node3 = build_omp_clause (input_location, - OMP_CLAUSE_MAP); + node3 = build_omp_clause (map_loc, OMP_CLAUSE_MAP); OMP_CLAUSE_SET_MAP_KIND (node3, kind); OMP_CLAUSE_DECL (node3) = se.string_length; OMP_CLAUSE_SIZE (node3) = TYPE_SIZE_UNIT (gfc_charlen_type_node); } + if (!openacc + && n->expr->ts.type == BT_DERIVED + && n->expr->ts.u.derived->attr.alloc_comp) + { + /* Save array descriptor for use in + gfc_omp_deep_mapping{,_p,_cnt}; force evaluate + to ensure that it is not gimplified + is a decl. */ + tree tmp = OMP_CLAUSE_SIZE (node); + if (tmp == NULL_TREE) + tmp = (DECL_P (se.expr) + ? DECL_SIZE_UNIT (se.expr) + : TYPE_SIZE_UNIT (TREE_TYPE (se.expr))); + tree var = gfc_create_var (TREE_TYPE (tmp), NULL); + gfc_add_modify_loc (input_location, block, var, tmp); + OMP_CLAUSE_SIZE (node) = var; + gfc_allocate_lang_decl (var); + if (TREE_CODE (se.expr) == INDIRECT_REF) + se.expr = TREE_OPERAND (se.expr, 0); + if (DECL_LANG_SPECIFIC (se.expr) + && GFC_DECL_SAVED_DESCRIPTOR (se.expr)) + GFC_DECL_SAVED_DESCRIPTOR (var) + = GFC_DECL_SAVED_DESCRIPTOR (se.expr); + else + GFC_DECL_SAVED_DESCRIPTOR (var) = se.expr; + } } } else if (n->expr @@ -3800,7 +4647,7 @@ gfc_trans_omp_clauses (stmtblock_t *block, gfc_omp_clauses *clauses, && (lastref->u.c.component->ts.type == BT_DERIVED || lastref->u.c.component->ts.type == BT_CLASS)) { - if (pointer || (openacc && allocatable)) + if (pointer || allocatable) { /* If it's a bare attach/detach clause, we just want to perform a single attach/detach operation, of the @@ -3880,8 +4727,7 @@ gfc_trans_omp_clauses (stmtblock_t *block, gfc_omp_clauses *clauses, OMP_CLAUSE_DECL (node) = data; OMP_CLAUSE_SIZE (node) = size; - node2 = build_omp_clause (input_location, - OMP_CLAUSE_MAP); + node2 = build_omp_clause (map_loc, OMP_CLAUSE_MAP); OMP_CLAUSE_SET_MAP_KIND (node2, GOMP_MAP_ATTACH_DETACH); OMP_CLAUSE_DECL (node2) = build_fold_addr_expr (data); @@ -3893,6 +4739,22 @@ gfc_trans_omp_clauses (stmtblock_t *block, gfc_omp_clauses *clauses, OMP_CLAUSE_SIZE (node) = TYPE_SIZE_UNIT (TREE_TYPE (inner)); } + if (!openacc + && n->expr->ts.type == BT_DERIVED + && n->expr->ts.u.derived->attr.alloc_comp) + { + /* Save array descriptor for use in + gfc_omp_deep_mapping{,_p,_cnt}; force evaluate + to ensure that it is not gimplified + is a decl. */ + tree tmp = OMP_CLAUSE_SIZE (node); + tree var = gfc_create_var (TREE_TYPE (tmp), NULL); + gfc_add_modify_loc (input_location, block, var, tmp); + OMP_CLAUSE_SIZE (node) = var; + gfc_allocate_lang_decl (var); + if (TREE_CODE (inner) == INDIRECT_REF) + inner = TREE_OPERAND (inner, 0); + GFC_DECL_SAVED_DESCRIPTOR (var) = inner; + } } else if (lastref->type == REF_ARRAY && lastref->u.ar.type == AR_FULL) @@ -3952,8 +4814,7 @@ gfc_trans_omp_clauses (stmtblock_t *block, gfc_omp_clauses *clauses, elemsz = TYPE_SIZE_UNIT (elemsz); elemsz = fold_build2 (MULT_EXPR, size_type_node, len, elemsz); - node4 = build_omp_clause (input_location, - OMP_CLAUSE_MAP); + node4 = build_omp_clause (map_loc, OMP_CLAUSE_MAP); OMP_CLAUSE_SET_MAP_KIND (node4, map_kind); OMP_CLAUSE_DECL (node4) = se.string_length; OMP_CLAUSE_SIZE (node4) @@ -3963,8 +4824,7 @@ gfc_trans_omp_clauses (stmtblock_t *block, gfc_omp_clauses *clauses, OMP_CLAUSE_SIZE (node) = fold_build2 (MULT_EXPR, gfc_array_index_type, OMP_CLAUSE_SIZE (node), elemsz); - node2 = build_omp_clause (input_location, - OMP_CLAUSE_MAP); + node2 = build_omp_clause (map_loc, OMP_CLAUSE_MAP); if (map_kind == GOMP_MAP_RELEASE || map_kind == GOMP_MAP_DELETE) { @@ -3978,6 +4838,23 @@ gfc_trans_omp_clauses (stmtblock_t *block, gfc_omp_clauses *clauses, OMP_CLAUSE_SIZE (node2) = TYPE_SIZE_UNIT (type); if (!openacc) { + if (n->expr->ts.type == BT_DERIVED + && n->expr->ts.u.derived->attr.alloc_comp) + { + /* Save array descriptor for use + in gfc_omp_deep_mapping{,_p,_cnt}; force + evaluate to ensure that it is + not gimplified + is a decl. */ + tree tmp = OMP_CLAUSE_SIZE (node); + tree var = gfc_create_var (TREE_TYPE (tmp), + NULL); + gfc_add_modify_loc (map_loc, block, + var, tmp); + OMP_CLAUSE_SIZE (node) = var; + gfc_allocate_lang_decl (var); + GFC_DECL_SAVED_DESCRIPTOR (var) = inner; + } + gfc_omp_namelist *n2 = clauses->lists[OMP_LIST_MAP]; @@ -4035,8 +4912,7 @@ gfc_trans_omp_clauses (stmtblock_t *block, gfc_omp_clauses *clauses, if (drop_mapping) continue; } - node3 = build_omp_clause (input_location, - OMP_CLAUSE_MAP); + node3 = build_omp_clause (map_loc, OMP_CLAUSE_MAP); OMP_CLAUSE_SET_MAP_KIND (node3, GOMP_MAP_ATTACH_DETACH); OMP_CLAUSE_DECL (node3) @@ -4107,7 +4983,8 @@ gfc_trans_omp_clauses (stmtblock_t *block, gfc_omp_clauses *clauses, default: gcc_unreachable (); } - tree node = build_omp_clause (input_location, clause_code); + tree node = build_omp_clause (gfc_get_location (&n->where), + clause_code); if (n->expr == NULL || (n->expr->ref->type == REF_ARRAY && n->expr->ref->u.ar.type == AR_FULL diff --git a/gcc/fortran/trans-stmt.cc b/gcc/fortran/trans-stmt.cc index 94ecde0..37f8aca 100644 --- a/gcc/fortran/trans-stmt.cc +++ b/gcc/fortran/trans-stmt.cc @@ -5154,7 +5154,7 @@ gfc_trans_concurrent_locality_spec (bool after_body, stmtblock_t *body, gfc_start_saved_local_decls (); cnt = 0; - static_assert (LOCALITY_LOCAL_INIT - LOCALITY_LOCAL == 1); + static_assert (LOCALITY_LOCAL_INIT - LOCALITY_LOCAL == 1, "locality_type"); for (int type = LOCALITY_LOCAL; type <= LOCALITY_LOCAL_INIT; type++) for (el = locality_list[type]; el; el = el->next) diff --git a/gcc/fortran/trans.h b/gcc/fortran/trans.h index 63a566a..ae7be9f 100644 --- a/gcc/fortran/trans.h +++ b/gcc/fortran/trans.h @@ -839,6 +839,10 @@ tree gfc_omp_clause_assign_op (tree, tree, tree); tree gfc_omp_clause_linear_ctor (tree, tree, tree, tree); tree gfc_omp_clause_dtor (tree, tree); void gfc_omp_finish_clause (tree, gimple_seq *, bool); +bool gfc_omp_deep_mapping_p (const gimple *, tree); +tree gfc_omp_deep_mapping_cnt (const gimple *, tree, gimple_seq *); +void gfc_omp_deep_mapping (const gimple *, tree, unsigned HOST_WIDE_INT, tree, + tree, tree, tree, tree, gimple_seq *); bool gfc_omp_allocatable_p (tree); bool gfc_omp_scalar_p (tree, bool); bool gfc_omp_scalar_target_p (tree); @@ -30,6 +30,9 @@ compilation is specified by a string called a "spec". */ #define INCLUDE_STRING #include "config.h" #include "system.h" +#ifdef HOST_HAS_PERSONALITY_ADDR_NO_RANDOMIZE +#include <sys/personality.h> +#endif #include "coretypes.h" #include "multilib.h" /* before tm.h */ #include "tm.h" @@ -7744,55 +7747,58 @@ print_configuration (FILE *file) #define RETRY_ICE_ATTEMPTS 3 -/* Returns true if FILE1 and FILE2 contain equivalent data, 0 otherwise. */ +/* Returns true if FILE1 and FILE2 contain equivalent data, 0 otherwise. + If lines start with 0x followed by 1-16 lowercase hexadecimal digits + followed by a space, ignore anything before that space. These are + typically function addresses from libbacktrace and those can differ + due to ASLR. */ static bool files_equal_p (char *file1, char *file2) { - struct stat st1, st2; - off_t n, len; - int fd1, fd2; - const int bufsize = 8192; - char *buf = XNEWVEC (char, bufsize); - - fd1 = open (file1, O_RDONLY); - fd2 = open (file2, O_RDONLY); - - if (fd1 < 0 || fd2 < 0) - goto error; - - if (fstat (fd1, &st1) < 0 || fstat (fd2, &st2) < 0) - goto error; + FILE *f1 = fopen (file1, "rb"); + FILE *f2 = fopen (file2, "rb"); + char line1[256], line2[256]; - if (st1.st_size != st2.st_size) - goto error; - - for (n = st1.st_size; n; n -= len) + bool line_start = true; + while (fgets (line1, sizeof (line1), f1)) { - len = n; - if ((int) len > bufsize / 2) - len = bufsize / 2; - - if (read (fd1, buf, len) != (int) len - || read (fd2, buf + bufsize / 2, len) != (int) len) + if (!fgets (line2, sizeof (line2), f2)) + goto error; + char *p1 = line1, *p2 = line2; + if (line_start + && line1[0] == '0' + && line1[1] == 'x' + && line2[0] == '0' + && line2[1] == 'x') { - goto error; + int i, j; + for (i = 0; i < 16; ++i) + if (!ISXDIGIT (line1[2 + i]) || ISUPPER (line1[2 + i])) + break; + for (j = 0; j < 16; ++j) + if (!ISXDIGIT (line2[2 + j]) || ISUPPER (line2[2 + j])) + break; + if (i && line1[2 + i] == ' ' && j && line2[2 + j] == ' ') + { + p1 = line1 + i + 3; + p2 = line2 + j + 3; + } } - - if (memcmp (buf, buf + bufsize / 2, len) != 0) + if (strcmp (p1, p2) != 0) goto error; + line_start = strchr (line1, '\n') != NULL; } + if (fgets (line2, sizeof (line2), f2)) + goto error; - free (buf); - close (fd1); - close (fd2); - + fclose (f1); + fclose (f2); return 1; error: - free (buf); - close (fd1); - close (fd2); + fclose (f1); + fclose (f2); return 0; } @@ -8000,6 +8006,10 @@ try_generate_repro (const char **argv) else new_argv[out_arg] = "-o-"; +#ifdef HOST_HAS_PERSONALITY_ADDR_NO_RANDOMIZE + personality (personality (0xffffffffU) | ADDR_NO_RANDOMIZE); +#endif + int status; for (attempt = 0; attempt < RETRY_ICE_ATTEMPTS; ++attempt) { diff --git a/gcc/gimple-expr.cc b/gcc/gimple-expr.cc index a670e46..c0367f4 100644 --- a/gcc/gimple-expr.cc +++ b/gcc/gimple-expr.cc @@ -884,7 +884,7 @@ bool is_gimple_mem_ref_addr (tree t) { return (is_gimple_reg (t) - || TREE_CODE (t) == INTEGER_CST + || poly_int_tree_p (t) || (TREE_CODE (t) == ADDR_EXPR && (CONSTANT_CLASS_P (TREE_OPERAND (t, 0)) || decl_address_invariant_p (TREE_OPERAND (t, 0))))); diff --git a/gcc/gimple-fold.cc b/gcc/gimple-fold.cc index b645613..94d5a1e 100644 --- a/gcc/gimple-fold.cc +++ b/gcc/gimple-fold.cc @@ -906,20 +906,60 @@ size_must_be_zero_p (tree size) static bool optimize_memcpy_to_memset (gimple_stmt_iterator *gsip, tree dest, tree src, tree len) { + ao_ref read; gimple *stmt = gsi_stmt (*gsip); if (gimple_has_volatile_ops (stmt)) return false; - tree vuse = gimple_vuse (stmt); - if (vuse == NULL || TREE_CODE (vuse) != SSA_NAME) - return false; - gimple *defstmt = SSA_NAME_DEF_STMT (vuse); tree src2 = NULL_TREE, len2 = NULL_TREE; poly_int64 offset, offset2; tree val = integer_zero_node; + bool len_was_null = len == NULL_TREE; + if (len == NULL_TREE) + len = (TREE_CODE (src) == COMPONENT_REF + ? DECL_SIZE_UNIT (TREE_OPERAND (src, 1)) + : TYPE_SIZE_UNIT (TREE_TYPE (src))); + if (len == NULL_TREE + || !poly_int_tree_p (len)) + return false; + + ao_ref_init (&read, src); + tree vuse = gimple_vuse (stmt); + gimple *defstmt; + do { + if (vuse == NULL || TREE_CODE (vuse) != SSA_NAME) + return false; + defstmt = SSA_NAME_DEF_STMT (vuse); + if (is_a <gphi*>(defstmt)) + return false; + + /* If the len was null, then we can use TBBA. */ + if (stmt_may_clobber_ref_p_1 (defstmt, &read, + /* tbaa_p = */ len_was_null)) + break; + vuse = gimple_vuse (defstmt); + } while (true); + if (gimple_store_p (defstmt) && gimple_assign_single_p (defstmt) + && TREE_CODE (gimple_assign_rhs1 (defstmt)) == STRING_CST + && !gimple_clobber_p (defstmt)) + { + tree str = gimple_assign_rhs1 (defstmt); + src2 = gimple_assign_lhs (defstmt); + /* The string must contain all null char's for now. */ + for (int i = 0; i < TREE_STRING_LENGTH (str); i++) + { + if (TREE_STRING_POINTER (str)[i] != 0) + { + src2 = NULL_TREE; + break; + } + } + } + else if (gimple_store_p (defstmt) + && gimple_assign_single_p (defstmt) && TREE_CODE (gimple_assign_rhs1 (defstmt)) == CONSTRUCTOR && !gimple_clobber_p (defstmt)) src2 = gimple_assign_lhs (defstmt); @@ -939,17 +979,11 @@ optimize_memcpy_to_memset (gimple_stmt_iterator *gsip, tree dest, tree src, tree if (src2 == NULL_TREE) return false; - if (len == NULL_TREE) - len = (TREE_CODE (src) == COMPONENT_REF - ? DECL_SIZE_UNIT (TREE_OPERAND (src, 1)) - : TYPE_SIZE_UNIT (TREE_TYPE (src))); if (len2 == NULL_TREE) len2 = (TREE_CODE (src2) == COMPONENT_REF ? DECL_SIZE_UNIT (TREE_OPERAND (src2, 1)) : TYPE_SIZE_UNIT (TREE_TYPE (src2))); - if (len == NULL_TREE - || !poly_int_tree_p (len) - || len2 == NULL_TREE + if (len2 == NULL_TREE || !poly_int_tree_p (len2)) return false; diff --git a/gcc/gimple-lower-bitint.cc b/gcc/gimple-lower-bitint.cc index cf30d8e..6fefc83 100644 --- a/gcc/gimple-lower-bitint.cc +++ b/gcc/gimple-lower-bitint.cc @@ -1547,14 +1547,15 @@ bitint_large_huge::handle_cast (tree lhs_type, tree rhs1, tree idx) } else { - if (tree_to_uhwi (idx) < low) + unsigned tidx = tree_to_uhwi (idx); + if (tidx < low) { t = handle_operand (rhs1, idx); if (m_first) m_data[save_data_cnt + 2] = build_int_cst (NULL_TREE, m_data_cnt); } - else if (tree_to_uhwi (idx) < high) + else if (tidx < high) { t = handle_operand (rhs1, size_int (low)); if (m_first) @@ -1587,7 +1588,9 @@ bitint_large_huge::handle_cast (tree lhs_type, tree rhs1, tree idx) m_data_cnt = tree_to_uhwi (m_data[save_data_cnt + 2]); if (TYPE_UNSIGNED (rhs_type)) t = build_zero_cst (m_limb_type); - else if (m_bb && m_data[save_data_cnt]) + else if (m_bb + && m_data[save_data_cnt] + && ((tidx & 1) == 0 || tidx != low + 1)) t = m_data[save_data_cnt]; else t = m_data[save_data_cnt + 1]; @@ -5916,7 +5919,8 @@ build_bitint_stmt_ssa_conflicts (gimple *stmt, live_track *live, ssa_conflicts *graph, bitmap names, void (*def) (live_track *, tree, ssa_conflicts *), - void (*use) (live_track *, tree)) + void (*use) (live_track *, tree), + void (*clear) (live_track *, tree)) { bool muldiv_p = false; tree lhs = NULL_TREE; @@ -5933,6 +5937,25 @@ build_bitint_stmt_ssa_conflicts (gimple *stmt, live_track *live, { if (!bitmap_bit_p (names, SSA_NAME_VERSION (lhs))) return; + + /* A copy between 2 partitions does not introduce an interference + by itself. If they did, you would never be able to coalesce + two things which are copied. If the two variables really do + conflict, they will conflict elsewhere in the program. + + This is handled by simply removing the SRC of the copy from + the live list, and processing the stmt normally. + + Don't do this if lhs is not in names though, in such cases + it is actually used at some point later in the basic + block. */ + if (gimple_assign_copy_p (stmt)) + { + tree rhs1 = gimple_assign_rhs1 (stmt); + if (TREE_CODE (rhs1) == SSA_NAME) + clear (live, rhs1); + } + switch (gimple_assign_rhs_code (stmt)) { case MULT_EXPR: @@ -6624,10 +6647,28 @@ gimple_lower_bitint (void) bitmap_set_bit (large_huge.m_names, SSA_NAME_VERSION (s)); if (has_single_use (s)) { - if (!large_huge.m_single_use_names) - large_huge.m_single_use_names = BITMAP_ALLOC (NULL); - bitmap_set_bit (large_huge.m_single_use_names, - SSA_NAME_VERSION (s)); + tree s2 = s; + /* The coalescing hook special cases SSA_NAME copies. + Make sure not to mark in m_single_use_names single + use SSA_NAMEs copied from non-single use SSA_NAMEs. */ + while (gimple_assign_copy_p (SSA_NAME_DEF_STMT (s2))) + { + s2 = gimple_assign_rhs1 (SSA_NAME_DEF_STMT (s2)); + if (TREE_CODE (s2) != SSA_NAME) + break; + if (!has_single_use (s2)) + { + s2 = NULL_TREE; + break; + } + } + if (s2) + { + if (!large_huge.m_single_use_names) + large_huge.m_single_use_names = BITMAP_ALLOC (NULL); + bitmap_set_bit (large_huge.m_single_use_names, + SSA_NAME_VERSION (s)); + } } if (SSA_NAME_VAR (s) && ((TREE_CODE (SSA_NAME_VAR (s)) == PARM_DECL diff --git a/gcc/gimple-lower-bitint.h b/gcc/gimple-lower-bitint.h index 8662c4b..4798484 100644 --- a/gcc/gimple-lower-bitint.h +++ b/gcc/gimple-lower-bitint.h @@ -26,6 +26,7 @@ extern void build_bitint_stmt_ssa_conflicts (gimple *, live_track *, ssa_conflicts *, bitmap, void (*) (live_track *, tree, ssa_conflicts *), + void (*) (live_track *, tree), void (*) (live_track *, tree)); #endif /* GCC_GIMPLE_LOWER_BITINT_H */ diff --git a/gcc/ginclude/stddef.h b/gcc/ginclude/stddef.h index 0d53103..bacf24d 100644 --- a/gcc/ginclude/stddef.h +++ b/gcc/ginclude/stddef.h @@ -89,6 +89,21 @@ see the files COPYING3 and COPYING.RUNTIME respectively. If not, see #undef _PTRDIFF_T_ #endif +/* When modular code is enabled with macOS SDKs from version 15, the + include guards are set in the includers of this code, rather than as + part of it. This means the we must unset them or the intended code + here will be bypassed (resulting in undefined values). */ +#if defined (__APPLE__) +# if defined(__has_feature) && __has_feature(modules) +# if defined (__need_ptrdiff_t) +# undef __PTRDIFF_T +# endif +# if defined (__need_size_t) +# undef __SIZE_T +# endif +# endif +#endif + /* On VxWorks, <type/vxTypesBase.h> may have defined macros like _TYPE_size_t which will typedef size_t. fixincludes patched the vxTypesBase.h so that this macro is only defined if _GCC_SIZE_T is diff --git a/gcc/ipa-cp.cc b/gcc/ipa-cp.cc index 2645689..806c2bd 100644 --- a/gcc/ipa-cp.cc +++ b/gcc/ipa-cp.cc @@ -307,14 +307,31 @@ ipcp_lattice<valtype>::print (FILE * f, bool dump_sources, bool dump_benefits) fprintf (f, "\n"); } -/* If VALUE has all bits set to one, print "-1" to F, otherwise simply print it - hexadecimally to F. */ +/* Print VALUE to F in a form which in usual cases does not take thousands of + characters. */ static void ipcp_print_widest_int (FILE *f, const widest_int &value) { - if (wi::eq_p (wi::bit_not (value), 0)) + if (value == -1) fprintf (f, "-1"); + else if (wi::arshift (value, 128) == -1) + { + char buf[35], *p = buf + 2; + widest_int v = wi::zext (value, 128); + size_t len; + print_hex (v, buf); + len = strlen (p); + if (len == 32) + { + fprintf (f, "0xf..f"); + while (*p == 'f') + ++p; + } + else + fprintf (f, "0xf..f%0*d", (int) (32 - len), 0); + fputs (p, f); + } else print_hex (value, f); } @@ -331,7 +348,7 @@ ipcp_bits_lattice::print (FILE *f) fprintf (f, " Bits: value = "); ipcp_print_widest_int (f, get_value ()); fprintf (f, ", mask = "); - print_hex (get_mask (), f); + ipcp_print_widest_int (f, get_mask ()); fprintf (f, "\n"); } } @@ -916,11 +933,13 @@ ipcp_bits_lattice::meet_with_1 (widest_int value, widest_int mask, m_mask = (m_mask | mask) | (m_value ^ value); if (drop_all_ones) m_mask |= m_value; - m_value &= ~m_mask; + widest_int cap_mask = wi::shifted_mask <widest_int> (0, precision, true); + m_mask |= cap_mask; if (wi::sext (m_mask, precision) == -1) return set_to_bottom (); + m_value &= ~m_mask; return m_mask != old_mask; } @@ -996,6 +1015,8 @@ ipcp_bits_lattice::meet_with (ipcp_bits_lattice& other, unsigned precision, adjusted_mask |= adjusted_value; adjusted_value &= ~adjusted_mask; } + widest_int cap_mask = wi::shifted_mask <widest_int> (0, precision, true); + adjusted_mask |= cap_mask; if (wi::sext (adjusted_mask, precision) == -1) return set_to_bottom (); return set_to_constant (adjusted_value, adjusted_mask); @@ -1467,10 +1488,12 @@ ipacp_value_safe_for_type (tree param_type, tree value) return NULL_TREE; } -/* Return the result of a (possibly arithmetic) operation on the constant value - INPUT. OPERAND is 2nd operand for binary operation. RES_TYPE is the type - in which any operation is to be performed. Return NULL_TREE if that cannot - be determined or be considered an interprocedural invariant. */ +/* Return the result of a (possibly arithmetic) operation determined by OPCODE + on the constant value INPUT. OPERAND is 2nd operand for binary operation + and is required for binary operations. RES_TYPE, required when opcode is + not NOP_EXPR, is the type in which any operation is to be performed. Return + NULL_TREE if that cannot be determined or be considered an interprocedural + invariant. */ static tree ipa_get_jf_arith_result (enum tree_code opcode, tree input, tree operand, @@ -1491,16 +1514,6 @@ ipa_get_jf_arith_result (enum tree_code opcode, tree input, tree operand, return NULL_TREE; } - if (!res_type) - { - if (TREE_CODE_CLASS (opcode) == tcc_comparison) - res_type = boolean_type_node; - else if (expr_type_first_operand_type_p (opcode)) - res_type = TREE_TYPE (input); - else - return NULL_TREE; - } - if (TREE_CODE_CLASS (opcode) == tcc_unary) res = fold_unary (opcode, res_type, input); else @@ -1584,7 +1597,10 @@ ipa_value_from_jfunc (class ipa_node_params *info, struct ipa_jump_func *jfunc, return NULL_TREE; enum tree_code opcode = ipa_get_jf_pass_through_operation (jfunc); tree op2 = ipa_get_jf_pass_through_operand (jfunc); - tree cstval = ipa_get_jf_arith_result (opcode, input, op2, NULL_TREE); + tree op_type + = (opcode == NOP_EXPR) ? NULL_TREE + : ipa_get_jf_pass_through_op_type (jfunc); + tree cstval = ipa_get_jf_arith_result (opcode, input, op2, op_type); return ipacp_value_safe_for_type (parm_type, cstval); } else @@ -1724,24 +1740,7 @@ ipa_vr_intersect_with_arith_jfunc (vrange &vr, const value_range *inter_vr; if (operation != NOP_EXPR) { - /* Since we construct arithmetic jump functions even when there is a - type conversion in between the operation encoded in the jump - function and when it is passed in a call argument, the IPA - propagation phase must also perform the operation and conversion - in two separate steps. - - TODO: In order to remove the use of expr_type_first_operand_type_p - predicate we would need to stream the operation type, ideally - encoding the whole jump function as a series of expr_eval_op - structures. */ - - tree operation_type; - if (expr_type_first_operand_type_p (operation)) - operation_type = src_type; - else if (operation == ABSU_EXPR) - operation_type = unsigned_type_for (src_type); - else - return; + tree operation_type = ipa_get_jf_pass_through_op_type (jfunc); op_res.set_varying (operation_type); if (!ipa_vr_operation_and_type_effects (op_res, src_vr, operation, operation_type, src_type)) @@ -1771,14 +1770,7 @@ ipa_vr_intersect_with_arith_jfunc (vrange &vr, value_range op_vr (TREE_TYPE (operand)); ipa_get_range_from_ip_invariant (op_vr, operand, context_node); - tree operation_type; - if (TREE_CODE_CLASS (operation) == tcc_comparison) - operation_type = boolean_type_node; - else if (expr_type_first_operand_type_p (operation)) - operation_type = src_type; - else - return; - + tree operation_type = ipa_get_jf_pass_through_op_type (jfunc); value_range op_res (operation_type); if (!ipa_vr_supported_type_p (operation_type) || !handler.operand_check_p (operation_type, src_type, op_vr.type ()) @@ -1918,10 +1910,11 @@ ipa_agg_value_from_jfunc (ipa_node_params *info, cgraph_node *node, return NULL_TREE; } - return ipa_get_jf_arith_result (item->value.pass_through.operation, - value, - item->value.pass_through.operand, - item->type); + tree cstval = ipa_get_jf_arith_result (item->value.pass_through.operation, + value, + item->value.pass_through.operand, + item->value.pass_through.op_type); + return ipacp_value_safe_for_type (item->type, cstval); } /* Process all items in AGG_JFUNC relative to caller (or the node the original @@ -2150,13 +2143,15 @@ ipcp_lattice<valtype>::add_value (valtype newval, cgraph_edge *cs, /* A helper function that returns result of operation specified by OPCODE on the value of SRC_VAL. If non-NULL, OPND1_TYPE is expected type for the value of SRC_VAL. If the operation is binary, OPND2 is a constant value - acting as its second operand. */ + acting as its second operand. OP_TYPE is the type in which the operation is + performed. */ static tree get_val_across_arith_op (enum tree_code opcode, tree opnd1_type, tree opnd2, - ipcp_value<tree> *src_val) + ipcp_value<tree> *src_val, + tree op_type) { tree opnd1 = src_val->value; @@ -2165,17 +2160,19 @@ get_val_across_arith_op (enum tree_code opcode, && !useless_type_conversion_p (opnd1_type, TREE_TYPE (opnd1))) return NULL_TREE; - return ipa_get_jf_arith_result (opcode, opnd1, opnd2, NULL_TREE); + return ipa_get_jf_arith_result (opcode, opnd1, opnd2, op_type); } /* Propagate values through an arithmetic transformation described by a jump function associated with edge CS, taking values from SRC_LAT and putting - them into DEST_LAT. OPND1_TYPE is expected type for the values in SRC_LAT. - OPND2 is a constant value if transformation is a binary operation. - SRC_OFFSET specifies offset in an aggregate if SRC_LAT describes lattice of - a part of the aggregate. SRC_IDX is the index of the source parameter. - RES_TYPE is the value type of result being propagated into. Return true if - DEST_LAT changed. */ + them into DEST_LAT. OPND1_TYPE, if non-NULL, is the expected type for the + values in SRC_LAT. OPND2 is a constant value if transformation is a binary + operation. SRC_OFFSET specifies offset in an aggregate if SRC_LAT describes + lattice of a part of an aggregate, otherwise it should be -1. SRC_IDX is + the index of the source parameter. OP_TYPE is the type in which the + operation is performed and can be NULL when OPCODE is NOP_EXPR. RES_TYPE is + the value type of result being propagated into. Return true if DEST_LAT + changed. */ static bool propagate_vals_across_arith_jfunc (cgraph_edge *cs, @@ -2186,6 +2183,7 @@ propagate_vals_across_arith_jfunc (cgraph_edge *cs, ipcp_lattice<tree> *dest_lat, HOST_WIDE_INT src_offset, int src_idx, + tree op_type, tree res_type) { ipcp_value<tree> *src_val; @@ -2241,7 +2239,7 @@ propagate_vals_across_arith_jfunc (cgraph_edge *cs, for (int j = 1; j < max_recursive_depth; j++) { tree cstval = get_val_across_arith_op (opcode, opnd1_type, opnd2, - src_val); + src_val, op_type); cstval = ipacp_value_safe_for_type (res_type, cstval); if (!cstval) break; @@ -2266,7 +2264,7 @@ propagate_vals_across_arith_jfunc (cgraph_edge *cs, } tree cstval = get_val_across_arith_op (opcode, opnd1_type, opnd2, - src_val); + src_val, op_type); cstval = ipacp_value_safe_for_type (res_type, cstval); if (cstval) ret |= dest_lat->add_value (cstval, cs, src_val, src_idx, @@ -2290,11 +2288,13 @@ propagate_vals_across_pass_through (cgraph_edge *cs, ipa_jump_func *jfunc, tree parm_type) { gcc_checking_assert (parm_type); - return propagate_vals_across_arith_jfunc (cs, - ipa_get_jf_pass_through_operation (jfunc), - NULL_TREE, + enum tree_code opcode = ipa_get_jf_pass_through_operation (jfunc); + tree op_type = (opcode == NOP_EXPR) ? NULL_TREE + : ipa_get_jf_pass_through_op_type (jfunc); + return propagate_vals_across_arith_jfunc (cs, opcode, NULL_TREE, ipa_get_jf_pass_through_operand (jfunc), - src_lat, dest_lat, -1, src_idx, parm_type); + src_lat, dest_lat, -1, src_idx, op_type, + parm_type); } /* Propagate values through an ancestor jump function JFUNC associated with @@ -2507,14 +2507,12 @@ propagate_bits_across_jump_function (cgraph_edge *cs, int idx, return dest_lattice->set_to_bottom (); } - unsigned precision = TYPE_PRECISION (parm_type); - signop sgn = TYPE_SIGN (parm_type); - if (jfunc->type == IPA_JF_PASS_THROUGH || jfunc->type == IPA_JF_ANCESTOR) { ipa_node_params *caller_info = ipa_node_params_sum->get (cs->caller); tree operand = NULL_TREE; + tree op_type = NULL_TREE; enum tree_code code; unsigned src_idx; bool keep_null = false; @@ -2524,7 +2522,10 @@ propagate_bits_across_jump_function (cgraph_edge *cs, int idx, code = ipa_get_jf_pass_through_operation (jfunc); src_idx = ipa_get_jf_pass_through_formal_id (jfunc); if (code != NOP_EXPR) - operand = ipa_get_jf_pass_through_operand (jfunc); + { + operand = ipa_get_jf_pass_through_operand (jfunc); + op_type = ipa_get_jf_pass_through_op_type (jfunc); + } } else { @@ -2551,6 +2552,11 @@ propagate_bits_across_jump_function (cgraph_edge *cs, int idx, if (!src_lats->bits_lattice.bottom_p ()) { + if (!op_type) + op_type = ipa_get_type (caller_info, src_idx); + + unsigned precision = TYPE_PRECISION (op_type); + signop sgn = TYPE_SIGN (op_type); bool drop_all_ones = keep_null && !src_lats->bits_lattice.known_nonzero_p (); @@ -2570,7 +2576,8 @@ propagate_bits_across_jump_function (cgraph_edge *cs, int idx, = widest_int::from (bm.mask (), TYPE_SIGN (parm_type)); widest_int value = widest_int::from (bm.value (), TYPE_SIGN (parm_type)); - return dest_lattice->meet_with (value, mask, precision); + return dest_lattice->meet_with (value, mask, + TYPE_PRECISION (parm_type)); } } return dest_lattice->set_to_bottom (); @@ -2869,6 +2876,7 @@ propagate_aggregate_lattice (struct cgraph_edge *cs, src_lat, aglat, src_offset, src_idx, + item->value.pass_through.op_type, item->type); if (src_lat->contains_variable) @@ -5394,11 +5402,14 @@ find_more_scalar_values_for_callers_subset (struct cgraph_node *node, if (self_recursive_pass_through_p (cs, jump_func, i, false)) { gcc_assert (newval); - t = ipa_get_jf_arith_result ( - ipa_get_jf_pass_through_operation (jump_func), - newval, + enum tree_code opcode + = ipa_get_jf_pass_through_operation (jump_func); + tree op_type = (opcode == NOP_EXPR) ? NULL_TREE + : ipa_get_jf_pass_through_op_type (jump_func); + t = ipa_get_jf_arith_result (opcode, newval, ipa_get_jf_pass_through_operand (jump_func), - type); + op_type); + t = ipacp_value_safe_for_type (type, t); } else t = ipa_value_from_jfunc (ipa_node_params_sum->get (cs->caller), @@ -5603,10 +5614,13 @@ push_agg_values_for_index_from_edge (struct cgraph_edge *cs, int index, && self_recursive_agg_pass_through_p (cs, &agg_jf, index, false) && (srcvalue = interim->get_value(index, agg_jf.offset / BITS_PER_UNIT))) - value = ipa_get_jf_arith_result (agg_jf.value.pass_through.operation, - srcvalue, - agg_jf.value.pass_through.operand, - agg_jf.type); + { + value = ipa_get_jf_arith_result (agg_jf.value.pass_through.operation, + srcvalue, + agg_jf.value.pass_through.operand, + agg_jf.value.pass_through.op_type); + value = ipacp_value_safe_for_type (agg_jf.type, value); + } else value = ipa_agg_value_from_jfunc (caller_info, cs->caller, &agg_jf); @@ -6426,7 +6440,7 @@ ipcp_store_vr_results (void) fprintf (dump_file, " param %i: value = ", i); ipcp_print_widest_int (dump_file, bits->get_value ()); fprintf (dump_file, ", mask = "); - print_hex (bits->get_mask (), dump_file); + ipcp_print_widest_int (dump_file, bits->get_mask ()); fprintf (dump_file, "\n"); } } diff --git a/gcc/ipa-locality-cloning.cc b/gcc/ipa-locality-cloning.cc new file mode 100644 index 0000000..2684046 --- /dev/null +++ b/gcc/ipa-locality-cloning.cc @@ -0,0 +1,1137 @@ +/* Code locality based function cloning. + Copyright The GNU Toolchain Authors + +This file is part of GCC. + +GCC is free software; you can redistribute it and/or modify it under +the terms of the GNU General Public License as published by the Free +Software Foundation; either version 3, or (at your option) any later +version. + +GCC is distributed in the hope that it will be useful, but WITHOUT ANY +WARRANTY; without even the implied warranty of MERCHANTABILITY or +FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License +for more details. + +You should have received a copy of the GNU General Public License +along with GCC; see the file COPYING3. If not see +<http://www.gnu.org/licenses/>. */ + +/* This file implements cloning required to improve partitioning of the + callgraph for locality considerations. + + Partitioning for improving code locality. + This pass aims to place frequently executed callchains closer together in + memory to improve performance through improved locality. If any frequent + callchains cannot be placed together because they are already placed + elsewhere, local function clones are created and all callers near to the + clones are redirected to use this copy. + + Locality code placement is done in 2 parts. + 1. IPA pass to be executed after ipa-inline and before ipa-pure-const. + Execute stage prepares the plan to place all nodes into partitions. + 2. WPA Partition stage actually implements the plan. + + Brief overview of the IPA pass: + 1. Create and sort callchains. If PGO is available, use real profile + counts. Otherwise, use a set of heuristics to sort the callchains. + 2. Create a partition plan for the callchains, processing them in the sorted + order. + 1. If a function is unpartitioned, place it in the current partition. + 2. If a function is already placed in a partition away from current + partition as part of another callchain: + Create a local clone in current partition, if cloning criteria is + satisfied. + 3. Redirect any new caller to a local clone if one exists. + Partition size is param controlled to fine tune per program behavior. */ + +#include "config.h" +#define INCLUDE_ALGORITHM +#include "system.h" +#include "coretypes.h" +#include "target.h" +#include "function.h" +#include "tree.h" +#include "alloc-pool.h" +#include "tree-pass.h" +#include "cgraph.h" +#include "symbol-summary.h" +#include "tree-vrp.h" +#include "symtab-thunks.h" +#include "sreal.h" +#include "ipa-cp.h" +#include "ipa-prop.h" +#include "ipa-fnsummary.h" +#include "ipa-modref-tree.h" +#include "ipa-modref.h" +#include "symtab-clones.h" +#include "ipa-locality-cloning.h" + +/* Locality partitions, assigns nodes to partitions. These are used later in + WPA partitioning. */ +vec<locality_partition> locality_partitions; + +/* Map from original node to its latest clone. Gets overwritten whenever a new + clone is created from the same node. */ +hash_map<cgraph_node *, cgraph_node *> node_to_clone; +/* Map from clone to its original node. */ +hash_map<cgraph_node *, cgraph_node *> clone_to_node; + +/* Data structure to hold static heuristics and orders for cgraph_nodes. */ +struct locality_order +{ + cgraph_node *node; + sreal order; + locality_order (cgraph_node *node, sreal order) : node (node), order (order) + {} +}; + +/* Return true if NODE is already in some partition. */ +static inline bool +node_partitioned_p (cgraph_node *node) +{ + return node->aux; +} + +/* Add symbol NODE to partition PART. */ +static void +add_node_to_partition (locality_partition part, cgraph_node *node) +{ + struct cgraph_edge *e; + if (node_partitioned_p (node)) + return; + + part->nodes.safe_push (node); + node->aux = (void *) (uintptr_t) (part->part_id); + + if (!node->alias && node->get_partitioning_class () == SYMBOL_PARTITION) + part->insns += ipa_size_summaries->get (node)->size; + + /* Add all inline clones and callees that are duplicated. */ + for (e = node->callees; e; e = e->next_callee) + if (!e->inline_failed) + add_node_to_partition (part, e->callee); + /* omp declare_variant_alt or transparent_alias with definition or linker + discardable (non-local comdat but not forced and not + used by non-LTO). */ + else if (e->callee->get_partitioning_class () == SYMBOL_DUPLICATE) + add_node_to_partition (part, e->callee); + + /* Add all thunks associated with the function. */ + for (e = node->callers; e; e = e->next_caller) + if (e->caller->thunk && !e->caller->inlined_to) + add_node_to_partition (part, e->caller); + + /* Add all aliases associated with the symbol. */ + struct ipa_ref *ref; + FOR_EACH_ALIAS (node, ref) + if (!ref->referring->transparent_alias) + { + cgraph_node *referring = dyn_cast<cgraph_node *> (ref->referring); + /* Only add function aliases. + Varpool refs are added later in LTO partitioning pass. */ + if (referring) + add_node_to_partition (part, referring); + } + else + { + struct ipa_ref *ref2; + /* We do not need to add transparent aliases if they are not used. + However we must add aliases of transparent aliases if they exist. */ + FOR_EACH_ALIAS (ref->referring, ref2) + { + /* Nested transparent aliases are not permitted. */ + gcc_checking_assert (!ref2->referring->transparent_alias); + cgraph_node *referring = dyn_cast<cgraph_node *> (ref2->referring); + if (referring) + add_node_to_partition (part, referring); + } + } +} + +/* Return TRUE if NODE is in PARTITION. */ +static bool +node_in_partition_p (locality_partition partition, cgraph_node *node) +{ + return ((uintptr_t) (partition->part_id) == (uintptr_t) (node->aux)); +} + +/* Helper function for qsort; to break ties. */ +static int +compare_node_uids (cgraph_node *n1, cgraph_node *n2) +{ + int res = n1->get_uid () - n2->get_uid (); + gcc_assert (res != 0); + return res > 0 ? 1 : -1; +} + +/* Helper function for qsort; sort nodes by order. */ +static int +static_profile_cmp (const void *pa, const void *pb) +{ + const locality_order *a = *static_cast<const locality_order *const *> (pa); + const locality_order *b = *static_cast<const locality_order *const *> (pb); + /* Ascending order. */ + if (b->order < a->order) + return 1; + if (b->order > a->order) + return -1; + return compare_node_uids (a->node, b->node); +} + +/* Helper function for qsort; sort nodes by profile count. */ +static int +compare_edge_profile_counts (const void *pa, const void *pb) +{ + const locality_order *a = *static_cast<const locality_order *const *> (pa); + const locality_order *b = *static_cast<const locality_order *const *> (pb); + + profile_count cnt1 = a->node->count.ipa (); + profile_count cnt2 = b->node->count.ipa (); + if (!cnt1.compatible_p (cnt2)) + return static_profile_cmp (pa, pb); + + if (cnt1 < cnt2) + return 1; + if (cnt1 > cnt2) + return -1; + return static_profile_cmp (pa, pb); +} + +/* Create and return a new partition and increment NPARTITIONS. */ + +static locality_partition +create_partition (int &npartitions) +{ + locality_partition part = XCNEW (struct locality_partition_def); + npartitions++; + part->part_id = npartitions; + part->nodes.create (1); + part->insns = 0; + locality_partitions.safe_push (part); + return part; +} + +/* Structure for holding profile count information of callers of a node. */ +struct profile_stats +{ + /* Sum of non-recursive call counts. */ + profile_count nonrec_count; + + /* Sum of recursive call counts. */ + profile_count rec_count; + + /* If non-NULL, this node is the target of alias or thunk and calls from this + should be count in rec_count. */ + cgraph_node *target; +}; + +/* Initialize fields of STATS. */ +static inline void +init_profile_stats (profile_stats *stats, cgraph_node *target = NULL) +{ + stats->nonrec_count = profile_count::zero (); + stats->rec_count = profile_count::zero (); + stats->target = target; +} + +/* Helper function of to accumulate call counts. */ +static bool +accumulate_profile_counts_after_cloning (cgraph_node *node, void *data) +{ + struct profile_stats *stats = (struct profile_stats *) data; + for (cgraph_edge *e = node->callers; e; e = e->next_caller) + { + if (!e->count.initialized_p ()) + continue; + + if (e->caller == stats->target) + stats->rec_count += e->count.ipa (); + else + stats->nonrec_count += e->count.ipa (); + } + return false; +} + +/* NEW_NODE is a previously created clone of ORIG_NODE already present in + current partition. EDGES contains newly redirected edges to NEW_NODE. + Adjust profile information for both nodes and the edge. */ + +static void +adjust_profile_info_for_non_self_rec_edges (auto_vec<cgraph_edge *> &edges, + cgraph_node *new_node, + cgraph_node *orig_node) +{ + profile_count orig_node_count = orig_node->count.ipa (); + profile_count edge_count = profile_count::zero (); + profile_count final_new_count = profile_count::zero (); + profile_count final_orig_count = profile_count::zero (); + + for (unsigned i = 0; i < edges.length (); ++i) + if (edges[i]->count.initialized_p ()) + edge_count += edges[i]->count.ipa (); + + final_orig_count = orig_node_count - edge_count; + + /* NEW_NODE->count was adjusted for other callers when the clone was + first created. Just add the new edge count. */ + final_new_count = new_node->count + edge_count; + + final_new_count = orig_node_count.combine_with_ipa_count (final_new_count); + orig_node->count = final_orig_count; + new_node->count = final_new_count; + + if (dump_file) + { + fprintf (dump_file, "Adjusting profile information for %s\n", + new_node->dump_asm_name ()); + fprintf (dump_file, "\tOriginal node %s\n", orig_node->dump_asm_name ()); + fprintf (dump_file, "\tOriginal count: "); + orig_node_count.dump (dump_file); + fprintf (dump_file, "\n\tAdjusted original count to: "); + final_orig_count.dump (dump_file); + fprintf (dump_file, "\n\tAdjusted clone count to: "); + final_new_count.dump (dump_file); + fprintf (dump_file, "\n"); + } + + /* Scale all callee edges according to adjusted counts. */ + profile_count orig_node_count_copy = orig_node_count; + profile_count::adjust_for_ipa_scaling (&final_new_count, + &orig_node_count_copy); + for (cgraph_edge *cs = new_node->callees; cs; cs = cs->next_callee) + cs->count = cs->count.apply_scale (final_new_count, orig_node_count_copy); + for (cgraph_edge *cs = new_node->indirect_calls; cs; cs = cs->next_callee) + cs->count = cs->count.apply_scale (final_new_count, orig_node_count_copy); + + profile_count::adjust_for_ipa_scaling (&final_orig_count, &orig_node_count); + for (cgraph_edge *cs = orig_node->callees; cs; cs = cs->next_callee) + cs->count = cs->count.apply_scale (final_orig_count, orig_node_count); + for (cgraph_edge *cs = orig_node->indirect_calls; cs; cs = cs->next_callee) + cs->count = cs->count.apply_scale (final_orig_count, orig_node_count); +} + +/* Adjust profile counts of NEW_NODE and ORIG_NODE, where NEW_NODE is a clone + of OLD_NODE. + Assumes that all eligible edges from current partition so far are redirected + to NEW_NODE and recursive edges are adjusted. */ + +static void +adjust_profile_info (cgraph_node *new_node, cgraph_node *orig_node) +{ + /* If all calls to NEW_NODE are non-recursive, subtract corresponding count + from ORIG_NODE and assign to NEW_NODE, any unexpected remainder stays with + ORIG_NODE. + Recursive calls if present, likely contribute to majority of count; + scale according to redirected callers' count. */ + + profile_count orig_node_count = orig_node->count.ipa (); + profile_stats new_stats, orig_stats; + + init_profile_stats (&new_stats); + init_profile_stats (&orig_stats); + + new_node->call_for_symbol_thunks_and_aliases + (accumulate_profile_counts_after_cloning, &new_stats, false); + orig_node->call_for_symbol_thunks_and_aliases + (accumulate_profile_counts_after_cloning, &orig_stats, false); + + profile_count orig_nonrec_count = orig_stats.nonrec_count; + profile_count orig_rec_count = orig_stats.rec_count; + profile_count new_nonrec_count = new_stats.nonrec_count; + profile_count new_rec_count = new_stats.rec_count; + + profile_count final_new_count = new_nonrec_count; + profile_count final_orig_count = profile_count::zero (); + + /* All calls to NEW_NODE are non-recursive or recursive calls have + zero count. */ + if (!new_rec_count.nonzero_p ()) + final_orig_count = orig_node_count - new_nonrec_count; + else + { + /* If ORIG_NODE is externally visible, indirect calls or calls from + another part of the code may contribute to the count. + update_profiling_info () from ipa-cp.cc pretends to have an extra + caller to represent the extra counts. */ + if (!orig_node->local) + { + profile_count pretend_count = (orig_node_count - new_nonrec_count - + orig_nonrec_count - orig_rec_count); + orig_nonrec_count += pretend_count; + } + + /* Remaining rec_count is assigned in proportion to clone's non-recursive + count. */ + profile_count rec_count = orig_node_count - new_nonrec_count + - orig_nonrec_count; + profile_count new_rec_scaled + = rec_count.apply_scale (new_nonrec_count, + new_nonrec_count + orig_nonrec_count); + final_new_count += new_rec_scaled; + final_orig_count = orig_node_count - final_new_count; + } + + final_new_count = orig_node_count.combine_with_ipa_count (final_new_count); + new_node->count = final_new_count; + orig_node->count = final_orig_count; + + if (dump_file) + { + fprintf (dump_file, "Adjusting profile information for %s\n", + new_node->dump_asm_name ()); + fprintf (dump_file, "\tOriginal node %s\n", orig_node->dump_asm_name ()); + fprintf (dump_file, "\tOriginal count: "); + orig_node_count.dump (dump_file); + fprintf (dump_file, "\n\tAdjusted original count to: "); + final_orig_count.dump (dump_file); + fprintf (dump_file, "\n\tAdjusted clone count to: "); + final_new_count.dump (dump_file); + fprintf (dump_file, "\n"); + } + + /* Scale all callee edges according to adjusted counts. */ + profile_count orig_node_count_copy = orig_node_count; + profile_count::adjust_for_ipa_scaling (&final_new_count, + &orig_node_count_copy); + for (cgraph_edge *cs = new_node->callees; cs; cs = cs->next_callee) + cs->count = cs->count.apply_scale (final_new_count, orig_node_count_copy); + for (cgraph_edge *cs = new_node->indirect_calls; cs; cs = cs->next_callee) + cs->count = cs->count.apply_scale (final_new_count, orig_node_count_copy); + + profile_count::adjust_for_ipa_scaling (&final_orig_count, &orig_node_count); + for (cgraph_edge *cs = orig_node->callees; cs; cs = cs->next_callee) + cs->count = cs->count.apply_scale (final_orig_count, orig_node_count); + for (cgraph_edge *cs = orig_node->indirect_calls; cs; cs = cs->next_callee) + cs->count = cs->count.apply_scale (final_orig_count, orig_node_count); +} + +/* Return true if EDGE can be safely redirected to another callee. */ +static inline bool +edge_redirectable_p (cgraph_edge *edge, lto_locality_cloning_model cm) +{ + if (cm == LTO_LOCALITY_NON_INTERPOSABLE_CLONING) + { + /* Interposability may change on edge basis. */ + enum availability avail; + avail = edge->callee->get_availability (edge->caller); + if (avail <= AVAIL_INTERPOSABLE) + return false; + } + return true; +} + +/* Create a locality clone of CNODE and redirect all callers present in + PARTITION. + Create a clone dpending on whether CNODE itself is a clone or not. */ + +static cgraph_node * +create_locality_clone (cgraph_node *cnode, + locality_partition partition, int &cl_num, + lto_locality_cloning_model cm) +{ + cgraph_node *cl_node = NULL; + vec<cgraph_edge *> redirect_callers = vNULL; + /* All callers of cnode in current partition are redirected. */ + struct cgraph_edge *edge; + for (edge = cnode->callers; edge; edge = edge->next_caller) + { + struct cgraph_node *caller = edge->caller; + if (node_in_partition_p (partition, caller) && caller->definition + && caller != cnode && edge_redirectable_p (edge, cm)) + redirect_callers.safe_push (edge); + } + + const char *suffix = "locality_clone"; + + tree old_decl = cnode->decl; + tree new_decl = copy_node (old_decl); + + /* Generate a new name for the new version. */ + const char *name = IDENTIFIER_POINTER (DECL_NAME (old_decl)); + DECL_NAME (new_decl) = clone_function_name (name, suffix, cl_num); + SET_DECL_ASSEMBLER_NAME (new_decl, + clone_function_name (old_decl, suffix, cl_num)); + cl_num++; + if (dump_file) + fprintf (dump_file, "\tNew name %s\n", + IDENTIFIER_POINTER (DECL_ASSEMBLER_NAME (new_decl))); + + cl_node = cnode->create_clone (new_decl, cnode->count /*profile_count*/, + false /*update_original*/, redirect_callers, + false /*call_duplication_hook*/, + NULL /*new_inlined_to*/, + NULL /*param_adjustments*/, suffix); + + set_new_clone_decl_and_node_flags (cl_node); + + if (cnode->ipa_transforms_to_apply.exists ()) + cl_node->ipa_transforms_to_apply + = cnode->ipa_transforms_to_apply.copy (); + + if (dump_file) + { + fprintf (dump_file, "Cloned Node: %s %s\n", cnode->dump_asm_name (), + cl_node->dump_asm_name ()); + + for (edge = cl_node->callers; edge; edge = edge->next_caller) + fprintf (dump_file, "Redirected callers: %s\n", + edge->caller->dump_asm_name ()); + + for (edge = cl_node->callees; edge; edge = edge->next_callee) + fprintf (dump_file, "Callees of clone: %s %d\n", + edge->callee->dump_asm_name (), edge->frequency ()); + } + return cl_node; +} + +/* Redirect recursive edges of CLONE to correctly point to CLONE. As part of + cloning process, all callee edges of a node are just duplicated but not + redirected. Therefore, these edges still call to original of CLONE. + + For non-inlined CLONEs, NEW_CALLEE == CLONE and ORIG_CALLEE is CLONE's + original node. + + For inlined node, self recursion to CLONE's original same as non-inlined, + additionally, calls to CLONE->inlined_to are also recursive: + NEW_CALLEE == CLONE->inlined_into and + ORIG_CALLEE == original node of CLONE->inlined_into. */ + +static void +adjust_recursive_callees (cgraph_node *clone, cgraph_node *new_callee, + cgraph_node *orig_callee) +{ + cgraph_node *alias = NULL; + for (cgraph_edge *e = clone->callees; e; e = e->next_callee) + { + if (!e->inline_failed) + continue; + + /* Only self-cycle or local alias are handled. */ + cgraph_node *callee = e->callee; + if (callee == orig_callee) + { + cgraph_node **cl = node_to_clone.get (orig_callee); + gcc_assert (cl && *cl == new_callee); + e->redirect_callee_duplicating_thunks (new_callee); + if (dump_file) + fprintf (dump_file, "recursive call from %s to %s orig %s\n", + e->caller->dump_asm_name (), e->callee->dump_asm_name (), + callee->dump_asm_name ()); + } + else if (callee->alias + && e->callee->ultimate_alias_target () == orig_callee) + { + if (!alias) + { + alias = dyn_cast<cgraph_node *> ( + new_callee->noninterposable_alias ()); + } + e->redirect_callee_duplicating_thunks (alias); + if (dump_file) + fprintf (dump_file, "recursive call from %s to %s orig %s\n", + e->caller->dump_asm_name (), e->callee->dump_asm_name (), + callee->dump_asm_name ()); + } + } + new_callee->expand_all_artificial_thunks (); + if (alias) + alias->expand_all_artificial_thunks (); +} + +/* Create clones for CALLER's inlined callees, ORIG_INLINED_TO is the original + node from clone_as_needed () such that new_inlined_to is a clone of it. */ + +static void +inline_clones (cgraph_node *caller, cgraph_node *orig_inlined_to) +{ + struct cgraph_edge *edge; + for (edge = caller->callees; edge; edge = edge->next_callee) + { + struct cgraph_node *callee = edge->callee; + if (edge->inline_failed) + continue; + + if (callee->inlined_to != orig_inlined_to) + continue; + + struct cgraph_node *new_inlined_to, *cl; + if (caller->inlined_to) + new_inlined_to = caller->inlined_to; + else + new_inlined_to = caller; + + cl = callee->create_clone (callee->decl, + edge->count /*profile_count*/, + true /*update_original*/, + vNULL /*redirect_callers*/, + false /*call_duplication_hook*/, + new_inlined_to /*new_inlined_to*/, + NULL /*param_adjustments*/, + "locality_clone" /*suffix*/); + edge->redirect_callee (cl); + + node_to_clone.put (callee, cl); + clone_to_node.put (cl, callee); + + if (callee->thunk) + { + thunk_info *info = thunk_info::get (callee); + *thunk_info::get_create (cl) = *info; + } + + adjust_recursive_callees (cl, new_inlined_to, orig_inlined_to); + adjust_recursive_callees (cl, cl, callee); + if (dump_file) + { + fprintf (dump_file, "Inline cloned\n"); + cl->dump (dump_file); + } + + /* Recursively inline till end of this callchain. */ + inline_clones (cl, orig_inlined_to); + } +} + +/* Clone EDGE->CALLEE if it or a clone of it is not already in PARTITION. + Redirect all callers of EDGE->CALLEE that are in PARTITION, not just the + EDGE. If a clone is already present in PARTITION, redirect all edges from + EDGE->CALLER to EDGE->CALLEE. This is because we only visit one edge per + caller to callee and redirect for all others from there. + + If cloning, also recursively clone inlined functions till the end of the + callchain because inlined clones have 1-1 exclusive copy and edge from + caller to inlined node. + + There are 2 flows possible: + 1. Only redirect + 1.1. cnode is already in current partition - cnode mustn't be a + locality_clone -> nothing to do + 1.2. A clone of cnode is in current partition - find out if it's the + correct clone for edge - must be a locality_clone but the exact same + kind as callee i.e. orig or cp/sra clone, if yes, redirect, else go to #2 + 1.3. Cnode/a clone of cnode is in current partition but caller is inlined + 2. Clone and redirect + 2.1. cnode is original node + 2.2. cnode itself is a clone + Clone inlines + Flavors of edges: + 1. Normal -> orig nodes, locality clones or cp/sra clones + 2. Recursive -> direct recursion + 3. Alias -> recursion via aliasing or as a result of IPA code duplication + 4. Inline -> shouldn't be included in callchain. */ + +static cgraph_node * +clone_node_as_needed (cgraph_edge *edge, locality_partition partition, + int &cl_num, lto_locality_cloning_model cm) +{ + /* suitable_for_locality_cloning_p () currently prohibits cloning aliases due + to potential versioning and materialization issues. Could be enabled in + the future. suitable_for_locality_cloning_p () also checks for + interposability for CNODE but not for edge redirection. */ + struct cgraph_node *cnode = edge->callee; + struct cgraph_node *caller = edge->caller; + + /* If clone of cnode is already in the partition + Get latest clone of cnode. If current partition has cloned cnode, that + clone should be returned. Otherwise, clone from previous partition is + returned + Original node and its clone shouldn't co-exist in current partition + + This is required if callee is partitioned via another edge before caller + was, and we are now visiting caller->callee edge + + 1) a -> b ==> a -> bc1; b was cloned say via d -> bc1, a is orig + 2) ac1 -> b ==> ac1 -> bc1; b was cloned and a was just cloned + 3) a -> bc1 and bc2 present, mustn't happen, b was cloned and a was + redirected without being partitioned first. + Why will we do this again - multiple edges and something's wrong in + partition_callchain () + 4) ac1 -> bc1 ==> ac1 -> bc2; a was cloned and we already got (1) in some + other partition + 5) ac1 -> bc1 but no clone present in this PARTITION. Create from b, not + from bc1? + 6) a -> b; a -> bc0; create new clone, no clone present + 7) ac0 -> b; ac0 -> bc0 same as (6) + 8) a -> bc0 and no clone present, mustn't happen, same as (3) + + Redirect when bc1 is present and: + a -> b or ac -> b or ac -> bc0 */ + + cgraph_node *orig_cnode = cnode; + cgraph_node **o_cnode = clone_to_node.get (cnode); + if (o_cnode) + orig_cnode = *o_cnode; + + cgraph_node **cnode_cl = node_to_clone.get (orig_cnode); + + if (cnode_cl && node_in_partition_p (partition, *cnode_cl)) + { + if (node_in_partition_p (partition, caller)) + { + bool clone_p = false; + auto_vec<cgraph_edge *> redirected_edges; + for (cgraph_edge *ec = caller->callees; ec; ec = ec->next_callee) + if (ec->callee == cnode && edge_redirectable_p (ec, cm)) + { + ec->redirect_callee_duplicating_thunks (*cnode_cl); + clone_p = true; + redirected_edges.safe_push (ec); + if (dump_file) + { + fprintf (dump_file, "clone present %s %s redirecting %s\n", + cnode->dump_asm_name (), + (*cnode_cl)->dump_asm_name (), + caller->dump_asm_name ()); + } + } + if (clone_p) + { + (*cnode_cl)->expand_all_artificial_thunks (); + adjust_profile_info_for_non_self_rec_edges (redirected_edges, + *cnode_cl, cnode); + return NULL; + } + } + } + + /* Create a new clone for a -> b, ac -> b. + For ac -> bc, should be done on bc or b? + bc could be from b_cp/b_sra or b. */ + + if (orig_cnode != cnode) + { + if (dump_file) + fprintf (dump_file, "Clone of clone %s %s\n", cnode->dump_asm_name (), + orig_cnode->dump_asm_name ()); + return NULL; + } + + struct cgraph_node *cloned_node + = create_locality_clone (cnode, partition, cl_num, cm); + + gcc_assert (cloned_node); + if (!cloned_node) + return NULL; + + node_to_clone.put (cnode, cloned_node); + clone_to_node.put (cloned_node, cnode); + + adjust_recursive_callees (cloned_node, cloned_node, cnode); + symtab->call_cgraph_duplication_hooks (cnode, cloned_node); + + adjust_profile_info (cloned_node, cnode); + /* Inline clones are created iff their inlined_to == CNODE. */ + inline_clones (cloned_node, cnode); + + return cloned_node; +} + +/* Accumulate frequency of all edges from EDGE->caller to EDGE->callee. */ + +static sreal +accumulate_incoming_edge_frequency (cgraph_edge *edge) +{ + sreal count = 0; + struct cgraph_edge *e; + for (e = edge->callee->callers; e; e = e->next_caller) + { + /* Make a local decision about all edges for EDGE->caller but not the + other nodes already in the partition. Their edges will be visited + later or may have been visited before and not fit the + cut-off criteria. */ + if (e->caller == edge->caller) + count += e->sreal_frequency (); + } + return count; +} + +/* Determine if EDGE->CALLEE is suitable for cloning. It is assummed that the + callee is not an inlined node. */ + +static bool +suitable_for_locality_cloning_p (cgraph_edge *edge, + lto_locality_cloning_model cm) +{ + cgraph_node *node = edge->callee; + if (!node->versionable) + return false; + + /* Out-of-line locality clones of ipcp or sra clones will be created in this + pass after IPA inline is run. A locality clone has the same function + body and the same updated signature as the ipcp/sra clone. + This fails or asserts based on how the clone is created: + 1. If param_adjustments and tree_map are not recorded for locality clone: + clone materialization (tree_function_versioning ()) fails when + updating signature and remapping calls because clone_of (ipcp/sra + clone) and locality clone differ in param information. + 2. If param_adjustments and tree_map are provided: asserts are triggered + in fnsummary duplication because IPA inline resets some summaries. + + One inelegant solution is to provide param_adjustments and tree_map, and + then set clone_of to ipcp/sra clone's clone_of. However, this sometimes + results in segmentation fault when the compiled program is run. + Disabling clone of clones altogether for now with an aim to resolve this + is future. */ + if (node->clone_of) + return false; + + if (node->alias) + return false; + + if (edge->recursive_p ()) + return false; + + if (!node->definition) + return false; + + /* Don't clone NODE if IPA count of NODE or EDGE is zero. */ + if (!node->count.ipa ().nonzero_p () || !edge->count.ipa ().nonzero_p ()) + return false; + + if (cm == LTO_LOCALITY_NON_INTERPOSABLE_CLONING) + { + /* Interposability may change on edge basis. */ + enum availability avail; + edge->callee->ultimate_alias_target (&avail, edge->caller); + if (avail <= AVAIL_INTERPOSABLE) + return false; + } + + return true; +} + +/* Map from caller to all callees already visited for partitioning. */ +hash_map<cgraph_node *, auto_vec<cgraph_node *> > caller_to_callees; + +/* Partition EDGE->CALLEE into PARTITION or clone if already partitioned and + satisfies cloning criteria such as CLONING_MODEL, REAL_FREQ and SIZE + cut-offs and CLONE_FURTHER_P set by previous caller. */ + +/* callgraph can have multiple caller to callee edges for multiple callsites + For the first such edge, we make decisions about cutoffs and cloning because + we redirect ALL callsites to cloned callee, not just one of them. */ + +static void +partition_callchain (cgraph_edge *edge, locality_partition partition, + bool clone_further_p, + lto_locality_cloning_model cloning_model, + double freq_cutoff, int size, int &cl_num) +{ + /* Aliases are added in the same partition as their targets. + Aliases are not cloned and their callees are not processed separately. */ + cgraph_node *node = edge->callee->ultimate_alias_target (); + cgraph_node *caller = edge->caller; + cgraph_node *caller_node = node, *cl_node = NULL; + + /* Already visited the caller to callee edges. */ + auto_vec<cgraph_node *> &callees = caller_to_callees.get_or_insert (caller); + if (std::find (callees.begin (), callees.end (), node) != callees.end ()) + return; + + callees.safe_push (node); + + if (node->get_partitioning_class () == SYMBOL_PARTITION) + { + if (!node_partitioned_p (node)) + { + add_node_to_partition (partition, node); + if (dump_file) + fprintf (dump_file, "Partitioned node: %s\n", + node->dump_asm_name ()); + } + else if (cloning_model >= LTO_LOCALITY_NON_INTERPOSABLE_CLONING + && !node_in_partition_p (partition, node)) + { + /* Non-inlined node, or alias, already partitioned + If cut-off, don't clone callees but partition unpartitioned + callees. + size is node + inlined nodes. */ + if (clone_further_p) + { + if (!node->alias) + if (ipa_size_summaries->get (node)->size >= size) + clone_further_p = false; + + if (freq_cutoff != 0.0) + { + sreal acc_freq = accumulate_incoming_edge_frequency (edge); + if (acc_freq.to_double () < freq_cutoff) + clone_further_p = false; + } + } + + if (!suitable_for_locality_cloning_p (edge, cloning_model)) + clone_further_p = false; + + if (clone_further_p) + { + /* Try to clone NODE and its inline chain. */ + if (dump_file) + fprintf (dump_file, "Cloning node: %s\n", + node->dump_asm_name ()); + cl_node = clone_node_as_needed (edge, partition, cl_num, + cloning_model); + if (cl_node) + { + add_node_to_partition (partition, cl_node); + caller_node = cl_node; + } + else + caller_node = NULL; + } + } + } + else if (!node->inlined_to) + return; + + if (caller_node) + for (cgraph_edge *e = caller_node->callees; e; e = e->next_callee) + partition_callchain (e, partition, clone_further_p, cloning_model, + freq_cutoff, size, cl_num); +} + +/* Determine whether NODE is an entrypoint to a callchain. */ + +static bool +is_entry_node_p (cgraph_node *node) +{ + /* node->inlined_to is returned as SYMBOL_DUPLICATE. */ + if (node->get_partitioning_class () != SYMBOL_PARTITION) + return false; + + if (!node->callers) + return true; + + for (cgraph_edge *e = node->callers; e; e = e->next_caller) + { + if (! e->recursive_p ()) + return false; + } + if (node->alias + && !is_entry_node_p (node->ultimate_alias_target ())) + return false; + return true; +} + +/* Determine order of all external nodes if PGO profile is available. + Store the order in ORDER. */ + +static bool +locality_determine_ipa_order (auto_vec<locality_order *> *order) +{ + struct cgraph_node *node; + auto_vec<locality_order *> non_comparable_nodes; + FOR_EACH_DEFINED_FUNCTION (node) + if (node->get_partitioning_class () == SYMBOL_PARTITION) + { + if (node->no_reorder) + { + if (dump_file) + fprintf (dump_file, "no reorder %s\n", node->dump_asm_name ()); + return false; + } + else if (is_entry_node_p (node)) + { + profile_count pcnt = node->count.ipa (); + if (!pcnt.initialized_p () || !pcnt.ipa_p ()) + { + sreal cnt = 0; + locality_order *lo = new locality_order (node, cnt); + non_comparable_nodes.safe_push (lo); + continue; + } + sreal count = 0; + struct cgraph_edge *edge; + for (edge = node->callees; edge; edge = edge->next_callee) + { + /* For PGO, frequency is not used in + compare_edge_profile_counts (), it's used only as part of + static profile order. */ + sreal freq = edge->sreal_frequency (); + count += freq; + } + locality_order *cl = new locality_order (node, count); + order->safe_push (cl); + } + } + order->qsort (compare_edge_profile_counts); + for (auto el : non_comparable_nodes) + order->safe_push (el); + return true; +} + +/* Determine order of all external nodes if only static profile is available. + Store the order in ORDER. */ + +static bool +locality_determine_static_order (auto_vec<locality_order *> *order) +{ + struct cgraph_node *node; + FOR_EACH_DEFINED_FUNCTION (node) + if (node->get_partitioning_class () == SYMBOL_PARTITION) + { + if (node->no_reorder) + { + if (dump_file) + fprintf (dump_file, "no reorder %s\n", node->dump_asm_name ()); + return false; + } + else if (is_entry_node_p (node)) + { + sreal count = 0; + struct cgraph_edge *edge; + for (edge = node->callees; edge; edge = edge->next_callee) + { + sreal freq = edge->sreal_frequency (); + count += freq; + } + locality_order *cl = new locality_order (node, count); + order->safe_push (cl); + } + } + order->qsort (static_profile_cmp); + return true; +} + +/* Partitioning for code locality. + 1. Create and sort callchains. If PGO is available, use real profile + counts. Otherwise, use a set of heuristics to sort the callchains. + 2. Partition the external nodes and their callchains in the determined order + 2.1. If !partition, partition, else try and clone if it satisfies cloning + criteria. + 3. Partition all other unpartitioned nodes. */ + +static void +locality_partition_and_clone (int max_locality_partition_size, + lto_locality_cloning_model cloning_model, + int freq_denominator, int size) +{ + locality_partition partition; + int npartitions = 0; + + auto_vec<locality_order *> order; + auto_vec<varpool_node *> varpool_order; + struct cgraph_node *node; + bool order_p; + + int cl_num = 0; + + double real_freq = 0.0; + if (freq_denominator > 0) + real_freq = 1.0 / (double) freq_denominator; + + cgraph_node *n = symtab->first_defined_function (); + if (n && n->count.ipa_p ()) + order_p = locality_determine_ipa_order (&order); + else + order_p = locality_determine_static_order (&order); + if (!order_p) + { + if (dump_file) + { + fprintf (dump_file, "Locality partition: falling back to balanced" + "model\n"); + } + + return; + } + + int64_t partition_size + = max_locality_partition_size + ? max_locality_partition_size : param_max_partition_size; + partition = create_partition (npartitions); + + for (unsigned i = 0; i < order.length (); i++) + { + node = order[i]->node; + if (node_partitioned_p (node)) + continue; + + if (partition->insns > partition_size) + partition = create_partition (npartitions); + if (dump_file) + fprintf (dump_file, "Partition id: %d\n", partition->part_id); + + add_node_to_partition (partition, node); + if (dump_file) + fprintf (dump_file, "Ordered Node: %s\n", node->dump_asm_name ()); + + for (cgraph_edge *edge = node->callees; edge; edge = edge->next_callee) + { + /* Recursively partition the callchain of edge->callee. */ + partition_callchain (edge, partition, true, cloning_model, real_freq, + size, cl_num); + } + } + + for (unsigned i = 0; i < order.length (); i++) + delete order[i]; + order = vNULL; +} + +/* Entry point to locality-clone pass. */ +static int +lc_execute (void) +{ + symtab_node *node; + FOR_EACH_SYMBOL (node) + node->aux = NULL; + + locality_partition_and_clone (param_max_locality_partition_size, + flag_lto_locality_cloning, + param_lto_locality_frequency, + param_lto_locality_size); + + FOR_EACH_SYMBOL (node) + node->aux = NULL; + return 0; +} + +namespace { + +const pass_data pass_data_ipa_locality_clone = { + IPA_PASS, /* type */ + "locality-clone", /* name */ + OPTGROUP_NONE, /* optinfo_flags */ + TV_IPA_LC, /* tv_id */ + 0, /* properties_required */ + 0, /* properties_provided */ + 0, /* properties_destroyed */ + 0, /* todo_flags_start */ + (TODO_dump_symtab | TODO_remove_functions), /* todo_flags_finish */ +}; + +class pass_ipa_locality_cloning : public ipa_opt_pass_d +{ +public: + pass_ipa_locality_cloning (gcc::context *ctxt) + : ipa_opt_pass_d (pass_data_ipa_locality_clone, ctxt, + NULL, /* generate_summary */ + NULL, /* write_summary */ + NULL, /* read_summary */ + NULL, /* write_optimization_summary */ + NULL, /* read_optimization_summary */ + NULL, /* stmt_fixup */ + 0, /* function_transform_todo_flags_start */ + NULL, /* function_transform */ + NULL) /* variable_transform */ + {} + + /* opt_pass methods: */ + virtual bool gate (function *) + { + return (flag_wpa && flag_ipa_reorder_for_locality); + } + + virtual unsigned int execute (function *) { return lc_execute (); } + +}; // class pass_ipa_locality_cloning + +} // namespace + +ipa_opt_pass_d * +make_pass_ipa_locality_cloning (gcc::context *ctxt) +{ + return new pass_ipa_locality_cloning (ctxt); +} diff --git a/gcc/ipa-locality-cloning.h b/gcc/ipa-locality-cloning.h new file mode 100644 index 0000000..591ce57 --- /dev/null +++ b/gcc/ipa-locality-cloning.h @@ -0,0 +1,35 @@ +/* LTO partitioning logic routines. + Copyright The GNU Toolchain Authors + +This file is part of GCC. + +GCC is free software; you can redistribute it and/or modify it under +the terms of the GNU General Public License as published by the Free +Software Foundation; either version 3, or (at your option) any later +version. + +GCC is distributed in the hope that it will be useful, but WITHOUT ANY +WARRANTY; without even the implied warranty of MERCHANTABILITY or +FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License +for more details. + +You should have received a copy of the GNU General Public License +along with GCC; see the file COPYING3. If not see +<http://www.gnu.org/licenses/>. */ + +#ifndef IPA_LOCALITY_CLONING_H +#define IPA_LOCALITY_CLONING_H + +/* Structure describing locality partitions. */ +struct locality_partition_def +{ + int part_id; + vec<cgraph_node *> nodes; + int insns; +}; + +typedef struct locality_partition_def *locality_partition; + +extern vec<locality_partition> locality_partitions; + +#endif /* IPA_LOCALITY_CLONING_H */ diff --git a/gcc/ipa-prop.cc b/gcc/ipa-prop.cc index a120f94..0398d69 100644 --- a/gcc/ipa-prop.cc +++ b/gcc/ipa-prop.cc @@ -60,6 +60,7 @@ along with GCC; see the file COPYING3. If not see #include "gimple-range.h" #include "value-range-storage.h" #include "vr-values.h" +#include "lto-streamer.h" /* Function summary where the parameter infos are actually stored. */ ipa_node_params_t *ipa_node_params_sum = NULL; @@ -454,7 +455,11 @@ ipa_dump_jump_function (FILE *f, ipa_jump_func *jump_func, if (jump_func->value.pass_through.operation != NOP_EXPR) { fprintf (f, " "); - print_generic_expr (f, jump_func->value.pass_through.operand); + if (jump_func->value.pass_through.operand) + print_generic_expr (f, jump_func->value.pass_through.operand); + fprintf (f, " (in type "); + print_generic_expr (f, jump_func->value.pass_through.op_type); + fprintf (f, ")"); } if (jump_func->value.pass_through.agg_preserved) fprintf (f, ", agg_preserved"); @@ -510,7 +515,11 @@ ipa_dump_jump_function (FILE *f, ipa_jump_func *jump_func, if (item->value.pass_through.operation != NOP_EXPR) { fprintf (f, " "); - print_generic_expr (f, item->value.pass_through.operand); + if (item->value.pass_through.operand) + print_generic_expr (f, item->value.pass_through.operand); + fprintf (f, " (in type "); + print_generic_expr (f, jump_func->value.pass_through.op_type); + fprintf (f, ")"); } } else if (item->jftype == IPA_JF_CONST) @@ -682,6 +691,7 @@ ipa_set_jf_simple_pass_through (struct ipa_jump_func *jfunc, int formal_id, { jfunc->type = IPA_JF_PASS_THROUGH; jfunc->value.pass_through.operand = NULL_TREE; + jfunc->value.pass_through.op_type = NULL_TREE; jfunc->value.pass_through.formal_id = formal_id; jfunc->value.pass_through.operation = NOP_EXPR; jfunc->value.pass_through.agg_preserved = agg_preserved; @@ -692,10 +702,11 @@ ipa_set_jf_simple_pass_through (struct ipa_jump_func *jfunc, int formal_id, static void ipa_set_jf_unary_pass_through (struct ipa_jump_func *jfunc, int formal_id, - enum tree_code operation) + enum tree_code operation, tree op_type) { jfunc->type = IPA_JF_PASS_THROUGH; jfunc->value.pass_through.operand = NULL_TREE; + jfunc->value.pass_through.op_type = op_type; jfunc->value.pass_through.formal_id = formal_id; jfunc->value.pass_through.operation = operation; jfunc->value.pass_through.agg_preserved = false; @@ -705,10 +716,12 @@ ipa_set_jf_unary_pass_through (struct ipa_jump_func *jfunc, int formal_id, static void ipa_set_jf_arith_pass_through (struct ipa_jump_func *jfunc, int formal_id, - tree operand, enum tree_code operation) + tree operand, enum tree_code operation, + tree op_type) { jfunc->type = IPA_JF_PASS_THROUGH; jfunc->value.pass_through.operand = unshare_expr_without_location (operand); + jfunc->value.pass_through.op_type = op_type; jfunc->value.pass_through.formal_id = formal_id; jfunc->value.pass_through.operation = operation; jfunc->value.pass_through.agg_preserved = false; @@ -1513,6 +1526,9 @@ compute_complex_assign_jump_func (struct ipa_func_body_info *fbi, if (index >= 0) { + if (lto_variably_modified_type_p (TREE_TYPE (name))) + return; + switch (gimple_assign_rhs_class (stmt)) { case GIMPLE_BINARY_RHS: @@ -1526,7 +1542,8 @@ compute_complex_assign_jump_func (struct ipa_func_body_info *fbi, return; ipa_set_jf_arith_pass_through (jfunc, index, op2, - gimple_assign_rhs_code (stmt)); + gimple_assign_rhs_code (stmt), + TREE_TYPE (name)); break; } case GIMPLE_SINGLE_RHS: @@ -1539,7 +1556,8 @@ compute_complex_assign_jump_func (struct ipa_func_body_info *fbi, case GIMPLE_UNARY_RHS: if (!CONVERT_EXPR_CODE_P (gimple_assign_rhs_code (stmt))) ipa_set_jf_unary_pass_through (jfunc, index, - gimple_assign_rhs_code (stmt)); + gimple_assign_rhs_code (stmt), + TREE_TYPE (name)); default:; } return; @@ -1912,6 +1930,7 @@ analyze_agg_content_value (struct ipa_func_body_info *fbi, if (!is_gimple_assign (stmt)) break; + lhs = gimple_assign_lhs (stmt); rhs1 = gimple_assign_rhs1 (stmt); } @@ -1931,7 +1950,8 @@ analyze_agg_content_value (struct ipa_func_body_info *fbi, PASS-THROUGH jump function with ASSERT_EXPR operation whith operand 1 (the constant from the PHI node). */ - if (gimple_phi_num_args (phi) != 2) + if (gimple_phi_num_args (phi) != 2 + || lto_variably_modified_type_p (TREE_TYPE (lhs))) return; tree arg0 = gimple_phi_arg_def (phi, 0); tree arg1 = gimple_phi_arg_def (phi, 1); @@ -1956,6 +1976,7 @@ analyze_agg_content_value (struct ipa_func_body_info *fbi, code = ASSERT_EXPR; agg_value->pass_through.operand = operand; + agg_value->pass_through.op_type = TREE_TYPE (lhs); } else if (is_gimple_assign (stmt)) { @@ -1980,10 +2001,12 @@ analyze_agg_content_value (struct ipa_func_body_info *fbi, with one operand, here we only allow tc_unary operation to avoid possible problem. Then we can use (opclass == tc_unary) or not to distinguish unary and binary. */ - if (TREE_CODE_CLASS (code) != tcc_unary || CONVERT_EXPR_CODE_P (code)) + if (TREE_CODE_CLASS (code) != tcc_unary || CONVERT_EXPR_CODE_P (code) + || lto_variably_modified_type_p (TREE_TYPE (lhs))) return; rhs1 = get_ssa_def_if_simple_copy (rhs1, &stmt); + agg_value->pass_through.op_type = TREE_TYPE (lhs); break; case GIMPLE_BINARY_RHS: @@ -1992,12 +2015,16 @@ analyze_agg_content_value (struct ipa_func_body_info *fbi, gimple *rhs2_stmt = stmt; tree rhs2 = gimple_assign_rhs2 (stmt); + if (lto_variably_modified_type_p (TREE_TYPE (lhs))) + return; + rhs1 = get_ssa_def_if_simple_copy (rhs1, &rhs1_stmt); rhs2 = get_ssa_def_if_simple_copy (rhs2, &rhs2_stmt); if (is_gimple_ip_invariant (rhs2)) { agg_value->pass_through.operand = rhs2; + agg_value->pass_through.op_type = TREE_TYPE (lhs); stmt = rhs1_stmt; } else if (is_gimple_ip_invariant (rhs1)) @@ -2008,6 +2035,7 @@ analyze_agg_content_value (struct ipa_func_body_info *fbi, return; agg_value->pass_through.operand = rhs1; + agg_value->pass_through.op_type = TREE_TYPE (lhs); stmt = rhs2_stmt; rhs1 = rhs2; } @@ -3520,12 +3548,17 @@ update_jump_functions_after_inlining (struct cgraph_edge *cs, ipa_set_jf_simple_pass_through (dst, formal_id, agg_p); } else if (TREE_CODE_CLASS (operation) == tcc_unary) - ipa_set_jf_unary_pass_through (dst, formal_id, operation); + { + tree op_t = ipa_get_jf_pass_through_op_type (src); + ipa_set_jf_unary_pass_through (dst, formal_id, operation, + op_t); + } else { tree operand = ipa_get_jf_pass_through_operand (src); + tree op_t = ipa_get_jf_pass_through_op_type (src); ipa_set_jf_arith_pass_through (dst, formal_id, operand, - operation); + operation, op_t); } break; } @@ -4935,9 +4968,13 @@ ipa_write_jump_function (struct output_block *ob, } else if (TREE_CODE_CLASS (jump_func->value.pass_through.operation) == tcc_unary) - streamer_write_uhwi (ob, jump_func->value.pass_through.formal_id); + { + stream_write_tree (ob, jump_func->value.pass_through.op_type, true); + streamer_write_uhwi (ob, jump_func->value.pass_through.formal_id); + } else { + stream_write_tree (ob, jump_func->value.pass_through.op_type, true); stream_write_tree (ob, jump_func->value.pass_through.operand, true); streamer_write_uhwi (ob, jump_func->value.pass_through.formal_id); } @@ -4979,6 +5016,8 @@ ipa_write_jump_function (struct output_block *ob, case IPA_JF_LOAD_AGG: streamer_write_uhwi (ob, item->value.pass_through.operation); streamer_write_uhwi (ob, item->value.pass_through.formal_id); + if (item->value.pass_through.operation != NOP_EXPR) + stream_write_tree (ob, item->value.pass_through.op_type, true); if (TREE_CODE_CLASS (item->value.pass_through.operation) != tcc_unary) stream_write_tree (ob, item->value.pass_through.operand, true); @@ -5047,15 +5086,18 @@ ipa_read_jump_function (class lto_input_block *ib, } else if (TREE_CODE_CLASS (operation) == tcc_unary) { + tree op_type = stream_read_tree (ib, data_in); int formal_id = streamer_read_uhwi (ib); - ipa_set_jf_unary_pass_through (jump_func, formal_id, operation); + ipa_set_jf_unary_pass_through (jump_func, formal_id, operation, + op_type); } else { + tree op_type = stream_read_tree (ib, data_in); tree operand = stream_read_tree (ib, data_in); int formal_id = streamer_read_uhwi (ib); ipa_set_jf_arith_pass_through (jump_func, formal_id, operand, - operation); + operation, op_type); } break; case IPA_JF_ANCESTOR: @@ -5103,6 +5145,10 @@ ipa_read_jump_function (class lto_input_block *ib, operation = (enum tree_code) streamer_read_uhwi (ib); item.value.pass_through.operation = operation; item.value.pass_through.formal_id = streamer_read_uhwi (ib); + if (operation != NOP_EXPR) + item.value.pass_through.op_type = stream_read_tree (ib, data_in); + else + item.value.pass_through.op_type = NULL_TREE; if (TREE_CODE_CLASS (operation) == tcc_unary) item.value.pass_through.operand = NULL_TREE; else @@ -5393,6 +5439,49 @@ ipa_read_node_info (class lto_input_block *ib, struct cgraph_node *node, } } +/* Stream out ipa_return_summary. */ +static void +ipa_write_return_summaries (output_block *ob) +{ + if (!ipa_return_value_sum) + { + streamer_write_uhwi (ob, 0); + return; + } + + lto_symtab_encoder_t encoder = ob->decl_state->symtab_node_encoder; + unsigned int count = 0; + for (int i = 0; i < lto_symtab_encoder_size (encoder); i++) + { + symtab_node *snode = lto_symtab_encoder_deref (encoder, i); + cgraph_node *cnode = dyn_cast <cgraph_node *> (snode); + ipa_return_value_summary *v; + + if (cnode && cnode->definition && !cnode->alias + && (v = ipa_return_value_sum->get (cnode)) + && v->vr) + count++; + } + streamer_write_uhwi (ob, count); + + for (int i = 0; i < lto_symtab_encoder_size (encoder); i++) + { + symtab_node *snode = lto_symtab_encoder_deref (encoder, i); + cgraph_node *cnode = dyn_cast <cgraph_node *> (snode); + ipa_return_value_summary *v; + + if (cnode && cnode->definition && !cnode->alias + && (v = ipa_return_value_sum->get (cnode)) + && v->vr) + { + streamer_write_uhwi + (ob, + lto_symtab_encoder_encode (encoder, cnode)); + v->vr->streamer_write (ob); + } + } +} + /* Write jump functions for nodes in SET. */ void @@ -5429,11 +5518,58 @@ ipa_prop_write_jump_functions (void) && ipa_node_params_sum->get (node) != NULL) ipa_write_node_info (ob, node); } - streamer_write_char_stream (ob->main_stream, 0); + ipa_write_return_summaries (ob); produce_asm (ob); destroy_output_block (ob); } +/* Record that return value range of N is VAL. */ + +static void +ipa_record_return_value_range_1 (cgraph_node *n, value_range val) +{ + if (!ipa_return_value_sum) + { + if (!ipa_vr_hash_table) + ipa_vr_hash_table = hash_table<ipa_vr_ggc_hash_traits>::create_ggc (37); + ipa_return_value_sum = new (ggc_alloc_no_dtor <ipa_return_value_sum_t> ()) + ipa_return_value_sum_t (symtab, true); + ipa_return_value_sum->disable_insertion_hook (); + } + ipa_return_value_sum->get_create (n)->vr = ipa_get_value_range (val); + if (dump_file && (dump_flags & TDF_DETAILS)) + { + fprintf (dump_file, "Recording return range of %s:", n->dump_name ()); + val.dump (dump_file); + fprintf (dump_file, "\n"); + } +} + +/* Stream out ipa_return_summary. */ +static void +ipa_read_return_summaries (lto_input_block *ib, + struct lto_file_decl_data *file_data, + class data_in *data_in) +{ + unsigned int f_count = streamer_read_uhwi (ib); + for (unsigned int i = 0; i < f_count; i++) + { + unsigned int index = streamer_read_uhwi (ib); + lto_symtab_encoder_t encoder = file_data->symtab_node_encoder; + struct cgraph_node *node + = dyn_cast <cgraph_node *> + (lto_symtab_encoder_deref (encoder, index)); + ipa_vr rvr; + rvr.streamer_read (ib, data_in); + if (node->prevailing_p ()) + { + value_range tmp; + rvr.get_vrange (tmp); + ipa_record_return_value_range_1 (node, tmp); + } + } +} + /* Read section in file FILE_DATA of length LEN with data DATA. */ static void @@ -5470,6 +5606,7 @@ ipa_prop_read_section (struct lto_file_decl_data *file_data, const char *data, gcc_assert (node->definition); ipa_read_node_info (&ib_main, node, data_in); } + ipa_read_return_summaries (&ib_main, file_data, data_in); lto_free_section_data (file_data, LTO_section_jump_functions, NULL, data, len); lto_data_in_delete (data_in); @@ -5589,6 +5726,7 @@ read_ipcp_transformation_info (lto_input_block *ib, cgraph_node *node, } } + /* Write all aggregate replacement for nodes in set. */ void @@ -5627,7 +5765,7 @@ ipcp_write_transformation_summaries (void) && lto_symtab_encoder_encode_body_p (encoder, cnode)) write_ipcp_transformation_info (ob, cnode, ts); } - streamer_write_char_stream (ob->main_stream, 0); + ipa_write_return_summaries (ob); produce_asm (ob); destroy_output_block (ob); } @@ -5668,6 +5806,7 @@ read_replacements_section (struct lto_file_decl_data *file_data, index)); read_ipcp_transformation_info (&ib_main, node, data_in); } + ipa_read_return_summaries (&ib_main, file_data, data_in); lto_free_section_data (file_data, LTO_section_jump_functions, NULL, data, len); lto_data_in_delete (data_in); @@ -6148,22 +6287,8 @@ ipcp_transform_function (struct cgraph_node *node) void ipa_record_return_value_range (value_range val) { - cgraph_node *n = cgraph_node::get (current_function_decl); - if (!ipa_return_value_sum) - { - if (!ipa_vr_hash_table) - ipa_vr_hash_table = hash_table<ipa_vr_ggc_hash_traits>::create_ggc (37); - ipa_return_value_sum = new (ggc_alloc_no_dtor <ipa_return_value_sum_t> ()) - ipa_return_value_sum_t (symtab, true); - ipa_return_value_sum->disable_insertion_hook (); - } - ipa_return_value_sum->get_create (n)->vr = ipa_get_value_range (val); - if (dump_file && (dump_flags & TDF_DETAILS)) - { - fprintf (dump_file, "Recording return range "); - val.dump (dump_file); - fprintf (dump_file, "\n"); - } + ipa_record_return_value_range_1 + (cgraph_node::get (current_function_decl), val); } /* Return true if value range of DECL is known and if so initialize RANGE. */ @@ -6224,6 +6349,10 @@ ipa_agg_pass_through_jf_equivalent_p (ipa_pass_through_data *ipt1, || ipt1->formal_id != ipt2->formal_id || (!agg_jf && (ipt1->agg_preserved != ipt2->agg_preserved))) return false; + if (ipt1->operation != NOP_EXPR + && (TYPE_MAIN_VARIANT (ipt1->op_type) + != TYPE_MAIN_VARIANT (ipt2->op_type))) + return false; if (((ipt1->operand != NULL_TREE) != (ipt2->operand != NULL_TREE)) || (ipt1->operand && !values_equal_for_ipcp_p (ipt1->operand, ipt2->operand))) diff --git a/gcc/ipa-prop.h b/gcc/ipa-prop.h index 7735b57..3bd442f 100644 --- a/gcc/ipa-prop.h +++ b/gcc/ipa-prop.h @@ -96,6 +96,9 @@ struct GTY(()) ipa_pass_through_data /* If an operation is to be performed on the original parameter, this is the second (constant) operand. */ tree operand; + /* The result type of the operation. In case of no operation (represented by + NOP_EXPR) it should be NULL_TREE. */ + tree op_type; /* Number of the caller's formal parameter being passed. */ int formal_id; /* Operation that is performed on the argument before it is passed on. @@ -387,6 +390,18 @@ ipa_get_jf_pass_through_operand (struct ipa_jump_func *jfunc) return jfunc->value.pass_through.operand; } +/* Return the type of the operation in a non-NOP pass through jmp function + JFUNC. */ + +inline tree +ipa_get_jf_pass_through_op_type (struct ipa_jump_func *jfunc) +{ + gcc_checking_assert (jfunc->type == IPA_JF_PASS_THROUGH + && jfunc->value.pass_through.operation != NOP_EXPR); + + return jfunc->value.pass_through.op_type; +} + /* Return the number of the caller's formal parameter that a pass through jump function JFUNC refers to. */ diff --git a/gcc/lto-cgraph.cc b/gcc/lto-cgraph.cc index ac835a4..8439c51 100644 --- a/gcc/lto-cgraph.cc +++ b/gcc/lto-cgraph.cc @@ -229,6 +229,8 @@ lto_set_symtab_encoder_in_partition (lto_symtab_encoder_t encoder, symtab_node *node) { int index = lto_symtab_encoder_encode (encoder, node); + if (dump_file) + fprintf(dump_file, "Node %s, index %d\n", node->asm_name(), index); encoder->nodes[index].in_partition = true; } diff --git a/gcc/lto-streamer-out.cc b/gcc/lto-streamer-out.cc index d5b6ee7..a055d12d 100644 --- a/gcc/lto-streamer-out.cc +++ b/gcc/lto-streamer-out.cc @@ -130,7 +130,7 @@ destroy_output_block (struct output_block *ob) /* Wrapper around variably_modified_type_p avoiding type modification during WPA streaming. */ -static bool +bool lto_variably_modified_type_p (tree type) { return (in_lto_p diff --git a/gcc/lto-streamer.h b/gcc/lto-streamer.h index ff33bf0..4b7209e3 100644 --- a/gcc/lto-streamer.h +++ b/gcc/lto-streamer.h @@ -906,6 +906,7 @@ void lto_output_decl_state_streams (struct output_block *, void lto_output_decl_state_refs (struct output_block *, struct lto_output_stream *, struct lto_out_decl_state *); +bool lto_variably_modified_type_p (tree); void lto_output_location (struct output_block *, struct bitpack_d *, location_t); void lto_output_location_and_block (struct output_block *, struct bitpack_d *, diff --git a/gcc/lto/ChangeLog b/gcc/lto/ChangeLog index ee53915..4da9ca3 100644 --- a/gcc/lto/ChangeLog +++ b/gcc/lto/ChangeLog @@ -1,3 +1,13 @@ +2025-04-15 Kyrylo Tkachov <ktkachov@nvidia.com> + + * lto-partition.cc (add_node_references_to_partition): Define. + (create_partition): Likewise. + (lto_locality_map): Likewise. + (lto_promote_cross_file_statics): Add extra dumping. + * lto-partition.h (lto_locality_map): Declare prototype. + * lto.cc (do_whole_program_analysis): Handle + flag_ipa_reorder_for_locality. + 2025-02-28 Richard Biener <rguenther@suse.de> PR lto/91299 diff --git a/gcc/lto/lto-partition.cc b/gcc/lto/lto-partition.cc index 3046951..c7e69ee 100644 --- a/gcc/lto/lto-partition.cc +++ b/gcc/lto/lto-partition.cc @@ -37,6 +37,7 @@ along with GCC; see the file COPYING3. If not see #include "ipa-prop.h" #include "ipa-fnsummary.h" #include "lto-partition.h" +#include "ipa-locality-cloning.h" #include <limits> @@ -1418,6 +1419,126 @@ lto_balanced_map (int n_lto_partitions, int max_partition_size) } } +/* Add all references of NODE into PARTITION. */ + +static void +add_node_references_to_partition (ltrans_partition partition, symtab_node *node) +{ + struct ipa_ref *ref = NULL; + varpool_node *vnode; + for (int j = 0; node->iterate_reference (j, ref); j++) + if (is_a <varpool_node *> (ref->referred)) + { + vnode = dyn_cast <varpool_node *> (ref->referred); + if (!symbol_partitioned_p (vnode) + && !vnode->no_reorder + && vnode->get_partitioning_class () == SYMBOL_PARTITION) + { + add_symbol_to_partition (partition, vnode); + if (dump_file) + fprintf (dump_file, "Varpool Node: %s\n", vnode->dump_asm_name ()); + add_node_references_to_partition (partition, vnode); + } + } + + for (int j = 0; node->iterate_referring (j, ref); j++) + if (is_a <varpool_node *> (ref->referring)) + { + vnode = dyn_cast <varpool_node *> (ref->referring); + gcc_assert (vnode->definition); + if (!symbol_partitioned_p (vnode) + && !vnode->no_reorder + && !vnode->can_remove_if_no_refs_p () + && vnode->get_partitioning_class () == SYMBOL_PARTITION) + { + add_symbol_to_partition (partition, vnode); + if (dump_file) + fprintf (dump_file, "Varpool Node: %s\n", vnode->dump_asm_name ()); + add_node_references_to_partition (partition, vnode); + } + } + if (cgraph_node *cnode = dyn_cast <cgraph_node *> (node)) + { + struct cgraph_edge *e; + + /* Add all inline clones and callees that are duplicated. */ + for (e = cnode->callees; e; e = e->next_callee) + if (e->callee->get_partitioning_class () == SYMBOL_DUPLICATE) + add_node_references_to_partition (partition, e->callee); + + /* Add all thunks associated with the function. */ + for (e = cnode->callers; e; e = e->next_caller) + if (e->caller->thunk && !e->caller->inlined_to) + add_node_references_to_partition (partition, e->caller); + } + +} + +/* Create and return the created partition of name NAME. */ + +static ltrans_partition +create_partition (int &npartitions, const char *name) +{ + npartitions++; + return new_partition (name); +} + +/* Partitioning for code locality. + The partitioning plan (and prerequisite cloning) will have been done by the + IPA locality cloning pass. This function just implements that plan by + assigning those partitions to ltrans_parititions. */ + +void +lto_locality_map (int max_partition_size) +{ + symtab_node *snode; + int npartitions = 0; + + auto_vec<varpool_node *> varpool_order; + struct cgraph_node *node; + + if (locality_partitions.length () == 0) + { + if (dump_file) + { + fprintf (dump_file, "Locality partition: falling back to balanced " + "model\n"); + } + lto_balanced_map (param_lto_partitions, param_max_partition_size); + return; + } + ltrans_partition partition = nullptr; + for (auto part : locality_partitions) + { + partition = create_partition (npartitions, ""); + for (unsigned j = 0; j < part->nodes.length (); j++) + { + node = part->nodes[j]; + if (symbol_partitioned_p (node)) + continue; + + add_symbol_to_partition (partition, node); + add_node_references_to_partition (partition, node); + } + } + + int64_t partition_size = max_partition_size; + /* All other unpartitioned symbols. */ + FOR_EACH_SYMBOL (snode) + { + if (snode->get_partitioning_class () == SYMBOL_PARTITION + && !symbol_partitioned_p (snode)) + { + if (partition->insns > partition_size) + partition = create_partition (npartitions, ""); + + add_symbol_to_partition (partition, snode); + if (dump_file) + fprintf (dump_file, "Un-ordered Node: %s\n", snode->dump_asm_name ()); + } + } +} + /* Return true if we must not change the name of the NODE. The name as extracted from the corresponding decl should be passed in NAME. */ @@ -1732,7 +1853,12 @@ lto_promote_cross_file_statics (void) { ltrans_partition part = ltrans_partitions[i]; + if (dump_file) + fprintf (dump_file, "lto_promote_cross_file_statics for part %s %p\n", + part->name, (void *)part->encoder); part->encoder = compute_ltrans_boundary (part->encoder); + if (dump_file) + fprintf (dump_file, "new encoder %p\n", (void *)part->encoder); } lto_clone_numbers = new hash_map<const char *, unsigned>; diff --git a/gcc/lto/lto-partition.h b/gcc/lto/lto-partition.h index 38b3f1e..a6a4195 100644 --- a/gcc/lto/lto-partition.h +++ b/gcc/lto/lto-partition.h @@ -37,6 +37,7 @@ void lto_1_to_1_map (void); void lto_max_map (void); void lto_cache_map (int, int); void lto_balanced_map (int, int); +void lto_locality_map (int); void lto_promote_cross_file_statics (void); void free_ltrans_partitions (void); void lto_promote_statics_nonwpa (void); diff --git a/gcc/lto/lto.cc b/gcc/lto/lto.cc index 18ca475..183634f 100644 --- a/gcc/lto/lto.cc +++ b/gcc/lto/lto.cc @@ -547,7 +547,9 @@ do_whole_program_analysis (void) symtab_node::checking_verify_symtab_nodes (); bitmap_obstack_release (NULL); - if (flag_lto_partition == LTO_PARTITION_1TO1) + if (flag_ipa_reorder_for_locality) + lto_locality_map (param_max_locality_partition_size); + else if (flag_lto_partition == LTO_PARTITION_1TO1) lto_1_to_1_map (); else if (flag_lto_partition == LTO_PARTITION_MAX) lto_max_map (); diff --git a/gcc/m2/ChangeLog b/gcc/m2/ChangeLog index 234578d..eeb5f66 100644 --- a/gcc/m2/ChangeLog +++ b/gcc/m2/ChangeLog @@ -1,3 +1,22 @@ +2025-04-11 Gaius Mulley <gaiusmod2@gmail.com> + + PR modula2/119735 + * gm2-compiler/M2MetaError.def: Hide %n from comment. + * gm2-compiler/SymbolTable.def (PutIncludedByDefinition): Remove ' + from comment. + * gm2-gcc/m2expr.def (init): Ditto. + * gm2-libiberty/pexecute.def: Ditto. + * gm2-libs-coroutines/Executive.def (InitSemaphore): Ditto. + (Wait): Ditto. + * gm2-libs-iso/ClientSocket.def: Ditto. + * gm2-libs-log/BlockOps.def (BlockMoveBackward): Ditto. + * gm2-libs-log/InOut.def: Ditto. + * mc/mcFileName.def: Ditto. + +2025-04-09 Jakub Jelinek <jakub@redhat.com> + + * gm2-compiler/M2MetaError.def: Fix comment typo, range" -> "range2". + 2025-03-30 Sandra Loosemore <sloosemore@baylibre.com> * lang.opt.urls: Regenerate. diff --git a/gcc/m2/gm2-compiler/M2MetaError.def b/gcc/m2/gm2-compiler/M2MetaError.def index 92c4ad2..cfe9195 100644 --- a/gcc/m2/gm2-compiler/M2MetaError.def +++ b/gcc/m2/gm2-compiler/M2MetaError.def @@ -73,7 +73,8 @@ FROM NameKey IMPORT Name ; {%kword} the string word is unquoted and rendered as a keyword. {%C} chain this error on the previous rooted error. {%R} this error will be the root of the future chained errors. - {%n} decimal number. Not quoted. + {% n} decimal number. Not quoted. There is no space between the + % and n (this has been added to hide this comment from gettext). {%N} count (number), for example, 1st, 2nd, 3rd, 4th. Not quoted. {%X} push contents of the output string onto the string stack. {%Yname} place contents of dictionary entry name onto the output string. diff --git a/gcc/m2/gm2-compiler/SymbolTable.def b/gcc/m2/gm2-compiler/SymbolTable.def index d9d4c87..85a3672 100644 --- a/gcc/m2/gm2-compiler/SymbolTable.def +++ b/gcc/m2/gm2-compiler/SymbolTable.def @@ -2025,7 +2025,7 @@ PROCEDURE PutIncludedByDefinition (Sym: CARDINAL) ; (* IsIncludedByDefinition - returns TRUE if definition module symbol, Sym, was included - by ModSym's definition module. + by ModSyms definition module. *) PROCEDURE IsIncludedByDefinition (ModSym, Sym: CARDINAL) : BOOLEAN ; diff --git a/gcc/m2/gm2-gcc/m2expr.def b/gcc/m2/gm2-gcc/m2expr.def index e9f48b8..a9f5f37 100644 --- a/gcc/m2/gm2-gcc/m2expr.def +++ b/gcc/m2/gm2-gcc/m2expr.def @@ -45,7 +45,7 @@ PROCEDURE init (location: location_t) ; (* CSTIntToString - return an integer string using base 10 and no padding. - The string returned will have been malloc'd. + The string returned will have been mallocd. *) PROCEDURE CSTIntToString (t: tree) : CharStar ; diff --git a/gcc/m2/gm2-libiberty/pexecute.def b/gcc/m2/gm2-libiberty/pexecute.def index 30a41e1..49af52c 100644 --- a/gcc/m2/gm2-libiberty/pexecute.def +++ b/gcc/m2/gm2-libiberty/pexecute.def @@ -31,16 +31,16 @@ EXPORT UNQUALIFIED pexecute ; THIS_PNAME is name of the calling program (i.e. argv[0]). TEMP_BASE is the path name, sans suffix, of a temporary file to use - if needed. This is currently only needed for MSDOS ports that don't use - GO32 (do any still exist?). Ports that don't need it can pass NULL. + if needed. This is currently only needed for MSDOS ports that dont use + GO32 (do any still exist?). Ports that dont need it can pass NULL. (FLAGS & PEXECUTE_SEARCH) is non-zero if $PATH should be searched - (??? It's not clear that GCC passes this flag correctly). + (??? Its not clear that GCC passes this flag correctly). (FLAGS & PEXECUTE_FIRST) is nonzero for the first process in chain. (FLAGS & PEXECUTE_FIRST) is nonzero for the last process in chain. FIRST_LAST could be simplified to only mark the last of a chain of processes but that requires the caller to always mark the last one (and not give up - early if some error occurs). It's more robust to require the caller to + early if some error occurs). Its more robust to require the caller to mark both ends of the chain. The result is the pid on systems like Unix where we fork/exec and on systems @@ -52,20 +52,20 @@ EXPORT UNQUALIFIED pexecute ; Upon failure, ERRMSG_FMT and ERRMSG_ARG are set to the text of the error message with an optional argument (if not needed, ERRMSG_ARG is set to - NULL), and -1 is returned. `errno' is available to the caller to use. + NULL), and -1 is returned. errno is available to the caller to use. pwait: cover function for wait. PID is the process id of the task to wait for. - STATUS is the `status' argument to wait. + STATUS is the status argument to wait. FLAGS is currently unused (allows future enhancement without breaking upward compatibility). Pass 0 for now. The result is the pid of the child reaped, or -1 for failure (errno says why). - On systems that don't support waiting for a particular child, PID is - ignored. On systems like MSDOS that don't really multitask pwait + On systems that dont support waiting for a particular child, PID is + ignored. On systems like MSDOS that dont really multitask pwait is just a mechanism to provide a consistent interface for the caller. pfinish: finish generation of script diff --git a/gcc/m2/gm2-libs-coroutines/Executive.def b/gcc/m2/gm2-libs-coroutines/Executive.def index 40eb8f1..f21a066 100644 --- a/gcc/m2/gm2-libs-coroutines/Executive.def +++ b/gcc/m2/gm2-libs-coroutines/Executive.def @@ -32,7 +32,7 @@ EXPORT QUALIFIED SEMAPHORE, DESCRIPTOR, RotateRunQueue, ProcessName, DebugProcess ; TYPE - SEMAPHORE ; (* defines Dijkstra's semaphores *) + SEMAPHORE ; (* defines Dijkstras semaphores *) DESCRIPTOR ; (* handle onto a process *) @@ -85,7 +85,7 @@ PROCEDURE InitSemaphore (v: CARDINAL; Name: ARRAY OF CHAR) : SEMAPHORE ; (* - Wait - performs dijkstra's P operation on a semaphore. + Wait - performs dijkstras P operation on a semaphore. A process which calls this procedure will wait until the value of the semaphore is > 0 and then it will decrement this value. @@ -95,7 +95,7 @@ PROCEDURE Wait (s: SEMAPHORE) ; (* - Signal - performs dijkstra's V operation on a semaphore. + Signal - performs dijkstras V operation on a semaphore. A process which calls the procedure will increment the semaphores value. *) diff --git a/gcc/m2/gm2-libs-iso/ClientSocket.def b/gcc/m2/gm2-libs-iso/ClientSocket.def index 293b53a..98aefc6 100644 --- a/gcc/m2/gm2-libs-iso/ClientSocket.def +++ b/gcc/m2/gm2-libs-iso/ClientSocket.def @@ -1,4 +1,4 @@ -(* ClientSocket.def provides a client TCP interface for ChanId's. +(* ClientSocket.def provides a client TCP interface for ChanIds. Copyright (C) 2008-2025 Free Software Foundation, Inc. Contributed by Gaius Mulley <gaius.mulley@southwales.ac.uk>. diff --git a/gcc/m2/gm2-libs-log/BlockOps.def b/gcc/m2/gm2-libs-log/BlockOps.def index 2978920..b770acc 100644 --- a/gcc/m2/gm2-libs-log/BlockOps.def +++ b/gcc/m2/gm2-libs-log/BlockOps.def @@ -50,7 +50,7 @@ PROCEDURE BlockMoveBackward (dest, src: ADDRESS; n: CARDINAL) ; (* - BlockClear - fills, block..block+n-1, with zero's. + BlockClear - fills, block..block+n-1, with zeros. *) PROCEDURE BlockClear (block: ADDRESS; n: CARDINAL) ; diff --git a/gcc/m2/gm2-libs-log/InOut.def b/gcc/m2/gm2-libs-log/InOut.def index 9335d0a..f2294e9 100644 --- a/gcc/m2/gm2-libs-log/InOut.def +++ b/gcc/m2/gm2-libs-log/InOut.def @@ -45,7 +45,7 @@ VAR (* OpenInput - reads a string from stdin as the filename for reading. - If the filename ends with `.' then it appends the defext + If the filename ends with '.' then it appends the defext extension. The global variable Done is set if all was successful. *) @@ -63,7 +63,7 @@ PROCEDURE CloseInput ; (* OpenOutput - reads a string from stdin as the filename for writing. - If the filename ends with `.' then it appends the defext + If the filename ends with '.' then it appends the defext extension. The global variable Done is set if all was successful. *) diff --git a/gcc/m2/mc/mcFileName.def b/gcc/m2/mc/mcFileName.def index da9db60..7768c2f 100644 --- a/gcc/m2/mc/mcFileName.def +++ b/gcc/m2/mc/mcFileName.def @@ -29,7 +29,7 @@ FROM DynamicStrings IMPORT String ; given a module and an extension. This file name length will be operating system specific. String, Extension, is concatenated onto - Module and thus it is safe to `Mark' the extension + Module and thus it is safe to Mark the extension for garbage collection. *) diff --git a/gcc/opts.cc b/gcc/opts.cc index 80c7a97..5e7b77d 100644 --- a/gcc/opts.cc +++ b/gcc/opts.cc @@ -1037,6 +1037,25 @@ report_conflicting_sanitizer_options (struct gcc_options *opts, location_t loc, } } +/* Validate from OPTS and OPTS_SET that when -fipa-reorder-for-locality is + enabled no explicit -flto-partition is also passed as the locality cloning + pass uses its own partitioning scheme. */ + +static void +validate_ipa_reorder_locality_lto_partition (struct gcc_options *opts, + struct gcc_options *opts_set) +{ + static bool validated_p = false; + + if (opts->x_flag_lto_partition != LTO_PARTITION_DEFAULT) + { + if (opts_set->x_flag_ipa_reorder_for_locality && !validated_p) + error ("%<-fipa-reorder-for-locality%> is incompatible with" + " an explicit %qs option", "-flto-partition"); + } + validated_p = true; +} + /* After all options at LOC have been read into OPTS and OPTS_SET, finalize settings of those options and diagnose incompatible combinations. */ @@ -1249,6 +1268,10 @@ finish_options (struct gcc_options *opts, struct gcc_options *opts_set, if (opts->x_flag_reorder_blocks_and_partition) SET_OPTION_IF_UNSET (opts, opts_set, flag_reorder_functions, 1); + validate_ipa_reorder_locality_lto_partition (opts, opts_set); + if (opts_set->x_flag_lto_partition != LTO_PARTITION_DEFAULT) + opts_set->x_flag_lto_partition = opts->x_flag_lto_partition = LTO_PARTITION_BALANCED; + /* The -gsplit-dwarf option requires -ggnu-pubnames. */ if (opts->x_dwarf_split_debug_info) opts->x_debug_generate_pub_sections = 2; diff --git a/gcc/params.opt b/gcc/params.opt index 422d082..a2b606f 100644 --- a/gcc/params.opt +++ b/gcc/params.opt @@ -469,6 +469,33 @@ Minimal size of a partition for LTO (in estimated instructions). Common Joined UInteger Var(param_lto_partitions) Init(128) IntegerRange(1, 65536) Param Number of partitions the program should be split to. +Enum +Name(lto_locality_cloning_model) Type(enum lto_locality_cloning_model) UnknownError(unknown LTO partitioning model %qs) + +EnumValue +Enum(lto_locality_cloning_model) String(no) Value(LTO_LOCALITY_NO_CLONING) + +EnumValue +Enum(lto_locality_cloning_model) String(non_interposable) Value(LTO_LOCALITY_NON_INTERPOSABLE_CLONING) + +EnumValue +Enum(lto_locality_cloning_model) String(maximal) Value(LTO_LOCALITY_MAXIMAL_CLONING) + +-param=lto-partition-locality-cloning= +Common Joined RejectNegative Enum(lto_locality_cloning_model) Var(flag_lto_locality_cloning) Init(LTO_LOCALITY_MAXIMAL_CLONING) Optimization + +-param=lto-partition-locality-frequency-cutoff= +Common Joined UInteger Var(param_lto_locality_frequency) Init(1) IntegerRange(0, 65536) Param Optimization +The denominator n of fraction 1/n of the execution frequency of callee to be cloned for a particular caller. Special value of 0 dictates to always clone without a cut-off. + +-param=lto-partition-locality-size-cutoff= +Common Joined UInteger Var(param_lto_locality_size) Init(1000) IntegerRange(1, 65536) Param Optimization +Size cut-off for callee including inlined calls to be cloned for a particular caller. + +-param=lto-max-locality-partition= +Common Joined UInteger Var(param_max_locality_partition_size) Init(1000000) Param +Maximal size of a locality partition for LTO (in estimated instructions). Value of 0 results in default value being used. + -param=max-average-unrolled-insns= Common Joined UInteger Var(param_max_average_unrolled_insns) Init(80) Param Optimization The maximum number of instructions to consider to unroll in a loop on average. diff --git a/gcc/passes.def b/gcc/passes.def index 9fd85a3..3b25105 100644 --- a/gcc/passes.def +++ b/gcc/passes.def @@ -162,6 +162,7 @@ along with GCC; see the file COPYING3. If not see NEXT_PASS (pass_ipa_sra); NEXT_PASS (pass_ipa_fn_summary); NEXT_PASS (pass_ipa_inline); + NEXT_PASS (pass_ipa_locality_cloning); NEXT_PASS (pass_ipa_pure_const); NEXT_PASS (pass_ipa_modref); NEXT_PASS (pass_ipa_free_fn_summary, false /* small_p */); diff --git a/gcc/po/ChangeLog b/gcc/po/ChangeLog index a3ea29f..b1537f7 100644 --- a/gcc/po/ChangeLog +++ b/gcc/po/ChangeLog @@ -1,3 +1,13 @@ +2025-04-09 Joseph Myers <josmyers@redhat.com> + + * de.po: Update. + +2025-04-09 Joseph Myers <josmyers@redhat.com> + + * be.po, da.po, de.po, el.po, es.po, fi.po, fr.po, hr.po, id.po, + ja.po, ka.po, nl.po, ru.po, sr.po, sv.po, tr.po, uk.po, vi.po, + zh_CN.po, zh_TW.po: Update. + 2025-04-07 Joseph Myers <josmyers@redhat.com> * sv.po: Update. diff --git a/gcc/rust/ChangeLog b/gcc/rust/ChangeLog index 64053d4..453b9f7 100644 --- a/gcc/rust/ChangeLog +++ b/gcc/rust/ChangeLog @@ -1,3 +1,162 @@ +2025-04-14 Arthur Cohen <arthur.cohen@embecosm.com> + + * util/rust-lang-item.h: Add new manually_drop lang item. + * util/rust-lang-item.cc: Likewise. + +2025-04-14 Arthur Cohen <arthur.cohen@embecosm.com> + + * util/rust-attribute-values.h: Add RUSTFMT value. + * util/rust-attributes.cc: Define the attribute. + * util/rust-attributes.h (enum CompilerPass): Add EXTERNAL variant. + * expand/rust-macro-builtins.cc: Fix formatting. + +2025-04-14 Arthur Cohen <arthur.cohen@embecosm.com> + + * resolve/rust-early-name-resolver-2.0.cc (Early::visit_attributes): Remove assertion. + +2025-04-14 Arthur Cohen <arthur.cohen@embecosm.com> + + * util/rust-attribute-values.h: Add missing attributes. + * util/rust-attributes.cc: Likewise. + * util/rust-attributes.h (enum CompilerPass): Mention adding something for const + functions. + +2025-04-14 beamandala <mandalapubhavesh@gmail.com> + + * expand/rust-macro-builtins.cc (MacroBuiltin::builtin_transcribers): + Add entry for track_caller. + * util/rust-attribute-values.h: add `TRACK_CALLER` attribute. + * util/rust-attributes.cc: add `track_caller` attribute definition. + +2025-04-14 Owen Avery <powerboat9.gamer@gmail.com> + + * checks/errors/rust-const-checker.cc + (ConstChecker::visit): Visit the enum items of enums. + * resolve/rust-ast-resolve-item.cc + (ResolveItem::visit): Resolve enum discriminants during nr1.0. + +2025-04-14 Arthur Cohen <arthur.cohen@embecosm.com> + + * expand/rust-macro-builtins-format-args.cc (format_args_parse_arguments): Improve safety, + allow extra commas after end of argument list. + +2025-04-14 Arthur Cohen <arthur.cohen@embecosm.com> + + * expand/rust-macro-expand.cc (MacroExpander::expand_decl_macro): Call into + TokenTreeDesugar. + * expand/rust-token-tree-desugar.cc: New file. + * expand/rust-token-tree-desugar.h: New file. + * Make-lang.in: Compile them. + +2025-04-14 Arthur Cohen <arthur.cohen@embecosm.com> + + * expand/rust-macro-expand.cc (MacroExpander::match_n_matches): Do not + insert fragments and substack fragments if the matcher failed. + +2025-04-14 Arthur Cohen <arthur.cohen@embecosm.com> + + * rust-session-manager.cc (Session::compile_crate): Call the visitor later in the pipeline. + +2025-04-14 Arthur Cohen <arthur.cohen@embecosm.com> + + * ast/rust-ast.h (DelimTokenTree::get_locus): New function. + +2025-04-14 Arthur Cohen <arthur.cohen@embecosm.com> + + * ast/rust-expr.h (class RangeExpr): Add empty outer attributes and allow getting them + and setting them. + +2025-04-14 Arthur Cohen <arthur.cohen@embecosm.com> + + * resolve/rust-toplevel-name-resolver-2.0.cc (TopLevel::visit): Return if module + is unloaded. + +2025-04-14 Arthur Cohen <arthur.cohen@embecosm.com> + + * typecheck/rust-hir-type-check-expr.cc (is_default_fn): New. + (emit_ambiguous_resolution_error): New. + (handle_multiple_candidates): Properly handle multiple candidates in + the case of specialization. + (TypeCheckExpr::visit): Call `handle_multiple_candidates`. + +2025-04-14 Andrew Pinski <quic_apinski@quicinc.com> + + PR rust/119342 + * rust-gcc.cc (block): Add comment on why chaining + the variables of the scope toether. + +2025-04-14 Andrew Pinski <quic_apinski@quicinc.com> + + PR rust/119341 + * rust-gcc.cc (function_type): Use range fors. + (function_type_variadic): Likewise. + (fill_in_fields): Likewise. + (statement_list): Likewise. + (block): Likewise. + (block_add_statements): Likewise. + (function_set_parameters): Likewise. + (write_global_definitions): Likewise. + +2025-04-14 Andrew Pinski <quic_apinski@quicinc.com> + + * rust-gcc.cc (Bvariable::get_tree): Use error_operand_p. + (pointer_type): Likewise. + (reference_type): Likewise. + (immutable_type): Likewise. + (function_type): Likewise. + (function_type_variadic): Likewise. + Cleanup the check for receiver.type first. + (function_ptr_type): Use error_operand_p. + (fill_in_fields): Likewise. + (fill_in_array): Likewise. + (named_type): Likewise. + (type_size): Likewise. + (type_alignment): Likewise. + (type_field_alignment): Likewise. + (type_field_offset): Likewise. + (zero_expression): Likewise. + (float_constant_expression): Likewise. + (convert_expression): Likewise. + (struct_field_expression): Likewise. + (compound_expression): Likewise. + (conditional_expression): Likewise. + (negation_expression): Likewise. + (arithmetic_or_logical_expression): Likewise. + (arithmetic_or_logical_expression_checked): Likewise. + (comparison_expression): Likewise. + (lazy_boolean_expression): Likewise. + (constructor_expression): Likewise. + (array_constructor_expression): Likewise. + (array_index_expression): Likewise. + (call_expression): Likewise. + (init_statement): Likewise. + (assignment_statement): Likewise. + (return_statement): Likewise. + (exception_handler_statement): Likewise. + (if_statement): Likewise. + (compound_statement): Likewise. + Tighten up the code, removing t variable. + (statement_list): Use error_operand_p. + (block): Likewise. + (block_add_statements): Likewise. + (convert_tree): Likewise. + (global_variable): Likewise. + (global_variable_set_init): Likewise. + (local_variable): Likewise. + (parameter_variable): Likewise. + (static_chain_variable): Likewise. + (temporary_variable): Likewise. + (function): Likewise. Tighten up the code. + (function_defer_statement): Use error_operand_p. + (function_set_parameters): Use error_operand_p. + (write_global_definitions): Use error_operand_p. + Tighten up the code around the loop. + +2025-04-14 Andrew Pinski <quic_apinski@quicinc.com> + + * rust-gcc.cc (is_floating_point): Use FLOAT_TYPE_P + instead of manually checking the type. + 2025-04-08 Matty Kuhn <matty.kuhn.1@gmail.com> * ast/rust-ast.h: (AST::Attribute): add empty_input function diff --git a/gcc/rust/Make-lang.in b/gcc/rust/Make-lang.in index 4028b47..835e113 100644 --- a/gcc/rust/Make-lang.in +++ b/gcc/rust/Make-lang.in @@ -115,6 +115,7 @@ GRS_OBJS = \ rust/rust-macro-builtins-format-args.o \ rust/rust-macro-builtins-location.o \ rust/rust-macro-builtins-include.o \ + rust/rust-token-tree-desugar.o \ rust/rust-fmt.o \ rust/rust-hir.o \ rust/rust-hir-map.o \ diff --git a/gcc/rust/ast/rust-ast.h b/gcc/rust/ast/rust-ast.h index 09e0fce..91611ec 100644 --- a/gcc/rust/ast/rust-ast.h +++ b/gcc/rust/ast/rust-ast.h @@ -1018,6 +1018,7 @@ public: } DelimType get_delim_type () const { return delim_type; } + location_t get_locus () const { return locus; } }; /* Forward decl - definition moved to rust-expr.h as it requires LiteralExpr diff --git a/gcc/rust/ast/rust-expr.h b/gcc/rust/ast/rust-expr.h index 84cdfdb..69538df 100644 --- a/gcc/rust/ast/rust-expr.h +++ b/gcc/rust/ast/rust-expr.h @@ -3004,6 +3004,10 @@ class RangeExpr : public ExprWithoutBlock { location_t locus; + // Some visitors still check for attributes on RangeExprs, and they will need + // to be supported in the future - so keep that for now + std::vector<Attribute> empty_attributes = {}; + protected: // outer attributes not allowed before range expressions RangeExpr (location_t locus) : locus (locus) {} @@ -3013,15 +3017,11 @@ public: std::vector<Attribute> &get_outer_attrs () override final { - // RangeExpr cannot have any outer attributes - rust_assert (false); + return empty_attributes; } // should never be called - error if called - void set_outer_attrs (std::vector<Attribute> /* new_attrs */) override - { - rust_assert (false); - } + void set_outer_attrs (std::vector<Attribute> /* new_attrs */) override {} Expr::Kind get_expr_kind () const override { return Expr::Kind::Range; } }; diff --git a/gcc/rust/checks/errors/rust-const-checker.cc b/gcc/rust/checks/errors/rust-const-checker.cc index 4904322..4c2257a 100644 --- a/gcc/rust/checks/errors/rust-const-checker.cc +++ b/gcc/rust/checks/errors/rust-const-checker.cc @@ -646,6 +646,9 @@ ConstChecker::visit (Enum &enum_item) { check_default_const_generics (enum_item.get_generic_params (), ConstGenericCtx::Enum); + + for (auto &item : enum_item.get_variants ()) + item->accept_vis (*this); } void diff --git a/gcc/rust/expand/rust-macro-builtins-format-args.cc b/gcc/rust/expand/rust-macro-builtins-format-args.cc index 8eb32d5..3e1249d 100644 --- a/gcc/rust/expand/rust-macro-builtins-format-args.cc +++ b/gcc/rust/expand/rust-macro-builtins-format-args.cc @@ -55,6 +55,8 @@ format_args_parse_arguments (AST::MacroInvocData &invoc) if (parser.peek_current_token ()->get_id () == STRING_LITERAL) format_expr = parser.parse_literal_expr (); + rust_assert (format_expr); + // TODO(Arthur): Clean this up - if we haven't parsed a string literal but a // macro invocation, what do we do here? return a tl::unexpected? auto format_str = static_cast<AST::LiteralExpr &> (*format_expr) @@ -81,6 +83,11 @@ format_args_parse_arguments (AST::MacroInvocData &invoc) { parser.skip_token (COMMA); + // Check in case of an extraneous comma in the args list, which is + // allowed - format_args!("fmt", arg, arg2,) + if (parser.peek_current_token ()->get_id () == last_token_id) + break; + if (parser.peek_current_token ()->get_id () == IDENTIFIER && parser.peek (1)->get_id () == EQUAL) { diff --git a/gcc/rust/expand/rust-macro-builtins.cc b/gcc/rust/expand/rust-macro-builtins.cc index 39c4c46..8b406ff 100644 --- a/gcc/rust/expand/rust-macro-builtins.cc +++ b/gcc/rust/expand/rust-macro-builtins.cc @@ -83,7 +83,6 @@ const BiMap<std::string, BuiltinMacro> MacroBuiltin::builtins = {{ {"Ord", BuiltinMacro::Ord}, {"PartialOrd", BuiltinMacro::PartialOrd}, {"Hash", BuiltinMacro::Hash}, - }}; AST::MacroTranscriberFunc @@ -137,6 +136,7 @@ std::unordered_map<std::string, AST::MacroTranscriberFunc> {"cfg_accessible", MacroBuiltin::sorry}, {"rustc_const_stable", MacroBuiltin::sorry}, {"rustc_const_unstable", MacroBuiltin::sorry}, + {"track_caller", MacroBuiltin::sorry}, /* Derive builtins do not need a real transcriber, but still need one. It should however never be called since builtin derive macros get expanded differently, and benefit from knowing on what kind of items they are diff --git a/gcc/rust/expand/rust-macro-expand.cc b/gcc/rust/expand/rust-macro-expand.cc index cd17a3f..673b8fb 100644 --- a/gcc/rust/expand/rust-macro-expand.cc +++ b/gcc/rust/expand/rust-macro-expand.cc @@ -28,6 +28,7 @@ #include "rust-cfg-strip.h" #include "rust-early-name-resolver.h" #include "rust-proc-macro.h" +#include "rust-token-tree-desugar.h" namespace Rust { @@ -78,7 +79,10 @@ MacroExpander::expand_decl_macro (location_t invoc_locus, * trees. */ - AST::DelimTokenTree &invoc_token_tree = invoc.get_delim_tok_tree (); + AST::DelimTokenTree &invoc_token_tree_sugar = invoc.get_delim_tok_tree (); + + // We must first desugar doc comments into proper attributes + auto invoc_token_tree = AST::TokenTreeDesugar ().go (invoc_token_tree_sugar); // find matching arm AST::MacroRule *matched_rule = nullptr; @@ -621,9 +625,10 @@ MacroExpander::match_n_matches (Parser<MacroInvocLexer> &parser, // matched fragment get the offset in the token stream size_t offs_end = source.get_offs (); - sub_stack.insert_metavar ( - MatchedFragment (fragment->get_ident ().as_string (), - offs_begin, offs_end)); + if (valid_current_match) + sub_stack.insert_metavar ( + MatchedFragment (fragment->get_ident ().as_string (), + offs_begin, offs_end)); } break; @@ -650,15 +655,15 @@ MacroExpander::match_n_matches (Parser<MacroInvocLexer> &parser, } auto old_stack = sub_stack.pop (); - // nest metavars into repetitions - for (auto &ent : old_stack) - sub_stack.append_fragment (ent.first, std::move (ent.second)); - // If we've encountered an error once, stop trying to match more // repetitions if (!valid_current_match) break; + // nest metavars into repetitions + for (auto &ent : old_stack) + sub_stack.append_fragment (ent.first, std::move (ent.second)); + match_amount++; // Break early if we notice there's too many expressions already diff --git a/gcc/rust/expand/rust-token-tree-desugar.cc b/gcc/rust/expand/rust-token-tree-desugar.cc new file mode 100644 index 0000000..3b47180 --- /dev/null +++ b/gcc/rust/expand/rust-token-tree-desugar.cc @@ -0,0 +1,72 @@ +// Copyright (C) 2025 Free Software Foundation, Inc. + +// This file is part of GCC. + +// GCC is free software; you can redistribute it and/or modify it under +// the terms of the GNU General Public License as published by the Free +// Software Foundation; either version 3, or (at your option) any later +// version. + +// GCC is distributed in the hope that it will be useful, but WITHOUT ANY +// WARRANTY; without even the implied warranty of MERCHANTABILITY or +// FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License +// for more details. + +// You should have received a copy of the GNU General Public License +// along with GCC; see the file COPYING3. If not see +// <http://www.gnu.org/licenses/>. + +#include "rust-token-tree-desugar.h" +#include "rust-ast.h" +#include "rust-token.h" + +namespace Rust { +namespace AST { + +DelimTokenTree +TokenTreeDesugar::go (DelimTokenTree &tts) +{ + tts.accept_vis (*this); + + return DelimTokenTree (tts.get_delim_type (), std::move (desugared), + tts.get_locus ()); +} + +void +TokenTreeDesugar::append (TokenPtr &&new_token) +{ + desugared.emplace_back (std::make_unique<Token> (std::move (new_token))); +} + +void +TokenTreeDesugar::append (std::unique_ptr<TokenTree> &&new_token) +{ + desugared.emplace_back (std::move (new_token)); +} + +void +TokenTreeDesugar::visit (Token &tts) +{ + if (tts.get_id () == TokenId::OUTER_DOC_COMMENT + || tts.get_id () == TokenId::INNER_DOC_COMMENT) + { + append (Rust::Token::make (TokenId::HASH, tts.get_locus ())); + + if (tts.get_id () == TokenId::INNER_DOC_COMMENT) + append (Rust::Token::make (EXCLAM, tts.get_locus ())); + + append (Rust::Token::make (TokenId::LEFT_SQUARE, tts.get_locus ())); + append (Rust::Token::make_identifier (tts.get_locus (), "doc")); + append (Rust::Token::make (TokenId::EQUAL, tts.get_locus ())); + append (Rust::Token::make_string (tts.get_locus (), + std::string (tts.get_str ()))); + append (Rust::Token::make (TokenId::RIGHT_SQUARE, tts.get_locus ())); + } + else + { + append (tts.clone_token ()); + } +} + +}; // namespace AST +}; // namespace Rust diff --git a/gcc/rust/expand/rust-token-tree-desugar.h b/gcc/rust/expand/rust-token-tree-desugar.h new file mode 100644 index 0000000..ccba53b --- /dev/null +++ b/gcc/rust/expand/rust-token-tree-desugar.h @@ -0,0 +1,55 @@ +// Copyright (C) 2025 Free Software Foundation, Inc. + +// This file is part of GCC. + +// GCC is free software; you can redistribute it and/or modify it under +// the terms of the GNU General Public License as published by the Free +// Software Foundation; either version 3, or (at your option) any later +// version. + +// GCC is distributed in the hope that it will be useful, but WITHOUT ANY +// WARRANTY; without even the implied warranty of MERCHANTABILITY or +// FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License +// for more details. + +// You should have received a copy of the GNU General Public License +// along with GCC; see the file COPYING3. If not see +// <http://www.gnu.org/licenses/>. + +#ifndef RUST_TOKEN_TREE_DESUGAR_H +#define RUST_TOKEN_TREE_DESUGAR_H + +#include "rust-ast-visitor.h" +#include "rust-system.h" +#include "rust-ast.h" + +namespace Rust { +namespace AST { + +/** + * Desugar a given token-tree before parsing it for a macro invocation. At the + * moment, the sole purpose of this desugar is to transform doc-comments into + * their attribute form (/// comment -> #[doc = "comment"]) + */ +class TokenTreeDesugar : public DefaultASTVisitor +{ +public: + TokenTreeDesugar () : desugared (std::vector<std::unique_ptr<TokenTree>> ()) + {} + + DelimTokenTree go (DelimTokenTree &tts); + +private: + std::vector<std::unique_ptr<TokenTree>> desugared; + void append (TokenPtr &&new_token); + void append (std::unique_ptr<TokenTree> &&new_token); + + using DefaultASTVisitor::visit; + + virtual void visit (Token &tts) override; +}; + +}; // namespace AST +}; // namespace Rust + +#endif //! RUST_TOKEN_TREE_DESUGAR_H diff --git a/gcc/rust/resolve/rust-ast-resolve-item.cc b/gcc/rust/resolve/rust-ast-resolve-item.cc index d584961..30f6d43 100644 --- a/gcc/rust/resolve/rust-ast-resolve-item.cc +++ b/gcc/rust/resolve/rust-ast-resolve-item.cc @@ -356,6 +356,8 @@ ResolveItem::visit (AST::EnumItemDiscriminant &item) auto cpath = canonical_prefix.append (decl); mappings.insert_canonical_path (item.get_node_id (), cpath); + + ResolveExpr::go (item.get_expr (), path, cpath); } void diff --git a/gcc/rust/resolve/rust-early-name-resolver-2.0.cc b/gcc/rust/resolve/rust-early-name-resolver-2.0.cc index afaca1f..36456e1 100644 --- a/gcc/rust/resolve/rust-early-name-resolver-2.0.cc +++ b/gcc/rust/resolve/rust-early-name-resolver-2.0.cc @@ -325,10 +325,9 @@ Early::visit_attributes (std::vector<AST::Attribute> &attrs) auto pm_def = mappings.lookup_derive_proc_macro_def ( definition->get_node_id ()); - rust_assert (pm_def.has_value ()); - - mappings.insert_derive_proc_macro_invocation (trait, - pm_def.value ()); + if (pm_def.has_value ()) + mappings.insert_derive_proc_macro_invocation (trait, + pm_def.value ()); } } else if (Analysis::BuiltinAttributeMappings::get () diff --git a/gcc/rust/resolve/rust-toplevel-name-resolver-2.0.cc b/gcc/rust/resolve/rust-toplevel-name-resolver-2.0.cc index 8863be7..ba37dee 100644 --- a/gcc/rust/resolve/rust-toplevel-name-resolver-2.0.cc +++ b/gcc/rust/resolve/rust-toplevel-name-resolver-2.0.cc @@ -113,7 +113,17 @@ TopLevel::visit (AST::Module &module) // This was copied from the old early resolver method // 'accumulate_escaped_macros' if (module.get_kind () == AST::Module::UNLOADED) - module.load_items (); + { + module.load_items (); + + // If the module was previously unloaded, then we don't want to visit it + // this time around as the CfgStrip hasn't run on its inner items yet. + // Skip it for now, mark the visitor as dirty and try again + + dirty = true; + + return; + } DefaultResolver::visit (module); diff --git a/gcc/rust/rust-gcc.cc b/gcc/rust/rust-gcc.cc index 72aef08..234721c 100644 --- a/gcc/rust/rust-gcc.cc +++ b/gcc/rust/rust-gcc.cc @@ -61,7 +61,7 @@ tree Bvariable::get_tree (location_t location) const { - if (this->t_ == error_mark_node) + if (error_operand_p (this->t_)) return error_mark_node; TREE_USED (this->t_) = 1; @@ -431,7 +431,7 @@ float_type (int bits) tree pointer_type (tree to_type) { - if (to_type == error_mark_node) + if (error_operand_p (to_type)) return error_mark_node; tree type = build_pointer_type (to_type); return type; @@ -442,7 +442,7 @@ pointer_type (tree to_type) tree reference_type (tree to_type) { - if (to_type == error_mark_node) + if (error_operand_p (to_type)) return error_mark_node; tree type = build_reference_type (to_type); return type; @@ -453,7 +453,7 @@ reference_type (tree to_type) tree immutable_type (tree base) { - if (base == error_mark_node) + if (error_operand_p (base)) return error_mark_node; tree constified = build_qualified_type (base, TYPE_QUAL_CONST); return constified; @@ -472,17 +472,16 @@ function_type (const typed_identifier &receiver, if (receiver.type != NULL_TREE) { tree t = receiver.type; - if (t == error_mark_node) + if (error_operand_p (t)) return error_mark_node; *pp = tree_cons (NULL_TREE, t, NULL_TREE); pp = &TREE_CHAIN (*pp); } - for (std::vector<typed_identifier>::const_iterator p = parameters.begin (); - p != parameters.end (); ++p) + for (const auto &p : parameters) { - tree t = p->type; - if (t == error_mark_node) + tree t = p.type; + if (error_operand_p (t)) return error_mark_node; *pp = tree_cons (NULL_TREE, t, NULL_TREE); pp = &TREE_CHAIN (*pp); @@ -502,11 +501,11 @@ function_type (const typed_identifier &receiver, gcc_assert (result_struct != NULL); result = result_struct; } - if (result == error_mark_node) + if (error_operand_p (result)) return error_mark_node; tree fntype = build_function_type (result, args); - if (fntype == error_mark_node) + if (error_operand_p (fntype)) return error_mark_node; return build_pointer_type (fntype); @@ -521,21 +520,16 @@ function_type_variadic (const typed_identifier &receiver, size_t n = parameters.size () + (receiver.type != NULL_TREE ? 1 : 0); tree *args = XALLOCAVEC (tree, n); size_t offs = 0; + if (error_operand_p (receiver.type)) + return error_mark_node; if (receiver.type != NULL_TREE) - { - tree t = receiver.type; - if (t == error_mark_node) - return error_mark_node; - - args[offs++] = t; - } + args[offs++] = receiver.type; - for (std::vector<typed_identifier>::const_iterator p = parameters.begin (); - p != parameters.end (); ++p) + for (const auto &p : parameters) { - tree t = p->type; - if (t == error_mark_node) + tree t = p.type; + if (error_operand_p (t)) return error_mark_node; args[offs++] = t; } @@ -550,11 +544,11 @@ function_type_variadic (const typed_identifier &receiver, gcc_assert (result_struct != NULL_TREE); result = result_struct; } - if (result == error_mark_node) + if (error_operand_p (result)) return error_mark_node; tree fntype = build_varargs_function_type_array (result, n, args); - if (fntype == error_mark_node) + if (error_operand_p (fntype)) return error_mark_node; return build_pointer_type (fntype); @@ -569,7 +563,7 @@ function_ptr_type (tree result_type, const std::vector<tree> ¶meters, for (auto ¶m : parameters) { - if (param == error_mark_node) + if (error_operand_p (param)) return error_mark_node; *pp = tree_cons (NULL_TREE, param, NULL_TREE); @@ -583,7 +577,7 @@ function_ptr_type (tree result_type, const std::vector<tree> ¶meters, result = void_type_node; tree fntype = build_function_type (result, args); - if (fntype == error_mark_node) + if (error_operand_p (fntype)) return error_mark_node; return build_pointer_type (fntype); @@ -613,14 +607,13 @@ fill_in_fields (tree fill, const std::vector<typed_identifier> &fields, { tree field_trees = NULL_TREE; tree *pp = &field_trees; - for (std::vector<typed_identifier>::const_iterator p = fields.begin (); - p != fields.end (); ++p) + for (const auto &p : fields) { - tree name_tree = get_identifier_from_string (p->name); - tree type_tree = p->type; - if (type_tree == error_mark_node) + tree name_tree = get_identifier_from_string (p.name); + tree type_tree = p.type; + if (error_operand_p (type_tree)) return error_mark_node; - tree field = build_decl (p->location, FIELD_DECL, name_tree, type_tree); + tree field = build_decl (p.location, FIELD_DECL, name_tree, type_tree); DECL_CONTEXT (field) = fill; *pp = field; pp = &DECL_CHAIN (field); @@ -652,7 +645,7 @@ array_type (tree element_type, tree length) tree fill_in_array (tree fill, tree element_type, tree length_tree) { - if (element_type == error_mark_node || length_tree == error_mark_node) + if (error_operand_p (element_type) || error_operand_p (length_tree)) return error_mark_node; gcc_assert (TYPE_SIZE (element_type) != NULL_TREE); @@ -684,7 +677,7 @@ fill_in_array (tree fill, tree element_type, tree length_tree) tree named_type (const std::string &name, tree type, location_t location) { - if (type == error_mark_node) + if (error_operand_p (type)) return error_mark_node; // The middle-end expects a basic type to have a name. In Rust every @@ -714,7 +707,7 @@ named_type (const std::string &name, tree type, location_t location) int64_t type_size (tree t) { - if (t == error_mark_node) + if (error_operand_p (t)) return 1; if (t == void_type_node) return 0; @@ -732,7 +725,7 @@ type_size (tree t) int64_t type_alignment (tree t) { - if (t == error_mark_node) + if (error_operand_p (t)) return 1; return TYPE_ALIGN_UNIT (t); } @@ -742,7 +735,7 @@ type_alignment (tree t) int64_t type_field_alignment (tree t) { - if (t == error_mark_node) + if (error_operand_p (t)) return 1; return rust_field_alignment (t); } @@ -752,7 +745,7 @@ type_field_alignment (tree t) int64_t type_field_offset (tree struct_tree, size_t index) { - if (struct_tree == error_mark_node) + if (error_operand_p (struct_tree)) return 0; gcc_assert (TREE_CODE (struct_tree) == RECORD_TYPE); tree field = TYPE_FIELDS (struct_tree); @@ -773,7 +766,7 @@ tree zero_expression (tree t) { tree ret; - if (t == error_mark_node) + if (error_operand_p (t)) ret = error_mark_node; else ret = build_zero_cst (t); @@ -794,7 +787,7 @@ tree float_constant_expression (tree t, mpfr_t val) { tree ret; - if (t == error_mark_node) + if (error_operand_p (t)) return error_mark_node; REAL_VALUE_TYPE r1; @@ -845,8 +838,7 @@ boolean_constant_expression (bool val) tree convert_expression (tree type_tree, tree expr_tree, location_t location) { - if (type_tree == error_mark_node || expr_tree == error_mark_node - || TREE_TYPE (expr_tree) == error_mark_node) + if (error_operand_p (type_tree) || error_operand_p (expr_tree)) return error_mark_node; tree ret; @@ -878,8 +870,7 @@ convert_expression (tree type_tree, tree expr_tree, location_t location) tree struct_field_expression (tree struct_tree, size_t index, location_t location) { - if (struct_tree == error_mark_node - || TREE_TYPE (struct_tree) == error_mark_node) + if (error_operand_p (struct_tree)) return error_mark_node; gcc_assert (TREE_CODE (TREE_TYPE (struct_tree)) == RECORD_TYPE || TREE_CODE (TREE_TYPE (struct_tree)) == UNION_TYPE); @@ -895,7 +886,7 @@ struct_field_expression (tree struct_tree, size_t index, location_t location) field = DECL_CHAIN (field); gcc_assert (field != NULL_TREE); } - if (TREE_TYPE (field) == error_mark_node) + if (error_operand_p (TREE_TYPE (field))) return error_mark_node; tree ret = fold_build3_loc (location, COMPONENT_REF, TREE_TYPE (field), struct_tree, field, NULL_TREE); @@ -909,7 +900,7 @@ struct_field_expression (tree struct_tree, size_t index, location_t location) tree compound_expression (tree stat, tree expr, location_t location) { - if (stat == error_mark_node || expr == error_mark_node) + if (error_operand_p (stat) || error_operand_p (expr)) return error_mark_node; tree ret = fold_build2_loc (location, COMPOUND_EXPR, TREE_TYPE (expr), stat, expr); @@ -923,8 +914,8 @@ tree conditional_expression (tree, tree type_tree, tree cond_expr, tree then_expr, tree else_expr, location_t location) { - if (type_tree == error_mark_node || cond_expr == error_mark_node - || then_expr == error_mark_node || else_expr == error_mark_node) + if (error_operand_p (type_tree) || error_operand_p (cond_expr) + || error_operand_p (then_expr) || error_operand_p (else_expr)) return error_mark_node; tree ret = build3_loc (location, COND_EXPR, type_tree, cond_expr, then_expr, else_expr); @@ -1021,12 +1012,12 @@ operator_to_tree_code (LazyBooleanOperator op) } } -/* Helper function for deciding if a tree is a floating point node. */ +/* Returns true if the type of EXP is a floating point type. + False otherwise. */ bool -is_floating_point (tree t) +is_floating_point (tree exp) { - auto tree_type = TREE_CODE (TREE_TYPE (t)); - return tree_type == REAL_TYPE || tree_type == COMPLEX_TYPE; + return FLOAT_TYPE_P (TREE_TYPE (exp)); } // Return an expression for the negation operation OP EXPR. @@ -1035,7 +1026,7 @@ negation_expression (NegationOperator op, tree expr_tree, location_t location) { /* Check if the expression is an error, in which case we return an error expression. */ - if (expr_tree == error_mark_node || TREE_TYPE (expr_tree) == error_mark_node) + if (error_operand_p (expr_tree)) return error_mark_node; /* For negation operators, the resulting type should be the same as its @@ -1071,7 +1062,7 @@ arithmetic_or_logical_expression (ArithmeticOrLogicalOperator op, tree left, { /* Check if either expression is an error, in which case we return an error expression. */ - if (left == error_mark_node || right == error_mark_node) + if (error_operand_p (left) || error_operand_p (right)) return error_mark_node; // unwrap the const decls if set @@ -1182,7 +1173,7 @@ arithmetic_or_logical_expression_checked (ArithmeticOrLogicalOperator op, { /* Check if either expression is an error, in which case we return an error expression. */ - if (left == error_mark_node || right == error_mark_node) + if (error_operand_p (left) || error_operand_p (right)) return error_mark_node; // FIXME: Add `if (!debug_mode)` @@ -1222,7 +1213,7 @@ comparison_expression (ComparisonOperator op, tree left_tree, tree right_tree, { /* Check if either expression is an error, in which case we return an error expression. */ - if (left_tree == error_mark_node || right_tree == error_mark_node) + if (error_operand_p (left_tree) || error_operand_p (right_tree)) return error_mark_node; /* For comparison operators, the resulting type should be boolean. */ @@ -1242,7 +1233,7 @@ lazy_boolean_expression (LazyBooleanOperator op, tree left_tree, { /* Check if either expression is an error, in which case we return an error expression. */ - if (left_tree == error_mark_node || right_tree == error_mark_node) + if (error_operand_p (left_tree) || error_operand_p (right_tree)) return error_mark_node; /* For lazy boolean operators, the resulting type should be the same as the @@ -1263,7 +1254,7 @@ constructor_expression (tree type_tree, bool is_variant, const std::vector<tree> &vals, int union_index, location_t location) { - if (type_tree == error_mark_node) + if (error_operand_p (type_tree)) return error_mark_node; vec<constructor_elt, va_gc> *init; @@ -1305,8 +1296,8 @@ constructor_expression (tree type_tree, bool is_variant, gcc_assert (field != NULL_TREE); field = DECL_CHAIN (field); } - if (TREE_TYPE (field) == error_mark_node || val == error_mark_node - || TREE_TYPE (val) == error_mark_node) + + if (TREE_TYPE (field) == error_mark_node || error_operand_p (val)) return error_mark_node; if (int_size_in_bytes (TREE_TYPE (field)) == 0) @@ -1336,8 +1327,7 @@ constructor_expression (tree type_tree, bool is_variant, { gcc_assert (field != NULL_TREE); tree val = (*p); - if (TREE_TYPE (field) == error_mark_node || val == error_mark_node - || TREE_TYPE (val) == error_mark_node) + if (TREE_TYPE (field) == error_mark_node || error_operand_p (val)) return error_mark_node; if (int_size_in_bytes (TREE_TYPE (field)) == 0) @@ -1376,7 +1366,7 @@ array_constructor_expression (tree type_tree, const std::vector<tree> &vals, location_t location) { - if (type_tree == error_mark_node) + if (error_operand_p (type_tree)) return error_mark_node; gcc_assert (indexes.size () == vals.size ()); @@ -1393,7 +1383,7 @@ array_constructor_expression (tree type_tree, tree index = size_int (indexes[i]); tree val = vals[i]; - if (index == error_mark_node || val == error_mark_node) + if (error_operand_p (index) || error_operand_p (val)) return error_mark_node; if (element_size == 0) @@ -1497,8 +1487,7 @@ array_initializer (tree fndecl, tree block, tree array_type, tree length, tree array_index_expression (tree array_tree, tree index_tree, location_t location) { - if (array_tree == error_mark_node || TREE_TYPE (array_tree) == error_mark_node - || index_tree == error_mark_node) + if (error_operand_p (array_tree) || error_operand_p (index_tree)) return error_mark_node; // A function call that returns a zero sized object will have been @@ -1520,7 +1509,7 @@ tree call_expression (tree fn, const std::vector<tree> &fn_args, tree chain_expr, location_t location) { - if (fn == error_mark_node || TREE_TYPE (fn) == error_mark_node) + if (error_operand_p (fn)) return error_mark_node; gcc_assert (FUNCTION_POINTER_TYPE_P (TREE_TYPE (fn))); @@ -1600,7 +1589,7 @@ tree init_statement (tree, Bvariable *var, tree init_tree) { tree var_tree = var->get_decl (); - if (var_tree == error_mark_node || init_tree == error_mark_node) + if (error_operand_p (var_tree) || error_operand_p (init_tree)) return error_mark_node; gcc_assert (TREE_CODE (var_tree) == VAR_DECL); @@ -1631,7 +1620,7 @@ init_statement (tree, Bvariable *var, tree init_tree) tree assignment_statement (tree lhs, tree rhs, location_t location) { - if (lhs == error_mark_node || rhs == error_mark_node) + if (error_operand_p (lhs) || error_operand_p (rhs)) return error_mark_node; // To avoid problems with GNU ld, we don't make zero-sized @@ -1656,14 +1645,14 @@ assignment_statement (tree lhs, tree rhs, location_t location) tree return_statement (tree fntree, tree val, location_t location) { - if (fntree == error_mark_node) + if (error_operand_p (fntree)) return error_mark_node; tree result = DECL_RESULT (fntree); - if (result == error_mark_node) + if (error_operand_p (result)) return error_mark_node; - if (val == error_mark_node) + if (error_operand_p (val)) return error_mark_node; tree set @@ -1681,8 +1670,8 @@ tree exception_handler_statement (tree try_stmt, tree except_stmt, tree finally_stmt, location_t location) { - if (try_stmt == error_mark_node || except_stmt == error_mark_node - || finally_stmt == error_mark_node) + if (error_operand_p (try_stmt) || error_operand_p (except_stmt) + || error_operand_p (finally_stmt)) return error_mark_node; if (except_stmt != NULL_TREE) @@ -1701,8 +1690,8 @@ tree if_statement (tree, tree cond_tree, tree then_tree, tree else_tree, location_t location) { - if (cond_tree == error_mark_node || then_tree == error_mark_node - || else_tree == error_mark_node) + if (error_operand_p (cond_tree) || error_operand_p (then_tree) + || error_operand_p (else_tree)) return error_mark_node; tree ret = build3_loc (location, COND_EXPR, void_type_node, cond_tree, then_tree, else_tree); @@ -1728,15 +1717,12 @@ exit_expression (tree cond_tree, location_t locus) tree compound_statement (tree s1, tree s2) { - tree stmt_list = NULL_TREE; - tree t = s1; - if (t == error_mark_node) - return error_mark_node; - append_to_statement_list (t, &stmt_list); - t = s2; - if (t == error_mark_node) + if (error_operand_p (s1) || error_operand_p (s2)) return error_mark_node; - append_to_statement_list (t, &stmt_list); + + tree stmt_list = NULL_TREE; + append_to_statement_list (s1, &stmt_list); + append_to_statement_list (s2, &stmt_list); // If neither statement has any side effects, stmt_list can be NULL // at this point. @@ -1752,11 +1738,9 @@ tree statement_list (const std::vector<tree> &statements) { tree stmt_list = NULL_TREE; - for (std::vector<tree>::const_iterator p = statements.begin (); - p != statements.end (); ++p) + for (tree t : statements) { - tree t = (*p); - if (t == error_mark_node) + if (error_operand_p (t)) return error_mark_node; append_to_statement_list (t, &stmt_list); } @@ -1808,12 +1792,13 @@ block (tree fndecl, tree enclosing, const std::vector<Bvariable *> &vars, *pp = block_tree; } + // Chain the variables of the scope together so they are all connected + // to the block. tree *pp = &BLOCK_VARS (block_tree); - for (std::vector<Bvariable *>::const_iterator pv = vars.begin (); - pv != vars.end (); ++pv) + for (Bvariable *bv : vars) { - *pp = (*pv)->get_decl (); - if (*pp != error_mark_node) + *pp = bv->get_decl (); + if (!error_operand_p (*pp)) pp = &DECL_CHAIN (*pp); } *pp = NULL_TREE; @@ -1832,11 +1817,9 @@ void block_add_statements (tree bind_tree, const std::vector<tree> &statements) { tree stmt_list = NULL_TREE; - for (std::vector<tree>::const_iterator p = statements.begin (); - p != statements.end (); ++p) + for (tree s : statements) { - tree s = (*p); - if (s != error_mark_node) + if (!error_operand_p (s)) append_to_statement_list (s, &stmt_list); } @@ -1914,8 +1897,7 @@ convert_tree (tree type_tree, tree expr_tree, location_t location) if (type_tree == TREE_TYPE (expr_tree)) return expr_tree; - if (type_tree == error_mark_node || expr_tree == error_mark_node - || TREE_TYPE (expr_tree) == error_mark_node) + if (error_operand_p (type_tree) || error_operand_p (expr_tree)) return error_mark_node; if (POINTER_TYPE_P (type_tree) || INTEGRAL_TYPE_P (type_tree) @@ -1944,7 +1926,7 @@ global_variable (const std::string &var_name, const std::string &asm_name, tree type_tree, bool is_external, bool is_hidden, bool in_unique_section, location_t location) { - if (type_tree == error_mark_node) + if (error_operand_p (type_tree)) return Bvariable::error_variable (); // The GNU linker does not like dynamic variables with zero size. @@ -1983,11 +1965,11 @@ global_variable (const std::string &var_name, const std::string &asm_name, void global_variable_set_init (Bvariable *var, tree expr_tree) { - if (expr_tree == error_mark_node) + if (error_operand_p (expr_tree)) return; gcc_assert (TREE_CONSTANT (expr_tree)); tree var_decl = var->get_decl (); - if (var_decl == error_mark_node) + if (error_operand_p (var_decl)) return; DECL_INITIAL (var_decl) = expr_tree; @@ -2008,7 +1990,7 @@ Bvariable * local_variable (tree function, const std::string &name, tree type_tree, Bvariable *decl_var, location_t location) { - if (type_tree == error_mark_node) + if (error_operand_p (type_tree)) return Bvariable::error_variable (); tree decl = build_decl (location, VAR_DECL, get_identifier_from_string (name), type_tree); @@ -2029,7 +2011,7 @@ Bvariable * parameter_variable (tree function, const std::string &name, tree type_tree, location_t location) { - if (type_tree == error_mark_node) + if (error_operand_p (type_tree)) return Bvariable::error_variable (); tree decl = build_decl (location, PARM_DECL, get_identifier_from_string (name), type_tree); @@ -2046,7 +2028,7 @@ Bvariable * static_chain_variable (tree fndecl, const std::string &name, tree type_tree, location_t location) { - if (type_tree == error_mark_node) + if (error_operand_p (type_tree)) return Bvariable::error_variable (); tree decl = build_decl (location, PARM_DECL, get_identifier_from_string (name), type_tree); @@ -2080,8 +2062,8 @@ temporary_variable (tree fndecl, tree bind_tree, tree type_tree, tree init_tree, tree *pstatement) { gcc_assert (fndecl != NULL_TREE); - if (type_tree == error_mark_node || init_tree == error_mark_node - || fndecl == error_mark_node) + if (error_operand_p (type_tree) || error_operand_p (init_tree) + || error_operand_p (fndecl)) { *pstatement = error_mark_node; return Bvariable::error_variable (); @@ -2198,13 +2180,13 @@ tree function (tree functype, const std::string &name, const std::string &asm_name, unsigned int flags, location_t location) { - if (functype != error_mark_node) - { - gcc_assert (FUNCTION_POINTER_TYPE_P (functype)); - functype = TREE_TYPE (functype); - } + if (error_operand_p (functype)) + return error_mark_node; + + gcc_assert (FUNCTION_POINTER_TYPE_P (functype)); + functype = TREE_TYPE (functype); tree id = get_identifier_from_string (name); - if (functype == error_mark_node || id == error_mark_node) + if (error_operand_p (id)) return error_mark_node; tree decl = build_decl (location, FUNCTION_DECL, id, functype); @@ -2242,8 +2224,8 @@ tree function_defer_statement (tree function, tree undefer_tree, tree defer_tree, location_t location) { - if (undefer_tree == error_mark_node || defer_tree == error_mark_node - || function == error_mark_node) + if (error_operand_p (undefer_tree) || error_operand_p (defer_tree) + || error_operand_p (function)) return error_mark_node; if (DECL_STRUCT_FUNCTION (function) == NULL) @@ -2275,16 +2257,15 @@ bool function_set_parameters (tree function, const std::vector<Bvariable *> ¶m_vars) { - if (function == error_mark_node) + if (error_operand_p (function)) return false; tree params = NULL_TREE; tree *pp = ¶ms; - for (std::vector<Bvariable *>::const_iterator pv = param_vars.begin (); - pv != param_vars.end (); ++pv) + for (Bvariable *bv : param_vars) { - *pp = (*pv)->get_decl (); - gcc_assert (*pp != error_mark_node); + *pp = bv->get_decl (); + gcc_assert (!error_operand_p (*pp)); pp = &DECL_CHAIN (*pp); } *pp = NULL_TREE; @@ -2309,23 +2290,19 @@ write_global_definitions (const std::vector<tree> &type_decls, // Convert all non-erroneous declarations into Gimple form. size_t i = 0; - for (std::vector<Bvariable *>::const_iterator p = variable_decls.begin (); - p != variable_decls.end (); ++p) + for (Bvariable *bv : variable_decls) { - tree v = (*p)->get_decl (); - if (v != error_mark_node) - { - defs[i] = v; - rust_preserve_from_gc (defs[i]); - ++i; - } + tree v = bv->get_decl (); + if (error_operand_p (v)) + continue; + defs[i] = v; + rust_preserve_from_gc (defs[i]); + ++i; } - for (std::vector<tree>::const_iterator p = type_decls.begin (); - p != type_decls.end (); ++p) + for (tree type_tree : type_decls) { - tree type_tree = (*p); - if (type_tree != error_mark_node && IS_TYPE_OR_DECL_P (type_tree)) + if (!error_operand_p (type_tree) && IS_TYPE_OR_DECL_P (type_tree)) { defs[i] = TYPE_NAME (type_tree); gcc_assert (defs[i] != NULL); @@ -2333,21 +2310,18 @@ write_global_definitions (const std::vector<tree> &type_decls, ++i; } } - for (std::vector<tree>::const_iterator p = constant_decls.begin (); - p != constant_decls.end (); ++p) + for (tree t : constant_decls) { - if ((*p) != error_mark_node) + if (!error_operand_p (t)) { - defs[i] = (*p); + defs[i] = t; rust_preserve_from_gc (defs[i]); ++i; } } - for (std::vector<tree>::const_iterator p = function_decls.begin (); - p != function_decls.end (); ++p) + for (tree decl : function_decls) { - tree decl = (*p); - if (decl != error_mark_node) + if (!error_operand_p (decl)) { rust_preserve_from_gc (decl); if (DECL_STRUCT_FUNCTION (decl) == NULL) diff --git a/gcc/rust/rust-session-manager.cc b/gcc/rust/rust-session-manager.cc index 15f21ef..48acbf34 100644 --- a/gcc/rust/rust-session-manager.cc +++ b/gcc/rust/rust-session-manager.cc @@ -611,7 +611,6 @@ Session::compile_crate (const char *filename) return; AST::CollectLangItems ().go (parsed_crate); - AST::DesugarQuestionMark ().go (parsed_crate); auto name_resolution_ctx = Resolver2_0::NameResolutionContext (); // expansion pipeline stage @@ -619,6 +618,7 @@ Session::compile_crate (const char *filename) expansion (parsed_crate, name_resolution_ctx); AST::DesugarForLoops ().go (parsed_crate); + AST::DesugarQuestionMark ().go (parsed_crate); rust_debug ("\033[0;31mSUCCESSFULLY FINISHED EXPANSION \033[0m"); if (options.dump_option_enabled (CompileOptions::EXPANSION_DUMP)) diff --git a/gcc/rust/typecheck/rust-hir-type-check-expr.cc b/gcc/rust/typecheck/rust-hir-type-check-expr.cc index 791795f..b2bcac0 100644 --- a/gcc/rust/typecheck/rust-hir-type-check-expr.cc +++ b/gcc/rust/typecheck/rust-hir-type-check-expr.cc @@ -16,6 +16,8 @@ // along with GCC; see the file COPYING3. If not see // <http://www.gnu.org/licenses/>. +#include "optional.h" +#include "rust-hir-expr.h" #include "rust-system.h" #include "rust-tyty-call.h" #include "rust-hir-type-check-struct-field.h" @@ -1154,6 +1156,94 @@ TypeCheckExpr::visit (HIR::FieldAccessExpr &expr) infered = lookup->get_field_type (); } +bool +is_default_fn (const MethodCandidate &candidate) +{ + if (candidate.candidate.is_impl_candidate ()) + { + auto *item = candidate.candidate.item.impl.impl_item; + + if (item->get_impl_item_type () == HIR::ImplItem::FUNCTION) + { + auto &fn = static_cast<HIR::Function &> (*item); + + return fn.is_default (); + } + } + + return false; +} + +void +emit_ambiguous_resolution_error (HIR::MethodCallExpr &expr, + std::set<MethodCandidate> &candidates) +{ + rich_location r (line_table, expr.get_method_name ().get_locus ()); + std::string rich_msg = "multiple " + + expr.get_method_name ().get_segment ().as_string () + + " found"; + + // We have to filter out default candidates + for (auto &c : candidates) + if (!is_default_fn (c)) + r.add_range (c.candidate.locus); + + r.add_fixit_replace (rich_msg.c_str ()); + + rust_error_at (r, ErrorCode::E0592, "duplicate definitions with name %qs", + expr.get_method_name ().get_segment ().as_string ().c_str ()); +} + +// We are allowed to have multiple candidates if they are all specializable +// functions or if all of them except one are specializable functions. +// In the later case, we just return a valid candidate without erroring out +// about ambiguity. If there are two or more specialized functions, then we +// error out. +// +// FIXME: The first case is not handled at the moment, so we error out +tl::optional<const MethodCandidate &> +handle_multiple_candidates (HIR::MethodCallExpr &expr, + std::set<MethodCandidate> &candidates) +{ + auto all_default = true; + tl::optional<const MethodCandidate &> found = tl::nullopt; + + for (auto &c : candidates) + { + if (!is_default_fn (c)) + { + all_default = false; + + // We haven't found a final candidate yet, so we can select + // this one. However, if we already have a candidate, then + // that means there are multiple non-default candidates - we + // must error out + if (!found) + { + found = c; + } + else + { + emit_ambiguous_resolution_error (expr, candidates); + return tl::nullopt; + } + } + } + + // None of the candidates were a non-default (specialized) function, so we + // error out + if (all_default) + { + rust_sorry_at (expr.get_locus (), + "cannot resolve method calls to non-specialized methods " + "(all function candidates are %qs)", + "default"); + return tl::nullopt; + } + + return found; +} + void TypeCheckExpr::visit (HIR::MethodCallExpr &expr) { @@ -1181,34 +1271,25 @@ TypeCheckExpr::visit (HIR::MethodCallExpr &expr) return; } - if (candidates.size () > 1) - { - rich_location r (line_table, expr.get_method_name ().get_locus ()); - std::string rich_msg - = "multiple " + expr.get_method_name ().get_segment ().as_string () - + " found"; + tl::optional<const MethodCandidate &> candidate = *candidates.begin (); - for (auto &c : candidates) - r.add_range (c.candidate.locus); + if (candidates.size () > 1) + candidate = handle_multiple_candidates (expr, candidates); - r.add_fixit_replace (rich_msg.c_str ()); + if (!candidate) + return; - rust_error_at ( - r, ErrorCode::E0592, "duplicate definitions with name %qs", - expr.get_method_name ().get_segment ().as_string ().c_str ()); - return; - } + auto found_candidate = *candidate; - auto candidate = *candidates.begin (); rust_debug_loc (expr.get_method_name ().get_locus (), "resolved method to: {%u} {%s} with [%lu] adjustments", - candidate.candidate.ty->get_ref (), - candidate.candidate.ty->debug_str ().c_str (), - (unsigned long) candidate.adjustments.size ()); + found_candidate.candidate.ty->get_ref (), + found_candidate.candidate.ty->debug_str ().c_str (), + (unsigned long) found_candidate.adjustments.size ()); // Get the adjusted self Adjuster adj (receiver_tyty); - TyTy::BaseType *adjusted_self = adj.adjust_type (candidate.adjustments); + TyTy::BaseType *adjusted_self = adj.adjust_type (found_candidate.adjustments); rust_debug ("receiver: %s adjusted self %s", receiver_tyty->debug_str ().c_str (), adjusted_self->debug_str ().c_str ()); @@ -1219,10 +1300,10 @@ TypeCheckExpr::visit (HIR::MethodCallExpr &expr) HirId autoderef_mappings_id = expr.get_receiver ().get_mappings ().get_hirid (); context->insert_autoderef_mappings (autoderef_mappings_id, - std::move (candidate.adjustments)); + std::move (found_candidate.adjustments)); - PathProbeCandidate &resolved_candidate = candidate.candidate; - TyTy::BaseType *lookup_tyty = candidate.candidate.ty; + PathProbeCandidate &resolved_candidate = found_candidate.candidate; + TyTy::BaseType *lookup_tyty = found_candidate.candidate.ty; NodeId resolved_node_id = resolved_candidate.is_impl_candidate () ? resolved_candidate.item.impl.impl_item->get_impl_mappings () @@ -1249,8 +1330,8 @@ TypeCheckExpr::visit (HIR::MethodCallExpr &expr) fn->prepare_higher_ranked_bounds (); rust_debug_loc (expr.get_locus (), "resolved method call to: {%u} {%s}", - candidate.candidate.ty->get_ref (), - candidate.candidate.ty->debug_str ().c_str ()); + found_candidate.candidate.ty->get_ref (), + found_candidate.candidate.ty->debug_str ().c_str ()); if (resolved_candidate.is_impl_candidate ()) { diff --git a/gcc/rust/util/rust-attribute-values.h b/gcc/rust/util/rust-attribute-values.h index 9ef5cc5..47e6a17 100644 --- a/gcc/rust/util/rust-attribute-values.h +++ b/gcc/rust/util/rust-attribute-values.h @@ -40,12 +40,14 @@ public: static constexpr auto &NO_MANGLE = "no_mangle"; static constexpr auto &REPR = "repr"; static constexpr auto &RUSTC_BUILTIN_MACRO = "rustc_builtin_macro"; + static constexpr auto &RUSTC_MACRO_TRANSPARENCY = "rustc_macro_transparency"; static constexpr auto &PATH = "path"; static constexpr auto &MACRO_USE = "macro_use"; static constexpr auto &MACRO_EXPORT = "macro_export"; static constexpr auto &PROC_MACRO = "proc_macro"; static constexpr auto &PROC_MACRO_DERIVE = "proc_macro_derive"; static constexpr auto &PROC_MACRO_ATTRIBUTE = "proc_macro_attribute"; + static constexpr auto &TARGET_FEATURE = "target_feature"; // From now on, these are reserved by the compiler and gated through // #![feature(rustc_attrs)] @@ -54,10 +56,35 @@ public: = "rustc_inherit_overflow_checks"; static constexpr auto &STABLE = "stable"; static constexpr auto &UNSTABLE = "unstable"; + + static constexpr auto &RUSTC_PROMOTABLE = "rustc_promotable"; static constexpr auto &RUSTC_CONST_STABLE = "rustc_const_stable"; static constexpr auto &RUSTC_CONST_UNSTABLE = "rustc_const_unstable"; + + static constexpr auto &RUSTC_SPECIALIZATION_TRAIT + = "rustc_specialization_trait"; + static constexpr auto &RUSTC_UNSAFE_SPECIALIZATION_MARKER + = "rustc_unsafe_specialization_marker"; + static constexpr auto &RUSTC_RESERVATION_IMPL = "rustc_reservation_impl"; + static constexpr auto &RUSTC_PAREN_SUGAR = "rustc_paren_sugar"; + static constexpr auto &RUSTC_NONNULL_OPTIMIZATION_GUARANTEED + = "rustc_nonnull_optimization_guaranteed"; + + static constexpr auto &RUSTC_LAYOUT_SCALAR_VALID_RANGE_START + = "rustc_layout_scalar_valid_range_start"; + static constexpr auto &MAY_DANGLE = "may_dangle"; static constexpr auto &PRELUDE_IMPORT = "prelude_import"; + static constexpr auto &TRACK_CALLER = "track_caller"; + + static constexpr auto &RUSTC_DIAGNOSTIC_ITEM = "rustc_diagnostic_item"; + static constexpr auto &RUSTC_ON_UNIMPLEMENTED = "rustc_on_unimplemented"; + + static constexpr auto &FUNDAMENTAL = "fundamental"; + + static constexpr auto &NON_EXHAUSTIVE = "non_exhaustive"; + + static constexpr auto &RUSTFMT = "rustfmt"; }; } // namespace Values } // namespace Rust diff --git a/gcc/rust/util/rust-attributes.cc b/gcc/rust/util/rust-attributes.cc index 03452c7..c77e99c 100644 --- a/gcc/rust/util/rust-attributes.cc +++ b/gcc/rust/util/rust-attributes.cc @@ -57,6 +57,7 @@ static const BuiltinAttrDefinition __definitions[] {Attrs::NO_MANGLE, CODE_GENERATION}, {Attrs::REPR, CODE_GENERATION}, {Attrs::RUSTC_BUILTIN_MACRO, EXPANSION}, + {Attrs::RUSTC_MACRO_TRANSPARENCY, EXPANSION}, {Attrs::PATH, EXPANSION}, {Attrs::MACRO_USE, NAME_RESOLUTION}, {Attrs::MACRO_EXPORT, NAME_RESOLUTION}, @@ -72,10 +73,29 @@ static const BuiltinAttrDefinition __definitions[] {Attrs::RUSTC_INHERIT_OVERFLOW_CHECKS, CODE_GENERATION}, {Attrs::STABLE, STATIC_ANALYSIS}, {Attrs::UNSTABLE, STATIC_ANALYSIS}, + // assuming we keep these for static analysis + {Attrs::RUSTC_PROMOTABLE, CODE_GENERATION}, {Attrs::RUSTC_CONST_STABLE, STATIC_ANALYSIS}, {Attrs::RUSTC_CONST_UNSTABLE, STATIC_ANALYSIS}, - {Attrs::PRELUDE_IMPORT, NAME_RESOLUTION}}; + {Attrs::PRELUDE_IMPORT, NAME_RESOLUTION}, + {Attrs::TRACK_CALLER, CODE_GENERATION}, + {Attrs::RUSTC_SPECIALIZATION_TRAIT, TYPE_CHECK}, + {Attrs::RUSTC_UNSAFE_SPECIALIZATION_MARKER, TYPE_CHECK}, + {Attrs::RUSTC_RESERVATION_IMPL, TYPE_CHECK}, + {Attrs::RUSTC_PAREN_SUGAR, TYPE_CHECK}, + {Attrs::RUSTC_NONNULL_OPTIMIZATION_GUARANTEED, TYPE_CHECK}, + + {Attrs::RUSTC_LAYOUT_SCALAR_VALID_RANGE_START, CODE_GENERATION}, + + {Attrs::PRELUDE_IMPORT, NAME_RESOLUTION}, + + {Attrs::RUSTC_DIAGNOSTIC_ITEM, STATIC_ANALYSIS}, + {Attrs::RUSTC_ON_UNIMPLEMENTED, STATIC_ANALYSIS}, + + {Attrs::FUNDAMENTAL, TYPE_CHECK}, + {Attrs::NON_EXHAUSTIVE, TYPE_CHECK}, + {Attrs::RUSTFMT, EXTERNAL}}; BuiltinAttributeMappings * BuiltinAttributeMappings::get () diff --git a/gcc/rust/util/rust-attributes.h b/gcc/rust/util/rust-attributes.h index c928c8e..7c7a1fc 100644 --- a/gcc/rust/util/rust-attributes.h +++ b/gcc/rust/util/rust-attributes.h @@ -40,7 +40,12 @@ enum CompilerPass HIR_LOWERING, TYPE_CHECK, STATIC_ANALYSIS, - CODE_GENERATION + CODE_GENERATION, + + // External Rust tooling attributes, like #[rustfmt::skip] + EXTERNAL, + + // Do we need to add something here for const fns? }; struct BuiltinAttrDefinition diff --git a/gcc/rust/util/rust-lang-item.cc b/gcc/rust/util/rust-lang-item.cc index a76cc7f..9aff31b 100644 --- a/gcc/rust/util/rust-lang-item.cc +++ b/gcc/rust/util/rust-lang-item.cc @@ -118,6 +118,7 @@ const BiMap<std::string, LangItem::Kind> Rust::LangItem::lang_items = {{ {"discriminant_kind", Kind::DISCRIMINANT_KIND}, {"discriminant_type", Kind::DISCRIMINANT_TYPE}, + {"manually_drop", Kind::MANUALLY_DROP}, }}; tl::optional<LangItem::Kind> diff --git a/gcc/rust/util/rust-lang-item.h b/gcc/rust/util/rust-lang-item.h index 8f3af36..67a5d9c 100644 --- a/gcc/rust/util/rust-lang-item.h +++ b/gcc/rust/util/rust-lang-item.h @@ -150,6 +150,8 @@ public: DISCRIMINANT_TYPE, DISCRIMINANT_KIND, + + MANUALLY_DROP, }; static const BiMap<std::string, Kind> lang_items; diff --git a/gcc/sanitizer.def b/gcc/sanitizer.def index 4b7c9dc..c5a9c2d 100644 --- a/gcc/sanitizer.def +++ b/gcc/sanitizer.def @@ -247,7 +247,7 @@ DEF_SANITIZER_BUILTIN(BUILT_IN_TSAN_INIT, "__tsan_init", DEF_SANITIZER_BUILTIN(BUILT_IN_TSAN_FUNC_ENTRY, "__tsan_func_entry", BT_FN_VOID_PTR, ATTR_NOTHROW_LEAF_LIST) DEF_SANITIZER_BUILTIN(BUILT_IN_TSAN_FUNC_EXIT, "__tsan_func_exit", - BT_FN_VOID_PTR, ATTR_NOTHROW_LEAF_LIST) + BT_FN_VOID, ATTR_NOTHROW_LEAF_LIST) DEF_SANITIZER_BUILTIN(BUILT_IN_TSAN_VPTR_UPDATE, "__tsan_vptr_update", BT_FN_VOID_PTR_PTR, ATTR_NOTHROW_LEAF_LIST) DEF_SANITIZER_BUILTIN(BUILT_IN_TSAN_READ1, "__tsan_read1", diff --git a/gcc/testsuite/ChangeLog b/gcc/testsuite/ChangeLog index c99e6df..521176a 100644 --- a/gcc/testsuite/ChangeLog +++ b/gcc/testsuite/ChangeLog @@ -1,3 +1,826 @@ +2025-04-20 H.J. Lu <hjl.tools@gmail.com> + + PR target/117863 + * gcc.dg/rtl/i386/vector_eq-2.c: New test. + * gcc.dg/rtl/i386/vector_eq-3.c: Likewise. + +2025-04-19 Thomas Schwinge <tschwinge@baylibre.com> + + PR testsuite/119508 + * rust/compile/nr2/compile.exp: Disable parallel testing. + +2025-04-19 Co-authored-by: Jeff Law <jlaw@ventanamicro.com> + + PR target/118410 + * gcc.target/riscv/pr118410-1.c: New test. + * gcc.target/riscv/pr118410-2.c: Likewise. + +2025-04-19 Andrew Pinski <quic_apinski@quicinc.com> + + * gcc.dg/pr118947-1.c: Use 1025 as the size of the buf. + * gcc.dg/pr78408-3.c: Likewise. + +2025-04-19 Andrew Pinski <quic_apinski@quicinc.com> + + PR rtl-optimization/111949 + * gcc.target/aarch64/bic-1.c: New test. + +2025-04-19 Jiaxun Yang <jiaxun.yang@flygoat.com> + + PR target/111814 + * gcc.target/sh/pr111814.c: New test. + +2025-04-19 Maciej W. Rozycki <macro@orcam.me.uk> + + * gcc.target/alpha/memcpy-nested-offset-long.c: New file. + * gcc.target/alpha/memcpy-nested-offset-quad.c: New file. + +2025-04-19 Steven G. Kargl <kargl@gcc.gnu.org> + + PR fortran/119836 + * gfortran.dg/do_concurrent_all_clauses.f90: Remove invalid + dg-error test. + * gfortran.dg/pr119836_1.f90: New test. + * gfortran.dg/pr119836_2.f90: New test. + * gfortran.dg/pr119836_3.f90: New test. + * gfortran.dg/pr119836_4.f90: New test. + +2025-04-18 Thomas Schwinge <tschwinge@baylibre.com> + + PR cobol/119818 + * cobol.dg/group2/FUNCTION_DATE___TIME_OMNIBUS.cob: + 'dg-set-target-env-var TZ UTC0'. + +2025-04-18 Jeff Law <jlaw@ventanamicro.com> + + * gcc.target/riscv/bext-ext-2.c: New test + +2025-04-18 Jonathan Yong <10walls@gmail.com> + + * g++.dg/abi/ref-temp1.C: Replicate some test based on + PE expectations. + * lib/target-supports.exp: New check_effective_target_pe. + +2025-04-18 Andrew Pinski <quic_apinski@quicinc.com> + + PR tree-optimization/87901 + * gcc.dg/tree-ssa/ssa-dse-53.c: New test. + * gcc.dg/tree-ssa/ssa-dse-54.c: New test. + +2025-04-18 Andrew Pinski <quic_apinski@quicinc.com> + + PR tree-optimization/87901 + * gcc.dg/tree-ssa/ssa-dse-52.c: New test. + +2025-04-18 Andrew Pinski <quic_apinski@quicinc.com> + + PR tree-optimization/118902 + * gcc.dg/tree-ssa/pr118902-1.c: New test. + +2025-04-18 Andrew Pinski <quic_apinski@quicinc.com> + + PR tree-optimization/118947 + * gcc.dg/pr118947-1.c: New test. + +2025-04-18 Andrew Pinski <quic_apinski@quicinc.com> + + PR tree-optimization/78408 + PR tree-optimization/118947 + * gcc.dg/pr78408-3.c: New test. + +2025-04-18 Dimitar Dimitrov <dimitar@dinux.eu> + + * gcc.dg/pr116357.c: Use sizeof(int) instead of alignof(int). + +2025-04-18 Alexey Merzlyakov <alexey.merzlyakov@samsung.com> + + PR middle-end/108016 + * gcc.target/riscv/pr108016.c: New test. + +2025-04-18 kelefth <konstantinos.eleftheriou@vrull.eu> + + PR rtl-optimization/119160 + * gcc.dg/pr119160.c: New test. + +2025-04-18 Xing Li <lixing@loongson.cn> + + * gcc.target/loongarch/vector/loongarch-vector.exp: Change + {dg-do-what-default} save and restore logical. + +2025-04-17 Jason Merrill <jason@redhat.com> + + * g++.dg/cpp2a/constexpr-dtor16.C: Adjust diagnostic. + * g++.dg/cpp2a/constexpr-dynamic10.C: Likewise. + +2025-04-17 Jason Merrill <jason@redhat.com> + + * g++.dg/cpp1y/constexpr-new.C: Adjust diagnostics. + * g++.dg/cpp1z/constexpr-asm-5.C: Likewise. + * g++.dg/cpp26/static_assert1.C: Likewise. + * g++.dg/cpp2a/constexpr-dtor7.C: Likewise. + * g++.dg/cpp2a/constexpr-new26.C: Likewise. + * g++.dg/cpp2a/constexpr-new3.C: Likewise. + * g++.dg/cpp2a/constinit14.C: Likewise. + +2025-04-17 Jason Merrill <jason@redhat.com> + + * g++.dg/cpp26/pack-indexing2.C: Adjust diagnostics. + * g++.dg/ext/type_pack_element2.C: Likewise. + * g++.dg/ext/type_pack_element4.C: Likewise. + +2025-04-17 Tamar Christina <tamar.christina@arm.com> + + PR tree-optimization/119351 + * gcc.target/aarch64/sve/pr119351.c: New test. + * gcc.target/aarch64/sve/pr119351_run.c: New test. + +2025-04-17 Jakub Jelinek <jakub@redhat.com> + + PR target/119834 + * g++.target/s390/pr119834.C: New test. + +2025-04-17 Iain Buclaw <ibuclaw@gdcproject.org> + + * gdc.test/fail_compilation/test21247.d: New test. + * gdc.test/fail_compilation/test21247b.d: New test. + +2025-04-17 Jason Merrill <jason@redhat.com> + + PR c++/113360 + * g++.dg/cpp23/constexpr-nonlit18.C: Remove redundant message. + * g++.dg/cpp1y/constexpr-diag2.C: New test. + * g++.dg/cpp1y/pr63996.C: Adjust expected errors. + * g++.dg/template/explicit-args6.C: Likewise. + * g++.dg/cpp0x/constexpr-ice21.C: Likewise. + +2025-04-16 Alexandre Oliva <oliva@adacore.com> + + * gcc.dg/ipa/ipa-sra-19.c: Add -Wno-psabi on ppc-elf too. + +2025-04-16 Peter Bergner <bergner@linux.ibm.com> + + PR tree-optimization/112822 + * g++.dg/pr112822.C: Replace altivec vector attribute with a generic + vector attribute. + +2025-04-16 Eric Botcazou <ebotcazou@gcc.gnu.org> + + * gnat.dg/opt105.adb: New test. + * gnat.dg/opt105_pkg.ads, gnat.dg/opt105_pkg.adb: New helper. + +2025-04-16 Jason Merrill <jason@redhat.com> + + PR c++/114772 + PR c++/101180 + * g++.dg/ext/pragma-target2.C: New test. + +2025-04-16 Jason Merrill <jason@redhat.com> + + PR c++/116954 + * g++.dg/warn/Wformat-3.C: New test. + +2025-04-16 Ard Biesheuvel <ardb@kernel.org> + + PR target/119386 + * gcc.target/i386/pr119386-3.c: New test. + +2025-04-16 Ard Biesheuvel <ardb@kernel.org> + + PR target/119386 + * gcc.target/i386/pr119386-1.c: New test. + * gcc.target/i386/pr119386-2.c: New test. + +2025-04-16 Harald Anlauf <anlauf@gmx.de> + + PR fortran/106948 + * gfortran.dg/pure_formal_proc_4.f90: New test. + +2025-04-16 Jan Hubicka <hubicka@ucw.cz> + + * g++.dg/lto/pr119614_0.C: New test. + +2025-04-16 Tamar Christina <tamar.christina@arm.com> + + PR target/119286 + * gcc.dg/vect/vect-early-break_18.c: Force -march=gfx908 for amdgcn. + +2025-04-16 Tamar Christina <tamar.christina@arm.com> + + PR tree-optimization/119351 + * gcc.target/aarch64/sve/peel_ind_10.c: New test. + * gcc.target/aarch64/sve/peel_ind_10_run.c: New test. + * gcc.target/aarch64/sve/peel_ind_5.c: New test. + * gcc.target/aarch64/sve/peel_ind_5_run.c: New test. + * gcc.target/aarch64/sve/peel_ind_6.c: New test. + * gcc.target/aarch64/sve/peel_ind_6_run.c: New test. + * gcc.target/aarch64/sve/peel_ind_7.c: New test. + * gcc.target/aarch64/sve/peel_ind_7_run.c: New test. + * gcc.target/aarch64/sve/peel_ind_8.c: New test. + * gcc.target/aarch64/sve/peel_ind_8_run.c: New test. + * gcc.target/aarch64/sve/peel_ind_9.c: New test. + * gcc.target/aarch64/sve/peel_ind_9_run.c: New test. + +2025-04-16 Jakub Jelinek <jakub@redhat.com> + + PR middle-end/119808 + * gcc.dg/bitint-121.c: New test. + +2025-04-16 Jesse Huang <jesse.huang@sifive.com> + + * gcc.target/riscv/gnu-property-align-rv32.c: New file. + * gcc.target/riscv/gnu-property-align-rv64.c: New file. + +2025-04-16 Kito Cheng <kito.cheng@sifive.com> + + * gcc.target/riscv/jump-table-large-code-model.c: New test. + +2025-04-16 Jakub Jelinek <jakub@redhat.com> + + PR tree-optimization/116093 + * gcc.dg/bitint-122.c: New test. + +2025-04-16 Alice Carlotti <alice.carlotti@arm.com> + + * gcc.target/aarch64/acle/rwsr-ungated.c: New test. + +2025-04-15 Iain Buclaw <ibuclaw@gdcproject.org> + + PR d/119826 + * gdc.dg/debug/imports/pr119826b.d: New test. + * gdc.dg/debug/pr119826.d: New test. + +2025-04-15 Nathaniel Shead <nathanieloshead@gmail.com> + + PR c++/119755 + * g++.dg/modules/lambda-10_a.H: New test. + * g++.dg/modules/lambda-10_b.C: New test. + +2025-04-15 Jakub Jelinek <jakub@redhat.com> + + * gcc.dg/completion-2.c: Expect also -flto-partition=default line. + +2025-04-15 Qing Zhao <qing.zhao@oracle.com> + + PR c/119717 + * gcc.dg/pr119717.c: New test. + +2025-04-15 H.J. Lu <hjl.tools@gmail.com> + + PR target/119784 + * gcc.target/i386/apx-interrupt-1.c: Expect 31 .cfi_restore + directives. + +2025-04-15 Vineet Gupta <vineetg@rivosinc.com> + + PR target/119533 + * go.dg/pr119533-riscv.go: New test. + * go.dg/pr119533-riscv-2.go: New test. + +2025-04-15 Robin Dapp <rdapp@ventanamicro.com> + + PR target/119547 + * gcc.target/riscv/rvv/vsetvl/avl_single-68.c: xfail. + * g++.target/riscv/rvv/autovec/pr119547.C: New test. + * g++.target/riscv/rvv/autovec/pr119547-2.C: New test. + * gcc.target/riscv/rvv/vsetvl/vlmax_switch_vtype-10.c: Adjust. + +2025-04-15 Tobias Burnus <tburnus@baylibre.com> + + * gfortran.dg/gomp/map-alloc-comp-1.f90: Remove dg-error. + * gfortran.dg/gomp/polymorphic-mapping-2.f90: Update warn wording. + * gfortran.dg/gomp/polymorphic-mapping.f90: Change expected + diagnostic; some tests moved to ... + * gfortran.dg/gomp/polymorphic-mapping-1.f90: ... here as new test. + * gfortran.dg/gomp/polymorphic-mapping-3.f90: New test. + * gfortran.dg/gomp/polymorphic-mapping-4.f90: New test. + * gfortran.dg/gomp/polymorphic-mapping-5.f90: New test. + +2025-04-15 Martin Jambor <mjambor@suse.cz> + Jakub Jelinek <jakub@redhat.com> + + PR ipa/119803 + * gcc.dg/ipa/pr119803.c: New test. + +2025-04-15 Iain Buclaw <ibuclaw@gdcproject.org> + + PR d/119799 + * gdc.dg/import-c/pr119799.d: New test. + * gdc.dg/import-c/pr119799c.c: New test. + +2025-04-15 Patrick Palka <ppalka@redhat.com> + + PR c++/119807 + PR c++/112288 + * g++.dg/template/friend86.C: New test. + * g++.dg/template/friend87.C: New test. + +2025-04-15 Iain Buclaw <ibuclaw@gdcproject.org> + + PR d/119817 + * gdc.dg/debug/imports/m119817/a.d: New test. + * gdc.dg/debug/imports/m119817/b.d: New test. + * gdc.dg/debug/imports/m119817/package.d: New test. + * gdc.dg/debug/pr119817.d: New test. + +2025-04-15 Jakub Jelinek <jakub@redhat.com> + + PR sanitizer/119801 + * c-c++-common/tsan/pr119801.c: New test. + +2025-04-15 Jonathan Yong <10walls@gmail.com> + + * gcc.dg/Wbuiltin-declaration-mismatch-4.c: Make diagnostic + accept long long. + +2025-04-15 Jakub Jelinek <jakub@redhat.com> + + PR ipa/119318 + * gcc.dg/ipa/pr119318.c: Remove dg-additional-options, add -w to + dg-options. + +2025-04-15 Jason Merrill <jason@redhat.com> + + PR c++/113835 + * g++.dg/cpp2a/constexpr-vector1.C: New test. + +2025-04-14 Thomas Schwinge <tschwinge@baylibre.com> + + PR target/118794 + * g++.target/gcn/exceptions-bad_cast-2.C: Set + '-mno-fake-exceptions'. + * g++.target/gcn/exceptions-pr118794-1.C: Likewise. + * g++.target/gcn/exceptions-throw-2.C: Likewise. + * g++.target/nvptx/exceptions-bad_cast-2.C: Likewise. + * g++.target/nvptx/exceptions-pr118794-1.C: Likewise. + * g++.target/nvptx/exceptions-throw-2.C: Likewise. + * g++.target/gcn/exceptions-bad_cast-2_-mfake-exceptions.C: New. + * g++.target/gcn/exceptions-pr118794-1_-mfake-exceptions.C: + Likewise. + * g++.target/gcn/exceptions-throw-2_-mfake-exceptions.C: Likewise. + * g++.target/nvptx/exceptions-bad_cast-2_-mfake-exceptions.C: + Likewise. + * g++.target/nvptx/exceptions-pr118794-1_-mfake-exceptions.C: + Likewise. + * g++.target/nvptx/exceptions-throw-2_-mfake-exceptions.C: + Likewise. + +2025-04-14 Thomas Schwinge <tschwinge@baylibre.com> + + * g++.target/gcn/exceptions-throw-3.C: New. + * g++.target/nvptx/exceptions-throw-3.C: Likewise. + +2025-04-14 Thomas Schwinge <tschwinge@baylibre.com> + + * g++.target/gcn/exceptions-throw-2.C: New. + * g++.target/nvptx/exceptions-throw-2.C: Likewise. + +2025-04-14 Thomas Schwinge <tschwinge@baylibre.com> + + * g++.target/gcn/exceptions-throw-1.C: New. + * g++.target/nvptx/exceptions-throw-1.C: Likewise. + +2025-04-14 Thomas Schwinge <tschwinge@baylibre.com> + + * g++.target/gcn/exceptions-bad_cast-3.C: New. + * g++.target/nvptx/exceptions-bad_cast-3.C: Likewise. + +2025-04-14 Thomas Schwinge <tschwinge@baylibre.com> + + * g++.target/gcn/exceptions-bad_cast-2.C: New. + * g++.target/nvptx/exceptions-bad_cast-2.C: Likewise. + +2025-04-14 Thomas Schwinge <tschwinge@baylibre.com> + + * g++.target/gcn/exceptions-bad_cast-1.C: New. + * g++.target/nvptx/exceptions-bad_cast-1.C: Likewise. + +2025-04-14 Thomas Schwinge <tschwinge@baylibre.com> + + PR target/118794 + * g++.target/gcn/exceptions-pr118794-1.C: New. + * g++.target/nvptx/exceptions-pr118794-1.C: Likewise. + +2025-04-14 Thomas Schwinge <tschwinge@baylibre.com> + + PR c++/119692 + * g++.target/gcn/pr119692-1-1.C: New. + * g++.target/nvptx/pr119692-1-1.C: Likewise. + +2025-04-14 Thomas Schwinge <tschwinge@baylibre.com> + + * g++.target/gcn/gcn.exp: New. + +2025-04-14 Thomas Schwinge <tschwinge@baylibre.com> + + * lib/gcc-dg.exp (${tool}_load): Polish 'dg-output-file' test + logs. + +2025-04-14 Jakub Jelinek <jakub@redhat.com> + + PR ipa/119318 + * gcc.dg/ipa/pr119530.c (d): Change type from char to signed char. + (e): Change argument type from long to long long. + +2025-04-14 beamandala <mandalapubhavesh@gmail.com> + + * rust/compile/track_caller.rs: New test. + +2025-04-14 Owen Avery <powerboat9.gamer@gmail.com> + + * rust/compile/enum_discriminant2.rs: New test. + +2025-04-14 Arthur Cohen <arthur.cohen@embecosm.com> + + * rust/compile/format_args_extra_comma.rs: New test. + +2025-04-14 Arthur Cohen <arthur.cohen@embecosm.com> + + * rust/compile/macros/mbe/macro-issue3709-1.rs: New test. + * rust/compile/macros/mbe/macro-issue3709-2.rs: New test. + * rust/compile/macros/mbe/macro-issue3693.rs: New file. + +2025-04-14 Arthur Cohen <arthur.cohen@embecosm.com> + + * rust/compile/macros/mbe/macro-issue3708.rs: New test. + +2025-04-14 Arthur Cohen <arthur.cohen@embecosm.com> + + * rust/execute/torture/min_specialization2.rs: New test. + * rust/execute/torture/min_specialization3.rs: New test. + +2025-04-14 Andrew Pinski <quic_apinski@quicinc.com> + + PR tree-optimization/118476 + * gcc.dg/torture/pr118476-1.c: New test. + +2025-04-14 Patrick Palka <ppalka@redhat.com> + + PR c++/99214 + * g++.dg/concepts/diagnostic20.C: New test. + +2025-04-14 H.J. Lu <hjl.tools@gmail.com> + + PR target/119784 + * gcc.target/i386/pr119784a.c: New test. + * gcc.target/i386/pr119784b.c: Likewise. + +2025-04-14 Martin Jambor <mjambor@suse.cz> + + PR ipa/119318 + * gcc.dg/ipa/pr119318.c: New test. + * gcc.dg/ipa/pr119530.c: Likwise. + +2025-04-14 Richard Biener <rguenther@suse.de> + + PR tree-optimization/119757 + * gcc.dg/vect/pr119757.c: New testcase. + +2025-04-14 Richard Biener <rguenther@suse.de> + + PR tree-optimization/119778 + * g++.dg/torture/pr119778.C: New testcase. + +2025-04-14 Gaius Mulley <gaiusmod2@gmail.com> + + PR modula2/119779 + * gm2.dg/doc/examples/pass/doc-examples-pass.exp: New test. + * gm2.dg/doc/examples/pass/exampleadd.mod: New test. + * gm2.dg/doc/examples/pass/exampleadd2.mod: New test. + * gm2.dg/doc/examples/pass/hello.mod: New test. + * gm2.dg/doc/examples/pass/hellopim.mod: New test. + +2025-04-14 Eric Botcazou <ebotcazou@adacore.com> + + PR lto/119792 + * gnat.dg/lto29.adb: New test. + * gnat.dg/lto29_pkg.ads: New helper. + +2025-04-13 Jerry DeLisle <jvdelisle@gcc.gnu.org> + + PR libfortran/119502 + * gfortran.dg/pr119502.f90: New test. + +2025-04-13 Nathaniel Shead <nathanieloshead@gmail.com> + + * g++.dg/modules/noexcept-4_a.H: New test. + * g++.dg/modules/noexcept-4_b.C: New test. + +2025-04-13 Nathaniel Shead <nathanieloshead@gmail.com> + + * g++.dg/modules/lambda-8_b.C: Adjust error. + * g++.dg/modules/leg-merge-4_c.C: Likewise. + +2025-04-13 Thomas Koenig <tkoenig@gcc.gnu.org> + + PR fortran/119669 + * gfortran.dg/interface_59.f90: New test. + +2025-04-12 Iain Buclaw <ibuclaw@gdcproject.org> + + PR d/119761 + * gdc.dg/import-c/import-c.exp: New test. + * gdc.dg/import-c/pr119761.d: New test. + * gdc.dg/import-c/pr119761c.c: New test. + +2025-04-12 Iain Buclaw <ibuclaw@gdcproject.org> + + PR d/109023 + * gdc.dg/torture/imports/pr109023.d: New test. + * gdc.dg/torture/pr109023.d: New test. + +2025-04-12 Iain Buclaw <ibuclaw@gdcproject.org> + + PR d/119758 + * gdc.dg/driver_fonly1.d: New test. + * gdc.dg/driver_fonly2.d: New test. + * gdc.dg/driver_fonly3.d: New test. + * gdc.dg/imports/fonly.d: New test. + +2025-04-12 Andrew Pinski <quic_apinski@quicinc.com> + + PR testsuite/117706 + * gcc.dg/ira-shrinkwrap-prep-1.c: Unxfail for i?68-*-* and x86_64-*-*. + * gcc.dg/ira-shrinkwrap-prep-2.c: Likewise. + +2025-04-12 Patrick Palka <ppalka@redhat.com> + + PR c++/116416 + * g++.dg/cpp1y/constexpr-prvalue1.C: Adjust to instead inspect + the 'original' dump. + * g++.dg/cpp1y/constexpr-prvalue1a.C: New test. + +2025-04-12 Jakub Jelinek <jakub@redhat.com> + + PR tree-optimization/119722 + * gcc.dg/torture/bitint-77.c: New test. + +2025-04-12 Eric Botcazou <ebotcazou@adacore.com> + + * gnat.dg/renaming17.adb: New test. + +2025-04-12 Bob Dubner <rdubner@symas.com> + + PR cobol/119694 + * cobol.dg/group2/ACCEPT_DATE___DAY_and_intrinsic_functions__2_.cob: GCOBOL_CURRENT_DATE. + * cobol.dg/group2/ACCEPT_FROM_TIME___DATE___DAY___DAY-OF-WEEK__2_.cob: Likewise + * cobol.dg/group2/FUNCTION_DATE___TIME_OMNIBUS.cob: Likewise + +2025-04-11 Christophe Lyon <christophe.lyon@linaro.org> + + * lib/target-supports.exp: Rename arm_v8_1_lob_ok into + arm_v8_1m_lob_hw. + Rename arm_thumb2_no_arm_v8_1_lob into + arm_thumb2_no_arm_v8_1m_lob. + Rename arm_thumb2_ok_no_arm_v8_1_lob into + arm_thumb2_ok_no_arm_v8_1m_lob. + * gcc.target/arm/lob1.c: Likewise. + * gcc.target/arm/lob6.c: Likewise. + * gcc.target/arm/ivopts.c: Likewise. + * gcc.target/arm/unsigned-extend-2.c: Likewise. + +2025-04-11 Andrew Pinski <quic_apinski@quicinc.com> + + PR rtl-optimization/118502 + * g++.dg/opt/shrink-wrapping-vector-1.C: New test. + +2025-04-11 Jeff Law <jlaw@ventanamicro.com> + + * gcc.target/riscv/rvv/base/pr115068-run.c: Turn off pedantic diagnostics. + * gcc.target/riscv/rvv/base/pr115068.c: Likewise. + * gcc.target/riscv/rvv/base/vwaddsub-1.c: Likewise. + +2025-04-11 Jason Merrill <jason@redhat.com> + + PR c++/114970 + * g++.dg/opt/is_constant_evaluated4.C: New test. + +2025-04-11 Jakub Jelinek <jakub@redhat.com> + + PR tree-optimization/119707 + * gcc.dg/torture/bitint-76.c: New test. + +2025-04-11 Jennifer Schmitz <jschmitz@nvidia.com> + + PR tree-optimization/119706 + * g++.target/aarch64/sve/pr119706.C: New test. + +2025-04-11 Jonathan Yong <10walls@gmail.com> + + PR target/113633 + * gcc.dg/bf-ms-attrib.c: Fix expected __ms_struct__ layout + size. + +2025-04-11 Jonathan Yong <10walls@gmail.com> + + * c-c++-common/analyzer/realloc-1.c: Make diagnostic accept + long long for __builtin_realloc warning. + +2025-04-10 Jason Merrill <jason@redhat.com> + + PR c++/119345 + * g++.dg/cpp2a/lambda-targ14.C: New test. + +2025-04-10 Patrick Palka <ppalka@redhat.com> + + PR c++/119687 + * g++.dg/cpp23/class-deduction-inherited8.C: New test. + +2025-04-10 Bob Dubner <rdubner@symas.com> + + * cobol.dg/group2/Dynamic_reference_modification.cob: New testcase. + * cobol.dg/group2/Length_overflow__1_.cob: Likewise. + * cobol.dg/group2/Length_overflow__2_.cob: Likewise. + * cobol.dg/group2/Length_overflow_with_offset__1_.cob: Likewise. + * cobol.dg/group2/Length_overflow_with_offset__2_.cob: Likewise. + * cobol.dg/group2/Length_overflow_with_offset__3_.cob: Likewise. + * cobol.dg/group2/Offset_overflow.cob: Likewise. + * cobol.dg/group2/Offset_underflow.cob: Likewise. + * cobol.dg/group2/Refmod__comparisons_inside_numeric-display.cob: Likewise. + * cobol.dg/group2/Refmod_sources_are_figurative_constants.cob: Likewise. + * cobol.dg/group2/Static_reference_modification.cob: Likewise. + * cobol.dg/group2/Dynamic_reference_modification.out: New known-good result. + * cobol.dg/group2/Length_overflow__1_.out: Likewise. + * cobol.dg/group2/Length_overflow__2_.out: Likewise. + * cobol.dg/group2/Length_overflow_with_offset__1_.out: Likewise. + * cobol.dg/group2/Length_overflow_with_offset__2_.out: Likewise. + * cobol.dg/group2/Length_overflow_with_offset__3_.out: Likewise. + * cobol.dg/group2/Offset_overflow.out: Likewise. + * cobol.dg/group2/Offset_underflow.out: Likewise. + * cobol.dg/group2/Refmod__comparisons_inside_numeric-display.out: Likewise. + * cobol.dg/group2/Refmod_sources_are_figurative_constants.out: Likewise. + * cobol.dg/group2/Static_reference_modification.out: Likewise. + +2025-04-10 Jason Merrill <jason@redhat.com> + + PR c++/119175 + * g++.dg/cpp2a/concepts-lambda23.C: New test. + +2025-04-10 Iain Sandoe <iain@sandoe.co.uk> + + * lib/cobol.exp: Add libquadmath paths. + +2025-04-10 Richard Sandiford <richard.sandiford@arm.com> + + PR tree-optimization/119399 + * gcc.dg/vect/pr119399.c: New test. + +2025-04-10 Kito Cheng <kito.cheng@sifive.com> + + * gcc.target/riscv/rvv/base/abi-callee-saved-1-fixed-1.c: Include local + riscv_vector.h. + * gcc.target/riscv/rvv/base/abi-callee-saved-1-fixed-2.c: Ditto. + * gcc.target/riscv/rvv/base/abi-callee-saved-1-save-restore.c: Ditto. + * gcc.target/riscv/rvv/base/abi-callee-saved-1-zcmp.c: Ditto. + * gcc.target/riscv/rvv/base/abi-callee-saved-1.c: Ditto. + * gcc.target/riscv/rvv/base/abi-callee-saved-2-save-restore.c: Ditto. + * gcc.target/riscv/rvv/base/abi-callee-saved-2-zcmp.c: Ditto. + * gcc.target/riscv/rvv/base/abi-callee-saved-2.c: Ditto. + * gcc.target/riscv/rvv/base/bug-10-2.c: Ditto. + * gcc.target/riscv/rvv/base/bug-10.c: Ditto. + * gcc.target/riscv/rvv/base/bug-7.c: Ditto. + * gcc.target/riscv/rvv/base/bug-8.c: Ditto. + * gcc.target/riscv/rvv/base/bug-9.c: Ditto. + * gcc.target/riscv/rvv/base/pr110943.c: Ditto. + * gcc.target/riscv/rvv/base/pr112431-21.c: Ditto. + * gcc.target/riscv/rvv/base/pr114639-1.c: Ditto. + * gcc.target/riscv/rvv/base/pr115068.c: Ditto. + * gcc.target/riscv/rvv/base/pr117286.c: Ditto. + * gcc.target/riscv/rvv/base/pr117544.c: Ditto. + * gcc.target/riscv/rvv/base/pr117955.c: Ditto. + * gcc.target/riscv/rvv/base/pr118872.c: Ditto. + * gcc.target/riscv/rvv/base/vlmul_ext-1.c: Ditto. + * gcc.target/riscv/rvv/base/vssubu-1.c: Ditto. + * gcc.target/riscv/rvv/base/vssubu-2.c: Ditto. + * gcc.target/riscv/rvv/base/vwaddsub-1.c: Ditto. + * gcc.target/riscv/rvv/vsetvl/pr111234.c: Ditto. + * gcc.target/riscv/rvv/vsetvl/pr115214.c: Ditto. + * gcc.target/riscv/rvv/vsetvl/vsetvl-24.c: Ditto. + * gcc.target/riscv/rvv/vsetvl/vsetvl_bug-3.c: Ditto. + * gcc.target/riscv/rvv/vsetvl/vsetvl_bug-4.c: Ditto. + * gcc.target/riscv/rvv/xtheadvector/pr116591.c: Ditto. + * gcc.target/riscv/rvv/xtheadvector/pr116592.c: Ditto. + * gcc.target/riscv/rvv/xtheadvector/pr118357.c: Ditto. + * gcc.target/riscv/rvv/xtheadvector/vsext.c: Ditto. + * gcc.target/riscv/rvv/xtheadvector/vzext.c: Ditto. + +2025-04-09 Patrick Palka <ppalka@redhat.com> + + PR c++/119574 + * g++.dg/cpp2a/lambda-targ13.C: New test. + * g++.dg/cpp2a/lambda-targ13a.C: New test. + * g++.dg/cpp2a/lambda-targ13b.C: New test. + +2025-04-09 Jakub Jelinek <jakub@redhat.com> + + PR target/119664 + * gcc.dg/pr119664.c: New test. + +2025-04-09 Andrew Pinski <quic_apinski@quicinc.com> + + PR middle-end/116595 + * g++.target/aarch64/sve/pr116595.C: New test. + +2025-04-09 Iain Buclaw <ibuclaw@gdcproject.org> + + PR d/118309 + * gdc.dg/debug/dwarf2/pr118309.d: New test. + +2025-04-09 Harald Anlauf <anlauf@gmx.de> + + PR fortran/119656 + * gfortran.dg/optional_absent_13.f90: New test. + +2025-04-09 Jeff Law <jlaw@ventanamicro.com> + + * gcc.target/riscv/sat/sat_s_sub-1-i64.c: Update expected output. + * gcc.target/riscv/sat/sat_s_sub-2-i64.c: Likewise. + * gcc.target/riscv/sat/sat_s_sub-3-i64.c: Likewise. + * gcc.target/riscv/sat/sat_s_sub-4-i64.c: Likewise. + +2025-04-09 Uros Bizjak <ubizjak@gmail.com> + + * gcc.target/i386/pr67215-1.c: Correctly escape + asterisk in scan-assembler dirctive. + * gcc.target/i386/pr67215-2.c: Ditto. + +2025-04-09 Jonathan Yong <10walls@gmail.com> + + PR analyzer/113253 + * gcc.dg/analyzer/deref-before-check-pr113253.c: + (ptrdiff_t): use stddef.h type. + (uintptr_t): ditto. + (EMACS_INT): ditto. + (set_marker_internal): Add dummy 0 to suppress -Wreturn-type. + +2025-04-09 Jeff Law <jlaw@ventanamicro.com> + + * gcc.target/riscv/rvv/autovec/pr117722.c: Adjust expected output. + +2025-04-09 Richard Biener <rguenther@suse.de> + + PR rtl-optimization/119689 + PR rtl-optimization/115568 + * g++.target/i386/pr119689.C: New testcase. + +2025-04-09 Pan Li <pan2.li@intel.com> + + Revert: + 2025-03-15 Pan Li <pan2.li@intel.com> + + * gcc.target/riscv/rvv/autovec/cond/cond_widen_complicate-3.c: Removed. + * gcc.target/riscv/rvv/autovec/cond/cond_widen_complicate-3-f16.c: New test. + * gcc.target/riscv/rvv/autovec/cond/cond_widen_complicate-3-f32.c: New test. + * gcc.target/riscv/rvv/autovec/cond/cond_widen_complicate-3-i16.c: New test. + * gcc.target/riscv/rvv/autovec/cond/cond_widen_complicate-3-i32.c: New test. + * gcc.target/riscv/rvv/autovec/cond/cond_widen_complicate-3-i8.c: New test. + * gcc.target/riscv/rvv/autovec/cond/cond_widen_complicate-3-u16.c: New test. + * gcc.target/riscv/rvv/autovec/cond/cond_widen_complicate-3-u32.c: New test. + * gcc.target/riscv/rvv/autovec/cond/cond_widen_complicate-3-u8.c: New test. + * gcc.target/riscv/rvv/autovec/cond/cond_widen_complicate-3.h: New test. + +2025-04-09 Jakub Jelinek <jakub@redhat.com> + + PR preprocessor/118674 + * gcc.dg/cpp/pr118674.c: New test. + +2025-04-09 Robin Dapp <rdapp@ventanamicro.com> + + * g++.target/riscv/rvv/autovec/pr116595.C: Add -mabi. + +2025-04-09 Robin Dapp <rdapp@ventanamicro.com> + + PR middle-end/116595 + * g++.target/riscv/rvv/autovec/pr116595.C: New test. + +2025-04-09 Paul Thomas <pault@gcc.gnu.org> + and Harald Anlauf <anlauf@gcc.gnu.org> + + PR fortran/119460 + * gfortran.dg/reduce_2.f90: Add test to check that deferred len + characters cannot slip through. + * gfortran.dg/reduce_3.f90: New test + * gfortran.dg/reduce_4.f90: New test + +2025-04-09 Tobias Burnus <tburnus@baylibre.com> + + PR fortran/101602 + * gfortran.dg/do_concurrent_8_f2023.f90: Update for removed 'sorry, + unimplemented'. + * gfortran.dg/do_concurrent_9.f90: Likewise. + * gfortran.dg/do_concurrent_all_clauses.f90: Likewise. + * gfortran.dg/do_concurrent_local_init.f90: Likewise. + * gfortran.dg/do_concurrent_locality_specs.f90: Likewise. + * gfortran.dg/do_concurrent_11.f90: New test. + * gfortran.dg/do_concurrent_12.f90: New test. + * gfortran.dg/do_concurrent_13.f90: New test. + * gfortran.dg/do_concurrent_14.f90: New test. + * gfortran.dg/do_concurrent_15.f90: New test. + +2025-04-09 Jason Merrill <jason@redhat.com> + + PR c++/118698 + * g++.dg/cpp2a/concepts-lambda22.C: New test. + 2025-04-08 Richard Sandiford <richard.sandiford@arm.com> * gcc.dg/guality/pr36728-3.c: Update XFAILs for aarch64. diff --git a/gcc/testsuite/c-c++-common/analyzer/realloc-1.c b/gcc/testsuite/c-c++-common/analyzer/realloc-1.c index 04925cf..0bb846c 100644 --- a/gcc/testsuite/c-c++-common/analyzer/realloc-1.c +++ b/gcc/testsuite/c-c++-common/analyzer/realloc-1.c @@ -92,5 +92,5 @@ void test_9 (void *p) void test_10 (char *s, int n) { __builtin_realloc(s, n); /* { dg-warning "ignoring return value of '__builtin_realloc' declared with attribute 'warn_unused_result'" "" { target c } } */ - /* { dg-warning "ignoring return value of 'void\\* __builtin_realloc\\(void\\*, (long )?unsigned int\\)' declared with attribute 'warn_unused_result'" "" { target c++ } .-1 } */ + /* { dg-warning "ignoring return value of 'void\\* __builtin_realloc\\(void\\*, (long )*unsigned int\\)' declared with attribute 'warn_unused_result'" "" { target c++ } .-1 } */ } /* { dg-warning "leak" } */ diff --git a/gcc/testsuite/c-c++-common/tsan/pr119801.c b/gcc/testsuite/c-c++-common/tsan/pr119801.c new file mode 100644 index 0000000..d3a6bb4 --- /dev/null +++ b/gcc/testsuite/c-c++-common/tsan/pr119801.c @@ -0,0 +1,24 @@ +/* PR sanitizer/119801 */ +/* { dg-do compile } */ +/* { dg-options "-fsanitize=thread" } */ + +[[gnu::noipa]] int +bar (int *p) +{ + return ++*p; +} + +int +foo (int *p) +{ + ++*p; + [[gnu::musttail]] return bar (p); +} + +[[gnu::noinline]] int +baz (int x) +{ + if (x < 10) + return x; + [[gnu::musttail]] return baz (x - 2); +} diff --git a/gcc/testsuite/cobol.dg/group2/ACCEPT_DATE___DAY_and_intrinsic_functions__2_.cob b/gcc/testsuite/cobol.dg/group2/ACCEPT_DATE___DAY_and_intrinsic_functions__2_.cob index 7a404fd..be58878 100644 --- a/gcc/testsuite/cobol.dg/group2/ACCEPT_DATE___DAY_and_intrinsic_functions__2_.cob +++ b/gcc/testsuite/cobol.dg/group2/ACCEPT_DATE___DAY_and_intrinsic_functions__2_.cob @@ -1,5 +1,5 @@ *> { dg-do run } - *> { dg-set-target-env-var COB_CURRENT_DATE "2020/06/12 18:45:22" } + *> { dg-set-target-env-var GCOBOL_CURRENT_DATE "2020/06/12 18:45:22" } IDENTIFICATION DIVISION. PROGRAM-ID. prog. diff --git a/gcc/testsuite/cobol.dg/group2/ACCEPT_FROM_TIME___DATE___DAY___DAY-OF-WEEK__2_.cob b/gcc/testsuite/cobol.dg/group2/ACCEPT_FROM_TIME___DATE___DAY___DAY-OF-WEEK__2_.cob index 6014220..665787d 100644 --- a/gcc/testsuite/cobol.dg/group2/ACCEPT_FROM_TIME___DATE___DAY___DAY-OF-WEEK__2_.cob +++ b/gcc/testsuite/cobol.dg/group2/ACCEPT_FROM_TIME___DATE___DAY___DAY-OF-WEEK__2_.cob @@ -1,5 +1,5 @@ *> { dg-do run } - *> { dg-set-target-env-var COB_CURRENT_DATE "2015/04/05 18:45:22" } + *> { dg-set-target-env-var GCOBOL_CURRENT_DATE "2015/04/05 18:45:22" } *> { dg-output-file "group2/ACCEPT_FROM_TIME___DATE___DAY___DAY-OF-WEEK__2_.out" } IDENTIFICATION DIVISION. diff --git a/gcc/testsuite/cobol.dg/group2/Dynamic_reference_modification.cob b/gcc/testsuite/cobol.dg/group2/Dynamic_reference_modification.cob new file mode 100644 index 0000000..99690da --- /dev/null +++ b/gcc/testsuite/cobol.dg/group2/Dynamic_reference_modification.cob @@ -0,0 +1,24 @@ + *> { dg-do run } + *> { dg-output-file "group2/Dynamic_reference_modification.out" } + + IDENTIFICATION DIVISION. + PROGRAM-ID. prog. + DATA DIVISION. + WORKING-STORAGE SECTION. + 01 X PIC X(4) VALUE "abcd". + 01 I PIC 9. + PROCEDURE DIVISION. + MOVE 1 TO I. + DISPLAY X(I:1) + END-DISPLAY. + MOVE 4 TO I. + DISPLAY X(I:1) + END-DISPLAY. + MOVE 1 TO I. + DISPLAY X(1:I) + END-DISPLAY. + MOVE 4 TO I. + DISPLAY X(1:I) + END-DISPLAY. + STOP RUN. + diff --git a/gcc/testsuite/cobol.dg/group2/Dynamic_reference_modification.out b/gcc/testsuite/cobol.dg/group2/Dynamic_reference_modification.out new file mode 100644 index 0000000..42a4b69 --- /dev/null +++ b/gcc/testsuite/cobol.dg/group2/Dynamic_reference_modification.out @@ -0,0 +1,5 @@ +a +d +a +abcd + diff --git a/gcc/testsuite/cobol.dg/group2/FUNCTION_DATE___TIME_OMNIBUS.cob b/gcc/testsuite/cobol.dg/group2/FUNCTION_DATE___TIME_OMNIBUS.cob index bb48bb0..88b1b84 100644 --- a/gcc/testsuite/cobol.dg/group2/FUNCTION_DATE___TIME_OMNIBUS.cob +++ b/gcc/testsuite/cobol.dg/group2/FUNCTION_DATE___TIME_OMNIBUS.cob @@ -1,4 +1,5 @@ *> { dg-do run } + *> { dg-set-target-env-var TZ UTC0 } identification division. program-id. test. @@ -79,7 +80,7 @@ 01 minus10 pic s99 value -10. - 01 forced_date_n pic X(64) VALUE Z"COB_CURRENT_DATE". + 01 forced_date_n pic X(64) VALUE Z"GCOBOL_CURRENT_DATE". 01 forced_date_v pic X(64) VALUE Z"1945/06/01 12:34:56". procedure division. diff --git a/gcc/testsuite/cobol.dg/group2/Length_overflow__1_.cob b/gcc/testsuite/cobol.dg/group2/Length_overflow__1_.cob new file mode 100644 index 0000000..6475356 --- /dev/null +++ b/gcc/testsuite/cobol.dg/group2/Length_overflow__1_.cob @@ -0,0 +1,16 @@ + *> { dg-do run } + *> { dg-xfail-run-if "" { *-*-* } } + *> { dg-output-file "group2/Length_overflow__1_.out" } + + IDENTIFICATION DIVISION. + PROGRAM-ID. prog. + DATA DIVISION. + WORKING-STORAGE SECTION. + 01 X PIC X(4) VALUE "abcd". + 01 I PIC 9 VALUE 5. + PROCEDURE DIVISION. + >>TURN EC-ALL CHECKING ON + DISPLAY X(1:I) NO ADVANCING + END-DISPLAY. + STOP RUN. + diff --git a/gcc/testsuite/cobol.dg/group2/Length_overflow__1_.out b/gcc/testsuite/cobol.dg/group2/Length_overflow__1_.out new file mode 100644 index 0000000..78981922 --- /dev/null +++ b/gcc/testsuite/cobol.dg/group2/Length_overflow__1_.out @@ -0,0 +1 @@ +a diff --git a/gcc/testsuite/cobol.dg/group2/Length_overflow__2_.cob b/gcc/testsuite/cobol.dg/group2/Length_overflow__2_.cob new file mode 100644 index 0000000..351c9df --- /dev/null +++ b/gcc/testsuite/cobol.dg/group2/Length_overflow__2_.cob @@ -0,0 +1,16 @@ + *> { dg-do run } + *> { dg-xfail-run-if "" { *-*-* } } + *> { dg-output-file "group2/Length_overflow__2_.out" } + + IDENTIFICATION DIVISION. + PROGRAM-ID. prog2. + DATA DIVISION. + WORKING-STORAGE SECTION. + 01 X PIC X(4) VALUE "abcd". + 01 I PIC 9 VALUE 5. + PROCEDURE DIVISION. + >>TURN EC-ALL CHECKING ON + DISPLAY X(3:I) NO ADVANCING + END-DISPLAY. + STOP RUN. + diff --git a/gcc/testsuite/cobol.dg/group2/Length_overflow__2_.out b/gcc/testsuite/cobol.dg/group2/Length_overflow__2_.out new file mode 100644 index 0000000..f2ad6c7 --- /dev/null +++ b/gcc/testsuite/cobol.dg/group2/Length_overflow__2_.out @@ -0,0 +1 @@ +c diff --git a/gcc/testsuite/cobol.dg/group2/Length_overflow_with_offset__1_.cob b/gcc/testsuite/cobol.dg/group2/Length_overflow_with_offset__1_.cob new file mode 100644 index 0000000..9f7fa83 --- /dev/null +++ b/gcc/testsuite/cobol.dg/group2/Length_overflow_with_offset__1_.cob @@ -0,0 +1,15 @@ + *> { dg-do run } + *> { dg-xfail-run-if "" { *-*-* } } + *> { dg-output-file "group2/Length_overflow_with_offset__1_.out" } + + IDENTIFICATION DIVISION. + PROGRAM-ID. prog. + DATA DIVISION. + WORKING-STORAGE SECTION. + 01 X PIC X(4) VALUE "abcd". + 01 I PIC 9 VALUE 3. + PROCEDURE DIVISION. + >>TURN EC-ALL CHECKING ON + DISPLAY X(3:I) NO ADVANCING. + STOP RUN. + diff --git a/gcc/testsuite/cobol.dg/group2/Length_overflow_with_offset__1_.out b/gcc/testsuite/cobol.dg/group2/Length_overflow_with_offset__1_.out new file mode 100644 index 0000000..f2ad6c7 --- /dev/null +++ b/gcc/testsuite/cobol.dg/group2/Length_overflow_with_offset__1_.out @@ -0,0 +1 @@ +c diff --git a/gcc/testsuite/cobol.dg/group2/Length_overflow_with_offset__2_.cob b/gcc/testsuite/cobol.dg/group2/Length_overflow_with_offset__2_.cob new file mode 100644 index 0000000..d077373 --- /dev/null +++ b/gcc/testsuite/cobol.dg/group2/Length_overflow_with_offset__2_.cob @@ -0,0 +1,16 @@ + *> { dg-do run } + *> { dg-xfail-run-if "" { *-*-* } } + *> { dg-output-file "group2/Length_overflow_with_offset__2_.out" } + + IDENTIFICATION DIVISION. + PROGRAM-ID. prog. + DATA DIVISION. + WORKING-STORAGE SECTION. + 01 X PIC X(4) VALUE "abcd". + 01 I PIC 9 VALUE 3. + PROCEDURE DIVISION. + >>TURN EC-ALL CHECKING ON + IF X(3:I) <> SPACES + DISPLAY X(3:I) NO ADVANCING. + STOP RUN. + diff --git a/gcc/testsuite/cobol.dg/group2/Length_overflow_with_offset__2_.out b/gcc/testsuite/cobol.dg/group2/Length_overflow_with_offset__2_.out new file mode 100644 index 0000000..8b13789 --- /dev/null +++ b/gcc/testsuite/cobol.dg/group2/Length_overflow_with_offset__2_.out @@ -0,0 +1 @@ + diff --git a/gcc/testsuite/cobol.dg/group2/Length_overflow_with_offset__3_.cob b/gcc/testsuite/cobol.dg/group2/Length_overflow_with_offset__3_.cob new file mode 100644 index 0000000..7fa9843 --- /dev/null +++ b/gcc/testsuite/cobol.dg/group2/Length_overflow_with_offset__3_.cob @@ -0,0 +1,22 @@ + *> { dg-do run } + *> { dg-xfail-run-if "" { *-*-* } } + *> { dg-output-file "group2/Length_overflow_with_offset__3_.out" } + + IDENTIFICATION DIVISION. + PROGRAM-ID. prog. + DATA DIVISION. + WORKING-STORAGE SECTION. + 01 X PIC X(4) VALUE "abcd". + 01 I PIC 9 VALUE 3. + PROCEDURE DIVISION. + >>TURN EC-ALL CHECKING ON + EVALUATE TRUE + WHEN I < 2 + AND X(3:I) <> SPACES + DISPLAY "1-" X(3:I) NO ADVANCING + WHEN I < 2 + WHEN X(3:I) <> SPACES + DISPLAY "2-" X(3:I) NO ADVANCING + END-EVALUATE + STOP RUN. + diff --git a/gcc/testsuite/cobol.dg/group2/Length_overflow_with_offset__3_.out b/gcc/testsuite/cobol.dg/group2/Length_overflow_with_offset__3_.out new file mode 100644 index 0000000..8b13789 --- /dev/null +++ b/gcc/testsuite/cobol.dg/group2/Length_overflow_with_offset__3_.out @@ -0,0 +1 @@ + diff --git a/gcc/testsuite/cobol.dg/group2/Offset_overflow.cob b/gcc/testsuite/cobol.dg/group2/Offset_overflow.cob new file mode 100644 index 0000000..8fd5421 --- /dev/null +++ b/gcc/testsuite/cobol.dg/group2/Offset_overflow.cob @@ -0,0 +1,16 @@ + *> { dg-do run } + *> { dg-xfail-run-if "" { *-*-* } } + *> { dg-output-file "group2/Offset_overflow.out" } + + IDENTIFICATION DIVISION. + PROGRAM-ID. prog. + DATA DIVISION. + WORKING-STORAGE SECTION. + 01. + 03 X PIC X(4) VALUE "abcd". + 03 I PIC 9 VALUE 5. + PROCEDURE DIVISION. + >>TURN EC-ALL CHECKING ON + DISPLAY X(I:1) NO ADVANCING. + STOP RUN. + diff --git a/gcc/testsuite/cobol.dg/group2/Offset_overflow.out b/gcc/testsuite/cobol.dg/group2/Offset_overflow.out new file mode 100644 index 0000000..7ed6ff8 --- /dev/null +++ b/gcc/testsuite/cobol.dg/group2/Offset_overflow.out @@ -0,0 +1 @@ +5 diff --git a/gcc/testsuite/cobol.dg/group2/Offset_underflow.cob b/gcc/testsuite/cobol.dg/group2/Offset_underflow.cob new file mode 100644 index 0000000..51100a8 --- /dev/null +++ b/gcc/testsuite/cobol.dg/group2/Offset_underflow.cob @@ -0,0 +1,16 @@ + *> { dg-do run } + *> { dg-xfail-run-if "" { *-*-* } } + *> { dg-output-file "group2/Offset_underflow.out" } + + IDENTIFICATION DIVISION. + PROGRAM-ID. prog. + DATA DIVISION. + WORKING-STORAGE SECTION. + 01 X PIC X(4) VALUE "abcd". + 01 I PIC 9 VALUE 0. + PROCEDURE DIVISION. + >>TURN EC-ALL CHECKING ON + DISPLAY X(I:1) NO ADVANCING + END-DISPLAY. + STOP RUN. + diff --git a/gcc/testsuite/cobol.dg/group2/Offset_underflow.out b/gcc/testsuite/cobol.dg/group2/Offset_underflow.out new file mode 100644 index 0000000..78981922 --- /dev/null +++ b/gcc/testsuite/cobol.dg/group2/Offset_underflow.out @@ -0,0 +1 @@ +a diff --git a/gcc/testsuite/cobol.dg/group2/Refmod__comparisons_inside_numeric-display.cob b/gcc/testsuite/cobol.dg/group2/Refmod__comparisons_inside_numeric-display.cob new file mode 100644 index 0000000..6fb70f4 --- /dev/null +++ b/gcc/testsuite/cobol.dg/group2/Refmod__comparisons_inside_numeric-display.cob @@ -0,0 +1,20 @@ + *> { dg-do run } + *> { dg-output-file "group2/Refmod__comparisons_inside_numeric-display.out" } + identification division. + program-id. prog. + data division. + working-storage section. + 01 n pic 9(9). + 01 i pic 99. + procedure division. + perform varying i from 1 by 1 until i > 8 + move 88888888 to n + move "12" to n(i:2) + display n + if n(i:2) not equal to "12" + display "Equality is flawed" + end-if + end-perform. + goback. + end program prog. + diff --git a/gcc/testsuite/cobol.dg/group2/Refmod__comparisons_inside_numeric-display.out b/gcc/testsuite/cobol.dg/group2/Refmod__comparisons_inside_numeric-display.out new file mode 100644 index 0000000..ac48dc8 --- /dev/null +++ b/gcc/testsuite/cobol.dg/group2/Refmod__comparisons_inside_numeric-display.out @@ -0,0 +1,9 @@ +128888888 +012888888 +081288888 +088128888 +088812888 +088881288 +088888128 +088888812 + diff --git a/gcc/testsuite/cobol.dg/group2/Refmod_sources_are_figurative_constants.cob b/gcc/testsuite/cobol.dg/group2/Refmod_sources_are_figurative_constants.cob new file mode 100644 index 0000000..c4af57d --- /dev/null +++ b/gcc/testsuite/cobol.dg/group2/Refmod_sources_are_figurative_constants.cob @@ -0,0 +1,29 @@ + *> { dg-do run } + *> { dg-output-file "group2/Refmod_sources_are_figurative_constants.out" } + + id division. + program-id. prog. + data division. + working-storage section. + 01 varx pic x(8) VALUE '""""""""'. + 01 varp redefines varx pointer. + procedure division. + move "12345678" to varx + display """" varx """" + move "999" to varx(4:3) + display """" varx """" + move LOW-VALUE to varx(4:3). + display """" varx """" + move ZERO to varx(4:3). + display """" varx """" + move SPACE to varx(4:3). + display """" varx """" + move QUOTE to varx(4:3). + display """" varx """" + move HIGH-VALUE to varx(4:3). + display varp + initialize varx all to value + display """" varx """" + . + end program prog. + diff --git a/gcc/testsuite/cobol.dg/group2/Refmod_sources_are_figurative_constants.out b/gcc/testsuite/cobol.dg/group2/Refmod_sources_are_figurative_constants.out new file mode 100644 index 0000000..2f5dadc --- /dev/null +++ b/gcc/testsuite/cobol.dg/group2/Refmod_sources_are_figurative_constants.out @@ -0,0 +1,9 @@ +"12345678" +"12399978" +"123" +"12300078" +"123 78" +"123"""78" +0x3837ffffff333231 +"""""""""" + diff --git a/gcc/testsuite/cobol.dg/group2/Static_reference_modification.cob b/gcc/testsuite/cobol.dg/group2/Static_reference_modification.cob new file mode 100644 index 0000000..919ddb3 --- /dev/null +++ b/gcc/testsuite/cobol.dg/group2/Static_reference_modification.cob @@ -0,0 +1,19 @@ + *> { dg-do run } + *> { dg-output-file "group2/Static_reference_modification.out" } + + IDENTIFICATION DIVISION. + PROGRAM-ID. prog. + DATA DIVISION. + WORKING-STORAGE SECTION. + 01 X PIC X(4) VALUE "abcd". + PROCEDURE DIVISION. + DISPLAY X(1:1) ":" X(1:2) ":" X(1:3) ":" X(1:4) ":" X(1:) + END-DISPLAY. + DISPLAY X(2:1) ":" X(2:2) ":" X(2:3) ":" X(2:) + END-DISPLAY. + DISPLAY X(3:1) ":" X(3:2) ":" X(3:) + END-DISPLAY. + DISPLAY X(4:1) ":" X(4:) + END-DISPLAY. + STOP RUN. + diff --git a/gcc/testsuite/cobol.dg/group2/Static_reference_modification.out b/gcc/testsuite/cobol.dg/group2/Static_reference_modification.out new file mode 100644 index 0000000..fe51165 --- /dev/null +++ b/gcc/testsuite/cobol.dg/group2/Static_reference_modification.out @@ -0,0 +1,5 @@ +a:ab:abc:abcd:abcd +b:bc:bcd:bcd +c:cd:cd +d:d + diff --git a/gcc/testsuite/g++.dg/abi/ref-temp1.C b/gcc/testsuite/g++.dg/abi/ref-temp1.C index 70c9a7a..b02dcf6 100644 --- a/gcc/testsuite/g++.dg/abi/ref-temp1.C +++ b/gcc/testsuite/g++.dg/abi/ref-temp1.C @@ -7,11 +7,16 @@ struct B { const A (&x)[2]; }; template <typename T> B &&b = { { { { 1, 2, 3 } }, { { 4, 5, 6 } } } }; B &temp = b<void>; -// { dg-final { scan-assembler ".weak\(_definition\)?\[ \t\]_?_ZGR1bIvE_" } } -// { dg-final { scan-assembler ".weak\(_definition\)?\[ \t\]_?_ZGR1bIvE0_" } } -// { dg-final { scan-assembler ".weak\(_definition\)?\[ \t\]_?_ZGR1bIvE1_" } } -// { dg-final { scan-assembler ".weak\(_definition\)?\[ \t\]_?_ZGR1bIvE2_" } } +// { dg-final { scan-assembler ".weak\(_definition\)?\[ \t\]_?_ZGR1bIvE_" { target { ! pe } } } } +// { dg-final { scan-assembler ".weak\(_definition\)?\[ \t\]_?_ZGR1bIvE0_" { target { ! pe } } } } +// { dg-final { scan-assembler ".weak\(_definition\)?\[ \t\]_?_ZGR1bIvE1_" { target { ! pe } } } } +// { dg-final { scan-assembler ".weak\(_definition\)?\[ \t\]_?_ZGR1bIvE2_" { target { ! pe } } } } +// { dg-final { scan-assembler "\.section\t\.data\\\$_ZGR1bIvE_,\"w\"\n\t\.linkonce same_size" { target pe } } } +// { dg-final { scan-assembler "\.section\t\.rdata\\\$_ZGR1bIvE0_,\"dr\"\n\t\.linkonce same_size" { target pe } } } +// { dg-final { scan-assembler "\.section\t\.rdata\\\$_ZGR1bIvE1_,\"dr\"\n\t\.linkonce same_size" { target pe } } } +// { dg-final { scan-assembler "\.section\t\.rdata\\\$_ZGR1bIvE2_,\"dr\"\n\t\.linkonce same_size" { target pe } } } +// // { dg-final { scan-assembler "_ZGR1bIvE_:\n\[^\n]+_ZGR1bIvE0_" } } // { dg-final { scan-assembler "_ZGR1bIvE0_:\n\[^\n]+_ZGR1bIvE1_" } } // { dg-final { scan-assembler "_ZGR1bIvE1_:\n\[^\n]+\[ \t\]1" } } diff --git a/gcc/testsuite/g++.dg/concepts/diagnostic20.C b/gcc/testsuite/g++.dg/concepts/diagnostic20.C new file mode 100644 index 0000000..2bb01db --- /dev/null +++ b/gcc/testsuite/g++.dg/concepts/diagnostic20.C @@ -0,0 +1,13 @@ +// PR c++/99214 +// { dg-do compile { target c++20 } } + +template <class T> +struct A { + template <class U> static void f() requires requires { T::fail; }; +}; + +int main() { + A<int>::f<char>(); // { dg-error "no match" } +} + +// { dg-message "In substitution of '\[^\r\n\]* \\\[with U = char\\\]'" "" { target *-*-* } 0 } diff --git a/gcc/testsuite/g++.dg/cpp0x/constexpr-ice21.C b/gcc/testsuite/g++.dg/cpp0x/constexpr-ice21.C index 4627365..dcc4044 100644 --- a/gcc/testsuite/g++.dg/cpp0x/constexpr-ice21.C +++ b/gcc/testsuite/g++.dg/cpp0x/constexpr-ice21.C @@ -3,7 +3,7 @@ struct NoMut1 { int a, b; }; struct NoMut3 : virtual NoMut1 { - constexpr NoMut3(int a, int b) // { dg-error "virtual base" "" { target c++23 } } + constexpr NoMut3(int a, int b) : NoMut1{a, b} {} // { dg-error "virtual base" } }; diff --git a/gcc/testsuite/g++.dg/cpp1y/constexpr-diag2.C b/gcc/testsuite/g++.dg/cpp1y/constexpr-diag2.C new file mode 100644 index 0000000..93f3f10 --- /dev/null +++ b/gcc/testsuite/g++.dg/cpp1y/constexpr-diag2.C @@ -0,0 +1,12 @@ +// PR c++/113360 +// { dg-do compile { target c++14 } } + +constexpr bool init_list() // { dg-bogus "because" } +{ + int total{}; + for (int x : {1, 2, 3}) // { dg-error "initializer list" } + total += x; + return total == 6; +} + +static_assert(init_list(), ""); // { dg-error "constant" } diff --git a/gcc/testsuite/g++.dg/cpp1y/constexpr-new.C b/gcc/testsuite/g++.dg/cpp1y/constexpr-new.C index d0ca0b7..f4c6d2e 100644 --- a/gcc/testsuite/g++.dg/cpp1y/constexpr-new.C +++ b/gcc/testsuite/g++.dg/cpp1y/constexpr-new.C @@ -6,7 +6,9 @@ constexpr int *f4(bool b) { return nullptr; } else { return new int{42}; // { dg-error "call to non-.constexpr." "" { target c++17_down } } - } // { dg-error "is not a constant expression because allocated storage has not been deallocated" "" { target c++2a } .-1 } + // { dg-message "allocated here" "" { target c++20 } .-1 } + } } static_assert(f4(true) == nullptr, ""); -static_assert(f4(false) == nullptr, ""); // { dg-error "non-.constant. condition|" } +static_assert(f4(false) == nullptr, ""); // { dg-error "non-constant condition" } +// { dg-error "is not a constant expression because allocated storage has not been deallocated" "" { target c++20 } .-1 } diff --git a/gcc/testsuite/g++.dg/cpp1y/constexpr-prvalue1.C b/gcc/testsuite/g++.dg/cpp1y/constexpr-prvalue1.C index ad31e30..6ad2ec8 100644 --- a/gcc/testsuite/g++.dg/cpp1y/constexpr-prvalue1.C +++ b/gcc/testsuite/g++.dg/cpp1y/constexpr-prvalue1.C @@ -1,6 +1,6 @@ // PR c++/116416 // { dg-do compile { target c++14 } } -// { dg-options "-O" } +// { dg-options "-O -fdump-tree-original" } struct Str { constexpr Str() {} @@ -17,14 +17,16 @@ extern void callback(Str str); void func1() { - callback(Str{"Test"}); + callback(Str{"Test1"}); } void func2() { - Str str{"Test"}; + Str str{"Test2"}; callback(str); } -// Check that we don't call Str::Str(char const*) -// { dg-final { scan-assembler-not "_ZN3StrC1EPKc" } } +// Check that the front end folds both the temporary initializer and +// that of 'str'. +// { dg-final { scan-tree-dump "{.str=\\(const char \\*\\) \"Test1\", .length=5}" "original" } } +// { dg-final { scan-tree-dump "{.str=\\(const char \\*\\) \"Test2\", .length=5}" "original" } } diff --git a/gcc/testsuite/g++.dg/cpp1y/constexpr-prvalue1a.C b/gcc/testsuite/g++.dg/cpp1y/constexpr-prvalue1a.C new file mode 100644 index 0000000..54176bf --- /dev/null +++ b/gcc/testsuite/g++.dg/cpp1y/constexpr-prvalue1a.C @@ -0,0 +1,33 @@ +// PR c++/116416 +// A version of constexpr-prvalue1.C that calls __builtin_is_constant_evaluated. +// { dg-do compile { target c++14 } } +// { dg-options "-O -fdump-tree-original" } + +struct Str { + constexpr Str() {} + constexpr Str(const char *instr) { + str = instr; length = 0; + for (auto index = 0; instr[index]; ++index) { + length += __builtin_is_constant_evaluated() ? 1 : 1; + } + } + const char *str = nullptr; + int length = 0; +}; +extern void callback(Str str); +void +func1() +{ + callback(Str{"Test1"}); +} +void +func2() +{ + Str str{"Test2"}; + callback(str); +} + +// Check that the front end folds both the temporary initializer and +// that of 'str'. +// { dg-final { scan-tree-dump "{.str=\\(const char \\*\\) \"Test1\", .length=5}" "original" } } +// { dg-final { scan-tree-dump "{.str=\\(const char \\*\\) \"Test2\", .length=5}" "original" } } diff --git a/gcc/testsuite/g++.dg/cpp1y/pr63996.C b/gcc/testsuite/g++.dg/cpp1y/pr63996.C index 8eee2e0..347c86c 100644 --- a/gcc/testsuite/g++.dg/cpp1y/pr63996.C +++ b/gcc/testsuite/g++.dg/cpp1y/pr63996.C @@ -1,5 +1,4 @@ // { dg-do compile { target c++14 } } -// { dg-additional-options "-Wno-return-type" } constexpr int foo (int i) @@ -8,4 +7,4 @@ foo (int i) if (i == 23) return 0; } -constexpr int j = foo (1); // { dg-error "flows off the end|in .constexpr. expansion of" } +constexpr int j = foo (1); diff --git a/gcc/testsuite/g++.dg/cpp1z/constexpr-asm-5.C b/gcc/testsuite/g++.dg/cpp1z/constexpr-asm-5.C index bcecea9..35beb27 100644 --- a/gcc/testsuite/g++.dg/cpp1z/constexpr-asm-5.C +++ b/gcc/testsuite/g++.dg/cpp1z/constexpr-asm-5.C @@ -28,7 +28,7 @@ struct M { constexpr K size () const { return {}; } constexpr L data () const { return {}; } }; #if __cpp_constexpr_dynamic_alloc >= 201907L struct N { constexpr int size () const { return 3; } - constexpr const char *data () const { return new char[3] { 'b', 'a', 'd' }; } }; // { dg-error "'\\\* N\\\(\\\).N::data\\\(\\\)' is not a constant expression because allocated storage has not been deallocated" "" { target c++20 } } + constexpr const char *data () const { return new char[3] { 'b', 'a', 'd' }; } }; #endif constexpr const char a[] = { 't', 'e', 's', 't' }; struct O { constexpr int size () const { return 4; } @@ -117,6 +117,7 @@ foo () asm ((M {})); #if __cpp_constexpr_dynamic_alloc >= 201907L asm ((N {})); // { dg-error "constexpr string 'data\\\(\\\)\\\[0\\\]' must be a constant expression" "" { target c++20 } } + // { dg-error "'\\\* N\\\(\\\).N::data\\\(\\\)' is not a constant expression because allocated storage has not been deallocated" "" { target c++20 } .-1 } #endif asm ((O {})); asm ((P (0))); @@ -190,6 +191,7 @@ bar () asm ((M {})); #if __cpp_constexpr_dynamic_alloc >= 201907L asm ((N {})); // { dg-error "constexpr string 'data\\\(\\\)\\\[0\\\]' must be a constant expression" "" { target c++20 } } + // { dg-error "'\\\* N\\\(\\\).N::data\\\(\\\)' is not a constant expression because allocated storage has not been deallocated" "" { target c++20 } .-1 } #endif asm ((O {})); asm ((P (0))); diff --git a/gcc/testsuite/g++.dg/cpp23/class-deduction-inherited8.C b/gcc/testsuite/g++.dg/cpp23/class-deduction-inherited8.C new file mode 100644 index 0000000..4494c70 --- /dev/null +++ b/gcc/testsuite/g++.dg/cpp23/class-deduction-inherited8.C @@ -0,0 +1,21 @@ +// PR c++/119687 +// { dg-do compile { target c++17 } } + +template <typename> class QFlagsStorage{}; + +template <typename Enum> struct QFlagsStorageHelper : QFlagsStorage<Enum> { + using QFlagsStorage<Enum>::QFlagsStorage; + +public: + QFlagsStorageHelper(Enum); +}; + +template <typename Enum> struct QFlags : public QFlagsStorageHelper<Enum> { + using Base = QFlagsStorageHelper<Enum>; + using Base::Base; + QFlags(Enum); +}; + +void f(int flag) { + QFlags{int{}}; +} diff --git a/gcc/testsuite/g++.dg/cpp23/constexpr-nonlit18.C b/gcc/testsuite/g++.dg/cpp23/constexpr-nonlit18.C index 8e230ef..f891814 100644 --- a/gcc/testsuite/g++.dg/cpp23/constexpr-nonlit18.C +++ b/gcc/testsuite/g++.dg/cpp23/constexpr-nonlit18.C @@ -24,7 +24,7 @@ f3 () } constexpr int -f4 () // { dg-message "declared here" "" { target c++20_down } } +f4 () { // { dg-message "is not usable as a 'constexpr' function because:" "" { target c++23 } .-1 } static const int a = f1 (1); // { dg-error "'a' defined 'static' in 'constexpr' function only available with" "" { target c++20_down } } return 0; // { dg-error "'a' defined 'static' in 'constexpr' context" "" { target c++23 } .-1 } diff --git a/gcc/testsuite/g++.dg/cpp26/pack-indexing2.C b/gcc/testsuite/g++.dg/cpp26/pack-indexing2.C index fdc8320..4a7e494 100644 --- a/gcc/testsuite/g++.dg/cpp26/pack-indexing2.C +++ b/gcc/testsuite/g++.dg/cpp26/pack-indexing2.C @@ -49,7 +49,7 @@ template<int N> int getT2 (auto... Ts) { - return Ts...[N]; // { dg-error "pack index is negative" } + return Ts...[N]; // { dg-error "pack index '-1' is negative" } } template<auto N, typename... Ts> @@ -63,7 +63,7 @@ template<auto N, typename... Ts> void badtype2 () { - Ts...[N] t; // { dg-error "pack index is out of range" } + Ts...[N] t; // { dg-error "pack index '1' is out of range for pack of length '1'" } } template<auto N, typename... Ts> @@ -77,7 +77,7 @@ template<auto N, typename... Ts> void badtype4 () { - Ts...[N] t; // { dg-error "pack index is negative" } + Ts...[N] t; // { dg-error "pack index '-1' is negative" } } int nonconst () { return 42; } diff --git a/gcc/testsuite/g++.dg/cpp26/static_assert1.C b/gcc/testsuite/g++.dg/cpp26/static_assert1.C index f9ac831..1d0e6f2 100644 --- a/gcc/testsuite/g++.dg/cpp26/static_assert1.C +++ b/gcc/testsuite/g++.dg/cpp26/static_assert1.C @@ -69,10 +69,11 @@ static_assert (false, M {}); // { dg-warning "'static_assert' with non-string me // { dg-error "static assertion failed: test" "" { target *-*-* } .-1 } #if __cpp_constexpr_dynamic_alloc >= 201907L struct N { constexpr int size () const { return 3; } - constexpr const char *data () const { return new char[3] { 'b', 'a', 'd' }; } }; // { dg-error "'\\\* N\\\(\\\).N::data\\\(\\\)' is not a constant expression because allocated storage has not been deallocated" "" { target c++20 } } + constexpr const char *data () const { return new char[3] { 'b', 'a', 'd' }; } }; static_assert (true, N {}); // { dg-warning "'static_assert' with non-string message only available with" "" { target { c++20 && c++23_down } } } static_assert (false, N {}); // { dg-warning "'static_assert' with non-string message only available with" "" { target { c++20 && c++23_down } } } // { dg-error "constexpr string 'data\\\(\\\)\\\[0\\\]' must be a constant expression" "" { target c++20 } .-1 } + // { dg-error "'\\\* N\\\(\\\).N::data\\\(\\\)' is not a constant expression because allocated storage has not been deallocated" "" { target c++20 } .-2 } #endif constexpr const char a[] = { 't', 'e', 's', 't' }; struct O { constexpr int size () const { return 4; } diff --git a/gcc/testsuite/g++.dg/cpp2a/concepts-lambda23.C b/gcc/testsuite/g++.dg/cpp2a/concepts-lambda23.C new file mode 100644 index 0000000..f442120 --- /dev/null +++ b/gcc/testsuite/g++.dg/cpp2a/concepts-lambda23.C @@ -0,0 +1,12 @@ +// PR c++/119175 +// { dg-do compile { target c++20 } } + +template<int = 0> +static void from() requires requires { + []<int> requires requires { [] {}; } {}; +} +{} + +int main() { + from(); +} diff --git a/gcc/testsuite/g++.dg/cpp2a/constexpr-dtor16.C b/gcc/testsuite/g++.dg/cpp2a/constexpr-dtor16.C index b84aaf9..99d1307 100644 --- a/gcc/testsuite/g++.dg/cpp2a/constexpr-dtor16.C +++ b/gcc/testsuite/g++.dg/cpp2a/constexpr-dtor16.C @@ -3,5 +3,5 @@ struct A { virtual ~A (); }; struct B : virtual A { constexpr ~B () {} }; -// { dg-error "'struct B' has virtual base classes" "" { target c++20 } .-1 } +// { dg-error "'constexpr' destructor in 'struct B' that has virtual base classes" "" { target c++20 } .-1 } // { dg-error "'constexpr' destructors only available with" "" { target c++17_down } .-2 } diff --git a/gcc/testsuite/g++.dg/cpp2a/constexpr-dtor7.C b/gcc/testsuite/g++.dg/cpp2a/constexpr-dtor7.C index 463eaca..f4546c1 100644 --- a/gcc/testsuite/g++.dg/cpp2a/constexpr-dtor7.C +++ b/gcc/testsuite/g++.dg/cpp2a/constexpr-dtor7.C @@ -3,7 +3,7 @@ struct S { int *s; - constexpr S () : s(new int) {} // { dg-error "is not a constant expression because allocated storage has not been deallocated" } + constexpr S () : s(new int) {} S (const S &) = delete; S &operator= (const S &) = delete; constexpr ~S () { delete s; } @@ -17,3 +17,4 @@ foo (S v) } static_assert (foo (S ())); // { dg-error "non-constant condition for static assertion" } +// { dg-error "is not a constant expression because allocated storage has not been deallocated" "" { target *-*-* } .-1 } diff --git a/gcc/testsuite/g++.dg/cpp2a/constexpr-dynamic10.C b/gcc/testsuite/g++.dg/cpp2a/constexpr-dynamic10.C index f9f8223..e543ce4 100644 --- a/gcc/testsuite/g++.dg/cpp2a/constexpr-dynamic10.C +++ b/gcc/testsuite/g++.dg/cpp2a/constexpr-dynamic10.C @@ -5,7 +5,7 @@ struct C { virtual void a(); }; struct B { virtual void b(); }; -struct A : virtual B, C { virtual void c(); }; // { dg-error ".struct A. has virtual base classes" } +struct A : virtual B, C { virtual void c(); }; // { dg-error "virtual base classes" } constexpr A a; // { dg-error "call" } diff --git a/gcc/testsuite/g++.dg/cpp2a/constexpr-new24.C b/gcc/testsuite/g++.dg/cpp2a/constexpr-new24.C index ee62f18..17c9f54 100644 --- a/gcc/testsuite/g++.dg/cpp2a/constexpr-new24.C +++ b/gcc/testsuite/g++.dg/cpp2a/constexpr-new24.C @@ -6,14 +6,14 @@ int a; constexpr char * f1 () { - constexpr auto p = new char[(long int) &a]; // { dg-error "size not constant" } + constexpr auto p = new char[(long int) &a]; // { dg-error "conversion from pointer" } return p; } constexpr char * f2 () { - auto p = new char[(long int) &a]; // { dg-error "size not constant" } + auto p = new char[(long int) &a]; // { dg-error "conversion from pointer" } return p; } diff --git a/gcc/testsuite/g++.dg/cpp2a/constexpr-new26.C b/gcc/testsuite/g++.dg/cpp2a/constexpr-new26.C index c82bd43..d8e53b2 100644 --- a/gcc/testsuite/g++.dg/cpp2a/constexpr-new26.C +++ b/gcc/testsuite/g++.dg/cpp2a/constexpr-new26.C @@ -4,7 +4,7 @@ constexpr int * f7 () { - int *p = new int (2); // { dg-error "is not a constant expression because it refers to a result of" } + int *p = new int (2); // { dg-message "allocated here" } delete p; return p; } @@ -12,6 +12,5 @@ f7 () void g () { - constexpr auto v7 = f7 (); + constexpr auto v7 = f7 (); // { dg-error "is not a constant expression because it refers to a result of" } } - diff --git a/gcc/testsuite/g++.dg/cpp2a/constexpr-new3.C b/gcc/testsuite/g++.dg/cpp2a/constexpr-new3.C index 5d9f1925..30e453e 100644 --- a/gcc/testsuite/g++.dg/cpp2a/constexpr-new3.C +++ b/gcc/testsuite/g++.dg/cpp2a/constexpr-new3.C @@ -5,19 +5,19 @@ constexpr int * f1 () { - return new int (2); // { dg-error "is not a constant expression because it refers to a result of" } + return new int (2); // { dg-message "allocated here" } } -constexpr auto v1 = f1 (); +constexpr auto v1 = f1 (); // { dg-error "is not a constant expression because it refers to a result of" } constexpr bool f2 () { - int *p = new int (3); // { dg-error "is not a constant expression because allocated storage has not been deallocated" } + int *p = new int (3); // { dg-message "allocated here" } return false; } -constexpr auto v2 = f2 (); +constexpr auto v2 = f2 (); // { dg-error "is not a constant expression because allocated storage has not been deallocated" } constexpr bool f3 () @@ -64,12 +64,12 @@ constexpr auto v6 = f6 (); // { dg-message "in 'constexpr' expansion of" } constexpr int * f7 () { - int *p = new int (2); // { dg-error "is not a constant expression because it refers to a result of" } + int *p = new int (2); // { dg-message "allocated here" } delete p; return p; } -constexpr auto v7 = f7 (); +constexpr auto v7 = f7 (); // { dg-error "is not a constant expression because it refers to a result of" } constexpr bool f8_impl (int *p) diff --git a/gcc/testsuite/g++.dg/cpp2a/constexpr-vector1.C b/gcc/testsuite/g++.dg/cpp2a/constexpr-vector1.C new file mode 100644 index 0000000..196c6ec --- /dev/null +++ b/gcc/testsuite/g++.dg/cpp2a/constexpr-vector1.C @@ -0,0 +1,8 @@ +// PR c++/113835 +// { dg-timeout-factor 0.05 } +// { dg-do compile { target c++20_only } } + +#include <vector> +const std::size_t N = 1'000'000; +std::vector<int> x(N); +int main() {} diff --git a/gcc/testsuite/g++.dg/cpp2a/constinit14.C b/gcc/testsuite/g++.dg/cpp2a/constinit14.C index 06c4cb4..26d82fe67 100644 --- a/gcc/testsuite/g++.dg/cpp2a/constinit14.C +++ b/gcc/testsuite/g++.dg/cpp2a/constinit14.C @@ -2,12 +2,13 @@ // { dg-do compile { target c++20 } } struct Value { - Value() : v{new int{42}} {} // { dg-error "result of 'operator new'" "" { target implicit_constexpr } } + Value() : v{new int{42}} {} int* v; }; struct S { static constinit inline Value v{}; // { dg-error "variable .S::v. does not have a constant initializer|call to non-.constexpr. function" } + // { dg-error "result of 'operator new'" "" { target implicit_constexpr } .-1 } }; int main() { return *S::v.v; } diff --git a/gcc/testsuite/g++.dg/cpp2a/lambda-targ14.C b/gcc/testsuite/g++.dg/cpp2a/lambda-targ14.C new file mode 100644 index 0000000..debb15e --- /dev/null +++ b/gcc/testsuite/g++.dg/cpp2a/lambda-targ14.C @@ -0,0 +1,12 @@ +// PR c++/119345 +// { dg-do compile { target c++20 } } + +void f(auto... args) { + [args...]<int... i> { + (..., [args...] { i; }); + }.template operator()<0>(); +} + +int main() { + f(); +} diff --git a/gcc/testsuite/g++.dg/eh/pr119507.C b/gcc/testsuite/g++.dg/eh/pr119507.C new file mode 100644 index 0000000..50afa75 --- /dev/null +++ b/gcc/testsuite/g++.dg/eh/pr119507.C @@ -0,0 +1,17 @@ +// { dg-do compile { target comdat_group } } +// Force off function sections +// Force on exceptions +// { dg-options "-fno-function-sections -fexceptions" } +// PR middle-end/119507 + + +inline int comdat() { try { throw 1; } catch (int) { return 1; } return 0; } +int another_func_with_exception() { try { throw 1; } catch (int) { return 1; } return 0; } +inline int comdat1() { try { throw 1; } catch (int) { return 1; } return 0; } +int foo() { return comdat() + comdat1(); } + +// Make sure the gcc puts the exception table for both comdat and comdat1 in their own section +// { dg-final { scan-assembler-times ".section\[\t \]\[^\n\]*.gcc_except_table._Z6comdatv" 1 } } +// { dg-final { scan-assembler-times ".section\[\t \]\[^\n\]*.gcc_except_table._Z7comdat1v" 1 } } +// There should be 3 exception tables, +// { dg-final { scan-assembler-times ".section\[\t \]\[^\n\]*.gcc_except_table" 3 } } diff --git a/gcc/testsuite/g++.dg/ext/pragma-target2.C b/gcc/testsuite/g++.dg/ext/pragma-target2.C new file mode 100644 index 0000000..53eb7dd --- /dev/null +++ b/gcc/testsuite/g++.dg/ext/pragma-target2.C @@ -0,0 +1,18 @@ +// PR c++/114772 +// { dg-do compile { target x86_64-*-* } } + +template<typename V, bool STREAMING> +inline __attribute__((always_inline)) +__attribute__((warn_unused_result)) +int walk_document(V visitor) {return 0;} + +template<bool STREAMING> +void parse_document() { + int r = walk_document<bool, STREAMING>(false); +} + +void stage2_next() { + parse_document<true>(); +} + +#pragma GCC target("pclmul") diff --git a/gcc/testsuite/g++.dg/ext/type_pack_element2.C b/gcc/testsuite/g++.dg/ext/type_pack_element2.C index 1bf7753..1b07673 100644 --- a/gcc/testsuite/g++.dg/ext/type_pack_element2.C +++ b/gcc/testsuite/g++.dg/ext/type_pack_element2.C @@ -2,7 +2,7 @@ int p; -using type = __type_pack_element<&p, int>; // { dg-error "not an integral constant" } +using type = __type_pack_element<&p, int>; // { dg-error "non-integral type" } using type = __type_pack_element<1, int>; // { dg-error "out of range" } using type = __type_pack_element<2, int, char>; // { dg-error "out of range" } using type = __type_pack_element<-1, int>; // { dg-error "negative" } diff --git a/gcc/testsuite/g++.dg/ext/type_pack_element4.C b/gcc/testsuite/g++.dg/ext/type_pack_element4.C index aa508c7..5a39194 100644 --- a/gcc/testsuite/g++.dg/ext/type_pack_element4.C +++ b/gcc/testsuite/g++.dg/ext/type_pack_element4.C @@ -3,7 +3,7 @@ template <typename... _Elements> class tuple{}; template <unsigned long __i, typename... _Elements> -__type_pack_element<__i, _Elements...> &get(tuple<_Elements...> &__t) noexcept; // { dg-error "index is out of range" } +__type_pack_element<__i, _Elements...> &get(tuple<_Elements...> &__t) noexcept; // { dg-error "out of range" } tuple<int,int> data; template <unsigned long Level> unsigned take_impl(unsigned idx) { diff --git a/gcc/testsuite/g++.dg/lto/pr119614_0.C b/gcc/testsuite/g++.dg/lto/pr119614_0.C new file mode 100644 index 0000000..09c07fb --- /dev/null +++ b/gcc/testsuite/g++.dg/lto/pr119614_0.C @@ -0,0 +1,34 @@ +// PR tree-optimization/119614 +// { dg-lto-do link } +// { dg-lto-options { { -O2 -fPIC -flto -flto-partition=max } } } +// { dg-require-effective-target shared } +// { dg-require-effective-target fpic } +// { dg-require-effective-target musttail } +// { dg-extra-ld-options "-shared" } + +struct S {} b; +char *foo (); +int e, g; +void bar (); +void corge (S); + +[[gnu::noinline]] static char * +baz () +{ + bar (); + return 0; +} + +const char * +qux () +{ + if (e) + { + S a = b; + corge (a); + if (g) + return 0; + [[gnu::musttail]] return baz (); + } + return foo (); +} diff --git a/gcc/testsuite/g++.dg/modules/lambda-10_a.H b/gcc/testsuite/g++.dg/modules/lambda-10_a.H new file mode 100644 index 0000000..1ad1a80 --- /dev/null +++ b/gcc/testsuite/g++.dg/modules/lambda-10_a.H @@ -0,0 +1,17 @@ +// PR c++/119755 +// { dg-additional-options "-fmodule-header" } +// { dg-module-cmi {} } + +template <typename _Out> void format(_Out) { + constexpr int __term = 1; + [&] { __term; }; + [&] { const int outer = __term; { __term; } }; + [&]() noexcept { __term; }; + [&]() noexcept { const int outer = __term; { __term; } }; + [&](auto) { int n[__term]; }(0); + [&](auto) noexcept { int n[__term]; }(0); +} + +inline void vformat() { + format(0); +} diff --git a/gcc/testsuite/g++.dg/modules/lambda-10_b.C b/gcc/testsuite/g++.dg/modules/lambda-10_b.C new file mode 100644 index 0000000..3556bce --- /dev/null +++ b/gcc/testsuite/g++.dg/modules/lambda-10_b.C @@ -0,0 +1,7 @@ +// PR c++/119755 +// { dg-additional-options "-fmodules" } + +import "lambda-10_a.H"; +int main() { + vformat(); +} diff --git a/gcc/testsuite/g++.dg/modules/lambda-8_b.C b/gcc/testsuite/g++.dg/modules/lambda-8_b.C index 7ace494..96578ba 100644 --- a/gcc/testsuite/g++.dg/modules/lambda-8_b.C +++ b/gcc/testsuite/g++.dg/modules/lambda-8_b.C @@ -4,4 +4,4 @@ #include "lambda-8.h" import "lambda-8_a.H"; -// { dg-error "conflicting global module declaration" "" { target *-*-* } 0 } +// { dg-error "conflicting imported declaration" "" { target *-*-* } 0 } diff --git a/gcc/testsuite/g++.dg/modules/leg-merge-4_c.C b/gcc/testsuite/g++.dg/modules/leg-merge-4_c.C index f1b1aeb..5756057 100644 --- a/gcc/testsuite/g++.dg/modules/leg-merge-4_c.C +++ b/gcc/testsuite/g++.dg/modules/leg-merge-4_c.C @@ -11,8 +11,8 @@ void foo () X *p; } -// { dg-regexp "\nIn module \[^\n]*leg-merge-4_b.H, imported at \[^\n]*leg-merge-4_c.C:\[0-9]*:\n\[^\n]*leg-merge-4_b.H:4:\[0-9]*: error: conflicting global module declaration 'float bob'\nIn module \[^\n]*leg-merge-4_a.H, imported at \[^\n]*leg-merge-4_c.C:\[0-9]*:\n\[^\n]*leg-merge-4_a.H:4:\[0-9]*: note: existing declaration 'int bob'\n\[^\n]*leg-merge-4_c.C:9:\[0-9]*: note: during load of binding '::bob'$" } +// { dg-regexp "\nIn module \[^\n]*leg-merge-4_b.H, imported at \[^\n]*leg-merge-4_c.C:\[0-9]*:\n\[^\n]*leg-merge-4_b.H:4:\[0-9]*: error: conflicting type for imported declaration 'float bob'\nIn module \[^\n]*leg-merge-4_a.H, imported at \[^\n]*leg-merge-4_c.C:\[0-9]*:\n\[^\n]*leg-merge-4_a.H:4:\[0-9]*: note: existing declaration 'int bob'\n\[^\n]*leg-merge-4_c.C:9:\[0-9]*: note: during load of binding '::bob'$" } -// { dg-regexp "\nIn module \[^\n]*leg-merge-4_b.H, imported at \[^\n]*leg-merge-4_c.C:\[0-9]*:\n\[^\n]*leg-merge-4_b.H:5:\[0-9]*: error: conflicting global module declaration 'int frob\\(\\)'\nIn module \[^\n]*leg-merge-4_a.H, imported at \[^\n]*leg-merge-4_c.C:\[0-9]*:\n\[^\n]*leg-merge-4_a.H:5:\[0-9]*: note: existing declaration 'void frob\\(\\)'\n\[^\n]*leg-merge-4_c.C:10:\[0-9]*: note: during load of binding '::frob'$" } +// { dg-regexp "\nIn module \[^\n]*leg-merge-4_b.H, imported at \[^\n]*leg-merge-4_c.C:\[0-9]*:\n\[^\n]*leg-merge-4_b.H:5:\[0-9]*: error: conflicting type for imported declaration 'int frob\\(\\)'\nIn module \[^\n]*leg-merge-4_a.H, imported at \[^\n]*leg-merge-4_c.C:\[0-9]*:\n\[^\n]*leg-merge-4_a.H:5:\[0-9]*: note: existing declaration 'void frob\\(\\)'\n\[^\n]*leg-merge-4_c.C:10:\[0-9]*: note: during load of binding '::frob'$" } -// { dg-regexp "In module \[^\n]*leg-merge-4_b.H, imported at \[^\n]*leg-merge-4_c.C:\[0-9]*:\n\[^\n]*leg-merge-4_b.H:6:\[0-9]*: error: conflicting global module declaration 'union X'\nIn module \[^\n]*leg-merge-4_a.H, imported at \[^\n]*leg-merge-4_c.C:\[0-9]*:\n\[^\n]*leg-merge-4_a.H:6:\[0-9]*: note: existing declaration 'class X'\n\[^\n]*leg-merge-4_c.C:11:\[0-9]*: note: during load of binding '::X'$" } +// { dg-regexp "In module \[^\n]*leg-merge-4_b.H, imported at \[^\n]*leg-merge-4_c.C:\[0-9]*:\n\[^\n]*leg-merge-4_b.H:6:\[0-9]*: error: conflicting type for imported declaration 'union X'\nIn module \[^\n]*leg-merge-4_a.H, imported at \[^\n]*leg-merge-4_c.C:\[0-9]*:\n\[^\n]*leg-merge-4_a.H:6:\[0-9]*: note: existing declaration 'class X'\n\[^\n]*leg-merge-4_c.C:11:\[0-9]*: note: during load of binding '::X'$" } diff --git a/gcc/testsuite/g++.dg/modules/noexcept-4_a.H b/gcc/testsuite/g++.dg/modules/noexcept-4_a.H new file mode 100644 index 0000000..b888a1b --- /dev/null +++ b/gcc/testsuite/g++.dg/modules/noexcept-4_a.H @@ -0,0 +1,6 @@ +// { dg-additional-options "-fmodule-header -std=c++20" } +// { dg-module-cmi {} } + +struct exception_ptr { + friend bool operator==(const exception_ptr&, const exception_ptr&) = default; +}; diff --git a/gcc/testsuite/g++.dg/modules/noexcept-4_b.C b/gcc/testsuite/g++.dg/modules/noexcept-4_b.C new file mode 100644 index 0000000..7cc5531 --- /dev/null +++ b/gcc/testsuite/g++.dg/modules/noexcept-4_b.C @@ -0,0 +1,18 @@ +// { dg-additional-options "-fmodules -std=c++20" } + +struct exception_ptr { + friend bool operator==(const exception_ptr&, const exception_ptr&) = default; +}; + +void enqueue() { + exception_ptr e; + e == e; +} + +import "noexcept-4_a.H"; + +int main() { + constexpr exception_ptr e; + static_assert(e == e); + static_assert(noexcept(e == e)); +} diff --git a/gcc/testsuite/g++.dg/opt/is_constant_evaluated4.C b/gcc/testsuite/g++.dg/opt/is_constant_evaluated4.C new file mode 100644 index 0000000..9650004 --- /dev/null +++ b/gcc/testsuite/g++.dg/opt/is_constant_evaluated4.C @@ -0,0 +1,20 @@ +// PR c++/114970 +// { dg-do compile { target c++17 } } +// { dg-additional-options "-O -Wunused-value" } + +struct sv +{ + const char* str; + unsigned len; + + constexpr sv(const char *p): str(p), len(0) + { + if (__builtin_is_constant_evaluated ()) { len = 42; } + } +}; + +int main() +{ + sv s ("foo"); + return s.len; +} diff --git a/gcc/testsuite/g++.dg/opt/shrink-wrapping-vector-1.C b/gcc/testsuite/g++.dg/opt/shrink-wrapping-vector-1.C new file mode 100644 index 0000000..8b1ad53 --- /dev/null +++ b/gcc/testsuite/g++.dg/opt/shrink-wrapping-vector-1.C @@ -0,0 +1,17 @@ +// { dg-do compile { target { { { i?86-*-* x86_64-*-* } && { ! ia32 } } || { powerpc*-*-* aarch64*-*-* riscv*-*-* } } } } +// { dg-options "-O2 -fdump-rtl-pro_and_epilogue" } +// { dg-skip-if "requires hosted libstdc++ for vector" { ! hostedlib } } + +// PR rtl-optimization/118502 + +// The shrink-wrapping should happen around the slow path of vector<int>::push_back, +// The fast path is just checking if there is enough space and doing a few stores. +// We want to verify that shrink wrapping always happens. + +#include <vector> + +void push_back(std::vector<int>& xs, unsigned char x) { + xs.push_back(x); +} + +/* { dg-final { scan-rtl-dump "Performing shrink-wrapping" "pro_and_epilogue" } } */ diff --git a/gcc/testsuite/g++.dg/pr112822.C b/gcc/testsuite/g++.dg/pr112822.C index a855752..f88bd83 100644 --- a/gcc/testsuite/g++.dg/pr112822.C +++ b/gcc/testsuite/g++.dg/pr112822.C @@ -89,7 +89,7 @@ template <typename aj, typename cm> struct cg<aj, cm> { typedef aj cn; }; namespace ai { template <typename cj, int> cj cp; template <typename bu, typename cj, int> void cl(bu *cr, cj cs) { ct(cr, cs); } -typedef __attribute__((altivec(vector__))) double co; +typedef double co __attribute__ ((vector_size (16))); void ct(double *cr, co cs) { *(co *)cr = cs; } struct cq { co q; diff --git a/gcc/testsuite/g++.dg/template/explicit-args6.C b/gcc/testsuite/g++.dg/template/explicit-args6.C index 18663d7b..0d9718c 100644 --- a/gcc/testsuite/g++.dg/template/explicit-args6.C +++ b/gcc/testsuite/g++.dg/template/explicit-args6.C @@ -24,10 +24,12 @@ frob() // narrowing check, reject negative values return unsigned{N}; // { dg-prune-output "narrowing" } -} // { dg-prune-output "flows off the end" } -// { dg-prune-output "not a return-statement" } +} -template<int N> void get_n(tuple& t) { get<frob<N>()>(t); } // { dg-error "" } +// This complains about calling frob only in C++11 because +// maybe_save_constexpr_fundef fails; in later standards it succeeds, +// and the evaluation failure is silent due to the earlier errors. +template<int N> void get_n(tuple& t) { get<frob<N>()>(t); } // { dg-error "" "" { target c++11_only } } int main() { diff --git a/gcc/testsuite/g++.dg/template/friend86.C b/gcc/testsuite/g++.dg/template/friend86.C new file mode 100644 index 0000000..9e2c1af --- /dev/null +++ b/gcc/testsuite/g++.dg/template/friend86.C @@ -0,0 +1,25 @@ +// PR c++/119807 +// { dg-do run } + +template<int N> +struct A { + template<class T> friend int f(A<N>, T); +}; + +template struct A<0>; +template struct A<1>; + +int main() { + A<0> x; + A<1> y; + if (f(x, true) != 0) __builtin_abort(); + if (f(y, true) != 1) __builtin_abort(); +} + +template<int N> +struct B { + template<class T> friend int f(A<N>, T) { return N; } +}; + +template struct B<0>; +template struct B<1>; diff --git a/gcc/testsuite/g++.dg/template/friend87.C b/gcc/testsuite/g++.dg/template/friend87.C new file mode 100644 index 0000000..94c0dfc --- /dev/null +++ b/gcc/testsuite/g++.dg/template/friend87.C @@ -0,0 +1,42 @@ +// PR c++/119807 +// { dg-do compile { target c++20 } } + +using size_t = decltype(sizeof(0)); + +template<auto tag, size_t current> +struct CounterReader { + template<typename> + friend auto counterFlag(CounterReader<tag, current>) noexcept; +}; + +template<auto tag, size_t current> +struct CounterWriter { + static constexpr size_t value = current; + + template<typename> + friend auto counterFlag(CounterReader<tag, current>) noexcept {} +}; + +template<auto tag, auto unique, size_t current = 0, size_t mask = size_t(1) << (sizeof(size_t) * 8 - 1)> +[[nodiscard]] constexpr size_t counterAdvance() noexcept { + if constexpr (!mask) { + return CounterWriter<tag, current + 1>::value; + } else if constexpr (requires { counterFlag<void>(CounterReader<tag, current | mask>()); }) { + return counterAdvance<tag, unique, current | mask, (mask >> 1)>(); + } + else { + return counterAdvance<tag, unique, current, (mask >> 1)>(); + } +} + +constexpr auto defaultCounterTag = [] {}; + +template<auto tag = defaultCounterTag, auto unique = [] {}> +constexpr size_t counter() noexcept { + return counterAdvance<tag, unique>(); +} + +int main() { + static_assert(counter() == 1); + static_assert(counter() == 2); +} diff --git a/gcc/testsuite/g++.dg/torture/pr119778.C b/gcc/testsuite/g++.dg/torture/pr119778.C new file mode 100644 index 0000000..4948056 --- /dev/null +++ b/gcc/testsuite/g++.dg/torture/pr119778.C @@ -0,0 +1,20 @@ +// { dg-do compile } +// { dg-additional-options "-Wall" } + +struct jmp_buf { long l[16]; }; +extern "C" int setjmp (jmp_buf *); +struct S { + void foo () { bar (); } + virtual char bar () { return 0; } +}; +void baz (); +jmp_buf *a; + +void +qux (bool x, S *y) +{ + if (x) + setjmp (a); + y->foo (); + baz (); +} diff --git a/gcc/testsuite/g++.dg/warn/Wformat-3.C b/gcc/testsuite/g++.dg/warn/Wformat-3.C new file mode 100644 index 0000000..e308530 --- /dev/null +++ b/gcc/testsuite/g++.dg/warn/Wformat-3.C @@ -0,0 +1,19 @@ +// PR c++/116954 +// { dg-additional-options -Wformat } + +#ifndef WORKS +template<int N> +int fn(char (&buf)[N], const char fmt[], ...) + __attribute__ ((__format__ (__printf__, 2, 3))); +#endif + +template<int N> +__attribute__ ((__format__ (__printf__, 2, 3))) +int fn(char (&)[N], const char [], ...) +{ return 0; } + +int main() +{ + char buf[20]; + return fn(buf, "%s", 42); /* { dg-warning "Wformat" } */ +} diff --git a/gcc/testsuite/g++.target/aarch64/sve/pr119706.C b/gcc/testsuite/g++.target/aarch64/sve/pr119706.C new file mode 100644 index 0000000..40fefe5 --- /dev/null +++ b/gcc/testsuite/g++.target/aarch64/sve/pr119706.C @@ -0,0 +1,178 @@ +/* { dg-do compile } */ +/* { dg-options "-O3 -mcpu=neoverse-v2 --param=aarch64-autovec-preference=sve-only -w" } */ + +namespace a { +typedef long unsigned b; +typedef int c; +template <bool, typename d> struct e { using f = d; }; +template <bool g, typename d = void> using h = typename e<g, d>::f; +template <typename aa, typename, template <typename> class> struct i { + using f = aa; +}; +template <typename aa, template <typename> class j> using k = i<aa, void, j>; +template <typename aa, template <typename> class j> +using l = typename k<aa, j>::f; +} // namespace a +inline void *operator new(a::b, void *ab) { return ab; } +namespace a { +template <typename> class ac { +public: + typedef b m; + template <typename ad, typename... n> void ae(ad *ab, n... w) { + new (ab) ad(w...); + } +}; +template <typename d> using x = ac<d>; +template <typename d> class af : public x<d> { +public: + typedef d o; + template <typename> struct ag { typedef af ah; }; +}; +struct ai {}; +struct aj : ai {}; +struct ak : aj {}; +template <typename> struct al; +template <typename d> struct al<d *> { + typedef ak an; + typedef c ao; + typedef d ap; +}; +template <typename aq> typename aq ::an ar(aq) { return typename aq ::an(); } +template <typename as> typename as ::ao at(as au, as av, ak) { return av - au; } +template <typename aw> typename aw ::ao ax(aw au, aw av) { + return at(au, av, ar(au)); +} +template <typename> struct ay { typedef c ao; }; +} // namespace a +namespace az { +template <typename am, typename> class ba { + am bb; + typedef a::al<am> bc; + +public: + typedef typename bc::an an; + typedef typename bc::ao ao; + typedef typename bc::ap ap; + ba(am bd) : bb(bd) {} + ap operator*() { return *bb; } + ba operator++() { + ++bb; + return *this; + } + am base() { return bb; } +}; +template <typename be, typename bf, typename bg> +bool operator!=(ba<be, bg> bh, ba<bf, bg> p) { + return bh.base() != p.base(); +} +template <typename be, typename bf, typename bg> +auto operator-(ba<be, bg> bh, ba<bf, bg> p) { + return bh.base() - p.base(); +} +} // namespace az +namespace a { +struct bi { + template <typename d, typename> struct bj { + using f = typename d::ag<d>::ah; + }; + template <typename> using bk = b; + template <typename...> static constexpr bool bl = false; + template <typename, typename> static constexpr bool bm = bl<>; + template <typename d, typename... n> static constexpr bool bn = bm<d, n...>; +}; +template <typename bo, typename ad> using bp = typename bi::bj<bo, ad>::f; +template <typename bo> struct bq : bi { + typedef typename bo::o o; + using br = l<o *, bk>; + template <typename, typename bs> struct bt { using f = typename ay<bs>::ao; }; + template <typename bu, typename> struct bv { using f = typename bu::m; }; + using ao = typename bt<bo, c>::f; + using m = typename bv<bo, ao>::f; + template <typename d> using bw = bp<bo, d>; + static br allocate(bo, m); + template <typename d, typename... n> + static h<bn<bo, d>> ae(bo ci, d ab, n... w) { + ci.ae(ab, w...); + } +}; +template <typename d> struct bx { + static bool by(d &bz) try { d(bz.begin(), bz.ca(), bz.cb()); } catch (...) { + } +}; +} // namespace a +namespace az { +template <typename bo> struct cc : a::bq<bo> { + typedef a::bq<bo> q; + template <typename d> struct ag { typedef typename q::bw<d> ah; }; +}; +} // namespace az +enum cd {}; +using ce = double; +namespace a { +template <typename aw, typename cf, typename cg, typename ch> +cg cj(aw au, cf av, cg ck, ch cl) { + typedef az::cc<ch> cx; + for (; au != av; ++au, ++ck) + cx::ae(cl, ck, *au); +} +template <typename d, typename bo> struct cm { + typedef typename az::cc<bo>::ag<d>::ah cn; + typedef typename az::cc<cn>::br br; + struct co { + br db; + br cp; + }; + struct cq : cn, co { + cq(cn) {} + } typedef cr; + cn cs(); + cr cb() noexcept; + cm(cr ci) : ct(ci) {} + cq ct; + br cu(b cv) { + typedef az::cc<cn> cw; + return cv ? cw::allocate(ct, cv) : c(); + } +}; +template <typename d, typename bo = af<d>> class cy : cm<d, bo> { + typedef cm<d, bo> cz; + +public: + typedef typename cz::br br; + typedef az::ba<br, cy> da; + typedef b m; + typedef bo cr; + cz::cs; + template <typename aw> cy(aw au, aw av, cr ci) : cz(ci) { + dg(au, av, ar(au)); + } + cz::cb; + da begin() { return this->ct.db; } + da ca() { return this->ct.cp; } + void r() { s(); } + void clear() { t(this->ct.db); } + template <typename cg> void dg(cg au, cg av, ai) { y(au, av, ax(au, av)); } + template <typename am, typename cf> void y(am au, cf av, m cv) { + br z = this->cu(dc(cv, cs())); + cj(au, av, z, cs()); + } + bool s(); + m dc(m cv, cr) { return cv; } + void t(br dd) { + if (this->ct.cp - dd) + this->ct.cp = dd; + } +}; +template <typename d, typename bo> bool cy<d, bo>::s() { bx<cy>::by(*this); } +namespace basic { +class u { + using de = ce; + void v(cd, b); + cy<de> df; +}; +void u::v(cd, b) { + df.clear(); + df.r(); +} +} // namespace basic +} // namespace a
\ No newline at end of file diff --git a/gcc/testsuite/g++.target/gcn/exceptions-bad_cast-1.C b/gcc/testsuite/g++.target/gcn/exceptions-bad_cast-1.C new file mode 100644 index 0000000..f3e3099 --- /dev/null +++ b/gcc/testsuite/g++.target/gcn/exceptions-bad_cast-1.C @@ -0,0 +1,15 @@ +/* 'std::bad_cast' exception. */ + +/* { dg-do run } */ +/* Via the magic string "-std=*++" indicate that testing one (the default) C++ standard is sufficient. */ +/* { dg-additional-options -fexceptions } */ +/* { dg-additional-options -fdump-tree-optimized-raw } */ + +#include "../../../../libgomp/testsuite/libgomp.oacc-c++/exceptions-bad_cast-1.C" + +/* { dg-output {CheCKpOInT[\r\n]+} } + + { dg-final { scan-tree-dump-times {gimple_call <__cxa_bad_cast, } 1 optimized } } + We don't print anything, but just 'abort'. + + { dg-shouldfail {'std::bad_cast' exception} } */ diff --git a/gcc/testsuite/g++.target/gcn/exceptions-bad_cast-2.C b/gcc/testsuite/g++.target/gcn/exceptions-bad_cast-2.C new file mode 100644 index 0000000..b047cbed --- /dev/null +++ b/gcc/testsuite/g++.target/gcn/exceptions-bad_cast-2.C @@ -0,0 +1,13 @@ +/* 'std::bad_cast' exception, caught. */ + +/* Via the magic string "-std=*++" indicate that testing one (the default) C++ standard is sufficient. */ +/* { dg-additional-options -fexceptions } */ +/* { dg-additional-options -mno-fake-exceptions } */ +/* { dg-additional-options -fdump-tree-optimized-raw } */ + +#include "../../../../libgomp/testsuite/libgomp.oacc-c++/exceptions-bad_cast-2.C" + +/* { dg-final { scan-tree-dump-times {gimple_call <__cxa_bad_cast, } 1 optimized } } + Compilation fails: + { dg-regexp {[^\r\n]+: In function 'int main\(\)':[\r\n]+(?:[^\r\n]+: sorry, unimplemented: exception handling not supported[\r\n]+)+} } + (Note, using 'dg-regexp' instead of 'dg-message', as the former runs before the auto-mark-UNSUPPORTED.) */ diff --git a/gcc/testsuite/g++.target/gcn/exceptions-bad_cast-2_-mfake-exceptions.C b/gcc/testsuite/g++.target/gcn/exceptions-bad_cast-2_-mfake-exceptions.C new file mode 100644 index 0000000..2904188 --- /dev/null +++ b/gcc/testsuite/g++.target/gcn/exceptions-bad_cast-2_-mfake-exceptions.C @@ -0,0 +1,18 @@ +/* 'std::bad_cast' exception, caught, '-mfake-exceptions'. */ + +/* { dg-do run } */ +/* Via the magic string "-std=*++" indicate that testing one (the default) C++ standard is sufficient. */ +/* { dg-additional-options -fexceptions } */ +/* { dg-additional-options -mfake-exceptions } + { dg-bogus {sorry, unimplemented: exception handling not supported} {} { target *-*-* } 0 } */ +/* { dg-additional-options -fdump-tree-optimized-raw } */ + +#include "exceptions-bad_cast-2.C" + +/* { dg-output {CheCKpOInT[\r\n]+} } + + { dg-final { scan-tree-dump-times {gimple_call <__cxa_bad_cast, } 1 optimized } } + We don't print anything, but just 'abort'. + + There is no 'catch'ing; any exception is fatal. + { dg-shouldfail {'std::bad_cast' exception} } */ diff --git a/gcc/testsuite/g++.target/gcn/exceptions-bad_cast-3.C b/gcc/testsuite/g++.target/gcn/exceptions-bad_cast-3.C new file mode 100644 index 0000000..3d0118c --- /dev/null +++ b/gcc/testsuite/g++.target/gcn/exceptions-bad_cast-3.C @@ -0,0 +1,10 @@ +/* 'std::bad_cast' exception, dead code. */ + +/* { dg-do run } */ +/* Via the magic string "-std=*++" indicate that testing one (the default) C++ standard is sufficient. */ +/* { dg-additional-options -fexceptions } */ +/* { dg-additional-options -fdump-tree-optimized-raw } */ + +#include "../../../../libgomp/testsuite/libgomp.oacc-c++/exceptions-bad_cast-3.C" + +/* { dg-final { scan-tree-dump-times {gimple_call <__cxa_bad_cast, } 1 optimized } } */ diff --git a/gcc/testsuite/g++.target/gcn/exceptions-pr118794-1.C b/gcc/testsuite/g++.target/gcn/exceptions-pr118794-1.C new file mode 100644 index 0000000..20f9d49 --- /dev/null +++ b/gcc/testsuite/g++.target/gcn/exceptions-pr118794-1.C @@ -0,0 +1,17 @@ +/* Exception handling constructs in dead code. */ + +/* Via the magic string "-std=*++" indicate that testing one (the default) C++ standard is sufficient. */ +/* { dg-additional-options -fexceptions } */ +/* { dg-additional-options -mno-fake-exceptions } */ +/* { dg-additional-options -O0 } */ +/* { dg-additional-options -fdump-tree-optimized-raw } */ + +#include "../../../../libgomp/testsuite/libgomp.c++/target-exceptions-pr118794-1.C" + +/* In this specific C++ arrangement, distilled from PR118794, GCC synthesizes + '__builtin_eh_pointer', '__builtin_unwind_resume' calls as dead code in 'f': + { dg-final { scan-tree-dump-times {gimple_call <__builtin_eh_pointer, } 1 optimized } } + { dg-final { scan-tree-dump-times {gimple_call <__builtin_unwind_resume, } 1 optimized } } + Given '-O0', compilation fails: + { dg-regexp {[^\r\n]+: In function 'void f\(\)':[\r\n]+(?:[^\r\n]+: sorry, unimplemented: exception handling not supported[\r\n]+)+} } + (Note, using 'dg-regexp' instead of 'dg-message', as the former runs before the auto-mark-UNSUPPORTED.) */ diff --git a/gcc/testsuite/g++.target/gcn/exceptions-pr118794-1_-mfake-exceptions.C b/gcc/testsuite/g++.target/gcn/exceptions-pr118794-1_-mfake-exceptions.C new file mode 100644 index 0000000..a5f0da2 --- /dev/null +++ b/gcc/testsuite/g++.target/gcn/exceptions-pr118794-1_-mfake-exceptions.C @@ -0,0 +1,16 @@ +/* Exception handling constructs in dead code, '-mfake-exceptions'. */ + +/* { dg-do run } */ +/* Via the magic string "-std=*++" indicate that testing one (the default) C++ standard is sufficient. */ +/* { dg-additional-options -fexceptions } */ +/* { dg-additional-options -mfake-exceptions } + { dg-bogus {sorry, unimplemented: exception handling not supported} {} { target *-*-* } 0 } */ +/* { dg-additional-options -O0 } */ +/* { dg-additional-options -fdump-tree-optimized-raw } */ + +#include "exceptions-pr118794-1.C" + +/* In this specific C++ arrangement, distilled from PR118794, GCC synthesizes + '__builtin_eh_pointer', '__builtin_unwind_resume' calls as dead code in 'f': + { dg-final { scan-tree-dump-times {gimple_call <__builtin_eh_pointer, } 1 optimized } } + { dg-final { scan-tree-dump-times {gimple_call <__builtin_unwind_resume, } 1 optimized } } */ diff --git a/gcc/testsuite/g++.target/gcn/exceptions-throw-1.C b/gcc/testsuite/g++.target/gcn/exceptions-throw-1.C new file mode 100644 index 0000000..6cadf58 --- /dev/null +++ b/gcc/testsuite/g++.target/gcn/exceptions-throw-1.C @@ -0,0 +1,16 @@ +/* 'throw'. */ + +/* { dg-do run } */ +/* Via the magic string "-std=*++" indicate that testing one (the default) C++ standard is sufficient. */ +/* { dg-additional-options -fexceptions } */ +/* { dg-additional-options -fdump-tree-optimized-raw } */ + +#include "../../../../libgomp/testsuite/libgomp.oacc-c++/exceptions-throw-1.C" + +/* { dg-output {CheCKpOInT[\r\n]+} } + + { dg-final { scan-tree-dump-times {gimple_call <__cxa_allocate_exception, } 1 optimized } } + { dg-final { scan-tree-dump-times {gimple_call <__cxa_throw, } 1 optimized } } + We don't print anything, but just 'abort'. + + { dg-shouldfail {'MyException' exception} } */ diff --git a/gcc/testsuite/g++.target/gcn/exceptions-throw-2.C b/gcc/testsuite/g++.target/gcn/exceptions-throw-2.C new file mode 100644 index 0000000..671c810 --- /dev/null +++ b/gcc/testsuite/g++.target/gcn/exceptions-throw-2.C @@ -0,0 +1,14 @@ +/* 'throw', caught. */ + +/* Via the magic string "-std=*++" indicate that testing one (the default) C++ standard is sufficient. */ +/* { dg-additional-options -fexceptions } */ +/* { dg-additional-options -mno-fake-exceptions } */ +/* { dg-additional-options -fdump-tree-optimized-raw } */ + +#include "../../../../libgomp/testsuite/libgomp.oacc-c++/exceptions-throw-2.C" + +/* { dg-final { scan-tree-dump-times {gimple_call <__cxa_allocate_exception, } 1 optimized } } + { dg-final { scan-tree-dump-times {gimple_call <__cxa_throw, } 1 optimized } } + Compilation fails: + { dg-regexp {[^\r\n]+: In function 'int main\(\)':[\r\n]+(?:[^\r\n]+: sorry, unimplemented: exception handling not supported[\r\n]+)+} } + (Note, using 'dg-regexp' instead of 'dg-message', as the former runs before the auto-mark-UNSUPPORTED.) */ diff --git a/gcc/testsuite/g++.target/gcn/exceptions-throw-2_-mfake-exceptions.C b/gcc/testsuite/g++.target/gcn/exceptions-throw-2_-mfake-exceptions.C new file mode 100644 index 0000000..f1fd505 --- /dev/null +++ b/gcc/testsuite/g++.target/gcn/exceptions-throw-2_-mfake-exceptions.C @@ -0,0 +1,19 @@ +/* 'throw', caught, '-mfake-exceptions'. */ + +/* { dg-do run } */ +/* Via the magic string "-std=*++" indicate that testing one (the default) C++ standard is sufficient. */ +/* { dg-additional-options -fexceptions } */ +/* { dg-additional-options -mfake-exceptions } + { dg-bogus {sorry, unimplemented: exception handling not supported} {} { target *-*-* } 0 } */ +/* { dg-additional-options -fdump-tree-optimized-raw } */ + +#include "exceptions-throw-2.C" + +/* { dg-output {CheCKpOInT[\r\n]+} } + + { dg-final { scan-tree-dump-times {gimple_call <__cxa_allocate_exception, } 1 optimized } } + { dg-final { scan-tree-dump-times {gimple_call <__cxa_throw, } 1 optimized } } + We don't print anything, but just 'abort'. + + There is no 'catch'ing; any exception is fatal. + { dg-shouldfail {'MyException' exception} } */ diff --git a/gcc/testsuite/g++.target/gcn/exceptions-throw-3.C b/gcc/testsuite/g++.target/gcn/exceptions-throw-3.C new file mode 100644 index 0000000..5c1ad7a --- /dev/null +++ b/gcc/testsuite/g++.target/gcn/exceptions-throw-3.C @@ -0,0 +1,11 @@ +/* 'throw', dead code. */ + +/* { dg-do run } */ +/* Via the magic string "-std=*++" indicate that testing one (the default) C++ standard is sufficient. */ +/* { dg-additional-options -fexceptions } */ +/* { dg-additional-options -fdump-tree-optimized-raw } */ + +#include "../../../../libgomp/testsuite/libgomp.oacc-c++/exceptions-throw-3.C" + +/* { dg-final { scan-tree-dump-times {gimple_call <__cxa_allocate_exception, } 1 optimized } } + { dg-final { scan-tree-dump-times {gimple_call <__cxa_throw, } 1 optimized } } */ diff --git a/gcc/testsuite/g++.target/gcn/gcn.exp b/gcc/testsuite/g++.target/gcn/gcn.exp new file mode 100644 index 0000000..a3bd75f --- /dev/null +++ b/gcc/testsuite/g++.target/gcn/gcn.exp @@ -0,0 +1,56 @@ +# Specific regression driver for GCN. +# Copyright (C) 2000-2025 Free Software Foundation, Inc. + +# This program is free software; you can redistribute it and/or modify +# it under the terms of the GNU General Public License as published by +# the Free Software Foundation; either version 3 of the License, or +# (at your option) any later version. +# +# This program is distributed in the hope that it will be useful, +# but WITHOUT ANY WARRANTY; without even the implied warranty of +# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the +# GNU General Public License for more details. +# +# You should have received a copy of the GNU General Public License +# along with GCC; see the file COPYING3. If not see +# <http://www.gnu.org/licenses/>. + +# G++ testsuite that uses the `dg.exp' driver. + +# Exit immediately if this isn't a GCN target. +if ![istarget amdgcn*-*-*] then { + return +} + +# Load support procs. +load_lib g++-dg.exp + +# If a testcase doesn't have special options, use these. +global DEFAULT_CXXFLAGS +if ![info exists DEFAULT_CXXFLAGS] then { + set DEFAULT_CXXFLAGS " -pedantic-errors -Wno-long-long" +} + +# Initialize `dg'. +dg-init + +# Recursively find files in $dir and subdirs, do not walk into subdirs +# that contain their own .exp file. +proc find-cxx-tests { dir suffix } { + set tests [lsort [glob -nocomplain -directory $dir "*.$suffix" ]] + foreach subdir [lsort [glob -nocomplain -type d -directory $dir *]] { + if { [glob -nocomplain -directory $subdir *.exp] eq "" } { + eval lappend tests [find-cxx-tests $subdir $suffix] + } + } + return $tests +} + +set tests [find-cxx-tests $srcdir/$subdir {C}] + +# Main loop. +g++-dg-runtest $tests "" $DEFAULT_CXXFLAGS + + +# All done. +dg-finish diff --git a/gcc/testsuite/g++.target/gcn/pr119692-1-1.C b/gcc/testsuite/g++.target/gcn/pr119692-1-1.C new file mode 100644 index 0000000..b44b08d --- /dev/null +++ b/gcc/testsuite/g++.target/gcn/pr119692-1-1.C @@ -0,0 +1,6 @@ +/* PR119692 "C++ 'typeinfo', 'vtable' vs. OpenACC, OpenMP 'target' offloading" */ + +/* { dg-do run } */ +/* Via the magic string "-std=*++" indicate that testing one (the default) C++ standard is sufficient. */ + +#include "../../../../libgomp/testsuite/libgomp.oacc-c++/pr119692-1-1.C" diff --git a/gcc/testsuite/g++.target/nvptx/exceptions-bad_cast-1.C b/gcc/testsuite/g++.target/nvptx/exceptions-bad_cast-1.C new file mode 100644 index 0000000..f3e3099 --- /dev/null +++ b/gcc/testsuite/g++.target/nvptx/exceptions-bad_cast-1.C @@ -0,0 +1,15 @@ +/* 'std::bad_cast' exception. */ + +/* { dg-do run } */ +/* Via the magic string "-std=*++" indicate that testing one (the default) C++ standard is sufficient. */ +/* { dg-additional-options -fexceptions } */ +/* { dg-additional-options -fdump-tree-optimized-raw } */ + +#include "../../../../libgomp/testsuite/libgomp.oacc-c++/exceptions-bad_cast-1.C" + +/* { dg-output {CheCKpOInT[\r\n]+} } + + { dg-final { scan-tree-dump-times {gimple_call <__cxa_bad_cast, } 1 optimized } } + We don't print anything, but just 'abort'. + + { dg-shouldfail {'std::bad_cast' exception} } */ diff --git a/gcc/testsuite/g++.target/nvptx/exceptions-bad_cast-2.C b/gcc/testsuite/g++.target/nvptx/exceptions-bad_cast-2.C new file mode 100644 index 0000000..b047cbed --- /dev/null +++ b/gcc/testsuite/g++.target/nvptx/exceptions-bad_cast-2.C @@ -0,0 +1,13 @@ +/* 'std::bad_cast' exception, caught. */ + +/* Via the magic string "-std=*++" indicate that testing one (the default) C++ standard is sufficient. */ +/* { dg-additional-options -fexceptions } */ +/* { dg-additional-options -mno-fake-exceptions } */ +/* { dg-additional-options -fdump-tree-optimized-raw } */ + +#include "../../../../libgomp/testsuite/libgomp.oacc-c++/exceptions-bad_cast-2.C" + +/* { dg-final { scan-tree-dump-times {gimple_call <__cxa_bad_cast, } 1 optimized } } + Compilation fails: + { dg-regexp {[^\r\n]+: In function 'int main\(\)':[\r\n]+(?:[^\r\n]+: sorry, unimplemented: exception handling not supported[\r\n]+)+} } + (Note, using 'dg-regexp' instead of 'dg-message', as the former runs before the auto-mark-UNSUPPORTED.) */ diff --git a/gcc/testsuite/g++.target/nvptx/exceptions-bad_cast-2_-mfake-exceptions.C b/gcc/testsuite/g++.target/nvptx/exceptions-bad_cast-2_-mfake-exceptions.C new file mode 100644 index 0000000..3f40951 --- /dev/null +++ b/gcc/testsuite/g++.target/nvptx/exceptions-bad_cast-2_-mfake-exceptions.C @@ -0,0 +1,19 @@ +/* 'std::bad_cast' exception, caught, '-mfake-exceptions'. */ + +/* { dg-do run } */ +/* Via the magic string "-std=*++" indicate that testing one (the default) C++ standard is sufficient. */ +/* { dg-additional-options -fexceptions } */ +/* { dg-additional-options -mfake-exceptions } + { dg-bogus {sorry, unimplemented: exception handling not supported} {} { target *-*-* } 0 } */ +/* { dg-additional-options -fdump-tree-optimized-raw } */ +/* { dg-bogus {_ZTISt8bad_cast} PR119734 { xfail *-*-* } 0 } */ + +#include "exceptions-bad_cast-2.C" + +/* { dg-output {CheCKpOInT[\r\n]+} } + + { dg-final { scan-tree-dump-times {gimple_call <__cxa_bad_cast, } 1 optimized } } + We don't print anything, but just 'abort'. + + There is no 'catch'ing; any exception is fatal. + { dg-shouldfail {'std::bad_cast' exception} } */ diff --git a/gcc/testsuite/g++.target/nvptx/exceptions-bad_cast-3.C b/gcc/testsuite/g++.target/nvptx/exceptions-bad_cast-3.C new file mode 100644 index 0000000..3d0118c --- /dev/null +++ b/gcc/testsuite/g++.target/nvptx/exceptions-bad_cast-3.C @@ -0,0 +1,10 @@ +/* 'std::bad_cast' exception, dead code. */ + +/* { dg-do run } */ +/* Via the magic string "-std=*++" indicate that testing one (the default) C++ standard is sufficient. */ +/* { dg-additional-options -fexceptions } */ +/* { dg-additional-options -fdump-tree-optimized-raw } */ + +#include "../../../../libgomp/testsuite/libgomp.oacc-c++/exceptions-bad_cast-3.C" + +/* { dg-final { scan-tree-dump-times {gimple_call <__cxa_bad_cast, } 1 optimized } } */ diff --git a/gcc/testsuite/g++.target/nvptx/exceptions-pr118794-1.C b/gcc/testsuite/g++.target/nvptx/exceptions-pr118794-1.C new file mode 100644 index 0000000..20f9d49 --- /dev/null +++ b/gcc/testsuite/g++.target/nvptx/exceptions-pr118794-1.C @@ -0,0 +1,17 @@ +/* Exception handling constructs in dead code. */ + +/* Via the magic string "-std=*++" indicate that testing one (the default) C++ standard is sufficient. */ +/* { dg-additional-options -fexceptions } */ +/* { dg-additional-options -mno-fake-exceptions } */ +/* { dg-additional-options -O0 } */ +/* { dg-additional-options -fdump-tree-optimized-raw } */ + +#include "../../../../libgomp/testsuite/libgomp.c++/target-exceptions-pr118794-1.C" + +/* In this specific C++ arrangement, distilled from PR118794, GCC synthesizes + '__builtin_eh_pointer', '__builtin_unwind_resume' calls as dead code in 'f': + { dg-final { scan-tree-dump-times {gimple_call <__builtin_eh_pointer, } 1 optimized } } + { dg-final { scan-tree-dump-times {gimple_call <__builtin_unwind_resume, } 1 optimized } } + Given '-O0', compilation fails: + { dg-regexp {[^\r\n]+: In function 'void f\(\)':[\r\n]+(?:[^\r\n]+: sorry, unimplemented: exception handling not supported[\r\n]+)+} } + (Note, using 'dg-regexp' instead of 'dg-message', as the former runs before the auto-mark-UNSUPPORTED.) */ diff --git a/gcc/testsuite/g++.target/nvptx/exceptions-pr118794-1_-mfake-exceptions.C b/gcc/testsuite/g++.target/nvptx/exceptions-pr118794-1_-mfake-exceptions.C new file mode 100644 index 0000000..a5f0da2 --- /dev/null +++ b/gcc/testsuite/g++.target/nvptx/exceptions-pr118794-1_-mfake-exceptions.C @@ -0,0 +1,16 @@ +/* Exception handling constructs in dead code, '-mfake-exceptions'. */ + +/* { dg-do run } */ +/* Via the magic string "-std=*++" indicate that testing one (the default) C++ standard is sufficient. */ +/* { dg-additional-options -fexceptions } */ +/* { dg-additional-options -mfake-exceptions } + { dg-bogus {sorry, unimplemented: exception handling not supported} {} { target *-*-* } 0 } */ +/* { dg-additional-options -O0 } */ +/* { dg-additional-options -fdump-tree-optimized-raw } */ + +#include "exceptions-pr118794-1.C" + +/* In this specific C++ arrangement, distilled from PR118794, GCC synthesizes + '__builtin_eh_pointer', '__builtin_unwind_resume' calls as dead code in 'f': + { dg-final { scan-tree-dump-times {gimple_call <__builtin_eh_pointer, } 1 optimized } } + { dg-final { scan-tree-dump-times {gimple_call <__builtin_unwind_resume, } 1 optimized } } */ diff --git a/gcc/testsuite/g++.target/nvptx/exceptions-throw-1.C b/gcc/testsuite/g++.target/nvptx/exceptions-throw-1.C new file mode 100644 index 0000000..6cadf58 --- /dev/null +++ b/gcc/testsuite/g++.target/nvptx/exceptions-throw-1.C @@ -0,0 +1,16 @@ +/* 'throw'. */ + +/* { dg-do run } */ +/* Via the magic string "-std=*++" indicate that testing one (the default) C++ standard is sufficient. */ +/* { dg-additional-options -fexceptions } */ +/* { dg-additional-options -fdump-tree-optimized-raw } */ + +#include "../../../../libgomp/testsuite/libgomp.oacc-c++/exceptions-throw-1.C" + +/* { dg-output {CheCKpOInT[\r\n]+} } + + { dg-final { scan-tree-dump-times {gimple_call <__cxa_allocate_exception, } 1 optimized } } + { dg-final { scan-tree-dump-times {gimple_call <__cxa_throw, } 1 optimized } } + We don't print anything, but just 'abort'. + + { dg-shouldfail {'MyException' exception} } */ diff --git a/gcc/testsuite/g++.target/nvptx/exceptions-throw-2.C b/gcc/testsuite/g++.target/nvptx/exceptions-throw-2.C new file mode 100644 index 0000000..671c810 --- /dev/null +++ b/gcc/testsuite/g++.target/nvptx/exceptions-throw-2.C @@ -0,0 +1,14 @@ +/* 'throw', caught. */ + +/* Via the magic string "-std=*++" indicate that testing one (the default) C++ standard is sufficient. */ +/* { dg-additional-options -fexceptions } */ +/* { dg-additional-options -mno-fake-exceptions } */ +/* { dg-additional-options -fdump-tree-optimized-raw } */ + +#include "../../../../libgomp/testsuite/libgomp.oacc-c++/exceptions-throw-2.C" + +/* { dg-final { scan-tree-dump-times {gimple_call <__cxa_allocate_exception, } 1 optimized } } + { dg-final { scan-tree-dump-times {gimple_call <__cxa_throw, } 1 optimized } } + Compilation fails: + { dg-regexp {[^\r\n]+: In function 'int main\(\)':[\r\n]+(?:[^\r\n]+: sorry, unimplemented: exception handling not supported[\r\n]+)+} } + (Note, using 'dg-regexp' instead of 'dg-message', as the former runs before the auto-mark-UNSUPPORTED.) */ diff --git a/gcc/testsuite/g++.target/nvptx/exceptions-throw-2_-mfake-exceptions.C b/gcc/testsuite/g++.target/nvptx/exceptions-throw-2_-mfake-exceptions.C new file mode 100644 index 0000000..f1fd505 --- /dev/null +++ b/gcc/testsuite/g++.target/nvptx/exceptions-throw-2_-mfake-exceptions.C @@ -0,0 +1,19 @@ +/* 'throw', caught, '-mfake-exceptions'. */ + +/* { dg-do run } */ +/* Via the magic string "-std=*++" indicate that testing one (the default) C++ standard is sufficient. */ +/* { dg-additional-options -fexceptions } */ +/* { dg-additional-options -mfake-exceptions } + { dg-bogus {sorry, unimplemented: exception handling not supported} {} { target *-*-* } 0 } */ +/* { dg-additional-options -fdump-tree-optimized-raw } */ + +#include "exceptions-throw-2.C" + +/* { dg-output {CheCKpOInT[\r\n]+} } + + { dg-final { scan-tree-dump-times {gimple_call <__cxa_allocate_exception, } 1 optimized } } + { dg-final { scan-tree-dump-times {gimple_call <__cxa_throw, } 1 optimized } } + We don't print anything, but just 'abort'. + + There is no 'catch'ing; any exception is fatal. + { dg-shouldfail {'MyException' exception} } */ diff --git a/gcc/testsuite/g++.target/nvptx/exceptions-throw-3.C b/gcc/testsuite/g++.target/nvptx/exceptions-throw-3.C new file mode 100644 index 0000000..5c1ad7a --- /dev/null +++ b/gcc/testsuite/g++.target/nvptx/exceptions-throw-3.C @@ -0,0 +1,11 @@ +/* 'throw', dead code. */ + +/* { dg-do run } */ +/* Via the magic string "-std=*++" indicate that testing one (the default) C++ standard is sufficient. */ +/* { dg-additional-options -fexceptions } */ +/* { dg-additional-options -fdump-tree-optimized-raw } */ + +#include "../../../../libgomp/testsuite/libgomp.oacc-c++/exceptions-throw-3.C" + +/* { dg-final { scan-tree-dump-times {gimple_call <__cxa_allocate_exception, } 1 optimized } } + { dg-final { scan-tree-dump-times {gimple_call <__cxa_throw, } 1 optimized } } */ diff --git a/gcc/testsuite/g++.target/nvptx/pr119692-1-1.C b/gcc/testsuite/g++.target/nvptx/pr119692-1-1.C new file mode 100644 index 0000000..b44b08d --- /dev/null +++ b/gcc/testsuite/g++.target/nvptx/pr119692-1-1.C @@ -0,0 +1,6 @@ +/* PR119692 "C++ 'typeinfo', 'vtable' vs. OpenACC, OpenMP 'target' offloading" */ + +/* { dg-do run } */ +/* Via the magic string "-std=*++" indicate that testing one (the default) C++ standard is sufficient. */ + +#include "../../../../libgomp/testsuite/libgomp.oacc-c++/pr119692-1-1.C" diff --git a/gcc/testsuite/g++.target/riscv/rvv/autovec/pr119547-2.C b/gcc/testsuite/g++.target/riscv/rvv/autovec/pr119547-2.C new file mode 100644 index 0000000..1b98d3d --- /dev/null +++ b/gcc/testsuite/g++.target/riscv/rvv/autovec/pr119547-2.C @@ -0,0 +1,212 @@ +/* { dg-do run { target rv64 } } */ +/* { dg-require-effective-target riscv_v_ok } */ +/* { dg-options "-O3 -march=rv64gcv -mabi=lp64d --param=logical-op-non-short-circuit=0" } */ + +#include <riscv_vector.h> + +using v_uint8 = vuint8m2_t; +using v_int8 = vint8m2_t; +using v_uint16 = vuint16m2_t; +using v_int16 = vint16m2_t; +using v_uint32 = vuint32m2_t; +using v_int32 = vint32m2_t; +using v_uint64 = vuint64m2_t; +using v_int64 = vint64m2_t; +using v_float32 = vfloat32m2_t; +using v_float64 = vfloat64m2_t; + +using uchar = unsigned char; +using schar = signed char; +using ushort = unsigned short; +using uint = unsigned int; +using uint64 = unsigned long int; +using int64 = long int; + +struct Size +{ + int width; + int height; +}; + +template <class T> struct VTraits; + +template <> struct VTraits<vint32m1_t> +{ + static inline int vlanes () { return __riscv_vsetvlmax_e32m1 (); } + using lane_type = int32_t; + static const int max_nlanes = 1024 / 32 * 2; +}; +template <> struct VTraits<vint32m2_t> +{ + static inline int vlanes () { return __riscv_vsetvlmax_e32m2 (); } + using lane_type = int32_t; + static const int max_nlanes = 1024 / 32 * 2; +}; +template <> struct VTraits<vint32m4_t> +{ + static inline int vlanes () { return __riscv_vsetvlmax_e32m4 (); } + using lane_type = int32_t; + static const int max_nlanes = 1024 / 32 * 2; +}; +template <> struct VTraits<vint32m8_t> +{ + static inline int vlanes () { return __riscv_vsetvlmax_e32m8 (); } + using lane_type = int32_t; + static const int max_nlanes = 1024 / 32 * 2; +}; + +template <> struct VTraits<vfloat64m1_t> +{ + static inline int vlanes () { return __riscv_vsetvlmax_e64m1 (); } + using lane_type = double; + static const int max_nlanes = 1024 / 64 * 2; +}; +template <> struct VTraits<vfloat64m2_t> +{ + static inline int vlanes () { return __riscv_vsetvlmax_e64m2 (); } + using lane_type = double; + static const int max_nlanes = 1024 / 64 * 2; +}; +template <> struct VTraits<vfloat64m4_t> +{ + static inline int vlanes () { return __riscv_vsetvlmax_e64m4 (); } + using lane_type = double; + static const int max_nlanes = 1024 / 64 * 2; +}; +template <> struct VTraits<vfloat64m8_t> +{ + static inline int vlanes () { return __riscv_vsetvlmax_e64m8 (); } + using lane_type = double; + static const int max_nlanes = 1024 / 64 * 2; +}; + +static inline v_float64 +v_setall_f64 (double v) +{ + return __riscv_vfmv_v_f_f64m2 (v, VTraits<v_float64>::vlanes ()); +} +static inline v_float64 +vx_setall_f64 (double v) +{ + return v_setall_f64 (v); +} + +inline v_int32 +v_load_expand_q (const schar *ptr) +{ + return __riscv_vwcvt_x ( + __riscv_vwcvt_x (__riscv_vle8_v_i8mf2 (ptr, VTraits<v_int32>::vlanes ()), + VTraits<v_int32>::vlanes ()), + VTraits<v_int32>::vlanes ()); +} + +static inline v_int32 +vx_load_expand_q (const schar *ptr) +{ + return v_load_expand_q (ptr); +} + +inline v_float64 +v_cvt_f64 (const v_int32 &a) +{ + return __riscv_vget_f64m2 (__riscv_vfwcvt_f (a, VTraits<v_int32>::vlanes ()), + 0); +} + +inline v_float64 +v_cvt_f64_high (const v_int32 &a) +{ + return __riscv_vget_f64m2 (__riscv_vfwcvt_f (a, VTraits<v_int32>::vlanes ()), + 1); +} + +inline void +v_store (double *ptr, const v_float64 &a) +{ + __riscv_vse64 (ptr, a, VTraits<v_float64>::vlanes ()); +} + +static inline void +v_store_pair_as (double *ptr, const v_float64 &a, const v_float64 &b) +{ + v_store (ptr, a); + v_store (ptr + VTraits<v_float64>::vlanes (), b); +} + +static inline void +vx_load_pair_as (const schar *ptr, v_float64 &a, v_float64 &b) +{ + v_int32 v0 = vx_load_expand_q (ptr); + a = v_cvt_f64 (v0); + b = v_cvt_f64_high (v0); +} + +inline v_float64 +v_fma (const v_float64 &a, const v_float64 &b, const v_float64 &c) +{ + return __riscv_vfmacc_vv_f64m2 (c, a, b, VTraits<v_float64>::vlanes ()); +} + +template <typename _Tp> +static inline _Tp +saturate_cast (double v) +{ + return _Tp (v); +} + +template <typename _Ts, typename _Td> +__attribute__ ((noipa)) void +cvt_64f (const _Ts *src, size_t sstep, _Td *dst, size_t dstep, Size size, + double a, double b) +{ + v_float64 va = vx_setall_f64 (a), vb = vx_setall_f64 (b); + const int VECSZ = VTraits<v_float64>::vlanes () * 2; + + sstep /= sizeof (src[0]); + dstep /= sizeof (dst[0]); + + for (int i = 0; i < size.height; i++, src += sstep, dst += dstep) + { + int j = 0; + + for (; j < size.width; j += VECSZ) + { + if (j > size.width - VECSZ) + { + if (j == 0 || src == (_Ts *) dst) + break; + j = size.width - VECSZ; + } + v_float64 v0, v1; + vx_load_pair_as (src + j, v0, v1); + v0 = v_fma (v0, va, vb); + v1 = v_fma (v1, va, vb); + v_store_pair_as (dst + j, v0, v1); + } + + for (; j < size.width; j++) + dst[j] = saturate_cast<_Td> (src[j] * a + b); + } +} + +void +__attribute__ ((noipa)) +cvtScale8s64f (const uchar *src_, size_t sstep, const uchar *, size_t, + uchar *dst_, size_t dstep, Size size, void *scale_) +{ + const schar *src = (const schar *) src_; + double *dst = (double *) dst_; + double *scale = (double *) scale_; + cvt_64f (src, sstep, dst, dstep, size, (double) scale[0], (double) scale[1]); +} + +int main () +{ + uchar src[1024]; + uchar dst[1024]; + + double scale[2] = {2.0, 3.0}; + Size size {4, 1}; + + cvtScale8s64f (src, 4, NULL, 0, dst, 32, size, (void *)scale); +} diff --git a/gcc/testsuite/g++.target/riscv/rvv/autovec/pr119547.C b/gcc/testsuite/g++.target/riscv/rvv/autovec/pr119547.C new file mode 100644 index 0000000..bac0fb1 --- /dev/null +++ b/gcc/testsuite/g++.target/riscv/rvv/autovec/pr119547.C @@ -0,0 +1,82 @@ +/* { dg-do run { target rv64 } } */ +/* { dg-require-effective-target riscv_v_ok } */ +/* { dg-options "-O3 -march=rv64gcv -mabi=lp64d --param=logical-op-non-short-circuit=0" } */ + +#include <riscv_vector.h> +using v_int32 = vint32m2_t; +using v_float64 = vfloat64m2_t; +struct Size +{ + int width; + int height; +}; +template <class> struct VTraits +{ + static int vlanes () { return __riscv_vsetvlmax_e32m2 (); } +}; +v_int32 +v_load_expand_q (const signed char *ptr) +{ + return __riscv_vwcvt_x ( + __riscv_vwcvt_x (__riscv_vle8_v_i8mf2 (ptr, VTraits<v_int32>::vlanes ()), + VTraits<v_int32>::vlanes ()), + VTraits<v_int32>::vlanes ()); +} +v_float64 +v_cvt_f64_high (v_int32 a) +{ + return __riscv_vget_f64m2 (__riscv_vfwcvt_f (a, VTraits<v_int32>::vlanes ()), + 1); +} +void +v_store (double *ptr, v_float64 a) +{ + __riscv_vse64 (ptr, a, __riscv_vsetvlmax_e64m2 ()); +} +void +v_store_pair_as (double *ptr, v_float64 b) +{ + v_store (ptr, b); +} +void +vx_load_pair_as (const signed char *ptr, v_float64, v_float64 &b) +{ + v_int32 v0; + b = v_cvt_f64_high (v0); +}; +void +cvt_64f (const signed char *src, double *dst, Size size) +{ + int VECSZ = __riscv_vsetvlmax_e64m2 (); + for (int i; i < size.height; i++) + { + int j; + for (;; j += VECSZ) + { + if (j > -VECSZ) + if (j == 0 || dst) + break; + v_float64 v0, v1; + vx_load_pair_as (src, v0, v1); + v_store_pair_as (dst, v1); + } + for (; j < size.width; j++) + dst[j] = (src[j]); + } +} +void +cvtScale8s64f (unsigned char *src_, unsigned char *dst_, + size_t, Size size, void *) +{ + signed char src; + double dst = *dst_; + cvt_64f (&src, &dst, size); +} +int main () +{ + unsigned char src[1]; + unsigned char dst[1024]; + double scale[1]; + Size size{4, 1}; + cvtScale8s64f (src, dst, 32, size, scale); +} diff --git a/gcc/testsuite/g++.target/s390/pr119834.C b/gcc/testsuite/g++.target/s390/pr119834.C new file mode 100644 index 0000000..66c0a69 --- /dev/null +++ b/gcc/testsuite/g++.target/s390/pr119834.C @@ -0,0 +1,76 @@ +// PR target/119834 +// { dg-do compile { target c++11 } } +// { dg-options "-O2 -march=z900" } + +int *a; +struct A; +struct B { + A begin (); + A end (); + operator bool * (); + void operator++ (); +}; +template <typename T> +auto operator| (int, T x) -> decltype (x (0)); +struct A : B { bool a; }; +struct C { A operator () (int); }; +enum D {} d; +int e; +void foo (); +struct E { + template <typename T> + T *garply () + { + if (d) + return 0; + if (e) + foo (); + return reinterpret_cast<T *> (f); + } + template <typename> + void bar (long x, bool) + { + if (&g - f) + __builtin_memset (a, 0, x); + f += x; + } + template <typename T> + T *baz (T *x, long y, bool z = true) + { + if (d) + return nullptr; + bar<T> ((char *)x + y - f, z); + return x; + } + template <typename T> + void qux (T x) { baz (x, x->j); } + char *f, g; +} *h; +struct F { + template <typename T> + int corge (T x) { x.freddy (this); return 0; } + template <typename T> + int boo (T x) { corge (x); return 0; } +} i; +template <typename T> +struct G { + template <typename U> friend T operator+ (U, G); + template <typename U> + void waldo (F *x, G y, U z) { x->boo (z + y); } + template <typename... Ts> + void plugh (E *y, Ts... z) { T *x = y->garply<T> (); x->thud (y, z...); } +}; +template <typename T> using H = G<T>; +struct I { + static constexpr unsigned j = 2; + void thud (E *x, A y) { x->qux (this); for (auto g : y) ; } +}; +H<I> k; +struct J { + void freddy (F *) { C a; auto b = 0 | a; k.plugh (h, b); } +}; +H<J> l; +struct K { + void freddy () { l.waldo (&i, l, this); } +}; +void grault () { K m; m.freddy (); } diff --git a/gcc/testsuite/gcc.dg/Wbuiltin-declaration-mismatch-4.c b/gcc/testsuite/gcc.dg/Wbuiltin-declaration-mismatch-4.c index c48fe5f..09aaaa6 100644 --- a/gcc/testsuite/gcc.dg/Wbuiltin-declaration-mismatch-4.c +++ b/gcc/testsuite/gcc.dg/Wbuiltin-declaration-mismatch-4.c @@ -77,9 +77,9 @@ void test_integer_conversion_memset (void *d) /* Passing a ptrdiff_t where size_t is expected may not be unsafe but because GCC may emits suboptimal code for such calls warning for them helps improve efficiency. */ - memset (d, 0, diffi); /* { dg-warning ".memset. argument 3 promotes to .ptrdiff_t. {aka .\(long \)?\(int\)?\(__int20\)?.} where .\(long \)?\(__int20 \)?unsigned\( int\)?. is expected" } */ + memset (d, 0, diffi); /* { dg-warning ".memset. argument 3 promotes to .ptrdiff_t. {aka .\(long \)*\(int\)?\(__int20\)?.} where .\(long \)*\(__int20 \)?unsigned\( int\)?. is expected" } */ - memset (d, 0, 2.0); /* { dg-warning ".memset. argument 3 type is .double. where '\(long \)?\(__int20 \)?unsigned\( int\)?' is expected" } */ + memset (d, 0, 2.0); /* { dg-warning ".memset. argument 3 type is .double. where '\(long \)*\(__int20 \)?unsigned\( int\)?' is expected" } */ /* Verify that the same call as above but to the built-in doesn't trigger a warning. */ diff --git a/gcc/testsuite/gcc.dg/bf-ms-attrib.c b/gcc/testsuite/gcc.dg/bf-ms-attrib.c index 2da4f03..5208c7f 100644 --- a/gcc/testsuite/gcc.dg/bf-ms-attrib.c +++ b/gcc/testsuite/gcc.dg/bf-ms-attrib.c @@ -32,7 +32,7 @@ main() /* As long as the sizes are as expected, we know attributes are working. bf-ms-layout.c makes sure the right thing happens when the attribute is on. */ - if (sizeof(struct one_ms) != 8) + if (sizeof(struct one_ms) != 12) abort(); if (sizeof(struct one_gcc) != 8) abort(); diff --git a/gcc/testsuite/gcc.dg/bitint-121.c b/gcc/testsuite/gcc.dg/bitint-121.c new file mode 100644 index 0000000..335b9a1 --- /dev/null +++ b/gcc/testsuite/gcc.dg/bitint-121.c @@ -0,0 +1,24 @@ +/* PR middle-end/119808 */ +/* { dg-do run { target { bitint && fstack_protector } } } */ +/* { dg-options "-O0 -ftree-coalesce-vars -fstack-protector-strong" } */ + +#if __BITINT_MAXWIDTH__ >= 129 +_BitInt(129) +foo () +{ + _BitInt(129) b = 0; + _BitInt(8) a + =__builtin_stdc_rotate_right (0x8c82111b5d2d37c57e9ada7213ed95a49uwb, b); + return b; +} +#endif + +int +main () +{ +#if __BITINT_MAXWIDTH__ >= 129 + _BitInt(129) x = foo (); + if (x) + __builtin_abort (); +#endif +} diff --git a/gcc/testsuite/gcc.dg/bitint-122.c b/gcc/testsuite/gcc.dg/bitint-122.c new file mode 100644 index 0000000..c791969 --- /dev/null +++ b/gcc/testsuite/gcc.dg/bitint-122.c @@ -0,0 +1,20 @@ +/* PR tree-optimization/116093 */ +/* { dg-do run { target bitint } } */ +/* { dg-options "-Og -ftree-vrp -fno-tree-dce" } */ + +#if __BITINT_MAXWIDTH__ >= 129 +char +foo (int a, _BitInt (129) b, char c) +{ + return c << (5 / b % (0xdb75dbf5 | a)); +} +#endif + +int +main () +{ +#if __BITINT_MAXWIDTH__ >= 129 + if (foo (0, 6, 1) != 1) + __builtin_abort (); +#endif +} diff --git a/gcc/testsuite/gcc.dg/completion-2.c b/gcc/testsuite/gcc.dg/completion-2.c index 99e6531..46c511c 100644 --- a/gcc/testsuite/gcc.dg/completion-2.c +++ b/gcc/testsuite/gcc.dg/completion-2.c @@ -5,6 +5,7 @@ -flto-partition=1to1 -flto-partition=balanced -flto-partition=cache +-flto-partition=default -flto-partition=max -flto-partition=none -flto-partition=one diff --git a/gcc/testsuite/gcc.dg/ipa/ipa-sra-19.c b/gcc/testsuite/gcc.dg/ipa/ipa-sra-19.c index c34c89e..39874ad 100644 --- a/gcc/testsuite/gcc.dg/ipa/ipa-sra-19.c +++ b/gcc/testsuite/gcc.dg/ipa/ipa-sra-19.c @@ -1,7 +1,7 @@ /* { dg-do compile } */ /* { dg-options "-O2" } */ /* { dg-additional-options "-msse2" { target ia32 } } */ -/* { dg-additional-options "-Wno-psabi" { target powerpc-ibm-aix* powerpc-wrs-vxworks* } } */ +/* { dg-additional-options "-Wno-psabi" { target powerpc-ibm-aix* powerpc-wrs-vxworks* powerpc-*-elf } } */ typedef int __attribute__((__vector_size__(16))) vectype; diff --git a/gcc/testsuite/gcc.dg/ipa/pr119318.c b/gcc/testsuite/gcc.dg/ipa/pr119318.c new file mode 100644 index 0000000..f179aed --- /dev/null +++ b/gcc/testsuite/gcc.dg/ipa/pr119318.c @@ -0,0 +1,37 @@ +/* { dg-do run } */ +/* { dg-require-effective-target int128 } */ +/* { dg-options "-Wno-psabi -w -O2" } */ + +typedef unsigned V __attribute__((vector_size (64))); +typedef unsigned __int128 W __attribute__((vector_size (64))); + +W a; +W b; +W c = { -0xffff, -0xffff, -0xffff, -0xffff }; + +static __attribute__((__noinline__, __noclone__)) W +bar (unsigned __int128 u) +{ + return u + c; +} + +static inline W +foo (unsigned short s, V v) +{ + V y = (V) bar ((unsigned short) ~s); + v >>= y; + b ^= (W) a; + v *= v; + return (W) v + b; +} + + +int +main () +{ + W x = foo (0, (V) { 0, 5 }); + for (unsigned i = 0; i < sizeof(x)/sizeof(x[0]); i++) + if (x[i] != (i ? 0 : 0x1900000000)) + __builtin_abort(); + return 0; +} diff --git a/gcc/testsuite/gcc.dg/ipa/pr119530.c b/gcc/testsuite/gcc.dg/ipa/pr119530.c new file mode 100644 index 0000000..f99c4fd --- /dev/null +++ b/gcc/testsuite/gcc.dg/ipa/pr119530.c @@ -0,0 +1,21 @@ +/* { dg-do run } */ +/* { dg-options "-O3 -fno-tree-vrp -fno-inline" } */ + +struct a { + int b; +}; +int c; +signed char d; +static int e(long long f) { return f < 0; } +static void g(unsigned f) { c = e(~f); } +int main() { + int h; + struct a i = {128}; + h = d > i.b; + g(h); + if (h) + __builtin_abort(); + if (c) + __builtin_abort(); + return 0; +} diff --git a/gcc/testsuite/gcc.dg/ipa/pr119803.c b/gcc/testsuite/gcc.dg/ipa/pr119803.c new file mode 100644 index 0000000..1a7bfd2 --- /dev/null +++ b/gcc/testsuite/gcc.dg/ipa/pr119803.c @@ -0,0 +1,16 @@ +/* { dg-do compile } */ +/* { dg-options "-O2" } */ + +extern void f(int p); +int a, b; +char c; +static int d(int e) { return !e || a == 1 ? 0 : a / e; } +static void h(short e) { + int g = d(e); + f(g); +} +void i() { + c = 128; + h(c); + b = d(65536); +} diff --git a/gcc/testsuite/gcc.dg/ira-shrinkwrap-prep-1.c b/gcc/testsuite/gcc.dg/ira-shrinkwrap-prep-1.c index 8c15097..2be31fa 100644 --- a/gcc/testsuite/gcc.dg/ira-shrinkwrap-prep-1.c +++ b/gcc/testsuite/gcc.dg/ira-shrinkwrap-prep-1.c @@ -25,5 +25,5 @@ bar (long a) } /* { dg-final { scan-rtl-dump "Will split live ranges of parameters" "ira" } } */ -/* { dg-final { scan-rtl-dump "Split live-range of register" "ira" { xfail { ! aarch64*-*-* } } } } */ +/* { dg-final { scan-rtl-dump "Split live-range of register" "ira" { xfail { ! { aarch64*-*-* i?86-*-* x86_64-*-* } } } } } */ /* { dg-final { scan-rtl-dump "Performing shrink-wrapping" "pro_and_epilogue" } } */ diff --git a/gcc/testsuite/gcc.dg/ira-shrinkwrap-prep-2.c b/gcc/testsuite/gcc.dg/ira-shrinkwrap-prep-2.c index 0690e03..61ee4c6 100644 --- a/gcc/testsuite/gcc.dg/ira-shrinkwrap-prep-2.c +++ b/gcc/testsuite/gcc.dg/ira-shrinkwrap-prep-2.c @@ -30,6 +30,6 @@ bar (long a) } /* { dg-final { scan-rtl-dump "Will split live ranges of parameters" "ira" } } */ -/* { dg-final { scan-rtl-dump "Split live-range of register" "ira" { xfail { ! aarch64*-*-* } } } } */ +/* { dg-final { scan-rtl-dump "Split live-range of register" "ira" { xfail { ! { aarch64*-*-* i?86-*-* x86_64-*-* } } } } } */ /* XFAIL due to PR70681. */ /* { dg-final { scan-rtl-dump "Performing shrink-wrapping" "pro_and_epilogue" { xfail arm*-*-* powerpc*-*-* } } } */ diff --git a/gcc/testsuite/gcc.dg/memcpy-4.c b/gcc/testsuite/gcc.dg/memcpy-4.c index 4c726f0..b17b369 100644 --- a/gcc/testsuite/gcc.dg/memcpy-4.c +++ b/gcc/testsuite/gcc.dg/memcpy-4.c @@ -1,13 +1,8 @@ /* { dg-do compile } */ -/* { dg-options "-O2 -fdump-rtl-expand" } */ +/* { dg-options "-O2" } */ -#ifdef __mips -__attribute__((nomips16)) -#endif void f1 (char *p) { __builtin_memcpy (p, "12345", 5); } - -/* { dg-final { scan-rtl-dump "mem/u.*mem/u" "expand" { target mips*-*-* } } } */ diff --git a/gcc/testsuite/gcc.dg/pr116357.c b/gcc/testsuite/gcc.dg/pr116357.c index 07effa1..12aaf62 100644 --- a/gcc/testsuite/gcc.dg/pr116357.c +++ b/gcc/testsuite/gcc.dg/pr116357.c @@ -2,9 +2,9 @@ /* { dg-do compile } */ /* { dg-options "" } */ -typedef int A __attribute__((aligned (2 * alignof (int)))); +typedef int A __attribute__((aligned (2 * sizeof (int)))); A a[4]; /* { dg-error "alignment of array elements is greater than element size" } */ -typedef volatile int B __attribute__((aligned (2 * alignof (int)))); +typedef volatile int B __attribute__((aligned (2 * sizeof (int)))); B b[4]; /* { dg-error "alignment of array elements is greater than element size" } */ -typedef const int C __attribute__((aligned (2 * alignof (int)))); +typedef const int C __attribute__((aligned (2 * sizeof (int)))); C c[4]; /* { dg-error "alignment of array elements is greater than element size" } */ diff --git a/gcc/testsuite/gcc.dg/pr118947-1.c b/gcc/testsuite/gcc.dg/pr118947-1.c new file mode 100644 index 0000000..8733e8d --- /dev/null +++ b/gcc/testsuite/gcc.dg/pr118947-1.c @@ -0,0 +1,15 @@ +/* PR tree-optimization/118947 */ +/* { dg-do compile { target size32plus } } */ +/* { dg-options "-O2 -fdump-tree-forwprop1-details" } */ +/* { dg-final { scan-tree-dump-times "after previous" 1 "forwprop1" } } */ + +void* aaa(); +void* bbb() +{ + char buf[1025] = {}; + /* Tha call to aaa should not matter and clobber buf. */ + void* ret = aaa(); + __builtin_memcpy(ret, buf, sizeof(buf)); + return ret; +} + diff --git a/gcc/testsuite/gcc.dg/pr119160.c b/gcc/testsuite/gcc.dg/pr119160.c new file mode 100644 index 0000000..b4629a1 --- /dev/null +++ b/gcc/testsuite/gcc.dg/pr119160.c @@ -0,0 +1,26 @@ +/* { dg-do run } */ +/* { dg-options "-O2 -finstrument-functions-once -favoid-store-forwarding -fnon-call-exceptions -fschedule-insns -mgeneral-regs-only -Wno-psabi" } */ + +typedef __attribute__((__vector_size__ (32))) int V; + +void +foo (V v, V, V, V *r) +{ + V u = (V){} + v[0]; + *r = u; +} + +__attribute__((__noipa__)) void +bar(int x) +{ + if (x != 2) __builtin_abort(); +} + +int +main () +{ + V x; + foo ((V){ 2, 3 }, (V){ }, (V){ }, &x); + for (unsigned i = 0; i < sizeof(x)/sizeof(x[0]); i++) + bar(x[i]); +}
\ No newline at end of file diff --git a/gcc/testsuite/gcc.dg/pr119717.c b/gcc/testsuite/gcc.dg/pr119717.c new file mode 100644 index 0000000..e5eedc5 --- /dev/null +++ b/gcc/testsuite/gcc.dg/pr119717.c @@ -0,0 +1,24 @@ +/* PR c/119717 */ +/* { dg-additional-options "-std=c23" } */ +/* { dg-do compile } */ + +struct annotated { + unsigned count; + [[gnu::counted_by(count)]] char array[]; +}; + +[[gnu::noinline,gnu::noipa]] +static unsigned +size_of (bool x, struct annotated *a) +{ + char *p = (x ? a : 0)->array; + return __builtin_dynamic_object_size (p, 1); +} + +int main() +{ + struct annotated *p = __builtin_malloc(sizeof *p); + p->count = 0; + __builtin_printf ("the bdos whole is %ld\n", size_of (0, p)); + return 0; +} diff --git a/gcc/testsuite/gcc.dg/pr78408-3.c b/gcc/testsuite/gcc.dg/pr78408-3.c new file mode 100644 index 0000000..5ea5458 --- /dev/null +++ b/gcc/testsuite/gcc.dg/pr78408-3.c @@ -0,0 +1,14 @@ +/* PR tree-optimization/78408 */ +/* { dg-do compile { target size32plus } } */ +/* { dg-options "-O2 -fdump-tree-forwprop1-details" } */ +/* { dg-final { scan-tree-dump-times "after previous" 1 "forwprop1" } } */ + +void* aaa(); +void* bbb() +{ + void* ret = aaa(); + char buf[1025] = {}; + __builtin_memcpy(ret, buf, sizeof(buf)); + return ret; +} + diff --git a/gcc/testsuite/gcc.dg/rtl/i386/vector_eq-2.c b/gcc/testsuite/gcc.dg/rtl/i386/vector_eq-2.c new file mode 100644 index 0000000..871d489 --- /dev/null +++ b/gcc/testsuite/gcc.dg/rtl/i386/vector_eq-2.c @@ -0,0 +1,71 @@ +/* { dg-do compile { target { i?86-*-* x86_64-*-* } } } */ +/* { dg-additional-options "-O2 -march=x86-64-v3" } */ + +typedef int v4si __attribute__((vector_size(16))); +typedef int v8si __attribute__((vector_size(32))); +typedef int v2di __attribute__((vector_size(16))); + +v4si __RTL (startwith ("vregs1")) foo1 (void) +{ +(function "foo1" + (insn-chain + (block 2 + (edge-from entry (flags "FALLTHRU")) + (cnote 1 [bb 2] NOTE_INSN_BASIC_BLOCK) + (cnote 2 NOTE_INSN_FUNCTION_BEG) + (cinsn 3 (set (reg:V4SI <0>) (const_vector:V4SI [(const_int -1) (const_int -1) (const_int -1) (const_int -1)]))) + (cinsn 4 (set (reg:V4SI <1>) (const_vector:V4SI [(const_int -1) (const_int -1) (const_int -1) (const_int -1)]))) + (cinsn 5 (set (reg:V4SI <2>) + (eq:V4SI (reg:V4SI <0>) (reg:V4SI <1>)))) + (cinsn 6 (set (reg:V4SI <3>) (reg:V4SI <2>))) + (cinsn 7 (set (reg:V4SI xmm0) (reg:V4SI <3>))) + (edge-to exit (flags "FALLTHRU")) + ) + ) + (crtl (return_rtx (reg/i:V4SI xmm0))) +) +} + +v8si __RTL (startwith ("vregs1")) foo2 (void) +{ +(function "foo2" + (insn-chain + (block 2 + (edge-from entry (flags "FALLTHRU")) + (cnote 1 [bb 2] NOTE_INSN_BASIC_BLOCK) + (cnote 2 NOTE_INSN_FUNCTION_BEG) + (cinsn 3 (set (reg:V8SI <0>) (const_vector:V8SI [(const_int -1) (const_int -1) (const_int -1) (const_int -1) (const_int -1) (const_int -1) (const_int -1) (const_int -1)]))) + (cinsn 4 (set (reg:V8SI <1>) (const_vector:V8SI [(const_int -1) (const_int -1) (const_int -1) (const_int -1) (const_int -1) (const_int -1) (const_int -1) (const_int -1)]))) + (cinsn 5 (set (reg:V8SI <2>) + (eq:V8SI (reg:V8SI <0>) (reg:V8SI <1>)))) + (cinsn 6 (set (reg:V8SI <3>) (reg:V8SI <2>))) + (cinsn 7 (set (reg:V8SI xmm0) (reg:V8SI <3>))) + (edge-to exit (flags "FALLTHRU")) + ) + ) + (crtl (return_rtx (reg/i:V8SI xmm0))) +) +} + +v2di __RTL (startwith ("vregs1")) foo3 (void) +{ +(function "foo3" + (insn-chain + (block 2 + (edge-from entry (flags "FALLTHRU")) + (cnote 1 [bb 2] NOTE_INSN_BASIC_BLOCK) + (cnote 2 NOTE_INSN_FUNCTION_BEG) + (cinsn 3 (set (reg:V2DI <0>) (const_vector:V2DI [(const_int -1) (const_int -1)]))) + (cinsn 4 (set (reg:V2DI <1>) (const_vector:V2DI [(const_int -1) (const_int -1)]))) + (cinsn 5 (set (reg:V2DI <2>) + (eq:V2DI (reg:V2DI <0>) (reg:V2DI <1>)))) + (cinsn 6 (set (reg:V2DI <3>) (reg:V2DI <2>))) + (cinsn 7 (set (reg:V2DI xmm0) (reg:V2DI <3>))) + (edge-to exit (flags "FALLTHRU")) + ) + ) + (crtl (return_rtx (reg/i:V2DI xmm0))) +) +} + +/* { dg-final { scan-assembler-times "vpcmpeq" 3 } } */ diff --git a/gcc/testsuite/gcc.dg/rtl/i386/vector_eq-3.c b/gcc/testsuite/gcc.dg/rtl/i386/vector_eq-3.c new file mode 100644 index 0000000..276c4c2 --- /dev/null +++ b/gcc/testsuite/gcc.dg/rtl/i386/vector_eq-3.c @@ -0,0 +1,74 @@ +/* { dg-do compile { target { i?86-*-* x86_64-*-* } } } */ +/* { dg-additional-options "-O2 -march=x86-64-v3" } */ + +typedef int v4si __attribute__((vector_size(16))); +typedef int v8si __attribute__((vector_size(32))); +typedef int v2di __attribute__((vector_size(16))); + +v4si __RTL (startwith ("vregs1")) foo1 (void) +{ +(function "foo1" + (insn-chain + (block 2 + (edge-from entry (flags "FALLTHRU")) + (cnote 1 [bb 2] NOTE_INSN_BASIC_BLOCK) + (cnote 2 NOTE_INSN_FUNCTION_BEG) + (cinsn 3 (set (reg:V4SI <1>) + (mem:V4SI (reg:SI di) [0 ptr S128 A128]))) + (cinsn 4 (set (reg:V4SI <2>) + (eq:V4SI (reg:V4SI <1>) + (mem:V4SI (reg:SI di) [0 ptr S128 A128])))) + (cinsn 5 (set (reg:V4SI <3>) (reg:V4SI <2>))) + (cinsn 6 (set (reg:V4SI xmm0) (reg:V4SI <3>))) + (edge-to exit (flags "FALLTHRU")) + ) + ) + (crtl (return_rtx (reg/i:V4SI xmm0))) +) +} + +v8si __RTL (startwith ("vregs1")) foo2 (void) +{ +(function "foo2" + (insn-chain + (block 2 + (edge-from entry (flags "FALLTHRU")) + (cnote 1 [bb 2] NOTE_INSN_BASIC_BLOCK) + (cnote 2 NOTE_INSN_FUNCTION_BEG) + (cinsn 3 (set (reg:V8SI <1>) + (mem:V8SI (reg:SI di) [0 ptr S256 A256]))) + (cinsn 4 (set (reg:V8SI <2>) + (eq:V8SI (mem:V8SI (reg:SI di) [0 ptr S256 A256]) + (reg:V8SI <1>)))) + (cinsn 5 (set (reg:V8SI <3>) (reg:V8SI <2>))) + (cinsn 6 (set (reg:V8SI xmm0) (reg:V8SI <3>))) + (edge-to exit (flags "FALLTHRU")) + ) + ) + (crtl (return_rtx (reg/i:V8SI xmm0))) +) +} + +v2di __RTL (startwith ("vregs1")) foo3 (void) +{ +(function "foo3" + (insn-chain + (block 2 + (edge-from entry (flags "FALLTHRU")) + (cnote 1 [bb 2] NOTE_INSN_BASIC_BLOCK) + (cnote 2 NOTE_INSN_FUNCTION_BEG) + (cinsn 3 (set (reg:V2DI <1>) + (mem:V2DI (reg:SI di) [0 ptr S128 A128]))) + (cinsn 4 (set (reg:V2DI <2>) + (eq:V2DI (reg:V2DI <1>) + (mem:V2DI (reg:SI di) [0 ptr S128 A128])))) + (cinsn 5 (set (reg:V2DI <3>) (reg:V2DI <2>))) + (cinsn 6 (set (reg:V2DI xmm0) (reg:V2DI <3>))) + (edge-to exit (flags "FALLTHRU")) + ) + ) + (crtl (return_rtx (reg/i:V2DI xmm0))) +) +} + +/* { dg-final { scan-assembler-times "vpcmpeq" 3 } } */ diff --git a/gcc/testsuite/gcc.dg/torture/bitint-76.c b/gcc/testsuite/gcc.dg/torture/bitint-76.c new file mode 100644 index 0000000..df47857 --- /dev/null +++ b/gcc/testsuite/gcc.dg/torture/bitint-76.c @@ -0,0 +1,19 @@ +/* PR tree-optimization/119707 */ +/* { dg-do run { target bitint } } */ + +#if __BITINT_MAXWIDTH__ >= 256 +__attribute__((noipa)) unsigned _BitInt(256) +foo (unsigned _BitInt(256) x, _BitInt(129) y) +{ + return x + (unsigned _BitInt(255)) y; +} +#endif + +int +main () +{ +#if __BITINT_MAXWIDTH__ >= 256 + if (foo (0, -1) != 0x7fffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffuwb) + __builtin_abort (); +#endif +} diff --git a/gcc/testsuite/gcc.dg/torture/bitint-77.c b/gcc/testsuite/gcc.dg/torture/bitint-77.c new file mode 100644 index 0000000..3e2523f --- /dev/null +++ b/gcc/testsuite/gcc.dg/torture/bitint-77.c @@ -0,0 +1,26 @@ +/* PR tree-optimization/119722 */ +/* { dg-do run { target bitint } } */ +/* { dg-options "-O2 -fno-tree-forwprop -fno-tree-copy-prop -fno-tree-fre" } */ + +#if __BITINT_MAXWIDTH__ >= 33300 +unsigned _BitInt(33300) g; + +unsigned +foo (long c) +{ + unsigned _BitInt(33300) b + = __builtin_stdc_rotate_left ((unsigned _BitInt(13)) 8, c); + return ((unsigned _BitInt(50)) (g >> 50) + + ({ unsigned _BitInt(300) unused; b; })); +} +#endif + +int +main () +{ +#if __BITINT_MAXWIDTH__ >= 33300 + unsigned x = foo (0); + if (x != 8) + __builtin_abort (); +#endif +} diff --git a/gcc/testsuite/gcc.dg/torture/pr118476-1.c b/gcc/testsuite/gcc.dg/torture/pr118476-1.c new file mode 100644 index 0000000..33509403 --- /dev/null +++ b/gcc/testsuite/gcc.dg/torture/pr118476-1.c @@ -0,0 +1,14 @@ +/* { dg-do compile } */ + +/* PR tree-optimization/118476 */ + +typedef unsigned long long poly64x1 __attribute__((__vector_size__(1*sizeof(long long)))); + +poly64x1 vext_p64(poly64x1 a, poly64x1 b, const int n) +{ + poly64x1 r = a; + unsigned src = (unsigned)n; + long long t = b[0]; + r[0] = (src < 1) ? a[src] : t; + return r; +} diff --git a/gcc/testsuite/gcc.dg/tree-ssa/pr118902-1.c b/gcc/testsuite/gcc.dg/tree-ssa/pr118902-1.c new file mode 100644 index 0000000..fa21b8a --- /dev/null +++ b/gcc/testsuite/gcc.dg/tree-ssa/pr118902-1.c @@ -0,0 +1,21 @@ +/* { dg-do compile } */ +/* { dg-options "-O2 -fdump-tree-optimized" } */ + +void foo(int); +void l(int**); +int f1(int j, int t) +{ + int p = 0; + int *a = &p; + l(&a); + if (a == &p) + return 0; + for(int i = 0; i < j; i++) + { + if (a == &p) foo(p); + } + return 0; +} + +/* We should be able to remove the call to foo because a is never equal to &p inside the loop. */ +/* { dg-final { scan-tree-dump-not "foo " "optimized"} } */ diff --git a/gcc/testsuite/gcc.dg/tree-ssa/ssa-dom-cse-2.c b/gcc/testsuite/gcc.dg/tree-ssa/ssa-dom-cse-2.c index a879d30..6fa52f6 100644 --- a/gcc/testsuite/gcc.dg/tree-ssa/ssa-dom-cse-2.c +++ b/gcc/testsuite/gcc.dg/tree-ssa/ssa-dom-cse-2.c @@ -27,4 +27,4 @@ foo () but the loop reads only one element at a time, and DOM cannot resolve these. The same happens on powerpc depending on the SIMD support available. */ -/* { dg-final { scan-tree-dump "return 28;" "optimized" { xfail { { alpha*-*-* hppa*64*-*-* nvptx*-*-* mmix-knuth-mmixware } || { { { lp64 && { powerpc*-*-* sparc*-*-* riscv*-*-* } } || aarch64_sve } || { arm*-*-* && { ! arm_neon } } } } } } } */ +/* { dg-final { scan-tree-dump "return 28;" "optimized" { xfail { { alpha*-*-* hppa*64*-*-* nvptx*-*-* mmix-knuth-mmixware } || { { { lp64 && { mips*-*-* powerpc*-*-* sparc*-*-* riscv*-*-* } } || aarch64_sve } || { arm*-*-* && { ! arm_neon } } } } } } } */ diff --git a/gcc/testsuite/gcc.dg/tree-ssa/ssa-dse-52.c b/gcc/testsuite/gcc.dg/tree-ssa/ssa-dse-52.c new file mode 100644 index 0000000..9e605ac --- /dev/null +++ b/gcc/testsuite/gcc.dg/tree-ssa/ssa-dse-52.c @@ -0,0 +1,30 @@ +/* { dg-options "-O2 -fdump-tree-dse-details -fno-tree-fre -fdump-tree-optimized" } */ + +/* PR tree-optimization/87901 */ + +char z[32]; +void foo1(void) +{ + char z1[17]; + char z2[24]; + __builtin_memset (z1, 0, 17); + __builtin_memcpy (z, z1, 17); + __builtin_memset (z2, 0, 24); + __builtin_memcpy (z+8, z2, 24); +} + +/* we should get: + MEM <unsigned char[8]> [(char * {ref-all})&z] = {}; + MEM <unsigned char[24]> [(char * {ref-all})&z + 8B] = {}; + after DSE; trimming the first memset to z (which was memcpy) to 8 bytes + from the original 17. + and not have a [17] in the IR after DSE. + The two memset to z1/z2 will also be removed. + */ +/* { dg-final { scan-tree-dump-not "\\\[17\\\]" "optimized" } } */ +/* { dg-final { scan-tree-dump "\\\[8\\\]" "dse1" } } */ + +/* { dg-final { scan-tree-dump-times "Trimming statement " 1 "dse1" } } */ +/* { dg-final { scan-tree-dump-times "Deleted dead call:" 2 "dse1" } } */ + + diff --git a/gcc/testsuite/gcc.dg/tree-ssa/ssa-dse-53.c b/gcc/testsuite/gcc.dg/tree-ssa/ssa-dse-53.c new file mode 100644 index 0000000..a2df591 --- /dev/null +++ b/gcc/testsuite/gcc.dg/tree-ssa/ssa-dse-53.c @@ -0,0 +1,26 @@ +/* { dg-options "-O2 -fno-strict-aliasing -fdump-tree-dse-details -fno-tree-fre -fdump-tree-optimized" } */ + +/* PR tree-optimization/87901 */ + + +int i; +int foo () +{ + i = 0; + *((short *)&i + 1) = 1; + return i; +} + +/* we should get: + MEM <char[2]> [(int *)&i] = {}; + MEM[(short int *)&i + 2B] = 1; + in DSE1. + + Note later on the stores will be merged. */ +/* { dg-final { scan-tree-dump "return 65536;" "optimized" { target le } } } */ +/* { dg-final { scan-tree-dump "return 1;" "optimized" { target be } } } */ +/* { dg-final { scan-tree-dump "\\\[2\\\]" "dse1" } } */ + +/* { dg-final { scan-tree-dump-times "Trimming statement " 1 "dse1" } } */ + + diff --git a/gcc/testsuite/gcc.dg/tree-ssa/ssa-dse-54.c b/gcc/testsuite/gcc.dg/tree-ssa/ssa-dse-54.c new file mode 100644 index 0000000..7e79a73 --- /dev/null +++ b/gcc/testsuite/gcc.dg/tree-ssa/ssa-dse-54.c @@ -0,0 +1,25 @@ +/* { dg-options "-O2 -fdump-tree-dse-details -fno-tree-fre -fdump-tree-optimized" } */ + +/* PR tree-optimization/87901 */ + +int z[128]; +void foo1(void) +{ + int z1; + int z2[24/sizeof(int)]; + __builtin_memset (&z1, 0, sizeof(int)); + __builtin_memcpy (z, &z1, sizeof(int)); + __builtin_memset (z2, 0, 24); + __builtin_memcpy (((char*)z)+1, z2, 24); +} + +/* we should get: + MEM[(char * {ref-all})&z] = {}; + __builtin_memset (&MEM <int[128]> [(void *)&z + 1B], 0, 24); + */ + +/* { dg-final { scan-tree-dump-not "MEM <unsigned int>" "optimized" } } */ +/* { dg-final { scan-tree-dump-not "MEM \\\[" "dse1" } } */ +/* { dg-final { scan-tree-dump-times "Trimming statement " 1 "dse1" } } */ +/* { dg-final { scan-tree-dump-times "Deleted dead call:" 1 "dse1" } } */ + diff --git a/gcc/testsuite/gcc.dg/vect/pr119399.c b/gcc/testsuite/gcc.dg/vect/pr119399.c new file mode 100644 index 0000000..8d868f4 --- /dev/null +++ b/gcc/testsuite/gcc.dg/vect/pr119399.c @@ -0,0 +1,10 @@ +/* { dg-do compile } */ +/* { dg-additional-options "-fdump-tree-vect-raw" } */ + +void foo(int *p, int *q, int n) +{ + for (int i = 0; i < n; i++) + p[i] = q[i] + 1; +} + +/* { dg-final { scan-tree-dump-not {<pointer_diff_expr,} "vect" } } */ diff --git a/gcc/testsuite/gcc.dg/vect/pr119757.c b/gcc/testsuite/gcc.dg/vect/pr119757.c new file mode 100644 index 0000000..8644299 --- /dev/null +++ b/gcc/testsuite/gcc.dg/vect/pr119757.c @@ -0,0 +1,17 @@ +/* { dg-do compile } */ + +void base64_encode(const char *table64, + const char *inputbuff, int insize, + char * __restrict output) +{ + const unsigned char *in = (const unsigned char *)inputbuff; + + while(insize >= 3) { + *output++ = table64[ in[0] >> 2 ]; + *output++ = table64[ ((in[0] & 0x03) << 4) | (in[1] >> 4) ]; + *output++ = table64[ ((in[1] & 0x0F) << 2) | ((in[2] & 0xC0) >> 6) ]; + *output++ = table64[ in[2] & 0x3F ]; + insize -= 3; + in += 3; + } +} diff --git a/gcc/testsuite/gcc.dg/vect/vect-early-break_18.c b/gcc/testsuite/gcc.dg/vect/vect-early-break_18.c index edddb44..cd39704 100644 --- a/gcc/testsuite/gcc.dg/vect/vect-early-break_18.c +++ b/gcc/testsuite/gcc.dg/vect/vect-early-break_18.c @@ -2,7 +2,7 @@ /* { dg-do compile } */ /* { dg-require-effective-target vect_early_break } */ /* { dg-require-effective-target vect_int } */ - +/* { dg-additional-options "-march=gfx908" { target amdgcn*-*-* } } */ /* { dg-additional-options "-Ofast" } */ /* { dg-final { scan-tree-dump "LOOP VECTORIZED" "vect" { target vect_load_lanes } } } */ diff --git a/gcc/testsuite/gcc.target/aarch64/acle/rwsr-ungated.c b/gcc/testsuite/gcc.target/aarch64/acle/rwsr-ungated.c new file mode 100644 index 0000000..d67a426 --- /dev/null +++ b/gcc/testsuite/gcc.target/aarch64/acle/rwsr-ungated.c @@ -0,0 +1,13 @@ +/* Test that __arm_[r,w]sr intrinsics aren't gated (by default). */ + +/* { dg-do compile } */ +/* { dg-options "-march=armv8-a" } */ + +#include <arm_acle.h> + +uint64_t +foo (uint64_t a) +{ + __arm_wsr64 ("zcr_el1", a); + return __arm_rsr64 ("smcr_el1"); +} diff --git a/gcc/testsuite/gcc.target/aarch64/bic-1.c b/gcc/testsuite/gcc.target/aarch64/bic-1.c new file mode 100644 index 0000000..65e1514 --- /dev/null +++ b/gcc/testsuite/gcc.target/aarch64/bic-1.c @@ -0,0 +1,40 @@ +/* { dg-do compile } */ +/* { dg-options "-O2" } */ +/* { dg-final { check-function-bodies "**" "" "" } } */ + +/* PR rtl-optmization/111949 */ + +/* +**func1: +** bic w([0-9]+), w0, w1 +** and w0, w\1, 1 +** ret +*/ + +unsigned func1(unsigned a, bool b) +{ + int c = a & b; + return (c ^ a)&1; +} + +/* +**func2: +** bic w([0-9]+), w1, w0 +** and w0, w\1, 255 +** ret +*/ +unsigned func2(bool a, bool b) +{ + return ~a & b; +} + +/* +**func3: +** bic w([0-9]+), w1, w0 +** and w0, w\1, 1 +** ret +*/ +bool func3(bool a, unsigned char b) +{ + return !a & b; +} diff --git a/gcc/testsuite/gcc.target/aarch64/sve/peel_ind_10.c b/gcc/testsuite/gcc.target/aarch64/sve/peel_ind_10.c new file mode 100644 index 0000000..b7a7bc5 --- /dev/null +++ b/gcc/testsuite/gcc.target/aarch64/sve/peel_ind_10.c @@ -0,0 +1,24 @@ +/* Fix for PR119351 alignment peeling with vectors and VLS. */ +/* { dg-do compile } */ +/* { dg-options "-Ofast -msve-vector-bits=256 --param aarch64-autovec-preference=sve-only -fdump-tree-vect-details" } */ + +#define N 512 +#define START 0 +#define END 505 + +int x[N] __attribute__((aligned(32))); + +int __attribute__((noipa)) +foo (int start) +{ + for (unsigned int i = start; i < END; ++i) + { + if (x[i] == 0) + return i; + } + return -1; +} + +/* { dg-final { scan-tree-dump "LOOP VECTORIZED" "vect" } } */ +/* { dg-final { scan-tree-dump "pfa_iv_offset" "vect" } } */ +/* { dg-final { scan-tree-dump "Alignment of access forced using peeling" "vect" } } */ diff --git a/gcc/testsuite/gcc.target/aarch64/sve/peel_ind_10_run.c b/gcc/testsuite/gcc.target/aarch64/sve/peel_ind_10_run.c new file mode 100644 index 0000000..6169aeb --- /dev/null +++ b/gcc/testsuite/gcc.target/aarch64/sve/peel_ind_10_run.c @@ -0,0 +1,17 @@ +/* Fix for PR119351 alignment peeling with vectors and VLS. */ +/* { dg-do run { target aarch64_sve_hw } } */ +/* { dg-options "-Ofast --param aarch64-autovec-preference=sve-only" } */ +/* { dg-additional-options "-msve-vector-bits=256" { target aarch64_sve256_hw } } */ +/* { dg-additional-options "-msve-vector-bits=128" { target aarch64_sve128_hw } } */ + +#include "peel_ind_10.c" + +int __attribute__ ((optimize (1))) +main (void) +{ + int res = foo (START); + asm volatile (""); + if (res != START) + __builtin_abort (); + return 0; +} diff --git a/gcc/testsuite/gcc.target/aarch64/sve/peel_ind_5.c b/gcc/testsuite/gcc.target/aarch64/sve/peel_ind_5.c new file mode 100644 index 0000000..a03bb1d --- /dev/null +++ b/gcc/testsuite/gcc.target/aarch64/sve/peel_ind_5.c @@ -0,0 +1,24 @@ +/* Fix for PR119351 alignment peeling with vectors and VLS. */ +/* { dg-do compile } */ +/* { dg-options "-Ofast -msve-vector-bits=256 --param aarch64-autovec-preference=sve-only -fdump-tree-vect-details" } */ + +#define N 512 +#define START 2 +#define END 505 + +int x[N] __attribute__((aligned(32))); + +int __attribute__((noipa)) +foo (void) +{ + for (signed int i = START; i < END; ++i) + { + if (x[i] == 0) + return i; + } + return -1; +} + +/* { dg-final { scan-tree-dump "LOOP VECTORIZED" "vect" } } */ +/* { dg-final { scan-tree-dump "pfa_iv_offset" "vect" } } */ +/* { dg-final { scan-tree-dump "Alignment of access forced using peeling" "vect" } } */ diff --git a/gcc/testsuite/gcc.target/aarch64/sve/peel_ind_5_run.c b/gcc/testsuite/gcc.target/aarch64/sve/peel_ind_5_run.c new file mode 100644 index 0000000..f26befe --- /dev/null +++ b/gcc/testsuite/gcc.target/aarch64/sve/peel_ind_5_run.c @@ -0,0 +1,17 @@ +/* Fix for PR119351 alignment peeling with vectors and VLS. */ +/* { dg-do run { target aarch64_sve_hw } } */ +/* { dg-options "-Ofast --param aarch64-autovec-preference=sve-only" } */ +/* { dg-additional-options "-msve-vector-bits=256" { target aarch64_sve256_hw } } */ +/* { dg-additional-options "-msve-vector-bits=128" { target aarch64_sve128_hw } } */ + +#include "peel_ind_5.c" + +int __attribute__ ((optimize (1))) +main (void) +{ + int res = foo (); + asm volatile (""); + if (res != START) + __builtin_abort (); + return 0; +} diff --git a/gcc/testsuite/gcc.target/aarch64/sve/peel_ind_6.c b/gcc/testsuite/gcc.target/aarch64/sve/peel_ind_6.c new file mode 100644 index 0000000..9bfd1a6 --- /dev/null +++ b/gcc/testsuite/gcc.target/aarch64/sve/peel_ind_6.c @@ -0,0 +1,24 @@ +/* Fix for PR119351 alignment peeling with vectors and VLS. */ +/* { dg-do compile } */ +/* { dg-options "-Ofast -msve-vector-bits=256 --param aarch64-autovec-preference=sve-only -fdump-tree-vect-details" } */ + +#define N 512 +#define START 1 +#define END 505 + +int x[N] __attribute__((aligned(32))); + +int __attribute__((noipa)) +foo (int start) +{ + for (unsigned int i = start; i < END; ++i) + { + if (x[i] == 0) + return i; + } + return -1; +} + +/* { dg-final { scan-tree-dump "LOOP VECTORIZED" "vect" } } */ +/* { dg-final { scan-tree-dump "pfa_iv_offset" "vect" } } */ +/* { dg-final { scan-tree-dump "Alignment of access forced using peeling" "vect" } } */ diff --git a/gcc/testsuite/gcc.target/aarch64/sve/peel_ind_6_run.c b/gcc/testsuite/gcc.target/aarch64/sve/peel_ind_6_run.c new file mode 100644 index 0000000..4fdf3e4 --- /dev/null +++ b/gcc/testsuite/gcc.target/aarch64/sve/peel_ind_6_run.c @@ -0,0 +1,17 @@ +/* Fix for PR119351 alignment peeling with vectors and VLS. */ +/* { dg-do run { target aarch64_sve_hw } } */ +/* { dg-options "-Ofast --param aarch64-autovec-preference=sve-only" } */ +/* { dg-additional-options "-msve-vector-bits=256" { target aarch64_sve256_hw } } */ +/* { dg-additional-options "-msve-vector-bits=128" { target aarch64_sve128_hw } } */ + +#include "peel_ind_6.c" + +int __attribute__ ((optimize (1))) +main (void) +{ + int res = foo (START); + asm volatile (""); + if (res != START) + __builtin_abort (); + return 0; +} diff --git a/gcc/testsuite/gcc.target/aarch64/sve/peel_ind_7.c b/gcc/testsuite/gcc.target/aarch64/sve/peel_ind_7.c new file mode 100644 index 0000000..0182e13 --- /dev/null +++ b/gcc/testsuite/gcc.target/aarch64/sve/peel_ind_7.c @@ -0,0 +1,24 @@ +/* Fix for PR119351 alignment peeling with vectors and VLS. */ +/* { dg-do compile } */ +/* { dg-options "-Ofast -msve-vector-bits=256 --param aarch64-autovec-preference=sve-only -fdump-tree-vect-details" } */ + +#define N 512 +#define START 1 +#define END 505 + +int x[N] __attribute__((aligned(32))); + +int __attribute__((noipa)) +foo (void) +{ + for (unsigned int i = START; i < END; ++i) + { + if (x[i] == 0) + return i; + } + return -1; +} + +/* { dg-final { scan-tree-dump "LOOP VECTORIZED" "vect" } } */ +/* { dg-final { scan-tree-dump "pfa_iv_offset" "vect" } } */ +/* { dg-final { scan-tree-dump "Alignment of access forced using peeling" "vect" } } */ diff --git a/gcc/testsuite/gcc.target/aarch64/sve/peel_ind_7_run.c b/gcc/testsuite/gcc.target/aarch64/sve/peel_ind_7_run.c new file mode 100644 index 0000000..05608dd --- /dev/null +++ b/gcc/testsuite/gcc.target/aarch64/sve/peel_ind_7_run.c @@ -0,0 +1,17 @@ +/* Fix for PR119351 alignment peeling with vectors and VLS. */ +/* { dg-do run { target aarch64_sve_hw } } */ +/* { dg-options "-Ofast --param aarch64-autovec-preference=sve-only" } */ +/* { dg-additional-options "-msve-vector-bits=256" { target aarch64_sve256_hw } } */ +/* { dg-additional-options "-msve-vector-bits=128" { target aarch64_sve128_hw } } */ + +#include "peel_ind_7.c" + +int __attribute__ ((optimize (1))) +main (void) +{ + int res = foo (); + asm volatile (""); + if (res != START) + __builtin_abort (); + return 0; +} diff --git a/gcc/testsuite/gcc.target/aarch64/sve/peel_ind_8.c b/gcc/testsuite/gcc.target/aarch64/sve/peel_ind_8.c new file mode 100644 index 0000000..043348b --- /dev/null +++ b/gcc/testsuite/gcc.target/aarch64/sve/peel_ind_8.c @@ -0,0 +1,24 @@ +/* Fix for PR119351 alignment peeling with vectors and VLS. */ +/* { dg-do compile } */ +/* { dg-options "-Ofast -msve-vector-bits=256 --param aarch64-autovec-preference=sve-only -fdump-tree-vect-details" } */ + +#define N 512 +#define START 1 +#define END 505 + +int x[N] __attribute__((aligned(32))); + +int __attribute__((noipa)) +foo (void) +{ + for (unsigned int i = START; i < END; i*=2) + { + if (x[i] == 0) + return i; + } + return -1; +} + +/* { dg-final { scan-tree-dump-not "LOOP VECTORIZED" "vect" } } */ +/* { dg-final { scan-tree-dump-not "pfa_iv_offset" "vect" } } */ +/* { dg-final { scan-tree-dump-not "Alignment of access forced using peeling" "vect" } } */ diff --git a/gcc/testsuite/gcc.target/aarch64/sve/peel_ind_8_run.c b/gcc/testsuite/gcc.target/aarch64/sve/peel_ind_8_run.c new file mode 100644 index 0000000..aa86122 --- /dev/null +++ b/gcc/testsuite/gcc.target/aarch64/sve/peel_ind_8_run.c @@ -0,0 +1,17 @@ +/* Fix for PR119351 alignment peeling with vectors and VLS. */ +/* { dg-do run { target aarch64_sve_hw } } */ +/* { dg-options "-Ofast --param aarch64-autovec-preference=sve-only" } */ +/* { dg-additional-options "-msve-vector-bits=256" { target aarch64_sve256_hw } } */ +/* { dg-additional-options "-msve-vector-bits=128" { target aarch64_sve128_hw } } */ + +#include "peel_ind_8.c" + +int __attribute__ ((optimize (1))) +main (void) +{ + int res = foo (); + asm volatile (""); + if (res != START) + __builtin_abort (); + return 0; +} diff --git a/gcc/testsuite/gcc.target/aarch64/sve/peel_ind_9.c b/gcc/testsuite/gcc.target/aarch64/sve/peel_ind_9.c new file mode 100644 index 0000000..cc904e8 --- /dev/null +++ b/gcc/testsuite/gcc.target/aarch64/sve/peel_ind_9.c @@ -0,0 +1,25 @@ +/* Fix for PR119351 alignment peeling with vectors and VLS. */ +/* { dg-do compile } */ +/* { dg-options "-Ofast -msve-vector-bits=256 --param aarch64-autovec-preference=sve-only -fdump-tree-vect-details" } */ + +#define N 512 +#define START 1 +#define END 505 + +int x[N] __attribute__((aligned(32))); + +int __attribute__((noipa)) +foo (void) +{ + for (int *p = x + START; p < x + END; p++) + { + if (*p == 0) + return START; + } + return -1; +} + +/* { dg-final { scan-tree-dump "LOOP VECTORIZED" "vect" } } */ +/* Peels using a scalar loop. */ +/* { dg-final { scan-tree-dump-not "pfa_iv_offset" "vect" } } */ +/* { dg-final { scan-tree-dump "Alignment of access forced using peeling" "vect" } } */ diff --git a/gcc/testsuite/gcc.target/aarch64/sve/peel_ind_9_run.c b/gcc/testsuite/gcc.target/aarch64/sve/peel_ind_9_run.c new file mode 100644 index 0000000..767f8bd --- /dev/null +++ b/gcc/testsuite/gcc.target/aarch64/sve/peel_ind_9_run.c @@ -0,0 +1,17 @@ +/* Fix for PR119351 alignment peeling with vectors and VLS. */ +/* { dg-do run { target aarch64_sve_hw } } */ +/* { dg-options "-Ofast --param aarch64-autovec-preference=sve-only" } */ +/* { dg-additional-options "-msve-vector-bits=256" { target aarch64_sve256_hw } } */ +/* { dg-additional-options "-msve-vector-bits=128" { target aarch64_sve128_hw } } */ + +#include "peel_ind_9.c" + +int __attribute__ ((optimize (1))) +main (void) +{ + int res = foo (); + asm volatile (""); + if (res != START) + __builtin_abort (); + return 0; +} diff --git a/gcc/testsuite/gcc.target/aarch64/sve/pr119351.c b/gcc/testsuite/gcc.target/aarch64/sve/pr119351.c new file mode 100644 index 0000000..85aab35 --- /dev/null +++ b/gcc/testsuite/gcc.target/aarch64/sve/pr119351.c @@ -0,0 +1,39 @@ +/* Fix for PR119351 alignment peeling with vectors and VLS. */ +/* { dg-do compile } */ +/* { dg-options "-Ofast -msve-vector-bits=256 --param aarch64-autovec-preference=sve-only -fdump-tree-vect-details" } */ +/* { dg-final { check-function-bodies "**" "" ""} } */ + +#define N 512 +#define START 1 +#define END 505 + +int x[N] __attribute__((aligned(32))); + +/* +** foo: +** ... +** ld1w z[0-9]+.s, p[0-9]+/z, \[x[0-9], x[0-9], lsl 2\] +** cmple p[0-9]+.s, p[0-9]+/z, z[0-9]+.s, #0 +** ptest p[0-9]+, p[0-9]+.b +** ... +*/ + +int __attribute__((noipa)) +foo (void) +{ + int z = 0; + for (unsigned int i = START; i < END; ++i) + { + z++; + if (x[i] > 0) + continue; + + return z; + } + return -1; +} + +/* { dg-final { scan-tree-dump "LOOP VECTORIZED" "vect" } } */ +/* { dg-final { scan-tree-dump "pfa_iv_offset" "vect" } } */ +/* { dg-final { scan-tree-dump "Alignment of access forced using peeling" "vect" } } */ + diff --git a/gcc/testsuite/gcc.target/aarch64/sve/pr119351_run.c b/gcc/testsuite/gcc.target/aarch64/sve/pr119351_run.c new file mode 100644 index 0000000..d36ab0e --- /dev/null +++ b/gcc/testsuite/gcc.target/aarch64/sve/pr119351_run.c @@ -0,0 +1,20 @@ +/* Fix for PR119351 alignment peeling with vectors and VLS. */ +/* { dg-do run { target aarch64_sve_hw } } */ +/* { dg-options "-Ofast --param aarch64-autovec-preference=sve-only" } */ +/* { dg-additional-options "-msve-vector-bits=256" { target aarch64_sve256_hw } } */ +/* { dg-additional-options "-msve-vector-bits=128" { target aarch64_sve128_hw } } */ + +#include "pr119351.c" + +int __attribute__ ((optimize (1))) +main (void) +{ + x[0] = 1; + x[1] = 21; + x[2] = 39; + x[3] = 59; + int res = foo (); + if (res != 4) + __builtin_abort (); + return 0; +} diff --git a/gcc/testsuite/gcc.target/alpha/memcpy-nested-offset-long.c b/gcc/testsuite/gcc.target/alpha/memcpy-nested-offset-long.c new file mode 100644 index 0000000..631d14f --- /dev/null +++ b/gcc/testsuite/gcc.target/alpha/memcpy-nested-offset-long.c @@ -0,0 +1,76 @@ +/* { dg-do compile } */ +/* { dg-options "" } */ +/* { dg-skip-if "" { *-*-* } { "-O0" } } */ + +typedef unsigned int __attribute__ ((mode (DI))) int64_t; +typedef unsigned int __attribute__ ((mode (SI))) int32_t; + +typedef union + { + int32_t l[8]; + } +val; + +typedef struct + { + int32_t l[2]; + val v; + } +tre; + +typedef struct + { + int32_t l[3]; + tre t; + } +due; + +typedef struct + { + val v; + int64_t q; + int32_t l[2]; + due d; + } +uno; + +void +memcpy_nested_offset_long (uno *u) +{ + u->d.t.v = u->v; +} + +/* Expect assembly such as: + + ldq $4,0($16) + ldq $3,8($16) + ldq $2,16($16) + srl $4,32,$7 + ldq $1,24($16) + srl $3,32,$6 + stl $4,68($16) + srl $2,32,$5 + stl $7,72($16) + srl $1,32,$4 + stl $3,76($16) + stl $6,80($16) + stl $2,84($16) + stl $5,88($16) + stl $1,92($16) + stl $4,96($16) + + that is with four quadword loads at offsets 0, 8, 16, 24 each and + eight longword stores at offsets 68, 72, 76, 80, 84, 88, 92, 96 each. */ + +/* { dg-final { scan-assembler-times "\\sldq\\s\\\$\[0-9\]+,0\\\(\\\$16\\\)\\s" 1 } } */ +/* { dg-final { scan-assembler-times "\\sldq\\s\\\$\[0-9\]+,8\\\(\\\$16\\\)\\s" 1 } } */ +/* { dg-final { scan-assembler-times "\\sldq\\s\\\$\[0-9\]+,16\\\(\\\$16\\\)\\s" 1 } } */ +/* { dg-final { scan-assembler-times "\\sldq\\s\\\$\[0-9\]+,24\\\(\\\$16\\\)\\s" 1 } } */ +/* { dg-final { scan-assembler-times "\\sstl\\s\\\$\[0-9\]+,68\\\(\\\$16\\\)\\s" 1 } } */ +/* { dg-final { scan-assembler-times "\\sstl\\s\\\$\[0-9\]+,72\\\(\\\$16\\\)\\s" 1 } } */ +/* { dg-final { scan-assembler-times "\\sstl\\s\\\$\[0-9\]+,76\\\(\\\$16\\\)\\s" 1 } } */ +/* { dg-final { scan-assembler-times "\\sstl\\s\\\$\[0-9\]+,80\\\(\\\$16\\\)\\s" 1 } } */ +/* { dg-final { scan-assembler-times "\\sstl\\s\\\$\[0-9\]+,84\\\(\\\$16\\\)\\s" 1 } } */ +/* { dg-final { scan-assembler-times "\\sstl\\s\\\$\[0-9\]+,88\\\(\\\$16\\\)\\s" 1 } } */ +/* { dg-final { scan-assembler-times "\\sstl\\s\\\$\[0-9\]+,92\\\(\\\$16\\\)\\s" 1 } } */ +/* { dg-final { scan-assembler-times "\\sstl\\s\\\$\[0-9\]+,96\\\(\\\$16\\\)\\s" 1 } } */ diff --git a/gcc/testsuite/gcc.target/alpha/memcpy-nested-offset-quad.c b/gcc/testsuite/gcc.target/alpha/memcpy-nested-offset-quad.c new file mode 100644 index 0000000..1d2227e --- /dev/null +++ b/gcc/testsuite/gcc.target/alpha/memcpy-nested-offset-quad.c @@ -0,0 +1,64 @@ +/* { dg-do compile } */ +/* { dg-options "" } */ +/* { dg-skip-if "" { *-*-* } { "-O0" } } */ + +typedef unsigned int __attribute__ ((mode (DI))) int64_t; +typedef unsigned int __attribute__ ((mode (SI))) int32_t; + +typedef union + { + int32_t l[8]; + } +val; + +typedef struct + { + int32_t l[2]; + val v; + } +tre; + +typedef struct + { + int32_t l[3]; + tre t; + } +due; + +typedef struct + { + val v; + int64_t q; + int32_t l[3]; + due d; + } +uno; + +void +memcpy_nested_offset_quad (uno *u) +{ + u->d.t.v = u->v; +} + +/* Expect assembly such as: + + ldq $4,0($16) + ldq $3,8($16) + ldq $2,16($16) + ldq $1,24($16) + stq $4,72($16) + stq $3,80($16) + stq $2,88($16) + stq $1,96($16) + + that is with four quadword loads at offsets 0, 8, 16, 24 each + and four quadword stores at offsets 72, 80, 88, 96 each. */ + +/* { dg-final { scan-assembler-times "\\sldq\\s\\\$\[0-9\]+,0\\\(\\\$16\\\)\\s" 1 } } */ +/* { dg-final { scan-assembler-times "\\sldq\\s\\\$\[0-9\]+,8\\\(\\\$16\\\)\\s" 1 } } */ +/* { dg-final { scan-assembler-times "\\sldq\\s\\\$\[0-9\]+,16\\\(\\\$16\\\)\\s" 1 } } */ +/* { dg-final { scan-assembler-times "\\sldq\\s\\\$\[0-9\]+,24\\\(\\\$16\\\)\\s" 1 } } */ +/* { dg-final { scan-assembler-times "\\sstq\\s\\\$\[0-9\]+,72\\\(\\\$16\\\)\\s" 1 } } */ +/* { dg-final { scan-assembler-times "\\sstq\\s\\\$\[0-9\]+,80\\\(\\\$16\\\)\\s" 1 } } */ +/* { dg-final { scan-assembler-times "\\sstq\\s\\\$\[0-9\]+,88\\\(\\\$16\\\)\\s" 1 } } */ +/* { dg-final { scan-assembler-times "\\sstq\\s\\\$\[0-9\]+,96\\\(\\\$16\\\)\\s" 1 } } */ diff --git a/gcc/testsuite/gcc.target/arm/ivopts.c b/gcc/testsuite/gcc.target/arm/ivopts.c index d7d72a5..582fdab 100644 --- a/gcc/testsuite/gcc.target/arm/ivopts.c +++ b/gcc/testsuite/gcc.target/arm/ivopts.c @@ -11,6 +11,6 @@ tr5 (short array[], int n) } /* { dg-final { scan-tree-dump-times "PHI <" 1 "ivopts"} } */ -/* { dg-final { object-size text <= 20 { target { arm_thumb2_no_arm_v8_1_lob } } } } */ +/* { dg-final { object-size text <= 20 { target { arm_thumb2_no_arm_v8_1m_lob } } } } */ /* { dg-final { object-size text <= 32 { target { arm_nothumb && { ! arm_iwmmxt_ok } } } } } */ /* { dg-final { object-size text <= 36 { target { arm_nothumb && arm_iwmmxt_ok } } } } */ diff --git a/gcc/testsuite/gcc.target/arm/lob1.c b/gcc/testsuite/gcc.target/arm/lob1.c index c8ce653..f42a367 100644 --- a/gcc/testsuite/gcc.target/arm/lob1.c +++ b/gcc/testsuite/gcc.target/arm/lob1.c @@ -1,7 +1,7 @@ /* Check that GCC generates Armv8.1-M low over head loop instructions for some simple loops. */ /* { dg-do run } */ -/* { dg-require-effective-target arm_v8_1_lob_ok } */ +/* { dg-require-effective-target arm_v8_1m_lob_hw } */ /* { dg-skip-if "avoid conflicting multilib options" { *-*-* } { "-marm" "-mcpu=*" } } */ /* { dg-options "-march=armv8.1-m.main -mthumb -O3 --save-temps" } */ #include <stdlib.h> diff --git a/gcc/testsuite/gcc.target/arm/lob6.c b/gcc/testsuite/gcc.target/arm/lob6.c index 4fe116e..e19635b 100644 --- a/gcc/testsuite/gcc.target/arm/lob6.c +++ b/gcc/testsuite/gcc.target/arm/lob6.c @@ -1,7 +1,7 @@ /* Check that GCC generates Armv8.1-M low over head loop instructions with some less trivial loops and the result is correct. */ /* { dg-do run } */ -/* { dg-require-effective-target arm_v8_1_lob_ok } */ +/* { dg-require-effective-target arm_v8_1m_lob_hw } */ /* { dg-skip-if "avoid conflicting multilib options" { *-*-* } { "-marm" "-mcpu=*" } } */ /* { dg-options "-march=armv8.1-m.main -mthumb -O3 --save-temps" } */ #include <stdlib.h> diff --git a/gcc/testsuite/gcc.target/arm/unsigned-extend-2.c b/gcc/testsuite/gcc.target/arm/unsigned-extend-2.c index 9272e4c..41ee994 100644 --- a/gcc/testsuite/gcc.target/arm/unsigned-extend-2.c +++ b/gcc/testsuite/gcc.target/arm/unsigned-extend-2.c @@ -1,5 +1,5 @@ /* { dg-do compile } */ -/* { dg-require-effective-target arm_thumb2_ok_no_arm_v8_1_lob } */ +/* { dg-require-effective-target arm_thumb2_ok_no_arm_v8_1m_lob } */ /* { dg-options "-O" } */ unsigned short foo (unsigned short x, unsigned short c) diff --git a/gcc/testsuite/gcc.target/i386/apx-interrupt-1.c b/gcc/testsuite/gcc.target/i386/apx-interrupt-1.c index fefe2e6..fa1acc7 100644 --- a/gcc/testsuite/gcc.target/i386/apx-interrupt-1.c +++ b/gcc/testsuite/gcc.target/i386/apx-interrupt-1.c @@ -66,7 +66,7 @@ void foo (void *frame) /* { dg-final { scan-assembler-times {\t\.cfi_offset 132, -120} 1 } } */ /* { dg-final { scan-assembler-times {\t\.cfi_offset 131, -128} 1 } } */ /* { dg-final { scan-assembler-times {\t\.cfi_offset 130, -136} 1 } } */ -/* { dg-final { scan-assembler-times ".cfi_restore" 15} } */ +/* { dg-final { scan-assembler-times ".cfi_restore" 31 } } */ /* { dg-final { scan-assembler-times "pop(?:l|q)\[\\t \]*%(?:e|r)ax" 1 } } */ /* { dg-final { scan-assembler-times "pop(?:l|q)\[\\t \]*%(?:e|r)bx" 1 } } */ /* { dg-final { scan-assembler-times "pop(?:l|q)\[\\t \]*%(?:e|r)cx" 1 } } */ diff --git a/gcc/testsuite/gcc.target/i386/pr119386-1.c b/gcc/testsuite/gcc.target/i386/pr119386-1.c new file mode 100644 index 0000000..9a0dc64 --- /dev/null +++ b/gcc/testsuite/gcc.target/i386/pr119386-1.c @@ -0,0 +1,10 @@ +/* PR target/119386 */ +/* { dg-do compile { target *-*-linux* } } */ +/* { dg-options "-O2 -fpic -pg" } */ +/* { dg-final { scan-assembler "call\[ \t\]+mcount@PLT" } } */ + +int +main () +{ + return 0; +} diff --git a/gcc/testsuite/gcc.target/i386/pr119386-2.c b/gcc/testsuite/gcc.target/i386/pr119386-2.c new file mode 100644 index 0000000..3ea978e --- /dev/null +++ b/gcc/testsuite/gcc.target/i386/pr119386-2.c @@ -0,0 +1,12 @@ +/* PR target/119386 */ +/* { dg-do compile { target *-*-linux* } } */ +/* { dg-options "-O2 -fpic -fno-plt -pg" } */ +/* { dg-final { scan-assembler "call\[ \t\]+\\*mcount@GOTPCREL\\(" { target { ! ia32 } } } } */ +/* { dg-final { scan-assembler "call\[ \t\]+\\*mcount@GOT\\(" { target ia32 } } } */ + + +int +main () +{ + return 0; +} diff --git a/gcc/testsuite/gcc.target/i386/pr119386-3.c b/gcc/testsuite/gcc.target/i386/pr119386-3.c new file mode 100644 index 0000000..287410b --- /dev/null +++ b/gcc/testsuite/gcc.target/i386/pr119386-3.c @@ -0,0 +1,10 @@ +/* PR target/119386 */ +/* { dg-do compile { target *-*-linux* } } */ +/* { dg-options "-O2 -fpic -pg -mnop-mcount" } */ +/* { dg-final { scan-assembler ".byte\[ \t\]+0x0f, 0x1f, 0x44, 0x00, 0x00" } } */ + +int +main () +{ + return 0; +} diff --git a/gcc/testsuite/gcc.target/i386/pr119784a.c b/gcc/testsuite/gcc.target/i386/pr119784a.c new file mode 100644 index 0000000..8a119d4 --- /dev/null +++ b/gcc/testsuite/gcc.target/i386/pr119784a.c @@ -0,0 +1,96 @@ +/* { dg-do compile { target { *-*-linux* && lp64 } } } */ +/* { dg-options "-O2 -fno-pic -mtune=generic -mgeneral-regs-only -mapxf -mtune-ctrl=prologue_using_move,epilogue_using_move" } */ +/* Keep labels and directives ('.cfi_startproc', '.cfi_endproc'). */ +/* { dg-final { check-function-bodies "**" "" "" { target "*-*-*" } {^\t?\.} } } */ + +/* start must save and restore all caller saved registers. */ + +/* +**start: +**.LFB[0-9]+: +** .cfi_startproc +** subq \$248, %rsp +**... +** movq %rax, \(%rsp\) +** movq %rdx, 8\(%rsp\) +** movq %rcx, 16\(%rsp\) +** movq %rbx, 24\(%rsp\) +** movq %rsi, 32\(%rsp\) +** movq %rdi, 40\(%rsp\) +**... +** movq %rbp, 48\(%rsp\) +** movq %r8, 56\(%rsp\) +** movq %r9, 64\(%rsp\) +** movq %r10, 72\(%rsp\) +** movq %r11, 80\(%rsp\) +** movq %r12, 88\(%rsp\) +** movq %r13, 96\(%rsp\) +** movq %r14, 104\(%rsp\) +** movq %r15, 112\(%rsp\) +** movq %r16, 120\(%rsp\) +** movq %r17, 128\(%rsp\) +** movq %r18, 136\(%rsp\) +** movq %r19, 144\(%rsp\) +** movq %r20, 152\(%rsp\) +** movq %r21, 160\(%rsp\) +** movq %r22, 168\(%rsp\) +** movq %r23, 176\(%rsp\) +** movq %r24, 184\(%rsp\) +** movq %r25, 192\(%rsp\) +** movq %r26, 200\(%rsp\) +** movq %r27, 208\(%rsp\) +** movq %r28, 216\(%rsp\) +** movq %r29, 224\(%rsp\) +** movq %r30, 232\(%rsp\) +** movq %r31, 240\(%rsp\) +**... +** call \*code\(%rip\) +** movq \(%rsp\), %rax +** movq 8\(%rsp\), %rdx +** movq 16\(%rsp\), %rcx +** movq 24\(%rsp\), %rbx +** movq 32\(%rsp\), %rsi +** movq 40\(%rsp\), %rdi +** movq 48\(%rsp\), %rbp +** movq 56\(%rsp\), %r8 +** movq 64\(%rsp\), %r9 +** movq 72\(%rsp\), %r10 +** movq 80\(%rsp\), %r11 +** movq 88\(%rsp\), %r12 +** movq 96\(%rsp\), %r13 +** movq 104\(%rsp\), %r14 +** movq 112\(%rsp\), %r15 +** movq 120\(%rsp\), %r16 +** movq 128\(%rsp\), %r17 +** movq 136\(%rsp\), %r18 +** movq 144\(%rsp\), %r19 +** movq 152\(%rsp\), %r20 +** movq 160\(%rsp\), %r21 +** movq 168\(%rsp\), %r22 +** movq 176\(%rsp\), %r23 +** movq 184\(%rsp\), %r24 +** movq 192\(%rsp\), %r25 +** movq 200\(%rsp\), %r26 +** movq 208\(%rsp\), %r27 +** movq 216\(%rsp\), %r28 +** movq 224\(%rsp\), %r29 +** movq 232\(%rsp\), %r30 +** movq 240\(%rsp\), %r31 +** addq \$248, %rsp +**... +** ret +** .cfi_endproc +**... +*/ + +#define DONT_SAVE_REGS __attribute__((no_callee_saved_registers)) +#define SAVE_REGS __attribute__((no_caller_saved_registers)) + +typedef DONT_SAVE_REGS void (*op_t)(void); + +extern op_t code[]; + +SAVE_REGS void start() +{ + code[0](); +} diff --git a/gcc/testsuite/gcc.target/i386/pr119784b.c b/gcc/testsuite/gcc.target/i386/pr119784b.c new file mode 100644 index 0000000..c676197 --- /dev/null +++ b/gcc/testsuite/gcc.target/i386/pr119784b.c @@ -0,0 +1,87 @@ +/* { dg-do compile { target { *-*-linux* && x32 } } } */ +/* { dg-options "-O2 -fno-pic -mtune=generic -mgeneral-regs-only -mapxf -mtune-ctrl=prologue_using_move,epilogue_using_move" } */ +/* Keep labels and directives ('.cfi_startproc', '.cfi_endproc'). */ +/* { dg-final { check-function-bodies "**" "" "" { target "*-*-*" } {^\t?\.} } } */ + +/* start must save and restore all caller saved registers. */ + +/* +**start: +**.LFB[0-9]+: +** .cfi_startproc +** subl \$248, %esp +**... +** movq %rax, \(%rsp\) +** movq %rdx, 8\(%rsp\) +** movq %rcx, 16\(%rsp\) +** movq %rbx, 24\(%rsp\) +** movq %rsi, 32\(%rsp\) +** movq %rdi, 40\(%rsp\) +**... +** movq %rbp, 48\(%rsp\) +** movq %r8, 56\(%rsp\) +** movq %r9, 64\(%rsp\) +** movq %r10, 72\(%rsp\) +** movq %r11, 80\(%rsp\) +** movq %r12, 88\(%rsp\) +** movq %r13, 96\(%rsp\) +** movq %r14, 104\(%rsp\) +** movq %r15, 112\(%rsp\) +** movq %r16, 120\(%rsp\) +** movq %r17, 128\(%rsp\) +** movq %r18, 136\(%rsp\) +** movq %r19, 144\(%rsp\) +** movq %r20, 152\(%rsp\) +** movq %r21, 160\(%rsp\) +** movq %r22, 168\(%rsp\) +** movq %r23, 176\(%rsp\) +** movq %r24, 184\(%rsp\) +** movq %r25, 192\(%rsp\) +** movq %r26, 200\(%rsp\) +** movq %r27, 208\(%rsp\) +** movq %r28, 216\(%rsp\) +** movq %r29, 224\(%rsp\) +** movq %r30, 232\(%rsp\) +** movq %r31, 240\(%rsp\) +**... +** movl code\(%rip\), %ebp +** call \*%rbp +** movq \(%rsp\), %rax +** movq 8\(%rsp\), %rdx +** movq 16\(%rsp\), %rcx +** movq 24\(%rsp\), %rbx +** movq 32\(%rsp\), %rsi +** movq 40\(%rsp\), %rdi +** movq 48\(%rsp\), %rbp +** movq 56\(%rsp\), %r8 +** movq 64\(%rsp\), %r9 +** movq 72\(%rsp\), %r10 +** movq 80\(%rsp\), %r11 +** movq 88\(%rsp\), %r12 +** movq 96\(%rsp\), %r13 +** movq 104\(%rsp\), %r14 +** movq 112\(%rsp\), %r15 +** movq 120\(%rsp\), %r16 +** movq 128\(%rsp\), %r17 +** movq 136\(%rsp\), %r18 +** movq 144\(%rsp\), %r19 +** movq 152\(%rsp\), %r20 +** movq 160\(%rsp\), %r21 +** movq 168\(%rsp\), %r22 +** movq 176\(%rsp\), %r23 +** movq 184\(%rsp\), %r24 +** movq 192\(%rsp\), %r25 +** movq 200\(%rsp\), %r26 +** movq 208\(%rsp\), %r27 +** movq 216\(%rsp\), %r28 +** movq 224\(%rsp\), %r29 +** movq 232\(%rsp\), %r30 +** movq 240\(%rsp\), %r31 +** addl \$248, %esp +**... +** ret +** .cfi_endproc +**... +*/ + +#include "pr119784a.c" diff --git a/gcc/testsuite/gcc.target/i386/recip-vec-divf-fma.c b/gcc/testsuite/gcc.target/i386/recip-vec-divf-fma.c new file mode 100644 index 0000000..ad9e07b --- /dev/null +++ b/gcc/testsuite/gcc.target/i386/recip-vec-divf-fma.c @@ -0,0 +1,12 @@ +/* { dg-do compile } */ +/* { dg-options "-Ofast -mfma -mavx2" } */ +/* { dg-final { scan-assembler-times {(?n)vfn?m(add|sub)[1-3]*ps} 2 } } */ + +typedef float v4sf __attribute__((vector_size(16))); +/* (a - (rcp(b) * a * b)) * rcp(b) + rcp(b) * a */ + +v4sf +foo (v4sf a, v4sf b) +{ + return a / b; +} diff --git a/gcc/testsuite/gcc.target/loongarch/vector/loongarch-vector.exp b/gcc/testsuite/gcc.target/loongarch/vector/loongarch-vector.exp index f56d2f1..9df3f29 100644 --- a/gcc/testsuite/gcc.target/loongarch/vector/loongarch-vector.exp +++ b/gcc/testsuite/gcc.target/loongarch/vector/loongarch-vector.exp @@ -35,7 +35,7 @@ dg-init # If the target hardware supports LSX, the default action is "run", otherwise # just "compile". -global dg-do-what-default +set saved-dg-do-what-default ${dg-do-what-default} if {[check_effective_target_loongarch_sx_hw]} then { set dg-do-what-default run } else { @@ -45,6 +45,7 @@ if {[check_effective_target_loongarch_sx_hw]} then { #Main loop. dg-runtest [lsort [glob -nocomplain $srcdir/$subdir/lsx/*.\[cS\]]] \ " -mlsx" $DEFAULT_CFLAGS +set dg-do-what-default ${saved-dg-do-what-default} dg-finish @@ -52,7 +53,7 @@ dg-init # If the target hardware supports LASX, the default action is "run", otherwise # just "compile". -global dg-do-what-default +set saved-dg-do-what-default ${dg-do-what-default} if {[check_effective_target_loongarch_asx_hw]} then { set dg-do-what-default run } else { @@ -61,5 +62,6 @@ if {[check_effective_target_loongarch_asx_hw]} then { dg-runtest [lsort [glob -nocomplain $srcdir/$subdir/lasx/*.\[cS\]]] \ " -mlasx" $DEFAULT_CFLAGS +set dg-do-what-default ${saved-dg-do-what-default} # All done. dg-finish diff --git a/gcc/testsuite/gcc.target/mips/clear-cache-1.c b/gcc/testsuite/gcc.target/mips/clear-cache-1.c index f1554f5..cd11c66 100644 --- a/gcc/testsuite/gcc.target/mips/clear-cache-1.c +++ b/gcc/testsuite/gcc.target/mips/clear-cache-1.c @@ -1,7 +1,7 @@ /* { dg-do compile } */ /* { dg-options "-msynci isa_rev>=2" } */ /* { dg-final { scan-assembler "\tsynci\t" } } */ -/* { dg-final { scan-assembler "\tjr.hb\t" } } */ +/* { dg-final { scan-assembler "\tjrc?.hb\t" } } */ /* { dg-final { scan-assembler-not "_flush_cache|mips_sync_icache|_cacheflush" } } */ NOMIPS16 void f() diff --git a/gcc/testsuite/gcc.target/mips/memcpy-2.c b/gcc/testsuite/gcc.target/mips/memcpy-2.c new file mode 100644 index 0000000..df0cd18 --- /dev/null +++ b/gcc/testsuite/gcc.target/mips/memcpy-2.c @@ -0,0 +1,12 @@ +/* { dg-do compile } */ +/* { dg-options "isa_rev<=5 -fdump-rtl-expand" } */ +/* { dg-skip-if "code quality test" { *-*-* } { "-Os" } { "" } } */ + +__attribute__((nomips16)) +void +f1 (char *p) +{ + __builtin_memcpy (p, "12345", 5); +} + +/* { dg-final { scan-rtl-dump "mem/u.*mem/u" "expand" } } */ diff --git a/gcc/testsuite/gcc.target/riscv/bext-ext-2.c b/gcc/testsuite/gcc.target/riscv/bext-ext-2.c new file mode 100644 index 0000000..aa170d0 --- /dev/null +++ b/gcc/testsuite/gcc.target/riscv/bext-ext-2.c @@ -0,0 +1,74 @@ +/* { dg-do compile } */ +/* { dg-options "-march=rv64gcb -mabi=lp64" } */ +/* { dg-skip-if "" { *-*-* } { "-O0" } } */ + +struct obstack; +struct bitmap_head_def; +typedef struct bitmap_head_def *bitmap; +struct obstack +{ + long chunk_size; + struct _obstack_chunk *chunk; + char *object_base; + char *next_free; + char *chunk_limit; + long int temp; + int alignment_mask; + + + + struct _obstack_chunk *(*chunkfun) (void *, long); + void (*freefun) (void *, struct _obstack_chunk *); + void *extra_arg; + unsigned use_extra_arg:1; + unsigned maybe_empty_object:1; + + + + unsigned alloc_failed:1; + + +}; + +typedef unsigned long BITMAP_WORD; +typedef struct bitmap_obstack { + struct bitmap_element_def *elements; + struct bitmap_head_def *heads; + struct obstack obstack; +} bitmap_obstack; +typedef struct bitmap_element_def { + struct bitmap_element_def *next; + struct bitmap_element_def *prev; + unsigned int indx; + BITMAP_WORD bits[((128 + (8 + * 8 * 1u) - 1) / (8 + * 8 * 1u))]; +} bitmap_element; +bitmap_element *bitmap_find_bit (bitmap, unsigned int); + + +int +bitmap_bit_p (bitmap head, int bit) +{ + bitmap_element *ptr; + unsigned bit_num; + unsigned word_num; + + ptr = bitmap_find_bit (head, bit); + if (ptr == 0) + return 0; + + bit_num = bit % (8 + * 8 * 1u); + word_num = bit / (8 + * 8 * 1u) % ((128 + (8 + * 8 * 1u) - 1) / (8 + * 8 * 1u)); + + return (ptr->bits[word_num] >> bit_num) & 1; +} + +/* { dg-final { scan-assembler-times "bext\t" 1 } } */ +/* { dg-final { scan-assembler-not "slr\t"} } */ +/* { dg-final { scan-assembler-not "andi\t"} } */ + diff --git a/gcc/testsuite/gcc.target/riscv/gnu-property-align-rv32.c b/gcc/testsuite/gcc.target/riscv/gnu-property-align-rv32.c new file mode 100644 index 0000000..4f48cff --- /dev/null +++ b/gcc/testsuite/gcc.target/riscv/gnu-property-align-rv32.c @@ -0,0 +1,7 @@ +/* { dg-do compile } */ +/* { dg-options "-march=rv32g_zicfiss -fcf-protection=return -mabi=ilp32d " } */ + +void foo() {} + +/* { dg-final { scan-assembler-times ".p2align\t2" 3 } } */ +/* { dg-final { scan-assembler-not ".p2align\t3" } } */ diff --git a/gcc/testsuite/gcc.target/riscv/gnu-property-align-rv64.c b/gcc/testsuite/gcc.target/riscv/gnu-property-align-rv64.c new file mode 100644 index 0000000..1bfd127 --- /dev/null +++ b/gcc/testsuite/gcc.target/riscv/gnu-property-align-rv64.c @@ -0,0 +1,7 @@ +/* { dg-do compile } */ +/* { dg-options "-march=rv64g_zicfiss -fcf-protection=return -mabi=lp64d " } */ + +void foo() {} + +/* { dg-final { scan-assembler-times ".p2align\t3" 3 } } */ +/* { dg-final { scan-assembler-not ".p2align\t2" } } */ diff --git a/gcc/testsuite/gcc.target/riscv/jump-table-large-code-model.c b/gcc/testsuite/gcc.target/riscv/jump-table-large-code-model.c new file mode 100644 index 0000000..1ee7f6c --- /dev/null +++ b/gcc/testsuite/gcc.target/riscv/jump-table-large-code-model.c @@ -0,0 +1,24 @@ +/* { dg-do compile } */ +/* { dg-options "-march=rv64gc -mabi=lp64 -mcmodel=large" } */ + +int foo(int x, int y) +{ + switch(x){ + case 0: + return 123 + y; + case 1: + return 456 + y; + case 2: + return 789 - y; + case 3: + return 12 * y; + case 4: + return 13 % y; + case 5: + return 11 *y; + } + return 0; +} + + +/* { dg-final { scan-assembler-not "\.section \.rodata" } } */ diff --git a/gcc/testsuite/gcc.target/riscv/pr108016.c b/gcc/testsuite/gcc.target/riscv/pr108016.c new file mode 100644 index 0000000..b60df42 --- /dev/null +++ b/gcc/testsuite/gcc.target/riscv/pr108016.c @@ -0,0 +1,33 @@ +/* { dg-do compile } */ +/* { dg-options "-march=rv64gc -mabi=lp64" } */ +/* { dg-skip-if "" { *-*-* } { "-O0" } } */ + +unsigned int addu (unsigned int a, unsigned int b) +{ + unsigned int out; + unsigned int overflow = __builtin_add_overflow (a, b, &out); + return overflow & out; +} + +int addi (int a, int b) +{ + int out; + int overflow = __builtin_add_overflow (a, b, &out); + return overflow & out; +} + +unsigned int subu (unsigned int a, unsigned int b) +{ + unsigned int out; + unsigned int overflow = __builtin_sub_overflow (a, b, &out); + return overflow & out; +} + +int subi (int a, int b) +{ + int out; + int overflow = __builtin_sub_overflow (a, b, &out); + return overflow & out; +} + +/* { dg-final { scan-assembler-not "sext\.w\t" } } */ diff --git a/gcc/testsuite/gcc.target/riscv/pr118410-1.c b/gcc/testsuite/gcc.target/riscv/pr118410-1.c new file mode 100644 index 0000000..4a8b847 --- /dev/null +++ b/gcc/testsuite/gcc.target/riscv/pr118410-1.c @@ -0,0 +1,9 @@ +/* { dg-do compile } */ +/* { dg-skip-if "" { *-*-* } { "-O0" "-Og" } } */ +/* { dg-options "-march=rv64gcb -mabi=lp64d" { target { rv64} } } */ +/* { dg-options "-march=rv32gcb -mabi=ilp32" { target { rv32} } } */ + +long orlow(long x) { return x | ((1L << 24) - 1); } + +/* { dg-final { scan-assembler-times "orn\t" 1 } } */ +/* { dg-final { scan-assembler-not "addi\t" } } */ diff --git a/gcc/testsuite/gcc.target/riscv/pr118410-2.c b/gcc/testsuite/gcc.target/riscv/pr118410-2.c new file mode 100644 index 0000000..b63a1d9 --- /dev/null +++ b/gcc/testsuite/gcc.target/riscv/pr118410-2.c @@ -0,0 +1,9 @@ +/* { dg-do compile } */ +/* { dg-skip-if "" { *-*-* } { "-O0" "-Og" } } */ +/* { dg-options "-march=rv64gcb -mabi=lp64d" { target { rv64} } } */ +/* { dg-options "-march=rv32gcb -mabi=ilp32" { target { rv32} } } */ + +long xorlow(long x) { return x ^ ((1L << 24) - 1); } + +/* { dg-final { scan-assembler-times "xnor\t" 1 } } */ +/* { dg-final { scan-assembler-not "addi\t" } } */ diff --git a/gcc/testsuite/gcc.target/riscv/rvv/base/abi-callee-saved-1-fixed-1.c b/gcc/testsuite/gcc.target/riscv/rvv/base/abi-callee-saved-1-fixed-1.c index 638e90f..69a94d5 100644 --- a/gcc/testsuite/gcc.target/riscv/rvv/base/abi-callee-saved-1-fixed-1.c +++ b/gcc/testsuite/gcc.target/riscv/rvv/base/abi-callee-saved-1-fixed-1.c @@ -2,7 +2,7 @@ /* { dg-options "-O1 -march=rv64gczve32x -mabi=lp64d -mrvv-vector-bits=zvl" } */ /* { dg-final { check-function-bodies "**" "" } } */ -#include <riscv_vector.h> +#include "riscv_vector.h" void bar (int8_t *data); diff --git a/gcc/testsuite/gcc.target/riscv/rvv/base/abi-callee-saved-1-fixed-2.c b/gcc/testsuite/gcc.target/riscv/rvv/base/abi-callee-saved-1-fixed-2.c index 380d0c1..5e0f136 100644 --- a/gcc/testsuite/gcc.target/riscv/rvv/base/abi-callee-saved-1-fixed-2.c +++ b/gcc/testsuite/gcc.target/riscv/rvv/base/abi-callee-saved-1-fixed-2.c @@ -2,7 +2,7 @@ /* { dg-options "-O1 -march=rv64gcv_zvl4096b -mabi=lp64d -mrvv-vector-bits=zvl" } */ /* { dg-final { check-function-bodies "**" "" } } */ -#include <riscv_vector.h> +#include "riscv_vector.h" void bar (int8_t *data); diff --git a/gcc/testsuite/gcc.target/riscv/rvv/base/abi-callee-saved-1-save-restore.c b/gcc/testsuite/gcc.target/riscv/rvv/base/abi-callee-saved-1-save-restore.c index 9ed72a6..a3c0a6d 100644 --- a/gcc/testsuite/gcc.target/riscv/rvv/base/abi-callee-saved-1-save-restore.c +++ b/gcc/testsuite/gcc.target/riscv/rvv/base/abi-callee-saved-1-save-restore.c @@ -2,7 +2,7 @@ /* { dg-options "-O1 -march=rv64gcv_zfh -mabi=lp64d -msave-restore" } */ /* { dg-final { check-function-bodies "**" "" } } */ -#include <riscv_vector.h> +#include "riscv_vector.h" void bar (int8_t *data); diff --git a/gcc/testsuite/gcc.target/riscv/rvv/base/abi-callee-saved-1-zcmp.c b/gcc/testsuite/gcc.target/riscv/rvv/base/abi-callee-saved-1-zcmp.c index b6b708f..b1cf6aa 100644 --- a/gcc/testsuite/gcc.target/riscv/rvv/base/abi-callee-saved-1-zcmp.c +++ b/gcc/testsuite/gcc.target/riscv/rvv/base/abi-callee-saved-1-zcmp.c @@ -2,7 +2,7 @@ /* { dg-options "-O1 -march=rv64gv_zfh_zca_zcmp -mabi=lp64d -fno-shrink-wrap-separate" } */ /* { dg-final { check-function-bodies "**" "" } } */ -#include <riscv_vector.h> +#include "riscv_vector.h" void bar (int8_t *data); diff --git a/gcc/testsuite/gcc.target/riscv/rvv/base/abi-callee-saved-1.c b/gcc/testsuite/gcc.target/riscv/rvv/base/abi-callee-saved-1.c index 13e3328..8838f0d 100644 --- a/gcc/testsuite/gcc.target/riscv/rvv/base/abi-callee-saved-1.c +++ b/gcc/testsuite/gcc.target/riscv/rvv/base/abi-callee-saved-1.c @@ -2,7 +2,7 @@ /* { dg-options "-O1 -march=rv64gcv_zfh -mabi=lp64d" } */ /* { dg-final { check-function-bodies "**" "" } } */ -#include <riscv_vector.h> +#include "riscv_vector.h" void bar (int8_t *data); diff --git a/gcc/testsuite/gcc.target/riscv/rvv/base/abi-callee-saved-2-save-restore.c b/gcc/testsuite/gcc.target/riscv/rvv/base/abi-callee-saved-2-save-restore.c index d21b810..77f1c7c 100644 --- a/gcc/testsuite/gcc.target/riscv/rvv/base/abi-callee-saved-2-save-restore.c +++ b/gcc/testsuite/gcc.target/riscv/rvv/base/abi-callee-saved-2-save-restore.c @@ -2,7 +2,7 @@ /* { dg-options "-O1 -march=rv64gcv_zfh -mabi=lp64d -msave-restore" } */ /* { dg-final { check-function-bodies "**" "" } } */ -#include <riscv_vector.h> +#include "riscv_vector.h" void bar1 (vint8m1_t a); void bar2 (); diff --git a/gcc/testsuite/gcc.target/riscv/rvv/base/abi-callee-saved-2-zcmp.c b/gcc/testsuite/gcc.target/riscv/rvv/base/abi-callee-saved-2-zcmp.c index 70a32d7..37127a8 100644 --- a/gcc/testsuite/gcc.target/riscv/rvv/base/abi-callee-saved-2-zcmp.c +++ b/gcc/testsuite/gcc.target/riscv/rvv/base/abi-callee-saved-2-zcmp.c @@ -2,7 +2,7 @@ /* { dg-options "-O1 -march=rv64gv_zfh_zca_zcmp -mabi=lp64d -fno-shrink-wrap-separate" } */ /* { dg-final { check-function-bodies "**" "" } } */ -#include <riscv_vector.h> +#include "riscv_vector.h" void bar1 (vint8m1_t a); void bar2 (); diff --git a/gcc/testsuite/gcc.target/riscv/rvv/base/abi-callee-saved-2.c b/gcc/testsuite/gcc.target/riscv/rvv/base/abi-callee-saved-2.c index 3f2cb2f..a8daeeb 100644 --- a/gcc/testsuite/gcc.target/riscv/rvv/base/abi-callee-saved-2.c +++ b/gcc/testsuite/gcc.target/riscv/rvv/base/abi-callee-saved-2.c @@ -2,7 +2,7 @@ /* { dg-options "-O1 -march=rv64gcv_zfh -mabi=lp64d" } */ /* { dg-final { check-function-bodies "**" "" } } */ -#include <riscv_vector.h> +#include "riscv_vector.h" void bar1 (vint8m1_t a); void bar2 (); diff --git a/gcc/testsuite/gcc.target/riscv/rvv/base/bug-10-2.c b/gcc/testsuite/gcc.target/riscv/rvv/base/bug-10-2.c index fe3a1ef..f8143b9 100644 --- a/gcc/testsuite/gcc.target/riscv/rvv/base/bug-10-2.c +++ b/gcc/testsuite/gcc.target/riscv/rvv/base/bug-10-2.c @@ -4,7 +4,7 @@ /* { dg-require-effective-target riscv_zvfh_ok } */ /* { dg-options " -march=rv64gcv_zvfh -mabi=lp64d -O2" } */ -#include <riscv_vector.h> +#include "riscv_vector.h" int8_t a[1]; uint16_t b[1]; diff --git a/gcc/testsuite/gcc.target/riscv/rvv/base/bug-10.c b/gcc/testsuite/gcc.target/riscv/rvv/base/bug-10.c index 60fdfc4..05628d5 100644 --- a/gcc/testsuite/gcc.target/riscv/rvv/base/bug-10.c +++ b/gcc/testsuite/gcc.target/riscv/rvv/base/bug-10.c @@ -4,7 +4,7 @@ /* { dg-require-effective-target riscv_zvfh_ok } */ /* { dg-options " -march=rv64gcv_zvfh -mabi=lp64d -O2 --param=vsetvl-strategy=optim -fno-schedule-insns -fno-schedule-insns2 -fno-schedule-fusion " } */ -#include <riscv_vector.h> +#include "riscv_vector.h" void __attribute__ ((noipa)) diff --git a/gcc/testsuite/gcc.target/riscv/rvv/base/bug-7.c b/gcc/testsuite/gcc.target/riscv/rvv/base/bug-7.c index 28766ce..3180f71 100644 --- a/gcc/testsuite/gcc.target/riscv/rvv/base/bug-7.c +++ b/gcc/testsuite/gcc.target/riscv/rvv/base/bug-7.c @@ -3,7 +3,7 @@ /* { dg-options "-march=rv64gcv -mabi=lp64d -O2" { target { rv64 } } } */ /* { dg-options "-march=rv32gcv -mabi=ilp32d -O2" { target { rv32 } } } */ -#include <riscv_vector.h> +#include "riscv_vector.h" vint64m1_t f1 (vint64m1_t vd, vint64m1_t vs2, size_t vl) { diff --git a/gcc/testsuite/gcc.target/riscv/rvv/base/bug-8.c b/gcc/testsuite/gcc.target/riscv/rvv/base/bug-8.c index 975f755..31b68c4 100644 --- a/gcc/testsuite/gcc.target/riscv/rvv/base/bug-8.c +++ b/gcc/testsuite/gcc.target/riscv/rvv/base/bug-8.c @@ -3,7 +3,7 @@ /* { dg-options "-march=rv64gcv -mabi=lp64d -O0" { target { rv64 } } } */ /* { dg-options "-march=rv32gcv -mabi=ilp32d -O0" { target { rv32 } } } */ -#include <riscv_vector.h> +#include "riscv_vector.h" vint64m1_t f1 (vint64m1_t vd, vint64m1_t vs2, size_t vl) { diff --git a/gcc/testsuite/gcc.target/riscv/rvv/base/bug-9.c b/gcc/testsuite/gcc.target/riscv/rvv/base/bug-9.c index 8cfe965..f7c9ad1 100644 --- a/gcc/testsuite/gcc.target/riscv/rvv/base/bug-9.c +++ b/gcc/testsuite/gcc.target/riscv/rvv/base/bug-9.c @@ -3,7 +3,7 @@ /* { dg-options "-march=rv64gcv -mabi=lp64d -O2" { target { rv64 } } } */ /* { dg-options "-march=rv32gcv -mabi=ilp32d -O2" { target { rv32 } } } */ -#include <riscv_vector.h> +#include "riscv_vector.h" vfloat16m1_t f0 (vfloat16m1_t vs2, vfloat16m1_t vs1, size_t vl) { diff --git a/gcc/testsuite/gcc.target/riscv/rvv/base/pr110943.c b/gcc/testsuite/gcc.target/riscv/rvv/base/pr110943.c index 8a6c00f..a08ac6e 100644 --- a/gcc/testsuite/gcc.target/riscv/rvv/base/pr110943.c +++ b/gcc/testsuite/gcc.target/riscv/rvv/base/pr110943.c @@ -2,7 +2,7 @@ /* { dg-options "-O3 -march=rv64gcv -mabi=lp64d" } */ /* { dg-final { check-function-bodies "**" "" } } */ -#include <riscv_vector.h> +#include "riscv_vector.h" /* ** foo9: diff --git a/gcc/testsuite/gcc.target/riscv/rvv/base/pr112431-21.c b/gcc/testsuite/gcc.target/riscv/rvv/base/pr112431-21.c index 3e43c94..a0ed793 100644 --- a/gcc/testsuite/gcc.target/riscv/rvv/base/pr112431-21.c +++ b/gcc/testsuite/gcc.target/riscv/rvv/base/pr112431-21.c @@ -1,7 +1,7 @@ /* { dg-do run { target { riscv_v } } } */ /* { dg-additional-options "-O3 -ansi -pedantic-errors -std=gnu99" } */ -#include <riscv_vector.h> +#include "riscv_vector.h" size_t __attribute__ ((noinline)) sumation (size_t sum0, size_t sum1, size_t sum2, size_t sum3, size_t sum4, diff --git a/gcc/testsuite/gcc.target/riscv/rvv/base/pr114639-1.c b/gcc/testsuite/gcc.target/riscv/rvv/base/pr114639-1.c index 3ad91db..c5b35c8 100644 --- a/gcc/testsuite/gcc.target/riscv/rvv/base/pr114639-1.c +++ b/gcc/testsuite/gcc.target/riscv/rvv/base/pr114639-1.c @@ -2,7 +2,7 @@ /* { dg-do compile } */ /* { dg-options "-march=rv64gcv -mabi=lp64d -O3" } */ -#include <riscv_vector.h> +#include "riscv_vector.h" extern size_t get_vl (); diff --git a/gcc/testsuite/gcc.target/riscv/rvv/base/pr115068-run.c b/gcc/testsuite/gcc.target/riscv/rvv/base/pr115068-run.c index d552eb5..e9e41f7 100644 --- a/gcc/testsuite/gcc.target/riscv/rvv/base/pr115068-run.c +++ b/gcc/testsuite/gcc.target/riscv/rvv/base/pr115068-run.c @@ -1,6 +1,6 @@ /* { dg-do run } */ /* { dg-require-effective-target riscv_v_ok } */ /* { dg-add-options riscv_v } */ -/* { dg-additional-options "-std=gnu99" } */ +/* { dg-additional-options "-std=gnu99 -Wno-pedantic" } */ #include "pr115068.c" diff --git a/gcc/testsuite/gcc.target/riscv/rvv/base/pr115068.c b/gcc/testsuite/gcc.target/riscv/rvv/base/pr115068.c index af2cba6..ce9a389 100644 --- a/gcc/testsuite/gcc.target/riscv/rvv/base/pr115068.c +++ b/gcc/testsuite/gcc.target/riscv/rvv/base/pr115068.c @@ -1,9 +1,9 @@ /* { dg-do compile { target { ! riscv_abi_e } } } */ /* { dg-add-options riscv_v } */ -/* { dg-additional-options "-std=gnu99" } */ +/* { dg-additional-options "-std=gnu99 -Wno-pedantic" } */ #include <stdint.h> -#include <riscv_vector.h> +#include "riscv_vector.h" vfloat64m8_t test_vfwadd_wf_f64m8_m (vbool8_t vm, vfloat64m8_t vs2, float rs1, size_t vl) diff --git a/gcc/testsuite/gcc.target/riscv/rvv/base/pr117286.c b/gcc/testsuite/gcc.target/riscv/rvv/base/pr117286.c index dabb8ae..7b6eefe 100644 --- a/gcc/testsuite/gcc.target/riscv/rvv/base/pr117286.c +++ b/gcc/testsuite/gcc.target/riscv/rvv/base/pr117286.c @@ -1,7 +1,7 @@ /* { dg-do compile } */ /* { dg-options "-march=rv64gcv_zvfh -mabi=lp64d -O1" } */ -#include <riscv_vector.h> +#include "riscv_vector.h" _Float16 a[10]; void func(){ int placeholder0 = 10; diff --git a/gcc/testsuite/gcc.target/riscv/rvv/base/pr117544.c b/gcc/testsuite/gcc.target/riscv/rvv/base/pr117544.c index af3532a..81e0ec3 100644 --- a/gcc/testsuite/gcc.target/riscv/rvv/base/pr117544.c +++ b/gcc/testsuite/gcc.target/riscv/rvv/base/pr117544.c @@ -1,7 +1,7 @@ /* { dg-do compile } */ /* { dg-options "-march=rv64gcv -mabi=lp64d -O3" } */ -#include <riscv_vector.h> +#include "riscv_vector.h" void bar() __attribute__((riscv_vector_cc)); vint32m1_t foo(vint32m1_t a, vint32m1_t b) { diff --git a/gcc/testsuite/gcc.target/riscv/rvv/base/pr117955.c b/gcc/testsuite/gcc.target/riscv/rvv/base/pr117955.c index 81e3a6e..4904c92 100644 --- a/gcc/testsuite/gcc.target/riscv/rvv/base/pr117955.c +++ b/gcc/testsuite/gcc.target/riscv/rvv/base/pr117955.c @@ -1,7 +1,7 @@ /* { dg-do compile { target { rv64 } } } */ /* { dg-options "-march=rv64gcv_zvfh -mabi=lp64d -O3" } */ -#include <riscv_vector.h> +#include "riscv_vector.h" _Float16 a (uint64_t); int8_t b () { diff --git a/gcc/testsuite/gcc.target/riscv/rvv/base/pr118872.c b/gcc/testsuite/gcc.target/riscv/rvv/base/pr118872.c index adb54d6..d62751e 100644 --- a/gcc/testsuite/gcc.target/riscv/rvv/base/pr118872.c +++ b/gcc/testsuite/gcc.target/riscv/rvv/base/pr118872.c @@ -3,7 +3,7 @@ /* { dg-options "-march=rv64gcv -mabi=lp64d -O2" { target { rv64 } } } */ /* { dg-options "-march=rv32gcv -mabi=ilp32d -O2" { target { rv32 } } } */ -#include <riscv_vector.h> +#include "riscv_vector.h" vfloat32m2_t foo (vfloat16m1_t a, size_t vl) { diff --git a/gcc/testsuite/gcc.target/riscv/rvv/base/vlmul_ext-1.c b/gcc/testsuite/gcc.target/riscv/rvv/base/vlmul_ext-1.c index 4253729..6bb7e1c 100644 --- a/gcc/testsuite/gcc.target/riscv/rvv/base/vlmul_ext-1.c +++ b/gcc/testsuite/gcc.target/riscv/rvv/base/vlmul_ext-1.c @@ -1,7 +1,7 @@ /* { dg-do compile } */ /* { dg-options "-march=rv64gcv -mabi=lp64d -O3 -fno-schedule-insns -fno-schedule-insns2" } */ -#include <riscv_vector.h> +#include "riscv_vector.h" vint16m8_t test_vlmul_ext_v_i16mf4_i16m8(vint16mf4_t op1) { return __riscv_vlmul_ext_v_i16mf4_i16m8(op1); diff --git a/gcc/testsuite/gcc.target/riscv/rvv/base/vssubu-1.c b/gcc/testsuite/gcc.target/riscv/rvv/base/vssubu-1.c index 606854b..a278709 100644 --- a/gcc/testsuite/gcc.target/riscv/rvv/base/vssubu-1.c +++ b/gcc/testsuite/gcc.target/riscv/rvv/base/vssubu-1.c @@ -1,7 +1,7 @@ /* { dg-do compile } */ /* { dg-options "-O2 -march=rv64gcv -mabi=lp64d" } */ -#include <riscv_vector.h> +#include "riscv_vector.h" vuint64m1_t test_vssubu_vx_u64m1(vuint64m1_t op1) { diff --git a/gcc/testsuite/gcc.target/riscv/rvv/base/vssubu-2.c b/gcc/testsuite/gcc.target/riscv/rvv/base/vssubu-2.c index 78abd09..2f8c146 100644 --- a/gcc/testsuite/gcc.target/riscv/rvv/base/vssubu-2.c +++ b/gcc/testsuite/gcc.target/riscv/rvv/base/vssubu-2.c @@ -1,7 +1,7 @@ /* { dg-do compile } */ /* { dg-options "-O2 -march=rv32gcv -mabi=ilp32d" } */ -#include <riscv_vector.h> +#include "riscv_vector.h" vuint64m1_t test_vssubu_vx_u64m1(vuint64m1_t op1) { diff --git a/gcc/testsuite/gcc.target/riscv/rvv/base/vwaddsub-1.c b/gcc/testsuite/gcc.target/riscv/rvv/base/vwaddsub-1.c index 84d3c4c..43be202 100644 --- a/gcc/testsuite/gcc.target/riscv/rvv/base/vwaddsub-1.c +++ b/gcc/testsuite/gcc.target/riscv/rvv/base/vwaddsub-1.c @@ -1,9 +1,9 @@ /* { dg-do compile { target { { ! riscv_abi_e } && rv64 } } } */ /* { dg-add-options riscv_v } */ -/* { dg-additional-options "-std=gnu99 -O3 -fno-schedule-insns -fno-schedule-insns2" } */ +/* { dg-additional-options "-std=gnu99 -O3 -fno-schedule-insns -fno-schedule-insns2 -Wno-pedantic" } */ #include <stdint.h> -#include <riscv_vector.h> +#include "riscv_vector.h" /* ** vwadd_wx_i64m8_m: diff --git a/gcc/testsuite/gcc.target/riscv/rvv/vsetvl/avl_single-68.c b/gcc/testsuite/gcc.target/riscv/rvv/vsetvl/avl_single-68.c index bf95e1c..64666d3 100644 --- a/gcc/testsuite/gcc.target/riscv/rvv/vsetvl/avl_single-68.c +++ b/gcc/testsuite/gcc.target/riscv/rvv/vsetvl/avl_single-68.c @@ -21,6 +21,12 @@ void f2 (void * restrict in, void * restrict out, int l, int n, int m) } } +/* The second check is XFAILed because we currently don't lift + vsetvls into non-transparent (in LCM parlance) blocks. + See PR119547. + In this test it is still possible because the conflicting + register only ever feeds vsetvls. */ + /* { dg-final { scan-assembler-times {vsetvli} 2 { target { no-opts "-O0" no-opts "-Os" no-opts "-Oz" no-opts "-g" no-opts "-funroll-loops" } } } } */ -/* { dg-final { scan-assembler-times {vsetvli\s+zero,\s*[a-x0-9]+,\s*e8,\s*mf8,\s*tu,\s*m[au]} 2 { target { no-opts "-O0" no-opts "-Os" no-opts "-Oz" no-opts "-g" no-opts "-funroll-loops" } } } } */ +/* { dg-final { scan-assembler-times {vsetvli\s+zero,\s*[a-x0-9]+,\s*e8,\s*mf8,\s*tu,\s*m[au]} 2 { target { no-opts "-O0" no-opts "-Os" no-opts "-Oz" no-opts "-g" no-opts "-funroll-loops" } xfail { *-*-* } } } } */ /* { dg-final { scan-assembler-times {addi\s+[a-x0-9]+,\s*[a-x0-9]+,\s*44} 1 { target { no-opts "-O0" no-opts "-Os" no-opts "-Oz" no-opts "-g" no-opts "-funroll-loops" } } } } */ diff --git a/gcc/testsuite/gcc.target/riscv/rvv/vsetvl/pr111234.c b/gcc/testsuite/gcc.target/riscv/rvv/vsetvl/pr111234.c index 871cf65..f594217 100644 --- a/gcc/testsuite/gcc.target/riscv/rvv/vsetvl/pr111234.c +++ b/gcc/testsuite/gcc.target/riscv/rvv/vsetvl/pr111234.c @@ -1,7 +1,7 @@ /* { dg-do compile } */ /* { dg-options "-mrvv-vector-bits=scalable -march=rv64gcv -mabi=lp64d -O3" } */ -#include <riscv_vector.h> +#include "riscv_vector.h" void f (vint32m1_t *in, vint64m2_t *out, vbool32_t *m, int b) diff --git a/gcc/testsuite/gcc.target/riscv/rvv/vsetvl/pr115214.c b/gcc/testsuite/gcc.target/riscv/rvv/vsetvl/pr115214.c index b76760b..48f200f 100644 --- a/gcc/testsuite/gcc.target/riscv/rvv/vsetvl/pr115214.c +++ b/gcc/testsuite/gcc.target/riscv/rvv/vsetvl/pr115214.c @@ -2,7 +2,7 @@ /* { dg-options "-mrvv-vector-bits=scalable -march=rv64gcv -mabi=lp64d -O3 -w -std=gnu17" } */ /* { dg-skip-if "" { *-*-* } { "-flto" } } */ -#include <riscv_vector.h> +#include "riscv_vector.h" static inline __attribute__(()) int vaddq_f32(); static inline __attribute__(()) int vload_tillz_f32(int nlane) { diff --git a/gcc/testsuite/gcc.target/riscv/rvv/vsetvl/vlmax_switch_vtype-10.c b/gcc/testsuite/gcc.target/riscv/rvv/vsetvl/vlmax_switch_vtype-10.c index ddf53ca..0dbf34a 100644 --- a/gcc/testsuite/gcc.target/riscv/rvv/vsetvl/vlmax_switch_vtype-10.c +++ b/gcc/testsuite/gcc.target/riscv/rvv/vsetvl/vlmax_switch_vtype-10.c @@ -43,6 +43,6 @@ void foo (int8_t * restrict in, int8_t * restrict out, int n, int cond) } } -/* { dg-final { scan-assembler-times {vsetvli} 15 { target { no-opts "-O0" no-opts "-O1" no-opts "-Os" no-opts "-Oz" no-opts "-funroll-loops" no-opts "-g" no-opts "-flto" } } } } */ +/* { dg-final { scan-assembler-times {vsetvli} 14 { target { no-opts "-O0" no-opts "-O1" no-opts "-Os" no-opts "-Oz" no-opts "-funroll-loops" no-opts "-g" no-opts "-flto" } } } } */ /* { dg-final { scan-assembler-times {vsetvli\s+[a-x0-9]+,\s*zero,\s*e32,\s*mf2,\s*t[au],\s*m[au]} 3 { target { no-opts "-O0" no-opts "-O1" no-opts "-Os" no-opts "-Oz" no-opts "-funroll-loops" no-opts "-g" } } } } */ -/* { dg-final { scan-assembler-times {vsetvli\s+[a-x0-9]+,\s*zero,\s*e16,\s*mf2,\s*t[au],\s*m[au]} 4 { target { no-opts "-O0" no-opts "-O1" no-opts "-Os" no-opts "-Oz" no-opts "-funroll-loops" no-opts "-g" } } } } */ +/* { dg-final { scan-assembler-times {vsetvli\s+[a-x0-9]+,\s*zero,\s*e16,\s*mf2,\s*t[au],\s*m[au]} 3 { target { no-opts "-O0" no-opts "-O1" no-opts "-Os" no-opts "-Oz" no-opts "-funroll-loops" no-opts "-g" } } } } */ diff --git a/gcc/testsuite/gcc.target/riscv/rvv/vsetvl/vsetvl-24.c b/gcc/testsuite/gcc.target/riscv/rvv/vsetvl/vsetvl-24.c index 7096159e..3867681 100644 --- a/gcc/testsuite/gcc.target/riscv/rvv/vsetvl/vsetvl-24.c +++ b/gcc/testsuite/gcc.target/riscv/rvv/vsetvl/vsetvl-24.c @@ -1,7 +1,7 @@ /* { dg-do compile } */ /* { dg-options "-mrvv-vector-bits=scalable -march=rv64gcv -mabi=lp64d" } */ -#include <riscv_vector.h> +#include "riscv_vector.h" size_t foo () { diff --git a/gcc/testsuite/gcc.target/riscv/rvv/vsetvl/vsetvl_bug-3.c b/gcc/testsuite/gcc.target/riscv/rvv/vsetvl/vsetvl_bug-3.c index c155f56..3acbc73 100644 --- a/gcc/testsuite/gcc.target/riscv/rvv/vsetvl/vsetvl_bug-3.c +++ b/gcc/testsuite/gcc.target/riscv/rvv/vsetvl/vsetvl_bug-3.c @@ -1,7 +1,7 @@ /* { dg-do compile } */ /* { dg-options "-march=rv32gcv -mabi=ilp32d -O2 -fdump-rtl-vsetvl-details" } */ -#include <riscv_vector.h> +#include "riscv_vector.h" uint64_t a[2], b[2]; diff --git a/gcc/testsuite/gcc.target/riscv/rvv/vsetvl/vsetvl_bug-4.c b/gcc/testsuite/gcc.target/riscv/rvv/vsetvl/vsetvl_bug-4.c index 04a8ff2..2b2fe27 100644 --- a/gcc/testsuite/gcc.target/riscv/rvv/vsetvl/vsetvl_bug-4.c +++ b/gcc/testsuite/gcc.target/riscv/rvv/vsetvl/vsetvl_bug-4.c @@ -1,7 +1,7 @@ /* { dg-do compile } */ /* { dg-options "-march=rv64gcv -mabi=lp64d -O2 -fno-schedule-insns -fdump-rtl-vsetvl-details" } */ -#include <riscv_vector.h> +#include "riscv_vector.h" vuint16m1_t foo (vuint16m1_t a, vuint16m1_t b, size_t avl) diff --git a/gcc/testsuite/gcc.target/riscv/rvv/xtheadvector/pr116591.c b/gcc/testsuite/gcc.target/riscv/rvv/xtheadvector/pr116591.c index dfaf82c..ad27c38 100644 --- a/gcc/testsuite/gcc.target/riscv/rvv/xtheadvector/pr116591.c +++ b/gcc/testsuite/gcc.target/riscv/rvv/xtheadvector/pr116591.c @@ -2,7 +2,7 @@ /* { dg-options "-march=rv32gc_xtheadvector -mabi=ilp32d -O2 -save-temps" { target { rv32 } } } */ /* { dg-options "-march=rv64gc_xtheadvector -mabi=lp64d -O2 -save-temps" { target { rv64 } } } */ -#include <riscv_vector.h> +#include "riscv_vector.h" void foo (float *a, int b) diff --git a/gcc/testsuite/gcc.target/riscv/rvv/xtheadvector/pr116592.c b/gcc/testsuite/gcc.target/riscv/rvv/xtheadvector/pr116592.c index a7cd8c5..c8056a8 100644 --- a/gcc/testsuite/gcc.target/riscv/rvv/xtheadvector/pr116592.c +++ b/gcc/testsuite/gcc.target/riscv/rvv/xtheadvector/pr116592.c @@ -3,7 +3,7 @@ /* { dg-options "-march=rv64gc_zfh_xtheadvector -mabi=lp64d -O2 -save-temps" { target { rv64 } } } */ #include <math.h> -#include <riscv_vector.h> +#include "riscv_vector.h" static vfloat32m8_t atan2_ps(vfloat32m8_t a, vfloat32m8_t b, size_t vl) { diff --git a/gcc/testsuite/gcc.target/riscv/rvv/xtheadvector/pr118357.c b/gcc/testsuite/gcc.target/riscv/rvv/xtheadvector/pr118357.c index aebb0e3..b3c3428 100644 --- a/gcc/testsuite/gcc.target/riscv/rvv/xtheadvector/pr118357.c +++ b/gcc/testsuite/gcc.target/riscv/rvv/xtheadvector/pr118357.c @@ -1,7 +1,7 @@ /* { dg-do compile { target { rv64 } } } */ /* { dg-options "-march=rv64gc_xtheadvector -mabi=lp64d -O2" } */ -#include <riscv_vector.h> +#include "riscv_vector.h" vfloat16m4_t foo (float *ptr, size_t vl) { diff --git a/gcc/testsuite/gcc.target/riscv/rvv/xtheadvector/vsext.c b/gcc/testsuite/gcc.target/riscv/rvv/xtheadvector/vsext.c index 55db283..42fa43e 100644 --- a/gcc/testsuite/gcc.target/riscv/rvv/xtheadvector/vsext.c +++ b/gcc/testsuite/gcc.target/riscv/rvv/xtheadvector/vsext.c @@ -1,7 +1,7 @@ /* { dg-do compile { target { rv64 } } } */ /* { dg-options "-march=rv64gc_xtheadvector -mabi=lp64d -O3" } */ -#include <riscv_vector.h> +#include "riscv_vector.h" struct a { diff --git a/gcc/testsuite/gcc.target/riscv/rvv/xtheadvector/vzext.c b/gcc/testsuite/gcc.target/riscv/rvv/xtheadvector/vzext.c index fcb5659..d622b72 100644 --- a/gcc/testsuite/gcc.target/riscv/rvv/xtheadvector/vzext.c +++ b/gcc/testsuite/gcc.target/riscv/rvv/xtheadvector/vzext.c @@ -1,7 +1,7 @@ /* { dg-do compile { target { rv64 } } } */ /* { dg-options "-march=rv64gc_xtheadvector -mabi=lp64d -O3" } */ -#include <riscv_vector.h> +#include "riscv_vector.h" struct a { diff --git a/gcc/testsuite/gcc.target/sh/pr111814.c b/gcc/testsuite/gcc.target/sh/pr111814.c new file mode 100644 index 0000000..a88e5d7 --- /dev/null +++ b/gcc/testsuite/gcc.target/sh/pr111814.c @@ -0,0 +1,7 @@ +/* Verify that __builtin_nan("") produces a constant matches + architecture specification. */ +/* { dg-do compile } */ + +double d = __builtin_nan (""); + +/* { dg-final { scan-assembler "\t.long\t-1\n\t.long\t2146959359\n" } } */ diff --git a/gcc/testsuite/gdc.dg/debug/imports/m119817/a.d b/gcc/testsuite/gdc.dg/debug/imports/m119817/a.d new file mode 100644 index 0000000..a137472 --- /dev/null +++ b/gcc/testsuite/gdc.dg/debug/imports/m119817/a.d @@ -0,0 +1,2 @@ +module imports.m119817.a; +void f119817()() { } diff --git a/gcc/testsuite/gdc.dg/debug/imports/m119817/b.d b/gcc/testsuite/gdc.dg/debug/imports/m119817/b.d new file mode 100644 index 0000000..aef0e37 --- /dev/null +++ b/gcc/testsuite/gdc.dg/debug/imports/m119817/b.d @@ -0,0 +1,2 @@ +module imports.m119817.b; +void f119817() { } diff --git a/gcc/testsuite/gdc.dg/debug/imports/m119817/package.d b/gcc/testsuite/gdc.dg/debug/imports/m119817/package.d new file mode 100644 index 0000000..188827e --- /dev/null +++ b/gcc/testsuite/gdc.dg/debug/imports/m119817/package.d @@ -0,0 +1,4 @@ +module imports.m119817; +public import + imports.m119817.a, + imports.m119817.b; diff --git a/gcc/testsuite/gdc.dg/debug/imports/pr119826b.d b/gcc/testsuite/gdc.dg/debug/imports/pr119826b.d new file mode 100644 index 0000000..3c5a6ac --- /dev/null +++ b/gcc/testsuite/gdc.dg/debug/imports/pr119826b.d @@ -0,0 +1,14 @@ +module imports.pr119826b; + +import pr119826 : t119826; + +class C119826 +{ + enum E119826 { Evalue } + const E119826 em = void; +} + +void f119826(C119826 c) +{ + t119826(c.em); +} diff --git a/gcc/testsuite/gdc.dg/debug/pr119817.d b/gcc/testsuite/gdc.dg/debug/pr119817.d new file mode 100644 index 0000000..3eea6ba --- /dev/null +++ b/gcc/testsuite/gdc.dg/debug/pr119817.d @@ -0,0 +1,6 @@ +// { dg-do compile } +// { dg-additional-sources "imports/m119817/package.d" } +// { dg-additional-sources "imports/m119817/a.d" } +// { dg-additional-sources "imports/m119817/b.d" } +module pr119817; +import imports.m119817 : f119817; diff --git a/gcc/testsuite/gdc.dg/debug/pr119826.d b/gcc/testsuite/gdc.dg/debug/pr119826.d new file mode 100644 index 0000000..2fb98c7 --- /dev/null +++ b/gcc/testsuite/gdc.dg/debug/pr119826.d @@ -0,0 +1,8 @@ +// { dg-do compile } +// { dg-additional-sources "imports/pr119826b.d" } +module pr119826; + +int t119826(A)(A args) +{ + assert(false); +} diff --git a/gcc/testsuite/gdc.dg/driver_fonly1.d b/gcc/testsuite/gdc.dg/driver_fonly1.d new file mode 100644 index 0000000..1af956a --- /dev/null +++ b/gcc/testsuite/gdc.dg/driver_fonly1.d @@ -0,0 +1,2 @@ +// { dg-additional-options "-fonly=not-a-file" } +// { dg-error "argument is different from first input file name" "" { target *-*-* } 0 } diff --git a/gcc/testsuite/gdc.dg/driver_fonly2.d b/gcc/testsuite/gdc.dg/driver_fonly2.d new file mode 100644 index 0000000..97cd93d --- /dev/null +++ b/gcc/testsuite/gdc.dg/driver_fonly2.d @@ -0,0 +1,8 @@ +// { dg-do "compile" } +// { dg-additional-options "-fonly=driver_fonly2.d" } +// { dg-additional-sources "imports/fonly.d" } +// { dg-final { scan-assembler "_D1a10fonly_testFZv" } } +// { dg-final { scan-assembler-not "_D1b10fonly_testFZv" } } +module a; + +void fonly_test() { } diff --git a/gcc/testsuite/gdc.dg/driver_fonly3.d b/gcc/testsuite/gdc.dg/driver_fonly3.d new file mode 100644 index 0000000..de2983f --- /dev/null +++ b/gcc/testsuite/gdc.dg/driver_fonly3.d @@ -0,0 +1,8 @@ +// { dg-do "compile" } +// { dg-additional-options "-fonly=imports/fonly" } +// { dg-additional-sources "imports/fonly.d" } +// { dg-final { scan-assembler-not "_D1a10fonly_testFZv" } } +// { dg-final { scan-assembler "_D1b10fonly_testFZv" } } +module a; + +void fonly_test() { } diff --git a/gcc/testsuite/gdc.dg/import-c/import-c.exp b/gcc/testsuite/gdc.dg/import-c/import-c.exp new file mode 100644 index 0000000..53d1478 --- /dev/null +++ b/gcc/testsuite/gdc.dg/import-c/import-c.exp @@ -0,0 +1,29 @@ +# Copyright (C) 2025 Free Software Foundation, Inc. + +# This program is free software; you can redistribute it and/or modify +# it under the terms of the GNU General Public License as published by +# the Free Software Foundation; either version 3 of the License, or +# (at your option) any later version. +# +# This program is distributed in the hope that it will be useful, +# but WITHOUT ANY WARRANTY; without even the implied warranty of +# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the +# GNU General Public License for more details. +# +# You should have received a copy of the GNU General Public License +# along with GCC; see the file COPYING3. If not see +# <http://www.gnu.org/licenses/>. + +# Load support procs. +load_lib gdc-dg.exp + +# Initialize `dg'. +dg-init + +# Main loop. +gdc-dg-runtest [lsort \ + [glob -nocomplain $srcdir/$subdir/*.d ] ] "" \ + "-I $srcdir/$subdir -finclude-imports" + +# All done. +dg-finish diff --git a/gcc/testsuite/gdc.dg/import-c/pr119761.d b/gcc/testsuite/gdc.dg/import-c/pr119761.d new file mode 100644 index 0000000..20eff31 --- /dev/null +++ b/gcc/testsuite/gdc.dg/import-c/pr119761.d @@ -0,0 +1,2 @@ +// { dg-do compile } +import pr119761c; diff --git a/gcc/testsuite/gdc.dg/import-c/pr119761c.c b/gcc/testsuite/gdc.dg/import-c/pr119761c.c new file mode 100644 index 0000000..522f1bf --- /dev/null +++ b/gcc/testsuite/gdc.dg/import-c/pr119761c.c @@ -0,0 +1,4 @@ +int f119761(const char *, ...) +{ + return 0; +} diff --git a/gcc/testsuite/gdc.dg/import-c/pr119799.d b/gcc/testsuite/gdc.dg/import-c/pr119799.d new file mode 100644 index 0000000..d8b0fa2 --- /dev/null +++ b/gcc/testsuite/gdc.dg/import-c/pr119799.d @@ -0,0 +1,2 @@ +// { dg-do compile } +import pr119799c; diff --git a/gcc/testsuite/gdc.dg/import-c/pr119799c.c b/gcc/testsuite/gdc.dg/import-c/pr119799c.c new file mode 100644 index 0000000..b80e856 --- /dev/null +++ b/gcc/testsuite/gdc.dg/import-c/pr119799c.c @@ -0,0 +1 @@ +static struct {} s119799; diff --git a/gcc/testsuite/gdc.dg/imports/fonly.d b/gcc/testsuite/gdc.dg/imports/fonly.d new file mode 100644 index 0000000..2b7755e --- /dev/null +++ b/gcc/testsuite/gdc.dg/imports/fonly.d @@ -0,0 +1,3 @@ +module b; + +void fonly_test() { } diff --git a/gcc/testsuite/gdc.dg/torture/imports/pr109023.d b/gcc/testsuite/gdc.dg/torture/imports/pr109023.d new file mode 100644 index 0000000..e85e0ed --- /dev/null +++ b/gcc/testsuite/gdc.dg/torture/imports/pr109023.d @@ -0,0 +1,3 @@ +module imports.pr109023; + +void f109023() { } diff --git a/gcc/testsuite/gdc.dg/torture/pr109023.d b/gcc/testsuite/gdc.dg/torture/pr109023.d new file mode 100644 index 0000000..3060446 --- /dev/null +++ b/gcc/testsuite/gdc.dg/torture/pr109023.d @@ -0,0 +1,6 @@ +// { dg-do "compile" } +// { dg-additional-options "-I[srcdir] -finclude-imports" } +// { dg-additional-files "imports/pr109023.d" } +// { dg-final { scan-assembler "_D7imports8pr1090237f109023FZv" } } +module pr109023; +import imports.pr109023; diff --git a/gcc/testsuite/gdc.test/compilable/test21179.d b/gcc/testsuite/gdc.test/compilable/test21179.d new file mode 100644 index 0000000..78bdffd --- /dev/null +++ b/gcc/testsuite/gdc.test/compilable/test21179.d @@ -0,0 +1,11 @@ +// https://github.com/dlang/dmd/issues/21179 + +void bigEndianToNative(ubyte[2] a) {} + +void main() +{ + ubyte[] arr; + const ubyte[2] bytes; + bigEndianToNative(bytes); + auto b = cast(const ubyte[2][]) arr; +} diff --git a/gcc/testsuite/gdc.test/fail_compilation/fail_pretty_errors.d b/gcc/testsuite/gdc.test/fail_compilation/fail_pretty_errors.d index 2016a50..79242b1 100644 --- a/gcc/testsuite/gdc.test/fail_compilation/fail_pretty_errors.d +++ b/gcc/testsuite/gdc.test/fail_compilation/fail_pretty_errors.d @@ -1,22 +1,24 @@ -/* +/* REQUIRED_ARGS: -verrors=context TEST_OUTPUT: --- -fail_compilation/fail_pretty_errors.d(27): Error: undefined identifier `a` +fail_compilation/fail_pretty_errors.d(29): Error: undefined identifier `a` a = 1; ^ -fail_compilation/fail_pretty_errors.d-mixin-32(32): Error: undefined identifier `b` -fail_compilation/fail_pretty_errors.d(37): Error: cannot implicitly convert expression `5` of type `int` to `string` +fail_compilation/fail_pretty_errors.d-mixin-34(34): Error: undefined identifier `b` +b = 1; +^ +fail_compilation/fail_pretty_errors.d(39): Error: cannot implicitly convert expression `5` of type `int` to `string` string x = 5; ^ -fail_compilation/fail_pretty_errors.d(42): Error: mixin `fail_pretty_errors.testMixin2.mixinTemplate!()` error instantiating +fail_compilation/fail_pretty_errors.d(44): Error: mixin `fail_pretty_errors.testMixin2.mixinTemplate!()` error instantiating mixin mixinTemplate; ^ -fail_compilation/fail_pretty_errors.d(48): Error: invalid array operation `"" + ""` (possible missing []) +fail_compilation/fail_pretty_errors.d(50): Error: invalid array operation `"" + ""` (possible missing []) auto x = ""+""; ^ -fail_compilation/fail_pretty_errors.d(48): did you mean to concatenate (`"" ~ ""`) instead ? -fail_compilation/fail_pretty_errors.d(51): Error: cannot implicitly convert expression `1111` of type `int` to `byte` +fail_compilation/fail_pretty_errors.d(50): did you mean to concatenate (`"" ~ ""`) instead ? +fail_compilation/fail_pretty_errors.d(53): Error: cannot implicitly convert expression `1111` of type `int` to `byte` byte ɑ = 1111; ^ --- diff --git a/gcc/testsuite/gdc.test/fail_compilation/test21247.d b/gcc/testsuite/gdc.test/fail_compilation/test21247.d new file mode 100644 index 0000000..c3e4105 --- /dev/null +++ b/gcc/testsuite/gdc.test/fail_compilation/test21247.d @@ -0,0 +1,20 @@ +/* +TEST_OUTPUT: +--- +fail_compilation/test21247.d(13): Error: anonymous union can only be a part of an aggregate, not function `hang_dmd` +fail_compilation/test21247.d(17): Error: undefined identifier `u` +fail_compilation/test21247.d(18): Error: undefined identifier `b` +fail_compilation/test21247.d(20): called from here: `hang_dmd(0u)` +--- + */ +// https://github.com/dlang/dmd/issues/21247 +ubyte[4] hang_dmd(uint a) +{ + union { + uint u = void; + ubyte[4] b; + } + u = a; + return b; +} +enum T = hang_dmd(0); diff --git a/gcc/testsuite/gdc.test/fail_compilation/test21247b.d b/gcc/testsuite/gdc.test/fail_compilation/test21247b.d new file mode 100644 index 0000000..ecd4603 --- /dev/null +++ b/gcc/testsuite/gdc.test/fail_compilation/test21247b.d @@ -0,0 +1,14 @@ +/* +TEST_OUTPUT: +--- +fail_compilation/test21247b.d(10): Error: anonymous union can only be a part of an aggregate, not function `test21247` +--- + */ +// https://github.com/dlang/dmd/issues/21247 +void test21247() +{ + union { + uint u = void; + ubyte[4] b; + } +} diff --git a/gcc/testsuite/gfortran.dg/do_concurrent_all_clauses.f90 b/gcc/testsuite/gfortran.dg/do_concurrent_all_clauses.f90 index 0c8a6ad..a7fa7c3 100644 --- a/gcc/testsuite/gfortran.dg/do_concurrent_all_clauses.f90 +++ b/gcc/testsuite/gfortran.dg/do_concurrent_all_clauses.f90 @@ -18,7 +18,7 @@ program do_concurrent_all_clauses squared = i * i arr(i) = temp2 + squared sum = sum + arr(i) - max_val = max(max_val, arr(i)) ! { dg-error "Reference to impure function" } + max_val = max(max_val, arr(i)) end block end do print *, arr, sum, max_val diff --git a/gcc/testsuite/gfortran.dg/gomp/map-alloc-comp-1.f90 b/gcc/testsuite/gfortran.dg/gomp/map-alloc-comp-1.f90 index 0c44296..f48addc 100644 --- a/gcc/testsuite/gfortran.dg/gomp/map-alloc-comp-1.f90 +++ b/gcc/testsuite/gfortran.dg/gomp/map-alloc-comp-1.f90 @@ -10,5 +10,5 @@ type sct end type type(sct) var -!$omp target enter data map(to:var) ! { dg-error "allocatable components is not permitted in map clause" } +!$omp target enter data map(to:var) end diff --git a/gcc/testsuite/gfortran.dg/gomp/polymorphic-mapping-1.f90 b/gcc/testsuite/gfortran.dg/gomp/polymorphic-mapping-1.f90 new file mode 100644 index 0000000..750cec9 --- /dev/null +++ b/gcc/testsuite/gfortran.dg/gomp/polymorphic-mapping-1.f90 @@ -0,0 +1,30 @@ +type t + integer :: t +end type t +class(t), target, allocatable :: c, ca(:) +class(t), pointer :: p, pa(:) +integer :: x +allocate( t :: c, ca(5)) +p => c +pa => ca + +! 11111111112222222222333333333344 +!2345678901234567890123456789012345678901 +!$omp target enter data map(c, ca, p, pa) +! { dg-warning "29:Mapping of polymorphic list item 'c' is unspecified behavior \\\[-Wopenmp\\\]" "" { target *-*-* } .-1 } +! { dg-warning "32:Mapping of polymorphic list item 'ca' is unspecified behavior \\\[-Wopenmp\\\]" "" { target *-*-* } .-2 } +! { dg-warning "36:Mapping of polymorphic list item 'p' is unspecified behavior \\\[-Wopenmp\\\]" "" { target *-*-* } .-3 } +! { dg-warning "39:Mapping of polymorphic list item 'pa' is unspecified behavior \\\[-Wopenmp\\\]" "" { target *-*-* } .-4 } + +! 11111111112222222222333333333344 +!2345678901234567890123456789012345678901 + +! 11111111112222222222333333333344 +!2345678901234567890123456789012345678901 +!$omp target update from(c,ca), to(p,pa) +! { dg-warning "26:Mapping of polymorphic list item 'c' is unspecified behavior \\\[-Wopenmp\\\]" "" { target *-*-* } .-1 } +! { dg-warning "28:Mapping of polymorphic list item 'ca' is unspecified behavior \\\[-Wopenmp\\\]" "" { target *-*-* } .-2 } +! { dg-warning "36:Mapping of polymorphic list item 'p' is unspecified behavior \\\[-Wopenmp\\\]" "" { target *-*-* } .-3 } +! { dg-warning "38:Mapping of polymorphic list item 'pa' is unspecified behavior \\\[-Wopenmp\\\]" "" { target *-*-* } .-4 } + +end diff --git a/gcc/testsuite/gfortran.dg/gomp/polymorphic-mapping-2.f90 b/gcc/testsuite/gfortran.dg/gomp/polymorphic-mapping-2.f90 index e25db68..3bedc9b 100644 --- a/gcc/testsuite/gfortran.dg/gomp/polymorphic-mapping-2.f90 +++ b/gcc/testsuite/gfortran.dg/gomp/polymorphic-mapping-2.f90 @@ -9,7 +9,7 @@ allocate( t :: c, ca(5)) p => c pa => ca -!$omp target ! { dg-warning "Implicit mapping of polymorphic variable 'ca' is unspecified behavior \\\[-Wopenmp\\\]" } +!$omp target ! { dg-warning "Mapping of polymorphic list item 'ca' is unspecified behavior \\\[-Wopenmp\\\]" } ll = allocated(ca) !$omp end target diff --git a/gcc/testsuite/gfortran.dg/gomp/polymorphic-mapping-3.f90 b/gcc/testsuite/gfortran.dg/gomp/polymorphic-mapping-3.f90 new file mode 100644 index 0000000..9777ecf --- /dev/null +++ b/gcc/testsuite/gfortran.dg/gomp/polymorphic-mapping-3.f90 @@ -0,0 +1,23 @@ +subroutine sub(var, var2) +type t + integer :: x +end type t + +type t2 + integer :: x + integer, allocatable :: y +end type + +class(t) var, var2 +type(t2) :: var3, var4 +!$omp target firstprivate(var) & ! { dg-error "Polymorphic list item 'var' at .1. in FIRSTPRIVATE clause has unspecified behavior and unsupported" } +!$omp& private(var2) ! { dg-error "Polymorphic list item 'var2' at .1. in PRIVATE clause has unspecified behavior and unsupported" } + var%x = 5 + var2%x = 5 +!$omp end target +!$omp target firstprivate(var3) & ! { dg-error "Sorry, list item 'var3' at .1. with allocatable components is not yet supported in FIRSTPRIVATE clause" } +!$omp& private(var4) ! { dg-error "Sorry, list item 'var4' at .1. with allocatable components is not yet supported in PRIVATE clause" } + var3%x = 5 + var4%x = 5 +!$omp end target +end diff --git a/gcc/testsuite/gfortran.dg/gomp/polymorphic-mapping-4.f90 b/gcc/testsuite/gfortran.dg/gomp/polymorphic-mapping-4.f90 new file mode 100644 index 0000000..5a1a70a --- /dev/null +++ b/gcc/testsuite/gfortran.dg/gomp/polymorphic-mapping-4.f90 @@ -0,0 +1,9 @@ +subroutine one +implicit none +type t + class(*), allocatable :: ul +end type + +type(t) :: var +!$omp target enter data map(to:var) ! { dg-error "Mapping of unlimited polymorphic list item 'var.ul' is unspecified behavior and unsupported" } +end diff --git a/gcc/testsuite/gfortran.dg/gomp/polymorphic-mapping-5.f90 b/gcc/testsuite/gfortran.dg/gomp/polymorphic-mapping-5.f90 new file mode 100644 index 0000000..4b5814e --- /dev/null +++ b/gcc/testsuite/gfortran.dg/gomp/polymorphic-mapping-5.f90 @@ -0,0 +1,9 @@ +subroutine one +implicit none +type t + class(*), allocatable :: ul +end type + +class(*), allocatable :: ul_var +!$omp target enter data map(to: ul_var) ! { dg-error "Mapping of unlimited polymorphic list item 'ul_var' is unspecified behavior and unsupported" } +end diff --git a/gcc/testsuite/gfortran.dg/gomp/polymorphic-mapping.f90 b/gcc/testsuite/gfortran.dg/gomp/polymorphic-mapping.f90 index dd7eb31..752cca2 100644 --- a/gcc/testsuite/gfortran.dg/gomp/polymorphic-mapping.f90 +++ b/gcc/testsuite/gfortran.dg/gomp/polymorphic-mapping.f90 @@ -10,37 +10,21 @@ pa => ca ! 11111111112222222222333333333344 !2345678901234567890123456789012345678901 -!$omp target enter data map(c, ca, p, pa) -! { dg-warning "29:Mapping polymorphic list item at .1. is unspecified behavior \\\[-Wopenmp\\\]" "" { target *-*-* } .-1 } -! { dg-warning "32:Mapping polymorphic list item at .1. is unspecified behavior \\\[-Wopenmp\\\]" "" { target *-*-* } .-2 } -! { dg-warning "36:Mapping polymorphic list item at .1. is unspecified behavior \\\[-Wopenmp\\\]" "" { target *-*-* } .-3 } -! { dg-warning "39:Mapping polymorphic list item at .1. is unspecified behavior \\\[-Wopenmp\\\]" "" { target *-*-* } .-4 } - -! 11111111112222222222333333333344 -!2345678901234567890123456789012345678901 -!$omp target firstprivate(ca) ! { dg-warning "27:FIRSTPRIVATE with polymorphic list item at .1. is unspecified behavior \\\[-Wopenmp\\\]" } +!$omp target firstprivate(ca) ! { dg-error "27:Polymorphic list item 'ca' at .1. in FIRSTPRIVATE clause has unspecified behavior and unsupported" } !$omp end target -!$omp target parallel do firstprivate(ca) ! { dg-warning "39:FIRSTPRIVATE with polymorphic list item at .1. is unspecified behavior \\\[-Wopenmp\\\]" } +!$omp target parallel do firstprivate(ca) ! { dg-error "39:Polymorphic list item 'ca' at .1. in FIRSTPRIVATE clause has unspecified behavior and unsupported" } do x = 0, 5 end do -!$omp target parallel do private(ca) ! OK; should map declared type +!$omp target parallel do private(ca) ! { dg-error "34:Polymorphic list item 'ca' at .1. in PRIVATE clause has unspecified behavior and unsupported" } do x = 0, 5 end do -!$omp target private(ca) ! OK; should map declared type +!$omp target private(ca) ! { dg-error "22:Polymorphic list item 'ca' at .1. in PRIVATE clause has unspecified behavior and unsupported" } block end block -! 11111111112222222222333333333344 -!2345678901234567890123456789012345678901 -!$omp target update from(c,ca), to(p,pa) -! { dg-warning "26:Mapping polymorphic list item at .1. is unspecified behavior \\\[-Wopenmp\\\]" "" { target *-*-* } .-1 } -! { dg-warning "28:Mapping polymorphic list item at .1. is unspecified behavior \\\[-Wopenmp\\\]" "" { target *-*-* } .-2 } -! { dg-warning "36:Mapping polymorphic list item at .1. is unspecified behavior \\\[-Wopenmp\\\]" "" { target *-*-* } .-3 } -! { dg-warning "38:Mapping polymorphic list item at .1. is unspecified behavior \\\[-Wopenmp\\\]" "" { target *-*-* } .-4 } - ! ------------------------- !$omp target parallel map(release: x) ! { dg-error "36:TARGET with map-type other than TO, FROM, TOFROM, or ALLOC on MAP clause" } diff --git a/gcc/testsuite/gfortran.dg/interface_59.f90 b/gcc/testsuite/gfortran.dg/interface_59.f90 new file mode 100644 index 0000000..c9ccd67 --- /dev/null +++ b/gcc/testsuite/gfortran.dg/interface_59.f90 @@ -0,0 +1,15 @@ +! { dg-do compile } +! PR fortran/119669 - this used to generate an ICE. + +program a + implicit real(a-h,o-z) + external abstract_caller, caller, func +! real func + call abstract_caller (caller, func, 1.5) + call abstract_caller (caller, func, 1.5) +end program a + +function func (x) + real func, x + func = x * x - 1. +end diff --git a/gcc/testsuite/gfortran.dg/pr119502.f90 b/gcc/testsuite/gfortran.dg/pr119502.f90 new file mode 100644 index 0000000..80d7c61 --- /dev/null +++ b/gcc/testsuite/gfortran.dg/pr119502.f90 @@ -0,0 +1,15 @@ +! { dg-do run } + +! PR119502, negative unit numbers are not allowed without using NEWUNIT + +program foo + integer :: iun = -1 + integer :: ios + open (iun, iostat=ios) + if (ios == 0) stop 1 + write(iun,*, iostat=ios) "This is a test." + if (ios == 0) stop 2 + close (iun, iostat=ios) + if (ios == 0) stop 3 +end + diff --git a/gcc/testsuite/gfortran.dg/pr119836_1.f90 b/gcc/testsuite/gfortran.dg/pr119836_1.f90 new file mode 100644 index 0000000..984e2d0 --- /dev/null +++ b/gcc/testsuite/gfortran.dg/pr119836_1.f90 @@ -0,0 +1,18 @@ +! +! { dg-do run } +! +! PR fortran/119836 +! +program p + implicit none + integer, parameter :: n = 4 + integer :: i + integer :: y(n), x(n) + do concurrent (i=1:n) + x(i) = shiftl (i,1) ! accepted + block + y(i) = shiftl (i,1) ! wrongly rejected + end block + end do + if (any(x /= y)) stop 1 +end program p diff --git a/gcc/testsuite/gfortran.dg/pr119836_2.f90 b/gcc/testsuite/gfortran.dg/pr119836_2.f90 new file mode 100644 index 0000000..5e2d0c9 --- /dev/null +++ b/gcc/testsuite/gfortran.dg/pr119836_2.f90 @@ -0,0 +1,21 @@ +! +! { dg-do compile } +! +! PR fortran/119836 +! +! Although intrinsic functions contained within the Fortran standard +! are pure procedures, many of the additional intrinsic functions +! supplied in libgfortran are impure. RAND() is one such function. +! +program foo + implicit none + integer i + real x(4) + do concurrent (i=1:4) + x = rand() ! { dg-error "Reference to impure function" } + block + x = rand() ! { dg-error "Reference to impure function" } + end block + end do + print *, x +end program foo diff --git a/gcc/testsuite/gfortran.dg/pr119836_3.f90 b/gcc/testsuite/gfortran.dg/pr119836_3.f90 new file mode 100644 index 0000000..69a5fcf --- /dev/null +++ b/gcc/testsuite/gfortran.dg/pr119836_3.f90 @@ -0,0 +1,30 @@ +! +! { dg-do run } +! +! PR fortran/119836 +! +program p + implicit none + integer, parameter :: n = 4 + integer :: i + integer :: y(n), x(n) + x = [(i,i=1,n)] + do concurrent (i=1:n) + call bar(x, y) + end do + if (any(x /= y)) stop 1 + x = 2 * x + do concurrent (i=1:n) + block + call bar(x, y) + end block + end do + if (any(x /= y)) stop 1 + + contains + elemental subroutine bar(x, y) + integer, intent(in) :: x + integer, intent(out) :: y + y = x + end subroutine +end program p diff --git a/gcc/testsuite/gfortran.dg/pr119836_4.f90 b/gcc/testsuite/gfortran.dg/pr119836_4.f90 new file mode 100644 index 0000000..dc6f72b --- /dev/null +++ b/gcc/testsuite/gfortran.dg/pr119836_4.f90 @@ -0,0 +1,30 @@ +! +! { dg-do compile } +! +! PR fortran/119836 +! +program p + implicit none + integer, parameter :: n = 4 + integer :: i + integer :: y(n), x(n) + x = [(i,i=1,n)] + do concurrent (i=1:n) + call bar(x, y) ! { dg-error "Subroutine call" } + end do + if (any(x /= y)) stop 1 + x = 2 * x + do concurrent (i=1:n) + block + call bar(x, y) ! { dg-error "Subroutine call" } + end block + end do + if (any(x /= y)) stop 1 + + contains + subroutine bar(x, y) + integer, intent(in) :: x(:) + integer, intent(out) :: y(:) + y = x + end subroutine +end program p diff --git a/gcc/testsuite/gfortran.dg/pure_formal_proc_4.f90 b/gcc/testsuite/gfortran.dg/pure_formal_proc_4.f90 new file mode 100644 index 0000000..92640e2 --- /dev/null +++ b/gcc/testsuite/gfortran.dg/pure_formal_proc_4.f90 @@ -0,0 +1,49 @@ +! { dg-do compile } +! PR fortran/106948 - check that passing of PURE procedures works +! +! Contributed by Jim Feng + +module a + implicit none + + interface new + pure module subroutine b(x, f) + integer, intent(inout) :: x + interface + pure function f(x) result(r) + real, intent(in) :: x + real :: r + end function f + end interface + end subroutine b + end interface new +end module a + +submodule(a) a_b + implicit none + +contains + module procedure b + x = int(f(real(x)) * 0.15) + end procedure b +end submodule a_b + +program test + use a + implicit none + + integer :: x + + x = 100 + call new(x, g) + print *, x + +contains + + pure function g(y) result(r) + real, intent(in) :: y + real :: r + + r = sqrt(y) + end function g +end program test diff --git a/gcc/testsuite/gm2.dg/doc/examples/pass/doc-examples-pass.exp b/gcc/testsuite/gm2.dg/doc/examples/pass/doc-examples-pass.exp new file mode 100644 index 0000000..0bfcea0 --- /dev/null +++ b/gcc/testsuite/gm2.dg/doc/examples/pass/doc-examples-pass.exp @@ -0,0 +1,18 @@ +# Compile tests, no torture testing. +# +# These tests should all pass. + +# Load support procs. +load_lib gm2-dg.exp + +gm2_init_pim4 $srcdir/$subdir + +# Initialize `dg'. +dg-init + +# Main loop. + +dg-runtest [lsort [glob -nocomplain $srcdir/$subdir/*.mod]] "" "" + +# All done. +dg-finish diff --git a/gcc/testsuite/gm2.dg/doc/examples/pass/exampleadd.mod b/gcc/testsuite/gm2.dg/doc/examples/pass/exampleadd.mod new file mode 100644 index 0000000..84020a8 --- /dev/null +++ b/gcc/testsuite/gm2.dg/doc/examples/pass/exampleadd.mod @@ -0,0 +1,32 @@ +(* { dg-do assemble { target { x86_64-*-* } } } *) +(* { dg-options "-g" } *) + +MODULE exampleadd ; + +FROM libc IMPORT printf, exit ; + + +PROCEDURE Example (foo, bar: CARDINAL) : CARDINAL ; +VAR + myout: CARDINAL ; +BEGIN + ASM VOLATILE ("movl %1,%%eax; addl %2,%%eax; movl %%eax,%0" + : "=rm" (myout) (* outputs *) + : "rm" (foo), "rm" (bar) (* inputs *) + : "eax") ; (* we trash *) + RETURN( myout ) +END Example ; + + +VAR + a, b, c: CARDINAL ; +BEGIN + a := 1 ; + b := 2 ; + c := Example (a, b) ; + IF c # 3 + THEN + printf ("Example procedure function failed to return 3, seen %d", c) ; + exit (1) + END +END exampleadd. diff --git a/gcc/testsuite/gm2.dg/doc/examples/pass/exampleadd2.mod b/gcc/testsuite/gm2.dg/doc/examples/pass/exampleadd2.mod new file mode 100644 index 0000000..f25397f --- /dev/null +++ b/gcc/testsuite/gm2.dg/doc/examples/pass/exampleadd2.mod @@ -0,0 +1,32 @@ +(* { dg-do assemble { target { x86_64-*-* } } } *) +(* { dg-options "-g" } *) + +MODULE exampleadd2 ; + +FROM libc IMPORT printf, exit ; + + +PROCEDURE Example (foo, bar: CARDINAL) : CARDINAL ; +VAR + myout: CARDINAL ; +BEGIN + ASM VOLATILE ( + "movl %[left],%%eax; addl %[right],%%eax; movl %%eax,%[output]" + : [output] "=rm" (myout) (* outputs *) + : [left] "rm" (foo), [right] "rm" (bar) (* inputs *) + : "eax") ; (* we trash *) + RETURN( myout ) +END Example ; + +VAR + a, b, c: CARDINAL ; +BEGIN + a := 1 ; + b := 2 ; + c := Example (a, b) ; + IF c # 3 + THEN + printf ("Example procedure function failed to return 3, seen %d", c) ; + exit (1) + END +END exampleadd2. diff --git a/gcc/testsuite/gm2.dg/doc/examples/pass/hello.mod b/gcc/testsuite/gm2.dg/doc/examples/pass/hello.mod new file mode 100644 index 0000000..f9770ec --- /dev/null +++ b/gcc/testsuite/gm2.dg/doc/examples/pass/hello.mod @@ -0,0 +1,10 @@ +(* { dg-do run } *) +(* { dg-options "-g -fno-scaffold-dynamic" } *) + +MODULE hello ; + +FROM libc IMPORT printf ; + +BEGIN + printf ("hello world\n") +END hello. diff --git a/gcc/testsuite/gm2.dg/doc/examples/pass/hellopim.mod b/gcc/testsuite/gm2.dg/doc/examples/pass/hellopim.mod new file mode 100644 index 0000000..b7876cd --- /dev/null +++ b/gcc/testsuite/gm2.dg/doc/examples/pass/hellopim.mod @@ -0,0 +1,10 @@ +(* { dg-do run } *) +(* { dg-options "-g -fno-scaffold-dynamic" } *) + +MODULE hellopim ; + +FROM StrIO IMPORT WriteString, WriteLn ; + +BEGIN + WriteString ("hello world") ; WriteLn +END hellopim. diff --git a/gcc/testsuite/gnat.dg/lto29.adb b/gcc/testsuite/gnat.dg/lto29.adb new file mode 100644 index 0000000..44f556f --- /dev/null +++ b/gcc/testsuite/gnat.dg/lto29.adb @@ -0,0 +1,9 @@ +-- { dg-do run } +-- { dg-options "-O -flto" { target lto } } + +with Lto29_Pkg; + +procedure Lto29 is +begin + null; +end; diff --git a/gcc/testsuite/gnat.dg/lto29_pkg.ads b/gcc/testsuite/gnat.dg/lto29_pkg.ads new file mode 100644 index 0000000..6008dc5 --- /dev/null +++ b/gcc/testsuite/gnat.dg/lto29_pkg.ads @@ -0,0 +1,15 @@ +with Ada.Strings.Bounded; + +package Lto29_Pkg is + + package M is new Ada.Strings.Bounded.Generic_Bounded_Length (10); + + type T is new M.Bounded_String; + + Null_T : constant T; + +private + + Null_T : constant T := To_Bounded_String (""); + +end Lto29_Pkg; diff --git a/gcc/testsuite/gnat.dg/opt105.adb b/gcc/testsuite/gnat.dg/opt105.adb new file mode 100644 index 0000000..eb2c197 --- /dev/null +++ b/gcc/testsuite/gnat.dg/opt105.adb @@ -0,0 +1,30 @@ +-- { dg-do run } +-- { dg-options "-O" } + +with Opt105_Pkg; use Opt105_Pkg; + +procedure Opt105 is + + Val : constant Enum := + (if Enabled then (if Disabled then Two else One) else Three); + +begin + if Cond1 then + return; + end if; + + if Cond2 then + return; + end if; + + case Val is + when One => + raise Program_Error; + + when Two => + raise Constraint_Error; + + when Three => + null; + end case; +end; diff --git a/gcc/testsuite/gnat.dg/opt105_pkg.adb b/gcc/testsuite/gnat.dg/opt105_pkg.adb new file mode 100644 index 0000000..e00de94 --- /dev/null +++ b/gcc/testsuite/gnat.dg/opt105_pkg.adb @@ -0,0 +1,6 @@ +package body Opt105_Pkg is + + function Cond1 return Boolean is (False); + function Cond2 return Boolean is (False); + +end Opt105_Pkg; diff --git a/gcc/testsuite/gnat.dg/opt105_pkg.ads b/gcc/testsuite/gnat.dg/opt105_pkg.ads new file mode 100644 index 0000000..2b373b7 --- /dev/null +++ b/gcc/testsuite/gnat.dg/opt105_pkg.ads @@ -0,0 +1,11 @@ +package Opt105_Pkg is + + type Enum is (One, Two, Three); + + Enabled : Boolean := False; + Disabled : Boolean := False; + + function Cond1 return Boolean; + function Cond2 return Boolean; + +end Opt105_Pkg; diff --git a/gcc/testsuite/gnat.dg/renaming17.adb b/gcc/testsuite/gnat.dg/renaming17.adb new file mode 100644 index 0000000..d826433 --- /dev/null +++ b/gcc/testsuite/gnat.dg/renaming17.adb @@ -0,0 +1,17 @@ +-- { dg-do run } + +procedure Renaming17 is + + function Incr (V : Integer; I : Integer := 1) return Integer is + (V + I); + + function Incr_Ren (V : Integer; I : Positive := 1) return Positive + renames Incr; + + I : Integer; + +begin + I := Incr_Ren (-3); + I := Incr_Ren (-3, 2); + I := Incr_Ren (-3, 0); +end; diff --git a/gcc/testsuite/go.dg/pr119533-riscv-2.go b/gcc/testsuite/go.dg/pr119533-riscv-2.go new file mode 100644 index 0000000..ce3ffaf --- /dev/null +++ b/gcc/testsuite/go.dg/pr119533-riscv-2.go @@ -0,0 +1,42 @@ +// { dg-do compile { target riscv64*-*-* } } +// { dg-options "-O2 -march=rv64gcv -mabi=lp64d" } + +package ast + +type as struct { + bt []struct{} + an string +} + +func bj(a *as) string { + if b := a.bt; len(a.an) == 1 { + _ = b[0] + } + return a.an +} + +func MergePackageFiles(f map[string][]interface{}, g uint) []interface{} { + bl := make([]string, len(f)) + var bo []interface{} + bu := make(map[string]int) + for _, bm := range bl { + a := f[bm] + for _, d := range a { + if g != 0 { + if a, p := d.(*as); p { + n := bj(a) + if j, bp := bu[n]; bp { + _ = j + } + } + } + } + } + for _, bm := range bl { + _ = bm + } + for _, bm := range bl { + _ = f[bm] + } + return bo +} diff --git a/gcc/testsuite/go.dg/pr119533-riscv.go b/gcc/testsuite/go.dg/pr119533-riscv.go new file mode 100644 index 0000000..30f52d2 --- /dev/null +++ b/gcc/testsuite/go.dg/pr119533-riscv.go @@ -0,0 +1,120 @@ +// { dg-do compile { target riscv64*-*-* } } +// { dg-options "-O2 -march=rv64gcv -mabi=lp64d" } + +// Reduced from libgo build (multi-file reduction, merged mnaully +// and hand reduced again). + +package ast +import ( + "go/token" + "go/scanner" + "reflect" +) +type v struct {} +type w func( string, reflect.Value) bool +func x( string, reflect.Value) bool +type r struct { + scanner.ErrorList +} +type ab interface {} +type ae interface {} +type af interface {} +type ag struct {} +func (ag) Pos() token.Pos +func (ag) ah() token.Pos +type c struct { + aj ae } +type ak struct { + al []c } +type ( + am struct { + an string } + bs struct { + Value string + } +) +func ao(string) *am +type ( + ap interface {} + aq struct { + ar bs } +as struct { + bt ak + an am } +) +type File struct { + *ag + token.Pos + *am + at []af + *v + au []*aq + av *am + aw []*ag } +type ax struct { + an string + *v + ay map[string]File } +func a(az *token.FileSet, b token.Pos) int +type k struct { + l token.Pos + ah token.Pos +} +type m struct { + bb bool + bc *ag +} + +type bi uint +func bj(a *as) string { + if b := a.bt; len(b.al) == 1 { + c := b.al[0].aj + if e := c; e != nil {} + } + return a.an.an +} +func MergePackageFiles(f ax, g bi) *File { + h := 0 + bk := 0 + k := 0 + bl := make([]string, len(f.ay)) + i := 0 + for bm, a := range f.ay { + bl[i] = bm + k += len(a.at) + } + var bn *ag + var l token.Pos + if h > 0 {} + var bo []af + bu := make(map[string]int) + m := 0 + for _, bm := range bl { + a := f.ay[bm] + for _, d := range a.at { + if g!= 0 { + if a, p := d.(*as); p { + n := bj(a) + if j, bp := bu[n]; bp { + if bo != nil && bo[j]== nil {} + } + } + } + } + } + if m > 0 {} + var bq []*aq + q := make(map[string]bool) + for _, bm := range bl { + a := f.ay[bm] + for _, br := range a.au { + if o := br.ar.Value; q[o] {} + } + } + var bh = make([]*ag, bk) + for _, bm := range bl { + a := f.ay[bm] + copy(bh, a.aw) + } + return &File{bn, l, ao(f.an), bo, f.v, bq, nil, bh} +} diff --git a/gcc/testsuite/lib/cobol.exp b/gcc/testsuite/lib/cobol.exp index 8ea91e3..723989e 100644 --- a/gcc/testsuite/lib/cobol.exp +++ b/gcc/testsuite/lib/cobol.exp @@ -122,6 +122,15 @@ proc cobol_link_flags { paths } { } append ld_library_path ":${gccpath}/libgcobol/.libs" } + if { [file exists "${gccpath}/libquadmath/.libs/libquadmath.a"] || + [file exists "${gccpath}/libquadmath/.libs/libquadmath.${shlib_ext}"] } { + if { $target_wants_B_option } { + append flags "-B${gccpath}/libquadmath/.libs " + } else { + append flags "-L${gccpath}/libquadmath/.libs " + } + append ld_library_path ":${gccpath}/libquadmath/.libs" + } if { [file exists "${gccpath}/libstdc++-v3/src/.libs/libstdc++.a"] || [file exists "${gccpath}/libstdc++-v3/src/.libs/libstdc++.${shlib_ext}"] } { if { $target_wants_B_option } { diff --git a/gcc/testsuite/lib/gcc-dg.exp b/gcc/testsuite/lib/gcc-dg.exp index eadc1cd..6dd8fa3 100644 --- a/gcc/testsuite/lib/gcc-dg.exp +++ b/gcc/testsuite/lib/gcc-dg.exp @@ -507,8 +507,7 @@ if { [info procs ${tool}_load] != [list] \ set linenum 1 set outfile [open [lindex ${output-file} 1]] set do_fail 0 - set name [file tail [lindex ${output-file} 1]] - verbose "output-file args is $args program is $program" 1 + set name [testname-for-summary] while { [gets $outfile line] >= 0 } { if { $linenum != 1 } { set c [string index $output $idx] diff --git a/gcc/testsuite/lib/target-supports.exp b/gcc/testsuite/lib/target-supports.exp index ee4138a..869d150 100644 --- a/gcc/testsuite/lib/target-supports.exp +++ b/gcc/testsuite/lib/target-supports.exp @@ -562,6 +562,16 @@ proc check_effective_target_elf { } { } } +# Returns 1 if the target uses the PE/COFF object format, 0 otherwise. + +proc check_effective_target_pe { } { + if { [gcc_target_object_format] == "pe" } { + return 1; + } else { + return 0; + } +} + # Returns 1 if the target toolchain supports ifunc, 0 otherwise. proc check_ifunc_available { } { @@ -13489,11 +13499,11 @@ proc check_effective_target_arm_v8_3a_bkey_directive { } { # Return 1 if the target supports executing the Armv8.1-M Mainline Low # Overhead Loop, 0 otherwise. The test is valid for ARM. -proc check_effective_target_arm_v8_1_lob_ok { } { +proc check_effective_target_arm_v8_1m_lob_hw { } { if { ![check_effective_target_arm_cortex_m] } { return 0; } else { - return [check_runtime arm_v8_1_lob_hw_available { + return [check_runtime arm_v8_1m_lob_hw_available { int main (void) { int i = 0; @@ -13513,9 +13523,9 @@ proc check_effective_target_arm_v8_1_lob_ok { } { # the Armv8.1-M Mainline Low Overhead Loop, 0 otherwise. The test is # valid for ARM. -proc check_effective_target_arm_thumb2_no_arm_v8_1_lob { } { +proc check_effective_target_arm_thumb2_no_arm_v8_1m_lob { } { if { [check_effective_target_arm_thumb2] - && ![check_effective_target_arm_v8_1_lob_ok] } { + && ![check_effective_target_arm_v8_1m_lob_hw] } { return 1 } return 0 @@ -13525,9 +13535,9 @@ proc check_effective_target_arm_thumb2_no_arm_v8_1_lob { } { # used and the target does not support executing the Armv8.1-M # Mainline Low Overhead Loop, 0 otherwise. The test is valid for ARM. -proc check_effective_target_arm_thumb2_ok_no_arm_v8_1_lob { } { +proc check_effective_target_arm_thumb2_ok_no_arm_v8_1m_lob { } { if { [check_effective_target_arm_thumb2_ok] - && ![check_effective_target_arm_v8_1_lob_ok] } { + && ![check_effective_target_arm_v8_1m_lob_hw] } { return 1 } return 0 diff --git a/gcc/testsuite/rust/compile/enum_discriminant2.rs b/gcc/testsuite/rust/compile/enum_discriminant2.rs new file mode 100644 index 0000000..351dfbb --- /dev/null +++ b/gcc/testsuite/rust/compile/enum_discriminant2.rs @@ -0,0 +1,9 @@ +fn test() -> isize { + 1 +} + +enum Foo { + Bar = test() // { dg-error "only functions marked as .const." } +} + +fn main() {} diff --git a/gcc/testsuite/rust/compile/format_args_extra_comma.rs b/gcc/testsuite/rust/compile/format_args_extra_comma.rs new file mode 100644 index 0000000..fcc435c --- /dev/null +++ b/gcc/testsuite/rust/compile/format_args_extra_comma.rs @@ -0,0 +1,47 @@ +#![feature(rustc_attrs)] + +#[rustc_builtin_macro] +macro_rules! format_args { + () => {}; +} + +#[lang = "sized"] +trait Sized {} + +pub mod core { + pub mod fmt { + pub struct Formatter; + pub struct Result; + + pub struct Arguments<'a>; + + impl<'a> Arguments<'a> { + pub fn new_v1(_: &'a [&'static str], _: &'a [ArgumentV1<'a>]) -> Arguments<'a> { + Arguments + } + } + + pub struct ArgumentV1<'a>; + + impl<'a> ArgumentV1<'a> { + pub fn new<'b, T>(_: &'b T, _: fn(&T, &mut Formatter) -> Result) -> ArgumentV1 { + ArgumentV1 + } + } + + pub trait Display { + fn fmt(&self, _: &mut Formatter) -> Result; + } + + impl Display for i32 { + fn fmt(&self, _: &mut Formatter) -> Result { + // { dg-warning "unused name .self." "" { target *-*-* } .-1 } + Result + } + } + } +} + +fn main() { + let _formatted = format_args!("extra commas {} {}", 15, 14,); +} diff --git a/gcc/testsuite/rust/compile/macros/mbe/macro-issue3693.rs b/gcc/testsuite/rust/compile/macros/mbe/macro-issue3693.rs new file mode 100644 index 0000000..e990c8b --- /dev/null +++ b/gcc/testsuite/rust/compile/macros/mbe/macro-issue3693.rs @@ -0,0 +1,10 @@ +macro_rules! generate_pattern_iterators { + { + $(#[$forward_iterator_attribute:meta])* + } => { + } +} + +generate_pattern_iterators! { + /// Created with the method [`split`]. +} diff --git a/gcc/testsuite/rust/compile/macros/mbe/macro-issue3708.rs b/gcc/testsuite/rust/compile/macros/mbe/macro-issue3708.rs new file mode 100644 index 0000000..e5b38bb --- /dev/null +++ b/gcc/testsuite/rust/compile/macros/mbe/macro-issue3708.rs @@ -0,0 +1,80 @@ +// { dg-additional-options "-frust-name-resolution-2.0 -frust-compile-until=lowering" } + +macro_rules! impl_fn_for_zst { + ($( + $( #[$attr: meta] )* + struct $Name: ident impl$( <$( $lifetime : lifetime ),+> )? Fn = + |$( $arg: ident: $ArgTy: ty ),*| -> $ReturnTy: ty + $body: block; + )+) => { + $( + $( #[$attr] )* + struct $Name; + + impl $( <$( $lifetime ),+> )? Fn<($( $ArgTy, )*)> for $Name { + #[inline] + extern "rust-call" fn call(&self, ($( $arg, )*): ($( $ArgTy, )*)) -> $ReturnTy { + $body + } + } + + impl $( <$( $lifetime ),+> )? FnMut<($( $ArgTy, )*)> for $Name { + #[inline] + extern "rust-call" fn call_mut( + &mut self, + ($( $arg, )*): ($( $ArgTy, )*) + ) -> $ReturnTy { + Fn::call(&*self, ($( $arg, )*)) + } + } + + impl $( <$( $lifetime ),+> )? FnOnce<($( $ArgTy, )*)> for $Name { + type Output = $ReturnTy; + + #[inline] + extern "rust-call" fn call_once(self, ($( $arg, )*): ($( $ArgTy, )*)) -> $ReturnTy { + Fn::call(&self, ($( $arg, )*)) + } + } + )+ + } +} + +#[lang = "sized"] +trait Sized {} + +#[lang = "copy"] +trait Copy {} + +#[lang = "fn"] +pub trait Fn<Args>: FnMut<Args> { + /// Performs the call operation. + #[unstable(feature = "fn_traits", issue = "29625")] + extern "rust-call" fn call(&self, args: Args) -> Self::Output; +} + +#[lang = "fn_mut"] +#[must_use = "closures are lazy and do nothing unless called"] +pub trait FnMut<Args>: FnOnce<Args> { + /// Performs the call operation. + #[unstable(feature = "fn_traits", issue = "29625")] + extern "rust-call" fn call_mut(&mut self, args: Args) -> Self::Output; +} + +#[lang = "fn_once"] +pub trait FnOnce<Args> { + /// The returned type after the call operator is used. + #[lang = "fn_once_output"] + #[stable(feature = "fn_once_output", since = "1.12.0")] + type Output; + + /// Performs the call operation. + #[unstable(feature = "fn_traits", issue = "29625")] + extern "rust-call" fn call_once(self, args: Args) -> Self::Output; +} + +impl_fn_for_zst! { + #[derive(Copy)] + struct LinesAnyMap impl<'a> Fn = |line: &'a str| -> () { + }; +} diff --git a/gcc/testsuite/rust/compile/macros/mbe/macro-issue3709-1.rs b/gcc/testsuite/rust/compile/macros/mbe/macro-issue3709-1.rs new file mode 100644 index 0000000..6fc3a31 --- /dev/null +++ b/gcc/testsuite/rust/compile/macros/mbe/macro-issue3709-1.rs @@ -0,0 +1,10 @@ +macro_rules! doc_comment { + (#[ $attr: meta ]) => { + #[$attr] + struct Generated; // { dg-warning "never constructed" } + }; +} + +doc_comment! { + /// This is a generated struct +} diff --git a/gcc/testsuite/rust/compile/macros/mbe/macro-issue3709-2.rs b/gcc/testsuite/rust/compile/macros/mbe/macro-issue3709-2.rs new file mode 100644 index 0000000..cfc8ab4 --- /dev/null +++ b/gcc/testsuite/rust/compile/macros/mbe/macro-issue3709-2.rs @@ -0,0 +1,81 @@ +// { dg-additional-options "-frust-name-resolution-2.0 -frust-compile-until=lowering" } + +macro_rules! impl_fn_for_zst { + ($( + $( #[$attr: meta] )* + struct $Name: ident impl$( <$( $lifetime : lifetime ),+> )? Fn = + |$( $arg: ident: $ArgTy: ty ),*| -> $ReturnTy: ty + $body: block; + )+) => { + $( + $( #[$attr] )* + struct $Name; + + impl $( <$( $lifetime ),+> )? Fn<($( $ArgTy, )*)> for $Name { + #[inline] + extern "rust-call" fn call(&self, ($( $arg, )*): ($( $ArgTy, )*)) -> $ReturnTy { + $body + } + } + + impl $( <$( $lifetime ),+> )? FnMut<($( $ArgTy, )*)> for $Name { + #[inline] + extern "rust-call" fn call_mut( + &mut self, + ($( $arg, )*): ($( $ArgTy, )*) + ) -> $ReturnTy { + Fn::call(&*self, ($( $arg, )*)) + } + } + + impl $( <$( $lifetime ),+> )? FnOnce<($( $ArgTy, )*)> for $Name { + type Output = $ReturnTy; + + #[inline] + extern "rust-call" fn call_once(self, ($( $arg, )*): ($( $ArgTy, )*)) -> $ReturnTy { + Fn::call(&self, ($( $arg, )*)) + } + } + )+ + } +} + +#[lang = "sized"] +trait Sized {} + +#[lang = "copy"] +trait Copy {} + +#[lang = "fn"] +pub trait Fn<Args>: FnMut<Args> { + /// Performs the call operation. + #[unstable(feature = "fn_traits", issue = "29625")] + extern "rust-call" fn call(&self, args: Args) -> Self::Output; +} + +#[lang = "fn_mut"] +#[must_use = "closures are lazy and do nothing unless called"] +pub trait FnMut<Args>: FnOnce<Args> { + /// Performs the call operation. + #[unstable(feature = "fn_traits", issue = "29625")] + extern "rust-call" fn call_mut(&mut self, args: Args) -> Self::Output; +} + +#[lang = "fn_once"] +pub trait FnOnce<Args> { + /// The returned type after the call operator is used. + #[lang = "fn_once_output"] + #[stable(feature = "fn_once_output", since = "1.12.0")] + type Output; + + /// Performs the call operation. + #[unstable(feature = "fn_traits", issue = "29625")] + extern "rust-call" fn call_once(self, args: Args) -> Self::Output; +} + +impl_fn_for_zst! { + /// Documentation for the zst + #[derive(Copy)] + struct LinesAnyMap impl<'a> Fn = |line: &'a str| -> () { + }; +} diff --git a/gcc/testsuite/rust/compile/nr2/compile.exp b/gcc/testsuite/rust/compile/nr2/compile.exp index 4d91dd0..9e15cdd 100644 --- a/gcc/testsuite/rust/compile/nr2/compile.exp +++ b/gcc/testsuite/rust/compile/nr2/compile.exp @@ -19,6 +19,15 @@ # Load support procs. load_lib rust-dg.exp +# These tests don't run runtest_file_p consistently if it +# doesn't return the same values, so disable parallelization +# of this *.exp file. The first parallel runtest to reach +# this will run all the tests serially. +if ![gcc_parallel_test_run_p compile] { + return +} +gcc_parallel_test_enable 0 + # Initialize `dg'. dg-init @@ -136,3 +145,5 @@ namespace eval rust-nr2-ns { # All done. dg-finish + +gcc_parallel_test_enable 1 diff --git a/gcc/testsuite/rust/compile/track_caller.rs b/gcc/testsuite/rust/compile/track_caller.rs new file mode 100644 index 0000000..fd1d842 --- /dev/null +++ b/gcc/testsuite/rust/compile/track_caller.rs @@ -0,0 +1,6 @@ +#[track_caller] +fn foo() {} + +fn main() { + foo(); +} diff --git a/gcc/testsuite/rust/execute/torture/min_specialization2.rs b/gcc/testsuite/rust/execute/torture/min_specialization2.rs new file mode 100644 index 0000000..d3239ee --- /dev/null +++ b/gcc/testsuite/rust/execute/torture/min_specialization2.rs @@ -0,0 +1,31 @@ +#![feature(min_specialization)] + +#[lang = "sized"] +trait Sized {} + +trait Foo { + fn foo(&self) -> i32; +} + +impl<T> Foo for T { + default fn foo(&self) -> i32 { // { dg-warning "unused" } + 15 + } +} + +impl Foo for bool { + fn foo(&self) -> i32 { + if *self { + 1 + } else { + 0 + } + } +} + +fn main() -> i32 { + let a = 1.foo() - 15; + let b = true.foo() - 1; + + a + b +} diff --git a/gcc/testsuite/rust/execute/torture/min_specialization3.rs b/gcc/testsuite/rust/execute/torture/min_specialization3.rs new file mode 100644 index 0000000..9eccd97 --- /dev/null +++ b/gcc/testsuite/rust/execute/torture/min_specialization3.rs @@ -0,0 +1,36 @@ +#![feature(min_specialization)] + +#[lang = "sized"] +trait Sized {} + +trait Foo { + fn foo(&self) -> i32; +} + +struct Wrap<T>(T); + +impl<T> Foo for T { + default fn foo(&self) -> i32 { + 15 + } +} + +impl<T> Foo for Wrap<T> { + default fn foo(&self) -> i32 { + 16 + } +} + +impl Foo for Wrap<bool> { + fn foo(&self) -> i32 { + if self.0 { + 1 + } else { + 0 + } + } +} + +fn main() -> i32 { + Wrap(true).foo() - 1 +} diff --git a/gcc/timevar.def b/gcc/timevar.def index c1029d9..02ace46 100644 --- a/gcc/timevar.def +++ b/gcc/timevar.def @@ -105,6 +105,7 @@ DEFTIMEVAR (TV_IPA_PURE_CONST , "ipa pure const") DEFTIMEVAR (TV_IPA_ICF , "ipa icf") DEFTIMEVAR (TV_IPA_PTA , "ipa points-to") DEFTIMEVAR (TV_IPA_SRA , "ipa SRA") +DEFTIMEVAR (TV_IPA_LC , "ipa locality clone") DEFTIMEVAR (TV_IPA_FREE_LANG_DATA , "ipa free lang data") DEFTIMEVAR (TV_IPA_FREE_INLINE_SUMMARY, "ipa free inline summary") DEFTIMEVAR (TV_IPA_MODREF , "ipa modref") diff --git a/gcc/tree-data-ref.cc b/gcc/tree-data-ref.cc index 18d9f0f..8a46e6d 100644 --- a/gcc/tree-data-ref.cc +++ b/gcc/tree-data-ref.cc @@ -2498,9 +2498,10 @@ create_waw_or_war_checks (tree *cond_expr, limit = fold_build2 (PLUS_EXPR, sizetype, limit, size_int (last_chunk_a + last_chunk_b)); - tree subject = fold_build2 (POINTER_DIFF_EXPR, ssizetype, addr_b, addr_a); - subject = fold_build2 (PLUS_EXPR, sizetype, - fold_convert (sizetype, subject), bias); + tree subject = fold_build2 (MINUS_EXPR, sizetype, + fold_convert (sizetype, addr_b), + fold_convert (sizetype, addr_a)); + subject = fold_build2 (PLUS_EXPR, sizetype, subject, bias); *cond_expr = fold_build2 (GT_EXPR, boolean_type_node, subject, limit); if (dump_enabled_p ()) diff --git a/gcc/tree-inline.cc b/gcc/tree-inline.cc index 05843b8..3289b4f 100644 --- a/gcc/tree-inline.cc +++ b/gcc/tree-inline.cc @@ -2729,8 +2729,11 @@ copy_edges_for_bb (basic_block bb, profile_count num, profile_count den, && gimple_call_arg (copy_stmt, 0) == boolean_true_node) nonlocal_goto = false; else - make_single_succ_edge (copy_stmt_bb, abnormal_goto_dest, - EDGE_ABNORMAL); + { + make_single_succ_edge (copy_stmt_bb, abnormal_goto_dest, + EDGE_ABNORMAL); + gimple_call_set_ctrl_altering (copy_stmt, true); + } } if ((can_throw || nonlocal_goto) diff --git a/gcc/tree-pass.h b/gcc/tree-pass.h index 217c31f..7cb5a12 100644 --- a/gcc/tree-pass.h +++ b/gcc/tree-pass.h @@ -551,6 +551,7 @@ extern ipa_opt_pass_d *make_pass_ipa_cdtor_merge (gcc::context *ctxt); extern ipa_opt_pass_d *make_pass_ipa_single_use (gcc::context *ctxt); extern ipa_opt_pass_d *make_pass_ipa_comdats (gcc::context *ctxt); extern ipa_opt_pass_d *make_pass_ipa_modref (gcc::context *ctxt); +extern ipa_opt_pass_d *make_pass_ipa_locality_cloning (gcc::context *ctxt); extern gimple_opt_pass *make_pass_cleanup_cfg_post_optimizing (gcc::context *ctxt); diff --git a/gcc/tree-pretty-print.cc b/gcc/tree-pretty-print.cc index a2a4f5b..c1a21e7 100644 --- a/gcc/tree-pretty-print.cc +++ b/gcc/tree-pretty-print.cc @@ -3201,6 +3201,8 @@ dump_generic_node (pretty_printer *pp, tree node, int spc, dump_flags_t flags, pp_string (pp, " [return slot optimization]"); if (CALL_EXPR_TAILCALL (node)) pp_string (pp, " [tail call]"); + if (CALL_EXPR_MUST_TAIL_CALL (node)) + pp_string (pp, " [must tail call]"); break; case WITH_CLEANUP_EXPR: diff --git a/gcc/tree-ssa-coalesce.cc b/gcc/tree-ssa-coalesce.cc index b78ffd7..7cc39f7 100644 --- a/gcc/tree-ssa-coalesce.cc +++ b/gcc/tree-ssa-coalesce.cc @@ -896,6 +896,18 @@ build_ssa_conflict_graph (tree_live_info_p liveinfo) tree var; gimple *stmt = gsi_stmt (gsi); + if (is_gimple_debug (stmt)) + continue; + + if (map->bitint) + { + build_bitint_stmt_ssa_conflicts (stmt, live, graph, map->bitint, + live_track_process_def, + live_track_process_use, + live_track_clear_var); + continue; + } + /* A copy between 2 partitions does not introduce an interference by itself. If they did, you would never be able to coalesce two things which are copied. If the two variables really do @@ -912,16 +924,6 @@ build_ssa_conflict_graph (tree_live_info_p liveinfo) && TREE_CODE (rhs1) == SSA_NAME) live_track_clear_var (live, rhs1); } - else if (is_gimple_debug (stmt)) - continue; - - if (map->bitint) - { - build_bitint_stmt_ssa_conflicts (stmt, live, graph, map->bitint, - live_track_process_def, - live_track_process_use); - continue; - } /* For stmts with more than one SSA_NAME definition pretend all the SSA_NAME outputs but the first one are live at this point, so diff --git a/gcc/tree-ssa-dse.cc b/gcc/tree-ssa-dse.cc index bc632e3..d1d58bf 100644 --- a/gcc/tree-ssa-dse.cc +++ b/gcc/tree-ssa-dse.cc @@ -566,16 +566,17 @@ maybe_trim_complex_store (ao_ref *ref, sbitmap live, gimple *stmt) The most common case for getting here is a CONSTRUCTOR with no elements being used to zero initialize an object. We do not try to handle other cases as those would force us to fully cover the object with the - CONSTRUCTOR node except for the components that are dead. */ + CONSTRUCTOR node except for the components that are dead. + Also handles integer stores of 0 which can happen with memset/memcpy optimizations. */ static void -maybe_trim_constructor_store (ao_ref *ref, sbitmap live, gimple *stmt) +maybe_trim_constructor_store (ao_ref *ref, sbitmap live, gimple *stmt, bool was_integer_cst) { tree ctor = gimple_assign_rhs1 (stmt); /* This is the only case we currently handle. It actually seems to catch most cases of actual interest. */ - gcc_assert (CONSTRUCTOR_NELTS (ctor) == 0); + gcc_assert (was_integer_cst ? integer_zerop (ctor) : CONSTRUCTOR_NELTS (ctor) == 0); int head_trim = 0; int tail_trim = 0; @@ -588,6 +589,8 @@ maybe_trim_constructor_store (ao_ref *ref, sbitmap live, gimple *stmt) /* We want &lhs for the MEM_REF expression. */ tree lhs_addr = build_fold_addr_expr (gimple_assign_lhs (stmt)); + STRIP_USELESS_TYPE_CONVERSION (lhs_addr); + if (! is_gimple_min_invariant (lhs_addr)) return; @@ -802,11 +805,16 @@ maybe_trim_partially_dead_store (ao_ref *ref, sbitmap live, gimple *stmt) switch (gimple_assign_rhs_code (stmt)) { case CONSTRUCTOR: - maybe_trim_constructor_store (ref, live, stmt); + maybe_trim_constructor_store (ref, live, stmt, false); break; case COMPLEX_CST: maybe_trim_complex_store (ref, live, stmt); break; + case INTEGER_CST: + if (integer_zerop (gimple_assign_rhs1 (stmt)) + && type_has_mode_precision_p (TREE_TYPE (gimple_assign_lhs (stmt)))) + maybe_trim_constructor_store (ref, live, stmt, true); + break; default: break; } diff --git a/gcc/tree-ssa-phiopt.cc b/gcc/tree-ssa-phiopt.cc index 7d2d169..a194bf6 100644 --- a/gcc/tree-ssa-phiopt.cc +++ b/gcc/tree-ssa-phiopt.cc @@ -403,12 +403,15 @@ factor_out_conditional_operation (edge e0, edge e1, basic_block merge, if (dominated_by_p (CDI_DOMINATORS, gimple_bb (phi), gimple_bb (arg0_def_stmt))) return false; - /* Only handle if arg1 is a INTEGER_CST and one that fits - into the new type or if it is the same precision. */ + /* If arg1 is an INTEGER_CST, fold it to new type if it fits, or else + if the bits will not be modified during the conversion, except for + boolean types whose precision is not 1 (see int_fits_type_p). */ if (!INTEGRAL_TYPE_P (TREE_TYPE (new_arg0)) || !(int_fits_type_p (arg1, TREE_TYPE (new_arg0)) || (TYPE_PRECISION (TREE_TYPE (new_arg0)) - == TYPE_PRECISION (TREE_TYPE (arg1))))) + == TYPE_PRECISION (TREE_TYPE (arg1)) + && (TREE_CODE (TREE_TYPE (new_arg0)) != BOOLEAN_TYPE + || TYPE_PRECISION (TREE_TYPE (new_arg0)) == 1)))) return false; /* For the INTEGER_CST case, we are just moving the diff --git a/gcc/tree-tailcall.cc b/gcc/tree-tailcall.cc index f51bb97..f593363 100644 --- a/gcc/tree-tailcall.cc +++ b/gcc/tree-tailcall.cc @@ -51,6 +51,8 @@ along with GCC; see the file COPYING3. If not see #include "symbol-summary.h" #include "ipa-cp.h" #include "ipa-prop.h" +#include "attribs.h" +#include "asan.h" /* The file implements the tail recursion elimination. It is also used to analyze the tail calls in general, passing the results to the rtl level @@ -122,6 +124,9 @@ struct tailcall /* True if it is a call to the current function. */ bool tail_recursion; + /* True if there is __tsan_func_exit call after the call. */ + bool has_tsan_func_exit; + /* The return value of the caller is mult * f + add, where f is the return value of the call. */ tree mult, add; @@ -492,10 +497,10 @@ maybe_error_musttail (gcall *call, const char *err, bool diag_musttail) gimple_call_set_must_tail (call, false); /* Avoid another error. */ gimple_call_set_tail (call, false); } - if (dump_file) + if (dump_file && (dump_flags & TDF_DETAILS)) { + fprintf (dump_file, "Cannot tail-call: %s: ", err); print_gimple_stmt (dump_file, call, 0, TDF_SLIM); - fprintf (dump_file, "Cannot convert: %s\n", err); } } @@ -504,7 +509,7 @@ maybe_error_musttail (gcall *call, const char *err, bool diag_musttail) Search at most CNT basic blocks (so that we don't need to do trivial loop discovery). */ static bool -empty_eh_cleanup (basic_block bb, int cnt) +empty_eh_cleanup (basic_block bb, int *eh_has_tsan_func_exit, int cnt) { if (EDGE_COUNT (bb->succs) > 1) return false; @@ -515,6 +520,14 @@ empty_eh_cleanup (basic_block bb, int cnt) gimple *g = gsi_stmt (gsi); if (is_gimple_debug (g) || gimple_clobber_p (g)) continue; + if (eh_has_tsan_func_exit + && !*eh_has_tsan_func_exit + && sanitize_flags_p (SANITIZE_THREAD) + && gimple_call_builtin_p (g, BUILT_IN_TSAN_FUNC_EXIT)) + { + *eh_has_tsan_func_exit = 1; + continue; + } if (is_gimple_resx (g) && stmt_can_throw_external (cfun, g)) return true; return false; @@ -523,7 +536,7 @@ empty_eh_cleanup (basic_block bb, int cnt) return false; if (cnt == 1) return false; - return empty_eh_cleanup (single_succ (bb), cnt - 1); + return empty_eh_cleanup (single_succ (bb), eh_has_tsan_func_exit, cnt - 1); } /* Argument for compute_live_vars/live_vars_at_stmt and what compute_live_vars @@ -531,14 +544,22 @@ empty_eh_cleanup (basic_block bb, int cnt) static live_vars_map *live_vars; static vec<bitmap_head> live_vars_vec; -/* Finds tailcalls falling into basic block BB. The list of found tailcalls is +/* Finds tailcalls falling into basic block BB. The list of found tailcalls is added to the start of RET. When ONLY_MUSTTAIL is set only handle musttail. Update OPT_TAILCALLS as output parameter. If DIAG_MUSTTAIL, diagnose - failures for musttail calls. */ + failures for musttail calls. RETRY_TSAN_FUNC_EXIT is initially 0 and + in that case the last call is attempted to be tail called, including + __tsan_func_exit with -fsanitize=thread. It is set to -1 if we + detect __tsan_func_exit call and in that case tree_optimize_tail_calls_1 + will retry with it set to 1 (regardless of whether turning the + __tsan_func_exit was successfully detected as tail call or not) and that + will allow turning musttail calls before that call into tail calls as well + by adding __tsan_func_exit call before the call. */ static void find_tail_calls (basic_block bb, struct tailcall **ret, bool only_musttail, - bool &opt_tailcalls, bool diag_musttail) + bool &opt_tailcalls, bool diag_musttail, + int &retry_tsan_func_exit) { tree ass_var = NULL_TREE, ret_var, func, param; gimple *stmt; @@ -552,6 +573,8 @@ find_tail_calls (basic_block bb, struct tailcall **ret, bool only_musttail, size_t idx; tree var; bool only_tailr = false; + bool has_tsan_func_exit = false; + int eh_has_tsan_func_exit = -1; if (!single_succ_p (bb) && (EDGE_COUNT (bb->succs) || !cfun->has_musttail || !diag_musttail)) @@ -585,6 +608,17 @@ find_tail_calls (basic_block bb, struct tailcall **ret, bool only_musttail, || is_gimple_debug (stmt)) continue; + if (cfun->has_musttail + && sanitize_flags_p (SANITIZE_THREAD) + && gimple_call_builtin_p (stmt, BUILT_IN_TSAN_FUNC_EXIT) + && diag_musttail) + { + if (retry_tsan_func_exit == 0) + retry_tsan_func_exit = -1; + else if (retry_tsan_func_exit == 1) + continue; + } + if (!last_stmt) last_stmt = stmt; /* Check for a call. */ @@ -596,9 +630,8 @@ find_tail_calls (basic_block bb, struct tailcall **ret, bool only_musttail, return; if (bad_stmt) { - maybe_error_musttail (call, - _("memory reference or volatile after " - "call"), diag_musttail); + maybe_error_musttail (call, _("memory reference or volatile " + "after call"), diag_musttail); return; } ass_var = gimple_call_lhs (call); @@ -636,7 +669,7 @@ find_tail_calls (basic_block bb, struct tailcall **ret, bool only_musttail, /* Recurse to the predecessors. */ FOR_EACH_EDGE (e, ei, bb->preds) find_tail_calls (e->src, ret, only_musttail, opt_tailcalls, - diag_musttail); + diag_musttail, retry_tsan_func_exit); return; } @@ -711,19 +744,22 @@ find_tail_calls (basic_block bb, struct tailcall **ret, bool only_musttail, if (!e) { - maybe_error_musttail (call, - _("call may throw exception that does not " - "propagate"), diag_musttail); + maybe_error_musttail (call, _("call may throw exception that does not " + "propagate"), diag_musttail); return; } + if (diag_musttail && gimple_call_must_tail_p (call)) + eh_has_tsan_func_exit = 0; if (!gimple_call_must_tail_p (call) - || !empty_eh_cleanup (e->dest, 20) + || !empty_eh_cleanup (e->dest, + eh_has_tsan_func_exit + ? NULL : &eh_has_tsan_func_exit, 20) || EDGE_COUNT (bb->succs) > 2) { - maybe_error_musttail (call, - _("call may throw exception caught locally " - "or perform cleanups"), diag_musttail); + maybe_error_musttail (call, _("call may throw exception caught " + "locally or perform cleanups"), + diag_musttail); return; } } @@ -854,9 +890,8 @@ find_tail_calls (basic_block bb, struct tailcall **ret, bool only_musttail, } if (local_live_vars) BITMAP_FREE (local_live_vars); - maybe_error_musttail (call, - _("call invocation refers to locals"), - diag_musttail); + maybe_error_musttail (call, _("call invocation refers to " + "locals"), diag_musttail); return; } else @@ -883,9 +918,8 @@ find_tail_calls (basic_block bb, struct tailcall **ret, bool only_musttail, continue; } BITMAP_FREE (local_live_vars); - maybe_error_musttail (call, - _("call invocation refers to locals"), - diag_musttail); + maybe_error_musttail (call, _("call invocation refers to " + "locals"), diag_musttail); return; } } @@ -951,6 +985,17 @@ find_tail_calls (basic_block bb, struct tailcall **ret, bool only_musttail, || is_gimple_debug (stmt)) continue; + if (cfun->has_musttail + && sanitize_flags_p (SANITIZE_THREAD) + && retry_tsan_func_exit == 1 + && gimple_call_builtin_p (stmt, BUILT_IN_TSAN_FUNC_EXIT) + && !has_tsan_func_exit + && gimple_call_must_tail_p (call)) + { + has_tsan_func_exit = true; + continue; + } + if (gimple_code (stmt) != GIMPLE_ASSIGN) { maybe_error_musttail (call, _("unhandled code after call"), @@ -1020,8 +1065,7 @@ find_tail_calls (basic_block bb, struct tailcall **ret, bool only_musttail, if (!VOID_TYPE_P (rettype) && !useless_type_conversion_p (rettype, calltype)) { - maybe_error_musttail (call, - _("call and return value are different"), + maybe_error_musttail (call, _("call and return value are different"), diag_musttail); return; } @@ -1092,8 +1136,7 @@ find_tail_calls (basic_block bb, struct tailcall **ret, bool only_musttail, } if (!ok) { - maybe_error_musttail (call, - _("call and return value are different"), + maybe_error_musttail (call, _("call and return value are different"), diag_musttail); return; } @@ -1103,18 +1146,29 @@ find_tail_calls (basic_block bb, struct tailcall **ret, bool only_musttail, multiplicands. */ if (!tail_recursion && (m || a)) { - maybe_error_musttail (call, - _("operations after non tail recursive call"), - diag_musttail); + maybe_error_musttail (call, _("operations after non tail recursive " + "call"), diag_musttail); return; } /* For pointers only allow additions. */ if (m && POINTER_TYPE_P (TREE_TYPE (DECL_RESULT (current_function_decl)))) { - maybe_error_musttail (call, - _("tail recursion with pointers can only use " - "additions"), diag_musttail); + maybe_error_musttail (call, _("tail recursion with pointers can only " + "use additions"), diag_musttail); + return; + } + + if (eh_has_tsan_func_exit != -1 + && eh_has_tsan_func_exit != has_tsan_func_exit) + { + if (eh_has_tsan_func_exit) + maybe_error_musttail (call, _("call may throw exception caught " + "locally or perform cleanups"), + diag_musttail); + else + maybe_error_musttail (call, _("exception cleanups omit " + "__tsan_func_exit call"), diag_musttail); return; } @@ -1146,6 +1200,7 @@ find_tail_calls (basic_block bb, struct tailcall **ret, bool only_musttail, nw->call_gsi = gsi; nw->tail_recursion = tail_recursion; + nw->has_tsan_func_exit = has_tsan_func_exit; nw->mult = m; nw->add = a; @@ -1480,6 +1535,14 @@ static bool optimize_tail_call (struct tailcall *t, bool opt_tailcalls, class loop *&new_loop) { + if (t->has_tsan_func_exit && (t->tail_recursion || opt_tailcalls)) + { + tree builtin_decl = builtin_decl_implicit (BUILT_IN_TSAN_FUNC_EXIT); + gimple *g = gimple_build_call (builtin_decl, 0); + gimple_set_location (g, cfun->function_end_locus); + gsi_insert_before (&t->call_gsi, g, GSI_SAME_STMT); + } + if (t->tail_recursion) { eliminate_tail_call (t, new_loop); @@ -1498,6 +1561,7 @@ optimize_tail_call (struct tailcall *t, bool opt_tailcalls, print_gimple_stmt (dump_file, stmt, 0, dump_flags); fprintf (dump_file, " in bb %i\n", (gsi_bb (t->call_gsi))->index); } + return t->has_tsan_func_exit; } return false; @@ -1547,12 +1611,23 @@ tree_optimize_tail_calls_1 (bool opt_tailcalls, bool only_musttail, /* Only traverse the normal exits, i.e. those that end with return statement. */ if (safe_is_a <greturn *> (*gsi_last_bb (e->src))) - find_tail_calls (e->src, &tailcalls, only_musttail, opt_tailcalls, - diag_musttail); + { + int retry_tsan_func_exit = 0; + find_tail_calls (e->src, &tailcalls, only_musttail, opt_tailcalls, + diag_musttail, retry_tsan_func_exit); + if (retry_tsan_func_exit == -1) + { + retry_tsan_func_exit = 1; + find_tail_calls (e->src, &tailcalls, only_musttail, + opt_tailcalls, diag_musttail, + retry_tsan_func_exit); + } + } } if (cfun->has_musttail && diag_musttail) { basic_block bb; + int retry_tsan_func_exit = 0; FOR_EACH_BB_FN (bb, cfun) if (EDGE_COUNT (bb->succs) == 0 || (single_succ_p (bb) @@ -1562,7 +1637,7 @@ tree_optimize_tail_calls_1 (bool opt_tailcalls, bool only_musttail, && gimple_call_must_tail_p (as_a <gcall *> (c)) && gimple_call_noreturn_p (as_a <gcall *> (c))) find_tail_calls (bb, &tailcalls, only_musttail, opt_tailcalls, - diag_musttail); + diag_musttail, retry_tsan_func_exit); } if (live_vars) @@ -1594,10 +1669,10 @@ tree_optimize_tail_calls_1 (bool opt_tailcalls, bool only_musttail, struct tailcall *a = *p; *p = (*p)->next; gcall *call = as_a <gcall *> (gsi_stmt (a->call_gsi)); - maybe_error_musttail (call, - _("tail recursion with accumulation " - "mixed with musttail " - "non-recursive call"), diag_musttail); + maybe_error_musttail (call, _("tail recursion with " + "accumulation mixed with " + "musttail non-recursive call"), + diag_musttail); free (a); } else diff --git a/gcc/tree-vect-loop.cc b/gcc/tree-vect-loop.cc index 9413dce..2d35fa1 100644 --- a/gcc/tree-vect-loop.cc +++ b/gcc/tree-vect-loop.cc @@ -653,6 +653,10 @@ vect_analyze_scalar_cycles_1 (loop_vec_info loop_vinfo, class loop *loop, if (dump_enabled_p ()) dump_printf_loc (MSG_NOTE, vect_location, "Detected induction.\n"); STMT_VINFO_DEF_TYPE (stmt_vinfo) = vect_induction_def; + + /* Mark if we have a non-linear IV. */ + LOOP_VINFO_NON_LINEAR_IV (loop_vinfo) + = STMT_VINFO_LOOP_PHI_EVOLUTION_TYPE (stmt_vinfo) != vect_step_op_add; } @@ -1046,12 +1050,14 @@ _loop_vec_info::_loop_vec_info (class loop *loop_in, vec_info_shared *shared) suggested_unroll_factor (1), max_vectorization_factor (0), mask_skip_niters (NULL_TREE), + mask_skip_niters_pfa_offset (NULL_TREE), rgroup_compare_type (NULL_TREE), simd_if_cond (NULL_TREE), partial_vector_style (vect_partial_vectors_none), unaligned_dr (NULL), peeling_for_alignment (0), ptr_mask (0), + nonlinear_iv (false), ivexpr_map (NULL), scan_map (NULL), slp_unrolling_factor (1), @@ -10678,6 +10684,54 @@ vectorizable_induction (loop_vec_info loop_vinfo, LOOP_VINFO_MASK_SKIP_NITERS (loop_vinfo)); peel_mul = gimple_build_vector_from_val (&init_stmts, step_vectype, peel_mul); + + /* If early break then we have to create a new PHI which we can use as + an offset to adjust the induction reduction in early exits. + + This is because when peeling for alignment using masking, the first + few elements of the vector can be inactive. As such if we find the + entry in the first iteration we have adjust the starting point of + the scalar code. + + We do this by creating a new scalar PHI that keeps track of whether + we are the first iteration of the loop (with the additional masking) + or whether we have taken a loop iteration already. + + The generated sequence: + + pre-header: + bb1: + i_1 = <number of leading inactive elements> + + header: + bb2: + i_2 = PHI <i_1(bb1), 0(latch)> + … + + early-exit: + bb3: + i_3 = iv_step * i_2 + PHI<vector-iv> + + The first part of the adjustment to create i_1 and i_2 are done here + and the last part creating i_3 is done in + vectorizable_live_operations when the induction extraction is + materialized. */ + if (LOOP_VINFO_EARLY_BREAKS (loop_vinfo) + && !LOOP_VINFO_MASK_NITERS_PFA_OFFSET (loop_vinfo)) + { + auto skip_niters = LOOP_VINFO_MASK_SKIP_NITERS (loop_vinfo); + tree ty_skip_niters = TREE_TYPE (skip_niters); + tree break_lhs_phi = vect_get_new_vect_var (ty_skip_niters, + vect_scalar_var, + "pfa_iv_offset"); + gphi *nphi = create_phi_node (break_lhs_phi, bb); + add_phi_arg (nphi, skip_niters, pe, UNKNOWN_LOCATION); + add_phi_arg (nphi, build_zero_cst (ty_skip_niters), + loop_latch_edge (iv_loop), UNKNOWN_LOCATION); + + LOOP_VINFO_MASK_NITERS_PFA_OFFSET (loop_vinfo) + = PHI_RESULT (nphi); + } } tree step_mul = NULL_TREE; unsigned ivn; @@ -11565,8 +11619,10 @@ vectorizable_live_operation (vec_info *vinfo, stmt_vec_info stmt_info, /* For early exit where the exit is not in the BB that leads to the latch then we're restarting the iteration in the scalar loop. So get the first live value. */ - if ((all_exits_as_early_p || !main_exit_edge) - && STMT_VINFO_DEF_TYPE (stmt_info) == vect_induction_def) + bool early_break_first_element_p + = (all_exits_as_early_p || !main_exit_edge) + && STMT_VINFO_DEF_TYPE (stmt_info) == vect_induction_def; + if (early_break_first_element_p) { tmp_vec_lhs = vec_lhs0; tmp_bitstart = build_zero_cst (TREE_TYPE (bitstart)); @@ -11581,6 +11637,45 @@ vectorizable_live_operation (vec_info *vinfo, stmt_vec_info stmt_info, lhs_type, &exit_gsi); auto gsi = gsi_for_stmt (use_stmt); + if (early_break_first_element_p + && LOOP_VINFO_MASK_NITERS_PFA_OFFSET (loop_vinfo)) + { + tree step_expr + = STMT_VINFO_LOOP_PHI_EVOLUTION_PART (stmt_info); + tree break_lhs_phi + = LOOP_VINFO_MASK_NITERS_PFA_OFFSET (loop_vinfo); + tree ty_skip_niters = TREE_TYPE (break_lhs_phi); + gimple_seq iv_stmts = NULL; + + /* Now create the PHI for the outside loop usage to + retrieve the value for the offset counter. */ + tree rphi_step + = gimple_convert (&iv_stmts, ty_skip_niters, step_expr); + tree tmp2 + = gimple_build (&iv_stmts, MULT_EXPR, + ty_skip_niters, rphi_step, + break_lhs_phi); + + if (POINTER_TYPE_P (TREE_TYPE (new_tree))) + { + tmp2 = gimple_convert (&iv_stmts, sizetype, tmp2); + tmp2 = gimple_build (&iv_stmts, POINTER_PLUS_EXPR, + TREE_TYPE (new_tree), new_tree, + tmp2); + } + else + { + tmp2 = gimple_convert (&iv_stmts, TREE_TYPE (new_tree), + tmp2); + tmp2 = gimple_build (&iv_stmts, PLUS_EXPR, + TREE_TYPE (new_tree), new_tree, + tmp2); + } + + new_tree = tmp2; + gsi_insert_seq_before (&exit_gsi, iv_stmts, GSI_SAME_STMT); + } + tree lhs_phi = gimple_phi_result (use_stmt); remove_phi_node (&gsi, false); gimple *copy = gimple_build_assign (lhs_phi, new_tree); diff --git a/gcc/tree-vect-slp.cc b/gcc/tree-vect-slp.cc index ecb4a65..19beeed 100644 --- a/gcc/tree-vect-slp.cc +++ b/gcc/tree-vect-slp.cc @@ -1099,7 +1099,7 @@ vect_build_slp_tree_1 (vec_info *vinfo, unsigned char *swap, tree first_lhs = NULL_TREE; tree first_op1 = NULL_TREE; stmt_vec_info first_load = NULL, prev_first_load = NULL; - bool first_stmt_ldst_p = false; + bool first_stmt_ldst_p = false, first_stmt_ldst_masklen_p = false; bool first_stmt_phi_p = false; int first_reduc_idx = -1; bool maybe_soft_fail = false; @@ -1133,6 +1133,7 @@ vect_build_slp_tree_1 (vec_info *vinfo, unsigned char *swap, FOR_EACH_VEC_ELT (stmts, i, stmt_info) { bool ldst_p = false; + bool ldst_masklen_p = false; bool phi_p = false; code_helper rhs_code = ERROR_MARK; @@ -1195,17 +1196,22 @@ vect_build_slp_tree_1 (vec_info *vinfo, unsigned char *swap, else rhs_code = CALL_EXPR; - if (cfn == CFN_MASK_LOAD - || cfn == CFN_GATHER_LOAD - || cfn == CFN_MASK_GATHER_LOAD - || cfn == CFN_MASK_LEN_GATHER_LOAD - || cfn == CFN_SCATTER_STORE - || cfn == CFN_MASK_SCATTER_STORE - || cfn == CFN_MASK_LEN_SCATTER_STORE) + if (cfn == CFN_GATHER_LOAD + || cfn == CFN_SCATTER_STORE) ldst_p = true; + else if (cfn == CFN_MASK_LOAD + || cfn == CFN_MASK_GATHER_LOAD + || cfn == CFN_MASK_LEN_GATHER_LOAD + || cfn == CFN_MASK_SCATTER_STORE + || cfn == CFN_MASK_LEN_SCATTER_STORE) + { + ldst_p = true; + ldst_masklen_p = true; + } else if (cfn == CFN_MASK_STORE) { ldst_p = true; + ldst_masklen_p = true; rhs_code = CFN_MASK_STORE; } else if (cfn == CFN_GOMP_SIMD_LANE) @@ -1246,6 +1252,7 @@ vect_build_slp_tree_1 (vec_info *vinfo, unsigned char *swap, first_lhs = lhs; first_stmt_code = rhs_code; first_stmt_ldst_p = ldst_p; + first_stmt_ldst_masklen_p = ldst_masklen_p; first_stmt_phi_p = phi_p; first_reduc_idx = STMT_VINFO_REDUC_IDX (stmt_info); @@ -1364,6 +1371,7 @@ vect_build_slp_tree_1 (vec_info *vinfo, unsigned char *swap, && (STMT_VINFO_GATHER_SCATTER_P (stmt_info) != STMT_VINFO_GATHER_SCATTER_P (first_stmt_info))) || first_stmt_ldst_p != ldst_p + || (ldst_p && first_stmt_ldst_masklen_p != ldst_masklen_p) || first_stmt_phi_p != phi_p) { if (dump_enabled_p ()) diff --git a/gcc/tree-vect-stmts.cc b/gcc/tree-vect-stmts.cc index 7f87435..5af1973 100644 --- a/gcc/tree-vect-stmts.cc +++ b/gcc/tree-vect-stmts.cc @@ -13615,29 +13615,23 @@ vectorizable_early_exit (vec_info *vinfo, stmt_vec_info stmt_info, codegen so we must replace the original insn. */ gimple *orig_stmt = STMT_VINFO_STMT (vect_orig_stmt (stmt_info)); gcond *cond_stmt = as_a <gcond *>(orig_stmt); + + tree cst = build_zero_cst (vectype); + auto bb = gimple_bb (cond_stmt); + edge exit_true_edge = EDGE_SUCC (bb, 0); + if (exit_true_edge->flags & EDGE_FALSE_VALUE) + exit_true_edge = EDGE_SUCC (bb, 1); + gcc_assert (exit_true_edge->flags & EDGE_TRUE_VALUE); + /* When vectorizing we assume that if the branch edge is taken that we're exiting the loop. This is not however always the case as the compiler will rewrite conditions to always be a comparison against 0. To do this it sometimes flips the edges. This is fine for scalar, but for vector we - then have to flip the test, as we're still assuming that if you take the - branch edge that we found the exit condition. i.e. we need to know whether - we are generating a `forall` or an `exist` condition. */ - auto new_code = NE_EXPR; - auto reduc_optab = ior_optab; - auto reduc_op = BIT_IOR_EXPR; - tree cst = build_zero_cst (vectype); - edge exit_true_edge = EDGE_SUCC (gimple_bb (cond_stmt), 0); - if (exit_true_edge->flags & EDGE_FALSE_VALUE) - exit_true_edge = EDGE_SUCC (gimple_bb (cond_stmt), 1); - gcc_assert (exit_true_edge->flags & EDGE_TRUE_VALUE); - if (flow_bb_inside_loop_p (LOOP_VINFO_LOOP (loop_vinfo), - exit_true_edge->dest)) - { - new_code = EQ_EXPR; - reduc_optab = and_optab; - reduc_op = BIT_AND_EXPR; - cst = build_minus_one_cst (vectype); - } + then have to negate the result of the test, as we're still assuming that if + you take the branch edge that we found the exit condition. i.e. we need to + know whether we are generating a `forall` or an `exist` condition. */ + bool flipped = flow_bb_inside_loop_p (LOOP_VINFO_LOOP (loop_vinfo), + exit_true_edge->dest); /* Analyze only. */ if (!vec_stmt) @@ -13653,14 +13647,13 @@ vectorizable_early_exit (vec_info *vinfo, stmt_vec_info stmt_info, } if (ncopies > 1 - && direct_optab_handler (reduc_optab, mode) == CODE_FOR_nothing) + && direct_optab_handler (ior_optab, mode) == CODE_FOR_nothing) { if (dump_enabled_p ()) dump_printf_loc (MSG_MISSED_OPTIMIZATION, vect_location, "can't vectorize early exit because the " - "target does not support boolean vector %s " + "target does not support boolean vector IOR " "for type %T.\n", - reduc_optab == ior_optab ? "OR" : "AND", vectype); return false; } @@ -13720,6 +13713,29 @@ vectorizable_early_exit (vec_info *vinfo, stmt_vec_info stmt_info, stmts.quick_push (gimple_assign_lhs (stmt)); } + /* If we're comparing against a previous forall we need to negate the resullts + before we do the final comparison or reduction. */ + if (flipped) + { + /* Rewrite the if(all(mask)) into if (!all(mask)) which is the same as + if (any(~mask)) by negating the masks and flipping the branches. + + 1. For unmasked loops we simply reduce the ~mask. + 2. For masked loops we reduce (~mask & loop_mask) which is the same as + doing (mask & loop_mask) ^ loop_mask. */ + for (unsigned i = 0; i < stmts.length (); i++) + { + tree inv_lhs = make_temp_ssa_name (vectype, NULL, "vexit_inv"); + auto inv_stmt = gimple_build_assign (inv_lhs, BIT_NOT_EXPR, stmts[i]); + vect_finish_stmt_generation (loop_vinfo, stmt_info, inv_stmt, + &cond_gsi); + stmts[i] = inv_lhs; + } + + EDGE_SUCC (bb, 0)->flags ^= (EDGE_TRUE_VALUE|EDGE_FALSE_VALUE); + EDGE_SUCC (bb, 1)->flags ^= (EDGE_TRUE_VALUE|EDGE_FALSE_VALUE); + } + /* Determine if we need to reduce the final value. */ if (stmts.length () > 1) { @@ -13758,7 +13774,7 @@ vectorizable_early_exit (vec_info *vinfo, stmt_vec_info stmt_info, new_temp = make_temp_ssa_name (vectype, NULL, "vexit_reduc"); tree arg0 = workset.pop (); tree arg1 = workset.pop (); - new_stmt = gimple_build_assign (new_temp, reduc_op, arg0, arg1); + new_stmt = gimple_build_assign (new_temp, BIT_IOR_EXPR, arg0, arg1); vect_finish_stmt_generation (loop_vinfo, stmt_info, new_stmt, &cond_gsi); workset.quick_insert (0, new_temp); @@ -13781,7 +13797,7 @@ vectorizable_early_exit (vec_info *vinfo, stmt_vec_info stmt_info, gcc_assert (new_temp); - gimple_cond_set_condition (cond_stmt, new_code, new_temp, cst); + gimple_cond_set_condition (cond_stmt, NE_EXPR, new_temp, cst); update_stmt (orig_stmt); if (slp_node) diff --git a/gcc/tree-vectorizer.h b/gcc/tree-vectorizer.h index 97caf61..01d19c7 100644 --- a/gcc/tree-vectorizer.h +++ b/gcc/tree-vectorizer.h @@ -818,6 +818,11 @@ public: elements that should be false in the first mask). */ tree mask_skip_niters; + /* If we are using a loop mask to align memory addresses and we're in an + early break loop then this variable contains the number of elements that + were skipped during the initial iteration of the loop. */ + tree mask_skip_niters_pfa_offset; + /* The type that the loop control IV should be converted to before testing which of the VF scalars are active and inactive. Only meaningful if LOOP_VINFO_USING_PARTIAL_VECTORS_P. */ @@ -854,6 +859,9 @@ public: /* The mask used to check the alignment of pointers or arrays. */ int ptr_mask; + /* Indicates whether the loop has any non-linear IV. */ + bool nonlinear_iv; + /* Data Dependence Relations defining address ranges that are candidates for a run-time aliasing check. */ auto_vec<ddr_p> may_alias_ddrs; @@ -1064,6 +1072,7 @@ public: #define LOOP_VINFO_MASKS(L) (L)->masks #define LOOP_VINFO_LENS(L) (L)->lens #define LOOP_VINFO_MASK_SKIP_NITERS(L) (L)->mask_skip_niters +#define LOOP_VINFO_MASK_NITERS_PFA_OFFSET(L) (L)->mask_skip_niters_pfa_offset #define LOOP_VINFO_RGROUP_COMPARE_TYPE(L) (L)->rgroup_compare_type #define LOOP_VINFO_RGROUP_IV_TYPE(L) (L)->rgroup_iv_type #define LOOP_VINFO_PARTIAL_VECTORS_STYLE(L) (L)->partial_vector_style @@ -1073,6 +1082,7 @@ public: #define LOOP_VINFO_DDRS(L) (L)->shared->ddrs #define LOOP_VINFO_INT_NITERS(L) (TREE_INT_CST_LOW ((L)->num_iters)) #define LOOP_VINFO_PEELING_FOR_ALIGNMENT(L) (L)->peeling_for_alignment +#define LOOP_VINFO_NON_LINEAR_IV(L) (L)->nonlinear_iv #define LOOP_VINFO_UNALIGNED_DR(L) (L)->unaligned_dr #define LOOP_VINFO_MAY_MISALIGN_STMTS(L) (L)->may_misalign_stmts #define LOOP_VINFO_MAY_ALIAS_DDRS(L) (L)->may_alias_ddrs @@ -2138,8 +2148,14 @@ unlimited_cost_model (loop_p loop) inline bool vect_use_loop_mask_for_alignment_p (loop_vec_info loop_vinfo) { + /* With early break vectorization we don't know whether the accesses will stay + inside the loop or not. TODO: The early break adjustment code can be + implemented the same way as vectorizable_linear_induction. However we + can't test this today so reject it. */ return (LOOP_VINFO_FULLY_MASKED_P (loop_vinfo) - && LOOP_VINFO_PEELING_FOR_ALIGNMENT (loop_vinfo)); + && LOOP_VINFO_PEELING_FOR_ALIGNMENT (loop_vinfo) + && !(LOOP_VINFO_NON_LINEAR_IV (loop_vinfo) + && LOOP_VINFO_EARLY_BREAKS (loop_vinfo))); } /* Return the number of vectors of type VECTYPE that are needed to get diff --git a/gcc/tree.def b/gcc/tree.def index c4ad8d0..2c37e44 100644 --- a/gcc/tree.def +++ b/gcc/tree.def @@ -135,7 +135,11 @@ DEFTREECODE (OFFSET_TYPE, "offset_type", tcc_type, 0) DEFTREECODE (ENUMERAL_TYPE, "enumeral_type", tcc_type, 0) /* Boolean type (true or false are the only values). Looks like an - INTEGRAL_TYPE. */ + INTEGER_TYPE, but must be dealt with specially because TYPE_PRECISION + may be arbitrary despite the restricted set of valid values (in other + words, boolean types with TYPE_PRECISION > 1 exist in some languages). + Similarly, TYPE_UNSIGNED may be false for components of vector masks, + as well as for boolean types in languages other than C. */ DEFTREECODE (BOOLEAN_TYPE, "boolean_type", tcc_type, 0) /* Integer types in all languages, including char in C. @@ -2395,11 +2395,11 @@ public: array_slice (vec<OtherT, A, vl_embed> *v) : m_base (v ? v->address () : nullptr), m_size (v ? v->length () : 0) {} - iterator begin () { return m_base; } - iterator end () { return m_base + m_size; } + iterator begin () { gcc_checking_assert (is_valid ()); return m_base; } + iterator end () { gcc_checking_assert (is_valid ()); return m_base + m_size; } - const_iterator begin () const { return m_base; } - const_iterator end () const { return m_base + m_size; } + const_iterator begin () const { gcc_checking_assert (is_valid ()); return m_base; } + const_iterator end () const { gcc_checking_assert (is_valid ()); return m_base + m_size; } value_type &front (); value_type &back (); |