{"payload":{"feedbackUrl":"https://github.com/orgs/community/discussions/53140","repo":{"id":176542129,"defaultBranch":"master","name":"gcc-git","ownerLogin":"iains","currentUserCanPush":false,"isFork":false,"isEmpty":false,"createdAt":"2019-03-19T15:23:55.000Z","ownerAvatar":"https://avatars.githubusercontent.com/u/4039407?v=4","public":true,"private":false,"isOrgOwned":false},"refInfo":{"name":"","listCacheKey":"v0:1720985328.0","currentOid":""},"activityList":{"items":[{"before":"40a990c8b512fd25bd7d7b45aa509e1880d77209","after":"e0d997e913f811ecf4b3e10891e6a4aab5b38a31","ref":"refs/heads/master","pushedAt":"2024-07-20T13:12:51.000Z","pushType":"push","commitsCount":97,"pusher":{"login":"iains","name":"Iain Sandoe","path":"/iains","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/4039407?s=80&v=4"},"commit":{"message":"Avoid undefined behaviour in build_option_suggestions\n\nThe inner loop in build_option_suggestions uses OPTION to take the\naddress of OPTB and use it across iterations, which is undefined\nbehaviour since OPTB is defined within the loop. Pull it outside the\nloop to make this defined.\n\ngcc/ChangeLog:\n\n\t* opt-suggestions.cc\n\t(option_proposer::build_option_suggestions): Pull OPTB\n\tdefinition out of the innermost loop.","shortMessageHtmlLink":"Avoid undefined behaviour in build_option_suggestions"}},{"before":"619f587f6852517889c216d4eb63728e9b062032","after":"40a990c8b512fd25bd7d7b45aa509e1880d77209","ref":"refs/heads/master","pushedAt":"2024-07-16T22:50:09.000Z","pushType":"push","commitsCount":112,"pusher":{"login":"iains","name":"Iain Sandoe","path":"/iains","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/4039407?s=80&v=4"},"commit":{"message":"c++/contracts: ICE in C++ Contracts with '-fno-exceptions' [PR 110159]\n\nWe currently only initialise terminate_fn if exceptions are enabled.\nHowever, contract handling requires terminate_fn when building the\ncontract because a contract failure may result in std::terminate call\nregardless of whether the exceptions are enabled. Refactored\ninit_exception_processing to extract the initialisation of\nterminate_fn. New function init_terminate_fn added that initialises\nterminate_fn if it hasn't already been initialised. Call to terminate_fn\nadded in cxx_init_decl_processing if contracts are enabled.\n\n\tPR c++/110159\n\ngcc/cp/ChangeLog:\n\n\t* cp-tree.h (init_terminate_fn): Declaration of a new function.\n\t* decl.cc (cxx_init_decl_processing): If contracts are enabled,\n\tcall init_terminate_fn.\n\t* except.cc (init_exception_processing): Function refactored to\n\tcall init_terminate_fn.\n\t(init_terminate_fn): Added new function that initializes\n\tterminate_fn if\tit hasn't already been initialised.\n\ngcc/testsuite/ChangeLog:\n\n\t* g++.dg/contracts/pr110159.C: New test.\n\nSigned-off-by: Nina Ranns ","shortMessageHtmlLink":"c++/contracts: ICE in C++ Contracts with '-fno-exceptions' [PR 110159]"}},{"before":null,"after":"14c1f1e92cdc16548b906c0cb6eec203b9af7c7b","ref":"refs/heads/master-wip-coro-tests","pushedAt":"2024-07-14T19:28:48.000Z","pushType":"branch_creation","commitsCount":0,"pusher":{"login":"iains","name":"Iain Sandoe","path":"/iains","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/4039407?s=80&v=4"},"commit":{"message":"testsuite, c++, coroutines: Misc tests for open PRs","shortMessageHtmlLink":"testsuite, c++, coroutines: Misc tests for open PRs"}},{"before":"d4be364625b091973d7b81caca87c2c454450ae6","after":"07bb2457812cd73ca0cdbde568000091c9e15e1d","ref":"refs/heads/master-wip-coro-dbg","pushedAt":"2024-07-14T15:21:53.000Z","pushType":"force_push","commitsCount":0,"pusher":{"login":"iains","name":"Iain Sandoe","path":"/iains","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/4039407?s=80&v=4"},"commit":{"message":"debug additions\nprint indication of bind expr in statement expr.\nfix pretty print of template ids, don't try to dump failed transforms\npretty-print: show sub-block lists, and the decl we built cleanup for.\nt-pretty-print-stmt-expr","shortMessageHtmlLink":"debug additions"}},{"before":"807e36d76e5105015afe0cf20e9a8837bb550f4b","after":"619f587f6852517889c216d4eb63728e9b062032","ref":"refs/heads/master","pushedAt":"2024-07-11T07:55:19.000Z","pushType":"push","commitsCount":97,"pusher":{"login":"iains","name":"Iain Sandoe","path":"/iains","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/4039407?s=80&v=4"},"commit":{"message":"Revert \"fixincludes: skip stdio_stdarg_h on darwin\"\n\nThis reverts commit 7d454cae9d7df1f2936ad02d0742674a85396736.\n\nThe change breaks bootstrap on some x86_64 OS versions.","shortMessageHtmlLink":"Revert \"fixincludes: skip stdio_stdarg_h on darwin\""}},{"before":"21e3565927eda5ce9907d91100623052fa8182cd","after":"807e36d76e5105015afe0cf20e9a8837bb550f4b","ref":"refs/heads/master","pushedAt":"2024-07-05T22:40:32.000Z","pushType":"push","commitsCount":146,"pusher":{"login":"iains","name":"Iain Sandoe","path":"/iains","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/4039407?s=80&v=4"},"commit":{"message":"x86, Darwin: Fix bootstrap for 32b multilibs/hosts.\n\nr15-1735-ge62ea4fb8ffcab06ddd contained changes that altered the\ncodegen for 32b Darwin (whether hosted on 64b or as 32b host) such\nthat the per function picbase load is called multiple times in some\ncases. Darwin's back end is not expecting this (and indeed some of\nthe handling depends on a single instance).\n\nThe fixes the issue by marking those instructions as not copyable\n(as suggested by Andrew Pinski).\n\nThe change is Darwin-specific.\n\ngcc/ChangeLog:\n\n\t* config/i386/i386.cc (ix86_cannot_copy_insn_p): New.\n\t(TARGET_CANNOT_COPY_INSN_P): New.\n\nSigned-off-by: Iain Sandoe ","shortMessageHtmlLink":"x86, Darwin: Fix bootstrap for 32b multilibs/hosts."}},{"before":"45e74d5dfa4c5f372df0d3545bc342b6a2505e71","after":"21e3565927eda5ce9907d91100623052fa8182cd","ref":"refs/heads/master","pushedAt":"2024-06-30T16:04:07.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"iains","name":"Iain Sandoe","path":"/iains","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/4039407?s=80&v=4"},"commit":{"message":"Match: Support imm form for unsigned scalar .SAT_ADD\n\nThis patch would like to support the form of unsigned scalar .SAT_ADD\nwhen one of the op is IMM. For example as below:\n\nForm IMM:\n #define DEF_SAT_U_ADD_IMM_FMT_1(T) \\\n T __attribute__((noinline)) \\\n sat_u_add_imm_##T##_fmt_1 (T x) \\\n { \\\n return (T)(x + 9) >= x ? (x + 9) : -1; \\\n }\n\nDEF_SAT_U_ADD_IMM_FMT_1(uint64_t)\n\nBefore this patch:\n__attribute__((noinline))\nuint64_t sat_u_add_imm_uint64_t_fmt_1 (uint64_t x)\n{\n long unsigned int _1;\n uint64_t _3;\n\n;; basic block 2, loop depth 0\n;; pred: ENTRY\n _1 = MIN_EXPR ;\n _3 = _1 + 9;\n return _3;\n;; succ: EXIT\n\n}\n\nAfter this patch:\n__attribute__((noinline))\nuint64_t sat_u_add_imm_uint64_t_fmt_1 (uint64_t x)\n{\n uint64_t _3;\n\n;; basic block 2, loop depth 0\n;; pred: ENTRY\n _3 = .SAT_ADD (x_2(D), 9); [tail call]\n return _3;\n;; succ: EXIT\n\n}\n\nThe below test suites are passed for this patch:\n1. The rv64gcv fully regression test with newlib.\n2. The x86 bootstrap test.\n3. The x86 fully regression test.\n\ngcc/ChangeLog:\n\n\t* match.pd: Add imm form for .SAT_ADD matching.\n\t* tree-ssa-math-opts.cc (math_opts_dom_walker::after_dom_children):\n\tAdd .SAT_ADD matching under PLUS_EXPR.\n\nSigned-off-by: Pan Li ","shortMessageHtmlLink":"Match: Support imm form for unsigned scalar .SAT_ADD"}},{"before":"e22b7f741ab54ff3a3f8a676ce9e7414fe174958","after":"45e74d5dfa4c5f372df0d3545bc342b6a2505e71","ref":"refs/heads/master","pushedAt":"2024-06-29T02:59:25.000Z","pushType":"push","commitsCount":609,"pusher":{"login":"iains","name":"Iain Sandoe","path":"/iains","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/4039407?s=80&v=4"},"commit":{"message":"jit: Fix Darwin bootstrap after r15-1699.\n\nr15-1699-g445c62ee492 contains changes that trigger two maybe-uninitialized\nwarnings on Darwin, which result in a bootstrap failure.\n\nNote that the warnings are false positives, in fact the variables should be\ninitialized in the cases of a switch (all values of the switch condition are\ncovered).\n\nFixed here by providing default initializations for the relevant variables.\n\ngcc/jit/ChangeLog:\n\n\t* jit-recording.cc\n\t(recording::memento_of_typeinfo::make_debug_string): Default the value\n\tof ident.\n\t(recording::memento_of_typeinfo::write_reproducer): Default the value\n\tof type.\n\nSigned-off-by: Iain Sandoe ","shortMessageHtmlLink":"jit: Fix Darwin bootstrap after r15-1699."}},{"before":null,"after":"47f73b93863a8b14dfe8ab102ac48a02898274b3","ref":"refs/heads/contracts-nonattr-rb","pushedAt":"2024-06-09T13:15:09.000Z","pushType":"branch_creation","commitsCount":0,"pusher":{"login":"iains","name":"Iain Sandoe","path":"/iains","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/4039407?s=80&v=4"},"commit":{"message":"*This pointer first attempt","shortMessageHtmlLink":"*This pointer first attempt"}},{"before":"28b508233a12c13295f960a2cb8a4864879acfb4","after":"e22b7f741ab54ff3a3f8a676ce9e7414fe174958","ref":"refs/heads/master","pushedAt":"2024-06-09T08:18:26.000Z","pushType":"push","commitsCount":281,"pusher":{"login":"iains","name":"Iain Sandoe","path":"/iains","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/4039407?s=80&v=4"},"commit":{"message":"analyzer: Restore g++ 4.8 bootstrap; use std::move to return std::unique_ptr.\n\nThis patch restores bootstrap when using g++ 4.8 as a host compiler.\nReturning a std::unique_ptr requires a std::move on C++ compilers\n(pre-C++17) that don't guarantee copy elision/return value optimization.\n\n2024-06-08 Roger Sayle \n\ngcc/analyzer/ChangeLog\n\t* constraint-manager.cc (equiv_class::make_dump_widget): Use\n\tstd::move to return a std::unique_ptr.\n\t(bounded_ranges_constraint::make_dump_widget): Likewise.\n\t(constraint_manager::make_dump_widget): Likewise.\n\t* program-state.cc (sm_state_map::make_dump_widget): Likewise.\n\t(program_state::make_dump_widget): Likewise.\n\t* region-model.cc (region_to_value_map::make_dump_widget): Likewise.\n\t(region_model::make_dump_widget): Likewise.\n\t* region.cc (region::make_dump_widget): Likewise.\n\t* store.cc (binding_cluster::make_dump_widget): Likewise.\n\t(store::make_dump_widget): Likewise.\n\t* svalue.cc (svalue::make_dump_widget): Likewise.","shortMessageHtmlLink":"analyzer: Restore g++ 4.8 bootstrap; use std::move to return std::uni…"}},{"before":"b9294757f82aae8de6d98c122cd4e3b98f685217","after":"28b508233a12c13295f960a2cb8a4864879acfb4","ref":"refs/heads/master","pushedAt":"2024-05-25T06:07:01.000Z","pushType":"push","commitsCount":444,"pusher":{"login":"iains","name":"Iain Sandoe","path":"/iains","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/4039407?s=80&v=4"},"commit":{"message":"c++/modules: Improve diagnostic when redeclaring builtin in module [PR102345]\n\nIf a user mistakenly includes a standard library header within the\nmodule purview, they currently get a confusing \"declaration conflicts\nwith builtin\" error. This patch updates the message to include \"in\nmodule\", to help guide the user towards the likely cause.\n\n\tPR c++/102345\n\ngcc/cp/ChangeLog:\n\n\t* module.cc (module_may_redeclare): Update error message.\n\ngcc/testsuite/ChangeLog:\n\n\t* g++.dg/modules/enum-12.C: Test for updated error.\n\nSigned-off-by: Nathaniel Shead ","shortMessageHtmlLink":"c++/modules: Improve diagnostic when redeclaring builtin in module [P…"}},{"before":"3dab8f8a542aeabfab1a244a7a601b220680c80b","after":"b9294757f82aae8de6d98c122cd4e3b98f685217","ref":"refs/heads/master","pushedAt":"2024-05-13T06:46:22.000Z","pushType":"push","commitsCount":9,"pusher":{"login":"iains","name":"Iain Sandoe","path":"/iains","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/4039407?s=80&v=4"},"commit":{"message":"Fortran: Unlimited polymorphic intrinsic function arguments [PR84006]\n\n2024-05-12 Paul Thomas \n\ngcc/fortran\n\tPR fortran/84006\n\tPR fortran/100027\n\tPR fortran/98534\n\t* iresolve.cc (gfc_resolve_transfer): Emit a TODO error for\n\tunlimited polymorphic mold.\n\t* trans-expr.cc (gfc_resize_class_size_with_len): Use the fold\n\teven if a block is not available in which to fix the result.\n\t(trans_class_assignment): Enable correct assignment of\n\tcharacter expressions to unlimited polymorphic variables using\n\tlhs _len field and rse string_length.\n\t* trans-intrinsic.cc (gfc_conv_intrinsic_storage_size): Extract\n\tthe class expression so that the unlimited polymorphic class\n\texpression can be used in gfc_resize_class_size_with_len to\n\tobtain the storage size for character payloads. Guard the use\n\tof GFC_DECL_SAVED_DESCRIPTOR by testing for DECL_LANG_SPECIFIC\n\tto prevent the ICE. Also, invert the order to use the class\n\texpression extracted from the argument.\n\t(gfc_conv_intrinsic_transfer): In same way as 'storage_size',\n\tuse the _len field to obtaining the correct length for arg 1.\n\tAdd a branch for the element size in bytes of class expressions\n\twith provision to make use of the unlimited polymorphic _len\n\tfield. Again, the class references are explicitly identified.\n\t'mold_expr' was already declared. Use it instead of 'arg'. Do\n\tnot fix 'dest_word_len' for deferred character sources because\n\treallocation on assign makes use of it before it is assigned.\n\ngcc/testsuite/\n\tPR fortran/84006\n\tPR fortran/100027\n\t* gfortran.dg/storage_size_7.f90: New test.\n\n\tPR fortran/98534\n\t* gfortran.dg/transfer_class_4.f90: New test.","shortMessageHtmlLink":"Fortran: Unlimited polymorphic intrinsic function arguments [PR84006]"}},{"before":"bb5f619a938567b316306bdc46f0efb8d18ddc59","after":"3dab8f8a542aeabfab1a244a7a601b220680c80b","ref":"refs/heads/master","pushedAt":"2024-05-11T08:43:25.000Z","pushType":"push","commitsCount":18,"pusher":{"login":"iains","name":"Iain Sandoe","path":"/iains","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/4039407?s=80&v=4"},"commit":{"message":"c++: Implement __is_nothrow_invocable built-in trait\n\nThis patch implements built-in trait for std::is_nothrow_invocable.\n\ngcc/cp/ChangeLog:\n\n\t* cp-trait.def: Define __is_nothrow_invocable.\n\t* constraint.cc (diagnose_trait_expr): Handle\n\tCPTK_IS_NOTHROW_INVOCABLE.\n\t* semantics.cc (trait_expr_value): Likewise.\n\t(finish_trait_expr): Likewise.\n\ngcc/testsuite/ChangeLog:\n\n\t* g++.dg/ext/has-builtin-1.C: Test existence of\n\t__is_nothrow_invocable.\n\t* g++.dg/ext/is_nothrow_invocable.C: New test.\n\nSigned-off-by: Ken Matsui \nReviewed-by: Jason Merrill ","shortMessageHtmlLink":"c++: Implement __is_nothrow_invocable built-in trait"}},{"before":"ffb521f157fec823907913602ed5cb73dd1c63e8","after":"bb5f619a938567b316306bdc46f0efb8d18ddc59","ref":"refs/heads/master","pushedAt":"2024-05-10T22:49:51.000Z","pushType":"push","commitsCount":193,"pusher":{"login":"iains","name":"Iain Sandoe","path":"/iains","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/4039407?s=80&v=4"},"commit":{"message":"bpf: fix printing of memory operands in pseudoc asm dialect\n\nThe BPF backend was emitting memory operands in pseudo-C syntax\nwithout surrounding parentheses. These were being provided in the\ncorresponding instruction templates.\n\nThis was causing GCC emitting invalid instructions when finding inline\nassembly with memory operands like:\n\n\tasm volatile (\n\t\"r1 = *(u64 *)%[ctx_a];\"\n\t\"if r1 != 42 goto 1f;\"\n\t\"r1 = *(u64 *)%[ctx_b];\"\n\t\"if r1 != 42 goto 1f;\"\n\t\"r1 = *(u64 *)%[ctx_c];\"\n\t\"if r1 != 7 goto 1f;\"\n\t\"r1 /= 0;\"\n\t\"1:\"\n\t:\n\t: [ctx_a]\"m\"(ctx.a),\n\t [ctx_b]\"m\"(ctx.b),\n\t [ctx_c]\"m\"(ctx.c)\n\t: \"r1\"\n\t);\n\nThis patch changes the backend to include the surrounding parentheses\nin the printed representation of the memory operands (much like\nsurrounding brackets are included in normal asm syntax) and adapts the\nimpacted instruction templates accordingly.\n\nTested in target bpf-unknown-none, host x86_64-linux-gnu.\n\ngcc/ChangeLog:\n\n\t* config/bpf/bpf.cc (bpf_print_operand_address): Include\n\tsurrounding parenthesis around mem operands in pseudoc asm\n\tdialect.\n\t* config/bpf/bpf.md (*mov): Adapt accordingly.\n\t(zero_extendhidi2): Likewise.\n\t(zero_extendqidi2): Likewise.\n\t(*extendsidi2): Likewise.\n\t(*extendsidi2): Likewise.\n\t(extendhidi2): Likewise.\n\t(extendqidi2): Likewise.\n\t(extendhisi2): Likewise.\n\t* config/bpf/atomic.md (atomic_add): Likewise.\n\t(atomic_and): Likewise.\n\t(atomic_or): Likewise.\n\t(atomic_xor): Likewise.\n\t(atomic_fetch_add): Likewise.\n\t(atomic_fetch_and): Likewise.\n\t(atomic_fetch_or): Likewise.\n\t(atomic_fetch_xor): Likewise.","shortMessageHtmlLink":"bpf: fix printing of memory operands in pseudoc asm dialect"}},{"before":"affd24bfc62203db9f9937c0d6cf8f1f75b80d72","after":"ffb521f157fec823907913602ed5cb73dd1c63e8","ref":"refs/heads/master","pushedAt":"2024-05-05T06:07:29.000Z","pushType":"push","commitsCount":47,"pusher":{"login":"iains","name":"Iain Sandoe","path":"/iains","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/4039407?s=80&v=4"},"commit":{"message":"Remove m_nloops field from loop_versioning\n\nThis is a small cleanup of loop_versioning where m_nloops\nis only used in the constructor so we can remove the whole\nfield.\n\nBootstrapped and tested on x86_64-linux-gnu.\n\ngcc/ChangeLog:\n\n\t* gimple-loop-versioning.cc (loop_versioning): Remove m_nloops field.\n\t(loop_versioning::loop_versioning): Remove initialization of\n\tm_nloops field and move it to be a local variable.\n\t(loop_versioning::analyze_blocks): Fix formating.\n\nSigned-off-by: Andrew Pinski ","shortMessageHtmlLink":"Remove m_nloops field from loop_versioning"}},{"before":"9b5c0be59d0f94df0517820f00b4520b5abddd8c","after":"affd24bfc62203db9f9937c0d6cf8f1f75b80d72","ref":"refs/heads/master","pushedAt":"2024-05-02T22:52:54.000Z","pushType":"push","commitsCount":8,"pusher":{"login":"iains","name":"Iain Sandoe","path":"/iains","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/4039407?s=80&v=4"},"commit":{"message":"libgfortran: Fix up the autoreconf warnings\n\nThis means using sub-dirs and amending some of the recipes accordingly.\n\nlibgfortran/ChangeLog:\n\n\t* Makefile.am: Use sub-dirs, amend recipies accordingly.\n\t* Makefile.in: Regenerate.\n\nSigned-off-by: Iain Sandoe ","shortMessageHtmlLink":"libgfortran: Fix up the autoreconf warnings"}},{"before":"39cb6b880f723780faeef06383e67cfed2e3458d","after":"9b5c0be59d0f94df0517820f00b4520b5abddd8c","ref":"refs/heads/master","pushedAt":"2024-05-02T13:07:31.000Z","pushType":"push","commitsCount":423,"pusher":{"login":"iains","name":"Iain Sandoe","path":"/iains","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/4039407?s=80&v=4"},"commit":{"message":"Objective-C, NeXT, v2: Correct a regression in code-gen.\n\nThere have been several changes in the ABI of Objective-C which\ndepend on the OS version targetted. In this case Protocols and\nLabelProtocols should be made weak/hidden/extern from macOS 10.7\nhowever there was a mistake in the code causing this to occur\nfrom macOS 10.6. Fixed thus.\n\ngcc/objc/ChangeLog:\n\n\t* objc-next-runtime-abi-02.cc (WEAK_PROTOCOLS_AFTER): New.\n\t(next_runtime_abi_02_protocol_decl): Use WEAK_PROTOCOLS_AFTER\n\tto determine this ABI change.\n\t(build_v2_protocol_list_address_table): Likewise.\n\nSigned-off-by: Iain Sandoe ","shortMessageHtmlLink":"Objective-C, NeXT, v2: Correct a regression in code-gen."}},{"before":"4e3c8257304c55f2ebfb24bd6de3236bda0f054e","after":"39cb6b880f723780faeef06383e67cfed2e3458d","ref":"refs/heads/master","pushedAt":"2024-04-08T11:27:15.000Z","pushType":"push","commitsCount":4,"pusher":{"login":"iains","name":"Iain Sandoe","path":"/iains","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/4039407?s=80&v=4"},"commit":{"message":"Darwin: Sync coverage specs with gcc/gcc.cc.\n\nThe specs for coverage ere out of date leading to test fails for\nfcondition-coverage cases. Fixed by updating to match the specs\nin gcc/gcc.cc.\n\ngcc/ChangeLog:\n\n\t* config/darwin.h (LINK_COMMAND_SPEC_A): Update coverage\n\tspecs.\n\nSigned-off-by: Iain Sandoe ","shortMessageHtmlLink":"Darwin: Sync coverage specs with gcc/gcc.cc."}},{"before":"75b49c0e9012f5ecef0d32f3f6a0d8da66517576","after":"4e3c8257304c55f2ebfb24bd6de3236bda0f054e","ref":"refs/heads/master","pushedAt":"2024-04-07T09:14:15.000Z","pushType":"push","commitsCount":7,"pusher":{"login":"iains","name":"Iain Sandoe","path":"/iains","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/4039407?s=80&v=4"},"commit":{"message":"PR modula2/114617 gm2 unable to resolve const expressions using relop ICE\n\nThis patch allows cc1gm2 to resolve constant expressions which use\nrelative operators. Previous to the patch the result of a relop\nwas stored in a temporary variable set by an if then else quadruple\nsequence. This patch marks a const expression in the quadruples\nand then reduces this sequence of quadruples into a single\nassignment to an internal constant.\n\ngcc/m2/ChangeLog:\n\n\tPR modula2/114617\n\t* gm2-compiler/M2GenGCC.mod (CodeStatememt): Add quad trace.\n\t(ResolveConstantExpressions): Add parameter p to FoldIfLess,\n\tFoldIfGre, FoldIfLessEqu, FoldIfGreEqu, FoldIfEqu, FoldIfNotEqu,\n\tFoldIfIn and FoldIfNotIn.\n\t(CodeInline): Add constExpr variable and pass it to GetQuadOtok.\n\t(CodeReturnValue): Ditto.\n\t(CodeParam): Ditto.\n\t(FoldStringLength): Ditto.\n\t(FoldStringConvertM2nul): Ditto.\n\t(FoldStringConvertCnul): Ditto.\n\t(DeclaredOperandsBecomes): Ditto.\n\t(TypeCheckBecomes): Ditto.\n\t(PerformFoldBecomes): Ditto.\n\t(CodeBecomes): Ditto.\n\t(CheckElementSetTypes): Ditto.\n\t(CodeBinarySet): Ditto.\n\t(PerformCodeIfLess): Ditto.\n\t(PerformCodeIfGre): Ditto.\n\t(PerformCodeIfLessEqu): Ditto.\n\t(PerformCodeIfGreEqu): Ditto.\n\t(PerformCodeIfEqu): Ditto.\n\t(PerformCodeIfNotEqu): Ditto.\n\t(IsValidExpressionRelOp): Ditto.\n\t(PerformCodeIfIn): Ditto.\n\t(PerformCodeIfNotIn): Ditto.\n\t(CodeXIndr): Ditto.\n\t(QuadCondition): New procedure function.\n\t(IsBooleanRelOpPattern): Ditto.\n\t(FoldBooleanRelopPattern): Ditto.\n\t(FoldIfGre): Check for boolean relop constant expression and\n\tadd parameter p.\n\t(FoldIfLessEqu): Ditto.\n\t(FoldIfIn): Ditto.\n\t(FoldIfEqu): Ditto.\n\t(FoldIfNotIn): Ditto.\n\t(FoldIfGreEqu): New procedure.\n\t(FoldIfNotEqu): Ditto.\n\t* gm2-compiler/M2Optimize.mod (ReduceBranch): Add constExpr\n\tvariable and pass it to GetQuadOtok.\n\t* gm2-compiler/M2Quads.def (IsBecomes): New procedure function.\n\t(IsDummy): Ditto.\n\t(IsQuadConstExpr): Ditto.\n\t(SetQuadConstExpr): Ditto.\n\t(GetQuadDest): New procedure.\n\t(GetQuadOp1): New procedure.\n\t(GetQuadOp2): New procedure.\n\t(GetQuadOp3): New procedure.\n\t(GetQuadOtok): New procedure.\n\t(GetQuadOTypetok): New procedure.\n\t(PutQuadOtok): New procedure.\n\t(IsInConstParameters): New procedure function.\n\t* gm2-compiler/M2Quads.mod (IsBecomes): New procedure function.\n\t(IsDummy): Ditto.\n\t(IsQuadConstExpr): Ditto.\n\t(SetQuadConstExpr): Ditto.\n\t(GetQuadDest): New procedure.\n\t(GetQuadOp1): New procedure.\n\t(GetQuadOp2): New procedure.\n\t(GetQuadOp3): New procedure.\n\t(GetQuadOtok): New procedure.\n\t(GetQuadOTypetok): New procedure.\n\t(PutQuadOtok): New procedure.\n\t(IsInConstParameters): New procedure function.\n\t(ConstStack): Remove to ...\n\t(ConstExprStack): ... this.\n\t(ConstParamStack): New variable and initialize.\n\t(QuadFrame): New field ConstExpr.\n\t(GetQuadOtok): Add parameter constExpr and assign.\n\t(PutQuadOtok): Add constExpr parameter and assign.\n\t(PutQuadOType): Ditto.\n\t(GetQuadOTypetok): Ditto.\n\t(EraseQuad): Assign ConstExpr to FALSE.\n\t(FoldSubrange): Set ConstExpr to FALSE in BecomesOp.\n\t(PushInConstParameters): New procedure.\n\t(PopInConstParameters): New procedure.\n\t(IsInConstParameters): New procedure function.\n\t* gm2-compiler/M2SymInit.mod (IssueConditional): Add\n\tconstExpr boolean variable.\n\t(CheckReadBeforeInitQuad): Ditto.\n\t(trashParam): Ditto.\n\t* gm2-compiler/P3Build.bnf (ConstExpression): Call\n\tPushInConstExpression and PopInConstExpression.\n\t(ConstSetOrQualidentOrFunction): Call\n\tPushInConstParameters and PopInConstParameters.\n\t* gm2-compiler/PCBuild.bnf (ConstExpression): Call\n\tPushInConstExpression and PopInConstExpression.\n\t* gm2-compiler/PHBuild.bnf: Ditto\n\t* gm2-gcc/m2expr.cc (m2expr_BuildCondIfExpression): New\n\tfunction.\n\t* gm2-gcc/m2expr.def (BuildCondIfExpression): New prototype.\n\t* gm2-gcc/m2expr.h (m2expr_BuildCondIfExpression): New function.\n\ngcc/testsuite/ChangeLog:\n\n\tPR modula2/114617\n\t* gm2/iso/const/pass/iso-const-pass.exp: New test.\n\nSigned-off-by: Gaius Mulley ","shortMessageHtmlLink":"PR modula2/114617 gm2 unable to resolve const expressions using relop…"}},{"before":"d60968de6961cef144a5cf8701ea0d3f4ea90f18","after":"75b49c0e9012f5ecef0d32f3f6a0d8da66517576","ref":"refs/heads/master","pushedAt":"2024-04-05T18:13:57.000Z","pushType":"push","commitsCount":43,"pusher":{"login":"iains","name":"Iain Sandoe","path":"/iains","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/4039407?s=80&v=4"},"commit":{"message":"testsuite, Darwin: Account for block labels in function body scans.\n\nWhen we have '-O3 -g' we emit a bunch of LB{B,E} local labels which\nwere not currently being discarded, leading to some test fails.\n\nFixed by adding this case to the ignored labels.\n\ngcc/testsuite/ChangeLog:\n\n\t* lib/scanasm.exp: Add 'LB*' to the local labels that are\n\tignored for Darwin.\n\nSigned-off-by: Iain Sandoe ","shortMessageHtmlLink":"testsuite, Darwin: Account for block labels in function body scans."}},{"before":"b120e355e59142bd15d6b010461d07236288d843","after":"d60968de6961cef144a5cf8701ea0d3f4ea90f18","ref":"refs/heads/master","pushedAt":"2024-04-03T12:21:18.000Z","pushType":"push","commitsCount":24,"pusher":{"login":"iains","name":"Iain Sandoe","path":"/iains","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/4039407?s=80&v=4"},"commit":{"message":"libphobos, Darwin: Enable libphobos for most Darwin.\n\nEarlier Darwin systems can be made to work too - but they need non-\nstandard 'binutils', so for now these must be enabled specifically.\n\nlibphobos/ChangeLog:\n\n\t* configure.tgt: Enable libphobos for Darwin >= 12.\n\nSigned-off-by: Iain Sandoe ","shortMessageHtmlLink":"libphobos, Darwin: Enable libphobos for most Darwin."}},{"before":"7907ff2bcb50d1d88dc8d22e740de33ab151d00c","after":"b120e355e59142bd15d6b010461d07236288d843","ref":"refs/heads/master","pushedAt":"2024-04-02T12:05:39.000Z","pushType":"push","commitsCount":47,"pusher":{"login":"iains","name":"Iain Sandoe","path":"/iains","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/4039407?s=80&v=4"},"commit":{"message":"jit, Darwin: Implement library exports list.\n\nCurrently, we have no exports list for libgccjit, which means that\nall symbols are exported, including those from libstdc++ which is\nlinked statically into the lib. This causes failures when the\nshared libstdc++ is used but some c++ symbols are satisfied from\nlibgccjit.\n\nThis implements an export file for Darwin (which is currently\nmanually created by cross-checking libgccjit.map). Ideally we'd\nscript this, at some point. Update libtool current and age to\nreflect the current ABI version (we are not bumping the SO name\nat this stage).\n\nThis fixes a number of new failures in jit testing.\n\ngcc/jit/ChangeLog:\n\n\t* Make-lang.in: Implement exports list, and use a shared\n\tlibgcc.\n\t* libgccjit.exports: New file.\n\nSigned-off-by: Iain Sandoe ","shortMessageHtmlLink":"jit, Darwin: Implement library exports list."}},{"before":"bb04a11418f54c4c062802d491ff2c3720b7313e","after":"7907ff2bcb50d1d88dc8d22e740de33ab151d00c","ref":"refs/heads/master","pushedAt":"2024-03-28T09:19:31.000Z","pushType":"push","commitsCount":53,"pusher":{"login":"iains","name":"Iain Sandoe","path":"/iains","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/4039407?s=80&v=4"},"commit":{"message":"modula2: Rebuild documentation sections for target independent libs\n\nThis patch rebuilds the documentation for the target independent\nlibrary sections.\n\ngcc/m2/ChangeLog:\n\n\t* target-independent/m2/Builtins.texi: Rebuilt.\n\t* target-independent/m2/gm2-libs.texi: Rebuilt.\n\nSigned-off-by: Gaius Mulley ","shortMessageHtmlLink":"modula2: Rebuild documentation sections for target independent libs"}},{"before":"4adb1a5839e7a3310a127c1776f1f95d7edaa6ff","after":"bb04a11418f54c4c062802d491ff2c3720b7313e","ref":"refs/heads/master","pushedAt":"2024-03-24T17:27:03.000Z","pushType":"push","commitsCount":106,"pusher":{"login":"iains","name":"Iain Sandoe","path":"/iains","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/4039407?s=80&v=4"},"commit":{"message":"Daily bump.","shortMessageHtmlLink":"Daily bump."}},{"before":null,"after":"2bd4e1b5716618ef9ac6be50f1dee60449323ced","ref":"refs/heads/master-wip-classic-expt","pushedAt":"2024-03-24T11:43:27.000Z","pushType":"branch_creation","commitsCount":0,"pusher":{"login":"iains","name":"Iain Sandoe","path":"/iains","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/4039407?s=80&v=4"},"commit":{"message":"collect2, darwin: Allow -fuse-ld=classic\n\nThis allows for a command line option of -fuse-ld=classic to switch to the ld64\nlinker implementation on systems with the dyld linker.\n\ngcc/ChangeLog:\n\n\t* collect2.cc (main):\n\t(printf):\n\t* common.opt:\n\t* config/darwin.h (GTY):\n\t* configure: Regenerate.\n\t* configure.ac:\n\t* exec-tool.in:\n\t* opts.cc (common_handle_option):\n\nSigned-off-by: Iain Sandoe ","shortMessageHtmlLink":"collect2, darwin: Allow -fuse-ld=classic"}},{"before":"2d454f982914c481a268f1c63e431b2682cc3be0","after":"4adb1a5839e7a3310a127c1776f1f95d7edaa6ff","ref":"refs/heads/master","pushedAt":"2024-03-19T13:04:53.000Z","pushType":"push","commitsCount":38,"pusher":{"login":"iains","name":"Iain Sandoe","path":"/iains","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/4039407?s=80&v=4"},"commit":{"message":"testsuite, Darwin: Use the IOKit framework in framework-1.c [PR114049].\n\nThe intent of the test is to show that we find a framework that\nis installed in /System/Library/Frameworks when the user has added\na '-F' option. The trick is to choose some header that is present\nfor all the Darwin versions we support and that does not contain any\ncontent we cannot parse. We had been using the Kernel framework for\nthis, but recent SDK versions have revealed that this is not suitable.\n\nReplacing with a use of IOKit.\n\n\tPR target/114049\n\ngcc/testsuite/ChangeLog:\n\n\t* gcc.dg/framework-1.c: Use an IOKit header instead of a\n\tKernel one.\n\nSigned-off-by: Iain Sandoe ","shortMessageHtmlLink":"testsuite, Darwin: Use the IOKit framework in framework-1.c [PR114049]."}},{"before":"e9753f4b633608ae4adc6efb747e638783cd6196","after":"2d454f982914c481a268f1c63e431b2682cc3be0","ref":"refs/heads/master","pushedAt":"2024-03-17T16:41:17.000Z","pushType":"push","commitsCount":96,"pusher":{"login":"iains","name":"Iain Sandoe","path":"/iains","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/4039407?s=80&v=4"},"commit":{"message":"d: Merge upstream dmd, druntime 855353a1d9\n\nD front-end changes:\n\n\t- Import dmd v2.108.0-rc.1.\n\t- Add support for Named Arguments for functions.\n\t- Hex strings now convert to integer arrays.\n\nD runtime changes:\n\n\t- Import druntime v2.108.0-rc.1.\n\ngcc/d/ChangeLog:\n\n\t* dmd/MERGE: Merge upstream dmd 855353a1d9.\n\t* dmd/VERSION:\n\nlibphobos/ChangeLog:\n\n\t* libdruntime/MERGE: Merge upstream druntime 855353a1d9.","shortMessageHtmlLink":"d: Merge upstream dmd, druntime 855353a1d9"}},{"before":"0c1ff8951c2f5ff5b0699bbfa7523f690deac713","after":"e9753f4b633608ae4adc6efb747e638783cd6196","ref":"refs/heads/master","pushedAt":"2024-03-09T08:58:05.000Z","pushType":"push","commitsCount":51,"pusher":{"login":"iains","name":"Iain Sandoe","path":"/iains","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/4039407?s=80&v=4"},"commit":{"message":"i386: Regenerate i386.opt.urls\n\nWhen I've added the -mnoreturn-no-callee-saved-registers option\nto i386.opt, I forgot to regenerate i386.opt.urls and Mark's\nCI kindly reminded me of that.\n\nFixed thusly.\n\n2024-03-09 Jakub Jelinek \n\n\t* config/i386/i386.opt.urls: Regenerate.","shortMessageHtmlLink":"i386: Regenerate i386.opt.urls"}},{"before":"282455d3132322885160d21e8d1da341de6b4484","after":"0c1ff8951c2f5ff5b0699bbfa7523f690deac713","ref":"refs/heads/master","pushedAt":"2024-03-07T16:00:31.000Z","pushType":"push","commitsCount":88,"pusher":{"login":"iains","name":"Iain Sandoe","path":"/iains","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/4039407?s=80&v=4"},"commit":{"message":"contrib: Update test_mklog to correspond to mklog\n\ncontrib/ChangeLog:\n\n\t* test_mklog.py: \"Moved to...\" -> \"Move to...\"\n\nSigned-off-by: Filip Kastl ","shortMessageHtmlLink":"contrib: Update test_mklog to correspond to mklog"}},{"before":null,"after":"d3a97cc2df1cbd0ffb30d9effea5b0cce3c1da83","ref":"refs/heads/master-wip-apple-si","pushedAt":"2024-03-02T20:55:22.000Z","pushType":"branch_creation","commitsCount":0,"pusher":{"login":"iains","name":"Iain Sandoe","path":"/iains","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/4039407?s=80&v=4"},"commit":{"message":"aarch64, Darwin: Set a default cpu type when none is configured.\n\nWe just choose apple-m1 here, since that should work on all released\narm64 [macOS] systems (so far). We are still not handling iOS/iPhoneOS\nso this might need revision should we eventually support them.\n\ngcc/ChangeLog:\n\n\t* config.gcc (aarch64-*-darwin*): Default to apple-m1 CPU.\n\nSigned-off-by: Iain Sandoe ","shortMessageHtmlLink":"aarch64, Darwin: Set a default cpu type when none is configured."}}],"hasNextPage":true,"hasPreviousPage":false,"activityType":"all","actor":null,"timePeriod":"all","sort":"DESC","perPage":30,"cursor":"djE6ks8AAAAEhJnKGAA","startCursor":null,"endCursor":null}},"title":"Activity · iains/gcc-git"}