diff options
| author | mrs <mrs@138bc75d-0d04-0410-961f-82ee72b054a4> | 2014-04-30 17:19:02 +0000 |
|---|---|---|
| committer | mrs <mrs@138bc75d-0d04-0410-961f-82ee72b054a4> | 2014-04-30 17:19:02 +0000 |
| commit | 951019127b9f3cf59989ceb74539d454d535d3b5 (patch) | |
| tree | 958443e6076f37a3ff7d07d9303f5c4171944638 | |
| parent | bedbed50ff969478b0c8878678d7c169a8a0cdaf (diff) | |
| parent | 8a21d39cf1ea8183eb1e95bbb4396d97ae0dc36d (diff) | |
| download | gcc-951019127b9f3cf59989ceb74539d454d535d3b5.tar.gz | |
Merge in trunk.
git-svn-id: svn+ssh://gcc.gnu.org/svn/gcc/branches/wide-int@209944 138bc75d-0d04-0410-961f-82ee72b054a4
218 files changed, 5212 insertions, 2386 deletions
diff --git a/ChangeLog b/ChangeLog index a0e55567d59..652c06d028b 100644 --- a/ChangeLog +++ b/ChangeLog @@ -1,3 +1,11 @@ +2014-04-29 Alan Lawrence <alan.lawrence@arm.com> + + * MAINTAINERS (Write After Approval): Put myself in correct order. + +2014-04-29 Alan Lawrence <alan.lawrence@arm.com> + + * MAINTAINERS (Write After Approval): Add myself. + 2014-04-24 Laurynas Biveinis <laurynas.biveinis@gmail.com> * MAINTAINERS: Move myself from Reviewers to Write After Approval diff --git a/MAINTAINERS b/MAINTAINERS index 36979e13785..b79588d3f23 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -441,6 +441,7 @@ Rask Ingemann Lambertsen ccc94453@vip.cybercity.dk Asher Langton langton2@llnl.gov Chris Lattner sabre@nondot.org Terry Laurenzo tlaurenzo@gmail.com +Alan Lawrence alan.lawrence@arm.com Georg-Johann Lay avr@gjlay.de Marc Lehmann pcg@goof.com James Lemke jwlemke@codesourcery.com diff --git a/gcc/ChangeLog b/gcc/ChangeLog index 67a4645a87a..e3ae8dcfc81 100644 --- a/gcc/ChangeLog +++ b/gcc/ChangeLog @@ -1,3 +1,251 @@ +2014-04-30 Alan Lawrence <alan.lawrence@arm.com> + + * config/aarch64/arm_neon.h (vuzp1_f32, vuzp1_p8, vuzp1_p16, vuzp1_s8, + vuzp1_s16, vuzp1_s32, vuzp1_u8, vuzp1_u16, vuzp1_u32, vuzp1q_f32, + vuzp1q_f64, vuzp1q_p8, vuzp1q_p16, vuzp1q_s8, vuzp1q_s16, vuzp1q_s32, + vuzp1q_s64, vuzp1q_u8, vuzp1q_u16, vuzp1q_u32, vuzp1q_u64, vuzp2_f32, + vuzp2_p8, vuzp2_p16, vuzp2_s8, vuzp2_s16, vuzp2_s32, vuzp2_u8, + vuzp2_u16, vuzp2_u32, vuzp2q_f32, vuzp2q_f64, vuzp2q_p8, vuzp2q_p16, + vuzp2q_s8, vuzp2q_s16, vuzp2q_s32, vuzp2q_s64, vuzp2q_u8, vuzp2q_u16, + vuzp2q_u32, vuzp2q_u64): Replace temporary asm with __builtin_shuffle. + +2014-04-30 Joern Rennecke <joern.rennecke@embecosm.com> + + * config/arc/arc.opt (mlra): Move comment above option name + to avoid mis-parsing as language options. + +2014-04-30 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE> + + * config/sol2-10.h (TARGET_LIBC_HAS_FUNCTION): Move ... + * config/sol2.h: ... here. + * config/sol2-10.h: Remove. + + * config/sol2-bi.h (WCHAR_TYPE, WCHAR_TYPE_SIZE, WINT_TYPE) + (WINT_TYPE_SIZE, MULTILIB_DEFAULTS, DEF_ARCH32_SPEC) + (DEF_ARCH64_SPEC, ASM_CPU_DEFAULT_SPEC, LINK_ARCH64_SPEC_BASE) + (LINK_ARCH64_SPEC, ARCH_DEFAULT_EMULATION, TARGET_LD_EMULATION) + (LINK_ARCH_SPEC, SUBTARGET_EXTRA_SPECS): Move ... + * config/sol2.h: ... here. + (SECTION_NAME_FORMAT): Don't redefine. + (STARTFILE_ARCH32_SPEC): Rename to ... + (STARTFILE_ARCH_SPEC): ... this. + (ASM_OUTPUT_ALIGNED_COMMON): Move ... + * config/sparc/sol2.h: ... here. + (SECTION_NAME_FORMAT): Don't undef. + * config/i386/sol2.h (ASM_CPU_DEFAULT_SPEC) + (SUBTARGET_EXTRA_SPECS): Remove. + * config/sparc/sol2.h (ASM_CPU_DEFAULT_SPEC): Remove. + + * config/i386/sol2-bi.h (TARGET_SUBTARGET_DEFAULT) + (MD_STARTFILE_PREFIX): Remove. + (SUBTARGET_OPTIMIZATION_OPTIONS, ASM_CPU32_DEFAULT_SPEC) + (ASM_CPU64_DEFAULT_SPEC, ASM_CPU_SPEC, ASM_SPEC, DEFAULT_ARCH32_P) + (ARCH64_SUBDIR, ARCH32_EMULATION, ARCH64_EMULATION) + (ASM_COMMENT_START, JUMP_TABLES_IN_TEXT_SECTION) + (ASM_OUTPUT_DWARF_PCREL, ASM_OUTPUT_ALIGNED_COMMON) + (USE_IX86_FRAME_POINTER, USE_X86_64_FRAME_POINTER): Move ... + * config/i386/sol2.h: ... here. + (TARGET_SUBTARGET_DEFAULT, SIZE_TYPE, PTRDIFF_TYPE): Remove. + * config/i386/sol2-bi.h: Remove. + * config/sol2.h (MD_STARTFILE_PREFIX): Remove. + (LINK_ARCH32_SPEC_BASE): Remove /usr/ccs/lib/libp, /usr/ccs/lib. + + * config/i386/t-sol2-64: Rename to ... + * config/i386/t-sol2: ... this. + * config/sparc/t-sol2-64: Rename to ... + * config/sparc/t-sol2: ... this. + + * config.gcc (*-*-solaris2*): Split sol2_tm_file into + sol2_tm_file_head, sol2_tm_file_tail. + Include ${cpu_type}/sol2.h before sol2.h. + Remove sol2-10.h. + (i[34567]86-*-solaris2* | x86_64-*-solaris2.1[0-9]*): Include + i386/x86-64.h between sol2_tm_file_head and sol2_tm_file_tail. + Remove i386/sol2-bi.h, sol2-bi.h from tm_file. + Reflect i386/t-sol2-64 renaming. + (sparc*-*-solaris2*): Remove sol2-bi.h from tm_file. + Reflect sparc/t-sol2-64 renaming. + +2014-04-30 Richard Biener <rguenther@suse.de> + + * passes.c (execute_function_todo): Move TODO_verify_stmts + and TODO_verify_ssa under the TODO_verify_il umbrella. + * tree-ssa.h (verify_ssa): Adjust prototype. + * tree-ssa.c (verify_ssa): Add parameter to tell whether + we should verify SSA operands. + * tree-cfg.h (verify_gimple_in_cfg): Adjust prototype. + * tree-cfg.c (verify_gimple_in_cfg): Add parameter to tell + whether we should verify whether not throwing stmts have EH info. + * graphite-scop-detection.c (create_sese_edges): Adjust. + * tree-ssa-loop-manip.c (verify_loop_closed_ssa): Likewise. + * tree-eh.c (lower_try_finally_switch): Do not add the + default case label twice. + +2014-04-30 Marek Polacek <polacek@redhat.com> + + * gcc.c (sanitize_spec_function): Handle SANITIZE_FLOAT_DIVIDE. + * builtins.def: Initialize builtins even for SANITIZE_FLOAT_DIVIDE. + * flag-types.h (enum sanitize_code): Add SANITIZE_FLOAT_DIVIDE. + * opts.c (common_handle_option): Add -fsanitize=float-divide-by-zero. + +2014-04-29 Alan Lawrence <alan.lawrence@arm.com> + + * config/aarch64/arm_neon.h (vzip1_f32, vzip1_p8, vzip1_p16, vzip1_s8, + vzip1_s16, vzip1_s32, vzip1_u8, vzip1_u16, vzip1_u32, vzip1q_f32, + vzip1q_f64, vzip1q_p8, vzip1q_p16, vzip1q_s8, vzip1q_s16, vzip1q_s32, + vzip1q_s64, vzip1q_u8, vzip1q_u16, vzip1q_u32, vzip1q_u64, vzip2_f32, + vzip2_p8, vzip2_p16, vzip2_s8, vzip2_s16, vzip2_s32, vzip2_u8, + vzip2_u16, vzip2_u32, vzip2q_f32, vzip2q_f64, vzip2q_p8, vzip2q_p16, + vzip2q_s8, vzip2q_s16, vzip2q_s32, vzip2q_s64, vzip2q_u8, vzip2q_u16, + vzip2q_u32, vzip2q_u64): Replace inline __asm__ with __builtin_shuffle. + +2014-04-29 David Malcolm <dmalcolm@redhat.com> + + * tree-cfg.c (dump_function_to_file): Dump the return type of + functions, in a line to itself before the function body, mimicking + the layout of a C function. + +2014-04-29 Jakub Jelinek <jakub@redhat.com> + + PR tree-optimization/60971 + * tree-tailcall.c (process_assignment): Reject conversions which + reduce precision. + +2014-04-29 James Greenhalgh <james.greenhalgh@arm.com> + + * calls.c (initialize_argument_information): Always treat + PUSH_ARGS_REVERSED as 1, simplify code accordingly. + (expand_call): Likewise. + (emit_library_call_calue_1): Likewise. + * expr.c (PUSH_ARGS_REVERSED): Do not define. + (emit_push_insn): Always treat PUSH_ARGS_REVERSED as 1, simplify + code accordingly. + +2014-04-29 Nick Clifton <nickc@redhat.com> + + * config/msp430/msp430.md (umulsidi): Fix typo. + (mulhisi3): Enable even inside interrupt handlers. + * config/msp430/msp430.c (msp430_print_operand): %O: Allow for the + bigger return address pushed in large mode. + +2014-04-29 Nick Clifton <nickc@redhat.com> + + * config/arc/arc.c (arc_select_cc_mode): Fix parentheses. + (arc_init_reg_tables): Use a machine_mode enum to iterate over + available modes. + * config/m32r/m32r.c (init_reg_tables): Likewise. + * config/m32c/m32c.c (m32c_illegal_subreg_p): Use a machine_mode + enum to hold the modes. + +2014-04-29 Richard Biener <rguenther@suse.de> + + * dominance.c (free_dominance_info): Add overload with + function parameter. + (dom_info_state): Likewise. + (dom_info_available_p): Likewise. + * basic-block.h (free_dominance_info, dom_info_state, + dom_info_available_p): Declare overloads. + * passes.c (execute_function_todo): Verify that verifiers + don't change dominator info state. Drop dominator info + for IPA pass invocations. + * cgraph.c (release_function_body): Restore asserts that + dominator information is released. + +2014-04-29 Patrick Palka <patrick@parcs.ath.cx> + + * doc/invoke.texi: Fix typo. + * tree-vrp.c: Fix typos. + * gimple.c (infer_nonnull_range): Reorder operands of an && + condition. + +2014-04-29 Zhenqiang Chen <zhenqiang.chen@linaro.org> + + * config/aarch64/aarch64.md (mov<mode>cc): New for GPF. + +2014-04-28 James Greenhalgh <james.greenhalgh@arm.com> + + * config/aarch64/aarch64-builtins.c + (aarch64_types_storestruct_lane_qualifiers): New. + (TYPES_STORESTRUCT_LANE): Likewise. + * config/aarch64/aarch64-simd-builtins.def (st2_lane): New. + (st3_lane): Likewise. + (st4_lane): Likewise. + * config/aarch64/aarch64-simd.md (vec_store_lanesoi_lane<mode>): New. + (vec_store_lanesci_lane<mode>): Likewise. + (vec_store_lanesxi_lane<mode>): Likewise. + (aarch64_st2_lane<VQ:mode>): Likewise. + (aarch64_st3_lane<VQ:mode>): Likewise. + (aarch64_st4_lane<VQ:mode>): Likewise. + * config/aarch64/aarch64.md (unspec): Add UNSPEC_ST{2,3,4}_LANE. + * config/aarch64/arm_neon.h + (__ST2_LANE_FUNC): Rewrite using builtins, update use points to + use new macro arguments. + (__ST3_LANE_FUNC): Likewise. + (__ST4_LANE_FUNC): Likewise. + * config/aarch64/iterators.md (V_TWO_ELEM): New. + (V_THREE_ELEM): Likewise. + (V_FOUR_ELEM): Likewise. + +2014-04-28 David Malcolm <dmalcolm@redhat.com> + + * doc/gimple.texi: Replace the description of the now-defunct + union gimple_statement_d with a diagram showing the + gimple_statement_base class hierarchy and its relationships to + the GSS_ and GIMPLE_ enums. + +2014-04-28 James Greenhalgh <james.greenhalgh@arm.com> + + * config/aarch64/aarch64-protos.h (aarch64_modes_tieable_p): New. + * config/aarch64/aarch64.c + (aarch64_cannot_change_mode_class): Weaken conditions. + (aarch64_modes_tieable_p): New. + * config/aarch64/aarch64.h (MODES_TIEABLE_P): Use it. + +2014-04-28 Pat Haugen <pthaugen@us.ibm.com> + + * config/rs6000/sync.md (AINT mode_iterator): Move definition. + (loadsync_<mode>): Change mode. + (load_quadpti, store_quadpti): New. + (atomic_load<mode>, atomic_store<mode>): Add support for TI mode. + * config/rs6000/rs6000.md (unspec enum): Add UNSPEC_LSQ. + +2014-04-28 Martin Jambor <mjambor@suse.cz> + + * tree-sra.c (sra_modify_expr): Generate new memory accesses with + same alias type as the original statement. + (subreplacement_assignment_data): New type. + (handle_unscalarized_data_in_subtree): New type of parameter, + generate new memory accesses with same alias type as the original + statement. + (load_assign_lhs_subreplacements): Likewise. + (sra_modify_constructor_assign): Generate new memory accesses with + same alias type as the original statement. + +2014-04-28 Richard Biener <rguenther@suse.de> + + * tree-pass.h (TODO_verify_il): Define. + (TODO_verify_all): Complete properly. + * passes.c (execute_function_todo): Move existing loop-closed + SSA verification under TODO_verify_il. + (execute_one_pass): Trigger TODO_verify_il at todo-after time. + * graphite-sese-to-poly.c (rewrite_cross_bb_scalar_deps): + Fix tree sharing issue. + +2014-04-28 Richard Biener <rguenther@suse.de> + + PR middle-end/60092 + * builtins.def (DEF_C11_BUILTIN): Add. + (BUILT_IN_ALIGNED_ALLOC): Likewise. + * coretypes.h (enum function_class): Add function_c11_misc. + * tree-ssa-alias.c (ref_maybe_used_by_call_p_1): Handle + BUILT_IN_ALIGNED_ALLOC like BUILT_IN_MALLOC. + (call_may_clobber_ref_p_1): Likewise. + * tree-ssa-dce.c (mark_stmt_if_obviously_necessary): Likewise. + (mark_all_reaching_defs_necessary_1): Likewise. + (propagate_necessity): Likewise. + (eliminate_unnecessary_stmts): Likewise. + * tree-ssa-ccp.c (evaluate_stmt): Handle BUILT_IN_ALIGNED_ALLOC. + 2014-04-28 Richard Biener <rguenther@suse.de> * tree-vrp.c (vrp_var_may_overflow): Remove. diff --git a/gcc/DATESTAMP b/gcc/DATESTAMP index 2eee80b2cf1..5907e7a3156 100644 --- a/gcc/DATESTAMP +++ b/gcc/DATESTAMP @@ -1 +1 @@ -20140428 +20140430 diff --git a/gcc/ada/ChangeLog b/gcc/ada/ChangeLog index a31cd3c2f11..fb57af5c0d4 100644 --- a/gcc/ada/ChangeLog +++ b/gcc/ada/ChangeLog @@ -1,3 +1,43 @@ +2014-04-28 Richard Henderson <rth@redhat.com> + + * gcc-interface/Makefile.in: Support aarch64-linux. + +2014-04-28 Eric Botcazou <ebotcazou@adacore.com> + + * exp_dbug.ads (Get_External_Name): Add 'False' default to Has_Suffix, + add 'Suffix' parameter and adjust comment. + (Get_External_Name_With_Suffix): Delete. + * exp_dbug.adb (Get_External_Name_With_Suffix): Merge into... + (Get_External_Name): ...here. Add 'False' default to Has_Suffix, add + 'Suffix' parameter. + (Get_Encoded_Name): Remove 2nd argument in call to Get_External_Name. + Call Get_External_Name instead of Get_External_Name_With_Suffix. + (Get_Secondary_DT_External_Name): Likewise. + * exp_cg.adb (Write_Call_Info): Likewise. + * exp_disp.adb (Export_DT): Likewise. + (Import_DT): Likewise. + * comperr.ads (Compiler_Abort): Remove Code parameter and add From_GCC + parameter with False default. + * comperr.adb (Compiler_Abort): Likewise. Adjust accordingly. + * types.h (Fat_Pointer): Rename into... + (String_Pointer): ...this. Add comment on interfacing rules. + * fe.h (Compiler_Abort): Adjust for above renaming. + (Error_Msg_N): Likewise. + (Error_Msg_NE): Likewise. + (Get_External_Name): Likewise. Add third parameter. + (Get_External_Name_With_Suffix): Delete. + * gcc-interface/decl.c (STDCALL_PREFIX): Define. + (create_concat_name): Adjust call to Get_External_Name, remove call to + Get_External_Name_With_Suffix, use STDCALL_PREFIX, adjust for renaming. + * gcc-interface/trans.c (post_error): Likewise. + (post_error_ne): Likewise. + * gcc-interface/misc.c (internal_error_function): Likewise. + +2014-04-28 Richard Biener <rguenther@suse.de> + + PR middle-end/60092 + * gcc-interface/utils.c: Define flag_isoc11. + 2014-04-26 Eric Botcazou <ebotcazou@adacore.com> * gnatvsn.ads (Library_Version): Bump to 4.10. diff --git a/gcc/ada/comperr.adb b/gcc/ada/comperr.adb index 13646a5c155..7a9d7070cde 100644 --- a/gcc/ada/comperr.adb +++ b/gcc/ada/comperr.adb @@ -6,7 +6,7 @@ -- -- -- B o d y -- -- -- --- Copyright (C) 1992-2013, Free Software Foundation, Inc. -- +-- Copyright (C) 1992-2014, Free Software Foundation, Inc. -- -- -- -- GNAT is free software; you can redistribute it and/or modify it under -- -- terms of the GNU General Public License as published by the Free Soft- -- @@ -74,8 +74,8 @@ package body Comperr is procedure Compiler_Abort (X : String; - Code : Integer := 0; - Fallback_Loc : String := "") + Fallback_Loc : String := ""; + From_GCC : Boolean := False) is -- The procedures below output a "bug box" with information about -- the cause of the compiler abort and about the preferred method @@ -206,7 +206,7 @@ package body Comperr is Write_Str (") "); if X'Length + Column > 76 then - if Code < 0 then + if From_GCC then Write_Str ("GCC error:"); end if; @@ -235,11 +235,7 @@ package body Comperr is Write_Str (X); end if; - if Code > 0 then - Write_Str (", Code="); - Write_Int (Int (Code)); - - elsif Code = 0 then + if not From_GCC then -- For exception case, get exception message from the TSD. Note -- that it would be neater and cleaner to pass the exception diff --git a/gcc/ada/comperr.ads b/gcc/ada/comperr.ads index ba3cb6b8f66..dccd8ef3433 100644 --- a/gcc/ada/comperr.ads +++ b/gcc/ada/comperr.ads @@ -6,7 +6,7 @@ -- -- -- S p e c -- -- -- --- Copyright (C) 1992-2013, Free Software Foundation, Inc. -- +-- Copyright (C) 1992-2014, Free Software Foundation, Inc. -- -- -- -- GNAT is free software; you can redistribute it and/or modify it under -- -- terms of the GNU General Public License as published by the Free Soft- -- @@ -31,8 +31,8 @@ package Comperr is procedure Compiler_Abort (X : String; - Code : Integer := 0; - Fallback_Loc : String := ""); + Fallback_Loc : String := ""; + From_GCC : Boolean := False); pragma No_Return (Compiler_Abort); -- Signals an internal compiler error. Never returns control. Depending on -- processing may end up raising Unrecoverable_Error, or exiting directly. @@ -46,10 +46,9 @@ package Comperr is -- Note that this is only used at the outer level (to handle constraint -- errors or assert errors etc.) In the normal logic of the compiler we -- always use pragma Assert to check for errors, and if necessary an - -- explicit abort is achieved by pragma Assert (False). Code is positive - -- for a gigi abort (giving the gigi abort code), zero for a front - -- end exception (with possible message stored in TSD.Current_Excep, - -- and negative (an unused value) for a GCC abort. + -- explicit abort is achieved by pragma Assert (False). From_GCC is true + -- for a GCC abort and false for a front end exception (with a possible + -- message stored in TSD.Current_Excep). procedure Delete_SCIL_Files; -- Delete SCIL files associated with the main unit diff --git a/gcc/ada/exp_cg.adb b/gcc/ada/exp_cg.adb index d8a7022e504..483f174efc6 100644 --- a/gcc/ada/exp_cg.adb +++ b/gcc/ada/exp_cg.adb @@ -6,7 +6,7 @@ -- -- -- B o d y -- -- -- --- Copyright (C) 2010-2013, Free Software Foundation, Inc. -- +-- Copyright (C) 2010-2014, Free Software Foundation, Inc. -- -- -- -- GNAT is free software; you can redistribute it and/or modify it under -- -- terms of the GNU General Public License as published by the Free Soft- -- @@ -437,10 +437,10 @@ package body Exp_CG is if Nkind (P) = N_Subprogram_Body and then not Acts_As_Spec (P) then - Get_External_Name (Corresponding_Spec (P), Has_Suffix => False); + Get_External_Name (Corresponding_Spec (P)); else - Get_External_Name (Defining_Entity (P), Has_Suffix => False); + Get_External_Name (Defining_Entity (P)); end if; Write_Str (Name_Buffer (1 .. Name_Len)); diff --git a/gcc/ada/exp_dbug.adb b/gcc/ada/exp_dbug.adb index 7dc4264cc59..13620290815 100644 --- a/gcc/ada/exp_dbug.adb +++ b/gcc/ada/exp_dbug.adb @@ -507,8 +507,8 @@ package body Exp_Dbug is begin -- If not generating code, there is no need to create encoded names, and -- problems when the back-end is called to annotate types without full - -- code generation. See comments in Get_External_Name_With_Suffix for - -- additional details. + -- code generation. See comments in Get_External_Name for additional + -- details. -- However we do create encoded names if the back end is active, even -- if Operating_Mode got reset. Otherwise any serious error reported @@ -556,7 +556,7 @@ package body Exp_Dbug is -- Fixed-point case if Is_Fixed_Point_Type (E) then - Get_External_Name_With_Suffix (E, "XF_"); + Get_External_Name (E, True, "XF_"); Add_Real_To_Buffer (Delta_Value (E)); if Small_Value (E) /= Delta_Value (E) then @@ -568,14 +568,14 @@ package body Exp_Dbug is elsif Vax_Float (E) then if Digits_Value (Base_Type (E)) = 6 then - Get_External_Name_With_Suffix (E, "XFF"); + Get_External_Name (E, True, "XFF"); elsif Digits_Value (Base_Type (E)) = 9 then - Get_External_Name_With_Suffix (E, "XFF"); + Get_External_Name (E, True, "XFF"); else pragma Assert (Digits_Value (Base_Type (E)) = 15); - Get_External_Name_With_Suffix (E, "XFG"); + Get_External_Name (E, True, "XFG"); end if; -- Discrete case where bounds do not match size @@ -607,9 +607,9 @@ package body Exp_Dbug is begin if Biased then - Get_External_Name_With_Suffix (E, "XB"); + Get_External_Name (E, True, "XB"); else - Get_External_Name_With_Suffix (E, "XD"); + Get_External_Name (E, True, "XD"); end if; if Lo_Encode or Hi_Encode then @@ -649,7 +649,7 @@ package body Exp_Dbug is else Has_Suffix := False; - Get_External_Name (E, Has_Suffix); + Get_External_Name (E); end if; if Debug_Flag_B and then Has_Suffix then @@ -667,7 +667,11 @@ package body Exp_Dbug is -- Get_External_Name -- ----------------------- - procedure Get_External_Name (Entity : Entity_Id; Has_Suffix : Boolean) is + procedure Get_External_Name + (Entity : Entity_Id; + Has_Suffix : Boolean := False; + Suffix : String := "") + is E : Entity_Id := Entity; Kind : Entity_Kind; @@ -704,6 +708,20 @@ package body Exp_Dbug is -- Start of processing for Get_External_Name begin + -- If we are not in code generation mode, this procedure may still be + -- called from Back_End (more specifically - from gigi for doing type + -- representation annotation or some representation-specific checks). + -- But in this mode there is no need to mess with external names. + + -- Furthermore, the call causes difficulties in this case because the + -- string representing the homonym number is not correctly reset as a + -- part of the call to Output_Homonym_Numbers_Suffix (which is not + -- called in gigi). + + if Operating_Mode /= Generate_Code then + return; + end if; + Reset_Buffers; -- If this is a child unit, we want the child @@ -762,42 +780,13 @@ package body Exp_Dbug is Get_Qualified_Name_And_Append (E); end if; - Name_Buffer (Name_Len + 1) := ASCII.NUL; - end Get_External_Name; - - ----------------------------------- - -- Get_External_Name_With_Suffix -- - ----------------------------------- - - procedure Get_External_Name_With_Suffix - (Entity : Entity_Id; - Suffix : String) - is - Has_Suffix : constant Boolean := (Suffix /= ""); - - begin - -- If we are not in code generation mode, this procedure may still be - -- called from Back_End (more specifically - from gigi for doing type - -- representation annotation or some representation-specific checks). - -- But in this mode there is no need to mess with external names. - - -- Furthermore, the call causes difficulties in this case because the - -- string representing the homonym number is not correctly reset as a - -- part of the call to Output_Homonym_Numbers_Suffix (which is not - -- called in gigi). - - if Operating_Mode /= Generate_Code then - return; - end if; - - Get_External_Name (Entity, Has_Suffix); - if Has_Suffix then Add_Str_To_Name_Buffer ("___"); Add_Str_To_Name_Buffer (Suffix); - Name_Buffer (Name_Len + 1) := ASCII.NUL; end if; - end Get_External_Name_With_Suffix; + + Name_Buffer (Name_Len + 1) := ASCII.NUL; + end Get_External_Name; -------------------------- -- Get_Variant_Encoding -- @@ -944,7 +933,7 @@ package body Exp_Dbug is Suffix_Index : Int) is begin - Get_External_Name (Typ, Has_Suffix => False); + Get_External_Name (Typ); if Ancestor_Typ /= Typ then declare @@ -952,7 +941,7 @@ package body Exp_Dbug is Save_Str : constant String (1 .. Name_Len) := Name_Buffer (1 .. Name_Len); begin - Get_External_Name (Ancestor_Typ, Has_Suffix => False); + Get_External_Name (Ancestor_Typ); -- Append the extended name of the ancestor to the -- extended name of Typ diff --git a/gcc/ada/exp_dbug.ads b/gcc/ada/exp_dbug.ads index 86099f66f75..6f27bfe0e3d 100644 --- a/gcc/ada/exp_dbug.ads +++ b/gcc/ada/exp_dbug.ads @@ -413,10 +413,11 @@ package Exp_Dbug is procedure Get_External_Name (Entity : Entity_Id; - Has_Suffix : Boolean); - -- Set Name_Buffer and Name_Len to the external name of entity E. The + Has_Suffix : Boolean := False; + Suffix : String := ""); + -- Set Name_Buffer and Name_Len to the external name of the entity. The -- external name is the Interface_Name, if specified, unless the entity - -- has an address clause or a suffix. + -- has an address clause or Has_Suffix is true. -- -- If the Interface is not present, or not used, the external name is the -- concatenation of: @@ -428,26 +429,11 @@ package Exp_Dbug is -- - the string "$" (or "__" if target does not allow "$"), followed -- by homonym suffix, if the entity is an overloaded subprogram -- or is defined within an overloaded subprogram. - - procedure Get_External_Name_With_Suffix - (Entity : Entity_Id; - Suffix : String); - -- Set Name_Buffer and Name_Len to the external name of entity E. If - -- Suffix is the empty string the external name is as above, otherwise - -- the external name is the concatenation of: - -- - -- - the string "_ada_", if the entity is a library subprogram, - -- - the names of any enclosing scopes, each followed by "__", - -- or "X_" if the next entity is a subunit) - -- - the name of the entity - -- - the string "$" (or "__" if target does not allow "$"), followed - -- by homonym suffix, if the entity is an overloaded subprogram - -- or is defined within an overloaded subprogram. - -- - the string "___" followed by Suffix + -- - the string "___" followed by Suffix if Has_Suffix is true. -- -- Note that a call to this procedure has no effect if we are not -- generating code, since the necessary information for computing the - -- proper encoded name is not available in this case. + -- proper external name is not available in this case. -------------------------------------------- -- Subprograms for Handling Qualification -- diff --git a/gcc/ada/exp_disp.adb b/gcc/ada/exp_disp.adb index 8ed3b3956c2..da2b55d3d9a 100644 --- a/gcc/ada/exp_disp.adb +++ b/gcc/ada/exp_disp.adb @@ -3913,10 +3913,7 @@ package body Exp_Disp is pragma Assert (Related_Type (Node (Elmt)) = Typ); - Get_External_Name - (Entity => Node (Elmt), - Has_Suffix => True); - + Get_External_Name (Node (Elmt)); Set_Interface_Name (DT, Make_String_Literal (Loc, Strval => String_From_Name_Buffer)); @@ -7088,7 +7085,7 @@ package body Exp_Disp is Set_Scope (DT, Current_Scope); - Get_External_Name (DT, True); + Get_External_Name (DT); Set_Interface_Name (DT, Make_String_Literal (Loc, Strval => String_From_Name_Buffer)); diff --git a/gcc/ada/fe.h b/gcc/ada/fe.h index d9fe48b5baa..3d670dc7bc3 100644 --- a/gcc/ada/fe.h +++ b/gcc/ada/fe.h @@ -29,17 +29,20 @@ * * ****************************************************************************/ -/* This file contains definitions to access front-end functions and - variables used by gigi. */ +/* This file contains declarations to access front-end functions and variables + used by gigi. + + WARNING: functions taking String_Pointer parameters must abide by the rule + documented alongside the definition of String_Pointer in types.h. */ #ifdef __cplusplus extern "C" { #endif -/* comperr: */ +/* comperr: */ #define Compiler_Abort comperr__compiler_abort -extern int Compiler_Abort (Fat_Pointer, int, Fat_Pointer) ATTRIBUTE_NORETURN; +extern int Compiler_Abort (String_Pointer, String_Pointer, Boolean) ATTRIBUTE_NORETURN; /* csets: */ @@ -72,8 +75,6 @@ extern void Set_Mechanism (Entity_Id, Mechanism_Type); extern void Set_RM_Size (Entity_Id, Uint); extern void Set_Present_Expr (Node_Id, Uint); -/* Test if the node N is the name of an entity (i.e. is an identifier, - expanded name, or an attribute reference that returns an entity). */ #define Is_Entity_Name einfo__is_entity_name extern Boolean Is_Entity_Name (Node_Id); @@ -90,8 +91,8 @@ extern Node_Id Get_Attribute_Definition_Clause (Entity_Id, char); #define Error_Msg_NE errout__error_msg_ne #define Set_Identifier_Casing errout__set_identifier_casing -extern void Error_Msg_N (Fat_Pointer, Node_Id); -extern void Error_Msg_NE (Fat_Pointer, Node_Id, Entity_Id); +extern void Error_Msg_N (String_Pointer, Node_Id); +extern void Error_Msg_NE (String_Pointer, Node_Id, Entity_Id); extern void Set_Identifier_Casing (Char *, const Char *); /* err_vars: */ @@ -147,11 +148,9 @@ extern void Setup_Asm_Outputs (Node_Id); #define Get_Encoded_Name exp_dbug__get_encoded_name #define Get_External_Name exp_dbug__get_external_name -#define Get_External_Name_With_Suffix exp_dbug__get_external_name_with_suffix -extern void Get_Encoded_Name (Entity_Id); -extern void Get_External_Name (Entity_Id, Boolean); -extern void Get_External_Name_With_Suffix (Entity_Id, Fat_Pointer); +extern void Get_Encoded_Name (Entity_Id); +extern void Get_External_Name (Entity_Id, Boolean, String_Pointer); /* exp_util: */ diff --git a/gcc/ada/gcc-interface/Makefile.in b/gcc/ada/gcc-interface/Makefile.in index 9af1967ce9f..5c36962ef3b 100644 --- a/gcc/ada/gcc-interface/Makefile.in +++ b/gcc/ada/gcc-interface/Makefile.in @@ -1988,6 +1988,44 @@ ifeq ($(strip $(filter-out arm% linux-gnueabi%,$(target_cpu) $(target_os))),) LIBRARY_VERSION := $(LIB_VERSION) endif +# AArch64 Linux +ifeq ($(strip $(filter-out aarch64% linux%,$(target_cpu) $(target_os))),) + LIBGNAT_TARGET_PAIRS = \ + a-exetim.adb<a-exetim-posix.adb \ + a-exetim.ads<a-exetim-default.ads \ + a-intnam.ads<a-intnam-linux.ads \ + a-synbar.adb<a-synbar-posix.adb \ + a-synbar.ads<a-synbar-posix.ads \ + s-inmaop.adb<s-inmaop-posix.adb \ + s-intman.adb<s-intman-posix.adb \ + s-linux.ads<s-linux.ads \ + s-mudido.adb<s-mudido-affinity.adb \ + s-osinte.ads<s-osinte-linux.ads \ + s-osinte.adb<s-osinte-posix.adb \ + s-osprim.adb<s-osprim-posix.adb \ + s-taprop.adb<s-taprop-linux.adb \ + s-tasinf.ads<s-tasinf-linux.ads \ + s-tasinf.adb<s-tasinf-linux.adb \ + s-tpopsp.adb<s-tpopsp-tls.adb \ + s-taspri.ads<s-taspri-posix.ads \ + g-sercom.adb<g-sercom-linux.adb \ + $(ATOMICS_TARGET_PAIRS) \ + $(ATOMICS_BUILTINS_TARGET_PAIRS) \ + system.ads<system-linux-x86_64.ads + ## ^^ Note the above is a pretty-close placeholder. + + TOOLS_TARGET_PAIRS = \ + mlib-tgt-specific.adb<mlib-tgt-specific-linux.adb \ + indepsw.adb<indepsw-gnu.adb + + EXTRA_GNATRTL_TASKING_OBJS=s-linux.o a-exetim.o + EH_MECHANISM=-gcc + THREADSLIB=-lpthread -lrt + GNATLIB_SHARED=gnatlib-shared-dual + GMEM_LIB = gmemlib + LIBRARY_VERSION := $(LIB_VERSION) +endif + # Sparc Linux ifeq ($(strip $(filter-out sparc% linux%,$(target_cpu) $(target_os))),) LIBGNAT_TARGET_PAIRS_COMMON = \ diff --git a/gcc/ada/gcc-interface/decl.c b/gcc/ada/gcc-interface/decl.c index ac2af06c45c..6d0b8b25038 100644 --- a/gcc/ada/gcc-interface/decl.c +++ b/gcc/ada/gcc-interface/decl.c @@ -72,6 +72,8 @@ #define Has_Thiscall_Convention(E) 0 #endif +#define STDCALL_PREFIX "_imp__" + /* Stack realignment is necessary for functions with foreign conventions when the ABI doesn't mandate as much as what the compiler assumes - that is, up to PREFERRED_STACK_BOUNDARY. @@ -8854,16 +8856,12 @@ get_entity_name (Entity_Id gnat_entity) tree create_concat_name (Entity_Id gnat_entity, const char *suffix) { - Entity_Kind kind = Ekind (gnat_entity); + const Entity_Kind kind = Ekind (gnat_entity); + const bool has_suffix = (suffix != NULL); + String_Template temp = {1, has_suffix ? strlen (suffix) : 0}; + String_Pointer sp = {suffix, &temp}; - if (suffix) - { - String_Template temp = {1, (int) strlen (suffix)}; - Fat_Pointer fp = {suffix, &temp}; - Get_External_Name_With_Suffix (gnat_entity, fp); - } - else - Get_External_Name (gnat_entity, 0); + Get_External_Name (gnat_entity, has_suffix, sp); /* A variable using the Stdcall convention lives in a DLL. We adjust its name to use the jump table, the _imp__NAME contains the address @@ -8871,9 +8869,9 @@ create_concat_name (Entity_Id gnat_entity, const char *suffix) if ((kind == E_Variable || kind == E_Constant) && Has_Stdcall_Convention (gnat_entity)) { - const int len = 6 + Name_Len; + const int len = strlen (STDCALL_PREFIX) + Name_Len; char *new_name = (char *) alloca (len + 1); - strcpy (new_name, "_imp__"); + strcpy (new_name, STDCALL_PREFIX); strcat (new_name, Name_Buffer); return get_identifier_with_length (new_name, len); } diff --git a/gcc/ada/gcc-interface/misc.c b/gcc/ada/gcc-interface/misc.c index a5f2881d697..fe44c6d5b3f 100644 --- a/gcc/ada/gcc-interface/misc.c +++ b/gcc/ada/gcc-interface/misc.c @@ -283,8 +283,8 @@ internal_error_function (diagnostic_context *context, text_info tinfo; char *buffer, *p, *loc; String_Template temp, temp_loc; - Fat_Pointer fp, fp_loc; - expanded_location s; + String_Pointer sp, sp_loc; + expanded_location xloc; /* Warn if plugins present. */ warn_if_plugins (); @@ -311,21 +311,21 @@ internal_error_function (diagnostic_context *context, temp.Low_Bound = 1; temp.High_Bound = p - buffer; - fp.Bounds = &temp; - fp.Array = buffer; + sp.Bounds = &temp; + sp.Array = buffer; - s = expand_location (input_location); - if (context->show_column && s.column != 0) - asprintf (&loc, "%s:%d:%d", s.file, s.line, s.column); + xloc = expand_location (input_location); + if (context->show_column && xloc.column != 0) + asprintf (&loc, "%s:%d:%d", xloc.file, xloc.line, xloc.column); else - asprintf (&loc, "%s:%d", s.file, s.line); + asprintf (&loc, "%s:%d", xloc.file, xloc.line); temp_loc.Low_Bound = 1; temp_loc.High_Bound = strlen (loc); - fp_loc.Bounds = &temp_loc; - fp_loc.Array = loc; + sp_loc.Bounds = &temp_loc; + sp_loc.Array = loc; Current_Error_Node = error_gnat_node; - Compiler_Abort (fp, -1, fp_loc); + Compiler_Abort (sp, sp_loc, true); } /* Perform all the initialization steps that are language-specific. */ diff --git a/gcc/ada/gcc-interface/trans.c b/gcc/ada/gcc-interface/trans.c index 813a1dce6f8..11b89825c23 100644 --- a/gcc/ada/gcc-interface/trans.c +++ b/gcc/ada/gcc-interface/trans.c @@ -9356,16 +9356,16 @@ void post_error (const char *msg, Node_Id node) { String_Template temp; - Fat_Pointer fp; + String_Pointer sp; if (No (node)) return; temp.Low_Bound = 1; temp.High_Bound = strlen (msg); - fp.Bounds = &temp; - fp.Array = msg; - Error_Msg_N (fp, node); + sp.Bounds = &temp; + sp.Array = msg; + Error_Msg_N (sp, node); } /* Similar to post_error, but NODE is the node at which to post the error and @@ -9375,16 +9375,16 @@ void post_error_ne (const char *msg, Node_Id node, Entity_Id ent) { String_Template temp; - Fat_Pointer fp; + String_Pointer sp; if (No (node)) return; temp.Low_Bound = 1; temp.High_Bound = strlen (msg); - fp.Bounds = &temp; - fp.Array = msg; - Error_Msg_NE (fp, node, ent); + sp.Bounds = &temp; + sp.Array = msg; + Error_Msg_NE (sp, node, ent); } /* Similar to post_error_ne, but NUM is the number to use for the '^'. */ diff --git a/gcc/ada/gcc-interface/utils.c b/gcc/ada/gcc-interface/utils.c index 9d7438c4c0b..b9b9dc1c52d 100644 --- a/gcc/ada/gcc-interface/utils.c +++ b/gcc/ada/gcc-interface/utils.c @@ -6518,6 +6518,7 @@ def_builtin_1 (enum built_in_function fncode, static int flag_isoc94 = 0; static int flag_isoc99 = 0; +static int flag_isoc11 = 0; /* Install what the common builtins.def offers. */ diff --git a/gcc/ada/types.h b/gcc/ada/types.h index dd049db908a..1330730b71b 100644 --- a/gcc/ada/types.h +++ b/gcc/ada/types.h @@ -76,11 +76,19 @@ typedef Char *Str; /* Pointer to string of Chars */ typedef Char *Str_Ptr; -/* Types for the fat pointer used for strings and the template it - points to. */ -typedef struct {int Low_Bound, High_Bound; } String_Template; -typedef struct {const char *Array; String_Template *Bounds; } - __attribute ((aligned (sizeof (char *) * 2))) Fat_Pointer; +/* Types for the fat pointer used for strings and the template it points to. + The fat pointer is conceptually a couple of pointers, but it is wrapped + up in a special record type. On the Ada side, the record is naturally + aligned (i.e. given pointer alignment) on regular platforms, but it is + given twice this alignment on strict-alignment platforms for performance + reasons. On the C side, for the sake of portability and simplicity, we + overalign it on all platforms (so the machine mode is always the same as + on the Ada side) but arrange to pass it in an even scalar position as a + parameter to functions (so the scalar parameter alignment is always the + same as on the Ada side). */ +typedef struct { int Low_Bound, High_Bound; } String_Template; +typedef struct { const char *Array; String_Template *Bounds; } + __attribute ((aligned (sizeof (char *) * 2))) String_Pointer; /* Types for Node/Entity Kinds: */ diff --git a/gcc/basic-block.h b/gcc/basic-block.h index 82729b4c810..0bf6e877145 100644 --- a/gcc/basic-block.h +++ b/gcc/basic-block.h @@ -826,10 +826,13 @@ enum cdi_direction CDI_POST_DOMINATORS = 2 }; +extern enum dom_state dom_info_state (function *, enum cdi_direction); extern enum dom_state dom_info_state (enum cdi_direction); extern void set_dom_info_availability (enum cdi_direction, enum dom_state); +extern bool dom_info_available_p (function *, enum cdi_direction); extern bool dom_info_available_p (enum cdi_direction); extern void calculate_dominance_info (enum cdi_direction); +extern void free_dominance_info (function *, enum cdi_direction); extern void free_dominance_info (enum cdi_direction); extern basic_block nearest_common_dominator (enum cdi_direction, basic_block, basic_block); diff --git a/gcc/builtins.def b/gcc/builtins.def index 5a76ba3291e..d400ecb4e8f 100644 --- a/gcc/builtins.def +++ b/gcc/builtins.def @@ -111,6 +111,13 @@ along with GCC; see the file COPYING3. If not see DEF_BUILTIN (ENUM, "__builtin_" NAME, BUILT_IN_NORMAL, TYPE, TYPE, \ true, true, !flag_isoc99, ATTRS, targetm.libc_has_function (function_c99_misc), true) +/* Like DEF_LIB_BUILTIN, except that the function is only a part of + the standard in C11 or above. */ +#undef DEF_C11_BUILTIN +#define DEF_C11_BUILTIN(ENUM, NAME, TYPE, ATTRS) \ + DEF_BUILTIN (ENUM, "__builtin_" NAME, BUILT_IN_NORMAL, TYPE, TYPE, \ + true, true, !flag_isoc11, ATTRS, targetm.libc_has_function (function_c11_misc), true) + /* Like DEF_C99_BUILTIN, but for complex math functions. */ #undef DEF_C99_COMPL_BUILTIN #define DEF_C99_COMPL_BUILTIN(ENUM, NAME, TYPE, ATTRS) \ @@ -169,7 +176,7 @@ along with GCC; see the file COPYING3. If not see DEF_BUILTIN (ENUM, "__builtin_" NAME, BUILT_IN_NORMAL, TYPE, TYPE, \ true, true, true, ATTRS, true, \ (flag_sanitize & (SANITIZE_ADDRESS | SANITIZE_THREAD \ - | SANITIZE_UNDEFINED))) + | SANITIZE_UNDEFINED | SANITIZE_FLOAT_DIVIDE))) #undef DEF_CILKPLUS_BUILTIN #define DEF_CILKPLUS_BUILTIN(ENUM, NAME, TYPE, ATTRS) \ @@ -223,6 +230,7 @@ DEF_C99_BUILTIN (BUILT_IN_ACOSH, "acosh", BT_FN_DOUBLE_DOUBLE, ATTR_MATHF DEF_C99_BUILTIN (BUILT_IN_ACOSHF, "acoshf", BT_FN_FLOAT_FLOAT, ATTR_MATHFN_FPROUNDING_ERRNO) DEF_C99_BUILTIN (BUILT_IN_ACOSHL, "acoshl", BT_FN_LONGDOUBLE_LONGDOUBLE, ATTR_MATHFN_FPROUNDING_ERRNO) DEF_C99_C90RES_BUILTIN (BUILT_IN_ACOSL, "acosl", BT_FN_LONGDOUBLE_LONGDOUBLE, ATTR_MATHFN_FPROUNDING_ERRNO) +DEF_C11_BUILTIN (BUILT_IN_ALIGNED_ALLOC, "aligned_alloc", BT_FN_PTR_SIZE_SIZE, ATTR_MALLOC_NOTHROW_LIST) DEF_LIB_BUILTIN (BUILT_IN_ASIN, "asin", BT_FN_DOUBLE_DOUBLE, ATTR_MATHFN_FPROUNDING_ERRNO) DEF_C99_C90RES_BUILTIN (BUILT_IN_ASINF, "asinf", BT_FN_FLOAT_FLOAT, ATTR_MATHFN_FPROUNDING_ERRNO) DEF_C99_BUILTIN (BUILT_IN_ASINH, "asinh", BT_FN_DOUBLE_DOUBLE, ATTR_MATHFN_FPROUNDING) diff --git a/gcc/c-family/ChangeLog b/gcc/c-family/ChangeLog index fb0d102eac3..47bb11438ad 100644 --- a/gcc/c-family/ChangeLog +++ b/gcc/c-family/ChangeLog @@ -1,3 +1,8 @@ +2014-04-30 Marek Polacek <polacek@redhat.com> + + * c-ubsan.c (ubsan_instrument_division): Handle REAL_TYPEs. Perform + INT_MIN / -1 sanitization only for integer types. + 2014-04-25 Marek Polacek <polacek@redhat.com> PR c/18079 diff --git a/gcc/c-family/c-ubsan.c b/gcc/c-family/c-ubsan.c index e4f6f327277..a0397925fe7 100644 --- a/gcc/c-family/c-ubsan.c +++ b/gcc/c-family/c-ubsan.c @@ -46,15 +46,21 @@ ubsan_instrument_division (location_t loc, tree op0, tree op1) gcc_assert (TYPE_MAIN_VARIANT (TREE_TYPE (op0)) == TYPE_MAIN_VARIANT (TREE_TYPE (op1))); - /* TODO: REAL_TYPE is not supported yet. */ - if (TREE_CODE (type) != INTEGER_TYPE) + if (TREE_CODE (type) == INTEGER_TYPE + && (flag_sanitize & SANITIZE_DIVIDE)) + t = fold_build2 (EQ_EXPR, boolean_type_node, + op1, build_int_cst (type, 0)); + else if (TREE_CODE (type) == REAL_TYPE + && (flag_sanitize & SANITIZE_FLOAT_DIVIDE)) + t = fold_build2 (EQ_EXPR, boolean_type_node, + op1, build_real (type, dconst0)); + else return NULL_TREE; - t = fold_build2 (EQ_EXPR, boolean_type_node, - op1, build_int_cst (type, 0)); - /* We check INT_MIN / -1 only for signed types. */ - if (!TYPE_UNSIGNED (type)) + if (TREE_CODE (type) == INTEGER_TYPE + && (flag_sanitize & SANITIZE_DIVIDE) + && !TYPE_UNSIGNED (type)) { tree x; tt = fold_build2 (EQ_EXPR, boolean_type_node, op1, diff --git a/gcc/c/ChangeLog b/gcc/c/ChangeLog index 80841af40ee..bf61610e37c 100644 --- a/gcc/c/ChangeLog +++ b/gcc/c/ChangeLog @@ -1,3 +1,20 @@ +2014-04-30 Marek Polacek <polacek@redhat.com> + + * c-typeck.c (build_binary_op): Call ubsan_instrument_division + also when SANITIZE_FLOAT_DIVIDE is on. + +2014-04-30 Marek Polacek <polacek@redhat.com> + + PR c/60139 + * c-typeck.c (output_init_element): Pass OPT_Wpedantic to pedwarn + and pedwarn_init. Use loc insted of input_location. + +2014-04-30 Marek Polacek <polacek@redhat.com> + + PR c/60351 + * c-typeck.c (build_binary_op): Use location when warning about + shift count. + 2014-04-25 Marek Polacek <polacek@redhat.com> PR c/18079 diff --git a/gcc/c/c-typeck.c b/gcc/c/c-typeck.c index e4293104d7b..11825a9218e 100644 --- a/gcc/c/c-typeck.c +++ b/gcc/c/c-typeck.c @@ -8256,12 +8256,12 @@ output_init_element (location_t loc, tree value, tree origtype, value = error_mark_node; } else if (require_constant_elements) - pedwarn (input_location, 0, + pedwarn (loc, OPT_Wpedantic, "initializer element is not computable at load time"); } else if (!maybe_const && (require_constant_value || require_constant_elements)) - pedwarn_init (input_location, 0, + pedwarn_init (loc, OPT_Wpedantic, "initializer element is not a constant expression"); /* Issue -Wc++-compat warnings about initializing a bitfield with @@ -10403,7 +10403,7 @@ build_binary_op (location_t location, enum tree_code code, { int_const = false; if (c_inhibit_evaluation_warnings == 0) - warning (0, "right shift count is negative"); + warning_at (location, 0, "right shift count is negative"); } else { @@ -10414,7 +10414,8 @@ build_binary_op (location_t location, enum tree_code code, { int_const = false; if (c_inhibit_evaluation_warnings == 0) - warning (0, "right shift count >= width of type"); + warning_at (location, 0, "right shift count >= width " + "of type"); } } } @@ -10456,14 +10457,15 @@ build_binary_op (location_t location, enum tree_code code, { int_const = false; if (c_inhibit_evaluation_warnings == 0) - warning (0, "left shift count is negative"); + warning_at (location, 0, "left shift count is negative"); } else if (compare_tree_int (op1, TYPE_PRECISION (type0)) >= 0) { int_const = false; if (c_inhibit_evaluation_warnings == 0) - warning (0, "left shift count >= width of type"); + warning_at (location, 0, "left shift count >= width of " + "type"); } } @@ -10996,7 +10998,8 @@ build_binary_op (location_t location, enum tree_code code, return error_mark_node; } - if ((flag_sanitize & (SANITIZE_SHIFT | SANITIZE_DIVIDE)) + if ((flag_sanitize & (SANITIZE_SHIFT | SANITIZE_DIVIDE + | SANITIZE_FLOAT_DIVIDE)) && current_function_decl != 0 && !lookup_attribute ("no_sanitize_undefined", DECL_ATTRIBUTES (current_function_decl)) @@ -11007,7 +11010,8 @@ build_binary_op (location_t location, enum tree_code code, op1 = c_save_expr (op1); op0 = c_fully_fold (op0, false, NULL); op1 = c_fully_fold (op1, false, NULL); - if (doing_div_or_mod && (flag_sanitize & SANITIZE_DIVIDE)) + if (doing_div_or_mod && (flag_sanitize & (SANITIZE_DIVIDE + | SANITIZE_FLOAT_DIVIDE))) instrument_expr = ubsan_instrument_division (location, op0, op1); else if (doing_shift && (flag_sanitize & SANITIZE_SHIFT)) instrument_expr = ubsan_instrument_shift (location, code, op0, op1); diff --git a/gcc/calls.c b/gcc/calls.c index e798c7a0349..78fe7d8525b 100644 --- a/gcc/calls.c +++ b/gcc/calls.c @@ -1104,8 +1104,6 @@ initialize_argument_information (int num_actuals ATTRIBUTE_UNUSED, { CUMULATIVE_ARGS *args_so_far_pnt = get_cumulative_args (args_so_far); location_t loc = EXPR_LOCATION (exp); - /* 1 if scanning parms front to back, -1 if scanning back to front. */ - int inc; /* Count arg position in order args appear. */ int argpos; @@ -1116,22 +1114,9 @@ initialize_argument_information (int num_actuals ATTRIBUTE_UNUSED, args_size->var = 0; /* In this loop, we consider args in the order they are written. - We fill up ARGS from the front or from the back if necessary - so that in any case the first arg to be pushed ends up at the front. */ + We fill up ARGS from the back. */ - if (PUSH_ARGS_REVERSED) - { - i = num_actuals - 1, inc = -1; - /* In this case, must reverse order of args - so that we compute and push the last arg first. */ - } - else - { - i = 0, inc = 1; - } - - /* First fill in the actual arguments in the ARGS array, splitting - complex arguments if necessary. */ + i = num_actuals - 1; { int j = i; call_expr_arg_iterator iter; @@ -1140,7 +1125,7 @@ initialize_argument_information (int num_actuals ATTRIBUTE_UNUSED, if (struct_value_addr_value) { args[j].tree_value = struct_value_addr_value; - j += inc; + j--; } FOR_EACH_CALL_EXPR_ARG (arg, iter, exp) { @@ -1152,17 +1137,17 @@ initialize_argument_information (int num_actuals ATTRIBUTE_UNUSED, { tree subtype = TREE_TYPE (argtype); args[j].tree_value = build1 (REALPART_EXPR, subtype, arg); - j += inc; + j--; args[j].tree_value = build1 (IMAGPART_EXPR, subtype, arg); } else args[j].tree_value = arg; - j += inc; + j--; } } /* I counts args in order (to be) pushed; ARGPOS counts in order written. */ - for (argpos = 0; argpos < num_actuals; i += inc, argpos++) + for (argpos = 0; argpos < num_actuals; i--, argpos++) { tree type = TREE_TYPE (args[i].tree_value); int unsignedp; @@ -2952,9 +2937,8 @@ expand_call (tree exp, rtx target, int ignore) compute_argument_addresses (args, argblock, num_actuals); - /* If we push args individually in reverse order, perform stack alignment - before the first push (the last arg). */ - if (PUSH_ARGS_REVERSED && argblock == 0 + /* Perform stack alignment before the first push (the last arg). */ + if (argblock == 0 && adjusted_args_size.constant > reg_parm_stack_space && adjusted_args_size.constant != unadjusted_args_size) { @@ -3097,12 +3081,6 @@ expand_call (tree exp, rtx target, int ignore) sibcall_failure = 1; } - /* If we pushed args in forward order, perform stack alignment - after pushing the last arg. */ - if (!PUSH_ARGS_REVERSED && argblock == 0) - anti_adjust_stack (GEN_INT (adjusted_args_size.constant - - unadjusted_args_size)); - /* If register arguments require space on the stack and stack space was not preallocated, allocate stack space here for arguments passed in registers. */ @@ -3152,8 +3130,7 @@ expand_call (tree exp, rtx target, int ignore) if (pass == 1 && (return_flags & ERF_RETURNS_ARG)) { int arg_nr = return_flags & ERF_RETURN_ARG_MASK; - if (PUSH_ARGS_REVERSED) - arg_nr = num_actuals - arg_nr - 1; + arg_nr = num_actuals - arg_nr - 1; if (arg_nr >= 0 && arg_nr < num_actuals && args[arg_nr].reg @@ -3610,7 +3587,6 @@ emit_library_call_value_1 (int retval, rtx orgfun, rtx value, isn't present here, so we default to native calling abi here. */ tree fndecl ATTRIBUTE_UNUSED = NULL_TREE; /* library calls default to host calling abi ? */ tree fntype ATTRIBUTE_UNUSED = NULL_TREE; /* library calls default to host calling abi ? */ - int inc; int count; rtx argblock = 0; CUMULATIVE_ARGS args_so_far_v; @@ -3959,22 +3935,13 @@ emit_library_call_value_1 (int retval, rtx orgfun, rtx value, argblock = push_block (GEN_INT (args_size.constant), 0, 0); } - /* If we push args individually in reverse order, perform stack alignment + /* We push args individually in reverse order, perform stack alignment before the first push (the last arg). */ - if (argblock == 0 && PUSH_ARGS_REVERSED) + if (argblock == 0) anti_adjust_stack (GEN_INT (args_size.constant - original_args_size.constant)); - if (PUSH_ARGS_REVERSED) - { - inc = -1; - argnum = nargs - 1; - } - else - { - inc = 1; - argnum = 0; - } + argnum = nargs - 1; #ifdef REG_PARM_STACK_SPACE if (ACCUMULATE_OUTGOING_ARGS) @@ -3991,7 +3958,7 @@ emit_library_call_value_1 (int retval, rtx orgfun, rtx value, /* ARGNUM indexes the ARGVEC array in the order in which the arguments are to be pushed. */ - for (count = 0; count < nargs; count++, argnum += inc) + for (count = 0; count < nargs; count++, argnum--) { enum machine_mode mode = argvec[argnum].mode; rtx val = argvec[argnum].value; @@ -4093,16 +4060,7 @@ emit_library_call_value_1 (int retval, rtx orgfun, rtx value, } } - /* If we pushed args in forward order, perform stack alignment - after pushing the last arg. */ - if (argblock == 0 && !PUSH_ARGS_REVERSED) - anti_adjust_stack (GEN_INT (args_size.constant - - original_args_size.constant)); - - if (PUSH_ARGS_REVERSED) - argnum = nargs - 1; - else - argnum = 0; + argnum = nargs - 1; fun = prepare_call_address (NULL, fun, NULL, &call_fusage, 0, 0); @@ -4110,7 +4068,7 @@ emit_library_call_value_1 (int retval, rtx orgfun, rtx value, /* ARGNUM indexes the ARGVEC array in the order in which the arguments are to be pushed. */ - for (count = 0; count < nargs; count++, argnum += inc) + for (count = 0; count < nargs; count++, argnum--) { enum machine_mode mode = argvec[argnum].mode; rtx val = argvec[argnum].value; diff --git a/gcc/cgraph.c b/gcc/cgraph.c index b5df572d039..2b4ce813c90 100644 --- a/gcc/cgraph.c +++ b/gcc/cgraph.c @@ -1695,8 +1695,8 @@ release_function_body (tree decl) } if (cfun->cfg) { - free_dominance_info (CDI_DOMINATORS); - free_dominance_info (CDI_POST_DOMINATORS); + gcc_assert (!dom_info_available_p (CDI_DOMINATORS)); + gcc_assert (!dom_info_available_p (CDI_POST_DOMINATORS)); clear_edges (); cfun->cfg = NULL; } diff --git a/gcc/config.gcc b/gcc/config.gcc index 92f8a94b3e6..81205ff0c8f 100644 --- a/gcc/config.gcc +++ b/gcc/config.gcc @@ -808,8 +808,9 @@ case ${target} in *-*-solaris2*) # i?86-*-solaris2* needs to insert headers between cpu default and # Solaris 2 specific ones. - sol2_tm_file="dbxelf.h elfos.h ${cpu_type}/sysv4.h sol2.h ${cpu_type}/sol2.h" - sol2_tm_file="${sol2_tm_file} sol2-10.h" + sol2_tm_file_head="dbxelf.h elfos.h ${cpu_type}/sysv4.h" + sol2_tm_file_tail="${cpu_type}/sol2.h sol2.h" + sol2_tm_file="${sol2_tm_file_head} ${sol2_tm_file_tail}" use_gcc_stdint=wrap if test x$gnu_ld = xyes; then tm_file="usegld.h ${tm_file}" @@ -1515,14 +1516,13 @@ i[34567]86-*-rtems*) tmake_file="${tmake_file} i386/t-rtems" ;; i[34567]86-*-solaris2* | x86_64-*-solaris2.1[0-9]*) - tm_file="${tm_file} i386/unix.h i386/att.h ${sol2_tm_file}" # Set default arch_32 to pentium4, tune_32 to generic like the other # i386 targets, although config.guess defaults to i386-pc-solaris2*. with_arch_32=${with_arch_32:-pentium4} with_tune_32=${with_tune_32:-generic} - tm_file="${tm_file} i386/x86-64.h i386/sol2-bi.h sol2-bi.h" + tm_file="${tm_file} i386/unix.h i386/att.h ${sol2_tm_file_head} i386/x86-64.h ${sol2_tm_file_tail}" tm_defines="${tm_defines} TARGET_BI_ARCH=1" - tmake_file="$tmake_file i386/t-sol2-64" + tmake_file="$tmake_file i386/t-sol2" need_64bit_isa=yes if test x$with_cpu = x; then if test x$with_cpu_64 = x; then @@ -2661,7 +2661,7 @@ sparc-*-netbsdelf*) tmake_file="${tmake_file} sparc/t-sparc" ;; sparc*-*-solaris2*) - tm_file="sparc/biarch64.h ${tm_file} ${sol2_tm_file} sol2-bi.h sparc/tso.h" + tm_file="sparc/biarch64.h ${tm_file} ${sol2_tm_file} sparc/tso.h" case ${target} in sparc64-*-* | sparcv9-*-*) tm_file="sparc/default-64.h ${tm_file}" @@ -2670,7 +2670,7 @@ sparc*-*-solaris2*) test x$with_cpu != x || with_cpu=v9 ;; esac - tmake_file="${tmake_file} sparc/t-sparc sparc/t-sol2-64" + tmake_file="${tmake_file} sparc/t-sparc sparc/t-sol2" ;; sparc-wrs-vxworks) tm_file="${tm_file} elfos.h sparc/sysv4.h vx-common.h vxworks.h sparc/vxworks.h" diff --git a/gcc/config/aarch64/aarch64-builtins.c b/gcc/config/aarch64/aarch64-builtins.c index 4616ad24c07..a3019828a93 100644 --- a/gcc/config/aarch64/aarch64-builtins.c +++ b/gcc/config/aarch64/aarch64-builtins.c @@ -246,6 +246,11 @@ aarch64_types_store1_qualifiers[SIMD_MAX_BUILTIN_ARGS] = { qualifier_void, qualifier_pointer_map_mode, qualifier_none }; #define TYPES_STORE1 (aarch64_types_store1_qualifiers) #define TYPES_STORESTRUCT (aarch64_types_store1_qualifiers) +static enum aarch64_type_qualifiers +aarch64_types_storestruct_lane_qualifiers[SIMD_MAX_BUILTIN_ARGS] + = { qualifier_void, qualifier_pointer_map_mode, + qualifier_none, qualifier_none }; +#define TYPES_STORESTRUCT_LANE (aarch64_types_storestruct_lane_qualifiers) #define CF0(N, X) CODE_FOR_aarch64_##N##X #define CF1(N, X) CODE_FOR_##N##X##1 diff --git a/gcc/config/aarch64/aarch64-protos.h b/gcc/config/aarch64/aarch64-protos.h index 5542f023b33..04cbc780da2 100644 --- a/gcc/config/aarch64/aarch64-protos.h +++ b/gcc/config/aarch64/aarch64-protos.h @@ -175,6 +175,8 @@ bool aarch64_is_extend_from_extract (enum machine_mode, rtx, rtx); bool aarch64_is_long_call_p (rtx); bool aarch64_label_mentioned_p (rtx); bool aarch64_legitimate_pic_operand_p (rtx); +bool aarch64_modes_tieable_p (enum machine_mode mode1, + enum machine_mode mode2); bool aarch64_move_imm (HOST_WIDE_INT, enum machine_mode); bool aarch64_mov_operand_p (rtx, enum aarch64_symbol_context, enum machine_mode); diff --git a/gcc/config/aarch64/aarch64-simd-builtins.def b/gcc/config/aarch64/aarch64-simd-builtins.def index fa332ae5948..339e8f86a4b 100644 --- a/gcc/config/aarch64/aarch64-simd-builtins.def +++ b/gcc/config/aarch64/aarch64-simd-builtins.def @@ -118,6 +118,10 @@ BUILTIN_VQ (STORESTRUCT, st3, 0) BUILTIN_VQ (STORESTRUCT, st4, 0) + BUILTIN_VQ (STORESTRUCT_LANE, st2_lane, 0) + BUILTIN_VQ (STORESTRUCT_LANE, st3_lane, 0) + BUILTIN_VQ (STORESTRUCT_LANE, st4_lane, 0) + BUILTIN_VQW (BINOP, saddl2, 0) BUILTIN_VQW (BINOP, uaddl2, 0) BUILTIN_VQW (BINOP, ssubl2, 0) diff --git a/gcc/config/aarch64/aarch64-simd.md b/gcc/config/aarch64/aarch64-simd.md index c05767b2045..108bc8d8893 100644 --- a/gcc/config/aarch64/aarch64-simd.md +++ b/gcc/config/aarch64/aarch64-simd.md @@ -3686,6 +3686,17 @@ [(set_attr "type" "neon_store2_2reg<q>")] ) +(define_insn "vec_store_lanesoi_lane<mode>" + [(set (match_operand:<V_TWO_ELEM> 0 "aarch64_simd_struct_operand" "=Utv") + (unspec:<V_TWO_ELEM> [(match_operand:OI 1 "register_operand" "w") + (unspec:VQ [(const_int 0)] UNSPEC_VSTRUCTDUMMY) + (match_operand:SI 2 "immediate_operand" "i")] + UNSPEC_ST2_LANE))] + "TARGET_SIMD" + "st2\\t{%S1.<Vetype> - %T1.<Vetype>}[%2], %0" + [(set_attr "type" "neon_store3_one_lane<q>")] +) + (define_insn "vec_load_lanesci<mode>" [(set (match_operand:CI 0 "register_operand" "=w") (unspec:CI [(match_operand:CI 1 "aarch64_simd_struct_operand" "Utv") @@ -3706,6 +3717,17 @@ [(set_attr "type" "neon_store3_3reg<q>")] ) +(define_insn "vec_store_lanesci_lane<mode>" + [(set (match_operand:<V_THREE_ELEM> 0 "aarch64_simd_struct_operand" "=Utv") + (unspec:<V_THREE_ELEM> [(match_operand:CI 1 "register_operand" "w") + (unspec:VQ [(const_int 0)] UNSPEC_VSTRUCTDUMMY) + (match_operand:SI 2 "immediate_operand" "i")] + UNSPEC_ST3_LANE))] + "TARGET_SIMD" + "st3\\t{%S1.<Vetype> - %U1.<Vetype>}[%2], %0" + [(set_attr "type" "neon_store3_one_lane<q>")] +) + (define_insn "vec_load_lanesxi<mode>" [(set (match_operand:XI 0 "register_operand" "=w") (unspec:XI [(match_operand:XI 1 "aarch64_simd_struct_operand" "Utv") @@ -3726,6 +3748,17 @@ [(set_attr "type" "neon_store4_4reg<q>")] ) +(define_insn "vec_store_lanesxi_lane<mode>" + [(set (match_operand:<V_FOUR_ELEM> 0 "aarch64_simd_struct_operand" "=Utv") + (unspec:<V_FOUR_ELEM> [(match_operand:XI 1 "register_operand" "w") + (unspec:VQ [(const_int 0)] UNSPEC_VSTRUCTDUMMY) + (match_operand:SI 2 "immediate_operand" "i")] + UNSPEC_ST4_LANE))] + "TARGET_SIMD" + "st4\\t{%S1.<Vetype> - %V1.<Vetype>}[%2], %0" + [(set_attr "type" "neon_store4_one_lane<q>")] +) + ;; Reload patterns for AdvSIMD register list operands. (define_expand "mov<mode>" @@ -4220,6 +4253,57 @@ DONE; }) +(define_expand "aarch64_st2_lane<VQ:mode>" + [(match_operand:DI 0 "register_operand" "r") + (match_operand:OI 1 "register_operand" "w") + (unspec:VQ [(const_int 0)] UNSPEC_VSTRUCTDUMMY) + (match_operand:SI 2 "immediate_operand")] + "TARGET_SIMD" +{ + enum machine_mode mode = <V_TWO_ELEM>mode; + rtx mem = gen_rtx_MEM (mode, operands[0]); + operands[2] = GEN_INT (ENDIAN_LANE_N (<MODE>mode, INTVAL (operands[2]))); + + emit_insn (gen_vec_store_lanesoi_lane<VQ:mode> (mem, + operands[1], + operands[2])); + DONE; +}) + +(define_expand "aarch64_st3_lane<VQ:mode>" + [(match_operand:DI 0 "register_operand" "r") + (match_operand:CI 1 "register_operand" "w") + (unspec:VQ [(const_int 0)] UNSPEC_VSTRUCTDUMMY) + (match_operand:SI 2 "immediate_operand")] + "TARGET_SIMD" +{ + enum machine_mode mode = <V_THREE_ELEM>mode; + rtx mem = gen_rtx_MEM (mode, operands[0]); + operands[2] = GEN_INT (ENDIAN_LANE_N (<MODE>mode, INTVAL (operands[2]))); + + emit_insn (gen_vec_store_lanesci_lane<VQ:mode> (mem, + operands[1], + operands[2])); + DONE; +}) + +(define_expand "aarch64_st4_lane<VQ:mode>" + [(match_operand:DI 0 "register_operand" "r") + (match_operand:XI 1 "register_operand" "w") + (unspec:VQ [(const_int 0)] UNSPEC_VSTRUCTDUMMY) + (match_operand:SI 2 "immediate_operand")] + "TARGET_SIMD" +{ + enum machine_mode mode = <V_FOUR_ELEM>mode; + rtx mem = gen_rtx_MEM (mode, operands[0]); + operands[2] = GEN_INT (ENDIAN_LANE_N (<MODE>mode, INTVAL (operands[2]))); + + emit_insn (gen_vec_store_lanesxi_lane<VQ:mode> (mem, + operands[1], + operands[2])); + DONE; +}) + (define_expand "aarch64_st1<VALL:mode>" [(match_operand:DI 0 "register_operand") (match_operand:VALL 1 "register_operand")] diff --git a/gcc/config/aarch64/aarch64.c b/gcc/config/aarch64/aarch64.c index 1d48108516d..d3d7d1e60d6 100644 --- a/gcc/config/aarch64/aarch64.c +++ b/gcc/config/aarch64/aarch64.c @@ -8316,7 +8316,8 @@ aarch64_cannot_change_mode_class (enum machine_mode from, /* Limited combinations of subregs are safe on FPREGs. Particularly, 1. Vector Mode to Scalar mode where 1 unit of the vector is accessed. 2. Scalar to Scalar for integer modes or same size float modes. - 3. Vector to Vector modes. */ + 3. Vector to Vector modes. + 4. On little-endian only, Vector-Structure to Vector modes. */ if (GET_MODE_SIZE (from) > GET_MODE_SIZE (to)) { if (aarch64_vector_mode_supported_p (from) @@ -8332,11 +8333,41 @@ aarch64_cannot_change_mode_class (enum machine_mode from, if (aarch64_vector_mode_supported_p (from) && aarch64_vector_mode_supported_p (to)) return false; + + /* Within an vector structure straddling multiple vector registers + we are in a mixed-endian representation. As such, we can't + easily change modes for BYTES_BIG_ENDIAN. Otherwise, we can + switch between vectors and vector structures cheaply. */ + if (!BYTES_BIG_ENDIAN) + if ((aarch64_vector_mode_supported_p (from) + && aarch64_vect_struct_mode_p (to)) + || (aarch64_vector_mode_supported_p (to) + && aarch64_vect_struct_mode_p (from))) + return false; } return true; } +/* Implement MODES_TIEABLE_P. */ + +bool +aarch64_modes_tieable_p (enum machine_mode mode1, enum machine_mode mode2) +{ + if (GET_MODE_CLASS (mode1) == GET_MODE_CLASS (mode2)) + return true; + + /* We specifically want to allow elements of "structure" modes to + be tieable to the structure. This more general condition allows + other rarer situations too. */ + if (TARGET_SIMD + && aarch64_vector_mode_p (mode1) + && aarch64_vector_mode_p (mode2)) + return true; + + return false; +} + #undef TARGET_ADDRESS_COST #define TARGET_ADDRESS_COST aarch64_address_cost diff --git a/gcc/config/aarch64/aarch64.h b/gcc/config/aarch64/aarch64.h index e2b6c8e2908..c9b30d01865 100644 --- a/gcc/config/aarch64/aarch64.h +++ b/gcc/config/aarch64/aarch64.h @@ -365,8 +365,7 @@ extern unsigned long aarch64_tune_flags; #define HARD_REGNO_MODE_OK(REGNO, MODE) aarch64_hard_regno_mode_ok (REGNO, MODE) -#define MODES_TIEABLE_P(MODE1, MODE2) \ - (GET_MODE_CLASS (MODE1) == GET_MODE_CLASS (MODE2)) +#define MODES_TIEABLE_P(MODE1, MODE2) aarch64_modes_tieable_p (MODE1, MODE2) #define DWARF2_UNWIND_INFO 1 diff --git a/gcc/config/aarch64/aarch64.md b/gcc/config/aarch64/aarch64.md index 7965db4c9c7..266d7873a5a 100644 --- a/gcc/config/aarch64/aarch64.md +++ b/gcc/config/aarch64/aarch64.md @@ -98,6 +98,9 @@ UNSPEC_ST2 UNSPEC_ST3 UNSPEC_ST4 + UNSPEC_ST2_LANE + UNSPEC_ST3_LANE + UNSPEC_ST4_LANE UNSPEC_TLS UNSPEC_TLSDESC UNSPEC_USHL_2S @@ -2426,6 +2429,25 @@ } ) +(define_expand "mov<mode>cc" + [(set (match_operand:GPF 0 "register_operand" "") + (if_then_else:GPF (match_operand 1 "aarch64_comparison_operator" "") + (match_operand:GPF 2 "register_operand" "") + (match_operand:GPF 3 "register_operand" "")))] + "" + { + rtx ccreg; + enum rtx_code code = GET_CODE (operands[1]); + + if (code == UNEQ || code == LTGT) + FAIL; + + ccreg = aarch64_gen_compare_reg (code, XEXP (operands[1], 0), + XEXP (operands[1], 1)); + operands[1] = gen_rtx_fmt_ee (code, VOIDmode, ccreg, const0_rtx); + } +) + (define_insn "*csinc2<mode>_insn" [(set (match_operand:GPI 0 "register_operand" "=r") (plus:GPI (match_operator:GPI 2 "aarch64_comparison_operator" diff --git a/gcc/config/aarch64/arm_neon.h b/gcc/config/aarch64/arm_neon.h index 9f1fa98e6fb..f6213ce2aea 100644 --- a/gcc/config/aarch64/arm_neon.h +++ b/gcc/config/aarch64/arm_neon.h @@ -13199,929 +13199,6 @@ vtstq_p16 (poly16x8_t a, poly16x8_t b) : /* No clobbers */); return result; } -__extension__ static __inline float32x2_t __attribute__ ((__always_inline__)) -vuzp1_f32 (float32x2_t a, float32x2_t b) -{ - float32x2_t result; - __asm__ ("uzp1 %0.2s,%1.2s,%2.2s" - : "=w"(result) - : "w"(a), "w"(b) - : /* No clobbers */); - return result; -} - -__extension__ static __inline poly8x8_t __attribute__ ((__always_inline__)) -vuzp1_p8 (poly8x8_t a, poly8x8_t b) -{ - poly8x8_t result; - __asm__ ("uzp1 %0.8b,%1.8b,%2.8b" - : "=w"(result) - : "w"(a), "w"(b) - : /* No clobbers */); - return result; -} - -__extension__ static __inline poly16x4_t __attribute__ ((__always_inline__)) -vuzp1_p16 (poly16x4_t a, poly16x4_t b) -{ - poly16x4_t result; - __asm__ ("uzp1 %0.4h,%1.4h,%2.4h" - : "=w"(result) - : "w"(a), "w"(b) - : /* No clobbers */); - return result; -} - -__extension__ static __inline int8x8_t __attribute__ ((__always_inline__)) -vuzp1_s8 (int8x8_t a, int8x8_t b) -{ - int8x8_t result; - __asm__ ("uzp1 %0.8b,%1.8b,%2.8b" - : "=w"(result) - : "w"(a), "w"(b) - : /* No clobbers */); - return result; -} - -__extension__ static __inline int16x4_t __attribute__ ((__always_inline__)) -vuzp1_s16 (int16x4_t a, int16x4_t b) -{ - int16x4_t result; - __asm__ ("uzp1 %0.4h,%1.4h,%2.4h" - : "=w"(result) - : "w"(a), "w"(b) - : /* No clobbers */); - return result; -} - -__extension__ static __inline int32x2_t __attribute__ ((__always_inline__)) -vuzp1_s32 (int32x2_t a, int32x2_t b) -{ - int32x2_t result; - __asm__ ("uzp1 %0.2s,%1.2s,%2.2s" - : "=w"(result) - : "w"(a), "w"(b) - : /* No clobbers */); - return result; -} - -__extension__ static __inline uint8x8_t __attribute__ ((__always_inline__)) -vuzp1_u8 (uint8x8_t a, uint8x8_t b) -{ - uint8x8_t result; - __asm__ ("uzp1 %0.8b,%1.8b,%2.8b" - : "=w"(result) - : "w"(a), "w"(b) - : /* No clobbers */); - return result; -} - -__extension__ static __inline uint16x4_t __attribute__ ((__always_inline__)) -vuzp1_u16 (uint16x4_t a, uint16x4_t b) -{ - uint16x4_t result; - __asm__ ("uzp1 %0.4h,%1.4h,%2.4h" - : "=w"(result) - : "w"(a), "w"(b) - : /* No clobbers */); - return result; -} - -__extension__ static __inline uint32x2_t __attribute__ ((__always_inline__)) -vuzp1_u32 (uint32x2_t a, uint32x2_t b) -{ - uint32x2_t result; - __asm__ ("uzp1 %0.2s,%1.2s,%2.2s" - : "=w"(result) - : "w"(a), "w"(b) - : /* No clobbers */); - return result; -} - -__extension__ static __inline float32x4_t __attribute__ ((__always_inline__)) -vuzp1q_f32 (float32x4_t a, float32x4_t b) -{ - float32x4_t result; - __asm__ ("uzp1 %0.4s,%1.4s,%2.4s" - : "=w"(result) - : "w"(a), "w"(b) - : /* No clobbers */); - return result; -} - -__extension__ static __inline float64x2_t __attribute__ ((__always_inline__)) -vuzp1q_f64 (float64x2_t a, float64x2_t b) -{ - float64x2_t result; - __asm__ ("uzp1 %0.2d,%1.2d,%2.2d" - : "=w"(result) - : "w"(a), "w"(b) - : /* No clobbers */); - return result; -} - -__extension__ static __inline poly8x16_t __attribute__ ((__always_inline__)) -vuzp1q_p8 (poly8x16_t a, poly8x16_t b) -{ - poly8x16_t result; - __asm__ ("uzp1 %0.16b,%1.16b,%2.16b" - : "=w"(result) - : "w"(a), "w"(b) - : /* No clobbers */); - return result; -} - -__extension__ static __inline poly16x8_t __attribute__ ((__always_inline__)) -vuzp1q_p16 (poly16x8_t a, poly16x8_t b) -{ - poly16x8_t result; - __asm__ ("uzp1 %0.8h,%1.8h,%2.8h" - : "=w"(result) - : "w"(a), "w"(b) - : /* No clobbers */); - return result; -} - -__extension__ static __inline int8x16_t __attribute__ ((__always_inline__)) -vuzp1q_s8 (int8x16_t a, int8x16_t b) -{ - int8x16_t result; - __asm__ ("uzp1 %0.16b,%1.16b,%2.16b" - : "=w"(result) - : "w"(a), "w"(b) - : /* No clobbers */); - return result; -} - -__extension__ static __inline int16x8_t __attribute__ ((__always_inline__)) -vuzp1q_s16 (int16x8_t a, int16x8_t b) -{ - int16x8_t result; - __asm__ ("uzp1 %0.8h,%1.8h,%2.8h" - : "=w"(result) - : "w"(a), "w"(b) - : /* No clobbers */); - return result; -} - -__extension__ static __inline int32x4_t __attribute__ ((__always_inline__)) -vuzp1q_s32 (int32x4_t a, int32x4_t b) -{ - int32x4_t result; - __asm__ ("uzp1 %0.4s,%1.4s,%2.4s" - : "=w"(result) - : "w"(a), "w"(b) - : /* No clobbers */); - return result; -} - -__extension__ static __inline int64x2_t __attribute__ ((__always_inline__)) -vuzp1q_s64 (int64x2_t a, int64x2_t b) -{ - int64x2_t result; - __asm__ ("uzp1 %0.2d,%1.2d,%2.2d" - : "=w"(result) - : "w"(a), "w"(b) - : /* No clobbers */); - return result; -} - -__extension__ static __inline uint8x16_t __attribute__ ((__always_inline__)) -vuzp1q_u8 (uint8x16_t a, uint8x16_t b) -{ - uint8x16_t result; - __asm__ ("uzp1 %0.16b,%1.16b,%2.16b" - : "=w"(result) - : "w"(a), "w"(b) - : /* No clobbers */); - return result; -} - -__extension__ static __inline uint16x8_t __attribute__ ((__always_inline__)) -vuzp1q_u16 (uint16x8_t a, uint16x8_t b) -{ - uint16x8_t result; - __asm__ ("uzp1 %0.8h,%1.8h,%2.8h" - : "=w"(result) - : "w"(a), "w"(b) - : /* No clobbers */); - return result; -} - -__extension__ static __inline uint32x4_t __attribute__ ((__always_inline__)) -vuzp1q_u32 (uint32x4_t a, uint32x4_t b) -{ - uint32x4_t result; - __asm__ ("uzp1 %0.4s,%1.4s,%2.4s" - : "=w"(result) - : "w"(a), "w"(b) - : /* No clobbers */); - return result; -} - -__extension__ static __inline uint64x2_t __attribute__ ((__always_inline__)) -vuzp1q_u64 (uint64x2_t a, uint64x2_t b) -{ - uint64x2_t result; - __asm__ ("uzp1 %0.2d,%1.2d,%2.2d" - : "=w"(result) - : "w"(a), "w"(b) - : /* No clobbers */); - return result; -} - -__extension__ static __inline float32x2_t __attribute__ ((__always_inline__)) -vuzp2_f32 (float32x2_t a, float32x2_t b) -{ - float32x2_t result; - __asm__ ("uzp2 %0.2s,%1.2s,%2.2s" - : "=w"(result) - : "w"(a), "w"(b) - : /* No clobbers */); - return result; -} - -__extension__ static __inline poly8x8_t __attribute__ ((__always_inline__)) -vuzp2_p8 (poly8x8_t a, poly8x8_t b) -{ - poly8x8_t result; - __asm__ ("uzp2 %0.8b,%1.8b,%2.8b" - : "=w"(result) - : "w"(a), "w"(b) - : /* No clobbers */); - return result; -} - -__extension__ static __inline poly16x4_t __attribute__ ((__always_inline__)) -vuzp2_p16 (poly16x4_t a, poly16x4_t b) -{ - poly16x4_t result; - __asm__ ("uzp2 %0.4h,%1.4h,%2.4h" - : "=w"(result) - : "w"(a), "w"(b) - : /* No clobbers */); - return result; -} - -__extension__ static __inline int8x8_t __attribute__ ((__always_inline__)) -vuzp2_s8 (int8x8_t a, int8x8_t b) -{ - int8x8_t result; - __asm__ ("uzp2 %0.8b,%1.8b,%2.8b" - : "=w"(result) - : "w"(a), "w"(b) - : /* No clobbers */); - return result; -} - -__extension__ static __inline int16x4_t __attribute__ ((__always_inline__)) -vuzp2_s16 (int16x4_t a, int16x4_t b) -{ - int16x4_t result; - __asm__ ("uzp2 %0.4h,%1.4h,%2.4h" - : "=w"(result) - : "w"(a), "w"(b) - : /* No clobbers */); - return result; -} - -__extension__ static __inline int32x2_t __attribute__ ((__always_inline__)) -vuzp2_s32 (int32x2_t a, int32x2_t b) -{ - int32x2_t result; - __asm__ ("uzp2 %0.2s,%1.2s,%2.2s" - : "=w"(result) - : "w"(a), "w"(b) - : /* No clobbers */); - return result; -} - -__extension__ static __inline uint8x8_t __attribute__ ((__always_inline__)) -vuzp2_u8 (uint8x8_t a, uint8x8_t b) -{ - uint8x8_t result; - __asm__ ("uzp2 %0.8b,%1.8b,%2.8b" - : "=w"(result) - : "w"(a), "w"(b) - : /* No clobbers */); - return result; -} - -__extension__ static __inline uint16x4_t __attribute__ ((__always_inline__)) -vuzp2_u16 (uint16x4_t a, uint16x4_t b) -{ - uint16x4_t result; - __asm__ ("uzp2 %0.4h,%1.4h,%2.4h" - : "=w"(result) - : "w"(a), "w"(b) - : /* No clobbers */); - return result; -} - -__extension__ static __inline uint32x2_t __attribute__ ((__always_inline__)) -vuzp2_u32 (uint32x2_t a, uint32x2_t b) -{ - uint32x2_t result; - __asm__ ("uzp2 %0.2s,%1.2s,%2.2s" - : "=w"(result) - : "w"(a), "w"(b) - : /* No clobbers */); - return result; -} - -__extension__ static __inline float32x4_t __attribute__ ((__always_inline__)) -vuzp2q_f32 (float32x4_t a, float32x4_t b) -{ - float32x4_t result; - __asm__ ("uzp2 %0.4s,%1.4s,%2.4s" - : "=w"(result) - : "w"(a), "w"(b) - : /* No clobbers */); - return result; -} - -__extension__ static __inline float64x2_t __attribute__ ((__always_inline__)) -vuzp2q_f64 (float64x2_t a, float64x2_t b) -{ - float64x2_t result; - __asm__ ("uzp2 %0.2d,%1.2d,%2.2d" - : "=w"(result) - : "w"(a), "w"(b) - : /* No clobbers */); - return result; -} - -__extension__ static __inline poly8x16_t __attribute__ ((__always_inline__)) -vuzp2q_p8 (poly8x16_t a, poly8x16_t b) -{ - poly8x16_t result; - __asm__ ("uzp2 %0.16b,%1.16b,%2.16b" - : "=w"(result) - : "w"(a), "w"(b) - : /* No clobbers */); - return result; -} - -__extension__ static __inline poly16x8_t __attribute__ ((__always_inline__)) -vuzp2q_p16 (poly16x8_t a, poly16x8_t b) -{ - poly16x8_t result; - __asm__ ("uzp2 %0.8h,%1.8h,%2.8h" - : "=w"(result) - : "w"(a), "w"(b) - : /* No clobbers */); - return result; -} - -__extension__ static __inline int8x16_t __attribute__ ((__always_inline__)) -vuzp2q_s8 (int8x16_t a, int8x16_t b) -{ - int8x16_t result; - __asm__ ("uzp2 %0.16b,%1.16b,%2.16b" - : "=w"(result) - : "w"(a), "w"(b) - : /* No clobbers */); - return result; -} - -__extension__ static __inline int16x8_t __attribute__ ((__always_inline__)) -vuzp2q_s16 (int16x8_t a, int16x8_t b) -{ - int16x8_t result; - __asm__ ("uzp2 %0.8h,%1.8h,%2.8h" - : "=w"(result) - : "w"(a), "w"(b) - : /* No clobbers */); - return result; -} - -__extension__ static __inline int32x4_t __attribute__ ((__always_inline__)) -vuzp2q_s32 (int32x4_t a, int32x4_t b) -{ - int32x4_t result; - __asm__ ("uzp2 %0.4s,%1.4s,%2.4s" - : "=w"(result) - : "w"(a), "w"(b) - : /* No clobbers */); - return result; -} - -__extension__ static __inline int64x2_t __attribute__ ((__always_inline__)) -vuzp2q_s64 (int64x2_t a, int64x2_t b) -{ - int64x2_t result; - __asm__ ("uzp2 %0.2d,%1.2d,%2.2d" - : "=w"(result) - : "w"(a), "w"(b) - : /* No clobbers */); - return result; -} - -__extension__ static __inline uint8x16_t __attribute__ ((__always_inline__)) -vuzp2q_u8 (uint8x16_t a, uint8x16_t b) -{ - uint8x16_t result; - __asm__ ("uzp2 %0.16b,%1.16b,%2.16b" - : "=w"(result) - : "w"(a), "w"(b) - : /* No clobbers */); - return result; -} - -__extension__ static __inline uint16x8_t __attribute__ ((__always_inline__)) -vuzp2q_u16 (uint16x8_t a, uint16x8_t b) -{ - uint16x8_t result; - __asm__ ("uzp2 %0.8h,%1.8h,%2.8h" - : "=w"(result) - : "w"(a), "w"(b) - : /* No clobbers */); - return result; -} - -__extension__ static __inline uint32x4_t __attribute__ ((__always_inline__)) -vuzp2q_u32 (uint32x4_t a, uint32x4_t b) -{ - uint32x4_t result; - __asm__ ("uzp2 %0.4s,%1.4s,%2.4s" - : "=w"(result) - : "w"(a), "w"(b) - : /* No clobbers */); - return result; -} - -__extension__ static __inline uint64x2_t __attribute__ ((__always_inline__)) -vuzp2q_u64 (uint64x2_t a, uint64x2_t b) -{ - uint64x2_t result; - __asm__ ("uzp2 %0.2d,%1.2d,%2.2d" - : "=w"(result) - : "w"(a), "w"(b) - : /* No clobbers */); - return result; -} - -__extension__ static __inline float32x2_t __attribute__ ((__always_inline__)) -vzip1_f32 (float32x2_t a, float32x2_t b) -{ - float32x2_t result; - __asm__ ("zip1 %0.2s,%1.2s,%2.2s" - : "=w"(result) - : "w"(a), "w"(b) - : /* No clobbers */); - return result; -} - -__extension__ static __inline poly8x8_t __attribute__ ((__always_inline__)) -vzip1_p8 (poly8x8_t a, poly8x8_t b) -{ - poly8x8_t result; - __asm__ ("zip1 %0.8b,%1.8b,%2.8b" - : "=w"(result) - : "w"(a), "w"(b) - : /* No clobbers */); - return result; -} - -__extension__ static __inline poly16x4_t __attribute__ ((__always_inline__)) -vzip1_p16 (poly16x4_t a, poly16x4_t b) -{ - poly16x4_t result; - __asm__ ("zip1 %0.4h,%1.4h,%2.4h" - : "=w"(result) - : "w"(a), "w"(b) - : /* No clobbers */); - return result; -} - -__extension__ static __inline int8x8_t __attribute__ ((__always_inline__)) -vzip1_s8 (int8x8_t a, int8x8_t b) -{ - int8x8_t result; - __asm__ ("zip1 %0.8b,%1.8b,%2.8b" - : "=w"(result) - : "w"(a), "w"(b) - : /* No clobbers */); - return result; -} - -__extension__ static __inline int16x4_t __attribute__ ((__always_inline__)) -vzip1_s16 (int16x4_t a, int16x4_t b) -{ - int16x4_t result; - __asm__ ("zip1 %0.4h,%1.4h,%2.4h" - : "=w"(result) - : "w"(a), "w"(b) - : /* No clobbers */); - return result; -} - -__extension__ static __inline int32x2_t __attribute__ ((__always_inline__)) -vzip1_s32 (int32x2_t a, int32x2_t b) -{ - int32x2_t result; - __asm__ ("zip1 %0.2s,%1.2s,%2.2s" - : "=w"(result) - : "w"(a), "w"(b) - : /* No clobbers */); - return result; -} - -__extension__ static __inline uint8x8_t __attribute__ ((__always_inline__)) -vzip1_u8 (uint8x8_t a, uint8x8_t b) -{ - uint8x8_t result; - __asm__ ("zip1 %0.8b,%1.8b,%2.8b" - : "=w"(result) - : "w"(a), "w"(b) - : /* No clobbers */); - return result; -} - -__extension__ static __inline uint16x4_t __attribute__ ((__always_inline__)) -vzip1_u16 (uint16x4_t a, uint16x4_t b) -{ - uint16x4_t result; - __asm__ ("zip1 %0.4h,%1.4h,%2.4h" - : "=w"(result) - : "w"(a), "w"(b) - : /* No clobbers */); - return result; -} - -__extension__ static __inline uint32x2_t __attribute__ ((__always_inline__)) -vzip1_u32 (uint32x2_t a, uint32x2_t b) -{ - uint32x2_t result; - __asm__ ("zip1 %0.2s,%1.2s,%2.2s" - : "=w"(result) - : "w"(a), "w"(b) - : /* No clobbers */); - return result; -} - -__extension__ static __inline float32x4_t __attribute__ ((__always_inline__)) -vzip1q_f32 (float32x4_t a, float32x4_t b) -{ - float32x4_t result; - __asm__ ("zip1 %0.4s,%1.4s,%2.4s" - : "=w"(result) - : "w"(a), "w"(b) - : /* No clobbers */); - return result; -} - -__extension__ static __inline float64x2_t __attribute__ ((__always_inline__)) -vzip1q_f64 (float64x2_t a, float64x2_t b) -{ - float64x2_t result; - __asm__ ("zip1 %0.2d,%1.2d,%2.2d" - : "=w"(result) - : "w"(a), "w"(b) - : /* No clobbers */); - return result; -} - -__extension__ static __inline poly8x16_t __attribute__ ((__always_inline__)) -vzip1q_p8 (poly8x16_t a, poly8x16_t b) -{ - poly8x16_t result; - __asm__ ("zip1 %0.16b,%1.16b,%2.16b" - : "=w"(result) - : "w"(a), "w"(b) - : /* No clobbers */); - return result; -} - -__extension__ static __inline poly16x8_t __attribute__ ((__always_inline__)) -vzip1q_p16 (poly16x8_t a, poly16x8_t b) -{ - poly16x8_t result; - __asm__ ("zip1 %0.8h,%1.8h,%2.8h" - : "=w"(result) - : "w"(a), "w"(b) - : /* No clobbers */); - return result; -} - -__extension__ static __inline int8x16_t __attribute__ ((__always_inline__)) -vzip1q_s8 (int8x16_t a, int8x16_t b) -{ - int8x16_t result; - __asm__ ("zip1 %0.16b,%1.16b,%2.16b" - : "=w"(result) - : "w"(a), "w"(b) - : /* No clobbers */); - return result; -} - -__extension__ static __inline int16x8_t __attribute__ ((__always_inline__)) -vzip1q_s16 (int16x8_t a, int16x8_t b) -{ - int16x8_t result; - __asm__ ("zip1 %0.8h,%1.8h,%2.8h" - : "=w"(result) - : "w"(a), "w"(b) - : /* No clobbers */); - return result; -} - -__extension__ static __inline int32x4_t __attribute__ ((__always_inline__)) -vzip1q_s32 (int32x4_t a, int32x4_t b) -{ - int32x4_t result; - __asm__ ("zip1 %0.4s,%1.4s,%2.4s" - : "=w"(result) - : "w"(a), "w"(b) - : /* No clobbers */); - return result; -} - -__extension__ static __inline int64x2_t __attribute__ ((__always_inline__)) -vzip1q_s64 (int64x2_t a, int64x2_t b) -{ - int64x2_t result; - __asm__ ("zip1 %0.2d,%1.2d,%2.2d" - : "=w"(result) - : "w"(a), "w"(b) - : /* No clobbers */); - return result; -} - -__extension__ static __inline uint8x16_t __attribute__ ((__always_inline__)) -vzip1q_u8 (uint8x16_t a, uint8x16_t b) -{ - uint8x16_t result; - __asm__ ("zip1 %0.16b,%1.16b,%2.16b" - : "=w"(result) - : "w"(a), "w"(b) - : /* No clobbers */); - return result; -} - -__extension__ static __inline uint16x8_t __attribute__ ((__always_inline__)) -vzip1q_u16 (uint16x8_t a, uint16x8_t b) -{ - uint16x8_t result; - __asm__ ("zip1 %0.8h,%1.8h,%2.8h" - : "=w"(result) - : "w"(a), "w"(b) - : /* No clobbers */); - return result; -} - -__extension__ static __inline uint32x4_t __attribute__ ((__always_inline__)) -vzip1q_u32 (uint32x4_t a, uint32x4_t b) -{ - uint32x4_t result; - __asm__ ("zip1 %0.4s,%1.4s,%2.4s" - : "=w"(result) - : "w"(a), "w"(b) - : /* No clobbers */); - return result; -} - -__extension__ static __inline uint64x2_t __attribute__ ((__always_inline__)) -vzip1q_u64 (uint64x2_t a, uint64x2_t b) -{ - uint64x2_t result; - __asm__ ("zip1 %0.2d,%1.2d,%2.2d" - : "=w"(result) - : "w"(a), "w"(b) - : /* No clobbers */); - return result; -} - -__extension__ static __inline float32x2_t __attribute__ ((__always_inline__)) -vzip2_f32 (float32x2_t a, float32x2_t b) -{ - float32x2_t result; - __asm__ ("zip2 %0.2s,%1.2s,%2.2s" - : "=w"(result) - : "w"(a), "w"(b) - : /* No clobbers */); - return result; -} - -__extension__ static __inline poly8x8_t __attribute__ ((__always_inline__)) -vzip2_p8 (poly8x8_t a, poly8x8_t b) -{ - poly8x8_t result; - __asm__ ("zip2 %0.8b,%1.8b,%2.8b" - : "=w"(result) - : "w"(a), "w"(b) - : /* No clobbers */); - return result; -} - -__extension__ static __inline poly16x4_t __attribute__ ((__always_inline__)) -vzip2_p16 (poly16x4_t a, poly16x4_t b) -{ - poly16x4_t result; - __asm__ ("zip2 %0.4h,%1.4h,%2.4h" - : "=w"(result) - : "w"(a), "w"(b) - : /* No clobbers */); - return result; -} - -__extension__ static __inline int8x8_t __attribute__ ((__always_inline__)) -vzip2_s8 (int8x8_t a, int8x8_t b) -{ - int8x8_t result; - __asm__ ("zip2 %0.8b,%1.8b,%2.8b" - : "=w"(result) - : "w"(a), "w"(b) - : /* No clobbers */); - return result; -} - -__extension__ static __inline int16x4_t __attribute__ ((__always_inline__)) -vzip2_s16 (int16x4_t a, int16x4_t b) -{ - int16x4_t result; - __asm__ ("zip2 %0.4h,%1.4h,%2.4h" - : "=w"(result) - : "w"(a), "w"(b) - : /* No clobbers */); - return result; -} - -__extension__ static __inline int32x2_t __attribute__ ((__always_inline__)) -vzip2_s32 (int32x2_t a, int32x2_t b) -{ - int32x2_t result; - __asm__ ("zip2 %0.2s,%1.2s,%2.2s" - : "=w"(result) - : "w"(a), "w"(b) - : /* No clobbers */); - return result; -} - -__extension__ static __inline uint8x8_t __attribute__ ((__always_inline__)) -vzip2_u8 (uint8x8_t a, uint8x8_t b) -{ - uint8x8_t result; - __asm__ ("zip2 %0.8b,%1.8b,%2.8b" - : "=w"(result) - : "w"(a), "w"(b) - : /* No clobbers */); - return result; -} - -__extension__ static __inline uint16x4_t __attribute__ ((__always_inline__)) -vzip2_u16 (uint16x4_t a, uint16x4_t b) -{ - uint16x4_t result; - __asm__ ("zip2 %0.4h,%1.4h,%2.4h" - : "=w"(result) - : "w"(a), "w"(b) - : /* No clobbers */); - return result; -} - -__extension__ static __inline uint32x2_t __attribute__ ((__always_inline__)) -vzip2_u32 (uint32x2_t a, uint32x2_t b) -{ - uint32x2_t result; - __asm__ ("zip2 %0.2s,%1.2s,%2.2s" - : "=w"(result) - : "w"(a), "w"(b) - : /* No clobbers */); - return result; -} - -__extension__ static __inline float32x4_t __attribute__ ((__always_inline__)) -vzip2q_f32 (float32x4_t a, float32x4_t b) -{ - float32x4_t result; - __asm__ ("zip2 %0.4s,%1.4s,%2.4s" - : "=w"(result) - : "w"(a), "w"(b) - : /* No clobbers */); - return result; -} - -__extension__ static __inline float64x2_t __attribute__ ((__always_inline__)) -vzip2q_f64 (float64x2_t a, float64x2_t b) -{ - float64x2_t result; - __asm__ ("zip2 %0.2d,%1.2d,%2.2d" - : "=w"(result) - : "w"(a), "w"(b) - : /* No clobbers */); - return result; -} - -__extension__ static __inline poly8x16_t __attribute__ ((__always_inline__)) -vzip2q_p8 (poly8x16_t a, poly8x16_t b) -{ - poly8x16_t result; - __asm__ ("zip2 %0.16b,%1.16b,%2.16b" - : "=w"(result) - : "w"(a), "w"(b) - : /* No clobbers */); - return result; -} - -__extension__ static __inline poly16x8_t __attribute__ ((__always_inline__)) -vzip2q_p16 (poly16x8_t a, poly16x8_t b) -{ - poly16x8_t result; - __asm__ ("zip2 %0.8h,%1.8h,%2.8h" - : "=w"(result) - : "w"(a), "w"(b) - : /* No clobbers */); - return result; -} - -__extension__ static __inline int8x16_t __attribute__ ((__always_inline__)) -vzip2q_s8 (int8x16_t a, int8x16_t b) -{ - int8x16_t result; - __asm__ ("zip2 %0.16b,%1.16b,%2.16b" - : "=w"(result) - : "w"(a), "w"(b) - : /* No clobbers */); - return result; -} - -__extension__ static __inline int16x8_t __attribute__ ((__always_inline__)) -vzip2q_s16 (int16x8_t a, int16x8_t b) -{ - int16x8_t result; - __asm__ ("zip2 %0.8h,%1.8h,%2.8h" - : "=w"(result) - : "w"(a), "w"(b) - : /* No clobbers */); - return result; -} - -__extension__ static __inline int32x4_t __attribute__ ((__always_inline__)) -vzip2q_s32 (int32x4_t a, int32x4_t b) -{ - int32x4_t result; - __asm__ ("zip2 %0.4s,%1.4s,%2.4s" - : "=w"(result) - : "w"(a), "w"(b) - : /* No clobbers */); - return result; -} - -__extension__ static __inline int64x2_t __attribute__ ((__always_inline__)) -vzip2q_s64 (int64x2_t a, int64x2_t b) -{ - int64x2_t result; - __asm__ ("zip2 %0.2d,%1.2d,%2.2d" - : "=w"(result) - : "w"(a), "w"(b) - : /* No clobbers */); - return result; -} - -__extension__ static __inline uint8x16_t __attribute__ ((__always_inline__)) -vzip2q_u8 (uint8x16_t a, uint8x16_t b) -{ - uint8x16_t result; - __asm__ ("zip2 %0.16b,%1.16b,%2.16b" - : "=w"(result) - : "w"(a), "w"(b) - : /* No clobbers */); - return result; -} - -__extension__ static __inline uint16x8_t __attribute__ ((__always_inline__)) -vzip2q_u16 (uint16x8_t a, uint16x8_t b) -{ - uint16x8_t result; - __asm__ ("zip2 %0.8h,%1.8h,%2.8h" - : "=w"(result) - : "w"(a), "w"(b) - : /* No clobbers */); - return result; -} - -__extension__ static __inline uint32x4_t __attribute__ ((__always_inline__)) -vzip2q_u32 (uint32x4_t a, uint32x4_t b) -{ - uint32x4_t result; - __asm__ ("zip2 %0.4s,%1.4s,%2.4s" - : "=w"(result) - : "w"(a), "w"(b) - : /* No clobbers */); - return result; -} - -__extension__ static __inline uint64x2_t __attribute__ ((__always_inline__)) -vzip2q_u64 (uint64x2_t a, uint64x2_t b) -{ - uint64x2_t result; - __asm__ ("zip2 %0.2d,%1.2d,%2.2d" - : "=w"(result) - : "w"(a), "w"(b) - : /* No clobbers */); - return result; -} /* End of temporary inline asm implementations. */ @@ -14452,131 +13529,224 @@ __LD4_LANE_FUNC (uint16x8x4_t, uint16_t, 8h, h, u16, q) __LD4_LANE_FUNC (uint32x4x4_t, uint32_t, 4s, s, u32, q) __LD4_LANE_FUNC (uint64x2x4_t, uint64_t, 2d, d, u64, q) -#define __ST2_LANE_FUNC(intype, ptrtype, regsuffix, \ - lnsuffix, funcsuffix, Q) \ - typedef struct { ptrtype __x[2]; } __ST2_LANE_STRUCTURE_##intype; \ - __extension__ static __inline void \ - __attribute__ ((__always_inline__)) \ - vst2 ## Q ## _lane_ ## funcsuffix (ptrtype *ptr, \ - intype b, const int c) \ - { \ - __ST2_LANE_STRUCTURE_##intype *__p = \ - (__ST2_LANE_STRUCTURE_##intype *)ptr; \ - __asm__ ("ld1 {v16." #regsuffix ", v17." #regsuffix "}, %1\n\t" \ - "st2 {v16." #lnsuffix ", v17." #lnsuffix "}[%2], %0\n\t" \ - : "=Q"(*__p) \ - : "Q"(b), "i"(c) \ - : "v16", "v17"); \ - } - -__ST2_LANE_FUNC (int8x8x2_t, int8_t, 8b, b, s8,) -__ST2_LANE_FUNC (float32x2x2_t, float32_t, 2s, s, f32,) -__ST2_LANE_FUNC (float64x1x2_t, float64_t, 1d, d, f64,) -__ST2_LANE_FUNC (poly8x8x2_t, poly8_t, 8b, b, p8,) -__ST2_LANE_FUNC (poly16x4x2_t, poly16_t, 4h, h, p16,) -__ST2_LANE_FUNC (int16x4x2_t, int16_t, 4h, h, s16,) -__ST2_LANE_FUNC (int32x2x2_t, int32_t, 2s, s, s32,) -__ST2_LANE_FUNC (int64x1x2_t, int64_t, 1d, d, s64,) -__ST2_LANE_FUNC (uint8x8x2_t, uint8_t, 8b, b, u8,) -__ST2_LANE_FUNC (uint16x4x2_t, uint16_t, 4h, h, u16,) -__ST2_LANE_FUNC (uint32x2x2_t, uint32_t, 2s, s, u32,) -__ST2_LANE_FUNC (uint64x1x2_t, uint64_t, 1d, d, u64,) -__ST2_LANE_FUNC (float32x4x2_t, float32_t, 4s, s, f32, q) -__ST2_LANE_FUNC (float64x2x2_t, float64_t, 2d, d, f64, q) -__ST2_LANE_FUNC (poly8x16x2_t, poly8_t, 16b, b, p8, q) -__ST2_LANE_FUNC (poly16x8x2_t, poly16_t, 8h, h, p16, q) -__ST2_LANE_FUNC (int8x16x2_t, int8_t, 16b, b, s8, q) -__ST2_LANE_FUNC (int16x8x2_t, int16_t, 8h, h, s16, q) -__ST2_LANE_FUNC (int32x4x2_t, int32_t, 4s, s, s32, q) -__ST2_LANE_FUNC (int64x2x2_t, int64_t, 2d, d, s64, q) -__ST2_LANE_FUNC (uint8x16x2_t, uint8_t, 16b, b, u8, q) -__ST2_LANE_FUNC (uint16x8x2_t, uint16_t, 8h, h, u16, q) -__ST2_LANE_FUNC (uint32x4x2_t, uint32_t, 4s, s, u32, q) -__ST2_LANE_FUNC (uint64x2x2_t, uint64_t, 2d, d, u64, q) - -#define __ST3_LANE_FUNC(intype, ptrtype, regsuffix, \ - lnsuffix, funcsuffix, Q) \ - typedef struct { ptrtype __x[3]; } __ST3_LANE_STRUCTURE_##intype; \ - __extension__ static __inline void \ - __attribute__ ((__always_inline__)) \ - vst3 ## Q ## _lane_ ## funcsuffix (ptrtype *ptr, \ - intype b, const int c) \ - { \ - __ST3_LANE_STRUCTURE_##intype *__p = \ - (__ST3_LANE_STRUCTURE_##intype *)ptr; \ - __asm__ ("ld1 {v16." #regsuffix " - v18." #regsuffix "}, %1\n\t" \ - "st3 {v16." #lnsuffix " - v18." #lnsuffix "}[%2], %0\n\t" \ - : "=Q"(*__p) \ - : "Q"(b), "i"(c) \ - : "v16", "v17", "v18"); \ - } - -__ST3_LANE_FUNC (int8x8x3_t, int8_t, 8b, b, s8,) -__ST3_LANE_FUNC (float32x2x3_t, float32_t, 2s, s, f32,) -__ST3_LANE_FUNC (float64x1x3_t, float64_t, 1d, d, f64,) -__ST3_LANE_FUNC (poly8x8x3_t, poly8_t, 8b, b, p8,) -__ST3_LANE_FUNC (poly16x4x3_t, poly16_t, 4h, h, p16,) -__ST3_LANE_FUNC (int16x4x3_t, int16_t, 4h, h, s16,) -__ST3_LANE_FUNC (int32x2x3_t, int32_t, 2s, s, s32,) -__ST3_LANE_FUNC (int64x1x3_t, int64_t, 1d, d, s64,) -__ST3_LANE_FUNC (uint8x8x3_t, uint8_t, 8b, b, u8,) -__ST3_LANE_FUNC (uint16x4x3_t, uint16_t, 4h, h, u16,) -__ST3_LANE_FUNC (uint32x2x3_t, uint32_t, 2s, s, u32,) -__ST3_LANE_FUNC (uint64x1x3_t, uint64_t, 1d, d, u64,) -__ST3_LANE_FUNC (float32x4x3_t, float32_t, 4s, s, f32, q) -__ST3_LANE_FUNC (float64x2x3_t, float64_t, 2d, d, f64, q) -__ST3_LANE_FUNC (poly8x16x3_t, poly8_t, 16b, b, p8, q) -__ST3_LANE_FUNC (poly16x8x3_t, poly16_t, 8h, h, p16, q) -__ST3_LANE_FUNC (int8x16x3_t, int8_t, 16b, b, s8, q) -__ST3_LANE_FUNC (int16x8x3_t, int16_t, 8h, h, s16, q) -__ST3_LANE_FUNC (int32x4x3_t, int32_t, 4s, s, s32, q) -__ST3_LANE_FUNC (int64x2x3_t, int64_t, 2d, d, s64, q) -__ST3_LANE_FUNC (uint8x16x3_t, uint8_t, 16b, b, u8, q) -__ST3_LANE_FUNC (uint16x8x3_t, uint16_t, 8h, h, u16, q) -__ST3_LANE_FUNC (uint32x4x3_t, uint32_t, 4s, s, u32, q) -__ST3_LANE_FUNC (uint64x2x3_t, uint64_t, 2d, d, u64, q) - -#define __ST4_LANE_FUNC(intype, ptrtype, regsuffix, \ - lnsuffix, funcsuffix, Q) \ - typedef struct { ptrtype __x[4]; } __ST4_LANE_STRUCTURE_##intype; \ - __extension__ static __inline void \ - __attribute__ ((__always_inline__)) \ - vst4 ## Q ## _lane_ ## funcsuffix (ptrtype *ptr, \ - intype b, const int c) \ - { \ - __ST4_LANE_STRUCTURE_##intype *__p = \ - (__ST4_LANE_STRUCTURE_##intype *)ptr; \ - __asm__ ("ld1 {v16." #regsuffix " - v19." #regsuffix "}, %1\n\t" \ - "st4 {v16." #lnsuffix " - v19." #lnsuffix "}[%2], %0\n\t" \ - : "=Q"(*__p) \ - : "Q"(b), "i"(c) \ - : "v16", "v17", "v18", "v19"); \ - } - -__ST4_LANE_FUNC (int8x8x4_t, int8_t, 8b, b, s8,) -__ST4_LANE_FUNC (float32x2x4_t, float32_t, 2s, s, f32,) -__ST4_LANE_FUNC (float64x1x4_t, float64_t, 1d, d, f64,) -__ST4_LANE_FUNC (poly8x8x4_t, poly8_t, 8b, b, p8,) -__ST4_LANE_FUNC (poly16x4x4_t, poly16_t, 4h, h, p16,) -__ST4_LANE_FUNC (int16x4x4_t, int16_t, 4h, h, s16,) -__ST4_LANE_FUNC (int32x2x4_t, int32_t, 2s, s, s32,) -__ST4_LANE_FUNC (int64x1x4_t, int64_t, 1d, d, s64,) -__ST4_LANE_FUNC (uint8x8x4_t, uint8_t, 8b, b, u8,) -__ST4_LANE_FUNC (uint16x4x4_t, uint16_t, 4h, h, u16,) -__ST4_LANE_FUNC (uint32x2x4_t, uint32_t, 2s, s, u32,) -__ST4_LANE_FUNC (uint64x1x4_t, uint64_t, 1d, d, u64,) -__ST4_LANE_FUNC (float32x4x4_t, float32_t, 4s, s, f32, q) -__ST4_LANE_FUNC (float64x2x4_t, float64_t, 2d, d, f64, q) -__ST4_LANE_FUNC (poly8x16x4_t, poly8_t, 16b, b, p8, q) -__ST4_LANE_FUNC (poly16x8x4_t, poly16_t, 8h, h, p16, q) -__ST4_LANE_FUNC (int8x16x4_t, int8_t, 16b, b, s8, q) -__ST4_LANE_FUNC (int16x8x4_t, int16_t, 8h, h, s16, q) -__ST4_LANE_FUNC (int32x4x4_t, int32_t, 4s, s, s32, q) -__ST4_LANE_FUNC (int64x2x4_t, int64_t, 2d, d, s64, q) -__ST4_LANE_FUNC (uint8x16x4_t, uint8_t, 16b, b, u8, q) -__ST4_LANE_FUNC (uint16x8x4_t, uint16_t, 8h, h, u16, q) -__ST4_LANE_FUNC (uint32x4x4_t, uint32_t, 4s, s, u32, q) -__ST4_LANE_FUNC (uint64x2x4_t, uint64_t, 2d, d, u64, q) +#define __ST2_LANE_FUNC(intype, largetype, ptrtype, \ + mode, ptr_mode, funcsuffix, signedtype) \ +__extension__ static __inline void \ +__attribute__ ((__always_inline__)) \ +vst2_lane_ ## funcsuffix (ptrtype *__ptr, \ + intype __b, const int __c) \ +{ \ + __builtin_aarch64_simd_oi __o; \ + largetype __temp; \ + __temp.val[0] \ + = vcombine_##funcsuffix (__b.val[0], \ + vcreate_##funcsuffix (__AARCH64_UINT64_C (0))); \ + __temp.val[1] \ + = vcombine_##funcsuffix (__b.val[1], \ + vcreate_##funcsuffix (__AARCH64_UINT64_C (0))); \ + __o = __builtin_aarch64_set_qregoi##mode (__o, \ + (signedtype) __temp.val[0], 0); \ + __o = __builtin_aarch64_set_qregoi##mode (__o, \ + (signedtype) __temp.val[1], 1); \ + __builtin_aarch64_st2_lane##mode ((__builtin_aarch64_simd_ ## ptr_mode *) \ + __ptr, __o, __c); \ +} + +__ST2_LANE_FUNC (float32x2x2_t, float32x4x2_t, float32_t, v4sf, sf, f32, + float32x4_t) +__ST2_LANE_FUNC (float64x1x2_t, float64x2x2_t, float64_t, v2df, df, f64, + float64x2_t) +__ST2_LANE_FUNC (poly8x8x2_t, poly8x16x2_t, poly8_t, v16qi, qi, p8, int8x16_t) +__ST2_LANE_FUNC (poly16x4x2_t, poly16x8x2_t, poly16_t, v8hi, hi, p16, + int16x8_t) +__ST2_LANE_FUNC (int8x8x2_t, int8x16x2_t, int8_t, v16qi, qi, s8, int8x16_t) +__ST2_LANE_FUNC (int16x4x2_t, int16x8x2_t, int16_t, v8hi, hi, s16, int16x8_t) +__ST2_LANE_FUNC (int32x2x2_t, int32x4x2_t, int32_t, v4si, si, s32, int32x4_t) +__ST2_LANE_FUNC (int64x1x2_t, int64x2x2_t, int64_t, v2di, di, s64, int64x2_t) +__ST2_LANE_FUNC (uint8x8x2_t, uint8x16x2_t, uint8_t, v16qi, qi, u8, int8x16_t) +__ST2_LANE_FUNC (uint16x4x2_t, uint16x8x2_t, uint16_t, v8hi, hi, u16, + int16x8_t) +__ST2_LANE_FUNC (uint32x2x2_t, uint32x4x2_t, uint32_t, v4si, si, u32, + int32x4_t) +__ST2_LANE_FUNC (uint64x1x2_t, uint64x2x2_t, uint64_t, v2di, di, u64, + int64x2_t) + +#undef __ST2_LANE_FUNC +#define __ST2_LANE_FUNC(intype, ptrtype, mode, ptr_mode, funcsuffix) \ +__extension__ static __inline void \ +__attribute__ ((__always_inline__)) \ +vst2q_lane_ ## funcsuffix (ptrtype *__ptr, \ + intype __b, const int __c) \ +{ \ + union { intype __i; \ + __builtin_aarch64_simd_oi __o; } __temp = { __b }; \ + __builtin_aarch64_st2_lane##mode ((__builtin_aarch64_simd_ ## ptr_mode *) \ + __ptr, __temp.__o, __c); \ +} + +__ST2_LANE_FUNC (float32x4x2_t, float32_t, v4sf, sf, f32) +__ST2_LANE_FUNC (float64x2x2_t, float64_t, v2df, df, f64) +__ST2_LANE_FUNC (poly8x16x2_t, poly8_t, v16qi, qi, p8) +__ST2_LANE_FUNC (poly16x8x2_t, poly16_t, v8hi, hi, p16) +__ST2_LANE_FUNC (int8x16x2_t, int8_t, v16qi, qi, s8) +__ST2_LANE_FUNC (int16x8x2_t, int16_t, v8hi, hi, s16) +__ST2_LANE_FUNC (int32x4x2_t, int32_t, v4si, si, s32) +__ST2_LANE_FUNC (int64x2x2_t, int64_t, v2di, di, s64) +__ST2_LANE_FUNC (uint8x16x2_t, uint8_t, v16qi, qi, u8) +__ST2_LANE_FUNC (uint16x8x2_t, uint16_t, v8hi, hi, u16) +__ST2_LANE_FUNC (uint32x4x2_t, uint32_t, v4si, si, u32) +__ST2_LANE_FUNC (uint64x2x2_t, uint64_t, v2di, di, u64) + +#define __ST3_LANE_FUNC(intype, largetype, ptrtype, \ + mode, ptr_mode, funcsuffix, signedtype) \ +__extension__ static __inline void \ +__attribute__ ((__always_inline__)) \ +vst3_lane_ ## funcsuffix (ptrtype *__ptr, \ + intype __b, const int __c) \ +{ \ + __builtin_aarch64_simd_ci __o; \ + largetype __temp; \ + __temp.val[0] \ + = vcombine_##funcsuffix (__b.val[0], \ + vcreate_##funcsuffix (__AARCH64_UINT64_C (0))); \ + __temp.val[1] \ + = vcombine_##funcsuffix (__b.val[1], \ + vcreate_##funcsuffix (__AARCH64_UINT64_C (0))); \ + __temp.val[2] \ + = vcombine_##funcsuffix (__b.val[2], \ + vcreate_##funcsuffix (__AARCH64_UINT64_C (0))); \ + __o = __builtin_aarch64_set_qregci##mode (__o, \ + (signedtype) __temp.val[0], 0); \ + __o = __builtin_aarch64_set_qregci##mode (__o, \ + (signedtype) __temp.val[1], 1); \ + __o = __builtin_aarch64_set_qregci##mode (__o, \ + (signedtype) __temp.val[2], 2); \ + __builtin_aarch64_st3_lane##mode ((__builtin_aarch64_simd_ ## ptr_mode *) \ + __ptr, __o, __c); \ +} + +__ST3_LANE_FUNC (float32x2x3_t, float32x4x3_t, float32_t, v4sf, sf, f32, + float32x4_t) +__ST3_LANE_FUNC (float64x1x3_t, float64x2x3_t, float64_t, v2df, df, f64, + float64x2_t) +__ST3_LANE_FUNC (poly8x8x3_t, poly8x16x3_t, poly8_t, v16qi, qi, p8, int8x16_t) +__ST3_LANE_FUNC (poly16x4x3_t, poly16x8x3_t, poly16_t, v8hi, hi, p16, + int16x8_t) +__ST3_LANE_FUNC (int8x8x3_t, int8x16x3_t, int8_t, v16qi, qi, s8, int8x16_t) +__ST3_LANE_FUNC (int16x4x3_t, int16x8x3_t, int16_t, v8hi, hi, s16, int16x8_t) +__ST3_LANE_FUNC (int32x2x3_t, int32x4x3_t, int32_t, v4si, si, s32, int32x4_t) +__ST3_LANE_FUNC (int64x1x3_t, int64x2x3_t, int64_t, v2di, di, s64, int64x2_t) +__ST3_LANE_FUNC (uint8x8x3_t, uint8x16x3_t, uint8_t, v16qi, qi, u8, int8x16_t) +__ST3_LANE_FUNC (uint16x4x3_t, uint16x8x3_t, uint16_t, v8hi, hi, u16, + int16x8_t) +__ST3_LANE_FUNC (uint32x2x3_t, uint32x4x3_t, uint32_t, v4si, si, u32, + int32x4_t) +__ST3_LANE_FUNC (uint64x1x3_t, uint64x2x3_t, uint64_t, v2di, di, u64, + int64x2_t) + +#undef __ST3_LANE_FUNC +#define __ST3_LANE_FUNC(intype, ptrtype, mode, ptr_mode, funcsuffix) \ +__extension__ static __inline void \ +__attribute__ ((__always_inline__)) \ +vst3q_lane_ ## funcsuffix (ptrtype *__ptr, \ + intype __b, const int __c) \ +{ \ + union { intype __i; \ + __builtin_aarch64_simd_ci __o; } __temp = { __b }; \ + __builtin_aarch64_st3_lane##mode ((__builtin_aarch64_simd_ ## ptr_mode *) \ + __ptr, __temp.__o, __c); \ +} + +__ST3_LANE_FUNC (float32x4x3_t, float32_t, v4sf, sf, f32) +__ST3_LANE_FUNC (float64x2x3_t, float64_t, v2df, df, f64) +__ST3_LANE_FUNC (poly8x16x3_t, poly8_t, v16qi, qi, p8) +__ST3_LANE_FUNC (poly16x8x3_t, poly16_t, v8hi, hi, p16) +__ST3_LANE_FUNC (int8x16x3_t, int8_t, v16qi, qi, s8) +__ST3_LANE_FUNC (int16x8x3_t, int16_t, v8hi, hi, s16) +__ST3_LANE_FUNC (int32x4x3_t, int32_t, v4si, si, s32) +__ST3_LANE_FUNC (int64x2x3_t, int64_t, v2di, di, s64) +__ST3_LANE_FUNC (uint8x16x3_t, uint8_t, v16qi, qi, u8) +__ST3_LANE_FUNC (uint16x8x3_t, uint16_t, v8hi, hi, u16) +__ST3_LANE_FUNC (uint32x4x3_t, uint32_t, v4si, si, u32) +__ST3_LANE_FUNC (uint64x2x3_t, uint64_t, v2di, di, u64) + +#define __ST4_LANE_FUNC(intype, largetype, ptrtype, \ + mode, ptr_mode, funcsuffix, signedtype) \ +__extension__ static __inline void \ +__attribute__ ((__always_inline__)) \ +vst4_lane_ ## funcsuffix (ptrtype *__ptr, \ + intype __b, const int __c) \ +{ \ + __builtin_aarch64_simd_xi __o; \ + largetype __temp; \ + __temp.val[0] \ + = vcombine_##funcsuffix (__b.val[0], \ + vcreate_##funcsuffix (__AARCH64_UINT64_C (0))); \ + __temp.val[1] \ + = vcombine_##funcsuffix (__b.val[1], \ + vcreate_##funcsuffix (__AARCH64_UINT64_C (0))); \ + __temp.val[2] \ + = vcombine_##funcsuffix (__b.val[2], \ + vcreate_##funcsuffix (__AARCH64_UINT64_C (0))); \ + __temp.val[3] \ + = vcombine_##funcsuffix (__b.val[3], \ + vcreate_##funcsuffix (__AARCH64_UINT64_C (0))); \ + __o = __builtin_aarch64_set_qregxi##mode (__o, \ + (signedtype) __temp.val[0], 0); \ + __o = __builtin_aarch64_set_qregxi##mode (__o, \ + (signedtype) __temp.val[1], 1); \ + __o = __builtin_aarch64_set_qregxi##mode (__o, \ + (signedtype) __temp.val[2], 2); \ + __o = __builtin_aarch64_set_qregxi##mode (__o, \ + (signedtype) __temp.val[3], 3); \ + __builtin_aarch64_st4_lane##mode ((__builtin_aarch64_simd_ ## ptr_mode *) \ + __ptr, __o, __c); \ +} + +__ST4_LANE_FUNC (float32x2x4_t, float32x4x4_t, float32_t, v4sf, sf, f32, + float32x4_t) +__ST4_LANE_FUNC (float64x1x4_t, float64x2x4_t, float64_t, v2df, df, f64, + float64x2_t) +__ST4_LANE_FUNC (poly8x8x4_t, poly8x16x4_t, poly8_t, v16qi, qi, p8, int8x16_t) +__ST4_LANE_FUNC (poly16x4x4_t, poly16x8x4_t, poly16_t, v8hi, hi, p16, + int16x8_t) +__ST4_LANE_FUNC (int8x8x4_t, int8x16x4_t, int8_t, v16qi, qi, s8, int8x16_t) +__ST4_LANE_FUNC (int16x4x4_t, int16x8x4_t, int16_t, v8hi, hi, s16, int16x8_t) +__ST4_LANE_FUNC (int32x2x4_t, int32x4x4_t, int32_t, v4si, si, s32, int32x4_t) +__ST4_LANE_FUNC (int64x1x4_t, int64x2x4_t, int64_t, v2di, di, s64, int64x2_t) +__ST4_LANE_FUNC (uint8x8x4_t, uint8x16x4_t, uint8_t, v16qi, qi, u8, int8x16_t) +__ST4_LANE_FUNC (uint16x4x4_t, uint16x8x4_t, uint16_t, v8hi, hi, u16, + int16x8_t) +__ST4_LANE_FUNC (uint32x2x4_t, uint32x4x4_t, uint32_t, v4si, si, u32, + int32x4_t) +__ST4_LANE_FUNC (uint64x1x4_t, uint64x2x4_t, uint64_t, v2di, di, u64, + int64x2_t) + +#undef __ST4_LANE_FUNC +#define __ST4_LANE_FUNC(intype, ptrtype, mode, ptr_mode, funcsuffix) \ +__extension__ static __inline void \ +__attribute__ ((__always_inline__)) \ +vst4q_lane_ ## funcsuffix (ptrtype *__ptr, \ + intype __b, const int __c) \ +{ \ + union { intype __i; \ + __builtin_aarch64_simd_xi __o; } __temp = { __b }; \ + __builtin_aarch64_st4_lane##mode ((__builtin_aarch64_simd_ ## ptr_mode *) \ + __ptr, __temp.__o, __c); \ +} + +__ST4_LANE_FUNC (float32x4x4_t, float32_t, v4sf, sf, f32) +__ST4_LANE_FUNC (float64x2x4_t, float64_t, v2df, df, f64) +__ST4_LANE_FUNC (poly8x16x4_t, poly8_t, v16qi, qi, p8) +__ST4_LANE_FUNC (poly16x8x4_t, poly16_t, v8hi, hi, p16) +__ST4_LANE_FUNC (int8x16x4_t, int8_t, v16qi, qi, s8) +__ST4_LANE_FUNC (int16x8x4_t, int16_t, v8hi, hi, s16) +__ST4_LANE_FUNC (int32x4x4_t, int32_t, v4si, si, s32) +__ST4_LANE_FUNC (int64x2x4_t, int64_t, v2di, di, s64) +__ST4_LANE_FUNC (uint8x16x4_t, uint8_t, v16qi, qi, u8) +__ST4_LANE_FUNC (uint16x8x4_t, uint16_t, v8hi, hi, u16) +__ST4_LANE_FUNC (uint32x4x4_t, uint32_t, v4si, si, u32) +__ST4_LANE_FUNC (uint64x2x4_t, uint64_t, v2di, di, u64) __extension__ static __inline int64_t __attribute__ ((__always_inline__)) vaddlv_s32 (int32x2_t a) @@ -25614,10 +24784,880 @@ vuqaddd_s64 (int64x1_t __a, uint64x1_t __b) /* vuzp */ +__extension__ static __inline float32x2_t __attribute__ ((__always_inline__)) +vuzp1_f32 (float32x2_t __a, float32x2_t __b) +{ +#ifdef __AARCH64EB__ + return __builtin_shuffle (__a, __b, (uint32x2_t) {3, 1}); +#else + return __builtin_shuffle (__a, __b, (uint32x2_t) {0, 2}); +#endif +} + +__extension__ static __inline poly8x8_t __attribute__ ((__always_inline__)) +vuzp1_p8 (poly8x8_t __a, poly8x8_t __b) +{ +#ifdef __AARCH64EB__ + return __builtin_shuffle (__a, __b, (uint8x8_t) {9, 11, 13, 15, 1, 3, 5, 7}); +#else + return __builtin_shuffle (__a, __b, (uint8x8_t) {0, 2, 4, 6, 8, 10, 12, 14}); +#endif +} + +__extension__ static __inline poly16x4_t __attribute__ ((__always_inline__)) +vuzp1_p16 (poly16x4_t __a, poly16x4_t __b) +{ +#ifdef __AARCH64EB__ + return __builtin_shuffle (__a, __b, (uint16x4_t) {5, 7, 1, 3}); +#else + return __builtin_shuffle (__a, __b, (uint16x4_t) {0, 2, 4, 6}); +#endif +} + +__extension__ static __inline int8x8_t __attribute__ ((__always_inline__)) +vuzp1_s8 (int8x8_t __a, int8x8_t __b) +{ +#ifdef __AARCH64EB__ + return __builtin_shuffle (__a, __b, (uint8x8_t) {9, 11, 13, 15, 1, 3, 5, 7}); +#else + return __builtin_shuffle (__a, __b, (uint8x8_t) {0, 2, 4, 6, 8, 10, 12, 14}); +#endif +} + +__extension__ static __inline int16x4_t __attribute__ ((__always_inline__)) +vuzp1_s16 (int16x4_t __a, int16x4_t __b) +{ +#ifdef __AARCH64EB__ + return __builtin_shuffle (__a, __b, (uint16x4_t) {5, 7, 1, 3}); +#else + return __builtin_shuffle (__a, __b, (uint16x4_t) {0, 2, 4, 6}); +#endif +} + +__extension__ static __inline int32x2_t __attribute__ ((__always_inline__)) +vuzp1_s32 (int32x2_t __a, int32x2_t __b) +{ +#ifdef __AARCH64EB__ + return __builtin_shuffle (__a, __b, (uint32x2_t) {3, 1}); +#else + return __builtin_shuffle (__a, __b, (uint32x2_t) {0, 2}); +#endif +} + +__extension__ static __inline uint8x8_t __attribute__ ((__always_inline__)) +vuzp1_u8 (uint8x8_t __a, uint8x8_t __b) +{ +#ifdef __AARCH64EB__ + return __builtin_shuffle (__a, __b, (uint8x8_t) {9, 11, 13, 15, 1, 3, 5, 7}); +#else + return __builtin_shuffle (__a, __b, (uint8x8_t) {0, 2, 4, 6, 8, 10, 12, 14}); +#endif +} + +__extension__ static __inline uint16x4_t __attribute__ ((__always_inline__)) +vuzp1_u16 (uint16x4_t __a, uint16x4_t __b) +{ +#ifdef __AARCH64EB__ + return __builtin_shuffle (__a, __b, (uint16x4_t) {5, 7, 1, 3}); +#else + return __builtin_shuffle (__a, __b, (uint16x4_t) {0, 2, 4, 6}); +#endif +} + +__extension__ static __inline uint32x2_t __attribute__ ((__always_inline__)) +vuzp1_u32 (uint32x2_t __a, uint32x2_t __b) +{ +#ifdef __AARCH64EB__ + return __builtin_shuffle (__a, __b, (uint32x2_t) {3, 1}); +#else + return __builtin_shuffle (__a, __b, (uint32x2_t) {0, 2}); +#endif +} + +__extension__ static __inline float32x4_t __attribute__ ((__always_inline__)) +vuzp1q_f32 (float32x4_t __a, float32x4_t __b) +{ +#ifdef __AARCH64EB__ + return __builtin_shuffle (__a, __b, (uint32x4_t) {5, 7, 1, 3}); +#else + return __builtin_shuffle (__a, __b, (uint32x4_t) {0, 2, 4, 6}); +#endif +} + +__extension__ static __inline float64x2_t __attribute__ ((__always_inline__)) +vuzp1q_f64 (float64x2_t __a, float64x2_t __b) +{ +#ifdef __AARCH64EB__ + return __builtin_shuffle (__a, __b, (uint64x2_t) {3, 1}); +#else + return __builtin_shuffle (__a, __b, (uint64x2_t) {0, 2}); +#endif +} + +__extension__ static __inline poly8x16_t __attribute__ ((__always_inline__)) +vuzp1q_p8 (poly8x16_t __a, poly8x16_t __b) +{ +#ifdef __AARCH64EB__ + return __builtin_shuffle (__a, __b, (uint8x16_t) + {17, 19, 21, 23, 25, 27, 29, 31, 1, 3, 5, 7, 9, 11, 13, 15}); +#else + return __builtin_shuffle (__a, __b, (uint8x16_t) + {0, 2, 4, 6, 8, 10, 12, 14, 16, 18, 20, 22, 24, 26, 28, 30}); +#endif +} + +__extension__ static __inline poly16x8_t __attribute__ ((__always_inline__)) +vuzp1q_p16 (poly16x8_t __a, poly16x8_t __b) +{ +#ifdef __AARCH64EB__ + return __builtin_shuffle (__a, __b, (uint16x8_t) {9, 11, 13, 15, 1, 3, 5, 7}); +#else + return __builtin_shuffle (__a, __b, (uint16x8_t) {0, 2, 4, 6, 8, 10, 12, 14}); +#endif +} + +__extension__ static __inline int8x16_t __attribute__ ((__always_inline__)) +vuzp1q_s8 (int8x16_t __a, int8x16_t __b) +{ +#ifdef __AARCH64EB__ + return __builtin_shuffle (__a, __b, + (uint8x16_t) {17, 19, 21, 23, 25, 27, 29, 31, 1, 3, 5, 7, 9, 11, 13, 15}); +#else + return __builtin_shuffle (__a, __b, + (uint8x16_t) {0, 2, 4, 6, 8, 10, 12, 14, 16, 18, 20, 22, 24, 26, 28, 30}); +#endif +} + +__extension__ static __inline int16x8_t __attribute__ ((__always_inline__)) +vuzp1q_s16 (int16x8_t __a, int16x8_t __b) +{ +#ifdef __AARCH64EB__ + return __builtin_shuffle (__a, __b, (uint16x8_t) {9, 11, 13, 15, 1, 3, 5, 7}); +#else + return __builtin_shuffle (__a, __b, (uint16x8_t) {0, 2, 4, 6, 8, 10, 12, 14}); +#endif +} + +__extension__ static __inline int32x4_t __attribute__ ((__always_inline__)) +vuzp1q_s32 (int32x4_t __a, int32x4_t __b) +{ +#ifdef __AARCH64EB__ + return __builtin_shuffle (__a, __b, (uint32x4_t) {5, 7, 1, 3}); +#else + return __builtin_shuffle (__a, __b, (uint32x4_t) {0, 2, 4, 6}); +#endif +} + +__extension__ static __inline int64x2_t __attribute__ ((__always_inline__)) +vuzp1q_s64 (int64x2_t __a, int64x2_t __b) +{ +#ifdef __AARCH64EB__ + return __builtin_shuffle (__a, __b, (uint64x2_t) {3, 1}); +#else + return __builtin_shuffle (__a, __b, (uint64x2_t) {0, 2}); +#endif +} + +__extension__ static __inline uint8x16_t __attribute__ ((__always_inline__)) +vuzp1q_u8 (uint8x16_t __a, uint8x16_t __b) +{ +#ifdef __AARCH64EB__ + return __builtin_shuffle (__a, __b, + (uint8x16_t) {17, 19, 21, 23, 25, 27, 29, 31, 1, 3, 5, 7, 9, 11, 13, 15}); +#else + return __builtin_shuffle (__a, __b, + (uint8x16_t) {0, 2, 4, 6, 8, 10, 12, 14, 16, 18, 20, 22, 24, 26, 28, 30}); +#endif +} + +__extension__ static __inline uint16x8_t __attribute__ ((__always_inline__)) +vuzp1q_u16 (uint16x8_t __a, uint16x8_t __b) +{ +#ifdef __AARCH64EB__ + return __builtin_shuffle (__a, __b, (uint16x8_t) {9, 11, 13, 15, 1, 3, 5, 7}); +#else + return __builtin_shuffle (__a, __b, (uint16x8_t) {0, 2, 4, 6, 8, 10, 12, 14}); +#endif +} + +__extension__ static __inline uint32x4_t __attribute__ ((__always_inline__)) +vuzp1q_u32 (uint32x4_t __a, uint32x4_t __b) +{ +#ifdef __AARCH64EB__ + return __builtin_shuffle (__a, __b, (uint32x4_t) {5, 7, 1, 3}); +#else + return __builtin_shuffle (__a, __b, (uint32x4_t) {0, 2, 4, 6}); +#endif +} + +__extension__ static __inline uint64x2_t __attribute__ ((__always_inline__)) +vuzp1q_u64 (uint64x2_t __a, uint64x2_t __b) +{ +#ifdef __AARCH64EB__ + return __builtin_shuffle (__a, __b, (uint64x2_t) {3, 1}); +#else + return __builtin_shuffle (__a, __b, (uint64x2_t) {0, 2}); +#endif +} + +__extension__ static __inline float32x2_t __attribute__ ((__always_inline__)) +vuzp2_f32 (float32x2_t __a, float32x2_t __b) +{ +#ifdef __AARCH64EB__ + return __builtin_shuffle (__a, __b, (uint32x2_t) {2, 0}); +#else + return __builtin_shuffle (__a, __b, (uint32x2_t) {1, 3}); +#endif +} + +__extension__ static __inline poly8x8_t __attribute__ ((__always_inline__)) +vuzp2_p8 (poly8x8_t __a, poly8x8_t __b) +{ +#ifdef __AARCH64EB__ + return __builtin_shuffle (__a, __b, (uint8x8_t) {8, 10, 12, 14, 0, 2, 4, 6}); +#else + return __builtin_shuffle (__a, __b, (uint8x8_t) {1, 3, 5, 7, 9, 11, 13, 15}); +#endif +} + +__extension__ static __inline poly16x4_t __attribute__ ((__always_inline__)) +vuzp2_p16 (poly16x4_t __a, poly16x4_t __b) +{ +#ifdef __AARCH64EB__ + return __builtin_shuffle (__a, __b, (uint16x4_t) {4, 6, 0, 2}); +#else + return __builtin_shuffle (__a, __b, (uint16x4_t) {1, 3, 5, 7}); +#endif +} + +__extension__ static __inline int8x8_t __attribute__ ((__always_inline__)) +vuzp2_s8 (int8x8_t __a, int8x8_t __b) +{ +#ifdef __AARCH64EB__ + return __builtin_shuffle (__a, __b, (uint8x8_t) {8, 10, 12, 14, 0, 2, 4, 6}); +#else + return __builtin_shuffle (__a, __b, (uint8x8_t) {1, 3, 5, 7, 9, 11, 13, 15}); +#endif +} + +__extension__ static __inline int16x4_t __attribute__ ((__always_inline__)) +vuzp2_s16 (int16x4_t __a, int16x4_t __b) +{ +#ifdef __AARCH64EB__ + return __builtin_shuffle (__a, __b, (uint16x4_t) {4, 6, 0, 2}); +#else + return __builtin_shuffle (__a, __b, (uint16x4_t) {1, 3, 5, 7}); +#endif +} + +__extension__ static __inline int32x2_t __attribute__ ((__always_inline__)) +vuzp2_s32 (int32x2_t __a, int32x2_t __b) +{ +#ifdef __AARCH64EB__ + return __builtin_shuffle (__a, __b, (uint32x2_t) {2, 0}); +#else + return __builtin_shuffle (__a, __b, (uint32x2_t) {1, 3}); +#endif +} + +__extension__ static __inline uint8x8_t __attribute__ ((__always_inline__)) +vuzp2_u8 (uint8x8_t __a, uint8x8_t __b) +{ +#ifdef __AARCH64EB__ + return __builtin_shuffle (__a, __b, (uint8x8_t) {8, 10, 12, 14, 0, 2, 4, 6}); +#else + return __builtin_shuffle (__a, __b, (uint8x8_t) {1, 3, 5, 7, 9, 11, 13, 15}); +#endif +} + +__extension__ static __inline uint16x4_t __attribute__ ((__always_inline__)) +vuzp2_u16 (uint16x4_t __a, uint16x4_t __b) +{ +#ifdef __AARCH64EB__ + return __builtin_shuffle (__a, __b, (uint16x4_t) {4, 6, 0, 2}); +#else + return __builtin_shuffle (__a, __b, (uint16x4_t) {1, 3, 5, 7}); +#endif +} + +__extension__ static __inline uint32x2_t __attribute__ ((__always_inline__)) +vuzp2_u32 (uint32x2_t __a, uint32x2_t __b) +{ +#ifdef __AARCH64EB__ + return __builtin_shuffle (__a, __b, (uint32x2_t) {2, 0}); +#else + return __builtin_shuffle (__a, __b, (uint32x2_t) {1, 3}); +#endif +} + +__extension__ static __inline float32x4_t __attribute__ ((__always_inline__)) +vuzp2q_f32 (float32x4_t __a, float32x4_t __b) +{ +#ifdef __AARCH64EB__ + return __builtin_shuffle (__a, __b, (uint32x4_t) {4, 6, 0, 2}); +#else + return __builtin_shuffle (__a, __b, (uint32x4_t) {1, 3, 5, 7}); +#endif +} + +__extension__ static __inline float64x2_t __attribute__ ((__always_inline__)) +vuzp2q_f64 (float64x2_t __a, float64x2_t __b) +{ +#ifdef __AARCH64EB__ + return __builtin_shuffle (__a, __b, (uint64x2_t) {2, 0}); +#else + return __builtin_shuffle (__a, __b, (uint64x2_t) {1, 3}); +#endif +} + +__extension__ static __inline poly8x16_t __attribute__ ((__always_inline__)) +vuzp2q_p8 (poly8x16_t __a, poly8x16_t __b) +{ +#ifdef __AARCH64EB__ + return __builtin_shuffle (__a, __b, + (uint8x16_t) {16, 18, 20, 22, 24, 26, 28, 30, 0, 2, 4, 6, 8, 10, 12, 14}); +#else + return __builtin_shuffle (__a, __b, + (uint8x16_t) {1, 3, 5, 7, 9, 11, 13, 15, 17, 19, 21, 23, 25, 27, 29, 31}); +#endif +} + +__extension__ static __inline poly16x8_t __attribute__ ((__always_inline__)) +vuzp2q_p16 (poly16x8_t __a, poly16x8_t __b) +{ +#ifdef __AARCH64EB__ + return __builtin_shuffle (__a, __b, (uint16x8_t) {8, 10, 12, 14, 0, 2, 4, 6}); +#else + return __builtin_shuffle (__a, __b, (uint16x8_t) {1, 3, 5, 7, 9, 11, 13, 15}); +#endif +} + +__extension__ static __inline int8x16_t __attribute__ ((__always_inline__)) +vuzp2q_s8 (int8x16_t __a, int8x16_t __b) +{ +#ifdef __AARCH64EB__ + return __builtin_shuffle (__a, __b, + (uint8x16_t) {16, 18, 20, 22, 24, 26, 28, 30, 0, 2, 4, 6, 8, 10, 12, 14}); +#else + return __builtin_shuffle (__a, __b, + (uint8x16_t) {1, 3, 5, 7, 9, 11, 13, 15, 17, 19, 21, 23, 25, 27, 29, 31}); +#endif +} + +__extension__ static __inline int16x8_t __attribute__ ((__always_inline__)) +vuzp2q_s16 (int16x8_t __a, int16x8_t __b) +{ +#ifdef __AARCH64EB__ + return __builtin_shuffle (__a, __b, (uint16x8_t) {8, 10, 12, 14, 0, 2, 4, 6}); +#else + return __builtin_shuffle (__a, __b, (uint16x8_t) {1, 3, 5, 7, 9, 11, 13, 15}); +#endif +} + +__extension__ static __inline int32x4_t __attribute__ ((__always_inline__)) +vuzp2q_s32 (int32x4_t __a, int32x4_t __b) +{ +#ifdef __AARCH64EB__ + return __builtin_shuffle (__a, __b, (uint32x4_t) {4, 6, 0, 2}); +#else + return __builtin_shuffle (__a, __b, (uint32x4_t) {1, 3, 5, 7}); +#endif +} + +__extension__ static __inline int64x2_t __attribute__ ((__always_inline__)) +vuzp2q_s64 (int64x2_t __a, int64x2_t __b) +{ +#ifdef __AARCH64EB__ + return __builtin_shuffle (__a, __b, (uint64x2_t) {2, 0}); +#else + return __builtin_shuffle (__a, __b, (uint64x2_t) {1, 3}); +#endif +} + +__extension__ static __inline uint8x16_t __attribute__ ((__always_inline__)) +vuzp2q_u8 (uint8x16_t __a, uint8x16_t __b) +{ +#ifdef __AARCH64EB__ + return __builtin_shuffle (__a, __b, (uint8x16_t) + {16, 18, 20, 22, 24, 26, 28, 30, 0, 2, 4, 6, 8, 10, 12, 14}); +#else + return __builtin_shuffle (__a, __b, (uint8x16_t) + {1, 3, 5, 7, 9, 11, 13, 15, 17, 19, 21, 23, 25, 27, 29, 31}); +#endif +} + +__extension__ static __inline uint16x8_t __attribute__ ((__always_inline__)) +vuzp2q_u16 (uint16x8_t __a, uint16x8_t __b) +{ +#ifdef __AARCH64EB__ + return __builtin_shuffle (__a, __b, (uint16x8_t) {8, 10, 12, 14, 0, 2, 4, 6}); +#else + return __builtin_shuffle (__a, __b, (uint16x8_t) {1, 3, 5, 7, 9, 11, 13, 15}); +#endif +} + +__extension__ static __inline uint32x4_t __attribute__ ((__always_inline__)) +vuzp2q_u32 (uint32x4_t __a, uint32x4_t __b) +{ +#ifdef __AARCH64EB__ + return __builtin_shuffle (__a, __b, (uint32x4_t) {4, 6, 0, 2}); +#else + return __builtin_shuffle (__a, __b, (uint32x4_t) {1, 3, 5, 7}); +#endif +} + +__extension__ static __inline uint64x2_t __attribute__ ((__always_inline__)) +vuzp2q_u64 (uint64x2_t __a, uint64x2_t __b) +{ +#ifdef __AARCH64EB__ + return __builtin_shuffle (__a, __b, (uint64x2_t) {2, 0}); +#else + return __builtin_shuffle (__a, __b, (uint64x2_t) {1, 3}); +#endif +} + __INTERLEAVE_LIST (uzp) /* vzip */ +__extension__ static __inline float32x2_t __attribute__ ((__always_inline__)) +vzip1_f32 (float32x2_t __a, float32x2_t __b) +{ +#ifdef __AARCH64EB__ + return __builtin_shuffle (__a, __b, (uint32x2_t) {3, 1}); +#else + return __builtin_shuffle (__a, __b, (uint32x2_t) {0, 2}); +#endif +} + +__extension__ static __inline poly8x8_t __attribute__ ((__always_inline__)) +vzip1_p8 (poly8x8_t __a, poly8x8_t __b) +{ +#ifdef __AARCH64EB__ + return __builtin_shuffle (__a, __b, (uint8x8_t) {12, 4, 13, 5, 14, 6, 15, 7}); +#else + return __builtin_shuffle (__a, __b, (uint8x8_t) {0, 8, 1, 9, 2, 10, 3, 11}); +#endif +} + +__extension__ static __inline poly16x4_t __attribute__ ((__always_inline__)) +vzip1_p16 (poly16x4_t __a, poly16x4_t __b) +{ +#ifdef __AARCH64EB__ + return __builtin_shuffle (__a, __b, (uint16x4_t) {6, 2, 7, 3}); +#else + return __builtin_shuffle (__a, __b, (uint16x4_t) {0, 4, 1, 5}); +#endif +} + +__extension__ static __inline int8x8_t __attribute__ ((__always_inline__)) +vzip1_s8 (int8x8_t __a, int8x8_t __b) +{ +#ifdef __AARCH64EB__ + return __builtin_shuffle (__a, __b, (uint8x8_t) {12, 4, 13, 5, 14, 6, 15, 7}); +#else + return __builtin_shuffle (__a, __b, (uint8x8_t) {0, 8, 1, 9, 2, 10, 3, 11}); +#endif +} + +__extension__ static __inline int16x4_t __attribute__ ((__always_inline__)) +vzip1_s16 (int16x4_t __a, int16x4_t __b) +{ +#ifdef __AARCH64EB__ + return __builtin_shuffle (__a, __b, (uint16x4_t) {6, 2, 7, 3}); +#else + return __builtin_shuffle (__a, __b, (uint16x4_t) {0, 4, 1, 5}); +#endif +} + +__extension__ static __inline int32x2_t __attribute__ ((__always_inline__)) +vzip1_s32 (int32x2_t __a, int32x2_t __b) +{ +#ifdef __AARCH64EB__ + return __builtin_shuffle (__a, __b, (uint32x2_t) {3, 1}); +#else + return __builtin_shuffle (__a, __b, (uint32x2_t) {0, 2}); +#endif +} + +__extension__ static __inline uint8x8_t __attribute__ ((__always_inline__)) +vzip1_u8 (uint8x8_t __a, uint8x8_t __b) +{ +#ifdef __AARCH64EB__ + return __builtin_shuffle (__a, __b, (uint8x8_t) {12, 4, 13, 5, 14, 6, 15, 7}); +#else + return __builtin_shuffle (__a, __b, (uint8x8_t) {0, 8, 1, 9, 2, 10, 3, 11}); +#endif +} + +__extension__ static __inline uint16x4_t __attribute__ ((__always_inline__)) +vzip1_u16 (uint16x4_t __a, uint16x4_t __b) +{ +#ifdef __AARCH64EB__ + return __builtin_shuffle (__a, __b, (uint16x4_t) {6, 2, 7, 3}); +#else + return __builtin_shuffle (__a, __b, (uint16x4_t) {0, 4, 1, 5}); +#endif +} + +__extension__ static __inline uint32x2_t __attribute__ ((__always_inline__)) +vzip1_u32 (uint32x2_t __a, uint32x2_t __b) +{ +#ifdef __AARCH64EB__ + return __builtin_shuffle (__a, __b, (uint32x2_t) {3, 1}); +#else + return __builtin_shuffle (__a, __b, (uint32x2_t) {0, 2}); +#endif +} + +__extension__ static __inline float32x4_t __attribute__ ((__always_inline__)) +vzip1q_f32 (float32x4_t __a, float32x4_t __b) +{ +#ifdef __AARCH64EB__ + return __builtin_shuffle (__a, __b, (uint32x4_t) {6, 2, 7, 3}); +#else + return __builtin_shuffle (__a, __b, (uint32x4_t) {0, 4, 1, 5}); +#endif +} + +__extension__ static __inline float64x2_t __attribute__ ((__always_inline__)) +vzip1q_f64 (float64x2_t __a, float64x2_t __b) +{ +#ifdef __AARCH64EB__ + return __builtin_shuffle (__a, __b, (uint64x2_t) {3, 1}); +#else + return __builtin_shuffle (__a, __b, (uint64x2_t) {0, 2}); +#endif +} + +__extension__ static __inline poly8x16_t __attribute__ ((__always_inline__)) +vzip1q_p8 (poly8x16_t __a, poly8x16_t __b) +{ +#ifdef __AARCH64EB__ + return __builtin_shuffle (__a, __b, (uint8x16_t) + {24, 8, 25, 9, 26, 10, 27, 11, 28, 12, 29, 13, 30, 14, 31, 15}); +#else + return __builtin_shuffle (__a, __b, (uint8x16_t) + {0, 16, 1, 17, 2, 18, 3, 19, 4, 20, 5, 21, 6, 22, 7, 23}); +#endif +} + +__extension__ static __inline poly16x8_t __attribute__ ((__always_inline__)) +vzip1q_p16 (poly16x8_t __a, poly16x8_t __b) +{ +#ifdef __AARCH64EB__ + return __builtin_shuffle (__a, __b, (uint16x8_t) + {12, 4, 13, 5, 14, 6, 15, 7}); +#else + return __builtin_shuffle (__a, __b, (uint16x8_t) {0, 8, 1, 9, 2, 10, 3, 11}); +#endif +} + +__extension__ static __inline int8x16_t __attribute__ ((__always_inline__)) +vzip1q_s8 (int8x16_t __a, int8x16_t __b) +{ +#ifdef __AARCH64EB__ + return __builtin_shuffle (__a, __b, (uint8x16_t) + {24, 8, 25, 9, 26, 10, 27, 11, 28, 12, 29, 13, 30, 14, 31, 15}); +#else + return __builtin_shuffle (__a, __b, (uint8x16_t) + {0, 16, 1, 17, 2, 18, 3, 19, 4, 20, 5, 21, 6, 22, 7, 23}); +#endif +} + +__extension__ static __inline int16x8_t __attribute__ ((__always_inline__)) +vzip1q_s16 (int16x8_t __a, int16x8_t __b) +{ +#ifdef __AARCH64EB__ + return __builtin_shuffle (__a, __b, (uint16x8_t) + {12, 4, 13, 5, 14, 6, 15, 7}); +#else + return __builtin_shuffle (__a, __b, (uint16x8_t) {0, 8, 1, 9, 2, 10, 3, 11}); +#endif +} + +__extension__ static __inline int32x4_t __attribute__ ((__always_inline__)) +vzip1q_s32 (int32x4_t __a, int32x4_t __b) +{ +#ifdef __AARCH64EB__ + return __builtin_shuffle (__a, __b, (uint32x4_t) {6, 2, 7, 3}); +#else + return __builtin_shuffle (__a, __b, (uint32x4_t) {0, 4, 1, 5}); +#endif +} + +__extension__ static __inline int64x2_t __attribute__ ((__always_inline__)) +vzip1q_s64 (int64x2_t __a, int64x2_t __b) +{ +#ifdef __AARCH64EB__ + return __builtin_shuffle (__a, __b, (uint64x2_t) {3, 1}); +#else + return __builtin_shuffle (__a, __b, (uint64x2_t) {0, 2}); +#endif +} + +__extension__ static __inline uint8x16_t __attribute__ ((__always_inline__)) +vzip1q_u8 (uint8x16_t __a, uint8x16_t __b) +{ +#ifdef __AARCH64EB__ + return __builtin_shuffle (__a, __b, (uint8x16_t) + {24, 8, 25, 9, 26, 10, 27, 11, 28, 12, 29, 13, 30, 14, 31, 15}); +#else + return __builtin_shuffle (__a, __b, (uint8x16_t) + {0, 16, 1, 17, 2, 18, 3, 19, 4, 20, 5, 21, 6, 22, 7, 23}); +#endif +} + +__extension__ static __inline uint16x8_t __attribute__ ((__always_inline__)) +vzip1q_u16 (uint16x8_t __a, uint16x8_t __b) +{ +#ifdef __AARCH64EB__ + return __builtin_shuffle (__a, __b, (uint16x8_t) + {12, 4, 13, 5, 14, 6, 15, 7}); +#else + return __builtin_shuffle (__a, __b, (uint16x8_t) {0, 8, 1, 9, 2, 10, 3, 11}); +#endif +} + +__extension__ static __inline uint32x4_t __attribute__ ((__always_inline__)) +vzip1q_u32 (uint32x4_t __a, uint32x4_t __b) +{ +#ifdef __AARCH64EB__ + return __builtin_shuffle (__a, __b, (uint32x4_t) {6, 2, 7, 3}); +#else + return __builtin_shuffle (__a, __b, (uint32x4_t) {0, 4, 1, 5}); +#endif +} + +__extension__ static __inline uint64x2_t __attribute__ ((__always_inline__)) +vzip1q_u64 (uint64x2_t __a, uint64x2_t __b) +{ +#ifdef __AARCH64EB__ + return __builtin_shuffle (__a, __b, (uint64x2_t) {3, 1}); +#else + return __builtin_shuffle (__a, __b, (uint64x2_t) {0, 2}); +#endif +} + +__extension__ static __inline float32x2_t __attribute__ ((__always_inline__)) +vzip2_f32 (float32x2_t __a, float32x2_t __b) +{ +#ifdef __AARCH64EB__ + return __builtin_shuffle (__a, __b, (uint32x2_t) {2, 0}); +#else + return __builtin_shuffle (__a, __b, (uint32x2_t) {1, 3}); +#endif +} + +__extension__ static __inline poly8x8_t __attribute__ ((__always_inline__)) +vzip2_p8 (poly8x8_t __a, poly8x8_t __b) +{ +#ifdef __AARCH64EB__ + return __builtin_shuffle (__a, __b, (uint8x8_t) {8, 0, 9, 1, 10, 2, 11, 3}); +#else + return __builtin_shuffle (__a, __b, (uint8x8_t) {4, 12, 5, 13, 6, 14, 7, 15}); +#endif +} + +__extension__ static __inline poly16x4_t __attribute__ ((__always_inline__)) +vzip2_p16 (poly16x4_t __a, poly16x4_t __b) +{ +#ifdef __AARCH64EB__ + return __builtin_shuffle (__a, __b, (uint16x4_t) {4, 0, 5, 1}); +#else + return __builtin_shuffle (__a, __b, (uint16x4_t) {2, 6, 3, 7}); +#endif +} + +__extension__ static __inline int8x8_t __attribute__ ((__always_inline__)) +vzip2_s8 (int8x8_t __a, int8x8_t __b) +{ +#ifdef __AARCH64EB__ + return __builtin_shuffle (__a, __b, (uint8x8_t) {8, 0, 9, 1, 10, 2, 11, 3}); +#else + return __builtin_shuffle (__a, __b, (uint8x8_t) {4, 12, 5, 13, 6, 14, 7, 15}); +#endif +} + +__extension__ static __inline int16x4_t __attribute__ ((__always_inline__)) +vzip2_s16 (int16x4_t __a, int16x4_t __b) +{ +#ifdef __AARCH64EB__ + return __builtin_shuffle (__a, __b, (uint16x4_t) {4, 0, 5, 1}); +#else + return __builtin_shuffle (__a, __b, (uint16x4_t) {2, 6, 3, 7}); +#endif +} + +__extension__ static __inline int32x2_t __attribute__ ((__always_inline__)) +vzip2_s32 (int32x2_t __a, int32x2_t __b) +{ +#ifdef __AARCH64EB__ + return __builtin_shuffle (__a, __b, (uint32x2_t) {2, 0}); +#else + return __builtin_shuffle (__a, __b, (uint32x2_t) {1, 3}); +#endif +} + +__extension__ static __inline uint8x8_t __attribute__ ((__always_inline__)) +vzip2_u8 (uint8x8_t __a, uint8x8_t __b) +{ +#ifdef __AARCH64EB__ + return __builtin_shuffle (__a, __b, (uint8x8_t) {8, 0, 9, 1, 10, 2, 11, 3}); +#else + return __builtin_shuffle (__a, __b, (uint8x8_t) {4, 12, 5, 13, 6, 14, 7, 15}); +#endif +} + +__extension__ static __inline uint16x4_t __attribute__ ((__always_inline__)) +vzip2_u16 (uint16x4_t __a, uint16x4_t __b) +{ +#ifdef __AARCH64EB__ + return __builtin_shuffle (__a, __b, (uint16x4_t) {4, 0, 5, 1}); +#else + return __builtin_shuffle (__a, __b, (uint16x4_t) {2, 6, 3, 7}); +#endif +} + +__extension__ static __inline uint32x2_t __attribute__ ((__always_inline__)) +vzip2_u32 (uint32x2_t __a, uint32x2_t __b) +{ +#ifdef __AARCH64EB__ + return __builtin_shuffle (__a, __b, (uint32x2_t) {2, 0}); +#else + return __builtin_shuffle (__a, __b, (uint32x2_t) {1, 3}); +#endif +} + +__extension__ static __inline float32x4_t __attribute__ ((__always_inline__)) +vzip2q_f32 (float32x4_t __a, float32x4_t __b) +{ +#ifdef __AARCH64EB__ + return __builtin_shuffle (__a, __b, (uint32x4_t) {4, 0, 5, 1}); +#else + return __builtin_shuffle (__a, __b, (uint32x4_t) {2, 6, 3, 7}); +#endif +} + +__extension__ static __inline float64x2_t __attribute__ ((__always_inline__)) +vzip2q_f64 (float64x2_t __a, float64x2_t __b) +{ +#ifdef __AARCH64EB__ + return __builtin_shuffle (__a, __b, (uint64x2_t) {2, 0}); +#else + return __builtin_shuffle (__a, __b, (uint64x2_t) {1, 3}); +#endif +} + +__extension__ static __inline poly8x16_t __attribute__ ((__always_inline__)) +vzip2q_p8 (poly8x16_t __a, poly8x16_t __b) +{ +#ifdef __AARCH64EB__ + return __builtin_shuffle (__a, __b, (uint8x16_t) + {16, 0, 17, 1, 18, 2, 19, 3, 20, 4, 21, 5, 22, 6, 23, 7}); +#else + return __builtin_shuffle (__a, __b, (uint8x16_t) + {8, 24, 9, 25, 10, 26, 11, 27, 12, 28, 13, 29, 14, 30, 15, 31}); +#endif +} + +__extension__ static __inline poly16x8_t __attribute__ ((__always_inline__)) +vzip2q_p16 (poly16x8_t __a, poly16x8_t __b) +{ +#ifdef __AARCH64EB__ + return __builtin_shuffle (__a, __b, (uint16x8_t) {8, 0, 9, 1, 10, 2, 11, 3}); +#else + return __builtin_shuffle (__a, __b, (uint16x8_t) + {4, 12, 5, 13, 6, 14, 7, 15}); +#endif +} + +__extension__ static __inline int8x16_t __attribute__ ((__always_inline__)) +vzip2q_s8 (int8x16_t __a, int8x16_t __b) +{ +#ifdef __AARCH64EB__ + return __builtin_shuffle (__a, __b, (uint8x16_t) + {16, 0, 17, 1, 18, 2, 19, 3, 20, 4, 21, 5, 22, 6, 23, 7}); +#else + return __builtin_shuffle (__a, __b, (uint8x16_t) + {8, 24, 9, 25, 10, 26, 11, 27, 12, 28, 13, 29, 14, 30, 15, 31}); +#endif +} + +__extension__ static __inline int16x8_t __attribute__ ((__always_inline__)) +vzip2q_s16 (int16x8_t __a, int16x8_t __b) +{ +#ifdef __AARCH64EB__ + return __builtin_shuffle (__a, __b, (uint16x8_t) {8, 0, 9, 1, 10, 2, 11, 3}); +#else + return __builtin_shuffle (__a, __b, (uint16x8_t) + {4, 12, 5, 13, 6, 14, 7, 15}); +#endif +} + +__extension__ static __inline int32x4_t __attribute__ ((__always_inline__)) +vzip2q_s32 (int32x4_t __a, int32x4_t __b) +{ +#ifdef __AARCH64EB__ + return __builtin_shuffle (__a, __b, (uint32x4_t) {4, 0, 5, 1}); +#else + return __builtin_shuffle (__a, __b, (uint32x4_t) {2, 6, 3, 7}); +#endif +} + +__extension__ static __inline int64x2_t __attribute__ ((__always_inline__)) +vzip2q_s64 (int64x2_t __a, int64x2_t __b) +{ +#ifdef __AARCH64EB__ + return __builtin_shuffle (__a, __b, (uint64x2_t) {2, 0}); +#else + return __builtin_shuffle (__a, __b, (uint64x2_t) {1, 3}); +#endif +} + +__extension__ static __inline uint8x16_t __attribute__ ((__always_inline__)) +vzip2q_u8 (uint8x16_t __a, uint8x16_t __b) +{ +#ifdef __AARCH64EB__ + return __builtin_shuffle (__a, __b, (uint8x16_t) + {16, 0, 17, 1, 18, 2, 19, 3, 20, 4, 21, 5, 22, 6, 23, 7}); +#else + return __builtin_shuffle (__a, __b, (uint8x16_t) + {8, 24, 9, 25, 10, 26, 11, 27, 12, 28, 13, 29, 14, 30, 15, 31}); +#endif +} + +__extension__ static __inline uint16x8_t __attribute__ ((__always_inline__)) +vzip2q_u16 (uint16x8_t __a, uint16x8_t __b) +{ +#ifdef __AARCH64EB__ + return __builtin_shuffle (__a, __b, (uint16x8_t) {8, 0, 9, 1, 10, 2, 11, 3}); +#else + return __builtin_shuffle (__a, __b, (uint16x8_t) + {4, 12, 5, 13, 6, 14, 7, 15}); +#endif +} + +__extension__ static __inline uint32x4_t __attribute__ ((__always_inline__)) +vzip2q_u32 (uint32x4_t __a, uint32x4_t __b) +{ +#ifdef __AARCH64EB__ + return __builtin_shuffle (__a, __b, (uint32x4_t) {4, 0, 5, 1}); +#else + return __builtin_shuffle (__a, __b, (uint32x4_t) {2, 6, 3, 7}); +#endif +} + +__extension__ static __inline uint64x2_t __attribute__ ((__always_inline__)) +vzip2q_u64 (uint64x2_t __a, uint64x2_t __b) +{ +#ifdef __AARCH64EB__ + return __builtin_shuffle (__a, __b, (uint64x2_t) {2, 0}); +#else + return __builtin_shuffle (__a, __b, (uint64x2_t) {1, 3}); +#endif +} + __INTERLEAVE_LIST (zip) #undef __INTERLEAVE_LIST diff --git a/gcc/config/aarch64/iterators.md b/gcc/config/aarch64/iterators.md index fd1eb482f0f..c537c3780ee 100644 --- a/gcc/config/aarch64/iterators.md +++ b/gcc/config/aarch64/iterators.md @@ -564,6 +564,32 @@ (define_mode_attr VSTRUCT_DREG [(OI "TI") (CI "EI") (XI "OI")]) +;; Mode of pair of elements for each vector mode, to define transfer +;; size for structure lane/dup loads and stores. +(define_mode_attr V_TWO_ELEM [(V8QI "HI") (V16QI "HI") + (V4HI "SI") (V8HI "SI") + (V2SI "V2SI") (V4SI "V2SI") + (DI "V2DI") (V2DI "V2DI") + (V2SF "V2SF") (V4SF "V2SF") + (DF "V2DI") (V2DF "V2DI")]) + +;; Similar, for three elements. +(define_mode_attr V_THREE_ELEM [(V8QI "BLK") (V16QI "BLK") + (V4HI "BLK") (V8HI "BLK") + (V2SI "BLK") (V4SI "BLK") + (DI "EI") (V2DI "EI") + (V2SF "BLK") (V4SF "BLK") + (DF "EI") (V2DF "EI")]) + +;; Similar, for four elements. +(define_mode_attr V_FOUR_ELEM [(V8QI "SI") (V16QI "SI") + (V4HI "V4HI") (V8HI "V4HI") + (V2SI "V4SI") (V4SI "V4SI") + (DI "OI") (V2DI "OI") + (V2SF "V4SF") (V4SF "V4SF") + (DF "OI") (V2DF "OI")]) + + ;; Mode for atomic operation suffixes (define_mode_attr atomic_sfx [(QI "b") (HI "h") (SI "") (DI "")]) diff --git a/gcc/config/arc/arc.c b/gcc/config/arc/arc.c index 987a7c9a55a..58d95d832c5 100644 --- a/gcc/config/arc/arc.c +++ b/gcc/config/arc/arc.c @@ -996,7 +996,7 @@ arc_select_cc_mode (enum rtx_code op, rtx x, rtx y) if (GET_MODE_CLASS (mode) == MODE_INT && y == const0_rtx && (op == EQ || op == NE - || ((op == LT || op == GE) && GET_MODE_SIZE (GET_MODE (x) <= 4)))) + || ((op == LT || op == GE) && GET_MODE_SIZE (GET_MODE (x)) <= 4)))) return CC_ZNmode; /* add.f for if (a+b) */ @@ -1135,31 +1135,33 @@ arc_init_reg_tables (void) for (i = 0; i < NUM_MACHINE_MODES; i++) { - switch (GET_MODE_CLASS (i)) + enum machine_mode m = (enum machine_mode) i; + + switch (GET_MODE_CLASS (m)) { case MODE_INT: case MODE_PARTIAL_INT: case MODE_COMPLEX_INT: - if (GET_MODE_SIZE (i) <= 4) + if (GET_MODE_SIZE (m) <= 4) arc_mode_class[i] = 1 << (int) S_MODE; - else if (GET_MODE_SIZE (i) == 8) + else if (GET_MODE_SIZE (m) == 8) arc_mode_class[i] = 1 << (int) D_MODE; - else if (GET_MODE_SIZE (i) == 16) + else if (GET_MODE_SIZE (m) == 16) arc_mode_class[i] = 1 << (int) T_MODE; - else if (GET_MODE_SIZE (i) == 32) + else if (GET_MODE_SIZE (m) == 32) arc_mode_class[i] = 1 << (int) O_MODE; else arc_mode_class[i] = 0; break; case MODE_FLOAT: case MODE_COMPLEX_FLOAT: - if (GET_MODE_SIZE (i) <= 4) + if (GET_MODE_SIZE (m) <= 4) arc_mode_class[i] = 1 << (int) SF_MODE; - else if (GET_MODE_SIZE (i) == 8) + else if (GET_MODE_SIZE (m) == 8) arc_mode_class[i] = 1 << (int) DF_MODE; - else if (GET_MODE_SIZE (i) == 16) + else if (GET_MODE_SIZE (m) == 16) arc_mode_class[i] = 1 << (int) TF_MODE; - else if (GET_MODE_SIZE (i) == 32) + else if (GET_MODE_SIZE (m) == 32) arc_mode_class[i] = 1 << (int) OF_MODE; else arc_mode_class[i] = 0; diff --git a/gcc/config/arc/arc.opt b/gcc/config/arc/arc.opt index 2deb9e77e13..1e98db97095 100644 --- a/gcc/config/arc/arc.opt +++ b/gcc/config/arc/arc.opt @@ -340,9 +340,9 @@ Pass -marclinux_prof option through to linker. ;; lra is still unproven for ARC, so allow to fall back to reload with -mno-lra. ;Target InverseMask(NO_LRA) -mlra ; lra still won't allow to configure libgcc; see PR rtl-optimization/55464. ; so don't enable by default. +mlra Target Mask(LRA) Enable lra diff --git a/gcc/config/i386/sol2-bi.h b/gcc/config/i386/sol2-bi.h deleted file mode 100644 index 66d17801f03..00000000000 --- a/gcc/config/i386/sol2-bi.h +++ /dev/null @@ -1,109 +0,0 @@ -/* Definitions of target machine for GCC, for bi-arch Solaris 2/x86. - Copyright (C) 2004-2014 Free Software Foundation, Inc. - Contributed by CodeSourcery, LLC. - -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/>. */ - -/* Override i386/sol2.h version: return 8-byte vectors in MMX registers if - possible, matching Sun Studio 12 Update 1+ compilers and other x86 - targets. */ -#undef TARGET_SUBTARGET_DEFAULT -#define TARGET_SUBTARGET_DEFAULT \ - (MASK_80387 | MASK_IEEE_FP | MASK_FLOAT_RETURNS) - -#define SUBTARGET_OPTIMIZATION_OPTIONS \ - { OPT_LEVELS_1_PLUS, OPT_momit_leaf_frame_pointer, NULL, 1 } - -/* GNU as understands --32 and --64, but the native Solaris - assembler requires -xarch=generic or -xarch=generic64 instead. */ -#ifdef USE_GAS -#define ASM_CPU32_DEFAULT_SPEC "--32" -#define ASM_CPU64_DEFAULT_SPEC "--64" -#else -#define ASM_CPU32_DEFAULT_SPEC "-xarch=generic" -#define ASM_CPU64_DEFAULT_SPEC "-xarch=generic64" -#endif - -#undef ASM_CPU_SPEC -#define ASM_CPU_SPEC "%(asm_cpu_default)" - -/* Don't let i386/x86-64.h override i386/sol2.h version. Still cannot use - -K PIC with the Solaris 10+ assembler, it gives many warnings: - Absolute relocation is used for symbol "<symbol>" */ -#undef ASM_SPEC -#define ASM_SPEC ASM_SPEC_BASE - -/* We do not need to search a special directory for startup files. */ -#undef MD_STARTFILE_PREFIX - -#define DEFAULT_ARCH32_P !TARGET_64BIT_DEFAULT - -#define ARCH64_SUBDIR "amd64" - -#ifdef USE_GLD -/* Since binutils 2.21, GNU ld supports new *_sol2 emulations to strictly - follow the Solaris 2 ABI. Prefer them if present. */ -#ifdef HAVE_LD_SOL2_EMULATION -#define ARCH32_EMULATION "elf_i386_sol2" -#define ARCH64_EMULATION "elf_x86_64_sol2" -#else -#define ARCH32_EMULATION "elf_i386" -#define ARCH64_EMULATION "elf_x86_64" -#endif -#endif - -#undef ASM_COMMENT_START -#define ASM_COMMENT_START "/" - -/* The native Solaris assembler can't calculate the difference between - symbols in different sections, which causes problems for -fPIC jump - tables in .rodata. */ -#ifndef HAVE_AS_IX86_DIFF_SECT_DELTA -#undef JUMP_TABLES_IN_TEXT_SECTION -#define JUMP_TABLES_IN_TEXT_SECTION 1 - -/* The native Solaris assembler cannot handle the SYMBOL-. syntax, but - requires SYMBOL@rel/@rel64 instead. */ -#define ASM_OUTPUT_DWARF_PCREL(FILE, SIZE, LABEL) \ - do { \ - fputs (integer_asm_op (SIZE, FALSE), FILE); \ - assemble_name (FILE, LABEL); \ - fputs (SIZE == 8 ? "@rel64" : "@rel", FILE); \ - } while (0) -#endif - -/* As in sol2.h, override the default from i386/x86-64.h to work around - Sun as TLS bug. */ -#undef ASM_OUTPUT_ALIGNED_COMMON -#define ASM_OUTPUT_ALIGNED_COMMON(FILE, NAME, SIZE, ALIGN) \ - do \ - { \ - if (TARGET_SUN_TLS \ - && in_section \ - && ((in_section->common.flags & SECTION_TLS) == SECTION_TLS)) \ - switch_to_section (bss_section); \ - x86_elf_aligned_common (FILE, NAME, SIZE, ALIGN); \ - } \ - while (0) - -#define USE_IX86_FRAME_POINTER 1 -#define USE_X86_64_FRAME_POINTER 1 - -#undef NO_PROFILE_COUNTERS - -#undef MCOUNT_NAME -#define MCOUNT_NAME "_mcount" diff --git a/gcc/config/i386/sol2.h b/gcc/config/i386/sol2.h index 6676941f67a..9c3a6f49662 100644 --- a/gcc/config/i386/sol2.h +++ b/gcc/config/i386/sol2.h @@ -18,12 +18,8 @@ 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/>. */ -/* Augment i386/unix.h version to return 8-byte vectors in memory, matching - Sun Studio compilers until version 12, the only ones supported on - Solaris 9. */ -#undef TARGET_SUBTARGET_DEFAULT -#define TARGET_SUBTARGET_DEFAULT \ - (MASK_80387 | MASK_IEEE_FP | MASK_FLOAT_RETURNS | MASK_VECT8_RETURNS) +#define SUBTARGET_OPTIMIZATION_OPTIONS \ + { OPT_LEVELS_1_PLUS, OPT_momit_leaf_frame_pointer, NULL, 1 } /* Old versions of the Solaris assembler can not handle the difference of labels in different sections, so force DW_EH_PE_datarel if so. */ @@ -50,27 +46,46 @@ along with GCC; see the file COPYING3. If not see #undef TARGET_SUN_TLS #define TARGET_SUN_TLS 1 -#undef SIZE_TYPE -#define SIZE_TYPE "unsigned int" - -#undef PTRDIFF_TYPE -#define PTRDIFF_TYPE "int" - /* Solaris 2/Intel as chokes on #line directives before Solaris 10. */ #undef CPP_SPEC #define CPP_SPEC "%{,assembler-with-cpp:-P} %(cpp_subtarget)" -#define ASM_CPU_DEFAULT_SPEC "" +/* GNU as understands --32 and --64, but the native Solaris + assembler requires -xarch=generic or -xarch=generic64 instead. */ +#ifdef USE_GAS +#define ASM_CPU32_DEFAULT_SPEC "--32" +#define ASM_CPU64_DEFAULT_SPEC "--64" +#else +#define ASM_CPU32_DEFAULT_SPEC "-xarch=generic" +#define ASM_CPU64_DEFAULT_SPEC "-xarch=generic64" +#endif + +#undef ASM_CPU_SPEC +#define ASM_CPU_SPEC "%(asm_cpu_default)" -#define ASM_CPU_SPEC "" - -/* Don't include ASM_PIC_SPEC. While the Solaris 9 assembler accepts - -K PIC, it gives many warnings: - R_386_32 relocation is used for symbol "<symbol>" +/* Don't include ASM_PIC_SPEC. While the Solaris 10+ assembler accepts -K PIC, + it gives many warnings: + Absolute relocation is used for symbol "<symbol>" GNU as doesn't recognize -K at all. */ #undef ASM_SPEC #define ASM_SPEC ASM_SPEC_BASE +#define DEFAULT_ARCH32_P !TARGET_64BIT_DEFAULT + +#define ARCH64_SUBDIR "amd64" + +#ifdef USE_GLD +/* Since binutils 2.21, GNU ld supports new *_sol2 emulations to strictly + follow the Solaris 2 ABI. Prefer them if present. */ +#ifdef HAVE_LD_SOL2_EMULATION +#define ARCH32_EMULATION "elf_i386_sol2" +#define ARCH64_EMULATION "elf_x86_64_sol2" +#else +#define ARCH32_EMULATION "elf_i386" +#define ARCH64_EMULATION "elf_x86_64" +#endif +#endif + #undef ENDFILE_SPEC #define ENDFILE_SPEC \ "%{Ofast|ffast-math|funsafe-math-optimizations:crtfastmath.o%s} \ @@ -84,23 +99,39 @@ along with GCC; see the file COPYING3. If not see { "asm_cpu", ASM_CPU_SPEC }, \ { "asm_cpu_default", ASM_CPU_DEFAULT_SPEC }, \ -#undef SUBTARGET_EXTRA_SPECS -#define SUBTARGET_EXTRA_SPECS \ - { "startfile_arch", STARTFILE_ARCH_SPEC }, \ - { "link_arch", LINK_ARCH_SPEC }, \ - SUBTARGET_CPU_EXTRA_SPECS - /* Register the Solaris-specific #pragma directives. */ #define REGISTER_SUBTARGET_PRAGMAS() solaris_register_pragmas () #undef LOCAL_LABEL_PREFIX #define LOCAL_LABEL_PREFIX "." +/* The Solaris 10 FCS as doesn't accept "#" comments, while later versions + do. */ +#undef ASM_COMMENT_START +#define ASM_COMMENT_START "/" + /* The 32-bit Solaris assembler does not support .quad. Do not use it. */ #ifndef HAVE_AS_IX86_QUAD #undef ASM_QUAD #endif +/* The native Solaris assembler can't calculate the difference between + symbols in different sections, which causes problems for -fPIC jump + tables in .rodata. */ +#ifndef HAVE_AS_IX86_DIFF_SECT_DELTA +#undef JUMP_TABLES_IN_TEXT_SECTION +#define JUMP_TABLES_IN_TEXT_SECTION 1 + +/* The native Solaris assembler cannot handle the SYMBOL-. syntax, but + requires SYMBOL@rel/@rel64 instead. */ +#define ASM_OUTPUT_DWARF_PCREL(FILE, SIZE, LABEL) \ + do { \ + fputs (integer_asm_op (SIZE, FALSE), FILE); \ + assemble_name (FILE, LABEL); \ + fputs (SIZE == 8 ? "@rel64" : "@rel", FILE); \ + } while (0) +#endif + /* The Solaris assembler wants a .local for non-exported aliases. */ #define ASM_OUTPUT_DEF_FROM_DECLS(FILE, DECL, TARGET) \ do { \ @@ -148,6 +179,20 @@ along with GCC; see the file COPYING3. If not see while (0) #endif /* !USE_GAS */ +/* As in sparc/sol2.h, override the default from i386/x86-64.h to work + around Sun as TLS bug. */ +#undef ASM_OUTPUT_ALIGNED_COMMON +#define ASM_OUTPUT_ALIGNED_COMMON(FILE, NAME, SIZE, ALIGN) \ + do \ + { \ + if (TARGET_SUN_TLS \ + && in_section \ + && ((in_section->common.flags & SECTION_TLS) == SECTION_TLS)) \ + switch_to_section (bss_section); \ + x86_elf_aligned_common (FILE, NAME, SIZE, ALIGN); \ + } \ + while (0) + /* Output a simple call for .init/.fini. */ #define ASM_OUTPUT_CALL(FILE, FN) \ do \ @@ -174,6 +219,14 @@ along with GCC; see the file COPYING3. If not see #define DTORS_SECTION_ASM_OP "\t.section\t.dtors, \"aw\"" #endif +#define USE_IX86_FRAME_POINTER 1 +#define USE_X86_64_FRAME_POINTER 1 + +#undef NO_PROFILE_COUNTERS + +#undef MCOUNT_NAME +#define MCOUNT_NAME "_mcount" + /* We do not need NT_VERSION notes. */ #undef X86_FILE_START_VERSION_DIRECTIVE #define X86_FILE_START_VERSION_DIRECTIVE false diff --git a/gcc/config/i386/t-sol2-64 b/gcc/config/i386/t-sol2 index 4e70f0bed27..4e70f0bed27 100644 --- a/gcc/config/i386/t-sol2-64 +++ b/gcc/config/i386/t-sol2 diff --git a/gcc/config/m32c/m32c.c b/gcc/config/m32c/m32c.c index 57cfb20ee16..837c22b2d77 100644 --- a/gcc/config/m32c/m32c.c +++ b/gcc/config/m32c/m32c.c @@ -3159,7 +3159,7 @@ m32c_illegal_subreg_p (rtx op) { int offset; unsigned int i; - int src_mode, dest_mode; + enum machine_mode src_mode, dest_mode; if (GET_CODE (op) == MEM && ! m32c_legitimate_address_p (Pmode, XEXP (op, 0), false)) diff --git a/gcc/config/m32r/m32r.c b/gcc/config/m32r/m32r.c index 83bc3a7bf3a..2b84b0ff1eb 100644 --- a/gcc/config/m32r/m32r.c +++ b/gcc/config/m32r/m32r.c @@ -282,31 +282,33 @@ init_reg_tables (void) for (i = 0; i < NUM_MACHINE_MODES; i++) { - switch (GET_MODE_CLASS (i)) + enum machine_mode m = (enum machine_mode) i; + + switch (GET_MODE_CLASS (m)) { case MODE_INT: case MODE_PARTIAL_INT: case MODE_COMPLEX_INT: - if (GET_MODE_SIZE (i) <= 4) + if (GET_MODE_SIZE (m) <= 4) m32r_mode_class[i] = 1 << (int) S_MODE; - else if (GET_MODE_SIZE (i) == 8) + else if (GET_MODE_SIZE (m) == 8) m32r_mode_class[i] = 1 << (int) D_MODE; - else if (GET_MODE_SIZE (i) == 16) + else if (GET_MODE_SIZE (m) == 16) m32r_mode_class[i] = 1 << (int) T_MODE; - else if (GET_MODE_SIZE (i) == 32) + else if (GET_MODE_SIZE (m) == 32) m32r_mode_class[i] = 1 << (int) O_MODE; else m32r_mode_class[i] = 0; break; case MODE_FLOAT: case MODE_COMPLEX_FLOAT: - if (GET_MODE_SIZE (i) <= 4) + if (GET_MODE_SIZE (m) <= 4) m32r_mode_class[i] = 1 << (int) SF_MODE; - else if (GET_MODE_SIZE (i) == 8) + else if (GET_MODE_SIZE (m) == 8) m32r_mode_class[i] = 1 << (int) DF_MODE; - else if (GET_MODE_SIZE (i) == 16) + else if (GET_MODE_SIZE (m) == 16) m32r_mode_class[i] = 1 << (int) TF_MODE; - else if (GET_MODE_SIZE (i) == 32) + else if (GET_MODE_SIZE (m) == 32) m32r_mode_class[i] = 1 << (int) OF_MODE; else m32r_mode_class[i] = 0; diff --git a/gcc/config/msp430/msp430.c b/gcc/config/msp430/msp430.c index a637e27d41b..1ec96526efd 100644 --- a/gcc/config/msp430/msp430.c +++ b/gcc/config/msp430/msp430.c @@ -2162,7 +2162,7 @@ msp430_print_operand (FILE * file, rtx op, int letter) because builtins are expanded before the frame layout is determined. */ fprintf (file, "%d", msp430_initial_elimination_offset (ARG_POINTER_REGNUM, STACK_POINTER_REGNUM) - - 2); + - (TARGET_LARGE ? 4 : 2)); return; case 'J': diff --git a/gcc/config/msp430/msp430.md b/gcc/config/msp430/msp430.md index 5d930c37901..74a98b48019 100644 --- a/gcc/config/msp430/msp430.md +++ b/gcc/config/msp430/msp430.md @@ -1321,7 +1321,7 @@ [(set (match_operand:SI 0 "register_operand" "=r") (mult:SI (sign_extend:SI (match_operand:HI 1 "register_operand" "%0")) (sign_extend:SI (match_operand:HI 2 "register_operand" "r"))))] - "optimize > 2 && msp430_hwmult_type != NONE && ! msp430_is_interrupt_func ()" + "optimize > 2 && msp430_hwmult_type != NONE" "* if (msp430_use_f5_series_hwmult ()) return \"PUSH.W sr { DINT { MOV.W %1, &0x04C2 { MOV.W %2, &0x04C8 { MOV.W &0x04CA, %L0 { MOV.W &0x04CC, %H0 { POP.W sr\"; @@ -1365,6 +1365,6 @@ if (msp430_use_f5_series_hwmult ()) return \"PUSH.W sr { DINT { MOV.W %L1, &0x04D0 { MOV.W %H1, &0x04D2 { MOV.W %L2, &0x04E0 { MOV.W %H2, &0x04E2 { MOV.W &0x04E4, %A0 { MOV.W &0x04E6, %B0 { MOV.W &0x04E8, %C0 { MOV.W &0x04EA, %D0 { POP.W sr\"; else - return \"PUSH.W sr { DINT { MOV.W %L1, &0x0140 { MOV.W %H1, &0x0141 { MOV.W %L2, &0x0150 { MOV.W %H2, &0x0152 { MOV.W &0x0154, %A0 { MOV.W &0x0156, %B0 { MOV.W &0x0158, %C0 { MOV.W &0x015A, %D0 { POP.W sr\"; + return \"PUSH.W sr { DINT { MOV.W %L1, &0x0140 { MOV.W %H1, &0x0142 { MOV.W %L2, &0x0150 { MOV.W %H2, &0x0152 { MOV.W &0x0154, %A0 { MOV.W &0x0156, %B0 { MOV.W &0x0158, %C0 { MOV.W &0x015A, %D0 { POP.W sr\"; " ) diff --git a/gcc/config/rs6000/predicates.md b/gcc/config/rs6000/predicates.md index 1616b888c9c..47050c3d03e 100644 --- a/gcc/config/rs6000/predicates.md +++ b/gcc/config/rs6000/predicates.md @@ -629,14 +629,14 @@ (match_test "offsettable_nonstrict_memref_p (op)"))) ;; Return 1 if the operand is suitable for load/store quad memory. -;; This predicate only checks for non-atomic loads/stores. +;; This predicate only checks for non-atomic loads/stores (not lqarx/stqcx). (define_predicate "quad_memory_operand" (match_code "mem") { rtx addr, op0, op1; int ret; - if (!TARGET_QUAD_MEMORY) + if (!TARGET_QUAD_MEMORY && !TARGET_SYNC_TI) ret = 0; else if (!memory_operand (op, mode)) diff --git a/gcc/config/rs6000/rs6000.md b/gcc/config/rs6000/rs6000.md index 937eabf3727..f6da9b3a382 100644 --- a/gcc/config/rs6000/rs6000.md +++ b/gcc/config/rs6000/rs6000.md @@ -134,6 +134,7 @@ UNSPEC_DIVEUO UNSPEC_UNPACK_128BIT UNSPEC_PACK_128BIT + UNSPEC_LSQ ]) ;; diff --git a/gcc/config/rs6000/sync.md b/gcc/config/rs6000/sync.md index 7db439074cd..63152ed04d2 100644 --- a/gcc/config/rs6000/sync.md +++ b/gcc/config/rs6000/sync.md @@ -107,10 +107,17 @@ "isync" [(set_attr "type" "isync")]) +;; Types that we should provide atomic instructions for. +(define_mode_iterator AINT [QI + HI + SI + (DI "TARGET_POWERPC64") + (TI "TARGET_SYNC_TI")]) + ;; The control dependency used for load dependency described ;; in B.2.3 of the Power ISA 2.06B. (define_insn "loadsync_<mode>" - [(unspec_volatile:BLK [(match_operand:INT1 0 "register_operand" "r")] + [(unspec_volatile:BLK [(match_operand:AINT 0 "register_operand" "r")] UNSPECV_ISYNC) (clobber (match_scratch:CC 1 "=y"))] "" @@ -118,18 +125,56 @@ [(set_attr "type" "isync") (set_attr "length" "12")]) +(define_insn "load_quadpti" + [(set (match_operand:PTI 0 "quad_int_reg_operand" "=&r") + (unspec:PTI + [(match_operand:TI 1 "quad_memory_operand" "wQ")] UNSPEC_LSQ))] + "TARGET_SYNC_TI + && !reg_mentioned_p (operands[0], operands[1])" + "lq %0,%1" + [(set_attr "type" "load") + (set_attr "length" "4")]) + (define_expand "atomic_load<mode>" - [(set (match_operand:INT1 0 "register_operand" "") ;; output - (match_operand:INT1 1 "memory_operand" "")) ;; memory + [(set (match_operand:AINT 0 "register_operand" "") ;; output + (match_operand:AINT 1 "memory_operand" "")) ;; memory (use (match_operand:SI 2 "const_int_operand" ""))] ;; model "" { + if (<MODE>mode == TImode && !TARGET_SYNC_TI) + FAIL; + enum memmodel model = (enum memmodel) INTVAL (operands[2]); if (model == MEMMODEL_SEQ_CST) emit_insn (gen_hwsync ()); - emit_move_insn (operands[0], operands[1]); + if (<MODE>mode != TImode) + emit_move_insn (operands[0], operands[1]); + else + { + rtx op0 = operands[0]; + rtx op1 = operands[1]; + rtx pti_reg = gen_reg_rtx (PTImode); + + // Can't have indexed address for 'lq' + if (indexed_address (XEXP (op1, 0), TImode)) + { + rtx old_addr = XEXP (op1, 0); + rtx new_addr = force_reg (Pmode, old_addr); + operands[1] = op1 = replace_equiv_address (op1, new_addr); + } + + emit_insn (gen_load_quadpti (pti_reg, op1)); + + if (WORDS_BIG_ENDIAN) + emit_move_insn (op0, gen_lowpart (TImode, pti_reg)); + else + { + emit_move_insn (gen_lowpart (DImode, op0), gen_highpart (DImode, pti_reg)); + emit_move_insn (gen_highpart (DImode, op0), gen_lowpart (DImode, pti_reg)); + } + } switch (model) { @@ -146,12 +191,24 @@ DONE; }) +(define_insn "store_quadpti" + [(set (match_operand:PTI 0 "quad_memory_operand" "=wQ") + (unspec:PTI + [(match_operand:PTI 1 "quad_int_reg_operand" "r")] UNSPEC_LSQ))] + "TARGET_SYNC_TI" + "stq %1,%0" + [(set_attr "type" "store") + (set_attr "length" "4")]) + (define_expand "atomic_store<mode>" - [(set (match_operand:INT1 0 "memory_operand" "") ;; memory - (match_operand:INT1 1 "register_operand" "")) ;; input + [(set (match_operand:AINT 0 "memory_operand" "") ;; memory + (match_operand:AINT 1 "register_operand" "")) ;; input (use (match_operand:SI 2 "const_int_operand" ""))] ;; model "" { + if (<MODE>mode == TImode && !TARGET_SYNC_TI) + FAIL; + enum memmodel model = (enum memmodel) INTVAL (operands[2]); switch (model) { @@ -166,7 +223,33 @@ default: gcc_unreachable (); } - emit_move_insn (operands[0], operands[1]); + if (<MODE>mode != TImode) + emit_move_insn (operands[0], operands[1]); + else + { + rtx op0 = operands[0]; + rtx op1 = operands[1]; + rtx pti_reg = gen_reg_rtx (PTImode); + + // Can't have indexed address for 'stq' + if (indexed_address (XEXP (op0, 0), TImode)) + { + rtx old_addr = XEXP (op0, 0); + rtx new_addr = force_reg (Pmode, old_addr); + operands[0] = op0 = replace_equiv_address (op0, new_addr); + } + + if (WORDS_BIG_ENDIAN) + emit_move_insn (pti_reg, gen_lowpart (PTImode, op1)); + else + { + emit_move_insn (gen_lowpart (DImode, pti_reg), gen_highpart (DImode, op1)); + emit_move_insn (gen_highpart (DImode, pti_reg), gen_lowpart (DImode, op1)); + } + + emit_insn (gen_store_quadpti (gen_lowpart (PTImode, op0), pti_reg)); + } + DONE; }) @@ -180,14 +263,6 @@ SI (DI "TARGET_POWERPC64")]) -;; Types that we should provide atomic instructions for. - -(define_mode_iterator AINT [QI - HI - SI - (DI "TARGET_POWERPC64") - (TI "TARGET_SYNC_TI")]) - (define_insn "load_locked<mode>" [(set (match_operand:ATOMIC 0 "int_reg_operand" "=r") (unspec_volatile:ATOMIC diff --git a/gcc/config/sol2-10.h b/gcc/config/sol2-10.h deleted file mode 100644 index 4488a40cba6..00000000000 --- a/gcc/config/sol2-10.h +++ /dev/null @@ -1,24 +0,0 @@ -/* Operating system specific defines to be used when targeting GCC for any - Solaris 2 system starting from Solaris 10. - Copyright (C) 2006-2014 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/>. */ - -/* Solaris 10 has the float and long double forms of math functions. - We redefine this hook so the version from elfos.h header won't be used. */ -#undef TARGET_LIBC_HAS_FUNCTION -#define TARGET_LIBC_HAS_FUNCTION default_libc_has_function diff --git a/gcc/config/sol2-bi.h b/gcc/config/sol2-bi.h deleted file mode 100644 index fdb2a28178c..00000000000 --- a/gcc/config/sol2-bi.h +++ /dev/null @@ -1,135 +0,0 @@ -/* Definitions of target machine for GCC, for bi-arch Solaris 2. - Copyright (C) 2011-2014 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. - - Under Section 7 of GPL version 3, you are granted additional - permissions described in the GCC Runtime Library Exception, version - 3.1, as published by the Free Software Foundation. - - You should have received a copy of the GNU General Public License and - a copy of the GCC Runtime Library Exception along with this program; - see the files COPYING3 and COPYING.RUNTIME respectively. If not, see - <http://www.gnu.org/licenses/>. */ - -/* wchar_t is called differently in <wchar.h> for 32 and 64-bit - compilations. This is called for by SCD 2.4.1, p. 6-83, Figure 6-65 - (32-bit) and p. 6P-10, Figure 6.38 (64-bit). */ - -#undef WCHAR_TYPE -#define WCHAR_TYPE (TARGET_64BIT ? "int" : "long int") - -#undef WCHAR_TYPE_SIZE -#define WCHAR_TYPE_SIZE 32 - -/* Same for wint_t. See SCD 2.4.1, p. 6-83, Figure 6-66 (32-bit). There's - no corresponding 64-bit definition, but this is what Solaris 8 - <iso/wchar_iso.h> uses. */ - -#undef WINT_TYPE -#define WINT_TYPE (TARGET_64BIT ? "int" : "long int") - -#undef WINT_TYPE_SIZE -#define WINT_TYPE_SIZE 32 - -#if DEFAULT_ARCH32_P -#define MULTILIB_DEFAULTS { "m32" } -#else -#define MULTILIB_DEFAULTS { "m64" } -#endif - -#if DEFAULT_ARCH32_P -#define DEF_ARCH32_SPEC(__str) "%{!m64:" __str "}" -#define DEF_ARCH64_SPEC(__str) "%{m64:" __str "}" -#else -#define DEF_ARCH32_SPEC(__str) "%{m32:" __str "}" -#define DEF_ARCH64_SPEC(__str) "%{!m32:" __str "}" -#endif - -#undef ASM_CPU_DEFAULT_SPEC -#define ASM_CPU_DEFAULT_SPEC \ -(DEFAULT_ARCH32_P ? "\ -%{m64:" ASM_CPU64_DEFAULT_SPEC "} \ -%{!m64:" ASM_CPU32_DEFAULT_SPEC "} \ -" : "\ -%{m32:" ASM_CPU32_DEFAULT_SPEC "} \ -%{!m32:" ASM_CPU64_DEFAULT_SPEC "} \ -") - -/* This should be the same as LINK_ARCH32_SPEC_BASE, except with - ARCH64_SUBDIR appended to the paths and /usr/ccs/lib is no longer - necessary. */ -#undef LINK_ARCH64_SPEC_BASE -#define LINK_ARCH64_SPEC_BASE \ - "%{G:-G} \ - %{YP,*} \ - %{R*} \ - %{!YP,*:%{p|pg:-Y P,%R/usr/lib/libp/" ARCH64_SUBDIR ":%R/lib/" ARCH64_SUBDIR ":%R/usr/lib/" ARCH64_SUBDIR "} \ - %{!p:%{!pg:-Y P,%R/lib/" ARCH64_SUBDIR ":%R/usr/lib/" ARCH64_SUBDIR "}}}" - -#undef LINK_ARCH64_SPEC -#ifndef USE_GLD -/* FIXME: Used to be SPARC-only. Not SPARC-specfic but for the model name! */ -#define LINK_ARCH64_SPEC \ - "%{mcmodel=medlow:-M /usr/lib/ld/" ARCH64_SUBDIR "/map.below4G} " \ - LINK_ARCH64_SPEC_BASE -#else -#define LINK_ARCH64_SPEC LINK_ARCH64_SPEC_BASE -#endif - -#ifdef USE_GLD -#if DEFAULT_ARCH32_P -#define ARCH_DEFAULT_EMULATION ARCH32_EMULATION -#else -#define ARCH_DEFAULT_EMULATION ARCH64_EMULATION -#endif -#define TARGET_LD_EMULATION "%{m32:-m " ARCH32_EMULATION "}" \ - "%{m64:-m " ARCH64_EMULATION "}" \ - "%{!m32:%{!m64:-m " ARCH_DEFAULT_EMULATION "}} " -#else -#define TARGET_LD_EMULATION "" -#endif - -#undef LINK_ARCH_SPEC -#if DISABLE_MULTILIB -#if DEFAULT_ARCH32_P -#define LINK_ARCH_SPEC TARGET_LD_EMULATION " \ -%{m32:%(link_arch32)} \ -%{m64:%edoes not support multilib} \ -%{!m32:%{!m64:%(link_arch_default)}} \ -" -#else -#define LINK_ARCH_SPEC TARGET_LD_EMULATION " \ -%{m32:%edoes not support multilib} \ -%{m64:%(link_arch64)} \ -%{!m32:%{!m64:%(link_arch_default)}} \ -" -#endif -#else -#define LINK_ARCH_SPEC TARGET_LD_EMULATION " \ -%{m32:%(link_arch32)} \ -%{m64:%(link_arch64)} \ -%{!m32:%{!m64:%(link_arch_default)}}" -#endif - -#define LINK_ARCH_DEFAULT_SPEC \ -(DEFAULT_ARCH32_P ? LINK_ARCH32_SPEC : LINK_ARCH64_SPEC) - -#undef SUBTARGET_EXTRA_SPECS -#define SUBTARGET_EXTRA_SPECS \ - { "startfile_arch", STARTFILE_ARCH_SPEC }, \ - { "link_arch32", LINK_ARCH32_SPEC }, \ - { "link_arch64", LINK_ARCH64_SPEC }, \ - { "link_arch_default", LINK_ARCH_DEFAULT_SPEC }, \ - { "link_arch", LINK_ARCH_SPEC }, \ - SUBTARGET_CPU_EXTRA_SPECS diff --git a/gcc/config/sol2.h b/gcc/config/sol2.h index a21c953b035..d88de37bd2e 100644 --- a/gcc/config/sol2.h +++ b/gcc/config/sol2.h @@ -21,20 +21,25 @@ along with GCC; see the file COPYING3. If not see /* We are compiling for Solaris 2 now. */ #define TARGET_SOLARIS 1 -/* Solaris 2 (at least as of 2.5.1) uses a 32-bit wchar_t. */ +/* wchar_t is called differently in <wchar.h> for 32 and 64-bit + compilations. This is called for by SCD 2.4.1, p. 6-83, Figure 6-65 + (32-bit) and p. 6P-10, Figure 6.38 (64-bit). */ + #undef WCHAR_TYPE -#define WCHAR_TYPE "long int" +#define WCHAR_TYPE (TARGET_64BIT ? "int" : "long int") #undef WCHAR_TYPE_SIZE -#define WCHAR_TYPE_SIZE BITS_PER_WORD +#define WCHAR_TYPE_SIZE 32 + +/* Same for wint_t. See SCD 2.4.1, p. 6-83, Figure 6-66 (32-bit). There's + no corresponding 64-bit definition, but this is what Solaris 8 + <iso/wchar_iso.h> uses. */ -/* Solaris 2 uses a wint_t different from the default. This is required - by the SCD 2.4.1, p. 6-83, Figure 6-66. */ -#undef WINT_TYPE -#define WINT_TYPE "long int" +#undef WINT_TYPE +#define WINT_TYPE (TARGET_64BIT ? "int" : "long int") -#undef WINT_TYPE_SIZE -#define WINT_TYPE_SIZE BITS_PER_WORD +#undef WINT_TYPE_SIZE +#define WINT_TYPE_SIZE 32 #define SIG_ATOMIC_TYPE "int" @@ -103,6 +108,19 @@ along with GCC; see the file COPYING3. If not see solaris_override_options (); \ } while (0) +#if DEFAULT_ARCH32_P +#define MULTILIB_DEFAULTS { "m32" } +#else +#define MULTILIB_DEFAULTS { "m64" } +#endif + +#if DEFAULT_ARCH32_P +#define DEF_ARCH32_SPEC(__str) "%{!m64:" __str "}" +#define DEF_ARCH64_SPEC(__str) "%{m64:" __str "}" +#else +#define DEF_ARCH32_SPEC(__str) "%{m32:" __str "}" +#define DEF_ARCH64_SPEC(__str) "%{!m32:" __str "}" +#endif /* It's safe to pass -s always, even if -g is not used. Those options are handled by both Sun as and GNU as. */ @@ -111,6 +129,16 @@ along with GCC; see the file COPYING3. If not see #define ASM_PIC_SPEC " %{fpic|fpie|fPIC|fPIE:-K PIC}" +#undef ASM_CPU_DEFAULT_SPEC +#define ASM_CPU_DEFAULT_SPEC \ +(DEFAULT_ARCH32_P ? "\ +%{m64:" ASM_CPU64_DEFAULT_SPEC "} \ +%{!m64:" ASM_CPU32_DEFAULT_SPEC "} \ +" : "\ +%{m32:" ASM_CPU32_DEFAULT_SPEC "} \ +%{!m32:" ASM_CPU64_DEFAULT_SPEC "} \ +") + #undef LIB_SPEC #define LIB_SPEC \ "%{!symbolic:\ @@ -120,17 +148,11 @@ along with GCC; see the file COPYING3. If not see #ifndef CROSS_DIRECTORY_STRUCTURE #undef MD_EXEC_PREFIX #define MD_EXEC_PREFIX "/usr/ccs/bin/" - -#undef MD_STARTFILE_PREFIX -#define MD_STARTFILE_PREFIX "/usr/ccs/lib/" #endif -#undef STARTFILE_ARCH32_SPEC -#define STARTFILE_ARCH32_SPEC "%{ansi:values-Xc.o%s} \ - %{!ansi:values-Xa.o%s}" - #undef STARTFILE_ARCH_SPEC -#define STARTFILE_ARCH_SPEC STARTFILE_ARCH32_SPEC +#define STARTFILE_ARCH_SPEC "%{ansi:values-Xc.o%s} \ + %{!ansi:values-Xa.o%s}" /* We don't use the standard svr4 STARTFILE_SPEC because it's wrong for us. */ #undef STARTFILE_SPEC @@ -153,14 +175,78 @@ along with GCC; see the file COPYING3. If not see "%{G:-G} \ %{YP,*} \ %{R*} \ - %{!YP,*:%{p|pg:-Y P,%R/usr/ccs/lib/libp:%R/usr/lib/libp:%R/usr/ccs/lib:%R/lib:%R/usr/lib} \ - %{!p:%{!pg:-Y P,%R/usr/ccs/lib:%R/lib:%R/usr/lib}}}" + %{!YP,*:%{p|pg:-Y P,%R/usr/lib/libp%R/lib:%R/usr/lib} \ + %{!p:%{!pg:-Y P,%R/lib:%R/usr/lib}}}" #undef LINK_ARCH32_SPEC #define LINK_ARCH32_SPEC LINK_ARCH32_SPEC_BASE +/* This should be the same as LINK_ARCH32_SPEC_BASE, except with + ARCH64_SUBDIR appended to the paths. */ +#undef LINK_ARCH64_SPEC_BASE +#define LINK_ARCH64_SPEC_BASE \ + "%{G:-G} \ + %{YP,*} \ + %{R*} \ + %{!YP,*:%{p|pg:-Y P,%R/usr/lib/libp/" ARCH64_SUBDIR ":%R/lib/" ARCH64_SUBDIR ":%R/usr/lib/" ARCH64_SUBDIR "} \ + %{!p:%{!pg:-Y P,%R/lib/" ARCH64_SUBDIR ":%R/usr/lib/" ARCH64_SUBDIR "}}}" + +#undef LINK_ARCH64_SPEC +#ifndef USE_GLD +/* FIXME: Used to be SPARC-only. Not SPARC-specfic but for the model name! */ +#define LINK_ARCH64_SPEC \ + "%{mcmodel=medlow:-M /usr/lib/ld/" ARCH64_SUBDIR "/map.below4G} " \ + LINK_ARCH64_SPEC_BASE +#else +#define LINK_ARCH64_SPEC LINK_ARCH64_SPEC_BASE +#endif + +#ifdef USE_GLD +#if DEFAULT_ARCH32_P +#define ARCH_DEFAULT_EMULATION ARCH32_EMULATION +#else +#define ARCH_DEFAULT_EMULATION ARCH64_EMULATION +#endif +#define TARGET_LD_EMULATION "%{m32:-m " ARCH32_EMULATION "}" \ + "%{m64:-m " ARCH64_EMULATION "}" \ + "%{!m32:%{!m64:-m " ARCH_DEFAULT_EMULATION "}} " +#else +#define TARGET_LD_EMULATION "" +#endif + #undef LINK_ARCH_SPEC -#define LINK_ARCH_SPEC LINK_ARCH32_SPEC +#if DISABLE_MULTILIB +#if DEFAULT_ARCH32_P +#define LINK_ARCH_SPEC TARGET_LD_EMULATION " \ +%{m32:%(link_arch32)} \ +%{m64:%edoes not support multilib} \ +%{!m32:%{!m64:%(link_arch_default)}} \ +" +#else +#define LINK_ARCH_SPEC TARGET_LD_EMULATION " \ +%{m32:%edoes not support multilib} \ +%{m64:%(link_arch64)} \ +%{!m32:%{!m64:%(link_arch_default)}} \ +" +#endif +#else +#define LINK_ARCH_SPEC TARGET_LD_EMULATION " \ +%{m32:%(link_arch32)} \ +%{m64:%(link_arch64)} \ +%{!m32:%{!m64:%(link_arch_default)}}" +#endif + +#define LINK_ARCH_DEFAULT_SPEC \ +(DEFAULT_ARCH32_P ? LINK_ARCH32_SPEC : LINK_ARCH64_SPEC) + +#undef SUBTARGET_EXTRA_SPECS +#define SUBTARGET_EXTRA_SPECS \ + { "startfile_arch", STARTFILE_ARCH_SPEC }, \ + { "link_arch32", LINK_ARCH32_SPEC }, \ + { "link_arch64", LINK_ARCH64_SPEC }, \ + { "link_arch_default", LINK_ARCH_DEFAULT_SPEC }, \ + { "link_arch", LINK_ARCH_SPEC }, \ + SUBTARGET_CPU_EXTRA_SPECS /* C++11 programs need -lrt for nanosleep. */ #define TIME_LIBRARY "rt" @@ -235,7 +321,9 @@ along with GCC; see the file COPYING3. If not see #define TARGET_CXX_DECL_MANGLING_CONTEXT solaris_cxx_decl_mangling_context /* Solaris/x86 as and gas support unquoted section names. */ +#ifndef SECTION_NAME_FORMAT #define SECTION_NAME_FORMAT "%s" +#endif /* This is how to declare the size of a function. For Solaris, we output any .init or .fini entries here. */ @@ -249,23 +337,6 @@ along with GCC; see the file COPYING3. If not see } \ while (0) -/* Solaris as has a bug: a .common directive in .tbss or .tdata section - behaves as .tls_common rather than normal non-TLS .common. */ -#undef ASM_OUTPUT_ALIGNED_COMMON -#define ASM_OUTPUT_ALIGNED_COMMON(FILE, NAME, SIZE, ALIGN) \ - do \ - { \ - if (TARGET_SUN_TLS \ - && in_section \ - && ((in_section->common.flags & SECTION_TLS) == SECTION_TLS)) \ - switch_to_section (bss_section); \ - fprintf ((FILE), "%s", COMMON_ASM_OP); \ - assemble_name ((FILE), (NAME)); \ - fprintf ((FILE), ","HOST_WIDE_INT_PRINT_UNSIGNED",%u\n", \ - (SIZE), (ALIGN) / BITS_PER_UNIT); \ - } \ - while (0) - #ifndef USE_GAS #undef TARGET_ASM_ASSEMBLE_VISIBILITY #define TARGET_ASM_ASSEMBLE_VISIBILITY solaris_assemble_visibility @@ -291,8 +362,10 @@ along with GCC; see the file COPYING3. If not see #define TARGET_POSIX_IO +/* Solaris 10 has the float and long double forms of math functions. + We redefine this hook so the version from elfos.h header won't be used. */ #undef TARGET_LIBC_HAS_FUNCTION -#define TARGET_LIBC_HAS_FUNCTION no_c99_libc_has_function +#define TARGET_LIBC_HAS_FUNCTION default_libc_has_function extern GTY(()) tree solaris_pending_aligns; extern GTY(()) tree solaris_pending_inits; diff --git a/gcc/config/sparc/sol2.h b/gcc/config/sparc/sol2.h index b50a937b26f..2fcdc85c81f 100644 --- a/gcc/config/sparc/sol2.h +++ b/gcc/config/sparc/sol2.h @@ -109,8 +109,6 @@ along with GCC; see the file COPYING3. If not see #define CPP_CPU64_DEFAULT_SPEC "" #undef ASM_CPU32_DEFAULT_SPEC #define ASM_CPU32_DEFAULT_SPEC "-xarch=v8plus" -#undef ASM_CPU_DEFAULT_SPEC -#define ASM_CPU_DEFAULT_SPEC ASM_CPU32_DEFAULT_SPEC #endif #if TARGET_CPU_DEFAULT == TARGET_CPU_ultrasparc @@ -120,8 +118,6 @@ along with GCC; see the file COPYING3. If not see #define ASM_CPU32_DEFAULT_SPEC "-xarch=v8plusa" #undef ASM_CPU64_DEFAULT_SPEC #define ASM_CPU64_DEFAULT_SPEC "-xarch=v9a" -#undef ASM_CPU_DEFAULT_SPEC -#define ASM_CPU_DEFAULT_SPEC ASM_CPU32_DEFAULT_SPEC #endif #if TARGET_CPU_DEFAULT == TARGET_CPU_ultrasparc3 @@ -131,8 +127,6 @@ along with GCC; see the file COPYING3. If not see #define ASM_CPU32_DEFAULT_SPEC "-xarch=v8plusb" #undef ASM_CPU64_DEFAULT_SPEC #define ASM_CPU64_DEFAULT_SPEC "-xarch=v9b" -#undef ASM_CPU_DEFAULT_SPEC -#define ASM_CPU_DEFAULT_SPEC ASM_CPU32_DEFAULT_SPEC #endif #if TARGET_CPU_DEFAULT == TARGET_CPU_niagara @@ -142,8 +136,6 @@ along with GCC; see the file COPYING3. If not see #define ASM_CPU32_DEFAULT_SPEC "-xarch=v8plusb" #undef ASM_CPU64_DEFAULT_SPEC #define ASM_CPU64_DEFAULT_SPEC "-xarch=v9b" -#undef ASM_CPU_DEFAULT_SPEC -#define ASM_CPU_DEFAULT_SPEC ASM_CPU32_DEFAULT_SPEC #endif #if TARGET_CPU_DEFAULT == TARGET_CPU_niagara2 @@ -153,8 +145,6 @@ along with GCC; see the file COPYING3. If not see #define ASM_CPU32_DEFAULT_SPEC "-xarch=v8plusb" #undef ASM_CPU64_DEFAULT_SPEC #define ASM_CPU64_DEFAULT_SPEC "-xarch=v9b" -#undef ASM_CPU_DEFAULT_SPEC -#define ASM_CPU_DEFAULT_SPEC ASM_CPU32_DEFAULT_SPEC #endif #if TARGET_CPU_DEFAULT == TARGET_CPU_niagara3 @@ -164,8 +154,6 @@ along with GCC; see the file COPYING3. If not see #define ASM_CPU32_DEFAULT_SPEC "-xarch=v8plus" AS_NIAGARA3_FLAG #undef ASM_CPU64_DEFAULT_SPEC #define ASM_CPU64_DEFAULT_SPEC "-xarch=v9" AS_NIAGARA3_FLAG -#undef ASM_CPU_DEFAULT_SPEC -#define ASM_CPU_DEFAULT_SPEC ASM_CPU32_DEFAULT_SPEC #endif #if TARGET_CPU_DEFAULT == TARGET_CPU_niagara4 @@ -175,8 +163,6 @@ along with GCC; see the file COPYING3. If not see #define ASM_CPU32_DEFAULT_SPEC AS_SPARC32_FLAG AS_NIAGARA4_FLAG #undef ASM_CPU64_DEFAULT_SPEC #define ASM_CPU64_DEFAULT_SPEC AS_SPARC64_FLAG AS_NIAGARA4_FLAG -#undef ASM_CPU_DEFAULT_SPEC -#define ASM_CPU_DEFAULT_SPEC ASM_CPU32_DEFAULT_SPEC #endif #undef CPP_CPU_SPEC @@ -361,6 +347,23 @@ extern const char *host_detect_local_cpu (int argc, const char **argv); } \ while (0) +/* Solaris as has a bug: a .common directive in .tbss or .tdata section + behaves as .tls_common rather than normal non-TLS .common. */ +#undef ASM_OUTPUT_ALIGNED_COMMON +#define ASM_OUTPUT_ALIGNED_COMMON(FILE, NAME, SIZE, ALIGN) \ + do \ + { \ + if (TARGET_SUN_TLS \ + && in_section \ + && ((in_section->common.flags & SECTION_TLS) == SECTION_TLS)) \ + switch_to_section (bss_section); \ + fprintf ((FILE), "%s", COMMON_ASM_OP); \ + assemble_name ((FILE), (NAME)); \ + fprintf ((FILE), ","HOST_WIDE_INT_PRINT_UNSIGNED",%u\n", \ + (SIZE), (ALIGN) / BITS_PER_UNIT); \ + } \ + while (0) + #ifndef USE_GAS /* This is how to output an assembler line that says to advance the location counter to a multiple of 2**LOG bytes using the @@ -376,7 +379,6 @@ extern const char *host_detect_local_cpu (int argc, const char **argv); /* Sun as requires doublequoted section names on SPARC. While GNU as supports that, too, we prefer the standard variant. */ -#undef SECTION_NAME_FORMAT #define SECTION_NAME_FORMAT "\"%s\"" #endif /* !USE_GAS */ diff --git a/gcc/config/sparc/t-sol2-64 b/gcc/config/sparc/t-sol2 index ec7e4eba6fd..ec7e4eba6fd 100644 --- a/gcc/config/sparc/t-sol2-64 +++ b/gcc/config/sparc/t-sol2 diff --git a/gcc/coretypes.h b/gcc/coretypes.h index 58f9af0ec98..41ad1a3b22d 100644 --- a/gcc/coretypes.h +++ b/gcc/coretypes.h @@ -197,7 +197,8 @@ enum function_class { function_c94, function_c99_misc, function_c99_math_complex, - function_sincos + function_sincos, + function_c11_misc }; /* Memory model types for the __atomic* builtins. diff --git a/gcc/cp/ChangeLog b/gcc/cp/ChangeLog index a5f3829d705..c078e0685cc 100644 --- a/gcc/cp/ChangeLog +++ b/gcc/cp/ChangeLog @@ -1,3 +1,47 @@ +2014-04-30 Jason Merrill <jason@redhat.com> + + PR c++/60980 + * init.c (build_value_init): Don't try to call an array constructor. + + PR c++/60951 + * typeck2.c (massage_init_elt): Use maybe_constant_init. + +2014-04-30 Marek Polacek <polacek@redhat.com> + + * typeck.c (cp_build_binary_op): Call ubsan_instrument_division + even when SANITIZE_FLOAT_DIVIDE is on. Set doing_div_or_mod even + for non-integer types. + +2014-04-29 Jason Merrill <jason@redhat.com> + + DR 1351 + Represent the unevaluated exception specification of an implicitly + declared or deleted function with a simple placeholder, not a list + of functions. + * cp-tree.h (UNEVALUATED_NOEXCEPT_SPEC_P): New. + * except.c (unevaluated_noexcept_spec): New. + * class.c (deduce_noexcept_on_destructor): Use it. + * decl.c (check_redeclaration_exception_specification): Call + maybe_instantiate_noexcept. + (duplicate_decls): Call it before merge_types. + (start_preparsed_function): Call maybe_instantiate_noexcept. + * decl2.c (mark_used): Call maybe_instantiate_noexcept earlier. + * init.c (get_nsdmi): Factor out of perform_member_init. + * method.c (process_subob_fn): Call maybe_instantiate_noexcept. + (walk_field_subobs): Consider NSDMI for EH spec. + (get_defaulted_eh_spec): New. + (implicitly_declare_fn): Use unevaluated_noexcept_spec. + (defaulted_late_check): Defer EH checking in non-template classes. + (after_nsdmi_defaulted_late_checks): New. + * parser.c (cp_parser_class_specifier_1): Use it. + (unparsed_classes): New macro. + * parser.h (cp_unparsed_functions_entry_d): Add classes field. + * pt.c (maybe_instantiate_noexcept): Use get_defaulted_eh_spec. + Remove list-of-functions handling. + * typeck2.c (merge_exception_specifiers): Remove list-of-functions + handling and FN parameter. + * typeck.c (merge_types): Adjust. + 2014-04-28 Paolo Carlini <paolo.carlini@oracle.com> PR c++/59120 diff --git a/gcc/cp/class.c b/gcc/cp/class.c index 67413761d8c..c96d79dbc82 100644 --- a/gcc/cp/class.c +++ b/gcc/cp/class.c @@ -4726,11 +4726,7 @@ deduce_noexcept_on_destructor (tree dtor) { if (!TYPE_RAISES_EXCEPTIONS (TREE_TYPE (dtor))) { - tree ctx = DECL_CONTEXT (dtor); - tree implicit_fn = implicitly_declare_fn (sfk_destructor, ctx, - /*const_p=*/false, - NULL, NULL); - tree eh_spec = TYPE_RAISES_EXCEPTIONS (TREE_TYPE (implicit_fn)); + tree eh_spec = unevaluated_noexcept_spec (); TREE_TYPE (dtor) = build_exception_variant (TREE_TYPE (dtor), eh_spec); } } diff --git a/gcc/cp/cp-tree.def b/gcc/cp/cp-tree.def index 057e7ea5ef7..b4a72d6e30a 100644 --- a/gcc/cp/cp-tree.def +++ b/gcc/cp/cp-tree.def @@ -212,9 +212,12 @@ DEFTREECODE (USING_STMT, "using_stmt", tcc_statement, 1) parsing had occurred. */ DEFTREECODE (DEFAULT_ARG, "default_arg", tcc_exceptional, 0) -/* An uninstantiated noexcept-specification. DEFERRED_NOEXCEPT_PATTERN is - the pattern from the template, and DEFERRED_NOEXCEPT_ARGS are the - template arguments to substitute into the pattern when needed. */ +/* An uninstantiated/unevaluated noexcept-specification. For the + uninstantiated case, DEFERRED_NOEXCEPT_PATTERN is the pattern from the + template, and DEFERRED_NOEXCEPT_ARGS are the template arguments to + substitute into the pattern when needed. For the unevaluated case, + those slots are NULL_TREE and we use get_defaulted_eh_spec to find + the exception-specification. */ DEFTREECODE (DEFERRED_NOEXCEPT, "deferred_noexcept", tcc_exceptional, 0) /* A template-id, like foo<int>. The first operand is the template. diff --git a/gcc/cp/cp-tree.h b/gcc/cp/cp-tree.h index f459e55bc84..55ecc4e5b5f 100644 --- a/gcc/cp/cp-tree.h +++ b/gcc/cp/cp-tree.h @@ -579,8 +579,10 @@ struct GTY (()) tree_default_arg { (((struct tree_deferred_noexcept *)DEFERRED_NOEXCEPT_CHECK (NODE))->args) #define DEFERRED_NOEXCEPT_SPEC_P(NODE) \ ((NODE) && (TREE_PURPOSE (NODE)) \ - && (TREE_CODE (TREE_PURPOSE (NODE)) == DEFERRED_NOEXCEPT \ - || is_overloaded_fn (TREE_PURPOSE (NODE)))) + && (TREE_CODE (TREE_PURPOSE (NODE)) == DEFERRED_NOEXCEPT)) +#define UNEVALUATED_NOEXCEPT_SPEC_P(NODE) \ + (DEFERRED_NOEXCEPT_SPEC_P (NODE) \ + && DEFERRED_NOEXCEPT_PATTERN (TREE_PURPOSE (NODE)) == NULL_TREE) struct GTY (()) tree_deferred_noexcept { struct tree_base base; @@ -4359,8 +4361,6 @@ extern int comparing_specializations; sizeof can be nested. */ extern int cp_unevaluated_operand; -extern tree cp_convert_range_for (tree, tree, tree, bool); -extern bool parsing_nsdmi (void); /* in pt.c */ @@ -5420,6 +5420,7 @@ extern tree get_type_value (tree); extern tree build_zero_init (tree, tree, bool); extern tree build_value_init (tree, tsubst_flags_t); extern tree build_value_init_noctor (tree, tsubst_flags_t); +extern tree get_nsdmi (tree, bool); extern tree build_offset_ref (tree, tree, bool, tsubst_flags_t); extern tree throw_bad_array_new_length (void); @@ -5468,6 +5469,9 @@ extern tree make_thunk (tree, bool, tree, tree); extern void finish_thunk (tree); extern void use_thunk (tree, bool); extern bool trivial_fn_p (tree); +extern tree get_defaulted_eh_spec (tree); +extern tree unevaluated_noexcept_spec (void); +extern void after_nsdmi_defaulted_late_checks (tree); extern bool maybe_explain_implicit_delete (tree); extern void explain_implicit_non_constexpr (tree); extern void deduce_inheriting_ctor (tree); @@ -5489,6 +5493,11 @@ extern tree implicitly_declare_fn (special_function_kind, tree, /* In optimize.c */ extern bool maybe_clone_body (tree); +/* In parser.c */ +extern tree cp_convert_range_for (tree, tree, tree, bool); +extern bool parsing_nsdmi (void); +extern void inject_this_parameter (tree, cp_cv_quals); + /* in pt.c */ extern bool check_template_shadow (tree); extern tree get_innermost_template_args (tree, int); @@ -6162,7 +6171,7 @@ extern tree build_x_arrow (location_t, tree, extern tree build_m_component_ref (tree, tree, tsubst_flags_t); extern tree build_functional_cast (tree, tree, tsubst_flags_t); extern tree add_exception_specifier (tree, tree, int); -extern tree merge_exception_specifiers (tree, tree, tree); +extern tree merge_exception_specifiers (tree, tree); /* in mangle.c */ extern void init_mangle (void); diff --git a/gcc/cp/decl.c b/gcc/cp/decl.c index 833570b1e72..bfd5395128a 100644 --- a/gcc/cp/decl.c +++ b/gcc/cp/decl.c @@ -1168,15 +1168,18 @@ static void check_redeclaration_exception_specification (tree new_decl, tree old_decl) { - tree new_type; - tree old_type; - tree new_exceptions; - tree old_exceptions; + tree new_exceptions = TYPE_RAISES_EXCEPTIONS (TREE_TYPE (new_decl)); + tree old_exceptions = TYPE_RAISES_EXCEPTIONS (TREE_TYPE (old_decl)); - new_type = TREE_TYPE (new_decl); - new_exceptions = TYPE_RAISES_EXCEPTIONS (new_type); - old_type = TREE_TYPE (old_decl); - old_exceptions = TYPE_RAISES_EXCEPTIONS (old_type); + /* Two default specs are equivalent, don't force evaluation. */ + if (UNEVALUATED_NOEXCEPT_SPEC_P (new_exceptions) + && UNEVALUATED_NOEXCEPT_SPEC_P (old_exceptions)) + return; + + maybe_instantiate_noexcept (new_decl); + maybe_instantiate_noexcept (old_decl); + new_exceptions = TYPE_RAISES_EXCEPTIONS (TREE_TYPE (new_decl)); + old_exceptions = TYPE_RAISES_EXCEPTIONS (TREE_TYPE (old_decl)); /* [except.spec] @@ -1916,13 +1919,13 @@ duplicate_decls (tree newdecl, tree olddecl, bool newdecl_is_friend) if (types_match) { + if (TREE_CODE (newdecl) == FUNCTION_DECL) + check_redeclaration_exception_specification (newdecl, olddecl); + /* Automatically handles default parameters. */ tree oldtype = TREE_TYPE (olddecl); tree newtype; - if (TREE_CODE (newdecl) == FUNCTION_DECL) - maybe_instantiate_noexcept (olddecl); - /* For typedefs use the old type, as the new type's DECL_NAME points at newdecl, which will be ggc_freed. */ if (TREE_CODE (newdecl) == TYPE_DECL) @@ -1953,10 +1956,6 @@ duplicate_decls (tree newdecl, tree olddecl, bool newdecl_is_friend) } } - /* Do this after calling `merge_types' so that default - parameters don't confuse us. */ - else if (TREE_CODE (newdecl) == FUNCTION_DECL) - check_redeclaration_exception_specification (newdecl, olddecl); TREE_TYPE (newdecl) = TREE_TYPE (olddecl) = newtype; if (TREE_CODE (newdecl) == FUNCTION_DECL) @@ -13435,6 +13434,9 @@ start_preparsed_function (tree decl1, tree attrs, int flags) if (!DECL_CLONED_FUNCTION_P (decl1)) determine_visibility (decl1); + if (!processing_template_decl) + maybe_instantiate_noexcept (decl1); + begin_scope (sk_function_parms, decl1); ++function_depth; diff --git a/gcc/cp/decl2.c b/gcc/cp/decl2.c index 8a7a8369f19..918ea2fc6d0 100644 --- a/gcc/cp/decl2.c +++ b/gcc/cp/decl2.c @@ -4806,6 +4806,9 @@ mark_used (tree decl, tsubst_flags_t complain) if (TREE_CODE (decl) == CONST_DECL) used_types_insert (DECL_CONTEXT (decl)); + if (TREE_CODE (decl) == FUNCTION_DECL) + maybe_instantiate_noexcept (decl); + if (TREE_CODE (decl) == FUNCTION_DECL && DECL_DELETED_FN (decl)) { @@ -4860,9 +4863,6 @@ mark_used (tree decl, tsubst_flags_t complain) return true; } - if (TREE_CODE (decl) == FUNCTION_DECL) - maybe_instantiate_noexcept (decl); - /* Normally, we can wait until instantiation-time to synthesize DECL. However, if DECL is a static data member initialized with a constant or a constexpr function, we need it right now because a reference to diff --git a/gcc/cp/except.c b/gcc/cp/except.c index 221971ac956..ead889c0658 100644 --- a/gcc/cp/except.c +++ b/gcc/cp/except.c @@ -1342,6 +1342,18 @@ build_noexcept_spec (tree expr, int complain) } } +/* Returns a noexcept-specifier to be evaluated later, for an + implicitly-declared or explicitly defaulted special member function. */ + +tree +unevaluated_noexcept_spec (void) +{ + static tree spec; + if (spec == NULL_TREE) + spec = build_noexcept_spec (make_node (DEFERRED_NOEXCEPT), tf_none); + return spec; +} + /* Returns a TRY_CATCH_EXPR that will put TRY_LIST and CATCH_LIST in the TRY and CATCH locations. CATCH_LIST must be a STATEMENT_LIST */ diff --git a/gcc/cp/init.c b/gcc/cp/init.c index 045044e4737..6838d2aadd0 100644 --- a/gcc/cp/init.c +++ b/gcc/cp/init.c @@ -340,7 +340,8 @@ build_value_init (tree type, tsubst_flags_t complain) gcc_assert (!processing_template_decl || (SCALAR_TYPE_P (type) || TREE_CODE (type) == ARRAY_TYPE)); - if (type_build_ctor_call (type)) + if (CLASS_TYPE_P (type) + && type_build_ctor_call (type)) { tree ctor = build_aggr_init_expr (type, @@ -522,6 +523,45 @@ perform_target_ctor (tree init) } } +/* Return the non-static data initializer for FIELD_DECL MEMBER. */ + +tree +get_nsdmi (tree member, bool in_ctor) +{ + tree init; + tree save_ccp = current_class_ptr; + tree save_ccr = current_class_ref; + if (!in_ctor) + inject_this_parameter (DECL_CONTEXT (member), TYPE_UNQUALIFIED); + if (DECL_LANG_SPECIFIC (member) && DECL_TEMPLATE_INFO (member)) + /* Do deferred instantiation of the NSDMI. */ + init = (tsubst_copy_and_build + (DECL_INITIAL (DECL_TI_TEMPLATE (member)), + DECL_TI_ARGS (member), + tf_warning_or_error, member, /*function_p=*/false, + /*integral_constant_expression_p=*/false)); + else + { + init = DECL_INITIAL (member); + if (init && TREE_CODE (init) == DEFAULT_ARG) + { + error ("constructor required before non-static data member " + "for %qD has been parsed", member); + DECL_INITIAL (member) = error_mark_node; + init = NULL_TREE; + } + /* Strip redundant TARGET_EXPR so we don't need to remap it, and + so the aggregate init code below will see a CONSTRUCTOR. */ + if (init && TREE_CODE (init) == TARGET_EXPR + && !VOID_TYPE_P (TREE_TYPE (TARGET_EXPR_INITIAL (init)))) + init = TARGET_EXPR_INITIAL (init); + init = break_out_target_exprs (init); + } + current_class_ptr = save_ccp; + current_class_ref = save_ccr; + return init; +} + /* Initialize MEMBER, a FIELD_DECL, with INIT, a TREE_LIST of arguments. If TREE_LIST is void_type_node, an empty initializer list was given; if NULL_TREE no initializer was given. */ @@ -535,31 +575,7 @@ perform_member_init (tree member, tree init) /* Use the non-static data member initializer if there was no mem-initializer for this field. */ if (init == NULL_TREE) - { - if (DECL_LANG_SPECIFIC (member) && DECL_TEMPLATE_INFO (member)) - /* Do deferred instantiation of the NSDMI. */ - init = (tsubst_copy_and_build - (DECL_INITIAL (DECL_TI_TEMPLATE (member)), - DECL_TI_ARGS (member), - tf_warning_or_error, member, /*function_p=*/false, - /*integral_constant_expression_p=*/false)); - else - { - init = DECL_INITIAL (member); - if (init && TREE_CODE (init) == DEFAULT_ARG) - { - error ("constructor required before non-static data member " - "for %qD has been parsed", member); - init = NULL_TREE; - } - /* Strip redundant TARGET_EXPR so we don't need to remap it, and - so the aggregate init code below will see a CONSTRUCTOR. */ - if (init && TREE_CODE (init) == TARGET_EXPR - && !VOID_TYPE_P (TREE_TYPE (TARGET_EXPR_INITIAL (init)))) - init = TARGET_EXPR_INITIAL (init); - init = break_out_target_exprs (init); - } - } + init = get_nsdmi (member, /*ctor*/true); if (init == error_mark_node) return; diff --git a/gcc/cp/method.c b/gcc/cp/method.c index 11bff7f4587..f8fc01ff531 100644 --- a/gcc/cp/method.c +++ b/gcc/cp/method.c @@ -1003,8 +1003,9 @@ process_subob_fn (tree fn, tree *spec_p, bool *trivial_p, if (spec_p) { + maybe_instantiate_noexcept (fn); tree raises = TYPE_RAISES_EXCEPTIONS (TREE_TYPE (fn)); - *spec_p = merge_exception_specifiers (*spec_p, raises, fn); + *spec_p = merge_exception_specifiers (*spec_p, raises); } if (!trivial_fn_p (fn)) @@ -1090,17 +1091,14 @@ walk_field_subobs (tree fields, tree fnname, special_function_kind sfk, inform (0, "initializer for %q+#D is invalid", field); if (trivial_p) *trivial_p = false; -#if 0 /* Core 1351: If the field has an NSDMI that could throw, the - default constructor is noexcept(false). FIXME this is - broken by deferred parsing and 1360 saying we can't lazily - declare a non-trivial default constructor. Also this - needs to do deferred instantiation. Disable until the - conflict between 1351 and 1360 is resolved. */ - if (spec_p && !expr_noexcept_p (DECL_INITIAL (field), complain)) - *spec_p = noexcept_false_spec; -#endif - + default constructor is noexcept(false). */ + if (spec_p) + { + tree nsdmi = get_nsdmi (field, /*ctor*/false); + if (!expr_noexcept_p (nsdmi, complain)) + *spec_p = noexcept_false_spec; + } /* Don't do the normal processing. */ continue; } @@ -1438,6 +1436,26 @@ synthesized_method_walk (tree ctype, special_function_kind sfk, bool const_p, --c_inhibit_evaluation_warnings; } +/* DECL is a defaulted function whose exception specification is now + needed. Return what it should be. */ + +tree +get_defaulted_eh_spec (tree decl) +{ + if (DECL_CLONED_FUNCTION_P (decl)) + decl = DECL_CLONED_FUNCTION (decl); + special_function_kind sfk = special_function_p (decl); + tree ctype = DECL_CONTEXT (decl); + tree parms = FUNCTION_FIRST_USER_PARMTYPE (decl); + tree parm_type = TREE_VALUE (parms); + bool const_p = CP_TYPE_CONST_P (non_reference (parm_type)); + tree spec = empty_except_spec; + synthesized_method_walk (ctype, sfk, const_p, &spec, NULL, NULL, + NULL, false, DECL_INHERITED_CTOR_BASE (decl), + parms); + return spec; +} + /* DECL is a deleted function. If it's implicitly deleted, explain why and return true; else return false. */ @@ -1675,6 +1693,13 @@ implicitly_declare_fn (special_function_kind kind, tree type, deleted_p = DECL_DELETED_FN (inherited_ctor); constexpr_p = DECL_DECLARED_CONSTEXPR_P (inherited_ctor); } + else if (cxx_dialect >= cxx11) + { + raises = unevaluated_noexcept_spec (); + synthesized_method_walk (type, kind, const_p, NULL, &trivial_p, + &deleted_p, &constexpr_p, false, + inherited_base, inherited_parms); + } else synthesized_method_walk (type, kind, const_p, &raises, &trivial_p, &deleted_p, &constexpr_p, false, @@ -1826,25 +1851,33 @@ defaulted_late_check (tree fn) is explicitly defaulted on its first declaration, (...) it is implicitly considered to have the same exception-specification as if it had been implicitly declared. */ - if (TYPE_RAISES_EXCEPTIONS (TREE_TYPE (fn))) + tree fn_spec = TYPE_RAISES_EXCEPTIONS (TREE_TYPE (fn)); + if (!fn_spec) { - maybe_instantiate_noexcept (fn); - if (!comp_except_specs (TYPE_RAISES_EXCEPTIONS (TREE_TYPE (fn)), - eh_spec, ce_normal)) + if (DECL_DEFAULTED_IN_CLASS_P (fn)) + TREE_TYPE (fn) = build_exception_variant (TREE_TYPE (fn), eh_spec); + } + else if (UNEVALUATED_NOEXCEPT_SPEC_P (fn_spec)) + /* Equivalent to the implicit spec. */; + else if (DECL_DEFAULTED_IN_CLASS_P (fn) + && !CLASSTYPE_TEMPLATE_INSTANTIATION (ctx)) + /* We can't compare an explicit exception-specification on a + constructor defaulted in the class body to the implicit + exception-specification until after we've parsed any NSDMI; see + after_nsdmi_defaulted_late_checks. */; + else + { + tree eh_spec = get_defaulted_eh_spec (fn); + if (!comp_except_specs (fn_spec, eh_spec, ce_normal)) { if (DECL_DEFAULTED_IN_CLASS_P (fn)) - { - DECL_DELETED_FN (fn) = true; - eh_spec = TYPE_RAISES_EXCEPTIONS (TREE_TYPE (fn)); - } + DECL_DELETED_FN (fn) = true; else error ("function %q+D defaulted on its redeclaration " "with an exception-specification that differs from " - "the implicit declaration %q#D", fn, implicit_fn); + "the implicit exception-specification %qX", fn, eh_spec); } } - if (DECL_DEFAULTED_IN_CLASS_P (fn)) - TREE_TYPE (fn) = build_exception_variant (TREE_TYPE (fn), eh_spec); if (DECL_DEFAULTED_IN_CLASS_P (fn) && DECL_DECLARED_CONSTEXPR_P (implicit_fn)) @@ -1874,6 +1907,30 @@ defaulted_late_check (tree fn) DECL_DELETED_FN (fn) = 1; } +/* OK, we've parsed the NSDMI for class T, now we can check any explicit + exception-specifications on functions defaulted in the class body. */ + +void +after_nsdmi_defaulted_late_checks (tree t) +{ + if (uses_template_parms (t)) + return; + if (t == error_mark_node) + return; + for (tree fn = TYPE_METHODS (t); fn; fn = DECL_CHAIN (fn)) + if (!DECL_ARTIFICIAL (fn) && DECL_DEFAULTED_IN_CLASS_P (fn)) + { + tree fn_spec = TYPE_RAISES_EXCEPTIONS (TREE_TYPE (fn)); + if (UNEVALUATED_NOEXCEPT_SPEC_P (fn_spec)) + continue; + + tree eh_spec = get_defaulted_eh_spec (fn); + if (!comp_except_specs (TYPE_RAISES_EXCEPTIONS (TREE_TYPE (fn)), + eh_spec, ce_normal)) + DECL_DELETED_FN (fn) = true; + } +} + /* Returns true iff FN can be explicitly defaulted, and gives any errors if defaulting FN is ill-formed. */ diff --git a/gcc/cp/parser.c b/gcc/cp/parser.c index 962cacedf80..5542dcd9ba3 100644 --- a/gcc/cp/parser.c +++ b/gcc/cp/parser.c @@ -1848,11 +1848,13 @@ cp_parser_context_new (cp_parser_context* next) parser->unparsed_queues->last ().funs_with_definitions #define unparsed_nsdmis \ parser->unparsed_queues->last ().nsdmis +#define unparsed_classes \ + parser->unparsed_queues->last ().classes static void push_unparsed_function_queues (cp_parser *parser) { - cp_unparsed_functions_entry e = {NULL, make_tree_vector (), NULL}; + cp_unparsed_functions_entry e = {NULL, make_tree_vector (), NULL, NULL}; vec_safe_push (parser->unparsed_queues, e); } @@ -17834,7 +17836,7 @@ cp_parser_virt_specifier_seq_opt (cp_parser* parser) /* Used by handling of trailing-return-types and NSDMI, in which 'this' is in scope even though it isn't real. */ -static void +void inject_this_parameter (tree ctype, cp_cv_quals quals) { tree this_parm; @@ -19505,6 +19507,13 @@ cp_parser_class_specifier_1 (cp_parser* parser) current_class_ref = save_ccr; if (pushed_scope) pop_scope (pushed_scope); + + /* Now do some post-NSDMI bookkeeping. */ + FOR_EACH_VEC_SAFE_ELT (unparsed_classes, ix, class_type) + after_nsdmi_defaulted_late_checks (class_type); + vec_safe_truncate (unparsed_classes, 0); + after_nsdmi_defaulted_late_checks (type); + /* Now parse the body of the functions. */ if (flag_openmp) { @@ -19521,6 +19530,8 @@ cp_parser_class_specifier_1 (cp_parser* parser) cp_parser_late_parsing_for_member (parser, decl); vec_safe_truncate (unparsed_funs_with_definitions, 0); } + else + vec_safe_push (unparsed_classes, type); /* Put back any saved access checks. */ pop_deferring_access_checks (); diff --git a/gcc/cp/parser.h b/gcc/cp/parser.h index 758c6df3c27..96a84534045 100644 --- a/gcc/cp/parser.h +++ b/gcc/cp/parser.h @@ -149,7 +149,7 @@ typedef struct GTY(()) cp_default_arg_entry_d { } cp_default_arg_entry; -/* An entry in a stack for member functions of local classes. */ +/* An entry in a stack for member functions defined within their classes. */ typedef struct GTY(()) cp_unparsed_functions_entry_d { /* Functions with default arguments that require post-processing. @@ -163,6 +163,10 @@ typedef struct GTY(()) cp_unparsed_functions_entry_d { /* Non-static data members with initializers that require post-processing. FIELD_DECLs appear in this list in declaration order. */ vec<tree, va_gc> *nsdmis; + + /* Nested classes go in this vector, so that we can do some final + processing after parsing any NSDMIs. */ + vec<tree, va_gc> *classes; } cp_unparsed_functions_entry; diff --git a/gcc/cp/pt.c b/gcc/cp/pt.c index c74e7ae7586..48cc2a9e9cf 100644 --- a/gcc/cp/pt.c +++ b/gcc/cp/pt.c @@ -19470,14 +19470,16 @@ maybe_instantiate_noexcept (tree fn) fntype = TREE_TYPE (fn); spec = TYPE_RAISES_EXCEPTIONS (fntype); - if (!DEFERRED_NOEXCEPT_SPEC_P (spec)) + if (!spec || !TREE_PURPOSE (spec)) return; noex = TREE_PURPOSE (spec); if (TREE_CODE (noex) == DEFERRED_NOEXCEPT) { - if (push_tinst_level (fn)) + if (DEFERRED_NOEXCEPT_PATTERN (noex) == NULL_TREE) + spec = get_defaulted_eh_spec (fn); + else if (push_tinst_level (fn)) { push_access_scope (fn); push_deferring_access_checks (dk_no_deferred); @@ -19496,24 +19498,9 @@ maybe_instantiate_noexcept (tree fn) } else spec = noexcept_false_spec; - } - else - { - /* This is an implicitly declared function, so NOEX is a list of - other functions to evaluate and merge. */ - tree elt; - spec = noexcept_true_spec; - for (elt = noex; elt; elt = OVL_NEXT (elt)) - { - tree fn = OVL_CURRENT (elt); - tree subspec; - maybe_instantiate_noexcept (fn); - subspec = TYPE_RAISES_EXCEPTIONS (TREE_TYPE (fn)); - spec = merge_exception_specifiers (spec, subspec, NULL_TREE); - } - } - TREE_TYPE (fn) = build_exception_variant (fntype, spec); + TREE_TYPE (fn) = build_exception_variant (fntype, spec); + } FOR_EACH_CLONE (clone, fn) { @@ -19664,9 +19651,6 @@ instantiate_decl (tree d, int defer_ok, SET_DECL_IMPLICIT_INSTANTIATION (d); } - if (TREE_CODE (d) == FUNCTION_DECL) - maybe_instantiate_noexcept (d); - /* Defer all other templates, unless we have been explicitly forbidden from doing so. */ if (/* If there is no definition, we cannot instantiate the diff --git a/gcc/cp/typeck.c b/gcc/cp/typeck.c index 9a80727dd8e..729e22eadc5 100644 --- a/gcc/cp/typeck.c +++ b/gcc/cp/typeck.c @@ -840,8 +840,7 @@ merge_types (tree t1, tree t2) type_memfn_quals (t1), type_memfn_rqual (t1)); raises = merge_exception_specifiers (TYPE_RAISES_EXCEPTIONS (t1), - TYPE_RAISES_EXCEPTIONS (t2), - NULL_TREE); + TYPE_RAISES_EXCEPTIONS (t2)); t1 = build_exception_variant (rval, raises); break; } @@ -852,8 +851,7 @@ merge_types (tree t1, tree t2) is just the main variant of this. */ tree basetype = class_of_this_parm (t2); tree raises = merge_exception_specifiers (TYPE_RAISES_EXCEPTIONS (t1), - TYPE_RAISES_EXCEPTIONS (t2), - NULL_TREE); + TYPE_RAISES_EXCEPTIONS (t2)); cp_ref_qualifier rqual = type_memfn_rqual (t1); tree t3; @@ -4112,10 +4110,7 @@ cp_build_binary_op (location_t location, enum tree_code tcode0 = code0, tcode1 = code1; tree cop1 = fold_non_dependent_expr_sfinae (op1, tf_none); cop1 = maybe_constant_value (cop1); - - if (tcode0 == INTEGER_TYPE) - doing_div_or_mod = true; - + doing_div_or_mod = true; warn_for_div_by_zero (location, cop1); if (tcode0 == COMPLEX_TYPE || tcode0 == VECTOR_TYPE) @@ -4155,9 +4150,7 @@ cp_build_binary_op (location_t location, { tree cop1 = fold_non_dependent_expr_sfinae (op1, tf_none); cop1 = maybe_constant_value (cop1); - - if (code0 == INTEGER_TYPE) - doing_div_or_mod = true; + doing_div_or_mod = true; warn_for_div_by_zero (location, cop1); } @@ -4904,7 +4897,8 @@ cp_build_binary_op (location_t location, if (build_type == NULL_TREE) build_type = result_type; - if ((flag_sanitize & (SANITIZE_SHIFT | SANITIZE_DIVIDE)) + if ((flag_sanitize & (SANITIZE_SHIFT | SANITIZE_DIVIDE + | SANITIZE_FLOAT_DIVIDE)) && !processing_template_decl && current_function_decl != 0 && !lookup_attribute ("no_sanitize_undefined", @@ -4918,7 +4912,8 @@ cp_build_binary_op (location_t location, tf_none)); op1 = maybe_constant_value (fold_non_dependent_expr_sfinae (op1, tf_none)); - if (doing_div_or_mod && (flag_sanitize & SANITIZE_DIVIDE)) + if (doing_div_or_mod && (flag_sanitize & (SANITIZE_DIVIDE + | SANITIZE_FLOAT_DIVIDE))) { /* For diagnostics we want to use the promoted types without shorten_binary_op. So convert the arguments to the diff --git a/gcc/cp/typeck2.c b/gcc/cp/typeck2.c index 70c06fb81bb..1d8b78278bf 100644 --- a/gcc/cp/typeck2.c +++ b/gcc/cp/typeck2.c @@ -1139,7 +1139,7 @@ massage_init_elt (tree type, tree init, tsubst_flags_t complain) /* When we defer constant folding within a statement, we may want to defer this folding as well. */ tree t = fold_non_dependent_expr_sfinae (init, complain); - t = maybe_constant_value (t); + t = maybe_constant_init (t); if (TREE_CONSTANT (t)) init = t; return init; @@ -1988,10 +1988,10 @@ nothrow_spec_p_uninst (const_tree spec) } /* Combine the two exceptions specifier lists LIST and ADD, and return - their union. If FN is non-null, it's the source of ADD. */ + their union. */ tree -merge_exception_specifiers (tree list, tree add, tree fn) +merge_exception_specifiers (tree list, tree add) { tree noex, orig_list; @@ -2007,22 +2007,18 @@ merge_exception_specifiers (tree list, tree add, tree fn) if (nothrow_spec_p_uninst (add)) return list; - noex = TREE_PURPOSE (list); - if (DEFERRED_NOEXCEPT_SPEC_P (add)) - { - /* If ADD is a deferred noexcept, we must have been called from - process_subob_fn. For implicitly declared functions, we build up - a list of functions to consider at instantiation time. */ - if (noex && operand_equal_p (noex, boolean_true_node, 0)) - noex = NULL_TREE; - gcc_assert (fn && (!noex || is_overloaded_fn (noex))); - noex = build_overload (fn, noex); - } - else if (nothrow_spec_p_uninst (list)) + /* Two implicit noexcept specs (e.g. on a destructor) are equivalent. */ + if (UNEVALUATED_NOEXCEPT_SPEC_P (add) + && UNEVALUATED_NOEXCEPT_SPEC_P (list)) + return list; + /* We should have instantiated other deferred noexcept specs by now. */ + gcc_assert (!DEFERRED_NOEXCEPT_SPEC_P (add)); + + if (nothrow_spec_p_uninst (list)) return add; - else - gcc_checking_assert (!TREE_PURPOSE (add) - || cp_tree_equal (noex, TREE_PURPOSE (add))); + noex = TREE_PURPOSE (list); + gcc_checking_assert (!TREE_PURPOSE (add) + || cp_tree_equal (noex, TREE_PURPOSE (add))); /* Combine the dynamic-exception-specifiers, if any. */ orig_list = list; diff --git a/gcc/doc/gimple.texi b/gcc/doc/gimple.texi index 9bb16e812dc..1e5473d109e 100644 --- a/gcc/doc/gimple.texi +++ b/gcc/doc/gimple.texi @@ -70,6 +70,7 @@ with the flag @option{-fdump-tree-gimple}. @menu * Tuple representation:: +* Class hierarchy of GIMPLE statements:: * GIMPLE instruction set:: * GIMPLE Exception Handling:: * Temporaries:: @@ -287,35 +288,135 @@ reduce memory utilization further by removing these sets). @end itemize All the other tuples are defined in terms of these three basic -ones. Each tuple will add some fields. The main gimple type -is defined to be the union of all these structures (@code{GTY} markers -elided for clarity): +ones. Each tuple will add some fields. + + +@node Class hierarchy of GIMPLE statements +@section Class hierarchy of GIMPLE statements +@cindex GIMPLE class hierarchy + +The following diagram shows the C++ inheritance hierarchy of statement +kinds, along with their relationships to @code{GSS_} values (layouts) and +@code{GIMPLE_} values (codes): @smallexample -union gimple_statement_d -@{ - struct gimple_statement_base gsbase; - struct gimple_statement_with_ops gsops; - struct gimple_statement_with_memory_ops gsmem; - struct gimple_statement_omp omp; - struct gimple_statement_bind gimple_bind; - struct gimple_statement_catch gimple_catch; - struct gimple_statement_eh_filter gimple_eh_filter; - struct gimple_statement_phi gimple_phi; - struct gimple_statement_resx gimple_resx; - struct gimple_statement_try gimple_try; - struct gimple_statement_wce gimple_wce; - struct gimple_statement_asm gimple_asm; - struct gimple_statement_omp_critical gimple_omp_critical; - struct gimple_statement_omp_for gimple_omp_for; - struct gimple_statement_omp_parallel gimple_omp_parallel; - struct gimple_statement_omp_task gimple_omp_task; - struct gimple_statement_omp_sections gimple_omp_sections; - struct gimple_statement_omp_single gimple_omp_single; - struct gimple_statement_omp_continue gimple_omp_continue; - struct gimple_statement_omp_atomic_load gimple_omp_atomic_load; - struct gimple_statement_omp_atomic_store gimple_omp_atomic_store; -@}; + gimple_statement_base + | layout: GSS_BASE + | used for 4 codes: GIMPLE_ERROR_MARK + | GIMPLE_NOP + | GIMPLE_OMP_SECTIONS_SWITCH + | GIMPLE_PREDICT + | + + gimple_statement_with_ops_base + | | (no GSS layout) + | | + | + gimple_statement_with_ops + | | layout: GSS_WITH_OPS + | | Used for 5 codes: GIMPLE_COND + | | GIMPLE_DEBUG + | | GIMPLE_GOTO + | | GIMPLE_LABEL + | | GIMPLE_SWITCH + | | + | + gimple_statement_with_memory_ops_base + | | layout: GSS_WITH_MEM_OPS_BASE + | | + | + gimple_statement_with_memory_ops + | | layout: GSS_WITH_MEM_OPS. + | | used for codes GIMPLE_ASSIGN and GIMPLE_RETURN. + | | + | + gimple_statement_call + | | layout: GSS_CALL, code: GIMPLE_CALL + | | + | + gimple_statement_asm + | | layout: GSS_ASM, code: GIMPLE_ASM + | | + | + gimple_statement_transaction + | layout: GSS_TRANSACTION, code: GIMPLE_TRANSACTION + | + + gimple_statement_omp + | | layout: GSS_OMP. Used for code GIMPLE_OMP_SECTION + | | + | + gimple_statement_omp_critical + | | layout: GSS_OMP_CRITICAL, code: GIMPLE_OMP_CRITICAL + | | + | + gimple_statement_omp_for + | | layout: GSS_OMP_FOR, code: GIMPLE_OMP_FOR + | | + | + gimple_statement_omp_parallel_layout + | | | layout: GSS_OMP_PARALLEL_LAYOUT + | | | + | | + gimple_statement_omp_taskreg + | | | | + | | | + gimple_statement_omp_parallel + | | | | code: GIMPLE_OMP_PARALLEL + | | | | + | | | + gimple_statement_omp_task + | | | code: GIMPLE_OMP_TASK + | | | + | | + gimple_statement_omp_target + | | code: GIMPLE_OMP_TARGET + | | + | + gimple_statement_omp_sections + | | layout: GSS_OMP_SECTIONS, code: GIMPLE_OMP_SECTIONS + | | + | + gimple_statement_omp_single_layout + | | layout: GSS_OMP_SINGLE_LAYOUT + | | + | + gimple_statement_omp_single + | | code: GIMPLE_OMP_SINGLE + | | + | + gimple_statement_omp_teams + | code: GIMPLE_OMP_TEAMS + | + + gimple_statement_bind + | layout: GSS_BIND, code: GIMPLE_BIND + | + + gimple_statement_catch + | layout: GSS_CATCH, code: GIMPLE_CATCH + | + + gimple_statement_eh_filter + | layout: GSS_EH_FILTER, code: GIMPLE_EH_FILTER + | + + gimple_statement_eh_else + | layout: GSS_EH_ELSE, code: GIMPLE_EH_ELSE + | + + gimple_statement_eh_mnt + | layout: GSS_EH_MNT, code: GIMPLE_EH_MUST_NOT_THROW + | + + gimple_statement_phi + | layout: GSS_PHI, code: GIMPLE_PHI + | + + gimple_statement_eh_ctrl + | | layout: GSS_EH_CTRL + | | + | + gimple_statement_resx + | | code: GIMPLE_RESX + | | + | + gimple_statement_eh_dispatch + | code: GIMPLE_EH_DISPATCH + | + + gimple_statement_try + | layout: GSS_TRY, code: GIMPLE_TRY + | + + gimple_statement_wce + | layout: GSS_WCE, code: GIMPLE_WITH_CLEANUP_EXPR + | + + gimple_statement_omp_continue + | layout: GSS_OMP_CONTINUE, code: GIMPLE_OMP_CONTINUE + | + + gimple_statement_omp_atomic_load + | layout: GSS_OMP_ATOMIC_LOAD, code: GIMPLE_OMP_ATOMIC_LOAD + | + + gimple_statement_omp_atomic_store_layout + | layout: GSS_OMP_ATOMIC_STORE_LAYOUT, + | code: GIMPLE_OMP_ATOMIC_STORE + | + + gimple_statement_omp_atomic_store + | code: GIMPLE_OMP_ATOMIC_STORE + | + + gimple_statement_omp_return + code: GIMPLE_OMP_RETURN @end smallexample diff --git a/gcc/doc/invoke.texi b/gcc/doc/invoke.texi index da7a00ed00c..cc6d405f168 100644 --- a/gcc/doc/invoke.texi +++ b/gcc/doc/invoke.texi @@ -4732,7 +4732,7 @@ Warn if a global function is defined without a previous declaration. Do so even if the definition itself provides a prototype. Use this option to detect global functions that are not declared in header files. In C, no warnings are issued for functions with previous -non-prototype declarations; use @option{-Wmissing-prototype} to detect +non-prototype declarations; use @option{-Wmissing-prototypes} to detect missing prototypes. In C++, no warnings are issued for function templates, or for inline functions, or for functions in anonymous namespaces. diff --git a/gcc/dominance.c b/gcc/dominance.c index ff0dfe6714f..7adec4f7376 100644 --- a/gcc/dominance.c +++ b/gcc/dominance.c @@ -681,24 +681,30 @@ calculate_dominance_info (enum cdi_direction dir) /* Free dominance information for direction DIR. */ void -free_dominance_info (enum cdi_direction dir) +free_dominance_info (function *fn, enum cdi_direction dir) { basic_block bb; unsigned int dir_index = dom_convert_dir_to_idx (dir); - if (!dom_info_available_p (dir)) + if (!dom_info_available_p (fn, dir)) return; - FOR_ALL_BB_FN (bb, cfun) + FOR_ALL_BB_FN (bb, fn) { et_free_tree_force (bb->dom[dir_index]); bb->dom[dir_index] = NULL; } et_free_pools (); - n_bbs_in_dom_tree[dir_index] = 0; + fn->cfg->x_n_bbs_in_dom_tree[dir_index] = 0; + + fn->cfg->x_dom_computed[dir_index] = DOM_NONE; +} - dom_computed[dir_index] = DOM_NONE; +void +free_dominance_info (enum cdi_direction dir) +{ + free_dominance_info (cfun, dir); } /* Return the immediate dominator of basic block BB. */ @@ -1461,11 +1467,19 @@ next_dom_son (enum cdi_direction dir, basic_block bb) /* Return dominance availability for dominance info DIR. */ enum dom_state -dom_info_state (enum cdi_direction dir) +dom_info_state (function *fn, enum cdi_direction dir) { + if (!fn->cfg) + return DOM_NONE; + unsigned int dir_index = dom_convert_dir_to_idx (dir); + return fn->cfg->x_dom_computed[dir_index]; +} - return dom_computed[dir_index]; +enum dom_state +dom_info_state (enum cdi_direction dir) +{ + return dom_info_state (cfun, dir); } /* Set the dominance availability for dominance info DIR to NEW_STATE. */ @@ -1481,11 +1495,15 @@ set_dom_info_availability (enum cdi_direction dir, enum dom_state new_state) /* Returns true if dominance information for direction DIR is available. */ bool -dom_info_available_p (enum cdi_direction dir) +dom_info_available_p (function *fn, enum cdi_direction dir) { - unsigned int dir_index = dom_convert_dir_to_idx (dir); + return dom_info_state (fn, dir) != DOM_NONE; +} - return dom_computed[dir_index] != DOM_NONE; +bool +dom_info_available_p (enum cdi_direction dir) +{ + return dom_info_available_p (cfun, dir); } DEBUG_FUNCTION void diff --git a/gcc/expr.c b/gcc/expr.c index 989a8780dc9..2868d9d3443 100644 --- a/gcc/expr.c +++ b/gcc/expr.c @@ -68,22 +68,6 @@ along with GCC; see the file COPYING3. If not see #include "tree-ssa-address.h" #include "cfgexpand.h" -/* Decide whether a function's arguments should be processed - from first to last or from last to first. - - They should if the stack and args grow in opposite directions, but - only if we have push insns. */ - -#ifdef PUSH_ROUNDING - -#ifndef PUSH_ARGS_REVERSED -#if defined (STACK_GROWS_DOWNWARD) != defined (ARGS_GROW_DOWNWARD) -#define PUSH_ARGS_REVERSED /* If it's last to first. */ -#endif -#endif - -#endif - #ifndef STACK_PUSH_CODE #ifdef STACK_GROWS_DOWNWARD #define STACK_PUSH_CODE PRE_DEC @@ -4334,11 +4318,7 @@ emit_push_insn (rtx x, enum machine_mode mode, tree type, rtx size, /* Loop over all the words allocated on the stack for this arg. */ /* We can do it by words, because any scalar bigger than a word has a size a multiple of a word. */ -#ifndef PUSH_ARGS_REVERSED - for (i = not_stack; i < size; i++) -#else for (i = size - 1; i >= not_stack; i--) -#endif if (i >= not_stack + offset) emit_push_insn (operand_subword_force (x, i, mode), word_mode, NULL_TREE, NULL_RTX, align, 0, NULL_RTX, diff --git a/gcc/flag-types.h b/gcc/flag-types.h index fc3261bbbb7..caf4039f9df 100644 --- a/gcc/flag-types.h +++ b/gcc/flag-types.h @@ -228,6 +228,7 @@ enum sanitize_code { SANITIZE_SI_OVERFLOW = 1 << 9, SANITIZE_BOOL = 1 << 10, SANITIZE_ENUM = 1 << 11, + SANITIZE_FLOAT_DIVIDE = 1 << 12, SANITIZE_UNDEFINED = SANITIZE_SHIFT | SANITIZE_DIVIDE | SANITIZE_UNREACHABLE | SANITIZE_VLA | SANITIZE_NULL | SANITIZE_RETURN | SANITIZE_SI_OVERFLOW | SANITIZE_BOOL | SANITIZE_ENUM diff --git a/gcc/gcc.c b/gcc/gcc.c index e5130d13535..7bea6d79ac2 100644 --- a/gcc/gcc.c +++ b/gcc/gcc.c @@ -8170,7 +8170,7 @@ sanitize_spec_function (int argc, const char **argv) if (strcmp (argv[0], "thread") == 0) return (flag_sanitize & SANITIZE_THREAD) ? "" : NULL; if (strcmp (argv[0], "undefined") == 0) - return ((flag_sanitize & SANITIZE_UNDEFINED) + return ((flag_sanitize & (SANITIZE_UNDEFINED | SANITIZE_FLOAT_DIVIDE)) && !flag_sanitize_undefined_trap_on_error) ? "" : NULL; if (strcmp (argv[0], "leak") == 0) return ((flag_sanitize diff --git a/gcc/gimple.c b/gcc/gimple.c index e51329f2702..d6bc15bf424 100644 --- a/gcc/gimple.c +++ b/gcc/gimple.c @@ -2565,8 +2565,8 @@ infer_nonnull_range (gimple stmt, tree op, bool dereference, bool attribute) { for (unsigned int i = 0; i < gimple_call_num_args (stmt); i++) { - if (operand_equal_p (op, gimple_call_arg (stmt, i), 0) - && POINTER_TYPE_P (TREE_TYPE (gimple_call_arg (stmt, i)))) + if (POINTER_TYPE_P (TREE_TYPE (gimple_call_arg (stmt, i))) + && operand_equal_p (op, gimple_call_arg (stmt, i), 0)) return true; } return false; diff --git a/gcc/go/ChangeLog b/gcc/go/ChangeLog index 1a863b04dbc..8d86e746d6e 100644 --- a/gcc/go/ChangeLog +++ b/gcc/go/ChangeLog @@ -1,3 +1,12 @@ +2014-04-30 Chris Manghane <cmang@google.com> + + * go-gcc.cc: #include "langhooks.h". + (Gcc_backend::Gcc_backend): Add constructor. + (Gcc_backend::lookup_function): New function. + (Gcc_backend::define_builtin): New private function. + (Gcc_backend::gcc_backend): Remove. + (go_get_backend): Use new to create new Gcc_backend. + 2014-04-25 Chris Manghane <cmang@google.com> * go-gcc.cc: Include "cgraph.h" and "gimplify.h". diff --git a/gcc/go/go-gcc.cc b/gcc/go/go-gcc.cc index a0283fe12e0..45827a08ea2 100644 --- a/gcc/go/go-gcc.cc +++ b/gcc/go/go-gcc.cc @@ -34,6 +34,7 @@ #include "basic-block.h" #include "gimple-expr.h" #include "gimplify.h" +#include "langhooks.h" #include "toplev.h" #include "output.h" #include "real.h" @@ -131,6 +132,8 @@ class Blabel : public Gcc_tree class Gcc_backend : public Backend { public: + Gcc_backend(); + // Types. Btype* @@ -425,6 +428,9 @@ class Gcc_backend : public Backend bool function_set_body(Bfunction* function, Bstatement* code_stmt); + Bfunction* + lookup_builtin(const std::string&); + void write_global_definitions(const std::vector<Btype*>&, const std::vector<Bexpression*>&, @@ -459,6 +465,14 @@ class Gcc_backend : public Backend tree non_zero_size_type(tree); + +private: + void + define_builtin(built_in_function bcode, const char* name, const char* libname, + tree fntype, bool const_p); + + // A mapping of the GCC built-ins exposed to GCCGo. + std::map<std::string, Bfunction*> builtin_functions_; }; // A helper function. @@ -469,6 +483,172 @@ get_identifier_from_string(const std::string& str) return get_identifier_with_length(str.data(), str.length()); } +// Define the built-in functions that are exposed to GCCGo. + +Gcc_backend::Gcc_backend() +{ + /* We need to define the fetch_and_add functions, since we use them + for ++ and --. */ + tree t = this->integer_type(BITS_PER_UNIT, 1)->get_tree(); + tree p = build_pointer_type(build_qualified_type(t, TYPE_QUAL_VOLATILE)); + this->define_builtin(BUILT_IN_SYNC_ADD_AND_FETCH_1, "__sync_fetch_and_add_1", + NULL, build_function_type_list(t, p, t, NULL_TREE), + false); + + t = this->integer_type(BITS_PER_UNIT * 2, 1)->get_tree(); + p = build_pointer_type(build_qualified_type(t, TYPE_QUAL_VOLATILE)); + this->define_builtin(BUILT_IN_SYNC_ADD_AND_FETCH_2, "__sync_fetch_and_add_2", + NULL, build_function_type_list(t, p, t, NULL_TREE), + false); + + t = this->integer_type(BITS_PER_UNIT * 4, 1)->get_tree(); + p = build_pointer_type(build_qualified_type(t, TYPE_QUAL_VOLATILE)); + this->define_builtin(BUILT_IN_SYNC_ADD_AND_FETCH_4, "__sync_fetch_and_add_4", + NULL, build_function_type_list(t, p, t, NULL_TREE), + false); + + t = this->integer_type(BITS_PER_UNIT * 8, 1)->get_tree(); + p = build_pointer_type(build_qualified_type(t, TYPE_QUAL_VOLATILE)); + this->define_builtin(BUILT_IN_SYNC_ADD_AND_FETCH_8, "__sync_fetch_and_add_8", + NULL, build_function_type_list(t, p, t, NULL_TREE), + false); + + // We use __builtin_expect for magic import functions. + this->define_builtin(BUILT_IN_EXPECT, "__builtin_expect", NULL, + build_function_type_list(long_integer_type_node, + long_integer_type_node, + long_integer_type_node, + NULL_TREE), + true); + + // We use __builtin_memcmp for struct comparisons. + this->define_builtin(BUILT_IN_MEMCMP, "__builtin_memcmp", "memcmp", + build_function_type_list(integer_type_node, + const_ptr_type_node, + const_ptr_type_node, + size_type_node, + NULL_TREE), + false); + + // We provide some functions for the math library. + tree math_function_type = build_function_type_list(double_type_node, + double_type_node, + NULL_TREE); + tree math_function_type_long = + build_function_type_list(long_double_type_node, long_double_type_node, + long_double_type_node, NULL_TREE); + tree math_function_type_two = build_function_type_list(double_type_node, + double_type_node, + double_type_node, + NULL_TREE); + tree math_function_type_long_two = + build_function_type_list(long_double_type_node, long_double_type_node, + long_double_type_node, NULL_TREE); + this->define_builtin(BUILT_IN_ACOS, "__builtin_acos", "acos", + math_function_type, true); + this->define_builtin(BUILT_IN_ACOSL, "__builtin_acosl", "acosl", + math_function_type_long, true); + this->define_builtin(BUILT_IN_ASIN, "__builtin_asin", "asin", + math_function_type, true); + this->define_builtin(BUILT_IN_ASINL, "__builtin_asinl", "asinl", + math_function_type_long, true); + this->define_builtin(BUILT_IN_ATAN, "__builtin_atan", "atan", + math_function_type, true); + this->define_builtin(BUILT_IN_ATANL, "__builtin_atanl", "atanl", + math_function_type_long, true); + this->define_builtin(BUILT_IN_ATAN2, "__builtin_atan2", "atan2", + math_function_type_two, true); + this->define_builtin(BUILT_IN_ATAN2L, "__builtin_atan2l", "atan2l", + math_function_type_long_two, true); + this->define_builtin(BUILT_IN_CEIL, "__builtin_ceil", "ceil", + math_function_type, true); + this->define_builtin(BUILT_IN_CEILL, "__builtin_ceill", "ceill", + math_function_type_long, true); + this->define_builtin(BUILT_IN_COS, "__builtin_cos", "cos", + math_function_type, true); + this->define_builtin(BUILT_IN_COSL, "__builtin_cosl", "cosl", + math_function_type_long, true); + this->define_builtin(BUILT_IN_EXP, "__builtin_exp", "exp", + math_function_type, true); + this->define_builtin(BUILT_IN_EXPL, "__builtin_expl", "expl", + math_function_type_long, true); + this->define_builtin(BUILT_IN_EXPM1, "__builtin_expm1", "expm1", + math_function_type, true); + this->define_builtin(BUILT_IN_EXPM1L, "__builtin_expm1l", "expm1l", + math_function_type_long, true); + this->define_builtin(BUILT_IN_FABS, "__builtin_fabs", "fabs", + math_function_type, true); + this->define_builtin(BUILT_IN_FABSL, "__builtin_fabsl", "fabsl", + math_function_type_long, true); + this->define_builtin(BUILT_IN_FLOOR, "__builtin_floor", "floor", + math_function_type, true); + this->define_builtin(BUILT_IN_FLOORL, "__builtin_floorl", "floorl", + math_function_type_long, true); + this->define_builtin(BUILT_IN_FMOD, "__builtin_fmod", "fmod", + math_function_type_two, true); + this->define_builtin(BUILT_IN_FMODL, "__builtin_fmodl", "fmodl", + math_function_type_long_two, true); + this->define_builtin(BUILT_IN_LDEXP, "__builtin_ldexp", "ldexp", + build_function_type_list(double_type_node, + double_type_node, + integer_type_node, + NULL_TREE), + true); + this->define_builtin(BUILT_IN_LDEXPL, "__builtin_ldexpl", "ldexpl", + build_function_type_list(long_double_type_node, + long_double_type_node, + integer_type_node, + NULL_TREE), + true); + this->define_builtin(BUILT_IN_LOG, "__builtin_log", "log", + math_function_type, true); + this->define_builtin(BUILT_IN_LOGL, "__builtin_logl", "logl", + math_function_type_long, true); + this->define_builtin(BUILT_IN_LOG1P, "__builtin_log1p", "log1p", + math_function_type, true); + this->define_builtin(BUILT_IN_LOG1PL, "__builtin_log1pl", "log1pl", + math_function_type_long, true); + this->define_builtin(BUILT_IN_LOG10, "__builtin_log10", "log10", + math_function_type, true); + this->define_builtin(BUILT_IN_LOG10L, "__builtin_log10l", "log10l", + math_function_type_long, true); + this->define_builtin(BUILT_IN_LOG2, "__builtin_log2", "log2", + math_function_type, true); + this->define_builtin(BUILT_IN_LOG2L, "__builtin_log2l", "log2l", + math_function_type_long, true); + this->define_builtin(BUILT_IN_SIN, "__builtin_sin", "sin", + math_function_type, true); + this->define_builtin(BUILT_IN_SINL, "__builtin_sinl", "sinl", + math_function_type_long, true); + this->define_builtin(BUILT_IN_SQRT, "__builtin_sqrt", "sqrt", + math_function_type, true); + this->define_builtin(BUILT_IN_SQRTL, "__builtin_sqrtl", "sqrtl", + math_function_type_long, true); + this->define_builtin(BUILT_IN_TAN, "__builtin_tan", "tan", + math_function_type, true); + this->define_builtin(BUILT_IN_TANL, "__builtin_tanl", "tanl", + math_function_type_long, true); + this->define_builtin(BUILT_IN_TRUNC, "__builtin_trunc", "trunc", + math_function_type, true); + this->define_builtin(BUILT_IN_TRUNCL, "__builtin_truncl", "truncl", + math_function_type_long, true); + + // We use __builtin_return_address in the thunk we build for + // functions which call recover. + this->define_builtin(BUILT_IN_RETURN_ADDRESS, "__builtin_return_address", + NULL, + build_function_type_list(ptr_type_node, + unsigned_type_node, + NULL_TREE), + false); + + // The compiler uses __builtin_trap for some exception handling + // cases. + this->define_builtin(BUILT_IN_TRAP, "__builtin_trap", NULL, + build_function_type(void_type_node, void_list_node), + false); +} + // Get an unnamed integer type. Btype* @@ -2597,6 +2777,17 @@ Gcc_backend::function_set_body(Bfunction* function, Bstatement* code_stmt) return true; } +// Look up a named built-in function in the current backend implementation. +// Returns NULL if no built-in function by that name exists. + +Bfunction* +Gcc_backend::lookup_builtin(const std::string& name) +{ + if (this->builtin_functions_.count(name) != 0) + return this->builtin_functions_[name]; + return NULL; +} + // Write the definitions for all TYPE_DECLS, CONSTANT_DECLS, // FUNCTION_DECLS, and VARIABLE_DECLS declared globally. @@ -2679,16 +2870,38 @@ Gcc_backend::write_global_definitions( delete[] defs; } -// The single backend. +// Define a builtin function. BCODE is the builtin function code +// defined by builtins.def. NAME is the name of the builtin function. +// LIBNAME is the name of the corresponding library function, and is +// NULL if there isn't one. FNTYPE is the type of the function. +// CONST_P is true if the function has the const attribute. -static Gcc_backend gcc_backend; +void +Gcc_backend::define_builtin(built_in_function bcode, const char* name, + const char* libname, tree fntype, bool const_p) +{ + tree decl = add_builtin_function(name, fntype, bcode, BUILT_IN_NORMAL, + libname, NULL_TREE); + if (const_p) + TREE_READONLY(decl) = 1; + set_builtin_decl(bcode, decl, true); + this->builtin_functions_[name] = this->make_function(decl); + if (libname != NULL) + { + decl = add_builtin_function(libname, fntype, bcode, BUILT_IN_NORMAL, + NULL, NULL_TREE); + if (const_p) + TREE_READONLY(decl) = 1; + this->builtin_functions_[libname] = this->make_function(decl); + } +} // Return the backend generator. Backend* go_get_backend() { - return &gcc_backend; + return new Gcc_backend(); } // FIXME: Temporary functions while converting to the new backend diff --git a/gcc/go/gofrontend/backend.h b/gcc/go/gofrontend/backend.h index aca3dc6f90e..786223fe736 100644 --- a/gcc/go/gofrontend/backend.h +++ b/gcc/go/gofrontend/backend.h @@ -660,6 +660,11 @@ class Backend virtual bool function_set_body(Bfunction* function, Bstatement* code_stmt) = 0; + // Look up a named built-in function in the current backend implementation. + // Returns NULL if no built-in function by that name exists. + virtual Bfunction* + lookup_builtin(const std::string&) = 0; + // Utility. // Write the definitions for all TYPE_DECLS, CONSTANT_DECLS, diff --git a/gcc/go/gofrontend/go.cc b/gcc/go/gofrontend/go.cc index ac772a095f7..d2331f3e0ae 100644 --- a/gcc/go/gofrontend/go.cc +++ b/gcc/go/gofrontend/go.cc @@ -34,9 +34,6 @@ go_create_gogo(int int_type_size, int pointer_size, const char *pkgpath, if (relative_import_path != NULL) ::gogo->set_relative_import_path(relative_import_path); - - // FIXME: This should be in the gcc dependent code. - ::gogo->define_builtin_function_trees(); } // Parse the input files. diff --git a/gcc/go/gofrontend/gogo-tree.cc b/gcc/go/gofrontend/gogo-tree.cc index 6b19a1d82e1..5b9a8180287 100644 --- a/gcc/go/gofrontend/gogo-tree.cc +++ b/gcc/go/gofrontend/gogo-tree.cc @@ -36,266 +36,6 @@ saw_errors() return errorcount != 0 || sorrycount != 0; } -// A helper function. - -static inline tree -get_identifier_from_string(const std::string& str) -{ - return get_identifier_with_length(str.data(), str.length()); -} - -// Builtin functions. - -static std::map<std::string, tree> builtin_functions; - -// Define a builtin function. BCODE is the builtin function code -// defined by builtins.def. NAME is the name of the builtin function. -// LIBNAME is the name of the corresponding library function, and is -// NULL if there isn't one. FNTYPE is the type of the function. -// CONST_P is true if the function has the const attribute. - -static void -define_builtin(built_in_function bcode, const char* name, const char* libname, - tree fntype, bool const_p) -{ - tree decl = add_builtin_function(name, fntype, bcode, BUILT_IN_NORMAL, - libname, NULL_TREE); - if (const_p) - TREE_READONLY(decl) = 1; - set_builtin_decl(bcode, decl, true); - builtin_functions[name] = decl; - if (libname != NULL) - { - decl = add_builtin_function(libname, fntype, bcode, BUILT_IN_NORMAL, - NULL, NULL_TREE); - if (const_p) - TREE_READONLY(decl) = 1; - builtin_functions[libname] = decl; - } -} - -// Create trees for implicit builtin functions. - -void -Gogo::define_builtin_function_trees() -{ - /* We need to define the fetch_and_add functions, since we use them - for ++ and --. */ - tree t = go_type_for_size(BITS_PER_UNIT, 1); - tree p = build_pointer_type(build_qualified_type(t, TYPE_QUAL_VOLATILE)); - define_builtin(BUILT_IN_SYNC_ADD_AND_FETCH_1, "__sync_fetch_and_add_1", NULL, - build_function_type_list(t, p, t, NULL_TREE), false); - - t = go_type_for_size(BITS_PER_UNIT * 2, 1); - p = build_pointer_type(build_qualified_type(t, TYPE_QUAL_VOLATILE)); - define_builtin (BUILT_IN_SYNC_ADD_AND_FETCH_2, "__sync_fetch_and_add_2", NULL, - build_function_type_list(t, p, t, NULL_TREE), false); - - t = go_type_for_size(BITS_PER_UNIT * 4, 1); - p = build_pointer_type(build_qualified_type(t, TYPE_QUAL_VOLATILE)); - define_builtin(BUILT_IN_SYNC_ADD_AND_FETCH_4, "__sync_fetch_and_add_4", NULL, - build_function_type_list(t, p, t, NULL_TREE), false); - - t = go_type_for_size(BITS_PER_UNIT * 8, 1); - p = build_pointer_type(build_qualified_type(t, TYPE_QUAL_VOLATILE)); - define_builtin(BUILT_IN_SYNC_ADD_AND_FETCH_8, "__sync_fetch_and_add_8", NULL, - build_function_type_list(t, p, t, NULL_TREE), false); - - // We use __builtin_expect for magic import functions. - define_builtin(BUILT_IN_EXPECT, "__builtin_expect", NULL, - build_function_type_list(long_integer_type_node, - long_integer_type_node, - long_integer_type_node, - NULL_TREE), - true); - - // We use __builtin_memcmp for struct comparisons. - define_builtin(BUILT_IN_MEMCMP, "__builtin_memcmp", "memcmp", - build_function_type_list(integer_type_node, - const_ptr_type_node, - const_ptr_type_node, - size_type_node, - NULL_TREE), - false); - - // We provide some functions for the math library. - tree math_function_type = build_function_type_list(double_type_node, - double_type_node, - NULL_TREE); - tree math_function_type_long = - build_function_type_list(long_double_type_node, long_double_type_node, - long_double_type_node, NULL_TREE); - tree math_function_type_two = build_function_type_list(double_type_node, - double_type_node, - double_type_node, - NULL_TREE); - tree math_function_type_long_two = - build_function_type_list(long_double_type_node, long_double_type_node, - long_double_type_node, NULL_TREE); - define_builtin(BUILT_IN_ACOS, "__builtin_acos", "acos", - math_function_type, true); - define_builtin(BUILT_IN_ACOSL, "__builtin_acosl", "acosl", - math_function_type_long, true); - define_builtin(BUILT_IN_ASIN, "__builtin_asin", "asin", - math_function_type, true); - define_builtin(BUILT_IN_ASINL, "__builtin_asinl", "asinl", - math_function_type_long, true); - define_builtin(BUILT_IN_ATAN, "__builtin_atan", "atan", - math_function_type, true); - define_builtin(BUILT_IN_ATANL, "__builtin_atanl", "atanl", - math_function_type_long, true); - define_builtin(BUILT_IN_ATAN2, "__builtin_atan2", "atan2", - math_function_type_two, true); - define_builtin(BUILT_IN_ATAN2L, "__builtin_atan2l", "atan2l", - math_function_type_long_two, true); - define_builtin(BUILT_IN_CEIL, "__builtin_ceil", "ceil", - math_function_type, true); - define_builtin(BUILT_IN_CEILL, "__builtin_ceill", "ceill", - math_function_type_long, true); - define_builtin(BUILT_IN_COS, "__builtin_cos", "cos", - math_function_type, true); - define_builtin(BUILT_IN_COSL, "__builtin_cosl", "cosl", - math_function_type_long, true); - define_builtin(BUILT_IN_EXP, "__builtin_exp", "exp", - math_function_type, true); - define_builtin(BUILT_IN_EXPL, "__builtin_expl", "expl", - math_function_type_long, true); - define_builtin(BUILT_IN_EXPM1, "__builtin_expm1", "expm1", - math_function_type, true); - define_builtin(BUILT_IN_EXPM1L, "__builtin_expm1l", "expm1l", - math_function_type_long, true); - define_builtin(BUILT_IN_FABS, "__builtin_fabs", "fabs", - math_function_type, true); - define_builtin(BUILT_IN_FABSL, "__builtin_fabsl", "fabsl", - math_function_type_long, true); - define_builtin(BUILT_IN_FLOOR, "__builtin_floor", "floor", - math_function_type, true); - define_builtin(BUILT_IN_FLOORL, "__builtin_floorl", "floorl", - math_function_type_long, true); - define_builtin(BUILT_IN_FMOD, "__builtin_fmod", "fmod", - math_function_type_two, true); - define_builtin(BUILT_IN_FMODL, "__builtin_fmodl", "fmodl", - math_function_type_long_two, true); - define_builtin(BUILT_IN_LDEXP, "__builtin_ldexp", "ldexp", - build_function_type_list(double_type_node, - double_type_node, - integer_type_node, - NULL_TREE), - true); - define_builtin(BUILT_IN_LDEXPL, "__builtin_ldexpl", "ldexpl", - build_function_type_list(long_double_type_node, - long_double_type_node, - integer_type_node, - NULL_TREE), - true); - define_builtin(BUILT_IN_LOG, "__builtin_log", "log", - math_function_type, true); - define_builtin(BUILT_IN_LOGL, "__builtin_logl", "logl", - math_function_type_long, true); - define_builtin(BUILT_IN_LOG1P, "__builtin_log1p", "log1p", - math_function_type, true); - define_builtin(BUILT_IN_LOG1PL, "__builtin_log1pl", "log1pl", - math_function_type_long, true); - define_builtin(BUILT_IN_LOG10, "__builtin_log10", "log10", - math_function_type, true); - define_builtin(BUILT_IN_LOG10L, "__builtin_log10l", "log10l", - math_function_type_long, true); - define_builtin(BUILT_IN_LOG2, "__builtin_log2", "log2", - math_function_type, true); - define_builtin(BUILT_IN_LOG2L, "__builtin_log2l", "log2l", - math_function_type_long, true); - define_builtin(BUILT_IN_SIN, "__builtin_sin", "sin", - math_function_type, true); - define_builtin(BUILT_IN_SINL, "__builtin_sinl", "sinl", - math_function_type_long, true); - define_builtin(BUILT_IN_SQRT, "__builtin_sqrt", "sqrt", - math_function_type, true); - define_builtin(BUILT_IN_SQRTL, "__builtin_sqrtl", "sqrtl", - math_function_type_long, true); - define_builtin(BUILT_IN_TAN, "__builtin_tan", "tan", - math_function_type, true); - define_builtin(BUILT_IN_TANL, "__builtin_tanl", "tanl", - math_function_type_long, true); - define_builtin(BUILT_IN_TRUNC, "__builtin_trunc", "trunc", - math_function_type, true); - define_builtin(BUILT_IN_TRUNCL, "__builtin_truncl", "truncl", - math_function_type_long, true); - - // We use __builtin_return_address in the thunk we build for - // functions which call recover. - define_builtin(BUILT_IN_RETURN_ADDRESS, "__builtin_return_address", NULL, - build_function_type_list(ptr_type_node, - unsigned_type_node, - NULL_TREE), - false); - - // The compiler uses __builtin_trap for some exception handling - // cases. - define_builtin(BUILT_IN_TRAP, "__builtin_trap", NULL, - build_function_type(void_type_node, void_list_node), - false); -} - -// Get the backend representation. - -Bfunction* -Function_declaration::get_or_make_decl(Gogo* gogo, Named_object* no) -{ - if (this->fndecl_ == NULL) - { - // Let Go code use an asm declaration to pick up a builtin - // function. - if (!this->asm_name_.empty()) - { - std::map<std::string, tree>::const_iterator p = - builtin_functions.find(this->asm_name_); - if (p != builtin_functions.end()) - { - this->fndecl_ = tree_to_function(p->second); - return this->fndecl_; - } - } - - std::string asm_name; - if (this->asm_name_.empty()) - { - asm_name = (no->package() == NULL - ? gogo->pkgpath_symbol() - : no->package()->pkgpath_symbol()); - asm_name.append(1, '.'); - asm_name.append(Gogo::unpack_hidden_name(no->name())); - if (this->fntype_->is_method()) - { - asm_name.append(1, '.'); - Type* rtype = this->fntype_->receiver()->type(); - asm_name.append(rtype->mangled_name(gogo)); - } - } - - Btype* functype = this->fntype_->get_backend_fntype(gogo); - this->fndecl_ = - gogo->backend()->function(functype, no->get_id(gogo), asm_name, - true, true, true, false, false, - this->location()); - } - - return this->fndecl_; -} - -// Build the descriptor for a function declaration. This won't -// necessarily happen if the package has just a declaration for the -// function and no other reference to it, but we may still need the -// descriptor for references from other packages. -void -Function_declaration::build_backend_descriptor(Gogo* gogo) -{ - if (this->descriptor_ != NULL) - { - Translate_context context(gogo, NULL, NULL, NULL); - this->descriptor_->get_tree(&context); - } -} - // Return the integer type to use for a size. GO_EXTERN_C @@ -380,121 +120,3 @@ go_type_for_mode(enum machine_mode mode, int unsignedp) else return NULL_TREE; } - -// Build a constructor for a slice. SLICE_TYPE_TREE is the type of -// the slice. VALUES is the value pointer and COUNT is the number of -// entries. If CAPACITY is not NULL, it is the capacity; otherwise -// the capacity and the count are the same. - -tree -Gogo::slice_constructor(tree slice_type_tree, tree values, tree count, - tree capacity) -{ - go_assert(TREE_CODE(slice_type_tree) == RECORD_TYPE); - - vec<constructor_elt, va_gc> *init; - vec_alloc(init, 3); - - tree field = TYPE_FIELDS(slice_type_tree); - go_assert(strcmp(IDENTIFIER_POINTER(DECL_NAME(field)), "__values") == 0); - constructor_elt empty = {NULL, NULL}; - constructor_elt* elt = init->quick_push(empty); - elt->index = field; - go_assert(TYPE_MAIN_VARIANT(TREE_TYPE(field)) - == TYPE_MAIN_VARIANT(TREE_TYPE(values))); - elt->value = values; - - count = fold_convert(sizetype, count); - if (capacity == NULL_TREE) - { - count = save_expr(count); - capacity = count; - } - - field = DECL_CHAIN(field); - go_assert(strcmp(IDENTIFIER_POINTER(DECL_NAME(field)), "__count") == 0); - elt = init->quick_push(empty); - elt->index = field; - elt->value = fold_convert(TREE_TYPE(field), count); - - field = DECL_CHAIN(field); - go_assert(strcmp(IDENTIFIER_POINTER(DECL_NAME(field)), "__capacity") == 0); - elt = init->quick_push(empty); - elt->index = field; - elt->value = fold_convert(TREE_TYPE(field), capacity); - - return build_constructor(slice_type_tree, init); -} - -// Mark a function as a builtin library function. - -void -Gogo::mark_fndecl_as_builtin_library(tree fndecl) -{ - DECL_EXTERNAL(fndecl) = 1; - TREE_PUBLIC(fndecl) = 1; - DECL_ARTIFICIAL(fndecl) = 1; - TREE_NOTHROW(fndecl) = 1; - DECL_VISIBILITY(fndecl) = VISIBILITY_DEFAULT; - DECL_VISIBILITY_SPECIFIED(fndecl) = 1; -} - -// Build a call to a builtin function. - -tree -Gogo::call_builtin(tree* pdecl, Location location, const char* name, - int nargs, tree rettype, ...) -{ - if (rettype == error_mark_node) - return error_mark_node; - - tree* types = new tree[nargs]; - tree* args = new tree[nargs]; - - va_list ap; - va_start(ap, rettype); - for (int i = 0; i < nargs; ++i) - { - types[i] = va_arg(ap, tree); - args[i] = va_arg(ap, tree); - if (types[i] == error_mark_node || args[i] == error_mark_node) - { - delete[] types; - delete[] args; - return error_mark_node; - } - } - va_end(ap); - - if (*pdecl == NULL_TREE) - { - tree fnid = get_identifier(name); - - tree argtypes = NULL_TREE; - tree* pp = &argtypes; - for (int i = 0; i < nargs; ++i) - { - *pp = tree_cons(NULL_TREE, types[i], NULL_TREE); - pp = &TREE_CHAIN(*pp); - } - *pp = void_list_node; - - tree fntype = build_function_type(rettype, argtypes); - - *pdecl = build_decl(BUILTINS_LOCATION, FUNCTION_DECL, fnid, fntype); - Gogo::mark_fndecl_as_builtin_library(*pdecl); - go_preserve_from_gc(*pdecl); - } - - tree fnptr = build_fold_addr_expr(*pdecl); - if (CAN_HAVE_LOCATION_P(fnptr)) - SET_EXPR_LOCATION(fnptr, location.gcc_location()); - - tree ret = build_call_array(rettype, fnptr, nargs, args); - SET_EXPR_LOCATION(ret, location.gcc_location()); - - delete[] types; - delete[] args; - - return ret; -} diff --git a/gcc/go/gofrontend/gogo.cc b/gcc/go/gofrontend/gogo.cc index c6ff9886090..995a4f2d2da 100644 --- a/gcc/go/gofrontend/gogo.cc +++ b/gcc/go/gofrontend/gogo.cc @@ -4855,6 +4855,66 @@ Function::get_or_make_decl(Gogo* gogo, Named_object* no) return this->fndecl_; } +// Get the backend representation. + +Bfunction* +Function_declaration::get_or_make_decl(Gogo* gogo, Named_object* no) +{ + if (this->fndecl_ == NULL) + { + // Let Go code use an asm declaration to pick up a builtin + // function. + if (!this->asm_name_.empty()) + { + Bfunction* builtin_decl = + gogo->backend()->lookup_builtin(this->asm_name_); + if (builtin_decl != NULL) + { + this->fndecl_ = builtin_decl; + return this->fndecl_; + } + } + + std::string asm_name; + if (this->asm_name_.empty()) + { + asm_name = (no->package() == NULL + ? gogo->pkgpath_symbol() + : no->package()->pkgpath_symbol()); + asm_name.append(1, '.'); + asm_name.append(Gogo::unpack_hidden_name(no->name())); + if (this->fntype_->is_method()) + { + asm_name.append(1, '.'); + Type* rtype = this->fntype_->receiver()->type(); + asm_name.append(rtype->mangled_name(gogo)); + } + } + + Btype* functype = this->fntype_->get_backend_fntype(gogo); + this->fndecl_ = + gogo->backend()->function(functype, no->get_id(gogo), asm_name, + true, true, true, false, false, + this->location()); + } + + return this->fndecl_; +} + +// Build the descriptor for a function declaration. This won't +// necessarily happen if the package has just a declaration for the +// function and no other reference to it, but we may still need the +// descriptor for references from other packages. +void +Function_declaration::build_backend_descriptor(Gogo* gogo) +{ + if (this->descriptor_ != NULL) + { + Translate_context context(gogo, NULL, NULL, NULL); + this->descriptor_->get_tree(&context); + } +} + // Return the function's decl after it has been built. Bfunction* diff --git a/gcc/go/gofrontend/gogo.h b/gcc/go/gofrontend/gogo.h index 0be81b2aafe..37cbbdf4411 100644 --- a/gcc/go/gofrontend/gogo.h +++ b/gcc/go/gofrontend/gogo.h @@ -575,35 +575,10 @@ class Gogo void write_globals(); - // Create trees for implicit builtin functions. - void - define_builtin_function_trees(); - - // Build a call to a builtin function. PDECL should point to a NULL - // initialized static pointer which will hold the fndecl. NAME is - // the name of the function. NARGS is the number of arguments. - // RETTYPE is the return type. It is followed by NARGS pairs of - // type and argument (both trees). - static tree - call_builtin(tree* pdecl, Location, const char* name, int nargs, - tree rettype, ...); - // Build a call to the runtime error function. Expression* runtime_error(int code, Location); - // Mark a function declaration as a builtin library function. - static void - mark_fndecl_as_builtin_library(tree fndecl); - - // Build a constructor for a slice. SLICE_TYPE_TREE is the type of - // the slice. VALUES points to the values. COUNT is the size, - // CAPACITY is the capacity. If CAPACITY is NULL, it is set to - // COUNT. - static tree - slice_constructor(tree slice_type_tree, tree values, tree count, - tree capacity); - // Build required interface method tables. void build_interface_method_tables(); diff --git a/gcc/go/gofrontend/import-archive.cc b/gcc/go/gofrontend/import-archive.cc index 9a1d5b3d7ef..34fb528ab5c 100644 --- a/gcc/go/gofrontend/import-archive.cc +++ b/gcc/go/gofrontend/import-archive.cc @@ -261,7 +261,7 @@ Archive_file::interpret_header(const Archive_header* hdr, off_t off, char size_string[size_string_size + 1]; memcpy(size_string, hdr->ar_size, size_string_size); char* ps = size_string + size_string_size; - while (ps[-1] == ' ') + while (ps > size_string && ps[-1] == ' ') --ps; *ps = '\0'; diff --git a/gcc/graphite-scop-detection.c b/gcc/graphite-scop-detection.c index 635e21a8519..5d1c96e353d 100644 --- a/gcc/graphite-scop-detection.c +++ b/gcc/graphite-scop-detection.c @@ -1056,7 +1056,7 @@ create_sese_edges (vec<sd_region> regions) #ifdef ENABLE_CHECKING verify_loop_structure (); - verify_ssa (false); + verify_ssa (false, true); #endif } diff --git a/gcc/graphite-sese-to-poly.c b/gcc/graphite-sese-to-poly.c index 3ad80410f9f..0bc443302c6 100644 --- a/gcc/graphite-sese-to-poly.c +++ b/gcc/graphite-sese-to-poly.c @@ -2469,7 +2469,7 @@ rewrite_cross_bb_scalar_deps (scop_p scop, gimple_stmt_iterator *gsi) gsi_next (gsi); } - rewrite_cross_bb_scalar_dependence (scop, zero_dim_array, + rewrite_cross_bb_scalar_dependence (scop, unshare_expr (zero_dim_array), def, use_stmt); } diff --git a/gcc/lto/ChangeLog b/gcc/lto/ChangeLog index f1ecec9e027..edb5bb9cfca 100644 --- a/gcc/lto/ChangeLog +++ b/gcc/lto/ChangeLog @@ -1,3 +1,8 @@ +2014-04-28 Richard Biener <rguenther@suse.de> + + PR middle-end/60092 + * lto-lang.c: Define flag_isoc11. + 2014-04-23 David Malcolm <dmalcolm@redhat.com> * lto-partition.c (add_references_to_partition): Update for diff --git a/gcc/lto/lto-lang.c b/gcc/lto/lto-lang.c index 1c00ec65be7..e213e923318 100644 --- a/gcc/lto/lto-lang.c +++ b/gcc/lto/lto-lang.c @@ -192,6 +192,7 @@ static GTY(()) tree signed_size_type_node; /* Flags needed to process builtins.def. */ int flag_isoc94; int flag_isoc99; +int flag_isoc11; /* Attribute handlers. */ diff --git a/gcc/opts.c b/gcc/opts.c index 1873b96a028..3c214f09701 100644 --- a/gcc/opts.c +++ b/gcc/opts.c @@ -1461,6 +1461,8 @@ common_handle_option (struct gcc_options *opts, sizeof "signed-integer-overflow" -1 }, { "bool", SANITIZE_BOOL, sizeof "bool" - 1 }, { "enum", SANITIZE_ENUM, sizeof "enum" - 1 }, + { "float-divide-by-zero", SANITIZE_FLOAT_DIVIDE, + sizeof "float-divide-by-zero" - 1 }, { NULL, 0, 0 } }; const char *comma; diff --git a/gcc/passes.c b/gcc/passes.c index c0a76d62d21..dbff5875a5d 100644 --- a/gcc/passes.c +++ b/gcc/passes.c @@ -1716,6 +1716,7 @@ pass_manager::dump_profile_report () const static void execute_function_todo (function *fn, void *data) { + bool from_ipa_pass = (cfun == NULL); unsigned int flags = (size_t)data; flags &= ~fn->last_verified; if (!flags) @@ -1761,32 +1762,56 @@ execute_function_todo (function *fn, void *data) rebuild_cgraph_edges (); /* If we've seen errors do not bother running any verifiers. */ - if (seen_error ()) + if (!seen_error ()) { - pop_cfun (); - return; - } - #if defined ENABLE_CHECKING - if (flags & TODO_verify_ssa - || (current_loops && loops_state_satisfies_p (LOOP_CLOSED_SSA))) - { - verify_gimple_in_cfg (cfun); - verify_ssa (true); - } - else if (flags & TODO_verify_stmts) - verify_gimple_in_cfg (cfun); - if (flags & TODO_verify_flow) - verify_flow_info (); - if (current_loops && loops_state_satisfies_p (LOOP_CLOSED_SSA)) - verify_loop_closed_ssa (false); - if (flags & TODO_verify_rtl_sharing) - verify_rtl_sharing (); + dom_state pre_verify_state = dom_info_state (fn, CDI_DOMINATORS); + dom_state pre_verify_pstate = dom_info_state (fn, CDI_POST_DOMINATORS); + + if (flags & TODO_verify_il) + { + if (cfun->curr_properties & PROP_trees) + { + if (cfun->curr_properties & PROP_cfg) + /* IPA passes leave stmts to be fixed up, so make sure to + not verify stmts really throw. */ + verify_gimple_in_cfg (cfun, !from_ipa_pass); + else + verify_gimple_in_seq (gimple_body (cfun->decl)); + } + if (cfun->curr_properties & PROP_ssa) + /* IPA passes leave stmts to be fixed up, so make sure to + not verify SSA operands whose verifier will choke on that. */ + verify_ssa (true, !from_ipa_pass); + } + if (flags & TODO_verify_flow) + verify_flow_info (); + if (flags & TODO_verify_il) + { + if (current_loops + && loops_state_satisfies_p (LOOP_CLOSED_SSA)) + verify_loop_closed_ssa (false); + } + if (flags & TODO_verify_rtl_sharing) + verify_rtl_sharing (); + + /* Make sure verifiers don't change dominator state. */ + gcc_assert (dom_info_state (fn, CDI_DOMINATORS) == pre_verify_state); + gcc_assert (dom_info_state (fn, CDI_POST_DOMINATORS) == pre_verify_pstate); #endif + } fn->last_verified = flags & TODO_verify_all; pop_cfun (); + + /* For IPA passes make sure to release dominator info, it can be + computed by non-verifying TODOs. */ + if (from_ipa_pass) + { + free_dominance_info (fn, CDI_DOMINATORS); + free_dominance_info (fn, CDI_POST_DOMINATORS); + } } /* Perform all TODO actions. */ @@ -2167,7 +2192,7 @@ execute_one_pass (opt_pass *pass) check_profile_consistency (pass->static_pass_number, 0, true); /* Run post-pass cleanup and verification. */ - execute_todo (todo_after | pass->todo_flags_finish); + execute_todo (todo_after | pass->todo_flags_finish | TODO_verify_il); if (profile_report && cfun && (cfun->curr_properties & PROP_cfg)) check_profile_consistency (pass->static_pass_number, 1, true); diff --git a/gcc/testsuite/ChangeLog b/gcc/testsuite/ChangeLog index 261bb98a3a4..00ae0d12f54 100644 --- a/gcc/testsuite/ChangeLog +++ b/gcc/testsuite/ChangeLog @@ -1,3 +1,166 @@ +2014-04-30 Alan Lawrence <alan.lawrence@arm.com> + + * gcc.target/aarch64/vuzps32_1.c: Expect zip1/2 insn rather than uzp1/2. + * gcc.target/aarch64/vuzpu32_1.c: Likewise. + * gcc.target/aarch64/vuzpf32_1.c: Likewise. + +2014-04-30 Alan Lawrence <alan.lawrence@arm.com> + + * gcc.target/aarch64/simd/vuzpf32_1.c: New file. + * gcc.target/aarch64/simd/vuzpf32.x: New file. + * gcc.target/aarch64/simd/vuzpp16_1.c: New file. + * gcc.target/aarch64/simd/vuzpp16.x: New file. + * gcc.target/aarch64/simd/vuzpp8_1.c: New file. + * gcc.target/aarch64/simd/vuzpp8.x: New file. + * gcc.target/aarch64/simd/vuzpqf32_1.c: New file. + * gcc.target/aarch64/simd/vuzpqf32.x: New file. + * gcc.target/aarch64/simd/vuzpqp16_1.c: New file. + * gcc.target/aarch64/simd/vuzpqp16.x: New file. + * gcc.target/aarch64/simd/vuzpqp8_1.c: New file. + * gcc.target/aarch64/simd/vuzpqp8.x: New file. + * gcc.target/aarch64/simd/vuzpqs16_1.c: New file. + * gcc.target/aarch64/simd/vuzpqs16.x: New file. + * gcc.target/aarch64/simd/vuzpqs32_1.c: New file. + * gcc.target/aarch64/simd/vuzpqs32.x: New file. + * gcc.target/aarch64/simd/vuzpqs8_1.c: New file. + * gcc.target/aarch64/simd/vuzpqs8.x: New file. + * gcc.target/aarch64/simd/vuzpqu16_1.c: New file. + * gcc.target/aarch64/simd/vuzpqu16.x: New file. + * gcc.target/aarch64/simd/vuzpqu32_1.c: New file. + * gcc.target/aarch64/simd/vuzpqu32.x: New file. + * gcc.target/aarch64/simd/vuzpqu8_1.c: New file. + * gcc.target/aarch64/simd/vuzpqu8.x: New file. + * gcc.target/aarch64/simd/vuzps16_1.c: New file. + * gcc.target/aarch64/simd/vuzps16.x: New file. + * gcc.target/aarch64/simd/vuzps32_1.c: New file. + * gcc.target/aarch64/simd/vuzps32.x: New file. + * gcc.target/aarch64/simd/vuzps8_1.c: New file. + * gcc.target/aarch64/simd/vuzps8.x: New file. + * gcc.target/aarch64/simd/vuzpu16_1.c: New file. + * gcc.target/aarch64/simd/vuzpu16.x: New file. + * gcc.target/aarch64/simd/vuzpu32_1.c: New file. + * gcc.target/aarch64/simd/vuzpu32.x: New file. + * gcc.target/aarch64/simd/vuzpu8_1.c: New file. + * gcc.target/aarch64/simd/vuzpu8.x: New file. + +2014-04-30 Richard Biener <rguenther@suse.de> + + PR tree-optimization/48329 + * gfortran.dg/vect/pr48329.f90: New testcase. + +2014-04-30 Marek Polacek <polacek@redhat.com> + + * c-c++-common/ubsan/div-by-zero-5.c: Fix formatting. + * c-c++-common/ubsan/float-div-by-zero-1.c: New test. + +2014-04-30 Marek Polacek <polacek@redhat.com> + + PR c/60139 + * gcc.dg/pr60139.c: New test. + +2014-04-30 Marek Polacek <polacek@redhat.com> + + PR c/60351 + * gcc.dg/pr60351.c: New test. + +2013-04-29 Alan Lawrence <alan.lawrence@arm.com> + + * gcc.target/arm/simd/simd.exp: New file. + * gcc.target/arm/simd/vzipqf32_1.c: New file. + * gcc.target/arm/simd/vzipqp16_1.c: New file. + * gcc.target/arm/simd/vzipqp8_1.c: New file. + * gcc.target/arm/simd/vzipqs16_1.c: New file. + * gcc.target/arm/simd/vzipqs32_1.c: New file. + * gcc.target/arm/simd/vzipqs8_1.c: New file. + * gcc.target/arm/simd/vzipqu16_1.c: New file. + * gcc.target/arm/simd/vzipqu32_1.c: New file. + * gcc.target/arm/simd/vzipqu8_1.c: New file. + * gcc.target/arm/simd/vzipf32_1.c: New file. + * gcc.target/arm/simd/vzipp16_1.c: New file. + * gcc.target/arm/simd/vzipp8_1.c: New file. + * gcc.target/arm/simd/vzips16_1.c: New file. + * gcc.target/arm/simd/vzips32_1.c: New file. + * gcc.target/arm/simd/vzips8_1.c: New file. + * gcc.target/arm/simd/vzipu16_1.c: New file. + * gcc.target/arm/simd/vzipu32_1.c: New file. + * gcc.target/arm/simd/vzipu8_1.c: New file. + +2014-04-29 Paolo Carlini <paolo.carlini@oracle.com> + + PR c++/51707 + * g++.dg/cpp0x/constexpr-51707.C: New. + +2014-04-29 David Malcolm <dmalcolm@redhat.com> + + * gcc.dg/tree-ssa/pr23401.c: Update the expected number of + occurrences of "int" in the gimple dump to reflect that the return + types of functions now show up in such dumps. + * gcc.dg/tree-ssa/pr27810.c: Likewise. + +2014-04-29 Jakub Jelinek <jakub@redhat.com> + + PR tree-optimization/60971 + * c-c++-common/turtore/pr60971.c: New test. + +2014-04-29 Alan Lawrence <alan.lawrence@arm.com> + + * gcc.target/aarch64/simd/simd.exp: New file. + * gcc.target/aarch64/simd/vzipf32_1.c: New file. + * gcc.target/aarch64/simd/vzipf32.x: New file. + * gcc.target/aarch64/simd/vzipp16_1.c: New file. + * gcc.target/aarch64/simd/vzipp16.x: New file. + * gcc.target/aarch64/simd/vzipp8_1.c: New file. + * gcc.target/aarch64/simd/vzipp8.x: New file. + * gcc.target/aarch64/simd/vzipqf32_1.c: New file. + * gcc.target/aarch64/simd/vzipqf32.x: New file. + * gcc.target/aarch64/simd/vzipqp16_1.c: New file. + * gcc.target/aarch64/simd/vzipqp16.x: New file. + * gcc.target/aarch64/simd/vzipqp8_1.c: New file. + * gcc.target/aarch64/simd/vzipqp8.x: New file. + * gcc.target/aarch64/simd/vzipqs16_1.c: New file. + * gcc.target/aarch64/simd/vzipqs16.x: New file. + * gcc.target/aarch64/simd/vzipqs32_1.c: New file. + * gcc.target/aarch64/simd/vzipqs32.x: New file. + * gcc.target/aarch64/simd/vzipqs8_1.c: New file. + * gcc.target/aarch64/simd/vzipqs8.x: New file. + * gcc.target/aarch64/simd/vzipqu16_1.c: New file. + * gcc.target/aarch64/simd/vzipqu16.x: New file. + * gcc.target/aarch64/simd/vzipqu32_1.c: New file. + * gcc.target/aarch64/simd/vzipqu32.x: New file. + * gcc.target/aarch64/simd/vzipqu8_1.c: New file. + * gcc.target/aarch64/simd/vzipqu8.x: New file. + * gcc.target/aarch64/simd/vzips16_1.c: New file. + * gcc.target/aarch64/simd/vzips16.x: New file. + * gcc.target/aarch64/simd/vzips32_1.c: New file. + * gcc.target/aarch64/simd/vzips32.x: New file. + * gcc.target/aarch64/simd/vzips8_1.c: New file. + * gcc.target/aarch64/simd/vzips8.x: New file. + * gcc.target/aarch64/simd/vzipu16_1.c: New file. + * gcc.target/aarch64/simd/vzipu16.x: New file. + * gcc.target/aarch64/simd/vzipu32_1.c: New file. + * gcc.target/aarch64/simd/vzipu32.x: New file. + * gcc.target/aarch64/simd/vzipu8_1.c: New file. + * gcc.target/aarch64/simd/vzipu8.x: New file. + +2014-04-29 Zhenqiang Chen <zhenqiang.chen@linaro.org> + + * gcc.target/aarch64/fcsel_1.c: New test case. + +2014-04-28 Jerry DeLisle <jvdelisle@gcc.gnu> + + PR libfortran/60810 + * gfortran.dg/arrayio_13.f90: New test. + +2014-04-28 Martin Jambor <mjambor@suse.cz> + + * gcc.dg/tree-ssa/sra-14.c: New test. + +2014-04-28 Richard Biener <rguenther@suse.de> + + PR middle-end/60092 + * gcc.dg/tree-ssa/alias-32.c: New testcase. + * gcc.dg/vect/pr60092.c: Likewise. + 2014-04-28 Richard Biener <rguenther@suse.de> * gcc.dg/tree-ssa/vrp91.c: New testcase. diff --git a/gcc/testsuite/c-c++-common/torture/pr60971.c b/gcc/testsuite/c-c++-common/torture/pr60971.c new file mode 100644 index 00000000000..b7a967dabb4 --- /dev/null +++ b/gcc/testsuite/c-c++-common/torture/pr60971.c @@ -0,0 +1,34 @@ +/* PR tree-optimization/60971 */ +/* { dg-do run } */ + +#ifndef __cplusplus +#define bool _Bool +#endif + +volatile unsigned char c; + +__attribute__((noinline)) unsigned char +foo (void) +{ + return c; +} + +__attribute__((noinline)) bool +bar (void) +{ + return foo () & 1; +} + +int +main () +{ + c = 0x41; + c = bar (); + if (c != 1) + __builtin_abort (); + c = 0x20; + c = bar (); + if (c != 0) + __builtin_abort (); + return 0; +} diff --git a/gcc/testsuite/c-c++-common/ubsan/div-by-zero-5.c b/gcc/testsuite/c-c++-common/ubsan/div-by-zero-5.c index 7a28bacd14b..bb391c5b36d 100644 --- a/gcc/testsuite/c-c++-common/ubsan/div-by-zero-5.c +++ b/gcc/testsuite/c-c++-common/ubsan/div-by-zero-5.c @@ -1,4 +1,4 @@ -/* { dg-do compile} */ +/* { dg-do compile } */ /* { dg-options "-fsanitize=integer-divide-by-zero" } */ void diff --git a/gcc/testsuite/c-c++-common/ubsan/float-div-by-zero-1.c b/gcc/testsuite/c-c++-common/ubsan/float-div-by-zero-1.c new file mode 100644 index 00000000000..2271ea9b776 --- /dev/null +++ b/gcc/testsuite/c-c++-common/ubsan/float-div-by-zero-1.c @@ -0,0 +1,26 @@ +/* { dg-do run } */ +/* { dg-options "-fsanitize=float-divide-by-zero" } */ + +int +main (void) +{ + volatile float a = 1.3f; + volatile double b = 0.0; + volatile int c = 4; + volatile float res; + + res = a / b; + res = a / 0.0; + res = 2.7f / b; + res = 3.6 / (b = 0.0, b); + res = c / b; + res = b / c; + + return 0; +} + +/* { dg-output "division by zero\[^\n\r]*(\n|\r\n|\r)" } */ +/* { dg-output "\[^\n\r]*division by zero\[^\n\r]*(\n|\r\n|\r)" } */ +/* { dg-output "\[^\n\r]*division by zero\[^\n\r]*(\n|\r\n|\r)" } */ +/* { dg-output "\[^\n\r]*division by zero\[^\n\r]*(\n|\r\n|\r)" } */ +/* { dg-output "\[^\n\r]*division by zero\[^\n\r]*" } */ diff --git a/gcc/testsuite/g++.dg/cpp0x/constexpr-51707.C b/gcc/testsuite/g++.dg/cpp0x/constexpr-51707.C new file mode 100644 index 00000000000..ae02a31c540 --- /dev/null +++ b/gcc/testsuite/g++.dg/cpp0x/constexpr-51707.C @@ -0,0 +1,14 @@ +// PR c++/51707 +// { dg-do compile { target c++11 } } + +struct S { + constexpr S() {} +}; + +struct T { + constexpr T(S const& s) : s{s} {} + S const& s; +}; + +constexpr S s {}; +constexpr T t { s }; diff --git a/gcc/testsuite/g++.dg/cpp0x/constexpr-aggr1.C b/gcc/testsuite/g++.dg/cpp0x/constexpr-aggr1.C new file mode 100644 index 00000000000..7e4da11a2df --- /dev/null +++ b/gcc/testsuite/g++.dg/cpp0x/constexpr-aggr1.C @@ -0,0 +1,17 @@ +// PR c++/60951 +// { dg-do compile { target c++11 } } + +struct Foo { + constexpr Foo(int x = 0) : memb(x) {} + int memb; +}; + +struct FooContainer { + Foo foo[2]; +}; + +void fubar() { + int nonConst = 0; + FooContainer fooContainer; + fooContainer = { { 0, nonConst } }; +} diff --git a/gcc/testsuite/g++.dg/cpp0x/defaulted49.C b/gcc/testsuite/g++.dg/cpp0x/defaulted49.C new file mode 100644 index 00000000000..357be419db7 --- /dev/null +++ b/gcc/testsuite/g++.dg/cpp0x/defaulted49.C @@ -0,0 +1,15 @@ +// PR c++/60980 +// { dg-do compile { target c++11 } } + +struct x0 +{ + x0 () = default; +}; +struct x1 +{ + x0 x2[2]; + void x3 () + { + x1 (); + } +}; diff --git a/gcc/testsuite/g++.dg/cpp0x/nsdmi-defer6.C b/gcc/testsuite/g++.dg/cpp0x/nsdmi-defer6.C index 033c14264dd..0f06343958b 100644 --- a/gcc/testsuite/g++.dg/cpp0x/nsdmi-defer6.C +++ b/gcc/testsuite/g++.dg/cpp0x/nsdmi-defer6.C @@ -1,8 +1,8 @@ // { dg-do compile { target c++11 } } -struct A // { dg-error "non-static data member" } +struct A { - int i = (A(), 42); // { dg-message "required here" } + int i = (A(), 42); // { dg-error "constructor required" } }; A a; diff --git a/gcc/testsuite/g++.dg/cpp0x/nsdmi-dr1397.C b/gcc/testsuite/g++.dg/cpp0x/nsdmi-dr1397.C new file mode 100644 index 00000000000..061af8b8c29 --- /dev/null +++ b/gcc/testsuite/g++.dg/cpp0x/nsdmi-dr1397.C @@ -0,0 +1,7 @@ +// DR 1397 +// { dg-require-effective-target c++11 } + +struct A +{ + int i = sizeof(A{}); // { dg-error "" } +}; diff --git a/gcc/testsuite/g++.dg/cpp0x/nsdmi-eh1.C b/gcc/testsuite/g++.dg/cpp0x/nsdmi-eh1.C index edcf5887db1..9bc632c4bc7 100644 --- a/gcc/testsuite/g++.dg/cpp0x/nsdmi-eh1.C +++ b/gcc/testsuite/g++.dg/cpp0x/nsdmi-eh1.C @@ -1,5 +1,5 @@ // Core issue 1351 -// { dg-do run { xfail *-*-* } } +// { dg-do run } // { dg-require-effective-target c++11 } bool fail; diff --git a/gcc/testsuite/gcc.dg/pr60139.c b/gcc/testsuite/gcc.dg/pr60139.c new file mode 100644 index 00000000000..a63d8b5b9bd --- /dev/null +++ b/gcc/testsuite/gcc.dg/pr60139.c @@ -0,0 +1,14 @@ +/* PR c/60139 */ +/* { dg-do compile } */ +/* { dg-options "-Wpedantic" } */ +/* { dg-prune-output ".*near initialization for.*" } */ + +double sin (double); +void +fn (int *p) +{ + int **a[] = { &p, /* { dg-warning "17:initializer element is not computable at load time" } */ + (void *) 0, &p }; /* { dg-warning "28:initializer element is not computable at load time" } */ + double d[] = { sin (1.0), /* { dg-warning "18:initializer element is not a constant expression" } */ + 8.8, sin (1.0), 2.6 }; /* { dg-warning "23:initializer element is not a constant expression" } */ +} diff --git a/gcc/testsuite/gcc.dg/pr60351.c b/gcc/testsuite/gcc.dg/pr60351.c new file mode 100644 index 00000000000..29184d94872 --- /dev/null +++ b/gcc/testsuite/gcc.dg/pr60351.c @@ -0,0 +1,11 @@ +/* PR c/60351 */ +/* { dg-do compile } */ + +void +f (int i) +{ + i >> -1; /* { dg-warning "5:right shift count is negative" } */ + i >> 250; /* { dg-warning "5:right shift count >= width of type" } */ + i << -1; /* { dg-warning "5:left shift count is negative" } */ + i << 250; /* { dg-warning "5:left shift count >= width of type" } */ +} diff --git a/gcc/testsuite/gcc.dg/tree-ssa/alias-32.c b/gcc/testsuite/gcc.dg/tree-ssa/alias-32.c new file mode 100644 index 00000000000..5d0dcc2937e --- /dev/null +++ b/gcc/testsuite/gcc.dg/tree-ssa/alias-32.c @@ -0,0 +1,20 @@ +/* { dg-do compile } */ +/* { dg-options "-O2 -fdump-tree-cddce1" } */ + +int bar (short *p) +{ + int res = *p; + struct { int *q1; int *q2; } q; + q.q1 = __builtin_aligned_alloc (128, 128 * sizeof (int)); + q.q2 = __builtin_aligned_alloc (128, 128 * sizeof (int)); + *q.q1 = 1; + *q.q2 = 2; + return res + *p + *q.q1 + *q.q2; +} + +/* There should be only one load from *p left. All stores and all + other loads should be removed. Likewise the calls to aligned_alloc. */ + +/* { dg-final { scan-tree-dump-times "\\\*\[^ \]" 1 "cddce1" } } */ +/* { dg-final { scan-tree-dump-not "aligned_alloc" "cddce1" } } */ +/* { dg-final { cleanup-tree-dump "cddce1" } } */ diff --git a/gcc/testsuite/gcc.dg/tree-ssa/pr23401.c b/gcc/testsuite/gcc.dg/tree-ssa/pr23401.c index 1d30ac7519f..3940692cd6c 100644 --- a/gcc/testsuite/gcc.dg/tree-ssa/pr23401.c +++ b/gcc/testsuite/gcc.dg/tree-ssa/pr23401.c @@ -19,6 +19,6 @@ int ffff(int i) /* We should not use extra temporaries apart from for i1 + i2. */ -/* { dg-final { scan-tree-dump-times "int" 5 "gimple" } } */ +/* { dg-final { scan-tree-dump-times "int" 6 "gimple" } } */ /* { dg-final { scan-tree-dump-times "int D\\\." 1 "gimple" } } */ /* { dg-final { cleanup-tree-dump "gimple" } } */ diff --git a/gcc/testsuite/gcc.dg/tree-ssa/pr27810.c b/gcc/testsuite/gcc.dg/tree-ssa/pr27810.c index c7da3bd5d06..6d0904bb06b 100644 --- a/gcc/testsuite/gcc.dg/tree-ssa/pr27810.c +++ b/gcc/testsuite/gcc.dg/tree-ssa/pr27810.c @@ -13,6 +13,6 @@ int qqq (int a) /* We should not use an extra temporary for the result of the function call. */ -/* { dg-final { scan-tree-dump-times "int" 3 "gimple" } } */ +/* { dg-final { scan-tree-dump-times "int" 4 "gimple" } } */ /* { dg-final { scan-tree-dump-times "int D\\\." 1 "gimple" } } */ /* { dg-final { cleanup-tree-dump "gimple" } } */ diff --git a/gcc/testsuite/gcc.dg/tree-ssa/sra-14.c b/gcc/testsuite/gcc.dg/tree-ssa/sra-14.c new file mode 100644 index 00000000000..6cbc0b43d58 --- /dev/null +++ b/gcc/testsuite/gcc.dg/tree-ssa/sra-14.c @@ -0,0 +1,70 @@ +/* { dg-do run } */ +/* { dg-options "-O1" } */ + +struct S +{ + int i, j; +}; + +struct Z +{ + struct S d, s; +}; + +struct S __attribute__ ((noinline, noclone)) +get_s (void) +{ + struct S s; + s.i = 5; + s.j = 6; + + return s; +} + +struct S __attribute__ ((noinline, noclone)) +get_d (void) +{ + struct S d; + d.i = 0; + d.j = 0; + + return d; +} + +int __attribute__ ((noinline, noclone)) +get_c (void) +{ + return 1; +} + +int __attribute__ ((noinline, noclone)) +my_nop (int i) +{ + return i; +} + +int __attribute__ ((noinline, noclone)) +foo (void) +{ + struct Z z; + int i, c = get_c (); + + z.d = get_d (); + z.s = get_s (); + + for (i = 0; i < c; i++) + { + z.s.i = my_nop (z.s.i); + z.s.j = my_nop (z.s.j); + } + + return z.s.i + z.s.j; +} + +int main (int argc, char *argv[]) +{ + if (foo () != 11) + __builtin_abort (); + return 0; +} + diff --git a/gcc/testsuite/gcc.dg/vect/pr60092.c b/gcc/testsuite/gcc.dg/vect/pr60092.c new file mode 100644 index 00000000000..e03c625b489 --- /dev/null +++ b/gcc/testsuite/gcc.dg/vect/pr60092.c @@ -0,0 +1,18 @@ +/* { dg-do compile } */ +/* { dg-require-effective-target vect_int } */ + +int *foo (int n) +{ + int *p = __builtin_aligned_alloc (256, n * sizeof (int)); + int *q = __builtin_aligned_alloc (256, n * sizeof (int)); + bar (q); + int i; + for (i = 0; i < n; ++i) + p[i] = q[i] + q[i]; + return p; +} + +/* { dg-final { scan-tree-dump "LOOP VECTORIZED" "vect" } } */ +/* { dg-final { scan-tree-dump-not "Peeling for alignment will be applied" "vect" } } */ +/* { dg-final { scan-tree-dump-not "Vectorizing an unaligned access" "vect" } } */ +/* { dg-final { cleanup-tree-dump "vect" } } */ diff --git a/gcc/testsuite/gcc.target/aarch64/fcsel_1.c b/gcc/testsuite/gcc.target/aarch64/fcsel_1.c new file mode 100644 index 00000000000..2704ee0ede7 --- /dev/null +++ b/gcc/testsuite/gcc.target/aarch64/fcsel_1.c @@ -0,0 +1,22 @@ +/* { dg-do compile } */ +/* { dg-options " -O2 " } */ + +float +f_1 (float a, float b, float c, float d) +{ + if (a > 0.0) + return c; + else + return 2.0; +} + +double +f_2 (double a, double b, double c, double d) +{ + if (a > b) + return c; + else + return d; +} + +/* { dg-final { scan-assembler-times "\tfcsel" 2 } } */ diff --git a/gcc/testsuite/gcc.target/aarch64/simd/simd.exp b/gcc/testsuite/gcc.target/aarch64/simd/simd.exp new file mode 100644 index 00000000000..097d29a9e13 --- /dev/null +++ b/gcc/testsuite/gcc.target/aarch64/simd/simd.exp @@ -0,0 +1,45 @@ +# Specific regression driver for AArch64 SIMD instructions. +# Copyright (C) 2014 Free Software Foundation, Inc. +# Contributed by ARM Ltd. +# +# 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/>. */ + +# GCC testsuite that uses the `dg.exp' driver. + +# Exit immediately if this isn't an AArch64 target. +if {![istarget aarch64*-*-*] } then { + return +} + +# Load support procs. +load_lib gcc-dg.exp + +# If a testcase doesn't have special options, use these. +global DEFAULT_CFLAGS +if ![info exists DEFAULT_CFLAGS] then { + set DEFAULT_CFLAGS " -ansi -pedantic-errors" +} + +# Initialize `dg'. +dg-init + +# Main loop. +dg-runtest [lsort [glob -nocomplain $srcdir/$subdir/*.\[cCS\]]] \ + "" $DEFAULT_CFLAGS + +# All done. +dg-finish diff --git a/gcc/testsuite/gcc.target/aarch64/simd/vuzpf32.x b/gcc/testsuite/gcc.target/aarch64/simd/vuzpf32.x new file mode 100644 index 00000000000..86c3700e522 --- /dev/null +++ b/gcc/testsuite/gcc.target/aarch64/simd/vuzpf32.x @@ -0,0 +1,26 @@ +extern void abort (void); + +float32x2x2_t +test_vuzpf32 (float32x2_t _a, float32x2_t _b) +{ + return vuzp_f32 (_a, _b); +} + +int +main (int argc, char **argv) +{ + int i; + float32_t first[] = {1, 2}; + float32_t second[] = {3, 4}; + float32x2x2_t result = test_vuzpf32 (vld1_f32 (first), vld1_f32 (second)); + float32_t exp1[] = {1, 3}; + float32_t exp2[] = {2, 4}; + float32x2_t expect1 = vld1_f32 (exp1); + float32x2_t expect2 = vld1_f32 (exp2); + + for (i = 0; i < 2; i++) + if ((result.val[0][i] != expect1[i]) || (result.val[1][i] != expect2[i])) + abort (); + + return 0; +} diff --git a/gcc/testsuite/gcc.target/aarch64/simd/vuzpf32_1.c b/gcc/testsuite/gcc.target/aarch64/simd/vuzpf32_1.c new file mode 100644 index 00000000000..0daba1c7f93 --- /dev/null +++ b/gcc/testsuite/gcc.target/aarch64/simd/vuzpf32_1.c @@ -0,0 +1,11 @@ +/* Test the `vuzp_f32' AArch64 SIMD intrinsic. */ + +/* { dg-do run } */ +/* { dg-options "-save-temps -fno-inline" } */ + +#include <arm_neon.h> +#include "vuzpf32.x" + +/* { dg-final { scan-assembler-times "zip1\[ \t\]+v\[0-9\]+\.2s, ?v\[0-9\]+\.2s, ?v\[0-9\]+\.2s!?\(?:\[ \t\]+@\[a-zA-Z0-9 \]+\)?\n" 1 } } */ +/* { dg-final { scan-assembler-times "zip2\[ \t\]+v\[0-9\]+\.2s, ?v\[0-9\]+\.2s, ?v\[0-9\]+\.2s!?\(?:\[ \t\]+@\[a-zA-Z0-9 \]+\)?\n" 1 } } */ +/* { dg-final { cleanup-saved-temps } } */ diff --git a/gcc/testsuite/gcc.target/aarch64/simd/vuzpp16.x b/gcc/testsuite/gcc.target/aarch64/simd/vuzpp16.x new file mode 100644 index 00000000000..bc45efcd965 --- /dev/null +++ b/gcc/testsuite/gcc.target/aarch64/simd/vuzpp16.x @@ -0,0 +1,26 @@ +extern void abort (void); + +poly16x4x2_t +test_vuzpp16 (poly16x4_t _a, poly16x4_t _b) +{ + return vuzp_p16 (_a, _b); +} + +int +main (int argc, char **argv) +{ + int i; + poly16_t first[] = {1, 2, 3, 4}; + poly16_t second[] = {5, 6, 7, 8}; + poly16x4x2_t result = test_vuzpp16 (vld1_p16 (first), vld1_p16 (second)); + poly16_t exp1[] = {1, 3, 5, 7}; + poly16_t exp2[] = {2, 4, 6, 8}; + poly16x4_t expect1 = vld1_p16 (exp1); + poly16x4_t expect2 = vld1_p16 (exp2); + + for (i = 0; i < 4; i++) + if ((result.val[0][i] != expect1[i]) || (result.val[1][i] != expect2[i])) + abort (); + + return 0; +} diff --git a/gcc/testsuite/gcc.target/aarch64/simd/vuzpp16_1.c b/gcc/testsuite/gcc.target/aarch64/simd/vuzpp16_1.c new file mode 100644 index 00000000000..03b07220640 --- /dev/null +++ b/gcc/testsuite/gcc.target/aarch64/simd/vuzpp16_1.c @@ -0,0 +1,11 @@ +/* Test the `vuzp_p16' AArch64 SIMD intrinsic. */ + +/* { dg-do run } */ +/* { dg-options "-save-temps -fno-inline" } */ + +#include <arm_neon.h> +#include "vuzpp16.x" + +/* { dg-final { scan-assembler-times "uzp1\[ \t\]+v\[0-9\]+\.4h, ?v\[0-9\]+\.4h, ?v\[0-9\]+\.4h!?\(?:\[ \t\]+@\[a-zA-Z0-9 \]+\)?\n" 1 } } */ +/* { dg-final { scan-assembler-times "uzp2\[ \t\]+v\[0-9\]+\.4h, ?v\[0-9\]+\.4h, ?v\[0-9\]+\.4h!?\(?:\[ \t\]+@\[a-zA-Z0-9 \]+\)?\n" 1 } } */ +/* { dg-final { cleanup-saved-temps } } */ diff --git a/gcc/testsuite/gcc.target/aarch64/simd/vuzpp8.x b/gcc/testsuite/gcc.target/aarch64/simd/vuzpp8.x new file mode 100644 index 00000000000..b4ef51cae74 --- /dev/null +++ b/gcc/testsuite/gcc.target/aarch64/simd/vuzpp8.x @@ -0,0 +1,26 @@ +extern void abort (void); + +poly8x8x2_t +test_vuzpp8 (poly8x8_t _a, poly8x8_t _b) +{ + return vuzp_p8 (_a, _b); +} + +int +main (int argc, char **argv) +{ + int i; + poly8_t first[] = {1, 2, 3, 4, 5, 6, 7, 8}; + poly8_t second[] = {9, 10, 11, 12, 13, 14, 15, 16}; + poly8x8x2_t result = test_vuzpp8 (vld1_p8 (first), vld1_p8 (second)); + poly8_t exp1[] = {1, 3, 5, 7, 9, 11, 13, 15}; + poly8_t exp2[] = {2, 4, 6, 8, 10, 12, 14, 16}; + poly8x8_t expect1 = vld1_p8 (exp1); + poly8x8_t expect2 = vld1_p8 (exp2); + + for (i = 0; i < 8; i++) + if ((result.val[0][i] != expect1[i]) || (result.val[1][i] != expect2[i])) + abort (); + + return 0; +} diff --git a/gcc/testsuite/gcc.target/aarch64/simd/vuzpp8_1.c b/gcc/testsuite/gcc.target/aarch64/simd/vuzpp8_1.c new file mode 100644 index 00000000000..5186b1f9166 --- /dev/null +++ b/gcc/testsuite/gcc.target/aarch64/simd/vuzpp8_1.c @@ -0,0 +1,11 @@ +/* Test the `vuzp_p8' AArch64 SIMD intrinsic. */ + +/* { dg-do run } */ +/* { dg-options "-save-temps -fno-inline" } */ + +#include <arm_neon.h> +#include "vuzpp8.x" + +/* { dg-final { scan-assembler-times "uzp1\[ \t\]+v\[0-9\]+\.8b, ?v\[0-9\]+\.8b, ?v\[0-9\]+\.8b!?\(?:\[ \t\]+@\[a-zA-Z0-9 \]+\)?\n" 1 } } */ +/* { dg-final { scan-assembler-times "uzp2\[ \t\]+v\[0-9\]+\.8b, ?v\[0-9\]+\.8b, ?v\[0-9\]+\.8b!?\(?:\[ \t\]+@\[a-zA-Z0-9 \]+\)?\n" 1 } } */ +/* { dg-final { cleanup-saved-temps } } */ diff --git a/gcc/testsuite/gcc.target/aarch64/simd/vuzpqf32.x b/gcc/testsuite/gcc.target/aarch64/simd/vuzpqf32.x new file mode 100644 index 00000000000..f1b48da315f --- /dev/null +++ b/gcc/testsuite/gcc.target/aarch64/simd/vuzpqf32.x @@ -0,0 +1,26 @@ +extern void abort (void); + +float32x4x2_t +test_vuzpqf32 (float32x4_t _a, float32x4_t _b) +{ + return vuzpq_f32 (_a, _b); +} + +int +main (int argc, char **argv) +{ + int i; + float32_t first[] = {1, 2, 3, 4}; + float32_t second[] = {5, 6, 7, 8}; + float32x4x2_t result = test_vuzpqf32 (vld1q_f32 (first), vld1q_f32 (second)); + float32_t exp1[] = {1, 3, 5, 7}; + float32_t exp2[] = {2, 4, 6, 8}; + float32x4_t expect1 = vld1q_f32 (exp1); + float32x4_t expect2 = vld1q_f32 (exp2); + + for (i = 0; i < 4; i++) + if ((result.val[0][i] != expect1[i]) || (result.val[1][i] != expect2[i])) + abort (); + + return 0; +} diff --git a/gcc/testsuite/gcc.target/aarch64/simd/vuzpqf32_1.c b/gcc/testsuite/gcc.target/aarch64/simd/vuzpqf32_1.c new file mode 100644 index 00000000000..1167f7bbe00 --- /dev/null +++ b/gcc/testsuite/gcc.target/aarch64/simd/vuzpqf32_1.c @@ -0,0 +1,11 @@ +/* Test the `vuzpq_f32' AArch64 SIMD intrinsic. */ + +/* { dg-do run } */ +/* { dg-options "-save-temps -fno-inline" } */ + +#include <arm_neon.h> +#include "vuzpqf32.x" + +/* { dg-final { scan-assembler-times "uzp1\[ \t\]+v\[0-9\]+\.4s, ?v\[0-9\]+\.4s, ?v\[0-9\]+\.4s!?\(?:\[ \t\]+@\[a-zA-Z0-9 \]+\)?\n" 1 } } */ +/* { dg-final { scan-assembler-times "uzp2\[ \t\]+v\[0-9\]+\.4s, ?v\[0-9\]+\.4s, ?v\[0-9\]+\.4s!?\(?:\[ \t\]+@\[a-zA-Z0-9 \]+\)?\n" 1 } } */ +/* { dg-final { cleanup-saved-temps } } */ diff --git a/gcc/testsuite/gcc.target/aarch64/simd/vuzpqp16.x b/gcc/testsuite/gcc.target/aarch64/simd/vuzpqp16.x new file mode 100644 index 00000000000..d4e08f74631 --- /dev/null +++ b/gcc/testsuite/gcc.target/aarch64/simd/vuzpqp16.x @@ -0,0 +1,26 @@ +extern void abort (void); + +poly16x8x2_t +test_vuzpqp16 (poly16x8_t _a, poly16x8_t _b) +{ + return vuzpq_p16 (_a, _b); +} + +int +main (int argc, char **argv) +{ + int i; + poly16_t first[] = {1, 2, 3, 4, 5, 6, 7, 8}; + poly16_t second[] = {9, 10, 11, 12, 13, 14, 15, 16}; + poly16x8x2_t result = test_vuzpqp16 (vld1q_p16 (first), vld1q_p16 (second)); + poly16_t exp1[] = {1, 3, 5, 7, 9, 11, 13, 15}; + poly16_t exp2[] = {2, 4, 6, 8, 10, 12, 14, 16}; + poly16x8_t expect1 = vld1q_p16 (exp1); + poly16x8_t expect2 = vld1q_p16 (exp2); + + for (i = 0; i < 8; i++) + if ((result.val[0][i] != expect1[i]) || (result.val[1][i] != expect2[i])) + abort (); + + return 0; +} diff --git a/gcc/testsuite/gcc.target/aarch64/simd/vuzpqp16_1.c b/gcc/testsuite/gcc.target/aarch64/simd/vuzpqp16_1.c new file mode 100644 index 00000000000..c6648045cf3 --- /dev/null +++ b/gcc/testsuite/gcc.target/aarch64/simd/vuzpqp16_1.c @@ -0,0 +1,11 @@ +/* Test the `vuzpq_p16' AArch64 SIMD intrinsic. */ + +/* { dg-do run } */ +/* { dg-options "-save-temps -fno-inline" } */ + +#include <arm_neon.h> +#include "vuzpqp16.x" + +/* { dg-final { scan-assembler-times "uzp1\[ \t\]+v\[0-9\]+\.8h, ?v\[0-9\]+\.8h, ?v\[0-9\]+\.8h!?\(?:\[ \t\]+@\[a-zA-Z0-9 \]+\)?\n" 1 } } */ +/* { dg-final { scan-assembler-times "uzp2\[ \t\]+v\[0-9\]+\.8h, ?v\[0-9\]+\.8h, ?v\[0-9\]+\.8h!?\(?:\[ \t\]+@\[a-zA-Z0-9 \]+\)?\n" 1 } } */ +/* { dg-final { cleanup-saved-temps } } */ diff --git a/gcc/testsuite/gcc.target/aarch64/simd/vuzpqp8.x b/gcc/testsuite/gcc.target/aarch64/simd/vuzpqp8.x new file mode 100644 index 00000000000..31541de7e14 --- /dev/null +++ b/gcc/testsuite/gcc.target/aarch64/simd/vuzpqp8.x @@ -0,0 +1,27 @@ +extern void abort (void); + +poly8x16x2_t +test_vuzpqp8 (poly8x16_t _a, poly8x16_t _b) +{ + return vuzpq_p8 (_a, _b); +} + +int +main (int argc, char **argv) +{ + int i; + poly8_t first[] = {1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16}; + poly8_t second[] = + {17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32}; + poly8x16x2_t result = test_vuzpqp8 (vld1q_p8 (first), vld1q_p8 (second)); + poly8_t exp1[] = {1, 3, 5, 7, 9, 11, 13, 15, 17, 19, 21, 23, 25, 27, 29, 31}; + poly8_t exp2[] = {2, 4, 6, 8, 10, 12, 14, 16, 18, 20, 22, 24, 26, 28, 30, 32}; + poly8x16_t expect1 = vld1q_p8 (exp1); + poly8x16_t expect2 = vld1q_p8 (exp2); + + for (i = 0; i < 16; i++) + if ((result.val[0][i] != expect1[i]) || (result.val[1][i] != expect2[i])) + abort (); + + return 0; +} diff --git a/gcc/testsuite/gcc.target/aarch64/simd/vuzpqp8_1.c b/gcc/testsuite/gcc.target/aarch64/simd/vuzpqp8_1.c new file mode 100644 index 00000000000..a9e6ce222e7 --- /dev/null +++ b/gcc/testsuite/gcc.target/aarch64/simd/vuzpqp8_1.c @@ -0,0 +1,11 @@ +/* Test the `vuzpq_p8' AArch64 SIMD intrinsic. */ + +/* { dg-do run } */ +/* { dg-options "-save-temps -fno-inline" } */ + +#include <arm_neon.h> +#include "vuzpqp8.x" + +/* { dg-final { scan-assembler-times "uzp1\[ \t\]+v\[0-9\]+\.16b, ?v\[0-9\]+\.16b, ?v\[0-9\]+\.16b!?\(?:\[ \t\]+@\[a-zA-Z0-9 \]+\)?\n" 1 } } */ +/* { dg-final { scan-assembler-times "uzp2\[ \t\]+v\[0-9\]+\.16b, ?v\[0-9\]+\.16b, ?v\[0-9\]+\.16b!?\(?:\[ \t\]+@\[a-zA-Z0-9 \]+\)?\n" 1 } } */ +/* { dg-final { cleanup-saved-temps } } */ diff --git a/gcc/testsuite/gcc.target/aarch64/simd/vuzpqs16.x b/gcc/testsuite/gcc.target/aarch64/simd/vuzpqs16.x new file mode 100644 index 00000000000..439107b2ec5 --- /dev/null +++ b/gcc/testsuite/gcc.target/aarch64/simd/vuzpqs16.x @@ -0,0 +1,26 @@ +extern void abort (void); + +int16x8x2_t +test_vuzpqs16 (int16x8_t _a, int16x8_t _b) +{ + return vuzpq_s16 (_a, _b); +} + +int +main (int argc, char **argv) +{ + int i; + int16_t first[] = {1, 2, 3, 4, 5, 6, 7, 8}; + int16_t second[] = {9, 10, 11, 12, 13, 14, 15, 16}; + int16x8x2_t result = test_vuzpqs16 (vld1q_s16 (first), vld1q_s16 (second)); + int16_t exp1[] = {1, 3, 5, 7, 9, 11, 13, 15}; + int16_t exp2[] = {2, 4, 6, 8, 10, 12, 14, 16}; + int16x8_t expect1 = vld1q_s16 (exp1); + int16x8_t expect2 = vld1q_s16 (exp2); + + for (i = 0; i < 8; i++) + if ((result.val[0][i] != expect1[i]) || (result.val[1][i] != expect2[i])) + abort (); + + return 0; +} diff --git a/gcc/testsuite/gcc.target/aarch64/simd/vuzpqs16_1.c b/gcc/testsuite/gcc.target/aarch64/simd/vuzpqs16_1.c new file mode 100644 index 00000000000..af1e28b7669 --- /dev/null +++ b/gcc/testsuite/gcc.target/aarch64/simd/vuzpqs16_1.c @@ -0,0 +1,11 @@ +/* Test the `vuzpq_s16' AArch64 SIMD intrinsic. */ + +/* { dg-do run } */ +/* { dg-options "-save-temps -fno-inline" } */ + +#include <arm_neon.h> +#include "vuzpqs16.x" + +/* { dg-final { scan-assembler-times "uzp1\[ \t\]+v\[0-9\]+\.8h, ?v\[0-9\]+\.8h, ?v\[0-9\]+\.8h!?\(?:\[ \t\]+@\[a-zA-Z0-9 \]+\)?\n" 1 } } */ +/* { dg-final { scan-assembler-times "uzp2\[ \t\]+v\[0-9\]+\.8h, ?v\[0-9\]+\.8h, ?v\[0-9\]+\.8h!?\(?:\[ \t\]+@\[a-zA-Z0-9 \]+\)?\n" 1 } } */ +/* { dg-final { cleanup-saved-temps } } */ diff --git a/gcc/testsuite/gcc.target/aarch64/simd/vuzpqs32.x b/gcc/testsuite/gcc.target/aarch64/simd/vuzpqs32.x new file mode 100644 index 00000000000..84463f038e5 --- /dev/null +++ b/gcc/testsuite/gcc.target/aarch64/simd/vuzpqs32.x @@ -0,0 +1,26 @@ +extern void abort (void); + +int32x4x2_t +test_vuzpqs32 (int32x4_t _a, int32x4_t _b) +{ + return vuzpq_s32 (_a, _b); +} + +int +main (int argc, char **argv) +{ + int i; + int32_t first[] = {1, 2, 3, 4}; + int32_t second[] = {5, 6, 7, 8}; + int32x4x2_t result = test_vuzpqs32 (vld1q_s32 (first), vld1q_s32 (second)); + int32_t exp1[] = {1, 3, 5, 7}; + int32_t exp2[] = {2, 4, 6, 8}; + int32x4_t expect1 = vld1q_s32 (exp1); + int32x4_t expect2 = vld1q_s32 (exp2); + + for (i = 0; i < 4; i++) + if ((result.val[0][i] != expect1[i]) || (result.val[1][i] != expect2[i])) + abort (); + + return 0; +} diff --git a/gcc/testsuite/gcc.target/aarch64/simd/vuzpqs32_1.c b/gcc/testsuite/gcc.target/aarch64/simd/vuzpqs32_1.c new file mode 100644 index 00000000000..a4bf7ac7e59 --- /dev/null +++ b/gcc/testsuite/gcc.target/aarch64/simd/vuzpqs32_1.c @@ -0,0 +1,11 @@ +/* Test the `vuzpq_s32' AArch64 SIMD intrinsic. */ + +/* { dg-do run } */ +/* { dg-options "-save-temps -fno-inline" } */ + +#include <arm_neon.h> +#include "vuzpqs32.x" + +/* { dg-final { scan-assembler-times "uzp1\[ \t\]+v\[0-9\]+\.4s, ?v\[0-9\]+\.4s, ?v\[0-9\]+\.4s!?\(?:\[ \t\]+@\[a-zA-Z0-9 \]+\)?\n" 1 } } */ +/* { dg-final { scan-assembler-times "uzp2\[ \t\]+v\[0-9\]+\.4s, ?v\[0-9\]+\.4s, ?v\[0-9\]+\.4s!?\(?:\[ \t\]+@\[a-zA-Z0-9 \]+\)?\n" 1 } } */ +/* { dg-final { cleanup-saved-temps } } */ diff --git a/gcc/testsuite/gcc.target/aarch64/simd/vuzpqs8.x b/gcc/testsuite/gcc.target/aarch64/simd/vuzpqs8.x new file mode 100644 index 00000000000..c8b916780d7 --- /dev/null +++ b/gcc/testsuite/gcc.target/aarch64/simd/vuzpqs8.x @@ -0,0 +1,27 @@ +extern void abort (void); + +int8x16x2_t +test_vuzpqs8 (int8x16_t _a, int8x16_t _b) +{ + return vuzpq_s8 (_a, _b); +} + +int +main (int argc, char **argv) +{ + int i; + int8_t first[] = {1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16}; + int8_t second[] = + {17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32}; + int8x16x2_t result = test_vuzpqs8 (vld1q_s8 (first), vld1q_s8 (second)); + int8_t exp1[] = {1, 3, 5, 7, 9, 11, 13, 15, 17, 19, 21, 23, 25, 27, 29, 31}; + int8_t exp2[] = {2, 4, 6, 8, 10, 12, 14, 16, 18, 20, 22, 24, 26, 28, 30, 32}; + int8x16_t expect1 = vld1q_s8 (exp1); + int8x16_t expect2 = vld1q_s8 (exp2); + + for (i = 0; i < 16; i++) + if ((result.val[0][i] != expect1[i]) || (result.val[1][i] != expect2[i])) + abort (); + + return 0; +} diff --git a/gcc/testsuite/gcc.target/aarch64/simd/vuzpqs8_1.c b/gcc/testsuite/gcc.target/aarch64/simd/vuzpqs8_1.c new file mode 100644 index 00000000000..234a3292823 --- /dev/null +++ b/gcc/testsuite/gcc.target/aarch64/simd/vuzpqs8_1.c @@ -0,0 +1,11 @@ +/* Test the `vuzpq_s8' AArch64 SIMD intrinsic. */ + +/* { dg-do run } */ +/* { dg-options "-save-temps -fno-inline" } */ + +#include <arm_neon.h> +#include "vuzpqs8.x" + +/* { dg-final { scan-assembler-times "uzp1\[ \t\]+v\[0-9\]+\.16b, ?v\[0-9\]+\.16b, ?v\[0-9\]+\.16b!?\(?:\[ \t\]+@\[a-zA-Z0-9 \]+\)?\n" 1 } } */ +/* { dg-final { scan-assembler-times "uzp2\[ \t\]+v\[0-9\]+\.16b, ?v\[0-9\]+\.16b, ?v\[0-9\]+\.16b!?\(?:\[ \t\]+@\[a-zA-Z0-9 \]+\)?\n" 1 } } */ +/* { dg-final { cleanup-saved-temps } } */ diff --git a/gcc/testsuite/gcc.target/aarch64/simd/vuzpqu16.x b/gcc/testsuite/gcc.target/aarch64/simd/vuzpqu16.x new file mode 100644 index 00000000000..1757467b467 --- /dev/null +++ b/gcc/testsuite/gcc.target/aarch64/simd/vuzpqu16.x @@ -0,0 +1,26 @@ +extern void abort (void); + +uint16x8x2_t +test_vuzpqu16 (uint16x8_t _a, uint16x8_t _b) +{ + return vuzpq_u16 (_a, _b); +} + +int +main (int argc, char **argv) +{ + int i; + uint16_t first[] = {1, 2, 3, 4, 5, 6, 7, 8}; + uint16_t second[] = {9, 10, 11, 12, 13, 14, 15, 16}; + uint16x8x2_t result = test_vuzpqu16 (vld1q_u16 (first), vld1q_u16 (second)); + uint16_t exp1[] = {1, 3, 5, 7, 9, 11, 13, 15}; + uint16_t exp2[] = {2, 4, 6, 8, 10, 12, 14, 16}; + uint16x8_t expect1 = vld1q_u16 (exp1); + uint16x8_t expect2 = vld1q_u16 (exp2); + + for (i = 0; i < 8; i++) + if ((result.val[0][i] != expect1[i]) || (result.val[1][i] != expect2[i])) + abort (); + + return 0; +} diff --git a/gcc/testsuite/gcc.target/aarch64/simd/vuzpqu16_1.c b/gcc/testsuite/gcc.target/aarch64/simd/vuzpqu16_1.c new file mode 100644 index 00000000000..3f029ed5439 --- /dev/null +++ b/gcc/testsuite/gcc.target/aarch64/simd/vuzpqu16_1.c @@ -0,0 +1,11 @@ +/* Test the `vuzpq_u16' AArch64 SIMD intrinsic. */ + +/* { dg-do run } */ +/* { dg-options "-save-temps -fno-inline" } */ + +#include <arm_neon.h> +#include "vuzpqu16.x" + +/* { dg-final { scan-assembler-times "uzp1\[ \t\]+v\[0-9\]+\.8h, ?v\[0-9\]+\.8h, ?v\[0-9\]+\.8h!?\(?:\[ \t\]+@\[a-zA-Z0-9 \]+\)?\n" 1 } } */ +/* { dg-final { scan-assembler-times "uzp2\[ \t\]+v\[0-9\]+\.8h, ?v\[0-9\]+\.8h, ?v\[0-9\]+\.8h!?\(?:\[ \t\]+@\[a-zA-Z0-9 \]+\)?\n" 1 } } */ +/* { dg-final { cleanup-saved-temps } } */ diff --git a/gcc/testsuite/gcc.target/aarch64/simd/vuzpqu32.x b/gcc/testsuite/gcc.target/aarch64/simd/vuzpqu32.x new file mode 100644 index 00000000000..9ff23694c29 --- /dev/null +++ b/gcc/testsuite/gcc.target/aarch64/simd/vuzpqu32.x @@ -0,0 +1,26 @@ +extern void abort (void); + +uint32x4x2_t +test_vuzpqu32 (uint32x4_t _a, uint32x4_t _b) +{ + return vuzpq_u32 (_a, _b); +} + +int +main (int argc, char **argv) +{ + int i; + uint32_t first[] = {1, 2, 3, 4}; + uint32_t second[] = {5, 6, 7, 8}; + uint32x4x2_t result = test_vuzpqu32 (vld1q_u32 (first), vld1q_u32 (second)); + uint32_t exp1[] = {1, 3, 5, 7}; + uint32_t exp2[] = {2, 4, 6, 8}; + uint32x4_t expect1 = vld1q_u32 (exp1); + uint32x4_t expect2 = vld1q_u32 (exp2); + + for (i = 0; i < 4; i++) + if ((result.val[0][i] != expect1[i]) || (result.val[1][i] != expect2[i])) + abort (); + + return 0; +} diff --git a/gcc/testsuite/gcc.target/aarch64/simd/vuzpqu32_1.c b/gcc/testsuite/gcc.target/aarch64/simd/vuzpqu32_1.c new file mode 100644 index 00000000000..16090eed712 --- /dev/null +++ b/gcc/testsuite/gcc.target/aarch64/simd/vuzpqu32_1.c @@ -0,0 +1,11 @@ +/* Test the `vuzpq_u32' AArch64 SIMD intrinsic. */ + +/* { dg-do run } */ +/* { dg-options "-save-temps -fno-inline" } */ + +#include <arm_neon.h> +#include "vuzpqu32.x" + +/* { dg-final { scan-assembler-times "uzp1\[ \t\]+v\[0-9\]+\.4s, ?v\[0-9\]+\.4s, ?v\[0-9\]+\.4s!?\(?:\[ \t\]+@\[a-zA-Z0-9 \]+\)?\n" 1 } } */ +/* { dg-final { scan-assembler-times "uzp2\[ \t\]+v\[0-9\]+\.4s, ?v\[0-9\]+\.4s, ?v\[0-9\]+\.4s!?\(?:\[ \t\]+@\[a-zA-Z0-9 \]+\)?\n" 1 } } */ +/* { dg-final { cleanup-saved-temps } } */ diff --git a/gcc/testsuite/gcc.target/aarch64/simd/vuzpqu8.x b/gcc/testsuite/gcc.target/aarch64/simd/vuzpqu8.x new file mode 100644 index 00000000000..1f5288d0ce3 --- /dev/null +++ b/gcc/testsuite/gcc.target/aarch64/simd/vuzpqu8.x @@ -0,0 +1,27 @@ +extern void abort (void); + +uint8x16x2_t +test_vuzpqu8 (uint8x16_t _a, uint8x16_t _b) +{ + return vuzpq_u8 (_a, _b); +} + +int +main (int argc, char **argv) +{ + int i; + uint8_t first[] = {1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16}; + uint8_t second[] = + {17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32}; + uint8x16x2_t result = test_vuzpqu8 (vld1q_u8 (first), vld1q_u8 (second)); + uint8_t exp1[] = {1, 3, 5, 7, 9, 11, 13, 15, 17, 19, 21, 23, 25, 27, 29, 31}; + uint8_t exp2[] = {2, 4, 6, 8, 10, 12, 14, 16, 18, 20, 22, 24, 26, 28, 30, 32}; + uint8x16_t expect1 = vld1q_u8 (exp1); + uint8x16_t expect2 = vld1q_u8 (exp2); + + for (i = 0; i < 16; i++) + if ((result.val[0][i] != expect1[i]) || (result.val[1][i] != expect2[i])) + abort (); + + return 0; +} diff --git a/gcc/testsuite/gcc.target/aarch64/simd/vuzpqu8_1.c b/gcc/testsuite/gcc.target/aarch64/simd/vuzpqu8_1.c new file mode 100644 index 00000000000..6313e4c9b5d --- /dev/null +++ b/gcc/testsuite/gcc.target/aarch64/simd/vuzpqu8_1.c @@ -0,0 +1,11 @@ +/* Test the `vuzpq_u8' AArch64 SIMD intrinsic. */ + +/* { dg-do run } */ +/* { dg-options "-save-temps -fno-inline" } */ + +#include <arm_neon.h> +#include "vuzpqu8.x" + +/* { dg-final { scan-assembler-times "uzp1\[ \t\]+v\[0-9\]+\.16b, ?v\[0-9\]+\.16b, ?v\[0-9\]+\.16b!?\(?:\[ \t\]+@\[a-zA-Z0-9 \]+\)?\n" 1 } } */ +/* { dg-final { scan-assembler-times "uzp2\[ \t\]+v\[0-9\]+\.16b, ?v\[0-9\]+\.16b, ?v\[0-9\]+\.16b!?\(?:\[ \t\]+@\[a-zA-Z0-9 \]+\)?\n" 1 } } */ +/* { dg-final { cleanup-saved-temps } } */ diff --git a/gcc/testsuite/gcc.target/aarch64/simd/vuzps16.x b/gcc/testsuite/gcc.target/aarch64/simd/vuzps16.x new file mode 100644 index 00000000000..4775135d842 --- /dev/null +++ b/gcc/testsuite/gcc.target/aarch64/simd/vuzps16.x @@ -0,0 +1,26 @@ +extern void abort (void); + +int16x4x2_t +test_vuzps16 (int16x4_t _a, int16x4_t _b) +{ + return vuzp_s16 (_a, _b); +} + +int +main (int argc, char **argv) +{ + int i; + int16_t first[] = {1, 2, 3, 4}; + int16_t second[] = {5, 6, 7, 8}; + int16x4x2_t result = test_vuzps16 (vld1_s16 (first), vld1_s16 (second)); + int16_t exp1[] = {1, 3, 5, 7}; + int16_t exp2[] = {2, 4, 6, 8}; + int16x4_t expect1 = vld1_s16 (exp1); + int16x4_t expect2 = vld1_s16 (exp2); + + for (i = 0; i < 4; i++) + if ((result.val[0][i] != expect1[i]) || (result.val[1][i] != expect2[i])) + abort (); + + return 0; +} diff --git a/gcc/testsuite/gcc.target/aarch64/simd/vuzps16_1.c b/gcc/testsuite/gcc.target/aarch64/simd/vuzps16_1.c new file mode 100644 index 00000000000..f31bd31d0bf --- /dev/null +++ b/gcc/testsuite/gcc.target/aarch64/simd/vuzps16_1.c @@ -0,0 +1,11 @@ +/* Test the `vuzp_s16' AArch64 SIMD intrinsic. */ + +/* { dg-do run } */ +/* { dg-options "-save-temps -fno-inline" } */ + +#include <arm_neon.h> +#include "vuzps16.x" + +/* { dg-final { scan-assembler-times "uzp1\[ \t\]+v\[0-9\]+\.4h, ?v\[0-9\]+\.4h, ?v\[0-9\]+\.4h!?\(?:\[ \t\]+@\[a-zA-Z0-9 \]+\)?\n" 1 } } */ +/* { dg-final { scan-assembler-times "uzp2\[ \t\]+v\[0-9\]+\.4h, ?v\[0-9\]+\.4h, ?v\[0-9\]+\.4h!?\(?:\[ \t\]+@\[a-zA-Z0-9 \]+\)?\n" 1 } } */ +/* { dg-final { cleanup-saved-temps } } */ diff --git a/gcc/testsuite/gcc.target/aarch64/simd/vuzps32.x b/gcc/testsuite/gcc.target/aarch64/simd/vuzps32.x new file mode 100644 index 00000000000..6f885ce083b --- /dev/null +++ b/gcc/testsuite/gcc.target/aarch64/simd/vuzps32.x @@ -0,0 +1,26 @@ +extern void abort (void); + +int32x2x2_t +test_vuzps32 (int32x2_t _a, int32x2_t _b) +{ + return vuzp_s32 (_a, _b); +} + +int +main (int argc, char **argv) +{ + int i; + int32_t first[] = {1, 2}; + int32_t second[] = {3, 4}; + int32x2x2_t result = test_vuzps32 (vld1_s32 (first), vld1_s32 (second)); + int32_t exp1[] = {1, 3}; + int32_t exp2[] = {2, 4}; + int32x2_t expect1 = vld1_s32 (exp1); + int32x2_t expect2 = vld1_s32 (exp2); + + for (i = 0; i < 2; i++) + if ((result.val[0][i] != expect1[i]) || (result.val[1][i] != expect2[i])) + abort (); + + return 0; +} diff --git a/gcc/testsuite/gcc.target/aarch64/simd/vuzps32_1.c b/gcc/testsuite/gcc.target/aarch64/simd/vuzps32_1.c new file mode 100644 index 00000000000..af48d63a67e --- /dev/null +++ b/gcc/testsuite/gcc.target/aarch64/simd/vuzps32_1.c @@ -0,0 +1,11 @@ +/* Test the `vuzp_s32' AArch64 SIMD intrinsic. */ + +/* { dg-do run } */ +/* { dg-options "-save-temps -fno-inline" } */ + +#include <arm_neon.h> +#include "vuzps32.x" + +/* { dg-final { scan-assembler-times "zip1\[ \t\]+v\[0-9\]+\.2s, ?v\[0-9\]+\.2s, ?v\[0-9\]+\.2s!?\(?:\[ \t\]+@\[a-zA-Z0-9 \]+\)?\n" 1 } } */ +/* { dg-final { scan-assembler-times "zip2\[ \t\]+v\[0-9\]+\.2s, ?v\[0-9\]+\.2s, ?v\[0-9\]+\.2s!?\(?:\[ \t\]+@\[a-zA-Z0-9 \]+\)?\n" 1 } } */ +/* { dg-final { cleanup-saved-temps } } */ diff --git a/gcc/testsuite/gcc.target/aarch64/simd/vuzps8.x b/gcc/testsuite/gcc.target/aarch64/simd/vuzps8.x new file mode 100644 index 00000000000..62ccad45779 --- /dev/null +++ b/gcc/testsuite/gcc.target/aarch64/simd/vuzps8.x @@ -0,0 +1,26 @@ +extern void abort (void); + +int8x8x2_t +test_vuzps8 (int8x8_t _a, int8x8_t _b) +{ + return vuzp_s8 (_a, _b); +} + +int +main (int argc, char **argv) +{ + int i; + int8_t first[] = {1, 2, 3, 4, 5, 6, 7, 8}; + int8_t second[] = {9, 10, 11, 12, 13, 14, 15, 16}; + int8x8x2_t result = test_vuzps8 (vld1_s8 (first), vld1_s8 (second)); + int8_t exp1[] = {1, 3, 5, 7, 9, 11, 13, 15}; + int8_t exp2[] = {2, 4, 6, 8, 10, 12, 14, 16}; + int8x8_t expect1 = vld1_s8 (exp1); + int8x8_t expect2 = vld1_s8 (exp2); + + for (i = 0; i < 8; i++) + if ((result.val[0][i] != expect1[i]) || (result.val[1][i] != expect2[i])) + abort (); + + return 0; +} diff --git a/gcc/testsuite/gcc.target/aarch64/simd/vuzps8_1.c b/gcc/testsuite/gcc.target/aarch64/simd/vuzps8_1.c new file mode 100644 index 00000000000..5962604ae42 --- /dev/null +++ b/gcc/testsuite/gcc.target/aarch64/simd/vuzps8_1.c @@ -0,0 +1,11 @@ +/* Test the `vuzp_s8' AArch64 SIMD intrinsic. */ + +/* { dg-do run } */ +/* { dg-options "-save-temps -fno-inline" } */ + +#include <arm_neon.h> +#include "vuzps8.x" + +/* { dg-final { scan-assembler-times "uzp1\[ \t\]+v\[0-9\]+\.8b, ?v\[0-9\]+\.8b, ?v\[0-9\]+\.8b!?\(?:\[ \t\]+@\[a-zA-Z0-9 \]+\)?\n" 1 } } */ +/* { dg-final { scan-assembler-times "uzp2\[ \t\]+v\[0-9\]+\.8b, ?v\[0-9\]+\.8b, ?v\[0-9\]+\.8b!?\(?:\[ \t\]+@\[a-zA-Z0-9 \]+\)?\n" 1 } } */ +/* { dg-final { cleanup-saved-temps } } */ diff --git a/gcc/testsuite/gcc.target/aarch64/simd/vuzpu16.x b/gcc/testsuite/gcc.target/aarch64/simd/vuzpu16.x new file mode 100644 index 00000000000..a5983f6f0b2 --- /dev/null +++ b/gcc/testsuite/gcc.target/aarch64/simd/vuzpu16.x @@ -0,0 +1,26 @@ +extern void abort (void); + +uint16x4x2_t +test_vuzpu16 (uint16x4_t _a, uint16x4_t _b) +{ + return vuzp_u16 (_a, _b); +} + +int +main (int argc, char **argv) +{ + int i; + uint16_t first[] = {1, 2, 3, 4}; + uint16_t second[] = {5, 6, 7, 8}; + uint16x4x2_t result = test_vuzpu16 (vld1_u16 (first), vld1_u16 (second)); + uint16_t exp1[] = {1, 3, 5, 7}; + uint16_t exp2[] = {2, 4, 6, 8}; + uint16x4_t expect1 = vld1_u16 (exp1); + uint16x4_t expect2 = vld1_u16 (exp2); + + for (i = 0; i < 4; i++) + if ((result.val[0][i] != expect1[i]) || (result.val[1][i] != expect2[i])) + abort (); + + return 0; +} diff --git a/gcc/testsuite/gcc.target/aarch64/simd/vuzpu16_1.c b/gcc/testsuite/gcc.target/aarch64/simd/vuzpu16_1.c new file mode 100644 index 00000000000..5025c5ff43e --- /dev/null +++ b/gcc/testsuite/gcc.target/aarch64/simd/vuzpu16_1.c @@ -0,0 +1,11 @@ +/* Test the `vuzp_u16' AArch64 SIMD intrinsic. */ + +/* { dg-do run } */ +/* { dg-options "-save-temps -fno-inline" } */ + +#include <arm_neon.h> +#include "vuzpu16.x" + +/* { dg-final { scan-assembler-times "uzp1\[ \t\]+v\[0-9\]+\.4h, ?v\[0-9\]+\.4h, ?v\[0-9\]+\.4h!?\(?:\[ \t\]+@\[a-zA-Z0-9 \]+\)?\n" 1 } } */ +/* { dg-final { scan-assembler-times "uzp2\[ \t\]+v\[0-9\]+\.4h, ?v\[0-9\]+\.4h, ?v\[0-9\]+\.4h!?\(?:\[ \t\]+@\[a-zA-Z0-9 \]+\)?\n" 1 } } */ +/* { dg-final { cleanup-saved-temps } } */ diff --git a/gcc/testsuite/gcc.target/aarch64/simd/vuzpu32.x b/gcc/testsuite/gcc.target/aarch64/simd/vuzpu32.x new file mode 100644 index 00000000000..6bf673130d9 --- /dev/null +++ b/gcc/testsuite/gcc.target/aarch64/simd/vuzpu32.x @@ -0,0 +1,26 @@ +extern void abort (void); + +uint32x2x2_t +test_vuzpu32 (uint32x2_t _a, uint32x2_t _b) +{ + return vuzp_u32 (_a, _b); +} + +int +main (int argc, char **argv) +{ + int i; + uint32_t first[] = {1, 2}; + uint32_t second[] = {3, 4}; + uint32x2x2_t result = test_vuzpu32 (vld1_u32 (first), vld1_u32 (second)); + uint32_t exp1[] = {1, 3}; + uint32_t exp2[] = {2, 4}; + uint32x2_t expect1 = vld1_u32 (exp1); + uint32x2_t expect2 = vld1_u32 (exp2); + + for (i = 0; i < 2; i++) + if ((result.val[0][i] != expect1[i]) || (result.val[1][i] != expect2[i])) + abort (); + + return 0; +} diff --git a/gcc/testsuite/gcc.target/aarch64/simd/vuzpu32_1.c b/gcc/testsuite/gcc.target/aarch64/simd/vuzpu32_1.c new file mode 100644 index 00000000000..05e1c95d42d --- /dev/null +++ b/gcc/testsuite/gcc.target/aarch64/simd/vuzpu32_1.c @@ -0,0 +1,11 @@ +/* Test the `vuzp_u32' AArch64 SIMD intrinsic. */ + +/* { dg-do run } */ +/* { dg-options "-save-temps -fno-inline" } */ + +#include <arm_neon.h> +#include "vuzpu32.x" + +/* { dg-final { scan-assembler-times "zip1\[ \t\]+v\[0-9\]+\.2s, ?v\[0-9\]+\.2s, ?v\[0-9\]+\.2s!?\(?:\[ \t\]+@\[a-zA-Z0-9 \]+\)?\n" 1 } } */ +/* { dg-final { scan-assembler-times "zip2\[ \t\]+v\[0-9\]+\.2s, ?v\[0-9\]+\.2s, ?v\[0-9\]+\.2s!?\(?:\[ \t\]+@\[a-zA-Z0-9 \]+\)?\n" 1 } } */ +/* { dg-final { cleanup-saved-temps } } */ diff --git a/gcc/testsuite/gcc.target/aarch64/simd/vuzpu8.x b/gcc/testsuite/gcc.target/aarch64/simd/vuzpu8.x new file mode 100644 index 00000000000..c3e67e8418f --- /dev/null +++ b/gcc/testsuite/gcc.target/aarch64/simd/vuzpu8.x @@ -0,0 +1,26 @@ +extern void abort (void); + +uint8x8x2_t +test_vuzpu8 (uint8x8_t _a, uint8x8_t _b) +{ + return vuzp_u8 (_a, _b); +} + +int +main (int argc, char **argv) +{ + int i; + uint8_t first[] = {1, 2, 3, 4, 5, 6, 7, 8}; + uint8_t second[] = {9, 10, 11, 12, 13, 14, 15, 16}; + uint8x8x2_t result = test_vuzpu8 (vld1_u8 (first), vld1_u8 (second)); + uint8_t exp1[] = {1, 3, 5, 7, 9, 11, 13, 15}; + uint8_t exp2[] = {2, 4, 6, 8, 10, 12, 14, 16}; + uint8x8_t expect1 = vld1_u8 (exp1); + uint8x8_t expect2 = vld1_u8 (exp2); + + for (i = 0; i < 8; i++) + if ((result.val[0][i] != expect1[i]) || (result.val[1][i] != expect2[i])) + abort (); + + return 0; +} diff --git a/gcc/testsuite/gcc.target/aarch64/simd/vuzpu8_1.c b/gcc/testsuite/gcc.target/aarch64/simd/vuzpu8_1.c new file mode 100644 index 00000000000..57aa49c9330 --- /dev/null +++ b/gcc/testsuite/gcc.target/aarch64/simd/vuzpu8_1.c @@ -0,0 +1,11 @@ +/* Test the `vuzp_u8' AArch64 SIMD intrinsic. */ + +/* { dg-do run } */ +/* { dg-options "-save-temps -fno-inline" } */ + +#include <arm_neon.h> +#include "vuzpu8.x" + +/* { dg-final { scan-assembler-times "uzp1\[ \t\]+v\[0-9\]+\.8b, ?v\[0-9\]+\.8b, ?v\[0-9\]+\.8b!?\(?:\[ \t\]+@\[a-zA-Z0-9 \]+\)?\n" 1 } } */ +/* { dg-final { scan-assembler-times "uzp2\[ \t\]+v\[0-9\]+\.8b, ?v\[0-9\]+\.8b, ?v\[0-9\]+\.8b!?\(?:\[ \t\]+@\[a-zA-Z0-9 \]+\)?\n" 1 } } */ +/* { dg-final { cleanup-saved-temps } } */ diff --git a/gcc/testsuite/gcc.target/aarch64/simd/vzipf32.x b/gcc/testsuite/gcc.target/aarch64/simd/vzipf32.x new file mode 100644 index 00000000000..cc69b892a02 --- /dev/null +++ b/gcc/testsuite/gcc.target/aarch64/simd/vzipf32.x @@ -0,0 +1,27 @@ +extern void abort (void); + +float32x2x2_t +test_vzipf32 (float32x2_t _a, float32x2_t _b) +{ + return vzip_f32 (_a, _b); +} + +int +main (int argc, char **argv) +{ + int i; + float32_t first[] = {1, 2}; + float32_t second[] = {3, 4}; + float32x2x2_t result = test_vzipf32 (vld1_f32 (first), vld1_f32 (second)); + float32x2_t res1 = result.val[0], res2 = result.val[1]; + float32_t exp1[] = {1, 3}; + float32_t exp2[] = {2, 4}; + float32x2_t expected1 = vld1_f32 (exp1); + float32x2_t expected2 = vld1_f32 (exp2); + + for (i = 0; i < 2; i++) + if ((res1[i] != expected1[i]) || (res2[i] != expected2[i])) + abort (); + + return 0; +} diff --git a/gcc/testsuite/gcc.target/aarch64/simd/vzipf32_1.c b/gcc/testsuite/gcc.target/aarch64/simd/vzipf32_1.c new file mode 100644 index 00000000000..df3395a034e --- /dev/null +++ b/gcc/testsuite/gcc.target/aarch64/simd/vzipf32_1.c @@ -0,0 +1,11 @@ +/* Test the `vzip_f32' AArch64 SIMD intrinsic. */ + +/* { dg-do run } */ +/* { dg-options "-save-temps -fno-inline" } */ + +#include <arm_neon.h> +#include "vzipf32.x" + +/* { dg-final { scan-assembler-times "zip1\[ \t\]+v\[0-9\]+\.2s, ?v\[0-9\]+\.2s, ?v\[0-9\]+\.2s!?\(?:\[ \t\]+@\[a-zA-Z0-9 \]+\)?\n" 1 } } */ +/* { dg-final { scan-assembler-times "zip2\[ \t\]+v\[0-9\]+\.2s, ?v\[0-9\]+\.2s, ?v\[0-9\]+\.2s!?\(?:\[ \t\]+@\[a-zA-Z0-9 \]+\)?\n" 1 } } */ +/* { dg-final { cleanup-saved-temps } } */ diff --git a/gcc/testsuite/gcc.target/aarch64/simd/vzipp16.x b/gcc/testsuite/gcc.target/aarch64/simd/vzipp16.x new file mode 100644 index 00000000000..6bdb3e66ac6 --- /dev/null +++ b/gcc/testsuite/gcc.target/aarch64/simd/vzipp16.x @@ -0,0 +1,27 @@ +extern void abort (void); + +poly16x4x2_t +test_vzipp16 (poly16x4_t _a, poly16x4_t _b) +{ + return vzip_p16 (_a, _b); +} + +int +main (int argc, char **argv) +{ + int i; + poly16_t first[] = {1, 2, 3, 4}; + poly16_t second[] = {5, 6, 7, 8}; + poly16x4x2_t result = test_vzipp16 (vld1_p16 (first), vld1_p16 (second)); + poly16x4_t res1 = result.val[0], res2 = result.val[1]; + poly16_t exp1[] = {1, 5, 2, 6}; + poly16_t exp2[] = {3, 7, 4, 8}; + poly16x4_t expected1 = vld1_p16 (exp1); + poly16x4_t expected2 = vld1_p16 (exp2); + + for (i = 0; i < 4; i++) + if ((res1[i] != expected1[i]) || (res2[i] != expected2[i])) + abort (); + + return 0; +} diff --git a/gcc/testsuite/gcc.target/aarch64/simd/vzipp16_1.c b/gcc/testsuite/gcc.target/aarch64/simd/vzipp16_1.c new file mode 100644 index 00000000000..e626a7877b3 --- /dev/null +++ b/gcc/testsuite/gcc.target/aarch64/simd/vzipp16_1.c @@ -0,0 +1,11 @@ +/* Test the `vzip_p16' AArch64 SIMD intrinsic. */ + +/* { dg-do run } */ +/* { dg-options "-save-temps -fno-inline" } */ + +#include <arm_neon.h> +#include "vzipp16.x" + +/* { dg-final { scan-assembler-times "zip1\[ \t\]+v\[0-9\]+\.4h, ?v\[0-9\]+\.4h, ?v\[0-9\]+\.4h!?\(?:\[ \t\]+@\[a-zA-Z0-9 \]+\)?\n" 1 } } */ +/* { dg-final { scan-assembler-times "zip2\[ \t\]+v\[0-9\]+\.4h, ?v\[0-9\]+\.4h, ?v\[0-9\]+\.4h!?\(?:\[ \t\]+@\[a-zA-Z0-9 \]+\)?\n" 1 } } */ +/* { dg-final { cleanup-saved-temps } } */ diff --git a/gcc/testsuite/gcc.target/aarch64/simd/vzipp8.x b/gcc/testsuite/gcc.target/aarch64/simd/vzipp8.x new file mode 100644 index 00000000000..5e8297eee4b --- /dev/null +++ b/gcc/testsuite/gcc.target/aarch64/simd/vzipp8.x @@ -0,0 +1,27 @@ +extern void abort (void); + +poly8x8x2_t +test_vzipp8 (poly8x8_t _a, poly8x8_t _b) +{ + return vzip_p8 (_a, _b); +} + +int +main (int argc, char **argv) +{ + int i; + poly8_t first[] = {1, 2, 3, 4, 5, 6, 7, 8}; + poly8_t second[] = {9, 10, 11, 12, 13, 14, 15, 16}; + poly8x8x2_t result = test_vzipp8 (vld1_p8 (first), vld1_p8 (second)); + poly8x8_t res1 = result.val[0], res2 = result.val[1]; + poly8_t exp1[] = {1, 9, 2, 10, 3, 11, 4, 12}; + poly8_t exp2[] = {5, 13, 6, 14, 7, 15, 8, 16}; + poly8x8_t expected1 = vld1_p8 (exp1); + poly8x8_t expected2 = vld1_p8 (exp2); + + for (i = 0; i < 8; i++) + if ((res1[i] != expected1[i]) || (res2[i] != expected2[i])) + abort (); + + return 0; +} diff --git a/gcc/testsuite/gcc.target/aarch64/simd/vzipp8_1.c b/gcc/testsuite/gcc.target/aarch64/simd/vzipp8_1.c new file mode 100644 index 00000000000..f99cb70211b --- /dev/null +++ b/gcc/testsuite/gcc.target/aarch64/simd/vzipp8_1.c @@ -0,0 +1,11 @@ +/* Test the `vzip_p8' AArch64 SIMD intrinsic. */ + +/* { dg-do run } */ +/* { dg-options "-save-temps -fno-inline" } */ + +#include <arm_neon.h> +#include "vzipp8.x" + +/* { dg-final { scan-assembler-times "zip1\[ \t\]+v\[0-9\]+\.8b, ?v\[0-9\]+\.8b, ?v\[0-9\]+\.8b!?\(?:\[ \t\]+@\[a-zA-Z0-9 \]+\)?\n" 1 } } */ +/* { dg-final { scan-assembler-times "zip2\[ \t\]+v\[0-9\]+\.8b, ?v\[0-9\]+\.8b, ?v\[0-9\]+\.8b!?\(?:\[ \t\]+@\[a-zA-Z0-9 \]+\)?\n" 1 } } */ +/* { dg-final { cleanup-saved-temps } } */ diff --git a/gcc/testsuite/gcc.target/aarch64/simd/vzipqf32.x b/gcc/testsuite/gcc.target/aarch64/simd/vzipqf32.x new file mode 100644 index 00000000000..e220aeaeb14 --- /dev/null +++ b/gcc/testsuite/gcc.target/aarch64/simd/vzipqf32.x @@ -0,0 +1,27 @@ +extern void abort (void); + +float32x4x2_t +test_vzipqf32 (float32x4_t _a, float32x4_t _b) +{ + return vzipq_f32 (_a, _b); +} + +int +main (int argc, char **argv) +{ + int i; + float32_t first[] = {1, 2, 3, 4}; + float32_t second[] = {5, 6, 7, 8}; + float32x4x2_t result = test_vzipqf32 (vld1q_f32 (first), vld1q_f32 (second)); + float32x4_t res1 = result.val[0], res2 = result.val[1]; + float32_t exp1[] = {1, 5, 2, 6}; + float32_t exp2[] = {3, 7, 4, 8}; + float32x4_t expected1 = vld1q_f32 (exp1); + float32x4_t expected2 = vld1q_f32 (exp2); + + for (i = 0; i < 4; i++) + if ((res1[i] != expected1[i]) || (res2[i] != expected2[i])) + abort (); + + return 0; +} diff --git a/gcc/testsuite/gcc.target/aarch64/simd/vzipqf32_1.c b/gcc/testsuite/gcc.target/aarch64/simd/vzipqf32_1.c new file mode 100644 index 00000000000..74dae27dda0 --- /dev/null +++ b/gcc/testsuite/gcc.target/aarch64/simd/vzipqf32_1.c @@ -0,0 +1,11 @@ +/* Test the `vzipq_f32' AArch64 SIMD intrinsic. */ + +/* { dg-do run } */ +/* { dg-options "-save-temps -fno-inline" } */ + +#include <arm_neon.h> +#include "vzipqf32.x" + +/* { dg-final { scan-assembler-times "zip1\[ \t\]+v\[0-9\]+\.4s, ?v\[0-9\]+\.4s, ?v\[0-9\]+\.4s!?\(?:\[ \t\]+@\[a-zA-Z0-9 \]+\)?\n" 1 } } */ +/* { dg-final { scan-assembler-times "zip2\[ \t\]+v\[0-9\]+\.4s, ?v\[0-9\]+\.4s, ?v\[0-9\]+\.4s!?\(?:\[ \t\]+@\[a-zA-Z0-9 \]+\)?\n" 1 } } */ +/* { dg-final { cleanup-saved-temps } } */ diff --git a/gcc/testsuite/gcc.target/aarch64/simd/vzipqp16.x b/gcc/testsuite/gcc.target/aarch64/simd/vzipqp16.x new file mode 100644 index 00000000000..640d7a2513f --- /dev/null +++ b/gcc/testsuite/gcc.target/aarch64/simd/vzipqp16.x @@ -0,0 +1,27 @@ +extern void abort (void); + +poly16x8x2_t +test_vzipqp16 (poly16x8_t _a, poly16x8_t _b) +{ + return vzipq_p16 (_a, _b); +} + +int +main (int argc, char **argv) +{ + int i; + poly16_t first[] = {1, 2, 3, 4, 5, 6, 7, 8}; + poly16_t second[] = {9, 10, 11, 12, 13, 14, 15, 16}; + poly16x8x2_t result = test_vzipqp16 (vld1q_p16 (first), vld1q_p16 (second)); + poly16x8_t res1 = result.val[0], res2 = result.val[1]; + poly16_t exp1[] = {1, 9, 2, 10, 3, 11, 4, 12}; + poly16_t exp2[] = {5, 13, 6, 14, 7, 15, 8, 16}; + poly16x8_t expected1 = vld1q_p16 (exp1); + poly16x8_t expected2 = vld1q_p16 (exp2); + + for (i = 0; i < 8; i++) + if ((res1[i] != expected1[i]) || (res2[i] != expected2[i])) + abort (); + + return 0; +} diff --git a/gcc/testsuite/gcc.target/aarch64/simd/vzipqp16_1.c b/gcc/testsuite/gcc.target/aarch64/simd/vzipqp16_1.c new file mode 100644 index 00000000000..0bfd4f1a53b --- /dev/null +++ b/gcc/testsuite/gcc.target/aarch64/simd/vzipqp16_1.c @@ -0,0 +1,11 @@ +/* Test the `vzipq_p16' AArch64 SIMD intrinsic. */ + +/* { dg-do run } */ +/* { dg-options "-save-temps -fno-inline" } */ + +#include <arm_neon.h> +#include "vzipqp16.x" + +/* { dg-final { scan-assembler-times "zip1\[ \t\]+v\[0-9\]+\.8h, ?v\[0-9\]+\.8h, ?v\[0-9\]+\.8h!?\(?:\[ \t\]+@\[a-zA-Z0-9 \]+\)?\n" 1 } } */ +/* { dg-final { scan-assembler-times "zip2\[ \t\]+v\[0-9\]+\.8h, ?v\[0-9\]+\.8h, ?v\[0-9\]+\.8h!?\(?:\[ \t\]+@\[a-zA-Z0-9 \]+\)?\n" 1 } } */ +/* { dg-final { cleanup-saved-temps } } */ diff --git a/gcc/testsuite/gcc.target/aarch64/simd/vzipqp8.x b/gcc/testsuite/gcc.target/aarch64/simd/vzipqp8.x new file mode 100644 index 00000000000..b211b4e532f --- /dev/null +++ b/gcc/testsuite/gcc.target/aarch64/simd/vzipqp8.x @@ -0,0 +1,29 @@ +extern void abort (void); + +poly8x16x2_t +test_vzipqp8 (poly8x16_t _a, poly8x16_t _b) +{ + return vzipq_p8 (_a, _b); +} + +int +main (int argc, char **argv) +{ + int i; + poly8_t first[] = {1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16}; + poly8_t second[] = + {17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32}; + poly8x16x2_t result = test_vzipqp8 (vld1q_p8 (first), vld1q_p8 (second)); + poly8x16_t res1 = result.val[0], res2 = result.val[1]; + poly8_t exp1[] = {1, 17, 2, 18, 3, 19, 4, 20, 5, 21, 6, 22, 7, 23, 8, 24}; + poly8_t exp2[] = + {9, 25, 10, 26, 11, 27, 12, 28, 13, 29, 14, 30, 15, 31, 16, 32}; + poly8x16_t expected1 = vld1q_p8 (exp1); + poly8x16_t expected2 = vld1q_p8 (exp2); + + for (i = 0; i < 16; i++) + if ((res1[i] != expected1[i]) || (res2[i] != expected2[i])) + abort (); + + return 0; +} diff --git a/gcc/testsuite/gcc.target/aarch64/simd/vzipqp8_1.c b/gcc/testsuite/gcc.target/aarch64/simd/vzipqp8_1.c new file mode 100644 index 00000000000..fb245063491 --- /dev/null +++ b/gcc/testsuite/gcc.target/aarch64/simd/vzipqp8_1.c @@ -0,0 +1,11 @@ +/* Test the `vzipq_p8' AArch64 SIMD intrinsic. */ + +/* { dg-do run } */ +/* { dg-options "-save-temps -fno-inline" } */ + +#include <arm_neon.h> +#include "vzipqp8.x" + +/* { dg-final { scan-assembler-times "zip1\[ \t\]+v\[0-9\]+\.16b, ?v\[0-9\]+\.16b, ?v\[0-9\]+\.16b!?\(?:\[ \t\]+@\[a-zA-Z0-9 \]+\)?\n" 1 } } */ +/* { dg-final { scan-assembler-times "zip2\[ \t\]+v\[0-9\]+\.16b, ?v\[0-9\]+\.16b, ?v\[0-9\]+\.16b!?\(?:\[ \t\]+@\[a-zA-Z0-9 \]+\)?\n" 1 } } */ +/* { dg-final { cleanup-saved-temps } } */ diff --git a/gcc/testsuite/gcc.target/aarch64/simd/vzipqs16.x b/gcc/testsuite/gcc.target/aarch64/simd/vzipqs16.x new file mode 100644 index 00000000000..97ee6b5bde6 --- /dev/null +++ b/gcc/testsuite/gcc.target/aarch64/simd/vzipqs16.x @@ -0,0 +1,27 @@ +extern void abort (void); + +int16x8x2_t +test_vzipqs16 (int16x8_t _a, int16x8_t _b) +{ + return vzipq_s16 (_a, _b); +} + +int +main (int argc, char **argv) +{ + int i; + int16_t first[] = {1, 2, 3, 4, 5, 6, 7, 8}; + int16_t second[] = {9, 10, 11, 12, 13, 14, 15, 16}; + int16x8x2_t result = test_vzipqs16 (vld1q_s16 (first), vld1q_s16 (second)); + int16x8_t res1 = result.val[0], res2 = result.val[1]; + int16_t exp1[] = {1, 9, 2, 10, 3, 11, 4, 12}; + int16_t exp2[] = {5, 13, 6, 14, 7, 15, 8, 16}; + int16x8_t expected1 = vld1q_s16 (exp1); + int16x8_t expected2 = vld1q_s16 (exp2); + + for (i = 0; i < 8; i++) + if ((res1[i] != expected1[i]) || (res2[i] != expected2[i])) + abort (); + + return 0; +} diff --git a/gcc/testsuite/gcc.target/aarch64/simd/vzipqs16_1.c b/gcc/testsuite/gcc.target/aarch64/simd/vzipqs16_1.c new file mode 100644 index 00000000000..3ff551cceb2 --- /dev/null +++ b/gcc/testsuite/gcc.target/aarch64/simd/vzipqs16_1.c @@ -0,0 +1,11 @@ +/* Test the `vzipq_s16' AArch64 SIMD intrinsic. */ + +/* { dg-do run } */ +/* { dg-options "-save-temps -fno-inline" } */ + +#include <arm_neon.h> +#include "vzipqs16.x" + +/* { dg-final { scan-assembler-times "zip1\[ \t\]+v\[0-9\]+\.8h, ?v\[0-9\]+\.8h, ?v\[0-9\]+\.8h!?\(?:\[ \t\]+@\[a-zA-Z0-9 \]+\)?\n" 1 } } */ +/* { dg-final { scan-assembler-times "zip2\[ \t\]+v\[0-9\]+\.8h, ?v\[0-9\]+\.8h, ?v\[0-9\]+\.8h!?\(?:\[ \t\]+@\[a-zA-Z0-9 \]+\)?\n" 1 } } */ +/* { dg-final { cleanup-saved-temps } } */ diff --git a/gcc/testsuite/gcc.target/aarch64/simd/vzipqs32.x b/gcc/testsuite/gcc.target/aarch64/simd/vzipqs32.x new file mode 100644 index 00000000000..45f490d7350 --- /dev/null +++ b/gcc/testsuite/gcc.target/aarch64/simd/vzipqs32.x @@ -0,0 +1,27 @@ +extern void abort (void); + +int32x4x2_t +test_vzipqs32 (int32x4_t _a, int32x4_t _b) +{ + return vzipq_s32 (_a, _b); +} + +int +main (int argc, char **argv) +{ + int i; + int32_t first[] = {1, 2, 3, 4}; + int32_t second[] = {5, 6, 7, 8}; + int32x4x2_t result = test_vzipqs32 (vld1q_s32 (first), vld1q_s32 (second)); + int32x4_t res1 = result.val[0], res2 = result.val[1]; + int32_t exp1[] = {1, 5, 2, 6}; + int32_t exp2[] = {3, 7, 4, 8}; + int32x4_t expected1 = vld1q_s32 (exp1); + int32x4_t expected2 = vld1q_s32 (exp2); + + for (i = 0; i < 4; i++) + if ((res1[i] != expected1[i]) || (res2[i] != expected2[i])) + abort (); + + return 0; +} diff --git a/gcc/testsuite/gcc.target/aarch64/simd/vzipqs32_1.c b/gcc/testsuite/gcc.target/aarch64/simd/vzipqs32_1.c new file mode 100644 index 00000000000..51681581bc9 --- /dev/null +++ b/gcc/testsuite/gcc.target/aarch64/simd/vzipqs32_1.c @@ -0,0 +1,11 @@ +/* Test the `vzipq_s32' AArch64 SIMD intrinsic. */ + +/* { dg-do run } */ +/* { dg-options "-save-temps -fno-inline" } */ + +#include <arm_neon.h> +#include "vzipqs32.x" + +/* { dg-final { scan-assembler-times "zip1\[ \t\]+v\[0-9\]+\.4s, ?v\[0-9\]+\.4s, ?v\[0-9\]+\.4s!?\(?:\[ \t\]+@\[a-zA-Z0-9 \]+\)?\n" 1 } } */ +/* { dg-final { scan-assembler-times "zip2\[ \t\]+v\[0-9\]+\.4s, ?v\[0-9\]+\.4s, ?v\[0-9\]+\.4s!?\(?:\[ \t\]+@\[a-zA-Z0-9 \]+\)?\n" 1 } } */ +/* { dg-final { cleanup-saved-temps } } */ diff --git a/gcc/testsuite/gcc.target/aarch64/simd/vzipqs8.x b/gcc/testsuite/gcc.target/aarch64/simd/vzipqs8.x new file mode 100644 index 00000000000..68cc84b7e86 --- /dev/null +++ b/gcc/testsuite/gcc.target/aarch64/simd/vzipqs8.x @@ -0,0 +1,29 @@ +extern void abort (void); + +int8x16x2_t +test_vzipqs8 (int8x16_t _a, int8x16_t _b) +{ + return vzipq_s8 (_a, _b); +} + +int +main (int argc, char **argv) +{ + int i; + int8_t first[] = {1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16}; + int8_t second[] = + {17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32}; + int8x16x2_t result = test_vzipqs8 (vld1q_s8 (first), vld1q_s8 (second)); + int8x16_t res1 = result.val[0], res2 = result.val[1]; + int8_t exp1[] = {1, 17, 2, 18, 3, 19, 4, 20, 5, 21, 6, 22, 7, 23, 8, 24}; + int8_t exp2[] = + {9, 25, 10, 26, 11, 27, 12, 28, 13, 29, 14, 30, 15, 31, 16, 32}; + int8x16_t expected1 = vld1q_s8 (exp1); + int8x16_t expected2 = vld1q_s8 (exp2); + + for (i = 0; i < 16; i++) + if ((res1[i] != expected1[i]) || (res2[i] != expected2[i])) + abort (); + + return 0; +} diff --git a/gcc/testsuite/gcc.target/aarch64/simd/vzipqs8_1.c b/gcc/testsuite/gcc.target/aarch64/simd/vzipqs8_1.c new file mode 100644 index 00000000000..ec035f3247d --- /dev/null +++ b/gcc/testsuite/gcc.target/aarch64/simd/vzipqs8_1.c @@ -0,0 +1,11 @@ +/* Test the `vzipq_s8' AArch64 SIMD intrinsic. */ + +/* { dg-do run } */ +/* { dg-options "-save-temps -fno-inline" } */ + +#include <arm_neon.h> +#include "vzipqs8.x" + +/* { dg-final { scan-assembler-times "zip1\[ \t\]+v\[0-9\]+\.16b, ?v\[0-9\]+\.16b, ?v\[0-9\]+\.16b!?\(?:\[ \t\]+@\[a-zA-Z0-9 \]+\)?\n" 1 } } */ +/* { dg-final { scan-assembler-times "zip2\[ \t\]+v\[0-9\]+\.16b, ?v\[0-9\]+\.16b, ?v\[0-9\]+\.16b!?\(?:\[ \t\]+@\[a-zA-Z0-9 \]+\)?\n" 1 } } */ +/* { dg-final { cleanup-saved-temps } } */ diff --git a/gcc/testsuite/gcc.target/aarch64/simd/vzipqu16.x b/gcc/testsuite/gcc.target/aarch64/simd/vzipqu16.x new file mode 100644 index 00000000000..dc4e1462d79 --- /dev/null +++ b/gcc/testsuite/gcc.target/aarch64/simd/vzipqu16.x @@ -0,0 +1,27 @@ +extern void abort (void); + +uint16x8x2_t +test_vzipqu16 (uint16x8_t _a, uint16x8_t _b) +{ + return vzipq_u16 (_a, _b); +} + +int +main (int argc, char **argv) +{ + int i; + uint16_t first[] = {1, 2, 3, 4, 5, 6, 7, 8}; + uint16_t second[] = {9, 10, 11, 12, 13, 14, 15, 16}; + uint16x8x2_t result = test_vzipqu16 (vld1q_u16 (first), vld1q_u16 (second)); + uint16x8_t res1 = result.val[0], res2 = result.val[1]; + uint16_t exp1[] = {1, 9, 2, 10, 3, 11, 4, 12}; + uint16_t exp2[] = {5, 13, 6, 14, 7, 15, 8, 16}; + uint16x8_t expected1 = vld1q_u16 (exp1); + uint16x8_t expected2 = vld1q_u16 (exp2); + + for (i = 0; i < 8; i++) + if ((res1[i] != expected1[i]) || (res2[i] != expected2[i])) + abort (); + + return 0; +} diff --git a/gcc/testsuite/gcc.target/aarch64/simd/vzipqu16_1.c b/gcc/testsuite/gcc.target/aarch64/simd/vzipqu16_1.c new file mode 100644 index 00000000000..b540c8236a6 --- /dev/null +++ b/gcc/testsuite/gcc.target/aarch64/simd/vzipqu16_1.c @@ -0,0 +1,11 @@ +/* Test the `vzipq_u16' AArch64 SIMD intrinsic. */ + +/* { dg-do run } */ +/* { dg-options "-save-temps -fno-inline" } */ + +#include <arm_neon.h> +#include "vzipqu16.x" + +/* { dg-final { scan-assembler-times "zip1\[ \t\]+v\[0-9\]+\.8h, ?v\[0-9\]+\.8h, ?v\[0-9\]+\.8h!?\(?:\[ \t\]+@\[a-zA-Z0-9 \]+\)?\n" 1 } } */ +/* { dg-final { scan-assembler-times "zip2\[ \t\]+v\[0-9\]+\.8h, ?v\[0-9\]+\.8h, ?v\[0-9\]+\.8h!?\(?:\[ \t\]+@\[a-zA-Z0-9 \]+\)?\n" 1 } } */ +/* { dg-final { cleanup-saved-temps } } */ diff --git a/gcc/testsuite/gcc.target/aarch64/simd/vzipqu32.x b/gcc/testsuite/gcc.target/aarch64/simd/vzipqu32.x new file mode 100644 index 00000000000..8dde7e9bd62 --- /dev/null +++ b/gcc/testsuite/gcc.target/aarch64/simd/vzipqu32.x @@ -0,0 +1,27 @@ +extern void abort (void); + +uint32x4x2_t +test_vzipqu32 (uint32x4_t _a, uint32x4_t _b) +{ + return vzipq_u32 (_a, _b); +} + +int +main (int argc, char **argv) +{ + int i; + uint32_t first[] = {1, 2, 3, 4}; + uint32_t second[] = {5, 6, 7, 8}; + uint32x4x2_t result = test_vzipqu32 (vld1q_u32 (first), vld1q_u32 (second)); + uint32x4_t res1 = result.val[0], res2 = result.val[1]; + uint32_t exp1[] = {1, 5, 2, 6}; + uint32_t exp2[] = {3, 7, 4, 8}; + uint32x4_t expected1 = vld1q_u32 (exp1); + uint32x4_t expected2 = vld1q_u32 (exp2); + + for (i = 0; i < 4; i++) + if ((res1[i] != expected1[i]) || (res2[i] != expected2[i])) + abort (); + + return 0; +} diff --git a/gcc/testsuite/gcc.target/aarch64/simd/vzipqu32_1.c b/gcc/testsuite/gcc.target/aarch64/simd/vzipqu32_1.c new file mode 100644 index 00000000000..ca907b34523 --- /dev/null +++ b/gcc/testsuite/gcc.target/aarch64/simd/vzipqu32_1.c @@ -0,0 +1,11 @@ +/* Test the `vzipq_u32' AArch64 SIMD intrinsic. */ + +/* { dg-do run } */ +/* { dg-options "-save-temps -fno-inline" } */ + +#include <arm_neon.h> +#include "vzipqu32.x" + +/* { dg-final { scan-assembler-times "zip1\[ \t\]+v\[0-9\]+\.4s, ?v\[0-9\]+\.4s, ?v\[0-9\]+\.4s!?\(?:\[ \t\]+@\[a-zA-Z0-9 \]+\)?\n" 1 } } */ +/* { dg-final { scan-assembler-times "zip2\[ \t\]+v\[0-9\]+\.4s, ?v\[0-9\]+\.4s, ?v\[0-9\]+\.4s!?\(?:\[ \t\]+@\[a-zA-Z0-9 \]+\)?\n" 1 } } */ +/* { dg-final { cleanup-saved-temps } } */ diff --git a/gcc/testsuite/gcc.target/aarch64/simd/vzipqu8.x b/gcc/testsuite/gcc.target/aarch64/simd/vzipqu8.x new file mode 100644 index 00000000000..8f2603bfebe --- /dev/null +++ b/gcc/testsuite/gcc.target/aarch64/simd/vzipqu8.x @@ -0,0 +1,29 @@ +extern void abort (void); + +uint8x16x2_t +test_vzipqu8 (uint8x16_t _a, uint8x16_t _b) +{ + return vzipq_u8 (_a, _b); +} + +int +main (int argc, char **argv) +{ + int i; + uint8_t first[] = {1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16}; + uint8_t second[] = + {17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32}; + uint8x16x2_t result = test_vzipqu8 (vld1q_u8 (first), vld1q_u8 (second)); + uint8x16_t res1 = result.val[0], res2 = result.val[1]; + uint8_t exp1[] = {1, 17, 2, 18, 3, 19, 4, 20, 5, 21, 6, 22, 7, 23, 8, 24}; + uint8_t exp2[] = + {9, 25, 10, 26, 11, 27, 12, 28, 13, 29, 14, 30, 15, 31, 16, 32}; + uint8x16_t expected1 = vld1q_u8 (exp1); + uint8x16_t expected2 = vld1q_u8 (exp2); + + for (i = 0; i < 16; i++) + if ((res1[i] != expected1[i]) || (res2[i] != expected2[i])) + abort (); + + return 0; +} diff --git a/gcc/testsuite/gcc.target/aarch64/simd/vzipqu8_1.c b/gcc/testsuite/gcc.target/aarch64/simd/vzipqu8_1.c new file mode 100644 index 00000000000..16ada581746 --- /dev/null +++ b/gcc/testsuite/gcc.target/aarch64/simd/vzipqu8_1.c @@ -0,0 +1,11 @@ +/* Test the `vzipq_u8' AArch64 SIMD intrinsic. */ + +/* { dg-do run } */ +/* { dg-options "-save-temps -fno-inline" } */ + +#include <arm_neon.h> +#include "vzipqu8.x" + +/* { dg-final { scan-assembler-times "zip1\[ \t\]+v\[0-9\]+\.16b, ?v\[0-9\]+\.16b, ?v\[0-9\]+\.16b!?\(?:\[ \t\]+@\[a-zA-Z0-9 \]+\)?\n" 1 } } */ +/* { dg-final { scan-assembler-times "zip2\[ \t\]+v\[0-9\]+\.16b, ?v\[0-9\]+\.16b, ?v\[0-9\]+\.16b!?\(?:\[ \t\]+@\[a-zA-Z0-9 \]+\)?\n" 1 } } */ +/* { dg-final { cleanup-saved-temps } } */ diff --git a/gcc/testsuite/gcc.target/aarch64/simd/vzips16.x b/gcc/testsuite/gcc.target/aarch64/simd/vzips16.x new file mode 100644 index 00000000000..71ee4687f3d --- /dev/null +++ b/gcc/testsuite/gcc.target/aarch64/simd/vzips16.x @@ -0,0 +1,27 @@ +extern void abort (void); + +int16x4x2_t +test_vzips16 (int16x4_t _a, int16x4_t _b) +{ + return vzip_s16 (_a, _b); +} + +int +main (int argc, char **argv) +{ + int i; + int16_t first[] = {1, 2, 3, 4}; + int16_t second[] = {5, 6, 7, 8}; + int16x4x2_t result = test_vzips16 (vld1_s16 (first), vld1_s16 (second)); + int16x4_t res1 = result.val[0], res2 = result.val[1]; + int16_t exp1[] = {1, 5, 2, 6}; + int16_t exp2[] = {3, 7, 4, 8}; + int16x4_t expected1 = vld1_s16 (exp1); + int16x4_t expected2 = vld1_s16 (exp2); + + for (i = 0; i < 4; i++) + if ((res1[i] != expected1[i]) || (res2[i] != expected2[i])) + abort (); + + return 0; +} diff --git a/gcc/testsuite/gcc.target/aarch64/simd/vzips16_1.c b/gcc/testsuite/gcc.target/aarch64/simd/vzips16_1.c new file mode 100644 index 00000000000..04a97546922 --- /dev/null +++ b/gcc/testsuite/gcc.target/aarch64/simd/vzips16_1.c @@ -0,0 +1,11 @@ +/* Test the `vzip_s16' AArch64 SIMD intrinsic. */ + +/* { dg-do run } */ +/* { dg-options "-save-temps -fno-inline" } */ + +#include <arm_neon.h> +#include "vzips16.x" + +/* { dg-final { scan-assembler-times "zip1\[ \t\]+v\[0-9\]+\.4h, ?v\[0-9\]+\.4h, ?v\[0-9\]+\.4h!?\(?:\[ \t\]+@\[a-zA-Z0-9 \]+\)?\n" 1 } } */ +/* { dg-final { scan-assembler-times "zip2\[ \t\]+v\[0-9\]+\.4h, ?v\[0-9\]+\.4h, ?v\[0-9\]+\.4h!?\(?:\[ \t\]+@\[a-zA-Z0-9 \]+\)?\n" 1 } } */ +/* { dg-final { cleanup-saved-temps } } */ diff --git a/gcc/testsuite/gcc.target/aarch64/simd/vzips32.x b/gcc/testsuite/gcc.target/aarch64/simd/vzips32.x new file mode 100644 index 00000000000..25bee1c2846 --- /dev/null +++ b/gcc/testsuite/gcc.target/aarch64/simd/vzips32.x @@ -0,0 +1,27 @@ +extern void abort (void); + +int32x2x2_t +test_vzips32 (int32x2_t _a, int32x2_t _b) +{ + return vzip_s32 (_a, _b); +} + +int +main (int argc, char **argv) +{ + int i; + int32_t first[] = {1, 2}; + int32_t second[] = {3, 4}; + int32x2x2_t result = test_vzips32 (vld1_s32 (first), vld1_s32 (second)); + int32x2_t res1 = result.val[0], res2 = result.val[1]; + int32_t exp1[] = {1, 3}; + int32_t exp2[] = {2, 4}; + int32x2_t expected1 = vld1_s32 (exp1); + int32x2_t expected2 = vld1_s32 (exp2); + + for (i = 0; i < 2; i++) + if ((res1[i] != expected1[i]) || (res2[i] != expected2[i])) + abort (); + + return 0; +} diff --git a/gcc/testsuite/gcc.target/aarch64/simd/vzips32_1.c b/gcc/testsuite/gcc.target/aarch64/simd/vzips32_1.c new file mode 100644 index 00000000000..1c44f64453e --- /dev/null +++ b/gcc/testsuite/gcc.target/aarch64/simd/vzips32_1.c @@ -0,0 +1,11 @@ +/* Test the `vzip_s32' AArch64 SIMD intrinsic. */ + +/* { dg-do run } */ +/* { dg-options "-save-temps -fno-inline" } */ + +#include <arm_neon.h> +#include "vzips32.x" + +/* { dg-final { scan-assembler-times "zip1\[ \t\]+v\[0-9\]+\.2s, ?v\[0-9\]+\.2s, ?v\[0-9\]+\.2s!?\(?:\[ \t\]+@\[a-zA-Z0-9 \]+\)?\n" 1 } } */ +/* { dg-final { scan-assembler-times "zip2\[ \t\]+v\[0-9\]+\.2s, ?v\[0-9\]+\.2s, ?v\[0-9\]+\.2s!?\(?:\[ \t\]+@\[a-zA-Z0-9 \]+\)?\n" 1 } } */ +/* { dg-final { cleanup-saved-temps } } */ diff --git a/gcc/testsuite/gcc.target/aarch64/simd/vzips8.x b/gcc/testsuite/gcc.target/aarch64/simd/vzips8.x new file mode 100644 index 00000000000..4f04d731abe --- /dev/null +++ b/gcc/testsuite/gcc.target/aarch64/simd/vzips8.x @@ -0,0 +1,27 @@ +extern void abort (void); + +int8x8x2_t +test_vzips8 (int8x8_t _a, int8x8_t _b) +{ + return vzip_s8 (_a, _b); +} + +int +main (int argc, char **argv) +{ + int i; + int8_t first[] = {1, 2, 3, 4, 5, 6, 7, 8}; + int8_t second[] = {9, 10, 11, 12, 13, 14, 15, 16}; + int8x8x2_t result = test_vzips8 (vld1_s8 (first), vld1_s8 (second)); + int8x8_t res1 = result.val[0], res2 = result.val[1]; + int8_t exp1[] = {1, 9, 2, 10, 3, 11, 4, 12}; + int8_t exp2[] = {5, 13, 6, 14, 7, 15, 8, 16}; + int8x8_t expected1 = vld1_s8 (exp1); + int8x8_t expected2 = vld1_s8 (exp2); + + for (i = 0; i < 8; i++) + if ((res1[i] != expected1[i]) || (res2[i] != expected2[i])) + abort (); + + return 0; +} diff --git a/gcc/testsuite/gcc.target/aarch64/simd/vzips8_1.c b/gcc/testsuite/gcc.target/aarch64/simd/vzips8_1.c new file mode 100644 index 00000000000..5ab7230aa04 --- /dev/null +++ b/gcc/testsuite/gcc.target/aarch64/simd/vzips8_1.c @@ -0,0 +1,11 @@ +/* Test the `vzip_s8' AArch64 SIMD intrinsic. */ + +/* { dg-do run } */ +/* { dg-options "-save-temps -fno-inline" } */ + +#include <arm_neon.h> +#include "vzips8.x" + +/* { dg-final { scan-assembler-times "zip1\[ \t\]+v\[0-9\]+\.8b, ?v\[0-9\]+\.8b, ?v\[0-9\]+\.8b!?\(?:\[ \t\]+@\[a-zA-Z0-9 \]+\)?\n" 1 } } */ +/* { dg-final { scan-assembler-times "zip2\[ \t\]+v\[0-9\]+\.8b, ?v\[0-9\]+\.8b, ?v\[0-9\]+\.8b!?\(?:\[ \t\]+@\[a-zA-Z0-9 \]+\)?\n" 1 } } */ +/* { dg-final { cleanup-saved-temps } } */ diff --git a/gcc/testsuite/gcc.target/aarch64/simd/vzipu16.x b/gcc/testsuite/gcc.target/aarch64/simd/vzipu16.x new file mode 100644 index 00000000000..f8dd2ceea98 --- /dev/null +++ b/gcc/testsuite/gcc.target/aarch64/simd/vzipu16.x @@ -0,0 +1,27 @@ +extern void abort (void); + +uint16x4x2_t +test_vzipu16 (uint16x4_t _a, uint16x4_t _b) +{ + return vzip_u16 (_a, _b); +} + +int +main (int argc, char **argv) +{ + int i; + uint16_t first[] = {1, 2, 3, 4}; + uint16_t second[] = {5, 6, 7, 8}; + uint16x4x2_t result = test_vzipu16 (vld1_u16 (first), vld1_u16 (second)); + uint16x4_t res1 = result.val[0], res2 = result.val[1]; + uint16_t exp1[] = {1, 5, 2, 6}; + uint16_t exp2[] = {3, 7, 4, 8}; + uint16x4_t expected1 = vld1_u16 (exp1); + uint16x4_t expected2 = vld1_u16 (exp2); + + for (i = 0; i < 4; i++) + if ((res1[i] != expected1[i]) || (res2[i] != expected2[i])) + abort (); + + return 0; +} diff --git a/gcc/testsuite/gcc.target/aarch64/simd/vzipu16_1.c b/gcc/testsuite/gcc.target/aarch64/simd/vzipu16_1.c new file mode 100644 index 00000000000..abf7365a733 --- /dev/null +++ b/gcc/testsuite/gcc.target/aarch64/simd/vzipu16_1.c @@ -0,0 +1,11 @@ +/* Test the `vzip_u16' AArch64 SIMD intrinsic. */ + +/* { dg-do run } */ +/* { dg-options "-save-temps -fno-inline" } */ + +#include <arm_neon.h> +#include "vzipu16.x" + +/* { dg-final { scan-assembler-times "zip1\[ \t\]+v\[0-9\]+\.4h, ?v\[0-9\]+\.4h, ?v\[0-9\]+\.4h!?\(?:\[ \t\]+@\[a-zA-Z0-9 \]+\)?\n" 1 } } */ +/* { dg-final { scan-assembler-times "zip2\[ \t\]+v\[0-9\]+\.4h, ?v\[0-9\]+\.4h, ?v\[0-9\]+\.4h!?\(?:\[ \t\]+@\[a-zA-Z0-9 \]+\)?\n" 1 } } */ +/* { dg-final { cleanup-saved-temps } } */ diff --git a/gcc/testsuite/gcc.target/aarch64/simd/vzipu32.x b/gcc/testsuite/gcc.target/aarch64/simd/vzipu32.x new file mode 100644 index 00000000000..0579fc4a8be --- /dev/null +++ b/gcc/testsuite/gcc.target/aarch64/simd/vzipu32.x @@ -0,0 +1,27 @@ +extern void abort (void); + +uint32x2x2_t +test_vzipu32 (uint32x2_t _a, uint32x2_t _b) +{ + return vzip_u32 (_a, _b); +} + +int +main (int argc, char **argv) +{ + int i; + uint32_t first[] = {1, 2}; + uint32_t second[] = {3, 4}; + uint32x2x2_t result = test_vzipu32 (vld1_u32 (first), vld1_u32 (second)); + uint32x2_t res1 = result.val[0], res2 = result.val[1]; + uint32_t exp1[] = {1, 3}; + uint32_t exp2[] = {2, 4}; + uint32x2_t expected1 = vld1_u32 (exp1); + uint32x2_t expected2 = vld1_u32 (exp2); + + for (i = 0; i < 2; i++) + if ((res1[i] != expected1[i]) || (res2[i] != expected2[i])) + abort (); + + return 0; +} diff --git a/gcc/testsuite/gcc.target/aarch64/simd/vzipu32_1.c b/gcc/testsuite/gcc.target/aarch64/simd/vzipu32_1.c new file mode 100644 index 00000000000..d994cb29c4f --- /dev/null +++ b/gcc/testsuite/gcc.target/aarch64/simd/vzipu32_1.c @@ -0,0 +1,11 @@ +/* Test the `vzip_u32' AArch64 SIMD intrinsic. */ + +/* { dg-do run } */ +/* { dg-options "-save-temps -fno-inline" } */ + +#include <arm_neon.h> +#include "vzipu32.x" + +/* { dg-final { scan-assembler-times "zip1\[ \t\]+v\[0-9\]+\.2s, ?v\[0-9\]+\.2s, ?v\[0-9\]+\.2s!?\(?:\[ \t\]+@\[a-zA-Z0-9 \]+\)?\n" 1 } } */ +/* { dg-final { scan-assembler-times "zip2\[ \t\]+v\[0-9\]+\.2s, ?v\[0-9\]+\.2s, ?v\[0-9\]+\.2s!?\(?:\[ \t\]+@\[a-zA-Z0-9 \]+\)?\n" 1 } } */ +/* { dg-final { cleanup-saved-temps } } */ diff --git a/gcc/testsuite/gcc.target/aarch64/simd/vzipu8.x b/gcc/testsuite/gcc.target/aarch64/simd/vzipu8.x new file mode 100644 index 00000000000..28d9205c4f5 --- /dev/null +++ b/gcc/testsuite/gcc.target/aarch64/simd/vzipu8.x @@ -0,0 +1,27 @@ +extern void abort (void); + +uint8x8x2_t +test_vzipu8 (uint8x8_t _a, uint8x8_t _b) +{ + return vzip_u8 (_a, _b); +} + +int +main (int argc, char **argv) +{ + int i; + uint8_t first[] = {1, 2, 3, 4, 5, 6, 7, 8}; + uint8_t second[] = {9, 10, 11, 12, 13, 14, 15, 16}; + uint8x8x2_t result = test_vzipu8 (vld1_u8 (first), vld1_u8 (second)); + uint8x8_t res1 = result.val[0], res2 = result.val[1]; + uint8_t exp1[] = {1, 9, 2, 10, 3, 11, 4, 12}; + uint8_t exp2[] = {5, 13, 6, 14, 7, 15, 8, 16}; + uint8x8_t expected1 = vld1_u8 (exp1); + uint8x8_t expected2 = vld1_u8 (exp2); + + for (i = 0; i < 8; i++) + if ((res1[i] != expected1[i]) || (res2[i] != expected2[i])) + abort (); + + return 0; +} diff --git a/gcc/testsuite/gcc.target/aarch64/simd/vzipu8_1.c b/gcc/testsuite/gcc.target/aarch64/simd/vzipu8_1.c new file mode 100644 index 00000000000..990186a33f1 --- /dev/null +++ b/gcc/testsuite/gcc.target/aarch64/simd/vzipu8_1.c @@ -0,0 +1,11 @@ +/* Test the `vzip_u8' AArch64 SIMD intrinsic. */ + +/* { dg-do run } */ +/* { dg-options "-save-temps -fno-inline" } */ + +#include <arm_neon.h> +#include "vzipu8.x" + +/* { dg-final { scan-assembler-times "zip1\[ \t\]+v\[0-9\]+\.8b, ?v\[0-9\]+\.8b, ?v\[0-9\]+\.8b!?\(?:\[ \t\]+@\[a-zA-Z0-9 \]+\)?\n" 1 } } */ +/* { dg-final { scan-assembler-times "zip2\[ \t\]+v\[0-9\]+\.8b, ?v\[0-9\]+\.8b, ?v\[0-9\]+\.8b!?\(?:\[ \t\]+@\[a-zA-Z0-9 \]+\)?\n" 1 } } */ +/* { dg-final { cleanup-saved-temps } } */ diff --git a/gcc/testsuite/gcc.target/arm/simd/simd.exp b/gcc/testsuite/gcc.target/arm/simd/simd.exp new file mode 100644 index 00000000000..746429dadf6 --- /dev/null +++ b/gcc/testsuite/gcc.target/arm/simd/simd.exp @@ -0,0 +1,35 @@ +# Copyright (C) 1997-2014 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/>. + +# GCC testsuite that uses the `dg.exp' driver. + +# Exit immediately if this isn't an ARM target. +if ![istarget arm*-*-*] then { + return +} + +# Load support procs. +load_lib gcc-dg.exp + +# Initialize `dg'. +dg-init + +# Main loop. +dg-runtest [lsort [glob -nocomplain $srcdir/$subdir/*.\[cCS\]]] \ + "" "" + +# All done. +dg-finish diff --git a/gcc/testsuite/gcc.target/arm/simd/vzipf32_1.c b/gcc/testsuite/gcc.target/arm/simd/vzipf32_1.c new file mode 100644 index 00000000000..efaa96ea955 --- /dev/null +++ b/gcc/testsuite/gcc.target/arm/simd/vzipf32_1.c @@ -0,0 +1,12 @@ +/* Test the `vzipf32' ARM Neon intrinsic. */ + +/* { dg-do run } */ +/* { dg-require-effective-target arm_neon_ok } */ +/* { dg-options "-save-temps -O1 -fno-inline" } */ +/* { dg-add-options arm_neon } */ + +#include "arm_neon.h" +#include "../../aarch64/simd/vzipf32.x" + +/* { dg-final { scan-assembler-times "vuzp\.32\[ \t\]+\[dD\]\[0-9\]+, ?\[dD\]\[0-9\]+!?\(?:\[ \t\]+@\[a-zA-Z0-9 \]+\)?\n" 1 } } */ +/* { dg-final { cleanup-saved-temps } } */ diff --git a/gcc/testsuite/gcc.target/arm/simd/vzipp16_1.c b/gcc/testsuite/gcc.target/arm/simd/vzipp16_1.c new file mode 100644 index 00000000000..4154333a7f7 --- /dev/null +++ b/gcc/testsuite/gcc.target/arm/simd/vzipp16_1.c @@ -0,0 +1,12 @@ +/* Test the `vzipp16' ARM Neon intrinsic. */ + +/* { dg-do run } */ +/* { dg-require-effective-target arm_neon_ok } */ +/* { dg-options "-save-temps -O1 -fno-inline" } */ +/* { dg-add-options arm_neon } */ + +#include "arm_neon.h" +#include "../../aarch64/simd/vzipp16.x" + +/* { dg-final { scan-assembler-times "vzip\.16\[ \t\]+\[dD\]\[0-9\]+, ?\[dD\]\[0-9\]+!?\(?:\[ \t\]+@\[a-zA-Z0-9 \]+\)?\n" 1 } } */ +/* { dg-final { cleanup-saved-temps } } */ diff --git a/gcc/testsuite/gcc.target/arm/simd/vzipp8_1.c b/gcc/testsuite/gcc.target/arm/simd/vzipp8_1.c new file mode 100644 index 00000000000..9fe2384c9f9 --- /dev/null +++ b/gcc/testsuite/gcc.target/arm/simd/vzipp8_1.c @@ -0,0 +1,12 @@ +/* Test the `vzipp8' ARM Neon intrinsic. */ + +/* { dg-do run } */ +/* { dg-require-effective-target arm_neon_ok } */ +/* { dg-options "-save-temps -O1 -fno-inline" } */ +/* { dg-add-options arm_neon } */ + +#include "arm_neon.h" +#include "../../aarch64/simd/vzipp8.x" + +/* { dg-final { scan-assembler-times "vzip\.8\[ \t\]+\[dD\]\[0-9\]+, ?\[dD\]\[0-9\]+!?\(?:\[ \t\]+@\[a-zA-Z0-9 \]+\)?\n" 1 } } */ +/* { dg-final { cleanup-saved-temps } } */ diff --git a/gcc/testsuite/gcc.target/arm/simd/vzipqf32_1.c b/gcc/testsuite/gcc.target/arm/simd/vzipqf32_1.c new file mode 100644 index 00000000000..8c547a79f5b --- /dev/null +++ b/gcc/testsuite/gcc.target/arm/simd/vzipqf32_1.c @@ -0,0 +1,12 @@ +/* Test the `vzipQf32' ARM Neon intrinsic. */ + +/* { dg-do run } */ +/* { dg-require-effective-target arm_neon_ok } */ +/* { dg-options "-save-temps -O1 -fno-inline" } */ +/* { dg-add-options arm_neon } */ + +#include "arm_neon.h" +#include "../../aarch64/simd/vzipqf32.x" + +/* { dg-final { scan-assembler-times "vzip\.32\[ \t\]+\[qQ\]\[0-9\]+, ?\[qQ\]\[0-9\]+!?\(?:\[ \t\]+@\[a-zA-Z0-9 \]+\)?\n" 1 } } */ +/* { dg-final { cleanup-saved-temps } } */ diff --git a/gcc/testsuite/gcc.target/arm/simd/vzipqp16_1.c b/gcc/testsuite/gcc.target/arm/simd/vzipqp16_1.c new file mode 100644 index 00000000000..e2af10b2af1 --- /dev/null +++ b/gcc/testsuite/gcc.target/arm/simd/vzipqp16_1.c @@ -0,0 +1,12 @@ +/* Test the `vzipQp16' ARM Neon intrinsic. */ + +/* { dg-do run } */ +/* { dg-require-effective-target arm_neon_ok } */ +/* { dg-options "-save-temps -O1 -fno-inline" } */ +/* { dg-add-options arm_neon } */ + +#include "arm_neon.h" +#include "../../aarch64/simd/vzipqp16.x" + +/* { dg-final { scan-assembler-times "vzip\.16\[ \t\]+\[qQ\]\[0-9\]+, ?\[qQ\]\[0-9\]+!?\(?:\[ \t\]+@\[a-zA-Z0-9 \]+\)?\n" 1 } } */ +/* { dg-final { cleanup-saved-temps } } */ diff --git a/gcc/testsuite/gcc.target/arm/simd/vzipqp8_1.c b/gcc/testsuite/gcc.target/arm/simd/vzipqp8_1.c new file mode 100644 index 00000000000..11a13298563 --- /dev/null +++ b/gcc/testsuite/gcc.target/arm/simd/vzipqp8_1.c @@ -0,0 +1,12 @@ +/* Test the `vzipQp8' ARM Neon intrinsic. */ + +/* { dg-do run } */ +/* { dg-require-effective-target arm_neon_ok } */ +/* { dg-options "-save-temps -O1 -fno-inline" } */ +/* { dg-add-options arm_neon } */ + +#include "arm_neon.h" +#include "../../aarch64/simd/vzipqp8.x" + +/* { dg-final { scan-assembler-times "vzip\.8\[ \t\]+\[qQ\]\[0-9\]+, ?\[qQ\]\[0-9\]+!?\(?:\[ \t\]+@\[a-zA-Z0-9 \]+\)?\n" 1 } } */ +/* { dg-final { cleanup-saved-temps } } */ diff --git a/gcc/testsuite/gcc.target/arm/simd/vzipqs16_1.c b/gcc/testsuite/gcc.target/arm/simd/vzipqs16_1.c new file mode 100644 index 00000000000..0576c0033e6 --- /dev/null +++ b/gcc/testsuite/gcc.target/arm/simd/vzipqs16_1.c @@ -0,0 +1,12 @@ +/* Test the `vzipQs16' ARM Neon intrinsic. */ + +/* { dg-do run } */ +/* { dg-require-effective-target arm_neon_ok } */ +/* { dg-options "-save-temps -O1 -fno-inline" } */ +/* { dg-add-options arm_neon } */ + +#include "arm_neon.h" +#include "../../aarch64/simd/vzipqs16.x" + +/* { dg-final { scan-assembler-times "vzip\.16\[ \t\]+\[qQ\]\[0-9\]+, ?\[qQ\]\[0-9\]+!?\(?:\[ \t\]+@\[a-zA-Z0-9 \]+\)?\n" 1 } } */ +/* { dg-final { cleanup-saved-temps } } */ diff --git a/gcc/testsuite/gcc.target/arm/simd/vzipqs32_1.c b/gcc/testsuite/gcc.target/arm/simd/vzipqs32_1.c new file mode 100644 index 00000000000..6cf24396d20 --- /dev/null +++ b/gcc/testsuite/gcc.target/arm/simd/vzipqs32_1.c @@ -0,0 +1,12 @@ +/* Test the `vzipQs32' ARM Neon intrinsic. */ + +/* { dg-do run } */ +/* { dg-require-effective-target arm_neon_ok } */ +/* { dg-options "-save-temps -O1 -fno-inline" } */ +/* { dg-add-options arm_neon } */ + +#include "arm_neon.h" +#include "../../aarch64/simd/vzipqs32.x" + +/* { dg-final { scan-assembler-times "vzip\.32\[ \t\]+\[qQ\]\[0-9\]+, ?\[qQ\]\[0-9\]+!?\(?:\[ \t\]+@\[a-zA-Z0-9 \]+\)?\n" 1 } } */ +/* { dg-final { cleanup-saved-temps } } */ diff --git a/gcc/testsuite/gcc.target/arm/simd/vzipqs8_1.c b/gcc/testsuite/gcc.target/arm/simd/vzipqs8_1.c new file mode 100644 index 00000000000..0244374e001 --- /dev/null +++ b/gcc/testsuite/gcc.target/arm/simd/vzipqs8_1.c @@ -0,0 +1,12 @@ +/* Test the `vzipQs8' ARM Neon intrinsic. */ + +/* { dg-do run } */ +/* { dg-require-effective-target arm_neon_ok } */ +/* { dg-options "-save-temps -O1 -fno-inline" } */ +/* { dg-add-options arm_neon } */ + +#include "arm_neon.h" +#include "../../aarch64/simd/vzipqs8.x" + +/* { dg-final { scan-assembler-times "vzip\.8\[ \t\]+\[qQ\]\[0-9\]+, ?\[qQ\]\[0-9\]+!?\(?:\[ \t\]+@\[a-zA-Z0-9 \]+\)?\n" 1 } } */ +/* { dg-final { cleanup-saved-temps } } */ diff --git a/gcc/testsuite/gcc.target/arm/simd/vzipqu16_1.c b/gcc/testsuite/gcc.target/arm/simd/vzipqu16_1.c new file mode 100644 index 00000000000..3c406f514d2 --- /dev/null +++ b/gcc/testsuite/gcc.target/arm/simd/vzipqu16_1.c @@ -0,0 +1,12 @@ +/* Test the `vzipQu16' ARM Neon intrinsic. */ + +/* { dg-do run } */ +/* { dg-require-effective-target arm_neon_ok } */ +/* { dg-options "-save-temps -O1 -fno-inline" } */ +/* { dg-add-options arm_neon } */ + +#include "arm_neon.h" +#include "../../aarch64/simd/vzipqu16.x" + +/* { dg-final { scan-assembler-times "vzip\.16\[ \t\]+\[qQ\]\[0-9\]+, ?\[qQ\]\[0-9\]+!?\(?:\[ \t\]+@\[a-zA-Z0-9 \]+\)?\n" 1 } } */ +/* { dg-final { cleanup-saved-temps } } */ diff --git a/gcc/testsuite/gcc.target/arm/simd/vzipqu32_1.c b/gcc/testsuite/gcc.target/arm/simd/vzipqu32_1.c new file mode 100644 index 00000000000..ba1393c6c92 --- /dev/null +++ b/gcc/testsuite/gcc.target/arm/simd/vzipqu32_1.c @@ -0,0 +1,12 @@ +/* Test the `vzipQu32' ARM Neon intrinsic. */ + +/* { dg-do run } */ +/* { dg-require-effective-target arm_neon_ok } */ +/* { dg-options "-save-temps -O1 -fno-inline" } */ +/* { dg-add-options arm_neon } */ + +#include "arm_neon.h" +#include "../../aarch64/simd/vzipqu32.x" + +/* { dg-final { scan-assembler-times "vzip\.32\[ \t\]+\[qQ\]\[0-9\]+, ?\[qQ\]\[0-9\]+!?\(?:\[ \t\]+@\[a-zA-Z0-9 \]+\)?\n" 1 } } */ +/* { dg-final { cleanup-saved-temps } } */ diff --git a/gcc/testsuite/gcc.target/arm/simd/vzipqu8_1.c b/gcc/testsuite/gcc.target/arm/simd/vzipqu8_1.c new file mode 100644 index 00000000000..023ecac3a52 --- /dev/null +++ b/gcc/testsuite/gcc.target/arm/simd/vzipqu8_1.c @@ -0,0 +1,12 @@ +/* Test the `vzipQu8' ARM Neon intrinsic. */ + +/* { dg-do run } */ +/* { dg-require-effective-target arm_neon_ok } */ +/* { dg-options "-save-temps -O1 -fno-inline" } */ +/* { dg-add-options arm_neon } */ + +#include "arm_neon.h" +#include "../../aarch64/simd/vzipqu8.x" + +/* { dg-final { scan-assembler-times "vzip\.8\[ \t\]+\[qQ\]\[0-9\]+, ?\[qQ\]\[0-9\]+!?\(?:\[ \t\]+@\[a-zA-Z0-9 \]+\)?\n" 1 } } */ +/* { dg-final { cleanup-saved-temps } } */ diff --git a/gcc/testsuite/gcc.target/arm/simd/vzips16_1.c b/gcc/testsuite/gcc.target/arm/simd/vzips16_1.c new file mode 100644 index 00000000000..b6c3c2fe897 --- /dev/null +++ b/gcc/testsuite/gcc.target/arm/simd/vzips16_1.c @@ -0,0 +1,12 @@ +/* Test the `vzips16' ARM Neon intrinsic. */ + +/* { dg-do run } */ +/* { dg-require-effective-target arm_neon_ok } */ +/* { dg-options "-save-temps -O1 -fno-inline" } */ +/* { dg-add-options arm_neon } */ + +#include "arm_neon.h" +#include "../../aarch64/simd/vzips16.x" + +/* { dg-final { scan-assembler-times "vzip\.16\[ \t\]+\[dD\]\[0-9\]+, ?\[dD\]\[0-9\]+!?\(?:\[ \t\]+@\[a-zA-Z0-9 \]+\)?\n" 1 } } */ +/* { dg-final { cleanup-saved-temps } } */ diff --git a/gcc/testsuite/gcc.target/arm/simd/vzips32_1.c b/gcc/testsuite/gcc.target/arm/simd/vzips32_1.c new file mode 100644 index 00000000000..1a6f1709342 --- /dev/null +++ b/gcc/testsuite/gcc.target/arm/simd/vzips32_1.c @@ -0,0 +1,12 @@ +/* Test the `vzips32' ARM Neon intrinsic. */ + +/* { dg-do run } */ +/* { dg-require-effective-target arm_neon_ok } */ +/* { dg-options "-save-temps -O1 -fno-inline" } */ +/* { dg-add-options arm_neon } */ + +#include "arm_neon.h" +#include "../../aarch64/simd/vzips32.x" + +/* { dg-final { scan-assembler-times "vuzp\.32\[ \t\]+\[dD\]\[0-9\]+, ?\[dD\]\[0-9\]+!?\(?:\[ \t\]+@\[a-zA-Z0-9 \]+\)?\n" 1 } } */ +/* { dg-final { cleanup-saved-temps } } */ diff --git a/gcc/testsuite/gcc.target/arm/simd/vzips8_1.c b/gcc/testsuite/gcc.target/arm/simd/vzips8_1.c new file mode 100644 index 00000000000..8569357817b --- /dev/null +++ b/gcc/testsuite/gcc.target/arm/simd/vzips8_1.c @@ -0,0 +1,12 @@ +/* Test the `vzips8' ARM Neon intrinsic. */ + +/* { dg-do run } */ +/* { dg-require-effective-target arm_neon_ok } */ +/* { dg-options "-save-temps -O1 -fno-inline" } */ +/* { dg-add-options arm_neon } */ + +#include "arm_neon.h" +#include "../../aarch64/simd/vzips8.x" + +/* { dg-final { scan-assembler-times "vzip\.8\[ \t\]+\[dD\]\[0-9\]+, ?\[dD\]\[0-9\]+!?\(?:\[ \t\]+@\[a-zA-Z0-9 \]+\)?\n" 1 } } */ +/* { dg-final { cleanup-saved-temps } } */ diff --git a/gcc/testsuite/gcc.target/arm/simd/vzipu16_1.c b/gcc/testsuite/gcc.target/arm/simd/vzipu16_1.c new file mode 100644 index 00000000000..23bfcc4d962 --- /dev/null +++ b/gcc/testsuite/gcc.target/arm/simd/vzipu16_1.c @@ -0,0 +1,12 @@ +/* Test the `vzipu16' ARM Neon intrinsic. */ + +/* { dg-do run } */ +/* { dg-require-effective-target arm_neon_ok } */ +/* { dg-options "-save-temps -O1 -fno-inline" } */ +/* { dg-add-options arm_neon } */ + +#include "arm_neon.h" +#include "../../aarch64/simd/vzipu16.x" + +/* { dg-final { scan-assembler-times "vzip\.16\[ \t\]+\[dD\]\[0-9\]+, ?\[dD\]\[0-9\]+!?\(?:\[ \t\]+@\[a-zA-Z0-9 \]+\)?\n" 1 } } */ +/* { dg-final { cleanup-saved-temps } } */ diff --git a/gcc/testsuite/gcc.target/arm/simd/vzipu32_1.c b/gcc/testsuite/gcc.target/arm/simd/vzipu32_1.c new file mode 100644 index 00000000000..6a753f25a9c --- /dev/null +++ b/gcc/testsuite/gcc.target/arm/simd/vzipu32_1.c @@ -0,0 +1,12 @@ +/* Test the `vzipu32' ARM Neon intrinsic. */ + +/* { dg-do run } */ +/* { dg-require-effective-target arm_neon_ok } */ +/* { dg-options "-save-temps -O1 -fno-inline" } */ +/* { dg-add-options arm_neon } */ + +#include "arm_neon.h" +#include "../../aarch64/simd/vzipu32.x" + +/* { dg-final { scan-assembler-times "vuzp\.32\[ \t\]+\[dD\]\[0-9\]+, ?\[dD\]\[0-9\]+!?\(?:\[ \t\]+@\[a-zA-Z0-9 \]+\)?\n" 1 } } */ +/* { dg-final { cleanup-saved-temps } } */ diff --git a/gcc/testsuite/gcc.target/arm/simd/vzipu8_1.c b/gcc/testsuite/gcc.target/arm/simd/vzipu8_1.c new file mode 100644 index 00000000000..972af74237f --- /dev/null +++ b/gcc/testsuite/gcc.target/arm/simd/vzipu8_1.c @@ -0,0 +1,12 @@ +/* Test the `vzipu8' ARM Neon intrinsic. */ + +/* { dg-do run } */ +/* { dg-require-effective-target arm_neon_ok } */ +/* { dg-options "-save-temps -O1 -fno-inline" } */ +/* { dg-add-options arm_neon } */ + +#include "arm_neon.h" +#include "../../aarch64/simd/vzipu8.x" + +/* { dg-final { scan-assembler-times "vzip\.8\[ \t\]+\[dD\]\[0-9\]+, ?\[dD\]\[0-9\]+!?\(?:\[ \t\]+@\[a-zA-Z0-9 \]+\)?\n" 1 } } */ +/* { dg-final { cleanup-saved-temps } } */ diff --git a/gcc/testsuite/gfortran.dg/arrayio_13.f90 b/gcc/testsuite/gfortran.dg/arrayio_13.f90 new file mode 100644 index 00000000000..92a856bc869 --- /dev/null +++ b/gcc/testsuite/gfortran.dg/arrayio_13.f90 @@ -0,0 +1,14 @@ +! { dg-do run } +! PR60810 Bogus end-of-file +program readstrlist + character(len=80), dimension(2) :: ver + integer :: a, b, c + a = 1 + b = 2 + c = 3 + ver(1) = '285 383' + ver(2) = '985' + read( ver, *) a, b, c + if (a /= 285 .or. b /= 383 .or. c /= 985) call abort + !write ( *, *) a, b, c +end diff --git a/gcc/testsuite/gfortran.dg/vect/pr48329.f90 b/gcc/testsuite/gfortran.dg/vect/pr48329.f90 new file mode 100644 index 00000000000..6ad03d4bd33 --- /dev/null +++ b/gcc/testsuite/gfortran.dg/vect/pr48329.f90 @@ -0,0 +1,29 @@ +! { dg-do compile } +! { dg-require-effective-target vect_float } +! { dg-require-effective-target vect_intfloat_cvt } +! { dg-additional-options "-ffast-math" } + +program calcpi + + implicit none + real(kind=4):: h,x,sum,pi + integer:: n,i + real(kind=4):: f + + f(x) = 4.0/(1.0+x**2) + + n = 2100000000 + + h= 1.0 / n + sum = 0.0 + DO i=1, n + x = h * (i-0.5) + sum = sum + f(x) + END DO + pi = h * sum + write(*,*) 'Pi=',pi + +end program calcpi + +! { dg-final { scan-tree-dump-times "vectorized 1 loops" 1 "vect" } } +! { dg-final { cleanup-tree-dump "vect" } } diff --git a/gcc/tree-cfg.c b/gcc/tree-cfg.c index f0845b66975..5793fcc11fc 100644 --- a/gcc/tree-cfg.c +++ b/gcc/tree-cfg.c @@ -4787,7 +4787,7 @@ collect_subblocks (pointer_set_t *blocks, tree block) /* Verify the GIMPLE statements in the CFG of FN. */ DEBUG_FUNCTION void -verify_gimple_in_cfg (struct function *fn) +verify_gimple_in_cfg (struct function *fn, bool verify_nothrow) { basic_block bb; bool err = false; @@ -4923,16 +4923,17 @@ verify_gimple_in_cfg (struct function *fn) that they cannot throw, that we update other data structures to match. */ lp_nr = lookup_stmt_eh_lp (stmt); - if (lp_nr != 0) + if (lp_nr > 0) { if (!stmt_could_throw_p (stmt)) { - error ("statement marked for throw, but doesn%'t"); - err2 |= true; + if (verify_nothrow) + { + error ("statement marked for throw, but doesn%'t"); + err2 |= true; + } } - else if (lp_nr > 0 - && !gsi_one_before_end_p (gsi) - && stmt_can_throw_internal (stmt)) + else if (!gsi_one_before_end_p (gsi)) { error ("statement marked for throw in middle of block"); err2 |= true; @@ -7088,6 +7089,12 @@ dump_function_to_file (tree fndecl, FILE *file, int flags) struct function *fun = DECL_STRUCT_FUNCTION (fndecl); current_function_decl = fndecl; + + /* Print the return type of the function: */ + print_generic_expr (file, TREE_TYPE (TREE_TYPE (fun->decl)), + dump_flags | TDF_SLIM); + fprintf (file, "\n"); + fprintf (file, "%s %s(", function_name (fun), tmclone ? "[tm-clone] " : ""); arg = DECL_ARGUMENTS (fndecl); diff --git a/gcc/tree-cfg.h b/gcc/tree-cfg.h index a115df58b9d..751d0a29e5a 100644 --- a/gcc/tree-cfg.h +++ b/gcc/tree-cfg.h @@ -58,7 +58,7 @@ extern gimple first_stmt (basic_block); extern gimple last_stmt (basic_block); extern gimple last_and_only_stmt (basic_block); extern void verify_gimple_in_seq (gimple_seq); -extern void verify_gimple_in_cfg (struct function *); +extern void verify_gimple_in_cfg (struct function *, bool); extern tree gimple_block_label (basic_block); extern void add_phi_args_after_copy_bb (basic_block); extern void add_phi_args_after_copy (basic_block *, unsigned, edge); diff --git a/gcc/tree-eh.c b/gcc/tree-eh.c index b83274b1fb8..e86c96b65e3 100644 --- a/gcc/tree-eh.c +++ b/gcc/tree-eh.c @@ -1550,6 +1550,8 @@ lower_try_finally_switch (struct leh_state *state, struct leh_tf_state *tf) /* Make sure that the last case is the default label, as one is required. Then sort the labels, which is also required in GIMPLE. */ CASE_LOW (last_case) = NULL; + tree tem = case_label_vec.pop (); + gcc_assert (tem == last_case); sort_case_labels (case_label_vec); /* Build the switch statement, setting last_case to be the default diff --git a/gcc/tree-pass.h b/gcc/tree-pass.h index 0d941019f0e..616e85bd5a0 100644 --- a/gcc/tree-pass.h +++ b/gcc/tree-pass.h @@ -234,6 +234,7 @@ protected: #define TODO_verify_flow (1 << 3) #define TODO_verify_stmts (1 << 4) #define TODO_cleanup_cfg (1 << 5) +#define TODO_verify_il (1 << 6) #define TODO_dump_symtab (1 << 7) #define TODO_remove_functions (1 << 8) #define TODO_rebuild_frequencies (1 << 9) @@ -309,7 +310,8 @@ protected: | TODO_update_ssa_only_virtuals) #define TODO_verify_all \ - (TODO_verify_ssa | TODO_verify_flow | TODO_verify_stmts) + (TODO_verify_ssa | TODO_verify_flow | TODO_verify_stmts | TODO_verify_il \ + | TODO_verify_rtl_sharing) /* Register pass info. */ diff --git a/gcc/tree-sra.c b/gcc/tree-sra.c index 49bbee3deb7..4a24e6a1b70 100644 --- a/gcc/tree-sra.c +++ b/gcc/tree-sra.c @@ -2769,7 +2769,7 @@ sra_modify_expr (tree *expr, gimple_stmt_iterator *gsi, bool write) { location_t loc; struct access *access; - tree type, bfr; + tree type, bfr, orig_expr; if (TREE_CODE (*expr) == BIT_FIELD_REF) { @@ -2785,6 +2785,7 @@ sra_modify_expr (tree *expr, gimple_stmt_iterator *gsi, bool write) if (!access) return false; type = TREE_TYPE (*expr); + orig_expr = *expr; loc = gimple_location (gsi_stmt (*gsi)); gimple_stmt_iterator alt_gsi = gsi_none (); @@ -2811,8 +2812,7 @@ sra_modify_expr (tree *expr, gimple_stmt_iterator *gsi, bool write) { tree ref; - ref = build_ref_for_model (loc, access->base, access->offset, access, - NULL, false); + ref = build_ref_for_model (loc, orig_expr, 0, access, NULL, false); if (write) { @@ -2863,7 +2863,7 @@ sra_modify_expr (tree *expr, gimple_stmt_iterator *gsi, bool write) else start_offset = chunk_size = 0; - generate_subtree_copies (access->first_child, access->base, 0, + generate_subtree_copies (access->first_child, orig_expr, access->offset, start_offset, chunk_size, gsi, write, write, loc); } @@ -2877,53 +2877,70 @@ enum unscalarized_data_handling { SRA_UDH_NONE, /* Nothing done so far. */ SRA_UDH_RIGHT, /* Data flushed to the RHS. */ SRA_UDH_LEFT }; /* Data flushed to the LHS. */ +struct subreplacement_assignment_data +{ + /* Offset of the access representing the lhs of the assignment. */ + HOST_WIDE_INT left_offset; + + /* LHS and RHS of the original assignment. */ + tree assignment_lhs, assignment_rhs; + + /* Access representing the rhs of the whole assignment. */ + struct access *top_racc; + + /* Stmt iterator used for statement insertions after the original assignment. + It points to the main GSI used to traverse a BB during function body + modification. */ + gimple_stmt_iterator *new_gsi; + + /* Stmt iterator used for statement insertions before the original + assignment. Keeps on pointing to the original statement. */ + gimple_stmt_iterator old_gsi; + + /* Location of the assignment. */ + location_t loc; + + /* Keeps the information whether we have needed to refresh replacements of + the LHS and from which side of the assignments this takes place. */ + enum unscalarized_data_handling refreshed; +}; + /* Store all replacements in the access tree rooted in TOP_RACC either to their base aggregate if there are unscalarized data or directly to LHS of the statement that is pointed to by GSI otherwise. */ -static enum unscalarized_data_handling -handle_unscalarized_data_in_subtree (struct access *top_racc, - gimple_stmt_iterator *gsi) +static void +handle_unscalarized_data_in_subtree (struct subreplacement_assignment_data *sad) { - if (top_racc->grp_unscalarized_data) + tree src; + if (sad->top_racc->grp_unscalarized_data) { - generate_subtree_copies (top_racc->first_child, top_racc->base, 0, 0, 0, - gsi, false, false, - gimple_location (gsi_stmt (*gsi))); - return SRA_UDH_RIGHT; + src = sad->assignment_rhs; + sad->refreshed = SRA_UDH_RIGHT; } else { - tree lhs = gimple_assign_lhs (gsi_stmt (*gsi)); - generate_subtree_copies (top_racc->first_child, lhs, top_racc->offset, - 0, 0, gsi, false, false, - gimple_location (gsi_stmt (*gsi))); - return SRA_UDH_LEFT; + src = sad->assignment_lhs; + sad->refreshed = SRA_UDH_LEFT; } + generate_subtree_copies (sad->top_racc->first_child, src, + sad->top_racc->offset, 0, 0, + &sad->old_gsi, false, false, sad->loc); } - /* Try to generate statements to load all sub-replacements in an access subtree - formed by children of LACC from scalar replacements in the TOP_RACC subtree. - If that is not possible, refresh the TOP_RACC base aggregate and load the - accesses from it. LEFT_OFFSET is the offset of the left whole subtree being - copied. NEW_GSI is stmt iterator used for statement insertions after the - original assignment, OLD_GSI is used to insert statements before the - assignment. *REFRESHED keeps the information whether we have needed to - refresh replacements of the LHS and from which side of the assignments this - takes place. */ + formed by children of LACC from scalar replacements in the SAD->top_racc + subtree. If that is not possible, refresh the SAD->top_racc base aggregate + and load the accesses from it. */ static void -load_assign_lhs_subreplacements (struct access *lacc, struct access *top_racc, - HOST_WIDE_INT left_offset, - gimple_stmt_iterator *old_gsi, - gimple_stmt_iterator *new_gsi, - enum unscalarized_data_handling *refreshed) +load_assign_lhs_subreplacements (struct access *lacc, + struct subreplacement_assignment_data *sad) { - location_t loc = gimple_location (gsi_stmt (*old_gsi)); for (lacc = lacc->first_child; lacc; lacc = lacc->next_sibling) { - HOST_WIDE_INT offset = lacc->offset - left_offset + top_racc->offset; + HOST_WIDE_INT offset; + offset = lacc->offset - sad->left_offset + sad->top_racc->offset; if (lacc->grp_to_be_replaced) { @@ -2931,53 +2948,57 @@ load_assign_lhs_subreplacements (struct access *lacc, struct access *top_racc, gimple stmt; tree rhs; - racc = find_access_in_subtree (top_racc, offset, lacc->size); + racc = find_access_in_subtree (sad->top_racc, offset, lacc->size); if (racc && racc->grp_to_be_replaced) { rhs = get_access_replacement (racc); if (!useless_type_conversion_p (lacc->type, racc->type)) - rhs = fold_build1_loc (loc, VIEW_CONVERT_EXPR, lacc->type, rhs); + rhs = fold_build1_loc (sad->loc, VIEW_CONVERT_EXPR, + lacc->type, rhs); if (racc->grp_partial_lhs && lacc->grp_partial_lhs) - rhs = force_gimple_operand_gsi (old_gsi, rhs, true, NULL_TREE, - true, GSI_SAME_STMT); + rhs = force_gimple_operand_gsi (&sad->old_gsi, rhs, true, + NULL_TREE, true, GSI_SAME_STMT); } else { /* No suitable access on the right hand side, need to load from the aggregate. See if we have to update it first... */ - if (*refreshed == SRA_UDH_NONE) - *refreshed = handle_unscalarized_data_in_subtree (top_racc, - old_gsi); + if (sad->refreshed == SRA_UDH_NONE) + handle_unscalarized_data_in_subtree (sad); - if (*refreshed == SRA_UDH_LEFT) - rhs = build_ref_for_model (loc, lacc->base, lacc->offset, lacc, - new_gsi, true); + if (sad->refreshed == SRA_UDH_LEFT) + rhs = build_ref_for_model (sad->loc, sad->assignment_lhs, + lacc->offset - sad->left_offset, + lacc, sad->new_gsi, true); else - rhs = build_ref_for_model (loc, top_racc->base, offset, lacc, - new_gsi, true); + rhs = build_ref_for_model (sad->loc, sad->assignment_rhs, + lacc->offset - sad->left_offset, + lacc, sad->new_gsi, true); if (lacc->grp_partial_lhs) - rhs = force_gimple_operand_gsi (new_gsi, rhs, true, NULL_TREE, + rhs = force_gimple_operand_gsi (sad->new_gsi, + rhs, true, NULL_TREE, false, GSI_NEW_STMT); } stmt = gimple_build_assign (get_access_replacement (lacc), rhs); - gsi_insert_after (new_gsi, stmt, GSI_NEW_STMT); - gimple_set_location (stmt, loc); + gsi_insert_after (sad->new_gsi, stmt, GSI_NEW_STMT); + gimple_set_location (stmt, sad->loc); update_stmt (stmt); sra_stats.subreplacements++; } else { - if (*refreshed == SRA_UDH_NONE + if (sad->refreshed == SRA_UDH_NONE && lacc->grp_read && !lacc->grp_covered) - *refreshed = handle_unscalarized_data_in_subtree (top_racc, - old_gsi); + handle_unscalarized_data_in_subtree (sad); + if (lacc && lacc->grp_to_be_debug_replaced) { gimple ds; tree drhs; - struct access *racc = find_access_in_subtree (top_racc, offset, + struct access *racc = find_access_in_subtree (sad->top_racc, + offset, lacc->size); if (racc && racc->grp_to_be_replaced) @@ -2987,27 +3008,26 @@ load_assign_lhs_subreplacements (struct access *lacc, struct access *top_racc, else drhs = NULL; } - else if (*refreshed == SRA_UDH_LEFT) - drhs = build_debug_ref_for_model (loc, lacc->base, lacc->offset, - lacc); - else if (*refreshed == SRA_UDH_RIGHT) - drhs = build_debug_ref_for_model (loc, top_racc->base, offset, - lacc); + else if (sad->refreshed == SRA_UDH_LEFT) + drhs = build_debug_ref_for_model (sad->loc, lacc->base, + lacc->offset, lacc); + else if (sad->refreshed == SRA_UDH_RIGHT) + drhs = build_debug_ref_for_model (sad->loc, sad->top_racc->base, + offset, lacc); else drhs = NULL_TREE; if (drhs && !useless_type_conversion_p (lacc->type, TREE_TYPE (drhs))) - drhs = fold_build1_loc (loc, VIEW_CONVERT_EXPR, + drhs = fold_build1_loc (sad->loc, VIEW_CONVERT_EXPR, lacc->type, drhs); ds = gimple_build_debug_bind (get_access_replacement (lacc), - drhs, gsi_stmt (*old_gsi)); - gsi_insert_after (new_gsi, ds, GSI_NEW_STMT); + drhs, gsi_stmt (sad->old_gsi)); + gsi_insert_after (sad->new_gsi, ds, GSI_NEW_STMT); } } if (lacc->first_child) - load_assign_lhs_subreplacements (lacc, top_racc, left_offset, - old_gsi, new_gsi, refreshed); + load_assign_lhs_subreplacements (lacc, sad); } } @@ -3053,7 +3073,7 @@ sra_modify_constructor_assign (gimple *stmt, gimple_stmt_iterator *gsi) /* I have never seen this code path trigger but if it can happen the following should handle it gracefully. */ if (access_has_children_p (acc)) - generate_subtree_copies (acc->first_child, acc->base, 0, 0, 0, gsi, + generate_subtree_copies (acc->first_child, lhs, acc->offset, 0, 0, gsi, true, true, loc); return SRA_AM_MODIFIED; } @@ -3261,7 +3281,7 @@ sra_modify_assign (gimple *stmt, gimple_stmt_iterator *gsi) || stmt_ends_bb_p (*stmt)) { if (access_has_children_p (racc)) - generate_subtree_copies (racc->first_child, racc->base, 0, 0, 0, + generate_subtree_copies (racc->first_child, rhs, racc->offset, 0, 0, gsi, false, false, loc); if (access_has_children_p (lacc)) { @@ -3271,7 +3291,7 @@ sra_modify_assign (gimple *stmt, gimple_stmt_iterator *gsi) alt_gsi = gsi_start_edge (single_non_eh_succ (gsi_bb (*gsi))); gsi = &alt_gsi; } - generate_subtree_copies (lacc->first_child, lacc->base, 0, 0, 0, + generate_subtree_copies (lacc->first_child, lhs, lacc->offset, 0, 0, gsi, true, true, loc); } sra_stats.separate_lhs_rhs_handling++; @@ -3301,21 +3321,26 @@ sra_modify_assign (gimple *stmt, gimple_stmt_iterator *gsi) && !lacc->grp_unscalarizable_region && !racc->grp_unscalarizable_region) { - gimple_stmt_iterator orig_gsi = *gsi; - enum unscalarized_data_handling refreshed; + struct subreplacement_assignment_data sad; + + sad.left_offset = lacc->offset; + sad.assignment_lhs = lhs; + sad.assignment_rhs = rhs; + sad.top_racc = racc; + sad.old_gsi = *gsi; + sad.new_gsi = gsi; + sad.loc = gimple_location (*stmt); + sad.refreshed = SRA_UDH_NONE; if (lacc->grp_read && !lacc->grp_covered) - refreshed = handle_unscalarized_data_in_subtree (racc, gsi); - else - refreshed = SRA_UDH_NONE; + handle_unscalarized_data_in_subtree (&sad); - load_assign_lhs_subreplacements (lacc, racc, lacc->offset, - &orig_gsi, gsi, &refreshed); - if (refreshed != SRA_UDH_RIGHT) + load_assign_lhs_subreplacements (lacc, &sad); + if (sad.refreshed != SRA_UDH_RIGHT) { gsi_next (gsi); unlink_stmt_vdef (*stmt); - gsi_remove (&orig_gsi, true); + gsi_remove (&sad.old_gsi, true); release_defs (*stmt); sra_stats.deleted++; return SRA_AM_REMOVED; @@ -3344,7 +3369,7 @@ sra_modify_assign (gimple *stmt, gimple_stmt_iterator *gsi) /* Restore the aggregate RHS from its components so the prevailing aggregate copy does the right thing. */ if (access_has_children_p (racc)) - generate_subtree_copies (racc->first_child, racc->base, 0, 0, 0, + generate_subtree_copies (racc->first_child, rhs, racc->offset, 0, 0, gsi, false, false, loc); /* Re-load the components of the aggregate copy destination. But use the RHS aggregate to load from to expose more diff --git a/gcc/tree-ssa-alias.c b/gcc/tree-ssa-alias.c index 4f21c4517d1..7781d632266 100644 --- a/gcc/tree-ssa-alias.c +++ b/gcc/tree-ssa-alias.c @@ -1645,6 +1645,7 @@ ref_maybe_used_by_call_p_1 (gimple call, ao_ref *ref) case BUILT_IN_FREE: case BUILT_IN_MALLOC: case BUILT_IN_POSIX_MEMALIGN: + case BUILT_IN_ALIGNED_ALLOC: case BUILT_IN_CALLOC: case BUILT_IN_ALLOCA: case BUILT_IN_ALLOCA_WITH_ALIGN: @@ -1955,6 +1956,7 @@ call_may_clobber_ref_p_1 (gimple call, ao_ref *ref) /* Allocating memory does not have any side-effects apart from being the definition point for the pointer. */ case BUILT_IN_MALLOC: + case BUILT_IN_ALIGNED_ALLOC: case BUILT_IN_CALLOC: case BUILT_IN_STRDUP: case BUILT_IN_STRNDUP: diff --git a/gcc/tree-ssa-ccp.c b/gcc/tree-ssa-ccp.c index b1e21b8b47c..99e1920a070 100644 --- a/gcc/tree-ssa-ccp.c +++ b/gcc/tree-ssa-ccp.c @@ -1803,6 +1803,25 @@ evaluate_stmt (gimple stmt) val = bit_value_assume_aligned (stmt, NULL_TREE, val, false); break; + case BUILT_IN_ALIGNED_ALLOC: + { + tree align = get_constant_value (gimple_call_arg (stmt, 0)); + if (align + && tree_fits_uhwi_p (align)) + { + unsigned HOST_WIDE_INT aligni = tree_to_uhwi (align); + if (aligni > 1 + /* align must be power-of-two */ + && (aligni & (aligni - 1)) == 0) + { + val.lattice_val = CONSTANT; + val.value = build_int_cst (ptr_type_node, 0); + val.mask = double_int::from_shwi (-aligni); + } + } + break; + } + default:; } } diff --git a/gcc/tree-ssa-dce.c b/gcc/tree-ssa-dce.c index 6389f238604..e1edb2ee3ac 100644 --- a/gcc/tree-ssa-dce.c +++ b/gcc/tree-ssa-dce.c @@ -231,6 +231,7 @@ mark_stmt_if_obviously_necessary (gimple stmt, bool aggressive) switch (DECL_FUNCTION_CODE (callee)) { case BUILT_IN_MALLOC: + case BUILT_IN_ALIGNED_ALLOC: case BUILT_IN_CALLOC: case BUILT_IN_ALLOCA: case BUILT_IN_ALLOCA_WITH_ALIGN: @@ -573,6 +574,7 @@ mark_all_reaching_defs_necessary_1 (ao_ref *ref ATTRIBUTE_UNUSED, switch (DECL_FUNCTION_CODE (callee)) { case BUILT_IN_MALLOC: + case BUILT_IN_ALIGNED_ALLOC: case BUILT_IN_CALLOC: case BUILT_IN_ALLOCA: case BUILT_IN_ALLOCA_WITH_ALIGN: @@ -776,7 +778,8 @@ propagate_necessity (bool aggressive) && is_gimple_call (def_stmt = SSA_NAME_DEF_STMT (ptr)) && (def_callee = gimple_call_fndecl (def_stmt)) && DECL_BUILT_IN_CLASS (def_callee) == BUILT_IN_NORMAL - && (DECL_FUNCTION_CODE (def_callee) == BUILT_IN_MALLOC + && (DECL_FUNCTION_CODE (def_callee) == BUILT_IN_ALIGNED_ALLOC + || DECL_FUNCTION_CODE (def_callee) == BUILT_IN_MALLOC || DECL_FUNCTION_CODE (def_callee) == BUILT_IN_CALLOC)) continue; } @@ -822,6 +825,7 @@ propagate_necessity (bool aggressive) && (DECL_FUNCTION_CODE (callee) == BUILT_IN_MEMSET || DECL_FUNCTION_CODE (callee) == BUILT_IN_MEMSET_CHK || DECL_FUNCTION_CODE (callee) == BUILT_IN_MALLOC + || DECL_FUNCTION_CODE (callee) == BUILT_IN_ALIGNED_ALLOC || DECL_FUNCTION_CODE (callee) == BUILT_IN_CALLOC || DECL_FUNCTION_CODE (callee) == BUILT_IN_FREE || DECL_FUNCTION_CODE (callee) == BUILT_IN_VA_END @@ -1229,7 +1233,8 @@ eliminate_unnecessary_stmts (void) special logic we apply to malloc/free pair removal. */ && (!(call = gimple_call_fndecl (stmt)) || DECL_BUILT_IN_CLASS (call) != BUILT_IN_NORMAL - || (DECL_FUNCTION_CODE (call) != BUILT_IN_MALLOC + || (DECL_FUNCTION_CODE (call) != BUILT_IN_ALIGNED_ALLOC + && DECL_FUNCTION_CODE (call) != BUILT_IN_MALLOC && DECL_FUNCTION_CODE (call) != BUILT_IN_CALLOC && DECL_FUNCTION_CODE (call) != BUILT_IN_ALLOCA && (DECL_FUNCTION_CODE (call) diff --git a/gcc/tree-ssa-loop-manip.c b/gcc/tree-ssa-loop-manip.c index 9dcbc530c36..6de2e4e40ec 100644 --- a/gcc/tree-ssa-loop-manip.c +++ b/gcc/tree-ssa-loop-manip.c @@ -598,7 +598,7 @@ verify_loop_closed_ssa (bool verify_ssa_p) return; if (verify_ssa_p) - verify_ssa (false); + verify_ssa (false, true); timevar_push (TV_VERIFY_LOOP_CLOSED); diff --git a/gcc/tree-ssa.c b/gcc/tree-ssa.c index e794dc1bfc3..856325e0de4 100644 --- a/gcc/tree-ssa.c +++ b/gcc/tree-ssa.c @@ -959,7 +959,7 @@ error: TODO: verify the variable annotations. */ DEBUG_FUNCTION void -verify_ssa (bool check_modified_stmt) +verify_ssa (bool check_modified_stmt, bool check_ssa_operands) { size_t i; basic_block bb; @@ -1042,7 +1042,7 @@ verify_ssa (bool check_modified_stmt) goto err; } - if (verify_ssa_operands (cfun, stmt)) + if (check_ssa_operands && verify_ssa_operands (cfun, stmt)) { print_gimple_stmt (stderr, stmt, 0, TDF_VOPS); goto err; diff --git a/gcc/tree-ssa.h b/gcc/tree-ssa.h index 1b1a9869062..c866206d522 100644 --- a/gcc/tree-ssa.h +++ b/gcc/tree-ssa.h @@ -45,7 +45,7 @@ extern void insert_debug_temp_for_var_def (gimple_stmt_iterator *, tree); extern void insert_debug_temps_for_defs (gimple_stmt_iterator *); extern void reset_debug_uses (gimple); extern void release_defs_bitset (bitmap toremove); -extern void verify_ssa (bool); +extern void verify_ssa (bool, bool); extern void init_tree_ssa (struct function *); extern void delete_tree_ssa (void); extern bool tree_ssa_useless_type_conversion (tree); diff --git a/gcc/tree-tailcall.c b/gcc/tree-tailcall.c index 11a29659bc5..9ad25d81c6d 100644 --- a/gcc/tree-tailcall.c +++ b/gcc/tree-tailcall.c @@ -285,9 +285,19 @@ process_assignment (gimple stmt, gimple_stmt_iterator call, tree *m, { /* Reject a tailcall if the type conversion might need additional code. */ - if (gimple_assign_cast_p (stmt) - && TYPE_MODE (TREE_TYPE (dest)) != TYPE_MODE (TREE_TYPE (src_var))) - return false; + if (gimple_assign_cast_p (stmt)) + { + if (TYPE_MODE (TREE_TYPE (dest)) != TYPE_MODE (TREE_TYPE (src_var))) + return false; + + /* Even if the type modes are the same, if the precision of the + type is smaller than mode's precision, + reduce_to_bit_field_precision would generate additional code. */ + if (INTEGRAL_TYPE_P (TREE_TYPE (dest)) + && (GET_MODE_PRECISION (TYPE_MODE (TREE_TYPE (dest))) + > TYPE_PRECISION (TREE_TYPE (dest)))) + return false; + } if (src_var != *ass_var) return false; diff --git a/gcc/tree-vrp.c b/gcc/tree-vrp.c index 890ee936fea..613c9e67103 100644 --- a/gcc/tree-vrp.c +++ b/gcc/tree-vrp.c @@ -4356,7 +4356,7 @@ debug_all_value_ranges (void) /* Given a COND_EXPR COND of the form 'V OP W', and an SSA name V, create a new SSA name N and return the assertion assignment - 'V = ASSERT_EXPR <V, V OP W>'. */ + 'N = ASSERT_EXPR <V, V OP W>'. */ static gimple build_assert_expr_for (tree cond, tree v) @@ -6037,7 +6037,7 @@ process_assert_insertions (void) } else { - y = ASSERT_EXPR <y, x <= y> + y = ASSERT_EXPR <y, x >= y> x = y + 3 } diff --git a/include/ChangeLog b/include/ChangeLog index 9d28cea1bbd..41103a4c8f9 100644 --- a/include/ChangeLog +++ b/include/ChangeLog @@ -1,3 +1,7 @@ +2014-04-30 Richard Sandiford <rdsandiford@googlemail.com> + + * longlong.h (__i386__): Remove W_TYPE_SIZE==64 handling. + 2014-04-22 Yufeng Zhang <yufeng.zhang@arm.com> * longlong.h: Merge from glibc. diff --git a/include/longlong.h b/include/longlong.h index d45dbe2b73b..0770290a58b 100644 --- a/include/longlong.h +++ b/include/longlong.h @@ -483,7 +483,7 @@ extern UDItype __umulsidi3 (USItype, USItype); #define UDIV_TIME 40 #endif /* 80x86 */ -#if (defined (__x86_64__) || defined (__i386__)) && W_TYPE_SIZE == 64 +#if defined (__x86_64__) && W_TYPE_SIZE == 64 #define add_ssaaaa(sh, sl, ah, al, bh, bl) \ __asm__ ("add{q} {%5,%1|%1,%5}\n\tadc{q} {%3,%0|%0,%3}" \ : "=r" ((UDItype) (sh)), \ diff --git a/libgcc/ChangeLog b/libgcc/ChangeLog index e2b28ee2c4d..e562c30235d 100644 --- a/libgcc/ChangeLog +++ b/libgcc/ChangeLog @@ -1,3 +1,10 @@ +2014-04-30 Bernd Edlinger <bernd.edlinger@hotmail.de> + + Work around for current cygwin32 build problems. + * config/i386/cygming-crtbegin.c (__register_frame_info, + __deregister_frame_info, _Jv_RegisterClasses): Compile weak default + functions only for 64-bit systems. + 2014-04-25 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE> * config/i386/crtfastmath.c [!__x86_64__ && __sun__ && __svr4__] diff --git a/libgcc/config/i386/cygming-crtbegin.c b/libgcc/config/i386/cygming-crtbegin.c index eeb51d4c573..195b4637637 100644 --- a/libgcc/config/i386/cygming-crtbegin.c +++ b/libgcc/config/i386/cygming-crtbegin.c @@ -54,6 +54,11 @@ extern void __register_frame_info (__attribute__((unused)) const void *, TARGET_ATTRIBUTE_WEAK; extern void *__deregister_frame_info (__attribute__((unused)) const void *) TARGET_ATTRIBUTE_WEAK; + +/* Work around for current cygwin32 build problems (Bug gas/16858). + Compile weak default functions only for 64-bit systems, + when absolutely necessary. */ +#ifdef __x86_64__ TARGET_ATTRIBUTE_WEAK void __register_frame_info (__attribute__((unused)) const void *p, __attribute__((unused)) struct object *o) @@ -65,16 +70,19 @@ __deregister_frame_info (__attribute__((unused)) const void *p) { return (void*) 0; } +#endif #endif /* DWARF2_UNWIND_INFO */ #if TARGET_USE_JCR_SECTION extern void _Jv_RegisterClasses (__attribute__((unused)) const void *) TARGET_ATTRIBUTE_WEAK; +#ifdef __x86_64__ TARGET_ATTRIBUTE_WEAK void _Jv_RegisterClasses (__attribute__((unused)) const void *p) { } +#endif #endif /* TARGET_USE_JCR_SECTION */ #if defined(HAVE_LD_RO_RW_SECTION_MIXING) |