md.texi (RTL Template): Correct equivalent match_operand expression.
authorAndreas Schwab <schwab@gcc.gnu.org>
Sun, 19 Oct 2014 17:47:29 +0000 (17:47 +0000)
committerAndreas Schwab <schwab@gcc.gnu.org>
Sun, 19 Oct 2014 17:47:29 +0000 (17:47 +0000)
* doc/md.texi (RTL Template) [match_scratch]: Correct equivalent
match_operand expression.

From-SVN: r216441

gcc/ChangeLog
gcc/doc/md.texi

index 9a7f7d2800c9a3b791d7d433e0dea2ea35f9a1d9..bee63fe9710f1212e53f59b4b58cb2d4ba85f53b 100644 (file)
@@ -1,3 +1,8 @@
+2014-10-19  Andreas Schwab  <schwab@linux-m68k.org>
+
+       * doc/md.texi (RTL Template) [match_scratch]: Correct equivalent
+       match_operand expression.
+
 2014-10-19  Adhemerval Zanella  <azanella@linux.vnet.ibm.com>
            David Edelsohn  <dje.gcc@gmail.com>
 
 
 2014-10-16  Andrew MacLeod  <amacleod@redhat.com>
 
-       * function.h: Flatten file.  Remove includes, adjust prototypes to 
+       * function.h: Flatten file.  Remove includes, adjust prototypes to
        reflect only what is in function.h.
        (enum direction, struct args_size, struct locate_and_pad_arg_data,
        ADD_PARM_SIZE, SUB_PARM_SIZE, ARGS_SIZE_TREE, ARGS_SIZE_RTX): Relocate
        Revert:
        2011-08-19  H.J. Lu  <hongjiu.lu@intel.com>
 
-        PR middle-end/49721
-        * explow.c (convert_memory_address_addr_space): Also permute the
-        conversion and addition of constant for zero-extend.
+       PR middle-end/49721
+       * explow.c (convert_memory_address_addr_space): Also permute the
+       conversion and addition of constant for zero-extend.
 
 2014-10-14  DJ Delorie  <dj@redhat.com>
 
            Kirill Yukhin  <kirill.yukhin@intel.com>
            Michael Zolotukhin  <michael.v.zolotukhin@intel.com>
 
-        * config/i386/sse.md
-        (define_mode_iterator REDUC_SMINMAX_MODE): Add V64QI and V32HI modes.
+       * config/i386/sse.md
+       (define_mode_iterator REDUC_SMINMAX_MODE): Add V64QI and V32HI modes.
 
 2014-10-14  Alexander Ivchenko  <alexander.ivchenko@intel.com>
            Maxim Kuznetsov  <maxim.kuznetsov@intel.com>
            Kirill Yukhin  <kirill.yukhin@intel.com>
            Michael Zolotukhin  <michael.v.zolotukhin@intel.com>
 
-        * config/i386/sse.md
-        (define_insn "avx512bw_umulhrswv32hi3<mask_name>"): New.
-        (define_expand "<ssse3_avx2>_pmulhrsw<mode>3_mask"): Ditto.
+       * config/i386/sse.md
+       (define_insn "avx512bw_umulhrswv32hi3<mask_name>"): New.
+       (define_expand "<ssse3_avx2>_pmulhrsw<mode>3_mask"): Ditto.
 
 2014-10-14  Alexander Ivchenko  <alexander.ivchenko@intel.com>
            Maxim Kuznetsov  <maxim.kuznetsov@intel.com>
 2014-10-13  John David Anglin  <dave.anglin@nrc-cnrc.gc.ca>
 
        PR libfortran/63471
-        * config/pa/pa-hpux11.h (TARGET_OS_CPP_BUILTINS): Define _REENTRANT
-        when _HPUX_SOURCE is defined.
+       * config/pa/pa-hpux11.h (TARGET_OS_CPP_BUILTINS): Define _REENTRANT
+       when _HPUX_SOURCE is defined.
 
 2014-10-13  Jan Hubicka  <hubicka@ucw.cz>
 
        * internal-fn.def (UBSAN_OBJECT_SIZE): Define.
        * opts.c (common_handle_option): Handle -fsanitize=object-size.
        * ubsan.c: Include tree-object-size.h.
-       (ubsan_type_descriptor): Call tree_to_uhwi instead of tree_to_shwi. 
+       (ubsan_type_descriptor): Call tree_to_uhwi instead of tree_to_shwi.
        (ubsan_expand_bounds_ifn): Use false instead of 0.
        (ubsan_expand_objsize_ifn): New function.
        (instrument_object_size): New function.
        TARGET_FIX_ERR_A53_835769_DEFAULT.
        * config/aarch64/aarch64.opt (mfix-cortex-a53-835769): Set Init
        value to 2.
-       * doc/install.texi (aarch64*-*-*): Document 
+       * doc/install.texi (aarch64*-*-*): Document
        new --enable-fix-cortex-a53-835769 option.
 
 2014-10-10  Kyrylo Tkachov  <kyrylo.tkachov@arm.com>
-            Ramana Radhakrishnan  <ramana.radhakrishnan@arm.com>
+           Ramana Radhakrishnan  <ramana.radhakrishnan@arm.com>
 
        * config/aarch64/aarch64.h (FINAL_PRESCAN_INSN): Define.
        (ADJUST_INSN_LENGTH): Define.
 2014-10-08  Rong Xu  <xur@google.com>
 
        * gcov-tool.c (profile_overlap): New driver function
-       to compute profile overlap. 
+       to compute profile overlap.
        (print_overlap_usage_message): New.
        (overlap_usage): New.
        (do_overlap): New.
 2014-09-26  Christophe Lyon  <christophe.lyon@linaro.org>
 
        * config/aarch64/aarch64-linux.h (ASAN_CC1_SPEC): Define.
-        (CC1_SPEC): Define.
-        * config/aarch64/aarch64.c (aarch64_asan_shadow_offset): New function.
-        (TARGET_ASAN_SHADOW_OFFSET): Define.
+       (CC1_SPEC): Define.
+       * config/aarch64/aarch64.c (aarch64_asan_shadow_offset): New function.
+       (TARGET_ASAN_SHADOW_OFFSET): Define.
 
 2014-09-26  Martin Liska  <mliska@suse.cz>
 
 
 2014-09-15  Trevor Saunders  <tsaunders@mozilla.com>
 
-       * cfgrtl.c, combine.c, config/arc/arc.c, config/mcore/mcore.c,
-       config/rs6000/rs6000.c, config/sh/sh.c, cprop.c, dwarf2out.c,
-       emit-rtl.c, final.c, function.c, gcse.c, jump.c, reg-stack.c,
-       reload1.c, reorg.c, resource.c, sel-sched-ir.c: Replace INSN_DELETED_P
-       macro with statically checked member functions.
-       * rtl.h (rtx_insn::deleted): New method.
-       (rtx_insn::set_deleted): Likewise.
-       (rtx_insn::set_undeleted): Likewise.
-       (INSN_DELETED_P): Remove.
+       * cfgrtl.c, combine.c, config/arc/arc.c, config/mcore/mcore.c,
+       config/rs6000/rs6000.c, config/sh/sh.c, cprop.c, dwarf2out.c,
+       emit-rtl.c, final.c, function.c, gcse.c, jump.c, reg-stack.c,
+       reload1.c, reorg.c, resource.c, sel-sched-ir.c: Replace INSN_DELETED_P
+       macro with statically checked member functions.
+       * rtl.h (rtx_insn::deleted): New method.
+       (rtx_insn::set_deleted): Likewise.
+       (rtx_insn::set_undeleted): Likewise.
+       (INSN_DELETED_P): Remove.
 
 2014-09-15  Trevor Saunders  <tsaunders@mozilla.com>
 
        Ramana Radhakrishnan  <ramana.radhakrishnan@arm.com>
 
         * config/aarch64/aarch64-elf-raw.h (ENDFILE_SPEC): Add crtfastmath.o.
-         * config/aarch64/aarch64-linux.h (GNU_USER_TARGET_MATH_ENDFILE_SPEC):
+        * config/aarch64/aarch64-linux.h (GNU_USER_TARGET_MATH_ENDFILE_SPEC):
        Define.
-        (ENDFILE_SPEC): Define and use GNU_USER_TARGET_MATH_ENDFILE_SPEC.
+       (ENDFILE_SPEC): Define and use GNU_USER_TARGET_MATH_ENDFILE_SPEC.
 
 2014-09-09  David Malcolm  <dmalcolm@redhat.com>
 
 
 2014-08-27  Kaz Kojima  <kkojima@gcc.gnu.org>
 
-        PR target/62261
-        * config/sh/sh.md (ashlsi3): Handle negative shift count for
+       PR target/62261
+       * config/sh/sh.md (ashlsi3): Handle negative shift count for
        TARGET_SHMEDIA.
        (ashldi3, ashrsi3, ashrdi3, lshrsi3, lshrdi3): Likewise.
 
            Kirill Yukhin  <kirill.yukhin@intel.com>
            Michael Zolotukhin  <michael.v.zolotukhin@intel.com>
 
-        * config/i386/sse.md
+       * config/i386/sse.md
        (define_mode_iterator VI48_AVX2_48_AVX512F): Delete.
        (define_mode_iterator VI48_AVX512BW): New.
        (define_insn "<avx2_avx512f>_<shift_insn>v<mode><mask_name>"): Delete.
 
        * sel-sched-ir.h (struct vinsn_def): Strengthen field "insn_rtx"
        from rtx to rtx_insn *.
-        (VINSN_INSN_RTX): Eliminate rvalue function and...
+       (VINSN_INSN_RTX): Eliminate rvalue function and...
        (SET_VINSN_INSN): ...lvalue function in favor of...
        (VINSN_INSN_RTX): reinstate this old macro.
 
            Kirill Yukhin  <kirill.yukhin@intel.com>
            Michael Zolotukhin  <michael.v.zolotukhin@intel.com>
 
-        * config/i386/sse.md (define_mode_attr avx512): New.
+       * config/i386/sse.md (define_mode_attr avx512): New.
        (define_mode_attr sse2_avx_avx512f): Allow V8HI, V16HI, V32HI, V2DI,
        V4DI modes.
        (define_mode_attr sse2_avx2): Allow V64QI, V32HI, V4TI modes.
            Kirill Yukhin  <kirill.yukhin@intel.com>
            Michael Zolotukhin  <michael.v.zolotukhin@intel.com>
 
-        * config/i386/i386.md
+       * config/i386/i386.md
        (define_attr "isa"): Add avx512bw,noavx512bw.
        (define_attr "enabled"): Ditto.
        (define_split): Add 32/64-bit mask logic.
        (compute_bounds_for_param): use isl_val instead of isl_int
        (compute_bounds_for_loop): likewise
        * graphite-interchange.c: include isl/val.h, isl/val_gmp.h
-        (build_linearized_memory_access): use isl_val instead of isl_int
-        (pdr_stride_in_loop): likewise
+       (build_linearized_memory_access): use isl_val instead of isl_int
+       (pdr_stride_in_loop): likewise
        * graphite-optimize-isl.c:
        (getPrevectorMap): use isl_val instead of isl_int
        * graphite-poly.c:
            Kirill Yukhin  <kirill.yukhin@intel.com>
            Michael Zolotukhin  <michael.v.zolotukhin@intel.com>
 
-        * common/config/i386/i386-common.c
+       * common/config/i386/i386-common.c
        (OPTION_MASK_ISA_AVX512VL_SET): Define.
        (OPTION_MASK_ISA_AVX512F_UNSET): Update.
        (ix86_handle_option): Handle OPT_mavx512vl.
            Kirill Yukhin  <kirill.yukhin@intel.com>
            Michael Zolotukhin  <michael.v.zolotukhin@intel.com>
 
-        * common/config/i386/i386-common.c
+       * common/config/i386/i386-common.c
        (OPTION_MASK_ISA_AVX512BW_SET) : Define.
        (OPTION_MASK_ISA_AVX512BW_UNSET): Ditto.
        (OPTION_MASK_ISA_AVX512VL_UNSET) : Ditto.
            Kirill Yukhin  <kirill.yukhin@intel.com>
            Michael Zolotukhin  <michael.v.zolotukhin@intel.com>
 
-        * common/config/i386/i386-common.c
+       * common/config/i386/i386-common.c
        (OPTION_MASK_ISA_AVX512DQ_SET): Define.
        (OPTION_MASK_ISA_AVX512DQ_UNSET): Ditto.
        (ix86_handle_option): Handle OPT_mavx512dq.
        TARGET_FLOAT and TARGET_SIMD.
 
 2014-08-07  Ian Bolton  <ian.bolton@arm.com>
-            Kyrylo Tkachov  <kyrylo.tkachov@arm.com>
+           Kyrylo Tkachov  <kyrylo.tkachov@arm.com>
 
        * config/aarch64/aarch64.c (aarch64_expand_mov_immediate):
        Use MOVN when one of the half-words is 0xffff.
 
 2014-08-05  David Malcolm  <dmalcolm@redhat.com>
 
-        * gdbhooks.py (find_gcc_source_dir): New helper function.
-        (class PassNames): New class, locating and parsing passes.def.
-        (class BreakOnPass): New command "break-on-pass".
+       * gdbhooks.py (find_gcc_source_dir): New helper function.
+       (class PassNames): New class, locating and parsing passes.def.
+       (class BreakOnPass): New command "break-on-pass".
 
 2014-08-05  Trevor Saunders  <tsaunders@mozilla.com>
 
        PTA_RDRND and PTA_MOVBE for bdver4.
 
 2014-08-04  Kyrylo Tkachov  <kyrylo.tkachov@arm.com>
-            James Greenhalgh  <james.greenhalgh@arm.com>
+           James Greenhalgh  <james.greenhalgh@arm.com>
 
        * doc/md.texi (clrsb): Document.
        (clz): Change reference to x into operand 1.
        move to subtarget in serial version if result is ignored.
 
 2014-08-04  Ramana Radhakrishnan <ramana.radhakrishnan@arm.com>
-            Kyrylo Tkachov  <kyrylo.tkachov@arm.com>
+           Kyrylo Tkachov  <kyrylo.tkachov@arm.com>
 
        * sched-deps.c (try_group_insn): Generalise macro fusion hook usage
        to any two insns.  Update comment.  Rename to sched_macro_fuse_insns.
        Idem.
        * config/i386/i386.c (x96_emit_mode_set): Idem.
        * config/sh/sh.c (sh_emit_mode_set): Likewise. Handle PR toggle.
-       * config/sh/sh.md (toggle_pr):  Defined if TARGET_FPU_SINGLE.
+       * config/sh/sh.md (toggle_pr):  Defined if TARGET_FPU_SINGLE.
        (fpscr_toggle) Disallow from delay slot.
        * target.def (emit_mode_set): Add prev_mode parameter.
        * doc/tm.texi: Regenerate.
index ccebc70be4bb62c5b238cf4de4b50f23c9bf31f4..2f2325e907a805ab87d1c853428b4d57bb25545c 100644 (file)
@@ -297,7 +297,7 @@ expression.
 When matching patterns, this is equivalent to
 
 @smallexample
-(match_operand:@var{m} @var{n} "scratch_operand" @var{pred})
+(match_operand:@var{m} @var{n} "scratch_operand" @var{constraint})
 @end smallexample
 
 but, when generating RTL, it produces a (@code{scratch}:@var{m})