Skip to content

R
riscv-gcc-1
Commit 897eef7b by Uros Bizjak
Options

extend.texi (Additional Floating Types): __float128 is also supported on i386 targets. 

	* doc/extend.texi (Additional Floating Types): __float128 is also
	supported on i386 targets.

From-SVN: r149054
parent 74d27244
Showing with 74 additions and 73 deletions
gcc/ChangeLog
2009-06-29 Uros Bizjak <ubizjak@gmail.com>
* doc/extend.texi (Additional Floating Types): __float128 is also
supported on i386 targets.
2009-06-29 Richard Guenther <rguenther@suse.de>
PR middle-end/14187
......@@ -84,83 +89,83 @@
2009-06-28 Paolo Bonzini <bonzini@gnu.org>
* config/i386/i386.h (enum ix86_fpcmp_strategy): New.
* config/i386/i386.md (cbranchxf4, cstorexf4, cbranch<MODEF>4,
cstore<MODEF>4, mov<X87MODEF>cc): Change predicate to
ix86_fp_comparison_operator.
(*fp_jcc_1_mixed, *fp_jcc_1_sse, *fp_jcc_1_387, *fp_jcc_2_mixed,
*fp_jcc_2_sse, *fp_jcc_2_387): Delete
(*fp_jcc_3_387, *fp_jcc_4_387, *fp_jcc_5_387, *fp_jcc_6_387,
*fp_jcc_7_387, *fp_jcc_8<MODEF>_387): Eliminate call to
!ix86_use_fcomi_compare, change ix86_fp_jump_nontrivial_p call
to !TARGET_CMOVE, change predicate to ix86_fp_comparison_operator.
(related splits): Change predicate to ix86_fp_comparison_operator.
* config/i386/predicates.md: Use ix86_trivial_fp_comparison_operator
instead of ix86_fp_comparison_codes.
(ix86_trivial_fp_comparison_operator,
ix86_fp_comparison_operator): New.
* config/i386/i386-protos.h (ix86_fp_comparison_strategy): New.
(ix86_expand_compare): Eliminate last two parameters.
(ix86_fp_jump_nontrivial_p): Kill.
* config/i386/i386.c (put_condition_code): Eliminate call to
ix86_fp_comparison_codes and subsequent assertion.
(ix86_fp_comparison_codes): Eliminate.
(ix86_fp_swap_condition): New.
(ix86_fp_comparison_arithmetics_cost, ix86_fp_comparison_fcomi_cost,
ix86_fp_comparison_sahf_cost, ix86_use_fcomi_compare): Consolidate
into ix86_fp_comparison_cost and ix86_fp_comparison_strategy.
(ix86_prepare_fp_compare_args): Use ix86_fp_comparison_strategy
and ix86_fp_swap_condition.
(ix86_expand_fp_compare): Eliminate code for second jump/bypass jump.
Use ix86_fp_comparison_strategy.
(ix86_expand_compare): Likewise. Eliminate last two arguments.
(ix86_fp_jump_nontrivial_p): Eliminate.
(ix86_expand_branch): Treat SFmode/DFmode/XFmode as simple. Adjust
call to ix86_expand_compare.
(ix86_split_fp_branch, ix86_expand_setcc,
ix86_expand_carry_flag_compare, ix86_expand_int_movcc,
ix86_expand_fp_movcc): Eliminate code for second jump/bypass jump.
* config/i386/i386.h (enum ix86_fpcmp_strategy): New.
* config/i386/i386.md (cbranchxf4, cstorexf4, cbranch<MODEF>4,
cstore<MODEF>4, mov<X87MODEF>cc): Change predicate to
ix86_fp_comparison_operator.
(*fp_jcc_1_mixed, *fp_jcc_1_sse, *fp_jcc_1_387, *fp_jcc_2_mixed,
*fp_jcc_2_sse, *fp_jcc_2_387): Delete
(*fp_jcc_3_387, *fp_jcc_4_387, *fp_jcc_5_387, *fp_jcc_6_387,
*fp_jcc_7_387, *fp_jcc_8<MODEF>_387): Eliminate call to
!ix86_use_fcomi_compare, change ix86_fp_jump_nontrivial_p call
to !TARGET_CMOVE, change predicate to ix86_fp_comparison_operator.
(related splits): Change predicate to ix86_fp_comparison_operator.
* config/i386/predicates.md: Use ix86_trivial_fp_comparison_operator
instead of ix86_fp_comparison_codes.
(ix86_trivial_fp_comparison_operator,
ix86_fp_comparison_operator): New.
* config/i386/i386-protos.h (ix86_fp_comparison_strategy): New.
(ix86_expand_compare): Eliminate last two parameters.
(ix86_fp_jump_nontrivial_p): Kill.
* config/i386/i386.c (put_condition_code): Eliminate call to
ix86_fp_comparison_codes and subsequent assertion.
(ix86_fp_comparison_codes): Eliminate.
(ix86_fp_swap_condition): New.
(ix86_fp_comparison_arithmetics_cost, ix86_fp_comparison_fcomi_cost,
ix86_fp_comparison_sahf_cost, ix86_use_fcomi_compare): Consolidate
into ix86_fp_comparison_cost and ix86_fp_comparison_strategy.
(ix86_prepare_fp_compare_args): Use ix86_fp_comparison_strategy
and ix86_fp_swap_condition.
(ix86_expand_fp_compare): Eliminate code for second jump/bypass jump.
Use ix86_fp_comparison_strategy.
(ix86_expand_compare): Likewise. Eliminate last two arguments.
(ix86_fp_jump_nontrivial_p): Eliminate.
(ix86_expand_branch): Treat SFmode/DFmode/XFmode as simple. Adjust
call to ix86_expand_compare.
(ix86_split_fp_branch, ix86_expand_setcc,
ix86_expand_carry_flag_compare, ix86_expand_int_movcc,
ix86_expand_fp_movcc): Eliminate code for second jump/bypass jump.
2009-06-28 Paolo Bonzini <bonzini@gnu.org>
* config/arm/arm.c (arm_final_prescan_ins): Eliminate code
related to jump_clobbers.
* config/arm/arm.md (conds): Remove jump_clob case.
(addsi3_cbranch, addsi3_cbranch_scratch, subsi3_cbranch, two
splits): Change comparison_operator to arm_comparison_operator.
(*arm_buneq, *arm_bltgt, *arm_buneq_reversed, *arm_bltgt_reversed):
Eliminate.
* config/arm/arm.c (arm_final_prescan_ins): Eliminate code
related to jump_clobbers.
* config/arm/arm.md (conds): Remove jump_clob case.
(addsi3_cbranch, addsi3_cbranch_scratch, subsi3_cbranch, two
splits): Change comparison_operator to arm_comparison_operator.
(*arm_buneq, *arm_bltgt, *arm_buneq_reversed, *arm_bltgt_reversed):
Eliminate.
2009-06-28 Paolo Bonzini <bonzini@gnu.org>
* dojump.c (do_compare_rtx_and_jump): Try swapping the
condition for floating point modes.
* expmed.c (emit_store_flag_1): Move here a bigger part
of emit_store_flag.
(emit_store_flag): Try swapping the condition for floating point
modes.
* optabs.c (emit_cmp_and_jump_insns): Cope with constant op0 better.
* dojump.c (do_compare_rtx_and_jump): Try swapping the
condition for floating point modes.
* expmed.c (emit_store_flag_1): Move here a bigger part
of emit_store_flag.
(emit_store_flag): Try swapping the condition for floating point
modes.
* optabs.c (emit_cmp_and_jump_insns): Cope with constant op0 better.
2009-06-28 Paolo Bonzini <bonzini@gnu.org>
* expr.c (expand_expr_real_1): Just use do_store_flag.
(do_store_flag): Drop support for TRUTH_NOT_EXPR. Use
emit_store_flag_force.
* expmed.c (emit_store_flag_force): Copy here trick
previously in expand_expr_real_1. Try reversing the comparison.
(emit_store_flag_1): Work if target is NULL.
(emit_store_flag): Work if target is NULL, using the result mode
from the comparison. Use split_comparison, restructure final part
to simplify conditionals.
* expr.c (expand_expr_real_1): Just use do_store_flag.
(do_store_flag): Drop support for TRUTH_NOT_EXPR. Use
emit_store_flag_force.
* expmed.c (emit_store_flag_force): Copy here trick
previously in expand_expr_real_1. Try reversing the comparison.
(emit_store_flag_1): Work if target is NULL.
(emit_store_flag): Work if target is NULL, using the result mode
from the comparison. Use split_comparison, restructure final part
to simplify conditionals.
2009-06-28 Paolo Bonzini <bonzini@gnu.org>
* builtins.c (expand_errno_check): Use do_compare_rtx_and_jump.
* dojump.c (do_jump): Change handling of floating-point
ops to use just do_compare_and_jump.
(split_comparison): New.
(do_compare_rtx_and_jump): Add here logic coming previously
in do_jump, using split_comparison.
* builtins.c (expand_errno_check): Use do_compare_rtx_and_jump.
* dojump.c (do_jump): Change handling of floating-point
ops to use just do_compare_and_jump.
(split_comparison): New.
(do_compare_rtx_and_jump): Add here logic coming previously
in do_jump, using split_comparison.
2009-06-27 H.J. Lu <hongjiu.lu@intel.com>
......@@ -914,8 +919,7 @@
2009-06-22 Ian Lance Taylor <iant@google.com>
* config/rs6000/rs6000.opt: Move msched-epilog before
msched-prolog.
* config/rs6000/rs6000.opt: Move msched-epilog before msched-prolog.
2009-06-22 Steven Bosscher <steven@gcc.gnu.org>
......@@ -1191,8 +1195,7 @@
2009-06-19 Ramana Radhakrishnan <ramana.radhakrishnan@arm.com>
PR target/40482
* config/arm/arm.c (thumb_shiftable_const): Truncate val to
32 bits.
* config/arm/arm.c (thumb_shiftable_const): Truncate val to 32 bits.
* config/arm/arm.md: Likewise.
2009-06-19 Ian Lance Taylor <iant@google.com>
......@@ -1201,8 +1204,7 @@
to GIMPLE_ERROR_MARK.
* c-typeck.c (build_conditional_expr): Add op1_original_type and
op2_original_type parameters. Warn about using different enum
types.
op2_original_type parameters. Warn about using different enum types.
* c-parser.c (c_parser_conditional_expression): Pass original
types to build_conditional_expr.
* c-tree.h (build_conditional_expr): Update declaration.
......
gcc/doc/extend.texi
......@@ -919,8 +919,7 @@ typedef _Complex float __attribute__((mode(XC))) _Complex80;
@end smallexample
Not all targets support additional floating point types. @code{__float80}
is supported on i386, x86_64 and ia64 targets and target @code{__float128}
is supported on x86_64 and ia64 targets.
and @code{__float128} types are supported on i386, x86_64 and ia64 targets.
@node Half-Precision
@section Half-Precision Floating Point
......
Markdown is supported
0% or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment