Commit ab57893b by Richard Earnshaw Committed by Richard Earnshaw

arm.h (ENABLE_XF_PATTERNS): Delete.

* arm.h (ENABLE_XF_PATTERNS): Delete.
* arm.md (addxf3, subxf3, mulxf3, divxf3, modxf3, negxf2, absxf2)
(sqrtxf2, floatsixf2, fix_truncxfsi2, truncxfsf2, truncxfdf2)
(extendsfxf2, extenddfxf2, movxf, cmpxf, cmpxf_insn)
(cmpxf_trap): Delete.
(movxf_hard_insn): Remove test of ENABLE_XF_PATTERNS.

From-SVN: r61040
parent 4298c66b
2003-01-08 Richard Earnshaw <rearnsha@arm.com>
* arm.h (ENABLE_XF_PATTERNS): Delete.
* arm.md (addxf3, subxf3, mulxf3, divxf3, modxf3, negxf2, absxf2)
(sqrtxf2, floatsixf2, fix_truncxfsi2, truncxfsf2, truncxfdf2)
(extendsfxf2, extenddfxf2, movxf, cmpxf, cmpxf_insn)
(cmpxf_trap): Delete.
(movxf_hard_insn): Remove test of ENABLE_XF_PATTERNS.
Wed Jan 8 12:10:57 CET 2003 Jan Hubicka <jh@suse.cz> Wed Jan 8 12:10:57 CET 2003 Jan Hubicka <jh@suse.cz>
* i386.md (adddi3_carry_rex64, subdi3_carry_rex64): Name pattern. * i386.md (adddi3_carry_rex64, subdi3_carry_rex64): Name pattern.
......
...@@ -641,20 +641,6 @@ extern int arm_is_6_or_7; ...@@ -641,20 +641,6 @@ extern int arm_is_6_or_7;
/* This is required to ensure that push insns always push a word. */ /* This is required to ensure that push insns always push a word. */
#define PROMOTE_FUNCTION_ARGS #define PROMOTE_FUNCTION_ARGS
/* For the ARM:
I think I have added all the code to make this work. Unfortunately,
early releases of the floating point emulation code on RISCiX used a
different format for extended precision numbers. On my RISCiX box there
is a bug somewhere which causes the machine to lock up when running enquire
with long doubles. There is the additional aspect that Norcroft C
treats long doubles as doubles and we ought to remain compatible.
Perhaps someone with an FPA coprocessor and not running RISCiX would like
to try this someday. */
/* #define LONG_DOUBLE_TYPE_SIZE 96 */
/* Disable XFmode patterns in md file */
#define ENABLE_XF_PATTERNS 0
/* Define this if most significant bit is lowest numbered /* Define this if most significant bit is lowest numbered
in instructions that operate on numbered bit-fields. */ in instructions that operate on numbered bit-fields. */
#define BITS_BIG_ENDIAN 0 #define BITS_BIG_ENDIAN 0
......
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