diff options
| author | Hu, Lin1 <lin1.hu@intel.com> | 2024-09-02 10:24:31 +0800 |
|---|---|---|
| committer | Haochen Jiang <haochen.jiang@intel.com> | 2024-09-02 10:24:31 +0800 |
| commit | 86f5031c804220274a9bbebd26b8ebf47a2207ac (patch) | |
| tree | a4318ed7c083e733ef8c69f3ffa0c7dcc6e1681e | |
| parent | b1f9fbb6da1a3ced57c3668cecc9f9449e1b237e (diff) | |
| download | gcc-86f5031c804220274a9bbebd26b8ebf47a2207ac.zip gcc-86f5031c804220274a9bbebd26b8ebf47a2207ac.tar.gz gcc-86f5031c804220274a9bbebd26b8ebf47a2207ac.tar.bz2 | |
i386: Optimize ordered and nonequal
Currently, when we input !__builtin_isunordered (a, b) && (a != b), gcc
will emit
ucomiss %xmm1, %xmm0
movl $1, %ecx
setp %dl
setnp %al
cmovne %ecx, %edx
andl %edx, %eax
movzbl %al, %eax
In fact,
xorl %eax, %eax
ucomiss %xmm1, %xmm0
setne %al
is better.
gcc/ChangeLog:
* match.pd: Optimize (and ordered non-equal) to
(not (or unordered equal))
gcc/testsuite/ChangeLog:
* gcc.target/i386/optimize_one.c: New test.
| -rw-r--r-- | gcc/match.pd | 3 | ||||
| -rw-r--r-- | gcc/testsuite/gcc.target/i386/optimize_one.c | 9 |
2 files changed, 12 insertions, 0 deletions
diff --git a/gcc/match.pd b/gcc/match.pd index be21153..4298e89 100644 --- a/gcc/match.pd +++ b/gcc/match.pd @@ -6651,6 +6651,9 @@ DEFINE_INT_AND_FLOAT_ROUND_FN (RINT) (ltgt @0 @0) (if (!flag_trapping_math || !tree_expr_maybe_nan_p (@0)) { constant_boolean_node (false, type); })) +(simplify + (bit_and (ordered @0 @1) (ne @0 @1)) + (bit_not (uneq @0 @1))) /* x == ~x -> false */ /* x != ~x -> true */ diff --git a/gcc/testsuite/gcc.target/i386/optimize_one.c b/gcc/testsuite/gcc.target/i386/optimize_one.c new file mode 100644 index 0000000..62728d3 --- /dev/null +++ b/gcc/testsuite/gcc.target/i386/optimize_one.c @@ -0,0 +1,9 @@ +/* { dg-do compile } */ +/* { dg-options "-O2 -mfpmath=sse" } */ +/* { dg-final { scan-assembler-times "comi" 1 } } */ +/* { dg-final { scan-assembler-times "set" 1 } } */ + +int is_ordered_or_nonequal_sh (float a, float b) +{ + return !__builtin_isunordered (a, b) && (a != b); +} |