The float64_to_uint32 has several flaws:
- for numbers between 2**32 and 2**64, the inexact exception flag
may get incorrectly set. In this case, only the invalid flag
should be set.
test pattern: 425F81378DC0CD1F / 0x1.f81378dc0cd1fp+38
- for numbers between 2**63 and 2**64, incorrect results may
be produced:
test pattern: 43EAAF73F1F0B8BD / 0x1.aaf73f1f0b8bdp+63
This patch re-implements float64_to_uint32 to re-use the
float64_to_uint64 routine (instead of float64_to_int64). For the
saturation case, the inexact bit is explicitly cleared before raising
the invalid flag.
Signed-off-by: Tom Musta <tommu...@gmail.com>
---
fpu/softfloat.c | 10 ++++------
1 files changed, 4 insertions(+), 6 deletions(-)
diff --git a/fpu/softfloat.c b/fpu/softfloat.c
index 1003e59..6a8b6f5 100644
--- a/fpu/softfloat.c
+++ b/fpu/softfloat.c
@@ -6578,15 +6578,13 @@ uint_fast16_t float32_to_uint16_round_to_zero(float32 a
STATUS_PARAM)
uint32 float64_to_uint32( float64 a STATUS_PARAM )
{
- int64_t v;
+ uint64_t v;
uint32 res;
- v = float64_to_int64(a STATUS_VAR);
- if (v < 0) {
- res = 0;
- float_raise( float_flag_invalid STATUS_VAR);
- } else if (v > 0xffffffff) {
+ v = float64_to_uint64(a STATUS_VAR);
+ if (v > 0xffffffff) {
res = 0xffffffff;
+ STATUS(float_exception_flags) &= ~float_flag_inexact;
float_raise( float_flag_invalid STATUS_VAR);
} else {
res = v;
--
1.7.1
