[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
[Qemu-devel] [PULL 62/76] softfloat: Fix float64_to_uint32
From: |
Peter Maydell |
Subject: |
[Qemu-devel] [PULL 62/76] softfloat: Fix float64_to_uint32 |
Date: |
Tue, 7 Jan 2014 20:03:58 +0000 |
From: Tom Musta <address@hidden>
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, we ignore any flags which the conversion routine
has set and raise only the invalid flag.
This contribution can be licensed under either the softfloat-2a or -2b
license.
Signed-off-by: Tom Musta <address@hidden>
Message-id: address@hidden
Signed-off-by: Peter Maydell <address@hidden>
Reviewed-by: Richard Henderson <address@hidden>
---
fpu/softfloat.c | 15 +++++++--------
1 file changed, 7 insertions(+), 8 deletions(-)
diff --git a/fpu/softfloat.c b/fpu/softfloat.c
index f782c24..e68a87b 100644
--- a/fpu/softfloat.c
+++ b/fpu/softfloat.c
@@ -6650,19 +6650,18 @@ 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;
+ int old_exc_flags = get_float_exception_flags(status);
- 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;
- float_raise( float_flag_invalid STATUS_VAR);
} else {
- res = v;
+ return v;
}
+ set_float_exception_flags(old_exc_flags, status);
+ float_raise(float_flag_invalid STATUS_VAR);
return res;
}
--
1.8.5
- [Qemu-devel] [PULL 61/76] softfloat: Fix float64_to_uint64_round_to_zero, (continued)
- [Qemu-devel] [PULL 61/76] softfloat: Fix float64_to_uint64_round_to_zero, Peter Maydell, 2014/01/07
- [Qemu-devel] [PULL 73/76] target-arm: A64: Add floating-point<->fixed-point instructions, Peter Maydell, 2014/01/07
- [Qemu-devel] [PULL 59/76] softfloat: Fix factor 2 error for scalbn on denormal inputs, Peter Maydell, 2014/01/07
- [Qemu-devel] [PULL 72/76] target-arm: A64: Add extra VFP fixed point conversion helpers, Peter Maydell, 2014/01/07
- [Qemu-devel] [PULL 60/76] softfloat: Add float32_to_uint64(), Peter Maydell, 2014/01/07
- [Qemu-devel] [PULL 76/76] target-arm: A64: Add support for FCVT between half, single and double, Peter Maydell, 2014/01/07
- [Qemu-devel] [PULL 51/76] arm_gic: Rename GIC_X_TRIGGER to GIC_X_EDGE_TRIGGER, Peter Maydell, 2014/01/07
- [Qemu-devel] [PULL 69/76] target-arm: Prepare VFP_CONV_FIX helpers for A64 uses, Peter Maydell, 2014/01/07
- [Qemu-devel] [PULL 54/76] softfloat: Add float to 16bit integer conversions., Peter Maydell, 2014/01/07
- [Qemu-devel] [PULL 49/76] arm/xilinx_zynq: Always instantiate the GEMs, Peter Maydell, 2014/01/07
- [Qemu-devel] [PULL 62/76] softfloat: Fix float64_to_uint32,
Peter Maydell <=
- [Qemu-devel] [PULL 63/76] softfloat: Fix float64_to_uint32_round_to_zero, Peter Maydell, 2014/01/07
- [Qemu-devel] [PULL 47/76] target-arm: use c13_context field for CONTEXTIDR, Peter Maydell, 2014/01/07
- [Qemu-devel] [PULL 64/76] softfloat: Provide complete set of accessors for fp state, Peter Maydell, 2014/01/07
- [Qemu-devel] [PULL 20/76] target-arm: Widen exclusive-access support struct fields to 64 bits, Peter Maydell, 2014/01/07
- [Qemu-devel] [PULL 48/76] target-arm: remove raw_read|write duplication, Peter Maydell, 2014/01/07
- [Qemu-devel] [PULL 19/76] target-arm: aarch64: add support for ld lit, Peter Maydell, 2014/01/07
- [Qemu-devel] [PULL 23/76] linux-user: AArch64: Use correct values for FPSR/FPCR in sigcontext, Peter Maydell, 2014/01/07
- [Qemu-devel] [PULL 35/76] target-arm: Give the FPSCR rounding modes names, Peter Maydell, 2014/01/07
- [Qemu-devel] [PULL 34/76] target-arm: A64: Add support for floating point cond select, Peter Maydell, 2014/01/07
- [Qemu-devel] [PULL 36/76] char/cadence_uart: Mark struct fields as public/private, Peter Maydell, 2014/01/07