在 2022/9/28 下午11:14, Richard Henderson 写道:
On 9/26/22 23:48, Song Gao wrote:
we just set high 32bit 0xffffffff as the other float instructions do.
Signed-off-by: Song Gao<gaos...@loongson.cn>
---
target/loongarch/fpu_helper.c | 18 +++++++++---------
1 file changed, 9 insertions(+), 9 deletions(-)
But the result in these cases is an integer, not a (single-precision)
float.
Is this really what hardware does?
The high 32bit value is not fixed as the manual 3.1.3.1 said:
' When the floating-point register records a single-precision
floating-point number or word integer, the data
always appears in the [31:0] bits of the floating-point register,
at this time the [63:32] bits of the
floating-point register can be any value.'
I do this just used for RISU test compare these instructions result value.
As the RISU patches not reviewed, I can drop this patch.
Thanks.
Song Gao
