T0b1/wasmtime
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Update comment on fcvt_low_from_sint instruction

Browse Source
This commit is contained in:
Johnnie Birch
2021-06-14 10:31:19 -07:00
parent 2d676d838f
commit 6dd2df4fb3
1 changed files with 3 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files

5
cranelift/codegen/meta/src/shared/instructions.rs
View File

@@ -4441,10 +4441,10 @@ pub(crate) fn define(
Inst::new( Inst::new(
"fcvt_low_from_sint", "fcvt_low_from_sint",
r#" r#"
Converts packed signed doubleword integers to packed double precision floating point. Converts packed signed 32-bit integers to packed double precision floating point.
Considering only the low half of the register, each lane in `x` is interpreted as a Considering only the low half of the register, each lane in `x` is interpreted as a
signed doubleword integer that is then converted to a double precision float. This signed 32-bit integer that is then converted to a double precision float. This
instruction differs from fcvt_from_sint in that it converts half the number of lanes instruction differs from fcvt_from_sint in that it converts half the number of lanes
which are converted to occupy twice the number of bits. No rounding should be needed which are converted to occupy twice the number of bits. No rounding should be needed
for the resulting float. for the resulting float.
@@ -4466,6 +4466,7 @@ pub(crate) fn define(
Considering only the low half of the register, each lane in `x` is interpreted as a Considering only the low half of the register, each lane in `x` is interpreted as a
unsigned 32-bit integer that is then converted to a double precision float. This unsigned 32-bit integer that is then converted to a double precision float. This
instruction differs from fcvt_from_uint in that it converts half the number of lanes
which are converted to occupy twice the number of bits. No rounding should be needed which are converted to occupy twice the number of bits. No rounding should be needed
for the resulting float. for the resulting float.