Minor correction to __m512d documentation.
A 512-bit register is f64x8, not f64x4. Likely a copy-paste error from the _m256d documentation, which seems correct.
This commit is contained in:
parent
1c136ddc5a
commit
a083e2a6c8
1 changed files with 1 additions and 1 deletions
|
|
@ -340,7 +340,7 @@ types! {
|
|||
///
|
||||
/// Note that unlike `__m512i`, the integer version of the 512-bit
|
||||
/// registers, this `__m512d` type has *one* interpretation. Each instance
|
||||
/// of `__m512d` always corresponds to `f64x4`, or eight `f64` types packed
|
||||
/// of `__m512d` always corresponds to `f64x8`, or eight `f64` types packed
|
||||
/// together.
|
||||
///
|
||||
/// The in-memory representation of this type is the same as the one of an
|
||||
|
|
|
|||
Loading…
Add table
Add a link
Reference in a new issue