[vector] ByteBuffer byte ordering
john.r.rose at oracle.com
Fri Apr 13 22:00:42 UTC 2018
On Apr 13, 2018, at 2:49 PM, John Rose <john.r.rose at oracle.com> wrote:
> On Apr 13, 2018, at 2:39 PM, Paul Sandoz <paul.sandoz at oracle.com <mailto:paul.sandoz at oracle.com>> wrote:
>> I am unsure about the need for such explanatory in the reshape/rebracket, ok for now, can revisit later.
> BTW, if a rebracket operation is given a byte-order parameter, then it becomes,
> as a special case, a pure byte swap operator if the rebracketing is done
> with the same lane size.
And by another way, a generalized byte-swap operation (on power-of-two
sized operands) is just a Hackers Delight hypercube reflection, specifiable
as a mask of index bits to complement. That all by itself is a candidate
operation for vectors, as are hypercube rotations, which correspond to
various kinds of transpose operations (possibly partitioned); see HD 7-8.
More information about the panama-dev