Has it been ever fixed for Big-endian platforms?

[barracuda156]

The original assumed LE archs: http://burtleburtle.net/bob/hash/spooky.html
I wonder if it has been fixed ever since.

[andikleen]

On Fri, Jun 21, 2024 at 05:14:19PM -0700, Sergey Fedorov wrote: The original assumed LE archs: http://burtleburtle.net/bob/hash/spooky.html I wonder if it has been fixed ever since.

Not that I know of. But perhaps it just works assuming you don't want to interoperate with a LE producer.? After all hash functions may be tolerant of some permutations. Perhaps just run it through smhasher?

[barracuda156]

Thank you. Yeah, upon reading a few references I came to a conclusion that it should just not be compared to LE value-wise. But algorithm itself should work fine.