On Thu, Apr 25, 2024 at 08:15:30PM +0200, Tim Düsterhus wrote: > Hi > > On 4/24/24 08:39, Willy Tarreau wrote: > > Just thinking about all the shifts above, I think you could have > > gone through less efforts by acting on 64-bit randoms (less shifts). > > But the difference is probably not that much anyway. > > I've used the existing implementation for UUIDv4 as the basis, that's why > the code looks like it does.
Ah OK makes sense ;-) Willy
