Re: [MirageOS-devel] Timestamp representation and CLOCK

[Daniel BÃnzli <daniel.buenzli@xxxxxxxxxxxx>]

Le jeudi, 9 juillet 2015 Ã 17:29, David Sheets a Ãcrit :
> This gives 1 word for tuple, 1 word for int64 pointer, 1 word for
> int64 header, 1 word for int, 8 bytes for int64 data = 4*word + 8
> bytes = 24 bytes on 32-bit = 40 bytes on 64-bit
>  
> :-( :-(
  
Still designing with nanosecond precision between ptime's boundaries in mind 
(since I will support that anyway on my side there) one could actually 
interpret int * int * int as being a POSIX day count from the epoch, a second 
in that day, and a nanosecond in that second. That would make it 4 words per 
timestamp, i.e. 16 bytes on 32-bit and 32 bytes on 64-bit. It's just a little 
bit painful to work with.

If people only care about 100ns precision then a single int64 will do which 
would be 12 bytes on 32-bit and 16 bytes on 64-bit.  

Daniel



_______________________________________________
MirageOS-devel mailing list
MirageOS-devel@xxxxxxxxxxxxxxxxxxxx
http://lists.xenproject.org/cgi-bin/mailman/listinfo/mirageos-devel