Andrew Morton wrote:
>
> Artur Skawina wrote:
> >
> > for example on p2 the extra checksumming cost (vs a plain copy) is ~7%,
> > and that's the worst case, ie everything cached. Add cache misses and
> > the difference won't be visible...
i didn't have the numbers in front of me and managed to remember the
wrong figure :( Sorry. It really looks more like this:
TIME-N+S TIME32 TIME33 TIME1480 TIMEXXXX FUNCTION
22109 9978 13303 18693 25684 csum_partial_copy_generic_686as2
17609 12194 12748 10207 22893 kernel_memcpy686as2
ie it depends on the size of frame, and for eth sized chunks overhead is ~83%.
> Ah. That's pretty convincing. But it would be a huge win for zero-copy
hmm, the quick check i just did seems to suggest the difference might not
disappear with a cold cache either. Did anyone try replacing the
checksum-copy with a plain copy, made the kernel ignore checksum errors
and benchmarked against a normal one? That would give an idea of the
true sw checksum impact (ignoring the extra hw support overhead).
|