It's working well on my box. I modified the crc32c methods to optionally take a null data pointer, in which case they behave as if it were zero-filled. For the sctp code this speeds it up by almost 2x. Overall, it's looking pretty good!
The one missing piece is that the intel optimized assembler code needs to be modified to handle the null buffer case before this can be merged. (As is it will just crash if sse4.2 is there.) sage -- To unsubscribe from this list: send the line "unsubscribe ceph-devel" in the body of a message to [email protected] More majordomo info at http://vger.kernel.org/majordomo-info.html
