On Mon, 13 Jan 2014, Mike Snitzer wrote: > The patch header should speak to why 12 is adequate. Your 0th patch > header said we should do further testing to arrive at a good value. 12 is optimal on my system (dm-loop on filesystem on 15k SCSI disk). > I think a follow-on patch should make the 'prefetch_chunks' configurable. I think the user wouldn't configure it anyway. Find a configuration scenario where this request for prefetch came from and find the optimal value for that scenario. > The dm_exception_store structure has a 4 byte hole before the context > member. Making prefetch_chunks configurable (either from table load or > dm-snapshot module param) would help future-proof this code. Mikulas -- dm-devel mailing list dm-devel@xxxxxxxxxx https://www.redhat.com/mailman/listinfo/dm-devel
