In testing the cifs-2.6.git tree for 3.2-pre, I noticed that dbench cifs dbench performance improved even over localhost mounts (to Samba) by more than 5% (60 processes). I don't know what part of that is due to the improved byte range lock caching (Pavel's patch set) or larger i/o sizes on read or better async read dispatch (Jeff's patch set), but suspect that the latter helped more. Relative performance should improve even more with fewer dbench processes as better async dispatch of large read requests helps utilize the Samba server more efficiently.
-- Thanks, Steve -- To unsubscribe from this list: send the line "unsubscribe linux-cifs" in the body of a message to [email protected] More majordomo info at http://vger.kernel.org/majordomo-info.html
