2011/4/24 Steve French <smfrench@xxxxxxxxx>: > On Sat, Apr 23, 2011 at 1:49 PM, Pavel Shilovsky <piastry@xxxxxxxxxxx> wrote: >> 2011/4/23 Steve French <smfrench@xxxxxxxxx>: >>> With this patch do you still measure a slight performance degradation >>> (e.g. with dbench) and how much? >>> >> >> I tested it several times and didn't notice any difference between >> this version and try #4 (without filemap_fdatawait calls) in dbench >> (dbench -D dir 1). But, anyway, this patch brings ~5% performance >> itself (in comparison with current state of for-next branch). > > Can you clarify what you mean by ~5%? Did it fix a perf problem? > Sorry for wasn't clear. I mean when I tested it more carefully (with several runs) I got the following results against Windows 7 on LAN (throughput): 1) for-next + try #4 and for-next + try #5 ~ 2.59 MB/sec; 2) for-next ~ 2.48 MB/sec. So, this patch increases performance by 5%. -- Best regards, Pavel Shilovsky. -- To unsubscribe from this list: send the line "unsubscribe linux-cifs" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
