On Wed, Jul 15, 2009 at 02:52:27AM +0800, Vladislav Bolkhovitin wrote: > > Wu Fengguang, on 07/13/2009 04:36 PM wrote: > >> Test done with XFS on both the target and the initiator. This confirms > >> your findings, using files instead of block devices is faster, but > >> only when using the io_context patch. > > > > It shows that the one really matters is the io_context patch, > > even when context readahead is running. I guess what happened > > in the tests are: > > - without readahead (or readahead algorithm failed to do proper > > sequential readaheads), the SCST processes will be submitting > > small but close to each other IOs. CFQ relies on the io_context > > patch to prevent unnecessary idling. > > - with proper readahead, the SCST processes will also be submitting > > close readahead IOs. For example, one file's 100-102MB pages is > > readahead by process A, while its 102-104MB pages may be > > readahead by process B. In this case CFQ will also idle waiting > > for process A to submit the next IO, but in fact that IO is being > > submitted by process B. So the io_context patch is still necessary > > even when context readahead is working fine. I guess context > > readahead do have the added value of possibly enlarging the IO size > > (however this benchmark seems to not very sensitive to IO size). > > Looks like the truth. Although with 2MB RA I expect CFQ to do idling >10 > times less, which should bring bigger improvement than few %%. > > For how long CFQ idles? For HZ/125, i.e. 8 ms with HZ 250? Yes, 8ms by default. Note that the 8ms idle time is armed when the last IO from current process completes. So it would be definitely a waste if the cooperative process submitted the next read/readahead IO within this 8ms idle window (without cfq_coop.patch). Thanks, Fengguang -- To unsubscribe from this list: send the line "unsubscribe linux-fsdevel" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
