Messages in this thread | | | Date | Tue, 18 Mar 2014 12:43:28 +0530 | From | Raghavendra K T <> | Subject | Re: [PATCH V6 ] mm readahead: Fix readahead fail for memoryless cpu and limit readahead pages |
| |
On 03/17/2014 07:37 AM, Madper Xie wrote: > > Raghavendra K T <raghavendra.kt@linux.vnet.ibm.com> writes: > >> On 02/18/2014 03:19 PM, Jan Kara wrote: >>> On Tue 18-02-14 12:55:38, Raghavendra K T wrote: > Hi. Just a concern. Will the performance reduce on some special storage > backend? E.g. tape. > The existent applications may using readahead for userspace I/O schedule > to decrease seeking time.
I have not tested the patch on such systems yet unfortunately :(. Sequential read with huge file has not suffered on disk based system, but I think, I should be honest enough not to guess the effect on tape.
| |