On Tue, 2012-01-31 at 09:48 +0100, Eric Dumazet wrote: > Le mardi 31 janvier 2012 à 15:59 +0800, Shaohua Li a écrit : > > Herbert Poetzl reported a performance regression since 2.6.39. The test > > is a simple dd read, but with big block size. The reason is: > > > > T1: ra (A, A+128k), (A+128k, A+256k) > > T2: lock_page for page A, submit the 256k > > T3: hit page A+128K, ra (A+256k, A+384). the range isn't submitted > > because of plug and there isn't any lock_page till we hit page A+256k > > because all pages from A to A+256k is in memory > > T4: hit page A+256k, ra (A+384, A+ 512). Because of plug, the range isn't > > submitted again. > > T5: lock_page A+256k, so (A+256k, A+512k) will be submitted. The task is > > waitting for (A+256k, A+512k) finish. > > > > There is no request to disk in T3 and T4, so readahead pipeline breaks. > > > > We really don't need block plug for generic_file_aio_read() for buffered > > I/O. The readahead already has plug and has fine grained control when I/O > > should be submitted. Deleting plug for buffered I/O fixes the regression. > > > > One side effect is plug makes the request size 256k, the size is 128k > > without it. This is because default ra size is 128k and not a reason we > > need plug here. > > > > Signed-off-by: Shaohua Li <shaohua.li@xxxxxxxxx> > > Tested-by: Herbert Poetzl <herbert@xxxxxxxxxxxx> > > Tested-by: Eric Dumazet <eric.dumazet@xxxxxxxxx> > > Hmm, this is not exactly the patch I tested from Wu Fengguang > > I'll test this one before adding my "Tested-by: ..." That added lines should not matter. We still need plug for direct-io case. Really sorry for this, I should ask you test it before adding the Tested-by. Thanks, Shaohua -- To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to majordomo@xxxxxxxxx. For more info on Linux MM, see: http://www.linux-mm.org/ . Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/ Don't email: <a href=mailto:"dont@xxxxxxxxx";> email@xxxxxxxxx </a>