> 在 2018年5月30日，上午10:16，Yan, Zheng <zyan@xxxxxxxxxx> 写道： > > > >> On May 30, 2018, at 10:13, Chengguang Xu <cgxu519@xxxxxxx> wrote: >> >> Fix improper rasize alignment. >> >> Based on code, rasize(max readahead) actually aligns to PAGESIZE not >> 1024 which is documented in below link, should fix the document as well. >> >> docs.ceph.com/docs/master/man/8/mount.ceph/ >> >> Signed-off-by: Chengguang Xu <cgxu519@xxxxxxx> >> --- >> fs/ceph/super.c | 2 +- >> 1 file changed, 1 insertion(+), 1 deletion(-) >> >> diff --git a/fs/ceph/super.c b/fs/ceph/super.c >> index b33082e6878f..465861959a6b 100644 >> --- a/fs/ceph/super.c >> +++ b/fs/ceph/super.c >> @@ -268,7 +268,7 @@ static int parse_fsopt_token(char *c, void *private) >> case Opt_rasize: >> if (intval < 0) >> return -EINVAL; >> - fsopt->rasize = ALIGN(intval + PAGE_SIZE - 1, PAGE_SIZE); >> + fsopt->rasize = ALIGN(intval, PAGE_SIZE); > > if intval is 2048, ALIGN(intval, PAGE_SIZE) is 0. why do you want to make this change? Hi Yan, Thanks for you quick reply. I think it aligns up, so it will be 4096 unless you specify rasize=0. The motivation of this patch is avoiding confusion about specifying rasize. On currently logic: when I specify rasize=0~1 then it will be 4096. when I specify rasize=2~4097 then it will be 8192. I think the result will easily cause user’s confusion. Thanks, Chengguang. -- To unsubscribe from this list: send the line "unsubscribe ceph-devel" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
