There's a hard limit on how large a read we can do from the OSD, defined as CEPH_MSG_MAX_DATA_LEN (currently, 64M). It's possible to create a file that is backed by larger objects than that (and indeed, xfstest ceph/001 does just that). Ensure we clamp the final length of a read to CEPH_MSG_MAX_DATA_LEN. Signed-off-by: Jeff Layton <jlayton@xxxxxxxxxx> --- fs/ceph/addr.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) I'm likely going to fold this patch into the one that introduces ceph_netfs_issue_op, so we don't have a regression in the series. diff --git a/fs/ceph/addr.c b/fs/ceph/addr.c index 9939100f9f9d..ba459b15604d 100644 --- a/fs/ceph/addr.c +++ b/fs/ceph/addr.c @@ -212,7 +212,7 @@ static bool ceph_netfs_clamp_length(struct netfs_read_subrequest *subreq) /* Truncate the extent at the end of the current block */ ceph_calc_file_object_mapping(&ci->i_layout, subreq->start, subreq->len, &objno, &objoff, &xlen); - subreq->len = xlen; + subreq->len = min(xlen, (u32)CEPH_MSG_MAX_DATA_LEN); return true; } -- 2.31.1
