Skip to content

Commit

Permalink
NFSv4.1: Get rid of NFS4CLNT_LAYOUTRECALL
Browse files Browse the repository at this point in the history
The NFS4CLNT_LAYOUTRECALL bit is a long-term impediment to scalability. It
basically stops all other recalls by a given server once any layout recall
is requested.

If the recall is for a different file, then we don't care.
If the recall applies to the same file, then we're in one of two situations:
Either we are in the case of a replay of an existing request, in which case
the session is supposed to deal with matters, or we are dealing with a
completely different request, in which case we should just try to process
it.

Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
  • Loading branch information
Trond Myklebust authored and Trond Myklebust committed Mar 1, 2012
1 parent a59c30a commit 0cb3284
Show file tree
Hide file tree
Showing 2 changed files with 1 addition and 6 deletions.
6 changes: 1 addition & 5 deletions fs/nfs/callback_proc.c
Original file line number Diff line number Diff line change
Expand Up @@ -233,17 +233,13 @@ static u32 initiate_bulk_draining(struct nfs_client *clp,
static u32 do_callback_layoutrecall(struct nfs_client *clp,
struct cb_layoutrecallargs *args)
{
u32 res = NFS4ERR_DELAY;
u32 res;

dprintk("%s enter, type=%i\n", __func__, args->cbl_recall_type);
if (test_and_set_bit(NFS4CLNT_LAYOUTRECALL, &clp->cl_state))
goto out;
if (args->cbl_recall_type == RETURN_FILE)
res = initiate_file_draining(clp, args);
else
res = initiate_bulk_draining(clp, args);
clear_bit(NFS4CLNT_LAYOUTRECALL, &clp->cl_state);
out:
dprintk("%s returning %i\n", __func__, res);
return res;

Expand Down
1 change: 0 additions & 1 deletion fs/nfs/nfs4_fs.h
Original file line number Diff line number Diff line change
Expand Up @@ -20,7 +20,6 @@ enum nfs4_client_state {
NFS4CLNT_RECLAIM_REBOOT,
NFS4CLNT_RECLAIM_NOGRACE,
NFS4CLNT_DELEGRETURN,
NFS4CLNT_LAYOUTRECALL,
NFS4CLNT_SESSION_RESET,
NFS4CLNT_RECALL_SLOT,
NFS4CLNT_LEASE_CONFIRM,
Expand Down

0 comments on commit 0cb3284

Please sign in to comment.