2023-09-13 15:12:41

by Dai Ngo

[permalink] [raw]
Subject: [PATCH 0/2] NFSD: use CB_GETATTR to handle GETATTR conflict with write delegation

Currently GETATTR conflict with a write delegation is handled by
recalling the delegation before replying to the GETATTR.

This patch series add supports for CB_GETATTR callback to get the latest
change_info and size information of the file from the client that holds
the delegation to reply to the GETATTR from the second client.


2023-09-13 15:26:51

by Dai Ngo

[permalink] [raw]
Subject: [PATCH 1/2] NFSD: add supports for CB_GETATTR callback

Includes:
. CB_GETATTR proc for nfs4_cb_procedures[]
. XDR encoding and decoding function for CB_GETATTR request/reply
. add nfs4_cb_fattr to nfs4_delegation for sending CB_GETATTR
and store file attributes from client's reply.

Signed-off-by: Dai Ngo <[email protected]>
---
fs/nfsd/nfs4callback.c | 97 +++++++++++++++++++++++++++++++++++++++++-
fs/nfsd/state.h | 14 ++++++
fs/nfsd/xdr4cb.h | 18 ++++++++
3 files changed, 128 insertions(+), 1 deletion(-)

diff --git a/fs/nfsd/nfs4callback.c b/fs/nfsd/nfs4callback.c
index 4039ffcf90ba..92bc109dabe6 100644
--- a/fs/nfsd/nfs4callback.c
+++ b/fs/nfsd/nfs4callback.c
@@ -84,7 +84,21 @@ static void encode_uint32(struct xdr_stream *xdr, u32 n)
static void encode_bitmap4(struct xdr_stream *xdr, const __u32 *bitmap,
size_t len)
{
- WARN_ON_ONCE(xdr_stream_encode_uint32_array(xdr, bitmap, len) < 0);
+ xdr_stream_encode_uint32_array(xdr, bitmap, len);
+}
+
+static int decode_cb_fattr4(struct xdr_stream *xdr, uint32_t *bitmap,
+ struct nfs4_cb_fattr *fattr)
+{
+ fattr->ncf_cb_change = 0;
+ fattr->ncf_cb_fsize = 0;
+ if (bitmap[0] & FATTR4_WORD0_CHANGE)
+ if (xdr_stream_decode_u64(xdr, &fattr->ncf_cb_change) < 0)
+ return -NFSERR_BAD_XDR;
+ if (bitmap[0] & FATTR4_WORD0_SIZE)
+ if (xdr_stream_decode_u64(xdr, &fattr->ncf_cb_fsize) < 0)
+ return -NFSERR_BAD_XDR;
+ return 0;
}

/*
@@ -357,6 +371,30 @@ encode_cb_recallany4args(struct xdr_stream *xdr,
hdr->nops++;
}

+/*
+ * CB_GETATTR4args
+ * struct CB_GETATTR4args {
+ * nfs_fh4 fh;
+ * bitmap4 attr_request;
+ * };
+ *
+ * The size and change attributes are the only one
+ * guaranteed to be serviced by the client.
+ */
+static void
+encode_cb_getattr4args(struct xdr_stream *xdr, struct nfs4_cb_compound_hdr *hdr,
+ struct nfs4_cb_fattr *fattr)
+{
+ struct nfs4_delegation *dp =
+ container_of(fattr, struct nfs4_delegation, dl_cb_fattr);
+ struct knfsd_fh *fh = &dp->dl_stid.sc_file->fi_fhandle;
+
+ encode_nfs_cb_opnum4(xdr, OP_CB_GETATTR);
+ encode_nfs_fh4(xdr, fh);
+ encode_bitmap4(xdr, fattr->ncf_cb_bmap, ARRAY_SIZE(fattr->ncf_cb_bmap));
+ hdr->nops++;
+}
+
/*
* CB_SEQUENCE4args
*
@@ -492,6 +530,26 @@ static void nfs4_xdr_enc_cb_null(struct rpc_rqst *req, struct xdr_stream *xdr,
xdr_reserve_space(xdr, 0);
}

+/*
+ * 20.1. Operation 3: CB_GETATTR - Get Attributes
+ */
+static void nfs4_xdr_enc_cb_getattr(struct rpc_rqst *req,
+ struct xdr_stream *xdr, const void *data)
+{
+ const struct nfsd4_callback *cb = data;
+ struct nfs4_cb_fattr *ncf =
+ container_of(cb, struct nfs4_cb_fattr, ncf_getattr);
+ struct nfs4_cb_compound_hdr hdr = {
+ .ident = cb->cb_clp->cl_cb_ident,
+ .minorversion = cb->cb_clp->cl_minorversion,
+ };
+
+ encode_cb_compound4args(xdr, &hdr);
+ encode_cb_sequence4args(xdr, cb, &hdr);
+ encode_cb_getattr4args(xdr, &hdr, ncf);
+ encode_cb_nops(&hdr);
+}
+
/*
* 20.2. Operation 4: CB_RECALL - Recall a Delegation
*/
@@ -547,6 +605,42 @@ static int nfs4_xdr_dec_cb_null(struct rpc_rqst *req, struct xdr_stream *xdr,
return 0;
}

+/*
+ * 20.1. Operation 3: CB_GETATTR - Get Attributes
+ */
+static int nfs4_xdr_dec_cb_getattr(struct rpc_rqst *rqstp,
+ struct xdr_stream *xdr,
+ void *data)
+{
+ struct nfsd4_callback *cb = data;
+ struct nfs4_cb_compound_hdr hdr;
+ int status;
+ u32 bitmap[3] = {0};
+ u32 attrlen;
+ struct nfs4_cb_fattr *ncf =
+ container_of(cb, struct nfs4_cb_fattr, ncf_getattr);
+
+ status = decode_cb_compound4res(xdr, &hdr);
+ if (unlikely(status))
+ return status;
+
+ status = decode_cb_sequence4res(xdr, cb);
+ if (unlikely(status || cb->cb_seq_status))
+ return status;
+
+ status = decode_cb_op_status(xdr, OP_CB_GETATTR, &cb->cb_status);
+ if (status)
+ return status;
+ if (xdr_stream_decode_uint32_array(xdr, bitmap, 3) < 0)
+ return -NFSERR_BAD_XDR;
+ if (xdr_stream_decode_u32(xdr, &attrlen) < 0)
+ return -NFSERR_BAD_XDR;
+ if (attrlen > (sizeof(ncf->ncf_cb_change) + sizeof(ncf->ncf_cb_fsize)))
+ return -NFSERR_BAD_XDR;
+ status = decode_cb_fattr4(xdr, bitmap, ncf);
+ return status;
+}
+
/*
* 20.2. Operation 4: CB_RECALL - Recall a Delegation
*/
@@ -855,6 +949,7 @@ static const struct rpc_procinfo nfs4_cb_procedures[] = {
PROC(CB_NOTIFY_LOCK, COMPOUND, cb_notify_lock, cb_notify_lock),
PROC(CB_OFFLOAD, COMPOUND, cb_offload, cb_offload),
PROC(CB_RECALL_ANY, COMPOUND, cb_recall_any, cb_recall_any),
+ PROC(CB_GETATTR, COMPOUND, cb_getattr, cb_getattr),
};

static unsigned int nfs4_cb_counts[ARRAY_SIZE(nfs4_cb_procedures)];
diff --git a/fs/nfsd/state.h b/fs/nfsd/state.h
index cbddcf484dba..82718d42f3b2 100644
--- a/fs/nfsd/state.h
+++ b/fs/nfsd/state.h
@@ -117,6 +117,16 @@ struct nfs4_cpntf_state {
time64_t cpntf_time; /* last time stateid used */
};

+struct nfs4_cb_fattr {
+ struct nfsd4_callback ncf_getattr;
+ u32 ncf_cb_status;
+ u32 ncf_cb_bmap[1];
+
+ /* from CB_GETATTR reply */
+ u64 ncf_cb_change;
+ u64 ncf_cb_fsize;
+};
+
/*
* Represents a delegation stateid. The nfs4_client holds references to these
* and they are put when it is being destroyed or when the delegation is
@@ -150,6 +160,9 @@ struct nfs4_delegation {
int dl_retries;
struct nfsd4_callback dl_recall;
bool dl_recalled;
+
+ /* for CB_GETATTR */
+ struct nfs4_cb_fattr dl_cb_fattr;
};

#define cb_to_delegation(cb) \
@@ -642,6 +655,7 @@ enum nfsd4_cb_op {
NFSPROC4_CLNT_CB_SEQUENCE,
NFSPROC4_CLNT_CB_NOTIFY_LOCK,
NFSPROC4_CLNT_CB_RECALL_ANY,
+ NFSPROC4_CLNT_CB_GETATTR,
};

/* Returns true iff a is later than b: */
diff --git a/fs/nfsd/xdr4cb.h b/fs/nfsd/xdr4cb.h
index 0d39af1b00a0..e8b00309c449 100644
--- a/fs/nfsd/xdr4cb.h
+++ b/fs/nfsd/xdr4cb.h
@@ -54,3 +54,21 @@
#define NFS4_dec_cb_recall_any_sz (cb_compound_dec_hdr_sz + \
cb_sequence_dec_sz + \
op_dec_sz)
+
+/*
+ * 1: CB_GETATTR opcode (32-bit)
+ * N: file_handle
+ * 1: number of entry in attribute array (32-bit)
+ * 1: entry 0 in attribute array (32-bit)
+ */
+#define NFS4_enc_cb_getattr_sz (cb_compound_enc_hdr_sz + \
+ cb_sequence_enc_sz + \
+ 1 + enc_nfs4_fh_sz + 1 + 1)
+/*
+ * 4: fattr_bitmap_maxsz
+ * 1: attribute array len
+ * 2: change attr (64-bit)
+ * 2: size (64-bit)
+ */
+#define NFS4_dec_cb_getattr_sz (cb_compound_dec_hdr_sz + \
+ cb_sequence_dec_sz + 4 + 1 + 2 + 2 + op_dec_sz)
--
2.39.3

2023-09-14 18:12:22

by Dai Ngo

[permalink] [raw]
Subject: [PATCH 2/2] NFSD: handle GETATTR conflict with write delegation

If the GETATTR request on a file that has write delegation in effect
and the request attributes include the change info and size attribute
then the request is handled as below:

Server sends CB_GETATTR to client to get the latest change info and file
size. If these values are the same as the server's cached values then
the GETATTR proceeds as normal.

If either the change info or file size is different from the server's
cached values, or the file was already marked as modified, then:

. update time_modify and time_metadata into file's metadata
with current time

. encode GETATTR as normal except the file size is encoded with
the value returned from CB_GETATTR

. mark the file as modified

If the CB_GETATTR fails for any reasons, the delegation is recalled
and NFS4ERR_DELAY is returned for the GETATTR.

Signed-off-by: Dai Ngo <[email protected]>
---
fs/nfsd/nfs4state.c | 106 ++++++++++++++++++++++++++++++++++++++++----
fs/nfsd/nfs4xdr.c | 10 ++++-
fs/nfsd/state.h | 11 ++++-
3 files changed, 115 insertions(+), 12 deletions(-)

diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c
index 8534693eb6a4..2151b89dfd2a 100644
--- a/fs/nfsd/nfs4state.c
+++ b/fs/nfsd/nfs4state.c
@@ -127,6 +127,7 @@ static void free_session(struct nfsd4_session *);

static const struct nfsd4_callback_ops nfsd4_cb_recall_ops;
static const struct nfsd4_callback_ops nfsd4_cb_notify_lock_ops;
+static const struct nfsd4_callback_ops nfsd4_cb_getattr_ops;

static struct workqueue_struct *laundry_wq;

@@ -1187,6 +1188,10 @@ alloc_init_deleg(struct nfs4_client *clp, struct nfs4_file *fp,
dp->dl_recalled = false;
nfsd4_init_cb(&dp->dl_recall, dp->dl_stid.sc_client,
&nfsd4_cb_recall_ops, NFSPROC4_CLNT_CB_RECALL);
+ nfsd4_init_cb(&dp->dl_cb_fattr.ncf_getattr, dp->dl_stid.sc_client,
+ &nfsd4_cb_getattr_ops, NFSPROC4_CLNT_CB_GETATTR);
+ dp->dl_cb_fattr.ncf_file_modified = false;
+ dp->dl_cb_fattr.ncf_cb_bmap[0] = FATTR4_WORD0_CHANGE | FATTR4_WORD0_SIZE;
get_nfs4_file(fp);
dp->dl_stid.sc_file = fp;
return dp;
@@ -2894,11 +2899,56 @@ nfsd4_cb_recall_any_release(struct nfsd4_callback *cb)
spin_unlock(&nn->client_lock);
}

+static int
+nfsd4_cb_getattr_done(struct nfsd4_callback *cb, struct rpc_task *task)
+{
+ struct nfs4_cb_fattr *ncf =
+ container_of(cb, struct nfs4_cb_fattr, ncf_getattr);
+
+ ncf->ncf_cb_status = task->tk_status;
+ switch (task->tk_status) {
+ case -NFS4ERR_DELAY:
+ rpc_delay(task, 2 * HZ);
+ return 0;
+ default:
+ return 1;
+ }
+}
+
+static void
+nfsd4_cb_getattr_release(struct nfsd4_callback *cb)
+{
+ struct nfs4_cb_fattr *ncf =
+ container_of(cb, struct nfs4_cb_fattr, ncf_getattr);
+ struct nfs4_delegation *dp =
+ container_of(ncf, struct nfs4_delegation, dl_cb_fattr);
+
+ nfs4_put_stid(&dp->dl_stid);
+ clear_bit(CB_GETATTR_BUSY, &ncf->ncf_cb_flags);
+ wake_up_bit(&ncf->ncf_cb_flags, CB_GETATTR_BUSY);
+}
+
static const struct nfsd4_callback_ops nfsd4_cb_recall_any_ops = {
.done = nfsd4_cb_recall_any_done,
.release = nfsd4_cb_recall_any_release,
};

+static const struct nfsd4_callback_ops nfsd4_cb_getattr_ops = {
+ .done = nfsd4_cb_getattr_done,
+ .release = nfsd4_cb_getattr_release,
+};
+
+void nfs4_cb_getattr(struct nfs4_cb_fattr *ncf)
+{
+ struct nfs4_delegation *dp =
+ container_of(ncf, struct nfs4_delegation, dl_cb_fattr);
+
+ if (test_and_set_bit(CB_GETATTR_BUSY, &ncf->ncf_cb_flags))
+ return;
+ refcount_inc(&dp->dl_stid.sc_count);
+ nfsd4_run_cb(&ncf->ncf_getattr);
+}
+
static struct nfs4_client *create_client(struct xdr_netobj name,
struct svc_rqst *rqstp, nfs4_verifier *verf)
{
@@ -5634,6 +5684,8 @@ nfs4_open_delegation(struct nfsd4_open *open, struct nfs4_ol_stateid *stp,
struct svc_fh *parent = NULL;
int cb_up;
int status = 0;
+ struct kstat stat;
+ struct path path;

cb_up = nfsd4_cb_channel_good(oo->oo_owner.so_client);
open->op_recall = 0;
@@ -5671,6 +5723,18 @@ nfs4_open_delegation(struct nfsd4_open *open, struct nfs4_ol_stateid *stp,
if (open->op_share_access & NFS4_SHARE_ACCESS_WRITE) {
open->op_delegate_type = NFS4_OPEN_DELEGATE_WRITE;
trace_nfsd_deleg_write(&dp->dl_stid.sc_stateid);
+ path.mnt = currentfh->fh_export->ex_path.mnt;
+ path.dentry = currentfh->fh_dentry;
+ if (vfs_getattr(&path, &stat,
+ (STATX_SIZE | STATX_CTIME | STATX_CHANGE_COOKIE),
+ AT_STATX_SYNC_AS_STAT)) {
+ nfs4_put_stid(&dp->dl_stid);
+ destroy_delegation(dp);
+ goto out_no_deleg;
+ }
+ dp->dl_cb_fattr.ncf_cur_fsize = stat.size;
+ dp->dl_cb_fattr.ncf_initial_cinfo =
+ nfsd4_change_attribute(&stat, d_inode(currentfh->fh_dentry));
} else {
open->op_delegate_type = NFS4_OPEN_DELEGATE_READ;
trace_nfsd_deleg_read(&dp->dl_stid.sc_stateid);
@@ -8411,21 +8475,21 @@ nfsd4_get_writestateid(struct nfsd4_compound_state *cstate,
* delegation before replying to the GETATTR. See RFC 8881 section
* 18.7.4.
*
- * The current implementation does not support CB_GETATTR yet. However
- * this can avoid recalling the delegation could be added in follow up
- * work.
- *
* Returns 0 if there is no conflict; otherwise an nfs_stat
* code is returned.
*/
__be32
-nfsd4_deleg_getattr_conflict(struct svc_rqst *rqstp, struct inode *inode)
+nfsd4_deleg_getattr_conflict(struct svc_rqst *rqstp, struct inode *inode,
+ bool *modified, u64 *size)
{
__be32 status;
struct file_lock_context *ctx;
struct file_lock *fl;
struct nfs4_delegation *dp;
+ struct iattr attrs;
+ struct nfs4_cb_fattr *ncf;

+ *modified = false;
ctx = locks_inode_context(inode);
if (!ctx)
return 0;
@@ -8452,10 +8516,34 @@ nfsd4_deleg_getattr_conflict(struct svc_rqst *rqstp, struct inode *inode)
break_lease:
spin_unlock(&ctx->flc_lock);
nfsd_stats_wdeleg_getattr_inc();
- status = nfserrno(nfsd_open_break_lease(inode, NFSD_MAY_READ));
- if (status != nfserr_jukebox ||
- !nfsd_wait_for_delegreturn(rqstp, inode))
- return status;
+
+ dp = fl->fl_owner;
+ ncf = &dp->dl_cb_fattr;
+ nfs4_cb_getattr(&dp->dl_cb_fattr);
+ wait_on_bit(&ncf->ncf_cb_flags, CB_GETATTR_BUSY, TASK_INTERRUPTIBLE);
+ if (ncf->ncf_cb_status) {
+ status = nfserrno(nfsd_open_break_lease(inode, NFSD_MAY_READ));
+ if (status != nfserr_jukebox ||
+ !nfsd_wait_for_delegreturn(rqstp, inode))
+ return status;
+ }
+ if (!ncf->ncf_file_modified &&
+ (ncf->ncf_initial_cinfo != ncf->ncf_cb_change ||
+ ncf->ncf_cur_fsize != ncf->ncf_cb_fsize))
+ ncf->ncf_file_modified = true;
+ if (ncf->ncf_file_modified) {
+ /*
+ * The server would not update the file's metadata
+ * with the client's modified size.
+ */
+ attrs.ia_mtime = attrs.ia_ctime = current_time(inode);
+ attrs.ia_valid = ATTR_MTIME | ATTR_CTIME;
+ setattr_copy(&nop_mnt_idmap, inode, &attrs);
+ mark_inode_dirty(inode);
+ ncf->ncf_cur_fsize = ncf->ncf_cb_fsize;
+ *size = ncf->ncf_cur_fsize;
+ *modified = true;
+ }
return 0;
}
break;
diff --git a/fs/nfsd/nfs4xdr.c b/fs/nfsd/nfs4xdr.c
index 2e40c74d2f72..69504da6c4d4 100644
--- a/fs/nfsd/nfs4xdr.c
+++ b/fs/nfsd/nfs4xdr.c
@@ -2975,6 +2975,8 @@ nfsd4_encode_fattr(struct xdr_stream *xdr, struct svc_fh *fhp,
.dentry = dentry,
};
struct nfsd_net *nn = net_generic(SVC_NET(rqstp), nfsd_net_id);
+ bool file_modified;
+ u64 size = 0;

BUG_ON(bmval1 & NFSD_WRITEONLY_ATTRS_WORD1);
BUG_ON(!nfsd_attrs_supported(minorversion, bmval));
@@ -2985,7 +2987,8 @@ nfsd4_encode_fattr(struct xdr_stream *xdr, struct svc_fh *fhp,
goto out;
}
if (bmval0 & (FATTR4_WORD0_CHANGE | FATTR4_WORD0_SIZE)) {
- status = nfsd4_deleg_getattr_conflict(rqstp, d_inode(dentry));
+ status = nfsd4_deleg_getattr_conflict(rqstp, d_inode(dentry),
+ &file_modified, &size);
if (status)
goto out;
}
@@ -3112,7 +3115,10 @@ nfsd4_encode_fattr(struct xdr_stream *xdr, struct svc_fh *fhp,
p = xdr_reserve_space(xdr, 8);
if (!p)
goto out_resource;
- p = xdr_encode_hyper(p, stat.size);
+ if (file_modified)
+ p = xdr_encode_hyper(p, size);
+ else
+ p = xdr_encode_hyper(p, stat.size);
}
if (bmval0 & FATTR4_WORD0_LINK_SUPPORT) {
p = xdr_reserve_space(xdr, 4);
diff --git a/fs/nfsd/state.h b/fs/nfsd/state.h
index 82718d42f3b2..6bbb1d0276b4 100644
--- a/fs/nfsd/state.h
+++ b/fs/nfsd/state.h
@@ -125,8 +125,16 @@ struct nfs4_cb_fattr {
/* from CB_GETATTR reply */
u64 ncf_cb_change;
u64 ncf_cb_fsize;
+
+ unsigned long ncf_cb_flags;
+ bool ncf_file_modified;
+ u64 ncf_initial_cinfo;
+ u64 ncf_cur_fsize;
};

+/* bits for ncf_cb_flags */
+#define CB_GETATTR_BUSY 0
+
/*
* Represents a delegation stateid. The nfs4_client holds references to these
* and they are put when it is being destroyed or when the delegation is
@@ -748,5 +756,6 @@ static inline bool try_to_expire_client(struct nfs4_client *clp)
}

extern __be32 nfsd4_deleg_getattr_conflict(struct svc_rqst *rqstp,
- struct inode *inode);
+ struct inode *inode, bool *file_modified, u64 *size);
+extern void nfs4_cb_getattr(struct nfs4_cb_fattr *ncf);
#endif /* NFSD4_STATE_H */
--
2.39.3