When triggering an nfs_xdr_status trace point, record the task ID and XID of the failing RPC to better pinpoint the problem. This feels like a bit of a layering violation. Suggested-by: Trond Myklebust <trondmy@xxxxxxxxxxxxxxx> Signed-off-by: Chuck Lever <chuck.lever@xxxxxxxxxx> --- fs/nfs/nfs2xdr.c | 2 +- fs/nfs/nfs3xdr.c | 2 +- fs/nfs/nfs4trace.h | 15 +++++++++++++-- fs/nfs/nfs4xdr.c | 2 +- fs/nfs/nfstrace.h | 15 +++++++++++++-- 5 files changed, 29 insertions(+), 7 deletions(-) diff --git a/fs/nfs/nfs2xdr.c b/fs/nfs/nfs2xdr.c index 572794d..cbc17a2 100644 --- a/fs/nfs/nfs2xdr.c +++ b/fs/nfs/nfs2xdr.c @@ -151,7 +151,7 @@ static int decode_stat(struct xdr_stream *xdr, enum nfs_stat *status) return 0; out_status: *status = be32_to_cpup(p); - trace_nfs_xdr_status((int)*status); + trace_nfs_xdr_status(xdr, (int)*status); return 0; } diff --git a/fs/nfs/nfs3xdr.c b/fs/nfs/nfs3xdr.c index abbbdde..6027678 100644 --- a/fs/nfs/nfs3xdr.c +++ b/fs/nfs/nfs3xdr.c @@ -343,7 +343,7 @@ static int decode_nfsstat3(struct xdr_stream *xdr, enum nfs_stat *status) return 0; out_status: *status = be32_to_cpup(p); - trace_nfs_xdr_status((int)*status); + trace_nfs_xdr_status(xdr, (int)*status); return 0; } diff --git a/fs/nfs/nfs4trace.h b/fs/nfs/nfs4trace.h index 9a39fd5..d85f209 100644 --- a/fs/nfs/nfs4trace.h +++ b/fs/nfs/nfs4trace.h @@ -564,24 +564,35 @@ TRACE_EVENT(nfs4_xdr_status, TP_PROTO( + const struct xdr_stream *xdr, u32 op, int error ), - TP_ARGS(op, error), + TP_ARGS(xdr, op, error), TP_STRUCT__entry( + __field(unsigned int, task_id) + __field(unsigned int, client_id) + __field(u32, xid) __field(u32, op) __field(unsigned long, error) ), TP_fast_assign( + const struct rpc_rqst *rqstp = xdr->rqst; + const struct rpc_task *task = rqstp->rq_task; + + __entry->task_id = task->tk_pid; + __entry->client_id = task->tk_client->cl_clid; + __entry->xid = be32_to_cpu(rqstp->rq_xid); __entry->op = op; __entry->error = error; ), TP_printk( - "error=%ld (%s) operation %d:", + "task:%u@%d xid=0x%08x error=%ld (%s) operation=%u", + __entry->task_id, __entry->client_id, __entry->xid, -__entry->error, show_nfsv4_errors(__entry->error), __entry->op ) diff --git a/fs/nfs/nfs4xdr.c b/fs/nfs/nfs4xdr.c index 6024461..d974ff3 100644 --- a/fs/nfs/nfs4xdr.c +++ b/fs/nfs/nfs4xdr.c @@ -3187,7 +3187,7 @@ static bool __decode_op_hdr(struct xdr_stream *xdr, enum nfs_opnum4 expected, return true; out_status: nfserr = be32_to_cpup(p); - trace_nfs4_xdr_status(opnum, nfserr); + trace_nfs4_xdr_status(xdr, opnum, nfserr); *nfs_retval = nfs4_stat_to_errno(nfserr); return true; out_bad_operation: diff --git a/fs/nfs/nfstrace.h b/fs/nfs/nfstrace.h index cd09356..976d408 100644 --- a/fs/nfs/nfstrace.h +++ b/fs/nfs/nfstrace.h @@ -1139,21 +1139,32 @@ TRACE_EVENT(nfs_xdr_status, TP_PROTO( + const struct xdr_stream *xdr, int error ), - TP_ARGS(error), + TP_ARGS(xdr, error), TP_STRUCT__entry( + __field(unsigned int, task_id) + __field(unsigned int, client_id) + __field(u32, xid) __field(unsigned long, error) ), TP_fast_assign( + const struct rpc_rqst *rqstp = xdr->rqst; + const struct rpc_task *task = rqstp->rq_task; + + __entry->task_id = task->tk_pid; + __entry->client_id = task->tk_client->cl_clid; + __entry->xid = be32_to_cpu(rqstp->rq_xid); __entry->error = error; ), TP_printk( - "error=%ld (%s)", + "task:%u@%d xid=0x%08x error=%ld (%s)", + __entry->task_id, __entry->client_id, __entry->xid, -__entry->error, nfs_show_status(__entry->error) ) );
