pNFS/flexfiles: Record the RPC errors in the I/O tracepoints

When debugging I/O issues, we want to see not just the NFS level errors,
but also the RPC level problems, so record both in the tracepoints.

Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
This commit is contained in:
Trond Myklebust
2025-04-07 14:36:41 +02:00
parent 440caf8ee2
commit bead8b4953
2 changed files with 25 additions and 15 deletions

View File

@@ -1329,7 +1329,7 @@ static int ff_layout_read_done_cb(struct rpc_task *task,
hdr->args.offset, hdr->args.count,
&hdr->res.op_status, OP_READ,
task->tk_status);
trace_ff_layout_read_error(hdr);
trace_ff_layout_read_error(hdr, task->tk_status);
}
err = ff_layout_async_handle_error(task, hdr->args.context->state,
@@ -1502,7 +1502,7 @@ static int ff_layout_write_done_cb(struct rpc_task *task,
hdr->args.offset, hdr->args.count,
&hdr->res.op_status, OP_WRITE,
task->tk_status);
trace_ff_layout_write_error(hdr);
trace_ff_layout_write_error(hdr, task->tk_status);
}
err = ff_layout_async_handle_error(task, hdr->args.context->state,
@@ -1551,7 +1551,7 @@ static int ff_layout_commit_done_cb(struct rpc_task *task,
data->args.offset, data->args.count,
&data->res.op_status, OP_COMMIT,
task->tk_status);
trace_ff_layout_commit_error(data);
trace_ff_layout_commit_error(data, task->tk_status);
}
err = ff_layout_async_handle_error(task, NULL, data->ds_clp,

View File

@@ -2051,13 +2051,15 @@ TRACE_EVENT(fl_getdevinfo,
DECLARE_EVENT_CLASS(nfs4_flexfiles_io_event,
TP_PROTO(
const struct nfs_pgio_header *hdr
const struct nfs_pgio_header *hdr,
int error
),
TP_ARGS(hdr),
TP_ARGS(hdr, error),
TP_STRUCT__entry(
__field(unsigned long, error)
__field(unsigned long, nfs_error)
__field(dev_t, dev)
__field(u32, fhandle)
__field(u64, fileid)
@@ -2073,7 +2075,8 @@ DECLARE_EVENT_CLASS(nfs4_flexfiles_io_event,
TP_fast_assign(
const struct inode *inode = hdr->inode;
__entry->error = hdr->res.op_status;
__entry->error = -error;
__entry->nfs_error = hdr->res.op_status;
__entry->fhandle = nfs_fhandle_hash(hdr->args.fh);
__entry->fileid = NFS_FILEID(inode);
__entry->dev = inode->i_sb->s_dev;
@@ -2088,7 +2091,8 @@ DECLARE_EVENT_CLASS(nfs4_flexfiles_io_event,
TP_printk(
"error=%ld (%s) fileid=%02x:%02x:%llu fhandle=0x%08x "
"offset=%llu count=%u stateid=%d:0x%08x dstaddr=%s",
"offset=%llu count=%u stateid=%d:0x%08x dstaddr=%s "
"nfs_error=%lu (%s)",
-__entry->error,
show_nfs4_status(__entry->error),
MAJOR(__entry->dev), MINOR(__entry->dev),
@@ -2096,28 +2100,32 @@ DECLARE_EVENT_CLASS(nfs4_flexfiles_io_event,
__entry->fhandle,
__entry->offset, __entry->count,
__entry->stateid_seq, __entry->stateid_hash,
__get_str(dstaddr)
__get_str(dstaddr), __entry->nfs_error,
show_nfs4_status(__entry->nfs_error)
)
);
#define DEFINE_NFS4_FLEXFILES_IO_EVENT(name) \
DEFINE_EVENT(nfs4_flexfiles_io_event, name, \
TP_PROTO( \
const struct nfs_pgio_header *hdr \
const struct nfs_pgio_header *hdr, \
int error \
), \
TP_ARGS(hdr))
TP_ARGS(hdr, error))
DEFINE_NFS4_FLEXFILES_IO_EVENT(ff_layout_read_error);
DEFINE_NFS4_FLEXFILES_IO_EVENT(ff_layout_write_error);
TRACE_EVENT(ff_layout_commit_error,
TP_PROTO(
const struct nfs_commit_data *data
const struct nfs_commit_data *data,
int error
),
TP_ARGS(data),
TP_ARGS(data, error),
TP_STRUCT__entry(
__field(unsigned long, error)
__field(unsigned long, nfs_error)
__field(dev_t, dev)
__field(u32, fhandle)
__field(u64, fileid)
@@ -2131,7 +2139,8 @@ TRACE_EVENT(ff_layout_commit_error,
TP_fast_assign(
const struct inode *inode = data->inode;
__entry->error = data->res.op_status;
__entry->error = -error;
__entry->nfs_error = data->res.op_status;
__entry->fhandle = nfs_fhandle_hash(data->args.fh);
__entry->fileid = NFS_FILEID(inode);
__entry->dev = inode->i_sb->s_dev;
@@ -2142,14 +2151,15 @@ TRACE_EVENT(ff_layout_commit_error,
TP_printk(
"error=%ld (%s) fileid=%02x:%02x:%llu fhandle=0x%08x "
"offset=%llu count=%u dstaddr=%s",
"offset=%llu count=%u dstaddr=%s nfs_error=%lu (%s)",
-__entry->error,
show_nfs4_status(__entry->error),
MAJOR(__entry->dev), MINOR(__entry->dev),
(unsigned long long)__entry->fileid,
__entry->fhandle,
__entry->offset, __entry->count,
__get_str(dstaddr)
__get_str(dstaddr), __entry->nfs_error,
show_nfs4_status(__entry->nfs_error)
)
);