Trond Myklebust wrote: > On Tue, 2009-06-09 at 14:18 -0400, Steve Dickson wrote: >> Hey Trond, >> >> Here is an updated version of the patch I sent out a while >> back... all that has changed was the addtion of 'rpc_' to >> the name of the trace points which coincides with the >> other trace points naming convention. >> >> Here is a pointer to the previous email that has the >> fairly long winded explanation of how all this works: >> >> http://marc.info/?l=linux-kernel&m=124058933520183&w=2 >> >> And below is the updated patch which is based on Ingo's >> linux-2.6-tip.git tree. >> >> Comments? >> >> steved. >> >> Author: Steve Dickson <steved@xxxxxxxxxx> >> Date: Tue Jun 9 14:12:07 EDT 2009 >> >> Adds three trace points to the status routines in the >> sunrpc state machine. The trace points will only execute >> in error conditions. >> >> The goal of these trace points it to make it easier to >> debug process that spin or get hung up in the NFS and >> related subsystems. > > Would this replace the existing dprint_status()? How does it differ from > the existing DYNAMIC_DEBUG functionality that was merged in 2.6.30? I believe it would be an addition to... Its my understanding the dynamic debug uses printks, which ultimately causes messages to be log to /var/log/messages... The trace points are used by a systemtap script which is much more efficent and lightweight... > >> Signed-off-by: Steve Dickson <steved@xxxxxxxxxx> >> >> diff --git a/include/trace/events/sunrpc.h b/include/trace/events/sunrpc.h >> new file mode 100644 >> index 0000000..f502eb4 >> --- /dev/null >> +++ b/include/trace/events/sunrpc.h >> @@ -0,0 +1,86 @@ >> +#if !defined(_TRACE_SUNRPC_H) || defined(TRACE_HEADER_MULTI_READ) >> +#define _TRACE_SUNRPC_H >> + >> +#include <linux/tracepoint.h> >> + >> +#undef TRACE_SYSTEM >> +#define TRACE_SYSTEM sunrpc >> + >> +TRACE_EVENT(rpc_call_status, >> + >> + TP_PROTO(struct rpc_task *task), >> + >> + TP_ARGS(task), >> + >> + TP_STRUCT__entry( >> + __field(u32, cl_prog) >> + __field(u32, cl_vers) >> + __field(u32, p_proc) >> + __field(int, tk_status) >> + ), >> + >> + TP_fast_assign( >> + __entry->cl_prog = task->tk_client->cl_prog; >> + __entry->cl_vers = task->tk_client->cl_vers; >> + __entry->p_proc = task->tk_client->cl_procinfo->p_proc; >> + __entry->cl_prog = task->tk_client->cl_prog; > > You should only need to set entry->cl_prog once, right? Point... its redundant... Here is the updated patch: Author: Steve Dickson <steved@xxxxxxxxxx> Date: Tue Jun 18 14:12:07 EDT 2009 Adds three trace points to the status routines in the sunrpc state machine. The trace points will only execute in error conditions. The goal of these trace points it to make it easier to debug process that spin or get hung up in the NFS and related subsystems. Signed-off-by: Steve Dickson <steved@xxxxxxxxxx> diff --git a/include/trace/events/sunrpc.h b/include/trace/events/sunrpc.h new file mode 100644 index 0000000..f502eb4 --- /dev/null +++ b/include/trace/events/sunrpc.h @@ -0,0 +1,86 @@ +#if !defined(_TRACE_SUNRPC_H) || defined(TRACE_HEADER_MULTI_READ) +#define _TRACE_SUNRPC_H + +#include <linux/tracepoint.h> + +#undef TRACE_SYSTEM +#define TRACE_SYSTEM sunrpc + +TRACE_EVENT(rpc_call_status, + + TP_PROTO(struct rpc_task *task), + + TP_ARGS(task), + + TP_STRUCT__entry( + __field(u32, cl_prog) + __field(u32, cl_vers) + __field(u32, p_proc) + __field(int, tk_status) + ), + + TP_fast_assign( + __entry->cl_prog = task->tk_client->cl_prog; + __entry->cl_vers = task->tk_client->cl_vers; + __entry->p_proc = task->tk_client->cl_procinfo->p_proc; + __entry->tk_status = task->tk_status; + ), + + TP_printk("%d[%d]:%d status=%d", __entry->cl_prog, + __entry->cl_vers,__entry->p_proc, __entry->tk_status) +); + +TRACE_EVENT(rpc_bind_status, + + TP_PROTO(struct rpc_task *task), + + TP_ARGS(task), + + TP_STRUCT__entry( + __field(u32, cl_prog) + __field(u32, cl_vers) + __field(u32, p_proc) + __field(int, tk_status) + ), + + TP_fast_assign( + __entry->cl_prog = task->tk_client->cl_prog; + __entry->cl_vers = task->tk_client->cl_vers; + __entry->p_proc = task->tk_client->cl_procinfo->p_proc; + __entry->tk_status = task->tk_status; + ), + + TP_printk("%d[%d]:%d status=%d", __entry->cl_prog, + __entry->cl_vers,__entry->p_proc, __entry->tk_status) +); +TRACE_EVENT(rpc_connect_status, + + TP_PROTO(struct rpc_task *task, int status), + + TP_ARGS(task, status), + + TP_STRUCT__entry( + __field(u32, cl_prog) + __field(u32, cl_vers) + __field(u32, p_proc) + __field(int, tk_status) + ), + + TP_fast_assign( + __entry->cl_prog = task->tk_client->cl_prog; + __entry->cl_vers = task->tk_client->cl_vers; + __entry->p_proc = task->tk_client->cl_procinfo->p_proc; + __entry->tk_status = status; + ), + + TP_printk("%d[%d]:%d status=%d", __entry->cl_prog, + __entry->cl_vers,__entry->p_proc, __entry->tk_status) + +); +#endif /* __TRACE_SUNRPC_H */ + +/* This part must be outside protection */ +#include <trace/define_trace.h> diff --git a/net/sunrpc/clnt.c b/net/sunrpc/clnt.c index 5abab09..b86ee2a 100644 --- a/net/sunrpc/clnt.c +++ b/net/sunrpc/clnt.c @@ -37,6 +37,9 @@ #include <linux/sunrpc/rpc_pipe_fs.h> #include <linux/sunrpc/metrics.h> +#include <linux/ftrace.h> +#define CREATE_TRACE_POINTS +#include <trace/events/sunrpc.h> #ifdef RPC_DEBUG # define RPCDBG_FACILITY RPCDBG_CALL @@ -957,6 +960,7 @@ call_bind_status(struct rpc_task *task) return; } + trace_rpc_bind_status(task); switch (task->tk_status) { case -ENOMEM: dprintk("RPC: %5u rpcbind out of memory\n", task->tk_pid); @@ -1038,6 +1042,7 @@ call_connect_status(struct rpc_task *task) return; } + trace_rpc_connect_status(task, status); switch (status) { /* if soft mounted, test if we've timed out */ case -ETIMEDOUT: @@ -1141,6 +1146,7 @@ call_status(struct rpc_task *task) return; } + trace_rpc_call_status(task); task->tk_status = 0; switch(status) { case -EHOSTDOWN: -- To unsubscribe from this list: send the line "unsubscribe linux-nfs" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
