[PATCH] Fix conversion of task state to char in latency tracer

[PATCH] Fix conversion of task state to char in latency tracer

[Ankita Garg]

Hi,

The conversion of task states to a character in the sched_switch tracer (part
of latency tracer infrastructure), seems to be incorrect. We currently do it
by indexing into the state_to_char array using the state value. The state
values do not map directly into the array index and are thus incorrect. The
following patch addresses this issue. This is also what is being done even
in the show_task() routine in kernel/sched.c

The patch has been compile and run tested.

Signed-off-by: Ankita Garg <ankita@in.ibm.com> 

Index: linux-2.6.24.3/kernel/trace/trace.c
===================================================================
--- linux-2.6.24.3.orig/kernel/trace/trace.c	2008-04-09 15:30:54.000000000 +0530
+++ linux-2.6.24.3/kernel/trace/trace.c	2008-04-09 23:37:03.000000000 +0530
@@ -1086,6 +1086,7 @@
 	char *comm;
 	unsigned long nr;
 	int S;
+	unsigned state;
 
 	if (!next_entry)
 		next_entry = entry;
@@ -1114,8 +1115,8 @@
 		seq_puts(m, ")\n");
 		break;
 	case TRACE_CTX:
-		S = entry->ctx.prev_state < sizeof(state_to_char) ?
-			state_to_char[entry->ctx.prev_state] : 'X';
+		state = entry->ctx.prev_state ? __ffs(entry->ctx.prev_state) + 1 : 0;
+		S = state < sizeof(state_to_char) - 1 ? state_to_char[state] : 'X';
 		comm = trace_find_cmdline(entry->ctx.next_pid);
 		seq_printf(m, " %d:%d:%c --> %d:%d %s\n",
 			   entry->ctx.prev_pid,


-- 
Regards,
Ankita Garg (ankita@in.ibm.com)
Linux Technology Center
IBM India Systems & Technology Labs, 
Bangalore, India   

Re: [PATCH] Fix conversion of task state to char in latency tracer

[Ingo Molnar]

* Ankita Garg <ankita@in.ibm.com> wrote:

> Hi,
> 
> The conversion of task states to a character in the sched_switch 
> tracer (part of latency tracer infrastructure), seems to be incorrect. 
> We currently do it by indexing into the state_to_char array using the 
> state value. The state values do not map directly into the array index 
> and are thus incorrect. The following patch addresses this issue. This 
> is also what is being done even in the show_task() routine in 
> kernel/sched.c
> 
> The patch has been compile and run tested.
> 
> Signed-off-by: Ankita Garg <ankita@in.ibm.com>

thanks Ankita, good catch. I applied your patch to sched-devel/latest as 
well (which carries the ftrace patches, and which patches -rt is 
tracking). You can track this upstream tree via:

  http://people.redhat.com/mingo/sched-devel.git/README

	Ingo