Skip to contentSkip to conversation

tracing: remove dump_ran check in __ftrace_dump

PATCH In review zhangwei(Jovi) 3 participants 4 messages Mar 11, 2013 Tracing

lore
Share & tools
lore
#0 lore raw

PATCH 6/13 tracing: remove dump_ran check in __ftrace_dump

It's reasonable to call __ftrace_dump function not only once, so remove the dump_ran variable checking.

---
 kernel/trace/trace.c |    6 ------
 1 file changed, 6 deletions(-)
kernel/trace/trace.c +0 -6
index 090eddb..4cec7b8 100644
@@ -5106,17 +5106,12 @@ __ftrace_dump(bool disable_tracing, enum ftrace_dump_mode oops_dump_mode)
 	/* use static because iter can be a bit big for the stack */
 	static struct trace_iterator iter;
 	unsigned int old_userobj;
-	static int dump_ran;
 	unsigned long flags;
 	int cnt = 0, cpu;
 
 	/* only one dump */
 	local_irq_save(flags);
 	arch_spin_lock(&ftrace_dump_lock);
-	if (dump_ran)
-		goto out;
-
-	dump_ran = 1;
 
 	tracing_off();
 
@@ -5206,7 +5201,6 @@ __ftrace_dump(bool disable_tracing, enum ftrace_dump_mode oops_dump_mode)
 		tracing_on();
 	}
 
- out:
 	arch_spin_unlock(&ftrace_dump_lock);
 	local_irq_restore(flags);
 }
Signature · 1 line
1.7.9.7
maintainer
#1 lore raw
Quoted 2 lines · zhangwei(Jovi)

It's reasonable to call __ftrace_dump function not only once, so remove the dump_ran variable checking.

This needs a little more work. On an oops, I only want it dumped once, because a crash can cause another crash while its dumping, and without that check in will corrupt the buffer.

Now, we have things like ctrl^z that also does a dump where we don't want to disable it. Cleaning this up has been on my todo list for a while. I may go ahead and clean that up myself.

-- Steve

Quoted 36 lines
---
 kernel/trace/trace.c |    6 ------
 1 file changed, 6 deletions(-)
kernel/trace/trace.c +0 -6
index 090eddb..4cec7b8 100644
@@ -5106,17 +5106,12 @@ __ftrace_dump(bool disable_tracing, enum ftrace_dump_mode oops_dump_mode)
 	/* use static because iter can be a bit big for the stack */
 	static struct trace_iterator iter;
 	unsigned int old_userobj;
-	static int dump_ran;
 	unsigned long flags;
 	int cnt = 0, cpu;
 
 	/* only one dump */
 	local_irq_save(flags);
 	arch_spin_lock(&ftrace_dump_lock);
-	if (dump_ran)
-		goto out;
-
-	dump_ran = 1;
 
 	tracing_off();
 
@@ -5206,7 +5201,6 @@ __ftrace_dump(bool disable_tracing, enum ftrace_dump_mode oops_dump_mode)
 		tracing_on();
 	}
 
- out:
 	arch_spin_unlock(&ftrace_dump_lock);
 	local_irq_restore(flags);
 }
#2 lore raw
Quoted 7 lines · Steven Rostedt
Quoted 2 lines · zhangwei(Jovi)

It's reasonable to call __ftrace_dump function not only once, so remove the dump_ran variable checking.

This needs a little more work. On an oops, I only want it dumped once, because a crash can cause another crash while its dumping, and without that check in will corrupt the buffer.

For reclusive dumping, it's already under the protection of ftrace_dump_lock spinlock,
I missed something? would you explain more on this case?

Quoted 4 lines

Now, we have things like ctrl^z that also does a dump where we don't want to disable it. Cleaning this up has been on my todo list for a while. I may go ahead and clean that up myself.

Nice, please ignore this patch if that code will be cleanup. Thanks.

Quoted 46 lines

-- Steve

Quoted 29 lines
---
 kernel/trace/trace.c |    6 ------
 1 file changed, 6 deletions(-)
kernel/trace/trace.c +0 -5
index 090eddb..4cec7b8 100644
@@ -5106,17 +5106,12 @@ __ftrace_dump(bool disable_tracing, enum ftrace_dump_mode oops_dump_mode)
      /* use static because iter can be a bit big for the stack */
      static struct trace_iterator iter;
      unsigned int old_userobj;
-     static int dump_ran;
      unsigned long flags;
      int cnt = 0, cpu;
 
      /* only one dump */
      local_irq_save(flags);
      arch_spin_lock(&ftrace_dump_lock);
-     if (dump_ran)
-             goto out;
-
-     dump_ran = 1;
 
      tracing_off();
 

kernel/trace/trace.c __ftrace_dump(bool disable_tracing, enum ftrace_dump_mode oops_dump_mode) · line 5206

Quoted 7 lines
kernel/trace/trace.c +0 -1
@@ -5206,7 +5201,6 @@ __ftrace_dump(bool disable_tracing, enum ftrace_dump_mode oops_dump_mode)
              tracing_on();
      }
 
- out:
      arch_spin_unlock(&ftrace_dump_lock);
      local_irq_restore(flags);
 }

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/

maintainer
#3 lore raw
Quoted 11 lines · Jovi Zhang
Quoted 7 lines · Steven Rostedt
Quoted 2 lines · zhangwei(Jovi)

It's reasonable to call __ftrace_dump function not only once, so remove the dump_ran variable checking.

This needs a little more work. On an oops, I only want it dumped once, because a crash can cause another crash while its dumping, and without that check in will corrupt the buffer.

For reclusive dumping, it's already under the protection of ftrace_dump_lock spinlock,
I missed something? would you explain more on this case?

Actually, it matters if it was called from NMI context or not. If an NMI triggered and did a dump while a reader was reading the buffer, the NMI can corrupt the buffer. It will print fine for the NMI, but if the reader continues, there's a chance it can get messed up and corrupt the buffer. The "dump once" is a paranoid method to say we only dump it once on oops and stop (hence the ftrace_kill() there too).

Now I think there's a possible deadlock here as well. If the dump was caused by something other than an NMI lock up, and while it is dumping the NMI goes off and triggers a bug, it too can enter this and that arch_spin_lock() will cause a deadlock. This will be something I need to clean up as well.

Thanks for calling my attention to this.

-- Steve