Messages in this thread | | | Date | Thu, 16 Apr 2009 13:45:03 -0400 (EDT) | From | Steven Rostedt <> | Subject | Re: [PATCH 2/2] tracing/events/lockdep: move tracepoints within recursive protection |
| |
On Thu, 16 Apr 2009, Peter Zijlstra wrote:
> On Thu, 2009-04-16 at 12:15 -0400, Steven Rostedt wrote: > > plain text document attachment > > (0002-tracing-events-lockdep-move-tracepoints-within-recu.patch) > > From: Steven Rostedt <srostedt@redhat.com> > > > > With the current location of the tracepoints in lockdep, the system > > can hard lockup in minutes when the tracepoints are enabled. > > > > Moving the tracepoints outside inside the lockdep protection solves > > the issue. > > NAK
Can we at least add this as a workaround. Basically, the lockdep tracepoints are broken as is. I do not plan on changing the logic of the events to prevent nesting. That's a feature I use. If we don't allow nesting, we must drop events, which is bad.
The other answer is simply to remove the trace points in lockdep, until they work again.
-- Steve
> > the idea is to eventually move lockdep on top of the tracepoints. The > tracer should grow to be more robust and handle recursion itself. > > Its likely a case of the tracer using a spinlock or mutex in the > tracepoint code. When I did the tracepoints I converted one such to a > raw_spinlock_t in the trace_print code.
| |