Re: [PATCH 1/8] perf: Allow to block process in syscall tracepoints

From: Peter Zijlstra
Date: Thu Dec 13 2018 - 05:05:36 EST

On Thu, Dec 13, 2018 at 11:01:49AM +0100, Peter Zijlstra wrote:
> One, very big maybe, would be to add a new tracepoint type that includes
> a might_sleep() and we very carefully undo all the preempt_disable and
> go sleep where we should. That also gives the tracepoint crud the
> information it needs to publish the capability to userspace.

Note that this includes handling fun things like the buffer getting
closed while we wait.