Re: HW breakpoints perf_events request

[Peter Zijlstra <peterz@infradead.org>]

On Wed, 2010-01-13 at 17:45 -0800, Joshua Pincus wrote:
> However, there's nothing yet in place to allow a
> signal to be sent to the thread when a breakpoint
> has been hit.  Put another way, there's nothing here
> which affects execution flow of the user-app when the CPU
> traps due to a HW breakpoint.
> 
> Currently, the perf events infrastructure allows me
> to mmap() a block of memory and to poll() on it, waiting
> and watching for events to be described in that mmap'd
> buffer.  That will not be sufficient for what we need
> to do.  We need to have the ability to specify that
> execution of the thread should be interrupted, just as it
> would under ptrace(), and have a signal be delivered.
> Delivery of the signal must be received and processed
> by the application before the thread will be allowed to
> proceed to the nPC after the PC which caused the
> HW breakpoint event.
> 
> Is this possible?  Can we architect this feature into
> the perf_events infrastructure? 

We have fnctl() managing signal support, but that will only interrupt
the observing thread, not the threads being observed (unless they are
one and the same -- but using inheritance precludes that from being
true).

I'm not sure I'm willing to go in this direction with perf, the idea is
to have a minimum impact on the observed threads, explicitly sending
them signals and disturbing their execution goes against this.