Re: OOPS in perf_mmap_close()

[Vince Weaver]

On Thu, 30 May 2013, Peter Zijlstra wrote:
> 
> Awesome! How specific is it to perf? I mean, would that tool work
> equally well for other tinity report?

No, it's fairly specific to my perf_fuzzer.  I only use trinity code to 
set up the perf_event_open() attr parameters, everything else I use my own 
code for.

The logging output from perf_fuzzer is fairly straightforward.  Just a 
single char indicating which syscall and a space separated list of 
the result (if relevant) and then all the syscall parameters.

The logging output can then be used by both the replayer and the 
sample-code generator.

While in theory this could be extended to something generic, currently it 
only works for perf_event_open/close/ioctl/read/write/mmap/munmap, and
even then it takes some shortcuts because I know it will only be run when 
finding perf_event problems.

The code for all of this is here if anyone else wants to play with it.
https://github.com/deater/perf_event_tests/tree/master/fuzzer

> I'll go prod, thanks again!

The bug looks related to hw breakpoints, not sure if the bugs Oleg 
Nesterov has been reporting in this area might be relevant or not.

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