Re: perf_fuzzer compiled for x32 causes reboot

[Vince Weaver]

On Sat, 1 Mar 2014, Andi Kleen wrote:

> Steven Rostedt <rostedt@xxxxxxxxxxx> writes:
> >
> > BTW, is the perf_fuzzer code posted somewhere? It sounds like it can be
> > really useful for us to do our own testing too.
> 
> I believe it's part of trinity.
> 
> http://codemonkey.org.uk/projects/trinity/
> 
> Perhaps it should have a "ftracer fuzzer" too?

It's not part of trinity, it's a separate project that re-uses some 
of trinity's codebase.

http://web.eece.maine.edu/~vweaver/projects/perf_events/fuzzer/

You can get the source as part of the perf_event_tests package
	https://github.com/deater/perf_event_tests

go into the fuzzer directory, run make, then "./perf_fuzzer".
You can also run the ./fast_repro.sh script instead, it's what I've
been using to track this kernel issue, although it tripped over the issue
a lot faster when compiled with -mx32 than it did when complied as native 
x86_64.

I think the code as found in the git tree should be working, I made a lot 
of changes when tracking this problem and I need to make sure I only check 
in the proper ones and not let any weird debugging patches slip in.

I'm aware of at least two other WARN messages and possibly one or two 
hard-lockup bugs in addition to the potential system reboot that can be 
triggered by the fuzzer, it just takes a while to isolate these things.

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/