The following steps should let you reproduce it. It segfaults for me every time. Though interestingly, the actual place it segfaults changes from run to run.

First checkout Krakatau a43b2e7e0a53bca9fe7c34d97b3b3738d662f8d5 (the latest)

Then run
pypy Krakatau/decompile.py -out <tempdir> -nauto -skip <path to rt.jar>

where <tempdir> is any writeable directory. For example
pypy Krakatau/decompile.py -out temp -nauto -skip /usr/lib/jvm/java-8-oracle/jre/lib/rt.jar


Decompiling almost anything segfaults, but this is the simplest setup. Feel free to respond if you still can't reproduce it.




2016-02-09 19:19 GMT-08:00 Robert Grosse <n210241048576@gmail.com>:
The segfaults only started last week. I suspect that one of my recent changes triggers the segfault behavior in Pypy. Can you try with the latest version of Krakatau? If you still can't reproduce it, I can try to figure out the steps in more detail.



2016-02-09 9:33 GMT-08:00 Maciej Fijalkowski <fijall@gmail.com>:
ok, I used krakatau in the past and it worked, so you need to be more
specific. Notably give me an example program, how to run it etc. I
need to be able to confirm your steps step by step

On Tue, Feb 9, 2016 at 4:54 PM, Robert Grosse <n210241048576@gmail.com> wrote:
> Pypy segfaults every time I try to decompile an app with Krakatau. The same
> code still works on CPython (it's just a lot slower obviously). I'll try to
> narrow down the circumstances and come up with better repro instructions
> later.
>
> This happened on my build from November too, so it's not a recent
> regression. And it happened on multiple computers, although I built Pypy
> from source in both cases, so there could be something going wrong there.
>
> 2016-02-08 0:51 GMT-08:00 Maciej Fijalkowski <fijall@gmail.com>:
>>
>> Hi Robert
>>
>> You need to explain in details what are you doing and how can we reproduce
>> it
>>
>> On Sun, Feb 7, 2016 at 10:26 PM, Robert Grosse <n210241048576@gmail.com>
>> wrote:
>> > I updated Pypy, but I'm still getting random segfaults. Is there any way
>> > to
>> > see what the problem might be? It just says segfault, so there's no
>> > information. Also, the same code works in CPython.
>> >
>> > 2016-02-07 9:05 GMT-08:00 Matti Picus <matti.picus@gmail.com>:
>> >>
>> >>
>> >> On 07/02/16 06:55, Robert Grosse wrote:
>> >>>
>> >>> I am trying to update Pypy. I want to build Pypy from source using the
>> >>> instructions at http://doc.pypy.org/en/latest/build.html in order to
>> >>> get the
>> >>> latest bugfixes and improvements.
>> >>>
>> >>> However, all of the last 5 nightly builds shown at
>> >>> http://buildbot.pypy.org/summary?branch=%3Ctrunk%3E failed their
>> >>> tests. So I
>> >>> can't tell what a good revision to update to is.
>> >>>
>> >>> What is the last good revision of Pypy?
>> >>>
>> >> The builds have been failing "only" for vmprof tests, everything else
>> >> should work.
>> >> Matti
>> >
>> >
>> >
>> > _______________________________________________
>> > pypy-dev mailing list
>> > pypy-dev@python.org
>> > https://mail.python.org/mailman/listinfo/pypy-dev
>> >
>
>