[Stackless] crash in stackless 2.5.2

Richard Tew richard.m.tew at gmail.com
Fri Feb 20 16:01:52 CET 2009


On Fri, Feb 20, 2009 at 12:07 AM, thejas c r <thejascr at gmail.com> wrote:
> Can anybody let me know if this is already fixed in later versions.

This bug is new.

> If not what additional information would be needed to fix this?

There are two possible causes off the top of my head.  The first is
your compiler.  The possibility exists that it does not handle the
inline assembly in the MIPS support well, and generates flawed code.
The second is the MIPS support.  It may be that this is flawed and
needs correction.  The way I would proceed to look into and fix this,
would be to find a MIPS machine and reproduce it repeatedly, debugging
it.

In the meantime, you might consider using different versions of GCC.

Note that the MIPS support for Stackless was taken as is from the PyPy
greenlets source code, greenlets being a stripped down version of the
code at the heart of Stackless.  If the support just is not working
and fixing this is not likely, then we might want to consider removing
this support.

If I get the time, I'll take a look at the possible code paths which
might lead to your stack trace.  Though it is unlikely I can guess
what went wrong from a code reading.

> Also is there any separate bug repo for stackless? I couldnt find any link
> in the website.

There is no bug reporting system for Stackless.

Cheers,
Richard.




More information about the Stackless mailing list