[Stackless] Stackless Linux/x86 and OS X test status
tismer at stackless.com
Mon Jan 26 22:07:54 CET 2004
Bob Ippolito wrote:
> Compiler warnings have been minimized, and I got rid of some code that
> doesn't belong anymore. I haven't tested on Win32 (do you have
> instructions for compiling with VS.net?), but things are looking pretty
> good on OS X and Linux/x86.
> All of the tests seem to pass, except for bajo_crash2.py (on both
> platforms, in the same way):
I know that crash.
It happens when a cstack object gets allocated. For some
reason, it is in an invalid GC chain.
I had that code running, already, but I wasn#t sure if
it was correct for every situation.
The old code used a nasty trick (but maybe healthy):
When a tasklet needs to be destructed and still has
runnable frames, then the tasklet will be duplicated.
The existing one will completely defunct, all properties
are moved into the new one.
This worked very well, even if a tasklet's __del__ method
called a stackless.schedule().
The bad thinbg about the latter was, that exactly that schedule()
in a __del__ caused to prevend the deletion from finishing.
Instead, you got one more runnable tasklet for every run of this
So I changed code, no longer duplicating and possible deferring
deletion, but I copied the object resurrection code and added
code that enforces deletion of the tasklet, "right here".
I am not sure, that that's my fear, if that opens a case where
the Python machinery is not re-entrant. The code might switch
to a different stack while some deletion cascade has not
Will try further to track this down.
ciao - chris
Christian Tismer :^) <mailto:tismer at stackless.com>
Mission Impossible 5oftware : Have a break! Take a ride on Python's
Johannes-Niemeyer-Weg 9a : *Starship* http://starship.python.net/
14109 Berlin : PGP key -> http://wwwkeys.pgp.net/
work +49 30 89 09 53 34 home +49 30 802 86 56 mobile +49 173 24 18 776
PGP 0x57F3BF04 9064 F4E1 D754 C2FF 1619 305B C09C 5A3B 57F3 BF04
whom do you want to sponsor today? http://www.stackless.com/
Stackless mailing list
Stackless at stackless.com
More information about the Stackless