From: Alexandre Oliva (oliva@dcc.unicamp.br)
Date: Thu Feb 04 1999 - 15:36:02 EST
On Feb 4, 1999, Godmar Back <gback@cs.utah.edu> wrote:
> Sometimes I feel I'm sending so much mail to kaffe-core that people
> don't have the time to read it anymore.
:-)
> What you suggest is what Alexandre suggested to, and it is what
> we're implementing for several years now. You just have to wrap
> to async-signal-unsafe with calls to jthread_spinon(0)/jthread_spinoff(0).
> (for now, at least)
But the issue is, (i) should we jthread_spinoff(0) whenever we enter
native mode and spinon when we leave, leaving it up to each particular
native method to allow context switching, or (ii) should each
potentially dangerous method be required to take care of it? Since we
cannot control what native code we're going to run, I fell we should
go for (i), and optimize existing methods later if we feel the need
for it.
And we can't forget softcalls that may run into dangerous execution
paths... :-(
Can calls of spinon and spinoff nest? I.e., can the garbage collector
spinoff just before it malloc()s and spinon upon return, so that, if
the caller called spinoff, the system remains so?
--
Alexandre Oliva http://www.dcc.unicamp.br/~oliva aoliva@{acm.org}
oliva@{dcc.unicamp.br,gnu.org,egcs.cygnus.com,samba.org}
Universidade Estadual de Campinas, SP, Brasil
This archive was generated by hypermail 2b29 : Sat Sep 23 2000 - 19:57:58 EDT