[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

RE: [DotGNU]Safely implementing Thread.Abort

From: Thong (Tum) Nguyen
Subject: RE: [DotGNU]Safely implementing Thread.Abort
Date: Thu, 20 May 2004 18:09:40 +1200

If an unmanaged/managed flag is acceptable then I think we can handle aborts
with the CVM in the same way right now.  The SEGV exception handler I worked
on does something similar in that it can longjump back to the CVM and throw
and exception from there.  I wasn't sure whether the flag check on every
internal call was worth it but it looks like the only choice considering my
code rewriting idea has no chance of working :(.



> -----Original Message-----
> From: Rhys Weatherley [mailto:address@hidden
> Sent: Thursday, 20 May 2004 5:02 p.m.
> To: Russell Stuart; Thong (Tum) Nguyen; DotGnu-Develop
> Subject: Re: [DotGNU]Safely implementing Thread.Abort
> On Thursday 20 May 2004 02:58 pm, Russell Stuart wrote:
> > And that assumes your signal() handler checked you were
> > in the interpreter to start with, so it wouldn't leap out
> > of some critical piece of code such as the GC.
> There are ways of dealing with that in the JIT.  e.g. Mono sets a flag to
> indicate if the current thread is in managed or unmanaged code.  The flag
> is
> changed upon transitions between the two worlds.  If the flag is set to
> "unmanaged", then the signal handler won't throw immediately, but will set
> the "abort requested" flag.  Upon returning to managed code, the abort
> flag
> is checked.
> > I am not sure if a function return would be expensive
> > under the JIT, but that is the JIT authors problem.
> > By all reports he is a very capable man.
> That he is. :-)
> Cheers,
> Rhys.
> _______________________________________________
> Developers mailing list
> address@hidden

reply via email to

[Prev in Thread] Current Thread [Next in Thread]