[Mono-dev] [Ximian-mono-list] Generics Sharing in the Debugger: Request for breaking the code freeze

Rodrigo Kumpera kumpera at gmail.com
Tue Jul 15 12:57:51 EDT 2008


How about disabling generic sharing for the 2.0 release and have a better
debugger experience for our users?

For me if gsharing breaks the debugger we can't say it's done. Enabling by
default something that breaks a
very important piece of our stack is not the way to go.

I would say that we go back to having generic sharing off by default and
make sure that the debugger refuses
to work if the user enabled it. This is, for me,  the saner way to go as it
would allow you guys to work on fixing it
without major disruption on the users and short deadlines.



On Tue, Jul 15, 2008 at 1:18 PM, Martin Baulig <martin at ximian.com> wrote:

> One word about attaching:
>
> The current situation is that attaching is officially disabled and not
> supported (as of July 4th, 2008).
>
> My understanding was that we'll enable generics sharing by default in
> Mono 2.0 and I didn't want to tell people that the debugger can do
> attaching if that feature doesn't work in the default configuration of
> Mono.
>
> >From a technical point of view, if you try to attach to a process:
>
> a) which is running without --debug
>
>   You won't be able to get managed backtraces, but nothing bad should
>   happen.
>
>   This "feature" is officially supported, but completely untested.  The
>   debugger may even crash at the moment, but that'd be considered a
>   bug and should be easy to fix.
>
> b) which is running with --debug and generics sharing
>
>   The debugger will crash.
>
> c) which is running without --debug and generics sharing
>
>   Should be exactly the same as in a).
>
> So basically, if we enable attaching in the debugger, we need to make
> sure that the target process is not running with generics sharing.
>
> So either, we keep attaching disabled, or we add a check to tell the
> debugger whether we're running with generics sharing or not.  Last week,
> I had no time to implement this check, so I turned it off.
>
> --
> Martin Baulig - martin at novell.com
> Novell GmbH, Düsseldorf
> GF: Volker Smid, Djamel Souici; HRB 21108 (AG Düsseldorf)
>
>
> _______________________________________________
> Ximian-mono-list mailing list
> Ximian-mono-list at lists.ximian.com
> http://lists.ximian.com/mailman/listinfo/ximian-mono-list
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.ximian.com/pipermail/mono-devel-list/attachments/20080715/dcbe0b85/attachment.html 


More information about the Mono-devel-list mailing list