[squeak-dev] Re: Is this a bug with Step "Over"?

Eliot Miranda eliot.miranda at gmail.com
Tue Jun 23 16:36:12 UTC 2015


Hi Marcel,

     AFAICT the idea if the simulation guards is to raise errors when doing something that "shouldn't happen during debugging".  But what should or shouldn't happen during debugging?  It's a value judgement.  Fir example, forking a new process.  IMO there no reason why a new process shouldn't be forked during debugging.  Indeed, before yield was a primitive, not bring able to fork would have prevented debugging code that contained a yield.

So unless doing something under debugging would really break the system I don't see that there should be simulation guards.  But that's my opinion.

Eliot (phone)

On Jun 21, 2015, at 12:56 AM, "marcel.taeumel" <Marcel.Taeumel at hpi.de> wrote:

> See Debugger class >> #openContext:label:contents:, which is in the control
> flow of Warning >> defaultAction. There, you can read "Processor
> activeProcess suspend", which seems to be ignored. Now go to pragma of that
> method. It says "<primitive: 19>", which is a simulation guard and hence
> important for a debugger's step-over.
> 
> Can someone explain, how those simulation guards work or are supposed to
> work? There actually were some changes in the simulation code... but
> where...
> 
> Best,
> Marcel
> 
> 
> 
> --
> View this message in context: http://forum.world.st/Is-this-a-bug-with-Step-Over-tp4830736p4833361.html
> Sent from the Squeak - Dev mailing list archive at Nabble.com.
> 


More information about the Squeak-dev mailing list