Please ignore this question. I traced the error to a mistake in my own code. An invalid access to a memory address was causing errors in subsequent calls.
Thanks, Ravi On Jun 25, 12:16 pm, Chinnu <[email protected]> wrote: > Hi Egorov, > > I'm using V8 from multiple threads with Locker and Unlocker around all > the calls, including the IdleNotification(). > > Thanks, > Ravi > > On Jun 25, 11:01 am, Vyacheslav Egorov <[email protected]> wrote: > > > > > Hi Ravi, > > > Are you using V8 from a single thread or from multiple threads? > > > -- > > Vyacheslav Egorov > > > On Fri, Jun 25, 2010 at 3:37 PM, Chinnu <[email protected]> wrote: > > > Hi, > > > > I'm trying to emulate a sequence of javascript codes using V8. After > > > emulating a script, I delete the context and invoke > > > V8::IdleNotification() to free any memory if possible. Before running > > > the next script, I create a new context. > > > > After running through several thousands of scripts this way, I'm > > > getting an assert while creating a context for a script. At that > > > point, the gc_stat_ seems to be in 'SCAVENGE' state. I'm wondering if > > > the previous IdleNotification() call is leaving the gc_state_ in this > > > 'SCAVENGE' state instead of setting it back to 'NOT_IN_GC' . But, the > > > IdleNotification() call seems to be a blocking call. > > > > Please let me know if anyone has seen such issues before and if there > > > is something I'm missing. > > > > Thank you, > > > Ravi > > > > -- > > > v8-users mailing list > > > [email protected] > > >http://groups.google.com/group/v8-users -- v8-users mailing list [email protected] http://groups.google.com/group/v8-users
