On 30.07.2010 17:31, Huihong Luo wrote:
if the debugger says it's NULL, it must be so, perhaps some overlooks? This is the first time I ever got this crash. What I did is very simple, create a new vm (XP Mode vm this case, shouldn't matter), boot it, take a snapshot, boot vm again, then immediately power if off, now try to restore the snapshot, and got that error.
Of course I believe that you hit a case where it was indeed NULL. However that doesn't give a hint what the root cause of the bug is. I'm not going to add a NULL check before the crashing line and blindly skip it without knowing WHAT is going wrong and WHY.
I am trying to recompile the latest svn code right now to double check this
Much appreciated, as the fact that this line caused a crash simply isn't providing enough information to do a proper fix. Tested restoring snapshots with a couple of VMs, and it worked.
Klaus
--- On *Fri, 7/30/10, Klaus Espenlaub /<[email protected]>/* wrote: From: Klaus Espenlaub <[email protected]> Subject: Re: [vbox-dev] crash when restoring a snapshot To: [email protected] Date: Friday, July 30, 2010, 7:16 AM On 30.07.2010 05:18, Huihong Luo wrote: > seems that in this function: mMediaData.backedUpData() might be NULL Do you have a reproduction scenario? Without one it's nearly impossible to find this bug. The call trace below provides no information which would give hints what could lead to this. We don't fix bugs by inserting a check based on the fact that the pointer was NULL. The cause needs to be investigated, as the pointer must not be NULL by design in this place. Klaus
_______________________________________________ vbox-dev mailing list [email protected] http://vbox.innotek.de/mailman/listinfo/vbox-dev
