Clear framebuffer memory on VM reset/reboot

This is for discussing general topics about how to use VirtualBox.
Dsen
Oracle Corporation
Posts: 170
Joined: 10. Sep 2007, 10:42

Re: Clear framebuffer memory on VM reset/reboot

Post by Dsen »

socratis wrote: Of course... I have a video if you care to look (too big to post, >128KB), I have opened a bug report (#12193, I'm pretty sure it's the same thing) ... AND ... it's reproducible. It happens ONLY when rebooting, not when shutting down the VM and restarting.
As I see #12193 related to 3D enabled case. It's a very important aspect. Does it (I mean this one issue with artifacts) happens with 3D disabled?
billsc26
Posts: 34
Joined: 4. Nov 2013, 06:57

Re: Clear framebuffer memory on VM reset/reboot

Post by billsc26 »

Dsen wrote:
socratis wrote: Of course... I have a video if you care to look (too big to post, >128KB), I have opened a bug report (#12193, I'm pretty sure it's the same thing) ... AND ... it's reproducible. It happens ONLY when rebooting, not when shutting down the VM and restarting.
As I see #12193 related to 3D enabled case. It's a very important aspect. Does it (I mean this one issue with artifacts) happens with 3D disabled?
I never have 3D enabled and it happens to me with my XP guests all the time when I reboot/reset the VM.
socratis
Site Moderator
Posts: 27329
Joined: 22. Oct 2010, 11:03
Primary OS: Mac OS X other
VBox Version: VirtualBox+Oracle ExtPack
Guest OSses: Win(*>98), Linux*, OSX>10.5
Location: Greece

Re: Clear framebuffer memory on VM reset/reboot

Post by socratis »

Dsen wrote:As I see #12193 related to 3D enabled case. It's a very important aspect. Does it (I mean this one issue with artifacts) happens with 3D disabled?
  • 3D disabled : 3 out of 3 restarts were normal. NO corruption.
  • 3D enabled : 2 out of 3 restarts were corrupted.
billsc26 wrote:I never have 3D enabled and it happens to me with my XP guests all the time when I reboot/reset the VM.
@billsc26 : Do you see the corruption only during startup ("VBox BIOS") or all the way through the guest starting? In my case during the "BIOS" phase, I've never seen the framebuffer issue that you describe. Maybe we're dealing with two different (but similar) issues???

EDIT: While I was trying to reliably reproduce the issue, I got video corruption during the "BIOS" phase as well... Only with 3D enabled...
@Dsen : Any hints as to how we can troubleshoot this? The normal logs don't show anything out of the order (but OTOH, I'm not an expert at reading tea leaves...)
Do NOT send me Personal Messages (PMs) for troubleshooting, they are simply deleted.
Do NOT reply with the "QUOTE" button, please use the "POST REPLY", at the bottom of the form.
If you obfuscate any information requested, I will obfuscate my response. These are virtual UUIDs, not real ones.
Dsen
Oracle Corporation
Posts: 170
Joined: 10. Sep 2007, 10:42

Re: Clear framebuffer memory on VM reset/reboot

Post by Dsen »

billsc26 wrote:I never have 3D enabled and it happens to me with my XP guests all the time when I reboot/reset the VM.
Yep, as I said that is another issue. I was able to reproduce it and I think fix. It will be available in nearest maintenance release 4.3.4. Thank you.
Dsen
Oracle Corporation
Posts: 170
Joined: 10. Sep 2007, 10:42

Re: Clear framebuffer memory on VM reset/reboot

Post by Dsen »

socratis wrote:
Dsen wrote:As I see #12193 related to 3D enabled case. It's a very important aspect. Does it (I mean this one issue with artifacts) happens with 3D disabled?
  • 3D disabled : 3 out of 3 restarts were normal. NO corruption.
  • 3D enabled : 2 out of 3 restarts were corrupted.
EDIT: While I was trying to reliably reproduce the issue, I got video corruption during the "BIOS" phase as well... Only with 3D enabled...
@Dsen : Any hints as to how we can troubleshoot this? The normal logs don't show anything out of the order (but OTOH, I'm not an expert at reading tea leaves...)
As I expected you confirmed that most probably related to 3D functionality. I will *summon* our 3D specialist to that discussion.. :)
billsc26
Posts: 34
Joined: 4. Nov 2013, 06:57

Re: Clear framebuffer memory on VM reset/reboot

Post by billsc26 »

socratis wrote:
Dsen wrote:As I see #12193 related to 3D enabled case. It's a very important aspect. Does it (I mean this one issue with artifacts) happens with 3D disabled?
  • 3D disabled : 3 out of 3 restarts were normal. NO corruption.
  • 3D enabled : 2 out of 3 restarts were corrupted.
billsc26 wrote:I never have 3D enabled and it happens to me with my XP guests all the time when I reboot/reset the VM.
@billsc26 : Do you see the corruption only during startup ("VBox BIOS") or all the way through the guest starting? In my case during the "BIOS" phase, I've never seen the framebuffer issue that you describe. Maybe we're dealing with two different (but similar) issues???

EDIT: While I was trying to reliably reproduce the issue, I got video corruption during the "BIOS" phase as well... Only with 3D enabled...
@Dsen : Any hints as to how we can troubleshoot this? The normal logs don't show anything out of the order (but OTOH, I'm not an expert at reading tea leaves...)
Correct - I only see my issue during reset/reboot, not during BIOS. Sounds like you found it, that's great! Thanks for your help and attention.
socratis
Site Moderator
Posts: 27329
Joined: 22. Oct 2010, 11:03
Primary OS: Mac OS X other
VBox Version: VirtualBox+Oracle ExtPack
Guest OSses: Win(*>98), Linux*, OSX>10.5
Location: Greece

Re: Clear framebuffer memory on VM reset/reboot

Post by socratis »

Dsen wrote:As I expected you confirmed that most probably related to 3D functionality. I will *summon* our 3D specialist to that discussion.. :)
Can't wait... It's a pain in the keister to reboot these guys. I just hope that it is the same issue as #12193. Two birds with one stone!!!
Do NOT send me Personal Messages (PMs) for troubleshooting, they are simply deleted.
Do NOT reply with the "QUOTE" button, please use the "POST REPLY", at the bottom of the form.
If you obfuscate any information requested, I will obfuscate my response. These are virtual UUIDs, not real ones.
Post Reply