Posts: 7,172
Joined: Dec 2003
Reputation:
4
kraqh3d
Retired Developer
Posts: 7,172
Your crash is different. It's not the enumeration of the display devices. But it does look like you may have a bad copy of the addons folder. I see the errors about the incorrect type fields for some of them, and lots of errors about missing skin files.
Posts: 1,132
Joined: Apr 2010
Reputation:
71
CrystalP
Team-Kodi Developer
Posts: 1,132
2010-05-14, 23:36
(This post was last modified: 2010-05-14, 23:39 by CrystalP.)
taxigps: it seemed you had issues with your computer earlier - please make sure that's taken care of first. The debug log you attached has proper screen enumeration, so I don't see how 29462 could break it.
Please follow the HOW-TO build to the letter, I recently updated it.
kimp93: same thing, first make sure your build environment is exactly as described in the wiki. You could be missing the XBMC_HOME variable.
kraqh3d: the addon type field error is nothing. It's something obsolete that doesn't do anything in Windows DX anyway.
Posts: 3,909
Joined: Dec 2004
Reputation:
20
Nuka1195
Skilled Python Coder
Posts: 3,909
30159 works fine for me with intel gfx. thanks
Posts: 1,132
Joined: Apr 2010
Reputation:
71
CrystalP
Team-Kodi Developer
Posts: 1,132
debug logs (working and non-working)!
Posts: 1,132
Joined: Apr 2010
Reputation:
71
CrystalP
Team-Kodi Developer
Posts: 1,132
nickritus, there is less than I expected. Do you unplug your screen when you remote in?
Can you trace what's happening in the screen enumeration (CWinSystemWin32::UpdateResolutionsInternal() inWinSystemWin32.cpp) or can you apply a patch that will add more to the debug log?
eduardor2k, not the best place but... Check the HOW-TO build and make sure you meet all requirements. Especially the DirectX SDK.
If that's ok, try a 'Solution Clean' before 'Solution Build'.
Posts: 30
Joined: Mar 2010
Reputation:
0
2010-05-21, 07:50
(This post was last modified: 2010-05-21, 08:59 by nickritus.)
Cheers for helping out CrystalP,
I don't physically unplug the TV, however I'm pretty sure when I turn the TV off it registers this as removing the device. I don't have proof of this on hande, only anectodal evidence (I hear the windows remove device sound...).
I don't have dev tools on that machine (and I really don't want them on there). I can see if I can stand a bit of remote debugging, but it might be quicker if I apply a patch.
Edit: OK so remote debugging has been somewhat fruitful. Here's what I've got for you so far:
Inside UpdateResolutionsInternal(),
The outer while loop call to enumDisplayDevices returns:
RDPDD Chained DD
RDP Encoder Mirror Driver
RDP Reflector Display Driver
Of these only the remote desktop chain driver passes the "Exclude displays that are not part of the windows desktop" check.
That device subsequently returns 0 on the inner call to EnumDisplayDevices. According to function documentation: "If the function fails, the return value is zero. The function fails if iDevNum is greater than the largest device index." iDevNum in that call is 0 so I guess there aren't any to enum. So anyway no screens are found.
I've hit a bit of an obstacle though, it's having trouble loading cpluff, which didn't happen in the release exe. I'm hitting an abort() there so I can't see where the real crash happens yet.
Hope this is helpful.
Posts: 30
Joined: Mar 2010
Reputation:
0
Thanks CrystalP, nice one!
Looking forward to trying it out.