[ros-dev] [ros-diffs] [fireball] 25667: - Comment out clearing of KeLoaderBlock (introduced by 25629), because it looks like someone is still calling IopLoadServiceModule() even after that point. 2nd stage boots with this change.

[Alex Ionescu]

Alex Ionescu wrote:
> Aleksey Bragin wrote:
> 
>>"2nd stage boots" != "2nd stage completes and reboots without  
>>errors" :-).
>>
>>I have no idea about that other hang -- I was tracking the hang which  
>>happens during system initalization (on a few vmwares it was just  
>>hanging, and on one it gave me proper exception), when even GUI does  
>>not appear. I narrowed it down to the 25629 revision, and then found  
>>out about the KeLoaderBlock thing.
>>
>>Maybe with this fix I can boot other revisions and do regression- 
>>testing regarding the "other" hang.
>>
> 
> 
> 629, 630, 631 or 632 cause it.
> 

631 breaks it. Probably the code that sets the process cookie which now
works, or any other thread code. I'll look for the actual bug.

-- 
Best regards,
Alex Ionescu
Project Lead, TinyKRNL
Kernel-Mode Software Design Engineer, ReactOS