cvs commit: src/sys/kern kern_fork.c

[David Xu]

Alfred Perlstein wrote:

>>thread_suspend_check() with SINGLE_BOUNDARY is used is only called
>>in userret() where I don't think any code is still using the p_fd.
> 
> 
> This is not an acceptable solution as it will deadlock unless all
> threads exit the kernel.
> 

Where is deadlock ? thread_single(SINGLE_BOUNDARY) does not suspend
sleeping thread, it will interrupt it.

> filedescs can be refcounted per thread I think.
>