RFC: proc.h p_status p_flags split

Matthew Dillon dillon at apollo.backplane.com
Tue Oct 11 20:18:40 PDT 2005


:
:This is a multi-part message in MIME format.
:--------------040206040000000201070107
:Content-Type: text/plain; charset=UTF-8; format=flowed
:Content-Transfer-Encoding: 7bit
:
:hey,
:
:I think we need to split the P_* flags and the S* stati between procs 
:and lwps.  This is how i see the P_* flags could be split.
:
:How should we do with the S* status?  Introduce PS* or LPS*?  Processes 
:can't sleep anymore, but they could be stopped.  Do we want a flag for 
:that or do we check p_nstopped == p_nthreads?
:
:cheers
:   simon

    Well, the job control code and signal() (SIGSTOP, SIGCONT, SIGTSTP, etc)
    is going to need process-wide stop/continue handling, but everything else
    associated with the kernel is going to want to deal with it on a
    per-thread basis.
    
						-Matt





More information about the Kernel mailing list