CLONE_DETACHED

CLONE_DETACHED

[Andries.Brouwer]

My current version of the clone.2 manual page says

       CLONE_THREAD
              (Linux 2.4 onwards) If  CLONE_THREAD  is  set,  the
              child  is  placed  in  the same thread group as the
              calling process.  Now in 2.6 also  the  meaningless
              CLONE_DETACHED  (introduced in 2.5.32) must be set.

I wonder whether it is too late to remove CLONE_DETACHED entirely.
There are precisely two occurrences: the definition and the check
that it is set when CLONE_THREAD is set. Seems fairly useless.

Andries

Re: CLONE_DETACHED

[Andrew Morton]

Andries.Brouwer@cwi.nl wrote:
>
> My current version of the clone.2 manual page says
> 
>        CLONE_THREAD
>               (Linux 2.4 onwards) If  CLONE_THREAD  is  set,  the
>               child  is  placed  in  the same thread group as the
>               calling process.  Now in 2.6 also  the  meaningless
>               CLONE_DETACHED  (introduced in 2.5.32) must be set.
> 
> I wonder whether it is too late to remove CLONE_DETACHED entirely.
> There are precisely two occurrences: the definition and the check
> that it is set when CLONE_THREAD is set. Seems fairly useless.
> 

Changelog says:

ChangeSet 1.1150 2003/08/14 11:11:39 torvalds@home.osdl.org
  Mark CLONE_DETACHED as being irrelevant: it must match CLONE_THREAD.
  
  CLONE_THREAD without CLONE_DETACHED will now return -EINVAL, and
  for a while we will warn about anything that uses it (there are no
  known users, but this will help pinpoint any problems if somebody
  used to care about the invalid combination).


It has been five months; I'd say we could lose it now.  Linus?