Re: cpuset_t vs. cpu_set_t (again)

[Anthony Mallet <anthony.mallet%useless-ficus.net@localhost>]

J.T. Conklin writes:
| I believe you mentioned this was in the ACE/TAO configure script. I do
| much/most of the maintenence on it, so if NetBSD keeps using cpuset_t,
| I'll probably end up improving the ACE feature test to handle it.

Yes, it's in ACE-5.6. And more specifically in ace/OS_NS_Thread.cpp.
The problem actually is that the ACE_HAS_CPU_SET_T is not tested (only
ACE_HAS_PTHREAD_SETAFFINITY_NP is tested).
The configure script is itself correct.

| FWIW, this may be a place where a typedef that maps cpuset_t to
| cpu_set_t for "compatibility" is appropriate, like u_intXX/uintXX,
| u_short/ushort, etc.

Ok, that's probably a good idea. But I really feel that a better solution
would be to change NetBSD typedef definition. Since it's a very new
feature and POSIX doesn't say anything here, I don't see the point in
making it incompatible since the beginning. There are probably other
applications where the problem will arise.

Thank you for your answer, let's see what happen to sched.h :)

-- 
Anthony