RE: threads madness

["Howard Chu" <hyc@symas.com>]

That sounds good. I'll try it out shortly.

  -- Howard Chu
  Chief Architect, Symas Corp.       Director, Highland Sun
  http://www.symas.com               http://highlandsun.com/hyc
  Symas: Premier OpenSource Development and Support 

> -----Original Message-----
> From: Kurt D. Zeilenga [mailto:Kurt@OpenLDAP.org]
> Sent: Friday, October 11, 2002 3:42 PM
> To: Howard Chu
> Cc: openldap-devel@OpenLDAP.org
> Subject: Re: threads madness
> 
> 
> How about:
>  AC_CHECK_FUNCS(sched_yield pthread_yield thr_yield)
> 
> if all are no, then look for sched_yeild() in -lrt (for Tru64).
> Otherwise bitch.
> 
> Then, in thr_posix, use thr_yield over sched_yeild over pthread_yield.
> 
> Kurt
> 
> 
> At 03:25 PM 2002-10-11, Howard Chu wrote:
> >Solaris provides sched_yield in a separate library librt, it also has
> >thr_yield in libc.
> 
> >Currently it looks like configure only tests for thr_yield
> >if sched_yield doesn't exist. I'd like to turn this test 
> around so that
> >thr_yield is the preferred symbol. Dragging in librt causes 
> many undesirable
> >side-effects, particularly in the context of nss_ldap/pam_ldap being
> >dynamically loaded into a process' context.
> >
> >  -- Howard Chu
> >  Chief Architect, Symas Corp.       Director, Highland Sun
> >  http://www.symas.com               http://highlandsun.com/hyc
> >  Symas: Premier OpenSource Development and Support
> 
> 
>