Re: Defect in XSH pthread_attr_getinheritsched()

[Dave Butenhof <yyyyyyyyyyyyyy@xxxxxx>]

yyyyyyy@xxxxxxxxxx wrote:

> 	Defect report from : Ulrich Drepper , Red Hat
>
> (Please direct followup comments direct to yyyyyyyyyyyyyy@xxxxxxxxxxxxx)
>
> @ page 985 line 31346 section pthread_attr_getinheritsched() objection {ptattr-1}
>
> Problem:
>
> Defect code :  2. Omission
>
> The standard does not mention anywhere the default value of the value retrieved via pthread_attr_getinheritsched() after a call to
> pthread_attr_init().
>
> The pthread_attr_getguardsize() man page does contain the appropriate information.
>
> I would guess the default value is PTHREAD_INHERIT_SCHED but I'm not sure at all.
>
> Action:
>
> Add new paragraph after line 31245.  Add either
>
>  The default value of the /inheritsched/ attribute is <t>PTHREAD_INHERIT_SCHED<t>
>
>
> or add
>
>  The default value of the /inheritsched/ attribute is
> <t>PTHREAD_EXPLICIT_SCHED<t>
While we could decide to change that, this is NOT a defect. The working 
group INTENDED that the defaults for some attributes (including all the 
scheduling attributes) be IMPLEMENTATION DEFINED. This was to allow 
support for the expected range of implementations, from embedded 
realtime systems to multiuser timeshare systems; no one set of defaults 
could be appropriate for all.

This should probably be pointed out on each of the attributes functions, 
but isn't. You will find a general reference on page 52 under "Thread 
Scheduling"... "The default values of other scheduling attributes are 
implementation defined."

So the Action here should be to add explicit notice to each of the 
scheduling attribute functions that "The default value of xxx attribute 
is implementation defined."

--
/--------------------[ yyyyyyyyyyyyyy@xxxxxx ]--------------------\
| Hewlett-Packard Company       Tru64 UNIX & VMS Thread Architect |
|     My book: http://www.awl.com/cseng/titles/0-201-63392-2/     |
\----[ http://homepage.mac.com/dbutenhof/Threads/Threads.html ]---/