Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > comp.programming.threads > #144

Re: pthread question

Show key headers only | View raw

Alexander Terekhov wrote:

> Geoff Clare wrote:
>> 
>> Alexander Terekhov wrote:
>> 
>> >
>> > Geoff Clare wrote:
>> > [...]
>> >> 9. When a running thread issues the sched_yield() function, the
>> >>    thread becomes the tail of the thread list for its priority.
>> >
>> > That makes sense for a uniprocessor.
>> >
>> > But nowadays even pocket gadgets are multiprocessors.
>> >
>> >> POSIX doesn't standardise internals, but it does standardise how
>> >> threads with the above mentioned scheduling policies are scheduled.
>> >
>> > http://pubs.opengroup.org/onlinepubs/009695399/functions/xsh_chap02_09.html
>> >
>> > "For application threads with scheduling allocation domains of size
>> > greater than one, the rules defined for SCHED_FIFO, SCHED_RR, [TSP] and
>> > SCHED_SPORADIC shall be used in an implementation-defined manner."
>> >
>> > Translation: no rules whatsoever.
>> 
>> The way the domain-size-one rules are applied to larger domains is
>> "implementation-defined", which means implementations are required
>> to document how they are applied.  Not quite the same as "no rules
>> whatsoever".
>
> In praxis "implementation-defined" means undocumented to begin with.

Huh?  XBD7 section 2.1.2:

    A conformance document with the following information shall be
    available for an implementation claiming conformance to POSIX.1-2008.
    [...]
    The conformance document shall describe the behavior of the
    implementation for all implementation-defined features defined in
    POSIX.1-2008. This requirement shall be met by listing these
    features and providing either a specific reference to the system
    documentation or providing full syntax and semantics of these
    features.

> Anyway since "implementation-defined" can change from version to version
> by even the same vendor it does mean "no rules whatsoever" in disguise.

No, it means each implementor chooses their own rules, but they must
tell you what those rules are.  It is then up to application writers
to decide whether to rely on those rules.  For applications targeted
at a small set of specific implementations, there might be some
advantage in doing so.

> http://pubs.opengroup.org/onlinepubs/9699919799/basedefs/V1_chap01.html#tag_01_05_02
>
> "implementation-defined
>
> Describes a value or behavior that is not defined by POSIX.1-2008 but is
> selected by an implementor. The value or behavior may vary among
> implementations that conform to POSIX.1-2008. An application should not
> rely on the existence of the value or behavior. An application that
> relies on such a value or behavior cannot be assured to be portable
> across conforming implementations."

This uses "should", which means it is a recommendation.  The point
is that for maximum portability, application writers should not rely
on any implementation-defined behaviour.  The application writer has
to decide whether portability is more important than some advantage
that might be gained by sacrificing a little portability.

-- 
Geoff Clare <netnews@gclare.org.uk>

Back to comp.programming.threads | Previous | Next — Previous in thread | Next in thread | Find similar | Unroll thread

Thread

pthread question p6 c6d6 <p6c6d6@gmail.com> - 2011-05-08 21:28 -0700
  Re: pthread question Vaclav Haisman <v.haisman@sh.cvut.cz> - 2011-05-09 07:11 +0200
  Re: pthread question Joshua Maurice <joshuamaurice@gmail.com> - 2011-05-09 11:49 -0700
    Re: pthread question David Schwartz <davids@webmaster.com> - 2011-05-09 12:24 -0700
      Re: pthread question Joshua Maurice <joshuamaurice@gmail.com> - 2011-05-09 14:22 -0700
        Re: pthread question "Chris M. Thomasson" <cristom@charter.net> - 2011-05-09 15:33 -0700
        Re: pthread question David Schwartz <davids@webmaster.com> - 2011-05-10 01:22 -0700
    Re: pthread question Geoff Clare <geoff@clare.See-My-Signature.invalid> - 2011-05-10 13:47 +0100
      Re: pthread question Alexander Terekhov <terekhov@web.de> - 2011-05-10 17:53 +0200
        Re: pthread question Geoff Clare <geoff@clare.See-My-Signature.invalid> - 2011-05-11 13:46 +0100
          Re: pthread question Alexander Terekhov <terekhov@web.de> - 2011-05-11 16:19 +0200
            Re: pthread question Geoff Clare <geoff@clare.See-My-Signature.invalid> - 2011-05-12 13:26 +0100
              Re: pthread question Alexander Terekhov <terekhov@web.de> - 2011-05-12 15:28 +0200
                Re: pthread question Geoff Clare <geoff@clare.See-My-Signature.invalid> - 2011-05-13 14:00 +0100
                Re: pthread question Alexander Terekhov <terekhov@web.de> - 2011-05-13 22:59 +0200
                Re: pthread question Geoff Clare <geoff@clare.See-My-Signature.invalid> - 2011-05-16 13:28 +0100

csiph-web