Draft pthread_spin_init(2) manual page
Michael Kerrisk (man-pages)
mtk.manpages@gmail.com
Thu Oct 19 20:08:00 GMT 2017
Hello Zack,
Hello On 10/18/2017 07:44 PM, Zack Weinberg wrote:
> On Wed, Oct 18, 2017 at 4:10 AM, Michael Kerrisk (man-pages)
> <mtk.manpages@gmail.com> wrote:
>> Hello all,
>>
>> I recently drafted a manual page for the pthread_spin_init() and
>> pthread_spin_destroy() APIs. Comments and suggestions for improvements
>> would be most welcome.
>
> This should probably cross-reference the pthread_mutex_* manpages and
> say when it is appropriate to use which. I see you already have some
> of that...
>
>> NOTES
>> Spin locks should be employed in conjunction with real-time schedâ
>> uling policies (SCHED_FIFO, or possibly SCHED_RR). Use of spin
>> locks with nondeterministic scheduling policies such as
>> SCHED_OTHER probably indicates a design mistake. The problem is
>> that if a thread operating under such a policy is scheduled off
>> the CPU while it holds a spin lock, then other threads will waste
>> time spinning on the lock until the lock holder is once more
>> rescheduled and releases the lock.
>>
>> Warning: if threads create a deadlock situation while employing
>> spin locks, those threads will spin forever consuming CPU time.
>
> but you might want to begin with something like "Most programs should
> use mutexes (pthread_mutex_init(3)) instead of spin locks. Spin locks
> are primarily useful in conjunction with real-time scheduling ..."
>
> Also, this is important enough that I'd put it in the DESCRIPTION.
Thanks. I think that's a great phrasing. I added the following at the start
of DESCRIPTION:
General note: Most programs should use mutexes instead of spin
locks. Spin locks are primarily useful in conjunction with real-
time scheduling policies. See NOTES.
Cheers,
Michael
--
Michael Kerrisk
Linux man-pages maintainer; http://www.kernel.org/doc/man-pages/
Linux/UNIX System Programming Training: http://man7.org/training/
More information about the Libc-alpha
mailing list