|
Home > Archive > Unix Programming > August 2005 > semtimedop portability?
You are viewing an archived Text-only version of the thread.
To view this thread in it's original format and/or if you want to reply to
this thread please [click here]
| Author |
semtimedop portability?
|
|
| Michael B Allen 2005-07-29, 6:01 pm |
| How portable is semtimedop(2)? I know Linux and HPUX has it. Is there a
standard way to wait on a Sys V semaphore for no longer than a specified
timeout?
Thanks,
Mike
| |
| Michael Kerrisk 2005-08-01, 2:55 am |
| On Fri, 29 Jul 2005 15:33:11 -0400, Michael B Allen
<mba2000@ioplex.com> wrote:
>How portable is semtimedop(2)? I know Linux and HPUX has it. Is there a
>standard way to wait on a Sys V semaphore for no longer than a specified
>timeout?
Hi Mike,
It isn't widely available (not in POSIX.1-2001). Solaris 9 has it.
(I didn't know that HP-UX had it... are you sure? Which version of
HP-UX?)
The portable way of doing thingsis to set a timer before semop() using
alarm() or setitimer().
Cheers,
Michael
| |
| Michael B Allen 2005-08-03, 6:00 pm |
| On Mon, 01 Aug 2005 10:27:36 +0200, Michael Kerrisk wrote:
> On Fri, 29 Jul 2005 15:33:11 -0400, Michael B Allen
> <mba2000@ioplex.com> wrote:
>
>
> Hi Mike,
>
> It isn't widely available (not in POSIX.1-2001). Solaris 9 has it.
> (I didn't know that HP-UX had it... are you sure? Which version of
> HP-UX?)
No but I found hp docs about it [1] while googling.
> The portable way of doing thingsis to set a timer before semop() using
> alarm() or setitimer().
I don't particularly like signals for this sort of thing - it's like
wacking someone over the head with a rubber mallet just to get their
attention. But I suppose I will do what I have to do. I guess in my
current case it would not be too bad to use a signal because the timeout
shouldn't happen too frequently.
Thanks,
Mike
[1] http://docs.hp.com/en/B2355-60103/semop.2.html
|
|
|
|
|