Hello,
Thanks.
You said
> If you really meant 1 microsecond (as opposed to 1 millisecond)
Yes ,I mean it. In short: it's for a video-audio app; this app can be
termed soft (or semi) real time; The terminology is not important at
all. The fact is that there are scenarios when
the timing is critical; I mean a packet that arrived at a certain time
should not be sent immediately but after some sleep interval; this
if for keeping a specified rate , which is important for a good quality.
So sometimes a microsec sleep is needed, sometimes 5 or more.
(Maybe it would have been better to phrase my question as
sleeping in a resolution of microsecond instead of 1 microsec).
You said:
>to do a busy wait loop for the duration of the time required.
may I ask ehat is the recommnded way to write a a busy wait loop?
If I am not wrong there is no "noop" command in "C" ; is there a noop
is asm to achieve this ? or simply a loop which just do dummy
operation
like incrementing some integer ?
I tried using outb() from user space in the following way:
According to chap. 9 of LDD3 (Communicating with HW),which generally
talks about I/O ports,you can access I/O ports from user space in
certain cases ; for that you need to call ioperm (or iopl) before.
It says there : (page 242)
"on the x86 the pause is achieved by performing an outb instruction to
port 0x80, or by busy waiting."
BTW , also "Linux I/O port programming mini-HOWTO" mentions this 0x80
port in this context.
I had tried outb() on port 0x80 and it gave a different results
on different platforms.
Regarding gettimeofday() discussion which evolved from this thread:
this is an interesting topic in itself and I am glad to see it here.
I have a little remark regarding this (I may be wrong)
Guy Keren said:
" for the gory details - arch/XXX/kernel/time.c -
do_gettimeofday()."
I had looked at the sys call wrapper which calls do_gettimeofday(),
which is in kernel/time.c (NOT under arch/... subtree)
and I saw there a put_user() call , which means
copying to user space if I am not wrong.
Copying to user space seems to me a heavy operation
(even if we only pass a little timeval struct).
Regards,
RG
On 8/18/05, Gilad Ben-Yossef <[EMAIL PROTECTED]> wrote:
> Rafi Gordon wrote:
> > Hello,
> >
> > Is there a way to create a delay of 1 microsec in a user space
> > applcation in 2.4
> > or 2.6 kernel?
>
> If you really meant 1 microsecond (as opposed to 1 millisecond) then the
> answer is to do a busy wait loop for the duration of the time required.
> Any system call will take more then 1 micro just to execute so they're
> out of the question.
>
> Cheers,
> Gilad
> --
> Gilad Ben-Yossef <[EMAIL PROTECTED]>
> Codefidence. A name you can trust(tm)
> Web: http://codefidence.com | SIP: [EMAIL PROTECTED]
> IL: +972.9.8650475 ext. 201 | Fax: +972.9.8850643
> US: +1.360.2275194 ext. 201 | Cel: +972.52.8260388
>
> "I am Jack's Overwritten Stack Pointer"
> -- Hackers Club, the movie
>
================================================================To unsubscribe,
send mail to [EMAIL PROTECTED] with
the word "unsubscribe" in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]
