Hi Domen,
I think you have found a bug. HZ is not constant by architecture
(for alpha 1024 or 1200).
this looks better for me:
schedule_timeout(usecs*HZ/10000);
can somebody with knowledge on HZ comment on this ?
regads,
walter
ps: i modeled this after 'device drivers' from A.Rubini see: Time Intervalls
but this is an old version of the book.
- - - - - - - - - - - - - - Original Message - - - - - - - - - - - - - -
From: Domen Puncer <domen@xxxxxxxxxxxx>
Subject: [Kernel-janitors] Re: Replaces yield() with nicedelay() on driv
Date: 05/24/04 15:24
On 02/05/04 22:24 -0300, Gustavo Franco wrote:
> Hi,
>
> It's the last REDIFF, but i'm not replacing yield() with
> schedule_timeout() because there's a nicedelay() (maybe the
> usecs need to be tuned below) there:
>
>
> static inline void nicedelay(unsigned long usecs)
> {
> current->state = TASK_INTERRUPTIBLE;
> schedule_timeout(HZ);
> return;
> }
And usecs stands for??
Funny thing is that there are actualy uses of this function.
Also, please make -p1 applyable patches.
_______________________________________________
Kernel-janitors mailing list
Kernel-janitors@xxxxxxxxxxxxxx
http://lists.osdl.org/mailman/listinfo/kernel-janitors
|
