On 20/07/04 21:01 +0000, Nishanth Aravamudan wrote:
...
> --- linux-vanilla/drivers/char/agp/sis-agp.c 2004-06-16 05:18:56.000000000
> +0000
...
> - set_current_state(TASK_UNINTERRUPTIBLE);
> - schedule_timeout (1+(HZ*10)/1000);
> + msleep(HZ); /* sleep 1s */
Umm... what?
I mean... HZ jiffies is one second, but HZ miliseconds is not necessary 1 s.
And that delay looks wierd, what was the point of HZ*10/1000 in the first
place? Why not HZ/100?
Anyway, IMO msleep(10) seems right.
Domen
_______________________________________________
Kernel-janitors mailing list
Kernel-janitors@xxxxxxxxxxxxxx
http://lists.osdl.org/mailman/listinfo/kernel-janitors
|