Andrea Arcangeli (andrea@e-mind.com)
Thu, 17 Dec 1998 23:23:42 +0100 (CET)
On Thu, 17 Dec 1998, Tim Waugh wrote:
>On Thu, 17 Dec 1998, Andrea Arcangeli wrote:
>
>> @@ -129,7 +123,7 @@
>> goto stop;
>>
>> /* Yield the port for a while. */
>> - if (count && polling (dev))
>> + if (count)
>> parport_release (dev);
>> timer.expires = jiffies + wait;
>> timer.data = (unsigned long) port;
>> @@ -137,10 +131,10 @@
>> add_timer (&timer);
>> down_interruptible (&port->ieee1284.sem);
>> del_timer (&timer);
>> - if (count && polling (dev))
>> + if (count)
>> parport_claim_or_block (dev);
>
>Andrea, this patch is wrong, as I've said before. If we are waiting for
>an interrupt, we need to have the port claimed.
It' s not wrong, it' s less efficient, but I' d like to be less effient
instead of kill ppa transfer rate in parport sharing.
>[Even if that weren't so, what would be the use of just checking to see if
> it's the first time round the loop ("if (count)"?), if it wasn't going to
> influence anything?]
Ah right ;) I' ll remove it now... Tim do you want to fix the parport
sharing playing with the preempt of the pardevice as I done with
lp_error()? Philip? Hints?
Andrea Arcangeli
-- To unsubscribe, send mail to: linux-parport-request@torque.net --
-- with the single word "unsubscribe" in the body of the message. --
This archive was generated by hypermail 2.0b3 on Wed 30 Dec 1998 - 10:18:58 EST