0x200-0x201 is the address of the game port on a standard PC. Perhaps the
chipset on your MediaGX board thinks it is talking to a joystick at that
address, and the three writes have something to do with that. Just a guess.
-----Original Message-----
From: Kyle Harris [mailto:[EMAIL PROTECTED]]
Sent: Friday, June 23, 2000 11:55 AM
To: linux embedded
Subject: Re: PC/104 timing
This is weird, maybe someone can explain. Originally, I had the LCD
registers mapped to 0x200-0x201 and saw the events described below. When
I change the IO space to something else (e.g., 208, 300, etc),
everything works as expected. /proc/ioports shows 200 to be available.
Does anyone have any ideas why this is so?
Thanks, Kyle.
Kyle Harris wrote:
>
> Hi folks,
>
> I'm trying to get a PC/104 card running on a MediaGX SBC. It's a simple
> LCD controller from Parvus that has registers for LCD data and control.
> I have a user space program using ioperm() and outb() for accessing the
> registers. When I perform a single outb(), I see 3 IOW cycles on the
> bus. The first latches data correctly into the register, but then the
> second comes along and clears it. My question is this. How is the OUTB
> instruction handled by the CPU. Does this go through the BIOS somehow?
> Shouldn't I see exactly what's in the disassembly listing? I realize the
> OS may do other things behind the scenes to other IO spaces, but doesn't
> make sense that it would mess with my LCD space. Any guidance is greatly
> appreciated.
>
> scratching my... uhhh.. head, Kyle.
--
To unsubscribe from this list, send a message to [EMAIL PROTECTED]
with the command "unsubscribe linux-embedded" in the message body.
For more information, see <http://waste.org/mail/linux-embedded>.
--
To unsubscribe from this list, send a message to [EMAIL PROTECTED]
with the command "unsubscribe linux-embedded" in the message body.
For more information, see <http://waste.org/mail/linux-embedded>.
