[prev in list] [next in list] [prev in thread] [next in thread] 

List:       linux-arm-kernel
Subject:    Re: [CFT] Mark IO as bufferable
From:       Matthias Welwarsky <mwelwarsky () web ! de>
Date:       2005-06-28 15:59:20
Message-ID: 200506281759.21080.mwelwarsky () web ! de
[Download RAW message or body]

On Monday 27 June 2005 18:37, Russell King - ARM Linux wrote:
> Can people please test out the effect of this change.  Potentially it
> should get slightly better IO performance.  Or it could prevent the
> kernel booting.

I wonder what this change does in detail. Does bufferable mean "use write-back 
cache" or is it just to use the write buffer?

Either way, I think it's quite a dangerous change. Imagine an IDE driver: The 
ATAPI4 spec says (not literally): "After writing to the command register, 
wait for at least 400ns before reading the status register". So the command 
sequence looks something like

	writeb(cmd, HD_COMMAND);
	nop();
	nop();
	...
	nop();
	stat = readb(HD_STATUS);

Can I be sure to not violate the protocol if the writes are buffered?

regards,
	matthias

-------------------------------------------------------------------
List admin: http://lists.arm.linux.org.uk/mailman/listinfo/linux-arm-kernel
FAQ:        http://www.arm.linux.org.uk/mailinglists/faq.php
Etiquette:  http://www.arm.linux.org.uk/mailinglists/etiquette.php
[prev in list] [next in list] [prev in thread] [next in thread]