"Fix ATAPI transfer lengths" causes CD writing regression

[Daniel Drake <dsd@gentoo.org>]

Hi Alan,

In 2.6.23 and previous, CD writing works fine on my system. I'm using 
ata_piix on:

00:1f.2 IDE interface: Intel Corporation 82801GBM/GHM (ICH7 Family) SATA 
IDE Controller (rev 01)

When I'm running CD writing utilities, I sometimes see this message in 
the kernel logs:
	ata2.00: 66 bytes trailing data
Things do work fine though.

With 2.6.24-rc1, I can't write CDs. Instead of the above message, I get:
ata2.00: exception Emask 0x0 SAct 0x0 SErr 0x0 action 0x2 frozen
ata2.00: cmd a0/00:00:00:0a:00/00:00:00:00:00/a0 tag 0 cdb 0x5a data 10 in
          res 58/00:02:00:0a:00/00:00:00:00:00/a0 Emask 0x2 (HSM violation)
ata2.00: status: { DRDY DRQ }
ata2: soft resetting link
ata2.00: configured for UDMA/33
ata2: EH complete

and the software acts oddly - for example Brasero tells me that the 
blank CD I have inserted has a capacity of 17,179,869,184gb and it 
doesn't let me write CDs as it says the medium is not writable.

git bisect lead me to commit 2db78dd302d26d242d3e8e5c4c5024b6c3ea93c2 as 
the culprit.

Author: Alan Cox <alan@lxorguk.ukuu.org.uk>
Date:   Tue Oct 2 13:53:04 2007 -0700

     libata_scsi: Fix ATAPI transfer lengths

     Some controller variants snoop the ATAPI length value for Packet
     transfers to do state machine and FIFO management. Thus we want to
     set it properly, even for cases where it is otherwise meaningless.

Any ideas?

Thanks!
Daniel