Re: [PATCH] fix for ide-scsi crash

[davidsen@tmr.com (bill davidsen)]

In article <E1AixLR-0000Hh-00@neptune.local>,
Pascal Schmidt  <der.eremit@email.de> wrote:
| On Tue, 20 Jan 2004 15:00:16 +0100, you wrote in linux.kernel:
| 
| >> I'd really like the ATA cdrom driver to handle different sector sizes 
| >> properly. There really is no excuse for a block device driver to hardcode 
| >> its blocksize if it can avoid it.
| > 
| > Agree, it's a bug. Pascal, care to take a stab at fixing it? You're the
| > most avid ide-cd non-2kb block size user at the moment :)
| 
| There's a lot of macros related to sector sizes in ide-cd.h. I suppose
| all that would need to be changed to depend on the real hardware sector
| size?
| 
| I don't have any idea about block layer internals or about sending
| commands to a drive.
| 
| In fact, I don't even know how to start or what uses of sector size
| in ide-cd.c really need to be changed. Much less do I know where
| information about hardware sector size could be gotten and how it should
| be stored for use by ide-cd.

To pour confusion on this, I thought this was working already. I just
(Sunday) tried burning audio CDs with the ide-cd interface, using
Joerg's new a25 version of cdrecord which claimed to use DMA for 2352
byte audio sectors. Worked! About 92% idle and 5% wait-io. only 1%
system while burning at 32x. And the CD does play just fine, I listened
to it all the way to work.

Is this a physical/logical sector thing, or is Joerg doing a magic
trick, or what? Definitely does work in terms of producing correct
output, is it really writing 2352 byte secotrs?
-- 
bill davidsen <davidsen@tmr.com>
  CTO, TMR Associates, Inc
Doing interesting things with little computers since 1979.