Re: "Fix ATAPI transfer lengths" causes CD writing regression

[Tejun Heo <htejun@gmail.com>]

Alan Cox wrote:
>> "one size" that you can supply to the SG_IO command (unless you want to 
>> use a stupidly large buffer) to retrieve all the data at once. Instead, 
>> as Tejun describes, you put a short read request in first, look at byte 
>> 1 of the page which tells you the length, and then read the whole lot.
> 
> ATAPI effectively requires you supply a stupidly large buffer. In theory
> you set the transfer size in the lba registers and it all works. In
> practice some drives ignore this and there isn't a nice reliable way to
> clean up.

The transfer size is specified in Allocation Length field inside CDB of
commands which can transfer variable length data.  I got confused about
this too but both the host and device know how much data can be transferred.

> Welcome to the wonderful world of IDE, where the spec sucks and the
> drives manage to do even worse things.
> 
> We can try and clean up better in these cases at least for PIO transfers
> by trying to drain the data beyond this point, on the controllers that
> cope with this but really - fix the app to reflect reality: ATAPI is SCSI
> as spoken by yokels

Not that I disagree to this point tho.  :-)

-- 
tejun