This message is from the T13 list server.

> From: Larry Barras [mailto:[EMAIL PROTECTED]]
> Sent: Wed 1/8/2003 5:08 PM
>
> I have no way to know about any other OS's, but you
> should have absolutely no problem accessing the
> last physical block of an ATA HDD on either the
> classic Mac OS or OS X on any Mac that shipped in
> the last 4-5 years.

Intriguingly enough, depending on technique, raw Mac Lba access turns out to be a 
wildly controversial topic.

I hear what works great on the Mac (and on Linux) are commands of the form `sudo dd 
if=/dev/disk0 ...`.  I figure somebody should volunteer to abstract that idea and port 
it back to DOS & Windows.

But the SCSI pass thru of DOS, Windows, Linux, BSD, VxWorks, & friends has suddenly 
gone missing with Mac OS X, as I document in some detail at:
http://members.aol.com/plscsi/mac.html

http://search.lists.apple.com/ata-scsi-dev?q=STUC
is a history of flame wars near here.

I say I find the controversy "intriguing" because I have failed to find a straight 
answer.  Apple even has a non-answer posted on the web for this:
http://developer.apple.com/qa/qa2001/qa1179.html

http://lists.apple.com/mhonarc/ata-scsi-dev/
at this moment shows you an ongoing work of trying to elaborate QA1179 til it makes 
sense, this time without having a flame ware stop the conversation mid-flow, in the 
thread titled "Re: SCSI pass thru lack still unexplained".

But at this moment, with Mac OS X, you can't say op x25 Read Capacity (i.e. the ATAPI 
analogoue of getting maxLba+1 from op xEC ATA Identify data), see the maxLba reported 
there, and write the maxLba, unless you're willing to patch the kernel yourself.  
Whooops.  Maybe "everybody knows" Mac OS X is BSD except that in more or less closed 
source Apple "improved" the GUI and the I/O.

> of an ATA HDD

I expect you mean an ATA HDD shipped inside the more or less sealed box that Apple 
ships.

But all the same in passing I will mention, what hardware connects the ATA HDD 
matters.  Certain bridges to the ATA of Compact Flash are thought to increment by one 
the reported last Lba e.g.
http://lxr.linux.no/ident?v=2.5.49;i=US_FL_FIX_CAPACITY

Pat LaVarre

Reply via email to