READ_TRACK command clarification

Bmcferrin at aol.com Bmcferrin at aol.com
Thu Apr 17 08:18:47 PDT 2003


* From the T10 Reflector (t10 at t10.org), posted by:
* Bmcferrin at aol.com
*

--part1_150.1e3adfef.2bd01fd7_boundary
Content-Type: text/plain; charset="US-ASCII"
Content-Transfer-Encoding: 7bit

Henry,
It is media and format dependent.




(A) For the READ_TRACK_INFORMATION command, it is possible to specify
the
track using LBA (Type == 0 in CDB).  For fixed-packet CD media, is this
LBA using Method 1 or Method 2 addressing mode?  This is not clear in
Fuji5 nor MMC3 are clear on this matter.  What are current
implementations
doing in this case?




LBA translation is defined in OB.  For CD-MRW DMA, method 3 should be
used.  For CD-MRW GAA and for non-MRW fixed packet, method 2 is used.
For all other CD media formats, method 1 is used.



(B) On a related note (m1 vs m2), how are current implementations
handling
method 2 addressing on the following type of disc:

Single session:
Track #1 == Fixed Packet, 0x20 packets (0x20*0x27 sectors)
Track #2 == Variable Packet, two packets of 0x400 sectors
Track #3 == Fixed Packet, 0x40 packets (0x20*0x27 sectors)

Specifically,
(B.1) What is returned for READ_CAPACITY data?
(B.2) Is the run-in for Track 2 Packet #1 hidden from LBA space?
(B.3) Is the run-out for Track 2 Packet #2 hidden from LBA space?




LBAs are translated by location AND recording type.  So, if you know the
starting LBA of a track and the recording type of the track, then you
know progression of LBAs.  MRW is a special case that always reports
track #1, but gives different numbers based upon the LBA space selected.
Let us not consider that case.
When a track is fixed packet, it should be translated according to
method 2 - regardless of mode.  All other recording types are linear
translations.

Your example appears to be some sort of test disc that has no particular
practical value, but...

B.1: READ CAPACITY is mostly useless on CD - except for MRW.  READ
CAPACITY should return the last lead-out start address - 1 with a
tolerance of 2 seconds.  Sorry.  If the last track is data rather than
audio, the LBA reported should match the physical location that is
reported as track stop address in the PMA entry for that track.  The
tolerance should be 2.

B.2 In the case of either fixed or variable, the link and run-ins of the
first packet are in the track pre-gap.  The address (MSF or LBA)
reported as track start address is the first data sector of the
packet...so, yes.

B.3 Variable packet address translations are always method 1...so, no.


How's that?
Bill McFerrin


--part1_150.1e3adfef.2bd01fd7_boundary
Content-Type: text/html; charset="US-ASCII"
Content-Transfer-Encoding: quoted-printable


Henry,
 It is media and format dependent.
 

track using LBA (Type =3D=3D 0 in CDB).  For fixed-packet CD = media, is this
 LBA using Method 1 or Method 2 addressing mode?  This is not clear = in
 Fuji5 nor MMC3 are clear on this matter.  What are current = implementations
 doing in this case?
 

LBA translation is defined in OB.  For CD-MRW DMA, method 3 should = be used.  For CD-MRW GAA and for non-MRW fixed packet, method 2 is = used.  For all other CD media formats, method 1 is = used. MARGIN-LEFT: 5px; MARGIN-RIGHT: 0px; PADDING-LEFT: 5px">(B) On a = related note (m1 vs m2), how are current implementations handling
 method 2 addressing on the following type of disc:
 
Single session:
 Track #1 =3D=3D Fixed Packet, 0x20 packets (0x20*0x27 sectors)
 Track #2 =3D=3D Variable Packet, two packets of 0x400 sectors
 Track #3 =3D=3D Fixed Packet, 0x40 packets (0x20*0x27 sectors)
 
Specifically,
 (B.1) What is returned for READ_CAPACITY data?
 (B.2) Is the run-in for Track 2 Packet #1 hidden from LBA space?
 (B.3) Is the run-out for Track 2 Packet #2 hidden from LBA space?
 

LBAs are = translated by location AND recording type.  So, if you know the = starting LBA of a track and the recording type of the track, then you = know progression of LBAs.  MRW is a special case that always = reports track #1, but gives different numbers based upon the LBA space = selected.  Let us not consider that case.
 When a track is fixed packet, it should be translated according to = method 2 - regardless of mode.  All other recording types are = linear translations.
 
Your example appears to be some sort of test disc that has no = particular practical value, but...
 
B.1: READ CAPACITY is mostly useless on CD - except for MRW.  READ = CAPACITY should return the last lead-out start address - 1 with a = tolerance of 2 seconds.  Sorry.  If the last track is data = rather than audio, the LBA reported should match the physical location = that is reported as track stop address in the PMA entry for that = track.  The tolerance should be 2.
 
B.2 In the case of either fixed or variable, the link and run-ins of = the first packet are in the track pre-gap.  The address (MSF or = LBA) reported as track start address is the first data sector of the = packet...so, yes.
 
B.3 Variable packet address translations are always method 1...so, = no.
 

How's that?
 Bill McFerrin
 
--part1_150.1e3adfef.2bd01fd7_boundary--




More information about the T10 mailing list