RL02-USB Controller Status/Problem

Pete Turnbull pete at dunnington.plus.com
Tue Apr 7 07:41:08 CDT 2015

On 06/04/2015 23:54, Christopher Parish wrote:
> Progress is good on the RL02-USB controller. I've gotten complete
> operation working as expected with the usual tools for disk access
> (badblocks, dd, etc.), and SIMH can access the real packs via the
> controller's block device (i.e. attach rl0 /dev/sdX).

Very good!  I want one :-)

> The issue is most obvious when backing up and restoring disk images.
> Suppose I backup a pack with 1 bad sector. I have two choices of what
> to do with the bad sector (specifically if it's a bad header), I
> could skip the sector (reporting an IO error), or I could report all
> zeros for the sector.

 > The usual trick of having the controller re-map the bad sectors will
 > not work because the platters in the RL02 are removable.

DEC does it by having the driver software remap sectors.

 > Writing some
 > mapping index on the disk pack or holding back sectors in reserve
 > will break compatibility with the original systems (PDP8s/11s/VAXes)
 > and all their software (an unacceptable solution).

Eh?  Use the existing bad block table which is on an RL01 or RL02, as on 
many other DEC disks.

Return an error.  Unless it's a block that's already in the bad block 
table at the end of the disk, in which case return the content of the 
remapped block, of course.  If you don't do that, you'll break things in 

> I vaguely remember reading about a program on RT-11 you would run
> (before?) backing up the filesystem. What was this program? How did
> it work? Did it make the data position independent?

There are several ways to detect or handle bad blocks in RT-11.  I 
suspect you're thinking of INIT/BAD or more likely DIR/BAD, or possibly 

INIT is is the monitor command to create an RT-11 filesystem on a disk; 
the /BAD option (full name BADBLOCKS, can be shortened to as little as 
/BA) tells it to scan the disk and report any bad blocks it finds.  With 
no other options, it creates directory entries for files called FILE.BAD 
covering those blocks.  INIT/BAD:RETAIN keeps any FILE.BAD it finds, and 
in most versions doesn't actually scan.  This is really meant for 
devices that don't have a DEC standard bad block table (like floppies).

Alternatively INIT/BAD/REPLACE doesn't write FILE.BAD but instead 
updates the bad block table at the end of the disk; /REPLACE overwrites 
the table, whereas REPLACE:RETAIN adds to it.  Note that the DL(X) 
drivers in RT-11, as in other DEC OSs, know about the bad block table.

But of course you wouldn't use INIT just before making a backup ;-)
And it wouldn't work if your controller lies and returns zeros for bad 
blocks; that would be a very bad idea and will cause things to fail badly.

The SQUEEZE command compresses a disk by shuffling up all the files 
(except FILE.BAD) to be contiguous (RT-11 always writes individual files 
as contiguous blocks on disk but gaps between files can arise as things 
are deleted or rewritten).  It honours the bad block table (and/or 
FILE.BAD), as do other disk utilities, because the drivers do that.  If 
your driver doesn't, it'll fail.

If you just want to quickly scan a disk for errors, a common way is to 
do COPY/DEV/IGN DLx: NL: which does a block for block copy of the entire 
disk to the NULL device, reporting errors but not stopping on them 
(/COPY/DEVICE/IGNORE).  It's common to do that just after formatting a 
disk, but before INIT, for example if you're working with a non-RT-11 
filesystem.  BTW, for other COPY operations, it ignores FILE.BAD files, 
for obvious reasons, unless you specifically include them.

You were probably thinking of the DIR command.  /BAD is also an option 
to DIR; it can tell also you which files are affected by bad blocks.  It 
scans the entire disk for bad blocks, and even works on non-RT-11 disks. 
  If it finds any, it will report the block number in octal and decimal, 
and tell you if it's "Replaced" or "Replaceable" - ie if it's already in 
the bad block table - but it won't change the table for you.  The full 
tell you what files are affected, assuming it's an RT-11 filesystem; 
/START and /END allow you specify a range rather than the whole disk. 
There's also a /WAIT option that pauses to let you change disks before 

BACKUP (or BUP) scans the target disk for bad blocks before copying a 
saveset to it (and stops if it finds any).

You can't FORMAT an RL01 or RL02, but you can FORMAT/VERIFY:ONLY one. 
That writes patterns over the surface to test it, much more thoroughly 


Pete Turnbull

More information about the cctalk mailing list