Re: Sider (Xebec/Datamac SASI) drive and UCSD Pascal 1.1 (Apple Pascal)

["John B. Matthews" <nospam@nospam.com>]

In article <40C20313.9338ABD3@REMOVETHIScybertheque.org>,
 msg <msg@cybertheque.org> wrote:

> John, Thanks much for your reply.
> 
> 
> > I seem to recall a utility named SYSTEM.ATTACH that was used to
> > bind drivers to unit numbers. I used it with a RAM disk driver
> > and OKS Cache Card Pascal driver. It appeared on the volume
> > named APPLE4: with Pascal 1.3. Previously it had been sold by
> > Call-APPLE. Sadly, I can't find the docs, but it might be the
> > missing piece.
> > 
> 
> Sadly, we don't have any Apple Pascal printed documentation; we're
> extrapolating from IBM UCSD P-system materials.  I now know
> that there was dynamic device driver support available for
> versions of Apple Pascal; this excerpt is from a Technote:
> 
> The Boot Process
> 
> The boot code (contained in blocks 0 and 1 of the boot disk) is
> loaded into memory by the Autostart ROM. It checks for the
> P-machine file and loads it into RAM. The P-machine, in turn,
> brings in and initializes the Run-time operating system.
> (In the case of a two-stage boot, the message "Insert boot disk with
> SYSTEM.PASCAL on it, then press RETURN" appears after the P-machine
> has been loaded. The user should then insert the second-stage boot
> disk and press the Return key, which results in the operating system
> being loaded and initialized.) The first noteworthy action taken by
> the operating system is to execute SYSTEM.ATTACH, if that utility
> program is available on the Vendor Product Disk. Remember that
> SYSTEM.ATTACH must not be present on the Vendor Product Disk unless
> special, low-level I/O drivers must be bound into the system. As
> explained more fully in Apple II Pascal Device and Interrupt Support
> Tools,
> SYSTEM.ATTACH uses two special data files and will fail if these files
> are not present on the boot disk.
> 
> However, there is no file 'SYSTEM.ATTACH' or its data files on the
> root volume (or any volume) of the original hard disk, yet the original
> system boots properly and recognizes the Xebec Pascal volumes.
> 
> Curiously, there also is no 'SYSTEM.ATTACH' as a directory entry on
> any of the Siderware disks from Wayne's site, and the contents of
> disk4 is a bit bizarre: it contains a single 2 block program called
> 'QUIT.CODE', however upon inspecting the remaining unused area on
> the disk one finds a great deal of code with imbedded strings
> containing references to 'SYSTEM.ATTACH' and I/O primitives, snippets
> of Assembler and Pascal source, and a menu for backup and restore
> operations.  Unfortunately, running 'QUIT.CODE' on our system merely
> corrupts and screen and exits to the monitor.
> 
> Unless there is some 'hidden file' mechanism at work on our original
> disk, there must have been some installation program which statically
> linked Xebec drivers into the P-system somehow.
> 
> Anyone have any Xebec support tools from about 1981 to 1983?
> 
> Michael

Here's an example of a simple driver using system.attach, 
attach.drivers and attach.data:

http://web.pdx.edu/~heiss/technotes/pasc/tn.pasc.16.html

BTW, I erred above: The disk is named "ATTACH:", not "APPLE4:". 
It's available here:

http://store.syndicomm.com/product_info.php?cPath=2_23_27_29&prod
ucts_id=55

Of course, Xebec may not have used this approach to bind their 
driver.

John
----
jmatthews at wright dot edu
www dot wright dot edu/~john.matthews/