Re: Re: standalone code for bootstrapping

On 3/16/2019 10:02 AM, kerravon86-/[email protected] [hercules-390] wrote:
> But I have a huge program to IPL. 112k.
> I want dasdload to put that huge program
> on disk, and (I believe) the way dasdload
> works, it will put that dataset on the next
> available track. So that means the program
> I want to IPL will be on cylinder 0, head 1,
> record 1.
> 
> The IPL2 record needs to know about that
> new location. Note that that is also existing
> behavior for a 2311 disk. I just need to do
> the same thing for a 3390.
> 
> So what I would like is for my IPLed
> program to be able to cope with the IPL2
> record booting it from either head 0
> record 4 or head 1 record 1. That's the
> design I would like for the assembler
> code in sapstart.asm. It's open to
> discussion though. I may be missing
> something.
What you are missing is flexibility. Depending on usage of a pack, the 
VTOC starts either on track one (e.g., SPOOL packs), or is about 1/3 of 
the way in the pack (lots of small data sets, as used by TSO and Wylbur).

A better design would be to have an intermediate loader program in 
record 4. That would locate your production module by data set name, 
rather than relying on fixed sizes and location.


Gerhard Postpischil
Bradford, VT

---
This email has been checked for viruses by AVG.
https://www.avg.com