Re: Re: standalone code for bootstrapping
| Newsgroups | gmane.comp.emulators.hercules390.general |
|---|---|
| Message-ID | <CAPcd4G9azE8Q_bk8T=HNSfrUuYEuqQiNezDDgQR+pNoBKZ0AKg@mail.gmail.com> |
I dont understand why youre not taking advantage of the capabilities of the hardware. Ever hear of something called a TIC (transfer-in-channel)? If you write a chained IPL PSW/CCW, you can TIC back to track 0, and then figure out what type of DASD you are on by SRCH to record* 4, then read record 4... For Example (Borrowed from another program) - IPLPSW DC XL8'0008000000000300' EC MODE RESTART PSW IPLCCW02 DC XL8'0600027060000090' RD DATA OF REC 2 IPLCCW03 DC XL8'0800027060000001' TIC TO 2ND IPL RECORD And then ... IPLCCW04 DC XL8'0700029060000006' SEEK TO TRK 0 IPLCCW05 DC XL8'3100029660000005' SRCH ID EQ FOR REC 4 IPLCCW06 DC XL8'0800027860000001' TIC BACK UNTIL FOUND IPLCCW07 DC XL8'0600030020004A00' RD DATA OF REC 4 All of that is done in the CHANNEL :) * Remember that in hardware language, a record is a block, so a SRCH to record 4 in HW language is a SRCH to block 4 in software. Joe On Sat, Mar 16, 2019 at 5:40 AM kerravon86-/[email protected] [hercules-390] < [email protected]> wrote: > > > ---In [email protected], <kerravon86@...> wrote : > > > Actually I think the logic in sapstart should > > be to start reading 32767 bytes from > > head 0 record 4, > > I think rereading the first block will clobber > modified code, so it will need to instead be > read to a different location, like 32768, just > to get the length and the start record > properly figured out. > > BFN. Paul. > >