
-------------------------
FILE: /src/paws/src/scsi/SCSIIF; REV: 56.4;
USER: jwh;  DATE: Thu Mar 11 12:53:27 MST 1993;



Changs to SyncTeac to solve the hfs floppy problem first noted in
Japan.


-------------------------
FILE: /src/paws/src/scsi/SCSIIF; REV: 56.5;
USER: jwh;  DATE: Thu Mar 11 15:50:24 MST 1993;

forgot to set the value of high_density.
ZZ

-------------------------
FILE: /src/paws/src/comp1/GENMOVE; REV: 56.4;
USER: jwh;  DATE: Thu Apr 15 11:56:00 MDT 1993;


  Disallowed maskboolexpr from working on packed boolean expressions.

-------------------------
FILE: /src/paws/src/scsi/SCSIIF; REV: 56.6;
USER: jwh;  DATE: Tue Jul  6 09:35:44 MDT 1993;

Changes for McBeth tape support with BACKUP.

-------------------------
FILE: /src/paws/src/scsi/SCSIIF; REV: 56.7;
USER: jwh;  DATE: Tue Jul  6 10:03:54 MDT 1993;

Changes for McBeth support, fixing editing errors from last time.

-------------------------
FILE: /src/paws/src/scsi/scsidisk; REV: 56.4;
USER: jwh;  DATE: Tue Jul  6 10:18:47 MDT 1993;

Added routine tape_transfer for McBeth support.

-------------------------
FILE: /src/paws/src/osfs/CTABLE; REV: 56.4;
USER: jwh;  DATE: Tue Jul  6 11:06:40 MDT 1993;

Changes to bring the McBeth drive on line at unit #41 or #42.

-------------------------
FILE: /src/paws/src/hprogs/cpio.c; REV: 56.2;
USER: jwh;  DATE: Tue Jul  6 11:28:06 MDT 1993;

Added changes for McBeth support.

-------------------------
FILE: /src/paws/src/hprogs/b_support.p; REV: 56.2;
USER: jwh;  DATE: Tue Jul  6 11:36:38 MDT 1993;

Changes for McBeth support.

-------------------------
FILE: /src/paws/src/hprogs/cpiomain.p; REV: 56.2;
USER: jwh;  DATE: Tue Jul  6 11:49:58 MDT 1993;

Changes for McBeth support.

-------------------------
FILE: /src/paws/src/hprogs/tapeio.p; REV: 56.2;
USER: jwh;  DATE: Tue Jul  6 13:19:41 MDT 1993;

Exported get_unum so it can be called from elsewhere.

-------------------------
FILE: /src/paws/src/hprogs/cpiomain.p; REV: 56.3;
USER: jwh;  DATE: Tue Jul  6 15:35:59 MDT 1993;

Moved the code to restore umaxbytes to 0 after the repeat.

-------------------------
FILE: /src/paws/src/scsi/scsidisk; REV: 56.5;
USER: jwh;  DATE: Tue Jul  6 15:52:09 MDT 1993;

Added call to tape_transfer for devid = -2 devices.

-------------------------
FILE: /src/paws/src/scsi/scsidisk; REV: 56.6;
USER: jwh;  DATE: Tue Jul  6 16:08:05 MDT 1993;

forgot a ;

-------------------------
FILE: /src/paws/src/hprogs/cpiomain.p; REV: 56.4;
USER: jwh;  DATE: Tue Jul  6 16:16:58 MDT 1993;

bumped copyright year to 1993.

-------------------------
FILE: /src/paws/src/hprogs/b_support.p; REV: 56.3;
USER: jwh;  DATE: Wed Jul  7 09:01:57 MDT 1993;

Changed request_vol to set umaxbytes back to 0 for a SCSI tape in the
event of a fatal ioresult.

-------------------------
FILE: /src/paws/src/hprogs/cpiomain.p; REV: 56.5;
USER: jwh;  DATE: Wed Jul  7 09:07:08 MDT 1993;

Added code to set umaxbytes to zero for SCSI tapes right at the beginning of
the main recover block. 

-------------------------
FILE: /src/paws/src/scsi/scsidisk; REV: 56.7;
USER: jwh;  DATE: Thu Jul  8 15:50:32 MDT 1993;

Cleaned up the routine tape_transfer.

-------------------------
FILE: /src/paws/src/hprogs/b_support.p; REV: 56.4;
USER: jwh;  DATE: Fri Sep  3 10:00:43 MDT 1993;

Added HP-HP35480A is a SCSI device BACKUP should work with.

-------------------------
FILE: /src/paws/src/scsi/scsidisk; REV: 56.8;
USER: jwh;  DATE: Sat Nov 20 11:40:12 MST 1993;


Handled a glitch by just going back to the "old" version of tape
transfer. Same as "transfer" but it uses ScsiTapeRead and ScsiTapeWrite
in place of ScsiDiskRead and ScsiDiskWrite. I thought I could make
tape_transfer smaller than transfer (with tape transfer, partial
blocks are impossible) but it's not worth the effort.

JWH 11/20/93.

-------------------------
FILE: /src/paws/src/init2/DEBUGGER; REV: 56.4;
USER: jwh;  DATE: Mon Jan 17 14:46:16 MST 1994;

Fix for named reboot problem spotted by the NSA, reported to us
by Bill Bock. I don't have an SR number as I type this.

Under certain circumstances, re-booting multiple times from an
HFS disk could result in a failed boot. The symptoms sure looked
like an uninitialized variable - possibly in the secondary loader.
Solution - forcibly zero memory up where the secondary loader runs -
in the space right below $fffffb00. I just picked 256k as an arbitrary
value. It seems to do the trick. 

Yes - this fix is a bit of a hack.

-------------------------
FILE: /src/paws/src/init2/DEBUGGER; REV: 56.5;
USER: jwh;  DATE: Mon Jan 17 16:08:44 MST 1994;

Same as last change but I decided to zero 128k not 256k.
