Re: [Simh] PDP-10/RP06 Disk Image Size

2015-04-13 Thread Timothe Litt
On 13-Apr-15 01:21, khandy21yo wrote:
 Bad blocks


No.   Bad blocks for MASSBUS disks are within the
non-maintenance cylinders, and tracked in the BAT blocks.

See my previous note for the correct explanation.

For MSCP disks, spare blocks (a.k.a. the RCT that the controller uses
replace bad blocks) are indeed allocated above the LBN space.  But
they are separate from maintenance cylinders and are transparent to
the OS addressing. More or less. MSCP is layered and has various
implementation options.  But that's beyond the scope of the OP's
question.

This communication may not represent my employer's views,
if any, on the matters discussed. 




smime.p7s
Description: S/MIME Cryptographic Signature
___
Simh mailing list
Simh@trailing-edge.com
http://mailman.trailing-edge.com/mailman/listinfo/simh

Re: [Simh] PDP-10/RP06 Disk Image Size

2015-04-13 Thread Timothe Litt

On 13-Apr-15 01:07, Rob Doyle wrote:
 I'm designing an RH11 disk controller for my KS10 FPGA. I'd like to
 use the SIMH data format for the media.

 I noticed that an RP06 disk image from SIMH is 315,187,200 bytes.

 Assuming an RP06 has 815 cylinders, 19 heads, 20 sectors (confirmed
 in SIMH code) - I calculate that disk image should be 317,132,800
 bytes.

 If use 810 cylinders (instead of 815), I get the size of the of the
 image file.

 SIMH doesn't pack data: there are 128 words per sector and each
 word occupies 64-bits or 8 bytes.  Therefore a sector is 1024 bytes.

 What's am I missing?

Maintenance mode.

The 5 highest cylinders of an RP06 are reserved for maintenance
operations -
providing a place for diagnostics to do writes and reads without
touching the
filesystem.  This is a software convention; the hardware doesn't have any
special support for this (though MSCP controllers do.)  The number of
maintenance cylinders was established for each drive type.

SimH won't have any reason to seek to them unless you run a timesharing
diagnostic that acesses
these cylinders, which you probably haven't.  So your host OS hasn't set
EOF out there.

There are special system calls that allow access under timesharing; e.g.
for TOPS-10, SU.SMN is set
in the SUSET. UUO to allow access to the maintenance cylinders.

From rhxkon.mac, the geometry tables:
|
||;BLOCKS PER UNIT||
||
||BLKPRU:DEC154280;(RP04/05) 406 CYLINDERS||
||DEC307800;(RP06) 810 CYLINDERS||
||DEC123150;(RM03) 821 CYLINDERS||
||DEC865504;(RP07) 629 CYL, 32 SURF, 43 SECT||
||
||;BLOCKS PER UNIT IN MAINTENANCE MODE||
||
||BLKPUM:DEC156180;(RP04/RP05) 411 CYLINDERS||
||DEC309700;(RP06) 815 CYLINDERS||
||DEC123450;(RM03) 823 CYLINDERS||
||DEC866880;(RP07) 630 CYLINDERS||
||
||;BLOCKS PER UNIT IN COMPATIBILITY MODE||
||
||BLKPUC:DEC171798;(RP04/RP05) 22*19*411||
||DEC340670;(RP06) 22*19*815||
||DEC131680;(RM03) 32*5*823||
||0;(RP07) NO COMPATIBILITY MODE||
||
||;BLOCKS PER TRACK,,BLOCKS PER CYLINDER||
||
||BLKPTC:XWD^D20,^D380;(RP04)||
||XWD^D20,^D380;(RP06)||
||XWD^D30,^D150;(RM03)||
||XWD^D43,^D1376;(RP07)||
|
For TOPS-20, IIRC, you get to use DSKOP%.

Stand-alone diagnostics knew to restrict their operations to the
maintenance cylinders unless told by the operator that scribbling
was OK (a scratch pack is mounted).

This communication may not represent my employer's views,
if any, on the matters discussed. 




smime.p7s
Description: S/MIME Cryptographic Signature
___
Simh mailing list
Simh@trailing-edge.com
http://mailman.trailing-edge.com/mailman/listinfo/simh

[Simh] PDP-10/RP06 Disk Image Size

2015-04-12 Thread Rob Doyle

I'm designing an RH11 disk controller for my KS10 FPGA. I'd like to
use the SIMH data format for the media.

I noticed that an RP06 disk image from SIMH is 315,187,200 bytes.

Assuming an RP06 has 815 cylinders, 19 heads, 20 sectors (confirmed
in SIMH code) - I calculate that disk image should be 317,132,800
bytes.

If use 810 cylinders (instead of 815), I get the size of the of the
image file.

SIMH doesn't pack data: there are 128 words per sector and each
word occupies 64-bits or 8 bytes.  Therefore a sector is 1024 bytes.

What's am I missing?

Rob

--
KS10 FPGA Project at: http://www.techtravels.org/KS10FPGA/
___
Simh mailing list
Simh@trailing-edge.com
http://mailman.trailing-edge.com/mailman/listinfo/simh

Re: [Simh] PDP-10/RP06 Disk Image Size

2015-04-12 Thread khandy21yo
Bad blocks

 Original message 
From Rob Doyle radioe...@gmail.com 
Date: 04/12/2015  11:07 PM  (GMT-07:00) 
To simh@trailing-edge.com 
Subject [Simh] PDP-10/RP06 Disk Image Size 
 
I'm designing an RH11 disk controller for my KS10 FPGA. I'd like to
use the SIMH data format for the media.

I noticed that an RP06 disk image from SIMH is 315,187,200 bytes.

Assuming an RP06 has 815 cylinders, 19 heads, 20 sectors (confirmed
in SIMH code) - I calculate that disk image should be 317,132,800
bytes.

If use 810 cylinders (instead of 815), I get the size of the of the
image file.

SIMH doesn't pack data: there are 128 words per sector and each
word occupies 64-bits or 8 bytes.  Therefore a sector is 1024 bytes.

What's am I missing?

Rob

--
KS10 FPGA Project at: http://www.techtravels.org/KS10FPGA/
___
Simh mailing list
Simh@trailing-edge.com
http://mailman.trailing-edge.com/mailman/listinfo/simh___
Simh mailing list
Simh@trailing-edge.com
http://mailman.trailing-edge.com/mailman/listinfo/simh