11 Aug
2021
11 Aug
'21
8:57 a.m.
Hi folks,
Could anybody spare a clue or some suggestions on how to access the contents of:
http://bitsavers.trailing-edge.com/bits/DEC/pdp11/floppyimages/rx02/BASIC-1…
under simh? I haven't had any luck mounting the contained BASIC.DSK e.g. on simh RY
under RT-11. Looking through a dump of the image, there seems to be an "RT11A"
signature. Tried putr under dosbox as well, but it seems to hang mounting the image.
Suggestions appreciated!
--FritzM.
11 Aug
11 Aug
1:12 p.m.
On Wed, Aug 11, 2021 at 8:57 AM Fritz Mueller via cctalk
<cctalk at classiccmp.org> wrote:
Hi folks,
Could anybody spare a clue or some suggestions on how to access the contents of:
http://bitsavers.trailing-edge.com/bits/DEC/pdp11/floppyimages/rx02/BASIC-1…
under simh? I haven't had any luck mounting the contained BASIC.DSK e.g. on simh RY
under RT-11. Looking through a dump of the image, there seems to be an "RT11A"
signature. Tried putr under dosbox as well, but it seems to hang mounting the image.
Suggestions appreciated!
--FritzM.
That BASIC.DSK image file has a size of 486,400 bytes. I don't know
where that size would come from.
A physical RX-02 floppy should have a sector size of 256 bytes, with
26 sectors per track, and 77 tracks, which would be a total of 512,512
bytes, or 505,856 bytes if the first physical track is ignored.
Indeed, the other RX-02 floppy images available here do have a size of
505,856 bytes:
http://www.bitsavers.org/bits/DEC/pdp11/floppyimages/rx02/
Hmm, maybe that BASIC.DSK image file was created by something that
only copied the initial allocated logical sectors and ignored unused
logical sectors at the end of the floppy, and maybe PUTR doesn't
handle disk image files that are not the expected full size?
Example of padding the 486,400 byte BASIC.DSK image file to a size of
512,512 bytes on a Windows system:
FSUTIL FILE CREATENEW BLANK 26112
COPY /B BASIC.DSK+BLANK TEST.DSK
C:\PUTR>DIR TEST.DSK
Volume in drive C has no label.
Volume Serial Number is 14CE-1A29
Directory of C:\PUTR
08/11/2021 12:55p 512,512 TEST.DSK
C:\PUTR>PUTR
PUTR V2.01 Copyright (C) 1995-2001 by John Wilson <wilson at dbit.com>.
All rights reserved. See www.dbit.com for other DEC-related software.
COPY mode is ASCII, SET COPY BINARY to change
(C:\PUTR)>MOUNT RX: TEST.DSK /RX02 /RT11 /RONLY
(C:\PUTR)>DIR RX:
Volume in drive RX is RT11A
Directory of RX:\*.*
11-Aug-2021
BSOT0D.EAE 12 04-Apr-1983
BSOT0S.EAE 10 04-Apr-1983
BSOT1D.EAE 9 04-Apr-1983
BSOT1S.EAE 6 04-Apr-1983
BSOT0D.EIS 12 04-Apr-1983
BSOT0S.EIS 9 04-Apr-1983
BSOT1D.EIS 9 04-Apr-1983
BSOT1S.EIS 6 04-Apr-1983
BSOT0S.FIS 7 04-Apr-1983
BSOT1S.FIS 6 04-Apr-1983
BSOT0D.FPU 7 04-Apr-1983
BSOT0S.FPU 7 04-Apr-1983
BSOT1D.FPU 4 04-Apr-1983
BSOT1S.FPU 3 04-Apr-1983
BSCLIN.OBJ 1 04-Apr-1983
BSCLLB.OBJ 5 04-Apr-1983
BSCLS .OBJ 6 04-Apr-1983
BSCLSN.OBJ 1 04-Apr-1983
BSCMP .OBJ 5 04-Apr-1983
BSERML.OBJ 8 04-Apr-1983
BSERMS.OBJ 5 04-Apr-1983
BSERR .OBJ 2 04-Apr-1983
BSE0 .OBJ 16 04-Apr-1983
BSE1 .OBJ 8 04-Apr-1983
BSFUNC.OBJ 9 04-Apr-1983
BSKEYS.OBJ 6 04-Apr-1983
BSOT0D.OBJ 12 04-Apr-1983
BSOT0S.OBJ 9 04-Apr-1983
BSOT1D.OBJ 9 04-Apr-1983
BSOT1S.OBJ 6 04-Apr-1983
BSPAT .OBJ 2 04-Apr-1983
BSPR0 .OBJ 3 04-Apr-1983
BSPR0N.OBJ 3 04-Apr-1983
BSPR1 .OBJ 7 04-Apr-1983
BSPR1N.OBJ 1 04-Apr-1983
BSPTRD.OBJ 6 04-Apr-1983
BSPTRS.OBJ 6 04-Apr-1983
BSRSQ .OBJ 2 04-Apr-1983
BSRSQN.OBJ 1 04-Apr-1983
BSR0D .OBJ 9 04-Apr-1983
BSR0S .OBJ 8 04-Apr-1983
BSR1D .OBJ 4 04-Apr-1983
BSR1S .OBJ 3 04-Apr-1983
BSSUB .OBJ 2 04-Apr-1983
BSSUBN.OBJ 1 04-Apr-1983
BSX0A .OBJ 11 04-Apr-1983
BSX0B .OBJ 8 04-Apr-1983
BSX1A .OBJ 8 04-Apr-1983
BSX1B .OBJ 7 04-Apr-1983
BSX2 .OBJ 7 04-Apr-1983
SUDTCM.OBJ 2 04-Apr-1983
SUEID .OBJ 1 04-Apr-1983
SUE1ID.OBJ 1 04-Apr-1983
SUIMP .OBJ 3 04-Apr-1983
SUIOID.OBJ 1 04-Apr-1983
SUIOPO.OBJ 11 04-Apr-1983
SUIOPR.OBJ 4 04-Apr-1983
SUOPT .OBJ 5 04-Apr-1983
SUO1ID.OBJ 1 04-Apr-1983
SUR1ID.OBJ 1 04-Apr-1983
SUXID .OBJ 1 04-Apr-1983
SUX2ID.OBJ 1 04-Apr-1983
SUCNFG.BAS 3 04-Apr-1983
SUCNF1.BAS 7 04-Apr-1983
SUCNF2.BAS 4 04-Apr-1983
SUCNF3.BAS 4 04-Apr-1983
SUCNF4.BAS 4 04-Apr-1983
SUCNF5.BAS 5 04-Apr-1983
SUCNF6.BAS 5 04-Apr-1983
SUCNF7.BAS 5 04-Apr-1983
SUCVT .BAS 6 04-Apr-1983
SUCVT1.BAS 14 04-Apr-1983
SUCNFG.DAT 7 04-Apr-1983
BSASM .MAC 5 04-Apr-1983
BSCLI .SAV 2 27-Mar-1987
BSMAC .MAC 10 04-Apr-1983
BASIC .SAV 53 04-Apr-1983
YSTART.OBJ 1 30-Mar-1987
ZSTART.OBJ 1 30-Mar-1987
BSASM .OLD 5 04-Apr-1983
BSCLI .OLD 5 04-Apr-1983
BSMAC .OLD 10 04-Apr-1983
BSCLIN.SAV 2 27-Mar-1987
BSASM .OBJ 1 27-Mar-1987
POSITI.LST 7 02-Apr-1987
XSTART.OBJ 1 30-Mar-1987
BSCLI .OK1 9 01-May-1987
JOYSTI.OK1 2 01-May-1987
CNC .OK1 1 01-May-1987
< UNUSED > 12
ACMOTO.MAC 2 13-May-1987
WINDOW.MAC 4 13-May-1987
POSITI.MAC 2 13-May-1987
JOYSTI.MAC 2 13-May-1987
SPEEDS.OLD 2 13-May-1987
< UNUSED > 1
XSTASP.OBJ 1 13-May-1987
YSTASP.OBJ 1 13-May-1987
ZSTASP.OBJ 1 13-May-1987
XDIREC.OBJ 1 13-May-1987
YDIREC.OBJ 1 13-May-1987
ZDIREC.OBJ 1 13-May-1987
ACMOTO.OBJ 1 13-May-1987
WINDOW.OBJ 1 13-May-1987
POSITI.OBJ 1 13-May-1987
STATUS.MAC 1 13-May-1987
STATUS.OBJ 1 13-May-1987
JOYSTI.OBJ 1 13-May-1987
CNC .SAV 58 30-Dec-1987
< UNUSED > 26
SPEEDS.OBJ 1 01-Aug-1987
< UNUSED > 55
VECTOR.MAC 18 30-Dec-1987
< UNUSED > 4
< UNUSED > 2
VECTOR.OBJ 2 30-Dec-1987
BSCLI .MAC 8 30-Dec-1987
CNC .COM 1 30-Dec-1987
XLINIE.MAC 13 30-Dec-1987
YLINIE.MAC 13 30-Dec-1987
ZLINIE.MAC 13 30-Dec-1987
XLINIE.OBJ 2 30-Dec-1987
YLINIE.OBJ 2 30-Dec-1987
ZLINIE.OBJ 2 30-Dec-1987
BSCLI .OBJ 2 30-Dec-1987
< UNUSED > 1
T .BAS 1
NULLPU.XYZ 1 31-Dec-1987
< UNUSED > 149
121 Files, 686 Blocks
250 Free blocks
(C:\PUTR)>TYPE RX:STATUS.MAC
STATUS.MAC
;+
; Basic callable subroutine to receive the status of
; the CNC-machine in one integer word.
; Basic-Call: Call STATUS(S%)
;-
;
.globl STATUS
.globl $ARGER
;
CSR = 167774
;
STATUS: cmpb #1,(r5)+ ; Number of arguments = 1
beq 10$ ; O.K. continue..
5$: jmp $ARGER ; ARGUMENT error
;
10$: tstb (r5)+ ; skip type checking
mov @#csr,r0 ; get status
bic #100000,r0 ; and clear bit 15 ( inhib neg.Int.)
mov r0,@(r5)+ ; save contens.
abo: rts pc ; return to caller
;
.END
(C:\PUTR)>TYPE RX:T.BAS
T .BAS
10 OPEN 'SY:NULLPU.XYZ' FOR INPUT AS FILE #1
20 INPUT #1,A1$,A2$,A3$
30 INPUT #1,B1$,B2$,B3$
40 CLOSE #1
50 PRINT A1$,A2$,A3$
60 PRINT B1$,B2$,B3$
4:06 p.m.
On Wed, Aug 11, 2021 at 8:57 AM Fritz Mueller via
cctalk
Could anybody spare a clue or some suggestions on how to access the contents of:
http://bitsavers.trailing-edge.com/bits/DEC/pdp11/floppyimages/rx02/BASIC-1…
On Aug 11, 2021, at 1:12 PM, Glen Slick via cctalk
<cctalk at classiccmp.org> wrote:
Hmm, maybe that BASIC.DSK image file was created by something that
only copied the initial allocated logical sectors and ignored unused
logical sectors at the end of the floppy, and maybe PUTR doesn't
handle disk image files that are not the expected full size?
Example of padding the 486,400 byte BASIC.DSK image file to a size of
512,512 bytes...[which works]
Ah ha! Thanks much, Glen!
simh is usually pretty good about padding out attached images, but maybe I'll try the
same thing there and see if RT-11 can then read the disk directly itself. If not, I can
extract/rewrite via PUTR.
5:59 p.m.
On Aug 11, 2021, at 7:06 PM, Fritz Mueller via cctalk
<cctalk at classiccmp.org> wrote:
No, I don't think that is true. What is true is that recent versions of SIMH will
create full size container files for disk containers. But sufficiently old versions did
not, so it is not too strange to run into, say, an RK05 image that's shorter than 4800
blocks.
But those normally work; all that happens is that reading too far either gives you zeroes
or a read error, I don't remember which. Since the blocks were never written they
should be marked as unused in the file system and nothing will read them, so that's
fine.
If there's an unused track at the start of a SIMH image file and in your copy that was
missing, that's a different case. Is that the issue?
Or is the RX02 emulation different from regular disks?
paul
...
Ah ha! Thanks much, Glen!
simh is usually pretty good about padding out attached images,
12 Aug
12 Aug
3:31 p.m.
On Wed, Aug 11, 2021 at 5:59 PM Paul Koning <paulkoning at comcast.net> wrote:
It's been a while since the last time I did anything with a SIMH
PDP-11. I just fired up the simh-4.0-Current--2020-06-09-0912a927
version that I last used to take a look at these RX-02 disk images.
As far as I can tell by default PUTR expects to work with logical
sector order RX-02 disk images that are 512,512 bytes in size. The
BASIC-11 RX-02 disk image available here is in logical sector order,
but is less than 512,512 bytes in size:
http://www.bitsavers.org/bits/DEC/pdp11/floppyimages/rx02/
PUTR appears to be unhappy with the disk image unless it is padded to
512,512 bytes in size.
On the other hand as far as I can tell by default SIMH expects to work
with physical sector order RX-02 disk images. If I mount the logical
sector order RX-02 disk image that works with PUTR in SIMH, then RT-11
gives a "?DIR-F-Invalid directory" error. If I translate the logical
sector order RX-02 disk image back into a physical sector order disk
image (dealing with track shifting, sector interleaving, and track to
track sector skewing) then RT-11 on SIMH is happy with the disk image.
On Aug 11, 2021, at 7:06 PM, Fritz Mueller via
cctalk <cctalk at classiccmp.org> wrote:
No, I don't think that is true. What is true is that recent versions of SIMH will
create full size container files for disk containers. But sufficiently old versions did
not, so it is not too strange to run into, say, an RK05 image that's shorter than 4800
blocks.
But those normally work; all that happens is that reading too far either gives you zeroes
or a read error, I don't remember which. Since the blocks were never written they
should be marked as unused in the file system and nothing will read them, so that's
fine.
If there's an unused track at the start of a SIMH image file and in your copy that
was missing, that's a different case. Is that the issue?
Or is the RX02 emulation different from regular disks?
paul ...
Ah ha! Thanks much, Glen!
simh is usually pretty good about padding out attached images,
5:29 p.m.
On Aug 12, 2021, at 3:31 PM, Glen Slick <glen.slick
at gmail.com> wrote:
If I translate the logical
sector order RX-02 disk image back into a physical sector order disk
image (dealing with track shifting, sector interleaving, and track to
track sector skewing) then RT-11 on SIMH is happy with the disk image.
Ah, interesting... Where's the best place to look for more information in the RX02
RT-11 logical/physical mapping?
Glen, if you wouldn't mind sending me a copy of your resulting simh-compatible image,
I'd be happy to post it in an upcoming blog article for others who might find it
useful as well...
cheers,
--FritzM.
5:50 p.m.
On Thu, Aug 12, 2021 at 5:29 PM Fritz Mueller <fritzm at fritzm.org> wrote:
One bit of information that I found helpful as a reference when I
looked at this quite a while ago was the 2.11BSD RX02 floppy disk
device driver source code, which can be viewed online here:
https://minnie.tuhs.org/cgi-bin/utree.pl?file=2.11BSD/sys/pdpuba/rx.c
In particular, the routine rxfactr(), which as the comment says it
calculates the physical sector and physical track on the disk for a
given logical sector.
I used that as a starting point to write a simple utility to read an
RX-02 disk image file in logical sector order and output an RX-02 disk
image file in physical sector order.
/*
* rxfactr -- calculates the physical sector and physical
* track on the disk for a given logical sector.
* call:
* rxfactr(logical_sector,&p_sector,&p_track);
* the logical sector number (0 - 2001) is converted
* to a physical sector number (1 - 26) and a physical
* track number (0 - 76).
* the logical sectors specify physical sectors that
* are interleaved with a factor of 2. thus the sectors
* are read in the following order for increasing
* logical sector numbers (1,3, ... 23,25,2,4, ... 24,26)
* There is also a 6 sector slew between tracks.
* Logical sectors start at track 1, sector 1; go to
* track 76 and then to track 0. Thus, for example, unix block number
* 498 starts at track 0, sector 25 and runs thru track 0, sector 2
* (or 6 depending on density).
*/
static
rxfactr(sectr, psectr, ptrck)
register int sectr;
int *psectr, *ptrck;
{
register int p1, p2;
p1 = sectr / 26;
p2 = sectr % 26;
/* 2 to 1 interleave */
p2 = (2 * p2 + (p2 >= 13 ? 1 : 0)) % 26;
/* 6 sector per track slew */
*psectr = 1 + (p2 + 6 * p1) % 26;
if (++p1 >= 77)
p1 = 0;
*ptrck = p1;
}
On Aug 12, 2021, at 3:31 PM, Glen Slick
<glen.slick at gmail.com> wrote:
If I translate the logical
sector order RX-02 disk image back into a physical sector order disk
image (dealing with track shifting, sector interleaving, and track to
track sector skewing) then RT-11 on SIMH is happy with the disk image.
Ah, interesting... Where's the best place to look for more information in the RX02
RT-11 logical/physical mapping?
Glen, if you wouldn't mind sending me a copy of your resulting simh-compatible image,
I'd be happy to post it in an upcoming blog article for others who might find it
useful as well...
cheers,
--FritzM.
993
days inactive
995
days old
6 comments
3 participants
participants (3)
-
fritzm＠fritzm.org -
glen.slick＠gmail.com -
paulkoning＠comcast.net