Re: 2.11 BSD on an 11/93 with an CMD SCSI Module
On Tue, Mar 26, 2019 at 5:10 PM Bill Gunshannon via cctalk
wrote:
>
> On 3/26/19 7:54 PM, Dennis Boone via cctalk wrote:
> > > Mine has two entries:
> >
> > > tms 0 174500 2605 tmsintr # tmscp driver
> > > tms 1 164334 0 5 tmsintr # tmscp driver
> >
> > You set the vector on the second one to zero. Should it be
> > something else?
> >
> > De
> >
>
> I didn't set it. It came that way in the 2.11 distro.
> It apparently picks a free vector when it is activated as
> demonstrated by the listing provided by init.
>
> tms 0 csr 164334 vector 774 vectorset attached
>
>
> bill
When in doubt, Use the Source.
src\sys\autoconfig\do_config.c
expect_intr(dp)
DTAB*dp;
{
HAND*hp;
register intaddr = dp->dt_vector;
/*
* A vector of 0 has special meaning for devices which support programmable
* (settable) vectors. If a xxVec() entry point is present in the driver and
* /etc/dtab has a value of 0 for the vector then 'autoconfig' will allocate
* one by calling the kernel routine 'nextiv'.
Re: 1/2" tape storage.
> On Mar 26, 2019, at 8:14 PM, Chuck Guzis via cctalk > wrote: > > Argh! I just got a batch of 10.5" open-reel tape in and about one-third > of the batch has vinyl "hanger strips" that heave either self-destructed > in shipment or are about to do so. > > Given that the durned things have lapsed into the land of the > unobtainium, I wonder if plastic (probably polypropylene) 16mm film cans > will work. The 1200' size looks to be about right. > > Has anyone tried this? Plastic cans were the other standard tape reel packaging, and I think predates the tape seals. Sure that works. Just make sure they are tall enough not to compress the reel sides. Ideally you'd have a hub in the middle of the can to center the reel hub, though it would work without that, just not as nicely. Metal (non-ferrous) cans would work too. Actually, even some ferrous materials would work. I once had a tape shipped to me in a box lined with mu-metal, because the shipping department saw "magnetic tape" on the shipping form so they applied the packaging rules for magnetic materials (i.e., permanent magnets). Nice, I saved it since that stuff is not easily found. paul
Re: 1/2" tape storage.
> > wonder if anyone on tapeheads.net could help u with ideas?
1/2" tape storage.
Argh! I just got a batch of 10.5" open-reel tape in and about one-third of the batch has vinyl "hanger strips" that heave either self-destructed in shipment or are about to do so. Given that the durned things have lapsed into the land of the unobtainium, I wonder if plastic (probably polypropylene) 16mm film cans will work. The 1200' size looks to be about right. Has anyone tried this? --Chuck
Re: 2.11 BSD on an 11/93 with an CMD SCSI Module
On 3/26/19 7:54 PM, Dennis Boone via cctalk wrote: > > Mine has two entries: > > > tms 0 174500 2605 tmsintr # tmscp driver > > tms 1 164334 0 5 tmsintr # tmscp driver > > You set the vector on the second one to zero. Should it be > something else? > > De > I didn't set it. It came that way in the 2.11 distro. It apparently picks a free vector when it is activated as demonstrated by the listing provided by init. tms 0 csr 164334 vector 774 vectorset attached bill
Re: 2.11 BSD on an 11/93 with an CMD SCSI Module
> Mine has two entries: > tms 0 174500 2605 tmsintr # tmscp driver > tms 1 164334 0 5 tmsintr # tmscp driver You set the vector on the second one to zero. Should it be something else? De
Re: 2.11 BSD on an 11/93 with an CMD SCSI Module
On Tue, Mar 26, 2019 at 12:35 PM Bill Gunshannon via cctalk wrote: > Yeah, got all that. The problem seems to be init not finding them > during the configure system stage. > > And even given the possible confusion over multiple TMSCP controllers, > it doesn't explain why the second I add a TMSCP drive all the others > seem to just go away in the eyes of 2.11. Are you using the full $$$ version of E11, or the demo version? I was going to experiment with multiple TMSCP controllers using SIMH, only to realize that while it is built to support up to 4 MSCP controllers, it is only built to support a single TMSCP controller. Then I thought I would try E11 as I assumed it must support multiple TMSCP controllers if you were trying it there. Gave it a try just now with the demo version of E11 V7.3 only to realize that the demo version supports only one of each disk/tape controller type. If you are also using the demo version of E11 instead of the full $$$ version then that would be a problem trying to experiment with 2.11BSD support of multiple TMSCP controllers. The only way I could try experimenting with 2.11BSD support for multiple TMSCP controllers is with real hardware (unless I went down the path of trying to figure out how to rebuild SIMH to support multiple TMSCP controllers, has anyone already done that?)
Re: 2.11 BSD on an 11/93 with an CMD SCSI Module
On 3/26/19 3:02 PM, Glen Slick via cctalk wrote:
> On Tue, Mar 26, 2019 at 11:23 AM Bill Gunshannon via cctalk
> wrote:
>>
>> The saga continues!!! I really don;t remember it being this
>> hard back in the good ole' days but then we had lots of real
>> hardware to work with and systems weren't kludges like what
>> I have in my house. :-)
>>
>> I am currently working with E11 V7.3. My prime interest is
>> in getting a 2.11 system running so I can use tcopy(1) to
>> copy tape images to my one real tape drive. It is SCSI so
>> it is going to be TMSCP on the real system. It is also
>> TMSCP on the E11 system.
>>
>> The problem is I don't seem to be able to do anything on E11
>> to make 2.11 BSD actually have more than one functional tape.
>>
>> I have tried two TMSCP tapes. I have tried two TMSCP controllers.
>> I have tried one TMSCP tape and various configurations of other
>> tape drives. the result is always the same. I hcae tried building
>> a kernel with all the tapes included. I have tried building kernels
>> with two tapes at a time with different combinations of TMSCP and
>> the others. None of them ever come up with more than one drive.
>>
>> March 26 05:11:44 init: configure system
>>
>> hk ? csr 177440 vector 210 skipped: No autoconfig routines.
>> ht ? csr 172440 vector 224 skipped: No autoconfig routines.
>> ra 0 csr 172150 vector 154 vectorset attached
>> rl ? csr 174400 vector 160 skipped: No autoconfig routines.
>> tm ? csr 172520 vector 224 skipped: No autoconfig routines.
>> tms 0 csr 174500 vector 260 skipped: No CSR.
>> tms 1 csr 164334 vector 774 vectorset attached
>> ts ? csr 172520 vector 224 skipped: No autoconfig routines.
>> xp ? csr 176700 vector 254 skipped: No autoconfig routines.
>>
>>
>> I especially don't understand the "No CSR" for tms Unit 0.
>> If I remove the entry for Unit 1 in the E11.ini file then it
>> recognizes Unit 0. But never both and never any other tape
>> drive.
>>
>> Is there something in 2.11 that I am missing?
>>
>> bill
>
> Do you need two separate TMSCP controllers, or is that just an
> experiment you were trying and would two tape drive units on a single
> TMSCP controller work equally well for your purpose?
I'm happy either way. But, from the E11 documentation:
"DEC generally didn’t have more than one tape drive per
TMSCP controller, but as with MSCP disks, E11 has no such artificial
limits. However this means it may be
possible to configure a system which is incompatible with the PDP-11
operating system’s TMSCP driver, so it is
best to define a separate controller for each drive anyway."
>
> Anyway, did you modify /etc/dtab at all? I believe the /etc/dtab file
> in the standard 2.11BSD contains this single entry for a TMSCP
> controller:
>
> tms ? 174500 260 5 tmsintr # tmscp driver
>
Mine has two entries:
tms 0 174500 2605 tmsintr # tmscp driver
tms 1 164334 0 5 tmsintr # tmscp driver
> If you want / need two separate controllers I believe you need to add
> another entry for the second controller.
>
> Also, did you go in /etc/dev and ./MAKDEV tu{unit number} for the
> second tape drive?
Yes, but it didn't work as I would expect being as the init message says
no controller found.
>
> The 2.11BSD installation guide tells you do this for TMSCP tape drives
> during installation. I believe that will only create device nodes for
> a single tape drive.
>
> # cd /dev; rm *mt*; ./MAKEDEV tu0; sync
>
> Without doing some experimenting myself I forget what the actual
> syntax would be for creating additional tape drive units on a single
> or multiple TMSCP controllers.
I have run it again for the additional devices but they end out being
dead.
>
> What do you see if you do "ls -l /dev/*mt*" ?
ls -l shows all the tape devices get created but they are none
functional.
>
> From the tmscp.4 man page the major / minor device numbers have the
> following meanings:
>
> major device number(s):
>
> raw: 23
> block: 12
>
> minor device encoding:
>
> bit:|7 | 6 | 5 | 4 | 3 | 2 | 1 | 0|
> ---
> C C X D D N U U
>
> C = Controller # (max of 4 controllers)
> D = Density
> N = Norewind on close
> U = Unit (drive) number (max of 4 drives per controller)
>
Yeah, got all that. The problem seems to be init not finding them
during the configure system stage.
And even given the possible confusion over multiple TMSCP controllers,
it doesn't explain why the second I add a TMSCP drive all the others
seem to just go away in the eyes of 2.11.
bill
Re: 2.11 BSD on an 11/93 with an CMD SCSI Module
On Tue, Mar 26, 2019 at 11:23 AM Bill Gunshannon via cctalk
wrote:
>
> The saga continues!!! I really don;t remember it being this
> hard back in the good ole' days but then we had lots of real
> hardware to work with and systems weren't kludges like what
> I have in my house. :-)
>
> I am currently working with E11 V7.3. My prime interest is
> in getting a 2.11 system running so I can use tcopy(1) to
> copy tape images to my one real tape drive. It is SCSI so
> it is going to be TMSCP on the real system. It is also
> TMSCP on the E11 system.
>
> The problem is I don't seem to be able to do anything on E11
> to make 2.11 BSD actually have more than one functional tape.
>
> I have tried two TMSCP tapes. I have tried two TMSCP controllers.
> I have tried one TMSCP tape and various configurations of other
> tape drives. the result is always the same. I hcae tried building
> a kernel with all the tapes included. I have tried building kernels
> with two tapes at a time with different combinations of TMSCP and
> the others. None of them ever come up with more than one drive.
>
> March 26 05:11:44 init: configure system
>
> hk ? csr 177440 vector 210 skipped: No autoconfig routines.
> ht ? csr 172440 vector 224 skipped: No autoconfig routines.
> ra 0 csr 172150 vector 154 vectorset attached
> rl ? csr 174400 vector 160 skipped: No autoconfig routines.
> tm ? csr 172520 vector 224 skipped: No autoconfig routines.
> tms 0 csr 174500 vector 260 skipped: No CSR.
> tms 1 csr 164334 vector 774 vectorset attached
> ts ? csr 172520 vector 224 skipped: No autoconfig routines.
> xp ? csr 176700 vector 254 skipped: No autoconfig routines.
>
>
> I especially don't understand the "No CSR" for tms Unit 0.
> If I remove the entry for Unit 1 in the E11.ini file then it
> recognizes Unit 0. But never both and never any other tape
> drive.
>
> Is there something in 2.11 that I am missing?
>
> bill
Do you need two separate TMSCP controllers, or is that just an
experiment you were trying and would two tape drive units on a single
TMSCP controller work equally well for your purpose?
Anyway, did you modify /etc/dtab at all? I believe the /etc/dtab file
in the standard 2.11BSD contains this single entry for a TMSCP
controller:
tms ? 174500 260 5 tmsintr # tmscp driver
If you want / need two separate controllers I believe you need to add
another entry for the second controller.
Also, did you go in /etc/dev and ./MAKDEV tu{unit number} for the
second tape drive?
The 2.11BSD installation guide tells you do this for TMSCP tape drives
during installation. I believe that will only create device nodes for
a single tape drive.
# cd /dev; rm *mt*; ./MAKEDEV tu0; sync
Without doing some experimenting myself I forget what the actual
syntax would be for creating additional tape drive units on a single
or multiple TMSCP controllers.
What do you see if you do "ls -l /dev/*mt*" ?
>From the tmscp.4 man page the major / minor device numbers have the
following meanings:
major device number(s):
raw: 23
block: 12
minor device encoding:
bit:|7 | 6 | 5 | 4 | 3 | 2 | 1 | 0|
---
C C X D D N U U
C = Controller # (max of 4 controllers)
D = Density
N = Norewind on close
U = Unit (drive) number (max of 4 drives per controller)
Re: 2.11 BSD on an 11/93 with an CMD SCSI Module
The saga continues!!! I really don;t remember it being this hard back in the good ole' days but then we had lots of real hardware to work with and systems weren't kludges like what I have in my house. :-) I am currently working with E11 V7.3. My prime interest is in getting a 2.11 system running so I can use tcopy(1) to copy tape images to my one real tape drive. It is SCSI so it is going to be TMSCP on the real system. It is also TMSCP on the E11 system. The problem is I don't seem to be able to do anything on E11 to make 2.11 BSD actually have more than one functional tape. I have tried two TMSCP tapes. I have tried two TMSCP controllers. I have tried one TMSCP tape and various configurations of other tape drives. the result is always the same. I hcae tried building a kernel with all the tapes included. I have tried building kernels with two tapes at a time with different combinations of TMSCP and the others. None of them ever come up with more than one drive. March 26 05:11:44 init: configure system hk ? csr 177440 vector 210 skipped: No autoconfig routines. ht ? csr 172440 vector 224 skipped: No autoconfig routines. ra 0 csr 172150 vector 154 vectorset attached rl ? csr 174400 vector 160 skipped: No autoconfig routines. tm ? csr 172520 vector 224 skipped: No autoconfig routines. tms 0 csr 174500 vector 260 skipped: No CSR. tms 1 csr 164334 vector 774 vectorset attached ts ? csr 172520 vector 224 skipped: No autoconfig routines. xp ? csr 176700 vector 254 skipped: No autoconfig routines. I especially don't understand the "No CSR" for tms Unit 0. If I remove the entry for Unit 1 in the E11.ini file then it recognizes Unit 0. But never both and never any other tape drive. Is there something in 2.11 that I am missing? bill
AGC restoration / was Re: PDP-8 signed overflow detection - Apollo guidace computer
On 2019-Mar-26, at 9:28 AM, Noel Chiappa via cctalk wrote: >> From: Ben Bfranchuk > >> Now they seem to have have found a SCRAPPED Apollo guidance >> computer and am rebuilding the missing pieces. > > Wow. What a great site (and that guy has mad skills, everything from > repairing old Teletypes, through designing boards, to repairing analog > stuff). Just 'wasted' a good chunk of the morning reading back through > it; tons of really neat things (including recovery of the very first > FORTH, along with a lot of Diablo drive - from the Alto - repairs). > > As a shortcut, here: > > http://rescue1130.blogspot.com/2018/11/ > > is the backstory on the AGC; about 1/3 of the way down, in "Restoring an > Apollo Guidance Computer, part V". This is the same AGC restoration that curious marc has been making videos about: https://www.youtube.com/watch?v=2KSahAoOLdU The project and videos were mentioned on the list back in December. There's an interview with Jimmie there (4:13) with more details about the backstory.
Re: George Comstock
On 3/26/19 8:53 AM, Al Kossow via cctalk wrote: > https://www.almanacnews.com/news/2019/03/25/george-comstock-silicon-valley-pioneer-and-portola-valley-civic-leader-dies Thanks for that--I remember George well. --Chuck
Re: PDP-8 signed overflow detection - Apollo guidace computer
> From: Ben Bfranchuk > Now they seem to have have found a SCRAPPED Apollo guidance > computer and am rebuilding the missing pieces. Wow. What a great site (and that guy has mad skills, everything from repairing old Teletypes, through designing boards, to repairing analog stuff). Just 'wasted' a good chunk of the morning reading back through it; tons of really neat things (including recovery of the very first FORTH, along with a lot of Diablo drive - from the Alto - repairs). As a shortcut, here: http://rescue1130.blogspot.com/2018/11/ is the backstory on the AGC; about 1/3 of the way down, in "Restoring an Apollo Guidance Computer, part V". Noel
George Comstock
https://www.almanacnews.com/news/2019/03/25/george-comstock-silicon-valley-pioneer-and-portola-valley-civic-leader-dies
Re: PDP-8 signed overflow detection - Apollo guidace computer
On 3/25/2019 7:27 PM, Charles Dickman via cctalk wrote: On Mon, Mar 25, 2019 at 1:26 AM Kyle Owen via cctalk wrote: All this to fix a ~45 year old bug in Spacewar! where a ship's velocity overflows, causing the ship to "bounce" off of nothing. I would have figured it was bouncing off a chunk of dark matter. Cool stuff! PDP-8 programming always seems a bit like puzzle solving. Kyle Some times the hardware is more fun than the software. I got looking at a BLOG I had forgotten about. First they planned on Emulating a IBM 1130 Later they found a real IBM 1130 in the middle of now where. Now they seem to have have found a SCRAPPED Apollo guidance computer and am rebuilding the missing pieces. http://rescue1130.blogspot.com/ Ben.
