开云体育


Re: GCCLIB

 

I have combined all the source and am trying to build in CMSUSER - there is some logic, I want the build to be kind of standalone (or at least I want to understand all the dependencies). I want to automate all these builds.

Anyway for MKGCCLIB - I get (first page)

ASSEMBLI CMSENTRY
?
ASSEMBLER (XF) DONE
? ? ? ? ? ? ? ?73? ? ? ? ? USING CMSCRAB,R13
IFO188 CMSCRAB IS AN UNDEFINED SYMBOL
IFO217 RELOCATABILITY ERROR NEAR OPERAND COLUMN 8
? ? ? ? ? ? ? ?74? ? ? ? ? LA? ? R2,MAINSTK? ? ?where the GCC stack begins
IFO188 MAINSTK IS AN UNDEFINED SYMBOL
? ? ? ? ? ? ? ?75? ? ? ? ? ST? ? R2,STACKNXT? ? next available spot in stack
IFO188 STACKNXT IS AN UNDEFINED SYMBOL
? ? ? ? ? ? ? ?76? ? ? ? ? ST? ? R13,GCCCRAB? ? address of CRAB
IFO188 GCCCRAB IS AN UNDEFINED SYMBOL
? ? ? ? ? ? ? ?80? ? ? ? ? LA? ? R2,CONIFILE? ? address of console input FILE st
ructure
IFO188 CONIFILE IS AN UNDEFINED SYMBOL
? ? ? ? ? ? ? ?81? ? ? ? ? ST? ? R2,CONFILIN? ? and store it here
IFO188 CONFILIN IS AN UNDEFINED SYMBOL
? ? ? ? ? ? ? 100? ? ? ? ? LA? ? R2,CONOFILE? ? address of console output FILE s
tructure
IFO188 CONOFILE IS AN UNDEFINED SYMBOL
?

Which tells me something important is missing!!

Any hints? :-)

Adrian


Re: IND$FILE Binary of Module

 

On Fri, Jan 3, 2020 at 12:23 AM, Drew Derbyshire wrote:
but when you're working with a 211 K line codebase, one needs to take take care lest installing brighter lights in the chicken coop causes a blackout in the cow barn
Noted!


Re: IND$FILE Binary of Module

 

开云体育

On 1/2/20 2:03 PM, adriansutherland67 wrote:
On Thu, Jan 2, 2020 at 08:01 PM, Drew Derbyshire wrote:
I said Hercules didn't support punches into pipes, whereas Hercules does support printers into pipes.
Is Hercules still supported by its amazing creators?
Hercules has owners. Check
I would have thought (without looking at the code!) that adding support to named pipes or TCP/IP sockets for the puncher or printer would be easy. In fact has anyone just tried to use named pipe, it might just work ...
Outside of the actual emulation code, it's all a simple matter of programming, but when you're working with a 211 K line codebase, one needs to take take care lest installing brighter lights in the chicken coop causes a blackout in the cow barn.?

-ahd-


Re: IND$FILE Binary of Module

 

开云体育

Drew,

?

Well there has been some divergence. So Roger is still putting bug fixes into release 3.xx but no real enhancements. Fish is building a “softdevlabs” fork of V4.xx. I am not sure about the original “4.xx”.

Its probably worth asking on hercules-390 why there isn’t support for a socket punch….

?

Dave

?

From: [email protected] <[email protected]> On Behalf Of adriansutherland67
Sent: 02 January 2020 22:04
To: [email protected]
Subject: Re: [h390-vm] IND$FILE Binary of Module

?

On Thu, Jan 2, 2020 at 08:01 PM, Drew Derbyshire wrote:

I said Hercules didn't support punches into pipes, whereas Hercules does support printers into pipes.

Is Hercules still supported by its amazing creators? I would have thought (without looking at the code!) that adding support to named pipes or TCP/IP sockets for the puncher or printer would be easy. In fact has anyone just tried to use named pipe, it might just work ...


Re: GCCLIB

 

开云体育

Yes, It just changes where the pointer is stored…

?

Dave

?

From: [email protected] <[email protected]> On Behalf Of adriansutherland67
Sent: 02 January 2020 17:12
To: [email protected]
Subject: Re: [h390-vm] GCCLIB

?

I will try now.

In fact the maclib i have only has a tiny change

A


Re: IND$FILE Binary of Module

 

On Thu, Jan 2, 2020 at 05:09 PM, hans.latz wrote:
well, i think i should know. But it's been a long time ago that i wrote this code and i simply don't remember why there is this restriction. Sorry...

Io-buffer has length 512 but 255 is hard coded as a check. I am thinking (considering cr lf and null) we should have it 512 - 3 as the max lrecl. In fact we should make it a define.

That's the file side ... What about the filetransfer? Unlimited or is there a max of 2k (I read something about 2048 somewhere in the docs?)

Thoughts?

Adrian


Re: IND$FILE Binary of Module

 

On Thu, Jan 2, 2020 at 08:01 PM, Drew Derbyshire wrote:
I said Hercules didn't support punches into pipes, whereas Hercules does support printers into pipes.
Is Hercules still supported by its amazing creators? I would have thought (without looking at the code!) that adding support to named pipes or TCP/IP sockets for the puncher or printer would be easy. In fact has anyone just tried to use named pipe, it might just work ...


Re: IND$FILE Binary of Module

 

开云体育

On 1/2/20 11:29 AM, Dave Wade wrote:

I believe that Hercules does support socket readers.

?

?

via NETCAT as well as FISH’s HERCRDR and of course the SIXPACK has VMARC (not sure if its on the

?

I know Hercules supports socket readers.? I said Hercules didn't support punches into pipes, whereas Hercules does support printers into pipes.? A punch -> pipe -> socket reader setup would allow automated RSCS-like forwarding between systems. (Nobody said I wasn't a wee bit twisted.)

<plug>
vmsubmit.py exploits Hercules socket readers, has a fairly decent (read: idiot resistant) UX, and has internal features like EBCDIC support and detecting VMARC files so it doesn't translate them.? vmsubmit.py's primary limitation is being written in Python, which makes it less appealing to Windows users.?

(vmsubmit.py even can act as a SIFT client, which is not a protocol I recommend; that RFC escaped into the wild too young, and it was obviously designed in a more innocent age.? That's for the curious; it's supported on later versions of RSCS.)
</plug>


Re: IND$FILE Binary of Module

 

On Thu, Jan 2, 2020 at 05:09 PM, hans.latz wrote:
well, i think i should know. But it's been a long time ago that i wrote this code and i simply don't remember why there is this restriction. Sorry...
Not to worry ... I can inspect the code myself :-)

Oh ... And sorry I am making my own home grown formatter ... I just wanted something super easy to move c source files back and forwards! ?


Re: IND$FILE Binary of Module

 

开云体育

Sorry, for the premature post been on Facebook and hit cntrl+enter..

Now for proper info…

?

Drew,

I believe that Hercules does support socket readers.

?

?

via NETCAT as well as FISH’s HERCRDR and of course the SIXPACK has VMARC (not sure if it is on the LCM machine)…

… but NETCAT is flagged as MALWARE by many Windows AV utilities…

?

.. the big advantages of these things is that they preserve the record format of the files without any work, which is what you need for some files…

.. VMARC also scores well as it can take input from the reader and output to the punch. This is really useful as disk files in VM/370 are limited to 32k records.

.. because it is not a CMS disk file, data in the SPOOL (so reader, print and punch files) can have more records….

?

Dave

(oh and whilst its not in VM/370 there is also NETDATA format….)

?

?

From: [email protected] <[email protected]> On Behalf Of Drew Derbyshire
Sent: 02 January 2020 19:20
To: [email protected]
Subject: Re: [h390-vm] IND$FILE Binary of Module

?

On 1/2/20 10:28 AM, Dave Wade wrote:

Its on the six pack...
.. and I can't see why we need yet another pack program. ...
There are utilities in the files section to read and write CMS TAPE files,
and DISK DUMP files both of which are native VM archive formats, on
windows/linux

Not to start a debate, but I can see why.? :-)

  • It's not trivial to mount a AWS tape on any system that you don't control the host
  • When you do control the host, it requires accessing not just your VM systems but two Hercules consoles.
  • VMARC data fits through card readers, and by extension host utilities that use TCP/IP to exploit the card reader, like vmsubmit.py in . (It's a pity Hercules card punches can't connect to pipes like printers can. I find this an odd omission.)
  • Issues that apply in special cases (i.e. the LCM+L 4361, which of course I'm rather fond of):
    • It's not trivial to mount a AWS tape on real hardware
    • DISK (DUMP/LOAD) is not entirely compatible between VM/370 R6 and VM/SP 5
  • IBM likes it.
    • It seems it's their preferred format for sharing free VM software over the ASCII friendly Internet.?
      (Pity that most the software itself is for XA and newer!)
    • It too has host utilities to read the format
      (Pity the MacOS version, being 32-bit, is broken on Catalina)

This is not to say the other toys should be tossed (except may the Windows specific one which requires shutting down VM and running a host script!), but it is useful.

-ahd-

p.s. I consider VMARC's command line syntax horrible, but I have to yet to decide what it should be (and if so, what wrapper script to write for it.)

-- 
Drew Derbyshire
?
 Alcohol, Tobacco, Firearms. Anyone remember the chips?


Re: IND$FILE Binary of Module

 

开云体育

Drew,

I believe that Hercules does support socket readers.

?

?

via NETCAT as well as FISH’s HERCRDR and of course the SIXPACK has VMARC (not sure if its on the

?

From: [email protected] <[email protected]> On Behalf Of Drew Derbyshire
Sent: 02 January 2020 19:20
To: [email protected]
Subject: Re: [h390-vm] IND$FILE Binary of Module

?

On 1/2/20 10:28 AM, Dave Wade wrote:

Its on the six pack...
.. and I can't see why we need yet another pack program. ...
There are utilities in the files section to read and write CMS TAPE files,
and DISK DUMP files both of which are native VM archive formats, on
windows/linux

Not to start a debate, but I can see why.? :-)

  • It's not trivial to mount a AWS tape on any system that you don't control the host
  • When you do control the host, it requires accessing not just your VM systems but two Hercules consoles.
  • VMARC data fits through card readers, and by extension host utilities that use TCP/IP to exploit the card reader, like vmsubmit.py in . (It's a pity Hercules card punches can't connect to pipes like printers can. I find this an odd omission.)
  • Issues that apply in special cases (i.e. the LCM+L 4361, which of course I'm rather fond of):
    • It's not trivial to mount a AWS tape on real hardware
    • DISK (DUMP/LOAD) is not entirely compatible between VM/370 R6 and VM/SP 5
  • IBM likes it.
    • It seems it's their preferred format for sharing free VM software over the ASCII friendly Internet.?
      (Pity that most the software itself is for XA and newer!)
    • It too has host utilities to read the format
      (Pity the MacOS version, being 32-bit, is broken on Catalina)

This is not to say the other toys should be tossed (except may the Windows specific one which requires shutting down VM and running a host script!), but it is useful.

-ahd-

p.s. I consider VMARC's command line syntax horrible, but I have to yet to decide what it should be (and if so, what wrapper script to write for it.)

-- 
Drew Derbyshire
?
 Alcohol, Tobacco, Firearms. Anyone remember the chips?


Re: IND$FILE Binary of Module

 


Its on the six pack...
.. and I can't see why we need yet another pack program. ...
There are utilities in the files section to read and write CMS TAPE files,
and DISK DUMP files both of which are native VM archive formats, on
windows/linux
It's just a suggestion :-)

I admit that the first time I came across a VMARC file, I moaned that here
was yet another file format that I needed to find tools to process.

However, I have very much come to like VMARC because it unlike CMS TAPE, it
does not have any filesystem type dependencies or CMS version dependencies
and unlike DISK DUMP, it can archive more than one file and unlike both, the
archive file is smaller than the original file(s). Also (and like DISK DUMP
archives), VMARC archives can trivially be sent through virtual and er, real
card readers and punches and across an NJE network.

I've even come up with a way to build VMARC on MUSIC to make it easier to
exchange files with that platform too (even though MUSIC has a utility to
read CMS TAPE files which seems to be more capable than CMS TAPE LOAD under
VM/370...)

Regards,
Peter Coghlan


Re: IND$FILE Binary of Module

 

开云体育

On 1/2/20 10:28 AM, Dave Wade wrote:
Its on the six pack...
.. and I can't see why we need yet another pack program. ...
There are utilities in the files section to read and write CMS TAPE files,
and DISK DUMP files both of which are native VM archive formats, on
windows/linux

Not to start a debate, but I can see why.? :-)

  • It's not trivial to mount a AWS tape on any system that you don't control the host
  • When you do control the host, it requires accessing not just your VM systems but two Hercules consoles.
  • VMARC data fits through card readers, and by extension host utilities that use TCP/IP to exploit the card reader, like vmsubmit.py in . (It's a pity Hercules card punches can't connect to pipes like printers can. I find this an odd omission.)
  • Issues that apply in special cases (i.e. the LCM+L 4361, which of course I'm rather fond of):
    • It's not trivial to mount a AWS tape on real hardware
    • DISK (DUMP/LOAD) is not entirely compatible between VM/370 R6 and VM/SP 5
  • IBM likes it.
    • It seems it's their preferred format for sharing free VM software over the ASCII friendly Internet.?
      (Pity that most the software itself is for XA and newer!)
    • It too has host utilities to read the format
      (Pity the MacOS version, being 32-bit, is broken on Catalina)

This is not to say the other toys should be tossed (except may the Windows specific one which requires shutting down VM and running a host script!), but it is useful.

-ahd-

p.s. I consider VMARC's command line syntax horrible, but I have to yet to decide what it should be (and if so, what wrapper script to write for it.)

-- 
Drew Derbyshire

 Alcohol, Tobacco, Firearms. Anyone remember the chips?


Re: IND$FILE Binary of Module

 

Its on the six pack...
.. and I can't see why we need yet another pack program. ...
There are utilities in the files section to read and write CMS TAPE files,
and DISK DUMP files both of which are native VM archive formats, on
windows/linux

Dave

-----Original Message-----
From: [email protected] <[email protected]> On Behalf Of Peter
Coghlan
Sent: 02 January 2020 17:59
To: [email protected]
Subject: Re: [h390-vm] IND$FILE Binary of Module


I get "Trans99 - LRECL > 255 unsupported: file transfer canceled" when
trying to receive a module file from the host.

Is there anything I can do or is this a known restriction?
Maybe you could get around the restriction by putting the module in a
VMARC archive file? These are 80 byte card images so there should be no
problems transferring them.

If you need to build the VMARC archive on the host, there is a C
implemention of VMARC called VMA which should be compilable on most
platforms.

Regards,
Peter Coghlan.


Re: IND$FILE Binary of Module

 


I get "Trans99 - LRECL > 255 unsupported: file transfer canceled" when
trying to receive a module file from the host.

Is there anything I can do or is this a known restriction?
Maybe you could get around the restriction by putting the module in a VMARC
archive file? These are 80 byte card images so there should be no problems
transferring them.

If you need to build the VMARC archive on the host, there is a C implemention
of VMARC called VMA which should be compilable on most platforms.

Regards,
Peter Coghlan.


Re: GCCLIB

 

I will try now.

In fact the maclib i have only has a tiny change

A


Re: IND$FILE Binary of Module

 

well, i think i should know. But it's been a long time ago that i wrote this code and i simply don't remember why there is this restriction. Sorry...


Re: GCCLIB

 

Hello,

try the file "GCCLIB-without-DOSTRANS-for-VM370-SixPack-1.2.zip" in the files section of this group.

The sources are in the tape "GCCLIB-without-DOSTRANS-modified-src.aws" in the ZIP. (there may be a missing source there, i believe to remember something about it... however, the missing file may be reconstructed by patching the original from the github-repo in analogy to the others...)

The other tapes contain the new binary GCC libraries and the MECAFF binaries using this new libraries. I don't know if you will need these, it depends on what versions were installed in your sixpack version (1.3?).

Good luck, Hans


GCCLIB

 

So I (with Dave's help) put the 0.6 source here -

However the "without DOSTRAN" tape does not seem to have any source - just 4 files

GCCLIB?? MACLIB?? E1? F??? 80??? 436???? 44? 2012-05-08 17:30? CMS193
GCCLIB?? TEXT???? E1? F??? 80?? 1623??? 163? 2012-05-08 17:34? CMS193
GCCLIB?? TXTLIB?? E1? F??? 80?? 1218??? 122? 2012-05-08 17:34? CMS193
GCCLIB_S TXTLIB?? E1? F??? 80?? 1640??? 164? 2012-05-08 17:34? CMS193

Which I think are intermediate build files ... (am I right?)

Does anyone know if/where the source exists and/or what was the needed change. There seems to be lots and lots of assembler files that reference DOSTRAn :-(

Thoughts anyone?!

Adrian


IND$FILE Binary of Module

 

Hello

I get "Trans99 - LRECL > 255 unsupported: file transfer canceled" when trying to receive a module file from the host.

Is there anything I can do or is this a known restriction??

In the end I used a tape - so no issue, just interested :-)

A