Re: Check out Ternary "flip-flap-flop"

[Martin Packer]

To my mind quad made more sense than tri. But as others have said the game 
moved on.

Cheers, Martin

Martin Packer,
zChampion, Principal Systems Investigator,
Worldwide Banking Center of Excellence, IBM

+44-7802-245-584

email: martin_pac...@uk.ibm.com

Twitter / Facebook IDs: MartinPacker
Blog: 
https://www.ibm.com/developerworks/mydeveloperworks/blogs/MartinPacker
Unless stated otherwise above:
IBM United Kingdom Limited - Registered in England and Wales with number 
741598. 
Registered office: PO Box 41, North Harbour, Portsmouth, Hampshire PO6 3AU

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

Re: Something to Think About - Optimal PDS Blocking

[nitz-...@gmx.net]

>   SAdump never writes short blocks.  When it wants to write a block
> which is not full, it pads the block with dummy records (which IPCS
> knows to ignore).
Then I don't understand why the utility used produced a copy significantly 
smaller than the full dump. Once it was initialized on the full 63000cyls I 
could see that sadump had successfully completed, so what chance of an I/O 
error is there?
Any utility would not care about the content of a 'short' block padded to full 
length.

The statistics at the end of the sadump messages were clearly showing that each 
of the volumes was between 30% and 35% full (they are mod27 and I forgot the 
real usage number). The copy was terminated after the third volume, is my guess.

>   SAdump does not support data sets whose stripe count is greater than 1.
My bad for using the wrong terminology.

Thanks for reminding us about the reasons for using copydump.

Barbara

ps: IBM was sent this sadump, too. It is 66181,077,724 and has a 'nice' 
SYSIEFSD Q10 deadlock intermixed with PDSE latch contention. Since I never saw 
the full sadump, I was unable to determine why Q10 was held and not released, 
effectively preventing SMSPDSE1 restart.

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

Re: "hexadecimal"?

[Gerhard Postpischil]

On 12/8/2013 3:54 PM, Scott Ford wrote:

Circuits work with 1s and 0s and Assembler, Cobol, etc translate into
machine code which is binary


Computers, depending on architecture, have components that process 
signals in parallel. For example, the old I/O hardware sent 8-bits (plus 
parity) concurrently. There are components that send 32 and 64 bits in 
parallel. For these I could make the case that they're not processing 
binary, but aggregates.


Gerhard Postpischil
Bradford, Vermont

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

English article on the setun machine, with pictures.

[David Boyes]

http://ternary.3neko.ru/setun.html

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

Technical details on the setun-70 trinary processor

[David Boyes]

http://ternary.3neko.ru/publications/setun70.pdf

In Russian, but google translate does a fair job. 

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

Re: Check out Ternary "flip-flap-flop"

[Harry Wahl]

Quantum computing is here. These people are already selling quantum mainframes 
with humungous capabilities to the government and industry (e.g. Google, NASA, 
Lockheed Martin):http://www.dwavesys.com/Nothing wrong with revisiting what we 
learned in school. The D-Wave people should give going old-school a shot and 
take a look at updating the Shannon-Hartley law we all learned in school. Maybe 
then they could get past their quantum decoherence deficiencies and quell 
doubters' objections.Unimagined in 1970s data processing is the quantum bit, 
the qubit, a bit with an additional "superposition state" (basically true and 
false at the same time). Would a qubit be considered binary or ternary? If 
ternary, it would have asymmetric values: two states plus a superposition of 
the other two states.The concepts of binary, ternary, decimal and hexadecimal 
are unified because they all always represent a whole definite integer value 
when considered outside of the context of any particular computing machine. 
Extended numerical representations such as twos-complement, decimal points, 
floating point, etc. are created at the whim of a machine's designers and 
programmers (firmware, OS and application).But, now qubits supersede classical 
data processing numeric representation with quantum, pending value, numeric 
representation and create the need for a whole new vocabulary. Schrödinger's 
cat demands a fresher pidgin (pun intended), so to speak.  With regard to the 
usage of the word "hexadecimal," I think you're arguing the evolving conceptual 
semantics of a polyseme. No winning when you debate something like that, a 
little dialectic however, goes a long way.Harry
> Date: Sun, 8 Dec 2013 21:36:20 -0600
> From: paulgboul...@aim.com
> Subject: Re: Check out Ternary "flip-flap-flop"
> To: IBM-MAIN@LISTSERV.UA.EDU
> 
> On Sun, 8 Dec 2013 20:58:27 -0600, Joel C. Ewing wrote:
> 
> >On 12/08/2013 05:23 PM, Ed Finnell wrote:
> >> _Ternary  "flip-flap-flop"_
> >> (http://www.goldenmuseum.com/1411FlipFlap_engl.html)
> >>
> >> This is pretty old stuff. Think the advantages were canceled out by
> >> 'indeterminate' states. Some of the new quantum stuff has similar 
> >> possibilities.
> >>
> >
> >The Soviet Union designed and built computers based on ternary logic
> >(Setun, Setun-70) for several decades starting in 1958.  They were less
> >expensive to produce and more electrically efficient to run than
> >binary-oriented hardware of comparable computational power, and were
> >actually an astute choice when the Soviet Union was trying to maximize
> >results with limited resources.
> > 
> I wonder what the states were, electronically?  Positive, ground, and 
> negative?
> And whether it was possible to build a ternary flipflapflop with fewer
> tubes/transistors than a binary flipflop.
> 
> Of course, "tri-state" bus transceivers are old stuff: 0, 1, and 
> high-impedance
> for uncommitted -- "let someone else decide."  I believe the original S/360 
> bus
> and tag cables didn't do that; rather they used NPN (I think) emitter 
> followers
> with resistive pulldowns in the terminators.
> 
> -- gil
> 
> --
> For IBM-MAIN subscribe / signoff / archive access instructions,
> send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
  
--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

Re: Something to Think About - Optimal PDS Blocking

[Jim Mulder]

> Think standalone dump written striped to  - say - 5 volumes. Each 
> volume has a data set in format FBS, but only one of the volumes can
> have a short record. SAdump knows that, and IPCS knows it, too. The 
> utilities don't. So assume that you took a complete sadump to 5 
> volumes and the sort record happens to be on the first volume. Then 
> you use a utility (ICEGENER is my favourite) to copy somewhere else.
> You end up with a severely truncated sadump. One fifth, to be exact.
> IPCS will read the truncated dump to the best of its abilities, but 
> you will get all kinds of 'storage not available' warnings when 
> looking at the dump.
> 
> Last time a customer sent me an sadump, it had 27000cyl. I got all 
> kinds of warnings and got lucky in that the sadump messages were 
> clearly truncated and didn't show the 'successfully finished' 
> message. It turned out that the wrong utility was used for copying, 
> and the actual dump had 63000 cyls. Visible when IPCS COPYDUMP was 
> used for copying. IPCS knows that a striped sadump can have the 
> short record "earlier".

  SAdump never writes short blocks.  When it wants to write a block
which is not full, it pads the block with dummy records (which IPCS
knows to ignore).

> I am somewhat at a loss to understand how some of the problems you are 
detailing happened.  The only way it could have would be with an ill 
behaving user written program or process.

>If I remember this correctly (and I am on shaky ground here), sadump 
writes a 'special striping', understood fully only by IPCS. Mind you, 
sadump is a standalone application that does not use standard >z/OS 
services because they are not available. Either here in old posts or even 
somewhere in the docs the behaviour I detailed is described, including the 
warning NOT to use standard utilities when >copying a 'striped' (multi 
DASD volume) sadump. Suffice it to say, the full 63000cyls (from my recent 
customer) could only be copied when the customer used IPCS COPYDUMP. I was 
not told how they copied >it when they sent me the 27000cyls.

>Just thought that a reminder about sadump was in order. After all, who 
would want to take an sadump only to be told by support that the necessary 
data are not in the dump?

  SAdump does not support data sets whose stripe count is greater than 1.
For multivolume data sets, sadump writes to all of the volumes 
concurrently,
but not in the same way that DFSMS does for a striped data set.
 
  We recommend that a multivolume sadump data set should always be 
copied using COPYDUMP before doing further processing, for 2 reasons:

1.  COPYDUMP OPENs all of the volumes concurrently, and merges them
  back into an approximation of the logical dumping order.  This 
  reduces the number of storage map entries which IPCS will need to 
  create in the dump directory, and improves IPCS performance.

2.  If the sadump failed to complete, in a manner which did not
   allow it to close the data set extent on each volume, IPCS
   dump initialization or a copying utility will most likely
   terminate when it encounters an error at the end of the data
   dumped on the first volume, or copy residual data from 
   a prior dump.  COPYDUMP will copy all of the relevant data 
   from each volume.

>Sorry Barbara, I forgot about SADUMP.  I wonder, does the same problem 
exist with a console dump or a SYSMDUMP? 

  SVC Dump (except when DCB= is specified on the SDUMP macro),
SYSMDUMP, and IEATDUMP all use BSAM to write to dump data sets.  So 
they support anything that BSAM supports (like DFSMS striped
data sets, and zHPF).  They is generally no requirement to use COPYDUMP
for these types of dumps, unless you need to process a data set which
was not closed because the system crashed while it was being written.
However, COPYDUMP may still be beneficial, if you want to make use
of the INIT or INITAPPEND options.  INITAPPEND is new in z/OS 2.1.


Jim Mulder   z/OS System Test   IBM Corp.  Poughkeepsie,  NY

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

Re: Check out Ternary "flip-flap-flop"

[Paul Gilmartin]

On Sun, 8 Dec 2013 20:58:27 -0600, Joel C. Ewing wrote:

>On 12/08/2013 05:23 PM, Ed Finnell wrote:
>> _Ternary  "flip-flap-flop"_
>> (http://www.goldenmuseum.com/1411FlipFlap_engl.html)
>>
>> This is pretty old stuff. Think the advantages were canceled out by
>> 'indeterminate' states. Some of the new quantum stuff has similar 
>> possibilities.
>>
>
>The Soviet Union designed and built computers based on ternary logic
>(Setun, Setun-70) for several decades starting in 1958.  They were less
>expensive to produce and more electrically efficient to run than
>binary-oriented hardware of comparable computational power, and were
>actually an astute choice when the Soviet Union was trying to maximize
>results with limited resources.
> 
I wonder what the states were, electronically?  Positive, ground, and negative?
And whether it was possible to build a ternary flipflapflop with fewer
tubes/transistors than a binary flipflop.

Of course, "tri-state" bus transceivers are old stuff: 0, 1, and high-impedance
for uncommitted -- "let someone else decide."  I believe the original S/360 bus
and tag cables didn't do that; rather they used NPN (I think) emitter followers
with resistive pulldowns in the terminators.

-- gil

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

Re: Check out Ternary "flip-flap-flop"

[Joel C. Ewing]

On 12/08/2013 05:23 PM, Ed Finnell wrote:
> _Ternary  "flip-flap-flop"_ 
> (http://www.goldenmuseum.com/1411FlipFlap_engl.html)  
>  
> This is pretty old stuff. Think the advantages were canceled out by  
> 'indeterminate' states. Some of the new quantum stuff has similar 
> possibilities. 
> 

The Soviet Union designed and built computers based on ternary logic
(Setun, Setun-70) for several decades starting in 1958.  They were less
expensive to produce and more electrically efficient to run than
binary-oriented hardware of comparable computational power, and were
actually an astute choice when the Soviet Union was trying to maximize
results with limited resources.

I suspect those ternary machines were eventually discontinued because
they were just overwhelmed by the success of S/360 and other binary,
byte oriented architectures, including S/360-compatible clones produced
in the Soviet Union as ES EVM starting in 1972. The dramatic cost
reductions from mass production of binary machines and specialized
binary Integrated Circuits probably erased the hardware cost advantage
of ternary machines, and there must also have been significant software
and manpower costs involved in trying to exchange data between
incompatible base-3 and base-2 data formats and from having to design
unique software for a ternary architecture.

While the ternary architecture does have a certain elegance, history has
numerous instances of elegance losing out to mass acceptance of
competing solutions.


-- 
Joel C. Ewing,Bentonville, AR   jcew...@acm.org 

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

Re: "hexadecimal"?

[Paul Gilmartin]

On Sun, 8 Dec 2013 20:01:20 -0600, Joel C. Ewing wrote:
>
>If the real requirement is that the parameter address must point to a
>location containing what the PoOp describes as a half-word binary value,
>then the manual should state precisely that.
>
Thanks for bringing this thread back to my original concern when I
started it (I never imagined! ...), however interesting the digressions
to HFP and ternary logic may be, and for stating that concern more
clearly than I did.

There should also be mention of the allowable range of values:
0

Re: "hexadecimal"?

[Joel C. Ewing]

On 12/08/2013 12:24 PM, Gerhard Postpischil wrote:
> On 12/8/2013 12:52 PM, John McKown wrote:
>> Agree. I have not checked, but will bat this is really a full word
>> integer
>> binary number.
> 
> Minor nit - DD= specifies a halfword.
> 
> 
> Gerhard Postpischil
> Bradford, Vermont

Which perfectly makes the point that a manual which refers to this
parameter value as "hexadecimal", while perhaps trivially true, is also
uselessly imprecise.  This should not be allowed in a well-edited manual.

The z-Arch PoOp describes many, many data formats used in the many
contexts of z/OS architecture, all of which data is ultimately stored in
some combination of bytes -- and any byte content may be represented or
described using either binary or hex notation.  So to say a parameter
may point to a hexadecimal value, while perhaps true, is totally
ambiguous.

If the real requirement is that the parameter address must point to a
location containing what the PoOp describes as a half-word binary value,
then the manual should state precisely that.

I have seen some functional routines that are clever enough to
automatically detect the difference between multiple parameter formats,
say between a binary value versus an EBCDIC character value, and behave
accordingly, but this is only plausible in specific instances. Typically
only one specific data format is expected, that format must be supplied,
and the documentation should reflect that requirement.

-- 
Joel C. Ewing,Bentonville, AR   jcew...@acm.org 

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

Re: Something to Think About - Optimal PDS Blocking

[Anne & Lynn Wheeler]

re:
http://www.garlic.com/~lynn/2013o.html#5 Something to Think About - Optimal PDS 
Blocking

original RAID patent from 1978 was by somebody in the san jose disk
group 
http://en.wikipedia.org/wiki/RAID

... whom I actually worked with some when they let me play disk engineer
in bldgs 14&15
http://www.garlic.com/~lynn/subtopic.html#disk

was originally added to s/38. s/38 is periodically referred to as vastly
simplified Future System implementation. 
http://www.garlic.com/~lynn/submain.html#futuresys

One of the simplifications was dynamic "scatter" block allocation across
all disks in the infrastructure ... treating all disks in the
configuration as a single resource. as a result the whole infrastructure
had to be backed up as single entity and restored as single entity.
common failure mode of the period was single disk failure ... for s/38
required stopping the whole system, replacing the failed disk and doing
a complete system restore ... which could be a 24hr event. RAID was used
to mask single disk failure and the associated major recovery event.

majority RAID implementations these days are at the hardware controller
level ... so single disk slices wouldn't be available at the software
level.

RAID has been used for both availability (masking disk failure) and
single thread throughput. One of the issues was single thread throughput
was degradation for large DBMS infrastructures. More recent RAID options
have tried to address both availability as well as multi-thread random
access.

over the years, part of commoditizing of industry standard disks is
driving MTBF from something like 80k hrs to 800k hrs (warrenty costs
across enormous number of disks).

MTBF has nearly doubled again. the major cloud operators do in-depth
studies of the issues involving component failures ... as part of
building their own servers (past news that chip manufacturers ship more
server chips directly to large cloud operators than to brand name server
vendors) and openly publish the information (something analogous to
mainframe industry group that published customer EREP information during
the heyday of mainframe clone processors).
http://en.wikipedia.org/wiki/Hard_disk_drive_failure

from above:

The mean time between failures (MTBF) of SATA drives is usually
specified to be about 1.2 million hours (some drives such as Western
Digital Raptor have rated 1.4 million hours MTBF),[17] while SAS/FC
drives are rated for upwards of 1.6 million hours.[18]

... and

A 2007 study published by Google suggested very little correlation
between failure rates and either high temperature or activity
level. Indeed, the Google study indicated that "lower temperatures are
associated with higher failure rates"

... snip ...

2007 google disk failure report
http://storagemojo.com/2007/02/19/googles-disk-failure-experience/

-- 
virtualization experience starting Jan1968, online at home since Mar1970

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

Re: "hexadecimal"?

[Anne & Lynn Wheeler]

dlc@gmail.com (David L. Craig) writes:
> GINYF when it doesn't equate trinary with trenary, a term
> I had forgotten (I'm really trying to avoid turning this
> thread into a career).  At least it wasn't just a rumor.
> +1 and thanks for causing me to be reminded of Minerva--
> perhaps someday Google will attain such capability.

for a little drift ... old (sql) post about 3-value logic and handling
of nulls. about same time i was involved in original relational
implementation ... some past posts
http://www.garlic.com/~lynn/submain.html#systemr

... i also dragged in helping with a similar but different kind of
relational implementation ... that had a different way of convention for
3-value logic (& nulls/unknowns).
http://www.garlic.com/~lynn/2003g.html#40 How to cope with missing values - 
NULLS?

part of the issue with sql 3-value logic and nulls/unknowns was that the
results could be the opposite of what people assumed.

3-value logic reference (also trivalent, ternary, trinary, trilean)
http://en.wikipedia.org/wiki/Three-valued_logic

includes section
http://en.wikipedia.org/wiki/Three-valued_logic#Application_in_SQL
and references null (sql)
http://en.wikipedia.org/wiki/Null_%28SQL%29

as well as references ternary computer
http://en.wikipedia.org/wiki/Ternary_computer

in the years since system/r & this other kind of relational ... i've
re-implemented versions a number of times from scratch ... i use it for
a number of things including ietf rfc (internet standards) index ... and
merged glossaries and taxonomies
http://www.garlic.com/~lynn/index.html

other posts mentioning 3-value logic
http://www.garlic.com/~lynn/2004l.html#75 NULL
http://www.garlic.com/~lynn/2005.html#15 Amusing acronym
http://www.garlic.com/~lynn/2005i.html#35 The Worth of Verisign's Brand
http://www.garlic.com/~lynn/2005m.html#19 Implementation of boolean types
http://www.garlic.com/~lynn/2005t.html#23 So what's null then if it's not 
nothing?
http://www.garlic.com/~lynn/2005t.html#33 What ever happened to Tandem and 
NonStop OS ?
http://www.garlic.com/~lynn/2006e.html#34 CJ Date on Missing Information
http://www.garlic.com/~lynn/2006q.html#22 3 value logic. Why is SQL so special?
http://www.garlic.com/~lynn/2006s.html#27 Why these original FORTRAN quirks?
http://www.garlic.com/~lynn/2006x.html#21 "The Elements of Programming Style"
http://www.garlic.com/~lynn/2006x.html#30 "The Elements of Programming Style"
http://www.garlic.com/~lynn/2006y.html#1 "The Elements of Programming Style"
http://www.garlic.com/~lynn/2009c.html#34 Is the Relational Database Doomed?
http://www.garlic.com/~lynn/2009g.html#32 Old-school programming techniques you 
probably don't miss
http://www.garlic.com/~lynn/2009r.html#65 You know you've been Lisp hacking to 
long when
http://www.garlic.com/~lynn/2012o.html#8 Initial ideas (orientation) constrain 
creativity

-- 
virtualization experience starting Jan1968, online at home since Mar1970

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

Re: "hexadecimal"?

[David L. Craig]

On 13Dec08:1717-0600, Paul Gilmartin wrote:

> On Sun, 8 Dec 2013 18:01:01 -0500, David L. Craig wrote:
> >
> >Boolean algebra is a mathematics based upon variables
> >that can take on only two possible values.  IIRC, the
> >Soviets were rumored to have probed the attributes
> >of trinary-state computing in the '60s or '70s and
> >discontinued the research, but I have not located a
> >source to confirm even the rumor.  Maybe Lynn Wheeler
> >has it in his archives.
> > 
> For a start:
> 
> http://en.wikipedia.org/wiki/Ternary_computer
> 
> GIYF,

GINYF when it doesn't equate trinary with trenary, a term
I had forgotten (I'm really trying to avoid turning this
thread into a career).  At least it wasn't just a rumor.
+1 and thanks for causing me to be reminded of Minerva--
perhaps someday Google will attain such capability.
-- 

May the LORD God bless you exceedingly abundantly!

Dave_Craig__
"So the universe is not quite as you thought it was.
 You'd better rearrange your beliefs, then.
 Because you certainly can't rearrange the universe."
__--from_Nightfall_by_Asimov/Silverberg_

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

Re: "hexadecimal"?

[Scott Ford]

David,

For one I am not trying to bust your chops as we say up here in the Northeast. 
But remember we are all products of school and experience. Because I haven't 
seen it doesn't mean it doesn't exist. But I have the experience of ppl making 
off the wall comments with no knowledge or experience to backup their comments, 
sales talk , whatever. I am a techie I know what I see and experience.

Scott ford
www.identityforge.com
from my IPAD

'Infinite wisdom through infinite means'


> On Dec 8, 2013, at 6:25 PM, "David L. Craig"  wrote:
> 
>> On 13Dec08:1449-0800, Scott Ford wrote:
>> 
>> I don't know what your profession is , but in 40+
>> yrs and about 20+ shops, many in the NYC area and very
>> very aggressive with their applications, including
>> z/OS, VM and VSE , I never ever had to worry about it
>> ..Give me an example of such  a concern ?
> 
> Back in the early '70s when I was a mere OS/MVT/HASP
> operator, I developed an assembler program to process
> FORTRAN statements containing literal non-integers.
> Converting C'123.45' into its HFP equivalent or back
> may not be something you ever needed to develop.
> 
>> BTW I have been commercially developing software
>> for 10+ years the other years with some of the world's
>> biggest companies..
> 
> I'd rather not run the risk of violating listserv policy
> by posting details, but you should be able to find my
> career details online if it really matters to you.
> 
>> I am trying to understand your point . I see fluff,
>> where's the example, real world, maybe NASA or JPL or
>> some government IBM software installation, I will give
>> you that ...
> 
> The NASA Goddard GRTS system certainly had such concerns
> but I was only responsible for the virtual machine in
> which it ran.
> -- 
> 
> May the LORD God bless you exceedingly abundantly!
> 
> Dave_Craig__
> "So the universe is not quite as you thought it was.
> You'd better rearrange your beliefs, then.
> Because you certainly can't rearrange the universe."
> __--from_Nightfall_by_Asimov/Silverberg_
> 
> --
> For IBM-MAIN subscribe / signoff / archive access instructions,
> send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

Re: "hexadecimal"?

[David L. Craig]

On 13Dec08:1449-0800, Scott Ford wrote:

> I don't know what your profession is , but in 40+
> yrs and about 20+ shops, many in the NYC area and very
> very aggressive with their applications, including
> z/OS, VM and VSE , I never ever had to worry about it
> ..Give me an example of such  a concern ?

Back in the early '70s when I was a mere OS/MVT/HASP
operator, I developed an assembler program to process
FORTRAN statements containing literal non-integers.
Converting C'123.45' into its HFP equivalent or back
may not be something you ever needed to develop.

> BTW I have been commercially developing software
> for 10+ years the other years with some of the world's
> biggest companies..

I'd rather not run the risk of violating listserv policy
by posting details, but you should be able to find my
career details online if it really matters to you.

> I am trying to understand your point . I see fluff,
> where's the example, real world, maybe NASA or JPL or
> some government IBM software installation, I will give
> you that ...

The NASA Goddard GRTS system certainly had such concerns
but I was only responsible for the virtual machine in
which it ran.
-- 

May the LORD God bless you exceedingly abundantly!

Dave_Craig__
"So the universe is not quite as you thought it was.
 You'd better rearrange your beliefs, then.
 Because you certainly can't rearrange the universe."
__--from_Nightfall_by_Asimov/Silverberg_

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

Check out Ternary "flip-flap-flop"

[Ed Finnell]

_Ternary  "flip-flap-flop"_ 
(http://www.goldenmuseum.com/1411FlipFlap_engl.html)  
 
This is pretty old stuff. Think the advantages were canceled out by  
'indeterminate' states. Some of the new quantum stuff has similar 
possibilities. 

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

Re: "hexadecimal"?

[Paul Gilmartin]

On Sun, 8 Dec 2013 18:01:01 -0500, David L. Craig wrote:
>
>Boolean algebra is a mathematics based upon variables
>that can take on only two possible values.  IIRC, the
>Soviets were rumored to have probed the attributes
>of trinary-state computing in the '60s or '70s and
>discontinued the research, but I have not located a
>source to confirm even the rumor.  Maybe Lynn Wheeler
>has it in his archives.
> 
For a start:

http://en.wikipedia.org/wiki/Ternary_computer

GIYF,
gil

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

Re: "hexadecimal"?

[David L. Craig]

On 13Dec08:1647-0500, Scott Ford wrote:

> > On Dec 8, 2013, at 4:39 PM, John Gilmore  wrote:
> > 
> > The system/360 and its mainframe successors are largely binary
> > machines.  In particular, addressing is entirely binary.  They can,
> > however, do both decimal and hexadecimal arithmetic in some situations
> > too.
> > 
> > All this can be reduced to boolean algebra and, finally, to assorted
> > configurations of NORs or NANDs: binary arithmetic can indeed be made
> > to disappear, replaced by boolean algebra.  These reductions are
> > theoretically important; but, as I have had occasion to point out here
> > before, you can cut yourself with Ockham's razor.
> > 
> > In most cases it is appropriate, qua programmer although not perhaps
> > qua circuit designer, to think of the z/Architecture HFP instructions
> > as doing hexadecimal arithmetic and of the decimal and DFP
> > instructions as doing decimal arithmetic
> 
> I understand the z/arch somewhat, I stepped out
> of the hardware arena many moons ago, my father
> was involved with hardware at Unisys. I was taught
> every converted down to its lowest common factor,
> binary..are you saying otherwise ? You comments I
> understand somewhat but I am a practical guy ..so
> help me understand here

Boolean algebra is a mathematics based upon variables
that can take on only two possible values.  IIRC, the
Soviets were rumored to have probed the attributes
of trinary-state computing in the '60s or '70s and
discontinued the research, but I have not located a
source to confirm even the rumor.  Maybe Lynn Wheeler
has it in his archives.
-- 

May the LORD God bless you exceedingly abundantly!

Dave_Craig__
"So the universe is not quite as you thought it was.
 You'd better rearrange your beliefs, then.
 Because you certainly can't rearrange the universe."
__--from_Nightfall_by_Asimov/Silverberg_

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

Re: "hexadecimal"?

[Scott Ford]

David:
 
I don't know what your profession is , but in 40+ yrs and about 20+ shops, many 
in the NYC area and very very aggressive with their applications, including 
z/OS, VM and VSE , I never ever had to worry about it ..Give me an example of 
such  a concern ?
 
BTW I have been commercially developing software for 10+ years the other years 
with some of the world's biggest companies..
 
I am trying to understand your point . I see fluff, where's the example, real 
world, maybe NASA or JPL or some government IBM software installation, I will 
give you that ...

Scott J Ford
Software Engineer
http://www.identityforge.com/
 



On Sunday, December 8, 2013 5:41 PM, David L. Craig  wrote:
  
On 13Dec08:0942-0800, Charles Mills wrote:

> Hex is not a kind of data. It is a convenient way of
> representing data. X'F1' is a clearer image in most
> cases than 0001 or 241. All data is potentially
> hex; that is, is representable in hex. That's the
> beauty of hex.

This is the paragraph I took (and still take) exception
to.  Before you assert the S/360 was digitally 100%
binary you need to review all the schematics to ensure
there aren't any non-binary components; e.g., three-state
logic gates.  Such is a question IBM Fellow Amdahl or any
competent CE can address.  But this is an area that even
the microcode could be ignorant of.  A BAL programmer had
better deal with what the POPs document.  If you program
HFP instructions as if your are dealing with IEEE hardware
your results are likely to become problematic at some
point.  At minimum, HFP documents a hexadecimal "kind of
data" and I rest my case.
-- 

May the LORD God bless you exceedingly abundantly!

Dave_Craig__
"So the universe is not quite as you thought it was.
You'd better rearrange your beliefs, then.
Because you certainly can't rearrange the universe."
__--from_Nightfall_by_Asimov/Silverberg_


--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

Re: "hexadecimal"?

[David L. Craig]

On 13Dec08:0942-0800, Charles Mills wrote:

> Hex is not a kind of data. It is a convenient way of
> representing data. X'F1' is a clearer image in most
> cases than 0001 or 241. All data is potentially
> hex; that is, is representable in hex. That's the
> beauty of hex.

This is the paragraph I took (and still take) exception
to.  Before you assert the S/360 was digitally 100%
binary you need to review all the schematics to ensure
there aren't any non-binary components; e.g., three-state
logic gates.  Such is a question IBM Fellow Amdahl or any
competent CE can address.  But this is an area that even
the microcode could be ignorant of.  A BAL programmer had
better deal with what the POPs document.  If you program
HFP instructions as if your are dealing with IEEE hardware
your results are likely to become problematic at some
point.  At minimum, HFP documents a hexadecimal "kind of
data" and I rest my case.
-- 

May the LORD God bless you exceedingly abundantly!

Dave_Craig__
"So the universe is not quite as you thought it was.
 You'd better rearrange your beliefs, then.
 Because you certainly can't rearrange the universe."
__--from_Nightfall_by_Asimov/Silverberg_

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

Re: "hexadecimal"?

[Scott Ford]

John,

I understand the z/arch somewhat, I stepped out of the hardware arena many 
moons ago, my father was involved with hardware at Unisys. I was taught every 
converted down to its lowest common factor, binary..are you saying otherwise ? 
You comments I understand somewhat but I am a practical guy ..so help me 
understand here

Scott ford
www.identityforge.com
from my IPAD

'Infinite wisdom through infinite means'


> On Dec 8, 2013, at 4:39 PM, John Gilmore  wrote:
> 
> The system/360 and its mainframe successors are largely binary
> machines.  In particular, addressing is entirely binary.  They can,
> however, do both decimal and hexadecimal arithmetic in some situations
> too.
> 
> All this can be reduced to boolean algebra and, finally, to assorted
> configurations of NORs or NANDs: binary arithmetic can indeed be made
> to disappear, replaced by boolean algebra.  These reductions are
> theoretically important; but, as I have had occasion to point out here
> before, you can cut yourself with Ockham's razor.
> 
> In most cases it is appropriate, qua programmer although not perhaps
> qua circuit designer, to think of the z/Architecture HFP instructions
> as doing hexadecimal arithmetic and of the decimal and DFP
> instructions as doing decimal arithmetic
> 
> John Gilmore, Ashland, MA 01721 - USA
> 
> --
> For IBM-MAIN subscribe / signoff / archive access instructions,
> send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

Re: "hexadecimal"?

[John Gilmore]

The system/360 and its mainframe successors are largely binary
machines.  In particular, addressing is entirely binary.  They can,
however, do both decimal and hexadecimal arithmetic in some situations
too.

All this can be reduced to boolean algebra and, finally, to assorted
configurations of NORs or NANDs: binary arithmetic can indeed be made
to disappear, replaced by boolean algebra.  These reductions are
theoretically important; but, as I have had occasion to point out here
before, you can cut yourself with Ockham's razor.

In most cases it is appropriate, qua programmer although not perhaps
qua circuit designer, to think of the z/Architecture HFP instructions
as doing hexadecimal arithmetic and of the decimal and DFP
instructions as doing decimal arithmetic

John Gilmore, Ashland, MA 01721 - USA

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

Re: "hexadecimal"?

[John McKown]

Ah, my mistake. Digital (binary) computers do deal in 0s and 1s. Analog
computers did not, but that's all I know about them.

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

Re: "hexadecimal"?

[Scott Ford]

John,

I agree but you can't compare apples and oranges ...right ? I don't know that 
much about Amdahl nor am I trying give the Gentleman a hard time. But 
hexadecimal, excuse me? 

Circuits work with 1s and 0s and Assembler, Cobol, etc translate into machine 
code which is binary .maybe I am over - reacting , if so I am sorry

Scott ford
www.identityforge.com
from my IPAD

'Infinite wisdom through infinite means'


> On Dec 8, 2013, at 3:05 PM, John McKown  wrote:
> 
> Amdahl was the main architect of the S/360. So I guess we can mention him
> on occasion.
> On Dec 8, 2013 1:41 PM, "Scott Ford"  wrote:
> 
>> Btw this isn't a Amdahl forum, it's IBM...
>> 
>> Scott ford
>> www.identityforge.com
>> from my IPAD
>> 
>> 'Infinite wisdom through infinite means'
>> 
>> 
>>> On Dec 8, 2013, at 2:38 PM, Scott Ford  wrote:
>>> 
>>> Then suggest you don't come out guns blazing , unless they are loaded,
>> because you don't have a clue
>>> 
>>> Scott ford
>>> www.identityforge.com
>>> from my IPAD
>>> 
>>> 'Infinite wisdom through infinite means'
>>> 
>>> 
> On Dec 8, 2013, at 2:35 PM, "David L. Craig" 
>> wrote:
> 
> On 13Dec08:1428-0500, Scott Ford wrote:
> 
> What do circuits work in  ...
 
 Sigh... I suggest you discuss it with Gene Amdahl, not me.
 --
 
 May the LORD God bless you exceedingly abundantly!
 
 Dave_Craig__
 "So the universe is not quite as you thought it was.
 You'd better rearrange your beliefs, then.
 Because you certainly can't rearrange the universe."
 __--from_Nightfall_by_Asimov/Silverberg_
 
 --
 For IBM-MAIN subscribe / signoff / archive access instructions,
 send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
>>> 
>>> --
>>> For IBM-MAIN subscribe / signoff / archive access instructions,
>>> send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
>> 
>> --
>> For IBM-MAIN subscribe / signoff / archive access instructions,
>> send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
> 
> --
> For IBM-MAIN subscribe / signoff / archive access instructions,
> send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

Re: "hexadecimal"?

[Robert A. Rosenberg]

At 10:55 -0800 on 12/08/2013, Warren Brown wrote about Re: "hexadecimal"?:


So , hexadecimal is not the same EBCDIC ?


Data in the computer is stored in 8-bit long bytes. Hex is a way of 
displaying the contents of the bytes as two characters representing 
the value of the first and second set of 4 bits. EBCDIC is a mapping 
of Glyphs into one of 256 values. Thus the glyph "1" is mapped to the 
value 241 or X"F1".


 From: David L. Craig 
 To: IBM-MAIN@LISTSERV.UA.EDU Sent: Sunday, 
December 8, 2013 1:44 PM Subject: Re: "hexadecimal"? On 
13Dec08:0942-0800, Charles Mills wrote: > It is a pet peeve of mine. 
People use "hex" sloppily > to mean "binary" (what I think IBM means 
in your > example) or "non-printable" ("does it look like a DD > 
name?" "Nyah, it's a bunch of hex."). > > Hex is not a kind of data. 
It is a convenient way of > representing data. X'F1' is a clearer 
image in most > cases than 0001 or 241. All data is 
potentially > hex; that is, is representable in hex. That's the > 
beauty of hex. I would not expect to read this on this maining list. 
The original floating-point hardware of the S/360 architecture is 
hexadecimal, not binary.  A normalized value may have up to three 
leading binary zeros as a consequence. --  May 
the LORD God bless you exceedingly abundantly! 
Dave_Craig__ "So the 
universe is not quite as you thought it was. You'd better rearrange 
your beliefs, then. Because you certainly can't rearrange the 
universe." __--from_Nightfall_by_Asimov/Silverberg_ 
-- 
For IBM-MAIN subscribe / signoff / archive access instructions, send 
email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN


--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN


--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

Re: "hexadecimal"?

[Ed Finnell]

Sleigh bells ring...We had an ol' V8 in the basement that only did SWIFT.  
It queued the requests until it got to a billion then would do the transfer. 
 We'd get the float on the interim. Good solid revenue producer. Then one 
day it  'thermalled'. They had boxes and boxes of components trucked in from 
Sunnyvale.  Still took about 22hrs to get it back online. There were so many 
'suits'  wandering around the OPs manager came and said 'Listen up, if 
you're not a CE  you'll be escorted out!' 
 
In a tight budget, amazing how fast funds can be found to reduce single  
points of failure! 
 
 
In a message dated 12/8/2013 2:05:56 P.M. Central Standard Time,  
john.archie.mck...@gmail.com writes:

Amdahl  was the main architect of the S/360. So I guess we can mention him
on  occasion.


--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

Re: "hexadecimal"?

[John McKown]

Amdahl was the main architect of the S/360. So I guess we can mention him
on occasion.
On Dec 8, 2013 1:41 PM, "Scott Ford"  wrote:

> Btw this isn't a Amdahl forum, it's IBM...
>
> Scott ford
> www.identityforge.com
> from my IPAD
>
> 'Infinite wisdom through infinite means'
>
>
> > On Dec 8, 2013, at 2:38 PM, Scott Ford  wrote:
> >
> > Then suggest you don't come out guns blazing , unless they are loaded,
> because you don't have a clue
> >
> > Scott ford
> > www.identityforge.com
> > from my IPAD
> >
> > 'Infinite wisdom through infinite means'
> >
> >
> >>> On Dec 8, 2013, at 2:35 PM, "David L. Craig" 
> wrote:
> >>>
> >>> On 13Dec08:1428-0500, Scott Ford wrote:
> >>>
> >>> What do circuits work in  ...
> >>
> >> Sigh... I suggest you discuss it with Gene Amdahl, not me.
> >> --
> >> 
> >> May the LORD God bless you exceedingly abundantly!
> >>
> >> Dave_Craig__
> >> "So the universe is not quite as you thought it was.
> >> You'd better rearrange your beliefs, then.
> >> Because you certainly can't rearrange the universe."
> >> __--from_Nightfall_by_Asimov/Silverberg_
> >>
> >> --
> >> For IBM-MAIN subscribe / signoff / archive access instructions,
> >> send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
> >
> > --
> > For IBM-MAIN subscribe / signoff / archive access instructions,
> > send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
>
> --
> For IBM-MAIN subscribe / signoff / archive access instructions,
> send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
>

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

Re: "hexadecimal"?

[Peter Sylvester]

On 12/08/2013 08:07 PM, David L. Craig wrote:

On 13Dec08:1352-0500, Scott Ford wrote:


Huh .maybe you should back to computer basics
...at machine level everything is binary


You, sir, are the one who needs to go back--specifically
to zPOPs page 9-3:

please :-)

- look into wikipedia about the meaning of hexadecimal

- HPF  magnitude is the product of the significant and the radix powered to the 
exponent
  the significant is a sequence of bits
  the representation table in 9-6 shows what happens when you divide
 (1 - 0.5 - 1/64)
  the table does not use a hexadecimal representation
  if you change the exponent in an hfp by one, you need to shift
  the signicant by four bits to roughly get the same number (~ 
rounding/overflow)
  thus it seems convenient to think about the radix 16 significant as a
  sequence of hexadecimal numbers (which would then be shifted by 1)

- =X'ABCD' in BAL is still a representation of 16 bits.


best regards
Peter Sylvester





: results are also available.


--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

Re: "hexadecimal"?

[Scott Ford]

Btw this isn't a Amdahl forum, it's IBM...

Scott ford
www.identityforge.com
from my IPAD

'Infinite wisdom through infinite means'


> On Dec 8, 2013, at 2:38 PM, Scott Ford  wrote:
> 
> Then suggest you don't come out guns blazing , unless they are loaded, 
> because you don't have a clue
> 
> Scott ford
> www.identityforge.com
> from my IPAD
> 
> 'Infinite wisdom through infinite means'
> 
> 
>>> On Dec 8, 2013, at 2:35 PM, "David L. Craig"  wrote:
>>> 
>>> On 13Dec08:1428-0500, Scott Ford wrote:
>>> 
>>> What do circuits work in  ...
>> 
>> Sigh... I suggest you discuss it with Gene Amdahl, not me.
>> -- 
>> 
>> May the LORD God bless you exceedingly abundantly!
>> 
>> Dave_Craig__
>> "So the universe is not quite as you thought it was.
>> You'd better rearrange your beliefs, then.
>> Because you certainly can't rearrange the universe."
>> __--from_Nightfall_by_Asimov/Silverberg_
>> 
>> --
>> For IBM-MAIN subscribe / signoff / archive access instructions,
>> send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
> 
> --
> For IBM-MAIN subscribe / signoff / archive access instructions,
> send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

Re: "hexadecimal"?

[Scott Ford]

Then suggest you don't come out guns blazing , unless they are loaded, because 
you don't have a clue

Scott ford
www.identityforge.com
from my IPAD

'Infinite wisdom through infinite means'


> On Dec 8, 2013, at 2:35 PM, "David L. Craig"  wrote:
> 
>> On 13Dec08:1428-0500, Scott Ford wrote:
>> 
>> What do circuits work in  ...
> 
> Sigh... I suggest you discuss it with Gene Amdahl, not me.
> -- 
> 
> May the LORD God bless you exceedingly abundantly!
> 
> Dave_Craig__
> "So the universe is not quite as you thought it was.
> You'd better rearrange your beliefs, then.
> Because you certainly can't rearrange the universe."
> __--from_Nightfall_by_Asimov/Silverberg_
> 
> --
> For IBM-MAIN subscribe / signoff / archive access instructions,
> send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

Re: Something to Think About - Optimal PDS Blocking

[Shmuel Metz (Seymour J.)]

In <20131208075537.8d8d814dcc2cd889368c0...@gmx.net>, on 12/08/2013
   at 07:55 AM, "nitz-...@gmx.net"  said:

>> Yes, however FBS stands for Fixed Block Standard, not Spanned.

>Exactly. And the last record in an FBS data set can be "short", 
>i.e. less than lrecl. 

ITYM "i.e., less than BLKSIZE"; the block can never be less than LRECL
for FBS.

>The short record denotes the end of the data set.
>And all the utility programs know it

It's the access method; the utility program doesn't need to do
anything special to get EOF after a short block. It's when the utility
wants to suppress the EOF that it needs more code, but it can always
override to RECFM=FB. 
 
-- 
 Shmuel (Seymour J.) Metz, SysProg and JOAT
 ISO position; see  
We don't care. We don't have to care, we're Congress.
(S877: The Shut up and Eat Your spam act of 2003)

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

Re: "hexadecimal"?

[David L. Craig]

On 13Dec08:1428-0500, Scott Ford wrote:

> What do circuits work in  ...

Sigh... I suggest you discuss it with Gene Amdahl, not me.
-- 

May the LORD God bless you exceedingly abundantly!

Dave_Craig__
"So the universe is not quite as you thought it was.
 You'd better rearrange your beliefs, then.
 Because you certainly can't rearrange the universe."
__--from_Nightfall_by_Asimov/Silverberg_

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

Re: "hexadecimal"?

[Scott Ford]

What do circuits work in  ...

Scott ford
www.identityforge.com
from my IPAD

'Infinite wisdom through infinite means'


> On Dec 8, 2013, at 2:07 PM, "David L. Craig"  wrote:
> 
>> On 13Dec08:1352-0500, Scott Ford wrote:
>> 
>> Huh .maybe you should back to computer basics
>> ...at machine level everything is binary
>> 
>> Scott ford
>> www.identityforge.com
>> from my IPAD
> 
> You, sir, are the one who needs to go back--specifically
> to zPOPs page 9-3:
> 
> : Hexadecimal-Floating-Point (HFP)
> : 
> : Hexadecimal-floating-point (HFP) operands have for-
> : mats which provide for exponents that specify pow-
> : ers of the radix 16 and significands that are
> : hexadecimal numbers. The exponent range is the
> : same for the short, long, and extended formats. The
> : results of most operations on HFP data are truncated
> : to fit into the target format, but there are instructions
> : available to round the result when converting to a
> : narrower format. For HFP operands, the implicit unit
> : digit of the significand is always zero. Since the value
> : of the significand and fraction are the same, HFP
> : operations are described in terms of the fraction, and
> : the term significand is not used.
> : 
> : Either normalized or unnormalized numbers may be
> : used as operands for any HFP or DFP operation,
> : where, for HFP, a normalized number is one having a
> : nonzero leftmost fraction digit, or, for DFP, a normal-
> : ized number is one having a nonzero leftmost signifi-
> : cand digit. Most HFP instructions generate
> : normalized results for greatest precision. HFP add
> : and subtract instructions that generate unnormalized
> : results are also available.
> -- 
> 
> May the LORD God bless you exceedingly abundantly!
> 
> Dave_Craig__
> "So the universe is not quite as you thought it was.
> You'd better rearrange your beliefs, then.
> Because you certainly can't rearrange the universe."
> __--from_Nightfall_by_Asimov/Silverberg_
> 
> --
> For IBM-MAIN subscribe / signoff / archive access instructions,
> send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

Re: "hexadecimal"?

[David L. Craig]

On 13Dec08:1352-0500, Scott Ford wrote:

> Huh .maybe you should back to computer basics
> ...at machine level everything is binary
> 
> Scott ford
> www.identityforge.com
> from my IPAD

You, sir, are the one who needs to go back--specifically
to zPOPs page 9-3:

: Hexadecimal-Floating-Point (HFP)
: 
: Hexadecimal-floating-point (HFP) operands have for-
: mats which provide for exponents that specify pow-
: ers of the radix 16 and significands that are
: hexadecimal numbers. The exponent range is the
: same for the short, long, and extended formats. The
: results of most operations on HFP data are truncated
: to fit into the target format, but there are instructions
: available to round the result when converting to a
: narrower format. For HFP operands, the implicit unit
: digit of the significand is always zero. Since the value
: of the significand and fraction are the same, HFP
: operations are described in terms of the fraction, and
: the term significand is not used.
: 
: Either normalized or unnormalized numbers may be
: used as operands for any HFP or DFP operation,
: where, for HFP, a normalized number is one having a
: nonzero leftmost fraction digit, or, for DFP, a normal-
: ized number is one having a nonzero leftmost signifi-
: cand digit. Most HFP instructions generate
: normalized results for greatest precision. HFP add
: and subtract instructions that generate unnormalized
: results are also available.
-- 

May the LORD God bless you exceedingly abundantly!

Dave_Craig__
"So the universe is not quite as you thought it was.
 You'd better rearrange your beliefs, then.
 Because you certainly can't rearrange the universe."
__--from_Nightfall_by_Asimov/Silverberg_

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

Re: JCL enhancement thought for UNIX files?

[Scott Ford]

Gil,

Yep, that's would be great to do

Scott ford
www.identityforge.com
from my IPAD

'Infinite wisdom through infinite means'


> On Dec 8, 2013, at 1:00 PM, Paul Gilmartin  wrote:
> 
>> On Sun, 8 Dec 2013 10:34:37 -0600, John McKown wrote:
>> 
>> The problem is that there is no way to dynamically create a subdirectory
>> via JCL. So I was thinking that a new PATHOPTS might be useful. Something
>> like O_MKDIR which would tell the initiator to do the equivalent of a
>> "mkdir -p /tmp2/&SYSUID/&JOBNAME" and then create the file in that
>> subdirectory.
> +1
> 
> The "-p" is important, but I wouldn't mind an additional DD statement
> to create the directory.
> 
> My preference, rather than extending PATHOPTS with a non-POSIX
> option, would be to have it a suboption of PATHDISP, which is non-POSIX
> anyway, and notionally similar to DISP.
> 
> A similar wish for links.  Why is there a DD option to create FIFOs,
> but none to create links?
> 
> -- gil
> 
> --
> For IBM-MAIN subscribe / signoff / archive access instructions,
> send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

Re: "hexadecimal"?

[Scott Ford]

Circuits talk 1s and 0s ...not hex

Scott ford
www.identityforge.com
from my IPAD

'Infinite wisdom through infinite means'


> On Dec 8, 2013, at 1:58 PM, Scott Ford  wrote:
> 
> Hexadecimal is a readable system for binary, if the hex characters are not 
> numbers per se they are EBCDIC...that's been since I wrote BAL on s360/20
> 
> Scott ford
> www.identityforge.com
> from my IPAD
> 
> 'Infinite wisdom through infinite means'
> 
> 
>> On Dec 8, 2013, at 1:55 PM, Warren Brown  wrote:
>> 
>> So , hexadecimal is not the same EBCDIC ?
>> 
>> 
>> 
>> From: David L. Craig 
>> To: IBM-MAIN@LISTSERV.UA.EDU 
>> Sent: Sunday, December 8, 2013 1:44 PM
>> Subject: Re: "hexadecimal"?
>> 
>> 
>>> On 13Dec08:0942-0800, Charles Mills wrote:
>>> 
>>> It is a pet peeve of mine. People use "hex" sloppily
>>> to mean "binary" (what I think IBM means in your
>>> example) or "non-printable" ("does it look like a DD
>>> name?" "Nyah, it's a bunch of hex.").
>>> 
>>> Hex is not a kind of data. It is a convenient way of
>>> representing data. X'F1' is a clearer image in most
>>> cases than 0001 or 241. All data is potentially
>>> hex; that is, is representable in hex. That's the
>>> beauty of hex.
>> 
>> I would not expect to read this on this maining list.
>> The original floating-point hardware of the S/360
>> architecture is hexadecimal, not binary.  A normalized
>> value may have up to three leading binary zeros as a
>> consequence.
>> -- 
>> 
>> May the LORD God bless you exceedingly abundantly!
>> 
>> Dave_Craig__
>> "So the universe is not quite as you thought it was.
>> You'd better rearrange your beliefs, then.
>> Because you certainly can't rearrange the universe."
>> __--from_Nightfall_by_Asimov/Silverberg_
>> 
>> --
>> For IBM-MAIN subscribe / signoff / archive access instructions,
>> send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
>> 
>> --
>> For IBM-MAIN subscribe / signoff / archive access instructions,
>> send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
> 
> --
> For IBM-MAIN subscribe / signoff / archive access instructions,
> send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

Re: "hexadecimal"?

[Scott Ford]

Hexadecimal is a readable system for binary, if the hex characters are not 
numbers per se they are EBCDIC...that's been since I wrote BAL on s360/20

Scott ford
www.identityforge.com
from my IPAD

'Infinite wisdom through infinite means'


> On Dec 8, 2013, at 1:55 PM, Warren Brown  wrote:
> 
> So , hexadecimal is not the same EBCDIC ?
> 
> 
> 
> From: David L. Craig 
> To: IBM-MAIN@LISTSERV.UA.EDU 
> Sent: Sunday, December 8, 2013 1:44 PM
> Subject: Re: "hexadecimal"?
> 
> 
>> On 13Dec08:0942-0800, Charles Mills wrote:
>> 
>> It is a pet peeve of mine. People use "hex" sloppily
>> to mean "binary" (what I think IBM means in your
>> example) or "non-printable" ("does it look like a DD
>> name?" "Nyah, it's a bunch of hex.").
>> 
>> Hex is not a kind of data. It is a convenient way of
>> representing data. X'F1' is a clearer image in most
>> cases than 0001 or 241. All data is potentially
>> hex; that is, is representable in hex. That's the
>> beauty of hex.
> 
> I would not expect to read this on this maining list.
> The original floating-point hardware of the S/360
> architecture is hexadecimal, not binary.  A normalized
> value may have up to three leading binary zeros as a
> consequence.
> -- 
> 
> May the LORD God bless you exceedingly abundantly!
> 
> Dave_Craig__
> "So the universe is not quite as you thought it was.
> You'd better rearrange your beliefs, then.
> Because you certainly can't rearrange the universe."
> __--from_Nightfall_by_Asimov/Silverberg_
> 
> --
> For IBM-MAIN subscribe / signoff / archive access instructions,
> send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
> 
> --
> For IBM-MAIN subscribe / signoff / archive access instructions,
> send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

Re: "hexadecimal"?

[Warren Brown]

So , hexadecimal is not the same EBCDIC ?



 From: David L. Craig 
To: IBM-MAIN@LISTSERV.UA.EDU 
Sent: Sunday, December 8, 2013 1:44 PM
Subject: Re: "hexadecimal"?
 

On 13Dec08:0942-0800, Charles Mills wrote:

> It is a pet peeve of mine. People use "hex" sloppily
> to mean "binary" (what I think IBM means in your
> example) or "non-printable" ("does it look like a DD
> name?" "Nyah, it's a bunch of hex.").
> 
> Hex is not a kind of data. It is a convenient way of
> representing data. X'F1' is a clearer image in most
> cases than 0001 or 241. All data is potentially
> hex; that is, is representable in hex. That's the
> beauty of hex.

I would not expect to read this on this maining list.
The original floating-point hardware of the S/360
architecture is hexadecimal, not binary.  A normalized
value may have up to three leading binary zeros as a
consequence.
-- 

May the LORD God bless you exceedingly abundantly!

Dave_Craig__
"So the universe is not quite as you thought it was.
You'd better rearrange your beliefs, then.
Because you certainly can't rearrange the universe."
__--from_Nightfall_by_Asimov/Silverberg_

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

Re: "hexadecimal"?

[Scott Ford]

Huh .maybe you should back to computer basics ...at machine level 
everything is binary

Scott ford
www.identityforge.com
from my IPAD

'Infinite wisdom through infinite means'


> On Dec 8, 2013, at 1:44 PM, "David L. Craig"  wrote:
> 
>> On 13Dec08:0942-0800, Charles Mills wrote:
>> 
>> It is a pet peeve of mine. People use "hex" sloppily
>> to mean "binary" (what I think IBM means in your
>> example) or "non-printable" ("does it look like a DD
>> name?" "Nyah, it's a bunch of hex.").
>> 
>> Hex is not a kind of data. It is a convenient way of
>> representing data. X'F1' is a clearer image in most
>> cases than 0001 or 241. All data is potentially
>> hex; that is, is representable in hex. That's the
>> beauty of hex.
> 
> I would not expect to read this on this maining list.
> The original floating-point hardware of the S/360
> architecture is hexadecimal, not binary.  A normalized
> value may have up to three leading binary zeros as a
> consequence.
> -- 
> 
> May the LORD God bless you exceedingly abundantly!
> 
> Dave_Craig__
> "So the universe is not quite as you thought it was.
> You'd better rearrange your beliefs, then.
> Because you certainly can't rearrange the universe."
> __--from_Nightfall_by_Asimov/Silverberg_
> 
> --
> For IBM-MAIN subscribe / signoff / archive access instructions,
> send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

Re: "hexadecimal"?

[David L. Craig]

On 13Dec08:0942-0800, Charles Mills wrote:

> It is a pet peeve of mine. People use "hex" sloppily
> to mean "binary" (what I think IBM means in your
> example) or "non-printable" ("does it look like a DD
> name?" "Nyah, it's a bunch of hex.").
> 
> Hex is not a kind of data. It is a convenient way of
> representing data. X'F1' is a clearer image in most
> cases than 0001 or 241. All data is potentially
> hex; that is, is representable in hex. That's the
> beauty of hex.

I would not expect to read this on this maining list.
The original floating-point hardware of the S/360
architecture is hexadecimal, not binary.  A normalized
value may have up to three leading binary zeros as a
consequence.
-- 

May the LORD God bless you exceedingly abundantly!

Dave_Craig__
"So the universe is not quite as you thought it was.
 You'd better rearrange your beliefs, then.
 Because you certainly can't rearrange the universe."
__--from_Nightfall_by_Asimov/Silverberg_

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

Re: "hexadecimal"?

[Gerhard Postpischil]

On 12/8/2013 12:52 PM, John McKown wrote:

Agree. I have not checked, but will bat this is really a full word integer
binary number.


Minor nit - DD= specifies a halfword.


Gerhard Postpischil
Bradford, Vermont

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

Re: "hexadecimal"?

[John Gilmore]

Charles Mills wrote

| Hex is not a kind of data.

and I think we all sympathize with the point he is making.  There is,
however, some real hexadecimal data around, beginning with its use in
IBM 7090 channel programming.  For the System/360 and its sequelæ the
exponent of an HFP number is represented in storage as a (biased)
hexadecimal number and operated upon using hexadecimal arithmetic.
(This is the rationale for the PL/I characterization of a fullword,
single-precision HFP value as binary float (21) and not binary
float(24) although the mantissa does in fact contain 24 bits.)

John Gilmore, Ashland, MA 01721 - USA

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

Re: JCL enhancement thought for UNIX files?

[Paul Gilmartin]

On Sun, 8 Dec 2013 10:34:37 -0600, John McKown wrote:
>
>The problem is that there is no way to dynamically create a subdirectory
>via JCL. So I was thinking that a new PATHOPTS might be useful. Something
>like O_MKDIR which would tell the initiator to do the equivalent of a
>"mkdir -p /tmp2/&SYSUID/&JOBNAME" and then create the file in that
>subdirectory.
> 
+1

The "-p" is important, but I wouldn't mind an additional DD statement
to create the directory.

My preference, rather than extending PATHOPTS with a non-POSIX
option, would be to have it a suboption of PATHDISP, which is non-POSIX
anyway, and notionally similar to DISP.

A similar wish for links.  Why is there a DD option to create FIFOs,
but none to create links?

-- gil

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

Re: "hexadecimal"?

[John McKown]

Agree. I have not checked, but will bat this is really a full word integer
binary number.
 On Dec 8, 2013 11:43 AM, "Charles Mills"  wrote:

> It is a pet peeve of mine. People use "hex" sloppily to mean "binary"
> (what I think IBM means in your example) or "non-printable" ("does it look
> like a DD name?" "Nyah, it's a bunch of hex.").
>
> Hex is not a kind of data. It is a convenient way of representing data.
> X'F1' is a clearer image in most cases than 0001 or 241. All data is
> potentially hex; that is, is representable in hex. That's the beauty of hex.
>
> Charles
>
> -Original Message-
> From: IBM Mainframe Discussion List [mailto:IBM-MAIN@LISTSERV.UA.EDU] On
> Behalf Of Paul Gilmartin
> Sent: Saturday, December 07, 2013 2:53 PM
> To: IBM-MAIN@LISTSERV.UA.EDU
> Subject: "hexadecimal"?
>
> From:
>
> Title: z/OS V1R13.0 DFSMSdfp Advanced Services
> Document Number: SC26-7400-14
>
> ... that I was reading lately:
>
> 7.5.3.1 TRKCALC--Standard Form
> ...
> DD=addr--RX-type address, (2-12), (0), (14), or n
> You can specify either the address of a field containing the
> hexadecimal value of the record's data length, ...
>
> Hexadecimal!?  Does this mean the value must be coded as hexadecimal
> display, e.g. C'50' to indicate 80?  Or must it be coded as a hexadecimal
> self-defining term (X'50')?  Why not a decimal self-defining term (80)?
> Or a binary self-defining term (B'0101')?  Or even an A()-constant
> (AL2(100-20))?
>
> Is "hexadecimal" otiose?  Is it time for an RCF?
>
> (I've seen other uses of "hexadecimal" that I find otiose in MVS
> documentation.  I suspect a tech writer's mother was traumatized during
> gestation by hexadecimal notation.)
>
> --
> For IBM-MAIN subscribe / signoff / archive access instructions,
> send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
>

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

Re: "hexadecimal"?

[Charles Mills]

It is a pet peeve of mine. People use "hex" sloppily to mean "binary" (what I 
think IBM means in your example) or "non-printable" ("does it look like a DD 
name?" "Nyah, it's a bunch of hex.").

Hex is not a kind of data. It is a convenient way of representing data. X'F1' 
is a clearer image in most cases than 0001 or 241. All data is potentially 
hex; that is, is representable in hex. That's the beauty of hex. 

Charles

-Original Message-
From: IBM Mainframe Discussion List [mailto:IBM-MAIN@LISTSERV.UA.EDU] On Behalf 
Of Paul Gilmartin
Sent: Saturday, December 07, 2013 2:53 PM
To: IBM-MAIN@LISTSERV.UA.EDU
Subject: "hexadecimal"?

From:

Title: z/OS V1R13.0 DFSMSdfp Advanced Services
Document Number: SC26-7400-14

... that I was reading lately:

7.5.3.1 TRKCALC--Standard Form
...
DD=addr--RX-type address, (2-12), (0), (14), or n
You can specify either the address of a field containing the
hexadecimal value of the record's data length, ...

Hexadecimal!?  Does this mean the value must be coded as hexadecimal display, 
e.g. C'50' to indicate 80?  Or must it be coded as a hexadecimal self-defining 
term (X'50')?  Why not a decimal self-defining term (80)?
Or a binary self-defining term (B'0101')?  Or even an A()-constant 
(AL2(100-20))?

Is "hexadecimal" otiose?  Is it time for an RCF?

(I've seen other uses of "hexadecimal" that I find otiose in MVS documentation. 
 I suspect a tech writer's mother was traumatized during gestation by 
hexadecimal notation.)

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

Re: Something to Think About - Optimal PDS Blocking

[Blaicher, Christopher Y.]

Sorry Barbara, I forgot about SADUMP.  I wonder, does the same problem exist 
with a console dump or a SYSMDUMP? 

Chris Blaicher
Principal Software Engineer, Software Development
Syncsort Incorporated
50 Tice Boulevard, Woodcliff Lake, NJ 07677
P: 201-930-8260  |  M: 512-627-3803    
E: cblaic...@syncsort.com

-Original Message-
From: IBM Mainframe Discussion List [mailto:IBM-MAIN@LISTSERV.UA.EDU] On Behalf 
Of nitz-...@gmx.net
Sent: Sunday, December 08, 2013 11:01 AM
To: IBM-MAIN@LISTSERV.UA.EDU
Subject: Re: Something to Think About - Optimal PDS Blocking

> I am somewhat at a loss to understand how some of the problems you are 
> detailing happened.  The only way it could have would be with an ill behaving 
> user written program or process.

If I remember this correctly (and I am on shaky ground here), sadump writes a 
'special striping', understood fully only by IPCS. Mind you, sadump is a 
standalone application that does not use standard z/OS services because they 
are not available. Either here in old posts or even somewhere in the docs the 
behaviour I detailed is described, including the warning NOT to use standard 
utilities when copying a 'striped' (multi DASD volume) sadump. Suffice it to 
say, the full 63000cyls (from my recent customer) could only be copied when the 
customer used IPCS COPYDUMP. I was not told how they copied it when they sent 
me the 27000cyls.

Just thought that a reminder about sadump was in order. After all, who would 
want to take an sadump only to be told by support that the necessary data are 
not in the dump?

Barbara

--
For IBM-MAIN subscribe / signoff / archive access instructions, send email to 
lists...@listserv.ua.edu with the message: INFO IBM-MAIN

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

Re: JCL enhancement thought for UNIX files?

[Scott Ford]

John,

What about removing the subdirectory ? I would assume like a br14 
mod,delete,delete...
Dynamic creation also would be nice somehow 

Scott ford
www.identityforge.com
from my IPAD

'Infinite wisdom through infinite means'


> On Dec 8, 2013, at 11:34 AM, John McKown  wrote:
> 
> This may be an "iffy" one.But I'll throw it out for discussion (and if
> nobody wants to discuss, then I have my answer).
> 
> I have an "automount" directory at /tmp2 . For a UNIX shell, I set all the
> TMPxxx type environment variables to "/tmp2/&SYSUID". This makes the
> default temporary UNIX subdirectory for each user to be unique. I do this
> so that some "hog" of a user can't use up the temp space for other users.
> I.e. they can only shoot themselves in their own foot, not somebody else's.
> 
> Therefore, in a batch job, I can also use
> PATH='/tmp2/&SYSUID/some-file.txt', ... However, what I would kind of like
> to do is create a separate subdirectory for a job based on the job name. In
> most shops, this would likely ensure that two concurrently running jobs
> would not contend over a file name. Yes, I know that I could use
> PATH='/tmp2/&SYSUID/&JOBNAME-some.file.txt', but I would prefer
> PATH='/tmp2/&SYSUID/&JOBNAME/some.file.txt'. Just another personal oddity.
> The problem is that there is no way to dynamically create a subdirectory
> via JCL. So I was thinking that a new PATHOPTS might be useful. Something
> like O_MKDIR which would tell the initiator to do the equivalent of a
> "mkdir -p /tmp2/&SYSUID/&JOBNAME" and then create the file in that
> subdirectory.
> 
> The only problem I can envision (and there are likely more), is that the
> JCL coders decide to just make O_MKDIR a standard for use on every DD which
> is for a UNIX file. Which could result in subdirectories which are not
> really wanted if the JCL coder mistypes something. Of course, the internal
> "mkdir" should be done using the RACF id / UID of the job's OWNER, and not
> the initiator. Which would hopefully stop somebody from creating UNIX
> subdirectories and files where they don't belong. Also, I'm not sure what
> the UNIX mode bits should be for any created subdirectories. At the least,
> they should be 7??. I.e. at least read-write-execute for OWNER.
> 
> In the interim, I guess I'm stuck with
> '/tmp2/&SYSUID/&JOBNAME-some.file.txt'.
> 
> -- 
> This is clearly another case of too many mad scientists, and not enough
> hunchbacks.
> 
> Maranatha! <><
> John McKown
> 
> --
> For IBM-MAIN subscribe / signoff / archive access instructions,
> send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

Re: Something to Think About - Optimal PDS Blocking

[John Gilmore]

Barbara's explanation of what happened is entirely clear.  The answer
to John Perryman's question


With the exception of DISP=MOD with multiple open/writes and that the
last block will always be, I don't see the difference between FB and
FBS. How would FB create a short block in the middle?


is thus clear enough for a striped FBS data set processed in naif
fashion, one read not in block rotation but seriatim by writing
'drive'.

John Gilmore, Ashland, MA 01721 - USA

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

JCL enhancement thought for UNIX files?

[John McKown]

This may be an "iffy" one.But I'll throw it out for discussion (and if
nobody wants to discuss, then I have my answer).

I have an "automount" directory at /tmp2 . For a UNIX shell, I set all the
TMPxxx type environment variables to "/tmp2/&SYSUID". This makes the
default temporary UNIX subdirectory for each user to be unique. I do this
so that some "hog" of a user can't use up the temp space for other users.
I.e. they can only shoot themselves in their own foot, not somebody else's.

Therefore, in a batch job, I can also use
PATH='/tmp2/&SYSUID/some-file.txt', ... However, what I would kind of like
to do is create a separate subdirectory for a job based on the job name. In
most shops, this would likely ensure that two concurrently running jobs
would not contend over a file name. Yes, I know that I could use
PATH='/tmp2/&SYSUID/&JOBNAME-some.file.txt', but I would prefer
PATH='/tmp2/&SYSUID/&JOBNAME/some.file.txt'. Just another personal oddity.
The problem is that there is no way to dynamically create a subdirectory
via JCL. So I was thinking that a new PATHOPTS might be useful. Something
like O_MKDIR which would tell the initiator to do the equivalent of a
"mkdir -p /tmp2/&SYSUID/&JOBNAME" and then create the file in that
subdirectory.

The only problem I can envision (and there are likely more), is that the
JCL coders decide to just make O_MKDIR a standard for use on every DD which
is for a UNIX file. Which could result in subdirectories which are not
really wanted if the JCL coder mistypes something. Of course, the internal
"mkdir" should be done using the RACF id / UID of the job's OWNER, and not
the initiator. Which would hopefully stop somebody from creating UNIX
subdirectories and files where they don't belong. Also, I'm not sure what
the UNIX mode bits should be for any created subdirectories. At the least,
they should be 7??. I.e. at least read-write-execute for OWNER.

In the interim, I guess I'm stuck with
'/tmp2/&SYSUID/&JOBNAME-some.file.txt'.

-- 
This is clearly another case of too many mad scientists, and not enough
hunchbacks.

Maranatha! <><
John McKown

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

Re: Something to Think About - Optimal PDS Blocking

[Steve Comstock]

On 12/8/2013 8:58 AM, Jon Perryman wrote:

With the exception of DISP=MOD with multiple open/writes and that the last

block will always be, I don't see the difference between FB and FBS. How would
FB create a short block in the middle?

Well, except for DISP=MOD, it won't: it is a DSIP=MOD situation
that allows the possibility.


-Steve Comstock




Jon Perryman


- Original Message -

From: "nitz-...@gmx.net" 
To: IBM-MAIN@LISTSERV.UA.EDU
Cc:
Sent: Saturday, December 7, 2013 10:55 PM
Subject: Re: Something to Think About - Optimal PDS Blocking


  Yes, however FBS stands for Fixed Block Standard, not Spanned.

Exactly. And the last record in an FBS data set can be "short", i.e.
less than lrecl. The short record denotes the end of the data set. And all the
utility programs know it and stop processing once they reach the short record.
That is all fine and well as long as we are not dealing with a multivolume data
set.

Think standalone dump written striped to  - say - 5 volumes. Each volume has a
data set in format FBS, but only one of the volumes can have a short record.
SAdump knows that, and IPCS knows it, too. The utilities don't. So assume
that you took a complete sadump to 5 volumes and the sort record happens to be
on the first volume. Then you use a utility (ICEGENER is my favourite) to copy
somewhere else. You end up with a severely truncated sadump. One fifth, to be
exact. IPCS will read the truncated dump to the best of its abilities, but you
will get all kinds of 'storage not available' warnings when looking at
the dump.

Last time a customer sent me an sadump, it had 27000cyl. I got all kinds of
warnings and got lucky in that the sadump messages were clearly truncated and
didn't show the 'successfully finished' message. It turned out that
the wrong utility was used for copying, and the actual dump had 63000 cyls.
Visible when IPCS COPYDUMP was used for copying. IPCS knows that a striped
sadump can have the short record "earlier".

Barbara

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN



--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN



--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

Re: Something to Think About - Optimal PDS Blocking

[nitz-...@gmx.net]

> I am somewhat at a loss to understand how some of the problems you are 
> detailing happened.  The only way it could have would be with an ill behaving 
> user written program or process.

If I remember this correctly (and I am on shaky ground here), sadump writes a 
'special striping', understood fully only by IPCS. Mind you, sadump is a 
standalone application that does not use standard z/OS services because they 
are not available. Either here in old posts or even somewhere in the docs the 
behaviour I detailed is described, including the warning NOT to use standard 
utilities when copying a 'striped' (multi DASD volume) sadump. Suffice it to 
say, the full 63000cyls (from my recent customer) could only be copied when the 
customer used IPCS COPYDUMP. I was not told how they copied it when they sent 
me the 27000cyls.

Just thought that a reminder about sadump was in order. After all, who would 
want to take an sadump only to be told by support that the necessary data are 
not in the dump?

Barbara

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

Re: Something to Think About - Optimal PDS Blocking

[Jon Perryman]

With the exception of DISP=MOD with multiple open/writes and that the last 
block will always be, I don't see the difference between FB and FBS. How would 
FB create a short block in the middle?

Jon Perryman


- Original Message -
> From: "nitz-...@gmx.net" 
> To: IBM-MAIN@LISTSERV.UA.EDU
> Cc: 
> Sent: Saturday, December 7, 2013 10:55 PM
> Subject: Re: Something to Think About - Optimal PDS Blocking
> 
>>  Yes, however FBS stands for Fixed Block Standard, not Spanned.
> Exactly. And the last record in an FBS data set can be "short", i.e. 
> less than lrecl. The short record denotes the end of the data set. And all 
> the 
> utility programs know it and stop processing once they reach the short 
> record. 
> That is all fine and well as long as we are not dealing with a multivolume 
> data 
> set.
> 
> Think standalone dump written striped to  - say - 5 volumes. Each volume has 
> a 
> data set in format FBS, but only one of the volumes can have a short record. 
> SAdump knows that, and IPCS knows it, too. The utilities don't. So assume 
> that you took a complete sadump to 5 volumes and the sort record happens to 
> be 
> on the first volume. Then you use a utility (ICEGENER is my favourite) to 
> copy 
> somewhere else. You end up with a severely truncated sadump. One fifth, to be 
> exact. IPCS will read the truncated dump to the best of its abilities, but 
> you 
> will get all kinds of 'storage not available' warnings when looking at 
> the dump.
> 
> Last time a customer sent me an sadump, it had 27000cyl. I got all kinds of 
> warnings and got lucky in that the sadump messages were clearly truncated and 
> didn't show the 'successfully finished' message. It turned out that 
> the wrong utility was used for copying, and the actual dump had 63000 cyls. 
> Visible when IPCS COPYDUMP was used for copying. IPCS knows that a striped 
> sadump can have the short record "earlier".
> 
> Barbara
> 
> --
> For IBM-MAIN subscribe / signoff / archive access instructions,
> send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
>

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

Re: Something to Think About - Optimal PDS Blocking

[Blaicher, Christopher Y.]

I am somewhat at a loss to understand how some of the problems you are 
detailing happened.  The only way it could have would be with an ill behaving 
user written program or process.

Any utility needs to read all 5 stripes in unison (my description, not a 
technical description), not volume by volume.  Striped data for a two stripe 
data set is written, basically, some on stripe 1, then some on stripe 2, then 
some more on stripe 1, then back and forth between the two.  Any attempt to 
read all of stripe 1 and then all of stripe 2 will end up with garbage, or at 
least with data out of order.

A user process of skipping the catalog and directly copying volume 1 to tape 1 
and volume 2 to tape 2, etc. in an attempt to 'speed up the process' will end 
up with an unusable bunch of data on 5 tapes.

The more common problem I have seen is a user using DISP=MOD to add data to a 
RECFM=FBS file.  On extending a file IOS does not read the last short block and 
re-write it as a full block, so unless you are very lucky and the last block 
written was a full block, you will not see the data added to a DISP=MOD FBS 
data set.

Chris Blaicher
Principal Software Engineer, Software Development
Syncsort Incorporated
50 Tice Boulevard, Woodcliff Lake, NJ 07677
P: 201-930-8260  |  M: 512-627-3803    
E: cblaic...@syncsort.com

-Original Message-
From: IBM Mainframe Discussion List [mailto:IBM-MAIN@LISTSERV.UA.EDU] On Behalf 
Of nitz-...@gmx.net
Sent: Sunday, December 08, 2013 1:56 AM
To: IBM-MAIN@LISTSERV.UA.EDU
Subject: Re: Something to Think About - Optimal PDS Blocking

> Yes, however FBS stands for Fixed Block Standard, not Spanned.
Exactly. And the last record in an FBS data set can be "short", i.e. less than 
lrecl. The short record denotes the end of the data set. And all the utility 
programs know it and stop processing once they reach the short record. That is 
all fine and well as long as we are not dealing with a multivolume data set.

Think standalone dump written striped to  - say - 5 volumes. Each volume has a 
data set in format FBS, but only one of the volumes can have a short record. 
SAdump knows that, and IPCS knows it, too. The utilities don't. So assume that 
you took a complete sadump to 5 volumes and the sort record happens to be on 
the first volume. Then you use a utility (ICEGENER is my favourite) to copy 
somewhere else. You end up with a severely truncated sadump. One fifth, to be 
exact. IPCS will read the truncated dump to the best of its abilities, but you 
will get all kinds of 'storage not available' warnings when looking at the dump.

Last time a customer sent me an sadump, it had 27000cyl. I got all kinds of 
warnings and got lucky in that the sadump messages were clearly truncated and 
didn't show the 'successfully finished' message. It turned out that the wrong 
utility was used for copying, and the actual dump had 63000 cyls. Visible when 
IPCS COPYDUMP was used for copying. IPCS knows that a striped sadump can have 
the short record "earlier".

Barbara

--
For IBM-MAIN subscribe / signoff / archive access instructions, send email to 
lists...@listserv.ua.edu with the message: INFO IBM-MAIN

--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN