Latest Principles of Operation
John R. Ehrman , 408-463-3543 T/543-
available at
http://publibz.boulder.ibm.com/epubs/pdf/a2278325.pdf
John Ehrman
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to list...@bama.ua.edu with the message: GET IBM-MAIN INFO
Search the archives at http://bama.ua.edu/archives/ibm-main.html
Binyamin Dissen
<ehr...@VNET.IBM.COM> wrote:
:>The new updated to the z/Architecture Principles of Operation is now
:>available at
:> http://publibz.boulder.ibm.com/epubs/pdf/a2278325.pdf
Some interesting new instructions.
Of course, I cannot use them until all customers are at the hardware level,
but ....
--
Binyamin Dissen <bdi...@dissensoftware.com>
http://www.dissensoftware.com
Director, Dissen Software, Bar & Grill - Israel
Should you use the mailblocks package and expect a response from me,
you should preauthorize the dissensoftware.com domain.
I very rarely bother responding to challenge/response systems,
especially those from irresponsible companies.
Tom Schmidt
>The new updated to the z/Architecture Principles of Operation is now
>available
..
So where is the MVCOS instruction? (I was expecting to find it in this
update.)
--
Tom Schmidt
Madison, WI
rkue...@ibm-main.lst
with it on!
IBM Mainframe Discussion List <IBM-...@BAMA.UA.EDU> wrote on 04/23/2007
06:13:09 PM:
> On Mon, 23 Apr 2007 14:10:15 -0700, John R. Ehrman wrote:
> >The new updated to the z/Architecture Principles of Operation is now
> >available
> ....
> So where is the MVCOS instruction? (I was expecting to find it in this
> update.)
> Tom Schmidt
> Madison, WI
-----------------------------------------
The information contained in this communication (including any
attachments hereto) is confidential and is intended solely for the
personal and confidential use of the individual or entity to whom
it is addressed. The information may also constitute a legally
privileged confidential communication. If the reader of this
message is not the intended recipient or an agent responsible for
delivering it to the intended recipient, you are hereby notified
that you have received this communication in error and that any
review, dissemination, copying, or unauthorized use of this
information, or the taking of any action in reliance on the
contents of this information is strictly prohibited. If you have
received this communication in error, please notify us immediately
by e-mail, and delete the original message. Thank you
Steve Comstock
> On Mon, 23 Apr 2007 14:10:15 -0700, John R. Ehrman wrote:
>
>
>>The new updated to the z/Architecture Principles of Operation is now
>>available
>
>
> So where is the MVCOS instruction? (I was expecting to find it in this
> update.)
>
Good point! I haven't gotten around to checking the
book out, but I, too, was expecting that write up.
Kind regards,
-Steve Comstock
The Trainer's Friend, Inc.
303-393-8716
http://www.trainersfriend.com
** Spring Promotion **
10% - 15% off for classes booked by April 30, 2007
to be taught by September 30, 2007
Call or email for details.
** Spring Promotion **
Tom Marchant
>The new updated to the z/Architecture Principles of Operation is now
>available at
>
> http://publibz.boulder.ibm.com/epubs/pdf/a2278325.pdf
Thanks, John. Will it be available as a BookManager book too?
--
Tom Marchant
Tom Marchant
>The new updated to the z/Architecture Principles of Operation is now
>available at
>
> http://publibz.boulder.ibm.com/epubs/pdf/a2278325.pdf
With the new instructions, I now count 751 instructions documented in the
POO. That's up a lot from the (IIRC) 143 for S/360.
Rick Fochtman
>>The new updated to the z/Architecture Principles of Operation is now
>>available at
>>
>> http://publibz.boulder.ibm.com/epubs/pdf/a2278325.pdf
>>
>>
>
>With the new instructions, I now count 751 instructions documented in the
>POO. That's up a lot from the (IIRC) 143 for S/360.
>
>
-------------------------<unsnip>-----------------------
It's called evolution, son. Like it or not, we're stuck with it and it
can be a good thing, in spite of personnal opinions to the contrary.
Tom Marchant
>>
>>With the new instructions, I now count 751 instructions documented in the
>>POO. That's up a lot from the (IIRC) 143 for S/360.
>>
>
>It's called evolution, son. Like it or not, we're stuck with it and it
>can be a good thing, in spite of personnal opinions to the contrary.
That wasn't a complaint. Actually, I do like it, even though it makes being an
Assembler programmer more challenging. It's one of the many details about
this platform that I find to be very impressive.
--
Tom Marchant
McKown, John
> From: IBM Mainframe Discussion List
> [mailto:IBM-...@BAMA.UA.EDU] On Behalf Of Tom Marchant
> Sent: Wednesday, April 25, 2007 2:34 PM
> To: IBM-...@BAMA.UA.EDU
> Subject: Re: Latest Principles of Operation
<snip>
>
> That wasn't a complaint. Actually, I do like it, even though
> it makes being an
> Assembler programmer more challenging. It's one of the many
> details about
> this platform that I find to be very impressive.
>
> --
> Tom Marchant
Perhaps to you, but not true in general according to one article.
Paraphrasing: Having so many instructions will simply confuse the
programmer! Better to have "one, true way" than many.
http://www.oreillynet.com/onlamp/blog/2007/04/the_virtues_of_monoculture
html
This article is basically about why Microsoft is better than Linux in
some ways. Linux has too many competing ways to do something whereas
Microsoft is usually monolithic. And "lack of choice" is superior
because choice leads to confusion.
--
John McKown
Senior Systems Programmer
HealthMarkets
Keeping the Promise of Affordable Coverage
Administrative Services Group
Information Technology
The information contained in this e-mail message may be privileged
and/or confidential. It is for intended addressee(s) only. If you are
not the intended recipient, you are hereby notified that any disclosure,
reproduction, distribution or other use of this communication is
strictly prohibited and could, in certain circumstances, be a criminal
offense. If you have received this e-mail in error, please notify the
sender by reply and delete this message without copying or disclosing
it.
Howard Brazee
wrote:
>Paraphrasing: Having so many instructions will simply confuse the
>programmer! Better to have "one, true way" than many.
>
>http://www.oreillynet.com/onlamp/blog/2007/04/the_virtues_of_monoculture
>html
>
>This article is basically about why Microsoft is better than Linux in
>some ways. Linux has too many competing ways to do something whereas
>Microsoft is usually monolithic. And "lack of choice" is superior
>because choice leads to confusion.
That makes sense. But continuing that thought, I see Apple, which
doesn't try to make its OS be all things for all people (and hardware
manufacturers). Even if it *is* UNIX.
DASD...@ibm-main.lst
In a message dated 4/25/2007 1:41:31 P.M. Central Daylight Time,
>With the new instructions, I now count 751 instructions documented in the
POO. That's up a lot from the (IIRC) 143 for S/360.
pages. With a PDF version, it is very simple - scroll down to the bottom and
look at PDF's page number. Version -05 is 1218. I remember being very
pleased about seven years ago when I discovered that the first POO that documented
64-bit architecture and grande instructions had 1024 pages, an integral power
of two.
Bill Fairchild
Plainfield, IL
************************************** See what's free at http://www.aol.com.
CICS Guy
-----Original Message-----
From: IBM Mainframe Discussion List [mailto:IBM-...@ibm-main.lst
Of (IBM Mainframe Discussion List)
Sent: Wednesday, April 25, 2007 4:07 PM
To: IBM-...@BAMA.UA.EDU
Subject: Re: Latest Principles of Operation
In a message dated 4/25/2007 1:41:31 P.M. Central Daylight Time,
m42tom-...@YAHOO.COM writes:
>With the new instructions, I now count 751 instructions documented in the
POO. That's up a lot from the (IIRC) 143 for S/360.
I do something similar, but less time-consuming. I look at the number of
pages. With a PDF version, it is very simple - scroll down to the bottom
and
look at PDF's page number. Version -05 is 1218. I remember being very
pleased about seven years ago when I discovered that the first POO that
documented
64-bit architecture and grande instructions had 1024 pages, an integral
power
of two.
Bill Fairchild
Plainfield, IL
_________________________________________________________________
The average US Credit Score is 675. The cost to see yours: $0 by Experian.
http://www.freecreditreport.com/pm/default.aspx?sc=660600&bcd=EMAILFOOTERAVERAGE
Efinn...@ibm-main.lst
In a message dated 4/25/2007 3:19:22 P.M. Central Daylight Time,
bst...@HOTMAIL.COM writes:
Gee, the POO I use (paper) tops out at around 175 pages.......
>>
How does it check for floating point decimal?
************************************** See what's free at http://www.aol.com.
----------------------------------------------------------------------
CICS Guy
-----Original Message-----
From: IBM Mainframe Discussion List [mailto:IBM-...@ibm-main.lst
Of Ed Finnell
Sent: Wednesday, April 25, 2007 4:24 PM
To: IBM-...@BAMA.UA.EDU
Subject: Re: Latest Principles of Operation
In a message dated 4/25/2007 3:19:22 P.M. Central Daylight Time,
bst...@HOTMAIL.COM writes:
Gee, the POO I use (paper) tops out at around 175 pages.......
>>
How does it check for floating point decimal?
_________________________________________________________________
Need a break? Find your escape route with Live Search Maps.
http://maps.live.com/default.aspx?ss=Restaurants~Hotels~Amusement%20Park&cp=33.832922~-117.915659&style=r&lvl=13&tilt=-90&dir=0&alt=-1000&scene=1118863&encType=1&FORM=MGAC01
Anne & Lynn Wheeler
> That makes sense. But continuing that thought, I see Apple, which
> doesn't try to make its OS be all things for all people (and hardware
> manufacturers). Even if it *is* UNIX.
nominally the argument is that complexity contributes to confusion and
failures ... KISS is frequently better because it minimizes confusion
which can be major source of failures, vulnerabilities, threats and
exploits. however, another argument is that the solution paradigm has to
match the environment ... that there can be enormous amount of
complexity introduced when the solution paradigm is a mismatch for the
environment that it is being applied to.
slightly related thread discussing f18/f14, f16/f15, as well as f20
(with even a little computer related stuff sprinkled in) ... warning
quite a bit of thread drift ... even tho there was a lot of numerical
intensive computing that went into f16, f18, f20, etc:
http://www.garlic.com/~lynn/2007i.html#3 John W. Backus, 82, Fortran developer, dies
http://www.garlic.com/~lynn/2007i.html#4 John W. Backus, 82, Fortran developer, dies
http://www.garlic.com/~lynn/2007i.html#6 John W. Backus, 82, Fortran developer, dies
http://www.garlic.com/~lynn/2007i.html#7 John W. Backus, 82, Fortran developer, dies
http://www.garlic.com/~lynn/2007i.html#8 John W. Backus, 82, Fortran developer, dies
http://www.garlic.com/~lynn/2007i.html#10 John W. Backus, 82, Fortran developer, dies
Anne & Lynn Wheeler
> That makes sense. But continuing that thought, I see Apple, which
> doesn't try to make its OS be all things for all people (and hardware
> manufacturers). Even if it *is* UNIX.
re:
http://www.garlic.com/~lynn/2007i.html#25 Latest Principles of Operation
apple os (and next before it) starts out with a MACH "microkernel" basis
... could be considered striving for KISS ... somewhat like the original
CP67, as an extremely well focused microkernel. The morph from CP67 to
VM370 included work by people with much more of traditional operating
system training. Over the years, many found that the extreme simplicity
of the kernel made it easy to change/add/modify on an adhoc basis.
Unfortunately many such years of such adhoc approach to dealing with
something that was suppose to be a microkernel (rather than operating
system) ... eventually results in a lot of bloat and spaghetti code
past reference to comment about simple can be frequently much harder
than complex ... and it should be done with there is nothing left to
remove ... as opposed to it being done when there is nothing left to
add.
http://www.garlic.com/~lynn/2007h.html#29 sizeof() was: The Perfect Computer - 36 bits?
http://www.garlic.com/~lynn/2007h.html#30 sizeof() was: The Perfect Computer - 36 bits?
http://www.garlic.com/~lynn/2007i.html#5 John W. Backus, 82, Fortran developer, dies
more recently there have been comments about simple virtual machine
microkernels may be solution to various significant security issues in
the personal computing market place ... dynamically opening up a
traditional operating system in a "padded cell" virtual machine when it
needs to do various kinds of internet/network operations ... and then
collapsing and discarding most of the environment when finished.
lots of past posts referring to various microkernel activities:
http://www.garlic.com/~lynn/94.html#54 How Do the Old Mainframes
http://www.garlic.com/~lynn/95.html#0 pathlengths
http://www.garlic.com/~lynn/2000e.html#42 IBM's Workplace OS (Was: .. Pink)
http://www.garlic.com/~lynn/2001f.html#23 MERT Operating System & Microkernels
http://www.garlic.com/~lynn/2001h.html#11 checking some myths.
http://www.garlic.com/~lynn/2001j.html#36 Proper ISA lifespan?
http://www.garlic.com/~lynn/2001k.html#45 SMP idea for the future
http://www.garlic.com/~lynn/2001l.html#25 mainframe question
http://www.garlic.com/~lynn/2001m.html#47 TSS/360
http://www.garlic.com/~lynn/2003.html#50 Origin of Kerberos
http://www.garlic.com/~lynn/2003.html#60 MIDAS
http://www.garlic.com/~lynn/2003h.html#37 Does PowerPC 970 has Tagged TLBs (Address Space Identifiers)
http://www.garlic.com/~lynn/2003j.html#72 Microkernels are not "all or nothing". Re: Multics Concepts For
http://www.garlic.com/~lynn/2003k.html#5 What is timesharing, anyway?
http://www.garlic.com/~lynn/2003k.html#9 What is timesharing, anyway?
http://www.garlic.com/~lynn/2003k.html#24 Microkernels are not "all or nothing". Re: Multics Concepts For
http://www.garlic.com/~lynn/2003k.html#26 Microkernels are not "all or nothing". Re: Multics Concepts For
http://www.garlic.com/~lynn/2003k.html#27 Microkernels are not "all or nothing". Re: Multics Concepts For
http://www.garlic.com/~lynn/2003k.html#28 Microkernels are not "all or nothing". Re: Multics Concepts For
http://www.garlic.com/~lynn/2003k.html#30 IBM channels, was Re: Microkernels are not "all or nothing"
http://www.garlic.com/~lynn/2003k.html#37 Microkernels are not "all or nothing". Re: Multics Concepts For
http://www.garlic.com/~lynn/2005b.html#22 The Mac is like a modern day Betamax
http://www.garlic.com/~lynn/2005c.html#44 [Lit.] Buffer overruns
http://www.garlic.com/~lynn/2005c.html#56 intel's Vanderpool and virtualization in general
http://www.garlic.com/~lynn/2005c.html#63 intel's Vanderpool and virtualization in general
http://www.garlic.com/~lynn/2005f.html#10 Where should the type information be: in tags and descriptors
http://www.garlic.com/~lynn/2006p.html#10 What part of z/OS is the OS?
http://www.garlic.com/~lynn/2006p.html#11 What part of z/OS is the OS?
http://www.garlic.com/~lynn/2007g.html#70 The Perfect Computer - 36 bits?
http://www.garlic.com/~lynn/2007g.html#83 IBM to the PCM market
Clark Morris
>On Wed, 25 Apr 2007 14:16:47 -0500, Rick Fochtman wrote:
>
>>>
>>>With the new instructions, I now count 751 instructions documented in the
>>>POO. That's up a lot from the (IIRC) 143 for S/360.
>>>
>>
>>It's called evolution, son. Like it or not, we're stuck with it and it
>>can be a good thing, in spite of personnal opinions to the contrary.
>
>That wasn't a complaint. Actually, I do like it, even though it makes being an
>Assembler programmer more challenging. It's one of the many details about
>this platform that I find to be very impressive.
The frustrating thing is that there is no compiler switch to tell most
of the compilers that these instructions can be used because the
target is known. ISV's normally have to hang back on using these
instructions because the target processor may not have the
instruction.
Of course, my feeling is that COBOL is on life support with no
strategic direction.
Rick Fochtman
nominally the argument is that complexity contributes to confusion and
failures ... KISS is frequently better because it minimizes confusion
which can be major source of failures, vulnerabilities, threats and
exploits. however, another argument is that the solution paradigm has to
match the environment ... that there can be enormous amount of
complexity introduced when the solution paradigm is a mismatch for the
environment that it is being applied to.
That increased instruction set allows for vastly increased capability,
in spite of the perceived complexity. Simple applications can still be
coded using simple instructions, but more complex requirements can be
met more simply and efficiently by using some of those "added
instructions" that seem to lead to complexity.
Complexity is far too often used as an excuse for incompetence or
laziness; not always or even most of the time, but still far too often.
You don't let a carpenter into your house if he doesn't know how to use
his tools, do you????
Edward Jaffe
> The frustrating thing is that there is no compiler switch to tell most
> of the compilers that these instructions can be used because the
> target is known. ISV's normally have to hang back on using these
> instructions because the target processor may not have the
> instruction.
>
ARCH
http://publibz.boulder.ibm.com/cgi-bin/bookmgr_OS390/BOOKS/cbcug150/2.3.15.5
TUNE
http://publibz.boulder.ibm.com/cgi-bin/bookmgr_OS390/BOOKS/cbcug150/2.3.15.94
ISVs do indeed have to "hang back". And, so do IBM developers. The life
saver is knows as an "Architectural Level Set". Some customers don't
like them because they tend to obsolete affordable hardware processor
models. But, IBM and ISV developers like them because they allow use of
the new facilities. For example, IBM no longer supports anything less
than z/OS 1.6. And, z/OS 1.6 requires z/Architecture. This means that
IBM developers can use z/Architecture instructions for _all_ new code
targeted for z/OS without the need for dual-path, bifurcated, or "lowest
common denominator" instruction paths. The same holds true for any ISV
that follows IBM's support schedule.
--
Edward E Jaffe
Phoenix Software International, Inc
5200 W Century Blvd, Suite 800
Los Angeles, CA 90045
310-338-0400 x318
edj...@phoenixsoftware.com
http://www.phoenixsoftware.com/
John Eells
> That increased instruction set allows for vastly increased capability,
> in spite of the perceived complexity. Simple applications can still be
> coded using simple instructions, but more complex requirements can be
> met more simply and efficiently by using some of those "added
> instructions" that seem to lead to complexity.
>
> Complexity is far too often used as an excuse for incompetence or
> laziness; not always or even most of the time, but still far too often.
> You don't let a carpenter into your house if he doesn't know how to use
> his tools, do you????
I think it's worth noting that subsystems and compilers (and of
course interpreters) can hide the underlying complexity. The
human-facing interface need not be more complex for the
underlying architecture to have better or more function.
--
John Eells
z/OS Technical Marketing
IBM Poughkeepsie
ee...@us.ibm.com
Jeffrey D. Smith
> From: IBM Mainframe Discussion List [mailto:IBM-...@ibm-main.lst
> Behalf Of Edward Jaffe
> Sent: Thursday, April 26, 2007 9:45 AM
> To: IBM-...@BAMA.UA.EDU
> Subject: Re: Latest Principles of Operation
>
> > The frustrating thing is that there is no compiler switch to tell most
> > of the compilers that these instructions can be used because the
> > target is known. ISV's normally have to hang back on using these
> > instructions because the target processor may not have the
> > instruction.
> >
>
>
> ISVs do indeed have to "hang back". And, so do IBM developers. The life
> saver is knows as an "Architectural Level Set". Some customers don't
> like them because they tend to obsolete affordable hardware processor
> models. But, IBM and ISV developers like them because they allow use of
> the new facilities. For example, IBM no longer supports anything less
> than z/OS 1.6. And, z/OS 1.6 requires z/Architecture. This means that
> IBM developers can use z/Architecture instructions for _all_ new code
> targeted for z/OS without the need for dual-path, bifurcated, or "lowest
> common denominator" instruction paths. The same holds true for any ISV
> that follows IBM's support schedule.
>
> --
> Edward E Jaffe
Unfortunately, there are still some customers (at least one that is a
"big name" in the corporate banking world) that uses a 31-bit Amdahl
box with OS/390 R10. I had to write my crypto stuff using some dual
pathing to accommodate them. sigh...
Jeffrey D. Smith
Principal Product Architect
Farsight Systems Corporation
700 KEN PRATT BLVD. #204-159
LONGMONT, CO 80501-6452
303-774-9381 direct
303-484-6170 FAX
http://www.farsight-systems.com/
see my résumé at my website (yes, I am looking for employment)
David Andrews
> The frustrating thing is that there is no compiler switch to tell most
> of the compilers that these instructions can be used
The IBMers contributing to GCC have been doing pretty well on this
front. GCC 4.1 uses 34 new instructions for z9 processors.
--
David Andrews
A. Duda and Sons, Inc.
david....@duda.com
Mark H. Young
<jeffre...@EARTHLINK.NET> wrote:
>> -----Original Message-----
>> From: IBM Mainframe Discussion List [mailto:IBM-...@ibm-main.lst
>> Behalf Of Edward Jaffe
>> Sent: Thursday, April 26, 2007 9:45 AM
>> To: IBM-...@BAMA.UA.EDU
>> Subject: Re: Latest Principles of Operation
>Unfortunately, there are still some customers (at least one that is a
>"big name" in the corporate banking world) that uses a 31-bit Amdahl
>box with OS/390 R10. I had to write my crypto stuff using some dual
>pathing to accommodate them. sigh...
>
>Jeffrey D. Smith
>Principal Product Architect
>Farsight Systems Corporation
>700 KEN PRATT BLVD. #204-159
>LONGMONT, CO 80501-6452
>303-774-9381 direct
>303-484-6170 FAX
>http://www.farsight-systems.com/
>see my résumé at my website (yes, I am looking for employment)
Speaking of dual pathing.....that brings to mind something I could not
remember recently about 3380 geometry DASD that had "internal pathing"???
Remember those? Anyone know what the device was called from IBM?
Like maybe the model name et al?? I'm experiencing my daily Alzheimer's spell.
We had them back in the day (mid-1980s) at Marriott Corp. in Bethesda, MD,
and had to consider what we placed where so it didn't cause performance
problems.
THANX,
Mark H. Young
Patrick...@ibm-main.lst
was told we had them and most IBM DASD at that time, 1979 or so, at NVIP
but I was not aware of all we had back then. I think the IBM 3380 had 4
internal path selection.
"Mark H. Young" <Mark....@FAIRFAXCOUNTY.GOV>
Sent by: IBM Mainframe Discussion List <IBM-...@BAMA.UA.EDU>
04/26/2007 12:10 PM
Please respond to
IBM Mainframe Discussion List <IBM-...@BAMA.UA.EDU>
Subject
Re: Latest Principles of Operation
Anne & Lynn Wheeler
> That increased instruction set allows for vastly increased capability,
> in spite of the perceived complexity. Simple applications can still be
> coded using simple instructions, but more complex requirements can be
> met more simply and efficiently by using some of those "added
> instructions" that seem to lead to complexity.
>
> Complexity is far too often used as an excuse for incompetence or
> laziness; not always or even most of the time, but still far too
> often. You don't let a carpenter into your house if he doesn't know
> how to use his tools, do you????
re:
http://www.garlic.com/~lynn/2007i.html#25 Latest Principles of Operation
http://www.garlic.com/~lynn/2007i.html#26 Latest Principles of Operation
well, i remember the hassle to get compare&swap instruction into the 370
architecture. Charlie had invented compare&swap instruction when he was
working on smp knerel fine-grain locking for cp67 at the science center
http://www.garlic.com/~lynn/subtopic.html#545tech
the "redbook" 370 architecture owners claimed that everybody (namely the
POK favorite son operating system people) thot that test&set was totally
adequate for all multiprocessor support. in order to justify getting
compare&swap instruction into 370 architecture, we had to come up with a
whole boatload of justifications for compare&swap instruction that
wasn't specific to multiprocessor operation. thus was born the stuff in
principle of operations about using compare&swap instruction for
application multithreaded operation (regardless of whether or not it was
multiprocessor environment). lots of past posts mentioning
multiprocessor support and/or compare&swap instruction
http://www.garlic.com/~lynn/subtopic.html#smp
it is sometimes relative. i've claimed that John came up with risc/801
as part of going to the opposite extreme after the extremely complex
future system project failed ... misc. past posts mentioning failed
future system project
http://www.garlic.com/~lynn/subtopic.html#futuresys
and lots of past post mentioning 801, romp, rios, iliad, fort knox,
somerset, power/pc, etc
http://www.garlic.com/~lynn/subtopic.html#801
and old email with 801 references
http://www.garlic.com/~lynn/lhwemail.html#801
Supposedly future system project was a countermeasure to clone
controllers ... something I got (at least partially) blamed for from a
project that I worked on as undergraduate in the 60s producing our own
clone controller
http://www.garlic.com/~lynn/subtopic.html#360pcm
also some "FS" quotes referenced in this post
http://www.garlic.com/~lynn/2007f.html#28 The Perfect Computer - 36 bits?
from article by former executive here
http://www.ecole.org/Crisis_and_change_1995_1.htm
The FS effort and subsequent failure ... can also be considered as
contributing to the uptake of clone processors (in part because of the
dearth of items in the 370 product pipeline) a few recent posts
http://www.garlic.com/~lynn/2007g.html#55 IBM to the PCM market(the sky is falling!!!the sky is falling!!)
http://www.garlic.com/~lynn/2007g.html#57 IBM to the PCM market(the sky is falling!!!the sky is falling!!)
http://www.garlic.com/~lynn/2007g.html#59 IBM to the PCM market(the sky is falling!!!the sky is falling!!)
and after the failure of the FS project ... there was a rush to get
stuff back into the 370 product pipeline. Recent post attributing that
as big part of the reason that the product group shipped so much of my
code (since I continued to develop 370-based software all thru the FS
activity)
http://www.garlic.com/~lynn/2007i.html#21 John W. Backus, 82, Fortran developer, dies
and another recent posting touching on some stuff that went on in the FS era
http://www.garlic.com/~lynn/2007i.html#20 Does anyone know of a documented case of VM being penetrated by hackers?
http://www.garlic.com/~lynn/2007i.html#29 Does anyone know of a documented case of VM being penetrated by hackers?
another stop-gap effort to try and quickly fill the 370 product pipeline
was 3033x (after failure of FS project) was 303x. The standard processor
development cycle was 7-8yrs ... and they needed to get something out
much quicker than that ... since starting the 370-xa/3081 wouldn't be
out before the early 80s.
So they took they 370/158 microcode engine ... and stripped out the 370
microcode support ... leaving just the integrated channel microcode and
packaged it as the 303x "channel director". Then the 3031 was a 370/158
microcode engine with just the 370 microcode support (and no integrated
channel microcode) paired with a "channel director". The 3032 was a
370/168-3 repackaged to work with "channel director". The 3033 started
out simply being 168 wiring diagram mapped to newer chip technology.
Recent posts about enormous effort to hurry up and get 303x out the
door after failed FS project:
http://www.garlic.com/~lynn/2007d.html#21 How many 36-bit Unix ports in the old days?
http://www.garlic.com/~lynn/2007d.html#62 Cycles per ASM instruction
http://www.garlic.com/~lynn/2007e.html#32 I/O in Emulated Mainframes
http://www.garlic.com/~lynn/2007e.html#40 FBA rant
http://www.garlic.com/~lynn/2007f.html#28 The Perfect Computer - 36 bits?
http://www.garlic.com/~lynn/2007f.html#73 Is computer history taught now?
http://www.garlic.com/~lynn/2007g.html#17 The Perfect Computer - 36 bits?
http://www.garlic.com/~lynn/2007g.html#23 The Perfect Computer - 36 bits?
http://www.garlic.com/~lynn/2007g.html#29 The Perfect Computer - 36 bits?
http://www.garlic.com/~lynn/2007g.html#57 IBM to the PCM market(the sky is falling!!!the sky is falling!!)
http://www.garlic.com/~lynn/2007h.html#1 21st Century ISA goals?
the other aftermath of the FS project failure was that POK lab convinced
corporate to shutdown the VM development group, decommit the product,
and transfer all the people to POK in order to work on MVS/XA ... also
attempting to shorten the typical development time. Endicott was
eventually able to salvage the product mission and one or two of the
people.
Clark Morris
wrote:
>Clark Morris wrote:
>> The frustrating thing is that there is no compiler switch to tell most
>> of the compilers that these instructions can be used because the
>> target is known. ISV's normally have to hang back on using these
>> instructions because the target processor may not have the
>> instruction.
>>
>
>ARCH
>http://publibz.boulder.ibm.com/cgi-bin/bookmgr_OS390/BOOKS/cbcug150/2.3.15.5
>
>TUNE
>http://publibz.boulder.ibm.com/cgi-bin/bookmgr_OS390/BOOKS/cbcug150/2.3.15.94
Now would it make a difference in COBOL performance if COBOL had those
options and if so why doesn't it. Of course the abysmal performance
of TRUNC(BIN) was due to incredibly bad programming decisions (convert
to and from decimal for the actual arithmetic). And then there is the
ridiculous IEEE to hex floating point conversion done for Java
interface even though there is a CLEAN way to define IEEE floating
point using the 2002 standard.
Clark Morris
----------------------------------------------------------------------
Richards.Bob
DASD was 3380-D, 3380-E, 3380-J, 3380-K, 3390-1, 3390-2, 3390-3 (today's
major image) and 3390-9 (a true SLED pulled by crippled and maimed dogs)
Yes, the 3380s had four paths. Do not recall if it was ever increased.
Bob Richards
-----Original Message-----
From: IBM Mainframe Discussion List [mailto:IBM-...@ibm-main.lst
Behalf Of Patrick...@COMBINED.COM
IBM 3350A2/B2 modules on IBM 3830 Controller? Not sure about IBM 3375, I
was told we had them and most IBM DASD at that time, 1979 or so, at NVIP
but I was not aware of all we had back then. I think the IBM 3380 had 4
internal path selection.
"Mark H. Young" <Mark....@FAIRFAXCOUNTY.GOV>
Speaking of dual pathing.....that brings to mind something I could not
remember recently about 3380 geometry DASD that had "internal
pathing"???
Remember those? Anyone know what the device was called from IBM?
Like maybe the model name et al?? I'm experiencing my daily Alzheimer's
spell.
We had them back in the day (mid-1980s) at Marriott Corp. in Bethesda,
MD,
and had to consider what we placed where so it didn't cause performance
problems.
LEGAL DISCLAIMER
The information transmitted is intended solely for the individual or entity to which it is addressed and may contain confidential and/or privileged material. Any review, retransmission, dissemination or other use of or taking action in reliance upon this information by persons or entities other than the intended recipient is prohibited. If you have received this email in error please contact the sender and delete the material from any computer.
SunTrust and Seeing beyond money are federally registered service marks of SunTrust Banks, Inc.
[ST:XCL]
Anne & Lynn Wheeler
Bob.Ri...@ibm-main.lst (Richards.Bob) writes:
> 3880-xx Controller series followed by the 3990-xxx series
>
> DASD was 3380-D, 3380-E, 3380-J, 3380-K, 3390-1, 3390-2, 3390-3 (today's
> major image) and 3390-9 (a true SLED pulled by crippled and maimed dogs)
>
> Yes, the 3380s had four paths. Do not recall if it was ever increased.
3830 disk controller had dual channel along with "string switch" for
3330 ... allowing for four paths total (i.e. string connecting to two
different controllers, each of which could connect to two different
channels)
3880 disk controller had four channel interfaces ... and 3380 "string
switch" (a-box) allowed 3380 string to be connected to two different
controllers for eight total paths.
Evolution of the DASD storage control
http://www.research.ibm.com/journal/sj/282/ibmsj2802C.pdf
not to be too harsh about comment in the above about 3880 having better
performance ... but some recent posts that touched on getting 3880 out
the door ... as well as little discussion of dynamic pathing
http://www.garlic.com/~lynn/2007h.html#1 21st Century ISA goals?
http://www.garlic.com/~lynn/2007h.html#3 21st Century ISA goals?
http://www.garlic.com/~lynn/2007h.html#4 21st Century ISA goals?
http://www.garlic.com/~lynn/2007h.html#5 21st Century ISA goals?
http://www.garlic.com/~lynn/2007h.html#6 21st Century ISA goals?
http://www.garlic.com/~lynn/2007h.html#9 21st Century ISA goals?
IBM Jargon:
A-Box n. A primary storage unit: the one closest to the controller.
Most 370 storage peripherals come in two flavours. The A-Box (also
called head of String, or Model A) either houses the controller
(e.g., 3422), is the controller (e.g., 3480), or connects to the
controller. The B-Box (Model B) is used for extending the string.
Some strings can be connected to the controller at both ends, in
which case the unit at the end of the string is usually a Model D.
... snip ...
3380 "history" here with A & B boxes:
http://www-03.ibm.com/ibm/history/exhibits/storage/storage_3380.html
http://www-03.ibm.com/ibm/history/exhibits/storage/storage_3380b.html
http://www-03.ibm.com/ibm/history/exhibits/storage/storage_3380c.html
minor mention of a-box, head-of-string function:
http://www-03.ibm.com/ibm/history/exhibits/storage/storage_3380d.html
http://www-03.ibm.com/ibm/history/exhibits/storage/storage_3380e.html
from above:
A string of Standard model 3380s can consist of a single 3380 Model A04
or AA4 and up to three 3380 Model B04 units. A string of Extended
Capability 3380s can consist of one 3380 Model AD4 or AE4 and up to
three 3380 Model BD4 and BE4 units, in any combination. Two strings of
3380s can be attached to each storage director of a 3880 Model 3 or
cache storage control Model 23. An A04 string is not supported by the
Model 23. Strings headed by an AA4, AD4 or AE4 must attach to two
storage directors (usually on two separate 3880 Storage Controls). An
AA4 string and an Extended Capability string can be attached to the same
storage directors.
... snip ...
at one point i made a forey into redoing the (3880) controller dynamic
pathing architecture implementation ... as part of significantly
extending & simplifying ("dynamic pathing simplification") capability
for virtualized operation ... but got into all sort of issues regarding
compatibility with implementation already in the field.
Howard Brazee
<ly...@garlic.com> wrote:
A bit off topic: I find your input to these threads in general to be
quite useful. They appear to take a fair amount of work - are they
part of your jobs, or is this just a style you are comfortable with?
Anne & Lynn Wheeler
> A bit off topic: I find your input to these threads in general to be
> quite useful. They appear to take a fair amount of work - are they
> part of your jobs, or is this just a style you are comfortable with?
re:
http://www.garlic.com/~lynn/2007i.html#33 Internal DASD Pathing
i had done some amount of the semi-automated computer conferencing
via email (copy list) mechanism on the internal network
http://www.garlic.com/~lynn/subnetwork.html#internalnet
in the late 70s and early 80s ... and got blaimed for something called
"tandem memos" (there was even a nov81 datamation article about tandem
memos). tandemo memos were somewhat spawned by various trip reports
visiting Tandem after Jim left SJR. somewhat recent postings referencing
Jim and his departure from SJR:
http://www.garlic.com/~lynn/2007.html#1 "The Elements of Programming Style"
http://www.garlic.com/~lynn/2007.html#13 "The Elements of Programming Style"
http://www.garlic.com/~lynn/2007d.html#17 Jim Gray Is Missing
somewhat as an outcome of various investigations into the "tandem memo"
phenomena ... there were decisions to deploy officially sanctioned
corporate computer conferencing capability.
there was also a researcher hired that sat in the back of my office for
nine months and studied how i communicated; phone, face-to-face, email,
instant messeging, etc. they had copies of my incoming and outgoing
email and logs of all instant messages. The resulting report was also
published as a stanford phd thesis (joint between ai and language) and
the subject matter for some number of papers and books. some related
posts mentioning computer mediated conferencing
http://www.garlic.com/~lynn/subtopic.html#cmc
never got paid for it ... frequently closer to the opposite ... minor
reference:
http://www.garlic.com/~lynn/2007e.html#48 time spent/day on a computer
there were sometimes jokes in the mid-80s about various subject threads
having gotten "wheeler'ized" ... i.e. when possibly 80-90% percent of
the characters in some corporate/world wide discussion were found to
originate from my keyboard. most consider that i have extremely mellowed
since then ... and have been doing it for 30 yrs or so
somewhat drifting back to the original topic ... misc. collected
posts about getting to play in the disk engineering (bldg. 14)
and disk product test (bldg. 15) labs.
http://www.garlic.com/~lynn/subtopic.html#disk
Steve Comstock
> The following message is a courtesy copy of an article
> that has been posted to bit.listserv.ibm-main,alt.folklore.computers as well.
Ted MacNEIL
You forgot the old 3380-1 (the original).
It had four on a stick and you could have internal path contention.
>Yes, the 3380s had four paths. Do not recall if it was ever increased.
D/E had dual path.
J/K had quad. It was never increased.
-
Too busy driving to stop for gas!
Anne & Lynn Wheeler
minor addenda ... to previous topic drift ... this index of old email
http://www.garlic.com/~lynn/lhwemail.html
also contains a URL for an eserver (since renamed IBM Systems) magazine
article that appeared two yrs ago ... although they got some of the
details slightly garbled. I actually haven't seen the copy of the
physical magazine ... they had sent a photographer to the house for
photo shoot ... and something from that supposedly shows up in the
magazine.
Scott Blackledge
What is MVCOS? (Where did you hear about it?)
Scott
Edward Jaffe
> What is MVCOS? (Where did you hear about it?)
>
http://shareew.prod.web.sba.com/client_files/callpapers/attach/SHARE_in_Tampa_Bay/S8162JE123550.pdf
--
Edward E Jaffe
Phoenix Software International, Inc
5200 W Century Blvd, Suite 800
Los Angeles, CA 90045
310-338-0400 x318
edj...@phoenixsoftware.com
http://www.phoenixsoftware.com/
----------------------------------------------------------------------
John P Baker
POP.
John P Baker
> -----Original Message-----
> From: IBM Mainframe Discussion List [mailto:IBM-...@ibm-main.lst
> Behalf Of Edward Jaffe
> Sent: Thursday, April 26, 2007 7:01 PM
> To: IBM-...@BAMA.UA.EDU
> Subject: Re: Latest Principles of Operation
>
Edward Jaffe
> It is disturbing that the MVCOS instruction is not described in the latest
> POP.
>
And, quite surprising.
Bob Rutledge
> John P Baker wrote:
>> It is disturbing that the MVCOS instruction is not described in the
>> latest
>> POP.
>>
>
> And, quite surprising.
>
developers had been using it.
Bob
Edward Jaffe
> Edward Jaffe wrote:
>> John P Baker wrote:
>>> It is disturbing that the MVCOS instruction is not described in the
>>> latest
>>> POP.
>>>
>>
>> And, quite surprising.
>>
> Not really--CSP didn't make it into the POP until some time after the
> Linux developers had been using it.
MVCOS has been discussed for quite some time now and in highly visible
settings. For example, by Bob Rogers at SHARE in Baltimore and by John
Dravnieks at SHARE in Tampa. IBM had plenty of time to put this in the
book. It *should* be there!
--
Edward E Jaffe
Phoenix Software International, Inc
5200 W Century Blvd, Suite 800
Los Angeles, CA 90045
310-338-0400 x318
edj...@phoenixsoftware.com
http://www.phoenixsoftware.com/
----------------------------------------------------------------------
Quadibloc
> > That increased instruction set allows for vastly increased capability,
> > in spite of the perceived complexity. Simple applications can still be
> > coded using simple instructions, but more complex requirements can be
> > met more simply and efficiently by using some of those "added
> > instructions" that seem to lead to complexity.
> >
> > Complexity is far too often used as an excuse for incompetence or
> > laziness; not always or even most of the time, but still far too
> > often. You don't let a carpenter into your house if he doesn't know
> > how to use his tools, do you????
> well, i remember the hassle to get compare&swap instruction into the 370
> architecture. Charlie had invented compare&swap instruction when he was
> working on smp knerel fine-grain locking for cp67 at the science center
> http://www.garlic.com/~lynn/subtopic.html#545tech
For amusement, a grossly complex instruction set architecture for a
computer is described on the pages starting at
http://www.quadibloc.com/arch/arcint.htm
It started out as a simple architecture, inspired by the System/360
and the Motorola 68020, but features were added in order to explain
various computer concepts.
On the page
http://www.quadibloc.com/arch/ar2335.htm
the new decimal floating point added on the z9000 is included; but I
had to modify the format to work with other word lengths as well.
John Savard
Dean Kent
news:4630C814...@ync.net...
> ----------------------<snip>---------------------------
>
> Complexity is far too often used as an excuse for incompetence or
> laziness; not always or even most of the time, but still far too often.
> You don't let a carpenter into your house if he doesn't know how to use
> his tools, do you????
The human mind has a limited capacity for organizing information into
something meaningful. It would be interesting to see a graph showing the
percentage of people able to properly handle various levels of complexity,
and I suspect that as the complexity increases the dropoff would become more
and more dramatic vs. something more linear. Therefore, rather than an
excuse, concern about complexity would seem to be a very real one. Were
the carpenter's tools to be come so complex that only the most talented of
people could use them, carpentry would likely become so expensive as to
eventually become a lost art, or at best an extremely arcane one. Do you
disagree?
Regards,
Dean
Howard Brazee
>The human mind has a limited capacity for organizing information into
>something meaningful. It would be interesting to see a graph showing the
>percentage of people able to properly handle various levels of complexity,
>and I suspect that as the complexity increases the dropoff would become more
>and more dramatic vs. something more linear. Therefore, rather than an
>excuse, concern about complexity would seem to be a very real one. Were
>the carpenter's tools to be come so complex that only the most talented of
>people could use them, carpentry would likely become so expensive as to
>eventually become a lost art, or at best an extremely arcane one. Do you
>disagree?
A lot of our smarts is in seeing patterns, simplifying what we are
looking for. Occasionally this kind of shortcut causes us to miss
things, but pattern recognition allows the chess master to ignore dead
ends that poorer players waste time on.
I see craftsmen and artists using tools that are difficult to master -
but supply and demand doesn't take that into consideration in setting
prices for their goods.
Dean Kent
news:h1e43315tmst7fr0v...@4ax.com...
>
> A lot of our smarts is in seeing patterns, simplifying what we are
> looking for. Occasionally this kind of shortcut causes us to miss
> things, but pattern recognition allows the chess master to ignore dead
> ends that poorer players waste time on.
>
> I see craftsmen and artists using tools that are difficult to master -
> but supply and demand doesn't take that into consideration in setting
> prices for their goods.
You are simplifying a complex subject and coming to an invalid conclusion, I
think. :-).
If one were to add one new square to the chessboard every few years and a
new piece on occasion, the number of people able to master the game would
diminish each time. At some point nobody would actually be able to master
the game. One might say that mastery is a relative thing, and those who
play best would still be masters, but you would still be reducing the number
of individuals actually attempting to master it. I would also point out
that if you *removed* a square on occasion, along with a piece, you would
start seeing a much greater number of people who can master the game, and
likely a greater number of people actually attempting it... thereby making
the value of a master much less. Which one is desireable depends entirely
on your objective - do you want to get more people using your game boards,
or do you want to make it a real achievement to be a master?
As far as supply and demand goes, prices are most certainly set by how many
people are able to master the skill. If anyone could build nice cabinets
because the tools and techniques were relatively simple, carpenters would
not be able to charge much at all - as the supply of 'skilled' carpenters
would equal or exceed the demand. Conversely, if the tools and techniques
are so difficult that only a very few could build something considered
reasonable quality, those few would be able to demand extremely high prices
that only the very wealthy could afford.
Having a child who is a figure skater, I can say that if the requirement to
win becomes a quintuple axle there are going to be far fewer individuals
able to compete unless the training techniques and tools allow the skills to
be learned more easily. As it stands, only the wealthiest (or most
fortunate by way of sponsorship) of people can afford to have their children
learn the skills to compete in this sport. It is likely one of the most
complex and difficult of skills, requiring 15-20 hours or more per week of
practice for many years by *kids*, expensive and specialized equipment, and
hours of personalized coaching. I daresay that this was almost certainly
not required when a double axle was the greatest skill level necessary.
The number of people that start has increased because the financial rewards
have increased dramatically - but the attrition rate is astounding and few
actually acquire the skills necessary to continue competing after the age of
13.
One might say that this is a good thing in an environment where you want to
limit the participants and only end up with the cream of the crop. But, if
you are in an environment where you actually wish to *expand* the use (as
IBM might wish, or those who want to have job opportunities might wish),
making it more complex seems to be the wrong way to go, if you have the
choice.
Anne & Lynn Wheeler
> A lot of our smarts is in seeing patterns, simplifying what we are
> looking for. Occasionally this kind of shortcut causes us to miss
> things, but pattern recognition allows the chess master to ignore dead
> ends that poorer players waste time on.
>
> I see craftsmen and artists using tools that are difficult to master -
> but supply and demand doesn't take that into consideration in setting
> prices for their goods.
long post warning ...
as an undergraduate ... i did fair share scheduler for cp67 ... or
actually dynamic adaptive resource management ... with default
policy being fair share; minor recent reference:
http://www.garlic.com/~lynn/2007h.html#77 Linux: The Completely Fair Scheduler
but also did a lot of highly optimized pathlength and fastpath stuff. I
made joke about being able to do stuff in zero instruction ... carefully
tweaking instructions all over the kernel so that various things
happened implicitly ... so scheduler didn't actually have to execute
instructions to obtain the desired results (secondary effects of the
order of how other things are occruing, of course it helps if you
effectively have memorized the source for the complete kernel).
over some yrs, there was complaints that nobody could understand how it
worked ... which probably contributed to a lot of being dropped
(simplified) in the morph from cp67 to vm370.
so possibly still part of the recovery in the aftermath of future system
project ... recent reference in this thread
http://www.garlic.com/~lynn/2007i.html#31 Latest Principles of Operation
I was given the opportunity to reintroduce the resource manager for
vm370.
... more topic drift ...
and they decided that the resource manager should be guinea pig to
starting to charge for kernel code ... so i had to spend some amount
of time with the business and legal people working on policy for charging
for kernel software; i.e. 23jun69 unbundling announcement started
charging for application software ... but they used the excuse that
kernel software was integral to hardware operation and should still
be "free" ... misc. past posts
http://www.garlic.com/~lynn/subtopic.html#unbundle
... slightly return to topic ...
now, i did fix up some of the more obtuse pieces of code ... but there
still was quite a bit of complexity in the resource manager ... and
people over the yrs would complain that they didn't understand how it
worked ... and periodically somebody would make changes in other parts
of the kernel resulting in unexplicable affects on scheduling (there was
still quite a bit of convoluted code that I justified on being able to
do things in fewer instructions and shorter pathlength).
so there was one fly in the oitment, somebody from corporate complained
that there weren't any tuning parameters ... which was the state of the
art in other products; ... namely the favorite son operating system of
the period had a massive table of tuning parameters ... and there were
this presentations at SHARE (could put everybody to sleep) about random
walks across the tuning parameter table landscape, varying the
parameters for all different kinds of workloads and configurations ...
attempting to find settings that made some difference.
So I was forced to put in some tuning parameters (before product ship),
document them and the backup formulas ... as well as detailed code
explanation.
so we roll forward nearly 20yrs ... we are on a marketing swing thru
the far east for our ha/cmp product
http://www.garlic.com/~lynn/subtopic.html#hacmp
also additional thread drift
http://www.garlic.com/~lynn/95.html#13
http://www.garlic.com/~lynn/96.html#15
and
http://www.garlic.com/~lynn/lhwemail.html#medusa
... and we are on a call at a large financial institution. One of the
people at the meeting (relatively recent graduate) pipes up and asks if
i'm the same person associated with the scheduler ... since they had
studied me/it at the Univ. of Waterloo.
So I respond politely and asked if the "joke" was discussed.
Now part of the issue with static tuning parameters is that workloads
tend to vary from minute-to-minute, day-to-day, week-to-week ... and
there was a whole lot of work that went into the dynamic adaptive
implementation to eliminate the requirement for any static tuning
parameters. Being force to add static tuning parameters seemed to be a
great step backward in the state-of-the-art. So as to the "joke" ... if
dynamic adaptive can adjust for changes in configurations and workloads
... as well as a lot of other things ... why couldn't the dynamic
adaptive implementation also adjust to compensate for changes in any
static tuning parameters.
misc. past posts discussing the resource manager "joke":
http://www.garlic.com/~lynn/2001b.html#18 Linux IA-64 interrupts [was Re: Itanium benchmarks ...]
http://www.garlic.com/~lynn/2001l.html#9 mainframe question
http://www.garlic.com/~lynn/2002c.html#16 OS Workloads : Interactive etc
http://www.garlic.com/~lynn/2002c.html#54 Swapper was Re: History of Login Names
http://www.garlic.com/~lynn/2002i.html#53 wrt code first, document later
http://www.garlic.com/~lynn/2004o.html#10 Multi-processor timing issue
http://www.garlic.com/~lynn/2005p.html#31 z/VM performance
http://www.garlic.com/~lynn/2006b.html#21 IBM 3090/VM Humor
http://www.garlic.com/~lynn/2006h.html#22 The Pankian Metaphor
http://www.garlic.com/~lynn/2006y.html#17 The Future of CPUs: What's After Multi-Core?
http://www.garlic.com/~lynn/2007g.html#56 The Perfect Computer - 36 bits?
Quadibloc
> the new decimal floating point added on the z9000 is included;
That should, of course, have been the z9.
John Savard