Instruction Set Evolution 
Author Message
 Instruction Set Evolution

One interesting characteristic of IBM is that once they have assigned
an op-code to a mnemonic, it stays that way.  BAS, for example, goes
back to the 360/20, and was assigned a different op-code to carry the
idea of link with address only, rather than link with address + ILC +
interrupt mask.

So, LRA was assigned in the 360/67 days, and continued on unchanged to the
present day.


Quote:

>>The same customer, who had 360/67 experience, noticed that the emulator
>>used a LRA instruction so the unannounced virtual addressing was out of
>>the bag.  Of course the control panel that displayed both real and
>>logical addresses was another hint.

>OK.... How does one figure this out.  The emulation of ICM & STCM would be easy
>to understand.  If you're looking at the emulator and you see the code
>necessary for the emulation of an new op-code (in this case, LRA) , how can you
>know your're looking at a page table if you don't know what one does?

-- Steve Myers

The E-mail addresses in this message are private property.  Any use of them
to  send  unsolicited  E-mail  messages  of  a  commerical  nature  will be
considered trespassing,  and the originator of the message will be  sued in
small claims court in Camden County,  New Jersey,  for the  maximum penalty
allowed by law.



Thu, 25 Jan 2001 03:00:00 GMT  
 Instruction Set Evolution
I thought that the BAS insruction on the 360/20 was called that to
differentiate it from the BAL instruction on other machines because
the 360/20 used halfword registers rather than fullword registers.

Judy Anderson
Product Development
Advanced Software Technologies Company, Ltd.


Quote:
>One interesting characteristic of IBM is that once they have assigned
>an op-code to a mnemonic, it stays that way.  BAS, for example, goes
>back to the 360/20, and was assigned a different op-code to carry the
>idea of link with address only, rather than link with address + ILC +
>interrupt mask.

>So, LRA was assigned in the 360/67 days, and continued on unchanged to the
>present day.


(LakeWin98) writes:

>>>The same customer, who had 360/67 experience, noticed that the emulator
>>>used a LRA instruction so the unannounced virtual addressing was out of
>>>the bag.  Of course the control panel that displayed both real and
>>>logical addresses was another hint.

>>OK.... How does one figure this out.  The emulation of ICM & STCM would be
easy
>>to understand.  If you're looking at the emulator and you see the code
>>necessary for the emulation of an new op-code (in this case, LRA) , how
can you
>>know your're looking at a page table if you don't know what one does?

>-- Steve Myers

>The E-mail addresses in this message are private property.  Any use of them
>to  send  unsolicited  E-mail  messages  of  a  commerical  nature  will be
>considered trespassing,  and the originator of the message will be  sued in
>small claims court in Camden County,  New Jersey,  for the  maximum penalty
>allowed by law.



Thu, 25 Jan 2001 03:00:00 GMT  
 Instruction Set Evolution

Quote:

> >The same customer, who had 360/67 experience, noticed that the emulator
> >used a LRA instruction so the unannounced virtual addressing was out of
> >the bag.  Of course the control panel that displayed both real and
> >logical addresses was another hint.

> OK.... How does one figure this out.  The emulation of ICM & STCM would be easy
> to understand.  If you're looking at the emulator and you see the code
> necessary for the emulation of an new op-code (in this case, LRA) , how can you
> know your're looking at a page table if you don't know what one does?

The LRA instruction was never emulated on the 370.  The
emulator/simulator itself made use of the instruction.  The LRA behaved
the same way that it did on the earlier 360/67 which did have page
tables.  That's how the customer, who had previously used a 360/67,
figured it out.

Bob



Thu, 25 Jan 2001 03:00:00 GMT  
 Instruction Set Evolution
In terms of when the Model 20 was first introduced, yes, I agree with Judy.
Underlying the concept, though, is that the entire link register was
devoted to the link address, whether 16 bit (as in the model 20) or 32 bit,
(as in the case of Extended Architecture), rather than the case with the
BAL, where 8 bits of the register were devoted to non address information.

Does anyone know if the 360/67 implemented BAS?  I very vaguely
recall that it did, but this memory is from examnination of Assembler G
op code tables, and it could be faulty.


Quote:
>I thought that the BAS insruction on the 360/20 was called that to
>differentiate it from the BAL instruction on other machines because
>the 360/20 used halfword registers rather than fullword registers.

>Judy Anderson
>Product Development
>Advanced Software Technologies Company, Ltd.

[snip]

-- Steve Myers

The E-mail addresses in this message are private property.  Any use of them
to  send  unsolicited  E-mail  messages  of  a  commerical  nature  will be
considered trespassing,  and the originator of the message will be  sued in
small claims court in Camden County,  New Jersey,  for the  maximum penalty
allowed by law.



Thu, 25 Jan 2001 03:00:00 GMT  
 Instruction Set Evolution

Quote:

>In terms of when the Model 20 was first introduced, yes, I agree with Judy.
>Underlying the concept, though, is that the entire link register was
>devoted to the link address, whether 16 bit (as in the model 20) or 32 bit,
>(as in the case of Extended Architecture), rather than the case with the
>BAL, where 8 bits of the register were devoted to non address information.

>Does anyone know if the 360/67 implemented BAS?  I very vaguely
>recall that it did, but this memory is from examnination of Assembler G
>op code tables, and it could be faulty.

GX20-1703-9 lists BAS and BASR, for "Model 67".

Andy Wood



Thu, 25 Jan 2001 03:00:00 GMT  
 Instruction Set Evolution

Quote:

>>The same customer, who had 360/67 experience, noticed that the emulator
>>used a LRA instruction so the unannounced virtual addressing was out of
>>the bag.  Of course the control panel that displayed both real and
>>logical addresses was another hint.

>OK.... How does one figure this out.  The emulation of ICM & STCM would be easy
>to understand.  If you're looking at the emulator and you see the code
>necessary for the emulation of an new op-code (in this case, LRA) , how can you
>know your're looking at a page table if you don't know what one does?

The 360/67 used page and segment tables, so that would supply a basic
pathway to understanding.

The S/370 145 came with microcode listings, so a clever reader could
figure things out...

john alvord



Fri, 26 Jan 2001 03:00:00 GMT  
 Instruction Set Evolution

Quote:

> One important case is TR, where the possible memory reference addresses
> cannot be known in advance.

No way. The possible addresses are in the range destination to destination + length
(really length-1) and table to table plus 255. That covers a maximum of four pages.

--

Shmuel (Seymour J.) Metz
Reply to host nsf (dot) gov, user smetz



Fri, 26 Jan 2001 03:00:00 GMT  
 Instruction Set Evolution

Quote:

> The S/370 135 and 145 had dynamic address translation, but it wasn't
> enabled by the microcode. Some University of Maine hackers studied the
> microcode, patched it to turn on virtual stuff, and then started porting
> CP/67. That is cool for 1972-3, eh?

Actually, the DAT hardware of the 370/145 was used by pre-VS "microcode"; take
a look at the CE manuals for the "DOS Emulation" feature.

--

Shmuel (Seymour J.) Metz
Reply to host nsf (dot) gov, user smetz



Fri, 26 Jan 2001 03:00:00 GMT  
 Instruction Set Evolution

Quote:


> >Not yet mentioned...  Data boundry alignment stopped being
> >required with S/370.  Only privileged instructions could get an
> >0C_    AHHHHGG!!!! Been so long since I got one, I forget what the
> >program interrupt code is for boundry alignment...  ;-)

> It was 0C6, "specification exception," and good riddance to it!

That's the associated ABEND code; the PI code was 6.

--

Shmuel (Seymour J.) Metz
Reply to host nsf (dot) gov, user smetz



Fri, 26 Jan 2001 03:00:00 GMT  
 Instruction Set Evolution

Quote:

> From the Table of Contents of the most recent POP --
> Of course, some of these facilies were not in the initial ESA/390, like
> Checksum, Compare and Move Extended, and maybe the string instructions.

More important, IMHO, is PLO, which was also not in the original S/390 PoOps.

--

Shmuel (Seymour J.) Metz
Reply to host nsf (dot) gov, user smetz



Fri, 26 Jan 2001 03:00:00 GMT  
 Instruction Set Evolution

Quote:

> snip

> I happen to like boundary checking, because aligned memory references
> perform much better than unaligned. My two cents worth... ;-)

I agree, but I may be old fashioned. Just how much does it matter
anymore? On a 370 it was a big deal. What about teoday's processors
(large and small).

Quote:
> Oh, BTW, I sometimes use a macro to generate a conditional branch to *+1
> to cause S0C6, when the "cannot happen" situation does happen. It's
> better than the ABEND macro (no registers or CC alteration), and slightly
> better than branching around "EX 0,*".

Haven't seen that one. I like it, under appropriate circumstances.

--

Beyond Software, Inc.      http://www.beyond-software.com
"Transforming Legacy Applications"



Fri, 26 Jan 2001 03:00:00 GMT  
 
 [ 109 post ]  Go to page: [1] [2] [3] [4] [5] [6] [7] [8]

 Relevant Pages 
 

 
Powered by phpBB® Forum Software