As stated, the ARCH(n) limits the assembler instructions the compiler can
generate based on hardware capabilities. You won't get z14 vector register
instructions with ARCH(11).
Your debugger may in fact have a problem We just installed the latest PTFs for
Expediter and have no ARCH(12) proble
What about PC tools such as Snagit or CNTL+WIN+S?
--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
As suggested, simply define four output files, open them, and then write to
them based on the data in the input file record, perhaps using an EVALUATE
statement. Opening them all at the start of your program has another
advantage: if you open only one output at a time and loop between them, it'
SELECT specifies the COBOL FILE name to be ASSIGNED to a DDNAME. The FILE name
is then used in an FD to nominate the I/O area(s).
There is no dynamic SELECT that I am aware of. You have to use some form of
SVC99. BPXWDYN can be used for this purpose.
---
As an enhancement to the HILITE command, it would be nice if the list of
keywords for any given language could be edited to customize them for
individual use.
I use PL/I to edit SQL statements, and a few tweaks would be nice...lacking an
SQL type, of course.
Cheers,
Jon.
--
I'm not certain in assembler, but in COBOL, there is a START statement that
will position a file at a given record which has a key that is EQ, GT, GE, or
NE to the one you supply. This does not READ the record, but simply sets the
pointer. Subsequent READ stamens will proceeded sequentially un
display 'Variable-Data = "' Variable-Data '"'
compute Char-Length = length(trim(input-array(1), trailing))
display 'Variable-Length = ' Char-Length
Variable-Data = &
Not to put too fine a point on encapsulation because what you say is true for
Working-Storage and Paragraphs or Sections. But you can restrict or make
variables available to internal subroutines with the GLOBAL attribute.
Likewise, variables can be made available to external subroutines with t
Has anyone considered that z/OS is knee-capped so it only runs on GPs? Are we
suggesting that people will buy GPs at 10 times the price rather than IFLs or
zIIPs just so they can run z/OS under Linux?
--
For IBM-MAIN subscribe
At the risk of being "WRONG" ;¬)) several times, I offer the following.
The Processor Units (GPs, CPU, etc.) are PCIe Gen 4, but the 16 slots in the
I/O drawer hold Gen 3 cards, up to 16 of them at 16GBps. Each card can support
a max of 32 lanes which can be multiplexed. The max theoretical tr
Depending on the actual work characteristics, you might consider buying a zIIP.
You could share this between PROD and DEV to offload a lot of "grunt" work
that runs in an SRB, including Sorts and many ISV products from BMC and
Broadcom. More good news, any work offloaded from the GPs may cut y
There will be a need for assembler programmers for quite a while, but mainly
because over the last forty years, and long after even COBOL II added functions
and a case construction in 1987, very, very clever people decided they would
write application modules in assembler... and not waste time
cic...@listserv.uga.edu
--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
I think the key is the use of GLOBAL variables: those available to all internal
subroutines. Not to be confused with EXTERNAL variables.
GLOBAL are available to all internal subroutines, as long as you don't redefine
the name in the subroutine. So if you want only one variable throughout,
def
The Linkage Section is how modules pass data from the top module in the chain
down, and then back. In fact, they don't "pass" the data at all, but use a
single address that is established by the top-most caller. All modules use the
same address which is located in the top-most module's DSECT,
Of course in my last example,
compute N = 1066
move 'ty' to A
goback.
Should be
compute N = 1066
move 'ty' to X
goback.
--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu
It's not libel if it's true...although they may know where you live ;¬))
--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
I have run the debug option with COBOL 6.1. You need OPT(0) or NOOPT as
compiler options, and PARM='/DEBUG' on the EXEC statement.
You do not need the CEEOPTS DD, unless that is disabled at your installation.
--
For IBM-MAIN
You are correct in that NOOPT has been officially replaced by OPT(0) for COBOL
5 and 6, although NOOPT still still valid.
You can use OPT(1) or (2). I would suggest OPT(0) so all the TEST options
work, and the LIST is easier to correlate with your source. After all, why are
we debugging?
---
Clark, I agree with you completely. However for initial debugging, especially
with COBOL 6.1...I never used 5.x... I recommend you run with OPT(0) to get the
wrinkles out. Of course you will also want ARCH(11) if you have a z13, or
whatever setting is appropriate. There are a lot of recommend
st that the IBM
compilers did not enforce them.
Cheers,
Jon Butler
--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Interestingly enough the article blames the hardware, "un ancien ordinateur",
not the software. I'm not sure the writer knows the difference, but it's hard
to believe the software quit if it had been running for 23 years.
--
Fo
The A14 had two head assemblies, while the A18 had 4, on up to a max of 6 with
the B1C. The A2x and B2x had the same head configurations as the A1x models,
but twice the disk capacity.
--
For IBM-MAIN subscribe / signoff / arch
Tom,
As at 0720 on 19 November the z/OSv2 link is not working for me. Nothing at
all, no 404, no nothing.
The other links work, and I love the new prototype. Very easy to find what you
need, and to grab the PDFs for local use.
Thanks.
AhPROFS...where is Ollie North these days?
--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Interesting. I'll have to hunt down the toolset.
--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Does anyone have a CA Datacom Programming Guide in PDF format they could share
with me?
This seems to be one of the few restricted manuals on the CA site.
TIA,
Jon.
--
For IBM-MAIN subscribe / signoff / archive access instruct
Santa came early !!!
Thanks to the person (nameless unless he wants to publish) for sending me the
Guide.
Happy Christmas to all.
Jon.
--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@l
I propose the following fix for the IBM web sites:
'All links will work without "Deleting all IBM cookies or opening a "private"
window on your browser." '
;-))
--
For IBM-MAIN subscribe / signoff / archive access instructions
Let me also suggest you do not use the stand-alone external DB2 translator, but
use the SQL compiler option. Then any errors will be included in-line and it
will be easier to debug. Likewise the CICS compiler option vs the translator.
Also today, all the DB2 artifacts are GLOBAL such as the SQ
It's also interesting that according to the OT that the value of PI used to be
3.00, or perhaps the definition of "round" was different back then?
--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email
Ah, WATFORused it on an IBM 1800 with 32 Kilobytes of memory in 1975.
--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
If we compare the maxed out base and high end models of the z10, z12, and z13
we do see a trend, but it’s a downward ratio…fewer SAPs per CP. Of course you
can vary the ratio to suit your business, but consider the following.
If my arithmetic is correct, the z10 base had a 4:1 CP to SAP ratio
The number of SAPs is is directly related to the number of available PUs in a
book/drawer, and therefore to the number of configured CPs, IFLs, etc.
In a one drawer z10, there are 17 PUs, of which 3 are standard SAPs and 2 are
reserved spares. This means the customer can characterize 12 of the
Another thing to watch out for with passed parameters is their lengths. If the
subroutine has a Linkage Section parameter that is longer than that in calling
module, you will get an LE run-time error "CEE0802C -- Heap storage control
information was damaged." This is another practice that the
If you want to change the compiler options, you can do that on the first
statement in the COBOL program. Starting in column 7, include the following:
PROCESS SSRANGE (followed by a space or comma, as you like) and then the other
parameters. You don't need period at the end.
As Bill wrote, man
I think you need to put the VAT data into a KSDS since you don't have a DBMS.
If you are really interested in CICS performance...and who isn't?...you could
also define a CICS data table to hold the data. If you make this a
User-Maintained Table so that it is not updated dynamically, then you co
>>We mount every tape and then try running restoration Job and read data from
>>that to make sure restoration process works.
I think you can make the process less painful with some tools, but there is no
way to guarantee a tape will be readable.
I was installing a z/VM system a few years ago a
to sequence its working dataset. Not too
bad for COBOL source, but a bummer for PL/I or assembler.
Any advice as to how I can get all 80 bytes of the source code?
Thanks in advance.
Jon Butler
--
For IBM-MAIN subscribe / signoff
not x'00'.
As suggested, I will try the ISPF-L list.
Cheers,
Jon Butler.
--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
As has been pointed out, there are many IBM tools such as zPCR that you can
download to help with this exercise. The tools require either a good estimate
or RMF data from the LPARs to give you an accurate comparison. In running one
job anything can happen to distort the figures.
However, I th
Charles,
Your summation is correct. But there are no more mainframes that run in REAL
modethey all run in LPAR mode. If you want to assume you are paying only
for CPU time, then you can probably estimate that your new bill will be
calculated as 1.48*T*c, where T is the z9 CPU Time, and c
There was a 3090-180J uni that was rated at 23.5. The 3090-300J (3+3) screamed
along at 117 MIPS (20MSU).
The 9672 G6 -- 9672-Z17 -- had 1 CP and was rated at 200 MIPS (35MSU).
The z900 -- 2064-1C1 -- had 1 CP and was rated at 250 MIPS (43MSU). Tho Turbo
2064-2C1 was 303 MIPS.
A z/196 -- 2817-7
In the Win/7 explorer (file list) click on the "preview pane" icon just to the
left of the Help question mark icon. This will bring up a window on the right
side that will give you a peek into the file you select...PDF, WORD, EXCEL, etc.
-
I won't argue APAR nor SPOOL, which is one acronym that is more meaningful than
its deriving phrase, Simultaneous Peripheral Operation On-Line, but I will have
to question Cadaver. According to the OED, Cadaver in English is derived from
the Latin word for a dead body, related to to cadere, to
I believe you are thinking of The Code Breakers. Danger UXB was about bomb
disposal, and played around 1987 in the US, if I remember. You might also
check out The Betchley Circle, about four women who, 5 years after the war,
solve crimes in post-war England applying methods they learnt during
It works for me on H, but not on ST. Running z/OS 2.01, JES2 2.1, ISPF 7.1.
--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
I can use regular expressions, but the "generic" attributes such as \d for any
digit do not appear to work in ISPF. If I use FIND r'\([0-9]\)' I can find
the string "(9)", but if I use FIND r'\(\d\)' I can not.
For me FIND r'd' and FIND r'\d' both find the letter "d", not a digit. Any
thou
The escaped parentheses, left and right, provide that character "as is", so
"\([0-9]\) will find the string "(9)".
--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the
I'd like to state that even after 30 years of MVS, I still use the term "file"
when referring to mainframe datasets. In fact, if you look at the IBM COBOL
manual, you will see "File Organization", not "Dataset Organization"; PL/I
refers to "FILE Attributes", etc. And I think I can safely say I
> And I think I can safely say I have never seen a DD NULLDATASET command ;-))
I'm sure you haven't. Neither have I ever seen a DD NULLFILE command. ;-))
touché ;-)
--
For IBM-MAIN subscribe / signoff / archive access ins
In fairness, since yesterday was Chinese New Year, Lenovo have decided to come
to the party after all:
http://support.lenovo.com/us/en/product_security/superfish
--
For IBM-MAIN subscribe / signoff / archive access instructions,
There is also quite an overhead...up to 30%...when you use an initial value
INITIMG(x'40') with a GETMAIN.
--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message
Of course there is the option to not allow any IIP work to go to CPs by setting
IIPHONORPRIORITY=NO. In this case IIP work is run only on IIPs in priority
order. This may cause unacceptable waits. The AHONORPRIORITY=NO works the
same way for zAAPs.
---
The books are great and I don't mind the ads, but I do have a thought:
Since the RedBooks are strictly for IBM hardware and Software or products that
run on IBM hardware... when was the last time you looked at the MLC bill your
company received?
Cheers,
Jon B
I am replying late in the game but my understanding regarding IFLs is as
follows:
You can have as many IFLs on a CEC as there are PUs available to support them.
You can mix zIIPs, zAAPs, IFLs, and CPs in an LPAR.
IFLs always run at full capacity based on the CEC model, e.g. zEC12 = 1511
MIPS
Paul, in short, no.
The operating system runs on the underlying logical and physical hardware. So
if I have z/VM running in an LPAR that has only IFLs assigned logically to it,
z/OS will not be able to run. If I have a mix of IFLs and CPs in the LPAR,
then you can start up z/OS as a z/VM gue
ave "Linux in LPAR
mode", and "Linux as a guest under z/VM". I know IBM wants to continue to
reassure customers that "it's real Linux"... but what's in a name?
Cheers,
Jon Butler
--
F
Rad,
Yes, I should have said "can make use of CPs characterized as SAPs." It's
still early in the morning in the US!
--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with
I think the use of a GP is recommended only to emulate the ICF in a z/VM region
if you don't have an ICF available...perhaps the real ICFs are on the CFs...or
are simply "trying out" a SysPlex.
--
For IBM-MAIN subscribe / signof
Is it possible to use lower case SQL commands with CA's Unload Utility. I
can't find any way other than to use all UPPERCASE.
TIA,
--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listse
Yes, they have been available for six months in the zE12...120 usable PUs...of
which, in maximum configuration, 16 are configured as SAPs, 2 are spares, 1 is
a reserve, and 101 are customer configurable as CPs, IFLs, zIIPs, zAAPs, ICFs
or additional SAPs.
---
You can add CPs or memory to an active LPAR, but you can't take either away
without stopping the LPAR. You don't want more CPs assigned than you (or
PR/SM) can use...it's just overhead. In fact, unused CPs will be "parked" on
newer CECs when underutilized.
---
I see the Linkage "Problem" as one that should continue to work as it did in
V4. After all, many languages use a simple pointer to refer back to the real
storage in the calling program.
Other things, such as overrunning an array's subscript, which was a favourite
trick of mine in my FORTRAN
I'm using IE 11.0.31 and the new Knowledge Center has worked fine all
week...including right now.
http://www.ibm.com/support/knowledgecenter
--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists.
Jack, Oy it is because I like it for the very reason you state: it's easier to
search. It appears to actually consider the terms you enter in the search
field. Did someone at IBM look at Google Scholar search?
However, like you, if I don't really know what I need it may be a bit dicey.
That
You might try the PRODUCT FINDER -- "Select a Product" option. If you enter
ISPF it will bring up the ISPF "home" page
--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu wit
Personally, I appreciated the extra sleep. ;-))
--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
An interesting but hardly earth-shattering article. I've seen advertisements
for "Senior" positions that required "3 years of experience and Intermediate
English". Let's see: we are going to turn these folks lose on users to gather
requirements, design the HMI, and build a bullet-proof system
Sorry, it's my Intermediate English. Try "loose".
--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Perhaps I missed it, but why not use DISP=(PASS) if you want to keep the
DataSet enqueued until the Job end?
--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the messa
As suggested, you should change your font when writing to the form so the
entire error message and the lines to which it refers are shown as displayed on
your 3270 emulator. I think the rules for COBOL continuation lines have been
pretty well covered.
I would add only that since you want to
Charles,
Point taken.
However, it is possible to show an eighty byte record in this forum without
wrapping:
=COLS>
+1+2+3+4+5+6+7--
Cheers,
Jon.
--
For IBM-MAIN subs
You did not say which version of CICS you are using, but the Application
Programming Guide has detailed instructions on running COBOL II modules under
CICS.
--
For IBM-MAIN subscribe / signoff / archive access instructions,
send
Did she realize 2000 was not a leap year?
--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Ask the Singapore stock exchange about Tandem "non-stop". ;-))
--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
It seems to me this is another abstraction layer that would enable an instance
of bare-metal Linux to spool up multiple instances in an LPAR without z/VM.
What would the effect be of running KVM in a Linux instance under z/VM? And
of course PR/SM is still dispatching logical IFLs onto real I
z/OS has no hardware limitations?
I guess IBM's new customer agreements as to what can run on IIPs and AAPs and
IFLs, their prohibition of running z/OS on Hercules, and their purchase of
Platform Solutions Inc. (PSI) to avoid a court decision regarding the use of
commodity chips such as Itani
Having consulted at several companies under HIPAA rules, let me chime in. It's
not as though there is someone standing behind you with a baseball bat just
waiting for the opportunity to swat your head.
Most of the practical rules are common sense.
Don't go peeking into personal data just bec
IBM has filed for a trademark of "Z SYSTEMS" but apparently there is some
clarification needed. Interestingly enough, IBM applied for, and abandoned
"ZSYSTEM" in 2006. There are any number of "Z-SYSTEM" and derivatives out
there.
For example there is an active "Z-SYSTEM" trademark for "a weig
Often there are Application DBAs who have data access. Someone has to fix data
problems!
--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
According to the IBM LRM,
"A group item includes all group and elementary items that follow it until a
level-number less than or equal to the level-number of that group is
encountered."
Why you want to do that escapes me at the moment, but the LRM goes on with an
example where the 05 and 04 va
While the low-cost software community would like you to think ' "hacking" means
programming for the joy of it.' in fact its meaning in English, according to
the OED, is quite the opposite and that which is generally accepted by the
public:
hack, v.3
[f. hack n.3]
1.1 trans. To make a hack of
Perhaps ILBOABN0 (zero)?
--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Ah...the old currency bugbear.
Actually the mainframe division had another good year...it's the mini-computer
and nebulous computing offerings that continue to cause the problems.
--
For IBM-MAIN subscribe / signoff / archive
You can code the following to allow any number of arguments: the compiler will
generate the pseudo variables for you. I'm doing this from memory as I don't
have access to a PL/I compiler right now so please verify.
Declare myFunc External('VSHPARR') Entry;
If you were to code this:
Declare m
To get a legal copy you must have signed a user agreement (since several years
ago) which precludes the use of specialty engines except as allowed by IBM.
--
For IBM-MAIN subscribe / signoff / archive access instructions,
send em
You are conflating MVS with JAVA. No guarantees expressed or implied in the
JAVA world.
--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Is this remote work or will extended travel be involved? ;-))
--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Janet,
I'm not sure, but I think you may be confused about the parameter descriptors
on the ENTRY statement.
The descriptors do not limit the parameters passed, but rather they cause any
parameters to be converted (if required) to dummy variables which match the
description on the ENTRY sta
LIST has been part of PL/I, and most high-level languages, since at least the
370...that's a far back as I go!
You can use LIST to get an listing of the assembler generated by the PL/I
statements in your module. You can also limit the output by using a LIST(n,m)
where n and m are the line numb
The name of the program is ESCHATON, and irrespective of the number or value of
parameters passed, always returns 42.
--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with
We are in the process of converting to COBOL 6.2 and there are a number of
compiler options you can use...many also work with 6.1 and 5.
I suggest you try :
rules(noendperiod,noevenpack,laxperf,noslackbytes). While the "noendperiod" is
not performance per se, it will help identify unstructured
I don't know the detail of what your module does, but you can do the same thing
with IEBPTPCH. Set the SYSIN to PUNCH MAXFLDS=1 and if the dataset is empty
the RC=4; if it contains data the RC=0.
--
For IBM-MAIN subscribe / sig
94 matches
Mail list logo
