Wow, that's interesting! It runs in 1.861138 cpu seconds on our z13s. We
have 800mips sliced up between 3 cores so 266mips per core.
So your zPDT core which is running emulated is faster at running REXX
than our z13s! Can that be right? No wonder IBM limit the zPDT
to 1 core imagine what workload you could run on an 8-core x86 blade
server with 256GB of RAM!
On 14/09/2018 2:50 AM, ITschak Mugzach wrote:
Hi David,
Happy to inform that it work just fine. Thanks. BTW, it took only
1.385268 cpu seconds on our machine, a zPdt, one CPU.
Tx again,
ITschak
On Thu, Sep 13, 2018 at 1:22 PM ITschak Mugzach <[email protected]> wrote:
Tx. Will report findings today.
ITschak
בתאריך יום ה׳, 13 בספט׳ 2018, 13:05, מאת David Crayford <
[email protected]>:
I didn't try it with System REXX as I don't have access to that. Let us
know how you go.
On 13/09/2018 4:58 PM, ITschak Mugzach wrote:
Thanks David.
Did u try this under SysremRexx? I'll try it tonight.
ITSCHAK
בתאריך יום ה׳, 13 בספט׳ 2018, 11:44, מאת David Crayford <
[email protected]>:
In the meantime you can do a quick and dirty using TCBTTIME
/* REXX */
main:
start = cputime()
do i = 1 to 1000000
nop
end
say 'CPU time:' cputime() - start
exit
cputime:
cvt = ptr(16)
tcbp = ptr(cvt)
tcb = ptr(tcbp + 4)
tcbttime = stg(tcb + 316, 8)
microsecs = x2d(substr(c2x(tcbttime),1,13))
cputime = microsecs / 1000000
return cputime
ptr: arg ptr, len
if len = '' then len = 4
return x2d(c2x(bitand(storage(d2x(ptr),len),x2c('7FFFFFF'))))
stg: arg ptr, len
return storage(d2x(ptr),len)
On 12/09/2018 7:29 PM, ITschak Mugzach wrote:
Will do. the partner site is under maintenance...
ITscha
On Wed, Sep 12, 2018 at 2:11 PM David Crayford <[email protected]>
wrote:
On 12/09/2018 7:01 PM, ITschak Mugzach wrote:
Sounds like a bug to ne...
Maybe in System REXX but not in TSO where you may have split screens
and
multiple tasks in the mix. Open an RFE.
ITschak
בתאריך יום ד׳, 12 בספט׳ 2018, 13:40, מאת David Crayford <
[email protected]>:
On 12/09/2018 6:07 PM, ITschak Mugzach wrote:
Boys, don't still the thread... I need to understand why is same
code
not
working under SystemRexx. Try it yourself.
I would take a guess that if you are running System REXX with
TSO=YES
then sysvar('SYSCPU') is returning the CPU time of
the parent task and not the CPU time of task that the exec is
running
in.
ITschak
On Wed, Sep 12, 2018 at 12:56 PM Mike Shorkend <
[email protected]>
wrote:
I use the procedure provided by IBM, taking all compiler option
defaults
// JCLLIB ORDER='FAN140.SVSC.PROCLIB'
//S1 EXEC REXXC
//SYSCEXEC DD DSN=MIKE.CREXX(LOOP2),DISP=SHR
//SYSIN DD DSN=MIKE.REXX(LOOP2),DISP=SHR
which expands to
XX*-----------------------------------------------------------------
XX* Compile REXX
program.
XX*-----------------------------------------------------------------
XX*
5 XXREXX EXEC
PGM=REXXCOMP,PARM='&OPTIONS'
IEFC653I SUBSTITUTION JCL - PGM=REXXCOMP,PARM='XREF
OBJECT'
6 XXSTEPLIB DD
DSN=&COMPDSN,DISP=SHR
IEFC653I SUBSTITUTION JCL -
DSN=FAN140.SFANLMD,DISP=SHR
7 XXSYSPRINT DD
SYSOUT=*
8 XXSYSTERM DD
SYSOUT=*
XX*SYSIEXEC DD
DUMMY
XX*SYSDUMP DD
DUMMY
9 //SYSCEXEC DD
DSN=MIKE.CREXX(LOOP2),DISP=SHR
X/SYSCEXEC DD
DSN=&&CEXEC(GO),DISP=(MOD,PASS),UNIT=SYSDA,
X/
SPACE=(800,(800,100,1))
10 XXSYSPUNCH DD
DSN=&&OBJECT,DISP=(MOD,PASS),UNIT=SYSDA,
XX
SPACE=(800,(800,100))
11 //SYSIN DD
DSN=MIKE.REXX(LOOP2),DISP=SHR
On Wed, 12 Sep 2018 at 12:44, David Crayford <
[email protected]>
wrote:
z/OS 2.3, IBM Compiler for REXX on zSeries 4.0 LVL PI76785.
Can you share you compiler JCL?
On 12/09/2018 5:32 PM, Mike Shorkend wrote:
David - I ran it on z/OS 2.2, the REXX compiler is 1.4
How about you?
On Wed, 12 Sep 2018 at 12:27, David Crayford <
[email protected]
wrote:
On 12/09/2018 4:56 PM, Mike Shorkend wrote:
The new test shows:
Non-compiled
CPU TIME = 77.09
Compiled
CPU TIME = 11.27
Your results are different to mine!
Interpreted: CPU time = 3.12
Compiled: CPU time = 7.99
Not so great, because of the inefficient memory management
with
stem
usage?
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access
instructions,
send email to [email protected] with the message: INFO
IBM-MAIN
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [email protected] with the message: INFO
IBM-MAIN
--
Mike Shorkend
[email protected]
www.shorkend.com
Tel: +972524208743
Fax: +97239772196
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [email protected] with the message: INFO
IBM-MAIN
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [email protected] with the message: INFO
IBM-MAIN
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [email protected] with the message: INFO
IBM-MAIN
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [email protected] with the message: INFO
IBM-MAIN
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [email protected] with the message: INFO IBM-MAIN
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [email protected] with the message: INFO IBM-MAIN
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [email protected] with the message: INFO IBM-MAIN
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [email protected] with the message: INFO IBM-MAIN
