Well, got some time unexpectedly. I redid my TRUNC tests. And my original test
code was a poor test. TRUNC(OPT) is significantly better than TRUNC(BIN) for
both fullword and halfword binary fields using the MULTIPLY and DIVIDE verbs.
TRUNC(OPT): DIVIDE LSS-FULLWORD-1 BY LSS-FULLWORD-2
L 6,8(0,5) LSS-FULLWORD-1
SRDA 6,32(0)
D 6,16(0,5) LSS-FULLWORD-2
ST 7,8(0,5) LSS-FULLWORD-1
Can't really get anything better.
TRUNC(BIN) of the same sentence:
L 6,16(0,5) LSS-FULLWORD-2
SRDA 6,32(0)
L 0,8(0,5) LSS-FULLWORD-1
SRDA 0,32(0)
D 0,0(0,12) SYSLIT AT +0
CVD 1,352(0,13) TS2=16
MVO 338(6,13),355(5,13) TS2=2
CVD 0,352(0,13) TS2=16
TM 343(13),X'10' TS2=7
MVC 343(5,13),355(13) TS2=7
BC 8,332(0,11) GN=19(000390)
OI 347(13),X'01' TS2=11
GN=19 EQU *
MVC 336(2,13),8(12) TS2=0
D 6,0(0,12) SYSLIT AT +0
CVD 7,368(0,13) TS2=32
MVO 352(6,13),371(5,13) TS2=16
CVD 6,368(0,13) TS2=32
TM 357(13),X'10' TS2=21
MVC 357(5,13),371(13) TS2=21
BC 8,374(0,11) GN=20(0003BA)
OI 361(13),X'01' TS2=25
GN=20 EQU *
DP 336(12,13),356(6,13) TS2=0
ZAP 376(10,13),336(6,13) TS2=40
MVC 368(8,13),89(10) TS2=32
MVC 371(5,13),381(13) TS2=35
CVB 4,368(0,13) TS2=32
MVO 371(5,13),376(5,13) TS2=35
CVB 7,368(0,13) TS2=32
M 6,0(0,12) SYSLIT AT +0
ALR 7,4
BC 12,426(0,11) GN=21(0003EE)
A 6,4(0,12) SYSLIT AT +4
GN=21 EQU *
LTR 4,4
BC 11,436(0,11) GN=22(0003F8)
S 6,4(0,12) SYSLIT AT +4
GN=22 EQU *
ST 7,8(0,5) LSS-FULLWORD-1
Really bad. MULTIPLY of a halfword is also very good for TRUNC(OPT) and equally
horrible for TRUNC(BIN).
--
John McKown
Systems Engineer IV
IT
Administrative Services Group
HealthMarkets(r)
9151 Boulevard 26 * N. Richland Hills * TX 76010
(817) 255-3225 phone *
[email protected] * www.HealthMarkets.com
Confidentiality Notice: This e-mail message may contain confidential or
proprietary information. If you are not the intended recipient, please contact
the sender by reply e-mail and destroy all copies of the original message.
HealthMarkets(r) is the brand name for products underwritten and issued by the
insurance subsidiaries of HealthMarkets, Inc. -The Chesapeake Life Insurance
Company(r), Mid-West National Life Insurance Company of TennesseeSM and The
MEGA Life and Health Insurance Company.SM
> -----Original Message-----
> From: IBM Mainframe Discussion List
> [mailto:[email protected]] On Behalf Of Clark Morris
> Sent: Friday, April 13, 2012 9:53 AM
> To: [email protected]
> Subject: Re: Modernizing the BCP code ?
>
> On 12 Apr 2012 09:48:17 -0700, in bit.listserv.ibm-main you wrote:
>
> >Now that you mention it, I remember that the C/C++ compiler
> has a architecture option to control the instructions
> generated. I should have known that the PL/X compiler would
> too. I didn't know that they both share the same back-end. I
> wish that the COBOL compiler did. I am constantly amazed at
> the amount of code generate by a simpe:
> >
> > ADD +1 TO WS-INTEGER.
> >
> >when WS-INTEGER is defined as PIC S9(9) BINARY or NATIVE.
> COBOL seems to have an inordinate love for PACKED-DECIMAL.
> Someone once said it was due to ANSI standards compliance.
> Might be worth it, in CPU saved, to license the C compiler
> and port the OpenCOBOL
> >
> >Unless I somehow have the wrong compile parameters.
> Make sure you have TRUNC(OPT) implied or expressed in the compile
> parameters. If you have TRUNC(STD) then WS-INTEGER must be in the
> range -999,999,999 through +999,999,999. TRUN(BIN) always has been
> fiasco.
>
> Clark Morris
>
> ----------------------------------------------------------------------
> 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
