True, I believe it is the SRST instruction:
Instruction Format: SRST R1,R2 Operation Code (Hex): B25E
Examples: SRST R9,R2 SEARCH STARTING AT ADDRESS IN
R2 UNTIL ADDRESS IN R9 IS REACHED,
OR A CPU-DETERMINED NUMBER OF
BYTES ARE PROCESSED, OR THE
CHARACTER SPECIFIED IN BITS 56-63
OF GENERAL REGISTER 0 IS FOUND
Description: The SEARCH STRING instruction searches for a specified
character starting at the address in the second operand
register. The search stops when:
1) the desired character is found, OR
2) the end of the string being searched is reached, OR
3) a CPU-determined number of bytes (at least 256) has
been searched.
When SRST begins execution, the second operand register
contains the address of the starting (leftmost) byte of
the string of bytes to search. The first operand
register contains the address of the first byte after
the end of the string of bytes to search. The character
to search for is contained in bits 56 through 63
(rightmost byte) of general register zero.
As the search progresses, the address in the second
operand register is incremented one byte at a time. When
the address in the second operand register is equal to
the address in the first operand register, then the end
of the string has been reached.
Bits 0 through 55 of register zero must be zero when
SRST executes; if they are not, a specification
exception (S0C6 ABEND) will occur.
The addresses in the first and second operand
even-numbered registers are handled differently,
depending upon the current addressing mode. In 24-bit
addressing mode, bits 40 through 63 of the first and
second operand registers contain the addresses of the
data and bits 0 through 39 of both registers are
ignored. In 31-bit addressing mode, the contents of bits
33-63 contain the addresses and bits 0-32 are ignored.
In 64-bit addressing mode, the full contents (bits 0-64)
of the registers contain the addresses of the bytes to
be operated upon.
In access register mode, the address space containing
the second operand is specified by the access register
associated with the R2 operand only. The access register
associated with the R1 operand is ignored.
When condition code 3 is set, you can resume execution
of SRST by simply branching back to execute the
instruction again. No registers need to be modified
before execution can resume.
Both the first and/or second operand may specify general
register 0; if either or both does, then register 0 is
considered to contain both an address and the character
to search for.
You can search a string of unknown length for a
particular character using only two registers by making
the first operand register zero; this will work only if:
1) the string to search does not start below location
256 in virtual storage,
2) the string does not wrap around to location 0 in
virtual storage, and
3) the end-of-string character in register zero does not
need to be preserved.
Condition Code: 0 - not set, (N/A)
1 - desired character found, address of found character
is in first operand register
2 - desired character not found, registers not changed
3 - CPU-determined number of bytes have been searched
but desired character was not found; address of next
string byte to start search with is in second
operand register
-----Original Message-----
From: IBM Mainframe Discussion List [mailto:[email protected]] On Behalf
Of Charles Mills
Sent: Friday, March 13, 2015 1:44 PM
To: [email protected]
Subject: Re: Need assembler trick to quickly get second and third DS qualifiers
The classic approach is TRT, which is a little bit of a PITA (R2 anyone?) but
does the job. It is a single machine instruction but that does not mean it's
fast!
There is a new instruction (new relative to a lot of us LOL) "search string"
(as I recall) that is more straightforward and might be faster. Again,
fundamentally a loop whether you code it or the machine does it under the
covers, so not as fast as you might guess. Faster than a CLI/JE loop I would
guess.
Watch out for the "no second qualifier" special case. FOOBAR is a valid dataset
name.
Charles
-----Original Message-----
From: IBM Mainframe Discussion List [mailto:[email protected]] On Behalf
Of Pinnacle
Sent: Friday, March 13, 2015 10:28 AM
To: [email protected]
Subject: Need assembler trick to quickly get second and third DS qualifiers
I have a dataset FIRST.SECOND.THIRD, and I want to get SECOND.THIRD.
Normally I would loop to one character at a time to get to the '.', but I need
performance here. I sure some of you assembler gurus could give me some
inscrutable assembler to get there in like one or two instructions. Thanks in
advance!
----------------------------------------------------------------------
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
