Large Block Interface (LBI)

The large block interface (LBI) lets your program handle much larger blocks with BSAM or QSAM. On the current level of the system you can use LBI with BSAM, BPAM, and QSAM for any kind of data set except unit record or a TSO/E terminal. Currently blocks of more than 32 760 bytes are supported only on tape, dummy data sets, and BSAM UNIX files.

You request LBI by coding a BLKSIZE value, even 0, in the DCBE macro or by turning on the DCBEULBI bit before completion of the DCB OPEN exit. Coding BLKSIZE causes the bit to be on. It is best if this bit is on before you issue the OPEN macro. That lets OPEN merge a large block size into the DCBE.

Your DCB OPEN exit can test bit DCBESLBI to learn if the access method supports LBI. If your program did not request unlike attributes processing (by turning on bit DCBOFPPC) before issuing OPEN, then DCBESLBI being on means that all the data sets in the concatenation support LBI. If your program requested unlike attributes processing before OPEN, then DCBESLBI being on each time that the system calls your DCB OPEN exit or JFCBE exit means only that the next data set supports LBI. After the exit, OPEN leaves DCBESLBI on only if DCBEULBI also is on. Your exit routine can change DCBEULBI. Never change DCBESLBI.

Another way to learn if the data set type supports LBI is to issue a DEVTYPE macro with INFO=AMCAP. See z/OS DFSMSdfp Advanced Services. After the DCB OPEN exit, the following items apply when DCBESLBI is on:

OPEN is honoring your request for LBI.
Do not use the BLKSIZE field in the DCB. The system uses it. Use the BLKSIZE field in the DCBE. For more information about DCBE field descriptions see z/OS DFSMS Macro Instructions for Data Sets.
You can use extended BDWs with format-V records. Format-V blocks longer than 32 760 bytes require an extended BDW. See Block Descriptor Word (BDW).
When reading with BSAM or BPAM, your program determines the length of the block differently. See Determining the Length of a Block when Reading with BSAM, BPAM, or BDAM.
When writing with BSAM or BPAM, your program sets the length of each block differently. See Writing a Short Format-FB Block with BSAM or BPAM.
When reading undefined-length records with QSAM, your program learns the length of the block differently. See the GET macro description in z/OS DFSMS Macro Instructions for Data Sets.
To write format-U or format-D blocks without BUFOFF=L, you must code the 'S' parameter for the length field on the WRITE macro. For more information, see z/OS DFSMS Macro Instructions for Data Sets.
When writing undefined-length records with QSAM, you store the record length in the DCBE before issuing each PUT. See z/OS DFSMS Macro Instructions for Data Sets.
After an I/O error, register 0 and the status area in the SYNAD routine are slightly different, and the beginning of the area returned by the SYNADAF macro is different. See Figure 1 and z/OS DFSMS Macro Instructions for Data Sets.
If the block size exceeds 32 760, you cannot use the BUILD, GETPOOL, or BUILDRCD macro or the BUFL parameter.
Your program cannot request exchange buffering (BFTEK=E), OPTCD=H (VSE embedded checkpoints) or open with the UPDAT option.
With LBI, fixed-length unblocked records greater than 32 760 bytes are not supported by QSAM.

BSAM and QSAM do not support reading format-V (variable-length) blocks that are created by the z/VSE operating system if the block length exceeds 32767. You can read them as format-U blocks. z/VSE programs cannot read format-V blocks that are longer than 32767 if the block is created by a z/OS program.