On Mon, Nov 19, 2012 at 11:13 AM, Paul Gilmartin <paulgboul...@aim.com> wrote: > On Mon, 19 Nov 2012 11:59:28 -0500, Martin, Larry D wrote: >> >>IBM Disk capacity tables show 3390-27 as 32760 Cylinders >> > That number is spookily close to 2^15. Was it chosen by design > for tolerance of code that might malfunction with a negative > halfword cylinder number? Has some such code been broken by > the 3390-54? > > -- gil
IBM DS8XXX requires volumes to be an equal multiple of 1113 cylinders. Could have created a 3390-29 for the highest multiple under 32,760. IBM ESS E/F20s allowed up to 32,670 , 800s allow up to 64K cylinders. Only certain versions of z/OS support 3390-28 to 54. Ceratin following version (z/OS 1.10?) support EAVs up to 4 * 3390-54. z/OS 1.13+? support 16 * 3390-54. Future incremental extentions up to 4096 * 3390-54 are expected per IBMs statement of direction. The EAV addressing is converting the 3390 C'CCHHR' addressing to x'ccccCCCHRR' format. The X'cccc' is the low 16 bits of the cylinder number, The X'CCC' is the high 12 bits of the cylinder number (currently 4 bits used for the 16X * 3390-54). The X'H' is the 4 bits for the track number (0-15), The X'RR' is the 8 bits for the record number. The old C'HH' used 16 bits for a 4 bits value, so they took the top 12 bits to extend the number of cylinders. -- Mike A Schwab, Springfield IL USA Where do Forest Rangers go to get away from it all? ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
