Once a device is selected, CP .ust determine on which cylinder to
allocate a record on that device. CP maintains a chain of cylinder
record maps, one for each allocatable cylinder on the device. For 2305 devices, CP attempts to keep cylinder map blocks at the head of the
chain. The only optimization done for a 2305 is an atte.pt to minimize the amount of processor time involved in the allocation process. For movable-arm DASD (that is, not 2305), CP attempts to allocate the first
available record found when scanning the cylinder map chain. I Whenever a changed page is selected for replacement, it must first be I copied onto DASD before the real page can be made available. In cases I wbere there is already a DASD record allocated for the page and it is on I a movaDie-arm the page selectiun vlu I record and requests that a new record be allocated. This occurs each I time a page is to be written and its current backing-store location is I on a movable-arm DASD. Although this represents overhead in teras of I processor use, it is justified because it should minimize arm movement I and reduce page wait tiae. PAGE I/O REQUEST QUEUEING ALGORITHM Tbe ordering of page I/O requests
initiation with one SIO is done on
priority is:
1. In-queue requests
2. Not-in-queue requests
3. Reads
4. Writes 5. Q1 requests
6. Q2 requests
that are
a priority
chained together
ordering basis.
for
The PCI flags are set for page I/O requests. For non-2305 requests, there
is an interruption after each request. For 2305 requests, the PCI flag
is set so that there is one interruption for each revolution of the drum (one interruption for every three requests).
Note: For installations that are .uch more constrained by a page I/O bottleneck (as opposed to processor bottleneck), the 2305 PCI .ode can
be changed to operate in the same way as the non-2305 processing, that
is, by allowing an interruption immediately after each request. The SET SRM PCI DISK command causes the PCI flag to be set so there is one
interruption for eacb 2305 page request. SET SR! PCI DRU! cbanges it
back to the default mode of operation. VIRTUAL STORAGE PAGING ERROR RECOVERY Errors encountered during virtual storage (as opposed to spooling)
paging operations can generally be classified as either soft or hard
errors. Soft errors allow the system to continue operation without delay
or degradation. Hard errors can cause noticeable effects such as the
1-116" IBM VM/310 System Logic and Problem Deteraination--Volume 1

abnormal termination of user tasks (abend) and response degradation.
Errors that are successfully retried or corrected are known only to the I/O supervisor and the I/O error retry and recording routines; they
appear to the second level interruption handlers (such as WAITP1GE) as
if the original operation completed normally. SOFT An I/O error that occurs on a page swap-out is
considered to be a soft error. DMKPTRAN calls DMKPGTPG to assign a
different DASD page slot and the page is re-queued for output. The slot
that caused the error is not de-allocated, and thus is not assigned to
another virtual machine. All other uncorrectable paging errors are hard
because they more drastically affect system performance. RECOVER!: Hard paging errors occur on either I/O errors for
page reads or upon eXhausting the system's spooling and paging space.
Recovery attempted on hard errors depends upon the nature of the task for which the read was being done. If the operation was an attempt to
place a page of a virtual machine's virtual storage into real storage,
the operation of that particular virtual machine is terminated by
setting the page frame in error to zero and placing the virtual machine
in console function mode. The user and operator are informed of the
condition, and the page frame causing the error is not de-allocated,
thereby ensuring that it is not allocated to another user& The control program functions that call DMKPTRAN (such as spooling,
pageable control program calls, and system directory management) have
the option of requesting that unrecoverable errors be returned to the
caller. In this case, the CP task may attempt some recovery to keep the
entire system from terminating Cabend). In general, every attempt is aade to at least allow the operator to bring the system to orderly
shutdown if continued operation is impossible.
Proper installation planning should make the occurrence of a space
exhaustion error an exception. An unusually heavy user load and a
backed-up spooling file could cause this to happen. The operator is
warned when 90% of the temporary (paging/spooling) space in the system
is exhausted. He should take immediate steps to alleviate the shortage.
Possible reaedies that exist include preventing more users from logging
on and requesting users to stop output spooling operations. More drastic measures might include the purging of low-priority spool files. If the
system's paging space is completely exhausted, the operation of virtual machines progressively slows as more and more users have paging requests
that cannot be satisfied and operator intervention is required. VIRTUAL RELOCATION CP provides the virtual aachine the capability of using the DAT feature
of the real System/370. Programming simulation and hardware features
are combined to allow usage of all of the available features in the real
hardware, (that is, 2K or 4K pages, 64K or 1M segments) • For clarification, soae tera definitions follow: The physical storage of the real CPU, in which CP resides. The virtual storage available to any virtual machine, maintained by CP. CP Introduction 1-117