|
Programming Interface Information:
The Monitor Records are intended Programming Interfaces that allow the customer to write programs to obtain services of z/VM. |
MRSYTPRP
PrologControl Block Contents
Cross Reference
MRSYTPRP Prolog
DSECT NAME - SYTPRP
FUNCTION - Map a Monitor record.
LOCATED BY -
Through the Monitor control area
REFERENCED CONTROL BLOCKS -
MRRECHDR Monitor Record Header
NAME - MRSYTPRP
DESCRIPTIVE NAME - Monitor Sample Record
Domain 0 - System Domain
Record 2 - Processor Data (Per Processor)
DESCRIPTION - Real System or CPU operation exception.
A separate record is generated for each online
processor.
When multithreading is active a separate
record is written per-CPU (thread). By default,
fields in this record represent per-CPU or thread
level data. Exceptionally, some fields are per-Core
or per-CPU type. Where the latter two apply, comments
tagged to the affected fields will indicate the
category of data.
IMPLICATIONS OF MULTITHREADING CONFIGURATION CHANGES.
- See the prolog of MRPRCSMT for details
GENERAL NOTES -
An approximate count of the number of SIE entries on
this processor can be calculated by summing the
following fields from this record:
PFXRUNCP, PLSFPPFSuccess, PFXFSTPX, PFXFSTXC,
PFXFSTSG, PFXFST44, and PLSWRUCP.
Note this sum is an upper bound because a SIE entry
might be counted and then not happen. The reason for
this is how CP must operate as it prepares to issue
SIE: first it must increment the count, then it must
enable for interrupts, and then it must issue SIE.
When CP enables for interrupts, an interrupt can arrive
and the counted SIE will not be issued. Thus the count
is not exact but rather is just an upper bound.
MRSYTPRP Control Block Contents
Offsets Dec Hex Type Len Name (Dim) Description 0 0 Structure 400+ SYTPRP Start of monitor record 0 0 Character 0 SYTPRP_MRHDR Record header. See MRRECHDR for details. 0 0 Character 20 MRHDR 0 0 Unsigned 2 MRHDRLEN 2 2 Unsigned 2 MRHDRZER 4 4 Unsigned 1 MRHDRDM 5 5 Unsigned 1 * 6 6 Unsigned 2 MRHDRRC 8 8 Character 8 MRHDRTOD 16 10 Character 4 * 20 14 Character 0 MRHDR_END 20 14 Unsigned 2 SYTPRP_PFXCPUAD Real processor's CPU address 22 16 Unsigned 2 SYTPRP_PLSCUHAF Count of times a VMDBK was dispatched with 'hard affinity' (i.e., dedicated to this processor) 24 18 Character 8 SYTPRP_PFXPRBTM Total raw CPU time spent in emulation mode on this processor for all users. This field starts at x'7FFFFFFFF_00000000' and advances down (in the negative direction). 32 20 Character 8 SYTPRP_PFXUTIME Total raw CPU time spent on this processor which was charged to users. It includes both time spent running the guest and time that CP spends on behalf of a user doing such things as instruction simulation or page translation. This is in raw CPU time, not elapsed (or wall clock) time. This field starts at x'7FFFFFFFF_00000000' and advances down (in the negative direction). 40 28 Character 8 SYTPRP_PFXTMSYS Total raw CPU time spent on this processor which has been charged to the system instead of the user. In general, this is processing that is inappropriate to charge to any one particular user. Examples include portions of the following functions: scheduling decisions, handling interruptions, monitoring, and page migration. This field starts at x'7FFFFFFFF_00000000' and advances down (in the negative direction). 48 30 Character 8 * Reserved for IBM use. Previous field is no longer meaningful. (SYTPRP_PLSVFVTM) 56 38 Character 8 * Reserved for IBM use. Previous field is no longer meaningful. (SYTPRP_PLSVFOTM) 64 40 Unsigned 4 * Reserved for IBM use. Previous field is no longer meaningful. (SYTPRP_PLSVFLOD) 68 44 Character 8 SYTPRP_PFXTOTWT Total system wait time on this CPU (includes both active wait and enabled CPU wait). It is the time this CPU had no work to do, measured in raw CPU time. This field starts at x'7FFFFFFFF_00000000' and advances down (in the negative direction). Time accumulated in SYTPRP_PFXPRKWT is not accumulated in SYTPRP_PFXTOTWT. 76 4C Unsigned 4 SYTPRP_PFXRUNCI Count of interceptions from SIE on this processor. It does NOT include interceptions from SIE instructions issued in order to run a 'Vguest', i.e., a guest of a guest, during guest SIE simulation. It does NOT include interceptions handled by the fast-path simulation support code. 80 50 Unsigned 4 SYTPRP_PFXRUNPF Count of host page faults on this processor resulting from guest references to storage paged out by CP. Does not include faults on RCP pages 84 54 Unsigned 4 SYTPRP_PFXRUNCP Count of SIE instructions on this processor. It does not include instructions issued in order to run a 'Vguest', i.e., a guest of a guest, during guest SIE simulation. It does not include instructions issued following a successful fast-path simulation support code. 88 58 Unsigned 4 SYTPRP_CALFSTPH Count of fast-path instruction simulations. PFXFSTSG + PFXFSTXC + PFXFST44 92 5C Character 8 SYTPRP_PFXSPINT Elapsed spin time on formal locks on this CPU starting at zero and counting up while spinning on a lock acquisition. Measured in TOD clock units. Reset to zero when a CPU is varied online. 100 64 Unsigned 4 SYTPRP_PFXSPINC Count of spin lock requests on this processor that spun on a formal lock because it was not immediately available. The value counts up from zero and is reset to zero when a CPU is varied online. 104 68 Unsigned 1 SYTPRP_PFXCPUTY CPU Type. Valid values: X'00' - General Purpose (CP) X'02' - zSeries Application Assist (zAAP) X'03' - Integrated Facility for Linux (IFL) X'04' - Internal Coupling Facility (ICF) X'05' - zSeries Integrated Information (zIIP) 105 69 Unsigned 1 SYTPRP_PFXPOLAR Current polarization of the CPU/core. Valid values: X'00' - Horizontally polarized X'01' - Vertically polarized with low entitlement X'02' - Vertically polarized with medium entitlement X'03' - Vertically polarized with high entitlement 106 6A Character 2 * Reserved for IBM use 108 6C Unsigned 4 SYTPRP_PFXFSTPX Count of fast path processing of partial execution interrupts 112 70 Unsigned 4 SYTPRP_PFXFSTXC Count of fast path reflections of guest External Call interrupts 116 74 Unsigned 4 SYTPRP_PFXFSTSG Count of fast path simulations of SIGP External Call instructions 120 78 Unsigned 4 SYTPRP_PFXFST44 Count of fast path simulations of Diagnose X'44' instructions 124 7C Unsigned 4 SYTPRP_PLS9CNR Count of times a guest issued Diagnose x'9C' and the target was in a VMDSTATE < VMDREADY 128 80 Unsigned 4 SYTPRP_PLS9CWT Count of times a guest issued Diag x'9C' and the target was found in wait state (PSW.W=1) 132 84 Unsigned 4 SYTPRP_PLS9CSWT Count of times a guest issued Diag x'9C' and the issuer had to soft-wait on the target. 136 88 Unsigned 4 SYTPRP_PLS9CDSP Count of times a guest issued Diag x'9C' and the target was already dispatched 140 8C Unsigned 4 SYTPRP_CALENTMT For vertical CPUs/cores (SYTPRP_PFXPOLAR <> X'00'), this is a scaled number between X'00000000' and X'00010000' representing the portion of a physical CPU/core to which this vertical CPU/core is entitled. This value is X'00000000' for a horizontal CPU/core. When MT is enabled, this core value is repeated in the monitor record of each CPU in the core. These values are not to be added up across the core. 144 90 Character 8 SYTPRP_PFXPRKWT Total parked wait time on this CPU in TOD clock units. This field starts at 0 and advances up (in the positive direction). Time accumulated in SYTPRP_PFXPRKWT is not accumulated in SYTPRP_PFXTOTWT.
TSGET and CAD fields start here.
Offsets Dec Hex Type Len Name (Dim) Description 152 98 Character 8 SYTPRP_PF2TSGTM Total elapsed wall-clock time consumed by TSGET requests 160 A0 Unsigned 4 SYTPRP_PF2TSCNT Count of TSGET requests 164 A4 Unsigned 4 SYTPRP_PF2TSCAD Count of CAD instructions executed during TSGET requests 168 A8 Character 8 SYTPRP_PF2CADCT Count of CAD instructions used during spin lock acquisitions that were not delayed
The following fields record multithreading metrics and related information
for the current monitor interval.
At the end of the monitor interval, measurements are taken and deltas
from the previous monitor interval are calculated. According to what
happened during the sampling and calculation, one of the following
results are placed into the metric fields of this monitor record:
1. When a metric can be calculated then a positive integer is stored
in the metric field.
2. When a metric cannot be calculated, a negative integer is
stored in the metric field in the form of a bit-mask.
Defined values and their meanings are:
SYTPRP_KNODATA_ERROR = X'80000000'
Indicates that an error has been returned. The following
additional bits define the specifc condition, which may
be combined using bitwise OR where more than one
condition applies.
SYTPRP_KNODATA_GENERIC is a synonym for SYTPRP_KNODATA_ERROR
SYTPRP_KNODATA_UNSPECIFIED = X'80000001'
No data is available for this metric for unspecified
reasons.
SYTPRP_KNODATA_LOWCOUNTS = X'80000002'
Insufficient data is available to allow a metric
to be calculated. This value is forced for all MT
metrics of the first sample record written
for each CPU following enablement of sampling.
SYTPRP_KNODATA_TRANSITION = X'80000004'
No data is available because a core was offline
for a period within the monitor interval.
SYTPRP_KNODATA_MTDATALOSS = X'80000008'
No data is available because of an MT counter data loss
condition.
SYTPRP_KNODATA_NOCORE = X'80000010'
No data is available because there are no cores from
which to extract counters for the metric in question.
SYTPRP_KNODATA_NOEXTRACTION = X'80000020'
No data is available because counters from which the
metric is calculated were not available during the
sample interval.
SYTPRP_KNODATA_INTERNAL = X'C0000000'
No data is available because of an internal error.
It is possible for configuration changes to occur that affect
the basis of comparison with previous monitor samples. An example
is where the speed of a core changes during the sample interval.
A multithreading configuration change may also occur due to
the processing of the SET MULTITHREAD command.
Configuration changes are indicated in SYTPRP_CAL_MTSFLGS when
SYTPRP_CAL_HISSFCC is set.
If an internal error occurs when calculating metrics, then a soft
abend is issued, CPU counter sampling is permanently disabled,
the metric values returned will be set to SYTPRP_NODATA_ERROR
and SYTPRP_CAL_HISSFNA is set in SYTPRP_CAL_MTSFLGS.
Metrics returned are in two categories:
1) CPU-type-category
This comprises: SYTPRP_CAL_INTERVALTIMEBYTYPE
SYTPRP_CAL_SAMPLEDCORESBYTYPE
SYTPRP_CAL_PRODBYTYPE
SYTPRP_CAL_BUSYTIMEBYTYPE
SYTPRP_CAL_CAPBYTYPE
SYTPRP_CAL_MAXCAPBYTYPE
SYTPRP_CAL_MTUTILBYTYPE
SYTPRP_CAL_AVGTDBYTYPE
These metrics are calculated using data pertaining to all
online cores of a particular CPU type.
The number of sampled cores participating in the calculation
given in SYTPRP_CAL_SAMPLEDCORESBYTYPE.
The time interval over which the metric applies is given by
SYTPRT_INTERVALTIMEBYTYPE.
SYTPRP_PFXCPUTY will indicate to which CPU type these
metrics apply.
Also, CPU-type-category data will appear in records for
other threads of the same CPU type for this monitor
interval.
2) Core-category
This comprises: SYTPRP_CAL_INTERVALTIMEBYCORE
SYTPRP_CAL_PRODBYCORE
SYTPRP_CAL_BUSYTIMEBYCORE
SYTPRP_CAL_MTUTILBYCORE
SYTPRP_CAL_AVGTDBYCORE
These metrics are calculated using data pertaining to all
threads of one core.
The time interval over which the metric applies is given by
SYTPRT_INTERVALTIMEBYCORE.
SYTPRP_CAL_CORID indicates the core from which these metrics
have been calculated.
SYTPRP_CAL_TID indicates the thread ID of the core for
which this record has been created.
The same core-category data will appear in all records for
threads of the same core for this monitor interval.
Aggregation of data.
When aggregating D0 R2 monitor record data for the same
monitor sample interval, core-category metrics for the same
core should be considered duplicate data.
When aggregating D0 R2 monitor record data for the same
monitor sample interval, CPU-type-category metrics for the
same CPU type should be considered duplicate data.
Scaling and units.
Time metrics are in units of milliseconds.
IMPLICATIONS OF MULTITHREADING CONFIGURATION CHANGES.
- See the prolog of MRPRCSMT for details.
Offsets
Dec Hex Type Len Name (Dim) Description
176 B0 Bitstring 1 SYTPRP_CAL_MTSFLGS Status flags settings:
1... .... SYTPRP_CAL_HISSFNA X'80' Facilities required
to calculate the
multithreading metrics are
not available. Measurements
have been suspended.
.1.. .... SYTPRP_CAL_HISSFCC X'40' A configuration
change occurred during the
measurement period. For
example, a CPU speed change
or multithreading configuration
change occurred. Metrics are
recorded but their basis for
comparison has changed.
..1. .... SYTPRP_CAL_SYSMTSTM X'20' Multithreading was
not requested
...1 .... SYTPRP_CAL_SYSMT X'10' Multithreading is
not enabled.
.... 1... *
.... .1.. *
.... ..1. *
.... ...1 *
177 B1 Unsigned 1 SYTPRP_CAL_TID Thread ID for this thread
178 B2 Unsigned 2 SYTPRP_CAL_CORID Core ID of the core to
which this thread belongs.
CPU-type metrics calculated across all cores for the CPU type specified
by SYTPRP_PFXCPUTY
Offsets
Dec Hex Type Len Name (Dim) Description
180 B4 Unsigned 4 SYTPRP_CAL_INTERVALTIMEBYTYPE Time
interval in milliseconds
since the previous extraction
of the metrics by core was
made. An initial extraction
was made to initialize D0R2
sampling so that the first
record to be written will
have a valid value for this
field. The field contains
either a positive integer
representing the described
metric value returned or a
negative indicating a kNoData
condition - see comments box
above for details.
184 B8 Unsigned 4 SYTPRP_CAL_SAMPLEDCORESBYTYPE The number
of cores sampled that
returned data for the
calculation of the
following per-CPU-type metrics
for this sample interval.
The field contains either
a positive integer
representing the described
metric value returned or a
negative value indicating a
kNoData condition - see comments
box above for details.
188 BC Unsigned 4 SYTPRP_CAL_PRODBYTYPE Productivity by CPU type:
For this logical CPU's CPU
type, a ratio times 1024 of how
much work was accomplished by all
cores to the estimated amount
of work that would have been
accomplished by all cores if
all core busy time had been spent
with all threads busy.
The field contains this ratio
unless a negative value is
present, in which case it
indicates a kNoData condition
- see comments box above for
details.
192 C0 Unsigned 4 SYTPRP_CAL_BUSYTIMEBYTYPE Busy Time by CPU
type: For this logical CPU's CPU
type, the sum of the dispatch
times of all logical cores of
this type, in milliseconds, over
the monitor sample interval. In
other words, it is the aggregate
of BUSYTIMEBYCORE for the cores of
this logical CPU's CPU type.
The field contains
either a positive integer
representing the described metric
value returned or a negative value
indicating a kNoData condition -
see comments box above for details.
196 C4 Unsigned 4 SYTPRP_CAL_CAPBYTYPE Capacity Factor by CPU
type: For this logical CPU's
CPU type, a ratio times 1024 of
the work rate the cores experienced
to the work rate the cores
experienced while running with
only one thread busy.
For example, if the CAPBYTYPE
field were 1280, the cores
experienced a work rate of
1280/1024 = 1.25 times the
work rate they experienced
when they were running with
only one thread busy.
The field contains this
ratio unless a negative
value is present,
in which case it indicates
a kNoData condition - see
comments box above for
details.
200 C8 Unsigned 4 SYTPRP_CAL_MAXCAPBYTYPE Max Capacity Factor by
CPU type: For this logical CPU's
CPU type, a ratio times 1024 of
the work rate the cores
experienced with all threads
busy to the work rate the
cores experienced with one
thread busy.
For example, if the
MAXCAPBYTYPE field were 1280,
the cores accomplished
1280/1024 = 1.25 times as much
work per unit time of all threads
busy as they did per unit
time of one thread busy.
The field contains
this ratio unless a
negative value is present,
in which case it indicates
a kNoData condition -
see comments box above for
details.
204 CC Unsigned 4 SYTPRP_CAL_MTUTILBYTYPE Processor Utilization by
CPU type: For this logical
CPU's CPU type, a ratio
times 1024 of the work the
cores accomplished during the
monitor sample interval to the
estimated work the cores would
have accomplished if they had
run with all threads busy for
the whole monitor sample interval.
The field contains this ratio
unless a negative value is
present, in which case it
indicates a kNoData returned
condition - see comment box
above for details.
208 D0 Unsigned 4 SYTPRP_CAL_AVGTDBYTYPE Average Thread Density by
CPU type: For this logical
CPU's CPU type, a ratio
expressing the average number
of threads active per core
when a core of this type
was dispatched and at least
one thread of the core was
active.
The field contains this ratio
times 1024 unless a negative
value is present, in which
case it indicates a kNoData
returned condition - see
comments box above for details.
Core metrics calculated across all threads of a specific core.
Offsets
Dec Hex Type Len Name (Dim) Description
212 D4 Unsigned 4 SYTPRP_CAL_INTERVALTIMEBYCORE Time
interval in milliseconds since
the previous extraction and
calculation of the metrics by
core was made. An initial
extraction was made to initialize
D0R2 sampling so that the first
record to be written will have a
value for this field.
The field contains either a
positive integer representing
the described metric value
returned or a negative value
indicating a kNoData condition
- see comments box above for
details.
216 D8 Unsigned 4 SYTPRP_CAL_PRODBYCORE Productivity by core:
For this logical CPU's core, a
ratio times 1024 of how much
work was accomplished to the
estimated amount of work that
would have been accomplished if
all core busy time had been
spent with all threads busy.
The field contains this ratio
unless a negative value is
present, in which case it
indicates a kNoData condition
- see comments box above for
details.
220 DC Unsigned 4 SYTPRP_CAL_BUSYTIMEBYCORE Busy Time by core: For
this logical CPU's core, the
amount of time the logical core
was dispatched to a physical
core over the sample
interval, in milliseconds.
The field contains either a
positive integer representing
the described metric value
returned or a negative value
indicating a kNoData condition
- see comments box above for
details.
224 E0 Unsigned 4 SYTPRP_CAL_MTUTILBYCORE MT Utilization by
core: For this logical CPU's
core, a ratio times 1024 of
the work accomplished to the
estimated work the core would
have accomplished if it had
run with all threads busy for
the whole monitor sample
interval.
This field contains either a
positive integer representing
the described metric value
returned or a negative value
indicating a kNoData condition
- see comments box above for
the details.
228 E4 Unsigned 4 SYTPRP_CAL_AVGTDBYCORE Core Average Thread
Density: For this logical
CPU's core, a ratio
expressing the average number
of threads active when the
core was dispatched and at
least one thread was active.
The field contains this
this ratio times 1024
unless a negative value
is present, in which case it
indicates a kNoData returned
condition - see comments box
above for details.
HOST SHARE IPTE INTERLOCK STATISTICS
There are two sets of IPTE Interlock statistics,
each of which is stored as a self-consistent set.
The first set is a quintuple relating to Interlock wait and
acquisition times. It comprises:
CAL_PLSIPTEI, PLSIIA, PLSIIADD, PLSIIWTM and PLSIIWTSSQ.
Derived from this are two triples from which we can
calculate the mean and variance of host interlock
acquisition wait times per interlock acquisition:
1) For Interlock acquisitions using method 1 use:
CAL_PLSIPTEI (total IPTE Interlock acquisitions
using method 1)
PLSIIWTM (accumulated time waiting for the
interlock using method 1 or 2)
PLSIIWTSSQ (accumulated sum of squares of time
waiting for the interlock using method 1
or 2)
2) For Interlock acquisitions using method 2 use:
PLSIIA (total IPTE Interlock acquisitions
using method 2)
PLSIIWTM (accumulated time waiting for the
interlock using method 1 or 2)
PLSIIWTSSQ (accumulated sum of squares of time
waiting for the interlock using method 1
or 2)
It is important to note that PLSIIA=0 when method 1
is in use. When method 2 is used then
PLSIIA=CAL_PLSIPTEI. Sums of wait times, squares of
wait times, hold times and squares of hold times are
accumulated for one method only.
The remaining metric, PLSIIADD, is incremented
whenever an interlock acquisition is granted and a
host share is already held. It may be used with the
total number of Interlock acquisitions to calculate
the mean number of additional host shares per initial
host interlock acquisition where its host share is not
held already.
To make this calculation, PLSIIADD and CAL_PLSIPTEI
(or PLSIIA) are summed across all logical CPUs.
See note below about calculating mean and variance of
hold times.
Thus:
delta-PLSIIADD/(delta-PLSIPTEI - delta-PLSIIADD)
provides the desired mean.
The second set is a triple relating to Interlock
hold time. It comprises:
CAL_PLSIINHLD, PLSIIHLD and PLSIIHDSSQ.
These three values are used to calculate the mean
and variance of the intervals over which the IPTE
Interlock was continuously held by the host.
Because an IPTE interlock continuous-hold interval
may begin on one logical CPU and terminate on another
then the three values should be summed over all logical
CPUs before calculating mean and variance.
Fields that provide sums of times are expressed in
TOD units; those that provide sums of squares of
times are expressed in TOD-units-squared.
If we carry out of the high-order bit position when
calculating any component of the two tuples then the
entire tuple affected is reset to zero.
The only time any metric component within a tuple is
observed not to increase monotonically between
successive sample records is where the tuple has been
reset. In all cases the recorded tuple is
self-consistent.
A tuple may be used as the base from which to calculate
a delta from the previous sample provided each
component value in that tuple is greater than or equal
to the corresponding component of the previous sample's
tuple.
Offsets Dec Hex Type Len Name (Dim) Description 232 E8 Unsigned 4 SYTPRP_CAL_PLSIPTEI Total count of IPTE Interlock acquisitions using methods 1 or 2. 236 EC Unsigned 4 SYTPRP_PLSIIA Count IPTE Interlock acquisitions using method 2. 240 F0 Unsigned 4 SYTPRP_PLSIIADD Count of times an IPTE Interlock was acquired as an additional host share. 244 F4 Character 8 SYTPRP_PLSIIWTM Cumulative amount of time spent waiting for an IPTE Interlock acquisition. 252 FC Character 16 SYTPRP_PLSIIWTSSQ Sum of squared time spent waiting for an IPTE Interlock acquisition. 268 10C Unsigned 4 SYTPRP_CAL_PLSIINHLD Count of intervals over which one or more host shares of an IPTE Interlock were continuously held. 272 110 Character 8 SYTPRP_PLSIIHLD Cumulative amount of time over which one or more host shares of an IPTE Interlock were continuously held. 280 118 Character 16 SYTPRP_PLSIIHDSSQ Sum of squared time over which one or more host shares of an IPTE Interlock were continuously held. 296 128 Unsigned 4 SYTPRP_COREXTCT Cnt times MT ctrs extracted 300 12C Character 8 SYTPRP_COREXTTT Total time consumed to extract MT counters for the CORBK for this core. 308 134 Unsigned 4 * Reserved for IBM use 312 138 Unsigned 4 SYTPRP_PLSPTLCL Count of calls to purge the TLB on a local processor. 316 13C Unsigned 4 SYTPRP_PLSPTLCD Count of calls to set a pending host TLB for a user. 320 140 Unsigned 4 SYTPRP_PLSPTLCA Count of calls to set a pending host TLB for a user (not the current user). 324 144 Unsigned 4 SYTPRP_PFXCPUCH Count of times this logical CPU ran a VMDBK using SIEIHCPU<>x'FFFF' and the VMDBK was switching logical CPUs. 328 148 Unsigned 4 SYTPRP_PFXPRGCT Count of times this logical CPU ran a VMDBK using SIEIHCPU=x'FFFF'. 332 14C Character 8 SYTPRP_CORTMTLT Timestamp at completion of the last threading level transition after IPL. 340 154 Unsigned 4 SYTPRP_CORTLSEQ Threading level transition sequence number. Odd during a threading level change and even otherwise. When 0, there have not been any threading level transitions since the core was last varied on. The value is twice the number of actual transitions. 344 158 Unsigned 2 SYTPRP_CAL_CORTMOFF Offset to SYTPRP_CORTMPTL array of core times per threading level. 346 15A Unsigned 2 SYTPRP_CAL_CORTMSIZ Size of an entry in the SYTPRP_CORTMPTL array. 348 15C Unsigned 1 SYTPRP_CAL_CORTMCNT Number of entries in the SYTPRP_CORTMPTL array. 349 15D Unsigned 1 SYTPRP_CORTHRDS Count of threads activated on this core. 350 15E Unsigned 2 * Reserved for IBM use 352 160 Unsigned 4 SYTPRP_PLSFPPFSUCCESS Count of times the pfault fast path was successful and returned directly to the SIE instruction 356 164 Unsigned 4 SYTPRP_PLSWRUCP Count of host SIE executions while simulating guest SIE instructions 360 168 Unsigned 4 SYTPRP_PLSWTIIN Cumulative count of WTIs recognized by interruption 364 16C Unsigned 4 SYTPRP_PLSWTITP Cumulative count of WTIs recognized by TPEI 368 170 Unsigned 4 SYTPRP_PLSWTSUSP Cumulative count of successful WTI suspensions 372 174 Unsigned 4 SYTPRP_PLSWTARDY Cumulative count of tardy WTI-suspend attempts 376 178 Character 8 SYTPRP_PLSWTI2SUSP Cumulative time from WTI-recognition to successful WTI-suspension in TOD clock units 384 180 Character 8 SYTPRP_PLSWTI2TRDY Cumulative time from WTI-recognition to tardy WTI-suspension in TOD clock units 392 188 Character 8 SYTPRP_PLSWTSUSTM Cumulative held in successful WTI-suspend in TOD clock units This field starts at 0 and counts up. Time accumulated in SYTPRP_PLSWTSUSTM is not accumulated in SYTPRP_PFXTOTWT. 400 190 Character 0 SYTPRP_END
Offsets Dec Hex Type Len Name (Dim) Description 0 0 Structure 8 SYTPRP_CORTMPTL Entry definition for the array of core times per threading level. 0 0 Character 8 SYTPRP_CAL_CORTMPRV Amount of time in TOD clock units that this core ran at the entry's corresponding threading level prior to the most recent threading level transition.
MRSYTPRP Cross Reference
Hex Hex
Name Offset Length Value
MRHDR 0 20
MRHDR_END 14 0
MRHDRDM 4 1
MRHDRLEN 0 2
MRHDRRC 6 2
MRHDRTOD 8 8
MRHDRZER 2 2
SYTPRP 0 400+
SYTPRP_CAL_AVGTDBYCORE E4 4
SYTPRP_CAL_AVGTDBYTYPE D0 4
SYTPRP_CAL_BUSYTIMEBYCORE DC 4
SYTPRP_CAL_BUSYTIMEBYTYPE C0 4
SYTPRP_CAL_CAPBYTYPE C4 4
SYTPRP_CAL_CORID B2 2
SYTPRP_CAL_CORTMCNT 15C 1
SYTPRP_CAL_CORTMOFF 158 2
SYTPRP_CAL_CORTMSIZ 15A 2
SYTPRP_CAL_HISSFCC B0 40
SYTPRP_CAL_HISSFNA B0 80
SYTPRP_CAL_INTERVALTIMEBYCORE D4 4
SYTPRP_CAL_INTERVALTIMEBYTYPE B4 4
SYTPRP_CAL_MAXCAPBYTYPE C8 4
SYTPRP_CAL_MTSFLGS B0 1
SYTPRP_CAL_MTUTILBYCORE E0 4
SYTPRP_CAL_MTUTILBYTYPE CC 4
SYTPRP_CAL_PLSIINHLD 10C 4
SYTPRP_CAL_PLSIPTEI E8 4
SYTPRP_CAL_PRODBYTYPE BC 4
SYTPRP_CAL_PRODBYCORE D8 4
SYTPRP_CAL_SAMPLEDCORESBYTYPE B8 4
SYTPRP_CAL_SYSMT B0 10
SYTPRP_CAL_SYSMTSTM B0 20
SYTPRP_CAL_TID B1 1
SYTPRP_CALENTMT 8C 4
SYTPRP_CALFSTPH 58 4
SYTPRP_COREXTCT 128 4
SYTPRP_COREXTTT 12C 8
SYTPRP_CORTLSEQ 154 4
SYTPRP_CORTMTLT 14C 8
SYTPRP_CORTHRDS 15D 1
SYTPRP_END 190 0
SYTPRP_MRHDR 0 0
SYTPRP_PFXCPUAD 14 2
SYTPRP_PFXCPUCH 144 4
SYTPRP_PFXCPUTY 68 1
SYTPRP_PFXFSTPX 6C 4
SYTPRP_PFXFSTSG 74 4
SYTPRP_PFXFSTXC 70 4
SYTPRP_PFXFST44 78 4
SYTPRP_PFXPOLAR 69 1
SYTPRP_PFXPRBTM 18 8
SYTPRP_PFXPRGCT 148 4
SYTPRP_PFXPRKWT 90 8
SYTPRP_PFXRUNCI 4C 4
SYTPRP_PFXRUNCP 54 4
SYTPRP_PFXRUNPF 50 4
SYTPRP_PFXSPINC 64 4
SYTPRP_PFXSPINT 5C 8
SYTPRP_PFXTMSYS 28 8
SYTPRP_PFXTOTWT 44 8
SYTPRP_PFXUTIME 20 8
SYTPRP_PF2CADCT A8 8
SYTPRP_PF2TSCAD A4 4
SYTPRP_PF2TSCNT A0 4
SYTPRP_PF2TSGTM 98 8
SYTPRP_PLSCUHAF 16 2
SYTPRP_PLSFPPFSUCCESS 160 4
SYTPRP_PLSIIA EC 4
SYTPRP_PLSIIADD F0 4
SYTPRP_PLSIIHDSSQ 118 16
SYTPRP_PLSIIHLD 110 8
SYTPRP_PLSIIWTM F4 8
SYTPRP_PLSIIWTSSQ FC 16
SYTPRP_PLSPTLCA 140 4
SYTPRP_PLSPTLCD 13C 4
SYTPRP_PLSPTLCL 138 4
SYTPRP_PLSWRUCP 164 4
SYTPRP_PLSWTARDY 174 4
SYTPRP_PLSWTIIN 168 4
SYTPRP_PLSWTITP 16C 4
SYTPRP_PLSWTI2SUSP 178 8
SYTPRP_PLSWTI2TRDY 180 8
SYTPRP_PLSWTSUSP 170 4
SYTPRP_PLSWTSUSTM 188 8
SYTPRP_PLS9CDSP 88 4
SYTPRP_PLS9CNR 7C 4
SYTPRP_PLS9CSWT 84 4
SYTPRP_PLS9CWT 80 4
This information is based on z/VM V7R3.0.
Last updated on 10 November 2023 at 16:18:05.
Copyright IBM Corporation, 1990, 2023