Purpose
Starts, resets, or stops generating
Performance Monitor events.
Syntax
pmctl [ { [ -E [ mode ] ] [ -f
interval ] {[ -y
command ]}} | [ -O
Options ] | [ -h ] | [ -r ] | [ -S
] ] [ -s ] [ { -a
-y
command [ -f
interval ] }]
Description
The pmctl command starts,
stops, or resets the generation of Performance Monitor events in the
PMAPI subsystem to support manual offline mode with the tprof -E command.
It also reports the current status of the PMAPI subsystem.
Flags
| Item |
Description |
| -a |
Turns on large page analysis. |
| -E [ mode ] |
Enables event-based profiling. You can specify one of the following modes:
- PM_event
- Specifies the hardware event to profile. If no mode is specified for the -E flag, the
default event is processor cycles (PM_CYC).
- EMULATION
- Enables the emulation profiling mode.
- ALIGNMENT
- Enables the alignment profiling mode.
- ISLBMISS
- Enables the Instruction Segment Lookaside Buffer miss profiling mode.
- DSLBMISS
- Enables the Data Segment Lookaside Buffer miss profiling mode.
|
| -f
interval |
Specifies the sampling interval to use.
- For processor cycle, EMULATION, ALIGNMENT, ISLBMISS, and DSLBMISS
events, specify 1 to 500 milliseconds (default = 10).
- For other Performance Monitor events, specify 10000 up to MAXINT occurrences (default =
10000).
If you use the -f flag with the -y flag, specify 1 up to MAXINT occurrences for
other Performance Monitor events (default = 10000). |
| -O
options |
This -O option can have the following values:
- threshold_cmp_val=value
- Sets a threshold-compare floating point value.
The Power® processor has a threshold counter facility. The threshold counter is different from the
primary performance monitor counters (PMCs) because the threshold counter can count the number of
events that occur between a separate set of designated start events and end events from the core,
cache, and memory subsystems.
This option specifies a threshold value that you can use to
compare against the number of events between the designated start events and end events.
- threshold_event_sel=name
- Specifies the events that must be considered for threshold counting. Valid values for the
threshold_event_sel option are as follows:
- THRESHOLD_COUNT_CYCLES
- Counts the number of events that occur or cycles on which the control register (CTRL) run latch
has been set. The AIX® operating system uses the control
register (CTRL) bit to indicate the idle or run state. The performance monitoring unit (PMU) also
use this bit to avoid counting events during idle periods. This bit is commonly called the run
latch. This option does not depend on freeze conditions. This means that the events are
counted even when the PMU has frozen the contents of event counters.
- THRESHOLD_COUNT_NUM_COMP_INSTR
- Counts the number of completed instructions when the CTRL run latch is set.
This option does
not depend on freeze conditions. This means that the events are counted even when the PMU has frozen
the contents of event counters.
- THRESHOLD_COUNT_PMC1_EVENTS
- Counts the PMC1 events.
PMC1 - PMC4 are 32-bit registers that are called programmable counters
because the events that can be counted are specified by the program. This option depends on freeze
conditions. This means that the events are not counted when the PMU has frozen the contents of event
counters.
- THRESHOLD_COUNT_PMC2_EVENTS
- Counts the PMC2 events. This option depends on freeze conditions. This means that the events are
not counted when the PMU has frozen the contents of event counters.
- THRESHOLD_COUNT_PMC3_EVENTS
- Counts the PMC3 events. This option depends on freeze conditions. This means that the events are
not counted when the PMU has frozen the contents of event counters.
- THRESHOLD_COUNT_PMC4_EVENTS
- Counts the PMC4 events. This option depends on freeze conditions. This means that the events are
not counted when the PMU has frozen the contents of event counters.
|
| -O
options (continued) |
- threshold_eve_start_sel=name
- Specifies the event to start the threshold counting. For valid values of this option, see threshold_eve_stop_sel.
- threshold_eve_stop_sel=name
- Specifies the event to stop the threshold counting. You can set the following values for the
threshold_eve_start_sel and threshold_eve_stop_sel options:
- PM_MRK_INST_DECODED
- Sampled instructions are decoded.
- PM_MRK_INST_DISP
- Sampled instructions are dispatched.
- PM_MRK_INST_ISSUED
- Sampled instructions are issued.
- PM_MRK_INST_FIN
- Sampled instructions are finished.
- PM_MRK_INST_CMPL
- Sampled instructions are completed.
- PM_MRK_LD_MISS_L1
- Sampled instruction level 1 (L1) load cache is missed.
- PM_MRK_L1_RELOAD_VALID
- Sampled instruction level 1(L1) reload is valid.
- EVE_SEL_PMC1
- Event that is selected in monitor mode control register (MMCR) 1 for PMC1 occurred.
- EVE_SEL_PMC2
- Event that is selected in MMCR1 for PMC2 occurred.
- EVE_SEL_PMC3
- Event that is selected in MMCR1 for PMC3 occurred.
- EVE_SEL_PMC4
- Event that is selected in MMCR1 for PMC4 occurred.
- PM_MRK_NTF_FIN
- Sampled instruction is in the next to finish state.
- PM_MRK_L2_RC_DISP
- RC machine dispatched for sampled instruction.
- PM_MRK_ST_DONE_L2
- RC machine done for sampled instruction.
|
| -O
options (continued) |
- random_samp_ele_crit=name
- Specifies the random criteria for selecting the instructions for sampling. Valid values for this
option are as follows:
- ALL_INSTR
- All instructions are eligible. This value is the default setting.
- LOAD_STORE
- The operation is routed to the Load Store Unit (LSU); for example, load, store.
- PROB_NOP
- Sample only special no-operation instructions, which are called Probe NOP events.
- LARX_STCX
- Sample only load and reserve indexed (LARX) instructions and store conditional indexed (STCX)
instructions.
- LOAD_SAMPLING
- Sample only load instructions.
- LONG_LATENCY_OP
- Sample only long latency instructions (div/sqrt/mul).
- STORE_SAMPLING
- Sample only store instructions.
- LOAD_MISSES
- Sample only load-miss instructions.
- LOAD_HIT_STORE
- Sample only load-hit-store instructions. This value is applicable only for cases where a load
operation can forward data from the store queue to a finish state. This value is not applicable to
cases where a load-hit-store instruction is in the store queue but the load operation partially
overlaps with the store.
- BRANCH_MISPREDICTS
- Sample only branch-mispredict instructions.
- BRANCH_MISPREDICTS_DIR_CR_CTR
- Sample only branch-mispredict events (Direction, CR, or CTR).
- BRANCH_MISPREDICTS_TA
- Sample only branch-mispredict events (TA).
- BRANCHES_RESOLVED_TAKEN
- Sample branches with resolution.
- NON_REPEATING_BRANCHES
- Sample only non-repeating branches.
- ALL_BRANCHES_REQ_PRED
- Sample all branches that require prediction.
- group_num=event_group
- Specifies a group number that must be used for the pmctl command. This option
is an optional attribute. If you specify this value, the group must have a threshold event and the
same event must be used with the -E flag.
|
| -h |
Prints man page information. |
| -r |
Releases and resets the PMAPI subsystem. |
| -S |
Stops generating Performance Monitor events. |
| -s |
Prints the current status of the PMAPI subsystem. |
| -y
command |
Turns on the event-based profiling only for the specified command and its
descendents. |
Security
Attention RBAC users: This command can perform privileged operations. Only privileged
users can run privileged operations. For more information about authorizations and privileges, see
Privileged Command Database in Security. For a list of privileges and the authorizations
associated with this command, see the lssecattr command or the getcmdattr
subcommand.
Examples
- To stop generating Performance Monitoring events, enter the following
command:
pmctl –S
- To reset generating Performance Monitoring events, enter the following
command:
pmctl –r
- To report the current status of the PMAPI subsystem, enter the
following command:
pmctl –s
- To start generating Performance Monitoring events, enter the following
command:
pmctl –E
- To start generating Performance Monitoring events only for the
specified workload command and its descendents, enter the following
command:
pmctl –E –y workload
- To support the tprof -E command in manual offline mode,
enter the following command:
trace -adf -o mydata.trc
trcon
pmctl –E
sleep 10; trcstop
gensyms > mydata.syms
tprof –suker mydata
- To support the tprof -E command in manual offline mode
profiling for the specified workload command and its descendents,
enter the following command:
trace -adf -o mydata.trc
trcon
pmctl –E –y workload
trcstop
gensyms > mydata.syms
tprof –suker mydata
- To count the number of threshold count events elapsed between a distinct pair of
threshold start events and end events and to specify the threshold value for comparison, enter the
following command:
trace -adf -o mydata.trc
trcon
pmctl -E PM_THRESH_MET-O threshold_cmp_val=10000 -O
threshold_event_sel=THRESHOLD_COUNT_CYCLES -O
threshold_eve_start_sel=PM_MRK_L1_RELOAD_VALID -O
threshold_eve_stop_sel=PM_MRK_ST_DONE_L2 -O
random_samp_ele_crit=ALL_INSTR -O group_num=607
sleep 10; trcstop
gensyms > mydata.syms
tprof -suker mydata