ISO8859-1  :0 > P \  3 %! -!7 !e!~!-!!"%""@"[M"=#)#g#y"#"###2$;$6;$r <$!@$"@%,#5%m$5%%5%&5&'5&E(5&{)5&*5&+5',5'S-1'.P'/>( 0>(K1>(2G(3I)4I)[5D)6D)7D*/8D*t9D*:D*;D+C<D+=D+>D,?,W@,iA,zB,C,D(,E),F-G-'H-DI-YJ-uK%-L-M%-N -O". P.CQ#.`R!.S!.T!.U!.V/ W/X/0Y/BZ/V[%/e\'/]/^0/_0`0a0,b0Bc0Xd0ge 0f&0g 0h0i#1j1%k14l1Sm1cn1o1p 1q1r 1s1t1u2v2#w2:x2Xy2rz2{2|2}(2~2$3$3=3b3|!3$333o4`4p4>4%5&J5LF5'5'6'6.'6V'6~'6'6'6'7'7F%7nH7*77848@:8u#8 8!8)99-9E!9_/99"99:B:5:`::5:6;>;>';}E;G;<3$ <> >\ >f&>r>>3>/? :?;7?v>?'?;@R@Q"@"@"@7A AEAYAuJA<A*B,B9-Bf3BB BC&C@CC CQQ(%QGQm?Q;QFRCRN^RR&SS+0SHSy#S%S'Susage: hpmcount [-adHkx] [-b time_base] [-D metrics] [-g event_groups] [-m metric_groups] [-o file] [-s set] command hpmcount [-h] where: command Program to be executed. -a Aggregates the counters on POE runs. -b time_base Selects a base for the data normalization. The available bases are as follows: time timebase purr PURR time (when available) spurr SPURR time (when available) The default value is "time". -d Adds detailed set counts for counter multiplexing mode. This option is ignored if the multiplexing mode is not specified. -D metrics Selects a list of derived metrics to be evaluated. Each derived metric can be qualified by a counting mode as follows: metric:counting_modes (See the -m option for available counting modes.) -g event_groups Lists a predefined group of events or a comma-separated list of event group names or numbers. When a comma-separated list of groups is used, the counter multiplexing mode is selected. Each event group can be qualified by a counting mode as follows: event_group:counting_modes (See the -m option for available counting modes.) -H Adds hypervisor activity on behalf of the process. -h Displays help message. -k Adds system activity on behalf of the process. -m metrics_groups Selects a list of derived metric groups to be evaluated. The "default" derived metric group refers to all derived metrics that do not belong to a specific derived metric group. Each metric group can be qualified by a counting mode as follows: metric_group_name:counting_modes The available counting modes are as follows: u user mode k kernel mode h hypervisor mode r runlatch mode n nointerrupt mode -o file Output file name. -s set Lists a predefined set of events or a comma-separated list of sets (1 to %d, or 0 to select all. See the pmlist command.) When a comma-separated list of sets is used, the counter multiplexing mode is selected. Each set can be qualified by a counting mode as follows: event_set:counting_modes (See the -m option for available counting modes.) -x Displays results in XML format. usage: hpmstat [-dHkrTUux] [-b time_base] [-D metrics] [-g event_groups] [-m metric_groups] [-o file] [-s set] interval count hpmstat [-h] where: interval Counting time interval in seconds or microseconds, with a default value of 1. count Number of iterations to count. The default is 1 with an interval in seconds, and infinity when the option -U is specified. -b time_base Selects a base for the data normalization. The available bases are as follows: time timebase purr PURR time (when available) spurr SPURR time (when available) The default value is "time". -d Adds detailed set counts for counter multiplexing mode. This option is ignored if the multiplexing mode is not specified. -D metrics Selects a list of derived metrics to be evaluated. Each derived metric can be qualified by a counting mode as follows: metric:counting_modes (See the -m option for available counting modes.) -g event_groups Lists a predefined group of events or a comma-separated list of event group names or numbers. When a comma-separated list of groups is used, the counter multiplexing mode is selected. Each event group can be qualified by a counting mode as follows: event_group:counting_modes (See the -m option for available counting modes.) -H Counts hypervisor activity only. -h Displays help message. -k Counts system activity only. -m metrics_groups Selects a list of derived metric groups to be evaluated. The "default" derived metric group refers to all derived metrics that do not belong to a specific derived metric group. Each metric group can be qualified by a counting mode as follows: metric_group_name:counting_modes. The available counting modes are as follows: u user mode k kernel mode h hypervisor mode r runlatch mode n nointerrupt mode -o file Output file name. -r Enables runlatch and disables counts while executing in idle cycle. -s set Lists a predefined set of events or a comma-separated list of sets (1 to %d, or 0 to select all. See the pmlist command.) When a comma-separated list of sets is used, the counter multiplexing mode is selected. Each set can be qualified by a counting mode as follows: event_set:counting_modes (See the -m option for available counting modes.) -T Write time stamps instead of time in seconds. -U Puts counting time interval in microseconds. This option is ignored if the counter multiplexing mode is specified. -u Counts user activity only. -x Displays results in XML format. HPM Fatal Error: Executing PE_Reduce with no MP_PARTITION Execution time (wall clock time): %.15g seconds on %d tasks HPM ERROR: hpm_init - pm_cycles returned invalid number: %lld ## HPM WARNING: Command Line Argument "-a" ignored. It is only valid with POE. Invalid set number %d Invalid group number %d HPM_EVENT_GROUP specifies an invalid group name %s Need to specify a program to execute ERROR: the time information is not available hpmstat ERROR - pm_stop:hpmstat ERROR - pm_get_data: Time Stamp: 0x%1$08x %2$08x <-> 0x%1$08x %2$08x seconds since 01/01/70 hpmstat ERROR - pm_tstart:ERROR - pm_stop_mygroup:hpmcount ERROR - pm_get_data_mygroup:ERROR - pm_tstart_mygroup: ## HPM WARNING: Environment variable "HPM_AGGREGATE_OUTPUT" ## ## Ignored. It is only valid with POE. ## ## HPM WARNING - Invalid environment variable HPM_LOG_DIR: "%s" - Ignored. ### hpmcount ERROR: node %1$d, getrusage FAILED cond = %2$d execvp error: %d ERROR - pm_init_private failed:hpmstat ERROR - pm_set_program_mm:ERROR - pm_set_program_mygroup_mm:ERROR - getrusage FAILED cond = %d ######## Resource Usage Statistics ######## Total amount of time in user mode : %f seconds Total amount of time in system mode : %f seconds Maximum resident set size : %lld Kbytes Average shared memory use in text segment : %lld Kbytes*sec Average unshared memory use in data segment : %lld Kbytes*sec Number of page faults without I/O activity : %lld Number of page faults with I/O activity : %lld Number of times process was swapped out : %lld Number of times file system performed INPUT : %lld Number of times file system performed OUTPUT : %lld Number of IPC messages sent : %lld Number of IPC messages received : %lld Number of signals delivered : %lld Number of voluntary context switches : %lld Number of involuntary context switches : %lld ####### End of Resource Statistics ######## ## Resource Usage Statistics - Average Total MIN MAX ## Wall Clock Time (in sec.) : %1$10f %2$10f %3$10f %4$10f s Time in user mode (in sec.) : %1$10f %2$10f %3$10f %4$10f s Time in system mode (in sec.): %1$10f %2$10f %3$10f %4$10f s Maximum resident set size : %1$10lld %2$10lld %3$10lld %4$10lld KB Shared mem use in text seg. : %1$10lld %2$10lld %3$10lld %4$10lld KB*s Unshared mem use in data seg.: %1$10lld %2$10lld %3$10lld %4$10lld KB*s Page faults w/out IO activity: %1$10lld %2$10lld %3$10lld %4$10lld Page faults with IO activity : %1$10lld %2$10lld %3$10lld %4$10lld Times process was swapped out: %1$10lld %2$10lld %3$10lld %4$10lld Times file system perf. INPUT: %1$10lld %2$10lld %3$10lld %4$10lld Times file system perf.OUTPUT: %1$10lld %2$10lld %3$10lld %4$10lld IPC messages sent : %1$10lld %2$10lld %3$10lld %4$10lld IPC messages received : %1$10lld %2$10lld %3$10lld %4$10lld signals delivered : %1$10lld %2$10lld %3$10lld %4$10lld voluntary context switches : %1$10lld %2$10lld %3$10lld %4$10lld involuntary context switches : %1$10lld %2$10lld %3$10lld %4$10lld Time in user modeUtilization rateInstructions per cycleMIPSInstructions per I Cache Miss% Instructions dispatched that completed% of cycles with 0 instructions completedTotal load and store operationsTotal l2 data cache accesses% Cycles LSU is idleInstructions per load/storenumber of loads per load missnumber of L1 LD misses per L2 LD missnumber of stores per store missnumber of L1 ST misses per L2 ST missnumber of loads per L2 load missnumber of stores per L2 store missnumber of loads per TLB missnumber of loads/stores per TLB missnumber of load/stores per D1 missnumber of load/stores per L1 missnumber of load/stores per D2 missnumber of load/stores per L2 missL1 cache hit rateL2 cache hit rateL3 cache hit rateL2 I cache hit rateSnoop hit rateHW Float point instructions per CycleHW floating point rate (HW Flops / WCT)HW floating point / user timeTotal Floating point instructions + FMAs (flips)Flip rate (flips / WCT)Flips / user timeFlips / avg user timeComputation intensityFMA percentageTotal Fixed point operationsFixed point operations per CycleFixed point operations per load/storesBranches mispredicted percentageMemory load trafficMemory load bandwidth per processorMemory trafficMemory bandwidth per processorL3 load trafficL3 load bandwidth per processorL2 load trafficL2 load bandwidth per processorL3 trafficL3 bandwidth per processorL2 trafficL2 bandwidth per processorL3 Load miss rateL2 Load miss rate% TLB misses per cycle% loads from memory per cycle% loads from L3 per cycle% accesses from L2 per cycle% loads from L2 per cycleTotal Loads from L3Total Loads from L2Estimated latency from loads from memoryEstimated latency from memoryEstimated latency from loads from L3Estimated latency from loads from L2Estimated latency from L3Estimated latency from L2Estimated latency from TLB missesWeighted Floating Point InstructionsWeighted flip rateWall clock timeWARNING: Please check file libHPMevents (line %1$d) counter %2$d, event %3$d not valid. (Event ignored) WARNING: counter %1$d, event %2$d not valid. (Event ignored) max pm counters = %3$d max events: counter %1$d: %2$d adding libHPMevents - line %1$d counter %2$d event %3$d %4$s adding counter %1$d event %2$d %3$s WARNING: File HPM_flags.env has incorrect format Line: %d - value ignored ERROR: File HPM_flags.env has incorrect format Line: %1$d - %2$s %3$s User input for memory latency: %d User input for division weight: %d User input for TLB Miss latency: %d User input for L3 latency: %d User input for L35 latency: %d User input for average L3 latency: %d User input for L2 latency: %d User input for L25 latency: %d User input for L275 latency: %d User input for average L2 latency: %d ####### End of User Input ######## WARNING: event set (%1$d) must be between 1 and %2$d. Using default set ERROR: inconsistent file %1$s (line %2$d) ERROR: file %1$s (line %2$d), counter %3$d, event %4$s ERROR: file %1$s (line %2$d), event group name %3$s Instrumented section: %1$d - Label: %2$s - process: %3$d file: %1$s, lines: %2$d <--> %3$d Count: %d Wall Clock Time: %.15g seconds Total time in user mode: %.15g seconds Average duration: %g Standard deviation: %g Exclusive duration: %g seconds WARNING: Measurement error for %s not removed HPM ERROR - hpmGetCounters:HPM ERROR - hpmGetTimeAndCounters:HPM ERROR - hpm_stop:HPM ERROR - hpm_tstop:HPM ERROR: hpmGetTimeAndCounters - time information not available HPM ERROR: hpm_init - time information not available HPM ERROR: hpm_terminate - time information not available HPM ERROR: hpm_stop - time information not available HPM ERROR: hpm_tstop - time information not available HPM ERROR: push_instPt_stack - time information not available HPM ERROR - compute_get_counters_error:HPM ERROR: hpm_init - invalid filename character in program name: %s Total execution time of instrumented code (wall time): %.15g seconds HPM output in %s HPM ERROR: hpm_stop, stack is empty HPM ERROR: hpm_tstop, stack is empty HPM WARNING: Stack out of order: Expecting inst pt %1$d instead of inst pt %2$d HPM ERROR - pm_stop_mythread:HPM ERROR - pm_get_data_mygroup:HPM ERROR - pm_get_data_mythread:HPM ERROR - pm_start_mythread:HPM WARNING: Inst ID %1$d ignored. Inst ID must be between %2$d and %3$d. 1FLOP instruction + 2 FMAs + 4 FDIVs + 4 FSQRTsFlop rate (flops / WCT)Flops / user timeFlops / avg user timeInvalid set number specified in the sequence at position %d Set: %d Group: %d Counting duration: %1d.%2.9d seconds HPM ERROR - pm_event_ndata_ts HPM ERROR - pm_dm_evaluate_mx HPM WARNING - Non partition migration DR signalled HPM WARNING - Partition migration %s signalled HPM ERROR - Partition migration signalled %s, retry later HPM ERROR - Partition migration changed processor type HPM ERROR - Reconfiguration signalled, dr_reconfig(%s) failed HPM ERROR - No DLPAR event in progress HPM ERROR - Cancellation of DLPAR event in the check phase HPM WARNING - Partition migration changed processor frequency, old %1$d, new %2$d hpmstat ERROR - pm_delete_program:ERROR - pm_delete_program_mygroup:Invalid set %s (must be a number) options -m and -D cannot be used with options -g or -s Workload context: Derived metric group: %s Counting mode: unsupported normalization base '%s': should be 'time', 'purr' or 'spurr'. normalization base '%s' is not supported on this processor. Invalid mode qualifier '%c' for set '%s'. Invalid mode qualifier '%c' for group '%s'. Invalid mode qualifier '%c' for metric '%s'. Invalid mode qualifier '%c' for metric group '%s'. Invalid metric name '%s'. Invalid metric group name '%s'. Normalization base: %s WARNING: not all events were counted. WARNING: low sample count (%d < %d), results may be inaccurate. usage: hpmstat [-dHkrTUux] [-b time_base] [-D metrics] [-g event_groups] [-m metric_groups] [-o file] [-s set] [-@ ALL | WparName] interval count hpmstat [-h] where: interval Counting time interval in seconds or microseconds, with a default value of 1. count Number of iterations to count. The default is 1 with an interval in seconds, and infinity when the option -U is specified. -b time_base Selects a base for the data normalization. The available bases are as follows: time timebase purr PURR time (when available) spurr SPURR time (when available) The default value is "time". -d Adds detailed set counts for counter multiplexing mode. This option is ignored if the multiplexing mode is not specified. -D metrics Selects a list of derived metrics to be evaluated. Each derived metric can be qualified by a counting mode as follows: metric:counting_modes (See the -m option for available counting modes.) -g event_groups Lists a predefined group of events or a comma-separated list of event group names or numbers. When a comma-separated list of groups is used, the counter multiplexing mode is selected. Each event group can be qualified by a counting mode as follows: event_group:counting_modes (See the -m option for available counting modes.) -H Counts hypervisor activity only. -h Displays help message. -k Counts system activity only. -m metrics_groups Selects a list of derived metric groups to be evaluated. The "default" derived metric group refers to all derived metrics that do not belong to a specific derived metric group. Each metric group can be qualified by a counting mode as follows: metric_group_name:counting_modes. The available counting modes are as follows: u user mode k kernel mode h hypervisor mode r runlatch mode n nointerrupt mode -o file Output file name. -r Enables runlatch and disables counts while executing in idle cycle. -s set Lists a predefined set of events or a comma-separated list of sets (1 to %d, or 0 to select all. See the pmlist command.) When a comma-separated list of sets is used, the counter multiplexing mode is selected. Each set can be qualified by a counting mode as follows: event_set:counting_modes (See the -m option for available counting modes.) -T Write time stamps instead of time in seconds. -U Puts counting time interval in microseconds. This option is ignored if the counter multiplexing mode is specified. -u Counts user activity only. -x Displays results in XML format. -@ ALL Counts all active WPARs and prints per-WPAR reports. -@ WparName Counts specified WPAR only. could not find WPAR '%s'.hpmstat ERROR - pm_get_wplist:hpmstat ERROR - pm_set_program_wp_mm:hpmstat ERROR - pm_tstart_wp:HPM WARNING - Non partition migration/hibernation DR signalled HPM WARNING - Partition migration/hibernation %s signalled HPM ERROR - Partition migration/hibernation signalled %s, retry later HPM ERROR - Partition migration/hibernation changed processor type HPM WARNING - Partition migration/hibernation changed processor frequency, old %1$d, new %2$d Set Not Supported PMU access is disabled in the system. No Authorization to run %s. hpmstat ERROR - Must be root to count globally. Failed to open %s file. hpmstat ERROR - Counters reserved. hpmstat ERROR - Cannot lock %s file. hpmstat ERROR - Cannot unlock %s file.