ISO8859-1 I^qL%  R   2w/b=fOMn M nZ S q #>C`}PU$DR@[+J SDefines the Lightweight Memory Trace (LMT) state. Default: 1; Range: 0 or 1; Value of 1 means LMT is enabled. To be effective, any change of state requires a subsequent bosboot and system reboot. Specifies the memory trace buffer size for common events of Lightweight Memory Trace (LMT) which provides system trace information for First Failure Data Capture (FFDC). Recorded events are saved in system dump and/or reported through user commands. Default: value based on data generation under a reference system-wide activity and HW and system characteristics; Range: 1 (unit of 4KB pagesize) to a value based on HW and system characteristics. User controlled dynamic buffer size changes are supported only on the 64 bit kernel. Specifies the memory trace buffer size for rare events of Lightweight Memory Trace (LMT) which provides system trace information for First Failure Data Capture (FFDC). Recorded events are saved in system dump and/or reported through user commands. Default: value based on data generation under a reference system-wide activity and HW and system characteristics; Range: 1 (unit of 4KB pagesize) to a value based on HW and system characteristics. User controlled dynamic buffer size changes are supported only on the 64 bit kernel. Specifies whether no-execute protection should be enabled for the kernel heap. With protection enabled, any attempt to execute code in the protected heap will result in a kernel exception. Default: 0; Range: 0 to 1. Specifies whether no-execute protection should be enabled for the mbuf heap. With protection enabled, any attempt to execute code in the protected heap will result in a kernel exception. Default: 0; Range: 0 to 1. Specifies whether no-execute protection should be enabled for kernel data regions. With protection enabled, any attempt to execute code in the protected regions will result in a kernel exception. Default: 1; Range: 0 to 1. Specifies the Performance Monitor PM_CYC and software event sampling frequency multiplier as a means to control the trace sampling frequency. Default: 1; Range: 1 to 100. Specifies the Performance Monitor PM_* event sampling frequency multiplier as a means to control the trace sampling frequency. Default: 1; Range: 1 to 10000. Specifies the minimum number of completed instructions between Performance Monitor event samples as a means to control the trace sampling frequency. Default: 1000; Range: 1 to 2G-1. Specifies whether kernel recovery should be enabled. With kernel recovery enabled, the system may recover and continue operation after some unexpected kernel errors. Specifies whether kernel recovery should initiate recovery or halt the system on kernel errors. Range 0 or 1; Value of 0 halts system. Value of 1 initiates recovery. This tunable is only effective if recovery_framework is enabled. Specifies whether kernel recovery should enter the kernel debugger on kernel errors. This tunable is only effective if recovery_framework is enabled. Default: 0; Range: -1 to 3; -1 : Enter KDB on any kernel error 0 : Do not enter KDB 1 : Enter KDB when error impact is greater than or equal to RWA_NO_IMPACT 2 : Enter KDB when error severity is greater than or equal to RWA_COMPONENT_INTEGRITY 3 : Enter KDB when error severity is greater than or equal to RWA_SYSTEM_INTEGRITY Defines the Lightweight Memory Trace (LMT) state. Value of 1 means LMT is enabled. To be effective, any change of state requires a subsequent bosboot and system reboot. Specifies the memory trace buffer size for common events of Lightweight Memory Trace (LMT) which provides system trace information for First Failure Data Capture (FFDC). The default value is based on data generation under a reference system-wide activity and HW and system characteristics. The range upper limit is based on HW and system characteristics and depends on the current value of mtrc_rarebufsize because they share the LMT resource. Recorded events are saved in system dump and/or reported through user commands. Specifies the memory trace buffer size for rare events of Lightweight Memory Trace (LMT) which provides system trace information for First Failure Data Capture (FFDC). The default value is based on data generation under a reference system-wide activity and HW and system characteristics. The range upper limit is based on HW and system characteristics and depends on the current value of mtrace_commonbufsize because they share the LMT resource. Recorded events are saved in system dump and/or reported through user commands. Specifies whether no-execute protection should be enabled for the kernel heap. With protection enabled, any attempt to execute code in the protected heap will result in a kernel exception. Specifies whether no-execute protection should be enabled for the mbuf heap. With protection enabled, any attempt to execute code in the protected heap will result in a kernel exception. Specifies whether no-execute protection should be enabled for kernel data regions. With protection enabled, any attempt to execute code in the protected regions will result in a kernel exception. Specifies the Performance Monitor PM_CYC and software event sampling frequency multiplier as a means to control the trace sampling frequency. N/A Specifies the Performance Monitor PM_* event sampling frequency multiplier as a means to control the trace sampling frequency. N/A Specifies the minimum number of completed instructions between Performance Monitor event samples as a means to control the trace sampling frequency. N/A Specifies whether kernel recovery should be enabled. With kernel recovery enabled, the system may recover and continue operation after some unexpected kernel errors. N/A Specifies whether kernel recovery should initiate recovery or halt the system on kernel errors. A value of 0 halts system. A value of 1 initiates recovery. This tunable is only effective if recovery_framework is enabled. Specifies whether kernel recovery should enter the kernel debugger on kernel errors. This tunable is only effective if recovery_framework is enabled. Possible choices are : -1 : Enter KDB on any kernel error 0 : Do not enter KDB 1 : Enter KDB when error impact is greater than or equal to RWA_NO_IMPACT 2 : Enter KDB when error severity is greater than or equal to RWA_COMPONENT_INTEGRITY 3 : Enter KDB when error severity is greater than or equal to RWA_SYSTEM_INTEGRITY Specifies the kernel recovery actions per minute average threshold. "This tunable is only effective if recovery_framework is enabled. The system stops if the average number of kernel recovery actions per minute exceeds this value. Specify a value of 0 to disable kernel recovery action threshold checking. "Specifies whether block IO device statistics collection should be enabled or not. This tunable is useful in analyzing performance/utilization of various block IO devices. If this tunable is enabled, we can use iostat -b to show IO statistics for various block IO devices. Possible Value: 1 : Enabled 0 : Disabled Allows/Dis-Allows system-wide Performance Monitor event-sampling for non-privileged users. With tprof_evt_system enabled (value 1), non-privileged users can use tprof and pmctl to do system-wide Performance Monitor event-sampling. When disabled (value 0), non-privileged users can do event-sampling only for processes started with -y option of tprof and pmctl. In the disabled mode, non-privileged users will not be able to do event-sampling of kernel and kernel extensions. Specifies whether timestamp update of /dev/null device should be delayed. This tunable provides performance improvement when multiple threads are trying to access the /dev/null device file concurrently.