Scheduling

disp_rechoose_interval

Description

Similar to the previous rechoose_interval parameter, this parameter specifies the amount of time before a process is deemed to have lost all affinity for the last CPU it ran on. However, this parameter is set in more granular time increments. This parameter should be used instead of the deprecated rechoose_interval parameter, but the rechoose_interval parameter is still accepted if it is set in the /etc/system file.

After this interval expires, any CPU is considered a candidate for scheduling a thread. This parameter does not apply to threads in the real-time class, but applies to threads in all other scheduling classes.

Use mdb if you want to change the value of this parameter by using the following steps:

1. Convert nanoseconds to unscaled time. For example, to convert a 5000000 nanosecond based value to unscaled time, use the following syntax:

   # mdb -kw
   .
   .
   .
   > 0t5000000::time -u
       0xb6a444

2. Set disp_rechoose_interval to the unscaled time value. For example, provide the value that was returned in preceding step.

       > disp_rechoose_interval /Z 0xb6a444
       disp_rechoose_interval:  0x447d998   =       0xb6a444

3. Verify that disp_rechoose_interval has been set to the right value. For example:

       > disp_rechoose_interval::print
       0xb6a444

Data Type

Signed integer

Default

3

Range

0 to MAXINT

Dynamic?

Yes

Validation

None

When to Change

When caches are large, or when the system is running a critical process or a set of processes that seem to suffer from excessive cache misses not caused by data access patterns.

Consider using the processor set capabilities or processor binding before changing this parameter. For more information, see psrset(1M) or pbind(1M).

Commitment Level

Unstable

Change History

For information, see disp_rechoose_interval.