The cachefs_readahead parameter specifies the number of blocks to read ahead of the current read block in the filesystem. The blocks are read asynchronously.

Default: 1 (0x1)

Range: 0 - 10

The cachefs_max_threads parameter specifies the maximum number of asynchronous I/O daemons that can be run per cachefs mount.

Default: 5 (0x5)

Range: 1 - 10

The nbuf parameter specifies the number of buffer headers in the filesystem buffer cache. The actual memory associated with each buffer header is dynamically allocated as needed. The memory can be of varying size: 1 to 128 blocks (512 to 64KB) for write operations, and 1 to 1024 blocks (512 to 512KB) for read operations.

The system uses the filesystem buffer cache to optimize filesystem I/O requests. The buffer memory caches blocks from the disk, and the blocks that are used frequently stay in the cache. This helps avoid excess disk activity.

Buffers are used only as transaction headers. When the input or output operation has finished, the buffer is detached from the memory it mapped and the buffer header becomes available for other uses. Because of this, a small number of buffer headers is sufficient for most systems. If nbuf is set to 0, the system automatically configures nbuf for average systems. There is little overhead in making it larger for non-average systems.

The nbuf parameter is defined in /var/sysgen/mtune.

Default: 0 (Automatically configured if set to 0)

Formula:

32-bit kernels: min(100 + (KiloBytes of memory) / 160, 6000)

64-bit kernels: min(100 + (KiloBytes of memory) / 160, 600000)

The automatic configuration is adequate for average systems. If you see dropping cache-hit rates in sar and osview output, increase this parameter. Also, if you have directories with a great number of files (over 1000), you may wish to raise this parameter.

The recommended initial value for large (64 processors or greater) systems is 2000.

The  min_bufmem parameter specifies the minimum amount of memory held by filesystem metadata that is cached in the buffer cache when the system runs into low memory conditions. This is expressed in pages of memory.

Default: two percent of physical memory

Change this value when there is a high I/O load on your machine but you still need to access directory information quickly.

The callout_himark parameter specifies the maximum number of callout table entries system-wide. The callout table is used by device drivers to provide a timeout to make sure that the system does not hang when a device does not respond to commands.

This parameter is defined in /var/sysgen/mtune and has the following formula:

nproc + 32

where:

nproc is the maximum number of processes, system-wide.

Default: 0 (Automatically configured if set to 0)

Formula: nproc + 32

Range: 42 - 1100

Increase this parameter if you see console error messages indicating that no more callouts are available.

The ncallout parameter specifies the number of callout table entries at boot time. The system will automatically allocate one new callout entry if it runs out of entries in the callout table. However, the maximum number of entries in the callout table is defined by the callout_himark parameter.

Default: 40

Range: 20–1000

Update ncallout if you are running an unusually high number of device drivers on your system. Note that the system automatically allocates additional entries when needed, and releases them when they are no longer needed.

The reserve_ncallout parameter specifies the number of reserved callout table entries. These reserved table entries exist for kernel interrupt routines when the system has run out of the normal callout table entries and cannot allocate additional entries.

Default: 5

Range: 0-30

The ncsize parameter controls the size of the name cache. The name cache is used to allow the system to bypass reading directory names out of the main buffer cache. A name cache entry contains a directory name, a pointer to the directory's in-core inode and version numbers, and a similar pointer to the directory's parent directory in-core inode and version number.

Default: 0 (Automatically configured if set to 0)

Range: 268-1M

The ndquot parameter controls disk quotas.

Default: 0 (Automatically configured if set to 0)

Range: 268-6200

The nproc parameter specifies the number of entries in the system process (proc) table. Each running process requires an in-core proc structure. Thus, nproc is the maximum number of processes that can exist in the system at any given time.

The default value of nproc is based on the amount of memory on your system. To find the currently auto-configured value of nproc, use the systune command.

The nproc parameter is defined in /var/sysgen/mtune.

Default: 0 (Automatically configured if set to 0)

Range: 30-1M

Increase this parameter if you see an overflow in the sar -v output for the proc -sz ov column or you receive the operating system message:

no more processes

This means that the total number of processes in the system has reached the current setting. If processes are prevented from forking (being created), increase this parameter. A related parameter is maxup.

The recommended initial value for large (64 processors or greater) systems is 8000.

If a process cannot fork, make sure that this is system-wide and not just a user ID problem (see the maxup parameter).

If nproc is too small, processes that try to fork receive the operating system error:

EAGAIN: No more processes

The shell also returns a message:

fork failed: too many processes

If a system daemon such as sched, vhand, init, or bdflush cannot allocate a process table entry, the system halts and displays:

No process slots

The maxpmem parameter specifies the amount of physical memory (in pages) that the system can recognize. The page size is 4 KB for 32 bit IRIX version 5 or 6 kernels and 16 KB for 64 bit IRIX version 6 kernels. If it is set to zero (0), the system will use all available pages in memory. A value other than zero defines the physical memory size (in pages) that the system will recognize.

This parameter is defined in /var/sysgen/mtune.

Default: 0 (Automatically configured if set to 0)

Range: 1024 pages—total amount of memory

You do not need to change this parameter, except when benchmarks require a specific system memory size less than the physical memory size of the system. This is primarily useful to kernel developers. You can also boot the system with this command added on the boot command line to achieve the same effect:

maxpmem = memory_size

The syssegez parameter specifies the maximum number of pages of dynamic system memory.

Default: 0 (Autoconfigured if set to 0)

32-bit Range: 0x2000 - 0x20000

64-bit Range: 0x2000 - 0x10000000

Increase this parameter correspondingly when maxdmasz is increased, or when you install a kernel driver that performs a lot of dynamic memory allocation.

The recommended initial value for large (64 processors or greater) systems is 0xfe800.

The maxdmasz parameter specifies the maximum DMA transfer expressed in pages of memory. This amount must be less than the value of syssegsz and maxpmem.

Default: 1025 (32-bit kernels)

Default: 257 (64-bit kernels)

32-bit Range: 1 - syssegsz (maximum 0x20000)

64-bit Range: 1 - syssegsz (maximum 0x10000000)

Change this parameter when you need to be able to use very large read or write system calls, or other system calls that perform large scale DMA. This situation typically arises when using optical scanners, film recorders or some printers.

Note that the actual DMA transfer size is one less page than the number of pages set in this parameter. Thus, to allow a full DMA transfer of 16 MB (1024 pages), set this parameter to 1025.

The recommended initial value for large (64 processors or greater) systems is 0x2001.

The mbmaxpages parameter specifies the maximum number of single page clusters that can be allocated to network buffers. This limits the total memory that network buffers (mbufs) can consume.

Default: 1/4 of physical memory

Range: Default - Total System Memory (in pages)

For most workstations and small servers the default is sufficient. For very large systems, larger values are appropriate. A 0 value tells the kernel to set the value (1/4 of physical memory) based on the amount of physical memory configured in the hardware.

The minimum that can be set with the systune command is 16 pages. The maximum is 8192 pages for 32-bit systems and 2097152 pages for 64-bit systems.

The ecc_recover_enable parameter, when set to 0, directs the system not to attempt to recover from multibit errors. If it is set greater than 0, the parameter value is taken as a number of seconds. The system is directed to attempt recovery, but not in the event of more than 32 errors in each number of seconds specified by this parameter.

Default: 60 seconds (0x3c)

utraces are a lightweight tracing mechanism used to collect kernel debugging information. The utrace_bufsize parameter selects the number of 48 byte utrace entries stored for each CPU. Setting this parameter to 0 disables trace collection. Only buffer sizes of 0 and 2048 are supported.

Default: 0

Range: 0-0x7fffffff

The dump_level parameter specifies the dump level and controls the data dumped in a system crash. Setting this parameter to 0 dumps only putbuf,errbuf, and low memory during a panic. 1 also dumps static, dynamic, and backtrace kernel pages along with pfdats (if applicable); 2 also dumps buffer cache pages; 3 also dump remaining in-use pages; and 4 also dumps free pages.

Default: 0

Range: 0-4

The recommended initial value for large (64 processors or greater) systems is 3.

The buf_relse_age parameter specifies when a buffer cache page needs to be released. If a file's data had been untouched for at least buf_relse_age seconds, then the data is released.

This run-time settable variable may be useful in systems in which some nodes have plenty of free memory and others have consumed all local memory. If the latter contains a lot of very aged buffer cache data, the normal buffer releasing system may not release the data because the system looks at memory globally and does not distinguish memory per node.

Default: 0 (0 means the feature is not used.)

Range: 0-0x7fffffff seconds

A recommended setting is difficult to provide as the load and needs of every system are different. Any setting under 30 seconds is too aggressive, and any under 10 seconds causes a lot of thrashing.

A workable solution is to set buf_relse_age to 600 seconds (5 minutes), to study the displays of the bufview and osview commands over time in order to observe how much memory is available, and then to adjust the buf_relse_age value up or down until it appears to be satisfactory.

The maxup parameter defines the number of processes allowed per user login. This value should always be at least 20 less than nproc.

Default: 150 processes

Range: 15–30000 (but never more than nproc minus 20)

Increase this parameter to allow more processes per user. In a heavily loaded time-sharing environment, you may want to decrease the value to reduce the number of processes per user.

The recommended initial value for large (64 processors or greater) systems is 8000.

The ngroups_max parameter specifies the maximum number of multiple groups to which a user may simultaneously belong.

The constants NGROUPS_UMIN <= ngroups_max <= NGROUPS_UMAX are defined in </usr/include/sys/param.h>. NGROUPS_UMIN is the minimum number of multiple groups that can be selected at lboot time. NGROUPS_UMAX is the maximum number of multiple groups that can be selected at lboot time and is the number of group-id slots for which space is set aside at compile time. NGROUPS, which is present for compatibility with networking code (defined in </usr/include/sys/param.h>), must not be larger than ngroups_max.

Default: 16

Range: 0-32

The default value is adequate for most systems. Increase this parameter if your system has users who need simultaneous access to more than 16 groups.

The maxwatchpoints parameter sets the maximum number of watchpoints per process. Watchpoints are set and used via the proc(4) filesystem. This parameter specifies the maximum number of virtual address segments to be watched in the traced process. This is typically used by debuggers.

Default: 100

Range: 1-1000

Raise maxwatchpoints if your debugger is running out of watchpoints.

The nprofile parameter specifies the maximum number of disjoint text spaces that can be profiled using the sprofil(2) system call. This is useful if you need to profile programs using shared libraries or profile an address space using different granularities for different sections of text.

Default: 100

Range: 100-10000

Change nprofile if you need to profile more text spaces than are currently configured.

The maxsymlinks parameter defines the maximum number of symbolic links that will be followed during filename lookups (for example, during the open(2) or stat(2) system calls) before ceasing the lookup. This limit is required to prevent loops where a chain of symbolic links points back to the original filename.

Default: 30

Range: 0-50

Change this parameter if you have pathnames with more than 30 symbolic links.

The ncargs parameter specifies the maximum size of arguments in bytes that may be passed during an exec(2) system call.

This parameter is specified in /var/sysgen/mtune.

Default: 20480

Range: 5120–262144

The default value is adequate for most systems. Increase this parameter if you get the following message from exec(2), shell(1), or make(1):

E2BIG arg list too long

Setting this parameter too large wastes memory (although this memory is pageable) and may cause some programs to function incorrectly. Also note that some shells may have independent limits smaller than ncargs.

The rlimit_core_cur parameter specifies the current limit to the size of core image files for the given process.

Default: 0x7fffffffffffffff (9223372036854775807 bytes)

Range: 0–0x7fffffffffffffff

Change this parameter when you want to place a cap on core file size.

The rlimit_core_max parameter specifies the maximum limit to the size of core image files.

Default: 0x7fffffffffffffff (9223372036854775807 bytes)

Range: 0–0x7fffffffffffffff

Change this parameter when you want to place a maximum restriction on core file size. rlimit_core_cur cannot be larger than this value.

The rlimit_cpu_cur parameter specifies the current limit to the amount of CPU time in seconds that may be used in executing the process.

Default: 0x7fffffffffffffff (9223372036854775807 seconds)

Range: 0–0x7fffffffffffffff

Change this parameter when you want to place a cap on CPU usage.

The rlimit_cpu_max parameter specifies the maximum limit to the amount of CPU time that may be used in executing a process.

Default: 0x7fffffffffffffff (9223372036854775807 minutes)

Range: 0–0x7fffffffffffffff

Change this parameter when you want to place a maximum restriction on general CPU usage.

The rlimit_data_cur parameter specifies the current limit to the data size of the process.

Default: 0 (auto-configured to rlimit_vmem_cur * NBPP) (0x20000000)

Range: 0–0x7fffffffffffffff

Change this parameter when you want to place a cap on data segment size.

The recommended initial value for large (64 processors or greater) systems is 0 ll.

The rlimit_data_max parameter specifies the maximum limit to the size of data that may be used in executing a process.

Default: 0 (auto-configured to rlimit_vmem_cur * NBPP) (0x20000000)

Range: 0–0x7fffffffffffffff

Change this parameter when you want to place a maximum restriction on the size of the data segment of any process.

The recommended initial value for large (64 processors or greater) systems is 0 ll.

The rlimit_fsize_cur parameter specifies the current limit to file size on the system for the process.

Default: 0x7fffffffffffffff (9223372036854775807 bytes)

Range: 0–0x7fffffffffffffff

Change this parameter when you want to place a limit on file size.

The rlimit_fsize_max parameter specifies the maximum limit to file size on the system.

Default: 0x7fffffffffffffff (9223372036854775807 bytes)

Range: 0–0x7fffffffffffffff

Change this parameter when you want to place a maximum size on all files.

The rlimit_nofile_cur parameter specifies the current limit to the number of open file descriptors that may be used in executing the process.

Default: 200

Range: 40–0x7fffffffffffffff

Change this parameter when you want to place a cap on the number of open file descriptors.

The rlimit_nofile_max parameter specifies the maximum limit to the number of open file descriptors that may be used in executing a process.

Default: 2500

Range: 0–0x7fffffffffffffff

Change this parameter when you want to place a maximum restriction on the number of open file descriptors.

The rlimit_pthread_cur parameter specifies the current amount of threads that a process creates.

Default: 1024 (0x400)

The rlimit_pthread_max parameter specifies the maximum value rlimit_pthread_cur may hold.

Default: 1024 (0x400)

The rlimit_rss_cur parameter specifies the current limit to the resident set size (the number of pages of memory in use at any given time) that may be used in executing the process. This limit is the larger of the results of the following two formulae:

physical_memory_size - 4 MB

physical_memory_size * 9/10

Default: 0. If rlimit_rss_cur is 0, it is automatically configured to the larger of the following formulae:

  physical_memory_size - 4 MB
  physical_memory_size * 9/10)

Range: 0–0x7fffffffffffffff

Change this parameter when you want to place a cap on the resident set size of a process.

The recommended initial value for large (64 processors or greater) systems is 0 ll.

The rlimit_rss_max parameter specifies the maximum limit to the resident set size that may be used in executing a process.

Default: 0. If rlimit_rss_max is 0, it is automatically configured to the larger of the following formulae:

  physical_memory_size - 4 MB
  physical_memory_size * 9/10)

Range: 0–0x7fffffffffffffff

Change this parameter when you want to place a maximum restriction on resident set size.

The recommended initial value for large (64 processors or greater) systems is 0 ll.

The rlimit_stack_cur parameter specifies the current limit to the amount of stack space that may be used in executing the process.

Default: 64 MB (0x04000000)

Range: 0–0x7fffffffffffffff)

Change this parameter when you want to place a limit on stack space usage.

The recommended initial value for large (64 processors or greater) systems is (0x04000000) ll.

The rlimit_stack_max parameter specifies the maximum limit to the amount of stack space that may be used in executing a process.

Default: rlimit_vmem_cur * NBPP (0x20000000)

Range: 0–0x7fffffffffffffff

Change this parameter when you want to place a maximum restriction on stack space usage.

The recommended initial value for large (64 processors or greater) systems is (0x20000000) ll.

The rlimit_vmem_cur parameter specifies the current limit to the amount of virtual memory that may be used in executing the process.

Default: 0. If rlimit_vmen_cur is 0, it is automatically configured to the following:

  0x7fffffffffffffff for Origin 3000, Origin 2000, Challenge, O200, and Onyx2 systems

  0x80000000 for Indgio 2, Power Indgio 2, Octane, and O2 systems

 0x40000000 for smaller systems

Range: 0–0x7fffffffffffffff

Change this parameter when you want to place a cap on virtual memory usage.

The recommended initial value for large (64 processors or greater) systems is 0 ll.

The rlimit_vmem_max parameter specifies the maximum limit to the amount of virtual memory that may be used in executing a process.

Default: 0. If rlimit_vmen_max is 0, it is automatically configured to the following:

  0x40000000 for smaller machines

  0x80000000 for larger machines

  0x7fffffffffffffff for the SN architecture

Range: 0–0x7fffffffffffffff

Change this parameter when you want to place a maximum restriction on virtual memory usage or if your swap space is larger than the current parameter value.

The recommended initial value for large (64 processors or greater) systems is 0 ll.

The rsshogfrac parameter specifies the percentage of memory allotted for resident pages.

The number of physical memory pages occupied by a process at any given time is called its resident set size (RSS). The limit to the RSS of a process is determined by its allowable memory-use resource limit. rsshogfrac is designed to guarantee that even if one or more processes are exceeding their RSS limit, some percentage of memory is always kept free so that good interactive response is maintained. The percentage of total memory hogged is 100% minus rsshogfrac times maximum memory or 100 MB, whichever is less.

Processes are permitted to exceed their RSS limit until either of the following occur:

Default: 75% of total memory

Range: 0–100% of total memory

The default value is adequate for most systems.

The recommended initial value for large (64 processors or greater) systems is 99.

The rsshogslop parameter specifies the number of pages above the resident set maximum that a process may use. To avoid thrashing (a condition where the computer devotes 100% of its CPU cycles to swapping and paging), a process can use up to rsshogslop more pages than its resident set maximum (see “ Resource Limits Parameters”).

This parameter is specified in /var/sysgen/mtune. For more information, see the rsshogfrac resource limit.

Default: 20

The default value is adequate for most systems.

The shlbmax parameter specifies the maximum number of shared libraries with which a process can link.

This parameter is specified in /var/sysgen/mtune.

Default: 8

Range: 3–32

The default value is adequate for most systems. Increase this parameter if you see the following message from exec(2):

ELIBMAX cannot link

The cpulimit_gracetime parameter establishes a grace period for processes that exceed the cpu time limit. It is the number of seconds that a process is allowed to run after exceeding the limit set in the rlimit_cpu_cur parameter. When cpulimit_gracetime is not set (that is, it is zero), any process that exceeds either the process or job cpu limit is sent a SIGXCPU signal. The kernel continues to send SIGXCPU signals periodically as long as that process continues to execute. Since a process can register to handle SIGXCPU, the process can effectively ignore the CPU limit.

If cpulimit_gracetime is set to a nonzero value via the systune(1M)command, the behavior changes. The kernel sends a SIGXCPU signal to a process one time when it exceeds the process cpu limit. The process can register for this signal and then perform any needed cleanup and shutdown operations. If the process is still running after accruing cpulimit_gracetime more seconds of cpu time, the kernel terminates the process with a SIGKILL signal.

Default: 0

Range: 0-86400 (seconds)

The recommended initial value is site dependent.

The bdflushr parameter specifies how many buffers will be examined each time bdflush runs; bdflush performs periodic flushes of dirty filesystem buffers. It is parallel and similar to vfs_syncr. The bdflush daemon runs once per second.

This parameter is specified in /var/sysgen/mtune. For more information, see the autoup kernel parameter.

Default: 5

Range: 1–31536000

The default value is adequate for most systems.

The gpgsmsk parameter specifies the mask used to determine if a given page may be swapped. Whenever the pager (vhand) is run, it decrements software reference bits for every active page. When a process subsequently references a page, the counter is reset to the limit (NDREF, as defined in /usr/include/sys/immu.h). When the pager is looking for pages to steal back (if memory is in short supply), it takes only pages whose reference counter has fallen to gpgsmsk or below.

This parameter is specified in /var/sysgen/mtune.

Also see /usr/include/sys/immu.h and /usr/include/sys/tuneable.h and the gpgshi and gpgslo kernel parameters for more information.

Default: 2

Range: 0–7

This value is adequate for most systems.

If the value is greater than 4, pages are written to the swap area earlier than they would be with the default value of gpgsmsk. Thus swapping/paging may occur before it should, unnecessarily using system resources.

When the vhand daemon (page handler) is stealing pages, it stops stealing when the amount of free pages is greater than gpgshi.

In other words, vhand starts stealing pages when there are fewer than gpgslo free pages in the system. Once vhand starts stealing pages, it continues until there are gpgshi pages.

If, at boot time, gpgslo and gpgshi are 0, the system sets gpgshi to 8% of the number of pages of memory in the system, and sets gpgslo to one half of gpgshi.

This parameter is specified in /var/sysgen/mtune. For more information, see the kernel parameters gpgsmsk and gpgslo.

Default: 0 (automatically configured to 8% of memory if set to 0)

Range: 30 pages – 1/2 of memory

This value is adequate for most systems.

The recommended initial value for large (64 processors or greater) systems is 2000.

If this parameter is too small, vhand cannot supply enough free memory for system-wide demand.

When the vhand daemon (page handler) executes, it would not start stealing back pages unless there are fewer than gpgslo free pages in the system. Once vhand starts stealing pages, it continues until there are gpgshi pages.

This parameter is specified in /var/sysgen/mtune. For more information, see the gpgshi and gpgsmsk kernel parameters.

Default: 0 (automatically configured to half of gpgshi if set to 0)

Range: 10 pages – 1/2 of memory

This value is adequate for most systems.

The recommended initial value for large (64 processors or greater) systems is 1000.

If this parameter is too small, vhand does not start swapping pages; thus entire processes must be swapped. If this parameter is too large, vhand swaps pages unnecessarily.

The maxlkmem parameter specifies the maximum number of physical pages that can be locked in memory (by mpin(2) or plock(2)) per non-superuser process.

This parameter is specified in /var/sysgen/mtune.

Default: 2000

Range: 0 pages – 3/4 of physical memory

Increase this parameter only if a particular application has a real need to lock more pages in memory.

On multi-user servers, you may want to decrease this parameter and also decrease rlimit_vmem_cur.

When pages are locked in memory, the system cannot reclaim those pages, and therefore cannot maintain the most efficient paging.

The maxfc parameter specifies the maximum number of pages that may be freed by the vhand daemon in a single operation. When the paging daemon (vhand) starts stealing pages, it collects pages that can be freed to the general page pool. It collects, at most, maxfc pages at a time before freeing them. Do not confuse this parameter with gpgshi, which sets the total number of pages that must be free before vhand stops stealing pages.

This parameter is specified in /var/sysgen/mtune.

Default: 100

Range: 50–100

This value is adequate for most systems.

The maxsc parameter specifies the maximum number of pages that may be swapped by the vhand daemon in a single operation. When the paging daemon starts tossing pages, it collects pages that must be written out to swap space before they are actually swapped and then freed into the general page pool. It collects at most maxsc pages at a time before swapping them out.

This parameter is specified in /var/sysgen/mtune.

Default: 100

Range: 8–100

You may want to decrease this parameter on systems that are swapping over NFS ( Network File System). This is always the case for diskless systems to increase performance.

The maxdc parameter is the maximum number of pages which can be saved up and written to the disk at one time.

Default: 100 pages

Range: 1-100

If the system is low on memory and consistently paging out user memory to remote swap space (for example, mounted via NFS), decrease this parameter by not more than ten pages at a time. However, this parameter's setting does not usually make any measurable difference in system performance.

The minarmem parameter represents the minimum available resident memory that must be maintained in order to avoid deadlock.

Default: 0 (Autoconfigured if set to 0)

The automatically configured value of this parameter should always be correct for each system. You should not have to change this parameter.

The minasmem parameter represents the minimum available swappable memory that must be maintained in order to avoid deadlock.

Default: 0 (Autoconfigured if set to 0)

The automatically configured value of this parameter should always be correct for each system. You should not have to change this parameter.

The numa_paging_node_freemem_low_threshold parameter specifies when to trigger a global system memory accounting procedure to check if vhand should be started.

Default: 40

The scache_pool_size parameter specifies the amount of memory always kept in reserve for use by the paging daemon. The value is the number of kilobytes reserved, which is always rounded up to the next page boundary.

Default: 32

Use caution when changing this parameter. Setting it too low will result in memory deadlocks. Setting it too high will waste memory. If the system panics with the message scache... out of memory, then this parameter should be increased. Otherwise, you should not have to change this parameter.

The tlbdrop parameter specifies the number of clock ticks before a process's wired entries are flushed.

Default: 100

If sar indicates a great deal of transaction lookaside buffer (utlbmiss) overhead in a very large application, you may need to increase this parameter. In general, the more the application changes the memory frame of reference in which it is executing, the more likely increasing tlbdrop will help performance. You may have to experiment somewhat to find the optimum value for your specific application.

The vfs_syncr parameter specifies the number of seconds between runs of the vfs_syncr process.

Default: 30

Range: 1 - 31536000 (1 year)

Under normal circumstances it is not necessary to change this parameter.

The maxpglst parameter specifies the maximum number of pages that can be held in each of the pager's pageout queues.

Default: 0 (Autoconfigured)

Range: 50 - 1000

The zone_accum_limit parameter specifies the percentage of memory (in a node on a NUMA system) that can be accumulated before shaked is kicked off to shake the zone of free memory back into the global pool. This parameter is set to 30%, which means if the amount of free memory kept in zones exceeds 30%, shaked is kicked off.

Default: 30

Range: 0 - 100

The percent_totalmem_64k_pages parameter specifies the percentage of total memory that can be used as an upper limit for the number of 64 KB pages.

Default: 0

Range: 0-100 (%)

Set dynamically to get 64 KB pages.

The recommended initial value for large (64 processors or greater) systems is 0.

---

Note: Other possible values are discussed in “Multiple Page Sizes” in Chapter 10.

---

The nlpages_64k parameter is used to reserve 64 KB pages at boot time. This increases the probability of getting large pages in a low memory system.

Default: 0

Range: 0-memory_size/64*1024, where memory_size is the total amount of memory in the system.

Not recommended unless absolutely necessary.

---

Note: Other possible values are discussed in “Multiple Page Sizes” in Chapter 10.

---

The msgmax parameter specifies the maximum size of a message.

This parameter is specified in /var/sysgen/mtune/msg.

Default: 16 * 1024 (0x4000)

Range: 512-0x8000

Increase this parameter if the maximum size of a message needs to be larger. Decrease the value to limit the size of messages.

The msgmnb parameter specifies the maximum length of a message queue.

This parameter is specified in /var/sysgen/mtune/msg.

Default: 32 * 1024 (0x8000)

Range: msgmax–1/2 of physical memory

Increase this parameter if the maximum number of bytes on a message queue needs to be longer. Decrease the value to limit the number of bytes per message queue.

The msgmni parameter specifies the maximum number of message queues system-wide.

This parameter is specified in /var/sysgen/mtune/msg.

Default: 50

Range: 10–1000

Increase this parameter if you want more message queues on the system. Decrease the value to limit the message queues.

If there are not enough message queues, a msgget(2) system call that attempts to create a new message queue returns the error:

ENOSPC: No space left on device

The msgseg parameter specifies the maximum number of message segments system-wide. A message on a message queue consists of one or more of these segments. The size of each segment is set by the msgssz parameter.

This parameter is specified in /var/sysgen/mtune/msg.

Default: 1536

Modify this parameter to reserve the appropriate amount of memory for messages. Increase this parameter if you need more memory for message segments on the system. Decrease the value to limit the amount of memory used for message segments.

If this parameter is too large, memory may be wasted (saved for messages but never used). If this parameter is too small, some messages that are sent will not fit into the reserved message buffer space. In this case, a msgsnd(2) system call waits until space becomes available.

The msgssz parameter specifies the size, in bytes, of a message segment. Messages consist of a contiguous set of message segments large enough to accommodate the text of the message. Using segments helps to eliminate fragmentation and speed the allocation of the message buffers.

This parameter is specified in /var/sysgen/mtune/msg.

Default: 8

This parameter is set to minimize wasted message buffer space. Change this parameter only if most messages do not fit into one segment with a minimum of wasted space.

If you modify this parameter, you may also need to change the msgseg parameter.

If this parameter is too large, message buffer space may be wasted by fragmentation, which in turn may cause processes that are sending messages to sleep while waiting for message buffer space.

The msgtql parameter specifies the maximum number of message headers system-wide, and thus the number of outstanding (unread) messages. One header is required for each outstanding message.

This parameter is specified in /var/sysgen/mtune/msg.

Default: 40

Range: 10–4000

Increase this parameter if you require more outstanding messages. Decrease the value to limit the number of outstanding messages.

If this parameter is too small, a msgsnd(2) system call attempting to send a message that would put msgtql over the limit waits until messages are received (read) from the queues.

The semmni parameter specifies the maximum number of semaphore identifiers in the kernel. This is the number of unique semaphore sets that can be active at any given time. Semaphores are created in sets; there may be more than one semaphore per set.

This parameter is specified in /var/sysgen/mtune/sem.

Default: 300

Increase this parameter if processes require more semaphore sets. Increasing this parameter to a large value requires more memory to keep track of semaphore sets. If you modify this parameter, you may need to modify other related parameters.

The recommended initial value for large (64 processors or greater) systems is 2000.

The semmsl parameter specifies the maximum number of semaphores per semaphore identifier.

This parameter is specified in /var/sysgen/mtune/sem.

Default: 100

Increase this parameter if processes require more semaphores per semaphore identifier.

The semopm parameter specifies the maximum number of semaphore operations that can be executed per semop() system call. This parameter permits the system to check or modify the value of more than one semaphore in a set with each semop() system call.

This parameter is specified in /var/sysgen/mtune/sem.

Default: 100

Change this parameter to increase/decrease the number of operations permitted per semop() system call. You may need to increase this parameter if you increase semmsl (the number of semaphore sets), so that a process can check/modify all the semaphores in a set with one system call.

The semvmx parameter specifies the maximum value that a semaphore can have.

This parameter is specified in /var/sysgen/mtune/sem.

Default: 32767 (maximum value)

Decrease this parameter if you want to limit the maximum value for a semaphore.

The semaem parameter specifies the adjustment on exit for maximum value, alias semadj. This value is used when a semaphore value becomes greater than or equal to the absolute value of semop(2), unless the program has set its own value.

This parameter is specified in /var/sysgen/mtune/sem.

Default: 16384 (maximum value)

Change this parameter to decrease the maximum value for the adjustment on exit value.

The shmmax parameter specifies the maximum size in bytes of an individual shared memory segment.

This parameter is specified in /var/sysgen/mtune/kernel.

Default: 0

32-bit Range: 0x1000 - 0x7fffffff

64-bit Range: 0x1000 - 0x7fffffffffffffff ll

If the value of shmmax is 0, it will be adjusted to be 80% of the system memory size at boot time. If the default value specifies a number, kernel will obey that.

Keep this parameter small if it is necessary that a single shared memory segment not use too much memory.

The recommended initial value for large (64 processors or greater) systems is (0x4000000) 11.

The shmmin parameter specifies the minimum size in bytes of a shared memory segment.

This parameter is specified in /var/sysgen/mtune/kernel.

Default: 1 byte

Range: 1

Increase this parameter if you want an error message to be generated when a process requests a shared memory segment that is too small.

The shmmni parameter specifies the maximum number of shared memory segments system-wide.

This parameter is specified in /var/sysgen/mtune/kernel.

Default: 0

Range: 5–100000

If shmmni default value is 0, it will be adjusted at boot time to be the same value as the nproc tunable variable. If the default value specifies a number, kernel will obey that.

Increase this parameter by one (1) for each additional shared memory segment that is required, and also if processes that use many shared memory segments reach the shmmni limit.

Decrease this parameter if you need to reduce the maximum number of shared memory segments of the system at any one time. Also decrease it to reduce the amount of kernel space taken for shared memory segments.

The sshmseg parameter specifies the maximum number of attached shared memory segments per process. A process must attach a shared memory segment before the data can be accessed.

This parameter is specified in /var/sysgen/mtune/kernel.

Default: 2000

Range: 1–10000

Increase this parameter if processes need to attach more than the default number of shared memory segments at one time.

The nstrpush parameter defines the maximum number of streams modules that can be pushed on a single stream.

Default: 9 (0x9)

Range: 9-10

Change nstrpush from 9 to 10 modules when you need an extra module.

The nstrintr parameter defines the streams buffers used at interrupt time.

Default: 1024 (0x400)

Range: 32-4096

The strctlsz parameter is the maximum size of the ctl buffer of a streams message. See the getmsg(2) or putmsg(2) man pages for a discussion of the ctl and data parts of a streams message.

Default: 1024 (0x400)

Change strctlsz when you need a larger buffer for the ctl portion of a streams message.

The strmsgsz parameter defines the maximum streams message size. This is the maximum allowable size of the ctl part plus the data part of a message. Use this parameter in conjunction with the strctlsz parameter described above to set the size of the data buffer of a streams message. See the getmsg(2) or putmsg(2) man pages for a discussion of the ctl and data parts of a streams message.

Default: 0x8000

Change this parameter in conjunction with the strctlsz parameter to adjust the sizes of the streams message as a whole and the data portion of the message.

The strholdtime parameter defines the STRHOLDTIME macro in <strsubr.h>. The purpose of this macro is to cut down on overhead on streams drivers. This happens at a cost to system latency.

Default: 50 (0x32)

Range: 0-1000

The strpmonmax parameter defines the maximum number of private streams monitors

Default: 4 (0x4)

Range: 0-1024

The maxsigq parameter specifies maximum number of signals that can be queued. Normally, multiple instances of the same signal result in only one signal being delivered. With the use of the SA_SIGINFO flag, outstanding signals of the same type are queued instead of being dropped.

Default: 64

Range: 32-32767

Raise maxsigq when a process expects to receive a great number of signals and 64 queue places may be insufficient to avoid losing signals before they can be processed. Change maxsigq to a value appropriate to your needs.

The memaff_sched parameter turns the scheduler's memory affinity preference on or off.

Default: 1

Range: 0-1

Change this parameter when you want to turn the scheduler's memory affinity preference on or off.

The slice_size parameter is the default process time slice, expressed as a number of ticks of the system clock. The frequency of the system clock is expressed by the constant Hz, which has a value of 100. Thus each unit of slice_size corresponds to 10 milliseconds. When a process is given control of the CPU, the kernel lets it run for slice_size ticks. When the time slice expires or when the process voluntarily gives up the CPU (for example, by calling pause(2) or by doing some other system call that causes the process to wait), the kernel examines the run queue and selects the process with the highest priority that is eligible to run on that CPU.

The slice_size parameter is defined in /var/sysgen/mtune/disp and has the following formula:

#define slice_size Hz / 30 int slice_size = slice_size

Default: 2 for single CPU systems, 10 for multiprocessor systems

Range: 1–100

If you use the system primarily for compute-intensive jobs and interactive response is not an important consideration, you can increase slice_size. For example, setting slice_size to 10 gives greater efficiency to the compute jobs, since each time they get control of the CPU, they are able to run for 100 milliseconds before getting switched out.

In situations where the system runs both compute jobs and interactive jobs, interactive response time will suffer as you increase the value of slice_size.

The rtcpus parameter controls the quality of the real-time scheduler. If this option is turned on, it enables the IRIX operating system to guarantee a one millisecond response time. Turning it on, especially on large systems, reduces system performance. It is expressed in the number of processors and is turned on if it is greater or equal to the number of processors in the system.

Default: 0 (Off)

Range: 0 - 512

For most machines, the default is sufficient. If the guaranteed response time is required, it should be turned on.

If the efs_inline parameter is set to any value other than 0, any symbolic link data is stored within a file's inode, rather than out-of-line.

Default: 0

Range: 0-1

It should not be necessary to change this variable. Increasing the value improves the lookup speed of symbolic links to some degree; however filesystems thus treated cannot be transferred to systems running releases of the IRIX system prior to version 5.3.

The bdflush_interval parameter specifies the interval at which the bdflush function executes to push dirty file data to disk. It is expressed in 10 microsecond intervals.

Default: 100 (Once per second)

For most machines, the default is sufficient

The cwcluster parameter sets the maximum number of commit write pages to cluster in each push. This parameters applies to NFS I/O only.

Default: 64

It should not be necessary to change this parameter. The automatically configured value is sufficient.

The dwcluster parameter sets the maximum number of delayed-write pages to cluster in each push. This amount cannot be greater than the value of maxdmasz.

Default: 64

It should not be necessary to change this parameter. The automatically configured value is sufficient.

The min_file_pages parameter sets the minimum number of file pages to keep in the cache when memory gets low. It is autoconfigured to 3% of the system's memory if it is 0. When setting this parameter, remember that the page size is 4K in 32-bit kernels and 16K in 64-bit kernels.

Default: 0

It should not be necessary to change this parameter. The automatically configured value is sufficient.

The min_free_pages parameter sets the minimum number of free pages when the number of file pages is above min_file_pages. The default value is gpgshi*2 for systems with less than 600M of memory. For larger memory systems, the default value is gpgshi*4.

Default: 0

The autoup parameter specifies the age, in seconds, that a buffer marked for delayed write must be before the bdflush daemon writes it to disk. This parameter is specified in /var/sysgen/mtune. For more information, see the entry for the bdflushr kernel parameter.

The autoup parameter is also used to determine the age of clean data that is marked inactive and shows up in the gr_osview rmemc command's freec label. The value is computed with the following algorithm: autoup*HZ/2, or a default of 10*100/2 = 500 seconds.

Default: 10

Range: 1–30

This value is adequate for most systems.

The bdevsw_extra parameter specifies an extra number of entries in the block device switch. This parameter is for use by loadable drivers only. If you configured a block device into the system at lboot(1M) time, you will not need to add extra entries to bdevsw.

Default: 21

Range: 1-254

Change this parameter when you have more than 21 block devices to load into dynamically the system. The IRIX system provides 21 spaces in the bdevsw by default.

The cdevsw_extra parameter specifies an extra number of entries in the character device switch. This parameter is for use by loadable drivers only. If you configured a character device into the system at lboot(1M) time, you will not need to add extra entries to cdevsw.

Default: 23

Range: 3-254

Change this parameter when you have more than 23 character devices to load dynamically into the system. The IRIX system provides 23 spaces in the cdevsw by default.

The fmodsw_extra parameter specifies an extra number of entries in the streams module switch. This parameter is for use by loadable drivers only. If you configured a streams module into the system at lboot(1M) time, you will not need to add extra entries to fmodsw.

Default: 20

Range: 0-254

Change this parameter when you have more than 20 streams modules to load dynamically into the system. The IRIX system provides 20 spaces in the fmodsw by default.

The vfssw_extra parameter specifies an extra number of entries in the vnode filesystem module switch. This parameter is for use by loadable drivers only. If you configured a vfs module into the system at lboot time, you will not need to add extra entries to vfssw.

Default: 5

Range: 0-254

Change this parameter when you have more than five virtual filesystem modules to load dynamically into the system. The IRIX system provides five spaces in the vfssw by default.

The munlddelay parameter specifies the timeout in minutes after which to auto-unload loadable modules.

Default: 5 (0x5) minutes

The panic_on_sbe parameter, when set to 1, turns on a special factory debugging mode called single bit errors. When set to 0, this feature is disabled.

Default: 0 (0x0)

Range: 0 or 1

The sbe_log_errors parameter, when set to 1, directs the system to log single bit errors to the SYSLOG. When set to 0, this feature is disabled.

Default: 0 (0x0)

Range: 0 or 1

This parameter applies only to systems containing ECC memory (R8000-based systems such as Challenge and Onyx systems).

The sbe_mfr_override parameter, when set to 1, overrides the default action of disabling single bit errors if the rate of single bit errors exceeds a predetermined limit. When set to 0, this feature is disabled.

Default: 0 (0x0)

Range: 0 or 1

This parameter applies only to systems containing ECC memory (R8000-based systems such as Challenge and Onyx systems).

The sbe_report_cons parameter, when set to 1, directs the system to display single bit errors on the system console. When set to 0, this feature is disabled.

Default: 0 (0x0)

Range: 0 or 1

This parameter applies only to systems containing ECC memory (R8000-based systems such as Challenge and Onyx systems).

The corepluspid parameter, when set to 1, directs the system to name core files with the process ID number appended. When set to 0, this feature is disabled.

Default: 0 (0x0)

Range: 0 or 1

The r4k_div_patch parameter, when set to 1, enables the exec patch code for binaries that have been processed with r4kpp for the divide in branch delay slot problem that occurs on R4000 SC rev 2.2 and 3.0 parts. When set to 0, this feature is disabled.

Default: 0 (0x0)

Range: 0 or 1

The mload_auto_rtsyms parameter, when set to 1, enables the standard automatic loading of the kernel's run-time symbol table. When set to 0, this feature is disabled.

Default: 1 (0x1)

Range: 0 or 1

The xpg4_sticky_dir parameter, when set to 1, specifies that write access to a file does not allow that file to be removed if the sticky bit is set in the directory in which it resides. When set to 0, such files can be removed.

Default: 1 (0x1)

Range: 0 or 1

The tty_auto_strhold parameter, when set to 1, automatically sets STRHOLD on ttys/ptys whenever the line discipline is in canonical & echo mode and automatically clears STRHOLD otherwise. When set to 0, this feature is under user control.

Default: 0 (0x0)

Range: 0 or 1

The reset_limits_on_exec parameter, when set to 1, directs the kernel to reset rlimit values on processes that run as root, in order to prevent non-root processes from enforcing resource limits. When set to 0, this feature is disabled and resource limits are not reset. Modifying this parameter may have security implications.

Default: 1 (0x1)

Range: 0 or 1

The ip26_allow_ucmem parameter, when set to 0, prevents users from accessing uncached system memory on IP26 (POWER Indigo2) systems. When set to 1, this feature is allowed.

Default: 0 (0x0)

Range: 0 or 1

If this parameter is set to 0, attempts to access uncached memory will cause a system panic. If the feature is allowed, there is a substantial memory performance degradation.

The restrict_fastprof parameter, when set to 0, allows users to do fast (1 ms) profiling of programs with prof. When set to 1, this feature is disallowed.

Default: 0 (0x0)

Range: 0 or 1

The reboot_on_panic parameter, when set to 1, specifies that the system should automatically reboot after a kernel panic. This is especially useful for servers or other systems that frequently go unattended or are used remotely, where the user may not conveniently be able to physically reset and reboot the system.

When set to 0, the system must be rebooted from the console.

Systems with IP19, IP20, IP21, IP22, IP25, IP26, IP27, IP28, IP30, IP32 and IP35 processors (check your hinv listing for your processor type) store an environment variable in the PROM monitor called rebound. If you have the rebound variable and set the reboot_on_panic parameter to -1, your system will check the PROM environment for instructions. For desktop systems (IP20, IP22, IP26, IP28, IP30, IP32), if the rebound variable is set to y, then the system will reboot automatically. If the rebound variable is set to n, the system will require manual reset. For the other platforms, IP19, IP21, IP25, IP27, IP35, if the rebound variable is not set to y (IP19, IP21, IP25) and the reboot_on_panic parameter is -1, the default, then the system behaves as though the setting is 1. If you do not have the rebound variable and set the reboot_on_panic parameter to -1, your system behaves as though the setting is 0 and does not automatically reboot. It is important to note that the rebound variable is always checked before the reboot_on_panic parameter and can override it. If the reboot_on_panic parameter is set to 0 and the rebound value is set to y, the reboot_on_panic parameter in effect becomes 1 and your system will automatically reboot.

---

Note: The default for SGI servers and large graphical systems is to reboot after a kernel panic.

---

Default: -1 (automatically reboot according to hardware platform implementation) or 0, depending on processor type. A value of 1 also automatically reboots a system.

Range: -1, 0, or 1

Change this parameter if you wish to automatically reboot after a system panic.

The svr3pipe parameter specifies whether the default SGI line disciplines are used. I parameter, when set to 1, specifies SVR3.2 style pipes, which are unidirectional. When set to 0, SVR4 style pipes are specified, which are bidirectional.

Default: 1 (SVR3.2 style pipes)

Range: 0 or 1

Change this parameter if you wish to take advantage of SVR4 style pipes. SVR3 pipes are the default because they provide faster performance.

Some programs are written so that they perform actions that require superuser privilege. In order to perform these actions, they create a shell in which the user has superuser privilege. Such shells pose a certain manageable risk to system security, but application developers are generally careful to limit the actions taken by the application in these shells. The nosuidshells switch, when set to 0, allows these applications to create superuser-privileged shells. When set to any value other than 0, such shells are not permitted.

Default: 1 (setuid shells not permitted)

Change this switch to allow setuid shells.

The posix_tty_default parameter specifies whether the default IRIX system of line disciplines and settings for serial lines is used. These default settings are different from those specified by POSIX. If the value of this switch is 0, the default SGI line disciplines are used. If the value is set to 1, POSIX line disciplines and settings are used.

Default: 0

Range: 0 or 1

Change this switch if you need to use POSIX line disciplines.

The restricted_chown parameter specifies whether the default SGI line disciplines are used. I switch allows you to decide whether you want to use a BSD UNIX style chown(2) system call or the System V style. Under the BSD version, only the superuser can use the chown system call to give away a file—to change the ownership to another user. Under the System V version, any user can give away a file or directory. If the value of the switch is 0, System V chown is enabled. If the value is not zero, BSD chown is enabled.

Default: 0

Range: 0 or 1

Change this switch to choose which behavior you prefer for the chown(2) system call.

The use_old_serialnum parameter forces the kernel to use the old method (before IRIX Version 5) of calculating a 32-bit serial number for sysinfo -s when the parameter is set to 1. This variable affects only Onyx and Challenge L or XL systems.

Default: 0

Range: 0 or 1

Change this parameter on your Challenge or Onyx system if you need to use some older software that requires a 32-bit serial number.

When the subnetsarelocal parameter is set to 1, all other subnets of a directly-connected subnetted network are considered to be local.

Default: 0

Range: 0 or 1

Change this parameter if no subnetted systems are directly connected to any external network, such as the Internet.

The process_core_cpu_logging parameter, when set to 1, enables a special factory debugging mode that enables the logging of the CPU on which an exception occurs causing a process to core dump. By default, this feature is turned off, that is, a value of 0. This feature is useful for isolating hardware failures.

Default: 0 (0x0)

Range: 0 or 1

Produces a syslog message of the following format: "Process [<program name>] <pid> core dumped from trap generated on CPU <cpu number>" where <program name> is the name of the process, <pid> is the process id of the process, and <cpu number> is an integer indicating the physical CPU on which the exception was trapped.

The fasthz parameter is used to set the profiling/fast itimer clock speed. fasthz may also be modified by using the systune command.

Default: 1000

Range: 500-2500

Change this parameter to give a finer or coarser grain for such system calls as gettimeofday, getitimer, and settimer.

On Indigo, Indigo2, and Indy systems, an acceptable value is 500, 1000, or 2500 (Hz). If the requested value is not one of these values, the default value of 1000 is used.

On Challenge systems, the resolution of the hardware timer is 21 nanoseconds; therefore, any value smaller than 47 (MHz) is possible.

On Octane systems, the fasthz value must be 1000.

See also the timers(5) man page.

The itimer_on_clkcpu parameter is set to either 0 or 1, to determine whether itimer requests are queued on the clock processor or on the running processor, respectively.

Default: 0

Range: 0 or 1

If a process uses the gettimeofday call to compare the accuracy of the itimer delivery, then you should set this parameter to 1, to take advantage of the clock processor. If the itimer request is for the purpose of implementing a user frequency-based scheduler, then set this parameter to 0 to queue the requests on the current running processor.

The timetrim parameter specifies a signed number of nanoseconds. The system clock is adjusted every second by the signed number of nanoseconds specified by it. This adjustment is limited to 3 milliseconds or 0.3%. The timed and timeslave utilities periodically place suggested values in /var/adm/SYSLOG.

Default: 0

Range: 0-0.3% of a second (3 milliseconds)

Change this parameter as suggested by timed and timeslave. Read the relevant man pages before taking any action.

The portmap_timeout parameter specifies the portmapper query timeout, in 1/10 second increments.

Default: 5 (0x5)

Range: 1 to 200 (20 seconds)

Note that decreasing timeouts can severely degrade system performance.

The sm_timeout parameter specifies the status monitor communication timeout in 1/10 second increments.

Default: 5 (0x5)

Range: 1 to 150 (15 seconds)

Note that decreasing normal and working timeouts can severely degrade system performance.

The GraceWaitTime parameter specifies the NLM grace period, in seconds.

Default: 5 (0x5)

Range: 1 to 60 (1 minute)

Note that decreasing normal and working timeouts can severely degrade system performance.

The first_retry parameter specifies the number of retries to be made for the first contact with the portmapper.

Default: 1 (0x1)

Range: 0 to 10000

The normal_retry parameter specifies the number of subsequent retries after communication has once been established with the portmapper.

Default: 1 (0x1)

Range: 0 to 10000

The lockd_grace_period parameter specifies the NLM grace period, in seconds.

Default: 45 (0x2d)

Range: 1 to 3600 (1 hour)

The lock_share_requests parameter determines whether or not the system will apply IRIX file locks to share requests.

Default: 0 (0x0)

Range: 0 or 1

The lockd_blocking_thresh parameter specifies the number of daemons allowed to block before a new lock daemon will be spawned.

Default: 0 (0x0)

Range: 0 to 1000

The nfs_portmon parameter determines whether or not a client must use a privileged port for NFS requests. Only processes with superuser privilege may bind to a privileged port. The nfs_portmon parameter is binary. If it is set to 0, clients may use any available port. If it is set to 1, clients must use only privileged ports.

Default: 0

Range: 0 or 1

You should change this parameter only if it is absolutely necessary to maintain root privilege on your NFS mounted filesystems and you have checked each NFS client to be sure that it requests a privileged port. If there are any clients requesting non-privileged ports, they will be unable to mount the filesystems.

Additionally, changing the value of nfs_portmon to 1 can give a false sense of security. A process must have root privilege in order to bind to a privileged port, but a single insecure machine compromises the security of this privilege check.

The svc_maxdupreqs parameter sets the number of cached NFS requests.

Default: 409

Range: 400-4096

This parameter should be adjusted to the service load so that there is likely to be a response entry when the first retransmission comes in.

The unpst_sendspace parameter controls the default size of the send buffer on streams sockets.

Default: 0x4000 (16KB)

It is generally recommended that you change the size of socket buffers individually, since changing this parameter changes the send buffer size on all streams sockets, using a tremendous amount of kernel memory. Also, increasing this parameter increases the amount of time necessary to wait for socket response, since all sockets will have more buffer space to read.

See the setsockopt(2) man page for more information on setting specific socket options.

The unpst_recvspace parameter controls the default size of the receive buffer of streams sockets.

Default: 0x4000 (16 KB)

It is generally recommended that you change the size of socket buffers on an individual basis, since changing this parameter changes the receive buffer size on all streams sockets, using a tremendous amount of kernel memory. Also, increasing this parameter will increase the amount of time necessary to wait for socket response, since all sockets will have more buffer space to read.

See the setsockopt(2) man page for more information on setting specific individual socket options.

The unpdg_sendspace parameter controls the size of a data-gram that can be sent over a socket.

Default: 0x2000 (8 KB)

Data-gram sockets operate slightly differently from streams sockets. When a streams socket fills the receive buffer, all data is then sent to the send buffer, and when the send buffer fills up, an error message is issued. Data-gram sockets allow data-grams to fill the receive buffer, and when the receive buffer is full, all future data-grams are discarded, and the error EWOULDBLOCK is generated. Therefore, the unpdg_sendspace parameter serves only to limit the size of a data-gram to not more than can be received by the receive buffer.

Note that the default data-gram socket receive buffers are twice the size of the default data-gram send buffers, thus allowing the process to appear the same as the streams sockets.

It is generally recommended that you not change the size of this parameter without also changing the default receive buffer size for data-gram sockets. If you raise the value of this parameter (unpdg_sendspace) without raising the receive buffer size (unpdg_recvspace), you will allow the sending half of the socket to send more data than can be received by the receiving half. Also it is generally recommended that socket buffer sizes be set individually via the setsockopt(2) system call. See the setsockopt(2) man page for more information on setting specific individual socket options.

The unpdg_recvspace parameter controls the default size of data-gram socket receive buffers.

Default: 0x4000 (16 KB)

It is generally recommended that you change the size of socket buffers individually, since changing this parameter changes the receive buffer size on all data-gram sockets, using a tremendous amount of kernel memory. Also, increasing this parameter increases the amount of time necessary to wait for socket response, since all sockets will have more buffer space to read.

See the setsockopt(2) man page for more information on setting specific individual socket options.

The udp_hashtablesz parameter controls the size of the UDP hash table.

Default: 0 (Auto-configure)

Range: 64-2048

This should not normally need to be changed, but might be changed in order to reduce the consumption of system RAM on systems that are not used to handle large numbers of networking applications, or to increase the number of hash buckets on systems that primarily provide networking services.Using a zero value specifies that the table is to be sized based on system RAM. A non-zero value specifies the number of hash buckets to use for UDP.

The system automatically adjusts this number to support the requirements of the hashing algorithm. For example, entering a value of 512 will create 513 buckets.

The system silently enforces a minimum value of 64 hash buckets and a maximum value of 2048 hash buckets.

The tcp_sendspace parameter controls the default size of the send buffer on TCP/IP sockets. There is a pair of buffers associated with each socket in use. There is a buffer on the receiving side of the socket, and on the sending side. The size of these buffers represent the maximum amount of data that can be queued.

Default: 60 KB

Range: 0-512 KB

The tcp_sendspace parameter may require tuning for slow throughput links, and for compatibility with TCP implementations based on 4.2BSD. For slow throughput links (such as lower speed PPP or SLIP connections), lower the value of this and the tcp_recvspace parameters until the best performance level is found. Very slow links such as PPP or SLIP over modems usually find a 3 KB or 4 KB buffer size to be the most efficient.

The tcp_sendspace parameter may require tuning for slow throughput links, and for compatibility with TCP implementations based on 4.2BSD. For slow throughput links (such as a Wide Area Net with a lower speed PPP or SLIP component), lower the value of this and the tcp_recvspace parameters on systems where the connection goes from fast (Ethernet, FDDI, etc.) to slow (PPP or SLIP) until the best performance level is found. Very slow links such as PPP or SLIP over modems usually find a 3 KB or 4 KB buffer size to be the most efficient.

Conversely, high speed connections may profit from increasing the buffer size incrementally from 60 KB until maximum performance is achieved.

See the setsockopt(2)man page for more information on setting specific socket options.

The tcp_recvspace parameter controls the default size of the TCP/IP receive buffer. There is a pair of buffers associated with each socket in use. There is a buffer on the receiving side of the socket, and on the sending side. The size of these buffers represent the maximum amount of data that can be queued.

Default: 60 KB

Range: 0-512 KB

The tcp_recvspace parameter may require tuning for slow throughput links, and for compatibility with TCP implementations based on 4.2BSD. For slow throughput links (such as a Wide Area Net with a lower speed PPP or SLIP component), lower the value of this and the tcp_sendspace parameters on systems where the connection goes from fast (ethernet, FDDI, and so on) to slow (PPP or SLIP) until the best performance level is found. Very slow links such as PPP or SLIP over modems usually find a 3 KB or 4 KB buffer size to be the most efficient.

Conversely, high speed connections may profit from increasing the buffer size incrementally from 60 KB until maximum performance is achieved.

See the setsockopt(2) man page for more information on setting specific individual socket options.

The tcp_hashtablesz parameter controls the size of the TCP hash table.

Default: 0 (Auto-configure)

Range: 64-8192

The tcp_hashtablesz parameter should not normally need to be changed, but might be changed in order to reduce the consumption of system RAM on systems that are not used to handle large numbers of networking applications, or to increase the number of hash buckets on systems that primarily provide networking services.

Using a zero value specifies that the table is to be sized based on system RAM. A non-zero value specifies the number of hash buckets to use for TCP. The system automatically adjusts this number to support the requirements of the hashing algorithm. For example, entering a value of 512 will create 1025 buckets, since TCP uses the extra space for processing connections in the TIME-WAIT state.

The system silently enforces a minimum value of 64 hash buckets and a maximum value of 8192 hash buckets.

The tcp_sack parameter enables selective acknowledgment for TCP, as per RFC 2018. This capability can provide significant increase in TCP performance over error prone or lossy connections.

Default: 1

Range: 0, 1

Turn this parameter off (set to 0) only if you have old network gear which does not properly process unknown TCP options.

The vino_mtune_dmrbpages parameter limits the size of the DMA descriptor table used by the VINO video capture. This in turn limits how many pages allocated to the Digital Media Ring Buffers (DMRB) are usable when capturing video. The number of pages in vino_mtune_dmrbpages is what the DMRB size specifically for VINO is limited to when allocated.

Default: 0 (auto-configure)

If this parameter is set to zero, then a default based on the amount of memory available at run time is used. If a limit of less than that is desired, set this variable to the maximum number of pages that the ring buffers can use. A value of -1 indicates do not preallocate any space for VINO DMA descriptors.

Note that if a DMA descriptor table size is needed that is greater than the size specified here and cannot be allocated contiguously at run time (possibly due to memory fragmentation), then the video transfer will not be started, and will return an ENOMEM error. The error recovery is to attempt the capture with a smaller DMRB.

The nlpages_256k parameter specifies the number of 256 KB pages to be allocated.

Default: 0

Range: 0-64

The nlpages_1m parameter specifies the number of 1 MB pages to be allocated.

Default: 0

Range: 0-64

The recommended initial value for large (64 processors or greater) systems is 0.

The nlpages_4m parameter specifies the number of 4 MB pages to be allocated.

Default: 0

Range: 0-64

The nlpages_16m parameter specifies the number of 16 MB pages to be allocated.

Default: 0

Range: 0-64

The do_procacct parameter, when set to 1, directs the system to perform BSD system accounting. When set to 0, the system does not perform the accounting, overriding the acct(2) system call.

Default: 1

Range: 0-1

The do_extpacct parameter, when set to 1, directs the system to perform extended process accounting. When set to 0, the system does not perform the accounting.

Default: 0

Range: 0-1

The do_sessacct parameter, when set to 1, directs the system to perform array session accounting when a process exits. When set to 0, the system does not perform the accounting.

Default: 0

Range: 0-1

The min_local_paggid parameter specifies the first array session handle that can be assigned by the kernel.

Default: 1

Range: 1-0x7ffffff00

The max_local_paggid parameter specifies the largest array session handle that can be assigned by the kernel before wrapping back to min_local_paggid.

Default: 65535

Range: 255 -0x7ffffffff

The asmachid parameter specifies the system ID for generating global ASHs. No other system in an array should have the same system ID. If 0 is specified, the kernel will only generate local ASHs.

Default: 0

Range: 0 -0x7fff

The dfltprid parameter specifies the default project ID for special system sessions and other sessions that bypass ordinary project ID assignment.

Default: 0

Range: 0 -0x7fffffffffffffff

The numa_migr_default_mode parameter defines the default migration mode.

Default: 2

Range: 0-4

0: MIGR_DEFMODE_DISABLED Migration is completely disabled. Users cannot use migration.

 1: MIGR_DEFMODE_ENABLED Migration is always enabled. Users cannot disable migration.

 2: MIGR_DEFMODE_NORMOFF Migration is normally off. Users can enable migration for an application.

3: MIGR_DEFMODE_NORMON Migration is normally on. Users can disable migration for an application.

4: MIGR_DEFMODE_LIMITED Migration is normally off for machine configurations with a maximum NUMAlink distance less than numa_migr_min_maxradius (defined below). Migration is normally on otherwise. Users can override this mode.

The numa_migr_default_threshold parameter defines the minimum difference between the local and any remote counter needed to generate a migration request interrupt.

if ( (remote_counter - local_counter) >= ((numa_migr_threshold_reference_value / 100) * numa_migr_default_threshold)){send_migration_request_intr(); }

Default: 50

Range: 0-100

The numa_migr_threshold_reference parameter defines the pegging value for the memory reference counters. It is machine configuration dependent. For Origin2000 systems, it can take the following values:

Default: 0

Range: 0-1

0: MIGR_THRESHREF_STANDARD = Threshold reference is 2048 (11 bit counters). This is the maximum threshold allowed for systems with STANDARD DIMMS.

1: MIGR_THRESHREF_PREMIUM = Threshold reference is 524288 (19-bit counters). This is the maximum threshold allowed for systems with *all* premium SIMMS.

The numa_migr_min_maxradius parameter specifies the minimum maxradius for migration to be turned on. This parameter is used if numa_migr_default_mode has been set to mode 4 (MIGR_DEFMODE_LIMITED). For this mode, migration is normally off for machine configurations with a maximum NUMAlink distance less than numa_migr_min_maxradius. Migration is normally on otherwise.

Default: 2

The numa_migr_min_distance parameter specifies the minimum distance required by the Node Distance Filter in order to accept a migration request.

Default: 1

The numa_zone_radius_max parameter specifies the maximum radius value for use by the zone allocator.

Default: 2

The numa_migr_vehicle parameter defines what device the system should use to migrate a page. The value 0 selects the Block Transfer Engine (BTE) and a value of 1 selects the processor. When the BTE is selected, and the system is equipped with the optional poison bits, the system automatically uses Lazy TLB Shootdown Algorithms.

MIGR_VEHICLE_IS_BTE           0
MIGR_VEHICLE_IS_TLBSYNC       1

Assume migr_vehicle to be BTE to start with. If the system has no BTE, NUMA initialization code will change this to TLBsync. In addition, it will be changed to TLBsync if the processor revision is not suitable to use BTE.

Default: 1

The numa_refcnt_default_mode parameter defines the default extended reference counter mode. It can take the following values:

Default: 0

Range: 0-3

 0: REFCNT_DEFMODE_DISABLED Extended reference counters are disabled; users cannot access the extended reference counters (refcnt(5)).

1: REFCNT_DEFMODE_ENABLED Extended reference counters are always enabled; users cannot disable them.

2: REFCNT_DEFMODE_NORMOFF Extended reference counters are normally disabled; users can disable or enable the counters for an application.

3: REFCNT_DEFMODE_NORMON Extended reference counters are normally enabled; users can disable or enable the counters for an application.

The numa_refcnt_overflow_threshold parameter defines the count at which the hardware reference counters notify the operating system of a counter overflow in order for the count to be transferred into the (software) extended reference counters. It is expressed as a percentage of the threshold reference value defined by numa_migr_threshold_reference.

Default: 50

The numa_migr_memory_low_threshold parameter defines the threshold at which the Memory Pressure Filter starts rejecting migration requests to a node. This threshold is expressed as a percentage of the total amount of physical memory in a node.

Default: 20

The numa_migr_memory_low_enabled parameter enables or disables the Memory Pressure Filter.

Default: 1

The numa_migr_freeze_enabled parameter enables or disables the freezing operation in the Bounce Control Filter.

Default: 0

The numa_migr_freeze_threshold parameter defines the threshold at which a page is frozen. This parameter is expressed as a percent of the maximum count supported by the migration counters (7% for Origin2000 systems).

Default: 20

The numa_migr_melt_enabled parameter enables or disables the melting operation in the Bounce Control Filter.

Default: 0

The numa_migr_melt_threshold parameter unfreezes a page when a migration counter goes below this threshold. This parameter is expressed as a percent of the maximum count supported by the migration counters (7% for Origin2000 systems).

Default: 90

The numa_migr_bounce_control_interval parameter defines the period for the loop that ages the migration counters and the dampening counters. It is expressed in terms of number of mem_ticks (the mem_tick unit is defined by mem_tick_base_period below). If it is set to 0, the system processes 4 pages per mem_tick. In this case, the actual period depends on the amount of physical memory present in a node.

Default: 0

The numa_migr_dampening_enabled parameter enables or disables migration dampening.

Default: 0

The numa_migr_dampening_factor parameter defines the number of migration requests needed for a page before migration is actually executed. It is expressed as a percent of the maximum count supported by the migration-request counters (3% for Origin2000 systems).

Default: 90

The mem_tick_enabled parameter enables or disables the loop that executes the Migration Control Periodic Operation from clock.c/.

Default: 1

The mem_tick_base_period parameter defines the number of 10 ms system ticks in one mem_tick.

Default: 1

The numa_migr_unpegging_control_enabled parameter enables or disables the unpegging periodic operation.

Default: 1

The numa_migr_unpegging_control_interval parameter defines the period for the loop that unpegs the hardware memory reference counters. It is expressed in terms of number of mem_ticks (the mem_tick unit is defined by mem_tick_base_period). If it is set to 0, the system processes 8 pages per mem_tick. In this case, the actual period depends on the amount of physical memory present in a node.

Default: 0

The numa_migr_unpegging_control_threshold parameter defines the hardware memory reference counter value at which the system considers the counter to be pegged. It is expressed as a percent of the maximum count defined by numa_migr_threshold_reference.

Default: 90

The numa_migr_traffic_control_enabled parameter enables or disables the Traffic Control Filter. This is an experimental module, and therefore it should always be disabled.

Default: 0

The numa_migr_traffic_control_interval parameter defines the traffic control period. This is an experimental module.

Default: 400

The numa_migr_traffic_control_threshold parameter defines the traffic control threshold for kicking the batch migration of enqueued migration requests. This is an experimental module.

Default: 80

The numa_kernel_replication_ratio parameter specifies the ratio of memory nodes to kernel text images. There will always be a copy on the master node and also on every other n of the other nodes. 0 means do not replicate. There is always a copy on the master node, and for a ratio n, copies are made on every n of the other nodes. The default value is 1, which causes copies of the kernel text to be made on every node.

Default: 1

The numa_migr_auto_migr_mech parameter defines the migration execution mode for memory coalescing migrations. A value of 0 specifies immediate, a value of 1 specifies delayed. Only the Immediate Mode (0) is currently available.

Default: 0

Range: 0-1

The numa_repl_mem_lowmark parameter defines the migration execution mode for user-requested migrations. A value of 0 specifies immediate, a value of 1 specifies delayed. Only the Immediate Mode (0) is currently available.

Default: 0

Range: 0-1

The numa_migr_auto_migr_mech parameter defines the migration execution mode for memory reference counter-triggered migrations. A value of 0 specifies immediate, a value of 1 specifies delayed. Only the Immediate Mode (0) is currently available.

Default: 0

Range: 0-1

The xvm_max_revive_threads parameter controls how many processes will revive at one time.

Default: 4

Range: 1 to 12

Decrease the number of threads if XLV and XVM mirrored logical volumes are used on the same system. This change will prevent the XVM Volume Manager from stealing too many resources from the XLV Volume Manager.

Or increase the number of threads to have more revives run in parallel.

The xvm_max_revive_rsc parameter controls the number of parallel I/O processes that are used for the revive process.

Default: 5

Range: 1 to 10

Decreasing the number of parallel I/O processes causes less interference with an open file system at the cost of increasing the total time to revive the data.

The coremask parameter is set in the /var/sysgen/mtune/kernel file to determine what kind of permissions a core file will receive.

Default: 0

Range: Minimum 0 - Maximum 0177

Currently, when a process creates a core file, it uses the umask of the owner of the process to determine the permissions on the core file. This can lead to core files containing sensitive information being created with world readable permissions. This systune variable allows the system administrator to limit access to core files. The default is 0, which is the current behavior. However, if the coremask variable is set to some other value, it uses that number as the umask and ignores the umask of the owner of the process. For example, if coremask is set to 0177 and the owner of the process has a umask of 022, the permissions set on the core file will be 600 instead of 644. This variable assumes that the number entered is octal.