Cache expiration is part of default Illumos cache behavior. The idea is that objects in magazines which have not been recently accessed should be returned to the slabs periodically. This is known as cache aging and when enabled objects will be typically returned after 15 seconds.

On the other hand Linux slabs are designed to never move objects back to the slabs unless there is memory pressure. This is possible because under Linux the cache will be notified when memory is low and objects can be released.

By default only the Linux method is enabled. It has been shown to improve responsiveness on low memory systems and not negatively impact the performance of systems with more memory. This policy may be changed by setting the spl_kmem_cache_expire bit mask as follows, both policies may be enabled concurrently.

0x01 - Aging (Illumos), 0x02 - Low memory (Linux)

Default value: 0x02

The number of threads created for the spl_kmem_cache task queue. This task queue is responsible for allocating new slabs for use by the kmem caches. For the majority of systems and workloads only a small number of threads are required.

Default value: 4

When this is set it prevents Linux from being able to rapidly reclaim all the memory held by the kmem caches. This may be useful in circumstances where it's preferable that Linux reclaim memory from some other subsystem first. Setting this will increase the likelihood out of memory events on a memory constrained system.

Default value: 0

The preferred number of objects per slab in the cache. In general, a larger value will increase the caches memory footprint while decreasing the time required to perform an allocation. Conversely, a smaller value will minimize the footprint and improve cache reclaim time but individual allocations may take longer.

Default value: 8

The minimum number of objects allowed per slab. Normally slabs will contain spl_kmem_cache_obj_per_slab objects but for caches that contain very large objects it's desirable to only have a few, or even just one, object per slab.

Default value: 1

The maximum size of a kmem cache slab in MiB. This effectively limits the maximum cache object size to spl_kmem_cache_max_size / spl_kmem_cache_obj_per_slab. Caches may not be created with object sized larger than this limit.

Default value: 32 (64-bit) or 4 (32-bit)

For small objects the Linux slab allocator should be used to make the most efficient use of the memory. However, large objects are not supported by the Linux slab and therefore the SPL implementation is preferred. This value is used to determine the cutoff between a small and large object.

Objects of spl_kmem_cache_slab_limit or smaller will be allocated using the Linux slab allocator, large objects use the SPL allocator. A cutoff of 16K was determined to be optimal for architectures using 4K pages.

Default value: 16,384

Depending on the size of a cache object it may be backed by kmalloc()'d or vmalloc()'d memory. This is because the size of the required allocation greatly impacts the best way to allocate the memory.

When objects are small and only a small number of memory pages need to be allocated, ideally just one, then kmalloc() is very efficient. However, when allocating multiple pages with kmalloc() it gets increasingly expensive because the pages must be physically contiguous.

For this reason we shift to vmalloc() for slabs of large objects which which removes the need for contiguous pages. We cannot use vmalloc() in all cases because there is significant locking overhead involved. This function takes a single global lock over the entire virtual address range which serializes all allocations. Using slightly different allocation functions for small and large objects allows us to handle a wide range of object sizes.

The spl_kmem_cache_kmem_limit value is used to determine this cutoff size. One quarter the PAGE_SIZE is used as the default value because spl_kmem_cache_obj_per_slab defaults to 16. This means that at most we will need to allocate four contiguous pages.

Default value: PAGE_SIZE/4

As a general rule kmem_alloc() allocations should be small, preferably just a few pages since they must by physically contiguous. Therefore, a rate limited warning will be printed to the console for any kmem_alloc() which exceeds a reasonable threshold.

The default warning threshold is set to eight pages but capped at 32K to accommodate systems using large pages. This value was selected to be small enough to ensure the largest allocations are quickly noticed and fixed. But large enough to avoid logging any warnings when a allocation size is larger than optimal but not a serious concern. Since this value is tunable, developers are encouraged to set it lower when testing so any new largish allocations are quickly caught. These warnings may be disabled by setting the threshold to zero.

Default value: 32,768

Large kmem_alloc() allocations will fail if they exceed KMALLOC_MAX_SIZE. Allocations which are marginally smaller than this limit may succeed but should still be avoided due to the expense of locating a contiguous range of free pages. Therefore, a maximum kmem size with reasonable safely margin of 4x is set. Kmem_alloc() allocations larger than this maximum will quickly fail. Vmem_alloc() allocations less than or equal to this value will use kmalloc(), but shift to vmalloc() when exceeding this value.

Default value: KMALLOC_MAX_SIZE/4

Cache magazines are an optimization designed to minimize the cost of allocating memory. They do this by keeping a per-cpu cache of recently freed objects, which can then be reallocated without taking a lock. This can improve performance on highly contended caches. However, because objects in magazines will prevent otherwise empty slabs from being immediately released this may not be ideal for low memory machines.

For this reason spl_kmem_cache_magazine_size can be used to set a maximum magazine size. When this value is set to 0 the magazine size will be automatically determined based on the object size. Otherwise magazines will be limited to 2-256 objects per magazine (i.e per cpu). Magazines may never be entirely disabled in this implementation.

Default value: 0

The system hostid, when set this can be used to uniquely identify a system. By default this value is set to zero which indicates the hostid is disabled. It can be explicitly enabled by placing a unique non-zero value in /etc/hostid/.

Default value: 0

The expected path to locate the system hostid when specified. This value may be overridden for non-standard configurations.

Default value: /etc/hostid

Cause a kernel panic on assertion failures. When not enabled, the thread is halted to facilitate further debugging.

Set to a non-zero value to enable.

Default value: 0

Kick stuck taskq to spawn threads. When writing a non-zero value to it, it will scan all the taskqs. If any of them have a pending task more than 5 seconds old, it will kick it to spawn more threads. This can be used if you find a rare deadlock occurs because one or more taskqs didn't spawn a thread when it should.

Default value: 0

Bind taskq threads to specific CPUs. When enabled all taskq threads will be distributed evenly over the available CPUs. By default, this behavior is disabled to allow the Linux scheduler the maximum flexibility to determine where a thread should run.

Default value: 0

Allow dynamic taskqs. When enabled taskqs which set the TASKQ_DYNAMIC flag will by default create only a single thread. New threads will be created on demand up to a maximum allowed number to facilitate the completion of outstanding tasks. Threads which are no longer needed will be promptly destroyed. By default this behavior is enabled but it can be disabled to aid performance analysis or troubleshooting.

Default value: 1

Allow newly created taskq threads to set a non-default scheduler priority. When enabled the priority specified when a taskq is created will be applied to all threads created by that taskq. When disabled all threads will use the default Linux kernel thread priority. By default, this behavior is enabled.

Default value: 1

The number of items a taskq worker thread must handle without interruption before requesting a new worker thread be spawned. This is used to control how quickly taskqs ramp up the number of threads processing the queue. Because Linux thread creation and destruction are relatively inexpensive a small default value has been selected. This means that normally threads will be created aggressively which is desirable. Increasing this value will result in a slower thread creation rate which may be preferable for some configurations.

Default value: 4

The maximum number of tasks per pending list in each taskq shown in /proc/spl/{taskq,taskq-all}. Write 0 to turn off the limit. The proc file will walk the lists with lock held, reading it could cause a lock up if the list grow too large without limiting the output. "(truncated)" will be shown if the list is larger than the limit.

Default value: 512