diff options
Diffstat (limited to 'Documentation/sysctl')
| -rw-r--r-- | Documentation/sysctl/kernel.txt | 230 | ||||
| -rw-r--r-- | Documentation/sysctl/net.txt | 63 | ||||
| -rw-r--r-- | Documentation/sysctl/vm.txt | 173 |
3 files changed, 412 insertions, 54 deletions
diff --git a/Documentation/sysctl/kernel.txt b/Documentation/sysctl/kernel.txt index ccd42589e12..c14374e7177 100644 --- a/Documentation/sysctl/kernel.txt +++ b/Documentation/sysctl/kernel.txt @@ -33,6 +33,11 @@ show up in /proc/sys/kernel: - domainname - hostname - hotplug +- hung_task_panic +- hung_task_check_count +- hung_task_timeout_secs +- hung_task_warnings +- kexec_load_disabled - kptr_restrict - kstack_depth_to_print [ X86 only ] - l2cr [ PPC only ] @@ -70,12 +75,14 @@ show up in /proc/sys/kernel: - shmall - shmmax [ sysv ipc ] - shmmni -- softlockup_thresh +- softlockup_all_cpu_backtrace - stop-a [ SPARC only ] - sysrq ==> Documentation/sysrq.txt +- sysctl_writes_strict - tainted - threads-max - unknown_nmi_panic +- watchdog_thresh - version ============================================================== @@ -182,6 +189,7 @@ core_pattern is used to specify a core dumpfile pattern name. %<NUL> '%' is dropped %% output one '%' %p pid + %P global pid (init PID namespace) %u uid %g gid %d dump mode, matches PR_SET_DUMPABLE and @@ -286,16 +294,79 @@ Default value is "/sbin/hotplug". ============================================================== +hung_task_panic: + +Controls the kernel's behavior when a hung task is detected. +This file shows up if CONFIG_DETECT_HUNG_TASK is enabled. + +0: continue operation. This is the default behavior. + +1: panic immediately. + +============================================================== + +hung_task_check_count: + +The upper bound on the number of tasks that are checked. +This file shows up if CONFIG_DETECT_HUNG_TASK is enabled. + +============================================================== + +hung_task_timeout_secs: + +Check interval. When a task in D state did not get scheduled +for more than this value report a warning. +This file shows up if CONFIG_DETECT_HUNG_TASK is enabled. + +0: means infinite timeout - no checking done. +Possible values to set are in range {0..LONG_MAX/HZ}. + +============================================================== + +hung_task_warnings: + +The maximum number of warnings to report. During a check interval +if a hung task is detected, this value is decreased by 1. +When this value reaches 0, no more warnings will be reported. +This file shows up if CONFIG_DETECT_HUNG_TASK is enabled. + +-1: report an infinite number of warnings. + +============================================================== + +kexec_load_disabled: + +A toggle indicating if the kexec_load syscall has been disabled. This +value defaults to 0 (false: kexec_load enabled), but can be set to 1 +(true: kexec_load disabled). Once true, kexec can no longer be used, and +the toggle cannot be set back to false. This allows a kexec image to be +loaded before disabling the syscall, allowing a system to set up (and +later use) an image without it being altered. Generally used together +with the "modules_disabled" sysctl. + +============================================================== + kptr_restrict: This toggle indicates whether restrictions are placed on -exposing kernel addresses via /proc and other interfaces. When -kptr_restrict is set to (0), there are no restrictions. When -kptr_restrict is set to (1), the default, kernel pointers -printed using the %pK format specifier will be replaced with 0's -unless the user has CAP_SYSLOG. When kptr_restrict is set to -(2), kernel pointers printed using %pK will be replaced with 0's -regardless of privileges. +exposing kernel addresses via /proc and other interfaces. + +When kptr_restrict is set to (0), the default, there are no restrictions. + +When kptr_restrict is set to (1), kernel pointers printed using the %pK +format specifier will be replaced with 0's unless the user has CAP_SYSLOG +and effective user and group ids are equal to the real ids. This is +because %pK checks are done at read() time rather than open() time, so +if permissions are elevated between the open() and the read() (e.g via +a setuid binary) then %pK will not leak kernel pointers to unprivileged +users. Note, this is a temporary solution only. The correct long-term +solution is to do the permission checks at open() time. Consider removing +world read permissions from files that use %pK, and using dmesg_restrict +to protect against uses of %pK in dmesg(8) if leaking kernel pointer +values to unprivileged users is a concern. + +When kptr_restrict is set to (2), kernel pointers printed using +%pK will be replaced with 0's regardless of privileges. ============================================================== @@ -319,7 +390,7 @@ A toggle value indicating if modules are allowed to be loaded in an otherwise modular kernel. This toggle defaults to off (0), but can be set true (1). Once true, modules can be neither loaded nor unloaded, and the toggle cannot be set back -to false. +to false. Generally used with the "kexec_load_disabled" toggle. ============================================================== @@ -354,6 +425,69 @@ utilize. ============================================================== +numa_balancing + +Enables/disables automatic page fault based NUMA memory +balancing. Memory is moved automatically to nodes +that access it often. + +Enables/disables automatic NUMA memory balancing. On NUMA machines, there +is a performance penalty if remote memory is accessed by a CPU. When this +feature is enabled the kernel samples what task thread is accessing memory +by periodically unmapping pages and later trapping a page fault. At the +time of the page fault, it is determined if the data being accessed should +be migrated to a local memory node. + +The unmapping of pages and trapping faults incur additional overhead that +ideally is offset by improved memory locality but there is no universal +guarantee. If the target workload is already bound to NUMA nodes then this +feature should be disabled. Otherwise, if the system overhead from the +feature is too high then the rate the kernel samples for NUMA hinting +faults may be controlled by the numa_balancing_scan_period_min_ms, +numa_balancing_scan_delay_ms, numa_balancing_scan_period_max_ms, +numa_balancing_scan_size_mb, and numa_balancing_settle_count sysctls. + +============================================================== + +numa_balancing_scan_period_min_ms, numa_balancing_scan_delay_ms, +numa_balancing_scan_period_max_ms, numa_balancing_scan_size_mb + +Automatic NUMA balancing scans tasks address space and unmaps pages to +detect if pages are properly placed or if the data should be migrated to a +memory node local to where the task is running. Every "scan delay" the task +scans the next "scan size" number of pages in its address space. When the +end of the address space is reached the scanner restarts from the beginning. + +In combination, the "scan delay" and "scan size" determine the scan rate. +When "scan delay" decreases, the scan rate increases. The scan delay and +hence the scan rate of every task is adaptive and depends on historical +behaviour. If pages are properly placed then the scan delay increases, +otherwise the scan delay decreases. The "scan size" is not adaptive but +the higher the "scan size", the higher the scan rate. + +Higher scan rates incur higher system overhead as page faults must be +trapped and potentially data must be migrated. However, the higher the scan +rate, the more quickly a tasks memory is migrated to a local node if the +workload pattern changes and minimises performance impact due to remote +memory accesses. These sysctls control the thresholds for scan delays and +the number of pages scanned. + +numa_balancing_scan_period_min_ms is the minimum time in milliseconds to +scan a tasks virtual memory. It effectively controls the maximum scanning +rate for each task. + +numa_balancing_scan_delay_ms is the starting "scan delay" used for a task +when it initially forks. + +numa_balancing_scan_period_max_ms is the maximum time in milliseconds to +scan a tasks virtual memory. It effectively controls the minimum scanning +rate for each task. + +numa_balancing_scan_size_mb is how many megabytes worth of pages are +scanned for a given scan. + +============================================================== + osrelease, ostype & version: # cat osrelease @@ -427,6 +561,32 @@ This file shows up if CONFIG_DEBUG_STACKOVERFLOW is enabled. ============================================================== +perf_cpu_time_max_percent: + +Hints to the kernel how much CPU time it should be allowed to +use to handle perf sampling events. If the perf subsystem +is informed that its samples are exceeding this limit, it +will drop its sampling frequency to attempt to reduce its CPU +usage. + +Some perf sampling happens in NMIs. If these samples +unexpectedly take too long to execute, the NMIs can become +stacked up next to each other so much that nothing else is +allowed to execute. + +0: disable the mechanism. Do not monitor or correct perf's + sampling rate no matter how CPU time it takes. + +1-100: attempt to throttle perf's sample rate to this + percentage of CPU. Note: the kernel calculates an + "expected" length of each sample event. 100 here means + 100% of that expected length. Even if this is set to + 100, you may still see sample throttling if this + length is exceeded. Set to 0 if you truly do not care + how much CPU is consumed. + +============================================================== + pid_max: @@ -604,12 +764,39 @@ without users and with a dead originative process will be destroyed. ============================================================== -softlockup_thresh: +sysctl_writes_strict: + +Control how file position affects the behavior of updating sysctl values +via the /proc/sys interface: + + -1 - Legacy per-write sysctl value handling, with no printk warnings. + Each write syscall must fully contain the sysctl value to be + written, and multiple writes on the same sysctl file descriptor + will rewrite the sysctl value, regardless of file position. + 0 - (default) Same behavior as above, but warn about processes that + perform writes to a sysctl file descriptor when the file position + is not 0. + 1 - Respect file position when writing sysctl strings. Multiple writes + will append to the sysctl value buffer. Anything past the max length + of the sysctl value buffer will be ignored. Writes to numeric sysctl + entries must always be at file position 0 and the value must be + fully contained in the buffer sent in the write syscall. + +============================================================== + +softlockup_all_cpu_backtrace: + +This value controls the soft lockup detector thread's behavior +when a soft lockup condition is detected as to whether or not +to gather further debug information. If enabled, each cpu will +be issued an NMI and instructed to capture stack trace. + +This feature is only applicable for architectures which support +NMI. -This value can be used to lower the softlockup tolerance threshold. The -default threshold is 60 seconds. If a cpu is locked up for 60 seconds, -the kernel complains. Valid values are 1-60 seconds. Setting this -tunable to zero will disable the softlockup detection altogether. +0: do nothing. This is the default behavior. + +1: on detection capture more debug information. ============================================================== @@ -637,6 +824,8 @@ can be ORed together: 1024 - A module from drivers/staging was loaded. 2048 - The system is working around a severe firmware bug. 4096 - An out-of-tree module has been loaded. +8192 - An unsigned module has been loaded in a kernel supporting module + signature. ============================================================== @@ -648,3 +837,16 @@ that time, kernel debugging information is displayed on console. NMI switch that most IA32 servers have fires unknown NMI up, for example. If a system hangs up, try pressing the NMI switch. + +============================================================== + +watchdog_thresh: + +This value can be used to control the frequency of hrtimer and NMI +events and the soft and hard lockup thresholds. The default threshold +is 10 seconds. + +The softlockup threshold is (2 * watchdog_thresh). Setting this +tunable to zero will disable lockup detection altogether. + +============================================================== diff --git a/Documentation/sysctl/net.txt b/Documentation/sysctl/net.txt index 98335b7a533..9a0319a8247 100644 --- a/Documentation/sysctl/net.txt +++ b/Documentation/sysctl/net.txt @@ -1,4 +1,4 @@ -Documentation for /proc/sys/net/* kernel version 2.4.0-test11-pre4 +Documentation for /proc/sys/net/* (c) 1999 Terrehon Bowden <terrehon@pacbell.net> Bodo Bauer <bb@ricochet.net> (c) 2000 Jorge Nerin <comandante@zaralinux.com> @@ -9,10 +9,10 @@ For general info and legal blurb, please look in README. ============================================================== This file contains the documentation for the sysctl files in -/proc/sys/net and is valid for Linux kernel version 2.4.0-test11-pre4. +/proc/sys/net The interface to the networking parts of the kernel is located in -/proc/sys/net. The following table shows all possible subdirectories.You may +/proc/sys/net. The following table shows all possible subdirectories. You may see only some of them, depending on your kernel's configuration. @@ -26,7 +26,7 @@ Table : Subdirectories in /proc/sys/net ipv4 IP version 4 x25 X.25 protocol ipx IPX token-ring IBM token ring bridge Bridging decnet DEC net - ipv6 IP version 6 + ipv6 IP version 6 tipc TIPC .............................................................................. 1. /proc/sys/net/core - Network core options @@ -50,6 +50,43 @@ The maximum number of packets that kernel can handle on a NAPI interrupt, it's a Per-CPU variable. Default: 64 +default_qdisc +-------------- + +The default queuing discipline to use for network devices. This allows +overriding the default queue discipline of pfifo_fast with an +alternative. Since the default queuing discipline is created with the +no additional parameters so is best suited to queuing disciplines that +work well without configuration like stochastic fair queue (sfq), +CoDel (codel) or fair queue CoDel (fq_codel). Don't use queuing disciplines +like Hierarchical Token Bucket or Deficit Round Robin which require setting +up classes and bandwidths. +Default: pfifo_fast + +busy_read +---------------- +Low latency busy poll timeout for socket reads. (needs CONFIG_NET_RX_BUSY_POLL) +Approximate time in us to busy loop waiting for packets on the device queue. +This sets the default value of the SO_BUSY_POLL socket option. +Can be set or overridden per socket by setting socket option SO_BUSY_POLL, +which is the preferred method of enabling. If you need to enable the feature +globally via sysctl, a value of 50 is recommended. +Will increase power usage. +Default: 0 (off) + +busy_poll +---------------- +Low latency busy poll timeout for poll and select. (needs CONFIG_NET_RX_BUSY_POLL) +Approximate time in us to busy loop waiting for events. +Recommended value depends on the number of sockets you poll on. +For several sockets 50, for several hundreds 100. +For more than that you probably want to use epoll. +Note that only sockets with SO_BUSY_POLL set will be busy polled, +so you want to either selectively set SO_BUSY_POLL on those sockets or set +sysctl.net.busy_read globally. +Will increase power usage. +Default: 0 (off) + rmem_default ------------ @@ -93,8 +130,7 @@ netdev_budget Maximum number of packets taken from all interfaces in one polling cycle (NAPI poll). In one polling cycle interfaces which are registered to polling are -probed in a round-robin manner. The limit of packets in one such probe can be -set per-device via sysfs class/net/<device>/weight . +probed in a round-robin manner. netdev_max_backlog ------------------ @@ -201,3 +237,18 @@ IPX. The /proc/net/ipx_route table holds a list of IPX routes. For each route it gives the destination network, the router node (or Directly) and the network address of the router (or Connected) for internal networks. + +6. TIPC +------------------------------------------------------- + +The TIPC protocol now has a tunable for the receive memory, similar to the +tcp_rmem - i.e. a vector of 3 INTEGERs: (min, default, max) + + # cat /proc/sys/net/tipc/tipc_rmem + 4252725 34021800 68043600 + # + +The max value is set to CONN_OVERLOAD_LIMIT, and the default and min values +are scaled (shifted) versions of that same value. Note that the min value +is not at this point in time used in any meaningful way, but the triplet is +preserved in order to be consistent with things like tcp_rmem. diff --git a/Documentation/sysctl/vm.txt b/Documentation/sysctl/vm.txt index 078701fdbd4..4415aa91568 100644 --- a/Documentation/sysctl/vm.txt +++ b/Documentation/sysctl/vm.txt @@ -18,6 +18,7 @@ files can be found in mm/swap.c. Currently, these files are in /proc/sys/vm: +- admin_reserve_kbytes - block_dump - compact_memory - dirty_background_bytes @@ -46,6 +47,7 @@ Currently, these files are in /proc/sys/vm: - numa_zonelist_order - oom_dump_tasks - oom_kill_allocating_task +- overcommit_kbytes - overcommit_memory - overcommit_ratio - page-cluster @@ -53,11 +55,41 @@ Currently, these files are in /proc/sys/vm: - percpu_pagelist_fraction - stat_interval - swappiness +- user_reserve_kbytes - vfs_cache_pressure - zone_reclaim_mode ============================================================== +admin_reserve_kbytes + +The amount of free memory in the system that should be reserved for users +with the capability cap_sys_admin. + +admin_reserve_kbytes defaults to min(3% of free pages, 8MB) + +That should provide enough for the admin to log in and kill a process, +if necessary, under the default overcommit 'guess' mode. + +Systems running under overcommit 'never' should increase this to account +for the full Virtual Memory Size of programs used to recover. Otherwise, +root may not be able to log in to recover the system. + +How do you calculate a minimum useful reserve? + +sshd or login + bash (or some other shell) + top (or ps, kill, etc.) + +For overcommit 'guess', we can sum resident set sizes (RSS). +On x86_64 this is about 8MB. + +For overcommit 'never', we can take the max of their virtual sizes (VSZ) +and add the sum of their RSS. +On x86_64 this is about 128MB. + +Changing this takes effect whenever an application requests memory. + +============================================================== + block_dump block_dump enables block I/O debugging when set to a nonzero value. More @@ -88,8 +120,11 @@ other appears as 0 when read. dirty_background_ratio -Contains, as a percentage of total system memory, the number of pages at which -the background kernel flusher threads will start writing out dirty data. +Contains, as a percentage of total available memory that contains free pages +and reclaimable pages, the number of pages at which the background kernel +flusher threads will start writing out dirty data. + +The total avaiable memory is not equal to total system memory. ============================================================== @@ -120,9 +155,11 @@ interval will be written out next time a flusher thread wakes up. dirty_ratio -Contains, as a percentage of total system memory, the number of pages at which -a process which is generating disk writes will itself start writing out dirty -data. +Contains, as a percentage of total available memory that contains free pages +and reclaimable pages, the number of pages at which a process which is +generating disk writes will itself start writing out dirty data. + +The total avaiable memory is not equal to total system memory. ============================================================== @@ -138,18 +175,39 @@ Setting this to zero disables periodic writeback altogether. drop_caches -Writing to this will cause the kernel to drop clean caches, dentries and -inodes from memory, causing that memory to become free. +Writing to this will cause the kernel to drop clean caches, as well as +reclaimable slab objects like dentries and inodes. Once dropped, their +memory becomes free. To free pagecache: echo 1 > /proc/sys/vm/drop_caches -To free dentries and inodes: +To free reclaimable slab objects (includes dentries and inodes): echo 2 > /proc/sys/vm/drop_caches -To free pagecache, dentries and inodes: +To free slab objects and pagecache: echo 3 > /proc/sys/vm/drop_caches -As this is a non-destructive operation and dirty objects are not freeable, the -user should run `sync' first. +This is a non-destructive operation and will not free any dirty objects. +To increase the number of objects freed by this operation, the user may run +`sync' prior to writing to /proc/sys/vm/drop_caches. This will minimize the +number of dirty objects on the system and create more candidates to be +dropped. + +This file is not a means to control the growth of the various kernel caches +(inodes, dentries, pagecache, etc...) These objects are automatically +reclaimed by the kernel when memory is needed elsewhere on the system. + +Use of this file can cause performance problems. Since it discards cached +objects, it may cost a significant amount of I/O and CPU to recreate the +dropped objects, especially if they were under heavy use. Because of this, +use outside of a testing or debugging environment is not recommended. + +You may see informational messages in your kernel log when this file is +used: + + cat (1234): drop_caches: 3 + +These are informational only. They do not mean that anything is wrong +with your system. To disable them, echo 4 (bit 3) into drop_caches. ============================================================== @@ -169,17 +227,25 @@ fragmentation index is <= extfrag_threshold. The default value is 500. hugepages_treat_as_movable -This parameter is only useful when kernelcore= is specified at boot time to -create ZONE_MOVABLE for pages that may be reclaimed or migrated. Huge pages -are not movable so are not normally allocated from ZONE_MOVABLE. A non-zero -value written to hugepages_treat_as_movable allows huge pages to be allocated -from ZONE_MOVABLE. +This parameter controls whether we can allocate hugepages from ZONE_MOVABLE +or not. If set to non-zero, hugepages can be allocated from ZONE_MOVABLE. +ZONE_MOVABLE is created when kernel boot parameter kernelcore= is specified, +so this parameter has no effect if used without kernelcore=. + +Hugepage migration is now available in some situations which depend on the +architecture and/or the hugepage size. If a hugepage supports migration, +allocation from ZONE_MOVABLE is always enabled for the hugepage regardless +of the value of this parameter. +IOW, this parameter affects only non-migratable hugepages. -Once enabled, the ZONE_MOVABLE is treated as an area of memory the huge -pages pool can easily grow or shrink within. Assuming that applications are -not running that mlock() a lot of memory, it is likely the huge pages pool -can grow to the size of ZONE_MOVABLE by repeatedly entering the desired value -into nr_hugepages and triggering page reclaim. +Assuming that hugepages are not migratable in your system, one usecase of +this parameter is that users can make hugepage pool more extensible by +enabling the allocation from ZONE_MOVABLE. This is because on ZONE_MOVABLE +page reclaim/migration/compaction work more and you can get contiguous +memory more likely. Note that using ZONE_MOVABLE for non-migratable +hugepages can do harm to other features like memory hotremove (because +memory hotremove expects that memory blocks on ZONE_MOVABLE are always +removable,) so it's a trade-off responsible for the users. ============================================================== @@ -479,7 +545,7 @@ Specify "[Dd]efault" to request automatic configuration. Autoconfiguration will select "node" order in following case. (1) if the DMA zone does not exist or (2) if the DMA zone comprises greater than 50% of the available memory or -(3) if any node's DMA zone comprises greater than 60% of its local memory and +(3) if any node's DMA zone comprises greater than 70% of its local memory and the amount of local memory is big enough. Otherwise, "zone" order will be selected. Default order is recommended unless @@ -530,6 +596,17 @@ The default value is 0. ============================================================== +overcommit_kbytes: + +When overcommit_memory is set to 2, the committed address space is not +permitted to exceed swap plus this amount of physical RAM. See below. + +Note: overcommit_kbytes is the counterpart of overcommit_ratio. Only one +of them may be specified at a time. Setting one disables the other (which +then appears as 0 when read). + +============================================================== + overcommit_memory: This value contains a flag that enables memory overcommitment. @@ -542,6 +619,7 @@ memory until it actually runs out. When this flag is 2, the kernel uses a "never overcommit" policy that attempts to prevent any overcommit of memory. +Note that user_reserve_kbytes affects this policy. This feature can be very useful because there are a lot of programs that malloc() huge amounts of memory "just-in-case" @@ -624,7 +702,8 @@ The batch value of each per cpu pagelist is also updated as a result. It is set to pcp->high/4. The upper limit of batch is (PAGE_SHIFT * 8) The initial value is zero. Kernel does not use this value at boot time to set -the high water marks for each per cpu page list. +the high water marks for each per cpu page list. If the user writes '0' to this +sysctl, it will revert to this default behavior. ============================================================== @@ -639,17 +718,37 @@ swappiness This control is used to define how aggressive the kernel will swap memory pages. Higher values will increase agressiveness, lower values -decrease the amount of swap. +decrease the amount of swap. A value of 0 instructs the kernel not to +initiate swap until the amount of free and file-backed pages is less +than the high water mark in a zone. The default value is 60. ============================================================== +- user_reserve_kbytes + +When overcommit_memory is set to 2, "never overommit" mode, reserve +min(3% of current process size, user_reserve_kbytes) of free memory. +This is intended to prevent a user from starting a single memory hogging +process, such that they cannot recover (kill the hog). + +user_reserve_kbytes defaults to min(3% of the current process size, 128MB). + +If this is reduced to zero, then the user will be allowed to allocate +all free memory with a single process, minus admin_reserve_kbytes. +Any subsequent attempts to execute a command will result in +"fork: Cannot allocate memory". + +Changing this takes effect whenever an application requests memory. + +============================================================== + vfs_cache_pressure ------------------ -Controls the tendency of the kernel to reclaim the memory which is used for -caching of directory and inode objects. +This percentage value controls the tendency of the kernel to reclaim +the memory which is used for caching of directory and inode objects. At the default value of vfs_cache_pressure=100 the kernel will attempt to reclaim dentries and inodes at a "fair" rate with respect to pagecache and @@ -659,6 +758,11 @@ never reclaim dentries and inodes due to memory pressure and this can easily lead to out-of-memory conditions. Increasing vfs_cache_pressure beyond 100 causes the kernel to prefer to reclaim dentries and inodes. +Increasing vfs_cache_pressure significantly beyond 100 may have negative +performance impact. Reclaim code needs to take various locks to find freeable +directory and inode objects. With vfs_cache_pressure=1000, it will look for +ten times more freeable objects than there are. + ============================================================== zone_reclaim_mode: @@ -674,16 +778,17 @@ This is value ORed together of 2 = Zone reclaim writes dirty pages out 4 = Zone reclaim swaps pages -zone_reclaim_mode is set during bootup to 1 if it is determined that pages -from remote zones will cause a measurable performance reduction. The -page allocator will then reclaim easily reusable pages (those page -cache pages that are currently not used) before allocating off node pages. - -It may be beneficial to switch off zone reclaim if the system is -used for a file server and all of memory should be used for caching files -from disk. In that case the caching effect is more important than +zone_reclaim_mode is disabled by default. For file servers or workloads +that benefit from having their data cached, zone_reclaim_mode should be +left disabled as the caching effect is likely to be more important than data locality. +zone_reclaim may be enabled if it's known that the workload is partitioned +such that each partition fits within a NUMA node and that accessing remote +memory would cause a measurable performance reduction. The page allocator +will then reclaim easily reusable pages (those page cache pages that are +currently not used) before allocating off node pages. + Allowing zone reclaim to write out pages stops processes that are writing large amounts of data from dirtying pages on other nodes. Zone reclaim will write out dirty pages if a zone fills up and so effectively |