aboutsummaryrefslogtreecommitdiff
path: root/include/linux/slub_def.h
AgeCommit message (Collapse)Author
2009-03-24Merge branches 'topic/slob/cleanups', 'topic/slob/fixes', 'topic/slub/core', ↵Pekka Enberg
'topic/slub/cleanups' and 'topic/slub/perf' into for-linus
2009-02-23slub: move min_partial to struct kmem_cacheDavid Rientjes
Although it allows for better cacheline use, it is unnecessary to save a copy of the cache's min_partial value in each kmem_cache_node. Cc: Christoph Lameter <cl@linux-foundation.org> Signed-off-by: David Rientjes <rientjes@google.com> Signed-off-by: Pekka Enberg <penberg@cs.helsinki.fi>
2009-02-20SLUB: Do not pass 8k objects through to the page allocatorPekka Enberg
Increase the maximum object size in SLUB so that 8k objects are not passed through to the page allocator anymore. The network stack uses 8k objects for performance critical operations. The patch is motivated by a SLAB vs. SLUB regression in the netperf benchmark. The problem is that the kfree(skb->head) call in skb_release_data() that is subject to page allocator pass-through as the size passed to __alloc_skb() is larger than 4 KB in this test. As explained by Yanmin Zhang: I use 2.6.29-rc2 kernel to run netperf UDP-U-4k CPU_NUM client/server pair loopback testing on x86-64 machines. Comparing with SLUB, SLAB's result is about 2.3 times of SLUB's. After applying the reverting patch, the result difference between SLUB and SLAB becomes 1% which we might consider as fluctuation. [ penberg@cs.helsinki.fi: fix oops in kmalloc() ] Reported-by: "Zhang, Yanmin" <yanmin_zhang@linux.intel.com> Tested-by: "Zhang, Yanmin" <yanmin_zhang@linux.intel.com> Signed-off-by: Christoph Lameter <cl@linux-foundation.org> Signed-off-by: Pekka Enberg <penberg@cs.helsinki.fi>
2009-02-20SLUB: Introduce and use SLUB_MAX_SIZE and SLUB_PAGE_SHIFT constantsChristoph Lameter
As a preparational patch to bump up page allocator pass-through threshold, introduce two new constants SLUB_MAX_SIZE and SLUB_PAGE_SHIFT and convert mm/slub.c to use them. Reported-by: "Zhang, Yanmin" <yanmin_zhang@linux.intel.com> Tested-by: "Zhang, Yanmin" <yanmin_zhang@linux.intel.com> Signed-off-by: Christoph Lameter <cl@linux-foundation.org> Signed-off-by: Pekka Enberg <penberg@cs.helsinki.fi>
2008-08-05SLUB: dynamic per-cache MIN_PARTIALPekka Enberg
This patch changes the static MIN_PARTIAL to a dynamic per-cache ->min_partial value that is calculated from object size. The bigger the object size, the more pages we keep on the partial list. I tested SLAB, SLUB, and SLUB with this patch on Jens Axboe's 'netio' example script of the fio benchmarking tool. The script stresses the networking subsystem which should also give a fairly good beating of kmalloc() et al. To run the test yourself, first clone the fio repository: git clone git://git.kernel.dk/fio.git and then run the following command n times on your machine: time ./fio examples/netio The results on my 2-way 64-bit x86 machine are as follows: [ the minimum, maximum, and average are captured from 50 individual runs ] real time (seconds) min max avg sd SLAB 22.76 23.38 22.98 0.17 SLUB 22.80 25.78 23.46 0.72 SLUB (dynamic) 22.74 23.54 23.00 0.20 sys time (seconds) min max avg sd SLAB 6.90 8.28 7.70 0.28 SLUB 7.42 16.95 8.89 2.28 SLUB (dynamic) 7.17 8.64 7.73 0.29 user time (seconds) min max avg sd SLAB 36.89 38.11 37.50 0.29 SLUB 30.85 37.99 37.06 1.67 SLUB (dynamic) 36.75 38.07 37.59 0.32 As you can see from the above numbers, this patch brings SLUB to the same level as SLAB for this particular workload fixing a ~2% regression. I'd expect this change to help similar workloads that allocate a lot of objects that are close to the size of a page. Cc: Matthew Wilcox <matthew@wil.cx> Cc: Andrew Morton <akpm@linux-foundation.org> Acked-by: Christoph Lameter <cl@linux-foundation.org> Signed-off-by: Pekka Enberg <penberg@cs.helsinki.fi>
2008-07-26SL*B: drop kmem cache argument from constructorAlexey Dobriyan
Kmem cache passed to constructor is only needed for constructors that are themselves multiplexeres. Nobody uses this "feature", nor does anybody uses passed kmem cache in non-trivial way, so pass only pointer to object. Non-trivial places are: arch/powerpc/mm/init_64.c arch/powerpc/mm/hugetlbpage.c This is flag day, yes. Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com> Acked-by: Pekka Enberg <penberg@cs.helsinki.fi> Acked-by: Christoph Lameter <cl@linux-foundation.org> Cc: Jon Tollefson <kniht@linux.vnet.ibm.com> Cc: Nick Piggin <nickpiggin@yahoo.com.au> Cc: Matt Mackall <mpm@selenic.com> [akpm@linux-foundation.org: fix arch/powerpc/mm/hugetlbpage.c] [akpm@linux-foundation.org: fix mm/slab.c] [akpm@linux-foundation.org: fix ubifs] Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-07-04Christoph has movedChristoph Lameter
Remove all clameter@sgi.com addresses from the kernel tree since they will become invalid on June 27th. Change my maintainer email address for the slab allocators to cl@linux-foundation.org (which will be the new email address for the future). Signed-off-by: Christoph Lameter <clameter@sgi.com> Signed-off-by: Christoph Lameter <cl@linux-foundation.org> Cc: Pekka Enberg <penberg@cs.helsinki.fi> Cc: Stephen Rothwell <sfr@canb.auug.org.au> Cc: Matt Mackall <mpm@selenic.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-07-03slub: Do not use 192 byte sized cache if minimum alignment is 128 byteChristoph Lameter
The 192 byte cache is not necessary if we have a basic alignment of 128 byte. If it would be used then the 192 would be aligned to the next 128 byte boundary which would result in another 256 byte cache. Two 256 kmalloc caches cause sysfs to complain about a duplicate entry. MIPS needs 128 byte aligned kmalloc caches and spits out warnings on boot without this patch. Signed-off-by: Christoph Lameter <cl@linux-foundation.org> Signed-off-by: Pekka Enberg <penberg@cs.helsinki.fi>
2008-04-27slub: Fallback to minimal order during slab page allocationChristoph Lameter
If any higher order allocation fails then fall back the smallest order necessary to contain at least one object. This enables fallback for all allocations to order 0 pages. The fallback will waste more memory (objects will not fit neatly) and the fallback slabs will be not as efficient as larger slabs since they contain less objects. Note that SLAB also depends on order 1 allocations for some slabs that waste too much memory if forced into PAGE_SIZE'd page. SLUB now can now deal with failing order 1 allocs which SLAB cannot do. Add a new field min that will contain the objects for the smallest possible order for a slab cache. Signed-off-by: Christoph Lameter <clameter@sgi.com> Signed-off-by: Pekka Enberg <penberg@cs.helsinki.fi>
2008-04-27slub: Update statistics handling for variable order slabsChristoph Lameter
Change the statistics to consider that slabs of the same slabcache can have different number of objects in them since they may be of different order. Provide a new sysfs field total_objects which shows the total objects that the allocated slabs of a slabcache could hold. Add a max field that holds the largest slab order that was ever used for a slab cache. Signed-off-by: Christoph Lameter <clameter@sgi.com> Signed-off-by: Pekka Enberg <penberg@cs.helsinki.fi>
2008-04-27slub: Add kmem_cache_order_objects structChristoph Lameter
Pack the order and the number of objects into a single word. This saves some memory in the kmem_cache_structure and more importantly allows us to fetch both values atomically. Later the slab orders become runtime configurable and we need to fetch these two items together in order to properly allocate a slab and initialize its objects. Fix the race by fetching the order and the number of objects in one word. [penberg@cs.helsinki.fi: fix memset() page order in new_slab()] Signed-off-by: Christoph Lameter <clameter@sgi.com> Signed-off-by: Pekka Enberg <penberg@cs.helsinki.fi>
2008-04-14slub: No need for per node slab counters if !SLUB_DEBUGChristoph Lameter
The per node counters are used mainly for showing data through the sysfs API. If that API is not compiled in then there is no point in keeping track of this data. Disable counters for the number of slabs and the number of total slabs if !SLUB_DEBUG. Incrementing the per node counters is also accessing a potentially contended cacheline so this could actually be a performance benefit to embedded systems. SLABINFO support is also affected. It now must depends on SLUB_DEBUG (which is on by default). Patch also avoids a check for a NULL kmem_cache_node pointer in new_slab() if the system is not compiled with NUMA support. [penberg@cs.helsinki.fi: fix oops and move ->nr_slabs into CONFIG_SLUB_DEBUG] Signed-off-by: Christoph Lameter <clameter@sgi.com> Signed-off-by: Pekka Enberg <penberg@cs.helsinki.fi>
2008-03-03slub: Fix up commentsChristoph Lameter
Provide comments and fix up various spelling / style issues. Signed-off-by: Christoph Lameter <clameter@sgi.com>
2008-02-14slub: Support 4k kmallocs again to compensate for page allocator slownessChristoph Lameter
Currently we hand off PAGE_SIZEd kmallocs to the page allocator in the mistaken belief that the page allocator can handle these allocations effectively. However, measurements indicate a minimum slowdown by the factor of 8 (and that is only SMP, NUMA is much worse) vs the slub fastpath which causes regressions in tbench. Increase the number of kmalloc caches by one so that we again handle 4k kmallocs directly from slub. 4k page buffering for the page allocator will be performed by slub like done by slab. At some point the page allocator fastpath should be fixed. A lot of the kernel would benefit from a faster ability to allocate a single page. If that is done then the 4k allocs may again be forwarded to the page allocator and this patch could be reverted. Reviewed-by: Pekka Enberg <penberg@cs.helsinki.fi> Acked-by: Mel Gorman <mel@csn.ul.ie> Signed-off-by: Christoph Lameter <clameter@sgi.com>
2008-02-14slub: Determine gfpflags once and not every time a slab is allocatedChristoph Lameter
Currently we determine the gfp flags to pass to the page allocator each time a slab is being allocated. Determine the bits to be set at the time the slab is created. Store in a new allocflags field and add the flags in allocate_slab(). Acked-by: Mel Gorman <mel@csn.ul.ie> Reviewed-by: Pekka Enberg <penberg@cs.helsinki.fi> Signed-off-by: Christoph Lameter <clameter@sgi.com>
2008-02-14slub: kmalloc page allocator pass-through cleanupPekka Enberg
This adds a proper function for kmalloc page allocator pass-through. While it simplifies any code that does slab tracing code a lot, I think it's a worthwhile cleanup in itself. Signed-off-by: Pekka Enberg <penberg@cs.helsinki.fi> Signed-off-by: Christoph Lameter <clameter@sgi.com>
2008-02-07SLUB: Support for performance statisticsChristoph Lameter
The statistics provided here allow the monitoring of allocator behavior but at the cost of some (minimal) loss of performance. Counters are placed in SLUB's per cpu data structure. The per cpu structure may be extended by the statistics to grow larger than one cacheline which will increase the cache footprint of SLUB. There is a compile option to enable/disable the inclusion of the runtime statistics and its off by default. The slabinfo tool is enhanced to support these statistics via two options: -D Switches the line of information displayed for a slab from size mode to activity mode. -A Sorts the slabs displayed by activity. This allows the display of the slabs most important to the performance of a certain load. -r Report option will report detailed statistics on Example (tbench load): slabinfo -AD ->Shows the most active slabs Name Objects Alloc Free %Fast skbuff_fclone_cache 33 111953835 111953835 99 99 :0000192 2666 5283688 5281047 99 99 :0001024 849 5247230 5246389 83 83 vm_area_struct 1349 119642 118355 91 22 :0004096 15 66753 66751 98 98 :0000064 2067 25297 23383 98 78 dentry 10259 28635 18464 91 45 :0000080 11004 18950 8089 98 98 :0000096 1703 12358 10784 99 98 :0000128 762 10582 9875 94 18 :0000512 184 9807 9647 95 81 :0002048 479 9669 9195 83 65 anon_vma 777 9461 9002 99 71 kmalloc-8 6492 9981 5624 99 97 :0000768 258 7174 6931 58 15 So the skbuff_fclone_cache is of highest importance for the tbench load. Pretty high load on the 192 sized slab. Look for the aliases slabinfo -a | grep 000192 :0000192 <- xfs_btree_cur filp kmalloc-192 uid_cache tw_sock_TCP request_sock_TCPv6 tw_sock_TCPv6 skbuff_head_cache xfs_ili Likely skbuff_head_cache. Looking into the statistics of the skbuff_fclone_cache is possible through slabinfo skbuff_fclone_cache ->-r option implied if cache name is mentioned .... Usual output ... Slab Perf Counter Alloc Free %Al %Fr -------------------------------------------------- Fastpath 111953360 111946981 99 99 Slowpath 1044 7423 0 0 Page Alloc 272 264 0 0 Add partial 25 325 0 0 Remove partial 86 264 0 0 RemoteObj/SlabFrozen 350 4832 0 0 Total 111954404 111954404 Flushes 49 Refill 0 Deactivate Full=325(92%) Empty=0(0%) ToHead=24(6%) ToTail=1(0%) Looks good because the fastpath is overwhelmingly taken. skbuff_head_cache: Slab Perf Counter Alloc Free %Al %Fr -------------------------------------------------- Fastpath 5297262 5259882 99 99 Slowpath 4477 39586 0 0 Page Alloc 937 824 0 0 Add partial 0 2515 0 0 Remove partial 1691 824 0 0 RemoteObj/SlabFrozen 2621 9684 0 0 Total 5301739 5299468 Deactivate Full=2620(100%) Empty=0(0%) ToHead=0(0%) ToTail=0(0%) Descriptions of the output: Total: The total number of allocation and frees that occurred for a slab Fastpath: The number of allocations/frees that used the fastpath. Slowpath: Other allocations Page Alloc: Number of calls to the page allocator as a result of slowpath processing Add Partial: Number of slabs added to the partial list through free or alloc (occurs during cpuslab flushes) Remove Partial: Number of slabs removed from the partial list as a result of allocations retrieving a partial slab or by a free freeing the last object of a slab. RemoteObj/Froz: How many times were remotely freed object encountered when a slab was about to be deactivated. Frozen: How many times was free able to skip list processing because the slab was in use as the cpuslab of another processor. Flushes: Number of times the cpuslab was flushed on request (kmem_cache_shrink, may result from races in __slab_alloc) Refill: Number of times we were able to refill the cpuslab from remotely freed objects for the same slab. Deactivate: Statistics how slabs were deactivated. Shows how they were put onto the partial list. In general fastpath is very good. Slowpath without partial list processing is also desirable. Any touching of partial list uses node specific locks which may potentially cause list lock contention. Signed-off-by: Christoph Lameter <clameter@sgi.com>
2008-02-04Explain kmem_cache_cpu fieldsChristoph Lameter
Add some comments explaining the fields of the kmem_cache_cpu structure. Signed-off-by: Christoph Lameter <clameter@sgi.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
2008-02-04SLUB: rename defrag to remote_node_defrag_ratioChristoph Lameter
The NUMA defrag works by allocating objects from partial slabs on remote nodes. Rename it to remote_node_defrag_ratio to be clear about this. Signed-off-by: Christoph Lameter <clameter@sgi.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
2008-01-02Unify /proc/slabinfo configurationLinus Torvalds
Both SLUB and SLAB really did almost exactly the same thing for /proc/slabinfo setup, using duplicate code and per-allocator #ifdef's. This just creates a common CONFIG_SLABINFO that is enabled by both SLUB and SLAB, and shares all the setup code. Maybe SLOB will want this some day too. Reviewed-by: Pekka Enberg <penberg@cs.helsinki.fi> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-01-01slub: provide /proc/slabinfoPekka J Enberg
This adds a read-only /proc/slabinfo file on SLUB, that makes slabtop work. [ mingo@elte.hu: build fix. ] Cc: Andi Kleen <andi@firstfloor.org> Cc: Christoph Lameter <clameter@sgi.com> Cc: Peter Zijlstra <a.p.zijlstra@chello.nl> Signed-off-by: Pekka Enberg <penberg@cs.helsinki.fi> Signed-off-by: Ingo Molnar <mingo@elte.hu> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-10-17Slab API: remove useless ctor parameter and reorder parametersChristoph Lameter
Slab constructors currently have a flags parameter that is never used. And the order of the arguments is opposite to other slab functions. The object pointer is placed before the kmem_cache pointer. Convert ctor(void *object, struct kmem_cache *s, unsigned long flags) to ctor(struct kmem_cache *s, void *object) throughout the kernel [akpm@linux-foundation.org: coupla fixes] Signed-off-by: Christoph Lameter <clameter@sgi.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-10-16SLUB: Optimize cacheline use for zeroingChristoph Lameter
We touch a cacheline in the kmem_cache structure for zeroing to get the size. However, the hot paths in slab_alloc and slab_free do not reference any other fields in kmem_cache, so we may have to just bring in the cacheline for this one access. Add a new field to kmem_cache_cpu that contains the object size. That cacheline must already be used in the hotpaths. So we save one cacheline on every slab_alloc if we zero. We need to update the kmem_cache_cpu object size if an aliasing operation changes the objsize of an non debug slab. Signed-off-by: Christoph Lameter <clameter@sgi.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-10-16SLUB: Place kmem_cache_cpu structures in a NUMA aware wayChristoph Lameter
The kmem_cache_cpu structures introduced are currently an array placed in the kmem_cache struct. Meaning the kmem_cache_cpu structures are overwhelmingly on the wrong node for systems with a higher amount of nodes. These are performance critical structures since the per node information has to be touched for every alloc and free in a slab. In order to place the kmem_cache_cpu structure optimally we put an array of pointers to kmem_cache_cpu structs in kmem_cache (similar to SLAB). However, the kmem_cache_cpu structures can now be allocated in a more intelligent way. We would like to put per cpu structures for the same cpu but different slab caches in cachelines together to save space and decrease the cache footprint. However, the slab allocators itself control only allocations per node. We set up a simple per cpu array for every processor with 100 per cpu structures which is usually enough to get them all set up right. If we run out then we fall back to kmalloc_node. This also solves the bootstrap problem since we do not have to use slab allocator functions early in boot to get memory for the small per cpu structures. Pro: - NUMA aware placement improves memory performance - All global structures in struct kmem_cache become readonly - Dense packing of per cpu structures reduces cacheline footprint in SMP and NUMA. - Potential avoidance of exclusive cacheline fetches on the free and alloc hotpath since multiple kmem_cache_cpu structures are in one cacheline. This is particularly important for the kmalloc array. Cons: - Additional reference to one read only cacheline (per cpu array of pointers to kmem_cache_cpu) in both slab_alloc() and slab_free(). [akinobu.mita@gmail.com: fix cpu hotplug offline/online path] Signed-off-by: Christoph Lameter <clameter@sgi.com> Cc: "Pekka Enberg" <penberg@cs.helsinki.fi> Cc: Akinobu Mita <akinobu.mita@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-10-16SLUB: Move page->offset to kmem_cache_cpu->offsetChristoph Lameter
We need the offset from the page struct during slab_alloc and slab_free. In both cases we also reference the cacheline of the kmem_cache_cpu structure. We can therefore move the offset field into the kmem_cache_cpu structure freeing up 16 bits in the page struct. Moving the offset allows an allocation from slab_alloc() without touching the page struct in the hot path. The only thing left in slab_free() that touches the page struct cacheline for per cpu freeing is the checking of SlabDebug(page). The next patch deals with that. Use the available 16 bits to broaden page->inuse. More than 64k objects per slab become possible and we can get rid of the checks for that limitation. No need anymore to shrink the order of slabs if we boot with 2M sized slabs (slub_min_order=9). No need anymore to switch off the offset calculation for very large slabs since the field in the kmem_cache_cpu structure is 32 bits and so the offset field can now handle slab sizes of up to 8GB. Signed-off-by: Christoph Lameter <clameter@sgi.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-10-16SLUB: Avoid page struct cacheline bouncing due to remote frees to cpu slabChristoph Lameter
A remote free may access the same page struct that also contains the lockless freelist for the cpu slab. If objects have a short lifetime and are freed by a different processor then remote frees back to the slab from which we are currently allocating are frequent. The cacheline with the page struct needs to be repeately acquired in exclusive mode by both the allocating thread and the freeing thread. If this is frequent enough then performance will suffer because of cacheline bouncing. This patchset puts the lockless_freelist pointer in its own cacheline. In order to make that happen we introduce a per cpu structure called kmem_cache_cpu. Instead of keeping an array of pointers to page structs we now keep an array to a per cpu structure that--among other things--contains the pointer to the lockless freelist. The freeing thread can then keep possession of exclusive access to the page struct cacheline while the allocating thread keeps its exclusive access to the cacheline containing the per cpu structure. This works as long as the allocating cpu is able to service its request from the lockless freelist. If the lockless freelist runs empty then the allocating thread needs to acquire exclusive access to the cacheline with the page struct lock the slab. The allocating thread will then check if new objects were freed to the per cpu slab. If so it will keep the slab as the cpu slab and continue with the recently remote freed objects. So the allocating thread can take a series of just freed remote pages and dish them out again. Ideally allocations could be just recycling objects in the same slab this way which will lead to an ideal allocation / remote free pattern. The number of objects that can be handled in this way is limited by the capacity of one slab. Increasing slab size via slub_min_objects/ slub_max_order may increase the number of objects and therefore performance. If the allocating thread runs out of objects and finds that no objects were put back by the remote processor then it will retrieve a new slab (from the partial lists or from the page allocator) and start with a whole new set of objects while the remote thread may still be freeing objects to the old cpu slab. This may then repeat until the new slab is also exhausted. If remote freeing has freed objects in the earlier slab then that earlier slab will now be on the partial freelist and the allocating thread will pick that slab next for allocation. So the loop is extended. However, both threads need to take the list_lock to make the swizzling via the partial list happen. It is likely that this kind of scheme will keep the objects being passed around to a small set that can be kept in the cpu caches leading to increased performance. More code cleanups become possible: - Instead of passing a cpu we can now pass a kmem_cache_cpu structure around. Allows reducing the number of parameters to various functions. - Can define a new node_match() function for NUMA to encapsulate locality checks. Effect on allocations: Cachelines touched before this patch: Write: page cache struct and first cacheline of object Cachelines touched after this patch: Write: kmem_cache_cpu cacheline and first cacheline of object Read: page cache struct (but see later patch that avoids touching that cacheline) The handling when the lockless alloc list runs empty gets to be a bit more complicated since another cacheline has now to be written to. But that is halfway out of the hot path. Effect on freeing: Cachelines touched before this patch: Write: page_struct and first cacheline of object Cachelines touched after this patch depending on how we free: Write(to cpu_slab): kmem_cache_cpu struct and first cacheline of object Write(to other): page struct and first cacheline of object Read(to cpu_slab): page struct to id slab etc. (but see later patch that avoids touching the page struct on free) Read(to other): cpu local kmem_cache_cpu struct to verify its not the cpu slab. Summary: Pro: - Distinct cachelines so that concurrent remote frees and local allocs on a cpuslab can occur without cacheline bouncing. - Avoids potential bouncing cachelines because of neighboring per cpu pointer updates in kmem_cache's cpu_slab structure since it now grows to a cacheline (Therefore remove the comment that talks about that concern). Cons: - Freeing objects now requires the reading of one additional cacheline. That can be mitigated for some cases by the following patches but its not possible to completely eliminate these references. - Memory usage grows slightly. The size of each per cpu object is blown up from one word (pointing to the page_struct) to one cacheline with various data. So this is NR_CPUS*NR_SLABS*L1_BYTES more memory use. Lets say NR_SLABS is 100 and a cache line size of 128 then we have just increased SLAB metadata requirements by 12.8k per cpu. (Another later patch reduces these requirements) Signed-off-by: Christoph Lameter <clameter@sgi.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-10-16SLUB: direct pass through of page size or higher kmalloc requestsChristoph Lameter
This gets rid of all kmalloc caches larger than page size. A kmalloc request larger than PAGE_SIZE > 2 is going to be passed through to the page allocator. This works both inline where we will call __get_free_pages instead of kmem_cache_alloc and in __kmalloc. kfree is modified to check if the object is in a slab page. If not then the page is freed via the page allocator instead. Roughly similar to what SLOB does. Advantages: - Reduces memory overhead for kmalloc array - Large kmalloc operations are faster since they do not need to pass through the slab allocator to get to the page allocator. - Performance increase of 10%-20% on alloc and 50% on free for PAGE_SIZEd allocations. SLUB must call page allocator for each alloc anyways since the higher order pages which that allowed avoiding the page alloc calls are not available in a reliable way anymore. So we are basically removing useless slab allocator overhead. - Large kmallocs yields page aligned object which is what SLAB did. Bad things like using page sized kmalloc allocations to stand in for page allocate allocs can be transparently handled and are not distinguishable from page allocator uses. - Checking for too large objects can be removed since it is done by the page allocator. Drawbacks: - No accounting for large kmalloc slab allocations anymore - No debugging of large kmalloc slab allocations. Signed-off-by: Christoph Lameter <clameter@sgi.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-08-31SLUB: Force inlining for functions in slub_def.hChristoph Lameter
Some compilers (especially older gcc releases) may skip inlining sometimes which will lead to link failures. Force the inlining of keyfunctions in slub_def.h to avoid these issues. Signed-off-by: Christoph Lameter <clameter@sgi.com> Acked-by: Jan Dittmer <jdi@l4x.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-07-20fix gfp_t annotations for slubAl Viro
Since we have use like ~SLUB_DMA, we ought to have the type set right in both cases. Signed-off-by: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-07-17Slab allocators: Cleanup zeroing allocationsChristoph Lameter
It becomes now easy to support the zeroing allocs with generic inline functions in slab.h. Provide inline definitions to allow the continued use of kzalloc, kmem_cache_zalloc etc but remove other definitions of zeroing functions from the slab allocators and util.c. Signed-off-by: Christoph Lameter <clameter@sgi.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-07-17SLUB: add some more inlines and #ifdef CONFIG_SLUB_DEBUGChristoph Lameter
Add #ifdefs around data structures only needed if debugging is compiled into SLUB. Add inlines to small functions to reduce code size. Signed-off-by: Christoph Lameter <clameter@sgi.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-07-17Slab allocators: consistent ZERO_SIZE_PTR support and NULL result semanticsChristoph Lameter
Define ZERO_OR_NULL_PTR macro to be able to remove the checks from the allocators. Move ZERO_SIZE_PTR related stuff into slab.h. Make ZERO_SIZE_PTR work for all slab allocators and get rid of the WARN_ON_ONCE(size == 0) that is still remaining in SLAB. Make slub return NULL like the other allocators if a too large memory segment is requested via __kmalloc. Signed-off-by: Christoph Lameter <clameter@sgi.com> Acked-by: Pekka Enberg <penberg@cs.helsinki.fi> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-07-16slob: initial NUMA supportPaul Mundt
This adds preliminary NUMA support to SLOB, primarily aimed at systems with small nodes (tested all the way down to a 128kB SRAM block), whether asymmetric or otherwise. We follow the same conventions as SLAB/SLUB, preferring current node placement for new pages, or with explicit placement, if a node has been specified. Presently on UP NUMA this has the side-effect of preferring node#0 allocations (since numa_node_id() == 0, though this could be reworked if we could hand off a pfn to determine node placement), so single-CPU NUMA systems will want to place smaller nodes further out in terms of node id. Once a page has been bound to a node (via explicit node id typing), we only do block allocations from partial free pages that have a matching node id in the page flags. The current implementation does have some scalability problems, in that all partial free pages are tracked in the global freelist (with contention due to the single spinlock). However, these are things that are being reworked for SMP scalability first, while things like per-node freelists can easily be built on top of this sort of functionality once it's been added. More background can be found in: http://marc.info/?l=linux-mm&m=118117916022379&w=2 http://marc.info/?l=linux-mm&m=118170446306199&w=2 http://marc.info/?l=linux-mm&m=118187859420048&w=2 and subsequent threads. Acked-by: Christoph Lameter <clameter@sgi.com> Acked-by: Matt Mackall <mpm@selenic.com> Signed-off-by: Paul Mundt <lethal@linux-sh.org> Acked-by: Nick Piggin <nickpiggin@yahoo.com.au> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-06-16SLUB: minimum alignment fixesChristoph Lameter
If ARCH_KMALLOC_MINALIGN is set to a value greater than 8 (SLUBs smallest kmalloc cache) then SLUB may generate duplicate slabs in sysfs (yes again) because the object size is padded to reach ARCH_KMALLOC_MINALIGN. Thus the size of the small slabs is all the same. No arch sets ARCH_KMALLOC_MINALIGN larger than 8 though except mips which for some reason wants a 128 byte alignment. This patch increases the size of the smallest cache if ARCH_KMALLOC_MINALIGN is greater than 8. In that case more and more of the smallest caches are disabled. If we do that then the count of the active general caches that is displayed on boot is not correct anymore since we may skip elements of the kmalloc array. So count them separately. This approach was tested by Havard yesterday. Signed-off-by: Christoph Lameter <clameter@sgi.com> Cc: Haavard Skinnemoen <hskinnemoen@atmel.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-06-08SLUB: return ZERO_SIZE_PTR for kmalloc(0)Christoph Lameter
Instead of returning the smallest available object return ZERO_SIZE_PTR. A ZERO_SIZE_PTR can be legitimately used as an object pointer as long as it is not deferenced. The dereference of ZERO_SIZE_PTR causes a distinctive fault. kfree can handle a ZERO_SIZE_PTR in the same way as NULL. This enables functions to use zero sized object. e.g. n = number of objects. objects = kmalloc(n * sizeof(object)); for (i = 0; i < n; i++) objects[i].x = y; kfree(objects); Signed-off-by: Christoph Lameter <clameter@sgi.com> Acked-by: Pekka Enberg <penberg@cs.helsinki.fi> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-05-17Slab allocators: define common size limitationsChristoph Lameter
Currently we have a maze of configuration variables that determine the maximum slab size. Worst of all it seems to vary between SLAB and SLUB. So define a common maximum size for kmalloc. For conveniences sake we use the maximum size ever supported which is 32 MB. We limit the maximum size to a lower limit if MAX_ORDER does not allow such large allocations. For many architectures this patch will have the effect of adding large kmalloc sizes. x86_64 adds 5 new kmalloc sizes. So a small amount of memory will be needed for these caches (contemporary SLAB has dynamically sizeable node and cpu structure so the waste is less than in the past) Most architectures will then be able to allocate object with sizes up to MAX_ORDER. We have had repeated breakage (in fact whenever we doubled the number of supported processors) on IA64 because one or the other struct grew beyond what the slab allocators supported. This will avoid future issues and f.e. avoid fixes for 2k and 4k cpu support. CONFIG_LARGE_ALLOCS is no longer necessary so drop it. It fixes sparc64 with SLAB. Signed-off-by: Christoph Lameter <clameter@sgi.com> Signed-off-by: "David S. Miller" <davem@davemloft.net> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-05-17slub: fix handling of oversized slabsAndrew Morton
I'm getting zillions of undefined references to __kmalloc_size_too_large on alpha. For some reason alpha is building out-of-line copies of kmalloc_slab() into lots of compilation units. It turns out that gcc just isn't smart enough to work out that __builtin_contant_p(size)==true implies that __builtin_contant_p(index)==true. So let's give it a bit of help. Cc: Christoph Lameter <clameter@sgi.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-05-17Slab allocators: Drop support for destructorsChristoph Lameter
There is no user of destructors left. There is no reason why we should keep checking for destructors calls in the slab allocators. The RFC for this patch was discussed at http://marc.info/?l=linux-kernel&m=117882364330705&w=2 Destructors were mainly used for list management which required them to take a spinlock. Taking a spinlock in a destructor is a bit risky since the slab allocators may run the destructors anytime they decide a slab is no longer needed. Patch drops destructor support. Any attempt to use a destructor will BUG(). Acked-by: Pekka Enberg <penberg@cs.helsinki.fi> Acked-by: Paul Mundt <lethal@linux-sh.org> Signed-off-by: Christoph Lameter <clameter@sgi.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-05-16SLUB: It is legit to allocate a slab of the maximum permitted sizeChristoph Lameter
Sorry I screwed up the comparison. It is only an error if we attempt to allocate a slab larger than the maximum allowed size. Signed-off-by: Christoph Lameter <clameter@sgi.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-05-15SLUB: CONFIG_LARGE_ALLOCS must consider MAX_ORDER limitChristoph Lameter
Take MAX_ORDER into consideration when determining KMALLOC_SHIFT_HIGH. Otherwise we may run into a situation where we attempt to create general slabs larger than MAX_ORDER. Signed-off-by: Christoph Lameter <clameter@sgi.com> Cc: "David S. Miller" <davem@davemloft.net> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-05-07slub: enable tracking of full slabsChristoph Lameter
If slab tracking is on then build a list of full slabs so that we can verify the integrity of all slabs and are also able to built list of alloc/free callers. Signed-off-by: Christoph Lameter <clameter@sgi.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-05-07SLUB: allocate smallest object size if the user asks for 0 bytesChristoph Lameter
Makes SLUB behave like SLAB in this area to avoid issues.... Throw a stack dump to alert people. At some point the behavior should be switched back. NULL is no memory as far as I can tell and if the use asked for 0 bytes then he need to get no memory. Signed-off-by: Christoph Lameter <clameter@sgi.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-05-07SLUB coreChristoph Lameter
This is a new slab allocator which was motivated by the complexity of the existing code in mm/slab.c. It attempts to address a variety of concerns with the existing implementation. A. Management of object queues A particular concern was the complex management of the numerous object queues in SLAB. SLUB has no such queues. Instead we dedicate a slab for each allocating CPU and use objects from a slab directly instead of queueing them up. B. Storage overhead of object queues SLAB Object queues exist per node, per CPU. The alien cache queue even has a queue array that contain a queue for each processor on each node. For very large systems the number of queues and the number of objects that may be caught in those queues grows exponentially. On our systems with 1k nodes / processors we have several gigabytes just tied up for storing references to objects for those queues This does not include the objects that could be on those queues. One fears that the whole memory of the machine could one day be consumed by those queues. C. SLAB meta data overhead SLAB has overhead at the beginning of each slab. This means that data cannot be naturally aligned at the beginning of a slab block. SLUB keeps all meta data in the corresponding page_struct. Objects can be naturally aligned in the slab. F.e. a 128 byte object will be aligned at 128 byte boundaries and can fit tightly into a 4k page with no bytes left over. SLAB cannot do this. D. SLAB has a complex cache reaper SLUB does not need a cache reaper for UP systems. On SMP systems the per CPU slab may be pushed back into partial list but that operation is simple and does not require an iteration over a list of objects. SLAB expires per CPU, shared and alien object queues during cache reaping which may cause strange hold offs. E. SLAB has complex NUMA policy layer support SLUB pushes NUMA policy handling into the page allocator. This means that allocation is coarser (SLUB does interleave on a page level) but that situation was also present before 2.6.13. SLABs application of policies to individual slab objects allocated in SLAB is certainly a performance concern due to the frequent references to memory policies which may lead a sequence of objects to come from one node after another. SLUB will get a slab full of objects from one node and then will switch to the next. F. Reduction of the size of partial slab lists SLAB has per node partial lists. This means that over time a large number of partial slabs may accumulate on those lists. These can only be reused if allocator occur on specific nodes. SLUB has a global pool of partial slabs and will consume slabs from that pool to decrease fragmentation. G. Tunables SLAB has sophisticated tuning abilities for each slab cache. One can manipulate the queue sizes in detail. However, filling the queues still requires the uses of the spin lock to check out slabs. SLUB has a global parameter (min_slab_order) for tuning. Increasing the minimum slab order can decrease the locking overhead. The bigger the slab order the less motions of pages between per CPU and partial lists occur and the better SLUB will be scaling. G. Slab merging We often have slab caches with similar parameters. SLUB detects those on boot up and merges them into the corresponding general caches. This leads to more effective memory use. About 50% of all caches can be eliminated through slab merging. This will also decrease slab fragmentation because partial allocated slabs can be filled up again. Slab merging can be switched off by specifying slub_nomerge on boot up. Note that merging can expose heretofore unknown bugs in the kernel because corrupted objects may now be placed differently and corrupt differing neighboring objects. Enable sanity checks to find those. H. Diagnostics The current slab diagnostics are difficult to use and require a recompilation of the kernel. SLUB contains debugging code that is always available (but is kept out of the hot code paths). SLUB diagnostics can be enabled via the "slab_debug" option. Parameters can be specified to select a single or a group of slab caches for diagnostics. This means that the system is running with the usual performance and it is much more likely that race conditions can be reproduced. I. Resiliency If basic sanity checks are on then SLUB is capable of detecting common error conditions and recover as best as possible to allow the system to continue. J. Tracing Tracing can be enabled via the slab_debug=T,<slabcache> option during boot. SLUB will then protocol all actions on that slabcache and dump the object contents on free. K. On demand DMA cache creation. Generally DMA caches are not needed. If a kmalloc is used with __GFP_DMA then just create this single slabcache that is needed. For systems that have no ZONE_DMA requirement the support is completely eliminated. L. Performance increase Some benchmarks have shown speed improvements on kernbench in the range of 5-10%. The locking overhead of slub is based on the underlying base allocation size. If we can reliably allocate larger order pages then it is possible to increase slub performance much further. The anti-fragmentation patches may enable further performance increases. Tested on: i386 UP + SMP, x86_64 UP + SMP + NUMA emulation, IA64 NUMA + Simulator SLUB Boot options slub_nomerge Disable merging of slabs slub_min_order=x Require a minimum order for slab caches. This increases the managed chunk size and therefore reduces meta data and locking overhead. slub_min_objects=x Mininum objects per slab. Default is 8. slub_max_order=x Avoid generating slabs larger than order specified. slub_debug Enable all diagnostics for all caches slub_debug=<options> Enable selective options for all caches slub_debug=<o>,<cache> Enable selective options for a certain set of caches Available Debug options F Double Free checking, sanity and resiliency R Red zoning P Object / padding poisoning U Track last free / alloc T Trace all allocs / frees (only use for individual slabs). To use SLUB: Apply this patch and then select SLUB as the default slab allocator. [hugh@veritas.com: fix an oops-causing locking error] [akpm@linux-foundation.org: various stupid cleanups and small fixes] Signed-off-by: Christoph Lameter <clameter@sgi.com> Signed-off-by: Hugh Dickins <hugh@veritas.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>