- 12 Nov, 2009 1 commit
-
-
Hugh Dickins authored
in page->mapping, with the higher bits a pointer to the anon_vma; and have defined PageKsm(page) as that with NULL anon_vma. But KSM swapping will need to store a pointer there: so in preparation for that, now define PAGE_MAPPING_FLAGS as the low two bits, including PAGE_MAPPING_KSM (always set along with PAGE_MAPPING_ANON, until some other use for the bit emerges). Declare page_rmapping(page) to return the pointer part of page->mapping, and page_anon_vma(page) to return the anon_vma pointer when that's what it is. Use these in a few appropriate places: notably, unuse_vma() has been testing page->mapping, but is better to be testing page_anon_vma() (cases may be added in which flag bits are set without any pointer). Signed-off-by:
Hugh Dickins <hugh.dickins@tiscali.co.uk> Cc: Izik Eidus <ieidus@redhat.com> Cc: Andrea Arcangeli <aarcange@redhat.com> Cc: Nick Piggin <npiggin@suse.de> Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> Cc: Rik van Riel <riel@redhat.com> Cc: Lee Schermerhorn <Lee.Schermerhorn@hp.com> Cc: Andi Kleen <andi@firstfloor.org> Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> Cc: Wu Fengguang <fengguang.wu@intel.com> Cc: Minchan Kim <minchan.kim@gmail.com> Signed-off-by:
Andrew Morton <akpm@linux-foundation.org>
-
- 14 Nov, 2009 1 commit
-
-
KOSAKI Motohiro authored
Once the priority is higher, kswapd starts waiting on congestion. However, if the zone is below the min watermark then kswapd needs to continue working without delay as there is a danger of an increased rate of GFP_ATOMIC allocation failure. This patch changes the conditions under which kswapd waits on congestion by only going to sleep if the min watermarks are being met. [mel@csn.ul.ie: Add stats to track how relevant the logic is] Signed-off-by:
KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> Signed-off-by:
Mel Gorman <mel@csn.ul.ie> Reviewed-by:
Rik van Riel <riel@redhat.com> Signed-off-by:
-
- 12 Nov, 2009 4 commits
-
-
Mel Gorman authored
event of no IO congestion, kswapd can go to sleep very shortly after the high watermark was reached. If there are a constant stream of allocations from parallel processes, it can mean that kswapd went to sleep too quickly and the high watermark is not being maintained for sufficient length time. This patch makes kswapd go to sleep as a two-stage process. It first tries to sleep for HZ/10. If it is woken up by another process or the high watermark is no longer met, it's considered a premature sleep and kswapd continues work. Otherwise it goes fully to sleep. This adds more counters to distinguish between fast and slow breaches of watermarks. A "fast" premature sleep is one where the low watermark was hit in a very short time after kswapd going to sleep. A "slow" premature sleep indicates that the high watermark was breached after a very short interval. Signed-off-by:
-
Mel Gorman authored
that the commits 373c0a7e 8aa7e847 dramatically increased the number of GFP_ATOMIC failures that were occuring within a wireless driver. Reverting this patch seemed to help a lot even though it was pointed out that the congestion changes were very far away from high-order atomic allocations. The key to why the revert makes such a big difference is down to timing and how long direct reclaimers wait versus kswapd. With the patch reverted, the congestion_wait() is on the SYNC queue instead of the ASYNC. As a significant part of the workload involved reads, it makes sense that the SYNC list is what was truely congested and with the revert processes were waiting on congestion as expected. Hence, direct reclaimers stalled properly and kswapd was able to do its job with fewer stalls. This patch aims to fix the congestion_wait() behaviour for SYNC and ASYNC for direct reclaimers. Instead of making the congestion_wait() on the SYNC queue which would only fix a particular type of workload, this patch adds a third type of congestion_wait - BLK_RW_BOTH which first waits on the ASYNC and then the SYNC queue if the timeout has not been reached. In tests, this counter-intuitively results in kswapd stalling less and freeing up pages resulting in fewer allocation failures and fewer direct-reclaim-orientated stalls. Signed-off-by:
-
Andrew Morton authored
#99: FILE: mm/oom_kill.c:209: + ^I * to kill current.We have to random task kill in this case.$ ERROR: code indent should use tabs where possible #100: FILE: mm/oom_kill.c:210: + ^I * Hopefully, CONSTRAINT_THISNODE...but no way to handle it, now.$ ERROR: code indent should use tabs where possible #101: FILE: mm/oom_kill.c:211: + ^I */$ ERROR: code indent should use tabs where possible #107: FILE: mm/oom_kill.c:216: + ^I * The nodemask here is a nodemask passed to alloc_pages(). Now,$ ERROR: code indent should use tabs where possible #108: FILE: mm/oom_kill.c:217: + ^I * cpuset doesn't use this nodemask for its hardwall/softwall/hierarchy$ ERROR: code indent should use tabs where possible #109: FILE: mm/oom_kill.c:218: + ^I * feature. mempolicy is an only user of nodemask here.$ ERROR: code indent should use tabs where possible #111: FILE: mm/oom_kill.c:220: + ^I */$ ERROR: code indent should use tabs where possible #169: FILE: mm/page_alloc.c:1672: +^I ^I* GFP_THISNODE contains __GFP_NORETRY and we never hit this.$ ERROR: code indent should use tabs where possible #170: FILE: mm/page_alloc.c:1673: +^I ^I* Sanity check for bare calls of __GFP_THISNODE, not real OOM.$ ERROR: code indent should use tabs where possible #171: FILE: mm/page_alloc.c:1674: +^I ^I* The caller should handle page allocation failure by itself if$ ERROR: code indent should use tabs where possible #172: FILE: mm/page_alloc.c:1675: +^I ^I* it specifies __GFP_THISNODE.$ ERROR: code indent should use tabs where possible #173: FILE: mm/page_alloc.c:1676: +^I ^I* Note: Hugepage uses it but will hit PAGE_ALLOC_COSTLY_ORDER.$ ERROR: code indent should use tabs where possible #174: FILE: mm/page_alloc.c:1677: +^I ^I*/$ total: 13 errors, 0 warnings, 125 lines checked ./patches/oom-kill-fix-numa-consraint-check-with-nodemask-v42.patch has style problems, please review. If any of these errors are false positives report them to the maintainer, see CHECKPATCH in MAINTAINERS. Please run checkpatch prior to sending patches Cc: Christoph Lameter <cl@linux-foundation.org> Cc: Daisuke Nishimura <nishimura@mxp.nes.nec.co.jp> Cc: David Rientjes <rientjes@google.com> Cc: KAMEZAWA Hiroyuki <kamezawa.hioryu@jp.fujitsu.com> Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> Signed-off-by: -
KAMEZAWA Hiroyuki authored
as a bugfix not as an ehnancement. In these days, things are changed as - alloc_pages() eats nodemask as its arguments, __alloc_pages_nodemask(). - mempolicy don't maintain its own private zonelists. (And cpuset doesn't use nodemask for __alloc_pages_nodemask()) So, current oom-killer's check function is wrong. This patch does - check nodemask, if nodemask && nodemask doesn't cover all node_states[N_HIGH_MEMORY], this is CONSTRAINT_MEMORY_POLICY. - Scan all zonelist under nodemask, if it hits cpuset's wall this faiulre is from cpuset. And - modifies the caller of out_of_memory not to call oom if __GFP_THISNODE. This doesn't change "current" behavior. If callers use __GFP_THISNODE it should handle "page allocation failure" by itself. - handle __GFP_NOFAIL+__GFP_THISNODE path. This is something like a FIXME but this gfpmask is not used now. Signed-off-by:KAMEZAWA Hiroyuki <kamezawa.hioryu@jp.fujitsu.com> Acked-by:
David Rientjes <rientjes@google.com> Cc: Daisuke Nishimura <nishimura@mxp.nes.nec.co.jp> Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> Cc: Christoph Lameter <cl@linux-foundation.org> Signed-off-by:
-
- 11 Nov, 2009 1 commit
-
-
David Rientjes authored
On Tue, 10 Nov 2009, akpm@linux-foundation.org wrote: > diff -puN mm/oom_kill.c~oom-kill-show-virtual-size-and-rss-information-of-the-killed-process mm/oom_kill.c > --- a/mm/oom_kill.c~oom-kill-show-virtual-size-and-rss-information-of-the-killed-process > +++ a/mm/oom_kill.c > @@ -352,6 +352,8 @@ static void dump_header(gfp_t gfp_mask, > dump_tasks(mem); > } > > +#define K(x) ((x) << (PAGE_SHIFT-10)) > + > /* > * Send SIGKILL to the selected process irrespective of CAP_SYS_RAW_IO > * flag though it's unlikely that we select a process with CAP_SYS_RAW_IO > @@ -371,9 +373,16 @@ static void __oom_kill_task(struct task_ > return; > } > > - if (verbose) > - printk(KERN_ERR "Killed process %d (%s)\n", > - task_pid_nr(p), p->comm); > + if (verbose) { > + task_lock(p); > + printk(KERN_ERR "Killed process %d (%s) " > + "vsz:%lukB, anon-rss:%lukB, file-rss:%lukB\n", > + task_pid_nr(p), p->comm, > + K(p->mm->total_vm), > + K(get_mm_counter(p->mm, anon_rss)), > + K(get_mm_counter(p->mm, file_rss))); > + task_unlock(p); > + } > > /* > * We give our sacrificial lamb high priority and access to There's a race there which can dereference a NULL p->mm. p->mm is protected by task_lock(), but there's no check added here that ensures p->mm is still valid. The previous check for !p->mm in __oom_kill_task() is not protected by task_lock(), so there's a race: select_bad_process() oom_kill_process(p) do_exit() exit_signals(p) /* PF_EXITING */ oom_kill_task(p) __oom_kill_task(p) exit_mm(p) task_lock(p) p->mm = NULL task_unlock(p) printk() of p->mm->total_vm Please merge this as a fix. Signed-off-by:
-
- 10 Nov, 2009 1 commit
-
-
KOSAKI Motohiro authored
killed process has a memory leak or not at the first step. This patch adds vsz and rss information to the oom log to help this analysis. To save time for the debugging. example: =================================================================== rsyslogd invoked oom-killer: gfp_mask=0x201da, order=0, oom_adj=0 Pid: 1308, comm: rsyslogd Not tainted 2.6.32-rc6 #24 Call Trace: [<ffffffff8132e35b>] ?_spin_unlock+0x2b/0x40 [<ffffffff810f186e>] oom_kill_process+0xbe/0x2b0 (snip) 492283 pages non-shared Out of memory: kill process 2341 (memhog) score 527276 or a child Killed process 2341 (memhog) vsz:1054552kB, anon-rss:970588kB, file-rss:4kB =========================================================================== ^ | here Signed-off-by:
-
- 30 Oct, 2009 1 commit
-
-
KAMEZAWA Hiroyuki authored
reproduce it easily. Now, oom-killer uses mm->total_vm as its base value. But in recent applications, there are a big gap between VM size and RSS size. Because - Applications attaches much dynamic libraries. (Gnome, KDE, etc...) - Applications may alloc big VM area but use small part of them. (Java, and multi-threaded applications has this tendency because of default-size of stack.) I think using mm->total_vm as score for oom-kill is not good. By the same reason, overcommit memory can't work as expected. (In other words, if we depends on total_vm, using overcommit more positive is a good choice.) This patch uses mm->anon_rss/file_rss as base value for calculating badness. Following is changes to OOM score(badness) on an environment with 1.6G memory plus memory-eater(500M & 1G). Top 10 of badness score. (The highest one is the first candidate to be killed) Before badness program 91228 gnome-settings- 94210 clock-applet 103202 mixer_applet2 106563 tomboy 112947 gnome-terminal 128944 mmap <----------- 500M malloc 129332 nautilus 215476 bash <----------- parent of 2 mallocs. 256944 mmap <----------- 1G malloc 423586 gnome-session After badness 1911 mixer_applet2 1955 clock-applet 1986 xinit 1989 gnome-session 2293 nautilus 2955 gnome-terminal 4113 tomboy 104163 mmap <----------- 500M malloc. 168577 bash <----------- parent of 2 mallocs 232375 mmap <----------- 1G malloc seems good for me. Maybe we can tweak this patch more, but this one will be a good one as a start point. Signed-off-by:KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> Reviewed-by:
Minchan Kim <minchan.kim@gmail.com> Cc: David Rientjes <rientjes@google.com> Cc: Hugh Dickins <hugh.dickins@tiscali.co.uk> Cc: Andrea Arcangeli <aarcange@redhat.com> Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> Signed-off-by:
-
- 03 Nov, 2009 2 commits
-
-
Huang Shijie authored
no need to check it. Signed-off-by:
Huang Shijie <shijie8@gmail.com> Acked-by:
Wu Fengguang <fengguang.wu@intel.com> Signed-off-by:
-
Huang Shijie authored
Signed-off-by:
-
- 16 Oct, 2009 2 commits
-
-
Huang Shijie authored
Signed-off-by:
-
Vincent Li authored
cleanups") removed generic_file_write() in filemap. Change the comment in vmscan pageout() to __generic_file_aio_write(). Signed-off-by:
Vincent Li <macli@brc.ubc.ca> Cc: Christoph Hellwig <hch@lst.de> Signed-off-by:
-
- 15 Oct, 2009 1 commit
-
-
Hugh Dickins authored
use of its cachelines: it's good for swap_duplicate() in particular if unsigned int max and swap_map are near the start. Signed-off-by:
-
- 11 Nov, 2009 5 commits
-
-
Hugh Dickins authored
value to shmem/tmpfs swap pages: their swap counts are never incremented, and it helps swapoff's try_to_unuse() a little if it can immediately distinguish those pages from process pages. Since we've no use for SWAP_MAP_BAD | COUNT_CONTINUED, we might as well use that 0xbf value for SWAP_MAP_SHMEM. Signed-off-by:
-
Hugh Dickins authored
swap page is inserted into another mm (when forking finds a swap entry in place of a pte, or when reclaim unmaps a pte to insert the swap entry). swap_info_struct's vmalloc'ed swap_map is the array of these reference counts: but what happens when the unsigned short (or unsigned char since the preceding patch) is full? (and its high bit is kept for a cache flag) We then lose track of it, never freeing, leaving it in use until swapoff: at which point we _hope_ that a single pass will have found all instances, assume there are no more, and will lose user data if we're wrong. Swapping of KSM pages has not yet been enabled; but it is implemented, and makes it very easy for a user to overflow the maximum swap count: possible with ordinary process pages, but unlikely, even when pid_max has been raised from PID_MAX_DEFAULT. This patch implements swap count continuations: when the count overflows, a continuation page is allocated and linked to the original vmalloc'ed map page, and this used to hold the continuation counts for that entry and its neighbours. These continuation pages are seldom referenced: the common paths all work on the original swap_map, only referring to a continuation page when the low "digit" of a count is incremented or decremented through SWAP_MAP_MAX. Signed-off-by:
-
Hugh Dickins authored
chars: it's still very unusual to reach a swap count of 126, and the next patch allows it to be extended indefinitely. Signed-off-by:
-
Hugh Dickins authored
encode_swapmap() obscure what happens in the swap_map entry, just at those points where I need to understand it. Remove them, and pass more usable "usage" values to scan_swap_map(), swap_entry_free() and __swap_duplicate(), instead of the SWAP_MAP and SWAP_CACHE enum. Signed-off-by:
-
Hugh Dickins authored
block, remove extraneous whitespace and return, fix typo in a comment. Signed-off-by:
-
- 03 Nov, 2009 2 commits
-
-
Hugh Dickins authored
initializing p->first_swap_extent.list at a point before p has been decided - we may kfree that newly allocated p and go on to reuse an existing free entry for p. Now, the patch is not actually wrong: an existing free entry will have a good empty first_swap_extent.list; but it looks suspicious, it seems strange to initialize a field in something we're about to kfree, and I'd rather we put that initialization back to where it was in 2.6.32. Signed-off-by:
-
Jiri Slaby authored
BUG: unable to handle kernel NULL pointer dereference at (null) IP: [<ffffffff810af160>] sys_swapon+0x1f0/0xc60 PGD 1dc0b067 PUD 1dc09067 PMD 0 Oops: 0000 [#1] SMP last sysfs file: CPU 1 Modules linked in: Pid: 562, comm: swapon Tainted: G W 2.6.32-rc5-mm1_64 #867 RIP: 0010:[<ffffffff810af160>] [<ffffffff810af160>] sys_swapon+0x1f0/0xc60 ... It is due to swap_info_struct->first_swap_extent.list not being initialized. ->next is NULL in such a situation and destroy_swap_extents fails to iterate over the list with the BUG above. Introduced by swap_info-include-first_swap_extent.patch. Revert the INIT_LIST_HEAD move. Signed-off-by:
Jiri Slaby <jirislaby@gmail.com> Acked-by:
-
- 11 Nov, 2009 3 commits
-
-
Hugh Dickins authored
swap_info_struct, instead of just the list_head: swap partitions need only that one, and for others it's used as a circular list anyway. Signed-off-by:
-
Hugh Dickins authored
to reserve an array of about 30 of them in bss, when most people will want only one. Change swap_info[] to an array of pointers. That does need a "type" field in the structure: pack it as a char with next type and short prio (aha, char is unsigned by default on PowerPC). Use the (admittedly peculiar) name "type" throughout for this index. /proc/swaps does not take swap_lock: I wouldn't want it to, but do take care with barriers when adding a new item to the array (never removed). Signed-off-by:
-
Hugh Dickins authored
one other in-tree user: get_swap_bio(). Adjust its interface to map_swap_page(), so that we can then remove get_swap_info_struct(). But there is a popular user out-of-tree, TuxOnIce: so leave the declaration of swap_info_struct in linux/swap.h. Signed-off-by:
-
- 13 Oct, 2009 2 commits
-
-
Jan Beulich authored
being called through vmalloc_32{,_user}() - explicitly allow using high memory here even if the outer allocation request doesn't allow it Signed-off-by:Jan Beulich <jbeulich@novell.com> Acked-by:
-
David Rientjes authored
backed by slab caches that are not of large order, traditionally never greater than PAGE_ALLOC_COSTLY_ORDER. Thus, using GFP_KERNEL for these allocations on large machines when CONFIG_NODES_SHIFT > 8 will cause the page allocator to loop endlessly in the allocation attempt, each time invoking both direct reclaim or the oom killer. This is of particular interest when using NODEMASK_ALLOC() from a mempolicy context (either directly in mm/mempolicy.c or the mempolicy constrained hugetlb allocations) since the oom killer always kills current when allocations are constrained by mempolicies. So for all present use cases in the kernel, current would end up being oom killed when direct reclaim fails. That would allow the NODEMASK_ALLOC() to succeed but current would have sacrificed itself upon returning. This patch adds gfp flags to NODEMASK_ALLOC() to pass to kmalloc() on CONFIG_NODES_SHIFT > 8; this parameter is a nop on other configurations. All current use cases either directly from hugetlb code or indirectly via NODEMASK_SCRATCH() union __GFP_NORETRY to avoid direct reclaim and the oom killer when the slab allocator needs to allocate additional pages. The side-effect of this change is that all current use cases of either NODEMASK_ALLOC() or NODEMASK_SCRATCH() need appropriate -ENOMEM handling when the allocation fails (never for CONFIG_NODES_SHIFT <= 8). All current use cases were audited and do have appropriate error handling at this time. Signed-off-by:
-
- 03 Nov, 2009 1 commit
-
-
Lee Schermerhorn authored
drivers/base/node.c:484: error: implicit declaration of function 'init_node_hugetlb_work' drivers/base/node.c:582: error: 'node_memory_callback' undeclared (first use in this function) Reported-by:
Randy Dunlap <randy.dunlap@oracle.com> Signed-off-by:
Lee Schermerhorn <lee.schermerhorn@hp.com> Signed-off-by:
-
- 13 Oct, 2009 2 commits
-
-
Lee Schermerhorn authored
attributes to a worker thread rather than attempt the allocation/attachment or detachment/freeing of the attributes in the context of the memory hotplug handler. I don't know that this is absolutely required, but the registration can sleep in allocations and other mem hot plug handlers do it this way. If it turns out this is NOT required, we can drop this patch. N.B., Only tested build, boot, libhugetlbfs regression. i.e., no memory hotplug testing. Signed-off-by:Andi Kleen <andi@firstfloor.org> Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> Cc: Lee Schermerhorn <lee.schermerhorn@hp.com> Cc: Mel Gorman <mel@csn.ul.ie> Cc: Randy Dunlap <randy.dunlap@oracle.com> Cc: Nishanth Aravamudan <nacc@us.ibm.com> Cc: David Rientjes <rientjes@google.com> Cc: Adam Litke <agl@us.ibm.com> Cc: Andy Whitcroft <apw@canonical.com> Cc: Eric Whitney <eric.whitney@hp.com> Cc: Christoph Lameter <cl@linux-foundation.org> Signed-off-by:
-
Lee Schermerhorn authored
suggested by David Rientjes. With Memory Hotplug, memory can be added to a memoryless node and a node with memory can become memoryless. Therefore, add a memory on/off-line notifier callback to [un]register a node's attributes on transition to/from memoryless state. N.B., Only tested build, boot, libhugetlbfs regression. i.e., no memory hotplug testing. Signed-off-by:
-
- 11 Nov, 2009 1 commit
-
-
David Rientjes authored
there are no present pages left. In such a situation, kswapd must also be stopped since it has nothing left to do. Signed-off-by:
-
- 13 Oct, 2009 4 commits
-
-
Lee Schermerhorn authored
Global replacement of 'all online nodes" with "all nodes with memory" in mm/hugetlb.c. Suggested by David Rientjes. A subsequent patch will handle adding/removing of per node hstate sysfs attributes when nodes transition to/from memoryless state via memory hotplug. NOTE: this patch has not been tested with memoryless nodes. Signed-off-by:
-
Lee Schermerhorn authored
policy based huge page management. Additionaly, the patch includes a fair amount of rework to improve consistency, eliminate duplication and set the context for documenting the memory policy interaction. Signed-off-by:
-
Lee Schermerhorn authored
sysdevs: /sys/devices/system/node/node<ID>/hugepages/hugepages-<size>/ nr_hugepages - r/w free_huge_pages - r/o surplus_huge_pages - r/o The patch attempts to re-use/share as much of the existing global hstate attribute initialization and handling, and the "nodes_allowed" constraint processing as possible. Calling set_max_huge_pages() with no node indicates a change to global hstate parameters. In this case, any non-default task mempolicy will be used to generate the nodes_allowed mask. A valid node id indicates an update to that node's hstate parameters, and the count argument specifies the target count for the specified node. From this info, we compute the target global count for the hstate and construct a nodes_allowed node mask contain only the specified node. Setting the node specific nr_hugepages via the per node attribute effectively ignores any task mempolicy or cpuset constraints. With this patch: (me):ls /sys/devices/system/node/node0/hugepages/hugepages-2048kB ./ ../ free_hugepages nr_hugepages surplus_hugepages Starting from: Node 0 HugePages_Total: 0 Node 0 HugePages_Free: 0 Node 0 HugePages_Surp: 0 Node 1 HugePages_Total: 0 Node 1 HugePages_Free: 0 Node 1 HugePages_Surp: 0 Node 2 HugePages_Total: 0 Node 2 HugePages_Free: 0 Node 2 HugePages_Surp: 0 Node 3 HugePages_Total: 0 Node 3 HugePages_Free: 0 Node 3 HugePages_Surp: 0 vm.nr_hugepages = 0 Allocate 16 persistent huge pages on node 2: (me):echo 16 >/sys/devices/system/node/node2/hugepages/hugepages-2048kB/nr_hugepages [Note that this is equivalent to: numactl -m 2 hugeadmin --pool-pages-min 2M:+16 ] Yields: Node 0 HugePages_Total: 0 Node 0 HugePages_Free: 0 Node 0 HugePages_Surp: 0 Node 1 HugePages_Total: 0 Node 1 HugePages_Free: 0 Node 1 HugePages_Surp: 0 Node 2 HugePages_Total: 16 Node 2 HugePages_Free: 16 Node 2 HugePages_Surp: 0 Node 3 HugePages_Total: 0 Node 3 HugePages_Free: 0 Node 3 HugePages_Surp: 0 vm.nr_hugepages = 16 Global controls work as expected--reduce pool to 8 persistent huge pages: (me):echo 8 >/sys/kernel/mm/hugepages/hugepages-2048kB/nr_hugepages Node 0 HugePages_Total: 0 Node 0 HugePages_Free: 0 Node 0 HugePages_Surp: 0 Node 1 HugePages_Total: 0 Node 1 HugePages_Free: 0 Node 1 HugePages_Surp: 0 Node 2 HugePages_Total: 8 Node 2 HugePages_Free: 8 Node 2 HugePages_Surp: 0 Node 3 HugePages_Total: 0 Node 3 HugePages_Free: 0 Node 3 HugePages_Surp: 0 Signed-off-by:
-
Lee Schermerhorn authored
to generic header 'linux/numa.h' for use in generic code. NUMA_NO_NODE replaces bare '-1' where it's used in this series to indicate "no node id specified". Ultimately, it can be used to replace the -1 elsewhere where it is used similarly. Signed-off-by:
-
- 13 Nov, 2009 1 commit
-
-
Lee Schermerhorn authored
the task modifying the number of persistent huge pages to control the allocation, freeing and adjusting of surplus huge pages when the pool page count is modified via the new sysctl or sysfs attribute "nr_hugepages_mempolicy". The nodes_allowed mask is derived as follows: * For "default" [NULL] task mempolicy, a NULL nodemask_t pointer is produced. This will cause the hugetlb subsystem to use node_online_map as the "nodes_allowed". This preserves the behavior before this patch. * For "preferred" mempolicy, including explicit local allocation, a nodemask with the single preferred node will be produced. "local" policy will NOT track any internode migrations of the task adjusting nr_hugepages. * For "bind" and "interleave" policy, the mempolicy's nodemask will be used. * Other than to inform the construction of the nodes_allowed node mask, the actual mempolicy mode is ignored. That is, all modes behave like interleave over the resulting nodes_allowed mask with no "fallback". See the updated documentation [next patch] for more information about the implications of this patch. Examples: Starting with: Node 0 HugePages_Total: 0 Node 1 HugePages_Total: 0 Node 2 HugePages_Total: 0 Node 3 HugePages_Total: 0 Default behavior [with or without this patch] balances persistent hugepage allocation across nodes [with sufficient contiguous memory]: sysctl vm.nr_hugepages[_mempolicy]=32 yields: Node 0 HugePages_Total: 8 Node 1 HugePages_Total: 8 Node 2 HugePages_Total: 8 Node 3 HugePages_Total: 8 Of course, we only have nr_hugepages_mempolicy with the patch, but with default mempolicy, nr_hugepages_mempolicy behaves the same as nr_hugepages. Applying mempolicy--e.g., with numactl [using '-m' a.k.a. '--membind' because it allows multiple nodes to be specified and it's easy to type]--we can allocate huge pages on individual nodes or sets of nodes. So, starting from the condition above, with 8 huge pages per node, add 8 more to node 2 using: numactl -m 2 sysctl vm.nr_hugepages_mempolicy=40 This yields: Node 0 HugePages_Total: 8 Node 1 HugePages_Total: 8 Node 2 HugePages_Total: 16 Node 3 HugePages_Total: 8 The incremental 8 huge pages were restricted to node 2 by the specified mempolicy. Similarly, we can use mempolicy to free persistent huge pages from specified nodes: numactl -m 0,1 sysctl vm.nr_hugepages_mempolicy=32 yields: Node 0 HugePages_Total: 4 Node 1 HugePages_Total: 4 Node 2 HugePages_Total: 16 Node 3 HugePages_Total: 8 The 8 huge pages freed were balanced over nodes 0 and 1. [rientjes@google.com: accomodate reworked NODEMASK_ALLOC] Signed-off-by:
-
- 13 Oct, 2009 5 commits
-
-
Lee Schermerhorn authored
This will be used to populate the huge pages "nodes_allowed" nodemask for a single node when basing nodes_allowed on a preferred/local mempolicy or when a persistent huge page pool page count is modified via a per node sysfs attribute. Signed-off-by:
-
David Rientjes authored
> @@ -1144,14 +1156,15 @@ static void __init report_hugepages(void > } > > #ifdef CONFIG_HIGHMEM > -static void try_to_free_low(struct hstate *h, unsigned long count) > +static void try_to_free_low(struct hstate *h, unsigned long count, > + nodemask_t *nodes_allowed) > { > int i; > > if (h->order >= MAX_ORDER) > return; > > - for (i = 0; i < MAX_NUMNODES; ++i) { > + for_each_node_mask(node, nodes_allowed_) { > struct page *page, *next; > struct list_head *freel = &h->hugepage_freelists[i]; > list_for_each_entry_safe(page, next, freel, lru) { That's not looking good for i386, Andrew please fold the following into this patch when it's merged into -mm: [rientjes@google.com: fix HIGHMEM compile error] Signed-off-by: -
Lee Schermerhorn authored
controlling task's numa mempolicy, add a "nodes_allowed" nodemask pointer to the allocate, free and surplus adjustment functions. For now, pass NULL to indicate default behavior--i.e., use node_online_map. A subsqeuent patch will derive a non-default mask from the controlling task's numa mempolicy. Note that this method of updating the global hstate nr_hugepages under the constraint of a nodemask simplifies keeping the global state consistent--especially the number of persistent and surplus pages relative to reservations and overcommit limits. There are undoubtedly other ways to do this, but this works for both interfaces: mempolicy and per node attributes. Signed-off-by:
-
Lee Schermerhorn authored
obtain the "start_nid". Then, whereas prior to this patch we unconditionally called hstate_next_node_to_{alloc|free}(), whether or not we successfully allocated/freed a huge page on the node, now we only call these functions on failure to alloc/free to advance to next allowed node. Factor out the next_node_allowed() function to handle wrap at end of node_online_map. In this version, the allowed nodes include all of the online nodes. Signed-off-by: -
David Rientjes authored
allocation and freeing of persistent huge pages on a NUMA platform. Please consider for merging into mmotm. This series uses two mechanisms to constrain the nodes from which persistent huge pages are allocated: 1) the task NUMA mempolicy of the task modifying a new sysctl "nr_hugepages_mempolicy", based on a suggestion by Mel Gorman; and 2) a subset of the hugepages hstate sysfs attributes have been added [in V4] to each node system device under: /sys/devices/node/node[0-9]*/hugepages The per node attibutes allow direct assignment of a huge page count on a specific node, regardless of the task's mempolicy or cpuset constraints. This patch: NODEMASK_ALLOC(x, m) assumes x is a type of struct, which is unnecessary. It's perfectly reasonable to use this macro to allocate a nodemask_t, which is anonymous, either dynamically or on the stack depending on NODES_SHIFT. Signed-off-by:
-
