- 10 Aug, 2010 40 commits
-
-
KAMEZAWA Hiroyuki authored
When taking a memory snapshot in hibernate_snapshot(), all (directly called) memory allocations use GFP_ATOMIC. Hence swap misusage during hibernation never occurs. But from a pessimistic point of view, there is no guarantee that no page allcation has __GFP_WAIT. It is better to have a global indication "we enter hibernation, don't use swap!". This patch tries to freeze new-swap-allocation during hibernation. (All user processes are frozenm so swapin is not a concern). This way, no updates will happen to swap_map[] between hibernate_snapshot() and save_image(). Swap is thawed when swsusp_free() is called. We can be assured that swap corruption will not occur. Signed-off-by:
KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> Cc: "Rafael J. Wysocki" <rjw@sisk.pl> Cc: Hugh Dickins <hughd@google.com> Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> Cc: Ondrej Zary <linux@rainbow-software.org> Cc: Balbir Singh <balbir@in.ibm.com> Cc: Andrea Arcangeli <aarcange@redhat.com> Signed-off-by:
Andrew Morton <akpm@linux-foundation.org> Signed-off-by:
Linus Torvalds <torvalds@linux-foundation.org>
-
KAMEZAWA Hiroyuki authored
Since 2.6.31, swap_map[]'s refcounting was changed to show that a used swap entry is just for swap-cache, can be reused. Then, while scanning free entry in swap_map[], a swap entry may be able to be reclaimed and reused. It was caused by commit c9e44410 ("mm: reuse unused swap entry if necessary"). But this caused deta corruption at resume. The scenario is - Assume a clean-swap cache, but mapped. - at hibernation_snapshot[], clean-swap-cache is saved as clean-swap-cache and swap_map[] is marked as SWAP_HAS_CACHE. - then, save_image() is called. And reuse SWAP_HAS_CACHE entry to save image, and break the contents. After resume: - the memory reclaim runs and finds clean-not-referenced-swap-cache and discards it because it's marked as clean. But here, the contents on disk and swap-cache is inconsistent. Hance memory is corrupted. This patch avoids the bug by not reclaiming swap-entry during hibernation. This is a quick fix for backporting. Signed-off-by:
Ondreg Zary <linux@rainbow-software.org> Tested-by:
Andrea Gelmini <andrea.gelmini@gmail.com> Acked-by:
Hugh Dickins <hughd@google.com> Cc: <stable@kernel.org> Signed-off-by:
-
Michael Cree authored
This implements hardware performance events for the EV67 and later CPUs within the Linux performance events subsystem. Only using the performance monitoring unit in HP/Compaq's so called "Aggregrate mode" is supported. The code has been implemented in a manner that makes extension to other older Alpha CPUs relatively straightforward should some mug wish to indulge themselves. Signed-off-by:
Michael Cree <mcree@orcon.net.nz> Cc: Richard Henderson <rth@twiddle.net> Cc: Ivan Kokshaysky <ink@jurassic.park.msu.ru> Cc: Matt Turner <mattst88@gmail.com> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Ingo Molnar <mingo@elte.hu> Cc: Jay Estabrook <jay.estabrook@hp.com> Signed-off-by:
-
Michael Cree authored
Signed-off-by:
-
Michael Cree authored
The following patches implement hardware performance events for the Alpha EV67 and later CPUs. I have had this running on a Compaq XP1000 (EV67, single CPU) for a few days now. Pretty cool -- discovered that the glibc exp2() library routine uses on average 985 cycles to execute 777 CPU instructions whereas Compaq's CPML library version of exp2() uses on average 32 cycles to execute 47 CPU instructions to achieve the same thing! This patch: Add performance monitor interrupt counternd and export the count to user space via /proc/interrupts. Signed-off-by:
-
Jan Kara authored
WB_SYNC_NONE writeback is done in rounds of 1024 pages so that we don't write out some huge inode for too long while starving writeout of other inodes. To avoid livelocks, we record time we started writeback in wbc->wb_start and do not write out inodes which were dirtied after this time. But currently, writeback_inodes_wb() resets wb_start each time it is called thus effectively invalidating this logic and making any WB_SYNC_NONE writeback prone to livelocks. This patch makes sure wb_start is set only once when we start writeback. Signed-off-by:
Jan Kara <jack@suse.cz> Reviewed-by:
Wu Fengguang <fengguang.wu@intel.com> Cc: Christoph Hellwig <hch@lst.de> Acked-by:
Jens Axboe <jaxboe@fusionio.com> Signed-off-by:
-
Lai Jiangshan authored
Use compile-allocated memory instead of dynamic allocated memory for mm_slots_hash. Use hash_ptr() instead divisions for bucket calculation. Signed-off-by:
Lai Jiangshan <laijs@cn.fujitsu.com> Signed-off-by:
Izik Eidus <ieidus@redhat.com> Cc: Avi Kivity <avi@redhat.com> Cc: Andrea Arcangeli <aarcange@redhat.com> Signed-off-by:
-
Wu Fengguang authored
Fix "system goes unresponsive under memory pressure and lots of dirty/writeback pages" bug. http://lkml.org/lkml/2010/4/4/86 In the above thread, Andreas Mohr described that Invoking any command locked up for minutes (note that I'm talking about attempted additional I/O to the _other_, _unaffected_ main system HDD - such as loading some shell binaries -, NOT the external SSD18M!!). This happens when the two conditions are both meet: - under memory pressure - writing heavily to a slow device OOM also happens in Andreas' system. The OOM trace shows that 3 processes are stuck in wait_on_page_writeback() in the direct reclaim path. One in do_fork() and the other two in unix_stream_sendmsg(). They are blocked on this condition: (sc->order && priority < DEF_PRIORITY - 2) which was introduced in commit 78dc583d (vmscan: low order lumpy reclaim also should use PAGEOUT_IO_SYNC) one year ago. That condition may be too permissive. In Andreas' case, 512MB/1024 = 512KB. If the direct reclaim for the order-1 fork() allocation runs into a range of 512KB hard-to-reclaim LRU pages, it will be stalled. It's a severe problem in three ways. Firstly, it can easily happen in daily desktop usage. vmscan priority can easily go below (DEF_PRIORITY - 2) on _local_ memory pressure. Even if the system has 50% globally reclaimable pages, it still has good opportunity to have 0.1% sized hard-to-reclaim ranges. For example, a simple dd can easily create a big range (up to 20%) of dirty pages in the LRU lists. And order-1 to order-3 allocations are more than common with SLUB. Try "grep -v '1 :' /proc/slabinfo" to get the list of high order slab caches. For example, the order-1 radix_tree_node slab cache may stall applications at swap-in time; the order-3 inode cache on most filesystems may stall applications when trying to read some file; the order-2 proc_inode_cache may stall applications when trying to open a /proc file. Secondly, once triggered, it will stall unrelated processes (not doing IO at all) in the system. This "one slow USB device stalls the whole system" avalanching effect is very bad. Thirdly, once stalled, the stall time could be intolerable long for the users. When there are 20MB queued writeback pages and USB 1.1 is writing them in 1MB/s, wait_on_page_writeback() will stuck for up to 20 seconds. Not to mention it may be called multiple times. So raise the bar to only enable PAGEOUT_IO_SYNC when priority goes below DEF_PRIORITY/3, or 6.25% LRU size. As the default dirty throttle ratio is 20%, it will hardly be triggered by pure dirty pages. We'd better treat PAGEOUT_IO_SYNC as some last resort workaround -- its stall time is so uncomfortably long (easily goes beyond 1s). The bar is only raised for (order < PAGE_ALLOC_COSTLY_ORDER) allocations, which are easy to satisfy in 1TB memory boxes. So, although 6.25% of memory could be an awful lot of pages to scan on a system with 1TB of memory, it won't really have to busy scan that much. Andreas tested an older version of this patch and reported that it mostly fixed his problem. Mel Gorman helped improve it and KOSAKI Motohiro will fix it further in the next patch. Reported-by:
Andreas Mohr <andi@lisas.de> Reviewed-by:
Minchan Kim <minchan.kim@gmail.com> Reviewed-by:
KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> Signed-off-by:
Mel Gorman <mel@csn.ul.ie> Signed-off-by:
-
Kulikov Vasiliy authored
kmalloc() may fail, if so return -ENOMEM. Signed-off-by:
Kulikov Vasiliy <segooon@gmail.com> Acked-by:
Pekka Enberg <penberg@cs.helsinki.fi> Signed-off-by:
-
KOSAKI Motohiro authored
Memcg also need to trace page isolation information as global reclaim. This patch does it. Signed-off-by:
Balbir Singh <balbir@linux.vnet.ibm.com> Signed-off-by:
-
KOSAKI Motohiro authored
Mel Gorman recently added some vmscan tracepoints. Unfortunately they are covered only global reclaim. But we want to trace memcg reclaim too. Thus, this patch convert them to DEFINE_TRACE macro. it help to reuse tracepoint definition for other similar usage (i.e. memcg). This patch have no functionally change. Signed-off-by:
-
KOSAKI Motohiro authored
Memcg also need to trace reclaim progress as direct reclaim. This patch add it. Signed-off-by:
-
KOSAKI Motohiro authored
Mel Gorman recently added some vmscan tracepoints. Unfortunately they are covered only global reclaim. But we want to trace memcg reclaim too. Thus, this patch convert them to DEFINE_TRACE macro. it help to reuse tracepoint definition for other similar usage (i.e. memcg). This patch have no functionally change. Signed-off-by:
-
KOSAKI Motohiro authored
Presently shrink_slab() has the following scanning equation. lru_scanned max_pass basic_scan_objects = 4 x ------------- x ----------------------------- lru_pages shrinker->seeks (default:2) scan_objects = min(basic_scan_objects, max_pass * 2) If we pass very small value as lru_pages instead real number of lru pages, shrink_slab() drop much objects rather than necessary. And now, __zone_reclaim() pass 'order' as lru_pages by mistake. That produces a bad result. For example, if we receive very low memory pressure (scan = 32, order = 0), shrink_slab() via zone_reclaim() always drop _all_ icache/dcache objects. (see above equation, very small lru_pages make very big scan_objects result). This patch fixes it. [akpm@linux-foundation.org: fix layout, typos] Signed-off-by:Christoph Lameter <cl@linux-foundation.org> Acked-by:
Rik van Riel <riel@redhat.com> Cc: Johannes Weiner <hannes@cmpxchg.org> Signed-off-by:
-
Jeremy Fitzhardinge authored
It is not appropriate for apply_to_page_range() to directly call any mmu notifiers, because it is a general purpose function whose effect depends on what context it is called in and what the callback function does. In particular, if it is being used as part of an mmu notifier implementation, the recursive calls can be particularly problematic. It is up to apply_to_page_range's caller to do any notifier calls if necessary. It does not affect any in-tree users because they all operate on init_mm, and mmu notifiers only pertain to usermode mappings. [stefano.stabellini@eu.citrix.com: remove unused local `start'] Signed-off-by:
Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com> Signed-off-by:
Stefano Stabellini <stefano.stabellini@eu.citrix.com> Cc: Andrea Arcangeli <aarcange@redhat.com> Cc: Stefano Stabellini <stefano.stabellini@eu.citrix.com> Cc: Avi Kivity <avi@qumranet.com> Signed-off-by:
-
KOSAKI Motohiro authored
Rik van Riel pointed out reading reclaim_stat should be protected lru_lock, otherwise vmscan might sweep 2x much pages. This fault was introduced by commit 4f98a2fe Author: Rik van Riel <riel@redhat.com> Date: Sat Oct 18 20:26:32 2008 -0700 vmscan: split LRU lists into anon & file sets Signed-off-by:
-
KOSAKI Motohiro authored
'slab_reclaimable' and 'nr_pages' are unsigned. Subtraction is unsafe because negative results would be misinterpreted. Signed-off-by:
-
KOSAKI Motohiro authored
drivers/base/node.c: In function 'node_read_meminfo': drivers/base/node.c:139: warning: the frame size of 848 bytes is larger than 512 bytes Fix it by splitting the sprintf() into three parts. It has no functional change. Signed-off-by:
-
Andrea Arcangeli authored
Set the flag if do_swap_page is decowing the page the same way do_wp_page would too. Signed-off-by:
Andrea Arcangeli <aarcange@redhat.com> Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> Cc: Rik van Riel <riel@redhat.com> Cc: Nick Piggin <nickpiggin@yahoo.com.au> Cc: Hugh Dickins <hughd@google.com> Signed-off-by:
-
Rik van Riel authored
On swapin it is fairly common for a page to be owned exclusively by one process. In that case we want to add the page to the anon_vma of that process's VMA, instead of to the root anon_vma. This will reduce the amount of rmap searching that the swapout code needs to do. Signed-off-by:
-
David Rientjes authored
/proc/pid/oom_adj is now deprecated so that that it may eventually be removed. The target date for removal is August 2012. A warning will be printed to the kernel log if a task attempts to use this interface. Future warning will be suppressed until the kernel is rebooted to prevent spamming the kernel log. Signed-off-by:
David Rientjes <rientjes@google.com> Cc: Nick Piggin <npiggin@suse.de> Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> Cc: Oleg Nesterov <oleg@redhat.com> Cc: Balbir Singh <balbir@in.ibm.com> Signed-off-by:
-
David Rientjes authored
This a complete rewrite of the oom killer's badness() heuristic which is used to determine which task to kill in oom conditions. The goal is to make it as simple and predictable as possible so the results are better understood and we end up killing the task which will lead to the most memory freeing while still respecting the fine-tuning from userspace. Instead of basing the heuristic on mm->total_vm for each task, the task's rss and swap space is used instead. This is a better indication of the amount of memory that will be freeable if the oom killed task is chosen and subsequently exits. This helps specifically in cases where KDE or GNOME is chosen for oom kill on desktop systems instead of a memory hogging task. The baseline for the heuristic is a proportion of memory that each task is currently using in memory plus swap compared to the amount of "allowable" memory. "Allowable," in this sense, means the system-wide resources for unconstrained oom conditions, the set of mempolicy nodes, the mems attached to current's cpuset, or a memory controller's limit. The proportion is given on a scale of 0 (never kill) to 1000 (always kill), roughly meaning that if a task has a badness() score of 500 that the task consumes approximately 50% of allowable memory resident in RAM or in swap space. The proportion is always relative to the amount of "allowable" memory and not the total amount of RAM systemwide so that mempolicies and cpusets may operate in isolation; they shall not need to know the true size of the machine on which they are running if they are bound to a specific set of nodes or mems, respectively. Root tasks are given 3% extra memory just like __vm_enough_memory() provides in LSMs. In the event of two tasks consuming similar amounts of memory, it is generally better to save root's task. Because of the change in the badness() heuristic's baseline, it is also necessary to introduce a new user interface to tune it. It's not possible to redefine the meaning of /proc/pid/oom_adj with a new scale since the ABI cannot be changed for backward compatability. Instead, a new tunable, /proc/pid/oom_score_adj, is added that ranges from -1000 to +1000. It may be used to polarize the heuristic such that certain tasks are never considered for oom kill while others may always be considered. The value is added directly into the badness() score so a value of -500, for example, means to discount 50% of its memory consumption in comparison to other tasks either on the system, bound to the mempolicy, in the cpuset, or sharing the same memory controller. /proc/pid/oom_adj is changed so that its meaning is rescaled into the units used by /proc/pid/oom_score_adj, and vice versa. Changing one of these per-task tunables will rescale the value of the other to an equivalent meaning. Although /proc/pid/oom_adj was originally defined as a bitshift on the badness score, it now shares the same linear growth as /proc/pid/oom_score_adj but with different granularity. This is required so the ABI is not broken with userspace applications and allows oom_adj to be deprecated for future removal. Signed-off-by:
-
Andrew Morton authored
Cc: Minchan Kim <minchan.kim@gmail.com> Cc: David Rientjes <rientjes@google.com> Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> Cc: Oleg Nesterov <oleg@redhat.com> Signed-off-by:
-
KOSAKI Motohiro authored
Oleg pointed out current PF_EXITING check is wrong. Because PF_EXITING is per-thread flag, not per-process flag. He said, Two threads, group-leader L and its sub-thread T. T dumps the code. In this case both threads have ->mm != NULL, L has PF_EXITING. The first problem is, select_bad_process() always return -1 in this case (even if the caller is T, this doesn't matter). The second problem is that we should add TIF_MEMDIE to T, not L. I think we can remove this dubious PF_EXITING check. but as first step, This patch add the protection of multi threaded issue. Signed-off-by:
-
Luis Claudio R. Goncalves authored
In a system under heavy load it was observed that even after the oom-killer selects a task to die, the task may take a long time to die. Right after sending a SIGKILL to the task selected by the oom-killer this task has its priority increased so that it can exit() soon, freeing memory. That is accomplished by: /* * We give our sacrificial lamb high priority and access to * all the memory it needs. That way it should be able to * exit() and clear out its resources quickly... */ p->rt.time_slice = HZ; set_tsk_thread_flag(p, TIF_MEMDIE); It sounds plausible giving the dying task an even higher priority to be sure it will be scheduled sooner and free the desired memory. It was suggested on LKML using SCHED_FIFO:1, the lowest RT priority so that this task won't interfere with any running RT task. If the dying task is already an RT task, leave it untouched. Another good suggestion, implemented here, was to avoid boosting the dying task priority in case of mem_cgroup OOM. Signed-off-by:Luis Claudio R. Goncalves <lclaudio@uudg.org> Signed-off-by:
-
KOSAKI Motohiro authored
The current "child->mm == p->mm" check prevents selection of vfork()ed task. But we don't have any reason to don't consider vfork(). Removed. Signed-off-by:
-
KOSAKI Motohiro authored
presently has_intersects_mems_allowed() has own thread iterate logic, but it should use while_each_thread(). It slightly improve the code readability. Signed-off-by:
-
KOSAKI Motohiro authored
Presently if oom_kill_allocating_task is enabled and current have OOM_DISABLED, following printk in oom_kill_process is called twice. pr_err("%s: Kill process %d (%s) score %lu or sacrifice child\n", message, task_pid_nr(p), p->comm, points); So, OOM_DISABLE check should be more early. Signed-off-by: -
KOSAKI Motohiro authored
select_bad_process() and badness() have the same OOM_DISABLE check. This patch kills one. Signed-off-by:
-
KOSAKI Motohiro authored
If a kernel thread is using use_mm(), badness() returns a positive value. This is not a big issue because caller take care of it correctly. But there is one exception, /proc/<pid>/oom_score calls badness() directly and doesn't care that the task is a regular process. Another example, /proc/1/oom_score return !0 value. But it's unkillable. This incorrectness makes administration a little confusing. This patch fixes it. Signed-off-by:
-
KOSAKI Motohiro authored
When oom_kill_allocating_task is enabled, an argument task of oom_kill_process is not selected by select_bad_process(), It's just out_of_memory() caller task. It mean the task can be unkillable. check it first. Signed-off-by:
-
KOSAKI Motohiro authored
Presently we have the same task check in two places. Unify it. Signed-off-by:
-
KOSAKI Motohiro authored
Presently select_bad_process() has a PF_KTHREAD check, but oom_kill_process doesn't. It mean oom_kill_process() may choose wrong task, especially, when the child are using use_mm(). Signed-off-by:
-
KOSAKI Motohiro authored
Presently, badness() doesn't care about either CPUSET nor mempolicy. Then if the victim child process have disjoint nodemask, OOM Killer might kill innocent process. This patch fixes it. [akpm@linux-foundation.org: coding-style fixes] Signed-off-by:
-
Mel Gorman authored
When shrink_inactive_list() isolates pages, it updates a number of counters using temporary variables to gather them. These consume stack and it's in the main path that calls ->writepage(). This patch moves the accounting updates outside of the main path to reduce stack usage. Signed-off-by:
Johannes Weiner <hannes@cmpxchg.org> Acked-by:
-
Mel Gorman authored
shrink_page_list() sets up a pagevec to release pages as according as they are free. It uses significant amounts of stack on the pagevec. This patch adds pages to be freed via pagevec to a linked list which is then freed en-masse at the end. This avoids using stack in the main path that potentially calls writepage(). Signed-off-by:
-
Mel Gorman authored
shrink_inactive_list() sets up a pagevec to release unfreeable pages. It uses significant amounts of stack doing this. This patch splits shrink_inactive_list() to take the stack usage out of the main path so that callers to writepage() do not contain an unused pagevec on the stack. Signed-off-by:
-
Mel Gorman authored
Remove temporary variable that is only used once and does not help clarify code. [akpm@linux-foundation.org: coding-style fixes] Signed-off-by:
-
KOSAKI Motohiro authored
Now, max_scan of shrink_inactive_list() is always passed less than SWAP_CLUSTER_MAX. then, we can remove scanning pages loop in it. This patch also help stack diet. detail - remove "while (nr_scanned < max_scan)" loop - remove nr_freed (now, we use nr_reclaimed directly) - remove nr_scan (now, we use nr_scanned directly) - rename max_scan to nr_to_scan - pass nr_to_scan into isolate_pages() directly instead using SWAP_CLUSTER_MAX [akpm@linux-foundation.org: coding-style fixes] Signed-off-by:
-
KOSAKI Motohiro authored
Since 2.6.28 zone->prev_priority is unused. Then it can be removed safely. It reduce stack usage slightly. Now I have to say that I'm sorry. 2 years ago, I thought prev_priority can be integrate again, it's useful. but four (or more) times trying haven't got good performance number. Thus I give up such approach. The rest of this changelog is notes on prev_priority and why it existed in the first place and why it might be not necessary any more. This information is based heavily on discussions between Andrew Morton, Rik van Riel and Kosaki Motohiro who is heavily quotes from. Historically prev_priority was important because it determined when the VM would start unmapping PTE pages. i.e. there are no balances of note within the VM, Anon vs File and Mapped vs Unmapped. Without prev_priority, there is a potential risk of unnecessarily increasing minor faults as a large amount of read activity of use-once pages could push mapped pages to the end of the LRU and get unmapped. There is no proof this is still a problem but currently it is not considered to be. Active files are not deactivated if the active file list is smaller than the inactive list reducing the liklihood that file-mapped pages are being pushed off the LRU and referenced executable pages are kept on the active list to avoid them getting pushed out by read activity. Even if it is a problem, prev_priority prev_priority wouldn't works nowadays. First of all, current vmscan still a lot of UP centric code. it expose some weakness on some dozens CPUs machine. I think we need more and more improvement. The problem is, current vmscan mix up per-system-pressure, per-zone-pressure and per-task-pressure a bit. example, prev_priority try to boost priority to other concurrent priority. but if the another task have mempolicy restriction, it is unnecessary, but also makes wrong big latency and exceeding reclaim. per-task based priority + prev_priority adjustment make the emulation of per-system pressure. but it have two issue 1) too rough and brutal emulation 2) we need per-zone pressure, not per-system. Another example, currently DEF_PRIORITY is 12. it mean the lru rotate about 2 cycle (1/4096 + 1/2048 + 1/1024 + .. + 1) before invoking OOM-Killer. but if 10,0000 thrreads enter DEF_PRIORITY reclaim at the same time, the system have higher memory pressure than priority==0 (1/4096*10,000 > 2). prev_priority can't solve such multithreads workload issue. In other word, prev_priority concept assume the sysmtem don't have lots threads." Signed-off-by:
-
