- 11 Sep, 2013 40 commits
-
-
Shaohua Li authored
The previous patch can expose races, according to Hugh: swapoff was sometimes failing with "Cannot allocate memory", coming from try_to_unuse()'s -ENOMEM: it needs to allow for swap_duplicate() failing on a free entry temporarily SWAP_MAP_BAD while being discarded. We should use ACCESS_ONCE() there, and whenever accessing swap_map locklessly; but rather than peppering it throughout try_to_unuse(), just declare *swap_map with volatile. try_to_unuse() is accustomed to *swap_map going down racily, but not necessarily to it jumping up from 0 to SWAP_MAP_BAD: we'll be safer to prevent that transition once SWP_WRITEOK is switched off, when it's a waste of time to issue discards anyway (swapon can do a whole discard). Another issue is: In swapin_readahead(), read_swap_cache_async() can read a bad swap entry, because we don't check if readahead swap entry is bad. This doesn't break anything but such swapin page is wasteful and can only be freed at page reclaim. We should avoid read such swap entry. And in discard, we mark swap entry SWAP_MAP_BAD and then switch it to normal when discard is finished. If readahead reads such swap entry, we have the same issue, so we much check if swap entry is bad too. Thanks Hugh to inspire swapin_readahead could use bad swap entry. [include Hugh's patch 'swap: fix swapoff ENOMEMs from discard'] Signed-off-by:
Shaohua Li <shli@fusionio.com> Signed-off-by:
Hugh Dickins <hughd@google.com> Cc: Rik van Riel <riel@redhat.com> Cc: Minchan Kim <minchan@kernel.org> Cc: Kyungmin Park <kmpark@infradead.org> Cc: Rafael Aquini <aquini@redhat.com> Signed-off-by:
Andrew Morton <akpm@linux-foundation.org> Signed-off-by:
Linus Torvalds <torvalds@linux-foundation.org>
-
Shaohua Li authored
swap can do cluster discard for SSD, which is good, but there are some problems here: 1. swap do the discard just before page reclaim gets a swap entry and writes the disk sectors. This is useless for high end SSD, because an overwrite to a sector implies a discard to original sector too. A discard + overwrite == overwrite. 2. the purpose of doing discard is to improve SSD firmware garbage collection. Idealy we should send discard as early as possible, so firmware can do something smart. Sending discard just after swap entry is freed is considered early compared to sending discard before write. Of course, if workload is already bound to gc speed, sending discard earlier or later doesn't make 3. block discard is a sync API, which will delay scan_swap_map() significantly. 4. Write and discard command can be executed parallel in PCIe SSD. Making swap discard async can make execution more efficiently. This patch makes swap discard async and moves discard to where swap entry is freed. Discard and write have no dependence now, so above issues can be avoided. Idealy we should do discard for any freed sectors, but some SSD discard is very slow. This patch still does discard for a whole cluster. My test does a several round of 'mmap, write, unmap', which will trigger a lot of swap discard. In a fusionio card, with this patch, the test runtime is reduced to 18% of the time without it, so around 5.5x faster. [akpm@linux-foundation.org: coding-style fixes] Signed-off-by:
-
Shaohua Li authored
I'm using a fast SSD to do swap. scan_swap_map() sometimes uses up to 20~30% CPU time (when cluster is hard to find, the CPU time can be up to 80%), which becomes a bottleneck. scan_swap_map() scans a byte array to search a 256 page cluster, which is very slow. Here I introduced a simple algorithm to search cluster. Since we only care about 256 pages cluster, we can just use a counter to track if a cluster is free. Every 256 pages use one int to store the counter. If the counter of a cluster is 0, the cluster is free. All free clusters will be added to a list, so searching cluster is very efficient. With this, scap_swap_map() overhead disappears. This might help low end SD card swap too. Because if the cluster is aligned, SD firmware can do flash erase more efficiently. We only enable the algorithm for SSD. Hard disk swap isn't fast enough and has downside with the algorithm which might introduce regression (see below). The patch slightly changes which cluster is choosen. It always adds free cluster to list tail. This can help wear leveling for low end SSD too. And if no cluster found, the scan_swap_map() will do search from the end of last cluster. So if no cluster found, the scan_swap_map() will do search from the end of last free cluster, which is random. For SSD, this isn't a problem at all. Another downside is the cluster must be aligned to 256 pages, which will reduce the chance to find a cluster. I would expect this isn't a big problem for SSD because of the non-seek penality. (And this is the reason I only enable the algorithm for SSD). Signed-off-by:
-
Chen Gang authored
set_pageblock_order() may be called when memory hotplug, so need use '__paginginit' instead of '__init'. The related warning: The function __meminit .free_area_init_node() references a function __init .set_pageblock_order(). If .set_pageblock_order is only used by .free_area_init_node then annotate .set_pageblock_order with a matching annotation. Signed-off-by:
Chen Gang <gang.chen@asianux.com> Signed-off-by:
-
Jerry Zhou authored
When PAGE_SHIFT > 20, the result of "20 - PAGE_SHIFT" is negative. The previous calculating here will generate an unexpected result. In addition, if PAGE_SIZE >= 1MB, The memory size of "numentries" was already integral multiple of 1MB. Signed-off-by:
Jerry Zhou <uulinux@gmail.com> Signed-off-by:
-
Jingoo Han authored
The use of strict_strtoul() is not preferred, because strict_strtoul() is obsolete. Thus, kstrtoul() should be used. Signed-off-by:
Jingoo Han <jg1.han@samsung.com> Signed-off-by:
-
Dave Hansen authored
The previous patch doing vmstats for TLB flushes ("mm: vmstats: tlb flush counters") effectively missed UP since arch/x86/mm/tlb.c is only compiled for SMP. UP systems do not do remote TLB flushes, so compile those counters out on UP. arch/x86/kernel/cpu/mtrr/generic.c calls __flush_tlb() directly. This is probably an optimization since both the mtrr code and __flush_tlb() write cr4. It would probably be safe to make that a flush_tlb_all() (and then get these statistics), but the mtrr code is ancient and I'm hesitant to touch it other than to just stick in the counters. [akpm@linux-foundation.org: tweak comments] Signed-off-by:Dave Hansen <dave.hansen@linux.intel.com> Cc: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Ingo Molnar <mingo@elte.hu> Cc: "H. Peter Anvin" <hpa@zytor.com> Cc: Thomas Gleixner <tglx@linutronix.de> Signed-off-by:
-
Dave Hansen authored
I was investigating some TLB flush scaling issues and realized that we do not have any good methods for figuring out how many TLB flushes we are doing. It would be nice to be able to do these in generic code, but the arch-independent calls don't explicitly specify whether we actually need to do remote flushes or not. In the end, we really need to know if we actually _did_ global vs. local invalidations, so that leaves us with few options other than to muck with the counters from arch-specific code. Signed-off-by:
-
Sunghan Suh authored
There is a proper macro to get the corresponding swapper address space from a swap entry. Instead of directly accessing "swapper_spaces" array, use the "swap_address_space" macro. Signed-off-by:
Sunghan Suh <sunghan.suh@samsung.com> Reviewed-by:
Bob Liu <bob.liu@oracle.com> Reviewed-by:
Wanpeng Li <liwanp@linux.vnet.ibm.com> Acked-by:
Seth Jennings <sjenning@linux.vnet.ibm.com> Signed-off-by:
-
Oleg Nesterov authored
correct_wcount and inode in mmap_region() just complicate the code. This boolean was needed previously, when deny_write_access() was called before vma_merge(), now we can simply check VM_DENYWRITE and do allow_write_access() if it is set. allow_write_access() checks file != NULL, so this is safe even if it was possible to use VM_DENYWRITE && !file. Just we need to ensure we use the same file which was deny_write_access()'ed, so the patch also moves "file = vma->vm_file" down after allow_write_access(). Signed-off-by:
Oleg Nesterov <oleg@redhat.com> Cc: Hugh Dickins <hughd@google.com> Cc: Al Viro <viro@zeniv.linux.org.uk> Cc: Colin Cross <ccross@android.com> Cc: David Rientjes <rientjes@google.com> Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> Signed-off-by:
-
Oleg Nesterov authored
Simple cleanup. Move "struct inode *inode" variable into "if (file)" block to simplify the code and avoid the unnecessary check. Signed-off-by:
-
Oleg Nesterov authored
mmap() doesn't allow the non-anonymous mappings with VM_GROWS* bit set. In particular this means that mmap_region()->vma_merge(file, vm_flags) must always fail if "vm_flags & VM_GROWS" is set incorrectly. So it does not make sense to check VM_GROWS* after we already allocated the new vma, the only caller, do_mmap_pgoff(), which can pass this flag can do the check itself. And this looks a bit more correct, mmap_region() already unmapped the old mapping at this stage. But if mmap() is going to fail, it should avoid do_munmap() if possible. Note: we check VM_GROWS at the end to ensure that do_mmap_pgoff() won't return EINVAL in the case when it currently returns another error code. Many thanks to Hugh who nacked the buggy v1. Signed-off-by:
-
Andrew Morton authored
A few 80-col gymnastics were cleaned up as a result. Cc: Hugh Dickins <hughd@google.com> Signed-off-by:
-
Raymond Jennings authored
It is possible to swapon a swap area that is too big for the pte width to handle. Presently this failure happens silently. Instead, emit a diagnostic to warn the user. Testing results, root prompt commands and kernel log messages: # lvresize /dev/system/swap --size 16G # mkswap /dev/system/swap # swapon /dev/system/swap Jul 7 04:27:22 warfang kernel: Adding 16777212k swap on /dev/mapper/system-swap. Priority:-1 extents:1 across:16777212k # lvresize /dev/system/swap --size 64G # mkswap /dev/system/swap # swapon /dev/system/swap Jul 7 04:27:22 warfang kernel: Truncating oversized swap area, only using 33554432k out of 67108860k Jul 7 04:27:22 warfang kernel: Adding 33554428k swap on /dev/mapper/system-swap. Priority:-1 extents:1 across:33554428k [akpm@linux-foundation.org: fix warning] Signed-off-by:
Raymond Jennings <shentino@gmail.com> Acked-by:
Valdis Kletnieks <valdis.kletnieks@vt.edu> Reviewed-by:
Rik van Riel <riel@redhat.com> Cc: Hugh Dickins <hughd@google.com> Signed-off-by:
-
Vladimir Cernov authored
This fixes following errors: - ERROR: "(foo*)" should be "(foo *)" - ERROR: "foo ** bar" should be "foo **bar" Signed-off-by:
Vladimir Cernov <gg.kaspersky@gmail.com> Reviewed-by:
Pekka Enberg <penberg@kernel.org> Acked-by:
David Rientjes <rientjes@google.com> Signed-off-by:
-
Oleg Nesterov authored
Simple cleanup. Every user of vma_set_policy() does the same work, this looks a bit annoying imho. And the new trivial helper which does mpol_dup() + vma_set_policy() to simplify the callers. Signed-off-by:
-
Jingoo Han authored
The driver core clears the driver data to NULL after device_release or on probe failure. Thus, it is not needed to manually clear the device driver data to NULL. Signed-off-by:
-
Mike Miller authored
At one time we used to set the maximum number of scatter gather elements on all Smart Array controllers to 32. At some point in time the firmware began to write the "appropriate" value for each controller into the config table. The cciss driver would then read that and set h->maxsgentries. h->maxsgentries = readl(&(h->cfgtable->MaxSGElements); On the P600 that value is 544. Under some workloads a significant performance reduction may result. This patch forces the P600 to use only 32 scatter gather elements. Other controllers are not affected. Signed-off-by:Mike Miller <mike.miller@hp.com> Signed-off-by:
Dwight (Bud) Brown <bubrown@redhat.com> Signed-off-by:
Tomas Henzl <thenzl@redhat.com> Acked-by:
Stephen M. Cameron <steve.cameron@hp.com> Cc: Jens Axboe <axboe@kernel.dk> Signed-off-by:
-
Jingoo Han authored
mg_times_out() is used only in this file. Fix the following sparse warning: drivers/block/mg_disk.c:639:6: warning: symbol 'mg_times_out' was not declared. Should it be static? Signed-off-by:
-
Cai Zhiyong authored
Read block device partition table from command line. The partition used for fixed block device (eMMC) embedded device. It is no MBR, save storage space. Bootloader can be easily accessed by absolute address of data on the block device. Users can easily change the partition. This code reference MTD partition, source "drivers/mtd/cmdlinepart.c" About the partition verbose reference "Documentation/block/cmdline-partition.txt" [akpm@linux-foundation.org: fix printk text] [yongjun_wei@trendmicro.com.cn: fix error return code in parse_parts()] Signed-off-by:
Cai Zhiyong <caizhiyong@huawei.com> Cc: Karel Zak <kzak@redhat.com> Cc: "Wanglin (Albert)" <albert.wanglin@huawei.com> Cc: Marius Groeger <mag@sysgo.de> Cc: David Woodhouse <dwmw2@infradead.org> Cc: Jens Axboe <axboe@kernel.dk> Cc: Brian Norris <computersforpeace@gmail.com> Cc: Artem Bityutskiy <dedekind@infradead.org> Signed-off-by:
Wei Yongjun <yongjun_wei@trendmicro.com.cn> Signed-off-by:
-
Jingoo Han authored
The usage of strict_strtoul() is not preferred, because strict_strtoul() is obsolete. Thus, kstrtoul() should be used. Signed-off-by:
-
Jingoo Han authored
The use of strict_strtoul() is not preferred, because strict_strtoul() is obsolete. Thus, kstrtoul() should be used. Signed-off-by:
-
Oleg Nesterov authored
task_struct->pid/tgid should go away. 1. Change same_thread_group() to use task->signal for comparison. 2. Change has_group_leader_pid(task) to compare task_pid(task) with signal->leader_pid. Signed-off-by:
"Eric W. Biederman" <ebiederm@xmission.com> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Ingo Molnar <mingo@elte.hu> Cc: Peter Zijlstra <a.p.zijlstra@chello.nl> Signed-off-by:
-
Jie Liu authored
Call fiemap ioctl(2) with given start offset as well as an desired mapping range should show extents if possible. However, we somehow figure out the end offset of mapping via 'mapping_end -= cpos' before iterating the extent records which would cause problems if the given fiemap length is too small to a cluster size, e.g, Cluster size 4096: debugfs.ocfs2 1.6.3 Block Size Bits: 12 Cluster Size Bits: 12 The extended fiemap test utility From David: https://gist.github.com/anonymous/6172331 # dd if=/dev/urandom of=/ocfs2/test_file bs=1M count=1000 # ./fiemap /ocfs2/test_file 4096 10 start: 4096, length: 10 File /ocfs2/test_file has 0 extents: # Logical Physical Length Flags ^^^^^ <-- No extent is shown In this case, at ocfs2_fiemap(): cpos == mapping_end == 1. Hence the loop of searching extent records was not executed at all. This patch remove the in question 'mapping_end -= cpos', and loops until the cpos is larger than the mapping_end as usual. # ./fiemap /ocfs2/test_file 4096 10 start: 4096, length: 10 File /ocfs2/test_file has 1 extents: # Logical Physical Length Flags 0: 0000000000000000 0000000056a01000 0000000006a00000 0000 Signed-off-by:Jie Liu <jeff.liu@oracle.com> Reported-by:
David Weber <wb@munzinger.de> Tested-by:
-
Joseph Qi authored
Variable ip in dlmfs_get_root_inode() is defined but not used. So clean it up. Signed-off-by:
Joseph Qi <joseph.qi@huawei.com> Reviewed-by:
-
Joyce authored
In o2hb_shutdown_slot() and o2hb_check_slot(), since event is defined as local, it is only valid during the call stack. So the following tiny race case may happen in a multi-volumes mounted environment: o2hb-vol1 o2hb-vol2 1) o2hb_shutdown_slot allocate local event1 2) queue_node_event add event1 to global o2hb_node_events 3) o2hb_shutdown_slot allocate local event2 4) queue_node_event add event2 to global o2hb_node_events 5) o2hb_run_event_list delete event1 from o2hb_node_events 6) o2hb_run_event_list event1 empty, return 7) o2hb_shutdown_slot event1 lifecycle ends 8) o2hb_fire_callbacks event1 is already *invalid* This patch lets it wait on o2hb_callback_sem when another thread is firing callbacks. And for performance consideration, we only call o2hb_run_event_list when there is an event queued. Signed-off-by:Joyce <xuejiufei@huawei.com> Signed-off-by:
-
Joseph Qi authored
Since o2nm_get_node_by_num() may return NULL, we add this check in o2net_accept_one() to avoid possible NULL pointer dereference. Signed-off-by:
-
Joseph Qi authored
Code in o2net_handler_tree_lookup() may be corrupted by mistake. So adjust it to promote readability. Signed-off-by:
-
Younger Liu authored
In ocfs2_remove_inode_range(), there is a memory leak. The variable path has allocated memory with ocfs2_new_path_from_et(), but it is not free. Signed-off-by:
Younger Liu <younger.liu@huawei.com> Reviewed-by:
-
Joseph Qi authored
In ocfs2_reflink_xattr_rec(), meta_ac and data_ac are allocated by calling ocfs2_lock_reflink_xattr_rec_allocators(). Once an error occurs when allocating *data_ac, it frees *meta_ac which is allocated before. Here it mistakenly sets meta_ac to NULL but *meta_ac. Then ocfs2_reflink_xattr_rec() will try to free meta_ac again which is already invalid. Signed-off-by:
-
Xue jiufei authored
dlm_do_local_recovery_cleanup() should force clean refmap if the owner of lockres is UNKNOWN. Otherwise node may hang when umounting filesystems. Here's the situation: Node1 Node2 dlmlock() -> dlm_get_lock_resource() send DLM_MASTER_REQUEST_MSG to other nodes. trying to master this lockres, return MAYBE. selected as the master of lockresA, set mle->master to Node1, and do assert_master, send DLM_ASSERT_MASTER_MSG to Node2. Node 2 has interest on lockresA and return DLM_ASSERT_RESPONSE_MASTERY_REF then something happened and Node2 crashed. Receiving DLM_ASSERT_RESPONSE_MASTERY_REF, set Node2 into refmap, and keep sending DLM_ASSERT_MASTER_MSG to other nodes o2hb found node2 down, calling dlm_hb_node_down() --> dlm_do_local_recovery_cleanup() the master of lockresA is still UNKNOWN, no need to call dlm_free_dead_locks(). Set the master of lockresA to Node1, but Node2 stills remains in refmap. When Node1 umount, it found that the refmap of lockresA is not empty and attempted to migrate it to Node2, But Node2 is already down, so umount hang, trying to migrate lockresA again and again. Signed-off-by: -
Younger Liu authored
In ocfs2_xattr_set(), if ocfs2_start_trans failed, meta_ac and data_ac should be free. Otherwise, It would lead to a memory leak. Signed-off-by:
-
Joseph Qi authored
In ocfs2_xattr_value_attach_refcount(), if error occurs when calling ocfs2_xattr_get_clusters(), it will go with unexpected behavior since local variables p_cluster, num_clusters and ext_flags are declared without initialization. Signed-off-by:
-
Jie Liu authored
The ocfs2 path is not properly freed which leads to a memory leak at __ocfs2_move_extents(). This patch stops the leaks of the ocfs2_path structure. Signed-off-by:
-
Joseph Qi authored
In ocfs2_attach_refcount_tree() and ocfs2_duplicate_extent_list(), if error occurs when calling ocfs2_get_clusters(), it will go with unexpected behavior as local variables p_cluster, num_clusters and ext_flags are declared without initialization. Signed-off-by:
-
Joseph Qi authored
In ocfs2_acl_from_xattr(), if size is less than sizeof(struct posix_acl_entry), it returns ERR_PTR(-EINVAL) directly. Then assign (size / sizeof(struct posix_acl_entry)) to count which will be at least 1, that means the following branch (count < 0) and (count == 0) will never be true. Signed-off-by:
Joel Becker <jlbec@evilplan.org> Signed-off-by:
-
Dong Fang authored
[dan.carpenter@oracle.com: fix up some NULL dereference bugs] Signed-off-by:
Dong Fang <yp.fangdong@gmail.com> Cc: Mark Fasheh <mfasheh@suse.com> Cc: Joel Becker <jlbec@evilplan.org> Cc: Jeff Liu <jeff.liu@oracle.com> Signed-off-by:
Dan Carpenter <dan.carpenter@oracle.com> Signed-off-by:
-
Sunil Mushran authored
Fix some possible null pointer dereferences that were detected by the static code analyser, smatch. Signed-off-by:
Sunil Mushran <sunil.mushran@oracle.com> Reported-by:
Dan Carpenter <error27@gmail.com> Reported-by:
Guozhonghua <guozhonghua@h3c.com> Cc: Sunil Mushran <sunil.mushran@gmail.com> Cc: Joseph Qi <joseph.qi@huawei.com> Cc: Mark Fasheh <mfasheh@suse.com> Cc: Joel Becker <jlbec@evilplan.org> Signed-off-by:
-
Younger Liu authored
There is an issue in reserving and claiming space for localalloc, When localalloc space is not enough, it would claim space from global_bitmap. And if there is not enough free space in global_bitmap, the size of claiming space would set to half of orignal size and retry. The issue is as follows: osb->local_alloc_bits is set to half of orignal size in ocfs2_recalc_la_window(), but ac->ac_bits_wanted is set to osb->local_alloc_default_bits which is not changed. localalloc always reserves and claims local_alloc_default_bits space and returns ENOSPC. So, ac->ac_bits_wanted should be osb->local_alloc_bits which would be changed. Signed-off-by:
-
Xue jiufei authored
dlm_request_all_locks() should deal with the status sent from target node if DLM_LOCK_REQUEST_MSG is sent successfully, or recovery master will fall into endless loop, waiting for other nodes to send locks and DLM_RECO_DATA_DONE_MSG to me. NodeA NodeB selected as recovery master dlm_remaster_locks() ->dlm_request_all_locks() send DLM_LOCK_REQUEST_MSG to nodeA It happened that NodeA cannot alloc memory when it processes this message. dlm_request_all_locks_handler() do not queue dlm_request_all_locks_worker and returns -ENOMEM. It will never send locks and DLM_RECO_DATA_DONE_MSG to NodeB. NodeB do not deal with the status sent from nodeA, and will fall in endless loop waiting for the recovery state of NodeA to be changed. Signed-off-by:
-
