From 3b2ba7b31d56c3d8f57cd5d32b8fb5101ab446e4 Mon Sep 17 00:00:00 2001 From: Filipe Manana Date: Thu, 5 Nov 2015 10:51:37 +0000 Subject: [PATCH] Btrfs: fix sleeping inside atomic context in qgroup rescan worker We are holding a btree path with spinning locks and then we attempt to clone an extent buffer, which calls kmem_cache_alloc() and this function can sleep, causing the following trace to be reported on a debug kernel: [107118.218536] BUG: sleeping function called from invalid context at mm/slab.c:2871 [107118.224110] in_atomic(): 1, irqs_disabled(): 0, pid: 19148, name: kworker/u32:3 [107118.226120] INFO: lockdep is turned off. [107118.226843] Preemption disabled at:[] btrfs_clear_lock_blocking_rw+0x96/0xea [btrfs] [107118.229175] CPU: 3 PID: 19148 Comm: kworker/u32:3 Tainted: G W 4.3.0-rc5-btrfs-next-17+ #1 [107118.231326] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.8.1-0-g4adadbd-20150316_085822-nilsson.home.kraxel.org 04/01/2014 [107118.233687] Workqueue: btrfs-qgroup-rescan btrfs_qgroup_rescan_helper [btrfs] [107118.236835] 0000000000000000 ffff880424bf3b78 ffffffff812566f4 0000000000000000 [107118.238369] ffff880424bf3ba0 ffffffff81070664 ffffffff817f1cd5 0000000000000b37 [107118.239769] 0000000000000000 ffff880424bf3bc8 ffffffff8107070a 0000000000008850 [107118.241244] Call Trace: [107118.241729] [] dump_stack+0x4e/0x79 [107118.242602] [] ___might_sleep+0x23a/0x241 [107118.243586] [] __might_sleep+0x9f/0xa6 [107118.244532] [] cache_alloc_debugcheck_before+0x25/0x36 [107118.245939] [] kmem_cache_alloc+0x50/0x215 [107118.246930] [] __alloc_extent_buffer+0x2a/0x11f [btrfs] [107118.248121] [] btrfs_clone_extent_buffer+0x3d/0xdd [btrfs] [107118.249451] [] btrfs_qgroup_rescan_worker+0x16d/0x434 [btrfs] [107118.250755] [] ? arch_local_irq_save+0x9/0xc [107118.251754] [] normal_work_helper+0x14c/0x32a [btrfs] [107118.252899] [] ? normal_work_helper+0x14c/0x32a [btrfs] [107118.254195] [] btrfs_qgroup_rescan_helper+0x12/0x14 [btrfs] [107118.255436] [] process_one_work+0x24a/0x4ac [107118.263690] [] worker_thread+0x206/0x2c2 [107118.264888] [] ? rescuer_thread+0x2cb/0x2cb [107118.267413] [] kthread+0xef/0xf7 [107118.268417] [] ? kthread_parkme+0x24/0x24 [107118.269505] [] ret_from_fork+0x3f/0x70 [107118.270491] [] ? kthread_parkme+0x24/0x24 So just use blocking locks for our path to solve this. This fixes the patch titled: "btrfs: qgroup: Don't copy extent buffer to do qgroup rescan" Signed-off-by: Filipe Manana --- fs/btrfs/qgroup.c | 1 - 1 file changed, 1 deletion(-) diff --git a/fs/btrfs/qgroup.c b/fs/btrfs/qgroup.c index 75bb4af98a93..93e12c18ffd7 100644 --- a/fs/btrfs/qgroup.c +++ b/fs/btrfs/qgroup.c @@ -2198,7 +2198,6 @@ qgroup_rescan_leaf(struct btrfs_fs_info *fs_info, struct btrfs_path *path, int slot; int ret; - path->leave_spinning = 1; mutex_lock(&fs_info->qgroup_rescan_lock); ret = btrfs_search_slot_for_read(fs_info->extent_root, &fs_info->qgroup_rescan_progress, -- 2.34.1