From 48be0dfba19f35637877ab63f58d52b6d51e759c Mon Sep 17 00:00:00 2001 From: jdike <52420226+jdike@users.noreply.github.com> Date: Wed, 31 Jul 2019 17:53:39 -0400 Subject: lockdep false positive - move txg_kick() outside of ->dp_lock This fixes a lockdep warning by breaking a link between ->tx_sync_lock and ->dp_lock. The deadlock envisioned by lockdep is this: thread 1 holds db->db_mtx and tries to get dp->dp_lock: dsl_pool_dirty_space+0x70/0x2d0 [zfs] dbuf_dirty+0x778/0x31d0 [zfs] thread 2 holds bpo->bpo_lock and tries to get db->db_mtx: dmu_buf_will_dirty_impl dmu_buf_will_dirty+0x6b/0x6c0 [zfs] bpobj_iterate_impl+0xbe6/0x1410 [zfs] thread 3 holds tx->tx_sync_lock and tries to get bpo->bpo_lock: bpobj_space+0x63/0x470 [zfs] dsl_scan_active+0x340/0x3d0 [zfs] txg_sync_thread+0x3f2/0x1370 [zfs] thread 4 holds dp->dp_lock and tries to get tx->tx_sync_lock txg_kick+0x61/0x420 [zfs] dsl_pool_need_dirty_delay+0x1c7/0x3f0 [zfs] This patch is orginally from Brian Behlendorf and slightly simplified by me. It breaks this cycle in thread 4 by moving the call from dsl_pool_need_dirty_delay to txg_kick outside the section controlled by dp->dp_lock. Reviewed-by: Brian Behlendorf Reviewed-by: Matt Ahrens Signed-off-by: Jeff Dike Closes #9094 --- module/zfs/dsl_pool.c | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) diff --git a/module/zfs/dsl_pool.c b/module/zfs/dsl_pool.c index c342f0c51..9fb3a061d 100644 --- a/module/zfs/dsl_pool.c +++ b/module/zfs/dsl_pool.c @@ -889,14 +889,14 @@ dsl_pool_need_dirty_delay(dsl_pool_t *dp) zfs_dirty_data_max * zfs_delay_min_dirty_percent / 100; uint64_t dirty_min_bytes = zfs_dirty_data_max * zfs_dirty_data_sync_percent / 100; - boolean_t rv; + uint64_t dirty; mutex_enter(&dp->dp_lock); - if (dp->dp_dirty_total > dirty_min_bytes) - txg_kick(dp); - rv = (dp->dp_dirty_total > delay_min_bytes); + dirty = dp->dp_dirty_total; mutex_exit(&dp->dp_lock); - return (rv); + if (dirty > dirty_min_bytes) + txg_kick(dp); + return (dirty > delay_min_bytes); } void -- cgit v1.2.3