Checksum errors may not be counted

Fix regression seen in issue #11545 where checksum errors 
where not being counted or showing up in a zpool event.

Reviewed-by: Matthew Ahrens <mahrens@delphix.com>
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Don Brady <don.brady@delphix.com>
Closes #11609

module/zfs/dsl_scan.c

@@ -20,7 +20,7 @@
  */
 /*
  * Copyright (c) 2008, 2010, Oracle and/or its affiliates. All rights reserved.
- * Copyright (c) 2011, 2018 by Delphix. All rights reserved.
+ * Copyright (c) 2011, 2021 by Delphix. All rights reserved.
  * Copyright 2016 Gary Mills
  * Copyright (c) 2017, 2019, Datto Inc. All rights reserved.
  * Copyright (c) 2015, Nexenta Systems, Inc. All rights reserved.
@@ -987,6 +987,10 @@ dsl_scan_done(dsl_scan_t *scn, boolean_t complete, dmu_tx_t *tx)
 			    (u_longlong_t)spa_get_errlog_size(spa));
 			spa_async_request(spa, SPA_ASYNC_RESILVER);
 		}
+
+		/* Clear recent error events (i.e. duplicate events tracking) */
+		if (complete)
+			zfs_ereport_clear(spa, NULL);
 	}
 
 	scn->scn_phys.scn_end_time = gethrestime_sec();

module/zfs/vdev.c

@@ -21,7 +21,7 @@
 
 /*
  * Copyright (c) 2005, 2010, Oracle and/or its affiliates. All rights reserved.
- * Copyright (c) 2011, 2020 by Delphix. All rights reserved.
+ * Copyright (c) 2011, 2021 by Delphix. All rights reserved.
  * Copyright 2017 Nexenta Systems, Inc.
  * Copyright (c) 2014 Integros [integros.com]
  * Copyright 2016 Toomas Soome <tsoome@me.com>
@@ -4170,6 +4170,9 @@ vdev_clear(spa_t *spa, vdev_t *vd)
 	    vd->vdev_parent->vdev_ops == &vdev_spare_ops &&
 	    vd->vdev_parent->vdev_child[0] == vd)
 		vd->vdev_unspare = B_TRUE;
+
+	/* Clear recent error events cache (i.e. duplicate events tracking) */
+	zfs_ereport_clear(spa, vd);
 }
 
 boolean_t

module/zfs/vdev_indirect.c

@@ -1485,14 +1485,12 @@ vdev_indirect_all_checksum_errors(zio_t *zio)
 
 			vdev_t *vd = ic->ic_vdev;
 
-			int ret = zfs_ereport_post_checksum(zio->io_spa, vd,
+			(void) zfs_ereport_post_checksum(zio->io_spa, vd,
 			    NULL, zio, is->is_target_offset, is->is_size,
 			    NULL, NULL, NULL);
-			if (ret != EALREADY) {
-				mutex_enter(&vd->vdev_stat_lock);
-				vd->vdev_stat.vs_checksum_errors++;
-				mutex_exit(&vd->vdev_stat_lock);
-			}
+			mutex_enter(&vd->vdev_stat_lock);
+			vd->vdev_stat.vs_checksum_errors++;
+			mutex_exit(&vd->vdev_stat_lock);
 		}
 	}
 }

module/zfs/vdev_raidz.c

@@ -1852,14 +1852,12 @@ raidz_checksum_error(zio_t *zio, raidz_col_t *rc, abd_t *bad_data)
 		zbc.zbc_has_cksum = 0;
 		zbc.zbc_injected = rm->rm_ecksuminjected;
 
-		int ret = zfs_ereport_post_checksum(zio->io_spa, vd,
+		(void) zfs_ereport_post_checksum(zio->io_spa, vd,
 		    &zio->io_bookmark, zio, rc->rc_offset, rc->rc_size,
 		    rc->rc_abd, bad_data, &zbc);
-		if (ret != EALREADY) {
-			mutex_enter(&vd->vdev_stat_lock);
-			vd->vdev_stat.vs_checksum_errors++;
-			mutex_exit(&vd->vdev_stat_lock);
-		}
+		mutex_enter(&vd->vdev_stat_lock);
+		vd->vdev_stat.vs_checksum_errors++;
+		mutex_exit(&vd->vdev_stat_lock);
 	}
 }
 
@@ -2453,14 +2451,12 @@ vdev_raidz_io_done_unrecoverable(zio_t *zio)
 			zbc.zbc_has_cksum = 0;
 			zbc.zbc_injected = rm->rm_ecksuminjected;
 
-			int ret = zfs_ereport_start_checksum(zio->io_spa,
+			(void) zfs_ereport_start_checksum(zio->io_spa,
 			    cvd, &zio->io_bookmark, zio, rc->rc_offset,
 			    rc->rc_size, (void *)(uintptr_t)c, &zbc);
-			if (ret != EALREADY) {
-				mutex_enter(&cvd->vdev_stat_lock);
-				cvd->vdev_stat.vs_checksum_errors++;
-				mutex_exit(&cvd->vdev_stat_lock);
-			}
+			mutex_enter(&cvd->vdev_stat_lock);
+			cvd->vdev_stat.vs_checksum_errors++;
+			mutex_exit(&cvd->vdev_stat_lock);
 		}
 	}
 }

module/zfs/vdev_rebuild.c

@@ -338,6 +338,9 @@ vdev_rebuild_complete_sync(void *arg, dmu_tx_t *tx)
 	}
 
 	cv_broadcast(&vd->vdev_rebuild_cv);
+
+	/* Clear recent error events (i.e. duplicate events tracking) */
+	zfs_ereport_clear(spa, NULL);
 }
 
 /*

module/zfs/zfs_fm.c

@@ -24,7 +24,7 @@
  */
 
 /*
- * Copyright (c) 2012,2020 by Delphix. All rights reserved.
+ * Copyright (c) 2012,2021 by Delphix. All rights reserved.
  */
 
 #include <sys/spa.h>
@@ -247,6 +247,44 @@ zfs_ereport_schedule_cleaner(void)
 	    ddi_get_lbolt() + NSEC_TO_TICK(timeout));
 }
 
+/*
+ * Clear entries for a given vdev or all vdevs in a pool when vdev == NULL
+ */
+void
+zfs_ereport_clear(spa_t *spa, vdev_t *vd)
+{
+	uint64_t vdev_guid, pool_guid;
+	int cnt = 0;
+
+	ASSERT(vd != NULL || spa != NULL);
+	if (vd == NULL) {
+		vdev_guid = 0;
+		pool_guid = spa_guid(spa);
+	} else {
+		vdev_guid = vd->vdev_guid;
+		pool_guid = 0;
+	}
+
+	mutex_enter(&recent_events_lock);
+
+	recent_events_node_t *next = list_head(&recent_events_list);
+	while (next != NULL) {
+		recent_events_node_t *entry = next;
+
+		next = list_next(&recent_events_list, next);
+
+		if (entry->re_vdev_guid == vdev_guid ||
+		    entry->re_pool_guid == pool_guid) {
+			avl_remove(&recent_events_tree, entry);
+			list_remove(&recent_events_list, entry);
+			kmem_free(entry, sizeof (*entry));
+			cnt++;
+		}
+	}
+
+	mutex_exit(&recent_events_lock);
+}
+
 /*
  * Check if an ereport would be a duplicate of one recently posted.
  *
@@ -951,6 +989,12 @@ annotate_ecksum(nvlist_t *ereport, zio_bad_cksum_t *info,
 	}
 	return (eip);
 }
+#else
+/*ARGSUSED*/
+void
+zfs_ereport_clear(spa_t *spa, vdev_t *vd)
+{
+}
 #endif
 
 /*

module/zfs/zio.c

@@ -4255,15 +4255,12 @@ zio_checksum_verify(zio_t *zio)
 		zio->io_error = error;
 		if (error == ECKSUM &&
 		    !(zio->io_flags & ZIO_FLAG_SPECULATIVE)) {
-			int ret = zfs_ereport_start_checksum(zio->io_spa,
+			(void) zfs_ereport_start_checksum(zio->io_spa,
 			    zio->io_vd, &zio->io_bookmark, zio,
 			    zio->io_offset, zio->io_size, NULL, &info);
-
-			if (ret != EALREADY) {
-				mutex_enter(&zio->io_vd->vdev_stat_lock);
-				zio->io_vd->vdev_stat.vs_checksum_errors++;
-				mutex_exit(&zio->io_vd->vdev_stat_lock);
-			}
+			mutex_enter(&zio->io_vd->vdev_stat_lock);
+			zio->io_vd->vdev_stat.vs_checksum_errors++;
+			mutex_exit(&zio->io_vd->vdev_stat_lock);
 		}
 	}