Print this page
2619 asynchronous destruction of ZFS file systems
2747 SPA versioning with zfs feature flags
Reviewed by: Matt Ahrens <mahrens@delphix.com>
Reviewed by: George Wilson <gwilson@delphix.com>
Reviewed by: Richard Lowe <richlowe@richlowe.net>
Reviewed by: Dan Kruchinin <dan.kruchinin@gmail.com>
Approved by: Dan McDonald <danmcd@nexenta.com>
@@ -16,12 +16,14 @@
* fields enclosed by brackets "[]" replaced with your own identifying
* information: Portions Copyright [yyyy] [name of copyright owner]
*
* CDDL HEADER END
*/
+
/*
* Copyright (c) 2005, 2010, Oracle and/or its affiliates. All rights reserved.
+ * Copyright (c) 2012 by Delphix. All rights reserved.
*/
/*
* Virtual Device Labels
* ---------------------
@@ -119,10 +121,12 @@
* name Pool name
* state Pool state
* txg Transaction group in which this label was written
* pool_guid Unique identifier for this pool
* vdev_tree An nvlist describing vdev tree.
+ * features_for_read
+ * An nvlist of the features necessary for reading the MOS.
*
* Each leaf device label also contains the following:
*
* top_guid Unique ID for top-level vdev in which this is contained
* guid Unique ID for the leaf vdev
@@ -426,12 +430,17 @@
rvd->vdev_children) == 0);
kmem_free(array, rvd->vdev_children * sizeof (uint64_t));
}
+/*
+ * Returns the configuration from the label of the given vdev. If 'label' is
+ * VDEV_BEST_LABEL, each label of the vdev will be read until a valid
+ * configuration is found; otherwise, only the specified label will be read.
+ */
nvlist_t *
-vdev_label_read_config(vdev_t *vd)
+vdev_label_read_config(vdev_t *vd, int label)
{
spa_t *spa = vd->vdev_spa;
nvlist_t *config = NULL;
vdev_phys_t *vp;
zio_t *zio;
@@ -445,10 +454,12 @@
vp = zio_buf_alloc(sizeof (vdev_phys_t));
retry:
for (int l = 0; l < VDEV_LABELS; l++) {
+ if (label >= 0 && label < VDEV_LABELS && label != l)
+ continue;
zio = zio_root(spa, NULL, NULL, flags);
vdev_label_read(zio, vd, l, vp,
offsetof(vdev_label_t, vl_vdev_phys),
@@ -494,11 +505,11 @@
*l2cache_guid = 0ULL;
/*
* Read the label, if any, and perform some basic sanity checks.
*/
- if ((label = vdev_label_read_config(vd)) == NULL)
+ if ((label = vdev_label_read_config(vd, VDEV_BEST_LABEL)) == NULL)
return (B_FALSE);
(void) nvlist_lookup_uint64(label, ZPOOL_CONFIG_CREATE_TXG,
&vdtxg);
@@ -831,11 +842,11 @@
* Consider the following situation: txg is safely synced to disk. We've
* written the first uberblock for txg + 1, and then we lose power. When we
* come back up, we fail to see the uberblock for txg + 1 because, say,
* it was on a mirrored device and the replica to which we wrote txg + 1
* is now offline. If we then make some changes and sync txg + 1, and then
- * the missing replica comes back, then for a new seconds we'll have two
+ * the missing replica comes back, then for a few seconds we'll have two
* conflicting uberblocks on disk with the same txg. The solution is simple:
* among uberblocks with equal txg, choose the one with the latest timestamp.
*/
static int
vdev_uberblock_compare(uberblock_t *ub1, uberblock_t *ub2)
@@ -851,50 +862,54 @@
return (1);
return (0);
}
+struct ubl_cbdata {
+ uberblock_t *ubl_ubbest; /* Best uberblock */
+ vdev_t *ubl_vd; /* vdev associated with the above */
+ int ubl_label; /* Label associated with the above */
+};
+
static void
vdev_uberblock_load_done(zio_t *zio)
{
+ vdev_t *vd = zio->io_vd;
spa_t *spa = zio->io_spa;
zio_t *rio = zio->io_private;
uberblock_t *ub = zio->io_data;
- uberblock_t *ubbest = rio->io_private;
+ struct ubl_cbdata *cbp = rio->io_private;
- ASSERT3U(zio->io_size, ==, VDEV_UBERBLOCK_SIZE(zio->io_vd));
+ ASSERT3U(zio->io_size, ==, VDEV_UBERBLOCK_SIZE(vd));
if (zio->io_error == 0 && uberblock_verify(ub) == 0) {
mutex_enter(&rio->io_lock);
if (ub->ub_txg <= spa->spa_load_max_txg &&
- vdev_uberblock_compare(ub, ubbest) > 0)
- *ubbest = *ub;
+ vdev_uberblock_compare(ub, cbp->ubl_ubbest) > 0) {
+ /*
+ * Keep track of the vdev and label in which this
+ * uberblock was found. We will use this information
+ * later to obtain the config nvlist associated with
+ * this uberblock.
+ */
+ *cbp->ubl_ubbest = *ub;
+ cbp->ubl_vd = vd;
+ cbp->ubl_label = vdev_label_number(vd->vdev_psize,
+ zio->io_offset);
+ }
mutex_exit(&rio->io_lock);
}
zio_buf_free(zio->io_data, zio->io_size);
}
-void
-vdev_uberblock_load(zio_t *zio, vdev_t *vd, uberblock_t *ubbest)
+static void
+vdev_uberblock_load_impl(zio_t *zio, vdev_t *vd, int flags,
+ struct ubl_cbdata *cbp)
{
- spa_t *spa = vd->vdev_spa;
- vdev_t *rvd = spa->spa_root_vdev;
- int flags = ZIO_FLAG_CONFIG_WRITER | ZIO_FLAG_CANFAIL |
- ZIO_FLAG_SPECULATIVE | ZIO_FLAG_TRYHARD;
-
- if (vd == rvd) {
- ASSERT(zio == NULL);
- spa_config_enter(spa, SCL_ALL, FTAG, RW_WRITER);
- zio = zio_root(spa, NULL, ubbest, flags);
- bzero(ubbest, sizeof (uberblock_t));
- }
-
- ASSERT(zio != NULL);
-
for (int c = 0; c < vd->vdev_children; c++)
- vdev_uberblock_load(zio, vd->vdev_child[c], ubbest);
+ vdev_uberblock_load_impl(zio, vd->vdev_child[c], flags, cbp);
if (vd->vdev_ops->vdev_op_leaf && vdev_readable(vd)) {
for (int l = 0; l < VDEV_LABELS; l++) {
for (int n = 0; n < VDEV_UBERBLOCK_COUNT(vd); n++) {
vdev_label_read(zio, vd, l,
@@ -903,15 +918,49 @@
VDEV_UBERBLOCK_SIZE(vd),
vdev_uberblock_load_done, zio, flags);
}
}
}
+}
- if (vd == rvd) {
+/*
+ * Reads the 'best' uberblock from disk along with its associated
+ * configuration. First, we read the uberblock array of each label of each
+ * vdev, keeping track of the uberblock with the highest txg in each array.
+ * Then, we read the configuration from the same label as the best uberblock.
+ */
+void
+vdev_uberblock_load(vdev_t *rvd, uberblock_t *ub, nvlist_t **config)
+{
+ int i;
+ zio_t *zio;
+ spa_t *spa = rvd->vdev_spa;
+ struct ubl_cbdata cb;
+ int flags = ZIO_FLAG_CONFIG_WRITER | ZIO_FLAG_CANFAIL |
+ ZIO_FLAG_SPECULATIVE | ZIO_FLAG_TRYHARD;
+
+ ASSERT(ub);
+ ASSERT(config);
+
+ bzero(ub, sizeof (uberblock_t));
+ *config = NULL;
+
+ cb.ubl_ubbest = ub;
+ cb.ubl_vd = NULL;
+
+ spa_config_enter(spa, SCL_ALL, FTAG, RW_WRITER);
+ zio = zio_root(spa, NULL, &cb, flags);
+ vdev_uberblock_load_impl(zio, rvd, flags, &cb);
(void) zio_wait(zio);
- spa_config_exit(spa, SCL_ALL, FTAG);
+ if (cb.ubl_vd != NULL) {
+ for (i = cb.ubl_label % 2; i < VDEV_LABELS; i += 2) {
+ *config = vdev_label_read_config(cb.ubl_vd, i);
+ if (*config != NULL)
+ break;
}
+ }
+ spa_config_exit(spa, SCL_ALL, FTAG);
}
/*
* On success, increment root zio's count of good writes.
* We only get credit for writes to known-visible vdevs; see spa_vdev_add().