MFV r336991, r337001:

9102 zfs should be able to initialize storage devices

The first access to a disk block can incur a performance penalty on some
platforms (e.g. AWS's EBS, VMware VMDKs). Therefore it is recommended that
volumes be "thick provisioned", where supported by the platform (VMware).
Thick provisioning is time consuming and often is ignored. If the thick
provision step is omitted, customers will see suboptimal performance until
we have written to all parts of the LUN. ZFS should be able to initialize
any unused storage to remove any first-write penalty that exists.

illumos/illumos-gate@094e47e980b0796b94b1b8f51f462a64d246e516

Reviewed by: John Wren Kennedy <john.kennedy@delphix.com>
Reviewed by: Matthew Ahrens <mahrens@delphix.com>
Reviewed by: Pavel Zakharov <pavel.zakharov@delphix.com>
Reviewed by: Prakash Surya <prakash.surya@delphix.com>
Approved by: Richard Lowe <richlowe@richlowe.net>
Author:     George Wilson <george.wilson@delphix.com>

3 files changed, 106 insertions, 0 deletions

diff --git a/cddl/contrib/opensolaris/lib/libzfs/common/libzfs.h b/cddl/contrib/opensolaris/lib/libzfs/common/libzfs.h
index d518abb86644..0c50eb89eaca 100644
--- a/cddl/contrib/opensolaris/lib/libzfs/common/libzfs.h
+++ b/cddl/contrib/opensolaris/lib/libzfs/common/libzfs.h
@@ -137,6 +137,9 @@ typedef enum zfs_error {
 	EZFS_NO_CHECKPOINT,	/* pool has no checkpoint */
 	EZFS_DEVRM_IN_PROGRESS,	/* a device is currently being removed */
 	EZFS_VDEV_TOO_BIG,	/* a device is too big to be used */
+	EZFS_TOOMANY,		/* argument list too long */
+	EZFS_INITIALIZING,	/* currently initializing */
+	EZFS_NO_INITIALIZE,	/* no active initialize */
 	EZFS_UNKNOWN
 } zfs_error_t;
 
@@ -262,6 +265,8 @@ typedef struct splitflags {
  * Functions to manipulate pool and vdev state
  */
 extern int zpool_scan(zpool_handle_t *, pool_scan_func_t, pool_scrub_cmd_t);
+extern int zpool_initialize(zpool_handle_t *, pool_initialize_func_t,
+    nvlist_t *);
 extern int zpool_clear(zpool_handle_t *, const char *, nvlist_t *);
 extern int zpool_reguid(zpool_handle_t *);
 extern int zpool_reopen(zpool_handle_t *);
diff --git a/cddl/contrib/opensolaris/lib/libzfs/common/libzfs_pool.c b/cddl/contrib/opensolaris/lib/libzfs/common/libzfs_pool.c
index c94963bf5713..392ed3cfa868 100644
--- a/cddl/contrib/opensolaris/lib/libzfs/common/libzfs_pool.c
+++ b/cddl/contrib/opensolaris/lib/libzfs/common/libzfs_pool.c
@@ -1981,6 +1981,100 @@ zpool_scan(zpool_handle_t *zhp, pool_scan_func_t func, pool_scrub_cmd_t cmd)
 	}
 }
 
+static int
+xlate_init_err(int err)
+{
+	switch (err) {
+	case ENODEV:
+		return (EZFS_NODEVICE);
+	case EINVAL:
+	case EROFS:
+		return (EZFS_BADDEV);
+	case EBUSY:
+		return (EZFS_INITIALIZING);
+	case ESRCH:
+		return (EZFS_NO_INITIALIZE);
+	}
+	return (err);
+}
+
+/*
+ * Begin, suspend, or cancel the initialization (initializing of all free
+ * blocks) for the given vdevs in the given pool.
+ */
+int
+zpool_initialize(zpool_handle_t *zhp, pool_initialize_func_t cmd_type,
+    nvlist_t *vds)
+{
+	char msg[1024];
+	libzfs_handle_t *hdl = zhp->zpool_hdl;
+
+	nvlist_t *errlist;
+
+	/* translate vdev names to guids */
+	nvlist_t *vdev_guids = fnvlist_alloc();
+	nvlist_t *guids_to_paths = fnvlist_alloc();
+	boolean_t spare, cache;
+	nvlist_t *tgt;
+	nvpair_t *elem;
+
+	for (elem = nvlist_next_nvpair(vds, NULL); elem != NULL;
+	    elem = nvlist_next_nvpair(vds, elem)) {
+		char *vd_path = nvpair_name(elem);
+		tgt = zpool_find_vdev(zhp, vd_path, &spare, &cache, NULL);
+
+		if ((tgt == NULL) || cache || spare) {
+			(void) snprintf(msg, sizeof (msg),
+			    dgettext(TEXT_DOMAIN, "cannot initialize '%s'"),
+			    vd_path);
+			int err = (tgt == NULL) ? EZFS_NODEVICE :
+			    (spare ? EZFS_ISSPARE : EZFS_ISL2CACHE);
+			fnvlist_free(vdev_guids);
+			fnvlist_free(guids_to_paths);
+			return (zfs_error(hdl, err, msg));
+		}
+
+		uint64_t guid = fnvlist_lookup_uint64(tgt, ZPOOL_CONFIG_GUID);
+		fnvlist_add_uint64(vdev_guids, vd_path, guid);
+
+		(void) snprintf(msg, sizeof (msg), "%llu", guid);
+		fnvlist_add_string(guids_to_paths, msg, vd_path);
+	}
+
+	int err = lzc_initialize(zhp->zpool_name, cmd_type, vdev_guids,
+	    &errlist);
+	fnvlist_free(vdev_guids);
+
+	if (err == 0) {
+		fnvlist_free(guids_to_paths);
+		return (0);
+	}
+
+	nvlist_t *vd_errlist = NULL;
+	if (errlist != NULL) {
+		vd_errlist = fnvlist_lookup_nvlist(errlist,
+		    ZPOOL_INITIALIZE_VDEVS);
+	}
+
+	(void) snprintf(msg, sizeof (msg),
+	    dgettext(TEXT_DOMAIN, "operation failed"));
+
+	for (elem = nvlist_next_nvpair(vd_errlist, NULL); elem != NULL;
+	    elem = nvlist_next_nvpair(vd_errlist, elem)) {
+		int64_t vd_error = xlate_init_err(fnvpair_value_int64(elem));
+		char *path = fnvlist_lookup_string(guids_to_paths,
+		    nvpair_name(elem));
+		(void) zfs_error_fmt(hdl, vd_error, "cannot initialize '%s'",
+		    path);
+	}
+
+	fnvlist_free(guids_to_paths);
+	if (vd_errlist != NULL)
+		return (-1);
+
+	return (zpool_standard_error(hdl, err, msg));
+}
+
 #ifdef illumos
 /*
  * This provides a very minimal check whether a given string is likely a
diff --git a/cddl/contrib/opensolaris/lib/libzfs/common/libzfs_util.c b/cddl/contrib/opensolaris/lib/libzfs/common/libzfs_util.c
index 6b52f9ad86d6..faf00486751c 100644
--- a/cddl/contrib/opensolaris/lib/libzfs/common/libzfs_util.c
+++ b/cddl/contrib/opensolaris/lib/libzfs/common/libzfs_util.c
@@ -254,6 +254,13 @@ libzfs_error_description(libzfs_handle_t *hdl)
 		return (dgettext(TEXT_DOMAIN, "device removal in progress"));
 	case EZFS_VDEV_TOO_BIG:
 		return (dgettext(TEXT_DOMAIN, "device exceeds supported size"));
+	case EZFS_TOOMANY:
+		return (dgettext(TEXT_DOMAIN, "argument list too long"));
+	case EZFS_INITIALIZING:
+		return (dgettext(TEXT_DOMAIN, "currently initializing"));
+	case EZFS_NO_INITIALIZE:
+		return (dgettext(TEXT_DOMAIN, "there is no active "
+		    "initialization"));
 	case EZFS_UNKNOWN:
 		return (dgettext(TEXT_DOMAIN, "unknown error"));
 	default: