OpenZFS 9963 - Separate tunable for disabling ZIL vdev flush

Porting Notes:
* Add options to zfs-module-parameters(5) man page.
* zfs_nocacheflush move to vdev.c instead of vdev_disk.c, since
  the latter doesn't get built for user space.

Authored by: Prakash Surya <prakash.surya@delphix.com>
Reviewed by: Matt Ahrens <matt@delphix.com>
Reviewed by: Brad Lewis <brad.lewis@delphix.com>
Reviewed by: Patrick Mooney <patrick.mooney@joyent.com>
Reviewed by: Tom Caputi <tcaputi@datto.com>
Reviewed by: George Melikov <mail@gmelikov.ru>
Approved by: Dan McDonald <danmcd@joyent.com>
Ported-by: Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
OpenZFS-issue: https://www.illumos.org/issues/9963
OpenZFS-commit: https://github.com/openzfs/openzfs/commit/f8fdf68125
Closes #8186

diff --git a/man/man5/zfs-module-parameters.5 b/man/man5/zfs-module-parameters.5
index 837a7d1c3d52aefcd4abe4611dc720a8492755b6..55115c2661c473462f731156223347850b46b5ba 100644 (file)
--- a/man/man5/zfs-module-parameters.5
+++ b/man/man5/zfs-module-parameters.5
@@ -1846,8 +1846,9 @@ Use \fB1\fR for yes and \fB0\fR for no (default).
 \fBzfs_nocacheflush\fR (int)
 .ad
 .RS 12n
-Disable cache flush operations on disks when writing. Beware, this may cause
-corruption if disks re-order writes.
+Disable cache flush operations on disks when writing.  Setting this will
+cause pool corruption on power loss if a volatile out-of-order write cache
+is enabled.
 .sp
 Use \fB1\fR for yes and \fB0\fR for no (default).
 .RE
@@ -1903,8 +1904,6 @@ A value of zero will disable this throttle.
 Default value: \fB30\fR and \fB0\fR to disable.
 .RE
 
-
-
 .sp
 .ne 2
 .na
@@ -2527,6 +2526,19 @@ value of 100% will create a maximum of one thread per cpu.
 Default value: \fB100\fR%.
 .RE
 
+.sp
+.ne 2
+.na
+\fBzil_nocacheflush\fR (int)
+.ad
+.RS 12n
+Disable the cache flush commands that are normally sent to the disk(s) by
+the ZIL after an LWB write has completed. Setting this will cause ZIL
+corruption on power loss if a volatile out-of-order write cache is enabled.
+.sp
+Use \fB1\fR for yes and \fB0\fR for no (default).
+.RE
+
 .sp
 .ne 2
 .na

diff --git a/module/zfs/vdev.c b/module/zfs/vdev.c
index ff5a15365720f459778f879449249d679409a5c9..8273e79079f147c249725a6557ca0beb2116c575 100644 (file)
--- a/module/zfs/vdev.c
+++ b/module/zfs/vdev.c
@@ -99,6 +99,13 @@ int zfs_scan_ignore_errors = 0;
  */
 int vdev_standard_sm_blksz = (1 << 17);
 
+/*
+ * Tunable parameter for debugging or performance analysis. Setting this
+ * will cause pool corruption on power loss if a volatile out-of-order
+ * write cache is enabled.
+ */
+int zfs_nocacheflush = 0;
+
 /*PRINTFLIKE2*/
 void
 vdev_dbgmsg(vdev_t *vd, const char *fmt, ...)
@@ -4650,5 +4657,8 @@ MODULE_PARM_DESC(zfs_scan_ignore_errors,
 module_param(vdev_validate_skip, int, 0644);
 MODULE_PARM_DESC(vdev_validate_skip,
        "Bypass vdev_validate()");
+
+module_param(zfs_nocacheflush, int, 0644);
+MODULE_PARM_DESC(zfs_nocacheflush, "Disable cache flushes");
 /* END CSTYLED */
 #endif

diff --git a/module/zfs/zil.c b/module/zfs/zil.c
index d5ebf7561f1bb1c844dbd463d12c1eedfc91a97a..a453c26c37eec7f523f2f625a24ee0fb736d42ee 100644 (file)
--- a/module/zfs/zil.c
+++ b/module/zfs/zil.c
@@ -118,11 +118,12 @@ static kstat_t *zil_ksp;
 int zil_replay_disable = 0;
 
 /*
- * Tunable parameter for debugging or performance analysis.  Setting
- * zfs_nocacheflush will cause corruption on power loss if a volatile
- * out-of-order write cache is enabled.
+ * Disable the DKIOCFLUSHWRITECACHE commands that are normally sent to
+ * the disk(s) by the ZIL after an LWB write has completed. Setting this
+ * will cause ZIL corruption on power loss if a volatile out-of-order
+ * write cache is enabled.
  */
-int zfs_nocacheflush = 0;
+int zil_nocacheflush = 0;
 
 /*
  * Limit SLOG write size per commit executed with synchronous priority.
@@ -1041,7 +1042,7 @@ zil_lwb_add_block(lwb_t *lwb, const blkptr_t *bp)
        int ndvas = BP_GET_NDVAS(bp);
        int i;
 
-       if (zfs_nocacheflush)
+       if (zil_nocacheflush)
                return;
 
        mutex_enter(&lwb->lwb_vdev_lock);
@@ -1065,7 +1066,7 @@ zil_lwb_add_txg(lwb_t *lwb, uint64_t txg)
 /*
  * This function is a called after all VDEVs associated with a given lwb
  * write have completed their DKIOCFLUSHWRITECACHE command; or as soon
- * as the lwb write completes, if "zfs_nocacheflush" is set.
+ * as the lwb write completes, if "zil_nocacheflush" is set.
  *
  * The intention is for this function to be called as soon as the
  * contents of an lwb are considered "stable" on disk, and will survive
@@ -3513,8 +3514,8 @@ MODULE_PARM_DESC(zfs_commit_timeout_pct, "ZIL block open timeout percentage");
 module_param(zil_replay_disable, int, 0644);
 MODULE_PARM_DESC(zil_replay_disable, "Disable intent logging replay");
 
-module_param(zfs_nocacheflush, int, 0644);
-MODULE_PARM_DESC(zfs_nocacheflush, "Disable cache flushes");
+module_param(zil_nocacheflush, int, 0644);
+MODULE_PARM_DESC(zil_nocacheflush, "Disable ZIL cache flushes");
 
 module_param(zil_slog_bulk, ulong, 0644);
 MODULE_PARM_DESC(zil_slog_bulk, "Limit in bytes slog sync writes per commit");