From: jsing <jsing@openbsd.org>
Date: Sat, 11 Apr 2015 16:37:34 +0000 (+0000)
Subject: Re-enable the RAID 5 discipline for softraid(4).
X-Git-Url: http://artulab.com/gitweb/?a=commitdiff_plain;h=74dd158f2d1a9ed7a2702a5bf45f589943528f6e;p=openbsd

Re-enable the RAID 5 discipline for softraid(4).

The RAID 5 implementation has been largely rewritten during the last two
hackathons in Dunedin - it now needs further testing and usage.
---

diff --git a/sbin/bioctl/bioctl.8 b/sbin/bioctl/bioctl.8
index 8cb965eaa25..2eb5bcba00c 100644
--- a/sbin/bioctl/bioctl.8
+++ b/sbin/bioctl/bioctl.8
@@ -1,4 +1,4 @@
-.\"	$OpenBSD: bioctl.8,v 1.94 2015/02/28 21:51:57 bentley Exp $
+.\"	$OpenBSD: bioctl.8,v 1.95 2015/04/11 16:37:34 jsing Exp $
 .\"
 .\" Copyright (c) 2004, 2005 Marco Peereboom
 .\"
@@ -23,7 +23,7 @@
 .\" OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
 .\" SUCH DAMAGE.
 .\"
-.Dd $Mdocdate: February 28 2015 $
+.Dd $Mdocdate: April 11 2015 $
 .Dt BIOCTL 8
 .Os
 .Sh NAME
@@ -194,9 +194,9 @@ A striping discipline.
 .It 1
 RAID 1:
 A mirroring discipline.
-.\" .It 5
-.\" RAID 5:
-.\" A striping discipline with floating parity chunk.
+.It 5
+RAID 5:
+A striping discipline with floating parity chunk.
 .It C
 CRYPTO:
 An encrypting discipline.
@@ -207,8 +207,8 @@ A concatenating discipline.
 .Pp
 The RAID 0, RAID 1 and CONCAT disciplines require a minimum of two devices to
 be provided via
-.Fl l
-.\" RAID 5 requires at least three devices,
+.Fl l .
+RAID 5 requires at least three devices
 and the CRYPTO discipline requires exactly one.
 .It Fl d
 Delete volume specified by device.
diff --git a/sbin/bioctl/bioctl.c b/sbin/bioctl/bioctl.c
index e823581c832..393f1d54942 100644
--- a/sbin/bioctl/bioctl.c
+++ b/sbin/bioctl/bioctl.c
@@ -1,4 +1,4 @@
-/* $OpenBSD: bioctl.c,v 1.124 2015/03/18 07:42:13 uebayasi Exp $       */
+/* $OpenBSD: bioctl.c,v 1.125 2015/04/11 16:37:34 jsing Exp $       */
 
 /*
  * Copyright (c) 2004, 2005 Marco Peereboom
@@ -785,11 +785,9 @@ bio_createraid(u_int16_t level, char *dev_list, char *key_disk)
 	case 1:
 		min_disks = 2;
 		break;
-#ifdef RAID5
 	case 5:
 		min_disks = 3;
 		break;
-#endif /* RAID5 */
 	case 'C':
 		min_disks = 1;
 		break;
diff --git a/share/man/man4/softraid.4 b/share/man/man4/softraid.4
index 9979102009f..d815bb0b95d 100644
--- a/share/man/man4/softraid.4
+++ b/share/man/man4/softraid.4
@@ -1,4 +1,4 @@
-.\"	$OpenBSD: softraid.4,v 1.39 2015/02/16 16:38:54 naddy Exp $
+.\"	$OpenBSD: softraid.4,v 1.40 2015/04/11 16:37:34 jsing Exp $
 .\"
 .\" Copyright (c) 2007 Todd T. Fries   <todd@OpenBSD.org>
 .\" Copyright (c) 2007 Marco Peereboom <marco@OpenBSD.org>
@@ -15,7 +15,7 @@
 .\" ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF
 .\" OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE.
 .\"
-.Dd $Mdocdate: February 16 2015 $
+.Dd $Mdocdate: April 11 2015 $
 .Dt SOFTRAID 4
 .Os
 .Sh NAME
@@ -78,14 +78,14 @@ though at the cost of write speed.
 Unlike traditional RAID 1,
 .Nm
 supports the use of more than two chunks in a RAID 1 setup.
-.\" .It RAID 5
-.\" A striping discipline with
-.\" .Em floating parity
-.\" across all chunks.
-.\" It stripes data across chunks and provides parity to prevent data loss of
-.\" a single chunk failure.
-.\" Read performance is increased;
-.\" write performance does incur additional overhead.
+.It RAID 5
+A striping discipline with
+.Em floating parity
+across all chunks.
+It stripes data across chunks and provides parity to prevent data loss of
+a single chunk failure.
+Read performance is increased;
+write performance does incur additional overhead.
 .It CRYPTO
 An
 .Em encrypting
@@ -198,9 +198,9 @@ The RAID 1 discipline does not initialize the mirror upon creation.
 This is by design because all sectors that are read are written first.
 There is no point in wasting a lot of time syncing random data.
 .Pp
-.\" The RAID 5 discipline do not initialize the parity upon creation.
-.\" This is due to the scrub functionality not being currently implemented.
-.\" .Pp
+The RAID 5 discipline does not initialize parity upon creation, instead parity
+is only updated upon write.
+.Pp
 Currently there is no automated mechanism to recover from failed disks.
 .Pp
 Sparc hardware needs to use fstype