Skip site navigation (1)Skip section navigation (2) 
Date:      Mon, 7 Feb 2011 15:04:54 +0000 (UTC)
From:      Alexander Motin <mav@FreeBSD.org>
To:        src-committers@freebsd.org, svn-src-projects@freebsd.org
Subject:   svn commit: r218401 - projects/graid/head/sys/geom/raid
Message-ID:  <201102071504.p17F4sq9013027@svn.freebsd.org>
next in thread | raw e-mail | index | archive | help 
Author: mav
Date: Mon Feb  7 15:04:53 2011
New Revision: 218401
URL: http://svn.freebsd.org/changeset/base/218401

Log:
  Improve rebuild aborting. Depending on the current status we may be unable
  to abort rebuild immediately. In that case set ABORT flag and let things
  happen slowly. When rebuild abort, check if we can/want restart it again
  for another subdisk or even same subdisk, but from different position (if
  rebuilding disk failed during rebuild and was replaced by spare).

Modified:
  projects/graid/head/sys/geom/raid/tr_raid1.c

Modified: projects/graid/head/sys/geom/raid/tr_raid1.c
==============================================================================
--- projects/graid/head/sys/geom/raid/tr_raid1.c	Mon Feb  7 15:04:23 2011	(r218400)
+++ projects/graid/head/sys/geom/raid/tr_raid1.c	Mon Feb  7 15:04:53 2011	(r218401)
@@ -91,6 +91,8 @@ static MALLOC_DEFINE(M_TR_RAID1, "tr_rai
 #define TR_RAID1_RESYNC 2
 
 #define TR_RAID1_F_DOING_SOME	0x1
+#define TR_RAID1_F_LOCKED	0x2
+#define TR_RAID1_F_ABORT	0x4
 
 struct g_raid_tr_raid1_object {
 	struct g_raid_tr_object	 trso_base;
@@ -252,6 +254,7 @@ g_raid_tr_raid1_rebuild_some(struct g_ra
 	trs->trso_recover_slabs = g_raid1_rebuild_cluster_idle;
 	trs->trso_fair_io = g_raid1_rebuild_fair_io;
 	trs->trso_flags |= TR_RAID1_F_DOING_SOME;
+	trs->trso_flags |= TR_RAID1_F_LOCKED;
 	g_raid_lock_range(sd->sd_volume,	/* Lock callback starts I/O */
 	   bp->bio_offset, bp->bio_length, bp);
 }
@@ -298,9 +301,19 @@ g_raid_tr_raid1_rebuild_abort(struct g_r
 	vol = tr->tro_volume;
 	trs = (struct g_raid_tr_raid1_object *)tr;
 	sd = trs->trso_failed_sd;
-	len = MIN(g_raid1_rebuild_slab, vol->v_mediasize - sd->sd_rebuild_pos);
-	g_raid_unlock_range(tr->tro_volume, sd->sd_rebuild_pos, len);
-	g_raid_tr_raid1_rebuild_done(trs);
+	if (trs->trso_flags & TR_RAID1_F_DOING_SOME) {
+		trs->trso_flags |= TR_RAID1_F_ABORT;
+	} else {
+		trs->trso_flags &= ~TR_RAID1_F_ABORT;
+		if (trs->trso_flags & TR_RAID1_F_LOCKED) {
+			trs->trso_flags &= ~TR_RAID1_F_LOCKED;
+			len = MIN(g_raid1_rebuild_slab,
+			    vol->v_mediasize - sd->sd_rebuild_pos);
+			g_raid_unlock_range(tr->tro_volume,
+			    sd->sd_rebuild_pos, len);
+		}
+		g_raid_tr_raid1_rebuild_done(trs);
+	}
 }
 
 static void
@@ -396,7 +409,7 @@ g_raid_tr_raid1_maybe_rebuild(struct g_r
 		g_raid_tr_raid1_rebuild_start(tr);
 		break;
 	case TR_RAID1_REBUILD:
-		if (na == 0 || nr == 0)
+		if (na == 0 || nr == 0 || trs->trso_failed_sd == sd)
 			g_raid_tr_raid1_rebuild_abort(tr);
 		break;
 	case TR_RAID1_RESYNC:
@@ -628,7 +641,9 @@ g_raid_tr_iodone_raid1(struct g_raid_tr_
 				 */
 				G_RAID_LOGREQ(4, bp, "rebuild read done. %d",
 				    bp->bio_error);
-				if (bp->bio_error != 0) {
+				if (bp->bio_error != 0 ||
+				    trs->trso_flags & TR_RAID1_F_ABORT) {
+					trs->trso_flags &= ~TR_RAID1_F_DOING_SOME;
 					g_raid_tr_raid1_rebuild_abort(tr);
 					return;
 				}
@@ -649,13 +664,18 @@ g_raid_tr_iodone_raid1(struct g_raid_tr_
 				    "rebuild write done. Error %d",
 				    bp->bio_error);
 				nsd = trs->trso_failed_sd;
-				if (bp->bio_error != 0) {
-					g_raid_fail_disk(sd->sd_softc, nsd,
-					    nsd->sd_disk);
+				if (bp->bio_error != 0 ||
+				    trs->trso_flags & TR_RAID1_F_ABORT) {
+					if (bp->bio_error != 0) {
+						g_raid_fail_disk(sd->sd_softc,
+						    nsd, nsd->sd_disk);
+					}
+					trs->trso_flags &= ~TR_RAID1_F_DOING_SOME;
 					g_raid_tr_raid1_rebuild_abort(tr);
 					return;
 				}
 /* XXX A lot of the following is needed when we kick of the work -- refactor */
+				trs->trso_flags &= ~TR_RAID1_F_LOCKED;
 				g_raid_unlock_range(sd->sd_volume,
 				    bp->bio_offset, bp->bio_length);
 				nsd->sd_rebuild_pos += bp->bio_length;
@@ -667,7 +687,7 @@ g_raid_tr_iodone_raid1(struct g_raid_tr_
 				/* Abort rebuild if we are stopping */
 				if (trs->trso_stopping) {
 					trs->trso_flags &= ~TR_RAID1_F_DOING_SOME;
-					g_raid_tr_update_state_raid1(vol, NULL);
+					g_raid_tr_raid1_rebuild_abort(tr);
 					return;
 				}
 
@@ -684,6 +704,7 @@ g_raid_tr_iodone_raid1(struct g_raid_tr_
 				good_sd = g_raid_get_subdisk(sd->sd_volume,
 				    G_RAID_SUBDISK_S_ACTIVE);
 				if (good_sd == NULL) {
+					trs->trso_flags &= ~TR_RAID1_F_DOING_SOME;
 					g_raid_tr_raid1_rebuild_abort(tr);
 					return;
 				}
@@ -696,6 +717,7 @@ g_raid_tr_iodone_raid1(struct g_raid_tr_
 				G_RAID_LOGREQ(4, bp,
 				    "Rebuild read at %jd.", bp->bio_offset);
 				/* Lock callback starts I/O */
+				trs->trso_flags |= TR_RAID1_F_LOCKED;
 				g_raid_lock_range(sd->sd_volume,
 				    bp->bio_offset, bp->bio_length, bp);
 			}

Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?201102071504.p17F4sq9013027>