Implement a hotspot for the sunlabel.

This means that you can no longer trash your opened partitions by writing to
the sunlabel through another partition.  This is similar to the semantics
implemented for BSD labels.

1 files changed, 38 insertions, 3 deletions

diff --git a/sys/geom/geom_sunlabel.c b/sys/geom/geom_sunlabel.c
index f7a66676c46b..8e09362e7079 100644
--- a/sys/geom/geom_sunlabel.c
+++ b/sys/geom/geom_sunlabel.c
@@ -103,6 +103,41 @@ g_sunlabel_modify(struct g_geom *gp, struct g_sunlabel_softc *ms, u_char *sec0)
 	return (0);
 }
 
+static void
+g_sunlabel_hotwrite(void *arg, int flag)
+{
+	struct bio *bp;
+	struct g_geom *gp;
+	struct g_slicer *gsp;
+	struct g_slice *gsl;
+	struct g_sunlabel_softc *ms;
+	u_char *p;
+	int error;
+
+	KASSERT(flag != EV_CANCEL, ("g_sunlabel_hotwrite cancelled"));
+	bp = arg;
+	gp = bp->bio_to->geom;
+	gsp = gp->softc;
+	ms = gsp->softc;
+	gsl = &gsp->slices[bp->bio_to->index];
+	/*
+	 * XXX: For all practical purposes, this whould be equvivalent to
+	 * XXX: "p = (u_char *)bp->bio_data;" because the label is always
+	 * XXX: in the first sector and we refuse sectors smaller than the
+	 * XXX: label.
+	 */
+	p = (u_char *)bp->bio_data - (bp->bio_offset + gsl->offset);
+
+	g_topology_unlock();
+	error = g_sunlabel_modify(gp, ms, p);
+	g_topology_lock();
+	if (error) {
+		g_io_deliver(bp, EPERM);
+		return;
+	}
+	g_slice_finish_hot(bp);
+}
+
 static int
 g_sunlabel_start(struct bio *bp)
 {
@@ -174,10 +209,10 @@ g_sunlabel_taste(struct g_class *mp, struct g_provider *pp, int flags)
 	if (LIST_EMPTY(&gp->provider)) {
 		g_std_spoiled(cp);
 		return (NULL);
-#ifdef notyet
 	} else {
-		g_slice_conf_hot(gp, 0, 0, ms->sectorsize);
-#endif
+		g_slice_conf_hot(gp, 0, 0, SUN_SIZE,
+		    G_SLICE_HOT_ALLOW, G_SLICE_HOT_DENY, G_SLICE_HOT_CALL);
+		gsp->hot = g_sunlabel_hotwrite;
 	}
 	return (gp);
 }