Skip site navigation (1)Skip section navigation (2)
Date:      Wed, 22 Jun 2016 14:39:14 +0000 (UTC)
From:      "Kenneth D. Merry" <ken@FreeBSD.org>
To:        src-committers@freebsd.org, svn-src-all@freebsd.org, svn-src-head@freebsd.org
Subject:   svn commit: r302087 - head/sys/geom
Message-ID:  <201606221439.u5MEdE3o042149@repo.freebsd.org>

next in thread | raw e-mail | index | archive | help
Author: ken
Date: Wed Jun 22 14:39:13 2016
New Revision: 302087
URL: https://svnweb.freebsd.org/changeset/base/302087

Log:
  Fix a bug that caused da(4) peripheral drivers to not fully go away
  after the underlying device went away.
  
  The problem was that callers who queue the GEOM resize provider
  event didn't check to make sure that the provider had not been
  withered.  For the other equivalent case, g_new_provider_event(),
  the code checks to see whether the provider has been withered
  before queueing a g_new_provider_event() to the event thread.
  
  In some cases, a resize provider event would come through after
  the provider had been withered and all of the existing consumers
  had been orphaned.  When the resize event triggered a taste of
  the provider, that would attach a new consumer to the now
  withered provider.  The wither washer (g_wither_washer() would
  never be able to completely tear down the GEOM because of the
  consumers that were hanging around.
  
  The solution was to check the G_PF_WITHER provider flag before
  queueing the g_resize_provider_event(), and add an assert to
  g_resize_provider_event() to insure that it isn't called on a
  withered provider.
  
  sys/geom/geom_subr.c:
  	In g_resize_provider(), don't try to continue if the
  	G_PF_WITHER flag is set.
  
  	In g_resize_provider_event(), add an assert that the
  	G_PF_WITHER flag is not set.
  
  	In g_access(), if a provider has an error, print out the
  	name of the provider with the error.
  
  Sponsored by:	Spectra Logic
  Approved by:	re (marius)
  MFC after:	3 days

Modified:
  head/sys/geom/geom_subr.c

Modified: head/sys/geom/geom_subr.c
==============================================================================
--- head/sys/geom/geom_subr.c	Wed Jun 22 12:53:10 2016	(r302086)
+++ head/sys/geom/geom_subr.c	Wed Jun 22 14:39:13 2016	(r302087)
@@ -620,6 +620,8 @@ g_resize_provider_event(void *arg, int f
 	g_free(hh);
 
 	G_VALID_PROVIDER(pp);
+	KASSERT(!(pp->flags & G_PF_WITHER),
+	    ("g_resize_provider_event but withered"));
 	g_trace(G_T_TOPOLOGY, "g_resize_provider_event(%p)", pp);
 
 	LIST_FOREACH_SAFE(cp, &pp->consumers, consumers, cp2) {
@@ -662,6 +664,8 @@ g_resize_provider(struct g_provider *pp,
 	struct g_hh00 *hh;
 
 	G_VALID_PROVIDER(pp);
+	if (pp->flags & G_PF_WITHER)
+		return;
 
 	if (size == pp->mediasize)
 		return;
@@ -909,8 +913,11 @@ g_access(struct g_consumer *cp, int dcr,
 	else if (dcw > 0 && pe > 0)
 		return (EPERM);
 	/* If we try to open more but provider is error'ed: fail */
-	else if ((dcr > 0 || dcw > 0 || dce > 0) && pp->error != 0)
+	else if ((dcr > 0 || dcw > 0 || dce > 0) && pp->error != 0) {
+		printf("%s(%d): provider %s has error\n",
+		       __func__, __LINE__, pp->name);
 		return (pp->error);
+	}
 
 	/* Ok then... */
 



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?201606221439.u5MEdE3o042149>