d/freebsd-skq
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
d6c1921f69
freebsd-skq/sys/geom/geom_subr.c

1571 lines
38 KiB
C
Raw Normal View History

First commit of the GEOM subsystem to make it easier for people to test and play with this. This is not yet production quality and should be run only on dedicated test boxes. For people who want to develop transformations for GEOM there exist a set of shims to run geom in userland (ask phk@freebsd.org). Reports of all kinds to: phk@freebsd.org Please include in report: dmesg sysctl debug.geomdot sysctl debug.geomconf Known significant limitations: no kernel dump facility. ioctls severely restricted. Sponsored by: DARPA, NAI Labs
2002-03-11 21:42:35 +00:00
/*-
* Copyright (c) 2002 Poul-Henning Kamp
* Copyright (c) 2002 Networks Associates Technology, Inc.
* All rights reserved.
*
* This software was developed for the FreeBSD Project by Poul-Henning Kamp
* and NAI Labs, the Security Research Division of Network Associates, Inc.
* under DARPA/SPAWAR contract N66001-01-C-8035 ("CBOSS"), as part of the
* DARPA CHATS research program.
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions
* are met:
* 1. Redistributions of source code must retain the above copyright
* notice, this list of conditions and the following disclaimer.
* 2. Redistributions in binary form must reproduce the above copyright
* notice, this list of conditions and the following disclaimer in the
* documentation and/or other materials provided with the distribution.
* 3. The names of the authors may not be used to endorse or promote
* products derived from this software without specific prior written
* permission.
*
* THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND
* ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
* IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
* ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE
* FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
* DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
* OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
* HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
* LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
* OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
* SUCH DAMAGE.
*/
Use __FBSDID(). Approved by: phk
2003-06-11 06:49:16 +00:00
#include <sys/cdefs.h>
__FBSDID("$FreeBSD$");
First commit of the GEOM subsystem to make it easier for people to test and play with this. This is not yet production quality and should be run only on dedicated test boxes. For people who want to develop transformations for GEOM there exist a set of shims to run geom in userland (ask phk@freebsd.org). Reports of all kinds to: phk@freebsd.org Please include in report: dmesg sysctl debug.geomdot sysctl debug.geomconf Known significant limitations: no kernel dump facility. ioctls severely restricted. Sponsored by: DARPA, NAI Labs
2002-03-11 21:42:35 +00:00
Add 'show geom [addr]' ddb(4) command, which prints entire GEOM topology if no additional argument is given or details about the given GEOM object (class, geom, provider or consumer). Approved by: phk
2006-09-15 16:36:45 +00:00
#include "opt_ddb.h"
First commit of the GEOM subsystem to make it easier for people to test and play with this. This is not yet production quality and should be run only on dedicated test boxes. For people who want to develop transformations for GEOM there exist a set of shims to run geom in userland (ask phk@freebsd.org). Reports of all kinds to: phk@freebsd.org Please include in report: dmesg sysctl debug.geomdot sysctl debug.geomconf Known significant limitations: no kernel dump facility. ioctls severely restricted. Sponsored by: DARPA, NAI Labs
2002-03-11 21:42:35 +00:00
#include <sys/param.h>
#include <sys/systm.h>
Retire the GEOM private statistics code and use devstat instead.
2003-03-18 09:42:33 +00:00
#include <sys/devicestat.h>
First commit of the GEOM subsystem to make it easier for people to test and play with this. This is not yet production quality and should be run only on dedicated test boxes. For people who want to develop transformations for GEOM there exist a set of shims to run geom in userland (ask phk@freebsd.org). Reports of all kinds to: phk@freebsd.org Please include in report: dmesg sysctl debug.geomdot sysctl debug.geomconf Known significant limitations: no kernel dump facility. ioctls severely restricted. Sponsored by: DARPA, NAI Labs
2002-03-11 21:42:35 +00:00
#include <sys/kernel.h>
#include <sys/malloc.h>
#include <sys/bio.h>
#include <sys/sysctl.h>
#include <sys/proc.h>
#include <sys/kthread.h>
#include <sys/lock.h>
#include <sys/mutex.h>
#include <sys/errno.h>
#include <sys/sbuf.h>
#include <geom/geom.h>
Eliminate some thread pointers which do not make sense anymore. Split private parts of geom.h into geom_int.h. The latter should never be included in class implemtations.
2002-03-26 22:07:38 +00:00
#include <geom/geom_int.h>
First commit of the GEOM subsystem to make it easier for people to test and play with this. This is not yet production quality and should be run only on dedicated test boxes. For people who want to develop transformations for GEOM there exist a set of shims to run geom in userland (ask phk@freebsd.org). Reports of all kinds to: phk@freebsd.org Please include in report: dmesg sysctl debug.geomdot sysctl debug.geomconf Known significant limitations: no kernel dump facility. ioctls severely restricted. Sponsored by: DARPA, NAI Labs
2002-03-11 21:42:35 +00:00
#include <machine/stdarg.h>
Add 'show geom [addr]' ddb(4) command, which prints entire GEOM topology if no additional argument is given or details about the given GEOM object (class, geom, provider or consumer). Approved by: phk
2006-09-15 16:36:45 +00:00
#ifdef DDB
#include <ddb/ddb.h>
#endif
Fix ddb(4) "show geom addr" command when INVARIANTS is enabled. Don't assert that the topology lock is held when g_valid_obj() is called from debugger. MFC after: 1 week
2010-04-19 20:07:35 +00:00
#ifdef KDB
#include <sys/kdb.h>
#endif
s/classs/classes/ to fixup grammer after the previous global renaming. Sponsored by: DARPA & NAI Labs
2002-04-04 09:41:47 +00:00
struct class_list_head g_classes = LIST_HEAD_INITIALIZER(g_classes);
First commit of the GEOM subsystem to make it easier for people to test and play with this. This is not yet production quality and should be run only on dedicated test boxes. For people who want to develop transformations for GEOM there exist a set of shims to run geom in userland (ask phk@freebsd.org). Reports of all kinds to: phk@freebsd.org Please include in report: dmesg sysctl debug.geomdot sysctl debug.geomconf Known significant limitations: no kernel dump facility. ioctls severely restricted. Sponsored by: DARPA, NAI Labs
2002-03-11 21:42:35 +00:00
static struct g_tailq_head geoms = TAILQ_HEAD_INITIALIZER(geoms);
char *g_wait_event, *g_wait_up, *g_wait_down, *g_wait_sim;
Introduce a init and fini member functions on a class. Use ->init() and ->fini() to handle the mutex in geom_disk.c Remove the g_add_class() function and replace it with a standardized g_modevent() function. This adds the basic infrastructure for loading/unloading GEOM classes
2003-05-31 18:13:07 +00:00
struct g_hh00 {
Add a new GEOM method, resize(), which is called after provider size changes. Add a new routine, g_resize_provider(), to use to notify GEOM about provider change. Reviewed by: mav Sponsored by: FreeBSD Foundation
2012-07-07 20:13:40 +00:00
struct g_class *mp;
struct g_provider *pp;
off_t size;
int error;
int post;
Introduce a init and fini member functions on a class. Use ->init() and ->fini() to handle the mutex in geom_disk.c Remove the g_add_class() function and replace it with a standardized g_modevent() function. This adds the basic infrastructure for loading/unloading GEOM classes
2003-05-31 18:13:07 +00:00
};
Turn the hardwired NEW_CLASS event into a g_call_me() event.
2003-04-23 19:34:38 +00:00
/*
* This event offers a new class a chance to taste all preexisting providers.
*/
static void
Introduce a init and fini member functions on a class. Use ->init() and ->fini() to handle the mutex in geom_disk.c Remove the g_add_class() function and replace it with a standardized g_modevent() function. This adds the basic infrastructure for loading/unloading GEOM classes
2003-05-31 18:13:07 +00:00
g_load_class(void *arg, int flag)
Turn the hardwired NEW_CLASS event into a g_call_me() event.
2003-04-23 19:34:38 +00:00
{
Introduce a init and fini member functions on a class. Use ->init() and ->fini() to handle the mutex in geom_disk.c Remove the g_add_class() function and replace it with a standardized g_modevent() function. This adds the basic infrastructure for loading/unloading GEOM classes
2003-05-31 18:13:07 +00:00
struct g_hh00 *hh;
Turn the hardwired NEW_CLASS event into a g_call_me() event.
2003-04-23 19:34:38 +00:00
struct g_class *mp2, *mp;
struct g_geom *gp;
struct g_provider *pp;
g_topology_assert();
Introduce a init and fini member functions on a class. Use ->init() and ->fini() to handle the mutex in geom_disk.c Remove the g_add_class() function and replace it with a standardized g_modevent() function. This adds the basic infrastructure for loading/unloading GEOM classes
2003-05-31 18:13:07 +00:00
if (flag == EV_CANCEL) /* XXX: can't happen ? */
Turn the hardwired NEW_CLASS event into a g_call_me() event.
2003-04-23 19:34:38 +00:00
return;
if (g_shutdown)
return;
Introduce a init and fini member functions on a class. Use ->init() and ->fini() to handle the mutex in geom_disk.c Remove the g_add_class() function and replace it with a standardized g_modevent() function. This adds the basic infrastructure for loading/unloading GEOM classes
2003-05-31 18:13:07 +00:00
hh = arg;
mp = hh->mp;
fix a "modify after free" bug which is practically impossible to experience. Found by: Coverity (id #540 #541)
2005-03-26 21:07:35 +00:00
hh->error = 0;
if (hh->post) {
- Turn KASSERT()s into warning printf()'s in the g_class_load() routine. This removes a panic that will occur if you build with GENERIC and attempt to kldload a GEOM module that is already in the kernel. Reviewed by: phk
2004-10-22 22:16:24 +00:00
g_free(hh);
fix a "modify after free" bug which is practically impossible to experience. Found by: Coverity (id #540 #541)
2005-03-26 21:07:35 +00:00
hh = NULL;
}
Introduce a init and fini member functions on a class. Use ->init() and ->fini() to handle the mutex in geom_disk.c Remove the g_add_class() function and replace it with a standardized g_modevent() function. This adds the basic infrastructure for loading/unloading GEOM classes
2003-05-31 18:13:07 +00:00
g_trace(G_T_TOPOLOGY, "g_load_class(%s)", mp->name);
Add KASSERTS. Submitted by: Pawel Jakub Dawidek <nick@garage.freebsd.pl>
2004-01-23 21:02:49 +00:00
KASSERT(mp->name != NULL && *mp->name != '\0',
("GEOM class has no name"));
Add a KASSERT to prevent the same GEOM class from being processed loaded twice. Enforce that classes should have different names while we are here.
2003-06-04 17:51:10 +00:00
LIST_FOREACH(mp2, &g_classes, class) {
- Turn KASSERT()s into warning printf()'s in the g_class_load() routine. This removes a panic that will occur if you build with GENERIC and attempt to kldload a GEOM module that is already in the kernel. Reviewed by: phk
2004-10-22 22:16:24 +00:00
if (mp2 == mp) {
printf("The GEOM class %s is already loaded.\n",
mp2->name);
fix a "modify after free" bug which is practically impossible to experience. Found by: Coverity (id #540 #541)
2005-03-26 21:07:35 +00:00
if (hh != NULL)
hh->error = EEXIST;
- Turn KASSERT()s into warning printf()'s in the g_class_load() routine. This removes a panic that will occur if you build with GENERIC and attempt to kldload a GEOM module that is already in the kernel. Reviewed by: phk
2004-10-22 22:16:24 +00:00
return;
} else if (strcmp(mp2->name, mp->name) == 0) {
printf("A GEOM class %s is already loaded.\n",
mp2->name);
fix a "modify after free" bug which is practically impossible to experience. Found by: Coverity (id #540 #541)
2005-03-26 21:07:35 +00:00
if (hh != NULL)
hh->error = EEXIST;
- Turn KASSERT()s into warning printf()'s in the g_class_load() routine. This removes a panic that will occur if you build with GENERIC and attempt to kldload a GEOM module that is already in the kernel. Reviewed by: phk
2004-10-22 22:16:24 +00:00
return;
}
Add a KASSERT to prevent the same GEOM class from being processed loaded twice. Enforce that classes should have different names while we are here.
2003-06-04 17:51:10 +00:00
}
Introduce a init and fini member functions on a class. Use ->init() and ->fini() to handle the mutex in geom_disk.c Remove the g_add_class() function and replace it with a standardized g_modevent() function. This adds the basic infrastructure for loading/unloading GEOM classes
2003-05-31 18:13:07 +00:00
LIST_INIT(&mp->geom);
LIST_INSERT_HEAD(&g_classes, mp, class);
Call class->init() an class->fini() while the class is hooked up, rather than right before and right after. This allows these routines to manipulate the mesh. KASSERT that nobody creates a geom on an alien class. Assert topology in g_valid_obj(). Approved by: re@
2003-11-18 18:17:39 +00:00
if (mp->init != NULL)
mp->init(mp);
Introduce a init and fini member functions on a class. Use ->init() and ->fini() to handle the mutex in geom_disk.c Remove the g_add_class() function and replace it with a standardized g_modevent() function. This adds the basic infrastructure for loading/unloading GEOM classes
2003-05-31 18:13:07 +00:00
if (mp->taste == NULL)
Turn the hardwired NEW_CLASS event into a g_call_me() event.
2003-04-23 19:34:38 +00:00
return;
Introduce a init and fini member functions on a class. Use ->init() and ->fini() to handle the mutex in geom_disk.c Remove the g_add_class() function and replace it with a standardized g_modevent() function. This adds the basic infrastructure for loading/unloading GEOM classes
2003-05-31 18:13:07 +00:00
LIST_FOREACH(mp2, &g_classes, class) {
if (mp == mp2)
Turn the hardwired NEW_CLASS event into a g_call_me() event.
2003-04-23 19:34:38 +00:00
continue;
Introduce a init and fini member functions on a class. Use ->init() and ->fini() to handle the mutex in geom_disk.c Remove the g_add_class() function and replace it with a standardized g_modevent() function. This adds the basic infrastructure for loading/unloading GEOM classes
2003-05-31 18:13:07 +00:00
LIST_FOREACH(gp, &mp2->geom, geom) {
Turn the hardwired NEW_CLASS event into a g_call_me() event.
2003-04-23 19:34:38 +00:00
LIST_FOREACH(pp, &gp->provider, provider) {
Introduce a init and fini member functions on a class. Use ->init() and ->fini() to handle the mutex in geom_disk.c Remove the g_add_class() function and replace it with a standardized g_modevent() function. This adds the basic infrastructure for loading/unloading GEOM classes
2003-05-31 18:13:07 +00:00
mp->taste(mp, pp, 0);
Turn the hardwired NEW_CLASS event into a g_call_me() event.
2003-04-23 19:34:38 +00:00
g_topology_assert();
}
}
}
}
Fix deadlock between GEOM class unloading and withering. Withering can't proceed while g_unload_class() blocks the event thread. Fix this by not running g_unload_class() as a GEOM event and dropping the topology lock when withering needs to proceed. PR: kern/139847 Silence on: freebsd-geom
2010-05-05 18:53:24 +00:00
static int
g_unload_class(struct g_class *mp)
First commit of the GEOM subsystem to make it easier for people to test and play with this. This is not yet production quality and should be run only on dedicated test boxes. For people who want to develop transformations for GEOM there exist a set of shims to run geom in userland (ask phk@freebsd.org). Reports of all kinds to: phk@freebsd.org Please include in report: dmesg sysctl debug.geomdot sysctl debug.geomconf Known significant limitations: no kernel dump facility. ioctls severely restricted. Sponsored by: DARPA, NAI Labs
2002-03-11 21:42:35 +00:00
{
Introduce a init and fini member functions on a class. Use ->init() and ->fini() to handle the mutex in geom_disk.c Remove the g_add_class() function and replace it with a standardized g_modevent() function. This adds the basic infrastructure for loading/unloading GEOM classes
2003-05-31 18:13:07 +00:00
struct g_geom *gp;
struct g_provider *pp;
struct g_consumer *cp;
int error;
Fix deadlock between GEOM class unloading and withering. Withering can't proceed while g_unload_class() blocks the event thread. Fix this by not running g_unload_class() as a GEOM event and dropping the topology lock when withering needs to proceed. PR: kern/139847 Silence on: freebsd-geom
2010-05-05 18:53:24 +00:00
g_topology_lock();
Introduce a init and fini member functions on a class. Use ->init() and ->fini() to handle the mutex in geom_disk.c Remove the g_add_class() function and replace it with a standardized g_modevent() function. This adds the basic infrastructure for loading/unloading GEOM classes
2003-05-31 18:13:07 +00:00
g_trace(G_T_TOPOLOGY, "g_unload_class(%s)", mp->name);
Fix deadlock between GEOM class unloading and withering. Withering can't proceed while g_unload_class() blocks the event thread. Fix this by not running g_unload_class() as a GEOM event and dropping the topology lock when withering needs to proceed. PR: kern/139847 Silence on: freebsd-geom
2010-05-05 18:53:24 +00:00
retry:
G_VALID_CLASS(mp);
Introduce a init and fini member functions on a class. Use ->init() and ->fini() to handle the mutex in geom_disk.c Remove the g_add_class() function and replace it with a standardized g_modevent() function. This adds the basic infrastructure for loading/unloading GEOM classes
2003-05-31 18:13:07 +00:00
LIST_FOREACH(gp, &mp->geom, geom) {
Fix deadlock between GEOM class unloading and withering. Withering can't proceed while g_unload_class() blocks the event thread. Fix this by not running g_unload_class() as a GEOM event and dropping the topology lock when withering needs to proceed. PR: kern/139847 Silence on: freebsd-geom
2010-05-05 18:53:24 +00:00
/* We refuse to unload if anything is open */
Introduce a init and fini member functions on a class. Use ->init() and ->fini() to handle the mutex in geom_disk.c Remove the g_add_class() function and replace it with a standardized g_modevent() function. This adds the basic infrastructure for loading/unloading GEOM classes
2003-05-31 18:13:07 +00:00
LIST_FOREACH(pp, &gp->provider, provider)
if (pp->acr || pp->acw || pp->ace) {
Fix deadlock between GEOM class unloading and withering. Withering can't proceed while g_unload_class() blocks the event thread. Fix this by not running g_unload_class() as a GEOM event and dropping the topology lock when withering needs to proceed. PR: kern/139847 Silence on: freebsd-geom
2010-05-05 18:53:24 +00:00
g_topology_unlock();
return (EBUSY);
Introduce a init and fini member functions on a class. Use ->init() and ->fini() to handle the mutex in geom_disk.c Remove the g_add_class() function and replace it with a standardized g_modevent() function. This adds the basic infrastructure for loading/unloading GEOM classes
2003-05-31 18:13:07 +00:00
}
LIST_FOREACH(cp, &gp->consumer, consumer)
if (cp->acr || cp->acw || cp->ace) {
Fix deadlock between GEOM class unloading and withering. Withering can't proceed while g_unload_class() blocks the event thread. Fix this by not running g_unload_class() as a GEOM event and dropping the topology lock when withering needs to proceed. PR: kern/139847 Silence on: freebsd-geom
2010-05-05 18:53:24 +00:00
g_topology_unlock();
return (EBUSY);
Introduce a init and fini member functions on a class. Use ->init() and ->fini() to handle the mutex in geom_disk.c Remove the g_add_class() function and replace it with a standardized g_modevent() function. This adds the basic infrastructure for loading/unloading GEOM classes
2003-05-31 18:13:07 +00:00
}
Fix deadlock between GEOM class unloading and withering. Withering can't proceed while g_unload_class() blocks the event thread. Fix this by not running g_unload_class() as a GEOM event and dropping the topology lock when withering needs to proceed. PR: kern/139847 Silence on: freebsd-geom
2010-05-05 18:53:24 +00:00
/* If the geom is withering, wait for it to finish. */
if (gp->flags & G_GEOM_WITHER) {
g_topology_sleep(mp, 1);
goto retry;
}
}
/*
* We allow unloading if we have no geoms, or a class
* method we can use to get rid of them.
*/
if (!LIST_EMPTY(&mp->geom) && mp->destroy_geom == NULL) {
g_topology_unlock();
return (EOPNOTSUPP);
Introduce a init and fini member functions on a class. Use ->init() and ->fini() to handle the mutex in geom_disk.c Remove the g_add_class() function and replace it with a standardized g_modevent() function. This adds the basic infrastructure for loading/unloading GEOM classes
2003-05-31 18:13:07 +00:00
}
/* Bar new entries */
mp->taste = NULL;
mp->config = NULL;
Fix deadlock between GEOM class unloading and withering. Withering can't proceed while g_unload_class() blocks the event thread. Fix this by not running g_unload_class() as a GEOM event and dropping the topology lock when withering needs to proceed. PR: kern/139847 Silence on: freebsd-geom
2010-05-05 18:53:24 +00:00
LIST_FOREACH(gp, &mp->geom, geom) {
error = mp->destroy_geom(NULL, mp, gp);
if (error != 0) {
g_topology_unlock();
return (error);
}
}
/* Wait for withering to finish. */
Be more insistent on destroying geoms at unload time. Still not perfect, but it will do (better) for now. KASSERT that to have providers a class must have an access method. Tag the new_provider event with the geom as well.
2004-03-11 08:16:23 +00:00
for (;;) {
gp = LIST_FIRST(&mp->geom);
if (gp == NULL)
break;
Fix deadlock between GEOM class unloading and withering. Withering can't proceed while g_unload_class() blocks the event thread. Fix this by not running g_unload_class() as a GEOM event and dropping the topology lock when withering needs to proceed. PR: kern/139847 Silence on: freebsd-geom
2010-05-05 18:53:24 +00:00
KASSERT(gp->flags & G_GEOM_WITHER,
("Non-withering geom in class %s", mp->name));
g_topology_sleep(mp, 1);
Introduce a init and fini member functions on a class. Use ->init() and ->fini() to handle the mutex in geom_disk.c Remove the g_add_class() function and replace it with a standardized g_modevent() function. This adds the basic infrastructure for loading/unloading GEOM classes
2003-05-31 18:13:07 +00:00
}
Fix deadlock between GEOM class unloading and withering. Withering can't proceed while g_unload_class() blocks the event thread. Fix this by not running g_unload_class() as a GEOM event and dropping the topology lock when withering needs to proceed. PR: kern/139847 Silence on: freebsd-geom
2010-05-05 18:53:24 +00:00
G_VALID_CLASS(mp);
if (mp->fini != NULL)
mp->fini(mp);
LIST_REMOVE(mp, class);
g_topology_unlock();
return (0);
Introduce a init and fini member functions on a class. Use ->init() and ->fini() to handle the mutex in geom_disk.c Remove the g_add_class() function and replace it with a standardized g_modevent() function. This adds the basic infrastructure for loading/unloading GEOM classes
2003-05-31 18:13:07 +00:00
}
int
g_modevent(module_t mod, int type, void *data)
{
struct g_hh00 *hh;
int error;
static int g_ignition;
Give classes a version number and refuse to touch classes which are not understood. This makes room for additional binary compatibility in the future. Put fields in the class for the geom's methods and initialize the methods of a new geom from these fields. This saves some code in all classes.
2004-08-08 06:46:27 +00:00
struct g_class *mp;
First commit of the GEOM subsystem to make it easier for people to test and play with this. This is not yet production quality and should be run only on dedicated test boxes. For people who want to develop transformations for GEOM there exist a set of shims to run geom in userland (ask phk@freebsd.org). Reports of all kinds to: phk@freebsd.org Please include in report: dmesg sysctl debug.geomdot sysctl debug.geomconf Known significant limitations: no kernel dump facility. ioctls severely restricted. Sponsored by: DARPA, NAI Labs
2002-03-11 21:42:35 +00:00
Give classes a version number and refuse to touch classes which are not understood. This makes room for additional binary compatibility in the future. Put fields in the class for the geom's methods and initialize the methods of a new geom from these fields. This saves some code in all classes.
2004-08-08 06:46:27 +00:00
mp = data;
OK, now check geom class version numbers.
2004-08-08 08:34:46 +00:00
if (mp->version != G_VERSION) {
Give classes a version number and refuse to touch classes which are not understood. This makes room for additional binary compatibility in the future. Put fields in the class for the geom's methods and initialize the methods of a new geom from these fields. This saves some code in all classes.
2004-08-08 06:46:27 +00:00
printf("GEOM class %s has Wrong version %x\n",
mp->name, mp->version);
return (EINVAL);
}
First commit of the GEOM subsystem to make it easier for people to test and play with this. This is not yet production quality and should be run only on dedicated test boxes. For people who want to develop transformations for GEOM there exist a set of shims to run geom in userland (ask phk@freebsd.org). Reports of all kinds to: phk@freebsd.org Please include in report: dmesg sysctl debug.geomdot sysctl debug.geomconf Known significant limitations: no kernel dump facility. ioctls severely restricted. Sponsored by: DARPA, NAI Labs
2002-03-11 21:42:35 +00:00
if (!g_ignition) {
g_ignition++;
g_init();
}
Introduce a init and fini member functions on a class. Use ->init() and ->fini() to handle the mutex in geom_disk.c Remove the g_add_class() function and replace it with a standardized g_modevent() function. This adds the basic infrastructure for loading/unloading GEOM classes
2003-05-31 18:13:07 +00:00
error = EOPNOTSUPP;
switch (type) {
case MOD_LOAD:
Fix deadlock between GEOM class unloading and withering. Withering can't proceed while g_unload_class() blocks the event thread. Fix this by not running g_unload_class() as a GEOM event and dropping the topology lock when withering needs to proceed. PR: kern/139847 Silence on: freebsd-geom
2010-05-05 18:53:24 +00:00
g_trace(G_T_TOPOLOGY, "g_modevent(%s, LOAD)", mp->name);
hh = g_malloc(sizeof *hh, M_WAITOK | M_ZERO);
hh->mp = mp;
When loading GEOM modules, we expect the actual load process to be done by the time that kldload(8) returns. Satisfy that by making the GEOM module load event -- only when the kernel is !cold -- wait until the GEOM module init function has finished instead of returning immediately. This is the other half of fixing md(8) (actually, "mfs" in fstab(5)) that is similar to r1.128 of src/sys/dev/md/md.c. This bug would be why RAM disks would often fail on boot and the first call to mdconfig(8) would probably fail. pjd has ideas for not requiring kldload(8) to work synchronously for control devices that could make this obsolete. Silence on: -arch
2004-10-12 04:44:54 +00:00
/*
* Once the system is not cold, MOD_LOAD calls will be
* from the userland and the g_event thread will be able
* to acknowledge their completion.
*/
- Turn KASSERT()s into warning printf()'s in the g_class_load() routine. This removes a panic that will occur if you build with GENERIC and attempt to kldload a GEOM module that is already in the kernel. Reviewed by: phk
2004-10-22 22:16:24 +00:00
if (cold) {
hh->post = 1;
error = g_post_event(g_load_class, hh, M_WAITOK, NULL);
} else {
error = g_waitfor_event(g_load_class, hh, M_WAITOK,
NULL);
if (error == 0)
error = hh->error;
g_free(hh);
}
Introduce a init and fini member functions on a class. Use ->init() and ->fini() to handle the mutex in geom_disk.c Remove the g_add_class() function and replace it with a standardized g_modevent() function. This adds the basic infrastructure for loading/unloading GEOM classes
2003-05-31 18:13:07 +00:00
break;
case MOD_UNLOAD:
Fix deadlock between GEOM class unloading and withering. Withering can't proceed while g_unload_class() blocks the event thread. Fix this by not running g_unload_class() as a GEOM event and dropping the topology lock when withering needs to proceed. PR: kern/139847 Silence on: freebsd-geom
2010-05-05 18:53:24 +00:00
g_trace(G_T_TOPOLOGY, "g_modevent(%s, UNLOAD)", mp->name);
error = g_unload_class(mp);
Don't panic over the fact that unloading failed if we already knew that.
2003-08-22 11:00:54 +00:00
if (error == 0) {
Fix deadlock between GEOM class unloading and withering. Withering can't proceed while g_unload_class() blocks the event thread. Fix this by not running g_unload_class() as a GEOM event and dropping the topology lock when withering needs to proceed. PR: kern/139847 Silence on: freebsd-geom
2010-05-05 18:53:24 +00:00
KASSERT(LIST_EMPTY(&mp->geom),
("Unloaded class (%s) still has geom", mp->name));
Don't panic over the fact that unloading failed if we already knew that.
2003-08-22 11:00:54 +00:00
}
Plug an insignificant memoryleak. Submitted by: Pawel Jakub Dawidek <nick@garage.freebsd.pl>
2004-01-23 20:40:25 +00:00
break;
Introduce a init and fini member functions on a class. Use ->init() and ->fini() to handle the mutex in geom_disk.c Remove the g_add_class() function and replace it with a standardized g_modevent() function. This adds the basic infrastructure for loading/unloading GEOM classes
2003-05-31 18:13:07 +00:00
}
return (error);
First commit of the GEOM subsystem to make it easier for people to test and play with this. This is not yet production quality and should be run only on dedicated test boxes. For people who want to develop transformations for GEOM there exist a set of shims to run geom in userland (ask phk@freebsd.org). Reports of all kinds to: phk@freebsd.org Please include in report: dmesg sysctl debug.geomdot sysctl debug.geomconf Known significant limitations: no kernel dump facility. ioctls severely restricted. Sponsored by: DARPA, NAI Labs
2002-03-11 21:42:35 +00:00
}
Add g_retaste(), which given a class will present all non-open providers to it for tasting. This is useful when the class, through means outside the scope of GEOM, can claim providers previously unclaimed. The g_retaste() function posts an event which is handled by the g_retaste_event(). Event suggested by: phk
2008-03-23 01:23:35 +00:00
static void
g_retaste_event(void *arg, int flag)
{
Implement media change notification for DA and CD removable media devices. It includes three parts: 1) Modifications to CAM to detect media media changes and report them to disk(9) layer. For modern SATA (and potentially UAS) devices it utilizes Asynchronous Notification mechanism to receive events from hardware. Active polling with TEST UNIT READY commands with 3 seconds period is used for incapable hardware. After that both CD and DA drivers work the same way, detecting two conditions: "NOT READY: Medium not present" after medium was detected previously, and "UNIT ATTENTION: Not ready to ready change, medium may have changed". First one reported to disk(9) as media removal, second as media insert/change. To reliably receive second event new AC_UNIT_ATTENTION async added to make UAs broadcasted to all periphs by generic error handling code in cam_periph_error(). 2) Modifications to GEOM core to handle media remove and change events. Media removal handled by spoiling all consumers attached to the provider. Media change event also schedules provider retaste after spoiling to probe new media. New flag G_CF_ORPHAN was added to consumers to reflect that consumer is in process of destruction. It allows retaste to create new geom instance of the same class, while previous one is still dying. 3) Modifications to some GEOM classes: DEV -- to report media change events to devd; VFS -- to handle spoiling same as orphan to prevent accessing replaced media. PART class already handles spoiling alike to orphan. Reviewed by: silence on geom@ and scsi@ Tested by: avg Sponsored by: iXsystems, Inc. / PC-BSD MFC after: 2 months
2012-07-29 11:51:48 +00:00
struct g_class *mp, *mp2;
struct g_geom *gp;
Add g_retaste(), which given a class will present all non-open providers to it for tasting. This is useful when the class, through means outside the scope of GEOM, can claim providers previously unclaimed. The g_retaste() function posts an event which is handled by the g_retaste_event(). Event suggested by: phk
2008-03-23 01:23:35 +00:00
struct g_hh00 *hh;
struct g_provider *pp;
Implement media change notification for DA and CD removable media devices. It includes three parts: 1) Modifications to CAM to detect media media changes and report them to disk(9) layer. For modern SATA (and potentially UAS) devices it utilizes Asynchronous Notification mechanism to receive events from hardware. Active polling with TEST UNIT READY commands with 3 seconds period is used for incapable hardware. After that both CD and DA drivers work the same way, detecting two conditions: "NOT READY: Medium not present" after medium was detected previously, and "UNIT ATTENTION: Not ready to ready change, medium may have changed". First one reported to disk(9) as media removal, second as media insert/change. To reliably receive second event new AC_UNIT_ATTENTION async added to make UAs broadcasted to all periphs by generic error handling code in cam_periph_error(). 2) Modifications to GEOM core to handle media remove and change events. Media removal handled by spoiling all consumers attached to the provider. Media change event also schedules provider retaste after spoiling to probe new media. New flag G_CF_ORPHAN was added to consumers to reflect that consumer is in process of destruction. It allows retaste to create new geom instance of the same class, while previous one is still dying. 3) Modifications to some GEOM classes: DEV -- to report media change events to devd; VFS -- to handle spoiling same as orphan to prevent accessing replaced media. PART class already handles spoiling alike to orphan. Reviewed by: silence on geom@ and scsi@ Tested by: avg Sponsored by: iXsystems, Inc. / PC-BSD MFC after: 2 months
2012-07-29 11:51:48 +00:00
struct g_consumer *cp;
Add g_retaste(), which given a class will present all non-open providers to it for tasting. This is useful when the class, through means outside the scope of GEOM, can claim providers previously unclaimed. The g_retaste() function posts an event which is handled by the g_retaste_event(). Event suggested by: phk
2008-03-23 01:23:35 +00:00
g_topology_assert();
if (flag == EV_CANCEL) /* XXX: can't happen ? */
return;
Introduce a kern.geom.notaste sysctl that can be used to temporarily disable GEOM tasting to avoid the "bouncing GEOM" problem where, when you shut down the consumer of a provider which can be viewed in multiple ways (typically a mirror whose members are labeled partitions), GEOM will immediately taste that provider's alter ego and reattach the consumer. Approved by: re (glebius)
2013-09-24 20:05:16 +00:00
if (g_shutdown || g_notaste)
Add g_retaste(), which given a class will present all non-open providers to it for tasting. This is useful when the class, through means outside the scope of GEOM, can claim providers previously unclaimed. The g_retaste() function posts an event which is handled by the g_retaste_event(). Event suggested by: phk
2008-03-23 01:23:35 +00:00
return;
hh = arg;
mp = hh->mp;
hh->error = 0;
if (hh->post) {
g_free(hh);
hh = NULL;
}
g_trace(G_T_TOPOLOGY, "g_retaste(%s)", mp->name);
Implement media change notification for DA and CD removable media devices. It includes three parts: 1) Modifications to CAM to detect media media changes and report them to disk(9) layer. For modern SATA (and potentially UAS) devices it utilizes Asynchronous Notification mechanism to receive events from hardware. Active polling with TEST UNIT READY commands with 3 seconds period is used for incapable hardware. After that both CD and DA drivers work the same way, detecting two conditions: "NOT READY: Medium not present" after medium was detected previously, and "UNIT ATTENTION: Not ready to ready change, medium may have changed". First one reported to disk(9) as media removal, second as media insert/change. To reliably receive second event new AC_UNIT_ATTENTION async added to make UAs broadcasted to all periphs by generic error handling code in cam_periph_error(). 2) Modifications to GEOM core to handle media remove and change events. Media removal handled by spoiling all consumers attached to the provider. Media change event also schedules provider retaste after spoiling to probe new media. New flag G_CF_ORPHAN was added to consumers to reflect that consumer is in process of destruction. It allows retaste to create new geom instance of the same class, while previous one is still dying. 3) Modifications to some GEOM classes: DEV -- to report media change events to devd; VFS -- to handle spoiling same as orphan to prevent accessing replaced media. PART class already handles spoiling alike to orphan. Reviewed by: silence on geom@ and scsi@ Tested by: avg Sponsored by: iXsystems, Inc. / PC-BSD MFC after: 2 months
2012-07-29 11:51:48 +00:00
LIST_FOREACH(mp2, &g_classes, class) {
LIST_FOREACH(gp, &mp2->geom, geom) {
Add g_retaste(), which given a class will present all non-open providers to it for tasting. This is useful when the class, through means outside the scope of GEOM, can claim providers previously unclaimed. The g_retaste() function posts an event which is handled by the g_retaste_event(). Event suggested by: phk
2008-03-23 01:23:35 +00:00
LIST_FOREACH(pp, &gp->provider, provider) {
if (pp->acr || pp->acw || pp->ace)
continue;
Implement media change notification for DA and CD removable media devices. It includes three parts: 1) Modifications to CAM to detect media media changes and report them to disk(9) layer. For modern SATA (and potentially UAS) devices it utilizes Asynchronous Notification mechanism to receive events from hardware. Active polling with TEST UNIT READY commands with 3 seconds period is used for incapable hardware. After that both CD and DA drivers work the same way, detecting two conditions: "NOT READY: Medium not present" after medium was detected previously, and "UNIT ATTENTION: Not ready to ready change, medium may have changed". First one reported to disk(9) as media removal, second as media insert/change. To reliably receive second event new AC_UNIT_ATTENTION async added to make UAs broadcasted to all periphs by generic error handling code in cam_periph_error(). 2) Modifications to GEOM core to handle media remove and change events. Media removal handled by spoiling all consumers attached to the provider. Media change event also schedules provider retaste after spoiling to probe new media. New flag G_CF_ORPHAN was added to consumers to reflect that consumer is in process of destruction. It allows retaste to create new geom instance of the same class, while previous one is still dying. 3) Modifications to some GEOM classes: DEV -- to report media change events to devd; VFS -- to handle spoiling same as orphan to prevent accessing replaced media. PART class already handles spoiling alike to orphan. Reviewed by: silence on geom@ and scsi@ Tested by: avg Sponsored by: iXsystems, Inc. / PC-BSD MFC after: 2 months
2012-07-29 11:51:48 +00:00
LIST_FOREACH(cp, &pp->consumers, consumers) {
if (cp->geom->class == mp &&
(cp->flags & G_CF_ORPHAN) == 0)
When retasting, wither any existing GEOMs of the same class. This allows the class to create a different GEOM for the same provider as well as avoid that we end up with multiple GEOMs of the same class with the same name. For example, when a disk contains a PC98 partition table but only MBR is supported, then the partition table can be treated as a MBR. If support for PC98 is later loaded as a module, the MBR scheme is pre-empted for the PC98 scheme as expected.
2008-03-28 06:31:12 +00:00
break;
}
Implement media change notification for DA and CD removable media devices. It includes three parts: 1) Modifications to CAM to detect media media changes and report them to disk(9) layer. For modern SATA (and potentially UAS) devices it utilizes Asynchronous Notification mechanism to receive events from hardware. Active polling with TEST UNIT READY commands with 3 seconds period is used for incapable hardware. After that both CD and DA drivers work the same way, detecting two conditions: "NOT READY: Medium not present" after medium was detected previously, and "UNIT ATTENTION: Not ready to ready change, medium may have changed". First one reported to disk(9) as media removal, second as media insert/change. To reliably receive second event new AC_UNIT_ATTENTION async added to make UAs broadcasted to all periphs by generic error handling code in cam_periph_error(). 2) Modifications to GEOM core to handle media remove and change events. Media removal handled by spoiling all consumers attached to the provider. Media change event also schedules provider retaste after spoiling to probe new media. New flag G_CF_ORPHAN was added to consumers to reflect that consumer is in process of destruction. It allows retaste to create new geom instance of the same class, while previous one is still dying. 3) Modifications to some GEOM classes: DEV -- to report media change events to devd; VFS -- to handle spoiling same as orphan to prevent accessing replaced media. PART class already handles spoiling alike to orphan. Reviewed by: silence on geom@ and scsi@ Tested by: avg Sponsored by: iXsystems, Inc. / PC-BSD MFC after: 2 months
2012-07-29 11:51:48 +00:00
if (cp != NULL) {
cp->flags |= G_CF_ORPHAN;
g_wither_geom(cp->geom, ENXIO);
}
Add g_retaste(), which given a class will present all non-open providers to it for tasting. This is useful when the class, through means outside the scope of GEOM, can claim providers previously unclaimed. The g_retaste() function posts an event which is handled by the g_retaste_event(). Event suggested by: phk
2008-03-23 01:23:35 +00:00
mp->taste(mp, pp, 0);
g_topology_assert();
}
}
}
}
int
g_retaste(struct g_class *mp)
{
struct g_hh00 *hh;
int error;
if (mp->taste == NULL)
return (EINVAL);
hh = g_malloc(sizeof *hh, M_WAITOK | M_ZERO);
hh->mp = mp;
if (cold) {
hh->post = 1;
error = g_post_event(g_retaste_event, hh, M_WAITOK, NULL);
} else {
error = g_waitfor_event(g_retaste_event, hh, M_WAITOK, NULL);
if (error == 0)
error = hh->error;
g_free(hh);
}
return (error);
}
First commit of the GEOM subsystem to make it easier for people to test and play with this. This is not yet production quality and should be run only on dedicated test boxes. For people who want to develop transformations for GEOM there exist a set of shims to run geom in userland (ask phk@freebsd.org). Reports of all kinds to: phk@freebsd.org Please include in report: dmesg sysctl debug.geomdot sysctl debug.geomconf Known significant limitations: no kernel dump facility. ioctls severely restricted. Sponsored by: DARPA, NAI Labs
2002-03-11 21:42:35 +00:00
struct g_geom *
Constification and some s/int/u_int/ changes.
2002-12-16 22:33:27 +00:00
g_new_geomf(struct g_class *mp, const char *fmt, ...)
First commit of the GEOM subsystem to make it easier for people to test and play with this. This is not yet production quality and should be run only on dedicated test boxes. For people who want to develop transformations for GEOM there exist a set of shims to run geom in userland (ask phk@freebsd.org). Reports of all kinds to: phk@freebsd.org Please include in report: dmesg sysctl debug.geomdot sysctl debug.geomconf Known significant limitations: no kernel dump facility. ioctls severely restricted. Sponsored by: DARPA, NAI Labs
2002-03-11 21:42:35 +00:00
{
struct g_geom *gp;
va_list ap;
struct sbuf *sb;
g_topology_assert();
Rearrange some of the GEOM debugging tools to be more structured. Retire g_sanity() and corresponding debugflag (0x8) Retire g_{stall,release}_events(). Under #ifdef DIAGNOSTIC: Make g_valid_obj() an official function and have it return an an non-zero integer which indicates the kind of object when found. Implement G_VALID_{CLASS,GEOM,CONSUMER,PROVIDER}() macros based on g_valid_obj(). Sprinkle calls to these macros liberally over the infrastructure. Always check that we do not free a live object.
2004-03-10 08:49:08 +00:00
G_VALID_CLASS(mp);
Add sbuf_new_auto as a shortcut for the very common case of creating a completely dynamic sbuf. Obtained from: Varnish MFC after: 2 weeks
2008-08-09 11:14:05 +00:00
sb = sbuf_new_auto();
Add missing va_end() calls. Noticed by: tmm
2003-06-07 10:16:53 +00:00
va_start(ap, fmt);
First commit of the GEOM subsystem to make it easier for people to test and play with this. This is not yet production quality and should be run only on dedicated test boxes. For people who want to develop transformations for GEOM there exist a set of shims to run geom in userland (ask phk@freebsd.org). Reports of all kinds to: phk@freebsd.org Please include in report: dmesg sysctl debug.geomdot sysctl debug.geomconf Known significant limitations: no kernel dump facility. ioctls severely restricted. Sponsored by: DARPA, NAI Labs
2002-03-11 21:42:35 +00:00
sbuf_vprintf(sb, fmt, ap);
Add missing va_end() calls. Noticed by: tmm
2003-06-07 10:16:53 +00:00
va_end(ap);
First commit of the GEOM subsystem to make it easier for people to test and play with this. This is not yet production quality and should be run only on dedicated test boxes. For people who want to develop transformations for GEOM there exist a set of shims to run geom in userland (ask phk@freebsd.org). Reports of all kinds to: phk@freebsd.org Please include in report: dmesg sysctl debug.geomdot sysctl debug.geomconf Known significant limitations: no kernel dump facility. ioctls severely restricted. Sponsored by: DARPA, NAI Labs
2002-03-11 21:42:35 +00:00
sbuf_finish(sb);
Back out M_* changes, per decision of the TRB. Approved by: trb
2003-02-19 05:47:46 +00:00
gp = g_malloc(sizeof *gp, M_WAITOK | M_ZERO);
gp->name = g_malloc(sbuf_len(sb) + 1, M_WAITOK | M_ZERO);
Cave in to tradition and rename "methods" to "classes".
2002-03-26 21:40:06 +00:00
gp->class = mp;
First commit of the GEOM subsystem to make it easier for people to test and play with this. This is not yet production quality and should be run only on dedicated test boxes. For people who want to develop transformations for GEOM there exist a set of shims to run geom in userland (ask phk@freebsd.org). Reports of all kinds to: phk@freebsd.org Please include in report: dmesg sysctl debug.geomdot sysctl debug.geomconf Known significant limitations: no kernel dump facility. ioctls severely restricted. Sponsored by: DARPA, NAI Labs
2002-03-11 21:42:35 +00:00
gp->rank = 1;
LIST_INIT(&gp->consumer);
LIST_INIT(&gp->provider);
Add aliasing concept to geom. Add an alias name list to geoms. Use them in geom_dev to create aliases. Previously, geom_dev would create an device node for the name of the geom. Now, additional nodes are created pointing back to the primary node with make_dev_alias_p. Aliases must be in place on the geom before any tasting occurs. Differential Revision: https://reviews.freebsd.org/D11873
2017-08-07 21:12:28 +00:00
LIST_INIT(&gp->aliases);
First commit of the GEOM subsystem to make it easier for people to test and play with this. This is not yet production quality and should be run only on dedicated test boxes. For people who want to develop transformations for GEOM there exist a set of shims to run geom in userland (ask phk@freebsd.org). Reports of all kinds to: phk@freebsd.org Please include in report: dmesg sysctl debug.geomdot sysctl debug.geomconf Known significant limitations: no kernel dump facility. ioctls severely restricted. Sponsored by: DARPA, NAI Labs
2002-03-11 21:42:35 +00:00
LIST_INSERT_HEAD(&mp->geom, gp, geom);
TAILQ_INSERT_HEAD(&geoms, gp, geoms);
strcpy(gp->name, sbuf_data(sb));
sbuf_delete(sb);
Give classes a version number and refuse to touch classes which are not understood. This makes room for additional binary compatibility in the future. Put fields in the class for the geom's methods and initialize the methods of a new geom from these fields. This saves some code in all classes.
2004-08-08 06:46:27 +00:00
/* Fill in defaults from class */
gp->start = mp->start;
gp->spoiled = mp->spoiled;
Plumb device physical path reporting from CAM devices, through GEOM and DEVFS, and make it accessible via the diskinfo utility. Extend GEOM's generic attribute query mechanism into generic disk consumers. sys/geom/geom_disk.c: sys/geom/geom_disk.h: sys/cam/scsi/scsi_da.c: sys/cam/ata/ata_da.c: - Allow disk providers to implement a new method which can override the default BIO_GETATTR response, d_getattr(struct bio *). This function returns -1 if not handled, otherwise it returns 0 or an errno to be passed to g_io_deliver(). sys/cam/scsi/scsi_da.c: sys/cam/ata/ata_da.c: - Don't copy the serial number to dp->d_ident anymore, as the CAM XPT is now responsible for returning this information via d_getattr()->(a)dagetattr()->xpt_getatr(). sys/geom/geom_dev.c: - Implement a new ioctl, DIOCGPHYSPATH, which returns the GEOM attribute "GEOM::physpath", if possible. If the attribute request returns a zero-length string, ENOENT is returned. usr.sbin/diskinfo/diskinfo.c: - If the DIOCGPHYSPATH ioctl is successful, report physical path data when diskinfo is executed with the '-v' option. Submitted by: will Reviewed by: gibbs Sponsored by: Spectra Logic Corporation Add generic attribute change notification support to GEOM. sys/sys/geom/geom.h: Add a new attrchanged method field to both g_class and g_geom. sys/sys/geom/geom.h: sys/geom/geom_event.c: - Provide the g_attr_changed() function that providers can use to advertise attribute changes. - Perform delivery of attribute change notifications from a thread context via the standard GEOM event mechanism. sys/geom/geom_subr.c: Inherit the attrchanged method from class to geom (class instance). sys/geom/geom_disk.c: Provide disk_attr_changed() to provide g_attr_changed() access to consumers of the disk API. sys/cam/scsi/scsi_pass.c: sys/cam/scsi/scsi_da.c: sys/geom/geom_dev.c: sys/geom/geom_disk.c: Use attribute changed events to track updates to physical path information. sys/cam/scsi/scsi_da.c: Add AC_ADVINFO_CHANGED to the registered asynchronous CAM events for this driver. When this event occurs, and the updated buffer type references our physical path attribute, emit a GEOM attribute changed event via the disk_attr_changed() API. sys/cam/scsi/scsi_pass.c: Add AC_ADVINFO_CHANGED to the registered asynchronous CAM events for this driver. When this event occurs, update the physical patch devfs alias for this pass instance. Submitted by: gibbs Sponsored by: Spectra Logic Corporation
2011-06-14 17:10:32 +00:00
gp->attrchanged = mp->attrchanged;
Fix a bug which causes a panic in daopen(). The panic is caused by a da(4) instance going away while GEOM is still probing it. In this case, the GEOM disk class instance has been created by disk_create(), and the taste of the disk is queued in the GEOM event queue. While that event is queued, the da(4) instance goes away. When the open call comes into the da(4) driver, it dereferences the freed (but non-NULL) peripheral pointer provided by GEOM, which results in a panic. The solution is to add a callback to the GEOM disk code that is called when all of its resources are cleaned up. This is implemented inside GEOM by adding an optional callback that is called when all consumers have detached from a provider, and the provider is about to be deleted. scsi_cd.c, scsi_da.c: In the register routine for the cd(4) and da(4) routines, acquire a reference to the CAM peripheral instance just before we call disk_create(). Use the new GEOM disk d_gone() callback to register a callback (dadiskgonecb()/cddiskgonecb()) that decrements the peripheral reference count once GEOM has finished cleaning up its resources. In the cd(4) driver, clean up open and close behavior slightly. GEOM makes sure we only get one open() and one close call, so there is no need to set an open flag and decrement the reference count if we are not the first open. In the cd(4) driver, use cam_periph_release_locked() in a couple of error scenarios to avoid extra mutex calls. geom.h: Add a new, optional, providergone callback that is called when a provider is about to be deleted. geom_disk.h: Add a new d_gone() callback to the GEOM disk interface. Bump the DISK_VERSION to version 2. This probably should have been done after a couple of previous changes, especially the addition of the d_getattr() callback. geom_disk.c: Add a providergone callback for the disk class, g_disk_providergone(), that calls the user's d_gone() callback if it exists. Bump the DISK_VERSION to 2. geom_subr.c: In g_destroy_provider(), call the providergone callback if it has been provided. In g_new_geomf(), propagate the class's providergone callback to the new geom instance. blkfront.c: Callers of disk_create() are supposed to pass in DISK_VERSION, not an explicit disk API version number. Update the blkfront driver to do that. disk.9: Update the disk(9) man page to include information on the new d_gone() callback, as well as the previously added d_getattr() callback, d_descr field, and HBA PCI ID fields. MFC after: 5 days
2012-06-24 04:29:03 +00:00
gp->providergone = mp->providergone;
Give classes a version number and refuse to touch classes which are not understood. This makes room for additional binary compatibility in the future. Put fields in the class for the geom's methods and initialize the methods of a new geom from these fields. This saves some code in all classes.
2004-08-08 06:46:27 +00:00
gp->dumpconf = mp->dumpconf;
gp->access = mp->access;
gp->orphan = mp->orphan;
gp->ioctl = mp->ioctl;
Add a new GEOM method, resize(), which is called after provider size changes. Add a new routine, g_resize_provider(), to use to notify GEOM about provider change. Reviewed by: mav Sponsored by: FreeBSD Foundation
2012-07-07 20:13:40 +00:00
gp->resize = mp->resize;
First commit of the GEOM subsystem to make it easier for people to test and play with this. This is not yet production quality and should be run only on dedicated test boxes. For people who want to develop transformations for GEOM there exist a set of shims to run geom in userland (ask phk@freebsd.org). Reports of all kinds to: phk@freebsd.org Please include in report: dmesg sysctl debug.geomdot sysctl debug.geomconf Known significant limitations: no kernel dump facility. ioctls severely restricted. Sponsored by: DARPA, NAI Labs
2002-03-11 21:42:35 +00:00
return (gp);
}
void
g_destroy_geom(struct g_geom *gp)
{
Add aliasing concept to geom. Add an alias name list to geoms. Use them in geom_dev to create aliases. Previously, geom_dev would create an device node for the name of the geom. Now, additional nodes are created pointing back to the primary node with make_dev_alias_p. Aliases must be in place on the geom before any tasting occurs. Differential Revision: https://reviews.freebsd.org/D11873
2017-08-07 21:12:28 +00:00
struct g_geom_alias *gap, *gaptmp;
First commit of the GEOM subsystem to make it easier for people to test and play with this. This is not yet production quality and should be run only on dedicated test boxes. For people who want to develop transformations for GEOM there exist a set of shims to run geom in userland (ask phk@freebsd.org). Reports of all kinds to: phk@freebsd.org Please include in report: dmesg sysctl debug.geomdot sysctl debug.geomconf Known significant limitations: no kernel dump facility. ioctls severely restricted. Sponsored by: DARPA, NAI Labs
2002-03-11 21:42:35 +00:00
g_topology_assert();
Rearrange some of the GEOM debugging tools to be more structured. Retire g_sanity() and corresponding debugflag (0x8) Retire g_{stall,release}_events(). Under #ifdef DIAGNOSTIC: Make g_valid_obj() an official function and have it return an an non-zero integer which indicates the kind of object when found. Implement G_VALID_{CLASS,GEOM,CONSUMER,PROVIDER}() macros based on g_valid_obj(). Sprinkle calls to these macros liberally over the infrastructure. Always check that we do not free a live object.
2004-03-10 08:49:08 +00:00
G_VALID_GEOM(gp);
g_trace(G_T_TOPOLOGY, "g_destroy_geom(%p(%s))", gp, gp->name);
First commit of the GEOM subsystem to make it easier for people to test and play with this. This is not yet production quality and should be run only on dedicated test boxes. For people who want to develop transformations for GEOM there exist a set of shims to run geom in userland (ask phk@freebsd.org). Reports of all kinds to: phk@freebsd.org Please include in report: dmesg sysctl debug.geomdot sysctl debug.geomconf Known significant limitations: no kernel dump facility. ioctls severely restricted. Sponsored by: DARPA, NAI Labs
2002-03-11 21:42:35 +00:00
KASSERT(LIST_EMPTY(&gp->consumer),
("g_destroy_geom(%s) with consumer(s) [%p]",
gp->name, LIST_FIRST(&gp->consumer)));
KASSERT(LIST_EMPTY(&gp->provider),
("g_destroy_geom(%s) with provider(s) [%p]",
Print the correct pointer in a KASSERT. Submitted by: Pawel Jakub Dawidek <nick@garage.freebsd.pl>
2004-01-11 10:02:42 +00:00
gp->name, LIST_FIRST(&gp->provider)));
Change events to have an array of "void *" references, and give the event posting functions varargs to fill these. Attribute g_call_me() to appropriate g_geom's where necessary. Add a flag argument to g_call_me() methods which will be used to signal cancellation of events in the future. This commit should be a no-op.
2003-04-02 20:41:18 +00:00
g_cancel_event(gp);
First commit of the GEOM subsystem to make it easier for people to test and play with this. This is not yet production quality and should be run only on dedicated test boxes. For people who want to develop transformations for GEOM there exist a set of shims to run geom in userland (ask phk@freebsd.org). Reports of all kinds to: phk@freebsd.org Please include in report: dmesg sysctl debug.geomdot sysctl debug.geomconf Known significant limitations: no kernel dump facility. ioctls severely restricted. Sponsored by: DARPA, NAI Labs
2002-03-11 21:42:35 +00:00
LIST_REMOVE(gp, geom);
TAILQ_REMOVE(&geoms, gp, geoms);
Add aliasing concept to geom. Add an alias name list to geoms. Use them in geom_dev to create aliases. Previously, geom_dev would create an device node for the name of the geom. Now, additional nodes are created pointing back to the primary node with make_dev_alias_p. Aliases must be in place on the geom before any tasting occurs. Differential Revision: https://reviews.freebsd.org/D11873
2017-08-07 21:12:28 +00:00
LIST_FOREACH_SAFE(gap, &gp->aliases, ga_next, gaptmp)
g_free(gap);
Introduce some serious paranoia to try to catch a memory overwrite problem as early as possible. Sponsored by: DARPA & NAI Labs
2002-04-23 11:48:45 +00:00
g_free(gp->name);
First commit of the GEOM subsystem to make it easier for people to test and play with this. This is not yet production quality and should be run only on dedicated test boxes. For people who want to develop transformations for GEOM there exist a set of shims to run geom in userland (ask phk@freebsd.org). Reports of all kinds to: phk@freebsd.org Please include in report: dmesg sysctl debug.geomdot sysctl debug.geomconf Known significant limitations: no kernel dump facility. ioctls severely restricted. Sponsored by: DARPA, NAI Labs
2002-03-11 21:42:35 +00:00
g_free(gp);
}
Rework the "withering" mechanism: Introduce g_wither_geom() to do the work in one single place.
2003-05-02 06:15:27 +00:00
/*
- Add missing word in comment.
2008-12-08 17:09:02 +00:00
* This function is called (repeatedly) until the geom has withered away.
Rework the "withering" mechanism: Introduce g_wither_geom() to do the work in one single place.
2003-05-02 06:15:27 +00:00
*/
void
g_wither_geom(struct g_geom *gp, int error)
{
Make withering water tight. When we orphan/wither a provider, an attached geom+consumer could end up being withered as a result and it may be in front of us in the normal object scanning order so we need to do multi-pass. On the other hand, there may be withering stuff we can't get rid off (yet), so we need to keep track of both the existence of withering stuff and if there is more we can do at this time.
2004-07-08 16:17:14 +00:00
struct g_provider *pp;
Rework the "withering" mechanism: Introduce g_wither_geom() to do the work in one single place.
2003-05-02 06:15:27 +00:00
Rearrange some of the GEOM debugging tools to be more structured. Retire g_sanity() and corresponding debugflag (0x8) Retire g_{stall,release}_events(). Under #ifdef DIAGNOSTIC: Make g_valid_obj() an official function and have it return an an non-zero integer which indicates the kind of object when found. Implement G_VALID_{CLASS,GEOM,CONSUMER,PROVIDER}() macros based on g_valid_obj(). Sprinkle calls to these macros liberally over the infrastructure. Always check that we do not free a live object.
2004-03-10 08:49:08 +00:00
g_topology_assert();
G_VALID_GEOM(gp);
Rework the "withering" mechanism: Introduce g_wither_geom() to do the work in one single place.
2003-05-02 06:15:27 +00:00
g_trace(G_T_TOPOLOGY, "g_wither_geom(%p(%s))", gp, gp->name);
if (!(gp->flags & G_GEOM_WITHER)) {
gp->flags |= G_GEOM_WITHER;
LIST_FOREACH(pp, &gp->provider, provider)
Don't panic on providers already withered when we wither a geom.
2004-03-07 17:33:15 +00:00
if (!(pp->flags & G_PF_ORPHAN))
g_orphan_provider(pp, error);
Rework the "withering" mechanism: Introduce g_wither_geom() to do the work in one single place.
2003-05-02 06:15:27 +00:00
}
Make withering water tight. When we orphan/wither a provider, an attached geom+consumer could end up being withered as a result and it may be in front of us in the normal object scanning order so we need to do multi-pass. On the other hand, there may be withering stuff we can't get rid off (yet), so we need to keep track of both the existence of withering stuff and if there is more we can do at this time.
2004-07-08 16:17:14 +00:00
g_do_wither();
}
Add g_wither_provider() to abstract the details of destroying a particular provider. Use this function where g_orphan_provider() is being called so that the flags are updated correctly and g_orphan_provider() is called only when allowed.
2006-04-10 03:55:13 +00:00
/*
* Convenience function to destroy a particular provider.
*/
void
g_wither_provider(struct g_provider *pp, int error)
{
pp->flags |= G_PF_WITHER;
if (!(pp->flags & G_PF_ORPHAN))
g_orphan_provider(pp, error);
}
Add g_wither_geom_close() function.
2004-10-29 09:19:03 +00:00
/*
* This function is called (repeatedly) until the has withered away.
*/
void
g_wither_geom_close(struct g_geom *gp, int error)
{
struct g_consumer *cp;
g_topology_assert();
G_VALID_GEOM(gp);
g_trace(G_T_TOPOLOGY, "g_wither_geom_close(%p(%s))", gp, gp->name);
LIST_FOREACH(cp, &gp->consumer, consumer)
if (cp->acr || cp->acw || cp->ace)
g_access(cp, -cp->acr, -cp->acw, -cp->ace);
g_wither_geom(gp, error);
}
Make withering water tight. When we orphan/wither a provider, an attached geom+consumer could end up being withered as a result and it may be in front of us in the normal object scanning order so we need to do multi-pass. On the other hand, there may be withering stuff we can't get rid off (yet), so we need to keep track of both the existence of withering stuff and if there is more we can do at this time.
2004-07-08 16:17:14 +00:00
/*
* This function is called (repeatedly) until we cant wash away more
Make g_wither_washer() to not loop by itself, but only when there was some more topology change done that may require its attention. Add few missing g_do_wither() calls in respective places to signal it. This fixes potential infinite loop here when some provider is withered, but still opened or connected for some reason and so can not be destroyed. For example, see r227009 and r227510.
2013-03-24 03:15:20 +00:00
* withered bits at present.
Make withering water tight. When we orphan/wither a provider, an attached geom+consumer could end up being withered as a result and it may be in front of us in the normal object scanning order so we need to do multi-pass. On the other hand, there may be withering stuff we can't get rid off (yet), so we need to keep track of both the existence of withering stuff and if there is more we can do at this time.
2004-07-08 16:17:14 +00:00
*/
Make g_wither_washer() to not loop by itself, but only when there was some more topology change done that may require its attention. Add few missing g_do_wither() calls in respective places to signal it. This fixes potential infinite loop here when some provider is withered, but still opened or connected for some reason and so can not be destroyed. For example, see r227009 and r227510.
2013-03-24 03:15:20 +00:00
void
Make withering water tight. When we orphan/wither a provider, an attached geom+consumer could end up being withered as a result and it may be in front of us in the normal object scanning order so we need to do multi-pass. On the other hand, there may be withering stuff we can't get rid off (yet), so we need to keep track of both the existence of withering stuff and if there is more we can do at this time.
2004-07-08 16:17:14 +00:00
g_wither_washer()
{
struct g_class *mp;
struct g_geom *gp, *gp2;
struct g_provider *pp, *pp2;
struct g_consumer *cp, *cp2;
g_topology_assert();
LIST_FOREACH(mp, &g_classes, class) {
LIST_FOREACH_SAFE(gp, &mp->geom, geom, gp2) {
LIST_FOREACH_SAFE(pp, &gp->provider, provider, pp2) {
if (!(pp->flags & G_PF_WITHER))
continue;
if (LIST_EMPTY(&pp->consumers))
g_destroy_provider(pp);
}
if (!(gp->flags & G_GEOM_WITHER))
continue;
LIST_FOREACH_SAFE(pp, &gp->provider, provider, pp2) {
if (LIST_EMPTY(&pp->consumers))
g_destroy_provider(pp);
}
LIST_FOREACH_SAFE(cp, &gp->consumer, consumer, cp2) {
Make g_wither_washer() to not loop by itself, but only when there was some more topology change done that may require its attention. Add few missing g_do_wither() calls in respective places to signal it. This fixes potential infinite loop here when some provider is withered, but still opened or connected for some reason and so can not be destroyed. For example, see r227009 and r227510.
2013-03-24 03:15:20 +00:00
if (cp->acr || cp->acw || cp->ace)
Make withering water tight. When we orphan/wither a provider, an attached geom+consumer could end up being withered as a result and it may be in front of us in the normal object scanning order so we need to do multi-pass. On the other hand, there may be withering stuff we can't get rid off (yet), so we need to keep track of both the existence of withering stuff and if there is more we can do at this time.
2004-07-08 16:17:14 +00:00
continue;
Only detach consumers which are attached when we wither stuff away. Pointed out by: pjd
2004-07-09 14:06:17 +00:00
if (cp->provider != NULL)
g_detach(cp);
Make withering water tight. When we orphan/wither a provider, an attached geom+consumer could end up being withered as a result and it may be in front of us in the normal object scanning order so we need to do multi-pass. On the other hand, there may be withering stuff we can't get rid off (yet), so we need to keep track of both the existence of withering stuff and if there is more we can do at this time.
2004-07-08 16:17:14 +00:00
g_destroy_consumer(cp);
}
if (LIST_EMPTY(&gp->provider) &&
LIST_EMPTY(&gp->consumer))
g_destroy_geom(gp);
}
Rework the "withering" mechanism: Introduce g_wither_geom() to do the work in one single place.
2003-05-02 06:15:27 +00:00
}
}
First commit of the GEOM subsystem to make it easier for people to test and play with this. This is not yet production quality and should be run only on dedicated test boxes. For people who want to develop transformations for GEOM there exist a set of shims to run geom in userland (ask phk@freebsd.org). Reports of all kinds to: phk@freebsd.org Please include in report: dmesg sysctl debug.geomdot sysctl debug.geomconf Known significant limitations: no kernel dump facility. ioctls severely restricted. Sponsored by: DARPA, NAI Labs
2002-03-11 21:42:35 +00:00
struct g_consumer *
g_new_consumer(struct g_geom *gp)
{
struct g_consumer *cp;
g_topology_assert();
Rearrange some of the GEOM debugging tools to be more structured. Retire g_sanity() and corresponding debugflag (0x8) Retire g_{stall,release}_events(). Under #ifdef DIAGNOSTIC: Make g_valid_obj() an official function and have it return an an non-zero integer which indicates the kind of object when found. Implement G_VALID_{CLASS,GEOM,CONSUMER,PROVIDER}() macros based on g_valid_obj(). Sprinkle calls to these macros liberally over the infrastructure. Always check that we do not free a live object.
2004-03-10 08:49:08 +00:00
G_VALID_GEOM(gp);
Add KASSERTS. Submitted by: Pawel Jakub Dawidek <nick@garage.freebsd.pl>
2004-01-23 21:02:49 +00:00
KASSERT(!(gp->flags & G_GEOM_WITHER),
("g_new_consumer on WITHERing geom(%s) (class %s)",
gp->name, gp->class->name));
Move access and orphan member functions from class to geom. Sponsored by: DARPA & NAI Labs
2002-04-04 09:54:13 +00:00
KASSERT(gp->orphan != NULL,
("g_new_consumer on geom(%s) (class %s) without orphan",
gp->name, gp->class->name));
First commit of the GEOM subsystem to make it easier for people to test and play with this. This is not yet production quality and should be run only on dedicated test boxes. For people who want to develop transformations for GEOM there exist a set of shims to run geom in userland (ask phk@freebsd.org). Reports of all kinds to: phk@freebsd.org Please include in report: dmesg sysctl debug.geomdot sysctl debug.geomconf Known significant limitations: no kernel dump facility. ioctls severely restricted. Sponsored by: DARPA, NAI Labs
2002-03-11 21:42:35 +00:00
Back out M_* changes, per decision of the TRB. Approved by: trb
2003-02-19 05:47:46 +00:00
cp = g_malloc(sizeof *cp, M_WAITOK | M_ZERO);
First commit of the GEOM subsystem to make it easier for people to test and play with this. This is not yet production quality and should be run only on dedicated test boxes. For people who want to develop transformations for GEOM there exist a set of shims to run geom in userland (ask phk@freebsd.org). Reports of all kinds to: phk@freebsd.org Please include in report: dmesg sysctl debug.geomdot sysctl debug.geomconf Known significant limitations: no kernel dump facility. ioctls severely restricted. Sponsored by: DARPA, NAI Labs
2002-03-11 21:42:35 +00:00
cp->geom = gp;
Retire the GEOM private statistics code and use devstat instead.
2003-03-18 09:42:33 +00:00
cp->stat = devstat_new_entry(cp, -1, 0, DEVSTAT_ALL_SUPPORTED,
DEVSTAT_TYPE_DIRECT, DEVSTAT_PRIORITY_MAX);
First commit of the GEOM subsystem to make it easier for people to test and play with this. This is not yet production quality and should be run only on dedicated test boxes. For people who want to develop transformations for GEOM there exist a set of shims to run geom in userland (ask phk@freebsd.org). Reports of all kinds to: phk@freebsd.org Please include in report: dmesg sysctl debug.geomdot sysctl debug.geomconf Known significant limitations: no kernel dump facility. ioctls severely restricted. Sponsored by: DARPA, NAI Labs
2002-03-11 21:42:35 +00:00
LIST_INSERT_HEAD(&gp->consumer, cp, consumer);
return(cp);
}
void
g_destroy_consumer(struct g_consumer *cp)
{
Rework the "withering" mechanism: Introduce g_wither_geom() to do the work in one single place.
2003-05-02 06:15:27 +00:00
struct g_geom *gp;
First commit of the GEOM subsystem to make it easier for people to test and play with this. This is not yet production quality and should be run only on dedicated test boxes. For people who want to develop transformations for GEOM there exist a set of shims to run geom in userland (ask phk@freebsd.org). Reports of all kinds to: phk@freebsd.org Please include in report: dmesg sysctl debug.geomdot sysctl debug.geomconf Known significant limitations: no kernel dump facility. ioctls severely restricted. Sponsored by: DARPA, NAI Labs
2002-03-11 21:42:35 +00:00
g_topology_assert();
Rearrange some of the GEOM debugging tools to be more structured. Retire g_sanity() and corresponding debugflag (0x8) Retire g_{stall,release}_events(). Under #ifdef DIAGNOSTIC: Make g_valid_obj() an official function and have it return an an non-zero integer which indicates the kind of object when found. Implement G_VALID_{CLASS,GEOM,CONSUMER,PROVIDER}() macros based on g_valid_obj(). Sprinkle calls to these macros liberally over the infrastructure. Always check that we do not free a live object.
2004-03-10 08:49:08 +00:00
G_VALID_CONSUMER(cp);
g_trace(G_T_TOPOLOGY, "g_destroy_consumer(%p)", cp);
First commit of the GEOM subsystem to make it easier for people to test and play with this. This is not yet production quality and should be run only on dedicated test boxes. For people who want to develop transformations for GEOM there exist a set of shims to run geom in userland (ask phk@freebsd.org). Reports of all kinds to: phk@freebsd.org Please include in report: dmesg sysctl debug.geomdot sysctl debug.geomconf Known significant limitations: no kernel dump facility. ioctls severely restricted. Sponsored by: DARPA, NAI Labs
2002-03-11 21:42:35 +00:00
KASSERT (cp->provider == NULL, ("g_destroy_consumer but attached"));
KASSERT (cp->acr == 0, ("g_destroy_consumer with acr"));
KASSERT (cp->acw == 0, ("g_destroy_consumer with acw"));
KASSERT (cp->ace == 0, ("g_destroy_consumer with ace"));
Change events to have an array of "void *" references, and give the event posting functions varargs to fill these. Attribute g_call_me() to appropriate g_geom's where necessary. Add a flag argument to g_call_me() methods which will be used to signal cancellation of events in the future. This commit should be a no-op.
2003-04-02 20:41:18 +00:00
g_cancel_event(cp);
Rework the "withering" mechanism: Introduce g_wither_geom() to do the work in one single place.
2003-05-02 06:15:27 +00:00
gp = cp->geom;
First commit of the GEOM subsystem to make it easier for people to test and play with this. This is not yet production quality and should be run only on dedicated test boxes. For people who want to develop transformations for GEOM there exist a set of shims to run geom in userland (ask phk@freebsd.org). Reports of all kinds to: phk@freebsd.org Please include in report: dmesg sysctl debug.geomdot sysctl debug.geomconf Known significant limitations: no kernel dump facility. ioctls severely restricted. Sponsored by: DARPA, NAI Labs
2002-03-11 21:42:35 +00:00
LIST_REMOVE(cp, consumer);
Retire the GEOM private statistics code and use devstat instead.
2003-03-18 09:42:33 +00:00
devstat_remove_entry(cp->stat);
First commit of the GEOM subsystem to make it easier for people to test and play with this. This is not yet production quality and should be run only on dedicated test boxes. For people who want to develop transformations for GEOM there exist a set of shims to run geom in userland (ask phk@freebsd.org). Reports of all kinds to: phk@freebsd.org Please include in report: dmesg sysctl debug.geomdot sysctl debug.geomconf Known significant limitations: no kernel dump facility. ioctls severely restricted. Sponsored by: DARPA, NAI Labs
2002-03-11 21:42:35 +00:00
g_free(cp);
Rework the "withering" mechanism: Introduce g_wither_geom() to do the work in one single place.
2003-05-02 06:15:27 +00:00
if (gp->flags & G_GEOM_WITHER)
Make withering water tight. When we orphan/wither a provider, an attached geom+consumer could end up being withered as a result and it may be in front of us in the normal object scanning order so we need to do multi-pass. On the other hand, there may be withering stuff we can't get rid off (yet), so we need to keep track of both the existence of withering stuff and if there is more we can do at this time.
2004-07-08 16:17:14 +00:00
g_do_wither();
First commit of the GEOM subsystem to make it easier for people to test and play with this. This is not yet production quality and should be run only on dedicated test boxes. For people who want to develop transformations for GEOM there exist a set of shims to run geom in userland (ask phk@freebsd.org). Reports of all kinds to: phk@freebsd.org Please include in report: dmesg sysctl debug.geomdot sysctl debug.geomconf Known significant limitations: no kernel dump facility. ioctls severely restricted. Sponsored by: DARPA, NAI Labs
2002-03-11 21:42:35 +00:00
}
Turn EV_NEW_PROVIDER into a g_call_me() event.
2003-04-23 20:16:13 +00:00
static void
g_new_provider_event(void *arg, int flag)
{
struct g_class *mp;
struct g_provider *pp;
Implement media change notification for DA and CD removable media devices. It includes three parts: 1) Modifications to CAM to detect media media changes and report them to disk(9) layer. For modern SATA (and potentially UAS) devices it utilizes Asynchronous Notification mechanism to receive events from hardware. Active polling with TEST UNIT READY commands with 3 seconds period is used for incapable hardware. After that both CD and DA drivers work the same way, detecting two conditions: "NOT READY: Medium not present" after medium was detected previously, and "UNIT ATTENTION: Not ready to ready change, medium may have changed". First one reported to disk(9) as media removal, second as media insert/change. To reliably receive second event new AC_UNIT_ATTENTION async added to make UAs broadcasted to all periphs by generic error handling code in cam_periph_error(). 2) Modifications to GEOM core to handle media remove and change events. Media removal handled by spoiling all consumers attached to the provider. Media change event also schedules provider retaste after spoiling to probe new media. New flag G_CF_ORPHAN was added to consumers to reflect that consumer is in process of destruction. It allows retaste to create new geom instance of the same class, while previous one is still dying. 3) Modifications to some GEOM classes: DEV -- to report media change events to devd; VFS -- to handle spoiling same as orphan to prevent accessing replaced media. PART class already handles spoiling alike to orphan. Reviewed by: silence on geom@ and scsi@ Tested by: avg Sponsored by: iXsystems, Inc. / PC-BSD MFC after: 2 months
2012-07-29 11:51:48 +00:00
struct g_consumer *cp, *next_cp;
Turn EV_NEW_PROVIDER into a g_call_me() event.
2003-04-23 20:16:13 +00:00
g_topology_assert();
if (flag == EV_CANCEL)
return;
if (g_shutdown)
return;
pp = arg;
Rearrange some of the GEOM debugging tools to be more structured. Retire g_sanity() and corresponding debugflag (0x8) Retire g_{stall,release}_events(). Under #ifdef DIAGNOSTIC: Make g_valid_obj() an official function and have it return an an non-zero integer which indicates the kind of object when found. Implement G_VALID_{CLASS,GEOM,CONSUMER,PROVIDER}() macros based on g_valid_obj(). Sprinkle calls to these macros liberally over the infrastructure. Always check that we do not free a live object.
2004-03-10 08:49:08 +00:00
G_VALID_PROVIDER(pp);
- Assert that we don't send new provider event for a provider which has G_PF_WITHER flag set. - Fix typo in assertion condition (sorry, but I forgot who report that).
2008-05-18 22:50:50 +00:00
KASSERT(!(pp->flags & G_PF_WITHER),
("g_new_provider_event but withered"));
Implement media change notification for DA and CD removable media devices. It includes three parts: 1) Modifications to CAM to detect media media changes and report them to disk(9) layer. For modern SATA (and potentially UAS) devices it utilizes Asynchronous Notification mechanism to receive events from hardware. Active polling with TEST UNIT READY commands with 3 seconds period is used for incapable hardware. After that both CD and DA drivers work the same way, detecting two conditions: "NOT READY: Medium not present" after medium was detected previously, and "UNIT ATTENTION: Not ready to ready change, medium may have changed". First one reported to disk(9) as media removal, second as media insert/change. To reliably receive second event new AC_UNIT_ATTENTION async added to make UAs broadcasted to all periphs by generic error handling code in cam_periph_error(). 2) Modifications to GEOM core to handle media remove and change events. Media removal handled by spoiling all consumers attached to the provider. Media change event also schedules provider retaste after spoiling to probe new media. New flag G_CF_ORPHAN was added to consumers to reflect that consumer is in process of destruction. It allows retaste to create new geom instance of the same class, while previous one is still dying. 3) Modifications to some GEOM classes: DEV -- to report media change events to devd; VFS -- to handle spoiling same as orphan to prevent accessing replaced media. PART class already handles spoiling alike to orphan. Reviewed by: silence on geom@ and scsi@ Tested by: avg Sponsored by: iXsystems, Inc. / PC-BSD MFC after: 2 months
2012-07-29 11:51:48 +00:00
LIST_FOREACH_SAFE(cp, &pp->consumers, consumers, next_cp) {
if ((cp->flags & G_CF_ORPHAN) == 0 &&
cp->geom->attrchanged != NULL)
cp->geom->attrchanged(cp, "GEOM::media");
}
Introduce a kern.geom.notaste sysctl that can be used to temporarily disable GEOM tasting to avoid the "bouncing GEOM" problem where, when you shut down the consumer of a provider which can be viewed in multiple ways (typically a mirror whose members are labeled partitions), GEOM will immediately taste that provider's alter ego and reattach the consumer. Approved by: re (glebius)
2013-09-24 20:05:16 +00:00
if (g_notaste)
return;
Turn EV_NEW_PROVIDER into a g_call_me() event.
2003-04-23 20:16:13 +00:00
LIST_FOREACH(mp, &g_classes, class) {
if (mp->taste == NULL)
continue;
LIST_FOREACH(cp, &pp->consumers, consumers)
Implement media change notification for DA and CD removable media devices. It includes three parts: 1) Modifications to CAM to detect media media changes and report them to disk(9) layer. For modern SATA (and potentially UAS) devices it utilizes Asynchronous Notification mechanism to receive events from hardware. Active polling with TEST UNIT READY commands with 3 seconds period is used for incapable hardware. After that both CD and DA drivers work the same way, detecting two conditions: "NOT READY: Medium not present" after medium was detected previously, and "UNIT ATTENTION: Not ready to ready change, medium may have changed". First one reported to disk(9) as media removal, second as media insert/change. To reliably receive second event new AC_UNIT_ATTENTION async added to make UAs broadcasted to all periphs by generic error handling code in cam_periph_error(). 2) Modifications to GEOM core to handle media remove and change events. Media removal handled by spoiling all consumers attached to the provider. Media change event also schedules provider retaste after spoiling to probe new media. New flag G_CF_ORPHAN was added to consumers to reflect that consumer is in process of destruction. It allows retaste to create new geom instance of the same class, while previous one is still dying. 3) Modifications to some GEOM classes: DEV -- to report media change events to devd; VFS -- to handle spoiling same as orphan to prevent accessing replaced media. PART class already handles spoiling alike to orphan. Reviewed by: silence on geom@ and scsi@ Tested by: avg Sponsored by: iXsystems, Inc. / PC-BSD MFC after: 2 months
2012-07-29 11:51:48 +00:00
if (cp->geom->class == mp &&
(cp->flags & G_CF_ORPHAN) == 0)
Simplify.
2009-06-05 23:35:43 +00:00
break;
if (cp != NULL)
Turn EV_NEW_PROVIDER into a g_call_me() event.
2003-04-23 20:16:13 +00:00
continue;
Back out all the stuff that didn't belong in the last commit.
2003-05-02 06:42:59 +00:00
mp->taste(mp, pp, 0);
Turn EV_NEW_PROVIDER into a g_call_me() event.
2003-04-23 20:16:13 +00:00
g_topology_assert();
}
}
First commit of the GEOM subsystem to make it easier for people to test and play with this. This is not yet production quality and should be run only on dedicated test boxes. For people who want to develop transformations for GEOM there exist a set of shims to run geom in userland (ask phk@freebsd.org). Reports of all kinds to: phk@freebsd.org Please include in report: dmesg sysctl debug.geomdot sysctl debug.geomconf Known significant limitations: no kernel dump facility. ioctls severely restricted. Sponsored by: DARPA, NAI Labs
2002-03-11 21:42:35 +00:00
struct g_provider *
Constification and some s/int/u_int/ changes.
2002-12-16 22:33:27 +00:00
g_new_providerf(struct g_geom *gp, const char *fmt, ...)
First commit of the GEOM subsystem to make it easier for people to test and play with this. This is not yet production quality and should be run only on dedicated test boxes. For people who want to develop transformations for GEOM there exist a set of shims to run geom in userland (ask phk@freebsd.org). Reports of all kinds to: phk@freebsd.org Please include in report: dmesg sysctl debug.geomdot sysctl debug.geomconf Known significant limitations: no kernel dump facility. ioctls severely restricted. Sponsored by: DARPA, NAI Labs
2002-03-11 21:42:35 +00:00
{
struct g_provider *pp;
struct sbuf *sb;
va_list ap;
g_topology_assert();
Rearrange some of the GEOM debugging tools to be more structured. Retire g_sanity() and corresponding debugflag (0x8) Retire g_{stall,release}_events(). Under #ifdef DIAGNOSTIC: Make g_valid_obj() an official function and have it return an an non-zero integer which indicates the kind of object when found. Implement G_VALID_{CLASS,GEOM,CONSUMER,PROVIDER}() macros based on g_valid_obj(). Sprinkle calls to these macros liberally over the infrastructure. Always check that we do not free a live object.
2004-03-10 08:49:08 +00:00
G_VALID_GEOM(gp);
Be more insistent on destroying geoms at unload time. Still not perfect, but it will do (better) for now. KASSERT that to have providers a class must have an access method. Tag the new_provider event with the geom as well.
2004-03-11 08:16:23 +00:00
KASSERT(gp->access != NULL,
("new provider on geom(%s) without ->access (class %s)",
gp->name, gp->class->name));
Add KASSERTS. Submitted by: Pawel Jakub Dawidek <nick@garage.freebsd.pl>
2004-01-23 21:02:49 +00:00
KASSERT(gp->start != NULL,
- Use proper names in KASSERTs. - Typos. Approved by: phk, scottl (mentor)
2004-02-02 17:50:09 +00:00
("new provider on geom(%s) without ->start (class %s)",
gp->name, gp->class->name));
Add KASSERTS. Submitted by: Pawel Jakub Dawidek <nick@garage.freebsd.pl>
2004-01-23 21:02:49 +00:00
KASSERT(!(gp->flags & G_GEOM_WITHER),
- Use proper names in KASSERTs. - Typos. Approved by: phk, scottl (mentor)
2004-02-02 17:50:09 +00:00
("new provider on WITHERing geom(%s) (class %s)",
gp->name, gp->class->name));
Add sbuf_new_auto as a shortcut for the very common case of creating a completely dynamic sbuf. Obtained from: Varnish MFC after: 2 weeks
2008-08-09 11:14:05 +00:00
sb = sbuf_new_auto();
Add missing va_end() calls. Noticed by: tmm
2003-06-07 10:16:53 +00:00
va_start(ap, fmt);
First commit of the GEOM subsystem to make it easier for people to test and play with this. This is not yet production quality and should be run only on dedicated test boxes. For people who want to develop transformations for GEOM there exist a set of shims to run geom in userland (ask phk@freebsd.org). Reports of all kinds to: phk@freebsd.org Please include in report: dmesg sysctl debug.geomdot sysctl debug.geomconf Known significant limitations: no kernel dump facility. ioctls severely restricted. Sponsored by: DARPA, NAI Labs
2002-03-11 21:42:35 +00:00
sbuf_vprintf(sb, fmt, ap);
Add missing va_end() calls. Noticed by: tmm
2003-06-07 10:16:53 +00:00
va_end(ap);
First commit of the GEOM subsystem to make it easier for people to test and play with this. This is not yet production quality and should be run only on dedicated test boxes. For people who want to develop transformations for GEOM there exist a set of shims to run geom in userland (ask phk@freebsd.org). Reports of all kinds to: phk@freebsd.org Please include in report: dmesg sysctl debug.geomdot sysctl debug.geomconf Known significant limitations: no kernel dump facility. ioctls severely restricted. Sponsored by: DARPA, NAI Labs
2002-03-11 21:42:35 +00:00
sbuf_finish(sb);
Back out M_* changes, per decision of the TRB. Approved by: trb
2003-02-19 05:47:46 +00:00
pp = g_malloc(sizeof *pp + sbuf_len(sb) + 1, M_WAITOK | M_ZERO);
First commit of the GEOM subsystem to make it easier for people to test and play with this. This is not yet production quality and should be run only on dedicated test boxes. For people who want to develop transformations for GEOM there exist a set of shims to run geom in userland (ask phk@freebsd.org). Reports of all kinds to: phk@freebsd.org Please include in report: dmesg sysctl debug.geomdot sysctl debug.geomconf Known significant limitations: no kernel dump facility. ioctls severely restricted. Sponsored by: DARPA, NAI Labs
2002-03-11 21:42:35 +00:00
pp->name = (char *)(pp + 1);
strcpy(pp->name, sbuf_data(sb));
sbuf_delete(sb);
LIST_INIT(&pp->consumers);
pp->error = ENXIO;
pp->geom = gp;
Retire the GEOM private statistics code and use devstat instead.
2003-03-18 09:42:33 +00:00
pp->stat = devstat_new_entry(pp, -1, 0, DEVSTAT_ALL_SUPPORTED,
DEVSTAT_TYPE_DIRECT, DEVSTAT_PRIORITY_MAX);
First commit of the GEOM subsystem to make it easier for people to test and play with this. This is not yet production quality and should be run only on dedicated test boxes. For people who want to develop transformations for GEOM there exist a set of shims to run geom in userland (ask phk@freebsd.org). Reports of all kinds to: phk@freebsd.org Please include in report: dmesg sysctl debug.geomdot sysctl debug.geomconf Known significant limitations: no kernel dump facility. ioctls severely restricted. Sponsored by: DARPA, NAI Labs
2002-03-11 21:42:35 +00:00
LIST_INSERT_HEAD(&gp->provider, pp, provider);
Be more insistent on destroying geoms at unload time. Still not perfect, but it will do (better) for now. KASSERT that to have providers a class must have an access method. Tag the new_provider event with the geom as well.
2004-03-11 08:16:23 +00:00
g_post_event(g_new_provider_event, pp, M_WAITOK, pp, gp, NULL);
First commit of the GEOM subsystem to make it easier for people to test and play with this. This is not yet production quality and should be run only on dedicated test boxes. For people who want to develop transformations for GEOM there exist a set of shims to run geom in userland (ask phk@freebsd.org). Reports of all kinds to: phk@freebsd.org Please include in report: dmesg sysctl debug.geomdot sysctl debug.geomconf Known significant limitations: no kernel dump facility. ioctls severely restricted. Sponsored by: DARPA, NAI Labs
2002-03-11 21:42:35 +00:00
return (pp);
}
void
g_error_provider(struct g_provider *pp, int error)
{
Rearrange some of the GEOM debugging tools to be more structured. Retire g_sanity() and corresponding debugflag (0x8) Retire g_{stall,release}_events(). Under #ifdef DIAGNOSTIC: Make g_valid_obj() an official function and have it return an an non-zero integer which indicates the kind of object when found. Implement G_VALID_{CLASS,GEOM,CONSUMER,PROVIDER}() macros based on g_valid_obj(). Sprinkle calls to these macros liberally over the infrastructure. Always check that we do not free a live object.
2004-03-10 08:49:08 +00:00
/* G_VALID_PROVIDER(pp); We may not have g_topology */
First commit of the GEOM subsystem to make it easier for people to test and play with this. This is not yet production quality and should be run only on dedicated test boxes. For people who want to develop transformations for GEOM there exist a set of shims to run geom in userland (ask phk@freebsd.org). Reports of all kinds to: phk@freebsd.org Please include in report: dmesg sysctl debug.geomdot sysctl debug.geomconf Known significant limitations: no kernel dump facility. ioctls severely restricted. Sponsored by: DARPA, NAI Labs
2002-03-11 21:42:35 +00:00
pp->error = error;
}
Add a new GEOM method, resize(), which is called after provider size changes. Add a new routine, g_resize_provider(), to use to notify GEOM about provider change. Reviewed by: mav Sponsored by: FreeBSD Foundation
2012-07-07 20:13:40 +00:00
static void
g_resize_provider_event(void *arg, int flag)
{
struct g_hh00 *hh;
struct g_class *mp;
struct g_geom *gp;
struct g_provider *pp;
struct g_consumer *cp, *cp2;
off_t size;
g_topology_assert();
if (g_shutdown)
return;
hh = arg;
pp = hh->pp;
size = hh->size;
Add missing free.
2012-07-18 07:26:20 +00:00
g_free(hh);
Add a new GEOM method, resize(), which is called after provider size changes. Add a new routine, g_resize_provider(), to use to notify GEOM about provider change. Reviewed by: mav Sponsored by: FreeBSD Foundation
2012-07-07 20:13:40 +00:00
G_VALID_PROVIDER(pp);
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
2016-06-22 14:39:13 +00:00
KASSERT(!(pp->flags & G_PF_WITHER),
("g_resize_provider_event but withered"));
Add a new GEOM method, resize(), which is called after provider size changes. Add a new routine, g_resize_provider(), to use to notify GEOM about provider change. Reviewed by: mav Sponsored by: FreeBSD Foundation
2012-07-07 20:13:40 +00:00
g_trace(G_T_TOPOLOGY, "g_resize_provider_event(%p)", pp);
LIST_FOREACH_SAFE(cp, &pp->consumers, consumers, cp2) {
gp = cp->geom;
Implement media change notification for DA and CD removable media devices. It includes three parts: 1) Modifications to CAM to detect media media changes and report them to disk(9) layer. For modern SATA (and potentially UAS) devices it utilizes Asynchronous Notification mechanism to receive events from hardware. Active polling with TEST UNIT READY commands with 3 seconds period is used for incapable hardware. After that both CD and DA drivers work the same way, detecting two conditions: "NOT READY: Medium not present" after medium was detected previously, and "UNIT ATTENTION: Not ready to ready change, medium may have changed". First one reported to disk(9) as media removal, second as media insert/change. To reliably receive second event new AC_UNIT_ATTENTION async added to make UAs broadcasted to all periphs by generic error handling code in cam_periph_error(). 2) Modifications to GEOM core to handle media remove and change events. Media removal handled by spoiling all consumers attached to the provider. Media change event also schedules provider retaste after spoiling to probe new media. New flag G_CF_ORPHAN was added to consumers to reflect that consumer is in process of destruction. It allows retaste to create new geom instance of the same class, while previous one is still dying. 3) Modifications to some GEOM classes: DEV -- to report media change events to devd; VFS -- to handle spoiling same as orphan to prevent accessing replaced media. PART class already handles spoiling alike to orphan. Reviewed by: silence on geom@ and scsi@ Tested by: avg Sponsored by: iXsystems, Inc. / PC-BSD MFC after: 2 months
2012-07-29 11:51:48 +00:00
if (gp->resize == NULL && size < pp->mediasize) {
g_resize_provider_event: Do not invoke orphan method twice Like r266444, g_resize_provider_event can attempt to orphan an already orphaned geom_dev consumer. This will cause a panic in g_dev_orphan. Apply the same fix as was applied to g_orphan_register. Reviewed by: ae Sponsored by: Dell EMC Isilon Differential Revision: https://reviews.freebsd.org/D12469
2017-09-24 19:59:26 +00:00
/*
* XXX: g_dev_orphan method does deferred destroying
* and it is possible, that other event could already
* call the orphan method. Check consumer's flags to
* do not schedule it twice.
*/
if (cp->flags & G_CF_ORPHAN)
continue;
Implement media change notification for DA and CD removable media devices. It includes three parts: 1) Modifications to CAM to detect media media changes and report them to disk(9) layer. For modern SATA (and potentially UAS) devices it utilizes Asynchronous Notification mechanism to receive events from hardware. Active polling with TEST UNIT READY commands with 3 seconds period is used for incapable hardware. After that both CD and DA drivers work the same way, detecting two conditions: "NOT READY: Medium not present" after medium was detected previously, and "UNIT ATTENTION: Not ready to ready change, medium may have changed". First one reported to disk(9) as media removal, second as media insert/change. To reliably receive second event new AC_UNIT_ATTENTION async added to make UAs broadcasted to all periphs by generic error handling code in cam_periph_error(). 2) Modifications to GEOM core to handle media remove and change events. Media removal handled by spoiling all consumers attached to the provider. Media change event also schedules provider retaste after spoiling to probe new media. New flag G_CF_ORPHAN was added to consumers to reflect that consumer is in process of destruction. It allows retaste to create new geom instance of the same class, while previous one is still dying. 3) Modifications to some GEOM classes: DEV -- to report media change events to devd; VFS -- to handle spoiling same as orphan to prevent accessing replaced media. PART class already handles spoiling alike to orphan. Reviewed by: silence on geom@ and scsi@ Tested by: avg Sponsored by: iXsystems, Inc. / PC-BSD MFC after: 2 months
2012-07-29 11:51:48 +00:00
cp->flags |= G_CF_ORPHAN;
Add a new GEOM method, resize(), which is called after provider size changes. Add a new routine, g_resize_provider(), to use to notify GEOM about provider change. Reviewed by: mav Sponsored by: FreeBSD Foundation
2012-07-07 20:13:40 +00:00
cp->geom->orphan(cp);
Implement media change notification for DA and CD removable media devices. It includes three parts: 1) Modifications to CAM to detect media media changes and report them to disk(9) layer. For modern SATA (and potentially UAS) devices it utilizes Asynchronous Notification mechanism to receive events from hardware. Active polling with TEST UNIT READY commands with 3 seconds period is used for incapable hardware. After that both CD and DA drivers work the same way, detecting two conditions: "NOT READY: Medium not present" after medium was detected previously, and "UNIT ATTENTION: Not ready to ready change, medium may have changed". First one reported to disk(9) as media removal, second as media insert/change. To reliably receive second event new AC_UNIT_ATTENTION async added to make UAs broadcasted to all periphs by generic error handling code in cam_periph_error(). 2) Modifications to GEOM core to handle media remove and change events. Media removal handled by spoiling all consumers attached to the provider. Media change event also schedules provider retaste after spoiling to probe new media. New flag G_CF_ORPHAN was added to consumers to reflect that consumer is in process of destruction. It allows retaste to create new geom instance of the same class, while previous one is still dying. 3) Modifications to some GEOM classes: DEV -- to report media change events to devd; VFS -- to handle spoiling same as orphan to prevent accessing replaced media. PART class already handles spoiling alike to orphan. Reviewed by: silence on geom@ and scsi@ Tested by: avg Sponsored by: iXsystems, Inc. / PC-BSD MFC after: 2 months
2012-07-29 11:51:48 +00:00
}
Add a new GEOM method, resize(), which is called after provider size changes. Add a new routine, g_resize_provider(), to use to notify GEOM about provider change. Reviewed by: mav Sponsored by: FreeBSD Foundation
2012-07-07 20:13:40 +00:00
}
pp->mediasize = size;
LIST_FOREACH_SAFE(cp, &pp->consumers, consumers, cp2) {
gp = cp->geom;
Do not invoke resize method if geom is being withered. PR: 211028 MFC after: 2 weeks
2016-07-25 09:12:08 +00:00
if ((gp->flags & G_GEOM_WITHER) == 0 && gp->resize != NULL)
Add a new GEOM method, resize(), which is called after provider size changes. Add a new routine, g_resize_provider(), to use to notify GEOM about provider change. Reviewed by: mav Sponsored by: FreeBSD Foundation
2012-07-07 20:13:40 +00:00
gp->resize(cp);
}
/*
* After resizing, the previously invalid GEOM class metadata
* might become valid. This means we should retaste.
*/
LIST_FOREACH(mp, &g_classes, class) {
if (mp->taste == NULL)
continue;
LIST_FOREACH(cp, &pp->consumers, consumers)
Implement media change notification for DA and CD removable media devices. It includes three parts: 1) Modifications to CAM to detect media media changes and report them to disk(9) layer. For modern SATA (and potentially UAS) devices it utilizes Asynchronous Notification mechanism to receive events from hardware. Active polling with TEST UNIT READY commands with 3 seconds period is used for incapable hardware. After that both CD and DA drivers work the same way, detecting two conditions: "NOT READY: Medium not present" after medium was detected previously, and "UNIT ATTENTION: Not ready to ready change, medium may have changed". First one reported to disk(9) as media removal, second as media insert/change. To reliably receive second event new AC_UNIT_ATTENTION async added to make UAs broadcasted to all periphs by generic error handling code in cam_periph_error(). 2) Modifications to GEOM core to handle media remove and change events. Media removal handled by spoiling all consumers attached to the provider. Media change event also schedules provider retaste after spoiling to probe new media. New flag G_CF_ORPHAN was added to consumers to reflect that consumer is in process of destruction. It allows retaste to create new geom instance of the same class, while previous one is still dying. 3) Modifications to some GEOM classes: DEV -- to report media change events to devd; VFS -- to handle spoiling same as orphan to prevent accessing replaced media. PART class already handles spoiling alike to orphan. Reviewed by: silence on geom@ and scsi@ Tested by: avg Sponsored by: iXsystems, Inc. / PC-BSD MFC after: 2 months
2012-07-29 11:51:48 +00:00
if (cp->geom->class == mp &&
(cp->flags & G_CF_ORPHAN) == 0)
Add a new GEOM method, resize(), which is called after provider size changes. Add a new routine, g_resize_provider(), to use to notify GEOM about provider change. Reviewed by: mav Sponsored by: FreeBSD Foundation
2012-07-07 20:13:40 +00:00
break;
if (cp != NULL)
continue;
mp->taste(mp, pp, 0);
g_topology_assert();
}
}
void
g_resize_provider(struct g_provider *pp, off_t size)
{
struct g_hh00 *hh;
G_VALID_PROVIDER(pp);
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
2016-06-22 14:39:13 +00:00
if (pp->flags & G_PF_WITHER)
return;
Add a new GEOM method, resize(), which is called after provider size changes. Add a new routine, g_resize_provider(), to use to notify GEOM about provider change. Reviewed by: mav Sponsored by: FreeBSD Foundation
2012-07-07 20:13:40 +00:00
if (size == pp->mediasize)
return;
hh = g_malloc(sizeof *hh, M_WAITOK | M_ZERO);
hh->pp = pp;
hh->size = size;
g_post_event(g_resize_provider_event, hh, M_WAITOK, NULL);
}
Allow to pass providers with /dev/ prefix to g_provider_by_name(). MFC after: 3 days
2012-09-01 10:52:19 +00:00
#ifndef _PATH_DEV
#define _PATH_DEV "/dev/"
#endif
Introduce g_provider_by_name() function, and use it.
2003-06-04 18:17:52 +00:00
struct g_provider *
g_provider_by_name(char const *arg)
{
struct g_class *cp;
struct g_geom *gp;
When searching for provider by name, prefer non-withered one. MFC after: 2 weeks
2015-03-26 11:02:29 +00:00
struct g_provider *pp, *wpp;
Introduce g_provider_by_name() function, and use it.
2003-06-04 18:17:52 +00:00
Allow to pass providers with /dev/ prefix to g_provider_by_name(). MFC after: 3 days
2012-09-01 10:52:19 +00:00
if (strncmp(arg, _PATH_DEV, sizeof(_PATH_DEV) - 1) == 0)
arg += sizeof(_PATH_DEV) - 1;
When searching for provider by name, prefer non-withered one. MFC after: 2 weeks
2015-03-26 11:02:29 +00:00
wpp = NULL;
Introduce g_provider_by_name() function, and use it.
2003-06-04 18:17:52 +00:00
LIST_FOREACH(cp, &g_classes, class) {
LIST_FOREACH(gp, &cp->geom, geom) {
LIST_FOREACH(pp, &gp->provider, provider) {
When searching for provider by name, prefer non-withered one. MFC after: 2 weeks
2015-03-26 11:02:29 +00:00
if (strcmp(arg, pp->name) != 0)
continue;
if ((gp->flags & G_GEOM_WITHER) == 0 &&
(pp->flags & G_PF_WITHER) == 0)
Introduce g_provider_by_name() function, and use it.
2003-06-04 18:17:52 +00:00
return (pp);
When searching for provider by name, prefer non-withered one. MFC after: 2 weeks
2015-03-26 11:02:29 +00:00
else
wpp = pp;
Introduce g_provider_by_name() function, and use it.
2003-06-04 18:17:52 +00:00
}
}
}
Allow to pass providers with /dev/ prefix to g_provider_by_name(). MFC after: 3 days
2012-09-01 10:52:19 +00:00
When searching for provider by name, prefer non-withered one. MFC after: 2 weeks
2015-03-26 11:02:29 +00:00
return (wpp);
Introduce g_provider_by_name() function, and use it.
2003-06-04 18:17:52 +00:00
}
First commit of the GEOM subsystem to make it easier for people to test and play with this. This is not yet production quality and should be run only on dedicated test boxes. For people who want to develop transformations for GEOM there exist a set of shims to run geom in userland (ask phk@freebsd.org). Reports of all kinds to: phk@freebsd.org Please include in report: dmesg sysctl debug.geomdot sysctl debug.geomconf Known significant limitations: no kernel dump facility. ioctls severely restricted. Sponsored by: DARPA, NAI Labs
2002-03-11 21:42:35 +00:00
void
g_destroy_provider(struct g_provider *pp)
{
struct g_geom *gp;
g_topology_assert();
Rearrange some of the GEOM debugging tools to be more structured. Retire g_sanity() and corresponding debugflag (0x8) Retire g_{stall,release}_events(). Under #ifdef DIAGNOSTIC: Make g_valid_obj() an official function and have it return an an non-zero integer which indicates the kind of object when found. Implement G_VALID_{CLASS,GEOM,CONSUMER,PROVIDER}() macros based on g_valid_obj(). Sprinkle calls to these macros liberally over the infrastructure. Always check that we do not free a live object.
2004-03-10 08:49:08 +00:00
G_VALID_PROVIDER(pp);
First commit of the GEOM subsystem to make it easier for people to test and play with this. This is not yet production quality and should be run only on dedicated test boxes. For people who want to develop transformations for GEOM there exist a set of shims to run geom in userland (ask phk@freebsd.org). Reports of all kinds to: phk@freebsd.org Please include in report: dmesg sysctl debug.geomdot sysctl debug.geomconf Known significant limitations: no kernel dump facility. ioctls severely restricted. Sponsored by: DARPA, NAI Labs
2002-03-11 21:42:35 +00:00
KASSERT(LIST_EMPTY(&pp->consumers),
("g_destroy_provider but attached"));
KASSERT (pp->acr == 0, ("g_destroy_provider with acr"));
KASSERT (pp->acw == 0, ("g_destroy_provider with acw"));
- Assert that we don't send new provider event for a provider which has G_PF_WITHER flag set. - Fix typo in assertion condition (sorry, but I forgot who report that).
2008-05-18 22:50:50 +00:00
KASSERT (pp->ace == 0, ("g_destroy_provider with ace"));
Change events to have an array of "void *" references, and give the event posting functions varargs to fill these. Attribute g_call_me() to appropriate g_geom's where necessary. Add a flag argument to g_call_me() methods which will be used to signal cancellation of events in the future. This commit should be a no-op.
2003-04-02 20:41:18 +00:00
g_cancel_event(pp);
First commit of the GEOM subsystem to make it easier for people to test and play with this. This is not yet production quality and should be run only on dedicated test boxes. For people who want to develop transformations for GEOM there exist a set of shims to run geom in userland (ask phk@freebsd.org). Reports of all kinds to: phk@freebsd.org Please include in report: dmesg sysctl debug.geomdot sysctl debug.geomconf Known significant limitations: no kernel dump facility. ioctls severely restricted. Sponsored by: DARPA, NAI Labs
2002-03-11 21:42:35 +00:00
LIST_REMOVE(pp, provider);
gp = pp->geom;
Retire the GEOM private statistics code and use devstat instead.
2003-03-18 09:42:33 +00:00
devstat_remove_entry(pp->stat);
Fix a bug which causes a panic in daopen(). The panic is caused by a da(4) instance going away while GEOM is still probing it. In this case, the GEOM disk class instance has been created by disk_create(), and the taste of the disk is queued in the GEOM event queue. While that event is queued, the da(4) instance goes away. When the open call comes into the da(4) driver, it dereferences the freed (but non-NULL) peripheral pointer provided by GEOM, which results in a panic. The solution is to add a callback to the GEOM disk code that is called when all of its resources are cleaned up. This is implemented inside GEOM by adding an optional callback that is called when all consumers have detached from a provider, and the provider is about to be deleted. scsi_cd.c, scsi_da.c: In the register routine for the cd(4) and da(4) routines, acquire a reference to the CAM peripheral instance just before we call disk_create(). Use the new GEOM disk d_gone() callback to register a callback (dadiskgonecb()/cddiskgonecb()) that decrements the peripheral reference count once GEOM has finished cleaning up its resources. In the cd(4) driver, clean up open and close behavior slightly. GEOM makes sure we only get one open() and one close call, so there is no need to set an open flag and decrement the reference count if we are not the first open. In the cd(4) driver, use cam_periph_release_locked() in a couple of error scenarios to avoid extra mutex calls. geom.h: Add a new, optional, providergone callback that is called when a provider is about to be deleted. geom_disk.h: Add a new d_gone() callback to the GEOM disk interface. Bump the DISK_VERSION to version 2. This probably should have been done after a couple of previous changes, especially the addition of the d_getattr() callback. geom_disk.c: Add a providergone callback for the disk class, g_disk_providergone(), that calls the user's d_gone() callback if it exists. Bump the DISK_VERSION to 2. geom_subr.c: In g_destroy_provider(), call the providergone callback if it has been provided. In g_new_geomf(), propagate the class's providergone callback to the new geom instance. blkfront.c: Callers of disk_create() are supposed to pass in DISK_VERSION, not an explicit disk API version number. Update the blkfront driver to do that. disk.9: Update the disk(9) man page to include information on the new d_gone() callback, as well as the previously added d_getattr() callback, d_descr field, and HBA PCI ID fields. MFC after: 5 days
2012-06-24 04:29:03 +00:00
/*
* If a callback was provided, send notification that the provider
* is now gone.
*/
if (gp->providergone != NULL)
gp->providergone(pp);
First commit of the GEOM subsystem to make it easier for people to test and play with this. This is not yet production quality and should be run only on dedicated test boxes. For people who want to develop transformations for GEOM there exist a set of shims to run geom in userland (ask phk@freebsd.org). Reports of all kinds to: phk@freebsd.org Please include in report: dmesg sysctl debug.geomdot sysctl debug.geomconf Known significant limitations: no kernel dump facility. ioctls severely restricted. Sponsored by: DARPA, NAI Labs
2002-03-11 21:42:35 +00:00
g_free(pp);
Rework the "withering" mechanism: Introduce g_wither_geom() to do the work in one single place.
2003-05-02 06:15:27 +00:00
if ((gp->flags & G_GEOM_WITHER))
Make withering water tight. When we orphan/wither a provider, an attached geom+consumer could end up being withered as a result and it may be in front of us in the normal object scanning order so we need to do multi-pass. On the other hand, there may be withering stuff we can't get rid off (yet), so we need to keep track of both the existence of withering stuff and if there is more we can do at this time.
2004-07-08 16:17:14 +00:00
g_do_wither();
First commit of the GEOM subsystem to make it easier for people to test and play with this. This is not yet production quality and should be run only on dedicated test boxes. For people who want to develop transformations for GEOM there exist a set of shims to run geom in userland (ask phk@freebsd.org). Reports of all kinds to: phk@freebsd.org Please include in report: dmesg sysctl debug.geomdot sysctl debug.geomconf Known significant limitations: no kernel dump facility. ioctls severely restricted. Sponsored by: DARPA, NAI Labs
2002-03-11 21:42:35 +00:00
}
/*
* We keep the "geoms" list sorted by topological order (== increasing
* numerical rank) at all times.
* When an attach is done, the attaching geoms rank is invalidated
* and it is moved to the tail of the list.
* All geoms later in the sequence has their ranks reevaluated in
* sequence. If we cannot assign rank to a geom because it's
* prerequisites do not have rank, we move that element to the tail
* of the sequence with invalid rank as well.
* At some point we encounter our original geom and if we stil fail
* to assign it a rank, there must be a loop and we fail back to
Improve some on the naming. Submitted by: iedowse
2002-06-09 10:57:34 +00:00
* g_attach() which detach again and calls redo_rank again
First commit of the GEOM subsystem to make it easier for people to test and play with this. This is not yet production quality and should be run only on dedicated test boxes. For people who want to develop transformations for GEOM there exist a set of shims to run geom in userland (ask phk@freebsd.org). Reports of all kinds to: phk@freebsd.org Please include in report: dmesg sysctl debug.geomdot sysctl debug.geomconf Known significant limitations: no kernel dump facility. ioctls severely restricted. Sponsored by: DARPA, NAI Labs
2002-03-11 21:42:35 +00:00
* to fix up the damage.
* It would be much simpler code wise to do it recursively, but we
* can't risk that on the kernel stack.
*/
static int
redo_rank(struct g_geom *gp)
{
struct g_consumer *cp;
struct g_geom *gp1, *gp2;
int n, m;
g_topology_assert();
Rearrange some of the GEOM debugging tools to be more structured. Retire g_sanity() and corresponding debugflag (0x8) Retire g_{stall,release}_events(). Under #ifdef DIAGNOSTIC: Make g_valid_obj() an official function and have it return an an non-zero integer which indicates the kind of object when found. Implement G_VALID_{CLASS,GEOM,CONSUMER,PROVIDER}() macros based on g_valid_obj(). Sprinkle calls to these macros liberally over the infrastructure. Always check that we do not free a live object.
2004-03-10 08:49:08 +00:00
G_VALID_GEOM(gp);
First commit of the GEOM subsystem to make it easier for people to test and play with this. This is not yet production quality and should be run only on dedicated test boxes. For people who want to develop transformations for GEOM there exist a set of shims to run geom in userland (ask phk@freebsd.org). Reports of all kinds to: phk@freebsd.org Please include in report: dmesg sysctl debug.geomdot sysctl debug.geomconf Known significant limitations: no kernel dump facility. ioctls severely restricted. Sponsored by: DARPA, NAI Labs
2002-03-11 21:42:35 +00:00
/* Invalidate this geoms rank and move it to the tail */
gp1 = TAILQ_NEXT(gp, geoms);
if (gp1 != NULL) {
gp->rank = 0;
TAILQ_REMOVE(&geoms, gp, geoms);
TAILQ_INSERT_TAIL(&geoms, gp, geoms);
} else {
gp1 = gp;
}
/* re-rank the rest of the sequence */
for (; gp1 != NULL; gp1 = gp2) {
gp1->rank = 0;
m = 1;
LIST_FOREACH(cp, &gp1->consumer, consumer) {
if (cp->provider == NULL)
continue;
n = cp->provider->geom->rank;
if (n == 0) {
m = 0;
break;
} else if (n >= m)
m = n + 1;
}
gp1->rank = m;
gp2 = TAILQ_NEXT(gp1, geoms);
/* got a rank, moving on */
if (m != 0)
continue;
/* no rank to original geom means loop */
Fix a {} bug which doesn't have any effect yet. Spotted by: jake
2002-04-27 07:07:37 +00:00
if (gp == gp1)
First commit of the GEOM subsystem to make it easier for people to test and play with this. This is not yet production quality and should be run only on dedicated test boxes. For people who want to develop transformations for GEOM there exist a set of shims to run geom in userland (ask phk@freebsd.org). Reports of all kinds to: phk@freebsd.org Please include in report: dmesg sysctl debug.geomdot sysctl debug.geomconf Known significant limitations: no kernel dump facility. ioctls severely restricted. Sponsored by: DARPA, NAI Labs
2002-03-11 21:42:35 +00:00
return (ELOOP);
/* no rank, put it at the end move on */
TAILQ_REMOVE(&geoms, gp1, geoms);
TAILQ_INSERT_TAIL(&geoms, gp1, geoms);
}
return (0);
}
int
g_attach(struct g_consumer *cp, struct g_provider *pp)
{
int error;
g_topology_assert();
Rearrange some of the GEOM debugging tools to be more structured. Retire g_sanity() and corresponding debugflag (0x8) Retire g_{stall,release}_events(). Under #ifdef DIAGNOSTIC: Make g_valid_obj() an official function and have it return an an non-zero integer which indicates the kind of object when found. Implement G_VALID_{CLASS,GEOM,CONSUMER,PROVIDER}() macros based on g_valid_obj(). Sprinkle calls to these macros liberally over the infrastructure. Always check that we do not free a live object.
2004-03-10 08:49:08 +00:00
G_VALID_CONSUMER(cp);
G_VALID_PROVIDER(pp);
Log attach just like we log detach.
2010-02-18 22:27:38 +00:00
g_trace(G_T_TOPOLOGY, "g_attach(%p, %p)", cp, pp);
First commit of the GEOM subsystem to make it easier for people to test and play with this. This is not yet production quality and should be run only on dedicated test boxes. For people who want to develop transformations for GEOM there exist a set of shims to run geom in userland (ask phk@freebsd.org). Reports of all kinds to: phk@freebsd.org Please include in report: dmesg sysctl debug.geomdot sysctl debug.geomconf Known significant limitations: no kernel dump facility. ioctls severely restricted. Sponsored by: DARPA, NAI Labs
2002-03-11 21:42:35 +00:00
KASSERT(cp->provider == NULL, ("attach but attached"));
cp->provider = pp;
Clear G_CF_ORPHAN when attaching. This fixes cases where the same GEOM consumer can be orphaned, and then reattach to another provider. From a user point of view, this makes gmountver(4) work again. Reviewed by: avg, mav MFC after: 2 weeks Sponsored by: DARPA, AFRL Differential Revision: https://reviews.freebsd.org/D12228
2017-10-02 11:57:00 +00:00
cp->flags &= ~G_CF_ORPHAN;
First commit of the GEOM subsystem to make it easier for people to test and play with this. This is not yet production quality and should be run only on dedicated test boxes. For people who want to develop transformations for GEOM there exist a set of shims to run geom in userland (ask phk@freebsd.org). Reports of all kinds to: phk@freebsd.org Please include in report: dmesg sysctl debug.geomdot sysctl debug.geomconf Known significant limitations: no kernel dump facility. ioctls severely restricted. Sponsored by: DARPA, NAI Labs
2002-03-11 21:42:35 +00:00
LIST_INSERT_HEAD(&pp->consumers, cp, consumers);
error = redo_rank(cp->geom);
if (error) {
LIST_REMOVE(cp, consumers);
cp->provider = NULL;
redo_rank(cp->geom);
}
return (error);
}
void
Improve some on the naming. Submitted by: iedowse
2002-06-09 10:57:34 +00:00
g_detach(struct g_consumer *cp)
First commit of the GEOM subsystem to make it easier for people to test and play with this. This is not yet production quality and should be run only on dedicated test boxes. For people who want to develop transformations for GEOM there exist a set of shims to run geom in userland (ask phk@freebsd.org). Reports of all kinds to: phk@freebsd.org Please include in report: dmesg sysctl debug.geomdot sysctl debug.geomconf Known significant limitations: no kernel dump facility. ioctls severely restricted. Sponsored by: DARPA, NAI Labs
2002-03-11 21:42:35 +00:00
{
struct g_provider *pp;
g_topology_assert();
Rearrange some of the GEOM debugging tools to be more structured. Retire g_sanity() and corresponding debugflag (0x8) Retire g_{stall,release}_events(). Under #ifdef DIAGNOSTIC: Make g_valid_obj() an official function and have it return an an non-zero integer which indicates the kind of object when found. Implement G_VALID_{CLASS,GEOM,CONSUMER,PROVIDER}() macros based on g_valid_obj(). Sprinkle calls to these macros liberally over the infrastructure. Always check that we do not free a live object.
2004-03-10 08:49:08 +00:00
G_VALID_CONSUMER(cp);
g_trace(G_T_TOPOLOGY, "g_detach(%p)", cp);
Improve some on the naming. Submitted by: iedowse
2002-06-09 10:57:34 +00:00
KASSERT(cp->provider != NULL, ("detach but not attached"));
KASSERT(cp->acr == 0, ("detach but nonzero acr"));
KASSERT(cp->acw == 0, ("detach but nonzero acw"));
KASSERT(cp->ace == 0, ("detach but nonzero ace"));
Don't use statistics counters to detect outstanding I/O.
2003-03-09 10:01:16 +00:00
KASSERT(cp->nstart == cp->nend,
Move the g_stat struct to its own .h file, we will export it to other code. Insted of embedding a struct g_stat in consumers and providers, merely include a pointer. Remove a couple of <sys/time.h> includes now unneeded. Add a special allocator for struct g_stat. This allocator will allocate entire pages and hand out g_stat functions from there. The "id" field indicates free/used status. Add "/dev/geom.stats" device driver whic exports the pages from the allocator to userland with mmap(2) in read-only mode. This mmap(2) interface should be considered a non-public interface and the functions in libgeom (not yet committed) should be used to access the statistics data.
2003-02-08 13:03:57 +00:00
("detach with active requests"));
First commit of the GEOM subsystem to make it easier for people to test and play with this. This is not yet production quality and should be run only on dedicated test boxes. For people who want to develop transformations for GEOM there exist a set of shims to run geom in userland (ask phk@freebsd.org). Reports of all kinds to: phk@freebsd.org Please include in report: dmesg sysctl debug.geomdot sysctl debug.geomconf Known significant limitations: no kernel dump facility. ioctls severely restricted. Sponsored by: DARPA, NAI Labs
2002-03-11 21:42:35 +00:00
pp = cp->provider;
LIST_REMOVE(cp, consumers);
cp->provider = NULL;
Make g_wither_washer() to not loop by itself, but only when there was some more topology change done that may require its attention. Add few missing g_do_wither() calls in respective places to signal it. This fixes potential infinite loop here when some provider is withered, but still opened or connected for some reason and so can not be destroyed. For example, see r227009 and r227510.
2013-03-24 03:15:20 +00:00
if ((cp->geom->flags & G_GEOM_WITHER) ||
(pp->geom->flags & G_GEOM_WITHER) ||
(pp->flags & G_PF_WITHER))
Make withering water tight. When we orphan/wither a provider, an attached geom+consumer could end up being withered as a result and it may be in front of us in the normal object scanning order so we need to do multi-pass. On the other hand, there may be withering stuff we can't get rid off (yet), so we need to keep track of both the existence of withering stuff and if there is more we can do at this time.
2004-07-08 16:17:14 +00:00
g_do_wither();
First commit of the GEOM subsystem to make it easier for people to test and play with this. This is not yet production quality and should be run only on dedicated test boxes. For people who want to develop transformations for GEOM there exist a set of shims to run geom in userland (ask phk@freebsd.org). Reports of all kinds to: phk@freebsd.org Please include in report: dmesg sysctl debug.geomdot sysctl debug.geomconf Known significant limitations: no kernel dump facility. ioctls severely restricted. Sponsored by: DARPA, NAI Labs
2002-03-11 21:42:35 +00:00
redo_rank(cp->geom);
}
/*
Remove the absolute count g_access_abs() function since experience has shown that it is not useful. Rename the relative count g_access_rel() function to g_access(), only the name has changed. Change all g_access_rel() calls in our CVS tree to call g_access() instead. Add an #ifndef BURN_BRIDGES #define of g_access_rel() for source code compatibility.
2004-02-12 22:42:11 +00:00
* g_access()
First commit of the GEOM subsystem to make it easier for people to test and play with this. This is not yet production quality and should be run only on dedicated test boxes. For people who want to develop transformations for GEOM there exist a set of shims to run geom in userland (ask phk@freebsd.org). Reports of all kinds to: phk@freebsd.org Please include in report: dmesg sysctl debug.geomdot sysctl debug.geomconf Known significant limitations: no kernel dump facility. ioctls severely restricted. Sponsored by: DARPA, NAI Labs
2002-03-11 21:42:35 +00:00
*
* Access-check with delta values. The question asked is "can provider
* "cp" change the access counters by the relative amounts dc[rwe] ?"
*/
int
Remove the absolute count g_access_abs() function since experience has shown that it is not useful. Rename the relative count g_access_rel() function to g_access(), only the name has changed. Change all g_access_rel() calls in our CVS tree to call g_access() instead. Add an #ifndef BURN_BRIDGES #define of g_access_rel() for source code compatibility.
2004-02-12 22:42:11 +00:00
g_access(struct g_consumer *cp, int dcr, int dcw, int dce)
First commit of the GEOM subsystem to make it easier for people to test and play with this. This is not yet production quality and should be run only on dedicated test boxes. For people who want to develop transformations for GEOM there exist a set of shims to run geom in userland (ask phk@freebsd.org). Reports of all kinds to: phk@freebsd.org Please include in report: dmesg sysctl debug.geomdot sysctl debug.geomconf Known significant limitations: no kernel dump facility. ioctls severely restricted. Sponsored by: DARPA, NAI Labs
2002-03-11 21:42:35 +00:00
{
struct g_provider *pp;
int pr,pw,pe;
int error;
Rearrange some of the GEOM debugging tools to be more structured. Retire g_sanity() and corresponding debugflag (0x8) Retire g_{stall,release}_events(). Under #ifdef DIAGNOSTIC: Make g_valid_obj() an official function and have it return an an non-zero integer which indicates the kind of object when found. Implement G_VALID_{CLASS,GEOM,CONSUMER,PROVIDER}() macros based on g_valid_obj(). Sprinkle calls to these macros liberally over the infrastructure. Always check that we do not free a live object.
2004-03-10 08:49:08 +00:00
g_topology_assert();
G_VALID_CONSUMER(cp);
First commit of the GEOM subsystem to make it easier for people to test and play with this. This is not yet production quality and should be run only on dedicated test boxes. For people who want to develop transformations for GEOM there exist a set of shims to run geom in userland (ask phk@freebsd.org). Reports of all kinds to: phk@freebsd.org Please include in report: dmesg sysctl debug.geomdot sysctl debug.geomconf Known significant limitations: no kernel dump facility. ioctls severely restricted. Sponsored by: DARPA, NAI Labs
2002-03-11 21:42:35 +00:00
pp = cp->provider;
Move "is consumer attached?" check before G_VALID_PROVIDER() check, because if consumer is not attached, its provider never will be valid, so we never reach this check. Approved by: phk
2004-03-18 07:17:10 +00:00
KASSERT(pp != NULL, ("access but not attached"));
Rearrange some of the GEOM debugging tools to be more structured. Retire g_sanity() and corresponding debugflag (0x8) Retire g_{stall,release}_events(). Under #ifdef DIAGNOSTIC: Make g_valid_obj() an official function and have it return an an non-zero integer which indicates the kind of object when found. Implement G_VALID_{CLASS,GEOM,CONSUMER,PROVIDER}() macros based on g_valid_obj(). Sprinkle calls to these macros liberally over the infrastructure. Always check that we do not free a live object.
2004-03-10 08:49:08 +00:00
G_VALID_PROVIDER(pp);
First commit of the GEOM subsystem to make it easier for people to test and play with this. This is not yet production quality and should be run only on dedicated test boxes. For people who want to develop transformations for GEOM there exist a set of shims to run geom in userland (ask phk@freebsd.org). Reports of all kinds to: phk@freebsd.org Please include in report: dmesg sysctl debug.geomdot sysctl debug.geomconf Known significant limitations: no kernel dump facility. ioctls severely restricted. Sponsored by: DARPA, NAI Labs
2002-03-11 21:42:35 +00:00
Remove the absolute count g_access_abs() function since experience has shown that it is not useful. Rename the relative count g_access_rel() function to g_access(), only the name has changed. Change all g_access_rel() calls in our CVS tree to call g_access() instead. Add an #ifndef BURN_BRIDGES #define of g_access_rel() for source code compatibility.
2004-02-12 22:42:11 +00:00
g_trace(G_T_ACCESS, "g_access(%p(%s), %d, %d, %d)",
First commit of the GEOM subsystem to make it easier for people to test and play with this. This is not yet production quality and should be run only on dedicated test boxes. For people who want to develop transformations for GEOM there exist a set of shims to run geom in userland (ask phk@freebsd.org). Reports of all kinds to: phk@freebsd.org Please include in report: dmesg sysctl debug.geomdot sysctl debug.geomconf Known significant limitations: no kernel dump facility. ioctls severely restricted. Sponsored by: DARPA, NAI Labs
2002-03-11 21:42:35 +00:00
cp, pp->name, dcr, dcw, dce);
KASSERT(cp->acr + dcr >= 0, ("access resulting in negative acr"));
KASSERT(cp->acw + dcw >= 0, ("access resulting in negative acw"));
KASSERT(cp->ace + dce >= 0, ("access resulting in negative ace"));
KASSERT against no-op access requests. Submitted by: Pawel Jakub Dawidek <nick@garage.freebsd.pl>
2004-01-09 16:10:32 +00:00
KASSERT(dcr != 0 || dcw != 0 || dce != 0, ("NOP access request"));
Move access and orphan member functions from class to geom. Sponsored by: DARPA & NAI Labs
2002-04-04 09:54:13 +00:00
KASSERT(pp->geom->access != NULL, ("NULL geom->access"));
First commit of the GEOM subsystem to make it easier for people to test and play with this. This is not yet production quality and should be run only on dedicated test boxes. For people who want to develop transformations for GEOM there exist a set of shims to run geom in userland (ask phk@freebsd.org). Reports of all kinds to: phk@freebsd.org Please include in report: dmesg sysctl debug.geomdot sysctl debug.geomconf Known significant limitations: no kernel dump facility. ioctls severely restricted. Sponsored by: DARPA, NAI Labs
2002-03-11 21:42:35 +00:00
/*
Cave in to tradition and rename "methods" to "classes".
2002-03-26 21:40:06 +00:00
* If our class cares about being spoiled, and we have been, we
First commit of the GEOM subsystem to make it easier for people to test and play with this. This is not yet production quality and should be run only on dedicated test boxes. For people who want to develop transformations for GEOM there exist a set of shims to run geom in userland (ask phk@freebsd.org). Reports of all kinds to: phk@freebsd.org Please include in report: dmesg sysctl debug.geomdot sysctl debug.geomconf Known significant limitations: no kernel dump facility. ioctls severely restricted. Sponsored by: DARPA, NAI Labs
2002-03-11 21:42:35 +00:00
* are probably just ahead of the event telling us that. Fail
* now rather than having to unravel this later.
*/
Implement media change notification for DA and CD removable media devices. It includes three parts: 1) Modifications to CAM to detect media media changes and report them to disk(9) layer. For modern SATA (and potentially UAS) devices it utilizes Asynchronous Notification mechanism to receive events from hardware. Active polling with TEST UNIT READY commands with 3 seconds period is used for incapable hardware. After that both CD and DA drivers work the same way, detecting two conditions: "NOT READY: Medium not present" after medium was detected previously, and "UNIT ATTENTION: Not ready to ready change, medium may have changed". First one reported to disk(9) as media removal, second as media insert/change. To reliably receive second event new AC_UNIT_ATTENTION async added to make UAs broadcasted to all periphs by generic error handling code in cam_periph_error(). 2) Modifications to GEOM core to handle media remove and change events. Media removal handled by spoiling all consumers attached to the provider. Media change event also schedules provider retaste after spoiling to probe new media. New flag G_CF_ORPHAN was added to consumers to reflect that consumer is in process of destruction. It allows retaste to create new geom instance of the same class, while previous one is still dying. 3) Modifications to some GEOM classes: DEV -- to report media change events to devd; VFS -- to handle spoiling same as orphan to prevent accessing replaced media. PART class already handles spoiling alike to orphan. Reviewed by: silence on geom@ and scsi@ Tested by: avg Sponsored by: iXsystems, Inc. / PC-BSD MFC after: 2 months
2012-07-29 11:51:48 +00:00
if (cp->geom->spoiled != NULL && (cp->flags & G_CF_SPOILED) &&
Fail normally rather than KASSERT if attempt to open a spoiled consumer.
2004-07-08 10:34:09 +00:00
(dcr > 0 || dcw > 0 || dce > 0))
return (ENXIO);
First commit of the GEOM subsystem to make it easier for people to test and play with this. This is not yet production quality and should be run only on dedicated test boxes. For people who want to develop transformations for GEOM there exist a set of shims to run geom in userland (ask phk@freebsd.org). Reports of all kinds to: phk@freebsd.org Please include in report: dmesg sysctl debug.geomdot sysctl debug.geomconf Known significant limitations: no kernel dump facility. ioctls severely restricted. Sponsored by: DARPA, NAI Labs
2002-03-11 21:42:35 +00:00
/*
* Figure out what counts the provider would have had, if this
* consumer had (r0w0e0) at this time.
*/
pr = pp->acr - cp->acr;
pw = pp->acw - cp->acw;
pe = pp->ace - cp->ace;
g_trace(G_T_ACCESS,
"open delta:[r%dw%de%d] old:[r%dw%de%d] provider:[r%dw%de%d] %p(%s)",
dcr, dcw, dce,
cp->acr, cp->acw, cp->ace,
pp->acr, pp->acw, pp->ace,
pp, pp->name);
Implement a handle for efficient implementation of perforations in lower extremities. Setting bit 4 in debugflags (sysctl kern.geom.debugflags=16) will allow any open to succeed on rank#1 providers. This will generally correspond to the physical disk devices: ad0, da0, md0 etc. This fundamentally violates the mechanics of GEOMs autoconfiguration, and is only provided as a debugging facility, so obviously error reports on GEOM where this bit is or has been set will not be accepted.
2003-02-12 09:48:27 +00:00
/* If foot-shooting is enabled, any open on rank#1 is OK */
if ((g_debugflags & 16) && pp->geom->rank == 1)
;
First commit of the GEOM subsystem to make it easier for people to test and play with this. This is not yet production quality and should be run only on dedicated test boxes. For people who want to develop transformations for GEOM there exist a set of shims to run geom in userland (ask phk@freebsd.org). Reports of all kinds to: phk@freebsd.org Please include in report: dmesg sysctl debug.geomdot sysctl debug.geomconf Known significant limitations: no kernel dump facility. ioctls severely restricted. Sponsored by: DARPA, NAI Labs
2002-03-11 21:42:35 +00:00
/* If we try exclusive but already write: fail */
Implement a handle for efficient implementation of perforations in lower extremities. Setting bit 4 in debugflags (sysctl kern.geom.debugflags=16) will allow any open to succeed on rank#1 providers. This will generally correspond to the physical disk devices: ad0, da0, md0 etc. This fundamentally violates the mechanics of GEOMs autoconfiguration, and is only provided as a debugging facility, so obviously error reports on GEOM where this bit is or has been set will not be accepted.
2003-02-12 09:48:27 +00:00
else if (dce > 0 && pw > 0)
First commit of the GEOM subsystem to make it easier for people to test and play with this. This is not yet production quality and should be run only on dedicated test boxes. For people who want to develop transformations for GEOM there exist a set of shims to run geom in userland (ask phk@freebsd.org). Reports of all kinds to: phk@freebsd.org Please include in report: dmesg sysctl debug.geomdot sysctl debug.geomconf Known significant limitations: no kernel dump facility. ioctls severely restricted. Sponsored by: DARPA, NAI Labs
2002-03-11 21:42:35 +00:00
return (EPERM);
/* If we try write but already exclusive: fail */
Implement a handle for efficient implementation of perforations in lower extremities. Setting bit 4 in debugflags (sysctl kern.geom.debugflags=16) will allow any open to succeed on rank#1 providers. This will generally correspond to the physical disk devices: ad0, da0, md0 etc. This fundamentally violates the mechanics of GEOMs autoconfiguration, and is only provided as a debugging facility, so obviously error reports on GEOM where this bit is or has been set will not be accepted.
2003-02-12 09:48:27 +00:00
else if (dcw > 0 && pe > 0)
First commit of the GEOM subsystem to make it easier for people to test and play with this. This is not yet production quality and should be run only on dedicated test boxes. For people who want to develop transformations for GEOM there exist a set of shims to run geom in userland (ask phk@freebsd.org). Reports of all kinds to: phk@freebsd.org Please include in report: dmesg sysctl debug.geomdot sysctl debug.geomconf Known significant limitations: no kernel dump facility. ioctls severely restricted. Sponsored by: DARPA, NAI Labs
2002-03-11 21:42:35 +00:00
return (EPERM);
/* If we try to open more but provider is error'ed: fail */
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
2016-06-22 14:39:13 +00:00
else if ((dcr > 0 || dcw > 0 || dce > 0) && pp->error != 0) {
Let g_access() log the actual error number. Submitted by: Fabian Keil <fk@fabiankeil.de> PR: 221855 MFC after: 1 week
2017-08-27 12:24:25 +00:00
printf("%s(%d): provider %s has error %d set\n",
__func__, __LINE__, pp->name, pp->error);
First commit of the GEOM subsystem to make it easier for people to test and play with this. This is not yet production quality and should be run only on dedicated test boxes. For people who want to develop transformations for GEOM there exist a set of shims to run geom in userland (ask phk@freebsd.org). Reports of all kinds to: phk@freebsd.org Please include in report: dmesg sysctl debug.geomdot sysctl debug.geomconf Known significant limitations: no kernel dump facility. ioctls severely restricted. Sponsored by: DARPA, NAI Labs
2002-03-11 21:42:35 +00:00
return (pp->error);
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
2016-06-22 14:39:13 +00:00
}
First commit of the GEOM subsystem to make it easier for people to test and play with this. This is not yet production quality and should be run only on dedicated test boxes. For people who want to develop transformations for GEOM there exist a set of shims to run geom in userland (ask phk@freebsd.org). Reports of all kinds to: phk@freebsd.org Please include in report: dmesg sysctl debug.geomdot sysctl debug.geomconf Known significant limitations: no kernel dump facility. ioctls severely restricted. Sponsored by: DARPA, NAI Labs
2002-03-11 21:42:35 +00:00
/* Ok then... */
Move access and orphan member functions from class to geom. Sponsored by: DARPA & NAI Labs
2002-04-04 09:54:13 +00:00
error = pp->geom->access(pp, dcr, dcw, dce);
Add a KASSERT which checks that a class never fails a closing ->access() call.
2004-02-14 17:58:57 +00:00
KASSERT(dcr > 0 || dcw > 0 || dce > 0 || error == 0,
Make g_access() KASSERT() more useful. Sponsored by: EMC / Isilon Storage Division Obtained from: Isilon OneFS MFC after: 2 weeks
2014-04-15 14:41:41 +00:00
("Geom provider %s::%s dcr=%d dcw=%d dce=%d error=%d failed "
"closing ->access()", pp->geom->class->name, pp->name, dcr, dcw,
dce, error));
First commit of the GEOM subsystem to make it easier for people to test and play with this. This is not yet production quality and should be run only on dedicated test boxes. For people who want to develop transformations for GEOM there exist a set of shims to run geom in userland (ask phk@freebsd.org). Reports of all kinds to: phk@freebsd.org Please include in report: dmesg sysctl debug.geomdot sysctl debug.geomconf Known significant limitations: no kernel dump facility. ioctls severely restricted. Sponsored by: DARPA, NAI Labs
2002-03-11 21:42:35 +00:00
if (!error) {
Only orphan things if the open/close actually succeeded.
2003-04-02 13:10:40 +00:00
/*
* If we open first write, spoil any partner consumers.
Fix a bug that caused some /dev entries to continue to exist after the underlying drive had been hot-unplugged from the system. Here is a specific example. Filesystem code had opened /dev/da1s1e. Subsequently, the drive was hot-unplugged. This (correctly) caused all of the associated /dev/da1* entries to be deleted. When the filesystem later realized that the drive was gone it closed the device, reducing the write-access counts to 0 on the geom providers for da1s1e, da1s1, and da1. This caused geom to re-taste the providers, resulting in the devices being created again. When the drive was hot-plugged back in, it resulted in duplicate /dev entries for da1s1e, da1s1, and da1. This fix adds a new disk_gone() function which is called by CAM when a drive goes away. It orphans all of the providers associated with the drive, setting an error condition of ENXIO in each one. In addition, we prevent a re-taste on last close for writing if an error condition has been set in the provider. Sponsored by: Isilon Systems Reviewed by: phk MFC after: 1 week
2005-11-18 02:43:49 +00:00
* If we close last write and provider is not errored,
* trigger re-taste.
Only orphan things if the open/close actually succeeded.
2003-04-02 13:10:40 +00:00
*/
if (pp->acw == 0 && dcw != 0)
g_spoil(pp, cp);
Fix a bug that caused some /dev entries to continue to exist after the underlying drive had been hot-unplugged from the system. Here is a specific example. Filesystem code had opened /dev/da1s1e. Subsequently, the drive was hot-unplugged. This (correctly) caused all of the associated /dev/da1* entries to be deleted. When the filesystem later realized that the drive was gone it closed the device, reducing the write-access counts to 0 on the geom providers for da1s1e, da1s1, and da1. This caused geom to re-taste the providers, resulting in the devices being created again. When the drive was hot-plugged back in, it resulted in duplicate /dev entries for da1s1e, da1s1, and da1. This fix adds a new disk_gone() function which is called by CAM when a drive goes away. It orphans all of the providers associated with the drive, setting an error condition of ENXIO in each one. In addition, we prevent a re-taste on last close for writing if an error condition has been set in the provider. Sponsored by: Isilon Systems Reviewed by: phk MFC after: 1 week
2005-11-18 02:43:49 +00:00
else if (pp->acw != 0 && pp->acw == -dcw && pp->error == 0 &&
Only orphan things if the open/close actually succeeded.
2003-04-02 13:10:40 +00:00
!(pp->geom->flags & G_GEOM_WITHER))
Rename g_call_me() to g_post_event(), and give it a flag argument to determine if we can M_WAITOK in malloc.
2003-04-23 20:46:12 +00:00
g_post_event(g_new_provider_event, pp, M_WAITOK,
pp, NULL);
Only orphan things if the open/close actually succeeded.
2003-04-02 13:10:40 +00:00
First commit of the GEOM subsystem to make it easier for people to test and play with this. This is not yet production quality and should be run only on dedicated test boxes. For people who want to develop transformations for GEOM there exist a set of shims to run geom in userland (ask phk@freebsd.org). Reports of all kinds to: phk@freebsd.org Please include in report: dmesg sysctl debug.geomdot sysctl debug.geomconf Known significant limitations: no kernel dump facility. ioctls severely restricted. Sponsored by: DARPA, NAI Labs
2002-03-11 21:42:35 +00:00
pp->acr += dcr;
pp->acw += dcw;
pp->ace += dce;
cp->acr += dcr;
cp->acw += dcw;
cp->ace += dce;
For removable devices without media we set a zero mediasize but a non-zero sectorsize in order to avoid a lot of checks around various divisions etc. Enforce the sectorsize being > 0 with a KASSERT on successful open. Fix scsi_cd.c to return 2k sectors when no media inserted.
2004-09-05 21:15:58 +00:00
if (pp->acr != 0 || pp->acw != 0 || pp->ace != 0)
KASSERT(pp->sectorsize > 0,
("Provider %s lacks sectorsize", pp->name));
Make g_wither_washer() to not loop by itself, but only when there was some more topology change done that may require its attention. Add few missing g_do_wither() calls in respective places to signal it. This fixes potential infinite loop here when some provider is withered, but still opened or connected for some reason and so can not be destroyed. For example, see r227009 and r227510.
2013-03-24 03:15:20 +00:00
if ((cp->geom->flags & G_GEOM_WITHER) &&
cp->acr == 0 && cp->acw == 0 && cp->ace == 0)
g_do_wither();
First commit of the GEOM subsystem to make it easier for people to test and play with this. This is not yet production quality and should be run only on dedicated test boxes. For people who want to develop transformations for GEOM there exist a set of shims to run geom in userland (ask phk@freebsd.org). Reports of all kinds to: phk@freebsd.org Please include in report: dmesg sysctl debug.geomdot sysctl debug.geomconf Known significant limitations: no kernel dump facility. ioctls severely restricted. Sponsored by: DARPA, NAI Labs
2002-03-11 21:42:35 +00:00
}
return (error);
}
int
Constification and some s/int/u_int/ changes.
2002-12-16 22:33:27 +00:00
g_handleattr_int(struct bio *bp, const char *attribute, int val)
First commit of the GEOM subsystem to make it easier for people to test and play with this. This is not yet production quality and should be run only on dedicated test boxes. For people who want to develop transformations for GEOM there exist a set of shims to run geom in userland (ask phk@freebsd.org). Reports of all kinds to: phk@freebsd.org Please include in report: dmesg sysctl debug.geomdot sysctl debug.geomconf Known significant limitations: no kernel dump facility. ioctls severely restricted. Sponsored by: DARPA, NAI Labs
2002-03-11 21:42:35 +00:00
{
Improve some on the naming. Submitted by: iedowse
2002-06-09 10:57:34 +00:00
return (g_handleattr(bp, attribute, &val, sizeof val));
First commit of the GEOM subsystem to make it easier for people to test and play with this. This is not yet production quality and should be run only on dedicated test boxes. For people who want to develop transformations for GEOM there exist a set of shims to run geom in userland (ask phk@freebsd.org). Reports of all kinds to: phk@freebsd.org Please include in report: dmesg sysctl debug.geomdot sysctl debug.geomconf Known significant limitations: no kernel dump facility. ioctls severely restricted. Sponsored by: DARPA, NAI Labs
2002-03-11 21:42:35 +00:00
}
Improve ZFS N-way mirror read performance by using load and locality information. The existing algorithm selects a preferred leaf vdev based on offset of the zio request modulo the number of members in the mirror. It assumes the devices are of equal performance and that spreading the requests randomly over both drives will be sufficient to saturate them. In practice this results in the leaf vdevs being under utilized. The new algorithm takes into the following additional factors: * Load of the vdevs (number outstanding I/O requests) * The locality of last queued I/O vs the new I/O request. Within the locality calculation additional knowledge about the underlying vdev is considered such as; is the device backing the vdev a rotating media device. This results in performance increases across the board as well as significant increases for predominantly streaming loads and for configurations which don't have evenly performing devices. The following are results from a setup with 3 Way Mirror with 2 x HD's and 1 x SSD from a basic test running multiple parrallel dd's. With pre-fetch disabled (vfs.zfs.prefetch_disable=1): == Stripe Balanced (default) == Read 15360MB using bs: 1048576, readers: 3, took 161 seconds @ 95 MB/s == Load Balanced (zfslinux) == Read 15360MB using bs: 1048576, readers: 3, took 297 seconds @ 51 MB/s == Load Balanced (locality freebsd) == Read 15360MB using bs: 1048576, readers: 3, took 54 seconds @ 284 MB/s With pre-fetch enabled (vfs.zfs.prefetch_disable=0): == Stripe Balanced (default) == Read 15360MB using bs: 1048576, readers: 3, took 91 seconds @ 168 MB/s == Load Balanced (zfslinux) == Read 15360MB using bs: 1048576, readers: 3, took 108 seconds @ 142 MB/s == Load Balanced (locality freebsd) == Read 15360MB using bs: 1048576, readers: 3, took 48 seconds @ 320 MB/s In addition to the performance changes the code was also restructured, with the help of Justin Gibbs, to provide a more logical flow which also ensures vdevs loads are only calculated from the set of valid candidates. The following additional sysctls where added to allow the administrator to tune the behaviour of the load algorithm: * vfs.zfs.vdev.mirror.rotating_inc * vfs.zfs.vdev.mirror.rotating_seek_inc * vfs.zfs.vdev.mirror.rotating_seek_offset * vfs.zfs.vdev.mirror.non_rotating_inc * vfs.zfs.vdev.mirror.non_rotating_seek_inc These changes where based on work started by the zfsonlinux developers: https://github.com/zfsonlinux/zfs/pull/1487 Reviewed by: gibbs, mav, will MFC after: 2 weeks Sponsored by: Multiplay
2013-10-23 09:54:58 +00:00
int
g_handleattr_uint16_t(struct bio *bp, const char *attribute, uint16_t val)
{
return (g_handleattr(bp, attribute, &val, sizeof val));
}
First commit of the GEOM subsystem to make it easier for people to test and play with this. This is not yet production quality and should be run only on dedicated test boxes. For people who want to develop transformations for GEOM there exist a set of shims to run geom in userland (ask phk@freebsd.org). Reports of all kinds to: phk@freebsd.org Please include in report: dmesg sysctl debug.geomdot sysctl debug.geomconf Known significant limitations: no kernel dump facility. ioctls severely restricted. Sponsored by: DARPA, NAI Labs
2002-03-11 21:42:35 +00:00
int
Constification and some s/int/u_int/ changes.
2002-12-16 22:33:27 +00:00
g_handleattr_off_t(struct bio *bp, const char *attribute, off_t val)
First commit of the GEOM subsystem to make it easier for people to test and play with this. This is not yet production quality and should be run only on dedicated test boxes. For people who want to develop transformations for GEOM there exist a set of shims to run geom in userland (ask phk@freebsd.org). Reports of all kinds to: phk@freebsd.org Please include in report: dmesg sysctl debug.geomdot sysctl debug.geomconf Known significant limitations: no kernel dump facility. ioctls severely restricted. Sponsored by: DARPA, NAI Labs
2002-03-11 21:42:35 +00:00
{
Improve some on the naming. Submitted by: iedowse
2002-06-09 10:57:34 +00:00
return (g_handleattr(bp, attribute, &val, sizeof val));
First commit of the GEOM subsystem to make it easier for people to test and play with this. This is not yet production quality and should be run only on dedicated test boxes. For people who want to develop transformations for GEOM there exist a set of shims to run geom in userland (ask phk@freebsd.org). Reports of all kinds to: phk@freebsd.org Please include in report: dmesg sysctl debug.geomdot sysctl debug.geomconf Known significant limitations: no kernel dump facility. ioctls severely restricted. Sponsored by: DARPA, NAI Labs
2002-03-11 21:42:35 +00:00
}
- Implement helper g_handleattr_str() function for string attributes handling. - Extend g_handleattr() to treat attribute as string when len=0. OK'ed by: phk
2007-05-05 16:33:44 +00:00
int
Constify val in g_handleattr() and str in g_handleattr_str(). This allows passing string constants to g_handleattr_str().
2009-02-01 01:50:09 +00:00
g_handleattr_str(struct bio *bp, const char *attribute, const char *str)
- Implement helper g_handleattr_str() function for string attributes handling. - Extend g_handleattr() to treat attribute as string when len=0. OK'ed by: phk
2007-05-05 16:33:44 +00:00
{
return (g_handleattr(bp, attribute, str, 0));
}
First commit of the GEOM subsystem to make it easier for people to test and play with this. This is not yet production quality and should be run only on dedicated test boxes. For people who want to develop transformations for GEOM there exist a set of shims to run geom in userland (ask phk@freebsd.org). Reports of all kinds to: phk@freebsd.org Please include in report: dmesg sysctl debug.geomdot sysctl debug.geomconf Known significant limitations: no kernel dump facility. ioctls severely restricted. Sponsored by: DARPA, NAI Labs
2002-03-11 21:42:35 +00:00
int
Constify val in g_handleattr() and str in g_handleattr_str(). This allows passing string constants to g_handleattr_str().
2009-02-01 01:50:09 +00:00
g_handleattr(struct bio *bp, const char *attribute, const void *val, int len)
First commit of the GEOM subsystem to make it easier for people to test and play with this. This is not yet production quality and should be run only on dedicated test boxes. For people who want to develop transformations for GEOM there exist a set of shims to run geom in userland (ask phk@freebsd.org). Reports of all kinds to: phk@freebsd.org Please include in report: dmesg sysctl debug.geomdot sysctl debug.geomconf Known significant limitations: no kernel dump facility. ioctls severely restricted. Sponsored by: DARPA, NAI Labs
2002-03-11 21:42:35 +00:00
{
- Implement helper g_handleattr_str() function for string attributes handling. - Extend g_handleattr() to treat attribute as string when len=0. OK'ed by: phk
2007-05-05 16:33:44 +00:00
int error = 0;
First commit of the GEOM subsystem to make it easier for people to test and play with this. This is not yet production quality and should be run only on dedicated test boxes. For people who want to develop transformations for GEOM there exist a set of shims to run geom in userland (ask phk@freebsd.org). Reports of all kinds to: phk@freebsd.org Please include in report: dmesg sysctl debug.geomdot sysctl debug.geomconf Known significant limitations: no kernel dump facility. ioctls severely restricted. Sponsored by: DARPA, NAI Labs
2002-03-11 21:42:35 +00:00
if (strcmp(bp->bio_attribute, attribute))
return (0);
- Implement helper g_handleattr_str() function for string attributes handling. - Extend g_handleattr() to treat attribute as string when len=0. OK'ed by: phk
2007-05-05 16:33:44 +00:00
if (len == 0) {
bzero(bp->bio_data, bp->bio_length);
if (strlcpy(bp->bio_data, val, bp->bio_length) >=
bp->bio_length) {
printf("%s: %s bio_length %jd len %zu -> EFAULT\n",
__func__, bp->bio_to->name,
(intmax_t)bp->bio_length, strlen(val));
error = EFAULT;
}
} else if (bp->bio_length == len) {
First commit of the GEOM subsystem to make it easier for people to test and play with this. This is not yet production quality and should be run only on dedicated test boxes. For people who want to develop transformations for GEOM there exist a set of shims to run geom in userland (ask phk@freebsd.org). Reports of all kinds to: phk@freebsd.org Please include in report: dmesg sysctl debug.geomdot sysctl debug.geomconf Known significant limitations: no kernel dump facility. ioctls severely restricted. Sponsored by: DARPA, NAI Labs
2002-03-11 21:42:35 +00:00
bcopy(val, bp->bio_data, len);
- Implement helper g_handleattr_str() function for string attributes handling. - Extend g_handleattr() to treat attribute as string when len=0. OK'ed by: phk
2007-05-05 16:33:44 +00:00
} else {
printf("%s: %s bio_length %jd len %d -> EFAULT\n", __func__,
bp->bio_to->name, (intmax_t)bp->bio_length, len);
error = EFAULT;
First commit of the GEOM subsystem to make it easier for people to test and play with this. This is not yet production quality and should be run only on dedicated test boxes. For people who want to develop transformations for GEOM there exist a set of shims to run geom in userland (ask phk@freebsd.org). Reports of all kinds to: phk@freebsd.org Please include in report: dmesg sysctl debug.geomdot sysctl debug.geomconf Known significant limitations: no kernel dump facility. ioctls severely restricted. Sponsored by: DARPA, NAI Labs
2002-03-11 21:42:35 +00:00
}
In g_handleattr(), set bp->bio_completed also for the case where len is 0. Otherwise g_getattr() will never succeed when it is handled by g_handleattr_str().
2009-02-03 07:07:13 +00:00
if (error == 0)
bp->bio_completed = bp->bio_length;
Retire g_io_fail() and let g_io_deliver() take an error argument instead. Sponsored by: DARPA & NAI Labs.
2002-09-30 08:54:46 +00:00
g_io_deliver(bp, error);
First commit of the GEOM subsystem to make it easier for people to test and play with this. This is not yet production quality and should be run only on dedicated test boxes. For people who want to develop transformations for GEOM there exist a set of shims to run geom in userland (ask phk@freebsd.org). Reports of all kinds to: phk@freebsd.org Please include in report: dmesg sysctl debug.geomdot sysctl debug.geomconf Known significant limitations: no kernel dump facility. ioctls severely restricted. Sponsored by: DARPA, NAI Labs
2002-03-11 21:42:35 +00:00
return (1);
}
int
Rearrange some of the GEOM debugging tools to be more structured. Retire g_sanity() and corresponding debugflag (0x8) Retire g_{stall,release}_events(). Under #ifdef DIAGNOSTIC: Make g_valid_obj() an official function and have it return an an non-zero integer which indicates the kind of object when found. Implement G_VALID_{CLASS,GEOM,CONSUMER,PROVIDER}() macros based on g_valid_obj(). Sprinkle calls to these macros liberally over the infrastructure. Always check that we do not free a live object.
2004-03-10 08:49:08 +00:00
g_std_access(struct g_provider *pp,
First commit of the GEOM subsystem to make it easier for people to test and play with this. This is not yet production quality and should be run only on dedicated test boxes. For people who want to develop transformations for GEOM there exist a set of shims to run geom in userland (ask phk@freebsd.org). Reports of all kinds to: phk@freebsd.org Please include in report: dmesg sysctl debug.geomdot sysctl debug.geomconf Known significant limitations: no kernel dump facility. ioctls severely restricted. Sponsored by: DARPA, NAI Labs
2002-03-11 21:42:35 +00:00
int dr __unused, int dw __unused, int de __unused)
{
Rearrange some of the GEOM debugging tools to be more structured. Retire g_sanity() and corresponding debugflag (0x8) Retire g_{stall,release}_events(). Under #ifdef DIAGNOSTIC: Make g_valid_obj() an official function and have it return an an non-zero integer which indicates the kind of object when found. Implement G_VALID_{CLASS,GEOM,CONSUMER,PROVIDER}() macros based on g_valid_obj(). Sprinkle calls to these macros liberally over the infrastructure. Always check that we do not free a live object.
2004-03-10 08:49:08 +00:00
g_topology_assert();
G_VALID_PROVIDER(pp);
First commit of the GEOM subsystem to make it easier for people to test and play with this. This is not yet production quality and should be run only on dedicated test boxes. For people who want to develop transformations for GEOM there exist a set of shims to run geom in userland (ask phk@freebsd.org). Reports of all kinds to: phk@freebsd.org Please include in report: dmesg sysctl debug.geomdot sysctl debug.geomconf Known significant limitations: no kernel dump facility. ioctls severely restricted. Sponsored by: DARPA, NAI Labs
2002-03-11 21:42:35 +00:00
return (0);
}
void
g_std_done(struct bio *bp)
{
struct bio *bp2;
Rename bio_linkage to the more obvious bio_parent. Add bio_t0 timestamp, and include <sys/time.h> where needed
2003-02-07 21:09:51 +00:00
bp2 = bp->bio_parent;
Add support g_clone_bio() and g_std_done() to spawn multiple children of a bio and correctly gather status when done. Sponsored by: DARPA & NAI Labs.
2002-10-09 07:11:59 +00:00
if (bp2->bio_error == 0)
bp2->bio_error = bp->bio_error;
bp2->bio_completed += bp->bio_completed;
First commit of the GEOM subsystem to make it easier for people to test and play with this. This is not yet production quality and should be run only on dedicated test boxes. For people who want to develop transformations for GEOM there exist a set of shims to run geom in userland (ask phk@freebsd.org). Reports of all kinds to: phk@freebsd.org Please include in report: dmesg sysctl debug.geomdot sysctl debug.geomconf Known significant limitations: no kernel dump facility. ioctls severely restricted. Sponsored by: DARPA, NAI Labs
2002-03-11 21:42:35 +00:00
g_destroy_bio(bp);
Commit the correct copy of the g_stat structure. Add debug.sizeof.g_stat sysctl. Set the id field of the g_stat when we create consumers and providers. Remove biocount from consumer, we will use the counters in the g_stat structure instead. Replace one field which will need to be atomically manipulated with two fields which will not (stat.nop and stat.nend). Change add companion field to bio_children: bio_inbed for the exact same reason. Don't output the biocount in the confdot output. Fix KASSERT in g_io_request(). Add sysctl kern.geom.collectstats defaulting to off. Collect the following raw statistics conditioned on this sysctl: for each consumer and provider { total number of operations started. total number of operations completed. time last operation completed. sum of idle-time. for each of BIO_READ, BIO_WRITE and BIO_DELETE { number of operations completed. number of bytes completed. number of ENOMEM errors. number of other errors. sum of transaction time. } } API for getting hold of these statistics data not included yet.
2003-02-07 23:08:24 +00:00
bp2->bio_inbed++;
if (bp2->bio_children == bp2->bio_inbed)
Add support g_clone_bio() and g_std_done() to spawn multiple children of a bio and correctly gather status when done. Sponsored by: DARPA & NAI Labs.
2002-10-09 07:11:59 +00:00
g_io_deliver(bp2, bp2->bio_error);
First commit of the GEOM subsystem to make it easier for people to test and play with this. This is not yet production quality and should be run only on dedicated test boxes. For people who want to develop transformations for GEOM there exist a set of shims to run geom in userland (ask phk@freebsd.org). Reports of all kinds to: phk@freebsd.org Please include in report: dmesg sysctl debug.geomdot sysctl debug.geomconf Known significant limitations: no kernel dump facility. ioctls severely restricted. Sponsored by: DARPA, NAI Labs
2002-03-11 21:42:35 +00:00
}
/* XXX: maybe this is only g_slice_spoiled */
void
g_std_spoiled(struct g_consumer *cp)
{
Back out all the stuff that didn't belong in the last commit.
2003-05-02 06:42:59 +00:00
struct g_geom *gp;
struct g_provider *pp;
First commit of the GEOM subsystem to make it easier for people to test and play with this. This is not yet production quality and should be run only on dedicated test boxes. For people who want to develop transformations for GEOM there exist a set of shims to run geom in userland (ask phk@freebsd.org). Reports of all kinds to: phk@freebsd.org Please include in report: dmesg sysctl debug.geomdot sysctl debug.geomconf Known significant limitations: no kernel dump facility. ioctls severely restricted. Sponsored by: DARPA, NAI Labs
2002-03-11 21:42:35 +00:00
g_topology_assert();
Rearrange some of the GEOM debugging tools to be more structured. Retire g_sanity() and corresponding debugflag (0x8) Retire g_{stall,release}_events(). Under #ifdef DIAGNOSTIC: Make g_valid_obj() an official function and have it return an an non-zero integer which indicates the kind of object when found. Implement G_VALID_{CLASS,GEOM,CONSUMER,PROVIDER}() macros based on g_valid_obj(). Sprinkle calls to these macros liberally over the infrastructure. Always check that we do not free a live object.
2004-03-10 08:49:08 +00:00
G_VALID_CONSUMER(cp);
g_trace(G_T_TOPOLOGY, "g_std_spoiled(%p)", cp);
Implement media change notification for DA and CD removable media devices. It includes three parts: 1) Modifications to CAM to detect media media changes and report them to disk(9) layer. For modern SATA (and potentially UAS) devices it utilizes Asynchronous Notification mechanism to receive events from hardware. Active polling with TEST UNIT READY commands with 3 seconds period is used for incapable hardware. After that both CD and DA drivers work the same way, detecting two conditions: "NOT READY: Medium not present" after medium was detected previously, and "UNIT ATTENTION: Not ready to ready change, medium may have changed". First one reported to disk(9) as media removal, second as media insert/change. To reliably receive second event new AC_UNIT_ATTENTION async added to make UAs broadcasted to all periphs by generic error handling code in cam_periph_error(). 2) Modifications to GEOM core to handle media remove and change events. Media removal handled by spoiling all consumers attached to the provider. Media change event also schedules provider retaste after spoiling to probe new media. New flag G_CF_ORPHAN was added to consumers to reflect that consumer is in process of destruction. It allows retaste to create new geom instance of the same class, while previous one is still dying. 3) Modifications to some GEOM classes: DEV -- to report media change events to devd; VFS -- to handle spoiling same as orphan to prevent accessing replaced media. PART class already handles spoiling alike to orphan. Reviewed by: silence on geom@ and scsi@ Tested by: avg Sponsored by: iXsystems, Inc. / PC-BSD MFC after: 2 months
2012-07-29 11:51:48 +00:00
cp->flags |= G_CF_ORPHAN;
Back out all the stuff that didn't belong in the last commit.
2003-05-02 06:42:59 +00:00
g_detach(cp);
gp = cp->geom;
LIST_FOREACH(pp, &gp->provider, provider)
g_orphan_provider(pp, ENXIO);
g_destroy_consumer(cp);
if (LIST_EMPTY(&gp->provider) && LIST_EMPTY(&gp->consumer))
g_destroy_geom(gp);
else
gp->flags |= G_GEOM_WITHER;
First commit of the GEOM subsystem to make it easier for people to test and play with this. This is not yet production quality and should be run only on dedicated test boxes. For people who want to develop transformations for GEOM there exist a set of shims to run geom in userland (ask phk@freebsd.org). Reports of all kinds to: phk@freebsd.org Please include in report: dmesg sysctl debug.geomdot sysctl debug.geomconf Known significant limitations: no kernel dump facility. ioctls severely restricted. Sponsored by: DARPA, NAI Labs
2002-03-11 21:42:35 +00:00
}
/*
* Spoiling happens when a provider is opened for writing, but consumers
* which are configured by in-band data are attached (slicers for instance).
* Since the write might potentially change the in-band data, such consumers
* need to re-evaluate their existence after the writing session closes.
* We do this by (offering to) tear them down when the open for write happens
* in return for a re-taste when it closes again.
* Together with the fact that such consumers grab an 'e' bit whenever they
* are open, regardless of mode, this ends up DTRT.
*/
Convert EV_SPOILED event to use g_call_me().
2003-04-23 20:06:38 +00:00
static void
g_spoil_event(void *arg, int flag)
{
struct g_provider *pp;
struct g_consumer *cp, *cp2;
g_topology_assert();
if (flag == EV_CANCEL)
return;
pp = arg;
Rearrange some of the GEOM debugging tools to be more structured. Retire g_sanity() and corresponding debugflag (0x8) Retire g_{stall,release}_events(). Under #ifdef DIAGNOSTIC: Make g_valid_obj() an official function and have it return an an non-zero integer which indicates the kind of object when found. Implement G_VALID_{CLASS,GEOM,CONSUMER,PROVIDER}() macros based on g_valid_obj(). Sprinkle calls to these macros liberally over the infrastructure. Always check that we do not free a live object.
2004-03-10 08:49:08 +00:00
G_VALID_PROVIDER(pp);
Add a topology trace to the g_spoil_event. MFC after: 1 week
2014-05-19 16:08:15 +00:00
g_trace(G_T_TOPOLOGY, "%s %p(%s:%s:%s)", __func__, pp,
pp->geom->class->name, pp->geom->name, pp->name);
Convert EV_SPOILED event to use g_call_me().
2003-04-23 20:06:38 +00:00
for (cp = LIST_FIRST(&pp->consumers); cp != NULL; cp = cp2) {
cp2 = LIST_NEXT(cp, consumers);
Implement media change notification for DA and CD removable media devices. It includes three parts: 1) Modifications to CAM to detect media media changes and report them to disk(9) layer. For modern SATA (and potentially UAS) devices it utilizes Asynchronous Notification mechanism to receive events from hardware. Active polling with TEST UNIT READY commands with 3 seconds period is used for incapable hardware. After that both CD and DA drivers work the same way, detecting two conditions: "NOT READY: Medium not present" after medium was detected previously, and "UNIT ATTENTION: Not ready to ready change, medium may have changed". First one reported to disk(9) as media removal, second as media insert/change. To reliably receive second event new AC_UNIT_ATTENTION async added to make UAs broadcasted to all periphs by generic error handling code in cam_periph_error(). 2) Modifications to GEOM core to handle media remove and change events. Media removal handled by spoiling all consumers attached to the provider. Media change event also schedules provider retaste after spoiling to probe new media. New flag G_CF_ORPHAN was added to consumers to reflect that consumer is in process of destruction. It allows retaste to create new geom instance of the same class, while previous one is still dying. 3) Modifications to some GEOM classes: DEV -- to report media change events to devd; VFS -- to handle spoiling same as orphan to prevent accessing replaced media. PART class already handles spoiling alike to orphan. Reviewed by: silence on geom@ and scsi@ Tested by: avg Sponsored by: iXsystems, Inc. / PC-BSD MFC after: 2 months
2012-07-29 11:51:48 +00:00
if ((cp->flags & G_CF_SPOILED) == 0)
Convert EV_SPOILED event to use g_call_me().
2003-04-23 20:06:38 +00:00
continue;
Implement media change notification for DA and CD removable media devices. It includes three parts: 1) Modifications to CAM to detect media media changes and report them to disk(9) layer. For modern SATA (and potentially UAS) devices it utilizes Asynchronous Notification mechanism to receive events from hardware. Active polling with TEST UNIT READY commands with 3 seconds period is used for incapable hardware. After that both CD and DA drivers work the same way, detecting two conditions: "NOT READY: Medium not present" after medium was detected previously, and "UNIT ATTENTION: Not ready to ready change, medium may have changed". First one reported to disk(9) as media removal, second as media insert/change. To reliably receive second event new AC_UNIT_ATTENTION async added to make UAs broadcasted to all periphs by generic error handling code in cam_periph_error(). 2) Modifications to GEOM core to handle media remove and change events. Media removal handled by spoiling all consumers attached to the provider. Media change event also schedules provider retaste after spoiling to probe new media. New flag G_CF_ORPHAN was added to consumers to reflect that consumer is in process of destruction. It allows retaste to create new geom instance of the same class, while previous one is still dying. 3) Modifications to some GEOM classes: DEV -- to report media change events to devd; VFS -- to handle spoiling same as orphan to prevent accessing replaced media. PART class already handles spoiling alike to orphan. Reviewed by: silence on geom@ and scsi@ Tested by: avg Sponsored by: iXsystems, Inc. / PC-BSD MFC after: 2 months
2012-07-29 11:51:48 +00:00
cp->flags &= ~G_CF_SPOILED;
Convert EV_SPOILED event to use g_call_me().
2003-04-23 20:06:38 +00:00
if (cp->geom->spoiled == NULL)
continue;
cp->geom->spoiled(cp);
g_topology_assert();
}
}
First commit of the GEOM subsystem to make it easier for people to test and play with this. This is not yet production quality and should be run only on dedicated test boxes. For people who want to develop transformations for GEOM there exist a set of shims to run geom in userland (ask phk@freebsd.org). Reports of all kinds to: phk@freebsd.org Please include in report: dmesg sysctl debug.geomdot sysctl debug.geomconf Known significant limitations: no kernel dump facility. ioctls severely restricted. Sponsored by: DARPA, NAI Labs
2002-03-11 21:42:35 +00:00
void
g_spoil(struct g_provider *pp, struct g_consumer *cp)
{
struct g_consumer *cp2;
g_topology_assert();
Rearrange some of the GEOM debugging tools to be more structured. Retire g_sanity() and corresponding debugflag (0x8) Retire g_{stall,release}_events(). Under #ifdef DIAGNOSTIC: Make g_valid_obj() an official function and have it return an an non-zero integer which indicates the kind of object when found. Implement G_VALID_{CLASS,GEOM,CONSUMER,PROVIDER}() macros based on g_valid_obj(). Sprinkle calls to these macros liberally over the infrastructure. Always check that we do not free a live object.
2004-03-10 08:49:08 +00:00
G_VALID_PROVIDER(pp);
G_VALID_CONSUMER(cp);
First commit of the GEOM subsystem to make it easier for people to test and play with this. This is not yet production quality and should be run only on dedicated test boxes. For people who want to develop transformations for GEOM there exist a set of shims to run geom in userland (ask phk@freebsd.org). Reports of all kinds to: phk@freebsd.org Please include in report: dmesg sysctl debug.geomdot sysctl debug.geomconf Known significant limitations: no kernel dump facility. ioctls severely restricted. Sponsored by: DARPA, NAI Labs
2002-03-11 21:42:35 +00:00
LIST_FOREACH(cp2, &pp->consumers, consumers) {
if (cp2 == cp)
continue;
/*
KASSERT(cp2->acr == 0, ("spoiling cp->acr = %d", cp2->acr));
KASSERT(cp2->acw == 0, ("spoiling cp->acw = %d", cp2->acw));
*/
KASSERT(cp2->ace == 0, ("spoiling cp->ace = %d", cp2->ace));
Implement media change notification for DA and CD removable media devices. It includes three parts: 1) Modifications to CAM to detect media media changes and report them to disk(9) layer. For modern SATA (and potentially UAS) devices it utilizes Asynchronous Notification mechanism to receive events from hardware. Active polling with TEST UNIT READY commands with 3 seconds period is used for incapable hardware. After that both CD and DA drivers work the same way, detecting two conditions: "NOT READY: Medium not present" after medium was detected previously, and "UNIT ATTENTION: Not ready to ready change, medium may have changed". First one reported to disk(9) as media removal, second as media insert/change. To reliably receive second event new AC_UNIT_ATTENTION async added to make UAs broadcasted to all periphs by generic error handling code in cam_periph_error(). 2) Modifications to GEOM core to handle media remove and change events. Media removal handled by spoiling all consumers attached to the provider. Media change event also schedules provider retaste after spoiling to probe new media. New flag G_CF_ORPHAN was added to consumers to reflect that consumer is in process of destruction. It allows retaste to create new geom instance of the same class, while previous one is still dying. 3) Modifications to some GEOM classes: DEV -- to report media change events to devd; VFS -- to handle spoiling same as orphan to prevent accessing replaced media. PART class already handles spoiling alike to orphan. Reviewed by: silence on geom@ and scsi@ Tested by: avg Sponsored by: iXsystems, Inc. / PC-BSD MFC after: 2 months
2012-07-29 11:51:48 +00:00
cp2->flags |= G_CF_SPOILED;
First commit of the GEOM subsystem to make it easier for people to test and play with this. This is not yet production quality and should be run only on dedicated test boxes. For people who want to develop transformations for GEOM there exist a set of shims to run geom in userland (ask phk@freebsd.org). Reports of all kinds to: phk@freebsd.org Please include in report: dmesg sysctl debug.geomdot sysctl debug.geomconf Known significant limitations: no kernel dump facility. ioctls severely restricted. Sponsored by: DARPA, NAI Labs
2002-03-11 21:42:35 +00:00
}
Rename g_call_me() to g_post_event(), and give it a flag argument to determine if we can M_WAITOK in malloc.
2003-04-23 20:46:12 +00:00
g_post_event(g_spoil_event, pp, M_WAITOK, pp, NULL);
First commit of the GEOM subsystem to make it easier for people to test and play with this. This is not yet production quality and should be run only on dedicated test boxes. For people who want to develop transformations for GEOM there exist a set of shims to run geom in userland (ask phk@freebsd.org). Reports of all kinds to: phk@freebsd.org Please include in report: dmesg sysctl debug.geomdot sysctl debug.geomconf Known significant limitations: no kernel dump facility. ioctls severely restricted. Sponsored by: DARPA, NAI Labs
2002-03-11 21:42:35 +00:00
}
Implement media change notification for DA and CD removable media devices. It includes three parts: 1) Modifications to CAM to detect media media changes and report them to disk(9) layer. For modern SATA (and potentially UAS) devices it utilizes Asynchronous Notification mechanism to receive events from hardware. Active polling with TEST UNIT READY commands with 3 seconds period is used for incapable hardware. After that both CD and DA drivers work the same way, detecting two conditions: "NOT READY: Medium not present" after medium was detected previously, and "UNIT ATTENTION: Not ready to ready change, medium may have changed". First one reported to disk(9) as media removal, second as media insert/change. To reliably receive second event new AC_UNIT_ATTENTION async added to make UAs broadcasted to all periphs by generic error handling code in cam_periph_error(). 2) Modifications to GEOM core to handle media remove and change events. Media removal handled by spoiling all consumers attached to the provider. Media change event also schedules provider retaste after spoiling to probe new media. New flag G_CF_ORPHAN was added to consumers to reflect that consumer is in process of destruction. It allows retaste to create new geom instance of the same class, while previous one is still dying. 3) Modifications to some GEOM classes: DEV -- to report media change events to devd; VFS -- to handle spoiling same as orphan to prevent accessing replaced media. PART class already handles spoiling alike to orphan. Reviewed by: silence on geom@ and scsi@ Tested by: avg Sponsored by: iXsystems, Inc. / PC-BSD MFC after: 2 months
2012-07-29 11:51:48 +00:00
static void
g_media_changed_event(void *arg, int flag)
{
struct g_provider *pp;
int retaste;
g_topology_assert();
if (flag == EV_CANCEL)
return;
pp = arg;
G_VALID_PROVIDER(pp);
/*
* If provider was not open for writing, queue retaste after spoiling.
* If it was, retaste will happen automatically on close.
*/
retaste = (pp->acw == 0 && pp->error == 0 &&
!(pp->geom->flags & G_GEOM_WITHER));
g_spoil_event(arg, flag);
if (retaste)
g_post_event(g_new_provider_event, pp, M_WAITOK, pp, NULL);
}
int
g_media_changed(struct g_provider *pp, int flag)
{
struct g_consumer *cp;
LIST_FOREACH(cp, &pp->consumers, consumers)
cp->flags |= G_CF_SPOILED;
return (g_post_event(g_media_changed_event, pp, flag, pp, NULL));
}
int
g_media_gone(struct g_provider *pp, int flag)
{
struct g_consumer *cp;
LIST_FOREACH(cp, &pp->consumers, consumers)
cp->flags |= G_CF_SPOILED;
return (g_post_event(g_spoil_event, pp, flag, pp, NULL));
}
Introduce the convenience function g_getattr() and make it DWIM. Sponsored by: DARPA & NAI Labs.
2002-04-09 15:13:42 +00:00
int
g_getattr__(const char *attr, struct g_consumer *cp, void *var, int len)
{
int error, i;
i = len;
error = g_io_getattr(attr, cp, &i, var);
if (error)
return (error);
if (i != len)
return (EINVAL);
return (0);
}
Introduce some serious paranoia to try to catch a memory overwrite problem as early as possible. Sponsored by: DARPA & NAI Labs
2002-04-23 11:48:45 +00:00
Implement relaxed comparision for hardcoded provider names to make it ignore adX/adaY difference in both directions to simplify migration to the CAM-based ATA or back.
2011-04-27 00:10:26 +00:00
static int
g_get_device_prefix_len(const char *name)
{
int len;
if (strncmp(name, "ada", 3) == 0)
len = 3;
else if (strncmp(name, "ad", 2) == 0)
len = 2;
else
return (0);
if (name[len] < '0' || name[len] > '9')
return (0);
do {
len++;
} while (name[len] >= '0' && name[len] <= '9');
return (len);
}
int
g_compare_names(const char *namea, const char *nameb)
{
int deva, devb;
if (strcmp(namea, nameb) == 0)
return (1);
deva = g_get_device_prefix_len(namea);
if (deva == 0)
return (0);
devb = g_get_device_prefix_len(nameb);
if (devb == 0)
return (0);
if (strcmp(namea + deva, nameb + devb) == 0)
return (1);
return (0);
}
Add aliasing concept to geom. Add an alias name list to geoms. Use them in geom_dev to create aliases. Previously, geom_dev would create an device node for the name of the geom. Now, additional nodes are created pointing back to the primary node with make_dev_alias_p. Aliases must be in place on the geom before any tasting occurs. Differential Revision: https://reviews.freebsd.org/D11873
2017-08-07 21:12:28 +00:00
void
g_geom_add_alias(struct g_geom *gp, const char *alias)
{
struct g_geom_alias *gap;
gap = (struct g_geom_alias *)g_malloc(
sizeof(struct g_geom_alias) + strlen(alias) + 1, M_WAITOK);
strcpy((char *)(gap + 1), alias);
gap->ga_alias = (const char *)(gap + 1);
LIST_INSERT_HEAD(&gp->aliases, gap, ga_next);
}
Add 'show geom [addr]' ddb(4) command, which prints entire GEOM topology if no additional argument is given or details about the given GEOM object (class, geom, provider or consumer). Approved by: phk
2006-09-15 16:36:45 +00:00
#if defined(DIAGNOSTIC) || defined(DDB)
This is a crude bandaid for 5.2 to protect against providers which disappear while being tasted. I can moderately easy trigger this with atapi-cd, but I do not fully understand the circumstances.
2003-11-15 18:44:43 +00:00
/*
Fix ddb(4) "show geom addr" command when INVARIANTS is enabled. Don't assert that the topology lock is held when g_valid_obj() is called from debugger. MFC after: 1 week
2010-04-19 20:07:35 +00:00
* This function walks the mesh and returns a non-zero integer if it
* finds the argument pointer is an object. The return value indicates
* which type of object it is believed to be. If topology is not locked,
* this function is potentially dangerous, but we don't assert that the
* topology lock is held when called from debugger.
This is a crude bandaid for 5.2 to protect against providers which disappear while being tasted. I can moderately easy trigger this with atapi-cd, but I do not fully understand the circumstances.
2003-11-15 18:44:43 +00:00
*/
Rearrange some of the GEOM debugging tools to be more structured. Retire g_sanity() and corresponding debugflag (0x8) Retire g_{stall,release}_events(). Under #ifdef DIAGNOSTIC: Make g_valid_obj() an official function and have it return an an non-zero integer which indicates the kind of object when found. Implement G_VALID_{CLASS,GEOM,CONSUMER,PROVIDER}() macros based on g_valid_obj(). Sprinkle calls to these macros liberally over the infrastructure. Always check that we do not free a live object.
2004-03-10 08:49:08 +00:00
int
This is a crude bandaid for 5.2 to protect against providers which disappear while being tasted. I can moderately easy trigger this with atapi-cd, but I do not fully understand the circumstances.
2003-11-15 18:44:43 +00:00
g_valid_obj(void const *ptr)
{
struct g_class *mp;
struct g_geom *gp;
struct g_consumer *cp;
struct g_provider *pp;
Fix ddb(4) "show geom addr" command when INVARIANTS is enabled. Don't assert that the topology lock is held when g_valid_obj() is called from debugger. MFC after: 1 week
2010-04-19 20:07:35 +00:00
#ifdef KDB
if (kdb_active == 0)
#endif
g_topology_assert();
Fix a panic which (reportedly) can happen when unmounting a filesystem with I/O requests in flight on kernels compiled with "options INVARIANTS". Also, make it obvious it's not right to call g_valid_obj() (and macros using it, e.g. G_VALID_CONSUMER()) without topology lock held. Approved by: re (kib) Reported by: pho
2009-07-01 20:16:29 +00:00
This is a crude bandaid for 5.2 to protect against providers which disappear while being tasted. I can moderately easy trigger this with atapi-cd, but I do not fully understand the circumstances.
2003-11-15 18:44:43 +00:00
LIST_FOREACH(mp, &g_classes, class) {
if (ptr == mp)
return (1);
LIST_FOREACH(gp, &mp->geom, geom) {
if (ptr == gp)
Rearrange some of the GEOM debugging tools to be more structured. Retire g_sanity() and corresponding debugflag (0x8) Retire g_{stall,release}_events(). Under #ifdef DIAGNOSTIC: Make g_valid_obj() an official function and have it return an an non-zero integer which indicates the kind of object when found. Implement G_VALID_{CLASS,GEOM,CONSUMER,PROVIDER}() macros based on g_valid_obj(). Sprinkle calls to these macros liberally over the infrastructure. Always check that we do not free a live object.
2004-03-10 08:49:08 +00:00
return (2);
This is a crude bandaid for 5.2 to protect against providers which disappear while being tasted. I can moderately easy trigger this with atapi-cd, but I do not fully understand the circumstances.
2003-11-15 18:44:43 +00:00
LIST_FOREACH(cp, &gp->consumer, consumer)
if (ptr == cp)
Rearrange some of the GEOM debugging tools to be more structured. Retire g_sanity() and corresponding debugflag (0x8) Retire g_{stall,release}_events(). Under #ifdef DIAGNOSTIC: Make g_valid_obj() an official function and have it return an an non-zero integer which indicates the kind of object when found. Implement G_VALID_{CLASS,GEOM,CONSUMER,PROVIDER}() macros based on g_valid_obj(). Sprinkle calls to these macros liberally over the infrastructure. Always check that we do not free a live object.
2004-03-10 08:49:08 +00:00
return (3);
This is a crude bandaid for 5.2 to protect against providers which disappear while being tasted. I can moderately easy trigger this with atapi-cd, but I do not fully understand the circumstances.
2003-11-15 18:44:43 +00:00
LIST_FOREACH(pp, &gp->provider, provider)
if (ptr == pp)
Rearrange some of the GEOM debugging tools to be more structured. Retire g_sanity() and corresponding debugflag (0x8) Retire g_{stall,release}_events(). Under #ifdef DIAGNOSTIC: Make g_valid_obj() an official function and have it return an an non-zero integer which indicates the kind of object when found. Implement G_VALID_{CLASS,GEOM,CONSUMER,PROVIDER}() macros based on g_valid_obj(). Sprinkle calls to these macros liberally over the infrastructure. Always check that we do not free a live object.
2004-03-10 08:49:08 +00:00
return (4);
This is a crude bandaid for 5.2 to protect against providers which disappear while being tasted. I can moderately easy trigger this with atapi-cd, but I do not fully understand the circumstances.
2003-11-15 18:44:43 +00:00
}
}
return(0);
}
Rearrange some of the GEOM debugging tools to be more structured. Retire g_sanity() and corresponding debugflag (0x8) Retire g_{stall,release}_events(). Under #ifdef DIAGNOSTIC: Make g_valid_obj() an official function and have it return an an non-zero integer which indicates the kind of object when found. Implement G_VALID_{CLASS,GEOM,CONSUMER,PROVIDER}() macros based on g_valid_obj(). Sprinkle calls to these macros liberally over the infrastructure. Always check that we do not free a live object.
2004-03-10 08:49:08 +00:00
#endif
Add 'show geom [addr]' ddb(4) command, which prints entire GEOM topology if no additional argument is given or details about the given GEOM object (class, geom, provider or consumer). Approved by: phk
2006-09-15 16:36:45 +00:00
#ifdef DDB
#define gprintf(...) do { \
- Fix wrong print on BIO_DONE. - Use db_printf instead of printf. While here, apply this to other ddb commands as well. Pointed out by: pjd
2009-05-26 10:03:44 +00:00
db_printf("%*s", indent, ""); \
db_printf(__VA_ARGS__); \
Add 'show geom [addr]' ddb(4) command, which prints entire GEOM topology if no additional argument is given or details about the given GEOM object (class, geom, provider or consumer). Approved by: phk
2006-09-15 16:36:45 +00:00
} while (0)
#define gprintln(...) do { \
gprintf(__VA_ARGS__); \
- Fix wrong print on BIO_DONE. - Use db_printf instead of printf. While here, apply this to other ddb commands as well. Pointed out by: pjd
2009-05-26 10:03:44 +00:00
db_printf("\n"); \
Add 'show geom [addr]' ddb(4) command, which prints entire GEOM topology if no additional argument is given or details about the given GEOM object (class, geom, provider or consumer). Approved by: phk
2006-09-15 16:36:45 +00:00
} while (0)
#define ADDFLAG(obj, flag, sflag) do { \
if ((obj)->flags & (flag)) { \
if (comma) \
strlcat(str, ",", size); \
strlcat(str, (sflag), size); \
comma = 1; \
} \
} while (0)
static char *
provider_flags_to_string(struct g_provider *pp, char *str, size_t size)
{
int comma = 0;
bzero(str, size);
if (pp->flags == 0) {
strlcpy(str, "NONE", size);
return (str);
}
ADDFLAG(pp, G_PF_WITHER, "G_PF_WITHER");
ADDFLAG(pp, G_PF_ORPHAN, "G_PF_ORPHAN");
return (str);
}
static char *
geom_flags_to_string(struct g_geom *gp, char *str, size_t size)
{
int comma = 0;
bzero(str, size);
if (gp->flags == 0) {
strlcpy(str, "NONE", size);
return (str);
}
ADDFLAG(gp, G_GEOM_WITHER, "G_GEOM_WITHER");
return (str);
}
static void
db_show_geom_consumer(int indent, struct g_consumer *cp)
{
if (indent == 0) {
gprintln("consumer: %p", cp);
gprintln(" class: %s (%p)", cp->geom->class->name,
cp->geom->class);
gprintln(" geom: %s (%p)", cp->geom->name, cp->geom);
if (cp->provider == NULL)
gprintln(" provider: none");
else {
gprintln(" provider: %s (%p)", cp->provider->name,
cp->provider);
}
gprintln(" access: r%dw%de%d", cp->acr, cp->acw, cp->ace);
Implement media change notification for DA and CD removable media devices. It includes three parts: 1) Modifications to CAM to detect media media changes and report them to disk(9) layer. For modern SATA (and potentially UAS) devices it utilizes Asynchronous Notification mechanism to receive events from hardware. Active polling with TEST UNIT READY commands with 3 seconds period is used for incapable hardware. After that both CD and DA drivers work the same way, detecting two conditions: "NOT READY: Medium not present" after medium was detected previously, and "UNIT ATTENTION: Not ready to ready change, medium may have changed". First one reported to disk(9) as media removal, second as media insert/change. To reliably receive second event new AC_UNIT_ATTENTION async added to make UAs broadcasted to all periphs by generic error handling code in cam_periph_error(). 2) Modifications to GEOM core to handle media remove and change events. Media removal handled by spoiling all consumers attached to the provider. Media change event also schedules provider retaste after spoiling to probe new media. New flag G_CF_ORPHAN was added to consumers to reflect that consumer is in process of destruction. It allows retaste to create new geom instance of the same class, while previous one is still dying. 3) Modifications to some GEOM classes: DEV -- to report media change events to devd; VFS -- to handle spoiling same as orphan to prevent accessing replaced media. PART class already handles spoiling alike to orphan. Reviewed by: silence on geom@ and scsi@ Tested by: avg Sponsored by: iXsystems, Inc. / PC-BSD MFC after: 2 months
2012-07-29 11:51:48 +00:00
gprintln(" flags: 0x%04x", cp->flags);
Add 'show geom [addr]' ddb(4) command, which prints entire GEOM topology if no additional argument is given or details about the given GEOM object (class, geom, provider or consumer). Approved by: phk
2006-09-15 16:36:45 +00:00
gprintln(" nstart: %u", cp->nstart);
gprintln(" nend: %u", cp->nend);
} else {
gprintf("consumer: %p (%s), access=r%dw%de%d", cp,
cp->provider != NULL ? cp->provider->name : "none",
cp->acr, cp->acw, cp->ace);
Implement media change notification for DA and CD removable media devices. It includes three parts: 1) Modifications to CAM to detect media media changes and report them to disk(9) layer. For modern SATA (and potentially UAS) devices it utilizes Asynchronous Notification mechanism to receive events from hardware. Active polling with TEST UNIT READY commands with 3 seconds period is used for incapable hardware. After that both CD and DA drivers work the same way, detecting two conditions: "NOT READY: Medium not present" after medium was detected previously, and "UNIT ATTENTION: Not ready to ready change, medium may have changed". First one reported to disk(9) as media removal, second as media insert/change. To reliably receive second event new AC_UNIT_ATTENTION async added to make UAs broadcasted to all periphs by generic error handling code in cam_periph_error(). 2) Modifications to GEOM core to handle media remove and change events. Media removal handled by spoiling all consumers attached to the provider. Media change event also schedules provider retaste after spoiling to probe new media. New flag G_CF_ORPHAN was added to consumers to reflect that consumer is in process of destruction. It allows retaste to create new geom instance of the same class, while previous one is still dying. 3) Modifications to some GEOM classes: DEV -- to report media change events to devd; VFS -- to handle spoiling same as orphan to prevent accessing replaced media. PART class already handles spoiling alike to orphan. Reviewed by: silence on geom@ and scsi@ Tested by: avg Sponsored by: iXsystems, Inc. / PC-BSD MFC after: 2 months
2012-07-29 11:51:48 +00:00
if (cp->flags)
db_printf(", flags=0x%04x", cp->flags);
- Fix wrong print on BIO_DONE. - Use db_printf instead of printf. While here, apply this to other ddb commands as well. Pointed out by: pjd
2009-05-26 10:03:44 +00:00
db_printf("\n");
Add 'show geom [addr]' ddb(4) command, which prints entire GEOM topology if no additional argument is given or details about the given GEOM object (class, geom, provider or consumer). Approved by: phk
2006-09-15 16:36:45 +00:00
}
}
static void
db_show_geom_provider(int indent, struct g_provider *pp)
{
struct g_consumer *cp;
char flags[64];
if (indent == 0) {
gprintln("provider: %s (%p)", pp->name, pp);
gprintln(" class: %s (%p)", pp->geom->class->name,
pp->geom->class);
gprintln(" geom: %s (%p)", pp->geom->name, pp->geom);
gprintln(" mediasize: %jd", (intmax_t)pp->mediasize);
gprintln(" sectorsize: %u", pp->sectorsize);
gprintln(" stripesize: %u", pp->stripesize);
gprintln(" stripeoffset: %u", pp->stripeoffset);
gprintln(" access: r%dw%de%d", pp->acr, pp->acw,
pp->ace);
gprintln(" flags: %s (0x%04x)",
provider_flags_to_string(pp, flags, sizeof(flags)),
pp->flags);
gprintln(" error: %d", pp->error);
gprintln(" nstart: %u", pp->nstart);
gprintln(" nend: %u", pp->nend);
if (LIST_EMPTY(&pp->consumers))
gprintln(" consumers: none");
} else {
gprintf("provider: %s (%p), access=r%dw%de%d",
pp->name, pp, pp->acr, pp->acw, pp->ace);
if (pp->flags != 0) {
- Fix wrong print on BIO_DONE. - Use db_printf instead of printf. While here, apply this to other ddb commands as well. Pointed out by: pjd
2009-05-26 10:03:44 +00:00
db_printf(", flags=%s (0x%04x)",
Add 'show geom [addr]' ddb(4) command, which prints entire GEOM topology if no additional argument is given or details about the given GEOM object (class, geom, provider or consumer). Approved by: phk
2006-09-15 16:36:45 +00:00
provider_flags_to_string(pp, flags, sizeof(flags)),
pp->flags);
}
- Fix wrong print on BIO_DONE. - Use db_printf instead of printf. While here, apply this to other ddb commands as well. Pointed out by: pjd
2009-05-26 10:03:44 +00:00
db_printf("\n");
Add 'show geom [addr]' ddb(4) command, which prints entire GEOM topology if no additional argument is given or details about the given GEOM object (class, geom, provider or consumer). Approved by: phk
2006-09-15 16:36:45 +00:00
}
if (!LIST_EMPTY(&pp->consumers)) {
Play nice with DDB pager. Educated by: jhb's BSDCan presentation
2008-05-18 21:13:10 +00:00
LIST_FOREACH(cp, &pp->consumers, consumers) {
Add 'show geom [addr]' ddb(4) command, which prints entire GEOM topology if no additional argument is given or details about the given GEOM object (class, geom, provider or consumer). Approved by: phk
2006-09-15 16:36:45 +00:00
db_show_geom_consumer(indent + 2, cp);
Play nice with DDB pager. Educated by: jhb's BSDCan presentation
2008-05-18 21:13:10 +00:00
if (db_pager_quit)
break;
}
Add 'show geom [addr]' ddb(4) command, which prints entire GEOM topology if no additional argument is given or details about the given GEOM object (class, geom, provider or consumer). Approved by: phk
2006-09-15 16:36:45 +00:00
}
}
static void
db_show_geom_geom(int indent, struct g_geom *gp)
{
struct g_provider *pp;
struct g_consumer *cp;
char flags[64];
if (indent == 0) {
gprintln("geom: %s (%p)", gp->name, gp);
gprintln(" class: %s (%p)", gp->class->name, gp->class);
gprintln(" flags: %s (0x%04x)",
geom_flags_to_string(gp, flags, sizeof(flags)), gp->flags);
gprintln(" rank: %d", gp->rank);
if (LIST_EMPTY(&gp->provider))
gprintln(" providers: none");
if (LIST_EMPTY(&gp->consumer))
gprintln(" consumers: none");
} else {
gprintf("geom: %s (%p), rank=%d", gp->name, gp, gp->rank);
if (gp->flags != 0) {
- Fix wrong print on BIO_DONE. - Use db_printf instead of printf. While here, apply this to other ddb commands as well. Pointed out by: pjd
2009-05-26 10:03:44 +00:00
db_printf(", flags=%s (0x%04x)",
Add 'show geom [addr]' ddb(4) command, which prints entire GEOM topology if no additional argument is given or details about the given GEOM object (class, geom, provider or consumer). Approved by: phk
2006-09-15 16:36:45 +00:00
geom_flags_to_string(gp, flags, sizeof(flags)),
gp->flags);
}
- Fix wrong print on BIO_DONE. - Use db_printf instead of printf. While here, apply this to other ddb commands as well. Pointed out by: pjd
2009-05-26 10:03:44 +00:00
db_printf("\n");
Add 'show geom [addr]' ddb(4) command, which prints entire GEOM topology if no additional argument is given or details about the given GEOM object (class, geom, provider or consumer). Approved by: phk
2006-09-15 16:36:45 +00:00
}
if (!LIST_EMPTY(&gp->provider)) {
Play nice with DDB pager. Educated by: jhb's BSDCan presentation
2008-05-18 21:13:10 +00:00
LIST_FOREACH(pp, &gp->provider, provider) {
Add 'show geom [addr]' ddb(4) command, which prints entire GEOM topology if no additional argument is given or details about the given GEOM object (class, geom, provider or consumer). Approved by: phk
2006-09-15 16:36:45 +00:00
db_show_geom_provider(indent + 2, pp);
Play nice with DDB pager. Educated by: jhb's BSDCan presentation
2008-05-18 21:13:10 +00:00
if (db_pager_quit)
break;
}
Add 'show geom [addr]' ddb(4) command, which prints entire GEOM topology if no additional argument is given or details about the given GEOM object (class, geom, provider or consumer). Approved by: phk
2006-09-15 16:36:45 +00:00
}
if (!LIST_EMPTY(&gp->consumer)) {
Play nice with DDB pager. Educated by: jhb's BSDCan presentation
2008-05-18 21:13:10 +00:00
LIST_FOREACH(cp, &gp->consumer, consumer) {
Add 'show geom [addr]' ddb(4) command, which prints entire GEOM topology if no additional argument is given or details about the given GEOM object (class, geom, provider or consumer). Approved by: phk
2006-09-15 16:36:45 +00:00
db_show_geom_consumer(indent + 2, cp);
Play nice with DDB pager. Educated by: jhb's BSDCan presentation
2008-05-18 21:13:10 +00:00
if (db_pager_quit)
break;
}
Add 'show geom [addr]' ddb(4) command, which prints entire GEOM topology if no additional argument is given or details about the given GEOM object (class, geom, provider or consumer). Approved by: phk
2006-09-15 16:36:45 +00:00
}
}
static void
db_show_geom_class(struct g_class *mp)
{
struct g_geom *gp;
- Fix wrong print on BIO_DONE. - Use db_printf instead of printf. While here, apply this to other ddb commands as well. Pointed out by: pjd
2009-05-26 10:03:44 +00:00
db_printf("class: %s (%p)\n", mp->name, mp);
Play nice with DDB pager. Educated by: jhb's BSDCan presentation
2008-05-18 21:13:10 +00:00
LIST_FOREACH(gp, &mp->geom, geom) {
Add 'show geom [addr]' ddb(4) command, which prints entire GEOM topology if no additional argument is given or details about the given GEOM object (class, geom, provider or consumer). Approved by: phk
2006-09-15 16:36:45 +00:00
db_show_geom_geom(2, gp);
Play nice with DDB pager. Educated by: jhb's BSDCan presentation
2008-05-18 21:13:10 +00:00
if (db_pager_quit)
break;
}
Add 'show geom [addr]' ddb(4) command, which prints entire GEOM topology if no additional argument is given or details about the given GEOM object (class, geom, provider or consumer). Approved by: phk
2006-09-15 16:36:45 +00:00
}
/*
* Print the GEOM topology or the given object.
*/
DB_SHOW_COMMAND(geom, db_show_geom)
{
struct g_class *mp;
if (!have_addr) {
/* No address given, print the entire topology. */
LIST_FOREACH(mp, &g_classes, class) {
db_show_geom_class(mp);
- Fix wrong print on BIO_DONE. - Use db_printf instead of printf. While here, apply this to other ddb commands as well. Pointed out by: pjd
2009-05-26 10:03:44 +00:00
db_printf("\n");
Play nice with DDB pager. Educated by: jhb's BSDCan presentation
2008-05-18 21:13:10 +00:00
if (db_pager_quit)
break;
Add 'show geom [addr]' ddb(4) command, which prints entire GEOM topology if no additional argument is given or details about the given GEOM object (class, geom, provider or consumer). Approved by: phk
2006-09-15 16:36:45 +00:00
}
} else {
switch (g_valid_obj((void *)addr)) {
case 1:
db_show_geom_class((struct g_class *)addr);
break;
case 2:
db_show_geom_geom(0, (struct g_geom *)addr);
break;
case 3:
db_show_geom_consumer(0, (struct g_consumer *)addr);
break;
case 4:
db_show_geom_provider(0, (struct g_provider *)addr);
break;
default:
- Fix wrong print on BIO_DONE. - Use db_printf instead of printf. While here, apply this to other ddb commands as well. Pointed out by: pjd
2009-05-26 10:03:44 +00:00
db_printf("Not a GEOM object.\n");
Add 'show geom [addr]' ddb(4) command, which prints entire GEOM topology if no additional argument is given or details about the given GEOM object (class, geom, provider or consumer). Approved by: phk
2006-09-15 16:36:45 +00:00
break;
}
}
}
- Add 'show bio' DDB command. MFC after: 3 weeks
2009-05-26 07:29:17 +00:00
static void
db_print_bio_cmd(struct bio *bp)
{
- Fix wrong print on BIO_DONE. - Use db_printf instead of printf. While here, apply this to other ddb commands as well. Pointed out by: pjd
2009-05-26 10:03:44 +00:00
db_printf(" cmd: ");
- Add 'show bio' DDB command. MFC after: 3 weeks
2009-05-26 07:29:17 +00:00
switch (bp->bio_cmd) {
- Fix wrong print on BIO_DONE. - Use db_printf instead of printf. While here, apply this to other ddb commands as well. Pointed out by: pjd
2009-05-26 10:03:44 +00:00
case BIO_READ: db_printf("BIO_READ"); break;
case BIO_WRITE: db_printf("BIO_WRITE"); break;
case BIO_DELETE: db_printf("BIO_DELETE"); break;
case BIO_GETATTR: db_printf("BIO_GETATTR"); break;
case BIO_FLUSH: db_printf("BIO_FLUSH"); break;
case BIO_CMD0: db_printf("BIO_CMD0"); break;
case BIO_CMD1: db_printf("BIO_CMD1"); break;
case BIO_CMD2: db_printf("BIO_CMD2"); break;
Add support for managing Shingled Magnetic Recording (SMR) drives. This change includes support for SCSI SMR drives (which conform to the Zoned Block Commands or ZBC spec) and ATA SMR drives (which conform to the Zoned ATA Command Set or ZAC spec) behind SAS expanders. This includes full management support through the GEOM BIO interface, and through a new userland utility, zonectl(8), and through camcontrol(8). This is now ready for filesystems to use to detect and manage zoned drives. (There is no work in progress that I know of to use this for ZFS or UFS, if anyone is interested, let me know and I may have some suggestions.) Also, improve ATA command passthrough and dispatch support, both via ATA and ATA passthrough over SCSI. Also, add support to camcontrol(8) for the ATA Extended Power Conditions feature set. You can now manage ATA device power states, and set various idle time thresholds for a drive to enter lower power states. Note that this change cannot be MFCed in full, because it depends on changes to the struct bio API that break compatilibity. In order to avoid breaking the stable API, only changes that don't touch or depend on the struct bio changes can be merged. For example, the camcontrol(8) changes don't depend on the new bio API, but zonectl(8) and the probe changes to the da(4) and ada(4) drivers do depend on it. Also note that the SMR changes have not yet been tested with an actual SCSI ZBC device, or a SCSI to ATA translation layer (SAT) that supports ZBC to ZAC translation. I have not yet gotten a suitable drive or SAT layer, so any testing help would be appreciated. These changes have been tested with Seagate Host Aware SATA drives attached to both SAS and SATA controllers. Also, I do not have any SATA Host Managed devices, and I suspect that it may take additional (hopefully minor) changes to support them. Thanks to Seagate for supplying the test hardware and answering questions. sbin/camcontrol/Makefile: Add epc.c and zone.c. sbin/camcontrol/camcontrol.8: Document the zone and epc subcommands. sbin/camcontrol/camcontrol.c: Add the zone and epc subcommands. Add auxiliary register support to build_ata_cmd(). Make sure to set the CAM_ATAIO_NEEDRESULT, CAM_ATAIO_DMA, and CAM_ATAIO_FPDMA flags as appropriate for ATA commands. Add a new get_ata_status() function to parse ATA result from SCSI sense descriptors (for ATA passthrough over SCSI) and ATA I/O requests. sbin/camcontrol/camcontrol.h: Update the build_ata_cmd() prototype Add get_ata_status(), zone(), and epc(). sbin/camcontrol/epc.c: Support for ATA Extended Power Conditions features. This includes support for all features documented in the ACS-4 Revision 12 specification from t13.org (dated February 18, 2016). The EPC feature set allows putting a drive into a power power mode immediately, or setting timeouts so that the drive will automatically enter progressively lower power states after various idle times. sbin/camcontrol/fwdownload.c: Update the firmware download code for the new build_ata_cmd() arguments. sbin/camcontrol/zone.c: Implement support for Shingled Magnetic Recording (SMR) drives via SCSI Zoned Block Commands (ZBC) and ATA Zoned Device ATA Command Set (ZAC). These specs were developed in concert, and are functionally identical. The primary differences are due to SCSI and ATA differences. (SCSI is big endian, ATA is little endian, for example.) This includes support for all commands defined in the ZBC and ZAC specs. sys/cam/ata/ata_all.c: Decode a number of additional ATA command names in ata_op_string(). Add a new CCB building function, ata_read_log(). Add ata_zac_mgmt_in() and ata_zac_mgmt_out() CCB building functions. These support both DMA and NCQ encapsulation. sys/cam/ata/ata_all.h: Add prototypes for ata_read_log(), ata_zac_mgmt_out(), and ata_zac_mgmt_in(). sys/cam/ata/ata_da.c: Revamp the ada(4) driver to support zoned devices. Add four new probe states to gather information needed for zone support. Add a new adasetflags() function to avoid duplication of large blocks of flag setting between the async handler and register functions. Add new sysctl variables that describe zone support and paramters. Add support for the new BIO_ZONE bio, and all of its subcommands: DISK_ZONE_OPEN, DISK_ZONE_CLOSE, DISK_ZONE_FINISH, DISK_ZONE_RWP, DISK_ZONE_REPORT_ZONES, and DISK_ZONE_GET_PARAMS. sys/cam/scsi/scsi_all.c: Add command descriptions for the ZBC IN/OUT commands. Add descriptions for ZBC Host Managed devices. Add a new function, scsi_ata_pass() to do ATA passthrough over SCSI. This will eventually replace scsi_ata_pass_16() -- it can create the 12, 16, and 32-byte variants of the ATA PASS-THROUGH command, and supports setting all of the registers defined as of SAT-4, Revision 5 (March 11, 2016). Change scsi_ata_identify() to use scsi_ata_pass() instead of scsi_ata_pass_16(). Add a new scsi_ata_read_log() function to facilitate reading ATA logs via SCSI. sys/cam/scsi/scsi_all.h: Add the new ATA PASS-THROUGH(32) command CDB. Add extended and variable CDB opcodes. Add Zoned Block Device Characteristics VPD page. Add ATA Return SCSI sense descriptor. Add prototypes for scsi_ata_read_log() and scsi_ata_pass(). sys/cam/scsi/scsi_da.c: Revamp the da(4) driver to support zoned devices. Add five new probe states, four of which are needed for ATA devices. Add five new sysctl variables that describe zone support and parameters. The da(4) driver supports SCSI ZBC devices, as well as ATA ZAC devices when they are attached via a SCSI to ATA Translation (SAT) layer. Since ZBC -> ZAC translation is a new feature in the T10 SAT-4 spec, most SATA drives will be supported via ATA commands sent via the SCSI ATA PASS-THROUGH command. The da(4) driver will prefer the ZBC interface, if it is available, for performance reasons, but will use the ATA PASS-THROUGH interface to the ZAC command set if the SAT layer doesn't support translation yet. As I mentioned above, ZBC command support is untested. Add support for the new BIO_ZONE bio, and all of its subcommands: DISK_ZONE_OPEN, DISK_ZONE_CLOSE, DISK_ZONE_FINISH, DISK_ZONE_RWP, DISK_ZONE_REPORT_ZONES, and DISK_ZONE_GET_PARAMS. Add scsi_zbc_in() and scsi_zbc_out() CCB building functions. Add scsi_ata_zac_mgmt_out() and scsi_ata_zac_mgmt_in() CCB/CDB building functions. Note that these have return values, unlike almost all other CCB building functions in CAM. The reason is that they can fail, depending upon the particular combination of input parameters. The primary failure case is if the user wants NCQ, but fails to specify additional CDB storage. NCQ requires using the 32-byte version of the SCSI ATA PASS-THROUGH command, and the current CAM CDB size is 16 bytes. sys/cam/scsi/scsi_da.h: Add ZBC IN and ZBC OUT CDBs and opcodes. Add SCSI Report Zones data structures. Add scsi_zbc_in(), scsi_zbc_out(), scsi_ata_zac_mgmt_out(), and scsi_ata_zac_mgmt_in() prototypes. sys/dev/ahci/ahci.c: Fix SEND / RECEIVE FPDMA QUEUED in the ahci(4) driver. ahci_setup_fis() previously set the top bits of the sector count register in the FIS to 0 for FPDMA commands. This is okay for read and write, because the PRIO field is in the only thing in those bits, and we don't implement that further up the stack. But, for SEND and RECEIVE FPDMA QUEUED, the subcommand is in that byte, so it needs to be transmitted to the drive. In ahci_setup_fis(), always set the the top 8 bits of the sector count register. We need it in both the standard and NCQ / FPDMA cases. sys/geom/eli/g_eli.c: Pass BIO_ZONE commands through the GELI class. sys/geom/geom.h: Add g_io_zonecmd() prototype. sys/geom/geom_dev.c: Add new DIOCZONECMD ioctl, which allows sending zone commands to disks. sys/geom/geom_disk.c: Add support for BIO_ZONE commands. sys/geom/geom_disk.h: Add a new flag, DISKFLAG_CANZONE, that indicates that a given GEOM disk client can handle BIO_ZONE commands. sys/geom/geom_io.c: Add a new function, g_io_zonecmd(), that handles execution of BIO_ZONE commands. Add permissions check for BIO_ZONE commands. Add command decoding for BIO_ZONE commands. sys/geom/geom_subr.c: Add DDB command decoding for BIO_ZONE commands. sys/kern/subr_devstat.c: Record statistics for REPORT ZONES commands. Note that the number of bytes transferred for REPORT ZONES won't quite match what is received from the harware. This is because we're necessarily counting bytes coming from the da(4) / ada(4) drivers, which are using the disk_zone.h interface to communicate up the stack. The structure sizes it uses are slightly different than the SCSI and ATA structure sizes. sys/sys/ata.h: Add many bit and structure definitions for ZAC, NCQ, and EPC command support. sys/sys/bio.h: Convert the bio_cmd field to a straight enumeration. This will yield more space for additional commands in the future. After change r297955 and other related changes, this is now possible. Converting to an enumeration will also prevent use as a bitmask in the future. sys/sys/disk.h: Define the DIOCZONECMD ioctl. sys/sys/disk_zone.h: Add a new API for managing zoned disks. This is very close to the SCSI ZBC and ATA ZAC standards, but uses integers in native byte order instead of big endian (SCSI) or little endian (ATA) byte arrays. This is intended to offer to the complete feature set of the ZBC and ZAC disk management without requiring the application developer to include SCSI or ATA headers. We also use one set of headers for ioctl consumers and kernel bio-level consumers. sys/sys/param.h: Bump __FreeBSD_version for sys/bio.h command changes, and inclusion of SMR support. usr.sbin/Makefile: Add the zonectl utility. usr.sbin/diskinfo/diskinfo.c Add disk zoning capability to the 'diskinfo -v' output. usr.sbin/zonectl/Makefile: Add zonectl makefile. usr.sbin/zonectl/zonectl.8 zonectl(8) man page. usr.sbin/zonectl/zonectl.c The zonectl(8) utility. This allows managing SCSI or ATA zoned disks via the disk_zone.h API. You can report zones, reset write pointers, get parameters, etc. Sponsored by: Spectra Logic Differential Revision: https://reviews.freebsd.org/D6147 Reviewed by: wblock (documentation)
2016-05-19 14:08:36 +00:00
case BIO_ZONE: db_printf("BIO_ZONE"); break;
- Fix wrong print on BIO_DONE. - Use db_printf instead of printf. While here, apply this to other ddb commands as well. Pointed out by: pjd
2009-05-26 10:03:44 +00:00
default: db_printf("UNKNOWN"); break;
- Add 'show bio' DDB command. MFC after: 3 weeks
2009-05-26 07:29:17 +00:00
}
- Fix wrong print on BIO_DONE. - Use db_printf instead of printf. While here, apply this to other ddb commands as well. Pointed out by: pjd
2009-05-26 10:03:44 +00:00
db_printf("\n");
- Add 'show bio' DDB command. MFC after: 3 weeks
2009-05-26 07:29:17 +00:00
}
static void
db_print_bio_flags(struct bio *bp)
{
int comma;
comma = 0;
- Fix wrong print on BIO_DONE. - Use db_printf instead of printf. While here, apply this to other ddb commands as well. Pointed out by: pjd
2009-05-26 10:03:44 +00:00
db_printf(" flags: ");
- Add 'show bio' DDB command. MFC after: 3 weeks
2009-05-26 07:29:17 +00:00
if (bp->bio_flags & BIO_ERROR) {
- Fix wrong print on BIO_DONE. - Use db_printf instead of printf. While here, apply this to other ddb commands as well. Pointed out by: pjd
2009-05-26 10:03:44 +00:00
db_printf("BIO_ERROR");
- Add 'show bio' DDB command. MFC after: 3 weeks
2009-05-26 07:29:17 +00:00
comma = 1;
}
if (bp->bio_flags & BIO_DONE) {
- Fix wrong print on BIO_DONE. - Use db_printf instead of printf. While here, apply this to other ddb commands as well. Pointed out by: pjd
2009-05-26 10:03:44 +00:00
db_printf("%sBIO_DONE", (comma ? ", " : ""));
- Add 'show bio' DDB command. MFC after: 3 weeks
2009-05-26 07:29:17 +00:00
comma = 1;
}
if (bp->bio_flags & BIO_ONQUEUE)
- Fix wrong print on BIO_DONE. - Use db_printf instead of printf. While here, apply this to other ddb commands as well. Pointed out by: pjd
2009-05-26 10:03:44 +00:00
db_printf("%sBIO_ONQUEUE", (comma ? ", " : ""));
db_printf("\n");
- Add 'show bio' DDB command. MFC after: 3 weeks
2009-05-26 07:29:17 +00:00
}
/*
* Print useful information in a BIO
*/
DB_SHOW_COMMAND(bio, db_show_bio)
{
struct bio *bp;
if (have_addr) {
bp = (struct bio *)addr;
- Fix wrong print on BIO_DONE. - Use db_printf instead of printf. While here, apply this to other ddb commands as well. Pointed out by: pjd
2009-05-26 10:03:44 +00:00
db_printf("BIO %p\n", bp);
- Add 'show bio' DDB command. MFC after: 3 weeks
2009-05-26 07:29:17 +00:00
db_print_bio_cmd(bp);
db_print_bio_flags(bp);
Bump bio_cmd and bio_*flags from 8 bits to 16. Differential Revision: https://reviews.freebsd.org/D5784
2016-04-14 05:10:41 +00:00
db_printf(" cflags: 0x%hx\n", bp->bio_cflags);
db_printf(" pflags: 0x%hx\n", bp->bio_pflags);
- Unbreak 64 bit platforms by casting off_t to intmax.
2009-05-26 14:15:06 +00:00
db_printf(" offset: %jd\n", (intmax_t)bp->bio_offset);
db_printf(" length: %jd\n", (intmax_t)bp->bio_length);
- Fix wrong print on BIO_DONE. - Use db_printf instead of printf. While here, apply this to other ddb commands as well. Pointed out by: pjd
2009-05-26 10:03:44 +00:00
db_printf(" bcount: %ld\n", bp->bio_bcount);
db_printf(" resid: %ld\n", bp->bio_resid);
- Unbreak 64 bit platforms by casting off_t to intmax.
2009-05-26 14:15:06 +00:00
db_printf(" completed: %jd\n", (intmax_t)bp->bio_completed);
- Fix wrong print on BIO_DONE. - Use db_printf instead of printf. While here, apply this to other ddb commands as well. Pointed out by: pjd
2009-05-26 10:03:44 +00:00
db_printf(" children: %u\n", bp->bio_children);
db_printf(" inbed: %u\n", bp->bio_inbed);
db_printf(" error: %d\n", bp->bio_error);
db_printf(" parent: %p\n", bp->bio_parent);
db_printf(" driver1: %p\n", bp->bio_driver1);
db_printf(" driver2: %p\n", bp->bio_driver2);
db_printf(" caller1: %p\n", bp->bio_caller1);
db_printf(" caller2: %p\n", bp->bio_caller2);
db_printf(" bio_from: %p\n", bp->bio_from);
db_printf(" bio_to: %p\n", bp->bio_to);
Add BUF_TRACKING and FULL_BUF_TRACKING buffer debugging Upstream the BUF_TRACKING and FULL_BUF_TRACKING buffer debugging code. This can be handy in tracking down what code touched hung bios and bufs last. The full history is especially useful, but adds enough bloat that it shouldn't be enabled in release builds. Function names (or arbitrary string constants) are tracked in a fixed-size ring in bufs. Bios gain a pointer to the upper buf for tracking. SCSI CCBs gain a pointer to the upper bio for tracking. Reviewed by: markj Sponsored by: Dell EMC Isilon Differential Revision: https://reviews.freebsd.org/D8366
2016-10-31 23:09:52 +00:00
#if defined(BUF_TRACKING) || defined(FULL_BUF_TRACKING)
db_printf(" bio_track_bp: %p\n", bp->bio_track_bp);
#endif
- Add 'show bio' DDB command. MFC after: 3 weeks
2009-05-26 07:29:17 +00:00
}
}
Add 'show geom [addr]' ddb(4) command, which prints entire GEOM topology if no additional argument is given or details about the given GEOM object (class, geom, provider or consumer). Approved by: phk
2006-09-15 16:36:45 +00:00
#undef gprintf
#undef gprintln
#undef ADDFLAG
#endif /* DDB */
Reference in New Issue Copy Permalink