freebsd-skq/sys/geom/vinum/geom_vinum_rm.c
le 18ba8315a7 Re-vamp how I/O is handled in volumes and plexes.
Analogous to the drive level, give each volume and plex a worker thread
that picks up and processes incoming and completed BIOs.

This should fix the data corruption issues that have come up a few
weeks ago and improve performance, especially of RAID5 plexes.

The volume level needs a little work, though.
2004-09-18 13:44:43 +00:00

439 lines
10 KiB
C

/*
* Copyright (c) 2004 Lukas Ertl
* All rights reserved.
*
* 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.
*
* THIS SOFTWARE IS PROVIDED BY 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 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.
*
*/
#include <sys/cdefs.h>
__FBSDID("$FreeBSD$");
#include <sys/param.h>
#include <sys/libkern.h>
#include <sys/kernel.h>
#include <sys/malloc.h>
#include <geom/geom.h>
#include <geom/vinum/geom_vinum_var.h>
#include <geom/vinum/geom_vinum.h>
#include <geom/vinum/geom_vinum_share.h>
static void gv_cleanup_pp(void *, int);
static void gv_free_sd(struct gv_sd *);
static int gv_rm_drive(struct gv_softc *, struct gctl_req *,
struct gv_drive *, int);
static int gv_rm_plex(struct gv_softc *, struct gctl_req *,
struct gv_plex *, int);
static int gv_rm_sd(struct gv_softc *, struct gctl_req *, struct gv_sd *,
int);
static int gv_rm_vol(struct gv_softc *, struct gctl_req *,
struct gv_volume *, int);
/* General 'remove' routine. */
void
gv_remove(struct g_geom *gp, struct gctl_req *req)
{
struct gv_softc *sc;
struct gv_volume *v;
struct gv_plex *p;
struct gv_sd *s;
struct gv_drive *d;
int *argc, *flags;
char *argv, buf[20];
int i, type, err;
argc = gctl_get_paraml(req, "argc", sizeof(*argc));
flags = gctl_get_paraml(req, "flags", sizeof(*flags));
if (argc == NULL || *argc == 0) {
gctl_error(req, "no arguments given");
return;
}
sc = gp->softc;
for (i = 0; i < *argc; i++) {
snprintf(buf, sizeof(buf), "argv%d", i);
argv = gctl_get_param(req, buf, NULL);
if (argv == NULL)
continue;
type = gv_object_type(sc, argv);
switch (type) {
case GV_TYPE_VOL:
v = gv_find_vol(sc, argv);
if (v == NULL) {
gctl_error(req, "unknown volume '%s'", argv);
return;
}
err = gv_rm_vol(sc, req, v, *flags);
if (err)
return;
break;
case GV_TYPE_PLEX:
p = gv_find_plex(sc, argv);
if (p == NULL) {
gctl_error(req, "unknown plex '%s'", argv);
return;
}
err = gv_rm_plex(sc, req, p, *flags);
if (err)
return;
break;
case GV_TYPE_SD:
s = gv_find_sd(sc, argv);
if (s == NULL) {
gctl_error(req, "unknown subdisk '%s'", argv);
return;
}
err = gv_rm_sd(sc, req, s, *flags);
if (err)
return;
break;
case GV_TYPE_DRIVE:
d = gv_find_drive(sc, argv);
if (d == NULL) {
gctl_error(req, "unknown drive '%s'", argv);
return;
}
err = gv_rm_drive(sc, req, d, *flags);
if (err)
return;
break;
default:
gctl_error(req, "unknown object '%s'", argv);
return;
}
}
gv_save_config_all(sc);
}
/* Remove a volume. */
static int
gv_rm_vol(struct gv_softc *sc, struct gctl_req *req, struct gv_volume *v, int flags)
{
struct g_geom *gp;
struct gv_plex *p, *p2;
int err;
g_topology_assert();
KASSERT(v != NULL, ("gv_rm_vol: NULL v"));
/* If this volume has plexes, we want a recursive removal. */
if (!LIST_EMPTY(&v->plexes) && !(flags & GV_FLAG_R)) {
gctl_error(req, "volume '%s' has attached plexes", v->name);
return (-1);
}
gp = v->geom;
/* Check if any of our consumers is open. */
if (gp != NULL && gv_is_open(gp)) {
gctl_error(req, "volume '%s' is busy", v->name);
return (-1);
}
/* Remove the plexes our volume has. */
LIST_FOREACH_SAFE(p, &v->plexes, in_volume, p2) {
v->plexcount--;
LIST_REMOVE(p, in_volume);
p->vol_sc = NULL;
err = gv_rm_plex(sc, req, p, flags);
if (err)
return (err);
}
/* Clean up and let our geom fade away. */
LIST_REMOVE(v, volume);
gv_kill_vol_thread(v);
g_free(v);
if (gp != NULL) {
gp->softc = NULL;
g_wither_geom(gp, ENXIO);
}
return (0);
}
/* Remove a plex. */
static int
gv_rm_plex(struct gv_softc *sc, struct gctl_req *req, struct gv_plex *p, int flags)
{
struct g_geom *gp;
struct gv_sd *s, *s2;
int err;
g_topology_assert();
KASSERT(p != NULL, ("gv_rm_plex: NULL p"));
/* If this plex has subdisks, we want a recursive removal. */
if (!LIST_EMPTY(&p->subdisks) && !(flags & GV_FLAG_R)) {
gctl_error(req, "plex '%s' has attached subdisks", p->name);
return (-1);
}
if (p->vol_sc != NULL && p->vol_sc->plexcount == 1) {
gctl_error(req, "plex '%s' is still attached to volume '%s'",
p->name, p->volume);
return (-1);
}
gp = p->geom;
/* Check if any of our consumers is open. */
if (gp != NULL && gv_is_open(gp)) {
gctl_error(req, "plex '%s' is busy", p->name);
return (-1);
}
/* Remove the subdisks our plex has. */
LIST_FOREACH_SAFE(s, &p->subdisks, in_plex, s2) {
p->sdcount--;
#if 0
LIST_REMOVE(s, in_plex);
s->plex_sc = NULL;
#endif
err = gv_rm_sd(sc, req, s, flags);
if (err)
return (err);
}
/* Clean up and let our geom fade away. */
LIST_REMOVE(p, plex);
if (p->vol_sc != NULL) {
p->vol_sc->plexcount--;
LIST_REMOVE(p, in_volume);
p->vol_sc = NULL;
}
gv_kill_plex_thread(p);
g_free(p);
if (gp != NULL) {
gp->softc = NULL;
g_wither_geom(gp, ENXIO);
}
return (0);
}
/* Remove a subdisk. */
static int
gv_rm_sd(struct gv_softc *sc, struct gctl_req *req, struct gv_sd *s, int flags)
{
struct gv_drive *d;
struct g_geom *gp;
struct g_provider *pp;
KASSERT(s != NULL, ("gv_rm_sd: NULL s"));
d = s->drive_sc;
KASSERT(d != NULL, ("gv_rm_sd: NULL d"));
gp = d->geom;
KASSERT(gp != NULL, ("gv_rm_sd: NULL gp"));
pp = s->provider;
/* Clean up. */
LIST_REMOVE(s, in_plex);
LIST_REMOVE(s, from_drive);
LIST_REMOVE(s, sd);
gv_free_sd(s);
g_free(s);
/* If the subdisk has a provider we need to clean up this one too. */
if (pp != NULL) {
g_orphan_provider(pp, ENXIO);
if (LIST_EMPTY(&pp->consumers))
g_destroy_provider(pp);
else
/* Schedule this left-over provider for destruction. */
g_post_event(gv_cleanup_pp, pp, M_WAITOK, pp, NULL);
}
return (0);
}
/* Remove a drive. */
static int
gv_rm_drive(struct gv_softc *sc, struct gctl_req *req, struct gv_drive *d, int flags)
{
struct g_geom *gp;
struct g_consumer *cp;
struct gv_freelist *fl, *fl2;
struct gv_plex *p;
struct gv_sd *s, *s2;
struct gv_volume *v;
int err;
KASSERT(d != NULL, ("gv_rm_drive: NULL d"));
gp = d->geom;
KASSERT(gp != NULL, ("gv_rm_drive: NULL gp"));
/* We don't allow to remove open drives. */
if (gv_is_open(gp)) {
gctl_error(req, "drive '%s' is open", d->name);
return (-1);
}
/* A drive with subdisks needs a recursive removal. */
if (!LIST_EMPTY(&d->subdisks) && !(flags & GV_FLAG_R)) {
gctl_error(req, "drive '%s' still has subdisks", d->name);
return (-1);
}
cp = LIST_FIRST(&gp->consumer);
err = g_access(cp, 0, 1, 0);
if (err) {
printf("GEOM_VINUM: gv_rm_drive: couldn't access '%s', errno: "
"%d\n", cp->provider->name, err);
return (err);
}
/* Clear the Vinum Magic. */
d->hdr->magic = GV_NOMAGIC;
g_topology_unlock();
err = g_write_data(cp, GV_HDR_OFFSET, d->hdr, GV_HDR_LEN);
if (err) {
printf("GEOM_VINUM: gv_rm_drive: couldn't write header to '%s'"
", errno: %d\n", cp->provider->name, err);
d->hdr->magic = GV_MAGIC;
}
g_topology_lock();
g_access(cp, 0, -1, 0);
/* Remove all associated subdisks, plexes, volumes. */
if (!LIST_EMPTY(&d->subdisks)) {
LIST_FOREACH_SAFE(s, &d->subdisks, from_drive, s2) {
p = s->plex_sc;
if (p != NULL) {
v = p->vol_sc;
if (v != NULL)
gv_rm_vol(sc, req, v, flags);
}
}
}
/* Clean up. */
LIST_FOREACH_SAFE(fl, &d->freelist, freelist, fl2) {
LIST_REMOVE(fl, freelist);
g_free(fl);
}
LIST_REMOVE(d, drive);
gv_kill_drive_thread(d);
gp = d->geom;
d->geom = NULL;
g_free(d->hdr);
g_free(d);
gv_save_config_all(sc);
g_wither_geom(gp, ENXIO);
return (err);
}
/*
* This function is called from the event queue to clean up left-over subdisk
* providers.
*/
static void
gv_cleanup_pp(void *arg, int flag)
{
struct g_provider *pp;
g_topology_assert();
if (flag == EV_CANCEL)
return;
pp = arg;
if (pp == NULL) {
printf("gv_cleanup_pp: provider has gone\n");
return;
}
if (!LIST_EMPTY(&pp->consumers)) {
printf("gv_cleanup_pp: provider still not empty\n");
return;
}
g_destroy_provider(pp);
}
static void
gv_free_sd(struct gv_sd *s)
{
struct gv_drive *d;
struct gv_freelist *fl, *fl2;
KASSERT(s != NULL, ("gv_free_sd: NULL s"));
d = s->drive_sc;
KASSERT(d != NULL, ("gv_free_sd: NULL d"));
/*
* First, find the free slot that's immediately before or after this
* subdisk.
*/
fl = NULL;
LIST_FOREACH(fl, &d->freelist, freelist) {
if (fl->offset == s->drive_offset + s->size)
break;
if (fl->offset + fl->size == s->drive_offset)
break;
}
/* If there is no free slot behind this subdisk, so create one. */
if (fl == NULL) {
fl = g_malloc(sizeof(*fl), M_WAITOK | M_ZERO);
fl->size = s->size;
fl->offset = s->drive_offset;
if (d->freelist_entries == 0) {
LIST_INSERT_HEAD(&d->freelist, fl, freelist);
} else {
LIST_FOREACH(fl2, &d->freelist, freelist) {
if (fl->offset < fl2->offset) {
LIST_INSERT_BEFORE(fl2, fl, freelist);
break;
} else if (LIST_NEXT(fl2, freelist) == NULL) {
LIST_INSERT_AFTER(fl2, fl, freelist);
break;
}
}
}
d->freelist_entries++;
/* Expand the free slot we just found. */
} else {
fl->size += s->size;
if (fl->offset > s->drive_offset)
fl->offset = s->drive_offset;
}
d->avail += s->size;
}