freebsd-dev/sys/geom/geom_dev.c
Kenneth D. Merry 9a6844d55f 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

752 lines
19 KiB
C

/*-
* 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.
*/
#include <sys/cdefs.h>
__FBSDID("$FreeBSD$");
#include <sys/param.h>
#include <sys/systm.h>
#include <sys/malloc.h>
#include <sys/kernel.h>
#include <sys/conf.h>
#include <sys/ctype.h>
#include <sys/bio.h>
#include <sys/bus.h>
#include <sys/lock.h>
#include <sys/mutex.h>
#include <sys/proc.h>
#include <sys/errno.h>
#include <sys/time.h>
#include <sys/disk.h>
#include <sys/fcntl.h>
#include <sys/limits.h>
#include <sys/sysctl.h>
#include <geom/geom.h>
#include <geom/geom_int.h>
#include <machine/stdarg.h>
struct g_dev_softc {
struct mtx sc_mtx;
struct cdev *sc_dev;
struct cdev *sc_alias;
int sc_open;
int sc_active;
};
static d_open_t g_dev_open;
static d_close_t g_dev_close;
static d_strategy_t g_dev_strategy;
static d_ioctl_t g_dev_ioctl;
static struct cdevsw g_dev_cdevsw = {
.d_version = D_VERSION,
.d_open = g_dev_open,
.d_close = g_dev_close,
.d_read = physread,
.d_write = physwrite,
.d_ioctl = g_dev_ioctl,
.d_strategy = g_dev_strategy,
.d_name = "g_dev",
.d_flags = D_DISK | D_TRACKCLOSE,
};
static g_init_t g_dev_init;
static g_fini_t g_dev_fini;
static g_taste_t g_dev_taste;
static g_orphan_t g_dev_orphan;
static g_attrchanged_t g_dev_attrchanged;
static struct g_class g_dev_class = {
.name = "DEV",
.version = G_VERSION,
.init = g_dev_init,
.fini = g_dev_fini,
.taste = g_dev_taste,
.orphan = g_dev_orphan,
.attrchanged = g_dev_attrchanged
};
/*
* We target 262144 (8 x 32768) sectors by default as this significantly
* increases the throughput on commonly used SSD's with a marginal
* increase in non-interruptible request latency.
*/
static uint64_t g_dev_del_max_sectors = 262144;
SYSCTL_DECL(_kern_geom);
SYSCTL_NODE(_kern_geom, OID_AUTO, dev, CTLFLAG_RW, 0, "GEOM_DEV stuff");
SYSCTL_QUAD(_kern_geom_dev, OID_AUTO, delete_max_sectors, CTLFLAG_RW,
&g_dev_del_max_sectors, 0, "Maximum number of sectors in a single "
"delete request sent to the provider. Larger requests are chunked "
"so they can be interrupted. (0 = disable chunking)");
static char *dumpdev = NULL;
static void
g_dev_init(struct g_class *mp)
{
dumpdev = kern_getenv("dumpdev");
}
static void
g_dev_fini(struct g_class *mp)
{
freeenv(dumpdev);
dumpdev = NULL;
}
static int
g_dev_setdumpdev(struct cdev *dev, struct thread *td)
{
struct g_kerneldump kd;
struct g_consumer *cp;
int error, len;
if (dev == NULL)
return (set_dumper(NULL, NULL, td));
cp = dev->si_drv2;
len = sizeof(kd);
kd.offset = 0;
kd.length = OFF_MAX;
error = g_io_getattr("GEOM::kerneldump", cp, &len, &kd);
if (error == 0) {
error = set_dumper(&kd.di, devtoname(dev), td);
if (error == 0)
dev->si_flags |= SI_DUMPDEV;
}
return (error);
}
static int
init_dumpdev(struct cdev *dev)
{
struct g_consumer *cp;
const char *devprefix = "/dev/", *devname;
int error;
size_t len;
if (dumpdev == NULL)
return (0);
len = strlen(devprefix);
devname = devtoname(dev);
if (strcmp(devname, dumpdev) != 0 &&
(strncmp(dumpdev, devprefix, len) != 0 ||
strcmp(devname, dumpdev + len) != 0))
return (0);
cp = (struct g_consumer *)dev->si_drv2;
error = g_access(cp, 1, 0, 0);
if (error != 0)
return (error);
error = g_dev_setdumpdev(dev, curthread);
if (error == 0) {
freeenv(dumpdev);
dumpdev = NULL;
}
(void)g_access(cp, -1, 0, 0);
return (error);
}
static void
g_dev_destroy(void *arg, int flags __unused)
{
struct g_consumer *cp;
struct g_geom *gp;
struct g_dev_softc *sc;
char buf[SPECNAMELEN + 6];
g_topology_assert();
cp = arg;
gp = cp->geom;
sc = cp->private;
g_trace(G_T_TOPOLOGY, "g_dev_destroy(%p(%s))", cp, gp->name);
snprintf(buf, sizeof(buf), "cdev=%s", gp->name);
devctl_notify_f("GEOM", "DEV", "DESTROY", buf, M_WAITOK);
if (cp->acr > 0 || cp->acw > 0 || cp->ace > 0)
g_access(cp, -cp->acr, -cp->acw, -cp->ace);
g_detach(cp);
g_destroy_consumer(cp);
g_destroy_geom(gp);
mtx_destroy(&sc->sc_mtx);
g_free(sc);
}
void
g_dev_print(void)
{
struct g_geom *gp;
char const *p = "";
LIST_FOREACH(gp, &g_dev_class.geom, geom) {
printf("%s%s", p, gp->name);
p = " ";
}
printf("\n");
}
static void
g_dev_attrchanged(struct g_consumer *cp, const char *attr)
{
struct g_dev_softc *sc;
struct cdev *dev;
char buf[SPECNAMELEN + 6];
sc = cp->private;
if (strcmp(attr, "GEOM::media") == 0) {
dev = sc->sc_dev;
snprintf(buf, sizeof(buf), "cdev=%s", dev->si_name);
devctl_notify_f("DEVFS", "CDEV", "MEDIACHANGE", buf, M_WAITOK);
devctl_notify_f("GEOM", "DEV", "MEDIACHANGE", buf, M_WAITOK);
dev = sc->sc_alias;
if (dev != NULL) {
snprintf(buf, sizeof(buf), "cdev=%s", dev->si_name);
devctl_notify_f("DEVFS", "CDEV", "MEDIACHANGE", buf,
M_WAITOK);
devctl_notify_f("GEOM", "DEV", "MEDIACHANGE", buf,
M_WAITOK);
}
return;
}
if (strcmp(attr, "GEOM::physpath") != 0)
return;
if (g_access(cp, 1, 0, 0) == 0) {
char *physpath;
int error, physpath_len;
physpath_len = MAXPATHLEN;
physpath = g_malloc(physpath_len, M_WAITOK|M_ZERO);
error =
g_io_getattr("GEOM::physpath", cp, &physpath_len, physpath);
g_access(cp, -1, 0, 0);
if (error == 0 && strlen(physpath) != 0) {
struct cdev *old_alias_dev;
struct cdev **alias_devp;
dev = sc->sc_dev;
old_alias_dev = sc->sc_alias;
alias_devp = (struct cdev **)&sc->sc_alias;
make_dev_physpath_alias(MAKEDEV_WAITOK, alias_devp,
dev, old_alias_dev, physpath);
} else if (sc->sc_alias) {
destroy_dev((struct cdev *)sc->sc_alias);
sc->sc_alias = NULL;
}
g_free(physpath);
}
}
struct g_provider *
g_dev_getprovider(struct cdev *dev)
{
struct g_consumer *cp;
g_topology_assert();
if (dev == NULL)
return (NULL);
if (dev->si_devsw != &g_dev_cdevsw)
return (NULL);
cp = dev->si_drv2;
return (cp->provider);
}
static struct g_geom *
g_dev_taste(struct g_class *mp, struct g_provider *pp, int insist __unused)
{
struct g_geom *gp;
struct g_consumer *cp;
struct g_dev_softc *sc;
int error;
struct cdev *dev;
char buf[SPECNAMELEN + 6];
g_trace(G_T_TOPOLOGY, "dev_taste(%s,%s)", mp->name, pp->name);
g_topology_assert();
gp = g_new_geomf(mp, "%s", pp->name);
sc = g_malloc(sizeof(*sc), M_WAITOK | M_ZERO);
mtx_init(&sc->sc_mtx, "g_dev", NULL, MTX_DEF);
cp = g_new_consumer(gp);
cp->private = sc;
cp->flags |= G_CF_DIRECT_SEND | G_CF_DIRECT_RECEIVE;
error = g_attach(cp, pp);
KASSERT(error == 0,
("g_dev_taste(%s) failed to g_attach, err=%d", pp->name, error));
error = make_dev_p(MAKEDEV_CHECKNAME | MAKEDEV_WAITOK, &dev,
&g_dev_cdevsw, NULL, UID_ROOT, GID_OPERATOR, 0640, "%s", gp->name);
if (error != 0) {
printf("%s: make_dev_p() failed (gp->name=%s, error=%d)\n",
__func__, gp->name, error);
g_detach(cp);
g_destroy_consumer(cp);
g_destroy_geom(gp);
mtx_destroy(&sc->sc_mtx);
g_free(sc);
return (NULL);
}
dev->si_flags |= SI_UNMAPPED;
sc->sc_dev = dev;
dev->si_iosize_max = MAXPHYS;
dev->si_drv2 = cp;
error = init_dumpdev(dev);
if (error != 0)
printf("%s: init_dumpdev() failed (gp->name=%s, error=%d)\n",
__func__, gp->name, error);
g_dev_attrchanged(cp, "GEOM::physpath");
snprintf(buf, sizeof(buf), "cdev=%s", gp->name);
devctl_notify_f("GEOM", "DEV", "CREATE", buf, M_WAITOK);
return (gp);
}
static int
g_dev_open(struct cdev *dev, int flags, int fmt, struct thread *td)
{
struct g_consumer *cp;
struct g_dev_softc *sc;
int error, r, w, e;
cp = dev->si_drv2;
if (cp == NULL)
return (ENXIO); /* g_dev_taste() not done yet */
g_trace(G_T_ACCESS, "g_dev_open(%s, %d, %d, %p)",
cp->geom->name, flags, fmt, td);
r = flags & FREAD ? 1 : 0;
w = flags & FWRITE ? 1 : 0;
#ifdef notyet
e = flags & O_EXCL ? 1 : 0;
#else
e = 0;
#endif
/*
* This happens on attempt to open a device node with O_EXEC.
*/
if (r + w + e == 0)
return (EINVAL);
if (w) {
/*
* When running in very secure mode, do not allow
* opens for writing of any disks.
*/
error = securelevel_ge(td->td_ucred, 2);
if (error)
return (error);
}
g_topology_lock();
error = g_access(cp, r, w, e);
g_topology_unlock();
if (error == 0) {
sc = cp->private;
mtx_lock(&sc->sc_mtx);
if (sc->sc_open == 0 && sc->sc_active != 0)
wakeup(&sc->sc_active);
sc->sc_open += r + w + e;
mtx_unlock(&sc->sc_mtx);
}
return (error);
}
static int
g_dev_close(struct cdev *dev, int flags, int fmt, struct thread *td)
{
struct g_consumer *cp;
struct g_dev_softc *sc;
int error, r, w, e;
cp = dev->si_drv2;
if (cp == NULL)
return (ENXIO);
g_trace(G_T_ACCESS, "g_dev_close(%s, %d, %d, %p)",
cp->geom->name, flags, fmt, td);
r = flags & FREAD ? -1 : 0;
w = flags & FWRITE ? -1 : 0;
#ifdef notyet
e = flags & O_EXCL ? -1 : 0;
#else
e = 0;
#endif
/*
* The vgonel(9) - caused by eg. forced unmount of devfs - calls
* VOP_CLOSE(9) on devfs vnode without any FREAD or FWRITE flags,
* which would result in zero deltas, which in turn would cause
* panic in g_access(9).
*
* Note that we cannot zero the counters (ie. do "r = cp->acr"
* etc) instead, because the consumer might be opened in another
* devfs instance.
*/
if (r + w + e == 0)
return (EINVAL);
sc = cp->private;
mtx_lock(&sc->sc_mtx);
sc->sc_open += r + w + e;
while (sc->sc_open == 0 && sc->sc_active != 0)
msleep(&sc->sc_active, &sc->sc_mtx, 0, "PRIBIO", 0);
mtx_unlock(&sc->sc_mtx);
g_topology_lock();
error = g_access(cp, r, w, e);
g_topology_unlock();
return (error);
}
/*
* XXX: Until we have unmessed the ioctl situation, there is a race against
* XXX: a concurrent orphanization. We cannot close it by holding topology
* XXX: since that would prevent us from doing our job, and stalling events
* XXX: will break (actually: stall) the BSD disklabel hacks.
*/
static int
g_dev_ioctl(struct cdev *dev, u_long cmd, caddr_t data, int fflag, struct thread *td)
{
struct g_consumer *cp;
struct g_provider *pp;
off_t offset, length, chunk;
int i, error;
cp = dev->si_drv2;
pp = cp->provider;
error = 0;
KASSERT(cp->acr || cp->acw,
("Consumer with zero access count in g_dev_ioctl"));
i = IOCPARM_LEN(cmd);
switch (cmd) {
case DIOCGSECTORSIZE:
*(u_int *)data = cp->provider->sectorsize;
if (*(u_int *)data == 0)
error = ENOENT;
break;
case DIOCGMEDIASIZE:
*(off_t *)data = cp->provider->mediasize;
if (*(off_t *)data == 0)
error = ENOENT;
break;
case DIOCGFWSECTORS:
error = g_io_getattr("GEOM::fwsectors", cp, &i, data);
if (error == 0 && *(u_int *)data == 0)
error = ENOENT;
break;
case DIOCGFWHEADS:
error = g_io_getattr("GEOM::fwheads", cp, &i, data);
if (error == 0 && *(u_int *)data == 0)
error = ENOENT;
break;
case DIOCGFRONTSTUFF:
error = g_io_getattr("GEOM::frontstuff", cp, &i, data);
break;
case DIOCSKERNELDUMP:
if (*(u_int *)data == 0)
error = g_dev_setdumpdev(NULL, td);
else
error = g_dev_setdumpdev(dev, td);
break;
case DIOCGFLUSH:
error = g_io_flush(cp);
break;
case DIOCGDELETE:
offset = ((off_t *)data)[0];
length = ((off_t *)data)[1];
if ((offset % cp->provider->sectorsize) != 0 ||
(length % cp->provider->sectorsize) != 0 || length <= 0) {
printf("%s: offset=%jd length=%jd\n", __func__, offset,
length);
error = EINVAL;
break;
}
while (length > 0) {
chunk = length;
if (g_dev_del_max_sectors != 0 && chunk >
g_dev_del_max_sectors * cp->provider->sectorsize) {
chunk = g_dev_del_max_sectors *
cp->provider->sectorsize;
}
error = g_delete_data(cp, offset, chunk);
length -= chunk;
offset += chunk;
if (error)
break;
/*
* Since the request size can be large, the service
* time can be is likewise. We make this ioctl
* interruptible by checking for signals for each bio.
*/
if (SIGPENDING(td))
break;
}
break;
case DIOCGIDENT:
error = g_io_getattr("GEOM::ident", cp, &i, data);
break;
case DIOCGPROVIDERNAME:
if (pp == NULL)
return (ENOENT);
strlcpy(data, pp->name, i);
break;
case DIOCGSTRIPESIZE:
*(off_t *)data = cp->provider->stripesize;
break;
case DIOCGSTRIPEOFFSET:
*(off_t *)data = cp->provider->stripeoffset;
break;
case DIOCGPHYSPATH:
error = g_io_getattr("GEOM::physpath", cp, &i, data);
if (error == 0 && *(char *)data == '\0')
error = ENOENT;
break;
case DIOCGATTR: {
struct diocgattr_arg *arg = (struct diocgattr_arg *)data;
if (arg->len > sizeof(arg->value)) {
error = EINVAL;
break;
}
error = g_io_getattr(arg->name, cp, &arg->len, &arg->value);
break;
}
case DIOCZONECMD: {
struct disk_zone_args *zone_args =(struct disk_zone_args *)data;
struct disk_zone_rep_entry *new_entries, *old_entries;
struct disk_zone_report *rep;
size_t alloc_size;
old_entries = NULL;
new_entries = NULL;
rep = NULL;
alloc_size = 0;
if (zone_args->zone_cmd == DISK_ZONE_REPORT_ZONES) {
rep = &zone_args->zone_params.report;
alloc_size = rep->entries_allocated *
sizeof(struct disk_zone_rep_entry);
if (alloc_size != 0)
new_entries = g_malloc(alloc_size,
M_WAITOK| M_ZERO);
old_entries = rep->entries;
rep->entries = new_entries;
}
error = g_io_zonecmd(zone_args, cp);
if ((zone_args->zone_cmd == DISK_ZONE_REPORT_ZONES)
&& (alloc_size != 0)
&& (error == 0)) {
error = copyout(new_entries, old_entries, alloc_size);
}
if ((old_entries != NULL)
&& (rep != NULL))
rep->entries = old_entries;
if (new_entries != NULL)
g_free(new_entries);
break;
}
default:
if (cp->provider->geom->ioctl != NULL) {
error = cp->provider->geom->ioctl(cp->provider, cmd, data, fflag, td);
} else {
error = ENOIOCTL;
}
}
return (error);
}
static void
g_dev_done(struct bio *bp2)
{
struct g_consumer *cp;
struct g_dev_softc *sc;
struct bio *bp;
int destroy;
cp = bp2->bio_from;
sc = cp->private;
bp = bp2->bio_parent;
bp->bio_error = bp2->bio_error;
bp->bio_completed = bp2->bio_completed;
bp->bio_resid = bp->bio_length - bp2->bio_completed;
if (bp2->bio_cmd == BIO_ZONE)
bcopy(&bp2->bio_zone, &bp->bio_zone, sizeof(bp->bio_zone));
if (bp2->bio_error != 0) {
g_trace(G_T_BIO, "g_dev_done(%p) had error %d",
bp2, bp2->bio_error);
bp->bio_flags |= BIO_ERROR;
} else {
g_trace(G_T_BIO, "g_dev_done(%p/%p) resid %ld completed %jd",
bp2, bp, bp2->bio_resid, (intmax_t)bp2->bio_completed);
}
g_destroy_bio(bp2);
destroy = 0;
mtx_lock(&sc->sc_mtx);
if ((--sc->sc_active) == 0) {
if (sc->sc_open == 0)
wakeup(&sc->sc_active);
if (sc->sc_dev == NULL)
destroy = 1;
}
mtx_unlock(&sc->sc_mtx);
if (destroy)
g_post_event(g_dev_destroy, cp, M_NOWAIT, NULL);
biodone(bp);
}
static void
g_dev_strategy(struct bio *bp)
{
struct g_consumer *cp;
struct bio *bp2;
struct cdev *dev;
struct g_dev_softc *sc;
KASSERT(bp->bio_cmd == BIO_READ ||
bp->bio_cmd == BIO_WRITE ||
bp->bio_cmd == BIO_DELETE ||
bp->bio_cmd == BIO_FLUSH ||
bp->bio_cmd == BIO_ZONE,
("Wrong bio_cmd bio=%p cmd=%d", bp, bp->bio_cmd));
dev = bp->bio_dev;
cp = dev->si_drv2;
sc = cp->private;
KASSERT(cp->acr || cp->acw,
("Consumer with zero access count in g_dev_strategy"));
#ifdef INVARIANTS
if ((bp->bio_offset % cp->provider->sectorsize) != 0 ||
(bp->bio_bcount % cp->provider->sectorsize) != 0) {
bp->bio_resid = bp->bio_bcount;
biofinish(bp, NULL, EINVAL);
return;
}
#endif
mtx_lock(&sc->sc_mtx);
KASSERT(sc->sc_open > 0, ("Closed device in g_dev_strategy"));
sc->sc_active++;
mtx_unlock(&sc->sc_mtx);
for (;;) {
/*
* XXX: This is not an ideal solution, but I believe it to
* XXX: deadlock safely, all things considered.
*/
bp2 = g_clone_bio(bp);
if (bp2 != NULL)
break;
pause("gdstrat", hz / 10);
}
KASSERT(bp2 != NULL, ("XXX: ENOMEM in a bad place"));
bp2->bio_done = g_dev_done;
g_trace(G_T_BIO,
"g_dev_strategy(%p/%p) offset %jd length %jd data %p cmd %d",
bp, bp2, (intmax_t)bp->bio_offset, (intmax_t)bp2->bio_length,
bp2->bio_data, bp2->bio_cmd);
g_io_request(bp2, cp);
KASSERT(cp->acr || cp->acw,
("g_dev_strategy raced with g_dev_close and lost"));
}
/*
* g_dev_callback()
*
* Called by devfs when asynchronous device destruction is completed.
* - Mark that we have no attached device any more.
* - If there are no outstanding requests, schedule geom destruction.
* Otherwise destruction will be scheduled later by g_dev_done().
*/
static void
g_dev_callback(void *arg)
{
struct g_consumer *cp;
struct g_dev_softc *sc;
int destroy;
cp = arg;
sc = cp->private;
g_trace(G_T_TOPOLOGY, "g_dev_callback(%p(%s))", cp, cp->geom->name);
mtx_lock(&sc->sc_mtx);
sc->sc_dev = NULL;
sc->sc_alias = NULL;
destroy = (sc->sc_active == 0);
mtx_unlock(&sc->sc_mtx);
if (destroy)
g_post_event(g_dev_destroy, cp, M_WAITOK, NULL);
}
/*
* g_dev_orphan()
*
* Called from below when the provider orphaned us.
* - Clear any dump settings.
* - Request asynchronous device destruction to prevent any more requests
* from coming in. The provider is already marked with an error, so
* anything which comes in the interim will be returned immediately.
*/
static void
g_dev_orphan(struct g_consumer *cp)
{
struct cdev *dev;
struct g_dev_softc *sc;
g_topology_assert();
sc = cp->private;
dev = sc->sc_dev;
g_trace(G_T_TOPOLOGY, "g_dev_orphan(%p(%s))", cp, cp->geom->name);
/* Reset any dump-area set on this device */
if (dev->si_flags & SI_DUMPDEV)
(void)set_dumper(NULL, NULL, curthread);
/* Destroy the struct cdev *so we get no more requests */
destroy_dev_sched_cb(dev, g_dev_callback, cp);
}
DECLARE_GEOM_CLASS(g_dev_class, g_dev);