[PATCH v6 6/7] sd: Implement support for ZBC devices

From: Shaun Tancheff
Date: Fri Sep 30 2016 - 15:15:12 EST


From: Hannes Reinecke <hare@xxxxxxx>

Implement ZBC support functions to setup zoned disks, both
host-managed and host-aware models. Only zoned disks that satisfy
the following conditions are supported:
1) All zones are the same size, with the exception of an eventual
last smaller runt zone.
2) For host-managed disks, reads are unrestricted (reads are not
failed due to zone or write pointer alignement constraints).
Zoned disks that do not satisfy these 2 conditions are setup with
a capacity of 0 to prevent their use.

The function sd_zbc_read_zones, called from sd_revalidate_disk,
checks that the device satisfies the above two constraints. This
function may also change the disk capacity previously set by
sd_read_capacity for devices reporting only the capacity of
conventional zones at the beginning of the LBA range (i.e. devices
reporting rc_basis set to 0).

The capacity message output was moved out of sd_read_capacity into
a new function sd_print_capacity to include this eventual capacity
change by sd_zbc_read_zones. This new function also includes a call
to sd_zbc_print_zones to display the number of zones and zone size
of the device.

Signed-off-by: Hannes Reinecke <hare@xxxxxxx>

[Damien: * Removed zone cache support
* Removed mapping of discard to reset write pointer command
* Modified sd_zbc_read_zones to include checks that the
device satisfies the kernel constraints
* Implemeted REPORT ZONES setup and post-processing based
on code from Shaun Tancheff <shaun.tancheff@xxxxxxxxxxx>
* Removed confusing use of 512B sector units in functions
interface]
Signed-off-by: Damien Le Moal <damien.lemoal@xxxxxxxx>
Reviewed-by: Christoph Hellwig <hch@xxxxxx>
Reviewed-by: Shaun Tancheff <shaun.tancheff@xxxxxxxxxxx>
Tested-by: Shaun Tancheff <shaun.tancheff@xxxxxxxxxxx>
---
Changes from v5:
* Rebased on Jens' for-4.9/block branch (v5 is based on next-20160928).

drivers/scsi/Makefile | 1 +
drivers/scsi/sd.c | 141 ++++++++---
drivers/scsi/sd.h | 67 +++++
drivers/scsi/sd_zbc.c | 627 ++++++++++++++++++++++++++++++++++++++++++++++
include/scsi/scsi_proto.h | 17 ++
5 files changed, 820 insertions(+), 33 deletions(-)
create mode 100644 drivers/scsi/sd_zbc.c

diff --git a/drivers/scsi/Makefile b/drivers/scsi/Makefile
index d539798..fabcb6d 100644
--- a/drivers/scsi/Makefile
+++ b/drivers/scsi/Makefile
@@ -179,6 +179,7 @@ hv_storvsc-y := storvsc_drv.o

sd_mod-objs := sd.o
sd_mod-$(CONFIG_BLK_DEV_INTEGRITY) += sd_dif.o
+sd_mod-$(CONFIG_BLK_DEV_ZONED) += sd_zbc.o

sr_mod-objs := sr.o sr_ioctl.o sr_vendor.o
ncr53c8xx-flags-$(CONFIG_SCSI_ZALON) \
diff --git a/drivers/scsi/sd.c b/drivers/scsi/sd.c
index d3e852a..fb324ac 100644
--- a/drivers/scsi/sd.c
+++ b/drivers/scsi/sd.c
@@ -92,6 +92,7 @@ MODULE_ALIAS_BLOCKDEV_MAJOR(SCSI_DISK15_MAJOR);
MODULE_ALIAS_SCSI_DEVICE(TYPE_DISK);
MODULE_ALIAS_SCSI_DEVICE(TYPE_MOD);
MODULE_ALIAS_SCSI_DEVICE(TYPE_RBC);
+MODULE_ALIAS_SCSI_DEVICE(TYPE_ZBC);

#if !defined(CONFIG_DEBUG_BLOCK_EXT_DEVT)
#define SD_MINORS 16
@@ -162,7 +163,7 @@ cache_type_store(struct device *dev, struct device_attribute *attr,
static const char temp[] = "temporary ";
int len;

- if (sdp->type != TYPE_DISK)
+ if (sdp->type != TYPE_DISK && sdp->type != TYPE_ZBC)
/* no cache control on RBC devices; theoretically they
* can do it, but there's probably so many exceptions
* it's not worth the risk */
@@ -261,7 +262,7 @@ allow_restart_store(struct device *dev, struct device_attribute *attr,
if (!capable(CAP_SYS_ADMIN))
return -EACCES;

- if (sdp->type != TYPE_DISK)
+ if (sdp->type != TYPE_DISK && sdp->type != TYPE_ZBC)
return -EINVAL;

sdp->allow_restart = simple_strtoul(buf, NULL, 10);
@@ -391,6 +392,11 @@ provisioning_mode_store(struct device *dev, struct device_attribute *attr,
if (!capable(CAP_SYS_ADMIN))
return -EACCES;

+ if (sd_is_zoned(sdkp)) {
+ sd_config_discard(sdkp, SD_LBP_DISABLE);
+ return count;
+ }
+
if (sdp->type != TYPE_DISK)
return -EINVAL;

@@ -458,7 +464,7 @@ max_write_same_blocks_store(struct device *dev, struct device_attribute *attr,
if (!capable(CAP_SYS_ADMIN))
return -EACCES;

- if (sdp->type != TYPE_DISK)
+ if (sdp->type != TYPE_DISK && sdp->type != TYPE_ZBC)
return -EINVAL;

err = kstrtoul(buf, 10, &max);
@@ -843,6 +849,12 @@ static int sd_setup_write_same_cmnd(struct scsi_cmnd *cmd)

BUG_ON(bio_offset(bio) || bio_iovec(bio).bv_len != sdp->sector_size);

+ if (sd_is_zoned(sdkp)) {
+ ret = sd_zbc_setup_read_write(cmd);
+ if (ret != BLKPREP_OK)
+ return ret;
+ }
+
sector >>= ilog2(sdp->sector_size) - 9;
nr_sectors >>= ilog2(sdp->sector_size) - 9;

@@ -962,6 +974,12 @@ static int sd_setup_read_write_cmnd(struct scsi_cmnd *SCpnt)
SCSI_LOG_HLQUEUE(2, scmd_printk(KERN_INFO, SCpnt, "block=%llu\n",
(unsigned long long)block));

+ if (sd_is_zoned(sdkp)) {
+ ret = sd_zbc_setup_read_write(SCpnt);
+ if (ret != BLKPREP_OK)
+ goto out;
+ }
+
/*
* If we have a 1K hardware sectorsize, prevent access to single
* 512 byte sectors. In theory we could handle this - in fact
@@ -1148,6 +1166,10 @@ static int sd_init_command(struct scsi_cmnd *cmd)
case REQ_OP_READ:
case REQ_OP_WRITE:
return sd_setup_read_write_cmnd(cmd);
+ case REQ_OP_ZONE_REPORT:
+ return sd_zbc_setup_report_cmnd(cmd);
+ case REQ_OP_ZONE_RESET:
+ return sd_zbc_setup_reset_cmnd(cmd);
default:
BUG();
}
@@ -1779,7 +1801,10 @@ static int sd_done(struct scsi_cmnd *SCpnt)
unsigned char op = SCpnt->cmnd[0];
unsigned char unmap = SCpnt->cmnd[1] & 8;

- if (req_op(req) == REQ_OP_DISCARD || req_op(req) == REQ_OP_WRITE_SAME) {
+ switch (req_op(req)) {
+ case REQ_OP_DISCARD:
+ case REQ_OP_WRITE_SAME:
+ case REQ_OP_ZONE_RESET:
if (!result) {
good_bytes = blk_rq_bytes(req);
scsi_set_resid(SCpnt, 0);
@@ -1787,6 +1812,17 @@ static int sd_done(struct scsi_cmnd *SCpnt)
good_bytes = 0;
scsi_set_resid(SCpnt, blk_rq_bytes(req));
}
+ break;
+ case REQ_OP_ZONE_REPORT:
+ if (!result) {
+ good_bytes = scsi_bufflen(SCpnt)
+ - scsi_get_resid(SCpnt);
+ scsi_set_resid(SCpnt, 0);
+ } else {
+ good_bytes = 0;
+ scsi_set_resid(SCpnt, blk_rq_bytes(req));
+ }
+ break;
}

if (result) {
@@ -1847,7 +1883,11 @@ static int sd_done(struct scsi_cmnd *SCpnt)
default:
break;
}
+
out:
+ if (sd_is_zoned(sdkp))
+ sd_zbc_complete(SCpnt, good_bytes, &sshdr);
+
SCSI_LOG_HLCOMPLETE(1, scmd_printk(KERN_INFO, SCpnt,
"sd_done: completed %d of %d bytes\n",
good_bytes, scsi_bufflen(SCpnt)));
@@ -1982,7 +2022,6 @@ sd_spinup_disk(struct scsi_disk *sdkp)
}
}

-
/*
* Determine whether disk supports Data Integrity Field.
*/
@@ -2132,6 +2171,9 @@ static int read_capacity_16(struct scsi_disk *sdkp, struct scsi_device *sdp,
/* Logical blocks per physical block exponent */
sdkp->physical_block_size = (1 << (buffer[13] & 0xf)) * sector_size;

+ /* RC basis */
+ sdkp->rc_basis = (buffer[12] >> 4) & 0x3;
+
/* Lowest aligned logical block */
alignment = ((buffer[14] & 0x3f) << 8 | buffer[15]) * sector_size;
blk_queue_alignment_offset(sdp->request_queue, alignment);
@@ -2241,7 +2283,6 @@ sd_read_capacity(struct scsi_disk *sdkp, unsigned char *buffer)
{
int sector_size;
struct scsi_device *sdp = sdkp->device;
- sector_t old_capacity = sdkp->capacity;

if (sd_try_rc16_first(sdp)) {
sector_size = read_capacity_16(sdkp, sdp, buffer);
@@ -2322,35 +2363,44 @@ got_data:
sector_size = 512;
}
blk_queue_logical_block_size(sdp->request_queue, sector_size);
+ blk_queue_physical_block_size(sdp->request_queue,
+ sdkp->physical_block_size);
+ sdkp->device->sector_size = sector_size;

- {
- char cap_str_2[10], cap_str_10[10];
+ if (sdkp->capacity > 0xffffffff)
+ sdp->use_16_for_rw = 1;

- string_get_size(sdkp->capacity, sector_size,
- STRING_UNITS_2, cap_str_2, sizeof(cap_str_2));
- string_get_size(sdkp->capacity, sector_size,
- STRING_UNITS_10, cap_str_10,
- sizeof(cap_str_10));
+}

- if (sdkp->first_scan || old_capacity != sdkp->capacity) {
- sd_printk(KERN_NOTICE, sdkp,
- "%llu %d-byte logical blocks: (%s/%s)\n",
- (unsigned long long)sdkp->capacity,
- sector_size, cap_str_10, cap_str_2);
+/*
+ * Print disk capacity
+ */
+static void
+sd_print_capacity(struct scsi_disk *sdkp,
+ sector_t old_capacity)
+{
+ int sector_size = sdkp->device->sector_size;
+ char cap_str_2[10], cap_str_10[10];

- if (sdkp->physical_block_size != sector_size)
- sd_printk(KERN_NOTICE, sdkp,
- "%u-byte physical blocks\n",
- sdkp->physical_block_size);
- }
- }
+ string_get_size(sdkp->capacity, sector_size,
+ STRING_UNITS_2, cap_str_2, sizeof(cap_str_2));
+ string_get_size(sdkp->capacity, sector_size,
+ STRING_UNITS_10, cap_str_10,
+ sizeof(cap_str_10));

- if (sdkp->capacity > 0xffffffff)
- sdp->use_16_for_rw = 1;
+ if (sdkp->first_scan || old_capacity != sdkp->capacity) {
+ sd_printk(KERN_NOTICE, sdkp,
+ "%llu %d-byte logical blocks: (%s/%s)\n",
+ (unsigned long long)sdkp->capacity,
+ sector_size, cap_str_10, cap_str_2);

- blk_queue_physical_block_size(sdp->request_queue,
- sdkp->physical_block_size);
- sdkp->device->sector_size = sector_size;
+ if (sdkp->physical_block_size != sector_size)
+ sd_printk(KERN_NOTICE, sdkp,
+ "%u-byte physical blocks\n",
+ sdkp->physical_block_size);
+
+ sd_zbc_print_zones(sdkp);
+ }
}

/* called with buffer of length 512 */
@@ -2612,7 +2662,7 @@ static void sd_read_app_tag_own(struct scsi_disk *sdkp, unsigned char *buffer)
struct scsi_mode_data data;
struct scsi_sense_hdr sshdr;

- if (sdp->type != TYPE_DISK)
+ if (sdp->type != TYPE_DISK && sdp->type != TYPE_ZBC)
return;

if (sdkp->protection_type == 0)
@@ -2719,6 +2769,7 @@ static void sd_read_block_limits(struct scsi_disk *sdkp)
*/
static void sd_read_block_characteristics(struct scsi_disk *sdkp)
{
+ struct request_queue *q = sdkp->disk->queue;
unsigned char *buffer;
u16 rot;
const int vpd_len = 64;
@@ -2733,10 +2784,21 @@ static void sd_read_block_characteristics(struct scsi_disk *sdkp)
rot = get_unaligned_be16(&buffer[4]);

if (rot == 1) {
- queue_flag_set_unlocked(QUEUE_FLAG_NONROT, sdkp->disk->queue);
- queue_flag_clear_unlocked(QUEUE_FLAG_ADD_RANDOM, sdkp->disk->queue);
+ queue_flag_set_unlocked(QUEUE_FLAG_NONROT, q);
+ queue_flag_clear_unlocked(QUEUE_FLAG_ADD_RANDOM, q);
}

+ sdkp->zoned = (buffer[8] >> 4) & 3;
+ if (sdkp->zoned == 1)
+ q->limits.zoned = BLK_ZONED_HA;
+ else if (sdkp->device->type == TYPE_ZBC)
+ q->limits.zoned = BLK_ZONED_HM;
+ else
+ q->limits.zoned = BLK_ZONED_NONE;
+ if (blk_queue_is_zoned(q) && sdkp->first_scan)
+ sd_printk(KERN_NOTICE, sdkp, "Host-%s zoned block device\n",
+ q->limits.zoned == BLK_ZONED_HM ? "managed" : "aware");
+
out:
kfree(buffer);
}
@@ -2808,6 +2870,7 @@ static int sd_revalidate_disk(struct gendisk *disk)
struct scsi_disk *sdkp = scsi_disk(disk);
struct scsi_device *sdp = sdkp->device;
struct request_queue *q = sdkp->disk->queue;
+ sector_t old_capacity = sdkp->capacity;
unsigned char *buffer;
unsigned int dev_max, rw_max;

@@ -2841,8 +2904,11 @@ static int sd_revalidate_disk(struct gendisk *disk)
sd_read_block_provisioning(sdkp);
sd_read_block_limits(sdkp);
sd_read_block_characteristics(sdkp);
+ sd_zbc_read_zones(sdkp, buffer);
}

+ sd_print_capacity(sdkp, old_capacity);
+
sd_read_write_protect_flag(sdkp, buffer);
sd_read_cache_type(sdkp, buffer);
sd_read_app_tag_own(sdkp, buffer);
@@ -3040,9 +3106,16 @@ static int sd_probe(struct device *dev)

scsi_autopm_get_device(sdp);
error = -ENODEV;
- if (sdp->type != TYPE_DISK && sdp->type != TYPE_MOD && sdp->type != TYPE_RBC)
+ if (sdp->type != TYPE_DISK &&
+ sdp->type != TYPE_ZBC &&
+ sdp->type != TYPE_MOD &&
+ sdp->type != TYPE_RBC)
goto out;

+#ifndef CONFIG_BLK_DEV_ZONED
+ if (sdp->type == TYPE_ZBC)
+ goto out;
+#endif
SCSI_LOG_HLQUEUE(3, sdev_printk(KERN_INFO, sdp,
"sd_probe\n"));

@@ -3146,6 +3219,8 @@ static int sd_remove(struct device *dev)
del_gendisk(sdkp->disk);
sd_shutdown(dev);

+ sd_zbc_remove(sdkp);
+
blk_register_region(devt, SD_MINORS, NULL,
sd_default_probe, NULL, NULL);

diff --git a/drivers/scsi/sd.h b/drivers/scsi/sd.h
index 765a6f1..4e140d2 100644
--- a/drivers/scsi/sd.h
+++ b/drivers/scsi/sd.h
@@ -64,6 +64,15 @@ struct scsi_disk {
struct scsi_device *device;
struct device dev;
struct gendisk *disk;
+#ifdef CONFIG_BLK_DEV_ZONED
+ unsigned int nr_zones;
+ unsigned int zone_blocks;
+ unsigned int zone_shift;
+ unsigned long *zones_wlock;
+ unsigned int zones_optimal_open;
+ unsigned int zones_optimal_nonseq;
+ unsigned int zones_max_open;
+#endif
atomic_t openers;
sector_t capacity; /* size in logical blocks */
u32 max_xfer_blocks;
@@ -94,6 +103,9 @@ struct scsi_disk {
unsigned lbpvpd : 1;
unsigned ws10 : 1;
unsigned ws16 : 1;
+ unsigned rc_basis: 2;
+ unsigned zoned: 2;
+ unsigned urswrz : 1;
};
#define to_scsi_disk(obj) container_of(obj,struct scsi_disk,dev)

@@ -156,6 +168,11 @@ static inline unsigned int logical_to_bytes(struct scsi_device *sdev, sector_t b
return blocks * sdev->sector_size;
}

+static inline sector_t sectors_to_logical(struct scsi_device *sdev, sector_t sector)
+{
+ return sector >> (ilog2(sdev->sector_size) - 9);
+}
+
/*
* A DIF-capable target device can be formatted with different
* protection schemes. Currently 0 through 3 are defined:
@@ -269,4 +286,54 @@ static inline void sd_dif_complete(struct scsi_cmnd *cmd, unsigned int a)

#endif /* CONFIG_BLK_DEV_INTEGRITY */

+static inline int sd_is_zoned(struct scsi_disk *sdkp)
+{
+ return sdkp->zoned == 1 || sdkp->device->type == TYPE_ZBC;
+}
+
+#ifdef CONFIG_BLK_DEV_ZONED
+
+extern int sd_zbc_read_zones(struct scsi_disk *sdkp, unsigned char *buffer);
+extern void sd_zbc_remove(struct scsi_disk *sdkp);
+extern void sd_zbc_print_zones(struct scsi_disk *sdkp);
+extern int sd_zbc_setup_read_write(struct scsi_cmnd *cmd);
+extern int sd_zbc_setup_report_cmnd(struct scsi_cmnd *cmd);
+extern int sd_zbc_setup_reset_cmnd(struct scsi_cmnd *cmd);
+extern void sd_zbc_complete(struct scsi_cmnd *cmd, unsigned int good_bytes,
+ struct scsi_sense_hdr *sshdr);
+
+#else /* CONFIG_BLK_DEV_ZONED */
+
+static inline int sd_zbc_read_zones(struct scsi_disk *sdkp,
+ unsigned char *buf)
+{
+ return 0;
+}
+
+static inline void sd_zbc_remove(struct scsi_disk *sdkp) {}
+
+static inline void sd_zbc_print_zones(struct scsi_disk *sdkp) {}
+
+static inline int sd_zbc_setup_read_write(struct scsi_cmnd *cmd)
+{
+ /* Let the drive fail requests */
+ return BLKPREP_OK;
+}
+
+static inline int sd_zbc_setup_report_cmnd(struct scsi_cmnd *cmd)
+{
+ return BLKPREP_KILL;
+}
+
+static inline int sd_zbc_setup_reset_cmnd(struct scsi_cmnd *cmd)
+{
+ return BLKPREP_KILL;
+}
+
+static inline void sd_zbc_complete(struct scsi_cmnd *cmd,
+ unsigned int good_bytes,
+ struct scsi_sense_hdr *sshdr) {}
+
+#endif /* CONFIG_BLK_DEV_ZONED */
+
#endif /* _SCSI_DISK_H */
diff --git a/drivers/scsi/sd_zbc.c b/drivers/scsi/sd_zbc.c
new file mode 100644
index 0000000..05ff208
--- /dev/null
+++ b/drivers/scsi/sd_zbc.c
@@ -0,0 +1,627 @@
+/*
+ * SCSI Zoned Block commands
+ *
+ * Copyright (C) 2014-2015 SUSE Linux GmbH
+ * Written by: Hannes Reinecke <hare@xxxxxxx>
+ * Modified by: Damien Le Moal <damien.lemoal@xxxxxxxx>
+ * Modified by: Shaun Tancheff <shaun.tancheff@xxxxxxxxxxx>
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License version
+ * 2 as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; see the file COPYING. If not, write to
+ * the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139,
+ * USA.
+ *
+ */
+
+#include <linux/blkdev.h>
+
+#include <asm/unaligned.h>
+
+#include <scsi/scsi.h>
+#include <scsi/scsi_cmnd.h>
+#include <scsi/scsi_dbg.h>
+#include <scsi/scsi_device.h>
+#include <scsi/scsi_driver.h>
+#include <scsi/scsi_host.h>
+#include <scsi/scsi_eh.h>
+
+#include "sd.h"
+#include "scsi_priv.h"
+
+enum zbc_zone_type {
+ ZBC_ZONE_TYPE_CONV = 0x1,
+ ZBC_ZONE_TYPE_SEQWRITE_REQ,
+ ZBC_ZONE_TYPE_SEQWRITE_PREF,
+ ZBC_ZONE_TYPE_RESERVED,
+};
+
+enum zbc_zone_cond {
+ ZBC_ZONE_COND_NO_WP,
+ ZBC_ZONE_COND_EMPTY,
+ ZBC_ZONE_COND_IMP_OPEN,
+ ZBC_ZONE_COND_EXP_OPEN,
+ ZBC_ZONE_COND_CLOSED,
+ ZBC_ZONE_COND_READONLY = 0xd,
+ ZBC_ZONE_COND_FULL,
+ ZBC_ZONE_COND_OFFLINE,
+};
+
+/**
+ * Convert a zone descriptor to a zone struct.
+ */
+static void sd_zbc_parse_report(struct scsi_disk *sdkp,
+ u8 *buf,
+ struct blk_zone *zone)
+{
+ struct scsi_device *sdp = sdkp->device;
+
+ memset(zone, 0, sizeof(struct blk_zone));
+
+ zone->type = buf[0] & 0x0f;
+ zone->cond = (buf[1] >> 4) & 0xf;
+ if (buf[1] & 0x01)
+ zone->reset = 1;
+ if (buf[1] & 0x02)
+ zone->non_seq = 1;
+
+ zone->len = logical_to_sectors(sdp, get_unaligned_be64(&buf[8]));
+ zone->start = logical_to_sectors(sdp, get_unaligned_be64(&buf[16]));
+ zone->wp = logical_to_sectors(sdp, get_unaligned_be64(&buf[24]));
+ if (zone->type != ZBC_ZONE_TYPE_CONV &&
+ zone->cond == ZBC_ZONE_COND_FULL)
+ zone->wp = zone->start + zone->len;
+}
+
+/**
+ * Issue a REPORT ZONES scsi command.
+ */
+static int sd_zbc_report_zones(struct scsi_disk *sdkp, unsigned char *buf,
+ unsigned int buflen, sector_t lba)
+{
+ struct scsi_device *sdp = sdkp->device;
+ const int timeout = sdp->request_queue->rq_timeout;
+ struct scsi_sense_hdr sshdr;
+ unsigned char cmd[16];
+ unsigned int rep_len;
+ int result;
+
+ memset(cmd, 0, 16);
+ cmd[0] = ZBC_IN;
+ cmd[1] = ZI_REPORT_ZONES;
+ put_unaligned_be64(lba, &cmd[2]);
+ put_unaligned_be32(buflen, &cmd[10]);
+ memset(buf, 0, buflen);
+
+ result = scsi_execute_req(sdp, cmd, DMA_FROM_DEVICE,
+ buf, buflen, &sshdr,
+ timeout, SD_MAX_RETRIES, NULL);
+ if (result) {
+ sd_printk(KERN_ERR, sdkp,
+ "REPORT ZONES lba %llu failed with %d/%d\n",
+ (unsigned long long)lba,
+ host_byte(result), driver_byte(result));
+ return -EIO;
+ }
+
+ rep_len = get_unaligned_be32(&buf[0]);
+ if (rep_len < 64) {
+ sd_printk(KERN_ERR, sdkp,
+ "REPORT ZONES report invalid length %u\n",
+ rep_len);
+ return -EIO;
+ }
+
+ return 0;
+}
+
+int sd_zbc_setup_report_cmnd(struct scsi_cmnd *cmd)
+{
+ struct request *rq = cmd->request;
+ struct scsi_disk *sdkp = scsi_disk(rq->rq_disk);
+ sector_t lba, sector = blk_rq_pos(rq);
+ unsigned int nr_bytes = blk_rq_bytes(rq);
+ int ret;
+
+ WARN_ON(nr_bytes == 0);
+
+ if (!sd_is_zoned(sdkp))
+ /* Not a zoned device */
+ return BLKPREP_KILL;
+
+ ret = scsi_init_io(cmd);
+ if (ret != BLKPREP_OK)
+ return ret;
+
+ cmd->cmd_len = 16;
+ memset(cmd->cmnd, 0, cmd->cmd_len);
+ cmd->cmnd[0] = ZBC_IN;
+ cmd->cmnd[1] = ZI_REPORT_ZONES;
+ lba = sectors_to_logical(sdkp->device, sector);
+ put_unaligned_be64(lba, &cmd->cmnd[2]);
+ put_unaligned_be32(nr_bytes, &cmd->cmnd[10]);
+ /* Do partial report for speeding things up */
+ cmd->cmnd[14] = ZBC_REPORT_ZONE_PARTIAL;
+
+ cmd->sc_data_direction = DMA_FROM_DEVICE;
+ cmd->sdb.length = nr_bytes;
+ cmd->transfersize = sdkp->device->sector_size;
+ cmd->allowed = 0;
+
+ /*
+ * Report may return less bytes than requested. Make sure
+ * to report completion on the entire initial request.
+ */
+ rq->__data_len = nr_bytes;
+
+ return BLKPREP_OK;
+}
+
+static void sd_zbc_report_zones_complete(struct scsi_cmnd *scmd,
+ unsigned int good_bytes)
+{
+ struct request *rq = scmd->request;
+ struct scsi_disk *sdkp = scsi_disk(rq->rq_disk);
+ struct sg_mapping_iter miter;
+ struct blk_zone_report_hdr hdr;
+ struct blk_zone zone;
+ unsigned int offset, bytes = 0;
+ unsigned long flags;
+ u8 *buf;
+
+ if (good_bytes < 64)
+ return;
+
+ memset(&hdr, 0, sizeof(struct blk_zone_report_hdr));
+
+ sg_miter_start(&miter, scsi_sglist(scmd), scsi_sg_count(scmd),
+ SG_MITER_TO_SG | SG_MITER_ATOMIC);
+
+ local_irq_save(flags);
+ while (sg_miter_next(&miter) && bytes < good_bytes) {
+
+ buf = miter.addr;
+ offset = 0;
+
+ if (bytes == 0) {
+ /* Set the report header */
+ hdr.nr_zones = min_t(unsigned int,
+ (good_bytes - 64) / 64,
+ get_unaligned_be32(&buf[0]) / 64);
+ memcpy(buf, &hdr, sizeof(struct blk_zone_report_hdr));
+ offset += 64;
+ bytes += 64;
+ }
+
+ /* Parse zone descriptors */
+ while (offset < miter.length && hdr.nr_zones) {
+ WARN_ON(offset > miter.length);
+ buf = miter.addr + offset;
+ sd_zbc_parse_report(sdkp, buf, &zone);
+ memcpy(buf, &zone, sizeof(struct blk_zone));
+ offset += 64;
+ bytes += 64;
+ hdr.nr_zones--;
+ }
+
+ if (!hdr.nr_zones)
+ break;
+
+ }
+ sg_miter_stop(&miter);
+ local_irq_restore(flags);
+}
+
+static inline sector_t sd_zbc_zone_sectors(struct scsi_disk *sdkp)
+{
+ return logical_to_sectors(sdkp->device, sdkp->zone_blocks);
+}
+
+static inline unsigned int sd_zbc_zone_no(struct scsi_disk *sdkp,
+ sector_t sector)
+{
+ return sectors_to_logical(sdkp->device, sector) >> sdkp->zone_shift;
+}
+
+int sd_zbc_setup_reset_cmnd(struct scsi_cmnd *cmd)
+{
+ struct request *rq = cmd->request;
+ struct scsi_disk *sdkp = scsi_disk(rq->rq_disk);
+ sector_t sector = blk_rq_pos(rq);
+ sector_t block = sectors_to_logical(sdkp->device, sector);
+
+ if (!sd_is_zoned(sdkp))
+ /* Not a zoned device */
+ return BLKPREP_KILL;
+
+ if (sdkp->device->changed)
+ return BLKPREP_KILL;
+
+ if (sector & (sd_zbc_zone_sectors(sdkp) - 1))
+ /* Unaligned request */
+ return BLKPREP_KILL;
+
+ /* Do not allow concurrent reset and writes */
+ if (!test_and_set_bit(sd_zbc_zone_no(sdkp, sector),
+ sdkp->zones_wlock))
+ return BLKPREP_DEFER;
+
+ cmd->cmd_len = 16;
+ memset(cmd->cmnd, 0, cmd->cmd_len);
+ cmd->cmnd[0] = ZBC_OUT;
+ cmd->cmnd[1] = ZO_RESET_WRITE_POINTER;
+ put_unaligned_be64(block, &cmd->cmnd[2]);
+
+ rq->timeout = SD_TIMEOUT;
+ cmd->sc_data_direction = DMA_NONE;
+ cmd->transfersize = 0;
+ cmd->allowed = 0;
+
+ return BLKPREP_OK;
+}
+
+int sd_zbc_setup_read_write(struct scsi_cmnd *cmd)
+{
+ struct request *rq = cmd->request;
+
+ /*
+ * Note: alignment of the read/write on logical blocks
+ * is done after this function returns in sd_setup_read_write.
+ */
+
+ /*
+ * Do not issue more than one write at a time per
+ * zone. This solves write ordering problems due to
+ * the unlocking of the request queue in the dispatch
+ * path in the non scsi-mq case. For scsi-mq, this
+ * also avoids potential write reordering when multiple
+ * threads running on different CPUs write to the same
+ * zone (with a synchronized sequential pattern).
+ */
+ if (req_op(rq) == REQ_OP_WRITE ||
+ req_op(rq) == REQ_OP_WRITE_SAME) {
+ struct scsi_disk *sdkp = scsi_disk(rq->rq_disk);
+ sector_t sector = blk_rq_pos(rq);
+ sector_t zone_sectors = sd_zbc_zone_sectors(sdkp);
+ sector_t zone_ofst = sector & (zone_sectors - 1);
+
+ /* Do not allow zone boundaries crossing */
+ if (zone_ofst + blk_rq_sectors(rq) > zone_sectors)
+ return BLKPREP_KILL;
+
+ if (!test_and_set_bit(sd_zbc_zone_no(sdkp, sector),
+ sdkp->zones_wlock))
+ return BLKPREP_DEFER;
+ }
+
+ return BLKPREP_OK;
+}
+
+void sd_zbc_complete(struct scsi_cmnd *cmd,
+ unsigned int good_bytes,
+ struct scsi_sense_hdr *sshdr)
+{
+ int result = cmd->result;
+ struct request *rq = cmd->request;
+ struct scsi_disk *sdkp = scsi_disk(rq->rq_disk);
+
+ switch (req_op(rq)) {
+ case REQ_OP_WRITE:
+ case REQ_OP_WRITE_SAME:
+
+ if (result &&
+ sshdr->sense_key == ILLEGAL_REQUEST &&
+ sshdr->asc == 0x21)
+ /*
+ * It is unlikely that retrying write requests failed
+ * with any kind of alignement error will result in
+ * success. So don't.
+ */
+ cmd->allowed = 0;
+
+ /* Fallthru */
+
+ case REQ_OP_ZONE_RESET:
+
+ /* Unlock the zone */
+ clear_bit_unlock(sd_zbc_zone_no(sdkp, blk_rq_pos(rq)),
+ sdkp->zones_wlock);
+ smp_mb__after_atomic();
+
+ if (result &&
+ sshdr->sense_key == ILLEGAL_REQUEST &&
+ sshdr->asc == 0x24)
+ /*
+ * INVALID FIELD IN CDB error: Reset of a conventional
+ * zone was attempted. Nothing to worry about,
+ * so be quiet about the error.
+ */
+ rq->cmd_flags |= REQ_QUIET;
+
+ break;
+
+ case REQ_OP_ZONE_REPORT:
+
+ if (!result)
+ sd_zbc_report_zones_complete(cmd, good_bytes);
+ break;
+
+ }
+}
+
+/**
+ * Read zoned block device characteristics (VPD page B6).
+ */
+static int sd_zbc_read_zoned_characteristics(struct scsi_disk *sdkp,
+ unsigned char *buf)
+{
+
+ if (scsi_get_vpd_page(sdkp->device, 0xb6, buf, 64)) {
+ sd_printk(KERN_NOTICE, sdkp,
+ "Unconstrained-read check failed\n");
+ return -ENODEV;
+ }
+
+ if (sdkp->device->type != TYPE_ZBC) {
+ /* Host-aware */
+ sdkp->urswrz = 1;
+ sdkp->zones_optimal_open = get_unaligned_be64(&buf[8]);
+ sdkp->zones_optimal_nonseq = get_unaligned_be64(&buf[12]);
+ sdkp->zones_max_open = 0;
+ } else {
+ /* Host-managed */
+ sdkp->urswrz = buf[4] & 1;
+ sdkp->zones_optimal_open = 0;
+ sdkp->zones_optimal_nonseq = 0;
+ sdkp->zones_max_open = get_unaligned_be64(&buf[16]);
+ }
+
+ return 0;
+}
+
+/**
+ * Check reported capacity.
+ */
+static int sd_zbc_check_capacity(struct scsi_disk *sdkp,
+ unsigned char *buf)
+{
+ sector_t lba;
+ int ret;
+
+ if (sdkp->rc_basis != 0)
+ return 0;
+
+ /* Do a report zone to get the maximum LBA to check capacity */
+ ret = sd_zbc_report_zones(sdkp, buf, SD_BUF_SIZE, 0);
+ if (ret)
+ return ret;
+
+ /* The max_lba field is the capacity of this device */
+ lba = get_unaligned_be64(&buf[8]);
+ if (lba + 1 == sdkp->capacity)
+ return 0;
+
+ if (sdkp->first_scan)
+ sd_printk(KERN_WARNING, sdkp,
+ "Changing capacity from %zu to max LBA+1 %llu\n",
+ sdkp->capacity,
+ (unsigned long long)lba + 1);
+ sdkp->capacity = lba + 1;
+
+ return 0;
+}
+
+#define SD_ZBC_BUF_SIZE 131072
+
+static int sd_zbc_check_zone_size(struct scsi_disk *sdkp)
+{
+ u64 zone_blocks;
+ sector_t block = 0;
+ unsigned char *buf;
+ unsigned char *rec;
+ unsigned int buf_len;
+ unsigned int list_length;
+ int ret;
+ u8 same;
+
+ sdkp->zone_blocks = 0;
+
+ /* Get a buffer */
+ buf = kmalloc(SD_ZBC_BUF_SIZE, GFP_KERNEL);
+ if (!buf)
+ return -ENOMEM;
+
+ /* Do a report zone to get the same field */
+ ret = sd_zbc_report_zones(sdkp, buf, SD_ZBC_BUF_SIZE, 0);
+ if (ret)
+ goto out;
+
+ same = buf[4] & 0x0f;
+ if (same > 0) {
+ rec = &buf[64];
+ zone_blocks = get_unaligned_be64(&rec[8]);
+ goto out;
+ }
+
+ /*
+ * Check the size of all zones: all zones must be of
+ * equal size, except the last zone which can be smaller
+ * than other zones.
+ */
+ do {
+
+ /* Parse REPORT ZONES header */
+ list_length = get_unaligned_be32(&buf[0]) + 64;
+ rec = buf + 64;
+ if (list_length < SD_ZBC_BUF_SIZE)
+ buf_len = list_length;
+ else
+ buf_len = SD_ZBC_BUF_SIZE;
+
+ /* Parse zone descriptors */
+ while (rec < buf + buf_len) {
+ zone_blocks = get_unaligned_be64(&rec[8]);
+ if (sdkp->zone_blocks == 0) {
+ sdkp->zone_blocks = zone_blocks;
+ } else if (zone_blocks != sdkp->zone_blocks &&
+ (block + zone_blocks < sdkp->capacity
+ || zone_blocks > sdkp->zone_blocks)) {
+ zone_blocks = 0;
+ goto out;
+ }
+ block += zone_blocks;
+ rec += 64;
+ }
+
+ if (block < sdkp->capacity) {
+ ret = sd_zbc_report_zones(sdkp, buf,
+ SD_ZBC_BUF_SIZE, block);
+ if (ret)
+ return ret;
+ }
+
+ } while (block < sdkp->capacity);
+
+ zone_blocks = sdkp->zone_blocks;
+
+out:
+ kfree(buf);
+
+ if (!zone_blocks) {
+ if (sdkp->first_scan)
+ sd_printk(KERN_NOTICE, sdkp,
+ "Devices with non constant zone "
+ "size are not supported\n");
+ return -ENODEV;
+ }
+
+ if (!is_power_of_2(zone_blocks)) {
+ if (sdkp->first_scan)
+ sd_printk(KERN_NOTICE, sdkp,
+ "Devices with non power of 2 zone "
+ "size are not supported\n");
+ return -ENODEV;
+ }
+
+ if (logical_to_sectors(sdkp->device, zone_blocks) > UINT_MAX) {
+ if (sdkp->first_scan)
+ sd_printk(KERN_NOTICE, sdkp,
+ "Zone size too large\n");
+ return -ENODEV;
+ }
+
+ sdkp->zone_blocks = zone_blocks;
+
+ return 0;
+}
+
+static int sd_zbc_setup(struct scsi_disk *sdkp)
+{
+
+ /* chunk_sectors indicates the zone size */
+ blk_queue_chunk_sectors(sdkp->disk->queue,
+ logical_to_sectors(sdkp->device, sdkp->zone_blocks));
+ sdkp->zone_shift = ilog2(sdkp->zone_blocks);
+ sdkp->nr_zones = sdkp->capacity >> sdkp->zone_shift;
+ if (sdkp->capacity & (sdkp->zone_blocks - 1))
+ sdkp->nr_zones++;
+
+ if (!sdkp->zones_wlock) {
+ sdkp->zones_wlock = kzalloc(BITS_TO_LONGS(sdkp->nr_zones),
+ GFP_KERNEL);
+ if (!sdkp->zones_wlock)
+ return -ENOMEM;
+ }
+
+ return 0;
+}
+
+int sd_zbc_read_zones(struct scsi_disk *sdkp,
+ unsigned char *buf)
+{
+ sector_t capacity;
+ int ret = 0;
+
+ if (!sd_is_zoned(sdkp))
+ /*
+ * Device managed or normal SCSI disk,
+ * no special handling required
+ */
+ return 0;
+
+
+ /* Get zoned block device characteristics */
+ ret = sd_zbc_read_zoned_characteristics(sdkp, buf);
+ if (ret)
+ goto err;
+
+ /*
+ * Check for unconstrained reads: host-managed devices with
+ * constrained reads (drives failing read after write pointer)
+ * are not supported.
+ */
+ if (!sdkp->urswrz) {
+ if (sdkp->first_scan)
+ sd_printk(KERN_NOTICE, sdkp,
+ "constrained reads devices are not supported\n");
+ ret = -ENODEV;
+ goto err;
+ }
+
+ /* Check capacity */
+ ret = sd_zbc_check_capacity(sdkp, buf);
+ if (ret)
+ goto err;
+ capacity = logical_to_sectors(sdkp->device, sdkp->capacity);
+
+ /*
+ * Check zone size: only devices with a constant zone size (except
+ * an eventual last runt zone) that is a power of 2 are supported.
+ */
+ ret = sd_zbc_check_zone_size(sdkp);
+ if (ret)
+ goto err;
+
+ /* The drive satisfies the kernel restrictions: set it up */
+ ret = sd_zbc_setup(sdkp);
+ if (ret)
+ goto err;
+
+ return 0;
+
+err:
+ sdkp->capacity = 0;
+
+ return ret;
+}
+
+void sd_zbc_remove(struct scsi_disk *sdkp)
+{
+ kfree(sdkp->zones_wlock);
+ sdkp->zones_wlock = NULL;
+}
+
+void sd_zbc_print_zones(struct scsi_disk *sdkp)
+{
+ if (!sd_is_zoned(sdkp) || !sdkp->capacity)
+ return;
+
+ if (sdkp->capacity & (sdkp->zone_blocks - 1))
+ sd_printk(KERN_NOTICE, sdkp,
+ "%u zones of %u logical blocks + 1 runt zone\n",
+ sdkp->nr_zones - 1,
+ sdkp->zone_blocks);
+ else
+ sd_printk(KERN_NOTICE, sdkp,
+ "%u zones of %u logical blocks\n",
+ sdkp->nr_zones,
+ sdkp->zone_blocks);
+}
diff --git a/include/scsi/scsi_proto.h b/include/scsi/scsi_proto.h
index d1defd1..6ba66e0 100644
--- a/include/scsi/scsi_proto.h
+++ b/include/scsi/scsi_proto.h
@@ -299,4 +299,21 @@ struct scsi_lun {
#define SCSI_ACCESS_STATE_MASK 0x0f
#define SCSI_ACCESS_STATE_PREFERRED 0x80

+/* Reporting options for REPORT ZONES */
+enum zbc_zone_reporting_options {
+ ZBC_ZONE_REPORTING_OPTION_ALL = 0,
+ ZBC_ZONE_REPORTING_OPTION_EMPTY,
+ ZBC_ZONE_REPORTING_OPTION_IMPLICIT_OPEN,
+ ZBC_ZONE_REPORTING_OPTION_EXPLICIT_OPEN,
+ ZBC_ZONE_REPORTING_OPTION_CLOSED,
+ ZBC_ZONE_REPORTING_OPTION_FULL,
+ ZBC_ZONE_REPORTING_OPTION_READONLY,
+ ZBC_ZONE_REPORTING_OPTION_OFFLINE,
+ ZBC_ZONE_REPORTING_OPTION_NEED_RESET_WP = 0x10,
+ ZBC_ZONE_REPORTING_OPTION_NON_SEQWRITE,
+ ZBC_ZONE_REPORTING_OPTION_NON_WP = 0x3f,
+};
+
+#define ZBC_REPORT_ZONE_PARTIAL 0x80
+
#endif /* _SCSI_PROTO_H_ */
--
2.9.3