[fio.git] / oslib / linux-blkzoned.c

/*
 * Copyright (C) 2020 Western Digital Corporation or its affiliates.
 *
 * This file is released under the GPL.
 */
#include <errno.h>
#include <string.h>
#include <stdlib.h>
#include <dirent.h>
#include <fcntl.h>
#include <sys/ioctl.h>
#include <sys/stat.h>
#include <unistd.h>

#include "file.h"
#include "fio.h"
#include "lib/pow2.h"
#include "log.h"
#include "oslib/asprintf.h"
#include "smalloc.h"
#include "verify.h"
#include "zbd_types.h"

#include <linux/blkzoned.h>

/*
 * If the uapi headers installed on the system lacks zone capacity support,
 * use our local versions. If the installed headers are recent enough to
 * support zone capacity, do not redefine any structs.
 */
#ifndef CONFIG_HAVE_REP_CAPACITY
#define BLK_ZONE_REP_CAPACITY	(1 << 0)

struct blk_zone_v2 {
	__u64	start;          /* Zone start sector */
	__u64	len;            /* Zone length in number of sectors */
	__u64	wp;             /* Zone write pointer position */
	__u8	type;           /* Zone type */
	__u8	cond;           /* Zone condition */
	__u8	non_seq;        /* Non-sequential write resources active */
	__u8	reset;          /* Reset write pointer recommended */
	__u8	resv[4];
	__u64	capacity;       /* Zone capacity in number of sectors */
	__u8	reserved[24];
};
#define blk_zone blk_zone_v2

struct blk_zone_report_v2 {
	__u64	sector;
	__u32	nr_zones;
	__u32	flags;
struct blk_zone zones[0];
};
#define blk_zone_report blk_zone_report_v2
#endif /* CONFIG_HAVE_REP_CAPACITY */

/*
 * Read up to 255 characters from the first line of a file. Strip the trailing
 * newline.
 */
static char *read_file(const char *path)
{
	char line[256], *p = line;
	FILE *f;

	f = fopen(path, "rb");
	if (!f)
		return NULL;
	if (!fgets(line, sizeof(line), f))
		line[0] = '\0';
	strsep(&p, "\n");
	fclose(f);

	return strdup(line);
}

int blkzoned_get_zoned_model(struct thread_data *td, struct fio_file *f,
			     enum zbd_zoned_model *model)
{
	const char *file_name = f->file_name;
	char *zoned_attr_path = NULL;
	char *model_str = NULL;
	struct stat statbuf;
	char *sys_devno_path = NULL;
	char *part_attr_path = NULL;
	char *part_str = NULL;
	char sys_path[PATH_MAX];
	ssize_t sz;
	char *delim = NULL;

	if (f->filetype != FIO_TYPE_BLOCK) {
		*model = ZBD_IGNORE;
		return 0;
	}

	*model = ZBD_NONE;

	if (stat(file_name, &statbuf) < 0)
		goto out;

	if (asprintf(&sys_devno_path, "/sys/dev/block/%d:%d",
		     major(statbuf.st_rdev), minor(statbuf.st_rdev)) < 0)
		goto out;

	sz = readlink(sys_devno_path, sys_path, sizeof(sys_path) - 1);
	if (sz < 0)
		goto out;
	sys_path[sz] = '\0';

	/*
	 * If the device is a partition device, cut the device name in the
	 * canonical sysfs path to obtain the sysfs path of the holder device.
	 *   e.g.:  /sys/devices/.../sda/sda1 -> /sys/devices/.../sda
	 */
	if (asprintf(&part_attr_path, "/sys/dev/block/%s/partition",
		     sys_path) < 0)
		goto out;
	part_str = read_file(part_attr_path);
	if (part_str && *part_str == '1') {
		delim = strrchr(sys_path, '/');
		if (!delim)
			goto out;
		*delim = '\0';
	}

	if (asprintf(&zoned_attr_path,
		     "/sys/dev/block/%s/queue/zoned", sys_path) < 0)
		goto out;

	model_str = read_file(zoned_attr_path);
	if (!model_str)
		goto out;
	dprint(FD_ZBD, "%s: zbd model string: %s\n", file_name, model_str);
	if (strcmp(model_str, "host-aware") == 0)
		*model = ZBD_HOST_AWARE;
	else if (strcmp(model_str, "host-managed") == 0)
		*model = ZBD_HOST_MANAGED;
out:
	free(model_str);
	free(zoned_attr_path);
	free(part_str);
	free(part_attr_path);
	free(sys_devno_path);
	return 0;
}

static uint64_t zone_capacity(struct blk_zone_report *hdr,
			      struct blk_zone *blkz)
{
	if (hdr->flags & BLK_ZONE_REP_CAPACITY)
		return blkz->capacity << 9;
	return blkz->len << 9;
}

int blkzoned_report_zones(struct thread_data *td, struct fio_file *f,
			  uint64_t offset, struct zbd_zone *zones,
			  unsigned int nr_zones)
{
	struct blk_zone_report *hdr = NULL;
	struct blk_zone *blkz;
	struct zbd_zone *z;
	unsigned int i;
	int fd = -1, ret;

	fd = open(f->file_name, O_RDONLY | O_LARGEFILE);
	if (fd < 0)
		return -errno;

	hdr = calloc(1, sizeof(struct blk_zone_report) +
			nr_zones * sizeof(struct blk_zone));
	if (!hdr) {
		ret = -ENOMEM;
		goto out;
	}

	hdr->nr_zones = nr_zones;
	hdr->sector = offset >> 9;
	ret = ioctl(fd, BLKREPORTZONE, hdr);
	if (ret) {
		ret = -errno;
		goto out;
	}

	nr_zones = hdr->nr_zones;
	blkz = (void *) hdr + sizeof(*hdr);
	z = &zones[0];
	for (i = 0; i < nr_zones; i++, z++, blkz++) {
		z->start = blkz->start << 9;
		z->wp = blkz->wp << 9;
		z->len = blkz->len << 9;
		z->capacity = zone_capacity(hdr, blkz);

		switch (blkz->type) {
		case BLK_ZONE_TYPE_CONVENTIONAL:
			z->type = ZBD_ZONE_TYPE_CNV;
			break;
		case BLK_ZONE_TYPE_SEQWRITE_REQ:
			z->type = ZBD_ZONE_TYPE_SWR;
			break;
		case BLK_ZONE_TYPE_SEQWRITE_PREF:
			z->type = ZBD_ZONE_TYPE_SWP;
			break;
		default:
			td_verror(td, errno, "invalid zone type");
			log_err("%s: invalid type for zone at sector %llu.\n",
				f->file_name, (unsigned long long)offset >> 9);
			ret = -EIO;
			goto out;
		}

		switch (blkz->cond) {
		case BLK_ZONE_COND_NOT_WP:
			z->cond = ZBD_ZONE_COND_NOT_WP;
			break;
		case BLK_ZONE_COND_EMPTY:
			z->cond = ZBD_ZONE_COND_EMPTY;
			break;
		case BLK_ZONE_COND_IMP_OPEN:
			z->cond = ZBD_ZONE_COND_IMP_OPEN;
			break;
		case BLK_ZONE_COND_EXP_OPEN:
			z->cond = ZBD_ZONE_COND_EXP_OPEN;
			break;
		case BLK_ZONE_COND_CLOSED:
			z->cond = ZBD_ZONE_COND_CLOSED;
			break;
		case BLK_ZONE_COND_FULL:
			z->cond = ZBD_ZONE_COND_FULL;
			break;
		case BLK_ZONE_COND_READONLY:
		case BLK_ZONE_COND_OFFLINE:
		default:
			/* Treat all these conditions as offline (don't use!) */
			z->cond = ZBD_ZONE_COND_OFFLINE;
			z->wp = z->start;
		}
	}

	ret = nr_zones;
out:
	free(hdr);
	close(fd);

	return ret;
}

int blkzoned_reset_wp(struct thread_data *td, struct fio_file *f,
		      uint64_t offset, uint64_t length)
{
	struct blk_zone_range zr = {
		.sector         = offset >> 9,
		.nr_sectors     = length >> 9,
	};
	int fd, ret = 0;

	/* If the file is not yet opened, open it for this function. */
	fd = f->fd;
	if (fd < 0) {
		fd = open(f->file_name, O_RDWR | O_LARGEFILE);
		if (fd < 0)
			return -errno;
	}

	if (ioctl(fd, BLKRESETZONE, &zr) < 0)
		ret = -errno;

	if (f->fd < 0)
		close(fd);

	return ret;
}
Commit	Line	Data
b7694961 DLM	1	/*
	2	* Copyright (C) 2020 Western Digital Corporation or its affiliates.
	3	*
	4	* This file is released under the GPL.
	5	*/
	6	#include <errno.h>
	7	#include <string.h>
	8	#include <stdlib.h>
	9	#include <dirent.h>
	10	#include <fcntl.h>
	11	#include <sys/ioctl.h>
	12	#include <sys/stat.h>
	13	#include <unistd.h>
	14
	15	#include "file.h"
	16	#include "fio.h"
	17	#include "lib/pow2.h"
	18	#include "log.h"
	19	#include "oslib/asprintf.h"
	20	#include "smalloc.h"
	21	#include "verify.h"
	22	#include "zbd_types.h"
	23
	24	#include <linux/blkzoned.h>
	25
6ee607ba NC	26	/*
	27	* If the uapi headers installed on the system lacks zone capacity support,
	28	* use our local versions. If the installed headers are recent enough to
	29	* support zone capacity, do not redefine any structs.
	30	*/
	31	#ifndef CONFIG_HAVE_REP_CAPACITY
	32	#define BLK_ZONE_REP_CAPACITY (1 << 0)
	33
	34	struct blk_zone_v2 {
	35	__u64 start; /* Zone start sector */
	36	__u64 len; /* Zone length in number of sectors */
	37	__u64 wp; /* Zone write pointer position */
	38	__u8 type; /* Zone type */
	39	__u8 cond; /* Zone condition */
	40	__u8 non_seq; /* Non-sequential write resources active */
	41	__u8 reset; /* Reset write pointer recommended */
	42	__u8 resv[4];
	43	__u64 capacity; /* Zone capacity in number of sectors */
	44	__u8 reserved[24];
	45	};
	46	#define blk_zone blk_zone_v2
	47
	48	struct blk_zone_report_v2 {
	49	__u64 sector;
	50	__u32 nr_zones;
	51	__u32 flags;
	52	struct blk_zone zones[0];
	53	};
	54	#define blk_zone_report blk_zone_report_v2
	55	#endif /* CONFIG_HAVE_REP_CAPACITY */
	56
b7694961 DLM	57	/*
	58	* Read up to 255 characters from the first line of a file. Strip the trailing
	59	* newline.
	60	*/
	61	static char read_file(const char path)
	62	{
	63	char line[256], *p = line;
	64	FILE *f;
	65
	66	f = fopen(path, "rb");
	67	if (!f)
	68	return NULL;
	69	if (!fgets(line, sizeof(line), f))
	70	line[0] = '\0';
	71	strsep(&p, "\n");
	72	fclose(f);
	73
	74	return strdup(line);
	75	}
	76
	77	int blkzoned_get_zoned_model(struct thread_data td, struct fio_file f,
	78	enum zbd_zoned_model *model)
	79	{
	80	const char *file_name = f->file_name;
	81	char *zoned_attr_path = NULL;
	82	char *model_str = NULL;
	83	struct stat statbuf;
	84	char *sys_devno_path = NULL;
	85	char *part_attr_path = NULL;
	86	char *part_str = NULL;
	87	char sys_path[PATH_MAX];
	88	ssize_t sz;
	89	char *delim = NULL;
	90
	91	if (f->filetype != FIO_TYPE_BLOCK) {
	92	*model = ZBD_IGNORE;
	93	return 0;
	94	}
	95
	96	*model = ZBD_NONE;
	97
	98	if (stat(file_name, &statbuf) < 0)
	99	goto out;
	100
	101	if (asprintf(&sys_devno_path, "/sys/dev/block/%d:%d",
	102	major(statbuf.st_rdev), minor(statbuf.st_rdev)) < 0)
	103	goto out;
	104
	105	sz = readlink(sys_devno_path, sys_path, sizeof(sys_path) - 1);
	106	if (sz < 0)
	107	goto out;
	108	sys_path[sz] = '\0';
	109
	110	/*
	111	* If the device is a partition device, cut the device name in the
	112	* canonical sysfs path to obtain the sysfs path of the holder device.
	113	* e.g.: /sys/devices/.../sda/sda1 -> /sys/devices/.../sda
	114	*/
	115	if (asprintf(&part_attr_path, "/sys/dev/block/%s/partition",
	116	sys_path) < 0)
	117	goto out;
	118	part_str = read_file(part_attr_path);
	119	if (part_str && *part_str == '1') {
	120	delim = strrchr(sys_path, '/');
121	if (!delim)
122	goto out;
123	*delim = '\0';
124	}
125
126	if (asprintf(&zoned_attr_path,
127	"/sys/dev/block/%s/queue/zoned", sys_path) < 0)
128	goto out;
129
130	model_str = read_file(zoned_attr_path);
131	if (!model_str)
132	goto out;
133	dprint(FD_ZBD, "%s: zbd model string: %s\n", file_name, model_str);
134	if (strcmp(model_str, "host-aware") == 0)
135	*model = ZBD_HOST_AWARE;
136	else if (strcmp(model_str, "host-managed") == 0)
137	*model = ZBD_HOST_MANAGED;
138	out:
139	free(model_str);
140	free(zoned_attr_path);
141	free(part_str);
142	free(part_attr_path);
143	free(sys_devno_path);
144	return 0;
145	}
146
236d23a8 SK	147	static uint64_t zone_capacity(struct blk_zone_report *hdr,
	148	struct blk_zone *blkz)
	149	{
236d23a8 SK	150	if (hdr->flags & BLK_ZONE_REP_CAPACITY)
236d23a8 SK	151	return blkz->capacity << 9;
236d23a8 SK	152	return blkz->len << 9;
	153	}
	154
b7694961 DLM	155	int blkzoned_report_zones(struct thread_data td, struct fio_file f,
	156	uint64_t offset, struct zbd_zone *zones,
	157	unsigned int nr_zones)
	158	{
	159	struct blk_zone_report *hdr = NULL;
	160	struct blk_zone *blkz;
	161	struct zbd_zone *z;
	162	unsigned int i;
	163	int fd = -1, ret;
	164
	165	fd = open(f->file_name, O_RDONLY \| O_LARGEFILE);
	166	if (fd < 0)
	167	return -errno;
	168
	169	hdr = calloc(1, sizeof(struct blk_zone_report) +
	170	nr_zones * sizeof(struct blk_zone));
	171	if (!hdr) {
	172	ret = -ENOMEM;
	173	goto out;
	174	}
	175
	176	hdr->nr_zones = nr_zones;
	177	hdr->sector = offset >> 9;
	178	ret = ioctl(fd, BLKREPORTZONE, hdr);
	179	if (ret) {
	180	ret = -errno;
	181	goto out;
	182	}
	183
	184	nr_zones = hdr->nr_zones;
f8779edf	185	blkz = (void ) hdr + sizeof(hdr);
b7694961 DLM	186	z = &zones[0];
	187	for (i = 0; i < nr_zones; i++, z++, blkz++) {
	188	z->start = blkz->start << 9;
	189	z->wp = blkz->wp << 9;
	190	z->len = blkz->len << 9;
236d23a8	191	z->capacity = zone_capacity(hdr, blkz);
b7694961 DLM	192
	193	switch (blkz->type) {
	194	case BLK_ZONE_TYPE_CONVENTIONAL:
	195	z->type = ZBD_ZONE_TYPE_CNV;
	196	break;
	197	case BLK_ZONE_TYPE_SEQWRITE_REQ:
	198	z->type = ZBD_ZONE_TYPE_SWR;
	199	break;
	200	case BLK_ZONE_TYPE_SEQWRITE_PREF:
	201	z->type = ZBD_ZONE_TYPE_SWP;
	202	break;
	203	default:
	204	td_verror(td, errno, "invalid zone type");
	205	log_err("%s: invalid type for zone at sector %llu.\n",
	206	f->file_name, (unsigned long long)offset >> 9);
	207	ret = -EIO;
	208	goto out;
	209	}
	210
	211	switch (blkz->cond) {
	212	case BLK_ZONE_COND_NOT_WP:
	213	z->cond = ZBD_ZONE_COND_NOT_WP;
	214	break;
	215	case BLK_ZONE_COND_EMPTY:
	216	z->cond = ZBD_ZONE_COND_EMPTY;
	217	break;
	218	case BLK_ZONE_COND_IMP_OPEN:
	219	z->cond = ZBD_ZONE_COND_IMP_OPEN;
	220	break;
	221	case BLK_ZONE_COND_EXP_OPEN:
	222	z->cond = ZBD_ZONE_COND_EXP_OPEN;
	223	break;
	224	case BLK_ZONE_COND_CLOSED:
	225	z->cond = ZBD_ZONE_COND_CLOSED;
	226	break;
	227	case BLK_ZONE_COND_FULL:
	228	z->cond = ZBD_ZONE_COND_FULL;
	229	break;
	230	case BLK_ZONE_COND_READONLY:
	231	case BLK_ZONE_COND_OFFLINE:
	232	default:
	233	/* Treat all these conditions as offline (don't use!) */
	234	z->cond = ZBD_ZONE_COND_OFFLINE;
59b67452	235	z->wp = z->start;
b7694961 DLM	236	}
	237	}
	238
	239	ret = nr_zones;
	240	out:
	241	free(hdr);
	242	close(fd);
	243
	244	return ret;
	245	}
	246
	247	int blkzoned_reset_wp(struct thread_data td, struct fio_file f,
	248	uint64_t offset, uint64_t length)
	249	{
	250	struct blk_zone_range zr = {
	251	.sector = offset >> 9,
	252	.nr_sectors = length >> 9,
	253	};
b4824992 SK	254	int fd, ret = 0;
	255
	256	/* If the file is not yet opened, open it for this function. */
	257	fd = f->fd;
	258	if (fd < 0) {
	259	fd = open(f->file_name, O_RDWR \| O_LARGEFILE);
	260	if (fd < 0)
	261	return -errno;
	262	}
b7694961	263
b4824992 SK	264	if (ioctl(fd, BLKRESETZONE, &zr) < 0)
b4824992 SK	265	ret = -errno;
b7694961	266
b4824992 SK	267	if (f->fd < 0)
	268	close(fd);
	269
	270	return ret;
b7694961	271	}