*/
#include "qemu-common.h"
-#include "nbd.h"
-#include "module.h"
+#include "block/nbd.h"
+#include "qemu/uri.h"
+#include "block/block_int.h"
+#include "qemu/module.h"
+#include "qemu/sockets.h"
#include <sys/types.h>
#include <unistd.h>
#define EN_OPTSTR ":exportname="
+/* #define DEBUG_NBD */
+
+#if defined(DEBUG_NBD)
+#define logout(fmt, ...) \
+ fprintf(stderr, "nbd\t%-24s" fmt, __func__, ##__VA_ARGS__)
+#else
+#define logout(fmt, ...) ((void)0)
+#endif
+
+#define MAX_NBD_REQUESTS 16
+#define HANDLE_TO_INDEX(bs, handle) ((handle) ^ ((uint64_t)(intptr_t)bs))
+#define INDEX_TO_HANDLE(bs, index) ((index) ^ ((uint64_t)(intptr_t)bs))
+
typedef struct BDRVNBDState {
int sock;
+ uint32_t nbdflags;
off_t size;
size_t blocksize;
+
+ CoMutex send_mutex;
+ CoMutex free_sema;
+ Coroutine *send_coroutine;
+ int in_flight;
+
+ Coroutine *recv_coroutine[MAX_NBD_REQUESTS];
+ struct nbd_reply reply;
+
+ int is_unix;
+ char *host_spec;
+ char *export_name; /* An NBD server may export several devices */
} BDRVNBDState;
-static int nbd_open(BlockDriverState *bs, const char* filename, int flags)
+static int nbd_parse_uri(BDRVNBDState *s, const char *filename)
{
- BDRVNBDState *s = bs->opaque;
- uint32_t nbdflags;
+ URI *uri;
+ const char *p;
+ QueryParams *qp = NULL;
+ int ret = 0;
+
+ uri = uri_parse(filename);
+ if (!uri) {
+ return -EINVAL;
+ }
+ /* transport */
+ if (!strcmp(uri->scheme, "nbd")) {
+ s->is_unix = false;
+ } else if (!strcmp(uri->scheme, "nbd+tcp")) {
+ s->is_unix = false;
+ } else if (!strcmp(uri->scheme, "nbd+unix")) {
+ s->is_unix = true;
+ } else {
+ ret = -EINVAL;
+ goto out;
+ }
+
+ p = uri->path ? uri->path : "/";
+ p += strspn(p, "/");
+ if (p[0]) {
+ s->export_name = g_strdup(p);
+ }
+
+ qp = query_params_parse(uri->query);
+ if (qp->n > 1 || (s->is_unix && !qp->n) || (!s->is_unix && qp->n)) {
+ ret = -EINVAL;
+ goto out;
+ }
+
+ if (s->is_unix) {
+ /* nbd+unix:///export?socket=path */
+ if (uri->server || uri->port || strcmp(qp->p[0].name, "socket")) {
+ ret = -EINVAL;
+ goto out;
+ }
+ s->host_spec = g_strdup(qp->p[0].value);
+ } else {
+ /* nbd[+tcp]://host:port/export */
+ if (!uri->server) {
+ ret = -EINVAL;
+ goto out;
+ }
+ if (!uri->port) {
+ uri->port = NBD_DEFAULT_PORT;
+ }
+ s->host_spec = g_strdup_printf("%s:%d", uri->server, uri->port);
+ }
+
+out:
+ if (qp) {
+ query_params_free(qp);
+ }
+ uri_free(uri);
+ return ret;
+}
+
+static int nbd_config(BDRVNBDState *s, const char *filename)
+{
char *file;
- char *name;
- const char *host;
+ char *export_name;
+ const char *host_spec;
const char *unixpath;
- int sock;
- off_t size;
- size_t blocksize;
- int ret;
int err = -EINVAL;
- file = qemu_strdup(filename);
+ if (strstr(filename, "://")) {
+ return nbd_parse_uri(s, filename);
+ }
+
+ file = g_strdup(filename);
- name = strstr(file, EN_OPTSTR);
- if (name) {
- if (name[strlen(EN_OPTSTR)] == 0) {
+ export_name = strstr(file, EN_OPTSTR);
+ if (export_name) {
+ if (export_name[strlen(EN_OPTSTR)] == 0) {
goto out;
}
- name[0] = 0;
- name += strlen(EN_OPTSTR);
+ export_name[0] = 0; /* truncate 'file' */
+ export_name += strlen(EN_OPTSTR);
+ s->export_name = g_strdup(export_name);
}
- if (!strstart(file, "nbd:", &host)) {
+ /* extract the host_spec - fail if it's not nbd:... */
+ if (!strstart(file, "nbd:", &host_spec)) {
goto out;
}
- if (strstart(host, "unix:", &unixpath)) {
+ /* are we a UNIX or TCP socket? */
+ if (strstart(host_spec, "unix:", &unixpath)) {
+ s->is_unix = true;
+ s->host_spec = g_strdup(unixpath);
+ } else {
+ s->is_unix = false;
+ s->host_spec = g_strdup(host_spec);
+ }
- if (unixpath[0] != '/') {
- goto out;
+ err = 0;
+
+out:
+ g_free(file);
+ if (err != 0) {
+ g_free(s->export_name);
+ g_free(s->host_spec);
+ }
+ return err;
+}
+
+static void nbd_coroutine_start(BDRVNBDState *s, struct nbd_request *request)
+{
+ int i;
+
+ /* Poor man semaphore. The free_sema is locked when no other request
+ * can be accepted, and unlocked after receiving one reply. */
+ if (s->in_flight >= MAX_NBD_REQUESTS - 1) {
+ qemu_co_mutex_lock(&s->free_sema);
+ assert(s->in_flight < MAX_NBD_REQUESTS);
+ }
+ s->in_flight++;
+
+ for (i = 0; i < MAX_NBD_REQUESTS; i++) {
+ if (s->recv_coroutine[i] == NULL) {
+ s->recv_coroutine[i] = qemu_coroutine_self();
+ break;
}
+ }
+
+ assert(i < MAX_NBD_REQUESTS);
+ request->handle = INDEX_TO_HANDLE(s, i);
+}
- sock = unix_socket_outgoing(unixpath);
+static int nbd_have_request(void *opaque)
+{
+ BDRVNBDState *s = opaque;
- } else {
- uint16_t port = NBD_DEFAULT_PORT;
- char *p, *r;
- char hostname[128];
+ return s->in_flight > 0;
+}
+
+static void nbd_reply_ready(void *opaque)
+{
+ BDRVNBDState *s = opaque;
+ uint64_t i;
+ int ret;
+
+ if (s->reply.handle == 0) {
+ /* No reply already in flight. Fetch a header. It is possible
+ * that another thread has done the same thing in parallel, so
+ * the socket is not readable anymore.
+ */
+ ret = nbd_receive_reply(s->sock, &s->reply);
+ if (ret == -EAGAIN) {
+ return;
+ }
+ if (ret < 0) {
+ s->reply.handle = 0;
+ goto fail;
+ }
+ }
- pstrcpy(hostname, 128, host);
+ /* There's no need for a mutex on the receive side, because the
+ * handler acts as a synchronization point and ensures that only
+ * one coroutine is called until the reply finishes. */
+ i = HANDLE_TO_INDEX(s, s->reply.handle);
+ if (i >= MAX_NBD_REQUESTS) {
+ goto fail;
+ }
- p = strchr(hostname, ':');
- if (p != NULL) {
- *p = '\0';
- p++;
+ if (s->recv_coroutine[i]) {
+ qemu_coroutine_enter(s->recv_coroutine[i], NULL);
+ return;
+ }
- port = strtol(p, &r, 0);
- if (r == p) {
- goto out;
+fail:
+ for (i = 0; i < MAX_NBD_REQUESTS; i++) {
+ if (s->recv_coroutine[i]) {
+ qemu_coroutine_enter(s->recv_coroutine[i], NULL);
+ }
+ }
+}
+
+static void nbd_restart_write(void *opaque)
+{
+ BDRVNBDState *s = opaque;
+ qemu_coroutine_enter(s->send_coroutine, NULL);
+}
+
+static int nbd_co_send_request(BDRVNBDState *s, struct nbd_request *request,
+ QEMUIOVector *qiov, int offset)
+{
+ int rc, ret;
+
+ qemu_co_mutex_lock(&s->send_mutex);
+ s->send_coroutine = qemu_coroutine_self();
+ qemu_aio_set_fd_handler(s->sock, nbd_reply_ready, nbd_restart_write,
+ nbd_have_request, s);
+ rc = nbd_send_request(s->sock, request);
+ if (rc >= 0 && qiov) {
+ ret = qemu_co_sendv(s->sock, qiov->iov, qiov->niov,
+ offset, request->len);
+ if (ret != request->len) {
+ return -EIO;
+ }
+ }
+ qemu_aio_set_fd_handler(s->sock, nbd_reply_ready, NULL,
+ nbd_have_request, s);
+ s->send_coroutine = NULL;
+ qemu_co_mutex_unlock(&s->send_mutex);
+ return rc;
+}
+
+static void nbd_co_receive_reply(BDRVNBDState *s, struct nbd_request *request,
+ struct nbd_reply *reply,
+ QEMUIOVector *qiov, int offset)
+{
+ int ret;
+
+ /* Wait until we're woken up by the read handler. TODO: perhaps
+ * peek at the next reply and avoid yielding if it's ours? */
+ qemu_coroutine_yield();
+ *reply = s->reply;
+ if (reply->handle != request->handle) {
+ reply->error = EIO;
+ } else {
+ if (qiov && reply->error == 0) {
+ ret = qemu_co_recvv(s->sock, qiov->iov, qiov->niov,
+ offset, request->len);
+ if (ret != request->len) {
+ reply->error = EIO;
}
}
- sock = tcp_socket_outgoing(hostname, port);
+ /* Tell the read handler to read another header. */
+ s->reply.handle = 0;
}
+}
- if (sock == -1) {
- err = -errno;
- goto out;
+static void nbd_coroutine_end(BDRVNBDState *s, struct nbd_request *request)
+{
+ int i = HANDLE_TO_INDEX(s, request->handle);
+ s->recv_coroutine[i] = NULL;
+ if (s->in_flight-- == MAX_NBD_REQUESTS) {
+ qemu_co_mutex_unlock(&s->free_sema);
}
+}
- ret = nbd_receive_negotiate(sock, name, &nbdflags, &size, &blocksize);
- if (ret == -1) {
- err = -errno;
- goto out;
+static int nbd_establish_connection(BlockDriverState *bs)
+{
+ BDRVNBDState *s = bs->opaque;
+ int sock;
+ int ret;
+ off_t size;
+ size_t blocksize;
+
+ if (s->is_unix) {
+ sock = unix_socket_outgoing(s->host_spec);
+ } else {
+ sock = tcp_socket_outgoing_spec(s->host_spec);
+ }
+
+ /* Failed to establish connection */
+ if (sock < 0) {
+ logout("Failed to establish connection to NBD server\n");
+ return -errno;
+ }
+
+ /* NBD handshake */
+ ret = nbd_receive_negotiate(sock, s->export_name, &s->nbdflags, &size,
+ &blocksize);
+ if (ret < 0) {
+ logout("Failed to negotiate with the NBD server\n");
+ closesocket(sock);
+ return ret;
}
+ /* Now that we're connected, set the socket to be non-blocking and
+ * kick the reply mechanism. */
+ socket_set_nonblock(sock);
+ qemu_aio_set_fd_handler(sock, nbd_reply_ready, NULL,
+ nbd_have_request, s);
+
s->sock = sock;
s->size = size;
s->blocksize = blocksize;
- err = 0;
-out:
- qemu_free(file);
- return err;
+ logout("Established connection with NBD server\n");
+ return 0;
}
-static int nbd_read(BlockDriverState *bs, int64_t sector_num,
- uint8_t *buf, int nb_sectors)
+static void nbd_teardown_connection(BlockDriverState *bs)
{
BDRVNBDState *s = bs->opaque;
struct nbd_request request;
- struct nbd_reply reply;
- request.type = NBD_CMD_READ;
- request.handle = (uint64_t)(intptr_t)bs;
- request.from = sector_num * 512;;
- request.len = nb_sectors * 512;
+ request.type = NBD_CMD_DISC;
+ request.from = 0;
+ request.len = 0;
+ nbd_send_request(s->sock, &request);
- if (nbd_send_request(s->sock, &request) == -1)
- return -errno;
+ qemu_aio_set_fd_handler(s->sock, NULL, NULL, NULL, NULL);
+ closesocket(s->sock);
+}
- if (nbd_receive_reply(s->sock, &reply) == -1)
- return -errno;
+static int nbd_open(BlockDriverState *bs, const char* filename, int flags)
+{
+ BDRVNBDState *s = bs->opaque;
+ int result;
- if (reply.error !=0)
- return -reply.error;
+ qemu_co_mutex_init(&s->send_mutex);
+ qemu_co_mutex_init(&s->free_sema);
- if (reply.handle != request.handle)
- return -EIO;
+ /* Pop the config into our state object. Exit if invalid. */
+ result = nbd_config(s, filename);
+ if (result != 0) {
+ return result;
+ }
- if (nbd_wr_sync(s->sock, buf, request.len, 1) != request.len)
- return -EIO;
+ /* establish TCP connection, return error if it fails
+ * TODO: Configurable retry-until-timeout behaviour.
+ */
+ result = nbd_establish_connection(bs);
- return 0;
+ return result;
}
-static int nbd_write(BlockDriverState *bs, int64_t sector_num,
- const uint8_t *buf, int nb_sectors)
+static int nbd_co_readv_1(BlockDriverState *bs, int64_t sector_num,
+ int nb_sectors, QEMUIOVector *qiov,
+ int offset)
{
BDRVNBDState *s = bs->opaque;
struct nbd_request request;
struct nbd_reply reply;
+ ssize_t ret;
- request.type = NBD_CMD_WRITE;
- request.handle = (uint64_t)(intptr_t)bs;
- request.from = sector_num * 512;;
+ request.type = NBD_CMD_READ;
+ request.from = sector_num * 512;
request.len = nb_sectors * 512;
- if (nbd_send_request(s->sock, &request) == -1)
- return -errno;
+ nbd_coroutine_start(s, &request);
+ ret = nbd_co_send_request(s, &request, NULL, 0);
+ if (ret < 0) {
+ reply.error = -ret;
+ } else {
+ nbd_co_receive_reply(s, &request, &reply, qiov, offset);
+ }
+ nbd_coroutine_end(s, &request);
+ return -reply.error;
- if (nbd_wr_sync(s->sock, (uint8_t*)buf, request.len, 0) != request.len)
- return -EIO;
+}
- if (nbd_receive_reply(s->sock, &reply) == -1)
- return -errno;
+static int nbd_co_writev_1(BlockDriverState *bs, int64_t sector_num,
+ int nb_sectors, QEMUIOVector *qiov,
+ int offset)
+{
+ BDRVNBDState *s = bs->opaque;
+ struct nbd_request request;
+ struct nbd_reply reply;
+ ssize_t ret;
- if (reply.error !=0)
- return -reply.error;
+ request.type = NBD_CMD_WRITE;
+ if (!bdrv_enable_write_cache(bs) && (s->nbdflags & NBD_FLAG_SEND_FUA)) {
+ request.type |= NBD_CMD_FLAG_FUA;
+ }
- if (reply.handle != request.handle)
- return -EIO;
+ request.from = sector_num * 512;
+ request.len = nb_sectors * 512;
- return 0;
+ nbd_coroutine_start(s, &request);
+ ret = nbd_co_send_request(s, &request, qiov, offset);
+ if (ret < 0) {
+ reply.error = -ret;
+ } else {
+ nbd_co_receive_reply(s, &request, &reply, NULL, 0);
+ }
+ nbd_coroutine_end(s, &request);
+ return -reply.error;
}
-static void nbd_close(BlockDriverState *bs)
+/* qemu-nbd has a limit of slightly less than 1M per request. Try to
+ * remain aligned to 4K. */
+#define NBD_MAX_SECTORS 2040
+
+static int nbd_co_readv(BlockDriverState *bs, int64_t sector_num,
+ int nb_sectors, QEMUIOVector *qiov)
+{
+ int offset = 0;
+ int ret;
+ while (nb_sectors > NBD_MAX_SECTORS) {
+ ret = nbd_co_readv_1(bs, sector_num, NBD_MAX_SECTORS, qiov, offset);
+ if (ret < 0) {
+ return ret;
+ }
+ offset += NBD_MAX_SECTORS * 512;
+ sector_num += NBD_MAX_SECTORS;
+ nb_sectors -= NBD_MAX_SECTORS;
+ }
+ return nbd_co_readv_1(bs, sector_num, nb_sectors, qiov, offset);
+}
+
+static int nbd_co_writev(BlockDriverState *bs, int64_t sector_num,
+ int nb_sectors, QEMUIOVector *qiov)
+{
+ int offset = 0;
+ int ret;
+ while (nb_sectors > NBD_MAX_SECTORS) {
+ ret = nbd_co_writev_1(bs, sector_num, NBD_MAX_SECTORS, qiov, offset);
+ if (ret < 0) {
+ return ret;
+ }
+ offset += NBD_MAX_SECTORS * 512;
+ sector_num += NBD_MAX_SECTORS;
+ nb_sectors -= NBD_MAX_SECTORS;
+ }
+ return nbd_co_writev_1(bs, sector_num, nb_sectors, qiov, offset);
+}
+
+static int nbd_co_flush(BlockDriverState *bs)
{
BDRVNBDState *s = bs->opaque;
struct nbd_request request;
+ struct nbd_reply reply;
+ ssize_t ret;
+
+ if (!(s->nbdflags & NBD_FLAG_SEND_FLUSH)) {
+ return 0;
+ }
+
+ request.type = NBD_CMD_FLUSH;
+ if (s->nbdflags & NBD_FLAG_SEND_FUA) {
+ request.type |= NBD_CMD_FLAG_FUA;
+ }
- request.type = NBD_CMD_DISC;
- request.handle = (uint64_t)(intptr_t)bs;
request.from = 0;
request.len = 0;
- nbd_send_request(s->sock, &request);
- close(s->sock);
+ nbd_coroutine_start(s, &request);
+ ret = nbd_co_send_request(s, &request, NULL, 0);
+ if (ret < 0) {
+ reply.error = -ret;
+ } else {
+ nbd_co_receive_reply(s, &request, &reply, NULL, 0);
+ }
+ nbd_coroutine_end(s, &request);
+ return -reply.error;
+}
+
+static int nbd_co_discard(BlockDriverState *bs, int64_t sector_num,
+ int nb_sectors)
+{
+ BDRVNBDState *s = bs->opaque;
+ struct nbd_request request;
+ struct nbd_reply reply;
+ ssize_t ret;
+
+ if (!(s->nbdflags & NBD_FLAG_SEND_TRIM)) {
+ return 0;
+ }
+ request.type = NBD_CMD_TRIM;
+ request.from = sector_num * 512;;
+ request.len = nb_sectors * 512;
+
+ nbd_coroutine_start(s, &request);
+ ret = nbd_co_send_request(s, &request, NULL, 0);
+ if (ret < 0) {
+ reply.error = -ret;
+ } else {
+ nbd_co_receive_reply(s, &request, &reply, NULL, 0);
+ }
+ nbd_coroutine_end(s, &request);
+ return -reply.error;
+}
+
+static void nbd_close(BlockDriverState *bs)
+{
+ BDRVNBDState *s = bs->opaque;
+ g_free(s->export_name);
+ g_free(s->host_spec);
+
+ nbd_teardown_connection(bs);
}
static int64_t nbd_getlength(BlockDriverState *bs)
}
static BlockDriver bdrv_nbd = {
- .format_name = "nbd",
- .instance_size = sizeof(BDRVNBDState),
- .bdrv_file_open = nbd_open,
- .bdrv_read = nbd_read,
- .bdrv_write = nbd_write,
- .bdrv_close = nbd_close,
- .bdrv_getlength = nbd_getlength,
- .protocol_name = "nbd",
+ .format_name = "nbd",
+ .protocol_name = "nbd",
+ .instance_size = sizeof(BDRVNBDState),
+ .bdrv_file_open = nbd_open,
+ .bdrv_co_readv = nbd_co_readv,
+ .bdrv_co_writev = nbd_co_writev,
+ .bdrv_close = nbd_close,
+ .bdrv_co_flush_to_os = nbd_co_flush,
+ .bdrv_co_discard = nbd_co_discard,
+ .bdrv_getlength = nbd_getlength,
+};
+
+static BlockDriver bdrv_nbd_tcp = {
+ .format_name = "nbd",
+ .protocol_name = "nbd+tcp",
+ .instance_size = sizeof(BDRVNBDState),
+ .bdrv_file_open = nbd_open,
+ .bdrv_co_readv = nbd_co_readv,
+ .bdrv_co_writev = nbd_co_writev,
+ .bdrv_close = nbd_close,
+ .bdrv_co_flush_to_os = nbd_co_flush,
+ .bdrv_co_discard = nbd_co_discard,
+ .bdrv_getlength = nbd_getlength,
+};
+
+static BlockDriver bdrv_nbd_unix = {
+ .format_name = "nbd",
+ .protocol_name = "nbd+unix",
+ .instance_size = sizeof(BDRVNBDState),
+ .bdrv_file_open = nbd_open,
+ .bdrv_co_readv = nbd_co_readv,
+ .bdrv_co_writev = nbd_co_writev,
+ .bdrv_close = nbd_close,
+ .bdrv_co_flush_to_os = nbd_co_flush,
+ .bdrv_co_discard = nbd_co_discard,
+ .bdrv_getlength = nbd_getlength,
};
static void bdrv_nbd_init(void)
{
bdrv_register(&bdrv_nbd);
+ bdrv_register(&bdrv_nbd_tcp);
+ bdrv_register(&bdrv_nbd_unix);
}
block_init(bdrv_nbd_init);
projects / qemu.git / blobdiff