/*
+ * Copyright (C) 2016 Red Hat, Inc.
* Copyright (C) 2005 Anthony Liguori <anthony@codemonkey.ws>
*
* Network Block Device Server Side
* along with this program; if not, see <http://www.gnu.org/licenses/>.
*/
+#include "qemu/osdep.h"
+#include "qapi/error.h"
#include "nbd-internal.h"
static int system_errno_to_nbd_errno(int err)
case 0:
return NBD_SUCCESS;
case EPERM:
+ case EROFS:
return NBD_EPERM;
case EIO:
return NBD_EIO;
/* Definitions for opaque data types */
-typedef struct NBDRequest NBDRequest;
+typedef struct NBDRequestData NBDRequestData;
-struct NBDRequest {
- QSIMPLEQ_ENTRY(NBDRequest) entry;
+struct NBDRequestData {
+ QSIMPLEQ_ENTRY(NBDRequestData) entry;
NBDClient *client;
uint8_t *data;
+ bool complete;
};
struct NBDExport {
BlockBackend *blk;
char *name;
+ char *description;
off_t dev_offset;
off_t size;
- uint32_t nbdflags;
+ uint16_t nbdflags;
QTAILQ_HEAD(, NBDClient) clients;
QTAILQ_ENTRY(NBDExport) next;
AioContext *ctx;
+
+ BlockBackend *eject_notifier_blk;
+ Notifier eject_notifier;
};
static QTAILQ_HEAD(, NBDExport) exports = QTAILQ_HEAD_INITIALIZER(exports);
int refcount;
void (*close)(NBDClient *client);
+ bool no_zeroes;
NBDExport *exp;
- int sock;
+ QCryptoTLSCreds *tlscreds;
+ char *tlsaclname;
+ QIOChannelSocket *sioc; /* The underlying data channel */
+ QIOChannel *ioc; /* The current I/O channel which may differ (eg TLS) */
Coroutine *recv_coroutine;
static void nbd_unset_handlers(NBDClient *client);
static void nbd_update_can_read(NBDClient *client);
-static ssize_t drop_sync(int fd, size_t size)
+static gboolean nbd_negotiate_continue(QIOChannel *ioc,
+ GIOCondition condition,
+ void *opaque)
+{
+ qemu_coroutine_enter(opaque);
+ return TRUE;
+}
+
+static ssize_t nbd_negotiate_read(QIOChannel *ioc, void *buffer, size_t size)
+{
+ ssize_t ret;
+ guint watch;
+
+ assert(qemu_in_coroutine());
+ /* Negotiation are always in main loop. */
+ watch = qio_channel_add_watch(ioc,
+ G_IO_IN,
+ nbd_negotiate_continue,
+ qemu_coroutine_self(),
+ NULL);
+ ret = read_sync(ioc, buffer, size);
+ g_source_remove(watch);
+ return ret;
+
+}
+
+static ssize_t nbd_negotiate_write(QIOChannel *ioc, const void *buffer,
+ size_t size)
+{
+ ssize_t ret;
+ guint watch;
+
+ assert(qemu_in_coroutine());
+ /* Negotiation are always in main loop. */
+ watch = qio_channel_add_watch(ioc,
+ G_IO_OUT,
+ nbd_negotiate_continue,
+ qemu_coroutine_self(),
+ NULL);
+ ret = write_sync(ioc, buffer, size);
+ g_source_remove(watch);
+ return ret;
+}
+
+static ssize_t nbd_negotiate_drop_sync(QIOChannel *ioc, size_t size)
{
ssize_t ret, dropped = size;
uint8_t *buffer = g_malloc(MIN(65536, size));
while (size > 0) {
- ret = read_sync(fd, buffer, MIN(65536, size));
+ ret = nbd_negotiate_read(ioc, buffer, MIN(65536, size));
if (ret < 0) {
g_free(buffer);
return ret;
*/
-static int nbd_send_rep(int csock, uint32_t type, uint32_t opt)
+/* Send a reply header, including length, but no payload.
+ * Return -errno on error, 0 on success. */
+static int nbd_negotiate_send_rep_len(QIOChannel *ioc, uint32_t type,
+ uint32_t opt, uint32_t len)
{
uint64_t magic;
- uint32_t len;
+
+ TRACE("Reply opt=%" PRIx32 " type=%" PRIx32 " len=%" PRIu32,
+ type, opt, len);
magic = cpu_to_be64(NBD_REP_MAGIC);
- if (write_sync(csock, &magic, sizeof(magic)) != sizeof(magic)) {
+ if (nbd_negotiate_write(ioc, &magic, sizeof(magic)) != sizeof(magic)) {
LOG("write failed (rep magic)");
return -EINVAL;
}
opt = cpu_to_be32(opt);
- if (write_sync(csock, &opt, sizeof(opt)) != sizeof(opt)) {
+ if (nbd_negotiate_write(ioc, &opt, sizeof(opt)) != sizeof(opt)) {
LOG("write failed (rep opt)");
return -EINVAL;
}
type = cpu_to_be32(type);
- if (write_sync(csock, &type, sizeof(type)) != sizeof(type)) {
+ if (nbd_negotiate_write(ioc, &type, sizeof(type)) != sizeof(type)) {
LOG("write failed (rep type)");
return -EINVAL;
}
- len = cpu_to_be32(0);
- if (write_sync(csock, &len, sizeof(len)) != sizeof(len)) {
+ len = cpu_to_be32(len);
+ if (nbd_negotiate_write(ioc, &len, sizeof(len)) != sizeof(len)) {
LOG("write failed (rep data length)");
return -EINVAL;
}
return 0;
}
-static int nbd_send_rep_list(int csock, NBDExport *exp)
+/* Send a reply header with default 0 length.
+ * Return -errno on error, 0 on success. */
+static int nbd_negotiate_send_rep(QIOChannel *ioc, uint32_t type, uint32_t opt)
{
- uint64_t magic, name_len;
- uint32_t opt, type, len;
+ return nbd_negotiate_send_rep_len(ioc, type, opt, 0);
+}
- name_len = strlen(exp->name);
- magic = cpu_to_be64(NBD_REP_MAGIC);
- if (write_sync(csock, &magic, sizeof(magic)) != sizeof(magic)) {
- LOG("write failed (magic)");
- return -EINVAL;
- }
- opt = cpu_to_be32(NBD_OPT_LIST);
- if (write_sync(csock, &opt, sizeof(opt)) != sizeof(opt)) {
- LOG("write failed (opt)");
- return -EINVAL;
+/* Send an error reply.
+ * Return -errno on error, 0 on success. */
+static int GCC_FMT_ATTR(4, 5)
+nbd_negotiate_send_rep_err(QIOChannel *ioc, uint32_t type,
+ uint32_t opt, const char *fmt, ...)
+{
+ va_list va;
+ char *msg;
+ int ret;
+ size_t len;
+
+ va_start(va, fmt);
+ msg = g_strdup_vprintf(fmt, va);
+ va_end(va);
+ len = strlen(msg);
+ assert(len < 4096);
+ TRACE("sending error message \"%s\"", msg);
+ ret = nbd_negotiate_send_rep_len(ioc, type, opt, len);
+ if (ret < 0) {
+ goto out;
}
- type = cpu_to_be32(NBD_REP_SERVER);
- if (write_sync(csock, &type, sizeof(type)) != sizeof(type)) {
- LOG("write failed (reply type)");
- return -EINVAL;
+ if (nbd_negotiate_write(ioc, msg, len) != len) {
+ LOG("write failed (error message)");
+ ret = -EIO;
+ } else {
+ ret = 0;
}
- len = cpu_to_be32(name_len + sizeof(len));
- if (write_sync(csock, &len, sizeof(len)) != sizeof(len)) {
- LOG("write failed (length)");
- return -EINVAL;
+out:
+ g_free(msg);
+ return ret;
+}
+
+/* Send a single NBD_REP_SERVER reply to NBD_OPT_LIST, including payload.
+ * Return -errno on error, 0 on success. */
+static int nbd_negotiate_send_rep_list(QIOChannel *ioc, NBDExport *exp)
+{
+ size_t name_len, desc_len;
+ uint32_t len;
+ const char *name = exp->name ? exp->name : "";
+ const char *desc = exp->description ? exp->description : "";
+ int rc;
+
+ TRACE("Advertising export name '%s' description '%s'", name, desc);
+ name_len = strlen(name);
+ desc_len = strlen(desc);
+ len = name_len + desc_len + sizeof(len);
+ rc = nbd_negotiate_send_rep_len(ioc, NBD_REP_SERVER, NBD_OPT_LIST, len);
+ if (rc < 0) {
+ return rc;
}
+
len = cpu_to_be32(name_len);
- if (write_sync(csock, &len, sizeof(len)) != sizeof(len)) {
- LOG("write failed (length)");
+ if (nbd_negotiate_write(ioc, &len, sizeof(len)) != sizeof(len)) {
+ LOG("write failed (name length)");
+ return -EINVAL;
+ }
+ if (nbd_negotiate_write(ioc, name, name_len) != name_len) {
+ LOG("write failed (name buffer)");
return -EINVAL;
}
- if (write_sync(csock, exp->name, name_len) != name_len) {
- LOG("write failed (buffer)");
+ if (nbd_negotiate_write(ioc, desc, desc_len) != desc_len) {
+ LOG("write failed (description buffer)");
return -EINVAL;
}
return 0;
}
-static int nbd_handle_list(NBDClient *client, uint32_t length)
+/* Process the NBD_OPT_LIST command, with a potential series of replies.
+ * Return -errno on error, 0 on success. */
+static int nbd_negotiate_handle_list(NBDClient *client, uint32_t length)
{
- int csock;
NBDExport *exp;
- csock = client->sock;
if (length) {
- if (drop_sync(csock, length) != length) {
+ if (nbd_negotiate_drop_sync(client->ioc, length) != length) {
return -EIO;
}
- return nbd_send_rep(csock, NBD_REP_ERR_INVALID, NBD_OPT_LIST);
+ return nbd_negotiate_send_rep_err(client->ioc,
+ NBD_REP_ERR_INVALID, NBD_OPT_LIST,
+ "OPT_LIST should not have length");
}
/* For each export, send a NBD_REP_SERVER reply. */
QTAILQ_FOREACH(exp, &exports, next) {
- if (nbd_send_rep_list(csock, exp)) {
+ if (nbd_negotiate_send_rep_list(client->ioc, exp)) {
return -EINVAL;
}
}
/* Finish with a NBD_REP_ACK. */
- return nbd_send_rep(csock, NBD_REP_ACK, NBD_OPT_LIST);
+ return nbd_negotiate_send_rep(client->ioc, NBD_REP_ACK, NBD_OPT_LIST);
}
-static int nbd_handle_export_name(NBDClient *client, uint32_t length)
+static int nbd_negotiate_handle_export_name(NBDClient *client, uint32_t length)
{
- int rc = -EINVAL, csock = client->sock;
- char name[256];
+ int rc = -EINVAL;
+ char name[NBD_MAX_NAME_SIZE + 1];
/* Client sends:
[20 .. xx] export name (length bytes)
*/
TRACE("Checking length");
- if (length > 255) {
+ if (length >= sizeof(name)) {
LOG("Bad length received");
goto fail;
}
- if (read_sync(csock, name, length) != length) {
+ if (nbd_negotiate_read(client->ioc, name, length) != length) {
LOG("read failed");
goto fail;
}
name[length] = '\0';
+ TRACE("Client requested export '%s'", name);
+
client->exp = nbd_export_find(name);
if (!client->exp) {
LOG("export not found");
return rc;
}
-static int nbd_receive_options(NBDClient *client)
+/* Handle NBD_OPT_STARTTLS. Return NULL to drop connection, or else the
+ * new channel for all further (now-encrypted) communication. */
+static QIOChannel *nbd_negotiate_handle_starttls(NBDClient *client,
+ uint32_t length)
+{
+ QIOChannel *ioc;
+ QIOChannelTLS *tioc;
+ struct NBDTLSHandshakeData data = { 0 };
+
+ TRACE("Setting up TLS");
+ ioc = client->ioc;
+ if (length) {
+ if (nbd_negotiate_drop_sync(ioc, length) != length) {
+ return NULL;
+ }
+ nbd_negotiate_send_rep_err(ioc, NBD_REP_ERR_INVALID, NBD_OPT_STARTTLS,
+ "OPT_STARTTLS should not have length");
+ return NULL;
+ }
+
+ if (nbd_negotiate_send_rep(client->ioc, NBD_REP_ACK,
+ NBD_OPT_STARTTLS) < 0) {
+ return NULL;
+ }
+
+ tioc = qio_channel_tls_new_server(ioc,
+ client->tlscreds,
+ client->tlsaclname,
+ NULL);
+ if (!tioc) {
+ return NULL;
+ }
+
+ qio_channel_set_name(QIO_CHANNEL(tioc), "nbd-server-tls");
+ TRACE("Starting TLS handshake");
+ data.loop = g_main_loop_new(g_main_context_default(), FALSE);
+ qio_channel_tls_handshake(tioc,
+ nbd_tls_handshake,
+ &data,
+ NULL);
+
+ if (!data.complete) {
+ g_main_loop_run(data.loop);
+ }
+ g_main_loop_unref(data.loop);
+ if (data.error) {
+ object_unref(OBJECT(tioc));
+ error_free(data.error);
+ return NULL;
+ }
+
+ return QIO_CHANNEL(tioc);
+}
+
+
+/* Process all NBD_OPT_* client option commands.
+ * Return -errno on error, 0 on success. */
+static int nbd_negotiate_options(NBDClient *client)
{
- int csock = client->sock;
uint32_t flags;
+ bool fixedNewstyle = false;
/* Client sends:
[ 0 .. 3] client flags
... Rest of request
*/
- if (read_sync(csock, &flags, sizeof(flags)) != sizeof(flags)) {
+ if (nbd_negotiate_read(client->ioc, &flags, sizeof(flags)) !=
+ sizeof(flags)) {
LOG("read failed");
return -EIO;
}
TRACE("Checking client flags");
be32_to_cpus(&flags);
- if (flags != 0 && flags != NBD_FLAG_C_FIXED_NEWSTYLE) {
- LOG("Bad client flags received");
+ if (flags & NBD_FLAG_C_FIXED_NEWSTYLE) {
+ TRACE("Client supports fixed newstyle handshake");
+ fixedNewstyle = true;
+ flags &= ~NBD_FLAG_C_FIXED_NEWSTYLE;
+ }
+ if (flags & NBD_FLAG_C_NO_ZEROES) {
+ TRACE("Client supports no zeroes at handshake end");
+ client->no_zeroes = true;
+ flags &= ~NBD_FLAG_C_NO_ZEROES;
+ }
+ if (flags != 0) {
+ TRACE("Unknown client flags 0x%" PRIx32 " received", flags);
return -EIO;
}
while (1) {
int ret;
- uint32_t tmp, length;
+ uint32_t clientflags, length;
uint64_t magic;
- if (read_sync(csock, &magic, sizeof(magic)) != sizeof(magic)) {
+ if (nbd_negotiate_read(client->ioc, &magic, sizeof(magic)) !=
+ sizeof(magic)) {
LOG("read failed");
return -EINVAL;
}
return -EINVAL;
}
- if (read_sync(csock, &tmp, sizeof(tmp)) != sizeof(tmp)) {
+ if (nbd_negotiate_read(client->ioc, &clientflags,
+ sizeof(clientflags)) != sizeof(clientflags)) {
LOG("read failed");
return -EINVAL;
}
+ clientflags = be32_to_cpu(clientflags);
- if (read_sync(csock, &length, sizeof(length)) != sizeof(length)) {
+ if (nbd_negotiate_read(client->ioc, &length, sizeof(length)) !=
+ sizeof(length)) {
LOG("read failed");
return -EINVAL;
}
length = be32_to_cpu(length);
- TRACE("Checking option");
- switch (be32_to_cpu(tmp)) {
- case NBD_OPT_LIST:
- ret = nbd_handle_list(client, length);
- if (ret < 0) {
- return ret;
+ TRACE("Checking option 0x%" PRIx32, clientflags);
+ if (client->tlscreds &&
+ client->ioc == (QIOChannel *)client->sioc) {
+ QIOChannel *tioc;
+ if (!fixedNewstyle) {
+ TRACE("Unsupported option 0x%" PRIx32, clientflags);
+ return -EINVAL;
+ }
+ switch (clientflags) {
+ case NBD_OPT_STARTTLS:
+ tioc = nbd_negotiate_handle_starttls(client, length);
+ if (!tioc) {
+ return -EIO;
+ }
+ object_unref(OBJECT(client->ioc));
+ client->ioc = QIO_CHANNEL(tioc);
+ break;
+
+ case NBD_OPT_EXPORT_NAME:
+ /* No way to return an error to client, so drop connection */
+ TRACE("Option 0x%x not permitted before TLS", clientflags);
+ return -EINVAL;
+
+ default:
+ if (nbd_negotiate_drop_sync(client->ioc, length) != length) {
+ return -EIO;
+ }
+ ret = nbd_negotiate_send_rep_err(client->ioc,
+ NBD_REP_ERR_TLS_REQD,
+ clientflags,
+ "Option 0x%" PRIx32
+ "not permitted before TLS",
+ clientflags);
+ if (ret < 0) {
+ return ret;
+ }
+ break;
+ }
+ } else if (fixedNewstyle) {
+ switch (clientflags) {
+ case NBD_OPT_LIST:
+ ret = nbd_negotiate_handle_list(client, length);
+ if (ret < 0) {
+ return ret;
+ }
+ break;
+
+ case NBD_OPT_ABORT:
+ return -EINVAL;
+
+ case NBD_OPT_EXPORT_NAME:
+ return nbd_negotiate_handle_export_name(client, length);
+
+ case NBD_OPT_STARTTLS:
+ if (nbd_negotiate_drop_sync(client->ioc, length) != length) {
+ return -EIO;
+ }
+ if (client->tlscreds) {
+ ret = nbd_negotiate_send_rep_err(client->ioc,
+ NBD_REP_ERR_INVALID,
+ clientflags,
+ "TLS already enabled");
+ } else {
+ ret = nbd_negotiate_send_rep_err(client->ioc,
+ NBD_REP_ERR_POLICY,
+ clientflags,
+ "TLS not configured");
+ }
+ if (ret < 0) {
+ return ret;
+ }
+ break;
+ default:
+ if (nbd_negotiate_drop_sync(client->ioc, length) != length) {
+ return -EIO;
+ }
+ ret = nbd_negotiate_send_rep_err(client->ioc,
+ NBD_REP_ERR_UNSUP,
+ clientflags,
+ "Unsupported option 0x%"
+ PRIx32,
+ clientflags);
+ if (ret < 0) {
+ return ret;
+ }
+ break;
+ }
+ } else {
+ /*
+ * If broken new-style we should drop the connection
+ * for anything except NBD_OPT_EXPORT_NAME
+ */
+ switch (clientflags) {
+ case NBD_OPT_EXPORT_NAME:
+ return nbd_negotiate_handle_export_name(client, length);
+
+ default:
+ TRACE("Unsupported option 0x%" PRIx32, clientflags);
+ return -EINVAL;
}
- break;
-
- case NBD_OPT_ABORT:
- return -EINVAL;
-
- case NBD_OPT_EXPORT_NAME:
- return nbd_handle_export_name(client, length);
-
- default:
- tmp = be32_to_cpu(tmp);
- LOG("Unsupported option 0x%x", tmp);
- nbd_send_rep(client->sock, NBD_REP_ERR_UNSUP, tmp);
- return -EINVAL;
}
}
}
-static int nbd_send_negotiate(NBDClient *client)
+typedef struct {
+ NBDClient *client;
+ Coroutine *co;
+} NBDClientNewData;
+
+static coroutine_fn int nbd_negotiate(NBDClientNewData *data)
{
- int csock = client->sock;
+ NBDClient *client = data->client;
char buf[8 + 8 + 8 + 128];
int rc;
- const int myflags = (NBD_FLAG_HAS_FLAGS | NBD_FLAG_SEND_TRIM |
- NBD_FLAG_SEND_FLUSH | NBD_FLAG_SEND_FUA);
+ const uint16_t myflags = (NBD_FLAG_HAS_FLAGS | NBD_FLAG_SEND_TRIM |
+ NBD_FLAG_SEND_FLUSH | NBD_FLAG_SEND_FUA);
+ bool oldStyle;
+ size_t len;
- /* Negotiation header without options:
+ /* Old style negotiation header without options
[ 0 .. 7] passwd ("NBDMAGIC")
[ 8 .. 15] magic (NBD_CLIENT_MAGIC)
[16 .. 23] size
[26 .. 27] export flags
[28 .. 151] reserved (0)
- Negotiation header with options, part 1:
+ New style negotiation header with options
[ 0 .. 7] passwd ("NBDMAGIC")
[ 8 .. 15] magic (NBD_OPTS_MAGIC)
[16 .. 17] server flags (0)
-
- part 2 (after options are sent):
+ ....options sent....
[18 .. 25] size
[26 .. 27] export flags
- [28 .. 151] reserved (0)
+ [28 .. 151] reserved (0, omit if no_zeroes)
*/
- qemu_set_block(csock);
+ qio_channel_set_blocking(client->ioc, false, NULL);
rc = -EINVAL;
TRACE("Beginning negotiation.");
memset(buf, 0, sizeof(buf));
memcpy(buf, "NBDMAGIC", 8);
- if (client->exp) {
- assert ((client->exp->nbdflags & ~65535) == 0);
- cpu_to_be64w((uint64_t*)(buf + 8), NBD_CLIENT_MAGIC);
- cpu_to_be64w((uint64_t*)(buf + 16), client->exp->size);
- cpu_to_be16w((uint16_t*)(buf + 26), client->exp->nbdflags | myflags);
+
+ oldStyle = client->exp != NULL && !client->tlscreds;
+ if (oldStyle) {
+ TRACE("advertising size %" PRIu64 " and flags %x",
+ client->exp->size, client->exp->nbdflags | myflags);
+ stq_be_p(buf + 8, NBD_CLIENT_MAGIC);
+ stq_be_p(buf + 16, client->exp->size);
+ stw_be_p(buf + 26, client->exp->nbdflags | myflags);
} else {
- cpu_to_be64w((uint64_t*)(buf + 8), NBD_OPTS_MAGIC);
- cpu_to_be16w((uint16_t *)(buf + 16), NBD_FLAG_FIXED_NEWSTYLE);
+ stq_be_p(buf + 8, NBD_OPTS_MAGIC);
+ stw_be_p(buf + 16, NBD_FLAG_FIXED_NEWSTYLE | NBD_FLAG_NO_ZEROES);
}
- if (client->exp) {
- if (write_sync(csock, buf, sizeof(buf)) != sizeof(buf)) {
+ if (oldStyle) {
+ if (client->tlscreds) {
+ TRACE("TLS cannot be enabled with oldstyle protocol");
+ goto fail;
+ }
+ if (nbd_negotiate_write(client->ioc, buf, sizeof(buf)) != sizeof(buf)) {
LOG("write failed");
goto fail;
}
} else {
- if (write_sync(csock, buf, 18) != 18) {
+ if (nbd_negotiate_write(client->ioc, buf, 18) != 18) {
LOG("write failed");
goto fail;
}
- rc = nbd_receive_options(client);
+ rc = nbd_negotiate_options(client);
if (rc != 0) {
LOG("option negotiation failed");
goto fail;
}
- assert ((client->exp->nbdflags & ~65535) == 0);
- cpu_to_be64w((uint64_t*)(buf + 18), client->exp->size);
- cpu_to_be16w((uint16_t*)(buf + 26), client->exp->nbdflags | myflags);
- if (write_sync(csock, buf + 18, sizeof(buf) - 18) != sizeof(buf) - 18) {
+ TRACE("advertising size %" PRIu64 " and flags %x",
+ client->exp->size, client->exp->nbdflags | myflags);
+ stq_be_p(buf + 18, client->exp->size);
+ stw_be_p(buf + 26, client->exp->nbdflags | myflags);
+ len = client->no_zeroes ? 10 : sizeof(buf) - 18;
+ if (nbd_negotiate_write(client->ioc, buf + 18, len) != len) {
LOG("write failed");
goto fail;
}
TRACE("Negotiation succeeded.");
rc = 0;
fail:
- qemu_set_nonblock(csock);
return rc;
}
-#ifdef __linux__
-
-int nbd_disconnect(int fd)
-{
- ioctl(fd, NBD_CLEAR_QUE);
- ioctl(fd, NBD_DISCONNECT);
- ioctl(fd, NBD_CLEAR_SOCK);
- return 0;
-}
-
-#else
-
-int nbd_disconnect(int fd)
-{
- return -ENOTSUP;
-}
-#endif
-
-static ssize_t nbd_receive_request(int csock, struct nbd_request *request)
+static ssize_t nbd_receive_request(QIOChannel *ioc, NBDRequest *request)
{
uint8_t buf[NBD_REQUEST_SIZE];
uint32_t magic;
ssize_t ret;
- ret = read_sync(csock, buf, sizeof(buf));
+ ret = read_sync(ioc, buf, sizeof(buf));
if (ret < 0) {
return ret;
}
/* Request
[ 0 .. 3] magic (NBD_REQUEST_MAGIC)
- [ 4 .. 7] type (0 == READ, 1 == WRITE)
+ [ 4 .. 5] flags (NBD_CMD_FLAG_FUA, ...)
+ [ 6 .. 7] type (NBD_CMD_READ, ...)
[ 8 .. 15] handle
[16 .. 23] from
[24 .. 27] len
*/
- magic = be32_to_cpup((uint32_t*)buf);
- request->type = be32_to_cpup((uint32_t*)(buf + 4));
- request->handle = be64_to_cpup((uint64_t*)(buf + 8));
- request->from = be64_to_cpup((uint64_t*)(buf + 16));
- request->len = be32_to_cpup((uint32_t*)(buf + 24));
+ magic = ldl_be_p(buf);
+ request->flags = lduw_be_p(buf + 4);
+ request->type = lduw_be_p(buf + 6);
+ request->handle = ldq_be_p(buf + 8);
+ request->from = ldq_be_p(buf + 16);
+ request->len = ldl_be_p(buf + 24);
- TRACE("Got request: "
- "{ magic = 0x%x, .type = %d, from = %" PRIu64" , len = %u }",
- magic, request->type, request->from, request->len);
+ TRACE("Got request: { magic = 0x%" PRIx32 ", .flags = %" PRIx16
+ ", .type = %" PRIx16 ", from = %" PRIu64 ", len = %" PRIu32 " }",
+ magic, request->flags, request->type, request->from, request->len);
if (magic != NBD_REQUEST_MAGIC) {
- LOG("invalid magic (got 0x%x)", magic);
+ LOG("invalid magic (got 0x%" PRIx32 ")", magic);
return -EINVAL;
}
return 0;
}
-static ssize_t nbd_send_reply(int csock, struct nbd_reply *reply)
+static ssize_t nbd_send_reply(QIOChannel *ioc, NBDReply *reply)
{
uint8_t buf[NBD_REPLY_SIZE];
ssize_t ret;
reply->error = system_errno_to_nbd_errno(reply->error);
+ TRACE("Sending response to client: { .error = %" PRId32
+ ", handle = %" PRIu64 " }",
+ reply->error, reply->handle);
+
/* Reply
[ 0 .. 3] magic (NBD_REPLY_MAGIC)
[ 4 .. 7] error (0 == no error)
[ 7 .. 15] handle
*/
- cpu_to_be32w((uint32_t*)buf, NBD_REPLY_MAGIC);
- cpu_to_be32w((uint32_t*)(buf + 4), reply->error);
- cpu_to_be64w((uint64_t*)(buf + 8), reply->handle);
+ stl_be_p(buf, NBD_REPLY_MAGIC);
+ stl_be_p(buf + 4, reply->error);
+ stq_be_p(buf + 8, reply->handle);
- TRACE("Sending response to client");
-
- ret = write_sync(csock, buf, sizeof(buf));
+ ret = write_sync(ioc, buf, sizeof(buf));
if (ret < 0) {
return ret;
}
assert(client->closing);
nbd_unset_handlers(client);
- close(client->sock);
- client->sock = -1;
+ object_unref(OBJECT(client->sioc));
+ object_unref(OBJECT(client->ioc));
+ if (client->tlscreds) {
+ object_unref(OBJECT(client->tlscreds));
+ }
+ g_free(client->tlsaclname);
if (client->exp) {
QTAILQ_REMOVE(&client->exp->clients, client, next);
nbd_export_put(client->exp);
/* Force requests to finish. They will drop their own references,
* then we'll close the socket and free the NBDClient.
*/
- shutdown(client->sock, 2);
+ qio_channel_shutdown(client->ioc, QIO_CHANNEL_SHUTDOWN_BOTH,
+ NULL);
/* Also tell the client, so that they release their reference. */
if (client->close) {
}
}
-static NBDRequest *nbd_request_get(NBDClient *client)
+static NBDRequestData *nbd_request_get(NBDClient *client)
{
- NBDRequest *req;
+ NBDRequestData *req;
assert(client->nb_requests <= MAX_NBD_REQUESTS - 1);
client->nb_requests++;
nbd_update_can_read(client);
- req = g_new0(NBDRequest, 1);
+ req = g_new0(NBDRequestData, 1);
nbd_client_get(client);
req->client = client;
return req;
}
-static void nbd_request_put(NBDRequest *req)
+static void nbd_request_put(NBDRequestData *req)
{
NBDClient *client = req->client;
exp->ctx = NULL;
}
-NBDExport *nbd_export_new(BlockBackend *blk, off_t dev_offset, off_t size,
- uint32_t nbdflags, void (*close)(NBDExport *),
+static void nbd_eject_notifier(Notifier *n, void *data)
+{
+ NBDExport *exp = container_of(n, NBDExport, eject_notifier);
+ nbd_export_close(exp);
+}
+
+NBDExport *nbd_export_new(BlockDriverState *bs, off_t dev_offset, off_t size,
+ uint16_t nbdflags, void (*close)(NBDExport *),
+ bool writethrough, BlockBackend *on_eject_blk,
Error **errp)
{
+ BlockBackend *blk;
NBDExport *exp = g_malloc0(sizeof(NBDExport));
+
+ blk = blk_new();
+ blk_insert_bs(blk, bs);
+ blk_set_enable_write_cache(blk, !writethrough);
+
exp->refcount = 1;
QTAILQ_INIT(&exp->clients);
exp->blk = blk;
exp->close = close;
exp->ctx = blk_get_aio_context(blk);
- blk_ref(blk);
blk_add_aio_context_notifier(blk, blk_aio_attached, blk_aio_detach, exp);
+
+ if (on_eject_blk) {
+ blk_ref(on_eject_blk);
+ exp->eject_notifier_blk = on_eject_blk;
+ exp->eject_notifier.notify = nbd_eject_notifier;
+ blk_add_remove_bs_notifier(on_eject_blk, &exp->eject_notifier);
+ }
+
/*
* NBD exports are used for non-shared storage migration. Make sure
- * that BDRV_O_INCOMING is cleared and the image is ready for write
+ * that BDRV_O_INACTIVE is cleared and the image is ready for write
* access since the export could be available before migration handover.
*/
+ aio_context_acquire(exp->ctx);
blk_invalidate_cache(blk, NULL);
+ aio_context_release(exp->ctx);
return exp;
fail:
+ blk_unref(blk);
g_free(exp);
return NULL;
}
nbd_export_put(exp);
}
+void nbd_export_set_description(NBDExport *exp, const char *description)
+{
+ g_free(exp->description);
+ exp->description = g_strdup(description);
+}
+
void nbd_export_close(NBDExport *exp)
{
NBDClient *client, *next;
client_close(client);
}
nbd_export_set_name(exp, NULL);
+ nbd_export_set_description(exp, NULL);
nbd_export_put(exp);
}
if (--exp->refcount == 0) {
assert(exp->name == NULL);
+ assert(exp->description == NULL);
if (exp->close) {
exp->close(exp);
}
if (exp->blk) {
+ if (exp->eject_notifier_blk) {
+ notifier_remove(&exp->eject_notifier);
+ blk_unref(exp->eject_notifier_blk);
+ }
blk_remove_aio_context_notifier(exp->blk, blk_aio_attached,
blk_aio_detach, exp);
blk_unref(exp->blk);
}
}
-static ssize_t nbd_co_send_reply(NBDRequest *req, struct nbd_reply *reply,
+static ssize_t nbd_co_send_reply(NBDRequestData *req, NBDReply *reply,
int len)
{
NBDClient *client = req->client;
- int csock = client->sock;
ssize_t rc, ret;
+ g_assert(qemu_in_coroutine());
qemu_co_mutex_lock(&client->send_lock);
client->send_coroutine = qemu_coroutine_self();
nbd_set_handlers(client);
if (!len) {
- rc = nbd_send_reply(csock, reply);
+ rc = nbd_send_reply(client->ioc, reply);
} else {
- socket_set_cork(csock, 1);
- rc = nbd_send_reply(csock, reply);
+ qio_channel_set_cork(client->ioc, true);
+ rc = nbd_send_reply(client->ioc, reply);
if (rc >= 0) {
- ret = qemu_co_send(csock, req->data, len);
+ ret = write_sync(client->ioc, req->data, len);
if (ret != len) {
rc = -EIO;
}
}
- socket_set_cork(csock, 0);
+ qio_channel_set_cork(client->ioc, false);
}
client->send_coroutine = NULL;
return rc;
}
-static ssize_t nbd_co_receive_request(NBDRequest *req, struct nbd_request *request)
+/* Collect a client request. Return 0 if request looks valid, -EAGAIN
+ * to keep trying the collection, -EIO to drop connection right away,
+ * and any other negative value to report an error to the client
+ * (although the caller may still need to disconnect after reporting
+ * the error). */
+static ssize_t nbd_co_receive_request(NBDRequestData *req,
+ NBDRequest *request)
{
NBDClient *client = req->client;
- int csock = client->sock;
- uint32_t command;
ssize_t rc;
+ g_assert(qemu_in_coroutine());
client->recv_coroutine = qemu_coroutine_self();
nbd_update_can_read(client);
- rc = nbd_receive_request(csock, request);
+ rc = nbd_receive_request(client->ioc, request);
if (rc < 0) {
if (rc != -EAGAIN) {
rc = -EIO;
goto out;
}
- if (request->len > NBD_MAX_BUFFER_SIZE) {
- LOG("len (%u) is larger than max len (%u)",
- request->len, NBD_MAX_BUFFER_SIZE);
- rc = -EINVAL;
+ TRACE("Decoding type");
+
+ if (request->type != NBD_CMD_WRITE) {
+ /* No payload, we are ready to read the next request. */
+ req->complete = true;
+ }
+
+ if (request->type == NBD_CMD_DISC) {
+ /* Special case: we're going to disconnect without a reply,
+ * whether or not flags, from, or len are bogus */
+ TRACE("Request type is DISCONNECT");
+ rc = -EIO;
goto out;
}
+ /* Check for sanity in the parameters, part 1. Defer as many
+ * checks as possible until after reading any NBD_CMD_WRITE
+ * payload, so we can try and keep the connection alive. */
if ((request->from + request->len) < request->from) {
- LOG("integer overflow detected! "
- "you're probably being attacked");
+ LOG("integer overflow detected, you're probably being attacked");
rc = -EINVAL;
goto out;
}
- TRACE("Decoding type");
+ if (request->type == NBD_CMD_READ || request->type == NBD_CMD_WRITE) {
+ if (request->len > NBD_MAX_BUFFER_SIZE) {
+ LOG("len (%" PRIu32" ) is larger than max len (%u)",
+ request->len, NBD_MAX_BUFFER_SIZE);
+ rc = -EINVAL;
+ goto out;
+ }
- command = request->type & NBD_CMD_MASK_COMMAND;
- if (command == NBD_CMD_READ || command == NBD_CMD_WRITE) {
- req->data = blk_blockalign(client->exp->blk, request->len);
+ req->data = blk_try_blockalign(client->exp->blk, request->len);
+ if (req->data == NULL) {
+ rc = -ENOMEM;
+ goto out;
+ }
}
- if (command == NBD_CMD_WRITE) {
- TRACE("Reading %u byte(s)", request->len);
+ if (request->type == NBD_CMD_WRITE) {
+ TRACE("Reading %" PRIu32 " byte(s)", request->len);
- if (qemu_co_recv(csock, req->data, request->len) != request->len) {
+ if (read_sync(client->ioc, req->data, request->len) != request->len) {
LOG("reading from socket failed");
rc = -EIO;
goto out;
}
+ req->complete = true;
+ }
+
+ /* Sanity checks, part 2. */
+ if (request->from + request->len > client->exp->size) {
+ LOG("operation past EOF; From: %" PRIu64 ", Len: %" PRIu32
+ ", Size: %" PRIu64, request->from, request->len,
+ (uint64_t)client->exp->size);
+ rc = request->type == NBD_CMD_WRITE ? -ENOSPC : -EINVAL;
+ goto out;
+ }
+ if (request->flags & ~NBD_CMD_FLAG_FUA) {
+ LOG("unsupported flags (got 0x%x)", request->flags);
+ rc = -EINVAL;
+ goto out;
}
+
rc = 0;
out:
{
NBDClient *client = opaque;
NBDExport *exp = client->exp;
- NBDRequest *req;
- struct nbd_request request;
- struct nbd_reply reply;
+ NBDRequestData *req;
+ NBDRequest request;
+ NBDReply reply;
ssize_t ret;
- uint32_t command;
+ int flags;
TRACE("Reading request.");
if (client->closing) {
reply.error = -ret;
goto error_reply;
}
- command = request.type & NBD_CMD_MASK_COMMAND;
- if (command != NBD_CMD_DISC && (request.from + request.len) > exp->size) {
- LOG("From: %" PRIu64 ", Len: %u, Size: %" PRIu64
- ", Offset: %" PRIu64 "\n",
- request.from, request.len,
- (uint64_t)exp->size, (uint64_t)exp->dev_offset);
- LOG("requested operation past EOF--bad client?");
- goto invalid_request;
- }
if (client->closing) {
/*
goto done;
}
- switch (command) {
+ switch (request.type) {
case NBD_CMD_READ:
TRACE("Request type is READ");
- if (request.type & NBD_CMD_FLAG_FUA) {
+ /* XXX: NBD Protocol only documents use of FUA with WRITE */
+ if (request.flags & NBD_CMD_FLAG_FUA) {
ret = blk_co_flush(exp->blk);
if (ret < 0) {
LOG("flush failed");
}
}
- ret = blk_read(exp->blk,
- (request.from + exp->dev_offset) / BDRV_SECTOR_SIZE,
- req->data, request.len / BDRV_SECTOR_SIZE);
+ ret = blk_pread(exp->blk, request.from + exp->dev_offset,
+ req->data, request.len);
if (ret < 0) {
LOG("reading from file failed");
reply.error = -ret;
goto error_reply;
}
- TRACE("Read %u byte(s)", request.len);
+ TRACE("Read %" PRIu32" byte(s)", request.len);
if (nbd_co_send_reply(req, &reply, request.len) < 0)
goto out;
break;
TRACE("Writing to device");
- ret = blk_write(exp->blk,
- (request.from + exp->dev_offset) / BDRV_SECTOR_SIZE,
- req->data, request.len / BDRV_SECTOR_SIZE);
+ flags = 0;
+ if (request.flags & NBD_CMD_FLAG_FUA) {
+ flags |= BDRV_REQ_FUA;
+ }
+ ret = blk_pwrite(exp->blk, request.from + exp->dev_offset,
+ req->data, request.len, flags);
if (ret < 0) {
LOG("writing to file failed");
reply.error = -ret;
goto error_reply;
}
- if (request.type & NBD_CMD_FLAG_FUA) {
- ret = blk_co_flush(exp->blk);
- if (ret < 0) {
- LOG("flush failed");
- reply.error = -ret;
- goto error_reply;
- }
- }
-
if (nbd_co_send_reply(req, &reply, 0) < 0) {
goto out;
}
break;
+
case NBD_CMD_DISC:
- TRACE("Request type is DISCONNECT");
- errno = 0;
- goto out;
+ /* unreachable, thanks to special case in nbd_co_receive_request() */
+ abort();
+
case NBD_CMD_FLUSH:
TRACE("Request type is FLUSH");
break;
case NBD_CMD_TRIM:
TRACE("Request type is TRIM");
- ret = blk_co_discard(exp->blk, (request.from + exp->dev_offset)
- / BDRV_SECTOR_SIZE,
- request.len / BDRV_SECTOR_SIZE);
+ ret = blk_co_pdiscard(exp->blk, request.from + exp->dev_offset,
+ request.len);
if (ret < 0) {
LOG("discard failed");
reply.error = -ret;
}
break;
default:
- LOG("invalid request type (%u) received", request.type);
- invalid_request:
+ LOG("invalid request type (%" PRIu32 ") received", request.type);
reply.error = EINVAL;
error_reply:
- if (nbd_co_send_reply(req, &reply, 0) < 0) {
+ /* We must disconnect after NBD_CMD_WRITE if we did not
+ * read the payload.
+ */
+ if (nbd_co_send_reply(req, &reply, 0) < 0 || !req->complete) {
goto out;
}
break;
NBDClient *client = opaque;
if (client->recv_coroutine) {
- qemu_coroutine_enter(client->recv_coroutine, NULL);
+ qemu_coroutine_enter(client->recv_coroutine);
} else {
- qemu_coroutine_enter(qemu_coroutine_create(nbd_trip), client);
+ qemu_coroutine_enter(qemu_coroutine_create(nbd_trip, client));
}
}
{
NBDClient *client = opaque;
- qemu_coroutine_enter(client->send_coroutine, NULL);
+ qemu_coroutine_enter(client->send_coroutine);
}
static void nbd_set_handlers(NBDClient *client)
{
if (client->exp && client->exp->ctx) {
- aio_set_fd_handler(client->exp->ctx, client->sock,
+ aio_set_fd_handler(client->exp->ctx, client->sioc->fd,
true,
client->can_read ? nbd_read : NULL,
client->send_coroutine ? nbd_restart_write : NULL,
static void nbd_unset_handlers(NBDClient *client)
{
if (client->exp && client->exp->ctx) {
- aio_set_fd_handler(client->exp->ctx, client->sock,
+ aio_set_fd_handler(client->exp->ctx, client->sioc->fd,
true, NULL, NULL, NULL);
}
}
}
}
-void nbd_client_new(NBDExport *exp, int csock, void (*close_fn)(NBDClient *))
+static coroutine_fn void nbd_co_client_start(void *opaque)
{
- NBDClient *client;
- client = g_malloc0(sizeof(NBDClient));
- client->refcount = 1;
- client->exp = exp;
- client->sock = csock;
- client->can_read = true;
- if (nbd_send_negotiate(client)) {
- shutdown(client->sock, 2);
- close_fn(client);
- return;
+ NBDClientNewData *data = opaque;
+ NBDClient *client = data->client;
+ NBDExport *exp = client->exp;
+
+ if (exp) {
+ nbd_export_get(exp);
+ }
+ if (nbd_negotiate(data)) {
+ client_close(client);
+ goto out;
}
- client->close = close_fn;
qemu_co_mutex_init(&client->send_lock);
nbd_set_handlers(client);
if (exp) {
QTAILQ_INSERT_TAIL(&exp->clients, client, next);
- nbd_export_get(exp);
}
+out:
+ g_free(data);
+}
+
+void nbd_client_new(NBDExport *exp,
+ QIOChannelSocket *sioc,
+ QCryptoTLSCreds *tlscreds,
+ const char *tlsaclname,
+ void (*close_fn)(NBDClient *))
+{
+ NBDClient *client;
+ NBDClientNewData *data = g_new(NBDClientNewData, 1);
+
+ client = g_malloc0(sizeof(NBDClient));
+ client->refcount = 1;
+ client->exp = exp;
+ client->tlscreds = tlscreds;
+ if (tlscreds) {
+ object_ref(OBJECT(client->tlscreds));
+ }
+ client->tlsaclname = g_strdup(tlsaclname);
+ client->sioc = sioc;
+ object_ref(OBJECT(client->sioc));
+ client->ioc = QIO_CHANNEL(sioc);
+ object_ref(OBJECT(client->ioc));
+ client->can_read = true;
+ client->close = close_fn;
+
+ data->client = client;
+ data->co = qemu_coroutine_create(nbd_co_client_start, data);
+ qemu_coroutine_enter(data->co);
}