#define SD_OP_RELEASE_VDI 0x13
#define SD_OP_GET_VDI_INFO 0x14
#define SD_OP_READ_VDIS 0x15
+#define SD_OP_FLUSH_VDI 0x16
#define SD_FLAG_CMD_WRITE 0x01
#define SD_FLAG_CMD_COW 0x02
+#define SD_FLAG_CMD_CACHE 0x04
#define SD_RES_SUCCESS 0x00 /* Success */
#define SD_RES_UNKNOWN 0x01 /* Unknown error */
uint8_t flags;
uint32_t id;
- QLIST_ENTRY(AIOReq) outstanding_aio_siblings;
- QLIST_ENTRY(AIOReq) aioreq_siblings;
+ QLIST_ENTRY(AIOReq) aio_siblings;
} AIOReq;
enum AIOCBState {
void (*aio_done_func)(SheepdogAIOCB *);
int canceled;
-
- QLIST_HEAD(aioreq_head, AIOReq) aioreq_head;
+ int nr_pending;
};
typedef struct BDRVSheepdogState {
char name[SD_MAX_VDI_LEN];
int is_snapshot;
+ uint8_t cache_enabled;
char *addr;
char *port;
int fd;
+ int flush_fd;
CoMutex lock;
Coroutine *co_send;
Coroutine *co_recv;
uint32_t aioreq_seq_num;
- QLIST_HEAD(outstanding_aio_head, AIOReq) outstanding_aio_head;
+ QLIST_HEAD(inflight_aio_head, AIOReq) inflight_aio_head;
+ QLIST_HEAD(pending_aio_head, AIOReq) pending_aio_head;
} BDRVSheepdogState;
static const char * sd_strerror(int err)
* Sheepdog I/O handling:
*
* 1. In sd_co_rw_vector, we send the I/O requests to the server and
- * link the requests to the outstanding_list in the
+ * link the requests to the inflight_list in the
* BDRVSheepdogState. The function exits without waiting for
* receiving the response.
*
aio_req->flags = flags;
aio_req->id = s->aioreq_seq_num++;
- QLIST_INSERT_HEAD(&s->outstanding_aio_head, aio_req,
- outstanding_aio_siblings);
- QLIST_INSERT_HEAD(&acb->aioreq_head, aio_req, aioreq_siblings);
-
+ acb->nr_pending++;
return aio_req;
}
-static inline int free_aio_req(BDRVSheepdogState *s, AIOReq *aio_req)
+static inline void free_aio_req(BDRVSheepdogState *s, AIOReq *aio_req)
{
SheepdogAIOCB *acb = aio_req->aiocb;
- QLIST_REMOVE(aio_req, outstanding_aio_siblings);
- QLIST_REMOVE(aio_req, aioreq_siblings);
+
+ QLIST_REMOVE(aio_req, aio_siblings);
g_free(aio_req);
- return !QLIST_EMPTY(&acb->aioreq_head);
+ acb->nr_pending--;
}
static void coroutine_fn sd_finish_aiocb(SheepdogAIOCB *acb)
acb->canceled = 0;
acb->coroutine = qemu_coroutine_self();
acb->ret = 0;
- QLIST_INIT(&acb->aioreq_head);
+ acb->nr_pending = 0;
return acb;
}
if (ret) {
error_report("unable to get address info %s, %s",
addr, strerror(errno));
- return -1;
+ return -errno;
}
for (res = res0; res; res = res->ai_next) {
dprintf("connected to %s:%s\n", addr, port);
goto success;
}
- fd = -1;
+ fd = -errno;
error_report("failed connect to %s:%s", addr, port);
success:
freeaddrinfo(res0);
ret = qemu_send_full(sockfd, hdr, sizeof(*hdr), 0);
if (ret < sizeof(*hdr)) {
error_report("failed to send a req, %s", strerror(errno));
+ return -errno;
}
ret = qemu_send_full(sockfd, data, *wlen, 0);
if (ret < *wlen) {
error_report("failed to send a req, %s", strerror(errno));
+ ret = -errno;
}
return ret;
}
+static coroutine_fn int send_co_req(int sockfd, SheepdogReq *hdr, void *data,
+ unsigned int *wlen)
+{
+ int ret;
+
+ ret = qemu_co_send(sockfd, hdr, sizeof(*hdr));
+ if (ret < sizeof(*hdr)) {
+ error_report("failed to send a req, %s", strerror(errno));
+ return ret;
+ }
+
+ ret = qemu_co_send(sockfd, data, *wlen);
+ if (ret < *wlen) {
+ error_report("failed to send a req, %s", strerror(errno));
+ }
+
+ return ret;
+}
+
+static coroutine_fn int do_co_req(int sockfd, SheepdogReq *hdr, void *data,
+ unsigned int *wlen, unsigned int *rlen);
+
static int do_req(int sockfd, SheepdogReq *hdr, void *data,
unsigned int *wlen, unsigned int *rlen)
{
int ret;
+ if (qemu_in_coroutine()) {
+ return do_co_req(sockfd, hdr, data, wlen, rlen);
+ }
+
socket_set_block(sockfd);
ret = send_req(sockfd, hdr, data, wlen);
if (ret < 0) {
ret = qemu_recv_full(sockfd, hdr, sizeof(*hdr), 0);
if (ret < sizeof(*hdr)) {
error_report("failed to get a rsp, %s", strerror(errno));
+ ret = -errno;
goto out;
}
ret = qemu_recv_full(sockfd, data, *rlen, 0);
if (ret < *rlen) {
error_report("failed to get the data, %s", strerror(errno));
+ ret = -errno;
+ goto out;
+ }
+ }
+ ret = 0;
+out:
+ socket_set_nonblock(sockfd);
+ return ret;
+}
+
+static void restart_co_req(void *opaque)
+{
+ Coroutine *co = opaque;
+
+ qemu_coroutine_enter(co, NULL);
+}
+
+static coroutine_fn int do_co_req(int sockfd, SheepdogReq *hdr, void *data,
+ unsigned int *wlen, unsigned int *rlen)
+{
+ int ret;
+ Coroutine *co;
+
+ co = qemu_coroutine_self();
+ qemu_aio_set_fd_handler(sockfd, NULL, restart_co_req, NULL, co);
+
+ socket_set_block(sockfd);
+ ret = send_co_req(sockfd, hdr, data, wlen);
+ if (ret < 0) {
+ goto out;
+ }
+
+ qemu_aio_set_fd_handler(sockfd, restart_co_req, NULL, NULL, co);
+
+ ret = qemu_co_recv(sockfd, hdr, sizeof(*hdr));
+ if (ret < sizeof(*hdr)) {
+ error_report("failed to get a rsp, %s", strerror(errno));
+ ret = -errno;
+ goto out;
+ }
+
+ if (*rlen > hdr->data_length) {
+ *rlen = hdr->data_length;
+ }
+
+ if (*rlen) {
+ ret = qemu_co_recv(sockfd, data, *rlen);
+ if (ret < *rlen) {
+ error_report("failed to get the data, %s", strerror(errno));
+ ret = -errno;
goto out;
}
}
ret = 0;
out:
+ qemu_aio_set_fd_handler(sockfd, NULL, NULL, NULL, NULL);
socket_set_nonblock(sockfd);
return ret;
}
struct iovec *iov, int niov, int create,
enum AIOCBState aiocb_type);
+
+static AIOReq *find_pending_req(BDRVSheepdogState *s, uint64_t oid)
+{
+ AIOReq *aio_req;
+
+ QLIST_FOREACH(aio_req, &s->pending_aio_head, aio_siblings) {
+ if (aio_req->oid == oid) {
+ return aio_req;
+ }
+ }
+
+ return NULL;
+}
+
/*
* This function searchs pending requests to the object `oid', and
* sends them.
*/
-static void coroutine_fn send_pending_req(BDRVSheepdogState *s, uint64_t oid, uint32_t id)
+static void coroutine_fn send_pending_req(BDRVSheepdogState *s, uint64_t oid)
{
- AIOReq *aio_req, *next;
+ AIOReq *aio_req;
SheepdogAIOCB *acb;
int ret;
- QLIST_FOREACH_SAFE(aio_req, &s->outstanding_aio_head,
- outstanding_aio_siblings, next) {
- if (id == aio_req->id) {
- continue;
- }
- if (aio_req->oid != oid) {
- continue;
- }
-
+ while ((aio_req = find_pending_req(s, oid)) != NULL) {
acb = aio_req->aiocb;
+ /* move aio_req from pending list to inflight one */
+ QLIST_REMOVE(aio_req, aio_siblings);
+ QLIST_INSERT_HEAD(&s->inflight_aio_head, aio_req, aio_siblings);
ret = add_aio_request(s, aio_req, acb->qiov->iov,
acb->qiov->niov, 0, acb->aiocb_type);
if (ret < 0) {
error_report("add_aio_request is failed");
free_aio_req(s, aio_req);
- if (QLIST_EMPTY(&acb->aioreq_head)) {
+ if (!acb->nr_pending) {
sd_finish_aiocb(acb);
}
}
int ret;
AIOReq *aio_req = NULL;
SheepdogAIOCB *acb;
- int rest;
unsigned long idx;
- if (QLIST_EMPTY(&s->outstanding_aio_head)) {
+ if (QLIST_EMPTY(&s->inflight_aio_head)) {
goto out;
}
goto out;
}
- /* find the right aio_req from the outstanding_aio list */
- QLIST_FOREACH(aio_req, &s->outstanding_aio_head, outstanding_aio_siblings) {
+ /* find the right aio_req from the inflight aio list */
+ QLIST_FOREACH(aio_req, &s->inflight_aio_head, aio_siblings) {
if (aio_req->id == rsp.id) {
break;
}
* create requests are not allowed, so we search the
* pending requests here.
*/
- send_pending_req(s, vid_to_data_oid(s->inode.vdi_id, idx), rsp.id);
+ send_pending_req(s, vid_to_data_oid(s->inode.vdi_id, idx));
}
break;
case AIOCB_READ_UDATA:
- ret = qemu_co_recvv(fd, acb->qiov->iov, rsp.data_length,
- aio_req->iov_offset);
+ ret = qemu_co_recvv(fd, acb->qiov->iov, acb->qiov->niov,
+ aio_req->iov_offset, rsp.data_length);
if (ret < 0) {
error_report("failed to get the data, %s", strerror(errno));
goto out;
error_report("%s", sd_strerror(rsp.result));
}
- rest = free_aio_req(s, aio_req);
- if (!rest) {
+ free_aio_req(s, aio_req);
+ if (!acb->nr_pending) {
/*
* We've finished all requests which belong to the AIOCB, so
* we can switch back to sd_co_readv/writev now.
{
BDRVSheepdogState *s = opaque;
- return !QLIST_EMPTY(&s->outstanding_aio_head);
+ return !QLIST_EMPTY(&s->inflight_aio_head) ||
+ !QLIST_EMPTY(&s->pending_aio_head);
}
static int set_nodelay(int fd)
fd = connect_to_sdog(s->addr, s->port);
if (fd < 0) {
error_report("%s", strerror(errno));
- return -1;
+ return fd;
}
socket_set_nonblock(fd);
if (ret) {
error_report("%s", strerror(errno));
closesocket(fd);
- return -1;
+ return -errno;
}
- qemu_aio_set_fd_handler(fd, co_read_response, NULL, aio_flush_request,
- NULL, s);
+ qemu_aio_set_fd_handler(fd, co_read_response, NULL, aio_flush_request, s);
return fd;
}
fd = connect_to_sdog(s->addr, s->port);
if (fd < 0) {
- return -1;
+ return fd;
}
memset(buf, 0, sizeof(buf));
ret = do_req(fd, (SheepdogReq *)&hdr, buf, &wlen, &rlen);
if (ret) {
- ret = -1;
goto out;
}
if (rsp->result != SD_RES_SUCCESS) {
error_report("cannot get vdi info, %s, %s %d %s",
sd_strerror(rsp->result), filename, snapid, tag);
- ret = -1;
+ if (rsp->result == SD_RES_NO_VDI) {
+ ret = -ENOENT;
+ } else {
+ ret = -EIO;
+ }
goto out;
}
*vid = rsp->vdi_id;
hdr.flags = SD_FLAG_CMD_WRITE | flags;
}
+ if (s->cache_enabled) {
+ hdr.flags |= SD_FLAG_CMD_CACHE;
+ }
+
hdr.oid = oid;
hdr.cow_oid = old_oid;
hdr.copies = s->inode.nr_copies;
qemu_co_mutex_lock(&s->lock);
s->co_send = qemu_coroutine_self();
qemu_aio_set_fd_handler(s->fd, co_read_response, co_write_request,
- aio_flush_request, NULL, s);
+ aio_flush_request, s);
socket_set_cork(s->fd, 1);
/* send a header */
if (ret < 0) {
qemu_co_mutex_unlock(&s->lock);
error_report("failed to send a req, %s", strerror(errno));
- return -EIO;
+ return -errno;
}
if (wlen) {
- ret = qemu_co_sendv(s->fd, iov, wlen, aio_req->iov_offset);
+ ret = qemu_co_sendv(s->fd, iov, niov, aio_req->iov_offset, wlen);
if (ret < 0) {
qemu_co_mutex_unlock(&s->lock);
error_report("failed to send a data, %s", strerror(errno));
- return -EIO;
+ return -errno;
}
}
socket_set_cork(s->fd, 0);
qemu_aio_set_fd_handler(s->fd, co_read_response, NULL,
- aio_flush_request, NULL, s);
+ aio_flush_request, s);
qemu_co_mutex_unlock(&s->lock);
return 0;
static int read_write_object(int fd, char *buf, uint64_t oid, int copies,
unsigned int datalen, uint64_t offset,
- int write, int create)
+ int write, int create, uint8_t cache)
{
SheepdogObjReq hdr;
SheepdogObjRsp *rsp = (SheepdogObjRsp *)&hdr;
rlen = datalen;
hdr.opcode = SD_OP_READ_OBJ;
}
+
+ if (cache) {
+ hdr.flags |= SD_FLAG_CMD_CACHE;
+ }
+
hdr.oid = oid;
hdr.data_length = datalen;
hdr.offset = offset;
ret = do_req(fd, (SheepdogReq *)&hdr, buf, &wlen, &rlen);
if (ret) {
error_report("failed to send a request to the sheep");
- return -1;
+ return ret;
}
switch (rsp->result) {
return 0;
default:
error_report("%s", sd_strerror(rsp->result));
- return -1;
+ return -EIO;
}
}
static int read_object(int fd, char *buf, uint64_t oid, int copies,
- unsigned int datalen, uint64_t offset)
+ unsigned int datalen, uint64_t offset, uint8_t cache)
{
- return read_write_object(fd, buf, oid, copies, datalen, offset, 0, 0);
+ return read_write_object(fd, buf, oid, copies, datalen, offset, 0, 0,
+ cache);
}
static int write_object(int fd, char *buf, uint64_t oid, int copies,
- unsigned int datalen, uint64_t offset, int create)
+ unsigned int datalen, uint64_t offset, int create,
+ uint8_t cache)
{
- return read_write_object(fd, buf, oid, copies, datalen, offset, 1, create);
+ return read_write_object(fd, buf, oid, copies, datalen, offset, 1, create,
+ cache);
}
static int sd_open(BlockDriverState *bs, const char *filename, int flags)
strstart(filename, "sheepdog:", (const char **)&filename);
- QLIST_INIT(&s->outstanding_aio_head);
+ QLIST_INIT(&s->inflight_aio_head);
+ QLIST_INIT(&s->pending_aio_head);
s->fd = -1;
memset(vdi, 0, sizeof(vdi));
memset(tag, 0, sizeof(tag));
if (parse_vdiname(s, filename, vdi, &snapid, tag) < 0) {
+ ret = -EINVAL;
goto out;
}
s->fd = get_sheep_fd(s);
if (s->fd < 0) {
+ ret = s->fd;
goto out;
}
goto out;
}
- if (snapid) {
+ if (flags & BDRV_O_CACHE_WB) {
+ s->cache_enabled = 1;
+ s->flush_fd = connect_to_sdog(s->addr, s->port);
+ if (s->flush_fd < 0) {
+ error_report("failed to connect");
+ ret = s->flush_fd;
+ goto out;
+ }
+ }
+
+ if (snapid || tag[0] != '\0') {
dprintf("%" PRIx32 " snapshot inode was open.\n", vid);
s->is_snapshot = 1;
}
fd = connect_to_sdog(s->addr, s->port);
if (fd < 0) {
error_report("failed to connect");
+ ret = fd;
goto out;
}
buf = g_malloc(SD_INODE_SIZE);
- ret = read_object(fd, buf, vid_to_vdi_oid(vid), 0, SD_INODE_SIZE, 0);
+ ret = read_object(fd, buf, vid_to_vdi_oid(vid), 0, SD_INODE_SIZE, 0,
+ s->cache_enabled);
closesocket(fd);
g_free(buf);
return 0;
out:
- qemu_aio_set_fd_handler(s->fd, NULL, NULL, NULL, NULL, NULL);
+ qemu_aio_set_fd_handler(s->fd, NULL, NULL, NULL, NULL);
if (s->fd >= 0) {
closesocket(s->fd);
}
g_free(buf);
- return -1;
+ return ret;
}
static int do_sd_create(char *filename, int64_t vdi_size,
fd = connect_to_sdog(addr, port);
if (fd < 0) {
- return -EIO;
+ return fd;
}
memset(buf, 0, sizeof(buf));
closesocket(fd);
if (ret) {
- return -EIO;
+ return ret;
}
if (rsp->result != SD_RES_SUCCESS) {
static int sd_create(const char *filename, QEMUOptionParameter *options)
{
- int ret;
+ int ret = 0;
uint32_t vid = 0, base_vid = 0;
int64_t vdi_size = 0;
char *backing_file = NULL;
- BDRVSheepdogState s;
+ BDRVSheepdogState *s;
char vdi[SD_MAX_VDI_LEN], tag[SD_MAX_VDI_TAG_LEN];
uint32_t snapid;
int prealloc = 0;
const char *vdiname;
+ s = g_malloc0(sizeof(BDRVSheepdogState));
+
strstart(filename, "sheepdog:", &vdiname);
- memset(&s, 0, sizeof(s));
memset(vdi, 0, sizeof(vdi));
memset(tag, 0, sizeof(tag));
- if (parse_vdiname(&s, vdiname, vdi, &snapid, tag) < 0) {
+ if (parse_vdiname(s, vdiname, vdi, &snapid, tag) < 0) {
error_report("invalid filename");
- return -EINVAL;
+ ret = -EINVAL;
+ goto out;
}
while (options && options->name) {
} else {
error_report("Invalid preallocation mode: '%s'",
options->value.s);
- return -EINVAL;
+ ret = -EINVAL;
+ goto out;
}
}
options++;
if (vdi_size > SD_MAX_VDI_SIZE) {
error_report("too big image size");
- return -EINVAL;
+ ret = -EINVAL;
+ goto out;
}
if (backing_file) {
drv = bdrv_find_protocol(backing_file);
if (!drv || strcmp(drv->protocol_name, "sheepdog") != 0) {
error_report("backing_file must be a sheepdog image");
- return -EINVAL;
+ ret = -EINVAL;
+ goto out;
}
ret = bdrv_file_open(&bs, backing_file, 0);
- if (ret < 0)
- return -EIO;
+ if (ret < 0) {
+ goto out;
+ }
s = bs->opaque;
if (!is_snapshot(&s->inode)) {
error_report("cannot clone from a non snapshot vdi");
bdrv_delete(bs);
- return -EINVAL;
+ ret = -EINVAL;
+ goto out;
}
base_vid = s->inode.vdi_id;
bdrv_delete(bs);
}
- ret = do_sd_create(vdi, vdi_size, base_vid, &vid, 0, s.addr, s.port);
+ ret = do_sd_create(vdi, vdi_size, base_vid, &vid, 0, s->addr, s->port);
if (!prealloc || ret) {
- return ret;
+ goto out;
}
- return sd_prealloc(filename);
+ ret = sd_prealloc(filename);
+out:
+ g_free(s);
+ return ret;
}
static void sd_close(BlockDriverState *bs)
error_report("%s, %s", sd_strerror(rsp->result), s->name);
}
- qemu_aio_set_fd_handler(s->fd, NULL, NULL, NULL, NULL, NULL);
+ qemu_aio_set_fd_handler(s->fd, NULL, NULL, NULL, NULL);
closesocket(s->fd);
+ if (s->cache_enabled) {
+ closesocket(s->flush_fd);
+ }
g_free(s->addr);
}
fd = connect_to_sdog(s->addr, s->port);
if (fd < 0) {
- return -EIO;
+ return fd;
}
/* we don't need to update entire object */
datalen = SD_INODE_SIZE - sizeof(s->inode.data_vdi_id);
s->inode.vdi_size = offset;
ret = write_object(fd, (char *)&s->inode, vid_to_vdi_oid(s->inode.vdi_id),
- s->inode.nr_copies, datalen, 0, 0);
+ s->inode.nr_copies, datalen, 0, 0, s->cache_enabled);
close(fd);
if (ret < 0) {
error_report("failed to update an inode.");
- return -EIO;
}
- return 0;
+ return ret;
}
/*
iov.iov_len = sizeof(s->inode);
aio_req = alloc_aio_req(s, acb, vid_to_vdi_oid(s->inode.vdi_id),
data_len, offset, 0, 0, offset);
+ QLIST_INSERT_HEAD(&s->inflight_aio_head, aio_req, aio_siblings);
ret = add_aio_request(s, aio_req, &iov, 1, 0, AIOCB_WRITE_UDATA);
if (ret) {
free_aio_req(s, aio_req);
fd = connect_to_sdog(s->addr, s->port);
if (fd < 0) {
error_report("failed to connect");
+ ret = fd;
goto out;
}
ret = read_object(fd, buf, vid_to_vdi_oid(vid), s->inode.nr_copies,
- SD_INODE_SIZE, 0);
+ SD_INODE_SIZE, 0, s->cache_enabled);
closesocket(fd);
* Send I/O requests to the server.
*
* This function sends requests to the server, links the requests to
- * the outstanding_list in BDRVSheepdogState, and exits without
+ * the inflight_list in BDRVSheepdogState, and exits without
* waiting the response. The responses are received in the
* `aio_read_response' function which is called from the main loop as
* a fd handler.
}
}
+ /*
+ * Make sure we don't free the aiocb before we are done with all requests.
+ * This additional reference is dropped at the end of this function.
+ */
+ acb->nr_pending++;
+
while (done != total) {
uint8_t flags = 0;
uint64_t old_oid = 0;
}
if (create) {
- dprintf("update ino (%" PRIu32") %" PRIu64 " %" PRIu64
- " %" PRIu64 "\n", inode->vdi_id, oid,
+ dprintf("update ino (%" PRIu32 ") %" PRIu64 " %" PRIu64 " %ld\n",
+ inode->vdi_id, oid,
vid_to_data_oid(inode->data_vdi_id[idx], idx), idx);
oid = vid_to_data_oid(inode->vdi_id, idx);
- dprintf("new oid %lx\n", oid);
+ dprintf("new oid %" PRIx64 "\n", oid);
}
aio_req = alloc_aio_req(s, acb, oid, len, offset, flags, old_oid, done);
if (create) {
AIOReq *areq;
- QLIST_FOREACH(areq, &s->outstanding_aio_head,
- outstanding_aio_siblings) {
- if (areq == aio_req) {
- continue;
- }
+ QLIST_FOREACH(areq, &s->inflight_aio_head, aio_siblings) {
if (areq->oid == oid) {
/*
* Sheepdog cannot handle simultaneous create
*/
aio_req->flags = 0;
aio_req->base_oid = 0;
+ QLIST_INSERT_HEAD(&s->pending_aio_head, aio_req,
+ aio_siblings);
goto done;
}
}
}
+ QLIST_INSERT_HEAD(&s->inflight_aio_head, aio_req, aio_siblings);
ret = add_aio_request(s, aio_req, acb->qiov->iov, acb->qiov->niov,
create, acb->aiocb_type);
if (ret < 0) {
done += len;
}
out:
- if (QLIST_EMPTY(&acb->aioreq_head)) {
+ if (!--acb->nr_pending) {
return acb->ret;
}
return 1;
int ret;
if (bs->growable && sector_num + nb_sectors > bs->total_sectors) {
- /* TODO: shouldn't block here */
- if (sd_truncate(bs, (sector_num + nb_sectors) * SECTOR_SIZE) < 0) {
- return -EIO;
+ ret = sd_truncate(bs, (sector_num + nb_sectors) * SECTOR_SIZE);
+ if (ret < 0) {
+ return ret;
}
bs->total_sectors = sector_num + nb_sectors;
}
return acb->ret;
}
+static int coroutine_fn sd_co_flush_to_disk(BlockDriverState *bs)
+{
+ BDRVSheepdogState *s = bs->opaque;
+ SheepdogObjReq hdr = { 0 };
+ SheepdogObjRsp *rsp = (SheepdogObjRsp *)&hdr;
+ SheepdogInode *inode = &s->inode;
+ int ret;
+ unsigned int wlen = 0, rlen = 0;
+
+ if (!s->cache_enabled) {
+ return 0;
+ }
+
+ hdr.opcode = SD_OP_FLUSH_VDI;
+ hdr.oid = vid_to_vdi_oid(inode->vdi_id);
+
+ ret = do_req(s->flush_fd, (SheepdogReq *)&hdr, NULL, &wlen, &rlen);
+ if (ret) {
+ error_report("failed to send a request to the sheep");
+ return ret;
+ }
+
+ if (rsp->result == SD_RES_INVALID_PARMS) {
+ dprintf("disable write cache since the server doesn't support it\n");
+
+ s->cache_enabled = 0;
+ closesocket(s->flush_fd);
+ return 0;
+ }
+
+ if (rsp->result != SD_RES_SUCCESS) {
+ error_report("%s", sd_strerror(rsp->result));
+ return -EIO;
+ }
+
+ return 0;
+}
+
static int sd_snapshot_create(BlockDriverState *bs, QEMUSnapshotInfo *sn_info)
{
BDRVSheepdogState *s = bs->opaque;
SheepdogInode *inode;
unsigned int datalen;
- dprintf("sn_info: name %s id_str %s s: name %s vm_state_size %d "
+ dprintf("sn_info: name %s id_str %s s: name %s vm_state_size %" PRId64 " "
"is_snapshot %d\n", sn_info->name, sn_info->id_str,
s->name, sn_info->vm_state_size, s->is_snapshot);
/* refresh inode. */
fd = connect_to_sdog(s->addr, s->port);
if (fd < 0) {
- ret = -EIO;
+ ret = fd;
goto cleanup;
}
ret = write_object(fd, (char *)&s->inode, vid_to_vdi_oid(s->inode.vdi_id),
- s->inode.nr_copies, datalen, 0, 0);
+ s->inode.nr_copies, datalen, 0, 0, s->cache_enabled);
if (ret < 0) {
error_report("failed to write snapshot's inode.");
- ret = -EIO;
goto cleanup;
}
if (ret < 0) {
error_report("failed to create inode for snapshot. %s",
strerror(errno));
- ret = -EIO;
goto cleanup;
}
inode = (SheepdogInode *)g_malloc(datalen);
ret = read_object(fd, (char *)inode, vid_to_vdi_oid(new_vid),
- s->inode.nr_copies, datalen, 0);
+ s->inode.nr_copies, datalen, 0, s->cache_enabled);
if (ret < 0) {
error_report("failed to read new inode info. %s", strerror(errno));
- ret = -EIO;
goto cleanup;
}
char *buf = NULL;
uint32_t vid;
uint32_t snapid = 0;
- int ret = -ENOENT, fd;
+ int ret = 0, fd;
old_s = g_malloc(sizeof(BDRVSheepdogState));
ret = find_vdi_name(s, vdi, snapid, tag, &vid, 1);
if (ret) {
error_report("Failed to find_vdi_name");
- ret = -ENOENT;
goto out;
}
fd = connect_to_sdog(s->addr, s->port);
if (fd < 0) {
error_report("failed to connect");
+ ret = fd;
goto out;
}
buf = g_malloc(SD_INODE_SIZE);
ret = read_object(fd, buf, vid_to_vdi_oid(vid), s->inode.nr_copies,
- SD_INODE_SIZE, 0);
+ SD_INODE_SIZE, 0, s->cache_enabled);
closesocket(fd);
if (ret) {
- ret = -ENOENT;
goto out;
}
fd = connect_to_sdog(s->addr, s->port);
if (fd < 0) {
+ ret = fd;
goto out;
}
fd = connect_to_sdog(s->addr, s->port);
if (fd < 0) {
error_report("failed to connect");
+ ret = fd;
goto out;
}
/* we don't need to read entire object */
ret = read_object(fd, (char *)&inode, vid_to_vdi_oid(vid),
- 0, SD_INODE_SIZE - sizeof(inode.data_vdi_id), 0);
+ 0, SD_INODE_SIZE - sizeof(inode.data_vdi_id), 0,
+ s->cache_enabled);
if (ret) {
continue;
g_free(vdi_inuse);
+ if (ret < 0) {
+ return ret;
+ }
+
return found;
}
int64_t pos, int size, int load)
{
int fd, create;
- int ret = 0;
+ int ret = 0, remaining = size;
unsigned int data_len;
uint64_t vmstate_oid;
uint32_t vdi_index;
fd = connect_to_sdog(s->addr, s->port);
if (fd < 0) {
- ret = -EIO;
- goto cleanup;
+ return fd;
}
- while (size) {
+ while (remaining) {
vdi_index = pos / SD_DATA_OBJ_SIZE;
offset = pos % SD_DATA_OBJ_SIZE;
- data_len = MIN(size, SD_DATA_OBJ_SIZE);
+ data_len = MIN(remaining, SD_DATA_OBJ_SIZE);
vmstate_oid = vid_to_vmstate_oid(s->inode.vdi_id, vdi_index);
create = (offset == 0);
if (load) {
ret = read_object(fd, (char *)data, vmstate_oid,
- s->inode.nr_copies, data_len, offset);
+ s->inode.nr_copies, data_len, offset,
+ s->cache_enabled);
} else {
ret = write_object(fd, (char *)data, vmstate_oid,
- s->inode.nr_copies, data_len, offset, create);
+ s->inode.nr_copies, data_len, offset, create,
+ s->cache_enabled);
}
if (ret < 0) {
error_report("failed to save vmstate %s", strerror(errno));
- ret = -EIO;
goto cleanup;
}
pos += data_len;
- size -= data_len;
- ret += data_len;
+ remaining -= data_len;
}
+ ret = size;
cleanup:
closesocket(fd);
return ret;
.bdrv_co_readv = sd_co_readv,
.bdrv_co_writev = sd_co_writev,
+ .bdrv_co_flush_to_disk = sd_co_flush_to_disk,
.bdrv_snapshot_create = sd_snapshot_create,
.bdrv_snapshot_goto = sd_snapshot_goto,