#define SD_OP_READ_VDIS 0x15
#define SD_OP_FLUSH_VDI 0x16
#define SD_OP_DEL_VDI 0x17
+#define SD_OP_GET_CLUSTER_DEFAULT 0x18
#define SD_FLAG_CMD_WRITE 0x01
#define SD_FLAG_CMD_COW 0x02
#define SD_NR_VDIS (1U << 24)
#define SD_DATA_OBJ_SIZE (UINT64_C(1) << 22)
#define SD_MAX_VDI_SIZE (SD_DATA_OBJ_SIZE * MAX_DATA_OBJS)
+#define SD_DEFAULT_BLOCK_SIZE_SHIFT 22
/*
* For erasure coding, we use at most SD_EC_MAX_STRIP for data strips and
* (SD_EC_MAX_STRIP - 1) for parity strips
#define SD_INODE_SIZE (sizeof(SheepdogInode))
#define CURRENT_VDI_ID 0
+#define LOCK_TYPE_NORMAL 0
+#define LOCK_TYPE_SHARED 1 /* for iSCSI multipath */
+
typedef struct SheepdogReq {
uint8_t proto_ver;
uint8_t opcode;
uint32_t base_vdi_id;
uint8_t copies;
uint8_t copy_policy;
- uint8_t reserved[2];
+ uint8_t store_policy;
+ uint8_t block_size_shift;
uint32_t snapid;
- uint32_t pad[3];
+ uint32_t type;
+ uint32_t pad[2];
} SheepdogVdiReq;
typedef struct SheepdogVdiRsp {
uint32_t pad[5];
} SheepdogVdiRsp;
+typedef struct SheepdogClusterRsp {
+ uint8_t proto_ver;
+ uint8_t opcode;
+ uint16_t flags;
+ uint32_t epoch;
+ uint32_t id;
+ uint32_t data_length;
+ uint32_t result;
+ uint8_t nr_copies;
+ uint8_t copy_policy;
+ uint8_t block_size_shift;
+ uint8_t __pad1;
+ uint32_t __pad2[6];
+} SheepdogClusterRsp;
+
typedef struct SheepdogInode {
char name[SD_MAX_VDI_LEN];
char tag[SD_MAX_VDI_TAG_LEN];
uint32_t data_vdi_id[MAX_DATA_OBJS];
} SheepdogInode;
+#define SD_INODE_HEADER_SIZE offsetof(SheepdogInode, data_vdi_id)
+
/*
* 64 bit FNV-1a non-zero initial basis
*/
unsigned int data_len;
uint8_t flags;
uint32_t id;
+ bool create;
QLIST_ENTRY(AIOReq) aio_siblings;
} AIOReq;
AIOCB_DISCARD_OBJ,
};
+#define AIOCBOverwrapping(x, y) \
+ (!(x->max_affect_data_idx < y->min_affect_data_idx \
+ || y->max_affect_data_idx < x->min_affect_data_idx))
+
struct SheepdogAIOCB {
- BlockDriverAIOCB common;
+ BlockAIOCB common;
QEMUIOVector *qiov;
void (*aio_done_func)(SheepdogAIOCB *);
bool cancelable;
- bool *finished;
int nr_pending;
+
+ uint32_t min_affect_data_idx;
+ uint32_t max_affect_data_idx;
+
+ QLIST_ENTRY(SheepdogAIOCB) aiocb_siblings;
};
typedef struct BDRVSheepdogState {
BlockDriverState *bs;
+ AioContext *aio_context;
SheepdogInode inode;
/* Every aio request must be linked to either of these queues. */
QLIST_HEAD(inflight_aio_head, AIOReq) inflight_aio_head;
- QLIST_HEAD(pending_aio_head, AIOReq) pending_aio_head;
QLIST_HEAD(failed_aio_head, AIOReq) failed_aio_head;
+
+ CoQueue overwrapping_queue;
+ QLIST_HEAD(inflight_aiocb_head, SheepdogAIOCB) inflight_aiocb_head;
} BDRVSheepdogState;
static const char * sd_strerror(int err)
static inline AIOReq *alloc_aio_req(BDRVSheepdogState *s, SheepdogAIOCB *acb,
uint64_t oid, unsigned int data_len,
- uint64_t offset, uint8_t flags,
+ uint64_t offset, uint8_t flags, bool create,
uint64_t base_oid, unsigned int iov_offset)
{
AIOReq *aio_req;
aio_req->data_len = data_len;
aio_req->flags = flags;
aio_req->id = s->aioreq_seq_num++;
+ aio_req->create = create;
acb->nr_pending++;
return aio_req;
static void coroutine_fn sd_finish_aiocb(SheepdogAIOCB *acb)
{
qemu_coroutine_enter(acb->coroutine, NULL);
- if (acb->finished) {
- *acb->finished = true;
- }
- qemu_aio_release(acb);
+ qemu_aio_unref(acb);
}
/*
return true;
}
-static void sd_aio_cancel(BlockDriverAIOCB *blockacb)
+static void sd_aio_cancel(BlockAIOCB *blockacb)
{
SheepdogAIOCB *acb = (SheepdogAIOCB *)blockacb;
BDRVSheepdogState *s = acb->common.bs->opaque;
AIOReq *aioreq, *next;
- bool finished = false;
-
- acb->finished = &finished;
- while (!finished) {
- if (sd_acb_cancelable(acb)) {
- /* Remove outstanding requests from pending and failed queues. */
- QLIST_FOREACH_SAFE(aioreq, &s->pending_aio_head, aio_siblings,
- next) {
- if (aioreq->aiocb == acb) {
- free_aio_req(s, aioreq);
- }
- }
- QLIST_FOREACH_SAFE(aioreq, &s->failed_aio_head, aio_siblings,
- next) {
- if (aioreq->aiocb == acb) {
- free_aio_req(s, aioreq);
- }
+
+ if (sd_acb_cancelable(acb)) {
+ /* Remove outstanding requests from failed queue. */
+ QLIST_FOREACH_SAFE(aioreq, &s->failed_aio_head, aio_siblings,
+ next) {
+ if (aioreq->aiocb == acb) {
+ free_aio_req(s, aioreq);
}
+ }
- assert(acb->nr_pending == 0);
- sd_finish_aiocb(acb);
- return;
+ assert(acb->nr_pending == 0);
+ if (acb->common.cb) {
+ acb->common.cb(acb->common.opaque, -ECANCELED);
}
- qemu_aio_wait();
+ sd_finish_aiocb(acb);
}
}
static const AIOCBInfo sd_aiocb_info = {
- .aiocb_size = sizeof(SheepdogAIOCB),
- .cancel = sd_aio_cancel,
+ .aiocb_size = sizeof(SheepdogAIOCB),
+ .cancel_async = sd_aio_cancel,
};
static SheepdogAIOCB *sd_aio_setup(BlockDriverState *bs, QEMUIOVector *qiov,
int64_t sector_num, int nb_sectors)
{
SheepdogAIOCB *acb;
+ uint32_t object_size;
+ BDRVSheepdogState *s = bs->opaque;
+
+ object_size = (UINT32_C(1) << s->inode.block_size_shift);
acb = qemu_aio_get(&sd_aiocb_info, bs, NULL, NULL);
acb->aio_done_func = NULL;
acb->cancelable = true;
- acb->finished = NULL;
acb->coroutine = qemu_coroutine_self();
acb->ret = 0;
acb->nr_pending = 0;
+
+ acb->min_affect_data_idx = acb->sector_num * BDRV_SECTOR_SIZE / object_size;
+ acb->max_affect_data_idx = (acb->sector_num * BDRV_SECTOR_SIZE +
+ acb->nb_sectors * BDRV_SECTOR_SIZE) / object_size;
+
return acb;
}
+/* Return -EIO in case of error, file descriptor on success */
static int connect_to_sdog(BDRVSheepdogState *s, Error **errp)
{
int fd;
if (fd >= 0) {
qemu_set_nonblock(fd);
+ } else {
+ fd = -EIO;
}
return fd;
}
+/* Return 0 on success and -errno in case of error */
static coroutine_fn int send_co_req(int sockfd, SheepdogReq *hdr, void *data,
unsigned int *wlen)
{
ret = qemu_co_send(sockfd, hdr, sizeof(*hdr));
if (ret != sizeof(*hdr)) {
error_report("failed to send a req, %s", strerror(errno));
+ ret = -socket_error();
return ret;
}
ret = qemu_co_send(sockfd, data, *wlen);
if (ret != *wlen) {
+ ret = -socket_error();
error_report("failed to send a req, %s", strerror(errno));
}
typedef struct SheepdogReqCo {
int sockfd;
+ AioContext *aio_context;
SheepdogReq *hdr;
void *data;
unsigned int *wlen;
unsigned int *rlen = srco->rlen;
co = qemu_coroutine_self();
- qemu_aio_set_fd_handler(sockfd, NULL, restart_co_req, co);
+ aio_set_fd_handler(srco->aio_context, sockfd, NULL, restart_co_req, co);
ret = send_co_req(sockfd, hdr, data, wlen);
if (ret < 0) {
goto out;
}
- qemu_aio_set_fd_handler(sockfd, restart_co_req, NULL, co);
+ aio_set_fd_handler(srco->aio_context, sockfd, restart_co_req, NULL, co);
ret = qemu_co_recv(sockfd, hdr, sizeof(*hdr));
if (ret != sizeof(*hdr)) {
out:
/* there is at most one request for this sockfd, so it is safe to
* set each handler to NULL. */
- qemu_aio_set_fd_handler(sockfd, NULL, NULL, NULL);
+ aio_set_fd_handler(srco->aio_context, sockfd, NULL, NULL, NULL);
srco->ret = ret;
srco->finished = true;
}
-static int do_req(int sockfd, SheepdogReq *hdr, void *data,
- unsigned int *wlen, unsigned int *rlen)
+/*
+ * Send the request to the sheep in a synchronous manner.
+ *
+ * Return 0 on success, -errno in case of error.
+ */
+static int do_req(int sockfd, AioContext *aio_context, SheepdogReq *hdr,
+ void *data, unsigned int *wlen, unsigned int *rlen)
{
Coroutine *co;
SheepdogReqCo srco = {
.sockfd = sockfd,
+ .aio_context = aio_context,
.hdr = hdr,
.data = data,
.wlen = wlen,
co = qemu_coroutine_create(do_co_req);
qemu_coroutine_enter(co, &srco);
while (!srco.finished) {
- qemu_aio_wait();
+ aio_poll(aio_context, true);
}
}
}
static void coroutine_fn add_aio_request(BDRVSheepdogState *s, AIOReq *aio_req,
- struct iovec *iov, int niov, bool create,
- enum AIOCBState aiocb_type);
+ struct iovec *iov, int niov,
+ enum AIOCBState aiocb_type);
static void coroutine_fn resend_aioreq(BDRVSheepdogState *s, AIOReq *aio_req);
static int reload_inode(BDRVSheepdogState *s, uint32_t snapid, const char *tag);
static int get_sheep_fd(BDRVSheepdogState *s, Error **errp);
static void co_write_request(void *opaque);
-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)
-{
- AIOReq *aio_req;
- SheepdogAIOCB *acb;
-
- 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);
- add_aio_request(s, aio_req, acb->qiov->iov, acb->qiov->niov, false,
- acb->aiocb_type);
- }
-}
-
static coroutine_fn void reconnect_to_sdog(void *opaque)
{
- Error *local_err = NULL;
BDRVSheepdogState *s = opaque;
AIOReq *aio_req, *next;
- qemu_aio_set_fd_handler(s->fd, NULL, NULL, NULL);
+ aio_set_fd_handler(s->aio_context, s->fd, NULL, NULL, NULL);
close(s->fd);
s->fd = -1;
/* Try to reconnect the sheepdog server every one second. */
while (s->fd < 0) {
+ Error *local_err = NULL;
s->fd = get_sheep_fd(s, &local_err);
if (s->fd < 0) {
DPRINTF("Wait for connection to be established\n");
- qerror_report_err(local_err);
- error_free(local_err);
+ error_report_err(local_err);
co_aio_sleep_ns(bdrv_get_aio_context(s->bs), QEMU_CLOCK_REALTIME,
1000000000ULL);
}
}
idx = data_oid_to_idx(aio_req->oid);
- if (s->inode.data_vdi_id[idx] != s->inode.vdi_id) {
+ if (aio_req->create) {
/*
* If the object is newly created one, we need to update
* the vdi object (metadata object). min_dirty_data_idx
s->max_dirty_data_idx = MAX(idx, s->max_dirty_data_idx);
s->min_dirty_data_idx = MIN(idx, s->min_dirty_data_idx);
}
- /*
- * Some requests may be blocked because simultaneous
- * create requests are not allowed, so we search the
- * pending requests here.
- */
- send_pending_req(s, aio_req->oid);
}
break;
case AIOCB_READ_UDATA:
return fd;
}
- qemu_aio_set_fd_handler(fd, co_read_response, NULL, s);
+ aio_set_fd_handler(s->aio_context, fd, co_read_response, NULL, s);
return fd;
}
memset(&hdr, 0, sizeof(hdr));
if (lock) {
hdr.opcode = SD_OP_LOCK_VDI;
+ hdr.type = LOCK_TYPE_NORMAL;
} else {
hdr.opcode = SD_OP_GET_VDI_INFO;
}
hdr.snapid = snapid;
hdr.flags = SD_FLAG_CMD_WRITE;
- ret = do_req(fd, (SheepdogReq *)&hdr, buf, &wlen, &rlen);
+ ret = do_req(fd, s->aio_context, (SheepdogReq *)&hdr, buf, &wlen, &rlen);
if (ret) {
error_setg_errno(errp, -ret, "cannot get vdi info");
goto out;
sd_strerror(rsp->result), filename, snapid, tag);
if (rsp->result == SD_RES_NO_VDI) {
ret = -ENOENT;
+ } else if (rsp->result == SD_RES_VDI_LOCKED) {
+ ret = -EBUSY;
} else {
ret = -EIO;
}
}
static void coroutine_fn add_aio_request(BDRVSheepdogState *s, AIOReq *aio_req,
- struct iovec *iov, int niov, bool create,
- enum AIOCBState aiocb_type)
+ struct iovec *iov, int niov,
+ enum AIOCBState aiocb_type)
{
int nr_copies = s->inode.nr_copies;
SheepdogObjReq hdr;
uint64_t offset = aio_req->offset;
uint8_t flags = aio_req->flags;
uint64_t old_oid = aio_req->base_oid;
+ bool create = aio_req->create;
if (!nr_copies) {
error_report("bug");
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, s);
+ aio_set_fd_handler(s->aio_context, s->fd,
+ co_read_response, co_write_request, s);
socket_set_cork(s->fd, 1);
/* send a header */
}
out:
socket_set_cork(s->fd, 0);
- qemu_aio_set_fd_handler(s->fd, co_read_response, NULL, s);
+ aio_set_fd_handler(s->aio_context, s->fd, co_read_response, NULL, s);
s->co_send = NULL;
qemu_co_mutex_unlock(&s->lock);
}
-static int read_write_object(int fd, char *buf, uint64_t oid, uint8_t copies,
+static int read_write_object(int fd, AioContext *aio_context, char *buf,
+ uint64_t oid, uint8_t copies,
unsigned int datalen, uint64_t offset,
bool write, bool create, uint32_t cache_flags)
{
hdr.offset = offset;
hdr.copies = copies;
- ret = do_req(fd, (SheepdogReq *)&hdr, buf, &wlen, &rlen);
+ ret = do_req(fd, aio_context, (SheepdogReq *)&hdr, buf, &wlen, &rlen);
if (ret) {
error_report("failed to send a request to the sheep");
return ret;
}
}
-static int read_object(int fd, char *buf, uint64_t oid, uint8_t copies,
+static int read_object(int fd, AioContext *aio_context, char *buf,
+ uint64_t oid, uint8_t copies,
unsigned int datalen, uint64_t offset,
uint32_t cache_flags)
{
- return read_write_object(fd, buf, oid, copies, datalen, offset, false,
+ return read_write_object(fd, aio_context, buf, oid, copies,
+ datalen, offset, false,
false, cache_flags);
}
-static int write_object(int fd, char *buf, uint64_t oid, uint8_t copies,
+static int write_object(int fd, AioContext *aio_context, char *buf,
+ uint64_t oid, uint8_t copies,
unsigned int datalen, uint64_t offset, bool create,
uint32_t cache_flags)
{
- return read_write_object(fd, buf, oid, copies, datalen, offset, true,
+ return read_write_object(fd, aio_context, buf, oid, copies,
+ datalen, offset, true,
create, cache_flags);
}
fd = connect_to_sdog(s, &local_err);
if (fd < 0) {
- qerror_report_err(local_err);
- error_free(local_err);
+ error_report_err(local_err);
return -EIO;
}
- inode = g_malloc(sizeof(s->inode));
+ inode = g_malloc(SD_INODE_HEADER_SIZE);
ret = find_vdi_name(s, s->name, snapid, tag, &vid, false, &local_err);
if (ret) {
- qerror_report_err(local_err);
- error_free(local_err);
+ error_report_err(local_err);
goto out;
}
- ret = read_object(fd, (char *)inode, vid_to_vdi_oid(vid),
- s->inode.nr_copies, sizeof(*inode), 0, s->cache_flags);
+ ret = read_object(fd, s->aio_context, (char *)inode, vid_to_vdi_oid(vid),
+ s->inode.nr_copies, SD_INODE_HEADER_SIZE, 0,
+ s->cache_flags);
if (ret < 0) {
goto out;
}
if (inode->vdi_id != s->inode.vdi_id) {
- memcpy(&s->inode, inode, sizeof(s->inode));
+ memcpy(&s->inode, inode, SD_INODE_HEADER_SIZE);
}
out:
return ret;
}
-/* Return true if the specified request is linked to the pending list. */
-static bool check_simultaneous_create(BDRVSheepdogState *s, AIOReq *aio_req)
-{
- AIOReq *areq;
- QLIST_FOREACH(areq, &s->inflight_aio_head, aio_siblings) {
- if (areq != aio_req && areq->oid == aio_req->oid) {
- /*
- * Sheepdog cannot handle simultaneous create requests to the same
- * object, so we cannot send the request until the previous request
- * finishes.
- */
- DPRINTF("simultaneous create to %" PRIx64 "\n", aio_req->oid);
- aio_req->flags = 0;
- aio_req->base_oid = 0;
- QLIST_REMOVE(aio_req, aio_siblings);
- QLIST_INSERT_HEAD(&s->pending_aio_head, aio_req, aio_siblings);
- return true;
- }
- }
-
- return false;
-}
-
static void coroutine_fn resend_aioreq(BDRVSheepdogState *s, AIOReq *aio_req)
{
SheepdogAIOCB *acb = aio_req->aiocb;
- bool create = false;
+
+ aio_req->create = false;
/* check whether this request becomes a CoW one */
if (acb->aiocb_type == AIOCB_WRITE_UDATA && is_data_obj(aio_req->oid)) {
goto out;
}
- if (check_simultaneous_create(s, aio_req)) {
- return;
- }
-
if (s->inode.data_vdi_id[idx]) {
aio_req->base_oid = vid_to_data_oid(s->inode.data_vdi_id[idx], idx);
aio_req->flags |= SD_FLAG_CMD_COW;
}
- create = true;
+ aio_req->create = true;
}
out:
if (is_data_obj(aio_req->oid)) {
- add_aio_request(s, aio_req, acb->qiov->iov, acb->qiov->niov, create,
+ add_aio_request(s, aio_req, acb->qiov->iov, acb->qiov->niov,
acb->aiocb_type);
} else {
struct iovec iov;
iov.iov_base = &s->inode;
iov.iov_len = sizeof(s->inode);
- add_aio_request(s, aio_req, &iov, 1, false, AIOCB_WRITE_UDATA);
+ add_aio_request(s, aio_req, &iov, 1, AIOCB_WRITE_UDATA);
}
}
+static void sd_detach_aio_context(BlockDriverState *bs)
+{
+ BDRVSheepdogState *s = bs->opaque;
+
+ aio_set_fd_handler(s->aio_context, s->fd, NULL, NULL, NULL);
+}
+
+static void sd_attach_aio_context(BlockDriverState *bs,
+ AioContext *new_context)
+{
+ BDRVSheepdogState *s = bs->opaque;
+
+ s->aio_context = new_context;
+ aio_set_fd_handler(new_context, s->fd, co_read_response, NULL, s);
+}
+
/* TODO Convert to fine grained options */
static QemuOptsList runtime_opts = {
.name = "sheepdog",
const char *filename;
s->bs = bs;
+ s->aio_context = bdrv_get_aio_context(bs);
opts = qemu_opts_create(&runtime_opts, NULL, 0, &error_abort);
qemu_opts_absorb_qdict(opts, options, &local_err);
if (local_err) {
- qerror_report_err(local_err);
- error_free(local_err);
+ error_propagate(errp, local_err);
ret = -EINVAL;
goto out;
}
filename = qemu_opt_get(opts, "filename");
QLIST_INIT(&s->inflight_aio_head);
- QLIST_INIT(&s->pending_aio_head);
QLIST_INIT(&s->failed_aio_head);
+ QLIST_INIT(&s->inflight_aiocb_head);
s->fd = -1;
memset(vdi, 0, sizeof(vdi));
ret = parse_vdiname(s, filename, vdi, &snapid, tag);
}
if (ret < 0) {
+ error_setg(errp, "Can't parse filename");
goto out;
}
- s->fd = get_sheep_fd(s, &local_err);
+ s->fd = get_sheep_fd(s, errp);
if (s->fd < 0) {
- qerror_report_err(local_err);
- error_free(local_err);
ret = s->fd;
goto out;
}
- ret = find_vdi_name(s, vdi, snapid, tag, &vid, true, &local_err);
+ ret = find_vdi_name(s, vdi, snapid, tag, &vid, true, errp);
if (ret) {
- qerror_report_err(local_err);
- error_free(local_err);
goto out;
}
s->is_snapshot = true;
}
- fd = connect_to_sdog(s, &local_err);
+ fd = connect_to_sdog(s, errp);
if (fd < 0) {
- qerror_report_err(local_err);
- error_free(local_err);
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,
- s->cache_flags);
+ ret = read_object(fd, s->aio_context, buf, vid_to_vdi_oid(vid),
+ 0, SD_INODE_SIZE, 0, s->cache_flags);
closesocket(fd);
if (ret) {
+ error_setg(errp, "Can't read snapshot inode");
goto out;
}
bs->total_sectors = s->inode.vdi_size / BDRV_SECTOR_SIZE;
pstrcpy(s->name, sizeof(s->name), vdi);
qemu_co_mutex_init(&s->lock);
+ qemu_co_queue_init(&s->overwrapping_queue);
qemu_opts_del(opts);
g_free(buf);
return 0;
out:
- qemu_aio_set_fd_handler(s->fd, NULL, NULL, NULL);
+ aio_set_fd_handler(bdrv_get_aio_context(bs), s->fd, NULL, NULL, NULL);
if (s->fd >= 0) {
closesocket(s->fd);
}
hdr.vdi_size = s->inode.vdi_size;
hdr.copy_policy = s->inode.copy_policy;
hdr.copies = s->inode.nr_copies;
+ hdr.block_size_shift = s->inode.block_size_shift;
- ret = do_req(fd, (SheepdogReq *)&hdr, buf, &wlen, &rlen);
+ ret = do_req(fd, s->aio_context, (SheepdogReq *)&hdr, buf, &wlen, &rlen);
closesocket(fd);
static int sd_prealloc(const char *filename, Error **errp)
{
BlockDriverState *bs = NULL;
+ BDRVSheepdogState *base = NULL;
+ unsigned long buf_size;
uint32_t idx, max_idx;
+ uint32_t object_size;
int64_t vdi_size;
- void *buf = g_malloc0(SD_DATA_OBJ_SIZE);
+ void *buf = NULL;
int ret;
ret = bdrv_open(&bs, filename, NULL, NULL, BDRV_O_RDWR | BDRV_O_PROTOCOL,
- NULL, errp);
+ errp);
if (ret < 0) {
goto out_with_err_set;
}
ret = vdi_size;
goto out;
}
- max_idx = DIV_ROUND_UP(vdi_size, SD_DATA_OBJ_SIZE);
+
+ base = bs->opaque;
+ object_size = (UINT32_C(1) << base->inode.block_size_shift);
+ buf_size = MIN(object_size, SD_DATA_OBJ_SIZE);
+ buf = g_malloc0(buf_size);
+
+ max_idx = DIV_ROUND_UP(vdi_size, buf_size);
for (idx = 0; idx < max_idx; idx++) {
/*
* The created image can be a cloned image, so we need to read
* a data from the source image.
*/
- ret = bdrv_pread(bs, idx * SD_DATA_OBJ_SIZE, buf, SD_DATA_OBJ_SIZE);
+ ret = bdrv_pread(bs, idx * buf_size, buf, buf_size);
if (ret < 0) {
goto out;
}
- ret = bdrv_pwrite(bs, idx * SD_DATA_OBJ_SIZE, buf, SD_DATA_OBJ_SIZE);
+ ret = bdrv_pwrite(bs, idx * buf_size, buf, buf_size);
if (ret < 0) {
goto out;
}
return 0;
}
-static int sd_create(const char *filename, QEMUOptionParameter *options,
+static int parse_block_size_shift(BDRVSheepdogState *s, QemuOpts *opt)
+{
+ struct SheepdogInode *inode = &s->inode;
+ uint64_t object_size;
+ int obj_order;
+
+ object_size = qemu_opt_get_size_del(opt, BLOCK_OPT_OBJECT_SIZE, 0);
+ if (object_size) {
+ if ((object_size - 1) & object_size) { /* not a power of 2? */
+ return -EINVAL;
+ }
+ obj_order = ctz32(object_size);
+ if (obj_order < 20 || obj_order > 31) {
+ return -EINVAL;
+ }
+ inode->block_size_shift = (uint8_t)obj_order;
+ }
+
+ return 0;
+}
+
+static int sd_create(const char *filename, QemuOpts *opts,
Error **errp)
{
int ret = 0;
uint32_t vid = 0;
char *backing_file = NULL;
+ char *buf = NULL;
BDRVSheepdogState *s;
char tag[SD_MAX_VDI_TAG_LEN];
uint32_t snapid;
+ uint64_t max_vdi_size;
bool prealloc = false;
- Error *local_err = NULL;
- s = g_malloc0(sizeof(BDRVSheepdogState));
+ s = g_new0(BDRVSheepdogState, 1);
memset(tag, 0, sizeof(tag));
if (strstr(filename, "://")) {
ret = parse_vdiname(s, filename, s->name, &snapid, tag);
}
if (ret < 0) {
+ error_setg(errp, "Can't parse filename");
goto out;
}
- while (options && options->name) {
- if (!strcmp(options->name, BLOCK_OPT_SIZE)) {
- s->inode.vdi_size = options->value.n;
- } else if (!strcmp(options->name, BLOCK_OPT_BACKING_FILE)) {
- backing_file = options->value.s;
- } else if (!strcmp(options->name, BLOCK_OPT_PREALLOC)) {
- if (!options->value.s || !strcmp(options->value.s, "off")) {
- prealloc = false;
- } else if (!strcmp(options->value.s, "full")) {
- prealloc = true;
- } else {
- error_report("Invalid preallocation mode: '%s'",
- options->value.s);
- ret = -EINVAL;
- goto out;
- }
- } else if (!strcmp(options->name, BLOCK_OPT_REDUNDANCY)) {
- if (options->value.s) {
- ret = parse_redundancy(s, options->value.s);
- if (ret < 0) {
- goto out;
- }
- }
- }
- options++;
+ s->inode.vdi_size = ROUND_UP(qemu_opt_get_size_del(opts, BLOCK_OPT_SIZE, 0),
+ BDRV_SECTOR_SIZE);
+ backing_file = qemu_opt_get_del(opts, BLOCK_OPT_BACKING_FILE);
+ buf = qemu_opt_get_del(opts, BLOCK_OPT_PREALLOC);
+ if (!buf || !strcmp(buf, "off")) {
+ prealloc = false;
+ } else if (!strcmp(buf, "full")) {
+ prealloc = true;
+ } else {
+ error_setg(errp, "Invalid preallocation mode: '%s'", buf);
+ ret = -EINVAL;
+ goto out;
}
- if (s->inode.vdi_size > SD_MAX_VDI_SIZE) {
- error_report("too big image size");
- ret = -EINVAL;
+ g_free(buf);
+ buf = qemu_opt_get_del(opts, BLOCK_OPT_REDUNDANCY);
+ if (buf) {
+ ret = parse_redundancy(s, buf);
+ if (ret < 0) {
+ error_setg(errp, "Invalid redundancy mode: '%s'", buf);
+ goto out;
+ }
+ }
+ ret = parse_block_size_shift(s, opts);
+ if (ret < 0) {
+ error_setg(errp, "Invalid object_size."
+ " obect_size needs to be power of 2"
+ " and be limited from 2^20 to 2^31");
goto out;
}
BlockDriver *drv;
/* Currently, only Sheepdog backing image is supported. */
- drv = bdrv_find_protocol(backing_file, true);
+ drv = bdrv_find_protocol(backing_file, true, NULL);
if (!drv || strcmp(drv->protocol_name, "sheepdog") != 0) {
- error_report("backing_file must be a sheepdog image");
+ error_setg(errp, "backing_file must be a sheepdog image");
ret = -EINVAL;
goto out;
}
bs = NULL;
- ret = bdrv_open(&bs, backing_file, NULL, NULL, BDRV_O_PROTOCOL, NULL,
- &local_err);
+ ret = bdrv_open(&bs, backing_file, NULL, NULL, BDRV_O_PROTOCOL, errp);
if (ret < 0) {
- qerror_report_err(local_err);
- error_free(local_err);
goto out;
}
base = bs->opaque;
if (!is_snapshot(&base->inode)) {
- error_report("cannot clone from a non snapshot vdi");
+ error_setg(errp, "cannot clone from a non snapshot vdi");
bdrv_unref(bs);
ret = -EINVAL;
goto out;
bdrv_unref(bs);
}
- ret = do_sd_create(s, &vid, 0, &local_err);
+ s->aio_context = qemu_get_aio_context();
+
+ /* if block_size_shift is not specified, get cluster default value */
+ if (s->inode.block_size_shift == 0) {
+ SheepdogVdiReq hdr;
+ SheepdogClusterRsp *rsp = (SheepdogClusterRsp *)&hdr;
+ Error *local_err = NULL;
+ int fd;
+ unsigned int wlen = 0, rlen = 0;
+
+ fd = connect_to_sdog(s, &local_err);
+ if (fd < 0) {
+ error_report("%s", error_get_pretty(local_err));
+ error_free(local_err);
+ ret = -EIO;
+ goto out;
+ }
+
+ memset(&hdr, 0, sizeof(hdr));
+ hdr.opcode = SD_OP_GET_CLUSTER_DEFAULT;
+ hdr.proto_ver = SD_PROTO_VER;
+
+ ret = do_req(fd, s->aio_context, (SheepdogReq *)&hdr,
+ NULL, &wlen, &rlen);
+ closesocket(fd);
+ if (ret) {
+ error_setg_errno(errp, -ret, "failed to get cluster default");
+ goto out;
+ }
+ if (rsp->result == SD_RES_SUCCESS) {
+ s->inode.block_size_shift = rsp->block_size_shift;
+ } else {
+ s->inode.block_size_shift = SD_DEFAULT_BLOCK_SIZE_SHIFT;
+ }
+ }
+
+ max_vdi_size = (UINT64_C(1) << s->inode.block_size_shift) * MAX_DATA_OBJS;
+
+ if (s->inode.vdi_size > max_vdi_size) {
+ error_setg(errp, "An image is too large."
+ " The maximum image size is %"PRIu64 "GB",
+ max_vdi_size / 1024 / 1024 / 1024);
+ ret = -EINVAL;
+ goto out;
+ }
+
+ ret = do_sd_create(s, &vid, 0, errp);
if (ret) {
- qerror_report_err(local_err);
- error_free(local_err);
goto out;
}
if (prealloc) {
- ret = sd_prealloc(filename, &local_err);
- if (ret < 0) {
- qerror_report_err(local_err);
- error_free(local_err);
- }
+ ret = sd_prealloc(filename, errp);
}
out:
+ g_free(backing_file);
+ g_free(buf);
g_free(s);
return ret;
}
fd = connect_to_sdog(s, &local_err);
if (fd < 0) {
- qerror_report_err(local_err);
- error_free(local_err);
+ error_report_err(local_err);
return;
}
memset(&hdr, 0, sizeof(hdr));
hdr.opcode = SD_OP_RELEASE_VDI;
+ hdr.type = LOCK_TYPE_NORMAL;
hdr.base_vdi_id = s->inode.vdi_id;
wlen = strlen(s->name) + 1;
hdr.data_length = wlen;
hdr.flags = SD_FLAG_CMD_WRITE;
- ret = do_req(fd, (SheepdogReq *)&hdr, s->name, &wlen, &rlen);
+ ret = do_req(fd, s->aio_context, (SheepdogReq *)&hdr,
+ s->name, &wlen, &rlen);
closesocket(fd);
error_report("%s, %s", sd_strerror(rsp->result), s->name);
}
- qemu_aio_set_fd_handler(s->fd, NULL, NULL, NULL);
+ aio_set_fd_handler(bdrv_get_aio_context(bs), s->fd, NULL, NULL, NULL);
closesocket(s->fd);
g_free(s->host_spec);
}
BDRVSheepdogState *s = bs->opaque;
int ret, fd;
unsigned int datalen;
+ uint64_t max_vdi_size;
+ max_vdi_size = (UINT64_C(1) << s->inode.block_size_shift) * MAX_DATA_OBJS;
if (offset < s->inode.vdi_size) {
error_report("shrinking is not supported");
return -EINVAL;
- } else if (offset > SD_MAX_VDI_SIZE) {
+ } else if (offset > max_vdi_size) {
error_report("too big image size");
return -EINVAL;
}
fd = connect_to_sdog(s, &local_err);
if (fd < 0) {
- qerror_report_err(local_err);
- error_free(local_err);
+ error_report_err(local_err);
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, false, s->cache_flags);
+ ret = write_object(fd, s->aio_context, (char *)&s->inode,
+ vid_to_vdi_oid(s->inode.vdi_id), s->inode.nr_copies,
+ datalen, 0, false, s->cache_flags);
close(fd);
if (ret < 0) {
iov.iov_base = &s->inode;
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);
+ data_len, offset, 0, false, 0, offset);
QLIST_INSERT_HEAD(&s->inflight_aio_head, aio_req, aio_siblings);
- add_aio_request(s, aio_req, &iov, 1, false, AIOCB_WRITE_UDATA);
+ add_aio_request(s, aio_req, &iov, 1, AIOCB_WRITE_UDATA);
acb->aio_done_func = sd_finish_aiocb;
acb->aiocb_type = AIOCB_WRITE_UDATA;
fd = connect_to_sdog(s, &local_err);
if (fd < 0) {
- qerror_report_err(local_err);
- error_free(local_err);
+ error_report_err(local_err);
return false;
}
- ret = do_req(fd, (SheepdogReq *)&hdr, s->name, &wlen, &rlen);
+ ret = do_req(fd, s->aio_context, (SheepdogReq *)&hdr,
+ s->name, &wlen, &rlen);
closesocket(fd);
if (ret) {
return false;
deleted = sd_delete(s);
ret = do_sd_create(s, &vid, !deleted, &local_err);
if (ret) {
- qerror_report_err(local_err);
- error_free(local_err);
+ error_report_err(local_err);
goto out;
}
fd = connect_to_sdog(s, &local_err);
if (fd < 0) {
- qerror_report_err(local_err);
- error_free(local_err);
+ error_report_err(local_err);
ret = fd;
goto out;
}
- ret = read_object(fd, buf, vid_to_vdi_oid(vid), s->inode.nr_copies,
- SD_INODE_SIZE, 0, s->cache_flags);
+ ret = read_object(fd, s->aio_context, buf, vid_to_vdi_oid(vid),
+ s->inode.nr_copies, SD_INODE_SIZE, 0, s->cache_flags);
closesocket(fd);
SheepdogAIOCB *acb = p;
int ret = 0;
unsigned long len, done = 0, total = acb->nb_sectors * BDRV_SECTOR_SIZE;
- unsigned long idx = acb->sector_num * BDRV_SECTOR_SIZE / SD_DATA_OBJ_SIZE;
+ unsigned long idx;
+ uint32_t object_size;
uint64_t oid;
- uint64_t offset = (acb->sector_num * BDRV_SECTOR_SIZE) % SD_DATA_OBJ_SIZE;
+ uint64_t offset;
BDRVSheepdogState *s = acb->common.bs->opaque;
SheepdogInode *inode = &s->inode;
AIOReq *aio_req;
}
}
+ object_size = (UINT32_C(1) << inode->block_size_shift);
+ idx = acb->sector_num * BDRV_SECTOR_SIZE / object_size;
+ offset = (acb->sector_num * BDRV_SECTOR_SIZE) % object_size;
+
/*
* 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.
oid = vid_to_data_oid(inode->data_vdi_id[idx], idx);
- len = MIN(total - done, SD_DATA_OBJ_SIZE - offset);
+ len = MIN(total - done, object_size - offset);
switch (acb->aiocb_type) {
case AIOCB_READ_UDATA:
* We discard the object only when the whole object is
* 1) allocated 2) trimmed. Otherwise, simply skip it.
*/
- if (len != SD_DATA_OBJ_SIZE || inode->data_vdi_id[idx] == 0) {
+ if (len != object_size || inode->data_vdi_id[idx] == 0) {
goto done;
}
break;
DPRINTF("new oid %" PRIx64 "\n", oid);
}
- aio_req = alloc_aio_req(s, acb, oid, len, offset, flags, old_oid, done);
+ aio_req = alloc_aio_req(s, acb, oid, len, offset, flags, create,
+ old_oid, done);
QLIST_INSERT_HEAD(&s->inflight_aio_head, aio_req, aio_siblings);
- if (create) {
- if (check_simultaneous_create(s, aio_req)) {
- goto done;
- }
- }
-
- add_aio_request(s, aio_req, acb->qiov->iov, acb->qiov->niov, create,
+ add_aio_request(s, aio_req, acb->qiov->iov, acb->qiov->niov,
acb->aiocb_type);
done:
offset = 0;
return 1;
}
+static bool check_overwrapping_aiocb(BDRVSheepdogState *s, SheepdogAIOCB *aiocb)
+{
+ SheepdogAIOCB *cb;
+
+ QLIST_FOREACH(cb, &s->inflight_aiocb_head, aiocb_siblings) {
+ if (AIOCBOverwrapping(aiocb, cb)) {
+ return true;
+ }
+ }
+
+ QLIST_INSERT_HEAD(&s->inflight_aiocb_head, aiocb, aiocb_siblings);
+ return false;
+}
+
static coroutine_fn int sd_co_writev(BlockDriverState *bs, int64_t sector_num,
int nb_sectors, QEMUIOVector *qiov)
{
int64_t offset = (sector_num + nb_sectors) * BDRV_SECTOR_SIZE;
BDRVSheepdogState *s = bs->opaque;
- if (bs->growable && offset > s->inode.vdi_size) {
+ if (offset > s->inode.vdi_size) {
ret = sd_truncate(bs, offset);
if (ret < 0) {
return ret;
acb->aio_done_func = sd_write_done;
acb->aiocb_type = AIOCB_WRITE_UDATA;
+retry:
+ if (check_overwrapping_aiocb(s, acb)) {
+ qemu_co_queue_wait(&s->overwrapping_queue);
+ goto retry;
+ }
+
ret = sd_co_rw_vector(acb);
if (ret <= 0) {
- qemu_aio_release(acb);
+ QLIST_REMOVE(acb, aiocb_siblings);
+ qemu_co_queue_restart_all(&s->overwrapping_queue);
+ qemu_aio_unref(acb);
return ret;
}
qemu_coroutine_yield();
+ QLIST_REMOVE(acb, aiocb_siblings);
+ qemu_co_queue_restart_all(&s->overwrapping_queue);
+
return acb->ret;
}
{
SheepdogAIOCB *acb;
int ret;
+ BDRVSheepdogState *s = bs->opaque;
acb = sd_aio_setup(bs, qiov, sector_num, nb_sectors);
acb->aiocb_type = AIOCB_READ_UDATA;
acb->aio_done_func = sd_finish_aiocb;
+retry:
+ if (check_overwrapping_aiocb(s, acb)) {
+ qemu_co_queue_wait(&s->overwrapping_queue);
+ goto retry;
+ }
+
ret = sd_co_rw_vector(acb);
if (ret <= 0) {
- qemu_aio_release(acb);
+ QLIST_REMOVE(acb, aiocb_siblings);
+ qemu_co_queue_restart_all(&s->overwrapping_queue);
+ qemu_aio_unref(acb);
return ret;
}
qemu_coroutine_yield();
+ QLIST_REMOVE(acb, aiocb_siblings);
+ qemu_co_queue_restart_all(&s->overwrapping_queue);
return acb->ret;
}
acb->aio_done_func = sd_finish_aiocb;
aio_req = alloc_aio_req(s, acb, vid_to_vdi_oid(s->inode.vdi_id),
- 0, 0, 0, 0, 0);
+ 0, 0, 0, false, 0, 0);
QLIST_INSERT_HEAD(&s->inflight_aio_head, aio_req, aio_siblings);
- add_aio_request(s, aio_req, NULL, 0, false, acb->aiocb_type);
+ add_aio_request(s, aio_req, NULL, 0, acb->aiocb_type);
qemu_coroutine_yield();
return acb->ret;
strncpy(s->inode.tag, sn_info->name, sizeof(s->inode.tag));
/* we don't need to update entire object */
datalen = SD_INODE_SIZE - sizeof(s->inode.data_vdi_id);
+ inode = g_malloc(datalen);
/* refresh inode. */
fd = connect_to_sdog(s, &local_err);
if (fd < 0) {
- qerror_report_err(local_err);
- error_free(local_err);
+ error_report_err(local_err);
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, false, s->cache_flags);
+ ret = write_object(fd, s->aio_context, (char *)&s->inode,
+ vid_to_vdi_oid(s->inode.vdi_id), s->inode.nr_copies,
+ datalen, 0, false, s->cache_flags);
if (ret < 0) {
error_report("failed to write snapshot's inode.");
goto cleanup;
ret = do_sd_create(s, &new_vid, 1, &local_err);
if (ret < 0) {
- qerror_report_err(local_err);
+ error_report("failed to create inode for snapshot: %s",
+ error_get_pretty(local_err));
error_free(local_err);
- error_report("failed to create inode for snapshot. %s",
- strerror(errno));
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->cache_flags);
+ ret = read_object(fd, s->aio_context, (char *)inode,
+ vid_to_vdi_oid(new_vid), s->inode.nr_copies, datalen, 0,
+ s->cache_flags);
if (ret < 0) {
error_report("failed to read new inode info. %s", strerror(errno));
s->inode.name, s->inode.snap_id, s->inode.vdi_id);
cleanup:
+ g_free(inode);
closesocket(fd);
return ret;
}
uint32_t snapid = 0;
int ret = 0;
- old_s = g_malloc(sizeof(BDRVSheepdogState));
+ old_s = g_new(BDRVSheepdogState, 1);
memcpy(old_s, s, sizeof(BDRVSheepdogState));
fd = connect_to_sdog(s, &local_err);
if (fd < 0) {
- qerror_report_err(local_err);
- error_free(local_err);
+ error_report_err(local_err);
ret = fd;
goto out;
}
req.opcode = SD_OP_READ_VDIS;
req.data_length = max;
- ret = do_req(fd, (SheepdogReq *)&req, vdi_inuse, &wlen, &rlen);
+ ret = do_req(fd, s->aio_context, (SheepdogReq *)&req,
+ vdi_inuse, &wlen, &rlen);
closesocket(fd);
if (ret) {
goto out;
}
- sn_tab = g_malloc0(nr * sizeof(*sn_tab));
+ sn_tab = g_new0(QEMUSnapshotInfo, nr);
/* calculate a vdi id with hash function */
hval = fnv_64a_buf(s->name, strlen(s->name), FNV1A_64_INIT);
fd = connect_to_sdog(s, &local_err);
if (fd < 0) {
- qerror_report_err(local_err);
- error_free(local_err);
+ error_report_err(local_err);
ret = fd;
goto out;
}
}
/* we don't need to read entire object */
- ret = read_object(fd, (char *)&inode, vid_to_vdi_oid(vid),
+ ret = read_object(fd, s->aio_context, (char *)&inode,
+ vid_to_vdi_oid(vid),
0, SD_INODE_SIZE - sizeof(inode.data_vdi_id), 0,
s->cache_flags);
uint64_t offset;
uint32_t vdi_index;
uint32_t vdi_id = load ? s->inode.parent_vdi_id : s->inode.vdi_id;
+ uint32_t object_size = (UINT32_C(1) << s->inode.block_size_shift);
fd = connect_to_sdog(s, &local_err);
if (fd < 0) {
- qerror_report_err(local_err);
- error_free(local_err);
+ error_report_err(local_err);
return fd;
}
while (remaining) {
- vdi_index = pos / SD_DATA_OBJ_SIZE;
- offset = pos % SD_DATA_OBJ_SIZE;
+ vdi_index = pos / object_size;
+ offset = pos % object_size;
- data_len = MIN(remaining, SD_DATA_OBJ_SIZE - offset);
+ data_len = MIN(remaining, object_size - offset);
vmstate_oid = vid_to_vmstate_oid(vdi_id, vdi_index);
create = (offset == 0);
if (load) {
- ret = read_object(fd, (char *)data, vmstate_oid,
+ ret = read_object(fd, s->aio_context, (char *)data, vmstate_oid,
s->inode.nr_copies, data_len, offset,
s->cache_flags);
} else {
- ret = write_object(fd, (char *)data, vmstate_oid,
+ ret = write_object(fd, s->aio_context, (char *)data, vmstate_oid,
s->inode.nr_copies, data_len, offset, create,
s->cache_flags);
}
acb->aiocb_type = AIOCB_DISCARD_OBJ;
acb->aio_done_func = sd_finish_aiocb;
+retry:
+ if (check_overwrapping_aiocb(s, acb)) {
+ qemu_co_queue_wait(&s->overwrapping_queue);
+ goto retry;
+ }
+
ret = sd_co_rw_vector(acb);
if (ret <= 0) {
- qemu_aio_release(acb);
+ QLIST_REMOVE(acb, aiocb_siblings);
+ qemu_co_queue_restart_all(&s->overwrapping_queue);
+ qemu_aio_unref(acb);
return ret;
}
qemu_coroutine_yield();
+ QLIST_REMOVE(acb, aiocb_siblings);
+ qemu_co_queue_restart_all(&s->overwrapping_queue);
+
return acb->ret;
}
{
BDRVSheepdogState *s = bs->opaque;
SheepdogInode *inode = &s->inode;
+ uint32_t object_size = (UINT32_C(1) << inode->block_size_shift);
uint64_t offset = sector_num * BDRV_SECTOR_SIZE;
- unsigned long start = offset / SD_DATA_OBJ_SIZE,
+ unsigned long start = offset / object_size,
end = DIV_ROUND_UP((sector_num + nb_sectors) *
- BDRV_SECTOR_SIZE, SD_DATA_OBJ_SIZE);
+ BDRV_SECTOR_SIZE, object_size);
unsigned long idx;
int64_t ret = BDRV_BLOCK_DATA | BDRV_BLOCK_OFFSET_VALID | offset;
}
}
- *pnum = (idx - start) * SD_DATA_OBJ_SIZE / BDRV_SECTOR_SIZE;
+ *pnum = (idx - start) * object_size / BDRV_SECTOR_SIZE;
if (*pnum > nb_sectors) {
*pnum = nb_sectors;
}
{
BDRVSheepdogState *s = bs->opaque;
SheepdogInode *inode = &s->inode;
- unsigned long i, last = DIV_ROUND_UP(inode->vdi_size, SD_DATA_OBJ_SIZE);
+ uint32_t object_size = (UINT32_C(1) << inode->block_size_shift);
+ unsigned long i, last = DIV_ROUND_UP(inode->vdi_size, object_size);
uint64_t size = 0;
for (i = 0; i < last; i++) {
if (inode->data_vdi_id[i] == 0) {
continue;
}
- size += SD_DATA_OBJ_SIZE;
+ size += object_size;
}
return size;
}
-static QEMUOptionParameter sd_create_options[] = {
- {
- .name = BLOCK_OPT_SIZE,
- .type = OPT_SIZE,
- .help = "Virtual disk size"
- },
- {
- .name = BLOCK_OPT_BACKING_FILE,
- .type = OPT_STRING,
- .help = "File name of a base image"
- },
- {
- .name = BLOCK_OPT_PREALLOC,
- .type = OPT_STRING,
- .help = "Preallocation mode (allowed values: off, full)"
- },
- {
- .name = BLOCK_OPT_REDUNDANCY,
- .type = OPT_STRING,
- .help = "Redundancy of the image"
- },
- { NULL }
+static QemuOptsList sd_create_opts = {
+ .name = "sheepdog-create-opts",
+ .head = QTAILQ_HEAD_INITIALIZER(sd_create_opts.head),
+ .desc = {
+ {
+ .name = BLOCK_OPT_SIZE,
+ .type = QEMU_OPT_SIZE,
+ .help = "Virtual disk size"
+ },
+ {
+ .name = BLOCK_OPT_BACKING_FILE,
+ .type = QEMU_OPT_STRING,
+ .help = "File name of a base image"
+ },
+ {
+ .name = BLOCK_OPT_PREALLOC,
+ .type = QEMU_OPT_STRING,
+ .help = "Preallocation mode (allowed values: off, full)"
+ },
+ {
+ .name = BLOCK_OPT_REDUNDANCY,
+ .type = QEMU_OPT_STRING,
+ .help = "Redundancy of the image"
+ },
+ {
+ .name = BLOCK_OPT_OBJECT_SIZE,
+ .type = QEMU_OPT_SIZE,
+ .help = "Object size of the image"
+ },
+ { /* end of list */ }
+ }
};
static BlockDriver bdrv_sheepdog = {
.bdrv_save_vmstate = sd_save_vmstate,
.bdrv_load_vmstate = sd_load_vmstate,
- .create_options = sd_create_options,
+ .bdrv_detach_aio_context = sd_detach_aio_context,
+ .bdrv_attach_aio_context = sd_attach_aio_context,
+
+ .create_opts = &sd_create_opts,
};
static BlockDriver bdrv_sheepdog_tcp = {
.bdrv_save_vmstate = sd_save_vmstate,
.bdrv_load_vmstate = sd_load_vmstate,
- .create_options = sd_create_options,
+ .bdrv_detach_aio_context = sd_detach_aio_context,
+ .bdrv_attach_aio_context = sd_attach_aio_context,
+
+ .create_opts = &sd_create_opts,
};
static BlockDriver bdrv_sheepdog_unix = {
.bdrv_save_vmstate = sd_save_vmstate,
.bdrv_load_vmstate = sd_load_vmstate,
- .create_options = sd_create_options,
+ .bdrv_detach_aio_context = sd_detach_aio_context,
+ .bdrv_attach_aio_context = sd_attach_aio_context,
+
+ .create_opts = &sd_create_opts,
};
static void bdrv_sheepdog_init(void)
projects / qemu.git / blobdiff