static int coroutine_fn bdrv_co_writev_em(BlockDriverState *bs,
int64_t sector_num, int nb_sectors,
QEMUIOVector *iov);
-static int coroutine_fn bdrv_co_do_readv(BlockDriverState *bs,
- int64_t sector_num, int nb_sectors, QEMUIOVector *qiov,
+static int coroutine_fn bdrv_co_do_preadv(BlockDriverState *bs,
+ int64_t offset, unsigned int bytes, QEMUIOVector *qiov,
BdrvRequestFlags flags);
-static int coroutine_fn bdrv_co_do_writev(BlockDriverState *bs,
- int64_t sector_num, int nb_sectors, QEMUIOVector *qiov,
+static int coroutine_fn bdrv_co_do_pwritev(BlockDriverState *bs,
+ int64_t offset, unsigned int bytes, QEMUIOVector *qiov,
BdrvRequestFlags flags);
static BlockDriverAIOCB *bdrv_co_aio_rw_vector(BlockDriverState *bs,
int64_t sector_num,
{
assert(!bs->io_limits_enabled);
throttle_init(&bs->throttle_state,
+ bdrv_get_aio_context(bs),
QEMU_CLOCK_VIRTUAL,
bdrv_throttle_read_timer_cb,
bdrv_throttle_write_timer_cb,
* @is_write: is the IO a write
*/
static void bdrv_io_limits_intercept(BlockDriverState *bs,
- int nb_sectors,
+ unsigned int bytes,
bool is_write)
{
/* does this io must wait */
}
/* the IO will be executed, do the accounting */
- throttle_account(&bs->throttle_state,
- is_write,
- nb_sectors * BDRV_SECTOR_SIZE);
+ throttle_account(&bs->throttle_state, is_write, bytes);
+
/* if the next request must wait -> do nothing */
if (throttle_schedule_timer(&bs->throttle_state, is_write)) {
}
/* create a new block device (by default it is empty) */
-BlockDriverState *bdrv_new(const char *device_name)
+BlockDriverState *bdrv_new(const char *device_name, Error **errp)
{
BlockDriverState *bs;
+ int i;
+
+ if (bdrv_find(device_name)) {
+ error_setg(errp, "Device with id '%s' already exists",
+ device_name);
+ return NULL;
+ }
+ if (bdrv_find_node(device_name)) {
+ error_setg(errp, "Device with node-name '%s' already exists",
+ device_name);
+ return NULL;
+ }
bs = g_malloc0(sizeof(BlockDriverState));
QLIST_INIT(&bs->dirty_bitmaps);
if (device_name[0] != '\0') {
QTAILQ_INSERT_TAIL(&bdrv_states, bs, device_list);
}
+ for (i = 0; i < BLOCK_OP_TYPE_MAX; i++) {
+ QLIST_INIT(&bs->op_blockers[i]);
+ }
bdrv_iostatus_disable(bs);
notifier_list_init(&bs->close_notifiers);
notifier_with_return_list_init(&bs->before_write_notifiers);
qemu_co_queue_init(&bs->throttled_reqs[0]);
qemu_co_queue_init(&bs->throttled_reqs[1]);
bs->refcnt = 1;
+ bs->aio_context = qemu_get_aio_context();
return bs;
}
assert(cco->drv);
ret = cco->drv->bdrv_create(cco->filename, cco->options, &local_err);
- if (error_is_set(&local_err)) {
+ if (local_err) {
error_propagate(&cco->err, local_err);
}
cco->ret = ret;
ret = cco.ret;
if (ret < 0) {
- if (error_is_set(&cco.err)) {
+ if (cco.err) {
error_propagate(errp, cco.err);
} else {
error_setg_errno(errp, -ret, "Could not create image");
}
ret = bdrv_create(drv, filename, options, &local_err);
- if (error_is_set(&local_err)) {
+ if (local_err) {
error_propagate(errp, local_err);
}
return ret;
int fd;
const char *tmpdir;
tmpdir = getenv("TMPDIR");
- if (!tmpdir)
- tmpdir = "/tmp";
+ if (!tmpdir) {
+ tmpdir = "/var/tmp";
+ }
if (snprintf(filename, size, "%s/vl.XXXXXX", tmpdir) >= size) {
return -EOVERFLOW;
}
bs->copy_on_read--;
}
+/*
+ * Returns the flags that a temporary snapshot should get, based on the
+ * originally requested flags (the originally requested image will have flags
+ * like a backing file)
+ */
+static int bdrv_temp_snapshot_flags(int flags)
+{
+ return (flags & ~BDRV_O_SNAPSHOT) | BDRV_O_TEMPORARY;
+}
+
+/*
+ * Returns the flags that bs->file should get, based on the given flags for
+ * the parent BDS
+ */
+static int bdrv_inherited_flags(int flags)
+{
+ /* Enable protocol handling, disable format probing for bs->file */
+ flags |= BDRV_O_PROTOCOL;
+
+ /* Our block drivers take care to send flushes and respect unmap policy,
+ * so we can enable both unconditionally on lower layers. */
+ flags |= BDRV_O_CACHE_WB | BDRV_O_UNMAP;
+
+ /* Clear flags that only apply to the top layer */
+ flags &= ~(BDRV_O_SNAPSHOT | BDRV_O_NO_BACKING | BDRV_O_COPY_ON_READ);
+
+ return flags;
+}
+
+/*
+ * Returns the flags that bs->backing_hd should get, based on the given flags
+ * for the parent BDS
+ */
+static int bdrv_backing_flags(int flags)
+{
+ /* backing files always opened read-only */
+ flags &= ~(BDRV_O_RDWR | BDRV_O_COPY_ON_READ);
+
+ /* snapshot=on is handled on the top layer */
+ flags &= ~(BDRV_O_SNAPSHOT | BDRV_O_TEMPORARY);
+
+ return flags;
+}
+
static int bdrv_open_flags(BlockDriverState *bs, int flags)
{
int open_flags = flags | BDRV_O_CACHE_WB;
/*
* Snapshots should be writable.
*/
- if (bs->is_temporary) {
+ if (flags & BDRV_O_TEMPORARY) {
open_flags |= BDRV_O_RDWR;
}
return open_flags;
}
-static int bdrv_assign_node_name(BlockDriverState *bs,
- const char *node_name,
- Error **errp)
+static void bdrv_assign_node_name(BlockDriverState *bs,
+ const char *node_name,
+ Error **errp)
{
if (!node_name) {
- return 0;
+ return;
}
/* empty string node name is invalid */
if (node_name[0] == '\0') {
error_setg(errp, "Empty node name");
- return -EINVAL;
+ return;
+ }
+
+ /* takes care of avoiding namespaces collisions */
+ if (bdrv_find(node_name)) {
+ error_setg(errp, "node-name=%s is conflicting with a device id",
+ node_name);
+ return;
}
/* takes care of avoiding duplicates node names */
if (bdrv_find_node(node_name)) {
error_setg(errp, "Duplicate node name");
- return -EINVAL;
+ return;
}
/* copy node name into the bs and insert it into the graph list */
pstrcpy(bs->node_name, sizeof(bs->node_name), node_name);
QTAILQ_INSERT_TAIL(&graph_bdrv_states, bs, node_list);
-
- return 0;
}
/*
filename = qdict_get_try_str(options, "filename");
}
+ if (drv->bdrv_needs_filename && !filename) {
+ error_setg(errp, "The '%s' block driver requires a file name",
+ drv->format_name);
+ return -EINVAL;
+ }
+
trace_bdrv_open_common(bs, filename ?: "", flags, drv->format_name);
node_name = qdict_get_try_str(options, "node-name");
- ret = bdrv_assign_node_name(bs, node_name, errp);
- if (ret < 0) {
- return ret;
+ bdrv_assign_node_name(bs, node_name, &local_err);
+ if (local_err) {
+ error_propagate(errp, local_err);
+ return -EINVAL;
}
qdict_del(options, "node-name");
}
if (ret < 0) {
- if (error_is_set(&local_err)) {
+ if (local_err) {
error_propagate(errp, local_err);
} else if (bs->filename[0]) {
error_setg_errno(errp, -ret, "Could not open '%s'", bs->filename);
bdrv_refresh_limits(bs);
assert(bdrv_opt_mem_align(bs) != 0);
- assert(bs->request_alignment != 0);
-
-#ifndef _WIN32
- if (bs->is_temporary) {
- assert(bs->filename[0] != '\0');
- unlink(bs->filename);
- }
-#endif
+ assert((bs->request_alignment != 0) || bs->sg);
return 0;
free_and_fail:
/*
* Opens a file using a protocol (file, host_device, nbd, ...)
*
- * options is a QDict of options to pass to the block drivers, or NULL for an
- * empty set of options. The reference to the QDict belongs to the block layer
- * after the call (even on failure), so if the caller intends to reuse the
- * dictionary, it needs to use QINCREF() before calling bdrv_file_open.
+ * options is an indirect pointer to a QDict of options to pass to the block
+ * drivers, or pointer to NULL for an empty set of options. If this function
+ * takes ownership of the QDict reference, it will set *options to NULL;
+ * otherwise, it will contain unused/unrecognized options after this function
+ * returns. Then, the caller is responsible for freeing it. If it intends to
+ * reuse the QDict, QINCREF() should be called beforehand.
*/
-int bdrv_file_open(BlockDriverState **pbs, const char *filename,
- const char *reference, QDict *options, int flags,
- Error **errp)
+static int bdrv_file_open(BlockDriverState *bs, const char *filename,
+ QDict **options, int flags, Error **errp)
{
- BlockDriverState *bs = NULL;
BlockDriver *drv;
const char *drvname;
- bool allow_protocol_prefix = false;
+ bool parse_filename = false;
Error *local_err = NULL;
int ret;
- /* NULL means an empty set of options */
- if (options == NULL) {
- options = qdict_new();
- }
-
- if (reference) {
- if (filename || qdict_size(options)) {
- error_setg(errp, "Cannot reference an existing block device with "
- "additional options or a new filename");
- return -EINVAL;
- }
- QDECREF(options);
-
- bs = bdrv_find(reference);
- if (!bs) {
- error_setg(errp, "Cannot find block device '%s'", reference);
- return -ENODEV;
- }
- bdrv_ref(bs);
- *pbs = bs;
- return 0;
- }
-
- bs = bdrv_new("");
- bs->options = options;
- options = qdict_clone_shallow(options);
-
/* Fetch the file name from the options QDict if necessary */
if (!filename) {
- filename = qdict_get_try_str(options, "filename");
- } else if (filename && !qdict_haskey(options, "filename")) {
- qdict_put(options, "filename", qstring_from_str(filename));
- allow_protocol_prefix = true;
+ filename = qdict_get_try_str(*options, "filename");
+ } else if (filename && !qdict_haskey(*options, "filename")) {
+ qdict_put(*options, "filename", qstring_from_str(filename));
+ parse_filename = true;
} else {
error_setg(errp, "Can't specify 'file' and 'filename' options at the "
"same time");
}
/* Find the right block driver */
- drvname = qdict_get_try_str(options, "driver");
+ drvname = qdict_get_try_str(*options, "driver");
if (drvname) {
drv = bdrv_find_format(drvname);
if (!drv) {
error_setg(errp, "Unknown driver '%s'", drvname);
}
- qdict_del(options, "driver");
+ qdict_del(*options, "driver");
} else if (filename) {
- drv = bdrv_find_protocol(filename, allow_protocol_prefix);
+ drv = bdrv_find_protocol(filename, parse_filename);
if (!drv) {
error_setg(errp, "Unknown protocol");
}
}
/* Parse the filename and open it */
- if (drv->bdrv_parse_filename && filename) {
- drv->bdrv_parse_filename(filename, options, &local_err);
- if (error_is_set(&local_err)) {
+ if (drv->bdrv_parse_filename && parse_filename) {
+ drv->bdrv_parse_filename(filename, *options, &local_err);
+ if (local_err) {
error_propagate(errp, local_err);
ret = -EINVAL;
goto fail;
}
- qdict_del(options, "filename");
- } else if (drv->bdrv_needs_filename && !filename) {
- error_setg(errp, "The '%s' block driver requires a file name",
- drv->format_name);
- ret = -EINVAL;
- goto fail;
+
+ if (!drv->bdrv_needs_filename) {
+ qdict_del(*options, "filename");
+ } else {
+ filename = qdict_get_str(*options, "filename");
+ }
}
if (!drv->bdrv_file_open) {
- ret = bdrv_open(bs, filename, options, flags, drv, &local_err);
- options = NULL;
+ ret = bdrv_open(&bs, filename, NULL, *options, flags, drv, &local_err);
+ *options = NULL;
} else {
- ret = bdrv_open_common(bs, NULL, options, flags, drv, &local_err);
+ ret = bdrv_open_common(bs, NULL, *options, flags, drv, &local_err);
}
if (ret < 0) {
error_propagate(errp, local_err);
goto fail;
}
- /* Check if any unknown options were used */
- if (options && (qdict_size(options) != 0)) {
- const QDictEntry *entry = qdict_first(options);
- error_setg(errp, "Block protocol '%s' doesn't support the option '%s'",
- drv->format_name, entry->key);
- ret = -EINVAL;
- goto fail;
- }
- QDECREF(options);
-
bs->growable = 1;
- *pbs = bs;
return 0;
fail:
- QDECREF(options);
- if (!bs->drv) {
- QDECREF(bs->options);
- }
- bdrv_unref(bs);
return ret;
}
+void bdrv_set_backing_hd(BlockDriverState *bs, BlockDriverState *backing_hd)
+{
+
+ if (bs->backing_hd) {
+ assert(bs->backing_blocker);
+ bdrv_op_unblock_all(bs->backing_hd, bs->backing_blocker);
+ } else if (backing_hd) {
+ error_setg(&bs->backing_blocker,
+ "device is used as backing hd of '%s'",
+ bs->device_name);
+ }
+
+ bs->backing_hd = backing_hd;
+ if (!backing_hd) {
+ error_free(bs->backing_blocker);
+ bs->backing_blocker = NULL;
+ goto out;
+ }
+ bs->open_flags &= ~BDRV_O_NO_BACKING;
+ pstrcpy(bs->backing_file, sizeof(bs->backing_file), backing_hd->filename);
+ pstrcpy(bs->backing_format, sizeof(bs->backing_format),
+ backing_hd->drv ? backing_hd->drv->format_name : "");
+
+ bdrv_op_block_all(bs->backing_hd, bs->backing_blocker);
+ /* Otherwise we won't be able to commit due to check in bdrv_commit */
+ bdrv_op_unblock(bs->backing_hd, BLOCK_OP_TYPE_COMMIT,
+ bs->backing_blocker);
+out:
+ bdrv_refresh_limits(bs);
+}
+
/*
* Opens the backing file for a BlockDriverState if not yet open
*
*/
int bdrv_open_backing_file(BlockDriverState *bs, QDict *options, Error **errp)
{
- char backing_filename[PATH_MAX];
- int back_flags, ret;
+ char *backing_filename = g_malloc0(PATH_MAX);
+ int ret = 0;
BlockDriver *back_drv = NULL;
+ BlockDriverState *backing_hd;
Error *local_err = NULL;
if (bs->backing_hd != NULL) {
QDECREF(options);
- return 0;
+ goto free_exit;
}
/* NULL means an empty set of options */
backing_filename[0] = '\0';
} else if (bs->backing_file[0] == '\0' && qdict_size(options) == 0) {
QDECREF(options);
- return 0;
+ goto free_exit;
} else {
- bdrv_get_full_backing_filename(bs, backing_filename,
- sizeof(backing_filename));
+ bdrv_get_full_backing_filename(bs, backing_filename, PATH_MAX);
}
- bs->backing_hd = bdrv_new("");
+ backing_hd = bdrv_new("", errp);
if (bs->backing_format[0] != '\0') {
back_drv = bdrv_find_format(bs->backing_format);
}
- /* backing files always opened read-only */
- back_flags = bs->open_flags & ~(BDRV_O_RDWR | BDRV_O_SNAPSHOT |
- BDRV_O_COPY_ON_READ);
-
- ret = bdrv_open(bs->backing_hd,
- *backing_filename ? backing_filename : NULL, options,
- back_flags, back_drv, &local_err);
+ assert(bs->backing_hd == NULL);
+ ret = bdrv_open(&backing_hd,
+ *backing_filename ? backing_filename : NULL, NULL, options,
+ bdrv_backing_flags(bs->open_flags), back_drv, &local_err);
if (ret < 0) {
- bdrv_unref(bs->backing_hd);
- bs->backing_hd = NULL;
+ bdrv_unref(backing_hd);
+ backing_hd = NULL;
bs->open_flags |= BDRV_O_NO_BACKING;
error_setg(errp, "Could not open backing file: %s",
error_get_pretty(local_err));
error_free(local_err);
- return ret;
+ goto free_exit;
}
+ bdrv_set_backing_hd(bs, backing_hd);
- if (bs->backing_hd->file) {
- pstrcpy(bs->backing_file, sizeof(bs->backing_file),
- bs->backing_hd->file->filename);
- }
-
- /* Recalculate the BlockLimits with the backing file */
- bdrv_refresh_limits(bs);
-
- return 0;
+free_exit:
+ g_free(backing_filename);
+ return ret;
}
/*
* Opens a disk image whose options are given as BlockdevRef in another block
* device's options.
*
- * If force_raw is true, bdrv_file_open() will be used, thereby preventing any
- * image format auto-detection. If it is false and a filename is given,
- * bdrv_open() will be used for auto-detection.
- *
* If allow_none is true, no image will be opened if filename is false and no
* BlockdevRef is given. *pbs will remain unchanged and 0 will be returned.
*
* BlockdevRef.
*
* The BlockdevRef will be removed from the options QDict.
+ *
+ * To conform with the behavior of bdrv_open(), *pbs has to be NULL.
*/
int bdrv_open_image(BlockDriverState **pbs, const char *filename,
QDict *options, const char *bdref_key, int flags,
- bool force_raw, bool allow_none, Error **errp)
+ bool allow_none, Error **errp)
{
QDict *image_options;
int ret;
char *bdref_key_dot;
const char *reference;
+ assert(pbs);
+ assert(*pbs == NULL);
+
bdref_key_dot = g_strdup_printf("%s.", bdref_key);
qdict_extract_subqdict(options, &image_options, bdref_key_dot);
g_free(bdref_key_dot);
bdref_key);
ret = -EINVAL;
}
+ QDECREF(image_options);
goto done;
}
- if (filename && !force_raw) {
- /* If a filename is given and the block driver should be detected
- automatically (instead of using none), use bdrv_open() in order to do
- that auto-detection. */
- BlockDriverState *bs;
-
- if (reference) {
- error_setg(errp, "Cannot reference an existing block device while "
- "giving a filename");
- ret = -EINVAL;
- goto done;
- }
-
- bs = bdrv_new("");
- ret = bdrv_open(bs, filename, image_options, flags, NULL, errp);
- if (ret < 0) {
- bdrv_unref(bs);
- } else {
- *pbs = bs;
- }
- } else {
- ret = bdrv_file_open(pbs, filename, reference, image_options, flags,
- errp);
- }
+ ret = bdrv_open(pbs, filename, reference, image_options, flags, NULL, errp);
done:
qdict_del(options, bdref_key);
return ret;
}
+void bdrv_append_temp_snapshot(BlockDriverState *bs, int flags, Error **errp)
+{
+ /* TODO: extra byte is a hack to ensure MAX_PATH space on Windows. */
+ char *tmp_filename = g_malloc0(PATH_MAX + 1);
+ int64_t total_size;
+ BlockDriver *bdrv_qcow2;
+ QEMUOptionParameter *create_options;
+ QDict *snapshot_options;
+ BlockDriverState *bs_snapshot;
+ Error *local_err;
+ int ret;
+
+ /* if snapshot, we create a temporary backing file and open it
+ instead of opening 'filename' directly */
+
+ /* Get the required size from the image */
+ total_size = bdrv_getlength(bs);
+ if (total_size < 0) {
+ error_setg_errno(errp, -total_size, "Could not get image size");
+ goto out;
+ }
+ total_size &= BDRV_SECTOR_MASK;
+
+ /* Create the temporary image */
+ ret = get_tmp_filename(tmp_filename, PATH_MAX + 1);
+ if (ret < 0) {
+ error_setg_errno(errp, -ret, "Could not get temporary filename");
+ goto out;
+ }
+
+ bdrv_qcow2 = bdrv_find_format("qcow2");
+ create_options = parse_option_parameters("", bdrv_qcow2->create_options,
+ NULL);
+
+ set_option_parameter_int(create_options, BLOCK_OPT_SIZE, total_size);
+
+ ret = bdrv_create(bdrv_qcow2, tmp_filename, create_options, &local_err);
+ free_option_parameters(create_options);
+ if (ret < 0) {
+ error_setg_errno(errp, -ret, "Could not create temporary overlay "
+ "'%s': %s", tmp_filename,
+ error_get_pretty(local_err));
+ error_free(local_err);
+ goto out;
+ }
+
+ /* Prepare a new options QDict for the temporary file */
+ snapshot_options = qdict_new();
+ qdict_put(snapshot_options, "file.driver",
+ qstring_from_str("file"));
+ qdict_put(snapshot_options, "file.filename",
+ qstring_from_str(tmp_filename));
+
+ bs_snapshot = bdrv_new("", &error_abort);
+
+ ret = bdrv_open(&bs_snapshot, NULL, NULL, snapshot_options,
+ flags, bdrv_qcow2, &local_err);
+ if (ret < 0) {
+ error_propagate(errp, local_err);
+ goto out;
+ }
+
+ bdrv_append(bs_snapshot, bs);
+
+out:
+ g_free(tmp_filename);
+}
+
+static QDict *parse_json_filename(const char *filename, Error **errp)
+{
+ QObject *options_obj;
+ QDict *options;
+ int ret;
+
+ ret = strstart(filename, "json:", &filename);
+ assert(ret);
+
+ options_obj = qobject_from_json(filename);
+ if (!options_obj) {
+ error_setg(errp, "Could not parse the JSON options");
+ return NULL;
+ }
+
+ if (qobject_type(options_obj) != QTYPE_QDICT) {
+ qobject_decref(options_obj);
+ error_setg(errp, "Invalid JSON object given");
+ return NULL;
+ }
+
+ options = qobject_to_qdict(options_obj);
+ qdict_flatten(options);
+
+ return options;
+}
+
/*
* Opens a disk image (raw, qcow2, vmdk, ...)
*
* empty set of options. The reference to the QDict belongs to the block layer
* after the call (even on failure), so if the caller intends to reuse the
* dictionary, it needs to use QINCREF() before calling bdrv_open.
+ *
+ * If *pbs is NULL, a new BDS will be created with a pointer to it stored there.
+ * If it is not NULL, the referenced BDS will be reused.
+ *
+ * The reference parameter may be used to specify an existing block device which
+ * should be opened. If specified, neither options nor a filename may be given,
+ * nor can an existing BDS be reused (that is, *pbs has to be NULL).
*/
-int bdrv_open(BlockDriverState *bs, const char *filename, QDict *options,
- int flags, BlockDriver *drv, Error **errp)
+int bdrv_open(BlockDriverState **pbs, const char *filename,
+ const char *reference, QDict *options, int flags,
+ BlockDriver *drv, Error **errp)
{
int ret;
- /* TODO: extra byte is a hack to ensure MAX_PATH space on Windows. */
- char tmp_filename[PATH_MAX + 1];
- BlockDriverState *file = NULL;
+ BlockDriverState *file = NULL, *bs;
const char *drvname;
Error *local_err = NULL;
+ int snapshot_flags = 0;
- /* NULL means an empty set of options */
- if (options == NULL) {
- options = qdict_new();
- }
+ assert(pbs);
- bs->options = options;
- options = qdict_clone_shallow(options);
+ if (reference) {
+ bool options_non_empty = options ? qdict_size(options) : false;
+ QDECREF(options);
- /* For snapshot=on, create a temporary qcow2 overlay */
- if (flags & BDRV_O_SNAPSHOT) {
- BlockDriverState *bs1;
- int64_t total_size;
- BlockDriver *bdrv_qcow2;
- QEMUOptionParameter *create_options;
- QDict *snapshot_options;
-
- /* if snapshot, we create a temporary backing file and open it
- instead of opening 'filename' directly */
-
- /* Get the required size from the image */
- bs1 = bdrv_new("");
- QINCREF(options);
- ret = bdrv_open(bs1, filename, options, BDRV_O_NO_BACKING,
- drv, &local_err);
- if (ret < 0) {
- bdrv_unref(bs1);
- goto fail;
+ if (*pbs) {
+ error_setg(errp, "Cannot reuse an existing BDS when referencing "
+ "another block device");
+ return -EINVAL;
}
- total_size = bdrv_getlength(bs1) & BDRV_SECTOR_MASK;
- bdrv_unref(bs1);
+ if (filename || options_non_empty) {
+ error_setg(errp, "Cannot reference an existing block device with "
+ "additional options or a new filename");
+ return -EINVAL;
+ }
- /* Create the temporary image */
- ret = get_tmp_filename(tmp_filename, sizeof(tmp_filename));
- if (ret < 0) {
- error_setg_errno(errp, -ret, "Could not get temporary filename");
- goto fail;
+ bs = bdrv_lookup_bs(reference, reference, errp);
+ if (!bs) {
+ return -ENODEV;
}
+ bdrv_ref(bs);
+ *pbs = bs;
+ return 0;
+ }
- bdrv_qcow2 = bdrv_find_format("qcow2");
- create_options = parse_option_parameters("", bdrv_qcow2->create_options,
- NULL);
+ if (*pbs) {
+ bs = *pbs;
+ } else {
+ bs = bdrv_new("", &error_abort);
+ }
- set_option_parameter_int(create_options, BLOCK_OPT_SIZE, total_size);
+ /* NULL means an empty set of options */
+ if (options == NULL) {
+ options = qdict_new();
+ }
- ret = bdrv_create(bdrv_qcow2, tmp_filename, create_options, &local_err);
- free_option_parameters(create_options);
- if (ret < 0) {
- error_setg_errno(errp, -ret, "Could not create temporary overlay "
- "'%s': %s", tmp_filename,
- error_get_pretty(local_err));
- error_free(local_err);
- local_err = NULL;
+ if (filename && g_str_has_prefix(filename, "json:")) {
+ QDict *json_options = parse_json_filename(filename, &local_err);
+ if (local_err) {
+ ret = -EINVAL;
goto fail;
}
- /* Prepare a new options QDict for the temporary file, where user
- * options refer to the backing file */
- if (filename) {
- qdict_put(options, "file.filename", qstring_from_str(filename));
- }
- if (drv) {
- qdict_put(options, "driver", qstring_from_str(drv->format_name));
- }
-
- snapshot_options = qdict_new();
- qdict_put(snapshot_options, "backing", options);
- qdict_flatten(snapshot_options);
+ /* Options given in the filename have lower priority than options
+ * specified directly */
+ qdict_join(options, json_options, false);
+ QDECREF(json_options);
+ filename = NULL;
+ }
- bs->options = snapshot_options;
- options = qdict_clone_shallow(bs->options);
+ bs->options = options;
+ options = qdict_clone_shallow(options);
- filename = tmp_filename;
- drv = bdrv_qcow2;
- bs->is_temporary = 1;
+ if (flags & BDRV_O_PROTOCOL) {
+ assert(!drv);
+ ret = bdrv_file_open(bs, filename, &options, flags & ~BDRV_O_PROTOCOL,
+ &local_err);
+ if (!ret) {
+ drv = bs->drv;
+ goto done;
+ } else if (bs->drv) {
+ goto close_and_fail;
+ } else {
+ goto fail;
+ }
}
/* Open image file without format layer */
if (flags & BDRV_O_RDWR) {
flags |= BDRV_O_ALLOW_RDWR;
}
+ if (flags & BDRV_O_SNAPSHOT) {
+ snapshot_flags = bdrv_temp_snapshot_flags(flags);
+ flags = bdrv_backing_flags(flags);
+ }
+ assert(file == NULL);
ret = bdrv_open_image(&file, filename, options, "file",
- bdrv_open_flags(bs, flags | BDRV_O_UNMAP), true, true,
- &local_err);
+ bdrv_inherited_flags(flags),
+ true, &local_err);
if (ret < 0) {
goto fail;
}
if (!drv) {
error_setg(errp, "Invalid driver: '%s'", drvname);
ret = -EINVAL;
- goto unlink_and_fail;
+ goto fail;
}
}
} else {
error_setg(errp, "Must specify either driver or file");
ret = -EINVAL;
- goto unlink_and_fail;
+ goto fail;
}
}
if (!drv) {
- goto unlink_and_fail;
+ goto fail;
}
/* Open the image */
ret = bdrv_open_common(bs, file, options, flags, drv, &local_err);
if (ret < 0) {
- goto unlink_and_fail;
+ goto fail;
}
if (file && (bs->file != file)) {
}
}
+ /* For snapshot=on, create a temporary qcow2 overlay. bs points to the
+ * temporary snapshot afterwards. */
+ if (snapshot_flags) {
+ bdrv_append_temp_snapshot(bs, snapshot_flags, &local_err);
+ if (local_err) {
+ error_propagate(errp, local_err);
+ goto close_and_fail;
+ }
+ }
+
+
+done:
/* Check if any unknown options were used */
- if (qdict_size(options) != 0) {
+ if (options && (qdict_size(options) != 0)) {
const QDictEntry *entry = qdict_first(options);
- error_setg(errp, "Block format '%s' used by device '%s' doesn't "
- "support the option '%s'", drv->format_name, bs->device_name,
- entry->key);
+ if (flags & BDRV_O_PROTOCOL) {
+ error_setg(errp, "Block protocol '%s' doesn't support the option "
+ "'%s'", drv->format_name, entry->key);
+ } else {
+ error_setg(errp, "Block format '%s' used by device '%s' doesn't "
+ "support the option '%s'", drv->format_name,
+ bs->device_name, entry->key);
+ }
ret = -EINVAL;
goto close_and_fail;
}
- QDECREF(options);
if (!bdrv_key_required(bs)) {
bdrv_dev_change_media_cb(bs, true);
+ } else if (!runstate_check(RUN_STATE_PRELAUNCH)
+ && !runstate_check(RUN_STATE_INMIGRATE)
+ && !runstate_check(RUN_STATE_PAUSED)) { /* HACK */
+ error_setg(errp,
+ "Guest must be stopped for opening of encrypted image");
+ ret = -EBUSY;
+ goto close_and_fail;
}
+ QDECREF(options);
+ *pbs = bs;
return 0;
-unlink_and_fail:
+fail:
if (file != NULL) {
bdrv_unref(file);
}
- if (bs->is_temporary) {
- unlink(filename);
- }
-fail:
QDECREF(bs->options);
QDECREF(options);
bs->options = NULL;
- if (error_is_set(&local_err)) {
+ if (!*pbs) {
+ /* If *pbs is NULL, a new BDS has been created in this function and
+ needs to be freed now. Otherwise, it does not need to be closed,
+ since it has not really been opened yet. */
+ bdrv_unref(bs);
+ }
+ if (local_err) {
error_propagate(errp, local_err);
}
return ret;
close_and_fail:
- bdrv_close(bs);
+ /* See fail path, but now the BDS has to be always closed */
+ if (*pbs) {
+ bdrv_close(bs);
+ } else {
+ bdrv_unref(bs);
+ }
QDECREF(options);
- if (error_is_set(&local_err)) {
+ if (local_err) {
error_propagate(errp, local_err);
}
return ret;
QSIMPLEQ_INIT(bs_queue);
}
+ /* bdrv_open() masks this flag out */
+ flags &= ~BDRV_O_PROTOCOL;
+
if (bs->file) {
- bdrv_reopen_queue(bs_queue, bs->file, flags);
+ bdrv_reopen_queue(bs_queue, bs->file, bdrv_inherited_flags(flags));
}
bs_entry = g_new0(BlockReopenQueueEntry, 1);
if (bs->drv) {
if (bs->backing_hd) {
- bdrv_unref(bs->backing_hd);
- bs->backing_hd = NULL;
+ BlockDriverState *backing_hd = bs->backing_hd;
+ bdrv_set_backing_hd(bs, NULL);
+ bdrv_unref(backing_hd);
}
bs->drv->bdrv_close(bs);
g_free(bs->opaque);
-#ifdef _WIN32
- if (bs->is_temporary) {
- unlink(bs->filename);
- }
-#endif
bs->opaque = NULL;
bs->drv = NULL;
bs->copy_on_read = 0;
BlockDriverState *bs;
QTAILQ_FOREACH(bs, &bdrv_states, device_list) {
+ AioContext *aio_context = bdrv_get_aio_context(bs);
+
+ aio_context_acquire(aio_context);
bdrv_close(bs);
+ aio_context_release(aio_context);
}
}
return false;
}
-static bool bdrv_requests_pending_all(void)
-{
- BlockDriverState *bs;
- QTAILQ_FOREACH(bs, &bdrv_states, device_list) {
- if (bdrv_requests_pending(bs)) {
- return true;
- }
- }
- return false;
-}
-
/*
* Wait for pending requests to complete across all BlockDriverStates
*
BlockDriverState *bs;
while (busy) {
+ busy = false;
+
QTAILQ_FOREACH(bs, &bdrv_states, device_list) {
+ AioContext *aio_context = bdrv_get_aio_context(bs);
+ bool bs_busy;
+
+ aio_context_acquire(aio_context);
bdrv_start_throttled_reqs(bs);
- }
+ bs_busy = bdrv_requests_pending(bs);
+ bs_busy |= aio_poll(aio_context, bs_busy);
+ aio_context_release(aio_context);
- busy = bdrv_requests_pending_all();
- busy |= aio_poll(qemu_get_aio_context(), busy);
+ busy |= bs_busy;
+ }
}
}
BlockDriverState *bs_src)
{
/* move some fields that need to stay attached to the device */
- bs_dest->open_flags = bs_src->open_flags;
/* dev info */
bs_dest->dev_ops = bs_src->dev_ops;
bs_dest->refcnt = bs_src->refcnt;
/* job */
- bs_dest->in_use = bs_src->in_use;
bs_dest->job = bs_src->job;
/* keep the same entry in bdrv_states */
pstrcpy(bs_dest->device_name, sizeof(bs_dest->device_name),
bs_src->device_name);
bs_dest->device_list = bs_src->device_list;
-
- /* keep the same entry in graph_bdrv_states
- * We do want to swap name but don't want to swap linked list entries
- */
- bs_dest->node_list = bs_src->node_list;
+ memcpy(bs_dest->op_blockers, bs_src->op_blockers,
+ sizeof(bs_dest->op_blockers));
}
/*
{
BlockDriverState tmp;
+ /* The code needs to swap the node_name but simply swapping node_list won't
+ * work so first remove the nodes from the graph list, do the swap then
+ * insert them back if needed.
+ */
+ if (bs_new->node_name[0] != '\0') {
+ QTAILQ_REMOVE(&graph_bdrv_states, bs_new, node_list);
+ }
+ if (bs_old->node_name[0] != '\0') {
+ QTAILQ_REMOVE(&graph_bdrv_states, bs_old, node_list);
+ }
+
/* bs_new must be anonymous and shouldn't have anything fancy enabled */
assert(bs_new->device_name[0] == '\0');
assert(QLIST_EMPTY(&bs_new->dirty_bitmaps));
assert(bs_new->job == NULL);
assert(bs_new->dev == NULL);
- assert(bs_new->in_use == 0);
assert(bs_new->io_limits_enabled == false);
assert(!throttle_have_timer(&bs_new->throttle_state));
/* Check a few fields that should remain attached to the device */
assert(bs_new->dev == NULL);
assert(bs_new->job == NULL);
- assert(bs_new->in_use == 0);
assert(bs_new->io_limits_enabled == false);
assert(!throttle_have_timer(&bs_new->throttle_state));
+ /* insert the nodes back into the graph node list if needed */
+ if (bs_new->node_name[0] != '\0') {
+ QTAILQ_INSERT_TAIL(&graph_bdrv_states, bs_new, node_list);
+ }
+ if (bs_old->node_name[0] != '\0') {
+ QTAILQ_INSERT_TAIL(&graph_bdrv_states, bs_old, node_list);
+ }
+
bdrv_rebind(bs_new);
bdrv_rebind(bs_old);
}
/* The contents of 'tmp' will become bs_top, as we are
* swapping bs_new and bs_top contents. */
- bs_top->backing_hd = bs_new;
- bs_top->open_flags &= ~BDRV_O_NO_BACKING;
- pstrcpy(bs_top->backing_file, sizeof(bs_top->backing_file),
- bs_new->filename);
- pstrcpy(bs_top->backing_format, sizeof(bs_top->backing_format),
- bs_new->drv ? bs_new->drv->format_name : "");
+ bdrv_set_backing_hd(bs_top, bs_new);
}
static void bdrv_delete(BlockDriverState *bs)
{
assert(!bs->dev);
assert(!bs->job);
- assert(!bs->in_use);
+ assert(bdrv_op_blocker_is_empty(bs));
assert(!bs->refcnt);
assert(QLIST_EMPTY(&bs->dirty_bitmaps));
return -ENOTSUP;
}
- if (bdrv_in_use(bs) || bdrv_in_use(bs->backing_hd)) {
+ if (bdrv_op_is_blocked(bs, BLOCK_OP_TYPE_COMMIT, NULL) ||
+ bdrv_op_is_blocked(bs->backing_hd, BLOCK_OP_TYPE_COMMIT, NULL)) {
return -EBUSY;
}
BlockDriverState *bs;
QTAILQ_FOREACH(bs, &bdrv_states, device_list) {
+ AioContext *aio_context = bdrv_get_aio_context(bs);
+
+ aio_context_acquire(aio_context);
if (bs->drv && bs->backing_hd) {
int ret = bdrv_commit(bs);
if (ret < 0) {
+ aio_context_release(aio_context);
return ret;
}
}
+ aio_context_release(aio_context);
}
return 0;
}
*/
static void tracked_request_end(BdrvTrackedRequest *req)
{
+ if (req->serialising) {
+ req->bs->serialising_in_flight--;
+ }
+
QLIST_REMOVE(req, list);
qemu_co_queue_restart_all(&req->wait_queue);
}
*/
static void tracked_request_begin(BdrvTrackedRequest *req,
BlockDriverState *bs,
- int64_t sector_num,
- int nb_sectors, bool is_write)
+ int64_t offset,
+ unsigned int bytes, bool is_write)
{
*req = (BdrvTrackedRequest){
.bs = bs,
- .sector_num = sector_num,
- .nb_sectors = nb_sectors,
- .is_write = is_write,
- .co = qemu_coroutine_self(),
+ .offset = offset,
+ .bytes = bytes,
+ .is_write = is_write,
+ .co = qemu_coroutine_self(),
+ .serialising = false,
+ .overlap_offset = offset,
+ .overlap_bytes = bytes,
};
qemu_co_queue_init(&req->wait_queue);
QLIST_INSERT_HEAD(&bs->tracked_requests, req, list);
}
+static void mark_request_serialising(BdrvTrackedRequest *req, uint64_t align)
+{
+ int64_t overlap_offset = req->offset & ~(align - 1);
+ unsigned int overlap_bytes = ROUND_UP(req->offset + req->bytes, align)
+ - overlap_offset;
+
+ if (!req->serialising) {
+ req->bs->serialising_in_flight++;
+ req->serialising = true;
+ }
+
+ req->overlap_offset = MIN(req->overlap_offset, overlap_offset);
+ req->overlap_bytes = MAX(req->overlap_bytes, overlap_bytes);
+}
+
/**
* Round a region to cluster boundaries
*/
}
}
+static int bdrv_get_cluster_size(BlockDriverState *bs)
+{
+ BlockDriverInfo bdi;
+ int ret;
+
+ ret = bdrv_get_info(bs, &bdi);
+ if (ret < 0 || bdi.cluster_size == 0) {
+ return bs->request_alignment;
+ } else {
+ return bdi.cluster_size;
+ }
+}
+
static bool tracked_request_overlaps(BdrvTrackedRequest *req,
- int64_t sector_num, int nb_sectors) {
+ int64_t offset, unsigned int bytes)
+{
/* aaaa bbbb */
- if (sector_num >= req->sector_num + req->nb_sectors) {
+ if (offset >= req->overlap_offset + req->overlap_bytes) {
return false;
}
/* bbbb aaaa */
- if (req->sector_num >= sector_num + nb_sectors) {
+ if (req->overlap_offset >= offset + bytes) {
return false;
}
return true;
}
-static void coroutine_fn wait_for_overlapping_requests(BlockDriverState *bs,
- int64_t sector_num, int nb_sectors)
+static bool coroutine_fn wait_serialising_requests(BdrvTrackedRequest *self)
{
+ BlockDriverState *bs = self->bs;
BdrvTrackedRequest *req;
- int64_t cluster_sector_num;
- int cluster_nb_sectors;
bool retry;
+ bool waited = false;
- /* If we touch the same cluster it counts as an overlap. This guarantees
- * that allocating writes will be serialized and not race with each other
- * for the same cluster. For example, in copy-on-read it ensures that the
- * CoR read and write operations are atomic and guest writes cannot
- * interleave between them.
- */
- bdrv_round_to_clusters(bs, sector_num, nb_sectors,
- &cluster_sector_num, &cluster_nb_sectors);
+ if (!bs->serialising_in_flight) {
+ return false;
+ }
do {
retry = false;
QLIST_FOREACH(req, &bs->tracked_requests, list) {
- if (tracked_request_overlaps(req, cluster_sector_num,
- cluster_nb_sectors)) {
+ if (req == self || (!req->serialising && !self->serialising)) {
+ continue;
+ }
+ if (tracked_request_overlaps(req, self->overlap_offset,
+ self->overlap_bytes))
+ {
/* Hitting this means there was a reentrant request, for
* example, a block driver issuing nested requests. This must
* never happen since it means deadlock.
*/
assert(qemu_coroutine_self() != req->co);
- qemu_co_queue_wait(&req->wait_queue);
- retry = true;
- break;
+ /* If the request is already (indirectly) waiting for us, or
+ * will wait for us as soon as it wakes up, then just go on
+ * (instead of producing a deadlock in the former case). */
+ if (!req->waiting_for) {
+ self->waiting_for = req;
+ qemu_co_queue_wait(&req->wait_queue);
+ self->waiting_for = NULL;
+ retry = true;
+ waited = true;
+ break;
+ }
}
}
} while (retry);
+
+ return waited;
}
/*
if (ret) {
goto exit;
}
- new_top_bs->backing_hd = base_bs;
-
- bdrv_refresh_limits(new_top_bs);
+ bdrv_set_backing_hd(new_top_bs, base_bs);
QSIMPLEQ_FOREACH_SAFE(intermediate_state, &states_to_delete, entry, next) {
/* so that bdrv_close() does not recursively close the chain */
- intermediate_state->bs->backing_hd = NULL;
+ bdrv_set_backing_hd(intermediate_state->bs, NULL);
bdrv_unref(intermediate_state->bs);
}
ret = 0;
{
int64_t len;
+ if (size > INT_MAX) {
+ return -EIO;
+ }
+
if (!bdrv_is_inserted(bs))
return -ENOMEDIUM;
static int bdrv_check_request(BlockDriverState *bs, int64_t sector_num,
int nb_sectors)
{
+ if (nb_sectors < 0 || nb_sectors > INT_MAX / BDRV_SECTOR_SIZE) {
+ return -EIO;
+ }
+
return bdrv_check_byte_request(bs, sector_num * BDRV_SECTOR_SIZE,
nb_sectors * BDRV_SECTOR_SIZE);
}
typedef struct RwCo {
BlockDriverState *bs;
- int64_t sector_num;
- int nb_sectors;
+ int64_t offset;
QEMUIOVector *qiov;
bool is_write;
int ret;
RwCo *rwco = opaque;
if (!rwco->is_write) {
- rwco->ret = bdrv_co_do_readv(rwco->bs, rwco->sector_num,
- rwco->nb_sectors, rwco->qiov,
- rwco->flags);
- } else {
- rwco->ret = bdrv_co_do_writev(rwco->bs, rwco->sector_num,
- rwco->nb_sectors, rwco->qiov,
+ rwco->ret = bdrv_co_do_preadv(rwco->bs, rwco->offset,
+ rwco->qiov->size, rwco->qiov,
rwco->flags);
+ } else {
+ rwco->ret = bdrv_co_do_pwritev(rwco->bs, rwco->offset,
+ rwco->qiov->size, rwco->qiov,
+ rwco->flags);
}
}
/*
* Process a vectored synchronous request using coroutines
*/
-static int bdrv_rwv_co(BlockDriverState *bs, int64_t sector_num,
- QEMUIOVector *qiov, bool is_write,
- BdrvRequestFlags flags)
+static int bdrv_prwv_co(BlockDriverState *bs, int64_t offset,
+ QEMUIOVector *qiov, bool is_write,
+ BdrvRequestFlags flags)
{
Coroutine *co;
RwCo rwco = {
.bs = bs,
- .sector_num = sector_num,
- .nb_sectors = qiov->size >> BDRV_SECTOR_BITS,
+ .offset = offset,
.qiov = qiov,
.is_write = is_write,
.ret = NOT_DONE,
.flags = flags,
};
- assert((qiov->size & (BDRV_SECTOR_SIZE - 1)) == 0);
/**
* In sync call context, when the vcpu is blocked, this throttling timer
/* Fast-path if already in coroutine context */
bdrv_rw_co_entry(&rwco);
} else {
+ AioContext *aio_context = bdrv_get_aio_context(bs);
+
co = qemu_coroutine_create(bdrv_rw_co_entry);
qemu_coroutine_enter(co, &rwco);
while (rwco.ret == NOT_DONE) {
- qemu_aio_wait();
+ aio_poll(aio_context, true);
}
}
return rwco.ret;
.iov_len = nb_sectors * BDRV_SECTOR_SIZE,
};
+ if (nb_sectors < 0 || nb_sectors > INT_MAX / BDRV_SECTOR_SIZE) {
+ return -EINVAL;
+ }
+
qemu_iovec_init_external(&qiov, &iov, 1);
- return bdrv_rwv_co(bs, sector_num, &qiov, is_write, flags);
+ return bdrv_prwv_co(bs, sector_num << BDRV_SECTOR_BITS,
+ &qiov, is_write, flags);
}
/* return < 0 if error. See bdrv_write() for the return codes */
return bdrv_rw_co(bs, sector_num, (uint8_t *)buf, nb_sectors, true, 0);
}
-int bdrv_writev(BlockDriverState *bs, int64_t sector_num, QEMUIOVector *qiov)
-{
- return bdrv_rwv_co(bs, sector_num, qiov, true, 0);
-}
-
int bdrv_write_zeroes(BlockDriverState *bs, int64_t sector_num,
int nb_sectors, BdrvRequestFlags flags)
{
*/
int bdrv_make_zero(BlockDriverState *bs, BdrvRequestFlags flags)
{
- int64_t target_size = bdrv_getlength(bs) / BDRV_SECTOR_SIZE;
+ int64_t target_size;
int64_t ret, nb_sectors, sector_num = 0;
int n;
+ target_size = bdrv_getlength(bs);
+ if (target_size < 0) {
+ return target_size;
+ }
+ target_size /= BDRV_SECTOR_SIZE;
+
for (;;) {
nb_sectors = target_size - sector_num;
if (nb_sectors <= 0) {
}
}
-int bdrv_pread(BlockDriverState *bs, int64_t offset,
- void *buf, int count1)
+int bdrv_pread(BlockDriverState *bs, int64_t offset, void *buf, int bytes)
{
- uint8_t tmp_buf[BDRV_SECTOR_SIZE];
- int len, nb_sectors, count;
- int64_t sector_num;
+ QEMUIOVector qiov;
+ struct iovec iov = {
+ .iov_base = (void *)buf,
+ .iov_len = bytes,
+ };
int ret;
- count = count1;
- /* first read to align to sector start */
- len = (BDRV_SECTOR_SIZE - offset) & (BDRV_SECTOR_SIZE - 1);
- if (len > count)
- len = count;
- sector_num = offset >> BDRV_SECTOR_BITS;
- if (len > 0) {
- if ((ret = bdrv_read(bs, sector_num, tmp_buf, 1)) < 0)
- return ret;
- memcpy(buf, tmp_buf + (offset & (BDRV_SECTOR_SIZE - 1)), len);
- count -= len;
- if (count == 0)
- return count1;
- sector_num++;
- buf += len;
- }
-
- /* read the sectors "in place" */
- nb_sectors = count >> BDRV_SECTOR_BITS;
- if (nb_sectors > 0) {
- if ((ret = bdrv_read(bs, sector_num, buf, nb_sectors)) < 0)
- return ret;
- sector_num += nb_sectors;
- len = nb_sectors << BDRV_SECTOR_BITS;
- buf += len;
- count -= len;
+ if (bytes < 0) {
+ return -EINVAL;
+ }
+
+ qemu_iovec_init_external(&qiov, &iov, 1);
+ ret = bdrv_prwv_co(bs, offset, &qiov, false, 0);
+ if (ret < 0) {
+ return ret;
}
- /* add data from the last sector */
- if (count > 0) {
- if ((ret = bdrv_read(bs, sector_num, tmp_buf, 1)) < 0)
- return ret;
- memcpy(buf, tmp_buf, count);
- }
- return count1;
+ return bytes;
}
int bdrv_pwritev(BlockDriverState *bs, int64_t offset, QEMUIOVector *qiov)
{
- uint8_t tmp_buf[BDRV_SECTOR_SIZE];
- int len, nb_sectors, count;
- int64_t sector_num;
int ret;
- count = qiov->size;
-
- /* first write to align to sector start */
- len = (BDRV_SECTOR_SIZE - offset) & (BDRV_SECTOR_SIZE - 1);
- if (len > count)
- len = count;
- sector_num = offset >> BDRV_SECTOR_BITS;
- if (len > 0) {
- if ((ret = bdrv_read(bs, sector_num, tmp_buf, 1)) < 0)
- return ret;
- qemu_iovec_to_buf(qiov, 0, tmp_buf + (offset & (BDRV_SECTOR_SIZE - 1)),
- len);
- if ((ret = bdrv_write(bs, sector_num, tmp_buf, 1)) < 0)
- return ret;
- count -= len;
- if (count == 0)
- return qiov->size;
- sector_num++;
- }
-
- /* write the sectors "in place" */
- nb_sectors = count >> BDRV_SECTOR_BITS;
- if (nb_sectors > 0) {
- QEMUIOVector qiov_inplace;
-
- qemu_iovec_init(&qiov_inplace, qiov->niov);
- qemu_iovec_concat(&qiov_inplace, qiov, len,
- nb_sectors << BDRV_SECTOR_BITS);
- ret = bdrv_writev(bs, sector_num, &qiov_inplace);
- qemu_iovec_destroy(&qiov_inplace);
- if (ret < 0) {
- return ret;
- }
-
- sector_num += nb_sectors;
- len = nb_sectors << BDRV_SECTOR_BITS;
- count -= len;
+ ret = bdrv_prwv_co(bs, offset, qiov, true, 0);
+ if (ret < 0) {
+ return ret;
}
- /* add data from the last sector */
- if (count > 0) {
- if ((ret = bdrv_read(bs, sector_num, tmp_buf, 1)) < 0)
- return ret;
- qemu_iovec_to_buf(qiov, qiov->size - count, tmp_buf, count);
- if ((ret = bdrv_write(bs, sector_num, tmp_buf, 1)) < 0)
- return ret;
- }
return qiov->size;
}
int bdrv_pwrite(BlockDriverState *bs, int64_t offset,
- const void *buf, int count1)
+ const void *buf, int bytes)
{
QEMUIOVector qiov;
struct iovec iov = {
.iov_base = (void *) buf,
- .iov_len = count1,
+ .iov_len = bytes,
};
+ if (bytes < 0) {
+ return -EINVAL;
+ }
+
qemu_iovec_init_external(&qiov, &iov, 1);
return bdrv_pwritev(bs, offset, &qiov);
}
* implemented by the caller.
*/
static int coroutine_fn bdrv_aligned_preadv(BlockDriverState *bs,
- int64_t offset, unsigned int bytes, QEMUIOVector *qiov, int flags)
+ BdrvTrackedRequest *req, int64_t offset, unsigned int bytes,
+ int64_t align, QEMUIOVector *qiov, int flags)
{
BlockDriver *drv = bs->drv;
- BdrvTrackedRequest req;
int ret;
int64_t sector_num = offset >> BDRV_SECTOR_BITS;
/* Handle Copy on Read and associated serialisation */
if (flags & BDRV_REQ_COPY_ON_READ) {
- bs->copy_on_read_in_flight++;
- }
-
- if (bs->copy_on_read_in_flight) {
- wait_for_overlapping_requests(bs, sector_num, nb_sectors);
+ /* If we touch the same cluster it counts as an overlap. This
+ * guarantees that allocating writes will be serialized and not race
+ * with each other for the same cluster. For example, in copy-on-read
+ * it ensures that the CoR read and write operations are atomic and
+ * guest writes cannot interleave between them. */
+ mark_request_serialising(req, bdrv_get_cluster_size(bs));
}
- tracked_request_begin(&req, bs, sector_num, nb_sectors, false);
+ wait_serialising_requests(req);
if (flags & BDRV_REQ_COPY_ON_READ) {
int pnum;
}
total_sectors = DIV_ROUND_UP(len, BDRV_SECTOR_SIZE);
- max_nb_sectors = MAX(0, total_sectors - sector_num);
+ max_nb_sectors = ROUND_UP(MAX(0, total_sectors - sector_num),
+ align >> BDRV_SECTOR_BITS);
if (max_nb_sectors > 0) {
ret = drv->bdrv_co_readv(bs, sector_num,
MIN(nb_sectors, max_nb_sectors), qiov);
}
out:
- tracked_request_end(&req);
-
- if (flags & BDRV_REQ_COPY_ON_READ) {
- bs->copy_on_read_in_flight--;
- }
-
return ret;
}
BdrvRequestFlags flags)
{
BlockDriver *drv = bs->drv;
+ BdrvTrackedRequest req;
+
/* TODO Lift BDRV_SECTOR_SIZE restriction in BlockDriver interface */
uint64_t align = MAX(BDRV_SECTOR_SIZE, bs->request_alignment);
uint8_t *head_buf = NULL;
/* throttling disk I/O */
if (bs->io_limits_enabled) {
- /* TODO Switch to byte granularity */
- bdrv_io_limits_intercept(bs, bytes >> BDRV_SECTOR_BITS, false);
+ bdrv_io_limits_intercept(bs, bytes, false);
}
/* Align read if necessary by padding qiov */
bytes = ROUND_UP(bytes, align);
}
- ret = bdrv_aligned_preadv(bs, offset, bytes,
+ tracked_request_begin(&req, bs, offset, bytes, false);
+ ret = bdrv_aligned_preadv(bs, &req, offset, bytes, align,
use_local_qiov ? &local_qiov : qiov,
flags);
+ tracked_request_end(&req);
if (use_local_qiov) {
qemu_iovec_destroy(&local_qiov);
* Forwards an already correctly aligned write request to the BlockDriver.
*/
static int coroutine_fn bdrv_aligned_pwritev(BlockDriverState *bs,
- int64_t offset, unsigned int bytes, QEMUIOVector *qiov, int flags)
+ BdrvTrackedRequest *req, int64_t offset, unsigned int bytes,
+ QEMUIOVector *qiov, int flags)
{
BlockDriver *drv = bs->drv;
- BdrvTrackedRequest req;
+ bool waited;
int ret;
int64_t sector_num = offset >> BDRV_SECTOR_BITS;
assert((offset & (BDRV_SECTOR_SIZE - 1)) == 0);
assert((bytes & (BDRV_SECTOR_SIZE - 1)) == 0);
- tracked_request_begin(&req, bs, sector_num, nb_sectors, true);
+ waited = wait_serialising_requests(req);
+ assert(!waited || !req->serialising);
+ assert(req->overlap_offset <= offset);
+ assert(offset + bytes <= req->overlap_offset + req->overlap_bytes);
- ret = notifier_with_return_list_notify(&bs->before_write_notifiers, &req);
+ ret = notifier_with_return_list_notify(&bs->before_write_notifiers, req);
+
+ if (!ret && bs->detect_zeroes != BLOCKDEV_DETECT_ZEROES_OPTIONS_OFF &&
+ !(flags & BDRV_REQ_ZERO_WRITE) && drv->bdrv_co_write_zeroes &&
+ qemu_iovec_is_zero(qiov)) {
+ flags |= BDRV_REQ_ZERO_WRITE;
+ if (bs->detect_zeroes == BLOCKDEV_DETECT_ZEROES_OPTIONS_UNMAP) {
+ flags |= BDRV_REQ_MAY_UNMAP;
+ }
+ }
if (ret < 0) {
/* Do nothing, write notifier decided to fail this request */
} else if (flags & BDRV_REQ_ZERO_WRITE) {
+ BLKDBG_EVENT(bs, BLKDBG_PWRITEV_ZERO);
ret = bdrv_co_do_write_zeroes(bs, sector_num, nb_sectors, flags);
} else {
+ BLKDBG_EVENT(bs, BLKDBG_PWRITEV);
ret = drv->bdrv_co_writev(bs, sector_num, nb_sectors, qiov);
}
+ BLKDBG_EVENT(bs, BLKDBG_PWRITEV_DONE);
if (ret == 0 && !bs->enable_write_cache) {
ret = bdrv_co_flush(bs);
bs->total_sectors = MAX(bs->total_sectors, sector_num + nb_sectors);
}
- tracked_request_end(&req);
-
return ret;
}
/*
* Handle a write request in coroutine context
*/
-static int coroutine_fn bdrv_co_do_writev(BlockDriverState *bs,
- int64_t sector_num, int nb_sectors, QEMUIOVector *qiov,
+static int coroutine_fn bdrv_co_do_pwritev(BlockDriverState *bs,
+ int64_t offset, unsigned int bytes, QEMUIOVector *qiov,
BdrvRequestFlags flags)
{
+ BdrvTrackedRequest req;
+ /* TODO Lift BDRV_SECTOR_SIZE restriction in BlockDriver interface */
+ uint64_t align = MAX(BDRV_SECTOR_SIZE, bs->request_alignment);
+ uint8_t *head_buf = NULL;
+ uint8_t *tail_buf = NULL;
+ QEMUIOVector local_qiov;
+ bool use_local_qiov = false;
int ret;
if (!bs->drv) {
if (bs->read_only) {
return -EACCES;
}
- if (bdrv_check_request(bs, sector_num, nb_sectors)) {
+ if (bdrv_check_byte_request(bs, offset, bytes)) {
return -EIO;
}
- if (bs->copy_on_read_in_flight) {
- wait_for_overlapping_requests(bs, sector_num, nb_sectors);
- }
-
/* throttling disk I/O */
if (bs->io_limits_enabled) {
- bdrv_io_limits_intercept(bs, nb_sectors, true);
+ bdrv_io_limits_intercept(bs, bytes, true);
+ }
+
+ /*
+ * Align write if necessary by performing a read-modify-write cycle.
+ * Pad qiov with the read parts and be sure to have a tracked request not
+ * only for bdrv_aligned_pwritev, but also for the reads of the RMW cycle.
+ */
+ tracked_request_begin(&req, bs, offset, bytes, true);
+
+ if (offset & (align - 1)) {
+ QEMUIOVector head_qiov;
+ struct iovec head_iov;
+
+ mark_request_serialising(&req, align);
+ wait_serialising_requests(&req);
+
+ head_buf = qemu_blockalign(bs, align);
+ head_iov = (struct iovec) {
+ .iov_base = head_buf,
+ .iov_len = align,
+ };
+ qemu_iovec_init_external(&head_qiov, &head_iov, 1);
+
+ BLKDBG_EVENT(bs, BLKDBG_PWRITEV_RMW_HEAD);
+ ret = bdrv_aligned_preadv(bs, &req, offset & ~(align - 1), align,
+ align, &head_qiov, 0);
+ if (ret < 0) {
+ goto fail;
+ }
+ BLKDBG_EVENT(bs, BLKDBG_PWRITEV_RMW_AFTER_HEAD);
+
+ qemu_iovec_init(&local_qiov, qiov->niov + 2);
+ qemu_iovec_add(&local_qiov, head_buf, offset & (align - 1));
+ qemu_iovec_concat(&local_qiov, qiov, 0, qiov->size);
+ use_local_qiov = true;
+
+ bytes += offset & (align - 1);
+ offset = offset & ~(align - 1);
+ }
+
+ if ((offset + bytes) & (align - 1)) {
+ QEMUIOVector tail_qiov;
+ struct iovec tail_iov;
+ size_t tail_bytes;
+ bool waited;
+
+ mark_request_serialising(&req, align);
+ waited = wait_serialising_requests(&req);
+ assert(!waited || !use_local_qiov);
+
+ tail_buf = qemu_blockalign(bs, align);
+ tail_iov = (struct iovec) {
+ .iov_base = tail_buf,
+ .iov_len = align,
+ };
+ qemu_iovec_init_external(&tail_qiov, &tail_iov, 1);
+
+ BLKDBG_EVENT(bs, BLKDBG_PWRITEV_RMW_TAIL);
+ ret = bdrv_aligned_preadv(bs, &req, (offset + bytes) & ~(align - 1), align,
+ align, &tail_qiov, 0);
+ if (ret < 0) {
+ goto fail;
+ }
+ BLKDBG_EVENT(bs, BLKDBG_PWRITEV_RMW_AFTER_TAIL);
+
+ if (!use_local_qiov) {
+ qemu_iovec_init(&local_qiov, qiov->niov + 1);
+ qemu_iovec_concat(&local_qiov, qiov, 0, qiov->size);
+ use_local_qiov = true;
+ }
+
+ tail_bytes = (offset + bytes) & (align - 1);
+ qemu_iovec_add(&local_qiov, tail_buf + tail_bytes, align - tail_bytes);
+
+ bytes = ROUND_UP(bytes, align);
}
- ret = bdrv_aligned_pwritev(bs, sector_num << BDRV_SECTOR_BITS,
- nb_sectors << BDRV_SECTOR_BITS, qiov, flags);
+ ret = bdrv_aligned_pwritev(bs, &req, offset, bytes,
+ use_local_qiov ? &local_qiov : qiov,
+ flags);
+
+fail:
+ tracked_request_end(&req);
+
+ if (use_local_qiov) {
+ qemu_iovec_destroy(&local_qiov);
+ }
+ qemu_vfree(head_buf);
+ qemu_vfree(tail_buf);
return ret;
}
+static int coroutine_fn bdrv_co_do_writev(BlockDriverState *bs,
+ int64_t sector_num, int nb_sectors, QEMUIOVector *qiov,
+ BdrvRequestFlags flags)
+{
+ if (nb_sectors < 0 || nb_sectors > (INT_MAX >> BDRV_SECTOR_BITS)) {
+ return -EINVAL;
+ }
+
+ return bdrv_co_do_pwritev(bs, sector_num << BDRV_SECTOR_BITS,
+ nb_sectors << BDRV_SECTOR_BITS, qiov, flags);
+}
+
int coroutine_fn bdrv_co_writev(BlockDriverState *bs, int64_t sector_num,
int nb_sectors, QEMUIOVector *qiov)
{
return -ENOTSUP;
if (bs->read_only)
return -EACCES;
- if (bdrv_in_use(bs))
+ if (bdrv_op_is_blocked(bs, BLOCK_OP_TYPE_RESIZE, NULL)) {
return -EBUSY;
+ }
ret = drv->bdrv_truncate(bs, offset);
if (ret == 0) {
ret = refresh_total_sectors(bs, offset >> BDRV_SECTOR_BITS);
void *opaque)
{
BlockDriver *drv;
+ int count = 0;
+ const char **formats = NULL;
QLIST_FOREACH(drv, &bdrv_drivers, list) {
- it(opaque, drv->format_name);
+ if (drv->format_name) {
+ bool found = false;
+ int i = count;
+ while (formats && i && !found) {
+ found = !strcmp(formats[--i], drv->format_name);
+ }
+
+ if (!found) {
+ formats = g_realloc(formats, (count + 1) * sizeof(char *));
+ formats[count++] = drv->format_name;
+ it(opaque, drv->format_name);
+ }
+ }
}
+ g_free(formats);
}
/* This function is to find block backend bs */
{
BlockDriverState *bs = NULL;
- if ((!device && !node_name) || (device && node_name)) {
- error_setg(errp, "Use either device or node-name but not both");
- return NULL;
- }
-
if (device) {
bs = bdrv_find(device);
- if (!bs) {
- error_set(errp, QERR_DEVICE_NOT_FOUND, device);
- return NULL;
+ if (bs) {
+ return bs;
}
-
- return bs;
}
- bs = bdrv_find_node(node_name);
+ if (node_name) {
+ bs = bdrv_find_node(node_name);
- if (!bs) {
- error_set(errp, QERR_DEVICE_NOT_FOUND, node_name);
- return NULL;
+ if (bs) {
+ return bs;
+ }
}
- return bs;
+ error_setg(errp, "Cannot find device=%s nor node_name=%s",
+ device ? device : "",
+ node_name ? node_name : "");
+ return NULL;
}
BlockDriverState *bdrv_next(BlockDriverState *bs)
int result = 0;
QTAILQ_FOREACH(bs, &bdrv_states, device_list) {
- int ret = bdrv_flush(bs);
+ AioContext *aio_context = bdrv_get_aio_context(bs);
+ int ret;
+
+ aio_context_acquire(aio_context);
+ ret = bdrv_flush(bs);
if (ret < 0 && !result) {
result = ret;
}
+ aio_context_release(aio_context);
}
return result;
if (!bs->drv->bdrv_co_get_block_status) {
*pnum = nb_sectors;
- ret = BDRV_BLOCK_DATA;
+ ret = BDRV_BLOCK_DATA | BDRV_BLOCK_ALLOCATED;
if (bs->drv->protocol_name) {
ret |= BDRV_BLOCK_OFFSET_VALID | (sector_num * BDRV_SECTOR_SIZE);
}
*pnum, pnum);
}
+ if (ret & (BDRV_BLOCK_DATA | BDRV_BLOCK_ZERO)) {
+ ret |= BDRV_BLOCK_ALLOCATED;
+ }
+
if (!(ret & BDRV_BLOCK_DATA) && !(ret & BDRV_BLOCK_ZERO)) {
if (bdrv_unallocated_blocks_are_zero(bs)) {
ret |= BDRV_BLOCK_ZERO;
/* Fast-path if already in coroutine context */
bdrv_get_block_status_co_entry(&data);
} else {
+ AioContext *aio_context = bdrv_get_aio_context(bs);
+
co = qemu_coroutine_create(bdrv_get_block_status_co_entry);
qemu_coroutine_enter(co, &data);
while (!data.done) {
- qemu_aio_wait();
+ aio_poll(aio_context, true);
}
}
return data.ret;
if (ret < 0) {
return ret;
}
- return
- (ret & BDRV_BLOCK_DATA) ||
- ((ret & BDRV_BLOCK_ZERO) && !bdrv_has_zero_init(bs));
+ return (ret & BDRV_BLOCK_ALLOCATED);
}
/*
int bdrv_debug_resume(BlockDriverState *bs, const char *tag)
{
- while (bs && bs->drv && !bs->drv->bdrv_debug_resume) {
+ while (bs && (!bs->drv || !bs->drv->bdrv_debug_resume)) {
bs = bs->file;
}
acb->is_write = is_write;
acb->qiov = qiov;
acb->bounce = qemu_blockalign(bs, qiov->size);
- acb->bh = qemu_bh_new(bdrv_aio_bh_cb, acb);
+ acb->bh = aio_bh_new(bdrv_get_aio_context(bs), bdrv_aio_bh_cb, acb);
if (is_write) {
qemu_iovec_to_buf(acb->qiov, 0, acb->bounce, qiov->size);
static void bdrv_aio_co_cancel_em(BlockDriverAIOCB *blockacb)
{
+ AioContext *aio_context = bdrv_get_aio_context(blockacb->bs);
BlockDriverAIOCBCoroutine *acb =
container_of(blockacb, BlockDriverAIOCBCoroutine, common);
bool done = false;
acb->done = &done;
while (!done) {
- qemu_aio_wait();
+ aio_poll(aio_context, true);
}
}
acb->req.nb_sectors, acb->req.qiov, acb->req.flags);
}
- acb->bh = qemu_bh_new(bdrv_co_em_bh, acb);
+ acb->bh = aio_bh_new(bdrv_get_aio_context(bs), bdrv_co_em_bh, acb);
qemu_bh_schedule(acb->bh);
}
BlockDriverState *bs = acb->common.bs;
acb->req.error = bdrv_co_flush(bs);
- acb->bh = qemu_bh_new(bdrv_co_em_bh, acb);
+ acb->bh = aio_bh_new(bdrv_get_aio_context(bs), bdrv_co_em_bh, acb);
qemu_bh_schedule(acb->bh);
}
BlockDriverState *bs = acb->common.bs;
acb->req.error = bdrv_co_discard(bs, acb->req.sector, acb->req.nb_sectors);
- acb->bh = qemu_bh_new(bdrv_co_em_bh, acb);
+ acb->bh = aio_bh_new(bdrv_get_aio_context(bs), bdrv_co_em_bh, acb);
qemu_bh_schedule(acb->bh);
}
return bdrv_co_flush(bs->file);
}
-void bdrv_invalidate_cache(BlockDriverState *bs)
+void bdrv_invalidate_cache(BlockDriverState *bs, Error **errp)
{
- if (bs->drv && bs->drv->bdrv_invalidate_cache) {
- bs->drv->bdrv_invalidate_cache(bs);
+ Error *local_err = NULL;
+ int ret;
+
+ if (!bs->drv) {
+ return;
+ }
+
+ if (bs->drv->bdrv_invalidate_cache) {
+ bs->drv->bdrv_invalidate_cache(bs, &local_err);
+ } else if (bs->file) {
+ bdrv_invalidate_cache(bs->file, &local_err);
+ }
+ if (local_err) {
+ error_propagate(errp, local_err);
+ return;
+ }
+
+ ret = refresh_total_sectors(bs, bs->total_sectors);
+ if (ret < 0) {
+ error_setg_errno(errp, -ret, "Could not refresh total sector count");
+ return;
}
}
-void bdrv_invalidate_cache_all(void)
+void bdrv_invalidate_cache_all(Error **errp)
{
BlockDriverState *bs;
+ Error *local_err = NULL;
QTAILQ_FOREACH(bs, &bdrv_states, device_list) {
- bdrv_invalidate_cache(bs);
+ AioContext *aio_context = bdrv_get_aio_context(bs);
+
+ aio_context_acquire(aio_context);
+ bdrv_invalidate_cache(bs, &local_err);
+ aio_context_release(aio_context);
+ if (local_err) {
+ error_propagate(errp, local_err);
+ return;
+ }
}
}
BlockDriverState *bs;
QTAILQ_FOREACH(bs, &bdrv_states, device_list) {
+ AioContext *aio_context = bdrv_get_aio_context(bs);
+
+ aio_context_acquire(aio_context);
bs->open_flags = bs->open_flags & ~(BDRV_O_INCOMING);
+ aio_context_release(aio_context);
}
}
/* Fast-path if already in coroutine context */
bdrv_flush_co_entry(&rwco);
} else {
+ AioContext *aio_context = bdrv_get_aio_context(bs);
+
co = qemu_coroutine_create(bdrv_flush_co_entry);
qemu_coroutine_enter(co, &rwco);
while (rwco.ret == NOT_DONE) {
- qemu_aio_wait();
+ aio_poll(aio_context, true);
}
}
return rwco.ret;
}
+typedef struct DiscardCo {
+ BlockDriverState *bs;
+ int64_t sector_num;
+ int nb_sectors;
+ int ret;
+} DiscardCo;
static void coroutine_fn bdrv_discard_co_entry(void *opaque)
{
- RwCo *rwco = opaque;
+ DiscardCo *rwco = opaque;
rwco->ret = bdrv_co_discard(rwco->bs, rwco->sector_num, rwco->nb_sectors);
}
int bdrv_discard(BlockDriverState *bs, int64_t sector_num, int nb_sectors)
{
Coroutine *co;
- RwCo rwco = {
+ DiscardCo rwco = {
.bs = bs,
.sector_num = sector_num,
.nb_sectors = nb_sectors,
/* Fast-path if already in coroutine context */
bdrv_discard_co_entry(&rwco);
} else {
+ AioContext *aio_context = bdrv_get_aio_context(bs);
+
co = qemu_coroutine_create(bdrv_discard_co_entry);
qemu_coroutine_enter(co, &rwco);
while (rwco.ret == NOT_DONE) {
- qemu_aio_wait();
+ aio_poll(aio_context, true);
}
}
return true;
}
-BdrvDirtyBitmap *bdrv_create_dirty_bitmap(BlockDriverState *bs, int granularity)
+BdrvDirtyBitmap *bdrv_create_dirty_bitmap(BlockDriverState *bs, int granularity,
+ Error **errp)
{
int64_t bitmap_size;
BdrvDirtyBitmap *bitmap;
granularity >>= BDRV_SECTOR_BITS;
assert(granularity);
- bitmap_size = (bdrv_getlength(bs) >> BDRV_SECTOR_BITS);
+ bitmap_size = bdrv_getlength(bs);
+ if (bitmap_size < 0) {
+ error_setg_errno(errp, -bitmap_size, "could not get length of device");
+ errno = -bitmap_size;
+ return NULL;
+ }
+ bitmap_size >>= BDRV_SECTOR_BITS;
bitmap = g_malloc0(sizeof(BdrvDirtyBitmap));
bitmap->bitmap = hbitmap_alloc(bitmap_size, ffs(granularity) - 1);
QLIST_INSERT_HEAD(&bs->dirty_bitmaps, bitmap, list);
}
}
-void bdrv_set_in_use(BlockDriverState *bs, int in_use)
+struct BdrvOpBlocker {
+ Error *reason;
+ QLIST_ENTRY(BdrvOpBlocker) list;
+};
+
+bool bdrv_op_is_blocked(BlockDriverState *bs, BlockOpType op, Error **errp)
+{
+ BdrvOpBlocker *blocker;
+ assert((int) op >= 0 && op < BLOCK_OP_TYPE_MAX);
+ if (!QLIST_EMPTY(&bs->op_blockers[op])) {
+ blocker = QLIST_FIRST(&bs->op_blockers[op]);
+ if (errp) {
+ error_setg(errp, "Device '%s' is busy: %s",
+ bs->device_name, error_get_pretty(blocker->reason));
+ }
+ return true;
+ }
+ return false;
+}
+
+void bdrv_op_block(BlockDriverState *bs, BlockOpType op, Error *reason)
+{
+ BdrvOpBlocker *blocker;
+ assert((int) op >= 0 && op < BLOCK_OP_TYPE_MAX);
+
+ blocker = g_malloc0(sizeof(BdrvOpBlocker));
+ blocker->reason = reason;
+ QLIST_INSERT_HEAD(&bs->op_blockers[op], blocker, list);
+}
+
+void bdrv_op_unblock(BlockDriverState *bs, BlockOpType op, Error *reason)
+{
+ BdrvOpBlocker *blocker, *next;
+ assert((int) op >= 0 && op < BLOCK_OP_TYPE_MAX);
+ QLIST_FOREACH_SAFE(blocker, &bs->op_blockers[op], list, next) {
+ if (blocker->reason == reason) {
+ QLIST_REMOVE(blocker, list);
+ g_free(blocker);
+ }
+ }
+}
+
+void bdrv_op_block_all(BlockDriverState *bs, Error *reason)
+{
+ int i;
+ for (i = 0; i < BLOCK_OP_TYPE_MAX; i++) {
+ bdrv_op_block(bs, i, reason);
+ }
+}
+
+void bdrv_op_unblock_all(BlockDriverState *bs, Error *reason)
{
- assert(bs->in_use != in_use);
- bs->in_use = in_use;
+ int i;
+ for (i = 0; i < BLOCK_OP_TYPE_MAX; i++) {
+ bdrv_op_unblock(bs, i, reason);
+ }
}
-int bdrv_in_use(BlockDriverState *bs)
+bool bdrv_op_blocker_is_empty(BlockDriverState *bs)
{
- return bs->in_use;
+ int i;
+
+ for (i = 0; i < BLOCK_OP_TYPE_MAX; i++) {
+ if (!QLIST_EMPTY(&bs->op_blockers[i])) {
+ return false;
+ }
+ }
+ return true;
}
void bdrv_iostatus_enable(BlockDriverState *bs)
back_flags =
flags & ~(BDRV_O_RDWR | BDRV_O_SNAPSHOT | BDRV_O_NO_BACKING);
- bs = bdrv_new("");
-
- ret = bdrv_open(bs, backing_file->value.s, NULL, back_flags,
+ bs = NULL;
+ ret = bdrv_open(&bs, backing_file->value.s, NULL, NULL, back_flags,
backing_drv, &local_err);
if (ret < 0) {
error_setg_errno(errp, -ret, "Could not open '%s': %s",
error_get_pretty(local_err));
error_free(local_err);
local_err = NULL;
- bdrv_unref(bs);
goto out;
}
bdrv_get_geometry(bs, &size);
free_option_parameters(create_options);
free_option_parameters(param);
- if (error_is_set(&local_err)) {
+ if (local_err) {
error_propagate(errp, local_err);
}
}
AioContext *bdrv_get_aio_context(BlockDriverState *bs)
{
- /* Currently BlockDriverState always uses the main loop AioContext */
- return qemu_get_aio_context();
+ return bs->aio_context;
+}
+
+void bdrv_detach_aio_context(BlockDriverState *bs)
+{
+ if (!bs->drv) {
+ return;
+ }
+
+ if (bs->io_limits_enabled) {
+ throttle_detach_aio_context(&bs->throttle_state);
+ }
+ if (bs->drv->bdrv_detach_aio_context) {
+ bs->drv->bdrv_detach_aio_context(bs);
+ }
+ if (bs->file) {
+ bdrv_detach_aio_context(bs->file);
+ }
+ if (bs->backing_hd) {
+ bdrv_detach_aio_context(bs->backing_hd);
+ }
+
+ bs->aio_context = NULL;
+}
+
+void bdrv_attach_aio_context(BlockDriverState *bs,
+ AioContext *new_context)
+{
+ if (!bs->drv) {
+ return;
+ }
+
+ bs->aio_context = new_context;
+
+ if (bs->backing_hd) {
+ bdrv_attach_aio_context(bs->backing_hd, new_context);
+ }
+ if (bs->file) {
+ bdrv_attach_aio_context(bs->file, new_context);
+ }
+ if (bs->drv->bdrv_attach_aio_context) {
+ bs->drv->bdrv_attach_aio_context(bs, new_context);
+ }
+ if (bs->io_limits_enabled) {
+ throttle_attach_aio_context(&bs->throttle_state, new_context);
+ }
+}
+
+void bdrv_set_aio_context(BlockDriverState *bs, AioContext *new_context)
+{
+ bdrv_drain_all(); /* ensure there are no in-flight requests */
+
+ bdrv_detach_aio_context(bs);
+
+ /* This function executes in the old AioContext so acquire the new one in
+ * case it runs in a different thread.
+ */
+ aio_context_acquire(new_context);
+ bdrv_attach_aio_context(bs, new_context);
+ aio_context_release(new_context);
}
void bdrv_add_before_write_notifier(BlockDriverState *bs,
return bs->drv->bdrv_amend_options(bs, options);
}
-/* Used to recurse on single child block filters.
- * Single child block filter will store their child in bs->file.
+/* This function will be called by the bdrv_recurse_is_first_non_filter method
+ * of block filter and by bdrv_is_first_non_filter.
+ * It is used to test if the given bs is the candidate or recurse more in the
+ * node graph.
*/
-bool bdrv_generic_is_first_non_filter(BlockDriverState *bs,
+bool bdrv_recurse_is_first_non_filter(BlockDriverState *bs,
BlockDriverState *candidate)
{
- if (!bs->drv) {
- return false;
- }
-
- if (!bs->drv->authorizations[BS_IS_A_FILTER]) {
- if (bs == candidate) {
- return true;
- } else {
- return false;
- }
- }
-
- if (!bs->drv->authorizations[BS_FILTER_PASS_DOWN]) {
+ /* return false if basic checks fails */
+ if (!bs || !bs->drv) {
return false;
}
- if (!bs->file) {
- return false;
+ /* the code reached a non block filter driver -> check if the bs is
+ * the same as the candidate. It's the recursion termination condition.
+ */
+ if (!bs->drv->is_filter) {
+ return bs == candidate;
}
+ /* Down this path the driver is a block filter driver */
- return bdrv_recurse_is_first_non_filter(bs->file, candidate);
-}
-
-bool bdrv_recurse_is_first_non_filter(BlockDriverState *bs,
- BlockDriverState *candidate)
-{
- if (bs->drv && bs->drv->bdrv_recurse_is_first_non_filter) {
+ /* If the block filter recursion method is defined use it to recurse down
+ * the node graph.
+ */
+ if (bs->drv->bdrv_recurse_is_first_non_filter) {
return bs->drv->bdrv_recurse_is_first_non_filter(bs, candidate);
}
- return bdrv_generic_is_first_non_filter(bs, candidate);
+ /* the driver is a block filter but don't allow to recurse -> return false
+ */
+ return false;
}
/* This function checks if the candidate is the first non filter bs down it's
QTAILQ_FOREACH(bs, &bdrv_states, device_list) {
bool perm;
- if (!bs->file) {
- continue;
- }
-
- perm = bdrv_recurse_is_first_non_filter(bs->file, candidate);
+ /* try to recurse in this top level bs */
+ perm = bdrv_recurse_is_first_non_filter(bs, candidate);
/* candidate is the first non filter */
if (perm) {
projects / qemu.git / blobdiff