* OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
* THE SOFTWARE.
*/
+
#include "qemu/osdep.h"
#include "block/trace.h"
#include "block/block_int.h"
#include "block/blockjob.h"
#include "block/nbd.h"
+#include "block/qdict.h"
#include "qemu/error-report.h"
#include "module_block.h"
#include "qemu/module.h"
-#include "qapi/qmp/qerror.h"
-#include "qapi/qmp/qbool.h"
+#include "qapi/error.h"
+#include "qapi/qmp/qdict.h"
#include "qapi/qmp/qjson.h"
+#include "qapi/qmp/qnull.h"
+#include "qapi/qmp/qstring.h"
+#include "qapi/qobject-output-visitor.h"
+#include "qapi/qapi-visit-block-core.h"
#include "sysemu/block-backend.h"
#include "sysemu/sysemu.h"
#include "qemu/notify.h"
+#include "qemu/option.h"
#include "qemu/coroutine.h"
#include "block/qapi.h"
-#include "qmp-commands.h"
#include "qemu/timer.h"
-#include "qapi-event.h"
#include "qemu/cutils.h"
#include "qemu/id.h"
return bs->read_only;
}
-/* Returns whether the image file can be written to right now */
-bool bdrv_is_writable(BlockDriverState *bs)
-{
- return !bdrv_is_read_only(bs) && !(bs->open_flags & BDRV_O_INACTIVE);
-}
-
int bdrv_can_set_read_only(BlockDriverState *bs, bool read_only,
bool ignore_allow_rdw, Error **errp)
{
return 0;
}
+/* TODO Remove (deprecated since 2.11)
+ * Block drivers are not supposed to automatically change bs->read_only.
+ * Instead, they should just check whether they can provide what the user
+ * explicitly requested and error out if read-write is requested, but they can
+ * only provide read-only access. */
int bdrv_set_read_only(BlockDriverState *bs, bool read_only, Error **errp)
{
int ret = 0;
qemu_co_queue_init(&bs->flush_queue);
+ for (i = 0; i < bdrv_drain_all_count; i++) {
+ bdrv_drained_begin(bs);
+ }
+
QTAILQ_INSERT_TAIL(&all_bdrv_states, bs, bs_list);
return bs;
return bdrv_do_find_format(format_name);
}
-static int bdrv_is_whitelisted(BlockDriver *drv, bool read_only)
+int bdrv_is_whitelisted(BlockDriver *drv, bool read_only)
{
static const char *whitelist_rw[] = {
CONFIG_BDRV_RW_WHITELIST
CreateCo *cco = opaque;
assert(cco->drv);
- ret = cco->drv->bdrv_create(cco->filename, cco->opts, &local_err);
+ ret = cco->drv->bdrv_co_create_opts(cco->filename, cco->opts, &local_err);
error_propagate(&cco->err, local_err);
cco->ret = ret;
}
.err = NULL,
};
- if (!drv->bdrv_create) {
+ if (!drv->bdrv_co_create_opts) {
error_setg(errp, "Driver '%s' does not support image creation", drv->format_name);
ret = -ENOTSUP;
goto out;
* Set the current 'total_sectors' value
* Return 0 on success, -errno on error.
*/
-static int refresh_total_sectors(BlockDriverState *bs, int64_t hint)
+int refresh_total_sectors(BlockDriverState *bs, int64_t hint)
{
BlockDriver *drv = bs->drv;
+ if (!drv) {
+ return -ENOMEDIUM;
+ }
+
/* Do not attempt drv->bdrv_getlength() on scsi-generic devices */
if (bdrv_is_sg(bs))
return 0;
}
}
+static BlockdevDetectZeroesOptions bdrv_parse_detect_zeroes(QemuOpts *opts,
+ int open_flags,
+ Error **errp)
+{
+ Error *local_err = NULL;
+ char *value = qemu_opt_get_del(opts, "detect-zeroes");
+ BlockdevDetectZeroesOptions detect_zeroes =
+ qapi_enum_parse(&BlockdevDetectZeroesOptions_lookup, value,
+ BLOCKDEV_DETECT_ZEROES_OPTIONS_OFF, &local_err);
+ g_free(value);
+ if (local_err) {
+ error_propagate(errp, local_err);
+ return detect_zeroes;
+ }
+
+ if (detect_zeroes == BLOCKDEV_DETECT_ZEROES_OPTIONS_UNMAP &&
+ !(open_flags & BDRV_O_UNMAP))
+ {
+ error_setg(errp, "setting detect-zeroes to unmap is not allowed "
+ "without setting discard operation to unmap");
+ }
+
+ return detect_zeroes;
+}
+
/**
* Set open flags for a given discard mode
*
static void bdrv_child_cb_drained_begin(BdrvChild *child)
{
BlockDriverState *bs = child->opaque;
- bdrv_drained_begin(bs);
+ bdrv_do_drained_begin_quiesce(bs, NULL, false);
+}
+
+static bool bdrv_child_cb_drained_poll(BdrvChild *child)
+{
+ BlockDriverState *bs = child->opaque;
+ return bdrv_drain_poll(bs, false, NULL, false);
}
static void bdrv_child_cb_drained_end(BdrvChild *child)
bdrv_drained_end(bs);
}
+static void bdrv_child_cb_attach(BdrvChild *child)
+{
+ BlockDriverState *bs = child->opaque;
+ bdrv_apply_subtree_drain(child, bs);
+}
+
+static void bdrv_child_cb_detach(BdrvChild *child)
+{
+ BlockDriverState *bs = child->opaque;
+ bdrv_unapply_subtree_drain(child, bs);
+}
+
static int bdrv_child_cb_inactivate(BdrvChild *child)
{
BlockDriverState *bs = child->opaque;
}
const BdrvChildRole child_file = {
+ .parent_is_bds = true,
.get_parent_desc = bdrv_child_get_parent_desc,
.inherit_options = bdrv_inherited_options,
.drained_begin = bdrv_child_cb_drained_begin,
+ .drained_poll = bdrv_child_cb_drained_poll,
.drained_end = bdrv_child_cb_drained_end,
+ .attach = bdrv_child_cb_attach,
+ .detach = bdrv_child_cb_detach,
.inactivate = bdrv_child_cb_inactivate,
};
}
const BdrvChildRole child_format = {
+ .parent_is_bds = true,
.get_parent_desc = bdrv_child_get_parent_desc,
.inherit_options = bdrv_inherited_fmt_options,
.drained_begin = bdrv_child_cb_drained_begin,
+ .drained_poll = bdrv_child_cb_drained_poll,
.drained_end = bdrv_child_cb_drained_end,
+ .attach = bdrv_child_cb_attach,
+ .detach = bdrv_child_cb_detach,
.inactivate = bdrv_child_cb_inactivate,
};
parent->backing_blocker);
bdrv_op_unblock(backing_hd, BLOCK_OP_TYPE_BACKUP_TARGET,
parent->backing_blocker);
+
+ bdrv_child_cb_attach(c);
}
static void bdrv_backing_detach(BdrvChild *c)
bdrv_op_unblock_all(c->bs, parent->backing_blocker);
error_free(parent->backing_blocker);
parent->backing_blocker = NULL;
+
+ bdrv_child_cb_detach(c);
}
/*
*child_flags = flags;
}
+static int bdrv_backing_update_filename(BdrvChild *c, BlockDriverState *base,
+ const char *filename, Error **errp)
+{
+ BlockDriverState *parent = c->opaque;
+ int orig_flags = bdrv_get_flags(parent);
+ int ret;
+
+ if (!(orig_flags & BDRV_O_RDWR)) {
+ ret = bdrv_reopen(parent, orig_flags | BDRV_O_RDWR, errp);
+ if (ret < 0) {
+ return ret;
+ }
+ }
+
+ ret = bdrv_change_backing_file(parent, filename,
+ base->drv ? base->drv->format_name : "");
+ if (ret < 0) {
+ error_setg_errno(errp, -ret, "Could not update backing file link");
+ }
+
+ if (!(orig_flags & BDRV_O_RDWR)) {
+ bdrv_reopen(parent, orig_flags, NULL);
+ }
+
+ return ret;
+}
+
const BdrvChildRole child_backing = {
+ .parent_is_bds = true,
.get_parent_desc = bdrv_child_get_parent_desc,
.attach = bdrv_backing_attach,
.detach = bdrv_backing_detach,
.inherit_options = bdrv_backing_options,
.drained_begin = bdrv_child_cb_drained_begin,
+ .drained_poll = bdrv_child_cb_drained_poll,
.drained_end = bdrv_child_cb_drained_end,
.inactivate = bdrv_child_cb_inactivate,
+ .update_filename = bdrv_backing_update_filename,
};
static int bdrv_open_flags(BlockDriverState *bs, int flags)
*flags &= ~BDRV_O_CACHE_MASK;
assert(qemu_opt_find(opts, BDRV_OPT_CACHE_NO_FLUSH));
- if (qemu_opt_get_bool(opts, BDRV_OPT_CACHE_NO_FLUSH, false)) {
+ if (qemu_opt_get_bool_del(opts, BDRV_OPT_CACHE_NO_FLUSH, false)) {
*flags |= BDRV_O_NO_FLUSH;
}
assert(qemu_opt_find(opts, BDRV_OPT_CACHE_DIRECT));
- if (qemu_opt_get_bool(opts, BDRV_OPT_CACHE_DIRECT, false)) {
+ if (qemu_opt_get_bool_del(opts, BDRV_OPT_CACHE_DIRECT, false)) {
*flags |= BDRV_O_NOCACHE;
}
*flags &= ~BDRV_O_RDWR;
assert(qemu_opt_find(opts, BDRV_OPT_READ_ONLY));
- if (!qemu_opt_get_bool(opts, BDRV_OPT_READ_ONLY, false)) {
+ if (!qemu_opt_get_bool_del(opts, BDRV_OPT_READ_ONLY, false)) {
*flags |= BDRV_O_RDWR;
}
goto out;
}
+ /* Make sure that the node name isn't truncated */
+ if (strlen(node_name) >= sizeof(bs->node_name)) {
+ error_setg(errp, "Node name too long");
+ goto out;
+ }
+
/* 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);
int open_flags, Error **errp)
{
Error *local_err = NULL;
- int ret;
+ int i, ret;
bdrv_assign_node_name(bs, node_name, &local_err);
if (local_err) {
assert(bdrv_min_mem_align(bs) != 0);
assert(is_power_of_2(bs->bl.request_alignment));
+ for (i = 0; i < bs->quiesce_counter; i++) {
+ if (drv->bdrv_co_drain_begin) {
+ drv->bdrv_co_drain_begin(bs);
+ }
+ }
+
return 0;
open_failed:
bs->drv = NULL;
ret = bdrv_open_driver(bs, drv, node_name, bs->options, flags, errp);
if (ret < 0) {
- QDECREF(bs->explicit_options);
+ qobject_unref(bs->explicit_options);
bs->explicit_options = NULL;
- QDECREF(bs->options);
+ qobject_unref(bs->options);
bs->options = NULL;
bdrv_unref(bs);
return NULL;
const char *driver_name = NULL;
const char *node_name = NULL;
const char *discard;
- const char *detect_zeroes;
QemuOpts *opts;
BlockDriver *drv;
Error *local_err = NULL;
}
}
- detect_zeroes = qemu_opt_get(opts, "detect-zeroes");
- if (detect_zeroes) {
- BlockdevDetectZeroesOptions value =
- qapi_enum_parse(&BlockdevDetectZeroesOptions_lookup,
- detect_zeroes,
- BLOCKDEV_DETECT_ZEROES_OPTIONS_OFF,
- &local_err);
- if (local_err) {
- error_propagate(errp, local_err);
- ret = -EINVAL;
- goto fail_opts;
- }
-
- if (value == BLOCKDEV_DETECT_ZEROES_OPTIONS_UNMAP &&
- !(bs->open_flags & BDRV_O_UNMAP))
- {
- error_setg(errp, "setting detect-zeroes to unmap is not allowed "
- "without setting discard operation to unmap");
- ret = -EINVAL;
- goto fail_opts;
- }
-
- bs->detect_zeroes = value;
+ bs->detect_zeroes =
+ bdrv_parse_detect_zeroes(opts, bs->open_flags, &local_err);
+ if (local_err) {
+ error_propagate(errp, local_err);
+ ret = -EINVAL;
+ goto fail_opts;
}
if (filename != NULL) {
options_obj = qobject_from_json(filename, errp);
if (!options_obj) {
- /* Work around qobject_from_json() lossage TODO fix that */
- if (errp && !*errp) {
- error_setg(errp, "Could not parse the JSON options");
- return NULL;
- }
error_prepend(errp, "Could not parse the JSON options: ");
return NULL;
}
- options = qobject_to_qdict(options_obj);
+ options = qobject_to(QDict, options_obj);
if (!options) {
- qobject_decref(options_obj);
+ qobject_unref(options_obj);
error_setg(errp, "Invalid JSON object given");
return NULL;
}
/* Options given in the filename have lower priority than options
* specified directly */
qdict_join(options, json_options, false);
- QDECREF(json_options);
+ qobject_unref(json_options);
*pfilename = NULL;
}
return 0;
}
-static int bdrv_child_check_perm(BdrvChild *c, uint64_t perm, uint64_t shared,
+static int bdrv_child_check_perm(BdrvChild *c, BlockReopenQueue *q,
+ uint64_t perm, uint64_t shared,
GSList *ignore_children, Error **errp);
static void bdrv_child_abort_perm_update(BdrvChild *c);
static void bdrv_child_set_perm(BdrvChild *c, uint64_t perm, uint64_t shared);
+typedef struct BlockReopenQueueEntry {
+ bool prepared;
+ BDRVReopenState state;
+ QSIMPLEQ_ENTRY(BlockReopenQueueEntry) entry;
+} BlockReopenQueueEntry;
+
+/*
+ * Return the flags that @bs will have after the reopens in @q have
+ * successfully completed. If @q is NULL (or @bs is not contained in @q),
+ * return the current flags.
+ */
+static int bdrv_reopen_get_flags(BlockReopenQueue *q, BlockDriverState *bs)
+{
+ BlockReopenQueueEntry *entry;
+
+ if (q != NULL) {
+ QSIMPLEQ_FOREACH(entry, q, entry) {
+ if (entry->state.bs == bs) {
+ return entry->state.flags;
+ }
+ }
+ }
+
+ return bs->open_flags;
+}
+
+/* Returns whether the image file can be written to after the reopen queue @q
+ * has been successfully applied, or right now if @q is NULL. */
+static bool bdrv_is_writable_after_reopen(BlockDriverState *bs,
+ BlockReopenQueue *q)
+{
+ int flags = bdrv_reopen_get_flags(q, bs);
+
+ return (flags & (BDRV_O_RDWR | BDRV_O_INACTIVE)) == BDRV_O_RDWR;
+}
+
+/*
+ * Return whether the BDS can be written to. This is not necessarily
+ * the same as !bdrv_is_read_only(bs), as inactivated images may not
+ * be written to but do not count as read-only images.
+ */
+bool bdrv_is_writable(BlockDriverState *bs)
+{
+ return bdrv_is_writable_after_reopen(bs, NULL);
+}
+
static void bdrv_child_perm(BlockDriverState *bs, BlockDriverState *child_bs,
- BdrvChild *c,
- const BdrvChildRole *role,
+ BdrvChild *c, const BdrvChildRole *role,
+ BlockReopenQueue *reopen_queue,
uint64_t parent_perm, uint64_t parent_shared,
uint64_t *nperm, uint64_t *nshared)
{
if (bs->drv && bs->drv->bdrv_child_perm) {
- bs->drv->bdrv_child_perm(bs, c, role,
+ bs->drv->bdrv_child_perm(bs, c, role, reopen_queue,
parent_perm, parent_shared,
nperm, nshared);
}
+ /* TODO Take force_share from reopen_queue */
if (child_bs && child_bs->force_share) {
*nshared = BLK_PERM_ALL;
}
* A call to this function must always be followed by a call to bdrv_set_perm()
* or bdrv_abort_perm_update().
*/
-static int bdrv_check_perm(BlockDriverState *bs, uint64_t cumulative_perms,
+static int bdrv_check_perm(BlockDriverState *bs, BlockReopenQueue *q,
+ uint64_t cumulative_perms,
uint64_t cumulative_shared_perms,
GSList *ignore_children, Error **errp)
{
/* Write permissions never work with read-only images */
if ((cumulative_perms & (BLK_PERM_WRITE | BLK_PERM_WRITE_UNCHANGED)) &&
- !bdrv_is_writable(bs))
+ !bdrv_is_writable_after_reopen(bs, q))
{
error_setg(errp, "Block node is read-only");
return -EPERM;
/* Check all children */
QLIST_FOREACH(c, &bs->children, next) {
uint64_t cur_perm, cur_shared;
- bdrv_child_perm(bs, c->bs, c, c->role,
+ bdrv_child_perm(bs, c->bs, c, c->role, q,
cumulative_perms, cumulative_shared_perms,
&cur_perm, &cur_shared);
- ret = bdrv_child_check_perm(c, cur_perm, cur_shared, ignore_children,
- errp);
+ ret = bdrv_child_check_perm(c, q, cur_perm, cur_shared,
+ ignore_children, errp);
if (ret < 0) {
return ret;
}
/* Update all children */
QLIST_FOREACH(c, &bs->children, next) {
uint64_t cur_perm, cur_shared;
- bdrv_child_perm(bs, c->bs, c, c->role,
+ bdrv_child_perm(bs, c->bs, c, c->role, NULL,
cumulative_perms, cumulative_shared_perms,
&cur_perm, &cur_shared);
bdrv_child_set_perm(c, cur_perm, cur_shared);
*
* Needs to be followed by a call to either bdrv_set_perm() or
* bdrv_abort_perm_update(). */
-static int bdrv_check_update_perm(BlockDriverState *bs, uint64_t new_used_perm,
+static int bdrv_check_update_perm(BlockDriverState *bs, BlockReopenQueue *q,
+ uint64_t new_used_perm,
uint64_t new_shared_perm,
GSList *ignore_children, Error **errp)
{
cumulative_shared_perms &= c->shared_perm;
}
- return bdrv_check_perm(bs, cumulative_perms, cumulative_shared_perms,
+ return bdrv_check_perm(bs, q, cumulative_perms, cumulative_shared_perms,
ignore_children, errp);
}
/* Needs to be followed by a call to either bdrv_child_set_perm() or
* bdrv_child_abort_perm_update(). */
-static int bdrv_child_check_perm(BdrvChild *c, uint64_t perm, uint64_t shared,
+static int bdrv_child_check_perm(BdrvChild *c, BlockReopenQueue *q,
+ uint64_t perm, uint64_t shared,
GSList *ignore_children, Error **errp)
{
int ret;
ignore_children = g_slist_prepend(g_slist_copy(ignore_children), c);
- ret = bdrv_check_update_perm(c->bs, perm, shared, ignore_children, errp);
+ ret = bdrv_check_update_perm(c->bs, q, perm, shared, ignore_children, errp);
g_slist_free(ignore_children);
return ret;
{
int ret;
- ret = bdrv_child_check_perm(c, perm, shared, NULL, errp);
+ ret = bdrv_child_check_perm(c, NULL, perm, shared, NULL, errp);
if (ret < 0) {
bdrv_child_abort_perm_update(c);
return ret;
return 0;
}
-#define DEFAULT_PERM_PASSTHROUGH (BLK_PERM_CONSISTENT_READ \
- | BLK_PERM_WRITE \
- | BLK_PERM_WRITE_UNCHANGED \
- | BLK_PERM_RESIZE)
-#define DEFAULT_PERM_UNCHANGED (BLK_PERM_ALL & ~DEFAULT_PERM_PASSTHROUGH)
-
void bdrv_filter_default_perms(BlockDriverState *bs, BdrvChild *c,
const BdrvChildRole *role,
+ BlockReopenQueue *reopen_queue,
uint64_t perm, uint64_t shared,
uint64_t *nperm, uint64_t *nshared)
{
void bdrv_format_default_perms(BlockDriverState *bs, BdrvChild *c,
const BdrvChildRole *role,
+ BlockReopenQueue *reopen_queue,
uint64_t perm, uint64_t shared,
uint64_t *nperm, uint64_t *nshared)
{
assert(role == &child_backing || role == &child_file);
if (!backing) {
+ int flags = bdrv_reopen_get_flags(reopen_queue, bs);
+
/* Apart from the modifications below, the same permissions are
* forwarded and left alone as for filters */
- bdrv_filter_default_perms(bs, c, role, perm, shared, &perm, &shared);
+ bdrv_filter_default_perms(bs, c, role, reopen_queue, perm, shared,
+ &perm, &shared);
/* Format drivers may touch metadata even if the guest doesn't write */
- if (bdrv_is_writable(bs)) {
+ if (bdrv_is_writable_after_reopen(bs, reopen_queue)) {
perm |= BLK_PERM_WRITE | BLK_PERM_RESIZE;
}
/* bs->file always needs to be consistent because of the metadata. We
* can never allow other users to resize or write to it. */
- perm |= BLK_PERM_CONSISTENT_READ;
+ if (!(flags & BDRV_O_NO_IO)) {
+ perm |= BLK_PERM_CONSISTENT_READ;
+ }
shared &= ~(BLK_PERM_WRITE | BLK_PERM_RESIZE);
} else {
/* We want consistent read from backing files if the parent needs it.
BlockDriverState *new_bs)
{
BlockDriverState *old_bs = child->bs;
+ int i;
if (old_bs && new_bs) {
assert(bdrv_get_aio_context(old_bs) == bdrv_get_aio_context(new_bs));
}
if (old_bs) {
- if (old_bs->quiesce_counter && child->role->drained_end) {
- child->role->drained_end(child);
- }
+ /* Detach first so that the recursive drain sections coming from @child
+ * are already gone and we only end the drain sections that came from
+ * elsewhere. */
if (child->role->detach) {
child->role->detach(child);
}
+ if (old_bs->quiesce_counter && child->role->drained_end) {
+ int num = old_bs->quiesce_counter;
+ if (child->role->parent_is_bds) {
+ num -= bdrv_drain_all_count;
+ }
+ assert(num >= 0);
+ for (i = 0; i < num; i++) {
+ child->role->drained_end(child);
+ }
+ }
QLIST_REMOVE(child, next_parent);
}
if (new_bs) {
QLIST_INSERT_HEAD(&new_bs->parents, child, next_parent);
if (new_bs->quiesce_counter && child->role->drained_begin) {
- child->role->drained_begin(child);
+ int num = new_bs->quiesce_counter;
+ if (child->role->parent_is_bds) {
+ num -= bdrv_drain_all_count;
+ }
+ assert(num >= 0);
+ for (i = 0; i < num; i++) {
+ bdrv_parent_drained_begin_single(child, true);
+ }
}
+ /* Attach only after starting new drained sections, so that recursive
+ * drain sections coming from @child don't get an extra .drained_begin
+ * callback. */
if (child->role->attach) {
child->role->attach(child);
}
* because we're just taking a parent away, so we're loosening
* restrictions. */
bdrv_get_cumulative_perm(old_bs, &perm, &shared_perm);
- bdrv_check_perm(old_bs, perm, shared_perm, NULL, &error_abort);
+ bdrv_check_perm(old_bs, NULL, perm, shared_perm, NULL, &error_abort);
bdrv_set_perm(old_bs, perm, shared_perm);
}
BdrvChild *child;
int ret;
- ret = bdrv_check_update_perm(child_bs, perm, shared_perm, NULL, errp);
+ ret = bdrv_check_update_perm(child_bs, NULL, perm, shared_perm, NULL, errp);
if (ret < 0) {
bdrv_abort_perm_update(child_bs);
return NULL;
assert(parent_bs->drv);
assert(bdrv_get_aio_context(parent_bs) == bdrv_get_aio_context(child_bs));
- bdrv_child_perm(parent_bs, child_bs, NULL, child_role,
+ bdrv_child_perm(parent_bs, child_bs, NULL, child_role, NULL,
perm, shared_perm, &perm, &shared_perm);
child = bdrv_root_attach_child(child_bs, child_name, child_role,
}
}
-static void bdrv_parent_cb_resize(BlockDriverState *bs)
-{
- BdrvChild *c;
- QLIST_FOREACH(c, &bs->parents, next_parent) {
- if (c->role->resize) {
- c->role->resize(c);
- }
- }
-}
-
/*
* Sets the backing file link of a BDS. A new reference is created; callers
* which don't need their own reference any more must call bdrv_unref().
if (reference || qdict_haskey(options, "file.filename")) {
backing_filename[0] = '\0';
} else if (bs->backing_file[0] == '\0' && qdict_size(options) == 0) {
- QDECREF(options);
+ qobject_unref(options);
goto free_exit;
} else {
bdrv_get_full_backing_filename(bs, backing_filename, PATH_MAX,
if (local_err) {
ret = -EINVAL;
error_propagate(errp, local_err);
- QDECREF(options);
+ qobject_unref(options);
goto free_exit;
}
}
if (!bs->drv || !bs->drv->supports_backing) {
ret = -EINVAL;
error_setg(errp, "Driver doesn't support backing files");
- QDECREF(options);
+ qobject_unref(options);
goto free_exit;
}
- if (bs->backing_format[0] != '\0' && !qdict_haskey(options, "driver")) {
+ if (!reference &&
+ bs->backing_format[0] != '\0' && !qdict_haskey(options, "driver")) {
qdict_put_str(options, "driver", bs->backing_format);
}
free_exit:
g_free(backing_filename);
- QDECREF(tmp_parent_options);
+ qobject_unref(tmp_parent_options);
return ret;
}
error_setg(errp, "A block device must be specified for \"%s\"",
bdref_key);
}
- QDECREF(image_options);
+ qobject_unref(image_options);
goto done;
}
return c;
}
+/* TODO Future callers may need to specify parent/child_role in order for
+ * option inheritance to work. Existing callers use it for the root node. */
+BlockDriverState *bdrv_open_blockdev_ref(BlockdevRef *ref, Error **errp)
+{
+ BlockDriverState *bs = NULL;
+ Error *local_err = NULL;
+ QObject *obj = NULL;
+ QDict *qdict = NULL;
+ const char *reference = NULL;
+ Visitor *v = NULL;
+
+ if (ref->type == QTYPE_QSTRING) {
+ reference = ref->u.reference;
+ } else {
+ BlockdevOptions *options = &ref->u.definition;
+ assert(ref->type == QTYPE_QDICT);
+
+ v = qobject_output_visitor_new(&obj);
+ visit_type_BlockdevOptions(v, NULL, &options, &local_err);
+ if (local_err) {
+ error_propagate(errp, local_err);
+ goto fail;
+ }
+ visit_complete(v, &obj);
+
+ qdict = qobject_to(QDict, obj);
+ qdict_flatten(qdict);
+
+ /* bdrv_open_inherit() defaults to the values in bdrv_flags (for
+ * compatibility with other callers) rather than what we want as the
+ * real defaults. Apply the defaults here instead. */
+ qdict_set_default_str(qdict, BDRV_OPT_CACHE_DIRECT, "off");
+ qdict_set_default_str(qdict, BDRV_OPT_CACHE_NO_FLUSH, "off");
+ qdict_set_default_str(qdict, BDRV_OPT_READ_ONLY, "off");
+ }
+
+ bs = bdrv_open_inherit(NULL, reference, qdict, 0, NULL, NULL, errp);
+ obj = NULL;
+
+fail:
+ qobject_unref(obj);
+ visit_free(v);
+ return bs;
+}
+
static BlockDriverState *bdrv_append_temp_snapshot(BlockDriverState *bs,
int flags,
QDict *snapshot_options,
}
out:
- QDECREF(snapshot_options);
+ qobject_unref(snapshot_options);
g_free(tmp_filename);
return bs_snapshot;
}
* 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_open.
+ * dictionary, it needs to use qobject_ref() 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.
BlockBackend *file = NULL;
BlockDriverState *bs;
BlockDriver *drv = NULL;
+ BdrvChild *child;
const char *drvname;
const char *backing;
Error *local_err = NULL;
if (reference) {
bool options_non_empty = options ? qdict_size(options) : false;
- QDECREF(options);
+ qobject_unref(options);
if (filename || options_non_empty) {
error_setg(errp, "Cannot reference an existing block device with "
/* See cautionary note on accessing @options above */
backing = qdict_get_try_str(options, "backing");
- if (backing && *backing == '\0') {
+ if (qobject_to(QNull, qdict_get(options, "backing")) != NULL ||
+ (backing && *backing == '\0'))
+ {
+ if (backing) {
+ warn_report("Use of \"backing\": \"\" is deprecated; "
+ "use \"backing\": null instead");
+ }
flags |= BDRV_O_NO_BACKING;
qdict_del(options, "backing");
}
goto fail;
}
if (file_bs != NULL) {
- file = blk_new(BLK_PERM_CONSISTENT_READ, BLK_PERM_ALL);
+ /* Not requesting BLK_PERM_CONSISTENT_READ because we're only
+ * looking at the header to guess the image format. This works even
+ * in cases where a guest would not see a consistent state. */
+ file = blk_new(0, BLK_PERM_ALL);
blk_insert_bs(file, file_bs, &local_err);
bdrv_unref(file_bs);
if (local_err) {
}
}
+ /* Remove all children options and references
+ * from bs->options and bs->explicit_options */
+ QLIST_FOREACH(child, &bs->children, next) {
+ char *child_key_dot;
+ child_key_dot = g_strdup_printf("%s.", child->name);
+ qdict_extract_subqdict(bs->explicit_options, NULL, child_key_dot);
+ qdict_extract_subqdict(bs->options, NULL, child_key_dot);
+ qdict_del(bs->explicit_options, child->name);
+ qdict_del(bs->options, child->name);
+ g_free(child_key_dot);
+ }
+
bdrv_refresh_filename(bs);
/* Check if any unknown options were used */
bdrv_parent_cb_change_media(bs, true);
- QDECREF(options);
+ qobject_unref(options);
+ options = NULL;
/* For snapshot=on, create a temporary qcow2 overlay. bs points to the
* temporary snapshot afterwards. */
fail:
blk_unref(file);
- QDECREF(snapshot_options);
- QDECREF(bs->explicit_options);
- QDECREF(bs->options);
- QDECREF(options);
+ qobject_unref(snapshot_options);
+ qobject_unref(bs->explicit_options);
+ qobject_unref(bs->options);
+ qobject_unref(options);
bs->options = NULL;
bs->explicit_options = NULL;
bdrv_unref(bs);
close_and_fail:
bdrv_unref(bs);
- QDECREF(snapshot_options);
- QDECREF(options);
+ qobject_unref(snapshot_options);
+ qobject_unref(options);
error_propagate(errp, local_err);
return NULL;
}
NULL, errp);
}
-typedef struct BlockReopenQueueEntry {
- bool prepared;
- BDRVReopenState state;
- QSIMPLEQ_ENTRY(BlockReopenQueueEntry) entry;
-} BlockReopenQueueEntry;
-
/*
* Adds a BlockDriverState to a simple queue for an atomic, transactional
* reopen of multiple devices.
* returns a pointer to bs_queue, which is either the newly allocated
* bs_queue, or the existing bs_queue being used.
*
+ * bs must be drained between bdrv_reopen_queue() and bdrv_reopen_multiple().
*/
static BlockReopenQueue *bdrv_reopen_queue_child(BlockReopenQueue *bs_queue,
BlockDriverState *bs,
BdrvChild *child;
QDict *old_options, *explicit_options;
+ /* Make sure that the caller remembered to use a drained section. This is
+ * important to avoid graph changes between the recursive queuing here and
+ * bdrv_reopen_multiple(). */
+ assert(bs->quiesce_counter > 0);
+
if (bs_queue == NULL) {
bs_queue = g_new0(BlockReopenQueue, 1);
QSIMPLEQ_INIT(bs_queue);
old_options = qdict_clone_shallow(bs->explicit_options);
}
bdrv_join_options(bs, options, old_options);
- QDECREF(old_options);
+ qobject_unref(old_options);
explicit_options = qdict_clone_shallow(options);
/* Inherit from parent node */
if (parent_options) {
+ QemuOpts *opts;
+ QDict *options_copy;
assert(!flags);
role->inherit_options(&flags, options, parent_flags, parent_options);
+ options_copy = qdict_clone_shallow(options);
+ opts = qemu_opts_create(&bdrv_runtime_opts, NULL, 0, &error_abort);
+ qemu_opts_absorb_qdict(opts, options_copy, NULL);
+ update_flags_from_options(&flags, opts);
+ qemu_opts_del(opts);
+ qobject_unref(options_copy);
}
/* Old values are used for options that aren't set yet */
old_options = qdict_clone_shallow(bs->options);
bdrv_join_options(bs, options, old_options);
- QDECREF(old_options);
+ qobject_unref(old_options);
/* bdrv_open_inherit() sets and clears some additional flags internally */
flags &= ~BDRV_O_PROTOCOL;
flags |= BDRV_O_ALLOW_RDWR;
}
+ if (!bs_entry) {
+ bs_entry = g_new0(BlockReopenQueueEntry, 1);
+ QSIMPLEQ_INSERT_TAIL(bs_queue, bs_entry, entry);
+ } else {
+ qobject_unref(bs_entry->state.options);
+ qobject_unref(bs_entry->state.explicit_options);
+ }
+
+ bs_entry->state.bs = bs;
+ bs_entry->state.options = options;
+ bs_entry->state.explicit_options = explicit_options;
+ bs_entry->state.flags = flags;
+
+ /* This needs to be overwritten in bdrv_reopen_prepare() */
+ bs_entry->state.perm = UINT64_MAX;
+ bs_entry->state.shared_perm = 0;
+
QLIST_FOREACH(child, &bs->children, next) {
QDict *new_child_options;
char *child_key_dot;
}
child_key_dot = g_strdup_printf("%s.", child->name);
+ qdict_extract_subqdict(explicit_options, NULL, child_key_dot);
qdict_extract_subqdict(options, &new_child_options, child_key_dot);
g_free(child_key_dot);
child->role, options, flags);
}
- if (!bs_entry) {
- bs_entry = g_new0(BlockReopenQueueEntry, 1);
- QSIMPLEQ_INSERT_TAIL(bs_queue, bs_entry, entry);
- } else {
- QDECREF(bs_entry->state.options);
- QDECREF(bs_entry->state.explicit_options);
- }
-
- bs_entry->state.bs = bs;
- bs_entry->state.options = options;
- bs_entry->state.explicit_options = explicit_options;
- bs_entry->state.flags = flags;
-
return bs_queue;
}
*
* Reopens all BDS specified in the queue, with the appropriate
* flags. All devices are prepared for reopen, and failure of any
- * device will cause all device changes to be abandonded, and intermediate
+ * device will cause all device changes to be abandoned, and intermediate
* data cleaned up.
*
* If all devices prepare successfully, then the changes are committed
* to all devices.
*
+ * All affected nodes must be drained between bdrv_reopen_queue() and
+ * bdrv_reopen_multiple().
*/
int bdrv_reopen_multiple(AioContext *ctx, BlockReopenQueue *bs_queue, Error **errp)
{
assert(bs_queue != NULL);
- aio_context_release(ctx);
- bdrv_drain_all_begin();
- aio_context_acquire(ctx);
-
QSIMPLEQ_FOREACH(bs_entry, bs_queue, entry) {
+ assert(bs_entry->state.bs->quiesce_counter > 0);
if (bdrv_reopen_prepare(&bs_entry->state, bs_queue, &local_err)) {
error_propagate(errp, local_err);
goto cleanup;
cleanup:
QSIMPLEQ_FOREACH_SAFE(bs_entry, bs_queue, entry, next) {
- if (ret && bs_entry->prepared) {
- bdrv_reopen_abort(&bs_entry->state);
- } else if (ret) {
- QDECREF(bs_entry->state.explicit_options);
+ if (ret) {
+ if (bs_entry->prepared) {
+ bdrv_reopen_abort(&bs_entry->state);
+ }
+ qobject_unref(bs_entry->state.explicit_options);
+ qobject_unref(bs_entry->state.options);
}
- QDECREF(bs_entry->state.options);
g_free(bs_entry);
}
g_free(bs_queue);
- bdrv_drain_all_end();
-
return ret;
}
{
int ret = -1;
Error *local_err = NULL;
- BlockReopenQueue *queue = bdrv_reopen_queue(NULL, bs, NULL, bdrv_flags);
+ BlockReopenQueue *queue;
+
+ bdrv_subtree_drained_begin(bs);
+ queue = bdrv_reopen_queue(NULL, bs, NULL, bdrv_flags);
ret = bdrv_reopen_multiple(bdrv_get_aio_context(bs), queue, &local_err);
if (local_err != NULL) {
error_propagate(errp, local_err);
}
+
+ bdrv_subtree_drained_end(bs);
+
return ret;
}
+static BlockReopenQueueEntry *find_parent_in_reopen_queue(BlockReopenQueue *q,
+ BdrvChild *c)
+{
+ BlockReopenQueueEntry *entry;
+
+ QSIMPLEQ_FOREACH(entry, q, entry) {
+ BlockDriverState *bs = entry->state.bs;
+ BdrvChild *child;
+
+ QLIST_FOREACH(child, &bs->children, next) {
+ if (child == c) {
+ return entry;
+ }
+ }
+ }
+
+ return NULL;
+}
+
+static void bdrv_reopen_perm(BlockReopenQueue *q, BlockDriverState *bs,
+ uint64_t *perm, uint64_t *shared)
+{
+ BdrvChild *c;
+ BlockReopenQueueEntry *parent;
+ uint64_t cumulative_perms = 0;
+ uint64_t cumulative_shared_perms = BLK_PERM_ALL;
+
+ QLIST_FOREACH(c, &bs->parents, next_parent) {
+ parent = find_parent_in_reopen_queue(q, c);
+ if (!parent) {
+ cumulative_perms |= c->perm;
+ cumulative_shared_perms &= c->shared_perm;
+ } else {
+ uint64_t nperm, nshared;
+
+ bdrv_child_perm(parent->state.bs, bs, c, c->role, q,
+ parent->state.perm, parent->state.shared_perm,
+ &nperm, &nshared);
+
+ cumulative_perms |= nperm;
+ cumulative_shared_perms &= nshared;
+ }
+ }
+ *perm = cumulative_perms;
+ *shared = cumulative_shared_perms;
+}
/*
* Prepares a BlockDriverState for reopen. All changes are staged in the
Error *local_err = NULL;
BlockDriver *drv;
QemuOpts *opts;
- const char *value;
+ QDict *orig_reopen_opts;
+ char *discard = NULL;
bool read_only;
assert(reopen_state != NULL);
assert(reopen_state->bs->drv != NULL);
drv = reopen_state->bs->drv;
+ /* This function and each driver's bdrv_reopen_prepare() remove
+ * entries from reopen_state->options as they are processed, so
+ * we need to make a copy of the original QDict. */
+ orig_reopen_opts = qdict_clone_shallow(reopen_state->options);
+
/* Process generic block layer options */
opts = qemu_opts_create(&bdrv_runtime_opts, NULL, 0, &error_abort);
qemu_opts_absorb_qdict(opts, reopen_state->options, &local_err);
update_flags_from_options(&reopen_state->flags, opts);
- /* node-name and driver must be unchanged. Put them back into the QDict, so
- * that they are checked at the end of this function. */
- value = qemu_opt_get(opts, "node-name");
- if (value) {
- qdict_put_str(reopen_state->options, "node-name", value);
+ discard = qemu_opt_get_del(opts, "discard");
+ if (discard != NULL) {
+ if (bdrv_parse_discard_flags(discard, &reopen_state->flags) != 0) {
+ error_setg(errp, "Invalid discard option");
+ ret = -EINVAL;
+ goto error;
+ }
}
- value = qemu_opt_get(opts, "driver");
- if (value) {
- qdict_put_str(reopen_state->options, "driver", value);
+ reopen_state->detect_zeroes =
+ bdrv_parse_detect_zeroes(opts, reopen_state->flags, &local_err);
+ if (local_err) {
+ error_propagate(errp, local_err);
+ ret = -EINVAL;
+ goto error;
}
+ /* All other options (including node-name and driver) must be unchanged.
+ * Put them back into the QDict, so that they are checked at the end
+ * of this function. */
+ qemu_opts_to_qdict(opts, reopen_state->options);
+
/* If we are to stay read-only, do not allow permission change
* to r/w. Attempting to set to r/w may fail if either BDRV_O_ALLOW_RDWR is
* not set, or if the BDS still has copy_on_read enabled */
goto error;
}
+ /* Calculate required permissions after reopening */
+ bdrv_reopen_perm(queue, reopen_state->bs,
+ &reopen_state->perm, &reopen_state->shared_perm);
ret = bdrv_flush(reopen_state->bs);
if (ret) {
const QDictEntry *entry = qdict_first(reopen_state->options);
do {
- QString *new_obj = qobject_to_qstring(entry->value);
- const char *new = qstring_get_str(new_obj);
+ QObject *new = entry->value;
+ QObject *old = qdict_get(reopen_state->bs->options, entry->key);
+
+ /* Allow child references (child_name=node_name) as long as they
+ * point to the current child (i.e. everything stays the same). */
+ if (qobject_type(new) == QTYPE_QSTRING) {
+ BdrvChild *child;
+ QLIST_FOREACH(child, &reopen_state->bs->children, next) {
+ if (!strcmp(child->name, entry->key)) {
+ break;
+ }
+ }
+
+ if (child) {
+ const char *str = qobject_get_try_str(new);
+ if (!strcmp(child->bs->node_name, str)) {
+ continue; /* Found child with this name, skip option */
+ }
+ }
+ }
+
/*
- * Caution: while qdict_get_try_str() is fine, getting
- * non-string types would require more care. When
- * bs->options come from -blockdev or blockdev_add, its
- * members are typed according to the QAPI schema, but
- * when they come from -drive, they're all QString.
+ * TODO: When using -drive to specify blockdev options, all values
+ * will be strings; however, when using -blockdev, blockdev-add or
+ * filenames using the json:{} pseudo-protocol, they will be
+ * correctly typed.
+ * In contrast, reopening options are (currently) always strings
+ * (because you can only specify them through qemu-io; all other
+ * callers do not specify any options).
+ * Therefore, when using anything other than -drive to create a BDS,
+ * this cannot detect non-string options as unchanged, because
+ * qobject_is_equal() always returns false for objects of different
+ * type. In the future, this should be remedied by correctly typing
+ * all options. For now, this is not too big of an issue because
+ * the user can simply omit options which cannot be changed anyway,
+ * so they will stay unchanged.
*/
- const char *old = qdict_get_try_str(reopen_state->bs->options,
- entry->key);
-
- if (!old || strcmp(new, old)) {
+ if (!qobject_is_equal(new, old)) {
error_setg(errp, "Cannot change the option '%s'", entry->key);
ret = -EINVAL;
goto error;
} while ((entry = qdict_next(reopen_state->options, entry)));
}
+ ret = bdrv_check_perm(reopen_state->bs, queue, reopen_state->perm,
+ reopen_state->shared_perm, NULL, errp);
+ if (ret < 0) {
+ goto error;
+ }
+
ret = 0;
+ /* Restore the original reopen_state->options QDict */
+ qobject_unref(reopen_state->options);
+ reopen_state->options = qobject_ref(orig_reopen_opts);
+
error:
qemu_opts_del(opts);
+ qobject_unref(orig_reopen_opts);
+ g_free(discard);
return ret;
}
{
BlockDriver *drv;
BlockDriverState *bs;
+ BdrvChild *child;
bool old_can_write, new_can_write;
assert(reopen_state != NULL);
}
/* set BDS specific flags now */
- QDECREF(bs->explicit_options);
+ qobject_unref(bs->explicit_options);
+ qobject_unref(bs->options);
bs->explicit_options = reopen_state->explicit_options;
+ bs->options = reopen_state->options;
bs->open_flags = reopen_state->flags;
bs->read_only = !(reopen_state->flags & BDRV_O_RDWR);
+ bs->detect_zeroes = reopen_state->detect_zeroes;
+
+ /* Remove child references from bs->options and bs->explicit_options.
+ * Child options were already removed in bdrv_reopen_queue_child() */
+ QLIST_FOREACH(child, &bs->children, next) {
+ qdict_del(bs->explicit_options, child->name);
+ qdict_del(bs->options, child->name);
+ }
bdrv_refresh_limits(bs, NULL);
+ bdrv_set_perm(reopen_state->bs, reopen_state->perm,
+ reopen_state->shared_perm);
+
new_can_write =
!bdrv_is_read_only(bs) && !(bdrv_get_flags(bs) & BDRV_O_INACTIVE);
if (!old_can_write && new_can_write && drv->bdrv_reopen_bitmaps_rw) {
drv->bdrv_reopen_abort(reopen_state);
}
- QDECREF(reopen_state->explicit_options);
+ bdrv_abort_perm_update(reopen_state->bs);
}
static void bdrv_close(BlockDriverState *bs)
{
BdrvAioNotifier *ban, *ban_next;
+ BdrvChild *child, *next;
assert(!bs->job);
assert(!bs->refcnt);
bdrv_drain(bs); /* in case flush left pending I/O */
if (bs->drv) {
- BdrvChild *child, *next;
-
- bs->drv->bdrv_close(bs);
+ if (bs->drv->bdrv_close) {
+ bs->drv->bdrv_close(bs);
+ }
bs->drv = NULL;
+ }
- bdrv_set_backing_hd(bs, NULL, &error_abort);
+ bdrv_set_backing_hd(bs, NULL, &error_abort);
- if (bs->file != NULL) {
- bdrv_unref_child(bs, bs->file);
- bs->file = NULL;
- }
+ if (bs->file != NULL) {
+ bdrv_unref_child(bs, bs->file);
+ bs->file = NULL;
+ }
- QLIST_FOREACH_SAFE(child, &bs->children, next, next) {
- /* TODO Remove bdrv_unref() from drivers' close function and use
- * bdrv_unref_child() here */
- if (child->bs->inherits_from == bs) {
- child->bs->inherits_from = NULL;
- }
- bdrv_detach_child(child);
+ QLIST_FOREACH_SAFE(child, &bs->children, next, next) {
+ /* TODO Remove bdrv_unref() from drivers' close function and use
+ * bdrv_unref_child() here */
+ if (child->bs->inherits_from == bs) {
+ child->bs->inherits_from = NULL;
}
-
- g_free(bs->opaque);
- bs->opaque = NULL;
- atomic_set(&bs->copy_on_read, 0);
- bs->backing_file[0] = '\0';
- bs->backing_format[0] = '\0';
- bs->total_sectors = 0;
- bs->encrypted = false;
- bs->sg = false;
- QDECREF(bs->options);
- QDECREF(bs->explicit_options);
- bs->options = NULL;
- bs->explicit_options = NULL;
- QDECREF(bs->full_open_options);
- bs->full_open_options = NULL;
+ bdrv_detach_child(child);
}
+ g_free(bs->opaque);
+ bs->opaque = NULL;
+ atomic_set(&bs->copy_on_read, 0);
+ bs->backing_file[0] = '\0';
+ bs->backing_format[0] = '\0';
+ bs->total_sectors = 0;
+ bs->encrypted = false;
+ bs->sg = false;
+ qobject_unref(bs->options);
+ qobject_unref(bs->explicit_options);
+ bs->options = NULL;
+ bs->explicit_options = NULL;
+ qobject_unref(bs->full_open_options);
+ bs->full_open_options = NULL;
+
bdrv_release_named_dirty_bitmaps(bs);
assert(QLIST_EMPTY(&bs->dirty_bitmaps));
void bdrv_close_all(void)
{
- block_job_cancel_sync_all();
+ assert(job_next(NULL) == NULL);
nbd_export_close_all();
/* Drop references from requests still in flight, such as canceled block
return false;
}
- if (c->role == &child_backing) {
- /* If @from is a backing file of @to, ignore the child to avoid
- * creating a loop. We only want to change the pointer of other
- * parents. */
- QLIST_FOREACH(to_c, &to->children, next) {
- if (to_c == c) {
- break;
- }
- }
- if (to_c) {
+ /* If the child @c belongs to the BDS @to, replacing the current
+ * c->bs by @to would mean to create a loop.
+ *
+ * Such a case occurs when appending a BDS to a backing chain.
+ * For instance, imagine the following chain:
+ *
+ * guest device -> node A -> further backing chain...
+ *
+ * Now we create a new BDS B which we want to put on top of this
+ * chain, so we first attach A as its backing node:
+ *
+ * node B
+ * |
+ * v
+ * guest device -> node A -> further backing chain...
+ *
+ * Finally we want to replace A by B. When doing that, we want to
+ * replace all pointers to A by pointers to B -- except for the
+ * pointer from B because (1) that would create a loop, and (2)
+ * that pointer should simply stay intact:
+ *
+ * guest device -> node B
+ * |
+ * v
+ * node A -> further backing chain...
+ *
+ * In general, when replacing a node A (c->bs) by a node B (@to),
+ * if A is a child of B, that means we cannot replace A by B there
+ * because that would create a loop. Silently detaching A from B
+ * is also not really an option. So overall just leaving A in
+ * place there is the most sensible choice. */
+ QLIST_FOREACH(to_c, &to->children, next) {
+ if (to_c == c) {
return false;
}
}
/* Put all parents into @list and calculate their cumulative permissions */
QLIST_FOREACH_SAFE(c, &from->parents, next_parent, next) {
+ assert(c->bs == from);
if (!should_update_child(c, to)) {
continue;
}
/* Check whether the required permissions can be granted on @to, ignoring
* all BdrvChild in @list so that they can't block themselves. */
- ret = bdrv_check_update_perm(to, perm, shared, list, errp);
+ ret = bdrv_check_update_perm(to, NULL, perm, shared, list, errp);
if (ret < 0) {
bdrv_abort_perm_update(to);
goto out;
* free of errors) or -errno when an internal error occurred. The results of the
* check are stored in res.
*/
-int bdrv_check(BlockDriverState *bs, BdrvCheckResult *res, BdrvCheckMode fix)
+static int coroutine_fn bdrv_co_check(BlockDriverState *bs,
+ BdrvCheckResult *res, BdrvCheckMode fix)
{
if (bs->drv == NULL) {
return -ENOMEDIUM;
}
- if (bs->drv->bdrv_check == NULL) {
+ if (bs->drv->bdrv_co_check == NULL) {
return -ENOTSUP;
}
memset(res, 0, sizeof(*res));
- return bs->drv->bdrv_check(bs, res, fix);
+ return bs->drv->bdrv_co_check(bs, res, fix);
+}
+
+typedef struct CheckCo {
+ BlockDriverState *bs;
+ BdrvCheckResult *res;
+ BdrvCheckMode fix;
+ int ret;
+} CheckCo;
+
+static void bdrv_check_co_entry(void *opaque)
+{
+ CheckCo *cco = opaque;
+ cco->ret = bdrv_co_check(cco->bs, cco->res, cco->fix);
+}
+
+int bdrv_check(BlockDriverState *bs,
+ BdrvCheckResult *res, BdrvCheckMode fix)
+{
+ Coroutine *co;
+ CheckCo cco = {
+ .bs = bs,
+ .res = res,
+ .ret = -EINPROGRESS,
+ .fix = fix,
+ };
+
+ if (qemu_in_coroutine()) {
+ /* Fast-path if already in coroutine context */
+ bdrv_check_co_entry(&cco);
+ } else {
+ co = qemu_coroutine_create(bdrv_check_co_entry, &cco);
+ qemu_coroutine_enter(co);
+ BDRV_POLL_WHILE(bs, cco.ret == -EINPROGRESS);
+ }
+
+ return cco.ret;
}
/*
BlockDriver *drv = bs->drv;
int ret;
+ if (!drv) {
+ return -ENOMEDIUM;
+ }
+
/* Backing file format doesn't make sense without a backing file */
if (backing_fmt && !backing_file) {
return -EINVAL;
* if active == top, that is considered an error
*
*/
-int bdrv_drop_intermediate(BlockDriverState *active, BlockDriverState *top,
- BlockDriverState *base, const char *backing_file_str)
+int bdrv_drop_intermediate(BlockDriverState *top, BlockDriverState *base,
+ const char *backing_file_str)
{
- BlockDriverState *new_top_bs = NULL;
+ BdrvChild *c, *next;
Error *local_err = NULL;
int ret = -EIO;
- if (!top->drv || !base->drv) {
- goto exit;
- }
-
- new_top_bs = bdrv_find_overlay(active, top);
-
- if (new_top_bs == NULL) {
- /* we could not find the image above 'top', this is an error */
- goto exit;
- }
+ bdrv_ref(top);
- /* special case of new_top_bs->backing->bs already pointing to base - nothing
- * to do, no intermediate images */
- if (backing_bs(new_top_bs) == base) {
- ret = 0;
+ if (!top->drv || !base->drv) {
goto exit;
}
}
/* success - we can delete the intermediate states, and link top->base */
+ /* TODO Check graph modification op blockers (BLK_PERM_GRAPH_MOD) once
+ * we've figured out how they should work. */
backing_file_str = backing_file_str ? backing_file_str : base->filename;
- ret = bdrv_change_backing_file(new_top_bs, backing_file_str,
- base->drv ? base->drv->format_name : "");
- if (ret) {
- goto exit;
- }
-
- bdrv_set_backing_hd(new_top_bs, base, &local_err);
- if (local_err) {
- ret = -EPERM;
- error_report_err(local_err);
- goto exit;
- }
- ret = 0;
-exit:
- return ret;
-}
-
-/**
- * Truncate file to 'offset' bytes (needed only for file protocols)
- */
-int bdrv_truncate(BdrvChild *child, int64_t offset, PreallocMode prealloc,
- Error **errp)
-{
- BlockDriverState *bs = child->bs;
- BlockDriver *drv = bs->drv;
- int ret;
-
- assert(child->perm & BLK_PERM_RESIZE);
-
- /* if bs->drv == NULL, bs is closed, so there's nothing to do here */
- if (!drv) {
- error_setg(errp, "No medium inserted");
- return -ENOMEDIUM;
- }
- if (!drv->bdrv_truncate) {
- if (bs->file && drv->is_filter) {
- return bdrv_truncate(bs->file, offset, prealloc, errp);
+ QLIST_FOREACH_SAFE(c, &top->parents, next_parent, next) {
+ /* Check whether we are allowed to switch c from top to base */
+ GSList *ignore_children = g_slist_prepend(NULL, c);
+ bdrv_check_update_perm(base, NULL, c->perm, c->shared_perm,
+ ignore_children, &local_err);
+ g_slist_free(ignore_children);
+ if (local_err) {
+ ret = -EPERM;
+ error_report_err(local_err);
+ goto exit;
}
- error_setg(errp, "Image format driver does not support resize");
- return -ENOTSUP;
- }
- if (bs->read_only) {
- error_setg(errp, "Image is read-only");
- return -EACCES;
- }
- assert(!(bs->open_flags & BDRV_O_INACTIVE));
+ /* If so, update the backing file path in the image file */
+ if (c->role->update_filename) {
+ ret = c->role->update_filename(c, base, backing_file_str,
+ &local_err);
+ if (ret < 0) {
+ bdrv_abort_perm_update(base);
+ error_report_err(local_err);
+ goto exit;
+ }
+ }
- ret = drv->bdrv_truncate(bs, offset, prealloc, errp);
- if (ret == 0) {
- ret = refresh_total_sectors(bs, offset >> BDRV_SECTOR_BITS);
- bdrv_dirty_bitmap_truncate(bs);
- bdrv_parent_cb_resize(bs);
- atomic_inc(&bs->write_gen);
+ /* Do the actual switch in the in-memory graph.
+ * Completes bdrv_check_update_perm() transaction internally. */
+ bdrv_ref(base);
+ bdrv_replace_child(c, base);
+ bdrv_unref(top);
}
+
+ ret = 0;
+exit:
+ bdrv_unref(top);
return ret;
}
return QTAILQ_NEXT(bs, node_list);
}
+BlockDriverState *bdrv_next_all_states(BlockDriverState *bs)
+{
+ if (!bs) {
+ return QTAILQ_FIRST(&all_bdrv_states);
+ }
+ return QTAILQ_NEXT(bs, bs_list);
+}
+
const char *bdrv_get_node_name(const BlockDriverState *bs)
{
return bs->node_name;
int bdrv_has_zero_init(BlockDriverState *bs)
{
- assert(bs->drv);
+ if (!bs->drv) {
+ return 0;
+ }
/* If BS is a copy on write image, it is initialized to
the contents of the base image, which may not be zeroes. */
bool bdrv_can_write_zeroes_with_unmap(BlockDriverState *bs)
{
- BlockDriverInfo bdi;
-
if (!(bs->open_flags & BDRV_O_UNMAP)) {
return false;
}
- if (bdrv_get_info(bs, &bdi) == 0) {
- return bdi.can_write_zeroes_with_unmap;
- }
-
- return false;
+ return bs->supported_zero_flags & BDRV_REQ_MAY_UNMAP;
}
const char *bdrv_get_encrypted_filename(BlockDriverState *bs)
bdrv_init();
}
-void bdrv_invalidate_cache(BlockDriverState *bs, Error **errp)
+static void coroutine_fn bdrv_co_invalidate_cache(BlockDriverState *bs,
+ Error **errp)
{
BdrvChild *child, *parent;
uint64_t perm, shared_perm;
}
QLIST_FOREACH(child, &bs->children, next) {
- bdrv_invalidate_cache(child->bs, &local_err);
+ bdrv_co_invalidate_cache(child->bs, &local_err);
if (local_err) {
error_propagate(errp, local_err);
return;
}
}
+ /*
+ * Update permissions, they may differ for inactive nodes.
+ *
+ * Note that the required permissions of inactive images are always a
+ * subset of the permissions required after activating the image. This
+ * allows us to just get the permissions upfront without restricting
+ * drv->bdrv_invalidate_cache().
+ *
+ * It also means that in error cases, we don't have to try and revert to
+ * the old permissions (which is an operation that could fail, too). We can
+ * just keep the extended permissions for the next time that an activation
+ * of the image is tried.
+ */
bs->open_flags &= ~BDRV_O_INACTIVE;
- if (bs->drv->bdrv_invalidate_cache) {
- bs->drv->bdrv_invalidate_cache(bs, &local_err);
+ bdrv_get_cumulative_perm(bs, &perm, &shared_perm);
+ ret = bdrv_check_perm(bs, NULL, perm, shared_perm, NULL, &local_err);
+ if (ret < 0) {
+ bs->open_flags |= BDRV_O_INACTIVE;
+ error_propagate(errp, local_err);
+ return;
+ }
+ bdrv_set_perm(bs, perm, shared_perm);
+
+ if (bs->drv->bdrv_co_invalidate_cache) {
+ bs->drv->bdrv_co_invalidate_cache(bs, &local_err);
if (local_err) {
bs->open_flags |= BDRV_O_INACTIVE;
error_propagate(errp, local_err);
return;
}
- /* Update permissions, they may differ for inactive nodes */
- bdrv_get_cumulative_perm(bs, &perm, &shared_perm);
- ret = bdrv_check_perm(bs, perm, shared_perm, NULL, &local_err);
- if (ret < 0) {
- bs->open_flags |= BDRV_O_INACTIVE;
- error_propagate(errp, local_err);
- return;
- }
- bdrv_set_perm(bs, perm, shared_perm);
-
QLIST_FOREACH(parent, &bs->parents, next_parent) {
if (parent->role->activate) {
parent->role->activate(parent, &local_err);
}
}
+typedef struct InvalidateCacheCo {
+ BlockDriverState *bs;
+ Error **errp;
+ bool done;
+} InvalidateCacheCo;
+
+static void coroutine_fn bdrv_invalidate_cache_co_entry(void *opaque)
+{
+ InvalidateCacheCo *ico = opaque;
+ bdrv_co_invalidate_cache(ico->bs, ico->errp);
+ ico->done = true;
+}
+
+void bdrv_invalidate_cache(BlockDriverState *bs, Error **errp)
+{
+ Coroutine *co;
+ InvalidateCacheCo ico = {
+ .bs = bs,
+ .done = false,
+ .errp = errp
+ };
+
+ if (qemu_in_coroutine()) {
+ /* Fast-path if already in coroutine context */
+ bdrv_invalidate_cache_co_entry(&ico);
+ } else {
+ co = qemu_coroutine_create(bdrv_invalidate_cache_co_entry, &ico);
+ qemu_coroutine_enter(co);
+ BDRV_POLL_WHILE(bs, !ico.done);
+ }
+}
+
void bdrv_invalidate_cache_all(Error **errp)
{
BlockDriverState *bs;
aio_context_release(aio_context);
if (local_err) {
error_propagate(errp, local_err);
+ bdrv_next_cleanup(&it);
return;
}
}
BdrvChild *child, *parent;
int ret;
+ if (!bs->drv) {
+ return -ENOMEDIUM;
+ }
+
if (!setting_flag && bs->drv->bdrv_inactivate) {
ret = bs->drv->bdrv_inactivate(bs);
if (ret < 0) {
/* Update permissions, they may differ for inactive nodes */
bdrv_get_cumulative_perm(bs, &perm, &shared_perm);
- bdrv_check_perm(bs, perm, shared_perm, NULL, &error_abort);
+ bdrv_check_perm(bs, NULL, perm, shared_perm, NULL, &error_abort);
bdrv_set_perm(bs, perm, shared_perm);
}
BdrvNextIterator it;
int ret = 0;
int pass;
+ GSList *aio_ctxs = NULL, *ctx;
for (bs = bdrv_first(&it); bs; bs = bdrv_next(&it)) {
- aio_context_acquire(bdrv_get_aio_context(bs));
+ AioContext *aio_context = bdrv_get_aio_context(bs);
+
+ if (!g_slist_find(aio_ctxs, aio_context)) {
+ aio_ctxs = g_slist_prepend(aio_ctxs, aio_context);
+ aio_context_acquire(aio_context);
+ }
}
/* We do two passes of inactivation. The first pass calls to drivers'
for (bs = bdrv_first(&it); bs; bs = bdrv_next(&it)) {
ret = bdrv_inactivate_recurse(bs, pass);
if (ret < 0) {
+ bdrv_next_cleanup(&it);
goto out;
}
}
}
out:
- for (bs = bdrv_first(&it); bs; bs = bdrv_next(&it)) {
- aio_context_release(bdrv_get_aio_context(bs));
+ for (ctx = aio_ctxs; ctx != NULL; ctx = ctx->next) {
+ AioContext *aio_context = ctx->data;
+ aio_context_release(aio_context);
}
+ g_slist_free(aio_ctxs);
return ret;
}
/* The size for the image must always be specified, unless we have a backing
* file and we have not been forbidden from opening it. */
- size = qemu_opt_get_size(opts, BLOCK_OPT_SIZE, 0);
+ size = qemu_opt_get_size(opts, BLOCK_OPT_SIZE, img_size);
if (backing_file && !(flags & BDRV_O_NO_BACKING)) {
BlockDriverState *bs;
char *full_backing = g_new0(char, PATH_MAX);
back_flags = flags;
back_flags &= ~(BDRV_O_RDWR | BDRV_O_SNAPSHOT | BDRV_O_NO_BACKING);
+ backing_options = qdict_new();
if (backing_fmt) {
- backing_options = qdict_new();
qdict_put_str(backing_options, "driver", backing_fmt);
}
+ qdict_put_bool(backing_options, BDRV_OPT_FORCE_SHARE, true);
bs = bdrv_open(full_backing, NULL, backing_options, back_flags,
&local_err);
AioContext *bdrv_get_aio_context(BlockDriverState *bs)
{
- return bs->aio_context;
+ return bs ? bs->aio_context : qemu_get_aio_context();
}
void bdrv_coroutine_enter(BlockDriverState *bs, Coroutine *co)
AioContext *ctx = bdrv_get_aio_context(bs);
aio_disable_external(ctx);
- bdrv_parent_drained_begin(bs);
+ bdrv_parent_drained_begin(bs, NULL, false);
bdrv_drain(bs); /* ensure there are no in-flight requests */
while (aio_poll(ctx, false)) {
*/
aio_context_acquire(new_context);
bdrv_attach_aio_context(bs, new_context);
- bdrv_parent_drained_end(bs);
+ bdrv_parent_drained_end(bs, NULL, false);
aio_enable_external(ctx);
aio_context_release(new_context);
}
}
int bdrv_amend_options(BlockDriverState *bs, QemuOpts *opts,
- BlockDriverAmendStatusCB *status_cb, void *cb_opaque)
+ BlockDriverAmendStatusCB *status_cb, void *cb_opaque,
+ Error **errp)
{
+ if (!bs->drv) {
+ error_setg(errp, "Node is ejected");
+ return -ENOMEDIUM;
+ }
if (!bs->drv->bdrv_amend_options) {
+ error_setg(errp, "Block driver '%s' does not support option amendment",
+ bs->drv->format_name);
return -ENOTSUP;
}
- return bs->drv->bdrv_amend_options(bs, opts, status_cb, cb_opaque);
+ return bs->drv->bdrv_amend_options(bs, opts, status_cb, cb_opaque, errp);
}
/* This function will be called by the bdrv_recurse_is_first_non_filter method
/* candidate is the first non filter */
if (perm) {
+ bdrv_next_cleanup(&it);
return true;
}
}
{
const QDictEntry *entry;
QemuOptDesc *desc;
- BdrvChild *child;
bool found_any = false;
- const char *p;
for (entry = qdict_first(bs->options); entry;
entry = qdict_next(bs->options, entry))
{
- /* Exclude options for children */
- QLIST_FOREACH(child, &bs->children, next) {
- if (strstart(qdict_entry_key(entry), child->name, &p)
- && (!*p || *p == '.'))
- {
- break;
- }
- }
- if (child) {
- continue;
- }
-
- /* And exclude all non-driver-specific options */
+ /* Exclude all non-driver-specific options */
for (desc = bdrv_runtime_opts.desc; desc->name; desc++) {
if (!strcmp(qdict_entry_key(entry), desc->name)) {
break;
continue;
}
- qobject_incref(qdict_entry_value(entry));
- qdict_put_obj(d, qdict_entry_key(entry), qdict_entry_value(entry));
+ qdict_put_obj(d, qdict_entry_key(entry),
+ qobject_ref(qdict_entry_value(entry)));
found_any = true;
}
* information before refreshing it */
bs->exact_filename[0] = '\0';
if (bs->full_open_options) {
- QDECREF(bs->full_open_options);
+ qobject_unref(bs->full_open_options);
bs->full_open_options = NULL;
}
opts = qdict_new();
append_open_options(opts, bs);
drv->bdrv_refresh_filename(bs, opts);
- QDECREF(opts);
+ qobject_unref(opts);
} else if (bs->file) {
/* Try to reconstruct valid information from the underlying file */
bool has_open_options;
bs->exact_filename[0] = '\0';
if (bs->full_open_options) {
- QDECREF(bs->full_open_options);
+ qobject_unref(bs->full_open_options);
bs->full_open_options = NULL;
}
* suffices without querying the (exact_)filename of this BDS. */
if (bs->file->bs->full_open_options) {
qdict_put_str(opts, "driver", drv->format_name);
- QINCREF(bs->file->bs->full_open_options);
- qdict_put(opts, "file", bs->file->bs->full_open_options);
+ qdict_put(opts, "file",
+ qobject_ref(bs->file->bs->full_open_options));
bs->full_open_options = opts;
} else {
- QDECREF(opts);
+ qobject_unref(opts);
}
} else if (!bs->full_open_options && qdict_size(bs->options)) {
/* There is no underlying file BDS (at least referenced by BDS.file),
QString *json = qobject_to_json(QOBJECT(bs->full_open_options));
snprintf(bs->filename, sizeof(bs->filename), "json:%s",
qstring_get_str(json));
- QDECREF(json);
+ qobject_unref(json);
}
}
projects / qemu.git / blobdiff