Merge tag 'for-upstream' of https://repo.or.cz/qemu/kevin into staging

Block layer patches

- Cleanup bs->backing and bs->file handling
- Refactor bdrv_try_set_aio_context using transactions
- Changes for improved coroutine_fn consistency
- vhost-user-blk: fix the resize crash
- io_uring: Use of io_uring_register_ring_fd() led to breakage, revert
- vvfat: Fix some problems with r/w mode
- Code cleanup
- MAINTAINERS: Fold "Block QAPI, monitor, ..." into "Block layer core"

# -----BEGIN PGP SIGNATURE-----
#
# iQJFBAABCAAvFiEE3D3rFZqa+V09dFb+fwmycsiPL9YFAmNazhIRHGt3b2xmQHJl
# ZGhhdC5jb20ACgkQfwmycsiPL9ZyTw/8Dfck/SuxfyeLlnQItkjaV4cnqWOU8vHs
# 9x0KhlptCs+HXdF/3iicpA0lHojn7mNnbdFGjPRY4E0LriQv91TQ5ycdEmrseFPf
# sgeQlgdKCVU/pHjZ2wYarm2pE43Cx85a5xuufmw+7w49dNNZn14l4t+DgviuClVM
# nuVaogfZFbYyetre+Qd2TgLl+gJ+0d4o7Zs5lSWLrT8t0L9AGkcWPA7Nrbl6loIE
# dOautV4G7jLjuMiCeJZOGcnuRVe3gCQ5rCGBFzzH4DUtz4BmiYx4hd3LMEsP0PMM
# CrsfDZS04Ztybl9M7TmJuwkAm1gx1JDMOuJuh18lbJocIOBvhkKKxY2wI5LIdZVI
# ZntmU36RowkX+GGu/PYpYyMjBDClJppZCl7vnjyLYsVt6r0Vu6SmlHpJhcRYabhe
# 96Kv1LXH9A6+ogKPU3Layw6JGjg01GNr1ALuT7PO3pGto/JshmOuBEJJDucoF84M
# 5AfxFCohMROVldwblA6M0eKnlQBgtr5BvtgbV54BBo88VlFJgDJFQn7R09cTFUEo
# UwaJoS+nIaiZ0bQQVZhZloVppUaTdVJojzfVRCZZctga96/tu1HSFnGLnbEFpUN3
# KOf+XnVNS6Ro+nPSDf9bMjbIom2JicGFfV+6yMgIoxY/d5UA2dTZfefil4TAlSod
# 6PsTgg+jrm8=
# =/Fw0
# -----END PGP SIGNATURE-----
# gpg: Signature made Thu 27 Oct 2022 14:29:38 EDT
# gpg:                using RSA key DC3DEB159A9AF95D3D7456FE7F09B272C88F2FD6
# gpg:                issuer "kwolf@redhat.com"
# gpg: Good signature from "Kevin Wolf <kwolf@redhat.com>" [full]
# Primary key fingerprint: DC3D EB15 9A9A F95D 3D74  56FE 7F09 B272 C88F 2FD6

* tag 'for-upstream' of https://repo.or.cz/qemu/kevin: (58 commits)
  block/block-backend: blk_set_enable_write_cache is IO_CODE
  monitor: switch to *_co_* functions
  vmdk: switch to *_co_* functions
  vhdx: switch to *_co_* functions
  vdi: switch to *_co_* functions
  qed: switch to *_co_* functions
  qcow2: switch to *_co_* functions
  qcow: switch to *_co_* functions
  parallels: switch to *_co_* functions
  mirror: switch to *_co_* functions
  block: switch to *_co_* functions
  commit: switch to *_co_* functions
  vmdk: manually add more coroutine_fn annotations
  qcow2: manually add more coroutine_fn annotations
  qcow: manually add more coroutine_fn annotations
  blkdebug: add missing coroutine_fn annotation for indirect-called functions
  qcow2: add coroutine_fn annotation for indirect-called functions
  block: add missing coroutine_fn annotation to BlockDriverState callbacks
  coroutine-io: add missing coroutine_fn annotation to prototypes
  coroutine-lock: add missing coroutine_fn annotation to prototypes
  ...

Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>

7 files changed, 98 insertions, 36 deletions

diff --git a/include/block/block-common.h b/include/block/block-common.h
index 061606e867..297704c1e9 100644
--- a/include/block/block-common.h
+++ b/include/block/block-common.h
@@ -322,6 +322,45 @@ enum {
  *
  * At least one of DATA, METADATA, FILTERED, or COW must be set for
  * every child.
+ *
+ *
+ * = Connection with bs->children, bs->file and bs->backing fields =
+ *
+ * 1. Filters
+ *
+ * Filter drivers have drv->is_filter = true.
+ *
+ * Filter node has exactly one FILTERED|PRIMARY child, and may have other
+ * children which must not have these bits (one example is the
+ * copy-before-write filter, which also has its target DATA child).
+ *
+ * Filter nodes never have COW children.
+ *
+ * For most filters, the filtered child is linked in bs->file, bs->backing is
+ * NULL.  For some filters (as an exception), it is the other way around; those
+ * drivers will have drv->filtered_child_is_backing set to true (see that
+ * field’s documentation for what drivers this concerns)
+ *
+ * 2. "raw" driver (block/raw-format.c)
+ *
+ * Formally it's not a filter (drv->is_filter = false)
+ *
+ * bs->backing is always NULL
+ *
+ * Only has one child, linked in bs->file. Its role is either FILTERED|PRIMARY
+ * (like filter) or DATA|PRIMARY depending on options.
+ *
+ * 3. Other drivers
+ *
+ * Don't have any FILTERED children.
+ *
+ * May have at most one COW child. In this case it's linked in bs->backing.
+ * Otherwise bs->backing is NULL. COW child is never PRIMARY.
+ *
+ * May have at most one PRIMARY child. In this case it's linked in bs->file.
+ * Otherwise bs->file is NULL.
+ *
+ * May also have some other children that don't have the PRIMARY or COW bit set.
  */
 enum BdrvChildRoleBits {
     /*
diff --git a/include/block/block-global-state.h b/include/block/block-global-state.h
index eba4ed23b4..bb42ed9559 100644
--- a/include/block/block-global-state.h
+++ b/include/block/block-global-state.h
@@ -76,6 +76,9 @@ BdrvChild *bdrv_open_child(const char *filename,
                            const BdrvChildClass *child_class,
                            BdrvChildRole child_role,
                            bool allow_none, Error **errp);
+int bdrv_open_file_child(const char *filename,
+                         QDict *options, const char *bdref_key,
+                         BlockDriverState *parent, Error **errp);
 BlockDriverState *bdrv_open_blockdev_ref(BlockdevRef *ref, Error **errp);
 int bdrv_set_backing_hd(BlockDriverState *bs, BlockDriverState *backing_hd,
                         Error **errp);
@@ -217,17 +220,12 @@ void coroutine_fn bdrv_co_lock(BlockDriverState *bs);
  */
 void coroutine_fn bdrv_co_unlock(BlockDriverState *bs);
 
-void bdrv_set_aio_context_ignore(BlockDriverState *bs,
-                                 AioContext *new_context, GSList **ignore);
-int bdrv_try_set_aio_context(BlockDriverState *bs, AioContext *ctx,
-                             Error **errp);
-int bdrv_child_try_set_aio_context(BlockDriverState *bs, AioContext *ctx,
-                                   BdrvChild *ignore_child, Error **errp);
-bool bdrv_child_can_set_aio_context(BdrvChild *c, AioContext *ctx,
-                                    GSList **ignore, Error **errp);
-bool bdrv_can_set_aio_context(BlockDriverState *bs, AioContext *ctx,
-                              GSList **ignore, Error **errp);
 AioContext *bdrv_child_get_parent_aio_context(BdrvChild *c);
+bool bdrv_child_change_aio_context(BdrvChild *c, AioContext *ctx,
+                                   GHashTable *visited, Transaction *tran,
+                                   Error **errp);
+int bdrv_try_change_aio_context(BlockDriverState *bs, AioContext *ctx,
+                                BdrvChild *ignore_child, Error **errp);
 
 int bdrv_probe_blocksizes(BlockDriverState *bs, BlockSizes *bsz);
 int bdrv_probe_geometry(BlockDriverState *bs, HDGeometry *geo);
diff --git a/include/block/block-hmp-cmds.h b/include/block/block-hmp-cmds.h
index 50ce0247c3..ba0593c440 100644
--- a/include/block/block-hmp-cmds.h
+++ b/include/block/block-hmp-cmds.h
@@ -38,7 +38,7 @@ void hmp_nbd_server_add(Monitor *mon, const QDict *qdict);
 void hmp_nbd_server_remove(Monitor *mon, const QDict *qdict);
 void hmp_nbd_server_stop(Monitor *mon, const QDict *qdict);
 
-void hmp_block_resize(Monitor *mon, const QDict *qdict);
+void coroutine_fn hmp_block_resize(Monitor *mon, const QDict *qdict);
 void hmp_block_stream(Monitor *mon, const QDict *qdict);
 void hmp_block_passwd(Monitor *mon, const QDict *qdict);
 void hmp_block_set_io_throttle(Monitor *mon, const QDict *qdict);
diff --git a/include/block/block-io.h b/include/block/block-io.h
index 492f95fc05..770ddeb7c8 100644
--- a/include/block/block-io.h
+++ b/include/block/block-io.h
@@ -83,12 +83,13 @@ void bdrv_aio_cancel(BlockAIOCB *acb);
 void bdrv_aio_cancel_async(BlockAIOCB *acb);
 
 /* sg packet commands */
-int bdrv_co_ioctl(BlockDriverState *bs, int req, void *buf);
+int coroutine_fn bdrv_co_ioctl(BlockDriverState *bs, int req, void *buf);
 
 /* Ensure contents are flushed to disk.  */
 int coroutine_fn bdrv_co_flush(BlockDriverState *bs);
 
-int bdrv_co_pdiscard(BdrvChild *child, int64_t offset, int64_t bytes);
+int coroutine_fn bdrv_co_pdiscard(BdrvChild *child, int64_t offset,
+                                  int64_t bytes);
 bool bdrv_can_write_zeroes_with_unmap(BlockDriverState *bs);
 int bdrv_block_status(BlockDriverState *bs, int64_t offset,
                       int64_t bytes, int64_t *pnum, int64_t *map,
diff --git a/include/block/block_int-common.h b/include/block/block_int-common.h
index 9c569be162..5a2cc077a0 100644
--- a/include/block/block_int-common.h
+++ b/include/block/block_int-common.h
@@ -120,6 +120,20 @@ struct BlockDriver {
      */
     bool is_filter;
     /*
+     * Only make sense for filter drivers, for others must be false.
+     * If true, filtered child is bs->backing. Otherwise it's bs->file.
+     * Two internal filters use bs->backing as filtered child and has this
+     * field set to true: mirror_top and commit_top. There also two such test
+     * filters in tests/unit/test-bdrv-graph-mod.c.
+     *
+     * Never create any more such filters!
+     *
+     * TODO: imagine how to deprecate this behavior and make all filters work
+     * similarly using bs->file as filtered child.
+     */
+    bool filtered_child_is_backing;
+
+    /*
      * Set to true if the BlockDriver is a format driver.  Format nodes
      * generally do not expect their children to be other format nodes
      * (except for backing files), and so format probing is disabled
@@ -734,13 +748,11 @@ struct BlockDriver {
     void coroutine_fn (*bdrv_co_drain_end)(BlockDriverState *bs);
 
     bool (*bdrv_supports_persistent_dirty_bitmap)(BlockDriverState *bs);
-    bool (*bdrv_co_can_store_new_dirty_bitmap)(BlockDriverState *bs,
-                                               const char *name,
-                                               uint32_t granularity,
-                                               Error **errp);
-    int (*bdrv_co_remove_persistent_dirty_bitmap)(BlockDriverState *bs,
-                                                  const char *name,
-                                                  Error **errp);
+    bool coroutine_fn (*bdrv_co_can_store_new_dirty_bitmap)(
+        BlockDriverState *bs, const char *name, uint32_t granularity,
+        Error **errp);
+    int coroutine_fn (*bdrv_co_remove_persistent_dirty_bitmap)(
+        BlockDriverState *bs, const char *name, Error **errp);
 };
 
 static inline bool block_driver_can_compress(BlockDriver *drv)
@@ -895,9 +907,9 @@ struct BdrvChildClass {
     int (*update_filename)(BdrvChild *child, BlockDriverState *new_base,
                            const char *filename, Error **errp);
 
-    bool (*can_set_aio_ctx)(BdrvChild *child, AioContext *ctx,
-                        GSList **ignore, Error **errp);
-    void (*set_aio_ctx)(BdrvChild *child, AioContext *ctx, GSList **ignore);
+    bool (*change_aio_ctx)(BdrvChild *child, AioContext *ctx,
+                           GHashTable *visited, Transaction *tran,
+                           Error **errp);
 
     AioContext *(*get_parent_aio_context)(BdrvChild *child);
 
@@ -1045,9 +1057,6 @@ struct BlockDriverState {
     QDict *full_open_options;
     char exact_filename[PATH_MAX];
 
-    BdrvChild *backing;
-    BdrvChild *file;
-
     /* I/O Limits */
     BlockLimits bl;
 
@@ -1106,7 +1115,19 @@ struct BlockDriverState {
      * parent node of this node.
      */
     BlockDriverState *inherits_from;
+
+    /*
+     * @backing and @file are some of @children or NULL. All these three fields
+     * (@file, @backing and @children) are modified only in
+     * bdrv_child_cb_attach() and bdrv_child_cb_detach().
+     *
+     * See also comment in include/block/block.h, to learn how backing and file
+     * are connected with BdrvChildRole.
+     */
     QLIST_HEAD(, BdrvChild) children;
+    BdrvChild *backing;
+    BdrvChild *file;
+
     QLIST_HEAD(, BdrvChild) parents;
 
     QDict *options;
@@ -1233,7 +1254,7 @@ static inline BlockDriverState *child_bs(BdrvChild *child)
 }
 
 int bdrv_check_request(int64_t offset, int64_t bytes, Error **errp);
-int get_tmp_filename(char *filename, int size);
+char *create_tmp_file(Error **errp);
 void bdrv_parse_filename_strip_prefix(const char *filename, const char *prefix,
                                       QDict *options);
 
diff --git a/include/monitor/hmp.h b/include/monitor/hmp.h
index a9cf064ee8..dfbc0c9a2f 100644
--- a/include/monitor/hmp.h
+++ b/include/monitor/hmp.h
@@ -15,6 +15,7 @@
 #define HMP_H
 
 #include "qemu/readline.h"
+#include "qemu/coroutine.h"
 #include "qapi/qapi-types-common.h"
 
 bool hmp_handle_error(Monitor *mon, Error *err);
@@ -81,7 +82,7 @@ void hmp_netdev_del(Monitor *mon, const QDict *qdict);
 void hmp_getfd(Monitor *mon, const QDict *qdict);
 void hmp_closefd(Monitor *mon, const QDict *qdict);
 void hmp_sendkey(Monitor *mon, const QDict *qdict);
-void hmp_screendump(Monitor *mon, const QDict *qdict);
+void coroutine_fn hmp_screendump(Monitor *mon, const QDict *qdict);
 void hmp_chardev_add(Monitor *mon, const QDict *qdict);
 void hmp_chardev_change(Monitor *mon, const QDict *qdict);
 void hmp_chardev_remove(Monitor *mon, const QDict *qdict);
diff --git a/include/qemu/coroutine.h b/include/qemu/coroutine.h
index 608fe45dcf..89650a2d7f 100644
--- a/include/qemu/coroutine.h
+++ b/include/qemu/coroutine.h
@@ -287,7 +287,7 @@ void qemu_co_rwlock_init(CoRwlock *lock);
  * of a parallel writer, control is transferred to the caller of the current
  * coroutine.
  */
-void qemu_co_rwlock_rdlock(CoRwlock *lock);
+void coroutine_fn qemu_co_rwlock_rdlock(CoRwlock *lock);
 
 /**
  * Write Locks the CoRwlock from a reader.  This is a bit more efficient than
@@ -296,7 +296,7 @@ void qemu_co_rwlock_rdlock(CoRwlock *lock);
  * to the caller of the current coroutine; another writer might run while
  * @qemu_co_rwlock_upgrade blocks.
  */
-void qemu_co_rwlock_upgrade(CoRwlock *lock);
+void coroutine_fn qemu_co_rwlock_upgrade(CoRwlock *lock);
 
 /**
  * Downgrades a write-side critical section to a reader.  Downgrading with
@@ -304,20 +304,20 @@ void qemu_co_rwlock_upgrade(CoRwlock *lock);
  * followed by @qemu_co_rwlock_rdlock.  This makes it more efficient, but
  * may also sometimes be necessary for correctness.
  */
-void qemu_co_rwlock_downgrade(CoRwlock *lock);
+void coroutine_fn qemu_co_rwlock_downgrade(CoRwlock *lock);
 
 /**
  * Write Locks the mutex. If the lock cannot be taken immediately because
  * of a parallel reader, control is transferred to the caller of the current
  * coroutine.
  */
-void qemu_co_rwlock_wrlock(CoRwlock *lock);
+void coroutine_fn qemu_co_rwlock_wrlock(CoRwlock *lock);
 
 /**
  * Unlocks the read/write lock and schedules the next coroutine that was
  * waiting for this lock to be run.
  */
-void qemu_co_rwlock_unlock(CoRwlock *lock);
+void coroutine_fn qemu_co_rwlock_unlock(CoRwlock *lock);
 
 typedef struct QemuCoSleep {
     Coroutine *to_wake;
@@ -389,8 +389,9 @@ void qemu_coroutine_dec_pool_size(unsigned int additional_pool_size);
  * The same interface as qemu_sendv_recvv(), with added yielding.
  * XXX should mark these as coroutine_fn
  */
-ssize_t qemu_co_sendv_recvv(int sockfd, struct iovec *iov, unsigned iov_cnt,
-                            size_t offset, size_t bytes, bool do_send);
+ssize_t coroutine_fn qemu_co_sendv_recvv(int sockfd, struct iovec *iov,
+                                         unsigned iov_cnt, size_t offset,
+                                         size_t bytes, bool do_send);
 #define qemu_co_recvv(sockfd, iov, iov_cnt, offset, bytes) \
   qemu_co_sendv_recvv(sockfd, iov, iov_cnt, offset, bytes, false)
 #define qemu_co_sendv(sockfd, iov, iov_cnt, offset, bytes) \
@@ -399,7 +400,8 @@ ssize_t qemu_co_sendv_recvv(int sockfd, struct iovec *iov, unsigned iov_cnt,
 /**
  * The same as above, but with just a single buffer
  */
-ssize_t qemu_co_send_recv(int sockfd, void *buf, size_t bytes, bool do_send);
+ssize_t coroutine_fn qemu_co_send_recv(int sockfd, void *buf, size_t bytes,
+                                       bool do_send);
 #define qemu_co_recv(sockfd, buf, bytes) \
   qemu_co_send_recv(sockfd, buf, bytes, false)
 #define qemu_co_send(sockfd, buf, bytes) \