[05/22] block: Mark drain related functions GRAPH_RDLOCK

Commit Message

Kevin Wolf Sept. 29, 2023, 2:51 p.m. UTC

From: Emanuele Giuseppe Esposito <[e-mail]>

Draining recursively traverses the graph, therefore we need to make sure
that also such accesses to the graph are protected by the graph rdlock.

There are 3 different drain callers to consider:
1. drain in the main loop: no issue at all, rdlock is nop.
2. drain in an iothread: rdlock only works in main loop or coroutines,
   so disallow it.
3. drain in a coroutine (regardless of AioContext): the drain mechanism
   takes care of scheduling a BH in the bs->aio_context that will
   then take care of perform the actual draining. This is wrong,
   because as pointed in (2) if bs->aio_context is an iothread then
   rdlock won't work. Therefore change bdrv_co_yield_to_drain to
   schedule the BH in the main loop.

Caller (2) also implies that we need to modify test-bdrv-drain.c to
disallow draining in the iothreads.

For some places, we know that they will hold the lock, but we don't have
the GRAPH_RDLOCK annotations yet. In this case, add assume_graph_lock()
with a FIXME comment. These places will be removed once everything is
properly annotated.

Signed-off-by: Emanuele Giuseppe Esposito <[e-mail]>
Signed-off-by: Kevin Wolf <[e-mail]>
---
 include/block/block-io.h         | 23 ++++++++++++++++++-----
 include/block/block_int-common.h |  6 +++---
 block.c                          |  6 +++---
 block/io.c                       | 32 ++++++++++++++++++++++++++++----
 tests/unit/test-bdrv-drain.c     |  4 ++--
 5 files changed, 54 insertions(+), 17 deletions(-)

Patch

diff --git a/include/block/block-io.h b/include/block/block-io.h
index f1c796a1ce..cf72e39717 100644
--- a/include/block/block-io.h
+++ b/include/block/block-io.h
@@ -363,7 +363,7 @@  bdrv_writev_vmstate(BlockDriverState *bs, QEMUIOVector *qiov, int64_t pos);
  *
  * Begin a quiesced section for the parent of @c.
  */
-void bdrv_parent_drained_begin_single(BdrvChild *c);
+void GRAPH_RDLOCK bdrv_parent_drained_begin_single(BdrvChild *c);
 
 /**
  * bdrv_parent_drained_poll_single:
@@ -371,14 +371,14 @@  void bdrv_parent_drained_begin_single(BdrvChild *c);
  * Returns true if there is any pending activity to cease before @c can be
  * called quiesced, false otherwise.
  */
-bool bdrv_parent_drained_poll_single(BdrvChild *c);
+bool GRAPH_RDLOCK bdrv_parent_drained_poll_single(BdrvChild *c);
 
 /**
  * bdrv_parent_drained_end_single:
  *
  * End a quiesced section for the parent of @c.
  */
-void bdrv_parent_drained_end_single(BdrvChild *c);
+void GRAPH_RDLOCK bdrv_parent_drained_end_single(BdrvChild *c);
 
 /**
  * bdrv_drain_poll:
@@ -391,8 +391,9 @@  void bdrv_parent_drained_end_single(BdrvChild *c);
  *
  * This is part of bdrv_drained_begin.
  */
-bool bdrv_drain_poll(BlockDriverState *bs, BdrvChild *ignore_parent,
-                     bool ignore_bds_parents);
+bool GRAPH_RDLOCK
+bdrv_drain_poll(BlockDriverState *bs, BdrvChild *ignore_parent,
+                bool ignore_bds_parents);
 
 /**
  * bdrv_drained_begin:
@@ -400,6 +401,12 @@  bool bdrv_drain_poll(BlockDriverState *bs, BdrvChild *ignore_parent,
  * Begin a quiesced section for exclusive access to the BDS, by disabling
  * external request sources including NBD server, block jobs, and device model.
  *
+ * This function can only be invoked by the main loop or a coroutine
+ * (regardless of the AioContext where it is running).
+ * If the coroutine is running in an Iothread AioContext, this function will
+ * just schedule a BH to run in the main loop.
+ * However, it cannot be directly called by an Iothread.
+ *
  * This function can be recursive.
  */
 void bdrv_drained_begin(BlockDriverState *bs);
@@ -416,6 +423,12 @@  void bdrv_do_drained_begin_quiesce(BlockDriverState *bs, BdrvChild *parent);
  * bdrv_drained_end:
  *
  * End a quiescent section started by bdrv_drained_begin().
+ *
+ * This function can only be invoked by the main loop or a coroutine
+ * (regardless of the AioContext where it is running).
+ * If the coroutine is running in an Iothread AioContext, this function will
+ * just schedule a BH to run in the main loop.
+ * However, it cannot be directly called by an Iothread.
  */
 void bdrv_drained_end(BlockDriverState *bs);
 
diff --git a/include/block/block_int-common.h b/include/block/block_int-common.h
index 2ca3758cb8..8ef68176a5 100644
--- a/include/block/block_int-common.h
+++ b/include/block/block_int-common.h
@@ -963,15 +963,15 @@  struct BdrvChildClass {
      * Note that this can be nested. If drained_begin() was called twice, new
      * I/O is allowed only after drained_end() was called twice, too.
      */
-    void (*drained_begin)(BdrvChild *child);
-    void (*drained_end)(BdrvChild *child);
+    void GRAPH_RDLOCK_PTR (*drained_begin)(BdrvChild *child);
+    void GRAPH_RDLOCK_PTR (*drained_end)(BdrvChild *child);
 
     /*
      * Returns whether the parent has pending requests for the child. This
      * callback is polled after .drained_begin() has been called until all
      * activity on the child has stopped.
      */
-    bool (*drained_poll)(BdrvChild *child);
+    bool GRAPH_RDLOCK_PTR (*drained_poll)(BdrvChild *child);
 
     /*
      * Notifies the parent that the filename of its child has changed (e.g.
diff --git a/block.c b/block.c
index 61705980dd..fb494f4a8b 100644
--- a/block.c
+++ b/block.c
@@ -1192,19 +1192,19 @@  static char *bdrv_child_get_parent_desc(BdrvChild *c)
     return g_strdup_printf("node '%s'", bdrv_get_node_name(parent));
 }
 
-static void bdrv_child_cb_drained_begin(BdrvChild *child)
+static void GRAPH_RDLOCK bdrv_child_cb_drained_begin(BdrvChild *child)
 {
     BlockDriverState *bs = child->opaque;
     bdrv_do_drained_begin_quiesce(bs, NULL);
 }
 
-static bool bdrv_child_cb_drained_poll(BdrvChild *child)
+static bool GRAPH_RDLOCK bdrv_child_cb_drained_poll(BdrvChild *child)
 {
     BlockDriverState *bs = child->opaque;
     return bdrv_drain_poll(bs, NULL, false);
 }
 
-static void bdrv_child_cb_drained_end(BdrvChild *child)
+static void GRAPH_RDLOCK bdrv_child_cb_drained_end(BdrvChild *child)
 {
     BlockDriverState *bs = child->opaque;
     bdrv_drained_end(bs);
diff --git a/block/io.c b/block/io.c
index ca5f018a7e..63f248d672 100644
--- a/block/io.c
+++ b/block/io.c
@@ -46,9 +46,12 @@  static void bdrv_parent_cb_resize(BlockDriverState *bs);
 static int coroutine_fn bdrv_co_do_pwrite_zeroes(BlockDriverState *bs,
     int64_t offset, int64_t bytes, BdrvRequestFlags flags);
 
-static void bdrv_parent_drained_begin(BlockDriverState *bs, BdrvChild *ignore)
+static void GRAPH_RDLOCK
+bdrv_parent_drained_begin(BlockDriverState *bs, BdrvChild *ignore)
 {
     BdrvChild *c, *next;
+    IO_OR_GS_CODE();
+    assert_bdrv_graph_readable();
 
     QLIST_FOREACH_SAFE(c, &bs->parents, next_parent, next) {
         if (c == ignore) {
@@ -70,9 +73,12 @@  void bdrv_parent_drained_end_single(BdrvChild *c)
     }
 }
 
-static void bdrv_parent_drained_end(BlockDriverState *bs, BdrvChild *ignore)
+static void GRAPH_RDLOCK
+bdrv_parent_drained_end(BlockDriverState *bs, BdrvChild *ignore)
 {
     BdrvChild *c;
+    IO_OR_GS_CODE();
+    assert_bdrv_graph_readable();
 
     QLIST_FOREACH(c, &bs->parents, next_parent) {
         if (c == ignore) {
@@ -84,17 +90,22 @@  static void bdrv_parent_drained_end(BlockDriverState *bs, BdrvChild *ignore)
 
 bool bdrv_parent_drained_poll_single(BdrvChild *c)
 {
+    IO_OR_GS_CODE();
+
     if (c->klass->drained_poll) {
         return c->klass->drained_poll(c);
     }
     return false;
 }
 
-static bool bdrv_parent_drained_poll(BlockDriverState *bs, BdrvChild *ignore,
-                                     bool ignore_bds_parents)
+static bool GRAPH_RDLOCK
+bdrv_parent_drained_poll(BlockDriverState *bs, BdrvChild *ignore,
+                         bool ignore_bds_parents)
 {
     BdrvChild *c, *next;
     bool busy = false;
+    IO_OR_GS_CODE();
+    assert_bdrv_graph_readable();
 
     QLIST_FOREACH_SAFE(c, &bs->parents, next_parent, next) {
         if (c == ignore || (ignore_bds_parents && c->klass->parent_is_bds)) {
@@ -114,6 +125,7 @@  void bdrv_parent_drained_begin_single(BdrvChild *c)
     c->quiesced_parent = true;
 
     if (c->klass->drained_begin) {
+        /* called with rdlock taken, but it doesn't really need it. */
         c->klass->drained_begin(c);
     }
 }
@@ -263,6 +275,9 @@  bool bdrv_drain_poll(BlockDriverState *bs, BdrvChild *ignore_parent,
 static bool bdrv_drain_poll_top_level(BlockDriverState *bs,
                                       BdrvChild *ignore_parent)
 {
+    GLOBAL_STATE_CODE();
+    GRAPH_RDLOCK_GUARD_MAINLOOP();
+
     return bdrv_drain_poll(bs, ignore_parent, false);
 }
 
@@ -362,6 +377,7 @@  static void bdrv_do_drained_begin(BlockDriverState *bs, BdrvChild *parent,
 
     /* Stop things in parent-to-child order */
     if (qatomic_fetch_inc(&bs->quiesce_counter) == 0) {
+        GRAPH_RDLOCK_GUARD_MAINLOOP();
         bdrv_parent_drained_begin(bs, parent);
         if (bs->drv && bs->drv->bdrv_drain_begin) {
             bs->drv->bdrv_drain_begin(bs);
@@ -407,12 +423,16 @@  static void bdrv_do_drained_end(BlockDriverState *bs, BdrvChild *parent)
         bdrv_co_yield_to_drain(bs, false, parent, false);
         return;
     }
+
+    /* At this point, we should be always running in the main loop. */
+    GLOBAL_STATE_CODE();
     assert(bs->quiesce_counter > 0);
     GLOBAL_STATE_CODE();
 
     /* Re-enable things in child-to-parent order */
     old_quiesce_counter = qatomic_fetch_dec(&bs->quiesce_counter);
     if (old_quiesce_counter == 1) {
+        GRAPH_RDLOCK_GUARD_MAINLOOP();
         if (bs->drv && bs->drv->bdrv_drain_end) {
             bs->drv->bdrv_drain_end(bs);
         }
@@ -436,6 +456,8 @@  void bdrv_drain(BlockDriverState *bs)
 static void bdrv_drain_assert_idle(BlockDriverState *bs)
 {
     BdrvChild *child, *next;
+    GLOBAL_STATE_CODE();
+    GRAPH_RDLOCK_GUARD_MAINLOOP();
 
     assert(qatomic_read(&bs->in_flight) == 0);
     QLIST_FOREACH_SAFE(child, &bs->children, next, next) {
@@ -449,7 +471,9 @@  static bool bdrv_drain_all_poll(void)
 {
     BlockDriverState *bs = NULL;
     bool result = false;
+
     GLOBAL_STATE_CODE();
+    GRAPH_RDLOCK_GUARD_MAINLOOP();
 
     /* bdrv_drain_poll() can't make changes to the graph and we are holding the
      * main AioContext lock, so iterating bdrv_next_all_states() is safe. */
diff --git a/tests/unit/test-bdrv-drain.c b/tests/unit/test-bdrv-drain.c
index bdd3757615..d734829778 100644
--- a/tests/unit/test-bdrv-drain.c
+++ b/tests/unit/test-bdrv-drain.c
@@ -1196,7 +1196,7 @@  static void coroutine_mixed_fn detach_by_parent_aio_cb(void *opaque, int ret)
     }
 }
 
-static void detach_by_driver_cb_drained_begin(BdrvChild *child)
+static void GRAPH_RDLOCK detach_by_driver_cb_drained_begin(BdrvChild *child)
 {
     struct detach_by_parent_data *data = &detach_by_parent_data;
 
@@ -1233,7 +1233,7 @@  static BdrvChildClass detach_by_driver_cb_class;
  *     state is messed up, but if it is only polled in the single
  *     BDRV_POLL_WHILE() at the end of the drain, this should work fine.
  */
-static void test_detach_indirect(bool by_parent_cb)
+static void TSA_NO_TSA test_detach_indirect(bool by_parent_cb)
 {
     BlockBackend *blk;
     BlockDriverState *parent_a, *parent_b, *a, *b, *c;