block: Use a single global AioWait

When draining a block node, we recurse to its parent and for subtree
drains also to its children. A single AIO_WAIT_WHILE() is then used to
wait for bdrv_drain_poll() to become true, which depends on all of the
nodes we recursed to. However, if the respective child or parent becomes
quiescent and calls bdrv_wakeup(), only the AioWait of the child/parent
is checked, while AIO_WAIT_WHILE() depends on the AioWait of the
original node.

Fix this by using a single AioWait for all callers of AIO_WAIT_WHILE().

This may mean that the draining thread gets a few more unnecessary
wakeups because an unrelated operation got completed, but we already
wake it up when something _could_ have changed rather than only if it
has certainly changed.

Apart from that, drain is a slow path anyway. In theory it would be
possible to use wakeups more selectively and still correctly, but the
gains are likely not worth the additional complexity. In fact, this
patch is a nice simplification for some places in the code.

Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Max Reitz <mreitz@redhat.com>

4 files changed, 12 insertions, 29 deletions

diff --git a/include/block/aio-wait.h b/include/block/aio-wait.h
index 600fad1..afd0ff7 100644
--- a/include/block/aio-wait.h
+++ b/include/block/aio-wait.h
@@ -30,14 +30,15 @@
 /**
  * AioWait:
  *
- * An object that facilitates synchronous waiting on a condition.  The main
- * loop can wait on an operation running in an IOThread as follows:
+ * An object that facilitates synchronous waiting on a condition. A single
+ * global AioWait object (global_aio_wait) is used internally.
+ *
+ * The main loop can wait on an operation running in an IOThread as follows:
  *
- *   AioWait *wait = ...;
  *   AioContext *ctx = ...;
  *   MyWork work = { .done = false };
  *   schedule_my_work_in_iothread(ctx, &work);
- *   AIO_WAIT_WHILE(wait, ctx, !work.done);
+ *   AIO_WAIT_WHILE(ctx, !work.done);
  *
  * The IOThread must call aio_wait_kick() to notify the main loop when
  * work.done changes:
@@ -46,7 +47,7 @@
  *   {
  *       ...
  *       work.done = true;
- *       aio_wait_kick(wait);
+ *       aio_wait_kick();
  *   }
  */
 typedef struct {
@@ -54,9 +55,10 @@ typedef struct {
     unsigned num_waiters;
 } AioWait;
 
+extern AioWait global_aio_wait;
+
 /**
  * AIO_WAIT_WHILE:
- * @wait: the aio wait object
  * @ctx: the aio context, or NULL if multiple aio contexts (for which the
  *       caller does not hold a lock) are involved in the polling condition.
  * @cond: wait while this conditional expression is true
@@ -72,9 +74,9 @@ typedef struct {
  * wait on conditions between two IOThreads since that could lead to deadlock,
  * go via the main loop instead.
  */
-#define AIO_WAIT_WHILE(wait, ctx, cond) ({                         \
+#define AIO_WAIT_WHILE(ctx, cond) ({                               \
     bool waited_ = false;                                          \
-    AioWait *wait_ = (wait);                                       \
+    AioWait *wait_ = &global_aio_wait;                             \
     AioContext *ctx_ = (ctx);                                      \
     /* Increment wait_->num_waiters before evaluating cond. */     \
     atomic_inc(&wait_->num_waiters);                               \
@@ -102,14 +104,12 @@ typedef struct {
 
 /**
  * aio_wait_kick:
- * @wait: the aio wait object that should re-evaluate its condition
- *
  * Wake up the main thread if it is waiting on AIO_WAIT_WHILE().  During
  * synchronous operations performed in an IOThread, the main thread lets the
  * IOThread's event loop run, waiting for the operation to complete.  A
  * aio_wait_kick() call will wake up the main thread.
  */
-void aio_wait_kick(AioWait *wait);
+void aio_wait_kick(void);
 
 /**
  * aio_wait_bh_oneshot:
diff --git a/include/block/block.h b/include/block/block.h
index 4e0871a..4edc1e8 100644
--- a/include/block/block.h
+++ b/include/block/block.h
@@ -410,13 +410,9 @@ void bdrv_drain_all_begin(void);
 void bdrv_drain_all_end(void);
 void bdrv_drain_all(void);
 
-/* Returns NULL when bs == NULL */
-AioWait *bdrv_get_aio_wait(BlockDriverState *bs);
-
 #define BDRV_POLL_WHILE(bs, cond) ({                       \
     BlockDriverState *bs_ = (bs);                          \
-    AIO_WAIT_WHILE(bdrv_get_aio_wait(bs_),                 \
-                   bdrv_get_aio_context(bs_),              \
+    AIO_WAIT_WHILE(bdrv_get_aio_context(bs_),              \
                    cond); })
 
 int bdrv_pdiscard(BdrvChild *child, int64_t offset, int bytes);
diff --git a/include/block/block_int.h b/include/block/block_int.h
index 4000d2a..92ecbd8 100644
--- a/include/block/block_int.h
+++ b/include/block/block_int.h
@@ -794,9 +794,6 @@ struct BlockDriverState {
     unsigned int in_flight;
     unsigned int serialising_in_flight;
 
-    /* Kicked to signal main loop when a request completes. */
-    AioWait wait;
-
     /* counter for nested bdrv_io_plug.
      * Accessed with atomic ops.
     */
diff --git a/include/block/blockjob.h b/include/block/blockjob.h
index 2290bbb..ede0bd8 100644
--- a/include/block/blockjob.h
+++ b/include/block/blockjob.h
@@ -122,16 +122,6 @@ int block_job_add_bdrv(BlockJob *job, const char *name, BlockDriverState *bs,
 void block_job_remove_all_bdrv(BlockJob *job);
 
 /**
- * block_job_wakeup_all_bdrv:
- * @job: The block job
- *
- * Calls bdrv_wakeup() for all BlockDriverStates that have been added to the
- * job. This function is to be called whenever child_job_drained_poll() would
- * go from true to false to notify waiting drain requests.
- */
-void block_job_wakeup_all_bdrv(BlockJob *job);
-
-/**
  * block_job_set_speed:
  * @job: The job to set the speed for.
  * @speed: The new value