[PATCH v3 0/5] block-copy: use aio-task-pool

[PATCH v3 0/5] block-copy: use aio-task-pool

[Vladimir Sementsov-Ogievskiy]

Hi all!

v3:
01: drop extra line "+    s->in_flight_bytes -= task->bytes - new_bytes;"
02: add Max's r-b
03: rebased on 01 fix (dropped line not updated now). keep Max's r-b
04: more refactoring:
     don't require offset argument of block_copy_task_create being dirty,
     use !bdrv_dirty_bitmap_next_dirty_area() in block_copy_task_create()
05: declare block_copy_task_entry (as moving patch dropped)
    rebased on 04 changes

==

This is the next step of improving block-copy: use aio task pool.

Async copying loop has better performance than linear, which is shown
in original series (was
"[RFC 00/24] backup performance: block_status + async".

Vladimir Sementsov-Ogievskiy (5):
  block/block-copy: rename in-flight requests to tasks
  block/block-copy: alloc task on each iteration
  block/block-copy: add state pointer to BlockCopyTask
  block/block-copy: refactor task creation
  block/block-copy: use aio-task-pool API

 block/block-copy.c | 262 ++++++++++++++++++++++++++++++---------------
 1 file changed, 177 insertions(+), 85 deletions(-)

-- 
2.21.0

[PATCH v3 1/5] block/block-copy: rename in-flight requests to tasks

[Vladimir Sementsov-Ogievskiy]

We are going to use aio-task-pool API and extend in-flight request
structure to be a successor of AioTask, so rename things appropriately.

Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
---
 block/block-copy.c | 98 +++++++++++++++++++++++-----------------------
 1 file changed, 48 insertions(+), 50 deletions(-)

diff --git a/block/block-copy.c b/block/block-copy.c
index 05227e18bf..bbb29366dc 100644
--- a/block/block-copy.c
+++ b/block/block-copy.c
@@ -24,12 +24,12 @@
 #define BLOCK_COPY_MAX_BUFFER (1 * MiB)
 #define BLOCK_COPY_MAX_MEM (128 * MiB)
 
-typedef struct BlockCopyInFlightReq {
+typedef struct BlockCopyTask {
     int64_t offset;
     int64_t bytes;
-    QLIST_ENTRY(BlockCopyInFlightReq) list;
-    CoQueue wait_queue; /* coroutines blocked on this request */
-} BlockCopyInFlightReq;
+    QLIST_ENTRY(BlockCopyTask) list;
+    CoQueue wait_queue; /* coroutines blocked on this task */
+} BlockCopyTask;
 
 typedef struct BlockCopyState {
     /*
@@ -45,7 +45,7 @@ typedef struct BlockCopyState {
     bool use_copy_range;
     int64_t copy_size;
     uint64_t len;
-    QLIST_HEAD(, BlockCopyInFlightReq) inflight_reqs;
+    QLIST_HEAD(, BlockCopyTask) tasks;
 
     BdrvRequestFlags write_flags;
 
@@ -73,15 +73,14 @@ typedef struct BlockCopyState {
     SharedResource *mem;
 } BlockCopyState;
 
-static BlockCopyInFlightReq *find_conflicting_inflight_req(BlockCopyState *s,
-                                                           int64_t offset,
-                                                           int64_t bytes)
+static BlockCopyTask *find_conflicting_task(BlockCopyState *s,
+                                            int64_t offset, int64_t bytes)
 {
-    BlockCopyInFlightReq *req;
+    BlockCopyTask *t;
 
-    QLIST_FOREACH(req, &s->inflight_reqs, list) {
-        if (offset + bytes > req->offset && offset < req->offset + req->bytes) {
-            return req;
+    QLIST_FOREACH(t, &s->tasks, list) {
+        if (offset + bytes > t->offset && offset < t->offset + t->bytes) {
+            return t;
         }
     }
 
@@ -89,73 +88,72 @@ static BlockCopyInFlightReq *find_conflicting_inflight_req(BlockCopyState *s,
 }
 
 /*
- * If there are no intersecting requests return false. Otherwise, wait for the
- * first found intersecting request to finish and return true.
+ * If there are no intersecting tasks return false. Otherwise, wait for the
+ * first found intersecting tasks to finish and return true.
  */
 static bool coroutine_fn block_copy_wait_one(BlockCopyState *s, int64_t offset,
                                              int64_t bytes)
 {
-    BlockCopyInFlightReq *req = find_conflicting_inflight_req(s, offset, bytes);
+    BlockCopyTask *task = find_conflicting_task(s, offset, bytes);
 
-    if (!req) {
+    if (!task) {
         return false;
     }
 
-    qemu_co_queue_wait(&req->wait_queue, NULL);
+    qemu_co_queue_wait(&task->wait_queue, NULL);
 
     return true;
 }
 
 /* Called only on full-dirty region */
-static void block_copy_inflight_req_begin(BlockCopyState *s,
-                                          BlockCopyInFlightReq *req,
-                                          int64_t offset, int64_t bytes)
+static void block_copy_task_begin(BlockCopyState *s, BlockCopyTask *task,
+                                  int64_t offset, int64_t bytes)
 {
-    assert(!find_conflicting_inflight_req(s, offset, bytes));
+    assert(!find_conflicting_task(s, offset, bytes));
 
     bdrv_reset_dirty_bitmap(s->copy_bitmap, offset, bytes);
     s->in_flight_bytes += bytes;
 
-    req->offset = offset;
-    req->bytes = bytes;
-    qemu_co_queue_init(&req->wait_queue);
-    QLIST_INSERT_HEAD(&s->inflight_reqs, req, list);
+    task->offset = offset;
+    task->bytes = bytes;
+    qemu_co_queue_init(&task->wait_queue);
+    QLIST_INSERT_HEAD(&s->tasks, task, list);
 }
 
 /*
- * block_copy_inflight_req_shrink
+ * block_copy_task_shrink
  *
- * Drop the tail of the request to be handled later. Set dirty bits back and
- * wake up all requests waiting for us (may be some of them are not intersecting
- * with shrunk request)
+ * Drop the tail of the task to be handled later. Set dirty bits back and
+ * wake up all tasks waiting for us (may be some of them are not intersecting
+ * with shrunk task)
  */
-static void coroutine_fn block_copy_inflight_req_shrink(BlockCopyState *s,
-        BlockCopyInFlightReq *req, int64_t new_bytes)
+static void coroutine_fn block_copy_task_shrink(BlockCopyState *s,
+                                                BlockCopyTask *task,
+                                                int64_t new_bytes)
 {
-    if (new_bytes == req->bytes) {
+    if (new_bytes == task->bytes) {
         return;
     }
 
-    assert(new_bytes > 0 && new_bytes < req->bytes);
+    assert(new_bytes > 0 && new_bytes < task->bytes);
 
-    s->in_flight_bytes -= req->bytes - new_bytes;
+    s->in_flight_bytes -= task->bytes - new_bytes;
     bdrv_set_dirty_bitmap(s->copy_bitmap,
-                          req->offset + new_bytes, req->bytes - new_bytes);
+                          task->offset + new_bytes, task->bytes - new_bytes);
 
-    req->bytes = new_bytes;
-    qemu_co_queue_restart_all(&req->wait_queue);
+    task->bytes = new_bytes;
+    qemu_co_queue_restart_all(&task->wait_queue);
 }
 
-static void coroutine_fn block_copy_inflight_req_end(BlockCopyState *s,
-                                                     BlockCopyInFlightReq *req,
-                                                     int ret)
+static void coroutine_fn block_copy_task_end(BlockCopyState *s,
+                                             BlockCopyTask *task, int ret)
 {
-    s->in_flight_bytes -= req->bytes;
+    s->in_flight_bytes -= task->bytes;
     if (ret < 0) {
-        bdrv_set_dirty_bitmap(s->copy_bitmap, req->offset, req->bytes);
+        bdrv_set_dirty_bitmap(s->copy_bitmap, task->offset, task->bytes);
     }
-    QLIST_REMOVE(req, list);
-    qemu_co_queue_restart_all(&req->wait_queue);
+    QLIST_REMOVE(task, list);
+    qemu_co_queue_restart_all(&task->wait_queue);
 }
 
 void block_copy_state_free(BlockCopyState *s)
@@ -223,7 +221,7 @@ BlockCopyState *block_copy_state_new(BdrvChild *source, BdrvChild *target,
         s->copy_size = MAX(s->cluster_size, BLOCK_COPY_MAX_BUFFER);
     }
 
-    QLIST_INIT(&s->inflight_reqs);
+    QLIST_INIT(&s->tasks);
 
     return s;
 }
@@ -474,7 +472,7 @@ static int coroutine_fn block_copy_dirty_clusters(BlockCopyState *s,
     assert(QEMU_IS_ALIGNED(bytes, s->cluster_size));
 
     while (bytes) {
-        BlockCopyInFlightReq req;
+        BlockCopyTask task;
         int64_t next_zero, cur_bytes, status_bytes;
 
         if (!bdrv_dirty_bitmap_get(s->copy_bitmap, offset)) {
@@ -495,14 +493,14 @@ static int coroutine_fn block_copy_dirty_clusters(BlockCopyState *s,
             assert(next_zero < offset + cur_bytes); /* no need to do MIN() */
             cur_bytes = next_zero - offset;
         }
-        block_copy_inflight_req_begin(s, &req, offset, cur_bytes);
+        block_copy_task_begin(s, &task, offset, cur_bytes);
 
         ret = block_copy_block_status(s, offset, cur_bytes, &status_bytes);
         assert(ret >= 0); /* never fail */
         cur_bytes = MIN(cur_bytes, status_bytes);
-        block_copy_inflight_req_shrink(s, &req, cur_bytes);
+        block_copy_task_shrink(s, &task, cur_bytes);
         if (s->skip_unallocated && !(ret & BDRV_BLOCK_ALLOCATED)) {
-            block_copy_inflight_req_end(s, &req, 0);
+            block_copy_task_end(s, &task, 0);
             progress_set_remaining(s->progress,
                                    bdrv_get_dirty_count(s->copy_bitmap) +
                                    s->in_flight_bytes);
@@ -518,7 +516,7 @@ static int coroutine_fn block_copy_dirty_clusters(BlockCopyState *s,
         ret = block_copy_do_copy(s, offset, cur_bytes, ret & BDRV_BLOCK_ZERO,
                                  error_is_read);
         co_put_to_shres(s->mem, cur_bytes);
-        block_copy_inflight_req_end(s, &req, ret);
+        block_copy_task_end(s, &task, ret);
         if (ret < 0) {
             return ret;
         }
-- 
2.21.0

[PATCH v3 2/5] block/block-copy: alloc task on each iteration

[Vladimir Sementsov-Ogievskiy]

We are going to use aio-task-pool API, so tasks will be handled in
parallel. We need therefore separate allocated task on each iteration.
Introduce this logic now.

Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
Reviewed-by: Max Reitz <mreitz@redhat.com>
---
 block/block-copy.c | 18 +++++++++++-------
 1 file changed, 11 insertions(+), 7 deletions(-)

diff --git a/block/block-copy.c b/block/block-copy.c
index bbb29366dc..8d1b9ab9f0 100644
--- a/block/block-copy.c
+++ b/block/block-copy.c
@@ -106,9 +106,11 @@ static bool coroutine_fn block_copy_wait_one(BlockCopyState *s, int64_t offset,
 }
 
 /* Called only on full-dirty region */
-static void block_copy_task_begin(BlockCopyState *s, BlockCopyTask *task,
-                                  int64_t offset, int64_t bytes)
+static BlockCopyTask *block_copy_task_create(BlockCopyState *s,
+                                             int64_t offset, int64_t bytes)
 {
+    BlockCopyTask *task = g_new(BlockCopyTask, 1);
+
     assert(!find_conflicting_task(s, offset, bytes));
 
     bdrv_reset_dirty_bitmap(s->copy_bitmap, offset, bytes);
@@ -118,6 +120,8 @@ static void block_copy_task_begin(BlockCopyState *s, BlockCopyTask *task,
     task->bytes = bytes;
     qemu_co_queue_init(&task->wait_queue);
     QLIST_INSERT_HEAD(&s->tasks, task, list);
+
+    return task;
 }
 
 /*
@@ -472,7 +476,7 @@ static int coroutine_fn block_copy_dirty_clusters(BlockCopyState *s,
     assert(QEMU_IS_ALIGNED(bytes, s->cluster_size));
 
     while (bytes) {
-        BlockCopyTask task;
+        g_autofree BlockCopyTask *task = NULL;
         int64_t next_zero, cur_bytes, status_bytes;
 
         if (!bdrv_dirty_bitmap_get(s->copy_bitmap, offset)) {
@@ -493,14 +497,14 @@ static int coroutine_fn block_copy_dirty_clusters(BlockCopyState *s,
             assert(next_zero < offset + cur_bytes); /* no need to do MIN() */
             cur_bytes = next_zero - offset;
         }
-        block_copy_task_begin(s, &task, offset, cur_bytes);
+        task = block_copy_task_create(s, offset, cur_bytes);
 
         ret = block_copy_block_status(s, offset, cur_bytes, &status_bytes);
         assert(ret >= 0); /* never fail */
         cur_bytes = MIN(cur_bytes, status_bytes);
-        block_copy_task_shrink(s, &task, cur_bytes);
+        block_copy_task_shrink(s, task, cur_bytes);
         if (s->skip_unallocated && !(ret & BDRV_BLOCK_ALLOCATED)) {
-            block_copy_task_end(s, &task, 0);
+            block_copy_task_end(s, task, 0);
             progress_set_remaining(s->progress,
                                    bdrv_get_dirty_count(s->copy_bitmap) +
                                    s->in_flight_bytes);
@@ -516,7 +520,7 @@ static int coroutine_fn block_copy_dirty_clusters(BlockCopyState *s,
         ret = block_copy_do_copy(s, offset, cur_bytes, ret & BDRV_BLOCK_ZERO,
                                  error_is_read);
         co_put_to_shres(s->mem, cur_bytes);
-        block_copy_task_end(s, &task, ret);
+        block_copy_task_end(s, task, ret);
         if (ret < 0) {
             return ret;
         }
-- 
2.21.0

[PATCH v3 3/5] block/block-copy: add state pointer to BlockCopyTask

[Vladimir Sementsov-Ogievskiy]

We are going to use aio-task-pool API, so we'll need state pointer in
BlockCopyTask anyway. Add it now and use where possible.

Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
Reviewed-by: Max Reitz <mreitz@redhat.com>
---
 block/block-copy.c | 28 +++++++++++++++-------------
 1 file changed, 15 insertions(+), 13 deletions(-)

diff --git a/block/block-copy.c b/block/block-copy.c
index 8d1b9ab9f0..35ff9cc3ef 100644
--- a/block/block-copy.c
+++ b/block/block-copy.c
@@ -25,6 +25,7 @@
 #define BLOCK_COPY_MAX_MEM (128 * MiB)
 
 typedef struct BlockCopyTask {
+    BlockCopyState *s;
     int64_t offset;
     int64_t bytes;
     QLIST_ENTRY(BlockCopyTask) list;
@@ -116,8 +117,11 @@ static BlockCopyTask *block_copy_task_create(BlockCopyState *s,
     bdrv_reset_dirty_bitmap(s->copy_bitmap, offset, bytes);
     s->in_flight_bytes += bytes;
 
-    task->offset = offset;
-    task->bytes = bytes;
+    *task = (BlockCopyTask) {
+        .s = s,
+        .offset = offset,
+        .bytes = bytes,
+    };
     qemu_co_queue_init(&task->wait_queue);
     QLIST_INSERT_HEAD(&s->tasks, task, list);
 
@@ -131,8 +135,7 @@ static BlockCopyTask *block_copy_task_create(BlockCopyState *s,
  * wake up all tasks waiting for us (may be some of them are not intersecting
  * with shrunk task)
  */
-static void coroutine_fn block_copy_task_shrink(BlockCopyState *s,
-                                                BlockCopyTask *task,
+static void coroutine_fn block_copy_task_shrink(BlockCopyTask *task,
                                                 int64_t new_bytes)
 {
     if (new_bytes == task->bytes) {
@@ -141,20 +144,19 @@ static void coroutine_fn block_copy_task_shrink(BlockCopyState *s,
 
     assert(new_bytes > 0 && new_bytes < task->bytes);
 
-    s->in_flight_bytes -= task->bytes - new_bytes;
-    bdrv_set_dirty_bitmap(s->copy_bitmap,
+    task->s->in_flight_bytes -= task->bytes - new_bytes;
+    bdrv_set_dirty_bitmap(task->s->copy_bitmap,
                           task->offset + new_bytes, task->bytes - new_bytes);
 
     task->bytes = new_bytes;
     qemu_co_queue_restart_all(&task->wait_queue);
 }
 
-static void coroutine_fn block_copy_task_end(BlockCopyState *s,
-                                             BlockCopyTask *task, int ret)
+static void coroutine_fn block_copy_task_end(BlockCopyTask *task, int ret)
 {
-    s->in_flight_bytes -= task->bytes;
+    task->s->in_flight_bytes -= task->bytes;
     if (ret < 0) {
-        bdrv_set_dirty_bitmap(s->copy_bitmap, task->offset, task->bytes);
+        bdrv_set_dirty_bitmap(task->s->copy_bitmap, task->offset, task->bytes);
     }
     QLIST_REMOVE(task, list);
     qemu_co_queue_restart_all(&task->wait_queue);
@@ -502,9 +504,9 @@ static int coroutine_fn block_copy_dirty_clusters(BlockCopyState *s,
         ret = block_copy_block_status(s, offset, cur_bytes, &status_bytes);
         assert(ret >= 0); /* never fail */
         cur_bytes = MIN(cur_bytes, status_bytes);
-        block_copy_task_shrink(s, task, cur_bytes);
+        block_copy_task_shrink(task, cur_bytes);
         if (s->skip_unallocated && !(ret & BDRV_BLOCK_ALLOCATED)) {
-            block_copy_task_end(s, task, 0);
+            block_copy_task_end(task, 0);
             progress_set_remaining(s->progress,
                                    bdrv_get_dirty_count(s->copy_bitmap) +
                                    s->in_flight_bytes);
@@ -520,7 +522,7 @@ static int coroutine_fn block_copy_dirty_clusters(BlockCopyState *s,
         ret = block_copy_do_copy(s, offset, cur_bytes, ret & BDRV_BLOCK_ZERO,
                                  error_is_read);
         co_put_to_shres(s->mem, cur_bytes);
-        block_copy_task_end(s, task, ret);
+        block_copy_task_end(task, ret);
         if (ret < 0) {
             return ret;
         }
-- 
2.21.0

[PATCH v3 4/5] block/block-copy: refactor task creation

[Vladimir Sementsov-Ogievskiy]

Instead of just relying on the comment "Called only on full-dirty
region" in block_copy_task_create() let's move initial dirty area
search directly to block_copy_task_create(). Let's also use effective
bdrv_dirty_bitmap_next_dirty_area instead of looping through all
non-dirty clusters.

Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
---
 block/block-copy.c | 78 ++++++++++++++++++++++++++--------------------
 1 file changed, 44 insertions(+), 34 deletions(-)

diff --git a/block/block-copy.c b/block/block-copy.c
index 35ff9cc3ef..5cf032c4d8 100644
--- a/block/block-copy.c
+++ b/block/block-copy.c
@@ -32,6 +32,11 @@ typedef struct BlockCopyTask {
     CoQueue wait_queue; /* coroutines blocked on this task */
 } BlockCopyTask;
 
+static int64_t task_end(BlockCopyTask *task)
+{
+    return task->offset + task->bytes;
+}
+
 typedef struct BlockCopyState {
     /*
      * BdrvChild objects are not owned or managed by block-copy. They are
@@ -106,17 +111,27 @@ static bool coroutine_fn block_copy_wait_one(BlockCopyState *s, int64_t offset,
     return true;
 }
 
-/* Called only on full-dirty region */
+/*
+ * Search for the first dirty area in offset/bytes range and create task at
+ * the beginning of it.
+ */
 static BlockCopyTask *block_copy_task_create(BlockCopyState *s,
                                              int64_t offset, int64_t bytes)
 {
-    BlockCopyTask *task = g_new(BlockCopyTask, 1);
+    if (!bdrv_dirty_bitmap_next_dirty_area(s->copy_bitmap,
+                                           offset, offset + bytes,
+                                           s->copy_size, &offset, &bytes))
+    {
+        return NULL;
+    }
 
+    /* region is dirty, so no existent tasks possible in it */
     assert(!find_conflicting_task(s, offset, bytes));
 
     bdrv_reset_dirty_bitmap(s->copy_bitmap, offset, bytes);
     s->in_flight_bytes += bytes;
 
+    BlockCopyTask *task = g_new(BlockCopyTask, 1);
     *task = (BlockCopyTask) {
         .s = s,
         .offset = offset,
@@ -466,6 +481,7 @@ static int coroutine_fn block_copy_dirty_clusters(BlockCopyState *s,
 {
     int ret = 0;
     bool found_dirty = false;
+    int64_t end = offset + bytes;
 
     /*
      * block_copy() user is responsible for keeping source and target in same
@@ -479,58 +495,52 @@ static int coroutine_fn block_copy_dirty_clusters(BlockCopyState *s,
 
     while (bytes) {
         g_autofree BlockCopyTask *task = NULL;
-        int64_t next_zero, cur_bytes, status_bytes;
+        int64_t status_bytes;
 
-        if (!bdrv_dirty_bitmap_get(s->copy_bitmap, offset)) {
-            trace_block_copy_skip(s, offset);
-            offset += s->cluster_size;
-            bytes -= s->cluster_size;
-            continue; /* already copied */
+        task = block_copy_task_create(s, offset, bytes);
+        if (!task) {
+            /* No more dirty bits in the bitmap */
+            trace_block_copy_skip_range(s, offset, bytes);
+            break;
+        }
+        if (task->offset > offset) {
+            trace_block_copy_skip_range(s, offset, task->offset - offset);
         }
 
         found_dirty = true;
 
-        cur_bytes = MIN(bytes, s->copy_size);
-
-        next_zero = bdrv_dirty_bitmap_next_zero(s->copy_bitmap, offset,
-                                                cur_bytes);
-        if (next_zero >= 0) {
-            assert(next_zero > offset); /* offset is dirty */
-            assert(next_zero < offset + cur_bytes); /* no need to do MIN() */
-            cur_bytes = next_zero - offset;
-        }
-        task = block_copy_task_create(s, offset, cur_bytes);
-
-        ret = block_copy_block_status(s, offset, cur_bytes, &status_bytes);
+        ret = block_copy_block_status(s, task->offset, task->bytes,
+                                      &status_bytes);
         assert(ret >= 0); /* never fail */
-        cur_bytes = MIN(cur_bytes, status_bytes);
-        block_copy_task_shrink(task, cur_bytes);
+        if (status_bytes < task->bytes) {
+            block_copy_task_shrink(task, status_bytes);
+        }
         if (s->skip_unallocated && !(ret & BDRV_BLOCK_ALLOCATED)) {
             block_copy_task_end(task, 0);
             progress_set_remaining(s->progress,
                                    bdrv_get_dirty_count(s->copy_bitmap) +
                                    s->in_flight_bytes);
-            trace_block_copy_skip_range(s, offset, status_bytes);
-            offset += status_bytes;
-            bytes -= status_bytes;
+            trace_block_copy_skip_range(s, task->offset, task->bytes);
+            offset = task_end(task);
+            bytes = end - offset;
             continue;
         }
 
-        trace_block_copy_process(s, offset);
+        trace_block_copy_process(s, task->offset);
 
-        co_get_from_shres(s->mem, cur_bytes);
-        ret = block_copy_do_copy(s, offset, cur_bytes, ret & BDRV_BLOCK_ZERO,
-                                 error_is_read);
-        co_put_to_shres(s->mem, cur_bytes);
+        co_get_from_shres(s->mem, task->bytes);
+        ret = block_copy_do_copy(s, task->offset, task->bytes,
+                                 ret & BDRV_BLOCK_ZERO, error_is_read);
+        co_put_to_shres(s->mem, task->bytes);
         block_copy_task_end(task, ret);
         if (ret < 0) {
             return ret;
         }
 
-        progress_work_done(s->progress, cur_bytes);
-        s->progress_bytes_callback(cur_bytes, s->progress_opaque);
-        offset += cur_bytes;
-        bytes -= cur_bytes;
+        progress_work_done(s->progress, task->bytes);
+        s->progress_bytes_callback(task->bytes, s->progress_opaque);
+        offset = task_end(task);
+        bytes = end - offset;
     }
 
     return found_dirty;
-- 
2.21.0

[PATCH v3 5/5] block/block-copy: use aio-task-pool API

[Vladimir Sementsov-Ogievskiy]

Run block_copy iterations in parallel in aio tasks.

Changes:
  - BlockCopyTask becomes aio task structure. Add zeroes field to pass
    it to block_copy_do_copy
  - add call state - it's a state of one call of block_copy(), shared
    between parallel tasks. For now used only to keep information about
    first error: is it read or not.
  - convert block_copy_dirty_clusters to aio-task loop.

Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
---
 block/block-copy.c | 104 +++++++++++++++++++++++++++++++++++++++------
 1 file changed, 91 insertions(+), 13 deletions(-)

diff --git a/block/block-copy.c b/block/block-copy.c
index 5cf032c4d8..f5ef91f292 100644
--- a/block/block-copy.c
+++ b/block/block-copy.c
@@ -19,15 +19,29 @@
 #include "block/block-copy.h"
 #include "sysemu/block-backend.h"
 #include "qemu/units.h"
+#include "qemu/coroutine.h"
+#include "block/aio_task.h"
 
 #define BLOCK_COPY_MAX_COPY_RANGE (16 * MiB)
 #define BLOCK_COPY_MAX_BUFFER (1 * MiB)
 #define BLOCK_COPY_MAX_MEM (128 * MiB)
+#define BLOCK_COPY_MAX_WORKERS 64
+
+static coroutine_fn int block_copy_task_entry(AioTask *task);
+
+typedef struct BlockCopyCallState {
+    bool failed;
+    bool error_is_read;
+} BlockCopyCallState;
 
 typedef struct BlockCopyTask {
+    AioTask task;
+
     BlockCopyState *s;
+    BlockCopyCallState *call_state;
     int64_t offset;
     int64_t bytes;
+    bool zeroes;
     QLIST_ENTRY(BlockCopyTask) list;
     CoQueue wait_queue; /* coroutines blocked on this task */
 } BlockCopyTask;
@@ -116,6 +130,7 @@ static bool coroutine_fn block_copy_wait_one(BlockCopyState *s, int64_t offset,
  * the beginning of it.
  */
 static BlockCopyTask *block_copy_task_create(BlockCopyState *s,
+                                             BlockCopyCallState *call_state,
                                              int64_t offset, int64_t bytes)
 {
     if (!bdrv_dirty_bitmap_next_dirty_area(s->copy_bitmap,
@@ -133,7 +148,9 @@ static BlockCopyTask *block_copy_task_create(BlockCopyState *s,
 
     BlockCopyTask *task = g_new(BlockCopyTask, 1);
     *task = (BlockCopyTask) {
+        .task.func = block_copy_task_entry,
         .s = s,
+        .call_state = call_state,
         .offset = offset,
         .bytes = bytes,
     };
@@ -261,6 +278,30 @@ void block_copy_set_progress_meter(BlockCopyState *s, ProgressMeter *pm)
     s->progress = pm;
 }
 
+/* Takes ownership on @task */
+static coroutine_fn int block_copy_task_run(AioTaskPool *pool,
+                                            BlockCopyTask *task)
+{
+    if (!pool) {
+        int ret = task->task.func(&task->task);
+
+        g_free(task);
+        return ret;
+    }
+
+    aio_task_pool_wait_slot(pool);
+    if (aio_task_pool_status(pool) < 0) {
+        co_put_to_shres(task->s->mem, task->bytes);
+        block_copy_task_end(task, -EAGAIN);
+        g_free(task);
+        return aio_task_pool_status(pool);
+    }
+
+    aio_task_pool_start_task(pool, &task->task);
+
+    return 0;
+}
+
 /*
  * block_copy_do_copy
  *
@@ -364,6 +405,27 @@ out:
     return ret;
 }
 
+static coroutine_fn int block_copy_task_entry(AioTask *task)
+{
+    BlockCopyTask *t = container_of(task, BlockCopyTask, task);
+    bool error_is_read;
+    int ret;
+
+    ret = block_copy_do_copy(t->s, t->offset, t->bytes, t->zeroes,
+                             &error_is_read);
+    if (ret < 0 && !t->call_state->failed) {
+        t->call_state->failed = true;
+        t->call_state->error_is_read = error_is_read;
+    } else {
+        progress_work_done(t->s->progress, t->bytes);
+        t->s->progress_bytes_callback(t->bytes, t->s->progress_opaque);
+    }
+    co_put_to_shres(t->s->mem, t->bytes);
+    block_copy_task_end(t, ret);
+
+    return ret;
+}
+
 static int block_copy_block_status(BlockCopyState *s, int64_t offset,
                                    int64_t bytes, int64_t *pnum)
 {
@@ -482,6 +544,8 @@ static int coroutine_fn block_copy_dirty_clusters(BlockCopyState *s,
     int ret = 0;
     bool found_dirty = false;
     int64_t end = offset + bytes;
+    AioTaskPool *aio = NULL;
+    BlockCopyCallState call_state = {false, false};
 
     /*
      * block_copy() user is responsible for keeping source and target in same
@@ -493,11 +557,11 @@ static int coroutine_fn block_copy_dirty_clusters(BlockCopyState *s,
     assert(QEMU_IS_ALIGNED(offset, s->cluster_size));
     assert(QEMU_IS_ALIGNED(bytes, s->cluster_size));
 
-    while (bytes) {
-        g_autofree BlockCopyTask *task = NULL;
+    while (bytes && aio_task_pool_status(aio) == 0) {
+        BlockCopyTask *task;
         int64_t status_bytes;
 
-        task = block_copy_task_create(s, offset, bytes);
+        task = block_copy_task_create(s, &call_state, offset, bytes);
         if (!task) {
             /* No more dirty bits in the bitmap */
             trace_block_copy_skip_range(s, offset, bytes);
@@ -517,6 +581,7 @@ static int coroutine_fn block_copy_dirty_clusters(BlockCopyState *s,
         }
         if (s->skip_unallocated && !(ret & BDRV_BLOCK_ALLOCATED)) {
             block_copy_task_end(task, 0);
+            g_free(task);
             progress_set_remaining(s->progress,
                                    bdrv_get_dirty_count(s->copy_bitmap) +
                                    s->in_flight_bytes);
@@ -525,25 +590,38 @@ static int coroutine_fn block_copy_dirty_clusters(BlockCopyState *s,
             bytes = end - offset;
             continue;
         }
+        task->zeroes = ret & BDRV_BLOCK_ZERO;
 
         trace_block_copy_process(s, task->offset);
 
         co_get_from_shres(s->mem, task->bytes);
-        ret = block_copy_do_copy(s, task->offset, task->bytes,
-                                 ret & BDRV_BLOCK_ZERO, error_is_read);
-        co_put_to_shres(s->mem, task->bytes);
-        block_copy_task_end(task, ret);
-        if (ret < 0) {
-            return ret;
-        }
 
-        progress_work_done(s->progress, task->bytes);
-        s->progress_bytes_callback(task->bytes, s->progress_opaque);
         offset = task_end(task);
         bytes = end - offset;
+
+        if (!aio && bytes) {
+            aio = aio_task_pool_new(BLOCK_COPY_MAX_WORKERS);
+        }
+
+        ret = block_copy_task_run(aio, task);
+        if (ret < 0) {
+            goto out;
+        }
+    }
+
+out:
+    if (aio) {
+        aio_task_pool_wait_all(aio);
+        if (ret == 0) {
+            ret = aio_task_pool_status(aio);
+        }
+        g_free(aio);
+    }
+    if (error_is_read && ret < 0) {
+        *error_is_read = call_state.error_is_read;
     }
 
-    return found_dirty;
+    return ret < 0 ? ret : found_dirty;
 }
 
 /*
-- 
2.21.0

Re: [PATCH v3 1/5] block/block-copy: rename in-flight requests to tasks

[Max Reitz]

[-- Attachment #1.1: Type: text/plain, Size: 463 bytes --]

On 29.04.20 08:10, Vladimir Sementsov-Ogievskiy wrote:
> We are going to use aio-task-pool API and extend in-flight request
> structure to be a successor of AioTask, so rename things appropriately.
> 
> Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
> ---
>  block/block-copy.c | 98 +++++++++++++++++++++++-----------------------
>  1 file changed, 48 insertions(+), 50 deletions(-)

Reviewed-by: Max Reitz <mreitz@redhat.com>


[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 488 bytes --]

Re: [PATCH v3 4/5] block/block-copy: refactor task creation

[Max Reitz]

[-- Attachment #1.1: Type: text/plain, Size: 2042 bytes --]

On 29.04.20 08:10, Vladimir Sementsov-Ogievskiy wrote:
> Instead of just relying on the comment "Called only on full-dirty
> region" in block_copy_task_create() let's move initial dirty area
> search directly to block_copy_task_create(). Let's also use effective
> bdrv_dirty_bitmap_next_dirty_area instead of looping through all
> non-dirty clusters.
> 
> Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
> ---
>  block/block-copy.c | 78 ++++++++++++++++++++++++++--------------------
>  1 file changed, 44 insertions(+), 34 deletions(-)
> 
> diff --git a/block/block-copy.c b/block/block-copy.c
> index 35ff9cc3ef..5cf032c4d8 100644
> --- a/block/block-copy.c
> +++ b/block/block-copy.c

[...]

> @@ -106,17 +111,27 @@ static bool coroutine_fn block_copy_wait_one(BlockCopyState *s, int64_t offset,
>      return true;
>  }
>  
> -/* Called only on full-dirty region */
> +/*
> + * Search for the first dirty area in offset/bytes range and create task at
> + * the beginning of it.

Oh, that’s even better.

> + */
>  static BlockCopyTask *block_copy_task_create(BlockCopyState *s,
>                                               int64_t offset, int64_t bytes)
>  {
> -    BlockCopyTask *task = g_new(BlockCopyTask, 1);
> +    if (!bdrv_dirty_bitmap_next_dirty_area(s->copy_bitmap,
> +                                           offset, offset + bytes,
> +                                           s->copy_size, &offset, &bytes))
> +    {
> +        return NULL;
> +    }
>  
> +    /* region is dirty, so no existent tasks possible in it */
>      assert(!find_conflicting_task(s, offset, bytes));
>  
>      bdrv_reset_dirty_bitmap(s->copy_bitmap, offset, bytes);
>      s->in_flight_bytes += bytes;
>  
> +    BlockCopyTask *task = g_new(BlockCopyTask, 1);

This should be declared at the top of the function.

With that fixed:

Reviewed-by: Max Reitz <mreitz@redhat.com>

>      *task = (BlockCopyTask) {
>          .s = s,
>          .offset = offset,


[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 488 bytes --]

Re: [PATCH v3 4/5] block/block-copy: refactor task creation

[Vladimir Sementsov-Ogievskiy]

29.04.2020 14:38, Max Reitz wrote:
> On 29.04.20 08:10, Vladimir Sementsov-Ogievskiy wrote:
>> Instead of just relying on the comment "Called only on full-dirty
>> region" in block_copy_task_create() let's move initial dirty area
>> search directly to block_copy_task_create(). Let's also use effective
>> bdrv_dirty_bitmap_next_dirty_area instead of looping through all
>> non-dirty clusters.
>>
>> Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
>> ---
>>   block/block-copy.c | 78 ++++++++++++++++++++++++++--------------------
>>   1 file changed, 44 insertions(+), 34 deletions(-)
>>
>> diff --git a/block/block-copy.c b/block/block-copy.c
>> index 35ff9cc3ef..5cf032c4d8 100644
>> --- a/block/block-copy.c
>> +++ b/block/block-copy.c
> 
> [...]
> 
>> @@ -106,17 +111,27 @@ static bool coroutine_fn block_copy_wait_one(BlockCopyState *s, int64_t offset,
>>       return true;
>>   }
>>   
>> -/* Called only on full-dirty region */
>> +/*
>> + * Search for the first dirty area in offset/bytes range and create task at
>> + * the beginning of it.
> 
> Oh, that’s even better.
> 
>> + */
>>   static BlockCopyTask *block_copy_task_create(BlockCopyState *s,
>>                                                int64_t offset, int64_t bytes)
>>   {
>> -    BlockCopyTask *task = g_new(BlockCopyTask, 1);
>> +    if (!bdrv_dirty_bitmap_next_dirty_area(s->copy_bitmap,
>> +                                           offset, offset + bytes,
>> +                                           s->copy_size, &offset, &bytes))
>> +    {
>> +        return NULL;
>> +    }
>>   
>> +    /* region is dirty, so no existent tasks possible in it */
>>       assert(!find_conflicting_task(s, offset, bytes));
>>   
>>       bdrv_reset_dirty_bitmap(s->copy_bitmap, offset, bytes);
>>       s->in_flight_bytes += bytes;
>>   
>> +    BlockCopyTask *task = g_new(BlockCopyTask, 1);
> 
> This should be declared at the top of the function.
> 

I just thought, why not to try another style? Are you against? Requirement to declare variables at start of block is obsolete, isn't it?

> 
> Reviewed-by: Max Reitz <mreitz@redhat.com>
> 
>>       *task = (BlockCopyTask) {
>>           .s = s,
>>           .offset = offset,
> 


-- 
Best regards,
Vladimir

Re: [PATCH v3 5/5] block/block-copy: use aio-task-pool API

[Max Reitz]

[-- Attachment #1.1: Type: text/plain, Size: 2448 bytes --]

On 29.04.20 08:10, Vladimir Sementsov-Ogievskiy wrote:
> Run block_copy iterations in parallel in aio tasks.
> 
> Changes:
>   - BlockCopyTask becomes aio task structure. Add zeroes field to pass
>     it to block_copy_do_copy
>   - add call state - it's a state of one call of block_copy(), shared
>     between parallel tasks. For now used only to keep information about
>     first error: is it read or not.
>   - convert block_copy_dirty_clusters to aio-task loop.
> 
> Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
> ---
>  block/block-copy.c | 104 +++++++++++++++++++++++++++++++++++++++------
>  1 file changed, 91 insertions(+), 13 deletions(-)
> 
> diff --git a/block/block-copy.c b/block/block-copy.c
> index 5cf032c4d8..f5ef91f292 100644
> --- a/block/block-copy.c
> +++ b/block/block-copy.c

[...]

> @@ -261,6 +278,30 @@ void block_copy_set_progress_meter(BlockCopyState *s, ProgressMeter *pm)
>      s->progress = pm;
>  }
>  
> +/* Takes ownership on @task */

Still *of

> +static coroutine_fn int block_copy_task_run(AioTaskPool *pool,
> +                                            BlockCopyTask *task)
> +{
> +    if (!pool) {
> +        int ret = task->task.func(&task->task);
> +
> +        g_free(task);
> +        return ret;
> +    }
> +
> +    aio_task_pool_wait_slot(pool);
> +    if (aio_task_pool_status(pool) < 0) {
> +        co_put_to_shres(task->s->mem, task->bytes);
> +        block_copy_task_end(task, -EAGAIN);

It looks like you may have missed my nit picks on v2 regarding this
patch, so I’m going to ask again whether -ECANCELED might be better here
(even though it still doesn’t really matter).

> +        g_free(task);
> +        return aio_task_pool_status(pool);

And whether it may be better to return a constant like -ECANCELED here,
because how a previous task failed shouldn’t really concern this task
(or its error code).

> +    }
> +
> +    aio_task_pool_start_task(pool, &task->task);
> +
> +    return 0;
> +}
> +
>  /*
>   * block_copy_do_copy
>   *

[...]

> @@ -525,25 +590,38 @@ static int coroutine_fn block_copy_dirty_clusters(BlockCopyState *s,

[...]

> +out:
> +    if (aio) {
> +        aio_task_pool_wait_all(aio);
> +        if (ret == 0) {
> +            ret = aio_task_pool_status(aio);
> +        }
> +        g_free(aio);

I’d still prefer aio_task_pool_free().

Max


[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 488 bytes --]

Re: [PATCH v3 4/5] block/block-copy: refactor task creation

[Max Reitz]

[-- Attachment #1.1: Type: text/plain, Size: 2838 bytes --]

On 29.04.20 13:54, Vladimir Sementsov-Ogievskiy wrote:
> 29.04.2020 14:38, Max Reitz wrote:
>> On 29.04.20 08:10, Vladimir Sementsov-Ogievskiy wrote:
>>> Instead of just relying on the comment "Called only on full-dirty
>>> region" in block_copy_task_create() let's move initial dirty area
>>> search directly to block_copy_task_create(). Let's also use effective
>>> bdrv_dirty_bitmap_next_dirty_area instead of looping through all
>>> non-dirty clusters.
>>>
>>> Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
>>> ---
>>>   block/block-copy.c | 78 ++++++++++++++++++++++++++--------------------
>>>   1 file changed, 44 insertions(+), 34 deletions(-)
>>>
>>> diff --git a/block/block-copy.c b/block/block-copy.c
>>> index 35ff9cc3ef..5cf032c4d8 100644
>>> --- a/block/block-copy.c
>>> +++ b/block/block-copy.c
>>
>> [...]
>>
>>> @@ -106,17 +111,27 @@ static bool coroutine_fn
>>> block_copy_wait_one(BlockCopyState *s, int64_t offset,
>>>       return true;
>>>   }
>>>   -/* Called only on full-dirty region */
>>> +/*
>>> + * Search for the first dirty area in offset/bytes range and create
>>> task at
>>> + * the beginning of it.
>>
>> Oh, that’s even better.
>>
>>> + */
>>>   static BlockCopyTask *block_copy_task_create(BlockCopyState *s,
>>>                                                int64_t offset,
>>> int64_t bytes)
>>>   {
>>> -    BlockCopyTask *task = g_new(BlockCopyTask, 1);
>>> +    if (!bdrv_dirty_bitmap_next_dirty_area(s->copy_bitmap,
>>> +                                           offset, offset + bytes,
>>> +                                           s->copy_size, &offset,
>>> &bytes))
>>> +    {
>>> +        return NULL;
>>> +    }
>>>   +    /* region is dirty, so no existent tasks possible in it */
>>>       assert(!find_conflicting_task(s, offset, bytes));
>>>         bdrv_reset_dirty_bitmap(s->copy_bitmap, offset, bytes);
>>>       s->in_flight_bytes += bytes;
>>>   +    BlockCopyTask *task = g_new(BlockCopyTask, 1);
>>
>> This should be declared at the top of the function.
>>
> 
> I just thought, why not to try another style? Are you against?
> Requirement to declare variables at start of block is obsolete, isn't it?

Oh, it absolutely is and personally I’m absolutely not against it, but
CODING_STYLE says:

> Mixed declarations (interleaving statements and declarations within
> blocks) are generally not allowed; declarations should be at the beginning                                                                                                                                                                   
> of blocks.

Max

>> Reviewed-by: Max Reitz <mreitz@redhat.com>
>>
>>>       *task = (BlockCopyTask) {
>>>           .s = s,
>>>           .offset = offset,
>>
> 
> 



[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 488 bytes --]

Re: [PATCH v3 4/5] block/block-copy: refactor task creation

[Vladimir Sementsov-Ogievskiy]

29.04.2020 14:56, Max Reitz wrote:
> On 29.04.20 13:54, Vladimir Sementsov-Ogievskiy wrote:
>> 29.04.2020 14:38, Max Reitz wrote:
>>> On 29.04.20 08:10, Vladimir Sementsov-Ogievskiy wrote:
>>>> Instead of just relying on the comment "Called only on full-dirty
>>>> region" in block_copy_task_create() let's move initial dirty area
>>>> search directly to block_copy_task_create(). Let's also use effective
>>>> bdrv_dirty_bitmap_next_dirty_area instead of looping through all
>>>> non-dirty clusters.
>>>>
>>>> Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
>>>> ---
>>>>    block/block-copy.c | 78 ++++++++++++++++++++++++++--------------------
>>>>    1 file changed, 44 insertions(+), 34 deletions(-)
>>>>
>>>> diff --git a/block/block-copy.c b/block/block-copy.c
>>>> index 35ff9cc3ef..5cf032c4d8 100644
>>>> --- a/block/block-copy.c
>>>> +++ b/block/block-copy.c
>>>
>>> [...]
>>>
>>>> @@ -106,17 +111,27 @@ static bool coroutine_fn
>>>> block_copy_wait_one(BlockCopyState *s, int64_t offset,
>>>>        return true;
>>>>    }
>>>>    -/* Called only on full-dirty region */
>>>> +/*
>>>> + * Search for the first dirty area in offset/bytes range and create
>>>> task at
>>>> + * the beginning of it.
>>>
>>> Oh, that’s even better.
>>>
>>>> + */
>>>>    static BlockCopyTask *block_copy_task_create(BlockCopyState *s,
>>>>                                                 int64_t offset,
>>>> int64_t bytes)
>>>>    {
>>>> -    BlockCopyTask *task = g_new(BlockCopyTask, 1);
>>>> +    if (!bdrv_dirty_bitmap_next_dirty_area(s->copy_bitmap,
>>>> +                                           offset, offset + bytes,
>>>> +                                           s->copy_size, &offset,
>>>> &bytes))
>>>> +    {
>>>> +        return NULL;
>>>> +    }
>>>>    +    /* region is dirty, so no existent tasks possible in it */
>>>>        assert(!find_conflicting_task(s, offset, bytes));
>>>>          bdrv_reset_dirty_bitmap(s->copy_bitmap, offset, bytes);
>>>>        s->in_flight_bytes += bytes;
>>>>    +    BlockCopyTask *task = g_new(BlockCopyTask, 1);
>>>
>>> This should be declared at the top of the function.
>>>
>>
>> I just thought, why not to try another style? Are you against?
>> Requirement to declare variables at start of block is obsolete, isn't it?
> 
> Oh, it absolutely is and personally I’m absolutely not against it, but
> CODING_STYLE says:
> 
>> Mixed declarations (interleaving statements and declarations within
>> blocks) are generally not allowed; declarations should be at the beginning
>> of blocks.
> 

Oh, missed (or forget). Let's fix it? :) Not in these series, of course. OK, I'll fix the patch.



-- 
Best regards,
Vladimir

Re: [PATCH v3 5/5] block/block-copy: use aio-task-pool API

[Vladimir Sementsov-Ogievskiy]

29.04.2020 14:55, Max Reitz wrote:
> On 29.04.20 08:10, Vladimir Sementsov-Ogievskiy wrote:
>> Run block_copy iterations in parallel in aio tasks.
>>
>> Changes:
>>    - BlockCopyTask becomes aio task structure. Add zeroes field to pass
>>      it to block_copy_do_copy
>>    - add call state - it's a state of one call of block_copy(), shared
>>      between parallel tasks. For now used only to keep information about
>>      first error: is it read or not.
>>    - convert block_copy_dirty_clusters to aio-task loop.
>>
>> Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
>> ---
>>   block/block-copy.c | 104 +++++++++++++++++++++++++++++++++++++++------
>>   1 file changed, 91 insertions(+), 13 deletions(-)
>>
>> diff --git a/block/block-copy.c b/block/block-copy.c
>> index 5cf032c4d8..f5ef91f292 100644
>> --- a/block/block-copy.c
>> +++ b/block/block-copy.c
> 
> [...]
> 
>> @@ -261,6 +278,30 @@ void block_copy_set_progress_meter(BlockCopyState *s, ProgressMeter *pm)
>>       s->progress = pm;
>>   }
>>   
>> +/* Takes ownership on @task */
> 
> Still *of

Ohhh, very sorry for it, I really forget to update the patch :(

> 
>> +static coroutine_fn int block_copy_task_run(AioTaskPool *pool,
>> +                                            BlockCopyTask *task)
>> +{
>> +    if (!pool) {
>> +        int ret = task->task.func(&task->task);
>> +
>> +        g_free(task);
>> +        return ret;
>> +    }
>> +
>> +    aio_task_pool_wait_slot(pool);
>> +    if (aio_task_pool_status(pool) < 0) {
>> +        co_put_to_shres(task->s->mem, task->bytes);
>> +        block_copy_task_end(task, -EAGAIN);
> 
> It looks like you may have missed my nit picks on v2 regarding this
> patch, so I’m going to ask again whether -ECANCELED might be better here
> (even though it still doesn’t really matter).

Hmm yes, sounds better in the context. And I don't see any specific usage of it, and don't remember why I've chosen EAGAIN :) Let's use ECANCELED.

> 
>> +        g_free(task);
>> +        return aio_task_pool_status(pool);
> 
> And whether it may be better to return a constant like -ECANCELED here,
> because how a previous task failed shouldn’t really concern this task
> (or its error code).

Looks correct, will change.

> 
>> +    }
>> +
>> +    aio_task_pool_start_task(pool, &task->task);
>> +
>> +    return 0;
>> +}
>> +
>>   /*
>>    * block_copy_do_copy
>>    *
> 
> [...]
> 
>> @@ -525,25 +590,38 @@ static int coroutine_fn block_copy_dirty_clusters(BlockCopyState *s,
> 
> [...]
> 
>> +out:
>> +    if (aio) {
>> +        aio_task_pool_wait_all(aio);
>> +        if (ret == 0) {
>> +            ret = aio_task_pool_status(aio);
>> +        }
>> +        g_free(aio);
> 
> I’d still prefer aio_task_pool_free().
> 

a thousand apologies :(


-- 
Best regards,
Vladimir