[Qemu-devel] [PATCH 08/19] block: Remove bdrv_drain_recurse()

Kevin Wolf posted 19 patches 7 years, 10 months ago
There is a newer version of this series
[Qemu-devel] [PATCH 08/19] block: Remove bdrv_drain_recurse()
Posted by Kevin Wolf 7 years, 10 months ago
For bdrv_drain(), recursively waiting for child node requests is
pointless because we didn't quiesce their parents, so new requests could
come in anyway. Letting the function work only on a single node makes it
more consistent.

For subtree drains and drain_all, we already have the recursion in
bdrv_do_drained_begin(), so the extra recursion doesn't add anything
either.

Remove the useless code.

Signed-off-by: Kevin Wolf <kwolf@redhat.com>
---
 block/io.c | 36 +++---------------------------------
 1 file changed, 3 insertions(+), 33 deletions(-)

diff --git a/block/io.c b/block/io.c
index 0a778eeff4..f24f39c278 100644
--- a/block/io.c
+++ b/block/io.c
@@ -211,38 +211,6 @@ bool bdrv_drain_poll(BlockDriverState *bs, bool top_level)
     return atomic_read(&bs->in_flight);
 }
 
-static bool bdrv_drain_recurse(BlockDriverState *bs)
-{
-    BdrvChild *child, *tmp;
-    bool waited;
-
-    /* Wait for drained requests to finish */
-    waited = BDRV_POLL_WHILE(bs, bdrv_drain_poll(bs, true));
-
-    QLIST_FOREACH_SAFE(child, &bs->children, next, tmp) {
-        BlockDriverState *bs = child->bs;
-        bool in_main_loop =
-            qemu_get_current_aio_context() == qemu_get_aio_context();
-        assert(bs->refcnt > 0);
-        if (in_main_loop) {
-            /* In case the recursive bdrv_drain_recurse processes a
-             * block_job_defer_to_main_loop BH and modifies the graph,
-             * let's hold a reference to bs until we are done.
-             *
-             * IOThread doesn't have such a BH, and it is not safe to call
-             * bdrv_unref without BQL, so skip doing it there.
-             */
-            bdrv_ref(bs);
-        }
-        waited |= bdrv_drain_recurse(bs);
-        if (in_main_loop) {
-            bdrv_unref(bs);
-        }
-    }
-
-    return waited;
-}
-
 static void bdrv_do_drained_begin(BlockDriverState *bs, bool recursive,
                                   BdrvChild *parent);
 static void bdrv_do_drained_end(BlockDriverState *bs, bool recursive,
@@ -310,7 +278,9 @@ void bdrv_do_drained_begin(BlockDriverState *bs, bool recursive,
 
     bdrv_parent_drained_begin(bs, parent);
     bdrv_drain_invoke(bs, true);
-    bdrv_drain_recurse(bs);
+
+    /* Wait for drained requests to finish */
+    BDRV_POLL_WHILE(bs, bdrv_drain_poll(bs, true));
 
     if (recursive) {
         bs->recursive_quiesce_counter++;
-- 
2.13.6


Re: [Qemu-devel] [PATCH 08/19] block: Remove bdrv_drain_recurse()
Posted by Paolo Bonzini 7 years, 10 months ago
On 11/04/2018 18:39, Kevin Wolf wrote:
> For bdrv_drain(), recursively waiting for child node requests is
> pointless because we didn't quiesce their parents, so new requests could
> come in anyway. Letting the function work only on a single node makes it
> more consistent.
> 
> For subtree drains and drain_all, we already have the recursion in
> bdrv_do_drained_begin(), so the extra recursion doesn't add anything
> either.
> 
> Remove the useless code.

This solves one of my questions, but leaves me the doubt about quadratic
complexity.

Paolo

> Signed-off-by: Kevin Wolf <kwolf@redhat.com>
> ---
>  block/io.c | 36 +++---------------------------------
>  1 file changed, 3 insertions(+), 33 deletions(-)
> 
> diff --git a/block/io.c b/block/io.c
> index 0a778eeff4..f24f39c278 100644
> --- a/block/io.c
> +++ b/block/io.c
> @@ -211,38 +211,6 @@ bool bdrv_drain_poll(BlockDriverState *bs, bool top_level)
>      return atomic_read(&bs->in_flight);
>  }
>  
> -static bool bdrv_drain_recurse(BlockDriverState *bs)
> -{
> -    BdrvChild *child, *tmp;
> -    bool waited;
> -
> -    /* Wait for drained requests to finish */
> -    waited = BDRV_POLL_WHILE(bs, bdrv_drain_poll(bs, true));
> -
> -    QLIST_FOREACH_SAFE(child, &bs->children, next, tmp) {
> -        BlockDriverState *bs = child->bs;
> -        bool in_main_loop =
> -            qemu_get_current_aio_context() == qemu_get_aio_context();
> -        assert(bs->refcnt > 0);
> -        if (in_main_loop) {
> -            /* In case the recursive bdrv_drain_recurse processes a
> -             * block_job_defer_to_main_loop BH and modifies the graph,
> -             * let's hold a reference to bs until we are done.
> -             *
> -             * IOThread doesn't have such a BH, and it is not safe to call
> -             * bdrv_unref without BQL, so skip doing it there.
> -             */
> -            bdrv_ref(bs);
> -        }
> -        waited |= bdrv_drain_recurse(bs);
> -        if (in_main_loop) {
> -            bdrv_unref(bs);
> -        }
> -    }
> -
> -    return waited;
> -}
> -
>  static void bdrv_do_drained_begin(BlockDriverState *bs, bool recursive,
>                                    BdrvChild *parent);
>  static void bdrv_do_drained_end(BlockDriverState *bs, bool recursive,
> @@ -310,7 +278,9 @@ void bdrv_do_drained_begin(BlockDriverState *bs, bool recursive,
>  
>      bdrv_parent_drained_begin(bs, parent);
>      bdrv_drain_invoke(bs, true);
> -    bdrv_drain_recurse(bs);
> +
> +    /* Wait for drained requests to finish */
> +    BDRV_POLL_WHILE(bs, bdrv_drain_poll(bs, true));
>  
>      if (recursive) {
>          bs->recursive_quiesce_counter++;
> 


Re: [Qemu-devel] [Qemu-block] [PATCH 08/19] block: Remove bdrv_drain_recurse()
Posted by Stefan Hajnoczi 7 years, 9 months ago
On Wed, Apr 11, 2018 at 06:39:29PM +0200, Kevin Wolf wrote:
> For bdrv_drain(), recursively waiting for child node requests is
> pointless because we didn't quiesce their parents, so new requests could
> come in anyway. Letting the function work only on a single node makes it
> more consistent.
> 
> For subtree drains and drain_all, we already have the recursion in
> bdrv_do_drained_begin(), so the extra recursion doesn't add anything
> either.
> 
> Remove the useless code.
> 
> Signed-off-by: Kevin Wolf <kwolf@redhat.com>
> ---
>  block/io.c | 36 +++---------------------------------
>  1 file changed, 3 insertions(+), 33 deletions(-)

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