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

Block layer patches - Graph locking part 4 (node management) - qemu-img map: report compressed data blocks - block-backend: process I/O in the current AioContext # -----BEGIN PGP SIGNATURE----- # # iQJFBAABCAAvFiEE3D3rFZqa+V09dFb+fwmycsiPL9YFAmUEbPkRHGt3b2xmQHJl # ZGhhdC5jb20ACgkQfwmycsiPL9bIaQ//QAwYKudsfXZXaeWxqDm88tKHzN5oTt5P # Fkb4MyHDqN0OboGwLbpHRFPDt5BPw3MhLzIyhsmWBD5Twyv+7+nTf7OzSp/ciNlg # hISy/eZwEbqNZUktiYD8vCeuDJuCHSPlDKGe1as7mqSfVg1KwScYK2Yf8HWoTNz6 # Ku/Dm9opEX5obaOfeyLWXIzoWj5w2PcK4mn6hOTOSew6Ia44usV+QQe+yNAhMrJf # sozoezaec6OUFXzoOlgz+nuEFcvxihST9ZBc/34msOSBGxDJuIHxBoAzsSTLZdmY # LlsnzfjfFwd1qTb+m5h/HL2oEC9fje79XFWn8AAINkenXY8qZMfzXo+286DlQN/M # YN6iq5pLmzOfbPLM6xq0EG/FBCm50mBmOgfyuLBuOhc3CTj/EzsGI7MAdeX32+cO # 4pu2hGnzvqmWuJi+bVo40d1xpUnhEztzDK0d/NTvQMqNdQ4ieiLqCw1pgDvW4w6f # 32xElO7wiut3SnylZoghXk2BACuK8SHeCgV2+iI9TRl8t0uBVsrYX/0mD4hNAtE0 # igNPQTfeZogqrFYhxrHftmL4pMfzvHXEaU1HbvJMTHnMdj+8bjtWfPpinmeLviP6 # fy1x90VJ7bSOgSgZSEGJTwCyOV+0M8bY5c+4uyRIRkye5QZtftYt5kJCvb0+4aTB # ADq6DjKxX/A= # =EgzU # -----END PGP SIGNATURE----- # gpg: Signature made Fri 15 Sep 2023 10:40:57 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: (28 commits) block-coroutine-wrapper: use qemu_get_current_aio_context() block-backend: process zoned requests in the current AioContext block-backend: process I/O in the current AioContext test-bdrv-drain: avoid race with BH in IOThread drain test block: remove AIOCBInfo->get_aio_context() qemu-img: map: report compressed data blocks block: add BDRV_BLOCK_COMPRESSED flag for bdrv_block_status() block: Mark bdrv_add/del_child() and caller GRAPH_WRLOCK block: Mark bdrv_unref_child() GRAPH_WRLOCK block: Mark bdrv_root_unref_child() GRAPH_WRLOCK block: Take graph rdlock in bdrv_change_aio_context() block: Take graph rdlock in bdrv_drop_intermediate() block: Mark bdrv_parent_cb_change_media() GRAPH_RDLOCK block: Mark bdrv_child_perm() GRAPH_RDLOCK block: Mark bdrv_get_cumulative_perm() and callers GRAPH_RDLOCK block: Mark bdrv_parent_perms_conflict() and callers GRAPH_RDLOCK block: Mark bdrv_attach_child() GRAPH_WRLOCK block: Call transaction callbacks with lock held block: Mark bdrv_attach_child_common() GRAPH_WRLOCK block: Mark bdrv_replace_child_tran() GRAPH_WRLOCK ... Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
qemu · Sep 18, 2023 · 1738b8c · 1738b8c
2 parents 8450c26 + 5d96864
commit 1738b8c
Show file tree

Hide file tree

Showing 50 changed files with 1,376 additions and 1,036 deletions.
diff --git a/block.c b/block.c
diff --git a/block/blklogwrites.c b/block/blklogwrites.c
@@ -251,7 +251,9 @@ static int blk_log_writes_open(BlockDriverState *bs, QDict *options, int flags,
     ret = 0;
 fail_log:
     if (ret < 0) {
+        bdrv_graph_wrlock(NULL);
         bdrv_unref_child(bs, s->log_file);
+        bdrv_graph_wrunlock();
         s->log_file = NULL;
     }
 fail:
@@ -263,8 +265,10 @@ static void blk_log_writes_close(BlockDriverState *bs)
 {
     BDRVBlkLogWritesState *s = bs->opaque;
 
+    bdrv_graph_wrlock(NULL);
     bdrv_unref_child(bs, s->log_file);
     s->log_file = NULL;
+    bdrv_graph_wrunlock();
 }
 
 static int64_t coroutine_fn GRAPH_RDLOCK

diff --git a/block/blkverify.c b/block/blkverify.c
@@ -151,8 +151,10 @@ static void blkverify_close(BlockDriverState *bs)
 {
     BDRVBlkverifyState *s = bs->opaque;
 
+    bdrv_graph_wrlock(NULL);
     bdrv_unref_child(bs, s->test_file);
     s->test_file = NULL;
+    bdrv_graph_wrunlock();
 }
 
 static int64_t coroutine_fn GRAPH_RDLOCK

diff --git a/block/block-backend.c b/block/block-backend.c
@@ -33,8 +33,6 @@
 
 #define NOT_DONE 0x7fffffff /* used while emulated sync operation in progress */
 
-static AioContext *blk_aiocb_get_aio_context(BlockAIOCB *acb);
-
 typedef struct BlockBackendAioNotifier {
     void (*attached_aio_context)(AioContext *new_context, void *opaque);
     void (*detach_aio_context)(void *opaque);
@@ -103,7 +101,6 @@ typedef struct BlockBackendAIOCB {
 } BlockBackendAIOCB;
 
 static const AIOCBInfo block_backend_aiocb_info = {
-    .get_aio_context = blk_aiocb_get_aio_context,
     .aiocb_size = sizeof(BlockBackendAIOCB),
 };
 
@@ -121,6 +118,10 @@ static QTAILQ_HEAD(, BlockBackend) block_backends =
 static QTAILQ_HEAD(, BlockBackend) monitor_block_backends =
     QTAILQ_HEAD_INITIALIZER(monitor_block_backends);
 
+static int coroutine_mixed_fn GRAPH_RDLOCK
+blk_set_perm_locked(BlockBackend *blk, uint64_t perm, uint64_t shared_perm,
+                    Error **errp);
+
 static void blk_root_inherit_options(BdrvChildRole role, bool parent_is_format,
                                      int *child_flags, QDict *child_options,
                                      int parent_flags, QDict *parent_options)
@@ -186,7 +187,7 @@ static void blk_vm_state_changed(void *opaque, bool running, RunState state)
  *
  * If an error is returned, the VM cannot be allowed to be resumed.
  */
-static void blk_root_activate(BdrvChild *child, Error **errp)
+static void GRAPH_RDLOCK blk_root_activate(BdrvChild *child, Error **errp)
 {
     BlockBackend *blk = child->opaque;
     Error *local_err = NULL;
@@ -207,7 +208,7 @@ static void blk_root_activate(BdrvChild *child, Error **errp)
      */
     saved_shared_perm = blk->shared_perm;
 
-    blk_set_perm(blk, blk->perm, BLK_PERM_ALL, &local_err);
+    blk_set_perm_locked(blk, blk->perm, BLK_PERM_ALL, &local_err);
     if (local_err) {
         error_propagate(errp, local_err);
         blk->disable_perm = true;
@@ -226,7 +227,7 @@ static void blk_root_activate(BdrvChild *child, Error **errp)
         return;
     }
 
-    blk_set_perm(blk, blk->perm, blk->shared_perm, &local_err);
+    blk_set_perm_locked(blk, blk->perm, blk->shared_perm, &local_err);
     if (local_err) {
         error_propagate(errp, local_err);
         blk->disable_perm = true;
@@ -259,7 +260,7 @@ static bool blk_can_inactivate(BlockBackend *blk)
     return blk->force_allow_inactivate;
 }
 
-static int blk_root_inactivate(BdrvChild *child)
+static int GRAPH_RDLOCK blk_root_inactivate(BdrvChild *child)
 {
     BlockBackend *blk = child->opaque;
 
@@ -911,7 +912,10 @@ void blk_remove_bs(BlockBackend *blk)
     blk_drain(blk);
     root = blk->root;
     blk->root = NULL;
+
+    bdrv_graph_wrlock(NULL);
     bdrv_root_unref_child(root);
+    bdrv_graph_wrunlock();
 }
 
 /*
@@ -953,8 +957,9 @@ int blk_replace_bs(BlockBackend *blk, BlockDriverState *new_bs, Error **errp)
 /*
  * Sets the permission bitmasks that the user of the BlockBackend needs.
  */
-int blk_set_perm(BlockBackend *blk, uint64_t perm, uint64_t shared_perm,
-                 Error **errp)
+static int coroutine_mixed_fn GRAPH_RDLOCK
+blk_set_perm_locked(BlockBackend *blk, uint64_t perm, uint64_t shared_perm,
+                    Error **errp)
 {
     int ret;
     GLOBAL_STATE_CODE();
@@ -972,6 +977,15 @@ int blk_set_perm(BlockBackend *blk, uint64_t perm, uint64_t shared_perm,
     return 0;
 }
 
+int blk_set_perm(BlockBackend *blk, uint64_t perm, uint64_t shared_perm,
+                 Error **errp)
+{
+    GLOBAL_STATE_CODE();
+    GRAPH_RDLOCK_GUARD_MAINLOOP();
+
+    return blk_set_perm_locked(blk, perm, shared_perm, errp);
+}
+
 void blk_get_perm(BlockBackend *blk, uint64_t *perm, uint64_t *shared_perm)
 {
     GLOBAL_STATE_CODE();
@@ -1533,7 +1547,7 @@ BlockAIOCB *blk_abort_aio_request(BlockBackend *blk,
     acb->blk = blk;
     acb->ret = ret;
 
-    replay_bh_schedule_oneshot_event(blk_get_aio_context(blk),
+    replay_bh_schedule_oneshot_event(qemu_get_current_aio_context(),
                                      error_callback_bh, acb);
     return &acb->common;
 }
@@ -1545,16 +1559,8 @@ typedef struct BlkAioEmAIOCB {
     bool has_returned;
 } BlkAioEmAIOCB;
 
-static AioContext *blk_aio_em_aiocb_get_aio_context(BlockAIOCB *acb_)
-{
-    BlkAioEmAIOCB *acb = container_of(acb_, BlkAioEmAIOCB, common);
-
-    return blk_get_aio_context(acb->rwco.blk);
-}
-
 static const AIOCBInfo blk_aio_em_aiocb_info = {
     .aiocb_size         = sizeof(BlkAioEmAIOCB),
-    .get_aio_context    = blk_aio_em_aiocb_get_aio_context,
 };
 
 static void blk_aio_complete(BlkAioEmAIOCB *acb)
@@ -1595,11 +1601,11 @@ static BlockAIOCB *blk_aio_prwv(BlockBackend *blk, int64_t offset,
     acb->has_returned = false;
 
     co = qemu_coroutine_create(co_entry, acb);
-    aio_co_enter(blk_get_aio_context(blk), co);
+    aio_co_enter(qemu_get_current_aio_context(), co);
 
     acb->has_returned = true;
     if (acb->rwco.ret != NOT_DONE) {
-        replay_bh_schedule_oneshot_event(blk_get_aio_context(blk),
+        replay_bh_schedule_oneshot_event(qemu_get_current_aio_context(),
                                          blk_aio_complete_bh, acb);
     }
 
@@ -1901,11 +1907,11 @@ BlockAIOCB *blk_aio_zone_report(BlockBackend *blk, int64_t offset,
     acb->has_returned = false;
 
     co = qemu_coroutine_create(blk_aio_zone_report_entry, acb);
-    aio_co_enter(blk_get_aio_context(blk), co);
+    aio_co_enter(qemu_get_current_aio_context(), co);
 
     acb->has_returned = true;
     if (acb->rwco.ret != NOT_DONE) {
-        replay_bh_schedule_oneshot_event(blk_get_aio_context(blk),
+        replay_bh_schedule_oneshot_event(qemu_get_current_aio_context(),
                                          blk_aio_complete_bh, acb);
     }
 
@@ -1942,11 +1948,11 @@ BlockAIOCB *blk_aio_zone_mgmt(BlockBackend *blk, BlockZoneOp op,
     acb->has_returned = false;
 
     co = qemu_coroutine_create(blk_aio_zone_mgmt_entry, acb);
-    aio_co_enter(blk_get_aio_context(blk), co);
+    aio_co_enter(qemu_get_current_aio_context(), co);
 
     acb->has_returned = true;
     if (acb->rwco.ret != NOT_DONE) {
-        replay_bh_schedule_oneshot_event(blk_get_aio_context(blk),
+        replay_bh_schedule_oneshot_event(qemu_get_current_aio_context(),
                                          blk_aio_complete_bh, acb);
     }
 
@@ -1982,10 +1988,10 @@ BlockAIOCB *blk_aio_zone_append(BlockBackend *blk, int64_t *offset,
     acb->has_returned = false;
 
     co = qemu_coroutine_create(blk_aio_zone_append_entry, acb);
-    aio_co_enter(blk_get_aio_context(blk), co);
+    aio_co_enter(qemu_get_current_aio_context(), co);
     acb->has_returned = true;
     if (acb->rwco.ret != NOT_DONE) {
-        replay_bh_schedule_oneshot_event(blk_get_aio_context(blk),
+        replay_bh_schedule_oneshot_event(qemu_get_current_aio_context(),
                                          blk_aio_complete_bh, acb);
     }
 
@@ -2434,12 +2440,6 @@ AioContext *blk_get_aio_context(BlockBackend *blk)
     return blk->ctx;
 }
 
-static AioContext *blk_aiocb_get_aio_context(BlockAIOCB *acb)
-{
-    BlockBackendAIOCB *blk_acb = DO_UPCAST(BlockBackendAIOCB, common, acb);
-    return blk_get_aio_context(blk_acb->blk);
-}
-
 int blk_set_aio_context(BlockBackend *blk, AioContext *new_context,
                         Error **errp)
 {

diff --git a/block/copy-before-write.c b/block/copy-before-write.c
@@ -341,11 +341,11 @@ static void cbw_refresh_filename(BlockDriverState *bs)
             bs->file->bs->filename);
 }
 
-static void cbw_child_perm(BlockDriverState *bs, BdrvChild *c,
-                           BdrvChildRole role,
-                           BlockReopenQueue *reopen_queue,
-                           uint64_t perm, uint64_t shared,
-                           uint64_t *nperm, uint64_t *nshared)
+static void GRAPH_RDLOCK
+cbw_child_perm(BlockDriverState *bs, BdrvChild *c, BdrvChildRole role,
+               BlockReopenQueue *reopen_queue,
+               uint64_t perm, uint64_t shared,
+               uint64_t *nperm, uint64_t *nshared)
 {
     if (!(role & BDRV_CHILD_FILTERED)) {
         /*

diff --git a/block/crypto.c b/block/crypto.c
@@ -777,7 +777,7 @@ block_crypto_get_specific_info_luks(BlockDriverState *bs, Error **errp)
     return spec_info;
 }
 
-static int
+static int GRAPH_RDLOCK
 block_crypto_amend_prepare(BlockDriverState *bs, Error **errp)
 {
     BlockCrypto *crypto = bs->opaque;
@@ -793,7 +793,7 @@ block_crypto_amend_prepare(BlockDriverState *bs, Error **errp)
     return ret;
 }
 
-static void
+static void GRAPH_RDLOCK
 block_crypto_amend_cleanup(BlockDriverState *bs)
 {
     BlockCrypto *crypto = bs->opaque;
@@ -841,6 +841,8 @@ block_crypto_amend_options_luks(BlockDriverState *bs,
     QCryptoBlockAmendOptions *amend_options = NULL;
     int ret = -EINVAL;
 
+    assume_graph_lock(); /* FIXME */
+
     assert(crypto);
     assert(crypto->block);
 

diff --git a/block/graph-lock.c b/block/graph-lock.c
@@ -163,17 +163,29 @@ void bdrv_graph_wrlock(BlockDriverState *bs)
 void bdrv_graph_wrunlock(void)
 {
     GLOBAL_STATE_CODE();
-    QEMU_LOCK_GUARD(&aio_context_list_lock);
     assert(qatomic_read(&has_writer));
 
+    WITH_QEMU_LOCK_GUARD(&aio_context_list_lock) {
+        /*
+         * No need for memory barriers, this works in pair with
+         * the slow path of rdlock() and both take the lock.
+         */
+        qatomic_store_release(&has_writer, 0);
+
+        /* Wake up all coroutines that are waiting to read the graph */
+        qemu_co_enter_all(&reader_queue, &aio_context_list_lock);
+    }
+
     /*
-     * No need for memory barriers, this works in pair with
-     * the slow path of rdlock() and both take the lock.
+     * Run any BHs that were scheduled during the wrlock section and that
+     * callers might expect to have finished (in particular, this is important
+     * for bdrv_schedule_unref()).
+     *
+     * Do this only after restarting coroutines so that nested event loops in
+     * BHs don't deadlock if their condition relies on the coroutine making
+     * progress.
      */
-    qatomic_store_release(&has_writer, 0);
-
-    /* Wake up all coroutine that are waiting to read the graph */
-    qemu_co_enter_all(&reader_queue, &aio_context_list_lock);
+    aio_bh_poll(qemu_get_aio_context());
 }
 
 void coroutine_fn bdrv_graph_co_rdlock(void)

diff --git a/block/io.c b/block/io.c
@@ -2950,25 +2950,18 @@ int bdrv_load_vmstate(BlockDriverState *bs, uint8_t *buf,
 /**************************************************************/
 /* async I/Os */
 
+/**
+ * Synchronously cancels an acb. Must be called with the BQL held and the acb
+ * must be processed with the BQL held too (IOThreads are not allowed).
+ *
+ * Use bdrv_aio_cancel_async() instead when possible.
+ */
 void bdrv_aio_cancel(BlockAIOCB *acb)
 {
-    IO_CODE();
+    GLOBAL_STATE_CODE();
     qemu_aio_ref(acb);
     bdrv_aio_cancel_async(acb);
-    while (acb->refcnt > 1) {
-        if (acb->aiocb_info->get_aio_context) {
-            aio_poll(acb->aiocb_info->get_aio_context(acb), true);
-        } else if (acb->bs) {
-            /* qemu_aio_ref and qemu_aio_unref are not thread-safe, so
-             * assert that we're not using an I/O thread.  Thread-safe
-             * code should use bdrv_aio_cancel_async exclusively.
-             */
-            assert(bdrv_get_aio_context(acb->bs) == qemu_get_aio_context());
-            aio_poll(bdrv_get_aio_context(acb->bs), true);
-        } else {
-            abort();
-        }
-    }
+    AIO_WAIT_WHILE_UNLOCKED(NULL, acb->refcnt > 1);
     qemu_aio_unref(acb);
 }
 

diff --git a/block/mirror.c b/block/mirror.c
@@ -702,8 +702,12 @@ static int mirror_exit_common(Job *job)
      * mirror_top_bs from now on, so keep it drained. */
     bdrv_drained_begin(mirror_top_bs);
     bs_opaque->stop = true;
+
+    bdrv_graph_rdlock_main_loop();
     bdrv_child_refresh_perms(mirror_top_bs, mirror_top_bs->backing,
                              &error_abort);
+    bdrv_graph_rdunlock_main_loop();
+
     if (!abort && s->backing_mode == MIRROR_SOURCE_BACKING_CHAIN) {
         BlockDriverState *backing = s->is_none_mode ? src : s->base;
         BlockDriverState *unfiltered_target = bdrv_skip_filters(target_bs);
@@ -1670,6 +1674,8 @@ static BlockJob *mirror_start_job(
     uint64_t target_perms, target_shared_perms;
     int ret;
 
+    GLOBAL_STATE_CODE();
+
     if (granularity == 0) {
         granularity = bdrv_get_default_bitmap_granularity(target);
     }
@@ -1906,8 +1912,10 @@ static BlockJob *mirror_start_job(
     }
 
     bs_opaque->stop = true;
+    bdrv_graph_rdlock_main_loop();
     bdrv_child_refresh_perms(mirror_top_bs, mirror_top_bs->backing,
                              &error_abort);
+    bdrv_graph_rdunlock_main_loop();
     bdrv_replace_node(mirror_top_bs, mirror_top_bs->backing->bs, &error_abort);
 
     bdrv_unref(mirror_top_bs);