Merge remote-tracking branch 'remotes/kevin/tags/for-upstream' into staging

Block layer patches

# gpg: Signature made Mon 10 Oct 2016 12:33:14 BST
# gpg:                using RSA key 0x7F09B272C88F2FD6
# gpg: Good signature from "Kevin Wolf <kwolf@redhat.com>"
# Primary key fingerprint: DC3D EB15 9A9A F95D 3D74  56FE 7F09 B272 C88F 2FD6

* remotes/kevin/tags/for-upstream:
  dmg: Move libbz2 code to dmg-bz2.so
  module: Don't load the same module if requested multiple times
  scripts: Allow block module to not define BlockDriver
  block: Add qdev ID to DEVICE_TRAY_MOVED
  block-backend: Remember if attached device is non-qdev
  block: Add node name to BLOCK_IO_ERROR event
  block: Add bdrv_runtime_opts to query-command-line-options
  block: use aio_bh_schedule_oneshot
  async: add aio_bh_schedule_oneshot
  block: use bdrv_add_before_write_notifier

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>

18 files changed, 220 insertions, 142 deletions

diff --git a/block/Makefile.objs b/block/Makefile.objs
index 7d4031dae5..67a036a1df 100644
--- a/block/Makefile.objs
+++ b/block/Makefile.objs
@@ -41,6 +41,7 @@ gluster.o-libs     := $(GLUSTERFS_LIBS)
 ssh.o-cflags       := $(LIBSSH2_CFLAGS)
 ssh.o-libs         := $(LIBSSH2_LIBS)
 archipelago.o-libs := $(ARCHIPELAGO_LIBS)
-dmg.o-libs         := $(BZIP2_LIBS)
+block-obj-$(if $(CONFIG_BZIP2),m,n) += dmg-bz2.o
+dmg-bz2.o-libs     := $(BZIP2_LIBS)
 qcow.o-libs        := -lz
 linux-aio.o-libs   := -laio
diff --git a/block/archipelago.c b/block/archipelago.c
index 37b8aca78d..2449cfc702 100644
--- a/block/archipelago.c
+++ b/block/archipelago.c
@@ -87,7 +87,6 @@ typedef enum {
 
 typedef struct ArchipelagoAIOCB {
     BlockAIOCB common;
-    QEMUBH *bh;
     struct BDRVArchipelagoState *s;
     QEMUIOVector *qiov;
     ARCHIPCmd cmd;
@@ -154,11 +153,10 @@ static void archipelago_finish_aiocb(AIORequestData *reqdata)
     } else if (reqdata->aio_cb->ret == reqdata->segreq->total) {
         reqdata->aio_cb->ret = 0;
     }
-    reqdata->aio_cb->bh = aio_bh_new(
+    aio_bh_schedule_oneshot(
                         bdrv_get_aio_context(reqdata->aio_cb->common.bs),
                         qemu_archipelago_complete_aio, reqdata
                         );
-    qemu_bh_schedule(reqdata->aio_cb->bh);
 }
 
 static int wait_reply(struct xseg *xseg, xport srcport, struct xseg_port *port,
@@ -313,7 +311,6 @@ static void qemu_archipelago_complete_aio(void *opaque)
     AIORequestData *reqdata = (AIORequestData *) opaque;
     ArchipelagoAIOCB *aio_cb = (ArchipelagoAIOCB *) reqdata->aio_cb;
 
-    qemu_bh_delete(aio_cb->bh);
     aio_cb->common.cb(aio_cb->common.opaque, aio_cb->ret);
     aio_cb->status = 0;
 
diff --git a/block/blkdebug.c b/block/blkdebug.c
index d5db166815..4127571454 100644
--- a/block/blkdebug.c
+++ b/block/blkdebug.c
@@ -49,7 +49,6 @@ typedef struct BDRVBlkdebugState {
 
 typedef struct BlkdebugAIOCB {
     BlockAIOCB common;
-    QEMUBH *bh;
     int ret;
 } BlkdebugAIOCB;
 
@@ -410,7 +409,6 @@ out:
 static void error_callback_bh(void *opaque)
 {
     struct BlkdebugAIOCB *acb = opaque;
-    qemu_bh_delete(acb->bh);
     acb->common.cb(acb->common.opaque, acb->ret);
     qemu_aio_unref(acb);
 }
@@ -421,7 +419,6 @@ static BlockAIOCB *inject_error(BlockDriverState *bs,
     BDRVBlkdebugState *s = bs->opaque;
     int error = rule->options.inject.error;
     struct BlkdebugAIOCB *acb;
-    QEMUBH *bh;
     bool immediately = rule->options.inject.immediately;
 
     if (rule->options.inject.once) {
@@ -436,9 +433,7 @@ static BlockAIOCB *inject_error(BlockDriverState *bs,
     acb = qemu_aio_get(&blkdebug_aiocb_info, bs, cb, opaque);
     acb->ret = -error;
 
-    bh = aio_bh_new(bdrv_get_aio_context(bs), error_callback_bh, acb);
-    acb->bh = bh;
-    qemu_bh_schedule(bh);
+    aio_bh_schedule_oneshot(bdrv_get_aio_context(bs), error_callback_bh, acb);
 
     return &acb->common;
 }
diff --git a/block/blkverify.c b/block/blkverify.c
index da62d75969..28f9af6dba 100644
--- a/block/blkverify.c
+++ b/block/blkverify.c
@@ -22,7 +22,6 @@ typedef struct {
 typedef struct BlkverifyAIOCB BlkverifyAIOCB;
 struct BlkverifyAIOCB {
     BlockAIOCB common;
-    QEMUBH *bh;
 
     /* Request metadata */
     bool is_write;
@@ -175,7 +174,6 @@ static BlkverifyAIOCB *blkverify_aio_get(BlockDriverState *bs, bool is_write,
 {
     BlkverifyAIOCB *acb = qemu_aio_get(&blkverify_aiocb_info, bs, cb, opaque);
 
-    acb->bh = NULL;
     acb->is_write = is_write;
     acb->sector_num = sector_num;
     acb->nb_sectors = nb_sectors;
@@ -191,7 +189,6 @@ static void blkverify_aio_bh(void *opaque)
 {
     BlkverifyAIOCB *acb = opaque;
 
-    qemu_bh_delete(acb->bh);
     if (acb->buf) {
         qemu_iovec_destroy(&acb->raw_qiov);
         qemu_vfree(acb->buf);
@@ -218,9 +215,8 @@ static void blkverify_aio_cb(void *opaque, int ret)
             acb->verify(acb);
         }
 
-        acb->bh = aio_bh_new(bdrv_get_aio_context(acb->common.bs),
-                             blkverify_aio_bh, acb);
-        qemu_bh_schedule(acb->bh);
+        aio_bh_schedule_oneshot(bdrv_get_aio_context(acb->common.bs),
+                                blkverify_aio_bh, acb);
         break;
     }
 }
diff --git a/block/block-backend.c b/block/block-backend.c
index 639294b8e6..1a724a8d89 100644
--- a/block/block-backend.c
+++ b/block/block-backend.c
@@ -38,6 +38,7 @@ struct BlockBackend {
     BlockBackendPublic public;
 
     void *dev;                  /* attached device model, if any */
+    bool legacy_dev;            /* true if dev is not a DeviceState */
     /* TODO change to DeviceState when all users are qdevified */
     const BlockDevOps *dev_ops;
     void *dev_opaque;
@@ -65,7 +66,6 @@ struct BlockBackend {
 
 typedef struct BlockBackendAIOCB {
     BlockAIOCB common;
-    QEMUBH *bh;
     BlockBackend *blk;
     int ret;
 } BlockBackendAIOCB;
@@ -507,32 +507,38 @@ void blk_insert_bs(BlockBackend *blk, BlockDriverState *bs)
     }
 }
 
-/*
- * Attach device model @dev to @blk.
- * Return 0 on success, -EBUSY when a device model is attached already.
- */
-int blk_attach_dev(BlockBackend *blk, void *dev)
-/* TODO change to DeviceState *dev when all users are qdevified */
+static int blk_do_attach_dev(BlockBackend *blk, void *dev)
 {
     if (blk->dev) {
         return -EBUSY;
     }
     blk_ref(blk);
     blk->dev = dev;
+    blk->legacy_dev = false;
     blk_iostatus_reset(blk);
     return 0;
 }
 
 /*
  * Attach device model @dev to @blk.
+ * Return 0 on success, -EBUSY when a device model is attached already.
+ */
+int blk_attach_dev(BlockBackend *blk, DeviceState *dev)
+{
+    return blk_do_attach_dev(blk, dev);
+}
+
+/*
+ * Attach device model @dev to @blk.
  * @blk must not have a device model attached already.
  * TODO qdevified devices don't use this, remove when devices are qdevified
  */
-void blk_attach_dev_nofail(BlockBackend *blk, void *dev)
+void blk_attach_dev_legacy(BlockBackend *blk, void *dev)
 {
-    if (blk_attach_dev(blk, dev) < 0) {
+    if (blk_do_attach_dev(blk, dev) < 0) {
         abort();
     }
+    blk->legacy_dev = true;
 }
 
 /*
@@ -559,6 +565,23 @@ void *blk_get_attached_dev(BlockBackend *blk)
     return blk->dev;
 }
 
+/* Return the qdev ID, or if no ID is assigned the QOM path, of the block
+ * device attached to the BlockBackend. */
+static char *blk_get_attached_dev_id(BlockBackend *blk)
+{
+    DeviceState *dev;
+
+    assert(!blk->legacy_dev);
+    dev = blk->dev;
+
+    if (!dev) {
+        return g_strdup("");
+    } else if (dev->id) {
+        return g_strdup(dev->id);
+    }
+    return object_get_canonical_path(OBJECT(dev));
+}
+
 /*
  * Return the BlockBackend which has the device model @dev attached if it
  * exists, else null.
@@ -586,6 +609,11 @@ BlockBackend *blk_by_dev(void *dev)
 void blk_set_dev_ops(BlockBackend *blk, const BlockDevOps *ops,
                      void *opaque)
 {
+    /* All drivers that use blk_set_dev_ops() are qdevified and we want to keep
+     * it that way, so we can assume blk->dev is a DeviceState if blk->dev_ops
+     * is set. */
+    assert(!blk->legacy_dev);
+
     blk->dev_ops = ops;
     blk->dev_opaque = opaque;
 }
@@ -601,13 +629,17 @@ void blk_dev_change_media_cb(BlockBackend *blk, bool load)
     if (blk->dev_ops && blk->dev_ops->change_media_cb) {
         bool tray_was_open, tray_is_open;
 
+        assert(!blk->legacy_dev);
+
         tray_was_open = blk_dev_is_tray_open(blk);
         blk->dev_ops->change_media_cb(blk->dev_opaque, load);
         tray_is_open = blk_dev_is_tray_open(blk);
 
         if (tray_was_open != tray_is_open) {
-            qapi_event_send_device_tray_moved(blk_name(blk), tray_is_open,
+            char *id = blk_get_attached_dev_id(blk);
+            qapi_event_send_device_tray_moved(blk_name(blk), id, tray_is_open,
                                               &error_abort);
+            g_free(id);
         }
     }
 }
@@ -898,7 +930,6 @@ int blk_make_zero(BlockBackend *blk, BdrvRequestFlags flags)
 static void error_callback_bh(void *opaque)
 {
     struct BlockBackendAIOCB *acb = opaque;
-    qemu_bh_delete(acb->bh);
     acb->common.cb(acb->common.opaque, acb->ret);
     qemu_aio_unref(acb);
 }
@@ -908,16 +939,12 @@ BlockAIOCB *blk_abort_aio_request(BlockBackend *blk,
                                   void *opaque, int ret)
 {
     struct BlockBackendAIOCB *acb;
-    QEMUBH *bh;
 
     acb = blk_aio_get(&block_backend_aiocb_info, blk, cb, opaque);
     acb->blk = blk;
     acb->ret = ret;
 
-    bh = aio_bh_new(blk_get_aio_context(blk), error_callback_bh, acb);
-    acb->bh = bh;
-    qemu_bh_schedule(bh);
-
+    aio_bh_schedule_oneshot(blk_get_aio_context(blk), error_callback_bh, acb);
     return &acb->common;
 }
 
@@ -926,7 +953,6 @@ typedef struct BlkAioEmAIOCB {
     BlkRwCo rwco;
     int bytes;
     bool has_returned;
-    QEMUBH* bh;
 } BlkAioEmAIOCB;
 
 static const AIOCBInfo blk_aio_em_aiocb_info = {
@@ -935,10 +961,6 @@ static const AIOCBInfo blk_aio_em_aiocb_info = {
 
 static void blk_aio_complete(BlkAioEmAIOCB *acb)
 {
-    if (acb->bh) {
-        assert(acb->has_returned);
-        qemu_bh_delete(acb->bh);
-    }
     if (acb->has_returned) {
         acb->common.cb(acb->common.opaque, acb->rwco.ret);
         qemu_aio_unref(acb);
@@ -947,7 +969,10 @@ static void blk_aio_complete(BlkAioEmAIOCB *acb)
 
 static void blk_aio_complete_bh(void *opaque)
 {
-    blk_aio_complete(opaque);
+    BlkAioEmAIOCB *acb = opaque;
+
+    assert(acb->has_returned);
+    blk_aio_complete(acb);
 }
 
 static BlockAIOCB *blk_aio_prwv(BlockBackend *blk, int64_t offset, int bytes,
@@ -967,7 +992,6 @@ static BlockAIOCB *blk_aio_prwv(BlockBackend *blk, int64_t offset, int bytes,
         .ret    = NOT_DONE,
     };
     acb->bytes = bytes;
-    acb->bh = NULL;
     acb->has_returned = false;
 
     co = qemu_coroutine_create(co_entry, acb);
@@ -975,8 +999,8 @@ static BlockAIOCB *blk_aio_prwv(BlockBackend *blk, int64_t offset, int bytes,
 
     acb->has_returned = true;
     if (acb->rwco.ret != NOT_DONE) {
-        acb->bh = aio_bh_new(blk_get_aio_context(blk), blk_aio_complete_bh, acb);
-        qemu_bh_schedule(acb->bh);
+        aio_bh_schedule_oneshot(blk_get_aio_context(blk),
+                                blk_aio_complete_bh, acb);
     }
 
     return &acb->common;
@@ -1206,8 +1230,9 @@ static void send_qmp_error_event(BlockBackend *blk,
     IoOperationType optype;
 
     optype = is_read ? IO_OPERATION_TYPE_READ : IO_OPERATION_TYPE_WRITE;
-    qapi_event_send_block_io_error(blk_name(blk), optype, action,
-                                   blk_iostatus_is_enabled(blk),
+    qapi_event_send_block_io_error(blk_name(blk),
+                                   bdrv_get_node_name(blk_bs(blk)), optype,
+                                   action, blk_iostatus_is_enabled(blk),
                                    error == ENOSPC, strerror(error),
                                    &error_abort);
 }
@@ -1312,9 +1337,19 @@ void blk_lock_medium(BlockBackend *blk, bool locked)
 void blk_eject(BlockBackend *blk, bool eject_flag)
 {
     BlockDriverState *bs = blk_bs(blk);
+    char *id;
+
+    /* blk_eject is only called by qdevified devices */
+    assert(!blk->legacy_dev);
 
     if (bs) {
         bdrv_eject(bs, eject_flag);
+
+        id = blk_get_attached_dev_id(blk);
+        qapi_event_send_device_tray_moved(blk_name(blk), id,
+                                          eject_flag, &error_abort);
+        g_free(id);
+
     }
 }
 
diff --git a/block/curl.c b/block/curl.c
index 571f24cac2..e5eaa7ba0a 100644
--- a/block/curl.c
+++ b/block/curl.c
@@ -96,7 +96,6 @@ struct BDRVCURLState;
 
 typedef struct CURLAIOCB {
     BlockAIOCB common;
-    QEMUBH *bh;
     QEMUIOVector *qiov;
 
     int64_t sector_num;
@@ -739,9 +738,6 @@ static void curl_readv_bh_cb(void *p)
     CURLAIOCB *acb = p;
     BDRVCURLState *s = acb->common.bs->opaque;
 
-    qemu_bh_delete(acb->bh);
-    acb->bh = NULL;
-
     size_t start = acb->sector_num * SECTOR_SIZE;
     size_t end;
 
@@ -805,8 +801,7 @@ static BlockAIOCB *curl_aio_readv(BlockDriverState *bs,
     acb->sector_num = sector_num;
     acb->nb_sectors = nb_sectors;
 
-    acb->bh = aio_bh_new(bdrv_get_aio_context(bs), curl_readv_bh_cb, acb);
-    qemu_bh_schedule(acb->bh);
+    aio_bh_schedule_oneshot(bdrv_get_aio_context(bs), curl_readv_bh_cb, acb);
     return &acb->common;
 }
 
diff --git a/block/dmg-bz2.c b/block/dmg-bz2.c
new file mode 100644
index 0000000000..9059492a9f
--- /dev/null
+++ b/block/dmg-bz2.c
@@ -0,0 +1,61 @@
+/*
+ * DMG bzip2 uncompression
+ *
+ * Copyright (c) 2004 Johannes E. Schindelin
+ * Copyright (c) 2016 Red Hat, Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a copy
+ * of this software and associated documentation files (the "Software"), to deal
+ * in the Software without restriction, including without limitation the rights
+ * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+ * copies of the Software, and to permit persons to whom the Software is
+ * furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
+ * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+ * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
+ * THE SOFTWARE.
+ */
+#include "qemu/osdep.h"
+#include "qemu-common.h"
+#include "dmg.h"
+#include <bzlib.h>
+
+static int dmg_uncompress_bz2_do(char *next_in, unsigned int avail_in,
+                                 char *next_out, unsigned int avail_out)
+{
+    int ret;
+    uint64_t total_out;
+    bz_stream bzstream = {};
+
+    ret = BZ2_bzDecompressInit(&bzstream, 0, 0);
+    if (ret != BZ_OK) {
+        return -1;
+    }
+    bzstream.next_in = next_in;
+    bzstream.avail_in = avail_in;
+    bzstream.next_out = next_out;
+    bzstream.avail_out = avail_out;
+    ret = BZ2_bzDecompress(&bzstream);
+    total_out = ((uint64_t)bzstream.total_out_hi32 << 32) +
+                bzstream.total_out_lo32;
+    BZ2_bzDecompressEnd(&bzstream);
+    if (ret != BZ_STREAM_END ||
+        total_out != avail_out) {
+        return -1;
+    }
+    return 0;
+}
+
+__attribute__((constructor))
+static void dmg_bz2_init(void)
+{
+    assert(!dmg_uncompress_bz2);
+    dmg_uncompress_bz2 = dmg_uncompress_bz2_do;
+}
diff --git a/block/dmg.c b/block/dmg.c
index b0ed89baa7..58a3ae86c1 100644
--- a/block/dmg.c
+++ b/block/dmg.c
@@ -28,10 +28,10 @@
 #include "qemu/bswap.h"
 #include "qemu/error-report.h"
 #include "qemu/module.h"
-#include <zlib.h>
-#ifdef CONFIG_BZIP2
-#include <bzlib.h>
-#endif
+#include "dmg.h"
+
+int (*dmg_uncompress_bz2)(char *next_in, unsigned int avail_in,
+                          char *next_out, unsigned int avail_out);
 
 enum {
     /* Limit chunk sizes to prevent unreasonable amounts of memory being used
@@ -41,31 +41,6 @@ enum {
     DMG_SECTORCOUNTS_MAX = DMG_LENGTHS_MAX / 512,
 };
 
-typedef struct BDRVDMGState {
-    CoMutex lock;
-    /* each chunk contains a certain number of sectors,
-     * offsets[i] is the offset in the .dmg file,
-     * lengths[i] is the length of the compressed chunk,
-     * sectors[i] is the sector beginning at offsets[i],
-     * sectorcounts[i] is the number of sectors in that chunk,
-     * the sectors array is ordered
-     * 0<=i<n_chunks */
-
-    uint32_t n_chunks;
-    uint32_t* types;
-    uint64_t* offsets;
-    uint64_t* lengths;
-    uint64_t* sectors;
-    uint64_t* sectorcounts;
-    uint32_t current_chunk;
-    uint8_t *compressed_chunk;
-    uint8_t *uncompressed_chunk;
-    z_stream zstream;
-#ifdef CONFIG_BZIP2
-    bz_stream bzstream;
-#endif
-} BDRVDMGState;
-
 static int dmg_probe(const uint8_t *buf, int buf_size, const char *filename)
 {
     int len;
@@ -210,10 +185,9 @@ static bool dmg_is_known_block_type(uint32_t entry_type)
     case 0x00000001:    /* uncompressed */
     case 0x00000002:    /* zeroes */
     case 0x80000005:    /* zlib */
-#ifdef CONFIG_BZIP2
-    case 0x80000006:    /* bzip2 */
-#endif
         return true;
+    case 0x80000006:    /* bzip2 */
+        return !!dmg_uncompress_bz2;
     default:
         return false;
     }
@@ -439,6 +413,7 @@ static int dmg_open(BlockDriverState *bs, QDict *options, int flags,
     int64_t offset;
     int ret;
 
+    block_module_load_one("dmg-bz2");
     bs->read_only = true;
 
     s->n_chunks = 0;
@@ -587,9 +562,6 @@ static inline int dmg_read_chunk(BlockDriverState *bs, uint64_t sector_num)
     if (!is_sector_in_chunk(s, s->current_chunk, sector_num)) {
         int ret;
         uint32_t chunk = search_chunk(s, sector_num);
-#ifdef CONFIG_BZIP2
-        uint64_t total_out;
-#endif
 
         if (chunk >= s->n_chunks) {
             return -1;
@@ -620,8 +592,10 @@ static inline int dmg_read_chunk(BlockDriverState *bs, uint64_t sector_num)
                 return -1;
             }
             break; }
-#ifdef CONFIG_BZIP2
         case 0x80000006: /* bzip2 compressed */
+            if (!dmg_uncompress_bz2) {
+                break;
+            }
             /* we need to buffer, because only the chunk as whole can be
              * inflated. */
             ret = bdrv_pread(bs->file, s->offsets[chunk],
@@ -630,24 +604,15 @@ static inline int dmg_read_chunk(BlockDriverState *bs, uint64_t sector_num)
                 return -1;
             }
 
-            ret = BZ2_bzDecompressInit(&s->bzstream, 0, 0);
-            if (ret != BZ_OK) {
-                return -1;
-            }
-            s->bzstream.next_in = (char *)s->compressed_chunk;
-            s->bzstream.avail_in = (unsigned int) s->lengths[chunk];
-            s->bzstream.next_out = (char *)s->uncompressed_chunk;
-            s->bzstream.avail_out = (unsigned int) 512 * s->sectorcounts[chunk];
-            ret = BZ2_bzDecompress(&s->bzstream);
-            total_out = ((uint64_t)s->bzstream.total_out_hi32 << 32) +
-                        s->bzstream.total_out_lo32;
-            BZ2_bzDecompressEnd(&s->bzstream);
-            if (ret != BZ_STREAM_END ||
-                total_out != 512 * s->sectorcounts[chunk]) {
-                return -1;
+            ret = dmg_uncompress_bz2((char *)s->compressed_chunk,
+                                     (unsigned int) s->lengths[chunk],
+                                     (char *)s->uncompressed_chunk,
+                                     (unsigned int)
+                                         (512 * s->sectorcounts[chunk]));
+            if (ret < 0) {
+                return ret;
             }
             break;
-#endif /* CONFIG_BZIP2 */
         case 1: /* copy */
             ret = bdrv_pread(bs->file, s->offsets[chunk],
                              s->uncompressed_chunk, s->lengths[chunk]);
diff --git a/block/dmg.h b/block/dmg.h
new file mode 100644
index 0000000000..b592d6fa8b
--- /dev/null
+++ b/block/dmg.h
@@ -0,0 +1,59 @@
+/*
+ * Header for DMG driver
+ *
+ * Copyright (c) 2004-2006 Fabrice Bellard
+ * Copyright (c) 2016 Red hat, Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a copy
+ * of this software and associated documentation files (the "Software"), to deal
+ * in the Software without restriction, including without limitation the rights
+ * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+ * copies of the Software, and to permit persons to whom the Software is
+ * furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
+ * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+ * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
+ * THE SOFTWARE.
+ */
+
+#ifndef BLOCK_DMG_H
+#define BLOCK_DMG_H
+
+#include "qemu/osdep.h"
+#include "qemu-common.h"
+#include "block/block_int.h"
+#include <zlib.h>
+
+typedef struct BDRVDMGState {
+    CoMutex lock;
+    /* each chunk contains a certain number of sectors,
+     * offsets[i] is the offset in the .dmg file,
+     * lengths[i] is the length of the compressed chunk,
+     * sectors[i] is the sector beginning at offsets[i],
+     * sectorcounts[i] is the number of sectors in that chunk,
+     * the sectors array is ordered
+     * 0<=i<n_chunks */
+
+    uint32_t n_chunks;
+    uint32_t *types;
+    uint64_t *offsets;
+    uint64_t *lengths;
+    uint64_t *sectors;
+    uint64_t *sectorcounts;
+    uint32_t current_chunk;
+    uint8_t *compressed_chunk;
+    uint8_t *uncompressed_chunk;
+    z_stream zstream;
+} BDRVDMGState;
+
+extern int (*dmg_uncompress_bz2)(char *next_in, unsigned int avail_in,
+                                 char *next_out, unsigned int avail_out);
+
+#endif
diff --git a/block/gluster.c b/block/gluster.c
index e7bd13cce3..af76d7d59a 100644
--- a/block/gluster.c
+++ b/block/gluster.c
@@ -38,7 +38,6 @@
 typedef struct GlusterAIOCB {
     int64_t size;
     int ret;
-    QEMUBH *bh;
     Coroutine *coroutine;
     AioContext *aio_context;
 } GlusterAIOCB;
@@ -622,8 +621,6 @@ static void qemu_gluster_complete_aio(void *opaque)
 {
     GlusterAIOCB *acb = (GlusterAIOCB *)opaque;
 
-    qemu_bh_delete(acb->bh);
-    acb->bh = NULL;
     qemu_coroutine_enter(acb->coroutine);
 }
 
@@ -642,8 +639,7 @@ static void gluster_finish_aiocb(struct glfs_fd *fd, ssize_t ret, void *arg)
         acb->ret = -EIO; /* Partial read/write - fail it */
     }
 
-    acb->bh = aio_bh_new(acb->aio_context, qemu_gluster_complete_aio, acb);
-    qemu_bh_schedule(acb->bh);
+    aio_bh_schedule_oneshot(acb->aio_context, qemu_gluster_complete_aio, acb);
 }
 
 static void qemu_gluster_parse_flags(int bdrv_flags, int *open_flags)
diff --git a/block/io.c b/block/io.c
index 57a2eeb512..b136c89ae0 100644
--- a/block/io.c
+++ b/block/io.c
@@ -171,7 +171,6 @@ static void bdrv_drain_recurse(BlockDriverState *bs)
 typedef struct {
     Coroutine *co;
     BlockDriverState *bs;
-    QEMUBH *bh;
     bool done;
 } BdrvCoDrainData;
 
@@ -191,7 +190,6 @@ static void bdrv_co_drain_bh_cb(void *opaque)
     BdrvCoDrainData *data = opaque;
     Coroutine *co = data->co;
 
-    qemu_bh_delete(data->bh);
     bdrv_drain_poll(data->bs);
     data->done = true;
     qemu_coroutine_enter(co);
@@ -210,9 +208,9 @@ static void coroutine_fn bdrv_co_yield_to_drain(BlockDriverState *bs)
         .co = qemu_coroutine_self(),
         .bs = bs,
         .done = false,
-        .bh = aio_bh_new(bdrv_get_aio_context(bs), bdrv_co_drain_bh_cb, &data),
     };
-    qemu_bh_schedule(data.bh);
+    aio_bh_schedule_oneshot(bdrv_get_aio_context(bs),
+                            bdrv_co_drain_bh_cb, &data);
 
     qemu_coroutine_yield();
     /* If we are resumed from some other event (such as an aio completion or a
@@ -2095,7 +2093,6 @@ typedef struct BlockAIOCBCoroutine {
     bool is_write;
     bool need_bh;
     bool *done;
-    QEMUBH* bh;
 } BlockAIOCBCoroutine;
 
 static const AIOCBInfo bdrv_em_co_aiocb_info = {
@@ -2115,7 +2112,6 @@ static void bdrv_co_em_bh(void *opaque)
     BlockAIOCBCoroutine *acb = opaque;
 
     assert(!acb->need_bh);
-    qemu_bh_delete(acb->bh);
     bdrv_co_complete(acb);
 }
 
@@ -2125,8 +2121,7 @@ static void bdrv_co_maybe_schedule_bh(BlockAIOCBCoroutine *acb)
     if (acb->req.error != -EINPROGRESS) {
         BlockDriverState *bs = acb->common.bs;
 
-        acb->bh = aio_bh_new(bdrv_get_aio_context(bs), bdrv_co_em_bh, acb);
-        qemu_bh_schedule(acb->bh);
+        aio_bh_schedule_oneshot(bdrv_get_aio_context(bs), bdrv_co_em_bh, acb);
     }
 }
 
diff --git a/block/iscsi.c b/block/iscsi.c
index dff548a139..46ddc355ac 100644
--- a/block/iscsi.c
+++ b/block/iscsi.c
@@ -95,7 +95,6 @@ typedef struct IscsiTask {
     int do_retry;
     struct scsi_task *task;
     Coroutine *co;
-    QEMUBH *bh;
     IscsiLun *iscsilun;
     QEMUTimer retry_timer;
     int err_code;
@@ -167,7 +166,6 @@ static void iscsi_co_generic_bh_cb(void *opaque)
 {
     struct IscsiTask *iTask = opaque;
     iTask->complete = 1;
-    qemu_bh_delete(iTask->bh);
     qemu_coroutine_enter(iTask->co);
 }
 
@@ -299,9 +297,8 @@ iscsi_co_generic_cb(struct iscsi_context *iscsi, int status,
 
 out:
     if (iTask->co) {
-        iTask->bh = aio_bh_new(iTask->iscsilun->aio_context,
-                               iscsi_co_generic_bh_cb, iTask);
-        qemu_bh_schedule(iTask->bh);
+        aio_bh_schedule_oneshot(iTask->iscsilun->aio_context,
+                                 iscsi_co_generic_bh_cb, iTask);
     } else {
         iTask->complete = 1;
     }
diff --git a/block/nfs.c b/block/nfs.c
index 8602a44211..c3db2ec58d 100644
--- a/block/nfs.c
+++ b/block/nfs.c
@@ -57,7 +57,6 @@ typedef struct NFSRPC {
     QEMUIOVector *iov;
     struct stat *st;
     Coroutine *co;
-    QEMUBH *bh;
     NFSClient *client;
 } NFSRPC;
 
@@ -103,7 +102,6 @@ static void nfs_co_generic_bh_cb(void *opaque)
 {
     NFSRPC *task = opaque;
     task->complete = 1;
-    qemu_bh_delete(task->bh);
     qemu_coroutine_enter(task->co);
 }
 
@@ -127,9 +125,8 @@ nfs_co_generic_cb(int ret, struct nfs_context *nfs, void *data,
         error_report("NFS Error: %s", nfs_get_error(nfs));
     }
     if (task->co) {
-        task->bh = aio_bh_new(task->client->aio_context,
-                              nfs_co_generic_bh_cb, task);
-        qemu_bh_schedule(task->bh);
+        aio_bh_schedule_oneshot(task->client->aio_context,
+                                nfs_co_generic_bh_cb, task);
     } else {
         task->complete = 1;
     }
diff --git a/block/null.c b/block/null.c
index b511010ba5..b300390944 100644
--- a/block/null.c
+++ b/block/null.c
@@ -124,7 +124,6 @@ static coroutine_fn int null_co_flush(BlockDriverState *bs)
 
 typedef struct {
     BlockAIOCB common;
-    QEMUBH *bh;
     QEMUTimer timer;
 } NullAIOCB;
 
@@ -136,7 +135,6 @@ static void null_bh_cb(void *opaque)
 {
     NullAIOCB *acb = opaque;
     acb->common.cb(acb->common.opaque, 0);
-    qemu_bh_delete(acb->bh);
     qemu_aio_unref(acb);
 }
 
@@ -164,8 +162,7 @@ static inline BlockAIOCB *null_aio_common(BlockDriverState *bs,
         timer_mod_ns(&acb->timer,
                      qemu_clock_get_ns(QEMU_CLOCK_REALTIME) + s->latency_ns);
     } else {
-        acb->bh = aio_bh_new(bdrv_get_aio_context(bs), null_bh_cb, acb);
-        qemu_bh_schedule(acb->bh);
+        aio_bh_schedule_oneshot(bdrv_get_aio_context(bs), null_bh_cb, acb);
     }
     return &acb->common;
 }
diff --git a/block/qed.c b/block/qed.c
index 426f3cb447..3ee879b52e 100644
--- a/block/qed.c
+++ b/block/qed.c
@@ -909,7 +909,6 @@ static void qed_aio_complete_bh(void *opaque)
     void *user_opaque = acb->common.opaque;
     int ret = acb->bh_ret;
 
-    qemu_bh_delete(acb->bh);
     qemu_aio_unref(acb);
 
     /* Invoke callback */
@@ -934,9 +933,8 @@ static void qed_aio_complete(QEDAIOCB *acb, int ret)
 
     /* Arrange for a bh to invoke the completion function */
     acb->bh_ret = ret;
-    acb->bh = aio_bh_new(bdrv_get_aio_context(acb->common.bs),
-                         qed_aio_complete_bh, acb);
-    qemu_bh_schedule(acb->bh);
+    aio_bh_schedule_oneshot(bdrv_get_aio_context(acb->common.bs),
+                            qed_aio_complete_bh, acb);
 
     /* Start next allocating write request waiting behind this one.  Note that
      * requests enqueue themselves when they first hit an unallocated cluster
diff --git a/block/qed.h b/block/qed.h
index 22b3198751..9676ab9479 100644
--- a/block/qed.h
+++ b/block/qed.h
@@ -130,7 +130,6 @@ enum {
 
 typedef struct QEDAIOCB {
     BlockAIOCB common;
-    QEMUBH *bh;
     int bh_ret;                     /* final return status for completion bh */
     QSIMPLEQ_ENTRY(QEDAIOCB) next;  /* next request */
     int flags;                      /* QED_AIOCB_* bits ORed together */
diff --git a/block/rbd.c b/block/rbd.c
index 0106fea45f..6f9eb6fb9c 100644
--- a/block/rbd.c
+++ b/block/rbd.c
@@ -71,7 +71,6 @@ typedef enum {
 
 typedef struct RBDAIOCB {
     BlockAIOCB common;
-    QEMUBH *bh;
     int64_t ret;
     QEMUIOVector *qiov;
     char *bounce;
@@ -602,7 +601,6 @@ static const AIOCBInfo rbd_aiocb_info = {
 static void rbd_finish_bh(void *opaque)
 {
     RADOSCB *rcb = opaque;
-    qemu_bh_delete(rcb->acb->bh);
     qemu_rbd_complete_aio(rcb);
 }
 
@@ -621,9 +619,8 @@ static void rbd_finish_aiocb(rbd_completion_t c, RADOSCB *rcb)
     rcb->ret = rbd_aio_get_return_value(c);
     rbd_aio_release(c);
 
-    acb->bh = aio_bh_new(bdrv_get_aio_context(acb->common.bs),
-                         rbd_finish_bh, rcb);
-    qemu_bh_schedule(acb->bh);
+    aio_bh_schedule_oneshot(bdrv_get_aio_context(acb->common.bs),
+                            rbd_finish_bh, rcb);
 }
 
 static int rbd_aio_discard_wrapper(rbd_image_t image,
@@ -679,7 +676,6 @@ static BlockAIOCB *rbd_start_aio(BlockDriverState *bs,
     acb->ret = 0;
     acb->error = 0;
     acb->s = s;
-    acb->bh = NULL;
 
     if (cmd == RBD_AIO_WRITE) {
         qemu_iovec_to_buf(acb->qiov, 0, acb->bounce, qiov->size);
diff --git a/block/write-threshold.c b/block/write-threshold.c
index cc2ca71835..0bd1a01c86 100644
--- a/block/write-threshold.c
+++ b/block/write-threshold.c
@@ -76,8 +76,7 @@ static int coroutine_fn before_write_notify(NotifierWithReturn *notifier,
 static void write_threshold_register_notifier(BlockDriverState *bs)
 {
     bs->write_threshold_notifier.notify = before_write_notify;
-    notifier_with_return_list_add(&bs->before_write_notifiers,
-                                  &bs->write_threshold_notifier);
+    bdrv_add_before_write_notifier(bs, &bs->write_threshold_notifier);
 }
 
 static void write_threshold_update(BlockDriverState *bs,