Don't change total_size when creating fixed vpc

After upstream commit #3f3f20d... vpc: Fix size in fixed image creation. Total_sectors is rounded up to match the geometry, total_size is also changed as well (Total_sectors * BDRV_SECTOR_SIZE). This brings problem for bnc#943446 - qemu-img convert doesn't create MB aligned VHDs anymore. Add patch to exclude the change. Signed-off-by: Chunyan Liu <cyliu@suse.com> [BR: BSC#943446] Signed-off-by: Bruce Rogers <brogers@suse.com>
e1000: Avoid infinite loop in processing transmit descriptor (CVE-2015-6815)
2015-11-04 09:45:55 -07:00 · 2015-10-08 18:56:57 +02:00 · 2015-10-08 18:56:56 +02:00 · 2015-10-08 18:56:56 +02:00 · 2015-10-08 18:56:56 +02:00 · 2015-10-08 18:56:56 +02:00
2888 changed files with 98707 additions and 226307 deletions
--- a/.dir-locals.el
+++ b/.dir-locals.el
@@ -1,2 +0,0 @@
-((c-mode . ((c-file-style . "stroustrup")
-	    (indent-tabs-mode . nil))))
--- a/.gitignore
+++ b/.gitignore
@@ -17,15 +17,12 @@
 /trace/generated-tcg-tracers.h
 /trace/generated-ust-provider.h
 /trace/generated-ust.c
-/ui/shader/texture-blit-frag.h
-/ui/shader/texture-blit-vert.h
+/libcacard/trace/generated-tracers.c
 *-timestamp
 /*-softmmu
 /*-darwin-user
 /*-linux-user
 /*-bsd-user
-/ivshmem-client
-/ivshmem-server
 /libdis*
 /libuser
 /linux-headers/asm
@@ -35,7 +32,6 @@
 /qapi-visit.[ch]
 /qapi-event.[ch]
 /qmp-commands.h
-/qmp-introspect.[ch]
 /qmp-marshal.c
 /qemu-doc.html
 /qemu-tech.html
@@ -51,7 +47,6 @@
 /qemu-ga
 /qemu-bridge-helper
 /qemu-monitor.texi
-/qemu-monitor-info.texi
 /qmp-commands.txt
 /vscclient
 /fsdev/virtfs-proxy-helper
@@ -61,7 +56,6 @@
 *.cp
 *.dvi
 *.exe
-*.msi
 *.dll
 *.so
 *.mo
--- a/.travis.yml
+++ b/.travis.yml
@@ -1,38 +1,9 @@
-sudo: false
 language: c
 python:
  - "2.4"
 compiler:
  - gcc
  - clang
-cache: ccache
-addons:
-  apt:
-    packages:
-      - libaio-dev
-      - libattr1-dev
-      - libbrlapi-dev
-      - libcap-ng-dev
-      - libgnutls-dev
-      - libgtk-3-dev
-      - libiscsi-dev
-      - liblttng-ust-dev
-      - libncurses5-dev
-      - libnss3-dev
-      - libpixman-1-dev
-      - libpng12-dev
-      - librados-dev
-      - libsdl1.2-dev
-      - libseccomp-dev
-      - libspice-protocol-dev
-      - libspice-server-dev
-      - libssh2-1-dev
-      - liburcu-dev
-      - libusb-1.0-0-dev
-      - libvte-2.90-dev
-      - sparse
-      - uuid-dev
-
 notifications:
  irc:
    channels:
@@ -41,16 +12,29 @@ notifications:
    on_failure: always
 env:
  global:
-    - TEST_CMD="make check"
+    - TEST_CMD=""
    - EXTRA_CONFIG=""
+    # Development packages, EXTRA_PKGS saved for additional builds
+    - CORE_PKGS="libusb-1.0-0-dev libiscsi-dev librados-dev libncurses5-dev"
+    - NET_PKGS="libseccomp-dev libgnutls-dev libssh2-1-dev  libspice-server-dev libspice-protocol-dev libnss3-dev"
+    - GUI_PKGS="libgtk-3-dev libvte-2.90-dev libsdl1.2-dev libpng12-dev libpixman-1-dev"
+    - EXTRA_PKGS=""
  matrix:
    # Group major targets together with their linux-user counterparts
-    - TARGETS=alpha-softmmu,alpha-linux-user,cris-softmmu,cris-linux-user,m68k-softmmu,m68k-linux-user,microblaze-softmmu,microblazeel-softmmu,microblaze-linux-user,microblazeel-linux-user
+    - TARGETS=alpha-softmmu,alpha-linux-user
    - TARGETS=arm-softmmu,arm-linux-user,armeb-linux-user,aarch64-softmmu,aarch64-linux-user
+    - TARGETS=cris-softmmu,cris-linux-user
    - TARGETS=i386-softmmu,i386-linux-user,x86_64-softmmu,x86_64-linux-user
-    - TARGETS=mips-softmmu,mips64-softmmu,mips64el-softmmu,mipsel-softmmu,mips-linux-user,mips64-linux-user,mips64el-linux-user,mipsel-linux-user,mipsn32-linux-user,mipsn32el-linux-user
-    - TARGETS=or32-softmmu,or32-linux-user,ppc-softmmu,ppc64-softmmu,ppcemb-softmmu,ppc-linux-user,ppc64-linux-user,ppc64abi32-linux-user,ppc64le-linux-user
-    - TARGETS=s390x-softmmu,s390x-linux-user,sh4-softmmu,sh4eb-softmmu,sh4-linux-user,sh4eb-linux-user,sparc-softmmu,sparc64-softmmu,sparc-linux-user,sparc32plus-linux-user,sparc64-linux-user,unicore32-softmmu,unicore32-linux-user
+    - TARGETS=m68k-softmmu,m68k-linux-user
+    - TARGETS=microblaze-softmmu,microblazeel-softmmu,microblaze-linux-user,microblazeel-linux-user
+    - TARGETS=mips-softmmu,mips64-softmmu,mips64el-softmmu,mipsel-softmmu
+    - TARGETS=mips-linux-user,mips64-linux-user,mips64el-linux-user,mipsel-linux-user,mipsn32-linux-user,mipsn32el-linux-user
+    - TARGETS=or32-softmmu,or32-linux-user
+    - TARGETS=ppc-softmmu,ppc64-softmmu,ppcemb-softmmu,ppc-linux-user,ppc64-linux-user,ppc64abi32-linux-user,ppc64le-linux-user
+    - TARGETS=s390x-softmmu,s390x-linux-user
+    - TARGETS=sh4-softmmu,sh4eb-softmmu,sh4-linux-user sh4eb-linux-user
+    - TARGETS=sparc-softmmu,sparc64-softmmu,sparc-linux-user,sparc32plus-linux-user,sparc64-linux-user
+    - TARGETS=unicore32-softmmu,unicore32-linux-user
    # Group remaining softmmu only targets into one build
    - TARGETS=lm32-softmmu,moxie-softmmu,tricore-softmmu,xtensa-softmmu,xtensaeb-softmmu
 git:
@@ -59,6 +43,8 @@ git:
 before_install:
  - wget -O - http://people.linaro.org/~alex.bennee/qemu-submodule-git-seed.tar.xz | tar -xvJ
  - git submodule update --init --recursive
+  - sudo apt-get update -qq
+  - sudo apt-get install -qq ${CORE_PKGS} ${NET_PKGS} ${GUI_PKGS} ${EXTRA_PKGS}
 before_script:
  - ./configure --target-list=${TARGETS} --enable-debug-tcg ${EXTRA_CONFIG}
 script:
@@ -66,59 +52,52 @@ script:
 matrix:
  # We manually include a number of additional build for non-standard bits
  include:
+    # Make check target (we only do this once)
+    - env:
+        - TARGETS=alpha-softmmu,arm-softmmu,aarch64-softmmu,cris-softmmu,
+                  i386-softmmu,x86_64-softmmu,m68k-softmmu,microblaze-softmmu,
+                  microblazeel-softmmu,mips-softmmu,mips64-softmmu,
+                  mips64el-softmmu,mipsel-softmmu,or32-softmmu,ppc-softmmu,
+                  ppc64-softmmu,ppcemb-softmmu,s390x-softmmu,sh4-softmmu,
+                  sh4eb-softmmu,sparc-softmmu,sparc64-softmmu,
+                  unicore32-softmmu,unicore32-linux-user,
+                  lm32-softmmu,moxie-softmmu,tricore-softmmu,xtensa-softmmu,
+                  xtensaeb-softmmu
+          TEST_CMD="make check"
+      compiler: gcc
    # Debug related options
-    - env: TARGETS=x86_64-softmmu
+    - env: TARGETS=i386-softmmu,x86_64-softmmu
           EXTRA_CONFIG="--enable-debug"
      compiler: gcc
-    # We currently disable "make check"
-    - env: TARGETS=alpha-softmmu
+    - env: TARGETS=i386-softmmu,x86_64-softmmu
           EXTRA_CONFIG="--enable-debug --enable-tcg-interpreter"
-           TEST_CMD=""
      compiler: gcc
-    # Disable a few of the optional features
-    - env: TARGETS=x86_64-softmmu
-           EXTRA_CONFIG="--disable-linux-aio --disable-cap-ng --disable-attr --disable-brlapi --disable-uuid --disable-libusb"
+    # All the extra -dev packages
+    - env: TARGETS=i386-softmmu,x86_64-softmmu
+           EXTRA_PKGS="libaio-dev libcap-ng-dev libattr1-dev libbrlapi-dev uuid-dev libusb-1.0.0-dev"
      compiler: gcc
    # Currently configure doesn't force --disable-pie
-    - env: TARGETS=x86_64-softmmu
+    - env: TARGETS=i386-softmmu,x86_64-softmmu
           EXTRA_CONFIG="--enable-gprof --enable-gcov --disable-pie"
      compiler: gcc
-    # Sparse
-    - env: TARGETS=x86_64-softmmu
+    - env: TARGETS=i386-softmmu,x86_64-softmmu
+           EXTRA_PKGS="sparse"
           EXTRA_CONFIG="--enable-sparse"
      compiler: gcc
-    # Modules
-    - env: TARGETS=arm-softmmu,x86_64-softmmu
+    # All the trace backends (apart from dtrace)
+    - env: TARGETS=i386-softmmu,x86_64-softmmu
+           EXTRA_CONFIG="--enable-trace-backends=stderr"
+      compiler: gcc
+    - env: TARGETS=i386-softmmu,x86_64-softmmu
+           EXTRA_CONFIG="--enable-trace-backends=simple"
+      compiler: gcc
+    - env: TARGETS=i386-softmmu,x86_64-softmmu
+           EXTRA_CONFIG="--enable-trace-backends=ftrace"
+      compiler: gcc
+    - env: TARGETS=i386-softmmu,x86_64-softmmu
+          EXTRA_PKGS="liblttng-ust-dev liburcu-dev"
+          EXTRA_CONFIG="--enable-trace-backends=ust"
+      compiler: gcc
+    - env: TARGETS=i386-softmmu,x86_64-softmmu
           EXTRA_CONFIG="--enable-modules"
      compiler: gcc
-    # All the trace backends (apart from dtrace)
-    - env: TARGETS=i386-softmmu
-           EXTRA_CONFIG="--enable-trace-backends=log"
-      compiler: gcc
-    # We currently disable "make check" (until 41fc57e44ed regression fixed)
-    - env: TARGETS=x86_64-softmmu
-           EXTRA_CONFIG="--enable-trace-backends=simple"
-           TEST_CMD=""
-      compiler: gcc
-    # We currently disable "make check"
-    - env: TARGETS=x86_64-softmmu
-           EXTRA_CONFIG="--enable-trace-backends=ftrace"
-           TEST_CMD=""
-      compiler: gcc
-    # We currently disable "make check"
-    - env: TARGETS=x86_64-softmmu
-           EXTRA_CONFIG="--enable-trace-backends=ust"
-           TEST_CMD=""
-      compiler: gcc
-    # All the co-routine backends (apart from windows)
-    # We currently disable "make check"
-    - env: TARGETS=x86_64-softmmu
-           EXTRA_CONFIG="--with-coroutine=gthread"
-           TEST_CMD=""
-      compiler: gcc
-    - env: TARGETS=x86_64-softmmu
-           EXTRA_CONFIG="--with-coroutine=ucontext"
-      compiler: gcc
-    - env: TARGETS=x86_64-softmmu
-           EXTRA_CONFIG="--with-coroutine=sigaltstack"
-      compiler: gcc
--- a/13
+++ b/13
@@ -87,15 +87,10 @@ Furthermore, it is the QEMU coding style.

 5. Declarations

-Mixed declarations (interleaving statements and declarations within
-blocks) are generally not allowed; declarations should be at the beginning
-of blocks.
-
-Every now and then, an exception is made for declarations inside a
-#ifdef or #ifndef block: if the code looks nicer, such declarations can
-be placed at the top of the block even if there are statements above.
-On the other hand, however, it's often best to move that #ifdef/#ifndef
-block to a separate function altogether.
+Mixed declarations (interleaving statements and declarations within blocks)
+are not allowed; declarations should be at the beginning of blocks.  In other
+words, the code should not generate warnings if using GCC's
+-Wdeclaration-after-statement option.

 6. Conditional statements

--- a/55
+++ b/55
@@ -157,58 +157,3 @@ painful. These are:
 * you may assume that integers are 2s complement representation
 * you may assume that right shift of a signed integer duplicates
   the sign bit (ie it is an arithmetic shift, not a logical shift)
-
-7. Error handling and reporting
-
-7.1 Reporting errors to the human user
-
-Do not use printf(), fprintf() or monitor_printf().  Instead, use
-error_report() or error_vreport() from error-report.h.  This ensures the
-error is reported in the right place (current monitor or stderr), and in
-a uniform format.
-
-Use error_printf() & friends to print additional information.
-
-error_report() prints the current location.  In certain common cases
-like command line parsing, the current location is tracked
-automatically.  To manipulate it manually, use the loc_*() from
-error-report.h.
-
-7.2 Propagating errors
-
-An error can't always be reported to the user right where it's detected,
-but often needs to be propagated up the call chain to a place that can
-handle it.  This can be done in various ways.
-
-The most flexible one is Error objects.  See error.h for usage
-information.
-
-Use the simplest suitable method to communicate success / failure to
-callers.  Stick to common methods: non-negative on success / -1 on
-error, non-negative / -errno, non-null / null, or Error objects.
-
-Example: when a function returns a non-null pointer on success, and it
-can fail only in one way (as far as the caller is concerned), returning
-null on failure is just fine, and certainly simpler and a lot easier on
-the eyes than propagating an Error object through an Error ** parameter.
-
-Example: when a function's callers need to report details on failure
-only the function really knows, use Error **, and set suitable errors.
-
-Do not report an error to the user when you're also returning an error
-for somebody else to handle.  Leave the reporting to the place that
-consumes the error returned.
-
-7.3 Handling errors
-
-Calling exit() is fine when handling configuration errors during
-startup.  It's problematic during normal operation.  In particular,
-monitor commands should never exit().
-
-Do not call exit() or abort() to handle an error that can be triggered
-by the guest (e.g., some unimplemented corner case in guest code
-translation or device emulation).  Guests should not be able to
-terminate QEMU.
-
-Note that &error_fatal is just another way to exit(1), and &error_abort
-is just another way to abort().
--- a/557
+++ b/557
--- a/152
+++ b/152
@@ -3,11 +3,6 @@
 # Always point to the root of the build tree (needs GNU make).
 BUILD_DIR=$(CURDIR)

-# Before including a proper config-host.mak, assume we are in the source tree
-SRC_PATH=.
-
-UNCHECKED_GOALS := %clean TAGS cscope ctags
-
 # All following code might depend on configuration variables
 ifneq ($(wildcard config-host.mak),)
 # Put the all: rule here so that config-host.mak can contain dependencies.
@@ -43,7 +38,7 @@ config-host.mak: $(SRC_PATH)/configure
 	fi
 else
 config-host.mak:
-ifneq ($(filter-out $(UNCHECKED_GOALS),$(MAKECMDGOALS)),$(if $(MAKECMDGOALS),,fail))
+ifneq ($(filter-out %clean,$(MAKECMDGOALS)),$(if $(MAKECMDGOALS),,fail))
 	@echo "Please call configure before running make!"
 	@exit 1
 endif
@@ -52,8 +47,6 @@ endif
 GENERATED_HEADERS = config-host.h qemu-options.def
 GENERATED_HEADERS += qmp-commands.h qapi-types.h qapi-visit.h qapi-event.h
 GENERATED_SOURCES += qmp-marshal.c qapi-types.c qapi-visit.c qapi-event.c
-GENERATED_HEADERS += qmp-introspect.h
-GENERATED_SOURCES += qmp-introspect.c

 GENERATED_HEADERS += trace/generated-events.h
 GENERATED_SOURCES += trace/generated-events.c
@@ -81,7 +74,7 @@ Makefile: ;
 configure: ;

 .PHONY: all clean cscope distclean dvi html info install install-doc \
-	pdf recurse-all speed test dist msi
+	pdf recurse-all speed test dist

 $(call set-vpath, $(SRC_PATH))

@@ -90,8 +83,7 @@ LIBS+=-lz $(LIBS_TOOLS)
 HELPERS-$(CONFIG_LINUX) = qemu-bridge-helper$(EXESUF)

 ifdef BUILD_DOCS
-DOCS=qemu-doc.html qemu-tech.html qemu.1 qemu-img.1 qemu-nbd.8 qemu-ga.8
-DOCS+=qmp-commands.txt
+DOCS=qemu-doc.html qemu-tech.html qemu.1 qemu-img.1 qemu-nbd.8 qmp-commands.txt
 ifdef CONFIG_LINUX
 DOCS+=kvm_stat.1
 endif
@@ -138,7 +130,7 @@ endif
 	 else \
 	  mv $@.tmp $@; \
 	  cp -p $@ $@.old; \
-	 fi, "  GEN   $@");
+	 fi, "  GEN  $@");

 defconfig:
 	rm -f config-all-devices.mak $(SUBDIR_DEVICES_MAK)
@@ -151,21 +143,18 @@ dummy := $(call unnest-vars,, \
                stub-obj-y \
                util-obj-y \
                qga-obj-y \
-                ivshmem-client-obj-y \
-                ivshmem-server-obj-y \
                qga-vss-dll-obj-y \
                block-obj-y \
                block-obj-m \
-                crypto-obj-y \
-                crypto-aes-obj-y \
-                qom-obj-y \
-                io-obj-y \
                common-obj-y \
                common-obj-m)

 ifneq ($(wildcard config-host.mak),)
 include $(SRC_PATH)/tests/Makefile
 endif
+ifeq ($(CONFIG_SMARTCARD_NSS),y)
+include $(SRC_PATH)/libcacard/Makefile
+endif

 all: $(DOCS) $(TOOLS) $(HELPERS-y) recurse-all modules

@@ -178,8 +167,6 @@ SUBDIR_RULES=$(patsubst %,subdir-%, $(TARGET_DIRS))
 SOFTMMU_SUBDIR_RULES=$(filter %-softmmu,$(SUBDIR_RULES))

 $(SOFTMMU_SUBDIR_RULES): $(block-obj-y)
-$(SOFTMMU_SUBDIR_RULES): $(crypto-obj-y)
-$(SOFTMMU_SUBDIR_RULES): $(io-obj-y)
 $(SOFTMMU_SUBDIR_RULES): config-all-devices.mak

 subdir-%:
@@ -204,7 +191,7 @@ subdir-dtc:dtc/libfdt dtc/tests
 dtc/%:
 	mkdir -p $@

-$(SUBDIR_RULES): libqemuutil.a libqemustub.a $(common-obj-y) $(qom-obj-y) $(crypto-aes-obj-$(CONFIG_USER_ONLY))
+$(SUBDIR_RULES): libqemuutil.a libqemustub.a $(common-obj-y)

 ROMSUBDIR_RULES=$(patsubst %,romsubdir-%, $(ROMS))
 romsubdir-%:
@@ -234,13 +221,13 @@ util/module.o-cflags = -D'CONFIG_BLOCK_MODULES=$(block-modules)'

 qemu-img.o: qemu-img-cmds.h

-qemu-img$(EXESUF): qemu-img.o $(block-obj-y) $(crypto-obj-y) $(io-obj-y) $(qom-obj-y) libqemuutil.a libqemustub.a
-qemu-nbd$(EXESUF): qemu-nbd.o $(block-obj-y) $(crypto-obj-y) $(io-obj-y) $(qom-obj-y) libqemuutil.a libqemustub.a
-qemu-io$(EXESUF): qemu-io.o $(block-obj-y) $(crypto-obj-y) $(io-obj-y) $(qom-obj-y) libqemuutil.a libqemustub.a
+qemu-img$(EXESUF): qemu-img.o $(block-obj-y) libqemuutil.a libqemustub.a
+qemu-nbd$(EXESUF): qemu-nbd.o $(block-obj-y) libqemuutil.a libqemustub.a
+qemu-io$(EXESUF): qemu-io.o $(block-obj-y) libqemuutil.a libqemustub.a

 qemu-bridge-helper$(EXESUF): qemu-bridge-helper.o

-fsdev/virtfs-proxy-helper$(EXESUF): fsdev/virtfs-proxy-helper.o fsdev/9p-marshal.o fsdev/9p-iov-marshal.o libqemuutil.a libqemustub.a
+fsdev/virtfs-proxy-helper$(EXESUF): fsdev/virtfs-proxy-helper.o fsdev/virtio-9p-marshal.o libqemuutil.a libqemustub.a
 fsdev/virtfs-proxy-helper$(EXESUF): LIBS += -lcap

 qemu-img-cmds.h: $(SRC_PATH)/qemu-img-cmds.hx
@@ -256,49 +243,42 @@ qapi-py = $(SRC_PATH)/scripts/qapi.py $(SRC_PATH)/scripts/ordereddict.py
 qga/qapi-generated/qga-qapi-types.c qga/qapi-generated/qga-qapi-types.h :\
 $(SRC_PATH)/qga/qapi-schema.json $(SRC_PATH)/scripts/qapi-types.py $(qapi-py)
 	$(call quiet-command,$(PYTHON) $(SRC_PATH)/scripts/qapi-types.py \
-		$(gen-out-type) -o qga/qapi-generated -p "qga-" $<, \
+		$(gen-out-type) -o qga/qapi-generated -p "qga-" -i $<, \
 		"  GEN   $@")
 qga/qapi-generated/qga-qapi-visit.c qga/qapi-generated/qga-qapi-visit.h :\
 $(SRC_PATH)/qga/qapi-schema.json $(SRC_PATH)/scripts/qapi-visit.py $(qapi-py)
 	$(call quiet-command,$(PYTHON) $(SRC_PATH)/scripts/qapi-visit.py \
-		$(gen-out-type) -o qga/qapi-generated -p "qga-" $<, \
+		$(gen-out-type) -o qga/qapi-generated -p "qga-" -i $<, \
 		"  GEN   $@")
 qga/qapi-generated/qga-qmp-commands.h qga/qapi-generated/qga-qmp-marshal.c :\
 $(SRC_PATH)/qga/qapi-schema.json $(SRC_PATH)/scripts/qapi-commands.py $(qapi-py)
 	$(call quiet-command,$(PYTHON) $(SRC_PATH)/scripts/qapi-commands.py \
-		$(gen-out-type) -o qga/qapi-generated -p "qga-" $<, \
+		$(gen-out-type) -o qga/qapi-generated -p "qga-" -i $<, \
 		"  GEN   $@")

 qapi-modules = $(SRC_PATH)/qapi-schema.json $(SRC_PATH)/qapi/common.json \
               $(SRC_PATH)/qapi/block.json $(SRC_PATH)/qapi/block-core.json \
-               $(SRC_PATH)/qapi/event.json $(SRC_PATH)/qapi/introspect.json \
-               $(SRC_PATH)/qapi/crypto.json $(SRC_PATH)/qapi/rocker.json \
-               $(SRC_PATH)/qapi/trace.json
+               $(SRC_PATH)/qapi/event.json

 qapi-types.c qapi-types.h :\
 $(qapi-modules) $(SRC_PATH)/scripts/qapi-types.py $(qapi-py)
 	$(call quiet-command,$(PYTHON) $(SRC_PATH)/scripts/qapi-types.py \
-		$(gen-out-type) -o "." -b $<, \
+		$(gen-out-type) -o "." -b -i $<, \
 		"  GEN   $@")
 qapi-visit.c qapi-visit.h :\
 $(qapi-modules) $(SRC_PATH)/scripts/qapi-visit.py $(qapi-py)
 	$(call quiet-command,$(PYTHON) $(SRC_PATH)/scripts/qapi-visit.py \
-		$(gen-out-type) -o "." -b $<, \
+		$(gen-out-type) -o "." -b -i $<, \
 		"  GEN   $@")
 qapi-event.c qapi-event.h :\
 $(qapi-modules) $(SRC_PATH)/scripts/qapi-event.py $(qapi-py)
 	$(call quiet-command,$(PYTHON) $(SRC_PATH)/scripts/qapi-event.py \
-		$(gen-out-type) -o "." $<, \
+		$(gen-out-type) -o "." -b -i $<, \
 		"  GEN   $@")
 qmp-commands.h qmp-marshal.c :\
 $(qapi-modules) $(SRC_PATH)/scripts/qapi-commands.py $(qapi-py)
 	$(call quiet-command,$(PYTHON) $(SRC_PATH)/scripts/qapi-commands.py \
-		$(gen-out-type) -o "." -m $<, \
-		"  GEN   $@")
-qmp-introspect.h qmp-introspect.c :\
-$(qapi-modules) $(SRC_PATH)/scripts/qapi-introspect.py $(qapi-py)
-	$(call quiet-command,$(PYTHON) $(SRC_PATH)/scripts/qapi-introspect.py \
-		$(gen-out-type) -o "." $<, \
+		$(gen-out-type) -o "." -m -i $<, \
 		"  GEN   $@")

 QGALIB_GEN=$(addprefix qga/qapi-generated/, qga-qapi-types.h qga-qapi-visit.h qga-qmp-commands.h)
@@ -307,44 +287,15 @@ $(qga-obj-y) qemu-ga.o: $(QGALIB_GEN)
 qemu-ga$(EXESUF): $(qga-obj-y) libqemuutil.a libqemustub.a
 	$(call LINK, $^)

-ifdef QEMU_GA_MSI_ENABLED
-QEMU_GA_MSI=qemu-ga-$(ARCH).msi
-
-msi: $(QEMU_GA_MSI)
-
-$(QEMU_GA_MSI): qemu-ga.exe $(QGA_VSS_PROVIDER)
-
-$(QEMU_GA_MSI): config-host.mak
-
-$(QEMU_GA_MSI):  $(SRC_PATH)/qga/installer/qemu-ga.wxs
-	$(call quiet-command,QEMU_GA_VERSION="$(QEMU_GA_VERSION)" QEMU_GA_MANUFACTURER="$(QEMU_GA_MANUFACTURER)" QEMU_GA_DISTRO="$(QEMU_GA_DISTRO)" BUILD_DIR="$(BUILD_DIR)" \
-	wixl -o $@ $(QEMU_GA_MSI_ARCH) $(QEMU_GA_MSI_WITH_VSS) $(QEMU_GA_MSI_MINGW_DLL_PATH) $<, "  WIXL  $@")
-else
-msi:
-	@echo "MSI build not configured or dependency resolution failed (reconfigure with --enable-guest-agent-msi option)"
-endif
-
-ifneq ($(EXESUF),)
-.PHONY: qemu-ga
-qemu-ga: qemu-ga$(EXESUF) $(QGA_VSS_PROVIDER) $(QEMU_GA_MSI)
-endif
-
-ivshmem-client$(EXESUF): $(ivshmem-client-obj-y)
-	$(call LINK, $^)
-ivshmem-server$(EXESUF): $(ivshmem-server-obj-y) libqemuutil.a libqemustub.a
-	$(call LINK, $^)
-
 clean:
 # avoid old build problems by removing potentially incorrect old files
 	rm -f config.mak op-i386.h opc-i386.h gen-op-i386.h op-arm.h opc-arm.h gen-op-arm.h
 	rm -f qemu-options.def
-	rm -f *.msi
 	find . \( -name '*.l[oa]' -o -name '*.so' -o -name '*.dll' -o -name '*.mo' -o -name '*.[oda]' \) -type f -exec rm {} +
 	rm -f $(filter-out %.tlb,$(TOOLS)) $(HELPERS-y) qemu-ga TAGS cscope.* *.pod *~ */*~
 	rm -f fsdev/*.pod
 	rm -rf .libs */.libs
 	rm -f qemu-img-cmds.h
-	rm -f ui/shader/*-vert.h ui/shader/*-frag.h
 	@# May not be present in GENERATED_HEADERS
 	rm -f trace/generated-tracers-dtrace.dtrace*
 	rm -f trace/generated-tracers-dtrace.h*
@@ -365,7 +316,7 @@ qemu-%.tar.bz2:
 	$(SRC_PATH)/scripts/make-release "$(SRC_PATH)" "$(patsubst qemu-%.tar.bz2,%,$@)"

 distclean: clean
-	rm -f config-host.mak config-host.h* config-host.ld $(DOCS) qemu-options.texi qemu-img-cmds.texi qemu-monitor.texi qemu-monitor-info.texi
+	rm -f config-host.mak config-host.h* config-host.ld $(DOCS) qemu-options.texi qemu-img-cmds.texi qemu-monitor.texi
 	rm -f config-all-devices.mak config-all-disas.mak config.status
 	rm -f po/*.mo tests/qemu-iotests/common.env
 	rm -f roms/seabios/config.mak roms/vgabios/config.mak
@@ -390,8 +341,8 @@ bepo    cz

 ifdef INSTALL_BLOBS
 BLOBS=bios.bin bios-256k.bin sgabios.bin vgabios.bin vgabios-cirrus.bin \
-vgabios-stdvga.bin vgabios-vmware.bin vgabios-qxl.bin vgabios-virtio.bin \
-acpi-dsdt.aml \
+vgabios-stdvga.bin vgabios-vmware.bin vgabios-qxl.bin \
+acpi-dsdt.aml q35-acpi-dsdt.aml \
 ppc_rom.bin openbios-sparc32 openbios-sparc64 openbios-ppc QEMU,tcx.bin QEMU,cgthree.bin \
 pxe-e1000.rom pxe-eepro100.rom pxe-ne2k_pci.rom \
 pxe-pcnet.rom pxe-rtl8139.rom pxe-virtio.rom \
@@ -400,6 +351,7 @@ efi-pcnet.rom efi-rtl8139.rom efi-virtio.rom \
 qemu-icon.bmp qemu_logo_no_text.svg \
 bamboo.dtb petalogix-s3adsp1800.dtb petalogix-ml605.dtb \
 multiboot.bin linuxboot.bin kvmvapic.bin \
+s390-zipl.rom \
 s390-ccw.img \
 spapr-rtas.bin slof.bin \
 palcode-clipper \
@@ -420,9 +372,6 @@ ifneq ($(TOOLS),)
 	$(INSTALL_DIR) "$(DESTDIR)$(mandir)/man8"
 	$(INSTALL_DATA) qemu-nbd.8 "$(DESTDIR)$(mandir)/man8"
 endif
-ifneq (,$(findstring qemu-ga,$(TOOLS)))
-	$(INSTALL_DATA) qemu-ga.8 "$(DESTDIR)$(mandir)/man8"
-endif
 endif
 ifdef CONFIG_VIRTFS
 	$(INSTALL_DIR) "$(DESTDIR)$(mandir)/man1"
@@ -439,11 +388,16 @@ ifneq (,$(findstring qemu-ga,$(TOOLS)))
 endif
 endif

+install-confdir:
+	$(INSTALL_DIR) "$(DESTDIR)$(qemu_confdir)"

-install: all $(if $(BUILD_DOCS),install-doc) \
+install-sysconfig: install-datadir install-confdir
+	$(INSTALL_DATA) $(SRC_PATH)/sysconfigs/target/target-x86_64.conf "$(DESTDIR)$(qemu_confdir)"
+
+install: all $(if $(BUILD_DOCS),install-doc) install-sysconfig \
 install-datadir install-localstatedir
 ifneq ($(TOOLS),)
-	$(call install-prog,$(subst qemu-ga,qemu-ga$(EXESUF),$(TOOLS)),$(DESTDIR)$(bindir))
+	$(call install-prog,$(TOOLS),$(DESTDIR)$(bindir))
 endif
 ifneq ($(CONFIG_MODULES),)
 	$(INSTALL_DIR) "$(DESTDIR)$(qemu_moddir)"
@@ -477,36 +431,15 @@ endif
 test speed: all
 	$(MAKE) -C tests/tcg $@

-.PHONY: ctags
-ctags:
-	rm -f $@
-	find "$(SRC_PATH)" -name '*.[hc]' -exec ctags --append {} +
-
 .PHONY: TAGS
 TAGS:
 	rm -f $@
 	find "$(SRC_PATH)" -name '*.[hc]' -exec etags --append {} +

 cscope:
-	rm -f "$(SRC_PATH)"/cscope.*
-	find "$(SRC_PATH)/" -name "*.[chsS]" -print | sed 's,^\./,,' > "$(SRC_PATH)/cscope.files"
-	cscope -b -i"$(SRC_PATH)/cscope.files"
-
-# opengl shader programs
-ui/shader/%-vert.h: $(SRC_PATH)/ui/shader/%.vert $(SRC_PATH)/scripts/shaderinclude.pl
-	@mkdir -p $(dir $@)
-	$(call quiet-command,\
-		perl $(SRC_PATH)/scripts/shaderinclude.pl $< > $@,\
-		"  VERT  $@")
-
-ui/shader/%-frag.h: $(SRC_PATH)/ui/shader/%.frag $(SRC_PATH)/scripts/shaderinclude.pl
-	@mkdir -p $(dir $@)
-	$(call quiet-command,\
-		perl $(SRC_PATH)/scripts/shaderinclude.pl $< > $@,\
-		"  FRAG  $@")
-
-ui/console-gl.o: $(SRC_PATH)/ui/console-gl.c \
-	ui/shader/texture-blit-vert.h ui/shader/texture-blit-frag.h
+	rm -f ./cscope.*
+	find "$(SRC_PATH)" -name "*.[chsS]" -print | sed 's,^\./,,' > ./cscope.files
+	cscope -b

 # documentation
 MAKEINFO=makeinfo
@@ -531,16 +464,13 @@ qemu-options.texi: $(SRC_PATH)/qemu-options.hx
 qemu-monitor.texi: $(SRC_PATH)/hmp-commands.hx
 	$(call quiet-command,sh $(SRC_PATH)/scripts/hxtool -t < $< > $@,"  GEN   $@")

-qemu-monitor-info.texi: $(SRC_PATH)/hmp-commands-info.hx
-	$(call quiet-command,sh $(SRC_PATH)/scripts/hxtool -t < $< > $@,"  GEN   $@")
-
 qmp-commands.txt: $(SRC_PATH)/qmp-commands.hx
 	$(call quiet-command,sh $(SRC_PATH)/scripts/hxtool -q < $< > $@,"  GEN   $@")

 qemu-img-cmds.texi: $(SRC_PATH)/qemu-img-cmds.hx
 	$(call quiet-command,sh $(SRC_PATH)/scripts/hxtool -t < $< > $@,"  GEN   $@")

-qemu.1: qemu-doc.texi qemu-options.texi qemu-monitor.texi qemu-monitor-info.texi
+qemu.1: qemu-doc.texi qemu-options.texi qemu-monitor.texi
 	$(call quiet-command, \
 	  perl -Ww -- $(SRC_PATH)/scripts/texi2pod.pl $< qemu.pod && \
 	  $(POD2MAN) --section=1 --center=" " --release=" " qemu.pod > $@, \
@@ -564,12 +494,6 @@ qemu-nbd.8: qemu-nbd.texi
 	  $(POD2MAN) --section=8 --center=" " --release=" " qemu-nbd.pod > $@, \
 	  "  GEN   $@")

-qemu-ga.8: qemu-ga.texi
-	$(call quiet-command, \
-	  perl -Ww -- $(SRC_PATH)/scripts/texi2pod.pl $< qemu-ga.pod && \
-	  $(POD2MAN) --section=8 --center=" " --release=" " qemu-ga.pod > $@, \
-	  "  GEN   $@")
-
 kvm_stat.1: scripts/kvm/kvm_stat.texi
 	$(call quiet-command, \
 	  perl -Ww -- $(SRC_PATH)/scripts/texi2pod.pl $< kvm_stat.pod && \
@@ -583,8 +507,7 @@ pdf: qemu-doc.pdf qemu-tech.pdf

 qemu-doc.dvi qemu-doc.html qemu-doc.info qemu-doc.pdf: \
 	qemu-img.texi qemu-nbd.texi qemu-options.texi \
-	qemu-monitor.texi qemu-img-cmds.texi qemu-ga.texi \
-	qemu-monitor-info.texi
+	qemu-monitor.texi qemu-img-cmds.texi

 ifdef CONFIG_WIN32

@@ -634,7 +557,6 @@ endif # SIGNCODE
                $(if $(DLL_PATH),-DDLLDIR="$(DLL_PATH)") \
                -DSRCDIR="$(SRC_PATH)" \
                -DOUTFILE="$(INSTALLER)" \
-                -DDISPLAYVERSION="$(VERSION)" \
                $(SRC_PATH)/qemu.nsi
 	rm -r ${INSTDIR}
 ifdef SIGNCODE
@@ -644,7 +566,7 @@ endif # CONFIG_WIN

 # Add a dependency on the generated files, so that they are always
 # rebuilt before other object files
-ifneq ($(filter-out $(UNCHECKED_GOALS),$(MAKECMDGOALS)),$(if $(MAKECMDGOALS),,fail))
+ifneq ($(filter-out %clean,$(MAKECMDGOALS)),$(if $(MAKECMDGOALS),,fail))
 Makefile: $(GENERATED_HEADERS)
 endif

--- a/Makefile.objs
+++ b/Makefile.objs
@@ -1,38 +1,37 @@
 #######################################################################
 # Common libraries for tools and emulators
 stub-obj-y = stubs/
-util-obj-y = util/ qobject/ qapi/
-util-obj-y += qmp-introspect.o qapi-types.o qapi-visit.o qapi-event.o
+util-obj-y = util/ qobject/ qapi/ qapi-types.o qapi-visit.o qapi-event.o

 #######################################################################
 # block-obj-y is code used by both qemu system emulation and qemu-img

 block-obj-y = async.o thread-pool.o
-block-obj-y += nbd/
-block-obj-y += block.o blockjob.o
+block-obj-y += nbd.o block.o blockjob.o
 block-obj-y += main-loop.o iohandler.o qemu-timer.o
 block-obj-$(CONFIG_POSIX) += aio-posix.o
 block-obj-$(CONFIG_WIN32) += aio-win32.o
 block-obj-y += block/
 block-obj-y += qemu-io-cmds.o

+block-obj-y += qemu-coroutine.o qemu-coroutine-lock.o qemu-coroutine-io.o
+block-obj-y += qemu-coroutine-sleep.o
+block-obj-y += coroutine-$(CONFIG_COROUTINE_BACKEND).o
+
 block-obj-m = block/

-#######################################################################
-# crypto-obj-y is code used by both qemu system emulation and qemu-img

-crypto-obj-y = crypto/
-crypto-aes-obj-y = crypto/
+######################################################################
+# smartcard

-#######################################################################
-# qom-obj-y is code used by both qemu system emulation and qemu-img
-
-qom-obj-y = qom/
-
-#######################################################################
-# io-obj-y is code used by both qemu system emulation and qemu-img
-
-io-obj-y = io/
+libcacard-y += libcacard/cac.o libcacard/event.o
+libcacard-y += libcacard/vcard.o libcacard/vreader.o
+libcacard-y += libcacard/vcard_emul_nss.o
+libcacard-y += libcacard/vcard_emul_type.o
+libcacard-y += libcacard/card_7816.o
+libcacard-y += libcacard/vcardt.o
+libcacard/vcard_emul_nss.o-cflags := $(NSS_CFLAGS)
+libcacard/vcard_emul_nss.o-libs := $(NSS_LIBS)

 ######################################################################
 # Target independent part of system emulation. The long term path is to
@@ -60,8 +59,6 @@ common-obj-y += audio/
 common-obj-y += hw/
 common-obj-y += accel.o

-common-obj-y += replay/
-
 common-obj-y += ui/
 common-obj-y += bt-host.o bt-vhci.o
 bt-host.o-cflags := $(BLUEZ_CFLAGS)
@@ -77,18 +74,18 @@ common-obj-y += backends/

 common-obj-$(CONFIG_SECCOMP) += qemu-seccomp.o

-common-obj-$(CONFIG_FDT) += device_tree.o
+common-obj-$(CONFIG_SMARTCARD_NSS) += $(libcacard-y)

 ######################################################################
 # qapi

 common-obj-y += qmp-marshal.o
-common-obj-y += qmp-introspect.o
 common-obj-y += qmp.o hmp.o
 endif

 #######################################################################
 # Target-independent parts used in system and user emulation
+common-obj-y += qemu-log.o
 common-obj-y += tcg-runtime.o
 common-obj-y += hw/
 common-obj-y += qom/
@@ -111,8 +108,3 @@ target-obj-y += trace/
 # by libqemuutil.a.  These should be moved to a separate .json schema.
 qga-obj-y = qga/
 qga-vss-dll-obj-y = qga/
-
-######################################################################
-# contrib
-ivshmem-client-obj-y = contrib/ivshmem-client/
-ivshmem-server-obj-y = contrib/ivshmem-server/
--- a/Makefile.target
+++ b/Makefile.target
@@ -1,13 +1,11 @@
 # -*- Mode: makefile -*-

-BUILD_DIR?=$(CURDIR)/..
-
 include ../config-host.mak
 include config-target.mak
 include config-devices.mak
 include $(SRC_PATH)/rules.mak

-$(call set-vpath, $(SRC_PATH):$(BUILD_DIR))
+$(call set-vpath, $(SRC_PATH))
 ifdef CONFIG_LINUX
 QEMU_CFLAGS += -I../linux-headers
 endif
@@ -36,6 +34,10 @@ endif
 PROGS=$(QEMU_PROG) $(QEMU_PROGW)
 STPFILES=

+ifdef CONFIG_LINUX_USER
+PROGS+=$(QEMU_PROG)-binfmt
+endif
+
 config-target.h: config-target.h-timestamp
 config-target.h-timestamp: config-target.mak

@@ -85,11 +87,8 @@ all: $(PROGS) stap
 #########################################################
 # cpu emulator library
 obj-y = exec.o translate-all.o cpu-exec.o
-obj-y += translate-common.o
-obj-y += cpu-exec-common.o
 obj-y += tcg/tcg.o tcg/tcg-op.o tcg/optimize.o
 obj-$(CONFIG_TCG_INTERPRETER) += tci.o
-obj-y += tcg/tcg-common.o
 obj-$(CONFIG_TCG_INTERPRETER) += disas/tci.o
 obj-y += fpu/softfloat.o
 obj-y += target-$(TARGET_BASE_ARCH)/
@@ -113,6 +112,8 @@ QEMU_CFLAGS+=-I$(SRC_PATH)/linux-user/$(TARGET_ABI_DIR) -I$(SRC_PATH)/linux-user
 obj-y += linux-user/
 obj-y += gdbstub.o thunk.o user-exec.o

+obj-binfmt-y += linux-user/
+
 endif #CONFIG_LINUX_USER

 #########################################################
@@ -134,12 +135,12 @@ ifdef CONFIG_SOFTMMU
 obj-y += arch_init.o cpus.o monitor.o gdbstub.o balloon.o ioport.o numa.o
 obj-y += qtest.o bootdevice.o
 obj-y += hw/
+obj-$(CONFIG_FDT) += device_tree.o
 obj-$(CONFIG_KVM) += kvm-all.o
-obj-y += memory.o cputlb.o
+obj-y += memory.o savevm.o cputlb.o
 obj-y += memory_mapping.o
 obj-y += dump.o
-obj-y += migration/ram.o migration/savevm.o
-LIBS := $(libs_softmmu) $(LIBS)
+LIBS+=$(libs_softmmu)

 # xen support
 obj-$(CONFIG_XEN) += xen-common.o
@@ -154,14 +155,18 @@ else
 obj-y += hw/$(TARGET_BASE_ARCH)/
 endif

-GENERATED_HEADERS += hmp-commands.h hmp-commands-info.h qmp-commands-old.h
+GENERATED_HEADERS += hmp-commands.h qmp-commands-old.h

 endif # CONFIG_SOFTMMU

 # Workaround for http://gcc.gnu.org/PR55489, see configure.
 %/translate.o: QEMU_CFLAGS += $(TRANSLATE_OPT_CFLAGS)

+ifdef CONFIG_LINUX_USER
+dummy := $(call unnest-vars,,obj-y obj-binfmt-y)
+else
 dummy := $(call unnest-vars,,obj-y)
+endif
 all-obj-y := $(obj-y)

 target-obj-y :=
@@ -173,30 +178,21 @@ target-obj-y-save := $(target-obj-y)
 dummy := $(call unnest-vars,.., \
               block-obj-y \
               block-obj-m \
-               crypto-obj-y \
-               crypto-aes-obj-y \
-               qom-obj-y \
-               io-obj-y \
               common-obj-y \
               common-obj-m)
 target-obj-y := $(target-obj-y-save)
 all-obj-y += $(common-obj-y)
 all-obj-y += $(target-obj-y)
-all-obj-y += $(qom-obj-y)
 all-obj-$(CONFIG_SOFTMMU) += $(block-obj-y)
-all-obj-$(CONFIG_USER_ONLY) += $(crypto-aes-obj-y)
-all-obj-$(CONFIG_SOFTMMU) += $(crypto-obj-y)
-all-obj-$(CONFIG_SOFTMMU) += $(io-obj-y)

 $(QEMU_PROG_BUILD): config-devices.mak

 # build either PROG or PROGW
 $(QEMU_PROG_BUILD): $(all-obj-y) ../libqemuutil.a ../libqemustub.a
 	$(call LINK, $(filter-out %.mak, $^))
-ifdef CONFIG_DARWIN
-	$(call quiet-command,Rez -append $(SRC_PATH)/pc-bios/qemu.rsrc -o $@,"  REZ   $(TARGET_DIR)$@")
-	$(call quiet-command,SetFile -a C $@,"  SETFILE $(TARGET_DIR)$@")
-endif
+
+$(QEMU_PROG)-binfmt: $(obj-binfmt-y)
+	$(call LINK,$^)

 gdbstub-xml.c: $(TARGET_XML_FILES) $(SRC_PATH)/scripts/feature_to_c.sh
 	$(call quiet-command,rm -f $@ && $(SHELL) $(SRC_PATH)/scripts/feature_to_c.sh $@ $(TARGET_XML_FILES),"  GEN   $(TARGET_DIR)$@")
@@ -204,9 +200,6 @@ gdbstub-xml.c: $(TARGET_XML_FILES) $(SRC_PATH)/scripts/feature_to_c.sh
 hmp-commands.h: $(SRC_PATH)/hmp-commands.hx
 	$(call quiet-command,sh $(SRC_PATH)/scripts/hxtool -h < $< > $@,"  GEN   $(TARGET_DIR)$@")

-hmp-commands-info.h: $(SRC_PATH)/hmp-commands-info.hx
-	$(call quiet-command,sh $(SRC_PATH)/scripts/hxtool -h < $< > $@,"  GEN   $(TARGET_DIR)$@")
-
 qmp-commands-old.h: $(SRC_PATH)/qmp-commands.hx
 	$(call quiet-command,sh $(SRC_PATH)/scripts/hxtool -h < $< > $@,"  GEN   $(TARGET_DIR)$@")

--- a/108
+++ b/108
@@ -1,107 +1,3 @@
-         QEMU README
-         ===========
+Read the documentation in qemu-doc.html or on http://wiki.qemu-project.org

-QEMU is a generic and open source machine & userspace emulator and
-virtualizer.
-
-QEMU is capable of emulating a complete machine in software without any
-need for hardware virtualization support. By using dynamic translation,
-it achieves very good performance. QEMU can also integrate with the Xen
-and KVM hypervisors to provide emulated hardware while allowing the
-hypervisor to manage the CPU. With hypervisor support, QEMU can achieve
-near native performance for CPUs. When QEMU emulates CPUs directly it is
-capable of running operating systems made for one machine (e.g. an ARMv7
-board) on a different machine (e.g. an x86_64 PC board).
-
-QEMU is also capable of providing userspace API virtualization for Linux
-and BSD kernel interfaces. This allows binaries compiled against one
-architecture ABI (e.g. the Linux PPC64 ABI) to be run on a host using a
-different architecture ABI (e.g. the Linux x86_64 ABI). This does not
-involve any hardware emulation, simply CPU and syscall emulation.
-
-QEMU aims to fit into a variety of use cases. It can be invoked directly
-by users wishing to have full control over its behaviour and settings.
-It also aims to facilitate integration into higher level management
-layers, by providing a stable command line interface and monitor API.
-It is commonly invoked indirectly via the libvirt library when using
-open source applications such as oVirt, OpenStack and virt-manager.
-
-QEMU as a whole is released under the GNU General Public License,
-version 2. For full licensing details, consult the LICENSE file.
-
-
-Building
-========
-
-QEMU is multi-platform software intended to be buildable on all modern
-Linux platforms, OS-X, Win32 (via the Mingw64 toolchain) and a variety
-of other UNIX targets. The simple steps to build QEMU are:
-
-  mkdir build
-  cd build
-  ../configure
-  make
-
-Complete details of the process for building and configuring QEMU for
-all supported host platforms can be found in the qemu-tech.html file.
-Additional information can also be found online via the QEMU website:
-
-  http://qemu-project.org/Hosts/Linux
-  http://qemu-project.org/Hosts/W32
-
-
-Submitting patches
-==================
-
-The QEMU source code is maintained under the GIT version control system.
-
-   git clone git://git.qemu-project.org/qemu.git
-
-When submitting patches, the preferred approach is to use 'git
-format-patch' and/or 'git send-email' to format & send the mail to the
-qemu-devel@nongnu.org mailing list. All patches submitted must contain
-a 'Signed-off-by' line from the author. Patches should follow the
-guidelines set out in the HACKING and CODING_STYLE files.
-
-Additional information on submitting patches can be found online via
-the QEMU website
-
-  http://qemu-project.org/Contribute/SubmitAPatch
-  http://qemu-project.org/Contribute/TrivialPatches
-
-
-Bug reporting
-=============
-
-The QEMU project uses Launchpad as its primary upstream bug tracker. Bugs
-found when running code built from QEMU git or upstream released sources
-should be reported via:
-
-  https://bugs.launchpad.net/qemu/
-
-If using QEMU via an operating system vendor pre-built binary package, it
-is preferable to report bugs to the vendor's own bug tracker first. If
-the bug is also known to affect latest upstream code, it can also be
-reported via launchpad.
-
-For additional information on bug reporting consult:
-
-  http://qemu-project.org/Contribute/ReportABug
-
-
-Contact
-=======
-
-The QEMU community can be contacted in a number of ways, with the two
-main methods being email and IRC
-
- - qemu-devel@nongnu.org
-   http://lists.nongnu.org/mailman/listinfo/qemu-devel
- - #qemu on irc.oftc.net
-
-Information on additional methods of contacting the community can be
-found online via the QEMU website:
-
-  http://qemu-project.org/Contribute/StartHere
-
-- End
+- QEMU team
--- a/2
+++ b/2
@@ -1 +1 @@
-2.5.50
+2.3.1
--- a/accel.c
+++ b/accel.c
@@ -23,7 +23,6 @@
 * THE SOFTWARE.
 */

-#include "qemu/osdep.h"
 #include "sysemu/accel.h"
 #include "hw/boards.h"
 #include "qemu-common.h"
--- a/aio-posix.c
+++ b/aio-posix.c
@@ -13,14 +13,10 @@
 * GNU GPL, version 2 or (at your option) any later version.
 */

-#include "qemu/osdep.h"
 #include "qemu-common.h"
 #include "block/block.h"
 #include "qemu/queue.h"
 #include "qemu/sockets.h"
-#ifdef CONFIG_EPOLL
-#include <sys/epoll.h>
-#endif

 struct AioHandler
 {
@@ -28,167 +24,11 @@ struct AioHandler
    IOHandler *io_read;
    IOHandler *io_write;
    int deleted;
+    int pollfds_idx;
    void *opaque;
-    bool is_external;
    QLIST_ENTRY(AioHandler) node;
 };

-#ifdef CONFIG_EPOLL
-
-/* The fd number threashold to switch to epoll */
-#define EPOLL_ENABLE_THRESHOLD 64
-
-static void aio_epoll_disable(AioContext *ctx)
-{
-    ctx->epoll_available = false;
-    if (!ctx->epoll_enabled) {
-        return;
-    }
-    ctx->epoll_enabled = false;
-    close(ctx->epollfd);
-}
-
-static inline int epoll_events_from_pfd(int pfd_events)
-{
-    return (pfd_events & G_IO_IN ? EPOLLIN : 0) |
-           (pfd_events & G_IO_OUT ? EPOLLOUT : 0) |
-           (pfd_events & G_IO_HUP ? EPOLLHUP : 0) |
-           (pfd_events & G_IO_ERR ? EPOLLERR : 0);
-}
-
-static bool aio_epoll_try_enable(AioContext *ctx)
-{
-    AioHandler *node;
-    struct epoll_event event;
-
-    QLIST_FOREACH(node, &ctx->aio_handlers, node) {
-        int r;
-        if (node->deleted || !node->pfd.events) {
-            continue;
-        }
-        event.events = epoll_events_from_pfd(node->pfd.events);
-        event.data.ptr = node;
-        r = epoll_ctl(ctx->epollfd, EPOLL_CTL_ADD, node->pfd.fd, &event);
-        if (r) {
-            return false;
-        }
-    }
-    ctx->epoll_enabled = true;
-    return true;
-}
-
-static void aio_epoll_update(AioContext *ctx, AioHandler *node, bool is_new)
-{
-    struct epoll_event event;
-    int r;
-
-    if (!ctx->epoll_enabled) {
-        return;
-    }
-    if (!node->pfd.events) {
-        r = epoll_ctl(ctx->epollfd, EPOLL_CTL_DEL, node->pfd.fd, &event);
-        if (r) {
-            aio_epoll_disable(ctx);
-        }
-    } else {
-        event.data.ptr = node;
-        event.events = epoll_events_from_pfd(node->pfd.events);
-        if (is_new) {
-            r = epoll_ctl(ctx->epollfd, EPOLL_CTL_ADD, node->pfd.fd, &event);
-            if (r) {
-                aio_epoll_disable(ctx);
-            }
-        } else {
-            r = epoll_ctl(ctx->epollfd, EPOLL_CTL_MOD, node->pfd.fd, &event);
-            if (r) {
-                aio_epoll_disable(ctx);
-            }
-        }
-    }
-}
-
-static int aio_epoll(AioContext *ctx, GPollFD *pfds,
-                     unsigned npfd, int64_t timeout)
-{
-    AioHandler *node;
-    int i, ret = 0;
-    struct epoll_event events[128];
-
-    assert(npfd == 1);
-    assert(pfds[0].fd == ctx->epollfd);
-    if (timeout > 0) {
-        ret = qemu_poll_ns(pfds, npfd, timeout);
-    }
-    if (timeout <= 0 || ret > 0) {
-        ret = epoll_wait(ctx->epollfd, events,
-                         sizeof(events) / sizeof(events[0]),
-                         timeout);
-        if (ret <= 0) {
-            goto out;
-        }
-        for (i = 0; i < ret; i++) {
-            int ev = events[i].events;
-            node = events[i].data.ptr;
-            node->pfd.revents = (ev & EPOLLIN ? G_IO_IN : 0) |
-                (ev & EPOLLOUT ? G_IO_OUT : 0) |
-                (ev & EPOLLHUP ? G_IO_HUP : 0) |
-                (ev & EPOLLERR ? G_IO_ERR : 0);
-        }
-    }
-out:
-    return ret;
-}
-
-static bool aio_epoll_enabled(AioContext *ctx)
-{
-    /* Fall back to ppoll when external clients are disabled. */
-    return !aio_external_disabled(ctx) && ctx->epoll_enabled;
-}
-
-static bool aio_epoll_check_poll(AioContext *ctx, GPollFD *pfds,
-                                 unsigned npfd, int64_t timeout)
-{
-    if (!ctx->epoll_available) {
-        return false;
-    }
-    if (aio_epoll_enabled(ctx)) {
-        return true;
-    }
-    if (npfd >= EPOLL_ENABLE_THRESHOLD) {
-        if (aio_epoll_try_enable(ctx)) {
-            return true;
-        } else {
-            aio_epoll_disable(ctx);
-        }
-    }
-    return false;
-}
-
-#else
-
-static void aio_epoll_update(AioContext *ctx, AioHandler *node, bool is_new)
-{
-}
-
-static int aio_epoll(AioContext *ctx, GPollFD *pfds,
-                     unsigned npfd, int64_t timeout)
-{
-    assert(false);
-}
-
-static bool aio_epoll_enabled(AioContext *ctx)
-{
-    return false;
-}
-
-static bool aio_epoll_check_poll(AioContext *ctx, GPollFD *pfds,
-                          unsigned npfd, int64_t timeout)
-{
-    return false;
-}
-
-#endif
-
 static AioHandler *find_aio_handler(AioContext *ctx, int fd)
 {
    AioHandler *node;
@@ -204,14 +44,11 @@ static AioHandler *find_aio_handler(AioContext *ctx, int fd)

 void aio_set_fd_handler(AioContext *ctx,
                        int fd,
-                        bool is_external,
                        IOHandler *io_read,
                        IOHandler *io_write,
                        void *opaque)
 {
    AioHandler *node;
-    bool is_new = false;
-    bool deleted = false;

    node = find_aio_handler(ctx, fd);

@@ -230,7 +67,7 @@ void aio_set_fd_handler(AioContext *ctx,
                 * releasing the walking_handlers lock.
                 */
                QLIST_REMOVE(node, node);
-                deleted = true;
+                g_free(node);
            }
        }
    } else {
@@ -241,32 +78,26 @@ void aio_set_fd_handler(AioContext *ctx,
            QLIST_INSERT_HEAD(&ctx->aio_handlers, node, node);

            g_source_add_poll(&ctx->source, &node->pfd);
-            is_new = true;
        }
        /* Update handler with latest information */
        node->io_read = io_read;
        node->io_write = io_write;
        node->opaque = opaque;
-        node->is_external = is_external;
+        node->pollfds_idx = -1;

        node->pfd.events = (io_read ? G_IO_IN | G_IO_HUP | G_IO_ERR : 0);
        node->pfd.events |= (io_write ? G_IO_OUT | G_IO_ERR : 0);
    }

-    aio_epoll_update(ctx, node, is_new);
    aio_notify(ctx);
-    if (deleted) {
-        g_free(node);
-    }
 }

 void aio_set_event_notifier(AioContext *ctx,
                            EventNotifier *notifier,
-                            bool is_external,
                            EventNotifierHandler *io_read)
 {
    aio_set_fd_handler(ctx, event_notifier_get_fd(notifier),
-                       is_external, (IOHandler *)io_read, NULL, notifier);
+                       (IOHandler *)io_read, NULL, notifier);
 }

 bool aio_prepare(AioContext *ctx)
@@ -355,141 +186,69 @@ bool aio_dispatch(AioContext *ctx)
    return progress;
 }

-/* These thread-local variables are used only in a small part of aio_poll
- * around the call to the poll() system call.  In particular they are not
- * used while aio_poll is performing callbacks, which makes it much easier
- * to think about reentrancy!
- *
- * Stack-allocated arrays would be perfect but they have size limitations;
- * heap allocation is expensive enough that we want to reuse arrays across
- * calls to aio_poll().  And because poll() has to be called without holding
- * any lock, the arrays cannot be stored in AioContext.  Thread-local data
- * has none of the disadvantages of these three options.
- */
-static __thread GPollFD *pollfds;
-static __thread AioHandler **nodes;
-static __thread unsigned npfd, nalloc;
-static __thread Notifier pollfds_cleanup_notifier;
-
-static void pollfds_cleanup(Notifier *n, void *unused)
-{
-    g_assert(npfd == 0);
-    g_free(pollfds);
-    g_free(nodes);
-    nalloc = 0;
-}
-
-static void add_pollfd(AioHandler *node)
-{
-    if (npfd == nalloc) {
-        if (nalloc == 0) {
-            pollfds_cleanup_notifier.notify = pollfds_cleanup;
-            qemu_thread_atexit_add(&pollfds_cleanup_notifier);
-            nalloc = 8;
-        } else {
-            g_assert(nalloc <= INT_MAX);
-            nalloc *= 2;
-        }
-        pollfds = g_renew(GPollFD, pollfds, nalloc);
-        nodes = g_renew(AioHandler *, nodes, nalloc);
-    }
-    nodes[npfd] = node;
-    pollfds[npfd] = (GPollFD) {
-        .fd = node->pfd.fd,
-        .events = node->pfd.events,
-    };
-    npfd++;
-}
-
 bool aio_poll(AioContext *ctx, bool blocking)
 {
    AioHandler *node;
-    int i, ret;
+    bool was_dispatching;
+    int ret;
    bool progress;
-    int64_t timeout;

-    aio_context_acquire(ctx);
+    was_dispatching = ctx->dispatching;
    progress = false;

    /* aio_notify can avoid the expensive event_notifier_set if
     * everything (file descriptors, bottom halves, timers) will
     * be re-evaluated before the next blocking poll().  This is
     * already true when aio_poll is called with blocking == false;
-     * if blocking == true, it is only true after poll() returns,
-     * so disable the optimization now.
+     * if blocking == true, it is only true after poll() returns.
+     *
+     * If we're in a nested event loop, ctx->dispatching might be true.
+     * In that case we can restore it just before returning, but we
+     * have to clear it now.
     */
-    if (blocking) {
-        atomic_add(&ctx->notify_me, 2);
-    }
+    aio_set_dispatching(ctx, !blocking);

    ctx->walking_handlers++;

-    assert(npfd == 0);
+    g_array_set_size(ctx->pollfds, 0);

    /* fill pollfds */
    QLIST_FOREACH(node, &ctx->aio_handlers, node) {
-        if (!node->deleted && node->pfd.events
-            && !aio_epoll_enabled(ctx)
-            && aio_node_check(ctx, node->is_external)) {
-            add_pollfd(node);
+        node->pollfds_idx = -1;
+        if (!node->deleted && node->pfd.events) {
+            GPollFD pfd = {
+                .fd = node->pfd.fd,
+                .events = node->pfd.events,
+            };
+            node->pollfds_idx = ctx->pollfds->len;
+            g_array_append_val(ctx->pollfds, pfd);
        }
    }

-    timeout = blocking ? aio_compute_timeout(ctx) : 0;
+    ctx->walking_handlers--;

    /* wait until next event */
-    if (timeout) {
-        aio_context_release(ctx);
-    }
-    if (aio_epoll_check_poll(ctx, pollfds, npfd, timeout)) {
-        AioHandler epoll_handler;
-
-        epoll_handler.pfd.fd = ctx->epollfd;
-        epoll_handler.pfd.events = G_IO_IN | G_IO_OUT | G_IO_HUP | G_IO_ERR;
-        npfd = 0;
-        add_pollfd(&epoll_handler);
-        ret = aio_epoll(ctx, pollfds, npfd, timeout);
-    } else  {
-        ret = qemu_poll_ns(pollfds, npfd, timeout);
-    }
-    if (blocking) {
-        atomic_sub(&ctx->notify_me, 2);
-    }
-    if (timeout) {
-        aio_context_acquire(ctx);
-    }
-
-    aio_notify_accept(ctx);
+    ret = qemu_poll_ns((GPollFD *)ctx->pollfds->data,
+                         ctx->pollfds->len,
+                         blocking ? aio_compute_timeout(ctx) : 0);

    /* if we have any readable fds, dispatch event */
    if (ret > 0) {
-        for (i = 0; i < npfd; i++) {
-            nodes[i]->pfd.revents = pollfds[i].revents;
+        QLIST_FOREACH(node, &ctx->aio_handlers, node) {
+            if (node->pollfds_idx != -1) {
+                GPollFD *pfd = &g_array_index(ctx->pollfds, GPollFD,
+                                              node->pollfds_idx);
+                node->pfd.revents = pfd->revents;
+            }
        }
    }

-    npfd = 0;
-    ctx->walking_handlers--;
-
    /* Run dispatch even if there were no readable fds to run timers */
+    aio_set_dispatching(ctx, true);
    if (aio_dispatch(ctx)) {
        progress = true;
    }

-    aio_context_release(ctx);
-
+    aio_set_dispatching(ctx, was_dispatching);
    return progress;
 }
-
-void aio_context_setup(AioContext *ctx, Error **errp)
-{
-#ifdef CONFIG_EPOLL
-    assert(!ctx->epollfd);
-    ctx->epollfd = epoll_create1(EPOLL_CLOEXEC);
-    if (ctx->epollfd == -1) {
-        ctx->epoll_available = false;
-    } else {
-        ctx->epoll_available = true;
-    }
-#endif
-}
--- a/aio-win32.c
+++ b/aio-win32.c
@@ -15,7 +15,6 @@
 * GNU GPL, version 2 or (at your option) any later version.
 */

-#include "qemu/osdep.h"
 #include "qemu-common.h"
 #include "block/block.h"
 #include "qemu/queue.h"
@@ -29,13 +28,11 @@ struct AioHandler {
    GPollFD pfd;
    int deleted;
    void *opaque;
-    bool is_external;
    QLIST_ENTRY(AioHandler) node;
 };

 void aio_set_fd_handler(AioContext *ctx,
                        int fd,
-                        bool is_external,
                        IOHandler *io_read,
                        IOHandler *io_write,
                        void *opaque)
@@ -89,7 +86,6 @@ void aio_set_fd_handler(AioContext *ctx,
        node->opaque = opaque;
        node->io_read = io_read;
        node->io_write = io_write;
-        node->is_external = is_external;

        event = event_notifier_get_handle(&ctx->notifier);
        WSAEventSelect(node->pfd.fd, event,
@@ -102,7 +98,6 @@ void aio_set_fd_handler(AioContext *ctx,

 void aio_set_event_notifier(AioContext *ctx,
                            EventNotifier *e,
-                            bool is_external,
                            EventNotifierHandler *io_notify)
 {
    AioHandler *node;
@@ -138,7 +133,6 @@ void aio_set_event_notifier(AioContext *ctx,
            node->e = e;
            node->pfd.fd = (uintptr_t)event_notifier_get_handle(e);
            node->pfd.events = G_IO_IN;
-            node->is_external = is_external;
            QLIST_INSERT_HEAD(&ctx->aio_handlers, node, node);

            g_source_add_poll(&ctx->source, &node->pfd);
@@ -285,33 +279,36 @@ bool aio_poll(AioContext *ctx, bool blocking)
 {
    AioHandler *node;
    HANDLE events[MAXIMUM_WAIT_OBJECTS + 1];
-    bool progress, have_select_revents, first;
+    bool was_dispatching, progress, have_select_revents, first;
    int count;
    int timeout;

-    aio_context_acquire(ctx);
+    have_select_revents = aio_prepare(ctx);
+    if (have_select_revents) {
+        blocking = false;
+    }
+
+    was_dispatching = ctx->dispatching;
    progress = false;

    /* aio_notify can avoid the expensive event_notifier_set if
     * everything (file descriptors, bottom halves, timers) will
     * be re-evaluated before the next blocking poll().  This is
     * already true when aio_poll is called with blocking == false;
-     * if blocking == true, it is only true after poll() returns,
-     * so disable the optimization now.
+     * if blocking == true, it is only true after poll() returns.
+     *
+     * If we're in a nested event loop, ctx->dispatching might be true.
+     * In that case we can restore it just before returning, but we
+     * have to clear it now.
     */
-    if (blocking) {
-        atomic_add(&ctx->notify_me, 2);
-    }
-
-    have_select_revents = aio_prepare(ctx);
+    aio_set_dispatching(ctx, !blocking);

    ctx->walking_handlers++;

    /* fill fd sets */
    count = 0;
    QLIST_FOREACH(node, &ctx->aio_handlers, node) {
-        if (!node->deleted && node->io_notify
-            && aio_node_check(ctx, node->is_external)) {
+        if (!node->deleted && node->io_notify) {
            events[count++] = event_notifier_get_handle(node->e);
        }
    }
@@ -319,36 +316,20 @@ bool aio_poll(AioContext *ctx, bool blocking)
    ctx->walking_handlers--;
    first = true;

-    /* ctx->notifier is always registered.  */
-    assert(count > 0);
-
-    /* Multiple iterations, all of them non-blocking except the first,
-     * may be necessary to process all pending events.  After the first
-     * WaitForMultipleObjects call ctx->notify_me will be decremented.
-     */
-    do {
+    /* wait until next event */
+    while (count > 0) {
        HANDLE event;
        int ret;

-        timeout = blocking && !have_select_revents
+        timeout = blocking
            ? qemu_timeout_ns_to_ms(aio_compute_timeout(ctx)) : 0;
-        if (timeout) {
-            aio_context_release(ctx);
-        }
        ret = WaitForMultipleObjects(count, events, FALSE, timeout);
-        if (blocking) {
-            assert(first);
-            atomic_sub(&ctx->notify_me, 2);
-        }
-        if (timeout) {
-            aio_context_acquire(ctx);
-        }
+        aio_set_dispatching(ctx, true);

-        if (first) {
-            aio_notify_accept(ctx);
-            progress |= aio_bh_poll(ctx);
-            first = false;
+        if (first && aio_bh_poll(ctx)) {
+            progress = true;
        }
+        first = false;

        /* if we have any signaled events, dispatch event */
        event = NULL;
@@ -363,14 +344,10 @@ bool aio_poll(AioContext *ctx, bool blocking)
        blocking = false;

        progress |= aio_dispatch_handlers(ctx, event);
-    } while (count > 0);
+    }

    progress |= timerlistgroup_run_timers(&ctx->tlg);

-    aio_context_release(ctx);
+    aio_set_dispatching(ctx, was_dispatching);
    return progress;
 }
-
-void aio_context_setup(AioContext *ctx, Error **errp)
-{
-}
--- a/arch_init.c
+++ b/arch_init.c
--- a/async.c
+++ b/async.c
@@ -22,7 +22,6 @@
 * THE SOFTWARE.
 */

-#include "qemu/osdep.h"
 #include "qemu-common.h"
 #include "block/aio.h"
 #include "block/thread-pool.h"
@@ -60,11 +59,6 @@ QEMUBH *aio_bh_new(AioContext *ctx, QEMUBHFunc *cb, void *opaque)
    return bh;
 }

-void aio_bh_call(QEMUBH *bh)
-{
-    bh->cb(bh->opaque);
-}
-
 /* Multiple occurrences of aio_bh_poll cannot be called concurrently */
 int aio_bh_poll(AioContext *ctx)
 {
@@ -85,12 +79,10 @@ int aio_bh_poll(AioContext *ctx)
         * aio_notify again if necessary.
         */
        if (!bh->deleted && atomic_xchg(&bh->scheduled, 0)) {
-            /* Idle BHs and the notify BH don't count as progress */
-            if (!bh->idle && bh != ctx->notify_dummy_bh) {
+            if (!bh->idle)
                ret = 1;
-            }
            bh->idle = 0;
-            aio_bh_call(bh);
+            bh->cb(bh->opaque);
        }
    }

@@ -192,8 +184,6 @@ aio_ctx_prepare(GSource *source, gint    *timeout)
 {
    AioContext *ctx = (AioContext *) source;

-    atomic_or(&ctx->notify_me, 1);
-
    /* We assume there is no timeout already supplied */
    *timeout = qemu_timeout_ns_to_ms(aio_compute_timeout(ctx));

@@ -210,9 +200,6 @@ aio_ctx_check(GSource *source)
    AioContext *ctx = (AioContext *) source;
    QEMUBH *bh;

-    atomic_and(&ctx->notify_me, ~1);
-    aio_notify_accept(ctx);
-
    for (bh = ctx->first_bh; bh; bh = bh->next) {
        if (!bh->deleted && bh->scheduled) {
            return true;
@@ -238,25 +225,12 @@ aio_ctx_finalize(GSource     *source)
 {
    AioContext *ctx = (AioContext *) source;

-    qemu_bh_delete(ctx->notify_dummy_bh);
    thread_pool_free(ctx->thread_pool);
-
-    qemu_mutex_lock(&ctx->bh_lock);
-    while (ctx->first_bh) {
-        QEMUBH *next = ctx->first_bh->next;
-
-        /* qemu_bh_delete() must have been called on BHs in this AioContext */
-        assert(ctx->first_bh->deleted);
-
-        g_free(ctx->first_bh);
-        ctx->first_bh = next;
-    }
-    qemu_mutex_unlock(&ctx->bh_lock);
-
-    aio_set_event_notifier(ctx, &ctx->notifier, false, NULL);
+    aio_set_event_notifier(ctx, &ctx->notifier, NULL);
    event_notifier_cleanup(&ctx->notifier);
    rfifolock_destroy(&ctx->lock);
    qemu_mutex_destroy(&ctx->bh_lock);
+    g_array_free(ctx->pollfds, TRUE);
    timerlistgroup_deinit(&ctx->tlg);
 }

@@ -281,22 +255,24 @@ ThreadPool *aio_get_thread_pool(AioContext *ctx)
    return ctx->thread_pool;
 }

-void aio_notify(AioContext *ctx)
+void aio_set_dispatching(AioContext *ctx, bool dispatching)
 {
-    /* Write e.g. bh->scheduled before reading ctx->notify_me.  Pairs
-     * with atomic_or in aio_ctx_prepare or atomic_add in aio_poll.
-     */
-    smp_mb();
-    if (ctx->notify_me) {
-        event_notifier_set(&ctx->notifier);
-        atomic_mb_set(&ctx->notified, true);
+    ctx->dispatching = dispatching;
+    if (!dispatching) {
+        /* Write ctx->dispatching before reading e.g. bh->scheduled.
+         * Optimization: this is only needed when we're entering the "unsafe"
+         * phase where other threads must call event_notifier_set.
+         */
+        smp_mb();
    }
 }

-void aio_notify_accept(AioContext *ctx)
+void aio_notify(AioContext *ctx)
 {
-    if (atomic_xchg(&ctx->notified, false)) {
-        event_notifier_test_and_clear(&ctx->notifier);
+    /* Write e.g. bh->scheduled before reading ctx->dispatching.  */
+    smp_mb();
+    if (!ctx->dispatching) {
+        event_notifier_set(&ctx->notifier);
    }
 }

@@ -307,54 +283,32 @@ static void aio_timerlist_notify(void *opaque)

 static void aio_rfifolock_cb(void *opaque)
 {
-    AioContext *ctx = opaque;
-
    /* Kick owner thread in case they are blocked in aio_poll() */
-    qemu_bh_schedule(ctx->notify_dummy_bh);
-}
-
-static void notify_dummy_bh(void *opaque)
-{
-    /* Do nothing, we were invoked just to force the event loop to iterate */
-}
-
-static void event_notifier_dummy_cb(EventNotifier *e)
-{
+    aio_notify(opaque);
 }

 AioContext *aio_context_new(Error **errp)
 {
    int ret;
    AioContext *ctx;
-    Error *local_err = NULL;
-
    ctx = (AioContext *) g_source_new(&aio_source_funcs, sizeof(AioContext));
-    aio_context_setup(ctx, &local_err);
-    if (local_err) {
-        error_propagate(errp, local_err);
-        goto fail;
-    }
    ret = event_notifier_init(&ctx->notifier, false);
    if (ret < 0) {
+        g_source_destroy(&ctx->source);
        error_setg_errno(errp, -ret, "Failed to initialize event notifier");
-        goto fail;
+        return NULL;
    }
    g_source_set_can_recurse(&ctx->source, true);
    aio_set_event_notifier(ctx, &ctx->notifier,
-                           false,
                           (EventNotifierHandler *)
-                           event_notifier_dummy_cb);
+                           event_notifier_test_and_clear);
+    ctx->pollfds = g_array_new(FALSE, FALSE, sizeof(GPollFD));
    ctx->thread_pool = NULL;
    qemu_mutex_init(&ctx->bh_lock);
    rfifolock_init(&ctx->lock, aio_rfifolock_cb, ctx);
    timerlistgroup_init(&ctx->tlg, aio_timerlist_notify, ctx);

-    ctx->notify_dummy_bh = aio_bh_new(ctx, notify_dummy_bh, NULL);
-
    return ctx;
-fail:
-    g_source_destroy(&ctx->source);
-    return NULL;
 }

 void aio_context_ref(AioContext *ctx)
--- a/audio/Makefile.objs
+++ b/audio/Makefile.objs
@@ -5,9 +5,13 @@ common-obj-$(CONFIG_SPICE) += spiceaudio.o
 common-obj-$(CONFIG_COREAUDIO) += coreaudio.o
 common-obj-$(CONFIG_ALSA) += alsaaudio.o
 common-obj-$(CONFIG_DSOUND) += dsoundaudio.o
+common-obj-$(CONFIG_FMOD) += fmodaudio.o
+common-obj-$(CONFIG_ESD) += esdaudio.o
 common-obj-$(CONFIG_PA) += paaudio.o
+common-obj-$(CONFIG_WINWAVE) += winwaveaudio.o
 common-obj-$(CONFIG_AUDIO_PT_INT) += audio_pt_int.o
 common-obj-$(CONFIG_AUDIO_WIN_INT) += audio_win_int.o
 common-obj-y += wavcapture.o

+$(obj)/audio.o $(obj)/fmodaudio.o: QEMU_CFLAGS += $(FMOD_CFLAGS)
 sdlaudio.o-cflags := $(SDL_CFLAGS)
--- a/audio/alsaaudio.c
+++ b/audio/alsaaudio.c
@@ -21,12 +21,10 @@
 * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
 * THE SOFTWARE.
 */
-#include "qemu/osdep.h"
 #include <alsa/asoundlib.h>
 #include "qemu-common.h"
 #include "qemu/main-loop.h"
 #include "audio.h"
-#include "trace.h"

 #if QEMU_GNUC_PREREQ(4, 3)
 #pragma GCC diagnostic ignored "-Waddress"
@@ -35,28 +33,9 @@
 #define AUDIO_CAP "alsa"
 #include "audio_int.h"

-typedef struct ALSAConf {
-    int size_in_usec_in;
-    int size_in_usec_out;
-    const char *pcm_name_in;
-    const char *pcm_name_out;
-    unsigned int buffer_size_in;
-    unsigned int period_size_in;
-    unsigned int buffer_size_out;
-    unsigned int period_size_out;
-    unsigned int threshold;
-
-    int buffer_size_in_overridden;
-    int period_size_in_overridden;
-
-    int buffer_size_out_overridden;
-    int period_size_out_overridden;
-} ALSAConf;
-
 struct pollhlp {
    snd_pcm_t *handle;
    struct pollfd *pfds;
-    ALSAConf *conf;
    int count;
    int mask;
 };
@@ -77,6 +56,30 @@ typedef struct ALSAVoiceIn {
    struct pollhlp pollhlp;
 } ALSAVoiceIn;

+static struct {
+    int size_in_usec_in;
+    int size_in_usec_out;
+    const char *pcm_name_in;
+    const char *pcm_name_out;
+    unsigned int buffer_size_in;
+    unsigned int period_size_in;
+    unsigned int buffer_size_out;
+    unsigned int period_size_out;
+    unsigned int threshold;
+
+    int buffer_size_in_overridden;
+    int period_size_in_overridden;
+
+    int buffer_size_out_overridden;
+    int period_size_out_overridden;
+    int verbose;
+} conf = {
+    .buffer_size_out = 4096,
+    .period_size_out = 1024,
+    .pcm_name_out = "default",
+    .pcm_name_in = "default",
+};
+
 struct alsa_params_req {
    int freq;
    snd_pcm_format_t fmt;
@@ -202,7 +205,9 @@ static void alsa_poll_handler (void *opaque)
    }

    if (!(revents & hlp->mask)) {
-        trace_alsa_revents(revents);
+        if (conf.verbose) {
+            dolog ("revents = %d\n", revents);
+        }
        return;
    }

@@ -261,14 +266,31 @@ static int alsa_poll_helper (snd_pcm_t *handle, struct pollhlp *hlp, int mask)

    for (i = 0; i < count; ++i) {
        if (pfds[i].events & POLLIN) {
-            qemu_set_fd_handler (pfds[i].fd, alsa_poll_handler, NULL, hlp);
+            err = qemu_set_fd_handler (pfds[i].fd, alsa_poll_handler,
+                                       NULL, hlp);
        }
        if (pfds[i].events & POLLOUT) {
-            trace_alsa_pollout(i, pfds[i].fd);
-            qemu_set_fd_handler (pfds[i].fd, NULL, alsa_poll_handler, hlp);
+            if (conf.verbose) {
+                dolog ("POLLOUT %d %d\n", i, pfds[i].fd);
+            }
+            err = qemu_set_fd_handler (pfds[i].fd, NULL,
+                                       alsa_poll_handler, hlp);
+        }
+        if (conf.verbose) {
+            dolog ("Set handler events=%#x index=%d fd=%d err=%d\n",
+                   pfds[i].events, i, pfds[i].fd, err);
        }
-        trace_alsa_set_handler(pfds[i].events, i, pfds[i].fd, err);

+        if (err) {
+            dolog ("Failed to set handler events=%#x index=%d fd=%d err=%d\n",
+                   pfds[i].events, i, pfds[i].fd, err);
+
+            while (i--) {
+                qemu_set_fd_handler (pfds[i].fd, NULL, NULL, NULL);
+            }
+            g_free (pfds);
+            return -1;
+        }
    }
    hlp->pfds = pfds;
    hlp->count = count;
@@ -454,15 +476,14 @@ static void alsa_set_threshold (snd_pcm_t *handle, snd_pcm_uframes_t threshold)
 }

 static int alsa_open (int in, struct alsa_params_req *req,
-                      struct alsa_params_obt *obt, snd_pcm_t **handlep,
-                      ALSAConf *conf)
+                      struct alsa_params_obt *obt, snd_pcm_t **handlep)
 {
    snd_pcm_t *handle;
    snd_pcm_hw_params_t *hw_params;
    int err;
    int size_in_usec;
    unsigned int freq, nchannels;
-    const char *pcm_name = in ? conf->pcm_name_in : conf->pcm_name_out;
+    const char *pcm_name = in ? conf.pcm_name_in : conf.pcm_name_out;
    snd_pcm_uframes_t obt_buffer_size;
    const char *typ = in ? "ADC" : "DAC";
    snd_pcm_format_t obtfmt;
@@ -501,7 +522,7 @@ static int alsa_open (int in, struct alsa_params_req *req,
    }

    err = snd_pcm_hw_params_set_format (handle, hw_params, req->fmt);
-    if (err < 0) {
+    if (err < 0 && conf.verbose) {
        alsa_logerr2 (err, typ, "Failed to set format %d\n", req->fmt);
    }

@@ -633,7 +654,7 @@ static int alsa_open (int in, struct alsa_params_req *req,
        goto err;
    }

-    if (!in && conf->threshold) {
+    if (!in && conf.threshold) {
        snd_pcm_uframes_t threshold;
        int bytes_per_sec;

@@ -655,7 +676,7 @@ static int alsa_open (int in, struct alsa_params_req *req,
            break;
        }

-        threshold = (conf->threshold * bytes_per_sec) / 1000;
+        threshold = (conf.threshold * bytes_per_sec) / 1000;
        alsa_set_threshold (handle, threshold);
    }

@@ -665,9 +686,10 @@ static int alsa_open (int in, struct alsa_params_req *req,

    *handlep = handle;

-    if (obtfmt != req->fmt ||
+    if (conf.verbose &&
+        (obtfmt != req->fmt ||
         obt->nchannels != req->nchannels ||
-         obt->freq != req->freq) {
+         obt->freq != req->freq)) {
        dolog ("Audio parameters for %s\n", typ);
        alsa_dump_info (req, obt, obtfmt);
    }
@@ -721,7 +743,9 @@ static void alsa_write_pending (ALSAVoiceOut *alsa)
            if (written <= 0) {
                switch (written) {
                case 0:
-                    trace_alsa_wrote_zero(len);
+                    if (conf.verbose) {
+                        dolog ("Failed to write %d frames (wrote zero)\n", len);
+                    }
                    return;

                case -EPIPE:
@@ -730,7 +754,9 @@ static void alsa_write_pending (ALSAVoiceOut *alsa)
                                     len);
                        return;
                    }
-                    trace_alsa_xrun_out();
+                    if (conf.verbose) {
+                        dolog ("Recovering from playback xrun\n");
+                    }
                    continue;

                case -ESTRPIPE:
@@ -741,7 +767,9 @@ static void alsa_write_pending (ALSAVoiceOut *alsa)
                                     len);
                        return;
                    }
-                    trace_alsa_resume_out();
+                    if (conf.verbose) {
+                        dolog ("Resuming suspended output stream\n");
+                    }
                    continue;

                case -EAGAIN:
@@ -791,27 +819,25 @@ static void alsa_fini_out (HWVoiceOut *hw)
    alsa->pcm_buf = NULL;
 }

-static int alsa_init_out(HWVoiceOut *hw, struct audsettings *as,
-                         void *drv_opaque)
+static int alsa_init_out (HWVoiceOut *hw, struct audsettings *as)
 {
    ALSAVoiceOut *alsa = (ALSAVoiceOut *) hw;
    struct alsa_params_req req;
    struct alsa_params_obt obt;
    snd_pcm_t *handle;
    struct audsettings obt_as;
-    ALSAConf *conf = drv_opaque;

    req.fmt = aud_to_alsafmt (as->fmt, as->endianness);
    req.freq = as->freq;
    req.nchannels = as->nchannels;
-    req.period_size = conf->period_size_out;
-    req.buffer_size = conf->buffer_size_out;
-    req.size_in_usec = conf->size_in_usec_out;
+    req.period_size = conf.period_size_out;
+    req.buffer_size = conf.buffer_size_out;
+    req.size_in_usec = conf.size_in_usec_out;
    req.override_mask =
-        (conf->period_size_out_overridden ? 1 : 0) |
-        (conf->buffer_size_out_overridden ? 2 : 0);
+        (conf.period_size_out_overridden ? 1 : 0) |
+        (conf.buffer_size_out_overridden ? 2 : 0);

-    if (alsa_open (0, &req, &obt, &handle, conf)) {
+    if (alsa_open (0, &req, &obt, &handle)) {
        return -1;
    }

@@ -832,7 +858,6 @@ static int alsa_init_out(HWVoiceOut *hw, struct audsettings *as,
    }

    alsa->handle = handle;
-    alsa->pollhlp.conf = conf;
    return 0;
 }

@@ -903,26 +928,25 @@ static int alsa_ctl_out (HWVoiceOut *hw, int cmd, ...)
    return -1;
 }

-static int alsa_init_in(HWVoiceIn *hw, struct audsettings *as, void *drv_opaque)
+static int alsa_init_in (HWVoiceIn *hw, struct audsettings *as)
 {
    ALSAVoiceIn *alsa = (ALSAVoiceIn *) hw;
    struct alsa_params_req req;
    struct alsa_params_obt obt;
    snd_pcm_t *handle;
    struct audsettings obt_as;
-    ALSAConf *conf = drv_opaque;

    req.fmt = aud_to_alsafmt (as->fmt, as->endianness);
    req.freq = as->freq;
    req.nchannels = as->nchannels;
-    req.period_size = conf->period_size_in;
-    req.buffer_size = conf->buffer_size_in;
-    req.size_in_usec = conf->size_in_usec_in;
+    req.period_size = conf.period_size_in;
+    req.buffer_size = conf.buffer_size_in;
+    req.size_in_usec = conf.size_in_usec_in;
    req.override_mask =
-        (conf->period_size_in_overridden ? 1 : 0) |
-        (conf->buffer_size_in_overridden ? 2 : 0);
+        (conf.period_size_in_overridden ? 1 : 0) |
+        (conf.buffer_size_in_overridden ? 2 : 0);

-    if (alsa_open (1, &req, &obt, &handle, conf)) {
+    if (alsa_open (1, &req, &obt, &handle)) {
        return -1;
    }

@@ -943,7 +967,6 @@ static int alsa_init_in(HWVoiceIn *hw, struct audsettings *as, void *drv_opaque)
    }

    alsa->handle = handle;
-    alsa->pollhlp.conf = conf;
    return 0;
 }

@@ -999,10 +1022,14 @@ static int alsa_run_in (HWVoiceIn *hw)
                dolog ("Failed to resume suspended input stream\n");
                return 0;
            }
-            trace_alsa_resume_in();
+            if (conf.verbose) {
+                dolog ("Resuming suspended input stream\n");
+            }
            break;
        default:
-            trace_alsa_no_frames(state);
+            if (conf.verbose) {
+                dolog ("No frames available and ALSA state is %d\n", state);
+            }
            return 0;
        }
    }
@@ -1037,7 +1064,9 @@ static int alsa_run_in (HWVoiceIn *hw)
            if (nread <= 0) {
                switch (nread) {
                case 0:
-                    trace_alsa_read_zero(len);
+                    if (conf.verbose) {
+                        dolog ("Failed to read %ld frames (read zero)\n", len);
+                    }
                    goto exit;

                case -EPIPE:
@@ -1045,7 +1074,9 @@ static int alsa_run_in (HWVoiceIn *hw)
                        alsa_logerr (nread, "Failed to read %ld frames\n", len);
                        goto exit;
                    }
-                    trace_alsa_xrun_in();
+                    if (conf.verbose) {
+                        dolog ("Recovering from capture xrun\n");
+                    }
                    continue;

                case -EAGAIN:
@@ -1117,85 +1148,82 @@ static int alsa_ctl_in (HWVoiceIn *hw, int cmd, ...)
    return -1;
 }

-static ALSAConf glob_conf = {
-    .buffer_size_out = 4096,
-    .period_size_out = 1024,
-    .pcm_name_out = "default",
-    .pcm_name_in = "default",
-};
-
 static void *alsa_audio_init (void)
 {
-    ALSAConf *conf = g_malloc(sizeof(ALSAConf));
-    *conf = glob_conf;
-    return conf;
+    return &conf;
 }

 static void alsa_audio_fini (void *opaque)
 {
-    g_free(opaque);
+    (void) opaque;
 }

 static struct audio_option alsa_options[] = {
    {
        .name        = "DAC_SIZE_IN_USEC",
        .tag         = AUD_OPT_BOOL,
-        .valp        = &glob_conf.size_in_usec_out,
+        .valp        = &conf.size_in_usec_out,
        .descr       = "DAC period/buffer size in microseconds (otherwise in frames)"
    },
    {
        .name        = "DAC_PERIOD_SIZE",
        .tag         = AUD_OPT_INT,
-        .valp        = &glob_conf.period_size_out,
+        .valp        = &conf.period_size_out,
        .descr       = "DAC period size (0 to go with system default)",
-        .overriddenp = &glob_conf.period_size_out_overridden
+        .overriddenp = &conf.period_size_out_overridden
    },
    {
        .name        = "DAC_BUFFER_SIZE",
        .tag         = AUD_OPT_INT,
-        .valp        = &glob_conf.buffer_size_out,
+        .valp        = &conf.buffer_size_out,
        .descr       = "DAC buffer size (0 to go with system default)",
-        .overriddenp = &glob_conf.buffer_size_out_overridden
+        .overriddenp = &conf.buffer_size_out_overridden
    },
    {
        .name        = "ADC_SIZE_IN_USEC",
        .tag         = AUD_OPT_BOOL,
-        .valp        = &glob_conf.size_in_usec_in,
+        .valp        = &conf.size_in_usec_in,
        .descr       =
        "ADC period/buffer size in microseconds (otherwise in frames)"
    },
    {
        .name        = "ADC_PERIOD_SIZE",
        .tag         = AUD_OPT_INT,
-        .valp        = &glob_conf.period_size_in,
+        .valp        = &conf.period_size_in,
        .descr       = "ADC period size (0 to go with system default)",
-        .overriddenp = &glob_conf.period_size_in_overridden
+        .overriddenp = &conf.period_size_in_overridden
    },
    {
        .name        = "ADC_BUFFER_SIZE",
        .tag         = AUD_OPT_INT,
-        .valp        = &glob_conf.buffer_size_in,
+        .valp        = &conf.buffer_size_in,
        .descr       = "ADC buffer size (0 to go with system default)",
-        .overriddenp = &glob_conf.buffer_size_in_overridden
+        .overriddenp = &conf.buffer_size_in_overridden
    },
    {
        .name        = "THRESHOLD",
        .tag         = AUD_OPT_INT,
-        .valp        = &glob_conf.threshold,
+        .valp        = &conf.threshold,
        .descr       = "(undocumented)"
    },
    {
        .name        = "DAC_DEV",
        .tag         = AUD_OPT_STR,
-        .valp        = &glob_conf.pcm_name_out,
+        .valp        = &conf.pcm_name_out,
        .descr       = "DAC device name (for instance dmix)"
    },
    {
        .name        = "ADC_DEV",
        .tag         = AUD_OPT_STR,
-        .valp        = &glob_conf.pcm_name_in,
+        .valp        = &conf.pcm_name_in,
        .descr       = "ADC device name"
    },
+    {
+        .name        = "VERBOSE",
+        .tag         = AUD_OPT_BOOL,
+        .valp        = &conf.verbose,
+        .descr       = "Behave in a more verbose way"
+    },
    { /* End of list */ }
 };

--- a/audio/audio.c
+++ b/audio/audio.c
@@ -21,7 +21,6 @@
 * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
 * THE SOFTWARE.
 */
-#include "qemu/osdep.h"
 #include "hw/hw.h"
 #include "audio.h"
 #include "monitor/monitor.h"
@@ -31,6 +30,7 @@
 #define AUDIO_CAP "audio"
 #include "audio_int.h"

+/* #define DEBUG_PLIVE */
 /* #define DEBUG_LIVE */
 /* #define DEBUG_OUT */
 /* #define DEBUG_CAPTURE */
@@ -66,6 +66,8 @@ static struct {
        int hertz;
        int64_t ticks;
    } period;
+    int plive;
+    int log_to_monitor;
    int try_poll_in;
    int try_poll_out;
 } conf = {
@@ -94,6 +96,8 @@ static struct {
    },

    .period = { .hertz = 100 },
+    .plive = 0,
+    .log_to_monitor = 0,
    .try_poll_in = 1,
    .try_poll_out = 1,
 };
@@ -327,11 +331,20 @@ static const char *audio_get_conf_str (const char *key,

 void AUD_vlog (const char *cap, const char *fmt, va_list ap)
 {
-    if (cap) {
-        fprintf(stderr, "%s: ", cap);
-    }
+    if (conf.log_to_monitor) {
+        if (cap) {
+            monitor_printf(default_mon, "%s: ", cap);
+        }

-    vfprintf(stderr, fmt, ap);
+        monitor_vprintf(default_mon, fmt, ap);
+    }
+    else {
+        if (cap) {
+            fprintf (stderr, "%s: ", cap);
+        }
+
+        vfprintf (stderr, fmt, ap);
+    }
 }

 void AUD_log (const char *cap, const char *fmt, ...)
@@ -1441,6 +1454,9 @@ static void audio_run_out (AudioState *s)
            while (sw) {
                sw1 = sw->entries.le_next;
                if (!sw->active && !sw->callback.fn) {
+#ifdef DEBUG_PLIVE
+                    dolog ("Finishing with old voice\n");
+#endif
                    audio_close_out (sw);
                }
                sw = sw1;
@@ -1632,6 +1648,18 @@ static struct audio_option audio_options[] = {
        .valp  = &conf.period.hertz,
        .descr = "Timer period in HZ (0 - use lowest possible)"
    },
+    {
+        .name  = "PLIVE",
+        .tag   = AUD_OPT_BOOL,
+        .valp  = &conf.plive,
+        .descr = "(undocumented)"
+    },
+    {
+        .name  = "LOG_TO_MONITOR",
+        .tag   = AUD_OPT_BOOL,
+        .valp  = &conf.log_to_monitor,
+        .descr = "Print logging messages to monitor instead of stderr"
+    },
    { /* End of list */ }
 };

@@ -1807,6 +1835,9 @@ static void audio_init (void)
    atexit (audio_atexit);

    s->ts = timer_new_ns(QEMU_CLOCK_VIRTUAL, audio_timer, s);
+    if (!s->ts) {
+        hw_error("Could not create audio timer\n");
+    }

    audio_process_options ("AUDIO", audio_options);

@@ -1857,8 +1888,12 @@ static void audio_init (void)

    if (!done) {
        done = !audio_driver_init (s, &no_audio_driver);
-        assert(done);
-        dolog("warning: Using timer based audio emulation\n");
+        if (!done) {
+            hw_error("Could not initialize audio subsystem\n");
+        }
+        else {
+            dolog ("warning: Using timer based audio emulation\n");
+        }
    }

    if (conf.period.hertz <= 0) {
--- a/audio/audio.h
+++ b/audio/audio.h
@@ -24,6 +24,7 @@
 #ifndef QEMU_AUDIO_H
 #define QEMU_AUDIO_H

+#include "config-host.h"
 #include "qemu/queue.h"

 typedef void (*audio_callback_fn) (void *opaque, int avail);
--- a/audio/audio_int.h
+++ b/audio/audio_int.h
@@ -156,13 +156,13 @@ struct audio_driver {
 };

 struct audio_pcm_ops {
-    int  (*init_out)(HWVoiceOut *hw, struct audsettings *as, void *drv_opaque);
+    int  (*init_out)(HWVoiceOut *hw, struct audsettings *as);
    void (*fini_out)(HWVoiceOut *hw);
    int  (*run_out) (HWVoiceOut *hw, int live);
    int  (*write)   (SWVoiceOut *sw, void *buf, int size);
    int  (*ctl_out) (HWVoiceOut *hw, int cmd, ...);

-    int  (*init_in) (HWVoiceIn *hw, struct audsettings *as, void *drv_opaque);
+    int  (*init_in) (HWVoiceIn *hw, struct audsettings *as);
    void (*fini_in) (HWVoiceIn *hw);
    int  (*run_in)  (HWVoiceIn *hw);
    int  (*read)    (SWVoiceIn *sw, void *buf, int size);
@@ -206,11 +206,14 @@ extern struct audio_driver no_audio_driver;
 extern struct audio_driver oss_audio_driver;
 extern struct audio_driver sdl_audio_driver;
 extern struct audio_driver wav_audio_driver;
+extern struct audio_driver fmod_audio_driver;
 extern struct audio_driver alsa_audio_driver;
 extern struct audio_driver coreaudio_audio_driver;
 extern struct audio_driver dsound_audio_driver;
+extern struct audio_driver esd_audio_driver;
 extern struct audio_driver pa_audio_driver;
 extern struct audio_driver spice_audio_driver;
+extern struct audio_driver winwave_audio_driver;
 extern const struct mixeng_volume nominal_volume;

 void audio_pcm_init_info (struct audio_pcm_info *info, struct audsettings *as);
--- a/audio/audio_pt_int.c
+++ b/audio/audio_pt_int.c
@@ -1,4 +1,3 @@
-#include "qemu/osdep.h"
 #include "qemu-common.h"
 #include "audio.h"

--- a/audio/audio_template.h
+++ b/audio/audio_template.h
@@ -262,7 +262,7 @@ static HW *glue (audio_pcm_hw_add_new_, TYPE) (struct audsettings *as)
 #ifdef DAC
    QLIST_INIT (&hw->cap_head);
 #endif
-    if (glue (hw->pcm_ops->init_, TYPE) (hw, as, s->drv_opaque)) {
+    if (glue (hw->pcm_ops->init_, TYPE) (hw, as)) {
        goto err0;
    }

@@ -398,6 +398,10 @@ SW *glue (AUD_open_, TYPE) (
    )
 {
    AudioState *s = &glob_audio_state;
+#ifdef DAC
+    int live = 0;
+    SW *old_sw = NULL;
+#endif

    if (audio_bug (AUDIO_FUNC, !card || !name || !callback_fn || !as)) {
        dolog ("card=%p name=%p callback_fn=%p as=%p\n",
@@ -422,6 +426,29 @@ SW *glue (AUD_open_, TYPE) (
        return sw;
    }

+#ifdef DAC
+    if (conf.plive && sw && (!sw->active && !sw->empty)) {
+        live = sw->total_hw_samples_mixed;
+
+#ifdef DEBUG_PLIVE
+        dolog ("Replacing voice %s with %d live samples\n", SW_NAME (sw), live);
+        dolog ("Old %s freq %d, bits %d, channels %d\n",
+               SW_NAME (sw), sw->info.freq, sw->info.bits, sw->info.nchannels);
+        dolog ("New %s freq %d, bits %d, channels %d\n",
+               name,
+               as->freq,
+               (as->fmt == AUD_FMT_S16 || as->fmt == AUD_FMT_U16) ? 16 : 8,
+               as->nchannels);
+#endif
+
+        if (live) {
+            old_sw = sw;
+            old_sw->callback.fn = NULL;
+            sw = NULL;
+        }
+    }
+#endif
+
    if (!glue (conf.fixed_, TYPE).enabled && sw) {
        glue (AUD_close_, TYPE) (card, sw);
        sw = NULL;
@@ -454,6 +481,20 @@ SW *glue (AUD_open_, TYPE) (
    sw->callback.fn = callback_fn;
    sw->callback.opaque = callback_opaque;

+#ifdef DAC
+    if (live) {
+        int mixed =
+            (live << old_sw->info.shift)
+            * old_sw->info.bytes_per_second
+            / sw->info.bytes_per_second;
+
+#ifdef DEBUG_PLIVE
+        dolog ("Silence will be mixed %d\n", mixed);
+#endif
+        sw->total_hw_samples_mixed += mixed;
+    }
+#endif
+
 #ifdef DEBUG_AUDIO
    dolog ("%s\n", name);
    audio_pcm_print_info ("hw", &sw->hw->info);
--- a/audio/audio_win_int.c
+++ b/audio/audio_win_int.c
@@ -1,6 +1,5 @@
 /* public domain */

-#include "qemu/osdep.h"
 #include "qemu-common.h"

 #define AUDIO_CAP "win-int"
--- a/audio/coreaudio.c
+++ b/audio/coreaudio.c
@@ -22,8 +22,8 @@
 * THE SOFTWARE.
 */

-#include "qemu/osdep.h"
 #include <CoreAudio/CoreAudio.h>
+#include <string.h>             /* strerror */
 #include <pthread.h>            /* pthread_X */

 #include "qemu-common.h"
@@ -32,250 +32,28 @@
 #define AUDIO_CAP "coreaudio"
 #include "audio_int.h"

-#ifndef MAC_OS_X_VERSION_10_6
-#define MAC_OS_X_VERSION_10_6 1060
-#endif
-
-static int isAtexit;
-
-typedef struct {
+struct {
    int buffer_frames;
    int nbuffers;
-} CoreaudioConf;
+    int isAtexit;
+} conf = {
+    .buffer_frames = 512,
+    .nbuffers = 4,
+    .isAtexit = 0
+};

 typedef struct coreaudioVoiceOut {
    HWVoiceOut hw;
    pthread_mutex_t mutex;
+    int isAtexit;
    AudioDeviceID outputDeviceID;
    UInt32 audioDevicePropertyBufferFrameSize;
    AudioStreamBasicDescription outputStreamBasicDescription;
-    AudioDeviceIOProcID ioprocid;
    int live;
    int decr;
    int rpos;
 } coreaudioVoiceOut;

-#if MAC_OS_X_VERSION_MAX_ALLOWED >= MAC_OS_X_VERSION_10_6
-/* The APIs used here only become available from 10.6 */
-
-static OSStatus coreaudio_get_voice(AudioDeviceID *id)
-{
-    UInt32 size = sizeof(*id);
-    AudioObjectPropertyAddress addr = {
-        kAudioHardwarePropertyDefaultOutputDevice,
-        kAudioObjectPropertyScopeGlobal,
-        kAudioObjectPropertyElementMaster
-    };
-
-    return AudioObjectGetPropertyData(kAudioObjectSystemObject,
-                                      &addr,
-                                      0,
-                                      NULL,
-                                      &size,
-                                      id);
-}
-
-static OSStatus coreaudio_get_framesizerange(AudioDeviceID id,
-                                             AudioValueRange *framerange)
-{
-    UInt32 size = sizeof(*framerange);
-    AudioObjectPropertyAddress addr = {
-        kAudioDevicePropertyBufferFrameSizeRange,
-        kAudioDevicePropertyScopeOutput,
-        kAudioObjectPropertyElementMaster
-    };
-
-    return AudioObjectGetPropertyData(id,
-                                      &addr,
-                                      0,
-                                      NULL,
-                                      &size,
-                                      framerange);
-}
-
-static OSStatus coreaudio_get_framesize(AudioDeviceID id, UInt32 *framesize)
-{
-    UInt32 size = sizeof(*framesize);
-    AudioObjectPropertyAddress addr = {
-        kAudioDevicePropertyBufferFrameSize,
-        kAudioDevicePropertyScopeOutput,
-        kAudioObjectPropertyElementMaster
-    };
-
-    return AudioObjectGetPropertyData(id,
-                                      &addr,
-                                      0,
-                                      NULL,
-                                      &size,
-                                      framesize);
-}
-
-static OSStatus coreaudio_set_framesize(AudioDeviceID id, UInt32 *framesize)
-{
-    UInt32 size = sizeof(*framesize);
-    AudioObjectPropertyAddress addr = {
-        kAudioDevicePropertyBufferFrameSize,
-        kAudioDevicePropertyScopeOutput,
-        kAudioObjectPropertyElementMaster
-    };
-
-    return AudioObjectSetPropertyData(id,
-                                      &addr,
-                                      0,
-                                      NULL,
-                                      size,
-                                      framesize);
-}
-
-static OSStatus coreaudio_get_streamformat(AudioDeviceID id,
-                                           AudioStreamBasicDescription *d)
-{
-    UInt32 size = sizeof(*d);
-    AudioObjectPropertyAddress addr = {
-        kAudioDevicePropertyStreamFormat,
-        kAudioDevicePropertyScopeOutput,
-        kAudioObjectPropertyElementMaster
-    };
-
-    return AudioObjectGetPropertyData(id,
-                                      &addr,
-                                      0,
-                                      NULL,
-                                      &size,
-                                      d);
-}
-
-static OSStatus coreaudio_set_streamformat(AudioDeviceID id,
-                                           AudioStreamBasicDescription *d)
-{
-    UInt32 size = sizeof(*d);
-    AudioObjectPropertyAddress addr = {
-        kAudioDevicePropertyStreamFormat,
-        kAudioDevicePropertyScopeOutput,
-        kAudioObjectPropertyElementMaster
-    };
-
-    return AudioObjectSetPropertyData(id,
-                                      &addr,
-                                      0,
-                                      NULL,
-                                      size,
-                                      d);
-}
-
-static OSStatus coreaudio_get_isrunning(AudioDeviceID id, UInt32 *result)
-{
-    UInt32 size = sizeof(*result);
-    AudioObjectPropertyAddress addr = {
-        kAudioDevicePropertyDeviceIsRunning,
-        kAudioDevicePropertyScopeOutput,
-        kAudioObjectPropertyElementMaster
-    };
-
-    return AudioObjectGetPropertyData(id,
-                                      &addr,
-                                      0,
-                                      NULL,
-                                      &size,
-                                      result);
-}
-#else
-/* Legacy versions of functions using deprecated APIs */
-
-static OSStatus coreaudio_get_voice(AudioDeviceID *id)
-{
-    UInt32 size = sizeof(*id);
-
-    return AudioHardwareGetProperty(
-        kAudioHardwarePropertyDefaultOutputDevice,
-        &size,
-        id);
-}
-
-static OSStatus coreaudio_get_framesizerange(AudioDeviceID id,
-                                             AudioValueRange *framerange)
-{
-    UInt32 size = sizeof(*framerange);
-
-    return AudioDeviceGetProperty(
-        id,
-        0,
-        0,
-        kAudioDevicePropertyBufferFrameSizeRange,
-        &size,
-        framerange);
-}
-
-static OSStatus coreaudio_get_framesize(AudioDeviceID id, UInt32 *framesize)
-{
-    UInt32 size = sizeof(*framesize);
-
-    return AudioDeviceGetProperty(
-        id,
-        0,
-        false,
-        kAudioDevicePropertyBufferFrameSize,
-        &size,
-        framesize);
-}
-
-static OSStatus coreaudio_set_framesize(AudioDeviceID id, UInt32 *framesize)
-{
-    UInt32 size = sizeof(*framesize);
-
-    return AudioDeviceSetProperty(
-        id,
-        NULL,
-        0,
-        false,
-        kAudioDevicePropertyBufferFrameSize,
-        size,
-        framesize);
-}
-
-static OSStatus coreaudio_get_streamformat(AudioDeviceID id,
-                                           AudioStreamBasicDescription *d)
-{
-    UInt32 size = sizeof(*d);
-
-    return AudioDeviceGetProperty(
-        id,
-        0,
-        false,
-        kAudioDevicePropertyStreamFormat,
-        &size,
-        d);
-}
-
-static OSStatus coreaudio_set_streamformat(AudioDeviceID id,
-                                           AudioStreamBasicDescription *d)
-{
-    UInt32 size = sizeof(*d);
-
-    return AudioDeviceSetProperty(
-        id,
-        0,
-        0,
-        0,
-        kAudioDevicePropertyStreamFormat,
-        size,
-        d);
-}
-
-static OSStatus coreaudio_get_isrunning(AudioDeviceID id, UInt32 *result)
-{
-    UInt32 size = sizeof(*result);
-
-    return AudioDeviceGetProperty(
-        id,
-        0,
-        0,
-        kAudioDevicePropertyDeviceIsRunning,
-        &size,
-        result);
-}
-#endif
-
 static void coreaudio_logstatus (OSStatus status)
 {
    const char *str = "BUG";
@@ -370,7 +148,10 @@ static inline UInt32 isPlaying (AudioDeviceID outputDeviceID)
 {
    OSStatus status;
    UInt32 result = 0;
-    status = coreaudio_get_isrunning(outputDeviceID, &result);
+    UInt32 propertySize = sizeof(outputDeviceID);
+    status = AudioDeviceGetProperty(
+        outputDeviceID, 0, 0,
+        kAudioDevicePropertyDeviceIsRunning, &propertySize, &result);
    if (status != kAudioHardwareNoError) {
        coreaudio_logerr(status,
                         "Could not determine whether Device is playing\n");
@@ -380,7 +161,7 @@ static inline UInt32 isPlaying (AudioDeviceID outputDeviceID)

 static void coreaudio_atexit (void)
 {
-    isAtexit = 1;
+    conf.isAtexit = 1;
 }

 static int coreaudio_lock (coreaudioVoiceOut *core, const char *fn_name)
@@ -506,15 +287,14 @@ static int coreaudio_write (SWVoiceOut *sw, void *buf, int len)
    return audio_pcm_sw_write (sw, buf, len);
 }

-static int coreaudio_init_out(HWVoiceOut *hw, struct audsettings *as,
-                              void *drv_opaque)
+static int coreaudio_init_out (HWVoiceOut *hw, struct audsettings *as)
 {
    OSStatus status;
    coreaudioVoiceOut *core = (coreaudioVoiceOut *) hw;
+    UInt32 propertySize;
    int err;
    const char *typ = "playback";
    AudioValueRange frameRange;
-    CoreaudioConf *conf = drv_opaque;

    /* create mutex */
    err = pthread_mutex_init(&core->mutex, NULL);
@@ -525,7 +305,12 @@ static int coreaudio_init_out(HWVoiceOut *hw, struct audsettings *as,

    audio_pcm_init_info (&hw->info, as);

-    status = coreaudio_get_voice(&core->outputDeviceID);
+    /* open default output device */
+    propertySize = sizeof(core->outputDeviceID);
+    status = AudioHardwareGetProperty(
+        kAudioHardwarePropertyDefaultOutputDevice,
+        &propertySize,
+        &core->outputDeviceID);
    if (status != kAudioHardwareNoError) {
        coreaudio_logerr2 (status, typ,
                           "Could not get default output Device\n");
@@ -537,29 +322,42 @@ static int coreaudio_init_out(HWVoiceOut *hw, struct audsettings *as,
    }

    /* get minimum and maximum buffer frame sizes */
-    status = coreaudio_get_framesizerange(core->outputDeviceID,
-                                          &frameRange);
+    propertySize = sizeof(frameRange);
+    status = AudioDeviceGetProperty(
+        core->outputDeviceID,
+        0,
+        0,
+        kAudioDevicePropertyBufferFrameSizeRange,
+        &propertySize,
+        &frameRange);
    if (status != kAudioHardwareNoError) {
        coreaudio_logerr2 (status, typ,
                           "Could not get device buffer frame range\n");
        return -1;
    }

-    if (frameRange.mMinimum > conf->buffer_frames) {
+    if (frameRange.mMinimum > conf.buffer_frames) {
        core->audioDevicePropertyBufferFrameSize = (UInt32) frameRange.mMinimum;
        dolog ("warning: Upsizing Buffer Frames to %f\n", frameRange.mMinimum);
    }
-    else if (frameRange.mMaximum < conf->buffer_frames) {
+    else if (frameRange.mMaximum < conf.buffer_frames) {
        core->audioDevicePropertyBufferFrameSize = (UInt32) frameRange.mMaximum;
        dolog ("warning: Downsizing Buffer Frames to %f\n", frameRange.mMaximum);
    }
    else {
-        core->audioDevicePropertyBufferFrameSize = conf->buffer_frames;
+        core->audioDevicePropertyBufferFrameSize = conf.buffer_frames;
    }

    /* set Buffer Frame Size */
-    status = coreaudio_set_framesize(core->outputDeviceID,
-                                     &core->audioDevicePropertyBufferFrameSize);
+    propertySize = sizeof(core->audioDevicePropertyBufferFrameSize);
+    status = AudioDeviceSetProperty(
+        core->outputDeviceID,
+        NULL,
+        0,
+        false,
+        kAudioDevicePropertyBufferFrameSize,
+        propertySize,
+        &core->audioDevicePropertyBufferFrameSize);
    if (status != kAudioHardwareNoError) {
        coreaudio_logerr2 (status, typ,
                           "Could not set device buffer frame size %" PRIu32 "\n",
@@ -568,18 +366,30 @@ static int coreaudio_init_out(HWVoiceOut *hw, struct audsettings *as,
    }

    /* get Buffer Frame Size */
-    status = coreaudio_get_framesize(core->outputDeviceID,
-                                     &core->audioDevicePropertyBufferFrameSize);
+    propertySize = sizeof(core->audioDevicePropertyBufferFrameSize);
+    status = AudioDeviceGetProperty(
+        core->outputDeviceID,
+        0,
+        false,
+        kAudioDevicePropertyBufferFrameSize,
+        &propertySize,
+        &core->audioDevicePropertyBufferFrameSize);
    if (status != kAudioHardwareNoError) {
        coreaudio_logerr2 (status, typ,
                           "Could not get device buffer frame size\n");
        return -1;
    }
-    hw->samples = conf->nbuffers * core->audioDevicePropertyBufferFrameSize;
+    hw->samples = conf.nbuffers * core->audioDevicePropertyBufferFrameSize;

    /* get StreamFormat */
-    status = coreaudio_get_streamformat(core->outputDeviceID,
-                                        &core->outputStreamBasicDescription);
+    propertySize = sizeof(core->outputStreamBasicDescription);
+    status = AudioDeviceGetProperty(
+        core->outputDeviceID,
+        0,
+        false,
+        kAudioDevicePropertyStreamFormat,
+        &propertySize,
+        &core->outputStreamBasicDescription);
    if (status != kAudioHardwareNoError) {
        coreaudio_logerr2 (status, typ,
                           "Could not get Device Stream properties\n");
@@ -589,8 +399,15 @@ static int coreaudio_init_out(HWVoiceOut *hw, struct audsettings *as,

    /* set Samplerate */
    core->outputStreamBasicDescription.mSampleRate = (Float64) as->freq;
-    status = coreaudio_set_streamformat(core->outputDeviceID,
-                                        &core->outputStreamBasicDescription);
+    propertySize = sizeof(core->outputStreamBasicDescription);
+    status = AudioDeviceSetProperty(
+        core->outputDeviceID,
+        0,
+        0,
+        0,
+        kAudioDevicePropertyStreamFormat,
+        propertySize,
+        &core->outputStreamBasicDescription);
    if (status != kAudioHardwareNoError) {
        coreaudio_logerr2 (status, typ, "Could not set samplerate %d\n",
                           as->freq);
@@ -599,12 +416,8 @@ static int coreaudio_init_out(HWVoiceOut *hw, struct audsettings *as,
    }

    /* set Callback */
-    core->ioprocid = NULL;
-    status = AudioDeviceCreateIOProcID(core->outputDeviceID,
-                                       audioDeviceIOProc,
-                                       hw,
-                                       &core->ioprocid);
-    if (status != kAudioHardwareNoError || core->ioprocid == NULL) {
+    status = AudioDeviceAddIOProc(core->outputDeviceID, audioDeviceIOProc, hw);
+    if (status != kAudioHardwareNoError) {
        coreaudio_logerr2 (status, typ, "Could not set IOProc\n");
        core->outputDeviceID = kAudioDeviceUnknown;
        return -1;
@@ -612,10 +425,10 @@ static int coreaudio_init_out(HWVoiceOut *hw, struct audsettings *as,

    /* start Playback */
    if (!isPlaying(core->outputDeviceID)) {
-        status = AudioDeviceStart(core->outputDeviceID, core->ioprocid);
+        status = AudioDeviceStart(core->outputDeviceID, audioDeviceIOProc);
        if (status != kAudioHardwareNoError) {
            coreaudio_logerr2 (status, typ, "Could not start playback\n");
-            AudioDeviceDestroyIOProcID(core->outputDeviceID, core->ioprocid);
+            AudioDeviceRemoveIOProc(core->outputDeviceID, audioDeviceIOProc);
            core->outputDeviceID = kAudioDeviceUnknown;
            return -1;
        }
@@ -630,18 +443,18 @@ static void coreaudio_fini_out (HWVoiceOut *hw)
    int err;
    coreaudioVoiceOut *core = (coreaudioVoiceOut *) hw;

-    if (!isAtexit) {
+    if (!conf.isAtexit) {
        /* stop playback */
        if (isPlaying(core->outputDeviceID)) {
-            status = AudioDeviceStop(core->outputDeviceID, core->ioprocid);
+            status = AudioDeviceStop(core->outputDeviceID, audioDeviceIOProc);
            if (status != kAudioHardwareNoError) {
                coreaudio_logerr (status, "Could not stop playback\n");
            }
        }

        /* remove callback */
-        status = AudioDeviceDestroyIOProcID(core->outputDeviceID,
-                                            core->ioprocid);
+        status = AudioDeviceRemoveIOProc(core->outputDeviceID,
+                                         audioDeviceIOProc);
        if (status != kAudioHardwareNoError) {
            coreaudio_logerr (status, "Could not remove IOProc\n");
        }
@@ -664,7 +477,7 @@ static int coreaudio_ctl_out (HWVoiceOut *hw, int cmd, ...)
    case VOICE_ENABLE:
        /* start playback */
        if (!isPlaying(core->outputDeviceID)) {
-            status = AudioDeviceStart(core->outputDeviceID, core->ioprocid);
+            status = AudioDeviceStart(core->outputDeviceID, audioDeviceIOProc);
            if (status != kAudioHardwareNoError) {
                coreaudio_logerr (status, "Could not resume playback\n");
            }
@@ -673,10 +486,9 @@ static int coreaudio_ctl_out (HWVoiceOut *hw, int cmd, ...)

    case VOICE_DISABLE:
        /* stop playback */
-        if (!isAtexit) {
+        if (!conf.isAtexit) {
            if (isPlaying(core->outputDeviceID)) {
-                status = AudioDeviceStop(core->outputDeviceID,
-                                         core->ioprocid);
+                status = AudioDeviceStop(core->outputDeviceID, audioDeviceIOProc);
                if (status != kAudioHardwareNoError) {
                    coreaudio_logerr (status, "Could not pause playback\n");
                }
@@ -687,36 +499,28 @@ static int coreaudio_ctl_out (HWVoiceOut *hw, int cmd, ...)
    return 0;
 }

-static CoreaudioConf glob_conf = {
-    .buffer_frames = 512,
-    .nbuffers = 4,
-};
-
 static void *coreaudio_audio_init (void)
 {
-    CoreaudioConf *conf = g_malloc(sizeof(CoreaudioConf));
-    *conf = glob_conf;
-
    atexit(coreaudio_atexit);
-    return conf;
+    return &coreaudio_audio_init;
 }

 static void coreaudio_audio_fini (void *opaque)
 {
-    g_free(opaque);
+    (void) opaque;
 }

 static struct audio_option coreaudio_options[] = {
    {
        .name  = "BUFFER_SIZE",
        .tag   = AUD_OPT_INT,
-        .valp  = &glob_conf.buffer_frames,
+        .valp  = &conf.buffer_frames,
        .descr = "Size of the buffer in frames"
    },
    {
        .name  = "BUFFER_COUNT",
        .tag   = AUD_OPT_INT,
-        .valp  = &glob_conf.nbuffers,
+        .valp  = &conf.nbuffers,
        .descr = "Number of buffers"
    },
    { /* End of list */ }
--- a/audio/dsound_template.h
+++ b/audio/dsound_template.h
@@ -67,11 +67,11 @@ static int glue (dsound_lock_, TYPE) (
    LPVOID *p2p,
    DWORD *blen1p,
    DWORD *blen2p,
-    int entire,
-    dsound *s
+    int entire
    )
 {
    HRESULT hr;
+    int i;
    LPVOID p1 = NULL, p2 = NULL;
    DWORD blen1 = 0, blen2 = 0;
    DWORD flag;
@@ -81,18 +81,37 @@ static int glue (dsound_lock_, TYPE) (
 #else
    flag = entire ? DSBLOCK_ENTIREBUFFER : 0;
 #endif
-    hr = glue(IFACE, _Lock)(buf, pos, len, &p1, &blen1, &p2, &blen2, flag);
+    for (i = 0; i < conf.lock_retries; ++i) {
+        hr = glue (IFACE, _Lock) (
+            buf,
+            pos,
+            len,
+            &p1,
+            &blen1,
+            &p2,
+            &blen2,
+            flag
+            );

-    if (FAILED (hr)) {
+        if (FAILED (hr)) {
 #ifndef DSBTYPE_IN
-        if (hr == DSERR_BUFFERLOST) {
-            if (glue (dsound_restore_, TYPE) (buf, s)) {
-                dsound_logerr (hr, "Could not lock " NAME "\n");
+            if (hr == DSERR_BUFFERLOST) {
+                if (glue (dsound_restore_, TYPE) (buf)) {
+                    dsound_logerr (hr, "Could not lock " NAME "\n");
+                    goto fail;
+                }
+                continue;
            }
+#endif
+            dsound_logerr (hr, "Could not lock " NAME "\n");
            goto fail;
        }
-#endif
-        dsound_logerr (hr, "Could not lock " NAME "\n");
+
+        break;
+    }
+
+    if (i == conf.lock_retries) {
+        dolog ("%d attempts to lock " NAME " failed\n", i);
        goto fail;
    }

@@ -155,19 +174,16 @@ static void dsound_fini_out (HWVoiceOut *hw)
 }

 #ifdef DSBTYPE_IN
-static int dsound_init_in(HWVoiceIn *hw, struct audsettings *as,
-                          void *drv_opaque)
+static int dsound_init_in (HWVoiceIn *hw, struct audsettings *as)
 #else
-static int dsound_init_out(HWVoiceOut *hw, struct audsettings *as,
-                           void *drv_opaque)
+static int dsound_init_out (HWVoiceOut *hw, struct audsettings *as)
 #endif
 {
    int err;
    HRESULT hr;
-    dsound *s = drv_opaque;
+    dsound *s = &glob_dsound;
    WAVEFORMATEX wfx;
    struct audsettings obt_as;
-    DSoundConf *conf = &s->conf;
 #ifdef DSBTYPE_IN
    const char *typ = "ADC";
    DSoundVoiceIn *ds = (DSoundVoiceIn *) hw;
@@ -194,7 +210,7 @@ static int dsound_init_out(HWVoiceOut *hw, struct audsettings *as,
    bd.dwSize = sizeof (bd);
    bd.lpwfxFormat = &wfx;
 #ifdef DSBTYPE_IN
-    bd.dwBufferBytes = conf->bufsize_in;
+    bd.dwBufferBytes = conf.bufsize_in;
    hr = IDirectSoundCapture_CreateCaptureBuffer (
        s->dsound_capture,
        &bd,
@@ -203,7 +219,7 @@ static int dsound_init_out(HWVoiceOut *hw, struct audsettings *as,
        );
 #else
    bd.dwFlags = DSBCAPS_STICKYFOCUS | DSBCAPS_GETCURRENTPOSITION2;
-    bd.dwBufferBytes = conf->bufsize_out;
+    bd.dwBufferBytes = conf.bufsize_out;
    hr = IDirectSound_CreateSoundBuffer (
        s->dsound,
        &bd,
@@ -253,7 +269,6 @@ static int dsound_init_out(HWVoiceOut *hw, struct audsettings *as,
            );
    }
    hw->samples = bc.dwBufferBytes >> hw->info.shift;
-    ds->s = s;

 #ifdef DEBUG_DSOUND
    dolog ("caps %ld, desc %ld\n",
--- a/audio/dsoundaudio.c
+++ b/audio/dsoundaudio.c
@@ -26,7 +26,6 @@
 * SEAL 1.07 by Carlos 'pel' Hasan was used as documentation
 */

-#include "qemu/osdep.h"
 #include "qemu-common.h"
 #include "audio.h"

@@ -42,25 +41,42 @@

 /* #define DEBUG_DSOUND */

-typedef struct {
+static struct {
+    int lock_retries;
+    int restore_retries;
+    int getstatus_retries;
+    int set_primary;
    int bufsize_in;
    int bufsize_out;
+    struct audsettings settings;
    int latency_millis;
-} DSoundConf;
+} conf = {
+    .lock_retries       = 1,
+    .restore_retries    = 1,
+    .getstatus_retries  = 1,
+    .set_primary        = 0,
+    .bufsize_in         = 16384,
+    .bufsize_out        = 16384,
+    .settings.freq      = 44100,
+    .settings.nchannels = 2,
+    .settings.fmt       = AUD_FMT_S16,
+    .latency_millis     = 10
+};

 typedef struct {
    LPDIRECTSOUND dsound;
    LPDIRECTSOUNDCAPTURE dsound_capture;
+    LPDIRECTSOUNDBUFFER dsound_primary_buffer;
    struct audsettings settings;
-    DSoundConf conf;
 } dsound;

+static dsound glob_dsound;
+
 typedef struct {
    HWVoiceOut hw;
    LPDIRECTSOUNDBUFFER dsound_buffer;
    DWORD old_pos;
    int first_time;
-    dsound *s;
 #ifdef DEBUG_DSOUND
    DWORD old_ppos;
    DWORD played;
@@ -72,7 +88,6 @@ typedef struct {
    HWVoiceIn hw;
    int first_time;
    LPDIRECTSOUNDCAPTUREBUFFER dsound_capture_buffer;
-    dsound *s;
 } DSoundVoiceIn;

 static void dsound_log_hresult (HRESULT hr)
@@ -266,17 +281,29 @@ static void print_wave_format (WAVEFORMATEX *wfx)
 }
 #endif

-static int dsound_restore_out (LPDIRECTSOUNDBUFFER dsb, dsound *s)
+static int dsound_restore_out (LPDIRECTSOUNDBUFFER dsb)
 {
    HRESULT hr;
+    int i;

-    hr = IDirectSoundBuffer_Restore (dsb);
+    for (i = 0; i < conf.restore_retries; ++i) {
+        hr = IDirectSoundBuffer_Restore (dsb);

-    if (hr != DS_OK) {
-        dsound_logerr (hr, "Could not restore playback buffer\n");
-        return -1;
+        switch (hr) {
+        case DS_OK:
+            return 0;
+
+        case DSERR_BUFFERLOST:
+            continue;
+
+        default:
+            dsound_logerr (hr, "Could not restore playback buffer\n");
+            return -1;
+        }
    }
-    return 0;
+
+    dolog ("%d attempts to restore playback buffer failed\n", i);
+    return -1;
 }

 #include "dsound_template.h"
@@ -284,20 +311,25 @@ static int dsound_restore_out (LPDIRECTSOUNDBUFFER dsb, dsound *s)
 #include "dsound_template.h"
 #undef DSBTYPE_IN

-static int dsound_get_status_out (LPDIRECTSOUNDBUFFER dsb, DWORD *statusp,
-                                  dsound *s)
+static int dsound_get_status_out (LPDIRECTSOUNDBUFFER dsb, DWORD *statusp)
 {
    HRESULT hr;
+    int i;

-    hr = IDirectSoundBuffer_GetStatus (dsb, statusp);
-    if (FAILED (hr)) {
-        dsound_logerr (hr, "Could not get playback buffer status\n");
-        return -1;
-    }
+    for (i = 0; i < conf.getstatus_retries; ++i) {
+        hr = IDirectSoundBuffer_GetStatus (dsb, statusp);
+        if (FAILED (hr)) {
+            dsound_logerr (hr, "Could not get playback buffer status\n");
+            return -1;
+        }

-    if (*statusp & DSERR_BUFFERLOST) {
-        dsound_restore_out(dsb, s);
-        return -1;
+        if (*statusp & DSERR_BUFFERLOST) {
+            if (dsound_restore_out (dsb)) {
+                return -1;
+            }
+            continue;
+        }
+        break;
    }

    return 0;
@@ -344,8 +376,7 @@ static void dsound_write_sample (HWVoiceOut *hw, uint8_t *dst, int dst_len)
    hw->rpos = pos % hw->samples;
 }

-static void dsound_clear_sample (HWVoiceOut *hw, LPDIRECTSOUNDBUFFER dsb,
-                                 dsound *s)
+static void dsound_clear_sample (HWVoiceOut *hw, LPDIRECTSOUNDBUFFER dsb)
 {
    int err;
    LPVOID p1, p2;
@@ -358,8 +389,7 @@ static void dsound_clear_sample (HWVoiceOut *hw, LPDIRECTSOUNDBUFFER dsb,
        hw->samples << hw->info.shift,
        &p1, &p2,
        &blen1, &blen2,
-        1,
-        s
+        1
        );
    if (err) {
        return;
@@ -385,9 +415,25 @@ static void dsound_clear_sample (HWVoiceOut *hw, LPDIRECTSOUNDBUFFER dsb,
    dsound_unlock_out (dsb, p1, p2, blen1, blen2);
 }

-static int dsound_open (dsound *s)
+static void dsound_close (dsound *s)
 {
    HRESULT hr;
+
+    if (s->dsound_primary_buffer) {
+        hr = IDirectSoundBuffer_Release (s->dsound_primary_buffer);
+        if (FAILED (hr)) {
+            dsound_logerr (hr, "Could not release primary buffer\n");
+        }
+        s->dsound_primary_buffer = NULL;
+    }
+}
+
+static int dsound_open (dsound *s)
+{
+    int err;
+    HRESULT hr;
+    WAVEFORMATEX wfx;
+    DSBUFFERDESC dsbd;
    HWND hwnd;

    hwnd = GetForegroundWindow ();
@@ -403,7 +449,63 @@ static int dsound_open (dsound *s)
        return -1;
    }

+    if (!conf.set_primary) {
+        return 0;
+    }
+
+    err = waveformat_from_audio_settings (&wfx, &conf.settings);
+    if (err) {
+        return -1;
+    }
+
+    memset (&dsbd, 0, sizeof (dsbd));
+    dsbd.dwSize = sizeof (dsbd);
+    dsbd.dwFlags = DSBCAPS_PRIMARYBUFFER;
+    dsbd.dwBufferBytes = 0;
+    dsbd.lpwfxFormat = NULL;
+
+    hr = IDirectSound_CreateSoundBuffer (
+        s->dsound,
+        &dsbd,
+        &s->dsound_primary_buffer,
+        NULL
+        );
+    if (FAILED (hr)) {
+        dsound_logerr (hr, "Could not create primary playback buffer\n");
+        return -1;
+    }
+
+    hr = IDirectSoundBuffer_SetFormat (s->dsound_primary_buffer, &wfx);
+    if (FAILED (hr)) {
+        dsound_logerr (hr, "Could not set primary playback buffer format\n");
+    }
+
+    hr = IDirectSoundBuffer_GetFormat (
+        s->dsound_primary_buffer,
+        &wfx,
+        sizeof (wfx),
+        NULL
+        );
+    if (FAILED (hr)) {
+        dsound_logerr (hr, "Could not get primary playback buffer format\n");
+        goto fail0;
+    }
+
+#ifdef DEBUG_DSOUND
+    dolog ("Primary\n");
+    print_wave_format (&wfx);
+#endif
+
+    err = waveformat_to_audio_settings (&wfx, &s->settings);
+    if (err) {
+        goto fail0;
+    }
+
    return 0;
+
+ fail0:
+    dsound_close (s);
+    return -1;
 }

 static int dsound_ctl_out (HWVoiceOut *hw, int cmd, ...)
@@ -412,7 +514,6 @@ static int dsound_ctl_out (HWVoiceOut *hw, int cmd, ...)
    DWORD status;
    DSoundVoiceOut *ds = (DSoundVoiceOut *) hw;
    LPDIRECTSOUNDBUFFER dsb = ds->dsound_buffer;
-    dsound *s = ds->s;

    if (!dsb) {
        dolog ("Attempt to control voice without a buffer\n");
@@ -421,7 +522,7 @@ static int dsound_ctl_out (HWVoiceOut *hw, int cmd, ...)

    switch (cmd) {
    case VOICE_ENABLE:
-        if (dsound_get_status_out (dsb, &status, s)) {
+        if (dsound_get_status_out (dsb, &status)) {
            return -1;
        }

@@ -430,7 +531,7 @@ static int dsound_ctl_out (HWVoiceOut *hw, int cmd, ...)
            return 0;
        }

-        dsound_clear_sample (hw, dsb, s);
+        dsound_clear_sample (hw, dsb);

        hr = IDirectSoundBuffer_Play (dsb, 0, 0, DSBPLAY_LOOPING);
        if (FAILED (hr)) {
@@ -440,7 +541,7 @@ static int dsound_ctl_out (HWVoiceOut *hw, int cmd, ...)
        break;

    case VOICE_DISABLE:
-        if (dsound_get_status_out (dsb, &status, s)) {
+        if (dsound_get_status_out (dsb, &status)) {
            return -1;
        }

@@ -477,8 +578,6 @@ static int dsound_run_out (HWVoiceOut *hw, int live)
    DWORD wpos, ppos, old_pos;
    LPVOID p1, p2;
    int bufsize;
-    dsound *s = ds->s;
-    DSoundConf *conf = &s->conf;

    if (!dsb) {
        dolog ("Attempt to run empty with playback buffer\n");
@@ -501,14 +600,14 @@ static int dsound_run_out (HWVoiceOut *hw, int live)
    len = live << hwshift;

    if (ds->first_time) {
-        if (conf->latency_millis) {
+        if (conf.latency_millis) {
            DWORD cur_blat;

            cur_blat = audio_ring_dist (wpos, ppos, bufsize);
            ds->first_time = 0;
            old_pos = wpos;
            old_pos +=
-                millis_to_bytes (&hw->info, conf->latency_millis) - cur_blat;
+                millis_to_bytes (&hw->info, conf.latency_millis) - cur_blat;
            old_pos %= bufsize;
            old_pos &= ~hw->info.align;
        }
@@ -564,8 +663,7 @@ static int dsound_run_out (HWVoiceOut *hw, int live)
        len,
        &p1, &p2,
        &blen1, &blen2,
-        0,
-        s
+        0
        );
    if (err) {
        return 0;
@@ -668,7 +766,6 @@ static int dsound_run_in (HWVoiceIn *hw)
    DWORD cpos, rpos;
    LPVOID p1, p2;
    int hwshift;
-    dsound *s = ds->s;

    if (!dscb) {
        dolog ("Attempt to run without capture buffer\n");
@@ -723,8 +820,7 @@ static int dsound_run_in (HWVoiceIn *hw)
        &p2,
        &blen1,
        &blen2,
-        0,
-        s
+        0
        );
    if (err) {
        return 0;
@@ -747,19 +843,12 @@ static int dsound_run_in (HWVoiceIn *hw)
    return decr;
 }

-static DSoundConf glob_conf = {
-    .bufsize_in         = 16384,
-    .bufsize_out        = 16384,
-    .latency_millis     = 10
-};
-
 static void dsound_audio_fini (void *opaque)
 {
    HRESULT hr;
    dsound *s = opaque;

    if (!s->dsound) {
-        g_free(s);
        return;
    }

@@ -770,7 +859,6 @@ static void dsound_audio_fini (void *opaque)
    s->dsound = NULL;

    if (!s->dsound_capture) {
-        g_free(s);
        return;
    }

@@ -779,21 +867,17 @@ static void dsound_audio_fini (void *opaque)
        dsound_logerr (hr, "Could not release DirectSoundCapture\n");
    }
    s->dsound_capture = NULL;
-
-    g_free(s);
 }

 static void *dsound_audio_init (void)
 {
    int err;
    HRESULT hr;
-    dsound *s = g_malloc0(sizeof(dsound));
+    dsound *s = &glob_dsound;

-    s->conf = glob_conf;
    hr = CoInitialize (NULL);
    if (FAILED (hr)) {
        dsound_logerr (hr, "Could not initialize COM\n");
-        g_free(s);
        return NULL;
    }

@@ -806,7 +890,6 @@ static void *dsound_audio_init (void)
        );
    if (FAILED (hr)) {
        dsound_logerr (hr, "Could not create DirectSound instance\n");
-        g_free(s);
        return NULL;
    }

@@ -818,7 +901,7 @@ static void *dsound_audio_init (void)
        if (FAILED (hr)) {
            dsound_logerr (hr, "Could not release DirectSound\n");
        }
-        g_free(s);
+        s->dsound = NULL;
        return NULL;
    }

@@ -855,22 +938,64 @@ static void *dsound_audio_init (void)
 }

 static struct audio_option dsound_options[] = {
+    {
+        .name  = "LOCK_RETRIES",
+        .tag   = AUD_OPT_INT,
+        .valp  = &conf.lock_retries,
+        .descr = "Number of times to attempt locking the buffer"
+    },
+    {
+        .name  = "RESTOURE_RETRIES",
+        .tag   = AUD_OPT_INT,
+        .valp  = &conf.restore_retries,
+        .descr = "Number of times to attempt restoring the buffer"
+    },
+    {
+        .name  = "GETSTATUS_RETRIES",
+        .tag   = AUD_OPT_INT,
+        .valp  = &conf.getstatus_retries,
+        .descr = "Number of times to attempt getting status of the buffer"
+    },
+    {
+        .name  = "SET_PRIMARY",
+        .tag   = AUD_OPT_BOOL,
+        .valp  = &conf.set_primary,
+        .descr = "Set the parameters of primary buffer"
+    },
    {
        .name  = "LATENCY_MILLIS",
        .tag   = AUD_OPT_INT,
-        .valp  = &glob_conf.latency_millis,
+        .valp  = &conf.latency_millis,
        .descr = "(undocumented)"
    },
+    {
+        .name  = "PRIMARY_FREQ",
+        .tag   = AUD_OPT_INT,
+        .valp  = &conf.settings.freq,
+        .descr = "Primary buffer frequency"
+    },
+    {
+        .name  = "PRIMARY_CHANNELS",
+        .tag   = AUD_OPT_INT,
+        .valp  = &conf.settings.nchannels,
+        .descr = "Primary buffer number of channels (1 - mono, 2 - stereo)"
+    },
+    {
+        .name  = "PRIMARY_FMT",
+        .tag   = AUD_OPT_FMT,
+        .valp  = &conf.settings.fmt,
+        .descr = "Primary buffer format"
+    },
    {
        .name  = "BUFSIZE_OUT",
        .tag   = AUD_OPT_INT,
-        .valp  = &glob_conf.bufsize_out,
+        .valp  = &conf.bufsize_out,
        .descr = "(undocumented)"
    },
    {
        .name  = "BUFSIZE_IN",
        .tag   = AUD_OPT_INT,
-        .valp  = &glob_conf.bufsize_in,
+        .valp  = &conf.bufsize_in,
        .descr = "(undocumented)"
    },
    { /* End of list */ }
--- a/audio/esdaudio.c
+++ b/audio/esdaudio.c
@@ -0,0 +1,557 @@
+/*
+ * QEMU ESD audio driver
+ *
+ * Copyright (c) 2006 Frederick Reeve (brushed up by malc)
+ *
+ * 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 <esd.h>
+#include "qemu-common.h"
+#include "audio.h"
+
+#define AUDIO_CAP "esd"
+#include "audio_int.h"
+#include "audio_pt_int.h"
+
+typedef struct {
+    HWVoiceOut hw;
+    int done;
+    int live;
+    int decr;
+    int rpos;
+    void *pcm_buf;
+    int fd;
+    struct audio_pt pt;
+} ESDVoiceOut;
+
+typedef struct {
+    HWVoiceIn hw;
+    int done;
+    int dead;
+    int incr;
+    int wpos;
+    void *pcm_buf;
+    int fd;
+    struct audio_pt pt;
+} ESDVoiceIn;
+
+static struct {
+    int samples;
+    int divisor;
+    char *dac_host;
+    char *adc_host;
+} conf = {
+    .samples = 1024,
+    .divisor = 2,
+};
+
+static void GCC_FMT_ATTR (2, 3) qesd_logerr (int err, const char *fmt, ...)
+{
+    va_list ap;
+
+    va_start (ap, fmt);
+    AUD_vlog (AUDIO_CAP, fmt, ap);
+    va_end (ap);
+
+    AUD_log (AUDIO_CAP, "Reason: %s\n", strerror (err));
+}
+
+/* playback */
+static void *qesd_thread_out (void *arg)
+{
+    ESDVoiceOut *esd = arg;
+    HWVoiceOut *hw = &esd->hw;
+    int threshold;
+
+    threshold = conf.divisor ? hw->samples / conf.divisor : 0;
+
+    if (audio_pt_lock (&esd->pt, AUDIO_FUNC)) {
+        return NULL;
+    }
+
+    for (;;) {
+        int decr, to_mix, rpos;
+
+        for (;;) {
+            if (esd->done) {
+                goto exit;
+            }
+
+            if (esd->live > threshold) {
+                break;
+            }
+
+            if (audio_pt_wait (&esd->pt, AUDIO_FUNC)) {
+                goto exit;
+            }
+        }
+
+        decr = to_mix = esd->live;
+        rpos = hw->rpos;
+
+        if (audio_pt_unlock (&esd->pt, AUDIO_FUNC)) {
+            return NULL;
+        }
+
+        while (to_mix) {
+            ssize_t written;
+            int chunk = audio_MIN (to_mix, hw->samples - rpos);
+            struct st_sample *src = hw->mix_buf + rpos;
+
+            hw->clip (esd->pcm_buf, src, chunk);
+
+        again:
+            written = write (esd->fd, esd->pcm_buf, chunk << hw->info.shift);
+            if (written == -1) {
+                if (errno == EINTR || errno == EAGAIN) {
+                    goto again;
+                }
+                qesd_logerr (errno, "write failed\n");
+                return NULL;
+            }
+
+            if (written != chunk << hw->info.shift) {
+                int wsamples = written >> hw->info.shift;
+                int wbytes = wsamples << hw->info.shift;
+                if (wbytes != written) {
+                    dolog ("warning: Misaligned write %d (requested %zd), "
+                           "alignment %d\n",
+                           wbytes, written, hw->info.align + 1);
+                }
+                to_mix -= wsamples;
+                rpos = (rpos + wsamples) % hw->samples;
+                break;
+            }
+
+            rpos = (rpos + chunk) % hw->samples;
+            to_mix -= chunk;
+        }
+
+        if (audio_pt_lock (&esd->pt, AUDIO_FUNC)) {
+            return NULL;
+        }
+
+        esd->rpos = rpos;
+        esd->live -= decr;
+        esd->decr += decr;
+    }
+
+ exit:
+    audio_pt_unlock (&esd->pt, AUDIO_FUNC);
+    return NULL;
+}
+
+static int qesd_run_out (HWVoiceOut *hw, int live)
+{
+    int decr;
+    ESDVoiceOut *esd = (ESDVoiceOut *) hw;
+
+    if (audio_pt_lock (&esd->pt, AUDIO_FUNC)) {
+        return 0;
+    }
+
+    decr = audio_MIN (live, esd->decr);
+    esd->decr -= decr;
+    esd->live = live - decr;
+    hw->rpos = esd->rpos;
+    if (esd->live > 0) {
+        audio_pt_unlock_and_signal (&esd->pt, AUDIO_FUNC);
+    }
+    else {
+        audio_pt_unlock (&esd->pt, AUDIO_FUNC);
+    }
+    return decr;
+}
+
+static int qesd_write (SWVoiceOut *sw, void *buf, int len)
+{
+    return audio_pcm_sw_write (sw, buf, len);
+}
+
+static int qesd_init_out (HWVoiceOut *hw, struct audsettings *as)
+{
+    ESDVoiceOut *esd = (ESDVoiceOut *) hw;
+    struct audsettings obt_as = *as;
+    int esdfmt = ESD_STREAM | ESD_PLAY;
+
+    esdfmt |= (as->nchannels == 2) ? ESD_STEREO : ESD_MONO;
+    switch (as->fmt) {
+    case AUD_FMT_S8:
+    case AUD_FMT_U8:
+        esdfmt |= ESD_BITS8;
+        obt_as.fmt = AUD_FMT_U8;
+        break;
+
+    case AUD_FMT_S32:
+    case AUD_FMT_U32:
+        dolog ("Will use 16 instead of 32 bit samples\n");
+        /* fall through */
+    case AUD_FMT_S16:
+    case AUD_FMT_U16:
+    deffmt:
+        esdfmt |= ESD_BITS16;
+        obt_as.fmt = AUD_FMT_S16;
+        break;
+
+    default:
+        dolog ("Internal logic error: Bad audio format %d\n", as->fmt);
+        goto deffmt;
+
+    }
+    obt_as.endianness = AUDIO_HOST_ENDIANNESS;
+
+    audio_pcm_init_info (&hw->info, &obt_as);
+
+    hw->samples = conf.samples;
+    esd->pcm_buf = audio_calloc (AUDIO_FUNC, hw->samples, 1 << hw->info.shift);
+    if (!esd->pcm_buf) {
+        dolog ("Could not allocate buffer (%d bytes)\n",
+               hw->samples << hw->info.shift);
+        return -1;
+    }
+
+    esd->fd = esd_play_stream (esdfmt, as->freq, conf.dac_host, NULL);
+    if (esd->fd < 0) {
+        qesd_logerr (errno, "esd_play_stream failed\n");
+        goto fail1;
+    }
+
+    if (audio_pt_init (&esd->pt, qesd_thread_out, esd, AUDIO_CAP, AUDIO_FUNC)) {
+        goto fail2;
+    }
+
+    return 0;
+
+ fail2:
+    if (close (esd->fd)) {
+        qesd_logerr (errno, "%s: close on esd socket(%d) failed\n",
+                     AUDIO_FUNC, esd->fd);
+    }
+    esd->fd = -1;
+
+ fail1:
+    g_free (esd->pcm_buf);
+    esd->pcm_buf = NULL;
+    return -1;
+}
+
+static void qesd_fini_out (HWVoiceOut *hw)
+{
+    void *ret;
+    ESDVoiceOut *esd = (ESDVoiceOut *) hw;
+
+    audio_pt_lock (&esd->pt, AUDIO_FUNC);
+    esd->done = 1;
+    audio_pt_unlock_and_signal (&esd->pt, AUDIO_FUNC);
+    audio_pt_join (&esd->pt, &ret, AUDIO_FUNC);
+
+    if (esd->fd >= 0) {
+        if (close (esd->fd)) {
+            qesd_logerr (errno, "failed to close esd socket\n");
+        }
+        esd->fd = -1;
+    }
+
+    audio_pt_fini (&esd->pt, AUDIO_FUNC);
+
+    g_free (esd->pcm_buf);
+    esd->pcm_buf = NULL;
+}
+
+static int qesd_ctl_out (HWVoiceOut *hw, int cmd, ...)
+{
+    (void) hw;
+    (void) cmd;
+    return 0;
+}
+
+/* capture */
+static void *qesd_thread_in (void *arg)
+{
+    ESDVoiceIn *esd = arg;
+    HWVoiceIn *hw = &esd->hw;
+    int threshold;
+
+    threshold = conf.divisor ? hw->samples / conf.divisor : 0;
+
+    if (audio_pt_lock (&esd->pt, AUDIO_FUNC)) {
+        return NULL;
+    }
+
+    for (;;) {
+        int incr, to_grab, wpos;
+
+        for (;;) {
+            if (esd->done) {
+                goto exit;
+            }
+
+            if (esd->dead > threshold) {
+                break;
+            }
+
+            if (audio_pt_wait (&esd->pt, AUDIO_FUNC)) {
+                goto exit;
+            }
+        }
+
+        incr = to_grab = esd->dead;
+        wpos = hw->wpos;
+
+        if (audio_pt_unlock (&esd->pt, AUDIO_FUNC)) {
+            return NULL;
+        }
+
+        while (to_grab) {
+            ssize_t nread;
+            int chunk = audio_MIN (to_grab, hw->samples - wpos);
+            void *buf = advance (esd->pcm_buf, wpos);
+
+        again:
+            nread = read (esd->fd, buf, chunk << hw->info.shift);
+            if (nread == -1) {
+                if (errno == EINTR || errno == EAGAIN) {
+                    goto again;
+                }
+                qesd_logerr (errno, "read failed\n");
+                return NULL;
+            }
+
+            if (nread != chunk << hw->info.shift) {
+                int rsamples = nread >> hw->info.shift;
+                int rbytes = rsamples << hw->info.shift;
+                if (rbytes != nread) {
+                    dolog ("warning: Misaligned write %d (requested %zd), "
+                           "alignment %d\n",
+                           rbytes, nread, hw->info.align + 1);
+                }
+                to_grab -= rsamples;
+                wpos = (wpos + rsamples) % hw->samples;
+                break;
+            }
+
+            hw->conv (hw->conv_buf + wpos, buf, nread >> hw->info.shift);
+            wpos = (wpos + chunk) % hw->samples;
+            to_grab -= chunk;
+        }
+
+        if (audio_pt_lock (&esd->pt, AUDIO_FUNC)) {
+            return NULL;
+        }
+
+        esd->wpos = wpos;
+        esd->dead -= incr;
+        esd->incr += incr;
+    }
+
+ exit:
+    audio_pt_unlock (&esd->pt, AUDIO_FUNC);
+    return NULL;
+}
+
+static int qesd_run_in (HWVoiceIn *hw)
+{
+    int live, incr, dead;
+    ESDVoiceIn *esd = (ESDVoiceIn *) hw;
+
+    if (audio_pt_lock (&esd->pt, AUDIO_FUNC)) {
+        return 0;
+    }
+
+    live = audio_pcm_hw_get_live_in (hw);
+    dead = hw->samples - live;
+    incr = audio_MIN (dead, esd->incr);
+    esd->incr -= incr;
+    esd->dead = dead - incr;
+    hw->wpos = esd->wpos;
+    if (esd->dead > 0) {
+        audio_pt_unlock_and_signal (&esd->pt, AUDIO_FUNC);
+    }
+    else {
+        audio_pt_unlock (&esd->pt, AUDIO_FUNC);
+    }
+    return incr;
+}
+
+static int qesd_read (SWVoiceIn *sw, void *buf, int len)
+{
+    return audio_pcm_sw_read (sw, buf, len);
+}
+
+static int qesd_init_in (HWVoiceIn *hw, struct audsettings *as)
+{
+    ESDVoiceIn *esd = (ESDVoiceIn *) hw;
+    struct audsettings obt_as = *as;
+    int esdfmt = ESD_STREAM | ESD_RECORD;
+
+    esdfmt |= (as->nchannels == 2) ? ESD_STEREO : ESD_MONO;
+    switch (as->fmt) {
+    case AUD_FMT_S8:
+    case AUD_FMT_U8:
+        esdfmt |= ESD_BITS8;
+        obt_as.fmt = AUD_FMT_U8;
+        break;
+
+    case AUD_FMT_S16:
+    case AUD_FMT_U16:
+        esdfmt |= ESD_BITS16;
+        obt_as.fmt = AUD_FMT_S16;
+        break;
+
+    case AUD_FMT_S32:
+    case AUD_FMT_U32:
+        dolog ("Will use 16 instead of 32 bit samples\n");
+        esdfmt |= ESD_BITS16;
+        obt_as.fmt = AUD_FMT_S16;
+        break;
+    }
+    obt_as.endianness = AUDIO_HOST_ENDIANNESS;
+
+    audio_pcm_init_info (&hw->info, &obt_as);
+
+    hw->samples = conf.samples;
+    esd->pcm_buf = audio_calloc (AUDIO_FUNC, hw->samples, 1 << hw->info.shift);
+    if (!esd->pcm_buf) {
+        dolog ("Could not allocate buffer (%d bytes)\n",
+               hw->samples << hw->info.shift);
+        return -1;
+    }
+
+    esd->fd = esd_record_stream (esdfmt, as->freq, conf.adc_host, NULL);
+    if (esd->fd < 0) {
+        qesd_logerr (errno, "esd_record_stream failed\n");
+        goto fail1;
+    }
+
+    if (audio_pt_init (&esd->pt, qesd_thread_in, esd, AUDIO_CAP, AUDIO_FUNC)) {
+        goto fail2;
+    }
+
+    return 0;
+
+ fail2:
+    if (close (esd->fd)) {
+        qesd_logerr (errno, "%s: close on esd socket(%d) failed\n",
+                     AUDIO_FUNC, esd->fd);
+    }
+    esd->fd = -1;
+
+ fail1:
+    g_free (esd->pcm_buf);
+    esd->pcm_buf = NULL;
+    return -1;
+}
+
+static void qesd_fini_in (HWVoiceIn *hw)
+{
+    void *ret;
+    ESDVoiceIn *esd = (ESDVoiceIn *) hw;
+
+    audio_pt_lock (&esd->pt, AUDIO_FUNC);
+    esd->done = 1;
+    audio_pt_unlock_and_signal (&esd->pt, AUDIO_FUNC);
+    audio_pt_join (&esd->pt, &ret, AUDIO_FUNC);
+
+    if (esd->fd >= 0) {
+        if (close (esd->fd)) {
+            qesd_logerr (errno, "failed to close esd socket\n");
+        }
+        esd->fd = -1;
+    }
+
+    audio_pt_fini (&esd->pt, AUDIO_FUNC);
+
+    g_free (esd->pcm_buf);
+    esd->pcm_buf = NULL;
+}
+
+static int qesd_ctl_in (HWVoiceIn *hw, int cmd, ...)
+{
+    (void) hw;
+    (void) cmd;
+    return 0;
+}
+
+/* common */
+static void *qesd_audio_init (void)
+{
+    return &conf;
+}
+
+static void qesd_audio_fini (void *opaque)
+{
+    (void) opaque;
+    ldebug ("esd_fini");
+}
+
+struct audio_option qesd_options[] = {
+    {
+        .name  = "SAMPLES",
+        .tag   = AUD_OPT_INT,
+        .valp  = &conf.samples,
+        .descr = "buffer size in samples"
+    },
+    {
+        .name  = "DIVISOR",
+        .tag   = AUD_OPT_INT,
+        .valp  = &conf.divisor,
+        .descr = "threshold divisor"
+    },
+    {
+        .name  = "DAC_HOST",
+        .tag   = AUD_OPT_STR,
+        .valp  = &conf.dac_host,
+        .descr = "playback host"
+    },
+    {
+        .name  = "ADC_HOST",
+        .tag   = AUD_OPT_STR,
+        .valp  = &conf.adc_host,
+        .descr = "capture host"
+    },
+    { /* End of list */ }
+};
+
+static struct audio_pcm_ops qesd_pcm_ops = {
+    .init_out = qesd_init_out,
+    .fini_out = qesd_fini_out,
+    .run_out  = qesd_run_out,
+    .write    = qesd_write,
+    .ctl_out  = qesd_ctl_out,
+
+    .init_in  = qesd_init_in,
+    .fini_in  = qesd_fini_in,
+    .run_in   = qesd_run_in,
+    .read     = qesd_read,
+    .ctl_in   = qesd_ctl_in,
+};
+
+struct audio_driver esd_audio_driver = {
+    .name           = "esd",
+    .descr          = "http://en.wikipedia.org/wiki/Esound",
+    .options        = qesd_options,
+    .init           = qesd_audio_init,
+    .fini           = qesd_audio_fini,
+    .pcm_ops        = &qesd_pcm_ops,
+    .can_be_default = 0,
+    .max_voices_out = INT_MAX,
+    .max_voices_in  = INT_MAX,
+    .voice_size_out = sizeof (ESDVoiceOut),
+    .voice_size_in  = sizeof (ESDVoiceIn)
+};
--- a/audio/fmodaudio.c
+++ b/audio/fmodaudio.c
@@ -0,0 +1,685 @@
+/*
+ * QEMU FMOD audio driver
+ *
+ * Copyright (c) 2004-2005 Vassili Karpov (malc)
+ *
+ * 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 <fmod.h>
+#include <fmod_errors.h>
+#include "qemu-common.h"
+#include "audio.h"
+
+#define AUDIO_CAP "fmod"
+#include "audio_int.h"
+
+typedef struct FMODVoiceOut {
+    HWVoiceOut hw;
+    unsigned int old_pos;
+    FSOUND_SAMPLE *fmod_sample;
+    int channel;
+} FMODVoiceOut;
+
+typedef struct FMODVoiceIn {
+    HWVoiceIn hw;
+    FSOUND_SAMPLE *fmod_sample;
+} FMODVoiceIn;
+
+static struct {
+    const char *drvname;
+    int nb_samples;
+    int freq;
+    int nb_channels;
+    int bufsize;
+    int broken_adc;
+} conf = {
+    .nb_samples  = 2048 * 2,
+    .freq        = 44100,
+    .nb_channels = 2,
+};
+
+static void GCC_FMT_ATTR (1, 2) fmod_logerr (const char *fmt, ...)
+{
+    va_list ap;
+
+    va_start (ap, fmt);
+    AUD_vlog (AUDIO_CAP, fmt, ap);
+    va_end (ap);
+
+    AUD_log (AUDIO_CAP, "Reason: %s\n",
+             FMOD_ErrorString (FSOUND_GetError ()));
+}
+
+static void GCC_FMT_ATTR (2, 3) fmod_logerr2 (
+    const char *typ,
+    const char *fmt,
+    ...
+    )
+{
+    va_list ap;
+
+    AUD_log (AUDIO_CAP, "Could not initialize %s\n", typ);
+
+    va_start (ap, fmt);
+    AUD_vlog (AUDIO_CAP, fmt, ap);
+    va_end (ap);
+
+    AUD_log (AUDIO_CAP, "Reason: %s\n",
+             FMOD_ErrorString (FSOUND_GetError ()));
+}
+
+static int fmod_write (SWVoiceOut *sw, void *buf, int len)
+{
+    return audio_pcm_sw_write (sw, buf, len);
+}
+
+static void fmod_clear_sample (FMODVoiceOut *fmd)
+{
+    HWVoiceOut *hw = &fmd->hw;
+    int status;
+    void *p1 = 0, *p2 = 0;
+    unsigned int len1 = 0, len2 = 0;
+
+    status = FSOUND_Sample_Lock (
+        fmd->fmod_sample,
+        0,
+        hw->samples << hw->info.shift,
+        &p1,
+        &p2,
+        &len1,
+        &len2
+        );
+
+    if (!status) {
+        fmod_logerr ("Failed to lock sample\n");
+        return;
+    }
+
+    if ((len1 & hw->info.align) || (len2 & hw->info.align)) {
+        dolog ("Lock returned misaligned length %d, %d, alignment %d\n",
+               len1, len2, hw->info.align + 1);
+        goto fail;
+    }
+
+    if ((len1 + len2) - (hw->samples << hw->info.shift)) {
+        dolog ("Lock returned incomplete length %d, %d\n",
+               len1 + len2, hw->samples << hw->info.shift);
+        goto fail;
+    }
+
+    audio_pcm_info_clear_buf (&hw->info, p1, hw->samples);
+
+ fail:
+    status = FSOUND_Sample_Unlock (fmd->fmod_sample, p1, p2, len1, len2);
+    if (!status) {
+        fmod_logerr ("Failed to unlock sample\n");
+    }
+}
+
+static void fmod_write_sample (HWVoiceOut *hw, uint8_t *dst, int dst_len)
+{
+    int src_len1 = dst_len;
+    int src_len2 = 0;
+    int pos = hw->rpos + dst_len;
+    struct st_sample *src1 = hw->mix_buf + hw->rpos;
+    struct st_sample *src2 = NULL;
+
+    if (pos > hw->samples) {
+        src_len1 = hw->samples - hw->rpos;
+        src2 = hw->mix_buf;
+        src_len2 = dst_len - src_len1;
+        pos = src_len2;
+    }
+
+    if (src_len1) {
+        hw->clip (dst, src1, src_len1);
+    }
+
+    if (src_len2) {
+        dst = advance (dst, src_len1 << hw->info.shift);
+        hw->clip (dst, src2, src_len2);
+    }
+
+    hw->rpos = pos % hw->samples;
+}
+
+static int fmod_unlock_sample (FSOUND_SAMPLE *sample, void *p1, void *p2,
+                               unsigned int blen1, unsigned int blen2)
+{
+    int status = FSOUND_Sample_Unlock (sample, p1, p2, blen1, blen2);
+    if (!status) {
+        fmod_logerr ("Failed to unlock sample\n");
+        return -1;
+    }
+    return 0;
+}
+
+static int fmod_lock_sample (
+    FSOUND_SAMPLE *sample,
+    struct audio_pcm_info *info,
+    int pos,
+    int len,
+    void **p1,
+    void **p2,
+    unsigned int *blen1,
+    unsigned int *blen2
+    )
+{
+    int status;
+
+    status = FSOUND_Sample_Lock (
+        sample,
+        pos << info->shift,
+        len << info->shift,
+        p1,
+        p2,
+        blen1,
+        blen2
+        );
+
+    if (!status) {
+        fmod_logerr ("Failed to lock sample\n");
+        return -1;
+    }
+
+    if ((*blen1 & info->align) || (*blen2 & info->align)) {
+        dolog ("Lock returned misaligned length %d, %d, alignment %d\n",
+               *blen1, *blen2, info->align + 1);
+
+        fmod_unlock_sample (sample, *p1, *p2, *blen1, *blen2);
+
+        *p1 = NULL - 1;
+        *p2 = NULL - 1;
+        *blen1 = ~0U;
+        *blen2 = ~0U;
+        return -1;
+    }
+
+    if (!*p1 && *blen1) {
+        dolog ("warning: !p1 && blen1=%d\n", *blen1);
+        *blen1 = 0;
+    }
+
+    if (!p2 && *blen2) {
+        dolog ("warning: !p2 && blen2=%d\n", *blen2);
+        *blen2 = 0;
+    }
+
+    return 0;
+}
+
+static int fmod_run_out (HWVoiceOut *hw, int live)
+{
+    FMODVoiceOut *fmd = (FMODVoiceOut *) hw;
+    int decr;
+    void *p1 = 0, *p2 = 0;
+    unsigned int blen1 = 0, blen2 = 0;
+    unsigned int len1 = 0, len2 = 0;
+
+    if (!hw->pending_disable) {
+        return 0;
+    }
+
+    decr = live;
+
+    if (fmd->channel >= 0) {
+        int len = decr;
+        int old_pos = fmd->old_pos;
+        int ppos = FSOUND_GetCurrentPosition (fmd->channel);
+
+        if (ppos == old_pos || !ppos) {
+            return 0;
+        }
+
+        if ((old_pos < ppos) && ((old_pos + len) > ppos)) {
+            len = ppos - old_pos;
+        }
+        else {
+            if ((old_pos > ppos) && ((old_pos + len) > (ppos + hw->samples))) {
+                len = hw->samples - old_pos + ppos;
+            }
+        }
+        decr = len;
+
+        if (audio_bug (AUDIO_FUNC, decr < 0)) {
+            dolog ("decr=%d live=%d ppos=%d old_pos=%d len=%d\n",
+                   decr, live, ppos, old_pos, len);
+            return 0;
+        }
+    }
+
+
+    if (!decr) {
+        return 0;
+    }
+
+    if (fmod_lock_sample (fmd->fmod_sample, &fmd->hw.info,
+                          fmd->old_pos, decr,
+                          &p1, &p2,
+                          &blen1, &blen2)) {
+        return 0;
+    }
+
+    len1 = blen1 >> hw->info.shift;
+    len2 = blen2 >> hw->info.shift;
+    ldebug ("%p %p %d %d %d %d\n", p1, p2, len1, len2, blen1, blen2);
+    decr = len1 + len2;
+
+    if (p1 && len1) {
+        fmod_write_sample (hw, p1, len1);
+    }
+
+    if (p2 && len2) {
+        fmod_write_sample (hw, p2, len2);
+    }
+
+    fmod_unlock_sample (fmd->fmod_sample, p1, p2, blen1, blen2);
+
+    fmd->old_pos = (fmd->old_pos + decr) % hw->samples;
+    return decr;
+}
+
+static int aud_to_fmodfmt (audfmt_e fmt, int stereo)
+{
+    int mode = FSOUND_LOOP_NORMAL;
+
+    switch (fmt) {
+    case AUD_FMT_S8:
+        mode |= FSOUND_SIGNED | FSOUND_8BITS;
+        break;
+
+    case AUD_FMT_U8:
+        mode |= FSOUND_UNSIGNED | FSOUND_8BITS;
+        break;
+
+    case AUD_FMT_S16:
+        mode |= FSOUND_SIGNED | FSOUND_16BITS;
+        break;
+
+    case AUD_FMT_U16:
+        mode |= FSOUND_UNSIGNED | FSOUND_16BITS;
+        break;
+
+    default:
+        dolog ("Internal logic error: Bad audio format %d\n", fmt);
+#ifdef DEBUG_FMOD
+        abort ();
+#endif
+        mode |= FSOUND_8BITS;
+    }
+    mode |= stereo ? FSOUND_STEREO : FSOUND_MONO;
+    return mode;
+}
+
+static void fmod_fini_out (HWVoiceOut *hw)
+{
+    FMODVoiceOut *fmd = (FMODVoiceOut *) hw;
+
+    if (fmd->fmod_sample) {
+        FSOUND_Sample_Free (fmd->fmod_sample);
+        fmd->fmod_sample = 0;
+
+        if (fmd->channel >= 0) {
+            FSOUND_StopSound (fmd->channel);
+        }
+    }
+}
+
+static int fmod_init_out (HWVoiceOut *hw, struct audsettings *as)
+{
+    int mode, channel;
+    FMODVoiceOut *fmd = (FMODVoiceOut *) hw;
+    struct audsettings obt_as = *as;
+
+    mode = aud_to_fmodfmt (as->fmt, as->nchannels == 2 ? 1 : 0);
+    fmd->fmod_sample = FSOUND_Sample_Alloc (
+        FSOUND_FREE,            /* index */
+        conf.nb_samples,        /* length */
+        mode,                   /* mode */
+        as->freq,               /* freq */
+        255,                    /* volume */
+        128,                    /* pan */
+        255                     /* priority */
+        );
+
+    if (!fmd->fmod_sample) {
+        fmod_logerr2 ("DAC", "Failed to allocate FMOD sample\n");
+        return -1;
+    }
+
+    channel = FSOUND_PlaySoundEx (FSOUND_FREE, fmd->fmod_sample, 0, 1);
+    if (channel < 0) {
+        fmod_logerr2 ("DAC", "Failed to start playing sound\n");
+        FSOUND_Sample_Free (fmd->fmod_sample);
+        return -1;
+    }
+    fmd->channel = channel;
+
+    /* FMOD always operates on little endian frames? */
+    obt_as.endianness = 0;
+    audio_pcm_init_info (&hw->info, &obt_as);
+    hw->samples = conf.nb_samples;
+    return 0;
+}
+
+static int fmod_ctl_out (HWVoiceOut *hw, int cmd, ...)
+{
+    int status;
+    FMODVoiceOut *fmd = (FMODVoiceOut *) hw;
+
+    switch (cmd) {
+    case VOICE_ENABLE:
+        fmod_clear_sample (fmd);
+        status = FSOUND_SetPaused (fmd->channel, 0);
+        if (!status) {
+            fmod_logerr ("Failed to resume channel %d\n", fmd->channel);
+        }
+        break;
+
+    case VOICE_DISABLE:
+        status = FSOUND_SetPaused (fmd->channel, 1);
+        if (!status) {
+            fmod_logerr ("Failed to pause channel %d\n", fmd->channel);
+        }
+        break;
+    }
+    return 0;
+}
+
+static int fmod_init_in (HWVoiceIn *hw, struct audsettings *as)
+{
+    int mode;
+    FMODVoiceIn *fmd = (FMODVoiceIn *) hw;
+    struct audsettings obt_as = *as;
+
+    if (conf.broken_adc) {
+        return -1;
+    }
+
+    mode = aud_to_fmodfmt (as->fmt, as->nchannels == 2 ? 1 : 0);
+    fmd->fmod_sample = FSOUND_Sample_Alloc (
+        FSOUND_FREE,            /* index */
+        conf.nb_samples,        /* length */
+        mode,                   /* mode */
+        as->freq,               /* freq */
+        255,                    /* volume */
+        128,                    /* pan */
+        255                     /* priority */
+        );
+
+    if (!fmd->fmod_sample) {
+        fmod_logerr2 ("ADC", "Failed to allocate FMOD sample\n");
+        return -1;
+    }
+
+    /* FMOD always operates on little endian frames? */
+    obt_as.endianness = 0;
+    audio_pcm_init_info (&hw->info, &obt_as);
+    hw->samples = conf.nb_samples;
+    return 0;
+}
+
+static void fmod_fini_in (HWVoiceIn *hw)
+{
+    FMODVoiceIn *fmd = (FMODVoiceIn *) hw;
+
+    if (fmd->fmod_sample) {
+        FSOUND_Record_Stop ();
+        FSOUND_Sample_Free (fmd->fmod_sample);
+        fmd->fmod_sample = 0;
+    }
+}
+
+static int fmod_run_in (HWVoiceIn *hw)
+{
+    FMODVoiceIn *fmd = (FMODVoiceIn *) hw;
+    int hwshift = hw->info.shift;
+    int live, dead, new_pos, len;
+    unsigned int blen1 = 0, blen2 = 0;
+    unsigned int len1, len2;
+    unsigned int decr;
+    void *p1, *p2;
+
+    live = audio_pcm_hw_get_live_in (hw);
+    dead = hw->samples - live;
+    if (!dead) {
+        return 0;
+    }
+
+    new_pos = FSOUND_Record_GetPosition ();
+    if (new_pos < 0) {
+        fmod_logerr ("Could not get recording position\n");
+        return 0;
+    }
+
+    len = audio_ring_dist (new_pos,  hw->wpos, hw->samples);
+    if (!len) {
+        return 0;
+    }
+    len = audio_MIN (len, dead);
+
+    if (fmod_lock_sample (fmd->fmod_sample, &fmd->hw.info,
+                          hw->wpos, len,
+                          &p1, &p2,
+                          &blen1, &blen2)) {
+        return 0;
+    }
+
+    len1 = blen1 >> hwshift;
+    len2 = blen2 >> hwshift;
+    decr = len1 + len2;
+
+    if (p1 && blen1) {
+        hw->conv (hw->conv_buf + hw->wpos, p1, len1);
+    }
+    if (p2 && len2) {
+        hw->conv (hw->conv_buf, p2, len2);
+    }
+
+    fmod_unlock_sample (fmd->fmod_sample, p1, p2, blen1, blen2);
+    hw->wpos = (hw->wpos + decr) % hw->samples;
+    return decr;
+}
+
+static struct {
+    const char *name;
+    int type;
+} drvtab[] = {
+    { .name = "none",   .type = FSOUND_OUTPUT_NOSOUND },
+#ifdef _WIN32
+    { .name = "winmm",  .type = FSOUND_OUTPUT_WINMM   },
+    { .name = "dsound", .type = FSOUND_OUTPUT_DSOUND  },
+    { .name = "a3d",    .type = FSOUND_OUTPUT_A3D     },
+    { .name = "asio",   .type = FSOUND_OUTPUT_ASIO    },
+#endif
+#ifdef __linux__
+    { .name = "oss",    .type = FSOUND_OUTPUT_OSS     },
+    { .name = "alsa",   .type = FSOUND_OUTPUT_ALSA    },
+    { .name = "esd",    .type = FSOUND_OUTPUT_ESD     },
+#endif
+#ifdef __APPLE__
+    { .name = "mac",    .type = FSOUND_OUTPUT_MAC     },
+#endif
+#if 0
+    { .name = "xbox",   .type = FSOUND_OUTPUT_XBOX    },
+    { .name = "ps2",    .type = FSOUND_OUTPUT_PS2     },
+    { .name = "gcube",  .type = FSOUND_OUTPUT_GC      },
+#endif
+    { .name = "none-realtime", .type = FSOUND_OUTPUT_NOSOUND_NONREALTIME }
+};
+
+static void *fmod_audio_init (void)
+{
+    size_t i;
+    double ver;
+    int status;
+    int output_type = -1;
+    const char *drv = conf.drvname;
+
+    ver = FSOUND_GetVersion ();
+    if (ver < FMOD_VERSION) {
+        dolog ("Wrong FMOD version %f, need at least %f\n", ver, FMOD_VERSION);
+        return NULL;
+    }
+
+#ifdef __linux__
+    if (ver < 3.75) {
+        dolog ("FMOD before 3.75 has bug preventing ADC from working\n"
+               "ADC will be disabled.\n");
+        conf.broken_adc = 1;
+    }
+#endif
+
+    if (drv) {
+        int found = 0;
+        for (i = 0; i < ARRAY_SIZE (drvtab); i++) {
+            if (!strcmp (drv, drvtab[i].name)) {
+                output_type = drvtab[i].type;
+                found = 1;
+                break;
+            }
+        }
+        if (!found) {
+            dolog ("Unknown FMOD driver `%s'\n", drv);
+            dolog ("Valid drivers:\n");
+            for (i = 0; i < ARRAY_SIZE (drvtab); i++) {
+                dolog ("  %s\n", drvtab[i].name);
+            }
+        }
+    }
+
+    if (output_type != -1) {
+        status = FSOUND_SetOutput (output_type);
+        if (!status) {
+            fmod_logerr ("FSOUND_SetOutput(%d) failed\n", output_type);
+            return NULL;
+        }
+    }
+
+    if (conf.bufsize) {
+        status = FSOUND_SetBufferSize (conf.bufsize);
+        if (!status) {
+            fmod_logerr ("FSOUND_SetBufferSize (%d) failed\n", conf.bufsize);
+        }
+    }
+
+    status = FSOUND_Init (conf.freq, conf.nb_channels, 0);
+    if (!status) {
+        fmod_logerr ("FSOUND_Init failed\n");
+        return NULL;
+    }
+
+    return &conf;
+}
+
+static int fmod_read (SWVoiceIn *sw, void *buf, int size)
+{
+    return audio_pcm_sw_read (sw, buf, size);
+}
+
+static int fmod_ctl_in (HWVoiceIn *hw, int cmd, ...)
+{
+    int status;
+    FMODVoiceIn *fmd = (FMODVoiceIn *) hw;
+
+    switch (cmd) {
+    case VOICE_ENABLE:
+        status = FSOUND_Record_StartSample (fmd->fmod_sample, 1);
+        if (!status) {
+            fmod_logerr ("Failed to start recording\n");
+        }
+        break;
+
+    case VOICE_DISABLE:
+        status = FSOUND_Record_Stop ();
+        if (!status) {
+            fmod_logerr ("Failed to stop recording\n");
+        }
+        break;
+    }
+    return 0;
+}
+
+static void fmod_audio_fini (void *opaque)
+{
+    (void) opaque;
+    FSOUND_Close ();
+}
+
+static struct audio_option fmod_options[] = {
+    {
+        .name  = "DRV",
+        .tag   = AUD_OPT_STR,
+        .valp  = &conf.drvname,
+        .descr = "FMOD driver"
+    },
+    {
+        .name  = "FREQ",
+        .tag   = AUD_OPT_INT,
+        .valp  = &conf.freq,
+        .descr = "Default frequency"
+    },
+    {
+        .name  = "SAMPLES",
+        .tag   = AUD_OPT_INT,
+        .valp  = &conf.nb_samples,
+        .descr = "Buffer size in samples"
+    },
+    {
+        .name  = "CHANNELS",
+        .tag   = AUD_OPT_INT,
+        .valp  = &conf.nb_channels,
+        .descr = "Number of default channels (1 - mono, 2 - stereo)"
+    },
+    {
+        .name  = "BUFSIZE",
+        .tag   = AUD_OPT_INT,
+        .valp  = &conf.bufsize,
+        .descr = "(undocumented)"
+    },
+    { /* End of list */ }
+};
+
+static struct audio_pcm_ops fmod_pcm_ops = {
+    .init_out = fmod_init_out,
+    .fini_out = fmod_fini_out,
+    .run_out  = fmod_run_out,
+    .write    = fmod_write,
+    .ctl_out  = fmod_ctl_out,
+
+    .init_in  = fmod_init_in,
+    .fini_in  = fmod_fini_in,
+    .run_in   = fmod_run_in,
+    .read     = fmod_read,
+    .ctl_in   = fmod_ctl_in
+};
+
+struct audio_driver fmod_audio_driver = {
+    .name           = "fmod",
+    .descr          = "FMOD 3.xx http://www.fmod.org",
+    .options        = fmod_options,
+    .init           = fmod_audio_init,
+    .fini           = fmod_audio_fini,
+    .pcm_ops        = &fmod_pcm_ops,
+    .can_be_default = 1,
+    .max_voices_out = INT_MAX,
+    .max_voices_in  = INT_MAX,
+    .voice_size_out = sizeof (FMODVoiceOut),
+    .voice_size_in  = sizeof (FMODVoiceIn)
+};
--- a/audio/mixeng.c
+++ b/audio/mixeng.c
@@ -22,7 +22,6 @@
 * 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 "audio.h"

--- a/audio/noaudio.c
+++ b/audio/noaudio.c
@@ -21,7 +21,6 @@
 * 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 "audio.h"
 #include "qemu/timer.h"
@@ -64,7 +63,7 @@ static int no_write (SWVoiceOut *sw, void *buf, int len)
    return audio_pcm_sw_write (sw, buf, len);
 }

-static int no_init_out(HWVoiceOut *hw, struct audsettings *as, void *drv_opaque)
+static int no_init_out (HWVoiceOut *hw, struct audsettings *as)
 {
    audio_pcm_init_info (&hw->info, as);
    hw->samples = 1024;
@@ -83,7 +82,7 @@ static int no_ctl_out (HWVoiceOut *hw, int cmd, ...)
    return 0;
 }

-static int no_init_in(HWVoiceIn *hw, struct audsettings *as, void *drv_opaque)
+static int no_init_in (HWVoiceIn *hw, struct audsettings *as)
 {
    audio_pcm_init_info (&hw->info, as);
    hw->samples = 1024;
--- a/audio/ossaudio.c
+++ b/audio/ossaudio.c
@@ -21,15 +21,15 @@
 * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
 * THE SOFTWARE.
 */
-#include "qemu/osdep.h"
+#include <stdlib.h>
 #include <sys/mman.h>
+#include <sys/types.h>
 #include <sys/ioctl.h>
 #include <sys/soundcard.h>
 #include "qemu-common.h"
 #include "qemu/main-loop.h"
 #include "qemu/host-utils.h"
 #include "audio.h"
-#include "trace.h"

 #define AUDIO_CAP "oss"
 #include "audio_int.h"
@@ -38,16 +38,6 @@
 #define USE_DSP_POLICY
 #endif

-typedef struct OSSConf {
-    int try_mmap;
-    int nfrags;
-    int fragsize;
-    const char *devpath_out;
-    const char *devpath_in;
-    int exclusive;
-    int policy;
-} OSSConf;
-
 typedef struct OSSVoiceOut {
    HWVoiceOut hw;
    void *pcm_buf;
@@ -57,7 +47,6 @@ typedef struct OSSVoiceOut {
    int fragsize;
    int mmapped;
    int pending;
-    OSSConf *conf;
 } OSSVoiceOut;

 typedef struct OSSVoiceIn {
@@ -66,9 +55,28 @@ typedef struct OSSVoiceIn {
    int fd;
    int nfrags;
    int fragsize;
-    OSSConf *conf;
 } OSSVoiceIn;

+static struct {
+    int try_mmap;
+    int nfrags;
+    int fragsize;
+    const char *devpath_out;
+    const char *devpath_in;
+    int debug;
+    int exclusive;
+    int policy;
+} conf = {
+    .try_mmap = 0,
+    .nfrags = 4,
+    .fragsize = 4096,
+    .devpath_out = "/dev/dsp",
+    .devpath_in = "/dev/dsp",
+    .debug = 0,
+    .exclusive = 0,
+    .policy = 5
+};
+
 struct oss_params {
    int freq;
    audfmt_e fmt;
@@ -130,18 +138,18 @@ static void oss_helper_poll_in (void *opaque)
    audio_run ("oss_poll_in");
 }

-static void oss_poll_out (HWVoiceOut *hw)
+static int oss_poll_out (HWVoiceOut *hw)
 {
    OSSVoiceOut *oss = (OSSVoiceOut *) hw;

-    qemu_set_fd_handler (oss->fd, NULL, oss_helper_poll_out, NULL);
+    return qemu_set_fd_handler (oss->fd, NULL, oss_helper_poll_out, NULL);
 }

-static void oss_poll_in (HWVoiceIn *hw)
+static int oss_poll_in (HWVoiceIn *hw)
 {
    OSSVoiceIn *oss = (OSSVoiceIn *) hw;

-    qemu_set_fd_handler (oss->fd, oss_helper_poll_in, NULL, NULL);
+    return qemu_set_fd_handler (oss->fd, oss_helper_poll_in, NULL, NULL);
 }

 static int oss_write (SWVoiceOut *sw, void *buf, int len)
@@ -264,18 +272,18 @@ static int oss_get_version (int fd, int *version, const char *typ)
 #endif

 static int oss_open (int in, struct oss_params *req,
-                     struct oss_params *obt, int *pfd, OSSConf* conf)
+                     struct oss_params *obt, int *pfd)
 {
    int fd;
-    int oflags = conf->exclusive ? O_EXCL : 0;
+    int oflags = conf.exclusive ? O_EXCL : 0;
    audio_buf_info abinfo;
    int fmt, freq, nchannels;
    int setfragment = 1;
-    const char *dspname = in ? conf->devpath_in : conf->devpath_out;
+    const char *dspname = in ? conf.devpath_in : conf.devpath_out;
    const char *typ = in ? "ADC" : "DAC";

    /* Kludge needed to have working mmap on Linux */
-    oflags |= conf->try_mmap ? O_RDWR : (in ? O_RDONLY : O_WRONLY);
+    oflags |= conf.try_mmap ? O_RDWR : (in ? O_RDONLY : O_WRONLY);

    fd = open (dspname, oflags | O_NONBLOCK);
    if (-1 == fd) {
@@ -309,18 +317,20 @@ static int oss_open (int in, struct oss_params *req,
    }

 #ifdef USE_DSP_POLICY
-    if (conf->policy >= 0) {
+    if (conf.policy >= 0) {
        int version;

        if (!oss_get_version (fd, &version, typ)) {
-            trace_oss_version(version);
+            if (conf.debug) {
+                dolog ("OSS version = %#x\n", version);
+            }

            if (version >= 0x040000) {
-                int policy = conf->policy;
+                int policy = conf.policy;
                if (ioctl (fd, SNDCTL_DSP_POLICY, &policy)) {
                    oss_logerr2 (errno, typ,
                                 "Failed to set timing policy to %d\n",
-                                 conf->policy);
+                                 conf.policy);
                    goto err;
                }
                setfragment = 0;
@@ -448,12 +458,19 @@ static int oss_run_out (HWVoiceOut *hw, int live)
        }

        if (abinfo.bytes > bufsize) {
-            trace_oss_invalid_available_size(abinfo.bytes, bufsize);
+            if (conf.debug) {
+                dolog ("warning: Invalid available size, size=%d bufsize=%d\n"
+                       "please report your OS/audio hw to av1474@comtv.ru\n",
+                       abinfo.bytes, bufsize);
+            }
            abinfo.bytes = bufsize;
        }

        if (abinfo.bytes < 0) {
-            trace_oss_invalid_available_size(abinfo.bytes, bufsize);
+            if (conf.debug) {
+                dolog ("warning: Invalid available size, size=%d bufsize=%d\n",
+                       abinfo.bytes, bufsize);
+            }
            return 0;
        }

@@ -493,8 +510,7 @@ static void oss_fini_out (HWVoiceOut *hw)
    }
 }

-static int oss_init_out(HWVoiceOut *hw, struct audsettings *as,
-                        void *drv_opaque)
+static int oss_init_out (HWVoiceOut *hw, struct audsettings *as)
 {
    OSSVoiceOut *oss = (OSSVoiceOut *) hw;
    struct oss_params req, obt;
@@ -503,17 +519,16 @@ static int oss_init_out(HWVoiceOut *hw, struct audsettings *as,
    int fd;
    audfmt_e effective_fmt;
    struct audsettings obt_as;
-    OSSConf *conf = drv_opaque;

    oss->fd = -1;

    req.fmt = aud_to_ossfmt (as->fmt, as->endianness);
    req.freq = as->freq;
    req.nchannels = as->nchannels;
-    req.fragsize = conf->fragsize;
-    req.nfrags = conf->nfrags;
+    req.fragsize = conf.fragsize;
+    req.nfrags = conf.nfrags;

-    if (oss_open (0, &req, &obt, &fd, conf)) {
+    if (oss_open (0, &req, &obt, &fd)) {
        return -1;
    }

@@ -540,7 +555,7 @@ static int oss_init_out(HWVoiceOut *hw, struct audsettings *as,
    hw->samples = (obt.nfrags * obt.fragsize) >> hw->info.shift;

    oss->mmapped = 0;
-    if (conf->try_mmap) {
+    if (conf.try_mmap) {
        oss->pcm_buf = mmap (
            NULL,
            hw->samples << hw->info.shift,
@@ -600,7 +615,6 @@ static int oss_init_out(HWVoiceOut *hw, struct audsettings *as,
    }

    oss->fd = fd;
-    oss->conf = conf;
    return 0;
 }

@@ -620,8 +634,7 @@ static int oss_ctl_out (HWVoiceOut *hw, int cmd, ...)
            va_end (ap);

            ldebug ("enabling voice\n");
-            if (poll_mode) {
-                oss_poll_out (hw);
+            if (poll_mode && oss_poll_out (hw)) {
                poll_mode = 0;
            }
            hw->poll_mode = poll_mode;
@@ -663,7 +676,7 @@ static int oss_ctl_out (HWVoiceOut *hw, int cmd, ...)
    return 0;
 }

-static int oss_init_in(HWVoiceIn *hw, struct audsettings *as, void *drv_opaque)
+static int oss_init_in (HWVoiceIn *hw, struct audsettings *as)
 {
    OSSVoiceIn *oss = (OSSVoiceIn *) hw;
    struct oss_params req, obt;
@@ -672,16 +685,15 @@ static int oss_init_in(HWVoiceIn *hw, struct audsettings *as, void *drv_opaque)
    int fd;
    audfmt_e effective_fmt;
    struct audsettings obt_as;
-    OSSConf *conf = drv_opaque;

    oss->fd = -1;

    req.fmt = aud_to_ossfmt (as->fmt, as->endianness);
    req.freq = as->freq;
    req.nchannels = as->nchannels;
-    req.fragsize = conf->fragsize;
-    req.nfrags = conf->nfrags;
-    if (oss_open (1, &req, &obt, &fd, conf)) {
+    req.fragsize = conf.fragsize;
+    req.nfrags = conf.nfrags;
+    if (oss_open (1, &req, &obt, &fd)) {
        return -1;
    }

@@ -715,7 +727,6 @@ static int oss_init_in(HWVoiceIn *hw, struct audsettings *as, void *drv_opaque)
    }

    oss->fd = fd;
-    oss->conf = conf;
    return 0;
 }

@@ -817,8 +828,7 @@ static int oss_ctl_in (HWVoiceIn *hw, int cmd, ...)
            poll_mode = va_arg (ap, int);
            va_end (ap);

-            if (poll_mode) {
-                oss_poll_in (hw);
+            if (poll_mode && oss_poll_in (hw)) {
                poll_mode = 0;
            }
            hw->poll_mode = poll_mode;
@@ -835,79 +845,71 @@ static int oss_ctl_in (HWVoiceIn *hw, int cmd, ...)
    return 0;
 }

-static OSSConf glob_conf = {
-    .try_mmap = 0,
-    .nfrags = 4,
-    .fragsize = 4096,
-    .devpath_out = "/dev/dsp",
-    .devpath_in = "/dev/dsp",
-    .exclusive = 0,
-    .policy = 5
-};
-
 static void *oss_audio_init (void)
 {
-    OSSConf *conf = g_malloc(sizeof(OSSConf));
-    *conf = glob_conf;
-
-    if (access(conf->devpath_in, R_OK | W_OK) < 0 ||
-        access(conf->devpath_out, R_OK | W_OK) < 0) {
-        g_free(conf);
+    if (access(conf.devpath_in, R_OK | W_OK) < 0 ||
+        access(conf.devpath_out, R_OK | W_OK) < 0) {
        return NULL;
    }
-    return conf;
+    return &conf;
 }

 static void oss_audio_fini (void *opaque)
 {
-    g_free(opaque);
+    (void) opaque;
 }

 static struct audio_option oss_options[] = {
    {
        .name  = "FRAGSIZE",
        .tag   = AUD_OPT_INT,
-        .valp  = &glob_conf.fragsize,
+        .valp  = &conf.fragsize,
        .descr = "Fragment size in bytes"
    },
    {
        .name  = "NFRAGS",
        .tag   = AUD_OPT_INT,
-        .valp  = &glob_conf.nfrags,
+        .valp  = &conf.nfrags,
        .descr = "Number of fragments"
    },
    {
        .name  = "MMAP",
        .tag   = AUD_OPT_BOOL,
-        .valp  = &glob_conf.try_mmap,
+        .valp  = &conf.try_mmap,
        .descr = "Try using memory mapped access"
    },
    {
        .name  = "DAC_DEV",
        .tag   = AUD_OPT_STR,
-        .valp  = &glob_conf.devpath_out,
+        .valp  = &conf.devpath_out,
        .descr = "Path to DAC device"
    },
    {
        .name  = "ADC_DEV",
        .tag   = AUD_OPT_STR,
-        .valp  = &glob_conf.devpath_in,
+        .valp  = &conf.devpath_in,
        .descr = "Path to ADC device"
    },
    {
        .name  = "EXCLUSIVE",
        .tag   = AUD_OPT_BOOL,
-        .valp  = &glob_conf.exclusive,
+        .valp  = &conf.exclusive,
        .descr = "Open device in exclusive mode (vmix wont work)"
    },
 #ifdef USE_DSP_POLICY
    {
        .name  = "POLICY",
        .tag   = AUD_OPT_INT,
-        .valp  = &glob_conf.policy,
+        .valp  = &conf.policy,
        .descr = "Set the timing policy of the device, -1 to use fragment mode",
    },
 #endif
+    {
+        .name  = "DEBUG",
+        .tag   = AUD_OPT_BOOL,
+        .valp  = &conf.debug,
+        .descr = "Turn on some debugging messages"
+    },
    { /* End of list */ }
 };

--- a/audio/paaudio.c
+++ b/audio/paaudio.c
@@ -1,5 +1,4 @@
 /* public domain */
-#include "qemu/osdep.h"
 #include "qemu-common.h"
 #include "audio.h"

@@ -9,19 +8,6 @@
 #include "audio_int.h"
 #include "audio_pt_int.h"

-typedef struct {
-    int samples;
-    char *server;
-    char *sink;
-    char *source;
-} PAConf;
-
-typedef struct {
-    PAConf conf;
-    pa_threaded_mainloop *mainloop;
-    pa_context *context;
-} paaudio;
-
 typedef struct {
    HWVoiceOut hw;
    int done;
@@ -31,7 +17,6 @@ typedef struct {
    pa_stream *stream;
    void *pcm_buf;
    struct audio_pt pt;
-    paaudio *g;
 } PAVoiceOut;

 typedef struct {
@@ -45,10 +30,20 @@ typedef struct {
    struct audio_pt pt;
    const void *read_data;
    size_t read_index, read_length;
-    paaudio *g;
 } PAVoiceIn;

-static void qpa_audio_fini(void *opaque);
+typedef struct {
+    int samples;
+    char *server;
+    char *sink;
+    char *source;
+    pa_threaded_mainloop *mainloop;
+    pa_context *context;
+} paaudio;
+
+static paaudio glob_paaudio = {
+    .samples = 4096,
+};

 static void GCC_FMT_ATTR (2, 3) qpa_logerr (int err, const char *fmt, ...)
 {
@@ -111,7 +106,7 @@ static inline int PA_STREAM_IS_GOOD(pa_stream_state_t x)

 static int qpa_simple_read (PAVoiceIn *p, void *data, size_t length, int *rerror)
 {
-    paaudio *g = p->g;
+    paaudio *g = &glob_paaudio;

    pa_threaded_mainloop_lock (g->mainloop);

@@ -165,7 +160,7 @@ unlock_and_fail:

 static int qpa_simple_write (PAVoiceOut *p, const void *data, size_t length, int *rerror)
 {
-    paaudio *g = p->g;
+    paaudio *g = &glob_paaudio;

    pa_threaded_mainloop_lock (g->mainloop);

@@ -227,7 +222,7 @@ static void *qpa_thread_out (void *arg)
            }
        }

-        decr = to_mix = audio_MIN (pa->live, pa->g->conf.samples >> 2);
+        decr = to_mix = audio_MIN (pa->live, glob_paaudio.samples >> 2);
        rpos = pa->rpos;

        if (audio_pt_unlock (&pa->pt, AUDIO_FUNC)) {
@@ -319,7 +314,7 @@ static void *qpa_thread_in (void *arg)
            }
        }

-        incr = to_grab = audio_MIN (pa->dead, pa->g->conf.samples >> 2);
+        incr = to_grab = audio_MIN (pa->dead, glob_paaudio.samples >> 2);
        wpos = pa->wpos;

        if (audio_pt_unlock (&pa->pt, AUDIO_FUNC)) {
@@ -435,7 +430,7 @@ static audfmt_e pa_to_audfmt (pa_sample_format_t fmt, int *endianness)

 static void context_state_cb (pa_context *c, void *userdata)
 {
-    paaudio *g = userdata;
+    paaudio *g = &glob_paaudio;

    switch (pa_context_get_state(c)) {
    case PA_CONTEXT_READY:
@@ -454,7 +449,7 @@ static void context_state_cb (pa_context *c, void *userdata)

 static void stream_state_cb (pa_stream *s, void * userdata)
 {
-    paaudio *g = userdata;
+    paaudio *g = &glob_paaudio;

    switch (pa_stream_get_state (s)) {

@@ -472,21 +467,23 @@ static void stream_state_cb (pa_stream *s, void * userdata)

 static void stream_request_cb (pa_stream *s, size_t length, void *userdata)
 {
-    paaudio *g = userdata;
+    paaudio *g = &glob_paaudio;

    pa_threaded_mainloop_signal (g->mainloop, 0);
 }

 static pa_stream *qpa_simple_new (
-        paaudio *g,
+        const char *server,
        const char *name,
        pa_stream_direction_t dir,
        const char *dev,
+        const char *stream_name,
        const pa_sample_spec *ss,
        const pa_channel_map *map,
        const pa_buffer_attr *attr,
        int *rerror)
 {
+    paaudio *g = &glob_paaudio;
    int r;
    pa_stream *stream;

@@ -537,15 +534,13 @@ fail:
    return NULL;
 }

-static int qpa_init_out(HWVoiceOut *hw, struct audsettings *as,
-                        void *drv_opaque)
+static int qpa_init_out (HWVoiceOut *hw, struct audsettings *as)
 {
    int error;
-    pa_sample_spec ss;
-    pa_buffer_attr ba;
+    static pa_sample_spec ss;
+    static pa_buffer_attr ba;
    struct audsettings obt_as = *as;
    PAVoiceOut *pa = (PAVoiceOut *) hw;
-    paaudio *g = pa->g = drv_opaque;

    ss.format = audfmt_to_pa (as->fmt, as->endianness);
    ss.channels = as->nchannels;
@@ -563,10 +558,11 @@ static int qpa_init_out(HWVoiceOut *hw, struct audsettings *as,
    obt_as.fmt = pa_to_audfmt (ss.format, &obt_as.endianness);

    pa->stream = qpa_simple_new (
-        g,
+        glob_paaudio.server,
        "qemu",
        PA_STREAM_PLAYBACK,
-        g->conf.sink,
+        glob_paaudio.sink,
+        "pcm.playback",
        &ss,
        NULL,                   /* channel map */
        &ba,                    /* buffering attributes */
@@ -578,7 +574,7 @@ static int qpa_init_out(HWVoiceOut *hw, struct audsettings *as,
    }

    audio_pcm_init_info (&hw->info, &obt_as);
-    hw->samples = g->conf.samples;
+    hw->samples = glob_paaudio.samples;
    pa->pcm_buf = audio_calloc (AUDIO_FUNC, hw->samples, 1 << hw->info.shift);
    pa->rpos = hw->rpos;
    if (!pa->pcm_buf) {
@@ -605,13 +601,12 @@ static int qpa_init_out(HWVoiceOut *hw, struct audsettings *as,
    return -1;
 }

-static int qpa_init_in(HWVoiceIn *hw, struct audsettings *as, void *drv_opaque)
+static int qpa_init_in (HWVoiceIn *hw, struct audsettings *as)
 {
    int error;
-    pa_sample_spec ss;
+    static pa_sample_spec ss;
    struct audsettings obt_as = *as;
    PAVoiceIn *pa = (PAVoiceIn *) hw;
-    paaudio *g = pa->g = drv_opaque;

    ss.format = audfmt_to_pa (as->fmt, as->endianness);
    ss.channels = as->nchannels;
@@ -620,10 +615,11 @@ static int qpa_init_in(HWVoiceIn *hw, struct audsettings *as, void *drv_opaque)
    obt_as.fmt = pa_to_audfmt (ss.format, &obt_as.endianness);

    pa->stream = qpa_simple_new (
-        g,
+        glob_paaudio.server,
        "qemu",
        PA_STREAM_RECORD,
-        g->conf.source,
+        glob_paaudio.source,
+        "pcm.capture",
        &ss,
        NULL,                   /* channel map */
        NULL,                   /* buffering attributes */
@@ -635,7 +631,7 @@ static int qpa_init_in(HWVoiceIn *hw, struct audsettings *as, void *drv_opaque)
    }

    audio_pcm_init_info (&hw->info, &obt_as);
-    hw->samples = g->conf.samples;
+    hw->samples = glob_paaudio.samples;
    pa->pcm_buf = audio_calloc (AUDIO_FUNC, hw->samples, 1 << hw->info.shift);
    pa->wpos = hw->wpos;
    if (!pa->pcm_buf) {
@@ -707,7 +703,7 @@ static int qpa_ctl_out (HWVoiceOut *hw, int cmd, ...)
    PAVoiceOut *pa = (PAVoiceOut *) hw;
    pa_operation *op;
    pa_cvolume v;
-    paaudio *g = pa->g;
+    paaudio *g = &glob_paaudio;

 #ifdef PA_CHECK_VERSION    /* macro is present in 0.9.16+ */
    pa_cvolume_init (&v);  /* function is present in 0.9.13+ */
@@ -759,7 +755,7 @@ static int qpa_ctl_in (HWVoiceIn *hw, int cmd, ...)
    PAVoiceIn *pa = (PAVoiceIn *) hw;
    pa_operation *op;
    pa_cvolume v;
-    paaudio *g = pa->g;
+    paaudio *g = &glob_paaudio;

 #ifdef PA_CHECK_VERSION
    pa_cvolume_init (&v);
@@ -809,31 +805,23 @@ static int qpa_ctl_in (HWVoiceIn *hw, int cmd, ...)
 }

 /* common */
-static PAConf glob_conf = {
-    .samples = 4096,
-};
-
 static void *qpa_audio_init (void)
 {
-    paaudio *g = g_malloc(sizeof(paaudio));
-    g->conf = glob_conf;
-    g->mainloop = NULL;
-    g->context = NULL;
+    paaudio *g = &glob_paaudio;

    g->mainloop = pa_threaded_mainloop_new ();
    if (!g->mainloop) {
        goto fail;
    }

-    g->context = pa_context_new (pa_threaded_mainloop_get_api (g->mainloop),
-                                 g->conf.server);
+    g->context = pa_context_new (pa_threaded_mainloop_get_api (g->mainloop), glob_paaudio.server);
    if (!g->context) {
        goto fail;
    }

    pa_context_set_state_callback (g->context, context_state_cb, g);

-    if (pa_context_connect (g->context, g->conf.server, 0, NULL) < 0) {
+    if (pa_context_connect (g->context, glob_paaudio.server, 0, NULL) < 0) {
        qpa_logerr (pa_context_errno (g->context),
                    "pa_context_connect() failed\n");
        goto fail;
@@ -866,13 +854,12 @@ static void *qpa_audio_init (void)

    pa_threaded_mainloop_unlock (g->mainloop);

-    return g;
+    return &glob_paaudio;

 unlock_and_fail:
    pa_threaded_mainloop_unlock (g->mainloop);
 fail:
    AUD_log (AUDIO_CAP, "Failed to initialize PA context");
-    qpa_audio_fini(g);
    return NULL;
 }

@@ -887,38 +874,39 @@ static void qpa_audio_fini (void *opaque)
    if (g->context) {
        pa_context_disconnect (g->context);
        pa_context_unref (g->context);
+        g->context = NULL;
    }

    if (g->mainloop) {
        pa_threaded_mainloop_free (g->mainloop);
    }

-    g_free(g);
+    g->mainloop = NULL;
 }

 struct audio_option qpa_options[] = {
    {
        .name  = "SAMPLES",
        .tag   = AUD_OPT_INT,
-        .valp  = &glob_conf.samples,
+        .valp  = &glob_paaudio.samples,
        .descr = "buffer size in samples"
    },
    {
        .name  = "SERVER",
        .tag   = AUD_OPT_STR,
-        .valp  = &glob_conf.server,
+        .valp  = &glob_paaudio.server,
        .descr = "server address"
    },
    {
        .name  = "SINK",
        .tag   = AUD_OPT_STR,
-        .valp  = &glob_conf.sink,
+        .valp  = &glob_paaudio.sink,
        .descr = "sink device name"
    },
    {
        .name  = "SOURCE",
        .tag   = AUD_OPT_STR,
-        .valp  = &glob_conf.source,
+        .valp  = &glob_paaudio.source,
        .descr = "source device name"
    },
    { /* End of list */ }
--- a/audio/sdlaudio.c
+++ b/audio/sdlaudio.c
@@ -21,7 +21,6 @@
 * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
 * THE SOFTWARE.
 */
-#include "qemu/osdep.h"
 #include <SDL.h>
 #include <SDL_thread.h>
 #include "qemu-common.h"
@@ -56,7 +55,6 @@ static struct SDLAudioState {
    SDL_mutex *mutex;
    SDL_sem *sem;
    int initialized;
-    bool driver_created;
 } glob_sdl;
 typedef struct SDLAudioState SDLAudioState;

@@ -334,8 +332,7 @@ static void sdl_fini_out (HWVoiceOut *hw)
    sdl_close (&glob_sdl);
 }

-static int sdl_init_out(HWVoiceOut *hw, struct audsettings *as,
-                        void *drv_opaque)
+static int sdl_init_out (HWVoiceOut *hw, struct audsettings *as)
 {
    SDLVoiceOut *sdl = (SDLVoiceOut *) hw;
    SDLAudioState *s = &glob_sdl;
@@ -395,10 +392,6 @@ static int sdl_ctl_out (HWVoiceOut *hw, int cmd, ...)
 static void *sdl_audio_init (void)
 {
    SDLAudioState *s = &glob_sdl;
-    if (s->driver_created) {
-        sdl_logerr("Can't create multiple sdl backends\n");
-        return NULL;
-    }

    if (SDL_InitSubSystem (SDL_INIT_AUDIO)) {
        sdl_logerr ("SDL failed to initialize audio subsystem\n");
@@ -420,7 +413,6 @@ static void *sdl_audio_init (void)
        return NULL;
    }

-    s->driver_created = true;
    return s;
 }

@@ -431,7 +423,6 @@ static void sdl_audio_fini (void *opaque)
    SDL_DestroySemaphore (s->sem);
    SDL_DestroyMutex (s->mutex);
    SDL_QuitSubSystem (SDL_INIT_AUDIO);
-    s->driver_created = false;
 }

 static struct audio_option sdl_options[] = {
--- a/audio/spiceaudio.c
+++ b/audio/spiceaudio.c
@@ -17,9 +17,7 @@
 * along with this program; if not, see <http://www.gnu.org/licenses/>.
 */

-#include "qemu/osdep.h"
 #include "hw/hw.h"
-#include "qemu/error-report.h"
 #include "qemu/timer.h"
 #include "ui/qemu-spice.h"

@@ -117,8 +115,7 @@ static int rate_get_samples (struct audio_pcm_info *info, SpiceRateCtl *rate)

 /* playback */

-static int line_out_init(HWVoiceOut *hw, struct audsettings *as,
-                         void *drv_opaque)
+static int line_out_init (HWVoiceOut *hw, struct audsettings *as)
 {
    SpiceVoiceOut *out = container_of (hw, SpiceVoiceOut, hw);
    struct audsettings settings;
@@ -246,7 +243,7 @@ static int line_out_ctl (HWVoiceOut *hw, int cmd, ...)

 /* record */

-static int line_in_init(HWVoiceIn *hw, struct audsettings *as, void *drv_opaque)
+static int line_in_init (HWVoiceIn *hw, struct audsettings *as)
 {
    SpiceVoiceIn *in = container_of (hw, SpiceVoiceIn, hw);
    struct audsettings settings;
--- a/audio/wavaudio.c
+++ b/audio/wavaudio.c
@@ -21,7 +21,6 @@
 * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
 * THE SOFTWARE.
 */
-#include "qemu/osdep.h"
 #include "hw/hw.h"
 #include "qemu/timer.h"
 #include "audio.h"
@@ -37,10 +36,15 @@ typedef struct WAVVoiceOut {
    int total_samples;
 } WAVVoiceOut;

-typedef struct {
+static struct {
    struct audsettings settings;
    const char *wav_path;
-} WAVConf;
+} conf = {
+    .settings.freq      = 44100,
+    .settings.nchannels = 2,
+    .settings.fmt       = AUD_FMT_S16,
+    .wav_path           = "qemu.wav"
+};

 static int wav_run_out (HWVoiceOut *hw, int live)
 {
@@ -101,8 +105,7 @@ static void le_store (uint8_t *buf, uint32_t val, int len)
    }
 }

-static int wav_init_out(HWVoiceOut *hw, struct audsettings *as,
-                        void *drv_opaque)
+static int wav_init_out (HWVoiceOut *hw, struct audsettings *as)
 {
    WAVVoiceOut *wav = (WAVVoiceOut *) hw;
    int bits16 = 0, stereo = 0;
@@ -112,8 +115,9 @@ static int wav_init_out(HWVoiceOut *hw, struct audsettings *as,
        0x02, 0x00, 0x44, 0xac, 0x00, 0x00, 0x10, 0xb1, 0x02, 0x00, 0x04,
        0x00, 0x10, 0x00, 0x64, 0x61, 0x74, 0x61, 0x00, 0x00, 0x00, 0x00
    };
-    WAVConf *conf = drv_opaque;
-    struct audsettings wav_as = conf->settings;
+    struct audsettings wav_as = conf.settings;
+
+    (void) as;

    stereo = wav_as.nchannels == 2;
    switch (wav_as.fmt) {
@@ -151,10 +155,10 @@ static int wav_init_out(HWVoiceOut *hw, struct audsettings *as,
    le_store (hdr + 28, hw->info.freq << (bits16 + stereo), 4);
    le_store (hdr + 32, 1 << (bits16 + stereo), 2);

-    wav->f = fopen (conf->wav_path, "wb");
+    wav->f = fopen (conf.wav_path, "wb");
    if (!wav->f) {
        dolog ("Failed to open wave file `%s'\nReason: %s\n",
-               conf->wav_path, strerror (errno));
+               conf.wav_path, strerror (errno));
        g_free (wav->pcm_buf);
        wav->pcm_buf = NULL;
        return -1;
@@ -222,49 +226,40 @@ static int wav_ctl_out (HWVoiceOut *hw, int cmd, ...)
    return 0;
 }

-static WAVConf glob_conf = {
-    .settings.freq      = 44100,
-    .settings.nchannels = 2,
-    .settings.fmt       = AUD_FMT_S16,
-    .wav_path           = "qemu.wav"
-};
-
 static void *wav_audio_init (void)
 {
-    WAVConf *conf = g_malloc(sizeof(WAVConf));
-    *conf = glob_conf;
-    return conf;
+    return &conf;
 }

 static void wav_audio_fini (void *opaque)
 {
+    (void) opaque;
    ldebug ("wav_fini");
-    g_free(opaque);
 }

 static struct audio_option wav_options[] = {
    {
        .name  = "FREQUENCY",
        .tag   = AUD_OPT_INT,
-        .valp  = &glob_conf.settings.freq,
+        .valp  = &conf.settings.freq,
        .descr = "Frequency"
    },
    {
        .name  = "FORMAT",
        .tag   = AUD_OPT_FMT,
-        .valp  = &glob_conf.settings.fmt,
+        .valp  = &conf.settings.fmt,
        .descr = "Format"
    },
    {
        .name  = "DAC_FIXED_CHANNELS",
        .tag   = AUD_OPT_INT,
-        .valp  = &glob_conf.settings.nchannels,
+        .valp  = &conf.settings.nchannels,
        .descr = "Number of channels (1 - mono, 2 - stereo)"
    },
    {
        .name  = "PATH",
        .tag   = AUD_OPT_STR,
-        .valp  = &glob_conf.wav_path,
+        .valp  = &conf.wav_path,
        .descr = "Path to wave file"
    },
    { /* End of list */ }
--- a/audio/wavcapture.c
+++ b/audio/wavcapture.c
@@ -1,7 +1,5 @@
-#include "qemu/osdep.h"
 #include "hw/hw.h"
 #include "monitor/monitor.h"
-#include "qemu/error-report.h"
 #include "audio.h"

 typedef struct {
--- a/audio/winwaveaudio.c
+++ b/audio/winwaveaudio.c
@@ -0,0 +1,717 @@
+/* public domain */
+
+#include "qemu-common.h"
+#include "sysemu/sysemu.h"
+#include "audio.h"
+
+#define AUDIO_CAP "winwave"
+#include "audio_int.h"
+
+#include <windows.h>
+#include <mmsystem.h>
+
+#include "audio_win_int.h"
+
+static struct {
+    int dac_headers;
+    int dac_samples;
+    int adc_headers;
+    int adc_samples;
+} conf = {
+    .dac_headers = 4,
+    .dac_samples = 1024,
+    .adc_headers = 4,
+    .adc_samples = 1024
+};
+
+typedef struct {
+    HWVoiceOut hw;
+    HWAVEOUT hwo;
+    WAVEHDR *hdrs;
+    HANDLE event;
+    void *pcm_buf;
+    int avail;
+    int pending;
+    int curhdr;
+    int paused;
+    CRITICAL_SECTION crit_sect;
+} WaveVoiceOut;
+
+typedef struct {
+    HWVoiceIn hw;
+    HWAVEIN hwi;
+    WAVEHDR *hdrs;
+    HANDLE event;
+    void *pcm_buf;
+    int curhdr;
+    int paused;
+    int rpos;
+    int avail;
+    CRITICAL_SECTION crit_sect;
+} WaveVoiceIn;
+
+static void winwave_log_mmresult (MMRESULT mr)
+{
+    const char *str = "BUG";
+
+    switch (mr) {
+    case MMSYSERR_NOERROR:
+        str = "Success";
+        break;
+
+    case MMSYSERR_INVALHANDLE:
+        str = "Specified device handle is invalid";
+        break;
+
+    case MMSYSERR_BADDEVICEID:
+        str = "Specified device id is out of range";
+        break;
+
+    case MMSYSERR_NODRIVER:
+        str = "No device driver is present";
+        break;
+
+    case MMSYSERR_NOMEM:
+        str = "Unable to allocate or lock memory";
+        break;
+
+    case WAVERR_SYNC:
+        str = "Device is synchronous but waveOutOpen was called "
+            "without using the WINWAVE_ALLOWSYNC flag";
+        break;
+
+    case WAVERR_UNPREPARED:
+        str = "The data block pointed to by the pwh parameter "
+            "hasn't been prepared";
+        break;
+
+    case WAVERR_STILLPLAYING:
+        str = "There are still buffers in the queue";
+        break;
+
+    default:
+        dolog ("Reason: Unknown (MMRESULT %#x)\n", mr);
+        return;
+    }
+
+    dolog ("Reason: %s\n", str);
+}
+
+static void GCC_FMT_ATTR (2, 3) winwave_logerr (
+    MMRESULT mr,
+    const char *fmt,
+    ...
+    )
+{
+    va_list ap;
+
+    va_start (ap, fmt);
+    AUD_vlog (AUDIO_CAP, fmt, ap);
+    va_end (ap);
+
+    AUD_log (NULL, " failed\n");
+    winwave_log_mmresult (mr);
+}
+
+static void winwave_anal_close_out (WaveVoiceOut *wave)
+{
+    MMRESULT mr;
+
+    mr = waveOutClose (wave->hwo);
+    if (mr != MMSYSERR_NOERROR) {
+        winwave_logerr (mr, "waveOutClose");
+    }
+    wave->hwo = NULL;
+}
+
+static void CALLBACK winwave_callback_out (
+    HWAVEOUT hwo,
+    UINT msg,
+    DWORD_PTR dwInstance,
+    DWORD_PTR dwParam1,
+    DWORD_PTR dwParam2
+    )
+{
+    WaveVoiceOut *wave = (WaveVoiceOut *) dwInstance;
+
+    switch (msg) {
+    case WOM_DONE:
+        {
+            WAVEHDR *h = (WAVEHDR *) dwParam1;
+            if (!h->dwUser) {
+                h->dwUser = 1;
+                EnterCriticalSection (&wave->crit_sect);
+                {
+                    wave->avail += conf.dac_samples;
+                }
+                LeaveCriticalSection (&wave->crit_sect);
+                if (wave->hw.poll_mode) {
+                    if (!SetEvent (wave->event)) {
+                        dolog ("DAC SetEvent failed %lx\n", GetLastError ());
+                    }
+                }
+            }
+        }
+        break;
+
+    case WOM_CLOSE:
+    case WOM_OPEN:
+        break;
+
+    default:
+        dolog ("unknown wave out callback msg %x\n", msg);
+    }
+}
+
+static int winwave_init_out (HWVoiceOut *hw, struct audsettings *as)
+{
+    int i;
+    int err;
+    MMRESULT mr;
+    WAVEFORMATEX wfx;
+    WaveVoiceOut *wave;
+
+    wave = (WaveVoiceOut *) hw;
+
+    InitializeCriticalSection (&wave->crit_sect);
+
+    err = waveformat_from_audio_settings (&wfx, as);
+    if (err) {
+        goto err0;
+    }
+
+    mr = waveOutOpen (&wave->hwo, WAVE_MAPPER, &wfx,
+                      (DWORD_PTR) winwave_callback_out,
+                      (DWORD_PTR) wave, CALLBACK_FUNCTION);
+    if (mr != MMSYSERR_NOERROR) {
+        winwave_logerr (mr, "waveOutOpen");
+        goto err1;
+    }
+
+    wave->hdrs = audio_calloc (AUDIO_FUNC, conf.dac_headers,
+                               sizeof (*wave->hdrs));
+    if (!wave->hdrs) {
+        goto err2;
+    }
+
+    audio_pcm_init_info (&hw->info, as);
+    hw->samples = conf.dac_samples * conf.dac_headers;
+    wave->avail = hw->samples;
+
+    wave->pcm_buf = audio_calloc (AUDIO_FUNC, conf.dac_samples,
+                                  conf.dac_headers << hw->info.shift);
+    if (!wave->pcm_buf) {
+        goto err3;
+    }
+
+    for (i = 0; i < conf.dac_headers; ++i) {
+        WAVEHDR *h = &wave->hdrs[i];
+
+        h->dwUser = 0;
+        h->dwBufferLength = conf.dac_samples << hw->info.shift;
+        h->lpData = advance (wave->pcm_buf, i * h->dwBufferLength);
+        h->dwFlags = 0;
+
+        mr = waveOutPrepareHeader (wave->hwo, h, sizeof (*h));
+        if (mr != MMSYSERR_NOERROR) {
+            winwave_logerr (mr, "waveOutPrepareHeader(%d)", i);
+            goto err4;
+        }
+    }
+
+    return 0;
+
+ err4:
+    g_free (wave->pcm_buf);
+ err3:
+    g_free (wave->hdrs);
+ err2:
+    winwave_anal_close_out (wave);
+ err1:
+ err0:
+    return -1;
+}
+
+static int winwave_write (SWVoiceOut *sw, void *buf, int len)
+{
+    return audio_pcm_sw_write (sw, buf, len);
+}
+
+static int winwave_run_out (HWVoiceOut *hw, int live)
+{
+    WaveVoiceOut *wave = (WaveVoiceOut *) hw;
+    int decr;
+    int doreset;
+
+    EnterCriticalSection (&wave->crit_sect);
+    {
+        decr = audio_MIN (live, wave->avail);
+        decr = audio_pcm_hw_clip_out (hw, wave->pcm_buf, decr, wave->pending);
+        wave->pending += decr;
+        wave->avail -= decr;
+    }
+    LeaveCriticalSection (&wave->crit_sect);
+
+    doreset = hw->poll_mode && (wave->pending >= conf.dac_samples);
+    if (doreset && !ResetEvent (wave->event)) {
+        dolog ("DAC ResetEvent failed %lx\n", GetLastError ());
+    }
+
+    while (wave->pending >= conf.dac_samples) {
+        MMRESULT mr;
+        WAVEHDR *h = &wave->hdrs[wave->curhdr];
+
+        h->dwUser = 0;
+        mr = waveOutWrite (wave->hwo, h, sizeof (*h));
+        if (mr != MMSYSERR_NOERROR) {
+            winwave_logerr (mr, "waveOutWrite(%d)", wave->curhdr);
+            break;
+        }
+
+        wave->pending -= conf.dac_samples;
+        wave->curhdr = (wave->curhdr + 1) % conf.dac_headers;
+    }
+
+    return decr;
+}
+
+static void winwave_poll (void *opaque)
+{
+    (void) opaque;
+    audio_run ("winwave_poll");
+}
+
+static void winwave_fini_out (HWVoiceOut *hw)
+{
+    int i;
+    MMRESULT mr;
+    WaveVoiceOut *wave = (WaveVoiceOut *) hw;
+
+    mr = waveOutReset (wave->hwo);
+    if (mr != MMSYSERR_NOERROR) {
+        winwave_logerr (mr, "waveOutReset");
+    }
+
+    for (i = 0; i < conf.dac_headers; ++i) {
+        mr = waveOutUnprepareHeader (wave->hwo, &wave->hdrs[i],
+                                     sizeof (wave->hdrs[i]));
+        if (mr != MMSYSERR_NOERROR) {
+            winwave_logerr (mr, "waveOutUnprepareHeader(%d)", i);
+        }
+    }
+
+    winwave_anal_close_out (wave);
+
+    if (wave->event) {
+        qemu_del_wait_object (wave->event, winwave_poll, wave);
+        if (!CloseHandle (wave->event)) {
+            dolog ("DAC CloseHandle failed %lx\n", GetLastError ());
+        }
+        wave->event = NULL;
+    }
+
+    g_free (wave->pcm_buf);
+    wave->pcm_buf = NULL;
+
+    g_free (wave->hdrs);
+    wave->hdrs = NULL;
+}
+
+static int winwave_ctl_out (HWVoiceOut *hw, int cmd, ...)
+{
+    MMRESULT mr;
+    WaveVoiceOut *wave = (WaveVoiceOut *) hw;
+
+    switch (cmd) {
+    case VOICE_ENABLE:
+        {
+            va_list ap;
+            int poll_mode;
+
+            va_start (ap, cmd);
+            poll_mode = va_arg (ap, int);
+            va_end (ap);
+
+            if (poll_mode && !wave->event) {
+                wave->event = CreateEvent (NULL, TRUE, TRUE, NULL);
+                if (!wave->event) {
+                    dolog ("DAC CreateEvent: %lx, poll mode will be disabled\n",
+                           GetLastError ());
+                }
+            }
+
+            if (wave->event) {
+                int ret;
+
+                ret = qemu_add_wait_object (wave->event, winwave_poll, wave);
+                hw->poll_mode = (ret == 0);
+            }
+            else {
+                hw->poll_mode = 0;
+            }
+            wave->paused = 0;
+        }
+        return 0;
+
+    case VOICE_DISABLE:
+        if (!wave->paused) {
+            mr = waveOutReset (wave->hwo);
+            if (mr != MMSYSERR_NOERROR) {
+                winwave_logerr (mr, "waveOutReset");
+            }
+            else {
+                wave->paused = 1;
+            }
+        }
+        if (wave->event) {
+            qemu_del_wait_object (wave->event, winwave_poll, wave);
+        }
+        return 0;
+    }
+    return -1;
+}
+
+static void winwave_anal_close_in (WaveVoiceIn *wave)
+{
+    MMRESULT mr;
+
+    mr = waveInClose (wave->hwi);
+    if (mr != MMSYSERR_NOERROR) {
+        winwave_logerr (mr, "waveInClose");
+    }
+    wave->hwi = NULL;
+}
+
+static void CALLBACK winwave_callback_in (
+    HWAVEIN *hwi,
+    UINT msg,
+    DWORD_PTR dwInstance,
+    DWORD_PTR dwParam1,
+    DWORD_PTR dwParam2
+    )
+{
+    WaveVoiceIn *wave = (WaveVoiceIn *) dwInstance;
+
+    switch (msg) {
+    case WIM_DATA:
+        {
+            WAVEHDR *h = (WAVEHDR *) dwParam1;
+            if (!h->dwUser) {
+                h->dwUser = 1;
+                EnterCriticalSection (&wave->crit_sect);
+                {
+                    wave->avail += conf.adc_samples;
+                }
+                LeaveCriticalSection (&wave->crit_sect);
+                if (wave->hw.poll_mode) {
+                    if (!SetEvent (wave->event)) {
+                        dolog ("ADC SetEvent failed %lx\n", GetLastError ());
+                    }
+                }
+            }
+        }
+        break;
+
+    case WIM_CLOSE:
+    case WIM_OPEN:
+        break;
+
+    default:
+        dolog ("unknown wave in callback msg %x\n", msg);
+    }
+}
+
+static void winwave_add_buffers (WaveVoiceIn *wave, int samples)
+{
+    int doreset;
+
+    doreset = wave->hw.poll_mode && (samples >= conf.adc_samples);
+    if (doreset && !ResetEvent (wave->event)) {
+        dolog ("ADC ResetEvent failed %lx\n", GetLastError ());
+    }
+
+    while (samples >= conf.adc_samples) {
+        MMRESULT mr;
+        WAVEHDR *h = &wave->hdrs[wave->curhdr];
+
+        h->dwUser = 0;
+        mr = waveInAddBuffer (wave->hwi, h, sizeof (*h));
+        if (mr != MMSYSERR_NOERROR) {
+            winwave_logerr (mr, "waveInAddBuffer(%d)", wave->curhdr);
+        }
+        wave->curhdr = (wave->curhdr + 1) % conf.adc_headers;
+        samples -= conf.adc_samples;
+    }
+}
+
+static int winwave_init_in (HWVoiceIn *hw, struct audsettings *as)
+{
+    int i;
+    int err;
+    MMRESULT mr;
+    WAVEFORMATEX wfx;
+    WaveVoiceIn *wave;
+
+    wave = (WaveVoiceIn *) hw;
+
+    InitializeCriticalSection (&wave->crit_sect);
+
+    err = waveformat_from_audio_settings (&wfx, as);
+    if (err) {
+        goto err0;
+    }
+
+    mr = waveInOpen (&wave->hwi, WAVE_MAPPER, &wfx,
+                     (DWORD_PTR) winwave_callback_in,
+                     (DWORD_PTR) wave, CALLBACK_FUNCTION);
+    if (mr != MMSYSERR_NOERROR) {
+        winwave_logerr (mr, "waveInOpen");
+        goto err1;
+    }
+
+    wave->hdrs = audio_calloc (AUDIO_FUNC, conf.dac_headers,
+                               sizeof (*wave->hdrs));
+    if (!wave->hdrs) {
+        goto err2;
+    }
+
+    audio_pcm_init_info (&hw->info, as);
+    hw->samples = conf.adc_samples * conf.adc_headers;
+    wave->avail = 0;
+
+    wave->pcm_buf = audio_calloc (AUDIO_FUNC, conf.adc_samples,
+                                  conf.adc_headers << hw->info.shift);
+    if (!wave->pcm_buf) {
+        goto err3;
+    }
+
+    for (i = 0; i < conf.adc_headers; ++i) {
+        WAVEHDR *h = &wave->hdrs[i];
+
+        h->dwUser = 0;
+        h->dwBufferLength = conf.adc_samples << hw->info.shift;
+        h->lpData = advance (wave->pcm_buf, i * h->dwBufferLength);
+        h->dwFlags = 0;
+
+        mr = waveInPrepareHeader (wave->hwi, h, sizeof (*h));
+        if (mr != MMSYSERR_NOERROR) {
+            winwave_logerr (mr, "waveInPrepareHeader(%d)", i);
+            goto err4;
+        }
+    }
+
+    wave->paused = 1;
+    winwave_add_buffers (wave, hw->samples);
+    return 0;
+
+ err4:
+    g_free (wave->pcm_buf);
+ err3:
+    g_free (wave->hdrs);
+ err2:
+    winwave_anal_close_in (wave);
+ err1:
+ err0:
+    return -1;
+}
+
+static void winwave_fini_in (HWVoiceIn *hw)
+{
+    int i;
+    MMRESULT mr;
+    WaveVoiceIn *wave = (WaveVoiceIn *) hw;
+
+    mr = waveInReset (wave->hwi);
+    if (mr != MMSYSERR_NOERROR) {
+        winwave_logerr (mr, "waveInReset");
+    }
+
+    for (i = 0; i < conf.adc_headers; ++i) {
+        mr = waveInUnprepareHeader (wave->hwi, &wave->hdrs[i],
+                                     sizeof (wave->hdrs[i]));
+        if (mr != MMSYSERR_NOERROR) {
+            winwave_logerr (mr, "waveInUnprepareHeader(%d)", i);
+        }
+    }
+
+    winwave_anal_close_in (wave);
+
+    if (wave->event) {
+        qemu_del_wait_object (wave->event, winwave_poll, wave);
+        if (!CloseHandle (wave->event)) {
+            dolog ("ADC CloseHandle failed %lx\n", GetLastError ());
+        }
+        wave->event = NULL;
+    }
+
+    g_free (wave->pcm_buf);
+    wave->pcm_buf = NULL;
+
+    g_free (wave->hdrs);
+    wave->hdrs = NULL;
+}
+
+static int winwave_run_in (HWVoiceIn *hw)
+{
+    WaveVoiceIn *wave = (WaveVoiceIn *) hw;
+    int live = audio_pcm_hw_get_live_in (hw);
+    int dead = hw->samples - live;
+    int decr, ret;
+
+    if (!dead) {
+        return 0;
+    }
+
+    EnterCriticalSection (&wave->crit_sect);
+    {
+        decr = audio_MIN (dead, wave->avail);
+        wave->avail -= decr;
+    }
+    LeaveCriticalSection (&wave->crit_sect);
+
+    ret = decr;
+    while (decr) {
+        int left = hw->samples - hw->wpos;
+        int conv = audio_MIN (left, decr);
+        hw->conv (hw->conv_buf + hw->wpos,
+                  advance (wave->pcm_buf, wave->rpos << hw->info.shift),
+                  conv);
+
+        wave->rpos = (wave->rpos + conv) % hw->samples;
+        hw->wpos = (hw->wpos + conv) % hw->samples;
+        decr -= conv;
+    }
+
+    winwave_add_buffers (wave, ret);
+    return ret;
+}
+
+static int winwave_read (SWVoiceIn *sw, void *buf, int size)
+{
+    return audio_pcm_sw_read (sw, buf, size);
+}
+
+static int winwave_ctl_in (HWVoiceIn *hw, int cmd, ...)
+{
+    MMRESULT mr;
+    WaveVoiceIn *wave = (WaveVoiceIn *) hw;
+
+    switch (cmd) {
+    case VOICE_ENABLE:
+        {
+            va_list ap;
+            int poll_mode;
+
+            va_start (ap, cmd);
+            poll_mode = va_arg (ap, int);
+            va_end (ap);
+
+            if (poll_mode && !wave->event) {
+                wave->event = CreateEvent (NULL, TRUE, TRUE, NULL);
+                if (!wave->event) {
+                    dolog ("ADC CreateEvent: %lx, poll mode will be disabled\n",
+                           GetLastError ());
+                }
+            }
+
+            if (wave->event) {
+                int ret;
+
+                ret = qemu_add_wait_object (wave->event, winwave_poll, wave);
+                hw->poll_mode = (ret == 0);
+            }
+            else {
+                hw->poll_mode = 0;
+            }
+            if (wave->paused) {
+                mr = waveInStart (wave->hwi);
+                if (mr != MMSYSERR_NOERROR) {
+                    winwave_logerr (mr, "waveInStart");
+                }
+                wave->paused = 0;
+            }
+        }
+        return 0;
+
+    case VOICE_DISABLE:
+        if (!wave->paused) {
+            mr = waveInStop (wave->hwi);
+            if (mr != MMSYSERR_NOERROR) {
+                winwave_logerr (mr, "waveInStop");
+            }
+            else {
+                wave->paused = 1;
+            }
+        }
+        if (wave->event) {
+            qemu_del_wait_object (wave->event, winwave_poll, wave);
+        }
+        return 0;
+    }
+    return 0;
+}
+
+static void *winwave_audio_init (void)
+{
+    return &conf;
+}
+
+static void winwave_audio_fini (void *opaque)
+{
+    (void) opaque;
+}
+
+static struct audio_option winwave_options[] = {
+    {
+        .name        = "DAC_HEADERS",
+        .tag         = AUD_OPT_INT,
+        .valp        = &conf.dac_headers,
+        .descr       = "DAC number of headers",
+    },
+    {
+        .name        = "DAC_SAMPLES",
+        .tag         = AUD_OPT_INT,
+        .valp        = &conf.dac_samples,
+        .descr       = "DAC number of samples per header",
+    },
+    {
+        .name        = "ADC_HEADERS",
+        .tag         = AUD_OPT_INT,
+        .valp        = &conf.adc_headers,
+        .descr       = "ADC number of headers",
+    },
+    {
+        .name        = "ADC_SAMPLES",
+        .tag         = AUD_OPT_INT,
+        .valp        = &conf.adc_samples,
+        .descr       = "ADC number of samples per header",
+    },
+    { /* End of list */ }
+};
+
+static struct audio_pcm_ops winwave_pcm_ops = {
+    .init_out = winwave_init_out,
+    .fini_out = winwave_fini_out,
+    .run_out  = winwave_run_out,
+    .write    = winwave_write,
+    .ctl_out  = winwave_ctl_out,
+    .init_in  = winwave_init_in,
+    .fini_in  = winwave_fini_in,
+    .run_in   = winwave_run_in,
+    .read     = winwave_read,
+    .ctl_in   = winwave_ctl_in
+};
+
+struct audio_driver winwave_audio_driver = {
+    .name           = "winwave",
+    .descr          = "Windows Waveform Audio http://msdn.microsoft.com",
+    .options        = winwave_options,
+    .init           = winwave_audio_init,
+    .fini           = winwave_audio_fini,
+    .pcm_ops        = &winwave_pcm_ops,
+    .can_be_default = 1,
+    .max_voices_out = INT_MAX,
+    .max_voices_in  = INT_MAX,
+    .voice_size_out = sizeof (WaveVoiceOut),
+    .voice_size_in  = sizeof (WaveVoiceIn)
+};
--- a/backends/baum.c
+++ b/backends/baum.c
@@ -21,7 +21,6 @@
 * 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 "sysemu/char.h"
 #include "qemu/timer.h"
@@ -304,7 +303,7 @@ static int baum_eat_packet(BaumDriverState *baum, const uint8_t *buf, int len)
                return 0;
            cur++;
        }
-        DPRINTF("Dropped %td bytes!\n", cur - buf);
+        DPRINTF("Dropped %d bytes!\n", cur - buf);
    }

 #define EAT(c) do {\
@@ -562,12 +561,8 @@ static void baum_close(struct CharDriverState *chr)
    g_free(baum);
 }

-static CharDriverState *chr_baum_init(const char *id,
-                                      ChardevBackend *backend,
-                                      ChardevReturn *ret,
-                                      Error **errp)
+CharDriverState *chr_baum_init(void)
 {
-    ChardevCommon *common = qapi_ChardevDummy_base(backend->u.braille);
    BaumDriverState *baum;
    CharDriverState *chr;
    brlapi_handle_t *handle;
@@ -578,12 +573,8 @@ static CharDriverState *chr_baum_init(const char *id,
 #endif
    int tty;

-    chr = qemu_chr_alloc(common, errp);
-    if (!chr) {
-        return NULL;
-    }
    baum = g_malloc0(sizeof(BaumDriverState));
-    baum->chr = chr;
+    baum->chr = chr = qemu_chr_alloc();

    chr->opaque = baum;
    chr->chr_write = baum_write;
@@ -595,16 +586,14 @@ static CharDriverState *chr_baum_init(const char *id,

    baum->brlapi_fd = brlapi__openConnection(handle, NULL, NULL);
    if (baum->brlapi_fd == -1) {
-        error_setg(errp, "brlapi__openConnection: %s",
-                   brlapi_strerror(brlapi_error_location()));
+        brlapi_perror("baum_init: brlapi_openConnection");
        goto fail_handle;
    }

    baum->cellCount_timer = timer_new_ns(QEMU_CLOCK_VIRTUAL, baum_cellCount_timer_cb, baum);

    if (brlapi__getDisplaySize(handle, &baum->x, &baum->y) == -1) {
-        error_setg(errp, "brlapi__getDisplaySize: %s",
-                   brlapi_strerror(brlapi_error_location()));
+        brlapi_perror("baum_init: brlapi_getDisplaySize");
        goto fail;
    }

@@ -620,8 +609,7 @@ static CharDriverState *chr_baum_init(const char *id,
        tty = BRLAPI_TTY_DEFAULT;

    if (brlapi__enterTtyMode(handle, tty, NULL) == -1) {
-        error_setg(errp, "brlapi__enterTtyMode: %s",
-                   brlapi_strerror(brlapi_error_location()));
+        brlapi_perror("baum_init: brlapi_enterTtyMode");
        goto fail;
    }

@@ -641,8 +629,7 @@ fail_handle:

 static void register_types(void)
 {
-    register_char_driver("braille", CHARDEV_BACKEND_KIND_BRAILLE, NULL,
-                         chr_baum_init);
+    register_char_driver("braille", CHARDEV_BACKEND_KIND_BRAILLE, NULL);
 }

 type_init(register_types);
--- a/backends/hostmem-file.c
+++ b/backends/hostmem-file.c
@@ -9,7 +9,6 @@
 * This work is licensed under the terms of the GNU GPL, version 2 or later.
 * See the COPYING file in the top-level directory.
 */
-#include "qemu/osdep.h"
 #include "qemu-common.h"
 #include "sysemu/hostmem.h"
 #include "sysemu/sysemu.h"
@@ -44,7 +43,7 @@ file_backend_memory_alloc(HostMemoryBackend *backend, Error **errp)
        return;
    }
    if (!fb->mem_path) {
-        error_setg(errp, "mem-path property not set");
+        error_setg(errp, "mem_path property not set");
        return;
    }
 #ifndef CONFIG_LINUX
@@ -84,7 +83,9 @@ static void set_mem_path(Object *o, const char *str, Error **errp)
        error_setg(errp, "cannot change property value");
        return;
    }
-    g_free(fb->mem_path);
+    if (fb->mem_path) {
+        g_free(fb->mem_path);
+    }
    fb->mem_path = g_strdup(str);
 }

--- a/backends/hostmem-ram.c
+++ b/backends/hostmem-ram.c
@@ -9,7 +9,6 @@
 * This work is licensed under the terms of the GNU GPL, version 2 or later.
 * See the COPYING file in the top-level directory.
 */
-#include "qemu/osdep.h"
 #include "sysemu/hostmem.h"
 #include "qom/object_interfaces.h"

--- a/backends/hostmem.c
+++ b/backends/hostmem.c
@@ -9,12 +9,11 @@
 * This work is licensed under the terms of the GNU GPL, version 2 or later.
 * See the COPYING file in the top-level directory.
 */
-#include "qemu/osdep.h"
 #include "sysemu/hostmem.h"
-#include "hw/boards.h"
 #include "qapi/visitor.h"
 #include "qapi-types.h"
 #include "qapi-visit.h"
+#include "qapi/qmp/qerror.h"
 #include "qemu/config-file.h"
 #include "qom/object_interfaces.h"

@@ -27,18 +26,18 @@ QEMU_BUILD_BUG_ON(HOST_MEM_POLICY_INTERLEAVE != MPOL_INTERLEAVE);
 #endif

 static void
-host_memory_backend_get_size(Object *obj, Visitor *v, const char *name,
-                             void *opaque, Error **errp)
+host_memory_backend_get_size(Object *obj, Visitor *v, void *opaque,
+                             const char *name, Error **errp)
 {
    HostMemoryBackend *backend = MEMORY_BACKEND(obj);
    uint64_t value = backend->size;

-    visit_type_size(v, name, &value, errp);
+    visit_type_size(v, &value, name, errp);
 }

 static void
-host_memory_backend_set_size(Object *obj, Visitor *v, const char *name,
-                             void *opaque, Error **errp)
+host_memory_backend_set_size(Object *obj, Visitor *v, void *opaque,
+                             const char *name, Error **errp)
 {
    HostMemoryBackend *backend = MEMORY_BACKEND(obj);
    Error *local_err = NULL;
@@ -49,7 +48,7 @@ host_memory_backend_set_size(Object *obj, Visitor *v, const char *name,
        goto out;
    }

-    visit_type_size(v, name, &value, &local_err);
+    visit_type_size(v, &value, name, &local_err);
    if (local_err) {
        goto out;
    }
@@ -64,8 +63,8 @@ out:
 }

 static void
-host_memory_backend_get_host_nodes(Object *obj, Visitor *v, const char *name,
-                                   void *opaque, Error **errp)
+host_memory_backend_get_host_nodes(Object *obj, Visitor *v, void *opaque,
+                                   const char *name, Error **errp)
 {
    HostMemoryBackend *backend = MEMORY_BACKEND(obj);
    uint16List *host_nodes = NULL;
@@ -92,18 +91,18 @@ host_memory_backend_get_host_nodes(Object *obj, Visitor *v, const char *name,
        node = &(*node)->next;
    } while (true);

-    visit_type_uint16List(v, name, &host_nodes, errp);
+    visit_type_uint16List(v, &host_nodes, name, errp);
 }

 static void
-host_memory_backend_set_host_nodes(Object *obj, Visitor *v, const char *name,
-                                   void *opaque, Error **errp)
+host_memory_backend_set_host_nodes(Object *obj, Visitor *v, void *opaque,
+                                   const char *name, Error **errp)
 {
 #ifdef CONFIG_NUMA
    HostMemoryBackend *backend = MEMORY_BACKEND(obj);
    uint16List *l = NULL;

-    visit_type_uint16List(v, name, &l, errp);
+    visit_type_uint16List(v, &l, name, errp);

    while (l) {
        bitmap_set(backend->host_nodes, l->value, 1);
@@ -114,17 +113,24 @@ host_memory_backend_set_host_nodes(Object *obj, Visitor *v, const char *name,
 #endif
 }

-static int
-host_memory_backend_get_policy(Object *obj, Error **errp G_GNUC_UNUSED)
+static void
+host_memory_backend_get_policy(Object *obj, Visitor *v, void *opaque,
+                               const char *name, Error **errp)
 {
    HostMemoryBackend *backend = MEMORY_BACKEND(obj);
-    return backend->policy;
+    int policy = backend->policy;
+
+    visit_type_enum(v, &policy, HostMemPolicy_lookup, NULL, name, errp);
 }

 static void
-host_memory_backend_set_policy(Object *obj, int policy, Error **errp)
+host_memory_backend_set_policy(Object *obj, Visitor *v, void *opaque,
+                               const char *name, Error **errp)
 {
    HostMemoryBackend *backend = MEMORY_BACKEND(obj);
+    int policy;
+
+    visit_type_enum(v, &policy, HostMemPolicy_lookup, NULL, name, errp);
    backend->policy = policy;

 #ifndef CONFIG_NUMA
@@ -224,10 +230,11 @@ static void host_memory_backend_set_prealloc(Object *obj, bool value,
 static void host_memory_backend_init(Object *obj)
 {
    HostMemoryBackend *backend = MEMORY_BACKEND(obj);
-    MachineState *machine = MACHINE(qdev_get_machine());

-    backend->merge = machine_mem_merge(machine);
-    backend->dump = machine_dump_guest_core(machine);
+    backend->merge = qemu_opt_get_bool(qemu_get_machine_opts(),
+                                       "mem-merge", true);
+    backend->dump = qemu_opt_get_bool(qemu_get_machine_opts(),
+                                      "dump-guest-core", true);
    backend->prealloc = mem_prealloc;

    object_property_add_bool(obj, "merge",
@@ -245,10 +252,9 @@ static void host_memory_backend_init(Object *obj)
    object_property_add(obj, "host-nodes", "int",
                        host_memory_backend_get_host_nodes,
                        host_memory_backend_set_host_nodes, NULL, NULL, NULL);
-    object_property_add_enum(obj, "policy", "HostMemPolicy",
-                             HostMemPolicy_lookup,
-                             host_memory_backend_get_policy,
-                             host_memory_backend_set_policy, NULL);
+    object_property_add(obj, "policy", "str",
+                        host_memory_backend_get_policy,
+                        host_memory_backend_set_policy, NULL, NULL, NULL);
 }

 MemoryRegion *
@@ -314,11 +320,9 @@ host_memory_backend_memory_complete(UserCreatable *uc, Error **errp)
        assert(maxnode <= MAX_NODES);
        if (mbind(ptr, sz, backend->policy,
                  maxnode ? backend->host_nodes : NULL, maxnode + 1, flags)) {
-            if (backend->policy != MPOL_DEFAULT || errno != ENOSYS) {
-                error_setg_errno(errp, errno,
-                                 "cannot bind memory to host NUMA nodes");
-                return;
-            }
+            error_setg_errno(errp, errno,
+                             "cannot bind memory to host NUMA nodes");
+            return;
        }
 #endif
        /* Preallocate memory after the NUMA policy has been instantiated.
--- a/backends/msmouse.c
+++ b/backends/msmouse.c
@@ -21,7 +21,7 @@
 * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
 * THE SOFTWARE.
 */
-#include "qemu/osdep.h"
+#include <stdlib.h>
 #include "qemu-common.h"
 #include "sysemu/char.h"
 #include "ui/console.h"
@@ -63,18 +63,11 @@ static void msmouse_chr_close (struct CharDriverState *chr)
    g_free (chr);
 }

-static CharDriverState *qemu_chr_open_msmouse(const char *id,
-                                              ChardevBackend *backend,
-                                              ChardevReturn *ret,
-                                              Error **errp)
+CharDriverState *qemu_chr_open_msmouse(void)
 {
-    ChardevCommon *common = qapi_ChardevDummy_base(backend->u.msmouse);
    CharDriverState *chr;

-    chr = qemu_chr_alloc(common, errp);
-    if (!chr) {
-        return NULL;
-    }
+    chr = qemu_chr_alloc();
    chr->chr_write = msmouse_chr_write;
    chr->chr_close = msmouse_chr_close;
    chr->explicit_be_open = true;
@@ -86,8 +79,7 @@ static CharDriverState *qemu_chr_open_msmouse(const char *id,

 static void register_types(void)
 {
-    register_char_driver("msmouse", CHARDEV_BACKEND_KIND_MSMOUSE, NULL,
-                         qemu_chr_open_msmouse);
+    register_char_driver("msmouse", CHARDEV_BACKEND_KIND_MSMOUSE, NULL);
 }

 type_init(register_types);
--- a/backends/rng-egd.c
+++ b/backends/rng-egd.c
@@ -10,7 +10,6 @@
 * See the COPYING file in the top-level directory.
 */

-#include "qemu/osdep.h"
 #include "sysemu/rng.h"
 #include "sysemu/char.h"
 #include "qapi/qmp/qerror.h"
@@ -141,20 +140,19 @@ static void rng_egd_opened(RngBackend *b, Error **errp)
    RngEgd *s = RNG_EGD(b);

    if (s->chr_name == NULL) {
-        error_setg(errp, QERR_INVALID_PARAMETER_VALUE,
-                   "chardev", "a valid character device");
+        error_set(errp, QERR_INVALID_PARAMETER_VALUE,
+                  "chardev", "a valid character device");
        return;
    }

    s->chr = qemu_chr_find(s->chr_name);
    if (s->chr == NULL) {
-        error_set(errp, ERROR_CLASS_DEVICE_NOT_FOUND,
-                  "Device '%s' not found", s->chr_name);
+        error_set(errp, QERR_DEVICE_NOT_FOUND, s->chr_name);
        return;
    }

    if (qemu_chr_fe_claim(s->chr) != 0) {
-        error_setg(errp, QERR_DEVICE_IN_USE, s->chr_name);
+        error_set(errp, QERR_DEVICE_IN_USE, s->chr_name);
        return;
    }

@@ -169,7 +167,7 @@ static void rng_egd_set_chardev(Object *obj, const char *value, Error **errp)
    RngEgd *s = RNG_EGD(b);

    if (b->opened) {
-        error_setg(errp, QERR_PERMISSION_DENIED);
+        error_set(errp, QERR_PERMISSION_DENIED);
    } else {
        g_free(s->chr_name);
        s->chr_name = g_strdup(value);
--- a/backends/rng-random.c
+++ b/backends/rng-random.c
@@ -10,7 +10,6 @@
 * See the COPYING file in the top-level directory.
 */

-#include "qemu/osdep.h"
 #include "sysemu/rng-random.h"
 #include "sysemu/rng.h"
 #include "qapi/qmp/qerror.h"
@@ -75,8 +74,8 @@ static void rng_random_opened(RngBackend *b, Error **errp)
    RndRandom *s = RNG_RANDOM(b);

    if (s->filename == NULL) {
-        error_setg(errp, QERR_INVALID_PARAMETER_VALUE,
-                   "filename", "a valid filename");
+        error_set(errp, QERR_INVALID_PARAMETER_VALUE,
+                  "filename", "a valid filename");
    } else {
        s->fd = qemu_open(s->filename, O_RDONLY | O_NONBLOCK);
        if (s->fd == -1) {
@@ -99,7 +98,7 @@ static void rng_random_set_filename(Object *obj, const char *filename,
    RndRandom *s = RNG_RANDOM(obj);

    if (b->opened) {
-        error_setg(errp, QERR_PERMISSION_DENIED);
+        error_set(errp, QERR_PERMISSION_DENIED);
        return;
    }

--- a/backends/rng.c
+++ b/backends/rng.c
@@ -10,7 +10,6 @@
 * See the COPYING file in the top-level directory.
 */

-#include "qemu/osdep.h"
 #include "sysemu/rng.h"
 #include "qapi/qmp/qerror.h"
 #include "qom/object_interfaces.h"
@@ -58,7 +57,7 @@ static void rng_backend_prop_set_opened(Object *obj, bool value, Error **errp)
    }

    if (!value && s->opened) {
-        error_setg(errp, QERR_PERMISSION_DENIED);
+        error_set(errp, QERR_PERMISSION_DENIED);
        return;
    }

--- a/backends/testdev.c
+++ b/backends/testdev.c
@@ -23,7 +23,6 @@
 * 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 "sysemu/char.h"

@@ -109,16 +108,13 @@ static void testdev_close(struct CharDriverState *chr)
    g_free(testdev);
 }

-static CharDriverState *chr_testdev_init(const char *id,
-                                         ChardevBackend *backend,
-                                         ChardevReturn *ret,
-                                         Error **errp)
+CharDriverState *chr_testdev_init(void)
 {
    TestdevCharState *testdev;
    CharDriverState *chr;

-    testdev = g_new0(TestdevCharState, 1);
-    testdev->chr = chr = g_new0(CharDriverState, 1);
+    testdev = g_malloc0(sizeof(TestdevCharState));
+    testdev->chr = chr = g_malloc0(sizeof(CharDriverState));

    chr->opaque = testdev;
    chr->chr_write = testdev_write;
@@ -129,8 +125,7 @@ static CharDriverState *chr_testdev_init(const char *id,

 static void register_types(void)
 {
-    register_char_driver("testdev", CHARDEV_BACKEND_KIND_TESTDEV, NULL,
-                         chr_testdev_init);
+    register_char_driver("testdev", CHARDEV_BACKEND_KIND_TESTDEV, NULL);
 }

 type_init(register_types);
--- a/backends/tpm.c
+++ b/backends/tpm.c
@@ -12,7 +12,6 @@
 * Based on backends/rng.c by Anthony Liguori
 */

-#include "qemu/osdep.h"
 #include "sysemu/tpm_backend.h"
 #include "qapi/qmp/qerror.h"
 #include "sysemu/tpm.h"
@@ -97,20 +96,6 @@ bool tpm_backend_get_tpm_established_flag(TPMBackend *s)
    return k->ops->get_tpm_established_flag(s);
 }

-int tpm_backend_reset_tpm_established_flag(TPMBackend *s, uint8_t locty)
-{
-    TPMBackendClass *k = TPM_BACKEND_GET_CLASS(s);
-
-    return k->ops->reset_tpm_established_flag(s, locty);
-}
-
-TPMVersion tpm_backend_get_tpm_version(TPMBackend *s)
-{
-    TPMBackendClass *k = TPM_BACKEND_GET_CLASS(s);
-
-    return k->ops->get_tpm_version(s);
-}
-
 static bool tpm_backend_prop_get_opened(Object *obj, Error **errp)
 {
    TPMBackend *s = TPM_BACKEND(obj);
@@ -134,7 +119,7 @@ static void tpm_backend_prop_set_opened(Object *obj, bool value, Error **errp)
    }

    if (!value && s->opened) {
-        error_setg(errp, QERR_PERMISSION_DENIED);
+        error_set(errp, QERR_PERMISSION_DENIED);
        return;
    }

@@ -180,6 +165,17 @@ void tpm_backend_thread_end(TPMBackendThread *tbt)
    }
 }

+void tpm_backend_thread_tpm_reset(TPMBackendThread *tbt,
+                                  GFunc func, gpointer user_data)
+{
+    if (!tbt->pool) {
+        tpm_backend_thread_create(tbt, func, user_data);
+    } else {
+        g_thread_pool_push(tbt->pool, (gpointer)TPM_BACKEND_CMD_TPM_RESET,
+                           NULL);
+    }
+}
+
 static const TypeInfo tpm_backend_info = {
    .name = TYPE_TPM_BACKEND,
    .parent = TYPE_OBJECT,
--- a/balloon.c
+++ b/balloon.c
@@ -24,30 +24,17 @@
 * THE SOFTWARE.
 */

-#include "qemu/osdep.h"
-#include "qemu-common.h"
+#include "monitor/monitor.h"
 #include "exec/cpu-common.h"
 #include "sysemu/kvm.h"
 #include "sysemu/balloon.h"
 #include "trace.h"
 #include "qmp-commands.h"
-#include "qapi/qmp/qerror.h"
 #include "qapi/qmp/qjson.h"

 static QEMUBalloonEvent *balloon_event_fn;
 static QEMUBalloonStatus *balloon_stat_fn;
 static void *balloon_opaque;
-static bool balloon_inhibited;
-
-bool qemu_balloon_is_inhibited(void)
-{
-    return balloon_inhibited;
-}
-
-void qemu_balloon_inhibit(bool state)
-{
-    balloon_inhibited = state;
-}

 static bool have_balloon(Error **errp)
 {
@@ -71,6 +58,7 @@ int qemu_add_balloon_handler(QEMUBalloonEvent *event_func,
        /* We're already registered one balloon handler.  How many can
         * a guest really have?
         */
+        error_report("Another balloon device already registered");
        return -1;
    }
    balloon_event_fn = event_func;
@@ -109,7 +97,7 @@ void qmp_balloon(int64_t target, Error **errp)
    }

    if (target <= 0) {
-        error_setg(errp, QERR_INVALID_PARAMETER_VALUE, "target", "a size");
+        error_set(errp, QERR_INVALID_PARAMETER_VALUE, "target", "a size");
        return;
    }

--- a/block.c
+++ b/block.c
--- a/block/Makefile.objs
+++ b/block/Makefile.objs
@@ -1,16 +1,15 @@
-block-obj-y += raw_bsd.o qcow.o vdi.o vmdk.o cloop.o bochs.o vpc.o vvfat.o
+block-obj-y += raw_bsd.o qcow.o vdi.o vmdk.o cloop.o dmg.o bochs.o vpc.o vvfat.o
 block-obj-y += qcow2.o qcow2-refcount.o qcow2-cluster.o qcow2-snapshot.o qcow2-cache.o
 block-obj-y += qed.o qed-gencb.o qed-l2-cache.o qed-table.o qed-cluster.o
 block-obj-y += qed-check.o
 block-obj-$(CONFIG_VHDX) += vhdx.o vhdx-endian.o vhdx-log.o
-block-obj-y += quorum.o
+block-obj-$(CONFIG_QUORUM) += quorum.o
 block-obj-y += parallels.o blkdebug.o blkverify.o
 block-obj-y += block-backend.o snapshot.o qapi.o
 block-obj-$(CONFIG_WIN32) += raw-win32.o win32-aio.o
 block-obj-$(CONFIG_POSIX) += raw-posix.o
 block-obj-$(CONFIG_LINUX_AIO) += linux-aio.o
-block-obj-y += null.o mirror.o io.o
-block-obj-y += throttle-groups.o
+block-obj-y += null.o mirror.o

 block-obj-y += nbd.o nbd-client.o sheepdog.o
 block-obj-$(CONFIG_LIBISCSI) += iscsi.o
@@ -20,6 +19,8 @@ block-obj-$(CONFIG_RBD) += rbd.o
 block-obj-$(CONFIG_GLUSTERFS) += gluster.o
 block-obj-$(CONFIG_ARCHIPELAGO) += archipelago.o
 block-obj-$(CONFIG_LIBSSH2) += ssh.o
+block-obj-y += dictzip.o
+block-obj-y += tar.o
 block-obj-y += accounting.o
 block-obj-y += write-threshold.o

@@ -38,7 +39,6 @@ gluster.o-libs     := $(GLUSTERFS_LIBS)
 ssh.o-cflags       := $(LIBSSH2_CFLAGS)
 ssh.o-libs         := $(LIBSSH2_LIBS)
 archipelago.o-libs := $(ARCHIPELAGO_LIBS)
-block-obj-m        += dmg.o
 dmg.o-libs         := $(BZIP2_LIBS)
 qcow.o-libs        := -lz
 linux-aio.o-libs   := -laio
--- a/block/accounting.c
+++ b/block/accounting.c
@@ -2,7 +2,6 @@
 * QEMU System Emulator block accounting
 *
 * Copyright (c) 2011 Christoph Hellwig
- * Copyright (c) 2015 Igalia, S.L.
 *
 * Permission is hereby granted, free of charge, to any person obtaining a copy
 * of this software and associated documentation files (the "Software"), to deal
@@ -23,58 +22,9 @@
 * THE SOFTWARE.
 */

-#include "qemu/osdep.h"
 #include "block/accounting.h"
 #include "block/block_int.h"
 #include "qemu/timer.h"
-#include "sysemu/qtest.h"
-
-static QEMUClockType clock_type = QEMU_CLOCK_REALTIME;
-static const int qtest_latency_ns = NANOSECONDS_PER_SECOND / 1000;
-
-void block_acct_init(BlockAcctStats *stats, bool account_invalid,
-                     bool account_failed)
-{
-    stats->account_invalid = account_invalid;
-    stats->account_failed = account_failed;
-
-    if (qtest_enabled()) {
-        clock_type = QEMU_CLOCK_VIRTUAL;
-    }
-}
-
-void block_acct_cleanup(BlockAcctStats *stats)
-{
-    BlockAcctTimedStats *s, *next;
-    QSLIST_FOREACH_SAFE(s, &stats->intervals, entries, next) {
-        g_free(s);
-    }
-}
-
-void block_acct_add_interval(BlockAcctStats *stats, unsigned interval_length)
-{
-    BlockAcctTimedStats *s;
-    unsigned i;
-
-    s = g_new0(BlockAcctTimedStats, 1);
-    s->interval_length = interval_length;
-    QSLIST_INSERT_HEAD(&stats->intervals, s, entries);
-
-    for (i = 0; i < BLOCK_MAX_IOTYPE; i++) {
-        timed_average_init(&s->latency[i], clock_type,
-                           (uint64_t) interval_length * NANOSECONDS_PER_SECOND);
-    }
-}
-
-BlockAcctTimedStats *block_acct_interval_next(BlockAcctStats *stats,
-                                              BlockAcctTimedStats *s)
-{
-    if (s == NULL) {
-        return QSLIST_FIRST(&stats->intervals);
-    } else {
-        return QSLIST_NEXT(s, entries);
-    }
-}

 void block_acct_start(BlockAcctStats *stats, BlockAcctCookie *cookie,
                      int64_t bytes, enum BlockAcctType type)
@@ -82,69 +32,26 @@ void block_acct_start(BlockAcctStats *stats, BlockAcctCookie *cookie,
    assert(type < BLOCK_MAX_IOTYPE);

    cookie->bytes = bytes;
-    cookie->start_time_ns = qemu_clock_get_ns(clock_type);
+    cookie->start_time_ns = qemu_clock_get_ns(QEMU_CLOCK_REALTIME);
    cookie->type = type;
 }

 void block_acct_done(BlockAcctStats *stats, BlockAcctCookie *cookie)
 {
-    BlockAcctTimedStats *s;
-    int64_t time_ns = qemu_clock_get_ns(clock_type);
-    int64_t latency_ns = time_ns - cookie->start_time_ns;
-
-    if (qtest_enabled()) {
-        latency_ns = qtest_latency_ns;
-    }
-
    assert(cookie->type < BLOCK_MAX_IOTYPE);

    stats->nr_bytes[cookie->type] += cookie->bytes;
    stats->nr_ops[cookie->type]++;
-    stats->total_time_ns[cookie->type] += latency_ns;
-    stats->last_access_time_ns = time_ns;
-
-    QSLIST_FOREACH(s, &stats->intervals, entries) {
-        timed_average_account(&s->latency[cookie->type], latency_ns);
-    }
+    stats->total_time_ns[cookie->type] +=
+        qemu_clock_get_ns(QEMU_CLOCK_REALTIME) - cookie->start_time_ns;
 }

-void block_acct_failed(BlockAcctStats *stats, BlockAcctCookie *cookie)
+
+void block_acct_highest_sector(BlockAcctStats *stats, int64_t sector_num,
+                               unsigned int nb_sectors)
 {
-    assert(cookie->type < BLOCK_MAX_IOTYPE);
-
-    stats->failed_ops[cookie->type]++;
-
-    if (stats->account_failed) {
-        BlockAcctTimedStats *s;
-        int64_t time_ns = qemu_clock_get_ns(clock_type);
-        int64_t latency_ns = time_ns - cookie->start_time_ns;
-
-        if (qtest_enabled()) {
-            latency_ns = qtest_latency_ns;
-        }
-
-        stats->total_time_ns[cookie->type] += latency_ns;
-        stats->last_access_time_ns = time_ns;
-
-        QSLIST_FOREACH(s, &stats->intervals, entries) {
-            timed_average_account(&s->latency[cookie->type], latency_ns);
-        }
-    }
-}
-
-void block_acct_invalid(BlockAcctStats *stats, enum BlockAcctType type)
-{
-    assert(type < BLOCK_MAX_IOTYPE);
-
-    /* block_acct_done() and block_acct_failed() update
-     * total_time_ns[], but this one does not. The reason is that
-     * invalid requests are accounted during their submission,
-     * therefore there's no actual I/O involved. */
-
-    stats->invalid_ops[type]++;
-
-    if (stats->account_invalid) {
-        stats->last_access_time_ns = qemu_clock_get_ns(clock_type);
+    if (stats->wr_highest_sector < sector_num + nb_sectors - 1) {
+        stats->wr_highest_sector = sector_num + nb_sectors - 1;
    }
 }

@@ -154,20 +61,3 @@ void block_acct_merge_done(BlockAcctStats *stats, enum BlockAcctType type,
    assert(type < BLOCK_MAX_IOTYPE);
    stats->merged[type] += num_requests;
 }
-
-int64_t block_acct_idle_time_ns(BlockAcctStats *stats)
-{
-    return qemu_clock_get_ns(clock_type) - stats->last_access_time_ns;
-}
-
-double block_acct_queue_depth(BlockAcctTimedStats *stats,
-                              enum BlockAcctType type)
-{
-    uint64_t sum, elapsed;
-
-    assert(type < BLOCK_MAX_IOTYPE);
-
-    sum = timed_average_sum(&stats->latency[type], &elapsed);
-
-    return (double) sum / elapsed;
-}
--- a/block/archipelago.c
+++ b/block/archipelago.c
@@ -50,7 +50,6 @@
 *
 */

-#include "qemu/osdep.h"
 #include "qemu-common.h"
 #include "block/block_int.h"
 #include "qemu/error-report.h"
@@ -60,6 +59,7 @@
 #include "qapi/qmp/qjson.h"
 #include "qemu/atomic.h"

+#include <inttypes.h>
 #include <xseg/xseg.h>
 #include <xseg/protocol.h>

--- a/block/backup.c
+++ b/block/backup.c
@@ -11,15 +11,15 @@
 *
 */

-#include "qemu/osdep.h"
+#include <stdio.h>
+#include <errno.h>
+#include <unistd.h>

 #include "trace.h"
 #include "block/block.h"
 #include "block/block_int.h"
 #include "block/blockjob.h"
-#include "qapi/qmp/qerror.h"
 #include "qemu/ratelimit.h"
-#include "sysemu/block-backend.h"

 #define BACKUP_CLUSTER_BITS 16
 #define BACKUP_CLUSTER_SIZE (1 << BACKUP_CLUSTER_BITS)
@@ -37,8 +37,6 @@ typedef struct CowRequest {
 typedef struct BackupBlockJob {
    BlockJob common;
    BlockDriverState *target;
-    /* bitmap for sync=incremental */
-    BdrvDirtyBitmap *sync_bitmap;
    MirrorSyncMode sync_mode;
    RateLimit limit;
    BlockdevOnError on_source_error;
@@ -88,8 +86,7 @@ static void cow_request_end(CowRequest *req)

 static int coroutine_fn backup_do_cow(BlockDriverState *bs,
                                      int64_t sector_num, int nb_sectors,
-                                      bool *error_is_read,
-                                      bool is_write_notifier)
+                                      bool *error_is_read)
 {
    BackupBlockJob *job = (BackupBlockJob *)bs->job;
    CowRequest cow_request;
@@ -129,14 +126,8 @@ static int coroutine_fn backup_do_cow(BlockDriverState *bs,
        iov.iov_len = n * BDRV_SECTOR_SIZE;
        qemu_iovec_init_external(&bounce_qiov, &iov, 1);

-        if (is_write_notifier) {
-            ret = bdrv_co_readv_no_serialising(bs,
-                                           start * BACKUP_SECTORS_PER_CLUSTER,
-                                           n, &bounce_qiov);
-        } else {
-            ret = bdrv_co_readv(bs, start * BACKUP_SECTORS_PER_CLUSTER, n,
-                                &bounce_qiov);
-        }
+        ret = bdrv_co_readv(bs, start * BACKUP_SECTORS_PER_CLUSTER, n,
+                            &bounce_qiov);
        if (ret < 0) {
            trace_backup_do_cow_read_fail(job, start, ret);
            if (error_is_read) {
@@ -196,7 +187,7 @@ static int coroutine_fn backup_before_write_notify(
    assert((req->offset & (BDRV_SECTOR_SIZE - 1)) == 0);
    assert((req->bytes & (BDRV_SECTOR_SIZE - 1)) == 0);

-    return backup_do_cow(req->bs, sector_num, nb_sectors, NULL, true);
+    return backup_do_cow(req->bs, sector_num, nb_sectors, NULL);
 }

 static void backup_set_speed(BlockJob *job, int64_t speed, Error **errp)
@@ -204,7 +195,7 @@ static void backup_set_speed(BlockJob *job, int64_t speed, Error **errp)
    BackupBlockJob *s = container_of(job, BackupBlockJob, common);

    if (speed < 0) {
-        error_setg(errp, QERR_INVALID_PARAMETER, "speed");
+        error_set(errp, QERR_INVALID_PARAMETER, "speed");
        return;
    }
    ratelimit_set_speed(&s->limit, speed / BDRV_SECTOR_SIZE, SLICE_TIME);
@@ -214,41 +205,7 @@ static void backup_iostatus_reset(BlockJob *job)
 {
    BackupBlockJob *s = container_of(job, BackupBlockJob, common);

-    if (s->target->blk) {
-        blk_iostatus_reset(s->target->blk);
-    }
-}
-
-static void backup_cleanup_sync_bitmap(BackupBlockJob *job, int ret)
-{
-    BdrvDirtyBitmap *bm;
-    BlockDriverState *bs = job->common.bs;
-
-    if (ret < 0 || block_job_is_cancelled(&job->common)) {
-        /* Merge the successor back into the parent, delete nothing. */
-        bm = bdrv_reclaim_dirty_bitmap(bs, job->sync_bitmap, NULL);
-        assert(bm);
-    } else {
-        /* Everything is fine, delete this bitmap and install the backup. */
-        bm = bdrv_dirty_bitmap_abdicate(bs, job->sync_bitmap, NULL);
-        assert(bm);
-    }
-}
-
-static void backup_commit(BlockJob *job)
-{
-    BackupBlockJob *s = container_of(job, BackupBlockJob, common);
-    if (s->sync_bitmap) {
-        backup_cleanup_sync_bitmap(s, 0);
-    }
-}
-
-static void backup_abort(BlockJob *job)
-{
-    BackupBlockJob *s = container_of(job, BackupBlockJob, common);
-    if (s->sync_bitmap) {
-        backup_cleanup_sync_bitmap(s, -1);
-    }
+    bdrv_iostatus_reset(s->target);
 }

 static const BlockJobDriver backup_job_driver = {
@@ -256,8 +213,6 @@ static const BlockJobDriver backup_job_driver = {
    .job_type       = BLOCK_JOB_TYPE_BACKUP,
    .set_speed      = backup_set_speed,
    .iostatus_reset = backup_iostatus_reset,
-    .commit         = backup_commit,
-    .abort          = backup_abort,
 };

 static BlockErrorAction backup_error_action(BackupBlockJob *job,
@@ -287,92 +242,6 @@ static void backup_complete(BlockJob *job, void *opaque)
    g_free(data);
 }

-static bool coroutine_fn yield_and_check(BackupBlockJob *job)
-{
-    if (block_job_is_cancelled(&job->common)) {
-        return true;
-    }
-
-    /* we need to yield so that bdrv_drain_all() returns.
-     * (without, VM does not reboot)
-     */
-    if (job->common.speed) {
-        uint64_t delay_ns = ratelimit_calculate_delay(&job->limit,
-                                                      job->sectors_read);
-        job->sectors_read = 0;
-        block_job_sleep_ns(&job->common, QEMU_CLOCK_REALTIME, delay_ns);
-    } else {
-        block_job_sleep_ns(&job->common, QEMU_CLOCK_REALTIME, 0);
-    }
-
-    if (block_job_is_cancelled(&job->common)) {
-        return true;
-    }
-
-    return false;
-}
-
-static int coroutine_fn backup_run_incremental(BackupBlockJob *job)
-{
-    bool error_is_read;
-    int ret = 0;
-    int clusters_per_iter;
-    uint32_t granularity;
-    int64_t sector;
-    int64_t cluster;
-    int64_t end;
-    int64_t last_cluster = -1;
-    BlockDriverState *bs = job->common.bs;
-    HBitmapIter hbi;
-
-    granularity = bdrv_dirty_bitmap_granularity(job->sync_bitmap);
-    clusters_per_iter = MAX((granularity / BACKUP_CLUSTER_SIZE), 1);
-    bdrv_dirty_iter_init(job->sync_bitmap, &hbi);
-
-    /* Find the next dirty sector(s) */
-    while ((sector = hbitmap_iter_next(&hbi)) != -1) {
-        cluster = sector / BACKUP_SECTORS_PER_CLUSTER;
-
-        /* Fake progress updates for any clusters we skipped */
-        if (cluster != last_cluster + 1) {
-            job->common.offset += ((cluster - last_cluster - 1) *
-                                   BACKUP_CLUSTER_SIZE);
-        }
-
-        for (end = cluster + clusters_per_iter; cluster < end; cluster++) {
-            do {
-                if (yield_and_check(job)) {
-                    return ret;
-                }
-                ret = backup_do_cow(bs, cluster * BACKUP_SECTORS_PER_CLUSTER,
-                                    BACKUP_SECTORS_PER_CLUSTER, &error_is_read,
-                                    false);
-                if ((ret < 0) &&
-                    backup_error_action(job, error_is_read, -ret) ==
-                    BLOCK_ERROR_ACTION_REPORT) {
-                    return ret;
-                }
-            } while (ret < 0);
-        }
-
-        /* If the bitmap granularity is smaller than the backup granularity,
-         * we need to advance the iterator pointer to the next cluster. */
-        if (granularity < BACKUP_CLUSTER_SIZE) {
-            bdrv_set_dirty_iter(&hbi, cluster * BACKUP_SECTORS_PER_CLUSTER);
-        }
-
-        last_cluster = cluster - 1;
-    }
-
-    /* Play some final catchup with the progress meter */
-    end = DIV_ROUND_UP(job->common.len, BACKUP_CLUSTER_SIZE);
-    if (last_cluster + 1 < end) {
-        job->common.offset += ((end - last_cluster - 1) * BACKUP_CLUSTER_SIZE);
-    }
-
-    return ret;
-}
-
 static void coroutine_fn backup_run(void *opaque)
 {
    BackupBlockJob *job = opaque;
@@ -390,15 +259,14 @@ static void coroutine_fn backup_run(void *opaque)
    qemu_co_rwlock_init(&job->flush_rwlock);

    start = 0;
-    end = DIV_ROUND_UP(job->common.len, BACKUP_CLUSTER_SIZE);
+    end = DIV_ROUND_UP(job->common.len / BDRV_SECTOR_SIZE,
+                       BACKUP_SECTORS_PER_CLUSTER);

    job->bitmap = hbitmap_alloc(end, 0);

    bdrv_set_enable_write_cache(target, true);
-    if (target->blk) {
-        blk_set_on_error(target->blk, on_target_error, on_target_error);
-        blk_iostatus_enable(target->blk);
-    }
+    bdrv_set_on_error(target, on_target_error, on_target_error);
+    bdrv_iostatus_enable(target);

    bdrv_add_before_write_notifier(bs, &before_write);

@@ -410,13 +278,28 @@ static void coroutine_fn backup_run(void *opaque)
            qemu_coroutine_yield();
            job->common.busy = true;
        }
-    } else if (job->sync_mode == MIRROR_SYNC_MODE_INCREMENTAL) {
-        ret = backup_run_incremental(job);
    } else {
        /* Both FULL and TOP SYNC_MODE's require copying.. */
        for (; start < end; start++) {
            bool error_is_read;
-            if (yield_and_check(job)) {
+
+            if (block_job_is_cancelled(&job->common)) {
+                break;
+            }
+
+            /* we need to yield so that qemu_aio_flush() returns.
+             * (without, VM does not reboot)
+             */
+            if (job->common.speed) {
+                uint64_t delay_ns = ratelimit_calculate_delay(
+                        &job->limit, job->sectors_read);
+                job->sectors_read = 0;
+                block_job_sleep_ns(&job->common, QEMU_CLOCK_REALTIME, delay_ns);
+            } else {
+                block_job_sleep_ns(&job->common, QEMU_CLOCK_REALTIME, 0);
+            }
+
+            if (block_job_is_cancelled(&job->common)) {
                break;
            }

@@ -453,7 +336,7 @@ static void coroutine_fn backup_run(void *opaque)
            }
            /* FULL sync mode we copy the whole drive. */
            ret = backup_do_cow(bs, start * BACKUP_SECTORS_PER_CLUSTER,
-                    BACKUP_SECTORS_PER_CLUSTER, &error_is_read, false);
+                    BACKUP_SECTORS_PER_CLUSTER, &error_is_read);
            if (ret < 0) {
                /* Depending on error action, fail now or retry cluster */
                BlockErrorAction action =
@@ -473,11 +356,10 @@ static void coroutine_fn backup_run(void *opaque)
    /* wait until pending backup_do_cow() calls have completed */
    qemu_co_rwlock_wrlock(&job->flush_rwlock);
    qemu_co_rwlock_unlock(&job->flush_rwlock);
+
    hbitmap_free(job->bitmap);

-    if (target->blk) {
-        blk_iostatus_disable(target->blk);
-    }
+    bdrv_iostatus_disable(target);
    bdrv_op_unblock_all(target, job->common.blocker);

    data = g_malloc(sizeof(*data));
@@ -487,11 +369,10 @@ static void coroutine_fn backup_run(void *opaque)

 void backup_start(BlockDriverState *bs, BlockDriverState *target,
                  int64_t speed, MirrorSyncMode sync_mode,
-                  BdrvDirtyBitmap *sync_bitmap,
                  BlockdevOnError on_source_error,
                  BlockdevOnError on_target_error,
                  BlockCompletionFunc *cb, void *opaque,
-                  BlockJobTxn *txn, Error **errp)
+                  Error **errp)
 {
    int64_t len;

@@ -506,8 +387,8 @@ void backup_start(BlockDriverState *bs, BlockDriverState *target,

    if ((on_source_error == BLOCKDEV_ON_ERROR_STOP ||
         on_source_error == BLOCKDEV_ON_ERROR_ENOSPC) &&
-        (!bs->blk || !blk_iostatus_is_enabled(bs->blk))) {
-        error_setg(errp, QERR_INVALID_PARAMETER, "on-source-error");
+        !bdrv_iostatus_is_enabled(bs)) {
+        error_set(errp, QERR_INVALID_PARAMETER, "on-source-error");
        return;
    }

@@ -531,36 +412,17 @@ void backup_start(BlockDriverState *bs, BlockDriverState *target,
        return;
    }

-    if (sync_mode == MIRROR_SYNC_MODE_INCREMENTAL) {
-        if (!sync_bitmap) {
-            error_setg(errp, "must provide a valid bitmap name for "
-                             "\"incremental\" sync mode");
-            return;
-        }
-
-        /* Create a new bitmap, and freeze/disable this one. */
-        if (bdrv_dirty_bitmap_create_successor(bs, sync_bitmap, errp) < 0) {
-            return;
-        }
-    } else if (sync_bitmap) {
-        error_setg(errp,
-                   "a sync_bitmap was provided to backup_run, "
-                   "but received an incompatible sync_mode (%s)",
-                   MirrorSyncMode_lookup[sync_mode]);
-        return;
-    }
-
    len = bdrv_getlength(bs);
    if (len < 0) {
        error_setg_errno(errp, -len, "unable to get length for '%s'",
                         bdrv_get_device_name(bs));
-        goto error;
+        return;
    }

    BackupBlockJob *job = block_job_create(&backup_job_driver, bs, speed,
                                           cb, opaque, errp);
    if (!job) {
-        goto error;
+        return;
    }

    bdrv_op_block_all(target, job->common.blocker);
@@ -569,16 +431,7 @@ void backup_start(BlockDriverState *bs, BlockDriverState *target,
    job->on_target_error = on_target_error;
    job->target = target;
    job->sync_mode = sync_mode;
-    job->sync_bitmap = sync_mode == MIRROR_SYNC_MODE_INCREMENTAL ?
-                       sync_bitmap : NULL;
    job->common.len = len;
    job->common.co = qemu_coroutine_create(backup_run);
-    block_job_txn_add_job(txn, &job->common);
    qemu_coroutine_enter(job->common.co, job);
-    return;
-
- error:
-    if (sync_bitmap) {
-        bdrv_reclaim_dirty_bitmap(bs, sync_bitmap, NULL);
-    }
 }
--- a/block/blkdebug.c
+++ b/block/blkdebug.c
@@ -22,7 +22,6 @@
 * THE SOFTWARE.
 */

-#include "qemu/osdep.h"
 #include "qemu-common.h"
 #include "qemu/config-file.h"
 #include "block/block_int.h"
@@ -31,13 +30,12 @@
 #include "qapi/qmp/qdict.h"
 #include "qapi/qmp/qint.h"
 #include "qapi/qmp/qstring.h"
-#include "sysemu/qtest.h"

 typedef struct BDRVBlkdebugState {
    int state;
    int new_state;

-    QLIST_HEAD(, BlkdebugRule) rules[BLKDBG__MAX];
+    QLIST_HEAD(, BlkdebugRule) rules[BLKDBG_EVENT_MAX];
    QSIMPLEQ_HEAD(, BlkdebugRule) active_rules;
    QLIST_HEAD(, BlkdebugSuspendedReq) suspended_reqs;
 } BDRVBlkdebugState;
@@ -65,7 +63,7 @@ enum {
 };

 typedef struct BlkdebugRule {
-    BlkdebugEvent event;
+    BlkDebugEvent event;
    int action;
    int state;
    union {
@@ -144,12 +142,69 @@ static QemuOptsList *config_groups[] = {
    NULL
 };

-static int get_event_by_name(const char *name, BlkdebugEvent *event)
+static const char *event_names[BLKDBG_EVENT_MAX] = {
+    [BLKDBG_L1_UPDATE]                      = "l1_update",
+    [BLKDBG_L1_GROW_ALLOC_TABLE]            = "l1_grow.alloc_table",
+    [BLKDBG_L1_GROW_WRITE_TABLE]            = "l1_grow.write_table",
+    [BLKDBG_L1_GROW_ACTIVATE_TABLE]         = "l1_grow.activate_table",
+
+    [BLKDBG_L2_LOAD]                        = "l2_load",
+    [BLKDBG_L2_UPDATE]                      = "l2_update",
+    [BLKDBG_L2_UPDATE_COMPRESSED]           = "l2_update_compressed",
+    [BLKDBG_L2_ALLOC_COW_READ]              = "l2_alloc.cow_read",
+    [BLKDBG_L2_ALLOC_WRITE]                 = "l2_alloc.write",
+
+    [BLKDBG_READ_AIO]                       = "read_aio",
+    [BLKDBG_READ_BACKING_AIO]               = "read_backing_aio",
+    [BLKDBG_READ_COMPRESSED]                = "read_compressed",
+
+    [BLKDBG_WRITE_AIO]                      = "write_aio",
+    [BLKDBG_WRITE_COMPRESSED]               = "write_compressed",
+
+    [BLKDBG_VMSTATE_LOAD]                   = "vmstate_load",
+    [BLKDBG_VMSTATE_SAVE]                   = "vmstate_save",
+
+    [BLKDBG_COW_READ]                       = "cow_read",
+    [BLKDBG_COW_WRITE]                      = "cow_write",
+
+    [BLKDBG_REFTABLE_LOAD]                  = "reftable_load",
+    [BLKDBG_REFTABLE_GROW]                  = "reftable_grow",
+    [BLKDBG_REFTABLE_UPDATE]                = "reftable_update",
+
+    [BLKDBG_REFBLOCK_LOAD]                  = "refblock_load",
+    [BLKDBG_REFBLOCK_UPDATE]                = "refblock_update",
+    [BLKDBG_REFBLOCK_UPDATE_PART]           = "refblock_update_part",
+    [BLKDBG_REFBLOCK_ALLOC]                 = "refblock_alloc",
+    [BLKDBG_REFBLOCK_ALLOC_HOOKUP]          = "refblock_alloc.hookup",
+    [BLKDBG_REFBLOCK_ALLOC_WRITE]           = "refblock_alloc.write",
+    [BLKDBG_REFBLOCK_ALLOC_WRITE_BLOCKS]    = "refblock_alloc.write_blocks",
+    [BLKDBG_REFBLOCK_ALLOC_WRITE_TABLE]     = "refblock_alloc.write_table",
+    [BLKDBG_REFBLOCK_ALLOC_SWITCH_TABLE]    = "refblock_alloc.switch_table",
+
+    [BLKDBG_CLUSTER_ALLOC]                  = "cluster_alloc",
+    [BLKDBG_CLUSTER_ALLOC_BYTES]            = "cluster_alloc_bytes",
+    [BLKDBG_CLUSTER_FREE]                   = "cluster_free",
+
+    [BLKDBG_FLUSH_TO_OS]                    = "flush_to_os",
+    [BLKDBG_FLUSH_TO_DISK]                  = "flush_to_disk",
+
+    [BLKDBG_PWRITEV_RMW_HEAD]               = "pwritev_rmw.head",
+    [BLKDBG_PWRITEV_RMW_AFTER_HEAD]         = "pwritev_rmw.after_head",
+    [BLKDBG_PWRITEV_RMW_TAIL]               = "pwritev_rmw.tail",
+    [BLKDBG_PWRITEV_RMW_AFTER_TAIL]         = "pwritev_rmw.after_tail",
+    [BLKDBG_PWRITEV]                        = "pwritev",
+    [BLKDBG_PWRITEV_ZERO]                   = "pwritev_zero",
+    [BLKDBG_PWRITEV_DONE]                   = "pwritev_done",
+
+    [BLKDBG_EMPTY_IMAGE_PREPARE]            = "empty_image_prepare",
+};
+
+static int get_event_by_name(const char *name, BlkDebugEvent *event)
 {
    int i;

-    for (i = 0; i < BLKDBG__MAX; i++) {
-        if (!strcmp(BlkdebugEvent_lookup[i], name)) {
+    for (i = 0; i < BLKDBG_EVENT_MAX; i++) {
+        if (!strcmp(event_names[i], name)) {
            *event = i;
            return 0;
        }
@@ -161,23 +216,24 @@ static int get_event_by_name(const char *name, BlkdebugEvent *event)
 struct add_rule_data {
    BDRVBlkdebugState *s;
    int action;
+    Error **errp;
 };

-static int add_rule(void *opaque, QemuOpts *opts, Error **errp)
+static int add_rule(QemuOpts *opts, void *opaque)
 {
    struct add_rule_data *d = opaque;
    BDRVBlkdebugState *s = d->s;
    const char* event_name;
-    BlkdebugEvent event;
+    BlkDebugEvent event;
    struct BlkdebugRule *rule;

    /* Find the right event for the rule */
    event_name = qemu_opt_get(opts, "event");
    if (!event_name) {
-        error_setg(errp, "Missing event name for rule");
+        error_setg(d->errp, "Missing event name for rule");
        return -1;
    } else if (get_event_by_name(event_name, &event) < 0) {
-        error_setg(errp, "Invalid event name \"%s\"", event_name);
+        error_setg(d->errp, "Invalid event name \"%s\"", event_name);
        return -1;
    }

@@ -263,7 +319,8 @@ static int read_config(BDRVBlkdebugState *s, const char *filename,

    d.s = s;
    d.action = ACTION_INJECT_ERROR;
-    qemu_opts_foreach(&inject_error_opts, add_rule, &d, &local_err);
+    d.errp = &local_err;
+    qemu_opts_foreach(&inject_error_opts, add_rule, &d, 1);
    if (local_err) {
        error_propagate(errp, local_err);
        ret = -EINVAL;
@@ -271,7 +328,7 @@ static int read_config(BDRVBlkdebugState *s, const char *filename,
    }

    d.action = ACTION_SET_STATE;
-    qemu_opts_foreach(&set_state_opts, add_rule, &d, &local_err);
+    qemu_opts_foreach(&set_state_opts, add_rule, &d, 1);
    if (local_err) {
        error_propagate(errp, local_err);
        ret = -EINVAL;
@@ -371,11 +428,11 @@ static int blkdebug_open(BlockDriverState *bs, QDict *options, int flags,
    /* Set initial state */
    s->state = 1;

-    /* Open the image file */
-    bs->file = bdrv_open_child(qemu_opt_get(opts, "x-image"), options, "image",
-                               bs, &child_file, false, &local_err);
-    if (local_err) {
-        ret = -EINVAL;
+    /* Open the backing file */
+    assert(bs->file == NULL);
+    ret = bdrv_open_image(&bs->file, qemu_opt_get(opts, "x-image"), options, "image",
+                          flags | BDRV_O_PROTOCOL, false, &local_err);
+    if (ret < 0) {
        error_propagate(errp, local_err);
        goto out;
    }
@@ -394,7 +451,7 @@ static int blkdebug_open(BlockDriverState *bs, QDict *options, int flags,
    goto out;

 fail_unref:
-    bdrv_unref_child(bs, bs->file);
+    bdrv_unref(bs->file);
 out:
    qemu_opts_del(opts);
    return ret;
@@ -455,8 +512,7 @@ static BlockAIOCB *blkdebug_aio_readv(BlockDriverState *bs,
        return inject_error(bs, cb, opaque, rule);
    }

-    return bdrv_aio_readv(bs->file->bs, sector_num, qiov, nb_sectors,
-                          cb, opaque);
+    return bdrv_aio_readv(bs->file, sector_num, qiov, nb_sectors, cb, opaque);
 }

 static BlockAIOCB *blkdebug_aio_writev(BlockDriverState *bs,
@@ -478,8 +534,7 @@ static BlockAIOCB *blkdebug_aio_writev(BlockDriverState *bs,
        return inject_error(bs, cb, opaque, rule);
    }

-    return bdrv_aio_writev(bs->file->bs, sector_num, qiov, nb_sectors,
-                           cb, opaque);
+    return bdrv_aio_writev(bs->file, sector_num, qiov, nb_sectors, cb, opaque);
 }

 static BlockAIOCB *blkdebug_aio_flush(BlockDriverState *bs,
@@ -498,7 +553,7 @@ static BlockAIOCB *blkdebug_aio_flush(BlockDriverState *bs,
        return inject_error(bs, cb, opaque, rule);
    }

-    return bdrv_aio_flush(bs->file->bs, cb, opaque);
+    return bdrv_aio_flush(bs->file, cb, opaque);
 }


@@ -508,7 +563,7 @@ static void blkdebug_close(BlockDriverState *bs)
    BlkdebugRule *rule, *next;
    int i;

-    for (i = 0; i < BLKDBG__MAX; i++) {
+    for (i = 0; i < BLKDBG_EVENT_MAX; i++) {
        QLIST_FOREACH_SAFE(rule, &s->rules[i], next, next) {
            remove_rule(rule);
        }
@@ -528,13 +583,9 @@ static void suspend_request(BlockDriverState *bs, BlkdebugRule *rule)
    remove_rule(rule);
    QLIST_INSERT_HEAD(&s->suspended_reqs, &r, next);

-    if (!qtest_enabled()) {
-        printf("blkdebug: Suspended request '%s'\n", r.tag);
-    }
+    printf("blkdebug: Suspended request '%s'\n", r.tag);
    qemu_coroutine_yield();
-    if (!qtest_enabled()) {
-        printf("blkdebug: Resuming request '%s'\n", r.tag);
-    }
+    printf("blkdebug: Resuming request '%s'\n", r.tag);

    QLIST_REMOVE(&r, next);
    g_free(r.tag);
@@ -571,13 +622,13 @@ static bool process_rule(BlockDriverState *bs, struct BlkdebugRule *rule,
    return injected;
 }

-static void blkdebug_debug_event(BlockDriverState *bs, BlkdebugEvent event)
+static void blkdebug_debug_event(BlockDriverState *bs, BlkDebugEvent event)
 {
    BDRVBlkdebugState *s = bs->opaque;
    struct BlkdebugRule *rule, *next;
    bool injected;

-    assert((int)event >= 0 && event < BLKDBG__MAX);
+    assert((int)event >= 0 && event < BLKDBG_EVENT_MAX);

    injected = false;
    s->new_state = s->state;
@@ -592,7 +643,7 @@ static int blkdebug_debug_breakpoint(BlockDriverState *bs, const char *event,
 {
    BDRVBlkdebugState *s = bs->opaque;
    struct BlkdebugRule *rule;
-    BlkdebugEvent blkdebug_event;
+    BlkDebugEvent blkdebug_event;

    if (get_event_by_name(event, &blkdebug_event) < 0) {
        return -ENOENT;
@@ -634,7 +685,7 @@ static int blkdebug_debug_remove_breakpoint(BlockDriverState *bs,
    BlkdebugRule *rule, *next;
    int i, ret = -ENOENT;

-    for (i = 0; i < BLKDBG__MAX; i++) {
+    for (i = 0; i < BLKDBG_EVENT_MAX; i++) {
        QLIST_FOREACH_SAFE(rule, &s->rules[i], next, next) {
            if (rule->action == ACTION_SUSPEND &&
                !strcmp(rule->options.suspend.tag, tag)) {
@@ -667,50 +718,50 @@ static bool blkdebug_debug_is_suspended(BlockDriverState *bs, const char *tag)

 static int64_t blkdebug_getlength(BlockDriverState *bs)
 {
-    return bdrv_getlength(bs->file->bs);
+    return bdrv_getlength(bs->file);
 }

-static int blkdebug_truncate(BlockDriverState *bs, int64_t offset)
-{
-    return bdrv_truncate(bs->file->bs, offset);
-}
-
-static void blkdebug_refresh_filename(BlockDriverState *bs, QDict *options)
+static void blkdebug_refresh_filename(BlockDriverState *bs)
 {
    QDict *opts;
    const QDictEntry *e;
    bool force_json = false;

-    for (e = qdict_first(options); e; e = qdict_next(options, e)) {
+    for (e = qdict_first(bs->options); e; e = qdict_next(bs->options, e)) {
        if (strcmp(qdict_entry_key(e), "config") &&
-            strcmp(qdict_entry_key(e), "x-image"))
+            strcmp(qdict_entry_key(e), "x-image") &&
+            strcmp(qdict_entry_key(e), "image") &&
+            strncmp(qdict_entry_key(e), "image.", strlen("image.")))
        {
            force_json = true;
            break;
        }
    }

-    if (force_json && !bs->file->bs->full_open_options) {
+    if (force_json && !bs->file->full_open_options) {
        /* The config file cannot be recreated, so creating a plain filename
         * is impossible */
        return;
    }

-    if (!force_json && bs->file->bs->exact_filename[0]) {
+    if (!force_json && bs->file->exact_filename[0]) {
        snprintf(bs->exact_filename, sizeof(bs->exact_filename),
                 "blkdebug:%s:%s",
-                 qdict_get_try_str(options, "config") ?: "",
-                 bs->file->bs->exact_filename);
+                 qdict_get_try_str(bs->options, "config") ?: "",
+                 bs->file->exact_filename);
    }

    opts = qdict_new();
    qdict_put_obj(opts, "driver", QOBJECT(qstring_from_str("blkdebug")));

-    QINCREF(bs->file->bs->full_open_options);
-    qdict_put_obj(opts, "image", QOBJECT(bs->file->bs->full_open_options));
+    QINCREF(bs->file->full_open_options);
+    qdict_put_obj(opts, "image", QOBJECT(bs->file->full_open_options));

-    for (e = qdict_first(options); e; e = qdict_next(options, e)) {
-        if (strcmp(qdict_entry_key(e), "x-image")) {
+    for (e = qdict_first(bs->options); e; e = qdict_next(bs->options, e)) {
+        if (strcmp(qdict_entry_key(e), "x-image") &&
+            strcmp(qdict_entry_key(e), "image") &&
+            strncmp(qdict_entry_key(e), "image.", strlen("image.")))
+        {
            qobject_incref(qdict_entry_value(e));
            qdict_put_obj(opts, qdict_entry_key(e), qdict_entry_value(e));
        }
@@ -719,12 +770,6 @@ static void blkdebug_refresh_filename(BlockDriverState *bs, QDict *options)
    bs->full_open_options = opts;
 }

-static int blkdebug_reopen_prepare(BDRVReopenState *reopen_state,
-                                   BlockReopenQueue *queue, Error **errp)
-{
-    return 0;
-}
-
 static BlockDriver bdrv_blkdebug = {
    .format_name            = "blkdebug",
    .protocol_name          = "blkdebug",
@@ -733,9 +778,7 @@ static BlockDriver bdrv_blkdebug = {
    .bdrv_parse_filename    = blkdebug_parse_filename,
    .bdrv_file_open         = blkdebug_open,
    .bdrv_close             = blkdebug_close,
-    .bdrv_reopen_prepare    = blkdebug_reopen_prepare,
    .bdrv_getlength         = blkdebug_getlength,
-    .bdrv_truncate          = blkdebug_truncate,
    .bdrv_refresh_filename  = blkdebug_refresh_filename,

    .bdrv_aio_readv         = blkdebug_aio_readv,
--- a/block/blkverify.c
+++ b/block/blkverify.c
@@ -7,14 +7,14 @@
 * See the COPYING file in the top-level directory.
 */

-#include "qemu/osdep.h"
+#include <stdarg.h>
 #include "qemu/sockets.h" /* for EINPROGRESS on Windows */
 #include "block/block_int.h"
 #include "qapi/qmp/qdict.h"
 #include "qapi/qmp/qstring.h"

 typedef struct {
-    BdrvChild *test_file;
+    BlockDriverState *test_file;
 } BDRVBlkverifyState;

 typedef struct BlkverifyAIOCB BlkverifyAIOCB;
@@ -123,29 +123,26 @@ static int blkverify_open(BlockDriverState *bs, QDict *options, int flags,
    }

    /* Open the raw file */
-    bs->file = bdrv_open_child(qemu_opt_get(opts, "x-raw"), options, "raw",
-                               bs, &child_file, false, &local_err);
-    if (local_err) {
-        ret = -EINVAL;
+    assert(bs->file == NULL);
+    ret = bdrv_open_image(&bs->file, qemu_opt_get(opts, "x-raw"), options,
+                          "raw", flags | BDRV_O_PROTOCOL, false, &local_err);
+    if (ret < 0) {
        error_propagate(errp, local_err);
        goto fail;
    }

    /* Open the test file */
-    s->test_file = bdrv_open_child(qemu_opt_get(opts, "x-image"), options,
-                                   "test", bs, &child_format, false,
-                                   &local_err);
-    if (local_err) {
-        ret = -EINVAL;
+    assert(s->test_file == NULL);
+    ret = bdrv_open_image(&s->test_file, qemu_opt_get(opts, "x-image"), options,
+                          "test", flags, false, &local_err);
+    if (ret < 0) {
        error_propagate(errp, local_err);
+        s->test_file = NULL;
        goto fail;
    }

    ret = 0;
 fail:
-    if (ret < 0) {
-        bdrv_unref_child(bs, bs->file);
-    }
    qemu_opts_del(opts);
    return ret;
 }
@@ -154,7 +151,7 @@ static void blkverify_close(BlockDriverState *bs)
 {
    BDRVBlkverifyState *s = bs->opaque;

-    bdrv_unref_child(bs, s->test_file);
+    bdrv_unref(s->test_file);
    s->test_file = NULL;
 }

@@ -162,7 +159,7 @@ static int64_t blkverify_getlength(BlockDriverState *bs)
 {
    BDRVBlkverifyState *s = bs->opaque;

-    return bdrv_getlength(s->test_file->bs);
+    return bdrv_getlength(s->test_file);
 }

 static BlkverifyAIOCB *blkverify_aio_get(BlockDriverState *bs, bool is_write,
@@ -241,13 +238,13 @@ static BlockAIOCB *blkverify_aio_readv(BlockDriverState *bs,
                                            nb_sectors, cb, opaque);

    acb->verify = blkverify_verify_readv;
-    acb->buf = qemu_blockalign(bs->file->bs, qiov->size);
+    acb->buf = qemu_blockalign(bs->file, qiov->size);
    qemu_iovec_init(&acb->raw_qiov, acb->qiov->niov);
    qemu_iovec_clone(&acb->raw_qiov, qiov, acb->buf);

-    bdrv_aio_readv(s->test_file->bs, sector_num, qiov, nb_sectors,
+    bdrv_aio_readv(s->test_file, sector_num, qiov, nb_sectors,
                   blkverify_aio_cb, acb);
-    bdrv_aio_readv(bs->file->bs, sector_num, &acb->raw_qiov, nb_sectors,
+    bdrv_aio_readv(bs->file, sector_num, &acb->raw_qiov, nb_sectors,
                   blkverify_aio_cb, acb);
    return &acb->common;
 }
@@ -260,9 +257,9 @@ static BlockAIOCB *blkverify_aio_writev(BlockDriverState *bs,
    BlkverifyAIOCB *acb = blkverify_aio_get(bs, true, sector_num, qiov,
                                            nb_sectors, cb, opaque);

-    bdrv_aio_writev(s->test_file->bs, sector_num, qiov, nb_sectors,
+    bdrv_aio_writev(s->test_file, sector_num, qiov, nb_sectors,
                    blkverify_aio_cb, acb);
-    bdrv_aio_writev(bs->file->bs, sector_num, qiov, nb_sectors,
+    bdrv_aio_writev(bs->file, sector_num, qiov, nb_sectors,
                    blkverify_aio_cb, acb);
    return &acb->common;
 }
@@ -274,7 +271,7 @@ static BlockAIOCB *blkverify_aio_flush(BlockDriverState *bs,
    BDRVBlkverifyState *s = bs->opaque;

    /* Only flush test file, the raw file is not important */
-    return bdrv_aio_flush(s->test_file->bs, cb, opaque);
+    return bdrv_aio_flush(s->test_file, cb, opaque);
 }

 static bool blkverify_recurse_is_first_non_filter(BlockDriverState *bs,
@@ -282,13 +279,13 @@ static bool blkverify_recurse_is_first_non_filter(BlockDriverState *bs,
 {
    BDRVBlkverifyState *s = bs->opaque;

-    bool perm = bdrv_recurse_is_first_non_filter(bs->file->bs, candidate);
+    bool perm = bdrv_recurse_is_first_non_filter(bs->file, candidate);

    if (perm) {
        return true;
    }

-    return bdrv_recurse_is_first_non_filter(s->test_file->bs, candidate);
+    return bdrv_recurse_is_first_non_filter(s->test_file, candidate);
 }

 /* Propagate AioContext changes to ->test_file */
@@ -296,7 +293,7 @@ static void blkverify_detach_aio_context(BlockDriverState *bs)
 {
    BDRVBlkverifyState *s = bs->opaque;

-    bdrv_detach_aio_context(s->test_file->bs);
+    bdrv_detach_aio_context(s->test_file);
 }

 static void blkverify_attach_aio_context(BlockDriverState *bs,
@@ -304,38 +301,32 @@ static void blkverify_attach_aio_context(BlockDriverState *bs,
 {
    BDRVBlkverifyState *s = bs->opaque;

-    bdrv_attach_aio_context(s->test_file->bs, new_context);
+    bdrv_attach_aio_context(s->test_file, new_context);
 }

-static void blkverify_refresh_filename(BlockDriverState *bs, QDict *options)
+static void blkverify_refresh_filename(BlockDriverState *bs)
 {
    BDRVBlkverifyState *s = bs->opaque;

-    /* bs->file->bs has already been refreshed */
-    bdrv_refresh_filename(s->test_file->bs);
+    /* bs->file has already been refreshed */
+    bdrv_refresh_filename(s->test_file);

-    if (bs->file->bs->full_open_options
-        && s->test_file->bs->full_open_options)
-    {
+    if (bs->file->full_open_options && s->test_file->full_open_options) {
        QDict *opts = qdict_new();
        qdict_put_obj(opts, "driver", QOBJECT(qstring_from_str("blkverify")));

-        QINCREF(bs->file->bs->full_open_options);
-        qdict_put_obj(opts, "raw", QOBJECT(bs->file->bs->full_open_options));
-        QINCREF(s->test_file->bs->full_open_options);
-        qdict_put_obj(opts, "test",
-                      QOBJECT(s->test_file->bs->full_open_options));
+        QINCREF(bs->file->full_open_options);
+        qdict_put_obj(opts, "raw", QOBJECT(bs->file->full_open_options));
+        QINCREF(s->test_file->full_open_options);
+        qdict_put_obj(opts, "test", QOBJECT(s->test_file->full_open_options));

        bs->full_open_options = opts;
    }

-    if (bs->file->bs->exact_filename[0]
-        && s->test_file->bs->exact_filename[0])
-    {
+    if (bs->file->exact_filename[0] && s->test_file->exact_filename[0]) {
        snprintf(bs->exact_filename, sizeof(bs->exact_filename),
                 "blkverify:%s:%s",
-                 bs->file->bs->exact_filename,
-                 s->test_file->bs->exact_filename);
+                 bs->file->exact_filename, s->test_file->exact_filename);
    }
 }

--- a/block/block-backend.c
+++ b/block/block-backend.c
@@ -10,20 +10,14 @@
 * or later.  See the COPYING.LIB file in the top-level directory.
 */

-#include "qemu/osdep.h"
 #include "sysemu/block-backend.h"
 #include "block/block_int.h"
-#include "block/blockjob.h"
-#include "block/throttle-groups.h"
 #include "sysemu/blockdev.h"
-#include "sysemu/sysemu.h"
 #include "qapi-event.h"

 /* Number of coroutines to reserve per attached device model */
 #define COROUTINE_POOL_RESERVATION 64

-static AioContext *blk_aiocb_get_aio_context(BlockAIOCB *acb);
-
 struct BlockBackend {
    char *name;
    int refcnt;
@@ -35,33 +29,15 @@ struct BlockBackend {
    /* TODO change to DeviceState when all users are qdevified */
    const BlockDevOps *dev_ops;
    void *dev_opaque;
-
-    /* the block size for which the guest device expects atomicity */
-    int guest_block_size;
-
-    /* If the BDS tree is removed, some of its options are stored here (which
-     * can be used to restore those options in the new BDS on insert) */
-    BlockBackendRootState root_state;
-
-    /* I/O stats (display with "info blockstats"). */
-    BlockAcctStats stats;
-
-    BlockdevOnError on_read_error, on_write_error;
-    bool iostatus_enabled;
-    BlockDeviceIoStatus iostatus;
-
-    NotifierList remove_bs_notifiers, insert_bs_notifiers;
 };

 typedef struct BlockBackendAIOCB {
    BlockAIOCB common;
    QEMUBH *bh;
-    BlockBackend *blk;
    int ret;
 } BlockBackendAIOCB;

 static const AIOCBInfo block_backend_aiocb_info = {
-    .get_aio_context = blk_aiocb_get_aio_context,
    .aiocb_size = sizeof(BlockBackendAIOCB),
 };

@@ -101,8 +77,6 @@ BlockBackend *blk_new(const char *name, Error **errp)
    blk = g_new0(BlockBackend, 1);
    blk->name = g_strdup(name);
    blk->refcnt = 1;
-    notifier_list_init(&blk->remove_bs_notifiers);
-    notifier_list_init(&blk->insert_bs_notifiers);
    QTAILQ_INSERT_TAIL(&blk_backends, blk, link);
    return blk;
 }
@@ -152,7 +126,7 @@ BlockBackend *blk_new_open(const char *name, const char *filename,
        return NULL;
    }

-    ret = bdrv_open(&blk->bs, filename, reference, options, flags, errp);
+    ret = bdrv_open(&blk->bs, filename, reference, options, flags, NULL, errp);
    if (ret < 0) {
        blk_unref(blk);
        return NULL;
@@ -166,13 +140,10 @@ static void blk_delete(BlockBackend *blk)
    assert(!blk->refcnt);
    assert(!blk->dev);
    if (blk->bs) {
-        blk_remove_bs(blk);
-    }
-    assert(QLIST_EMPTY(&blk->remove_bs_notifiers.notifiers));
-    assert(QLIST_EMPTY(&blk->insert_bs_notifiers.notifiers));
-    if (blk->root_state.throttle_state) {
-        g_free(blk->root_state.throttle_group);
-        throttle_group_unref(blk->root_state.throttle_state);
+        assert(blk->bs->blk == blk);
+        blk->bs->blk = NULL;
+        bdrv_unref(blk->bs);
+        blk->bs = NULL;
    }
    /* Avoid double-remove after blk_hide_on_behalf_of_hmp_drive_del() */
    if (blk->name[0]) {
@@ -180,7 +151,6 @@ static void blk_delete(BlockBackend *blk)
    }
    g_free(blk->name);
    drive_info_del(blk->legacy_dinfo);
-    block_acct_cleanup(&blk->stats);
    g_free(blk);
 }

@@ -194,11 +164,6 @@ static void drive_info_del(DriveInfo *dinfo)
    g_free(dinfo);
 }

-int blk_get_refcnt(BlockBackend *blk)
-{
-    return blk ? blk->refcnt : 0;
-}
-
 /*
 * Increment @blk's reference count.
 * @blk must not be null.
@@ -223,21 +188,6 @@ void blk_unref(BlockBackend *blk)
    }
 }

-void blk_remove_all_bs(void)
-{
-    BlockBackend *blk;
-
-    QTAILQ_FOREACH(blk, &blk_backends, link) {
-        AioContext *ctx = blk_get_aio_context(blk);
-
-        aio_context_acquire(ctx);
-        if (blk->bs) {
-            blk_remove_bs(blk);
-        }
-        aio_context_release(ctx);
-    }
-}
-
 /*
 * Return the BlockBackend after @blk.
 * If @blk is null, return the first one.
@@ -288,23 +238,6 @@ BlockDriverState *blk_bs(BlockBackend *blk)
    return blk->bs;
 }

-/*
- * Changes the BlockDriverState attached to @blk
- */
-void blk_set_bs(BlockBackend *blk, BlockDriverState *bs)
-{
-    bdrv_ref(bs);
-
-    if (blk->bs) {
-        blk->bs->blk = NULL;
-        bdrv_unref(blk->bs);
-    }
-    assert(bs->blk == NULL);
-
-    blk->bs = bs;
-    bs->blk = blk;
-}
-
 /*
 * Return @blk's DriveInfo if any, else null.
 */
@@ -358,35 +291,6 @@ void blk_hide_on_behalf_of_hmp_drive_del(BlockBackend *blk)
    }
 }

-/*
- * Disassociates the currently associated BlockDriverState from @blk.
- */
-void blk_remove_bs(BlockBackend *blk)
-{
-    assert(blk->bs->blk == blk);
-
-    notifier_list_notify(&blk->remove_bs_notifiers, blk);
-
-    blk_update_root_state(blk);
-
-    blk->bs->blk = NULL;
-    bdrv_unref(blk->bs);
-    blk->bs = NULL;
-}
-
-/*
- * Associates a new BlockDriverState with @blk.
- */
-void blk_insert_bs(BlockBackend *blk, BlockDriverState *bs)
-{
-    assert(!blk->bs && !bs->blk);
-    bdrv_ref(bs);
-    blk->bs = bs;
-    bs->blk = blk;
-
-    notifier_list_notify(&blk->insert_bs_notifiers, blk);
-}
-
 /*
 * Attach device model @dev to @blk.
 * Return 0 on success, -EBUSY when a device model is attached already.
@@ -399,7 +303,7 @@ int blk_attach_dev(BlockBackend *blk, void *dev)
    }
    blk_ref(blk);
    blk->dev = dev;
-    blk_iostatus_reset(blk);
+    bdrv_iostatus_reset(blk->bs);
    return 0;
 }

@@ -426,7 +330,7 @@ void blk_detach_dev(BlockBackend *blk, void *dev)
    blk->dev = NULL;
    blk->dev_ops = NULL;
    blk->dev_opaque = NULL;
-    blk->guest_block_size = 512;
+    bdrv_set_guest_block_size(blk->bs, 512);
    blk_unref(blk);
 }

@@ -460,15 +364,18 @@ void blk_set_dev_ops(BlockBackend *blk, const BlockDevOps *ops,
 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;
+        bool tray_was_closed = !blk_dev_is_tray_open(blk);

-        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,
-                                              &error_abort);
+        if (tray_was_closed) {
+            /* tray open */
+            qapi_event_send_device_tray_moved(blk_name(blk),
+                                              true, &error_abort);
+        }
+        if (load) {
+            /* tray close */
+            qapi_event_send_device_tray_moved(blk_name(blk),
+                                              false, &error_abort);
        }
    }
 }
@@ -482,14 +389,6 @@ bool blk_dev_has_removable_media(BlockBackend *blk)
    return !blk->dev || (blk->dev_ops && blk->dev_ops->change_media_cb);
 }

-/*
- * Does @blk's attached device model have a tray?
- */
-bool blk_dev_has_tray(BlockBackend *blk)
-{
-    return blk->dev_ops && blk->dev_ops->is_tray_open;
-}
-
 /*
 * Notify @blk's attached device model of a media eject request.
 * If @force is true, the medium is about to be yanked out forcefully.
@@ -506,7 +405,7 @@ void blk_dev_eject_request(BlockBackend *blk, bool force)
 */
 bool blk_dev_is_tray_open(BlockBackend *blk)
 {
-    if (blk_dev_has_tray(blk)) {
+    if (blk->dev_ops && blk->dev_ops->is_tray_open) {
        return blk->dev_ops->is_tray_open(blk->dev_opaque);
    }
    return false;
@@ -536,47 +435,7 @@ void blk_dev_resize_cb(BlockBackend *blk)

 void blk_iostatus_enable(BlockBackend *blk)
 {
-    blk->iostatus_enabled = true;
-    blk->iostatus = BLOCK_DEVICE_IO_STATUS_OK;
-}
-
-/* The I/O status is only enabled if the drive explicitly
- * enables it _and_ the VM is configured to stop on errors */
-bool blk_iostatus_is_enabled(const BlockBackend *blk)
-{
-    return (blk->iostatus_enabled &&
-           (blk->on_write_error == BLOCKDEV_ON_ERROR_ENOSPC ||
-            blk->on_write_error == BLOCKDEV_ON_ERROR_STOP   ||
-            blk->on_read_error == BLOCKDEV_ON_ERROR_STOP));
-}
-
-BlockDeviceIoStatus blk_iostatus(const BlockBackend *blk)
-{
-    return blk->iostatus;
-}
-
-void blk_iostatus_disable(BlockBackend *blk)
-{
-    blk->iostatus_enabled = false;
-}
-
-void blk_iostatus_reset(BlockBackend *blk)
-{
-    if (blk_iostatus_is_enabled(blk)) {
-        blk->iostatus = BLOCK_DEVICE_IO_STATUS_OK;
-        if (blk->bs && blk->bs->job) {
-            block_job_iostatus_reset(blk->bs->job);
-        }
-    }
-}
-
-void blk_iostatus_set_err(BlockBackend *blk, int error)
-{
-    assert(blk_iostatus_is_enabled(blk));
-    if (blk->iostatus == BLOCK_DEVICE_IO_STATUS_OK) {
-        blk->iostatus = error == ENOSPC ? BLOCK_DEVICE_IO_STATUS_NOSPACE :
-                                          BLOCK_DEVICE_IO_STATUS_FAILED;
-    }
+    bdrv_iostatus_enable(blk->bs);
 }

 static int blk_check_byte_request(BlockBackend *blk, int64_t offset,
@@ -588,7 +447,7 @@ static int blk_check_byte_request(BlockBackend *blk, int64_t offset,
        return -EIO;
    }

-    if (!blk_is_available(blk)) {
+    if (!blk_is_inserted(blk)) {
        return -ENOMEDIUM;
    }

@@ -656,17 +515,6 @@ int blk_write(BlockBackend *blk, int64_t sector_num, const uint8_t *buf,
    return bdrv_write(blk->bs, sector_num, buf, nb_sectors);
 }

-int blk_write_zeroes(BlockBackend *blk, int64_t sector_num,
-                     int nb_sectors, BdrvRequestFlags flags)
-{
-    int ret = blk_check_request(blk, sector_num, nb_sectors);
-    if (ret < 0) {
-        return ret;
-    }
-
-    return bdrv_write_zeroes(blk->bs, sector_num, nb_sectors, flags);
-}
-
 static void error_callback_bh(void *opaque)
 {
    struct BlockBackendAIOCB *acb = opaque;
@@ -675,15 +523,13 @@ static void error_callback_bh(void *opaque)
    qemu_aio_unref(acb);
 }

-BlockAIOCB *blk_abort_aio_request(BlockBackend *blk,
-                                  BlockCompletionFunc *cb,
-                                  void *opaque, int ret)
+static BlockAIOCB *abort_aio_request(BlockBackend *blk, BlockCompletionFunc *cb,
+                                     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);
@@ -699,7 +545,7 @@ BlockAIOCB *blk_aio_write_zeroes(BlockBackend *blk, int64_t sector_num,
 {
    int ret = blk_check_request(blk, sector_num, nb_sectors);
    if (ret < 0) {
-        return blk_abort_aio_request(blk, cb, opaque, ret);
+        return abort_aio_request(blk, cb, opaque, ret);
    }

    return bdrv_aio_write_zeroes(blk->bs, sector_num, nb_sectors, flags,
@@ -728,28 +574,16 @@ int blk_pwrite(BlockBackend *blk, int64_t offset, const void *buf, int count)

 int64_t blk_getlength(BlockBackend *blk)
 {
-    if (!blk_is_available(blk)) {
-        return -ENOMEDIUM;
-    }
-
    return bdrv_getlength(blk->bs);
 }

 void blk_get_geometry(BlockBackend *blk, uint64_t *nb_sectors_ptr)
 {
-    if (!blk->bs) {
-        *nb_sectors_ptr = 0;
-    } else {
-        bdrv_get_geometry(blk->bs, nb_sectors_ptr);
-    }
+    bdrv_get_geometry(blk->bs, nb_sectors_ptr);
 }

 int64_t blk_nb_sectors(BlockBackend *blk)
 {
-    if (!blk_is_available(blk)) {
-        return -ENOMEDIUM;
-    }
-
    return bdrv_nb_sectors(blk->bs);
 }

@@ -759,7 +593,7 @@ BlockAIOCB *blk_aio_readv(BlockBackend *blk, int64_t sector_num,
 {
    int ret = blk_check_request(blk, sector_num, nb_sectors);
    if (ret < 0) {
-        return blk_abort_aio_request(blk, cb, opaque, ret);
+        return abort_aio_request(blk, cb, opaque, ret);
    }

    return bdrv_aio_readv(blk->bs, sector_num, iov, nb_sectors, cb, opaque);
@@ -771,7 +605,7 @@ BlockAIOCB *blk_aio_writev(BlockBackend *blk, int64_t sector_num,
 {
    int ret = blk_check_request(blk, sector_num, nb_sectors);
    if (ret < 0) {
-        return blk_abort_aio_request(blk, cb, opaque, ret);
+        return abort_aio_request(blk, cb, opaque, ret);
    }

    return bdrv_aio_writev(blk->bs, sector_num, iov, nb_sectors, cb, opaque);
@@ -780,10 +614,6 @@ BlockAIOCB *blk_aio_writev(BlockBackend *blk, int64_t sector_num,
 BlockAIOCB *blk_aio_flush(BlockBackend *blk,
                          BlockCompletionFunc *cb, void *opaque)
 {
-    if (!blk_is_available(blk)) {
-        return blk_abort_aio_request(blk, cb, opaque, -ENOMEDIUM);
-    }
-
    return bdrv_aio_flush(blk->bs, cb, opaque);
 }

@@ -793,7 +623,7 @@ BlockAIOCB *blk_aio_discard(BlockBackend *blk,
 {
    int ret = blk_check_request(blk, sector_num, nb_sectors);
    if (ret < 0) {
-        return blk_abort_aio_request(blk, cb, opaque, ret);
+        return abort_aio_request(blk, cb, opaque, ret);
    }

    return bdrv_aio_discard(blk->bs, sector_num, nb_sectors, cb, opaque);
@@ -825,20 +655,12 @@ int blk_aio_multiwrite(BlockBackend *blk, BlockRequest *reqs, int num_reqs)

 int blk_ioctl(BlockBackend *blk, unsigned long int req, void *buf)
 {
-    if (!blk_is_available(blk)) {
-        return -ENOMEDIUM;
-    }
-
    return bdrv_ioctl(blk->bs, req, buf);
 }

 BlockAIOCB *blk_aio_ioctl(BlockBackend *blk, unsigned long int req, void *buf,
                          BlockCompletionFunc *cb, void *opaque)
 {
-    if (!blk_is_available(blk)) {
-        return blk_abort_aio_request(blk, cb, opaque, -ENOMEDIUM);
-    }
-
    return bdrv_aio_ioctl(blk->bs, req, buf, cb, opaque);
 }

@@ -854,19 +676,11 @@ int blk_co_discard(BlockBackend *blk, int64_t sector_num, int nb_sectors)

 int blk_co_flush(BlockBackend *blk)
 {
-    if (!blk_is_available(blk)) {
-        return -ENOMEDIUM;
-    }
-
    return bdrv_co_flush(blk->bs);
 }

 int blk_flush(BlockBackend *blk)
 {
-    if (!blk_is_available(blk)) {
-        return -ENOMEDIUM;
-    }
-
    return bdrv_flush(blk->bs);
 }

@@ -875,200 +689,81 @@ int blk_flush_all(void)
    return bdrv_flush_all();
 }

-void blk_drain(BlockBackend *blk)
-{
-    if (blk->bs) {
-        bdrv_drain(blk->bs);
-    }
-}
-
 void blk_drain_all(void)
 {
    bdrv_drain_all();
 }

-void blk_set_on_error(BlockBackend *blk, BlockdevOnError on_read_error,
-                      BlockdevOnError on_write_error)
-{
-    blk->on_read_error = on_read_error;
-    blk->on_write_error = on_write_error;
-}
-
 BlockdevOnError blk_get_on_error(BlockBackend *blk, bool is_read)
 {
-    return is_read ? blk->on_read_error : blk->on_write_error;
+    return bdrv_get_on_error(blk->bs, is_read);
 }

 BlockErrorAction blk_get_error_action(BlockBackend *blk, bool is_read,
                                      int error)
 {
-    BlockdevOnError on_err = blk_get_on_error(blk, is_read);
-
-    switch (on_err) {
-    case BLOCKDEV_ON_ERROR_ENOSPC:
-        return (error == ENOSPC) ?
-               BLOCK_ERROR_ACTION_STOP : BLOCK_ERROR_ACTION_REPORT;
-    case BLOCKDEV_ON_ERROR_STOP:
-        return BLOCK_ERROR_ACTION_STOP;
-    case BLOCKDEV_ON_ERROR_REPORT:
-        return BLOCK_ERROR_ACTION_REPORT;
-    case BLOCKDEV_ON_ERROR_IGNORE:
-        return BLOCK_ERROR_ACTION_IGNORE;
-    default:
-        abort();
-    }
+    return bdrv_get_error_action(blk->bs, is_read, error);
 }

-static void send_qmp_error_event(BlockBackend *blk,
-                                 BlockErrorAction action,
-                                 bool is_read, int error)
-{
-    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),
-                                   error == ENOSPC, strerror(error),
-                                   &error_abort);
-}
-
-/* This is done by device models because, while the block layer knows
- * about the error, it does not know whether an operation comes from
- * the device or the block layer (from a job, for example).
- */
 void blk_error_action(BlockBackend *blk, BlockErrorAction action,
                      bool is_read, int error)
 {
-    assert(error >= 0);
-
-    if (action == BLOCK_ERROR_ACTION_STOP) {
-        /* First set the iostatus, so that "info block" returns an iostatus
-         * that matches the events raised so far (an additional error iostatus
-         * is fine, but not a lost one).
-         */
-        blk_iostatus_set_err(blk, error);
-
-        /* Then raise the request to stop the VM and the event.
-         * qemu_system_vmstop_request_prepare has two effects.  First,
-         * it ensures that the STOP event always comes after the
-         * BLOCK_IO_ERROR event.  Second, it ensures that even if management
-         * can observe the STOP event and do a "cont" before the STOP
-         * event is issued, the VM will not stop.  In this case, vm_start()
-         * also ensures that the STOP/RESUME pair of events is emitted.
-         */
-        qemu_system_vmstop_request_prepare();
-        send_qmp_error_event(blk, action, is_read, error);
-        qemu_system_vmstop_request(RUN_STATE_IO_ERROR);
-    } else {
-        send_qmp_error_event(blk, action, is_read, error);
-    }
+    bdrv_error_action(blk->bs, action, is_read, error);
 }

 int blk_is_read_only(BlockBackend *blk)
 {
-    if (blk->bs) {
-        return bdrv_is_read_only(blk->bs);
-    } else {
-        return blk->root_state.read_only;
-    }
+    return bdrv_is_read_only(blk->bs);
 }

 int blk_is_sg(BlockBackend *blk)
 {
-    if (!blk->bs) {
-        return 0;
-    }
-
    return bdrv_is_sg(blk->bs);
 }

 int blk_enable_write_cache(BlockBackend *blk)
 {
-    if (blk->bs) {
-        return bdrv_enable_write_cache(blk->bs);
-    } else {
-        return !!(blk->root_state.open_flags & BDRV_O_CACHE_WB);
-    }
+    return bdrv_enable_write_cache(blk->bs);
 }

 void blk_set_enable_write_cache(BlockBackend *blk, bool wce)
 {
-    if (blk->bs) {
-        bdrv_set_enable_write_cache(blk->bs, wce);
-    } else {
-        if (wce) {
-            blk->root_state.open_flags |= BDRV_O_CACHE_WB;
-        } else {
-            blk->root_state.open_flags &= ~BDRV_O_CACHE_WB;
-        }
-    }
+    bdrv_set_enable_write_cache(blk->bs, wce);
 }

 void blk_invalidate_cache(BlockBackend *blk, Error **errp)
 {
-    if (!blk->bs) {
-        error_setg(errp, "Device '%s' has no medium", blk->name);
-        return;
-    }
-
    bdrv_invalidate_cache(blk->bs, errp);
 }

-bool blk_is_inserted(BlockBackend *blk)
+int blk_is_inserted(BlockBackend *blk)
 {
-    return blk->bs && bdrv_is_inserted(blk->bs);
-}
-
-bool blk_is_available(BlockBackend *blk)
-{
-    return blk_is_inserted(blk) && !blk_dev_is_tray_open(blk);
+    return bdrv_is_inserted(blk->bs);
 }

 void blk_lock_medium(BlockBackend *blk, bool locked)
 {
-    if (blk->bs) {
-        bdrv_lock_medium(blk->bs, locked);
-    }
+    bdrv_lock_medium(blk->bs, locked);
 }

 void blk_eject(BlockBackend *blk, bool eject_flag)
 {
-    if (blk->bs) {
-        bdrv_eject(blk->bs, eject_flag);
-    }
+    bdrv_eject(blk->bs, eject_flag);
 }

 int blk_get_flags(BlockBackend *blk)
 {
-    if (blk->bs) {
-        return bdrv_get_flags(blk->bs);
-    } else {
-        return blk->root_state.open_flags;
-    }
+    return bdrv_get_flags(blk->bs);
 }

 int blk_get_max_transfer_length(BlockBackend *blk)
 {
-    if (blk->bs) {
-        return blk->bs->bl.max_transfer_length;
-    } else {
-        return 0;
-    }
-}
-
-int blk_get_max_iov(BlockBackend *blk)
-{
-    return blk->bs->bl.max_iov;
+    return blk->bs->bl.max_transfer_length;
 }

 void blk_set_guest_block_size(BlockBackend *blk, int align)
 {
-    blk->guest_block_size = align;
-}
-
-void *blk_try_blockalign(BlockBackend *blk, size_t size)
-{
-    return qemu_try_blockalign(blk ? blk->bs : NULL, size);
+    bdrv_set_guest_block_size(blk->bs, align);
 }

 void *blk_blockalign(BlockBackend *blk, size_t size)
@@ -1078,64 +773,40 @@ void *blk_blockalign(BlockBackend *blk, size_t size)

 bool blk_op_is_blocked(BlockBackend *blk, BlockOpType op, Error **errp)
 {
-    if (!blk->bs) {
-        return false;
-    }
-
    return bdrv_op_is_blocked(blk->bs, op, errp);
 }

 void blk_op_unblock(BlockBackend *blk, BlockOpType op, Error *reason)
 {
-    if (blk->bs) {
-        bdrv_op_unblock(blk->bs, op, reason);
-    }
+    bdrv_op_unblock(blk->bs, op, reason);
 }

 void blk_op_block_all(BlockBackend *blk, Error *reason)
 {
-    if (blk->bs) {
-        bdrv_op_block_all(blk->bs, reason);
-    }
+    bdrv_op_block_all(blk->bs, reason);
 }

 void blk_op_unblock_all(BlockBackend *blk, Error *reason)
 {
-    if (blk->bs) {
-        bdrv_op_unblock_all(blk->bs, reason);
-    }
+    bdrv_op_unblock_all(blk->bs, reason);
 }

 AioContext *blk_get_aio_context(BlockBackend *blk)
 {
-    if (blk->bs) {
-        return bdrv_get_aio_context(blk->bs);
-    } else {
-        return qemu_get_aio_context();
-    }
-}
-
-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);
+    return bdrv_get_aio_context(blk->bs);
 }

 void blk_set_aio_context(BlockBackend *blk, AioContext *new_context)
 {
-    if (blk->bs) {
-        bdrv_set_aio_context(blk->bs, new_context);
-    }
+    bdrv_set_aio_context(blk->bs, new_context);
 }

 void blk_add_aio_context_notifier(BlockBackend *blk,
        void (*attached_aio_context)(AioContext *new_context, void *opaque),
        void (*detach_aio_context)(void *opaque), void *opaque)
 {
-    if (blk->bs) {
-        bdrv_add_aio_context_notifier(blk->bs, attached_aio_context,
-                                      detach_aio_context, opaque);
-    }
+    bdrv_add_aio_context_notifier(blk->bs, attached_aio_context,
+                                  detach_aio_context, opaque);
 }

 void blk_remove_aio_context_notifier(BlockBackend *blk,
@@ -1144,39 +815,28 @@ void blk_remove_aio_context_notifier(BlockBackend *blk,
                                     void (*detach_aio_context)(void *),
                                     void *opaque)
 {
-    if (blk->bs) {
-        bdrv_remove_aio_context_notifier(blk->bs, attached_aio_context,
-                                         detach_aio_context, opaque);
-    }
+    bdrv_remove_aio_context_notifier(blk->bs, attached_aio_context,
+                                     detach_aio_context, opaque);
 }

-void blk_add_remove_bs_notifier(BlockBackend *blk, Notifier *notify)
+void blk_add_close_notifier(BlockBackend *blk, Notifier *notify)
 {
-    notifier_list_add(&blk->remove_bs_notifiers, notify);
-}
-
-void blk_add_insert_bs_notifier(BlockBackend *blk, Notifier *notify)
-{
-    notifier_list_add(&blk->insert_bs_notifiers, notify);
+    bdrv_add_close_notifier(blk->bs, notify);
 }

 void blk_io_plug(BlockBackend *blk)
 {
-    if (blk->bs) {
-        bdrv_io_plug(blk->bs);
-    }
+    bdrv_io_plug(blk->bs);
 }

 void blk_io_unplug(BlockBackend *blk)
 {
-    if (blk->bs) {
-        bdrv_io_unplug(blk->bs);
-    }
+    bdrv_io_unplug(blk->bs);
 }

 BlockAcctStats *blk_get_stats(BlockBackend *blk)
 {
-    return &blk->stats;
+    return bdrv_get_stats(blk->bs);
 }

 void *blk_aio_get(const AIOCBInfo *aiocb_info, BlockBackend *blk,
@@ -1209,10 +869,6 @@ int blk_write_compressed(BlockBackend *blk, int64_t sector_num,

 int blk_truncate(BlockBackend *blk, int64_t offset)
 {
-    if (!blk_is_available(blk)) {
-        return -ENOMEDIUM;
-    }
-
    return bdrv_truncate(blk->bs, offset);
 }

@@ -1229,94 +885,20 @@ int blk_discard(BlockBackend *blk, int64_t sector_num, int nb_sectors)
 int blk_save_vmstate(BlockBackend *blk, const uint8_t *buf,
                     int64_t pos, int size)
 {
-    if (!blk_is_available(blk)) {
-        return -ENOMEDIUM;
-    }
-
    return bdrv_save_vmstate(blk->bs, buf, pos, size);
 }

 int blk_load_vmstate(BlockBackend *blk, uint8_t *buf, int64_t pos, int size)
 {
-    if (!blk_is_available(blk)) {
-        return -ENOMEDIUM;
-    }
-
    return bdrv_load_vmstate(blk->bs, buf, pos, size);
 }

 int blk_probe_blocksizes(BlockBackend *blk, BlockSizes *bsz)
 {
-    if (!blk_is_available(blk)) {
-        return -ENOMEDIUM;
-    }
-
    return bdrv_probe_blocksizes(blk->bs, bsz);
 }

 int blk_probe_geometry(BlockBackend *blk, HDGeometry *geo)
 {
-    if (!blk_is_available(blk)) {
-        return -ENOMEDIUM;
-    }
-
    return bdrv_probe_geometry(blk->bs, geo);
 }
-
-/*
- * Updates the BlockBackendRootState object with data from the currently
- * attached BlockDriverState.
- */
-void blk_update_root_state(BlockBackend *blk)
-{
-    assert(blk->bs);
-
-    blk->root_state.open_flags    = blk->bs->open_flags;
-    blk->root_state.read_only     = blk->bs->read_only;
-    blk->root_state.detect_zeroes = blk->bs->detect_zeroes;
-
-    if (blk->root_state.throttle_group) {
-        g_free(blk->root_state.throttle_group);
-        throttle_group_unref(blk->root_state.throttle_state);
-    }
-    if (blk->bs->throttle_state) {
-        const char *name = throttle_group_get_name(blk->bs);
-        blk->root_state.throttle_group = g_strdup(name);
-        blk->root_state.throttle_state = throttle_group_incref(name);
-    } else {
-        blk->root_state.throttle_group = NULL;
-        blk->root_state.throttle_state = NULL;
-    }
-}
-
-/*
- * Applies the information in the root state to the given BlockDriverState. This
- * does not include the flags which have to be specified for bdrv_open(), use
- * blk_get_open_flags_from_root_state() to inquire them.
- */
-void blk_apply_root_state(BlockBackend *blk, BlockDriverState *bs)
-{
-    bs->detect_zeroes = blk->root_state.detect_zeroes;
-    if (blk->root_state.throttle_group) {
-        bdrv_io_limits_enable(bs, blk->root_state.throttle_group);
-    }
-}
-
-/*
- * Returns the flags to be used for bdrv_open() of a BlockDriverState which is
- * supposed to inherit the root state.
- */
-int blk_get_open_flags_from_root_state(BlockBackend *blk)
-{
-    int bs_flags;
-
-    bs_flags = blk->root_state.read_only ? 0 : BDRV_O_RDWR;
-    bs_flags |= blk->root_state.open_flags & ~BDRV_O_RDWR;
-
-    return bs_flags;
-}
-
-BlockBackendRootState *blk_get_root_state(BlockBackend *blk)
-{
-    return &blk->root_state;
-}
--- a/block/bochs.c
+++ b/block/bochs.c
@@ -22,7 +22,6 @@
 * 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 "block/block_int.h"
 #include "qemu/module.h"
@@ -104,7 +103,7 @@ static int bochs_open(BlockDriverState *bs, QDict *options, int flags,

    bs->read_only = 1; // no write support yet

-    ret = bdrv_pread(bs->file->bs, 0, &bochs, sizeof(bochs));
+    ret = bdrv_pread(bs->file, 0, &bochs, sizeof(bochs));
    if (ret < 0) {
        return ret;
    }
@@ -138,7 +137,7 @@ static int bochs_open(BlockDriverState *bs, QDict *options, int flags,
        return -ENOMEM;
    }

-    ret = bdrv_pread(bs->file->bs, le32_to_cpu(bochs.header), s->catalog_bitmap,
+    ret = bdrv_pread(bs->file, le32_to_cpu(bochs.header), s->catalog_bitmap,
                     s->catalog_size * 4);
    if (ret < 0) {
        goto fail;
@@ -207,7 +206,7 @@ static int64_t seek_to_sector(BlockDriverState *bs, int64_t sector_num)
        (s->extent_blocks + s->bitmap_blocks));

    /* read in bitmap for current extent */
-    ret = bdrv_pread(bs->file->bs, bitmap_offset + (extent_offset / 8),
+    ret = bdrv_pread(bs->file, bitmap_offset + (extent_offset / 8),
                     &bitmap_entry, 1);
    if (ret < 0) {
        return ret;
@@ -230,7 +229,7 @@ static int bochs_read(BlockDriverState *bs, int64_t sector_num,
        if (block_offset < 0) {
            return block_offset;
        } else if (block_offset > 0) {
-            ret = bdrv_pread(bs->file->bs, block_offset, buf, 512);
+            ret = bdrv_pread(bs->file, block_offset, buf, 512);
            if (ret < 0) {
                return ret;
            }
--- a/block/cloop.c
+++ b/block/cloop.c
@@ -21,7 +21,6 @@
 * 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 "block/block_int.h"
 #include "qemu/module.h"
@@ -67,7 +66,7 @@ static int cloop_open(BlockDriverState *bs, QDict *options, int flags,
    bs->read_only = 1;

    /* read header */
-    ret = bdrv_pread(bs->file->bs, 128, &s->block_size, 4);
+    ret = bdrv_pread(bs->file, 128, &s->block_size, 4);
    if (ret < 0) {
        return ret;
    }
@@ -93,7 +92,7 @@ static int cloop_open(BlockDriverState *bs, QDict *options, int flags,
        return -EINVAL;
    }

-    ret = bdrv_pread(bs->file->bs, 128 + 4, &s->n_blocks, 4);
+    ret = bdrv_pread(bs->file, 128 + 4, &s->n_blocks, 4);
    if (ret < 0) {
        return ret;
    }
@@ -124,7 +123,7 @@ static int cloop_open(BlockDriverState *bs, QDict *options, int flags,
        return -ENOMEM;
    }

-    ret = bdrv_pread(bs->file->bs, 128 + 4 + 4, s->offsets, offsets_size);
+    ret = bdrv_pread(bs->file, 128 + 4 + 4, s->offsets, offsets_size);
    if (ret < 0) {
        goto fail;
    }
@@ -204,8 +203,8 @@ static inline int cloop_read_block(BlockDriverState *bs, int block_num)
        int ret;
        uint32_t bytes = s->offsets[block_num + 1] - s->offsets[block_num];

-        ret = bdrv_pread(bs->file->bs, s->offsets[block_num],
-                         s->compressed_block, bytes);
+        ret = bdrv_pread(bs->file, s->offsets[block_num], s->compressed_block,
+                         bytes);
        if (ret != bytes) {
            return -1;
        }
--- a/block/commit.c
+++ b/block/commit.c
@@ -12,13 +12,10 @@
 *
 */

-#include "qemu/osdep.h"
 #include "trace.h"
 #include "block/block_int.h"
 #include "block/blockjob.h"
-#include "qapi/qmp/qerror.h"
 #include "qemu/ratelimit.h"
-#include "sysemu/block-backend.h"

 enum {
    /*
@@ -189,7 +186,7 @@ static void commit_set_speed(BlockJob *job, int64_t speed, Error **errp)
    CommitBlockJob *s = container_of(job, CommitBlockJob, common);

    if (speed < 0) {
-        error_setg(errp, QERR_INVALID_PARAMETER, "speed");
+        error_set(errp, QERR_INVALID_PARAMETER, "speed");
        return;
    }
    ratelimit_set_speed(&s->limit, speed / BDRV_SECTOR_SIZE, SLICE_TIME);
@@ -215,7 +212,7 @@ void commit_start(BlockDriverState *bs, BlockDriverState *base,

    if ((on_error == BLOCKDEV_ON_ERROR_STOP ||
         on_error == BLOCKDEV_ON_ERROR_ENOSPC) &&
-        (!bs->blk || !blk_iostatus_is_enabled(bs->blk))) {
+        !bdrv_iostatus_is_enabled(bs)) {
        error_setg(errp, "Invalid parameter combination");
        return;
    }
@@ -237,14 +234,14 @@ void commit_start(BlockDriverState *bs, BlockDriverState *base,
    orig_overlay_flags = bdrv_get_flags(overlay_bs);

    /* convert base & overlay_bs to r/w, if necessary */
-    if (!(orig_overlay_flags & BDRV_O_RDWR)) {
-        reopen_queue = bdrv_reopen_queue(reopen_queue, overlay_bs, NULL,
-                                         orig_overlay_flags | BDRV_O_RDWR);
-    }
    if (!(orig_base_flags & BDRV_O_RDWR)) {
-        reopen_queue = bdrv_reopen_queue(reopen_queue, base, NULL,
+        reopen_queue = bdrv_reopen_queue(reopen_queue, base,
                                         orig_base_flags | BDRV_O_RDWR);
    }
+    if (!(orig_overlay_flags & BDRV_O_RDWR)) {
+        reopen_queue = bdrv_reopen_queue(reopen_queue, overlay_bs,
+                                         orig_overlay_flags | BDRV_O_RDWR);
+    }
    if (reopen_queue) {
        bdrv_reopen_multiple(reopen_queue, &local_err);
        if (local_err != NULL) {
--- a/block/curl.c
+++ b/block/curl.c
@@ -21,12 +21,9 @@
 * 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 "qemu/error-report.h"
 #include "block/block_int.h"
 #include "qapi/qmp/qbool.h"
-#include "qapi/qmp/qstring.h"
 #include <curl/curl.h>

 // #define DEBUG_CURL
@@ -155,20 +152,18 @@ static int curl_sock_cb(CURL *curl, curl_socket_t fd, int action,
    DPRINTF("CURL (AIO): Sock action %d on fd %d\n", action, fd);
    switch (action) {
        case CURL_POLL_IN:
-            aio_set_fd_handler(s->aio_context, fd, false,
-                               curl_multi_read, NULL, state);
+            aio_set_fd_handler(s->aio_context, fd, curl_multi_read,
+                               NULL, state);
            break;
        case CURL_POLL_OUT:
-            aio_set_fd_handler(s->aio_context, fd, false,
-                               NULL, curl_multi_do, state);
+            aio_set_fd_handler(s->aio_context, fd, NULL, curl_multi_do, state);
            break;
        case CURL_POLL_INOUT:
-            aio_set_fd_handler(s->aio_context, fd, false,
-                               curl_multi_read, curl_multi_do, state);
+            aio_set_fd_handler(s->aio_context, fd, curl_multi_read,
+                               curl_multi_do, state);
            break;
        case CURL_POLL_REMOVE:
-            aio_set_fd_handler(s->aio_context, fd, false,
-                               NULL, NULL, NULL);
+            aio_set_fd_handler(s->aio_context, fd, NULL, NULL, NULL);
            break;
    }

@@ -302,18 +297,6 @@ static void curl_multi_check_completion(BDRVCURLState *s)
            /* ACBs for successful messages get completed in curl_read_cb */
            if (msg->data.result != CURLE_OK) {
                int i;
-                static int errcount = 100;
-
-                /* Don't lose the original error message from curl, since
-                 * it contains extra data.
-                 */
-                if (errcount > 0) {
-                    error_report("curl: %s", state->errmsg);
-                    if (--errcount == 0) {
-                        error_report("curl: further errors suppressed");
-                    }
-                }
-
                for (i = 0; i < CURL_NUM_ACB; i++) {
                    CURLAIOCB *acb = state->acb[i];

@@ -321,7 +304,7 @@ static void curl_multi_check_completion(BDRVCURLState *s)
                        continue;
                    }

-                    acb->common.cb(acb->common.opaque, -EPROTO);
+                    acb->common.cb(acb->common.opaque, -EIO);
                    qemu_aio_unref(acb);
                    state->acb[i] = NULL;
                }
--- a/block/dictzip.c
+++ b/block/dictzip.c
@@ -0,0 +1,584 @@
+/*
+ * DictZip Block driver for dictzip enabled gzip files
+ *
+ * Use the "dictzip" tool from the "dictd" package to create gzip files that
+ * contain the extra DictZip headers.
+ *
+ * dictzip(1) is a compression program which creates compressed files in the
+ * gzip format (see RFC 1952). However, unlike gzip(1), dictzip(1) compresses
+ * the file in pieces and stores an index to the pieces in the gzip header.
+ * This allows random access to the file at the granularity of the compressed
+ * pieces (currently about 64kB) while maintaining good compression ratios
+ * (within 5% of the expected ratio for dictionary data).
+ * dictd(8) uses files stored in this format.
+ *
+ * For details on DictZip see http://dict.org/.
+ *
+ * Copyright (c) 2009 Alexander Graf <agraf@suse.de>
+ *
+ * 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-common.h"
+#include "block/block_int.h"
+#include <zlib.h>
+
+// #define DEBUG
+
+#ifdef DEBUG
+#define dprintf(fmt, ...) do { printf("dzip: " fmt, ## __VA_ARGS__); } while (0)
+#else
+#define dprintf(fmt, ...) do { } while (0)
+#endif
+
+#define SECTOR_SIZE 512
+#define Z_STREAM_COUNT 4
+#define CACHE_COUNT 20
+
+/* magic values */
+
+#define GZ_MAGIC1     0x1f
+#define GZ_MAGIC2     0x8b
+#define DZ_MAGIC1      'R'
+#define DZ_MAGIC2      'A'
+
+#define GZ_FEXTRA     0x04      /* Optional field (random access index)    */
+#define GZ_FNAME      0x08      /* Original name                           */
+#define GZ_COMMENT    0x10      /* Zero-terminated, human-readable comment */
+#define GZ_FHCRC      0x02      /* Header CRC16                            */
+
+/* offsets */
+
+#define GZ_ID            0      /* GZ_MAGIC (16bit)                        */
+#define GZ_FLG           3      /* FLaGs (see above)                       */
+#define GZ_XLEN         10      /* eXtra LENgth (16bit)                    */
+#define GZ_SI           12      /* Subfield ID (16bit)                     */
+#define GZ_VERSION      16      /* Version for subfield format             */
+#define GZ_CHUNKSIZE    18      /* Chunk size (16bit)                      */
+#define GZ_CHUNKCNT     20      /* Number of chunks (16bit)                */
+#define GZ_RNDDATA      22      /* Random access data (16bit)              */
+
+#define GZ_99_CHUNKSIZE 18      /* Chunk size (32bit)                      */
+#define GZ_99_CHUNKCNT  22      /* Number of chunks (32bit)                */
+#define GZ_99_FILESIZE  26      /* Size of unpacked file (64bit)           */
+#define GZ_99_RNDDATA   34      /* Random access data (32bit)              */
+
+struct BDRVDictZipState;
+
+typedef struct DictZipAIOCB {
+    BlockAIOCB common;
+    struct BDRVDictZipState *s;
+    QEMUIOVector *qiov;          /* QIOV of the original request */
+    QEMUIOVector *qiov_gz;       /* QIOV of the gz subrequest */
+    QEMUBH *bh;                  /* BH for cache */
+    z_stream *zStream;           /* stream to use for decoding */
+    int zStream_id;              /* stream id of the above pointer */
+    size_t start;                /* offset into the uncompressed file */
+    size_t len;                  /* uncompressed bytes to read */
+    uint8_t *gzipped;            /* the gzipped data */
+    uint8_t *buf;                /* cached result */
+    size_t gz_len;               /* amount of gzip data */
+    size_t gz_start;             /* uncompressed starting point of gzip data */
+    uint64_t offset;             /* offset for "start" into the uncompressed chunk */
+    int chunks_len;              /* amount of uncompressed data in all gzip data */
+} DictZipAIOCB;
+
+typedef struct dict_cache {
+    size_t start;
+    size_t len;
+    uint8_t *buf;
+} DictCache;
+
+typedef struct BDRVDictZipState {
+    BlockDriverState *hd;
+    z_stream zStream[Z_STREAM_COUNT];
+    DictCache cache[CACHE_COUNT];
+    int cache_index;
+    uint8_t  stream_in_use;
+    uint64_t chunk_len;
+    uint32_t chunk_cnt;
+    uint16_t *chunks;
+    uint32_t *chunks32;
+    uint64_t *offsets;
+    int64_t file_len;
+} BDRVDictZipState;
+
+static int start_zStream(z_stream *zStream)
+{
+    zStream->zalloc    = NULL;
+    zStream->zfree     = NULL;
+    zStream->opaque    = NULL;
+    zStream->next_in   = 0;
+    zStream->avail_in  = 0;
+    zStream->next_out  = NULL;
+    zStream->avail_out = 0;
+
+    return inflateInit2( zStream, -15 );
+}
+
+static QemuOptsList runtime_opts = {
+    .name = "dzip",
+    .head = QTAILQ_HEAD_INITIALIZER(runtime_opts.head),
+    .desc = {
+        {
+            .name = "filename",
+            .type = QEMU_OPT_STRING,
+            .help = "URL to the dictzip file",
+        },
+        { /* end of list */ }
+    },
+};
+
+static int dictzip_open(BlockDriverState *bs, QDict *options, int flags, Error **errp)
+{
+    BDRVDictZipState *s = bs->opaque;
+    const char *err = "Unknown (read error?)";
+    uint8_t magic[2];
+    char buf[100];
+    uint8_t header_flags;
+    uint16_t chunk_len16;
+    uint16_t chunk_cnt16;
+    uint32_t chunk_len32;
+    uint16_t header_ver;
+    uint16_t tmp_short;
+    uint64_t offset;
+    int chunks_len;
+    int headerLength = GZ_XLEN - 1;
+    int rnd_offs;
+    int ret;
+    int i;
+    QemuOpts *opts;
+    Error *local_err = NULL;
+    const char *filename;
+
+    opts = qemu_opts_create(&runtime_opts, NULL, 0, &error_abort);
+    qemu_opts_absorb_qdict(opts, options, &local_err);
+    if (local_err != NULL) {
+        error_propagate(errp, local_err);
+        ret = -EINVAL;
+        goto fail;
+    }
+
+    filename = qemu_opt_get(opts, "filename");
+
+    if (!strncmp(filename, "dzip://", 7))
+        filename += 7;
+    else if (!strncmp(filename, "dzip:", 5))
+        filename += 5;
+
+    ret = bdrv_open(&s->hd, filename, NULL, NULL, flags | BDRV_O_PROTOCOL, NULL, &local_err);
+    if (ret < 0) {
+        error_propagate(errp, local_err);
+        qemu_opts_del(opts);
+        return ret;
+    }
+
+    /* initialize zlib streams */
+    for (i = 0; i < Z_STREAM_COUNT; i++) {
+        if (start_zStream( &s->zStream[i] ) != Z_OK) {
+            err = s->zStream[i].msg;
+            goto fail;
+        }
+    }
+
+    /* gzip header */
+    if (bdrv_pread(s->hd, GZ_ID, &magic, sizeof(magic)) != sizeof(magic))
+        goto fail;
+
+    if (!((magic[0] == GZ_MAGIC1) && (magic[1] == GZ_MAGIC2))) {
+        err = "No gzip file";
+        goto fail;
+    }
+
+    /* dzip header */
+    if (bdrv_pread(s->hd, GZ_FLG, &header_flags, 1) != 1)
+        goto fail;
+
+    if (!(header_flags & GZ_FEXTRA)) {
+        err = "Not a dictzip file (wrong flags)";
+        goto fail;
+    }
+
+    /* extra length */
+    if (bdrv_pread(s->hd, GZ_XLEN, &tmp_short, 2) != 2)
+        goto fail;
+
+    headerLength += le16_to_cpu(tmp_short) + 2;
+
+    /* DictZip magic */
+    if (bdrv_pread(s->hd, GZ_SI, &magic, 2) != 2)
+        goto fail;
+
+    if (magic[0] != DZ_MAGIC1 || magic[1] != DZ_MAGIC2) {
+        err = "Not a dictzip file (missing extra magic)";
+        goto fail;
+    }
+
+    /* DictZip version */
+    if (bdrv_pread(s->hd, GZ_VERSION, &header_ver, 2) != 2)
+        goto fail;
+
+    header_ver = le16_to_cpu(header_ver);
+
+    switch (header_ver) {
+        case 1: /* Normal DictZip */
+            /* number of chunks */
+            if (bdrv_pread(s->hd, GZ_CHUNKSIZE, &chunk_len16, 2) != 2)
+                goto fail;
+
+            s->chunk_len = le16_to_cpu(chunk_len16);
+
+            /* chunk count */
+            if (bdrv_pread(s->hd, GZ_CHUNKCNT, &chunk_cnt16, 2) != 2)
+                goto fail;
+
+            s->chunk_cnt = le16_to_cpu(chunk_cnt16);
+            chunks_len = sizeof(short) * s->chunk_cnt;
+            rnd_offs = GZ_RNDDATA;
+            break;
+        case 99: /* Special Alex pigz version */
+            /* number of chunks */
+            if (bdrv_pread(s->hd, GZ_99_CHUNKSIZE, &chunk_len32, 4) != 4)
+                goto fail;
+
+            dprintf("chunk len [%#x] = %d\n", GZ_99_CHUNKSIZE, chunk_len32);
+            s->chunk_len = le32_to_cpu(chunk_len32);
+
+            /* chunk count */
+            if (bdrv_pread(s->hd, GZ_99_CHUNKCNT, &s->chunk_cnt, 4) != 4)
+                goto fail;
+
+            s->chunk_cnt = le32_to_cpu(s->chunk_cnt);
+
+            dprintf("chunk len | count = %"PRId64" | %d\n", s->chunk_len, s->chunk_cnt);
+
+            /* file size */
+            if (bdrv_pread(s->hd, GZ_99_FILESIZE, &s->file_len, 8) != 8)
+                goto fail;
+
+            s->file_len = le64_to_cpu(s->file_len);
+            chunks_len = sizeof(int) * s->chunk_cnt;
+            rnd_offs = GZ_99_RNDDATA;
+            break;
+        default:
+            err = "Invalid DictZip version";
+            goto fail;
+    }
+
+    /* random access data */
+    s->chunks = g_malloc(chunks_len);
+    if (header_ver == 99)
+        s->chunks32 = (uint32_t *)s->chunks;
+
+    if (bdrv_pread(s->hd, rnd_offs, s->chunks, chunks_len) != chunks_len)
+        goto fail;
+
+    /* orig filename */
+    if (header_flags & GZ_FNAME) {
+        if (bdrv_pread(s->hd, headerLength + 1, buf, sizeof(buf)) != sizeof(buf))
+            goto fail;
+
+        buf[sizeof(buf) - 1] = '\0';
+        headerLength += strlen(buf) + 1;
+
+        if (strlen(buf) == sizeof(buf))
+            goto fail;
+
+        dprintf("filename: %s\n", buf);
+    }
+
+    /* comment field */
+    if (header_flags & GZ_COMMENT) {
+        if (bdrv_pread(s->hd, headerLength, buf, sizeof(buf)) != sizeof(buf))
+            goto fail;
+
+        buf[sizeof(buf) - 1] = '\0';
+        headerLength += strlen(buf) + 1;
+
+        if (strlen(buf) == sizeof(buf))
+            goto fail;
+
+        dprintf("comment: %s\n", buf);
+    }
+
+    if (header_flags & GZ_FHCRC)
+        headerLength += 2;
+
+    /* uncompressed file length*/
+    if (!s->file_len) {
+        uint32_t file_len;
+
+        if (bdrv_pread(s->hd, bdrv_getlength(s->hd) - 4, &file_len, 4) != 4)
+            goto fail;
+
+        s->file_len = le32_to_cpu(file_len);
+    }
+
+    /* compute offsets */
+    s->offsets = g_malloc(sizeof( *s->offsets ) * s->chunk_cnt);
+
+    for (offset = headerLength + 1, i = 0; i < s->chunk_cnt; i++) {
+        s->offsets[i] = offset;
+        switch (header_ver) {
+        case 1:
+            offset += le16_to_cpu(s->chunks[i]);
+            break;
+        case 99:
+            offset += le32_to_cpu(s->chunks32[i]);
+            break;
+        }
+
+        dprintf("chunk %#"PRIx64" - %#"PRIx64" = offset %#"PRIx64" -> %#"PRIx64"\n", i * s->chunk_len, (i+1) * s->chunk_len, s->offsets[i], offset);
+    }
+    qemu_opts_del(opts);
+
+    return 0;
+
+fail:
+    fprintf(stderr, "DictZip: Error opening file: %s\n", err);
+    bdrv_unref(s->hd);
+    if (s->chunks)
+        g_free(s->chunks);
+    qemu_opts_del(opts);
+    return -EINVAL;
+}
+
+/* This callback gets invoked when we have the result in cache already */
+static void dictzip_cache_cb(void *opaque)
+{
+    DictZipAIOCB *acb = (DictZipAIOCB *)opaque;
+
+    qemu_iovec_from_buf(acb->qiov, 0, acb->buf, acb->len);
+    acb->common.cb(acb->common.opaque, 0);
+    qemu_bh_delete(acb->bh);
+    qemu_aio_unref(acb);
+}
+
+/* This callback gets invoked by the underlying block reader when we have
+ * all compressed data. We uncompress in here. */
+static void dictzip_read_cb(void *opaque, int ret)
+{
+    DictZipAIOCB *acb = (DictZipAIOCB *)opaque;
+    struct BDRVDictZipState *s = acb->s;
+    uint8_t *buf;
+    DictCache *cache;
+    int r, i;
+
+    buf = g_malloc(acb->chunks_len);
+
+    /* try to find zlib stream for decoding */
+    do {
+        for (i = 0; i < Z_STREAM_COUNT; i++) {
+            if (!(s->stream_in_use & (1 << i))) {
+                s->stream_in_use |= (1 << i);
+                acb->zStream_id = i;
+                acb->zStream = &s->zStream[i];
+                break;
+            }
+        }
+    } while(!acb->zStream);
+
+    /* sure, we could handle more streams, but this callback should be single
+       threaded and when it's not, we really want to know! */
+    assert(i == 0);
+
+    /* uncompress the chunk */
+    acb->zStream->next_in   = acb->gzipped;
+    acb->zStream->avail_in  = acb->gz_len;
+    acb->zStream->next_out  = buf;
+    acb->zStream->avail_out = acb->chunks_len;
+
+    r = inflate( acb->zStream,  Z_PARTIAL_FLUSH );
+    if ( (r != Z_OK) && (r != Z_STREAM_END) )
+        fprintf(stderr, "Error inflating: [%d] %s\n", r, acb->zStream->msg);
+
+    if ( r == Z_STREAM_END )
+        inflateReset(acb->zStream);
+
+    dprintf("inflating [%d] left: %d | %d bytes\n", r, acb->zStream->avail_in, acb->zStream->avail_out);
+    s->stream_in_use &= ~(1 << acb->zStream_id);
+
+    /* nofity the caller */
+    qemu_iovec_from_buf(acb->qiov, 0, buf + acb->offset, acb->len);
+    acb->common.cb(acb->common.opaque, 0);
+
+    /* fill the cache */
+    cache = &s->cache[s->cache_index];
+    s->cache_index++;
+    if (s->cache_index == CACHE_COUNT)
+        s->cache_index = 0;
+
+    cache->len = 0;
+    if (cache->buf)
+        g_free(cache->buf);
+    cache->start = acb->gz_start;
+    cache->buf = buf;
+    cache->len = acb->chunks_len;
+
+    /* free occupied ressources */
+    g_free(acb->qiov_gz);
+    qemu_aio_unref(acb);
+}
+
+static const AIOCBInfo dictzip_aiocb_info = {
+    .aiocb_size         = sizeof(DictZipAIOCB),
+};
+
+/* This is where we get a request from a caller to read something */
+static BlockAIOCB *dictzip_aio_readv(BlockDriverState *bs,
+        int64_t sector_num, QEMUIOVector *qiov, int nb_sectors,
+        BlockCompletionFunc *cb, void *opaque)
+{
+    BDRVDictZipState *s = bs->opaque;
+    DictZipAIOCB *acb;
+    QEMUIOVector *qiov_gz;
+    struct iovec *iov;
+    uint8_t *buf;
+    size_t  start = sector_num * SECTOR_SIZE;
+    size_t  len = nb_sectors * SECTOR_SIZE;
+    size_t  end = start + len;
+    size_t  gz_start;
+    size_t  gz_len;
+    int64_t gz_sector_num;
+    int     gz_nb_sectors;
+    int     first_chunk, last_chunk;
+    int     first_offset;
+    int     i;
+
+    acb = qemu_aio_get(&dictzip_aiocb_info, bs, cb, opaque);
+    if (!acb)
+        return NULL;
+
+    /* Search Cache */
+    for (i = 0; i < CACHE_COUNT; i++) {
+        if (!s->cache[i].len)
+            continue;
+
+        if ((start >= s->cache[i].start) &&
+            (end <= (s->cache[i].start + s->cache[i].len))) {
+            acb->buf = s->cache[i].buf + (start - s->cache[i].start);
+            acb->len = len;
+            acb->qiov = qiov;
+            acb->bh = qemu_bh_new(dictzip_cache_cb, acb);
+            qemu_bh_schedule(acb->bh);
+
+            return &acb->common;
+        }
+    }
+
+    /* No cache, so let's decode */
+    /* We need to read these chunks */
+    first_chunk  = start / s->chunk_len;
+    first_offset = start - first_chunk * s->chunk_len;
+    last_chunk   = end / s->chunk_len;
+
+    gz_start = s->offsets[first_chunk];
+    gz_len = 0;
+    for (i = first_chunk; i <= last_chunk; i++) {
+        if (s->chunks32)
+            gz_len += le32_to_cpu(s->chunks32[i]);
+        else
+            gz_len += le16_to_cpu(s->chunks[i]);
+    }
+
+    gz_sector_num = gz_start / SECTOR_SIZE;
+    gz_nb_sectors = (gz_len / SECTOR_SIZE);
+
+    /* account for tail and heads */
+    while ((gz_start + gz_len) > ((gz_sector_num + gz_nb_sectors) * SECTOR_SIZE))
+        gz_nb_sectors++;
+
+    /* Allocate qiov, iov and buf in one chunk so we only need to free qiov */
+    qiov_gz = g_malloc0(sizeof(QEMUIOVector) + sizeof(struct iovec) +
+                           (gz_nb_sectors * SECTOR_SIZE));
+    iov = (struct iovec *)(((char *)qiov_gz) + sizeof(QEMUIOVector));
+    buf = ((uint8_t *)iov) + sizeof(struct iovec *);
+
+    /* Kick off the read by the backing file, so we can start decompressing */
+    iov->iov_base = (void *)buf;
+    iov->iov_len = gz_nb_sectors * 512;
+    qemu_iovec_init_external(qiov_gz, iov, 1);
+
+    dprintf("read %zd - %zd => %zd - %zd\n", start, end, gz_start, gz_start + gz_len);
+
+    acb->s = s;
+    acb->qiov = qiov;
+    acb->qiov_gz = qiov_gz;
+    acb->start = start;
+    acb->len = len;
+    acb->gzipped = buf + (gz_start % SECTOR_SIZE);
+    acb->gz_len = gz_len;
+    acb->gz_start = first_chunk * s->chunk_len;
+    acb->offset = first_offset;
+    acb->chunks_len = (last_chunk - first_chunk + 1) * s->chunk_len;
+
+    return bdrv_aio_readv(s->hd, gz_sector_num, qiov_gz, gz_nb_sectors,
+                          dictzip_read_cb, acb);
+}
+
+static void dictzip_close(BlockDriverState *bs)
+{
+    BDRVDictZipState *s = bs->opaque;
+    int i;
+
+    for (i = 0; i < CACHE_COUNT; i++) {
+        if (!s->cache[i].len)
+            continue;
+
+        g_free(s->cache[i].buf);
+    }
+
+    for (i = 0; i < Z_STREAM_COUNT; i++) {
+        inflateEnd(&s->zStream[i]);
+    }
+
+    if (s->chunks)
+        g_free(s->chunks);
+
+    if (s->offsets)
+        g_free(s->offsets);
+
+    dprintf("Close\n");
+}
+
+static int64_t dictzip_getlength(BlockDriverState *bs)
+{
+    BDRVDictZipState *s = bs->opaque;
+    dprintf("getlength -> %ld\n", s->file_len);
+    return s->file_len;
+}
+
+static BlockDriver bdrv_dictzip = {
+    .format_name     = "dzip",
+    .protocol_name   = "dzip",
+
+    .instance_size   = sizeof(BDRVDictZipState),
+    .bdrv_file_open  = dictzip_open,
+    .bdrv_close      = dictzip_close,
+    .bdrv_getlength  = dictzip_getlength,
+
+    .bdrv_aio_readv  = dictzip_aio_readv,
+};
+
+static void dictzip_block_init(void)
+{
+    bdrv_register(&bdrv_dictzip);
+}
+
+block_init(dictzip_block_init);
--- a/block/dmg.c
+++ b/block/dmg.c
@@ -21,11 +21,9 @@
 * 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 "block/block_int.h"
 #include "qemu/bswap.h"
-#include "qemu/error-report.h"
 #include "qemu/module.h"
 #include <zlib.h>
 #ifdef CONFIG_BZIP2
@@ -86,7 +84,7 @@ static int read_uint64(BlockDriverState *bs, int64_t offset, uint64_t *result)
    uint64_t buffer;
    int ret;

-    ret = bdrv_pread(bs->file->bs, offset, &buffer, 8);
+    ret = bdrv_pread(bs->file, offset, &buffer, 8);
    if (ret < 0) {
        return ret;
    }
@@ -100,7 +98,7 @@ static int read_uint32(BlockDriverState *bs, int64_t offset, uint32_t *result)
    uint32_t buffer;
    int ret;

-    ret = bdrv_pread(bs->file->bs, offset, &buffer, 4);
+    ret = bdrv_pread(bs->file, offset, &buffer, 4);
    if (ret < 0) {
        return ret;
    }
@@ -355,7 +353,7 @@ static int dmg_read_resource_fork(BlockDriverState *bs, DmgHeaderState *ds,
        offset += 4;

        buffer = g_realloc(buffer, count);
-        ret = bdrv_pread(bs->file->bs, offset, buffer, count);
+        ret = bdrv_pread(bs->file, offset, buffer, count);
        if (ret < 0) {
            goto fail;
        }
@@ -392,7 +390,7 @@ static int dmg_read_plist_xml(BlockDriverState *bs, DmgHeaderState *ds,

    buffer = g_malloc(info_length + 1);
    buffer[info_length] = '\0';
-    ret = bdrv_pread(bs->file->bs, info_begin, buffer, info_length);
+    ret = bdrv_pread(bs->file, info_begin, buffer, info_length);
    if (ret != info_length) {
        ret = -EINVAL;
        goto fail;
@@ -447,7 +445,7 @@ static int dmg_open(BlockDriverState *bs, QDict *options, int flags,
    ds.max_sectors_per_chunk = 1;

    /* locate the UDIF trailer */
-    offset = dmg_find_koly_offset(bs->file->bs, errp);
+    offset = dmg_find_koly_offset(bs->file, errp);
    if (offset < 0) {
        ret = offset;
        goto fail;
@@ -515,9 +513,9 @@ static int dmg_open(BlockDriverState *bs, QDict *options, int flags,
    }

    /* initialize zlib engine */
-    s->compressed_chunk = qemu_try_blockalign(bs->file->bs,
+    s->compressed_chunk = qemu_try_blockalign(bs->file,
                                              ds.max_compressed_size + 1);
-    s->uncompressed_chunk = qemu_try_blockalign(bs->file->bs,
+    s->uncompressed_chunk = qemu_try_blockalign(bs->file,
                                                512 * ds.max_sectors_per_chunk);
    if (s->compressed_chunk == NULL || s->uncompressed_chunk == NULL) {
        ret = -ENOMEM;
@@ -593,7 +591,7 @@ static inline int dmg_read_chunk(BlockDriverState *bs, uint64_t sector_num)
        case 0x80000005: { /* zlib compressed */
            /* we need to buffer, because only the chunk as whole can be
             * inflated. */
-            ret = bdrv_pread(bs->file->bs, s->offsets[chunk],
+            ret = bdrv_pread(bs->file, s->offsets[chunk],
                             s->compressed_chunk, s->lengths[chunk]);
            if (ret != s->lengths[chunk]) {
                return -1;
@@ -617,7 +615,7 @@ static inline int dmg_read_chunk(BlockDriverState *bs, uint64_t sector_num)
        case 0x80000006: /* bzip2 compressed */
            /* we need to buffer, because only the chunk as whole can be
             * inflated. */
-            ret = bdrv_pread(bs->file->bs, s->offsets[chunk],
+            ret = bdrv_pread(bs->file, s->offsets[chunk],
                             s->compressed_chunk, s->lengths[chunk]);
            if (ret != s->lengths[chunk]) {
                return -1;
@@ -642,7 +640,7 @@ static inline int dmg_read_chunk(BlockDriverState *bs, uint64_t sector_num)
            break;
 #endif /* CONFIG_BZIP2 */
        case 1: /* copy */
-            ret = bdrv_pread(bs->file->bs, s->offsets[chunk],
+            ret = bdrv_pread(bs->file, s->offsets[chunk],
                             s->uncompressed_chunk, s->lengths[chunk]);
            if (ret != s->lengths[chunk]) {
                return -1;
--- a/block/gluster.c
+++ b/block/gluster.c
@@ -7,7 +7,6 @@
 * See the COPYING file in the top-level directory.
 *
 */
-#include "qemu/osdep.h"
 #include <glusterfs/api/glfs.h>
 #include "block/block_int.h"
 #include "qemu/uri.h"
@@ -430,23 +429,28 @@ static coroutine_fn int qemu_gluster_co_write_zeroes(BlockDriverState *bs,
        int64_t sector_num, int nb_sectors, BdrvRequestFlags flags)
 {
    int ret;
-    GlusterAIOCB acb;
+    GlusterAIOCB *acb = g_slice_new(GlusterAIOCB);
    BDRVGlusterState *s = bs->opaque;
    off_t size = nb_sectors * BDRV_SECTOR_SIZE;
    off_t offset = sector_num * BDRV_SECTOR_SIZE;

-    acb.size = size;
-    acb.ret = 0;
-    acb.coroutine = qemu_coroutine_self();
-    acb.aio_context = bdrv_get_aio_context(bs);
+    acb->size = size;
+    acb->ret = 0;
+    acb->coroutine = qemu_coroutine_self();
+    acb->aio_context = bdrv_get_aio_context(bs);

-    ret = glfs_zerofill_async(s->fd, offset, size, gluster_finish_aiocb, &acb);
+    ret = glfs_zerofill_async(s->fd, offset, size, &gluster_finish_aiocb, acb);
    if (ret < 0) {
-        return -errno;
+        ret = -errno;
+        goto out;
    }

    qemu_coroutine_yield();
-    return acb.ret;
+    ret = acb->ret;
+
+out:
+    g_slice_free(GlusterAIOCB, acb);
+    return ret;
 }

 static inline bool gluster_supports_zerofill(void)
@@ -537,30 +541,35 @@ static coroutine_fn int qemu_gluster_co_rw(BlockDriverState *bs,
        int64_t sector_num, int nb_sectors, QEMUIOVector *qiov, int write)
 {
    int ret;
-    GlusterAIOCB acb;
+    GlusterAIOCB *acb = g_slice_new(GlusterAIOCB);
    BDRVGlusterState *s = bs->opaque;
    size_t size = nb_sectors * BDRV_SECTOR_SIZE;
    off_t offset = sector_num * BDRV_SECTOR_SIZE;

-    acb.size = size;
-    acb.ret = 0;
-    acb.coroutine = qemu_coroutine_self();
-    acb.aio_context = bdrv_get_aio_context(bs);
+    acb->size = size;
+    acb->ret = 0;
+    acb->coroutine = qemu_coroutine_self();
+    acb->aio_context = bdrv_get_aio_context(bs);

    if (write) {
        ret = glfs_pwritev_async(s->fd, qiov->iov, qiov->niov, offset, 0,
-            gluster_finish_aiocb, &acb);
+            &gluster_finish_aiocb, acb);
    } else {
        ret = glfs_preadv_async(s->fd, qiov->iov, qiov->niov, offset, 0,
-            gluster_finish_aiocb, &acb);
+            &gluster_finish_aiocb, acb);
    }

    if (ret < 0) {
-        return -errno;
+        ret = -errno;
+        goto out;
    }

    qemu_coroutine_yield();
-    return acb.ret;
+    ret = acb->ret;
+
+out:
+    g_slice_free(GlusterAIOCB, acb);
+    return ret;
 }

 static int qemu_gluster_truncate(BlockDriverState *bs, int64_t offset)
@@ -591,21 +600,26 @@ static coroutine_fn int qemu_gluster_co_writev(BlockDriverState *bs,
 static coroutine_fn int qemu_gluster_co_flush_to_disk(BlockDriverState *bs)
 {
    int ret;
-    GlusterAIOCB acb;
+    GlusterAIOCB *acb = g_slice_new(GlusterAIOCB);
    BDRVGlusterState *s = bs->opaque;

-    acb.size = 0;
-    acb.ret = 0;
-    acb.coroutine = qemu_coroutine_self();
-    acb.aio_context = bdrv_get_aio_context(bs);
+    acb->size = 0;
+    acb->ret = 0;
+    acb->coroutine = qemu_coroutine_self();
+    acb->aio_context = bdrv_get_aio_context(bs);

-    ret = glfs_fsync_async(s->fd, gluster_finish_aiocb, &acb);
+    ret = glfs_fsync_async(s->fd, &gluster_finish_aiocb, acb);
    if (ret < 0) {
-        return -errno;
+        ret = -errno;
+        goto out;
    }

    qemu_coroutine_yield();
-    return acb.ret;
+    ret = acb->ret;
+
+out:
+    g_slice_free(GlusterAIOCB, acb);
+    return ret;
 }

 #ifdef CONFIG_GLUSTERFS_DISCARD
@@ -613,23 +627,28 @@ static coroutine_fn int qemu_gluster_co_discard(BlockDriverState *bs,
        int64_t sector_num, int nb_sectors)
 {
    int ret;
-    GlusterAIOCB acb;
+    GlusterAIOCB *acb = g_slice_new(GlusterAIOCB);
    BDRVGlusterState *s = bs->opaque;
    size_t size = nb_sectors * BDRV_SECTOR_SIZE;
    off_t offset = sector_num * BDRV_SECTOR_SIZE;

-    acb.size = 0;
-    acb.ret = 0;
-    acb.coroutine = qemu_coroutine_self();
-    acb.aio_context = bdrv_get_aio_context(bs);
+    acb->size = 0;
+    acb->ret = 0;
+    acb->coroutine = qemu_coroutine_self();
+    acb->aio_context = bdrv_get_aio_context(bs);

-    ret = glfs_discard_async(s->fd, offset, size, gluster_finish_aiocb, &acb);
+    ret = glfs_discard_async(s->fd, offset, size, &gluster_finish_aiocb, acb);
    if (ret < 0) {
-        return -errno;
+        ret = -errno;
+        goto out;
    }

    qemu_coroutine_yield();
-    return acb.ret;
+    ret = acb->ret;
+
+out:
+    g_slice_free(GlusterAIOCB, acb);
+    return ret;
 }
 #endif

--- a/block/io.c
+++ b/block/io.c
--- a/block/iscsi.c
+++ b/block/iscsi.c
@@ -2,7 +2,7 @@
 * QEMU Block driver for iSCSI images
 *
 * Copyright (c) 2010-2011 Ronnie Sahlberg <ronniesahlberg@gmail.com>
- * Copyright (c) 2012-2015 Peter Lieven <pl@kamp.de>
+ * Copyright (c) 2012-2014 Peter Lieven <pl@kamp.de>
 *
 * Permission is hereby granted, free of charge, to any person obtaining a copy
 * of this software and associated documentation files (the "Software"), to deal
@@ -23,7 +23,7 @@
 * THE SOFTWARE.
 */

-#include "qemu/osdep.h"
+#include "config-host.h"

 #include <poll.h>
 #include <math.h>
@@ -38,7 +38,6 @@
 #include "qemu/iov.h"
 #include "sysemu/sysemu.h"
 #include "qmp-commands.h"
-#include "qapi/qmp/qstring.h"

 #include <iscsi/iscsi.h>
 #include <iscsi/scsi-lowlevel.h>
@@ -58,6 +57,9 @@ typedef struct IscsiLun {
    int events;
    QEMUTimer *nop_timer;
    QEMUTimer *event_timer;
+    uint8_t lbpme;
+    uint8_t lbprz;
+    uint8_t has_write_same;
    struct scsi_inquiry_logical_block_provisioning lbp;
    struct scsi_inquiry_block_limits bl;
    unsigned char *zeroblock;
@@ -65,12 +67,6 @@ typedef struct IscsiLun {
    int cluster_sectors;
    bool use_16_for_rw;
    bool write_protected;
-    bool lbpme;
-    bool lbprz;
-    bool dpofua;
-    bool has_write_same;
-    bool force_next_flush;
-    bool request_timed_out;
 } IscsiLun;

 typedef struct IscsiTask {
@@ -83,8 +79,6 @@ typedef struct IscsiTask {
    QEMUBH *bh;
    IscsiLun *iscsilun;
    QEMUTimer retry_timer;
-    bool force_next_flush;
-    int err_code;
 } IscsiTask;

 typedef struct IscsiAIOCB {
@@ -97,18 +91,16 @@ typedef struct IscsiAIOCB {
    int status;
    int64_t sector_num;
    int nb_sectors;
-    int ret;
 #ifdef __linux__
    sg_io_hdr_t *ioh;
 #endif
 } IscsiAIOCB;

-/* libiscsi uses time_t so its enough to process events every second */
-#define EVENT_INTERVAL 1000
+#define EVENT_INTERVAL 250
 #define NOP_INTERVAL 5000
 #define MAX_NOP_FAILURES 3
 #define ISCSI_CMD_RETRIES ARRAY_SIZE(iscsi_retry_times)
-static const unsigned iscsi_retry_times[] = {8, 32, 128, 512, 2048, 8192, 32768};
+static const unsigned iscsi_retry_times[] = {8, 32, 128, 512, 2048};

 /* this threshold is a trade-off knob to choose between
 * the potential additional overhead of an extra GET_LBA_STATUS request
@@ -171,70 +163,6 @@ static inline unsigned exp_random(double mean)
    return -mean * log((double)rand() / RAND_MAX);
 }

-/* SCSI_SENSE_ASCQ_INVALID_FIELD_IN_PARAMETER_LIST was introduced in
- * libiscsi 1.10.0, together with other constants we need.  Use it as
- * a hint that we have to define them ourselves if needed, to keep the
- * minimum required libiscsi version at 1.9.0.  We use an ASCQ macro for
- * the test because SCSI_STATUS_* is an enum.
- *
- * To guard against future changes where SCSI_SENSE_ASCQ_* also becomes
- * an enum, check against the LIBISCSI_API_VERSION macro, which was
- * introduced in 1.11.0.  If it is present, there is no need to define
- * anything.
- */
-#if !defined(SCSI_SENSE_ASCQ_INVALID_FIELD_IN_PARAMETER_LIST) && \
-    !defined(LIBISCSI_API_VERSION)
-#define SCSI_STATUS_TASK_SET_FULL                          0x28
-#define SCSI_STATUS_TIMEOUT                                0x0f000002
-#define SCSI_SENSE_ASCQ_INVALID_FIELD_IN_PARAMETER_LIST    0x2600
-#define SCSI_SENSE_ASCQ_PARAMETER_LIST_LENGTH_ERROR        0x1a00
-#endif
-
-static int iscsi_translate_sense(struct scsi_sense *sense)
-{
-    int ret;
-
-    switch (sense->key) {
-    case SCSI_SENSE_NOT_READY:
-        return -EBUSY;
-    case SCSI_SENSE_DATA_PROTECTION:
-        return -EACCES;
-    case SCSI_SENSE_COMMAND_ABORTED:
-        return -ECANCELED;
-    case SCSI_SENSE_ILLEGAL_REQUEST:
-        /* Parse ASCQ */
-        break;
-    default:
-        return -EIO;
-    }
-    switch (sense->ascq) {
-    case SCSI_SENSE_ASCQ_PARAMETER_LIST_LENGTH_ERROR:
-    case SCSI_SENSE_ASCQ_INVALID_OPERATION_CODE:
-    case SCSI_SENSE_ASCQ_INVALID_FIELD_IN_CDB:
-    case SCSI_SENSE_ASCQ_INVALID_FIELD_IN_PARAMETER_LIST:
-        ret = -EINVAL;
-        break;
-    case SCSI_SENSE_ASCQ_LBA_OUT_OF_RANGE:
-        ret = -ENOSPC;
-        break;
-    case SCSI_SENSE_ASCQ_LOGICAL_UNIT_NOT_SUPPORTED:
-        ret = -ENOTSUP;
-        break;
-    case SCSI_SENSE_ASCQ_MEDIUM_NOT_PRESENT:
-    case SCSI_SENSE_ASCQ_MEDIUM_NOT_PRESENT_TRAY_CLOSED:
-    case SCSI_SENSE_ASCQ_MEDIUM_NOT_PRESENT_TRAY_OPEN:
-        ret = -ENOMEDIUM;
-        break;
-    case SCSI_SENSE_ASCQ_WRITE_PROTECTED:
-        ret = -EACCES;
-        break;
-    default:
-        ret = -EIO;
-        break;
-    }
-    return ret;
-}
-
 static void
 iscsi_co_generic_cb(struct iscsi_context *iscsi, int status,
                        void *command_data, void *opaque)
@@ -255,19 +183,10 @@ iscsi_co_generic_cb(struct iscsi_context *iscsi, int status,
                iTask->do_retry = 1;
                goto out;
            }
-            if (status == SCSI_STATUS_BUSY ||
-                status == SCSI_STATUS_TIMEOUT ||
-                status == SCSI_STATUS_TASK_SET_FULL) {
+            if (status == SCSI_STATUS_BUSY) {
                unsigned retry_time =
                    exp_random(iscsi_retry_times[iTask->retries - 1]);
-                if (status == SCSI_STATUS_TIMEOUT) {
-                    /* make sure the request is rescheduled AFTER the
-                     * reconnect is initiated */
-                    retry_time = EVENT_INTERVAL * 2;
-                    iTask->iscsilun->request_timed_out = true;
-                }
-                error_report("iSCSI Busy/TaskSetFull/TimeOut"
-                             " (retry #%u in %u ms): %s",
+                error_report("iSCSI Busy (retry #%u in %u ms): %s",
                             iTask->retries, retry_time,
                             iscsi_get_error(iscsi));
                aio_timer_init(iTask->iscsilun->aio_context,
@@ -279,10 +198,7 @@ iscsi_co_generic_cb(struct iscsi_context *iscsi, int status,
                return;
            }
        }
-        iTask->err_code = iscsi_translate_sense(&task->sense);
        error_report("iSCSI Failure: %s", iscsi_get_error(iscsi));
-    } else {
-        iTask->iscsilun->force_next_flush |= iTask->force_next_flush;
    }

 out:
@@ -345,33 +261,27 @@ iscsi_set_events(IscsiLun *iscsilun)
    int ev = iscsi_which_events(iscsi);

    if (ev != iscsilun->events) {
-        aio_set_fd_handler(iscsilun->aio_context, iscsi_get_fd(iscsi),
-                           false,
+        aio_set_fd_handler(iscsilun->aio_context,
+                           iscsi_get_fd(iscsi),
                           (ev & POLLIN) ? iscsi_process_read : NULL,
                           (ev & POLLOUT) ? iscsi_process_write : NULL,
                           iscsilun);
        iscsilun->events = ev;
    }
+
+    /* newer versions of libiscsi may return zero events. In this
+     * case start a timer to ensure we are able to return to service
+     * once this situation changes. */
+    if (!ev) {
+        timer_mod(iscsilun->event_timer,
+                  qemu_clock_get_ms(QEMU_CLOCK_REALTIME) + EVENT_INTERVAL);
+    }
 }

-static void iscsi_timed_check_events(void *opaque)
+static void iscsi_timed_set_events(void *opaque)
 {
    IscsiLun *iscsilun = opaque;
-
-    /* check for timed out requests */
-    iscsi_service(iscsilun->iscsi, 0);
-
-    if (iscsilun->request_timed_out) {
-        iscsilun->request_timed_out = false;
-        iscsi_reconnect(iscsilun->iscsi);
-    }
-
-    /* newer versions of libiscsi may return zero events. Ensure we are able
-     * to return to service once this situation changes. */
    iscsi_set_events(iscsilun);
-
-    timer_mod(iscsilun->event_timer,
-              qemu_clock_get_ms(QEMU_CLOCK_REALTIME) + EVENT_INTERVAL);
 }

 static void
@@ -459,7 +369,6 @@ static int coroutine_fn iscsi_co_writev(BlockDriverState *bs,
    struct IscsiTask iTask;
    uint64_t lba;
    uint32_t num_sectors;
-    int fua;

    if (!is_request_lun_aligned(sector_num, nb_sectors, iscsilun)) {
        return -EINVAL;
@@ -475,17 +384,15 @@ static int coroutine_fn iscsi_co_writev(BlockDriverState *bs,
    num_sectors = sector_qemu2lun(nb_sectors, iscsilun);
    iscsi_co_init_iscsitask(iscsilun, &iTask);
 retry:
-    fua = iscsilun->dpofua && !bs->enable_write_cache;
-    iTask.force_next_flush = !fua;
    if (iscsilun->use_16_for_rw) {
        iTask.task = iscsi_write16_task(iscsilun->iscsi, iscsilun->lun, lba,
                                        NULL, num_sectors * iscsilun->block_size,
-                                        iscsilun->block_size, 0, 0, fua, 0, 0,
+                                        iscsilun->block_size, 0, 0, 0, 0, 0,
                                        iscsi_co_generic_cb, &iTask);
    } else {
        iTask.task = iscsi_write10_task(iscsilun->iscsi, iscsilun->lun, lba,
                                        NULL, num_sectors * iscsilun->block_size,
-                                        iscsilun->block_size, 0, 0, fua, 0, 0,
+                                        iscsilun->block_size, 0, 0, 0, 0, 0,
                                        iscsi_co_generic_cb, &iTask);
    }
    if (iTask.task == NULL) {
@@ -509,7 +416,7 @@ retry:
    }

    if (iTask.status != SCSI_STATUS_GOOD) {
-        return iTask.err_code;
+        return -EIO;
    }

    iscsi_allocationmap_set(iscsilun, sector_num, nb_sectors);
@@ -532,8 +439,7 @@ static bool iscsi_allocationmap_is_allocated(IscsiLun *iscsilun,

 static int64_t coroutine_fn iscsi_co_get_block_status(BlockDriverState *bs,
                                                  int64_t sector_num,
-                                                  int nb_sectors, int *pnum,
-                                                  BlockDriverState **file)
+                                                  int nb_sectors, int *pnum)
 {
    IscsiLun *iscsilun = bs->opaque;
    struct scsi_get_lba_status *lbas = NULL;
@@ -554,7 +460,7 @@ static int64_t coroutine_fn iscsi_co_get_block_status(BlockDriverState *bs,
    *pnum = nb_sectors;

    /* LUN does not support logical block provisioning */
-    if (!iscsilun->lbpme) {
+    if (iscsilun->lbpme == 0) {
        goto out;
    }

@@ -625,9 +531,6 @@ out:
    if (iTask.task != NULL) {
        scsi_free_scsi_task(iTask.task);
    }
-    if (ret > 0 && ret & BDRV_BLOCK_OFFSET_VALID) {
-        *file = bs;
-    }
    return ret;
 }

@@ -654,8 +557,7 @@ static int coroutine_fn iscsi_co_readv(BlockDriverState *bs,
        !iscsi_allocationmap_is_allocated(iscsilun, sector_num, nb_sectors)) {
        int64_t ret;
        int pnum;
-        BlockDriverState *file;
-        ret = iscsi_co_get_block_status(bs, sector_num, INT_MAX, &pnum, &file);
+        ret = iscsi_co_get_block_status(bs, sector_num, INT_MAX, &pnum);
        if (ret < 0) {
            return ret;
        }
@@ -703,7 +605,7 @@ retry:
    }

    if (iTask.status != SCSI_STATUS_GOOD) {
-        return iTask.err_code;
+        return -EIO;
    }

    return 0;
@@ -714,12 +616,12 @@ static int coroutine_fn iscsi_co_flush(BlockDriverState *bs)
    IscsiLun *iscsilun = bs->opaque;
    struct IscsiTask iTask;

-    if (!iscsilun->force_next_flush) {
+    if (bs->sg) {
        return 0;
    }
-    iscsilun->force_next_flush = false;

    iscsi_co_init_iscsitask(iscsilun, &iTask);
+
 retry:
    if (iscsi_synchronizecache10_task(iscsilun->iscsi, iscsilun->lun, 0, 0, 0,
                                      0, iscsi_co_generic_cb, &iTask) == NULL) {
@@ -742,7 +644,7 @@ retry:
    }

    if (iTask.status != SCSI_STATUS_GOOD) {
-        return iTask.err_code;
+        return -EIO;
    }

    return 0;
@@ -762,7 +664,7 @@ iscsi_aio_ioctl_cb(struct iscsi_context *iscsi, int status,
    if (status < 0) {
        error_report("Failed to ioctl(SG_IO) to iSCSI lun. %s",
                     iscsi_get_error(iscsi));
-        acb->status = iscsi_translate_sense(&acb->task->sense);
+        acb->status = -EIO;
    }

    acb->ioh->driver_status = 0;
@@ -785,38 +687,6 @@ iscsi_aio_ioctl_cb(struct iscsi_context *iscsi, int status,
    iscsi_schedule_bh(acb);
 }

-static void iscsi_ioctl_bh_completion(void *opaque)
-{
-    IscsiAIOCB *acb = opaque;
-
-    qemu_bh_delete(acb->bh);
-    acb->common.cb(acb->common.opaque, acb->ret);
-    qemu_aio_unref(acb);
-}
-
-static void iscsi_ioctl_handle_emulated(IscsiAIOCB *acb, int req, void *buf)
-{
-    BlockDriverState *bs = acb->common.bs;
-    IscsiLun *iscsilun = bs->opaque;
-    int ret = 0;
-
-    switch (req) {
-    case SG_GET_VERSION_NUM:
-        *(int *)buf = 30000;
-        break;
-    case SG_GET_SCSI_ID:
-        ((struct sg_scsi_id *)buf)->scsi_type = iscsilun->type;
-        break;
-    default:
-        ret = -EINVAL;
-    }
-    assert(!acb->bh);
-    acb->bh = aio_bh_new(bdrv_get_aio_context(bs),
-                         iscsi_ioctl_bh_completion, acb);
-    acb->ret = ret;
-    qemu_bh_schedule(acb->bh);
-}
-
 static BlockAIOCB *iscsi_aio_ioctl(BlockDriverState *bs,
        unsigned long int req, void *buf,
        BlockCompletionFunc *cb, void *opaque)
@@ -826,6 +696,8 @@ static BlockAIOCB *iscsi_aio_ioctl(BlockDriverState *bs,
    struct iscsi_data data;
    IscsiAIOCB *acb;

+    assert(req == SG_IO);
+
    acb = qemu_aio_get(&iscsi_aiocb_info, bs, cb, opaque);

    acb->iscsilun = iscsilun;
@@ -834,11 +706,6 @@ static BlockAIOCB *iscsi_aio_ioctl(BlockDriverState *bs,
    acb->buf         = NULL;
    acb->ioh         = buf;

-    if (req != SG_IO) {
-        iscsi_ioctl_handle_emulated(acb, req, buf);
-        return &acb->common;
-    }
-
    acb->task = malloc(sizeof(struct scsi_task));
    if (acb->task == NULL) {
        error_report("iSCSI: Failed to allocate task for scsi command. %s",
@@ -903,6 +770,38 @@ static BlockAIOCB *iscsi_aio_ioctl(BlockDriverState *bs,
    return &acb->common;
 }

+static void ioctl_cb(void *opaque, int status)
+{
+    int *p_status = opaque;
+    *p_status = status;
+}
+
+static int iscsi_ioctl(BlockDriverState *bs, unsigned long int req, void *buf)
+{
+    IscsiLun *iscsilun = bs->opaque;
+    int status;
+
+    switch (req) {
+    case SG_GET_VERSION_NUM:
+        *(int *)buf = 30000;
+        break;
+    case SG_GET_SCSI_ID:
+        ((struct sg_scsi_id *)buf)->scsi_type = iscsilun->type;
+        break;
+    case SG_IO:
+        status = -EINPROGRESS;
+        iscsi_aio_ioctl(bs, req, buf, ioctl_cb, &status);
+
+        while (status == -EINPROGRESS) {
+            aio_poll(iscsilun->aio_context, true);
+        }
+
+        return 0;
+    default:
+        return -1;
+    }
+    return 0;
+}
 #endif

 static int64_t
@@ -967,7 +866,7 @@ retry:
    }

    if (iTask.status != SCSI_STATUS_GOOD) {
-        return iTask.err_code;
+        return -EIO;
    }

    iscsi_allocationmap_clear(iscsilun, sector_num, nb_sectors);
@@ -1018,7 +917,6 @@ coroutine_fn iscsi_co_write_zeroes(BlockDriverState *bs, int64_t sector_num,
    }

    iscsi_co_init_iscsitask(iscsilun, &iTask);
-    iTask.force_next_flush = true;
 retry:
    if (use_16_for_ws) {
        iTask.task = iscsi_writesame16_task(iscsilun->iscsi, iscsilun->lun, lba,
@@ -1061,7 +959,7 @@ retry:
    }

    if (iTask.status != SCSI_STATUS_GOOD) {
-        return iTask.err_code;
+        return -EIO;
    }

    if (flags & BDRV_REQ_MAY_UNMAP) {
@@ -1182,37 +1080,16 @@ static char *parse_initiator_name(const char *target)
    return iscsi_name;
 }

-static int parse_timeout(const char *target)
-{
-    QemuOptsList *list;
-    QemuOpts *opts;
-    const char *timeout;
-
-    list = qemu_find_opts("iscsi");
-    if (list) {
-        opts = qemu_opts_find(list, target);
-        if (!opts) {
-            opts = QTAILQ_FIRST(&list->head);
-        }
-        if (opts) {
-            timeout = qemu_opt_get(opts, "timeout");
-            if (timeout) {
-                return atoi(timeout);
-            }
-        }
-    }
-
-    return 0;
-}
-
 static void iscsi_nop_timed_event(void *opaque)
 {
    IscsiLun *iscsilun = opaque;

-    if (iscsi_get_nops_in_flight(iscsilun->iscsi) >= MAX_NOP_FAILURES) {
+    if (iscsi_get_nops_in_flight(iscsilun->iscsi) > MAX_NOP_FAILURES) {
        error_report("iSCSI: NOP timeout. Reconnecting...");
-        iscsilun->request_timed_out = true;
-    } else if (iscsi_nop_out_async(iscsilun->iscsi, NULL, NULL, 0, NULL) != 0) {
+        iscsi_reconnect(iscsilun->iscsi);
+    }
+
+    if (iscsi_nop_out_async(iscsilun->iscsi, NULL, NULL, 0, NULL) != 0) {
        error_report("iSCSI: failed to sent NOP-Out. Disabling NOP messages.");
        return;
    }
@@ -1244,17 +1121,12 @@ static void iscsi_readcapacity_sync(IscsiLun *iscsilun, Error **errp)
                } else {
                    iscsilun->block_size = rc16->block_length;
                    iscsilun->num_blocks = rc16->returned_lba + 1;
-                    iscsilun->lbpme = !!rc16->lbpme;
-                    iscsilun->lbprz = !!rc16->lbprz;
+                    iscsilun->lbpme = rc16->lbpme;
+                    iscsilun->lbprz = rc16->lbprz;
                    iscsilun->use_16_for_rw = (rc16->returned_lba > 0xffffffff);
                }
-                break;
            }
-            if (task != NULL && task->status == SCSI_STATUS_CHECK_CONDITION
-                && task->sense.key == SCSI_SENSE_UNIT_ATTENTION) {
-                break;
-            }
-            /* Fall through and try READ CAPACITY(10) instead.  */
+            break;
        case TYPE_ROM:
            task = iscsi_readcapacity10_sync(iscsilun->iscsi, iscsilun->lun, 0, 0);
            if (task != NULL && task->status == SCSI_STATUS_GOOD) {
@@ -1280,11 +1152,7 @@ static void iscsi_readcapacity_sync(IscsiLun *iscsilun, Error **errp)
             && retries-- > 0);

    if (task == NULL || task->status != SCSI_STATUS_GOOD) {
-        error_setg(errp, "iSCSI: failed to send readcapacity10/16 command");
-    } else if (!iscsilun->block_size ||
-               iscsilun->block_size % BDRV_SECTOR_SIZE) {
-        error_setg(errp, "iSCSI: the target returned an invalid "
-                   "block size of %d.", iscsilun->block_size);
+        error_setg(errp, "iSCSI: failed to send readcapacity10 command.");
    }
    if (task) {
        scsi_free_scsi_task(task);
@@ -1347,8 +1215,9 @@ static void iscsi_detach_aio_context(BlockDriverState *bs)
 {
    IscsiLun *iscsilun = bs->opaque;

-    aio_set_fd_handler(iscsilun->aio_context, iscsi_get_fd(iscsilun->iscsi),
-                       false, NULL, NULL, NULL);
+    aio_set_fd_handler(iscsilun->aio_context,
+                       iscsi_get_fd(iscsilun->iscsi),
+                       NULL, NULL, NULL);
    iscsilun->events = 0;

    if (iscsilun->nop_timer) {
@@ -1378,21 +1247,17 @@ static void iscsi_attach_aio_context(BlockDriverState *bs,
    timer_mod(iscsilun->nop_timer,
              qemu_clock_get_ms(QEMU_CLOCK_REALTIME) + NOP_INTERVAL);

-    /* Set up a timer for periodic calls to iscsi_set_events and to
-     * scan for command timeout */
+    /* Prepare a timer for a delayed call to iscsi_set_events */
    iscsilun->event_timer = aio_timer_new(iscsilun->aio_context,
                                          QEMU_CLOCK_REALTIME, SCALE_MS,
-                                          iscsi_timed_check_events, iscsilun);
-    timer_mod(iscsilun->event_timer,
-              qemu_clock_get_ms(QEMU_CLOCK_REALTIME) + EVENT_INTERVAL);
+                                          iscsi_timed_set_events, iscsilun);
 }

-static void iscsi_modesense_sync(IscsiLun *iscsilun)
+static bool iscsi_is_write_protected(IscsiLun *iscsilun)
 {
    struct scsi_task *task;
    struct scsi_mode_sense *ms = NULL;
-    iscsilun->write_protected = false;
-    iscsilun->dpofua = false;
+    bool wrprotected = false;

    task = iscsi_modesense6_sync(iscsilun->iscsi, iscsilun->lun,
                                 1, SCSI_MODESENSE_PC_CURRENT,
@@ -1413,13 +1278,13 @@ static void iscsi_modesense_sync(IscsiLun *iscsilun)
                     iscsi_get_error(iscsilun->iscsi));
        goto out;
    }
-    iscsilun->write_protected = ms->device_specific_parameter & 0x80;
-    iscsilun->dpofua          = ms->device_specific_parameter & 0x10;
+    wrprotected = ms->device_specific_parameter & 0x80;

 out:
    if (task) {
        scsi_free_scsi_task(task);
    }
+    return wrprotected;
 }

 /*
@@ -1439,7 +1304,14 @@ static int iscsi_open(BlockDriverState *bs, QDict *options, int flags,
    QemuOpts *opts;
    Error *local_err = NULL;
    const char *filename;
-    int i, ret = 0, timeout = 0;
+    int i, ret = 0;
+
+    if ((BDRV_SECTOR_SIZE % 512) != 0) {
+        error_setg(errp, "iSCSI: Invalid BDRV_SECTOR_SIZE. "
+                   "BDRV_SECTOR_SIZE(%lld) is not a multiple "
+                   "of 512", BDRV_SECTOR_SIZE);
+        return -EINVAL;
+    }

    opts = qemu_opts_create(&runtime_opts, NULL, 0, &error_abort);
    qemu_opts_absorb_qdict(opts, options, &local_err);
@@ -1509,16 +1381,6 @@ static int iscsi_open(BlockDriverState *bs, QDict *options, int flags,
        goto out;
    }

-    /* timeout handling is broken in libiscsi before 1.15.0 */
-    timeout = parse_timeout(iscsi_url->target);
-#if defined(LIBISCSI_API_VERSION) && LIBISCSI_API_VERSION >= 20150621
-    iscsi_set_timeout(iscsi, timeout);
-#else
-    if (timeout) {
-        error_report("iSCSI: ignoring timeout value for libiscsi <1.15.0");
-    }
-#endif
-
    if (iscsi_full_connect_sync(iscsi, iscsi_url->portal, iscsi_url->lun) != 0) {
        error_setg(errp, "iSCSI: Failed to connect to LUN : %s",
            iscsi_get_error(iscsi));
@@ -1541,8 +1403,7 @@ static int iscsi_open(BlockDriverState *bs, QDict *options, int flags,
    scsi_free_scsi_task(task);
    task = NULL;

-    iscsi_modesense_sync(iscsilun);
-
+    iscsilun->write_protected = iscsi_is_write_protected(iscsilun);
    /* Check the write protect flag of the LUN if we want to write */
    if (iscsilun->type == TYPE_DISK && (flags & BDRV_O_RDWR) &&
        iscsilun->write_protected) {
@@ -1620,7 +1481,7 @@ static int iscsi_open(BlockDriverState *bs, QDict *options, int flags,
        iscsilun->bl.opt_unmap_gran * iscsilun->block_size <= 16 * 1024 * 1024) {
        iscsilun->cluster_sectors = (iscsilun->bl.opt_unmap_gran *
                                     iscsilun->block_size) >> BDRV_SECTOR_BITS;
-        if (iscsilun->lbprz) {
+        if (iscsilun->lbprz && !(bs->open_flags & BDRV_O_NOCACHE)) {
            iscsilun->allocationmap = iscsi_allocationmap_init(iscsilun);
            if (iscsilun->allocationmap == NULL) {
                ret = -ENOMEM;
@@ -1794,7 +1655,7 @@ out:
 static int iscsi_get_info(BlockDriverState *bs, BlockDriverInfo *bdi)
 {
    IscsiLun *iscsilun = bs->opaque;
-    bdi->unallocated_blocks_are_zero = iscsilun->lbprz;
+    bdi->unallocated_blocks_are_zero = !!iscsilun->lbprz;
    bdi->can_write_zeroes_with_unmap = iscsilun->lbprz && iscsilun->lbp.lbpws;
    bdi->cluster_size = iscsilun->cluster_sectors * BDRV_SECTOR_SIZE;
    return 0;
@@ -1838,6 +1699,7 @@ static BlockDriver bdrv_iscsi = {
    .bdrv_co_flush_to_disk = iscsi_co_flush,

 #ifdef __linux__
+    .bdrv_ioctl       = iscsi_ioctl,
    .bdrv_aio_ioctl   = iscsi_aio_ioctl,
 #endif

@@ -1866,10 +1728,6 @@ static QemuOptsList qemu_iscsi_opts = {
            .name = "initiator-name",
            .type = QEMU_OPT_STRING,
            .help = "Initiator iqn name to use when connecting",
-        },{
-            .name = "timeout",
-            .type = QEMU_OPT_NUMBER,
-            .help = "Request timeout in seconds (default 0 = no timeout)",
        },
        { /* end of list */ }
    },
--- a/block/linux-aio.c
+++ b/block/linux-aio.c
@@ -7,7 +7,6 @@
 * This work is licensed under the terms of the GNU GPL, version 2 or later.
 * See the COPYING file in the top-level directory.
 */
-#include "qemu/osdep.h"
 #include "qemu-common.h"
 #include "block/aio.h"
 #include "qemu/queue.h"
@@ -288,7 +287,7 @@ void laio_detach_aio_context(void *s_, AioContext *old_context)
 {
    struct qemu_laio_state *s = s_;

-    aio_set_event_notifier(old_context, &s->e, false, NULL);
+    aio_set_event_notifier(old_context, &s->e, NULL);
    qemu_bh_delete(s->completion_bh);
 }

@@ -297,8 +296,7 @@ void laio_attach_aio_context(void *s_, AioContext *new_context)
    struct qemu_laio_state *s = s_;

    s->completion_bh = aio_bh_new(new_context, qemu_laio_completion_bh, s);
-    aio_set_event_notifier(new_context, &s->e, false,
-                           qemu_laio_completion_cb);
+    aio_set_event_notifier(new_context, &s->e, qemu_laio_completion_cb);
 }

 void *laio_init(void)
--- a/block/mirror.c
+++ b/block/mirror.c
@@ -11,19 +11,14 @@
 *
 */

-#include "qemu/osdep.h"
 #include "trace.h"
 #include "block/blockjob.h"
 #include "block/block_int.h"
-#include "sysemu/block-backend.h"
-#include "qapi/qmp/qerror.h"
 #include "qemu/ratelimit.h"
 #include "qemu/bitmap.h"
-#include "qemu/error-report.h"

 #define SLICE_TIME    100000000ULL /* ns */
 #define MAX_IN_FLIGHT 16
-#define DEFAULT_MIRROR_BUF_SIZE   (10 << 20)

 /* The mirroring buffer is a list of granularity-sized chunks.
 * Free chunks are organized in a list.
@@ -63,7 +58,6 @@ typedef struct MirrorBlockJob {
    int sectors_in_flight;
    int ret;
    bool unmap;
-    bool waiting_for_io;
 } MirrorBlockJob;

 typedef struct MirrorOp {
@@ -116,9 +110,13 @@ static void mirror_iteration_done(MirrorOp *op, int ret)
    }

    qemu_iovec_destroy(&op->qiov);
-    g_free(op);
+    g_slice_free(MirrorOp, op);

-    if (s->waiting_for_io) {
+    /* Enter coroutine when it is not sleeping.  The coroutine sleeps to
+     * rate-limit itself.  The coroutine will eventually resume since there is
+     * a sleep timeout so don't wake it early.
+     */
+    if (s->common.busy) {
        qemu_coroutine_enter(s->common.co, NULL);
    }
 }
@@ -128,9 +126,11 @@ static void mirror_write_complete(void *opaque, int ret)
    MirrorOp *op = opaque;
    MirrorBlockJob *s = op->s;
    if (ret < 0) {
+        BlockDriverState *source = s->common.bs;
        BlockErrorAction action;

-        bdrv_set_dirty_bitmap(s->dirty_bitmap, op->sector_num, op->nb_sectors);
+        bdrv_set_dirty_bitmap(source, s->dirty_bitmap, op->sector_num,
+                              op->nb_sectors);
        action = mirror_error_action(s, false, -ret);
        if (action == BLOCK_ERROR_ACTION_REPORT && s->ret >= 0) {
            s->ret = ret;
@@ -144,9 +144,11 @@ static void mirror_read_complete(void *opaque, int ret)
    MirrorOp *op = opaque;
    MirrorBlockJob *s = op->s;
    if (ret < 0) {
+        BlockDriverState *source = s->common.bs;
        BlockErrorAction action;

-        bdrv_set_dirty_bitmap(s->dirty_bitmap, op->sector_num, op->nb_sectors);
+        bdrv_set_dirty_bitmap(source, s->dirty_bitmap, op->sector_num,
+                              op->nb_sectors);
        action = mirror_error_action(s, true, -ret);
        if (action == BLOCK_ERROR_ACTION_REPORT && s->ret >= 0) {
            s->ret = ret;
@@ -162,21 +164,19 @@ static void mirror_read_complete(void *opaque, int ret)
 static uint64_t coroutine_fn mirror_iteration(MirrorBlockJob *s)
 {
    BlockDriverState *source = s->common.bs;
-    int nb_sectors, sectors_per_chunk, nb_chunks, max_iov;
+    int nb_sectors, sectors_per_chunk, nb_chunks;
    int64_t end, sector_num, next_chunk, next_sector, hbitmap_next_sector;
    uint64_t delay_ns = 0;
    MirrorOp *op;
    int pnum;
    int64_t ret;
-    BlockDriverState *file;
-
-    max_iov = MIN(source->bl.max_iov, s->target->bl.max_iov);

    s->sector_num = hbitmap_iter_next(&s->hbi);
    if (s->sector_num < 0) {
-        bdrv_dirty_iter_init(s->dirty_bitmap, &s->hbi);
+        bdrv_dirty_iter_init(source, s->dirty_bitmap, &s->hbi);
        s->sector_num = hbitmap_iter_next(&s->hbi);
-        trace_mirror_restart_iter(s, bdrv_get_dirty_count(s->dirty_bitmap));
+        trace_mirror_restart_iter(s,
+                                  bdrv_get_dirty_count(source, s->dirty_bitmap));
        assert(s->sector_num >= 0);
    }

@@ -206,9 +206,7 @@ static uint64_t coroutine_fn mirror_iteration(MirrorBlockJob *s)
    /* Wait for I/O to this cluster (from a previous iteration) to be done.  */
    while (test_bit(next_chunk, s->in_flight_bitmap)) {
        trace_mirror_yield_in_flight(s, sector_num, s->in_flight);
-        s->waiting_for_io = true;
        qemu_coroutine_yield();
-        s->waiting_for_io = false;
    }

    do {
@@ -244,18 +242,12 @@ static uint64_t coroutine_fn mirror_iteration(MirrorBlockJob *s)
         */
        while (nb_chunks == 0 && s->buf_free_count < added_chunks) {
            trace_mirror_yield_buf_busy(s, nb_chunks, s->in_flight);
-            s->waiting_for_io = true;
            qemu_coroutine_yield();
-            s->waiting_for_io = false;
        }
        if (s->buf_free_count < nb_chunks + added_chunks) {
            trace_mirror_break_buf_busy(s, nb_chunks, s->in_flight);
            break;
        }
-        if (max_iov < nb_chunks + added_chunks) {
-            trace_mirror_break_iov_max(s, nb_chunks, added_chunks);
-            break;
-        }

        /* We have enough free space to copy these sectors.  */
        bitmap_set(s->in_flight_bitmap, next_chunk, added_chunks);
@@ -270,7 +262,7 @@ static uint64_t coroutine_fn mirror_iteration(MirrorBlockJob *s)
    } while (delay_ns == 0 && next_sector < end);

    /* Allocate a MirrorOp that is used as an AIO callback.  */
-    op = g_new(MirrorOp, 1);
+    op = g_slice_new(MirrorOp);
    op->s = s;
    op->sector_num = sector_num;
    op->nb_sectors = nb_sectors;
@@ -299,7 +291,8 @@ static uint64_t coroutine_fn mirror_iteration(MirrorBlockJob *s)
        next_sector += sectors_per_chunk;
    }

-    bdrv_reset_dirty_bitmap(s->dirty_bitmap, sector_num, nb_sectors);
+    bdrv_reset_dirty_bitmap(source, s->dirty_bitmap, sector_num,
+                            nb_sectors);

    /* Copy the dirty cluster.  */
    s->in_flight++;
@@ -307,7 +300,7 @@ static uint64_t coroutine_fn mirror_iteration(MirrorBlockJob *s)
    trace_mirror_one_iteration(s, sector_num, nb_sectors);

    ret = bdrv_get_block_status_above(source, NULL, sector_num,
-                                      nb_sectors, &pnum, &file);
+                                      nb_sectors, &pnum);
    if (ret < 0 || pnum < nb_sectors ||
            (ret & BDRV_BLOCK_DATA && !(ret & BDRV_BLOCK_ZERO))) {
        bdrv_aio_readv(source, sector_num, &op->qiov, nb_sectors,
@@ -344,9 +337,7 @@ static void mirror_free_init(MirrorBlockJob *s)
 static void mirror_drain(MirrorBlockJob *s)
 {
    while (s->in_flight > 0) {
-        s->waiting_for_io = true;
        qemu_coroutine_yield();
-        s->waiting_for_io = false;
    }
 }

@@ -359,11 +350,6 @@ static void mirror_exit(BlockJob *job, void *opaque)
    MirrorBlockJob *s = container_of(job, MirrorBlockJob, common);
    MirrorExitData *data = opaque;
    AioContext *replace_aio_context = NULL;
-    BlockDriverState *src = s->common.bs;
-
-    /* Make sure that the source BDS doesn't go away before we called
-     * block_job_completed(). */
-    bdrv_ref(src);

    if (s->to_replace) {
        replace_aio_context = bdrv_get_aio_context(s->to_replace);
@@ -375,22 +361,18 @@ static void mirror_exit(BlockJob *job, void *opaque)
        if (s->to_replace) {
            to_replace = s->to_replace;
        }
-
-        /* This was checked in mirror_start_job(), but meanwhile one of the
-         * nodes could have been newly attached to a BlockBackend. */
-        if (to_replace->blk && s->target->blk) {
-            error_report("block job: Can't create node with two BlockBackends");
-            data->ret = -EINVAL;
-            goto out;
-        }
-
        if (bdrv_get_flags(s->target) != bdrv_get_flags(to_replace)) {
            bdrv_reopen(s->target, bdrv_get_flags(to_replace), NULL);
        }
-        bdrv_replace_in_backing_chain(to_replace, s->target);
+        bdrv_swap(s->target, to_replace);
+        if (s->common.driver->job_type == BLOCK_JOB_TYPE_COMMIT) {
+            /* drop the bs loop chain formed by the swap: break the loop then
+             * trigger the unref from the top one */
+            BlockDriverState *p = s->base->backing_hd;
+            bdrv_set_backing_hd(s->base, NULL);
+            bdrv_unref(p);
+        }
    }
-
-out:
    if (s->to_replace) {
        bdrv_op_unblock_all(s->to_replace, s->replace_blocker);
        error_free(s->replace_blocker);
@@ -400,12 +382,9 @@ out:
        aio_context_release(replace_aio_context);
    }
    g_free(s->replaces);
-    bdrv_op_unblock_all(s->target, s->common.blocker);
    bdrv_unref(s->target);
    block_job_completed(&s->common, data->ret);
    g_free(data);
-    bdrv_drained_end(src);
-    bdrv_unref(src);
 }

 static void coroutine_fn mirror_run(void *opaque)
@@ -413,7 +392,7 @@ static void coroutine_fn mirror_run(void *opaque)
    MirrorBlockJob *s = opaque;
    MirrorExitData *data;
    BlockDriverState *bs = s->common.bs;
-    int64_t sector_num, end, length;
+    int64_t sector_num, end, sectors_per_chunk, length;
    uint64_t last_pause_ns;
    BlockDriverInfo bdi;
    char backing_filename[2]; /* we only need 2 characters because we are only
@@ -449,7 +428,7 @@ static void coroutine_fn mirror_run(void *opaque)
     */
    bdrv_get_backing_filename(s->target, backing_filename,
                              sizeof(backing_filename));
-    if (backing_filename[0] && !s->target->backing) {
+    if (backing_filename[0] && !s->target->backing_hd) {
        ret = bdrv_get_info(s->target, &bdi);
        if (ret < 0) {
            goto immediate_exit;
@@ -467,44 +446,33 @@ static void coroutine_fn mirror_run(void *opaque)
        goto immediate_exit;
    }

+    sectors_per_chunk = s->granularity >> BDRV_SECTOR_BITS;
    mirror_free_init(s);

-    last_pause_ns = qemu_clock_get_ns(QEMU_CLOCK_REALTIME);
    if (!s->is_none_mode) {
        /* First part, loop on the sectors and initialize the dirty bitmap.  */
        BlockDriverState *base = s->base;
-        bool mark_all_dirty = s->base == NULL && !bdrv_has_zero_init(s->target);
-
        for (sector_num = 0; sector_num < end; ) {
-            /* Just to make sure we are not exceeding int limit. */
-            int nb_sectors = MIN(INT_MAX >> BDRV_SECTOR_BITS,
-                                 end - sector_num);
-            int64_t now = qemu_clock_get_ns(QEMU_CLOCK_REALTIME);
-
-            if (now - last_pause_ns > SLICE_TIME) {
-                last_pause_ns = now;
-                block_job_sleep_ns(&s->common, QEMU_CLOCK_REALTIME, 0);
-            }
-
-            if (block_job_is_cancelled(&s->common)) {
-                goto immediate_exit;
-            }
-
-            ret = bdrv_is_allocated_above(bs, base, sector_num, nb_sectors, &n);
+            int64_t next = (sector_num | (sectors_per_chunk - 1)) + 1;
+            ret = bdrv_is_allocated_above(bs, base,
+                                          sector_num, next - sector_num, &n);

            if (ret < 0) {
                goto immediate_exit;
            }

            assert(n > 0);
-            if (ret == 1 || mark_all_dirty) {
-                bdrv_set_dirty_bitmap(s->dirty_bitmap, sector_num, n);
+            if (ret == 1) {
+                bdrv_set_dirty_bitmap(bs, s->dirty_bitmap, sector_num, n);
+                sector_num = next;
+            } else {
+                sector_num += n;
            }
-            sector_num += n;
        }
    }

-    bdrv_dirty_iter_init(s->dirty_bitmap, &s->hbi);
+    bdrv_dirty_iter_init(bs, s->dirty_bitmap, &s->hbi);
+    last_pause_ns = qemu_clock_get_ns(QEMU_CLOCK_REALTIME);
    for (;;) {
        uint64_t delay_ns = 0;
        int64_t cnt;
@@ -515,7 +483,7 @@ static void coroutine_fn mirror_run(void *opaque)
            goto immediate_exit;
        }

-        cnt = bdrv_get_dirty_count(s->dirty_bitmap);
+        cnt = bdrv_get_dirty_count(bs, s->dirty_bitmap);
        /* s->common.offset contains the number of bytes already processed so
         * far, cnt is the number of dirty sectors remaining and
         * s->sectors_in_flight is the number of sectors currently being
@@ -524,7 +492,7 @@ static void coroutine_fn mirror_run(void *opaque)
                        (cnt + s->sectors_in_flight) * BDRV_SECTOR_SIZE;

        /* Note that even when no rate limit is applied we need to yield
-         * periodically with no pending I/O so that bdrv_drain_all() returns.
+         * periodically with no pending I/O so that qemu_aio_flush() returns.
         * We do so every SLICE_TIME nanoseconds, or when there is an error,
         * or when the source is clean, whichever comes first.
         */
@@ -533,12 +501,13 @@ static void coroutine_fn mirror_run(void *opaque)
            if (s->in_flight == MAX_IN_FLIGHT || s->buf_free_count == 0 ||
                (cnt == 0 && s->in_flight > 0)) {
                trace_mirror_yield(s, s->in_flight, s->buf_free_count, cnt);
-                s->waiting_for_io = true;
                qemu_coroutine_yield();
-                s->waiting_for_io = false;
                continue;
            } else if (cnt != 0) {
                delay_ns = mirror_iteration(s);
+                if (delay_ns == 0) {
+                    continue;
+                }
            }
        }

@@ -564,7 +533,7 @@ static void coroutine_fn mirror_run(void *opaque)

                should_complete = s->should_complete ||
                    block_job_is_cancelled(&s->common);
-                cnt = bdrv_get_dirty_count(s->dirty_bitmap);
+                cnt = bdrv_get_dirty_count(bs, s->dirty_bitmap);
            }
        }

@@ -579,7 +548,7 @@ static void coroutine_fn mirror_run(void *opaque)
             */
            trace_mirror_before_drain(s, cnt);
            bdrv_drain(bs);
-            cnt = bdrv_get_dirty_count(s->dirty_bitmap);
+            cnt = bdrv_get_dirty_count(bs, s->dirty_bitmap);
        }

        ret = 0;
@@ -618,15 +587,10 @@ immediate_exit:
    g_free(s->cow_bitmap);
    g_free(s->in_flight_bitmap);
    bdrv_release_dirty_bitmap(bs, s->dirty_bitmap);
-    if (s->target->blk) {
-        blk_iostatus_disable(s->target->blk);
-    }
+    bdrv_iostatus_disable(s->target);

    data = g_malloc(sizeof(*data));
    data->ret = ret;
-    /* Before we switch to target in mirror_exit, make sure data doesn't
-     * change. */
-    bdrv_drained_begin(s->common.bs);
    block_job_defer_to_main_loop(&s->common, mirror_exit, data);
 }

@@ -635,7 +599,7 @@ static void mirror_set_speed(BlockJob *job, int64_t speed, Error **errp)
    MirrorBlockJob *s = container_of(job, MirrorBlockJob, common);

    if (speed < 0) {
-        error_setg(errp, QERR_INVALID_PARAMETER, "speed");
+        error_set(errp, QERR_INVALID_PARAMETER, "speed");
        return;
    }
    ratelimit_set_speed(&s->limit, speed / BDRV_SECTOR_SIZE, SLICE_TIME);
@@ -645,9 +609,7 @@ static void mirror_iostatus_reset(BlockJob *job)
 {
    MirrorBlockJob *s = container_of(job, MirrorBlockJob, common);

-    if (s->target->blk) {
-        blk_iostatus_reset(s->target->blk);
-    }
+    bdrv_iostatus_reset(s->target);
 }

 static void mirror_complete(BlockJob *job, Error **errp)
@@ -656,13 +618,14 @@ static void mirror_complete(BlockJob *job, Error **errp)
    Error *local_err = NULL;
    int ret;

-    ret = bdrv_open_backing_file(s->target, NULL, "backing", &local_err);
+    ret = bdrv_open_backing_file(s->target, NULL, &local_err);
    if (ret < 0) {
        error_propagate(errp, local_err);
        return;
    }
    if (!s->synced) {
-        error_setg(errp, QERR_BLOCK_JOB_NOT_READY, job->id);
+        error_set(errp, QERR_BLOCK_JOB_NOT_READY,
+                  bdrv_get_device_name(job->bs));
        return;
    }

@@ -670,9 +633,9 @@ static void mirror_complete(BlockJob *job, Error **errp)
    if (s->replaces) {
        AioContext *replace_aio_context;

-        s->to_replace = bdrv_find_node(s->replaces);
+        s->to_replace = check_to_replace_node(s->replaces, &local_err);
        if (!s->to_replace) {
-            error_setg(errp, "Node name '%s' not found", s->replaces);
+            error_propagate(errp, local_err);
            return;
        }

@@ -688,7 +651,7 @@ static void mirror_complete(BlockJob *job, Error **errp)
    }

    s->should_complete = true;
-    block_job_enter(&s->common);
+    block_job_resume(job);
 }

 static const BlockJobDriver mirror_job_driver = {
@@ -710,7 +673,7 @@ static const BlockJobDriver commit_active_job_driver = {

 static void mirror_start_job(BlockDriverState *bs, BlockDriverState *target,
                             const char *replaces,
-                             int64_t speed, uint32_t granularity,
+                             int64_t speed, int64_t granularity,
                             int64_t buf_size,
                             BlockdevOnError on_source_error,
                             BlockdevOnError on_target_error,
@@ -721,44 +684,28 @@ static void mirror_start_job(BlockDriverState *bs, BlockDriverState *target,
                             bool is_none_mode, BlockDriverState *base)
 {
    MirrorBlockJob *s;
-    BlockDriverState *replaced_bs;

    if (granularity == 0) {
-        granularity = bdrv_get_default_bitmap_granularity(target);
+        /* Choose the default granularity based on the target file's cluster
+         * size, clamped between 4k and 64k.  */
+        BlockDriverInfo bdi;
+        if (bdrv_get_info(target, &bdi) >= 0 && bdi.cluster_size != 0) {
+            granularity = MAX(4096, bdi.cluster_size);
+            granularity = MIN(65536, granularity);
+        } else {
+            granularity = 65536;
+        }
    }

    assert ((granularity & (granularity - 1)) == 0);

    if ((on_source_error == BLOCKDEV_ON_ERROR_STOP ||
         on_source_error == BLOCKDEV_ON_ERROR_ENOSPC) &&
-        (!bs->blk || !blk_iostatus_is_enabled(bs->blk))) {
-        error_setg(errp, QERR_INVALID_PARAMETER, "on-source-error");
+        !bdrv_iostatus_is_enabled(bs)) {
+        error_set(errp, QERR_INVALID_PARAMETER, "on-source-error");
        return;
    }

-    if (buf_size < 0) {
-        error_setg(errp, "Invalid parameter 'buf-size'");
-        return;
-    }
-
-    if (buf_size == 0) {
-        buf_size = DEFAULT_MIRROR_BUF_SIZE;
-    }
-
-    /* We can't support this case as long as the block layer can't handle
-     * multiple BlockBackends per BlockDriverState. */
-    if (replaces) {
-        replaced_bs = bdrv_lookup_bs(replaces, replaces, errp);
-        if (replaced_bs == NULL) {
-            return;
-        }
-    } else {
-        replaced_bs = bs;
-    }
-    if (replaced_bs->blk && target->blk) {
-        error_setg(errp, "Can't create node with two BlockBackends");
-        return;
-    }

    s = block_job_create(driver, bs, speed, cb, opaque, errp);
    if (!s) {
@@ -772,23 +719,16 @@ static void mirror_start_job(BlockDriverState *bs, BlockDriverState *target,
    s->is_none_mode = is_none_mode;
    s->base = base;
    s->granularity = granularity;
-    s->buf_size = ROUND_UP(buf_size, granularity);
+    s->buf_size = MAX(buf_size, granularity);
    s->unmap = unmap;

-    s->dirty_bitmap = bdrv_create_dirty_bitmap(bs, granularity, NULL, errp);
+    s->dirty_bitmap = bdrv_create_dirty_bitmap(bs, granularity, errp);
    if (!s->dirty_bitmap) {
-        g_free(s->replaces);
-        block_job_unref(&s->common);
        return;
    }
-
-    bdrv_op_block_all(s->target, s->common.blocker);
-
    bdrv_set_enable_write_cache(s->target, true);
-    if (s->target->blk) {
-        blk_set_on_error(s->target->blk, on_target_error, on_target_error);
-        blk_iostatus_enable(s->target->blk);
-    }
+    bdrv_set_on_error(s->target, on_target_error, on_target_error);
+    bdrv_iostatus_enable(s->target);
    s->common.co = qemu_coroutine_create(mirror_run);
    trace_mirror_start(bs, s, s->common.co, opaque);
    qemu_coroutine_enter(s->common.co, s);
@@ -796,7 +736,7 @@ static void mirror_start_job(BlockDriverState *bs, BlockDriverState *target,

 void mirror_start(BlockDriverState *bs, BlockDriverState *target,
                  const char *replaces,
-                  int64_t speed, uint32_t granularity, int64_t buf_size,
+                  int64_t speed, int64_t granularity, int64_t buf_size,
                  MirrorSyncMode mode, BlockdevOnError on_source_error,
                  BlockdevOnError on_target_error,
                  bool unmap,
@@ -806,12 +746,8 @@ void mirror_start(BlockDriverState *bs, BlockDriverState *target,
    bool is_none_mode;
    BlockDriverState *base;

-    if (mode == MIRROR_SYNC_MODE_INCREMENTAL) {
-        error_setg(errp, "Sync mode 'incremental' not supported");
-        return;
-    }
    is_none_mode = mode == MIRROR_SYNC_MODE_NONE;
-    base = mode == MIRROR_SYNC_MODE_TOP ? backing_bs(bs) : NULL;
+    base = mode == MIRROR_SYNC_MODE_TOP ? bs->backing_hd : NULL;
    mirror_start_job(bs, target, replaces,
                     speed, granularity, buf_size,
                     on_source_error, on_target_error, unmap, cb, opaque, errp,
--- a/block/nbd-client.c
+++ b/block/nbd-client.c
@@ -26,8 +26,8 @@
 * THE SOFTWARE.
 */

-#include "qemu/osdep.h"
 #include "nbd-client.h"
+#include "qemu/sockets.h"

 #define HANDLE_TO_INDEX(bs, handle) ((handle) ^ ((uint64_t)(intptr_t)bs))
 #define INDEX_TO_HANDLE(bs, index)  ((index)  ^ ((uint64_t)(intptr_t)bs))
@@ -47,21 +47,13 @@ static void nbd_teardown_connection(BlockDriverState *bs)
 {
    NbdClientSession *client = nbd_get_client_session(bs);

-    if (!client->ioc) { /* Already closed */
-        return;
-    }
-
    /* finish any pending coroutines */
-    qio_channel_shutdown(client->ioc,
-                         QIO_CHANNEL_SHUTDOWN_BOTH,
-                         NULL);
+    shutdown(client->sock, 2);
    nbd_recv_coroutines_enter_all(client);

    nbd_client_detach_aio_context(bs);
-    object_unref(OBJECT(client->sioc));
-    client->sioc = NULL;
-    object_unref(OBJECT(client->ioc));
-    client->ioc = NULL;
+    closesocket(client->sock);
+    client->sock = -1;
 }

 static void nbd_reply_ready(void *opaque)
@@ -71,16 +63,12 @@ static void nbd_reply_ready(void *opaque)
    uint64_t i;
    int ret;

-    if (!s->ioc) { /* Already closed */
-        return;
-    }
-
    if (s->reply.handle == 0) {
        /* No reply already in flight.  Fetch a header.  It is possible
         * that another thread has done the same thing in parallel, so
         * the socket is not readable anymore.
         */
-        ret = nbd_receive_reply(s->ioc, &s->reply);
+        ret = nbd_receive_reply(s->sock, &s->reply);
        if (ret == -EAGAIN) {
            return;
        }
@@ -131,36 +119,32 @@ static int nbd_co_send_request(BlockDriverState *bs,
        }
    }

-    g_assert(qemu_in_coroutine());
    assert(i < MAX_NBD_REQUESTS);
    request->handle = INDEX_TO_HANDLE(s, i);
-
-    if (!s->ioc) {
-        qemu_co_mutex_unlock(&s->send_mutex);
-        return -EPIPE;
-    }
-
    s->send_coroutine = qemu_coroutine_self();
    aio_context = bdrv_get_aio_context(bs);

-    aio_set_fd_handler(aio_context, s->sioc->fd, false,
+    aio_set_fd_handler(aio_context, s->sock,
                       nbd_reply_ready, nbd_restart_write, bs);
    if (qiov) {
-        qio_channel_set_cork(s->ioc, true);
-        rc = nbd_send_request(s->ioc, request);
+        if (!s->is_unix) {
+            socket_set_cork(s->sock, 1);
+        }
+        rc = nbd_send_request(s->sock, request);
        if (rc >= 0) {
-            ret = nbd_wr_syncv(s->ioc, qiov->iov, qiov->niov,
-                               offset, request->len, 0);
+            ret = qemu_co_sendv(s->sock, qiov->iov, qiov->niov,
+                                offset, request->len);
            if (ret != request->len) {
                rc = -EIO;
            }
        }
-        qio_channel_set_cork(s->ioc, false);
+        if (!s->is_unix) {
+            socket_set_cork(s->sock, 0);
+        }
    } else {
-        rc = nbd_send_request(s->ioc, request);
+        rc = nbd_send_request(s->sock, request);
    }
-    aio_set_fd_handler(aio_context, s->sioc->fd, false,
-                       nbd_reply_ready, NULL, bs);
+    aio_set_fd_handler(aio_context, s->sock, nbd_reply_ready, NULL, bs);
    s->send_coroutine = NULL;
    qemu_co_mutex_unlock(&s->send_mutex);
    return rc;
@@ -176,13 +160,12 @@ static void nbd_co_receive_reply(NbdClientSession *s,
     * peek at the next reply and avoid yielding if it's ours?  */
    qemu_coroutine_yield();
    *reply = s->reply;
-    if (reply->handle != request->handle ||
-        !s->ioc) {
+    if (reply->handle != request->handle) {
        reply->error = EIO;
    } else {
        if (qiov && reply->error == 0) {
-            ret = nbd_wr_syncv(s->ioc, qiov->iov, qiov->niov,
-                               offset, request->len, 1);
+            ret = qemu_co_recvv(s->sock, qiov->iov, qiov->niov,
+                                offset, request->len);
            if (ret != request->len) {
                reply->error = EIO;
            }
@@ -365,15 +348,14 @@ int nbd_client_co_discard(BlockDriverState *bs, int64_t sector_num,
 void nbd_client_detach_aio_context(BlockDriverState *bs)
 {
    aio_set_fd_handler(bdrv_get_aio_context(bs),
-                       nbd_get_client_session(bs)->sioc->fd,
-                       false, NULL, NULL, NULL);
+                       nbd_get_client_session(bs)->sock, NULL, NULL, NULL);
 }

 void nbd_client_attach_aio_context(BlockDriverState *bs,
                                   AioContext *new_context)
 {
-    aio_set_fd_handler(new_context, nbd_get_client_session(bs)->sioc->fd,
-                       false, nbd_reply_ready, NULL, bs);
+    aio_set_fd_handler(new_context, nbd_get_client_session(bs)->sock,
+                       nbd_reply_ready, NULL, bs);
 }

 void nbd_client_close(BlockDriverState *bs)
@@ -385,20 +367,16 @@ void nbd_client_close(BlockDriverState *bs)
        .len = 0
    };

-    if (client->ioc == NULL) {
+    if (client->sock == -1) {
        return;
    }

-    nbd_send_request(client->ioc, &request);
+    nbd_send_request(client->sock, &request);

    nbd_teardown_connection(bs);
 }

-int nbd_client_init(BlockDriverState *bs,
-                    QIOChannelSocket *sioc,
-                    const char *export,
-                    QCryptoTLSCreds *tlscreds,
-                    const char *hostname,
+int nbd_client_init(BlockDriverState *bs, int sock, const char *export,
                    Error **errp)
 {
    NbdClientSession *client = nbd_get_client_session(bs);
@@ -406,32 +384,22 @@ int nbd_client_init(BlockDriverState *bs,

    /* NBD handshake */
    logout("session init %s\n", export);
-    qio_channel_set_blocking(QIO_CHANNEL(sioc), true, NULL);
-
-    ret = nbd_receive_negotiate(QIO_CHANNEL(sioc), export,
-                                &client->nbdflags,
-                                tlscreds, hostname,
-                                &client->ioc,
-                                &client->size, errp);
+    qemu_set_block(sock);
+    ret = nbd_receive_negotiate(sock, export,
+                                &client->nbdflags, &client->size, errp);
    if (ret < 0) {
        logout("Failed to negotiate with the NBD server\n");
+        closesocket(sock);
        return ret;
    }

    qemu_co_mutex_init(&client->send_mutex);
    qemu_co_mutex_init(&client->free_sema);
-    client->sioc = sioc;
-    object_ref(OBJECT(client->sioc));
-
-    if (!client->ioc) {
-        client->ioc = QIO_CHANNEL(sioc);
-        object_ref(OBJECT(client->ioc));
-    }
+    client->sock = sock;

    /* Now that we're connected, set the socket to be non-blocking and
     * kick the reply mechanism.  */
-    qio_channel_set_blocking(QIO_CHANNEL(sioc), false, NULL);
-
+    qemu_set_nonblock(sock);
    nbd_client_attach_aio_context(bs, bdrv_get_aio_context(bs));

    logout("Established connection with NBD server\n");
--- a/block/nbd-client.h
+++ b/block/nbd-client.h
@@ -4,7 +4,6 @@
 #include "qemu-common.h"
 #include "block/nbd.h"
 #include "block/block_int.h"
-#include "io/channel-socket.h"

 /* #define DEBUG_NBD */

@@ -18,8 +17,7 @@
 #define MAX_NBD_REQUESTS    16

 typedef struct NbdClientSession {
-    QIOChannelSocket *sioc; /* The master data channel */
-    QIOChannel *ioc; /* The current I/O channel which may differ (eg TLS) */
+    int sock;
    uint32_t nbdflags;
    off_t size;

@@ -36,11 +34,7 @@ typedef struct NbdClientSession {

 NbdClientSession *nbd_get_client_session(BlockDriverState *bs);

-int nbd_client_init(BlockDriverState *bs,
-                    QIOChannelSocket *sock,
-                    const char *export_name,
-                    QCryptoTLSCreds *tlscreds,
-                    const char *hostname,
+int nbd_client_init(BlockDriverState *bs, int sock, const char *export_name,
                    Error **errp);
 void nbd_client_close(BlockDriverState *bs);

--- a/block/nbd.c
+++ b/block/nbd.c
@@ -26,21 +26,24 @@
 * THE SOFTWARE.
 */

-#include "qemu/osdep.h"
 #include "block/nbd-client.h"
 #include "qemu/uri.h"
 #include "block/block_int.h"
 #include "qemu/module.h"
+#include "qemu/sockets.h"
 #include "qapi/qmp/qdict.h"
 #include "qapi/qmp/qjson.h"
 #include "qapi/qmp/qint.h"
 #include "qapi/qmp/qstring.h"

+#include <sys/types.h>
+#include <unistd.h>

 #define EN_OPTSTR ":exportname="

 typedef struct BDRVNBDState {
    NbdClientSession client;
+    QemuOpts *socket_opts;
 } BDRVNBDState;

 static int nbd_parse_uri(const char *filename, QDict *options)
@@ -187,10 +190,10 @@ out:
    g_free(file);
 }

-static SocketAddress *nbd_config(BDRVNBDState *s, QDict *options, char **export,
-                                 Error **errp)
+static void nbd_config(BDRVNBDState *s, QDict *options, char **export,
+                       Error **errp)
 {
-    SocketAddress *saddr;
+    Error *local_err = NULL;

    if (qdict_haskey(options, "path") == qdict_haskey(options, "host")) {
        if (qdict_haskey(options, "path")) {
@@ -198,37 +201,28 @@ static SocketAddress *nbd_config(BDRVNBDState *s, QDict *options, char **export,
        } else {
            error_setg(errp, "one of path and host must be specified.");
        }
-        return NULL;
+        return;
    }

-    saddr = g_new0(SocketAddress, 1);
+    s->client.is_unix = qdict_haskey(options, "path");
+    s->socket_opts = qemu_opts_create(&socket_optslist, NULL, 0,
+                                      &error_abort);

-    if (qdict_haskey(options, "path")) {
-        saddr->type = SOCKET_ADDRESS_KIND_UNIX;
-        saddr->u.q_unix = g_new0(UnixSocketAddress, 1);
-        saddr->u.q_unix->path = g_strdup(qdict_get_str(options, "path"));
-        qdict_del(options, "path");
-    } else {
-        saddr->type = SOCKET_ADDRESS_KIND_INET;
-        saddr->u.inet = g_new0(InetSocketAddress, 1);
-        saddr->u.inet->host = g_strdup(qdict_get_str(options, "host"));
-        if (!qdict_get_try_str(options, "port")) {
-            saddr->u.inet->port = g_strdup_printf("%d", NBD_DEFAULT_PORT);
-        } else {
-            saddr->u.inet->port = g_strdup(qdict_get_str(options, "port"));
-        }
-        qdict_del(options, "host");
-        qdict_del(options, "port");
+    qemu_opts_absorb_qdict(s->socket_opts, options, &local_err);
+    if (local_err) {
+        error_propagate(errp, local_err);
+        return;
    }

-    s->client.is_unix = saddr->type == SOCKET_ADDRESS_KIND_UNIX;
+    if (!qemu_opt_get(s->socket_opts, "port")) {
+        qemu_opt_set_number(s->socket_opts, "port", NBD_DEFAULT_PORT,
+                            &error_abort);
+    }

    *export = g_strdup(qdict_get_try_str(options, "export"));
    if (*export) {
        qdict_del(options, "export");
    }
-
-    return saddr;
 }

 NbdClientSession *nbd_get_client_session(BlockDriverState *bs)
@@ -237,113 +231,57 @@ NbdClientSession *nbd_get_client_session(BlockDriverState *bs)
    return &s->client;
 }

-static QIOChannelSocket *nbd_establish_connection(SocketAddress *saddr,
-                                                  Error **errp)
+static int nbd_establish_connection(BlockDriverState *bs, Error **errp)
 {
-    QIOChannelSocket *sioc;
-    Error *local_err = NULL;
+    BDRVNBDState *s = bs->opaque;
+    int sock;

-    sioc = qio_channel_socket_new();
-
-    qio_channel_socket_connect_sync(sioc,
-                                    saddr,
-                                    &local_err);
-    if (local_err) {
-        error_propagate(errp, local_err);
-        return NULL;
+    if (s->client.is_unix) {
+        sock = unix_connect_opts(s->socket_opts, errp, NULL, NULL);
+    } else {
+        sock = inet_connect_opts(s->socket_opts, errp, NULL, NULL);
+        if (sock >= 0) {
+            socket_set_nodelay(sock);
+        }
    }

-    qio_channel_set_delay(QIO_CHANNEL(sioc), false);
+    /* Failed to establish connection */
+    if (sock < 0) {
+        logout("Failed to establish connection to NBD server\n");
+        return -EIO;
+    }

-    return sioc;
+    return sock;
 }

-
-static QCryptoTLSCreds *nbd_get_tls_creds(const char *id, Error **errp)
-{
-    Object *obj;
-    QCryptoTLSCreds *creds;
-
-    obj = object_resolve_path_component(
-        object_get_objects_root(), id);
-    if (!obj) {
-        error_setg(errp, "No TLS credentials with id '%s'",
-                   id);
-        return NULL;
-    }
-    creds = (QCryptoTLSCreds *)
-        object_dynamic_cast(obj, TYPE_QCRYPTO_TLS_CREDS);
-    if (!creds) {
-        error_setg(errp, "Object with id '%s' is not TLS credentials",
-                   id);
-        return NULL;
-    }
-
-    if (creds->endpoint != QCRYPTO_TLS_CREDS_ENDPOINT_CLIENT) {
-        error_setg(errp,
-                   "Expecting TLS credentials with a client endpoint");
-        return NULL;
-    }
-    object_ref(obj);
-    return creds;
-}
-
-
 static int nbd_open(BlockDriverState *bs, QDict *options, int flags,
                    Error **errp)
 {
    BDRVNBDState *s = bs->opaque;
    char *export = NULL;
-    QIOChannelSocket *sioc = NULL;
-    SocketAddress *saddr;
-    const char *tlscredsid;
-    QCryptoTLSCreds *tlscreds = NULL;
-    const char *hostname = NULL;
-    int ret = -EINVAL;
+    int result, sock;
+    Error *local_err = NULL;

    /* Pop the config into our state object. Exit if invalid. */
-    saddr = nbd_config(s, options, &export, errp);
-    if (!saddr) {
-        goto error;
-    }
-
-    tlscredsid = g_strdup(qdict_get_try_str(options, "tls-creds"));
-    if (tlscredsid) {
-        qdict_del(options, "tls-creds");
-        tlscreds = nbd_get_tls_creds(tlscredsid, errp);
-        if (!tlscreds) {
-            goto error;
-        }
-
-        if (saddr->type != SOCKET_ADDRESS_KIND_INET) {
-            error_setg(errp, "TLS only supported over IP sockets");
-            goto error;
-        }
-        hostname = saddr->u.inet->host;
+    nbd_config(s, options, &export, &local_err);
+    if (local_err) {
+        error_propagate(errp, local_err);
+        return -EINVAL;
    }

    /* establish TCP connection, return error if it fails
     * TODO: Configurable retry-until-timeout behaviour.
     */
-    sioc = nbd_establish_connection(saddr, errp);
-    if (!sioc) {
-        ret = -ECONNREFUSED;
-        goto error;
+    sock = nbd_establish_connection(bs, errp);
+    if (sock < 0) {
+        g_free(export);
+        return sock;
    }

    /* NBD handshake */
-    ret = nbd_client_init(bs, sioc, export,
-                          tlscreds, hostname, errp);
- error:
-    if (sioc) {
-        object_unref(OBJECT(sioc));
-    }
-    if (tlscreds) {
-        object_unref(OBJECT(tlscreds));
-    }
-    qapi_free_SocketAddress(saddr);
+    result = nbd_client_init(bs, sock, export, errp);
    g_free(export);
-    return ret;
+    return result;
 }

 static int nbd_co_readv(BlockDriverState *bs, int64_t sector_num,
@@ -377,6 +315,9 @@ static int nbd_co_discard(BlockDriverState *bs, int64_t sector_num,

 static void nbd_close(BlockDriverState *bs)
 {
+    BDRVNBDState *s = bs->opaque;
+
+    qemu_opts_del(s->socket_opts);
    nbd_client_close(bs);
 }

@@ -398,14 +339,13 @@ static void nbd_attach_aio_context(BlockDriverState *bs,
    nbd_client_attach_aio_context(bs, new_context);
 }

-static void nbd_refresh_filename(BlockDriverState *bs, QDict *options)
+static void nbd_refresh_filename(BlockDriverState *bs)
 {
    QDict *opts = qdict_new();
-    const char *path   = qdict_get_try_str(options, "path");
-    const char *host   = qdict_get_try_str(options, "host");
-    const char *port   = qdict_get_try_str(options, "port");
-    const char *export = qdict_get_try_str(options, "export");
-    const char *tlscreds = qdict_get_try_str(options, "tls-creds");
+    const char *path   = qdict_get_try_str(bs->options, "path");
+    const char *host   = qdict_get_try_str(bs->options, "host");
+    const char *port   = qdict_get_try_str(bs->options, "port");
+    const char *export = qdict_get_try_str(bs->options, "export");

    qdict_put_obj(opts, "driver", QOBJECT(qstring_from_str("nbd")));

@@ -440,9 +380,6 @@ static void nbd_refresh_filename(BlockDriverState *bs, QDict *options)
    if (export) {
        qdict_put_obj(opts, "export", QOBJECT(qstring_from_str(export)));
    }
-    if (tlscreds) {
-        qdict_put_obj(opts, "tls-creds", QOBJECT(qstring_from_str(tlscreds)));
-    }

    bs->full_open_options = opts;
 }
--- a/block/nfs.c
+++ b/block/nfs.c
@@ -22,7 +22,7 @@
 * THE SOFTWARE.
 */

-#include "qemu/osdep.h"
+#include "config-host.h"

 #include <poll.h>
 #include "qemu-common.h"
@@ -43,7 +43,6 @@ typedef struct NFSClient {
    int events;
    bool has_zero_init;
    AioContext *aio_context;
-    blkcnt_t st_blocks;
 } NFSClient;

 typedef struct NFSRPC {
@@ -63,10 +62,11 @@ static void nfs_set_events(NFSClient *client)
 {
    int ev = nfs_which_events(client->context);
    if (ev != client->events) {
-        aio_set_fd_handler(client->aio_context, nfs_get_fd(client->context),
-                           false,
+        aio_set_fd_handler(client->aio_context,
+                           nfs_get_fd(client->context),
                           (ev & POLLIN) ? nfs_process_read : NULL,
-                           (ev & POLLOUT) ? nfs_process_write : NULL, client);
+                           (ev & POLLOUT) ? nfs_process_write : NULL,
+                           client);

    }
    client->events = ev;
@@ -241,8 +241,9 @@ static void nfs_detach_aio_context(BlockDriverState *bs)
 {
    NFSClient *client = bs->opaque;

-    aio_set_fd_handler(client->aio_context, nfs_get_fd(client->context),
-                       false, NULL, NULL, NULL);
+    aio_set_fd_handler(client->aio_context,
+                       nfs_get_fd(client->context),
+                       NULL, NULL, NULL);
    client->events = 0;
 }

@@ -261,8 +262,9 @@ static void nfs_client_close(NFSClient *client)
        if (client->fh) {
            nfs_close(client->context, client->fh);
        }
-        aio_set_fd_handler(client->aio_context, nfs_get_fd(client->context),
-                           false, NULL, NULL, NULL);
+        aio_set_fd_handler(client->aio_context,
+                           nfs_get_fd(client->context),
+                           NULL, NULL, NULL);
        nfs_destroy_context(client->context);
    }
    memset(client, 0, sizeof(NFSClient));
@@ -372,7 +374,6 @@ static int64_t nfs_client_open(NFSClient *client, const char *filename,
    }

    ret = DIV_ROUND_UP(st.st_size, BDRV_SECTOR_SIZE);
-    client->st_blocks = st.st_blocks;
    client->has_zero_init = S_ISREG(st.st_mode);
    goto out;
 fail:
@@ -463,11 +464,6 @@ static int64_t nfs_get_allocated_file_size(BlockDriverState *bs)
    NFSRPC task = {0};
    struct stat st;

-    if (bdrv_is_read_only(bs) &&
-        !(bs->open_flags & BDRV_O_NOCACHE)) {
-        return client->st_blocks * 512;
-    }
-
    task.st = &st;
    if (nfs_fstat_async(client->context, client->fh, nfs_co_generic_cb,
                        &task) != 0) {
@@ -479,7 +475,7 @@ static int64_t nfs_get_allocated_file_size(BlockDriverState *bs)
        aio_poll(client->aio_context, true);
    }

-    return (task.ret < 0 ? task.ret : st.st_blocks * 512);
+    return (task.ret < 0 ? task.ret : st.st_blocks * st.st_blksize);
 }

 static int nfs_file_truncate(BlockDriverState *bs, int64_t offset)
@@ -488,34 +484,6 @@ static int nfs_file_truncate(BlockDriverState *bs, int64_t offset)
    return nfs_ftruncate(client->context, client->fh, offset);
 }

-/* Note that this will not re-establish a connection with the NFS server
- * - it is effectively a NOP.  */
-static int nfs_reopen_prepare(BDRVReopenState *state,
-                              BlockReopenQueue *queue, Error **errp)
-{
-    NFSClient *client = state->bs->opaque;
-    struct stat st;
-    int ret = 0;
-
-    if (state->flags & BDRV_O_RDWR && bdrv_is_read_only(state->bs)) {
-        error_setg(errp, "Cannot open a read-only mount as read-write");
-        return -EACCES;
-    }
-
-    /* Update cache for read-only reopens */
-    if (!(state->flags & BDRV_O_RDWR)) {
-        ret = nfs_fstat(client->context, client->fh, &st);
-        if (ret < 0) {
-            error_setg(errp, "Failed to fstat file: %s",
-                       nfs_get_error(client->context));
-            return ret;
-        }
-        client->st_blocks = st.st_blocks;
-    }
-
-    return 0;
-}
-
 static BlockDriver bdrv_nfs = {
    .format_name                    = "nfs",
    .protocol_name                  = "nfs",
@@ -531,7 +499,6 @@ static BlockDriver bdrv_nfs = {
    .bdrv_file_open                 = nfs_file_open,
    .bdrv_close                     = nfs_file_close,
    .bdrv_create                    = nfs_file_create,
-    .bdrv_reopen_prepare            = nfs_reopen_prepare,

    .bdrv_co_readv                  = nfs_co_readv,
    .bdrv_co_writev                 = nfs_co_writev,
--- a/block/null.c
+++ b/block/null.c
@@ -10,14 +10,10 @@
 * See the COPYING file in the top-level directory.
 */

-#include "qemu/osdep.h"
 #include "block/block_int.h"

-#define NULL_OPT_LATENCY "latency-ns"
-
 typedef struct {
    int64_t length;
-    int64_t latency_ns;
 } BDRVNullState;

 static QemuOptsList runtime_opts = {
@@ -34,12 +30,6 @@ static QemuOptsList runtime_opts = {
            .type = QEMU_OPT_SIZE,
            .help = "size of the null block",
        },
-        {
-            .name = NULL_OPT_LATENCY,
-            .type = QEMU_OPT_NUMBER,
-            .help = "nanoseconds (approximated) to wait "
-                    "before completing request",
-        },
        { /* end of list */ }
    },
 };
@@ -49,20 +39,13 @@ static int null_file_open(BlockDriverState *bs, QDict *options, int flags,
 {
    QemuOpts *opts;
    BDRVNullState *s = bs->opaque;
-    int ret = 0;

    opts = qemu_opts_create(&runtime_opts, NULL, 0, &error_abort);
    qemu_opts_absorb_qdict(opts, options, &error_abort);
    s->length =
        qemu_opt_get_size(opts, BLOCK_OPT_SIZE, 1 << 30);
-    s->latency_ns =
-        qemu_opt_get_number(opts, NULL_OPT_LATENCY, 0);
-    if (s->latency_ns < 0) {
-        error_setg(errp, "latency-ns is invalid");
-        ret = -EINVAL;
-    }
    qemu_opts_del(opts);
-    return ret;
+    return 0;
 }

 static void null_close(BlockDriverState *bs)
@@ -75,40 +58,28 @@ static int64_t null_getlength(BlockDriverState *bs)
    return s->length;
 }

-static coroutine_fn int null_co_common(BlockDriverState *bs)
-{
-    BDRVNullState *s = bs->opaque;
-
-    if (s->latency_ns) {
-        co_aio_sleep_ns(bdrv_get_aio_context(bs), QEMU_CLOCK_REALTIME,
-                        s->latency_ns);
-    }
-    return 0;
-}
-
 static coroutine_fn int null_co_readv(BlockDriverState *bs,
                                      int64_t sector_num, int nb_sectors,
                                      QEMUIOVector *qiov)
 {
-    return null_co_common(bs);
+    return 0;
 }

 static coroutine_fn int null_co_writev(BlockDriverState *bs,
                                       int64_t sector_num, int nb_sectors,
                                       QEMUIOVector *qiov)
 {
-    return null_co_common(bs);
+    return 0;
 }

 static coroutine_fn int null_co_flush(BlockDriverState *bs)
 {
-    return null_co_common(bs);
+    return 0;
 }

 typedef struct {
    BlockAIOCB common;
    QEMUBH *bh;
-    QEMUTimer timer;
 } NullAIOCB;

 static const AIOCBInfo null_aiocb_info = {
@@ -123,33 +94,15 @@ static void null_bh_cb(void *opaque)
    qemu_aio_unref(acb);
 }

-static void null_timer_cb(void *opaque)
-{
-    NullAIOCB *acb = opaque;
-    acb->common.cb(acb->common.opaque, 0);
-    timer_deinit(&acb->timer);
-    qemu_aio_unref(acb);
-}
-
 static inline BlockAIOCB *null_aio_common(BlockDriverState *bs,
                                          BlockCompletionFunc *cb,
                                          void *opaque)
 {
    NullAIOCB *acb;
-    BDRVNullState *s = bs->opaque;

    acb = qemu_aio_get(&null_aiocb_info, bs, cb, opaque);
-    /* Only emulate latency after vcpu is running. */
-    if (s->latency_ns) {
-        aio_timer_init(bdrv_get_aio_context(bs), &acb->timer,
-                       QEMU_CLOCK_REALTIME, SCALE_NS,
-                       null_timer_cb, acb);
-        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);
-    }
+    acb->bh = aio_bh_new(bdrv_get_aio_context(bs), null_bh_cb, acb);
+    qemu_bh_schedule(acb->bh);
    return &acb->common;
 }

@@ -178,12 +131,6 @@ static BlockAIOCB *null_aio_flush(BlockDriverState *bs,
    return null_aio_common(bs, cb, opaque);
 }

-static int null_reopen_prepare(BDRVReopenState *reopen_state,
-                               BlockReopenQueue *queue, Error **errp)
-{
-    return 0;
-}
-
 static BlockDriver bdrv_null_co = {
    .format_name            = "null-co",
    .protocol_name          = "null-co",
@@ -196,7 +143,6 @@ static BlockDriver bdrv_null_co = {
    .bdrv_co_readv          = null_co_readv,
    .bdrv_co_writev         = null_co_writev,
    .bdrv_co_flush_to_disk  = null_co_flush,
-    .bdrv_reopen_prepare    = null_reopen_prepare,
 };

 static BlockDriver bdrv_null_aio = {
@@ -211,7 +157,6 @@ static BlockDriver bdrv_null_aio = {
    .bdrv_aio_readv         = null_aio_readv,
    .bdrv_aio_writev        = null_aio_writev,
    .bdrv_aio_flush         = null_aio_flush,
-    .bdrv_reopen_prepare    = null_reopen_prepare,
 };

 static void bdrv_null_init(void)
--- a/block/parallels.c
+++ b/block/parallels.c
@@ -2,12 +2,8 @@
 * Block driver for Parallels disk image format
 *
 * Copyright (c) 2007 Alex Beregszaszi
- * Copyright (c) 2015 Denis V. Lunev <den@openvz.org>
 *
- * This code was originally based on comparing different disk images created
- * by Parallels. Currently it is based on opened OpenVZ sources
- * available at
- *     http://git.openvz.org/?p=ploop;a=summary
+ * This code is based on comparing different disk images created by Parallels.
 *
 * Permission is hereby granted, free of charge, to any person obtaining a copy
 * of this software and associated documentation files (the "Software"), to deal
@@ -27,548 +23,68 @@
 * 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 "block/block_int.h"
 #include "qemu/module.h"
-#include "qemu/bitmap.h"
-#include "qapi/util.h"

 /**************************************************************/

 #define HEADER_MAGIC "WithoutFreeSpace"
 #define HEADER_MAGIC2 "WithouFreSpacExt"
 #define HEADER_VERSION 2
-#define HEADER_INUSE_MAGIC  (0x746F6E59)
-
-#define DEFAULT_CLUSTER_SIZE 1048576        /* 1 MiB */
-
+#define HEADER_SIZE 64

 // always little-endian
-typedef struct ParallelsHeader {
+struct parallels_header {
    char magic[16]; // "WithoutFreeSpace"
    uint32_t version;
    uint32_t heads;
    uint32_t cylinders;
    uint32_t tracks;
-    uint32_t bat_entries;
+    uint32_t catalog_entries;
    uint64_t nb_sectors;
    uint32_t inuse;
    uint32_t data_off;
    char padding[12];
-} QEMU_PACKED ParallelsHeader;
-
-
-typedef enum ParallelsPreallocMode {
-    PRL_PREALLOC_MODE_FALLOCATE = 0,
-    PRL_PREALLOC_MODE_TRUNCATE = 1,
-    PRL_PREALLOC_MODE__MAX = 2,
-} ParallelsPreallocMode;
-
-static const char *prealloc_mode_lookup[] = {
-    "falloc",
-    "truncate",
-    NULL,
-};
-
+} QEMU_PACKED;

 typedef struct BDRVParallelsState {
-    /** Locking is conservative, the lock protects
-     *   - image file extending (truncate, fallocate)
-     *   - any access to block allocation table
-     */
    CoMutex lock;

-    ParallelsHeader *header;
-    uint32_t header_size;
-    bool header_unclean;
-
-    unsigned long *bat_dirty_bmap;
-    unsigned int  bat_dirty_block;
-
-    uint32_t *bat_bitmap;
-    unsigned int bat_size;
-
-    int64_t  data_end;
-    uint64_t prealloc_size;
-    ParallelsPreallocMode prealloc_mode;
+    uint32_t *catalog_bitmap;
+    unsigned int catalog_size;

    unsigned int tracks;

    unsigned int off_multiplier;
 } BDRVParallelsState;

-
-#define PARALLELS_OPT_PREALLOC_MODE     "prealloc-mode"
-#define PARALLELS_OPT_PREALLOC_SIZE     "prealloc-size"
-
-static QemuOptsList parallels_runtime_opts = {
-    .name = "parallels",
-    .head = QTAILQ_HEAD_INITIALIZER(parallels_runtime_opts.head),
-    .desc = {
-        {
-            .name = PARALLELS_OPT_PREALLOC_SIZE,
-            .type = QEMU_OPT_SIZE,
-            .help = "Preallocation size on image expansion",
-            .def_value_str = "128MiB",
-        },
-        {
-            .name = PARALLELS_OPT_PREALLOC_MODE,
-            .type = QEMU_OPT_STRING,
-            .help = "Preallocation mode on image expansion "
-                    "(allowed values: falloc, truncate)",
-            .def_value_str = "falloc",
-        },
-        { /* end of list */ },
-    },
-};
-
-
-static int64_t bat2sect(BDRVParallelsState *s, uint32_t idx)
+static int parallels_probe(const uint8_t *buf, int buf_size, const char *filename)
 {
-    return (uint64_t)le32_to_cpu(s->bat_bitmap[idx]) * s->off_multiplier;
-}
+    const struct parallels_header *ph = (const void *)buf;

-static uint32_t bat_entry_off(uint32_t idx)
-{
-    return sizeof(ParallelsHeader) + sizeof(uint32_t) * idx;
-}
-
-static int64_t seek_to_sector(BDRVParallelsState *s, int64_t sector_num)
-{
-    uint32_t index, offset;
-
-    index = sector_num / s->tracks;
-    offset = sector_num % s->tracks;
-
-    /* not allocated */
-    if ((index >= s->bat_size) || (s->bat_bitmap[index] == 0)) {
-        return -1;
-    }
-    return bat2sect(s, index) + offset;
-}
-
-static int cluster_remainder(BDRVParallelsState *s, int64_t sector_num,
-        int nb_sectors)
-{
-    int ret = s->tracks - sector_num % s->tracks;
-    return MIN(nb_sectors, ret);
-}
-
-static int64_t block_status(BDRVParallelsState *s, int64_t sector_num,
-                            int nb_sectors, int *pnum)
-{
-    int64_t start_off = -2, prev_end_off = -2;
-
-    *pnum = 0;
-    while (nb_sectors > 0 || start_off == -2) {
-        int64_t offset = seek_to_sector(s, sector_num);
-        int to_end;
-
-        if (start_off == -2) {
-            start_off = offset;
-            prev_end_off = offset;
-        } else if (offset != prev_end_off) {
-            break;
-        }
-
-        to_end = cluster_remainder(s, sector_num, nb_sectors);
-        nb_sectors -= to_end;
-        sector_num += to_end;
-        *pnum += to_end;
-
-        if (offset > 0) {
-            prev_end_off += to_end;
-        }
-    }
-    return start_off;
-}
-
-static int64_t allocate_clusters(BlockDriverState *bs, int64_t sector_num,
-                                 int nb_sectors, int *pnum)
-{
-    BDRVParallelsState *s = bs->opaque;
-    uint32_t idx, to_allocate, i;
-    int64_t pos, space;
-
-    pos = block_status(s, sector_num, nb_sectors, pnum);
-    if (pos > 0) {
-        return pos;
-    }
-
-    idx = sector_num / s->tracks;
-    if (idx >= s->bat_size) {
-        return -EINVAL;
-    }
-
-    to_allocate = (sector_num + *pnum + s->tracks - 1) / s->tracks - idx;
-    space = to_allocate * s->tracks;
-    if (s->data_end + space > bdrv_getlength(bs->file->bs) >> BDRV_SECTOR_BITS) {
-        int ret;
-        space += s->prealloc_size;
-        if (s->prealloc_mode == PRL_PREALLOC_MODE_FALLOCATE) {
-            ret = bdrv_write_zeroes(bs->file->bs, s->data_end, space, 0);
-        } else {
-            ret = bdrv_truncate(bs->file->bs,
-                                (s->data_end + space) << BDRV_SECTOR_BITS);
-        }
-        if (ret < 0) {
-            return ret;
-        }
-    }
-
-    for (i = 0; i < to_allocate; i++) {
-        s->bat_bitmap[idx + i] = cpu_to_le32(s->data_end / s->off_multiplier);
-        s->data_end += s->tracks;
-        bitmap_set(s->bat_dirty_bmap,
-                   bat_entry_off(idx + i) / s->bat_dirty_block, 1);
-    }
-
-    return bat2sect(s, idx) + sector_num % s->tracks;
-}
-
-
-static coroutine_fn int parallels_co_flush_to_os(BlockDriverState *bs)
-{
-    BDRVParallelsState *s = bs->opaque;
-    unsigned long size = DIV_ROUND_UP(s->header_size, s->bat_dirty_block);
-    unsigned long bit;
-
-    qemu_co_mutex_lock(&s->lock);
-
-    bit = find_first_bit(s->bat_dirty_bmap, size);
-    while (bit < size) {
-        uint32_t off = bit * s->bat_dirty_block;
-        uint32_t to_write = s->bat_dirty_block;
-        int ret;
-
-        if (off + to_write > s->header_size) {
-            to_write = s->header_size - off;
-        }
-        ret = bdrv_pwrite(bs->file->bs, off, (uint8_t *)s->header + off,
-                          to_write);
-        if (ret < 0) {
-            qemu_co_mutex_unlock(&s->lock);
-            return ret;
-        }
-        bit = find_next_bit(s->bat_dirty_bmap, size, bit + 1);
-    }
-    bitmap_zero(s->bat_dirty_bmap, size);
-
-    qemu_co_mutex_unlock(&s->lock);
-    return 0;
-}
-
-
-static int64_t coroutine_fn parallels_co_get_block_status(BlockDriverState *bs,
-        int64_t sector_num, int nb_sectors, int *pnum, BlockDriverState **file)
-{
-    BDRVParallelsState *s = bs->opaque;
-    int64_t offset;
-
-    qemu_co_mutex_lock(&s->lock);
-    offset = block_status(s, sector_num, nb_sectors, pnum);
-    qemu_co_mutex_unlock(&s->lock);
-
-    if (offset < 0) {
+    if (buf_size < HEADER_SIZE)
        return 0;
-    }
-
-    *file = bs->file->bs;
-    return (offset << BDRV_SECTOR_BITS) |
-        BDRV_BLOCK_DATA | BDRV_BLOCK_OFFSET_VALID;
-}
-
-static coroutine_fn int parallels_co_writev(BlockDriverState *bs,
-        int64_t sector_num, int nb_sectors, QEMUIOVector *qiov)
-{
-    BDRVParallelsState *s = bs->opaque;
-    uint64_t bytes_done = 0;
-    QEMUIOVector hd_qiov;
-    int ret = 0;
-
-    qemu_iovec_init(&hd_qiov, qiov->niov);
-
-    while (nb_sectors > 0) {
-        int64_t position;
-        int n, nbytes;
-
-        qemu_co_mutex_lock(&s->lock);
-        position = allocate_clusters(bs, sector_num, nb_sectors, &n);
-        qemu_co_mutex_unlock(&s->lock);
-        if (position < 0) {
-            ret = (int)position;
-            break;
-        }
-
-        nbytes = n << BDRV_SECTOR_BITS;
-
-        qemu_iovec_reset(&hd_qiov);
-        qemu_iovec_concat(&hd_qiov, qiov, bytes_done, nbytes);
-
-        ret = bdrv_co_writev(bs->file->bs, position, n, &hd_qiov);
-        if (ret < 0) {
-            break;
-        }
-
-        nb_sectors -= n;
-        sector_num += n;
-        bytes_done += nbytes;
-    }
-
-    qemu_iovec_destroy(&hd_qiov);
-    return ret;
-}
-
-static coroutine_fn int parallels_co_readv(BlockDriverState *bs,
-        int64_t sector_num, int nb_sectors, QEMUIOVector *qiov)
-{
-    BDRVParallelsState *s = bs->opaque;
-    uint64_t bytes_done = 0;
-    QEMUIOVector hd_qiov;
-    int ret = 0;
-
-    qemu_iovec_init(&hd_qiov, qiov->niov);
-
-    while (nb_sectors > 0) {
-        int64_t position;
-        int n, nbytes;
-
-        qemu_co_mutex_lock(&s->lock);
-        position = block_status(s, sector_num, nb_sectors, &n);
-        qemu_co_mutex_unlock(&s->lock);
-
-        nbytes = n << BDRV_SECTOR_BITS;
-
-        if (position < 0) {
-            qemu_iovec_memset(qiov, bytes_done, 0, nbytes);
-        } else {
-            qemu_iovec_reset(&hd_qiov);
-            qemu_iovec_concat(&hd_qiov, qiov, bytes_done, nbytes);
-
-            ret = bdrv_co_readv(bs->file->bs, position, n, &hd_qiov);
-            if (ret < 0) {
-                break;
-            }
-        }
-
-        nb_sectors -= n;
-        sector_num += n;
-        bytes_done += nbytes;
-    }
-
-    qemu_iovec_destroy(&hd_qiov);
-    return ret;
-}
-
-
-static int parallels_check(BlockDriverState *bs, BdrvCheckResult *res,
-                           BdrvCheckMode fix)
-{
-    BDRVParallelsState *s = bs->opaque;
-    int64_t size, prev_off, high_off;
-    int ret;
-    uint32_t i;
-    bool flush_bat = false;
-    int cluster_size = s->tracks << BDRV_SECTOR_BITS;
-
-    size = bdrv_getlength(bs->file->bs);
-    if (size < 0) {
-        res->check_errors++;
-        return size;
-    }
-
-    if (s->header_unclean) {
-        fprintf(stderr, "%s image was not closed correctly\n",
-                fix & BDRV_FIX_ERRORS ? "Repairing" : "ERROR");
-        res->corruptions++;
-        if (fix & BDRV_FIX_ERRORS) {
-            /* parallels_close will do the job right */
-            res->corruptions_fixed++;
-            s->header_unclean = false;
-        }
-    }
-
-    res->bfi.total_clusters = s->bat_size;
-    res->bfi.compressed_clusters = 0; /* compression is not supported */
-
-    high_off = 0;
-    prev_off = 0;
-    for (i = 0; i < s->bat_size; i++) {
-        int64_t off = bat2sect(s, i) << BDRV_SECTOR_BITS;
-        if (off == 0) {
-            prev_off = 0;
-            continue;
-        }
-
-        /* cluster outside the image */
-        if (off > size) {
-            fprintf(stderr, "%s cluster %u is outside image\n",
-                    fix & BDRV_FIX_ERRORS ? "Repairing" : "ERROR", i);
-            res->corruptions++;
-            if (fix & BDRV_FIX_ERRORS) {
-                prev_off = 0;
-                s->bat_bitmap[i] = 0;
-                res->corruptions_fixed++;
-                flush_bat = true;
-                continue;
-            }
-        }
-
-        res->bfi.allocated_clusters++;
-        if (off > high_off) {
-            high_off = off;
-        }
-
-        if (prev_off != 0 && (prev_off + cluster_size) != off) {
-            res->bfi.fragmented_clusters++;
-        }
-        prev_off = off;
-    }
-
-    if (flush_bat) {
-        ret = bdrv_pwrite_sync(bs->file->bs, 0, s->header, s->header_size);
-        if (ret < 0) {
-            res->check_errors++;
-            return ret;
-        }
-    }
-
-    res->image_end_offset = high_off + cluster_size;
-    if (size > res->image_end_offset) {
-        int64_t count;
-        count = DIV_ROUND_UP(size - res->image_end_offset, cluster_size);
-        fprintf(stderr, "%s space leaked at the end of the image %" PRId64 "\n",
-                fix & BDRV_FIX_LEAKS ? "Repairing" : "ERROR",
-                size - res->image_end_offset);
-        res->leaks += count;
-        if (fix & BDRV_FIX_LEAKS) {
-            ret = bdrv_truncate(bs->file->bs, res->image_end_offset);
-            if (ret < 0) {
-                res->check_errors++;
-                return ret;
-            }
-            res->leaks_fixed += count;
-        }
-    }
-
-    return 0;
-}
-
-
-static int parallels_create(const char *filename, QemuOpts *opts, Error **errp)
-{
-    int64_t total_size, cl_size;
-    uint8_t tmp[BDRV_SECTOR_SIZE];
-    Error *local_err = NULL;
-    BlockDriverState *file;
-    uint32_t bat_entries, bat_sectors;
-    ParallelsHeader header;
-    int ret;
-
-    total_size = ROUND_UP(qemu_opt_get_size_del(opts, BLOCK_OPT_SIZE, 0),
-                          BDRV_SECTOR_SIZE);
-    cl_size = ROUND_UP(qemu_opt_get_size_del(opts, BLOCK_OPT_CLUSTER_SIZE,
-                          DEFAULT_CLUSTER_SIZE), BDRV_SECTOR_SIZE);
-
-    ret = bdrv_create_file(filename, opts, &local_err);
-    if (ret < 0) {
-        error_propagate(errp, local_err);
-        return ret;
-    }
-
-    file = NULL;
-    ret = bdrv_open(&file, filename, NULL, NULL,
-                    BDRV_O_RDWR | BDRV_O_PROTOCOL, &local_err);
-    if (ret < 0) {
-        error_propagate(errp, local_err);
-        return ret;
-    }
-    ret = bdrv_truncate(file, 0);
-    if (ret < 0) {
-        goto exit;
-    }
-
-    bat_entries = DIV_ROUND_UP(total_size, cl_size);
-    bat_sectors = DIV_ROUND_UP(bat_entry_off(bat_entries), cl_size);
-    bat_sectors = (bat_sectors *  cl_size) >> BDRV_SECTOR_BITS;
-
-    memset(&header, 0, sizeof(header));
-    memcpy(header.magic, HEADER_MAGIC2, sizeof(header.magic));
-    header.version = cpu_to_le32(HEADER_VERSION);
-    /* don't care much about geometry, it is not used on image level */
-    header.heads = cpu_to_le32(16);
-    header.cylinders = cpu_to_le32(total_size / BDRV_SECTOR_SIZE / 16 / 32);
-    header.tracks = cpu_to_le32(cl_size >> BDRV_SECTOR_BITS);
-    header.bat_entries = cpu_to_le32(bat_entries);
-    header.nb_sectors = cpu_to_le64(DIV_ROUND_UP(total_size, BDRV_SECTOR_SIZE));
-    header.data_off = cpu_to_le32(bat_sectors);
-
-    /* write all the data */
-    memset(tmp, 0, sizeof(tmp));
-    memcpy(tmp, &header, sizeof(header));
-
-    ret = bdrv_pwrite(file, 0, tmp, BDRV_SECTOR_SIZE);
-    if (ret < 0) {
-        goto exit;
-    }
-    ret = bdrv_write_zeroes(file, 1, bat_sectors - 1, 0);
-    if (ret < 0) {
-        goto exit;
-    }
-    ret = 0;
-
-done:
-    bdrv_unref(file);
-    return ret;
-
-exit:
-    error_setg_errno(errp, -ret, "Failed to create Parallels image");
-    goto done;
-}
-
-
-static int parallels_probe(const uint8_t *buf, int buf_size,
-                           const char *filename)
-{
-    const ParallelsHeader *ph = (const void *)buf;
-
-    if (buf_size < sizeof(ParallelsHeader)) {
-        return 0;
-    }

    if ((!memcmp(ph->magic, HEADER_MAGIC, 16) ||
-           !memcmp(ph->magic, HEADER_MAGIC2, 16)) &&
-           (le32_to_cpu(ph->version) == HEADER_VERSION)) {
+        !memcmp(ph->magic, HEADER_MAGIC2, 16)) &&
+        (le32_to_cpu(ph->version) == HEADER_VERSION))
        return 100;
-    }

    return 0;
 }

-static int parallels_update_header(BlockDriverState *bs)
-{
-    BDRVParallelsState *s = bs->opaque;
-    unsigned size = MAX(bdrv_opt_mem_align(bs->file->bs),
-                        sizeof(ParallelsHeader));
-
-    if (size > s->header_size) {
-        size = s->header_size;
-    }
-    return bdrv_pwrite_sync(bs->file->bs, 0, s->header, size);
-}
-
 static int parallels_open(BlockDriverState *bs, QDict *options, int flags,
                          Error **errp)
 {
    BDRVParallelsState *s = bs->opaque;
-    ParallelsHeader ph;
-    int ret, size, i;
-    QemuOpts *opts = NULL;
-    Error *local_err = NULL;
-    char *buf;
+    int i;
+    struct parallels_header ph;
+    int ret;

-    ret = bdrv_pread(bs->file->bs, 0, &ph, sizeof(ph));
+    bs->read_only = 1; // no write support yet
+
+    ret = bdrv_pread(bs->file, 0, &ph, sizeof(ph));
    if (ret < 0) {
        goto fail;
    }
@@ -599,90 +115,25 @@ static int parallels_open(BlockDriverState *bs, QDict *options, int flags,
        goto fail;
    }

-    s->bat_size = le32_to_cpu(ph.bat_entries);
-    if (s->bat_size > INT_MAX / sizeof(uint32_t)) {
+    s->catalog_size = le32_to_cpu(ph.catalog_entries);
+    if (s->catalog_size > INT_MAX / 4) {
        error_setg(errp, "Catalog too large");
        ret = -EFBIG;
        goto fail;
    }
-
-    size = bat_entry_off(s->bat_size);
-    s->header_size = ROUND_UP(size, bdrv_opt_mem_align(bs->file->bs));
-    s->header = qemu_try_blockalign(bs->file->bs, s->header_size);
-    if (s->header == NULL) {
+    s->catalog_bitmap = g_try_new(uint32_t, s->catalog_size);
+    if (s->catalog_size && s->catalog_bitmap == NULL) {
        ret = -ENOMEM;
        goto fail;
    }
-    s->data_end = le32_to_cpu(ph.data_off);
-    if (s->data_end == 0) {
-        s->data_end = ROUND_UP(bat_entry_off(s->bat_size), BDRV_SECTOR_SIZE);
-    }
-    if (s->data_end < s->header_size) {
-        /* there is not enough unused space to fit to block align between BAT
-           and actual data. We can't avoid read-modify-write... */
-        s->header_size = size;
-    }

-    ret = bdrv_pread(bs->file->bs, 0, s->header, s->header_size);
+    ret = bdrv_pread(bs->file, 64, s->catalog_bitmap, s->catalog_size * 4);
    if (ret < 0) {
        goto fail;
    }
-    s->bat_bitmap = (uint32_t *)(s->header + 1);

-    for (i = 0; i < s->bat_size; i++) {
-        int64_t off = bat2sect(s, i);
-        if (off >= s->data_end) {
-            s->data_end = off + s->tracks;
-        }
-    }
-
-    if (le32_to_cpu(ph.inuse) == HEADER_INUSE_MAGIC) {
-        /* Image was not closed correctly. The check is mandatory */
-        s->header_unclean = true;
-        if ((flags & BDRV_O_RDWR) && !(flags & BDRV_O_CHECK)) {
-            error_setg(errp, "parallels: Image was not closed correctly; "
-                       "cannot be opened read/write");
-            ret = -EACCES;
-            goto fail;
-        }
-    }
-
-    opts = qemu_opts_create(&parallels_runtime_opts, NULL, 0, &local_err);
-    if (local_err != NULL) {
-        goto fail_options;
-    }
-
-    qemu_opts_absorb_qdict(opts, options, &local_err);
-    if (local_err != NULL) {
-        goto fail_options;
-    }
-
-    s->prealloc_size =
-        qemu_opt_get_size_del(opts, PARALLELS_OPT_PREALLOC_SIZE, 0);
-    s->prealloc_size = MAX(s->tracks, s->prealloc_size >> BDRV_SECTOR_BITS);
-    buf = qemu_opt_get_del(opts, PARALLELS_OPT_PREALLOC_MODE);
-    s->prealloc_mode = qapi_enum_parse(prealloc_mode_lookup, buf,
-            PRL_PREALLOC_MODE__MAX, PRL_PREALLOC_MODE_FALLOCATE, &local_err);
-    g_free(buf);
-    if (local_err != NULL) {
-        goto fail_options;
-    }
-    if (!bdrv_has_zero_init(bs->file->bs) ||
-            bdrv_truncate(bs->file->bs, bdrv_getlength(bs->file->bs)) != 0) {
-        s->prealloc_mode = PRL_PREALLOC_MODE_FALLOCATE;
-    }
-
-    if (flags & BDRV_O_RDWR) {
-        s->header->inuse = cpu_to_le32(HEADER_INUSE_MAGIC);
-        ret = parallels_update_header(bs);
-        if (ret < 0) {
-            goto fail;
-        }
-    }
-
-    s->bat_dirty_block = 4 * getpagesize();
-    s->bat_dirty_bmap =
-        bitmap_new(DIV_ROUND_UP(s->header_size, s->bat_dirty_block));
+    for (i = 0; i < s->catalog_size; i++)
+        le32_to_cpus(&s->catalog_bitmap[i]);

    qemu_co_mutex_init(&s->lock);
    return 0;
@@ -691,67 +142,67 @@ fail_format:
    error_setg(errp, "Image not in Parallels format");
    ret = -EINVAL;
 fail:
-    qemu_vfree(s->header);
+    g_free(s->catalog_bitmap);
    return ret;
-
-fail_options:
-    error_propagate(errp, local_err);
-    ret = -EINVAL;
-    goto fail;
 }

+static int64_t seek_to_sector(BlockDriverState *bs, int64_t sector_num)
+{
+    BDRVParallelsState *s = bs->opaque;
+    uint32_t index, offset;
+
+    index = sector_num / s->tracks;
+    offset = sector_num % s->tracks;
+
+    /* not allocated */
+    if ((index >= s->catalog_size) || (s->catalog_bitmap[index] == 0))
+        return -1;
+    return
+        ((uint64_t)s->catalog_bitmap[index] * s->off_multiplier + offset) * 512;
+}
+
+static int parallels_read(BlockDriverState *bs, int64_t sector_num,
+                    uint8_t *buf, int nb_sectors)
+{
+    while (nb_sectors > 0) {
+        int64_t position = seek_to_sector(bs, sector_num);
+        if (position >= 0) {
+            if (bdrv_pread(bs->file, position, buf, 512) != 512)
+                return -1;
+        } else {
+            memset(buf, 0, 512);
+        }
+        nb_sectors--;
+        sector_num++;
+        buf += 512;
+    }
+    return 0;
+}
+
+static coroutine_fn int parallels_co_read(BlockDriverState *bs, int64_t sector_num,
+                                          uint8_t *buf, int nb_sectors)
+{
+    int ret;
+    BDRVParallelsState *s = bs->opaque;
+    qemu_co_mutex_lock(&s->lock);
+    ret = parallels_read(bs, sector_num, buf, nb_sectors);
+    qemu_co_mutex_unlock(&s->lock);
+    return ret;
+}

 static void parallels_close(BlockDriverState *bs)
 {
    BDRVParallelsState *s = bs->opaque;
-
-    if (bs->open_flags & BDRV_O_RDWR) {
-        s->header->inuse = 0;
-        parallels_update_header(bs);
-    }
-
-    if (bs->open_flags & BDRV_O_RDWR) {
-        bdrv_truncate(bs->file->bs, s->data_end << BDRV_SECTOR_BITS);
-    }
-
-    g_free(s->bat_dirty_bmap);
-    qemu_vfree(s->header);
+    g_free(s->catalog_bitmap);
 }

-static QemuOptsList parallels_create_opts = {
-    .name = "parallels-create-opts",
-    .head = QTAILQ_HEAD_INITIALIZER(parallels_create_opts.head),
-    .desc = {
-        {
-            .name = BLOCK_OPT_SIZE,
-            .type = QEMU_OPT_SIZE,
-            .help = "Virtual disk size",
-        },
-        {
-            .name = BLOCK_OPT_CLUSTER_SIZE,
-            .type = QEMU_OPT_SIZE,
-            .help = "Parallels image cluster size",
-            .def_value_str = stringify(DEFAULT_CLUSTER_SIZE),
-        },
-        { /* end of list */ }
-    }
-};
-
 static BlockDriver bdrv_parallels = {
    .format_name	= "parallels",
    .instance_size	= sizeof(BDRVParallelsState),
    .bdrv_probe		= parallels_probe,
    .bdrv_open		= parallels_open,
+    .bdrv_read          = parallels_co_read,
    .bdrv_close		= parallels_close,
-    .bdrv_co_get_block_status = parallels_co_get_block_status,
-    .bdrv_has_zero_init       = bdrv_has_zero_init_1,
-    .bdrv_co_flush_to_os      = parallels_co_flush_to_os,
-    .bdrv_co_readv  = parallels_co_readv,
-    .bdrv_co_writev = parallels_co_writev,
-
-    .bdrv_create    = parallels_create,
-    .bdrv_check     = parallels_check,
-    .create_opts    = &parallels_create_opts,
 };

 static void bdrv_parallels_init(void)
--- a/block/qapi.c
+++ b/block/qapi.c
@@ -22,10 +22,8 @@
 * THE SOFTWARE.
 */

-#include "qemu/osdep.h"
 #include "block/qapi.h"
 #include "block/block_int.h"
-#include "block/throttle-groups.h"
 #include "block/write-threshold.h"
 #include "qmp-commands.h"
 #include "qapi-visit.h"
@@ -33,10 +31,8 @@
 #include "qapi/qmp/types.h"
 #include "sysemu/block-backend.h"

-BlockDeviceInfo *bdrv_block_device_info(BlockDriverState *bs, Error **errp)
+BlockDeviceInfo *bdrv_block_device_info(BlockDriverState *bs)
 {
-    ImageInfo **p_image_info;
-    BlockDriverState *bs0;
    BlockDeviceInfo *info = g_malloc0(sizeof(*info));

    info->file                   = g_strdup(bs->filename);
@@ -65,11 +61,9 @@ BlockDeviceInfo *bdrv_block_device_info(BlockDriverState *bs, Error **errp)
    info->backing_file_depth = bdrv_get_backing_file_depth(bs);
    info->detect_zeroes = bs->detect_zeroes;

-    if (bs->throttle_state) {
+    if (bs->io_limits_enabled) {
        ThrottleConfig cfg;
-
-        throttle_group_get_config(bs, &cfg);
-
+        throttle_get_config(&bs->throttle_state, &cfg);
        info->bps     = cfg.buckets[THROTTLE_BPS_TOTAL].avg;
        info->bps_rd  = cfg.buckets[THROTTLE_BPS_READ].avg;
        info->bps_wr  = cfg.buckets[THROTTLE_BPS_WRITE].avg;
@@ -92,54 +86,12 @@ BlockDeviceInfo *bdrv_block_device_info(BlockDriverState *bs, Error **errp)
        info->has_iops_wr_max = cfg.buckets[THROTTLE_OPS_WRITE].max;
        info->iops_wr_max     = cfg.buckets[THROTTLE_OPS_WRITE].max;

-        info->has_bps_max_length     = info->has_bps_max;
-        info->bps_max_length         =
-            cfg.buckets[THROTTLE_BPS_TOTAL].burst_length;
-        info->has_bps_rd_max_length  = info->has_bps_rd_max;
-        info->bps_rd_max_length      =
-            cfg.buckets[THROTTLE_BPS_READ].burst_length;
-        info->has_bps_wr_max_length  = info->has_bps_wr_max;
-        info->bps_wr_max_length      =
-            cfg.buckets[THROTTLE_BPS_WRITE].burst_length;
-
-        info->has_iops_max_length    = info->has_iops_max;
-        info->iops_max_length        =
-            cfg.buckets[THROTTLE_OPS_TOTAL].burst_length;
-        info->has_iops_rd_max_length = info->has_iops_rd_max;
-        info->iops_rd_max_length     =
-            cfg.buckets[THROTTLE_OPS_READ].burst_length;
-        info->has_iops_wr_max_length = info->has_iops_wr_max;
-        info->iops_wr_max_length     =
-            cfg.buckets[THROTTLE_OPS_WRITE].burst_length;
-
        info->has_iops_size = cfg.op_size;
        info->iops_size = cfg.op_size;
-
-        info->has_group = true;
-        info->group = g_strdup(throttle_group_get_name(bs));
    }

    info->write_threshold = bdrv_write_threshold_get(bs);

-    bs0 = bs;
-    p_image_info = &info->image;
-    while (1) {
-        Error *local_err = NULL;
-        bdrv_query_image_info(bs0, p_image_info, &local_err);
-        if (local_err) {
-            error_propagate(errp, local_err);
-            qapi_free_BlockDeviceInfo(info);
-            return NULL;
-        }
-        if (bs0->drv && bs0->backing) {
-            bs0 = bs0->backing->bs;
-            (*p_image_info)->has_backing_image = true;
-            p_image_info = &((*p_image_info)->backing_image);
-        } else {
-            break;
-        }
-    }
-
    return info;
 }

@@ -231,13 +183,11 @@ void bdrv_query_image_info(BlockDriverState *bs,
    Error *err = NULL;
    ImageInfo *info;

-    aio_context_acquire(bdrv_get_aio_context(bs));
-
    size = bdrv_getlength(bs);
    if (size < 0) {
        error_setg_errno(errp, -size, "Can't get size of device '%s'",
                         bdrv_get_device_name(bs));
-        goto out;
+        return;
    }

    info = g_new0(ImageInfo, 1);
@@ -268,18 +218,15 @@ void bdrv_query_image_info(BlockDriverState *bs,
        info->has_backing_filename = true;
        bdrv_get_full_backing_filename(bs, backing_filename2, PATH_MAX, &err);
        if (err) {
-            /* Can't reconstruct the full backing filename, so we must omit
-             * this field and apply a Best Effort to this query. */
+            error_propagate(errp, err);
+            qapi_free_ImageInfo(info);
            g_free(backing_filename2);
-            backing_filename2 = NULL;
-            error_free(err);
-            err = NULL;
+            return;
        }

-        /* Always report the full_backing_filename if present, even if it's the
-         * same as backing_filename. That they are same is useful info. */
-        if (backing_filename2) {
-            info->full_backing_filename = g_strdup(backing_filename2);
+        if (strcmp(backing_filename, backing_filename2) != 0) {
+            info->full_backing_filename =
+                        g_strdup(backing_filename2);
            info->has_full_backing_filename = true;
        }

@@ -305,13 +252,10 @@ void bdrv_query_image_info(BlockDriverState *bs,
    default:
        error_propagate(errp, err);
        qapi_free_ImageInfo(info);
-        goto out;
+        return;
    }

    *p_info = info;
-
-out:
-    aio_context_release(bdrv_get_aio_context(bs));
 }

 /* @p_info will be set only on success. */
@@ -320,31 +264,48 @@ static void bdrv_query_info(BlockBackend *blk, BlockInfo **p_info,
 {
    BlockInfo *info = g_malloc0(sizeof(*info));
    BlockDriverState *bs = blk_bs(blk);
+    BlockDriverState *bs0;
+    ImageInfo **p_image_info;
+    Error *local_err = NULL;
    info->device = g_strdup(blk_name(blk));
    info->type = g_strdup("unknown");
    info->locked = blk_dev_is_medium_locked(blk);
    info->removable = blk_dev_has_removable_media(blk);

-    if (blk_dev_has_tray(blk)) {
+    if (blk_dev_has_removable_media(blk)) {
        info->has_tray_open = true;
        info->tray_open = blk_dev_is_tray_open(blk);
    }

-    if (blk_iostatus_is_enabled(blk)) {
+    if (bdrv_iostatus_is_enabled(bs)) {
        info->has_io_status = true;
-        info->io_status = blk_iostatus(blk);
+        info->io_status = bs->iostatus;
    }

-    if (bs && !QLIST_EMPTY(&bs->dirty_bitmaps)) {
+    if (!QLIST_EMPTY(&bs->dirty_bitmaps)) {
        info->has_dirty_bitmaps = true;
        info->dirty_bitmaps = bdrv_query_dirty_bitmaps(bs);
    }

-    if (bs && bs->drv) {
+    if (bs->drv) {
        info->has_inserted = true;
-        info->inserted = bdrv_block_device_info(bs, errp);
-        if (info->inserted == NULL) {
-            goto err;
+        info->inserted = bdrv_block_device_info(bs);
+
+        bs0 = bs;
+        p_image_info = &info->inserted->image;
+        while (1) {
+            bdrv_query_image_info(bs0, p_image_info, &local_err);
+            if (local_err) {
+                error_propagate(errp, local_err);
+                goto err;
+            }
+            if (bs0->drv && bs0->backing_hd) {
+                bs0 = bs0->backing_hd;
+                (*p_image_info)->has_backing_image = true;
+                p_image_info = &((*p_image_info)->backing_image);
+            } else {
+                break;
+            }
        }
    }

@@ -373,82 +334,27 @@ static BlockStats *bdrv_query_stats(const BlockDriverState *bs,
    }

    s->stats = g_malloc0(sizeof(*s->stats));
-    if (bs->blk) {
-        BlockAcctStats *stats = blk_get_stats(bs->blk);
-        BlockAcctTimedStats *ts = NULL;
-
-        s->stats->rd_bytes = stats->nr_bytes[BLOCK_ACCT_READ];
-        s->stats->wr_bytes = stats->nr_bytes[BLOCK_ACCT_WRITE];
-        s->stats->rd_operations = stats->nr_ops[BLOCK_ACCT_READ];
-        s->stats->wr_operations = stats->nr_ops[BLOCK_ACCT_WRITE];
-
-        s->stats->failed_rd_operations = stats->failed_ops[BLOCK_ACCT_READ];
-        s->stats->failed_wr_operations = stats->failed_ops[BLOCK_ACCT_WRITE];
-        s->stats->failed_flush_operations = stats->failed_ops[BLOCK_ACCT_FLUSH];
-
-        s->stats->invalid_rd_operations = stats->invalid_ops[BLOCK_ACCT_READ];
-        s->stats->invalid_wr_operations = stats->invalid_ops[BLOCK_ACCT_WRITE];
-        s->stats->invalid_flush_operations =
-            stats->invalid_ops[BLOCK_ACCT_FLUSH];
-
-        s->stats->rd_merged = stats->merged[BLOCK_ACCT_READ];
-        s->stats->wr_merged = stats->merged[BLOCK_ACCT_WRITE];
-        s->stats->flush_operations = stats->nr_ops[BLOCK_ACCT_FLUSH];
-        s->stats->wr_total_time_ns = stats->total_time_ns[BLOCK_ACCT_WRITE];
-        s->stats->rd_total_time_ns = stats->total_time_ns[BLOCK_ACCT_READ];
-        s->stats->flush_total_time_ns = stats->total_time_ns[BLOCK_ACCT_FLUSH];
-
-        s->stats->has_idle_time_ns = stats->last_access_time_ns > 0;
-        if (s->stats->has_idle_time_ns) {
-            s->stats->idle_time_ns = block_acct_idle_time_ns(stats);
-        }
-
-        s->stats->account_invalid = stats->account_invalid;
-        s->stats->account_failed = stats->account_failed;
-
-        while ((ts = block_acct_interval_next(stats, ts))) {
-            BlockDeviceTimedStatsList *timed_stats =
-                g_malloc0(sizeof(*timed_stats));
-            BlockDeviceTimedStats *dev_stats = g_malloc0(sizeof(*dev_stats));
-            timed_stats->next = s->stats->timed_stats;
-            timed_stats->value = dev_stats;
-            s->stats->timed_stats = timed_stats;
-
-            TimedAverage *rd = &ts->latency[BLOCK_ACCT_READ];
-            TimedAverage *wr = &ts->latency[BLOCK_ACCT_WRITE];
-            TimedAverage *fl = &ts->latency[BLOCK_ACCT_FLUSH];
-
-            dev_stats->interval_length = ts->interval_length;
-
-            dev_stats->min_rd_latency_ns = timed_average_min(rd);
-            dev_stats->max_rd_latency_ns = timed_average_max(rd);
-            dev_stats->avg_rd_latency_ns = timed_average_avg(rd);
-
-            dev_stats->min_wr_latency_ns = timed_average_min(wr);
-            dev_stats->max_wr_latency_ns = timed_average_max(wr);
-            dev_stats->avg_wr_latency_ns = timed_average_avg(wr);
-
-            dev_stats->min_flush_latency_ns = timed_average_min(fl);
-            dev_stats->max_flush_latency_ns = timed_average_max(fl);
-            dev_stats->avg_flush_latency_ns = timed_average_avg(fl);
-
-            dev_stats->avg_rd_queue_depth =
-                block_acct_queue_depth(ts, BLOCK_ACCT_READ);
-            dev_stats->avg_wr_queue_depth =
-                block_acct_queue_depth(ts, BLOCK_ACCT_WRITE);
-        }
-    }
-
-    s->stats->wr_highest_offset = bs->wr_highest_offset;
+    s->stats->rd_bytes = bs->stats.nr_bytes[BLOCK_ACCT_READ];
+    s->stats->wr_bytes = bs->stats.nr_bytes[BLOCK_ACCT_WRITE];
+    s->stats->rd_operations = bs->stats.nr_ops[BLOCK_ACCT_READ];
+    s->stats->wr_operations = bs->stats.nr_ops[BLOCK_ACCT_WRITE];
+    s->stats->rd_merged = bs->stats.merged[BLOCK_ACCT_READ];
+    s->stats->wr_merged = bs->stats.merged[BLOCK_ACCT_WRITE];
+    s->stats->wr_highest_offset =
+        bs->stats.wr_highest_sector * BDRV_SECTOR_SIZE;
+    s->stats->flush_operations = bs->stats.nr_ops[BLOCK_ACCT_FLUSH];
+    s->stats->wr_total_time_ns = bs->stats.total_time_ns[BLOCK_ACCT_WRITE];
+    s->stats->rd_total_time_ns = bs->stats.total_time_ns[BLOCK_ACCT_READ];
+    s->stats->flush_total_time_ns = bs->stats.total_time_ns[BLOCK_ACCT_FLUSH];

    if (bs->file) {
        s->has_parent = true;
-        s->parent = bdrv_query_stats(bs->file->bs, query_backing);
+        s->parent = bdrv_query_stats(bs->file, query_backing);
    }

-    if (query_backing && bs->backing) {
+    if (query_backing && bs->backing_hd) {
        s->has_backing = true;
-        s->backing = bdrv_query_stats(bs->backing->bs, query_backing);
+        s->backing = bdrv_query_stats(bs->backing_hd, query_backing);
    }

    return s;
@@ -465,9 +371,7 @@ BlockInfoList *qmp_query_block(Error **errp)
        bdrv_query_info(blk, &info->value, &local_err);
        if (local_err) {
            error_propagate(errp, local_err);
-            g_free(info);
-            qapi_free_BlockInfoList(head);
-            return NULL;
+            goto err;
        }

        *p_next = info;
@@ -475,6 +379,10 @@ BlockInfoList *qmp_query_block(Error **errp)
    }

    return head;
+
+ err:
+    qapi_free_BlockInfoList(head);
+    return NULL;
 }

 BlockStatsList *qmp_query_blockstats(bool has_query_nodes,
@@ -602,9 +510,18 @@ static void dump_qobject(fprintf_function func_fprintf, void *f,
        }
        case QTYPE_QBOOL: {
            QBool *value = qobject_to_qbool(obj);
-            func_fprintf(f, "%s", qbool_get_bool(value) ? "true" : "false");
+            func_fprintf(f, "%s", qbool_get_int(value) ? "true" : "false");
            break;
        }
+        case QTYPE_QERROR: {
+            QString *value = qerror_human((QError *)obj);
+            func_fprintf(f, "%s", qstring_get_str(value));
+            QDECREF(value);
+            break;
+        }
+        case QTYPE_NONE:
+            break;
+        case QTYPE_MAX:
        default:
            abort();
    }
@@ -617,7 +534,7 @@ static void dump_qlist(fprintf_function func_fprintf, void *f, int indentation,
    int i = 0;

    for (entry = qlist_first(list); entry; entry = qlist_next(entry), i++) {
-        QType type = qobject_type(entry->value);
+        qtype_code type = qobject_type(entry->value);
        bool composite = (type == QTYPE_QDICT || type == QTYPE_QLIST);
        const char *format = composite ? "%*s[%i]:\n" : "%*s[%i]: ";

@@ -635,7 +552,7 @@ static void dump_qdict(fprintf_function func_fprintf, void *f, int indentation,
    const QDictEntry *entry;

    for (entry = qdict_first(dict); entry; entry = qdict_next(dict, entry)) {
-        QType type = qobject_type(entry->value);
+        qtype_code type = qobject_type(entry->value);
        bool composite = (type == QTYPE_QDICT || type == QTYPE_QLIST);
        const char *format = composite ? "%*s%s:\n" : "%*s%s: ";
        char key[strlen(entry->key) + 1];
@@ -661,7 +578,7 @@ void bdrv_image_info_specific_dump(fprintf_function func_fprintf, void *f,
    QmpOutputVisitor *ov = qmp_output_visitor_new();
    QObject *obj, *data;

-    visit_type_ImageInfoSpecific(qmp_output_get_visitor(ov), NULL, &info_spec,
+    visit_type_ImageInfoSpecific(qmp_output_get_visitor(ov), &info_spec, NULL,
                                 &error_abort);
    obj = qmp_output_get_qobject(ov);
    assert(qobject_type(obj) == QTYPE_QDICT);
@@ -705,10 +622,7 @@ void bdrv_image_info_dump(fprintf_function func_fprintf, void *f,

    if (info->has_backing_filename) {
        func_fprintf(f, "backing file: %s", info->backing_filename);
-        if (!info->has_full_backing_filename) {
-            func_fprintf(f, " (cannot determine actual path)");
-        } else if (strcmp(info->backing_filename,
-                          info->full_backing_filename) != 0) {
+        if (info->has_full_backing_filename) {
            func_fprintf(f, " (actual path: %s)", info->full_backing_filename);
        }
        func_fprintf(f, "\n");
--- a/block/qcow.c
+++ b/block/qcow.c
@@ -21,13 +21,11 @@
 * 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 "block/block_int.h"
 #include "qemu/module.h"
 #include <zlib.h>
-#include "qapi/qmp/qerror.h"
-#include "crypto/cipher.h"
+#include "qemu/aes.h"
 #include "migration/migration.h"

 /**************************************************************/
@@ -73,8 +71,10 @@ typedef struct BDRVQcowState {
    uint8_t *cluster_cache;
    uint8_t *cluster_data;
    uint64_t cluster_cache_offset;
-    QCryptoCipher *cipher; /* NULL if no key yet */
+    uint32_t crypt_method; /* current crypt method, 0 if no key yet */
    uint32_t crypt_method_header;
+    AES_KEY aes_encrypt_key;
+    AES_KEY aes_decrypt_key;
    CoMutex lock;
    Error *migration_blocker;
 } BDRVQcowState;
@@ -101,7 +101,7 @@ static int qcow_open(BlockDriverState *bs, QDict *options, int flags,
    int ret;
    QCowHeader header;

-    ret = bdrv_pread(bs->file->bs, 0, &header, sizeof(header));
+    ret = bdrv_pread(bs->file, 0, &header, sizeof(header));
    if (ret < 0) {
        goto fail;
    }
@@ -123,8 +123,8 @@ static int qcow_open(BlockDriverState *bs, QDict *options, int flags,
        char version[64];
        snprintf(version, sizeof(version), "QCOW version %" PRIu32,
                 header.version);
-        error_setg(errp, QERR_UNKNOWN_BLOCK_FORMAT_FEATURE,
-                   bdrv_get_device_or_node_name(bs), "qcow", version);
+        error_set(errp, QERR_UNKNOWN_BLOCK_FORMAT_FEATURE,
+                  bdrv_get_device_name(bs), "qcow", version);
        ret = -ENOTSUP;
        goto fail;
    }
@@ -153,11 +153,6 @@ static int qcow_open(BlockDriverState *bs, QDict *options, int flags,
        ret = -EINVAL;
        goto fail;
    }
-    if (!qcrypto_cipher_supports(QCRYPTO_CIPHER_ALG_AES_128)) {
-        error_setg(errp, "AES cipher not available");
-        ret = -EINVAL;
-        goto fail;
-    }
    s->crypt_method_header = header.crypt_method;
    if (s->crypt_method_header) {
        bs->encrypted = 1;
@@ -194,7 +189,7 @@ static int qcow_open(BlockDriverState *bs, QDict *options, int flags,
        goto fail;
    }

-    ret = bdrv_pread(bs->file->bs, s->l1_table_offset, s->l1_table,
+    ret = bdrv_pread(bs->file, s->l1_table_offset, s->l1_table,
               s->l1_size * sizeof(uint64_t));
    if (ret < 0) {
        goto fail;
@@ -206,7 +201,7 @@ static int qcow_open(BlockDriverState *bs, QDict *options, int flags,

    /* alloc L2 cache (max. 64k * 16 * 8 = 8 MB) */
    s->l2_cache =
-        qemu_try_blockalign(bs->file->bs,
+        qemu_try_blockalign(bs->file,
                            s->l2_size * L2_CACHE_SIZE * sizeof(uint64_t));
    if (s->l2_cache == NULL) {
        error_setg(errp, "Could not allocate L2 table cache");
@@ -225,7 +220,7 @@ static int qcow_open(BlockDriverState *bs, QDict *options, int flags,
            ret = -EINVAL;
            goto fail;
        }
-        ret = bdrv_pread(bs->file->bs, header.backing_file_offset,
+        ret = bdrv_pread(bs->file, header.backing_file_offset,
                   bs->backing_file, len);
        if (ret < 0) {
            goto fail;
@@ -234,9 +229,9 @@ static int qcow_open(BlockDriverState *bs, QDict *options, int flags,
    }

    /* Disable migration when qcow images are used */
-    error_setg(&s->migration_blocker, "The qcow format used by node '%s' "
-               "does not support live migration",
-               bdrv_get_device_or_node_name(bs));
+    error_set(&s->migration_blocker,
+              QERR_BLOCK_FORMAT_FEATURE_NOT_SUPPORTED,
+              "qcow", bdrv_get_device_name(bs), "live migration");
    migrate_add_blocker(s->migration_blocker);

    qemu_co_mutex_init(&s->lock);
@@ -264,7 +259,6 @@ static int qcow_set_key(BlockDriverState *bs, const char *key)
    BDRVQcowState *s = bs->opaque;
    uint8_t keybuf[16];
    int len, i;
-    Error *err;

    memset(keybuf, 0, 16);
    len = strlen(key);
@@ -275,68 +269,38 @@ static int qcow_set_key(BlockDriverState *bs, const char *key)
    for(i = 0;i < len;i++) {
        keybuf[i] = key[i];
    }
-    assert(bs->encrypted);
+    s->crypt_method = s->crypt_method_header;

-    qcrypto_cipher_free(s->cipher);
-    s->cipher = qcrypto_cipher_new(
-        QCRYPTO_CIPHER_ALG_AES_128,
-        QCRYPTO_CIPHER_MODE_CBC,
-        keybuf, G_N_ELEMENTS(keybuf),
-        &err);
-
-    if (!s->cipher) {
-        /* XXX would be nice if errors in this method could
-         * be properly propagate to the caller. Would need
-         * the bdrv_set_key() API signature to be fixed. */
-        error_free(err);
+    if (AES_set_encrypt_key(keybuf, 128, &s->aes_encrypt_key) != 0)
+        return -1;
+    if (AES_set_decrypt_key(keybuf, 128, &s->aes_decrypt_key) != 0)
        return -1;
-    }
    return 0;
 }

 /* The crypt function is compatible with the linux cryptoloop
   algorithm for < 4 GB images. NOTE: out_buf == in_buf is
   supported */
-static int encrypt_sectors(BDRVQcowState *s, int64_t sector_num,
-                           uint8_t *out_buf, const uint8_t *in_buf,
-                           int nb_sectors, bool enc, Error **errp)
+static void encrypt_sectors(BDRVQcowState *s, int64_t sector_num,
+                            uint8_t *out_buf, const uint8_t *in_buf,
+                            int nb_sectors, int enc,
+                            const AES_KEY *key)
 {
    union {
        uint64_t ll[2];
        uint8_t b[16];
    } ivec;
    int i;
-    int ret;

    for(i = 0; i < nb_sectors; i++) {
        ivec.ll[0] = cpu_to_le64(sector_num);
        ivec.ll[1] = 0;
-        if (qcrypto_cipher_setiv(s->cipher,
-                                 ivec.b, G_N_ELEMENTS(ivec.b),
-                                 errp) < 0) {
-            return -1;
-        }
-        if (enc) {
-            ret = qcrypto_cipher_encrypt(s->cipher,
-                                         in_buf,
-                                         out_buf,
-                                         512,
-                                         errp);
-        } else {
-            ret = qcrypto_cipher_decrypt(s->cipher,
-                                         in_buf,
-                                         out_buf,
-                                         512,
-                                         errp);
-        }
-        if (ret < 0) {
-            return -1;
-        }
+        AES_cbc_encrypt(in_buf, out_buf, 512, key,
+                        ivec.b, enc);
        sector_num++;
        in_buf += 512;
        out_buf += 512;
    }
-    return 0;
 }

 /* 'allocate' is:
@@ -370,13 +334,13 @@ static uint64_t get_cluster_offset(BlockDriverState *bs,
        if (!allocate)
            return 0;
        /* allocate a new l2 entry */
-        l2_offset = bdrv_getlength(bs->file->bs);
+        l2_offset = bdrv_getlength(bs->file);
        /* round to cluster size */
        l2_offset = (l2_offset + s->cluster_size - 1) & ~(s->cluster_size - 1);
        /* update the L1 entry */
        s->l1_table[l1_index] = l2_offset;
        tmp = cpu_to_be64(l2_offset);
-        if (bdrv_pwrite_sync(bs->file->bs,
+        if (bdrv_pwrite_sync(bs->file,
                s->l1_table_offset + l1_index * sizeof(tmp),
                &tmp, sizeof(tmp)) < 0)
            return 0;
@@ -406,12 +370,11 @@ static uint64_t get_cluster_offset(BlockDriverState *bs,
    l2_table = s->l2_cache + (min_index << s->l2_bits);
    if (new_l2_table) {
        memset(l2_table, 0, s->l2_size * sizeof(uint64_t));
-        if (bdrv_pwrite_sync(bs->file->bs, l2_offset, l2_table,
+        if (bdrv_pwrite_sync(bs->file, l2_offset, l2_table,
                s->l2_size * sizeof(uint64_t)) < 0)
            return 0;
    } else {
-        if (bdrv_pread(bs->file->bs, l2_offset, l2_table,
-                       s->l2_size * sizeof(uint64_t)) !=
+        if (bdrv_pread(bs->file, l2_offset, l2_table, s->l2_size * sizeof(uint64_t)) !=
            s->l2_size * sizeof(uint64_t))
            return 0;
    }
@@ -432,42 +395,34 @@ static uint64_t get_cluster_offset(BlockDriverState *bs,
               overwritten */
            if (decompress_cluster(bs, cluster_offset) < 0)
                return 0;
-            cluster_offset = bdrv_getlength(bs->file->bs);
+            cluster_offset = bdrv_getlength(bs->file);
            cluster_offset = (cluster_offset + s->cluster_size - 1) &
                ~(s->cluster_size - 1);
            /* write the cluster content */
-            if (bdrv_pwrite(bs->file->bs, cluster_offset, s->cluster_cache,
-                            s->cluster_size) !=
+            if (bdrv_pwrite(bs->file, cluster_offset, s->cluster_cache, s->cluster_size) !=
                s->cluster_size)
                return -1;
        } else {
-            cluster_offset = bdrv_getlength(bs->file->bs);
+            cluster_offset = bdrv_getlength(bs->file);
            if (allocate == 1) {
                /* round to cluster size */
                cluster_offset = (cluster_offset + s->cluster_size - 1) &
                    ~(s->cluster_size - 1);
-                bdrv_truncate(bs->file->bs, cluster_offset + s->cluster_size);
+                bdrv_truncate(bs->file, cluster_offset + s->cluster_size);
                /* if encrypted, we must initialize the cluster
                   content which won't be written */
-                if (bs->encrypted &&
+                if (s->crypt_method &&
                    (n_end - n_start) < s->cluster_sectors) {
                    uint64_t start_sect;
-                    assert(s->cipher);
                    start_sect = (offset & ~(s->cluster_size - 1)) >> 9;
                    memset(s->cluster_data + 512, 0x00, 512);
                    for(i = 0; i < s->cluster_sectors; i++) {
                        if (i < n_start || i >= n_end) {
-                            Error *err = NULL;
-                            if (encrypt_sectors(s, start_sect + i,
-                                                s->cluster_data,
-                                                s->cluster_data + 512, 1,
-                                                true, &err) < 0) {
-                                error_free(err);
-                                errno = EIO;
-                                return -1;
-                            }
-                            if (bdrv_pwrite(bs->file->bs,
-                                            cluster_offset + i * 512,
+                            encrypt_sectors(s, start_sect + i,
+                                            s->cluster_data,
+                                            s->cluster_data + 512, 1, 1,
+                                            &s->aes_encrypt_key);
+                            if (bdrv_pwrite(bs->file, cluster_offset + i * 512,
                                            s->cluster_data, 512) != 512)
                                return -1;
                        }
@@ -481,7 +436,7 @@ static uint64_t get_cluster_offset(BlockDriverState *bs,
        /* update L2 table */
        tmp = cpu_to_be64(cluster_offset);
        l2_table[l2_index] = tmp;
-        if (bdrv_pwrite_sync(bs->file->bs, l2_offset + l2_index * sizeof(tmp),
+        if (bdrv_pwrite_sync(bs->file, l2_offset + l2_index * sizeof(tmp),
                &tmp, sizeof(tmp)) < 0)
            return 0;
    }
@@ -489,7 +444,7 @@ static uint64_t get_cluster_offset(BlockDriverState *bs,
 }

 static int64_t coroutine_fn qcow_co_get_block_status(BlockDriverState *bs,
-        int64_t sector_num, int nb_sectors, int *pnum, BlockDriverState **file)
+        int64_t sector_num, int nb_sectors, int *pnum)
 {
    BDRVQcowState *s = bs->opaque;
    int index_in_cluster, n;
@@ -506,11 +461,10 @@ static int64_t coroutine_fn qcow_co_get_block_status(BlockDriverState *bs,
    if (!cluster_offset) {
        return 0;
    }
-    if ((cluster_offset & QCOW_OFLAG_COMPRESSED) || s->cipher) {
+    if ((cluster_offset & QCOW_OFLAG_COMPRESSED) || s->crypt_method) {
        return BDRV_BLOCK_DATA;
    }
    cluster_offset |= (index_in_cluster << BDRV_SECTOR_BITS);
-    *file = bs->file->bs;
    return BDRV_BLOCK_DATA | BDRV_BLOCK_OFFSET_VALID | cluster_offset;
 }

@@ -551,7 +505,7 @@ static int decompress_cluster(BlockDriverState *bs, uint64_t cluster_offset)
    if (s->cluster_cache_offset != coffset) {
        csize = cluster_offset >> (63 - s->cluster_bits);
        csize &= (s->cluster_size - 1);
-        ret = bdrv_pread(bs->file->bs, coffset, s->cluster_data, csize);
+        ret = bdrv_pread(bs->file, coffset, s->cluster_data, csize);
        if (ret != csize)
            return -1;
        if (decompress_buffer(s->cluster_cache, s->cluster_size,
@@ -574,7 +528,6 @@ static coroutine_fn int qcow_co_readv(BlockDriverState *bs, int64_t sector_num,
    QEMUIOVector hd_qiov;
    uint8_t *buf;
    void *orig_buf;
-    Error *err = NULL;

    if (qiov->niov > 1) {
        buf = orig_buf = qemu_try_blockalign(bs, qiov->size);
@@ -599,13 +552,13 @@ static coroutine_fn int qcow_co_readv(BlockDriverState *bs, int64_t sector_num,
        }

        if (!cluster_offset) {
-            if (bs->backing) {
+            if (bs->backing_hd) {
                /* read from the base image */
                hd_iov.iov_base = (void *)buf;
                hd_iov.iov_len = n * 512;
                qemu_iovec_init_external(&hd_qiov, &hd_iov, 1);
                qemu_co_mutex_unlock(&s->lock);
-                ret = bdrv_co_readv(bs->backing->bs, sector_num,
+                ret = bdrv_co_readv(bs->backing_hd, sector_num,
                                    n, &hd_qiov);
                qemu_co_mutex_lock(&s->lock);
                if (ret < 0) {
@@ -630,19 +583,17 @@ static coroutine_fn int qcow_co_readv(BlockDriverState *bs, int64_t sector_num,
            hd_iov.iov_len = n * 512;
            qemu_iovec_init_external(&hd_qiov, &hd_iov, 1);
            qemu_co_mutex_unlock(&s->lock);
-            ret = bdrv_co_readv(bs->file->bs,
+            ret = bdrv_co_readv(bs->file,
                                (cluster_offset >> 9) + index_in_cluster,
                                n, &hd_qiov);
            qemu_co_mutex_lock(&s->lock);
            if (ret < 0) {
                break;
            }
-            if (bs->encrypted) {
-                assert(s->cipher);
-                if (encrypt_sectors(s, sector_num, buf, buf,
-                                    n, false, &err) < 0) {
-                    goto fail;
-                }
+            if (s->crypt_method) {
+                encrypt_sectors(s, sector_num, buf, buf,
+                                n, 0,
+                                &s->aes_decrypt_key);
            }
        }
        ret = 0;
@@ -663,7 +614,6 @@ done:
    return ret;

 fail:
-    error_free(err);
    ret = -EIO;
    goto done;
 }
@@ -711,18 +661,12 @@ static coroutine_fn int qcow_co_writev(BlockDriverState *bs, int64_t sector_num,
            ret = -EIO;
            break;
        }
-        if (bs->encrypted) {
-            Error *err = NULL;
-            assert(s->cipher);
+        if (s->crypt_method) {
            if (!cluster_data) {
                cluster_data = g_malloc0(s->cluster_size);
            }
-            if (encrypt_sectors(s, sector_num, cluster_data, buf,
-                                n, true, &err) < 0) {
-                error_free(err);
-                ret = -EIO;
-                break;
-            }
+            encrypt_sectors(s, sector_num, cluster_data, buf,
+                            n, 1, &s->aes_encrypt_key);
            src_buf = cluster_data;
        } else {
            src_buf = buf;
@@ -732,7 +676,7 @@ static coroutine_fn int qcow_co_writev(BlockDriverState *bs, int64_t sector_num,
        hd_iov.iov_len = n * 512;
        qemu_iovec_init_external(&hd_qiov, &hd_iov, 1);
        qemu_co_mutex_unlock(&s->lock);
-        ret = bdrv_co_writev(bs->file->bs,
+        ret = bdrv_co_writev(bs->file,
                             (cluster_offset >> 9) + index_in_cluster,
                             n, &hd_qiov);
        qemu_co_mutex_lock(&s->lock);
@@ -759,8 +703,6 @@ static void qcow_close(BlockDriverState *bs)
 {
    BDRVQcowState *s = bs->opaque;

-    qcrypto_cipher_free(s->cipher);
-    s->cipher = NULL;
    g_free(s->l1_table);
    qemu_vfree(s->l2_cache);
    g_free(s->cluster_cache);
@@ -798,7 +740,7 @@ static int qcow_create(const char *filename, QemuOpts *opts, Error **errp)

    qcow_bs = NULL;
    ret = bdrv_open(&qcow_bs, filename, NULL, NULL,
-                    BDRV_O_RDWR | BDRV_O_PROTOCOL, &local_err);
+                    BDRV_O_RDWR | BDRV_O_PROTOCOL, NULL, &local_err);
    if (ret < 0) {
        error_propagate(errp, local_err);
        goto cleanup;
@@ -884,10 +826,10 @@ static int qcow_make_empty(BlockDriverState *bs)
    int ret;

    memset(s->l1_table, 0, l1_length);
-    if (bdrv_pwrite_sync(bs->file->bs, s->l1_table_offset, s->l1_table,
+    if (bdrv_pwrite_sync(bs->file, s->l1_table_offset, s->l1_table,
            l1_length) < 0)
        return -1;
-    ret = bdrv_truncate(bs->file->bs, s->l1_table_offset + l1_length);
+    ret = bdrv_truncate(bs->file, s->l1_table_offset + l1_length);
    if (ret < 0)
        return ret;

@@ -967,7 +909,7 @@ static int qcow_write_compressed(BlockDriverState *bs, int64_t sector_num,
        }

        cluster_offset &= s->cluster_offset_mask;
-        ret = bdrv_pwrite(bs->file->bs, cluster_offset, out_buf, out_len);
+        ret = bdrv_pwrite(bs->file, cluster_offset, out_buf, out_len);
        if (ret < 0) {
            goto fail;
        }
--- a/block/qcow2-cache.c
+++ b/block/qcow2-cache.c
@@ -22,132 +22,68 @@
 * THE SOFTWARE.
 */

-/* Needed for CONFIG_MADVISE */
-#include "qemu/osdep.h"
-
-#if defined(CONFIG_MADVISE) || defined(CONFIG_POSIX_MADVISE)
-#include <sys/mman.h>
-#endif
-
 #include "block/block_int.h"
 #include "qemu-common.h"
 #include "qcow2.h"
 #include "trace.h"

 typedef struct Qcow2CachedTable {
-    int64_t  offset;
-    uint64_t lru_counter;
-    int      ref;
-    bool     dirty;
+    void*   table;
+    int64_t offset;
+    bool    dirty;
+    int     cache_hits;
+    int     ref;
 } Qcow2CachedTable;

 struct Qcow2Cache {
-    Qcow2CachedTable       *entries;
-    struct Qcow2Cache      *depends;
+    Qcow2CachedTable*       entries;
+    struct Qcow2Cache*      depends;
    int                     size;
    bool                    depends_on_flush;
-    void                   *table_array;
-    uint64_t                lru_counter;
-    uint64_t                cache_clean_lru_counter;
 };

-static inline void *qcow2_cache_get_table_addr(BlockDriverState *bs,
-                    Qcow2Cache *c, int table)
-{
-    BDRVQcow2State *s = bs->opaque;
-    return (uint8_t *) c->table_array + (size_t) table * s->cluster_size;
-}
-
-static inline int qcow2_cache_get_table_idx(BlockDriverState *bs,
-                  Qcow2Cache *c, void *table)
-{
-    BDRVQcow2State *s = bs->opaque;
-    ptrdiff_t table_offset = (uint8_t *) table - (uint8_t *) c->table_array;
-    int idx = table_offset / s->cluster_size;
-    assert(idx >= 0 && idx < c->size && table_offset % s->cluster_size == 0);
-    return idx;
-}
-
-static void qcow2_cache_table_release(BlockDriverState *bs, Qcow2Cache *c,
-                                      int i, int num_tables)
-{
-#if QEMU_MADV_DONTNEED != QEMU_MADV_INVALID
-    BDRVQcow2State *s = bs->opaque;
-    void *t = qcow2_cache_get_table_addr(bs, c, i);
-    int align = getpagesize();
-    size_t mem_size = (size_t) s->cluster_size * num_tables;
-    size_t offset = QEMU_ALIGN_UP((uintptr_t) t, align) - (uintptr_t) t;
-    size_t length = QEMU_ALIGN_DOWN(mem_size - offset, align);
-    if (length > 0) {
-        qemu_madvise((uint8_t *) t + offset, length, QEMU_MADV_DONTNEED);
-    }
-#endif
-}
-
-static inline bool can_clean_entry(Qcow2Cache *c, int i)
-{
-    Qcow2CachedTable *t = &c->entries[i];
-    return t->ref == 0 && !t->dirty && t->offset != 0 &&
-        t->lru_counter <= c->cache_clean_lru_counter;
-}
-
-void qcow2_cache_clean_unused(BlockDriverState *bs, Qcow2Cache *c)
-{
-    int i = 0;
-    while (i < c->size) {
-        int to_clean = 0;
-
-        /* Skip the entries that we don't need to clean */
-        while (i < c->size && !can_clean_entry(c, i)) {
-            i++;
-        }
-
-        /* And count how many we can clean in a row */
-        while (i < c->size && can_clean_entry(c, i)) {
-            c->entries[i].offset = 0;
-            c->entries[i].lru_counter = 0;
-            i++;
-            to_clean++;
-        }
-
-        if (to_clean > 0) {
-            qcow2_cache_table_release(bs, c, i - to_clean, to_clean);
-        }
-    }
-
-    c->cache_clean_lru_counter = c->lru_counter;
-}
-
 Qcow2Cache *qcow2_cache_create(BlockDriverState *bs, int num_tables)
 {
-    BDRVQcow2State *s = bs->opaque;
+    BDRVQcowState *s = bs->opaque;
    Qcow2Cache *c;
+    int i;

    c = g_new0(Qcow2Cache, 1);
    c->size = num_tables;
    c->entries = g_try_new0(Qcow2CachedTable, num_tables);
-    c->table_array = qemu_try_blockalign(bs->file->bs,
-                                         (size_t) num_tables * s->cluster_size);
+    if (!c->entries) {
+        goto fail;
+    }

-    if (!c->entries || !c->table_array) {
-        qemu_vfree(c->table_array);
-        g_free(c->entries);
-        g_free(c);
-        c = NULL;
+    for (i = 0; i < c->size; i++) {
+        c->entries[i].table = qemu_try_blockalign(bs->file, s->cluster_size);
+        if (c->entries[i].table == NULL) {
+            goto fail;
+        }
    }

    return c;
+
+fail:
+    if (c->entries) {
+        for (i = 0; i < c->size; i++) {
+            qemu_vfree(c->entries[i].table);
+        }
+    }
+    g_free(c->entries);
+    g_free(c);
+    return NULL;
 }

-int qcow2_cache_destroy(BlockDriverState *bs, Qcow2Cache *c)
+int qcow2_cache_destroy(BlockDriverState* bs, Qcow2Cache *c)
 {
    int i;

    for (i = 0; i < c->size; i++) {
        assert(c->entries[i].ref == 0);
+        qemu_vfree(c->entries[i].table);
    }

-    qemu_vfree(c->table_array);
    g_free(c->entries);
    g_free(c);

@@ -171,7 +107,7 @@ static int qcow2_cache_flush_dependency(BlockDriverState *bs, Qcow2Cache *c)

 static int qcow2_cache_entry_flush(BlockDriverState *bs, Qcow2Cache *c, int i)
 {
-    BDRVQcow2State *s = bs->opaque;
+    BDRVQcowState *s = bs->opaque;
    int ret = 0;

    if (!c->entries[i].dirty || !c->entries[i].offset) {
@@ -184,7 +120,7 @@ static int qcow2_cache_entry_flush(BlockDriverState *bs, Qcow2Cache *c, int i)
    if (c->depends) {
        ret = qcow2_cache_flush_dependency(bs, c);
    } else if (c->depends_on_flush) {
-        ret = bdrv_flush(bs->file->bs);
+        ret = bdrv_flush(bs->file);
        if (ret >= 0) {
            c->depends_on_flush = false;
        }
@@ -215,8 +151,8 @@ static int qcow2_cache_entry_flush(BlockDriverState *bs, Qcow2Cache *c, int i)
        BLKDBG_EVENT(bs->file, BLKDBG_L2_UPDATE);
    }

-    ret = bdrv_pwrite(bs->file->bs, c->entries[i].offset,
-                      qcow2_cache_get_table_addr(bs, c, i), s->cluster_size);
+    ret = bdrv_pwrite(bs->file, c->entries[i].offset, c->entries[i].table,
+        s->cluster_size);
    if (ret < 0) {
        return ret;
    }
@@ -228,7 +164,7 @@ static int qcow2_cache_entry_flush(BlockDriverState *bs, Qcow2Cache *c, int i)

 int qcow2_cache_flush(BlockDriverState *bs, Qcow2Cache *c)
 {
-    BDRVQcow2State *s = bs->opaque;
+    BDRVQcowState *s = bs->opaque;
    int result = 0;
    int ret;
    int i;
@@ -243,7 +179,7 @@ int qcow2_cache_flush(BlockDriverState *bs, Qcow2Cache *c)
    }

    if (result == 0) {
-        ret = bdrv_flush(bs->file->bs);
+        ret = bdrv_flush(bs->file);
        if (ret < 0) {
            result = ret;
        }
@@ -292,55 +228,68 @@ int qcow2_cache_empty(BlockDriverState *bs, Qcow2Cache *c)
    for (i = 0; i < c->size; i++) {
        assert(c->entries[i].ref == 0);
        c->entries[i].offset = 0;
-        c->entries[i].lru_counter = 0;
+        c->entries[i].cache_hits = 0;
    }

-    qcow2_cache_table_release(bs, c, 0, c->size);
-
-    c->lru_counter = 0;
-
    return 0;
 }

+static int qcow2_cache_find_entry_to_replace(Qcow2Cache *c)
+{
+    int i;
+    int min_count = INT_MAX;
+    int min_index = -1;
+
+
+    for (i = 0; i < c->size; i++) {
+        if (c->entries[i].ref) {
+            continue;
+        }
+
+        if (c->entries[i].cache_hits < min_count) {
+            min_index = i;
+            min_count = c->entries[i].cache_hits;
+        }
+
+        /* Give newer hits priority */
+        /* TODO Check how to optimize the replacement strategy */
+        if (c->entries[i].cache_hits > 1) {
+            c->entries[i].cache_hits /= 2;
+        }
+    }
+
+    if (min_index == -1) {
+        /* This can't happen in current synchronous code, but leave the check
+         * here as a reminder for whoever starts using AIO with the cache */
+        abort();
+    }
+    return min_index;
+}
+
 static int qcow2_cache_do_get(BlockDriverState *bs, Qcow2Cache *c,
    uint64_t offset, void **table, bool read_from_disk)
 {
-    BDRVQcow2State *s = bs->opaque;
+    BDRVQcowState *s = bs->opaque;
    int i;
    int ret;
-    int lookup_index;
-    uint64_t min_lru_counter = UINT64_MAX;
-    int min_lru_index = -1;

    trace_qcow2_cache_get(qemu_coroutine_self(), c == s->l2_table_cache,
                          offset, read_from_disk);

    /* Check if the table is already cached */
-    i = lookup_index = (offset / s->cluster_size * 4) % c->size;
-    do {
-        const Qcow2CachedTable *t = &c->entries[i];
-        if (t->offset == offset) {
+    for (i = 0; i < c->size; i++) {
+        if (c->entries[i].offset == offset) {
            goto found;
        }
-        if (t->ref == 0 && t->lru_counter < min_lru_counter) {
-            min_lru_counter = t->lru_counter;
-            min_lru_index = i;
-        }
-        if (++i == c->size) {
-            i = 0;
-        }
-    } while (i != lookup_index);
-
-    if (min_lru_index == -1) {
-        /* This can't happen in current synchronous code, but leave the check
-         * here as a reminder for whoever starts using AIO with the cache */
-        abort();
    }

-    /* Cache miss: write a table back and replace it */
-    i = min_lru_index;
+    /* If not, write a table back and replace it */
+    i = qcow2_cache_find_entry_to_replace(c);
    trace_qcow2_cache_get_replace_entry(qemu_coroutine_self(),
                                        c == s->l2_table_cache, i);
+    if (i < 0) {
+        return i;
+    }

    ret = qcow2_cache_entry_flush(bs, c, i);
    if (ret < 0) {
@@ -355,20 +304,22 @@ static int qcow2_cache_do_get(BlockDriverState *bs, Qcow2Cache *c,
            BLKDBG_EVENT(bs->file, BLKDBG_L2_LOAD);
        }

-        ret = bdrv_pread(bs->file->bs, offset,
-                         qcow2_cache_get_table_addr(bs, c, i),
-                         s->cluster_size);
+        ret = bdrv_pread(bs->file, offset, c->entries[i].table, s->cluster_size);
        if (ret < 0) {
            return ret;
        }
    }

+    /* Give the table some hits for the start so that it won't be replaced
+     * immediately. The number 32 is completely arbitrary. */
+    c->entries[i].cache_hits = 32;
    c->entries[i].offset = offset;

    /* And return the right table */
 found:
+    c->entries[i].cache_hits++;
    c->entries[i].ref++;
-    *table = qcow2_cache_get_table_addr(bs, c, i);
+    *table = c->entries[i].table;

    trace_qcow2_cache_get_done(qemu_coroutine_self(),
                               c == s->l2_table_cache, i);
@@ -388,24 +339,36 @@ int qcow2_cache_get_empty(BlockDriverState *bs, Qcow2Cache *c, uint64_t offset,
    return qcow2_cache_do_get(bs, c, offset, table, false);
 }

-void qcow2_cache_put(BlockDriverState *bs, Qcow2Cache *c, void **table)
+int qcow2_cache_put(BlockDriverState *bs, Qcow2Cache *c, void **table)
 {
-    int i = qcow2_cache_get_table_idx(bs, c, *table);
+    int i;

+    for (i = 0; i < c->size; i++) {
+        if (c->entries[i].table == *table) {
+            goto found;
+        }
+    }
+    return -ENOENT;
+
+found:
    c->entries[i].ref--;
    *table = NULL;

-    if (c->entries[i].ref == 0) {
-        c->entries[i].lru_counter = ++c->lru_counter;
-    }
-
    assert(c->entries[i].ref >= 0);
+    return 0;
 }

-void qcow2_cache_entry_mark_dirty(BlockDriverState *bs, Qcow2Cache *c,
-     void *table)
+void qcow2_cache_entry_mark_dirty(Qcow2Cache *c, void *table)
 {
-    int i = qcow2_cache_get_table_idx(bs, c, table);
-    assert(c->entries[i].offset != 0);
+    int i;
+
+    for (i = 0; i < c->size; i++) {
+        if (c->entries[i].table == table) {
+            goto found;
+        }
+    }
+    abort();
+
+found:
    c->entries[i].dirty = true;
 }
--- a/block/qcow2-cluster.c
+++ b/block/qcow2-cluster.c
@@ -22,7 +22,6 @@
 * THE SOFTWARE.
 */

-#include "qemu/osdep.h"
 #include <zlib.h>

 #include "qemu-common.h"
@@ -33,7 +32,7 @@
 int qcow2_grow_l1_table(BlockDriverState *bs, uint64_t min_size,
                        bool exact_size)
 {
-    BDRVQcow2State *s = bs->opaque;
+    BDRVQcowState *s = bs->opaque;
    int new_l1_size2, ret, i;
    uint64_t *new_l1_table;
    int64_t old_l1_table_offset, old_l1_size;
@@ -73,7 +72,7 @@ int qcow2_grow_l1_table(BlockDriverState *bs, uint64_t min_size,
 #endif

    new_l1_size2 = sizeof(uint64_t) * new_l1_size;
-    new_l1_table = qemu_try_blockalign(bs->file->bs,
+    new_l1_table = qemu_try_blockalign(bs->file,
                                       align_offset(new_l1_size2, 512));
    if (new_l1_table == NULL) {
        return -ENOMEM;
@@ -106,8 +105,7 @@ int qcow2_grow_l1_table(BlockDriverState *bs, uint64_t min_size,
    BLKDBG_EVENT(bs->file, BLKDBG_L1_GROW_WRITE_TABLE);
    for(i = 0; i < s->l1_size; i++)
        new_l1_table[i] = cpu_to_be64(new_l1_table[i]);
-    ret = bdrv_pwrite_sync(bs->file->bs, new_l1_table_offset,
-                           new_l1_table, new_l1_size2);
+    ret = bdrv_pwrite_sync(bs->file, new_l1_table_offset, new_l1_table, new_l1_size2);
    if (ret < 0)
        goto fail;
    for(i = 0; i < s->l1_size; i++)
@@ -117,8 +115,7 @@ int qcow2_grow_l1_table(BlockDriverState *bs, uint64_t min_size,
    BLKDBG_EVENT(bs->file, BLKDBG_L1_GROW_ACTIVATE_TABLE);
    cpu_to_be32w((uint32_t*)data, new_l1_size);
    stq_be_p(data + 4, new_l1_table_offset);
-    ret = bdrv_pwrite_sync(bs->file->bs, offsetof(QCowHeader, l1_size),
-                           data, sizeof(data));
+    ret = bdrv_pwrite_sync(bs->file, offsetof(QCowHeader, l1_size), data,sizeof(data));
    if (ret < 0) {
        goto fail;
    }
@@ -151,7 +148,7 @@ int qcow2_grow_l1_table(BlockDriverState *bs, uint64_t min_size,
 static int l2_load(BlockDriverState *bs, uint64_t l2_offset,
    uint64_t **l2_table)
 {
-    BDRVQcow2State *s = bs->opaque;
+    BDRVQcowState *s = bs->opaque;
    int ret;

    ret = qcow2_cache_get(bs, s->l2_table_cache, l2_offset, (void**) l2_table);
@@ -166,7 +163,7 @@ static int l2_load(BlockDriverState *bs, uint64_t l2_offset,
 #define L1_ENTRIES_PER_SECTOR (512 / 8)
 int qcow2_write_l1_entry(BlockDriverState *bs, int l1_index)
 {
-    BDRVQcow2State *s = bs->opaque;
+    BDRVQcowState *s = bs->opaque;
    uint64_t buf[L1_ENTRIES_PER_SECTOR] = { 0 };
    int l1_start_index;
    int i, ret;
@@ -185,9 +182,8 @@ int qcow2_write_l1_entry(BlockDriverState *bs, int l1_index)
    }

    BLKDBG_EVENT(bs->file, BLKDBG_L1_UPDATE);
-    ret = bdrv_pwrite_sync(bs->file->bs,
-                           s->l1_table_offset + 8 * l1_start_index,
-                           buf, sizeof(buf));
+    ret = bdrv_pwrite_sync(bs->file, s->l1_table_offset + 8 * l1_start_index,
+        buf, sizeof(buf));
    if (ret < 0) {
        return ret;
    }
@@ -207,7 +203,7 @@ int qcow2_write_l1_entry(BlockDriverState *bs, int l1_index)

 static int l2_allocate(BlockDriverState *bs, int l1_index, uint64_t **table)
 {
-    BDRVQcow2State *s = bs->opaque;
+    BDRVQcowState *s = bs->opaque;
    uint64_t old_l2_offset;
    uint64_t *l2_table = NULL;
    int64_t l2_offset;
@@ -257,14 +253,17 @@ static int l2_allocate(BlockDriverState *bs, int l1_index, uint64_t **table)

        memcpy(l2_table, old_table, s->cluster_size);

-        qcow2_cache_put(bs, s->l2_table_cache, (void **) &old_table);
+        ret = qcow2_cache_put(bs, s->l2_table_cache, (void**) &old_table);
+        if (ret < 0) {
+            goto fail;
+        }
    }

    /* write the l2 table to the file */
    BLKDBG_EVENT(bs->file, BLKDBG_L2_ALLOC_WRITE);

    trace_qcow2_l2_allocate_write_l2(bs, l1_index);
-    qcow2_cache_entry_mark_dirty(bs, s->l2_table_cache, l2_table);
+    qcow2_cache_entry_mark_dirty(s->l2_table_cache, l2_table);
    ret = qcow2_cache_flush(bs, s->l2_table_cache);
    if (ret < 0) {
        goto fail;
@@ -302,7 +301,7 @@ fail:
 * as contiguous. (This allows it, for example, to stop at the first compressed
 * cluster which may require a different handling)
 */
-static int count_contiguous_clusters(int nb_clusters, int cluster_size,
+static int count_contiguous_clusters(uint64_t nb_clusters, int cluster_size,
        uint64_t *l2_table, uint64_t stop_flags)
 {
    int i;
@@ -313,7 +312,7 @@ static int count_contiguous_clusters(int nb_clusters, int cluster_size,
    if (!offset)
        return 0;

-    assert(qcow2_get_cluster_type(first_entry) == QCOW2_CLUSTER_NORMAL);
+    assert(qcow2_get_cluster_type(first_entry) != QCOW2_CLUSTER_COMPRESSED);

    for (i = 0; i < nb_clusters; i++) {
        uint64_t l2_entry = be64_to_cpu(l2_table[i]) & mask;
@@ -325,16 +324,14 @@ static int count_contiguous_clusters(int nb_clusters, int cluster_size,
 	return i;
 }

-static int count_contiguous_clusters_by_type(int nb_clusters,
-                                             uint64_t *l2_table,
-                                             int wanted_type)
+static int count_contiguous_free_clusters(uint64_t nb_clusters, uint64_t *l2_table)
 {
    int i;

    for (i = 0; i < nb_clusters; i++) {
        int type = qcow2_get_cluster_type(be64_to_cpu(l2_table[i]));

-        if (type != wanted_type) {
+        if (type != QCOW2_CLUSTER_UNALLOCATED) {
            break;
        }
    }
@@ -345,47 +342,26 @@ static int count_contiguous_clusters_by_type(int nb_clusters,
 /* The crypt function is compatible with the linux cryptoloop
   algorithm for < 4 GB images. NOTE: out_buf == in_buf is
   supported */
-int qcow2_encrypt_sectors(BDRVQcow2State *s, int64_t sector_num,
-                          uint8_t *out_buf, const uint8_t *in_buf,
-                          int nb_sectors, bool enc,
-                          Error **errp)
+void qcow2_encrypt_sectors(BDRVQcowState *s, int64_t sector_num,
+                           uint8_t *out_buf, const uint8_t *in_buf,
+                           int nb_sectors, int enc,
+                           const AES_KEY *key)
 {
    union {
        uint64_t ll[2];
        uint8_t b[16];
    } ivec;
    int i;
-    int ret;

    for(i = 0; i < nb_sectors; i++) {
        ivec.ll[0] = cpu_to_le64(sector_num);
        ivec.ll[1] = 0;
-        if (qcrypto_cipher_setiv(s->cipher,
-                                 ivec.b, G_N_ELEMENTS(ivec.b),
-                                 errp) < 0) {
-            return -1;
-        }
-        if (enc) {
-            ret = qcrypto_cipher_encrypt(s->cipher,
-                                         in_buf,
-                                         out_buf,
-                                         512,
-                                         errp);
-        } else {
-            ret = qcrypto_cipher_decrypt(s->cipher,
-                                         in_buf,
-                                         out_buf,
-                                         512,
-                                         errp);
-        }
-        if (ret < 0) {
-            return -1;
-        }
+        AES_cbc_encrypt(in_buf, out_buf, 512, key,
+                        ivec.b, enc);
        sector_num++;
        in_buf += 512;
        out_buf += 512;
    }
-    return 0;
 }

 static int coroutine_fn copy_sectors(BlockDriverState *bs,
@@ -393,7 +369,7 @@ static int coroutine_fn copy_sectors(BlockDriverState *bs,
                                     uint64_t cluster_offset,
                                     int n_start, int n_end)
 {
-    BDRVQcow2State *s = bs->opaque;
+    BDRVQcowState *s = bs->opaque;
    QEMUIOVector qiov;
    struct iovec iov;
    int n, ret;
@@ -427,16 +403,10 @@ static int coroutine_fn copy_sectors(BlockDriverState *bs,
        goto out;
    }

-    if (bs->encrypted) {
-        Error *err = NULL;
-        assert(s->cipher);
-        if (qcow2_encrypt_sectors(s, start_sect + n_start,
-                                  iov.iov_base, iov.iov_base, n,
-                                  true, &err) < 0) {
-            ret = -EIO;
-            error_free(err);
-            goto out;
-        }
+    if (s->crypt_method) {
+        qcow2_encrypt_sectors(s, start_sect + n_start,
+                        iov.iov_base, iov.iov_base, n, 1,
+                        &s->aes_encrypt_key);
    }

    ret = qcow2_pre_write_overlap_check(bs, 0,
@@ -446,8 +416,7 @@ static int coroutine_fn copy_sectors(BlockDriverState *bs,
    }

    BLKDBG_EVENT(bs->file, BLKDBG_COW_WRITE);
-    ret = bdrv_co_writev(bs->file->bs, (cluster_offset >> 9) + n_start, n,
-                         &qiov);
+    ret = bdrv_co_writev(bs->file, (cluster_offset >> 9) + n_start, n, &qiov);
    if (ret < 0) {
        goto out;
    }
@@ -476,7 +445,7 @@ out:
 int qcow2_get_cluster_offset(BlockDriverState *bs, uint64_t offset,
    int *num, uint64_t *cluster_offset)
 {
-    BDRVQcow2State *s = bs->opaque;
+    BDRVQcowState *s = bs->opaque;
    unsigned int l2_index;
    uint64_t l1_index, l2_offset, *l2_table;
    int l1_bits, c;
@@ -502,11 +471,10 @@ int qcow2_get_cluster_offset(BlockDriverState *bs, uint64_t offset,
    if (nb_needed > nb_available) {
        nb_needed = nb_available;
    }
-    assert(nb_needed <= INT_MAX);

    *cluster_offset = 0;

-    /* seek to the l2 offset in the l1 table */
+    /* seek the the l2 offset in the l1 table */

    l1_index = offset >> l1_bits;
    if (l1_index >= s->l1_size) {
@@ -538,8 +506,6 @@ int qcow2_get_cluster_offset(BlockDriverState *bs, uint64_t offset,

    l2_index = (offset >> s->cluster_bits) & (s->l2_size - 1);
    *cluster_offset = be64_to_cpu(l2_table[l2_index]);
-
-    /* nb_needed <= INT_MAX, thus nb_clusters <= INT_MAX, too */
    nb_clusters = size_to_clusters(s, nb_needed << 9);

    ret = qcow2_get_cluster_type(*cluster_offset);
@@ -557,14 +523,13 @@ int qcow2_get_cluster_offset(BlockDriverState *bs, uint64_t offset,
            ret = -EIO;
            goto fail;
        }
-        c = count_contiguous_clusters_by_type(nb_clusters, &l2_table[l2_index],
-                                              QCOW2_CLUSTER_ZERO);
+        c = count_contiguous_clusters(nb_clusters, s->cluster_size,
+                &l2_table[l2_index], QCOW_OFLAG_ZERO);
        *cluster_offset = 0;
        break;
    case QCOW2_CLUSTER_UNALLOCATED:
        /* how many empty clusters ? */
-        c = count_contiguous_clusters_by_type(nb_clusters, &l2_table[l2_index],
-                                              QCOW2_CLUSTER_UNALLOCATED);
+        c = count_contiguous_free_clusters(nb_clusters, &l2_table[l2_index]);
        *cluster_offset = 0;
        break;
    case QCOW2_CLUSTER_NORMAL:
@@ -617,13 +582,13 @@ static int get_cluster_table(BlockDriverState *bs, uint64_t offset,
                             uint64_t **new_l2_table,
                             int *new_l2_index)
 {
-    BDRVQcow2State *s = bs->opaque;
+    BDRVQcowState *s = bs->opaque;
    unsigned int l2_index;
    uint64_t l1_index, l2_offset;
    uint64_t *l2_table = NULL;
    int ret;

-    /* seek to the l2 offset in the l1 table */
+    /* seek the the l2 offset in the l1 table */

    l1_index = offset >> (s->l2_bits + s->cluster_bits);
    if (l1_index >= s->l1_size) {
@@ -691,7 +656,7 @@ uint64_t qcow2_alloc_compressed_cluster_offset(BlockDriverState *bs,
                                               uint64_t offset,
                                               int compressed_size)
 {
-    BDRVQcow2State *s = bs->opaque;
+    BDRVQcowState *s = bs->opaque;
    int l2_index, ret;
    uint64_t *l2_table;
    int64_t cluster_offset;
@@ -727,16 +692,19 @@ uint64_t qcow2_alloc_compressed_cluster_offset(BlockDriverState *bs,
    /* compressed clusters never have the copied flag */

    BLKDBG_EVENT(bs->file, BLKDBG_L2_UPDATE_COMPRESSED);
-    qcow2_cache_entry_mark_dirty(bs, s->l2_table_cache, l2_table);
+    qcow2_cache_entry_mark_dirty(s->l2_table_cache, l2_table);
    l2_table[l2_index] = cpu_to_be64(cluster_offset);
-    qcow2_cache_put(bs, s->l2_table_cache, (void **) &l2_table);
+    ret = qcow2_cache_put(bs, s->l2_table_cache, (void**) &l2_table);
+    if (ret < 0) {
+        return 0;
+    }

    return cluster_offset;
 }

 static int perform_cow(BlockDriverState *bs, QCowL2Meta *m, Qcow2COWRegion *r)
 {
-    BDRVQcow2State *s = bs->opaque;
+    BDRVQcowState *s = bs->opaque;
    int ret;

    if (r->nb_sectors == 0) {
@@ -765,7 +733,7 @@ static int perform_cow(BlockDriverState *bs, QCowL2Meta *m, Qcow2COWRegion *r)

 int qcow2_alloc_cluster_link_l2(BlockDriverState *bs, QCowL2Meta *m)
 {
-    BDRVQcow2State *s = bs->opaque;
+    BDRVQcowState *s = bs->opaque;
    int i, j = 0, l2_index, ret;
    uint64_t *old_cluster, *l2_table;
    uint64_t cluster_offset = m->alloc_offset;
@@ -803,7 +771,7 @@ int qcow2_alloc_cluster_link_l2(BlockDriverState *bs, QCowL2Meta *m)
    if (ret < 0) {
        goto err;
    }
-    qcow2_cache_entry_mark_dirty(bs, s->l2_table_cache, l2_table);
+    qcow2_cache_entry_mark_dirty(s->l2_table_cache, l2_table);

    assert(l2_index + m->nb_clusters <= s->l2_size);
    for (i = 0; i < m->nb_clusters; i++) {
@@ -821,10 +789,14 @@ int qcow2_alloc_cluster_link_l2(BlockDriverState *bs, QCowL2Meta *m)
     }


-    qcow2_cache_put(bs, s->l2_table_cache, (void **) &l2_table);
+    ret = qcow2_cache_put(bs, s->l2_table_cache, (void**) &l2_table);
+    if (ret < 0) {
+        goto err;
+    }

    /*
     * If this was a COW, we need to decrease the refcount of the old cluster.
+     * Also flush bs->file to get the right order for L2 and refcount update.
     *
     * Don't discard clusters that reach a refcount of 0 (e.g. compressed
     * clusters), the next write will reuse them anyway.
@@ -847,7 +819,7 @@ err:
 * write, but require COW to be performed (this includes yet unallocated space,
 * which must copy from the backing file)
 */
-static int count_cow_clusters(BDRVQcow2State *s, int nb_clusters,
+static int count_cow_clusters(BDRVQcowState *s, int nb_clusters,
    uint64_t *l2_table, int l2_index)
 {
    int i;
@@ -893,7 +865,7 @@ out:
 static int handle_dependencies(BlockDriverState *bs, uint64_t guest_offset,
    uint64_t *cur_bytes, QCowL2Meta **m)
 {
-    BDRVQcow2State *s = bs->opaque;
+    BDRVQcowState *s = bs->opaque;
    QCowL2Meta *old_alloc;
    uint64_t bytes = *cur_bytes;

@@ -966,13 +938,13 @@ static int handle_dependencies(BlockDriverState *bs, uint64_t guest_offset,
 static int handle_copied(BlockDriverState *bs, uint64_t guest_offset,
    uint64_t *host_offset, uint64_t *bytes, QCowL2Meta **m)
 {
-    BDRVQcow2State *s = bs->opaque;
+    BDRVQcowState *s = bs->opaque;
    int l2_index;
    uint64_t cluster_offset;
    uint64_t *l2_table;
-    uint64_t nb_clusters;
+    unsigned int nb_clusters;
    unsigned int keep_clusters;
-    int ret;
+    int ret, pret;

    trace_qcow2_handle_copied(qemu_coroutine_self(), guest_offset, *host_offset,
                              *bytes);
@@ -989,7 +961,6 @@ static int handle_copied(BlockDriverState *bs, uint64_t guest_offset,

    l2_index = offset_to_l2_index(s, guest_offset);
    nb_clusters = MIN(nb_clusters, s->l2_size - l2_index);
-    assert(nb_clusters <= INT_MAX);

    /* Find L2 entry for the first involved cluster */
    ret = get_cluster_table(bs, guest_offset, &l2_table, &l2_index);
@@ -1040,7 +1011,10 @@ static int handle_copied(BlockDriverState *bs, uint64_t guest_offset,

    /* Cleanup */
 out:
-    qcow2_cache_put(bs, s->l2_table_cache, (void **) &l2_table);
+    pret = qcow2_cache_put(bs, s->l2_table_cache, (void**) &l2_table);
+    if (pret < 0) {
+        return pret;
+    }

    /* Only return a host offset if we actually made progress. Otherwise we
     * would make requirements for handle_alloc() that it can't fulfill */
@@ -1072,9 +1046,9 @@ out:
 * restarted, but the whole request should not be failed.
 */
 static int do_alloc_cluster_offset(BlockDriverState *bs, uint64_t guest_offset,
-                                   uint64_t *host_offset, uint64_t *nb_clusters)
+    uint64_t *host_offset, unsigned int *nb_clusters)
 {
-    BDRVQcow2State *s = bs->opaque;
+    BDRVQcowState *s = bs->opaque;

    trace_qcow2_do_alloc_clusters_offset(qemu_coroutine_self(), guest_offset,
                                         *host_offset, *nb_clusters);
@@ -1090,7 +1064,7 @@ static int do_alloc_cluster_offset(BlockDriverState *bs, uint64_t guest_offset,
        *host_offset = cluster_offset;
        return 0;
    } else {
-        int64_t ret = qcow2_alloc_clusters_at(bs, *host_offset, *nb_clusters);
+        int ret = qcow2_alloc_clusters_at(bs, *host_offset, *nb_clusters);
        if (ret < 0) {
            return ret;
        }
@@ -1122,11 +1096,11 @@ static int do_alloc_cluster_offset(BlockDriverState *bs, uint64_t guest_offset,
 static int handle_alloc(BlockDriverState *bs, uint64_t guest_offset,
    uint64_t *host_offset, uint64_t *bytes, QCowL2Meta **m)
 {
-    BDRVQcow2State *s = bs->opaque;
+    BDRVQcowState *s = bs->opaque;
    int l2_index;
    uint64_t *l2_table;
    uint64_t entry;
-    uint64_t nb_clusters;
+    unsigned int nb_clusters;
    int ret;

    uint64_t alloc_cluster_offset;
@@ -1144,7 +1118,6 @@ static int handle_alloc(BlockDriverState *bs, uint64_t guest_offset,

    l2_index = offset_to_l2_index(s, guest_offset);
    nb_clusters = MIN(nb_clusters, s->l2_size - l2_index);
-    assert(nb_clusters <= INT_MAX);

    /* Find L2 entry for the first involved cluster */
    ret = get_cluster_table(bs, guest_offset, &l2_table, &l2_index);
@@ -1166,7 +1139,10 @@ static int handle_alloc(BlockDriverState *bs, uint64_t guest_offset,
     * wrong with our code. */
    assert(nb_clusters > 0);

-    qcow2_cache_put(bs, s->l2_table_cache, (void **) &l2_table);
+    ret = qcow2_cache_put(bs, s->l2_table_cache, (void**) &l2_table);
+    if (ret < 0) {
+        return ret;
+    }

    /* Allocate, if necessary at a given offset in the image file */
    alloc_cluster_offset = start_of_cluster(s, *host_offset);
@@ -1275,7 +1251,7 @@ fail:
 int qcow2_alloc_cluster_offset(BlockDriverState *bs, uint64_t offset,
    int *num, uint64_t *host_offset, QCowL2Meta **m)
 {
-    BDRVQcow2State *s = bs->opaque;
+    BDRVQcowState *s = bs->opaque;
    uint64_t start, remaining;
    uint64_t cluster_offset;
    uint64_t cur_bytes;
@@ -1409,7 +1385,7 @@ static int decompress_buffer(uint8_t *out_buf, int out_buf_size,

 int qcow2_decompress_cluster(BlockDriverState *bs, uint64_t cluster_offset)
 {
-    BDRVQcow2State *s = bs->opaque;
+    BDRVQcowState *s = bs->opaque;
    int ret, csize, nb_csectors, sector_offset;
    uint64_t coffset;

@@ -1419,8 +1395,7 @@ int qcow2_decompress_cluster(BlockDriverState *bs, uint64_t cluster_offset)
        sector_offset = coffset & 511;
        csize = nb_csectors * 512 - sector_offset;
        BLKDBG_EVENT(bs->file, BLKDBG_READ_COMPRESSED);
-        ret = bdrv_read(bs->file->bs, coffset >> 9, s->cluster_data,
-                        nb_csectors);
+        ret = bdrv_read(bs->file, coffset >> 9, s->cluster_data, nb_csectors);
        if (ret < 0) {
            return ret;
        }
@@ -1439,10 +1414,9 @@ int qcow2_decompress_cluster(BlockDriverState *bs, uint64_t cluster_offset)
 * clusters.
 */
 static int discard_single_l2(BlockDriverState *bs, uint64_t offset,
-                             uint64_t nb_clusters, enum qcow2_discard_type type,
-                             bool full_discard)
+    unsigned int nb_clusters, enum qcow2_discard_type type, bool full_discard)
 {
-    BDRVQcow2State *s = bs->opaque;
+    BDRVQcowState *s = bs->opaque;
    uint64_t *l2_table;
    int l2_index;
    int ret;
@@ -1455,7 +1429,6 @@ static int discard_single_l2(BlockDriverState *bs, uint64_t offset,

    /* Limit nb_clusters to one L2 table */
    nb_clusters = MIN(nb_clusters, s->l2_size - l2_index);
-    assert(nb_clusters <= INT_MAX);

    for (i = 0; i < nb_clusters; i++) {
        uint64_t old_l2_entry;
@@ -1477,7 +1450,7 @@ static int discard_single_l2(BlockDriverState *bs, uint64_t offset,
         */
        switch (qcow2_get_cluster_type(old_l2_entry)) {
            case QCOW2_CLUSTER_UNALLOCATED:
-                if (full_discard || !bs->backing) {
+                if (full_discard || !bs->backing_hd) {
                    continue;
                }
                break;
@@ -1497,7 +1470,7 @@ static int discard_single_l2(BlockDriverState *bs, uint64_t offset,
        }

        /* First remove L2 entries */
-        qcow2_cache_entry_mark_dirty(bs, s->l2_table_cache, l2_table);
+        qcow2_cache_entry_mark_dirty(s->l2_table_cache, l2_table);
        if (!full_discard && s->qcow_version >= 3) {
            l2_table[l2_index + i] = cpu_to_be64(QCOW_OFLAG_ZERO);
        } else {
@@ -1508,7 +1481,10 @@ static int discard_single_l2(BlockDriverState *bs, uint64_t offset,
        qcow2_free_any_clusters(bs, old_l2_entry, 1, type);
    }

-    qcow2_cache_put(bs, s->l2_table_cache, (void **) &l2_table);
+    ret = qcow2_cache_put(bs, s->l2_table_cache, (void**) &l2_table);
+    if (ret < 0) {
+        return ret;
+    }

    return nb_clusters;
 }
@@ -1516,9 +1492,9 @@ static int discard_single_l2(BlockDriverState *bs, uint64_t offset,
 int qcow2_discard_clusters(BlockDriverState *bs, uint64_t offset,
    int nb_sectors, enum qcow2_discard_type type, bool full_discard)
 {
-    BDRVQcow2State *s = bs->opaque;
+    BDRVQcowState *s = bs->opaque;
    uint64_t end_offset;
-    uint64_t nb_clusters;
+    unsigned int nb_clusters;
    int ret;

    end_offset = offset + (nb_sectors << BDRV_SECTOR_BITS);
@@ -1560,9 +1536,9 @@ fail:
 * clusters.
 */
 static int zero_single_l2(BlockDriverState *bs, uint64_t offset,
-                          uint64_t nb_clusters)
+    unsigned int nb_clusters)
 {
-    BDRVQcow2State *s = bs->opaque;
+    BDRVQcowState *s = bs->opaque;
    uint64_t *l2_table;
    int l2_index;
    int ret;
@@ -1575,7 +1551,6 @@ static int zero_single_l2(BlockDriverState *bs, uint64_t offset,

    /* Limit nb_clusters to one L2 table */
    nb_clusters = MIN(nb_clusters, s->l2_size - l2_index);
-    assert(nb_clusters <= INT_MAX);

    for (i = 0; i < nb_clusters; i++) {
        uint64_t old_offset;
@@ -1583,7 +1558,7 @@ static int zero_single_l2(BlockDriverState *bs, uint64_t offset,
        old_offset = be64_to_cpu(l2_table[l2_index + i]);

        /* Update L2 entries */
-        qcow2_cache_entry_mark_dirty(bs, s->l2_table_cache, l2_table);
+        qcow2_cache_entry_mark_dirty(s->l2_table_cache, l2_table);
        if (old_offset & QCOW_OFLAG_COMPRESSED) {
            l2_table[l2_index + i] = cpu_to_be64(QCOW_OFLAG_ZERO);
            qcow2_free_any_clusters(bs, old_offset, 1, QCOW2_DISCARD_REQUEST);
@@ -1592,15 +1567,18 @@ static int zero_single_l2(BlockDriverState *bs, uint64_t offset,
        }
    }

-    qcow2_cache_put(bs, s->l2_table_cache, (void **) &l2_table);
+    ret = qcow2_cache_put(bs, s->l2_table_cache, (void**) &l2_table);
+    if (ret < 0) {
+        return ret;
+    }

    return nb_clusters;
 }

 int qcow2_zero_clusters(BlockDriverState *bs, uint64_t offset, int nb_sectors)
 {
-    BDRVQcow2State *s = bs->opaque;
-    uint64_t nb_clusters;
+    BDRVQcowState *s = bs->opaque;
+    unsigned int nb_clusters;
    int ret;

    /* The zero flag is only supported by version 3 and newer */
@@ -1642,10 +1620,9 @@ fail:
 static int expand_zero_clusters_in_l1(BlockDriverState *bs, uint64_t *l1_table,
                                      int l1_size, int64_t *visited_l1_entries,
                                      int64_t l1_entries,
-                                      BlockDriverAmendStatusCB *status_cb,
-                                      void *cb_opaque)
+                                      BlockDriverAmendStatusCB *status_cb)
 {
-    BDRVQcow2State *s = bs->opaque;
+    BDRVQcowState *s = bs->opaque;
    bool is_active_l1 = (l1_table == s->l1_table);
    uint64_t *l2_table = NULL;
    int ret;
@@ -1654,7 +1631,7 @@ static int expand_zero_clusters_in_l1(BlockDriverState *bs, uint64_t *l1_table,
    if (!is_active_l1) {
        /* inactive L2 tables require a buffer to be stored in when loading
         * them from disk */
-        l2_table = qemu_try_blockalign(bs->file->bs, s->cluster_size);
+        l2_table = qemu_try_blockalign(bs->file, s->cluster_size);
        if (l2_table == NULL) {
            return -ENOMEM;
        }
@@ -1669,7 +1646,7 @@ static int expand_zero_clusters_in_l1(BlockDriverState *bs, uint64_t *l1_table,
            /* unallocated */
            (*visited_l1_entries)++;
            if (status_cb) {
-                status_cb(bs, *visited_l1_entries, l1_entries, cb_opaque);
+                status_cb(bs, *visited_l1_entries, l1_entries);
            }
            continue;
        }
@@ -1688,8 +1665,8 @@ static int expand_zero_clusters_in_l1(BlockDriverState *bs, uint64_t *l1_table,
                    (void **)&l2_table);
        } else {
            /* load inactive L2 tables from disk */
-            ret = bdrv_read(bs->file->bs, l2_offset / BDRV_SECTOR_SIZE,
-                            (void *)l2_table, s->cluster_sectors);
+            ret = bdrv_read(bs->file, l2_offset / BDRV_SECTOR_SIZE,
+                    (void *)l2_table, s->cluster_sectors);
        }
        if (ret < 0) {
            goto fail;
@@ -1712,7 +1689,7 @@ static int expand_zero_clusters_in_l1(BlockDriverState *bs, uint64_t *l1_table,
            }

            if (!preallocated) {
-                if (!bs->backing) {
+                if (!bs->backing_hd) {
                    /* not backed; therefore we can simply deallocate the
                     * cluster */
                    l2_table[j] = 0;
@@ -1763,7 +1740,7 @@ static int expand_zero_clusters_in_l1(BlockDriverState *bs, uint64_t *l1_table,
                goto fail;
            }

-            ret = bdrv_write_zeroes(bs->file->bs, offset / BDRV_SECTOR_SIZE,
+            ret = bdrv_write_zeroes(bs->file, offset / BDRV_SECTOR_SIZE,
                                    s->cluster_sectors, 0);
            if (ret < 0) {
                if (!preallocated) {
@@ -1783,10 +1760,14 @@ static int expand_zero_clusters_in_l1(BlockDriverState *bs, uint64_t *l1_table,

        if (is_active_l1) {
            if (l2_dirty) {
-                qcow2_cache_entry_mark_dirty(bs, s->l2_table_cache, l2_table);
+                qcow2_cache_entry_mark_dirty(s->l2_table_cache, l2_table);
                qcow2_cache_depends_on_flush(s->l2_table_cache);
            }
-            qcow2_cache_put(bs, s->l2_table_cache, (void **) &l2_table);
+            ret = qcow2_cache_put(bs, s->l2_table_cache, (void **)&l2_table);
+            if (ret < 0) {
+                l2_table = NULL;
+                goto fail;
+            }
        } else {
            if (l2_dirty) {
                ret = qcow2_pre_write_overlap_check(bs,
@@ -1796,8 +1777,8 @@ static int expand_zero_clusters_in_l1(BlockDriverState *bs, uint64_t *l1_table,
                    goto fail;
                }

-                ret = bdrv_write(bs->file->bs, l2_offset / BDRV_SECTOR_SIZE,
-                                 (void *)l2_table, s->cluster_sectors);
+                ret = bdrv_write(bs->file, l2_offset / BDRV_SECTOR_SIZE,
+                        (void *)l2_table, s->cluster_sectors);
                if (ret < 0) {
                    goto fail;
                }
@@ -1806,7 +1787,7 @@ static int expand_zero_clusters_in_l1(BlockDriverState *bs, uint64_t *l1_table,

        (*visited_l1_entries)++;
        if (status_cb) {
-            status_cb(bs, *visited_l1_entries, l1_entries, cb_opaque);
+            status_cb(bs, *visited_l1_entries, l1_entries);
        }
    }

@@ -1817,7 +1798,12 @@ fail:
        if (!is_active_l1) {
            qemu_vfree(l2_table);
        } else {
-            qcow2_cache_put(bs, s->l2_table_cache, (void **) &l2_table);
+            if (ret < 0) {
+                qcow2_cache_put(bs, s->l2_table_cache, (void **)&l2_table);
+            } else {
+                ret = qcow2_cache_put(bs, s->l2_table_cache,
+                        (void **)&l2_table);
+            }
        }
    }
    return ret;
@@ -1830,10 +1816,9 @@ fail:
 * qcow2 version which doesn't yet support metadata zero clusters.
 */
 int qcow2_expand_zero_clusters(BlockDriverState *bs,
-                               BlockDriverAmendStatusCB *status_cb,
-                               void *cb_opaque)
+                               BlockDriverAmendStatusCB *status_cb)
 {
-    BDRVQcow2State *s = bs->opaque;
+    BDRVQcowState *s = bs->opaque;
    uint64_t *l1_table = NULL;
    int64_t l1_entries = 0, visited_l1_entries = 0;
    int ret;
@@ -1848,7 +1833,7 @@ int qcow2_expand_zero_clusters(BlockDriverState *bs,

    ret = expand_zero_clusters_in_l1(bs, s->l1_table, s->l1_size,
                                     &visited_l1_entries, l1_entries,
-                                     status_cb, cb_opaque);
+                                     status_cb);
    if (ret < 0) {
        goto fail;
    }
@@ -1871,9 +1856,8 @@ int qcow2_expand_zero_clusters(BlockDriverState *bs,

        l1_table = g_realloc(l1_table, l1_sectors * BDRV_SECTOR_SIZE);

-        ret = bdrv_read(bs->file->bs,
-                        s->snapshots[i].l1_table_offset / BDRV_SECTOR_SIZE,
-                        (void *)l1_table, l1_sectors);
+        ret = bdrv_read(bs->file, s->snapshots[i].l1_table_offset /
+                BDRV_SECTOR_SIZE, (void *)l1_table, l1_sectors);
        if (ret < 0) {
            goto fail;
        }
@@ -1884,7 +1868,7 @@ int qcow2_expand_zero_clusters(BlockDriverState *bs,

        ret = expand_zero_clusters_in_l1(bs, l1_table, s->snapshots[i].l1_size,
                                         &visited_l1_entries, l1_entries,
-                                         status_cb, cb_opaque);
+                                         status_cb);
        if (ret < 0) {
            goto fail;
        }
--- a/block/qcow2-refcount.c
+++ b/block/qcow2-refcount.c
@@ -22,7 +22,6 @@
 * THE SOFTWARE.
 */

-#include "qemu/osdep.h"
 #include "qemu-common.h"
 #include "block/block_int.h"
 #include "block/qcow2.h"
@@ -83,7 +82,7 @@ static Qcow2SetRefcountFunc *const set_refcount_funcs[] = {

 int qcow2_refcount_init(BlockDriverState *bs)
 {
-    BDRVQcow2State *s = bs->opaque;
+    BDRVQcowState *s = bs->opaque;
    unsigned int refcount_table_size2, i;
    int ret;

@@ -102,7 +101,7 @@ int qcow2_refcount_init(BlockDriverState *bs)
            goto fail;
        }
        BLKDBG_EVENT(bs->file, BLKDBG_REFTABLE_LOAD);
-        ret = bdrv_pread(bs->file->bs, s->refcount_table_offset,
+        ret = bdrv_pread(bs->file, s->refcount_table_offset,
                         s->refcount_table, refcount_table_size2);
        if (ret < 0) {
            goto fail;
@@ -117,7 +116,7 @@ int qcow2_refcount_init(BlockDriverState *bs)

 void qcow2_refcount_close(BlockDriverState *bs)
 {
-    BDRVQcow2State *s = bs->opaque;
+    BDRVQcowState *s = bs->opaque;
    g_free(s->refcount_table);
 }

@@ -215,7 +214,7 @@ static int load_refcount_block(BlockDriverState *bs,
                               int64_t refcount_block_offset,
                               void **refcount_block)
 {
-    BDRVQcow2State *s = bs->opaque;
+    BDRVQcowState *s = bs->opaque;
    int ret;

    BLKDBG_EVENT(bs->file, BLKDBG_REFBLOCK_LOAD);
@@ -232,7 +231,7 @@ static int load_refcount_block(BlockDriverState *bs,
 int qcow2_get_refcount(BlockDriverState *bs, int64_t cluster_index,
                       uint64_t *refcount)
 {
-    BDRVQcow2State *s = bs->opaque;
+    BDRVQcowState *s = bs->opaque;
    uint64_t refcount_table_index, block_index;
    int64_t refcount_block_offset;
    int ret;
@@ -266,7 +265,10 @@ int qcow2_get_refcount(BlockDriverState *bs, int64_t cluster_index,
    block_index = cluster_index & (s->refcount_block_size - 1);
    *refcount = s->get_refcount(refcount_block, block_index);

-    qcow2_cache_put(bs, s->refcount_block_cache, &refcount_block);
+    ret = qcow2_cache_put(bs, s->refcount_block_cache, &refcount_block);
+    if (ret < 0) {
+        return ret;
+    }

    return 0;
 }
@@ -275,7 +277,7 @@ int qcow2_get_refcount(BlockDriverState *bs, int64_t cluster_index,
 * Rounds the refcount table size up to avoid growing the table for each single
 * refcount block that is allocated.
 */
-static unsigned int next_refcount_table_size(BDRVQcow2State *s,
+static unsigned int next_refcount_table_size(BDRVQcowState *s,
    unsigned int min_size)
 {
    unsigned int min_clusters = (min_size >> (s->cluster_bits - 3)) + 1;
@@ -291,7 +293,7 @@ static unsigned int next_refcount_table_size(BDRVQcow2State *s,


 /* Checks if two offsets are described by the same refcount block */
-static int in_same_refcount_block(BDRVQcow2State *s, uint64_t offset_a,
+static int in_same_refcount_block(BDRVQcowState *s, uint64_t offset_a,
    uint64_t offset_b)
 {
    uint64_t block_a = offset_a >> (s->cluster_bits + s->refcount_block_bits);
@@ -309,7 +311,7 @@ static int in_same_refcount_block(BDRVQcow2State *s, uint64_t offset_a,
 static int alloc_refcount_block(BlockDriverState *bs,
                                int64_t cluster_index, void **refcount_block)
 {
-    BDRVQcow2State *s = bs->opaque;
+    BDRVQcowState *s = bs->opaque;
    unsigned int refcount_table_index;
    int ret;

@@ -422,7 +424,7 @@ static int alloc_refcount_block(BlockDriverState *bs,

    /* Now the new refcount block needs to be written to disk */
    BLKDBG_EVENT(bs->file, BLKDBG_REFBLOCK_ALLOC_WRITE);
-    qcow2_cache_entry_mark_dirty(bs, s->refcount_block_cache, *refcount_block);
+    qcow2_cache_entry_mark_dirty(s->refcount_block_cache, *refcount_block);
    ret = qcow2_cache_flush(bs, s->refcount_block_cache);
    if (ret < 0) {
        goto fail_block;
@@ -432,7 +434,7 @@ static int alloc_refcount_block(BlockDriverState *bs,
    if (refcount_table_index < s->refcount_table_size) {
        uint64_t data64 = cpu_to_be64(new_block);
        BLKDBG_EVENT(bs->file, BLKDBG_REFBLOCK_ALLOC_HOOKUP);
-        ret = bdrv_pwrite_sync(bs->file->bs,
+        ret = bdrv_pwrite_sync(bs->file,
            s->refcount_table_offset + refcount_table_index * sizeof(uint64_t),
            &data64, sizeof(data64));
        if (ret < 0) {
@@ -446,7 +448,10 @@ static int alloc_refcount_block(BlockDriverState *bs,
        return -EAGAIN;
    }

-    qcow2_cache_put(bs, s->refcount_block_cache, refcount_block);
+    ret = qcow2_cache_put(bs, s->refcount_block_cache, refcount_block);
+    if (ret < 0) {
+        goto fail_block;
+    }

    /*
     * If we come here, we need to grow the refcount table. Again, a new
@@ -536,7 +541,7 @@ static int alloc_refcount_block(BlockDriverState *bs,

    /* Write refcount blocks to disk */
    BLKDBG_EVENT(bs->file, BLKDBG_REFBLOCK_ALLOC_WRITE_BLOCKS);
-    ret = bdrv_pwrite_sync(bs->file->bs, meta_offset, new_blocks,
+    ret = bdrv_pwrite_sync(bs->file, meta_offset, new_blocks,
        blocks_clusters * s->cluster_size);
    g_free(new_blocks);
    new_blocks = NULL;
@@ -550,7 +555,7 @@ static int alloc_refcount_block(BlockDriverState *bs,
    }

    BLKDBG_EVENT(bs->file, BLKDBG_REFBLOCK_ALLOC_WRITE_TABLE);
-    ret = bdrv_pwrite_sync(bs->file->bs, table_offset, new_table,
+    ret = bdrv_pwrite_sync(bs->file, table_offset, new_table,
        table_size * sizeof(uint64_t));
    if (ret < 0) {
        goto fail_table;
@@ -561,16 +566,12 @@ static int alloc_refcount_block(BlockDriverState *bs,
    }

    /* Hook up the new refcount table in the qcow2 header */
-    struct QEMU_PACKED {
-        uint64_t d64;
-        uint32_t d32;
-    } data;
-    cpu_to_be64w(&data.d64, table_offset);
-    cpu_to_be32w(&data.d32, table_clusters);
+    uint8_t data[12];
+    cpu_to_be64w((uint64_t*)data, table_offset);
+    cpu_to_be32w((uint32_t*)(data + 8), table_clusters);
    BLKDBG_EVENT(bs->file, BLKDBG_REFBLOCK_ALLOC_SWITCH_TABLE);
-    ret = bdrv_pwrite_sync(bs->file->bs,
-                           offsetof(QCowHeader, refcount_table_offset),
-                           &data, sizeof(data));
+    ret = bdrv_pwrite_sync(bs->file, offsetof(QCowHeader, refcount_table_offset),
+        data, sizeof(data));
    if (ret < 0) {
        goto fail_table;
    }
@@ -610,7 +611,7 @@ fail_block:

 void qcow2_process_discards(BlockDriverState *bs, int ret)
 {
-    BDRVQcow2State *s = bs->opaque;
+    BDRVQcowState *s = bs->opaque;
    Qcow2DiscardRegion *d, *next;

    QTAILQ_FOREACH_SAFE(d, &s->discards, next, next) {
@@ -618,7 +619,7 @@ void qcow2_process_discards(BlockDriverState *bs, int ret)

        /* Discard is optional, ignore the return value */
        if (ret >= 0) {
-            bdrv_discard(bs->file->bs,
+            bdrv_discard(bs->file,
                         d->offset >> BDRV_SECTOR_BITS,
                         d->bytes >> BDRV_SECTOR_BITS);
        }
@@ -630,7 +631,7 @@ void qcow2_process_discards(BlockDriverState *bs, int ret)
 static void update_refcount_discard(BlockDriverState *bs,
                                    uint64_t offset, uint64_t length)
 {
-    BDRVQcow2State *s = bs->opaque;
+    BDRVQcowState *s = bs->opaque;
    Qcow2DiscardRegion *d, *p, *next;

    QTAILQ_FOREACH(d, &s->discards, next) {
@@ -687,7 +688,7 @@ static int QEMU_WARN_UNUSED_RESULT update_refcount(BlockDriverState *bs,
                                                   bool decrease,
                                                   enum qcow2_discard_type type)
 {
-    BDRVQcow2State *s = bs->opaque;
+    BDRVQcowState *s = bs->opaque;
    int64_t start, last, cluster_offset;
    void *refcount_block = NULL;
    int64_t old_table_index = -1;
@@ -722,8 +723,13 @@ static int QEMU_WARN_UNUSED_RESULT update_refcount(BlockDriverState *bs,
        /* Load the refcount block and allocate it if needed */
        if (table_index != old_table_index) {
            if (refcount_block) {
-                qcow2_cache_put(bs, s->refcount_block_cache, &refcount_block);
+                ret = qcow2_cache_put(bs, s->refcount_block_cache,
+                                      &refcount_block);
+                if (ret < 0) {
+                    goto fail;
+                }
            }
+
            ret = alloc_refcount_block(bs, cluster_index, &refcount_block);
            if (ret < 0) {
                goto fail;
@@ -731,8 +737,7 @@ static int QEMU_WARN_UNUSED_RESULT update_refcount(BlockDriverState *bs,
        }
        old_table_index = table_index;

-        qcow2_cache_entry_mark_dirty(bs, s->refcount_block_cache,
-                                     refcount_block);
+        qcow2_cache_entry_mark_dirty(s->refcount_block_cache, refcount_block);

        /* we can update the count and save it */
        block_index = cluster_index & (s->refcount_block_size - 1);
@@ -768,7 +773,11 @@ fail:

    /* Write last changed block to disk */
    if (refcount_block) {
-        qcow2_cache_put(bs, s->refcount_block_cache, &refcount_block);
+        int wret;
+        wret = qcow2_cache_put(bs, s->refcount_block_cache, &refcount_block);
+        if (wret < 0) {
+            return ret < 0 ? ret : wret;
+        }
    }

    /*
@@ -798,7 +807,7 @@ int qcow2_update_cluster_refcount(BlockDriverState *bs,
                                  uint64_t addend, bool decrease,
                                  enum qcow2_discard_type type)
 {
-    BDRVQcow2State *s = bs->opaque;
+    BDRVQcowState *s = bs->opaque;
    int ret;

    ret = update_refcount(bs, cluster_index << s->cluster_bits, 1, addend,
@@ -820,7 +829,7 @@ int qcow2_update_cluster_refcount(BlockDriverState *bs,
 /* return < 0 if error */
 static int64_t alloc_clusters_noref(BlockDriverState *bs, uint64_t size)
 {
-    BDRVQcow2State *s = bs->opaque;
+    BDRVQcowState *s = bs->opaque;
    uint64_t i, nb_clusters, refcount;
    int ret;

@@ -880,10 +889,10 @@ int64_t qcow2_alloc_clusters(BlockDriverState *bs, uint64_t size)
    return offset;
 }

-int64_t qcow2_alloc_clusters_at(BlockDriverState *bs, uint64_t offset,
-                                int64_t nb_clusters)
+int qcow2_alloc_clusters_at(BlockDriverState *bs, uint64_t offset,
+    int nb_clusters)
 {
-    BDRVQcow2State *s = bs->opaque;
+    BDRVQcowState *s = bs->opaque;
    uint64_t cluster_index, refcount;
    uint64_t i;
    int ret;
@@ -921,7 +930,7 @@ int64_t qcow2_alloc_clusters_at(BlockDriverState *bs, uint64_t offset,
   contiguous sectors. size must be <= cluster_size */
 int64_t qcow2_alloc_bytes(BlockDriverState *bs, int size)
 {
-    BDRVQcow2State *s = bs->opaque;
+    BDRVQcowState *s = bs->opaque;
    int64_t offset;
    size_t free_in_cluster;
    int ret;
@@ -1003,7 +1012,7 @@ void qcow2_free_clusters(BlockDriverState *bs,
 void qcow2_free_any_clusters(BlockDriverState *bs, uint64_t l2_entry,
                             int nb_clusters, enum qcow2_discard_type type)
 {
-    BDRVQcow2State *s = bs->opaque;
+    BDRVQcowState *s = bs->opaque;

    switch (qcow2_get_cluster_type(l2_entry)) {
    case QCOW2_CLUSTER_COMPRESSED:
@@ -1047,7 +1056,7 @@ void qcow2_free_any_clusters(BlockDriverState *bs, uint64_t l2_entry,
 int qcow2_update_snapshot_refcount(BlockDriverState *bs,
    int64_t l1_table_offset, int l1_size, int addend)
 {
-    BDRVQcow2State *s = bs->opaque;
+    BDRVQcowState *s = bs->opaque;
    uint64_t *l1_table, *l2_table, l2_offset, offset, l1_size2, refcount;
    bool l1_allocated = false;
    int64_t old_offset, old_l2_offset;
@@ -1073,7 +1082,7 @@ int qcow2_update_snapshot_refcount(BlockDriverState *bs,
        }
        l1_allocated = true;

-        ret = bdrv_pread(bs->file->bs, l1_table_offset, l1_table, l1_size2);
+        ret = bdrv_pread(bs->file, l1_table_offset, l1_table, l1_size2);
        if (ret < 0) {
            goto fail;
        }
@@ -1181,12 +1190,15 @@ int qcow2_update_snapshot_refcount(BlockDriverState *bs,
                            s->refcount_block_cache);
                    }
                    l2_table[j] = cpu_to_be64(offset);
-                    qcow2_cache_entry_mark_dirty(bs, s->l2_table_cache,
-                                                 l2_table);
+                    qcow2_cache_entry_mark_dirty(s->l2_table_cache, l2_table);
                }
            }

-            qcow2_cache_put(bs, s->l2_table_cache, (void **) &l2_table);
+            ret = qcow2_cache_put(bs, s->l2_table_cache, (void**) &l2_table);
+            if (ret < 0) {
+                goto fail;
+            }
+

            if (addend != 0) {
                ret = qcow2_update_cluster_refcount(bs, l2_offset >>
@@ -1226,8 +1238,7 @@ fail:
            cpu_to_be64s(&l1_table[i]);
        }

-        ret = bdrv_pwrite_sync(bs->file->bs, l1_table_offset,
-                               l1_table, l1_size2);
+        ret = bdrv_pwrite_sync(bs->file, l1_table_offset, l1_table, l1_size2);

        for (i = 0; i < l1_size; i++) {
            be64_to_cpus(&l1_table[i]);
@@ -1245,7 +1256,7 @@ fail:
 /* refcount checking functions */


-static uint64_t refcount_array_byte_size(BDRVQcow2State *s, uint64_t entries)
+static size_t refcount_array_byte_size(BDRVQcowState *s, uint64_t entries)
 {
    /* This assertion holds because there is no way we can address more than
     * 2^(64 - 9) clusters at once (with cluster size 512 = 2^9, and because
@@ -1268,10 +1279,10 @@ static uint64_t refcount_array_byte_size(BDRVQcow2State *s, uint64_t entries)
 * refcount array buffer will be aligned to a cluster boundary, and the newly
 * allocated area will be zeroed.
 */
-static int realloc_refcount_array(BDRVQcow2State *s, void **array,
+static int realloc_refcount_array(BDRVQcowState *s, void **array,
                                  int64_t *size, int64_t new_size)
 {
-    int64_t old_byte_size, new_byte_size;
+    size_t old_byte_size, new_byte_size;
    void *new_ptr;

    /* Round to clusters so the array can be directly written to disk */
@@ -1287,17 +1298,13 @@ static int realloc_refcount_array(BDRVQcow2State *s, void **array,

    assert(new_byte_size > 0);

-    if (new_byte_size > SIZE_MAX) {
-        return -ENOMEM;
-    }
-
    new_ptr = g_try_realloc(*array, new_byte_size);
    if (!new_ptr) {
        return -ENOMEM;
    }

    if (new_byte_size > old_byte_size) {
-        memset((char *)new_ptr + old_byte_size, 0,
+        memset((void *)((uintptr_t)new_ptr + old_byte_size), 0,
               new_byte_size - old_byte_size);
    }

@@ -1310,7 +1317,7 @@ static int realloc_refcount_array(BDRVQcow2State *s, void **array,
 /*
 * Increases the refcount for a range of clusters in a given refcount table.
 * This is used to construct a temporary refcount table out of L1 and L2 tables
- * which can be compared to the refcount table saved in the image.
+ * which can be compared the the refcount table saved in the image.
 *
 * Modifies the number of errors in res.
 */
@@ -1320,7 +1327,7 @@ static int inc_refcounts(BlockDriverState *bs,
                         int64_t *refcount_table_size,
                         int64_t offset, int64_t size)
 {
-    BDRVQcow2State *s = bs->opaque;
+    BDRVQcowState *s = bs->opaque;
    uint64_t start, last, cluster_offset, k, refcount;
    int ret;

@@ -1346,9 +1353,6 @@ static int inc_refcounts(BlockDriverState *bs,
        if (refcount == s->refcount_max) {
            fprintf(stderr, "ERROR: overflow cluster offset=0x%" PRIx64
                    "\n", cluster_offset);
-            fprintf(stderr, "Use qemu-img amend to increase the refcount entry "
-                    "width or qemu-img convert to create a clean copy if the "
-                    "image cannot be opened for writing\n");
            res->corruptions++;
            continue;
        }
@@ -1376,7 +1380,7 @@ static int check_refcounts_l2(BlockDriverState *bs, BdrvCheckResult *res,
                              int64_t *refcount_table_size, int64_t l2_offset,
                              int flags)
 {
-    BDRVQcow2State *s = bs->opaque;
+    BDRVQcowState *s = bs->opaque;
    uint64_t *l2_table, l2_entry;
    uint64_t next_contiguous_offset = 0;
    int i, l2_size, nb_csectors, ret;
@@ -1385,7 +1389,7 @@ static int check_refcounts_l2(BlockDriverState *bs, BdrvCheckResult *res,
    l2_size = s->l2_size * sizeof(uint64_t);
    l2_table = g_malloc(l2_size);

-    ret = bdrv_pread(bs->file->bs, l2_offset, l2_table, l2_size);
+    ret = bdrv_pread(bs->file, l2_offset, l2_table, l2_size);
    if (ret < 0) {
        fprintf(stderr, "ERROR: I/O error in check_refcounts_l2\n");
        res->check_errors++;
@@ -1496,7 +1500,7 @@ static int check_refcounts_l1(BlockDriverState *bs,
                              int64_t l1_table_offset, int l1_size,
                              int flags)
 {
-    BDRVQcow2State *s = bs->opaque;
+    BDRVQcowState *s = bs->opaque;
    uint64_t *l1_table = NULL, l2_offset, l1_size2;
    int i, ret;

@@ -1517,7 +1521,7 @@ static int check_refcounts_l1(BlockDriverState *bs,
            res->check_errors++;
            goto fail;
        }
-        ret = bdrv_pread(bs->file->bs, l1_table_offset, l1_table, l1_size2);
+        ret = bdrv_pread(bs->file, l1_table_offset, l1_table, l1_size2);
        if (ret < 0) {
            fprintf(stderr, "ERROR: I/O error in check_refcounts_l1\n");
            res->check_errors++;
@@ -1573,7 +1577,7 @@ fail:
 static int check_oflag_copied(BlockDriverState *bs, BdrvCheckResult *res,
                              BdrvCheckMode fix)
 {
-    BDRVQcow2State *s = bs->opaque;
+    BDRVQcowState *s = bs->opaque;
    uint64_t *l2_table = qemu_blockalign(bs, s->cluster_size);
    int ret;
    uint64_t refcount;
@@ -1615,7 +1619,7 @@ static int check_oflag_copied(BlockDriverState *bs, BdrvCheckResult *res,
            }
        }

-        ret = bdrv_pread(bs->file->bs, l2_offset, l2_table,
+        ret = bdrv_pread(bs->file, l2_offset, l2_table,
                         s->l2_size * sizeof(uint64_t));
        if (ret < 0) {
            fprintf(stderr, "ERROR: Could not read L2 table: %s\n",
@@ -1667,8 +1671,7 @@ static int check_oflag_copied(BlockDriverState *bs, BdrvCheckResult *res,
                goto fail;
            }

-            ret = bdrv_pwrite(bs->file->bs, l2_offset, l2_table,
-                              s->cluster_size);
+            ret = bdrv_pwrite(bs->file, l2_offset, l2_table, s->cluster_size);
            if (ret < 0) {
                fprintf(stderr, "ERROR: Could not write L2 table: %s\n",
                        strerror(-ret));
@@ -1693,7 +1696,7 @@ static int check_refblocks(BlockDriverState *bs, BdrvCheckResult *res,
                           BdrvCheckMode fix, bool *rebuild,
                           void **refcount_table, int64_t *nb_clusters)
 {
-    BDRVQcow2State *s = bs->opaque;
+    BDRVQcowState *s = bs->opaque;
    int64_t i, size;
    int ret;

@@ -1723,11 +1726,11 @@ static int check_refblocks(BlockDriverState *bs, BdrvCheckResult *res,
                    goto resize_fail;
                }

-                ret = bdrv_truncate(bs->file->bs, offset + s->cluster_size);
+                ret = bdrv_truncate(bs->file, offset + s->cluster_size);
                if (ret < 0) {
                    goto resize_fail;
                }
-                size = bdrv_getlength(bs->file->bs);
+                size = bdrv_getlength(bs->file);
                if (size < 0) {
                    ret = size;
                    goto resize_fail;
@@ -1796,7 +1799,7 @@ static int calculate_refcounts(BlockDriverState *bs, BdrvCheckResult *res,
                               BdrvCheckMode fix, bool *rebuild,
                               void **refcount_table, int64_t *nb_clusters)
 {
-    BDRVQcow2State *s = bs->opaque;
+    BDRVQcowState *s = bs->opaque;
    int64_t i;
    QCowSnapshot *sn;
    int ret;
@@ -1860,7 +1863,7 @@ static void compare_refcounts(BlockDriverState *bs, BdrvCheckResult *res,
                              int64_t *highest_cluster,
                              void *refcount_table, int64_t nb_clusters)
 {
-    BDRVQcow2State *s = bs->opaque;
+    BDRVQcowState *s = bs->opaque;
    int64_t i;
    uint64_t refcount1, refcount2;
    int ret;
@@ -1937,7 +1940,7 @@ static int64_t alloc_clusters_imrt(BlockDriverState *bs,
                                   int64_t *imrt_nb_clusters,
                                   int64_t *first_free_cluster)
 {
-    BDRVQcow2State *s = bs->opaque;
+    BDRVQcowState *s = bs->opaque;
    int64_t cluster = *first_free_cluster, i;
    bool first_gap = true;
    int contiguous_free_clusters;
@@ -2007,7 +2010,7 @@ static int rebuild_refcount_structure(BlockDriverState *bs,
                                      void **refcount_table,
                                      int64_t *nb_clusters)
 {
-    BDRVQcow2State *s = bs->opaque;
+    BDRVQcowState *s = bs->opaque;
    int64_t first_free_cluster = 0, reftable_offset = -1, cluster = 0;
    int64_t refblock_offset, refblock_start, refblock_index;
    uint32_t reftable_size = 0;
@@ -2101,7 +2104,7 @@ write_refblocks:
        on_disk_refblock = (void *)((char *) *refcount_table +
                                    refblock_index * s->cluster_size);

-        ret = bdrv_write(bs->file->bs, refblock_offset / BDRV_SECTOR_SIZE,
+        ret = bdrv_write(bs->file, refblock_offset / BDRV_SECTOR_SIZE,
                         on_disk_refblock, s->cluster_sectors);
        if (ret < 0) {
            fprintf(stderr, "ERROR writing refblock: %s\n", strerror(-ret));
@@ -2150,7 +2153,7 @@ write_refblocks:
    }

    assert(reftable_size < INT_MAX / sizeof(uint64_t));
-    ret = bdrv_pwrite(bs->file->bs, reftable_offset, on_disk_reftable,
+    ret = bdrv_pwrite(bs->file, reftable_offset, on_disk_reftable,
                      reftable_size * sizeof(uint64_t));
    if (ret < 0) {
        fprintf(stderr, "ERROR writing reftable: %s\n", strerror(-ret));
@@ -2162,8 +2165,8 @@ write_refblocks:
                 reftable_offset);
    cpu_to_be32w(&reftable_offset_and_clusters.reftable_clusters,
                 size_to_clusters(s, reftable_size * sizeof(uint64_t)));
-    ret = bdrv_pwrite_sync(bs->file->bs, offsetof(QCowHeader,
-                                                  refcount_table_offset),
+    ret = bdrv_pwrite_sync(bs->file, offsetof(QCowHeader,
+                                              refcount_table_offset),
                           &reftable_offset_and_clusters,
                           sizeof(reftable_offset_and_clusters));
    if (ret < 0) {
@@ -2194,14 +2197,14 @@ fail:
 int qcow2_check_refcounts(BlockDriverState *bs, BdrvCheckResult *res,
                          BdrvCheckMode fix)
 {
-    BDRVQcow2State *s = bs->opaque;
+    BDRVQcowState *s = bs->opaque;
    BdrvCheckResult pre_compare_res;
    int64_t size, highest_cluster, nb_clusters;
    void *refcount_table = NULL;
    bool rebuild = false;
    int ret;

-    size = bdrv_getlength(bs->file->bs);
+    size = bdrv_getlength(bs->file);
    if (size < 0) {
        res->check_errors++;
        return size;
@@ -2331,7 +2334,7 @@ fail:
 int qcow2_check_metadata_overlap(BlockDriverState *bs, int ign, int64_t offset,
                                 int64_t size)
 {
-    BDRVQcow2State *s = bs->opaque;
+    BDRVQcowState *s = bs->opaque;
    int chk = s->overlap_check & ~ign;
    int i, j;

@@ -2410,7 +2413,7 @@ int qcow2_check_metadata_overlap(BlockDriverState *bs, int ign, int64_t offset,
                return -ENOMEM;
            }

-            ret = bdrv_pread(bs->file->bs, l1_ofs, l1, l1_sz2);
+            ret = bdrv_pread(bs->file, l1_ofs, l1, l1_sz2);
            if (ret < 0) {
                g_free(l1);
                return ret;
@@ -2460,7 +2463,7 @@ int qcow2_pre_write_overlap_check(BlockDriverState *bs, int ign, int64_t offset,
    if (ret < 0) {
        return ret;
    } else if (ret > 0) {
-        int metadata_ol_bitnr = ctz32(ret);
+        int metadata_ol_bitnr = ffs(ret) - 1;
        assert(metadata_ol_bitnr < QCOW2_OL_MAX_BITNR);

        qcow2_signal_corruption(bs, true, offset, size, "Preventing invalid "
@@ -2471,450 +2474,3 @@ int qcow2_pre_write_overlap_check(BlockDriverState *bs, int ign, int64_t offset,

    return 0;
 }
-
-/* A pointer to a function of this type is given to walk_over_reftable(). That
- * function will create refblocks and pass them to a RefblockFinishOp once they
- * are completed (@refblock). @refblock_empty is set if the refblock is
- * completely empty.
- *
- * Along with the refblock, a corresponding reftable entry is passed, in the
- * reftable @reftable (which may be reallocated) at @reftable_index.
- *
- * @allocated should be set to true if a new cluster has been allocated.
- */
-typedef int (RefblockFinishOp)(BlockDriverState *bs, uint64_t **reftable,
-                               uint64_t reftable_index, uint64_t *reftable_size,
-                               void *refblock, bool refblock_empty,
-                               bool *allocated, Error **errp);
-
-/**
- * This "operation" for walk_over_reftable() allocates the refblock on disk (if
- * it is not empty) and inserts its offset into the new reftable. The size of
- * this new reftable is increased as required.
- */
-static int alloc_refblock(BlockDriverState *bs, uint64_t **reftable,
-                          uint64_t reftable_index, uint64_t *reftable_size,
-                          void *refblock, bool refblock_empty, bool *allocated,
-                          Error **errp)
-{
-    BDRVQcow2State *s = bs->opaque;
-    int64_t offset;
-
-    if (!refblock_empty && reftable_index >= *reftable_size) {
-        uint64_t *new_reftable;
-        uint64_t new_reftable_size;
-
-        new_reftable_size = ROUND_UP(reftable_index + 1,
-                                     s->cluster_size / sizeof(uint64_t));
-        if (new_reftable_size > QCOW_MAX_REFTABLE_SIZE / sizeof(uint64_t)) {
-            error_setg(errp,
-                       "This operation would make the refcount table grow "
-                       "beyond the maximum size supported by QEMU, aborting");
-            return -ENOTSUP;
-        }
-
-        new_reftable = g_try_realloc(*reftable, new_reftable_size *
-                                                sizeof(uint64_t));
-        if (!new_reftable) {
-            error_setg(errp, "Failed to increase reftable buffer size");
-            return -ENOMEM;
-        }
-
-        memset(new_reftable + *reftable_size, 0,
-               (new_reftable_size - *reftable_size) * sizeof(uint64_t));
-
-        *reftable      = new_reftable;
-        *reftable_size = new_reftable_size;
-    }
-
-    if (!refblock_empty && !(*reftable)[reftable_index]) {
-        offset = qcow2_alloc_clusters(bs, s->cluster_size);
-        if (offset < 0) {
-            error_setg_errno(errp, -offset, "Failed to allocate refblock");
-            return offset;
-        }
-        (*reftable)[reftable_index] = offset;
-        *allocated = true;
-    }
-
-    return 0;
-}
-
-/**
- * This "operation" for walk_over_reftable() writes the refblock to disk at the
- * offset specified by the new reftable's entry. It does not modify the new
- * reftable or change any refcounts.
- */
-static int flush_refblock(BlockDriverState *bs, uint64_t **reftable,
-                          uint64_t reftable_index, uint64_t *reftable_size,
-                          void *refblock, bool refblock_empty, bool *allocated,
-                          Error **errp)
-{
-    BDRVQcow2State *s = bs->opaque;
-    int64_t offset;
-    int ret;
-
-    if (reftable_index < *reftable_size && (*reftable)[reftable_index]) {
-        offset = (*reftable)[reftable_index];
-
-        ret = qcow2_pre_write_overlap_check(bs, 0, offset, s->cluster_size);
-        if (ret < 0) {
-            error_setg_errno(errp, -ret, "Overlap check failed");
-            return ret;
-        }
-
-        ret = bdrv_pwrite(bs->file->bs, offset, refblock, s->cluster_size);
-        if (ret < 0) {
-            error_setg_errno(errp, -ret, "Failed to write refblock");
-            return ret;
-        }
-    } else {
-        assert(refblock_empty);
-    }
-
-    return 0;
-}
-
-/**
- * This function walks over the existing reftable and every referenced refblock;
- * if @new_set_refcount is non-NULL, it is called for every refcount entry to
- * create an equal new entry in the passed @new_refblock. Once that
- * @new_refblock is completely filled, @operation will be called.
- *
- * @status_cb and @cb_opaque are used for the amend operation's status callback.
- * @index is the index of the walk_over_reftable() calls and @total is the total
- * number of walk_over_reftable() calls per amend operation. Both are used for
- * calculating the parameters for the status callback.
- *
- * @allocated is set to true if a new cluster has been allocated.
- */
-static int walk_over_reftable(BlockDriverState *bs, uint64_t **new_reftable,
-                              uint64_t *new_reftable_index,
-                              uint64_t *new_reftable_size,
-                              void *new_refblock, int new_refblock_size,
-                              int new_refcount_bits,
-                              RefblockFinishOp *operation, bool *allocated,
-                              Qcow2SetRefcountFunc *new_set_refcount,
-                              BlockDriverAmendStatusCB *status_cb,
-                              void *cb_opaque, int index, int total,
-                              Error **errp)
-{
-    BDRVQcow2State *s = bs->opaque;
-    uint64_t reftable_index;
-    bool new_refblock_empty = true;
-    int refblock_index;
-    int new_refblock_index = 0;
-    int ret;
-
-    for (reftable_index = 0; reftable_index < s->refcount_table_size;
-         reftable_index++)
-    {
-        uint64_t refblock_offset = s->refcount_table[reftable_index]
-                                 & REFT_OFFSET_MASK;
-
-        status_cb(bs, (uint64_t)index * s->refcount_table_size + reftable_index,
-                  (uint64_t)total * s->refcount_table_size, cb_opaque);
-
-        if (refblock_offset) {
-            void *refblock;
-
-            if (offset_into_cluster(s, refblock_offset)) {
-                qcow2_signal_corruption(bs, true, -1, -1, "Refblock offset %#"
-                                        PRIx64 " unaligned (reftable index: %#"
-                                        PRIx64 ")", refblock_offset,
-                                        reftable_index);
-                error_setg(errp,
-                           "Image is corrupt (unaligned refblock offset)");
-                return -EIO;
-            }
-
-            ret = qcow2_cache_get(bs, s->refcount_block_cache, refblock_offset,
-                                  &refblock);
-            if (ret < 0) {
-                error_setg_errno(errp, -ret, "Failed to retrieve refblock");
-                return ret;
-            }
-
-            for (refblock_index = 0; refblock_index < s->refcount_block_size;
-                 refblock_index++)
-            {
-                uint64_t refcount;
-
-                if (new_refblock_index >= new_refblock_size) {
-                    /* new_refblock is now complete */
-                    ret = operation(bs, new_reftable, *new_reftable_index,
-                                    new_reftable_size, new_refblock,
-                                    new_refblock_empty, allocated, errp);
-                    if (ret < 0) {
-                        qcow2_cache_put(bs, s->refcount_block_cache, &refblock);
-                        return ret;
-                    }
-
-                    (*new_reftable_index)++;
-                    new_refblock_index = 0;
-                    new_refblock_empty = true;
-                }
-
-                refcount = s->get_refcount(refblock, refblock_index);
-                if (new_refcount_bits < 64 && refcount >> new_refcount_bits) {
-                    uint64_t offset;
-
-                    qcow2_cache_put(bs, s->refcount_block_cache, &refblock);
-
-                    offset = ((reftable_index << s->refcount_block_bits)
-                              + refblock_index) << s->cluster_bits;
-
-                    error_setg(errp, "Cannot decrease refcount entry width to "
-                               "%i bits: Cluster at offset %#" PRIx64 " has a "
-                               "refcount of %" PRIu64, new_refcount_bits,
-                               offset, refcount);
-                    return -EINVAL;
-                }
-
-                if (new_set_refcount) {
-                    new_set_refcount(new_refblock, new_refblock_index++,
-                                     refcount);
-                } else {
-                    new_refblock_index++;
-                }
-                new_refblock_empty = new_refblock_empty && refcount == 0;
-            }
-
-            qcow2_cache_put(bs, s->refcount_block_cache, &refblock);
-        } else {
-            /* No refblock means every refcount is 0 */
-            for (refblock_index = 0; refblock_index < s->refcount_block_size;
-                 refblock_index++)
-            {
-                if (new_refblock_index >= new_refblock_size) {
-                    /* new_refblock is now complete */
-                    ret = operation(bs, new_reftable, *new_reftable_index,
-                                    new_reftable_size, new_refblock,
-                                    new_refblock_empty, allocated, errp);
-                    if (ret < 0) {
-                        return ret;
-                    }
-
-                    (*new_reftable_index)++;
-                    new_refblock_index = 0;
-                    new_refblock_empty = true;
-                }
-
-                if (new_set_refcount) {
-                    new_set_refcount(new_refblock, new_refblock_index++, 0);
-                } else {
-                    new_refblock_index++;
-                }
-            }
-        }
-    }
-
-    if (new_refblock_index > 0) {
-        /* Complete the potentially existing partially filled final refblock */
-        if (new_set_refcount) {
-            for (; new_refblock_index < new_refblock_size;
-                 new_refblock_index++)
-            {
-                new_set_refcount(new_refblock, new_refblock_index, 0);
-            }
-        }
-
-        ret = operation(bs, new_reftable, *new_reftable_index,
-                        new_reftable_size, new_refblock, new_refblock_empty,
-                        allocated, errp);
-        if (ret < 0) {
-            return ret;
-        }
-
-        (*new_reftable_index)++;
-    }
-
-    status_cb(bs, (uint64_t)(index + 1) * s->refcount_table_size,
-              (uint64_t)total * s->refcount_table_size, cb_opaque);
-
-    return 0;
-}
-
-int qcow2_change_refcount_order(BlockDriverState *bs, int refcount_order,
-                                BlockDriverAmendStatusCB *status_cb,
-                                void *cb_opaque, Error **errp)
-{
-    BDRVQcow2State *s = bs->opaque;
-    Qcow2GetRefcountFunc *new_get_refcount;
-    Qcow2SetRefcountFunc *new_set_refcount;
-    void *new_refblock = qemu_blockalign(bs->file->bs, s->cluster_size);
-    uint64_t *new_reftable = NULL, new_reftable_size = 0;
-    uint64_t *old_reftable, old_reftable_size, old_reftable_offset;
-    uint64_t new_reftable_index = 0;
-    uint64_t i;
-    int64_t new_reftable_offset = 0, allocated_reftable_size = 0;
-    int new_refblock_size, new_refcount_bits = 1 << refcount_order;
-    int old_refcount_order;
-    int walk_index = 0;
-    int ret;
-    bool new_allocation;
-
-    assert(s->qcow_version >= 3);
-    assert(refcount_order >= 0 && refcount_order <= 6);
-
-    /* see qcow2_open() */
-    new_refblock_size = 1 << (s->cluster_bits - (refcount_order - 3));
-
-    new_get_refcount = get_refcount_funcs[refcount_order];
-    new_set_refcount = set_refcount_funcs[refcount_order];
-
-
-    do {
-        int total_walks;
-
-        new_allocation = false;
-
-        /* At least we have to do this walk and the one which writes the
-         * refblocks; also, at least we have to do this loop here at least
-         * twice (normally), first to do the allocations, and second to
-         * determine that everything is correctly allocated, this then makes
-         * three walks in total */
-        total_walks = MAX(walk_index + 2, 3);
-
-        /* First, allocate the structures so they are present in the refcount
-         * structures */
-        ret = walk_over_reftable(bs, &new_reftable, &new_reftable_index,
-                                 &new_reftable_size, NULL, new_refblock_size,
-                                 new_refcount_bits, &alloc_refblock,
-                                 &new_allocation, NULL, status_cb, cb_opaque,
-                                 walk_index++, total_walks, errp);
-        if (ret < 0) {
-            goto done;
-        }
-
-        new_reftable_index = 0;
-
-        if (new_allocation) {
-            if (new_reftable_offset) {
-                qcow2_free_clusters(bs, new_reftable_offset,
-                                    allocated_reftable_size * sizeof(uint64_t),
-                                    QCOW2_DISCARD_NEVER);
-            }
-
-            new_reftable_offset = qcow2_alloc_clusters(bs, new_reftable_size *
-                                                           sizeof(uint64_t));
-            if (new_reftable_offset < 0) {
-                error_setg_errno(errp, -new_reftable_offset,
-                                 "Failed to allocate the new reftable");
-                ret = new_reftable_offset;
-                goto done;
-            }
-            allocated_reftable_size = new_reftable_size;
-        }
-    } while (new_allocation);
-
-    /* Second, write the new refblocks */
-    ret = walk_over_reftable(bs, &new_reftable, &new_reftable_index,
-                             &new_reftable_size, new_refblock,
-                             new_refblock_size, new_refcount_bits,
-                             &flush_refblock, &new_allocation, new_set_refcount,
-                             status_cb, cb_opaque, walk_index, walk_index + 1,
-                             errp);
-    if (ret < 0) {
-        goto done;
-    }
-    assert(!new_allocation);
-
-
-    /* Write the new reftable */
-    ret = qcow2_pre_write_overlap_check(bs, 0, new_reftable_offset,
-                                        new_reftable_size * sizeof(uint64_t));
-    if (ret < 0) {
-        error_setg_errno(errp, -ret, "Overlap check failed");
-        goto done;
-    }
-
-    for (i = 0; i < new_reftable_size; i++) {
-        cpu_to_be64s(&new_reftable[i]);
-    }
-
-    ret = bdrv_pwrite(bs->file->bs, new_reftable_offset, new_reftable,
-                      new_reftable_size * sizeof(uint64_t));
-
-    for (i = 0; i < new_reftable_size; i++) {
-        be64_to_cpus(&new_reftable[i]);
-    }
-
-    if (ret < 0) {
-        error_setg_errno(errp, -ret, "Failed to write the new reftable");
-        goto done;
-    }
-
-
-    /* Empty the refcount cache */
-    ret = qcow2_cache_flush(bs, s->refcount_block_cache);
-    if (ret < 0) {
-        error_setg_errno(errp, -ret, "Failed to flush the refblock cache");
-        goto done;
-    }
-
-    /* Update the image header to point to the new reftable; this only updates
-     * the fields which are relevant to qcow2_update_header(); other fields
-     * such as s->refcount_table or s->refcount_bits stay stale for now
-     * (because we have to restore everything if qcow2_update_header() fails) */
-    old_refcount_order  = s->refcount_order;
-    old_reftable_size   = s->refcount_table_size;
-    old_reftable_offset = s->refcount_table_offset;
-
-    s->refcount_order        = refcount_order;
-    s->refcount_table_size   = new_reftable_size;
-    s->refcount_table_offset = new_reftable_offset;
-
-    ret = qcow2_update_header(bs);
-    if (ret < 0) {
-        s->refcount_order        = old_refcount_order;
-        s->refcount_table_size   = old_reftable_size;
-        s->refcount_table_offset = old_reftable_offset;
-        error_setg_errno(errp, -ret, "Failed to update the qcow2 header");
-        goto done;
-    }
-
-    /* Now update the rest of the in-memory information */
-    old_reftable = s->refcount_table;
-    s->refcount_table = new_reftable;
-
-    s->refcount_bits = 1 << refcount_order;
-    s->refcount_max = UINT64_C(1) << (s->refcount_bits - 1);
-    s->refcount_max += s->refcount_max - 1;
-
-    s->refcount_block_bits = s->cluster_bits - (refcount_order - 3);
-    s->refcount_block_size = 1 << s->refcount_block_bits;
-
-    s->get_refcount = new_get_refcount;
-    s->set_refcount = new_set_refcount;
-
-    /* For cleaning up all old refblocks and the old reftable below the "done"
-     * label */
-    new_reftable        = old_reftable;
-    new_reftable_size   = old_reftable_size;
-    new_reftable_offset = old_reftable_offset;
-
-done:
-    if (new_reftable) {
-        /* On success, new_reftable actually points to the old reftable (and
-         * new_reftable_size is the old reftable's size); but that is just
-         * fine */
-        for (i = 0; i < new_reftable_size; i++) {
-            uint64_t offset = new_reftable[i] & REFT_OFFSET_MASK;
-            if (offset) {
-                qcow2_free_clusters(bs, offset, s->cluster_size,
-                                    QCOW2_DISCARD_OTHER);
-            }
-        }
-        g_free(new_reftable);
-
-        if (new_reftable_offset > 0) {
-            qcow2_free_clusters(bs, new_reftable_offset,
-                                new_reftable_size * sizeof(uint64_t),
-                                QCOW2_DISCARD_OTHER);
-        }
-    }
-
-    qemu_vfree(new_refblock);
-    return ret;
-}
--- a/block/qcow2-snapshot.c
+++ b/block/qcow2-snapshot.c
@@ -22,15 +22,13 @@
 * THE SOFTWARE.
 */

-#include "qemu/osdep.h"
 #include "qemu-common.h"
 #include "block/block_int.h"
 #include "block/qcow2.h"
-#include "qemu/error-report.h"

 void qcow2_free_snapshots(BlockDriverState *bs)
 {
-    BDRVQcow2State *s = bs->opaque;
+    BDRVQcowState *s = bs->opaque;
    int i;

    for(i = 0; i < s->nb_snapshots; i++) {
@@ -44,7 +42,7 @@ void qcow2_free_snapshots(BlockDriverState *bs)

 int qcow2_read_snapshots(BlockDriverState *bs)
 {
-    BDRVQcow2State *s = bs->opaque;
+    BDRVQcowState *s = bs->opaque;
    QCowSnapshotHeader h;
    QCowSnapshotExtraData extra;
    QCowSnapshot *sn;
@@ -65,7 +63,7 @@ int qcow2_read_snapshots(BlockDriverState *bs)
    for(i = 0; i < s->nb_snapshots; i++) {
        /* Read statically sized part of the snapshot header */
        offset = align_offset(offset, 8);
-        ret = bdrv_pread(bs->file->bs, offset, &h, sizeof(h));
+        ret = bdrv_pread(bs->file, offset, &h, sizeof(h));
        if (ret < 0) {
            goto fail;
        }
@@ -84,7 +82,7 @@ int qcow2_read_snapshots(BlockDriverState *bs)
        name_size = be16_to_cpu(h.name_size);

        /* Read extra data */
-        ret = bdrv_pread(bs->file->bs, offset, &extra,
+        ret = bdrv_pread(bs->file, offset, &extra,
                         MIN(sizeof(extra), extra_data_size));
        if (ret < 0) {
            goto fail;
@@ -103,7 +101,7 @@ int qcow2_read_snapshots(BlockDriverState *bs)

        /* Read snapshot ID */
        sn->id_str = g_malloc(id_str_size + 1);
-        ret = bdrv_pread(bs->file->bs, offset, sn->id_str, id_str_size);
+        ret = bdrv_pread(bs->file, offset, sn->id_str, id_str_size);
        if (ret < 0) {
            goto fail;
        }
@@ -112,7 +110,7 @@ int qcow2_read_snapshots(BlockDriverState *bs)

        /* Read snapshot name */
        sn->name = g_malloc(name_size + 1);
-        ret = bdrv_pread(bs->file->bs, offset, sn->name, name_size);
+        ret = bdrv_pread(bs->file, offset, sn->name, name_size);
        if (ret < 0) {
            goto fail;
        }
@@ -137,7 +135,7 @@ fail:
 /* add at the end of the file a new list of snapshots */
 static int qcow2_write_snapshots(BlockDriverState *bs)
 {
-    BDRVQcow2State *s = bs->opaque;
+    BDRVQcowState *s = bs->opaque;
    QCowSnapshot *sn;
    QCowSnapshotHeader h;
    QCowSnapshotExtraData extra;
@@ -215,25 +213,25 @@ static int qcow2_write_snapshots(BlockDriverState *bs)
        h.name_size = cpu_to_be16(name_size);
        offset = align_offset(offset, 8);

-        ret = bdrv_pwrite(bs->file->bs, offset, &h, sizeof(h));
+        ret = bdrv_pwrite(bs->file, offset, &h, sizeof(h));
        if (ret < 0) {
            goto fail;
        }
        offset += sizeof(h);

-        ret = bdrv_pwrite(bs->file->bs, offset, &extra, sizeof(extra));
+        ret = bdrv_pwrite(bs->file, offset, &extra, sizeof(extra));
        if (ret < 0) {
            goto fail;
        }
        offset += sizeof(extra);

-        ret = bdrv_pwrite(bs->file->bs, offset, sn->id_str, id_str_size);
+        ret = bdrv_pwrite(bs->file, offset, sn->id_str, id_str_size);
        if (ret < 0) {
            goto fail;
        }
        offset += id_str_size;

-        ret = bdrv_pwrite(bs->file->bs, offset, sn->name, name_size);
+        ret = bdrv_pwrite(bs->file, offset, sn->name, name_size);
        if (ret < 0) {
            goto fail;
        }
@@ -255,7 +253,7 @@ static int qcow2_write_snapshots(BlockDriverState *bs)
    header_data.nb_snapshots        = cpu_to_be32(s->nb_snapshots);
    header_data.snapshots_offset    = cpu_to_be64(snapshots_offset);

-    ret = bdrv_pwrite_sync(bs->file->bs, offsetof(QCowHeader, nb_snapshots),
+    ret = bdrv_pwrite_sync(bs->file, offsetof(QCowHeader, nb_snapshots),
                           &header_data, sizeof(header_data));
    if (ret < 0) {
        goto fail;
@@ -279,7 +277,7 @@ fail:
 static void find_new_snapshot_id(BlockDriverState *bs,
                                 char *id_str, int id_str_size)
 {
-    BDRVQcow2State *s = bs->opaque;
+    BDRVQcowState *s = bs->opaque;
    QCowSnapshot *sn;
    int i;
    unsigned long id, id_max = 0;
@@ -297,7 +295,7 @@ static int find_snapshot_by_id_and_name(BlockDriverState *bs,
                                        const char *id,
                                        const char *name)
 {
-    BDRVQcow2State *s = bs->opaque;
+    BDRVQcowState *s = bs->opaque;
    int i;

    if (id && name) {
@@ -339,7 +337,7 @@ static int find_snapshot_by_id_or_name(BlockDriverState *bs,
 /* if no id is provided, a new one is constructed */
 int qcow2_snapshot_create(BlockDriverState *bs, QEMUSnapshotInfo *sn_info)
 {
-    BDRVQcow2State *s = bs->opaque;
+    BDRVQcowState *s = bs->opaque;
    QCowSnapshot *new_snapshot_list = NULL;
    QCowSnapshot *old_snapshot_list = NULL;
    QCowSnapshot sn1, *sn = &sn1;
@@ -353,8 +351,10 @@ int qcow2_snapshot_create(BlockDriverState *bs, QEMUSnapshotInfo *sn_info)

    memset(sn, 0, sizeof(*sn));

-    /* Generate an ID */
-    find_new_snapshot_id(bs, sn_info->id_str, sizeof(sn_info->id_str));
+    /* Generate an ID if it wasn't passed */
+    if (sn_info->id_str[0] == '\0') {
+        find_new_snapshot_id(bs, sn_info->id_str, sizeof(sn_info->id_str));
+    }

    /* Check that the ID is unique */
    if (find_snapshot_by_id_and_name(bs, sn_info->id_str, NULL) >= 0) {
@@ -397,7 +397,7 @@ int qcow2_snapshot_create(BlockDriverState *bs, QEMUSnapshotInfo *sn_info)
        goto fail;
    }

-    ret = bdrv_pwrite(bs->file->bs, sn->l1_table_offset, l1_table,
+    ret = bdrv_pwrite(bs->file, sn->l1_table_offset, l1_table,
                      s->l1_size * sizeof(uint64_t));
    if (ret < 0) {
        goto fail;
@@ -462,7 +462,7 @@ fail:
 /* copy the snapshot 'snapshot_name' into the current disk image */
 int qcow2_snapshot_goto(BlockDriverState *bs, const char *snapshot_id)
 {
-    BDRVQcow2State *s = bs->opaque;
+    BDRVQcowState *s = bs->opaque;
    QCowSnapshot *sn;
    int i, snapshot_index;
    int cur_l1_bytes, sn_l1_bytes;
@@ -510,8 +510,7 @@ int qcow2_snapshot_goto(BlockDriverState *bs, const char *snapshot_id)
        goto fail;
    }

-    ret = bdrv_pread(bs->file->bs, sn->l1_table_offset,
-                     sn_l1_table, sn_l1_bytes);
+    ret = bdrv_pread(bs->file, sn->l1_table_offset, sn_l1_table, sn_l1_bytes);
    if (ret < 0) {
        goto fail;
    }
@@ -528,7 +527,7 @@ int qcow2_snapshot_goto(BlockDriverState *bs, const char *snapshot_id)
        goto fail;
    }

-    ret = bdrv_pwrite_sync(bs->file->bs, s->l1_table_offset, sn_l1_table,
+    ret = bdrv_pwrite_sync(bs->file, s->l1_table_offset, sn_l1_table,
                           cur_l1_bytes);
    if (ret < 0) {
        goto fail;
@@ -589,7 +588,7 @@ int qcow2_snapshot_delete(BlockDriverState *bs,
                          const char *name,
                          Error **errp)
 {
-    BDRVQcow2State *s = bs->opaque;
+    BDRVQcowState *s = bs->opaque;
    QCowSnapshot sn;
    int snapshot_index, ret;

@@ -652,7 +651,7 @@ int qcow2_snapshot_delete(BlockDriverState *bs,

 int qcow2_snapshot_list(BlockDriverState *bs, QEMUSnapshotInfo **psn_tab)
 {
-    BDRVQcow2State *s = bs->opaque;
+    BDRVQcowState *s = bs->opaque;
    QEMUSnapshotInfo *sn_tab, *sn_info;
    QCowSnapshot *sn;
    int i;
@@ -685,7 +684,7 @@ int qcow2_snapshot_load_tmp(BlockDriverState *bs,
                            Error **errp)
 {
    int i, snapshot_index;
-    BDRVQcow2State *s = bs->opaque;
+    BDRVQcowState *s = bs->opaque;
    QCowSnapshot *sn;
    uint64_t *new_l1_table;
    int new_l1_bytes;
@@ -708,14 +707,13 @@ int qcow2_snapshot_load_tmp(BlockDriverState *bs,
        return -EFBIG;
    }
    new_l1_bytes = sn->l1_size * sizeof(uint64_t);
-    new_l1_table = qemu_try_blockalign(bs->file->bs,
+    new_l1_table = qemu_try_blockalign(bs->file,
                                       align_offset(new_l1_bytes, 512));
    if (new_l1_table == NULL) {
        return -ENOMEM;
    }

-    ret = bdrv_pread(bs->file->bs, sn->l1_table_offset,
-                     new_l1_table, new_l1_bytes);
+    ret = bdrv_pread(bs->file, sn->l1_table_offset, new_l1_table, new_l1_bytes);
    if (ret < 0) {
        error_setg(errp, "Failed to read l1 table for snapshot");
        qemu_vfree(new_l1_table);
--- a/block/qcow2.c
+++ b/block/qcow2.c
--- a/block/qcow2.h
+++ b/block/qcow2.h
@@ -25,8 +25,8 @@
 #ifndef BLOCK_QCOW2_H
 #define BLOCK_QCOW2_H

-#include "crypto/cipher.h"
-#include "qemu/coroutine.h"
+#include "qemu/aes.h"
+#include "block/coroutine.h"

 //#define DEBUG_ALLOC
 //#define DEBUG_ALLOC2
@@ -68,8 +68,6 @@
 /* Must be at least 4 to cover all cases of refcount table growth */
 #define MIN_REFCOUNT_CACHE_SIZE 4 /* clusters */

-/* Whichever is more */
-#define DEFAULT_L2_CACHE_CLUSTERS 8 /* clusters */
 #define DEFAULT_L2_CACHE_BYTE_SIZE 1048576 /* bytes */

 /* The refblock cache needs only a fourth of the L2 cache size to cover as many
@@ -96,7 +94,6 @@
 #define QCOW2_OPT_CACHE_SIZE "cache-size"
 #define QCOW2_OPT_L2_CACHE_SIZE "l2-cache-size"
 #define QCOW2_OPT_REFCOUNT_CACHE_SIZE "refcount-cache-size"
-#define QCOW2_OPT_CACHE_CLEAN_INTERVAL "cache-clean-interval"

 typedef struct QCowHeader {
    uint32_t magic;
@@ -222,7 +219,7 @@ typedef uint64_t Qcow2GetRefcountFunc(const void *refcount_array,
 typedef void Qcow2SetRefcountFunc(void *refcount_array,
                                  uint64_t index, uint64_t value);

-typedef struct BDRVQcow2State {
+typedef struct BDRVQcowState {
    int cluster_bits;
    int cluster_size;
    int cluster_sectors;
@@ -240,8 +237,6 @@ typedef struct BDRVQcow2State {

    Qcow2Cache* l2_table_cache;
    Qcow2Cache* refcount_block_cache;
-    QEMUTimer *cache_clean_timer;
-    unsigned cache_clean_interval;

    uint8_t *cluster_cache;
    uint8_t *cluster_data;
@@ -256,8 +251,10 @@ typedef struct BDRVQcow2State {

    CoMutex lock;

-    QCryptoCipher *cipher; /* current cipher, NULL if no key yet */
+    uint32_t crypt_method; /* current crypt method, 0 if no key yet */
    uint32_t crypt_method_header;
+    AES_KEY aes_encrypt_key;
+    AES_KEY aes_decrypt_key;
    uint64_t snapshots_offset;
    int snapshots_size;
    unsigned int nb_snapshots;
@@ -293,7 +290,9 @@ typedef struct BDRVQcow2State {
     * override) */
    char *image_backing_file;
    char *image_backing_format;
-} BDRVQcow2State;
+} BDRVQcowState;
+
+struct QCowAIOCB;

 typedef struct Qcow2COWRegion {
    /**
@@ -403,28 +402,28 @@ typedef enum QCow2MetadataOverlap {

 #define REFT_OFFSET_MASK 0xfffffffffffffe00ULL

-static inline int64_t start_of_cluster(BDRVQcow2State *s, int64_t offset)
+static inline int64_t start_of_cluster(BDRVQcowState *s, int64_t offset)
 {
    return offset & ~(s->cluster_size - 1);
 }

-static inline int64_t offset_into_cluster(BDRVQcow2State *s, int64_t offset)
+static inline int64_t offset_into_cluster(BDRVQcowState *s, int64_t offset)
 {
    return offset & (s->cluster_size - 1);
 }

-static inline uint64_t size_to_clusters(BDRVQcow2State *s, uint64_t size)
+static inline int size_to_clusters(BDRVQcowState *s, int64_t size)
 {
    return (size + (s->cluster_size - 1)) >> s->cluster_bits;
 }

-static inline int64_t size_to_l1(BDRVQcow2State *s, int64_t size)
+static inline int64_t size_to_l1(BDRVQcowState *s, int64_t size)
 {
    int shift = s->cluster_bits + s->l2_bits;
    return (size + (1ULL << shift) - 1) >> shift;
 }

-static inline int offset_to_l2_index(BDRVQcow2State *s, int64_t offset)
+static inline int offset_to_l2_index(BDRVQcowState *s, int64_t offset)
 {
    return (offset >> s->cluster_bits) & (s->l2_size - 1);
 }
@@ -435,12 +434,12 @@ static inline int64_t align_offset(int64_t offset, int n)
    return offset;
 }

-static inline int64_t qcow2_vm_state_offset(BDRVQcow2State *s)
+static inline int64_t qcow2_vm_state_offset(BDRVQcowState *s)
 {
    return (int64_t)s->l1_vm_state_index << (s->cluster_bits + s->l2_bits);
 }

-static inline uint64_t qcow2_max_refcount_clusters(BDRVQcow2State *s)
+static inline uint64_t qcow2_max_refcount_clusters(BDRVQcowState *s)
 {
    return QCOW_MAX_REFTABLE_SIZE >> s->cluster_bits;
 }
@@ -459,7 +458,7 @@ static inline int qcow2_get_cluster_type(uint64_t l2_entry)
 }

 /* Check whether refcounts are eager or lazy */
-static inline bool qcow2_need_accurate_refcounts(BDRVQcow2State *s)
+static inline bool qcow2_need_accurate_refcounts(BDRVQcowState *s)
 {
    return !(s->incompatible_features & QCOW2_INCOMPAT_DIRTY);
 }
@@ -507,8 +506,8 @@ int qcow2_update_cluster_refcount(BlockDriverState *bs, int64_t cluster_index,
                                  enum qcow2_discard_type type);

 int64_t qcow2_alloc_clusters(BlockDriverState *bs, uint64_t size);
-int64_t qcow2_alloc_clusters_at(BlockDriverState *bs, uint64_t offset,
-                                int64_t nb_clusters);
+int qcow2_alloc_clusters_at(BlockDriverState *bs, uint64_t offset,
+    int nb_clusters);
 int64_t qcow2_alloc_bytes(BlockDriverState *bs, int size);
 void qcow2_free_clusters(BlockDriverState *bs,
                          int64_t offset, int64_t size,
@@ -529,19 +528,16 @@ int qcow2_check_metadata_overlap(BlockDriverState *bs, int ign, int64_t offset,
 int qcow2_pre_write_overlap_check(BlockDriverState *bs, int ign, int64_t offset,
                                  int64_t size);

-int qcow2_change_refcount_order(BlockDriverState *bs, int refcount_order,
-                                BlockDriverAmendStatusCB *status_cb,
-                                void *cb_opaque, Error **errp);
-
 /* qcow2-cluster.c functions */
 int qcow2_grow_l1_table(BlockDriverState *bs, uint64_t min_size,
                        bool exact_size);
 int qcow2_write_l1_entry(BlockDriverState *bs, int l1_index);
 void qcow2_l2_cache_reset(BlockDriverState *bs);
 int qcow2_decompress_cluster(BlockDriverState *bs, uint64_t cluster_offset);
-int qcow2_encrypt_sectors(BDRVQcow2State *s, int64_t sector_num,
-                          uint8_t *out_buf, const uint8_t *in_buf,
-                          int nb_sectors, bool enc, Error **errp);
+void qcow2_encrypt_sectors(BDRVQcowState *s, int64_t sector_num,
+                     uint8_t *out_buf, const uint8_t *in_buf,
+                     int nb_sectors, int enc,
+                     const AES_KEY *key);

 int qcow2_get_cluster_offset(BlockDriverState *bs, uint64_t offset,
    int *num, uint64_t *cluster_offset);
@@ -557,8 +553,7 @@ int qcow2_discard_clusters(BlockDriverState *bs, uint64_t offset,
 int qcow2_zero_clusters(BlockDriverState *bs, uint64_t offset, int nb_sectors);

 int qcow2_expand_zero_clusters(BlockDriverState *bs,
-                               BlockDriverAmendStatusCB *status_cb,
-                               void *cb_opaque);
+                               BlockDriverAmendStatusCB *status_cb);

 /* qcow2-snapshot.c functions */
 int qcow2_snapshot_create(BlockDriverState *bs, QEMUSnapshotInfo *sn_info);
@@ -580,20 +575,18 @@ int qcow2_read_snapshots(BlockDriverState *bs);
 Qcow2Cache *qcow2_cache_create(BlockDriverState *bs, int num_tables);
 int qcow2_cache_destroy(BlockDriverState* bs, Qcow2Cache *c);

-void qcow2_cache_entry_mark_dirty(BlockDriverState *bs, Qcow2Cache *c,
-     void *table);
+void qcow2_cache_entry_mark_dirty(Qcow2Cache *c, void *table);
 int qcow2_cache_flush(BlockDriverState *bs, Qcow2Cache *c);
 int qcow2_cache_set_dependency(BlockDriverState *bs, Qcow2Cache *c,
    Qcow2Cache *dependency);
 void qcow2_cache_depends_on_flush(Qcow2Cache *c);

-void qcow2_cache_clean_unused(BlockDriverState *bs, Qcow2Cache *c);
 int qcow2_cache_empty(BlockDriverState *bs, Qcow2Cache *c);

 int qcow2_cache_get(BlockDriverState *bs, Qcow2Cache *c, uint64_t offset,
    void **table);
 int qcow2_cache_get_empty(BlockDriverState *bs, Qcow2Cache *c, uint64_t offset,
    void **table);
-void qcow2_cache_put(BlockDriverState *bs, Qcow2Cache *c, void **table);
+int qcow2_cache_put(BlockDriverState *bs, Qcow2Cache *c, void **table);

 #endif
--- a/block/qed-check.c
+++ b/block/qed-check.c
@@ -11,7 +11,6 @@
 *
 */

-#include "qemu/osdep.h"
 #include "qed.h"

 typedef struct {
--- a/block/qed-cluster.c
+++ b/block/qed-cluster.c
@@ -12,7 +12,6 @@
 *
 */

-#include "qemu/osdep.h"
 #include "qed.h"

 /**
--- a/block/qed-gencb.c
+++ b/block/qed-gencb.c
@@ -11,7 +11,6 @@
 *
 */

-#include "qemu/osdep.h"
 #include "qed.h"

 void *gencb_alloc(size_t len, BlockCompletionFunc *cb, void *opaque)
--- a/block/qed-l2-cache.c
+++ b/block/qed-l2-cache.c
@@ -50,7 +50,6 @@
 * table will be deleted in favor of the existing cache entry.
 */

-#include "qemu/osdep.h"
 #include "trace.h"
 #include "qed.h"

--- a/block/qed-table.c
+++ b/block/qed-table.c
@@ -12,7 +12,6 @@
 *
 */

-#include "qemu/osdep.h"
 #include "trace.h"
 #include "qemu/sockets.h" /* for EINPROGRESS on Windows */
 #include "qed.h"
@@ -64,7 +63,7 @@ static void qed_read_table(BDRVQEDState *s, uint64_t offset, QEDTable *table,
    read_table_cb->iov.iov_len = s->header.cluster_size * s->header.table_size,

    qemu_iovec_init_external(qiov, &read_table_cb->iov, 1);
-    bdrv_aio_readv(s->bs->file->bs, offset / BDRV_SECTOR_SIZE, qiov,
+    bdrv_aio_readv(s->bs->file, offset / BDRV_SECTOR_SIZE, qiov,
                   qiov->size / BDRV_SECTOR_SIZE,
                   qed_read_table_cb, read_table_cb);
 }
@@ -153,7 +152,7 @@ static void qed_write_table(BDRVQEDState *s, uint64_t offset, QEDTable *table,
    /* Adjust for offset into table */
    offset += start * sizeof(uint64_t);

-    bdrv_aio_writev(s->bs->file->bs, offset / BDRV_SECTOR_SIZE,
+    bdrv_aio_writev(s->bs->file, offset / BDRV_SECTOR_SIZE,
                    &write_table_cb->qiov,
                    write_table_cb->qiov.size / BDRV_SECTOR_SIZE,
                    qed_write_table_cb, write_table_cb);
--- a/block/qed.c
+++ b/block/qed.c
@@ -12,7 +12,6 @@
 *
 */

-#include "qemu/osdep.h"
 #include "qemu/timer.h"
 #include "trace.h"
 #include "qed.h"
@@ -83,7 +82,7 @@ int qed_write_header_sync(BDRVQEDState *s)
    int ret;

    qed_header_cpu_to_le(&s->header, &le);
-    ret = bdrv_pwrite(s->bs->file->bs, 0, &le, sizeof(le));
+    ret = bdrv_pwrite(s->bs->file, 0, &le, sizeof(le));
    if (ret != sizeof(le)) {
        return ret;
    }
@@ -120,7 +119,7 @@ static void qed_write_header_read_cb(void *opaque, int ret)
    /* Update header */
    qed_header_cpu_to_le(&s->header, (QEDHeader *)write_header_cb->buf);

-    bdrv_aio_writev(s->bs->file->bs, 0, &write_header_cb->qiov,
+    bdrv_aio_writev(s->bs->file, 0, &write_header_cb->qiov,
                    write_header_cb->nsectors, qed_write_header_cb,
                    write_header_cb);
 }
@@ -153,7 +152,7 @@ static void qed_write_header(BDRVQEDState *s, BlockCompletionFunc cb,
    write_header_cb->iov.iov_len = len;
    qemu_iovec_init_external(&write_header_cb->qiov, &write_header_cb->iov, 1);

-    bdrv_aio_readv(s->bs->file->bs, 0, &write_header_cb->qiov, nsectors,
+    bdrv_aio_readv(s->bs->file, 0, &write_header_cb->qiov, nsectors,
                   qed_write_header_read_cb, write_header_cb);
 }

@@ -355,6 +354,12 @@ static void qed_cancel_need_check_timer(BDRVQEDState *s)
    timer_del(s->need_check_timer);
 }

+static void bdrv_qed_rebind(BlockDriverState *bs)
+{
+    BDRVQEDState *s = bs->opaque;
+    s->bs = bs;
+}
+
 static void bdrv_qed_detach_aio_context(BlockDriverState *bs)
 {
    BDRVQEDState *s = bs->opaque;
@@ -376,18 +381,6 @@ static void bdrv_qed_attach_aio_context(BlockDriverState *bs,
    }
 }

-static void bdrv_qed_drain(BlockDriverState *bs)
-{
-    BDRVQEDState *s = bs->opaque;
-
-    /* Cancel timer and start doing I/O that were meant to happen as if it
-     * fired, that way we get bdrv_drain() taking care of the ongoing requests
-     * correctly. */
-    qed_cancel_need_check_timer(s);
-    qed_plug_allocating_write_reqs(s);
-    bdrv_aio_flush(s->bs, qed_clear_need_check, s);
-}
-
 static int bdrv_qed_open(BlockDriverState *bs, QDict *options, int flags,
                         Error **errp)
 {
@@ -399,7 +392,7 @@ static int bdrv_qed_open(BlockDriverState *bs, QDict *options, int flags,
    s->bs = bs;
    QSIMPLEQ_INIT(&s->allocating_write_reqs);

-    ret = bdrv_pread(bs->file->bs, 0, &le_header, sizeof(le_header));
+    ret = bdrv_pread(bs->file, 0, &le_header, sizeof(le_header));
    if (ret < 0) {
        return ret;
    }
@@ -414,8 +407,8 @@ static int bdrv_qed_open(BlockDriverState *bs, QDict *options, int flags,
        char buf[64];
        snprintf(buf, sizeof(buf), "%" PRIx64,
            s->header.features & ~QED_FEATURE_MASK);
-        error_setg(errp, QERR_UNKNOWN_BLOCK_FORMAT_FEATURE,
-                   bdrv_get_device_or_node_name(bs), "QED", buf);
+        error_set(errp, QERR_UNKNOWN_BLOCK_FORMAT_FEATURE,
+            bdrv_get_device_name(bs), "QED", buf);
        return -ENOTSUP;
    }
    if (!qed_is_cluster_size_valid(s->header.cluster_size)) {
@@ -423,7 +416,7 @@ static int bdrv_qed_open(BlockDriverState *bs, QDict *options, int flags,
    }

    /* Round down file size to the last cluster */
-    file_size = bdrv_getlength(bs->file->bs);
+    file_size = bdrv_getlength(bs->file);
    if (file_size < 0) {
        return file_size;
    }
@@ -443,9 +436,9 @@ static int bdrv_qed_open(BlockDriverState *bs, QDict *options, int flags,

    s->table_nelems = (s->header.cluster_size * s->header.table_size) /
                      sizeof(uint64_t);
-    s->l2_shift = ctz32(s->header.cluster_size);
+    s->l2_shift = ffs(s->header.cluster_size) - 1;
    s->l2_mask = s->table_nelems - 1;
-    s->l1_shift = s->l2_shift + ctz32(s->table_nelems);
+    s->l1_shift = s->l2_shift + ffs(s->table_nelems) - 1;

    /* Header size calculation must not overflow uint32_t */
    if (s->header.header_size > UINT32_MAX / s->header.cluster_size) {
@@ -459,7 +452,7 @@ static int bdrv_qed_open(BlockDriverState *bs, QDict *options, int flags,
            return -EINVAL;
        }

-        ret = qed_read_string(bs->file->bs, s->header.backing_filename_offset,
+        ret = qed_read_string(bs->file, s->header.backing_filename_offset,
                              s->header.backing_filename_size, bs->backing_file,
                              sizeof(bs->backing_file));
        if (ret < 0) {
@@ -478,7 +471,7 @@ static int bdrv_qed_open(BlockDriverState *bs, QDict *options, int flags,
     * feature is no longer valid.
     */
    if ((s->header.autoclear_features & ~QED_AUTOCLEAR_FEATURE_MASK) != 0 &&
-        !bdrv_is_read_only(bs->file->bs) && !(flags & BDRV_O_INACTIVE)) {
+        !bdrv_is_read_only(bs->file) && !(flags & BDRV_O_INCOMING)) {
        s->header.autoclear_features &= QED_AUTOCLEAR_FEATURE_MASK;

        ret = qed_write_header_sync(s);
@@ -487,7 +480,7 @@ static int bdrv_qed_open(BlockDriverState *bs, QDict *options, int flags,
        }

        /* From here on only known autoclear feature bits are valid */
-        bdrv_flush(bs->file->bs);
+        bdrv_flush(bs->file);
    }

    s->l1_table = qed_alloc_table(s);
@@ -505,8 +498,8 @@ static int bdrv_qed_open(BlockDriverState *bs, QDict *options, int flags,
         * potentially inconsistent images to be opened read-only.  This can
         * aid data recovery from an otherwise inconsistent image.
         */
-        if (!bdrv_is_read_only(bs->file->bs) &&
-            !(flags & BDRV_O_INACTIVE)) {
+        if (!bdrv_is_read_only(bs->file) &&
+            !(flags & BDRV_O_INCOMING)) {
            BdrvCheckResult result = {0};

            ret = qed_check(s, &result, true);
@@ -548,7 +541,7 @@ static void bdrv_qed_close(BlockDriverState *bs)
    bdrv_qed_detach_aio_context(bs);

    /* Ensure writes reach stable storage */
-    bdrv_flush(bs->file->bs);
+    bdrv_flush(bs->file);

    /* Clean shutdown, no check required on next open */
    if (s->header.features & QED_F_NEED_CHECK) {
@@ -590,7 +583,7 @@ static int qed_create(const char *filename, uint32_t cluster_size,

    bs = NULL;
    ret = bdrv_open(&bs, filename, NULL, NULL,
-                    BDRV_O_RDWR | BDRV_O_CACHE_WB | BDRV_O_PROTOCOL,
+                    BDRV_O_RDWR | BDRV_O_CACHE_WB | BDRV_O_PROTOCOL, NULL,
                    &local_err);
    if (ret < 0) {
        error_propagate(errp, local_err);
@@ -693,7 +686,6 @@ typedef struct {
    uint64_t pos;
    int64_t status;
    int *pnum;
-    BlockDriverState **file;
 } QEDIsAllocatedCB;

 static void qed_is_allocated_cb(void *opaque, int ret, uint64_t offset, size_t len)
@@ -705,7 +697,6 @@ static void qed_is_allocated_cb(void *opaque, int ret, uint64_t offset, size_t l
    case QED_CLUSTER_FOUND:
        offset |= qed_offset_into_cluster(s, cb->pos);
        cb->status = BDRV_BLOCK_DATA | BDRV_BLOCK_OFFSET_VALID | offset;
-        *cb->file = cb->bs->file->bs;
        break;
    case QED_CLUSTER_ZERO:
        cb->status = BDRV_BLOCK_ZERO;
@@ -727,8 +718,7 @@ static void qed_is_allocated_cb(void *opaque, int ret, uint64_t offset, size_t l

 static int64_t coroutine_fn bdrv_qed_co_get_block_status(BlockDriverState *bs,
                                                 int64_t sector_num,
-                                                 int nb_sectors, int *pnum,
-                                                 BlockDriverState **file)
+                                                 int nb_sectors, int *pnum)
 {
    BDRVQEDState *s = bs->opaque;
    size_t len = (size_t)nb_sectors * BDRV_SECTOR_SIZE;
@@ -737,7 +727,6 @@ static int64_t coroutine_fn bdrv_qed_co_get_block_status(BlockDriverState *bs,
        .pos = (uint64_t)sector_num * BDRV_SECTOR_SIZE,
        .status = BDRV_BLOCK_OFFSET_MASK,
        .pnum = pnum,
-        .file = file,
    };
    QEDRequest request = { .l2_table = NULL };

@@ -783,8 +772,8 @@ static void qed_read_backing_file(BDRVQEDState *s, uint64_t pos,
    /* If there is a backing file, get its length.  Treat the absence of a
     * backing file like a zero length backing file.
     */
-    if (s->bs->backing) {
-        int64_t l = bdrv_getlength(s->bs->backing->bs);
+    if (s->bs->backing_hd) {
+        int64_t l = bdrv_getlength(s->bs->backing_hd);
        if (l < 0) {
            cb(opaque, l);
            return;
@@ -813,7 +802,7 @@ static void qed_read_backing_file(BDRVQEDState *s, uint64_t pos,
    qemu_iovec_concat(*backing_qiov, qiov, 0, size);

    BLKDBG_EVENT(s->bs->file, BLKDBG_READ_BACKING_AIO);
-    bdrv_aio_readv(s->bs->backing->bs, pos / BDRV_SECTOR_SIZE,
+    bdrv_aio_readv(s->bs->backing_hd, pos / BDRV_SECTOR_SIZE,
                   *backing_qiov, size / BDRV_SECTOR_SIZE, cb, opaque);
 }

@@ -850,7 +839,7 @@ static void qed_copy_from_backing_file_write(void *opaque, int ret)
    }

    BLKDBG_EVENT(s->bs->file, BLKDBG_COW_WRITE);
-    bdrv_aio_writev(s->bs->file->bs, copy_cb->offset / BDRV_SECTOR_SIZE,
+    bdrv_aio_writev(s->bs->file, copy_cb->offset / BDRV_SECTOR_SIZE,
                    &copy_cb->qiov, copy_cb->qiov.size / BDRV_SECTOR_SIZE,
                    qed_copy_from_backing_file_cb, copy_cb);
 }
@@ -1066,7 +1055,7 @@ static void qed_aio_write_flush_before_l2_update(void *opaque, int ret)
    QEDAIOCB *acb = opaque;
    BDRVQEDState *s = acb_to_s(acb);

-    if (!bdrv_aio_flush(s->bs->file->bs, qed_aio_write_l2_update_cb, opaque)) {
+    if (!bdrv_aio_flush(s->bs->file, qed_aio_write_l2_update_cb, opaque)) {
        qed_aio_complete(acb, -EIO);
    }
 }
@@ -1092,7 +1081,7 @@ static void qed_aio_write_main(void *opaque, int ret)
    if (acb->find_cluster_ret == QED_CLUSTER_FOUND) {
        next_fn = qed_aio_next_io;
    } else {
-        if (s->bs->backing) {
+        if (s->bs->backing_hd) {
            next_fn = qed_aio_write_flush_before_l2_update;
        } else {
            next_fn = qed_aio_write_l2_update_cb;
@@ -1100,7 +1089,7 @@ static void qed_aio_write_main(void *opaque, int ret)
    }

    BLKDBG_EVENT(s->bs->file, BLKDBG_WRITE_AIO);
-    bdrv_aio_writev(s->bs->file->bs, offset / BDRV_SECTOR_SIZE,
+    bdrv_aio_writev(s->bs->file, offset / BDRV_SECTOR_SIZE,
                    &acb->cur_qiov, acb->cur_qiov.size / BDRV_SECTOR_SIZE,
                    next_fn, acb);
 }
@@ -1150,7 +1139,7 @@ static void qed_aio_write_prefill(void *opaque, int ret)
 static bool qed_should_set_need_check(BDRVQEDState *s)
 {
    /* The flush before L2 update path ensures consistency */
-    if (s->bs->backing) {
+    if (s->bs->backing_hd) {
        return false;
    }

@@ -1332,7 +1321,7 @@ static void qed_aio_read_data(void *opaque, int ret,
    }

    BLKDBG_EVENT(bs->file, BLKDBG_READ_AIO);
-    bdrv_aio_readv(bs->file->bs, offset / BDRV_SECTOR_SIZE,
+    bdrv_aio_readv(bs->file, offset / BDRV_SECTOR_SIZE,
                   &acb->cur_qiov, acb->cur_qiov.size / BDRV_SECTOR_SIZE,
                   qed_aio_next_io, acb);
    return;
@@ -1454,7 +1443,7 @@ static int coroutine_fn bdrv_qed_co_write_zeroes(BlockDriverState *bs,
    struct iovec iov;

    /* Refuse if there are untouched backing file sectors */
-    if (bs->backing) {
+    if (bs->backing_hd) {
        if (qed_offset_into_cluster(s, sector_num * BDRV_SECTOR_SIZE) != 0) {
            return -ENOTSUP;
        }
@@ -1591,7 +1580,7 @@ static int bdrv_qed_change_backing_file(BlockDriverState *bs,
    }

    /* Write new header */
-    ret = bdrv_pwrite_sync(bs->file->bs, 0, buffer, buffer_len);
+    ret = bdrv_pwrite_sync(bs->file, 0, buffer, buffer_len);
    g_free(buffer);
    if (ret == 0) {
        memcpy(&s->header, &new_header, sizeof(new_header));
@@ -1607,7 +1596,7 @@ static void bdrv_qed_invalidate_cache(BlockDriverState *bs, Error **errp)

    bdrv_qed_close(bs);

-    bdrv_invalidate_cache(bs->file->bs, &local_err);
+    bdrv_invalidate_cache(bs->file, &local_err);
    if (local_err) {
        error_propagate(errp, local_err);
        return;
@@ -1616,8 +1605,9 @@ static void bdrv_qed_invalidate_cache(BlockDriverState *bs, Error **errp)
    memset(s, 0, sizeof(BDRVQEDState));
    ret = bdrv_qed_open(bs, NULL, bs->open_flags, &local_err);
    if (local_err) {
-        error_propagate(errp, local_err);
-        error_prepend(errp, "Could not reopen qed layer: ");
+        error_setg(errp, "Could not reopen qed layer: %s",
+                   error_get_pretty(local_err));
+        error_free(local_err);
        return;
    } else if (ret < 0) {
        error_setg_errno(errp, -ret, "Could not reopen qed layer");
@@ -1674,6 +1664,7 @@ static BlockDriver bdrv_qed = {
    .supports_backing         = true,

    .bdrv_probe               = bdrv_qed_probe,
+    .bdrv_rebind              = bdrv_qed_rebind,
    .bdrv_open                = bdrv_qed_open,
    .bdrv_close               = bdrv_qed_close,
    .bdrv_reopen_prepare      = bdrv_qed_reopen_prepare,
@@ -1692,7 +1683,6 @@ static BlockDriver bdrv_qed = {
    .bdrv_check               = bdrv_qed_check,
    .bdrv_detach_aio_context  = bdrv_qed_detach_aio_context,
    .bdrv_attach_aio_context  = bdrv_qed_attach_aio_context,
-    .bdrv_drain               = bdrv_qed_drain,
 };

 static void bdrv_qed_init(void)
--- a/block/quorum.c
+++ b/block/quorum.c
@@ -13,17 +13,16 @@
 * See the COPYING file in the top-level directory.
 */

-#include "qemu/osdep.h"
+#include <gnutls/gnutls.h>
+#include <gnutls/crypto.h>
 #include "block/block_int.h"
 #include "qapi/qmp/qbool.h"
 #include "qapi/qmp/qdict.h"
-#include "qapi/qmp/qerror.h"
 #include "qapi/qmp/qint.h"
 #include "qapi/qmp/qjson.h"
 #include "qapi/qmp/qlist.h"
 #include "qapi/qmp/qstring.h"
 #include "qapi-event.h"
-#include "crypto/hash.h"

 #define HASH_LENGTH 32

@@ -34,7 +33,7 @@

 /* This union holds a vote hash value */
 typedef union QuorumVoteValue {
-    uint8_t h[HASH_LENGTH];    /* SHA-256 hash */
+    char h[HASH_LENGTH];       /* SHA-256 hash */
    int64_t l;                 /* simpler 64 bits hash */
 } QuorumVoteValue;

@@ -65,7 +64,7 @@ typedef struct QuorumVotes {

 /* the following structure holds the state of one quorum instance */
 typedef struct BDRVQuorumState {
-    BdrvChild **children;  /* children BlockDriverStates */
+    BlockDriverState **bs; /* children BlockDriverStates */
    int num_children;      /* children count */
    int threshold;         /* if less than threshold children reads gave the
                            * same result a quorum error occurs.
@@ -227,7 +226,10 @@ static void quorum_report_bad(QuorumAIOCB *acb, char *node_name, int ret)

 static void quorum_report_failure(QuorumAIOCB *acb)
 {
-    const char *reference = bdrv_get_device_or_node_name(acb->common.bs);
+    const char *reference = bdrv_get_device_name(acb->common.bs)[0] ?
+                            bdrv_get_device_name(acb->common.bs) :
+                            acb->common.bs->node_name;
+
    qapi_event_send_quorum_failure(reference, acb->sector_num,
                                   acb->nb_sectors, &error_abort);
 }
@@ -286,8 +288,7 @@ static void quorum_aio_cb(void *opaque, int ret)

    if (acb->is_read && s->read_pattern == QUORUM_READ_PATTERN_FIFO) {
        /* We try to read next child in FIFO order if we fail to read */
-        if (ret < 0 && (acb->child_iter + 1) < s->num_children) {
-            acb->child_iter++;
+        if (ret < 0 && ++acb->child_iter < s->num_children) {
            read_fifo_child(acb);
            return;
        }
@@ -338,7 +339,7 @@ static void quorum_report_bad_versions(BDRVQuorumState *s,
            continue;
        }
        QLIST_FOREACH(item, &version->items, next) {
-            quorum_report_bad(acb, s->children[item->index]->bs->node_name, 0);
+            quorum_report_bad(acb, s->bs[item->index]->node_name, 0);
        }
    }
 }
@@ -371,9 +372,8 @@ static bool quorum_rewrite_bad_versions(BDRVQuorumState *s, QuorumAIOCB *acb,
            continue;
        }
        QLIST_FOREACH(item, &version->items, next) {
-            bdrv_aio_writev(s->children[item->index]->bs, acb->sector_num,
-                            acb->qiov, acb->nb_sectors, quorum_rewrite_aio_cb,
-                            acb);
+            bdrv_aio_writev(s->bs[item->index], acb->sector_num, acb->qiov,
+                            acb->nb_sectors, quorum_rewrite_aio_cb, acb);
        }
    }

@@ -430,21 +430,25 @@ static void quorum_free_vote_list(QuorumVotes *votes)

 static int quorum_compute_hash(QuorumAIOCB *acb, int i, QuorumVoteValue *hash)
 {
+    int j, ret;
+    gnutls_hash_hd_t dig;
    QEMUIOVector *qiov = &acb->qcrs[i].qiov;
-    size_t len = sizeof(hash->h);
-    uint8_t *data = hash->h;

-    /* XXX - would be nice if we could pass in the Error **
-     * and propagate that back, but this quorum code is
-     * restricted to just errno values currently */
-    if (qcrypto_hash_bytesv(QCRYPTO_HASH_ALG_SHA256,
-                            qiov->iov, qiov->niov,
-                            &data, &len,
-                            NULL) < 0) {
-        return -EINVAL;
+    ret = gnutls_hash_init(&dig, GNUTLS_DIG_SHA256);
+
+    if (ret < 0) {
+        return ret;
    }

-    return 0;
+    for (j = 0; j < qiov->niov; j++) {
+        ret = gnutls_hash(dig, qiov->iov[j].iov_base, qiov->iov[j].iov_len);
+        if (ret < 0) {
+            break;
+        }
+    }
+
+    gnutls_hash_deinit(dig, (void *) hash);
+    return ret;
 }

 static QuorumVoteVersion *quorum_get_vote_winner(QuorumVotes *votes)
@@ -642,13 +646,13 @@ static BlockAIOCB *read_quorum_children(QuorumAIOCB *acb)
    int i;

    for (i = 0; i < s->num_children; i++) {
-        acb->qcrs[i].buf = qemu_blockalign(s->children[i]->bs, acb->qiov->size);
+        acb->qcrs[i].buf = qemu_blockalign(s->bs[i], acb->qiov->size);
        qemu_iovec_init(&acb->qcrs[i].qiov, acb->qiov->niov);
        qemu_iovec_clone(&acb->qcrs[i].qiov, acb->qiov, acb->qcrs[i].buf);
    }

    for (i = 0; i < s->num_children; i++) {
-        bdrv_aio_readv(s->children[i]->bs, acb->sector_num, &acb->qcrs[i].qiov,
+        bdrv_aio_readv(s->bs[i], acb->sector_num, &acb->qcrs[i].qiov,
                       acb->nb_sectors, quorum_aio_cb, &acb->qcrs[i]);
    }

@@ -659,12 +663,12 @@ static BlockAIOCB *read_fifo_child(QuorumAIOCB *acb)
 {
    BDRVQuorumState *s = acb->common.bs->opaque;

-    acb->qcrs[acb->child_iter].buf =
-        qemu_blockalign(s->children[acb->child_iter]->bs, acb->qiov->size);
+    acb->qcrs[acb->child_iter].buf = qemu_blockalign(s->bs[acb->child_iter],
+                                                     acb->qiov->size);
    qemu_iovec_init(&acb->qcrs[acb->child_iter].qiov, acb->qiov->niov);
    qemu_iovec_clone(&acb->qcrs[acb->child_iter].qiov, acb->qiov,
                     acb->qcrs[acb->child_iter].buf);
-    bdrv_aio_readv(s->children[acb->child_iter]->bs, acb->sector_num,
+    bdrv_aio_readv(s->bs[acb->child_iter], acb->sector_num,
                   &acb->qcrs[acb->child_iter].qiov, acb->nb_sectors,
                   quorum_aio_cb, &acb->qcrs[acb->child_iter]);

@@ -705,8 +709,8 @@ static BlockAIOCB *quorum_aio_writev(BlockDriverState *bs,
    int i;

    for (i = 0; i < s->num_children; i++) {
-        acb->qcrs[i].aiocb = bdrv_aio_writev(s->children[i]->bs, sector_num,
-                                             qiov, nb_sectors, &quorum_aio_cb,
+        acb->qcrs[i].aiocb = bdrv_aio_writev(s->bs[i], sector_num, qiov,
+                                             nb_sectors, &quorum_aio_cb,
                                             &acb->qcrs[i]);
    }

@@ -720,12 +724,12 @@ static int64_t quorum_getlength(BlockDriverState *bs)
    int i;

    /* check that all file have the same length */
-    result = bdrv_getlength(s->children[0]->bs);
+    result = bdrv_getlength(s->bs[0]);
    if (result < 0) {
        return result;
    }
    for (i = 1; i < s->num_children; i++) {
-        int64_t value = bdrv_getlength(s->children[i]->bs);
+        int64_t value = bdrv_getlength(s->bs[i]);
        if (value < 0) {
            return value;
        }
@@ -744,7 +748,7 @@ static void quorum_invalidate_cache(BlockDriverState *bs, Error **errp)
    int i;

    for (i = 0; i < s->num_children; i++) {
-        bdrv_invalidate_cache(s->children[i]->bs, &local_err);
+        bdrv_invalidate_cache(s->bs[i], &local_err);
        if (local_err) {
            error_propagate(errp, local_err);
            return;
@@ -765,7 +769,7 @@ static coroutine_fn int quorum_co_flush(BlockDriverState *bs)
    error_votes.compare = quorum_64bits_compare;

    for (i = 0; i < s->num_children; i++) {
-        result = bdrv_co_flush(s->children[i]->bs);
+        result = bdrv_co_flush(s->bs[i]);
        result_value.l = result;
        quorum_count_vote(&error_votes, &result_value, i);
    }
@@ -785,7 +789,7 @@ static bool quorum_recurse_is_first_non_filter(BlockDriverState *bs,
    int i;

    for (i = 0; i < s->num_children; i++) {
-        bool perm = bdrv_recurse_is_first_non_filter(s->children[i]->bs,
+        bool perm = bdrv_recurse_is_first_non_filter(s->bs[i],
                                                     candidate);
        if (perm) {
            return true;
@@ -799,8 +803,8 @@ static int quorum_valid_threshold(int threshold, int num_children, Error **errp)
 {

    if (threshold < 1) {
-        error_setg(errp, QERR_INVALID_PARAMETER_VALUE,
-                   "vote-threshold", "value >= 1");
+        error_set(errp, QERR_INVALID_PARAMETER_VALUE,
+                  "vote-threshold", "value >= 1");
        return -ERANGE;
    }

@@ -849,7 +853,7 @@ static int parse_read_pattern(const char *opt)
        return QUORUM_READ_PATTERN_QUORUM;
    }

-    for (i = 0; i < QUORUM_READ_PATTERN__MAX; i++) {
+    for (i = 0; i < QUORUM_READ_PATTERN_MAX; i++) {
        if (!strcmp(opt, QuorumReadPattern_lookup[i])) {
            return i;
        }
@@ -865,18 +869,25 @@ static int quorum_open(BlockDriverState *bs, QDict *options, int flags,
    Error *local_err = NULL;
    QemuOpts *opts = NULL;
    bool *opened;
+    QDict *sub = NULL;
+    QList *list = NULL;
+    const QListEntry *lentry;
    int i;
    int ret = 0;

    qdict_flatten(options);
+    qdict_extract_subqdict(options, &sub, "children.");
+    qdict_array_split(sub, &list);

-    /* count how many different children are present */
-    s->num_children = qdict_array_entries(options, "children.");
-    if (s->num_children < 0) {
-        error_setg(&local_err, "Option children is not a valid array");
+    if (qdict_size(sub)) {
+        error_setg(&local_err, "Invalid option children.%s",
+                   qdict_first(sub)->key);
        ret = -EINVAL;
        goto exit;
    }
+
+    /* count how many different children are present */
+    s->num_children = qlist_size(list);
    if (s->num_children < 2) {
        error_setg(&local_err,
                   "Number of provided children must be greater than 1");
@@ -892,12 +903,6 @@ static int quorum_open(BlockDriverState *bs, QDict *options, int flags,
    }

    s->threshold = qemu_opt_get_number(opts, QUORUM_OPT_VOTE_THRESHOLD, 0);
-    /* and validate it against s->num_children */
-    ret = quorum_valid_threshold(s->threshold, s->num_children, &local_err);
-    if (ret < 0) {
-        goto exit;
-    }
-
    ret = parse_read_pattern(qemu_opt_get(opts, QUORUM_OPT_READ_PATTERN));
    if (ret < 0) {
        error_setg(&local_err, "Please set read-pattern as fifo or quorum");
@@ -906,6 +911,12 @@ static int quorum_open(BlockDriverState *bs, QDict *options, int flags,
    s->read_pattern = ret;

    if (s->read_pattern == QUORUM_READ_PATTERN_QUORUM) {
+        /* and validate it against s->num_children */
+        ret = quorum_valid_threshold(s->threshold, s->num_children, &local_err);
+        if (ret < 0) {
+            goto exit;
+        }
+
        /* is the driver in blkverify mode */
        if (qemu_opt_get_bool(opts, QUORUM_OPT_BLKVERIFY, false) &&
            s->num_children == 2 && s->threshold == 2) {
@@ -925,22 +936,41 @@ static int quorum_open(BlockDriverState *bs, QDict *options, int flags,
        }
    }

-    /* allocate the children array */
-    s->children = g_new0(BdrvChild *, s->num_children);
+    /* allocate the children BlockDriverState array */
+    s->bs = g_new0(BlockDriverState *, s->num_children);
    opened = g_new0(bool, s->num_children);

-    for (i = 0; i < s->num_children; i++) {
-        char indexstr[32];
-        ret = snprintf(indexstr, 32, "children.%d", i);
-        assert(ret < 32);
+    for (i = 0, lentry = qlist_first(list); lentry;
+         lentry = qlist_next(lentry), i++) {
+        QDict *d;
+        QString *string;

-        s->children[i] = bdrv_open_child(NULL, options, indexstr, bs,
-                                         &child_format, false, &local_err);
-        if (local_err) {
-            ret = -EINVAL;
-            goto close_exit;
+        switch (qobject_type(lentry->value))
+        {
+            /* List of options */
+            case QTYPE_QDICT:
+                d = qobject_to_qdict(lentry->value);
+                QINCREF(d);
+                ret = bdrv_open(&s->bs[i], NULL, NULL, d, flags, NULL,
+                                &local_err);
+                break;
+
+            /* QMP reference */
+            case QTYPE_QSTRING:
+                string = qobject_to_qstring(lentry->value);
+                ret = bdrv_open(&s->bs[i], NULL, qstring_get_str(string), NULL,
+                                flags, NULL, &local_err);
+                break;
+
+            default:
+                error_setg(&local_err, "Specification of child block device %i "
+                           "is invalid", i);
+                ret = -EINVAL;
        }

+        if (ret < 0) {
+            goto close_exit;
+        }
        opened[i] = true;
    }

@@ -953,9 +983,9 @@ close_exit:
        if (!opened[i]) {
            continue;
        }
-        bdrv_unref_child(bs, s->children[i]);
+        bdrv_unref(s->bs[i]);
    }
-    g_free(s->children);
+    g_free(s->bs);
    g_free(opened);
 exit:
    qemu_opts_del(opts);
@@ -963,6 +993,8 @@ exit:
    if (local_err) {
        error_propagate(errp, local_err);
    }
+    QDECREF(list);
+    QDECREF(sub);
    return ret;
 }

@@ -972,10 +1004,10 @@ static void quorum_close(BlockDriverState *bs)
    int i;

    for (i = 0; i < s->num_children; i++) {
-        bdrv_unref_child(bs, s->children[i]);
+        bdrv_unref(s->bs[i]);
    }

-    g_free(s->children);
+    g_free(s->bs);
 }

 static void quorum_detach_aio_context(BlockDriverState *bs)
@@ -984,7 +1016,7 @@ static void quorum_detach_aio_context(BlockDriverState *bs)
    int i;

    for (i = 0; i < s->num_children; i++) {
-        bdrv_detach_aio_context(s->children[i]->bs);
+        bdrv_detach_aio_context(s->bs[i]);
    }
 }

@@ -995,11 +1027,11 @@ static void quorum_attach_aio_context(BlockDriverState *bs,
    int i;

    for (i = 0; i < s->num_children; i++) {
-        bdrv_attach_aio_context(s->children[i]->bs, new_context);
+        bdrv_attach_aio_context(s->bs[i], new_context);
    }
 }

-static void quorum_refresh_filename(BlockDriverState *bs, QDict *options)
+static void quorum_refresh_filename(BlockDriverState *bs)
 {
    BDRVQuorumState *s = bs->opaque;
    QDict *opts;
@@ -1007,17 +1039,16 @@ static void quorum_refresh_filename(BlockDriverState *bs, QDict *options)
    int i;

    for (i = 0; i < s->num_children; i++) {
-        bdrv_refresh_filename(s->children[i]->bs);
-        if (!s->children[i]->bs->full_open_options) {
+        bdrv_refresh_filename(s->bs[i]);
+        if (!s->bs[i]->full_open_options) {
            return;
        }
    }

    children = qlist_new();
    for (i = 0; i < s->num_children; i++) {
-        QINCREF(s->children[i]->bs->full_open_options);
-        qlist_append_obj(children,
-                         QOBJECT(s->children[i]->bs->full_open_options));
+        QINCREF(s->bs[i]->full_open_options);
+        qlist_append_obj(children, QOBJECT(s->bs[i]->full_open_options));
    }

    opts = qdict_new();
@@ -1025,9 +1056,9 @@ static void quorum_refresh_filename(BlockDriverState *bs, QDict *options)
    qdict_put_obj(opts, QUORUM_OPT_VOTE_THRESHOLD,
                  QOBJECT(qint_from_int(s->threshold)));
    qdict_put_obj(opts, QUORUM_OPT_BLKVERIFY,
-                  QOBJECT(qbool_from_bool(s->is_blkverify)));
+                  QOBJECT(qbool_from_int(s->is_blkverify)));
    qdict_put_obj(opts, QUORUM_OPT_REWRITE,
-                  QOBJECT(qbool_from_bool(s->rewrite_corrupted)));
+                  QOBJECT(qbool_from_int(s->rewrite_corrupted)));
    qdict_put_obj(opts, "children", QOBJECT(children));

    bs->full_open_options = opts;
@@ -1060,10 +1091,6 @@ static BlockDriver bdrv_quorum = {

 static void bdrv_quorum_init(void)
 {
-    if (!qcrypto_hash_supports(QCRYPTO_HASH_ALG_SHA256)) {
-        /* SHA256 hash support is required for quorum device */
-        return;
-    }
    bdrv_register(&bdrv_quorum);
 }

--- a/block/raw-posix.c
+++ b/block/raw-posix.c
@@ -21,9 +21,7 @@
 * 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 "qemu/error-report.h"
 #include "qemu/timer.h"
 #include "qemu/log.h"
 #include "block/block_int.h"
@@ -33,7 +31,6 @@
 #include "qemu/iov.h"
 #include "raw-aio.h"
 #include "qapi/util.h"
-#include "qapi/qmp/qstring.h"

 #if defined(__APPLE__) && (__MACH__)
 #include <paths.h>
@@ -52,13 +49,14 @@
 #include <sys/dkio.h>
 #endif
 #ifdef __linux__
+#include <sys/types.h>
+#include <sys/stat.h>
 #include <sys/ioctl.h>
 #include <sys/param.h>
 #include <linux/cdrom.h>
 #include <linux/fd.h>
 #include <linux/fs.h>
 #include <linux/hdreg.h>
-#include <scsi/sg.h>
 #ifdef __s390__
 #include <asm/dasd.h>
 #endif
@@ -96,19 +94,15 @@
 #include <xfs/xfs.h>
 #endif

-//#define DEBUG_BLOCK
+//#define DEBUG_FLOPPY

-#ifdef DEBUG_BLOCK
-# define DEBUG_BLOCK_PRINT 1
+//#define DEBUG_BLOCK
+#if defined(DEBUG_BLOCK)
+#define DEBUG_BLOCK_PRINT(formatCstr, ...) do { if (qemu_log_enabled()) \
+    { qemu_log(formatCstr, ## __VA_ARGS__); qemu_log_flush(); } } while (0)
 #else
-# define DEBUG_BLOCK_PRINT 0
+#define DEBUG_BLOCK_PRINT(formatCstr, ...)
 #endif
-#define DPRINTF(fmt, ...) \
-do { \
-    if (DEBUG_BLOCK_PRINT) { \
-        printf(fmt, ## __VA_ARGS__); \
-    } \
-} while (0)

 /* OS X does not have O_DSYNC */
 #ifndef O_DSYNC
@@ -126,6 +120,11 @@ do { \

 #define FTYPE_FILE   0
 #define FTYPE_CD     1
+#define FTYPE_FD     2
+
+/* if the FD is not accessed during that time (in ns), we try to
+   reopen it to see if the disk has been changed */
+#define FD_OPEN_TIMEOUT (1000000000)

 #define MAX_BLOCKSIZE	4096

@@ -135,6 +134,13 @@ typedef struct BDRVRawState {
    int open_flags;
    size_t buf_align;

+#if defined(__linux__)
+    /* linux floppy specific */
+    int64_t fd_open_time;
+    int64_t fd_error_time;
+    int fd_got_error;
+    int fd_media_changed;
+#endif
 #ifdef CONFIG_LINUX_AIO
    int use_aio;
    void *aio_ctx;
@@ -295,11 +301,10 @@ static void raw_probe_alignment(BlockDriverState *bs, int fd, Error **errp)
 {
    BDRVRawState *s = bs->opaque;
    char *buf;
-    size_t max_align = MAX(MAX_BLOCKSIZE, getpagesize());

-    /* For SCSI generic devices the alignment is not really used.
+    /* For /dev/sg devices the alignment is not really used.
       With buffered I/O, we don't have any restrictions. */
-    if (bdrv_is_sg(bs) || !s->needs_alignment) {
+    if (bs->sg || !s->needs_alignment) {
        bs->request_alignment = 1;
        s->buf_align = 1;
        return;
@@ -325,9 +330,9 @@ static void raw_probe_alignment(BlockDriverState *bs, int fd, Error **errp)
    /* If we could not get the sizes so far, we can only guess them */
    if (!s->buf_align) {
        size_t align;
-        buf = qemu_memalign(max_align, 2 * max_align);
-        for (align = 512; align <= max_align; align <<= 1) {
-            if (raw_is_io_aligned(fd, buf + align, max_align)) {
+        buf = qemu_memalign(MAX_BLOCKSIZE, 2 * MAX_BLOCKSIZE);
+        for (align = 512; align <= MAX_BLOCKSIZE; align <<= 1) {
+            if (raw_is_io_aligned(fd, buf + align, MAX_BLOCKSIZE)) {
                s->buf_align = align;
                break;
            }
@@ -337,8 +342,8 @@ static void raw_probe_alignment(BlockDriverState *bs, int fd, Error **errp)

    if (!bs->request_alignment) {
        size_t align;
-        buf = qemu_memalign(s->buf_align, max_align);
-        for (align = 512; align <= max_align; align <<= 1) {
+        buf = qemu_memalign(s->buf_align, MAX_BLOCKSIZE);
+        for (align = 512; align <= MAX_BLOCKSIZE; align <<= 1) {
            if (raw_is_io_aligned(fd, buf, align)) {
                bs->request_alignment = align;
                break;
@@ -499,19 +504,14 @@ static int raw_open_common(BlockDriverState *bs, QDict *options,
        goto fail;
    }
    if (!s->use_aio && (bdrv_flags & BDRV_O_NATIVE_AIO)) {
-        error_setg(errp, "aio=native was specified, but it requires "
-                         "cache.direct=on, which was not specified.");
-        ret = -EINVAL;
-        goto fail;
+        error_printf("WARNING: aio=native was specified for '%s', but "
+                     "it requires cache.direct=on, which was not "
+                     "specified. Falling back to aio=threads.\n"
+                     "         This will become an error condition in "
+                     "future QEMU versions.\n",
+                     bs->filename);
    }
-#else
-    if (bdrv_flags & BDRV_O_NATIVE_AIO) {
-        error_setg(errp, "aio=native was specified, but is not supported "
-                         "in this build.");
-        ret = -EINVAL;
-        goto fail;
-    }
-#endif /* !defined(CONFIG_LINUX_AIO) */
+#endif

    s->has_discard = true;
    s->has_write_zeroes = true;
@@ -618,7 +618,7 @@ static int raw_reopen_prepare(BDRVReopenState *state,
    }
 #endif

-    if (s->type == FTYPE_CD) {
+    if (s->type == FTYPE_FD || s->type == FTYPE_CD) {
        raw_s->open_flags |= O_NONBLOCK;
    }

@@ -662,17 +662,11 @@ static int raw_reopen_prepare(BDRVReopenState *state,

    /* If we cannot use fcntl, or fcntl failed, fall back to qemu_open() */
    if (raw_s->fd == -1) {
-        const char *normalized_filename = state->bs->filename;
-        ret = raw_normalize_devicepath(&normalized_filename);
-        if (ret < 0) {
-            error_setg_errno(errp, -ret, "Could not normalize device path");
-        } else {
-            assert(!(raw_s->open_flags & O_CREAT));
-            raw_s->fd = qemu_open(normalized_filename, raw_s->open_flags);
-            if (raw_s->fd == -1) {
-                error_setg_errno(errp, errno, "Could not reopen file");
-                ret = -1;
-            }
+        assert(!(raw_s->open_flags & O_CREAT));
+        raw_s->fd = qemu_open(state->bs->filename, raw_s->open_flags);
+        if (raw_s->fd == -1) {
+            error_setg_errno(errp, errno, "Could not reopen file");
+            ret = -1;
        }
    }

@@ -731,8 +725,7 @@ static void raw_refresh_limits(BlockDriverState *bs, Error **errp)
    BDRVRawState *s = bs->opaque;

    raw_probe_alignment(bs, s->fd, errp);
-    bs->bl.min_mem_alignment = s->buf_align;
-    bs->bl.opt_mem_alignment = MAX(s->buf_align, getpagesize());
+    bs->bl.opt_mem_alignment = s->buf_align;
 }

 static int check_for_dasd(int fd)
@@ -778,6 +771,7 @@ static int hdev_probe_geometry(BlockDriverState *bs, HDGeometry *geo)
 {
    BDRVRawState *s = bs->opaque;
    struct hd_geometry ioctl_geo = {0};
+    uint32_t blksize;

    /* If DASD, get its geometry */
    if (check_for_dasd(s->fd) < 0) {
@@ -797,6 +791,12 @@ static int hdev_probe_geometry(BlockDriverState *bs, HDGeometry *geo)
    }
    geo->heads = ioctl_geo.heads;
    geo->sectors = ioctl_geo.sectors;
+    if (!probe_physical_blocksize(s->fd, &blksize)) {
+        /* overwrite cyls: HDIO_GETGEO result is incorrect for big drives */
+        geo->cylinders = bdrv_nb_sectors(bs) / (blksize / BDRV_SECTOR_SIZE)
+                                             / (geo->heads * geo->sectors);
+        return 0;
+    }
    geo->cylinders = ioctl_geo.cylinders;

    return 0;
@@ -1016,7 +1016,6 @@ static ssize_t handle_aiocb_rw(RawPosixAIOData *aiocb)
 static int xfs_write_zeroes(BDRVRawState *s, int64_t offset, uint64_t bytes)
 {
    struct xfs_flock64 fl;
-    int err;

    memset(&fl, 0, sizeof(fl));
    fl.l_whence = SEEK_SET;
@@ -1024,9 +1023,8 @@ static int xfs_write_zeroes(BDRVRawState *s, int64_t offset, uint64_t bytes)
    fl.l_len = bytes;

    if (xfsctl(NULL, s->fd, XFS_IOC_ZERO_RANGE, &fl) < 0) {
-        err = errno;
-        DPRINTF("cannot write zero range (%s)\n", strerror(errno));
-        return -err;
+        DEBUG_BLOCK_PRINT("cannot write zero range (%s)\n", strerror(errno));
+        return -errno;
    }

    return 0;
@@ -1035,7 +1033,6 @@ static int xfs_write_zeroes(BDRVRawState *s, int64_t offset, uint64_t bytes)
 static int xfs_discard(BDRVRawState *s, int64_t offset, uint64_t bytes)
 {
    struct xfs_flock64 fl;
-    int err;

    memset(&fl, 0, sizeof(fl));
    fl.l_whence = SEEK_SET;
@@ -1043,9 +1040,8 @@ static int xfs_discard(BDRVRawState *s, int64_t offset, uint64_t bytes)
    fl.l_len = bytes;

    if (xfsctl(NULL, s->fd, XFS_IOC_UNRESVSP64, &fl) < 0) {
-        err = errno;
-        DPRINTF("cannot punch hole (%s)\n", strerror(errno));
-        return -err;
+        DEBUG_BLOCK_PRINT("cannot punch hole (%s)\n", strerror(errno));
+        return -errno;
    }

    return 0;
@@ -1244,7 +1240,7 @@ static int aio_worker(void *arg)
        break;
    }

-    g_free(aiocb);
+    g_slice_free(RawPosixAIOData, aiocb);
    return ret;
 }

@@ -1252,7 +1248,7 @@ static int paio_submit_co(BlockDriverState *bs, int fd,
        int64_t sector_num, QEMUIOVector *qiov, int nb_sectors,
        int type)
 {
-    RawPosixAIOData *acb = g_new(RawPosixAIOData, 1);
+    RawPosixAIOData *acb = g_slice_new(RawPosixAIOData);
    ThreadPool *pool;

    acb->bs = bs;
@@ -1277,7 +1273,7 @@ static BlockAIOCB *paio_submit(BlockDriverState *bs, int fd,
        int64_t sector_num, QEMUIOVector *qiov, int nb_sectors,
        BlockCompletionFunc *cb, void *opaque, int type)
 {
-    RawPosixAIOData *acb = g_new(RawPosixAIOData, 1);
+    RawPosixAIOData *acb = g_slice_new(RawPosixAIOData);
    ThreadPool *pool;

    acb->bs = bs;
@@ -1624,7 +1620,7 @@ static int raw_create(const char *filename, QemuOpts *opts, Error **errp)
    nocow = qemu_opt_get_bool(opts, BLOCK_OPT_NOCOW, false);
    buf = qemu_opt_get_del(opts, BLOCK_OPT_PREALLOC);
    prealloc = qapi_enum_parse(PreallocMode_lookup, buf,
-                               PREALLOC_MODE__MAX, PREALLOC_MODE_OFF,
+                               PREALLOC_MODE_MAX, PREALLOC_MODE_OFF,
                               &local_err);
    g_free(buf);
    if (local_err) {
@@ -1633,7 +1629,7 @@ static int raw_create(const char *filename, QemuOpts *opts, Error **errp)
        goto out;
    }

-    fd = qemu_open(filename, O_RDWR | O_CREAT | O_TRUNC | O_BINARY,
+    fd = qemu_open(filename, O_WRONLY | O_CREAT | O_TRUNC | O_BINARY,
                   0644);
    if (fd < 0) {
        result = -errno;
@@ -1818,8 +1814,7 @@ static int find_allocation(BlockDriverState *bs, off_t start,
 */
 static int64_t coroutine_fn raw_co_get_block_status(BlockDriverState *bs,
                                                    int64_t sector_num,
-                                                    int nb_sectors, int *pnum,
-                                                    BlockDriverState **file)
+                                                    int nb_sectors, int *pnum)
 {
    off_t start, data = 0, hole = 0;
    int64_t total_size;
@@ -1851,9 +1846,8 @@ static int64_t coroutine_fn raw_co_get_block_status(BlockDriverState *bs,
        *pnum = nb_sectors;
        ret = BDRV_BLOCK_DATA;
    } else if (data == start) {
-        /* On a data extent, compute sectors to the end of the extent,
-         * possibly including a partial sector at EOF. */
-        *pnum = MIN(nb_sectors, DIV_ROUND_UP(hole - start, BDRV_SECTOR_SIZE));
+        /* On a data extent, compute sectors to the end of the extent.  */
+        *pnum = MIN(nb_sectors, (hole - start) / BDRV_SECTOR_SIZE);
        ret = BDRV_BLOCK_DATA;
    } else {
        /* On a hole, compute sectors to the beginning of the next extent.  */
@@ -1861,7 +1855,6 @@ static int64_t coroutine_fn raw_co_get_block_status(BlockDriverState *bs,
        *pnum = MIN(nb_sectors, (data - start) / BDRV_SECTOR_SIZE);
        ret = BDRV_BLOCK_ZERO;
    }
-    *file = bs;
    return ret | BDRV_BLOCK_OFFSET_VALID | start;
 }

@@ -1966,8 +1959,8 @@ BlockDriver bdrv_file = {

 #if defined(__APPLE__) && defined(__MACH__)
 static kern_return_t FindEjectableCDMedia( io_iterator_t *mediaIterator );
-static kern_return_t GetBSDPath(io_iterator_t mediaIterator, char *bsdPath,
-                                CFIndex maxPathSize, int flags);
+static kern_return_t GetBSDPath( io_iterator_t mediaIterator, char *bsdPath, CFIndex maxPathSize );
+
 kern_return_t FindEjectableCDMedia( io_iterator_t *mediaIterator )
 {
    kern_return_t       kernResult;
@@ -1994,8 +1987,7 @@ kern_return_t FindEjectableCDMedia( io_iterator_t *mediaIterator )
    return kernResult;
 }

-kern_return_t GetBSDPath(io_iterator_t mediaIterator, char *bsdPath,
-                         CFIndex maxPathSize, int flags)
+kern_return_t GetBSDPath( io_iterator_t mediaIterator, char *bsdPath, CFIndex maxPathSize )
 {
    io_object_t     nextMedia;
    kern_return_t   kernResult = KERN_FAILURE;
@@ -2008,9 +2000,7 @@ kern_return_t GetBSDPath(io_iterator_t mediaIterator, char *bsdPath,
        if ( bsdPathAsCFString ) {
            size_t devPathLength;
            strcpy( bsdPath, _PATH_DEV );
-            if (flags & BDRV_O_NOCACHE) {
-                strcat(bsdPath, "r");
-            }
+            strcat( bsdPath, "r" );
            devPathLength = strlen( bsdPath );
            if ( CFStringGetCString( bsdPathAsCFString, bsdPath + devPathLength, maxPathSize - devPathLength, kCFStringEncodingASCII ) ) {
                kernResult = KERN_SUCCESS;
@@ -2083,38 +2073,15 @@ static void hdev_parse_filename(const char *filename, QDict *options,
    qdict_put_obj(options, "filename", QOBJECT(qstring_from_str(filename)));
 }

-static bool hdev_is_sg(BlockDriverState *bs)
-{
-
-#if defined(__linux__)
-
-    struct stat st;
-    struct sg_scsi_id scsiid;
-    int sg_version;
-
-    if (stat(bs->filename, &st) >= 0 && S_ISCHR(st.st_mode) &&
-        !bdrv_ioctl(bs, SG_GET_VERSION_NUM, &sg_version) &&
-        !bdrv_ioctl(bs, SG_GET_SCSI_ID, &scsiid)) {
-        DPRINTF("SG device found: type=%d, version=%d\n",
-            scsiid.scsi_type, sg_version);
-        return true;
-    }
-
-#endif
-
-    return false;
-}
-
 static int hdev_open(BlockDriverState *bs, QDict *options, int flags,
                     Error **errp)
 {
    BDRVRawState *s = bs->opaque;
    Error *local_err = NULL;
    int ret;
-
-#if defined(__APPLE__) && defined(__MACH__)
    const char *filename = qdict_get_str(options, "filename");

+#if defined(__APPLE__) && defined(__MACH__)
    if (strstart(filename, "/dev/cdrom", NULL)) {
        kern_return_t kernResult;
        io_iterator_t mediaIterator;
@@ -2122,8 +2089,8 @@ static int hdev_open(BlockDriverState *bs, QDict *options, int flags,
        int fd;

        kernResult = FindEjectableCDMedia( &mediaIterator );
-        kernResult = GetBSDPath(mediaIterator, bsdPath, sizeof(bsdPath),
-                                flags);
+        kernResult = GetBSDPath( mediaIterator, bsdPath, sizeof( bsdPath ) );
+
        if ( bsdPath[ 0 ] != '\0' ) {
            strcat(bsdPath,"s0");
            /* some CDs don't have a partition 0 */
@@ -2143,6 +2110,16 @@ static int hdev_open(BlockDriverState *bs, QDict *options, int flags,
 #endif

    s->type = FTYPE_FILE;
+#if defined(__linux__)
+    {
+        char resolved_path[ MAXPATHLEN ], *temp;
+
+        temp = realpath(filename, resolved_path);
+        if (temp && strstart(temp, "/dev/sg", NULL)) {
+            bs->sg = 1;
+        }
+    }
+#endif

    ret = raw_open_common(bs, options, flags, 0, &local_err);
    if (ret < 0) {
@@ -2152,9 +2129,6 @@ static int hdev_open(BlockDriverState *bs, QDict *options, int flags,
        return ret;
    }

-    /* Since this does ioctl the device must be already opened */
-    bs->sg = hdev_is_sg(bs);
-
    if (flags & BDRV_O_RDWR) {
        ret = check_hdev_writable(s);
        if (ret < 0) {
@@ -2168,6 +2142,61 @@ static int hdev_open(BlockDriverState *bs, QDict *options, int flags,
 }

 #if defined(__linux__)
+/* Note: we do not have a reliable method to detect if the floppy is
+   present. The current method is to try to open the floppy at every
+   I/O and to keep it opened during a few hundreds of ms. */
+static int fd_open(BlockDriverState *bs)
+{
+    BDRVRawState *s = bs->opaque;
+    int last_media_present;
+
+    if (s->type != FTYPE_FD)
+        return 0;
+    last_media_present = (s->fd >= 0);
+    if (s->fd >= 0 &&
+        (qemu_clock_get_ns(QEMU_CLOCK_REALTIME) - s->fd_open_time) >= FD_OPEN_TIMEOUT) {
+        qemu_close(s->fd);
+        s->fd = -1;
+#ifdef DEBUG_FLOPPY
+        printf("Floppy closed\n");
+#endif
+    }
+    if (s->fd < 0) {
+        if (s->fd_got_error &&
+            (qemu_clock_get_ns(QEMU_CLOCK_REALTIME) - s->fd_error_time) < FD_OPEN_TIMEOUT) {
+#ifdef DEBUG_FLOPPY
+            printf("No floppy (open delayed)\n");
+#endif
+            return -EIO;
+        }
+        s->fd = qemu_open(bs->filename, s->open_flags & ~O_NONBLOCK);
+        if (s->fd < 0) {
+            s->fd_error_time = qemu_clock_get_ns(QEMU_CLOCK_REALTIME);
+            s->fd_got_error = 1;
+            if (last_media_present)
+                s->fd_media_changed = 1;
+#ifdef DEBUG_FLOPPY
+            printf("No floppy\n");
+#endif
+            return -EIO;
+        }
+#ifdef DEBUG_FLOPPY
+        printf("Floppy opened\n");
+#endif
+    }
+    if (!last_media_present)
+        s->fd_media_changed = 1;
+    s->fd_open_time = qemu_clock_get_ns(QEMU_CLOCK_REALTIME);
+    s->fd_got_error = 0;
+    return 0;
+}
+
+static int hdev_ioctl(BlockDriverState *bs, unsigned long int req, void *buf)
+{
+    BDRVRawState *s = bs->opaque;
+
+    return ioctl(s->fd, req, buf);
+}

 static BlockAIOCB *hdev_aio_ioctl(BlockDriverState *bs,
        unsigned long int req, void *buf,
@@ -2180,7 +2209,7 @@ static BlockAIOCB *hdev_aio_ioctl(BlockDriverState *bs,
    if (fd_open(bs) < 0)
        return NULL;

-    acb = g_new(RawPosixAIOData, 1);
+    acb = g_slice_new(RawPosixAIOData);
    acb->bs = bs;
    acb->aio_type = QEMU_AIO_IOCTL;
    acb->aio_fildes = s->fd;
@@ -2190,8 +2219,8 @@ static BlockAIOCB *hdev_aio_ioctl(BlockDriverState *bs,
    pool = aio_get_thread_pool(bdrv_get_aio_context(bs));
    return thread_pool_submit_aio(pool, aio_worker, acb, cb, opaque);
 }
-#endif /* linux */

+#elif defined(__FreeBSD__) || defined(__FreeBSD_kernel__)
 static int fd_open(BlockDriverState *bs)
 {
    BDRVRawState *s = bs->opaque;
@@ -2201,6 +2230,14 @@ static int fd_open(BlockDriverState *bs)
        return 0;
    return -EIO;
 }
+#else /* !linux && !FreeBSD */
+
+static int fd_open(BlockDriverState *bs)
+{
+    return 0;
+}
+
+#endif /* !linux && !FreeBSD */

 static coroutine_fn BlockAIOCB *hdev_aio_discard(BlockDriverState *bs,
    int64_t sector_num, int nb_sectors,
@@ -2244,22 +2281,17 @@ static int hdev_create(const char *filename, QemuOpts *opts,
    int64_t total_size = 0;
    bool has_prefix;

-    /* This function is used by both protocol block drivers and therefore either
-     * of these prefixes may be given.
+    /* This function is used by all three protocol block drivers and therefore
+     * any of these three prefixes may be given.
     * The return value has to be stored somewhere, otherwise this is an error
     * due to -Werror=unused-value. */
    has_prefix =
        strstart(filename, "host_device:", &filename) ||
-        strstart(filename, "host_cdrom:" , &filename);
+        strstart(filename, "host_cdrom:" , &filename) ||
+        strstart(filename, "host_floppy:", &filename);

    (void)has_prefix;

-    ret = raw_normalize_devicepath(&filename);
-    if (ret < 0) {
-        error_setg_errno(errp, -ret, "Could not normalize device path");
-        return ret;
-    }
-
    /* Read out options */
    total_size = ROUND_UP(qemu_opt_get_size_del(opts, BLOCK_OPT_SIZE, 0),
                          BDRV_SECTOR_SIZE);
@@ -2325,10 +2357,161 @@ static BlockDriver bdrv_host_device = {

    /* generic scsi device */
 #ifdef __linux__
+    .bdrv_ioctl         = hdev_ioctl,
    .bdrv_aio_ioctl     = hdev_aio_ioctl,
 #endif
 };

+#ifdef __linux__
+static void floppy_parse_filename(const char *filename, QDict *options,
+                                  Error **errp)
+{
+    /* The prefix is optional, just as for "file". */
+    strstart(filename, "host_floppy:", &filename);
+
+    qdict_put_obj(options, "filename", QOBJECT(qstring_from_str(filename)));
+}
+
+static int floppy_open(BlockDriverState *bs, QDict *options, int flags,
+                       Error **errp)
+{
+    BDRVRawState *s = bs->opaque;
+    Error *local_err = NULL;
+    int ret;
+
+    s->type = FTYPE_FD;
+
+    /* open will not fail even if no floppy is inserted, so add O_NONBLOCK */
+    ret = raw_open_common(bs, options, flags, O_NONBLOCK, &local_err);
+    if (ret) {
+        if (local_err) {
+            error_propagate(errp, local_err);
+        }
+        return ret;
+    }
+
+    /* close fd so that we can reopen it as needed */
+    qemu_close(s->fd);
+    s->fd = -1;
+    s->fd_media_changed = 1;
+
+    error_report("Host floppy pass-through is deprecated");
+    error_printf("Support for it will be removed in a future release.\n");
+    return 0;
+}
+
+static int floppy_probe_device(const char *filename)
+{
+    int fd, ret;
+    int prio = 0;
+    struct floppy_struct fdparam;
+    struct stat st;
+
+    if (strstart(filename, "/dev/fd", NULL) &&
+        !strstart(filename, "/dev/fdset/", NULL)) {
+        prio = 50;
+    }
+
+    fd = qemu_open(filename, O_RDONLY | O_NONBLOCK);
+    if (fd < 0) {
+        goto out;
+    }
+    ret = fstat(fd, &st);
+    if (ret == -1 || !S_ISBLK(st.st_mode)) {
+        goto outc;
+    }
+
+    /* Attempt to detect via a floppy specific ioctl */
+    ret = ioctl(fd, FDGETPRM, &fdparam);
+    if (ret >= 0)
+        prio = 100;
+
+outc:
+    qemu_close(fd);
+out:
+    return prio;
+}
+
+
+static int floppy_is_inserted(BlockDriverState *bs)
+{
+    return fd_open(bs) >= 0;
+}
+
+static int floppy_media_changed(BlockDriverState *bs)
+{
+    BDRVRawState *s = bs->opaque;
+    int ret;
+
+    /*
+     * XXX: we do not have a true media changed indication.
+     * It does not work if the floppy is changed without trying to read it.
+     */
+    fd_open(bs);
+    ret = s->fd_media_changed;
+    s->fd_media_changed = 0;
+#ifdef DEBUG_FLOPPY
+    printf("Floppy changed=%d\n", ret);
+#endif
+    return ret;
+}
+
+static void floppy_eject(BlockDriverState *bs, bool eject_flag)
+{
+    BDRVRawState *s = bs->opaque;
+    int fd;
+
+    if (s->fd >= 0) {
+        qemu_close(s->fd);
+        s->fd = -1;
+    }
+    fd = qemu_open(bs->filename, s->open_flags | O_NONBLOCK);
+    if (fd >= 0) {
+        if (ioctl(fd, FDEJECT, 0) < 0)
+            perror("FDEJECT");
+        qemu_close(fd);
+    }
+}
+
+static BlockDriver bdrv_host_floppy = {
+    .format_name        = "host_floppy",
+    .protocol_name      = "host_floppy",
+    .instance_size      = sizeof(BDRVRawState),
+    .bdrv_needs_filename = true,
+    .bdrv_probe_device	= floppy_probe_device,
+    .bdrv_parse_filename = floppy_parse_filename,
+    .bdrv_file_open     = floppy_open,
+    .bdrv_close         = raw_close,
+    .bdrv_reopen_prepare = raw_reopen_prepare,
+    .bdrv_reopen_commit  = raw_reopen_commit,
+    .bdrv_reopen_abort   = raw_reopen_abort,
+    .bdrv_create         = hdev_create,
+    .create_opts         = &raw_create_opts,
+
+    .bdrv_aio_readv     = raw_aio_readv,
+    .bdrv_aio_writev    = raw_aio_writev,
+    .bdrv_aio_flush	= raw_aio_flush,
+    .bdrv_refresh_limits = raw_refresh_limits,
+    .bdrv_io_plug = raw_aio_plug,
+    .bdrv_io_unplug = raw_aio_unplug,
+    .bdrv_flush_io_queue = raw_aio_flush_io_queue,
+
+    .bdrv_truncate      = raw_truncate,
+    .bdrv_getlength      = raw_getlength,
+    .has_variable_length = true,
+    .bdrv_get_allocated_file_size
+                        = raw_get_allocated_file_size,
+
+    .bdrv_detach_aio_context = raw_detach_aio_context,
+    .bdrv_attach_aio_context = raw_attach_aio_context,
+
+    /* removable device support */
+    .bdrv_is_inserted   = floppy_is_inserted,
+    .bdrv_media_changed = floppy_media_changed,
+    .bdrv_eject         = floppy_eject,
+};
+#endif
+
 #if defined(__linux__) || defined(__FreeBSD__) || defined(__FreeBSD_kernel__)
 static void cdrom_parse_filename(const char *filename, QDict *options,
                                 Error **errp)
@@ -2384,13 +2567,15 @@ out:
    return prio;
 }

-static bool cdrom_is_inserted(BlockDriverState *bs)
+static int cdrom_is_inserted(BlockDriverState *bs)
 {
    BDRVRawState *s = bs->opaque;
    int ret;

    ret = ioctl(s->fd, CDROM_DRIVE_STATUS, CDSL_CURRENT);
-    return ret == CDS_DISC_OK;
+    if (ret == CDS_DISC_OK)
+        return 1;
+    return 0;
 }

 static void cdrom_eject(BlockDriverState *bs, bool eject_flag)
@@ -2457,6 +2642,7 @@ static BlockDriver bdrv_host_cdrom = {
    .bdrv_lock_medium   = cdrom_lock_medium,

    /* generic scsi device */
+    .bdrv_ioctl         = hdev_ioctl,
    .bdrv_aio_ioctl     = hdev_aio_ioctl,
 };
 #endif /* __linux__ */
@@ -2515,7 +2701,7 @@ static int cdrom_reopen(BlockDriverState *bs)
    return 0;
 }

-static bool cdrom_is_inserted(BlockDriverState *bs)
+static int cdrom_is_inserted(BlockDriverState *bs)
 {
    return raw_getlength(bs) > 0;
 }
@@ -2603,6 +2789,7 @@ static void bdrv_file_init(void)
    bdrv_register(&bdrv_file);
    bdrv_register(&bdrv_host_device);
 #ifdef __linux__
+    bdrv_register(&bdrv_host_floppy);
    bdrv_register(&bdrv_host_cdrom);
 #endif
 #if defined(__FreeBSD__) || defined(__FreeBSD_kernel__)
--- a/block/raw-win32.c
+++ b/block/raw-win32.c
@@ -21,7 +21,6 @@
 * 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 "qemu/timer.h"
 #include "block/block_int.h"
@@ -30,7 +29,6 @@
 #include "trace.h"
 #include "block/thread-pool.h"
 #include "qemu/iov.h"
-#include "qapi/qmp/qstring.h"
 #include <windows.h>
 #include <winioctl.h>

@@ -120,9 +118,9 @@ static int aio_worker(void *arg)
    case QEMU_AIO_WRITE:
        count = handle_aiocb_rw(aiocb);
        if (count == aiocb->aio_nbytes) {
-            ret = 0;
+            count = 0;
        } else {
-            ret = -EINVAL;
+            count = -EINVAL;
        }
        break;
    case QEMU_AIO_FLUSH:
@@ -136,7 +134,7 @@ static int aio_worker(void *arg)
        break;
    }

-    g_free(aiocb);
+    g_slice_free(RawWin32AIOData, aiocb);
    return ret;
 }

@@ -144,7 +142,7 @@ static BlockAIOCB *paio_submit(BlockDriverState *bs, HANDLE hfile,
        int64_t sector_num, QEMUIOVector *qiov, int nb_sectors,
        BlockCompletionFunc *cb, void *opaque, int type)
 {
-    RawWin32AIOData *acb = g_new(RawWin32AIOData, 1);
+    RawWin32AIOData *acb = g_slice_new(RawWin32AIOData);
    ThreadPool *pool;

    acb->bs = bs;
--- a/block/raw_bsd.c
+++ b/block/raw_bsd.c
@@ -26,7 +26,6 @@
 * IN THE SOFTWARE.
 */

-#include "qemu/osdep.h"
 #include "block/block_int.h"
 #include "qemu/option.h"

@@ -53,7 +52,7 @@ static int coroutine_fn raw_co_readv(BlockDriverState *bs, int64_t sector_num,
                                     int nb_sectors, QEMUIOVector *qiov)
 {
    BLKDBG_EVENT(bs->file, BLKDBG_READ_AIO);
-    return bdrv_co_readv(bs->file->bs, sector_num, nb_sectors, qiov);
+    return bdrv_co_readv(bs->file, sector_num, nb_sectors, qiov);
 }

 static int coroutine_fn raw_co_writev(BlockDriverState *bs, int64_t sector_num,
@@ -76,7 +75,7 @@ static int coroutine_fn raw_co_writev(BlockDriverState *bs, int64_t sector_num,
            return 0;
        }

-        buf = qemu_try_blockalign(bs->file->bs, 512);
+        buf = qemu_try_blockalign(bs->file, 512);
        if (!buf) {
            ret = -ENOMEM;
            goto fail;
@@ -103,7 +102,7 @@ static int coroutine_fn raw_co_writev(BlockDriverState *bs, int64_t sector_num,
    }

    BLKDBG_EVENT(bs->file, BLKDBG_WRITE_AIO);
-    ret = bdrv_co_writev(bs->file->bs, sector_num, nb_sectors, qiov);
+    ret = bdrv_co_writev(bs->file, sector_num, nb_sectors, qiov);

 fail:
    if (qiov == &local_qiov) {
@@ -115,11 +114,9 @@ fail:

 static int64_t coroutine_fn raw_co_get_block_status(BlockDriverState *bs,
                                            int64_t sector_num,
-                                            int nb_sectors, int *pnum,
-                                            BlockDriverState **file)
+                                            int nb_sectors, int *pnum)
 {
    *pnum = nb_sectors;
-    *file = bs->file->bs;
    return BDRV_BLOCK_RAW | BDRV_BLOCK_OFFSET_VALID | BDRV_BLOCK_DATA |
           (sector_num << BDRV_SECTOR_BITS);
 }
@@ -128,48 +125,58 @@ static int coroutine_fn raw_co_write_zeroes(BlockDriverState *bs,
                                            int64_t sector_num, int nb_sectors,
                                            BdrvRequestFlags flags)
 {
-    return bdrv_co_write_zeroes(bs->file->bs, sector_num, nb_sectors, flags);
+    return bdrv_co_write_zeroes(bs->file, sector_num, nb_sectors, flags);
 }

 static int coroutine_fn raw_co_discard(BlockDriverState *bs,
                                       int64_t sector_num, int nb_sectors)
 {
-    return bdrv_co_discard(bs->file->bs, sector_num, nb_sectors);
+    return bdrv_co_discard(bs->file, sector_num, nb_sectors);
 }

 static int64_t raw_getlength(BlockDriverState *bs)
 {
-    return bdrv_getlength(bs->file->bs);
+    return bdrv_getlength(bs->file);
 }

 static int raw_get_info(BlockDriverState *bs, BlockDriverInfo *bdi)
 {
-    return bdrv_get_info(bs->file->bs, bdi);
+    return bdrv_get_info(bs->file, bdi);
 }

 static void raw_refresh_limits(BlockDriverState *bs, Error **errp)
 {
-    bs->bl = bs->file->bs->bl;
+    bs->bl = bs->file->bl;
 }

 static int raw_truncate(BlockDriverState *bs, int64_t offset)
 {
-    return bdrv_truncate(bs->file->bs, offset);
+    return bdrv_truncate(bs->file, offset);
+}
+
+static int raw_is_inserted(BlockDriverState *bs)
+{
+    return bdrv_is_inserted(bs->file);
 }

 static int raw_media_changed(BlockDriverState *bs)
 {
-    return bdrv_media_changed(bs->file->bs);
+    return bdrv_media_changed(bs->file);
 }

 static void raw_eject(BlockDriverState *bs, bool eject_flag)
 {
-    bdrv_eject(bs->file->bs, eject_flag);
+    bdrv_eject(bs->file, eject_flag);
 }

 static void raw_lock_medium(BlockDriverState *bs, bool locked)
 {
-    bdrv_lock_medium(bs->file->bs, locked);
+    bdrv_lock_medium(bs->file, locked);
+}
+
+static int raw_ioctl(BlockDriverState *bs, unsigned long int req, void *buf)
+{
+    return bdrv_ioctl(bs->file, req, buf);
 }

 static BlockAIOCB *raw_aio_ioctl(BlockDriverState *bs,
@@ -177,12 +184,12 @@ static BlockAIOCB *raw_aio_ioctl(BlockDriverState *bs,
                                 BlockCompletionFunc *cb,
                                 void *opaque)
 {
-    return bdrv_aio_ioctl(bs->file->bs, req, buf, cb, opaque);
+    return bdrv_aio_ioctl(bs->file, req, buf, cb, opaque);
 }

 static int raw_has_zero_init(BlockDriverState *bs)
 {
-    return bdrv_has_zero_init(bs->file->bs);
+    return bdrv_has_zero_init(bs->file);
 }

 static int raw_create(const char *filename, QemuOpts *opts, Error **errp)
@@ -200,7 +207,7 @@ static int raw_create(const char *filename, QemuOpts *opts, Error **errp)
 static int raw_open(BlockDriverState *bs, QDict *options, int flags,
                    Error **errp)
 {
-    bs->sg = bs->file->bs->sg;
+    bs->sg = bs->file->sg;

    if (bs->probed && !bdrv_is_read_only(bs)) {
        fprintf(stderr,
@@ -210,7 +217,7 @@ static int raw_open(BlockDriverState *bs, QDict *options, int flags,
                "raw images, write operations on block 0 will be restricted.\n"
                "         Specify the 'raw' format explicitly to remove the "
                "restrictions.\n",
-                bs->file->bs->filename);
+                bs->file->filename);
    }

    return 0;
@@ -230,12 +237,12 @@ static int raw_probe(const uint8_t *buf, int buf_size, const char *filename)

 static int raw_probe_blocksizes(BlockDriverState *bs, BlockSizes *bsz)
 {
-    return bdrv_probe_blocksizes(bs->file->bs, bsz);
+    return bdrv_probe_blocksizes(bs->file, bsz);
 }

 static int raw_probe_geometry(BlockDriverState *bs, HDGeometry *geo)
 {
-    return bdrv_probe_geometry(bs->file->bs, geo);
+    return bdrv_probe_geometry(bs->file, geo);
 }

 BlockDriver bdrv_raw = {
@@ -257,9 +264,11 @@ BlockDriver bdrv_raw = {
    .bdrv_refresh_limits  = &raw_refresh_limits,
    .bdrv_probe_blocksizes = &raw_probe_blocksizes,
    .bdrv_probe_geometry  = &raw_probe_geometry,
+    .bdrv_is_inserted     = &raw_is_inserted,
    .bdrv_media_changed   = &raw_media_changed,
    .bdrv_eject           = &raw_eject,
    .bdrv_lock_medium     = &raw_lock_medium,
+    .bdrv_ioctl           = &raw_ioctl,
    .bdrv_aio_ioctl       = &raw_aio_ioctl,
    .create_opts          = &raw_create_opts,
    .bdrv_has_zero_init   = &raw_has_zero_init
--- a/block/rbd.c
+++ b/block/rbd.c
@@ -11,7 +11,7 @@
 * GNU GPL, version 2 or (at your option) any later version.
 */

-#include "qemu/osdep.h"
+#include <inttypes.h>

 #include "qemu-common.h"
 #include "qemu/error-report.h"
@@ -74,18 +74,25 @@ typedef struct RBDAIOCB {
    QEMUIOVector *qiov;
    char *bounce;
    RBDAIOCmd cmd;
+    int64_t sector_num;
    int error;
    struct BDRVRBDState *s;
+    int status;
 } RBDAIOCB;

 typedef struct RADOSCB {
+    int rcbid;
    RBDAIOCB *acb;
    struct BDRVRBDState *s;
+    int done;
    int64_t size;
    char *buf;
    int64_t ret;
 } RADOSCB;

+#define RBD_FD_READ 0
+#define RBD_FD_WRITE 1
+
 typedef struct BDRVRBDState {
    rados_t cluster;
    rados_ioctx_t io_ctx;
@@ -228,9 +235,7 @@ static char *qemu_rbd_parse_clientname(const char *conf, char *clientname)
    return NULL;
 }

-static int qemu_rbd_set_conf(rados_t cluster, const char *conf,
-                             bool only_read_conf_file,
-                             Error **errp)
+static int qemu_rbd_set_conf(rados_t cluster, const char *conf, Error **errp)
 {
    char *p, *buf;
    char name[RBD_MAX_CONF_NAME_SIZE];
@@ -262,18 +267,14 @@ static int qemu_rbd_set_conf(rados_t cluster, const char *conf,
        qemu_rbd_unescape(value);

        if (strcmp(name, "conf") == 0) {
-            /* read the conf file alone, so it doesn't override more
-               specific settings for a particular device */
-            if (only_read_conf_file) {
-                ret = rados_conf_read_file(cluster, value);
-                if (ret < 0) {
-                    error_setg(errp, "error reading conf file %s", value);
-                    break;
-                }
+            ret = rados_conf_read_file(cluster, value);
+            if (ret < 0) {
+                error_setg(errp, "error reading conf file %s", value);
+                break;
            }
        } else if (strcmp(name, "id") == 0) {
            /* ignore, this is parsed by qemu_rbd_parse_clientname() */
-        } else if (!only_read_conf_file) {
+        } else {
            ret = rados_conf_set(cluster, name, value);
            if (ret < 0) {
                error_setg(errp, "invalid conf option %s", name);
@@ -324,7 +325,7 @@ static int qemu_rbd_create(const char *filename, QemuOpts *opts, Error **errp)
            error_setg(errp, "obj size too small");
            return -EINVAL;
        }
-        obj_order = ctz32(objsize);
+        obj_order = ffs(objsize) - 1;
    }

    clientname = qemu_rbd_parse_clientname(conf, clientname_buf);
@@ -336,15 +337,10 @@ static int qemu_rbd_create(const char *filename, QemuOpts *opts, Error **errp)
    if (strstr(conf, "conf=") == NULL) {
        /* try default location, but ignore failure */
        rados_conf_read_file(cluster, NULL);
-    } else if (conf[0] != '\0' &&
-               qemu_rbd_set_conf(cluster, conf, true, &local_err) < 0) {
-        rados_shutdown(cluster);
-        error_propagate(errp, local_err);
-        return -EIO;
    }

    if (conf[0] != '\0' &&
-        qemu_rbd_set_conf(cluster, conf, false, &local_err) < 0) {
+        qemu_rbd_set_conf(cluster, conf, &local_err) < 0) {
        rados_shutdown(cluster);
        error_propagate(errp, local_err);
        return -EIO;
@@ -409,6 +405,7 @@ static void qemu_rbd_complete_aio(RADOSCB *rcb)
    }
    qemu_vfree(acb->bounce);
    acb->common.cb(acb->common.opaque, (acb->ret > 0 ? 0 : acb->ret));
+    acb->status = 0;

    qemu_aio_unref(acb);
 }
@@ -471,23 +468,6 @@ static int qemu_rbd_open(BlockDriverState *bs, QDict *options, int flags,
        s->snap = g_strdup(snap_buf);
    }

-    if (strstr(conf, "conf=") == NULL) {
-        /* try default location, but ignore failure */
-        rados_conf_read_file(s->cluster, NULL);
-    } else if (conf[0] != '\0') {
-        r = qemu_rbd_set_conf(s->cluster, conf, true, errp);
-        if (r < 0) {
-            goto failed_shutdown;
-        }
-    }
-
-    if (conf[0] != '\0') {
-        r = qemu_rbd_set_conf(s->cluster, conf, false, errp);
-        if (r < 0) {
-            goto failed_shutdown;
-        }
-    }
-
    /*
     * Fallback to more conservative semantics if setting cache
     * options fails. Ignore errors from setting rbd_cache because the
@@ -501,6 +481,18 @@ static int qemu_rbd_open(BlockDriverState *bs, QDict *options, int flags,
        rados_conf_set(s->cluster, "rbd_cache", "true");
    }

+    if (strstr(conf, "conf=") == NULL) {
+        /* try default location, but ignore failure */
+        rados_conf_read_file(s->cluster, NULL);
+    }
+
+    if (conf[0] != '\0') {
+        r = qemu_rbd_set_conf(s->cluster, conf, errp);
+        if (r < 0) {
+            goto failed_shutdown;
+        }
+    }
+
    r = rados_connect(s->cluster);
    if (r < 0) {
        error_setg(errp, "error connecting");
@@ -629,6 +621,7 @@ static BlockAIOCB *rbd_start_aio(BlockDriverState *bs,
    acb->error = 0;
    acb->s = s;
    acb->bh = NULL;
+    acb->status = -EINPROGRESS;

    if (cmd == RBD_AIO_WRITE) {
        qemu_iovec_to_buf(acb->qiov, 0, acb->bounce, qiov->size);
@@ -640,6 +633,7 @@ static BlockAIOCB *rbd_start_aio(BlockDriverState *bs,
    size = nb_sectors * BDRV_SECTOR_SIZE;

    rcb = g_new(RADOSCB, 1);
+    rcb->done = 0;
    rcb->acb = acb;
    rcb->buf = buf;
    rcb->s = acb->s;
--- a/block/sheepdog.c
+++ b/block/sheepdog.c
@@ -12,7 +12,6 @@
 * GNU GPL, version 2 or (at your option) any later version.
 */

-#include "qemu/osdep.h"
 #include "qemu-common.h"
 #include "qemu/uri.h"
 #include "qemu/error-report.h"
@@ -29,6 +28,7 @@
 #define SD_OP_READ_OBJ       0x02
 #define SD_OP_WRITE_OBJ      0x03
 /* 0x04 is used internally by Sheepdog */
+#define SD_OP_DISCARD_OBJ    0x05

 #define SD_OP_NEW_VDI        0x11
 #define SD_OP_LOCK_VDI       0x12
@@ -318,10 +318,6 @@ enum AIOCBState {
    AIOCB_DISCARD_OBJ,
 };

-#define AIOCBOverlapping(x, y)                                 \
-    (!(x->max_affect_data_idx < y->min_affect_data_idx          \
-       || y->max_affect_data_idx < x->min_affect_data_idx))
-
 struct SheepdogAIOCB {
    BlockAIOCB common;

@@ -338,20 +334,6 @@ struct SheepdogAIOCB {

    bool cancelable;
    int nr_pending;
-
-    uint32_t min_affect_data_idx;
-    uint32_t max_affect_data_idx;
-
-    /*
-     * The difference between affect_data_idx and dirty_data_idx:
-     * affect_data_idx represents range of index of all request types.
-     * dirty_data_idx represents range of index updated by COW requests.
-     * dirty_data_idx is used for updating an inode object.
-     */
-    uint32_t min_dirty_data_idx;
-    uint32_t max_dirty_data_idx;
-
-    QLIST_ENTRY(SheepdogAIOCB) aiocb_siblings;
 };

 typedef struct BDRVSheepdogState {
@@ -360,6 +342,9 @@ typedef struct BDRVSheepdogState {

    SheepdogInode inode;

+    uint32_t min_dirty_data_idx;
+    uint32_t max_dirty_data_idx;
+
    char name[SD_MAX_VDI_LEN];
    bool is_snapshot;
    uint32_t cache_flags;
@@ -377,17 +362,10 @@ typedef struct BDRVSheepdogState {

    /* Every aio request must be linked to either of these queues. */
    QLIST_HEAD(inflight_aio_head, AIOReq) inflight_aio_head;
+    QLIST_HEAD(pending_aio_head, AIOReq) pending_aio_head;
    QLIST_HEAD(failed_aio_head, AIOReq) failed_aio_head;
-
-    CoQueue overlapping_queue;
-    QLIST_HEAD(inflight_aiocb_head, SheepdogAIOCB) inflight_aiocb_head;
 } BDRVSheepdogState;

-typedef struct BDRVSheepdogReopenState {
-    int fd;
-    int cache_flags;
-} BDRVSheepdogReopenState;
-
 static const char * sd_strerror(int err)
 {
    int i;
@@ -520,7 +498,13 @@ static void sd_aio_cancel(BlockAIOCB *blockacb)
    AIOReq *aioreq, *next;

    if (sd_acb_cancelable(acb)) {
-        /* Remove outstanding requests from failed queue.  */
+        /* Remove outstanding requests from pending and failed queues.  */
+        QLIST_FOREACH_SAFE(aioreq, &s->pending_aio_head, aio_siblings,
+                           next) {
+            if (aioreq->aiocb == acb) {
+                free_aio_req(s, aioreq);
+            }
+        }
        QLIST_FOREACH_SAFE(aioreq, &s->failed_aio_head, aio_siblings,
                           next) {
            if (aioreq->aiocb == acb) {
@@ -545,10 +529,6 @@ static SheepdogAIOCB *sd_aio_setup(BlockDriverState *bs, QEMUIOVector *qiov,
                                   int64_t sector_num, int nb_sectors)
 {
    SheepdogAIOCB *acb;
-    uint32_t object_size;
-    BDRVSheepdogState *s = bs->opaque;
-
-    object_size = (UINT32_C(1) << s->inode.block_size_shift);

    acb = qemu_aio_get(&sd_aiocb_info, bs, NULL, NULL);

@@ -562,14 +542,6 @@ static SheepdogAIOCB *sd_aio_setup(BlockDriverState *bs, QEMUIOVector *qiov,
    acb->coroutine = qemu_coroutine_self();
    acb->ret = 0;
    acb->nr_pending = 0;
-
-    acb->min_affect_data_idx = acb->sector_num * BDRV_SECTOR_SIZE / object_size;
-    acb->max_affect_data_idx = (acb->sector_num * BDRV_SECTOR_SIZE +
-                              acb->nb_sectors * BDRV_SECTOR_SIZE) / object_size;
-
-    acb->min_dirty_data_idx = UINT32_MAX;
-    acb->max_dirty_data_idx = 0;
-
    return acb;
 }

@@ -652,16 +624,14 @@ static coroutine_fn void do_co_req(void *opaque)
    unsigned int *rlen = srco->rlen;

    co = qemu_coroutine_self();
-    aio_set_fd_handler(srco->aio_context, sockfd, false,
-                       NULL, restart_co_req, co);
+    aio_set_fd_handler(srco->aio_context, sockfd, NULL, restart_co_req, co);

    ret = send_co_req(sockfd, hdr, data, wlen);
    if (ret < 0) {
        goto out;
    }

-    aio_set_fd_handler(srco->aio_context, sockfd, false,
-                       restart_co_req, NULL, co);
+    aio_set_fd_handler(srco->aio_context, sockfd, restart_co_req, NULL, co);

    ret = qemu_co_recv(sockfd, hdr, sizeof(*hdr));
    if (ret != sizeof(*hdr)) {
@@ -686,8 +656,7 @@ static coroutine_fn void do_co_req(void *opaque)
 out:
    /* there is at most one request for this sockfd, so it is safe to
     * set each handler to NULL. */
-    aio_set_fd_handler(srco->aio_context, sockfd, false,
-                       NULL, NULL, NULL);
+    aio_set_fd_handler(srco->aio_context, sockfd, NULL, NULL, NULL);

    srco->ret = ret;
    srco->finished = true;
@@ -734,13 +703,44 @@ static int reload_inode(BDRVSheepdogState *s, uint32_t snapid, const char *tag);
 static int get_sheep_fd(BDRVSheepdogState *s, Error **errp);
 static void co_write_request(void *opaque);

+static AIOReq *find_pending_req(BDRVSheepdogState *s, uint64_t oid)
+{
+    AIOReq *aio_req;
+
+    QLIST_FOREACH(aio_req, &s->pending_aio_head, aio_siblings) {
+        if (aio_req->oid == oid) {
+            return aio_req;
+        }
+    }
+
+    return NULL;
+}
+
+/*
+ * This function searchs pending requests to the object `oid', and
+ * sends them.
+ */
+static void coroutine_fn send_pending_req(BDRVSheepdogState *s, uint64_t oid)
+{
+    AIOReq *aio_req;
+    SheepdogAIOCB *acb;
+
+    while ((aio_req = find_pending_req(s, oid)) != NULL) {
+        acb = aio_req->aiocb;
+        /* move aio_req from pending list to inflight one */
+        QLIST_REMOVE(aio_req, aio_siblings);
+        QLIST_INSERT_HEAD(&s->inflight_aio_head, aio_req, aio_siblings);
+        add_aio_request(s, aio_req, acb->qiov->iov, acb->qiov->niov,
+                        acb->aiocb_type);
+    }
+}
+
 static coroutine_fn void reconnect_to_sdog(void *opaque)
 {
    BDRVSheepdogState *s = opaque;
    AIOReq *aio_req, *next;

-    aio_set_fd_handler(s->aio_context, s->fd, false, NULL,
-                       NULL, NULL);
+    aio_set_fd_handler(s->aio_context, s->fd, NULL, NULL, NULL);
    close(s->fd);
    s->fd = -1;

@@ -837,9 +837,15 @@ static void coroutine_fn aio_read_response(void *opaque)
             */
            if (rsp.result == SD_RES_SUCCESS) {
                s->inode.data_vdi_id[idx] = s->inode.vdi_id;
-                acb->max_dirty_data_idx = MAX(idx, acb->max_dirty_data_idx);
-                acb->min_dirty_data_idx = MIN(idx, acb->min_dirty_data_idx);
+                s->max_dirty_data_idx = MAX(idx, s->max_dirty_data_idx);
+                s->min_dirty_data_idx = MIN(idx, s->min_dirty_data_idx);
            }
+            /*
+             * Some requests may be blocked because simultaneous
+             * create requests are not allowed, so we search the
+             * pending requests here.
+             */
+            send_pending_req(s, aio_req->oid);
        }
        break;
    case AIOCB_READ_UDATA:
@@ -865,6 +871,10 @@ static void coroutine_fn aio_read_response(void *opaque)
            rsp.result = SD_RES_SUCCESS;
            s->discard_supported = false;
            break;
+        case SD_RES_SUCCESS:
+            idx = data_oid_to_idx(aio_req->oid);
+            s->inode.data_vdi_id[idx] = 0;
+            break;
        default:
            break;
        }
@@ -943,8 +953,7 @@ static int get_sheep_fd(BDRVSheepdogState *s, Error **errp)
        return fd;
    }

-    aio_set_fd_handler(s->aio_context, fd, false,
-                       co_read_response, NULL, s);
+    aio_set_fd_handler(s->aio_context, fd, co_read_response, NULL, s);
    return fd;
 }

@@ -1180,13 +1189,7 @@ static void coroutine_fn add_aio_request(BDRVSheepdogState *s, AIOReq *aio_req,
        hdr.flags = SD_FLAG_CMD_WRITE | flags;
        break;
    case AIOCB_DISCARD_OBJ:
-        hdr.opcode = SD_OP_WRITE_OBJ;
-        hdr.flags = SD_FLAG_CMD_WRITE | flags;
-        s->inode.data_vdi_id[data_oid_to_idx(oid)] = 0;
-        offset = offsetof(SheepdogInode,
-                          data_vdi_id[data_oid_to_idx(oid)]);
-        oid = vid_to_vdi_oid(s->inode.vdi_id);
-        wlen = datalen = sizeof(uint32_t);
+        hdr.opcode = SD_OP_DISCARD_OBJ;
        break;
    }

@@ -1205,7 +1208,7 @@ static void coroutine_fn add_aio_request(BDRVSheepdogState *s, AIOReq *aio_req,

    qemu_co_mutex_lock(&s->lock);
    s->co_send = qemu_coroutine_self();
-    aio_set_fd_handler(s->aio_context, s->fd, false,
+    aio_set_fd_handler(s->aio_context, s->fd,
                       co_read_response, co_write_request, s);
    socket_set_cork(s->fd, 1);

@@ -1224,8 +1227,7 @@ static void coroutine_fn add_aio_request(BDRVSheepdogState *s, AIOReq *aio_req,
    }
 out:
    socket_set_cork(s->fd, 0);
-    aio_set_fd_handler(s->aio_context, s->fd, false,
-                       co_read_response, NULL, s);
+    aio_set_fd_handler(s->aio_context, s->fd, co_read_response, NULL, s);
    s->co_send = NULL;
    qemu_co_mutex_unlock(&s->lock);
 }
@@ -1339,6 +1341,30 @@ out:
    return ret;
 }

+/* Return true if the specified request is linked to the pending list. */
+static bool check_simultaneous_create(BDRVSheepdogState *s, AIOReq *aio_req)
+{
+    AIOReq *areq;
+    QLIST_FOREACH(areq, &s->inflight_aio_head, aio_siblings) {
+        if (areq != aio_req && areq->oid == aio_req->oid) {
+            /*
+             * Sheepdog cannot handle simultaneous create requests to the same
+             * object, so we cannot send the request until the previous request
+             * finishes.
+             */
+            DPRINTF("simultaneous create to %" PRIx64 "\n", aio_req->oid);
+            aio_req->flags = 0;
+            aio_req->base_oid = 0;
+            aio_req->create = false;
+            QLIST_REMOVE(aio_req, aio_siblings);
+            QLIST_INSERT_HEAD(&s->pending_aio_head, aio_req, aio_siblings);
+            return true;
+        }
+    }
+
+    return false;
+}
+
 static void coroutine_fn resend_aioreq(BDRVSheepdogState *s, AIOReq *aio_req)
 {
    SheepdogAIOCB *acb = aio_req->aiocb;
@@ -1353,6 +1379,10 @@ static void coroutine_fn resend_aioreq(BDRVSheepdogState *s, AIOReq *aio_req)
            goto out;
        }

+        if (check_simultaneous_create(s, aio_req)) {
+            return;
+        }
+
        if (s->inode.data_vdi_id[idx]) {
            aio_req->base_oid = vid_to_data_oid(s->inode.data_vdi_id[idx], idx);
            aio_req->flags |= SD_FLAG_CMD_COW;
@@ -1375,8 +1405,7 @@ static void sd_detach_aio_context(BlockDriverState *bs)
 {
    BDRVSheepdogState *s = bs->opaque;

-    aio_set_fd_handler(s->aio_context, s->fd, false, NULL,
-                       NULL, NULL);
+    aio_set_fd_handler(s->aio_context, s->fd, NULL, NULL, NULL);
 }

 static void sd_attach_aio_context(BlockDriverState *bs,
@@ -1385,8 +1414,7 @@ static void sd_attach_aio_context(BlockDriverState *bs,
    BDRVSheepdogState *s = bs->opaque;

    s->aio_context = new_context;
-    aio_set_fd_handler(new_context, s->fd, false,
-                       co_read_response, NULL, s);
+    aio_set_fd_handler(new_context, s->fd, co_read_response, NULL, s);
 }

 /* TODO Convert to fine grained options */
@@ -1430,8 +1458,8 @@ static int sd_open(BlockDriverState *bs, QDict *options, int flags,
    filename = qemu_opt_get(opts, "filename");

    QLIST_INIT(&s->inflight_aio_head);
+    QLIST_INIT(&s->pending_aio_head);
    QLIST_INIT(&s->failed_aio_head);
-    QLIST_INIT(&s->inflight_aiocb_head);
    s->fd = -1;

    memset(vdi, 0, sizeof(vdi));
@@ -1490,17 +1518,17 @@ static int sd_open(BlockDriverState *bs, QDict *options, int flags,
    }

    memcpy(&s->inode, buf, sizeof(s->inode));
+    s->min_dirty_data_idx = UINT32_MAX;
+    s->max_dirty_data_idx = 0;

    bs->total_sectors = s->inode.vdi_size / BDRV_SECTOR_SIZE;
    pstrcpy(s->name, sizeof(s->name), vdi);
    qemu_co_mutex_init(&s->lock);
-    qemu_co_queue_init(&s->overlapping_queue);
    qemu_opts_del(opts);
    g_free(buf);
    return 0;
 out:
-    aio_set_fd_handler(bdrv_get_aio_context(bs), s->fd,
-                       false, NULL, NULL, NULL);
+    aio_set_fd_handler(bdrv_get_aio_context(bs), s->fd, NULL, NULL, NULL);
    if (s->fd >= 0) {
        closesocket(s->fd);
    }
@@ -1509,70 +1537,6 @@ out:
    return ret;
 }

-static int sd_reopen_prepare(BDRVReopenState *state, BlockReopenQueue *queue,
-                             Error **errp)
-{
-    BDRVSheepdogState *s = state->bs->opaque;
-    BDRVSheepdogReopenState *re_s;
-    int ret = 0;
-
-    re_s = state->opaque = g_new0(BDRVSheepdogReopenState, 1);
-
-    re_s->cache_flags = SD_FLAG_CMD_CACHE;
-    if (state->flags & BDRV_O_NOCACHE) {
-        re_s->cache_flags = SD_FLAG_CMD_DIRECT;
-    }
-
-    re_s->fd = get_sheep_fd(s, errp);
-    if (re_s->fd < 0) {
-        ret = re_s->fd;
-        return ret;
-    }
-
-    return ret;
-}
-
-static void sd_reopen_commit(BDRVReopenState *state)
-{
-    BDRVSheepdogReopenState *re_s = state->opaque;
-    BDRVSheepdogState *s = state->bs->opaque;
-
-    if (s->fd) {
-        aio_set_fd_handler(s->aio_context, s->fd, false,
-                           NULL, NULL, NULL);
-        closesocket(s->fd);
-    }
-
-    s->fd = re_s->fd;
-    s->cache_flags = re_s->cache_flags;
-
-    g_free(state->opaque);
-    state->opaque = NULL;
-
-    return;
-}
-
-static void sd_reopen_abort(BDRVReopenState *state)
-{
-    BDRVSheepdogReopenState *re_s = state->opaque;
-    BDRVSheepdogState *s = state->bs->opaque;
-
-    if (re_s == NULL) {
-        return;
-    }
-
-    if (re_s->fd) {
-        aio_set_fd_handler(s->aio_context, re_s->fd, false,
-                           NULL, NULL, NULL);
-        closesocket(re_s->fd);
-    }
-
-    g_free(state->opaque);
-    state->opaque = NULL;
-
-    return;
-}
-
 static int do_sd_create(BDRVSheepdogState *s, uint32_t *vdi_id, int snapshot,
                        Error **errp)
 {
@@ -1641,7 +1605,7 @@ static int sd_prealloc(const char *filename, Error **errp)
    int ret;

    ret = bdrv_open(&bs, filename, NULL, NULL, BDRV_O_RDWR | BDRV_O_PROTOCOL,
-                    errp);
+                    NULL, errp);
    if (ret < 0) {
        goto out_with_err_set;
    }
@@ -1752,7 +1716,7 @@ static int parse_block_size_shift(BDRVSheepdogState *s, QemuOpts *opt)
        if ((object_size - 1) & object_size) {    /* not a power of 2? */
            return -EINVAL;
        }
-        obj_order = ctz32(object_size);
+        obj_order = ffs(object_size) - 1;
        if (obj_order < 20 || obj_order > 31) {
            return -EINVAL;
        }
@@ -1833,7 +1797,8 @@ static int sd_create(const char *filename, QemuOpts *opts,
        }

        bs = NULL;
-        ret = bdrv_open(&bs, backing_file, NULL, NULL, BDRV_O_PROTOCOL, errp);
+        ret = bdrv_open(&bs, backing_file, NULL, NULL, BDRV_O_PROTOCOL, NULL,
+                        errp);
        if (ret < 0) {
            goto out;
        }
@@ -1862,7 +1827,8 @@ static int sd_create(const char *filename, QemuOpts *opts,

        fd = connect_to_sdog(s, &local_err);
        if (fd < 0) {
-            error_report_err(local_err);
+            error_report("%s", error_get_pretty(local_err));
+            error_free(local_err);
            ret = -EIO;
            goto out;
        }
@@ -1946,8 +1912,7 @@ static void sd_close(BlockDriverState *bs)
        error_report("%s, %s", sd_strerror(rsp->result), s->name);
    }

-    aio_set_fd_handler(bdrv_get_aio_context(bs), s->fd,
-                       false, NULL, NULL, NULL);
+    aio_set_fd_handler(bdrv_get_aio_context(bs), s->fd, NULL, NULL, NULL);
    closesocket(s->fd);
    g_free(s->host_spec);
 }
@@ -2009,16 +1974,16 @@ static void coroutine_fn sd_write_done(SheepdogAIOCB *acb)
    AIOReq *aio_req;
    uint32_t offset, data_len, mn, mx;

-    mn = acb->min_dirty_data_idx;
-    mx = acb->max_dirty_data_idx;
+    mn = s->min_dirty_data_idx;
+    mx = s->max_dirty_data_idx;
    if (mn <= mx) {
        /* we need to update the vdi object. */
        offset = sizeof(s->inode) - sizeof(s->inode.data_vdi_id) +
            mn * sizeof(s->inode.data_vdi_id[0]);
        data_len = (mx - mn + 1) * sizeof(s->inode.data_vdi_id[0]);

-        acb->min_dirty_data_idx = UINT32_MAX;
-        acb->max_dirty_data_idx = 0;
+        s->min_dirty_data_idx = UINT32_MAX;
+        s->max_dirty_data_idx = 0;

        iov.iov_base = &s->inode;
        iov.iov_len = sizeof(s->inode);
@@ -2227,11 +2192,15 @@ static int coroutine_fn sd_co_rw_vector(void *p)
        }

        aio_req = alloc_aio_req(s, acb, oid, len, offset, flags, create,
-                                old_oid,
-                                acb->aiocb_type == AIOCB_DISCARD_OBJ ?
-                                0 : done);
+                                old_oid, done);
        QLIST_INSERT_HEAD(&s->inflight_aio_head, aio_req, aio_siblings);

+        if (create) {
+            if (check_simultaneous_create(s, aio_req)) {
+                goto done;
+            }
+        }
+
        add_aio_request(s, aio_req, acb->qiov->iov, acb->qiov->niov,
                        acb->aiocb_type);
    done:
@@ -2246,20 +2215,6 @@ out:
    return 1;
 }

-static bool check_overlapping_aiocb(BDRVSheepdogState *s, SheepdogAIOCB *aiocb)
-{
-    SheepdogAIOCB *cb;
-
-    QLIST_FOREACH(cb, &s->inflight_aiocb_head, aiocb_siblings) {
-        if (AIOCBOverlapping(aiocb, cb)) {
-            return true;
-        }
-    }
-
-    QLIST_INSERT_HEAD(&s->inflight_aiocb_head, aiocb, aiocb_siblings);
-    return false;
-}
-
 static coroutine_fn int sd_co_writev(BlockDriverState *bs, int64_t sector_num,
                        int nb_sectors, QEMUIOVector *qiov)
 {
@@ -2279,25 +2234,14 @@ static coroutine_fn int sd_co_writev(BlockDriverState *bs, int64_t sector_num,
    acb->aio_done_func = sd_write_done;
    acb->aiocb_type = AIOCB_WRITE_UDATA;

-retry:
-    if (check_overlapping_aiocb(s, acb)) {
-        qemu_co_queue_wait(&s->overlapping_queue);
-        goto retry;
-    }
-
    ret = sd_co_rw_vector(acb);
    if (ret <= 0) {
-        QLIST_REMOVE(acb, aiocb_siblings);
-        qemu_co_queue_restart_all(&s->overlapping_queue);
        qemu_aio_unref(acb);
        return ret;
    }

    qemu_coroutine_yield();

-    QLIST_REMOVE(acb, aiocb_siblings);
-    qemu_co_queue_restart_all(&s->overlapping_queue);
-
    return acb->ret;
 }

@@ -2306,30 +2250,19 @@ static coroutine_fn int sd_co_readv(BlockDriverState *bs, int64_t sector_num,
 {
    SheepdogAIOCB *acb;
    int ret;
-    BDRVSheepdogState *s = bs->opaque;

    acb = sd_aio_setup(bs, qiov, sector_num, nb_sectors);
    acb->aiocb_type = AIOCB_READ_UDATA;
    acb->aio_done_func = sd_finish_aiocb;

-retry:
-    if (check_overlapping_aiocb(s, acb)) {
-        qemu_co_queue_wait(&s->overlapping_queue);
-        goto retry;
-    }
-
    ret = sd_co_rw_vector(acb);
    if (ret <= 0) {
-        QLIST_REMOVE(acb, aiocb_siblings);
-        qemu_co_queue_restart_all(&s->overlapping_queue);
        qemu_aio_unref(acb);
        return ret;
    }

    qemu_coroutine_yield();

-    QLIST_REMOVE(acb, aiocb_siblings);
-    qemu_co_queue_restart_all(&s->overlapping_queue);
    return acb->ret;
 }

@@ -2406,8 +2339,8 @@ static int sd_snapshot_create(BlockDriverState *bs, QEMUSnapshotInfo *sn_info)

    ret = do_sd_create(s, &new_vid, 1, &local_err);
    if (ret < 0) {
-        error_reportf_err(local_err,
-                          "failed to create inode for snapshot: ");
+        error_report("failed to create inode for snapshot: %s",
+                     error_get_pretty(local_err));
        goto cleanup;
    }

@@ -2664,51 +2597,32 @@ static coroutine_fn int sd_co_discard(BlockDriverState *bs, int64_t sector_num,
                                      int nb_sectors)
 {
    SheepdogAIOCB *acb;
+    QEMUIOVector dummy;
    BDRVSheepdogState *s = bs->opaque;
    int ret;
-    QEMUIOVector discard_iov;
-    struct iovec iov;
-    uint32_t zero = 0;

    if (!s->discard_supported) {
            return 0;
    }

-    memset(&discard_iov, 0, sizeof(discard_iov));
-    memset(&iov, 0, sizeof(iov));
-    iov.iov_base = &zero;
-    iov.iov_len = sizeof(zero);
-    discard_iov.iov = &iov;
-    discard_iov.niov = 1;
-    acb = sd_aio_setup(bs, &discard_iov, sector_num, nb_sectors);
+    acb = sd_aio_setup(bs, &dummy, sector_num, nb_sectors);
    acb->aiocb_type = AIOCB_DISCARD_OBJ;
    acb->aio_done_func = sd_finish_aiocb;

-retry:
-    if (check_overlapping_aiocb(s, acb)) {
-        qemu_co_queue_wait(&s->overlapping_queue);
-        goto retry;
-    }
-
    ret = sd_co_rw_vector(acb);
    if (ret <= 0) {
-        QLIST_REMOVE(acb, aiocb_siblings);
-        qemu_co_queue_restart_all(&s->overlapping_queue);
        qemu_aio_unref(acb);
        return ret;
    }

    qemu_coroutine_yield();

-    QLIST_REMOVE(acb, aiocb_siblings);
-    qemu_co_queue_restart_all(&s->overlapping_queue);
-
    return acb->ret;
 }

 static coroutine_fn int64_t
 sd_co_get_block_status(BlockDriverState *bs, int64_t sector_num, int nb_sectors,
-                       int *pnum, BlockDriverState **file)
+                       int *pnum)
 {
    BDRVSheepdogState *s = bs->opaque;
    SheepdogInode *inode = &s->inode;
@@ -2739,9 +2653,6 @@ sd_co_get_block_status(BlockDriverState *bs, int64_t sector_num, int nb_sectors,
    if (*pnum > nb_sectors) {
        *pnum = nb_sectors;
    }
-    if (ret > 0 && ret & BDRV_BLOCK_OFFSET_VALID) {
-        *file = bs;
-    }
    return ret;
 }

@@ -2801,9 +2712,6 @@ static BlockDriver bdrv_sheepdog = {
    .instance_size  = sizeof(BDRVSheepdogState),
    .bdrv_needs_filename = true,
    .bdrv_file_open = sd_open,
-    .bdrv_reopen_prepare    = sd_reopen_prepare,
-    .bdrv_reopen_commit     = sd_reopen_commit,
-    .bdrv_reopen_abort      = sd_reopen_abort,
    .bdrv_close     = sd_close,
    .bdrv_create    = sd_create,
    .bdrv_has_zero_init = bdrv_has_zero_init_1,
@@ -2837,9 +2745,6 @@ static BlockDriver bdrv_sheepdog_tcp = {
    .instance_size  = sizeof(BDRVSheepdogState),
    .bdrv_needs_filename = true,
    .bdrv_file_open = sd_open,
-    .bdrv_reopen_prepare    = sd_reopen_prepare,
-    .bdrv_reopen_commit     = sd_reopen_commit,
-    .bdrv_reopen_abort      = sd_reopen_abort,
    .bdrv_close     = sd_close,
    .bdrv_create    = sd_create,
    .bdrv_has_zero_init = bdrv_has_zero_init_1,
@@ -2873,9 +2778,6 @@ static BlockDriver bdrv_sheepdog_unix = {
    .instance_size  = sizeof(BDRVSheepdogState),
    .bdrv_needs_filename = true,
    .bdrv_file_open = sd_open,
-    .bdrv_reopen_prepare    = sd_reopen_prepare,
-    .bdrv_reopen_commit     = sd_reopen_commit,
-    .bdrv_reopen_abort      = sd_reopen_abort,
    .bdrv_close     = sd_close,
    .bdrv_create    = sd_create,
    .bdrv_has_zero_init = bdrv_has_zero_init_1,
--- a/block/snapshot.c
+++ b/block/snapshot.c
@@ -22,10 +22,8 @@
 * THE SOFTWARE.
 */

-#include "qemu/osdep.h"
 #include "block/snapshot.h"
 #include "block/block_int.h"
-#include "qapi/qmp/qerror.h"

 QemuOptsList internal_snapshot_opts = {
    .name = "snapshot",
@@ -150,7 +148,7 @@ int bdrv_can_snapshot(BlockDriverState *bs)

    if (!drv->bdrv_snapshot_create) {
        if (bs->file != NULL) {
-            return bdrv_can_snapshot(bs->file->bs);
+            return bdrv_can_snapshot(bs->file);
        }
        return 0;
    }
@@ -169,7 +167,7 @@ int bdrv_snapshot_create(BlockDriverState *bs,
        return drv->bdrv_snapshot_create(bs, sn_info);
    }
    if (bs->file) {
-        return bdrv_snapshot_create(bs->file->bs, sn_info);
+        return bdrv_snapshot_create(bs->file, sn_info);
    }
    return -ENOTSUP;
 }
@@ -189,10 +187,10 @@ int bdrv_snapshot_goto(BlockDriverState *bs,

    if (bs->file) {
        drv->bdrv_close(bs);
-        ret = bdrv_snapshot_goto(bs->file->bs, snapshot_id);
+        ret = bdrv_snapshot_goto(bs->file, snapshot_id);
        open_ret = drv->bdrv_open(bs, NULL, bs->open_flags, NULL);
        if (open_ret < 0) {
-            bdrv_unref(bs->file->bs);
+            bdrv_unref(bs->file);
            bs->drv = NULL;
            return open_ret;
        }
@@ -230,10 +228,8 @@ int bdrv_snapshot_delete(BlockDriverState *bs,
                         Error **errp)
 {
    BlockDriver *drv = bs->drv;
-    int ret;
-
    if (!drv) {
-        error_setg(errp, QERR_DEVICE_HAS_NO_MEDIUM, bdrv_get_device_name(bs));
+        error_set(errp, QERR_DEVICE_HAS_NO_MEDIUM, bdrv_get_device_name(bs));
        return -ENOMEDIUM;
    }
    if (!snapshot_id && !name) {
@@ -242,26 +238,23 @@ int bdrv_snapshot_delete(BlockDriverState *bs,
    }

    /* drain all pending i/o before deleting snapshot */
-    bdrv_drained_begin(bs);
+    bdrv_drain_all();

    if (drv->bdrv_snapshot_delete) {
-        ret = drv->bdrv_snapshot_delete(bs, snapshot_id, name, errp);
-    } else if (bs->file) {
-        ret = bdrv_snapshot_delete(bs->file->bs, snapshot_id, name, errp);
-    } else {
-        error_setg(errp, "Block format '%s' used by device '%s' "
-                   "does not support internal snapshot deletion",
-                   drv->format_name, bdrv_get_device_name(bs));
-        ret = -ENOTSUP;
+        return drv->bdrv_snapshot_delete(bs, snapshot_id, name, errp);
    }
-
-    bdrv_drained_end(bs);
-    return ret;
+    if (bs->file) {
+        return bdrv_snapshot_delete(bs->file, snapshot_id, name, errp);
+    }
+    error_set(errp, QERR_BLOCK_FORMAT_FEATURE_NOT_SUPPORTED,
+              drv->format_name, bdrv_get_device_name(bs),
+              "internal snapshot deletion");
+    return -ENOTSUP;
 }

-int bdrv_snapshot_delete_by_id_or_name(BlockDriverState *bs,
-                                       const char *id_or_name,
-                                       Error **errp)
+void bdrv_snapshot_delete_by_id_or_name(BlockDriverState *bs,
+                                        const char *id_or_name,
+                                        Error **errp)
 {
    int ret;
    Error *local_err = NULL;
@@ -276,7 +269,6 @@ int bdrv_snapshot_delete_by_id_or_name(BlockDriverState *bs,
    if (ret < 0) {
        error_propagate(errp, local_err);
    }
-    return ret;
 }

 int bdrv_snapshot_list(BlockDriverState *bs,
@@ -290,7 +282,7 @@ int bdrv_snapshot_list(BlockDriverState *bs,
        return drv->bdrv_snapshot_list(bs, psn_info);
    }
    if (bs->file) {
-        return bdrv_snapshot_list(bs->file->bs, psn_info);
+        return bdrv_snapshot_list(bs->file, psn_info);
    }
    return -ENOTSUP;
 }
@@ -323,7 +315,7 @@ int bdrv_snapshot_load_tmp(BlockDriverState *bs,
    BlockDriver *drv = bs->drv;

    if (!drv) {
-        error_setg(errp, QERR_DEVICE_HAS_NO_MEDIUM, bdrv_get_device_name(bs));
+        error_set(errp, QERR_DEVICE_HAS_NO_MEDIUM, bdrv_get_device_name(bs));
        return -ENOMEDIUM;
    }
    if (!snapshot_id && !name) {
@@ -337,9 +329,9 @@ int bdrv_snapshot_load_tmp(BlockDriverState *bs,
    if (drv->bdrv_snapshot_load_tmp) {
        return drv->bdrv_snapshot_load_tmp(bs, snapshot_id, name, errp);
    }
-    error_setg(errp, "Block format '%s' used by device '%s' "
-               "does not support temporarily loading internal snapshots",
-               drv->format_name, bdrv_get_device_name(bs));
+    error_set(errp, QERR_BLOCK_FORMAT_FEATURE_NOT_SUPPORTED,
+              drv->format_name, bdrv_get_device_name(bs),
+              "temporarily load internal snapshot");
    return -ENOTSUP;
 }

@@ -363,130 +355,3 @@ int bdrv_snapshot_load_tmp_by_id_or_name(BlockDriverState *bs,

    return ret;
 }
-
-
-/* Group operations. All block drivers are involved.
- * These functions will properly handle dataplane (take aio_context_acquire
- * when appropriate for appropriate block drivers) */
-
-bool bdrv_all_can_snapshot(BlockDriverState **first_bad_bs)
-{
-    bool ok = true;
-    BlockDriverState *bs = NULL;
-
-    while (ok && (bs = bdrv_next(bs))) {
-        AioContext *ctx = bdrv_get_aio_context(bs);
-
-        aio_context_acquire(ctx);
-        if (bdrv_is_inserted(bs) && !bdrv_is_read_only(bs)) {
-            ok = bdrv_can_snapshot(bs);
-        }
-        aio_context_release(ctx);
-    }
-
-    *first_bad_bs = bs;
-    return ok;
-}
-
-int bdrv_all_delete_snapshot(const char *name, BlockDriverState **first_bad_bs,
-                             Error **err)
-{
-    int ret = 0;
-    BlockDriverState *bs = NULL;
-    QEMUSnapshotInfo sn1, *snapshot = &sn1;
-
-    while (ret == 0 && (bs = bdrv_next(bs))) {
-        AioContext *ctx = bdrv_get_aio_context(bs);
-
-        aio_context_acquire(ctx);
-        if (bdrv_can_snapshot(bs) &&
-                bdrv_snapshot_find(bs, snapshot, name) >= 0) {
-            ret = bdrv_snapshot_delete_by_id_or_name(bs, name, err);
-        }
-        aio_context_release(ctx);
-    }
-
-    *first_bad_bs = bs;
-    return ret;
-}
-
-
-int bdrv_all_goto_snapshot(const char *name, BlockDriverState **first_bad_bs)
-{
-    int err = 0;
-    BlockDriverState *bs = NULL;
-
-    while (err == 0 && (bs = bdrv_next(bs))) {
-        AioContext *ctx = bdrv_get_aio_context(bs);
-
-        aio_context_acquire(ctx);
-        if (bdrv_can_snapshot(bs)) {
-            err = bdrv_snapshot_goto(bs, name);
-        }
-        aio_context_release(ctx);
-    }
-
-    *first_bad_bs = bs;
-    return err;
-}
-
-int bdrv_all_find_snapshot(const char *name, BlockDriverState **first_bad_bs)
-{
-    QEMUSnapshotInfo sn;
-    int err = 0;
-    BlockDriverState *bs = NULL;
-
-    while (err == 0 && (bs = bdrv_next(bs))) {
-        AioContext *ctx = bdrv_get_aio_context(bs);
-
-        aio_context_acquire(ctx);
-        if (bdrv_can_snapshot(bs)) {
-            err = bdrv_snapshot_find(bs, &sn, name);
-        }
-        aio_context_release(ctx);
-    }
-
-    *first_bad_bs = bs;
-    return err;
-}
-
-int bdrv_all_create_snapshot(QEMUSnapshotInfo *sn,
-                             BlockDriverState *vm_state_bs,
-                             uint64_t vm_state_size,
-                             BlockDriverState **first_bad_bs)
-{
-    int err = 0;
-    BlockDriverState *bs = NULL;
-
-    while (err == 0 && (bs = bdrv_next(bs))) {
-        AioContext *ctx = bdrv_get_aio_context(bs);
-
-        aio_context_acquire(ctx);
-        if (bs == vm_state_bs) {
-            sn->vm_state_size = vm_state_size;
-            err = bdrv_snapshot_create(bs, sn);
-        } else if (bdrv_can_snapshot(bs)) {
-            sn->vm_state_size = 0;
-            err = bdrv_snapshot_create(bs, sn);
-        }
-        aio_context_release(ctx);
-    }
-
-    *first_bad_bs = bs;
-    return err;
-}
-
-BlockDriverState *bdrv_all_find_vmstate_bs(void)
-{
-    bool not_found = true;
-    BlockDriverState *bs = NULL;
-
-    while (not_found && (bs = bdrv_next(bs))) {
-        AioContext *ctx = bdrv_get_aio_context(bs);
-
-        aio_context_acquire(ctx);
-        not_found = !bdrv_can_snapshot(bs);
-        aio_context_release(ctx);
-    }
-    return bs;
-}
--- a/block/ssh.c
+++ b/block/ssh.c
@@ -22,17 +22,17 @@
 * THE SOFTWARE.
 */

-#include "qemu/osdep.h"
+#include <stdio.h>
+#include <stdlib.h>
+#include <stdarg.h>

 #include <libssh2.h>
 #include <libssh2_sftp.h>

 #include "block/block_int.h"
-#include "qemu/error-report.h"
 #include "qemu/sockets.h"
 #include "qemu/uri.h"
 #include "qapi/qmp/qint.h"
-#include "qapi/qmp/qstring.h"

 /* DEBUG_SSH=1 enables the DPRINTF (debugging printf) statements in
 * this block driver code.
@@ -191,7 +191,7 @@ sftp_error_report(BDRVSSHState *s, const char *fs, ...)
 static int parse_uri(const char *filename, QDict *options, Error **errp)
 {
    URI *uri = NULL;
-    QueryParams *qp;
+    QueryParams *qp = NULL;
    int i;

    uri = uri_parse(filename);
@@ -247,6 +247,9 @@ static int parse_uri(const char *filename, QDict *options, Error **errp)
    return 0;

 err:
+    if (qp) {
+      query_params_free(qp);
+    }
    if (uri) {
      uri_free(uri);
    }
@@ -558,7 +561,7 @@ static int connect_to_ssh(BDRVSSHState *s, QDict *options,
    /* Open the socket and connect. */
    s->sock = inet_connect(s->hostport, errp);
    if (s->sock < 0) {
-        ret = -EIO;
+        ret = -errno;
        goto err;
    }

@@ -798,15 +801,14 @@ static coroutine_fn void set_fd_handler(BDRVSSHState *s, BlockDriverState *bs)
            rd_handler, wr_handler);

    aio_set_fd_handler(bdrv_get_aio_context(bs), s->sock,
-                       false, rd_handler, wr_handler, co);
+                       rd_handler, wr_handler, co);
 }

 static coroutine_fn void clear_fd_handler(BDRVSSHState *s,
                                          BlockDriverState *bs)
 {
    DPRINTF("s->sock=%d", s->sock);
-    aio_set_fd_handler(bdrv_get_aio_context(bs), s->sock,
-                       false, NULL, NULL, NULL);
+    aio_set_fd_handler(bdrv_get_aio_context(bs), s->sock, NULL, NULL, NULL);
 }

 /* A non-blocking call returned EAGAIN, so yield, ensuring the
--- a/block/stream.c
+++ b/block/stream.c
@@ -11,13 +11,10 @@
 *
 */

-#include "qemu/osdep.h"
 #include "trace.h"
 #include "block/block_int.h"
 #include "block/blockjob.h"
-#include "qapi/qmp/qerror.h"
 #include "qemu/ratelimit.h"
-#include "sysemu/block-backend.h"

 enum {
    /*
@@ -54,6 +51,34 @@ static int coroutine_fn stream_populate(BlockDriverState *bs,
    return bdrv_co_copy_on_readv(bs, sector_num, nb_sectors, &qiov);
 }

+static void close_unused_images(BlockDriverState *top, BlockDriverState *base,
+                                const char *base_id)
+{
+    BlockDriverState *intermediate;
+    intermediate = top->backing_hd;
+
+    /* Must assign before bdrv_delete() to prevent traversing dangling pointer
+     * while we delete backing image instances.
+     */
+    bdrv_set_backing_hd(top, base);
+
+    while (intermediate) {
+        BlockDriverState *unused;
+
+        /* reached base */
+        if (intermediate == base) {
+            break;
+        }
+
+        unused = intermediate;
+        intermediate = intermediate->backing_hd;
+        bdrv_set_backing_hd(unused, NULL);
+        bdrv_unref(unused);
+    }
+
+    bdrv_refresh_limits(top, NULL);
+}
+
 typedef struct {
    int ret;
    bool reached_end;
@@ -75,7 +100,7 @@ static void stream_complete(BlockJob *job, void *opaque)
            }
        }
        data->ret = bdrv_change_backing_file(job->bs, base_id, base_fmt);
-        bdrv_set_backing_hd(job->bs, base);
+        close_unused_images(job->bs, base, base_id);
    }

    g_free(s->backing_file_str);
@@ -95,7 +120,7 @@ static void coroutine_fn stream_run(void *opaque)
    int n = 0;
    void *buf;

-    if (!bs->backing) {
+    if (!bs->backing_hd) {
        block_job_completed(&s->common, 0);
        return;
    }
@@ -140,7 +165,7 @@ wait:
        } else if (ret >= 0) {
            /* Copy if allocated in the intermediate images.  Limit to the
             * known-unallocated area [sector_num, sector_num+n).  */
-            ret = bdrv_is_allocated_above(backing_bs(bs), base,
+            ret = bdrv_is_allocated_above(bs->backing_hd, base,
                                          sector_num, n, &n);

            /* Finish early if end of backing file has been reached */
@@ -202,7 +227,7 @@ static void stream_set_speed(BlockJob *job, int64_t speed, Error **errp)
    StreamBlockJob *s = container_of(job, StreamBlockJob, common);

    if (speed < 0) {
-        error_setg(errp, QERR_INVALID_PARAMETER, "speed");
+        error_set(errp, QERR_INVALID_PARAMETER, "speed");
        return;
    }
    ratelimit_set_speed(&s->limit, speed / BDRV_SECTOR_SIZE, SLICE_TIME);
@@ -224,8 +249,8 @@ void stream_start(BlockDriverState *bs, BlockDriverState *base,

    if ((on_error == BLOCKDEV_ON_ERROR_STOP ||
         on_error == BLOCKDEV_ON_ERROR_ENOSPC) &&
-        (!bs->blk || !blk_iostatus_is_enabled(bs->blk))) {
-        error_setg(errp, QERR_INVALID_PARAMETER, "on-error");
+        !bdrv_iostatus_is_enabled(bs)) {
+        error_set(errp, QERR_INVALID_PARAMETER, "on-error");
        return;
    }

--- a/block/tar.c
+++ b/block/tar.c
@@ -0,0 +1,377 @@
+/*
+ * Tar block driver
+ *
+ * Copyright (c) 2009 Alexander Graf <agraf@suse.de>
+ *
+ * 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-common.h"
+#include "block/block_int.h"
+
+// #define DEBUG
+
+#ifdef DEBUG
+#define dprintf(fmt, ...) do { printf("tar: " fmt, ## __VA_ARGS__); } while (0)
+#else
+#define dprintf(fmt, ...) do { } while (0)
+#endif
+
+#define SECTOR_SIZE      512
+
+#define POSIX_TAR_MAGIC  "ustar"
+#define OFFS_LENGTH      0x7c
+#define OFFS_TYPE        0x9c
+#define OFFS_MAGIC       0x101
+
+#define OFFS_S_SP        0x182
+#define OFFS_S_EXT       0x1e2
+#define OFFS_S_LENGTH    0x1e3
+#define OFFS_SX_EXT      0x1f8
+
+typedef struct SparseCache {
+    uint64_t start;
+    uint64_t end;
+} SparseCache;
+
+typedef struct BDRVTarState {
+    BlockDriverState *hd;
+    size_t file_sec;
+    uint64_t file_len;
+    SparseCache *sparse;
+    int sparse_num;
+    uint64_t last_end;
+    char longfile[2048];
+} BDRVTarState;
+
+static int str_ends(char *str, const char *end)
+{
+    int end_len = strlen(end);
+    int str_len = strlen(str);
+
+    if (str_len < end_len)
+        return 0;
+
+    return !strncmp(str + str_len - end_len, end, end_len);
+}
+
+static int is_target_file(BlockDriverState *bs, char *filename,
+                          char *header)
+{
+    int retval = 0;
+
+    if (str_ends(filename, ".raw"))
+        retval = 1;
+
+    if (str_ends(filename, ".qcow"))
+        retval = 1;
+
+    if (str_ends(filename, ".qcow2"))
+        retval = 1;
+
+    if (str_ends(filename, ".vmdk"))
+        retval = 1;
+
+    if (retval &&
+        (header[OFFS_TYPE] != '0') &&
+        (header[OFFS_TYPE] != 'S')) {
+        retval = 0;
+    }
+
+    dprintf("does filename %s match? %s\n", filename, retval ? "yes" : "no");
+
+    /* make sure we're not using this name again */
+    filename[0] = '\0';
+
+    return retval;
+}
+
+static uint64_t tar2u64(char *ptr)
+{
+    uint64_t retval;
+    char oldend = ptr[12];
+
+    ptr[12] = '\0';
+    if (*ptr & 0x80) {
+        /* XXX we only support files up to 64 bit length */
+        retval = be64_to_cpu(*(uint64_t *)(ptr+4));
+        dprintf("Convert %lx -> %#lx\n", *(uint64_t*)(ptr+4), retval);
+    } else {
+        retval = strtol(ptr, NULL, 8);
+        dprintf("Convert %s -> %#lx\n", ptr, retval);
+    }
+
+    ptr[12] = oldend;
+
+    return retval;
+}
+
+static void tar_sparse(BDRVTarState *s, uint64_t offs, uint64_t len)
+{
+    SparseCache *sparse;
+
+    if (!len)
+        return;
+    if (!(offs - s->last_end)) {
+        s->last_end += len;
+        return;
+    }
+    if (s->last_end > offs)
+        return;
+
+    dprintf("Last chunk until %lx new chunk at %lx\n", s->last_end, offs);
+
+    s->sparse = g_realloc(s->sparse, (s->sparse_num + 1) * sizeof(SparseCache));
+    sparse = &s->sparse[s->sparse_num];
+    sparse->start = s->last_end;
+    sparse->end = offs;
+    s->last_end = offs + len;
+    s->sparse_num++;
+    dprintf("Sparse at %lx end=%lx\n", sparse->start,
+                                       sparse->end);
+}
+
+static QemuOptsList runtime_opts = {
+    .name = "tar",
+    .head = QTAILQ_HEAD_INITIALIZER(runtime_opts.head),
+    .desc = {
+        {
+            .name = "filename",
+            .type = QEMU_OPT_STRING,
+            .help = "URL to the tar file",
+        },
+        { /* end of list */ }
+    },
+};
+
+static int tar_open(BlockDriverState *bs, QDict *options, int flags, Error **errp)
+{
+    BDRVTarState *s = bs->opaque;
+    char header[SECTOR_SIZE];
+    char *real_file = header;
+    char *magic;
+    size_t header_offs = 0;
+    int ret;
+    QemuOpts *opts;
+    Error *local_err = NULL;
+    const char *filename;
+
+    opts = qemu_opts_create(&runtime_opts, NULL, 0, &error_abort);
+    qemu_opts_absorb_qdict(opts, options, &local_err);
+    if (local_err != NULL) {
+        error_propagate(errp, local_err);
+        ret = -EINVAL;
+        goto fail;
+    }
+
+    filename = qemu_opt_get(opts, "filename");
+
+    if (!strncmp(filename, "tar://", 6))
+        filename += 6;
+    else if (!strncmp(filename, "tar:", 4))
+        filename += 4;
+
+    ret = bdrv_open(&s->hd, filename, NULL, NULL, flags | BDRV_O_PROTOCOL, NULL, &local_err);
+    if (ret < 0) {
+        error_propagate(errp, local_err);
+        qemu_opts_del(opts);
+        return ret;
+    }
+
+    /* Search the file for an image */
+
+    do {
+        /* tar header */
+        if (bdrv_pread(s->hd, header_offs, header, SECTOR_SIZE) != SECTOR_SIZE)
+            goto fail;
+
+        if ((header_offs > 1) && !header[0]) {
+            fprintf(stderr, "Tar: No image file found in archive\n");
+            goto fail;
+        }
+
+        magic = &header[OFFS_MAGIC];
+        if (strncmp(magic, POSIX_TAR_MAGIC, 5)) {
+            fprintf(stderr, "Tar: Invalid magic: %s\n", magic);
+            goto fail;
+        }
+
+        dprintf("file type: %c\n", header[OFFS_TYPE]);
+
+        /* file length*/
+        s->file_len = (tar2u64(&header[OFFS_LENGTH]) + (SECTOR_SIZE - 1)) &
+                      ~(SECTOR_SIZE - 1);
+        s->file_sec = (header_offs / SECTOR_SIZE) + 1;
+
+        header_offs += s->file_len + SECTOR_SIZE;
+
+        if (header[OFFS_TYPE] == 'L') {
+            bdrv_pread(s->hd, header_offs - s->file_len, s->longfile,
+                       sizeof(s->longfile));
+            s->longfile[sizeof(s->longfile)-1] = '\0';
+            real_file = header;
+        } else if (s->longfile[0]) {
+            real_file = s->longfile;
+        } else {
+            real_file = header;
+        }
+    } while(!is_target_file(bs, real_file, header));
+
+    /* We found an image! */
+
+    if (header[OFFS_TYPE] == 'S') {
+        uint8_t isextended;
+        int i;
+
+        for (i = OFFS_S_SP; i < (OFFS_S_SP + (4 * 24)); i += 24)
+            tar_sparse(s, tar2u64(&header[i]), tar2u64(&header[i+12]));
+
+        s->file_len = tar2u64(&header[OFFS_S_LENGTH]);
+        isextended = header[OFFS_S_EXT];
+
+        while (isextended) {
+            if (bdrv_pread(s->hd, s->file_sec * SECTOR_SIZE, header,
+                           SECTOR_SIZE) != SECTOR_SIZE)
+                goto fail;
+
+            for (i = 0; i < (21 * 24); i += 24)
+                tar_sparse(s, tar2u64(&header[i]), tar2u64(&header[i+12]));
+            isextended = header[OFFS_SX_EXT];
+            s->file_sec++;
+        }
+        tar_sparse(s, s->file_len, 1);
+    }
+    qemu_opts_del(opts);
+
+    return 0;
+
+fail:
+    fprintf(stderr, "Tar: Error opening file\n");
+    bdrv_unref(s->hd);
+    qemu_opts_del(opts);
+    return -EINVAL;
+}
+
+typedef struct TarAIOCB {
+    BlockAIOCB common;
+    QEMUBH *bh;
+} TarAIOCB;
+
+/* This callback gets invoked when we have pure sparseness */
+static void tar_sparse_cb(void *opaque)
+{
+    TarAIOCB *acb = (TarAIOCB *)opaque;
+
+    acb->common.cb(acb->common.opaque, 0);
+    qemu_bh_delete(acb->bh);
+    qemu_aio_unref(acb);
+}
+
+static AIOCBInfo tar_aiocb_info = {
+    .aiocb_size         = sizeof(TarAIOCB),
+};
+
+/* This is where we get a request from a caller to read something */
+static BlockAIOCB *tar_aio_readv(BlockDriverState *bs,
+        int64_t sector_num, QEMUIOVector *qiov, int nb_sectors,
+        BlockCompletionFunc *cb, void *opaque)
+{
+    BDRVTarState *s = bs->opaque;
+    SparseCache *sparse;
+    int64_t sec_file = sector_num + s->file_sec;
+    int64_t start = sector_num * SECTOR_SIZE;
+    int64_t end = start + (nb_sectors * SECTOR_SIZE);
+    int i;
+    TarAIOCB *acb;
+
+    for (i = 0; i < s->sparse_num; i++) {
+        sparse = &s->sparse[i];
+        if (sparse->start > end) {
+            /* We expect the cache to be start increasing */
+            break;
+        } else if ((sparse->start < start) && (sparse->end <= start)) {
+            /* sparse before our offset */
+            sec_file -= (sparse->end - sparse->start) / SECTOR_SIZE;
+        } else if ((sparse->start <= start) && (sparse->end >= end)) {
+            /* all our sectors are sparse */
+            char *buf = g_malloc0(nb_sectors * SECTOR_SIZE);
+
+            acb = qemu_aio_get(&tar_aiocb_info, bs, cb, opaque);
+            qemu_iovec_from_buf(qiov, 0, buf, nb_sectors * SECTOR_SIZE);
+            g_free(buf);
+            acb->bh = qemu_bh_new(tar_sparse_cb, acb);
+            qemu_bh_schedule(acb->bh);
+
+            return &acb->common;
+        } else if (((sparse->start >= start) && (sparse->start < end)) ||
+                   ((sparse->end >= start) && (sparse->end < end))) {
+            /* we're semi-sparse (worst case) */
+            /* let's go synchronous and read all sectors individually */
+            char *buf = g_malloc(nb_sectors * SECTOR_SIZE);
+            uint64_t offs;
+
+            for (offs = 0; offs < (nb_sectors * SECTOR_SIZE);
+                 offs += SECTOR_SIZE) {
+                bdrv_pread(bs, (sector_num * SECTOR_SIZE) + offs,
+                           buf + offs, SECTOR_SIZE);
+            }
+
+            qemu_iovec_from_buf(qiov, 0, buf, nb_sectors * SECTOR_SIZE);
+            acb = qemu_aio_get(&tar_aiocb_info, bs, cb, opaque);
+            acb->bh = qemu_bh_new(tar_sparse_cb, acb);
+            qemu_bh_schedule(acb->bh);
+
+            return &acb->common;
+        }
+    }
+
+    return bdrv_aio_readv(s->hd, sec_file, qiov, nb_sectors,
+                          cb, opaque);
+}
+
+static void tar_close(BlockDriverState *bs)
+{
+    dprintf("Close\n");
+}
+
+static int64_t tar_getlength(BlockDriverState *bs)
+{
+    BDRVTarState *s = bs->opaque;
+    dprintf("getlength -> %ld\n", s->file_len);
+    return s->file_len;
+}
+
+static BlockDriver bdrv_tar = {
+    .format_name     = "tar",
+    .protocol_name   = "tar",
+
+    .instance_size   = sizeof(BDRVTarState),
+    .bdrv_file_open  = tar_open,
+    .bdrv_close      = tar_close,
+    .bdrv_getlength  = tar_getlength,
+
+    .bdrv_aio_readv  = tar_aio_readv,
+};
+
+static void tar_block_init(void)
+{
+    bdrv_register(&bdrv_tar);
+}
+
+block_init(tar_block_init);
--- a/block/throttle-groups.c
+++ b/block/throttle-groups.c
@@ -1,483 +0,0 @@
-/*
- * QEMU block throttling group infrastructure
- *
- * Copyright (C) Nodalink, EURL. 2014
- * Copyright (C) Igalia, S.L. 2015
- *
- * Authors:
- *   Benoît Canet <benoit.canet@nodalink.com>
- *   Alberto Garcia <berto@igalia.com>
- *
- * This program is free software; you can redistribute it and/or
- * modify it under the terms of the GNU General Public License as
- * published by the Free Software Foundation; either version 2 or
- * (at your option) version 3 of the License.
- *
- * This program is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program; if not, see <http://www.gnu.org/licenses/>.
- */
-
-#include "qemu/osdep.h"
-#include "block/throttle-groups.h"
-#include "qemu/queue.h"
-#include "qemu/thread.h"
-#include "sysemu/qtest.h"
-
-/* The ThrottleGroup structure (with its ThrottleState) is shared
- * among different BlockDriverState and it's independent from
- * AioContext, so in order to use it from different threads it needs
- * its own locking.
- *
- * This locking is however handled internally in this file, so it's
- * transparent to outside users.
- *
- * The whole ThrottleGroup structure is private and invisible to
- * outside users, that only use it through its ThrottleState.
- *
- * In addition to the ThrottleGroup structure, BlockDriverState has
- * fields that need to be accessed by other members of the group and
- * therefore also need to be protected by this lock. Once a BDS is
- * registered in a group those fields can be accessed by other threads
- * any time.
- *
- * Again, all this is handled internally and is mostly transparent to
- * the outside. The 'throttle_timers' field however has an additional
- * constraint because it may be temporarily invalid (see for example
- * bdrv_set_aio_context()). Therefore in this file a thread will
- * access some other BDS's timers only after verifying that that BDS
- * has throttled requests in the queue.
- */
-typedef struct ThrottleGroup {
-    char *name; /* This is constant during the lifetime of the group */
-
-    QemuMutex lock; /* This lock protects the following four fields */
-    ThrottleState ts;
-    QLIST_HEAD(, BlockDriverState) head;
-    BlockDriverState *tokens[2];
-    bool any_timer_armed[2];
-
-    /* These two are protected by the global throttle_groups_lock */
-    unsigned refcount;
-    QTAILQ_ENTRY(ThrottleGroup) list;
-} ThrottleGroup;
-
-static QemuMutex throttle_groups_lock;
-static QTAILQ_HEAD(, ThrottleGroup) throttle_groups =
-    QTAILQ_HEAD_INITIALIZER(throttle_groups);
-
-/* Increments the reference count of a ThrottleGroup given its name.
- *
- * If no ThrottleGroup is found with the given name a new one is
- * created.
- *
- * @name: the name of the ThrottleGroup
- * @ret:  the ThrottleState member of the ThrottleGroup
- */
-ThrottleState *throttle_group_incref(const char *name)
-{
-    ThrottleGroup *tg = NULL;
-    ThrottleGroup *iter;
-
-    qemu_mutex_lock(&throttle_groups_lock);
-
-    /* Look for an existing group with that name */
-    QTAILQ_FOREACH(iter, &throttle_groups, list) {
-        if (!strcmp(name, iter->name)) {
-            tg = iter;
-            break;
-        }
-    }
-
-    /* Create a new one if not found */
-    if (!tg) {
-        tg = g_new0(ThrottleGroup, 1);
-        tg->name = g_strdup(name);
-        qemu_mutex_init(&tg->lock);
-        throttle_init(&tg->ts);
-        QLIST_INIT(&tg->head);
-
-        QTAILQ_INSERT_TAIL(&throttle_groups, tg, list);
-    }
-
-    tg->refcount++;
-
-    qemu_mutex_unlock(&throttle_groups_lock);
-
-    return &tg->ts;
-}
-
-/* Decrease the reference count of a ThrottleGroup.
- *
- * When the reference count reaches zero the ThrottleGroup is
- * destroyed.
- *
- * @ts:  The ThrottleGroup to unref, given by its ThrottleState member
- */
-void throttle_group_unref(ThrottleState *ts)
-{
-    ThrottleGroup *tg = container_of(ts, ThrottleGroup, ts);
-
-    qemu_mutex_lock(&throttle_groups_lock);
-    if (--tg->refcount == 0) {
-        QTAILQ_REMOVE(&throttle_groups, tg, list);
-        qemu_mutex_destroy(&tg->lock);
-        g_free(tg->name);
-        g_free(tg);
-    }
-    qemu_mutex_unlock(&throttle_groups_lock);
-}
-
-/* Get the name from a BlockDriverState's ThrottleGroup. The name (and
- * the pointer) is guaranteed to remain constant during the lifetime
- * of the group.
- *
- * @bs:   a BlockDriverState that is member of a throttling group
- * @ret:  the name of the group.
- */
-const char *throttle_group_get_name(BlockDriverState *bs)
-{
-    ThrottleGroup *tg = container_of(bs->throttle_state, ThrottleGroup, ts);
-    return tg->name;
-}
-
-/* Return the next BlockDriverState in the round-robin sequence,
- * simulating a circular list.
- *
- * This assumes that tg->lock is held.
- *
- * @bs:  the current BlockDriverState
- * @ret: the next BlockDriverState in the sequence
- */
-static BlockDriverState *throttle_group_next_bs(BlockDriverState *bs)
-{
-    ThrottleState *ts = bs->throttle_state;
-    ThrottleGroup *tg = container_of(ts, ThrottleGroup, ts);
-    BlockDriverState *next = QLIST_NEXT(bs, round_robin);
-
-    if (!next) {
-        return QLIST_FIRST(&tg->head);
-    }
-
-    return next;
-}
-
-/* Return the next BlockDriverState in the round-robin sequence with
- * pending I/O requests.
- *
- * This assumes that tg->lock is held.
- *
- * @bs:        the current BlockDriverState
- * @is_write:  the type of operation (read/write)
- * @ret:       the next BlockDriverState with pending requests, or bs
- *             if there is none.
- */
-static BlockDriverState *next_throttle_token(BlockDriverState *bs,
-                                             bool is_write)
-{
-    ThrottleGroup *tg = container_of(bs->throttle_state, ThrottleGroup, ts);
-    BlockDriverState *token, *start;
-
-    start = token = tg->tokens[is_write];
-
-    /* get next bs round in round robin style */
-    token = throttle_group_next_bs(token);
-    while (token != start && !token->pending_reqs[is_write]) {
-        token = throttle_group_next_bs(token);
-    }
-
-    /* If no IO are queued for scheduling on the next round robin token
-     * then decide the token is the current bs because chances are
-     * the current bs get the current request queued.
-     */
-    if (token == start && !token->pending_reqs[is_write]) {
-        token = bs;
-    }
-
-    return token;
-}
-
-/* Check if the next I/O request for a BlockDriverState needs to be
- * throttled or not. If there's no timer set in this group, set one
- * and update the token accordingly.
- *
- * This assumes that tg->lock is held.
- *
- * @bs:         the current BlockDriverState
- * @is_write:   the type of operation (read/write)
- * @ret:        whether the I/O request needs to be throttled or not
- */
-static bool throttle_group_schedule_timer(BlockDriverState *bs,
-                                          bool is_write)
-{
-    ThrottleState *ts = bs->throttle_state;
-    ThrottleTimers *tt = &bs->throttle_timers;
-    ThrottleGroup *tg = container_of(ts, ThrottleGroup, ts);
-    bool must_wait;
-
-    /* Check if any of the timers in this group is already armed */
-    if (tg->any_timer_armed[is_write]) {
-        return true;
-    }
-
-    must_wait = throttle_schedule_timer(ts, tt, is_write);
-
-    /* If a timer just got armed, set bs as the current token */
-    if (must_wait) {
-        tg->tokens[is_write] = bs;
-        tg->any_timer_armed[is_write] = true;
-    }
-
-    return must_wait;
-}
-
-/* Look for the next pending I/O request and schedule it.
- *
- * This assumes that tg->lock is held.
- *
- * @bs:        the current BlockDriverState
- * @is_write:  the type of operation (read/write)
- */
-static void schedule_next_request(BlockDriverState *bs, bool is_write)
-{
-    ThrottleGroup *tg = container_of(bs->throttle_state, ThrottleGroup, ts);
-    bool must_wait;
-    BlockDriverState *token;
-
-    /* Check if there's any pending request to schedule next */
-    token = next_throttle_token(bs, is_write);
-    if (!token->pending_reqs[is_write]) {
-        return;
-    }
-
-    /* Set a timer for the request if it needs to be throttled */
-    must_wait = throttle_group_schedule_timer(token, is_write);
-
-    /* If it doesn't have to wait, queue it for immediate execution */
-    if (!must_wait) {
-        /* Give preference to requests from the current bs */
-        if (qemu_in_coroutine() &&
-            qemu_co_queue_next(&bs->throttled_reqs[is_write])) {
-            token = bs;
-        } else {
-            ThrottleTimers *tt = &token->throttle_timers;
-            int64_t now = qemu_clock_get_ns(tt->clock_type);
-            timer_mod(tt->timers[is_write], now + 1);
-            tg->any_timer_armed[is_write] = true;
-        }
-        tg->tokens[is_write] = token;
-    }
-}
-
-/* Check if an I/O request needs to be throttled, wait and set a timer
- * if necessary, and schedule the next request using a round robin
- * algorithm.
- *
- * @bs:        the current BlockDriverState
- * @bytes:     the number of bytes for this I/O
- * @is_write:  the type of operation (read/write)
- */
-void coroutine_fn throttle_group_co_io_limits_intercept(BlockDriverState *bs,
-                                                        unsigned int bytes,
-                                                        bool is_write)
-{
-    bool must_wait;
-    BlockDriverState *token;
-
-    ThrottleGroup *tg = container_of(bs->throttle_state, ThrottleGroup, ts);
-    qemu_mutex_lock(&tg->lock);
-
-    /* First we check if this I/O has to be throttled. */
-    token = next_throttle_token(bs, is_write);
-    must_wait = throttle_group_schedule_timer(token, is_write);
-
-    /* Wait if there's a timer set or queued requests of this type */
-    if (must_wait || bs->pending_reqs[is_write]) {
-        bs->pending_reqs[is_write]++;
-        qemu_mutex_unlock(&tg->lock);
-        qemu_co_queue_wait(&bs->throttled_reqs[is_write]);
-        qemu_mutex_lock(&tg->lock);
-        bs->pending_reqs[is_write]--;
-    }
-
-    /* The I/O will be executed, so do the accounting */
-    throttle_account(bs->throttle_state, is_write, bytes);
-
-    /* Schedule the next request */
-    schedule_next_request(bs, is_write);
-
-    qemu_mutex_unlock(&tg->lock);
-}
-
-/* Update the throttle configuration for a particular group. Similar
- * to throttle_config(), but guarantees atomicity within the
- * throttling group.
- *
- * @bs:  a BlockDriverState that is member of the group
- * @cfg: the configuration to set
- */
-void throttle_group_config(BlockDriverState *bs, ThrottleConfig *cfg)
-{
-    ThrottleTimers *tt = &bs->throttle_timers;
-    ThrottleState *ts = bs->throttle_state;
-    ThrottleGroup *tg = container_of(ts, ThrottleGroup, ts);
-    qemu_mutex_lock(&tg->lock);
-    /* throttle_config() cancels the timers */
-    if (timer_pending(tt->timers[0])) {
-        tg->any_timer_armed[0] = false;
-    }
-    if (timer_pending(tt->timers[1])) {
-        tg->any_timer_armed[1] = false;
-    }
-    throttle_config(ts, tt, cfg);
-    qemu_mutex_unlock(&tg->lock);
-}
-
-/* Get the throttle configuration from a particular group. Similar to
- * throttle_get_config(), but guarantees atomicity within the
- * throttling group.
- *
- * @bs:  a BlockDriverState that is member of the group
- * @cfg: the configuration will be written here
- */
-void throttle_group_get_config(BlockDriverState *bs, ThrottleConfig *cfg)
-{
-    ThrottleState *ts = bs->throttle_state;
-    ThrottleGroup *tg = container_of(ts, ThrottleGroup, ts);
-    qemu_mutex_lock(&tg->lock);
-    throttle_get_config(ts, cfg);
-    qemu_mutex_unlock(&tg->lock);
-}
-
-/* ThrottleTimers callback. This wakes up a request that was waiting
- * because it had been throttled.
- *
- * @bs:        the BlockDriverState whose request had been throttled
- * @is_write:  the type of operation (read/write)
- */
-static void timer_cb(BlockDriverState *bs, bool is_write)
-{
-    ThrottleState *ts = bs->throttle_state;
-    ThrottleGroup *tg = container_of(ts, ThrottleGroup, ts);
-    bool empty_queue;
-
-    /* The timer has just been fired, so we can update the flag */
-    qemu_mutex_lock(&tg->lock);
-    tg->any_timer_armed[is_write] = false;
-    qemu_mutex_unlock(&tg->lock);
-
-    /* Run the request that was waiting for this timer */
-    empty_queue = !qemu_co_enter_next(&bs->throttled_reqs[is_write]);
-
-    /* If the request queue was empty then we have to take care of
-     * scheduling the next one */
-    if (empty_queue) {
-        qemu_mutex_lock(&tg->lock);
-        schedule_next_request(bs, is_write);
-        qemu_mutex_unlock(&tg->lock);
-    }
-}
-
-static void read_timer_cb(void *opaque)
-{
-    timer_cb(opaque, false);
-}
-
-static void write_timer_cb(void *opaque)
-{
-    timer_cb(opaque, true);
-}
-
-/* Register a BlockDriverState in the throttling group, also
- * initializing its timers and updating its throttle_state pointer to
- * point to it. If a throttling group with that name does not exist
- * yet, it will be created.
- *
- * @bs:        the BlockDriverState to insert
- * @groupname: the name of the group
- */
-void throttle_group_register_bs(BlockDriverState *bs, const char *groupname)
-{
-    int i;
-    ThrottleState *ts = throttle_group_incref(groupname);
-    ThrottleGroup *tg = container_of(ts, ThrottleGroup, ts);
-    int clock_type = QEMU_CLOCK_REALTIME;
-
-    if (qtest_enabled()) {
-        /* For testing block IO throttling only */
-        clock_type = QEMU_CLOCK_VIRTUAL;
-    }
-
-    bs->throttle_state = ts;
-
-    qemu_mutex_lock(&tg->lock);
-    /* If the ThrottleGroup is new set this BlockDriverState as the token */
-    for (i = 0; i < 2; i++) {
-        if (!tg->tokens[i]) {
-            tg->tokens[i] = bs;
-        }
-    }
-
-    QLIST_INSERT_HEAD(&tg->head, bs, round_robin);
-
-    throttle_timers_init(&bs->throttle_timers,
-                         bdrv_get_aio_context(bs),
-                         clock_type,
-                         read_timer_cb,
-                         write_timer_cb,
-                         bs);
-
-    qemu_mutex_unlock(&tg->lock);
-}
-
-/* Unregister a BlockDriverState from its group, removing it from the
- * list, destroying the timers and setting the throttle_state pointer
- * to NULL.
- *
- * The BlockDriverState must not have pending throttled requests, so
- * the caller has to drain them first.
- *
- * The group will be destroyed if it's empty after this operation.
- *
- * @bs: the BlockDriverState to remove
- */
-void throttle_group_unregister_bs(BlockDriverState *bs)
-{
-    ThrottleGroup *tg = container_of(bs->throttle_state, ThrottleGroup, ts);
-    int i;
-
-    assert(bs->pending_reqs[0] == 0 && bs->pending_reqs[1] == 0);
-    assert(qemu_co_queue_empty(&bs->throttled_reqs[0]));
-    assert(qemu_co_queue_empty(&bs->throttled_reqs[1]));
-
-    qemu_mutex_lock(&tg->lock);
-    for (i = 0; i < 2; i++) {
-        if (tg->tokens[i] == bs) {
-            BlockDriverState *token = throttle_group_next_bs(bs);
-            /* Take care of the case where this is the last bs in the group */
-            if (token == bs) {
-                token = NULL;
-            }
-            tg->tokens[i] = token;
-        }
-    }
-
-    /* remove the current bs from the list */
-    QLIST_REMOVE(bs, round_robin);
-    throttle_timers_destroy(&bs->throttle_timers);
-    qemu_mutex_unlock(&tg->lock);
-
-    throttle_group_unref(&tg->ts);
-    bs->throttle_state = NULL;
-}
-
-static void throttle_groups_init(void)
-{
-    qemu_mutex_init(&throttle_groups_lock);
-}
-
-block_init(throttle_groups_init);
--- a/block/vdi.c
+++ b/block/vdi.c
@@ -49,12 +49,11 @@
 * so this seems to be reasonable.
 */

-#include "qemu/osdep.h"
 #include "qemu-common.h"
 #include "block/block_int.h"
 #include "qemu/module.h"
 #include "migration/migration.h"
-#include "qemu/coroutine.h"
+#include "block/coroutine.h"

 #if defined(CONFIG_UUID)
 #include <uuid/uuid.h>
@@ -400,7 +399,7 @@ static int vdi_open(BlockDriverState *bs, QDict *options, int flags,

    logout("\n");

-    ret = bdrv_read(bs->file->bs, 0, (uint8_t *)&header, 1);
+    ret = bdrv_read(bs->file, 0, (uint8_t *)&header, 1);
    if (ret < 0) {
        goto fail;
    }
@@ -491,22 +490,21 @@ static int vdi_open(BlockDriverState *bs, QDict *options, int flags,

    bmap_size = header.blocks_in_image * sizeof(uint32_t);
    bmap_size = DIV_ROUND_UP(bmap_size, SECTOR_SIZE);
-    s->bmap = qemu_try_blockalign(bs->file->bs, bmap_size * SECTOR_SIZE);
+    s->bmap = qemu_try_blockalign(bs->file, bmap_size * SECTOR_SIZE);
    if (s->bmap == NULL) {
        ret = -ENOMEM;
        goto fail;
    }

-    ret = bdrv_read(bs->file->bs, s->bmap_sector, (uint8_t *)s->bmap,
-                    bmap_size);
+    ret = bdrv_read(bs->file, s->bmap_sector, (uint8_t *)s->bmap, bmap_size);
    if (ret < 0) {
        goto fail_free_bmap;
    }

    /* Disable migration when vdi images are used */
-    error_setg(&s->migration_blocker, "The vdi format used by node '%s' "
-               "does not support live migration",
-               bdrv_get_device_or_node_name(bs));
+    error_set(&s->migration_blocker,
+              QERR_BLOCK_FORMAT_FEATURE_NOT_SUPPORTED,
+              "vdi", bdrv_get_device_name(bs), "live migration");
    migrate_add_blocker(s->migration_blocker);

    qemu_co_mutex_init(&s->write_lock);
@@ -527,7 +525,7 @@ static int vdi_reopen_prepare(BDRVReopenState *state,
 }

 static int64_t coroutine_fn vdi_co_get_block_status(BlockDriverState *bs,
-        int64_t sector_num, int nb_sectors, int *pnum, BlockDriverState **file)
+        int64_t sector_num, int nb_sectors, int *pnum)
 {
    /* TODO: Check for too large sector_num (in bdrv_is_allocated or here). */
    BDRVVdiState *s = (BDRVVdiState *)bs->opaque;
@@ -551,7 +549,6 @@ static int64_t coroutine_fn vdi_co_get_block_status(BlockDriverState *bs,
    offset = s->header.offset_data +
                              (uint64_t)bmap_entry * s->block_size +
                              sector_in_block * SECTOR_SIZE;
-    *file = bs->file->bs;
    return BDRV_BLOCK_DATA | BDRV_BLOCK_OFFSET_VALID | offset;
 }

@@ -588,7 +585,7 @@ static int vdi_co_read(BlockDriverState *bs,
            uint64_t offset = s->header.offset_data / SECTOR_SIZE +
                              (uint64_t)bmap_entry * s->block_sectors +
                              sector_in_block;
-            ret = bdrv_read(bs->file->bs, offset, buf, n_sectors);
+            ret = bdrv_read(bs->file, offset, buf, n_sectors);
        }
        logout("%u sectors read\n", n_sectors);

@@ -656,7 +653,7 @@ static int vdi_co_write(BlockDriverState *bs,
             * acquire the lock and thus the padded cluster is written before
             * the other coroutines can write to the affected area. */
            qemu_co_mutex_lock(&s->write_lock);
-            ret = bdrv_write(bs->file->bs, offset, block, s->block_sectors);
+            ret = bdrv_write(bs->file, offset, block, s->block_sectors);
            qemu_co_mutex_unlock(&s->write_lock);
        } else {
            uint64_t offset = s->header.offset_data / SECTOR_SIZE +
@@ -672,7 +669,7 @@ static int vdi_co_write(BlockDriverState *bs,
             * that that write operation has returned (there may be other writes
             * in flight, but they do not concern this very operation). */
            qemu_co_mutex_unlock(&s->write_lock);
-            ret = bdrv_write(bs->file->bs, offset, buf, n_sectors);
+            ret = bdrv_write(bs->file, offset, buf, n_sectors);
        }

        nb_sectors -= n_sectors;
@@ -697,7 +694,7 @@ static int vdi_co_write(BlockDriverState *bs,
        assert(VDI_IS_ALLOCATED(bmap_first));
        *header = s->header;
        vdi_header_to_le(header);
-        ret = bdrv_write(bs->file->bs, 0, block, 1);
+        ret = bdrv_write(bs->file, 0, block, 1);
        g_free(block);
        block = NULL;

@@ -715,7 +712,7 @@ static int vdi_co_write(BlockDriverState *bs,
        base = ((uint8_t *)&s->bmap[0]) + bmap_first * SECTOR_SIZE;
        logout("will write %u block map sectors starting from entry %u\n",
               n_sectors, bmap_first);
-        ret = bdrv_write(bs->file->bs, offset, base, n_sectors);
+        ret = bdrv_write(bs->file, offset, base, n_sectors);
    }

    return ret;
@@ -767,7 +764,7 @@ static int vdi_create(const char *filename, QemuOpts *opts, Error **errp)
        goto exit;
    }
    ret = bdrv_open(&bs, filename, NULL, NULL, BDRV_O_RDWR | BDRV_O_PROTOCOL,
-                    &local_err);
+                    NULL, &local_err);
    if (ret < 0) {
        error_propagate(errp, local_err);
        goto exit;
--- a/Show More
+++ b/Show More
@@ -1 +1 @@
 .5.50
 .3.1