summary refs log tree commit diff stats
diff options
context:
space:
mode:
-rw-r--r--.gitignore2
-rw-r--r--.gitlab-ci.d/buildtest.yml29
-rw-r--r--.gitlab-ci.d/cirrus.yml15
-rw-r--r--.gitlab-ci.d/opensbi.yml28
-rw-r--r--.mailmap4
-rw-r--r--MAINTAINERS6
-rw-r--r--VERSION2
-rw-r--r--accel/tcg/atomic_common.c.inc107
-rw-r--r--accel/tcg/atomic_template.h141
-rw-r--r--accel/tcg/cpu-exec.c207
-rw-r--r--accel/tcg/cputlb.c49
-rw-r--r--accel/tcg/tcg-runtime.h46
-rw-r--r--accel/tcg/translate-all.c7
-rw-r--r--accel/tcg/translator.c39
-rw-r--r--accel/tcg/user-exec.c41
-rw-r--r--block.c3
-rw-r--r--block/blkdebug.c136
-rw-r--r--block/dirty-bitmap.c2
-rw-r--r--block/export/export.c5
-rw-r--r--block/linux-aio.c9
-rw-r--r--block/mirror.c13
-rw-r--r--block/replication.c111
-rw-r--r--block/vvfat.c43
-rw-r--r--bsd-user/syscall.c6
-rw-r--r--chardev/spice.c1
-rw-r--r--configs/devices/mips64el-softmmu/default.mak4
-rw-r--r--configs/devices/ppc-softmmu/default.mak1
-rw-r--r--configs/devices/tricore-softmmu/default.mak1
-rwxr-xr-xconfigure51
-rw-r--r--contrib/gitdm/aliases6
-rw-r--r--contrib/gitdm/domain-map7
-rw-r--r--contrib/gitdm/group-map-academics3
-rw-r--r--contrib/gitdm/group-map-individuals5
-rw-r--r--contrib/gitdm/group-map-interns13
-rw-r--r--contrib/gitdm/group-map-netflix5
-rw-r--r--contrib/gitdm/group-map-robots7
-rw-r--r--contrib/plugins/cache.c11
-rw-r--r--cpu.c34
-rw-r--r--docs/interop/vhost-user.rst2
-rw-r--r--docs/system/device-emulation.rst90
-rw-r--r--docs/system/devices/ivshmem.rst (renamed from docs/system/ivshmem.rst)0
-rw-r--r--docs/system/devices/net.rst (renamed from docs/system/net.rst)0
-rw-r--r--docs/system/devices/nvme.rst (renamed from docs/system/nvme.rst)0
-rw-r--r--docs/system/devices/usb.rst (renamed from docs/system/usb.rst)0
-rw-r--r--docs/system/devices/vhost-user.rst59
-rw-r--r--docs/system/devices/virtio-pmem.rst (renamed from docs/system/virtio-pmem.rst)0
-rw-r--r--docs/system/index.rst6
-rw-r--r--docs/tools/qemu-img.rst8
-rw-r--r--gitdm.config13
-rw-r--r--hw/acpi/Kconfig4
-rw-r--r--hw/acpi/meson.build3
-rw-r--r--hw/arm/Kconfig3
-rw-r--r--hw/char/Kconfig1
-rw-r--r--hw/display/qxl.c35
-rw-r--r--hw/display/qxl.h1
-rw-r--r--hw/display/virtio-gpu-gl.c27
-rw-r--r--hw/display/virtio-gpu-virgl.c8
-rw-r--r--hw/display/virtio-gpu.c26
-rw-r--r--hw/i386/Kconfig1
-rw-r--r--hw/ide/Kconfig2
-rw-r--r--hw/isa/Kconfig5
-rw-r--r--hw/mips/Kconfig6
-rw-r--r--hw/pci-host/Kconfig1
-rw-r--r--hw/ppc/Kconfig2
-rw-r--r--hw/remote/memory.c2
-rw-r--r--hw/riscv/Kconfig5
-rw-r--r--hw/riscv/meson.build2
-rw-r--r--hw/tricore/Kconfig3
-rw-r--r--hw/tricore/meson.build4
-rw-r--r--hw/usb/meson.build2
-rw-r--r--include/block/aio.h12
-rw-r--r--include/exec/exec-all.h24
-rw-r--r--include/exec/translator.h11
-rw-r--r--include/hw/core/cpu.h4
-rw-r--r--include/hw/core/tcg-cpu-ops.h6
-rw-r--r--include/hw/virtio/virtio-gpu.h1
-rw-r--r--include/qapi/forward-visitor.h27
-rw-r--r--include/qemu/atomic.h247
-rw-r--r--include/qemu/plugin.h12
-rw-r--r--include/qemu/qemu-plugin.h13
-rw-r--r--include/qemu/stats64.h2
-rw-r--r--include/sysemu/iothread.h3
-rw-r--r--include/tcg/tcg.h80
-rw-r--r--iothread.c82
-rw-r--r--linux-user/exit.c2
-rw-r--r--linux-user/hppa/cpu_loop.c2
-rw-r--r--meson.build11
-rw-r--r--monitor/hmp-cmds.c2
-rw-r--r--plugins/api.c2
-rw-r--r--plugins/core.c41
-rw-r--r--qapi/meson.build1
-rw-r--r--qapi/misc.json6
-rw-r--r--qapi/qapi-forward-visitor.c326
-rw-r--r--qapi/qom.json7
-rw-r--r--qemu-img.c50
-rw-r--r--qemu-options.hx8
-rw-r--r--qom/object.c9
-rw-r--r--softmmu/timers-state.h2
-rw-r--r--softmmu/vl.c2
-rw-r--r--target/alpha/translate.c31
-rw-r--r--target/arm/cpu.c1
-rw-r--r--target/arm/cpu_tcg.c1
-rw-r--r--target/arm/debug_helper.c12
-rw-r--r--target/arm/helper-a64.c8
-rw-r--r--target/arm/helper.h2
-rw-r--r--target/arm/internals.h3
-rw-r--r--target/arm/translate-a64.c25
-rw-r--r--target/arm/translate.c29
-rw-r--r--target/avr/cpu.c1
-rw-r--r--target/avr/cpu.h1
-rw-r--r--target/avr/gdbstub.c13
-rw-r--r--target/avr/translate.c32
-rw-r--r--target/cris/translate.c20
-rw-r--r--target/hexagon/translate.c17
-rw-r--r--target/hppa/translate.c11
-rw-r--r--target/i386/cpu.c19
-rw-r--r--target/i386/cpu.h44
-rw-r--r--target/i386/host-cpu.c13
-rw-r--r--target/i386/kvm/kvm-cpu.c105
-rw-r--r--target/i386/tcg/mem_helper.c15
-rw-r--r--target/i386/tcg/sysemu/misc_helper.c10
-rw-r--r--target/i386/tcg/sysemu/svm_helper.c70
-rw-r--r--target/i386/tcg/tcg-cpu.c23
-rw-r--r--target/i386/tcg/translate.c28
-rw-r--r--target/m68k/op_helper.c19
-rw-r--r--target/m68k/translate.c18
-rw-r--r--target/microblaze/translate.c18
-rw-r--r--target/mips/tcg/translate.c19
-rw-r--r--target/nios2/translate.c27
-rw-r--r--target/openrisc/translate.c17
-rw-r--r--target/ppc/mem_helper.c16
-rw-r--r--target/ppc/translate.c18
-rw-r--r--target/riscv/translate.c17
-rw-r--r--target/rx/translate.c14
-rw-r--r--target/s390x/tcg/mem_helper.c19
-rw-r--r--target/s390x/tcg/translate.c24
-rw-r--r--target/sh4/translate.c18
-rw-r--r--target/sparc/translate.c17
-rw-r--r--target/tricore/translate.c16
-rw-r--r--target/xtensa/translate.c17
-rw-r--r--tcg/tcg-op.c79
-rw-r--r--tests/acceptance/virtio-gpu.py42
-rw-r--r--tests/docker/dockerfiles/debian-tricore-cross.docker34
-rwxr-xr-xtests/qemu-iotests/15154
-rw-r--r--tests/qemu-iotests/151.out4
-rwxr-xr-xtests/qemu-iotests/30715
-rw-r--r--tests/qemu-iotests/307.out8
-rwxr-xr-xtests/qemu-iotests/tests/qemu-img-bitmaps (renamed from tests/qemu-iotests/291)34
-rw-r--r--tests/qemu-iotests/tests/qemu-img-bitmaps.out (renamed from tests/qemu-iotests/291.out)67
-rwxr-xr-xtests/tcg/configure.sh18
-rw-r--r--tests/unit/meson.build1
-rw-r--r--tests/unit/test-forward-visitor.c197
-rw-r--r--trace/mem-internal.h50
-rw-r--r--trace/mem.h46
-rw-r--r--util/aio-posix.c12
-rw-r--r--util/aio-win32.c5
-rw-r--r--util/async.c2
-rw-r--r--util/qemu-config.c16
-rw-r--r--util/qsp.c4
159 files changed, 2404 insertions, 1628 deletions
diff --git a/.gitignore b/.gitignore
index 75a4be0724..eb2553026c 100644
--- a/.gitignore
+++ b/.gitignore
@@ -13,3 +13,5 @@ GTAGS
 *~
 *.ast_raw
 *.depend_raw
+*.swp
+*.patch
diff --git a/.gitlab-ci.d/buildtest.yml b/.gitlab-ci.d/buildtest.yml
index 89df51517c..ee0c83b600 100644
--- a/.gitlab-ci.d/buildtest.yml
+++ b/.gitlab-ci.d/buildtest.yml
@@ -354,6 +354,17 @@ build-some-softmmu:
     TARGETS: xtensa-softmmu arm-softmmu aarch64-softmmu alpha-softmmu
     MAKE_CHECK_ARGS: check-tcg
 
+# We build tricore in a very minimal tricore only container
+build-tricore-softmmu:
+  extends: .native_build_job_template
+  needs:
+    job: tricore-debian-cross-container
+  variables:
+    IMAGE: debian-tricore-cross
+    CONFIGURE_ARGS: --disable-tools --disable-fdt --enable-debug
+    TARGETS: tricore-softmmu
+    MAKE_CHECK_ARGS: check-tcg
+
 clang-system:
   extends: .native_build_job_template
   needs:
@@ -663,6 +674,17 @@ build-tools-and-docs-debian:
 
 # Prepare for GitLab pages deployment. Anything copied into the
 # "public" directory will be deployed to $USER.gitlab.io/$PROJECT
+#
+# GitLab publishes from any branch that triggers a CI pipeline
+#
+# For the main repo we don't want to publish from 'staging'
+# since that content may not be pushed, nor do we wish to
+# publish from 'stable-NNN' branches as that content is outdated.
+# Thus we restrict to just the default branch
+#
+# For contributor forks we want to publish from any repo so
+# that users can see the results of their commits, regardless
+# of what topic branch they're currently using
 pages:
   image: $CI_REGISTRY_IMAGE/qemu/debian-amd64:latest
   stage: test
@@ -681,3 +703,10 @@ pages:
   artifacts:
     paths:
       - public
+  rules:
+    - if: '$CI_PROJECT_NAMESPACE == "qemu-project" && $CI_COMMIT_BRANCH == $CI_DEFAULT_BRANCH'
+      when: on_success
+    - if: '$CI_PROJECT_NAMESPACE == "qemu-project"'
+      when: never
+    - if: '$CI_PROJECT_NAMESPACE != "qemu-project"'
+      when: on_success
diff --git a/.gitlab-ci.d/cirrus.yml b/.gitlab-ci.d/cirrus.yml
index 60b13ed83f..675db69622 100644
--- a/.gitlab-ci.d/cirrus.yml
+++ b/.gitlab-ci.d/cirrus.yml
@@ -85,18 +85,3 @@ x64-macos-11-base-build:
     PATH_EXTRA: /usr/local/opt/ccache/libexec:/usr/local/opt/gettext/bin
     PKG_CONFIG_PATH: /usr/local/opt/curl/lib/pkgconfig:/usr/local/opt/ncurses/lib/pkgconfig:/usr/local/opt/readline/lib/pkgconfig
     TEST_TARGETS: check-unit check-block check-qapi-schema check-softfloat check-qtest-x86_64
-
-x64-macos-11-xcode-build:
-  extends: .cirrus_build_job
-  variables:
-    NAME: macos-11
-    CIRRUS_VM_INSTANCE_TYPE: osx_instance
-    CIRRUS_VM_IMAGE_SELECTOR: image
-    CIRRUS_VM_IMAGE_NAME: big-sur-xcode
-    CIRRUS_VM_CPUS: 12
-    CIRRUS_VM_RAM: 24G
-    UPDATE_COMMAND: brew update
-    INSTALL_COMMAND: brew install
-    PATH_EXTRA: /usr/local/opt/ccache/libexec:/usr/local/opt/gettext/bin
-    PKG_CONFIG_PATH: /usr/local/opt/curl/lib/pkgconfig:/usr/local/opt/ncurses/lib/pkgconfig:/usr/local/opt/readline/lib/pkgconfig
-    TEST_TARGETS: check-unit check-block check-qapi-schema check-softfloat check-qtest-x86_64
diff --git a/.gitlab-ci.d/opensbi.yml b/.gitlab-ci.d/opensbi.yml
index f66cd1d908..d8a0456679 100644
--- a/.gitlab-ci.d/opensbi.yml
+++ b/.gitlab-ci.d/opensbi.yml
@@ -1,10 +1,23 @@
-docker-opensbi:
- stage: containers
- rules: # Only run this job when the Dockerfile is modified
+# All jobs needing docker-opensbi must use the same rules it uses.
+.opensbi_job_rules:
+ rules: # Only run this job when ...
  - changes:
+   # this file is modified
    - .gitlab-ci.d/opensbi.yml
+   # or the Dockerfile is modified
    - .gitlab-ci.d/opensbi/Dockerfile
    when: always
+ - changes: # or roms/opensbi/ is modified (submodule updated)
+   - roms/opensbi/*
+   when: always
+ - if: '$CI_COMMIT_REF_NAME =~ /^opensbi/' # or the branch/tag starts with 'opensbi'
+   when: always
+ - if: '$CI_COMMIT_MESSAGE =~ /opensbi/i' # or last commit description contains 'OpenSBI'
+   when: always
+
+docker-opensbi:
+ extends: .opensbi_job_rules
+ stage: containers
  image: docker:19.03.1
  services:
  - docker:19.03.1-dind
@@ -24,16 +37,9 @@ docker-opensbi:
  - docker push $IMAGE_TAG
 
 build-opensbi:
+ extends: .opensbi_job_rules
  stage: build
  needs: ['docker-opensbi']
- rules: # Only run this job when ...
- - changes: # ... roms/opensbi/ is modified (submodule updated)
-   - roms/opensbi/*
-   when: always
- - if: '$CI_COMMIT_REF_NAME =~ /^opensbi/' # or the branch/tag starts with 'opensbi'
-   when: always
- - if: '$CI_COMMIT_MESSAGE =~ /opensbi/i' # or last commit description contains 'OpenSBI'
-   when: always
  artifacts:
    paths: # 'artifacts.zip' will contains the following files:
    - pc-bios/opensbi-riscv32-generic-fw_dynamic.bin
diff --git a/.mailmap b/.mailmap
index a1bd659817..082ff893ab 100644
--- a/.mailmap
+++ b/.mailmap
@@ -27,6 +27,10 @@ Paul Brook <paul@codesourcery.com> pbrook <pbrook@c046a42c-6fe2-441c-8c8c-714662
 Thiemo Seufer <ths@networkno.de> ths <ths@c046a42c-6fe2-441c-8c8c-71466251a162>
 malc <av1474@comtv.ru> malc <malc@c046a42c-6fe2-441c-8c8c-71466251a162>
 
+# Corrupted Author fields
+Marek Dolata <mkdolata@us.ibm.com> mkdolata@us.ibm.com <mkdolata@us.ibm.com>
+Nick Hudson <hnick@vmware.com> hnick@vmware.com <hnick@vmware.com>
+
 # There is also a:
 #    (no author) <(no author)@c046a42c-6fe2-441c-8c8c-71466251a162>
 # for the cvs2svn initialization commit e63c3dc74bf.
diff --git a/MAINTAINERS b/MAINTAINERS
index 4256ad1adb..445f7fe2d1 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -2452,15 +2452,19 @@ F: tests/tcg/multiarch/gdbstub/
 
 Memory API
 M: Paolo Bonzini <pbonzini@redhat.com>
+M: Peter Xu <peterx@redhat.com>
+M: David Hildenbrand <david@redhat.com>
 S: Supported
 F: include/exec/ioport.h
 F: include/exec/memop.h
 F: include/exec/memory.h
 F: include/exec/ram_addr.h
 F: include/exec/ramblock.h
+F: include/sysemu/memory_mapping.h
 F: softmmu/dma-helpers.c
 F: softmmu/ioport.c
 F: softmmu/memory.c
+F: softmmu/memory_mapping.c
 F: softmmu/physmem.c
 F: include/exec/memory-internal.h
 F: scripts/coccinelle/memory-region-housekeeping.cocci
@@ -2547,7 +2551,7 @@ S: Maintained
 F: net/netmap.c
 
 Host Memory Backends
-M: Eduardo Habkost <ehabkost@redhat.com>
+M: David Hildenbrand <david@redhat.com>
 M: Igor Mammedov <imammedo@redhat.com>
 S: Maintained
 F: backends/hostmem*.c
diff --git a/VERSION b/VERSION
index cc94f6b803..7d5e1d8670 100644
--- a/VERSION
+++ b/VERSION
@@ -1 +1 @@
-6.0.50
+6.0.90
diff --git a/accel/tcg/atomic_common.c.inc b/accel/tcg/atomic_common.c.inc
index 344525b0bb..6c0339f610 100644
--- a/accel/tcg/atomic_common.c.inc
+++ b/accel/tcg/atomic_common.c.inc
@@ -13,42 +13,125 @@
  * See the COPYING file in the top-level directory.
  */
 
-static inline
-void atomic_trace_rmw_pre(CPUArchState *env, target_ulong addr, uint16_t info)
+static uint16_t atomic_trace_rmw_pre(CPUArchState *env, target_ulong addr,
+                                     TCGMemOpIdx oi)
 {
     CPUState *cpu = env_cpu(env);
+    uint16_t info = trace_mem_get_info(get_memop(oi), get_mmuidx(oi), false);
 
     trace_guest_mem_before_exec(cpu, addr, info);
     trace_guest_mem_before_exec(cpu, addr, info | TRACE_MEM_ST);
+
+    return info;
 }
 
-static inline void
-atomic_trace_rmw_post(CPUArchState *env, target_ulong addr, uint16_t info)
+static void atomic_trace_rmw_post(CPUArchState *env, target_ulong addr,
+                                  uint16_t info)
 {
     qemu_plugin_vcpu_mem_cb(env_cpu(env), addr, info);
     qemu_plugin_vcpu_mem_cb(env_cpu(env), addr, info | TRACE_MEM_ST);
 }
 
-static inline
-void atomic_trace_ld_pre(CPUArchState *env, target_ulong addr, uint16_t info)
+#if HAVE_ATOMIC128
+static uint16_t atomic_trace_ld_pre(CPUArchState *env, target_ulong addr,
+                                    TCGMemOpIdx oi)
 {
+    uint16_t info = trace_mem_get_info(get_memop(oi), get_mmuidx(oi), false);
+
     trace_guest_mem_before_exec(env_cpu(env), addr, info);
+
+    return info;
 }
 
-static inline
-void atomic_trace_ld_post(CPUArchState *env, target_ulong addr, uint16_t info)
+static void atomic_trace_ld_post(CPUArchState *env, target_ulong addr,
+                                 uint16_t info)
 {
     qemu_plugin_vcpu_mem_cb(env_cpu(env), addr, info);
 }
 
-static inline
-void atomic_trace_st_pre(CPUArchState *env, target_ulong addr, uint16_t info)
+static uint16_t atomic_trace_st_pre(CPUArchState *env, target_ulong addr,
+                                    TCGMemOpIdx oi)
 {
+    uint16_t info = trace_mem_get_info(get_memop(oi), get_mmuidx(oi), true);
+
     trace_guest_mem_before_exec(env_cpu(env), addr, info);
+
+    return info;
 }
 
-static inline
-void atomic_trace_st_post(CPUArchState *env, target_ulong addr, uint16_t info)
+static void atomic_trace_st_post(CPUArchState *env, target_ulong addr,
+                                 uint16_t info)
 {
     qemu_plugin_vcpu_mem_cb(env_cpu(env), addr, info);
 }
+#endif
+
+/*
+ * Atomic helpers callable from TCG.
+ * These have a common interface and all defer to cpu_atomic_*
+ * using the host return address from GETPC().
+ */
+
+#define CMPXCHG_HELPER(OP, TYPE) \
+    TYPE HELPER(atomic_##OP)(CPUArchState *env, target_ulong addr,  \
+                             TYPE oldv, TYPE newv, uint32_t oi)     \
+    { return cpu_atomic_##OP##_mmu(env, addr, oldv, newv, oi, GETPC()); }
+
+CMPXCHG_HELPER(cmpxchgb, uint32_t)
+CMPXCHG_HELPER(cmpxchgw_be, uint32_t)
+CMPXCHG_HELPER(cmpxchgw_le, uint32_t)
+CMPXCHG_HELPER(cmpxchgl_be, uint32_t)
+CMPXCHG_HELPER(cmpxchgl_le, uint32_t)
+
+#ifdef CONFIG_ATOMIC64
+CMPXCHG_HELPER(cmpxchgq_be, uint64_t)
+CMPXCHG_HELPER(cmpxchgq_le, uint64_t)
+#endif
+
+#undef CMPXCHG_HELPER
+
+#define ATOMIC_HELPER(OP, TYPE) \
+    TYPE HELPER(glue(atomic_,OP))(CPUArchState *env, target_ulong addr,  \
+                                  TYPE val, uint32_t oi)                 \
+    { return glue(glue(cpu_atomic_,OP),_mmu)(env, addr, val, oi, GETPC()); }
+
+#ifdef CONFIG_ATOMIC64
+#define GEN_ATOMIC_HELPERS(OP)              \
+    ATOMIC_HELPER(glue(OP,b), uint32_t)     \
+    ATOMIC_HELPER(glue(OP,w_be), uint32_t)  \
+    ATOMIC_HELPER(glue(OP,w_le), uint32_t)  \
+    ATOMIC_HELPER(glue(OP,l_be), uint32_t)  \
+    ATOMIC_HELPER(glue(OP,l_le), uint32_t)  \
+    ATOMIC_HELPER(glue(OP,q_be), uint64_t)  \
+    ATOMIC_HELPER(glue(OP,q_le), uint64_t)
+#else
+#define GEN_ATOMIC_HELPERS(OP)              \
+    ATOMIC_HELPER(glue(OP,b), uint32_t)     \
+    ATOMIC_HELPER(glue(OP,w_be), uint32_t)  \
+    ATOMIC_HELPER(glue(OP,w_le), uint32_t)  \
+    ATOMIC_HELPER(glue(OP,l_be), uint32_t)  \
+    ATOMIC_HELPER(glue(OP,l_le), uint32_t)
+#endif
+
+GEN_ATOMIC_HELPERS(fetch_add)
+GEN_ATOMIC_HELPERS(fetch_and)
+GEN_ATOMIC_HELPERS(fetch_or)
+GEN_ATOMIC_HELPERS(fetch_xor)
+GEN_ATOMIC_HELPERS(fetch_smin)
+GEN_ATOMIC_HELPERS(fetch_umin)
+GEN_ATOMIC_HELPERS(fetch_smax)
+GEN_ATOMIC_HELPERS(fetch_umax)
+
+GEN_ATOMIC_HELPERS(add_fetch)
+GEN_ATOMIC_HELPERS(and_fetch)
+GEN_ATOMIC_HELPERS(or_fetch)
+GEN_ATOMIC_HELPERS(xor_fetch)
+GEN_ATOMIC_HELPERS(smin_fetch)
+GEN_ATOMIC_HELPERS(umin_fetch)
+GEN_ATOMIC_HELPERS(smax_fetch)
+GEN_ATOMIC_HELPERS(umax_fetch)
+
+GEN_ATOMIC_HELPERS(xchg)
+
+#undef ATOMIC_HELPER
+#undef GEN_ATOMIC_HELPERS
diff --git a/accel/tcg/atomic_template.h b/accel/tcg/atomic_template.h
index afa8a9daf3..d89af4cc1e 100644
--- a/accel/tcg/atomic_template.h
+++ b/accel/tcg/atomic_template.h
@@ -28,8 +28,8 @@
 # define SHIFT      4
 #elif DATA_SIZE == 8
 # define SUFFIX     q
-# define DATA_TYPE  uint64_t
-# define SDATA_TYPE int64_t
+# define DATA_TYPE  aligned_uint64_t
+# define SDATA_TYPE aligned_int64_t
 # define BSWAP      bswap64
 # define SHIFT      3
 #elif DATA_SIZE == 4
@@ -71,15 +71,14 @@
 #endif
 
 ABI_TYPE ATOMIC_NAME(cmpxchg)(CPUArchState *env, target_ulong addr,
-                              ABI_TYPE cmpv, ABI_TYPE newv EXTRA_ARGS)
+                              ABI_TYPE cmpv, ABI_TYPE newv,
+                              TCGMemOpIdx oi, uintptr_t retaddr)
 {
-    ATOMIC_MMU_DECLS;
-    DATA_TYPE *haddr = ATOMIC_MMU_LOOKUP_RW;
+    DATA_TYPE *haddr = atomic_mmu_lookup(env, addr, oi, DATA_SIZE,
+                                         PAGE_READ | PAGE_WRITE, retaddr);
     DATA_TYPE ret;
-    uint16_t info = trace_mem_build_info(SHIFT, false, 0, false,
-                                         ATOMIC_MMU_IDX);
+    uint16_t info = atomic_trace_rmw_pre(env, addr, oi);
 
-    atomic_trace_rmw_pre(env, addr, info);
 #if DATA_SIZE == 16
     ret = atomic16_cmpxchg(haddr, cmpv, newv);
 #else
@@ -92,45 +91,41 @@ ABI_TYPE ATOMIC_NAME(cmpxchg)(CPUArchState *env, target_ulong addr,
 
 #if DATA_SIZE >= 16
 #if HAVE_ATOMIC128
-ABI_TYPE ATOMIC_NAME(ld)(CPUArchState *env, target_ulong addr EXTRA_ARGS)
+ABI_TYPE ATOMIC_NAME(ld)(CPUArchState *env, target_ulong addr,
+                         TCGMemOpIdx oi, uintptr_t retaddr)
 {
-    ATOMIC_MMU_DECLS;
-    DATA_TYPE val, *haddr = ATOMIC_MMU_LOOKUP_R;
-    uint16_t info = trace_mem_build_info(SHIFT, false, 0, false,
-                                         ATOMIC_MMU_IDX);
+    DATA_TYPE *haddr = atomic_mmu_lookup(env, addr, oi, DATA_SIZE,
+                                         PAGE_READ, retaddr);
+    DATA_TYPE val;
+    uint16_t info = atomic_trace_ld_pre(env, addr, oi);
 
-    atomic_trace_ld_pre(env, addr, info);
     val = atomic16_read(haddr);
     ATOMIC_MMU_CLEANUP;
     atomic_trace_ld_post(env, addr, info);
     return val;
 }
 
-void ATOMIC_NAME(st)(CPUArchState *env, target_ulong addr,
-                     ABI_TYPE val EXTRA_ARGS)
+void ATOMIC_NAME(st)(CPUArchState *env, target_ulong addr, ABI_TYPE val,
+                     TCGMemOpIdx oi, uintptr_t retaddr)
 {
-    ATOMIC_MMU_DECLS;
-    DATA_TYPE *haddr = ATOMIC_MMU_LOOKUP_W;
-    uint16_t info = trace_mem_build_info(SHIFT, false, 0, true,
-                                         ATOMIC_MMU_IDX);
+    DATA_TYPE *haddr = atomic_mmu_lookup(env, addr, oi, DATA_SIZE,
+                                         PAGE_WRITE, retaddr);
+    uint16_t info = atomic_trace_st_pre(env, addr, oi);
 
-    atomic_trace_st_pre(env, addr, info);
     atomic16_set(haddr, val);
     ATOMIC_MMU_CLEANUP;
     atomic_trace_st_post(env, addr, info);
 }
 #endif
 #else
-ABI_TYPE ATOMIC_NAME(xchg)(CPUArchState *env, target_ulong addr,
-                           ABI_TYPE val EXTRA_ARGS)
+ABI_TYPE ATOMIC_NAME(xchg)(CPUArchState *env, target_ulong addr, ABI_TYPE val,
+                           TCGMemOpIdx oi, uintptr_t retaddr)
 {
-    ATOMIC_MMU_DECLS;
-    DATA_TYPE *haddr = ATOMIC_MMU_LOOKUP_RW;
+    DATA_TYPE *haddr = atomic_mmu_lookup(env, addr, oi, DATA_SIZE,
+                                         PAGE_READ | PAGE_WRITE, retaddr);
     DATA_TYPE ret;
-    uint16_t info = trace_mem_build_info(SHIFT, false, 0, false,
-                                         ATOMIC_MMU_IDX);
+    uint16_t info = atomic_trace_rmw_pre(env, addr, oi);
 
-    atomic_trace_rmw_pre(env, addr, info);
     ret = qatomic_xchg__nocheck(haddr, val);
     ATOMIC_MMU_CLEANUP;
     atomic_trace_rmw_post(env, addr, info);
@@ -139,14 +134,12 @@ ABI_TYPE ATOMIC_NAME(xchg)(CPUArchState *env, target_ulong addr,
 
 #define GEN_ATOMIC_HELPER(X)                                        \
 ABI_TYPE ATOMIC_NAME(X)(CPUArchState *env, target_ulong addr,       \
-                        ABI_TYPE val EXTRA_ARGS)                    \
+                        ABI_TYPE val, TCGMemOpIdx oi, uintptr_t retaddr) \
 {                                                                   \
-    ATOMIC_MMU_DECLS;                                               \
-    DATA_TYPE *haddr = ATOMIC_MMU_LOOKUP_RW;                        \
+    DATA_TYPE *haddr = atomic_mmu_lookup(env, addr, oi, DATA_SIZE,  \
+                                         PAGE_READ | PAGE_WRITE, retaddr); \
     DATA_TYPE ret;                                                  \
-    uint16_t info = trace_mem_build_info(SHIFT, false, 0, false,    \
-                                         ATOMIC_MMU_IDX);           \
-    atomic_trace_rmw_pre(env, addr, info);                          \
+    uint16_t info = atomic_trace_rmw_pre(env, addr, oi);            \
     ret = qatomic_##X(haddr, val);                                  \
     ATOMIC_MMU_CLEANUP;                                             \
     atomic_trace_rmw_post(env, addr, info);                         \
@@ -164,7 +157,8 @@ GEN_ATOMIC_HELPER(xor_fetch)
 
 #undef GEN_ATOMIC_HELPER
 
-/* These helpers are, as a whole, full barriers.  Within the helper,
+/*
+ * These helpers are, as a whole, full barriers.  Within the helper,
  * the leading barrier is explicit and the trailing barrier is within
  * cmpxchg primitive.
  *
@@ -173,14 +167,12 @@ GEN_ATOMIC_HELPER(xor_fetch)
  */
 #define GEN_ATOMIC_HELPER_FN(X, FN, XDATA_TYPE, RET)                \
 ABI_TYPE ATOMIC_NAME(X)(CPUArchState *env, target_ulong addr,       \
-                        ABI_TYPE xval EXTRA_ARGS)                   \
+                        ABI_TYPE xval, TCGMemOpIdx oi, uintptr_t retaddr) \
 {                                                                   \
-    ATOMIC_MMU_DECLS;                                               \
-    XDATA_TYPE *haddr = ATOMIC_MMU_LOOKUP_RW;                       \
+    XDATA_TYPE *haddr = atomic_mmu_lookup(env, addr, oi, DATA_SIZE, \
+                                          PAGE_READ | PAGE_WRITE, retaddr); \
     XDATA_TYPE cmp, old, new, val = xval;                           \
-    uint16_t info = trace_mem_build_info(SHIFT, false, 0, false,    \
-                                         ATOMIC_MMU_IDX);           \
-    atomic_trace_rmw_pre(env, addr, info);                          \
+    uint16_t info = atomic_trace_rmw_pre(env, addr, oi);            \
     smp_mb();                                                       \
     cmp = qatomic_read__nocheck(haddr);                             \
     do {                                                            \
@@ -218,15 +210,14 @@ GEN_ATOMIC_HELPER_FN(umax_fetch, MAX,  DATA_TYPE, new)
 #endif
 
 ABI_TYPE ATOMIC_NAME(cmpxchg)(CPUArchState *env, target_ulong addr,
-                              ABI_TYPE cmpv, ABI_TYPE newv EXTRA_ARGS)
+                              ABI_TYPE cmpv, ABI_TYPE newv,
+                              TCGMemOpIdx oi, uintptr_t retaddr)
 {
-    ATOMIC_MMU_DECLS;
-    DATA_TYPE *haddr = ATOMIC_MMU_LOOKUP_RW;
+    DATA_TYPE *haddr = atomic_mmu_lookup(env, addr, oi, DATA_SIZE,
+                                         PAGE_READ | PAGE_WRITE, retaddr);
     DATA_TYPE ret;
-    uint16_t info = trace_mem_build_info(SHIFT, false, MO_BSWAP, false,
-                                         ATOMIC_MMU_IDX);
+    uint16_t info = atomic_trace_rmw_pre(env, addr, oi);
 
-    atomic_trace_rmw_pre(env, addr, info);
 #if DATA_SIZE == 16
     ret = atomic16_cmpxchg(haddr, BSWAP(cmpv), BSWAP(newv));
 #else
@@ -239,30 +230,28 @@ ABI_TYPE ATOMIC_NAME(cmpxchg)(CPUArchState *env, target_ulong addr,
 
 #if DATA_SIZE >= 16
 #if HAVE_ATOMIC128
-ABI_TYPE ATOMIC_NAME(ld)(CPUArchState *env, target_ulong addr EXTRA_ARGS)
+ABI_TYPE ATOMIC_NAME(ld)(CPUArchState *env, target_ulong addr,
+                         TCGMemOpIdx oi, uintptr_t retaddr)
 {
-    ATOMIC_MMU_DECLS;
-    DATA_TYPE val, *haddr = ATOMIC_MMU_LOOKUP_R;
-    uint16_t info = trace_mem_build_info(SHIFT, false, MO_BSWAP, false,
-                                         ATOMIC_MMU_IDX);
+    DATA_TYPE *haddr = atomic_mmu_lookup(env, addr, oi, DATA_SIZE,
+                                         PAGE_READ, retaddr);
+    DATA_TYPE val;
+    uint16_t info = atomic_trace_ld_pre(env, addr, oi);
 
-    atomic_trace_ld_pre(env, addr, info);
     val = atomic16_read(haddr);
     ATOMIC_MMU_CLEANUP;
     atomic_trace_ld_post(env, addr, info);
     return BSWAP(val);
 }
 
-void ATOMIC_NAME(st)(CPUArchState *env, target_ulong addr,
-                     ABI_TYPE val EXTRA_ARGS)
+void ATOMIC_NAME(st)(CPUArchState *env, target_ulong addr, ABI_TYPE val,
+                     TCGMemOpIdx oi, uintptr_t retaddr)
 {
-    ATOMIC_MMU_DECLS;
-    DATA_TYPE *haddr = ATOMIC_MMU_LOOKUP_W;
-    uint16_t info = trace_mem_build_info(SHIFT, false, MO_BSWAP, true,
-                                         ATOMIC_MMU_IDX);
+    DATA_TYPE *haddr = atomic_mmu_lookup(env, addr, oi, DATA_SIZE,
+                                         PAGE_WRITE, retaddr);
+    uint16_t info = atomic_trace_st_pre(env, addr, oi);
 
     val = BSWAP(val);
-    atomic_trace_st_pre(env, addr, info);
     val = BSWAP(val);
     atomic16_set(haddr, val);
     ATOMIC_MMU_CLEANUP;
@@ -270,16 +259,14 @@ void ATOMIC_NAME(st)(CPUArchState *env, target_ulong addr,
 }
 #endif
 #else
-ABI_TYPE ATOMIC_NAME(xchg)(CPUArchState *env, target_ulong addr,
-                           ABI_TYPE val EXTRA_ARGS)
+ABI_TYPE ATOMIC_NAME(xchg)(CPUArchState *env, target_ulong addr, ABI_TYPE val,
+                           TCGMemOpIdx oi, uintptr_t retaddr)
 {
-    ATOMIC_MMU_DECLS;
-    DATA_TYPE *haddr = ATOMIC_MMU_LOOKUP_RW;
+    DATA_TYPE *haddr = atomic_mmu_lookup(env, addr, oi, DATA_SIZE,
+                                         PAGE_READ | PAGE_WRITE, retaddr);
     ABI_TYPE ret;
-    uint16_t info = trace_mem_build_info(SHIFT, false, MO_BSWAP, false,
-                                         ATOMIC_MMU_IDX);
+    uint16_t info = atomic_trace_rmw_pre(env, addr, oi);
 
-    atomic_trace_rmw_pre(env, addr, info);
     ret = qatomic_xchg__nocheck(haddr, BSWAP(val));
     ATOMIC_MMU_CLEANUP;
     atomic_trace_rmw_post(env, addr, info);
@@ -288,14 +275,12 @@ ABI_TYPE ATOMIC_NAME(xchg)(CPUArchState *env, target_ulong addr,
 
 #define GEN_ATOMIC_HELPER(X)                                        \
 ABI_TYPE ATOMIC_NAME(X)(CPUArchState *env, target_ulong addr,       \
-                        ABI_TYPE val EXTRA_ARGS)                    \
+                        ABI_TYPE val, TCGMemOpIdx oi, uintptr_t retaddr) \
 {                                                                   \
-    ATOMIC_MMU_DECLS;                                               \
-    DATA_TYPE *haddr = ATOMIC_MMU_LOOKUP_RW;                        \
+    DATA_TYPE *haddr = atomic_mmu_lookup(env, addr, oi, DATA_SIZE,  \
+                                         PAGE_READ | PAGE_WRITE, retaddr); \
     DATA_TYPE ret;                                                  \
-    uint16_t info = trace_mem_build_info(SHIFT, false, MO_BSWAP,    \
-                                         false, ATOMIC_MMU_IDX);    \
-    atomic_trace_rmw_pre(env, addr, info);                          \
+    uint16_t info = atomic_trace_rmw_pre(env, addr, oi);            \
     ret = qatomic_##X(haddr, BSWAP(val));                           \
     ATOMIC_MMU_CLEANUP;                                             \
     atomic_trace_rmw_post(env, addr, info);                         \
@@ -320,14 +305,12 @@ GEN_ATOMIC_HELPER(xor_fetch)
  */
 #define GEN_ATOMIC_HELPER_FN(X, FN, XDATA_TYPE, RET)                \
 ABI_TYPE ATOMIC_NAME(X)(CPUArchState *env, target_ulong addr,       \
-                        ABI_TYPE xval EXTRA_ARGS)                   \
+                        ABI_TYPE xval, TCGMemOpIdx oi, uintptr_t retaddr) \
 {                                                                   \
-    ATOMIC_MMU_DECLS;                                               \
-    XDATA_TYPE *haddr = ATOMIC_MMU_LOOKUP_RW;                       \
+    XDATA_TYPE *haddr = atomic_mmu_lookup(env, addr, oi, DATA_SIZE, \
+                                          PAGE_READ | PAGE_WRITE, retaddr); \
     XDATA_TYPE ldo, ldn, old, new, val = xval;                      \
-    uint16_t info = trace_mem_build_info(SHIFT, false, MO_BSWAP,    \
-                                         false, ATOMIC_MMU_IDX);    \
-    atomic_trace_rmw_pre(env, addr, info);                          \
+    uint16_t info = atomic_trace_rmw_pre(env, addr, oi);            \
     smp_mb();                                                       \
     ldn = qatomic_read__nocheck(haddr);                             \
     do {                                                            \
diff --git a/accel/tcg/cpu-exec.c b/accel/tcg/cpu-exec.c
index e22bcb99f7..fc895cf51e 100644
--- a/accel/tcg/cpu-exec.c
+++ b/accel/tcg/cpu-exec.c
@@ -145,6 +145,28 @@ static void init_delay_params(SyncClocks *sc, const CPUState *cpu)
 }
 #endif /* CONFIG USER ONLY */
 
+uint32_t curr_cflags(CPUState *cpu)
+{
+    uint32_t cflags = cpu->tcg_cflags;
+
+    /*
+     * Record gdb single-step.  We should be exiting the TB by raising
+     * EXCP_DEBUG, but to simplify other tests, disable chaining too.
+     *
+     * For singlestep and -d nochain, suppress goto_tb so that
+     * we can log -d cpu,exec after every TB.
+     */
+    if (unlikely(cpu->singlestep_enabled)) {
+        cflags |= CF_NO_GOTO_TB | CF_NO_GOTO_PTR | CF_SINGLE_STEP | 1;
+    } else if (singlestep) {
+        cflags |= CF_NO_GOTO_TB | 1;
+    } else if (qemu_loglevel_mask(CPU_LOG_TB_NOCHAIN)) {
+        cflags |= CF_NO_GOTO_TB;
+    }
+
+    return cflags;
+}
+
 /* Might cause an exception, so have a longjmp destination ready */
 static inline TranslationBlock *tb_lookup(CPUState *cpu, target_ulong pc,
                                           target_ulong cs_base,
@@ -205,6 +227,76 @@ static inline void log_cpu_exec(target_ulong pc, CPUState *cpu,
     }
 }
 
+static bool check_for_breakpoints(CPUState *cpu, target_ulong pc,
+                                  uint32_t *cflags)
+{
+    CPUBreakpoint *bp;
+    bool match_page = false;
+
+    if (likely(QTAILQ_EMPTY(&cpu->breakpoints))) {
+        return false;
+    }
+
+    /*
+     * Singlestep overrides breakpoints.
+     * This requirement is visible in the record-replay tests, where
+     * we would fail to make forward progress in reverse-continue.
+     *
+     * TODO: gdb singlestep should only override gdb breakpoints,
+     * so that one could (gdb) singlestep into the guest kernel's
+     * architectural breakpoint handler.
+     */
+    if (cpu->singlestep_enabled) {
+        return false;
+    }
+
+    QTAILQ_FOREACH(bp, &cpu->breakpoints, entry) {
+        /*
+         * If we have an exact pc match, trigger the breakpoint.
+         * Otherwise, note matches within the page.
+         */
+        if (pc == bp->pc) {
+            bool match_bp = false;
+
+            if (bp->flags & BP_GDB) {
+                match_bp = true;
+            } else if (bp->flags & BP_CPU) {
+#ifdef CONFIG_USER_ONLY
+                g_assert_not_reached();
+#else
+                CPUClass *cc = CPU_GET_CLASS(cpu);
+                assert(cc->tcg_ops->debug_check_breakpoint);
+                match_bp = cc->tcg_ops->debug_check_breakpoint(cpu);
+#endif
+            }
+
+            if (match_bp) {
+                cpu->exception_index = EXCP_DEBUG;
+                return true;
+            }
+        } else if (((pc ^ bp->pc) & TARGET_PAGE_MASK) == 0) {
+            match_page = true;
+        }
+    }
+
+    /*
+     * Within the same page as a breakpoint, single-step,
+     * returning to helper_lookup_tb_ptr after each insn looking
+     * for the actual breakpoint.
+     *
+     * TODO: Perhaps better to record all of the TBs associated
+     * with a given virtual page that contains a breakpoint, and
+     * then invalidate them when a new overlapping breakpoint is
+     * set on the page.  Non-overlapping TBs would not be
+     * invalidated, nor would any TB need to be invalidated as
+     * breakpoints are removed.
+     */
+    if (match_page) {
+        *cflags = (*cflags & ~CF_COUNT_MASK) | CF_NO_GOTO_TB | 1;
+    }
+    return false;
+}
+
 /**
  * helper_lookup_tb_ptr: quick check for next tb
  * @env: current cpu state
@@ -218,11 +310,16 @@ const void *HELPER(lookup_tb_ptr)(CPUArchState *env)
     CPUState *cpu = env_cpu(env);
     TranslationBlock *tb;
     target_ulong cs_base, pc;
-    uint32_t flags;
+    uint32_t flags, cflags;
 
     cpu_get_tb_cpu_state(env, &pc, &cs_base, &flags);
 
-    tb = tb_lookup(cpu, pc, cs_base, flags, curr_cflags(cpu));
+    cflags = curr_cflags(cpu);
+    if (check_for_breakpoints(cpu, pc, &cflags)) {
+        cpu_loop_exit(cpu);
+    }
+
+    tb = tb_lookup(cpu, pc, cs_base, flags, cflags);
     if (tb == NULL) {
         return tcg_code_gen_epilogue;
     }
@@ -313,8 +410,7 @@ void cpu_exec_step_atomic(CPUState *cpu)
     CPUArchState *env = (CPUArchState *)cpu->env_ptr;
     TranslationBlock *tb;
     target_ulong cs_base, pc;
-    uint32_t flags;
-    uint32_t cflags = (curr_cflags(cpu) & ~CF_PARALLEL) | 1;
+    uint32_t flags, cflags;
     int tb_exit;
 
     if (sigsetjmp(cpu->jmp_env, 0) == 0) {
@@ -324,8 +420,20 @@ void cpu_exec_step_atomic(CPUState *cpu)
         cpu->running = true;
 
         cpu_get_tb_cpu_state(env, &pc, &cs_base, &flags);
-        tb = tb_lookup(cpu, pc, cs_base, flags, cflags);
 
+        cflags = curr_cflags(cpu);
+        /* Execute in a serial context. */
+        cflags &= ~CF_PARALLEL;
+        /* After 1 insn, return and release the exclusive lock. */
+        cflags |= CF_NO_GOTO_TB | CF_NO_GOTO_PTR | 1;
+        /*
+         * No need to check_for_breakpoints here.
+         * We only arrive in cpu_exec_step_atomic after beginning execution
+         * of an insn that includes an atomic operation we can't handle.
+         * Any breakpoint for this insn will have been recognized earlier.
+         */
+
+        tb = tb_lookup(cpu, pc, cs_base, flags, cflags);
         if (tb == NULL) {
             mmap_lock();
             tb = tb_gen_code(cpu, pc, cs_base, flags, cflags);
@@ -478,41 +586,6 @@ static inline void tb_add_jump(TranslationBlock *tb, int n,
     return;
 }
 
-static inline TranslationBlock *tb_find(CPUState *cpu,
-                                        TranslationBlock *last_tb,
-                                        int tb_exit, uint32_t cflags)
-{
-    CPUArchState *env = (CPUArchState *)cpu->env_ptr;
-    TranslationBlock *tb;
-    target_ulong cs_base, pc;
-    uint32_t flags;
-
-    cpu_get_tb_cpu_state(env, &pc, &cs_base, &flags);
-
-    tb = tb_lookup(cpu, pc, cs_base, flags, cflags);
-    if (tb == NULL) {
-        mmap_lock();
-        tb = tb_gen_code(cpu, pc, cs_base, flags, cflags);
-        mmap_unlock();
-        /* We add the TB in the virtual pc hash table for the fast lookup */
-        qatomic_set(&cpu->tb_jmp_cache[tb_jmp_cache_hash_func(pc)], tb);
-    }
-#ifndef CONFIG_USER_ONLY
-    /* We don't take care of direct jumps when address mapping changes in
-     * system emulation. So it's not safe to make a direct jump to a TB
-     * spanning two pages because the mapping for the second page can change.
-     */
-    if (tb->page_addr[1] != -1) {
-        last_tb = NULL;
-    }
-#endif
-    /* See if we can patch the calling TB. */
-    if (last_tb) {
-        tb_add_jump(last_tb, tb_exit, tb);
-    }
-    return tb;
-}
-
 static inline bool cpu_handle_halt(CPUState *cpu)
 {
     if (cpu->halted) {
@@ -846,22 +919,60 @@ int cpu_exec(CPUState *cpu)
         int tb_exit = 0;
 
         while (!cpu_handle_interrupt(cpu, &last_tb)) {
-            uint32_t cflags = cpu->cflags_next_tb;
             TranslationBlock *tb;
-
-            /* When requested, use an exact setting for cflags for the next
-               execution.  This is used for icount, precise smc, and stop-
-               after-access watchpoints.  Since this request should never
-               have CF_INVALID set, -1 is a convenient invalid value that
-               does not require tcg headers for cpu_common_reset.  */
+            target_ulong cs_base, pc;
+            uint32_t flags, cflags;
+
+            cpu_get_tb_cpu_state(cpu->env_ptr, &pc, &cs_base, &flags);
+
+            /*
+             * When requested, use an exact setting for cflags for the next
+             * execution.  This is used for icount, precise smc, and stop-
+             * after-access watchpoints.  Since this request should never
+             * have CF_INVALID set, -1 is a convenient invalid value that
+             * does not require tcg headers for cpu_common_reset.
+             */
+            cflags = cpu->cflags_next_tb;
             if (cflags == -1) {
                 cflags = curr_cflags(cpu);
             } else {
                 cpu->cflags_next_tb = -1;
             }
 
-            tb = tb_find(cpu, last_tb, tb_exit, cflags);
+            if (check_for_breakpoints(cpu, pc, &cflags)) {
+                break;
+            }
+
+            tb = tb_lookup(cpu, pc, cs_base, flags, cflags);
+            if (tb == NULL) {
+                mmap_lock();
+                tb = tb_gen_code(cpu, pc, cs_base, flags, cflags);
+                mmap_unlock();
+                /*
+                 * We add the TB in the virtual pc hash table
+                 * for the fast lookup
+                 */
+                qatomic_set(&cpu->tb_jmp_cache[tb_jmp_cache_hash_func(pc)], tb);
+            }
+
+#ifndef CONFIG_USER_ONLY
+            /*
+             * We don't take care of direct jumps when address mapping
+             * changes in system emulation.  So it's not safe to make a
+             * direct jump to a TB spanning two pages because the mapping
+             * for the second page can change.
+             */
+            if (tb->page_addr[1] != -1) {
+                last_tb = NULL;
+            }
+#endif
+            /* See if we can patch the calling TB. */
+            if (last_tb) {
+                tb_add_jump(last_tb, tb_exit, tb);
+            }
+
             cpu_loop_exec_tb(cpu, tb, &last_tb, &tb_exit);
+
             /* Try to align the host and virtual clocks
                if the guest is in advance */
             align_clocks(&sc, cpu);
diff --git a/accel/tcg/cputlb.c b/accel/tcg/cputlb.c
index b4e15b6aad..b1e5471f94 100644
--- a/accel/tcg/cputlb.c
+++ b/accel/tcg/cputlb.c
@@ -2686,19 +2686,14 @@ void cpu_stq_le_data(CPUArchState *env, target_ulong ptr, uint64_t val)
     cpu_stq_le_data_ra(env, ptr, val, 0);
 }
 
-/* First set of helpers allows passing in of OI and RETADDR.  This makes
-   them callable from other helpers.  */
+/*
+ * First set of functions passes in OI and RETADDR.
+ * This makes them callable from other helpers.
+ */
 
-#define EXTRA_ARGS     , TCGMemOpIdx oi, uintptr_t retaddr
 #define ATOMIC_NAME(X) \
-    HELPER(glue(glue(glue(atomic_ ## X, SUFFIX), END), _mmu))
-#define ATOMIC_MMU_DECLS
-#define ATOMIC_MMU_LOOKUP_RW \
-    atomic_mmu_lookup(env, addr, oi, DATA_SIZE, PAGE_READ | PAGE_WRITE, retaddr)
-#define ATOMIC_MMU_LOOKUP_R \
-    atomic_mmu_lookup(env, addr, oi, DATA_SIZE, PAGE_READ, retaddr)
-#define ATOMIC_MMU_LOOKUP_W \
-    atomic_mmu_lookup(env, addr, oi, DATA_SIZE, PAGE_WRITE, retaddr)
+    glue(glue(glue(cpu_atomic_ ## X, SUFFIX), END), _mmu)
+
 #define ATOMIC_MMU_CLEANUP
 #define ATOMIC_MMU_IDX   get_mmuidx(oi)
 
@@ -2723,38 +2718,6 @@ void cpu_stq_le_data(CPUArchState *env, target_ulong ptr, uint64_t val)
 #include "atomic_template.h"
 #endif
 
-/* Second set of helpers are directly callable from TCG as helpers.  */
-
-#undef EXTRA_ARGS
-#undef ATOMIC_NAME
-#undef ATOMIC_MMU_LOOKUP_RW
-#undef ATOMIC_MMU_LOOKUP_R
-#undef ATOMIC_MMU_LOOKUP_W
-
-#define EXTRA_ARGS         , TCGMemOpIdx oi
-#define ATOMIC_NAME(X)     HELPER(glue(glue(atomic_ ## X, SUFFIX), END))
-#define ATOMIC_MMU_LOOKUP_RW \
-    atomic_mmu_lookup(env, addr, oi, DATA_SIZE, PAGE_READ | PAGE_WRITE, GETPC())
-#define ATOMIC_MMU_LOOKUP_R \
-    atomic_mmu_lookup(env, addr, oi, DATA_SIZE, PAGE_READ, GETPC())
-#define ATOMIC_MMU_LOOKUP_W \
-    atomic_mmu_lookup(env, addr, oi, DATA_SIZE, PAGE_WRITE, GETPC())
-
-#define DATA_SIZE 1
-#include "atomic_template.h"
-
-#define DATA_SIZE 2
-#include "atomic_template.h"
-
-#define DATA_SIZE 4
-#include "atomic_template.h"
-
-#ifdef CONFIG_ATOMIC64
-#define DATA_SIZE 8
-#include "atomic_template.h"
-#endif
-#undef ATOMIC_MMU_IDX
-
 /* Code access functions.  */
 
 static uint64_t full_ldub_code(CPUArchState *env, target_ulong addr,
diff --git a/accel/tcg/tcg-runtime.h b/accel/tcg/tcg-runtime.h
index 91a5b7e85f..37cbd722bf 100644
--- a/accel/tcg/tcg-runtime.h
+++ b/accel/tcg/tcg-runtime.h
@@ -39,8 +39,6 @@ DEF_HELPER_FLAGS_1(exit_atomic, TCG_CALL_NO_WG, noreturn, env)
 DEF_HELPER_FLAGS_3(memset, TCG_CALL_NO_RWG, ptr, ptr, int, ptr)
 #endif /* IN_HELPER_PROTO */
 
-#ifdef CONFIG_SOFTMMU
-
 DEF_HELPER_FLAGS_5(atomic_cmpxchgb, TCG_CALL_NO_WG,
                    i32, env, tl, i32, i32, i32)
 DEF_HELPER_FLAGS_5(atomic_cmpxchgw_be, TCG_CALL_NO_WG,
@@ -88,50 +86,6 @@ DEF_HELPER_FLAGS_5(atomic_cmpxchgq_le, TCG_CALL_NO_WG,
                        TCG_CALL_NO_WG, i32, env, tl, i32, i32)
 #endif /* CONFIG_ATOMIC64 */
 
-#else
-
-DEF_HELPER_FLAGS_4(atomic_cmpxchgb, TCG_CALL_NO_WG, i32, env, tl, i32, i32)
-DEF_HELPER_FLAGS_4(atomic_cmpxchgw_be, TCG_CALL_NO_WG, i32, env, tl, i32, i32)
-DEF_HELPER_FLAGS_4(atomic_cmpxchgw_le, TCG_CALL_NO_WG, i32, env, tl, i32, i32)
-DEF_HELPER_FLAGS_4(atomic_cmpxchgl_be, TCG_CALL_NO_WG, i32, env, tl, i32, i32)
-DEF_HELPER_FLAGS_4(atomic_cmpxchgl_le, TCG_CALL_NO_WG, i32, env, tl, i32, i32)
-#ifdef CONFIG_ATOMIC64
-DEF_HELPER_FLAGS_4(atomic_cmpxchgq_be, TCG_CALL_NO_WG, i64, env, tl, i64, i64)
-DEF_HELPER_FLAGS_4(atomic_cmpxchgq_le, TCG_CALL_NO_WG, i64, env, tl, i64, i64)
-#endif
-
-#ifdef CONFIG_ATOMIC64
-#define GEN_ATOMIC_HELPERS(NAME)                             \
-    DEF_HELPER_FLAGS_3(glue(glue(atomic_, NAME), b),         \
-                       TCG_CALL_NO_WG, i32, env, tl, i32)    \
-    DEF_HELPER_FLAGS_3(glue(glue(atomic_, NAME), w_le),      \
-                       TCG_CALL_NO_WG, i32, env, tl, i32)    \
-    DEF_HELPER_FLAGS_3(glue(glue(atomic_, NAME), w_be),      \
-                       TCG_CALL_NO_WG, i32, env, tl, i32)    \
-    DEF_HELPER_FLAGS_3(glue(glue(atomic_, NAME), l_le),      \
-                       TCG_CALL_NO_WG, i32, env, tl, i32)    \
-    DEF_HELPER_FLAGS_3(glue(glue(atomic_, NAME), l_be),      \
-                       TCG_CALL_NO_WG, i32, env, tl, i32)    \
-    DEF_HELPER_FLAGS_3(glue(glue(atomic_, NAME), q_le),      \
-                       TCG_CALL_NO_WG, i64, env, tl, i64)    \
-    DEF_HELPER_FLAGS_3(glue(glue(atomic_, NAME), q_be),      \
-                       TCG_CALL_NO_WG, i64, env, tl, i64)
-#else
-#define GEN_ATOMIC_HELPERS(NAME)                             \
-    DEF_HELPER_FLAGS_3(glue(glue(atomic_, NAME), b),         \
-                       TCG_CALL_NO_WG, i32, env, tl, i32)    \
-    DEF_HELPER_FLAGS_3(glue(glue(atomic_, NAME), w_le),      \
-                       TCG_CALL_NO_WG, i32, env, tl, i32)    \
-    DEF_HELPER_FLAGS_3(glue(glue(atomic_, NAME), w_be),      \
-                       TCG_CALL_NO_WG, i32, env, tl, i32)    \
-    DEF_HELPER_FLAGS_3(glue(glue(atomic_, NAME), l_le),      \
-                       TCG_CALL_NO_WG, i32, env, tl, i32)    \
-    DEF_HELPER_FLAGS_3(glue(glue(atomic_, NAME), l_be),      \
-                       TCG_CALL_NO_WG, i32, env, tl, i32)
-#endif /* CONFIG_ATOMIC64 */
-
-#endif /* CONFIG_SOFTMMU */
-
 GEN_ATOMIC_HELPERS(fetch_add)
 GEN_ATOMIC_HELPERS(fetch_and)
 GEN_ATOMIC_HELPERS(fetch_or)
diff --git a/accel/tcg/translate-all.c b/accel/tcg/translate-all.c
index 4df26de858..bbfcfb698c 100644
--- a/accel/tcg/translate-all.c
+++ b/accel/tcg/translate-all.c
@@ -1428,14 +1428,9 @@ TranslationBlock *tb_gen_code(CPUState *cpu,
 
     max_insns = cflags & CF_COUNT_MASK;
     if (max_insns == 0) {
-        max_insns = CF_COUNT_MASK;
-    }
-    if (max_insns > TCG_MAX_INSNS) {
         max_insns = TCG_MAX_INSNS;
     }
-    if (cpu->singlestep_enabled || singlestep) {
-        max_insns = 1;
-    }
+    QEMU_BUILD_BUG_ON(CF_COUNT_MASK + 1 != TCG_MAX_INSNS);
 
  buffer_overflow:
     tb = tcg_tb_alloc(tcg_ctx);
diff --git a/accel/tcg/translator.c b/accel/tcg/translator.c
index 59804af37b..c53a7f8e44 100644
--- a/accel/tcg/translator.c
+++ b/accel/tcg/translator.c
@@ -33,8 +33,8 @@ void translator_loop_temp_check(DisasContextBase *db)
 
 bool translator_use_goto_tb(DisasContextBase *db, target_ulong dest)
 {
-    /* Suppress goto_tb in the case of single-steping.  */
-    if (db->singlestep_enabled || singlestep) {
+    /* Suppress goto_tb if requested. */
+    if (tb_cflags(db->tb) & CF_NO_GOTO_TB) {
         return false;
     }
 
@@ -45,7 +45,7 @@ bool translator_use_goto_tb(DisasContextBase *db, target_ulong dest)
 void translator_loop(const TranslatorOps *ops, DisasContextBase *db,
                      CPUState *cpu, TranslationBlock *tb, int max_insns)
 {
-    int bp_insn = 0;
+    uint32_t cflags = tb_cflags(tb);
     bool plugin_enabled;
 
     /* Initialize DisasContext */
@@ -55,7 +55,7 @@ void translator_loop(const TranslatorOps *ops, DisasContextBase *db,
     db->is_jmp = DISAS_NEXT;
     db->num_insns = 0;
     db->max_insns = max_insns;
-    db->singlestep_enabled = cpu->singlestep_enabled;
+    db->singlestep_enabled = cflags & CF_SINGLE_STEP;
 
     ops->init_disas_context(db, cpu);
     tcg_debug_assert(db->is_jmp == DISAS_NEXT);  /* no early exit */
@@ -68,8 +68,7 @@ void translator_loop(const TranslatorOps *ops, DisasContextBase *db,
     ops->tb_start(db, cpu);
     tcg_debug_assert(db->is_jmp == DISAS_NEXT);  /* no early exit */
 
-    plugin_enabled = plugin_gen_tb_start(cpu, tb,
-                                         tb_cflags(db->tb) & CF_MEMI_ONLY);
+    plugin_enabled = plugin_gen_tb_start(cpu, tb, cflags & CF_MEMI_ONLY);
 
     while (true) {
         db->num_insns++;
@@ -80,39 +79,17 @@ void translator_loop(const TranslatorOps *ops, DisasContextBase *db,
             plugin_gen_insn_start(cpu, db);
         }
 
-        /* Pass breakpoint hits to target for further processing */
-        if (!db->singlestep_enabled
-            && unlikely(!QTAILQ_EMPTY(&cpu->breakpoints))) {
-            CPUBreakpoint *bp;
-            QTAILQ_FOREACH(bp, &cpu->breakpoints, entry) {
-                if (bp->pc == db->pc_next) {
-                    if (ops->breakpoint_check(db, cpu, bp)) {
-                        bp_insn = 1;
-                        break;
-                    }
-                }
-            }
-            /* The breakpoint_check hook may use DISAS_TOO_MANY to indicate
-               that only one more instruction is to be executed.  Otherwise
-               it should use DISAS_NORETURN when generating an exception,
-               but may use a DISAS_TARGET_* value for Something Else.  */
-            if (db->is_jmp > DISAS_TOO_MANY) {
-                break;
-            }
-        }
-
         /* Disassemble one instruction.  The translate_insn hook should
            update db->pc_next and db->is_jmp to indicate what should be
            done next -- either exiting this loop or locate the start of
            the next instruction.  */
-        if (db->num_insns == db->max_insns
-            && (tb_cflags(db->tb) & CF_LAST_IO)) {
+        if (db->num_insns == db->max_insns && (cflags & CF_LAST_IO)) {
             /* Accept I/O on the last instruction.  */
             gen_io_start();
             ops->translate_insn(db, cpu);
         } else {
             /* we should only see CF_MEMI_ONLY for io_recompile */
-            tcg_debug_assert(!(tb_cflags(db->tb) & CF_MEMI_ONLY));
+            tcg_debug_assert(!(cflags & CF_MEMI_ONLY));
             ops->translate_insn(db, cpu);
         }
 
@@ -139,7 +116,7 @@ void translator_loop(const TranslatorOps *ops, DisasContextBase *db,
 
     /* Emit code to exit the TB, as indicated by db->is_jmp.  */
     ops->tb_stop(db, cpu);
-    gen_tb_end(db->tb, db->num_insns - bp_insn);
+    gen_tb_end(db->tb, db->num_insns);
 
     if (plugin_enabled) {
         plugin_gen_tb_end(cpu);
diff --git a/accel/tcg/user-exec.c b/accel/tcg/user-exec.c
index ba09fd0413..90d1a2d327 100644
--- a/accel/tcg/user-exec.c
+++ b/accel/tcg/user-exec.c
@@ -1221,9 +1221,14 @@ uint64_t cpu_ldq_code(CPUArchState *env, abi_ptr ptr)
     return ret;
 }
 
-/* Do not allow unaligned operations to proceed.  Return the host address.  */
+/*
+ * Do not allow unaligned operations to proceed.  Return the host address.
+ *
+ * @prot may be PAGE_READ, PAGE_WRITE, or PAGE_READ|PAGE_WRITE.
+ */
 static void *atomic_mmu_lookup(CPUArchState *env, target_ulong addr,
-                               int size, uintptr_t retaddr)
+                               TCGMemOpIdx oi, int size, int prot,
+                               uintptr_t retaddr)
 {
     /* Enforce qemu required alignment.  */
     if (unlikely(addr & (size - 1))) {
@@ -1234,18 +1239,17 @@ static void *atomic_mmu_lookup(CPUArchState *env, target_ulong addr,
     return ret;
 }
 
-/* Macro to call the above, with local variables from the use context.  */
-#define ATOMIC_MMU_DECLS do {} while (0)
-#define ATOMIC_MMU_LOOKUP_RW  atomic_mmu_lookup(env, addr, DATA_SIZE, GETPC())
-#define ATOMIC_MMU_LOOKUP_R   ATOMIC_MMU_LOOKUP_RW
-#define ATOMIC_MMU_LOOKUP_W   ATOMIC_MMU_LOOKUP_RW
-#define ATOMIC_MMU_CLEANUP do { clear_helper_retaddr(); } while (0)
-#define ATOMIC_MMU_IDX MMU_USER_IDX
+#include "atomic_common.c.inc"
 
-#define ATOMIC_NAME(X)   HELPER(glue(glue(atomic_ ## X, SUFFIX), END))
-#define EXTRA_ARGS
+/*
+ * First set of functions passes in OI and RETADDR.
+ * This makes them callable from other helpers.
+ */
 
-#include "atomic_common.c.inc"
+#define ATOMIC_NAME(X) \
+    glue(glue(glue(cpu_atomic_ ## X, SUFFIX), END), _mmu)
+#define ATOMIC_MMU_CLEANUP do { clear_helper_retaddr(); } while (0)
+#define ATOMIC_MMU_IDX MMU_USER_IDX
 
 #define DATA_SIZE 1
 #include "atomic_template.h"
@@ -1261,20 +1265,7 @@ static void *atomic_mmu_lookup(CPUArchState *env, target_ulong addr,
 #include "atomic_template.h"
 #endif
 
-/* The following is only callable from other helpers, and matches up
-   with the softmmu version.  */
-
 #if HAVE_ATOMIC128 || HAVE_CMPXCHG128
-
-#undef EXTRA_ARGS
-#undef ATOMIC_NAME
-#undef ATOMIC_MMU_LOOKUP_RW
-
-#define EXTRA_ARGS     , TCGMemOpIdx oi, uintptr_t retaddr
-#define ATOMIC_NAME(X) \
-    HELPER(glue(glue(glue(atomic_ ## X, SUFFIX), END), _mmu))
-#define ATOMIC_MMU_LOOKUP_RW  atomic_mmu_lookup(env, addr, DATA_SIZE, retaddr)
-
 #define DATA_SIZE 16
 #include "atomic_template.h"
 #endif
diff --git a/block.c b/block.c
index be083f389e..e97ce0b1c8 100644
--- a/block.c
+++ b/block.c
@@ -6162,6 +6162,9 @@ BlockDriverState *bdrv_find_backing_image(BlockDriverState *bs,
 
 void bdrv_init(void)
 {
+#ifdef CONFIG_BDRV_WHITELIST_TOOLS
+    use_bdrv_whitelist = 1;
+#endif
     module_call_init(MODULE_INIT_BLOCK);
 }
 
diff --git a/block/blkdebug.c b/block/blkdebug.c
index 2c0b9b0ee8..8b67554bec 100644
--- a/block/blkdebug.c
+++ b/block/blkdebug.c
@@ -38,25 +38,27 @@
 #include "qapi/qobject-input-visitor.h"
 #include "sysemu/qtest.h"
 
+/* All APIs are thread-safe */
+
 typedef struct BDRVBlkdebugState {
-    int state;
-    int new_state;
+    /* IN: initialized in blkdebug_open() and never changed */
     uint64_t align;
     uint64_t max_transfer;
     uint64_t opt_write_zero;
     uint64_t max_write_zero;
     uint64_t opt_discard;
     uint64_t max_discard;
-
+    char *config_file; /* For blkdebug_refresh_filename() */
+    /* initialized in blkdebug_parse_perms() */
     uint64_t take_child_perms;
     uint64_t unshare_child_perms;
 
-    /* For blkdebug_refresh_filename() */
-    char *config_file;
-
+    /* State. Protected by lock */
+    int state;
     QLIST_HEAD(, BlkdebugRule) rules[BLKDBG__MAX];
     QSIMPLEQ_HEAD(, BlkdebugRule) active_rules;
     QLIST_HEAD(, BlkdebugSuspendedReq) suspended_reqs;
+    QemuMutex lock;
 } BDRVBlkdebugState;
 
 typedef struct BlkdebugAIOCB {
@@ -65,8 +67,11 @@ typedef struct BlkdebugAIOCB {
 } BlkdebugAIOCB;
 
 typedef struct BlkdebugSuspendedReq {
+    /* IN: initialized in suspend_request() */
     Coroutine *co;
     char *tag;
+
+    /* List entry protected BDRVBlkdebugState's lock */
     QLIST_ENTRY(BlkdebugSuspendedReq) next;
 } BlkdebugSuspendedReq;
 
@@ -74,9 +79,11 @@ enum {
     ACTION_INJECT_ERROR,
     ACTION_SET_STATE,
     ACTION_SUSPEND,
+    ACTION__MAX,
 };
 
 typedef struct BlkdebugRule {
+    /* IN: initialized in add_rule() or blkdebug_debug_breakpoint() */
     BlkdebugEvent event;
     int action;
     int state;
@@ -95,6 +102,8 @@ typedef struct BlkdebugRule {
             char *tag;
         } suspend;
     } options;
+
+    /* List entries protected BDRVBlkdebugState's lock */
     QLIST_ENTRY(BlkdebugRule) next;
     QSIMPLEQ_ENTRY(BlkdebugRule) active_next;
 } BlkdebugRule;
@@ -244,11 +253,14 @@ static int add_rule(void *opaque, QemuOpts *opts, Error **errp)
     };
 
     /* Add the rule */
+    qemu_mutex_lock(&s->lock);
     QLIST_INSERT_HEAD(&s->rules[event], rule, next);
+    qemu_mutex_unlock(&s->lock);
 
     return 0;
 }
 
+/* Called with lock held or from .bdrv_close */
 static void remove_rule(BlkdebugRule *rule)
 {
     switch (rule->action) {
@@ -467,6 +479,7 @@ static int blkdebug_open(BlockDriverState *bs, QDict *options, int flags,
     int ret;
     uint64_t align;
 
+    qemu_mutex_init(&s->lock);
     opts = qemu_opts_create(&runtime_opts, NULL, 0, &error_abort);
     if (!qemu_opts_absorb_qdict(opts, options, errp)) {
         ret = -EINVAL;
@@ -567,6 +580,7 @@ static int blkdebug_open(BlockDriverState *bs, QDict *options, int flags,
     ret = 0;
 out:
     if (ret < 0) {
+        qemu_mutex_destroy(&s->lock);
         g_free(s->config_file);
     }
     qemu_opts_del(opts);
@@ -581,6 +595,7 @@ static int rule_check(BlockDriverState *bs, uint64_t offset, uint64_t bytes,
     int error;
     bool immediately;
 
+    qemu_mutex_lock(&s->lock);
     QSIMPLEQ_FOREACH(rule, &s->active_rules, active_next) {
         uint64_t inject_offset = rule->options.inject.offset;
 
@@ -594,6 +609,7 @@ static int rule_check(BlockDriverState *bs, uint64_t offset, uint64_t bytes,
     }
 
     if (!rule || !rule->options.inject.error) {
+        qemu_mutex_unlock(&s->lock);
         return 0;
     }
 
@@ -605,6 +621,7 @@ static int rule_check(BlockDriverState *bs, uint64_t offset, uint64_t bytes,
         remove_rule(rule);
     }
 
+    qemu_mutex_unlock(&s->lock);
     if (!immediately) {
         aio_co_schedule(qemu_get_current_aio_context(), qemu_coroutine_self());
         qemu_coroutine_yield();
@@ -770,78 +787,80 @@ static void blkdebug_close(BlockDriverState *bs)
     }
 
     g_free(s->config_file);
+    qemu_mutex_destroy(&s->lock);
 }
 
+/* Called with lock held.  */
 static void suspend_request(BlockDriverState *bs, BlkdebugRule *rule)
 {
     BDRVBlkdebugState *s = bs->opaque;
-    BlkdebugSuspendedReq r;
+    BlkdebugSuspendedReq *r;
 
-    r = (BlkdebugSuspendedReq) {
-        .co         = qemu_coroutine_self(),
-        .tag        = g_strdup(rule->options.suspend.tag),
-    };
+    r = g_new(BlkdebugSuspendedReq, 1);
+
+    r->co         = qemu_coroutine_self();
+    r->tag        = g_strdup(rule->options.suspend.tag);
 
     remove_rule(rule);
-    QLIST_INSERT_HEAD(&s->suspended_reqs, &r, next);
+    QLIST_INSERT_HEAD(&s->suspended_reqs, r, next);
 
     if (!qtest_enabled()) {
-        printf("blkdebug: Suspended request '%s'\n", r.tag);
-    }
-    qemu_coroutine_yield();
-    if (!qtest_enabled()) {
-        printf("blkdebug: Resuming request '%s'\n", r.tag);
+        printf("blkdebug: Suspended request '%s'\n", r->tag);
     }
-
-    QLIST_REMOVE(&r, next);
-    g_free(r.tag);
 }
 
-static bool process_rule(BlockDriverState *bs, struct BlkdebugRule *rule,
-    bool injected)
+/* Called with lock held.  */
+static void process_rule(BlockDriverState *bs, struct BlkdebugRule *rule,
+                         int *action_count, int *new_state)
 {
     BDRVBlkdebugState *s = bs->opaque;
 
     /* Only process rules for the current state */
     if (rule->state && rule->state != s->state) {
-        return injected;
+        return;
     }
 
     /* Take the action */
+    action_count[rule->action]++;
     switch (rule->action) {
     case ACTION_INJECT_ERROR:
-        if (!injected) {
+        if (action_count[ACTION_INJECT_ERROR] == 1) {
             QSIMPLEQ_INIT(&s->active_rules);
-            injected = true;
         }
         QSIMPLEQ_INSERT_HEAD(&s->active_rules, rule, active_next);
         break;
 
     case ACTION_SET_STATE:
-        s->new_state = rule->options.set_state.new_state;
+        *new_state = rule->options.set_state.new_state;
         break;
 
     case ACTION_SUSPEND:
         suspend_request(bs, rule);
         break;
     }
-    return injected;
 }
 
 static void blkdebug_debug_event(BlockDriverState *bs, BlkdebugEvent event)
 {
     BDRVBlkdebugState *s = bs->opaque;
     struct BlkdebugRule *rule, *next;
-    bool injected;
+    int new_state;
+    int actions_count[ACTION__MAX] = { 0 };
 
     assert((int)event >= 0 && event < BLKDBG__MAX);
 
-    injected = false;
-    s->new_state = s->state;
-    QLIST_FOREACH_SAFE(rule, &s->rules[event], next, next) {
-        injected = process_rule(bs, rule, injected);
+    WITH_QEMU_LOCK_GUARD(&s->lock) {
+        new_state = s->state;
+        QLIST_FOREACH_SAFE(rule, &s->rules[event], next, next) {
+            process_rule(bs, rule, actions_count, &new_state);
+        }
+        s->state = new_state;
+    }
+
+    while (actions_count[ACTION_SUSPEND] > 0) {
+        qemu_coroutine_yield();
+        actions_count[ACTION_SUSPEND]--;
     }
-    s->state = s->new_state;
 }
 
 static int blkdebug_debug_breakpoint(BlockDriverState *bs, const char *event,
@@ -864,33 +883,64 @@ static int blkdebug_debug_breakpoint(BlockDriverState *bs, const char *event,
         .options.suspend.tag = g_strdup(tag),
     };
 
+    qemu_mutex_lock(&s->lock);
     QLIST_INSERT_HEAD(&s->rules[blkdebug_event], rule, next);
+    qemu_mutex_unlock(&s->lock);
 
     return 0;
 }
 
-static int blkdebug_debug_resume(BlockDriverState *bs, const char *tag)
+/* Called with lock held. May temporarily release lock. */
+static int resume_req_by_tag(BDRVBlkdebugState *s, const char *tag, bool all)
 {
-    BDRVBlkdebugState *s = bs->opaque;
-    BlkdebugSuspendedReq *r, *next;
+    BlkdebugSuspendedReq *r;
 
-    QLIST_FOREACH_SAFE(r, &s->suspended_reqs, next, next) {
+retry:
+    /*
+     * No need for _SAFE, since a different coroutine can remove another node
+     * (not the current one) in this list, and when the current one is removed
+     * the iteration starts back from beginning anyways.
+     */
+    QLIST_FOREACH(r, &s->suspended_reqs, next) {
         if (!strcmp(r->tag, tag)) {
-            qemu_coroutine_enter(r->co);
+            Coroutine *co = r->co;
+
+            if (!qtest_enabled()) {
+                printf("blkdebug: Resuming request '%s'\n", r->tag);
+            }
+
+            QLIST_REMOVE(r, next);
+            g_free(r->tag);
+            g_free(r);
+
+            qemu_mutex_unlock(&s->lock);
+            qemu_coroutine_enter(co);
+            qemu_mutex_lock(&s->lock);
+
+            if (all) {
+                goto retry;
+            }
             return 0;
         }
     }
     return -ENOENT;
 }
 
+static int blkdebug_debug_resume(BlockDriverState *bs, const char *tag)
+{
+    BDRVBlkdebugState *s = bs->opaque;
+    QEMU_LOCK_GUARD(&s->lock);
+    return resume_req_by_tag(s, tag, false);
+}
+
 static int blkdebug_debug_remove_breakpoint(BlockDriverState *bs,
                                             const char *tag)
 {
     BDRVBlkdebugState *s = bs->opaque;
-    BlkdebugSuspendedReq *r, *r_next;
     BlkdebugRule *rule, *next;
     int i, ret = -ENOENT;
 
+    QEMU_LOCK_GUARD(&s->lock);
     for (i = 0; i < BLKDBG__MAX; i++) {
         QLIST_FOREACH_SAFE(rule, &s->rules[i], next, next) {
             if (rule->action == ACTION_SUSPEND &&
@@ -900,11 +950,8 @@ static int blkdebug_debug_remove_breakpoint(BlockDriverState *bs,
             }
         }
     }
-    QLIST_FOREACH_SAFE(r, &s->suspended_reqs, next, r_next) {
-        if (!strcmp(r->tag, tag)) {
-            qemu_coroutine_enter(r->co);
-            ret = 0;
-        }
+    if (resume_req_by_tag(s, tag, true) == 0) {
+        ret = 0;
     }
     return ret;
 }
@@ -914,6 +961,7 @@ static bool blkdebug_debug_is_suspended(BlockDriverState *bs, const char *tag)
     BDRVBlkdebugState *s = bs->opaque;
     BlkdebugSuspendedReq *r;
 
+    QEMU_LOCK_GUARD(&s->lock);
     QLIST_FOREACH(r, &s->suspended_reqs, next) {
         if (!strcmp(r->tag, tag)) {
             return true;
diff --git a/block/dirty-bitmap.c b/block/dirty-bitmap.c
index 68d295d6e3..0ef46163e3 100644
--- a/block/dirty-bitmap.c
+++ b/block/dirty-bitmap.c
@@ -193,7 +193,7 @@ int bdrv_dirty_bitmap_check(const BdrvDirtyBitmap *bitmap, uint32_t flags,
         error_setg(errp, "Bitmap '%s' is inconsistent and cannot be used",
                    bitmap->name);
         error_append_hint(errp, "Try block-dirty-bitmap-remove to delete"
-                          " this bitmap from disk");
+                          " this bitmap from disk\n");
         return -1;
     }
 
diff --git a/block/export/export.c b/block/export/export.c
index fec7d9f738..6d3b9964c8 100644
--- a/block/export/export.c
+++ b/block/export/export.c
@@ -111,6 +111,7 @@ BlockExport *blk_exp_add(BlockExportOptions *export, Error **errp)
     if (export->has_iothread) {
         IOThread *iothread;
         AioContext *new_ctx;
+        Error **set_context_errp;
 
         iothread = iothread_by_id(export->iothread);
         if (!iothread) {
@@ -120,7 +121,9 @@ BlockExport *blk_exp_add(BlockExportOptions *export, Error **errp)
 
         new_ctx = iothread_get_aio_context(iothread);
 
-        ret = bdrv_try_set_aio_context(bs, new_ctx, errp);
+        /* Ignore errors with fixed-iothread=false */
+        set_context_errp = fixed_iothread ? errp : NULL;
+        ret = bdrv_try_set_aio_context(bs, new_ctx, set_context_errp);
         if (ret == 0) {
             aio_context_release(ctx);
             aio_context_acquire(new_ctx);
diff --git a/block/linux-aio.c b/block/linux-aio.c
index 3c0527c2bf..0dab507b71 100644
--- a/block/linux-aio.c
+++ b/block/linux-aio.c
@@ -28,6 +28,9 @@
  */
 #define MAX_EVENTS 1024
 
+/* Maximum number of requests in a batch. (default value) */
+#define DEFAULT_MAX_BATCH 32
+
 struct qemu_laiocb {
     Coroutine *co;
     LinuxAioState *ctx;
@@ -351,6 +354,10 @@ static int laio_do_submit(int fd, struct qemu_laiocb *laiocb, off_t offset,
     LinuxAioState *s = laiocb->ctx;
     struct iocb *iocbs = &laiocb->iocb;
     QEMUIOVector *qiov = laiocb->qiov;
+    int64_t max_batch = s->aio_context->aio_max_batch ?: DEFAULT_MAX_BATCH;
+
+    /* limit the batch with the number of available events */
+    max_batch = MIN_NON_ZERO(MAX_EVENTS - s->io_q.in_flight, max_batch);
 
     switch (type) {
     case QEMU_AIO_WRITE:
@@ -371,7 +378,7 @@ static int laio_do_submit(int fd, struct qemu_laiocb *laiocb, off_t offset,
     s->io_q.in_queue++;
     if (!s->io_q.blocked &&
         (!s->io_q.plugged ||
-         s->io_q.in_flight + s->io_q.in_queue >= MAX_EVENTS)) {
+         s->io_q.in_queue >= max_batch)) {
         ioq_submit(s);
     }
 
diff --git a/block/mirror.c b/block/mirror.c
index 019f6deaa5..98fc66eabf 100644
--- a/block/mirror.c
+++ b/block/mirror.c
@@ -107,6 +107,7 @@ struct MirrorOp {
     bool is_in_flight;
     CoQueue waiting_requests;
     Coroutine *co;
+    MirrorOp *waiting_for_op;
 
     QTAILQ_ENTRY(MirrorOp) next;
 };
@@ -159,7 +160,18 @@ static void coroutine_fn mirror_wait_on_conflicts(MirrorOp *self,
             if (ranges_overlap(self_start_chunk, self_nb_chunks,
                                op_start_chunk, op_nb_chunks))
             {
+                /*
+                 * If the operation is already (indirectly) waiting for us, or
+                 * will wait for us as soon as it wakes up, then just go on
+                 * (instead of producing a deadlock in the former case).
+                 */
+                if (op->waiting_for_op) {
+                    continue;
+                }
+
+                self->waiting_for_op = op;
                 qemu_co_queue_wait(&op->waiting_requests, NULL);
+                self->waiting_for_op = NULL;
                 break;
             }
         }
@@ -1343,6 +1355,7 @@ static MirrorOp *coroutine_fn active_write_prepare(MirrorBlockJob *s,
         .bytes              = bytes,
         .is_active_write    = true,
         .is_in_flight       = true,
+        .co                 = qemu_coroutine_self(),
     };
     qemu_co_queue_init(&op->waiting_requests);
     QTAILQ_INSERT_TAIL(&s->ops_in_flight, op, next);
diff --git a/block/replication.c b/block/replication.c
index 774e15df16..32444b9a8f 100644
--- a/block/replication.c
+++ b/block/replication.c
@@ -35,7 +35,6 @@ typedef enum {
 typedef struct BDRVReplicationState {
     ReplicationMode mode;
     ReplicationStage stage;
-    BdrvChild *active_disk;
     BlockJob *commit_job;
     BdrvChild *hidden_disk;
     BdrvChild *secondary_disk;
@@ -166,7 +165,12 @@ static void replication_child_perm(BlockDriverState *bs, BdrvChild *c,
                                    uint64_t perm, uint64_t shared,
                                    uint64_t *nperm, uint64_t *nshared)
 {
-    *nperm = BLK_PERM_CONSISTENT_READ;
+    if (role & BDRV_CHILD_PRIMARY) {
+        *nperm = BLK_PERM_CONSISTENT_READ;
+    } else {
+        *nperm = 0;
+    }
+
     if ((bs->open_flags & (BDRV_O_INACTIVE | BDRV_O_RDWR)) == BDRV_O_RDWR) {
         *nperm |= BLK_PERM_WRITE;
     }
@@ -307,8 +311,10 @@ out:
     return ret;
 }
 
-static void secondary_do_checkpoint(BDRVReplicationState *s, Error **errp)
+static void secondary_do_checkpoint(BlockDriverState *bs, Error **errp)
 {
+    BDRVReplicationState *s = bs->opaque;
+    BdrvChild *active_disk = bs->file;
     Error *local_err = NULL;
     int ret;
 
@@ -323,13 +329,13 @@ static void secondary_do_checkpoint(BDRVReplicationState *s, Error **errp)
         return;
     }
 
-    if (!s->active_disk->bs->drv) {
+    if (!active_disk->bs->drv) {
         error_setg(errp, "Active disk %s is ejected",
-                   s->active_disk->bs->node_name);
+                   active_disk->bs->node_name);
         return;
     }
 
-    ret = bdrv_make_empty(s->active_disk, errp);
+    ret = bdrv_make_empty(active_disk, errp);
     if (ret < 0) {
         return;
     }
@@ -340,17 +346,7 @@ static void secondary_do_checkpoint(BDRVReplicationState *s, Error **errp)
         return;
     }
 
-    BlockBackend *blk = blk_new(qemu_get_current_aio_context(),
-                                BLK_PERM_WRITE, BLK_PERM_ALL);
-    blk_insert_bs(blk, s->hidden_disk->bs, &local_err);
-    if (local_err) {
-        error_propagate(errp, local_err);
-        blk_unref(blk);
-        return;
-    }
-
-    ret = blk_make_empty(blk, errp);
-    blk_unref(blk);
+    ret = bdrv_make_empty(s->hidden_disk, errp);
     if (ret < 0) {
         return;
     }
@@ -365,27 +361,35 @@ static void reopen_backing_file(BlockDriverState *bs, bool writable,
                                 Error **errp)
 {
     BDRVReplicationState *s = bs->opaque;
+    BdrvChild *hidden_disk, *secondary_disk;
     BlockReopenQueue *reopen_queue = NULL;
 
+    /*
+     * s->hidden_disk and s->secondary_disk may not be set yet, as they will
+     * only be set after the children are writable.
+     */
+    hidden_disk = bs->file->bs->backing;
+    secondary_disk = hidden_disk->bs->backing;
+
     if (writable) {
-        s->orig_hidden_read_only = bdrv_is_read_only(s->hidden_disk->bs);
-        s->orig_secondary_read_only = bdrv_is_read_only(s->secondary_disk->bs);
+        s->orig_hidden_read_only = bdrv_is_read_only(hidden_disk->bs);
+        s->orig_secondary_read_only = bdrv_is_read_only(secondary_disk->bs);
     }
 
-    bdrv_subtree_drained_begin(s->hidden_disk->bs);
-    bdrv_subtree_drained_begin(s->secondary_disk->bs);
+    bdrv_subtree_drained_begin(hidden_disk->bs);
+    bdrv_subtree_drained_begin(secondary_disk->bs);
 
     if (s->orig_hidden_read_only) {
         QDict *opts = qdict_new();
         qdict_put_bool(opts, BDRV_OPT_READ_ONLY, !writable);
-        reopen_queue = bdrv_reopen_queue(reopen_queue, s->hidden_disk->bs,
+        reopen_queue = bdrv_reopen_queue(reopen_queue, hidden_disk->bs,
                                          opts, true);
     }
 
     if (s->orig_secondary_read_only) {
         QDict *opts = qdict_new();
         qdict_put_bool(opts, BDRV_OPT_READ_ONLY, !writable);
-        reopen_queue = bdrv_reopen_queue(reopen_queue, s->secondary_disk->bs,
+        reopen_queue = bdrv_reopen_queue(reopen_queue, secondary_disk->bs,
                                          opts, true);
     }
 
@@ -400,8 +404,8 @@ static void reopen_backing_file(BlockDriverState *bs, bool writable,
         }
     }
 
-    bdrv_subtree_drained_end(s->hidden_disk->bs);
-    bdrv_subtree_drained_end(s->secondary_disk->bs);
+    bdrv_subtree_drained_end(hidden_disk->bs);
+    bdrv_subtree_drained_end(secondary_disk->bs);
 }
 
 static void backup_job_cleanup(BlockDriverState *bs)
@@ -458,6 +462,7 @@ static void replication_start(ReplicationState *rs, ReplicationMode mode,
     BlockDriverState *bs = rs->opaque;
     BDRVReplicationState *s;
     BlockDriverState *top_bs;
+    BdrvChild *active_disk, *hidden_disk, *secondary_disk;
     int64_t active_length, hidden_length, disk_length;
     AioContext *aio_context;
     Error *local_err = NULL;
@@ -495,32 +500,31 @@ static void replication_start(ReplicationState *rs, ReplicationMode mode,
     case REPLICATION_MODE_PRIMARY:
         break;
     case REPLICATION_MODE_SECONDARY:
-        s->active_disk = bs->file;
-        if (!s->active_disk || !s->active_disk->bs ||
-                                    !s->active_disk->bs->backing) {
+        active_disk = bs->file;
+        if (!active_disk || !active_disk->bs || !active_disk->bs->backing) {
             error_setg(errp, "Active disk doesn't have backing file");
             aio_context_release(aio_context);
             return;
         }
 
-        s->hidden_disk = s->active_disk->bs->backing;
-        if (!s->hidden_disk->bs || !s->hidden_disk->bs->backing) {
+        hidden_disk = active_disk->bs->backing;
+        if (!hidden_disk->bs || !hidden_disk->bs->backing) {
             error_setg(errp, "Hidden disk doesn't have backing file");
             aio_context_release(aio_context);
             return;
         }
 
-        s->secondary_disk = s->hidden_disk->bs->backing;
-        if (!s->secondary_disk->bs || !bdrv_has_blk(s->secondary_disk->bs)) {
+        secondary_disk = hidden_disk->bs->backing;
+        if (!secondary_disk->bs || !bdrv_has_blk(secondary_disk->bs)) {
             error_setg(errp, "The secondary disk doesn't have block backend");
             aio_context_release(aio_context);
             return;
         }
 
         /* verify the length */
-        active_length = bdrv_getlength(s->active_disk->bs);
-        hidden_length = bdrv_getlength(s->hidden_disk->bs);
-        disk_length = bdrv_getlength(s->secondary_disk->bs);
+        active_length = bdrv_getlength(active_disk->bs);
+        hidden_length = bdrv_getlength(hidden_disk->bs);
+        disk_length = bdrv_getlength(secondary_disk->bs);
         if (active_length < 0 || hidden_length < 0 || disk_length < 0 ||
             active_length != hidden_length || hidden_length != disk_length) {
             error_setg(errp, "Active disk, hidden disk, secondary disk's length"
@@ -530,10 +534,10 @@ static void replication_start(ReplicationState *rs, ReplicationMode mode,
         }
 
         /* Must be true, or the bdrv_getlength() calls would have failed */
-        assert(s->active_disk->bs->drv && s->hidden_disk->bs->drv);
+        assert(active_disk->bs->drv && hidden_disk->bs->drv);
 
-        if (!s->active_disk->bs->drv->bdrv_make_empty ||
-            !s->hidden_disk->bs->drv->bdrv_make_empty) {
+        if (!active_disk->bs->drv->bdrv_make_empty ||
+            !hidden_disk->bs->drv->bdrv_make_empty) {
             error_setg(errp,
                        "Active disk or hidden disk doesn't support make_empty");
             aio_context_release(aio_context);
@@ -548,6 +552,26 @@ static void replication_start(ReplicationState *rs, ReplicationMode mode,
             return;
         }
 
+        bdrv_ref(hidden_disk->bs);
+        s->hidden_disk = bdrv_attach_child(bs, hidden_disk->bs, "hidden disk",
+                                           &child_of_bds, BDRV_CHILD_DATA,
+                                           &local_err);
+        if (local_err) {
+            error_propagate(errp, local_err);
+            aio_context_release(aio_context);
+            return;
+        }
+
+        bdrv_ref(secondary_disk->bs);
+        s->secondary_disk = bdrv_attach_child(bs, secondary_disk->bs,
+                                              "secondary disk", &child_of_bds,
+                                              BDRV_CHILD_DATA, &local_err);
+        if (local_err) {
+            error_propagate(errp, local_err);
+            aio_context_release(aio_context);
+            return;
+        }
+
         /* start backup job now */
         error_setg(&s->blocker,
                    "Block device is in use by internal backup job");
@@ -586,7 +610,7 @@ static void replication_start(ReplicationState *rs, ReplicationMode mode,
     s->stage = BLOCK_REPLICATION_RUNNING;
 
     if (s->mode == REPLICATION_MODE_SECONDARY) {
-        secondary_do_checkpoint(s, errp);
+        secondary_do_checkpoint(bs, errp);
     }
 
     s->error = 0;
@@ -615,7 +639,7 @@ static void replication_do_checkpoint(ReplicationState *rs, Error **errp)
     }
 
     if (s->mode == REPLICATION_MODE_SECONDARY) {
-        secondary_do_checkpoint(s, errp);
+        secondary_do_checkpoint(bs, errp);
     }
     aio_context_release(aio_context);
 }
@@ -652,8 +676,9 @@ static void replication_done(void *opaque, int ret)
     if (ret == 0) {
         s->stage = BLOCK_REPLICATION_DONE;
 
-        s->active_disk = NULL;
+        bdrv_unref_child(bs, s->secondary_disk);
         s->secondary_disk = NULL;
+        bdrv_unref_child(bs, s->hidden_disk);
         s->hidden_disk = NULL;
         s->error = 0;
     } else {
@@ -705,7 +730,7 @@ static void replication_stop(ReplicationState *rs, bool failover, Error **errp)
         }
 
         if (!failover) {
-            secondary_do_checkpoint(s, errp);
+            secondary_do_checkpoint(bs, errp);
             s->stage = BLOCK_REPLICATION_DONE;
             aio_context_release(aio_context);
             return;
@@ -713,7 +738,7 @@ static void replication_stop(ReplicationState *rs, bool failover, Error **errp)
 
         s->stage = BLOCK_REPLICATION_FAILOVER;
         s->commit_job = commit_active_start(
-                            NULL, s->active_disk->bs, s->secondary_disk->bs,
+                            NULL, bs->file->bs, s->secondary_disk->bs,
                             JOB_INTERNAL, 0, BLOCKDEV_ON_ERROR_REPORT,
                             NULL, replication_done, bs, true, errp);
         break;
diff --git a/block/vvfat.c b/block/vvfat.c
index ae9d387da7..34bf1e3a86 100644
--- a/block/vvfat.c
+++ b/block/vvfat.c
@@ -3098,26 +3098,6 @@ static int coroutine_fn vvfat_co_block_status(BlockDriverState *bs,
     return BDRV_BLOCK_DATA;
 }
 
-static int coroutine_fn
-write_target_commit(BlockDriverState *bs, uint64_t offset, uint64_t bytes,
-                    QEMUIOVector *qiov, int flags)
-{
-    int ret;
-
-    BDRVVVFATState* s = *((BDRVVVFATState**) bs->opaque);
-    qemu_co_mutex_lock(&s->lock);
-    ret = try_commit(s);
-    qemu_co_mutex_unlock(&s->lock);
-
-    return ret;
-}
-
-static BlockDriver vvfat_write_target = {
-    .format_name        = "vvfat_write_target",
-    .instance_size      = sizeof(void*),
-    .bdrv_co_pwritev    = write_target_commit,
-};
-
 static void vvfat_qcow_options(BdrvChildRole role, bool parent_is_format,
                                int *child_flags, QDict *child_options,
                                int parent_flags, QDict *parent_options)
@@ -3133,7 +3113,6 @@ static int enable_write_target(BlockDriverState *bs, Error **errp)
 {
     BDRVVVFATState *s = bs->opaque;
     BlockDriver *bdrv_qcow = NULL;
-    BlockDriverState *backing;
     QemuOpts *opts = NULL;
     int ret;
     int size = sector2cluster(s, s->sector_count);
@@ -3184,13 +3163,6 @@ static int enable_write_target(BlockDriverState *bs, Error **errp)
     unlink(s->qcow_filename);
 #endif
 
-    backing = bdrv_new_open_driver(&vvfat_write_target, NULL, BDRV_O_ALLOW_RDWR,
-                                   &error_abort);
-    *(void**) backing->opaque = s;
-
-    bdrv_set_backing_hd(s->bs, backing, &error_abort);
-    bdrv_unref(backing);
-
     return 0;
 
 err:
@@ -3205,17 +3177,10 @@ static void vvfat_child_perm(BlockDriverState *bs, BdrvChild *c,
                              uint64_t perm, uint64_t shared,
                              uint64_t *nperm, uint64_t *nshared)
 {
-    if (role & BDRV_CHILD_DATA) {
-        /* This is a private node, nobody should try to attach to it */
-        *nperm = BLK_PERM_CONSISTENT_READ | BLK_PERM_WRITE;
-        *nshared = BLK_PERM_WRITE_UNCHANGED;
-    } else {
-        assert(role & BDRV_CHILD_COW);
-        /* The backing file is there so 'commit' can use it. vvfat doesn't
-         * access it in any way. */
-        *nperm = 0;
-        *nshared = BLK_PERM_ALL;
-    }
+    assert(role & BDRV_CHILD_DATA);
+    /* This is a private node, nobody should try to attach to it */
+    *nperm = BLK_PERM_CONSISTENT_READ | BLK_PERM_WRITE;
+    *nshared = BLK_PERM_WRITE_UNCHANGED;
 }
 
 static void vvfat_close(BlockDriverState *bs)
diff --git a/bsd-user/syscall.c b/bsd-user/syscall.c
index 7d986e9700..3f44311396 100644
--- a/bsd-user/syscall.c
+++ b/bsd-user/syscall.c
@@ -335,7 +335,7 @@ abi_long do_freebsd_syscall(void *cpu_env, int num, abi_long arg1,
         _mcleanup();
 #endif
         gdb_exit(arg1);
-        qemu_plugin_atexit_cb();
+        qemu_plugin_user_exit();
         /* XXX: should free thread stack and CPU env */
         _exit(arg1);
         ret = 0; /* avoid warning */
@@ -437,7 +437,7 @@ abi_long do_netbsd_syscall(void *cpu_env, int num, abi_long arg1,
         _mcleanup();
 #endif
         gdb_exit(arg1);
-        qemu_plugin_atexit_cb();
+        qemu_plugin_user_exit();
         /* XXX: should free thread stack and CPU env */
         _exit(arg1);
         ret = 0; /* avoid warning */
@@ -516,7 +516,7 @@ abi_long do_openbsd_syscall(void *cpu_env, int num, abi_long arg1,
         _mcleanup();
 #endif
         gdb_exit(arg1);
-        qemu_plugin_atexit_cb();
+        qemu_plugin_user_exit();
         /* XXX: should free thread stack and CPU env */
         _exit(arg1);
         ret = 0; /* avoid warning */
diff --git a/chardev/spice.c b/chardev/spice.c
index 3ffb3fdc0d..bbffef4913 100644
--- a/chardev/spice.c
+++ b/chardev/spice.c
@@ -382,6 +382,7 @@ static const TypeInfo char_spicevmc_type_info = {
     .parent = TYPE_CHARDEV_SPICE,
     .class_init = char_spicevmc_class_init,
 };
+module_obj(TYPE_CHARDEV_SPICEVMC);
 
 static void char_spiceport_class_init(ObjectClass *oc, void *data)
 {
diff --git a/configs/devices/mips64el-softmmu/default.mak b/configs/devices/mips64el-softmmu/default.mak
index c511a061ba..c610749ac1 100644
--- a/configs/devices/mips64el-softmmu/default.mak
+++ b/configs/devices/mips64el-softmmu/default.mak
@@ -8,8 +8,4 @@ CONFIG_ATI_VGA=y
 CONFIG_RTL8139_PCI=y
 CONFIG_JAZZ=y
 CONFIG_VT82C686=y
-CONFIG_AHCI=y
 CONFIG_MIPS_BOSTON=y
-CONFIG_FITLOADER=y
-CONFIG_PCI_EXPRESS=y
-CONFIG_PCI_EXPRESS_XILINX=y
diff --git a/configs/devices/ppc-softmmu/default.mak b/configs/devices/ppc-softmmu/default.mak
index 4535993d8d..658a454426 100644
--- a/configs/devices/ppc-softmmu/default.mak
+++ b/configs/devices/ppc-softmmu/default.mak
@@ -1,7 +1,6 @@
 # Default configuration for ppc-softmmu
 
 # For embedded PPCs:
-CONFIG_DS1338=y
 CONFIG_E500=y
 CONFIG_PPC405=y
 CONFIG_PPC440=y
diff --git a/configs/devices/tricore-softmmu/default.mak b/configs/devices/tricore-softmmu/default.mak
index 5cc91cebce..cb8fc286eb 100644
--- a/configs/devices/tricore-softmmu/default.mak
+++ b/configs/devices/tricore-softmmu/default.mak
@@ -1 +1,2 @@
+CONFIG_TRICORE_TESTBOARD=y
 CONFIG_TRIBOARD=y
diff --git a/configure b/configure
index 63f38fa94c..69cef68861 100755
--- a/configure
+++ b/configure
@@ -243,6 +243,7 @@ cross_prefix=""
 audio_drv_list=""
 block_drv_rw_whitelist=""
 block_drv_ro_whitelist=""
+block_drv_whitelist_tools="no"
 host_cc="cc"
 audio_win_int=""
 libs_qga=""
@@ -304,14 +305,14 @@ virtiofsd="auto"
 virtfs="auto"
 libudev="auto"
 mpath="auto"
-vnc="enabled"
+vnc="auto"
 sparse="auto"
 vde="$default_feature"
 vnc_sasl="auto"
 vnc_jpeg="auto"
 vnc_png="auto"
 xkbcommon="auto"
-xen="$default_feature"
+xen=${default_feature:+disabled}
 xen_ctrl_version="$default_feature"
 xen_pci_passthrough="auto"
 linux_aio="$default_feature"
@@ -321,6 +322,7 @@ attr="auto"
 xfs="$default_feature"
 tcg="enabled"
 membarrier="$default_feature"
+vhost_kernel="$default_feature"
 vhost_net="$default_feature"
 vhost_crypto="$default_feature"
 vhost_scsi="$default_feature"
@@ -328,6 +330,7 @@ vhost_vsock="$default_feature"
 vhost_user="no"
 vhost_user_blk_server="auto"
 vhost_user_fs="$default_feature"
+vhost_vdpa="$default_feature"
 bpf="auto"
 kvm="auto"
 hax="auto"
@@ -1016,6 +1019,10 @@ for opt do
   ;;
   --block-drv-ro-whitelist=*) block_drv_ro_whitelist=$(echo "$optarg" | sed -e 's/,/ /g')
   ;;
+  --enable-block-drv-whitelist-in-tools) block_drv_whitelist_tools="yes"
+  ;;
+  --disable-block-drv-whitelist-in-tools) block_drv_whitelist_tools="no"
+  ;;
   --enable-debug-tcg) debug_tcg="yes"
   ;;
   --disable-debug-tcg) debug_tcg="no"
@@ -1800,10 +1807,12 @@ Advanced options (experts only):
   --block-drv-whitelist=L  Same as --block-drv-rw-whitelist=L
   --block-drv-rw-whitelist=L
                            set block driver read-write whitelist
-                           (affects only QEMU, not qemu-img)
+                           (by default affects only QEMU, not tools like qemu-img)
   --block-drv-ro-whitelist=L
                            set block driver read-only whitelist
-                           (affects only QEMU, not qemu-img)
+                           (by default affects only QEMU, not tools like qemu-img)
+  --enable-block-drv-whitelist-in-tools
+                           use block whitelist also in tools instead of only QEMU
   --enable-trace-backends=B Set trace backend
                            Available backends: $trace_backend_list
   --with-trace-file=NAME   Full PATH,NAME of file to store traces
@@ -3266,18 +3275,6 @@ if ! compile_prog "$glib_cflags" "$glib_libs" ; then
 	       "build target"
 fi
 
-# Silence clang 3.5.0 warnings about glib attribute __alloc_size__ usage
-cat > $TMPC << EOF
-#include <glib.h>
-int main(void) { return 0; }
-EOF
-if ! compile_prog "$glib_cflags -Werror" "$glib_libs" ; then
-    if cc_has_warning_flag "-Wno-unknown-attributes"; then
-        glib_cflags="-Wno-unknown-attributes $glib_cflags"
-        CONFIGURE_CFLAGS="-Wno-unknown-attributes $CONFIGURE_CFLAGS"
-    fi
-fi
-
 # Silence clang warnings triggered by glib < 2.57.2
 cat > $TMPC << EOF
 #include <glib.h>
@@ -3984,18 +3981,11 @@ cat > $TMPC << EOF
 int main(void)
 {
   uint64_t x = 0, y = 0;
-#ifdef __ATOMIC_RELAXED
   y = __atomic_load_n(&x, __ATOMIC_RELAXED);
   __atomic_store_n(&x, y, __ATOMIC_RELAXED);
   __atomic_compare_exchange_n(&x, &y, x, 0, __ATOMIC_RELAXED, __ATOMIC_RELAXED);
   __atomic_exchange_n(&x, y, __ATOMIC_RELAXED);
   __atomic_fetch_add(&x, y, __ATOMIC_RELAXED);
-#else
-  typedef char is_host64[sizeof(void *) >= sizeof(uint64_t) ? 1 : -1];
-  __sync_lock_test_and_set(&x, y);
-  __sync_val_compare_and_swap(&x, y, 0);
-  __sync_fetch_and_add(&x, y);
-#endif
   return 0;
 }
 EOF
@@ -4583,6 +4573,9 @@ if test "$audio_win_int" = "yes" ; then
 fi
 echo "CONFIG_BDRV_RW_WHITELIST=$block_drv_rw_whitelist" >> $config_host_mak
 echo "CONFIG_BDRV_RO_WHITELIST=$block_drv_ro_whitelist" >> $config_host_mak
+if test "$block_drv_whitelist_tools" = "yes" ; then
+  echo "CONFIG_BDRV_WHITELIST_TOOLS=y" >> $config_host_mak
+fi
 if test "$xfs" = "yes" ; then
   echo "CONFIG_XFS=y" >> $config_host_mak
 fi
@@ -5110,12 +5103,10 @@ if test "$skip_meson" = no; then
   echo "[properties]" >> $cross
 
   # unroll any custom device configs
-  if test -n "$device_archs"; then
-      for a in $device_archs; do
-          eval "c=\$devices_${a}"
-          echo "${a}-softmmu = '$c'" >> $cross
-      done
-  fi
+  for a in $device_archs; do
+      eval "c=\$devices_${a}"
+      echo "${a}-softmmu = '$c'" >> $cross
+  done
 
   test -z "$cxx" && echo "link_language = 'c'" >> $cross
   echo "[built-in options]" >> $cross
@@ -5218,7 +5209,7 @@ if test "$skip_meson" = no; then
         -Ddocs=$docs -Dsphinx_build=$sphinx_build -Dinstall_blobs=$blobs \
         -Dvhost_user_blk_server=$vhost_user_blk_server -Dmultiprocess=$multiprocess \
         -Dfuse=$fuse -Dfuse_lseek=$fuse_lseek -Dguest_agent_msi=$guest_agent_msi -Dbpf=$bpf\
-        $(if test "$default_features" = no; then echo "-Dauto_features=disabled"; fi) \
+        $(if test "$default_feature" = no; then echo "-Dauto_features=disabled"; fi) \
 	-Dtcg_interpreter=$tcg_interpreter \
         $cross_arg \
         "$PWD" "$source_path"
diff --git a/contrib/gitdm/aliases b/contrib/gitdm/aliases
index c1e744312f..4792413ce7 100644
--- a/contrib/gitdm/aliases
+++ b/contrib/gitdm/aliases
@@ -31,6 +31,12 @@ pbrook@c046a42c-6fe2-441c-8c8c-71466251a162 paul@codesourcery.com
 ths@c046a42c-6fe2-441c-8c8c-71466251a162 ths@networkno.de
 malc@c046a42c-6fe2-441c-8c8c-71466251a162 av1474@comtv.ru
 
+# canonical emails
+liq3ea@163.com liq3ea@gmail.com
+
+# some broken tags
+yuval.shaia.ml.gmail.com yuval.shaia.ml@gmail.com
+
 # There is also a:
 #    (no author) <(no author)@c046a42c-6fe2-441c-8c8c-71466251a162>
 # for the cvs2svn initialization commit e63c3dc74bf.
diff --git a/contrib/gitdm/domain-map b/contrib/gitdm/domain-map
index 0074da618f..2800d9f986 100644
--- a/contrib/gitdm/domain-map
+++ b/contrib/gitdm/domain-map
@@ -9,6 +9,8 @@ baidu.com       Baidu
 bytedance.com   ByteDance
 cmss.chinamobile.com China Mobile
 citrix.com      Citrix
+crudebyte.com   Crudebyte
+eldorado.org.br Instituto de Pesquisas Eldorado
 fujitsu.com     Fujitsu
 google.com      Google
 greensocs.com   GreenSocs
@@ -17,20 +19,25 @@ ibm.com         IBM
 igalia.com      Igalia
 intel.com       Intel
 linaro.org      Linaro
+lwn.net         LWN
 microsoft.com   Microsoft
+mvista.com      MontaVista
 nokia.com       Nokia
 nuviainc.com    NUVIA
+nvidia.com      NVIDIA
 oracle.com      Oracle
 proxmox.com     Proxmox
 quicinc.com     Qualcomm Innovation Center
 redhat.com      Red Hat
 rt-rk.com       RT-RK
+samsung.com     Samsung
 siemens.com     Siemens
 sifive.com      SiFive
 suse.com        SUSE
 suse.de         SUSE
 virtuozzo.com   Virtuozzo
 wdc.com         Western Digital
+windriver.com   Wind River
 xilinx.com      Xilinx
 yadro.com       YADRO
 yandex-team.ru  Yandex
diff --git a/contrib/gitdm/group-map-academics b/contrib/gitdm/group-map-academics
index bf3c894821..44745ca85b 100644
--- a/contrib/gitdm/group-map-academics
+++ b/contrib/gitdm/group-map-academics
@@ -16,3 +16,6 @@ cota@braap.org
 uni-paderborn.de
 edu
 edu.cn
+
+# Boston University
+bu.edu
diff --git a/contrib/gitdm/group-map-individuals b/contrib/gitdm/group-map-individuals
index 36bbb77c39..f816aa8770 100644
--- a/contrib/gitdm/group-map-individuals
+++ b/contrib/gitdm/group-map-individuals
@@ -29,3 +29,8 @@ mrolnik@gmail.com
 huth@tuxfamily.org
 jhogan@kernel.org
 atar4qemu@gmail.com
+minwoo.im.dev@gmail.com
+bmeng.cn@gmail.com
+liq3ea@gmail.com
+chetan4windows@gmail.com
+akihiko.odaki@gmail.com
diff --git a/contrib/gitdm/group-map-interns b/contrib/gitdm/group-map-interns
new file mode 100644
index 0000000000..fe33a3231e
--- /dev/null
+++ b/contrib/gitdm/group-map-interns
@@ -0,0 +1,13 @@
+#
+# Group together everyone working as an intern via one of the various
+# outreach programs.
+#
+
+# GSoC 2020 Virtual FIDO/U2F security key
+cesar.belley@lse.epita.fr
+
+# GSoC 2020 TCG performance
+ahmedkhaledkaraman@gmail.com
+
+# GSoC 2021 TCG plugins
+ma.mandourr@gmail.com
diff --git a/contrib/gitdm/group-map-netflix b/contrib/gitdm/group-map-netflix
new file mode 100644
index 0000000000..468f95dcb2
--- /dev/null
+++ b/contrib/gitdm/group-map-netflix
@@ -0,0 +1,5 @@
+#
+# Netflix contributors using their personal emails
+#
+
+imp@bsdimp.com
diff --git a/contrib/gitdm/group-map-robots b/contrib/gitdm/group-map-robots
new file mode 100644
index 0000000000..ffd956c2eb
--- /dev/null
+++ b/contrib/gitdm/group-map-robots
@@ -0,0 +1,7 @@
+#
+# There are various automatic robots that occasionally scan and report
+# bugs. Let's group them together here.
+#
+
+# Euler Robot
+euler.robot@huawei.com
diff --git a/contrib/plugins/cache.c b/contrib/plugins/cache.c
index bf0d2f6097..066ea6d8ec 100644
--- a/contrib/plugins/cache.c
+++ b/contrib/plugins/cache.c
@@ -200,7 +200,7 @@ static void fifo_destroy(Cache *cache)
 {
     int i;
 
-    for (i = 0; i < cache->assoc; i++) {
+    for (i = 0; i < cache->num_sets; i++) {
         g_queue_free(cache->sets[i].fifo_queue);
     }
 }
@@ -355,15 +355,14 @@ static void vcpu_mem_access(unsigned int vcpu_index, qemu_plugin_meminfo_t info,
     struct qemu_plugin_hwaddr *hwaddr;
     InsnData *insn;
 
-    g_mutex_lock(&mtx);
     hwaddr = qemu_plugin_get_hwaddr(info, vaddr);
     if (hwaddr && qemu_plugin_hwaddr_is_io(hwaddr)) {
-        g_mutex_unlock(&mtx);
         return;
     }
 
     effective_addr = hwaddr ? qemu_plugin_hwaddr_phys_addr(hwaddr) : vaddr;
 
+    g_mutex_lock(&mtx);
     if (!access_cache(dcache, effective_addr)) {
         insn = (InsnData *) userdata;
         insn->dmisses++;
@@ -470,7 +469,7 @@ static int icmp(gconstpointer a, gconstpointer b)
     return insn_a->imisses < insn_b->imisses ? 1 : -1;
 }
 
-static void log_stats()
+static void log_stats(void)
 {
     g_autoptr(GString) rep = g_string_new("");
     g_string_append_printf(rep,
@@ -488,7 +487,7 @@ static void log_stats()
     qemu_plugin_outs(rep->str);
 }
 
-static void log_top_insns()
+static void log_top_insns(void)
 {
     int i;
     GList *curr, *miss_insns;
@@ -537,7 +536,7 @@ static void plugin_exit(qemu_plugin_id_t id, void *p)
     g_hash_table_destroy(miss_ht);
 }
 
-static void policy_init()
+static void policy_init(void)
 {
     switch (policy) {
     case LRU:
diff --git a/cpu.c b/cpu.c
index 83059537d7..e1799a15bc 100644
--- a/cpu.c
+++ b/cpu.c
@@ -225,11 +225,6 @@ void tb_invalidate_phys_addr(target_ulong addr)
     tb_invalidate_phys_page_range(addr, addr + 1);
     mmap_unlock();
 }
-
-static void breakpoint_invalidate(CPUState *cpu, target_ulong pc)
-{
-    tb_invalidate_phys_addr(pc);
-}
 #else
 void tb_invalidate_phys_addr(AddressSpace *as, hwaddr addr, MemTxAttrs attrs)
 {
@@ -250,25 +245,19 @@ void tb_invalidate_phys_addr(AddressSpace *as, hwaddr addr, MemTxAttrs attrs)
     ram_addr = memory_region_get_ram_addr(mr) + addr;
     tb_invalidate_phys_page_range(ram_addr, ram_addr + 1);
 }
-
-static void breakpoint_invalidate(CPUState *cpu, target_ulong pc)
-{
-    /*
-     * There may not be a virtual to physical translation for the pc
-     * right now, but there may exist cached TB for this pc.
-     * Flush the whole TB cache to force re-translation of such TBs.
-     * This is heavyweight, but we're debugging anyway.
-     */
-    tb_flush(cpu);
-}
 #endif
 
 /* Add a breakpoint.  */
 int cpu_breakpoint_insert(CPUState *cpu, vaddr pc, int flags,
                           CPUBreakpoint **breakpoint)
 {
+    CPUClass *cc = CPU_GET_CLASS(cpu);
     CPUBreakpoint *bp;
 
+    if (cc->gdb_adjust_breakpoint) {
+        pc = cc->gdb_adjust_breakpoint(cpu, pc);
+    }
+
     bp = g_malloc(sizeof(*bp));
 
     bp->pc = pc;
@@ -281,8 +270,6 @@ int cpu_breakpoint_insert(CPUState *cpu, vaddr pc, int flags,
         QTAILQ_INSERT_TAIL(&cpu->breakpoints, bp, entry);
     }
 
-    breakpoint_invalidate(cpu, pc);
-
     if (breakpoint) {
         *breakpoint = bp;
     }
@@ -294,8 +281,13 @@ int cpu_breakpoint_insert(CPUState *cpu, vaddr pc, int flags,
 /* Remove a specific breakpoint.  */
 int cpu_breakpoint_remove(CPUState *cpu, vaddr pc, int flags)
 {
+    CPUClass *cc = CPU_GET_CLASS(cpu);
     CPUBreakpoint *bp;
 
+    if (cc->gdb_adjust_breakpoint) {
+        pc = cc->gdb_adjust_breakpoint(cpu, pc);
+    }
+
     QTAILQ_FOREACH(bp, &cpu->breakpoints, entry) {
         if (bp->pc == pc && bp->flags == flags) {
             cpu_breakpoint_remove_by_ref(cpu, bp);
@@ -310,8 +302,6 @@ void cpu_breakpoint_remove_by_ref(CPUState *cpu, CPUBreakpoint *bp)
 {
     QTAILQ_REMOVE(&cpu->breakpoints, bp, entry);
 
-    breakpoint_invalidate(cpu, bp->pc);
-
     trace_breakpoint_remove(cpu->cpu_index, bp->pc, bp->flags);
     g_free(bp);
 }
@@ -336,10 +326,6 @@ void cpu_single_step(CPUState *cpu, int enabled)
         cpu->singlestep_enabled = enabled;
         if (kvm_enabled()) {
             kvm_update_guest_debug(cpu, 0);
-        } else {
-            /* must flush all the translated code to avoid inconsistencies */
-            /* XXX: only flush what is necessary */
-            tb_flush(cpu);
         }
         trace_breakpoint_singlestep(cpu->cpu_index, enabled);
     }
diff --git a/docs/interop/vhost-user.rst b/docs/interop/vhost-user.rst
index d6085f7045..7fc693521e 100644
--- a/docs/interop/vhost-user.rst
+++ b/docs/interop/vhost-user.rst
@@ -1,3 +1,5 @@
+.. _vhost_user_proto:
+
 ===================
 Vhost-user Protocol
 ===================
diff --git a/docs/system/device-emulation.rst b/docs/system/device-emulation.rst
new file mode 100644
index 0000000000..7afcfd8064
--- /dev/null
+++ b/docs/system/device-emulation.rst
@@ -0,0 +1,90 @@
+.. _device-emulation:
+
+Device Emulation
+----------------
+
+QEMU supports the emulation of a large number of devices from
+peripherals such network cards and USB devices to integrated systems
+on a chip (SoCs). Configuration of these is often a source of
+confusion so it helps to have an understanding of some of the terms
+used to describes devices within QEMU.
+
+Common Terms
+~~~~~~~~~~~~
+
+Device Front End
+================
+
+A device front end is how a device is presented to the guest. The type
+of device presented should match the hardware that the guest operating
+system is expecting to see. All devices can be specified with the
+``--device`` command line option. Running QEMU with the command line
+options ``--device help`` will list all devices it is aware of. Using
+the command line ``--device foo,help`` will list the additional
+configuration options available for that device.
+
+A front end is often paired with a back end, which describes how the
+host's resources are used in the emulation.
+
+Device Buses
+============
+
+Most devices will exist on a BUS of some sort. Depending on the
+machine model you choose (``-M foo``) a number of buses will have been
+automatically created. In most cases the BUS a device is attached to
+can be inferred, for example PCI devices are generally automatically
+allocated to the next free address of first PCI bus found. However in
+complicated configurations you can explicitly specify what bus
+(``bus=ID``) a device is attached to along with its address
+(``addr=N``).
+
+Some devices, for example a PCI SCSI host controller, will add an
+additional buses to the system that other devices can be attached to.
+A hypothetical chain of devices might look like:
+
+  --device foo,bus=pci.0,addr=0,id=foo
+  --device bar,bus=foo.0,addr=1,id=baz
+
+which would be a bar device (with the ID of baz) which is attached to
+the first foo bus (foo.0) at address 1. The foo device which provides
+that bus is itself is attached to the first PCI bus (pci.0).
+
+
+Device Back End
+===============
+
+The back end describes how the data from the emulated device will be
+processed by QEMU. The configuration of the back end is usually
+specific to the class of device being emulated. For example serial
+devices will be backed by a ``--chardev`` which can redirect the data
+to a file or socket or some other system. Storage devices are handled
+by ``--blockdev`` which will specify how blocks are handled, for
+example being stored in a qcow2 file or accessing a raw host disk
+partition. Back ends can sometimes be stacked to implement features
+like snapshots.
+
+While the choice of back end is generally transparent to the guest,
+there are cases where features will not be reported to the guest if
+the back end is unable to support it.
+
+Device Pass Through
+===================
+
+Device pass through is where the device is actually given access to
+the underlying hardware. This can be as simple as exposing a single
+USB device on the host system to the guest or dedicating a video card
+in a PCI slot to the exclusive use of the guest.
+
+
+Emulated Devices
+~~~~~~~~~~~~~~~~
+
+.. toctree::
+   :maxdepth: 1
+
+   devices/ivshmem.rst
+   devices/net.rst
+   devices/nvme.rst
+   devices/usb.rst
+   devices/vhost-user.rst
+   devices/virtio-pmem.rst
diff --git a/docs/system/ivshmem.rst b/docs/system/devices/ivshmem.rst
index b03a48afa3..b03a48afa3 100644
--- a/docs/system/ivshmem.rst
+++ b/docs/system/devices/ivshmem.rst
diff --git a/docs/system/net.rst b/docs/system/devices/net.rst
index 4b2640c448..4b2640c448 100644
--- a/docs/system/net.rst
+++ b/docs/system/devices/net.rst
diff --git a/docs/system/nvme.rst b/docs/system/devices/nvme.rst
index bff72d1c24..bff72d1c24 100644
--- a/docs/system/nvme.rst
+++ b/docs/system/devices/nvme.rst
diff --git a/docs/system/usb.rst b/docs/system/devices/usb.rst
index eeab78dcfb..eeab78dcfb 100644
--- a/docs/system/usb.rst
+++ b/docs/system/devices/usb.rst
diff --git a/docs/system/devices/vhost-user.rst b/docs/system/devices/vhost-user.rst
new file mode 100644
index 0000000000..86128114fa
--- /dev/null
+++ b/docs/system/devices/vhost-user.rst
@@ -0,0 +1,59 @@
+.. _vhost_user:
+
+vhost-user back ends
+--------------------
+
+vhost-user back ends are way to service the request of VirtIO devices
+outside of QEMU itself. To do this there are a number of things
+required.
+
+vhost-user device
+===================
+
+These are simple stub devices that ensure the VirtIO device is visible
+to the guest. The code is mostly boilerplate although each device has
+a ``chardev`` option which specifies the ID of the ``--chardev``
+device that connects via a socket to the vhost-user *daemon*.
+
+vhost-user daemon
+=================
+
+This is a separate process that is connected to by QEMU via a socket
+following the :ref:`vhost_user_proto`. There are a number of daemons
+that can be built when enabled by the project although any daemon that
+meets the specification for a given device can be used.
+
+Shared memory object
+====================
+
+In order for the daemon to access the VirtIO queues to process the
+requests it needs access to the guest's address space. This is
+achieved via the ``memory-backend-file`` or ``memory-backend-memfd``
+objects. A reference to a file-descriptor which can access this object
+will be passed via the socket as part of the protocol negotiation.
+
+Currently the shared memory object needs to match the size of the main
+system memory as defined by the ``-m`` argument.
+
+Example
+=======
+
+First start you daemon.
+
+.. parsed-literal::
+
+  $ virtio-foo --socket-path=/var/run/foo.sock $OTHER_ARGS
+
+The you start your QEMU instance specifying the device, chardev and
+memory objects.
+
+.. parsed-literal::
+
+  $ |qemu_system| \\
+      -m 4096 \\
+      -chardev socket,id=ba1,path=/var/run/foo.sock \\
+      -device vhost-user-foo,chardev=ba1,$OTHER_ARGS \\
+      -object memory-backend-memfd,id=mem,size=4G,share=on \\
+      -numa node,memdev=mem \\
+        ...
+
diff --git a/docs/system/virtio-pmem.rst b/docs/system/devices/virtio-pmem.rst
index c82ac06731..c82ac06731 100644
--- a/docs/system/virtio-pmem.rst
+++ b/docs/system/devices/virtio-pmem.rst
diff --git a/docs/system/index.rst b/docs/system/index.rst
index fda4b1b705..64a424ae99 100644
--- a/docs/system/index.rst
+++ b/docs/system/index.rst
@@ -11,15 +11,12 @@ or Hypervisor.Framework.
 
    quickstart
    invocation
+   device-emulation
    keys
    mux-chardev
    monitor
    images
-   net
    virtio-net-failover
-   usb
-   nvme
-   ivshmem
    linuxboot
    generic-loader
    guest-loader
@@ -30,7 +27,6 @@ or Hypervisor.Framework.
    gdb
    managed-startup
    cpu-hotplug
-   virtio-pmem
    pr-manager
    targets
    security
diff --git a/docs/tools/qemu-img.rst b/docs/tools/qemu-img.rst
index 1d8470eada..b7d602a288 100644
--- a/docs/tools/qemu-img.rst
+++ b/docs/tools/qemu-img.rst
@@ -414,7 +414,7 @@ Command description:
   4
     Error on reading data
 
-.. option:: convert [--object OBJECTDEF] [--image-opts] [--target-image-opts] [--target-is-zero] [--bitmaps] [-U] [-C] [-c] [-p] [-q] [-n] [-f FMT] [-t CACHE] [-T SRC_CACHE] [-O OUTPUT_FMT] [-B BACKING_FILE] [-o OPTIONS] [-l SNAPSHOT_PARAM] [-S SPARSE_SIZE] [-r RATE_LIMIT] [-m NUM_COROUTINES] [-W] FILENAME [FILENAME2 [...]] OUTPUT_FILENAME
+.. option:: convert [--object OBJECTDEF] [--image-opts] [--target-image-opts] [--target-is-zero] [--bitmaps [--skip-broken-bitmaps]] [-U] [-C] [-c] [-p] [-q] [-n] [-f FMT] [-t CACHE] [-T SRC_CACHE] [-O OUTPUT_FMT] [-B BACKING_FILE] [-o OPTIONS] [-l SNAPSHOT_PARAM] [-S SPARSE_SIZE] [-r RATE_LIMIT] [-m NUM_COROUTINES] [-W] FILENAME [FILENAME2 [...]] OUTPUT_FILENAME
 
   Convert the disk image *FILENAME* or a snapshot *SNAPSHOT_PARAM*
   to disk image *OUTPUT_FILENAME* using format *OUTPUT_FMT*. It can
@@ -456,6 +456,12 @@ Command description:
   *NUM_COROUTINES* specifies how many coroutines work in parallel during
   the convert process (defaults to 8).
 
+  Use of ``--bitmaps`` requests that any persistent bitmaps present in
+  the original are also copied to the destination.  If any bitmap is
+  inconsistent in the source, the conversion will fail unless
+  ``--skip-broken-bitmaps`` is also specified to copy only the
+  consistent bitmaps.
+
 .. option:: create [--object OBJECTDEF] [-q] [-f FMT] [-b BACKING_FILE] [-F BACKING_FMT] [-u] [-o OPTIONS] FILENAME [SIZE]
 
   Create the new disk image *FILENAME* of size *SIZE* and format
diff --git a/gitdm.config b/gitdm.config
index c01c219078..288b100d89 100644
--- a/gitdm.config
+++ b/gitdm.config
@@ -28,20 +28,25 @@ EmailMap contrib/gitdm/domain-map
 #
 # Use GroupMap to map a file full of addresses to the
 # same employer. This is used for people that don't post from easily
-# identifiable corporate emails.
+# identifiable corporate emails. Please keep this list sorted.
 #
 
-GroupMap contrib/gitdm/group-map-redhat Red Hat
-GroupMap contrib/gitdm/group-map-wavecomp Wave Computing
 GroupMap contrib/gitdm/group-map-cadence Cadence Design Systems
 GroupMap contrib/gitdm/group-map-codeweavers CodeWeavers
 GroupMap contrib/gitdm/group-map-ibm IBM
 GroupMap contrib/gitdm/group-map-janustech Janus Technologies
+GroupMap contrib/gitdm/group-map-netflix Netflix
+GroupMap contrib/gitdm/group-map-redhat Red Hat
+GroupMap contrib/gitdm/group-map-wavecomp Wave Computing
 
 # Also group together our prolific individual contributors
-# and those working under academic auspices
+# and those working under academic or intern auspices
 GroupMap contrib/gitdm/group-map-individuals (None)
 GroupMap contrib/gitdm/group-map-academics Academics (various)
+GroupMap contrib/gitdm/group-map-interns GSoC/Outreachy Interns
+
+# Group together robots and other auto-reporters
+GroupMap contrib/gitdm/group-map-robots Robots (various)
 
 #
 #
diff --git a/hw/acpi/Kconfig b/hw/acpi/Kconfig
index 1932f66af8..cfc4ede8d9 100644
--- a/hw/acpi/Kconfig
+++ b/hw/acpi/Kconfig
@@ -42,3 +42,7 @@ config ACPI_VMGENID
     depends on PC
 
 config ACPI_HW_REDUCED
+    bool
+    select ACPI
+    select ACPI_MEMORY_HOTPLUG
+    select ACPI_NVDIMM
diff --git a/hw/acpi/meson.build b/hw/acpi/meson.build
index 9b7fa75719..29f804d13e 100644
--- a/hw/acpi/meson.build
+++ b/hw/acpi/meson.build
@@ -3,6 +3,7 @@ acpi_ss.add(files(
   'acpi_interface.c',
   'aml-build.c',
   'bios-linker-loader.c',
+  'core.c',
   'utils.c',
 ))
 acpi_ss.add(when: 'CONFIG_ACPI_CPU_HOTPLUG', if_true: files('cpu.c'))
@@ -14,7 +15,7 @@ acpi_ss.add(when: 'CONFIG_ACPI_VMGENID', if_true: files('vmgenid.c'))
 acpi_ss.add(when: 'CONFIG_ACPI_HW_REDUCED', if_true: files('generic_event_device.c'))
 acpi_ss.add(when: 'CONFIG_ACPI_HMAT', if_true: files('hmat.c'))
 acpi_ss.add(when: 'CONFIG_ACPI_APEI', if_true: files('ghes.c'), if_false: files('ghes-stub.c'))
-acpi_ss.add(when: 'CONFIG_ACPI_X86', if_true: files('core.c', 'piix4.c', 'pcihp.c'), if_false: files('acpi-stub.c'))
+acpi_ss.add(when: 'CONFIG_ACPI_X86', if_true: files('piix4.c', 'pcihp.c'))
 acpi_ss.add(when: 'CONFIG_ACPI_X86_ICH', if_true: files('ich9.c', 'tco.c'))
 acpi_ss.add(when: 'CONFIG_IPMI', if_true: files('ipmi.c'), if_false: files('ipmi-stub.c'))
 acpi_ss.add(when: 'CONFIG_PC', if_false: files('acpi-x86-stub.c'))
diff --git a/hw/arm/Kconfig b/hw/arm/Kconfig
index 90b19c0861..4ba0aca067 100644
--- a/hw/arm/Kconfig
+++ b/hw/arm/Kconfig
@@ -388,6 +388,7 @@ config NPCM7XX
     select A9MPCORE
     select ADM1272
     select ARM_GIC
+    select SMBUS
     select AT24C  # EEPROM
     select MAX34451
     select PL310  # cache controller
@@ -403,7 +404,7 @@ config FSL_IMX25
     select IMX_FEC
     select IMX_I2C
     select WDT_IMX2
-    select DS1338
+    select SDHCI
 
 config FSL_IMX31
     bool
diff --git a/hw/char/Kconfig b/hw/char/Kconfig
index 4cf36ac637..2e4f620b13 100644
--- a/hw/char/Kconfig
+++ b/hw/char/Kconfig
@@ -61,6 +61,7 @@ config AVR_USART
 
 config MCHP_PFSOC_MMUART
     bool
+    select SERIAL
 
 config SIFIVE_UART
     bool
diff --git a/hw/display/qxl.c b/hw/display/qxl.c
index 84f99088e0..43482d4364 100644
--- a/hw/display/qxl.c
+++ b/hw/display/qxl.c
@@ -30,7 +30,6 @@
 #include "qemu/module.h"
 #include "hw/qdev-properties.h"
 #include "sysemu/runstate.h"
-#include "migration/blocker.h"
 #include "migration/vmstate.h"
 #include "trace.h"
 
@@ -666,30 +665,6 @@ static int interface_get_command(QXLInstance *sin, struct QXLCommandExt *ext)
         qxl->guest_primary.commands++;
         qxl_track_command(qxl, ext);
         qxl_log_command(qxl, "cmd", ext);
-        {
-            /*
-             * Windows 8 drivers place qxl commands in the vram
-             * (instead of the ram) bar.  We can't live migrate such a
-             * guest, so add a migration blocker in case we detect
-             * this, to avoid triggering the assert in pre_save().
-             *
-             * https://cgit.freedesktop.org/spice/win32/qxl-wddm-dod/commit/?id=f6e099db39e7d0787f294d5fd0dce328b5210faa
-             */
-            void *msg = qxl_phys2virt(qxl, ext->cmd.data, ext->group_id);
-            if (msg != NULL && (
-                    msg < (void *)qxl->vga.vram_ptr ||
-                    msg > ((void *)qxl->vga.vram_ptr + qxl->vga.vram_size))) {
-                if (!qxl->migration_blocker) {
-                    Error *local_err = NULL;
-                    error_setg(&qxl->migration_blocker,
-                               "qxl: guest bug: command not in ram bar");
-                    migrate_add_blocker(qxl->migration_blocker, &local_err);
-                    if (local_err) {
-                        error_report_err(local_err);
-                    }
-                }
-            }
-        }
         trace_qxl_ring_command_get(qxl->id, qxl_mode_to_string(qxl->mode));
         return true;
     default:
@@ -1283,12 +1258,6 @@ static void qxl_hard_reset(PCIQXLDevice *d, int loadvm)
     qemu_spice_create_host_memslot(&d->ssd);
     qxl_soft_reset(d);
 
-    if (d->migration_blocker) {
-        migrate_del_blocker(d->migration_blocker);
-        error_free(d->migration_blocker);
-        d->migration_blocker = NULL;
-    }
-
     if (startstop) {
         qemu_spice_display_start();
     }
@@ -2283,7 +2252,9 @@ static int qxl_pre_save(void *opaque)
     } else {
         d->last_release_offset = (uint8_t *)d->last_release - ram_start;
     }
-    assert(d->last_release_offset < d->vga.vram_size);
+    if (d->last_release_offset < d->vga.vram_size) {
+        return 1;
+    }
 
     return 0;
 }
diff --git a/hw/display/qxl.h b/hw/display/qxl.h
index 379d3304ab..30d21f4d0b 100644
--- a/hw/display/qxl.h
+++ b/hw/display/qxl.h
@@ -39,7 +39,6 @@ struct PCIQXLDevice {
     uint32_t           cmdlog;
 
     uint32_t           guest_bug;
-    Error              *migration_blocker;
 
     enum qxl_mode      mode;
     uint32_t           cmdflags;
diff --git a/hw/display/virtio-gpu-gl.c b/hw/display/virtio-gpu-gl.c
index 7ab93bf8c8..6cc4313b1a 100644
--- a/hw/display/virtio-gpu-gl.c
+++ b/hw/display/virtio-gpu-gl.c
@@ -51,12 +51,7 @@ static void virtio_gpu_gl_update_cursor_data(VirtIOGPU *g,
 static void virtio_gpu_gl_flushed(VirtIOGPUBase *b)
 {
     VirtIOGPU *g = VIRTIO_GPU(b);
-    VirtIOGPUGL *gl = VIRTIO_GPU_GL(b);
 
-    if (gl->renderer_reset) {
-        gl->renderer_reset = false;
-        virtio_gpu_virgl_reset(g);
-    }
     virtio_gpu_process_cmdq(g);
 }
 
@@ -74,6 +69,10 @@ static void virtio_gpu_gl_handle_ctrl(VirtIODevice *vdev, VirtQueue *vq)
         virtio_gpu_virgl_init(g);
         gl->renderer_inited = true;
     }
+    if (gl->renderer_reset) {
+        gl->renderer_reset = false;
+        virtio_gpu_virgl_reset(g);
+    }
 
     cmd = virtqueue_pop(vq, sizeof(struct virtio_gpu_ctrl_command));
     while (cmd) {
@@ -95,12 +94,13 @@ static void virtio_gpu_gl_reset(VirtIODevice *vdev)
 
     virtio_gpu_reset(vdev);
 
-    if (gl->renderer_inited) {
-        if (g->parent_obj.renderer_blocked) {
-            gl->renderer_reset = true;
-        } else {
-            virtio_gpu_virgl_reset(g);
-        }
+    /*
+     * GL functions must be called with the associated GL context in main
+     * thread, and when the renderer is unblocked.
+     */
+    if (gl->renderer_inited && !gl->renderer_reset) {
+        virtio_gpu_virgl_reset_scanout(g);
+        gl->renderer_reset = true;
     }
 }
 
@@ -113,6 +113,11 @@ static void virtio_gpu_gl_device_realize(DeviceState *qdev, Error **errp)
     return;
 #endif
 
+    if (!object_resolve_path_type("", TYPE_VIRTIO_GPU_GL, NULL)) {
+        error_setg(errp, "at most one %s device is permitted", TYPE_VIRTIO_GPU_GL);
+        return;
+    }
+
     if (!display_opengl) {
         error_setg(errp, "opengl is not available");
         return;
diff --git a/hw/display/virtio-gpu-virgl.c b/hw/display/virtio-gpu-virgl.c
index 092c6dc380..18d054922f 100644
--- a/hw/display/virtio-gpu-virgl.c
+++ b/hw/display/virtio-gpu-virgl.c
@@ -588,17 +588,21 @@ void virtio_gpu_virgl_fence_poll(VirtIOGPU *g)
     virtio_gpu_fence_poll(g);
 }
 
-void virtio_gpu_virgl_reset(VirtIOGPU *g)
+void virtio_gpu_virgl_reset_scanout(VirtIOGPU *g)
 {
     int i;
 
-    virgl_renderer_reset();
     for (i = 0; i < g->parent_obj.conf.max_outputs; i++) {
         dpy_gfx_replace_surface(g->parent_obj.scanout[i].con, NULL);
         dpy_gl_scanout_disable(g->parent_obj.scanout[i].con);
     }
 }
 
+void virtio_gpu_virgl_reset(VirtIOGPU *g)
+{
+    virgl_renderer_reset();
+}
+
 int virtio_gpu_virgl_init(VirtIOGPU *g)
 {
     int ret;
diff --git a/hw/display/virtio-gpu.c b/hw/display/virtio-gpu.c
index 6b7f643951..990e71fd40 100644
--- a/hw/display/virtio-gpu.c
+++ b/hw/display/virtio-gpu.c
@@ -340,37 +340,31 @@ static void virtio_gpu_resource_create_blob(VirtIOGPU *g,
         return;
     }
 
-    res = virtio_gpu_find_resource(g, cblob.resource_id);
-    if (res) {
-        qemu_log_mask(LOG_GUEST_ERROR, "%s: resource already exists %d\n",
-                      __func__, cblob.resource_id);
-        cmd->error = VIRTIO_GPU_RESP_ERR_INVALID_RESOURCE_ID;
-        return;
-    }
-
-    res = g_new0(struct virtio_gpu_simple_resource, 1);
-    res->resource_id = cblob.resource_id;
-    res->blob_size = cblob.size;
-
     if (cblob.blob_mem != VIRTIO_GPU_BLOB_MEM_GUEST &&
         cblob.blob_flags != VIRTIO_GPU_BLOB_FLAG_USE_SHAREABLE) {
         qemu_log_mask(LOG_GUEST_ERROR, "%s: invalid memory type\n",
                       __func__);
         cmd->error = VIRTIO_GPU_RESP_ERR_INVALID_PARAMETER;
-        g_free(res);
         return;
     }
 
-    if (res->iov) {
-        cmd->error = VIRTIO_GPU_RESP_ERR_UNSPEC;
+    if (virtio_gpu_find_resource(g, cblob.resource_id)) {
+        qemu_log_mask(LOG_GUEST_ERROR, "%s: resource already exists %d\n",
+                      __func__, cblob.resource_id);
+        cmd->error = VIRTIO_GPU_RESP_ERR_INVALID_RESOURCE_ID;
         return;
     }
 
+    res = g_new0(struct virtio_gpu_simple_resource, 1);
+    res->resource_id = cblob.resource_id;
+    res->blob_size = cblob.size;
+
     ret = virtio_gpu_create_mapping_iov(g, cblob.nr_entries, sizeof(cblob),
                                         cmd, &res->addrs, &res->iov,
                                         &res->iov_cnt);
-    if (ret != 0) {
+    if (ret != 0 || res->iov) {
         cmd->error = VIRTIO_GPU_RESP_ERR_UNSPEC;
+        g_free(res);
         return;
     }
 
diff --git a/hw/i386/Kconfig b/hw/i386/Kconfig
index bad6cf5b4e..ddedcef0b2 100644
--- a/hw/i386/Kconfig
+++ b/hw/i386/Kconfig
@@ -111,6 +111,7 @@ config MICROVM
     select ACPI_HW_REDUCED
     select PCI_EXPRESS_GENERIC_BRIDGE
     select USB_XHCI_SYSBUS
+    select I8254
 
 config X86_IOMMU
     bool
diff --git a/hw/ide/Kconfig b/hw/ide/Kconfig
index 8e2c893454..dd85fa3619 100644
--- a/hw/ide/Kconfig
+++ b/hw/ide/Kconfig
@@ -8,7 +8,7 @@ config IDE_QDEV
 config IDE_PCI
     bool
     depends on PCI
-    select IDE_CORE
+    select IDE_QDEV
 
 config IDE_ISA
     bool
diff --git a/hw/isa/Kconfig b/hw/isa/Kconfig
index 96db170eff..d42143a991 100644
--- a/hw/isa/Kconfig
+++ b/hw/isa/Kconfig
@@ -50,6 +50,11 @@ config VT82C686
     select FDC_ISA
     select USB_UHCI
     select APM
+    select I8254
+    select I8257
+    select I8259
+    select MC146818RTC
+    select PARALLEL
 
 config SMC37C669
     bool
diff --git a/hw/mips/Kconfig b/hw/mips/Kconfig
index c245e881a2..b4c5549ce8 100644
--- a/hw/mips/Kconfig
+++ b/hw/mips/Kconfig
@@ -47,9 +47,15 @@ config LOONGSON3V
 config MIPS_CPS
     bool
     select PTIMER
+    select MIPS_ITU
 
 config MIPS_BOSTON
     bool
+    select FITLOADER
+    select MIPS_CPS
+    select PCI_EXPRESS_XILINX
+    select AHCI_ICH9
+    select SERIAL
 
 config FW_CFG_MIPS
     bool
diff --git a/hw/pci-host/Kconfig b/hw/pci-host/Kconfig
index 84494400b8..2b5f7d58cc 100644
--- a/hw/pci-host/Kconfig
+++ b/hw/pci-host/Kconfig
@@ -76,3 +76,4 @@ config SH_PCI
 config MV64361
     bool
     select PCI
+    select I8259
diff --git a/hw/ppc/Kconfig b/hw/ppc/Kconfig
index 322a7eb031..400511c6b7 100644
--- a/hw/ppc/Kconfig
+++ b/hw/ppc/Kconfig
@@ -79,6 +79,7 @@ config PEGASOS2
     select VOF
 # This should come with VT82C686
     select ACPI_X86
+    imply ATI_VGA
 
 config PREP
     bool
@@ -130,6 +131,7 @@ config E500
     select SERIAL
     select MPC_I2C
     select FDT_PPC
+    select DS1338
 
 config VIRTEX
     bool
diff --git a/hw/remote/memory.c b/hw/remote/memory.c
index 472ed2a272..6e21ab1a45 100644
--- a/hw/remote/memory.c
+++ b/hw/remote/memory.c
@@ -46,7 +46,7 @@ void remote_sysmem_reconfig(MPQemuMsg *msg, Error **errp)
         subregion = g_new(MemoryRegion, 1);
         memory_region_init_ram_from_fd(subregion, NULL,
                                        name, sysmem_info->sizes[region],
-                                       true, msg->fds[region],
+                                       RAM_SHARED, msg->fds[region],
                                        sysmem_info->offsets[region],
                                        errp);
 
diff --git a/hw/riscv/Kconfig b/hw/riscv/Kconfig
index 86957ec7b0..0590f443fd 100644
--- a/hw/riscv/Kconfig
+++ b/hw/riscv/Kconfig
@@ -1,3 +1,6 @@
+config RISCV_NUMA
+    bool
+
 config IBEX
     bool
 
@@ -34,6 +37,7 @@ config RISCV_VIRT
     imply PCI_DEVICES
     imply VIRTIO_VGA
     imply TEST_DEVICES
+    select RISCV_NUMA
     select GOLDFISH_RTC
     select MSI_NONBROKEN
     select PCI
@@ -74,6 +78,7 @@ config SIFIVE_U
 
 config SPIKE
     bool
+    select RISCV_NUMA
     select HTIF
     select MSI_NONBROKEN
     select SIFIVE_CLINT
diff --git a/hw/riscv/meson.build b/hw/riscv/meson.build
index a97454661c..ab6cae57ea 100644
--- a/hw/riscv/meson.build
+++ b/hw/riscv/meson.build
@@ -1,6 +1,6 @@
 riscv_ss = ss.source_set()
 riscv_ss.add(files('boot.c'), fdt)
-riscv_ss.add(files('numa.c'))
+riscv_ss.add(when: 'CONFIG_RISCV_NUMA', if_true: files('numa.c'))
 riscv_ss.add(files('riscv_hart.c'))
 riscv_ss.add(when: 'CONFIG_OPENTITAN', if_true: files('opentitan.c'))
 riscv_ss.add(when: 'CONFIG_RISCV_VIRT', if_true: files('virt.c'))
diff --git a/hw/tricore/Kconfig b/hw/tricore/Kconfig
index 506e6183c1..33c1e852c3 100644
--- a/hw/tricore/Kconfig
+++ b/hw/tricore/Kconfig
@@ -1,9 +1,8 @@
-config TRICORE
+config TRICORE_TESTBOARD
     bool
 
 config TRIBOARD
     bool
-    select TRICORE
     select TC27X_SOC
 
 config TC27X_SOC
diff --git a/hw/tricore/meson.build b/hw/tricore/meson.build
index 47e36bb077..7e3585daf8 100644
--- a/hw/tricore/meson.build
+++ b/hw/tricore/meson.build
@@ -1,6 +1,6 @@
 tricore_ss = ss.source_set()
-tricore_ss.add(when: 'CONFIG_TRICORE', if_true: files('tricore_testboard.c'))
-tricore_ss.add(when: 'CONFIG_TRICORE', if_true: files('tricore_testdevice.c'))
+tricore_ss.add(when: 'CONFIG_TRICORE_TESTBOARD', if_true: files('tricore_testboard.c'))
+tricore_ss.add(when: 'CONFIG_TRICORE_TESTBOARD', if_true: files('tricore_testdevice.c'))
 tricore_ss.add(when: 'CONFIG_TRIBOARD', if_true: files('triboard.c'))
 tricore_ss.add(when: 'CONFIG_TC27X_SOC', if_true: files('tc27x_soc.c'))
 
diff --git a/hw/usb/meson.build b/hw/usb/meson.build
index 3ca6127937..de853d780d 100644
--- a/hw/usb/meson.build
+++ b/hw/usb/meson.build
@@ -72,7 +72,7 @@ if usbredir.found()
 endif
 
 # usb pass-through
-if config_host.has_key('CONFIG_USB_LIBUSB')
+if libusb.found()
   usbhost_ss = ss.source_set()
   usbhost_ss.add(when: ['CONFIG_USB', libusb],
                  if_true: files('host-libusb.c'))
diff --git a/include/block/aio.h b/include/block/aio.h
index 807edce9b5..47fbe9d81f 100644
--- a/include/block/aio.h
+++ b/include/block/aio.h
@@ -232,6 +232,9 @@ struct AioContext {
     int64_t poll_grow;      /* polling time growth factor */
     int64_t poll_shrink;    /* polling time shrink factor */
 
+    /* AIO engine parameters */
+    int64_t aio_max_batch;  /* maximum number of requests in a batch */
+
     /*
      * List of handlers participating in userspace polling.  Protected by
      * ctx->list_lock.  Iterated and modified mostly by the event loop thread
@@ -755,4 +758,13 @@ void aio_context_set_poll_params(AioContext *ctx, int64_t max_ns,
                                  int64_t grow, int64_t shrink,
                                  Error **errp);
 
+/**
+ * aio_context_set_aio_params:
+ * @ctx: the aio context
+ * @max_batch: maximum number of requests in a batch, 0 means that the
+ *             engine will use its default
+ */
+void aio_context_set_aio_params(AioContext *ctx, int64_t max_batch,
+                                Error **errp);
+
 #endif
diff --git a/include/exec/exec-all.h b/include/exec/exec-all.h
index 754f4130c9..5d1b6d80fb 100644
--- a/include/exec/exec-all.h
+++ b/include/exec/exec-all.h
@@ -492,13 +492,18 @@ struct TranslationBlock {
     target_ulong cs_base; /* CS base for this block */
     uint32_t flags; /* flags defining in which context the code was generated */
     uint32_t cflags;    /* compile flags */
-#define CF_COUNT_MASK  0x00007fff
-#define CF_LAST_IO     0x00008000 /* Last insn may be an IO access.  */
-#define CF_MEMI_ONLY   0x00010000 /* Only instrument memory ops */
-#define CF_USE_ICOUNT  0x00020000
-#define CF_INVALID     0x00040000 /* TB is stale. Set with @jmp_lock held */
-#define CF_PARALLEL    0x00080000 /* Generate code for a parallel context */
-#define CF_CLUSTER_MASK 0xff000000 /* Top 8 bits are cluster ID */
+
+/* Note that TCG_MAX_INSNS is 512; we validate this match elsewhere. */
+#define CF_COUNT_MASK    0x000001ff
+#define CF_NO_GOTO_TB    0x00000200 /* Do not chain with goto_tb */
+#define CF_NO_GOTO_PTR   0x00000400 /* Do not chain with goto_ptr */
+#define CF_SINGLE_STEP   0x00000800 /* gdbstub single-step in effect */
+#define CF_LAST_IO       0x00008000 /* Last insn may be an IO access.  */
+#define CF_MEMI_ONLY     0x00010000 /* Only instrument memory ops */
+#define CF_USE_ICOUNT    0x00020000
+#define CF_INVALID       0x00040000 /* TB is stale. Set with @jmp_lock held */
+#define CF_PARALLEL      0x00080000 /* Generate code for a parallel context */
+#define CF_CLUSTER_MASK  0xff000000 /* Top 8 bits are cluster ID */
 #define CF_CLUSTER_SHIFT 24
 
     /* Per-vCPU dynamic tracing state used to generate this TB */
@@ -563,10 +568,7 @@ static inline uint32_t tb_cflags(const TranslationBlock *tb)
 }
 
 /* current cflags for hashing/comparison */
-static inline uint32_t curr_cflags(CPUState *cpu)
-{
-    return cpu->tcg_cflags;
-}
+uint32_t curr_cflags(CPUState *cpu);
 
 /* TranslationBlock invalidate API */
 #if defined(CONFIG_USER_ONLY)
diff --git a/include/exec/translator.h b/include/exec/translator.h
index dd9c06d40d..d318803267 100644
--- a/include/exec/translator.h
+++ b/include/exec/translator.h
@@ -89,15 +89,6 @@ typedef struct DisasContextBase {
  * @insn_start:
  *      Emit the tcg_gen_insn_start opcode.
  *
- * @breakpoint_check:
- *      When called, the breakpoint has already been checked to match the PC,
- *      but the target may decide the breakpoint missed the address
- *      (e.g., due to conditions encoded in their flags).  Return true to
- *      indicate that the breakpoint did hit, in which case no more breakpoints
- *      are checked.  If the breakpoint did hit, emit any code required to
- *      signal the exception, and set db->is_jmp as necessary to terminate
- *      the main loop.
- *
  * @translate_insn:
  *      Disassemble one instruction and set db->pc_next for the start
  *      of the following instruction.  Set db->is_jmp as necessary to
@@ -113,8 +104,6 @@ typedef struct TranslatorOps {
     void (*init_disas_context)(DisasContextBase *db, CPUState *cpu);
     void (*tb_start)(DisasContextBase *db, CPUState *cpu);
     void (*insn_start)(DisasContextBase *db, CPUState *cpu);
-    bool (*breakpoint_check)(DisasContextBase *db, CPUState *cpu,
-                             const CPUBreakpoint *bp);
     void (*translate_insn)(DisasContextBase *db, CPUState *cpu);
     void (*tb_stop)(DisasContextBase *db, CPUState *cpu);
     void (*disas_log)(const DisasContextBase *db, CPUState *cpu);
diff --git a/include/hw/core/cpu.h b/include/hw/core/cpu.h
index 4e0ea68efc..bc864564ce 100644
--- a/include/hw/core/cpu.h
+++ b/include/hw/core/cpu.h
@@ -103,6 +103,9 @@ struct SysemuCPUOps;
  *       also implement the synchronize_from_tb hook.
  * @gdb_read_register: Callback for letting GDB read a register.
  * @gdb_write_register: Callback for letting GDB write a register.
+ * @gdb_adjust_breakpoint: Callback for adjusting the address of a
+ *       breakpoint.  Used by AVR to handle a gdb mis-feature with
+ *       its Harvard architecture split code and data.
  * @gdb_num_core_regs: Number of core registers accessible to GDB.
  * @gdb_core_xml_file: File name for core registers GDB XML description.
  * @gdb_stop_before_watchpoint: Indicates whether GDB expects the CPU to stop
@@ -137,6 +140,7 @@ struct CPUClass {
     void (*set_pc)(CPUState *cpu, vaddr value);
     int (*gdb_read_register)(CPUState *cpu, GByteArray *buf, int reg);
     int (*gdb_write_register)(CPUState *cpu, uint8_t *buf, int reg);
+    vaddr (*gdb_adjust_breakpoint)(CPUState *cpu, vaddr addr);
 
     const char *gdb_core_xml_file;
     gchar * (*gdb_arch_name)(CPUState *cpu);
diff --git a/include/hw/core/tcg-cpu-ops.h b/include/hw/core/tcg-cpu-ops.h
index 72d791438c..eab27d0c03 100644
--- a/include/hw/core/tcg-cpu-ops.h
+++ b/include/hw/core/tcg-cpu-ops.h
@@ -89,6 +89,12 @@ struct TCGCPUOps {
     bool (*debug_check_watchpoint)(CPUState *cpu, CPUWatchpoint *wp);
 
     /**
+     * @debug_check_breakpoint: return true if the architectural
+     * breakpoint whose PC has matched should really fire.
+     */
+    bool (*debug_check_breakpoint)(CPUState *cpu);
+
+    /**
      * @io_recompile_replay_branch: Callback for cpu_io_recompile.
      *
      * The cpu has been stopped, and cpu_restore_state_from_tb has been
diff --git a/include/hw/virtio/virtio-gpu.h b/include/hw/virtio/virtio-gpu.h
index bcf54d970f..24c6628944 100644
--- a/include/hw/virtio/virtio-gpu.h
+++ b/include/hw/virtio/virtio-gpu.h
@@ -279,6 +279,7 @@ int virtio_gpu_update_dmabuf(VirtIOGPU *g,
 void virtio_gpu_virgl_process_cmd(VirtIOGPU *g,
                                   struct virtio_gpu_ctrl_command *cmd);
 void virtio_gpu_virgl_fence_poll(VirtIOGPU *g);
+void virtio_gpu_virgl_reset_scanout(VirtIOGPU *g);
 void virtio_gpu_virgl_reset(VirtIOGPU *g);
 int virtio_gpu_virgl_init(VirtIOGPU *g);
 int virtio_gpu_virgl_get_num_capsets(VirtIOGPU *g);
diff --git a/include/qapi/forward-visitor.h b/include/qapi/forward-visitor.h
new file mode 100644
index 0000000000..50fb3e9d50
--- /dev/null
+++ b/include/qapi/forward-visitor.h
@@ -0,0 +1,27 @@
+/*
+ * Forwarding visitor
+ *
+ * Copyright Red Hat, Inc. 2021
+ *
+ * Author: Paolo Bonzini <pbonzini@redhat.com>
+ *
+ * This work is licensed under the terms of the GNU LGPL, version 2.1 or later.
+ * See the COPYING.LIB file in the top-level directory.
+ *
+ */
+
+#ifndef FORWARD_VISITOR_H
+#define FORWARD_VISITOR_H
+
+#include "qapi/visitor.h"
+
+typedef struct ForwardFieldVisitor ForwardFieldVisitor;
+
+/*
+ * The forwarding visitor only expects a single name, @from, to be passed for
+ * toplevel fields.  It is converted to @to and forwarded to the @target visitor.
+ * Calls within a struct are forwarded without changing the name.
+ */
+Visitor *visitor_forward_field(Visitor *target, const char *from, const char *to);
+
+#endif
diff --git a/include/qemu/atomic.h b/include/qemu/atomic.h
index 3ccf84fd46..112a29910b 100644
--- a/include/qemu/atomic.h
+++ b/include/qemu/atomic.h
@@ -60,8 +60,9 @@
         (unsigned short)1,                                                         \
       (expr)+0))))))
 
-#ifdef __ATOMIC_RELAXED
-/* For C11 atomic ops */
+#ifndef __ATOMIC_RELAXED
+#error "Expecting C11 atomic ops"
+#endif
 
 /* Manual memory barriers
  *
@@ -239,193 +240,8 @@
 #define qatomic_xor(ptr, n) \
     ((void) __atomic_fetch_xor(ptr, n, __ATOMIC_SEQ_CST))
 
-#else /* __ATOMIC_RELAXED */
-
-#ifdef __alpha__
-#define smp_read_barrier_depends()   asm volatile("mb":::"memory")
-#endif
-
-#if defined(__i386__) || defined(__x86_64__) || defined(__s390x__)
-
-/*
- * Because of the strongly ordered storage model, wmb() and rmb() are nops
- * here (a compiler barrier only).  QEMU doesn't do accesses to write-combining
- * qemu memory or non-temporal load/stores from C code.
- */
-#define smp_mb_release()   barrier()
-#define smp_mb_acquire()   barrier()
-
-/*
- * __sync_lock_test_and_set() is documented to be an acquire barrier only,
- * but it is a full barrier at the hardware level.  Add a compiler barrier
- * to make it a full barrier also at the compiler level.
- */
-#define qatomic_xchg(ptr, i)    (barrier(), __sync_lock_test_and_set(ptr, i))
-
-#elif defined(_ARCH_PPC)
-
-/*
- * We use an eieio() for wmb() on powerpc.  This assumes we don't
- * need to order cacheable and non-cacheable stores with respect to
- * each other.
- *
- * smp_mb has the same problem as on x86 for not-very-new GCC
- * (http://patchwork.ozlabs.org/patch/126184/, Nov 2011).
- */
-#define smp_wmb()          ({ asm volatile("eieio" ::: "memory"); (void)0; })
-#if defined(__powerpc64__)
-#define smp_mb_release()   ({ asm volatile("lwsync" ::: "memory"); (void)0; })
-#define smp_mb_acquire()   ({ asm volatile("lwsync" ::: "memory"); (void)0; })
-#else
-#define smp_mb_release()   ({ asm volatile("sync" ::: "memory"); (void)0; })
-#define smp_mb_acquire()   ({ asm volatile("sync" ::: "memory"); (void)0; })
-#endif
-#define smp_mb()           ({ asm volatile("sync" ::: "memory"); (void)0; })
-
-#endif /* _ARCH_PPC */
-
-/*
- * For (host) platforms we don't have explicit barrier definitions
- * for, we use the gcc __sync_synchronize() primitive to generate a
- * full barrier.  This should be safe on all platforms, though it may
- * be overkill for smp_mb_acquire() and smp_mb_release().
- */
-#ifndef smp_mb
-#define smp_mb()           __sync_synchronize()
-#endif
-
-#ifndef smp_mb_acquire
-#define smp_mb_acquire()   __sync_synchronize()
-#endif
-
-#ifndef smp_mb_release
-#define smp_mb_release()   __sync_synchronize()
-#endif
-
-#ifndef smp_read_barrier_depends
-#define smp_read_barrier_depends()   barrier()
-#endif
-
-#ifndef signal_barrier
-#define signal_barrier()    barrier()
-#endif
-
-/* These will only be atomic if the processor does the fetch or store
- * in a single issue memory operation
- */
-#define qatomic_read__nocheck(p)   (*(__typeof__(*(p)) volatile*) (p))
-#define qatomic_set__nocheck(p, i) ((*(__typeof__(*(p)) volatile*) (p)) = (i))
-
-#define qatomic_read(ptr)       qatomic_read__nocheck(ptr)
-#define qatomic_set(ptr, i)     qatomic_set__nocheck(ptr,i)
-
-/**
- * qatomic_rcu_read - reads a RCU-protected pointer to a local variable
- * into a RCU read-side critical section. The pointer can later be safely
- * dereferenced within the critical section.
- *
- * This ensures that the pointer copy is invariant thorough the whole critical
- * section.
- *
- * Inserts memory barriers on architectures that require them (currently only
- * Alpha) and documents which pointers are protected by RCU.
- *
- * qatomic_rcu_read also includes a compiler barrier to ensure that
- * value-speculative optimizations (e.g. VSS: Value Speculation
- * Scheduling) does not perform the data read before the pointer read
- * by speculating the value of the pointer.
- *
- * Should match qatomic_rcu_set(), qatomic_xchg(), qatomic_cmpxchg().
- */
-#define qatomic_rcu_read(ptr)    ({               \
-    typeof(*ptr) _val = qatomic_read(ptr);        \
-    smp_read_barrier_depends();                   \
-    _val;                                         \
-})
-
-/**
- * qatomic_rcu_set - assigns (publicizes) a pointer to a new data structure
- * meant to be read by RCU read-side critical sections.
- *
- * Documents which pointers will be dereferenced by RCU read-side critical
- * sections and adds the required memory barriers on architectures requiring
- * them. It also makes sure the compiler does not reorder code initializing the
- * data structure before its publication.
- *
- * Should match qatomic_rcu_read().
- */
-#define qatomic_rcu_set(ptr, i)  do {             \
-    smp_wmb();                                    \
-    qatomic_set(ptr, i);                          \
-} while (0)
-
-#define qatomic_load_acquire(ptr)    ({     \
-    typeof(*ptr) _val = qatomic_read(ptr);  \
-    smp_mb_acquire();                       \
-    _val;                                   \
-})
-
-#define qatomic_store_release(ptr, i)  do { \
-    smp_mb_release();                       \
-    qatomic_set(ptr, i);                    \
-} while (0)
-
-#ifndef qatomic_xchg
-#if defined(__clang__)
-#define qatomic_xchg(ptr, i)    __sync_swap(ptr, i)
-#else
-/* __sync_lock_test_and_set() is documented to be an acquire barrier only.  */
-#define qatomic_xchg(ptr, i)    (smp_mb(), __sync_lock_test_and_set(ptr, i))
-#endif
-#endif
-#define qatomic_xchg__nocheck  qatomic_xchg
-
-/* Provide shorter names for GCC atomic builtins.  */
-#define qatomic_fetch_inc(ptr)  __sync_fetch_and_add(ptr, 1)
-#define qatomic_fetch_dec(ptr)  __sync_fetch_and_add(ptr, -1)
-
-#define qatomic_fetch_add(ptr, n) __sync_fetch_and_add(ptr, n)
-#define qatomic_fetch_sub(ptr, n) __sync_fetch_and_sub(ptr, n)
-#define qatomic_fetch_and(ptr, n) __sync_fetch_and_and(ptr, n)
-#define qatomic_fetch_or(ptr, n) __sync_fetch_and_or(ptr, n)
-#define qatomic_fetch_xor(ptr, n) __sync_fetch_and_xor(ptr, n)
-
-#define qatomic_inc_fetch(ptr)  __sync_add_and_fetch(ptr, 1)
-#define qatomic_dec_fetch(ptr)  __sync_add_and_fetch(ptr, -1)
-#define qatomic_add_fetch(ptr, n) __sync_add_and_fetch(ptr, n)
-#define qatomic_sub_fetch(ptr, n) __sync_sub_and_fetch(ptr, n)
-#define qatomic_and_fetch(ptr, n) __sync_and_and_fetch(ptr, n)
-#define qatomic_or_fetch(ptr, n) __sync_or_and_fetch(ptr, n)
-#define qatomic_xor_fetch(ptr, n) __sync_xor_and_fetch(ptr, n)
-
-#define qatomic_cmpxchg(ptr, old, new) \
-    __sync_val_compare_and_swap(ptr, old, new)
-#define qatomic_cmpxchg__nocheck(ptr, old, new)  qatomic_cmpxchg(ptr, old, new)
-
-/* And even shorter names that return void.  */
-#define qatomic_inc(ptr)        ((void) __sync_fetch_and_add(ptr, 1))
-#define qatomic_dec(ptr)        ((void) __sync_fetch_and_add(ptr, -1))
-#define qatomic_add(ptr, n)     ((void) __sync_fetch_and_add(ptr, n))
-#define qatomic_sub(ptr, n)     ((void) __sync_fetch_and_sub(ptr, n))
-#define qatomic_and(ptr, n)     ((void) __sync_fetch_and_and(ptr, n))
-#define qatomic_or(ptr, n)      ((void) __sync_fetch_and_or(ptr, n))
-#define qatomic_xor(ptr, n)     ((void) __sync_fetch_and_xor(ptr, n))
-
-#endif /* __ATOMIC_RELAXED */
-
-#ifndef smp_wmb
 #define smp_wmb()   smp_mb_release()
-#endif
-#ifndef smp_rmb
 #define smp_rmb()   smp_mb_acquire()
-#endif
-
-/* This is more efficient than a store plus a fence.  */
-#if !defined(__SANITIZE_THREAD__)
-#if defined(__i386__) || defined(__x86_64__) || defined(__s390x__)
-#define qatomic_mb_set(ptr, i)  ((void)qatomic_xchg(ptr, i))
-#endif
-#endif
 
 /* qatomic_mb_read/set semantics map Java volatile variables. They are
  * less expensive on some platforms (notably POWER) than fully
@@ -435,16 +251,16 @@
  * use. See docs/devel/atomics.rst for more discussion.
  */
 
-#ifndef qatomic_mb_read
 #define qatomic_mb_read(ptr)                             \
     qatomic_load_acquire(ptr)
-#endif
 
-#ifndef qatomic_mb_set
-#define qatomic_mb_set(ptr, i)  do {                    \
-    qatomic_store_release(ptr, i);                      \
-    smp_mb();                                           \
-} while(0)
+#if !defined(__SANITIZE_THREAD__) && \
+    (defined(__i386__) || defined(__x86_64__) || defined(__s390x__))
+/* This is more efficient than a store plus a fence.  */
+# define qatomic_mb_set(ptr, i)  ((void)qatomic_xchg(ptr, i))
+#else
+# define qatomic_mb_set(ptr, i) \
+   ({ qatomic_store_release(ptr, i); smp_mb(); })
 #endif
 
 #define qatomic_fetch_inc_nonzero(ptr) ({                               \
@@ -455,28 +271,29 @@
     _oldn;                                                              \
 })
 
-/* Abstractions to access atomically (i.e. "once") i64/u64 variables */
-#ifdef CONFIG_ATOMIC64
-static inline int64_t qatomic_read_i64(const int64_t *ptr)
-{
-    /* use __nocheck because sizeof(void *) might be < sizeof(u64) */
-    return qatomic_read__nocheck(ptr);
-}
-
-static inline uint64_t qatomic_read_u64(const uint64_t *ptr)
-{
-    return qatomic_read__nocheck(ptr);
-}
-
-static inline void qatomic_set_i64(int64_t *ptr, int64_t val)
-{
-    qatomic_set__nocheck(ptr, val);
-}
+/*
+ * Abstractions to access atomically (i.e. "once") i64/u64 variables.
+ *
+ * The i386 abi is odd in that by default members are only aligned to
+ * 4 bytes, which means that 8-byte types can wind up mis-aligned.
+ * Clang will then warn about this, and emit a call into libatomic.
+ *
+ * Use of these types in structures when they will be used with atomic
+ * operations can avoid this.
+ */
+typedef int64_t aligned_int64_t __attribute__((aligned(8)));
+typedef uint64_t aligned_uint64_t __attribute__((aligned(8)));
 
-static inline void qatomic_set_u64(uint64_t *ptr, uint64_t val)
-{
-    qatomic_set__nocheck(ptr, val);
-}
+#ifdef CONFIG_ATOMIC64
+/* Use __nocheck because sizeof(void *) might be < sizeof(u64) */
+#define qatomic_read_i64(P) \
+    _Generic(*(P), int64_t: qatomic_read__nocheck(P))
+#define qatomic_read_u64(P) \
+    _Generic(*(P), uint64_t: qatomic_read__nocheck(P))
+#define qatomic_set_i64(P, V) \
+    _Generic(*(P), int64_t: qatomic_set__nocheck(P, V))
+#define qatomic_set_u64(P, V) \
+    _Generic(*(P), uint64_t: qatomic_set__nocheck(P, V))
 
 static inline void qatomic64_init(void)
 {
diff --git a/include/qemu/plugin.h b/include/qemu/plugin.h
index 0fefbc6084..9a8438f683 100644
--- a/include/qemu/plugin.h
+++ b/include/qemu/plugin.h
@@ -190,6 +190,16 @@ void qemu_plugin_add_dyn_cb_arr(GArray *arr);
 
 void qemu_plugin_disable_mem_helpers(CPUState *cpu);
 
+/**
+ * qemu_plugin_user_exit(): clean-up callbacks before calling exit callbacks
+ *
+ * This is a user-mode only helper that ensure we have fully cleared
+ * callbacks from all threads before calling the exit callbacks. This
+ * is so the plugins themselves don't have to jump through hoops to
+ * guard against race conditions.
+ */
+void qemu_plugin_user_exit(void);
+
 #else /* !CONFIG_PLUGIN */
 
 static inline void qemu_plugin_add_opts(void)
@@ -250,6 +260,8 @@ void qemu_plugin_add_dyn_cb_arr(GArray *arr)
 static inline void qemu_plugin_disable_mem_helpers(CPUState *cpu)
 { }
 
+static inline void qemu_plugin_user_exit(void)
+{ }
 #endif /* !CONFIG_PLUGIN */
 
 #endif /* QEMU_PLUGIN_H */
diff --git a/include/qemu/qemu-plugin.h b/include/qemu/qemu-plugin.h
index dc3496f36c..e6e815abc5 100644
--- a/include/qemu/qemu-plugin.h
+++ b/include/qemu/qemu-plugin.h
@@ -549,6 +549,19 @@ void qemu_plugin_vcpu_for_each(qemu_plugin_id_t id,
 void qemu_plugin_register_flush_cb(qemu_plugin_id_t id,
                                    qemu_plugin_simple_cb_t cb);
 
+/**
+ * qemu_plugin_register_atexit_cb() - register exit callback
+ * @id: plugin ID
+ * @cb: callback
+ * @userdata: user data for callback
+ *
+ * The @cb function is called once execution has finished. Plugins
+ * should be able to free all their resources at this point much like
+ * after a reset/uninstall callback is called.
+ *
+ * In user-mode it is possible a few un-instrumented instructions from
+ * child threads may run before the host kernel reaps the threads.
+ */
 void qemu_plugin_register_atexit_cb(qemu_plugin_id_t id,
                                     qemu_plugin_udata_cb_t cb, void *userdata);
 
diff --git a/include/qemu/stats64.h b/include/qemu/stats64.h
index fdd3d1b8f9..802402254b 100644
--- a/include/qemu/stats64.h
+++ b/include/qemu/stats64.h
@@ -21,7 +21,7 @@
 
 typedef struct Stat64 {
 #ifdef CONFIG_ATOMIC64
-    uint64_t value;
+    aligned_uint64_t value;
 #else
     uint32_t low, high;
     uint32_t lock;
diff --git a/include/sysemu/iothread.h b/include/sysemu/iothread.h
index f177142f16..7f714bd136 100644
--- a/include/sysemu/iothread.h
+++ b/include/sysemu/iothread.h
@@ -37,6 +37,9 @@ struct IOThread {
     int64_t poll_max_ns;
     int64_t poll_grow;
     int64_t poll_shrink;
+
+    /* AioContext AIO engine parameters */
+    int64_t aio_max_batch;
 };
 typedef struct IOThread IOThread;
 
diff --git a/include/tcg/tcg.h b/include/tcg/tcg.h
index 25dd19d6e1..44ccd86f3e 100644
--- a/include/tcg/tcg.h
+++ b/include/tcg/tcg.h
@@ -1341,31 +1341,32 @@ void helper_be_stq_mmu(CPUArchState *env, target_ulong addr, uint64_t val,
 # define helper_ret_stl_mmu   helper_le_stl_mmu
 # define helper_ret_stq_mmu   helper_le_stq_mmu
 #endif
+#endif /* CONFIG_SOFTMMU */
 
-uint32_t helper_atomic_cmpxchgb_mmu(CPUArchState *env, target_ulong addr,
+uint32_t cpu_atomic_cmpxchgb_mmu(CPUArchState *env, target_ulong addr,
+                                 uint32_t cmpv, uint32_t newv,
+                                 TCGMemOpIdx oi, uintptr_t retaddr);
+uint32_t cpu_atomic_cmpxchgw_le_mmu(CPUArchState *env, target_ulong addr,
+                                    uint32_t cmpv, uint32_t newv,
+                                    TCGMemOpIdx oi, uintptr_t retaddr);
+uint32_t cpu_atomic_cmpxchgl_le_mmu(CPUArchState *env, target_ulong addr,
                                     uint32_t cmpv, uint32_t newv,
                                     TCGMemOpIdx oi, uintptr_t retaddr);
-uint32_t helper_atomic_cmpxchgw_le_mmu(CPUArchState *env, target_ulong addr,
-                                       uint32_t cmpv, uint32_t newv,
-                                       TCGMemOpIdx oi, uintptr_t retaddr);
-uint32_t helper_atomic_cmpxchgl_le_mmu(CPUArchState *env, target_ulong addr,
-                                       uint32_t cmpv, uint32_t newv,
-                                       TCGMemOpIdx oi, uintptr_t retaddr);
-uint64_t helper_atomic_cmpxchgq_le_mmu(CPUArchState *env, target_ulong addr,
-                                       uint64_t cmpv, uint64_t newv,
-                                       TCGMemOpIdx oi, uintptr_t retaddr);
-uint32_t helper_atomic_cmpxchgw_be_mmu(CPUArchState *env, target_ulong addr,
-                                       uint32_t cmpv, uint32_t newv,
-                                       TCGMemOpIdx oi, uintptr_t retaddr);
-uint32_t helper_atomic_cmpxchgl_be_mmu(CPUArchState *env, target_ulong addr,
-                                       uint32_t cmpv, uint32_t newv,
-                                       TCGMemOpIdx oi, uintptr_t retaddr);
-uint64_t helper_atomic_cmpxchgq_be_mmu(CPUArchState *env, target_ulong addr,
-                                       uint64_t cmpv, uint64_t newv,
-                                       TCGMemOpIdx oi, uintptr_t retaddr);
+uint64_t cpu_atomic_cmpxchgq_le_mmu(CPUArchState *env, target_ulong addr,
+                                    uint64_t cmpv, uint64_t newv,
+                                    TCGMemOpIdx oi, uintptr_t retaddr);
+uint32_t cpu_atomic_cmpxchgw_be_mmu(CPUArchState *env, target_ulong addr,
+                                    uint32_t cmpv, uint32_t newv,
+                                    TCGMemOpIdx oi, uintptr_t retaddr);
+uint32_t cpu_atomic_cmpxchgl_be_mmu(CPUArchState *env, target_ulong addr,
+                                    uint32_t cmpv, uint32_t newv,
+                                    TCGMemOpIdx oi, uintptr_t retaddr);
+uint64_t cpu_atomic_cmpxchgq_be_mmu(CPUArchState *env, target_ulong addr,
+                                    uint64_t cmpv, uint64_t newv,
+                                    TCGMemOpIdx oi, uintptr_t retaddr);
 
 #define GEN_ATOMIC_HELPER(NAME, TYPE, SUFFIX)         \
-TYPE helper_atomic_ ## NAME ## SUFFIX ## _mmu         \
+TYPE cpu_atomic_ ## NAME ## SUFFIX ## _mmu            \
     (CPUArchState *env, target_ulong addr, TYPE val,  \
      TCGMemOpIdx oi, uintptr_t retaddr);
 
@@ -1411,31 +1412,22 @@ GEN_ATOMIC_HELPER_ALL(xchg)
 
 #undef GEN_ATOMIC_HELPER_ALL
 #undef GEN_ATOMIC_HELPER
-#endif /* CONFIG_SOFTMMU */
 
-/*
- * These aren't really a "proper" helpers because TCG cannot manage Int128.
- * However, use the same format as the others, for use by the backends.
- *
- * The cmpxchg functions are only defined if HAVE_CMPXCHG128;
- * the ld/st functions are only defined if HAVE_ATOMIC128,
- * as defined by <qemu/atomic128.h>.
- */
-Int128 helper_atomic_cmpxchgo_le_mmu(CPUArchState *env, target_ulong addr,
-                                     Int128 cmpv, Int128 newv,
-                                     TCGMemOpIdx oi, uintptr_t retaddr);
-Int128 helper_atomic_cmpxchgo_be_mmu(CPUArchState *env, target_ulong addr,
-                                     Int128 cmpv, Int128 newv,
-                                     TCGMemOpIdx oi, uintptr_t retaddr);
-
-Int128 helper_atomic_ldo_le_mmu(CPUArchState *env, target_ulong addr,
-                                TCGMemOpIdx oi, uintptr_t retaddr);
-Int128 helper_atomic_ldo_be_mmu(CPUArchState *env, target_ulong addr,
-                                TCGMemOpIdx oi, uintptr_t retaddr);
-void helper_atomic_sto_le_mmu(CPUArchState *env, target_ulong addr, Int128 val,
-                              TCGMemOpIdx oi, uintptr_t retaddr);
-void helper_atomic_sto_be_mmu(CPUArchState *env, target_ulong addr, Int128 val,
-                              TCGMemOpIdx oi, uintptr_t retaddr);
+Int128 cpu_atomic_cmpxchgo_le_mmu(CPUArchState *env, target_ulong addr,
+                                  Int128 cmpv, Int128 newv,
+                                  TCGMemOpIdx oi, uintptr_t retaddr);
+Int128 cpu_atomic_cmpxchgo_be_mmu(CPUArchState *env, target_ulong addr,
+                                  Int128 cmpv, Int128 newv,
+                                  TCGMemOpIdx oi, uintptr_t retaddr);
+
+Int128 cpu_atomic_ldo_le_mmu(CPUArchState *env, target_ulong addr,
+                             TCGMemOpIdx oi, uintptr_t retaddr);
+Int128 cpu_atomic_ldo_be_mmu(CPUArchState *env, target_ulong addr,
+                             TCGMemOpIdx oi, uintptr_t retaddr);
+void cpu_atomic_sto_le_mmu(CPUArchState *env, target_ulong addr, Int128 val,
+                           TCGMemOpIdx oi, uintptr_t retaddr);
+void cpu_atomic_sto_be_mmu(CPUArchState *env, target_ulong addr, Int128 val,
+                           TCGMemOpIdx oi, uintptr_t retaddr);
 
 #ifdef CONFIG_DEBUG_TCG
 void tcg_assert_listed_vecop(TCGOpcode);
diff --git a/iothread.c b/iothread.c
index 2c5ccd7367..ddbbde61f7 100644
--- a/iothread.c
+++ b/iothread.c
@@ -152,6 +152,24 @@ static void iothread_init_gcontext(IOThread *iothread)
     iothread->main_loop = g_main_loop_new(iothread->worker_context, TRUE);
 }
 
+static void iothread_set_aio_context_params(IOThread *iothread, Error **errp)
+{
+    ERRP_GUARD();
+
+    aio_context_set_poll_params(iothread->ctx,
+                                iothread->poll_max_ns,
+                                iothread->poll_grow,
+                                iothread->poll_shrink,
+                                errp);
+    if (*errp) {
+        return;
+    }
+
+    aio_context_set_aio_params(iothread->ctx,
+                               iothread->aio_max_batch,
+                               errp);
+}
+
 static void iothread_complete(UserCreatable *obj, Error **errp)
 {
     Error *local_error = NULL;
@@ -171,11 +189,7 @@ static void iothread_complete(UserCreatable *obj, Error **errp)
      */
     iothread_init_gcontext(iothread);
 
-    aio_context_set_poll_params(iothread->ctx,
-                                iothread->poll_max_ns,
-                                iothread->poll_grow,
-                                iothread->poll_shrink,
-                                &local_error);
+    iothread_set_aio_context_params(iothread, &local_error);
     if (local_error) {
         error_propagate(errp, local_error);
         aio_context_unref(iothread->ctx);
@@ -212,8 +226,11 @@ static PollParamInfo poll_grow_info = {
 static PollParamInfo poll_shrink_info = {
     "poll-shrink", offsetof(IOThread, poll_shrink),
 };
+static PollParamInfo aio_max_batch_info = {
+    "aio-max-batch", offsetof(IOThread, aio_max_batch),
+};
 
-static void iothread_get_poll_param(Object *obj, Visitor *v,
+static void iothread_get_param(Object *obj, Visitor *v,
         const char *name, void *opaque, Error **errp)
 {
     IOThread *iothread = IOTHREAD(obj);
@@ -223,7 +240,7 @@ static void iothread_get_poll_param(Object *obj, Visitor *v,
     visit_type_int64(v, name, field, errp);
 }
 
-static void iothread_set_poll_param(Object *obj, Visitor *v,
+static bool iothread_set_param(Object *obj, Visitor *v,
         const char *name, void *opaque, Error **errp)
 {
     IOThread *iothread = IOTHREAD(obj);
@@ -232,17 +249,36 @@ static void iothread_set_poll_param(Object *obj, Visitor *v,
     int64_t value;
 
     if (!visit_type_int64(v, name, &value, errp)) {
-        return;
+        return false;
     }
 
     if (value < 0) {
         error_setg(errp, "%s value must be in range [0, %" PRId64 "]",
                    info->name, INT64_MAX);
-        return;
+        return false;
     }
 
     *field = value;
 
+    return true;
+}
+
+static void iothread_get_poll_param(Object *obj, Visitor *v,
+        const char *name, void *opaque, Error **errp)
+{
+
+    iothread_get_param(obj, v, name, opaque, errp);
+}
+
+static void iothread_set_poll_param(Object *obj, Visitor *v,
+        const char *name, void *opaque, Error **errp)
+{
+    IOThread *iothread = IOTHREAD(obj);
+
+    if (!iothread_set_param(obj, v, name, opaque, errp)) {
+        return;
+    }
+
     if (iothread->ctx) {
         aio_context_set_poll_params(iothread->ctx,
                                     iothread->poll_max_ns,
@@ -252,6 +288,29 @@ static void iothread_set_poll_param(Object *obj, Visitor *v,
     }
 }
 
+static void iothread_get_aio_param(Object *obj, Visitor *v,
+        const char *name, void *opaque, Error **errp)
+{
+
+    iothread_get_param(obj, v, name, opaque, errp);
+}
+
+static void iothread_set_aio_param(Object *obj, Visitor *v,
+        const char *name, void *opaque, Error **errp)
+{
+    IOThread *iothread = IOTHREAD(obj);
+
+    if (!iothread_set_param(obj, v, name, opaque, errp)) {
+        return;
+    }
+
+    if (iothread->ctx) {
+        aio_context_set_aio_params(iothread->ctx,
+                                   iothread->aio_max_batch,
+                                   errp);
+    }
+}
+
 static void iothread_class_init(ObjectClass *klass, void *class_data)
 {
     UserCreatableClass *ucc = USER_CREATABLE_CLASS(klass);
@@ -269,6 +328,10 @@ static void iothread_class_init(ObjectClass *klass, void *class_data)
                               iothread_get_poll_param,
                               iothread_set_poll_param,
                               NULL, &poll_shrink_info);
+    object_class_property_add(klass, "aio-max-batch", "int",
+                              iothread_get_aio_param,
+                              iothread_set_aio_param,
+                              NULL, &aio_max_batch_info);
 }
 
 static const TypeInfo iothread_info = {
@@ -318,6 +381,7 @@ static int query_one_iothread(Object *object, void *opaque)
     info->poll_max_ns = iothread->poll_max_ns;
     info->poll_grow = iothread->poll_grow;
     info->poll_shrink = iothread->poll_shrink;
+    info->aio_max_batch = iothread->aio_max_batch;
 
     QAPI_LIST_APPEND(*tail, info);
     return 0;
diff --git a/linux-user/exit.c b/linux-user/exit.c
index 70b344048c..527e29cbc1 100644
--- a/linux-user/exit.c
+++ b/linux-user/exit.c
@@ -35,5 +35,5 @@ void preexit_cleanup(CPUArchState *env, int code)
         __gcov_dump();
 #endif
         gdb_exit(code);
-        qemu_plugin_atexit_cb();
+        qemu_plugin_user_exit();
 }
diff --git a/linux-user/hppa/cpu_loop.c b/linux-user/hppa/cpu_loop.c
index 3aaaf3337c..82d8183821 100644
--- a/linux-user/hppa/cpu_loop.c
+++ b/linux-user/hppa/cpu_loop.c
@@ -82,7 +82,7 @@ static abi_ulong hppa_lws(CPUHPPAState *env)
                 o64 = *(uint64_t *)g2h(cs, old);
                 n64 = *(uint64_t *)g2h(cs, new);
 #ifdef CONFIG_ATOMIC64
-                r64 = qatomic_cmpxchg__nocheck((uint64_t *)g2h(cs, addr),
+                r64 = qatomic_cmpxchg__nocheck((aligned_uint64_t *)g2h(cs, addr),
                                                o64, n64);
                 ret = r64 != o64;
 #else
diff --git a/meson.build b/meson.build
index 6e4d2d8034..f2e148eaf9 100644
--- a/meson.build
+++ b/meson.build
@@ -930,7 +930,7 @@ vnc = not_found
 png = not_found
 jpeg = not_found
 sasl = not_found
-if get_option('vnc').enabled()
+if not get_option('vnc').disabled()
   vnc = declare_dependency() # dummy dependency
   png = dependency('libpng', required: get_option('vnc_png'),
                    method: 'pkg-config', kwargs: static_kwargs)
@@ -2335,9 +2335,9 @@ foreach d, list : modules
         # https://github.com/mesonbuild/meson/pull/8900
         modinfo_files += custom_target(d + '-' + m + '.modinfo',
                                        output: d + '-' + m + '.modinfo',
-                                       input: module_ss.sources(),
+                                       input: module_ss.sources() + genh,
                                        capture: true,
-                                       command: [modinfo_collect, '@INPUT@'])
+                                       command: [modinfo_collect, module_ss.sources()])
       endif
     else
       if d == 'block'
@@ -2373,9 +2373,9 @@ foreach d, list : target_modules
             # FIXME: Should use sl.extract_all_objects(recursive: true) too.
             modinfo_files += custom_target(module_name + '.modinfo',
                                            output: module_name + '.modinfo',
-                                           input: target_module_ss.sources(),
+                                           input: target_module_ss.sources() + genh,
                                            capture: true,
-                                           command: [modinfo_collect, '--target', target, '@INPUT@'])
+                                           command: [modinfo_collect, '--target', target, target_module_ss.sources()])
           endif
         endif
       endforeach
@@ -2996,6 +2996,7 @@ summary_info += {'coroutine pool':    config_host['CONFIG_COROUTINE_POOL'] == '1
 if have_block
   summary_info += {'Block whitelist (rw)': config_host['CONFIG_BDRV_RW_WHITELIST']}
   summary_info += {'Block whitelist (ro)': config_host['CONFIG_BDRV_RO_WHITELIST']}
+  summary_info += {'Use block whitelist in tools': config_host.has_key('CONFIG_BDRV_WHITELIST_TOOLS')}
   summary_info += {'VirtFS support':    have_virtfs}
   summary_info += {'build virtiofs daemon': have_virtiofsd}
   summary_info += {'Live block migration': config_host.has_key('CONFIG_LIVE_BLOCK_MIGRATION')}
diff --git a/monitor/hmp-cmds.c b/monitor/hmp-cmds.c
index 0942027208..e00255f7ee 100644
--- a/monitor/hmp-cmds.c
+++ b/monitor/hmp-cmds.c
@@ -1893,6 +1893,8 @@ void hmp_info_iothreads(Monitor *mon, const QDict *qdict)
         monitor_printf(mon, "  poll-max-ns=%" PRId64 "\n", value->poll_max_ns);
         monitor_printf(mon, "  poll-grow=%" PRId64 "\n", value->poll_grow);
         monitor_printf(mon, "  poll-shrink=%" PRId64 "\n", value->poll_shrink);
+        monitor_printf(mon, "  aio-max-batch=%" PRId64 "\n",
+                       value->aio_max_batch);
     }
 
     qapi_free_IOThreadInfoList(info_list);
diff --git a/plugins/api.c b/plugins/api.c
index 78b563c5c5..2d521e6ba8 100644
--- a/plugins/api.c
+++ b/plugins/api.c
@@ -319,7 +319,7 @@ uint64_t qemu_plugin_hwaddr_phys_addr(const struct qemu_plugin_hwaddr *haddr)
             return block->offset + offset + block->mr->addr;
         } else {
             MemoryRegionSection *mrs = haddr->v.io.section;
-            return haddr->v.io.offset + mrs->mr->addr;
+            return mrs->offset_within_address_space + haddr->v.io.offset;
         }
     }
 #endif
diff --git a/plugins/core.c b/plugins/core.c
index e1bcdb570d..6b2490f973 100644
--- a/plugins/core.c
+++ b/plugins/core.c
@@ -27,7 +27,7 @@
 #include "exec/helper-proto.h"
 #include "tcg/tcg.h"
 #include "tcg/tcg-op.h"
-#include "trace/mem-internal.h" /* mem_info macros */
+#include "trace/mem.h" /* mem_info macros */
 #include "plugin.h"
 #include "qemu/compiler.h"
 
@@ -488,6 +488,45 @@ void qemu_plugin_register_atexit_cb(qemu_plugin_id_t id,
 }
 
 /*
+ * Handle exit from linux-user. Unlike the normal atexit() mechanism
+ * we need to handle the clean-up manually as it's possible threads
+ * are still running. We need to remove all callbacks from code
+ * generation, flush the current translations and then we can safely
+ * trigger the exit callbacks.
+ */
+
+void qemu_plugin_user_exit(void)
+{
+    enum qemu_plugin_event ev;
+    CPUState *cpu;
+
+    QEMU_LOCK_GUARD(&plugin.lock);
+
+    start_exclusive();
+
+    /* un-register all callbacks except the final AT_EXIT one */
+    for (ev = 0; ev < QEMU_PLUGIN_EV_MAX; ev++) {
+        if (ev != QEMU_PLUGIN_EV_ATEXIT) {
+            struct qemu_plugin_ctx *ctx;
+            QTAILQ_FOREACH(ctx, &plugin.ctxs, entry) {
+                plugin_unregister_cb__locked(ctx, ev);
+            }
+        }
+    }
+
+    tb_flush(current_cpu);
+
+    CPU_FOREACH(cpu) {
+        qemu_plugin_disable_mem_helpers(cpu);
+    }
+
+    end_exclusive();
+
+    /* now it's safe to handle the exit case */
+    qemu_plugin_atexit_cb();
+}
+
+/*
  * Call this function after longjmp'ing to the main loop. It's possible that the
  * last instruction of a TB might have used helpers, and therefore the
  * "disable" instruction will never execute because it ended up as dead code.
diff --git a/qapi/meson.build b/qapi/meson.build
index 376f4ceafe..c356a385e3 100644
--- a/qapi/meson.build
+++ b/qapi/meson.build
@@ -2,6 +2,7 @@ util_ss.add(files(
   'opts-visitor.c',
   'qapi-clone-visitor.c',
   'qapi-dealloc-visitor.c',
+  'qapi-forward-visitor.c',
   'qapi-util.c',
   'qapi-visit-core.c',
   'qobject-input-visitor.c',
diff --git a/qapi/misc.json b/qapi/misc.json
index 156f98203e..5c2ca3b556 100644
--- a/qapi/misc.json
+++ b/qapi/misc.json
@@ -86,6 +86,9 @@
 # @poll-shrink: how many ns will be removed from polling time, 0 means that
 #               it's not configured (since 2.9)
 #
+# @aio-max-batch: maximum number of requests in a batch for the AIO engine,
+#                 0 means that the engine will use its default (since 6.1)
+#
 # Since: 2.0
 ##
 { 'struct': 'IOThreadInfo',
@@ -93,7 +96,8 @@
            'thread-id': 'int',
            'poll-max-ns': 'int',
            'poll-grow': 'int',
-           'poll-shrink': 'int' } }
+           'poll-shrink': 'int',
+           'aio-max-batch': 'int' } }
 
 ##
 # @query-iothreads:
diff --git a/qapi/qapi-forward-visitor.c b/qapi/qapi-forward-visitor.c
new file mode 100644
index 0000000000..a4b111e22a
--- /dev/null
+++ b/qapi/qapi-forward-visitor.c
@@ -0,0 +1,326 @@
+/*
+ * Forward Visitor
+ *
+ * Copyright (C) 2021 Red Hat, Inc.
+ *
+ * This work is licensed under the terms of the GNU LGPL, version 2.1 or later.
+ * See the COPYING.LIB file in the top-level directory.
+ *
+ */
+
+#include "qemu/osdep.h"
+#include "qapi/compat-policy.h"
+#include "qapi/error.h"
+#include "qapi/forward-visitor.h"
+#include "qapi/visitor-impl.h"
+#include "qemu/queue.h"
+#include "qapi/qmp/qjson.h"
+#include "qapi/qmp/qbool.h"
+#include "qapi/qmp/qdict.h"
+#include "qapi/qmp/qerror.h"
+#include "qapi/qmp/qlist.h"
+#include "qapi/qmp/qnull.h"
+#include "qapi/qmp/qnum.h"
+#include "qapi/qmp/qstring.h"
+#include "qemu/cutils.h"
+#include "qemu/option.h"
+
+struct ForwardFieldVisitor {
+    Visitor visitor;
+
+    Visitor *target;
+    char *from;
+    char *to;
+
+    int depth;
+};
+
+static ForwardFieldVisitor *to_ffv(Visitor *v)
+{
+    return container_of(v, ForwardFieldVisitor, visitor);
+}
+
+static bool forward_field_translate_name(ForwardFieldVisitor *v, const char **name,
+                                         Error **errp)
+{
+    if (v->depth) {
+        return true;
+    }
+    if (g_str_equal(*name, v->from)) {
+        *name = v->to;
+        return true;
+    }
+    error_setg(errp, QERR_MISSING_PARAMETER, *name);
+    return false;
+}
+
+static bool forward_field_check_struct(Visitor *v, Error **errp)
+{
+    ForwardFieldVisitor *ffv = to_ffv(v);
+
+    return visit_check_struct(ffv->target, errp);
+}
+
+static bool forward_field_start_struct(Visitor *v, const char *name, void **obj,
+                                       size_t size, Error **errp)
+{
+    ForwardFieldVisitor *ffv = to_ffv(v);
+
+    if (!forward_field_translate_name(ffv, &name, errp)) {
+        return false;
+    }
+    if (!visit_start_struct(ffv->target, name, obj, size, errp)) {
+        return false;
+    }
+    ffv->depth++;
+    return true;
+}
+
+static void forward_field_end_struct(Visitor *v, void **obj)
+{
+    ForwardFieldVisitor *ffv = to_ffv(v);
+
+    assert(ffv->depth);
+    ffv->depth--;
+    visit_end_struct(ffv->target, obj);
+}
+
+static bool forward_field_start_list(Visitor *v, const char *name,
+                                     GenericList **list, size_t size,
+                                     Error **errp)
+{
+    ForwardFieldVisitor *ffv = to_ffv(v);
+
+    if (!forward_field_translate_name(ffv, &name, errp)) {
+        return false;
+    }
+    ffv->depth++;
+    return visit_start_list(ffv->target, name, list, size, errp);
+}
+
+static GenericList *forward_field_next_list(Visitor *v, GenericList *tail,
+                                            size_t size)
+{
+    ForwardFieldVisitor *ffv = to_ffv(v);
+
+    assert(ffv->depth);
+    return visit_next_list(ffv->target, tail, size);
+}
+
+static bool forward_field_check_list(Visitor *v, Error **errp)
+{
+    ForwardFieldVisitor *ffv = to_ffv(v);
+
+    assert(ffv->depth);
+    return visit_check_list(ffv->target, errp);
+}
+
+static void forward_field_end_list(Visitor *v, void **obj)
+{
+    ForwardFieldVisitor *ffv = to_ffv(v);
+
+    assert(ffv->depth);
+    ffv->depth--;
+    visit_end_list(ffv->target, obj);
+}
+
+static bool forward_field_start_alternate(Visitor *v, const char *name,
+                                          GenericAlternate **obj, size_t size,
+                                          Error **errp)
+{
+    ForwardFieldVisitor *ffv = to_ffv(v);
+
+    if (!forward_field_translate_name(ffv, &name, errp)) {
+        return false;
+    }
+    /*
+     * The name passed to start_alternate is used also in the visit_type_* calls
+     * that retrieve the alternate's content; so, do not increase depth here.
+     */
+    return visit_start_alternate(ffv->target, name, obj, size, errp);
+}
+
+static void forward_field_end_alternate(Visitor *v, void **obj)
+{
+    ForwardFieldVisitor *ffv = to_ffv(v);
+
+    visit_end_alternate(ffv->target, obj);
+}
+
+static bool forward_field_type_int64(Visitor *v, const char *name, int64_t *obj,
+                                     Error **errp)
+{
+    ForwardFieldVisitor *ffv = to_ffv(v);
+
+    if (!forward_field_translate_name(ffv, &name, errp)) {
+        return false;
+    }
+    return visit_type_int64(ffv->target, name, obj, errp);
+}
+
+static bool forward_field_type_uint64(Visitor *v, const char *name,
+                                      uint64_t *obj, Error **errp)
+{
+    ForwardFieldVisitor *ffv = to_ffv(v);
+
+    if (!forward_field_translate_name(ffv, &name, errp)) {
+        return false;
+    }
+    return visit_type_uint64(ffv->target, name, obj, errp);
+}
+
+static bool forward_field_type_bool(Visitor *v, const char *name, bool *obj,
+                                    Error **errp)
+{
+    ForwardFieldVisitor *ffv = to_ffv(v);
+
+    if (!forward_field_translate_name(ffv, &name, errp)) {
+        return false;
+    }
+    return visit_type_bool(ffv->target, name, obj, errp);
+}
+
+static bool forward_field_type_str(Visitor *v, const char *name, char **obj,
+                                   Error **errp)
+{
+    ForwardFieldVisitor *ffv = to_ffv(v);
+
+    if (!forward_field_translate_name(ffv, &name, errp)) {
+        return false;
+    }
+    return visit_type_str(ffv->target, name, obj, errp);
+}
+
+static bool forward_field_type_size(Visitor *v, const char *name, uint64_t *obj,
+                                    Error **errp)
+{
+    ForwardFieldVisitor *ffv = to_ffv(v);
+
+    if (!forward_field_translate_name(ffv, &name, errp)) {
+        return false;
+    }
+    return visit_type_size(ffv->target, name, obj, errp);
+}
+
+static bool forward_field_type_number(Visitor *v, const char *name, double *obj,
+                                      Error **errp)
+{
+    ForwardFieldVisitor *ffv = to_ffv(v);
+
+    if (!forward_field_translate_name(ffv, &name, errp)) {
+        return false;
+    }
+    return visit_type_number(ffv->target, name, obj, errp);
+}
+
+static bool forward_field_type_any(Visitor *v, const char *name, QObject **obj,
+                                   Error **errp)
+{
+    ForwardFieldVisitor *ffv = to_ffv(v);
+
+    if (!forward_field_translate_name(ffv, &name, errp)) {
+        return false;
+    }
+    return visit_type_any(ffv->target, name, obj, errp);
+}
+
+static bool forward_field_type_null(Visitor *v, const char *name,
+                                    QNull **obj, Error **errp)
+{
+    ForwardFieldVisitor *ffv = to_ffv(v);
+
+    if (!forward_field_translate_name(ffv, &name, errp)) {
+        return false;
+    }
+    return visit_type_null(ffv->target, name, obj, errp);
+}
+
+static void forward_field_optional(Visitor *v, const char *name, bool *present)
+{
+    ForwardFieldVisitor *ffv = to_ffv(v);
+
+    if (!forward_field_translate_name(ffv, &name, NULL)) {
+        *present = false;
+        return;
+    }
+    visit_optional(ffv->target, name, present);
+}
+
+static bool forward_field_deprecated_accept(Visitor *v, const char *name,
+                                            Error **errp)
+{
+    ForwardFieldVisitor *ffv = to_ffv(v);
+
+    if (!forward_field_translate_name(ffv, &name, errp)) {
+        return false;
+    }
+    return visit_deprecated_accept(ffv->target, name, errp);
+}
+
+static bool forward_field_deprecated(Visitor *v, const char *name)
+{
+    ForwardFieldVisitor *ffv = to_ffv(v);
+
+    if (!forward_field_translate_name(ffv, &name, NULL)) {
+        return false;
+    }
+    return visit_deprecated(ffv->target, name);
+}
+
+static void forward_field_complete(Visitor *v, void *opaque)
+{
+    /*
+     * Do nothing, the complete method will be called in due time
+     * on the target visitor.
+     */
+}
+
+static void forward_field_free(Visitor *v)
+{
+    ForwardFieldVisitor *ffv = to_ffv(v);
+
+    g_free(ffv->from);
+    g_free(ffv->to);
+    g_free(ffv);
+}
+
+Visitor *visitor_forward_field(Visitor *target, const char *from, const char *to)
+{
+    ForwardFieldVisitor *v = g_new0(ForwardFieldVisitor, 1);
+
+    /*
+     * Clone and dealloc visitors don't use a name for the toplevel
+     * visit, so they make no sense here.
+     */
+    assert(target->type == VISITOR_OUTPUT || target->type == VISITOR_INPUT);
+
+    v->visitor.type = target->type;
+    v->visitor.start_struct = forward_field_start_struct;
+    v->visitor.check_struct = forward_field_check_struct;
+    v->visitor.end_struct = forward_field_end_struct;
+    v->visitor.start_list = forward_field_start_list;
+    v->visitor.next_list = forward_field_next_list;
+    v->visitor.check_list = forward_field_check_list;
+    v->visitor.end_list = forward_field_end_list;
+    v->visitor.start_alternate = forward_field_start_alternate;
+    v->visitor.end_alternate = forward_field_end_alternate;
+    v->visitor.type_int64 = forward_field_type_int64;
+    v->visitor.type_uint64 = forward_field_type_uint64;
+    v->visitor.type_size = forward_field_type_size;
+    v->visitor.type_bool = forward_field_type_bool;
+    v->visitor.type_str = forward_field_type_str;
+    v->visitor.type_number = forward_field_type_number;
+    v->visitor.type_any = forward_field_type_any;
+    v->visitor.type_null = forward_field_type_null;
+    v->visitor.optional = forward_field_optional;
+    v->visitor.deprecated_accept = forward_field_deprecated_accept;
+    v->visitor.deprecated = forward_field_deprecated;
+    v->visitor.complete = forward_field_complete;
+    v->visitor.free = forward_field_free;
+
+    v->target = target;
+    v->from = g_strdup(from);
+    v->to = g_strdup(to);
+
+    return &v->visitor;
+}
diff --git a/qapi/qom.json b/qapi/qom.json
index 652be317b8..6d5f4a88e6 100644
--- a/qapi/qom.json
+++ b/qapi/qom.json
@@ -516,12 +516,17 @@
 #               algorithm detects it is spending too long polling without
 #               encountering events. 0 selects a default behaviour (default: 0)
 #
+# @aio-max-batch: maximum number of requests in a batch for the AIO engine,
+#                 0 means that the engine will use its default
+#                 (default:0, since 6.1)
+#
 # Since: 2.0
 ##
 { 'struct': 'IothreadProperties',
   'data': { '*poll-max-ns': 'int',
             '*poll-grow': 'int',
-            '*poll-shrink': 'int' } }
+            '*poll-shrink': 'int',
+            '*aio-max-batch': 'int' } }
 
 ##
 # @MemoryBackendProperties:
diff --git a/qemu-img.c b/qemu-img.c
index 797742a443..908fd0cce5 100644
--- a/qemu-img.c
+++ b/qemu-img.c
@@ -82,6 +82,7 @@ enum {
     OPTION_MERGE = 274,
     OPTION_BITMAPS = 275,
     OPTION_FORCE = 276,
+    OPTION_SKIP_BROKEN = 277,
 };
 
 typedef enum OutputFormat {
@@ -2101,7 +2102,32 @@ static int convert_do_copy(ImgConvertState *s)
     return s->ret;
 }
 
-static int convert_copy_bitmaps(BlockDriverState *src, BlockDriverState *dst)
+/* Check that bitmaps can be copied, or output an error */
+static int convert_check_bitmaps(BlockDriverState *src, bool skip_broken)
+{
+    BdrvDirtyBitmap *bm;
+
+    if (!bdrv_supports_persistent_dirty_bitmap(src)) {
+        error_report("Source lacks bitmap support");
+        return -1;
+    }
+    FOR_EACH_DIRTY_BITMAP(src, bm) {
+        if (!bdrv_dirty_bitmap_get_persistence(bm)) {
+            continue;
+        }
+        if (!skip_broken && bdrv_dirty_bitmap_inconsistent(bm)) {
+            error_report("Cannot copy inconsistent bitmap '%s'",
+                         bdrv_dirty_bitmap_name(bm));
+            error_printf("Try --skip-broken-bitmaps, or "
+                         "use 'qemu-img bitmap --remove' to delete it\n");
+            return -1;
+        }
+    }
+    return 0;
+}
+
+static int convert_copy_bitmaps(BlockDriverState *src, BlockDriverState *dst,
+                                bool skip_broken)
 {
     BdrvDirtyBitmap *bm;
     Error *err = NULL;
@@ -2113,6 +2139,10 @@ static int convert_copy_bitmaps(BlockDriverState *src, BlockDriverState *dst)
             continue;
         }
         name = bdrv_dirty_bitmap_name(bm);
+        if (skip_broken && bdrv_dirty_bitmap_inconsistent(bm)) {
+            warn_report("Skipping inconsistent bitmap '%s'", name);
+            continue;
+        }
         qmp_block_dirty_bitmap_add(dst->node_name, name,
                                    true, bdrv_dirty_bitmap_granularity(bm),
                                    true, true,
@@ -2127,6 +2157,7 @@ static int convert_copy_bitmaps(BlockDriverState *src, BlockDriverState *dst)
                               &err);
         if (err) {
             error_reportf_err(err, "Failed to populate bitmap %s: ", name);
+            qmp_block_dirty_bitmap_remove(dst->node_name, name, NULL);
             return -1;
         }
     }
@@ -2167,6 +2198,7 @@ static int img_convert(int argc, char **argv)
     bool force_share = false;
     bool explict_min_sparse = false;
     bool bitmaps = false;
+    bool skip_broken = false;
     int64_t rate_limit = 0;
 
     ImgConvertState s = (ImgConvertState) {
@@ -2188,6 +2220,7 @@ static int img_convert(int argc, char **argv)
             {"salvage", no_argument, 0, OPTION_SALVAGE},
             {"target-is-zero", no_argument, 0, OPTION_TARGET_IS_ZERO},
             {"bitmaps", no_argument, 0, OPTION_BITMAPS},
+            {"skip-broken-bitmaps", no_argument, 0, OPTION_SKIP_BROKEN},
             {0, 0, 0, 0}
         };
         c = getopt_long(argc, argv, ":hf:O:B:Cco:l:S:pt:T:qnm:WUr:",
@@ -2316,6 +2349,9 @@ static int img_convert(int argc, char **argv)
         case OPTION_BITMAPS:
             bitmaps = true;
             break;
+        case OPTION_SKIP_BROKEN:
+            skip_broken = true;
+            break;
         }
     }
 
@@ -2323,6 +2359,11 @@ static int img_convert(int argc, char **argv)
         out_fmt = "raw";
     }
 
+    if (skip_broken && !bitmaps) {
+        error_report("Use of --skip-broken-bitmaps requires --bitmaps");
+        goto fail_getopt;
+    }
+
     if (s.compressed && s.copy_range) {
         error_report("Cannot enable copy offloading when -c is used");
         goto fail_getopt;
@@ -2554,9 +2595,8 @@ static int img_convert(int argc, char **argv)
             ret = -1;
             goto out;
         }
-        if (!bdrv_supports_persistent_dirty_bitmap(blk_bs(s.src[0]))) {
-            error_report("Source lacks bitmap support");
-            ret = -1;
+        ret = convert_check_bitmaps(blk_bs(s.src[0]), skip_broken);
+        if (ret < 0) {
             goto out;
         }
     }
@@ -2680,7 +2720,7 @@ static int img_convert(int argc, char **argv)
 
     /* Now copy the bitmaps */
     if (bitmaps && ret == 0) {
-        ret = convert_copy_bitmaps(blk_bs(s.src[0]), out_bs);
+        ret = convert_copy_bitmaps(blk_bs(s.src[0]), out_bs, skip_broken);
     }
 
 out:
diff --git a/qemu-options.hx b/qemu-options.hx
index 0c9ddc0274..99ed5ec5f1 100644
--- a/qemu-options.hx
+++ b/qemu-options.hx
@@ -5301,7 +5301,7 @@ SRST
 
             CN=laptop.example.com,O=Example Home,L=London,ST=London,C=GB
 
-    ``-object iothread,id=id,poll-max-ns=poll-max-ns,poll-grow=poll-grow,poll-shrink=poll-shrink``
+    ``-object iothread,id=id,poll-max-ns=poll-max-ns,poll-grow=poll-grow,poll-shrink=poll-shrink,aio-max-batch=aio-max-batch``
         Creates a dedicated event loop thread that devices can be
         assigned to. This is known as an IOThread. By default device
         emulation happens in vCPU threads or the main event loop thread.
@@ -5337,7 +5337,11 @@ SRST
         the polling time when the algorithm detects it is spending too
         long polling without encountering events.
 
-        The polling parameters can be modified at run-time using the
+        The ``aio-max-batch`` parameter is the maximum number of requests
+        in a batch for the AIO engine, 0 means that the engine will use
+        its default.
+
+        The IOThread parameters can be modified at run-time using the
         ``qom-set`` command (where ``iothread1`` is the IOThread's
         ``id``):
 
diff --git a/qom/object.c b/qom/object.c
index 6a01d56546..e86cb05b84 100644
--- a/qom/object.c
+++ b/qom/object.c
@@ -20,6 +20,7 @@
 #include "qapi/string-input-visitor.h"
 #include "qapi/string-output-visitor.h"
 #include "qapi/qobject-input-visitor.h"
+#include "qapi/forward-visitor.h"
 #include "qapi/qapi-builtin-visit.h"
 #include "qapi/qmp/qerror.h"
 #include "qapi/qmp/qjson.h"
@@ -2683,16 +2684,20 @@ static void property_get_alias(Object *obj, Visitor *v, const char *name,
                                void *opaque, Error **errp)
 {
     AliasProperty *prop = opaque;
+    Visitor *alias_v = visitor_forward_field(v, prop->target_name, name);
 
-    object_property_get(prop->target_obj, prop->target_name, v, errp);
+    object_property_get(prop->target_obj, prop->target_name, alias_v, errp);
+    visit_free(alias_v);
 }
 
 static void property_set_alias(Object *obj, Visitor *v, const char *name,
                                void *opaque, Error **errp)
 {
     AliasProperty *prop = opaque;
+    Visitor *alias_v = visitor_forward_field(v, prop->target_name, name);
 
-    object_property_set(prop->target_obj, prop->target_name, v, errp);
+    object_property_set(prop->target_obj, prop->target_name, alias_v, errp);
+    visit_free(alias_v);
 }
 
 static Object *property_resolve_alias(Object *obj, void *opaque,
diff --git a/softmmu/timers-state.h b/softmmu/timers-state.h
index 8c262ce139..94bb7394c5 100644
--- a/softmmu/timers-state.h
+++ b/softmmu/timers-state.h
@@ -47,7 +47,7 @@ typedef struct TimersState {
     int64_t last_delta;
 
     /* Compensate for varying guest execution speed.  */
-    int64_t qemu_icount_bias;
+    aligned_int64_t qemu_icount_bias;
 
     int64_t vm_clock_warp_start;
     int64_t cpu_clock_offset;
diff --git a/softmmu/vl.c b/softmmu/vl.c
index 4df1496101..f815acccaa 100644
--- a/softmmu/vl.c
+++ b/softmmu/vl.c
@@ -202,6 +202,7 @@ static struct {
     { .driver = "virtio-vga",           .flag = &default_vga       },
     { .driver = "ati-vga",              .flag = &default_vga       },
     { .driver = "vhost-user-vga",       .flag = &default_vga       },
+    { .driver = "virtio-vga-gl",        .flag = &default_vga       },
 };
 
 static QemuOptsList qemu_rtc_opts = {
@@ -2126,6 +2127,7 @@ static void qemu_create_machine(QDict *qdict)
         QDict *default_opts =
             keyval_parse(machine_class->default_machine_opts, NULL, NULL,
                          &error_abort);
+        qemu_apply_legacy_machine_options(default_opts);
         object_set_properties_from_keyval(OBJECT(current_machine), default_opts,
                                           false, &error_abort);
         qobject_unref(default_opts);
diff --git a/target/alpha/translate.c b/target/alpha/translate.c
index 103c6326a2..de6c0a8439 100644
--- a/target/alpha/translate.c
+++ b/target/alpha/translate.c
@@ -1207,19 +1207,8 @@ static DisasJumpType gen_call_pal(DisasContext *ctx, int palcode)
                   ? 0x2000 + (palcode - 0x80) * 64
                   : 0x1000 + palcode * 64);
 
-        /* Since the destination is running in PALmode, we don't really
-           need the page permissions check.  We'll see the existence of
-           the page when we create the TB, and we'll flush all TBs if
-           we change the PAL base register.  */
-        if (!ctx->base.singlestep_enabled) {
-            tcg_gen_goto_tb(0);
-            tcg_gen_movi_i64(cpu_pc, entry);
-            tcg_gen_exit_tb(ctx->base.tb, 0);
-            return DISAS_NORETURN;
-        } else {
-            tcg_gen_movi_i64(cpu_pc, entry);
-            return DISAS_PC_UPDATED;
-        }
+        tcg_gen_movi_i64(cpu_pc, entry);
+        return DISAS_PC_UPDATED;
     }
 #endif
 }
@@ -2978,21 +2967,6 @@ static void alpha_tr_insn_start(DisasContextBase *dcbase, CPUState *cpu)
     tcg_gen_insn_start(dcbase->pc_next);
 }
 
-static bool alpha_tr_breakpoint_check(DisasContextBase *dcbase, CPUState *cpu,
-                                      const CPUBreakpoint *bp)
-{
-    DisasContext *ctx = container_of(dcbase, DisasContext, base);
-
-    ctx->base.is_jmp = gen_excp(ctx, EXCP_DEBUG, 0);
-
-    /* The address covered by the breakpoint must be included in
-       [tb->pc, tb->pc + tb->size) in order to for it to be
-       properly cleared -- thus we increment the PC here so that
-       the logic setting tb->size below does the right thing.  */
-    ctx->base.pc_next += 4;
-    return true;
-}
-
 static void alpha_tr_translate_insn(DisasContextBase *dcbase, CPUState *cpu)
 {
     DisasContext *ctx = container_of(dcbase, DisasContext, base);
@@ -3051,7 +3025,6 @@ static const TranslatorOps alpha_tr_ops = {
     .init_disas_context = alpha_tr_init_disas_context,
     .tb_start           = alpha_tr_tb_start,
     .insn_start         = alpha_tr_insn_start,
-    .breakpoint_check   = alpha_tr_breakpoint_check,
     .translate_insn     = alpha_tr_translate_insn,
     .tb_stop            = alpha_tr_tb_stop,
     .disas_log          = alpha_tr_disas_log,
diff --git a/target/arm/cpu.c b/target/arm/cpu.c
index 9cddfd6a44..752b15bb79 100644
--- a/target/arm/cpu.c
+++ b/target/arm/cpu.c
@@ -1984,6 +1984,7 @@ static const struct TCGCPUOps arm_tcg_ops = {
     .do_unaligned_access = arm_cpu_do_unaligned_access,
     .adjust_watchpoint_address = arm_adjust_watchpoint_address,
     .debug_check_watchpoint = arm_debug_check_watchpoint,
+    .debug_check_breakpoint = arm_debug_check_breakpoint,
 #endif /* !CONFIG_USER_ONLY */
 };
 #endif /* CONFIG_TCG */
diff --git a/target/arm/cpu_tcg.c b/target/arm/cpu_tcg.c
index d2d97115ea..ed444bf436 100644
--- a/target/arm/cpu_tcg.c
+++ b/target/arm/cpu_tcg.c
@@ -911,6 +911,7 @@ static const struct TCGCPUOps arm_v7m_tcg_ops = {
     .do_unaligned_access = arm_cpu_do_unaligned_access,
     .adjust_watchpoint_address = arm_adjust_watchpoint_address,
     .debug_check_watchpoint = arm_debug_check_watchpoint,
+    .debug_check_breakpoint = arm_debug_check_breakpoint,
 #endif /* !CONFIG_USER_ONLY */
 };
 #endif /* CONFIG_TCG */
diff --git a/target/arm/debug_helper.c b/target/arm/debug_helper.c
index 2ff72d47d1..2983e36dd3 100644
--- a/target/arm/debug_helper.c
+++ b/target/arm/debug_helper.c
@@ -216,8 +216,9 @@ static bool check_watchpoints(ARMCPU *cpu)
     return false;
 }
 
-static bool check_breakpoints(ARMCPU *cpu)
+bool arm_debug_check_breakpoint(CPUState *cs)
 {
+    ARMCPU *cpu = ARM_CPU(cs);
     CPUARMState *env = &cpu->env;
     int n;
 
@@ -238,15 +239,6 @@ static bool check_breakpoints(ARMCPU *cpu)
     return false;
 }
 
-void HELPER(check_breakpoints)(CPUARMState *env)
-{
-    ARMCPU *cpu = env_archcpu(env);
-
-    if (check_breakpoints(cpu)) {
-        HELPER(exception_internal(env, EXCP_DEBUG));
-    }
-}
-
 bool arm_debug_check_watchpoint(CPUState *cs, CPUWatchpoint *wp)
 {
     /*
diff --git a/target/arm/helper-a64.c b/target/arm/helper-a64.c
index ac5c4452d5..26f79f9141 100644
--- a/target/arm/helper-a64.c
+++ b/target/arm/helper-a64.c
@@ -564,7 +564,7 @@ uint64_t HELPER(paired_cmpxchg64_le_parallel)(CPUARMState *env, uint64_t addr,
 
     cmpv = int128_make128(env->exclusive_val, env->exclusive_high);
     newv = int128_make128(new_lo, new_hi);
-    oldv = helper_atomic_cmpxchgo_le_mmu(env, addr, cmpv, newv, oi, ra);
+    oldv = cpu_atomic_cmpxchgo_le_mmu(env, addr, cmpv, newv, oi, ra);
 
     success = int128_eq(oldv, cmpv);
     return !success;
@@ -638,7 +638,7 @@ uint64_t HELPER(paired_cmpxchg64_be_parallel)(CPUARMState *env, uint64_t addr,
      */
     cmpv = int128_make128(env->exclusive_high, env->exclusive_val);
     newv = int128_make128(new_hi, new_lo);
-    oldv = helper_atomic_cmpxchgo_be_mmu(env, addr, cmpv, newv, oi, ra);
+    oldv = cpu_atomic_cmpxchgo_be_mmu(env, addr, cmpv, newv, oi, ra);
 
     success = int128_eq(oldv, cmpv);
     return !success;
@@ -660,7 +660,7 @@ void HELPER(casp_le_parallel)(CPUARMState *env, uint32_t rs, uint64_t addr,
 
     cmpv = int128_make128(env->xregs[rs], env->xregs[rs + 1]);
     newv = int128_make128(new_lo, new_hi);
-    oldv = helper_atomic_cmpxchgo_le_mmu(env, addr, cmpv, newv, oi, ra);
+    oldv = cpu_atomic_cmpxchgo_le_mmu(env, addr, cmpv, newv, oi, ra);
 
     env->xregs[rs] = int128_getlo(oldv);
     env->xregs[rs + 1] = int128_gethi(oldv);
@@ -681,7 +681,7 @@ void HELPER(casp_be_parallel)(CPUARMState *env, uint32_t rs, uint64_t addr,
 
     cmpv = int128_make128(env->xregs[rs + 1], env->xregs[rs]);
     newv = int128_make128(new_lo, new_hi);
-    oldv = helper_atomic_cmpxchgo_be_mmu(env, addr, cmpv, newv, oi, ra);
+    oldv = cpu_atomic_cmpxchgo_be_mmu(env, addr, cmpv, newv, oi, ra);
 
     env->xregs[rs + 1] = int128_getlo(oldv);
     env->xregs[rs] = int128_gethi(oldv);
diff --git a/target/arm/helper.h b/target/arm/helper.h
index db87d7d537..248569b0cd 100644
--- a/target/arm/helper.h
+++ b/target/arm/helper.h
@@ -54,8 +54,6 @@ DEF_HELPER_1(yield, void, env)
 DEF_HELPER_1(pre_hvc, void, env)
 DEF_HELPER_2(pre_smc, void, env, i32)
 
-DEF_HELPER_1(check_breakpoints, void, env)
-
 DEF_HELPER_3(cpsr_write, void, env, i32, i32)
 DEF_HELPER_2(cpsr_write_eret, void, env, i32)
 DEF_HELPER_1(cpsr_read, i32, env)
diff --git a/target/arm/internals.h b/target/arm/internals.h
index 3ba86e8af8..11a72013f5 100644
--- a/target/arm/internals.h
+++ b/target/arm/internals.h
@@ -282,6 +282,9 @@ void hw_breakpoint_update(ARMCPU *cpu, int n);
  */
 void hw_breakpoint_update_all(ARMCPU *cpu);
 
+/* Callback function for checking if a breakpoint should trigger. */
+bool arm_debug_check_breakpoint(CPUState *cs);
+
 /* Callback function for checking if a watchpoint should trigger. */
 bool arm_debug_check_watchpoint(CPUState *cs, CPUWatchpoint *wp);
 
diff --git a/target/arm/translate-a64.c b/target/arm/translate-a64.c
index ca11a5fecd..422e2ac0c9 100644
--- a/target/arm/translate-a64.c
+++ b/target/arm/translate-a64.c
@@ -14844,30 +14844,6 @@ static void aarch64_tr_insn_start(DisasContextBase *dcbase, CPUState *cpu)
     dc->insn_start = tcg_last_op();
 }
 
-static bool aarch64_tr_breakpoint_check(DisasContextBase *dcbase, CPUState *cpu,
-                                        const CPUBreakpoint *bp)
-{
-    DisasContext *dc = container_of(dcbase, DisasContext, base);
-
-    if (bp->flags & BP_CPU) {
-        gen_a64_set_pc_im(dc->base.pc_next);
-        gen_helper_check_breakpoints(cpu_env);
-        /* End the TB early; it likely won't be executed */
-        dc->base.is_jmp = DISAS_TOO_MANY;
-    } else {
-        gen_exception_internal_insn(dc, dc->base.pc_next, EXCP_DEBUG);
-        /* The address covered by the breakpoint must be
-           included in [tb->pc, tb->pc + tb->size) in order
-           to for it to be properly cleared -- thus we
-           increment the PC here so that the logic setting
-           tb->size below does the right thing.  */
-        dc->base.pc_next += 4;
-        dc->base.is_jmp = DISAS_NORETURN;
-    }
-
-    return true;
-}
-
 static void aarch64_tr_translate_insn(DisasContextBase *dcbase, CPUState *cpu)
 {
     DisasContext *dc = container_of(dcbase, DisasContext, base);
@@ -14982,7 +14958,6 @@ const TranslatorOps aarch64_translator_ops = {
     .init_disas_context = aarch64_tr_init_disas_context,
     .tb_start           = aarch64_tr_tb_start,
     .insn_start         = aarch64_tr_insn_start,
-    .breakpoint_check   = aarch64_tr_breakpoint_check,
     .translate_insn     = aarch64_tr_translate_insn,
     .tb_stop            = aarch64_tr_tb_stop,
     .disas_log          = aarch64_tr_disas_log,
diff --git a/target/arm/translate.c b/target/arm/translate.c
index e1a8152598..351afa43a2 100644
--- a/target/arm/translate.c
+++ b/target/arm/translate.c
@@ -9438,33 +9438,6 @@ static void arm_tr_insn_start(DisasContextBase *dcbase, CPUState *cpu)
     dc->insn_start = tcg_last_op();
 }
 
-static bool arm_tr_breakpoint_check(DisasContextBase *dcbase, CPUState *cpu,
-                                    const CPUBreakpoint *bp)
-{
-    DisasContext *dc = container_of(dcbase, DisasContext, base);
-
-    if (bp->flags & BP_CPU) {
-        gen_set_condexec(dc);
-        gen_set_pc_im(dc, dc->base.pc_next);
-        gen_helper_check_breakpoints(cpu_env);
-        /* End the TB early; it's likely not going to be executed */
-        dc->base.is_jmp = DISAS_TOO_MANY;
-    } else {
-        gen_exception_internal_insn(dc, dc->base.pc_next, EXCP_DEBUG);
-        /* The address covered by the breakpoint must be
-           included in [tb->pc, tb->pc + tb->size) in order
-           to for it to be properly cleared -- thus we
-           increment the PC here so that the logic setting
-           tb->size below does the right thing.  */
-        /* TODO: Advance PC by correct instruction length to
-         * avoid disassembler error messages */
-        dc->base.pc_next += 2;
-        dc->base.is_jmp = DISAS_NORETURN;
-    }
-
-    return true;
-}
-
 static bool arm_pre_translate_insn(DisasContext *dc)
 {
 #ifdef CONFIG_USER_ONLY
@@ -9827,7 +9800,6 @@ static const TranslatorOps arm_translator_ops = {
     .init_disas_context = arm_tr_init_disas_context,
     .tb_start           = arm_tr_tb_start,
     .insn_start         = arm_tr_insn_start,
-    .breakpoint_check   = arm_tr_breakpoint_check,
     .translate_insn     = arm_tr_translate_insn,
     .tb_stop            = arm_tr_tb_stop,
     .disas_log          = arm_tr_disas_log,
@@ -9837,7 +9809,6 @@ static const TranslatorOps thumb_translator_ops = {
     .init_disas_context = arm_tr_init_disas_context,
     .tb_start           = arm_tr_tb_start,
     .insn_start         = arm_tr_insn_start,
-    .breakpoint_check   = arm_tr_breakpoint_check,
     .translate_insn     = thumb_tr_translate_insn,
     .tb_stop            = arm_tr_tb_stop,
     .disas_log          = arm_tr_disas_log,
diff --git a/target/avr/cpu.c b/target/avr/cpu.c
index 57e3fab4a0..ea14175ca5 100644
--- a/target/avr/cpu.c
+++ b/target/avr/cpu.c
@@ -223,6 +223,7 @@ static void avr_cpu_class_init(ObjectClass *oc, void *data)
     cc->disas_set_info = avr_cpu_disas_set_info;
     cc->gdb_read_register = avr_cpu_gdb_read_register;
     cc->gdb_write_register = avr_cpu_gdb_write_register;
+    cc->gdb_adjust_breakpoint = avr_cpu_gdb_adjust_breakpoint;
     cc->gdb_num_core_regs = 35;
     cc->gdb_core_xml_file = "avr-cpu.xml";
     cc->tcg_ops = &avr_tcg_ops;
diff --git a/target/avr/cpu.h b/target/avr/cpu.h
index d148e8c75a..93e3faa0a9 100644
--- a/target/avr/cpu.h
+++ b/target/avr/cpu.h
@@ -162,6 +162,7 @@ hwaddr avr_cpu_get_phys_page_debug(CPUState *cpu, vaddr addr);
 int avr_cpu_gdb_read_register(CPUState *cpu, GByteArray *buf, int reg);
 int avr_cpu_gdb_write_register(CPUState *cpu, uint8_t *buf, int reg);
 int avr_print_insn(bfd_vma addr, disassemble_info *info);
+vaddr avr_cpu_gdb_adjust_breakpoint(CPUState *cpu, vaddr addr);
 
 static inline int avr_feature(CPUAVRState *env, AVRFeature feature)
 {
diff --git a/target/avr/gdbstub.c b/target/avr/gdbstub.c
index c28ed67efe..1c1b908c92 100644
--- a/target/avr/gdbstub.c
+++ b/target/avr/gdbstub.c
@@ -82,3 +82,16 @@ int avr_cpu_gdb_write_register(CPUState *cs, uint8_t *mem_buf, int n)
 
     return 0;
 }
+
+vaddr avr_cpu_gdb_adjust_breakpoint(CPUState *cpu, vaddr addr)
+{
+    /*
+     * This is due to some strange GDB behavior
+     * Let's assume main has address 0x100:
+     * b main   - sets breakpoint at address 0x00000100 (code)
+     * b *0x100 - sets breakpoint at address 0x00800100 (data)
+     *
+     * Force all breakpoints into code space.
+     */
+    return addr % OFFSET_DATA;
+}
diff --git a/target/avr/translate.c b/target/avr/translate.c
index 8237a03c23..1111e08b83 100644
--- a/target/avr/translate.c
+++ b/target/avr/translate.c
@@ -2900,14 +2900,6 @@ static bool canonicalize_skip(DisasContext *ctx)
     return true;
 }
 
-static void gen_breakpoint(DisasContext *ctx)
-{
-    canonicalize_skip(ctx);
-    tcg_gen_movi_tl(cpu_pc, ctx->npc);
-    gen_helper_debug(cpu_env);
-    ctx->base.is_jmp = DISAS_NORETURN;
-}
-
 static void avr_tr_init_disas_context(DisasContextBase *dcbase, CPUState *cs)
 {
     DisasContext *ctx = container_of(dcbase, DisasContext, base);
@@ -2944,34 +2936,11 @@ static void avr_tr_insn_start(DisasContextBase *dcbase, CPUState *cs)
     tcg_gen_insn_start(ctx->npc);
 }
 
-static bool avr_tr_breakpoint_check(DisasContextBase *dcbase, CPUState *cs,
-                                    const CPUBreakpoint *bp)
-{
-    DisasContext *ctx = container_of(dcbase, DisasContext, base);
-
-    gen_breakpoint(ctx);
-    return true;
-}
-
 static void avr_tr_translate_insn(DisasContextBase *dcbase, CPUState *cs)
 {
     DisasContext *ctx = container_of(dcbase, DisasContext, base);
     TCGLabel *skip_label = NULL;
 
-    /*
-     * This is due to some strange GDB behavior
-     * Let's assume main has address 0x100:
-     * b main   - sets breakpoint at address 0x00000100 (code)
-     * b *0x100 - sets breakpoint at address 0x00800100 (data)
-     *
-     * The translator driver has already taken care of the code pointer.
-     */
-    if (!ctx->base.singlestep_enabled &&
-        cpu_breakpoint_test(cs, OFFSET_DATA + ctx->base.pc_next, BP_ANY)) {
-        gen_breakpoint(ctx);
-        return;
-    }
-
     /* Conditionally skip the next instruction, if indicated.  */
     if (ctx->skip_cond != TCG_COND_NEVER) {
         skip_label = gen_new_label();
@@ -3069,7 +3038,6 @@ static const TranslatorOps avr_tr_ops = {
     .init_disas_context = avr_tr_init_disas_context,
     .tb_start           = avr_tr_tb_start,
     .insn_start         = avr_tr_insn_start,
-    .breakpoint_check   = avr_tr_breakpoint_check,
     .translate_insn     = avr_tr_translate_insn,
     .tb_stop            = avr_tr_tb_stop,
     .disas_log          = avr_tr_disas_log,
diff --git a/target/cris/translate.c b/target/cris/translate.c
index 9258c13e9f..a84b753349 100644
--- a/target/cris/translate.c
+++ b/target/cris/translate.c
@@ -3118,25 +3118,6 @@ static void cris_tr_insn_start(DisasContextBase *dcbase, CPUState *cpu)
     tcg_gen_insn_start(dc->delayed_branch == 1 ? dc->ppc | 1 : dc->pc);
 }
 
-static bool cris_tr_breakpoint_check(DisasContextBase *dcbase, CPUState *cpu,
-                                     const CPUBreakpoint *bp)
-{
-    DisasContext *dc = container_of(dcbase, DisasContext, base);
-
-    cris_evaluate_flags(dc);
-    tcg_gen_movi_tl(env_pc, dc->pc);
-    t_gen_raise_exception(EXCP_DEBUG);
-    dc->base.is_jmp = DISAS_NORETURN;
-    /*
-     * The address covered by the breakpoint must be included in
-     * [tb->pc, tb->pc + tb->size) in order to for it to be
-     * properly cleared -- thus we increment the PC here so that
-     * the logic setting tb->size below does the right thing.
-     */
-    dc->pc += 2;
-    return true;
-}
-
 static void cris_tr_translate_insn(DisasContextBase *dcbase, CPUState *cs)
 {
     DisasContext *dc = container_of(dcbase, DisasContext, base);
@@ -3315,7 +3296,6 @@ static const TranslatorOps cris_tr_ops = {
     .init_disas_context = cris_tr_init_disas_context,
     .tb_start           = cris_tr_tb_start,
     .insn_start         = cris_tr_insn_start,
-    .breakpoint_check   = cris_tr_breakpoint_check,
     .translate_insn     = cris_tr_translate_insn,
     .tb_stop            = cris_tr_tb_stop,
     .disas_log          = cris_tr_disas_log,
diff --git a/target/hexagon/translate.c b/target/hexagon/translate.c
index b23d36adf5..54fdcaa5e8 100644
--- a/target/hexagon/translate.c
+++ b/target/hexagon/translate.c
@@ -540,22 +540,6 @@ static void hexagon_tr_insn_start(DisasContextBase *dcbase, CPUState *cpu)
     tcg_gen_insn_start(ctx->base.pc_next);
 }
 
-static bool hexagon_tr_breakpoint_check(DisasContextBase *dcbase, CPUState *cpu,
-                                        const CPUBreakpoint *bp)
-{
-    DisasContext *ctx = container_of(dcbase, DisasContext, base);
-
-    gen_exception_end_tb(ctx, EXCP_DEBUG);
-    /*
-     * The address covered by the breakpoint must be included in
-     * [tb->pc, tb->pc + tb->size) in order to for it to be
-     * properly cleared -- thus we increment the PC here so that
-     * the logic setting tb->size below does the right thing.
-     */
-    ctx->base.pc_next += 4;
-    return true;
-}
-
 static bool pkt_crosses_page(CPUHexagonState *env, DisasContext *ctx)
 {
     target_ulong page_start = ctx->base.pc_first & TARGET_PAGE_MASK;
@@ -631,7 +615,6 @@ static const TranslatorOps hexagon_tr_ops = {
     .init_disas_context = hexagon_tr_init_disas_context,
     .tb_start           = hexagon_tr_tb_start,
     .insn_start         = hexagon_tr_insn_start,
-    .breakpoint_check   = hexagon_tr_breakpoint_check,
     .translate_insn     = hexagon_tr_translate_packet,
     .tb_stop            = hexagon_tr_tb_stop,
     .disas_log          = hexagon_tr_disas_log,
diff --git a/target/hppa/translate.c b/target/hppa/translate.c
index 2552747138..b18150ef8d 100644
--- a/target/hppa/translate.c
+++ b/target/hppa/translate.c
@@ -4159,16 +4159,6 @@ static void hppa_tr_insn_start(DisasContextBase *dcbase, CPUState *cs)
     tcg_gen_insn_start(ctx->iaoq_f, ctx->iaoq_b);
 }
 
-static bool hppa_tr_breakpoint_check(DisasContextBase *dcbase, CPUState *cs,
-                                      const CPUBreakpoint *bp)
-{
-    DisasContext *ctx = container_of(dcbase, DisasContext, base);
-
-    gen_excp(ctx, EXCP_DEBUG);
-    ctx->base.pc_next += 4;
-    return true;
-}
-
 static void hppa_tr_translate_insn(DisasContextBase *dcbase, CPUState *cs)
 {
     DisasContext *ctx = container_of(dcbase, DisasContext, base);
@@ -4330,7 +4320,6 @@ static const TranslatorOps hppa_tr_ops = {
     .init_disas_context = hppa_tr_init_disas_context,
     .tb_start           = hppa_tr_tb_start,
     .insn_start         = hppa_tr_insn_start,
-    .breakpoint_check   = hppa_tr_breakpoint_check,
     .translate_insn     = hppa_tr_translate_insn,
     .tb_stop            = hppa_tr_tb_stop,
     .disas_log          = hppa_tr_disas_log,
diff --git a/target/i386/cpu.c b/target/i386/cpu.c
index 48b55ebd0a..edb97ebbbe 100644
--- a/target/i386/cpu.c
+++ b/target/i386/cpu.c
@@ -4919,6 +4919,9 @@ static uint64_t x86_cpu_get_supported_feature_word(FeatureWord w,
     return r;
 }
 
+/*
+ * Only for builtin_x86_defs models initialized with x86_register_cpudef_types.
+ */
 void x86_cpu_apply_props(X86CPU *cpu, PropValue *props)
 {
     PropValue *pv;
@@ -4931,7 +4934,11 @@ void x86_cpu_apply_props(X86CPU *cpu, PropValue *props)
     }
 }
 
-/* Apply properties for the CPU model version specified in model */
+/*
+ * Apply properties for the CPU model version specified in model.
+ * Only for builtin_x86_defs models initialized with x86_register_cpudef_types.
+ */
+
 static void x86_cpu_apply_version_props(X86CPU *cpu, X86CPUModel *model)
 {
     const X86CPUVersionDefinition *vdef;
@@ -4960,7 +4967,9 @@ static void x86_cpu_apply_version_props(X86CPU *cpu, X86CPUModel *model)
     assert(vdef->version == version);
 }
 
-/* Load data from X86CPUDefinition into a X86CPU object
+/*
+ * Load data from X86CPUDefinition into a X86CPU object.
+ * Only for builtin_x86_defs models initialized with x86_register_cpudef_types.
  */
 static void x86_cpu_load_model(X86CPU *cpu, X86CPUModel *model)
 {
@@ -5051,6 +5060,12 @@ static void x86_register_cpu_model_type(const char *name, X86CPUModel *model)
     type_register(&ti);
 }
 
+
+/*
+ * register builtin_x86_defs;
+ * "max", "base" and subclasses ("host") are not registered here.
+ * See x86_cpu_register_types for all model registrations.
+ */
 static void x86_register_cpudef_types(const X86CPUDefinition *def)
 {
     X86CPUModel *m;
diff --git a/target/i386/cpu.h b/target/i386/cpu.h
index 5d98a4e7c0..6c50d3ab4f 100644
--- a/target/i386/cpu.h
+++ b/target/i386/cpu.h
@@ -240,6 +240,7 @@ typedef enum X86Seg {
 #define CR4_OSFXSR_SHIFT 9
 #define CR4_OSFXSR_MASK (1U << CR4_OSFXSR_SHIFT)
 #define CR4_OSXMMEXCPT_MASK  (1U << 10)
+#define CR4_UMIP_MASK   (1U << 11)
 #define CR4_LA57_MASK   (1U << 12)
 #define CR4_VMXE_MASK   (1U << 13)
 #define CR4_SMXE_MASK   (1U << 14)
@@ -251,6 +252,14 @@ typedef enum X86Seg {
 #define CR4_PKE_MASK   (1U << 22)
 #define CR4_PKS_MASK   (1U << 24)
 
+#define CR4_RESERVED_MASK \
+(~(target_ulong)(CR4_VME_MASK | CR4_PVI_MASK | CR4_TSD_MASK \
+                | CR4_DE_MASK | CR4_PSE_MASK | CR4_PAE_MASK \
+                | CR4_MCE_MASK | CR4_PGE_MASK | CR4_PCE_MASK \
+                | CR4_OSFXSR_MASK | CR4_OSXMMEXCPT_MASK |CR4_UMIP_MASK \
+                | CR4_FSGSBASE_MASK | CR4_PCIDE_MASK | CR4_OSXSAVE_MASK \
+                | CR4_SMEP_MASK | CR4_SMAP_MASK | CR4_PKE_MASK | CR4_PKS_MASK))
+
 #define DR6_BD          (1 << 13)
 #define DR6_BS          (1 << 14)
 #define DR6_BT          (1 << 15)
@@ -466,6 +475,11 @@ typedef enum X86Seg {
 #define MSR_EFER_SVME  (1 << 12)
 #define MSR_EFER_FFXSR (1 << 14)
 
+#define MSR_EFER_RESERVED\
+        (~(target_ulong)(MSR_EFER_SCE | MSR_EFER_LME\
+            | MSR_EFER_LMA | MSR_EFER_NXE | MSR_EFER_SVME\
+            | MSR_EFER_FFXSR))
+
 #define MSR_STAR                        0xc0000081
 #define MSR_LSTAR                       0xc0000082
 #define MSR_CSTAR                       0xc0000083
@@ -2196,6 +2210,36 @@ static inline bool hyperv_feat_enabled(X86CPU *cpu, int feat)
     return !!(cpu->hyperv_features & BIT(feat));
 }
 
+static inline uint64_t cr4_reserved_bits(CPUX86State *env)
+{
+    uint64_t reserved_bits = CR4_RESERVED_MASK;
+    if (!env->features[FEAT_XSAVE]) {
+        reserved_bits |= CR4_OSXSAVE_MASK;
+    }
+    if (!(env->features[FEAT_7_0_EBX] & CPUID_7_0_EBX_SMEP)) {
+        reserved_bits |= CR4_SMEP_MASK;
+    }
+    if (!(env->features[FEAT_7_0_EBX] & CPUID_7_0_EBX_SMAP)) {
+        reserved_bits |= CR4_SMAP_MASK;
+    }
+    if (!(env->features[FEAT_7_0_EBX] & CPUID_7_0_EBX_FSGSBASE)) {
+        reserved_bits |= CR4_FSGSBASE_MASK;
+    }
+    if (!(env->features[FEAT_7_0_ECX] & CPUID_7_0_ECX_PKU)) {
+        reserved_bits |= CR4_PKE_MASK;
+    }
+    if (!(env->features[FEAT_7_0_ECX] & CPUID_7_0_ECX_LA57)) {
+        reserved_bits |= CR4_LA57_MASK;
+    }
+    if (!(env->features[FEAT_7_0_ECX] & CPUID_7_0_ECX_UMIP)) {
+        reserved_bits |= CR4_UMIP_MASK;
+    }
+    if (!(env->features[FEAT_7_0_ECX] & CPUID_7_0_ECX_PKS)) {
+        reserved_bits |= CR4_PKS_MASK;
+    }
+    return reserved_bits;
+}
+
 #if defined(TARGET_X86_64) && \
     defined(CONFIG_USER_ONLY) && \
     defined(CONFIG_LINUX)
diff --git a/target/i386/host-cpu.c b/target/i386/host-cpu.c
index 4ea9e354ea..10f8aba86e 100644
--- a/target/i386/host-cpu.c
+++ b/target/i386/host-cpu.c
@@ -150,13 +150,16 @@ void host_cpu_vendor_fms(char *vendor, int *family, int *model, int *stepping)
 
 void host_cpu_instance_init(X86CPU *cpu)
 {
-    uint32_t ebx = 0, ecx = 0, edx = 0;
-    char vendor[CPUID_VENDOR_SZ + 1];
+    X86CPUClass *xcc = X86_CPU_GET_CLASS(cpu);
 
-    host_cpuid(0, 0, NULL, &ebx, &ecx, &edx);
-    x86_cpu_vendor_words2str(vendor, ebx, edx, ecx);
+    if (xcc->model) {
+        uint32_t ebx = 0, ecx = 0, edx = 0;
+        char vendor[CPUID_VENDOR_SZ + 1];
 
-    object_property_set_str(OBJECT(cpu), "vendor", vendor, &error_abort);
+        host_cpuid(0, 0, NULL, &ebx, &ecx, &edx);
+        x86_cpu_vendor_words2str(vendor, ebx, edx, ecx);
+        object_property_set_str(OBJECT(cpu), "vendor", vendor, &error_abort);
+    }
 }
 
 void host_cpu_max_instance_init(X86CPU *cpu)
diff --git a/target/i386/kvm/kvm-cpu.c b/target/i386/kvm/kvm-cpu.c
index bbe817764d..d95028018e 100644
--- a/target/i386/kvm/kvm-cpu.c
+++ b/target/i386/kvm/kvm-cpu.c
@@ -52,47 +52,6 @@ static bool kvm_cpu_realizefn(CPUState *cs, Error **errp)
     return host_cpu_realizefn(cs, errp);
 }
 
-/*
- * KVM-specific features that are automatically added/removed
- * from all CPU models when KVM is enabled.
- *
- * NOTE: features can be enabled by default only if they were
- *       already available in the oldest kernel version supported
- *       by the KVM accelerator (see "OS requirements" section at
- *       docs/system/target-i386.rst)
- */
-static PropValue kvm_default_props[] = {
-    { "kvmclock", "on" },
-    { "kvm-nopiodelay", "on" },
-    { "kvm-asyncpf", "on" },
-    { "kvm-steal-time", "on" },
-    { "kvm-pv-eoi", "on" },
-    { "kvmclock-stable-bit", "on" },
-    { "x2apic", "on" },
-    { "kvm-msi-ext-dest-id", "off" },
-    { "acpi", "off" },
-    { "monitor", "off" },
-    { "svm", "off" },
-    { NULL, NULL },
-};
-
-void x86_cpu_change_kvm_default(const char *prop, const char *value)
-{
-    PropValue *pv;
-    for (pv = kvm_default_props; pv->prop; pv++) {
-        if (!strcmp(pv->prop, prop)) {
-            pv->value = value;
-            break;
-        }
-    }
-
-    /*
-     * It is valid to call this function only for properties that
-     * are already present in the kvm_default_props table.
-     */
-    assert(pv->prop);
-}
-
 static bool lmce_supported(void)
 {
     uint64_t mce_cap = 0;
@@ -150,21 +109,69 @@ static void kvm_cpu_xsave_init(void)
     }
 }
 
+/*
+ * KVM-specific features that are automatically added/removed
+ * from cpudef models when KVM is enabled.
+ * Only for builtin_x86_defs models initialized with x86_register_cpudef_types.
+ *
+ * NOTE: features can be enabled by default only if they were
+ *       already available in the oldest kernel version supported
+ *       by the KVM accelerator (see "OS requirements" section at
+ *       docs/system/target-i386.rst)
+ */
+static PropValue kvm_default_props[] = {
+    { "kvmclock", "on" },
+    { "kvm-nopiodelay", "on" },
+    { "kvm-asyncpf", "on" },
+    { "kvm-steal-time", "on" },
+    { "kvm-pv-eoi", "on" },
+    { "kvmclock-stable-bit", "on" },
+    { "x2apic", "on" },
+    { "kvm-msi-ext-dest-id", "off" },
+    { "acpi", "off" },
+    { "monitor", "off" },
+    { "svm", "off" },
+    { NULL, NULL },
+};
+
+/*
+ * Only for builtin_x86_defs models initialized with x86_register_cpudef_types.
+ */
+void x86_cpu_change_kvm_default(const char *prop, const char *value)
+{
+    PropValue *pv;
+    for (pv = kvm_default_props; pv->prop; pv++) {
+        if (!strcmp(pv->prop, prop)) {
+            pv->value = value;
+            break;
+        }
+    }
+
+    /*
+     * It is valid to call this function only for properties that
+     * are already present in the kvm_default_props table.
+     */
+    assert(pv->prop);
+}
+
 static void kvm_cpu_instance_init(CPUState *cs)
 {
     X86CPU *cpu = X86_CPU(cs);
+    X86CPUClass *xcc = X86_CPU_GET_CLASS(cpu);
 
     host_cpu_instance_init(cpu);
 
-    if (!kvm_irqchip_in_kernel()) {
-        x86_cpu_change_kvm_default("x2apic", "off");
-    } else if (kvm_irqchip_is_split() && kvm_enable_x2apic()) {
-        x86_cpu_change_kvm_default("kvm-msi-ext-dest-id", "on");
-    }
-
-    /* Special cases not set in the X86CPUDefinition structs: */
+    if (xcc->model) {
+        /* only applies to builtin_x86_defs cpus */
+        if (!kvm_irqchip_in_kernel()) {
+            x86_cpu_change_kvm_default("x2apic", "off");
+        } else if (kvm_irqchip_is_split() && kvm_enable_x2apic()) {
+            x86_cpu_change_kvm_default("kvm-msi-ext-dest-id", "on");
+        }
 
-    x86_cpu_apply_props(cpu, kvm_default_props);
+        /* Special cases not set in the X86CPUDefinition structs: */
+        x86_cpu_apply_props(cpu, kvm_default_props);
+    }
 
     if (cpu->max_features) {
         kvm_cpu_max_instance_init(cpu);
diff --git a/target/i386/tcg/mem_helper.c b/target/i386/tcg/mem_helper.c
index 591f512bff..2da3cd14b6 100644
--- a/target/i386/tcg/mem_helper.c
+++ b/target/i386/tcg/mem_helper.c
@@ -64,22 +64,12 @@ void helper_cmpxchg8b(CPUX86State *env, target_ulong a0)
     cmpv = deposit64(env->regs[R_EAX], 32, 32, env->regs[R_EDX]);
     newv = deposit64(env->regs[R_EBX], 32, 32, env->regs[R_ECX]);
 
-#ifdef CONFIG_USER_ONLY
-    {
-        uint64_t *haddr = g2h(env_cpu(env), a0);
-        cmpv = cpu_to_le64(cmpv);
-        newv = cpu_to_le64(newv);
-        oldv = qatomic_cmpxchg__nocheck(haddr, cmpv, newv);
-        oldv = le64_to_cpu(oldv);
-    }
-#else
     {
         uintptr_t ra = GETPC();
         int mem_idx = cpu_mmu_index(env, false);
         TCGMemOpIdx oi = make_memop_idx(MO_TEQ, mem_idx);
-        oldv = helper_atomic_cmpxchgq_le_mmu(env, a0, cmpv, newv, oi, ra);
+        oldv = cpu_atomic_cmpxchgq_le_mmu(env, a0, cmpv, newv, oi, ra);
     }
-#endif
 
     if (oldv == cmpv) {
         eflags |= CC_Z;
@@ -147,8 +137,7 @@ void helper_cmpxchg16b(CPUX86State *env, target_ulong a0)
 
         int mem_idx = cpu_mmu_index(env, false);
         TCGMemOpIdx oi = make_memop_idx(MO_TEQ | MO_ALIGN_16, mem_idx);
-        Int128 oldv = helper_atomic_cmpxchgo_le_mmu(env, a0, cmpv,
-                                                    newv, oi, ra);
+        Int128 oldv = cpu_atomic_cmpxchgo_le_mmu(env, a0, cmpv, newv, oi, ra);
 
         if (int128_eq(oldv, cmpv)) {
             eflags |= CC_Z;
diff --git a/target/i386/tcg/sysemu/misc_helper.c b/target/i386/tcg/sysemu/misc_helper.c
index db0d8a9d79..d347af2a99 100644
--- a/target/i386/tcg/sysemu/misc_helper.c
+++ b/target/i386/tcg/sysemu/misc_helper.c
@@ -96,9 +96,19 @@ void helper_write_crN(CPUX86State *env, int reg, target_ulong t0)
         cpu_x86_update_cr0(env, t0);
         break;
     case 3:
+        if ((env->efer & MSR_EFER_LMA) &&
+                (t0 & ((~0UL) << env_archcpu(env)->phys_bits))) {
+            cpu_vmexit(env, SVM_EXIT_ERR, 0, GETPC());
+        }
+        if (!(env->efer & MSR_EFER_LMA)) {
+            t0 &= 0xffffffffUL;
+        }
         cpu_x86_update_cr3(env, t0);
         break;
     case 4:
+        if (t0 & cr4_reserved_bits(env)) {
+            cpu_vmexit(env, SVM_EXIT_ERR, 0, GETPC());
+        }
         if (((t0 ^ env->cr[4]) & CR4_LA57_MASK) &&
             (env->hflags & HF_CS64_MASK)) {
             raise_exception_ra(env, EXCP0D_GPF, GETPC());
diff --git a/target/i386/tcg/sysemu/svm_helper.c b/target/i386/tcg/sysemu/svm_helper.c
index 00618cff23..145511d635 100644
--- a/target/i386/tcg/sysemu/svm_helper.c
+++ b/target/i386/tcg/sysemu/svm_helper.c
@@ -65,6 +65,51 @@ static inline void svm_load_seg_cache(CPUX86State *env, hwaddr addr,
                            sc->base, sc->limit, sc->flags);
 }
 
+static inline bool ctl_has_irq(uint32_t int_ctl)
+{
+    uint32_t int_prio;
+    uint32_t tpr;
+
+    int_prio = (int_ctl & V_INTR_PRIO_MASK) >> V_INTR_MASKING_SHIFT;
+    tpr = int_ctl & V_TPR_MASK;
+    return (int_ctl & V_IRQ_MASK) && (int_prio >= tpr);
+}
+
+static inline bool is_efer_invalid_state (CPUX86State *env)
+{
+    if (!(env->efer & MSR_EFER_SVME)) {
+        return true;
+    }
+
+    if (env->efer & MSR_EFER_RESERVED) {
+        return true;
+    }
+
+    if ((env->efer & (MSR_EFER_LMA | MSR_EFER_LME)) &&
+            !(env->features[FEAT_8000_0001_EDX] & CPUID_EXT2_LM)) {
+        return true;
+    }
+
+    if ((env->efer & MSR_EFER_LME) && (env->cr[0] & CR0_PG_MASK)
+                                && !(env->cr[4] & CR4_PAE_MASK)) {
+        return true;
+    }
+
+    if ((env->efer & MSR_EFER_LME) && (env->cr[0] & CR0_PG_MASK)
+                                && !(env->cr[0] & CR0_PE_MASK)) {
+        return true;
+    }
+
+    if ((env->efer & MSR_EFER_LME) && (env->cr[0] & CR0_PG_MASK)
+                                && (env->cr[4] & CR4_PAE_MASK)
+                                && (env->segs[R_CS].flags & DESC_L_MASK)
+                                && (env->segs[R_CS].flags & DESC_B_MASK)) {
+        return true;
+    }
+
+    return false;
+}
+
 void helper_vmrun(CPUX86State *env, int aflag, int next_eip_addend)
 {
     CPUState *cs = env_cpu(env);
@@ -75,6 +120,8 @@ void helper_vmrun(CPUX86State *env, int aflag, int next_eip_addend)
     uint32_t int_ctl;
     uint32_t asid;
     uint64_t new_cr0;
+    uint64_t new_cr3;
+    uint64_t new_cr4;
 
     cpu_svm_check_intercept_param(env, SVM_EXIT_VMRUN, 0, GETPC());
 
@@ -215,17 +262,22 @@ void helper_vmrun(CPUX86State *env, int aflag, int next_eip_addend)
     if ((new_cr0 & CR0_NW_MASK) && !(new_cr0 & CR0_CD_MASK)) {
         cpu_vmexit(env, SVM_EXIT_ERR, 0, GETPC());
     }
+    new_cr3 = x86_ldq_phys(cs, env->vm_vmcb + offsetof(struct vmcb, save.cr3));
+    if ((env->efer & MSR_EFER_LMA) &&
+            (new_cr3 & ((~0UL) << cpu->phys_bits))) {
+        cpu_vmexit(env, SVM_EXIT_ERR, 0, GETPC());
+    }
+    new_cr4 = x86_ldq_phys(cs, env->vm_vmcb + offsetof(struct vmcb, save.cr4));
+    if (new_cr4 & cr4_reserved_bits(env)) {
+        cpu_vmexit(env, SVM_EXIT_ERR, 0, GETPC());
+    }
     /* clear exit_info_2 so we behave like the real hardware */
     x86_stq_phys(cs,
              env->vm_vmcb + offsetof(struct vmcb, control.exit_info_2), 0);
 
     cpu_x86_update_cr0(env, new_cr0);
-    cpu_x86_update_cr4(env, x86_ldq_phys(cs,
-                                     env->vm_vmcb + offsetof(struct vmcb,
-                                                             save.cr4)));
-    cpu_x86_update_cr3(env, x86_ldq_phys(cs,
-                                     env->vm_vmcb + offsetof(struct vmcb,
-                                                             save.cr3)));
+    cpu_x86_update_cr4(env, new_cr4);
+    cpu_x86_update_cr3(env, new_cr3);
     env->cr[2] = x86_ldq_phys(cs,
                           env->vm_vmcb + offsetof(struct vmcb, save.cr2));
     int_ctl = x86_ldl_phys(cs,
@@ -278,6 +330,10 @@ void helper_vmrun(CPUX86State *env, int aflag, int next_eip_addend)
     }
 #endif
 
+    if (is_efer_invalid_state(env)) {
+        cpu_vmexit(env, SVM_EXIT_ERR, 0, GETPC());
+    }
+
     switch (x86_ldub_phys(cs,
                       env->vm_vmcb + offsetof(struct vmcb, control.tlb_ctl))) {
     case TLB_CONTROL_DO_NOTHING:
@@ -290,7 +346,7 @@ void helper_vmrun(CPUX86State *env, int aflag, int next_eip_addend)
 
     env->hflags2 |= HF2_GIF_MASK;
 
-    if (int_ctl & V_IRQ_MASK) {
+    if (ctl_has_irq(int_ctl)) {
         CPUState *cs = env_cpu(env);
 
         cs->interrupt_request |= CPU_INTERRUPT_VIRQ;
diff --git a/target/i386/tcg/tcg-cpu.c b/target/i386/tcg/tcg-cpu.c
index e96ec9bbcc..93a79a5741 100644
--- a/target/i386/tcg/tcg-cpu.c
+++ b/target/i386/tcg/tcg-cpu.c
@@ -54,6 +54,17 @@ static void x86_cpu_synchronize_from_tb(CPUState *cs,
     cpu->env.eip = tb->pc - tb->cs_base;
 }
 
+#ifndef CONFIG_USER_ONLY
+static bool x86_debug_check_breakpoint(CPUState *cs)
+{
+    X86CPU *cpu = X86_CPU(cs);
+    CPUX86State *env = &cpu->env;
+
+    /* RF disables all architectural breakpoints. */
+    return !(env->eflags & RF_MASK);
+}
+#endif
+
 #include "hw/core/tcg-cpu-ops.h"
 
 static const struct TCGCPUOps x86_tcg_ops = {
@@ -66,6 +77,7 @@ static const struct TCGCPUOps x86_tcg_ops = {
     .tlb_fill = x86_cpu_tlb_fill,
 #ifndef CONFIG_USER_ONLY
     .debug_excp_handler = breakpoint_handler,
+    .debug_check_breakpoint = x86_debug_check_breakpoint,
 #endif /* !CONFIG_USER_ONLY */
 };
 
@@ -99,7 +111,8 @@ static void tcg_cpu_xsave_init(void)
 }
 
 /*
- * TCG-specific defaults that override all CPU models when using TCG
+ * TCG-specific defaults that override cpudef models when using TCG.
+ * Only for builtin_x86_defs models initialized with x86_register_cpudef_types.
  */
 static PropValue tcg_default_props[] = {
     { "vme", "off" },
@@ -109,8 +122,12 @@ static PropValue tcg_default_props[] = {
 static void tcg_cpu_instance_init(CPUState *cs)
 {
     X86CPU *cpu = X86_CPU(cs);
-    /* Special cases not set in the X86CPUDefinition structs: */
-    x86_cpu_apply_props(cpu, tcg_default_props);
+    X86CPUClass *xcc = X86_CPU_GET_CLASS(cpu);
+
+    if (xcc->model) {
+        /* Special cases not set in the X86CPUDefinition structs: */
+        x86_cpu_apply_props(cpu, tcg_default_props);
+    }
 
     tcg_cpu_xsave_init();
 }
diff --git a/target/i386/tcg/translate.c b/target/i386/tcg/translate.c
index 8520d5a1e2..aacb605eee 100644
--- a/target/i386/tcg/translate.c
+++ b/target/i386/tcg/translate.c
@@ -2604,14 +2604,6 @@ static void gen_interrupt(DisasContext *s, int intno,
     s->base.is_jmp = DISAS_NORETURN;
 }
 
-static void gen_debug(DisasContext *s)
-{
-    gen_update_cc_op(s);
-    gen_jmp_im(s, s->base.pc_next - s->cs_base);
-    gen_helper_debug(cpu_env);
-    s->base.is_jmp = DISAS_NORETURN;
-}
-
 static void gen_set_hflag(DisasContext *s, uint32_t mask)
 {
     if ((s->flags & mask) == 0) {
@@ -8635,25 +8627,6 @@ static void i386_tr_insn_start(DisasContextBase *dcbase, CPUState *cpu)
     tcg_gen_insn_start(dc->base.pc_next, dc->cc_op);
 }
 
-static bool i386_tr_breakpoint_check(DisasContextBase *dcbase, CPUState *cpu,
-                                     const CPUBreakpoint *bp)
-{
-    DisasContext *dc = container_of(dcbase, DisasContext, base);
-    /* If RF is set, suppress an internally generated breakpoint.  */
-    int flags = dc->base.tb->flags & HF_RF_MASK ? BP_GDB : BP_ANY;
-    if (bp->flags & flags) {
-        gen_debug(dc);
-        /* The address covered by the breakpoint must be included in
-           [tb->pc, tb->pc + tb->size) in order to for it to be
-           properly cleared -- thus we increment the PC here so that
-           the generic logic setting tb->size later does the right thing.  */
-        dc->base.pc_next += 1;
-        return true;
-    } else {
-        return false;
-    }
-}
-
 static void i386_tr_translate_insn(DisasContextBase *dcbase, CPUState *cpu)
 {
     DisasContext *dc = container_of(dcbase, DisasContext, base);
@@ -8721,7 +8694,6 @@ static const TranslatorOps i386_tr_ops = {
     .init_disas_context = i386_tr_init_disas_context,
     .tb_start           = i386_tr_tb_start,
     .insn_start         = i386_tr_insn_start,
-    .breakpoint_check   = i386_tr_breakpoint_check,
     .translate_insn     = i386_tr_translate_insn,
     .tb_stop            = i386_tr_tb_stop,
     .disas_log          = i386_tr_disas_log,
diff --git a/target/m68k/op_helper.c b/target/m68k/op_helper.c
index ae1ba4b437..d006d1cb3e 100644
--- a/target/m68k/op_helper.c
+++ b/target/m68k/op_helper.c
@@ -22,6 +22,7 @@
 #include "exec/exec-all.h"
 #include "exec/cpu_ldst.h"
 #include "semihosting/semihost.h"
+#include "tcg/tcg.h"
 
 #if defined(CONFIG_USER_ONLY)
 
@@ -782,9 +783,9 @@ static void do_cas2l(CPUM68KState *env, uint32_t regs, uint32_t a1, uint32_t a2,
     uint32_t u2 = env->dregs[Du2];
     uint32_t l1, l2;
     uintptr_t ra = GETPC();
-#if defined(CONFIG_ATOMIC64) && !defined(CONFIG_USER_ONLY)
+#if defined(CONFIG_ATOMIC64)
     int mmu_idx = cpu_mmu_index(env, 0);
-    TCGMemOpIdx oi;
+    TCGMemOpIdx oi = make_memop_idx(MO_BEQ, mmu_idx);
 #endif
 
     if (parallel) {
@@ -794,23 +795,13 @@ static void do_cas2l(CPUM68KState *env, uint32_t regs, uint32_t a1, uint32_t a2,
         if ((a1 & 7) == 0 && a2 == a1 + 4) {
             c = deposit64(c2, 32, 32, c1);
             u = deposit64(u2, 32, 32, u1);
-#ifdef CONFIG_USER_ONLY
-            l = helper_atomic_cmpxchgq_be(env, a1, c, u);
-#else
-            oi = make_memop_idx(MO_BEQ, mmu_idx);
-            l = helper_atomic_cmpxchgq_be_mmu(env, a1, c, u, oi, ra);
-#endif
+            l = cpu_atomic_cmpxchgq_be_mmu(env, a1, c, u, oi, ra);
             l1 = l >> 32;
             l2 = l;
         } else if ((a2 & 7) == 0 && a1 == a2 + 4) {
             c = deposit64(c1, 32, 32, c2);
             u = deposit64(u1, 32, 32, u2);
-#ifdef CONFIG_USER_ONLY
-            l = helper_atomic_cmpxchgq_be(env, a2, c, u);
-#else
-            oi = make_memop_idx(MO_BEQ, mmu_idx);
-            l = helper_atomic_cmpxchgq_be_mmu(env, a2, c, u, oi, ra);
-#endif
+            l = cpu_atomic_cmpxchgq_be_mmu(env, a2, c, u, oi, ra);
             l2 = l >> 32;
             l1 = l;
         } else
diff --git a/target/m68k/translate.c b/target/m68k/translate.c
index 1fee04b8dd..c34d9aed61 100644
--- a/target/m68k/translate.c
+++ b/target/m68k/translate.c
@@ -6208,23 +6208,6 @@ static void m68k_tr_insn_start(DisasContextBase *dcbase, CPUState *cpu)
     tcg_gen_insn_start(dc->base.pc_next, dc->cc_op);
 }
 
-static bool m68k_tr_breakpoint_check(DisasContextBase *dcbase, CPUState *cpu,
-                                     const CPUBreakpoint *bp)
-{
-    DisasContext *dc = container_of(dcbase, DisasContext, base);
-
-    gen_exception(dc, dc->base.pc_next, EXCP_DEBUG);
-    /*
-     * The address covered by the breakpoint must be included in
-     * [tb->pc, tb->pc + tb->size) in order to for it to be
-     * properly cleared -- thus we increment the PC here so that
-     * the logic setting tb->size below does the right thing.
-     */
-    dc->base.pc_next += 2;
-
-    return true;
-}
-
 static void m68k_tr_translate_insn(DisasContextBase *dcbase, CPUState *cpu)
 {
     DisasContext *dc = container_of(dcbase, DisasContext, base);
@@ -6310,7 +6293,6 @@ static const TranslatorOps m68k_tr_ops = {
     .init_disas_context = m68k_tr_init_disas_context,
     .tb_start           = m68k_tr_tb_start,
     .insn_start         = m68k_tr_insn_start,
-    .breakpoint_check   = m68k_tr_breakpoint_check,
     .translate_insn     = m68k_tr_translate_insn,
     .tb_stop            = m68k_tr_tb_stop,
     .disas_log          = m68k_tr_disas_log,
diff --git a/target/microblaze/translate.c b/target/microblaze/translate.c
index c68a84a219..a14ffed784 100644
--- a/target/microblaze/translate.c
+++ b/target/microblaze/translate.c
@@ -1673,23 +1673,6 @@ static void mb_tr_insn_start(DisasContextBase *dcb, CPUState *cs)
     dc->insn_start = tcg_last_op();
 }
 
-static bool mb_tr_breakpoint_check(DisasContextBase *dcb, CPUState *cs,
-                                   const CPUBreakpoint *bp)
-{
-    DisasContext *dc = container_of(dcb, DisasContext, base);
-
-    gen_raise_exception_sync(dc, EXCP_DEBUG);
-
-    /*
-     * The address covered by the breakpoint must be included in
-     * [tb->pc, tb->pc + tb->size) in order to for it to be
-     * properly cleared -- thus we increment the PC here so that
-     * the logic setting tb->size below does the right thing.
-     */
-    dc->base.pc_next += 4;
-    return true;
-}
-
 static void mb_tr_translate_insn(DisasContextBase *dcb, CPUState *cs)
 {
     DisasContext *dc = container_of(dcb, DisasContext, base);
@@ -1854,7 +1837,6 @@ static const TranslatorOps mb_tr_ops = {
     .init_disas_context = mb_tr_init_disas_context,
     .tb_start           = mb_tr_tb_start,
     .insn_start         = mb_tr_insn_start,
-    .breakpoint_check   = mb_tr_breakpoint_check,
     .translate_insn     = mb_tr_translate_insn,
     .tb_stop            = mb_tr_tb_stop,
     .disas_log          = mb_tr_disas_log,
diff --git a/target/mips/tcg/translate.c b/target/mips/tcg/translate.c
index fd980ea966..5b03545f09 100644
--- a/target/mips/tcg/translate.c
+++ b/target/mips/tcg/translate.c
@@ -16178,24 +16178,6 @@ static void mips_tr_insn_start(DisasContextBase *dcbase, CPUState *cs)
                        ctx->btarget);
 }
 
-static bool mips_tr_breakpoint_check(DisasContextBase *dcbase, CPUState *cs,
-                                     const CPUBreakpoint *bp)
-{
-    DisasContext *ctx = container_of(dcbase, DisasContext, base);
-
-    save_cpu_state(ctx, 1);
-    ctx->base.is_jmp = DISAS_NORETURN;
-    gen_helper_raise_exception_debug(cpu_env);
-    /*
-     * The address covered by the breakpoint must be included in
-     * [tb->pc, tb->pc + tb->size) in order to for it to be
-     * properly cleared -- thus we increment the PC here so that
-     * the logic setting tb->size below does the right thing.
-     */
-    ctx->base.pc_next += 4;
-    return true;
-}
-
 static void mips_tr_translate_insn(DisasContextBase *dcbase, CPUState *cs)
 {
     CPUMIPSState *env = cs->env_ptr;
@@ -16303,7 +16285,6 @@ static const TranslatorOps mips_tr_ops = {
     .init_disas_context = mips_tr_init_disas_context,
     .tb_start           = mips_tr_tb_start,
     .insn_start         = mips_tr_insn_start,
-    .breakpoint_check   = mips_tr_breakpoint_check,
     .translate_insn     = mips_tr_translate_insn,
     .tb_stop            = mips_tr_tb_stop,
     .disas_log          = mips_tr_disas_log,
diff --git a/target/nios2/translate.c b/target/nios2/translate.c
index 17742cebc7..08d7ac5398 100644
--- a/target/nios2/translate.c
+++ b/target/nios2/translate.c
@@ -744,16 +744,6 @@ static const char * const regnames[] = {
 
 #include "exec/gen-icount.h"
 
-static void gen_exception(DisasContext *dc, uint32_t excp)
-{
-    TCGv_i32 tmp = tcg_const_i32(excp);
-
-    tcg_gen_movi_tl(cpu_R[R_PC], dc->pc);
-    gen_helper_raise_exception(cpu_env, tmp);
-    tcg_temp_free_i32(tmp);
-    dc->base.is_jmp = DISAS_NORETURN;
-}
-
 /* generate intermediate code for basic block 'tb'.  */
 static void nios2_tr_init_disas_context(DisasContextBase *dcbase, CPUState *cs)
 {
@@ -777,22 +767,6 @@ static void nios2_tr_insn_start(DisasContextBase *dcbase, CPUState *cs)
     tcg_gen_insn_start(dcbase->pc_next);
 }
 
-static bool nios2_tr_breakpoint_check(DisasContextBase *dcbase, CPUState *cs,
-                                      const CPUBreakpoint *bp)
-{
-    DisasContext *dc = container_of(dcbase, DisasContext, base);
-
-    gen_exception(dc, EXCP_DEBUG);
-    /*
-     * The address covered by the breakpoint must be included in
-     * [tb->pc, tb->pc + tb->size) in order to for it to be
-     * properly cleared -- thus we increment the PC here so that
-     * the logic setting tb->size below does the right thing.
-     */
-    dc->base.pc_next += 4;
-    return true;
-}
-
 static void nios2_tr_translate_insn(DisasContextBase *dcbase, CPUState *cs)
 {
     DisasContext *dc = container_of(dcbase, DisasContext, base);
@@ -870,7 +844,6 @@ static const TranslatorOps nios2_tr_ops = {
     .init_disas_context = nios2_tr_init_disas_context,
     .tb_start           = nios2_tr_tb_start,
     .insn_start         = nios2_tr_insn_start,
-    .breakpoint_check   = nios2_tr_breakpoint_check,
     .translate_insn     = nios2_tr_translate_insn,
     .tb_stop            = nios2_tr_tb_stop,
     .disas_log          = nios2_tr_disas_log,
diff --git a/target/openrisc/translate.c b/target/openrisc/translate.c
index 059da48475..d6ea536744 100644
--- a/target/openrisc/translate.c
+++ b/target/openrisc/translate.c
@@ -1609,22 +1609,6 @@ static void openrisc_tr_insn_start(DisasContextBase *dcbase, CPUState *cs)
                        | (dc->base.num_insns > 1 ? 2 : 0));
 }
 
-static bool openrisc_tr_breakpoint_check(DisasContextBase *dcbase, CPUState *cs,
-                                         const CPUBreakpoint *bp)
-{
-    DisasContext *dc = container_of(dcbase, DisasContext, base);
-
-    tcg_gen_movi_tl(cpu_pc, dc->base.pc_next);
-    gen_exception(dc, EXCP_DEBUG);
-    dc->base.is_jmp = DISAS_NORETURN;
-    /* The address covered by the breakpoint must be included in
-       [tb->pc, tb->pc + tb->size) in order to for it to be
-       properly cleared -- thus we increment the PC here so that
-       the logic setting tb->size below does the right thing.  */
-    dc->base.pc_next += 4;
-    return true;
-}
-
 static void openrisc_tr_translate_insn(DisasContextBase *dcbase, CPUState *cs)
 {
     DisasContext *dc = container_of(dcbase, DisasContext, base);
@@ -1727,7 +1711,6 @@ static const TranslatorOps openrisc_tr_ops = {
     .init_disas_context = openrisc_tr_init_disas_context,
     .tb_start           = openrisc_tr_tb_start,
     .insn_start         = openrisc_tr_insn_start,
-    .breakpoint_check   = openrisc_tr_breakpoint_check,
     .translate_insn     = openrisc_tr_translate_insn,
     .tb_stop            = openrisc_tr_tb_stop,
     .disas_log          = openrisc_tr_disas_log,
diff --git a/target/ppc/mem_helper.c b/target/ppc/mem_helper.c
index 444b2a30ef..e2282baa8d 100644
--- a/target/ppc/mem_helper.c
+++ b/target/ppc/mem_helper.c
@@ -376,7 +376,7 @@ uint64_t helper_lq_le_parallel(CPUPPCState *env, target_ulong addr,
 
     /* We will have raised EXCP_ATOMIC from the translator.  */
     assert(HAVE_ATOMIC128);
-    ret = helper_atomic_ldo_le_mmu(env, addr, opidx, GETPC());
+    ret = cpu_atomic_ldo_le_mmu(env, addr, opidx, GETPC());
     env->retxh = int128_gethi(ret);
     return int128_getlo(ret);
 }
@@ -388,7 +388,7 @@ uint64_t helper_lq_be_parallel(CPUPPCState *env, target_ulong addr,
 
     /* We will have raised EXCP_ATOMIC from the translator.  */
     assert(HAVE_ATOMIC128);
-    ret = helper_atomic_ldo_be_mmu(env, addr, opidx, GETPC());
+    ret = cpu_atomic_ldo_be_mmu(env, addr, opidx, GETPC());
     env->retxh = int128_gethi(ret);
     return int128_getlo(ret);
 }
@@ -401,7 +401,7 @@ void helper_stq_le_parallel(CPUPPCState *env, target_ulong addr,
     /* We will have raised EXCP_ATOMIC from the translator.  */
     assert(HAVE_ATOMIC128);
     val = int128_make128(lo, hi);
-    helper_atomic_sto_le_mmu(env, addr, val, opidx, GETPC());
+    cpu_atomic_sto_le_mmu(env, addr, val, opidx, GETPC());
 }
 
 void helper_stq_be_parallel(CPUPPCState *env, target_ulong addr,
@@ -412,7 +412,7 @@ void helper_stq_be_parallel(CPUPPCState *env, target_ulong addr,
     /* We will have raised EXCP_ATOMIC from the translator.  */
     assert(HAVE_ATOMIC128);
     val = int128_make128(lo, hi);
-    helper_atomic_sto_be_mmu(env, addr, val, opidx, GETPC());
+    cpu_atomic_sto_be_mmu(env, addr, val, opidx, GETPC());
 }
 
 uint32_t helper_stqcx_le_parallel(CPUPPCState *env, target_ulong addr,
@@ -429,8 +429,8 @@ uint32_t helper_stqcx_le_parallel(CPUPPCState *env, target_ulong addr,
 
         cmpv = int128_make128(env->reserve_val2, env->reserve_val);
         newv = int128_make128(new_lo, new_hi);
-        oldv = helper_atomic_cmpxchgo_le_mmu(env, addr, cmpv, newv,
-                                             opidx, GETPC());
+        oldv = cpu_atomic_cmpxchgo_le_mmu(env, addr, cmpv, newv,
+                                          opidx, GETPC());
         success = int128_eq(oldv, cmpv);
     }
     env->reserve_addr = -1;
@@ -451,8 +451,8 @@ uint32_t helper_stqcx_be_parallel(CPUPPCState *env, target_ulong addr,
 
         cmpv = int128_make128(env->reserve_val2, env->reserve_val);
         newv = int128_make128(new_lo, new_hi);
-        oldv = helper_atomic_cmpxchgo_be_mmu(env, addr, cmpv, newv,
-                                             opidx, GETPC());
+        oldv = cpu_atomic_cmpxchgo_be_mmu(env, addr, cmpv, newv,
+                                          opidx, GETPC());
         success = int128_eq(oldv, cmpv);
     }
     env->reserve_addr = -1;
diff --git a/target/ppc/translate.c b/target/ppc/translate.c
index 0a55cb7181..171b216e17 100644
--- a/target/ppc/translate.c
+++ b/target/ppc/translate.c
@@ -8565,23 +8565,6 @@ static void ppc_tr_insn_start(DisasContextBase *dcbase, CPUState *cs)
     tcg_gen_insn_start(dcbase->pc_next);
 }
 
-static bool ppc_tr_breakpoint_check(DisasContextBase *dcbase, CPUState *cs,
-                                    const CPUBreakpoint *bp)
-{
-    DisasContext *ctx = container_of(dcbase, DisasContext, base);
-
-    gen_update_nip(ctx, ctx->base.pc_next);
-    gen_debug_exception(ctx);
-    /*
-     * The address covered by the breakpoint must be included in
-     * [tb->pc, tb->pc + tb->size) in order to for it to be properly
-     * cleared -- thus we increment the PC here so that the logic
-     * setting tb->size below does the right thing.
-     */
-    ctx->base.pc_next += 4;
-    return true;
-}
-
 static bool is_prefix_insn(DisasContext *ctx, uint32_t insn)
 {
     REQUIRE_INSNS_FLAGS2(ctx, ISA310);
@@ -8710,7 +8693,6 @@ static const TranslatorOps ppc_tr_ops = {
     .init_disas_context = ppc_tr_init_disas_context,
     .tb_start           = ppc_tr_tb_start,
     .insn_start         = ppc_tr_insn_start,
-    .breakpoint_check   = ppc_tr_breakpoint_check,
     .translate_insn     = ppc_tr_translate_insn,
     .tb_stop            = ppc_tr_tb_stop,
     .disas_log          = ppc_tr_disas_log,
diff --git a/target/riscv/translate.c b/target/riscv/translate.c
index deda0c8a44..6983be5723 100644
--- a/target/riscv/translate.c
+++ b/target/riscv/translate.c
@@ -961,22 +961,6 @@ static void riscv_tr_insn_start(DisasContextBase *dcbase, CPUState *cpu)
     tcg_gen_insn_start(ctx->base.pc_next);
 }
 
-static bool riscv_tr_breakpoint_check(DisasContextBase *dcbase, CPUState *cpu,
-                                      const CPUBreakpoint *bp)
-{
-    DisasContext *ctx = container_of(dcbase, DisasContext, base);
-
-    tcg_gen_movi_tl(cpu_pc, ctx->base.pc_next);
-    ctx->base.is_jmp = DISAS_NORETURN;
-    gen_exception_debug();
-    /* The address covered by the breakpoint must be included in
-       [tb->pc, tb->pc + tb->size) in order to for it to be
-       properly cleared -- thus we increment the PC here so that
-       the logic setting tb->size below does the right thing.  */
-    ctx->base.pc_next += 4;
-    return true;
-}
-
 static void riscv_tr_translate_insn(DisasContextBase *dcbase, CPUState *cpu)
 {
     DisasContext *ctx = container_of(dcbase, DisasContext, base);
@@ -1029,7 +1013,6 @@ static const TranslatorOps riscv_tr_ops = {
     .init_disas_context = riscv_tr_init_disas_context,
     .tb_start           = riscv_tr_tb_start,
     .insn_start         = riscv_tr_insn_start,
-    .breakpoint_check   = riscv_tr_breakpoint_check,
     .translate_insn     = riscv_tr_translate_insn,
     .tb_stop            = riscv_tr_tb_stop,
     .disas_log          = riscv_tr_disas_log,
diff --git a/target/rx/translate.c b/target/rx/translate.c
index 23a626438a..a3cf720455 100644
--- a/target/rx/translate.c
+++ b/target/rx/translate.c
@@ -2309,19 +2309,6 @@ static void rx_tr_insn_start(DisasContextBase *dcbase, CPUState *cs)
     tcg_gen_insn_start(ctx->base.pc_next);
 }
 
-static bool rx_tr_breakpoint_check(DisasContextBase *dcbase, CPUState *cs,
-                                    const CPUBreakpoint *bp)
-{
-    DisasContext *ctx = container_of(dcbase, DisasContext, base);
-
-    /* We have hit a breakpoint - make sure PC is up-to-date */
-    tcg_gen_movi_i32(cpu_pc, ctx->base.pc_next);
-    gen_helper_debug(cpu_env);
-    ctx->base.is_jmp = DISAS_NORETURN;
-    ctx->base.pc_next += 1;
-    return true;
-}
-
 static void rx_tr_translate_insn(DisasContextBase *dcbase, CPUState *cs)
 {
     DisasContext *ctx = container_of(dcbase, DisasContext, base);
@@ -2373,7 +2360,6 @@ static const TranslatorOps rx_tr_ops = {
     .init_disas_context = rx_tr_init_disas_context,
     .tb_start           = rx_tr_tb_start,
     .insn_start         = rx_tr_insn_start,
-    .breakpoint_check   = rx_tr_breakpoint_check,
     .translate_insn     = rx_tr_translate_insn,
     .tb_stop            = rx_tr_tb_stop,
     .disas_log          = rx_tr_disas_log,
diff --git a/target/s390x/tcg/mem_helper.c b/target/s390x/tcg/mem_helper.c
index 9bae13ecf0..21a4de4067 100644
--- a/target/s390x/tcg/mem_helper.c
+++ b/target/s390x/tcg/mem_helper.c
@@ -1811,7 +1811,7 @@ void HELPER(cdsg_parallel)(CPUS390XState *env, uint64_t addr,
 
     mem_idx = cpu_mmu_index(env, false);
     oi = make_memop_idx(MO_TEQ | MO_ALIGN_16, mem_idx);
-    oldv = helper_atomic_cmpxchgo_be_mmu(env, addr, cmpv, newv, oi, ra);
+    oldv = cpu_atomic_cmpxchgo_be_mmu(env, addr, cmpv, newv, oi, ra);
     fail = !int128_eq(oldv, cmpv);
 
     env->cc_op = fail;
@@ -1884,7 +1884,7 @@ static uint32_t do_csst(CPUS390XState *env, uint32_t r3, uint64_t a1,
                 ov = qatomic_cmpxchg__nocheck(haddr, cv, nv);
 #else
                 TCGMemOpIdx oi = make_memop_idx(MO_TEUL | MO_ALIGN, mem_idx);
-                ov = helper_atomic_cmpxchgl_be_mmu(env, a1, cv, nv, oi, ra);
+                ov = cpu_atomic_cmpxchgl_be_mmu(env, a1, cv, nv, oi, ra);
 #endif
             } else {
                 ov = cpu_ldl_data_ra(env, a1, ra);
@@ -1903,13 +1903,8 @@ static uint32_t do_csst(CPUS390XState *env, uint32_t r3, uint64_t a1,
 
             if (parallel) {
 #ifdef CONFIG_ATOMIC64
-# ifdef CONFIG_USER_ONLY
-                uint64_t *haddr = g2h(env_cpu(env), a1);
-                ov = qatomic_cmpxchg__nocheck(haddr, cv, nv);
-# else
                 TCGMemOpIdx oi = make_memop_idx(MO_TEQ | MO_ALIGN, mem_idx);
-                ov = helper_atomic_cmpxchgq_be_mmu(env, a1, cv, nv, oi, ra);
-# endif
+                ov = cpu_atomic_cmpxchgq_be_mmu(env, a1, cv, nv, oi, ra);
 #else
                 /* Note that we asserted !parallel above.  */
                 g_assert_not_reached();
@@ -1945,7 +1940,7 @@ static uint32_t do_csst(CPUS390XState *env, uint32_t r3, uint64_t a1,
                 cpu_stq_data_ra(env, a1 + 8, int128_getlo(nv), ra);
             } else if (HAVE_CMPXCHG128) {
                 TCGMemOpIdx oi = make_memop_idx(MO_TEQ | MO_ALIGN_16, mem_idx);
-                ov = helper_atomic_cmpxchgo_be_mmu(env, a1, cv, nv, oi, ra);
+                ov = cpu_atomic_cmpxchgo_be_mmu(env, a1, cv, nv, oi, ra);
                 cc = !int128_eq(ov, cv);
             } else {
                 /* Note that we asserted !parallel above.  */
@@ -1985,7 +1980,7 @@ static uint32_t do_csst(CPUS390XState *env, uint32_t r3, uint64_t a1,
             } else if (HAVE_ATOMIC128) {
                 TCGMemOpIdx oi = make_memop_idx(MO_TEQ | MO_ALIGN_16, mem_idx);
                 Int128 sv = int128_make128(svl, svh);
-                helper_atomic_sto_be_mmu(env, a2, sv, oi, ra);
+                cpu_atomic_sto_be_mmu(env, a2, sv, oi, ra);
             } else {
                 /* Note that we asserted !parallel above.  */
                 g_assert_not_reached();
@@ -2486,7 +2481,7 @@ uint64_t HELPER(lpq_parallel)(CPUS390XState *env, uint64_t addr)
 
     mem_idx = cpu_mmu_index(env, false);
     oi = make_memop_idx(MO_TEQ | MO_ALIGN_16, mem_idx);
-    v = helper_atomic_ldo_be_mmu(env, addr, oi, ra);
+    v = cpu_atomic_ldo_be_mmu(env, addr, oi, ra);
     hi = int128_gethi(v);
     lo = int128_getlo(v);
 
@@ -2518,7 +2513,7 @@ void HELPER(stpq_parallel)(CPUS390XState *env, uint64_t addr,
     mem_idx = cpu_mmu_index(env, false);
     oi = make_memop_idx(MO_TEQ | MO_ALIGN_16, mem_idx);
     v = int128_make128(low, high);
-    helper_atomic_sto_be_mmu(env, addr, v, oi, ra);
+    cpu_atomic_sto_be_mmu(env, addr, v, oi, ra);
 }
 
 /* Execute instruction.  This instruction executes an insn modified with
diff --git a/target/s390x/tcg/translate.c b/target/s390x/tcg/translate.c
index 92fa7656c2..0632b0374b 100644
--- a/target/s390x/tcg/translate.c
+++ b/target/s390x/tcg/translate.c
@@ -6552,29 +6552,6 @@ static void s390x_tr_insn_start(DisasContextBase *dcbase, CPUState *cs)
 {
 }
 
-static bool s390x_tr_breakpoint_check(DisasContextBase *dcbase, CPUState *cs,
-                                      const CPUBreakpoint *bp)
-{
-    DisasContext *dc = container_of(dcbase, DisasContext, base);
-
-    /*
-     * Emit an insn_start to accompany the breakpoint exception.
-     * The ILEN value is a dummy, since this does not result in
-     * an s390x exception, but an internal qemu exception which
-     * brings us back to interact with the gdbstub.
-     */
-    tcg_gen_insn_start(dc->base.pc_next, dc->cc_op, 2);
-
-    dc->base.is_jmp = DISAS_PC_STALE;
-    dc->do_debug = true;
-    /* The address covered by the breakpoint must be included in
-       [tb->pc, tb->pc + tb->size) in order to for it to be
-       properly cleared -- thus we increment the PC here so that
-       the logic setting tb->size does the right thing.  */
-    dc->base.pc_next += 2;
-    return true;
-}
-
 static void s390x_tr_translate_insn(DisasContextBase *dcbase, CPUState *cs)
 {
     CPUS390XState *env = cs->env_ptr;
@@ -6642,7 +6619,6 @@ static const TranslatorOps s390x_tr_ops = {
     .init_disas_context = s390x_tr_init_disas_context,
     .tb_start           = s390x_tr_tb_start,
     .insn_start         = s390x_tr_insn_start,
-    .breakpoint_check   = s390x_tr_breakpoint_check,
     .translate_insn     = s390x_tr_translate_insn,
     .tb_stop            = s390x_tr_tb_stop,
     .disas_log          = s390x_tr_disas_log,
diff --git a/target/sh4/translate.c b/target/sh4/translate.c
index 40898e2393..8704fea1ca 100644
--- a/target/sh4/translate.c
+++ b/target/sh4/translate.c
@@ -2289,23 +2289,6 @@ static void sh4_tr_insn_start(DisasContextBase *dcbase, CPUState *cs)
     tcg_gen_insn_start(ctx->base.pc_next, ctx->envflags);
 }
 
-static bool sh4_tr_breakpoint_check(DisasContextBase *dcbase, CPUState *cs,
-                                    const CPUBreakpoint *bp)
-{
-    DisasContext *ctx = container_of(dcbase, DisasContext, base);
-
-    /* We have hit a breakpoint - make sure PC is up-to-date */
-    gen_save_cpu_state(ctx, true);
-    gen_helper_debug(cpu_env);
-    ctx->base.is_jmp = DISAS_NORETURN;
-    /* The address covered by the breakpoint must be included in
-       [tb->pc, tb->pc + tb->size) in order to for it to be
-       properly cleared -- thus we increment the PC here so that
-       the logic setting tb->size below does the right thing.  */
-    ctx->base.pc_next += 2;
-    return true;
-}
-
 static void sh4_tr_translate_insn(DisasContextBase *dcbase, CPUState *cs)
 {
     CPUSH4State *env = cs->env_ptr;
@@ -2369,7 +2352,6 @@ static const TranslatorOps sh4_tr_ops = {
     .init_disas_context = sh4_tr_init_disas_context,
     .tb_start           = sh4_tr_tb_start,
     .insn_start         = sh4_tr_insn_start,
-    .breakpoint_check   = sh4_tr_breakpoint_check,
     .translate_insn     = sh4_tr_translate_insn,
     .tb_stop            = sh4_tr_tb_stop,
     .disas_log          = sh4_tr_disas_log,
diff --git a/target/sparc/translate.c b/target/sparc/translate.c
index e530cb4aa8..11de5a4963 100644
--- a/target/sparc/translate.c
+++ b/target/sparc/translate.c
@@ -5854,22 +5854,6 @@ static void sparc_tr_insn_start(DisasContextBase *dcbase, CPUState *cs)
     }
 }
 
-static bool sparc_tr_breakpoint_check(DisasContextBase *dcbase, CPUState *cs,
-                                      const CPUBreakpoint *bp)
-{
-    DisasContext *dc = container_of(dcbase, DisasContext, base);
-
-    if (dc->pc != dc->base.pc_first) {
-        save_state(dc);
-    }
-    gen_helper_debug(cpu_env);
-    tcg_gen_exit_tb(NULL, 0);
-    dc->base.is_jmp = DISAS_NORETURN;
-    /* update pc_next so that the current instruction is included in tb->size */
-    dc->base.pc_next += 4;
-    return true;
-}
-
 static void sparc_tr_translate_insn(DisasContextBase *dcbase, CPUState *cs)
 {
     DisasContext *dc = container_of(dcbase, DisasContext, base);
@@ -5932,7 +5916,6 @@ static const TranslatorOps sparc_tr_ops = {
     .init_disas_context = sparc_tr_init_disas_context,
     .tb_start           = sparc_tr_tb_start,
     .insn_start         = sparc_tr_insn_start,
-    .breakpoint_check   = sparc_tr_breakpoint_check,
     .translate_insn     = sparc_tr_translate_insn,
     .tb_stop            = sparc_tr_tb_stop,
     .disas_log          = sparc_tr_disas_log,
diff --git a/target/tricore/translate.c b/target/tricore/translate.c
index 865020754d..a0cc0f1cb3 100644
--- a/target/tricore/translate.c
+++ b/target/tricore/translate.c
@@ -8810,21 +8810,6 @@ static void tricore_tr_insn_start(DisasContextBase *dcbase, CPUState *cpu)
     tcg_gen_insn_start(ctx->base.pc_next);
 }
 
-static bool tricore_tr_breakpoint_check(DisasContextBase *dcbase, CPUState *cpu,
-                                      const CPUBreakpoint *bp)
-{
-    DisasContext *ctx = container_of(dcbase, DisasContext, base);
-    generate_qemu_excp(ctx, EXCP_DEBUG);
-    /*
-     * The address covered by the breakpoint must be included in
-     * [tb->pc, tb->pc + tb->size) in order to for it to be
-     * properly cleared -- thus we increment the PC here so that
-     * the logic setting tb->size below does the right thing.
-     */
-    ctx->base.pc_next += 4;
-    return true;
-}
-
 static bool insn_crosses_page(CPUTriCoreState *env, DisasContext *ctx)
 {
     /*
@@ -8898,7 +8883,6 @@ static const TranslatorOps tricore_tr_ops = {
     .init_disas_context = tricore_tr_init_disas_context,
     .tb_start           = tricore_tr_tb_start,
     .insn_start         = tricore_tr_insn_start,
-    .breakpoint_check   = tricore_tr_breakpoint_check,
     .translate_insn     = tricore_tr_translate_insn,
     .tb_stop            = tricore_tr_tb_stop,
     .disas_log          = tricore_tr_disas_log,
diff --git a/target/xtensa/translate.c b/target/xtensa/translate.c
index 7094cfcf1d..20399d6a04 100644
--- a/target/xtensa/translate.c
+++ b/target/xtensa/translate.c
@@ -1232,22 +1232,6 @@ static void xtensa_tr_insn_start(DisasContextBase *dcbase, CPUState *cpu)
     tcg_gen_insn_start(dcbase->pc_next);
 }
 
-static bool xtensa_tr_breakpoint_check(DisasContextBase *dcbase, CPUState *cpu,
-                                       const CPUBreakpoint *bp)
-{
-    DisasContext *dc = container_of(dcbase, DisasContext, base);
-
-    tcg_gen_movi_i32(cpu_pc, dc->base.pc_next);
-    gen_exception(dc, EXCP_DEBUG);
-    dc->base.is_jmp = DISAS_NORETURN;
-    /* The address covered by the breakpoint must be included in
-       [tb->pc, tb->pc + tb->size) in order to for it to be
-       properly cleared -- thus we increment the PC here so that
-       the logic setting tb->size below does the right thing.  */
-    dc->base.pc_next += 2;
-    return true;
-}
-
 static void xtensa_tr_translate_insn(DisasContextBase *dcbase, CPUState *cpu)
 {
     DisasContext *dc = container_of(dcbase, DisasContext, base);
@@ -1330,7 +1314,6 @@ static const TranslatorOps xtensa_translator_ops = {
     .init_disas_context = xtensa_tr_init_disas_context,
     .tb_start           = xtensa_tr_tb_start,
     .insn_start         = xtensa_tr_insn_start,
-    .breakpoint_check   = xtensa_tr_breakpoint_check,
     .translate_insn     = xtensa_tr_translate_insn,
     .tb_stop            = xtensa_tr_tb_stop,
     .disas_log          = xtensa_tr_disas_log,
diff --git a/tcg/tcg-op.c b/tcg/tcg-op.c
index 0c561fb253..c754396575 100644
--- a/tcg/tcg-op.c
+++ b/tcg/tcg-op.c
@@ -2723,10 +2723,6 @@ void tcg_gen_exit_tb(const TranslationBlock *tb, unsigned idx)
            seen this numbered exit before, via tcg_gen_goto_tb.  */
         tcg_debug_assert(tcg_ctx->goto_tb_issue_mask & (1 << idx));
 #endif
-        /* When not chaining, exit without indicating a link.  */
-        if (qemu_loglevel_mask(CPU_LOG_TB_NOCHAIN)) {
-            val = 0;
-        }
     } else {
         /* This is an exit via the exitreq label.  */
         tcg_debug_assert(idx == TB_EXIT_REQUESTED);
@@ -2738,6 +2734,8 @@ void tcg_gen_exit_tb(const TranslationBlock *tb, unsigned idx)
 
 void tcg_gen_goto_tb(unsigned idx)
 {
+    /* We tested CF_NO_GOTO_TB in translator_use_goto_tb. */
+    tcg_debug_assert(!(tcg_ctx->tb_cflags & CF_NO_GOTO_TB));
     /* We only support two chained exits.  */
     tcg_debug_assert(idx <= TB_EXIT_IDXMAX);
 #ifdef CONFIG_DEBUG_TCG
@@ -2746,25 +2744,23 @@ void tcg_gen_goto_tb(unsigned idx)
     tcg_ctx->goto_tb_issue_mask |= 1 << idx;
 #endif
     plugin_gen_disable_mem_helpers();
-    /* When not chaining, we simply fall through to the "fallback" exit.  */
-    if (!qemu_loglevel_mask(CPU_LOG_TB_NOCHAIN)) {
-        tcg_gen_op1i(INDEX_op_goto_tb, idx);
-    }
+    tcg_gen_op1i(INDEX_op_goto_tb, idx);
 }
 
 void tcg_gen_lookup_and_goto_ptr(void)
 {
-    if (!qemu_loglevel_mask(CPU_LOG_TB_NOCHAIN)) {
-        TCGv_ptr ptr;
+    TCGv_ptr ptr;
 
-        plugin_gen_disable_mem_helpers();
-        ptr = tcg_temp_new_ptr();
-        gen_helper_lookup_tb_ptr(ptr, cpu_env);
-        tcg_gen_op1i(INDEX_op_goto_ptr, tcgv_ptr_arg(ptr));
-        tcg_temp_free_ptr(ptr);
-    } else {
+    if (tcg_ctx->tb_cflags & CF_NO_GOTO_PTR) {
         tcg_gen_exit_tb(NULL, 0);
+        return;
     }
+
+    plugin_gen_disable_mem_helpers();
+    ptr = tcg_temp_new_ptr();
+    gen_helper_lookup_tb_ptr(ptr, cpu_env);
+    tcg_gen_op1i(INDEX_op_goto_ptr, tcgv_ptr_arg(ptr));
+    tcg_temp_free_ptr(ptr);
 }
 
 static inline MemOp tcg_canonicalize_memop(MemOp op, bool is64, bool st)
@@ -3084,7 +3080,6 @@ static void tcg_gen_ext_i64(TCGv_i64 ret, TCGv_i64 val, MemOp opc)
     }
 }
 
-#ifdef CONFIG_SOFTMMU
 typedef void (*gen_atomic_cx_i32)(TCGv_i32, TCGv_env, TCGv,
                                   TCGv_i32, TCGv_i32, TCGv_i32);
 typedef void (*gen_atomic_cx_i64)(TCGv_i64, TCGv_env, TCGv,
@@ -3093,12 +3088,6 @@ typedef void (*gen_atomic_op_i32)(TCGv_i32, TCGv_env, TCGv,
                                   TCGv_i32, TCGv_i32);
 typedef void (*gen_atomic_op_i64)(TCGv_i64, TCGv_env, TCGv,
                                   TCGv_i64, TCGv_i32);
-#else
-typedef void (*gen_atomic_cx_i32)(TCGv_i32, TCGv_env, TCGv, TCGv_i32, TCGv_i32);
-typedef void (*gen_atomic_cx_i64)(TCGv_i64, TCGv_env, TCGv, TCGv_i64, TCGv_i64);
-typedef void (*gen_atomic_op_i32)(TCGv_i32, TCGv_env, TCGv, TCGv_i32);
-typedef void (*gen_atomic_op_i64)(TCGv_i64, TCGv_env, TCGv, TCGv_i64);
-#endif
 
 #ifdef CONFIG_ATOMIC64
 # define WITH_ATOMIC64(X) X,
@@ -3140,18 +3129,13 @@ void tcg_gen_atomic_cmpxchg_i32(TCGv_i32 retv, TCGv addr, TCGv_i32 cmpv,
         tcg_temp_free_i32(t1);
     } else {
         gen_atomic_cx_i32 gen;
+        TCGMemOpIdx oi;
 
         gen = table_cmpxchg[memop & (MO_SIZE | MO_BSWAP)];
         tcg_debug_assert(gen != NULL);
 
-#ifdef CONFIG_SOFTMMU
-        {
-            TCGMemOpIdx oi = make_memop_idx(memop & ~MO_SIGN, idx);
-            gen(retv, cpu_env, addr, cmpv, newv, tcg_constant_i32(oi));
-        }
-#else
-        gen(retv, cpu_env, addr, cmpv, newv);
-#endif
+        oi = make_memop_idx(memop & ~MO_SIGN, idx);
+        gen(retv, cpu_env, addr, cmpv, newv, tcg_constant_i32(oi));
 
         if (memop & MO_SIGN) {
             tcg_gen_ext_i32(retv, retv, memop);
@@ -3184,18 +3168,13 @@ void tcg_gen_atomic_cmpxchg_i64(TCGv_i64 retv, TCGv addr, TCGv_i64 cmpv,
     } else if ((memop & MO_SIZE) == MO_64) {
 #ifdef CONFIG_ATOMIC64
         gen_atomic_cx_i64 gen;
+        TCGMemOpIdx oi;
 
         gen = table_cmpxchg[memop & (MO_SIZE | MO_BSWAP)];
         tcg_debug_assert(gen != NULL);
 
-#ifdef CONFIG_SOFTMMU
-        {
-            TCGMemOpIdx oi = make_memop_idx(memop, idx);
-            gen(retv, cpu_env, addr, cmpv, newv, tcg_constant_i32(oi));
-        }
-#else
-        gen(retv, cpu_env, addr, cmpv, newv);
-#endif
+        oi = make_memop_idx(memop, idx);
+        gen(retv, cpu_env, addr, cmpv, newv, tcg_constant_i32(oi));
 #else
         gen_helper_exit_atomic(cpu_env);
         /* Produce a result, so that we have a well-formed opcode stream
@@ -3245,20 +3224,15 @@ static void do_atomic_op_i32(TCGv_i32 ret, TCGv addr, TCGv_i32 val,
                              TCGArg idx, MemOp memop, void * const table[])
 {
     gen_atomic_op_i32 gen;
+    TCGMemOpIdx oi;
 
     memop = tcg_canonicalize_memop(memop, 0, 0);
 
     gen = table[memop & (MO_SIZE | MO_BSWAP)];
     tcg_debug_assert(gen != NULL);
 
-#ifdef CONFIG_SOFTMMU
-    {
-        TCGMemOpIdx oi = make_memop_idx(memop & ~MO_SIGN, idx);
-        gen(ret, cpu_env, addr, val, tcg_constant_i32(oi));
-    }
-#else
-    gen(ret, cpu_env, addr, val);
-#endif
+    oi = make_memop_idx(memop & ~MO_SIGN, idx);
+    gen(ret, cpu_env, addr, val, tcg_constant_i32(oi));
 
     if (memop & MO_SIGN) {
         tcg_gen_ext_i32(ret, ret, memop);
@@ -3292,18 +3266,13 @@ static void do_atomic_op_i64(TCGv_i64 ret, TCGv addr, TCGv_i64 val,
     if ((memop & MO_SIZE) == MO_64) {
 #ifdef CONFIG_ATOMIC64
         gen_atomic_op_i64 gen;
+        TCGMemOpIdx oi;
 
         gen = table[memop & (MO_SIZE | MO_BSWAP)];
         tcg_debug_assert(gen != NULL);
 
-#ifdef CONFIG_SOFTMMU
-        {
-            TCGMemOpIdx oi = make_memop_idx(memop & ~MO_SIGN, idx);
-            gen(ret, cpu_env, addr, val, tcg_constant_i32(oi));
-        }
-#else
-        gen(ret, cpu_env, addr, val);
-#endif
+        oi = make_memop_idx(memop & ~MO_SIGN, idx);
+        gen(ret, cpu_env, addr, val, tcg_constant_i32(oi));
 #else
         gen_helper_exit_atomic(cpu_env);
         /* Produce a result, so that we have a well-formed opcode stream
diff --git a/tests/acceptance/virtio-gpu.py b/tests/acceptance/virtio-gpu.py
index 589332c1b7..4acc1e6d5f 100644
--- a/tests/acceptance/virtio-gpu.py
+++ b/tests/acceptance/virtio-gpu.py
@@ -17,10 +17,6 @@ import socket
 import subprocess
 
 
-ACCEL_NOT_AVAILABLE_FMT = "%s accelerator does not seem to be available"
-KVM_NOT_AVAILABLE = ACCEL_NOT_AVAILABLE_FMT % "KVM"
-
-
 def pick_default_vug_bin():
     relative_path = "./contrib/vhost-user-gpu/vhost-user-gpu"
     if is_readable_executable_file(relative_path):
@@ -34,19 +30,23 @@ def pick_default_vug_bin():
 class VirtioGPUx86(Test):
     """
     :avocado: tags=virtio-gpu
+    :avocado: tags=arch:x86_64
+    :avocado: tags=cpu:host
     """
 
-    KERNEL_COMMON_COMMAND_LINE = "printk.time=0 "
+    KERNEL_COMMAND_LINE = "printk.time=0 console=ttyS0 rdinit=/bin/bash"
     KERNEL_URL = (
         "https://archives.fedoraproject.org/pub/fedora"
         "/linux/releases/33/Everything/x86_64/os/images"
         "/pxeboot/vmlinuz"
     )
+    KERNEL_HASH = '1433cfe3f2ffaa44de4ecfb57ec25dc2399cdecf'
     INITRD_URL = (
         "https://archives.fedoraproject.org/pub/fedora"
         "/linux/releases/33/Everything/x86_64/os/images"
         "/pxeboot/initrd.img"
     )
+    INITRD_HASH = 'c828d68a027b53e5220536585efe03412332c2d9'
 
     def wait_for_console_pattern(self, success_message, vm=None):
         wait_for_console_pattern(
@@ -58,24 +58,18 @@ class VirtioGPUx86(Test):
 
     def test_virtio_vga_virgl(self):
         """
-        :avocado: tags=arch:x86_64
-        :avocado: tags=device:virtio-vga
-        :avocado: tags=cpu:host
+        :avocado: tags=device:virtio-vga-gl
         """
-        kernel_command_line = (
-            self.KERNEL_COMMON_COMMAND_LINE + "console=ttyS0 rdinit=/bin/bash"
-        )
         # FIXME: should check presence of virtio, virgl etc
-        if not kvm_available(self.arch, self.qemu_bin):
-            self.cancel(KVM_NOT_AVAILABLE)
+        self.require_accelerator('kvm')
 
-        kernel_path = self.fetch_asset(self.KERNEL_URL)
-        initrd_path = self.fetch_asset(self.INITRD_URL)
+        kernel_path = self.fetch_asset(self.KERNEL_URL, self.KERNEL_HASH)
+        initrd_path = self.fetch_asset(self.INITRD_URL, self.INITRD_HASH)
 
         self.vm.set_console()
         self.vm.add_args("-m", "2G")
         self.vm.add_args("-machine", "pc,accel=kvm")
-        self.vm.add_args("-device", "virtio-vga,virgl=on")
+        self.vm.add_args("-device", "virtio-vga-gl")
         self.vm.add_args("-display", "egl-headless")
         self.vm.add_args(
             "-kernel",
@@ -83,7 +77,7 @@ class VirtioGPUx86(Test):
             "-initrd",
             initrd_path,
             "-append",
-            kernel_command_line,
+            self.KERNEL_COMMAND_LINE,
         )
         try:
             self.vm.launch()
@@ -99,23 +93,17 @@ class VirtioGPUx86(Test):
 
     def test_vhost_user_vga_virgl(self):
         """
-        :avocado: tags=arch:x86_64
         :avocado: tags=device:vhost-user-vga
-        :avocado: tags=cpu:host
         """
-        kernel_command_line = (
-            self.KERNEL_COMMON_COMMAND_LINE + "console=ttyS0 rdinit=/bin/bash"
-        )
         # FIXME: should check presence of vhost-user-gpu, virgl, memfd etc
-        if not kvm_available(self.arch, self.qemu_bin):
-            self.cancel(KVM_NOT_AVAILABLE)
+        self.require_accelerator('kvm')
 
         vug = pick_default_vug_bin()
         if not vug:
             self.cancel("Could not find vhost-user-gpu")
 
-        kernel_path = self.fetch_asset(self.KERNEL_URL)
-        initrd_path = self.fetch_asset(self.INITRD_URL)
+        kernel_path = self.fetch_asset(self.KERNEL_URL, self.KERNEL_HASH)
+        initrd_path = self.fetch_asset(self.INITRD_URL, self.INITRD_HASH)
 
         # Create socketpair to connect proxy and remote processes
         qemu_sock, vug_sock = socket.socketpair(
@@ -153,7 +141,7 @@ class VirtioGPUx86(Test):
             "-initrd",
             initrd_path,
             "-append",
-            kernel_command_line,
+            self.KERNEL_COMMAND_LINE,
         )
         self.vm.launch()
         self.wait_for_console_pattern("as init process")
diff --git a/tests/docker/dockerfiles/debian-tricore-cross.docker b/tests/docker/dockerfiles/debian-tricore-cross.docker
index 985925134c..d8df2c6117 100644
--- a/tests/docker/dockerfiles/debian-tricore-cross.docker
+++ b/tests/docker/dockerfiles/debian-tricore-cross.docker
@@ -1,23 +1,47 @@
 #
 # Docker TriCore cross-compiler target
 #
-# This docker target builds on the debian Stretch base image.
+# This docker target builds on the Debian Buster base image but
+# doesn't inherit from the common one to avoid bringing in unneeded
+# dependencies.
 #
 # Copyright (c) 2018 Philippe Mathieu-Daudé
 #
 # SPDX-License-Identifier: GPL-2.0-or-later
 #
-FROM qemu/debian10
+FROM docker.io/library/debian:buster-slim
 
 MAINTAINER Philippe Mathieu-Daudé <f4bug@amsat.org>
 
+RUN apt update && \
+    DEBIAN_FRONTEND=noninteractive apt install -yy eatmydata && \
+    DEBIAN_FRONTEND=noninteractive eatmydata apt install -yy \
+       bzip2 \
+       ca-certificates \
+       ccache \
+       g++ \
+       gcc \
+       git \
+       libglib2.0-dev \
+       libpixman-1-dev \
+       libtest-harness-perl \
+       locales \
+       make \
+       ninja-build \
+       perl-base \
+       pkgconf \
+       python3-pip \
+       python3-setuptools \
+       python3-wheel
+
 RUN git clone --single-branch \
         https://github.com/bkoppelmann/tricore-binutils.git \
         /usr/src/binutils && \
     cd /usr/src/binutils && chmod +x missing && \
-    CFLAGS=-w ./configure --prefix=/usr --disable-nls --target=tricore && \
+    CFLAGS=-w ./configure --prefix=/usr/local --disable-nls --target=tricore && \
     make && make install && \
     rm -rf /usr/src/binutils
 
-# This image isn't designed for building QEMU but building tests
-ENV QEMU_CONFIGURE_OPTS --disable-system --disable-user
+# This image can only build a very minimal QEMU as well as the tests
+ENV DEF_TARGET_LIST tricore-softmmu
+ENV QEMU_CONFIGURE_OPTS --disable-user --disable-tools --disable-fdt
diff --git a/tests/qemu-iotests/151 b/tests/qemu-iotests/151
index 182f6b5321..93d14193d0 100755
--- a/tests/qemu-iotests/151
+++ b/tests/qemu-iotests/151
@@ -38,8 +38,9 @@ class TestActiveMirror(iotests.QMPTestCase):
                       'if': 'none',
                       'node-name': 'source-node',
                       'driver': iotests.imgfmt,
-                      'file': {'driver': 'file',
-                               'filename': source_img}}
+                      'file': {'driver': 'blkdebug',
+                               'image': {'driver': 'file',
+                                         'filename': source_img}}}
 
         blk_target = {'node-name': 'target-node',
                       'driver': iotests.imgfmt,
@@ -141,6 +142,55 @@ class TestActiveMirror(iotests.QMPTestCase):
 
         self.potential_writes_in_flight = False
 
+    def testIntersectingActiveIO(self):
+        # Fill the source image
+        result = self.vm.hmp_qemu_io('source', 'write -P 1 0 2M')
+
+        # Start the block job (very slowly)
+        result = self.vm.qmp('blockdev-mirror',
+                             job_id='mirror',
+                             filter_node_name='mirror-node',
+                             device='source-node',
+                             target='target-node',
+                             sync='full',
+                             copy_mode='write-blocking',
+                             speed=1)
+
+        self.vm.hmp_qemu_io('source', 'break write_aio A')
+        self.vm.hmp_qemu_io('source', 'aio_write 0 1M')  # 1
+        self.vm.hmp_qemu_io('source', 'wait_break A')
+        self.vm.hmp_qemu_io('source', 'aio_write 0 2M')  # 2
+        self.vm.hmp_qemu_io('source', 'aio_write 0 2M')  # 3
+
+        # Now 2 and 3 are in mirror_wait_on_conflicts, waiting for 1
+
+        self.vm.hmp_qemu_io('source', 'break write_aio B')
+        self.vm.hmp_qemu_io('source', 'aio_write 1M 2M')  # 4
+        self.vm.hmp_qemu_io('source', 'wait_break B')
+
+        # 4 doesn't wait for 2 and 3, because they didn't yet set
+        # in_flight_bitmap. So, nothing prevents 4 to go except for our
+        # break-point B.
+
+        self.vm.hmp_qemu_io('source', 'resume A')
+
+        # Now we resumed 1, so 2 and 3 goes to the next iteration of while loop
+        # in mirror_wait_on_conflicts(). They don't exit, as bitmap is dirty
+        # due to request 4.
+        # In the past at that point 2 and 3 would wait for each other producing
+        # a dead-lock. Now this is fixed and they will wait for request 4.
+
+        self.vm.hmp_qemu_io('source', 'resume B')
+
+        # After resuming 4, one of 2 and 3 goes first and set in_flight_bitmap,
+        # so the other will wait for it.
+
+        result = self.vm.qmp('block-job-set-speed', device='mirror', speed=0)
+        self.assert_qmp(result, 'return', {})
+        self.complete_and_wait(drive='mirror')
+
+        self.potential_writes_in_flight = False
+
 
 if __name__ == '__main__':
     iotests.main(supported_fmts=['qcow2', 'raw'],
diff --git a/tests/qemu-iotests/151.out b/tests/qemu-iotests/151.out
index 8d7e996700..89968f35d7 100644
--- a/tests/qemu-iotests/151.out
+++ b/tests/qemu-iotests/151.out
@@ -1,5 +1,5 @@
-...
+....
 ----------------------------------------------------------------------
-Ran 3 tests
+Ran 4 tests
 
 OK
diff --git a/tests/qemu-iotests/307 b/tests/qemu-iotests/307
index c7685347bc..b429b5aa50 100755
--- a/tests/qemu-iotests/307
+++ b/tests/qemu-iotests/307
@@ -41,9 +41,11 @@ with iotests.FilePath('image') as img, \
     iotests.log('=== Launch VM ===')
 
     vm.add_object('iothread,id=iothread0')
+    vm.add_object('iothread,id=iothread1')
     vm.add_blockdev(f'file,filename={img},node-name=file')
     vm.add_blockdev(f'{iotests.imgfmt},file=file,node-name=fmt')
     vm.add_blockdev('raw,file=file,node-name=ro,read-only=on')
+    vm.add_blockdev('null-co,node-name=null')
     vm.add_device(f'id=scsi0,driver=virtio-scsi,iothread=iothread0')
     vm.launch()
 
@@ -74,6 +76,19 @@ with iotests.FilePath('image') as img, \
     vm.qmp_log('query-block-exports')
     iotests.qemu_nbd_list_log('-k', socket)
 
+    iotests.log('\n=== Add export with conflicting iothread ===')
+
+    vm.qmp_log('device_add', id='sdb', driver='scsi-hd', drive='null')
+
+    # Should fail because of fixed-iothread
+    vm.qmp_log('block-export-add', id='export1', type='nbd', node_name='null',
+               iothread='iothread1', fixed_iothread=True, writable=True)
+
+    # Should ignore the iothread conflict, but then fail because of the
+    # permission conflict (and not crash)
+    vm.qmp_log('block-export-add', id='export1', type='nbd', node_name='null',
+               iothread='iothread1', fixed_iothread=False, writable=True)
+
     iotests.log('\n=== Add a writable export ===')
 
     # This fails because share-rw=off
diff --git a/tests/qemu-iotests/307.out b/tests/qemu-iotests/307.out
index 4b0c7e155a..ec8d2be0e0 100644
--- a/tests/qemu-iotests/307.out
+++ b/tests/qemu-iotests/307.out
@@ -51,6 +51,14 @@ exports available: 1
    base:allocation
 
 
+=== Add export with conflicting iothread ===
+{"execute": "device_add", "arguments": {"drive": "null", "driver": "scsi-hd", "id": "sdb"}}
+{"return": {}}
+{"execute": "block-export-add", "arguments": {"fixed-iothread": true, "id": "export1", "iothread": "iothread1", "node-name": "null", "type": "nbd", "writable": true}}
+{"error": {"class": "GenericError", "desc": "Cannot change iothread of active block backend"}}
+{"execute": "block-export-add", "arguments": {"fixed-iothread": false, "id": "export1", "iothread": "iothread1", "node-name": "null", "type": "nbd", "writable": true}}
+{"error": {"class": "GenericError", "desc": "Permission conflict on node 'null': permissions 'write' are both required by an unnamed block device (uses node 'null' as 'root' child) and unshared by block device 'sdb' (uses node 'null' as 'root' child)."}}
+
 === Add a writable export ===
 {"execute": "block-export-add", "arguments": {"description": "This is the writable second export", "id": "export1", "name": "export1", "node-name": "fmt", "type": "nbd", "writable": true, "writethrough": true}}
 {"error": {"class": "GenericError", "desc": "Permission conflict on node 'fmt': permissions 'write' are both required by an unnamed block device (uses node 'fmt' as 'root' child) and unshared by block device 'sda' (uses node 'fmt' as 'root' child)."}}
diff --git a/tests/qemu-iotests/291 b/tests/qemu-iotests/tests/qemu-img-bitmaps
index 20efb080a6..7a3fe8c3d3 100755
--- a/tests/qemu-iotests/291
+++ b/tests/qemu-iotests/tests/qemu-img-bitmaps
@@ -3,7 +3,7 @@
 #
 # Test qemu-img bitmap handling
 #
-# Copyright (C) 2018-2020 Red Hat, Inc.
+# Copyright (C) 2018-2021 Red Hat, Inc.
 #
 # 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
@@ -27,11 +27,13 @@ status=1 # failure is the default!
 _cleanup()
 {
     _cleanup_test_img
+    _rm_test_img "$TEST_IMG.copy"
     nbd_server_stop
 }
 trap "_cleanup; exit \$status" 0 1 2 3 15
 
 # get standard environment, filters and checks
+cd ..
 . ./common.rc
 . ./common.filter
 . ./common.nbd
@@ -129,6 +131,36 @@ $QEMU_IMG map --output=json --image-opts \
 
 nbd_server_stop
 
+echo
+echo "=== Check handling of inconsistent bitmap ==="
+echo
+
+# Prepare image with corrupted bitmap
+$QEMU_IO -c abort "$TEST_IMG" 2>/dev/null
+$QEMU_IMG bitmap --add "$TEST_IMG" b4
+$QEMU_IMG bitmap --remove "$TEST_IMG" b1
+_img_info --format-specific | _filter_irrelevant_img_info
+# Proof that we fail fast if bitmaps can't be copied
+echo
+$QEMU_IMG convert --bitmaps -O qcow2 "$TEST_IMG" "$TEST_IMG.copy" &&
+    echo "unexpected success"
+TEST_IMG="$TEST_IMG.copy" _img_info --format-specific \
+    | _filter_irrelevant_img_info
+# Skipping the broken bitmaps works,...
+echo
+$QEMU_IMG convert --bitmaps --skip-broken-bitmaps \
+    -O qcow2 "$TEST_IMG" "$TEST_IMG.copy"
+TEST_IMG="$TEST_IMG.copy" _img_info --format-specific \
+    | _filter_irrelevant_img_info
+# ...as does removing them
+echo
+_rm_test_img "$TEST_IMG.copy"
+$QEMU_IMG bitmap --remove "$TEST_IMG" b0
+$QEMU_IMG bitmap --remove --add "$TEST_IMG" b2
+$QEMU_IMG convert --bitmaps -O qcow2 "$TEST_IMG" "$TEST_IMG.copy"
+TEST_IMG="$TEST_IMG.copy" _img_info --format-specific \
+    | _filter_irrelevant_img_info
+
 # success, all done
 echo '*** done'
 rm -f $seq.full
diff --git a/tests/qemu-iotests/291.out b/tests/qemu-iotests/tests/qemu-img-bitmaps.out
index 018d6b103f..e851f0320e 100644
--- a/tests/qemu-iotests/291.out
+++ b/tests/qemu-iotests/tests/qemu-img-bitmaps.out
@@ -1,4 +1,4 @@
-QA output created by 291
+QA output created by qemu-img-bitmaps
 
 === Initial image setup ===
 
@@ -115,4 +115,69 @@ Format specific information:
 [{ "start": 0, "length": 2097152, "depth": 0, "present": true, "zero": false, "data": true, "offset": OFFSET},
 { "start": 2097152, "length": 1048576, "depth": 0, "present": false, "zero": false, "data": false},
 { "start": 3145728, "length": 7340032, "depth": 0, "present": true, "zero": false, "data": true, "offset": OFFSET}]
+
+=== Check handling of inconsistent bitmap ===
+
+image: TEST_DIR/t.IMGFMT
+file format: IMGFMT
+virtual size: 10 MiB (10485760 bytes)
+cluster_size: 65536
+backing file: TEST_DIR/t.IMGFMT.base
+backing file format: IMGFMT
+Format specific information:
+    bitmaps:
+        [0]:
+            flags:
+                [0]: in-use
+                [1]: auto
+            name: b2
+            granularity: 65536
+        [1]:
+            flags:
+                [0]: in-use
+            name: b0
+            granularity: 65536
+        [2]:
+            flags:
+                [0]: auto
+            name: b4
+            granularity: 65536
+    corrupt: false
+
+qemu-img: Cannot copy inconsistent bitmap 'b0'
+Try --skip-broken-bitmaps, or use 'qemu-img bitmap --remove' to delete it
+qemu-img: Could not open 'TEST_DIR/t.IMGFMT.copy': Could not open 'TEST_DIR/t.IMGFMT.copy': No such file or directory
+
+qemu-img: warning: Skipping inconsistent bitmap 'b0'
+qemu-img: warning: Skipping inconsistent bitmap 'b2'
+image: TEST_DIR/t.IMGFMT.copy
+file format: IMGFMT
+virtual size: 10 MiB (10485760 bytes)
+cluster_size: 65536
+Format specific information:
+    bitmaps:
+        [0]:
+            flags:
+                [0]: auto
+            name: b4
+            granularity: 65536
+    corrupt: false
+
+image: TEST_DIR/t.IMGFMT.copy
+file format: IMGFMT
+virtual size: 10 MiB (10485760 bytes)
+cluster_size: 65536
+Format specific information:
+    bitmaps:
+        [0]:
+            flags:
+                [0]: auto
+            name: b4
+            granularity: 65536
+        [1]:
+            flags:
+                [0]: auto
+            name: b2
+            granularity: 65536
+    corrupt: false
 *** done
diff --git a/tests/tcg/configure.sh b/tests/tcg/configure.sh
index aa7c24328a..1f985ccfc0 100755
--- a/tests/tcg/configure.sh
+++ b/tests/tcg/configure.sh
@@ -72,6 +72,10 @@ fi
 : ${cross_cc_x86_64="x86_64-linux-gnu-gcc"}
 : ${cross_cc_cflags_x86_64="-m64"}
 
+# tricore is special as it doesn't have a compiler
+: ${cross_as_tricore="tricore-as"}
+: ${cross_ld_tricore="tricore-ld"}
+
 for target in $target_list; do
   arch=${target%%-*}
 
@@ -247,6 +251,20 @@ for target in $target_list; do
               fi
           fi
       fi
+
+      # Special handling for assembler only tests
+      eval "target_as=\"\${cross_as_$arch}\""
+      eval "target_ld=\"\${cross_ld_$arch}\""
+      if has $target_as && has $target_ld; then
+          case $target in
+              tricore-softmmu)
+                  echo "CROSS_CC_GUEST=$target_as" >> $config_target_mak
+                  echo "CROSS_AS_GUEST=$target_as" >> $config_target_mak
+                  echo "CROSS_LD_GUEST=$target_ld" >> $config_target_mak
+                  got_cross_cc=yes
+                  ;;
+          esac
+      fi
   fi
 
   if test $got_cross_cc = yes; then
diff --git a/tests/unit/meson.build b/tests/unit/meson.build
index 3e0504dd21..5736d285b2 100644
--- a/tests/unit/meson.build
+++ b/tests/unit/meson.build
@@ -14,6 +14,7 @@ tests = {
   'test-qobject-output-visitor': [testqapi],
   'test-clone-visitor': [testqapi],
   'test-qobject-input-visitor': [testqapi],
+  'test-forward-visitor': [testqapi],
   'test-string-input-visitor': [testqapi],
   'test-string-output-visitor': [testqapi],
   'test-opts-visitor': [testqapi],
diff --git a/tests/unit/test-forward-visitor.c b/tests/unit/test-forward-visitor.c
new file mode 100644
index 0000000000..348f7e4e81
--- /dev/null
+++ b/tests/unit/test-forward-visitor.c
@@ -0,0 +1,197 @@
+/*
+ * QAPI Forwarding Visitor unit-tests.
+ *
+ * Copyright (C) 2021 Red Hat Inc.
+ *
+ * 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 "qapi/forward-visitor.h"
+#include "qapi/qobject-input-visitor.h"
+#include "qapi/error.h"
+#include "qapi/qmp/qobject.h"
+#include "qapi/qmp/qdict.h"
+#include "test-qapi-visit.h"
+#include "qemu/option.h"
+
+typedef bool GenericVisitor (Visitor *, const char *, void **, Error **);
+#define CAST_VISIT_TYPE(fn) ((GenericVisitor *)(fn))
+
+/*
+ * Parse @srcstr and wrap it with a ForwardFieldVisitor converting "src" to
+ * "dst". Check that visiting the result with "src" name fails, and return
+ * the result of visiting "dst".
+ */
+static void *visit_with_forward(const char *srcstr, GenericVisitor *fn)
+{
+    bool help = false;
+    QDict *src = keyval_parse(srcstr, NULL, &help, &error_abort);
+    Visitor *v, *alias_v;
+    Error *err = NULL;
+    void *result = NULL;
+
+    v = qobject_input_visitor_new_keyval(QOBJECT(src));
+    visit_start_struct(v, NULL, NULL, 0, &error_abort);
+
+    alias_v = visitor_forward_field(v, "dst", "src");
+    fn(alias_v, "src", &result, &err);
+    error_free_or_abort(&err);
+    assert(!result);
+    fn(alias_v, "dst", &result, &err);
+    assert(err == NULL);
+    visit_free(alias_v);
+
+    visit_end_struct(v, NULL);
+    visit_free(v);
+    qobject_unref(QOBJECT(src));
+    return result;
+}
+
+static void test_forward_any(void)
+{
+    QObject *src = visit_with_forward("src.integer=42,src.string=Hello,src.enum1=value2",
+                                      CAST_VISIT_TYPE(visit_type_any));
+    Visitor *v = qobject_input_visitor_new_keyval(src);
+    Error *err = NULL;
+    UserDefOne *dst;
+
+    visit_type_UserDefOne(v, NULL, &dst, &err);
+    assert(err == NULL);
+    visit_free(v);
+
+    g_assert_cmpint(dst->integer, ==, 42);
+    g_assert_cmpstr(dst->string, ==, "Hello");
+    g_assert_cmpint(dst->has_enum1, ==, true);
+    g_assert_cmpint(dst->enum1, ==, ENUM_ONE_VALUE2);
+    qapi_free_UserDefOne(dst);
+    qobject_unref(QOBJECT(src));
+}
+
+static void test_forward_size(void)
+{
+    /*
+     * visit_type_size does not return a pointer, so visit_with_forward
+     * cannot be used.
+     */
+    bool help = false;
+    QDict *src = keyval_parse("src=1.5M", NULL, &help, &error_abort);
+    Visitor *v, *alias_v;
+    Error *err = NULL;
+    uint64_t result = 0;
+
+    v = qobject_input_visitor_new_keyval(QOBJECT(src));
+    visit_start_struct(v, NULL, NULL, 0, &error_abort);
+
+    alias_v = visitor_forward_field(v, "dst", "src");
+    visit_type_size(alias_v, "src", &result, &err);
+    error_free_or_abort(&err);
+    visit_type_size(alias_v, "dst", &result, &err);
+    assert(result == 3 << 19);
+    assert(err == NULL);
+    visit_free(alias_v);
+
+    visit_end_struct(v, NULL);
+    visit_free(v);
+    qobject_unref(QOBJECT(src));
+}
+
+static void test_forward_number(void)
+{
+    /*
+     * visit_type_number does not return a pointer, so visit_with_forward
+     * cannot be used.
+     */
+    bool help = false;
+    QDict *src = keyval_parse("src=1.5", NULL, &help, &error_abort);
+    Visitor *v, *alias_v;
+    Error *err = NULL;
+    double result = 0.0;
+
+    v = qobject_input_visitor_new_keyval(QOBJECT(src));
+    visit_start_struct(v, NULL, NULL, 0, &error_abort);
+
+    alias_v = visitor_forward_field(v, "dst", "src");
+    visit_type_number(alias_v, "src", &result, &err);
+    error_free_or_abort(&err);
+    visit_type_number(alias_v, "dst", &result, &err);
+    assert(result == 1.5);
+    assert(err == NULL);
+    visit_free(alias_v);
+
+    visit_end_struct(v, NULL);
+    visit_free(v);
+    qobject_unref(QOBJECT(src));
+}
+
+static void test_forward_string(void)
+{
+    char *dst = visit_with_forward("src=Hello",
+                                   CAST_VISIT_TYPE(visit_type_str));
+
+    g_assert_cmpstr(dst, ==, "Hello");
+    g_free(dst);
+}
+
+static void test_forward_struct(void)
+{
+    UserDefOne *dst = visit_with_forward("src.integer=42,src.string=Hello",
+                                         CAST_VISIT_TYPE(visit_type_UserDefOne));
+
+    g_assert_cmpint(dst->integer, ==, 42);
+    g_assert_cmpstr(dst->string, ==, "Hello");
+    g_assert_cmpint(dst->has_enum1, ==, false);
+    qapi_free_UserDefOne(dst);
+}
+
+static void test_forward_alternate(void)
+{
+    AltStrObj *s_dst = visit_with_forward("src=hello",
+                                          CAST_VISIT_TYPE(visit_type_AltStrObj));
+    AltStrObj *o_dst = visit_with_forward("src.integer=42,src.boolean=true,src.string=world",
+                                          CAST_VISIT_TYPE(visit_type_AltStrObj));
+
+    g_assert_cmpint(s_dst->type, ==, QTYPE_QSTRING);
+    g_assert_cmpstr(s_dst->u.s, ==, "hello");
+    g_assert_cmpint(o_dst->type, ==, QTYPE_QDICT);
+    g_assert_cmpint(o_dst->u.o.integer, ==, 42);
+    g_assert_cmpint(o_dst->u.o.boolean, ==, true);
+    g_assert_cmpstr(o_dst->u.o.string, ==, "world");
+
+    qapi_free_AltStrObj(s_dst);
+    qapi_free_AltStrObj(o_dst);
+}
+
+static void test_forward_list(void)
+{
+    uint8List *dst = visit_with_forward("src.0=1,src.1=2,src.2=3,src.3=4",
+                                        CAST_VISIT_TYPE(visit_type_uint8List));
+    uint8List *tmp;
+    int i;
+
+    for (tmp = dst, i = 1; i <= 4; i++) {
+        g_assert(tmp);
+        g_assert_cmpint(tmp->value, ==, i);
+        tmp = tmp->next;
+    }
+    g_assert(!tmp);
+    qapi_free_uint8List(dst);
+}
+
+int main(int argc, char **argv)
+{
+    g_test_init(&argc, &argv, NULL);
+
+    g_test_add_func("/visitor/forward/struct", test_forward_struct);
+    g_test_add_func("/visitor/forward/alternate", test_forward_alternate);
+    g_test_add_func("/visitor/forward/string", test_forward_string);
+    g_test_add_func("/visitor/forward/size", test_forward_size);
+    g_test_add_func("/visitor/forward/number", test_forward_number);
+    g_test_add_func("/visitor/forward/any", test_forward_any);
+    g_test_add_func("/visitor/forward/list", test_forward_list);
+
+    return g_test_run();
+}
diff --git a/trace/mem-internal.h b/trace/mem-internal.h
deleted file mode 100644
index 8b72b678fa..0000000000
--- a/trace/mem-internal.h
+++ /dev/null
@@ -1,50 +0,0 @@
-/*
- * Helper functions for guest memory tracing
- *
- * Copyright (C) 2016 Lluís Vilanova <vilanova@ac.upc.edu>
- *
- * This work is licensed under the terms of the GNU GPL, version 2 or later.
- * See the COPYING file in the top-level directory.
- */
-
-#ifndef TRACE__MEM_INTERNAL_H
-#define TRACE__MEM_INTERNAL_H
-
-#define TRACE_MEM_SZ_SHIFT_MASK 0xf /* size shift mask */
-#define TRACE_MEM_SE (1ULL << 4)    /* sign extended (y/n) */
-#define TRACE_MEM_BE (1ULL << 5)    /* big endian (y/n) */
-#define TRACE_MEM_ST (1ULL << 6)    /* store (y/n) */
-#define TRACE_MEM_MMU_SHIFT 8       /* mmu idx */
-
-static inline uint16_t trace_mem_build_info(
-    int size_shift, bool sign_extend, MemOp endianness,
-    bool store, unsigned int mmu_idx)
-{
-    uint16_t res;
-
-    res = size_shift & TRACE_MEM_SZ_SHIFT_MASK;
-    if (sign_extend) {
-        res |= TRACE_MEM_SE;
-    }
-    if (endianness == MO_BE) {
-        res |= TRACE_MEM_BE;
-    }
-    if (store) {
-        res |= TRACE_MEM_ST;
-    }
-#ifdef CONFIG_SOFTMMU
-    res |= mmu_idx << TRACE_MEM_MMU_SHIFT;
-#endif
-    return res;
-}
-
-static inline uint16_t trace_mem_get_info(MemOp op,
-                                          unsigned int mmu_idx,
-                                          bool store)
-{
-    return trace_mem_build_info(op & MO_SIZE, !!(op & MO_SIGN),
-                                op & MO_BSWAP, store,
-                                mmu_idx);
-}
-
-#endif /* TRACE__MEM_INTERNAL_H */
diff --git a/trace/mem.h b/trace/mem.h
index 9644f592b4..2f27e7bdf0 100644
--- a/trace/mem.h
+++ b/trace/mem.h
@@ -12,24 +12,52 @@
 
 #include "tcg/tcg.h"
 
+#define TRACE_MEM_SZ_SHIFT_MASK 0xf /* size shift mask */
+#define TRACE_MEM_SE (1ULL << 4)    /* sign extended (y/n) */
+#define TRACE_MEM_BE (1ULL << 5)    /* big endian (y/n) */
+#define TRACE_MEM_ST (1ULL << 6)    /* store (y/n) */
+#define TRACE_MEM_MMU_SHIFT 8       /* mmu idx */
 
 /**
- * trace_mem_get_info:
+ * trace_mem_build_info:
  *
  * Return a value for the 'info' argument in guest memory access traces.
  */
-static uint16_t trace_mem_get_info(MemOp op, unsigned int mmu_idx, bool store);
+static inline uint16_t trace_mem_build_info(int size_shift, bool sign_extend,
+                                            MemOp endianness, bool store,
+                                            unsigned int mmu_idx)
+{
+    uint16_t res;
+
+    res = size_shift & TRACE_MEM_SZ_SHIFT_MASK;
+    if (sign_extend) {
+        res |= TRACE_MEM_SE;
+    }
+    if (endianness == MO_BE) {
+        res |= TRACE_MEM_BE;
+    }
+    if (store) {
+        res |= TRACE_MEM_ST;
+    }
+#ifdef CONFIG_SOFTMMU
+    res |= mmu_idx << TRACE_MEM_MMU_SHIFT;
+#endif
+    return res;
+}
+
 
 /**
- * trace_mem_build_info:
+ * trace_mem_get_info:
  *
  * Return a value for the 'info' argument in guest memory access traces.
  */
-static uint16_t trace_mem_build_info(int size_shift, bool sign_extend,
-                                     MemOp endianness, bool store,
-                                     unsigned int mmuidx);
-
-
-#include "trace/mem-internal.h"
+static inline uint16_t trace_mem_get_info(MemOp op,
+                                          unsigned int mmu_idx,
+                                          bool store)
+{
+    return trace_mem_build_info(op & MO_SIZE, !!(op & MO_SIGN),
+                                op & MO_BSWAP, store,
+                                mmu_idx);
+}
 
 #endif /* TRACE__MEM_H */
diff --git a/util/aio-posix.c b/util/aio-posix.c
index 30f5354b1e..2b86777e91 100644
--- a/util/aio-posix.c
+++ b/util/aio-posix.c
@@ -716,3 +716,15 @@ void aio_context_set_poll_params(AioContext *ctx, int64_t max_ns,
 
     aio_notify(ctx);
 }
+
+void aio_context_set_aio_params(AioContext *ctx, int64_t max_batch,
+                                Error **errp)
+{
+    /*
+     * No thread synchronization here, it doesn't matter if an incorrect value
+     * is used once.
+     */
+    ctx->aio_max_batch = max_batch;
+
+    aio_notify(ctx);
+}
diff --git a/util/aio-win32.c b/util/aio-win32.c
index 168717b51b..d5b09a1193 100644
--- a/util/aio-win32.c
+++ b/util/aio-win32.c
@@ -440,3 +440,8 @@ void aio_context_set_poll_params(AioContext *ctx, int64_t max_ns,
         error_setg(errp, "AioContext polling is not implemented on Windows");
     }
 }
+
+void aio_context_set_aio_params(AioContext *ctx, int64_t max_batch,
+                                Error **errp)
+{
+}
diff --git a/util/async.c b/util/async.c
index 9a41591319..6f6717a34b 100644
--- a/util/async.c
+++ b/util/async.c
@@ -554,6 +554,8 @@ AioContext *aio_context_new(Error **errp)
     ctx->poll_grow = 0;
     ctx->poll_shrink = 0;
 
+    ctx->aio_max_batch = 0;
+
     return ctx;
 fail:
     g_source_destroy(&ctx->source);
diff --git a/util/qemu-config.c b/util/qemu-config.c
index 84ee6dc4ea..436ab63b16 100644
--- a/util/qemu-config.c
+++ b/util/qemu-config.c
@@ -255,8 +255,6 @@ CommandLineOptionInfoList *qmp_query_command_line_options(bool has_option,
             info->option = g_strdup(vm_config_groups[i]->name);
             if (!strcmp("drive", vm_config_groups[i]->name)) {
                 info->parameters = get_drive_infolist();
-            } else if (!strcmp("machine", vm_config_groups[i]->name)) {
-                info->parameters = query_option_descs(machine_opts.desc);
             } else {
                 info->parameters =
                     query_option_descs(vm_config_groups[i]->desc);
@@ -265,6 +263,13 @@ CommandLineOptionInfoList *qmp_query_command_line_options(bool has_option,
         }
     }
 
+    if (!has_option || !strcmp(option, "machine")) {
+        info = g_malloc0(sizeof(*info));
+        info->option = g_strdup("machine");
+        info->parameters = query_option_descs(machine_opts.desc);
+        QAPI_LIST_PREPEND(conf_list, info);
+    }
+
     if (conf_list == NULL) {
         error_setg(errp, "invalid option name: %s", option);
     }
@@ -414,15 +419,16 @@ static int qemu_config_foreach(FILE *fp, QEMUConfigCB *cb, void *opaque,
     if (ferror(fp)) {
         loc_pop(&loc);
         error_setg_errno(errp, errno, "Cannot read config file");
-        return res;
+        goto out_no_loc;
     }
     res = count;
-out:
     if (qdict) {
         cb(group, qdict, opaque, errp);
-        qobject_unref(qdict);
     }
+out:
     loc_pop(&loc);
+out_no_loc:
+    qobject_unref(qdict);
     return res;
 }
 
diff --git a/util/qsp.c b/util/qsp.c
index bacc5fa2f6..8562b14a87 100644
--- a/util/qsp.c
+++ b/util/qsp.c
@@ -83,8 +83,8 @@ typedef struct QSPCallSite QSPCallSite;
 struct QSPEntry {
     void *thread_ptr;
     const QSPCallSite *callsite;
-    uint64_t n_acqs;
-    uint64_t ns;
+    aligned_uint64_t n_acqs;
+    aligned_uint64_t ns;
     unsigned int n_objs; /* count of coalesced objs; only used for reporting */
 };
 typedef struct QSPEntry QSPEntry;