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

* Speed up AddressSpaceDispatch creation (Alexey)
* Fix kvm.c assert (David)
* Memory fixes and further speedup (me)
* Persistent reservation manager infrastructure (me)
* virtio-serial: add enable_backend callback (Pavel)
* chardev GMainContext fixes (Peter)

# gpg: Signature made Fri 22 Sep 2017 20:07:33 BST
# gpg:                using RSA key 0xBFFBD25F78C7AE83
# gpg: Good signature from "Paolo Bonzini <bonzini@gnu.org>"
# gpg:                 aka "Paolo Bonzini <pbonzini@redhat.com>"
# Primary key fingerprint: 46F5 9FBD 57D6 12E7 BFD4  E2F7 7E15 100C CD36 69B1
#      Subkey fingerprint: F133 3857 4B66 2389 866C  7682 BFFB D25F 78C7 AE83

* remotes/bonzini/tags/for-upstream: (32 commits)
  chardev: remove context in chr_update_read_handler
  chardev: use per-dev context for io_add_watch_poll
  chardev: add Chardev.gcontext field
  chardev: new qemu_chr_be_update_read_handlers()
  scsi: add persistent reservation manager using qemu-pr-helper
  scsi: add multipath support to qemu-pr-helper
  scsi: build qemu-pr-helper
  scsi, file-posix: add support for persistent reservation management
  memory: Share special empty FlatView
  memory: seek FlatView sharing candidates among children subregions
  memory: trace FlatView creation and destruction
  memory: Create FlatView directly
  memory: Get rid of address_space_init_shareable
  memory: Rework "info mtree" to print flat views and dispatch trees
  memory: Do not allocate FlatView in address_space_init
  memory: Share FlatView's and dispatch trees between address spaces
  memory: Move address_space_update_ioeventfds
  memory: Alloc dispatch tree where topology is generared
  memory: Store physical root MR in FlatView
  memory: Rename mem_begin/mem_commit/mem_add helpers
  ...

# Conflicts:
#	configure

Makefile

@@ -372,6 +372,11 @@ qemu-bridge-helper$(EXESUF): qemu-bridge-helper.o $(COMMON_LDADDS)
 fsdev/virtfs-proxy-helper$(EXESUF): fsdev/virtfs-proxy-helper.o fsdev/9p-marshal.o fsdev/9p-iov-marshal.o $(COMMON_LDADDS)
 fsdev/virtfs-proxy-helper$(EXESUF): LIBS += -lcap
 
+scsi/qemu-pr-helper$(EXESUF): scsi/qemu-pr-helper.o scsi/utils.o $(crypto-obj-y) $(io-obj-y) $(qom-obj-y) $(COMMON_LDADDS)
+ifdef CONFIG_MPATH
+scsi/qemu-pr-helper$(EXESUF): LIBS += -ludev -lmultipath -lmpathpersist
+endif
+
 qemu-img-cmds.h: $(SRC_PATH)/qemu-img-cmds.hx $(SRC_PATH)/scripts/hxtool
 	$(call quiet-command,sh $(SRC_PATH)/scripts/hxtool -h < $< > $@,"GEN","$@")
 
@@ -488,7 +493,7 @@ clean:
 	rm -f *.msi
 	find . \( -name '*.so' -o -name '*.dll' -o -name '*.mo' -o -name '*.[oda]' \) -type f -exec rm {} +
 	rm -f $(filter-out %.tlb,$(TOOLS)) $(HELPERS-y) qemu-ga TAGS cscope.* *.pod *~ */*~
-	rm -f fsdev/*.pod
+	rm -f fsdev/*.pod scsi/*.pod
 	rm -f qemu-img-cmds.h
 	rm -f ui/shader/*-vert.h ui/shader/*-frag.h
 	@# May not be present in GENERATED_FILES

Makefile.objs

@@ -171,6 +171,7 @@ trace-events-subdirs += qapi
 trace-events-subdirs += accel/tcg
 trace-events-subdirs += accel/kvm
 trace-events-subdirs += nbd
+trace-events-subdirs += scsi
 
 trace-events-files = $(SRC_PATH)/trace-events $(trace-events-subdirs:%=$(SRC_PATH)/%/trace-events)
 

accel/kvm/kvm-all.c

@@ -722,7 +722,6 @@ static void kvm_set_phys_mem(KVMMemoryListener *kml,
     mem = kvm_lookup_matching_slot(kml, start_addr, size);
     if (!add) {
         if (!mem) {
-            g_assert(!memory_region_is_ram(mr) && !writeable && !mr->romd_mode);
             return;
         }
         if (mem->flags & KVM_MEM_LOG_DIRTY_PAGES) {

block/file-posix.c

@@ -33,6 +33,9 @@
 #include "block/raw-aio.h"
 #include "qapi/qmp/qstring.h"
 
+#include "scsi/pr-manager.h"
+#include "scsi/constants.h"
+
 #if defined(__APPLE__) && (__MACH__)
 #include <paths.h>
 #include <sys/param.h>
@@ -155,6 +158,8 @@ typedef struct BDRVRawState {
     bool page_cache_inconsistent:1;
     bool has_fallocate;
     bool needs_alignment;
+
+    PRManager *pr_mgr;
 } BDRVRawState;
 
 typedef struct BDRVRawReopenState {
@@ -402,6 +407,11 @@ static QemuOptsList raw_runtime_opts = {
             .type = QEMU_OPT_STRING,
             .help = "file locking mode (on/off/auto, default: auto)",
         },
+        {
+            .name = "pr-manager",
+            .type = QEMU_OPT_STRING,
+            .help = "id of persistent reservation manager object (default: none)",
+        },
         { /* end of list */ }
     },
 };
@@ -413,6 +423,7 @@ static int raw_open_common(BlockDriverState *bs, QDict *options,
     QemuOpts *opts;
     Error *local_err = NULL;
     const char *filename = NULL;
+    const char *str;
     BlockdevAioOptions aio, aio_default;
     int fd, ret;
     struct stat st;
@@ -476,6 +487,16 @@ static int raw_open_common(BlockDriverState *bs, QDict *options,
         abort();
     }
 
+    str = qemu_opt_get(opts, "pr-manager");
+    if (str) {
+        s->pr_mgr = pr_manager_lookup(str, &local_err);
+        if (local_err) {
+            error_propagate(errp, local_err);
+            ret = -EINVAL;
+            goto fail;
+        }
+    }
+
     s->open_flags = open_flags;
     raw_parse_flags(bdrv_flags, &s->open_flags);
 
@@ -2597,6 +2618,15 @@ static BlockAIOCB *hdev_aio_ioctl(BlockDriverState *bs,
     if (fd_open(bs) < 0)
         return NULL;
 
+    if (req == SG_IO && s->pr_mgr) {
+        struct sg_io_hdr *io_hdr = buf;
+        if (io_hdr->cmdp[0] == PERSISTENT_RESERVE_OUT ||
+            io_hdr->cmdp[0] == PERSISTENT_RESERVE_IN) {
+            return pr_manager_execute(s->pr_mgr, bdrv_get_aio_context(bs),
+                                      s->fd, io_hdr, cb, opaque);
+        }
+    }
+
     acb = g_new(RawPosixAIOData, 1);
     acb->bs = bs;
     acb->aio_type = QEMU_AIO_IOCTL;

chardev/char-fd.c

@@ -84,8 +84,7 @@ static GSource *fd_chr_add_watch(Chardev *chr, GIOCondition cond)
     return qio_channel_create_watch(s->ioc_out, cond);
 }
 
-static void fd_chr_update_read_handler(Chardev *chr,
-                                       GMainContext *context)
+static void fd_chr_update_read_handler(Chardev *chr)
 {
     FDChardev *s = FD_CHARDEV(chr);
 
@@ -94,7 +93,7 @@ static void fd_chr_update_read_handler(Chardev *chr,
         chr->gsource = io_add_watch_poll(chr, s->ioc_in,
                                            fd_chr_read_poll,
                                            fd_chr_read, chr,
-                                           context);
+                                           chr->gcontext);
     }
 }
 

chardev/char-fe.c

@@ -253,7 +253,6 @@ void qemu_chr_fe_set_handlers(CharBackend *b,
                               bool set_open)
 {
     Chardev *s;
-    ChardevClass *cc;
     int fe_open;
 
     s = b->chr;
@@ -261,7 +260,6 @@ void qemu_chr_fe_set_handlers(CharBackend *b,
         return;
     }
 
-    cc = CHARDEV_GET_CLASS(s);
     if (!opaque && !fd_can_read && !fd_read && !fd_event) {
         fe_open = 0;
         remove_fd_in_watch(s);
@@ -273,9 +271,8 @@ void qemu_chr_fe_set_handlers(CharBackend *b,
     b->chr_event = fd_event;
     b->chr_be_change = be_change;
     b->opaque = opaque;
-    if (cc->chr_update_read_handler) {
-        cc->chr_update_read_handler(s, context);
-    }
+
+    qemu_chr_be_update_read_handlers(s, context);
 
     if (set_open) {
         qemu_chr_fe_set_open(b, fe_open);

chardev/char-pty.c

@@ -112,8 +112,7 @@ static void pty_chr_update_read_handler_locked(Chardev *chr)
     }
 }
 
-static void pty_chr_update_read_handler(Chardev *chr,
-                                        GMainContext *context)
+static void pty_chr_update_read_handler(Chardev *chr)
 {
     qemu_mutex_lock(&chr->chr_write_lock);
     pty_chr_update_read_handler_locked(chr);
@@ -219,7 +218,7 @@ static void pty_chr_state(Chardev *chr, int connected)
             chr->gsource = io_add_watch_poll(chr, s->ioc,
                                                pty_chr_read_poll,
                                                pty_chr_read,
-                                               chr, NULL);
+                                               chr, chr->gcontext);
         }
     }
 }

chardev/char-socket.c

@@ -516,13 +516,12 @@ static void tcp_chr_connect(void *opaque)
         chr->gsource = io_add_watch_poll(chr, s->ioc,
                                            tcp_chr_read_poll,
                                            tcp_chr_read,
-                                           chr, NULL);
+                                           chr, chr->gcontext);
     }
     qemu_chr_be_event(chr, CHR_EVENT_OPENED);
 }
 
-static void tcp_chr_update_read_handler(Chardev *chr,
-                                        GMainContext *context)
+static void tcp_chr_update_read_handler(Chardev *chr)
 {
     SocketChardev *s = SOCKET_CHARDEV(chr);
 
@@ -535,7 +534,7 @@ static void tcp_chr_update_read_handler(Chardev *chr,
         chr->gsource = io_add_watch_poll(chr, s->ioc,
                                            tcp_chr_read_poll,
                                            tcp_chr_read, chr,
-                                           context);
+                                           chr->gcontext);
     }
 }
 

chardev/char-udp.c

@@ -100,8 +100,7 @@ static gboolean udp_chr_read(QIOChannel *chan, GIOCondition cond, void *opaque)
     return TRUE;
 }
 
-static void udp_chr_update_read_handler(Chardev *chr,
-                                        GMainContext *context)
+static void udp_chr_update_read_handler(Chardev *chr)
 {
     UdpChardev *s = UDP_CHARDEV(chr);
 
@@ -110,7 +109,7 @@ static void udp_chr_update_read_handler(Chardev *chr,
         chr->gsource = io_add_watch_poll(chr, s->ioc,
                                            udp_chr_read_poll,
                                            udp_chr_read, chr,
-                                           context);
+                                           chr->gcontext);
     }
 }
 

chardev/char.c

@@ -180,6 +180,17 @@ void qemu_chr_be_write(Chardev *s, uint8_t *buf, int len)
     }
 }
 
+void qemu_chr_be_update_read_handlers(Chardev *s,
+                                      GMainContext *context)
+{
+    ChardevClass *cc = CHARDEV_GET_CLASS(s);
+
+    s->gcontext = context;
+    if (cc->chr_update_read_handler) {
+        cc->chr_update_read_handler(s);
+    }
+}
+
 int qemu_chr_add_client(Chardev *s, int fd)
 {
     return CHARDEV_GET_CLASS(s)->chr_add_client ?

configure

@@ -290,6 +290,7 @@ netmap="no"
 sdl=""
 sdlabi=""
 virtfs=""
+mpath=""
 vnc="yes"
 sparse="no"
 vde=""
@@ -936,6 +937,10 @@ for opt do
   ;;
   --enable-virtfs) virtfs="yes"
   ;;
+  --disable-mpath) mpath="no"
+  ;;
+  --enable-mpath) mpath="yes"
+  ;;
   --disable-vnc) vnc="no"
   ;;
   --enable-vnc) vnc="yes"
@@ -1479,6 +1484,7 @@ disabled with --disable-FEATURE, default is enabled if available:
   vnc-png         PNG compression for VNC server
   cocoa           Cocoa UI (Mac OS X only)
   virtfs          VirtFS
+  mpath           Multipath persistent reservation passthrough
   xen             xen backend driver support
   xen-pci-passthrough
   brlapi          BrlAPI (Braile)
@@ -3294,6 +3300,30 @@ else
       "Please install the pixman devel package."
 fi
 
+##########################################
+# libmpathpersist probe
+
+if test "$mpath" != "no" ; then
+  cat > $TMPC <<EOF
+#include <libudev.h>
+#include <mpath_persist.h>
+unsigned mpath_mx_alloc_len = 1024;
+int logsink;
+int main(void) {
+    struct udev *udev = udev_new();
+    mpath_lib_init(udev);
+    return 0;
+}
+EOF
+  if compile_prog "" "-ludev -lmultipath -lmpathpersist" ; then
+    mpathpersist=yes
+  else
+    mpathpersist=no
+  fi
+else
+  mpathpersist=no
+fi
+
 ##########################################
 # libcap probe
 
@@ -5023,16 +5053,34 @@ if test "$want_tools" = "yes" ; then
   fi
 fi
 if test "$softmmu" = yes ; then
-  if test "$virtfs" != no ; then
-    if test "$cap" = yes && test "$linux" = yes && test "$attr" = yes ; then
+  if test "$linux" = yes; then
+    if test "$virtfs" != no && test "$cap" = yes && test "$attr" = yes ; then
       virtfs=yes
       tools="$tools fsdev/virtfs-proxy-helper\$(EXESUF)"
     else
       if test "$virtfs" = yes; then
-        error_exit "VirtFS is supported only on Linux and requires libcap devel and libattr devel"
+        error_exit "VirtFS requires libcap devel and libattr devel"
       fi
       virtfs=no
     fi
+    if test "$mpath" != no && test "$mpathpersist" = yes ; then
+      mpath=yes
+    else
+      if test "$mpath" = yes; then
+        error_exit "Multipath requires libmpathpersist devel"
+      fi
+      mpath=no
+    fi
+    tools="$tools scsi/qemu-pr-helper\$(EXESUF)"
+  else
+    if test "$virtfs" = yes; then
+      error_exit "VirtFS is supported only on Linux"
+    fi
+    virtfs=no
+    if test "$mpath" = yes; then
+      error_exit "Multipath is supported only on Linux"
+    fi
+    mpath=no
   fi
 fi
 
@@ -5278,6 +5326,7 @@ echo "Audio drivers     $audio_drv_list"
 echo "Block whitelist (rw) $block_drv_rw_whitelist"
 echo "Block whitelist (ro) $block_drv_ro_whitelist"
 echo "VirtFS support    $virtfs"
+echo "Multipath support $mpath"
 echo "VNC support       $vnc"
 if test "$vnc" = "yes" ; then
     echo "VNC SASL support  $vnc_sasl"
@@ -5729,6 +5778,9 @@ fi
 if test "$virtfs" = "yes" ; then
   echo "CONFIG_VIRTFS=y" >> $config_host_mak
 fi
+if test "$mpath" = "yes" ; then
+  echo "CONFIG_MPATH=y" >> $config_host_mak
+fi
 if test "$vhost_scsi" = "yes" ; then
   echo "CONFIG_VHOST_SCSI=y" >> $config_host_mak
 fi
@@ -6510,7 +6562,7 @@ fi
 
 # build tree in object directory in case the source is not in the current directory
 DIRS="tests tests/tcg tests/tcg/cris tests/tcg/lm32 tests/libqos tests/qapi-schema tests/tcg/xtensa tests/qemu-iotests tests/vm"
-DIRS="$DIRS docs docs/interop fsdev"
+DIRS="$DIRS docs docs/interop fsdev scsi"
 DIRS="$DIRS pc-bios/optionrom pc-bios/spapr-rtas pc-bios/s390-ccw"
 DIRS="$DIRS roms/seabios roms/vgabios"
 DIRS="$DIRS qapi-generated"

cpus.c

@@ -1764,8 +1764,9 @@ void qemu_init_vcpu(CPUState *cpu)
         /* If the target cpu hasn't set up any address spaces itself,
          * give it the default one.
          */
-        AddressSpace *as = address_space_init_shareable(cpu->memory,
-                                                        "cpu-memory");
+        AddressSpace *as = g_new0(AddressSpace, 1);
+
+        address_space_init(as, cpu->memory, "cpu-memory");
         cpu->num_ases = 1;
         cpu_address_space_init(cpu, as, 0);
     }

docs/devel/atomics.txt

@@ -63,11 +63,23 @@ operations:
     typeof(*ptr) atomic_fetch_sub(ptr, val)
     typeof(*ptr) atomic_fetch_and(ptr, val)
     typeof(*ptr) atomic_fetch_or(ptr, val)
+    typeof(*ptr) atomic_fetch_xor(ptr, val)
+    typeof(*ptr) atomic_fetch_inc_nonzero(ptr)
     typeof(*ptr) atomic_xchg(ptr, val)
     typeof(*ptr) atomic_cmpxchg(ptr, old, new)
 
 all of which return the old value of *ptr.  These operations are
-polymorphic; they operate on any type that is as wide as an int.
+polymorphic; they operate on any type that is as wide as a pointer.
+
+Similar operations return the new value of *ptr:
+
+    typeof(*ptr) atomic_inc_fetch(ptr)
+    typeof(*ptr) atomic_dec_fetch(ptr)
+    typeof(*ptr) atomic_add_fetch(ptr, val)
+    typeof(*ptr) atomic_sub_fetch(ptr, val)
+    typeof(*ptr) atomic_and_fetch(ptr, val)
+    typeof(*ptr) atomic_or_fetch(ptr, val)
+    typeof(*ptr) atomic_xor_fetch(ptr, val)
 
 Sequentially consistent loads and stores can be done using:
 

docs/interop/pr-helper.rst

+..
+
+======================================
+Persistent reservation helper protocol
+======================================
+
+QEMU's SCSI passthrough devices, ``scsi-block`` and ``scsi-generic``,
+can delegate implementation of persistent reservations to an external
+(and typically privileged) program.  Persistent Reservations allow
+restricting access to block devices to specific initiators in a shared
+storage setup.
+
+For a more detailed reference please refer the the SCSI Primary
+Commands standard, specifically the section on Reservations and the
+"PERSISTENT RESERVE IN" and "PERSISTENT RESERVE OUT" commands.
+
+This document describes the socket protocol used between QEMU's
+``pr-manager-helper`` object and the external program.
+
+.. contents::
+
+Connection and initialization
+-----------------------------
+
+All data transmitted on the socket is big-endian.
+
+After connecting to the helper program's socket, the helper starts a simple
+feature negotiation process by writing four bytes corresponding to
+the features it exposes (``supported_features``).  QEMU reads it,
+then writes four bytes corresponding to the desired features of the
+helper program (``requested_features``).
+
+If a bit is 1 in ``requested_features`` and 0 in ``supported_features``,
+the corresponding feature is not supported by the helper and the connection
+is closed.  On the other hand, it is acceptable for a bit to be 0 in
+``requested_features`` and 1 in ``supported_features``; in this case,
+the helper will not enable the feature.
+
+Right now no feature is defined, so the two parties always write four
+zero bytes.
+
+Command format
+--------------
+
+It is invalid to send multiple commands concurrently on the same
+socket.  It is however possible to connect multiple sockets to the
+helper and send multiple commands to the helper for one or more
+file descriptors.
+
+A command consists of a request and a response.  A request consists
+of a 16-byte SCSI CDB.  A file descriptor must be passed to the helper
+together with the SCSI CDB using ancillary data.
+
+The CDB has the following limitations:
+
+- the command (stored in the first byte) must be one of 0x5E
+  (PERSISTENT RESERVE IN) or 0x5F (PERSISTENT RESERVE OUT).
+
+- the allocation length (stored in bytes 7-8 of the CDB for PERSISTENT
+  RESERVE IN) or parameter list length (stored in bytes 5-8 of the CDB
+  for PERSISTENT RESERVE OUT) is limited to 8 KiB.
+
+For PERSISTENT RESERVE OUT, the parameter list is sent right after the
+CDB.  The length of the parameter list is taken from the CDB itself.
+
+The helper's reply has the following structure:
+
+- 4 bytes for the SCSI status
+
+- 4 bytes for the payload size (nonzero only for PERSISTENT RESERVE IN
+  and only if the SCSI status is 0x00, i.e. GOOD)
+
+- 96 bytes for the SCSI sense data
+
+- if the size is nonzero, the payload follows
+
+The sense data is always sent to keep the protocol simple, even though
+it is only valid if the SCSI status is CHECK CONDITION (0x02).
+
+The payload size is always less than or equal to the allocation length
+specified in the CDB for the PERSISTENT RESERVE IN command.
+
+If the protocol is violated, the helper closes the socket.

docs/pr-manager.rst

+======================================
+Persistent reservation managers
+======================================
+
+SCSI persistent Reservations allow restricting access to block devices
+to specific initiators in a shared storage setup.  When implementing
+clustering of virtual machines, it is a common requirement for virtual
+machines to send persistent reservation SCSI commands.  However,
+the operating system restricts sending these commands to unprivileged
+programs because incorrect usage can disrupt regular operation of the
+storage fabric.
+
+For this reason, QEMU's SCSI passthrough devices, ``scsi-block``
+and ``scsi-generic`` (both are only available on Linux) can delegate
+implementation of persistent reservations to a separate object,
+the "persistent reservation manager".  Only PERSISTENT RESERVE OUT and
+PERSISTENT RESERVE IN commands are passed to the persistent reservation
+manager object; other commands are processed by QEMU as usual.
+
+-----------------------------------------
+Defining a persistent reservation manager
+-----------------------------------------
+
+A persistent reservation manager is an instance of a subclass of the
+"pr-manager" QOM class.
+
+Right now only one subclass is defined, ``pr-manager-helper``, which
+forwards the commands to an external privileged helper program
+over Unix sockets.  The helper program only allows sending persistent
+reservation commands to devices for which QEMU has a file descriptor,
+so that QEMU will not be able to effect persistent reservations
+unless it has access to both the socket and the device.
+
+``pr-manager-helper`` has a single string property, ``path``, which
+accepts the path to the helper program's Unix socket.  For example,
+the following command line defines a ``pr-manager-helper`` object and
+attaches it to a SCSI passthrough device::
+
+      $ qemu-system-x86_64
+          -device virtio-scsi \
+          -object pr-manager-helper,id=helper0,path=/var/run/qemu-pr-helper.sock
+          -drive if=none,id=hd,driver=raw,file.filename=/dev/sdb,file.pr-manager=helper0
+          -device scsi-block,drive=hd
+
+Alternatively, using ``-blockdev``::
+
+      $ qemu-system-x86_64
+          -device virtio-scsi \
+          -object pr-manager-helper,id=helper0,path=/var/run/qemu-pr-helper.sock
+          -blockdev node-name=hd,driver=raw,file.driver=host_device,file.filename=/dev/sdb,file.pr-manager=helper0
+          -device scsi-block,drive=hd
+
+----------------------------------
+Invoking :program:`qemu-pr-helper`
+----------------------------------
+
+QEMU provides an implementation of the persistent reservation helper,
+called :program:`qemu-pr-helper`.  The helper should be started as a
+system service and supports the following option:
+
+-d, --daemon              run in the background
+-q, --quiet               decrease verbosity
+-v, --verbose             increase verbosity
+-f, --pidfile=path        PID file when running as a daemon
+-k, --socket=path         path to the socket
+-T, --trace=trace-opts    tracing options
+
+By default, the socket and PID file are placed in the runtime state
+directory, for example :file:`/var/run/qemu-pr-helper.sock` and
+:file:`/var/run/qemu-pr-helper.pid`.  The PID file is not created
+unless :option:`-d` is passed too.
+
+:program:`qemu-pr-helper` can also use the systemd socket activation
+protocol.  In this case, the systemd socket unit should specify a
+Unix stream socket, like this::
+
+    [Socket]
+    ListenStream=/var/run/qemu-pr-helper.sock
+
+After connecting to the socket, :program:`qemu-pr-helper`` can optionally drop
+root privileges, except for those capabilities that are needed for
+its operation.  To do this, add the following options:
+
+-u, --user=user           user to drop privileges to
+-g, --group=group         group to drop privileges to
+
+---------------------------------------------
+Multipath devices and persistent reservations
+---------------------------------------------
+
+Proper support of persistent reservation for multipath devices requires
+communication with the multipath daemon, so that the reservation is
+registered and applied when a path is newly discovered or becomes online
+again.  :command:`qemu-pr-helper` can do this if the ``libmpathpersist``
+library was available on the system at build time.
+
+As of August 2017, a reservation key must be specified in ``multipath.conf``
+for ``multipathd`` to check for persistent reservation for newly
+discovered paths or reinstated paths.  The attribute can be added
+to the ``defaults`` section or the ``multipaths`` section; for example::
+
+    multipaths {
+        multipath {
+            wwid   XXXXXXXXXXXXXXXX
+            alias      yellow
+            reservation_key  0x123abc
+        }
+    }
+
+Linking :program:`qemu-pr-helper` to ``libmpathpersist`` does not impede
+its usage on regular SCSI devices.

hmp-commands-info.hx

@@ -250,9 +250,10 @@ ETEXI
 
     {
         .name       = "mtree",
-        .args_type  = "flatview:-f",
-        .params     = "[-f]",
-        .help       = "show memory tree (-f: dump flat view for address spaces)",
+        .args_type  = "flatview:-f,dispatch_tree:-d",
+        .params     = "[-f][-d]",
+        .help       = "show memory tree (-f: dump flat view for address spaces;"
+                      "-d: dump dispatch tree, valid with -f only)",
         .cmd        = hmp_info_mtree,
     },
 

hw/arm/armv7m.c

@@ -41,7 +41,7 @@ static MemTxResult bitband_read(void *opaque, hwaddr offset,
 
     /* Find address in underlying memory and round down to multiple of size */
     addr = bitband_addr(s, offset) & (-size);
-    res = address_space_read(s->source_as, addr, attrs, buf, size);
+    res = address_space_read(&s->source_as, addr, attrs, buf, size);
     if (res) {
         return res;
     }
@@ -66,7 +66,7 @@ static MemTxResult bitband_write(void *opaque, hwaddr offset, uint64_t value,
 
     /* Find address in underlying memory and round down to multiple of size */
     addr = bitband_addr(s, offset) & (-size);
-    res = address_space_read(s->source_as, addr, attrs, buf, size);
+    res = address_space_read(&s->source_as, addr, attrs, buf, size);
     if (res) {
         return res;
     }
@@ -79,7 +79,7 @@ static MemTxResult bitband_write(void *opaque, hwaddr offset, uint64_t value,
     } else {
         buf[bitpos >> 3] &= ~bit;
     }
-    return address_space_write(s->source_as, addr, attrs, buf, size);
+    return address_space_write(&s->source_as, addr, attrs, buf, size);
 }
 
 static const MemoryRegionOps bitband_ops = {
@@ -111,8 +111,7 @@ static void bitband_realize(DeviceState *dev, Error **errp)
         return;
     }
 
-    s->source_as = address_space_init_shareable(s->source_memory,
-                                                "bitband-source");
+    address_space_init(&s->source_as, s->source_memory, "bitband-source");
 }
 
 /* Board init.  */

hw/char/virtio-console.c

@@ -187,6 +187,26 @@ static int chr_be_change(void *opaque)
     return 0;
 }
 
+static void virtconsole_enable_backend(VirtIOSerialPort *port, bool enable)
+{
+    VirtConsole *vcon = VIRTIO_CONSOLE(port);
+
+    if (!qemu_chr_fe_backend_connected(&vcon->chr)) {
+        return;
+    }
+
+    if (enable) {
+        VirtIOSerialPortClass *k = VIRTIO_SERIAL_PORT_GET_CLASS(port);
+
+        qemu_chr_fe_set_handlers(&vcon->chr, chr_can_read, chr_read,
+                                 k->is_console ? NULL : chr_event,
+                                 chr_be_change, vcon, NULL, false);
+    } else {
+        qemu_chr_fe_set_handlers(&vcon->chr, NULL, NULL, NULL,
+                                 NULL, NULL, NULL, false);
+    }
+}
+
 static void virtconsole_realize(DeviceState *dev, Error **errp)
 {
     VirtIOSerialPort *port = VIRTIO_SERIAL_PORT(dev);
@@ -258,6 +278,7 @@ static void virtserialport_class_init(ObjectClass *klass, void *data)
     k->unrealize = virtconsole_unrealize;
     k->have_data = flush_buf;
     k->set_guest_connected = set_guest_connected;
+    k->enable_backend = virtconsole_enable_backend;
     k->guest_writable = guest_writable;
     dc->props = virtserialport_properties;
 }

hw/char/virtio-serial-bus.c

@@ -637,6 +637,13 @@ static void set_status(VirtIODevice *vdev, uint8_t status)
     if (!(status & VIRTIO_CONFIG_S_DRIVER_OK)) {
         guest_reset(vser);
     }
+
+    QTAILQ_FOREACH(port, &vser->ports, next) {
+        VirtIOSerialPortClass *vsc = VIRTIO_SERIAL_PORT_GET_CLASS(port);
+        if (vsc->enable_backend) {
+            vsc->enable_backend(port, vdev->vm_running);
+        }
+    }
 }
 
 static void vser_reset(VirtIODevice *vdev)

hw/intc/openpic_kvm.c

@@ -124,7 +124,7 @@ static void kvm_openpic_region_add(MemoryListener *listener,
     uint64_t reg_base;
     int ret;
 
-    if (section->address_space != &address_space_memory) {
+    if (section->fv != address_space_to_flatview(&address_space_memory)) {
         abort();
     }
 

include/chardev/char.h

@@ -55,6 +55,7 @@ struct Chardev {
     int logfd;
     int be_open;
     GSource *gsource;
+    GMainContext *gcontext;
     DECLARE_BITMAP(features, QEMU_CHAR_FEATURE_LAST);
 };
 
@@ -168,6 +169,16 @@ void qemu_chr_be_write(Chardev *s, uint8_t *buf, int len);
  */
 void qemu_chr_be_write_impl(Chardev *s, uint8_t *buf, int len);
 
+/**
+ * @qemu_chr_be_update_read_handlers:
+ *
+ * Invoked when frontend read handlers are setup
+ *
+ * @context the gcontext that will be used to attach the watch sources
+ */
+void qemu_chr_be_update_read_handlers(Chardev *s,
+                                      GMainContext *context);
+
 /**
  * @qemu_chr_be_event:
  *
@@ -227,7 +238,7 @@ typedef struct ChardevClass {
     int (*chr_write)(Chardev *s, const uint8_t *buf, int len);
     int (*chr_sync_read)(Chardev *s, const uint8_t *buf, int len);
     GSource *(*chr_add_watch)(Chardev *s, GIOCondition cond);
-    void (*chr_update_read_handler)(Chardev *s, GMainContext *context);
+    void (*chr_update_read_handler)(Chardev *s);
     int (*chr_ioctl)(Chardev *s, int cmd, void *arg);
     int (*get_msgfds)(Chardev *s, int* fds, int num);
     int (*set_msgfds)(Chardev *s, int *fds, int num);

include/exec/memory-internal.h

@@ -22,14 +22,22 @@
 #ifndef CONFIG_USER_ONLY
 typedef struct AddressSpaceDispatch AddressSpaceDispatch;
 
-void address_space_init_dispatch(AddressSpace *as);
-void address_space_unregister(AddressSpace *as);
-void address_space_destroy_dispatch(AddressSpace *as);
-
 extern const MemoryRegionOps unassigned_mem_ops;
 
 bool memory_region_access_valid(MemoryRegion *mr, hwaddr addr,
                                 unsigned size, bool is_write);
 
+void flatview_add_to_dispatch(FlatView *fv, MemoryRegionSection *section);
+AddressSpaceDispatch *address_space_dispatch_new(FlatView *fv);
+void address_space_dispatch_compact(AddressSpaceDispatch *d);
+
+AddressSpaceDispatch *address_space_to_dispatch(AddressSpace *as);
+AddressSpaceDispatch *flatview_to_dispatch(FlatView *fv);
+void address_space_dispatch_free(AddressSpaceDispatch *d);
+
+void mtree_print_dispatch(fprintf_function mon, void *f,
+                          struct AddressSpaceDispatch *d,
+                          MemoryRegion *root);
+
 #endif
 #endif

include/exec/memory.h

@@ -308,21 +308,18 @@ struct AddressSpace {
     struct rcu_head rcu;
     char *name;
     MemoryRegion *root;
-    int ref_count;
-    bool malloced;
 
     /* Accessed via RCU.  */
     struct FlatView *current_map;
 
     int ioeventfd_nb;
     struct MemoryRegionIoeventfd *ioeventfds;
-    struct AddressSpaceDispatch *dispatch;
-    struct AddressSpaceDispatch *next_dispatch;
-    MemoryListener dispatch_listener;
     QTAILQ_HEAD(memory_listeners_as, MemoryListener) listeners;
     QTAILQ_ENTRY(AddressSpace) address_spaces_link;
 };
 
+FlatView *address_space_to_flatview(AddressSpace *as);
+
 /**
  * MemoryRegionSection: describes a fragment of a #MemoryRegion
  *
@@ -336,7 +333,7 @@ struct AddressSpace {
  */
 struct MemoryRegionSection {
     MemoryRegion *mr;
-    AddressSpace *address_space;
+    FlatView *fv;
     hwaddr offset_within_region;
     Int128 size;
     hwaddr offset_within_address_space;
@@ -1515,7 +1512,8 @@ void memory_global_dirty_log_start(void);
  */
 void memory_global_dirty_log_stop(void);
 
-void mtree_info(fprintf_function mon_printf, void *f, bool flatview);
+void mtree_info(fprintf_function mon_printf, void *f, bool flatview,
+                bool dispatch_tree);
 
 /**
  * memory_region_request_mmio_ptr: request a pointer to an mmio
@@ -1584,23 +1582,6 @@ MemTxResult memory_region_dispatch_write(MemoryRegion *mr,
  */
 void address_space_init(AddressSpace *as, MemoryRegion *root, const char *name);
 
-/**
- * address_space_init_shareable: return an address space for a memory region,
- *                               creating it if it does not already exist
- *
- * @root: a #MemoryRegion that routes addresses for the address space
- * @name: an address space name.  The name is only used for debugging
- *        output.
- *
- * This function will return a pointer to an existing AddressSpace
- * which was initialized with the specified MemoryRegion, or it will
- * create and initialize one if it does not already exist. The ASes
- * are reference-counted, so the memory will be freed automatically
- * when the AddressSpace is destroyed via address_space_destroy.
- */
-AddressSpace *address_space_init_shareable(MemoryRegion *root,
-                                           const char *name);
-
 /**
  * address_space_destroy: destroy an address space
  *
@@ -1845,9 +1826,17 @@ IOMMUTLBEntry address_space_get_iotlb_entry(AddressSpace *as, hwaddr addr,
  * @len: pointer to length
  * @is_write: indicates the transfer direction
  */
-MemoryRegion *address_space_translate(AddressSpace *as, hwaddr addr,
-                                      hwaddr *xlat, hwaddr *len,
-                                      bool is_write);
+MemoryRegion *flatview_translate(FlatView *fv,
+                                 hwaddr addr, hwaddr *xlat,
+                                 hwaddr *len, bool is_write);
+
+static inline MemoryRegion *address_space_translate(AddressSpace *as,
+                                                    hwaddr addr, hwaddr *xlat,
+                                                    hwaddr *len, bool is_write)
+{
+    return flatview_translate(address_space_to_flatview(as),
+                              addr, xlat, len, is_write);
+}
 
 /* address_space_access_valid: check for validity of accessing an address
  * space range
@@ -1898,12 +1887,13 @@ void address_space_unmap(AddressSpace *as, void *buffer, hwaddr len,
 
 
 /* Internal functions, part of the implementation of address_space_read.  */
-MemTxResult address_space_read_continue(AddressSpace *as, hwaddr addr,
-                                        MemTxAttrs attrs, uint8_t *buf,
-                                        int len, hwaddr addr1, hwaddr l,
-					MemoryRegion *mr);
-MemTxResult address_space_read_full(AddressSpace *as, hwaddr addr,
-                                    MemTxAttrs attrs, uint8_t *buf, int len);
+MemTxResult flatview_read_continue(FlatView *fv, hwaddr addr,
+                                   MemTxAttrs attrs, uint8_t *buf,
+                                   int len, hwaddr addr1, hwaddr l,
+                                   MemoryRegion *mr);
+
+MemTxResult flatview_read_full(FlatView *fv, hwaddr addr,
+                               MemTxAttrs attrs, uint8_t *buf, int len);
 void *qemu_map_ram_ptr(RAMBlock *ram_block, ram_addr_t addr);
 
 static inline bool memory_access_is_direct(MemoryRegion *mr, bool is_write)
@@ -1930,8 +1920,8 @@ static inline bool memory_access_is_direct(MemoryRegion *mr, bool is_write)
  * @buf: buffer with the data transferred
  */
 static inline __attribute__((__always_inline__))
-MemTxResult address_space_read(AddressSpace *as, hwaddr addr, MemTxAttrs attrs,
-                               uint8_t *buf, int len)
+MemTxResult flatview_read(FlatView *fv, hwaddr addr, MemTxAttrs attrs,
+                          uint8_t *buf, int len)
 {
     MemTxResult result = MEMTX_OK;
     hwaddr l, addr1;
@@ -1942,22 +1932,29 @@ MemTxResult address_space_read(AddressSpace *as, hwaddr addr, MemTxAttrs attrs,
         if (len) {
             rcu_read_lock();
             l = len;
-            mr = address_space_translate(as, addr, &addr1, &l, false);
+            mr = flatview_translate(fv, addr, &addr1, &l, false);
             if (len == l && memory_access_is_direct(mr, false)) {
                 ptr = qemu_map_ram_ptr(mr->ram_block, addr1);
                 memcpy(buf, ptr, len);
             } else {
-                result = address_space_read_continue(as, addr, attrs, buf, len,
-                                                     addr1, l, mr);
+                result = flatview_read_continue(fv, addr, attrs, buf, len,
+                                                addr1, l, mr);
             }
             rcu_read_unlock();
         }
     } else {
-        result = address_space_read_full(as, addr, attrs, buf, len);
+        result = flatview_read_full(fv, addr, attrs, buf, len);
     }
     return result;
 }
 
+static inline MemTxResult address_space_read(AddressSpace *as, hwaddr addr,
+                                             MemTxAttrs attrs, uint8_t *buf,
+                                             int len)
+{
+    return flatview_read(address_space_to_flatview(as), addr, attrs, buf, len);
+}
+
 /**
  * address_space_read_cached: read from a cached RAM region
  *

include/hw/arm/armv7m.h

@@ -21,7 +21,7 @@ typedef struct {
     SysBusDevice parent_obj;
     /*< public >*/
 
-    AddressSpace *source_as;
+    AddressSpace source_as;
     MemoryRegion iomem;
     uint32_t base;
     MemoryRegion *source_memory;

include/hw/virtio/virtio-serial.h

@@ -58,6 +58,9 @@ typedef struct VirtIOSerialPortClass {
         /* Guest opened/closed device. */
     void (*set_guest_connected)(VirtIOSerialPort *port, int guest_connected);
 
+    /* Enable/disable backend for virtio serial port */
+    void (*enable_backend)(VirtIOSerialPort *port, bool enable);
+
         /* Guest is now ready to accept data (virtqueues set up). */
     void (*guest_ready)(VirtIOSerialPort *port);
 

include/qemu/atomic.h

@@ -442,4 +442,12 @@
 } while(0)
 #endif
 
+#define atomic_fetch_inc_nonzero(ptr) ({                                \
+    typeof_strip_qual(*ptr) _oldn = atomic_read(ptr);                   \
+    while (_oldn && atomic_cmpxchg(ptr, _oldn, _oldn + 1) != _oldn) {   \
+        _oldn = atomic_read(ptr);                                       \
+    }                                                                   \
+    _oldn;                                                              \
+})
+
 #endif /* QEMU_ATOMIC_H */

include/qemu/typedefs.h

@@ -30,6 +30,7 @@ typedef struct DisplaySurface DisplaySurface;
 typedef struct DriveInfo DriveInfo;
 typedef struct Error Error;
 typedef struct EventNotifier EventNotifier;
+typedef struct FlatView FlatView;
 typedef struct FWCfgEntry FWCfgEntry;
 typedef struct FWCfgIoState FWCfgIoState;
 typedef struct FWCfgMemState FWCfgMemState;

include/scsi/pr-manager.h

+#ifndef PR_MANAGER_H
+#define PR_MANAGER_H
+
+#include "qom/object.h"
+#include "qapi/qmp/qdict.h"
+#include "qapi/visitor.h"
+#include "qom/object_interfaces.h"
+#include "block/aio.h"
+
+#define TYPE_PR_MANAGER "pr-manager"
+
+#define PR_MANAGER_CLASS(klass) \
+     OBJECT_CLASS_CHECK(PRManagerClass, (klass), TYPE_PR_MANAGER)
+#define PR_MANAGER_GET_CLASS(obj) \
+     OBJECT_GET_CLASS(PRManagerClass, (obj), TYPE_PR_MANAGER)
+#define PR_MANAGER(obj) \
+     OBJECT_CHECK(PRManager, (obj), TYPE_PR_MANAGER)
+
+struct sg_io_hdr;
+
+typedef struct PRManager {
+    /* <private> */
+    Object parent;
+} PRManager;
+
+/**
+ * PRManagerClass:
+ * @parent_class: the base class
+ * @run: callback invoked in thread pool context
+ */
+typedef struct PRManagerClass {
+    /* <private> */
+    ObjectClass parent_class;
+
+    /* <public> */
+    int (*run)(PRManager *pr_mgr, int fd, struct sg_io_hdr *hdr);
+} PRManagerClass;
+
+BlockAIOCB *pr_manager_execute(PRManager *pr_mgr,
+                               AioContext *ctx, int fd,
+                               struct sg_io_hdr *hdr,
+                               BlockCompletionFunc *complete,
+                               void *opaque);
+
+#ifdef CONFIG_LINUX
+PRManager *pr_manager_lookup(const char *id, Error **errp);
+#else
+static inline PRManager *pr_manager_lookup(const char *id, Error **errp)
+{
+    /* The classes do not exist at all!  */
+    error_setg(errp, "No persistent reservation manager with id '%s'", id);
+    return NULL;
+}
+#endif
+
+#endif

include/scsi/utils.h

@@ -72,10 +72,14 @@ extern const struct SCSISense sense_code_IO_ERROR;
 extern const struct SCSISense sense_code_I_T_NEXUS_LOSS;
 /* Command aborted, Logical Unit failure */
 extern const struct SCSISense sense_code_LUN_FAILURE;
+/* Command aborted, LUN Communication failure */
+extern const struct SCSISense sense_code_LUN_COMM_FAILURE;
 /* Command aborted, Overlapped Commands Attempted */
 extern const struct SCSISense sense_code_OVERLAPPED_COMMANDS;
 /* LUN not ready, Capacity data has changed */
 extern const struct SCSISense sense_code_CAPACITY_CHANGED;
+/* Unit attention, SCSI bus reset */
+extern const struct SCSISense sense_code_SCSI_BUS_RESET;
 /* LUN not ready, Medium not present */
 extern const struct SCSISense sense_code_UNIT_ATTENTION_NO_MEDIUM;
 /* Unit attention, Power on, reset or bus device reset occurred */

monitor.c

@@ -1703,8 +1703,9 @@ static void hmp_boot_set(Monitor *mon, const QDict *qdict)
 static void hmp_info_mtree(Monitor *mon, const QDict *qdict)
 {
     bool flatview = qdict_get_try_bool(qdict, "flatview", false);
+    bool dispatch_tree = qdict_get_try_bool(qdict, "dispatch_tree", false);
 
-    mtree_info((fprintf_function)monitor_printf, mon, flatview);
+    mtree_info((fprintf_function)monitor_printf, mon, flatview, dispatch_tree);
 }
 
 static void hmp_info_numa(Monitor *mon, const QDict *qdict)

qapi/block-core.json

@@ -2241,6 +2241,9 @@
 # Driver specific block device options for the file backend.
 #
 # @filename:    path to the image file
+# @pr-manager:  the id for the object that will handle persistent reservations
+#               for this device (default: none, forward the commands via SG_IO;
+#               since 2.11)
 # @aio:         AIO backend (default: threads) (since: 2.8)
 # @locking:     whether to enable file locking. If set to 'auto', only enable
 #               when Open File Descriptor (OFD) locking API is available
@@ -2250,6 +2253,7 @@
 ##
 { 'struct': 'BlockdevOptionsFile',
   'data': { 'filename': 'str',
+            '*pr-manager': 'str',
             '*locking': 'OnOffAuto',
             '*aio': 'BlockdevAioOptions' } }
 

scsi/Makefile.objs

 block-obj-y += utils.o
+
+block-obj-$(CONFIG_LINUX) += pr-manager.o pr-manager-helper.o

scsi/pr-helper.h

+/* Definitions for QEMU's persistent reservation helper daemon
+ *
+ * Copyright (C) 2017 Red Hat, Inc.
+ *
+ * Author:
+ *   Paolo Bonzini <pbonzini@redhat.com>
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a copy
+ * of this software and associated documentation files (the "Software"), to
+ * deal in the Software without restriction, including without limitation the
+ * rights to use, copy, modify, merge, publish, distribute, sublicense, and/or
+ * sell copies of the Software, and to permit persons to whom the Software is
+ * furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+ * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
+ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
+ * IN THE SOFTWARE.
+ */
+#ifndef QEMU_PR_HELPER_H
+#define QEMU_PR_HELPER_H 1
+
+#include <stdint.h>
+
+#define PR_HELPER_CDB_SIZE     16
+#define PR_HELPER_SENSE_SIZE   96
+#define PR_HELPER_DATA_SIZE    8192
+
+typedef struct PRHelperResponse {
+    int32_t result;
+    int32_t sz;
+    uint8_t sense[PR_HELPER_SENSE_SIZE];
+} PRHelperResponse;
+
+#endif

scsi/pr-manager-helper.c

+/*
+ * Persistent reservation manager that talks to qemu-pr-helper
+ *
+ * Copyright (c) 2017 Red Hat, Inc.
+ *
+ * Author: Paolo Bonzini <pbonzini@redhat.com>
+ *
+ * This code is licensed under the LGPL v2.1 or later.
+ *
+ */
+
+#include "qemu/osdep.h"
+#include "qapi/error.h"
+#include "scsi/constants.h"
+#include "scsi/pr-manager.h"
+#include "scsi/utils.h"
+#include "io/channel.h"
+#include "io/channel-socket.h"
+#include "pr-helper.h"
+
+#include <scsi/sg.h>
+
+#define PR_MAX_RECONNECT_ATTEMPTS 5
+
+#define TYPE_PR_MANAGER_HELPER "pr-manager-helper"
+
+#define PR_MANAGER_HELPER(obj) \
+     OBJECT_CHECK(PRManagerHelper, (obj), \
+                  TYPE_PR_MANAGER_HELPER)
+
+typedef struct PRManagerHelper {
+    /* <private> */
+    PRManager parent;
+
+    char *path;
+
+    QemuMutex lock;
+    QIOChannel *ioc;
+} PRManagerHelper;
+
+/* Called with lock held.  */
+static int pr_manager_helper_read(PRManagerHelper *pr_mgr,
+                                  void *buf, int sz, Error **errp)
+{
+    ssize_t r = qio_channel_read_all(pr_mgr->ioc, buf, sz, errp);
+
+    if (r < 0) {
+        object_unref(OBJECT(pr_mgr->ioc));
+        pr_mgr->ioc = NULL;
+        return -EINVAL;
+    }
+
+    return 0;
+}
+
+/* Called with lock held.  */
+static int pr_manager_helper_write(PRManagerHelper *pr_mgr,
+                                   int fd,
+                                   const void *buf, int sz, Error **errp)
+{
+    size_t nfds = (fd != -1);
+    while (sz > 0) {
+        struct iovec iov;
+        ssize_t n_written;
+
+        iov.iov_base = (void *)buf;
+        iov.iov_len = sz;
+        n_written = qio_channel_writev_full(QIO_CHANNEL(pr_mgr->ioc), &iov, 1,
+                                            nfds ? &fd : NULL, nfds, errp);
+
+        if (n_written <= 0) {
+            assert(n_written != QIO_CHANNEL_ERR_BLOCK);
+            object_unref(OBJECT(pr_mgr->ioc));
+            return n_written < 0 ? -EINVAL : 0;
+        }
+
+        nfds = 0;
+        buf += n_written;
+        sz -= n_written;
+    }
+
+    return 0;
+}
+
+/* Called with lock held.  */
+static int pr_manager_helper_initialize(PRManagerHelper *pr_mgr,
+                                        Error **errp)
+{
+    char *path = g_strdup(pr_mgr->path);
+    SocketAddress saddr = {
+        .type = SOCKET_ADDRESS_TYPE_UNIX,
+        .u.q_unix.path = path
+    };
+    QIOChannelSocket *sioc = qio_channel_socket_new();
+    Error *local_err = NULL;
+
+    uint32_t flags;
+    int r;
+
+    assert(!pr_mgr->ioc);
+    qio_channel_set_name(QIO_CHANNEL(sioc), "pr-manager-helper");
+    qio_channel_socket_connect_sync(sioc,
+                                    &saddr,
+                                    &local_err);
+    g_free(path);
+    if (local_err) {
+        object_unref(OBJECT(sioc));
+        error_propagate(errp, local_err);
+        return -ENOTCONN;
+    }
+
+    qio_channel_set_delay(QIO_CHANNEL(sioc), false);
+    pr_mgr->ioc = QIO_CHANNEL(sioc);
+
+    /* A simple feature negotation protocol, even though there is
+     * no optional feature right now.
+     */
+    r = pr_manager_helper_read(pr_mgr, &flags, sizeof(flags), errp);
+    if (r < 0) {
+        goto out_close;
+    }
+
+    flags = 0;
+    r = pr_manager_helper_write(pr_mgr, -1, &flags, sizeof(flags), errp);
+    if (r < 0) {
+        goto out_close;
+    }
+
+    return 0;
+
+out_close:
+    object_unref(OBJECT(pr_mgr->ioc));
+    pr_mgr->ioc = NULL;
+    return r;
+}
+
+static int pr_manager_helper_run(PRManager *p,
+                                 int fd, struct sg_io_hdr *io_hdr)
+{
+    PRManagerHelper *pr_mgr = PR_MANAGER_HELPER(p);
+
+    uint32_t len;
+    PRHelperResponse resp;
+    int ret;
+    int expected_dir;
+    int attempts;
+    uint8_t cdb[PR_HELPER_CDB_SIZE] = { 0 };
+
+    if (!io_hdr->cmd_len || io_hdr->cmd_len > PR_HELPER_CDB_SIZE) {
+        return -EINVAL;
+    }
+
+    memcpy(cdb, io_hdr->cmdp, io_hdr->cmd_len);
+    assert(cdb[0] == PERSISTENT_RESERVE_OUT || cdb[0] == PERSISTENT_RESERVE_IN);
+    expected_dir =
+        (cdb[0] == PERSISTENT_RESERVE_OUT ? SG_DXFER_TO_DEV : SG_DXFER_FROM_DEV);
+    if (io_hdr->dxfer_direction != expected_dir) {
+        return -EINVAL;
+    }
+
+    len = scsi_cdb_xfer(cdb);
+    if (io_hdr->dxfer_len < len || len > PR_HELPER_DATA_SIZE) {
+        return -EINVAL;
+    }
+
+    qemu_mutex_lock(&pr_mgr->lock);
+
+    /* Try to reconnect while sending the CDB.  */
+    for (attempts = 0; attempts < PR_MAX_RECONNECT_ATTEMPTS; attempts++) {
+        if (!pr_mgr->ioc) {
+            ret = pr_manager_helper_initialize(pr_mgr, NULL);
+            if (ret < 0) {
+                qemu_mutex_unlock(&pr_mgr->lock);
+                g_usleep(G_USEC_PER_SEC);
+                qemu_mutex_lock(&pr_mgr->lock);
+                continue;
+            }
+        }
+
+        ret = pr_manager_helper_write(pr_mgr, fd, cdb, ARRAY_SIZE(cdb), NULL);
+        if (ret >= 0) {
+            break;
+        }
+    }
+    if (ret < 0) {
+        goto out;
+    }
+
+    /* After sending the CDB, any communications failure causes the
+     * command to fail.  The failure is transient, retrying the command
+     * will invoke pr_manager_helper_initialize again.
+     */
+    if (expected_dir == SG_DXFER_TO_DEV) {
+        io_hdr->resid = io_hdr->dxfer_len - len;
+        ret = pr_manager_helper_write(pr_mgr, -1, io_hdr->dxferp, len, NULL);
+        if (ret < 0) {
+            goto out;
+        }
+    }
+    ret = pr_manager_helper_read(pr_mgr, &resp, sizeof(resp), NULL);
+    if (ret < 0) {
+        goto out;
+    }
+
+    resp.result = be32_to_cpu(resp.result);
+    resp.sz = be32_to_cpu(resp.sz);
+    if (io_hdr->dxfer_direction == SG_DXFER_FROM_DEV) {
+        assert(resp.sz <= io_hdr->dxfer_len);
+        ret = pr_manager_helper_read(pr_mgr, io_hdr->dxferp, resp.sz, NULL);
+        if (ret < 0) {
+            goto out;
+        }
+        io_hdr->resid = io_hdr->dxfer_len - resp.sz;
+    } else {
+        assert(resp.sz == 0);
+    }
+
+    io_hdr->status = resp.result;
+    if (resp.result == CHECK_CONDITION) {
+        io_hdr->driver_status = SG_ERR_DRIVER_SENSE;
+        io_hdr->sb_len_wr = MIN(io_hdr->mx_sb_len, PR_HELPER_SENSE_SIZE);
+        memcpy(io_hdr->sbp, resp.sense, io_hdr->sb_len_wr);
+    }
+
+out:
+    if (ret < 0) {
+        int sense_len = scsi_build_sense(io_hdr->sbp,
+                                         SENSE_CODE(LUN_COMM_FAILURE));
+        io_hdr->driver_status = SG_ERR_DRIVER_SENSE;
+        io_hdr->sb_len_wr = MIN(io_hdr->mx_sb_len, sense_len);
+        io_hdr->status = CHECK_CONDITION;
+    }
+    qemu_mutex_unlock(&pr_mgr->lock);
+    return ret;
+}
+
+static void pr_manager_helper_complete(UserCreatable *uc, Error **errp)
+{
+    PRManagerHelper *pr_mgr = PR_MANAGER_HELPER(uc);
+
+    qemu_mutex_lock(&pr_mgr->lock);
+    pr_manager_helper_initialize(pr_mgr, errp);
+    qemu_mutex_unlock(&pr_mgr->lock);
+}
+
+static char *get_path(Object *obj, Error **errp)
+{
+    PRManagerHelper *pr_mgr = PR_MANAGER_HELPER(obj);
+
+    return g_strdup(pr_mgr->path);
+}
+
+static void set_path(Object *obj, const char *str, Error **errp)
+{
+    PRManagerHelper *pr_mgr = PR_MANAGER_HELPER(obj);
+
+    g_free(pr_mgr->path);
+    pr_mgr->path = g_strdup(str);
+}
+
+static void pr_manager_helper_instance_finalize(Object *obj)
+{
+    PRManagerHelper *pr_mgr = PR_MANAGER_HELPER(obj);
+
+    object_unref(OBJECT(pr_mgr->ioc));
+    qemu_mutex_destroy(&pr_mgr->lock);
+}
+
+static void pr_manager_helper_instance_init(Object *obj)
+{
+    PRManagerHelper *pr_mgr = PR_MANAGER_HELPER(obj);
+
+    qemu_mutex_init(&pr_mgr->lock);
+}
+
+static void pr_manager_helper_class_init(ObjectClass *klass,
+                                         void *class_data G_GNUC_UNUSED)
+{
+    PRManagerClass *prmgr_klass = PR_MANAGER_CLASS(klass);
+    UserCreatableClass *uc_klass = USER_CREATABLE_CLASS(klass);
+
+    object_class_property_add_str(klass, "path", get_path, set_path,
+                                  &error_abort);
+    uc_klass->complete = pr_manager_helper_complete;
+    prmgr_klass->run = pr_manager_helper_run;
+}
+
+static const TypeInfo pr_manager_helper_info = {
+    .parent = TYPE_PR_MANAGER,
+    .name = TYPE_PR_MANAGER_HELPER,
+    .instance_size = sizeof(PRManagerHelper),
+    .instance_init = pr_manager_helper_instance_init,
+    .instance_finalize = pr_manager_helper_instance_finalize,
+    .class_init = pr_manager_helper_class_init,
+};
+
+static void pr_manager_helper_register_types(void)
+{
+    type_register_static(&pr_manager_helper_info);
+}
+
+type_init(pr_manager_helper_register_types);

scsi/pr-manager.c

+/*
+ * Persistent reservation manager abstract class
+ *
+ * Copyright (c) 2017 Red Hat, Inc.
+ *
+ * Author: Paolo Bonzini <pbonzini@redhat.com>
+ *
+ * This code is licensed under the LGPL.
+ *
+ */
+
+#include "qemu/osdep.h"
+#include <scsi/sg.h>
+
+#include "qapi/error.h"
+#include "block/aio.h"
+#include "block/thread-pool.h"
+#include "scsi/pr-manager.h"
+#include "trace.h"
+
+typedef struct PRManagerData {
+    PRManager *pr_mgr;
+    struct sg_io_hdr *hdr;
+    int fd;
+} PRManagerData;
+
+static int pr_manager_worker(void *opaque)
+{
+    PRManagerData *data = opaque;
+    PRManager *pr_mgr = data->pr_mgr;
+    PRManagerClass *pr_mgr_class =
+        PR_MANAGER_GET_CLASS(pr_mgr);
+    struct sg_io_hdr *hdr = data->hdr;
+    int fd = data->fd;
+    int r;
+
+    g_free(data);
+    trace_pr_manager_run(fd, hdr->cmdp[0], hdr->cmdp[1]);
+
+    /* The reference was taken in pr_manager_execute.  */
+    r = pr_mgr_class->run(pr_mgr, fd, hdr);
+    object_unref(OBJECT(pr_mgr));
+    return r;
+}
+
+
+BlockAIOCB *pr_manager_execute(PRManager *pr_mgr,
+                               AioContext *ctx, int fd,
+                               struct sg_io_hdr *hdr,
+                               BlockCompletionFunc *complete,
+                               void *opaque)
+{
+    PRManagerData *data = g_new(PRManagerData, 1);
+    ThreadPool *pool = aio_get_thread_pool(ctx);
+
+    trace_pr_manager_execute(fd, hdr->cmdp[0], hdr->cmdp[1], opaque);
+    data->pr_mgr = pr_mgr;
+    data->fd = fd;
+    data->hdr = hdr;
+
+    /* The matching object_unref is in pr_manager_worker.  */
+    object_ref(OBJECT(pr_mgr));
+    return thread_pool_submit_aio(pool, pr_manager_worker,
+                                  data, complete, opaque);
+}
+
+static const TypeInfo pr_manager_info = {
+    .parent = TYPE_OBJECT,
+    .name = TYPE_PR_MANAGER,
+    .class_size = sizeof(PRManagerClass),
+    .abstract = true,
+    .interfaces = (InterfaceInfo[]) {
+        { TYPE_USER_CREATABLE },
+        { }
+    }
+};
+
+PRManager *pr_manager_lookup(const char *id, Error **errp)
+{
+    Object *obj;
+    PRManager *pr_mgr;
+
+    obj = object_resolve_path_component(object_get_objects_root(), id);
+    if (!obj) {
+        error_setg(errp, "No persistent reservation manager with id '%s'", id);
+        return NULL;
+    }
+
+    pr_mgr = (PRManager *)
+        object_dynamic_cast(obj,
+                            TYPE_PR_MANAGER);
+    if (!pr_mgr) {
+        error_setg(errp,
+                   "Object with id '%s' is not a persistent reservation manager",
+                   id);
+        return NULL;
+    }
+
+    return pr_mgr;
+}
+
+static void
+pr_manager_register_types(void)
+{
+    type_register_static(&pr_manager_info);
+}
+
+
+type_init(pr_manager_register_types);

scsi/trace-events

+# scsi/pr-manager.c
+pr_manager_execute(int fd, int cmd, int sa, void *opaque) "fd=%d cmd=0x%02x service action=0x%02x opaque=%p"
+pr_manager_run(int fd, int cmd, int sa) "fd=%d cmd=0x%02x service action=0x%02x"

scsi/utils.c

@@ -206,6 +206,11 @@ const struct SCSISense sense_code_OVERLAPPED_COMMANDS = {
     .key = ABORTED_COMMAND, .asc = 0x4e, .ascq = 0x00
 };
 
+/* Command aborted, LUN Communication Failure */
+const struct SCSISense sense_code_LUN_COMM_FAILURE = {
+    .key = ABORTED_COMMAND, .asc = 0x08, .ascq = 0x00
+};
+
 /* Unit attention, Capacity data has changed */
 const struct SCSISense sense_code_CAPACITY_CHANGED = {
     .key = UNIT_ATTENTION, .asc = 0x2a, .ascq = 0x09
@@ -216,6 +221,11 @@ const struct SCSISense sense_code_RESET = {
     .key = UNIT_ATTENTION, .asc = 0x29, .ascq = 0x00
 };
 
+/* Unit attention, SCSI bus reset */
+const struct SCSISense sense_code_SCSI_BUS_RESET = {
+    .key = UNIT_ATTENTION, .asc = 0x29, .ascq = 0x02
+};
+
 /* Unit attention, No medium */
 const struct SCSISense sense_code_UNIT_ATTENTION_NO_MEDIUM = {
     .key = UNIT_ATTENTION, .asc = 0x3a, .ascq = 0x00

target/arm/cpu.c

@@ -691,6 +691,9 @@ static void arm_cpu_realizefn(DeviceState *dev, Error **errp)
     CPUARMState *env = &cpu->env;
     int pagebits;
     Error *local_err = NULL;
+#ifndef CONFIG_USER_ONLY
+    AddressSpace *as;
+#endif
 
     cpu_exec_realizefn(cs, &local_err);
     if (local_err != NULL) {
@@ -881,24 +884,21 @@ static void arm_cpu_realizefn(DeviceState *dev, Error **errp)
 
 #ifndef CONFIG_USER_ONLY
     if (cpu->has_el3 || arm_feature(env, ARM_FEATURE_M_SECURITY)) {
-        AddressSpace *as;
+        as = g_new0(AddressSpace, 1);
 
         cs->num_ases = 2;
 
         if (!cpu->secure_memory) {
             cpu->secure_memory = cs->memory;
         }
-        as = address_space_init_shareable(cpu->secure_memory,
-                                          "cpu-secure-memory");
+        address_space_init(as, cpu->secure_memory, "cpu-secure-memory");
         cpu_address_space_init(cs, as, ARMASIdx_S);
     } else {
         cs->num_ases = 1;
     }
-
-    cpu_address_space_init(cs,
-                           address_space_init_shareable(cs->memory,
-                                                        "cpu-memory"),
-                           ARMASIdx_NS);
+    as = g_new0(AddressSpace, 1);
+    address_space_init(as, cs->memory, "cpu-memory");
+    cpu_address_space_init(cs, as, ARMASIdx_NS);
 #endif
 
     qemu_init_vcpu(cs);

target/i386/cpu.c

@@ -3738,10 +3738,11 @@ static void x86_cpu_realizefn(DeviceState *dev, Error **errp)
 
 #ifndef CONFIG_USER_ONLY
     if (tcg_enabled()) {
-        AddressSpace *as_normal = address_space_init_shareable(cs->memory,
-                                                               "cpu-memory");
+        AddressSpace *as_normal = g_new0(AddressSpace, 1);
         AddressSpace *as_smm = g_new(AddressSpace, 1);
 
+        address_space_init(as_normal, cs->memory, "cpu-memory");
+
         cpu->cpu_as_mem = g_new(MemoryRegion, 1);
         cpu->cpu_as_root = g_new(MemoryRegion, 1);
 

trace-events

@@ -64,6 +64,9 @@ memory_region_tb_read(int cpu_index, uint64_t addr, uint64_t value, unsigned siz
 memory_region_tb_write(int cpu_index, uint64_t addr, uint64_t value, unsigned size) "cpu %d addr 0x%"PRIx64" value 0x%"PRIx64" size %u"
 memory_region_ram_device_read(int cpu_index, void *mr, uint64_t addr, uint64_t value, unsigned size) "cpu %d mr %p addr 0x%"PRIx64" value 0x%"PRIx64" size %u"
 memory_region_ram_device_write(int cpu_index, void *mr, uint64_t addr, uint64_t value, unsigned size) "cpu %d mr %p addr 0x%"PRIx64" value 0x%"PRIx64" size %u"
+flatview_new(FlatView *view, MemoryRegion *root) "%p (root %p)"
+flatview_destroy(FlatView *view, MemoryRegion *root) "%p (root %p)"
+flatview_destroy_rcu(FlatView *view, MemoryRegion *root) "%p (root %p)"
 
 ### Guest events, keep at bottom
 

vl.c

@@ -2889,7 +2889,8 @@ static int machine_set_property(void *opaque,
  */
 static bool object_create_initial(const char *type)
 {
-    if (g_str_equal(type, "rng-egd")) {
+    if (g_str_equal(type, "rng-egd") ||
+        g_str_has_prefix(type, "pr-manager-")) {
         return false;
     }