diff --git a/Changelog b/Changelog
index 1c41e14..28a69af 100644
--- a/Changelog
+++ b/Changelog
@@ -1,3 +1,9 @@
+This file documents changes for QEMU releases 0.12 and earlier.
+For changelog information for later releases, see
+http://wiki.qemu.org/ChangeLog or look at the git history for
+more detailed information.
+
+
 version 0.12.0:
 
   - Update to SeaBIOS 0.5.0
diff --git a/MAINTAINERS b/MAINTAINERS
index e6f853d..35d4496 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -56,8 +56,8 @@
 Guest CPU cores (TCG):
 ----------------------
 Alpha
-M: qemu-devel@nongnu.org
-S: Orphan
+M: Richard Henderson <rth@twiddle.net>
+S: Maintained
 F: target-alpha/
 
 ARM
diff --git a/Makefile b/Makefile
index 6d58710..b3ffbe2 100644
--- a/Makefile
+++ b/Makefile
@@ -151,14 +151,14 @@
 
 check-qint.o check-qstring.o check-qdict.o check-qlist.o check-qfloat.o check-qjson.o: $(GENERATED_HEADERS)
 
-CHECK_PROG_DEPS = qemu-malloc.o $(oslib-obj-y) $(trace-obj-y)
+CHECK_PROG_DEPS = qemu-malloc.o $(oslib-obj-y) $(trace-obj-y) qemu-tool.o
 
 check-qint: check-qint.o qint.o $(CHECK_PROG_DEPS)
 check-qstring: check-qstring.o qstring.o $(CHECK_PROG_DEPS)
 check-qdict: check-qdict.o qdict.o qfloat.o qint.o qstring.o qbool.o qlist.o $(CHECK_PROG_DEPS)
 check-qlist: check-qlist.o qlist.o qint.o $(CHECK_PROG_DEPS)
 check-qfloat: check-qfloat.o qfloat.o $(CHECK_PROG_DEPS)
-check-qjson: check-qjson.o qfloat.o qint.o qdict.o qstring.o qlist.o qbool.o qjson.o json-streamer.o json-lexer.o json-parser.o $(CHECK_PROG_DEPS)
+check-qjson: check-qjson.o qfloat.o qint.o qdict.o qstring.o qlist.o qbool.o qjson.o json-streamer.o json-lexer.o json-parser.o error.o qerror.o qemu-error.o $(CHECK_PROG_DEPS)
 
 QEMULIBS=libhw32 libhw64 libuser libdis libdis-user
 
@@ -203,6 +203,7 @@
 pxe-e1000.rom pxe-eepro100.rom pxe-ne2k_pci.rom \
 pxe-pcnet.rom pxe-rtl8139.rom pxe-virtio.rom \
 bamboo.dtb petalogix-s3adsp1800.dtb petalogix-ml605.dtb \
+mpc8544ds.dtb \
 multiboot.bin linuxboot.bin \
 s390-zipl.rom \
 spapr-rtas.bin slof.bin
diff --git a/Makefile.objs b/Makefile.objs
index 0b235db..f617ed5 100644
--- a/Makefile.objs
+++ b/Makefile.objs
@@ -2,7 +2,7 @@
 # QObject
 qobject-obj-y = qint.o qstring.o qdict.o qlist.o qfloat.o qbool.o
 qobject-obj-y += qjson.o json-lexer.o json-streamer.o json-parser.o
-qobject-obj-y += qerror.o
+qobject-obj-y += qerror.o error.o
 
 #######################################################################
 # oslib-obj-y is code depending on the OS (win32 vs posix)
@@ -45,12 +45,14 @@
 net-nested-$(CONFIG_VDE) += vde.o
 net-obj-y += $(addprefix net/, $(net-nested-y))
 
-ifeq ($(CONFIG_VIRTIO)$(CONFIG_VIRTFS),yy)
+ifeq ($(CONFIG_VIRTIO)$(CONFIG_VIRTFS)$(CONFIG_PCI),yyy)
 # Lots of the fsdev/9pcode is pulled in by vl.c via qemu_fsdev_add.
 # only pull in the actual virtio-9p device if we also enabled virtio.
 CONFIG_REALLY_VIRTFS=y
+fsdev-nested-y = qemu-fsdev.o
+else
+fsdev-nested-y = qemu-fsdev-dummy.o
 endif
-fsdev-nested-$(CONFIG_VIRTFS) = qemu-fsdev.o
 fsdev-obj-$(CONFIG_VIRTFS) += $(addprefix fsdev/, $(fsdev-nested-y))
 
 ######################################################################
@@ -170,6 +172,7 @@
 hw-obj-y =
 hw-obj-y += vl.o loader.o
 hw-obj-$(CONFIG_VIRTIO) += virtio.o virtio-console.o
+hw-obj-$(CONFIG_VIRTIO_PCI) += virtio-pci.o
 hw-obj-y += fw_cfg.o
 hw-obj-$(CONFIG_PCI) += pci.o pci_bridge.o
 hw-obj-$(CONFIG_PCI) += msix.o msi.o
@@ -285,12 +288,11 @@
 adlib.o fmopl.o: QEMU_CFLAGS += -DBUILD_Y8950=0
 hw-obj-$(CONFIG_SOUND) += $(sound-obj-y)
 
-9pfs-nested-$(CONFIG_REALLY_VIRTFS) = virtio-9p-debug.o
+9pfs-nested-$(CONFIG_VIRTFS) = virtio-9p.o virtio-9p-debug.o
 9pfs-nested-$(CONFIG_VIRTFS) +=  virtio-9p-local.o virtio-9p-xattr.o
 9pfs-nested-$(CONFIG_VIRTFS) +=   virtio-9p-xattr-user.o virtio-9p-posix-acl.o
 
-hw-obj-$(CONFIG_VIRTFS) += $(addprefix 9pfs/, $(9pfs-nested-y))
-$(addprefix 9pfs/, $(9pfs-nested-y)): CFLAGS +=  -I$(SRC_PATH)/hw/
+hw-obj-$(CONFIG_REALLY_VIRTFS) += $(addprefix 9pfs/, $(9pfs-nested-y))
 
 
 ######################################################################
diff --git a/Makefile.target b/Makefile.target
index 602d50d..b1a0f6d 100644
--- a/Makefile.target
+++ b/Makefile.target
@@ -71,8 +71,7 @@
 # cpu emulator library
 libobj-y = exec.o translate-all.o cpu-exec.o translate.o
 libobj-y += tcg/tcg.o
-libobj-$(CONFIG_SOFTFLOAT) += fpu/softfloat.o
-libobj-$(CONFIG_NOSOFTFLOAT) += fpu/softfloat-native.o
+libobj-y += fpu/softfloat.o
 libobj-y += op_helper.o helper.o
 ifeq ($(TARGET_BASE_ARCH), i386)
 libobj-y += cpuid.o
@@ -191,10 +190,9 @@
 # need to fix this properly
 obj-$(CONFIG_NO_PCI) += pci-stub.o
 obj-$(CONFIG_VIRTIO) += virtio-blk.o virtio-balloon.o virtio-net.o virtio-serial-bus.o
-obj-$(CONFIG_VIRTIO_PCI) += virtio-pci.o
 obj-y += vhost_net.o
 obj-$(CONFIG_VHOST_NET) += vhost.o
-obj-$(CONFIG_REALLY_VIRTFS) += 9pfs/virtio-9p.o
+obj-$(CONFIG_REALLY_VIRTFS) += 9pfs/virtio-9p-device.o
 obj-y += rwhandler.o
 obj-$(CONFIG_KVM) += kvm.o kvm-all.o
 obj-$(CONFIG_NO_KVM) += kvm-stub.o
@@ -232,7 +230,7 @@
 # Hardware support
 obj-i386-y += vga.o
 obj-i386-y += mc146818rtc.o i8259.o pc.o
-obj-i386-y += cirrus_vga.o apic.o ioapic.o piix_pci.o
+obj-i386-y += cirrus_vga.o sga.o apic.o ioapic.o piix_pci.o
 obj-i386-y += vmport.o
 obj-i386-y += device-hotplug.o pci-hotplug.o smbios.o wdt_ib700.o
 obj-i386-y += debugcon.o multiboot.o
@@ -376,7 +374,8 @@
 
 obj-s390x-y = s390-virtio-bus.o s390-virtio.o
 
-obj-alpha-y = alpha_palcode.o
+obj-alpha-y = i8259.o mc146818rtc.o
+obj-alpha-y += vga.o cirrus_vga.o
 
 main.o: QEMU_CFLAGS+=$(GPROF_CFLAGS)
 
@@ -413,8 +412,6 @@
 qmp-commands.h: $(SRC_PATH)/qmp-commands.hx
 	$(call quiet-command,sh $(SRC_PATH)/scripts/hxtool -h < $< > $@,"  GEN   $(TARGET_DIR)$@")
 
-9pfs/virtio-9p.o: CFLAGS +=  -I$(SRC_PATH)/hw/
-
 clean:
 	rm -f *.o *.a *~ $(PROGS) nwfpe/*.o fpu/*.o
 	rm -f *.d */*.d tcg/*.o ide/*.o 9pfs/*.o
diff --git a/QMP/qmp.py b/QMP/qmp.py
index 14ce8b0..c7dbea0 100644
--- a/QMP/qmp.py
+++ b/QMP/qmp.py
@@ -22,19 +22,24 @@
     pass
 
 class QEMUMonitorProtocol:
-    def __init__(self, address):
+    def __init__(self, address, server=False):
         """
         Create a QEMUMonitorProtocol class.
 
         @param address: QEMU address, can be either a unix socket path (string)
                         or a tuple in the form ( address, port ) for a TCP
                         connection
-        @note No connection is established, this is done by the connect() method
+        @param server: server mode listens on the socket (bool)
+        @raise socket.error on socket connection errors
+        @note No connection is established, this is done by the connect() or
+              accept() methods
         """
         self.__events = []
         self.__address = address
         self.__sock = self.__get_sock()
-        self.__sockfile = self.__sock.makefile()
+        if server:
+            self.__sock.bind(self.__address)
+            self.__sock.listen(1)
 
     def __get_sock(self):
         if isinstance(self.__address, tuple):
@@ -43,7 +48,18 @@
             family = socket.AF_UNIX
         return socket.socket(family, socket.SOCK_STREAM)
 
-    def __json_read(self):
+    def __negotiate_capabilities(self):
+        self.__sockfile = self.__sock.makefile()
+        greeting = self.__json_read()
+        if greeting is None or not greeting.has_key('QMP'):
+            raise QMPConnectError
+        # Greeting seems ok, negotiate capabilities
+        resp = self.cmd('qmp_capabilities')
+        if "return" in resp:
+            return greeting
+        raise QMPCapabilitiesError
+
+    def __json_read(self, only_event=False):
         while True:
             data = self.__sockfile.readline()
             if not data:
@@ -51,7 +67,8 @@
             resp = json.loads(data)
             if 'event' in resp:
                 self.__events.append(resp)
-                continue
+                if not only_event:
+                    continue
             return resp
 
     error = socket.error
@@ -66,14 +83,19 @@
         @raise QMPCapabilitiesError if fails to negotiate capabilities
         """
         self.__sock.connect(self.__address)
-        greeting = self.__json_read()
-        if greeting is None or not greeting.has_key('QMP'):
-            raise QMPConnectError
-        # Greeting seems ok, negotiate capabilities
-        resp = self.cmd('qmp_capabilities')
-        if "return" in resp:
-            return greeting
-        raise QMPCapabilitiesError
+        return self.__negotiate_capabilities()
+
+    def accept(self):
+        """
+        Await connection from QMP Monitor and perform capabilities negotiation.
+
+        @return QMP greeting dict
+        @raise socket.error on socket connection errors
+        @raise QMPConnectError if the greeting is not received
+        @raise QMPCapabilitiesError if fails to negotiate capabilities
+        """
+        self.__sock, _ = self.__sock.accept()
+        return self.__negotiate_capabilities()
 
     def cmd_obj(self, qmp_cmd):
         """
@@ -106,9 +128,11 @@
             qmp_cmd['id'] = id
         return self.cmd_obj(qmp_cmd)
 
-    def get_events(self):
+    def get_events(self, wait=False):
         """
         Get a list of available QMP events.
+
+        @param wait: block until an event is available (bool)
         """
         self.__sock.setblocking(0)
         try:
@@ -118,6 +142,8 @@
                 # No data available
                 pass
         self.__sock.setblocking(1)
+        if not self.__events and wait:
+            self.__json_read(only_event=True)
         return self.__events
 
     def clear_events(self):
diff --git a/alpha-dis.c b/alpha-dis.c
index 8a2411e..ae331b3 100644
--- a/alpha-dis.c
+++ b/alpha-dis.c
@@ -238,10 +238,6 @@
 #define AXP_REG_SP	30
 #define AXP_REG_ZERO	31
 
-#define bfd_mach_alpha_ev4  0x10
-#define bfd_mach_alpha_ev5  0x20
-#define bfd_mach_alpha_ev6  0x30
-
 enum bfd_reloc_code_real {
     BFD_RELOC_23_PCREL_S2,
     BFD_RELOC_ALPHA_HINT
diff --git a/async.c b/async.c
index 57ac3a8..fd313df 100644
--- a/async.c
+++ b/async.c
@@ -137,11 +137,12 @@
 
 int qemu_bh_poll(void)
 {
-    QEMUBH *bh, **bhp;
+    QEMUBH *bh, **bhp, *next;
     int ret;
 
     ret = 0;
-    for (bh = async_context->first_bh; bh; bh = bh->next) {
+    for (bh = async_context->first_bh; bh; bh = next) {
+        next = bh->next;
         if (!bh->deleted && bh->scheduled) {
             bh->scheduled = 0;
             if (!bh->idle)
diff --git a/audio/audio_pt_int.c b/audio/audio_pt_int.c
index 908c569..9a9c306 100644
--- a/audio/audio_pt_int.c
+++ b/audio/audio_pt_int.c
@@ -6,8 +6,6 @@
 #include "audio_int.h"
 #include "audio_pt_int.h"
 
-#include <signal.h>
-
 static void GCC_FMT_ATTR(3, 4) logerr (struct audio_pt *pt, int err,
                                        const char *fmt, ...)
 {
diff --git a/audio/mixeng_template.h b/audio/mixeng_template.h
index a2d0ef8..e644c23 100644
--- a/audio/mixeng_template.h
+++ b/audio/mixeng_template.h
@@ -46,7 +46,7 @@
 #endif
 #else  /* !RECIPROCAL */
 #ifdef SIGNED
-    return nv / (mixeng_real) (IN_MAX - IN_MIN);
+    return nv / (mixeng_real) ((mixeng_real) IN_MAX - IN_MIN);
 #else
     return (nv - HALF) / (mixeng_real) IN_MAX;
 #endif
@@ -63,7 +63,7 @@
     }
 
 #ifdef SIGNED
-    return ENDIAN_CONVERT ((IN_T) (v * (IN_MAX - IN_MIN)));
+    return ENDIAN_CONVERT ((IN_T) (v * ((mixeng_real) IN_MAX - IN_MIN)));
 #else
     return ENDIAN_CONVERT ((IN_T) ((v * IN_MAX) + HALF));
 #endif
diff --git a/audio/sdlaudio.c b/audio/sdlaudio.c
index a847aa9..d24daa5 100644
--- a/audio/sdlaudio.c
+++ b/audio/sdlaudio.c
@@ -32,7 +32,6 @@
 #elif defined(__OpenBSD__) || defined(__FreeBSD__) || defined(__DragonFly__)
 #include <pthread.h>
 #endif
-#include <signal.h>
 #endif
 
 #define AUDIO_CAP "sdl"
diff --git a/block.c b/block.c
index effa86f..24a25d5 100644
--- a/block.c
+++ b/block.c
@@ -439,13 +439,7 @@
     bs->drv = drv;
     bs->opaque = qemu_mallocz(drv->instance_size);
 
-    /*
-     * Yes, BDRV_O_NOCACHE aka O_DIRECT means we have to present a
-     * write cache to the guest.  We do need the fdatasync to flush
-     * out transactions for block allocations, and we maybe have a
-     * volatile write cache in our backing device to deal with.
-     */
-    if (flags & (BDRV_O_CACHE_WB|BDRV_O_NOCACHE))
+    if (flags & BDRV_O_CACHE_WB)
         bs->enable_write_cache = 1;
 
     /*
@@ -2887,7 +2881,7 @@
                     char *options, uint64_t img_size, int flags)
 {
     QEMUOptionParameter *param = NULL, *create_options = NULL;
-    QEMUOptionParameter *backing_fmt, *backing_file;
+    QEMUOptionParameter *backing_fmt, *backing_file, *size;
     BlockDriverState *bs = NULL;
     BlockDriver *drv, *proto_drv;
     BlockDriver *backing_drv = NULL;
@@ -2970,7 +2964,8 @@
 
     // The size for the image must always be specified, with one exception:
     // If we are using a backing file, we can obtain the size from there
-    if (get_option_parameter(param, BLOCK_OPT_SIZE)->value.n == -1) {
+    size = get_option_parameter(param, BLOCK_OPT_SIZE);
+    if (size && size->value.n == -1) {
         if (backing_file && backing_file->value.s) {
             uint64_t size;
             char buf[32];
diff --git a/block.h b/block.h
index da7d39c..859d1d9 100644
--- a/block.h
+++ b/block.h
@@ -110,7 +110,7 @@
 typedef struct BlockDriverAIOCB BlockDriverAIOCB;
 typedef void BlockDriverCompletionFunc(void *opaque, int ret);
 typedef void BlockDriverDirtyHandler(BlockDriverState *bs, int64_t sector,
-				     int sector_num);
+                                     int sector_num);
 BlockDriverAIOCB *bdrv_aio_readv(BlockDriverState *bs, int64_t sector_num,
                                  QEMUIOVector *iov, int nb_sectors,
                                  BlockDriverCompletionFunc *cb, void *opaque);
@@ -118,7 +118,7 @@
                                   QEMUIOVector *iov, int nb_sectors,
                                   BlockDriverCompletionFunc *cb, void *opaque);
 BlockDriverAIOCB *bdrv_aio_flush(BlockDriverState *bs,
-				 BlockDriverCompletionFunc *cb, void *opaque);
+                                 BlockDriverCompletionFunc *cb, void *opaque);
 void bdrv_aio_cancel(BlockDriverAIOCB *acb);
 
 typedef struct BlockRequest {
@@ -150,7 +150,7 @@
 int bdrv_discard(BlockDriverState *bs, int64_t sector_num, int nb_sectors);
 int bdrv_has_zero_init(BlockDriverState *bs);
 int bdrv_is_allocated(BlockDriverState *bs, int64_t sector_num, int nb_sectors,
-	int *pnum);
+                      int *pnum);
 
 #define BIOS_ATA_TRANSLATION_AUTO   0
 #define BIOS_ATA_TRANSLATION_NONE   1
diff --git a/block/qcow.c b/block/qcow.c
index a26c886..227b104 100644
--- a/block/qcow.c
+++ b/block/qcow.c
@@ -496,6 +496,8 @@
     uint64_t cluster_offset;
     uint8_t *cluster_data;
     struct iovec hd_iov;
+    bool is_write;
+    QEMUBH *bh;
     QEMUIOVector hd_qiov;
     BlockDriverAIOCB *hd_aiocb;
 } QCowAIOCB;
@@ -525,6 +527,8 @@
     acb->hd_aiocb = NULL;
     acb->sector_num = sector_num;
     acb->qiov = qiov;
+    acb->is_write = is_write;
+
     if (qiov->niov > 1) {
         acb->buf = acb->orig_buf = qemu_blockalign(bs, qiov->size);
         if (is_write)
@@ -538,6 +542,38 @@
     return acb;
 }
 
+static void qcow_aio_read_cb(void *opaque, int ret);
+static void qcow_aio_write_cb(void *opaque, int ret);
+
+static void qcow_aio_rw_bh(void *opaque)
+{
+    QCowAIOCB *acb = opaque;
+    qemu_bh_delete(acb->bh);
+    acb->bh = NULL;
+
+    if (acb->is_write) {
+        qcow_aio_write_cb(opaque, 0);
+    } else {
+        qcow_aio_read_cb(opaque, 0);
+    }
+}
+
+static int qcow_schedule_bh(QEMUBHFunc *cb, QCowAIOCB *acb)
+{
+    if (acb->bh) {
+        return -EIO;
+    }
+
+    acb->bh = qemu_bh_new(cb, acb);
+    if (!acb->bh) {
+        return -EIO;
+    }
+
+    qemu_bh_schedule(acb->bh);
+
+    return 0;
+}
+
 static void qcow_aio_read_cb(void *opaque, int ret)
 {
     QCowAIOCB *acb = opaque;
@@ -640,12 +676,21 @@
         BlockDriverCompletionFunc *cb, void *opaque)
 {
     QCowAIOCB *acb;
+    int ret;
 
     acb = qcow_aio_setup(bs, sector_num, qiov, nb_sectors, cb, opaque, 0);
     if (!acb)
         return NULL;
 
-    qcow_aio_read_cb(acb, 0);
+    ret = qcow_schedule_bh(qcow_aio_rw_bh, acb);
+    if (ret < 0) {
+        if (acb->qiov->niov > 1) {
+            qemu_vfree(acb->orig_buf);
+        }
+        qemu_aio_release(acb);
+        return NULL;
+    }
+
     return &acb->common;
 }
 
@@ -725,6 +770,7 @@
 {
     BDRVQcowState *s = bs->opaque;
     QCowAIOCB *acb;
+    int ret;
 
     s->cluster_cache_offset = -1; /* disable compressed cache */
 
@@ -733,7 +779,15 @@
         return NULL;
 
 
-    qcow_aio_write_cb(acb, 0);
+    ret = qcow_schedule_bh(qcow_aio_rw_bh, acb);
+    if (ret < 0) {
+        if (acb->qiov->niov > 1) {
+            qemu_vfree(acb->orig_buf);
+        }
+        qemu_aio_release(acb);
+        return NULL;
+    }
+
     return &acb->common;
 }
 
diff --git a/block/qcow2-cluster.c b/block/qcow2-cluster.c
index 750abe3..882f50a 100644
--- a/block/qcow2-cluster.c
+++ b/block/qcow2-cluster.c
@@ -70,7 +70,7 @@
 
     ret = qcow2_cache_flush(bs, s->refcount_block_cache);
     if (ret < 0) {
-        return ret;
+        goto fail;
     }
 
     BLKDBG_EVENT(bs->file, BLKDBG_L1_GROW_WRITE_TABLE);
@@ -796,8 +796,8 @@
                 m->depends_on = old_alloc;
                 m->nb_clusters = 0;
                 *num = 0;
-                ret = 0;
-                goto fail;
+
+                goto out_wait_dependency;
             }
         }
     }
@@ -812,7 +812,6 @@
 
     cluster_offset = qcow2_alloc_clusters(bs, nb_clusters * s->cluster_size);
     if (cluster_offset < 0) {
-        QLIST_REMOVE(m, next_in_flight);
         ret = cluster_offset;
         goto fail;
     }
@@ -825,7 +824,7 @@
 out:
     ret = qcow2_cache_put(bs, s->l2_table_cache, (void**) &l2_table);
     if (ret < 0) {
-        return ret;
+        goto fail_put;
     }
 
     m->nb_available = MIN(nb_clusters << (s->cluster_bits - 9), n_end);
@@ -835,8 +834,13 @@
 
     return 0;
 
+out_wait_dependency:
+    return qcow2_cache_put(bs, s->l2_table_cache, (void**) &l2_table);
+
 fail:
     qcow2_cache_put(bs, s->l2_table_cache, (void**) &l2_table);
+fail_put:
+    QLIST_REMOVE(m, next_in_flight);
     return ret;
 }
 
diff --git a/block/qcow2-refcount.c b/block/qcow2-refcount.c
index d62dc1c..ac95b88 100644
--- a/block/qcow2-refcount.c
+++ b/block/qcow2-refcount.c
@@ -1086,7 +1086,7 @@
     ret = check_refcounts_l1(bs, res, refcount_table, nb_clusters,
                        s->l1_table_offset, s->l1_size, 1);
     if (ret < 0) {
-        return ret;
+        goto fail;
     }
 
     /* snapshots */
@@ -1095,7 +1095,7 @@
         ret = check_refcounts_l1(bs, res, refcount_table, nb_clusters,
             sn->l1_table_offset, sn->l1_size, 0);
         if (ret < 0) {
-            return ret;
+            goto fail;
         }
     }
     inc_refcounts(bs, res, refcount_table, nb_clusters,
@@ -1159,8 +1159,11 @@
         }
     }
 
+    ret = 0;
+
+fail:
     qemu_free(refcount_table);
 
-    return 0;
+    return ret;
 }
 
diff --git a/block/qcow2.c b/block/qcow2.c
index 75b8bec..2c51e7c 100644
--- a/block/qcow2.c
+++ b/block/qcow2.c
@@ -229,7 +229,7 @@
     }
 
     /* alloc L2 table/refcount block cache */
-    writethrough = ((flags & BDRV_O_CACHE_MASK) == 0);
+    writethrough = ((flags & BDRV_O_CACHE_WB) == 0);
     s->l2_table_cache = qcow2_cache_create(bs, L2_CACHE_SIZE, writethrough);
     s->refcount_block_cache = qcow2_cache_create(bs, REFCOUNT_CACHE_SIZE,
         writethrough);
@@ -378,6 +378,7 @@
     uint64_t bytes_done;
     uint64_t cluster_offset;
     uint8_t *cluster_data;
+    bool is_write;
     BlockDriverAIOCB *hd_aiocb;
     QEMUIOVector hd_qiov;
     QEMUBH *bh;
@@ -399,12 +400,19 @@
 };
 
 static void qcow2_aio_read_cb(void *opaque, int ret);
-static void qcow2_aio_read_bh(void *opaque)
+static void qcow2_aio_write_cb(void *opaque, int ret);
+
+static void qcow2_aio_rw_bh(void *opaque)
 {
     QCowAIOCB *acb = opaque;
     qemu_bh_delete(acb->bh);
     acb->bh = NULL;
-    qcow2_aio_read_cb(opaque, 0);
+
+    if (acb->is_write) {
+        qcow2_aio_write_cb(opaque, 0);
+    } else {
+        qcow2_aio_read_cb(opaque, 0);
+    }
 }
 
 static int qcow2_schedule_bh(QEMUBHFunc *cb, QCowAIOCB *acb)
@@ -493,14 +501,14 @@
                     goto done;
                 }
             } else {
-                ret = qcow2_schedule_bh(qcow2_aio_read_bh, acb);
+                ret = qcow2_schedule_bh(qcow2_aio_rw_bh, acb);
                 if (ret < 0)
                     goto done;
             }
         } else {
             /* Note: in this case, no need to wait */
             qemu_iovec_memset(&acb->hd_qiov, 0, 512 * acb->cur_nr_sectors);
-            ret = qcow2_schedule_bh(qcow2_aio_read_bh, acb);
+            ret = qcow2_schedule_bh(qcow2_aio_rw_bh, acb);
             if (ret < 0)
                 goto done;
         }
@@ -515,7 +523,7 @@
             s->cluster_cache + index_in_cluster * 512,
             512 * acb->cur_nr_sectors);
 
-        ret = qcow2_schedule_bh(qcow2_aio_read_bh, acb);
+        ret = qcow2_schedule_bh(qcow2_aio_rw_bh, acb);
         if (ret < 0)
             goto done;
     } else {
@@ -572,6 +580,7 @@
     acb->hd_aiocb = NULL;
     acb->sector_num = sector_num;
     acb->qiov = qiov;
+    acb->is_write = is_write;
 
     qemu_iovec_init(&acb->hd_qiov, qiov->niov);
 
@@ -591,17 +600,22 @@
                                          void *opaque)
 {
     QCowAIOCB *acb;
+    int ret;
 
     acb = qcow2_aio_setup(bs, sector_num, qiov, nb_sectors, cb, opaque, 0);
     if (!acb)
         return NULL;
 
-    qcow2_aio_read_cb(acb, 0);
+    ret = qcow2_schedule_bh(qcow2_aio_rw_bh, acb);
+    if (ret < 0) {
+        qemu_iovec_destroy(&acb->hd_qiov);
+        qemu_aio_release(acb);
+        return NULL;
+    }
+
     return &acb->common;
 }
 
-static void qcow2_aio_write_cb(void *opaque, int ret);
-
 static void run_dependent_requests(QCowL2Meta *m)
 {
     QCowAIOCB *req;
@@ -724,6 +738,7 @@
 {
     BDRVQcowState *s = bs->opaque;
     QCowAIOCB *acb;
+    int ret;
 
     s->cluster_cache_offset = -1; /* disable compressed cache */
 
@@ -731,7 +746,13 @@
     if (!acb)
         return NULL;
 
-    qcow2_aio_write_cb(acb, 0);
+    ret = qcow2_schedule_bh(qcow2_aio_rw_bh, acb);
+    if (ret < 0) {
+        qemu_iovec_destroy(&acb->hd_qiov);
+        qemu_aio_release(acb);
+        return NULL;
+    }
+
     return &acb->common;
 }
 
@@ -1036,7 +1057,7 @@
     const char *backing_fmt = NULL;
     uint64_t sectors = 0;
     int flags = 0;
-    size_t cluster_size = 65536;
+    size_t cluster_size = DEFAULT_CLUSTER_SIZE;
     int prealloc = 0;
 
     /* Read out options */
@@ -1343,7 +1364,8 @@
     {
         .name = BLOCK_OPT_CLUSTER_SIZE,
         .type = OPT_SIZE,
-        .help = "qcow2 cluster size"
+        .help = "qcow2 cluster size",
+        .value = { .n = DEFAULT_CLUSTER_SIZE },
     },
     {
         .name = BLOCK_OPT_PREALLOC,
diff --git a/block/qcow2.h b/block/qcow2.h
index a019831..e1ae3e8 100644
--- a/block/qcow2.h
+++ b/block/qcow2.h
@@ -54,6 +54,8 @@
 /* Must be at least 4 to cover all cases of refcount table growth */
 #define REFCOUNT_CACHE_SIZE 4
 
+#define DEFAULT_CLUSTER_SIZE 65536
+
 typedef struct QCowHeader {
     uint32_t magic;
     uint32_t version;
diff --git a/block/qed.c b/block/qed.c
index da0bf31..3970379 100644
--- a/block/qed.c
+++ b/block/qed.c
@@ -1464,7 +1464,8 @@
     }, {
         .name = BLOCK_OPT_CLUSTER_SIZE,
         .type = OPT_SIZE,
-        .help = "Cluster size (in bytes)"
+        .help = "Cluster size (in bytes)",
+        .value = { .n = QED_DEFAULT_CLUSTER_SIZE },
     }, {
         .name = BLOCK_OPT_TABLE_SIZE,
         .type = OPT_SIZE,
diff --git a/block/raw-posix.c b/block/raw-posix.c
index 6b72470..4cd7d7a 100644
--- a/block/raw-posix.c
+++ b/block/raw-posix.c
@@ -43,7 +43,6 @@
 
 #ifdef __sun__
 #define _POSIX_PTHREAD_SEMANTICS 1
-#include <signal.h>
 #include <sys/dkio.h>
 #endif
 #ifdef __linux__
@@ -53,7 +52,6 @@
 #include <linux/fd.h>
 #endif
 #if defined (__FreeBSD__) || defined(__FreeBSD_kernel__)
-#include <signal.h>
 #include <sys/disk.h>
 #include <sys/cdio.h>
 #endif
@@ -64,6 +62,13 @@
 #include <sys/dkio.h>
 #endif
 
+#ifdef __NetBSD__
+#include <sys/ioctl.h>
+#include <sys/disklabel.h>
+#include <sys/dkio.h>
+#include <sys/disk.h>
+#endif
+
 #ifdef __DragonFly__
 #include <sys/ioctl.h>
 #include <sys/diskslice.h>
@@ -136,12 +141,55 @@
 static int cdrom_reopen(BlockDriverState *bs);
 #endif
 
+#if defined(__NetBSD__)
+static int raw_normalize_devicepath(const char **filename)
+{
+    static char namebuf[PATH_MAX];
+    const char *dp, *fname;
+    struct stat sb;
+
+    fname = *filename;
+    dp = strrchr(fname, '/');
+    if (lstat(fname, &sb) < 0) {
+        fprintf(stderr, "%s: stat failed: %s\n",
+            fname, strerror(errno));
+        return -errno;
+    }
+
+    if (!S_ISBLK(sb.st_mode)) {
+        return 0;
+    }
+
+    if (dp == NULL) {
+        snprintf(namebuf, PATH_MAX, "r%s", fname);
+    } else {
+        snprintf(namebuf, PATH_MAX, "%.*s/r%s",
+            (int)(dp - fname), fname, dp + 1);
+    }
+    fprintf(stderr, "%s is a block device", fname);
+    *filename = namebuf;
+    fprintf(stderr, ", using %s\n", *filename);
+
+    return 0;
+}
+#else
+static int raw_normalize_devicepath(const char **filename)
+{
+    return 0;
+}
+#endif
+
 static int raw_open_common(BlockDriverState *bs, const char *filename,
                            int bdrv_flags, int open_flags)
 {
     BDRVRawState *s = bs->opaque;
     int fd, ret;
 
+    ret = raw_normalize_devicepath(&filename);
+    if (ret != 0) {
+        return ret;
+    }
+
     s->open_flags = open_flags | O_BINARY;
     s->open_flags &= ~O_ACCMODE;
     if (bdrv_flags & BDRV_O_RDWR) {
@@ -154,7 +202,7 @@
      * and O_DIRECT for no caching. */
     if ((bdrv_flags & BDRV_O_NOCACHE))
         s->open_flags |= O_DIRECT;
-    else if (!(bdrv_flags & BDRV_O_CACHE_WB))
+    if (!(bdrv_flags & BDRV_O_CACHE_WB))
         s->open_flags |= O_DSYNC;
 
     s->fd = -1;
@@ -622,6 +670,31 @@
     } else
         return st.st_size;
 }
+#elif defined(__NetBSD__)
+static int64_t raw_getlength(BlockDriverState *bs)
+{
+    BDRVRawState *s = bs->opaque;
+    int fd = s->fd;
+    struct stat st;
+
+    if (fstat(fd, &st))
+        return -1;
+    if (S_ISCHR(st.st_mode) || S_ISBLK(st.st_mode)) {
+        struct dkwedge_info dkw;
+
+        if (ioctl(fd, DIOCGWEDGEINFO, &dkw) != -1) {
+            return dkw.dkw_size * 512;
+        } else {
+            struct disklabel dl;
+
+            if (ioctl(fd, DIOCGDINFO, &dl))
+                return -1;
+            return (uint64_t)dl.d_secsize *
+                dl.d_partitions[DISKPART(st.st_rdev)].p_size;
+        }
+    } else
+        return st.st_size;
+}
 #elif defined(__sun__)
 static int64_t raw_getlength(BlockDriverState *bs)
 {
diff --git a/block/raw-win32.c b/block/raw-win32.c
index c204a80..56bd719 100644
--- a/block/raw-win32.c
+++ b/block/raw-win32.c
@@ -88,9 +88,9 @@
     }
 
     overlapped = FILE_ATTRIBUTE_NORMAL;
-    if ((flags & BDRV_O_NOCACHE))
-        overlapped |= FILE_FLAG_NO_BUFFERING | FILE_FLAG_WRITE_THROUGH;
-    else if (!(flags & BDRV_O_CACHE_WB))
+    if (flags & BDRV_O_NOCACHE)
+        overlapped |= FILE_FLAG_NO_BUFFERING;
+    if (!(flags & BDRV_O_CACHE_WB))
         overlapped |= FILE_FLAG_WRITE_THROUGH;
     s->hfile = CreateFile(filename, access_flags,
                           FILE_SHARE_READ, NULL,
@@ -349,9 +349,9 @@
     create_flags = OPEN_EXISTING;
 
     overlapped = FILE_ATTRIBUTE_NORMAL;
-    if ((flags & BDRV_O_NOCACHE))
-        overlapped |= FILE_FLAG_NO_BUFFERING | FILE_FLAG_WRITE_THROUGH;
-    else if (!(flags & BDRV_O_CACHE_WB))
+    if (flags & BDRV_O_NOCACHE)
+        overlapped |= FILE_FLAG_NO_BUFFERING;
+    if (!(flags & BDRV_O_CACHE_WB))
         overlapped |= FILE_FLAG_WRITE_THROUGH;
     s->hfile = CreateFile(filename, access_flags,
                           FILE_SHARE_READ, NULL,
diff --git a/block/rbd.c b/block/rbd.c
index 249a590..d5659cd 100644
--- a/block/rbd.c
+++ b/block/rbd.c
@@ -1,33 +1,39 @@
 /*
  * QEMU Block driver for RADOS (Ceph)
  *
- * Copyright (C) 2010 Christian Brunner <chb@muc.de>
+ * Copyright (C) 2010-2011 Christian Brunner <chb@muc.de>,
+ *                         Josh Durgin <josh.durgin@dreamhost.com>
  *
  * This work is licensed under the terms of the GNU GPL, version 2.  See
  * the COPYING file in the top-level directory.
  *
  */
 
+#include <inttypes.h>
+
 #include "qemu-common.h"
 #include "qemu-error.h"
 
-#include "rbd_types.h"
 #include "block_int.h"
 
-#include <rados/librados.h>
+#include <rbd/librbd.h>
 
 
 
 /*
  * When specifying the image filename use:
  *
- * rbd:poolname/devicename
+ * rbd:poolname/devicename[@snapshotname][:option1=value1[:option2=value2...]]
  *
  * poolname must be the name of an existing rados pool
  *
  * devicename is the basename for all objects used to
  * emulate the raw device.
  *
+ * Each option given is used to configure rados, and may be
+ * any Ceph option, or "conf". The "conf" option specifies
+ * a Ceph configuration file to read.
+ *
  * Metadata information (image size, ...) is stored in an
  * object with the name "devicename.rbd".
  *
@@ -40,6 +46,13 @@
 
 #define OBJ_MAX_SIZE (1UL << OBJ_DEFAULT_OBJ_ORDER)
 
+#define RBD_MAX_CONF_NAME_SIZE 128
+#define RBD_MAX_CONF_VAL_SIZE 512
+#define RBD_MAX_CONF_SIZE 1024
+#define RBD_MAX_POOL_NAME_SIZE 128
+#define RBD_MAX_SNAP_NAME_SIZE 128
+#define RBD_MAX_SNAPS 100
+
 typedef struct RBDAIOCB {
     BlockDriverAIOCB common;
     QEMUBH *bh;
@@ -48,7 +61,6 @@
     char *bounce;
     int write;
     int64_t sector_num;
-    int aiocnt;
     int error;
     struct BDRVRBDState *s;
     int cancelled;
@@ -59,7 +71,7 @@
     RBDAIOCB *acb;
     struct BDRVRBDState *s;
     int done;
-    int64_t segsize;
+    int64_t size;
     char *buf;
     int ret;
 } RADOSCB;
@@ -69,25 +81,22 @@
 
 typedef struct BDRVRBDState {
     int fds[2];
-    rados_pool_t pool;
-    rados_pool_t header_pool;
-    char name[RBD_MAX_OBJ_NAME_SIZE];
-    char block_name[RBD_MAX_BLOCK_NAME_SIZE];
-    uint64_t size;
-    uint64_t objsize;
+    rados_t cluster;
+    rados_ioctx_t io_ctx;
+    rbd_image_t image;
+    char name[RBD_MAX_IMAGE_NAME_SIZE];
     int qemu_aio_count;
+    char *snap;
     int event_reader_pos;
     RADOSCB *event_rcb;
 } BDRVRBDState;
 
-typedef struct rbd_obj_header_ondisk RbdHeader1;
-
 static void rbd_aio_bh_cb(void *opaque);
 
-static int rbd_next_tok(char *dst, int dst_len,
-                        char *src, char delim,
-                        const char *name,
-                        char **p)
+static int qemu_rbd_next_tok(char *dst, int dst_len,
+                             char *src, char delim,
+                             const char *name,
+                             char **p)
 {
     int l;
     char *end;
@@ -115,10 +124,11 @@
     return 0;
 }
 
-static int rbd_parsename(const char *filename,
-                         char *pool, int pool_len,
-                         char *snap, int snap_len,
-                         char *name, int name_len)
+static int qemu_rbd_parsename(const char *filename,
+                              char *pool, int pool_len,
+                              char *snap, int snap_len,
+                              char *name, int name_len,
+                              char *conf, int conf_len)
 {
     const char *start;
     char *p, *buf;
@@ -130,137 +140,103 @@
 
     buf = qemu_strdup(start);
     p = buf;
+    *snap = '\0';
+    *conf = '\0';
 
-    ret = rbd_next_tok(pool, pool_len, p, '/', "pool name", &p);
+    ret = qemu_rbd_next_tok(pool, pool_len, p, '/', "pool name", &p);
     if (ret < 0 || !p) {
         ret = -EINVAL;
         goto done;
     }
-    ret = rbd_next_tok(name, name_len, p, '@', "object name", &p);
-    if (ret < 0) {
-        goto done;
+
+    if (strchr(p, '@')) {
+        ret = qemu_rbd_next_tok(name, name_len, p, '@', "object name", &p);
+        if (ret < 0) {
+            goto done;
+        }
+        ret = qemu_rbd_next_tok(snap, snap_len, p, ':', "snap name", &p);
+    } else {
+        ret = qemu_rbd_next_tok(name, name_len, p, ':', "object name", &p);
     }
-    if (!p) {
-        *snap = '\0';
+    if (ret < 0 || !p) {
         goto done;
     }
 
-    ret = rbd_next_tok(snap, snap_len, p, '\0', "snap name", &p);
+    ret = qemu_rbd_next_tok(conf, conf_len, p, '\0', "configuration", &p);
 
 done:
     qemu_free(buf);
     return ret;
 }
 
-static int create_tmap_op(uint8_t op, const char *name, char **tmap_desc)
+static int qemu_rbd_set_conf(rados_t cluster, const char *conf)
 {
-    uint32_t len = strlen(name);
-    uint32_t len_le = cpu_to_le32(len);
-    /* total_len = encoding op + name + empty buffer */
-    uint32_t total_len = 1 + (sizeof(uint32_t) + len) + sizeof(uint32_t);
-    uint8_t *desc = NULL;
+    char *p, *buf;
+    char name[RBD_MAX_CONF_NAME_SIZE];
+    char value[RBD_MAX_CONF_VAL_SIZE];
+    int ret = 0;
 
-    desc = qemu_malloc(total_len);
+    buf = qemu_strdup(conf);
+    p = buf;
 
-    *tmap_desc = (char *)desc;
+    while (p) {
+        ret = qemu_rbd_next_tok(name, sizeof(name), p,
+                                '=', "conf option name", &p);
+        if (ret < 0) {
+            break;
+        }
 
-    *desc = op;
-    desc++;
-    memcpy(desc, &len_le, sizeof(len_le));
-    desc += sizeof(len_le);
-    memcpy(desc, name, len);
-    desc += len;
-    len = 0; /* no need for endian conversion for 0 */
-    memcpy(desc, &len, sizeof(len));
-    desc += sizeof(len);
+        if (!p) {
+            error_report("conf option %s has no value", name);
+            ret = -EINVAL;
+            break;
+        }
 
-    return (char *)desc - *tmap_desc;
-}
+        ret = qemu_rbd_next_tok(value, sizeof(value), p,
+                                ':', "conf option value", &p);
+        if (ret < 0) {
+            break;
+        }
 
-static void free_tmap_op(char *tmap_desc)
-{
-    qemu_free(tmap_desc);
-}
-
-static int rbd_register_image(rados_pool_t pool, const char *name)
-{
-    char *tmap_desc;
-    const char *dir = RBD_DIRECTORY;
-    int ret;
-
-    ret = create_tmap_op(CEPH_OSD_TMAP_SET, name, &tmap_desc);
-    if (ret < 0) {
-        return ret;
+        if (strcmp(name, "conf")) {
+            ret = rados_conf_set(cluster, name, value);
+            if (ret < 0) {
+                error_report("invalid conf option %s", name);
+                ret = -EINVAL;
+                break;
+            }
+        } else {
+            ret = rados_conf_read_file(cluster, value);
+            if (ret < 0) {
+                error_report("error reading conf file %s", value);
+                break;
+            }
+        }
     }
 
-    ret = rados_tmap_update(pool, dir, tmap_desc, ret);
-    free_tmap_op(tmap_desc);
-
+    qemu_free(buf);
     return ret;
 }
 
-static int touch_rbd_info(rados_pool_t pool, const char *info_oid)
-{
-    int r = rados_write(pool, info_oid, 0, NULL, 0);
-    if (r < 0) {
-        return r;
-    }
-    return 0;
-}
-
-static int rbd_assign_bid(rados_pool_t pool, uint64_t *id)
-{
-    uint64_t out[1];
-    const char *info_oid = RBD_INFO;
-
-    *id = 0;
-
-    int r = touch_rbd_info(pool, info_oid);
-    if (r < 0) {
-        return r;
-    }
-
-    r = rados_exec(pool, info_oid, "rbd", "assign_bid", NULL,
-                   0, (char *)out, sizeof(out));
-    if (r < 0) {
-        return r;
-    }
-
-    le64_to_cpus(out);
-    *id = out[0];
-
-    return 0;
-}
-
-static int rbd_create(const char *filename, QEMUOptionParameter *options)
+static int qemu_rbd_create(const char *filename, QEMUOptionParameter *options)
 {
     int64_t bytes = 0;
     int64_t objsize;
-    uint64_t size;
-    time_t mtime;
-    uint8_t obj_order = RBD_DEFAULT_OBJ_ORDER;
-    char pool[RBD_MAX_SEG_NAME_SIZE];
-    char n[RBD_MAX_SEG_NAME_SIZE];
-    char name[RBD_MAX_OBJ_NAME_SIZE];
-    char snap_buf[RBD_MAX_SEG_NAME_SIZE];
-    char *snap = NULL;
-    RbdHeader1 header;
-    rados_pool_t p;
-    uint64_t bid;
-    uint32_t hi, lo;
+    int obj_order = 0;
+    char pool[RBD_MAX_POOL_NAME_SIZE];
+    char name[RBD_MAX_IMAGE_NAME_SIZE];
+    char snap_buf[RBD_MAX_SNAP_NAME_SIZE];
+    char conf[RBD_MAX_CONF_SIZE];
+    rados_t cluster;
+    rados_ioctx_t io_ctx;
     int ret;
 
-    if (rbd_parsename(filename,
-                      pool, sizeof(pool),
-                      snap_buf, sizeof(snap_buf),
-                      name, sizeof(name)) < 0) {
+    if (qemu_rbd_parsename(filename, pool, sizeof(pool),
+                           snap_buf, sizeof(snap_buf),
+                           name, sizeof(name),
+                           conf, sizeof(conf)) < 0) {
         return -EINVAL;
     }
-    if (snap_buf[0] != '\0') {
-        snap = snap_buf;
-    }
-
-    snprintf(n, sizeof(n), "%s%s", name, RBD_SUFFIX);
 
     /* Read out options */
     while (options && options->name) {
@@ -277,82 +253,64 @@
                     error_report("obj size too small");
                     return -EINVAL;
                 }
-		obj_order = ffs(objsize) - 1;
+                obj_order = ffs(objsize) - 1;
             }
         }
         options++;
     }
 
-    memset(&header, 0, sizeof(header));
-    pstrcpy(header.text, sizeof(header.text), RBD_HEADER_TEXT);
-    pstrcpy(header.signature, sizeof(header.signature), RBD_HEADER_SIGNATURE);
-    pstrcpy(header.version, sizeof(header.version), RBD_HEADER_VERSION);
-    header.image_size = cpu_to_le64(bytes);
-    header.options.order = obj_order;
-    header.options.crypt_type = RBD_CRYPT_NONE;
-    header.options.comp_type = RBD_COMP_NONE;
-    header.snap_seq = 0;
-    header.snap_count = 0;
-
-    if (rados_initialize(0, NULL) < 0) {
+    if (rados_create(&cluster, NULL) < 0) {
         error_report("error initializing");
         return -EIO;
     }
 
-    if (rados_open_pool(pool, &p)) {
+    if (strstr(conf, "conf=") == NULL) {
+        if (rados_conf_read_file(cluster, NULL) < 0) {
+            error_report("error reading config file");
+            rados_shutdown(cluster);
+            return -EIO;
+        }
+    }
+
+    if (conf[0] != '\0' &&
+        qemu_rbd_set_conf(cluster, conf) < 0) {
+        error_report("error setting config options");
+        rados_shutdown(cluster);
+        return -EIO;
+    }
+
+    if (rados_connect(cluster) < 0) {
+        error_report("error connecting");
+        rados_shutdown(cluster);
+        return -EIO;
+    }
+
+    if (rados_ioctx_create(cluster, pool, &io_ctx) < 0) {
         error_report("error opening pool %s", pool);
-        rados_deinitialize();
+        rados_shutdown(cluster);
         return -EIO;
     }
 
-    /* check for existing rbd header file */
-    ret = rados_stat(p, n, &size, &mtime);
-    if (ret == 0) {
-        ret=-EEXIST;
-        goto done;
-    }
-
-    ret = rbd_assign_bid(p, &bid);
-    if (ret < 0) {
-        error_report("failed assigning block id");
-        rados_deinitialize();
-        return -EIO;
-    }
-    hi = bid >> 32;
-    lo = bid & 0xFFFFFFFF;
-    snprintf(header.block_name, sizeof(header.block_name), "rb.%x.%x", hi, lo);
-
-    /* create header file */
-    ret = rados_write(p, n, 0, (const char *)&header, sizeof(header));
-    if (ret < 0) {
-        goto done;
-    }
-
-    ret = rbd_register_image(p, name);
-done:
-    rados_close_pool(p);
-    rados_deinitialize();
+    ret = rbd_create(io_ctx, name, bytes, &obj_order);
+    rados_ioctx_destroy(io_ctx);
+    rados_shutdown(cluster);
 
     return ret;
 }
 
 /*
- * This aio completion is being called from rbd_aio_event_reader() and
- * runs in qemu context. It schedules a bh, but just in case the aio
+ * This aio completion is being called from qemu_rbd_aio_event_reader()
+ * and runs in qemu context. It schedules a bh, but just in case the aio
  * was not cancelled before.
  */
-static void rbd_complete_aio(RADOSCB *rcb)
+static void qemu_rbd_complete_aio(RADOSCB *rcb)
 {
     RBDAIOCB *acb = rcb->acb;
     int64_t r;
 
-    acb->aiocnt--;
-
     if (acb->cancelled) {
-        if (!acb->aiocnt) {
-            qemu_vfree(acb->bounce);
-            qemu_aio_release(acb);
-        }
+        qemu_vfree(acb->bounce);
+        qemu_aio_release(acb);
         goto done;
     }
 
@@ -363,32 +321,25 @@
             acb->ret = r;
             acb->error = 1;
         } else if (!acb->error) {
-            acb->ret += rcb->segsize;
+            acb->ret = rcb->size;
         }
     } else {
-        if (r == -ENOENT) {
-            memset(rcb->buf, 0, rcb->segsize);
-            if (!acb->error) {
-                acb->ret += rcb->segsize;
-            }
-        } else if (r < 0) {
-	    memset(rcb->buf, 0, rcb->segsize);
+        if (r < 0) {
+            memset(rcb->buf, 0, rcb->size);
             acb->ret = r;
             acb->error = 1;
-        } else if (r < rcb->segsize) {
-            memset(rcb->buf + r, 0, rcb->segsize - r);
+        } else if (r < rcb->size) {
+            memset(rcb->buf + r, 0, rcb->size - r);
             if (!acb->error) {
-                acb->ret += rcb->segsize;
+                acb->ret = rcb->size;
             }
         } else if (!acb->error) {
-            acb->ret += r;
+            acb->ret = r;
         }
     }
     /* Note that acb->bh can be NULL in case where the aio was cancelled */
-    if (!acb->aiocnt) {
-        acb->bh = qemu_bh_new(rbd_aio_bh_cb, acb);
-        qemu_bh_schedule(acb->bh);
-    }
+    acb->bh = qemu_bh_new(rbd_aio_bh_cb, acb);
+    qemu_bh_schedule(acb->bh);
 done:
     qemu_free(rcb);
 }
@@ -397,7 +348,7 @@
  * aio fd read handler. It runs in the qemu context and calls the
  * completion handling of completed rados aio operations.
  */
-static void rbd_aio_event_reader(void *opaque)
+static void qemu_rbd_aio_event_reader(void *opaque)
 {
     BDRVRBDState *s = opaque;
 
@@ -413,176 +364,87 @@
                 s->event_reader_pos += ret;
                 if (s->event_reader_pos == sizeof(s->event_rcb)) {
                     s->event_reader_pos = 0;
-                    rbd_complete_aio(s->event_rcb);
-                    s->qemu_aio_count --;
+                    qemu_rbd_complete_aio(s->event_rcb);
+                    s->qemu_aio_count--;
                 }
             }
         }
     } while (ret < 0 && errno == EINTR);
 }
 
-static int rbd_aio_flush_cb(void *opaque)
+static int qemu_rbd_aio_flush_cb(void *opaque)
 {
     BDRVRBDState *s = opaque;
 
     return (s->qemu_aio_count > 0);
 }
 
-
-static int rbd_set_snapc(rados_pool_t pool, const char *snap, RbdHeader1 *header)
-{
-    uint32_t snap_count = le32_to_cpu(header->snap_count);
-    rados_snap_t *snaps = NULL;
-    rados_snap_t seq;
-    uint32_t i;
-    uint64_t snap_names_len = le64_to_cpu(header->snap_names_len);
-    int r;
-    rados_snap_t snapid = 0;
-
-    if (snap_count) {
-        const char *header_snap = (const char *)&header->snaps[snap_count];
-        const char *end = header_snap + snap_names_len;
-        snaps = qemu_malloc(sizeof(rados_snap_t) * header->snap_count);
-
-        for (i=0; i < snap_count; i++) {
-            snaps[i] = le64_to_cpu(header->snaps[i].id);
-
-            if (snap && strcmp(snap, header_snap) == 0) {
-                snapid = snaps[i];
-            }
-
-            header_snap += strlen(header_snap) + 1;
-            if (header_snap > end) {
-                error_report("bad header, snapshot list broken");
-            }
-        }
-    }
-
-    if (snap && !snapid) {
-        error_report("snapshot not found");
-        qemu_free(snaps);
-        return -ENOENT;
-    }
-    seq = le32_to_cpu(header->snap_seq);
-
-    r = rados_set_snap_context(pool, seq, snaps, snap_count);
-
-    rados_set_snap(pool, snapid);
-
-    qemu_free(snaps);
-
-    return r;
-}
-
-#define BUF_READ_START_LEN    4096
-
-static int rbd_read_header(BDRVRBDState *s, char **hbuf)
-{
-    char *buf = NULL;
-    char n[RBD_MAX_SEG_NAME_SIZE];
-    uint64_t len = BUF_READ_START_LEN;
-    int r;
-
-    snprintf(n, sizeof(n), "%s%s", s->name, RBD_SUFFIX);
-
-    buf = qemu_malloc(len);
-
-    r = rados_read(s->header_pool, n, 0, buf, len);
-    if (r < 0) {
-        goto failed;
-    }
-
-    if (r < len) {
-        goto done;
-    }
-
-    qemu_free(buf);
-    buf = qemu_malloc(len);
-
-    r = rados_stat(s->header_pool, n, &len, NULL);
-    if (r < 0) {
-        goto failed;
-    }
-
-    r = rados_read(s->header_pool, n, 0, buf, len);
-    if (r < 0) {
-        goto failed;
-    }
-
-done:
-    *hbuf = buf;
-    return 0;
-
-failed:
-    qemu_free(buf);
-    return r;
-}
-
-static int rbd_open(BlockDriverState *bs, const char *filename, int flags)
+static int qemu_rbd_open(BlockDriverState *bs, const char *filename, int flags)
 {
     BDRVRBDState *s = bs->opaque;
-    RbdHeader1 *header;
-    char pool[RBD_MAX_SEG_NAME_SIZE];
-    char snap_buf[RBD_MAX_SEG_NAME_SIZE];
-    char *snap = NULL;
-    char *hbuf = NULL;
+    char pool[RBD_MAX_POOL_NAME_SIZE];
+    char snap_buf[RBD_MAX_SNAP_NAME_SIZE];
+    char conf[RBD_MAX_CONF_SIZE];
     int r;
 
-    if (rbd_parsename(filename, pool, sizeof(pool),
-                      snap_buf, sizeof(snap_buf),
-                      s->name, sizeof(s->name)) < 0) {
+    if (qemu_rbd_parsename(filename, pool, sizeof(pool),
+                           snap_buf, sizeof(snap_buf),
+                           s->name, sizeof(s->name),
+                           conf, sizeof(conf)) < 0) {
         return -EINVAL;
     }
+    s->snap = NULL;
     if (snap_buf[0] != '\0') {
-        snap = snap_buf;
+        s->snap = qemu_strdup(snap_buf);
     }
 
-    if ((r = rados_initialize(0, NULL)) < 0) {
+    r = rados_create(&s->cluster, NULL);
+    if (r < 0) {
         error_report("error initializing");
         return r;
     }
 
-    if ((r = rados_open_pool(pool, &s->pool))) {
-        error_report("error opening pool %s", pool);
-        rados_deinitialize();
-        return r;
+    if (strstr(conf, "conf=") == NULL) {
+        r = rados_conf_read_file(s->cluster, NULL);
+        if (r < 0) {
+            error_report("error reading config file");
+            rados_shutdown(s->cluster);
+            return r;
+        }
     }
 
-    if ((r = rados_open_pool(pool, &s->header_pool))) {
-        error_report("error opening pool %s", pool);
-        rados_deinitialize();
-        return r;
+    if (conf[0] != '\0') {
+        r = qemu_rbd_set_conf(s->cluster, conf);
+        if (r < 0) {
+            error_report("error setting config options");
+            rados_shutdown(s->cluster);
+            return r;
+        }
     }
 
-    if ((r = rbd_read_header(s, &hbuf)) < 0) {
-        error_report("error reading header from %s", s->name);
-        goto failed;
-    }
-
-    if (memcmp(hbuf + 64, RBD_HEADER_SIGNATURE, 4)) {
-        error_report("Invalid header signature");
-        r = -EMEDIUMTYPE;
-        goto failed;
-    }
-
-    if (memcmp(hbuf + 68, RBD_HEADER_VERSION, 8)) {
-        error_report("Unknown image version");
-        r = -EMEDIUMTYPE;
-        goto failed;
-    }
-
-    header = (RbdHeader1 *) hbuf;
-    s->size = le64_to_cpu(header->image_size);
-    s->objsize = 1ULL << header->options.order;
-    memcpy(s->block_name, header->block_name, sizeof(header->block_name));
-
-    r = rbd_set_snapc(s->pool, snap, header);
+    r = rados_connect(s->cluster);
     if (r < 0) {
-        error_report("failed setting snap context: %s", strerror(-r));
-        goto failed;
+        error_report("error connecting");
+        rados_shutdown(s->cluster);
+        return r;
     }
 
-    bs->read_only = (snap != NULL);
+    r = rados_ioctx_create(s->cluster, pool, &s->io_ctx);
+    if (r < 0) {
+        error_report("error opening pool %s", pool);
+        rados_shutdown(s->cluster);
+        return r;
+    }
+
+    r = rbd_open(s->io_ctx, s->name, &s->image, s->snap);
+    if (r < 0) {
+        error_report("error reading header from %s", s->name);
+        rados_ioctx_destroy(s->io_ctx);
+        rados_shutdown(s->cluster);
+        return r;
+    }
+
+    bs->read_only = (s->snap != NULL);
 
     s->event_reader_pos = 0;
     r = qemu_pipe(s->fds);
@@ -592,23 +454,20 @@
     }
     fcntl(s->fds[0], F_SETFL, O_NONBLOCK);
     fcntl(s->fds[1], F_SETFL, O_NONBLOCK);
-    qemu_aio_set_fd_handler(s->fds[RBD_FD_READ], rbd_aio_event_reader, NULL,
-        rbd_aio_flush_cb, NULL, s);
+    qemu_aio_set_fd_handler(s->fds[RBD_FD_READ], qemu_rbd_aio_event_reader,
+                            NULL, qemu_rbd_aio_flush_cb, NULL, s);
 
-    qemu_free(hbuf);
 
     return 0;
 
 failed:
-    qemu_free(hbuf);
-
-    rados_close_pool(s->header_pool);
-    rados_close_pool(s->pool);
-    rados_deinitialize();
+    rbd_close(s->image);
+    rados_ioctx_destroy(s->io_ctx);
+    rados_shutdown(s->cluster);
     return r;
 }
 
-static void rbd_close(BlockDriverState *bs)
+static void qemu_rbd_close(BlockDriverState *bs)
 {
     BDRVRBDState *s = bs->opaque;
 
@@ -617,16 +476,17 @@
     qemu_aio_set_fd_handler(s->fds[RBD_FD_READ], NULL , NULL, NULL, NULL,
         NULL);
 
-    rados_close_pool(s->header_pool);
-    rados_close_pool(s->pool);
-    rados_deinitialize();
+    rbd_close(s->image);
+    rados_ioctx_destroy(s->io_ctx);
+    qemu_free(s->snap);
+    rados_shutdown(s->cluster);
 }
 
 /*
  * Cancel aio. Since we don't reference acb in a non qemu threads,
  * it is safe to access it here.
  */
-static void rbd_aio_cancel(BlockDriverAIOCB *blockacb)
+static void qemu_rbd_aio_cancel(BlockDriverAIOCB *blockacb)
 {
     RBDAIOCB *acb = (RBDAIOCB *) blockacb;
     acb->cancelled = 1;
@@ -634,39 +494,28 @@
 
 static AIOPool rbd_aio_pool = {
     .aiocb_size = sizeof(RBDAIOCB),
-    .cancel = rbd_aio_cancel,
+    .cancel = qemu_rbd_aio_cancel,
 };
 
-/*
- * This is the callback function for rados_aio_read and _write
- *
- * Note: this function is being called from a non qemu thread so
- * we need to be careful about what we do here. Generally we only
- * write to the block notification pipe, and do the rest of the
- * io completion handling from rbd_aio_event_reader() which
- * runs in a qemu context.
- */
-static void rbd_finish_aiocb(rados_completion_t c, RADOSCB *rcb)
+static int qemu_rbd_send_pipe(BDRVRBDState *s, RADOSCB *rcb)
 {
-    int ret;
-    rcb->ret = rados_aio_get_return_value(c);
-    rados_aio_release(c);
+    int ret = 0;
     while (1) {
         fd_set wfd;
-        int fd = rcb->s->fds[RBD_FD_WRITE];
+        int fd = s->fds[RBD_FD_WRITE];
 
-        /* send the rcb pointer to the qemu thread that is responsible
-           for the aio completion. Must do it in a qemu thread context */
+        /* send the op pointer to the qemu thread that is responsible
+           for the aio/op completion. Must do it in a qemu thread context */
         ret = write(fd, (void *)&rcb, sizeof(rcb));
         if (ret >= 0) {
             break;
         }
         if (errno == EINTR) {
             continue;
-	}
+        }
         if (errno != EAGAIN) {
             break;
-	}
+        }
 
         FD_ZERO(&wfd);
         FD_SET(fd, &wfd);
@@ -675,13 +524,31 @@
         } while (ret < 0 && errno == EINTR);
     }
 
+    return ret;
+}
+
+/*
+ * This is the callback function for rbd_aio_read and _write
+ *
+ * Note: this function is being called from a non qemu thread so
+ * we need to be careful about what we do here. Generally we only
+ * write to the block notification pipe, and do the rest of the
+ * io completion handling from qemu_rbd_aio_event_reader() which
+ * runs in a qemu context.
+ */
+static void rbd_finish_aiocb(rbd_completion_t c, RADOSCB *rcb)
+{
+    int ret;
+    rcb->ret = rbd_aio_get_return_value(c);
+    rbd_aio_release(c);
+    ret = qemu_rbd_send_pipe(rcb->s, rcb);
     if (ret < 0) {
-        error_report("failed writing to acb->s->fds\n");
+        error_report("failed writing to acb->s->fds");
         qemu_free(rcb);
     }
 }
 
-/* Callback when all queued rados_aio requests are complete */
+/* Callback when all queued rbd_aio requests are complete */
 
 static void rbd_aio_bh_cb(void *opaque)
 {
@@ -707,19 +574,20 @@
 {
     RBDAIOCB *acb;
     RADOSCB *rcb;
-    rados_completion_t c;
-    char n[RBD_MAX_SEG_NAME_SIZE];
-    int64_t segnr, segoffs, segsize, last_segnr;
+    rbd_completion_t c;
     int64_t off, size;
     char *buf;
+    int r;
 
     BDRVRBDState *s = bs->opaque;
 
     acb = qemu_aio_get(&rbd_aio_pool, bs, cb, opaque);
+    if (!acb) {
+        return NULL;
+    }
     acb->write = write;
     acb->qiov = qiov;
     acb->bounce = qemu_blockalign(bs, qiov->size);
-    acb->aiocnt = 0;
     acb->ret = 0;
     acb->error = 0;
     acb->s = s;
@@ -734,95 +602,106 @@
 
     off = sector_num * BDRV_SECTOR_SIZE;
     size = nb_sectors * BDRV_SECTOR_SIZE;
-    segnr = off / s->objsize;
-    segoffs = off % s->objsize;
-    segsize = s->objsize - segoffs;
 
-    last_segnr = ((off + size - 1) / s->objsize);
-    acb->aiocnt = (last_segnr - segnr) + 1;
+    s->qemu_aio_count++; /* All the RADOSCB */
 
-    s->qemu_aio_count += acb->aiocnt; /* All the RADOSCB */
+    rcb = qemu_malloc(sizeof(RADOSCB));
+    rcb->done = 0;
+    rcb->acb = acb;
+    rcb->buf = buf;
+    rcb->s = acb->s;
+    rcb->size = size;
+    r = rbd_aio_create_completion(rcb, (rbd_callback_t) rbd_finish_aiocb, &c);
+    if (r < 0) {
+        goto failed;
+    }
 
-    while (size > 0) {
-        if (size < segsize) {
-            segsize = size;
-        }
+    if (write) {
+        r = rbd_aio_write(s->image, off, size, buf, c);
+    } else {
+        r = rbd_aio_read(s->image, off, size, buf, c);
+    }
 
-        snprintf(n, sizeof(n), "%s.%012" PRIx64, s->block_name,
-                 segnr);
-
-        rcb = qemu_malloc(sizeof(RADOSCB));
-        rcb->done = 0;
-        rcb->acb = acb;
-        rcb->segsize = segsize;
-        rcb->buf = buf;
-        rcb->s = acb->s;
-
-        if (write) {
-            rados_aio_create_completion(rcb, NULL,
-                                        (rados_callback_t) rbd_finish_aiocb,
-                                        &c);
-            rados_aio_write(s->pool, n, segoffs, buf, segsize, c);
-        } else {
-            rados_aio_create_completion(rcb,
-                                        (rados_callback_t) rbd_finish_aiocb,
-                                        NULL, &c);
-            rados_aio_read(s->pool, n, segoffs, buf, segsize, c);
-        }
-
-        buf += segsize;
-        size -= segsize;
-        segoffs = 0;
-        segsize = s->objsize;
-        segnr++;
+    if (r < 0) {
+        goto failed;
     }
 
     return &acb->common;
+
+failed:
+    qemu_free(rcb);
+    s->qemu_aio_count--;
+    qemu_aio_release(acb);
+    return NULL;
 }
 
-static BlockDriverAIOCB *rbd_aio_readv(BlockDriverState * bs,
-                                       int64_t sector_num, QEMUIOVector * qiov,
-                                       int nb_sectors,
-                                       BlockDriverCompletionFunc * cb,
-                                       void *opaque)
+static BlockDriverAIOCB *qemu_rbd_aio_readv(BlockDriverState *bs,
+                                            int64_t sector_num,
+                                            QEMUIOVector *qiov,
+                                            int nb_sectors,
+                                            BlockDriverCompletionFunc *cb,
+                                            void *opaque)
 {
     return rbd_aio_rw_vector(bs, sector_num, qiov, nb_sectors, cb, opaque, 0);
 }
 
-static BlockDriverAIOCB *rbd_aio_writev(BlockDriverState * bs,
-                                        int64_t sector_num, QEMUIOVector * qiov,
-                                        int nb_sectors,
-                                        BlockDriverCompletionFunc * cb,
-                                        void *opaque)
+static BlockDriverAIOCB *qemu_rbd_aio_writev(BlockDriverState *bs,
+                                             int64_t sector_num,
+                                             QEMUIOVector *qiov,
+                                             int nb_sectors,
+                                             BlockDriverCompletionFunc *cb,
+                                             void *opaque)
 {
     return rbd_aio_rw_vector(bs, sector_num, qiov, nb_sectors, cb, opaque, 1);
 }
 
-static int rbd_getinfo(BlockDriverState * bs, BlockDriverInfo * bdi)
+static int qemu_rbd_getinfo(BlockDriverState *bs, BlockDriverInfo *bdi)
 {
     BDRVRBDState *s = bs->opaque;
-    bdi->cluster_size = s->objsize;
+    rbd_image_info_t info;
+    int r;
+
+    r = rbd_stat(s->image, &info, sizeof(info));
+    if (r < 0) {
+        return r;
+    }
+
+    bdi->cluster_size = info.obj_size;
     return 0;
 }
 
-static int64_t rbd_getlength(BlockDriverState * bs)
+static int64_t qemu_rbd_getlength(BlockDriverState *bs)
 {
     BDRVRBDState *s = bs->opaque;
+    rbd_image_info_t info;
+    int r;
 
-    return s->size;
+    r = rbd_stat(s->image, &info, sizeof(info));
+    if (r < 0) {
+        return r;
+    }
+
+    return info.size;
 }
 
-static int rbd_snap_create(BlockDriverState *bs, QEMUSnapshotInfo *sn_info)
+static int qemu_rbd_truncate(BlockDriverState *bs, int64_t offset)
 {
     BDRVRBDState *s = bs->opaque;
-    char inbuf[512], outbuf[128];
-    uint64_t snap_id;
     int r;
-    char *p = inbuf;
-    char *end = inbuf + sizeof(inbuf);
-    char n[RBD_MAX_SEG_NAME_SIZE];
-    char *hbuf = NULL;
-    RbdHeader1 *header;
+
+    r = rbd_resize(s->image, offset);
+    if (r < 0) {
+        return r;
+    }
+
+    return 0;
+}
+
+static int qemu_rbd_snap_create(BlockDriverState *bs,
+                                QEMUSnapshotInfo *sn_info)
+{
+    BDRVRBDState *s = bs->opaque;
+    int r;
 
     if (sn_info->name[0] == '\0') {
         return -EINVAL; /* we need a name for rbd snapshots */
@@ -841,185 +720,57 @@
         return -ERANGE;
     }
 
-    r = rados_selfmanaged_snap_create(s->header_pool, &snap_id);
+    r = rbd_snap_create(s->image, sn_info->name);
     if (r < 0) {
-        error_report("failed to create snap id: %s", strerror(-r));
+        error_report("failed to create snap: %s", strerror(-r));
         return r;
     }
 
-    *(uint32_t *)p = strlen(sn_info->name);
-    cpu_to_le32s((uint32_t *)p);
-    p += sizeof(uint32_t);
-    strncpy(p, sn_info->name, end - p);
-    p += strlen(p);
-    if (p + sizeof(snap_id) > end) {
-        error_report("invalid input parameter");
-        return -EINVAL;
-    }
-
-    *(uint64_t *)p = snap_id;
-    cpu_to_le64s((uint64_t *)p);
-
-    snprintf(n, sizeof(n), "%s%s", s->name, RBD_SUFFIX);
-
-    r = rados_exec(s->header_pool, n, "rbd", "snap_add", inbuf,
-                   sizeof(inbuf), outbuf, sizeof(outbuf));
-    if (r < 0) {
-        error_report("rbd.snap_add execution failed failed: %s", strerror(-r));
-        return r;
-    }
-
-    sprintf(sn_info->id_str, "%s", sn_info->name);
-
-    r = rbd_read_header(s, &hbuf);
-    if (r < 0) {
-        error_report("failed reading header: %s", strerror(-r));
-        return r;
-    }
-
-    header = (RbdHeader1 *) hbuf;
-    r = rbd_set_snapc(s->pool, sn_info->name, header);
-    if (r < 0) {
-        error_report("failed setting snap context: %s", strerror(-r));
-        goto failed;
-    }
-
     return 0;
-
-failed:
-    qemu_free(header);
-    return r;
 }
 
-static int decode32(char **p, const char *end, uint32_t *v)
-{
-    if (*p + 4 > end) {
-	return -ERANGE;
-    }
-
-    *v = *(uint32_t *)(*p);
-    le32_to_cpus(v);
-    *p += 4;
-    return 0;
-}
-
-static int decode64(char **p, const char *end, uint64_t *v)
-{
-    if (*p + 8 > end) {
-        return -ERANGE;
-    }
-
-    *v = *(uint64_t *)(*p);
-    le64_to_cpus(v);
-    *p += 8;
-    return 0;
-}
-
-static int decode_str(char **p, const char *end, char **s)
-{
-    uint32_t len;
-    int r;
-
-    if ((r = decode32(p, end, &len)) < 0) {
-        return r;
-    }
-
-    *s = qemu_malloc(len + 1);
-    memcpy(*s, *p, len);
-    *p += len;
-    (*s)[len] = '\0';
-
-    return len;
-}
-
-static int rbd_snap_list(BlockDriverState *bs, QEMUSnapshotInfo **psn_tab)
+static int qemu_rbd_snap_list(BlockDriverState *bs,
+                              QEMUSnapshotInfo **psn_tab)
 {
     BDRVRBDState *s = bs->opaque;
-    char n[RBD_MAX_SEG_NAME_SIZE];
     QEMUSnapshotInfo *sn_info, *sn_tab = NULL;
-    RbdHeader1 *header;
-    char *hbuf = NULL;
-    char *outbuf = NULL, *end, *buf;
-    uint64_t len;
-    uint64_t snap_seq;
-    uint32_t snap_count;
-    int r, i;
+    int i, snap_count;
+    rbd_snap_info_t *snaps;
+    int max_snaps = RBD_MAX_SNAPS;
 
-    /* read header to estimate how much space we need to read the snap
-     * list */
-    if ((r = rbd_read_header(s, &hbuf)) < 0) {
-        goto done_err;
-    }
-    header = (RbdHeader1 *)hbuf;
-    len = le64_to_cpu(header->snap_names_len);
-    len += 1024; /* should have already been enough, but new snapshots might
-                    already been created since we read the header. just allocate
-                    a bit more, so that in most cases it'll suffice anyway */
-    qemu_free(hbuf);
-
-    snprintf(n, sizeof(n), "%s%s", s->name, RBD_SUFFIX);
-    while (1) {
-        qemu_free(outbuf);
-        outbuf = qemu_malloc(len);
-
-        r = rados_exec(s->header_pool, n, "rbd", "snap_list", NULL, 0,
-                       outbuf, len);
-        if (r < 0) {
-            error_report("rbd.snap_list execution failed failed: %s", strerror(-r));
-            goto done_err;
+    do {
+        snaps = qemu_malloc(sizeof(*snaps) * max_snaps);
+        snap_count = rbd_snap_list(s->image, snaps, &max_snaps);
+        if (snap_count < 0) {
+            qemu_free(snaps);
         }
-        if (r != len) {
-            break;
-	}
+    } while (snap_count == -ERANGE);
 
-        /* if we're here, we probably raced with some snaps creation */
-        len *= 2;
-    }
-    buf = outbuf;
-    end = buf + len;
-
-    if ((r = decode64(&buf, end, &snap_seq)) < 0) {
-        goto done_err;
-    }
-    if ((r = decode32(&buf, end, &snap_count)) < 0) {
-        goto done_err;
+    if (snap_count <= 0) {
+        return snap_count;
     }
 
     sn_tab = qemu_mallocz(snap_count * sizeof(QEMUSnapshotInfo));
-    for (i = 0; i < snap_count; i++) {
-        uint64_t id, image_size;
-        char *snap_name;
 
-        if ((r = decode64(&buf, end, &id)) < 0) {
-            goto done_err;
-        }
-        if ((r = decode64(&buf, end, &image_size)) < 0) {
-            goto done_err;
-        }
-        if ((r = decode_str(&buf, end, &snap_name)) < 0) {
-            goto done_err;
-        }
+    for (i = 0; i < snap_count; i++) {
+        const char *snap_name = snaps[i].name;
 
         sn_info = sn_tab + i;
         pstrcpy(sn_info->id_str, sizeof(sn_info->id_str), snap_name);
         pstrcpy(sn_info->name, sizeof(sn_info->name), snap_name);
-        qemu_free(snap_name);
 
-        sn_info->vm_state_size = image_size;
+        sn_info->vm_state_size = snaps[i].size;
         sn_info->date_sec = 0;
         sn_info->date_nsec = 0;
         sn_info->vm_clock_nsec = 0;
     }
+    rbd_snap_list_end(snaps);
+
     *psn_tab = sn_tab;
-    qemu_free(outbuf);
     return snap_count;
-done_err:
-    qemu_free(sn_tab);
-    qemu_free(outbuf);
-    return r;
 }
 
-static QEMUOptionParameter rbd_create_options[] = {
+static QEMUOptionParameter qemu_rbd_create_options[] = {
     {
      .name = BLOCK_OPT_SIZE,
      .type = OPT_SIZE,
@@ -1036,19 +787,20 @@
 static BlockDriver bdrv_rbd = {
     .format_name        = "rbd",
     .instance_size      = sizeof(BDRVRBDState),
-    .bdrv_file_open     = rbd_open,
-    .bdrv_close         = rbd_close,
-    .bdrv_create        = rbd_create,
-    .bdrv_get_info      = rbd_getinfo,
-    .create_options     = rbd_create_options,
-    .bdrv_getlength     = rbd_getlength,
+    .bdrv_file_open     = qemu_rbd_open,
+    .bdrv_close         = qemu_rbd_close,
+    .bdrv_create        = qemu_rbd_create,
+    .bdrv_get_info      = qemu_rbd_getinfo,
+    .create_options     = qemu_rbd_create_options,
+    .bdrv_getlength     = qemu_rbd_getlength,
+    .bdrv_truncate      = qemu_rbd_truncate,
     .protocol_name      = "rbd",
 
-    .bdrv_aio_readv     = rbd_aio_readv,
-    .bdrv_aio_writev    = rbd_aio_writev,
+    .bdrv_aio_readv     = qemu_rbd_aio_readv,
+    .bdrv_aio_writev    = qemu_rbd_aio_writev,
 
-    .bdrv_snapshot_create = rbd_snap_create,
-    .bdrv_snapshot_list = rbd_snap_list,
+    .bdrv_snapshot_create = qemu_rbd_snap_create,
+    .bdrv_snapshot_list = qemu_rbd_snap_list,
 };
 
 static void bdrv_rbd_init(void)
diff --git a/block/rbd_types.h b/block/rbd_types.h
deleted file mode 100644
index f4cca99..0000000
--- a/block/rbd_types.h
+++ /dev/null
@@ -1,71 +0,0 @@
-/*
- * Ceph - scalable distributed file system
- *
- * Copyright (C) 2004-2010 Sage Weil <sage@newdream.net>
- *
- * This is free software; you can redistribute it and/or
- * modify it under the terms of the GNU Lesser General Public
- * License version 2.1, as published by the Free Software
- * Foundation.  See file COPYING.LIB.
- *
- */
-
-#ifndef CEPH_RBD_TYPES_H
-#define CEPH_RBD_TYPES_H
-
-
-/*
- * rbd image 'foo' consists of objects
- *   foo.rbd      - image metadata
- *   foo.00000000
- *   foo.00000001
- *   ...          - data
- */
-
-#define RBD_SUFFIX              ".rbd"
-#define RBD_DIRECTORY           "rbd_directory"
-#define RBD_INFO                "rbd_info"
-
-#define RBD_DEFAULT_OBJ_ORDER   22   /* 4MB */
-
-#define RBD_MAX_OBJ_NAME_SIZE   96
-#define RBD_MAX_BLOCK_NAME_SIZE 24
-#define RBD_MAX_SEG_NAME_SIZE   128
-
-#define RBD_COMP_NONE           0
-#define RBD_CRYPT_NONE          0
-
-#define RBD_HEADER_TEXT         "<<< Rados Block Device Image >>>\n"
-#define RBD_HEADER_SIGNATURE    "RBD"
-#define RBD_HEADER_VERSION      "001.005"
-
-struct rbd_info {
-    uint64_t max_id;
-} __attribute__ ((packed));
-
-struct rbd_obj_snap_ondisk {
-    uint64_t id;
-    uint64_t image_size;
-} __attribute__((packed));
-
-struct rbd_obj_header_ondisk {
-    char text[40];
-    char block_name[RBD_MAX_BLOCK_NAME_SIZE];
-    char signature[4];
-    char version[8];
-    struct {
-        uint8_t order;
-        uint8_t crypt_type;
-        uint8_t comp_type;
-        uint8_t unused;
-    } __attribute__((packed)) options;
-    uint64_t image_size;
-    uint64_t snap_seq;
-    uint32_t snap_count;
-    uint32_t reserved;
-    uint64_t snap_names_len;
-    struct rbd_obj_snap_ondisk snaps[0];
-} __attribute__((packed));
-
-
-#endif
diff --git a/block/vdi.c b/block/vdi.c
index 701745b..261cf9b 100644
--- a/block/vdi.c
+++ b/block/vdi.c
@@ -87,6 +87,7 @@
 #define MiB     (KiB * KiB)
 
 #define SECTOR_SIZE 512
+#define DEFAULT_CLUSTER_SIZE (1 * MiB)
 
 #if defined(CONFIG_VDI_DEBUG)
 #define logout(fmt, ...) \
@@ -151,6 +152,7 @@
     /* Buffer for new allocated block. */
     void *block_buffer;
     void *orig_buf;
+    bool is_write;
     int header_modified;
     BlockDriverAIOCB *hd_aiocb;
     struct iovec hd_iov;
@@ -503,6 +505,8 @@
         acb->hd_aiocb = NULL;
         acb->sector_num = sector_num;
         acb->qiov = qiov;
+        acb->is_write = is_write;
+
         if (qiov->niov > 1) {
             acb->buf = qemu_blockalign(bs, qiov->size);
             acb->orig_buf = acb->buf;
@@ -541,14 +545,20 @@
 }
 
 static void vdi_aio_read_cb(void *opaque, int ret);
+static void vdi_aio_write_cb(void *opaque, int ret);
 
-static void vdi_aio_read_bh(void *opaque)
+static void vdi_aio_rw_bh(void *opaque)
 {
     VdiAIOCB *acb = opaque;
     logout("\n");
     qemu_bh_delete(acb->bh);
     acb->bh = NULL;
-    vdi_aio_read_cb(opaque, 0);
+
+    if (acb->is_write) {
+        vdi_aio_write_cb(opaque, 0);
+    } else {
+        vdi_aio_read_cb(opaque, 0);
+    }
 }
 
 static void vdi_aio_read_cb(void *opaque, int ret)
@@ -596,7 +606,7 @@
     if (bmap_entry == VDI_UNALLOCATED) {
         /* Block not allocated, return zeros, no need to wait. */
         memset(acb->buf, 0, n_sectors * SECTOR_SIZE);
-        ret = vdi_schedule_bh(vdi_aio_read_bh, acb);
+        ret = vdi_schedule_bh(vdi_aio_rw_bh, acb);
         if (ret < 0) {
             goto done;
         }
@@ -629,12 +639,23 @@
         BlockDriverCompletionFunc *cb, void *opaque)
 {
     VdiAIOCB *acb;
+    int ret;
+
     logout("\n");
     acb = vdi_aio_setup(bs, sector_num, qiov, nb_sectors, cb, opaque, 0);
     if (!acb) {
         return NULL;
     }
-    vdi_aio_read_cb(acb, 0);
+
+    ret = vdi_schedule_bh(vdi_aio_rw_bh, acb);
+    if (ret < 0) {
+        if (acb->qiov->niov > 1) {
+            qemu_vfree(acb->orig_buf);
+        }
+        qemu_aio_release(acb);
+        return NULL;
+    }
+
     return &acb->common;
 }
 
@@ -788,12 +809,23 @@
         BlockDriverCompletionFunc *cb, void *opaque)
 {
     VdiAIOCB *acb;
+    int ret;
+
     logout("\n");
     acb = vdi_aio_setup(bs, sector_num, qiov, nb_sectors, cb, opaque, 1);
     if (!acb) {
         return NULL;
     }
-    vdi_aio_write_cb(acb, 0);
+
+    ret = vdi_schedule_bh(vdi_aio_rw_bh, acb);
+    if (ret < 0) {
+        if (acb->qiov->niov > 1) {
+            qemu_vfree(acb->orig_buf);
+        }
+        qemu_aio_release(acb);
+        return NULL;
+    }
+
     return &acb->common;
 }
 
@@ -803,7 +835,7 @@
     int result = 0;
     uint64_t bytes = 0;
     uint32_t blocks;
-    size_t block_size = 1 * MiB;
+    size_t block_size = DEFAULT_CLUSTER_SIZE;
     uint32_t image_type = VDI_TYPE_DYNAMIC;
     VdiHeader header;
     size_t i;
@@ -921,7 +953,8 @@
     {
         .name = BLOCK_OPT_CLUSTER_SIZE,
         .type = OPT_SIZE,
-        .help = "VDI cluster (block) size"
+        .help = "VDI cluster (block) size",
+        .value = { .n = DEFAULT_CLUSTER_SIZE },
     },
 #endif
 #if defined(CONFIG_VDI_STATIC_IMAGE)
diff --git a/block/vmdk.c b/block/vmdk.c
index 8fc9d67..922b23d 100644
--- a/block/vmdk.c
+++ b/block/vmdk.c
@@ -716,11 +716,11 @@
         return -errno;
     magic = cpu_to_be32(VMDK4_MAGIC);
     memset(&header, 0, sizeof(header));
-    header.version = cpu_to_le32(1);
-    header.flags = cpu_to_le32(3); /* ?? */
-    header.capacity = cpu_to_le64(total_size);
-    header.granularity = cpu_to_le64(128);
-    header.num_gtes_per_gte = cpu_to_le32(512);
+    header.version = 1;
+    header.flags = 3; /* ?? */
+    header.capacity = total_size;
+    header.granularity = 128;
+    header.num_gtes_per_gte = 512;
 
     grains = (total_size + header.granularity - 1) / header.granularity;
     gt_size = ((header.num_gtes_per_gte * sizeof(uint32_t)) + 511) >> 9;
@@ -736,6 +736,12 @@
          header.granularity - 1) / header.granularity) *
         header.granularity;
 
+    /* swap endianness for all header fields */
+    header.version = cpu_to_le32(header.version);
+    header.flags = cpu_to_le32(header.flags);
+    header.capacity = cpu_to_le64(header.capacity);
+    header.granularity = cpu_to_le64(header.granularity);
+    header.num_gtes_per_gte = cpu_to_le32(header.num_gtes_per_gte);
     header.desc_offset = cpu_to_le64(header.desc_offset);
     header.desc_size = cpu_to_le64(header.desc_size);
     header.rgd_offset = cpu_to_le64(header.rgd_offset);
@@ -759,7 +765,7 @@
         goto exit;
     }
 
-    ret = ftruncate(fd, header.grain_offset << 9);
+    ret = ftruncate(fd, le64_to_cpu(header.grain_offset) << 9);
     if (ret < 0) {
         ret = -errno;
         goto exit;
@@ -767,7 +773,7 @@
 
     /* write grain directory */
     lseek(fd, le64_to_cpu(header.rgd_offset) << 9, SEEK_SET);
-    for (i = 0, tmp = header.rgd_offset + gd_size;
+    for (i = 0, tmp = le64_to_cpu(header.rgd_offset) + gd_size;
          i < gt_count; i++, tmp += gt_size) {
         ret = qemu_write_full(fd, &tmp, sizeof(tmp));
         if (ret != sizeof(tmp)) {
@@ -778,7 +784,7 @@
 
     /* write backup grain directory */
     lseek(fd, le64_to_cpu(header.gd_offset) << 9, SEEK_SET);
-    for (i = 0, tmp = header.gd_offset + gd_size;
+    for (i = 0, tmp = le64_to_cpu(header.gd_offset) + gd_size;
          i < gt_count; i++, tmp += gt_size) {
         ret = qemu_write_full(fd, &tmp, sizeof(tmp));
         if (ret != sizeof(tmp)) {
diff --git a/block_int.h b/block_int.h
index fa91337..1e265d2 100644
--- a/block_int.h
+++ b/block_int.h
@@ -203,8 +203,8 @@
     void *private;
 };
 
-#define CHANGE_MEDIA	0x01
-#define CHANGE_SIZE	0x02
+#define CHANGE_MEDIA    0x01
+#define CHANGE_SIZE     0x02
 
 struct BlockDriverAIOCB {
     AIOPool *pool;
diff --git a/blockdev.c b/blockdev.c
index 6e0eb83..1502575 100644
--- a/blockdev.c
+++ b/blockdev.c
@@ -326,7 +326,7 @@
 
     if ((buf = qemu_opt_get(opts, "cache")) != NULL) {
         if (!strcmp(buf, "off") || !strcmp(buf, "none")) {
-            bdrv_flags |= BDRV_O_NOCACHE;
+            bdrv_flags |= BDRV_O_NOCACHE | BDRV_O_CACHE_WB;
         } else if (!strcmp(buf, "writeback")) {
             bdrv_flags |= BDRV_O_CACHE_WB;
         } else if (!strcmp(buf, "unsafe")) {
diff --git a/bsd-user/syscall.c b/bsd-user/syscall.c
index eb1cdf2..d4d039a 100644
--- a/bsd-user/syscall.c
+++ b/bsd-user/syscall.c
@@ -31,7 +31,6 @@
 #include <sys/syscall.h>
 #include <sys/param.h>
 #include <sys/sysctl.h>
-#include <signal.h>
 #include <utime.h>
 
 #include "qemu.h"
diff --git a/compatfd.c b/compatfd.c
index bd377c4..41586ce 100644
--- a/compatfd.c
+++ b/compatfd.c
@@ -29,7 +29,7 @@
     sigset_t all;
 
     sigfillset(&all);
-    sigprocmask(SIG_BLOCK, &all, NULL);
+    pthread_sigmask(SIG_BLOCK, &all, NULL);
 
     while (1) {
         int sig;
diff --git a/configure b/configure
index 79e2ac9..a6e9d1c 100755
--- a/configure
+++ b/configure
@@ -230,7 +230,7 @@
 # default flags for all hosts
 QEMU_CFLAGS="-fno-strict-aliasing $QEMU_CFLAGS"
 CFLAGS="-g $CFLAGS"
-QEMU_CFLAGS="-Wall -Wundef -Wendif-labels -Wwrite-strings -Wmissing-prototypes $QEMU_CFLAGS"
+QEMU_CFLAGS="-Wall -Wundef -Wwrite-strings -Wmissing-prototypes $QEMU_CFLAGS"
 QEMU_CFLAGS="-Wstrict-prototypes -Wredundant-decls $QEMU_CFLAGS"
 QEMU_CFLAGS="-D_GNU_SOURCE -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE $QEMU_CFLAGS"
 QEMU_CFLAGS="-D_FORTIFY_SOURCE=2 $QEMU_CFLAGS"
@@ -834,6 +834,7 @@
     default_target_list="\
 i386-softmmu \
 x86_64-softmmu \
+alpha-softmmu \
 arm-softmmu \
 cris-softmmu \
 lm32-softmmu \
@@ -1040,7 +1041,7 @@
 gcc_flags="-Wold-style-declaration -Wold-style-definition -Wtype-limits"
 gcc_flags="-Wformat-security -Wformat-y2k -Winit-self -Wignored-qualifiers $gcc_flags"
 gcc_flags="-Wmissing-include-dirs -Wempty-body -Wnested-externs $gcc_flags"
-gcc_flags="-fstack-protector-all $gcc_flags"
+gcc_flags="-fstack-protector-all -Wendif-labels $gcc_flags"
 cat > $TMPC << EOF
 int main(void) { return 0; }
 EOF
@@ -1721,7 +1722,7 @@
 if test "$curl" != "no" ; then
   cat > $TMPC << EOF
 #include <curl/curl.h>
-int main(void) { return curl_easy_init(); }
+int main(void) { curl_easy_init(); curl_multi_setopt(0, 0, 0); return 0; }
 EOF
   curl_cflags=`$curlconfig --cflags 2>/dev/null`
   curl_libs=`$curlconfig --libs 2>/dev/null`
@@ -1929,41 +1930,24 @@
 if test "$rbd" != "no" ; then
   cat > $TMPC <<EOF
 #include <stdio.h>
-#include <rados/librados.h>
-int main(void) { rados_initialize(0, NULL); return 0; }
-EOF
-  rbd_libs="-lrados"
-  if compile_prog "" "$rbd_libs" ; then
-    librados_too_old=no
-    cat > $TMPC <<EOF
-#include <stdio.h>
-#include <rados/librados.h>
-#ifndef CEPH_OSD_TMAP_SET
-#error missing CEPH_OSD_TMAP_SET
-#endif
+#include <rbd/librbd.h>
 int main(void) {
-    int (*func)(const rados_pool_t pool, uint64_t *snapid) = rados_selfmanaged_snap_create;
-    rados_initialize(0, NULL);
+    rados_t cluster;
+    rados_create(&cluster, NULL);
     return 0;
 }
 EOF
-    if compile_prog "" "$rbd_libs" ; then
-      rbd=yes
-      libs_tools="$rbd_libs $libs_tools"
-      libs_softmmu="$rbd_libs $libs_softmmu"
-    else
-      rbd=no
-      librados_too_old=yes
-    fi
+  rbd_libs="-lrbd -lrados"
+  if compile_prog "" "$rbd_libs" ; then
+    rbd=yes
+    libs_tools="$rbd_libs $libs_tools"
+    libs_softmmu="$rbd_libs $libs_softmmu"
   else
     if test "$rbd" = "yes" ; then
       feature_not_found "rados block device"
     fi
     rbd=no
   fi
-  if test "$librados_too_old" = "yes" ; then
-    echo "-> Your librados version is too old - upgrade needed to have rbd support"
-  fi
 fi
 
 ##########################################
@@ -2443,7 +2427,7 @@
 EOF
   spice_cflags=$($pkg_config --cflags spice-protocol spice-server 2>/dev/null)
   spice_libs=$($pkg_config --libs spice-protocol spice-server 2>/dev/null)
-  if $pkg_config --atleast-version=0.5.3 spice-server >/dev/null 2>&1 && \
+  if $pkg_config --atleast-version=0.6.0 spice-server >/dev/null 2>&1 && \
      compile_prog "$spice_cflags" "$spice_libs" ; then
     spice="yes"
     libs_softmmu="$libs_softmmu $spice_libs"
@@ -3400,8 +3384,6 @@
   echo "TARGET_XML_FILES=$list" >> $config_target_mak
 fi
 
-echo "CONFIG_SOFTFLOAT=y" >> $config_target_mak
-
 if test "$target_user_only" = "yes" -a "$bflt" = "yes"; then
   echo "TARGET_HAS_BFLT=y" >> $config_target_mak
 fi
diff --git a/console.c b/console.c
index 871c1d4..9c6addf 100644
--- a/console.c
+++ b/console.c
@@ -180,7 +180,7 @@
     active_console = consoles[0];
     /* There is currently no way of specifying which screen we want to dump,
        so always dump the first one.  */
-    if (consoles[0]->hw_screen_dump)
+    if (consoles[0] && consoles[0]->hw_screen_dump)
         consoles[0]->hw_screen_dump(consoles[0]->hw, filename);
     active_console = previous_active_console;
 }
diff --git a/cpu-all.h b/cpu-all.h
index 54df1d3..880f570 100644
--- a/cpu-all.h
+++ b/cpu-all.h
@@ -123,8 +123,7 @@
    endian ! */
 typedef union {
     float64 d;
-#if defined(HOST_WORDS_BIGENDIAN) \
-    || (defined(__arm__) && !defined(__VFP_FP__) && !defined(CONFIG_SOFTFLOAT))
+#if defined(HOST_WORDS_BIGENDIAN)
     struct {
         uint32_t upper;
         uint32_t lower;
@@ -138,7 +137,6 @@
     uint64_t ll;
 } CPU_DoubleU;
 
-#if defined(FLOATX80)
 typedef union {
      floatx80 d;
      struct {
@@ -146,9 +144,7 @@
          uint16_t upper;
      } l;
 } CPU_LDoubleU;
-#endif
 
-#if defined(CONFIG_SOFTFLOAT)
 typedef union {
     float128 q;
 #if defined(HOST_WORDS_BIGENDIAN)
@@ -175,7 +171,6 @@
     } ll;
 #endif
 } CPU_QuadU;
-#endif
 
 /* CPU memory access without any memory or io remapping */
 
diff --git a/cpu-common.h b/cpu-common.h
index 151c32c..9f59172 100644
--- a/cpu-common.h
+++ b/cpu-common.h
@@ -61,6 +61,7 @@
                         ram_addr_t size, void *host);
 ram_addr_t qemu_ram_alloc(DeviceState *dev, const char *name, ram_addr_t size);
 void qemu_ram_free(ram_addr_t addr);
+void qemu_ram_free_from_ptr(ram_addr_t addr);
 void qemu_ram_remap(ram_addr_t addr, ram_addr_t length);
 /* This should only be used for ram local to a device.  */
 void *qemu_get_ram_ptr(ram_addr_t addr);
diff --git a/cpu-exec.c b/cpu-exec.c
index 6ddd8dd..e1de56b 100644
--- a/cpu-exec.c
+++ b/cpu-exec.c
@@ -488,9 +488,36 @@
                         next_tb = 0;
                     }
 #elif defined(TARGET_ALPHA)
-                    if (interrupt_request & CPU_INTERRUPT_HARD) {
-                        do_interrupt(env);
-                        next_tb = 0;
+                    {
+                        int idx = -1;
+                        /* ??? This hard-codes the OSF/1 interrupt levels.  */
+		        switch (env->pal_mode ? 7 : env->ps & PS_INT_MASK) {
+                        case 0 ... 3:
+                            if (interrupt_request & CPU_INTERRUPT_HARD) {
+                                idx = EXCP_DEV_INTERRUPT;
+                            }
+                            /* FALLTHRU */
+                        case 4:
+                            if (interrupt_request & CPU_INTERRUPT_TIMER) {
+                                idx = EXCP_CLK_INTERRUPT;
+                            }
+                            /* FALLTHRU */
+                        case 5:
+                            if (interrupt_request & CPU_INTERRUPT_SMP) {
+                                idx = EXCP_SMP_INTERRUPT;
+                            }
+                            /* FALLTHRU */
+                        case 6:
+                            if (interrupt_request & CPU_INTERRUPT_MCHK) {
+                                idx = EXCP_MCHK;
+                            }
+                        }
+                        if (idx >= 0) {
+                            env->exception_index = idx;
+                            env->error_code = 0;
+                            do_interrupt(env);
+                            next_tb = 0;
+                        }
                     }
 #elif defined(TARGET_CRIS)
                     if (interrupt_request & CPU_INTERRUPT_HARD
diff --git a/darwin-user/signal.c b/darwin-user/signal.c
index 4862018..e2adca3 100644
--- a/darwin-user/signal.c
+++ b/darwin-user/signal.c
@@ -21,7 +21,6 @@
 #include <string.h>
 #include <stdarg.h>
 #include <unistd.h>
-#include <signal.h>
 #include <errno.h>
 #include <sys/ucontext.h>
 
@@ -32,8 +31,6 @@
 #undef uc_link
 #endif
 
-#include <signal.h>
-
 #include "qemu.h"
 #include "qemu-common.h"
 
diff --git a/default-configs/alpha-softmmu.mak b/default-configs/alpha-softmmu.mak
new file mode 100644
index 0000000..abadcff
--- /dev/null
+++ b/default-configs/alpha-softmmu.mak
@@ -0,0 +1,9 @@
+# Default configuration for alpha-softmmu
+
+include pci.mak
+CONFIG_SERIAL=y
+CONFIG_I8254=y
+CONFIG_VGA_PCI=y
+CONFIG_IDE_CORE=y
+CONFIG_IDE_QDEV=y
+CONFIG_VMWARE_VGA=y
diff --git a/dis-asm.h b/dis-asm.h
index 296537a..5b07d7f 100644
--- a/dis-asm.h
+++ b/dis-asm.h
@@ -184,6 +184,9 @@
 #define bfd_mach_sh5        0x50
   bfd_arch_alpha,      /* Dec Alpha */
 #define bfd_mach_alpha 1
+#define bfd_mach_alpha_ev4  0x10
+#define bfd_mach_alpha_ev5  0x20
+#define bfd_mach_alpha_ev6  0x30
   bfd_arch_arm,        /* Advanced Risc Machines ARM */
 #define bfd_mach_arm_unknown	0
 #define bfd_mach_arm_2		1
diff --git a/disas.c b/disas.c
index 223606c..d208c52 100644
--- a/disas.c
+++ b/disas.c
@@ -205,7 +205,7 @@
     disasm_info.mach = bfd_mach_sh4;
     print_insn = print_insn_sh;
 #elif defined(TARGET_ALPHA)
-    disasm_info.mach = bfd_mach_alpha;
+    disasm_info.mach = bfd_mach_alpha_ev6;
     print_insn = print_insn_alpha;
 #elif defined(TARGET_CRIS)
     if (flags != 32) {
diff --git a/dma-helpers.c b/dma-helpers.c
index 712ed89..ba7f897 100644
--- a/dma-helpers.c
+++ b/dma-helpers.c
@@ -47,6 +47,7 @@
     target_phys_addr_t sg_cur_byte;
     QEMUIOVector iov;
     QEMUBH *bh;
+    DMAIOFunc *io_func;
 } DMAAIOCB;
 
 static void dma_bdrv_cb(void *opaque, int ret);
@@ -116,13 +117,8 @@
         return;
     }
 
-    if (dbs->is_write) {
-        dbs->acb = bdrv_aio_writev(dbs->bs, dbs->sector_num, &dbs->iov,
-                                   dbs->iov.size / 512, dma_bdrv_cb, dbs);
-    } else {
-        dbs->acb = bdrv_aio_readv(dbs->bs, dbs->sector_num, &dbs->iov,
-                                  dbs->iov.size / 512, dma_bdrv_cb, dbs);
-    }
+    dbs->acb = dbs->io_func(dbs->bs, dbs->sector_num, &dbs->iov,
+                            dbs->iov.size / 512, dma_bdrv_cb, dbs);
     if (!dbs->acb) {
         dma_bdrv_unmap(dbs);
         qemu_iovec_destroy(&dbs->iov);
@@ -144,12 +140,12 @@
     .cancel             = dma_aio_cancel,
 };
 
-static BlockDriverAIOCB *dma_bdrv_io(
+BlockDriverAIOCB *dma_bdrv_io(
     BlockDriverState *bs, QEMUSGList *sg, uint64_t sector_num,
-    BlockDriverCompletionFunc *cb, void *opaque,
-    int is_write)
+    DMAIOFunc *io_func, BlockDriverCompletionFunc *cb,
+    void *opaque, int is_write)
 {
-    DMAAIOCB *dbs =  qemu_aio_get(&dma_aio_pool, bs, cb, opaque);
+    DMAAIOCB *dbs = qemu_aio_get(&dma_aio_pool, bs, cb, opaque);
 
     dbs->acb = NULL;
     dbs->bs = bs;
@@ -158,6 +154,7 @@
     dbs->sg_cur_index = 0;
     dbs->sg_cur_byte = 0;
     dbs->is_write = is_write;
+    dbs->io_func = io_func;
     dbs->bh = NULL;
     qemu_iovec_init(&dbs->iov, sg->nsg);
     dma_bdrv_cb(dbs, 0);
@@ -173,12 +170,12 @@
                                 QEMUSGList *sg, uint64_t sector,
                                 void (*cb)(void *opaque, int ret), void *opaque)
 {
-    return dma_bdrv_io(bs, sg, sector, cb, opaque, 0);
+    return dma_bdrv_io(bs, sg, sector, bdrv_aio_readv, cb, opaque, 0);
 }
 
 BlockDriverAIOCB *dma_bdrv_write(BlockDriverState *bs,
                                  QEMUSGList *sg, uint64_t sector,
                                  void (*cb)(void *opaque, int ret), void *opaque)
 {
-    return dma_bdrv_io(bs, sg, sector, cb, opaque, 1);
+    return dma_bdrv_io(bs, sg, sector, bdrv_aio_writev, cb, opaque, 1);
 }
diff --git a/dma.h b/dma.h
index f3bb275..3d8324b 100644
--- a/dma.h
+++ b/dma.h
@@ -32,6 +32,14 @@
                      target_phys_addr_t len);
 void qemu_sglist_destroy(QEMUSGList *qsg);
 
+typedef BlockDriverAIOCB *DMAIOFunc(BlockDriverState *bs, int64_t sector_num,
+                                 QEMUIOVector *iov, int nb_sectors,
+                                 BlockDriverCompletionFunc *cb, void *opaque);
+
+BlockDriverAIOCB *dma_bdrv_io(BlockDriverState *bs,
+                              QEMUSGList *sg, uint64_t sector_num,
+                              DMAIOFunc *io_func, BlockDriverCompletionFunc *cb,
+                              void *opaque, int is_write);
 BlockDriverAIOCB *dma_bdrv_read(BlockDriverState *bs,
                                 QEMUSGList *sg, uint64_t sector,
                                 BlockDriverCompletionFunc *cb, void *opaque);
diff --git a/docs/qdev-device-use.txt b/docs/qdev-device-use.txt
index 4bb2be8..057c322 100644
--- a/docs/qdev-device-use.txt
+++ b/docs/qdev-device-use.txt
@@ -8,20 +8,23 @@
 
 A device typically has a device address on its parent bus.  For buses
 where this address can be configured, devices provide a bus-specific
-property.  These are
+property.  Examples:
 
-    bus     property name       value format
-    PCI     addr                %x.%x (dev.fn, .fn optional)
-    I2C     address             %u
-    SCSI    scsi-id             %u
+    bus         property name       value format
+    PCI         addr                %x.%x    (dev.fn, .fn optional)
+    I2C         address             %u
+    SCSI        scsi-id             %u
+    IDE         unit                %u
+    HDA         cad                 %u
+    virtio-serial-bus  nr           %u
+    ccid-bus    slot                %u
+    USB         port                %d(.%d)*    (port.port...)
 
 Example: device i440FX-pcihost is on the root bus, and provides a PCI
 bus named pci.0.  To put a FOO device into its slot 4, use -device
 FOO,bus=/i440FX-pcihost/pci.0,addr=4.  The abbreviated form bus=pci.0
 also works as long as the bus name is unique.
 
-Note: the USB device address can't be controlled at this time.
-
 === Block Devices ===
 
 A QEMU block device (drive) has a host and a guest part.
@@ -44,28 +47,43 @@
 
 The various old ways to define drives all boil down to the common form
 
-    -drive if=TYPE,index=IDX,bus=BUS,unit=UNIT,HOST-OPTS...
+    -drive if=TYPE,bus=BUS,unit=UNIT,OPTS...
 
 TYPE, BUS and UNIT identify the controller device, which of its buses
 to use, and the drive's address on that bus.  Details depend on TYPE.
-IDX is an alternative way to specify BUS and UNIT.
+
+Instead of bus=BUS,unit=UNIT, you can also say index=IDX.
 
 In the new way, this becomes something like
 
    -drive if=none,id=DRIVE-ID,HOST-OPTS...
    -device DEVNAME,drive=DRIVE-ID,DEV-OPTS...
 
-The -device argument differs in detail for each kind of drive:
+The old OPTS get split into HOST-OPTS and DEV-OPTS as follows:
+
+* file, format, snapshot, cache, aio, readonly, rerror, werror go into
+  HOST-OPTS.
+
+* cyls, head, secs and trans go into HOST-OPTS.  Future work: they
+  should go into DEV-OPTS instead.
+
+* serial goes into DEV-OPTS, for devices supporting serial numbers.
+  For other devices, it goes nowhere.
+
+* media is special.  In the old way, it selects disk vs. CD-ROM with
+  if=ide, if=scsi and if=xen.  The new way uses DEVNAME for that.
+  Additionally, readonly=on goes into HOST-OPTS.
+
+* addr is special, see if=virtio below.
+
+The -device argument differs in detail for each type of drive:
 
 * if=ide
 
-  -device ide-drive,drive=DRIVE-ID,bus=IDE-BUS,unit=UNIT
+  -device DEVNAME,drive=DRIVE-ID,bus=IDE-BUS,unit=UNIT
 
-  where IDE-BUS identifies an IDE bus, normally either ide.0 or ide.1,
-  and UNIT is either 0 or 1.
-
-  Bug: new way does not work for ide.1 unit 0 (in old terms: index=2)
-  unless you disable the default CD-ROM with -nodefaults.
+  where DEVNAME is either ide-hd or ide-cd, IDE-BUS identifies an IDE
+  bus, normally either ide.0 or ide.1, and UNIT is either 0 or 1.
 
 * if=scsi
 
@@ -77,27 +95,25 @@
   As for all PCI devices, you can add bus=PCI-BUS,addr=DEVFN to
   control the PCI device address.
 
-  This SCSI controller a single SCSI bus, named ID.0.  Put a disk on
-  it:
+  This SCSI controller provides a single SCSI bus, named ID.0.  Put a
+  disk on it:
 
-  -device scsi-disk,drive=DRIVE-ID,bus=ID.0,scsi-id=SCSI-ID,removable=RMB
+  -device DEVNAME,drive=DRIVE-ID,bus=ID.0,scsi-id=UNIT
 
-  The (optional) removable parameter lets you override the SCSI INQUIRY
-  removable (RMB) bit for non CD-ROM devices.  It is ignored for CD-ROM devices
-  which are always removable.  RMB is "on" or "off".
+  where DEVNAME is either scsi-hd, scsi-cd or scsi-generic.
 
 * if=floppy
 
-  -global isa-fdc,driveA=DRIVE-ID,driveB=DRIVE-ID
+  -global isa-fdc.driveA=DRIVE-ID
+  -global isa-fdc.driveB=DRIVE-ID
 
   This is -global instead of -device, because the floppy controller is
   created automatically, and we want to configure that one, not create
   a second one (which isn't possible anyway).
 
-  Omitting a drive parameter makes that drive empty.
-
-  Bug: driveA works only if you disable the default floppy drive with
-  -nodefaults.
+  Without any -global isa-fdc,... you get an empty driveA and no
+  driveB.  You can use -nodefaults to suppress the default driveA, see
+  "Default Devices".
 
 * if=virtio
 
@@ -105,11 +121,12 @@
 
   This lets you control PCI device class and MSI-X vectors.
 
-  IOEVENTFD controls whether or not ioeventfd is used for virtqueue notify.  It
-  can be set to on (default) or off.
+  IOEVENTFD controls whether or not ioeventfd is used for virtqueue
+  notify.  It can be set to on (default) or off.
 
   As for all PCI devices, you can add bus=PCI-BUS,addr=DEVFN to
-  control the PCI device address.
+  control the PCI device address.  This replaces option addr available
+  with -drive if=virtio.
 
 * if=pflash, if=mtd, if=sd, if=xen are not yet available with -device
 
@@ -117,15 +134,20 @@
 
     -usbdevice disk:format=FMT:FILENAME
 
-Provides much less control than -drive's HOST-OPTS...  The new way
-fixes that:
+Provides much less control than -drive's OPTS...  The new way fixes
+that:
 
     -device usb-storage,drive=DRIVE-ID,removable=RMB
 
-The removable parameter gives control over the SCSI INQUIRY removable (RMB)
-bit.  USB thumbdrives usually set removable=on, while USB hard disks set
-removable=off.  See the if=scsi description above for details on the removable
-parameter, which applies only to scsi-disk devices and not to scsi-generic.
+The removable parameter gives control over the SCSI INQUIRY removable
+(RMB) bit.  USB thumbdrives usually set removable=on, while USB hard
+disks set removable=off.
+
+Bug: usb-storage pretends to be a block device, but it's really a SCSI
+controller that can serve only a single device, which it creates
+automatically.  The automatic creation guesses what kind of guest part
+to create from the host part, like -drive if=scsi.  Host and guest
+part are not cleanly separated.
 
 === Character Devices ===
 
@@ -170,7 +192,9 @@
   -device usb-braille,chardev=braille,vendorid=VID,productid=PRID
   -chardev braille,id=braille
 
-* -virtioconsole is still being worked on
+* -virtioconsole becomes
+  -device virtio-serial-pci,class=C,vectors=V,ioeventfd=IOEVENTFD,max_ports=N
+  -device virtconsole,is_console=NUM,nr=NR,name=NAME
 
 LEGACY-CHARDEV translates to -chardev HOST-OPTS... as follows:
 
@@ -219,38 +243,29 @@
 
 === Network Devices ===
 
-A QEMU network device (NIC) has a host and a guest part.
+Host and guest part of network devices have always been separate.
 
-The old ways to define NICs define host and guest part together.  It
-looks like this:
+The old way to define the guest part looks like this:
 
-    -net nic,vlan=VLAN,macaddr=MACADDR,model=MODEL,name=ID,addr=STR,vectors=V
+    -net nic,netdev=NET-ID,macaddr=MACADDR,model=MODEL,name=ID,addr=STR,vectors=V
 
 Except for USB it looks like this:
 
-    -usbdevice net:vlan=VLAN,macaddr=MACADDR,name=ID,addr=STR,vectors=V
+    -usbdevice net:netdev=NET-ID,macaddr=MACADDR,name=ID
 
-The new way keeps the parts separate: you create the host part with
--netdev, and the guest device with -device, like this:
+The new way is -device:
 
-    -netdev type=TYPE,id=NET-ID
     -device DEVNAME,netdev=NET-ID,mac=MACADDR,DEV-OPTS...
 
-Unlike the old way, this creates just a network device, not a VLAN.
-If you really want a VLAN, create it the usual way, then create the
-guest device like this:
-
-    -device DEVNAME,vlan=VLAN,mac=MACADDR,DEV-OPTS...
-
 DEVNAME equals MODEL, except for virtio you have to name the virtio
 device appropriate for the bus (virtio-net-pci for PCI), and for USB
-NIC you have to use usb-net.
+you have to use usb-net.
 
 The old name=ID parameter becomes the usual id=ID with -device.
 
 For PCI devices, you can add bus=PCI-BUS,addr=DEVFN to control the PCI
 device address, as usual.  The old -net nic provides parameter addr
-for that, it is silently ignored when the NIC is not a PCI device.
+for that, which is silently ignored when the NIC is not a PCI device.
 
 For virtio-net-pci, you can control whether or not ioeventfd is used for
 virtqueue notify by setting ioeventfd= to on or off (default).
@@ -264,20 +279,25 @@
 
 Some PCI devices aren't available with -net nic, e.g. i82558a.
 
-Bug: usb-net does not work, yet.  Patch posted.
+To connect to a VLAN instead of an ordinary host part, replace
+netdev=NET-ID by vlan=VLAN.
 
 === Graphics Devices ===
 
 Host and guest part of graphics devices have always been separate.
 
-The old way to define the guest graphics device is -vga VGA.
+The old way to define the guest graphics device is -vga VGA.  Not all
+machines support all -vga options.
 
-The new way is -device.  Map from -vga argument to -device:
+The new way is -device.  The mapping from -vga argument to -device
+depends on the machine type.  For machine "pc", it's:
 
     std         -device VGA
     cirrus      -device cirrus-vga
     vmware      -device vmware-svga
-    xenfb       not yet available with -device
+    qxl         -device qxl-vga
+    none        -nodefaults
+                disables more than just VGA, see "Default Devices"
 
 As for all PCI devices, you can add bus=PCI-BUS,addr=DEVFN to control
 the PCI device address.
@@ -285,13 +305,16 @@
 -device VGA supports properties bios-offset and bios-size, but they
 aren't used with machine type "pc".
 
-Bug: -device cirrus-vga and -device vmware-svga require -nodefaults.
+For machine "isapc", it's
 
-Bug: the new way requires PCI; ISA VGA is not yet available with
--device.
+    std         -device isa-vga
+    cirrus      not yet available with -device
+    none        -nodefaults
+                disables more than just VGA, see "Default Devices"
 
-Bug: the new way doesn't work for machine type "pc", because it
-violates obscure device initialization ordering constraints.
+Bug: the new way doesn't work for machine types "pc" and "isapc",
+because it violates obscure device initialization ordering
+constraints.
 
 === Audio Devices ===
 
@@ -308,6 +331,7 @@
     cs4231a     -device cs4231a,iobase=IOADDR,irq=IRQ,dma=DMA
     es1370      -device ES1370
     gus         -device gus,iobase=IOADDR,irq=IRQ,dma=DMA,freq=F
+    hda         -device intel-hda,msi=MSI -device hda-duplex
     sb16        -device sb16,iobase=IOADDR,irq=IRQ,dma=DMA,dma16=DMA16,version=V
     adlib       not yet available with -device
     pcspk       not yet available with -device
@@ -321,9 +345,10 @@
 
 The new way is -device DEVNAME,DEV-OPTS...  Details depend on DRIVER:
 
+* ccid            -device usb-ccid
+* keyboard        -device usb-kbd
 * mouse           -device usb-mouse
 * tablet          -device usb-tablet
-* keyboard        -device usb-kdb
 * wacom-tablet    -device usb-wacom-tablet
 * host:...        See "Host Device Assignment"
 * disk:...        See "Block Devices"
@@ -353,7 +378,7 @@
 
     -device pci-assign,host=ADDR,iommu=IOMMU,id=ID
 
-The old dma=none becomes iommu=0 with -device.
+The old dma=none becomes iommu=off with -device.
 
 The old way to assign a host USB device is
 
@@ -365,4 +390,27 @@
 
     -device usb-host,hostbus=BUS,hostaddr=ADDR,vendorid=VID,productid=PRID
 
-where left out or zero BUS, ADDR, VID, PRID serve as wildcard.
+Omitted options match anything, just like the old way's wildcard.
+
+=== Default Devices ===
+
+QEMU creates a number of devices by default, depending on the machine
+type.
+
+-device DEVNAME... and global DEVNAME... suppress default devices for
+some DEVNAMEs:
+
+    default device      suppressing DEVNAMEs
+    CD-ROM              ide-cd, ide-drive, scsi-cd
+    isa-fdc's driveA    isa-fdc
+    parallel            isa-parallel
+    serial              isa-serial
+    VGA                 VGA, cirrus-vga, vmware-svga
+    virtioconsole       virtio-serial-pci, virtio-serial-s390, virtio-serial
+
+The default NIC is connected to a default part created along with it.
+It is *not* suppressed by configuring a NIC with -device (you may call
+that a bug).  -net and -netdev suppress the default NIC.
+
+-nodefaults suppresses all the default devices mentioned above, plus a
+few other things such as default SD-Card drive and default monitor.
diff --git a/docs/usb2.txt b/docs/usb2.txt
index b283c13..5950c71 100644
--- a/docs/usb2.txt
+++ b/docs/usb2.txt
@@ -31,6 +31,91 @@
 This attaches a usb tablet to the UHCI adapter and a usb mass storage
 device to the EHCI adapter.
 
+
+More USB tips & tricks
+======================
+
+Recently the usb pass through driver (also known as usb-host) and the
+qemu usb subsystem gained a few capabilities which are available only
+via qdev properties, i,e. when using '-device'.
+
+
+physical port addressing
+------------------------
+
+First you can (for all usb devices) specify the physical port where
+the device will show up in the guest.  This can be done using the
+"port" property.  UHCI has two root ports (1,2).  EHCI has four root
+ports (1-4), the emulated (1.1) USB hub has eight ports.
+
+Plugging a tablet into UHCI port 1 works like this:
+
+        -device usb-tablet,bus=usb.0,port=1
+
+Plugging a hub into UHCI port 2 works like this:
+
+        -device usb-hub,bus=usb.0,port=2
+
+Plugging a virtual usb stick into port 4 of the hub just plugged works
+this way:
+
+        -device usb-storage,bus=usb.0,port=2.4,drive=...
+
+You can do basically the same in the monitor using the device_add
+command.  If you want to unplug devices too you should specify some
+unique id which you can use to refer to the device ...
+
+        (qemu) device_add usb-tablet,bus=usb.0,port=1,id=my-tablet
+        (qemu) device_del my-tablet
+
+... when unplugging it with device_del.
+
+
+USB pass through hints
+----------------------
+
+The usb-host driver has a bunch of properties to specify the device
+which should be passed to the guest:
+
+  hostbus=<nr> -- Specifies the bus number the device must be attached
+  to.
+
+  hostaddr=<nr> -- Specifies the device address the device got
+  assigned by the guest os.
+
+  hostport=<str> -- Specifies the physical port the device is attached
+  to.
+
+  vendorid=<hexnr> -- Specifies the vendor ID of the device.
+  productid=<hexnr> -- Specifies the product ID of the device.
+
+In theory you can combine all these properties as you like.  In
+practice only a few combinations are useful:
+
+  (1) vendorid+productid -- match for a specific device, pass it to
+      the guest when it shows up somewhere in the host.
+
+  (2) hostbus+hostport -- match for a specific physical port in the
+      host, any device which is plugged in there gets passed to the
+      guest.
+
+  (3) hostbus+hostaddr -- most useful for ad-hoc pass through as the
+      hostaddr isn't stable, the next time you plug in the device it
+      gets a new one ...
+
+Note that USB 1.1 devices are handled by UHCI/OHCI and USB 2.0 by
+EHCI.  That means a device plugged into the very same physical port
+may show up on different busses depending on the speed.  The port I'm
+using for testing is bus 1 + port 1 for 2.0 devices and bus 3 + port 1
+for 1.1 devices.  Passing through any device plugged into that port
+and also assign them to the correct bus can be done this way:
+
+    qemu -M pc ${otheroptions}                           \
+        -usb                                             \
+        -device usb-ehci,id=ehci                         \
+        -device usb-host,bus=usb.0,hostbus=3,hostport=1  \
+        -device usb-host,bus=ehci.0,hostbus=1,hostport=1
+
 enjoy,
   Gerd
 
diff --git a/error.c b/error.c
new file mode 100644
index 0000000..867eec2
--- /dev/null
+++ b/error.c
@@ -0,0 +1,140 @@
+/*
+ * QEMU Error Objects
+ *
+ * Copyright IBM, Corp. 2011
+ *
+ * Authors:
+ *  Anthony Liguori   <aliguori@us.ibm.com>
+ *
+ * This work is licensed under the terms of the GNU LGPL, version 2.  See
+ * the COPYING.LIB file in the top-level directory.
+ */
+#include "error.h"
+#include "error_int.h"
+#include "qemu-objects.h"
+#include "qerror.h"
+#include <assert.h>
+
+struct Error
+{
+    QDict *obj;
+    const char *fmt;
+    char *msg;
+};
+
+void error_set(Error **errp, const char *fmt, ...)
+{
+    Error *err;
+    va_list ap;
+
+    if (errp == NULL) {
+        return;
+    }
+
+    err = qemu_mallocz(sizeof(*err));
+
+    va_start(ap, fmt);
+    err->obj = qobject_to_qdict(qobject_from_jsonv(fmt, &ap));
+    va_end(ap);
+    err->fmt = fmt;
+
+    *errp = err;
+}
+
+bool error_is_set(Error **errp)
+{
+    return (errp && *errp);
+}
+
+const char *error_get_pretty(Error *err)
+{
+    if (err->msg == NULL) {
+        QString *str;
+        str = qerror_format(err->fmt, err->obj);
+        err->msg = qemu_strdup(qstring_get_str(str));
+        QDECREF(str);
+    }
+
+    return err->msg;
+}
+
+const char *error_get_field(Error *err, const char *field)
+{
+    if (strcmp(field, "class") == 0) {
+        return qdict_get_str(err->obj, field);
+    } else {
+        QDict *dict = qdict_get_qdict(err->obj, "data");
+        return qdict_get_str(dict, field);
+    }
+}
+
+QDict *error_get_data(Error *err)
+{
+    QDict *data = qdict_get_qdict(err->obj, "data");
+    QINCREF(data);
+    return data;
+}
+
+void error_set_field(Error *err, const char *field, const char *value)
+{
+    QDict *dict = qdict_get_qdict(err->obj, "data");
+    return qdict_put(dict, field, qstring_from_str(value));
+}
+
+void error_free(Error *err)
+{
+    if (err) {
+        QDECREF(err->obj);
+        qemu_free(err->msg);
+        qemu_free(err);
+    }
+}
+
+bool error_is_type(Error *err, const char *fmt)
+{
+    const char *error_class;
+    char *ptr;
+    char *end;
+
+    ptr = strstr(fmt, "'class': '");
+    assert(ptr != NULL);
+    ptr += strlen("'class': '");
+
+    end = strchr(ptr, '\'');
+    assert(end != NULL);
+
+    error_class = error_get_field(err, "class");
+    if (strlen(error_class) != end - ptr) {
+        return false;
+    }
+
+    return strncmp(ptr, error_class, end - ptr) == 0;
+}
+
+void error_propagate(Error **dst_err, Error *local_err)
+{
+    if (dst_err) {
+        *dst_err = local_err;
+    } else if (local_err) {
+        error_free(local_err);
+    }
+}
+
+QObject *error_get_qobject(Error *err)
+{
+    QINCREF(err->obj);
+    return QOBJECT(err->obj);
+}
+
+void error_set_qobject(Error **errp, QObject *obj)
+{
+    Error *err;
+    if (errp == NULL) {
+        return;
+    }
+    err = qemu_mallocz(sizeof(*err));
+    err->obj = qobject_to_qdict(obj);
+    qobject_incref(obj);
+
+    *errp = err;
+}
diff --git a/error.h b/error.h
new file mode 100644
index 0000000..003c855
--- /dev/null
+++ b/error.h
@@ -0,0 +1,70 @@
+/*
+ * QEMU Error Objects
+ *
+ * Copyright IBM, Corp. 2011
+ *
+ * Authors:
+ *  Anthony Liguori   <aliguori@us.ibm.com>
+ *
+ * This work is licensed under the terms of the GNU LGPL, version 2.  See
+ * the COPYING.LIB file in the top-level directory.
+ */
+#ifndef ERROR_H
+#define ERROR_H
+
+#include <stdbool.h>
+
+/**
+ * A class representing internal errors within QEMU.  An error has a string
+ * typename and optionally a set of named string parameters.
+ */
+typedef struct Error Error;
+
+/**
+ * Set an indirect pointer to an error given a printf-style format parameter.
+ * Currently, qerror.h defines these error formats.  This function is not
+ * meant to be used outside of QEMU.
+ */
+void error_set(Error **err, const char *fmt, ...)
+    __attribute__((format(printf, 2, 3)));
+
+/**
+ * Returns true if an indirect pointer to an error is pointing to a valid
+ * error object.
+ */
+bool error_is_set(Error **err);
+
+/**
+ * Get a human readable representation of an error object.
+ */
+const char *error_get_pretty(Error *err);
+
+/**
+ * Get an individual named error field.
+ */
+const char *error_get_field(Error *err, const char *field);
+
+/**
+ * Get an individual named error field.
+ */
+void error_set_field(Error *err, const char *field, const char *value);
+
+/**
+ * Propagate an error to an indirect pointer to an error.  This function will
+ * always transfer ownership of the error reference and handles the case where
+ * dst_err is NULL correctly.
+ */
+void error_propagate(Error **dst_err, Error *local_err);
+
+/**
+ * Free an error object.
+ */
+void error_free(Error *err);
+
+/**
+ * Determine if an error is of a speific type (based on the qerror format).
+ * Non-QEMU users should get the `class' field to identify the error type.
+ */
+bool error_is_type(Error *err, const char *fmt);
+
+#endif
diff --git a/error_int.h b/error_int.h
new file mode 100644
index 0000000..5e39424
--- /dev/null
+++ b/error_int.h
@@ -0,0 +1,29 @@
+/*
+ * QEMU Error Objects
+ *
+ * Copyright IBM, Corp. 2011
+ *
+ * Authors:
+ *  Anthony Liguori   <aliguori@us.ibm.com>
+ *
+ * This work is licensed under the terms of the GNU LGPL, version 2.  See
+ * the COPYING.LIB file in the top-level directory.
+ */
+#ifndef QEMU_ERROR_INT_H
+#define QEMU_ERROR_INT_H
+
+#include "qemu-common.h"
+#include "qobject.h"
+#include "qdict.h"
+#include "error.h"
+
+/**
+ * Internal QEMU functions for working with Error.
+ *
+ * These are used to convert QErrors to Errors
+ */
+QDict *error_get_data(Error *err);
+QObject *error_get_qobject(Error *err);
+void error_set_qobject(Error **errp, QObject *obj);
+  
+#endif
diff --git a/exec-all.h b/exec-all.h
index 026864e..2a13a95 100644
--- a/exec-all.h
+++ b/exec-all.h
@@ -325,7 +325,7 @@
     }
     pd = env1->tlb_table[mmu_idx][page_index].addr_code & ~TARGET_PAGE_MASK;
     if (pd > IO_MEM_ROM && !(pd & IO_MEM_ROMD)) {
-#if defined(TARGET_SPARC) || defined(TARGET_MIPS)
+#if defined(TARGET_ALPHA) || defined(TARGET_MIPS) || defined(TARGET_SPARC)
         do_unassigned_access(addr, 0, 1, 0, 4);
 #else
         cpu_abort(env1, "Trying to execute code outside RAM or ROM at 0x" TARGET_FMT_lx "\n", addr);
diff --git a/exec.c b/exec.c
index 8529390..09928a3 100644
--- a/exec.c
+++ b/exec.c
@@ -36,7 +36,6 @@
 #include "qemu-timer.h"
 #if defined(CONFIG_USER_ONLY)
 #include <qemu.h>
-#include <signal.h>
 #if defined(__FreeBSD__) || defined(__FreeBSD_kernel__)
 #include <sys/param.h>
 #if __FreeBSD_version >= 700104
@@ -2952,6 +2951,19 @@
     return qemu_ram_alloc_from_ptr(dev, name, size, NULL);
 }
 
+void qemu_ram_free_from_ptr(ram_addr_t addr)
+{
+    RAMBlock *block;
+
+    QLIST_FOREACH(block, &ram_list.blocks, next) {
+        if (addr == block->offset) {
+            QLIST_REMOVE(block, next);
+            qemu_free(block);
+            return;
+        }
+    }
+}
+
 void qemu_ram_free(ram_addr_t addr)
 {
     RAMBlock *block;
@@ -3181,7 +3193,7 @@
 #ifdef DEBUG_UNASSIGNED
     printf("Unassigned mem read " TARGET_FMT_plx "\n", addr);
 #endif
-#if defined(TARGET_SPARC) || defined(TARGET_MICROBLAZE)
+#if defined(TARGET_ALPHA) || defined(TARGET_SPARC) || defined(TARGET_MICROBLAZE)
     do_unassigned_access(addr, 0, 0, 0, 1);
 #endif
     return 0;
@@ -3192,7 +3204,7 @@
 #ifdef DEBUG_UNASSIGNED
     printf("Unassigned mem read " TARGET_FMT_plx "\n", addr);
 #endif
-#if defined(TARGET_SPARC) || defined(TARGET_MICROBLAZE)
+#if defined(TARGET_ALPHA) || defined(TARGET_SPARC) || defined(TARGET_MICROBLAZE)
     do_unassigned_access(addr, 0, 0, 0, 2);
 #endif
     return 0;
@@ -3203,7 +3215,7 @@
 #ifdef DEBUG_UNASSIGNED
     printf("Unassigned mem read " TARGET_FMT_plx "\n", addr);
 #endif
-#if defined(TARGET_SPARC) || defined(TARGET_MICROBLAZE)
+#if defined(TARGET_ALPHA) || defined(TARGET_SPARC) || defined(TARGET_MICROBLAZE)
     do_unassigned_access(addr, 0, 0, 0, 4);
 #endif
     return 0;
@@ -3214,7 +3226,7 @@
 #ifdef DEBUG_UNASSIGNED
     printf("Unassigned mem write " TARGET_FMT_plx " = 0x%x\n", addr, val);
 #endif
-#if defined(TARGET_SPARC) || defined(TARGET_MICROBLAZE)
+#if defined(TARGET_ALPHA) || defined(TARGET_SPARC) || defined(TARGET_MICROBLAZE)
     do_unassigned_access(addr, 1, 0, 0, 1);
 #endif
 }
@@ -3224,7 +3236,7 @@
 #ifdef DEBUG_UNASSIGNED
     printf("Unassigned mem write " TARGET_FMT_plx " = 0x%x\n", addr, val);
 #endif
-#if defined(TARGET_SPARC) || defined(TARGET_MICROBLAZE)
+#if defined(TARGET_ALPHA) || defined(TARGET_SPARC) || defined(TARGET_MICROBLAZE)
     do_unassigned_access(addr, 1, 0, 0, 2);
 #endif
 }
@@ -3234,7 +3246,7 @@
 #ifdef DEBUG_UNASSIGNED
     printf("Unassigned mem write " TARGET_FMT_plx " = 0x%x\n", addr, val);
 #endif
-#if defined(TARGET_SPARC) || defined(TARGET_MICROBLAZE)
+#if defined(TARGET_ALPHA) || defined(TARGET_SPARC) || defined(TARGET_MICROBLAZE)
     do_unassigned_access(addr, 1, 0, 0, 4);
 #endif
 }
diff --git a/fpu/softfloat-native.c b/fpu/softfloat-native.c
deleted file mode 100644
index 8848651..0000000
--- a/fpu/softfloat-native.c
+++ /dev/null
@@ -1,540 +0,0 @@
-/* Native implementation of soft float functions. Only a single status
-   context is supported */
-#include "softfloat.h"
-#include <math.h>
-#if defined(CONFIG_SOLARIS)
-#include <fenv.h>
-#endif
-
-void set_float_rounding_mode(int val STATUS_PARAM)
-{
-    STATUS(float_rounding_mode) = val;
-#if (defined(CONFIG_BSD) && !defined(__APPLE__) && !defined(__GLIBC__)) || \
-    (defined(CONFIG_SOLARIS) && CONFIG_SOLARIS_VERSION < 10)
-    fpsetround(val);
-#else
-    fesetround(val);
-#endif
-}
-
-#ifdef FLOATX80
-void set_floatx80_rounding_precision(int val STATUS_PARAM)
-{
-    STATUS(floatx80_rounding_precision) = val;
-}
-#endif
-
-#if defined(CONFIG_BSD) || \
-    (defined(CONFIG_SOLARIS) && CONFIG_SOLARIS_VERSION < 10)
-#define lrint(d)		((int32_t)rint(d))
-#define llrint(d)		((int64_t)rint(d))
-#define lrintf(f)		((int32_t)rint(f))
-#define llrintf(f)		((int64_t)rint(f))
-#define sqrtf(f)		((float)sqrt(f))
-#define remainderf(fa, fb)	((float)remainder(fa, fb))
-#define rintf(f)		((float)rint(f))
-#if !defined(__sparc__) && \
-    (defined(CONFIG_SOLARIS) && CONFIG_SOLARIS_VERSION < 10)
-extern long double rintl(long double);
-extern long double scalbnl(long double, int);
-
-long long
-llrintl(long double x) {
-	return ((long long) rintl(x));
-}
-
-long
-lrintl(long double x) {
-	return ((long) rintl(x));
-}
-
-long double
-ldexpl(long double x, int n) {
-	return (scalbnl(x, n));
-}
-#endif
-#endif
-
-#if defined(_ARCH_PPC)
-
-/* correct (but slow) PowerPC rint() (glibc version is incorrect) */
-static double qemu_rint(double x)
-{
-    double y = 4503599627370496.0;
-    if (fabs(x) >= y)
-        return x;
-    if (x < 0)
-        y = -y;
-    y = (x + y) - y;
-    if (y == 0.0)
-        y = copysign(y, x);
-    return y;
-}
-
-#define rint qemu_rint
-#endif
-
-/*----------------------------------------------------------------------------
-| Software IEC/IEEE integer-to-floating-point conversion routines.
-*----------------------------------------------------------------------------*/
-float32 int32_to_float32(int v STATUS_PARAM)
-{
-    return (float32)v;
-}
-
-float32 uint32_to_float32(unsigned int v STATUS_PARAM)
-{
-    return (float32)v;
-}
-
-float64 int32_to_float64(int v STATUS_PARAM)
-{
-    return (float64)v;
-}
-
-float64 uint32_to_float64(unsigned int v STATUS_PARAM)
-{
-    return (float64)v;
-}
-
-#ifdef FLOATX80
-floatx80 int32_to_floatx80(int v STATUS_PARAM)
-{
-    return (floatx80)v;
-}
-#endif
-float32 int64_to_float32( int64_t v STATUS_PARAM)
-{
-    return (float32)v;
-}
-float32 uint64_to_float32( uint64_t v STATUS_PARAM)
-{
-    return (float32)v;
-}
-float64 int64_to_float64( int64_t v STATUS_PARAM)
-{
-    return (float64)v;
-}
-float64 uint64_to_float64( uint64_t v STATUS_PARAM)
-{
-    return (float64)v;
-}
-#ifdef FLOATX80
-floatx80 int64_to_floatx80( int64_t v STATUS_PARAM)
-{
-    return (floatx80)v;
-}
-#endif
-
-/* XXX: this code implements the x86 behaviour, not the IEEE one.  */
-#if HOST_LONG_BITS == 32
-static inline int long_to_int32(long a)
-{
-    return a;
-}
-#else
-static inline int long_to_int32(long a)
-{
-    if (a != (int32_t)a)
-        a = 0x80000000;
-    return a;
-}
-#endif
-
-/*----------------------------------------------------------------------------
-| Software IEC/IEEE single-precision conversion routines.
-*----------------------------------------------------------------------------*/
-int float32_to_int32( float32 a STATUS_PARAM)
-{
-    return long_to_int32(lrintf(a));
-}
-int float32_to_int32_round_to_zero( float32 a STATUS_PARAM)
-{
-    return (int)a;
-}
-int64_t float32_to_int64( float32 a STATUS_PARAM)
-{
-    return llrintf(a);
-}
-
-int64_t float32_to_int64_round_to_zero( float32 a STATUS_PARAM)
-{
-    return (int64_t)a;
-}
-
-float64 float32_to_float64( float32 a STATUS_PARAM)
-{
-    return a;
-}
-#ifdef FLOATX80
-floatx80 float32_to_floatx80( float32 a STATUS_PARAM)
-{
-    return a;
-}
-#endif
-
-unsigned int float32_to_uint32( float32 a STATUS_PARAM)
-{
-    int64_t v;
-    unsigned int res;
-
-    v = llrintf(a);
-    if (v < 0) {
-        res = 0;
-    } else if (v > 0xffffffff) {
-        res = 0xffffffff;
-    } else {
-        res = v;
-    }
-    return res;
-}
-unsigned int float32_to_uint32_round_to_zero( float32 a STATUS_PARAM)
-{
-    int64_t v;
-    unsigned int res;
-
-    v = (int64_t)a;
-    if (v < 0) {
-        res = 0;
-    } else if (v > 0xffffffff) {
-        res = 0xffffffff;
-    } else {
-        res = v;
-    }
-    return res;
-}
-
-/*----------------------------------------------------------------------------
-| Software IEC/IEEE single-precision operations.
-*----------------------------------------------------------------------------*/
-float32 float32_round_to_int( float32 a STATUS_PARAM)
-{
-    return rintf(a);
-}
-
-float32 float32_rem( float32 a, float32 b STATUS_PARAM)
-{
-    return remainderf(a, b);
-}
-
-float32 float32_sqrt( float32 a STATUS_PARAM)
-{
-    return sqrtf(a);
-}
-int float32_compare( float32 a, float32 b STATUS_PARAM )
-{
-    if (a < b) {
-        return float_relation_less;
-    } else if (a == b) {
-        return float_relation_equal;
-    } else if (a > b) {
-        return float_relation_greater;
-    } else {
-        return float_relation_unordered;
-    }
-}
-int float32_compare_quiet( float32 a, float32 b STATUS_PARAM )
-{
-    if (isless(a, b)) {
-        return float_relation_less;
-    } else if (a == b) {
-        return float_relation_equal;
-    } else if (isgreater(a, b)) {
-        return float_relation_greater;
-    } else {
-        return float_relation_unordered;
-    }
-}
-int float32_is_signaling_nan( float32 a1)
-{
-    float32u u;
-    uint32_t a;
-    u.f = a1;
-    a = u.i;
-    return ( ( ( a>>22 ) & 0x1FF ) == 0x1FE ) && ( a & 0x003FFFFF );
-}
-
-int float32_is_quiet_nan( float32 a1 )
-{
-    float32u u;
-    uint64_t a;
-    u.f = a1;
-    a = u.i;
-    return ( 0xFF800000 < ( a<<1 ) );
-}
-
-int float32_is_any_nan( float32 a1 )
-{
-    float32u u;
-    uint32_t a;
-    u.f = a1;
-    a = u.i;
-    return (a & ~(1 << 31)) > 0x7f800000U;
-}
-
-/*----------------------------------------------------------------------------
-| Software IEC/IEEE double-precision conversion routines.
-*----------------------------------------------------------------------------*/
-int float64_to_int32( float64 a STATUS_PARAM)
-{
-    return long_to_int32(lrint(a));
-}
-int float64_to_int32_round_to_zero( float64 a STATUS_PARAM)
-{
-    return (int)a;
-}
-int64_t float64_to_int64( float64 a STATUS_PARAM)
-{
-    return llrint(a);
-}
-int64_t float64_to_int64_round_to_zero( float64 a STATUS_PARAM)
-{
-    return (int64_t)a;
-}
-float32 float64_to_float32( float64 a STATUS_PARAM)
-{
-    return a;
-}
-#ifdef FLOATX80
-floatx80 float64_to_floatx80( float64 a STATUS_PARAM)
-{
-    return a;
-}
-#endif
-#ifdef FLOAT128
-float128 float64_to_float128( float64 a STATUS_PARAM)
-{
-    return a;
-}
-#endif
-
-unsigned int float64_to_uint32( float64 a STATUS_PARAM)
-{
-    int64_t v;
-    unsigned int res;
-
-    v = llrint(a);
-    if (v < 0) {
-        res = 0;
-    } else if (v > 0xffffffff) {
-        res = 0xffffffff;
-    } else {
-        res = v;
-    }
-    return res;
-}
-unsigned int float64_to_uint32_round_to_zero( float64 a STATUS_PARAM)
-{
-    int64_t v;
-    unsigned int res;
-
-    v = (int64_t)a;
-    if (v < 0) {
-        res = 0;
-    } else if (v > 0xffffffff) {
-        res = 0xffffffff;
-    } else {
-        res = v;
-    }
-    return res;
-}
-uint64_t float64_to_uint64 (float64 a STATUS_PARAM)
-{
-    int64_t v;
-
-    v = llrint(a + (float64)INT64_MIN);
-
-    return v - INT64_MIN;
-}
-uint64_t float64_to_uint64_round_to_zero (float64 a STATUS_PARAM)
-{
-    int64_t v;
-
-    v = (int64_t)(a + (float64)INT64_MIN);
-
-    return v - INT64_MIN;
-}
-
-/*----------------------------------------------------------------------------
-| Software IEC/IEEE double-precision operations.
-*----------------------------------------------------------------------------*/
-#if defined(__sun__) && \
-    (defined(CONFIG_SOLARIS) && CONFIG_SOLARIS_VERSION < 10)
-static inline float64 trunc(float64 x)
-{
-    return x < 0 ? -floor(-x) : floor(x);
-}
-#endif
-float64 float64_trunc_to_int( float64 a STATUS_PARAM )
-{
-    return trunc(a);
-}
-
-float64 float64_round_to_int( float64 a STATUS_PARAM )
-{
-    return rint(a);
-}
-
-float64 float64_rem( float64 a, float64 b STATUS_PARAM)
-{
-    return remainder(a, b);
-}
-
-float64 float64_sqrt( float64 a STATUS_PARAM)
-{
-    return sqrt(a);
-}
-int float64_compare( float64 a, float64 b STATUS_PARAM )
-{
-    if (a < b) {
-        return float_relation_less;
-    } else if (a == b) {
-        return float_relation_equal;
-    } else if (a > b) {
-        return float_relation_greater;
-    } else {
-        return float_relation_unordered;
-    }
-}
-int float64_compare_quiet( float64 a, float64 b STATUS_PARAM )
-{
-    if (isless(a, b)) {
-        return float_relation_less;
-    } else if (a == b) {
-        return float_relation_equal;
-    } else if (isgreater(a, b)) {
-        return float_relation_greater;
-    } else {
-        return float_relation_unordered;
-    }
-}
-int float64_is_signaling_nan( float64 a1)
-{
-    float64u u;
-    uint64_t a;
-    u.f = a1;
-    a = u.i;
-    return
-           ( ( ( a>>51 ) & 0xFFF ) == 0xFFE )
-        && ( a & LIT64( 0x0007FFFFFFFFFFFF ) );
-
-}
-
-int float64_is_quiet_nan( float64 a1 )
-{
-    float64u u;
-    uint64_t a;
-    u.f = a1;
-    a = u.i;
-
-    return ( LIT64( 0xFFF0000000000000 ) < (uint64_t) ( a<<1 ) );
-
-}
-
-int float64_is_any_nan( float64 a1 )
-{
-    float64u u;
-    uint64_t a;
-    u.f = a1;
-    a = u.i;
-
-    return (a & ~(1ULL << 63)) > LIT64 (0x7FF0000000000000 );
-}
-
-#ifdef FLOATX80
-
-/*----------------------------------------------------------------------------
-| Software IEC/IEEE extended double-precision conversion routines.
-*----------------------------------------------------------------------------*/
-int floatx80_to_int32( floatx80 a STATUS_PARAM)
-{
-    return long_to_int32(lrintl(a));
-}
-int floatx80_to_int32_round_to_zero( floatx80 a STATUS_PARAM)
-{
-    return (int)a;
-}
-int64_t floatx80_to_int64( floatx80 a STATUS_PARAM)
-{
-    return llrintl(a);
-}
-int64_t floatx80_to_int64_round_to_zero( floatx80 a STATUS_PARAM)
-{
-    return (int64_t)a;
-}
-float32 floatx80_to_float32( floatx80 a STATUS_PARAM)
-{
-    return a;
-}
-float64 floatx80_to_float64( floatx80 a STATUS_PARAM)
-{
-    return a;
-}
-
-/*----------------------------------------------------------------------------
-| Software IEC/IEEE extended double-precision operations.
-*----------------------------------------------------------------------------*/
-floatx80 floatx80_round_to_int( floatx80 a STATUS_PARAM)
-{
-    return rintl(a);
-}
-floatx80 floatx80_rem( floatx80 a, floatx80 b STATUS_PARAM)
-{
-    return remainderl(a, b);
-}
-floatx80 floatx80_sqrt( floatx80 a STATUS_PARAM)
-{
-    return sqrtl(a);
-}
-int floatx80_compare( floatx80 a, floatx80 b STATUS_PARAM )
-{
-    if (a < b) {
-        return float_relation_less;
-    } else if (a == b) {
-        return float_relation_equal;
-    } else if (a > b) {
-        return float_relation_greater;
-    } else {
-        return float_relation_unordered;
-    }
-}
-int floatx80_compare_quiet( floatx80 a, floatx80 b STATUS_PARAM )
-{
-    if (isless(a, b)) {
-        return float_relation_less;
-    } else if (a == b) {
-        return float_relation_equal;
-    } else if (isgreater(a, b)) {
-        return float_relation_greater;
-    } else {
-        return float_relation_unordered;
-    }
-}
-int floatx80_is_signaling_nan( floatx80 a1)
-{
-    floatx80u u;
-    uint64_t aLow;
-    u.f = a1;
-
-    aLow = u.i.low & ~ LIT64( 0x4000000000000000 );
-    return
-           ( ( u.i.high & 0x7FFF ) == 0x7FFF )
-        && (uint64_t) ( aLow<<1 )
-        && ( u.i.low == aLow );
-}
-
-int floatx80_is_quiet_nan( floatx80 a1 )
-{
-    floatx80u u;
-    u.f = a1;
-    return ( ( u.i.high & 0x7FFF ) == 0x7FFF ) && (uint64_t) ( u.i.low<<1 );
-}
-
-int floatx80_is_any_nan( floatx80 a1 )
-{
-    floatx80u u;
-    u.f = a1;
-    return ((u.i.high & 0x7FFF) == 0x7FFF) && ( u.i.low<<1 );
-}
-
-#endif
diff --git a/fpu/softfloat-native.h b/fpu/softfloat-native.h
deleted file mode 100644
index 6afb74a..0000000
--- a/fpu/softfloat-native.h
+++ /dev/null
@@ -1,531 +0,0 @@
-/* Native implementation of soft float functions */
-#include <math.h>
-
-#if (defined(CONFIG_BSD) && !defined(__APPLE__) && !defined(__GLIBC__)) \
-    || defined(CONFIG_SOLARIS)
-#include <ieeefp.h>
-#define fabsf(f) ((float)fabs(f))
-#else
-#include <fenv.h>
-#endif
-
-#if defined(__OpenBSD__) || defined(__NetBSD__)
-#include <sys/param.h>
-#endif
-
-/*
- * Define some C99-7.12.3 classification macros and
- *        some C99-.12.4 for Solaris systems OS less than 10,
- *        or Solaris 10 systems running GCC 3.x or less.
- *   Solaris 10 with GCC4 does not need these macros as they
- *   are defined in <iso/math_c99.h> with a compiler directive
- */
-#if defined(CONFIG_SOLARIS) && \
-           ((CONFIG_SOLARIS_VERSION <= 9 ) || \
-           ((CONFIG_SOLARIS_VERSION == 10) && (__GNUC__ < 4))) \
-    || (defined(__OpenBSD__) && (OpenBSD < 200811))
-/*
- * C99 7.12.3 classification macros
- * and
- * C99 7.12.14 comparison macros
- *
- * ... do not work on Solaris 10 using GNU CC 3.4.x.
- * Try to workaround the missing / broken C99 math macros.
- */
-#if defined(__OpenBSD__)
-#define unordered(x, y) (isnan(x) || isnan(y))
-#endif
-
-#ifdef __NetBSD__
-#ifndef isgreater
-#define isgreater(x, y)		__builtin_isgreater(x, y)
-#endif
-#ifndef isgreaterequal
-#define isgreaterequal(x, y)	__builtin_isgreaterequal(x, y)
-#endif
-#ifndef isless
-#define isless(x, y)		__builtin_isless(x, y)
-#endif
-#ifndef islessequal
-#define islessequal(x, y)	__builtin_islessequal(x, y)
-#endif
-#ifndef isunordered
-#define isunordered(x, y)	__builtin_isunordered(x, y)
-#endif
-#endif
-
-
-#define isnormal(x)             (fpclass(x) >= FP_NZERO)
-#define isgreater(x, y)         ((!unordered(x, y)) && ((x) > (y)))
-#define isgreaterequal(x, y)    ((!unordered(x, y)) && ((x) >= (y)))
-#define isless(x, y)            ((!unordered(x, y)) && ((x) < (y)))
-#define islessequal(x, y)       ((!unordered(x, y)) && ((x) <= (y)))
-#define isunordered(x,y)        unordered(x, y)
-#endif
-
-#if defined(__sun__) && !defined(CONFIG_NEEDS_LIBSUNMATH)
-
-#ifndef isnan
-# define isnan(x) \
-    (sizeof (x) == sizeof (long double) ? isnan_ld (x) \
-     : sizeof (x) == sizeof (double) ? isnan_d (x) \
-     : isnan_f (x))
-static inline int isnan_f  (float       x) { return x != x; }
-static inline int isnan_d  (double      x) { return x != x; }
-static inline int isnan_ld (long double x) { return x != x; }
-#endif
-
-#ifndef isinf
-# define isinf(x) \
-    (sizeof (x) == sizeof (long double) ? isinf_ld (x) \
-     : sizeof (x) == sizeof (double) ? isinf_d (x) \
-     : isinf_f (x))
-static inline int isinf_f  (float       x) { return isnan (x - x); }
-static inline int isinf_d  (double      x) { return isnan (x - x); }
-static inline int isinf_ld (long double x) { return isnan (x - x); }
-#endif
-#endif
-
-typedef float float32;
-typedef double float64;
-#ifdef FLOATX80
-typedef long double floatx80;
-#endif
-
-typedef union {
-    float32 f;
-    uint32_t i;
-} float32u;
-typedef union {
-    float64 f;
-    uint64_t i;
-} float64u;
-#ifdef FLOATX80
-typedef union {
-    floatx80 f;
-    struct {
-        uint64_t low;
-        uint16_t high;
-    } i;
-} floatx80u;
-#endif
-
-/*----------------------------------------------------------------------------
-| Software IEC/IEEE floating-point rounding mode.
-*----------------------------------------------------------------------------*/
-#if (defined(CONFIG_BSD) && !defined(__APPLE__) && !defined(__GLIBC__)) \
-    || defined(CONFIG_SOLARIS)
-#if defined(__OpenBSD__)
-#define FE_RM FP_RM
-#define FE_RP FP_RP
-#define FE_RZ FP_RZ
-#endif
-enum {
-    float_round_nearest_even = FP_RN,
-    float_round_down         = FP_RM,
-    float_round_up           = FP_RP,
-    float_round_to_zero      = FP_RZ
-};
-#else
-enum {
-    float_round_nearest_even = FE_TONEAREST,
-    float_round_down         = FE_DOWNWARD,
-    float_round_up           = FE_UPWARD,
-    float_round_to_zero      = FE_TOWARDZERO
-};
-#endif
-
-typedef struct float_status {
-    int float_rounding_mode;
-#ifdef FLOATX80
-    int floatx80_rounding_precision;
-#endif
-} float_status;
-
-void set_float_rounding_mode(int val STATUS_PARAM);
-#ifdef FLOATX80
-void set_floatx80_rounding_precision(int val STATUS_PARAM);
-#endif
-
-/*----------------------------------------------------------------------------
-| Software IEC/IEEE integer-to-floating-point conversion routines.
-*----------------------------------------------------------------------------*/
-float32 int32_to_float32( int STATUS_PARAM);
-float32 uint32_to_float32( unsigned int STATUS_PARAM);
-float64 int32_to_float64( int STATUS_PARAM);
-float64 uint32_to_float64( unsigned int STATUS_PARAM);
-#ifdef FLOATX80
-floatx80 int32_to_floatx80( int STATUS_PARAM);
-#endif
-#ifdef FLOAT128
-float128 int32_to_float128( int STATUS_PARAM);
-#endif
-float32 int64_to_float32( int64_t STATUS_PARAM);
-float32 uint64_to_float32( uint64_t STATUS_PARAM);
-float64 int64_to_float64( int64_t STATUS_PARAM);
-float64 uint64_to_float64( uint64_t v STATUS_PARAM);
-#ifdef FLOATX80
-floatx80 int64_to_floatx80( int64_t STATUS_PARAM);
-#endif
-#ifdef FLOAT128
-float128 int64_to_float128( int64_t STATUS_PARAM);
-#endif
-
-/*----------------------------------------------------------------------------
-| Software IEC/IEEE single-precision conversion constants.
-*----------------------------------------------------------------------------*/
-#define float32_zero (0.0)
-#define float32_one (1.0)
-#define float32_ln2 (0.6931471)
-#define float32_pi (3.1415926)
-#define float32_half (0.5)
-
-/*----------------------------------------------------------------------------
-| Software IEC/IEEE single-precision conversion routines.
-*----------------------------------------------------------------------------*/
-int float32_to_int32( float32  STATUS_PARAM);
-int float32_to_int32_round_to_zero( float32  STATUS_PARAM);
-unsigned int float32_to_uint32( float32 a STATUS_PARAM);
-unsigned int float32_to_uint32_round_to_zero( float32 a STATUS_PARAM);
-int64_t float32_to_int64( float32  STATUS_PARAM);
-int64_t float32_to_int64_round_to_zero( float32  STATUS_PARAM);
-float64 float32_to_float64( float32  STATUS_PARAM);
-#ifdef FLOATX80
-floatx80 float32_to_floatx80( float32  STATUS_PARAM);
-#endif
-#ifdef FLOAT128
-float128 float32_to_float128( float32  STATUS_PARAM);
-#endif
-
-/*----------------------------------------------------------------------------
-| Software IEC/IEEE single-precision operations.
-*----------------------------------------------------------------------------*/
-float32 float32_round_to_int( float32  STATUS_PARAM);
-INLINE float32 float32_add( float32 a, float32 b STATUS_PARAM)
-{
-    return a + b;
-}
-INLINE float32 float32_sub( float32 a, float32 b STATUS_PARAM)
-{
-    return a - b;
-}
-INLINE float32 float32_mul( float32 a, float32 b STATUS_PARAM)
-{
-    return a * b;
-}
-INLINE float32 float32_div( float32 a, float32 b STATUS_PARAM)
-{
-    return a / b;
-}
-float32 float32_rem( float32, float32  STATUS_PARAM);
-float32 float32_sqrt( float32  STATUS_PARAM);
-INLINE int float32_eq_quiet( float32 a, float32 b STATUS_PARAM)
-{
-    return a == b;
-}
-INLINE int float32_le( float32 a, float32 b STATUS_PARAM)
-{
-    return a <= b;
-}
-INLINE int float32_lt( float32 a, float32 b STATUS_PARAM)
-{
-    return a < b;
-}
-INLINE int float32_eq( float32 a, float32 b STATUS_PARAM)
-{
-    return a <= b && a >= b;
-}
-INLINE int float32_le_quiet( float32 a, float32 b STATUS_PARAM)
-{
-    return islessequal(a, b);
-}
-INLINE int float32_lt_quiet( float32 a, float32 b STATUS_PARAM)
-{
-    return isless(a, b);
-}
-INLINE int float32_unordered( float32 a, float32 b STATUS_PARAM)
-{
-    return isunordered(a, b);
-}
-INLINE int float32_unordered_quiet( float32 a, float32 b STATUS_PARAM)
-{
-    return isunordered(a, b);
-}
-int float32_compare( float32, float32 STATUS_PARAM );
-int float32_compare_quiet( float32, float32 STATUS_PARAM );
-int float32_is_signaling_nan( float32 );
-int float32_is_quiet_nan( float32 );
-int float32_is_any_nan( float32 );
-
-INLINE float32 float32_abs(float32 a)
-{
-    return fabsf(a);
-}
-
-INLINE float32 float32_chs(float32 a)
-{
-    return -a;
-}
-
-INLINE float32 float32_is_infinity(float32 a)
-{
-    return fpclassify(a) == FP_INFINITE;
-}
-
-INLINE float32 float32_is_neg(float32 a)
-{
-    float32u u;
-    u.f = a;
-    return u.i >> 31;
-}
-
-INLINE float32 float32_is_zero(float32 a)
-{
-    return fpclassify(a) == FP_ZERO;
-}
-
-INLINE float32 float32_scalbn(float32 a, int n STATUS_PARAM)
-{
-    return scalbnf(a, n);
-}
-
-/*----------------------------------------------------------------------------
-| Software IEC/IEEE double-precision conversion constants.
-*----------------------------------------------------------------------------*/
-#define float64_zero (0.0)
-#define float64_one (1.0)
-#define float64_ln2 (0.693147180559945)
-#define float64_pi (3.141592653589793)
-#define float64_half (0.5)
-
-/*----------------------------------------------------------------------------
-| Software IEC/IEEE double-precision conversion routines.
-*----------------------------------------------------------------------------*/
-int float64_to_int32( float64 STATUS_PARAM );
-int float64_to_int32_round_to_zero( float64 STATUS_PARAM );
-unsigned int float64_to_uint32( float64 STATUS_PARAM );
-unsigned int float64_to_uint32_round_to_zero( float64 STATUS_PARAM );
-int64_t float64_to_int64( float64 STATUS_PARAM );
-int64_t float64_to_int64_round_to_zero( float64 STATUS_PARAM );
-uint64_t float64_to_uint64( float64 STATUS_PARAM );
-uint64_t float64_to_uint64_round_to_zero( float64 STATUS_PARAM );
-float32 float64_to_float32( float64 STATUS_PARAM );
-#ifdef FLOATX80
-floatx80 float64_to_floatx80( float64 STATUS_PARAM );
-#endif
-#ifdef FLOAT128
-float128 float64_to_float128( float64 STATUS_PARAM );
-#endif
-
-/*----------------------------------------------------------------------------
-| Software IEC/IEEE double-precision operations.
-*----------------------------------------------------------------------------*/
-float64 float64_round_to_int( float64 STATUS_PARAM );
-float64 float64_trunc_to_int( float64 STATUS_PARAM );
-INLINE float64 float64_add( float64 a, float64 b STATUS_PARAM)
-{
-    return a + b;
-}
-INLINE float64 float64_sub( float64 a, float64 b STATUS_PARAM)
-{
-    return a - b;
-}
-INLINE float64 float64_mul( float64 a, float64 b STATUS_PARAM)
-{
-    return a * b;
-}
-INLINE float64 float64_div( float64 a, float64 b STATUS_PARAM)
-{
-    return a / b;
-}
-float64 float64_rem( float64, float64 STATUS_PARAM );
-float64 float64_sqrt( float64 STATUS_PARAM );
-INLINE int float64_eq_quiet( float64 a, float64 b STATUS_PARAM)
-{
-    return a == b;
-}
-INLINE int float64_le( float64 a, float64 b STATUS_PARAM)
-{
-    return a <= b;
-}
-INLINE int float64_lt( float64 a, float64 b STATUS_PARAM)
-{
-    return a < b;
-}
-INLINE int float64_eq( float64 a, float64 b STATUS_PARAM)
-{
-    return a <= b && a >= b;
-}
-INLINE int float64_le_quiet( float64 a, float64 b STATUS_PARAM)
-{
-    return islessequal(a, b);
-}
-INLINE int float64_lt_quiet( float64 a, float64 b STATUS_PARAM)
-{
-    return isless(a, b);
-
-}
-INLINE int float64_unordered( float64 a, float64 b STATUS_PARAM)
-{
-    return isunordered(a, b);
-}
-INLINE int float64_unordered_quiet( float64 a, float64 b STATUS_PARAM)
-{
-    return isunordered(a, b);
-}
-int float64_compare( float64, float64 STATUS_PARAM );
-int float64_compare_quiet( float64, float64 STATUS_PARAM );
-int float64_is_signaling_nan( float64 );
-int float64_is_any_nan( float64 );
-int float64_is_quiet_nan( float64 );
-
-INLINE float64 float64_abs(float64 a)
-{
-    return fabs(a);
-}
-
-INLINE float64 float64_chs(float64 a)
-{
-    return -a;
-}
-
-INLINE float64 float64_is_infinity(float64 a)
-{
-    return fpclassify(a) == FP_INFINITE;
-}
-
-INLINE float64 float64_is_neg(float64 a)
-{
-    float64u u;
-    u.f = a;
-    return u.i >> 63;
-}
-
-INLINE float64 float64_is_zero(float64 a)
-{
-    return fpclassify(a) == FP_ZERO;
-}
-
-INLINE float64 float64_scalbn(float64 a, int n STATUS_PARAM)
-{
-    return scalbn(a, n);
-}
-
-#ifdef FLOATX80
-
-/*----------------------------------------------------------------------------
-| Software IEC/IEEE extended double-precision conversion constants.
-*----------------------------------------------------------------------------*/
-#define floatx80_zero (0.0L)
-#define floatx80_one (1.0L)
-#define floatx80_ln2 (0.69314718055994530943L)
-#define floatx80_pi (3.14159265358979323851L)
-#define floatx80_half (0.5L)
-
-/*----------------------------------------------------------------------------
-| Software IEC/IEEE extended double-precision conversion routines.
-*----------------------------------------------------------------------------*/
-int floatx80_to_int32( floatx80 STATUS_PARAM );
-int floatx80_to_int32_round_to_zero( floatx80 STATUS_PARAM );
-int64_t floatx80_to_int64( floatx80 STATUS_PARAM);
-int64_t floatx80_to_int64_round_to_zero( floatx80 STATUS_PARAM);
-float32 floatx80_to_float32( floatx80 STATUS_PARAM );
-float64 floatx80_to_float64( floatx80 STATUS_PARAM );
-#ifdef FLOAT128
-float128 floatx80_to_float128( floatx80 STATUS_PARAM );
-#endif
-
-/*----------------------------------------------------------------------------
-| Software IEC/IEEE extended double-precision operations.
-*----------------------------------------------------------------------------*/
-floatx80 floatx80_round_to_int( floatx80 STATUS_PARAM );
-INLINE floatx80 floatx80_add( floatx80 a, floatx80 b STATUS_PARAM)
-{
-    return a + b;
-}
-INLINE floatx80 floatx80_sub( floatx80 a, floatx80 b STATUS_PARAM)
-{
-    return a - b;
-}
-INLINE floatx80 floatx80_mul( floatx80 a, floatx80 b STATUS_PARAM)
-{
-    return a * b;
-}
-INLINE floatx80 floatx80_div( floatx80 a, floatx80 b STATUS_PARAM)
-{
-    return a / b;
-}
-floatx80 floatx80_rem( floatx80, floatx80 STATUS_PARAM );
-floatx80 floatx80_sqrt( floatx80 STATUS_PARAM );
-INLINE int floatx80_eq_quiet( floatx80 a, floatx80 b STATUS_PARAM)
-{
-    return a == b;
-}
-INLINE int floatx80_le( floatx80 a, floatx80 b STATUS_PARAM)
-{
-    return a <= b;
-}
-INLINE int floatx80_lt( floatx80 a, floatx80 b STATUS_PARAM)
-{
-    return a < b;
-}
-INLINE int floatx80_eq( floatx80 a, floatx80 b STATUS_PARAM)
-{
-    return a <= b && a >= b;
-}
-INLINE int floatx80_le_quiet( floatx80 a, floatx80 b STATUS_PARAM)
-{
-    return islessequal(a, b);
-}
-INLINE int floatx80_lt_quiet( floatx80 a, floatx80 b STATUS_PARAM)
-{
-    return isless(a, b);
-
-}
-INLINE int floatx80_unordered( floatx80 a, floatx80 b STATUS_PARAM)
-{
-    return isunordered(a, b);
-}
-INLINE int floatx80_unordered_quiet( floatx80 a, floatx80 b STATUS_PARAM)
-{
-    return isunordered(a, b);
-}
-int floatx80_compare( floatx80, floatx80 STATUS_PARAM );
-int floatx80_compare_quiet( floatx80, floatx80 STATUS_PARAM );
-int floatx80_is_signaling_nan( floatx80 );
-int floatx80_is_quiet_nan( floatx80 );
-int floatx80_is_any_nan( floatx80 );
-
-INLINE floatx80 floatx80_abs(floatx80 a)
-{
-    return fabsl(a);
-}
-
-INLINE floatx80 floatx80_chs(floatx80 a)
-{
-    return -a;
-}
-
-INLINE floatx80 floatx80_is_infinity(floatx80 a)
-{
-    return fpclassify(a) == FP_INFINITE;
-}
-
-INLINE floatx80 floatx80_is_neg(floatx80 a)
-{
-    floatx80u u;
-    u.f = a;
-    return u.i.high >> 15;
-}
-
-INLINE floatx80 floatx80_is_zero(floatx80 a)
-{
-    return fpclassify(a) == FP_ZERO;
-}
-
-INLINE floatx80 floatx80_scalbn(floatx80 a, int n STATUS_PARAM)
-{
-    return scalbnl(a, n);
-}
-
-#endif
diff --git a/fpu/softfloat-specialize.h b/fpu/softfloat-specialize.h
index 9d68aae..c7d35a1 100644
--- a/fpu/softfloat-specialize.h
+++ b/fpu/softfloat-specialize.h
@@ -523,8 +523,6 @@
     }
 }
 
-#ifdef FLOATX80
-
 /*----------------------------------------------------------------------------
 | Returns 1 if the extended double-precision floating-point value `a' is a
 | quiet NaN; otherwise returns 0. This slightly differs from the same
@@ -681,10 +679,6 @@
     }
 }
 
-#endif
-
-#ifdef FLOAT128
-
 /*----------------------------------------------------------------------------
 | Returns 1 if the quadruple-precision floating-point value `a' is a quiet
 | NaN; otherwise returns 0.
@@ -820,4 +814,3 @@
     }
 }
 
-#endif
diff --git a/fpu/softfloat.c b/fpu/softfloat.c
index e3cd8a7..7951a0e 100644
--- a/fpu/softfloat.c
+++ b/fpu/softfloat.c
@@ -64,12 +64,10 @@
     STATUS(float_exception_flags) = val;
 }
 
-#ifdef FLOATX80
 void set_floatx80_rounding_precision(int val STATUS_PARAM)
 {
     STATUS(floatx80_rounding_precision) = val;
 }
-#endif
 
 /*----------------------------------------------------------------------------
 | Returns the fraction bits of the half-precision floating-point value `a'.
@@ -564,8 +562,6 @@
 
 }
 
-#ifdef FLOATX80
-
 /*----------------------------------------------------------------------------
 | Returns the fraction bits of the extended double-precision floating-point
 | value `a'.
@@ -851,10 +847,6 @@
 
 }
 
-#endif
-
-#ifdef FLOAT128
-
 /*----------------------------------------------------------------------------
 | Returns the least-significant 64 fraction bits of the quadruple-precision
 | floating-point value `a'.
@@ -1118,8 +1110,6 @@
 
 }
 
-#endif
-
 /*----------------------------------------------------------------------------
 | Returns the result of converting the 32-bit two's complement integer `a'
 | to the single-precision floating-point format.  The conversion is performed
@@ -1159,8 +1149,6 @@
 
 }
 
-#ifdef FLOATX80
-
 /*----------------------------------------------------------------------------
 | Returns the result of converting the 32-bit two's complement integer `a'
 | to the extended double-precision floating-point format.  The conversion
@@ -1184,10 +1172,6 @@
 
 }
 
-#endif
-
-#ifdef FLOAT128
-
 /*----------------------------------------------------------------------------
 | Returns the result of converting the 32-bit two's complement integer `a' to
 | the quadruple-precision floating-point format.  The conversion is performed
@@ -1210,8 +1194,6 @@
 
 }
 
-#endif
-
 /*----------------------------------------------------------------------------
 | Returns the result of converting the 64-bit two's complement integer `a'
 | to the single-precision floating-point format.  The conversion is performed
@@ -1291,8 +1273,6 @@
 
 }
 
-#ifdef FLOATX80
-
 /*----------------------------------------------------------------------------
 | Returns the result of converting the 64-bit two's complement integer `a'
 | to the extended double-precision floating-point format.  The conversion
@@ -1314,10 +1294,6 @@
 
 }
 
-#endif
-
-#ifdef FLOAT128
-
 /*----------------------------------------------------------------------------
 | Returns the result of converting the 64-bit two's complement integer `a' to
 | the quadruple-precision floating-point format.  The conversion is performed
@@ -1351,8 +1327,6 @@
 
 }
 
-#endif
-
 /*----------------------------------------------------------------------------
 | Returns the result of converting the single-precision floating-point value
 | `a' to the 32-bit two's complement integer format.  The conversion is
@@ -1590,8 +1564,6 @@
 
 }
 
-#ifdef FLOATX80
-
 /*----------------------------------------------------------------------------
 | Returns the result of converting the single-precision floating-point value
 | `a' to the extended double-precision floating-point format.  The conversion
@@ -1622,10 +1594,6 @@
 
 }
 
-#endif
-
-#ifdef FLOAT128
-
 /*----------------------------------------------------------------------------
 | Returns the result of converting the single-precision floating-point value
 | `a' to the double-precision floating-point format.  The conversion is
@@ -1656,8 +1624,6 @@
 
 }
 
-#endif
-
 /*----------------------------------------------------------------------------
 | Rounds the single-precision floating-point value `a' to an integer, and
 | returns the result as a single-precision floating-point value.  The
@@ -2939,8 +2905,6 @@
     return packFloat16(aSign, aExp + 14, aSig >> 13);
 }
 
-#ifdef FLOATX80
-
 /*----------------------------------------------------------------------------
 | Returns the result of converting the double-precision floating-point value
 | `a' to the extended double-precision floating-point format.  The conversion
@@ -2972,10 +2936,6 @@
 
 }
 
-#endif
-
-#ifdef FLOAT128
-
 /*----------------------------------------------------------------------------
 | Returns the result of converting the double-precision floating-point value
 | `a' to the quadruple-precision floating-point format.  The conversion is
@@ -3007,8 +2967,6 @@
 
 }
 
-#endif
-
 /*----------------------------------------------------------------------------
 | Rounds the double-precision floating-point value `a' to an integer, and
 | returns the result as a double-precision floating-point value.  The
@@ -3816,8 +3774,6 @@
     return 0;
 }
 
-#ifdef FLOATX80
-
 /*----------------------------------------------------------------------------
 | Returns the result of converting the extended double-precision floating-
 | point value `a' to the 32-bit two's complement integer format.  The
@@ -4030,8 +3986,6 @@
 
 }
 
-#ifdef FLOAT128
-
 /*----------------------------------------------------------------------------
 | Returns the result of converting the extended double-precision floating-
 | point value `a' to the quadruple-precision floating-point format.  The
@@ -4056,8 +4010,6 @@
 
 }
 
-#endif
-
 /*----------------------------------------------------------------------------
 | Rounds the extended double-precision floating-point value `a' to an integer,
 | and returns the result as an extended quadruple-precision floating-point
@@ -4849,10 +4801,6 @@
     return 0;
 }
 
-#endif
-
-#ifdef FLOAT128
-
 /*----------------------------------------------------------------------------
 | Returns the result of converting the quadruple-precision floating-point
 | value `a' to the 32-bit two's complement integer format.  The conversion
@@ -5102,8 +5050,6 @@
 
 }
 
-#ifdef FLOATX80
-
 /*----------------------------------------------------------------------------
 | Returns the result of converting the quadruple-precision floating-point
 | value `a' to the extended double-precision floating-point format.  The
@@ -5139,8 +5085,6 @@
 
 }
 
-#endif
-
 /*----------------------------------------------------------------------------
 | Rounds the quadruple-precision floating-point value `a' to an integer, and
 | returns the result as a quadruple-precision floating-point value.  The
@@ -6020,8 +5964,6 @@
     return 0;
 }
 
-#endif
-
 /* misc functions */
 float32 uint32_to_float32( unsigned int a STATUS_PARAM )
 {
@@ -6423,7 +6365,6 @@
     return normalizeRoundAndPackFloat64( aSign, aExp, aSig STATUS_VAR );
 }
 
-#ifdef FLOATX80
 floatx80 floatx80_scalbn( floatx80 a, int n STATUS_PARAM )
 {
     flag aSign;
@@ -6454,9 +6395,7 @@
     return normalizeRoundAndPackFloatx80( STATUS(floatx80_rounding_precision),
                                           aSign, aExp, aSig, 0 STATUS_VAR );
 }
-#endif
 
-#ifdef FLOAT128
 float128 float128_scalbn( float128 a, int n STATUS_PARAM )
 {
     flag aSign;
@@ -6489,4 +6428,3 @@
                                           STATUS_VAR );
 
 }
-#endif
diff --git a/fpu/softfloat.h b/fpu/softfloat.h
index 58c9b7b..bde2500 100644
--- a/fpu/softfloat.h
+++ b/fpu/softfloat.h
@@ -74,24 +74,6 @@
 #define SNAN_BIT_IS_ONE		0
 #endif
 
-/*----------------------------------------------------------------------------
-| The macro `FLOATX80' must be defined to enable the extended double-precision
-| floating-point format `floatx80'.  If this macro is not defined, the
-| `floatx80' type will not be defined, and none of the functions that either
-| input or output the `floatx80' type will be defined.  The same applies to
-| the `FLOAT128' macro and the quadruple-precision format `float128'.
-*----------------------------------------------------------------------------*/
-#ifdef CONFIG_SOFTFLOAT
-/* bit exact soft float support */
-#define FLOATX80
-#define FLOAT128
-#else
-/* native float support */
-#if (defined(__i386__) || defined(__x86_64__)) && !defined(CONFIG_BSD)
-#define FLOATX80
-#endif
-#endif /* !CONFIG_SOFTFLOAT */
-
 #define STATUS_PARAM , float_status *status
 #define STATUS(field) status->field
 #define STATUS_VAR , status
@@ -106,7 +88,6 @@
     float_relation_unordered =  2
 };
 
-#ifdef CONFIG_SOFTFLOAT
 /*----------------------------------------------------------------------------
 | Software IEC/IEEE floating-point types.
 *----------------------------------------------------------------------------*/
@@ -149,14 +130,11 @@
 #define const_float32(x) (x)
 #define const_float64(x) (x)
 #endif
-#ifdef FLOATX80
 typedef struct {
     uint64_t low;
     uint16_t high;
 } floatx80;
 #define make_floatx80(exp, mant) ((floatx80) { mant, exp })
-#endif
-#ifdef FLOAT128
 typedef struct {
 #ifdef HOST_WORDS_BIGENDIAN
     uint64_t high, low;
@@ -164,7 +142,6 @@
     uint64_t low, high;
 #endif
 } float128;
-#endif
 
 /*----------------------------------------------------------------------------
 | Software IEC/IEEE floating-point underflow tininess-detection mode.
@@ -201,9 +178,7 @@
     signed char float_detect_tininess;
     signed char float_rounding_mode;
     signed char float_exception_flags;
-#ifdef FLOATX80
     signed char floatx80_rounding_precision;
-#endif
     /* should denormalised results go to zero and set the inexact flag? */
     flag flush_to_zero;
     /* should denormalised inputs go to zero and set the input_denormal flag? */
@@ -233,9 +208,7 @@
 {
     return STATUS(float_exception_flags);
 }
-#ifdef FLOATX80
 void set_floatx80_rounding_precision(int val STATUS_PARAM);
-#endif
 
 /*----------------------------------------------------------------------------
 | Routine to raise any or all of the software IEC/IEEE floating-point
@@ -250,22 +223,14 @@
 float64 int32_to_float64( int32 STATUS_PARAM );
 float32 uint32_to_float32( unsigned int STATUS_PARAM );
 float64 uint32_to_float64( unsigned int STATUS_PARAM );
-#ifdef FLOATX80
 floatx80 int32_to_floatx80( int32 STATUS_PARAM );
-#endif
-#ifdef FLOAT128
 float128 int32_to_float128( int32 STATUS_PARAM );
-#endif
 float32 int64_to_float32( int64 STATUS_PARAM );
 float32 uint64_to_float32( uint64 STATUS_PARAM );
 float64 int64_to_float64( int64 STATUS_PARAM );
 float64 uint64_to_float64( uint64 STATUS_PARAM );
-#ifdef FLOATX80
 floatx80 int64_to_floatx80( int64 STATUS_PARAM );
-#endif
-#ifdef FLOAT128
 float128 int64_to_float128( int64 STATUS_PARAM );
-#endif
 
 /*----------------------------------------------------------------------------
 | Software half-precision conversion routines.
@@ -303,12 +268,8 @@
 int64 float32_to_int64( float32 STATUS_PARAM );
 int64 float32_to_int64_round_to_zero( float32 STATUS_PARAM );
 float64 float32_to_float64( float32 STATUS_PARAM );
-#ifdef FLOATX80
 floatx80 float32_to_floatx80( float32 STATUS_PARAM );
-#endif
-#ifdef FLOAT128
 float128 float32_to_float128( float32 STATUS_PARAM );
-#endif
 
 /*----------------------------------------------------------------------------
 | Software IEC/IEEE single-precision operations.
@@ -420,12 +381,8 @@
 uint64 float64_to_uint64 (float64 a STATUS_PARAM);
 uint64 float64_to_uint64_round_to_zero (float64 a STATUS_PARAM);
 float32 float64_to_float32( float64 STATUS_PARAM );
-#ifdef FLOATX80
 floatx80 float64_to_floatx80( float64 STATUS_PARAM );
-#endif
-#ifdef FLOAT128
 float128 float64_to_float128( float64 STATUS_PARAM );
-#endif
 
 /*----------------------------------------------------------------------------
 | Software IEC/IEEE double-precision operations.
@@ -492,6 +449,11 @@
     return ((float64_val(a) & ~(1ULL << 63)) > 0x7ff0000000000000ULL);
 }
 
+INLINE int float64_is_zero_or_denormal(float64 a)
+{
+    return (float64_val(a) & 0x7ff0000000000000LL) == 0;
+}
+
 INLINE float64 float64_set_sign(float64 a, int sign)
 {
     return make_float64((float64_val(a) & 0x7fffffffffffffffULL)
@@ -518,8 +480,6 @@
 #define float64_default_nan make_float64(LIT64( 0xFFF8000000000000 ))
 #endif
 
-#ifdef FLOATX80
-
 /*----------------------------------------------------------------------------
 | Software IEC/IEEE extended double-precision conversion routines.
 *----------------------------------------------------------------------------*/
@@ -529,9 +489,7 @@
 int64 floatx80_to_int64_round_to_zero( floatx80 STATUS_PARAM );
 float32 floatx80_to_float32( floatx80 STATUS_PARAM );
 float64 floatx80_to_float64( floatx80 STATUS_PARAM );
-#ifdef FLOAT128
 float128 floatx80_to_float128( floatx80 STATUS_PARAM );
-#endif
 
 /*----------------------------------------------------------------------------
 | Software IEC/IEEE extended double-precision operations.
@@ -585,6 +543,11 @@
     return (a.high & 0x7fff) == 0 && a.low == 0;
 }
 
+INLINE int floatx80_is_zero_or_denormal(floatx80 a)
+{
+    return (a.high & 0x7fff) == 0;
+}
+
 INLINE int floatx80_is_any_nan(floatx80 a)
 {
     return ((a.high & 0x7fff) == 0x7fff) && (a.low<<1);
@@ -610,10 +573,6 @@
 #define floatx80_default_nan_low  LIT64( 0xC000000000000000 )
 #endif
 
-#endif
-
-#ifdef FLOAT128
-
 /*----------------------------------------------------------------------------
 | Software IEC/IEEE quadruple-precision conversion routines.
 *----------------------------------------------------------------------------*/
@@ -623,9 +582,7 @@
 int64 float128_to_int64_round_to_zero( float128 STATUS_PARAM );
 float32 float128_to_float32( float128 STATUS_PARAM );
 float64 float128_to_float64( float128 STATUS_PARAM );
-#ifdef FLOATX80
 floatx80 float128_to_floatx80( float128 STATUS_PARAM );
-#endif
 
 /*----------------------------------------------------------------------------
 | Software IEC/IEEE quadruple-precision operations.
@@ -679,6 +636,11 @@
     return (a.high & 0x7fffffffffffffffLL) == 0 && a.low == 0;
 }
 
+INLINE int float128_is_zero_or_denormal(float128 a)
+{
+    return (a.high & 0x7fff000000000000LL) == 0;
+}
+
 INLINE int float128_is_any_nan(float128 a)
 {
     return ((a.high >> 48) & 0x7fff) == 0x7fff &&
@@ -697,12 +659,4 @@
 #define float128_default_nan_low  LIT64( 0x0000000000000000 )
 #endif
 
-#endif
-
-#else /* CONFIG_SOFTFLOAT */
-
-#include "softfloat-native.h"
-
-#endif /* !CONFIG_SOFTFLOAT */
-
 #endif /* !SOFTFLOAT_H */
diff --git a/fsdev/file-op-9p.h b/fsdev/file-op-9p.h
index 126e60e..af9daf7 100644
--- a/fsdev/file-op-9p.h
+++ b/fsdev/file-op-9p.h
@@ -97,11 +97,4 @@
     void *opaque;
 } FileOperations;
 
-static inline const char *rpath(FsContext *ctx, const char *path)
-{
-    /* FIXME: so wrong... */
-    static char buffer[4096];
-    snprintf(buffer, sizeof(buffer), "%s/%s", ctx->fs_root, path);
-    return buffer;
-}
 #endif
diff --git a/fsdev/qemu-fsdev-dummy.c b/fsdev/qemu-fsdev-dummy.c
new file mode 100644
index 0000000..4e700dd
--- /dev/null
+++ b/fsdev/qemu-fsdev-dummy.c
@@ -0,0 +1,28 @@
+/*
+ * Virtio 9p
+ *
+ * Copyright IBM, Corp. 2010
+ *
+ * Authors:
+ *  Gautham R Shenoy <ego@in.ibm.com>
+ *
+ * This work is licensed under the terms of the GNU GPL, version 2.  See
+ * the COPYING file in the top-level directory.
+ *
+ */
+#include <stdio.h>
+#include <string.h>
+#include "qemu-fsdev.h"
+#include "qemu-config.h"
+
+int qemu_fsdev_add(QemuOpts *opts)
+{
+    return 0;
+}
+
+static void fsdev_register_config(void)
+{
+    qemu_add_opts(&qemu_fsdev_opts);
+    qemu_add_opts(&qemu_virtfs_opts);
+}
+machine_init(fsdev_register_config);
diff --git a/gdbstub.c b/gdbstub.c
index ae856f9..b9ae30d 100644
--- a/gdbstub.c
+++ b/gdbstub.c
@@ -1105,10 +1105,6 @@
             env->active_fpu.fcr31 = tmp & 0xFF83FFFF;
             /* set rounding mode */
             RESTORE_ROUNDING_MODE;
-#ifndef CONFIG_SOFTFLOAT
-            /* no floating point exception for native float */
-            SET_FP_ENABLE(env->active_fpu.fcr31, 0);
-#endif
             break;
         case 71: env->active_fpu.fcr0 = tmp; break;
         }
diff --git a/hmp-commands.hx b/hmp-commands.hx
index 834e6a8..6ad8806 100644
--- a/hmp-commands.hx
+++ b/hmp-commands.hx
@@ -740,10 +740,11 @@
 #if defined(TARGET_I386)
     {
         .name       = "nmi",
-        .args_type  = "cpu_index:i",
-        .params     = "cpu",
-        .help       = "inject an NMI on the given CPU",
-        .mhandler.cmd = do_inject_nmi,
+        .args_type  = "",
+        .params     = "",
+        .help       = "inject an NMI on all guest's CPUs",
+        .user_print = monitor_user_noop,
+        .mhandler.cmd_new = do_inject_nmi,
     },
 #endif
 STEXI
diff --git a/hw/9pfs/virtio-9p-debug.c b/hw/9pfs/virtio-9p-debug.c
index 6b18842..4636ad5 100644
--- a/hw/9pfs/virtio-9p-debug.c
+++ b/hw/9pfs/virtio-9p-debug.c
@@ -10,8 +10,9 @@
  * the COPYING file in the top-level directory.
  *
  */
-#include "virtio.h"
-#include "pc.h"
+
+#include "hw/virtio.h"
+#include "hw/pc.h"
 #include "virtio-9p.h"
 #include "virtio-9p-debug.h"
 
diff --git a/hw/9pfs/virtio-9p-device.c b/hw/9pfs/virtio-9p-device.c
new file mode 100644
index 0000000..a2b6acc
--- /dev/null
+++ b/hw/9pfs/virtio-9p-device.c
@@ -0,0 +1,173 @@
+/*
+ * Virtio 9p backend
+ *
+ * Copyright IBM, Corp. 2010
+ *
+ * Authors:
+ *  Anthony Liguori   <aliguori@us.ibm.com>
+ *
+ * This work is licensed under the terms of the GNU GPL, version 2.  See
+ * the COPYING file in the top-level directory.
+ *
+ */
+
+#include "hw/virtio.h"
+#include "hw/pc.h"
+#include "qemu_socket.h"
+#include "hw/virtio-pci.h"
+#include "virtio-9p.h"
+#include "fsdev/qemu-fsdev.h"
+#include "virtio-9p-xattr.h"
+
+static uint32_t virtio_9p_get_features(VirtIODevice *vdev, uint32_t features)
+{
+    features |= 1 << VIRTIO_9P_MOUNT_TAG;
+    return features;
+}
+
+static V9fsState *to_virtio_9p(VirtIODevice *vdev)
+{
+    return (V9fsState *)vdev;
+}
+
+static void virtio_9p_get_config(VirtIODevice *vdev, uint8_t *config)
+{
+    struct virtio_9p_config *cfg;
+    V9fsState *s = to_virtio_9p(vdev);
+
+    cfg = qemu_mallocz(sizeof(struct virtio_9p_config) +
+                        s->tag_len);
+    stw_raw(&cfg->tag_len, s->tag_len);
+    memcpy(cfg->tag, s->tag, s->tag_len);
+    memcpy(config, cfg, s->config_size);
+    qemu_free(cfg);
+}
+
+VirtIODevice *virtio_9p_init(DeviceState *dev, V9fsConf *conf)
+ {
+    V9fsState *s;
+    int i, len;
+    struct stat stat;
+    FsTypeEntry *fse;
+
+
+    s = (V9fsState *)virtio_common_init("virtio-9p",
+                                    VIRTIO_ID_9P,
+                                    sizeof(struct virtio_9p_config)+
+                                    MAX_TAG_LEN,
+                                    sizeof(V9fsState));
+
+    /* initialize pdu allocator */
+    QLIST_INIT(&s->free_list);
+    for (i = 0; i < (MAX_REQ - 1); i++) {
+        QLIST_INSERT_HEAD(&s->free_list, &s->pdus[i], next);
+    }
+
+    s->vq = virtio_add_queue(&s->vdev, MAX_REQ, handle_9p_output);
+
+    fse = get_fsdev_fsentry(conf->fsdev_id);
+
+    if (!fse) {
+        /* We don't have a fsdev identified by fsdev_id */
+        fprintf(stderr, "Virtio-9p device couldn't find fsdev with the "
+                "id = %s\n", conf->fsdev_id ? conf->fsdev_id : "NULL");
+        exit(1);
+    }
+
+    if (!fse->path || !conf->tag) {
+        /* we haven't specified a mount_tag or the path */
+        fprintf(stderr, "fsdev with id %s needs path "
+                "and Virtio-9p device needs mount_tag arguments\n",
+                conf->fsdev_id);
+        exit(1);
+    }
+
+    if (!strcmp(fse->security_model, "passthrough")) {
+        /* Files on the Fileserver set to client user credentials */
+        s->ctx.fs_sm = SM_PASSTHROUGH;
+        s->ctx.xops = passthrough_xattr_ops;
+    } else if (!strcmp(fse->security_model, "mapped")) {
+        /* Files on the fileserver are set to QEMU credentials.
+         * Client user credentials are saved in extended attributes.
+         */
+        s->ctx.fs_sm = SM_MAPPED;
+        s->ctx.xops = mapped_xattr_ops;
+    } else if (!strcmp(fse->security_model, "none")) {
+        /*
+         * Files on the fileserver are set to QEMU credentials.
+         */
+        s->ctx.fs_sm = SM_NONE;
+        s->ctx.xops = none_xattr_ops;
+    } else {
+        fprintf(stderr, "Default to security_model=none. You may want"
+                " enable advanced security model using "
+                "security option:\n\t security_model=passthrough\n\t "
+                "security_model=mapped\n");
+        s->ctx.fs_sm = SM_NONE;
+        s->ctx.xops = none_xattr_ops;
+    }
+
+    if (lstat(fse->path, &stat)) {
+        fprintf(stderr, "share path %s does not exist\n", fse->path);
+        exit(1);
+    } else if (!S_ISDIR(stat.st_mode)) {
+        fprintf(stderr, "share path %s is not a directory\n", fse->path);
+        exit(1);
+    }
+
+    s->ctx.fs_root = qemu_strdup(fse->path);
+    len = strlen(conf->tag);
+    if (len > MAX_TAG_LEN) {
+        len = MAX_TAG_LEN;
+    }
+    /* s->tag is non-NULL terminated string */
+    s->tag = qemu_malloc(len);
+    memcpy(s->tag, conf->tag, len);
+    s->tag_len = len;
+    s->ctx.uid = -1;
+
+    s->ops = fse->ops;
+    s->vdev.get_features = virtio_9p_get_features;
+    s->config_size = sizeof(struct virtio_9p_config) +
+                        s->tag_len;
+    s->vdev.get_config = virtio_9p_get_config;
+
+    return &s->vdev;
+}
+
+static int virtio_9p_init_pci(PCIDevice *pci_dev)
+{
+    VirtIOPCIProxy *proxy = DO_UPCAST(VirtIOPCIProxy, pci_dev, pci_dev);
+    VirtIODevice *vdev;
+
+    vdev = virtio_9p_init(&pci_dev->qdev, &proxy->fsconf);
+    vdev->nvectors = proxy->nvectors;
+    virtio_init_pci(proxy, vdev,
+                    PCI_VENDOR_ID_REDHAT_QUMRANET,
+                    0x1009,
+                    0x2,
+                    0x00);
+    /* make the actual value visible */
+    proxy->nvectors = vdev->nvectors;
+    return 0;
+}
+
+static PCIDeviceInfo virtio_9p_info = {
+    .qdev.name = "virtio-9p-pci",
+    .qdev.size = sizeof(VirtIOPCIProxy),
+    .init      = virtio_9p_init_pci,
+    .qdev.props = (Property[]) {
+        DEFINE_PROP_UINT32("vectors", VirtIOPCIProxy, nvectors, 2),
+        DEFINE_VIRTIO_COMMON_FEATURES(VirtIOPCIProxy, host_features),
+        DEFINE_PROP_STRING("mount_tag", VirtIOPCIProxy, fsconf.tag),
+        DEFINE_PROP_STRING("fsdev", VirtIOPCIProxy, fsconf.fsdev_id),
+        DEFINE_PROP_END_OF_LIST(),
+    }
+};
+
+static void virtio_9p_register_devices(void)
+{
+    pci_qdev_register(&virtio_9p_info);
+}
+
+device_init(virtio_9p_register_devices)
diff --git a/hw/9pfs/virtio-9p-local.c b/hw/9pfs/virtio-9p-local.c
index 0a015de..77904c3 100644
--- a/hw/9pfs/virtio-9p-local.c
+++ b/hw/9pfs/virtio-9p-local.c
@@ -10,7 +10,8 @@
  * the COPYING file in the top-level directory.
  *
  */
-#include "virtio.h"
+
+#include "hw/virtio.h"
 #include "virtio-9p.h"
 #include "virtio-9p-xattr.h"
 #include <arpa/inet.h>
@@ -24,7 +25,8 @@
 static int local_lstat(FsContext *fs_ctx, const char *path, struct stat *stbuf)
 {
     int err;
-    err =  lstat(rpath(fs_ctx, path), stbuf);
+    char buffer[PATH_MAX];
+    err =  lstat(rpath(fs_ctx, path, buffer), stbuf);
     if (err) {
         return err;
     }
@@ -34,19 +36,19 @@
         gid_t tmp_gid;
         mode_t tmp_mode;
         dev_t tmp_dev;
-        if (getxattr(rpath(fs_ctx, path), "user.virtfs.uid", &tmp_uid,
+        if (getxattr(rpath(fs_ctx, path, buffer), "user.virtfs.uid", &tmp_uid,
                     sizeof(uid_t)) > 0) {
             stbuf->st_uid = tmp_uid;
         }
-        if (getxattr(rpath(fs_ctx, path), "user.virtfs.gid", &tmp_gid,
+        if (getxattr(rpath(fs_ctx, path, buffer), "user.virtfs.gid", &tmp_gid,
                     sizeof(gid_t)) > 0) {
             stbuf->st_gid = tmp_gid;
         }
-        if (getxattr(rpath(fs_ctx, path), "user.virtfs.mode", &tmp_mode,
-                    sizeof(mode_t)) > 0) {
+        if (getxattr(rpath(fs_ctx, path, buffer), "user.virtfs.mode",
+                    &tmp_mode, sizeof(mode_t)) > 0) {
             stbuf->st_mode = tmp_mode;
         }
-        if (getxattr(rpath(fs_ctx, path), "user.virtfs.rdev", &tmp_dev,
+        if (getxattr(rpath(fs_ctx, path, buffer), "user.virtfs.rdev", &tmp_dev,
                         sizeof(dev_t)) > 0) {
                 stbuf->st_rdev = tmp_dev;
         }
@@ -91,10 +93,12 @@
 static int local_post_create_passthrough(FsContext *fs_ctx, const char *path,
         FsCred *credp)
 {
-    if (chmod(rpath(fs_ctx, path), credp->fc_mode & 07777) < 0) {
+    char buffer[PATH_MAX];
+    if (chmod(rpath(fs_ctx, path, buffer), credp->fc_mode & 07777) < 0) {
         return -1;
     }
-    if (lchown(rpath(fs_ctx, path), credp->fc_uid, credp->fc_gid) < 0) {
+    if (lchown(rpath(fs_ctx, path, buffer), credp->fc_uid,
+                credp->fc_gid) < 0) {
         /*
          * If we fail to change ownership and if we are
          * using security model none. Ignore the error
@@ -110,9 +114,10 @@
         char *buf, size_t bufsz)
 {
     ssize_t tsize = -1;
+    char buffer[PATH_MAX];
     if (fs_ctx->fs_sm == SM_MAPPED) {
         int fd;
-        fd = open(rpath(fs_ctx, path), O_RDONLY);
+        fd = open(rpath(fs_ctx, path, buffer), O_RDONLY);
         if (fd == -1) {
             return -1;
         }
@@ -123,7 +128,7 @@
         return tsize;
     } else if ((fs_ctx->fs_sm == SM_PASSTHROUGH) ||
                (fs_ctx->fs_sm == SM_NONE)) {
-        tsize = readlink(rpath(fs_ctx, path), buf, bufsz);
+        tsize = readlink(rpath(fs_ctx, path, buffer), buf, bufsz);
     }
     return tsize;
 }
@@ -140,12 +145,14 @@
 
 static int local_open(FsContext *ctx, const char *path, int flags)
 {
-    return open(rpath(ctx, path), flags);
+    char buffer[PATH_MAX];
+    return open(rpath(ctx, path, buffer), flags);
 }
 
 static DIR *local_opendir(FsContext *ctx, const char *path)
 {
-    return opendir(rpath(ctx, path));
+    char buffer[PATH_MAX];
+    return opendir(rpath(ctx, path, buffer));
 }
 
 static void local_rewinddir(FsContext *ctx, DIR *dir)
@@ -200,11 +207,12 @@
 
 static int local_chmod(FsContext *fs_ctx, const char *path, FsCred *credp)
 {
+    char buffer[PATH_MAX];
     if (fs_ctx->fs_sm == SM_MAPPED) {
-        return local_set_xattr(rpath(fs_ctx, path), credp);
+        return local_set_xattr(rpath(fs_ctx, path, buffer), credp);
     } else if ((fs_ctx->fs_sm == SM_PASSTHROUGH) ||
                (fs_ctx->fs_sm == SM_NONE)) {
-        return chmod(rpath(fs_ctx, path), credp->fc_mode);
+        return chmod(rpath(fs_ctx, path, buffer), credp->fc_mode);
     }
     return -1;
 }
@@ -213,21 +221,24 @@
 {
     int err = -1;
     int serrno = 0;
+    char buffer[PATH_MAX];
 
     /* Determine the security model */
     if (fs_ctx->fs_sm == SM_MAPPED) {
-        err = mknod(rpath(fs_ctx, path), SM_LOCAL_MODE_BITS|S_IFREG, 0);
+        err = mknod(rpath(fs_ctx, path, buffer),
+                SM_LOCAL_MODE_BITS|S_IFREG, 0);
         if (err == -1) {
             return err;
         }
-        local_set_xattr(rpath(fs_ctx, path), credp);
+        local_set_xattr(rpath(fs_ctx, path, buffer), credp);
         if (err == -1) {
             serrno = errno;
             goto err_end;
         }
     } else if ((fs_ctx->fs_sm == SM_PASSTHROUGH) ||
                (fs_ctx->fs_sm == SM_NONE)) {
-        err = mknod(rpath(fs_ctx, path), credp->fc_mode, credp->fc_rdev);
+        err = mknod(rpath(fs_ctx, path, buffer), credp->fc_mode,
+                credp->fc_rdev);
         if (err == -1) {
             return err;
         }
@@ -240,7 +251,7 @@
     return err;
 
 err_end:
-    remove(rpath(fs_ctx, path));
+    remove(rpath(fs_ctx, path, buffer));
     errno = serrno;
     return err;
 }
@@ -249,22 +260,23 @@
 {
     int err = -1;
     int serrno = 0;
+    char buffer[PATH_MAX];
 
     /* Determine the security model */
     if (fs_ctx->fs_sm == SM_MAPPED) {
-        err = mkdir(rpath(fs_ctx, path), SM_LOCAL_DIR_MODE_BITS);
+        err = mkdir(rpath(fs_ctx, path, buffer), SM_LOCAL_DIR_MODE_BITS);
         if (err == -1) {
             return err;
         }
         credp->fc_mode = credp->fc_mode|S_IFDIR;
-        err = local_set_xattr(rpath(fs_ctx, path), credp);
+        err = local_set_xattr(rpath(fs_ctx, path, buffer), credp);
         if (err == -1) {
             serrno = errno;
             goto err_end;
         }
     } else if ((fs_ctx->fs_sm == SM_PASSTHROUGH) ||
                (fs_ctx->fs_sm == SM_NONE)) {
-        err = mkdir(rpath(fs_ctx, path), credp->fc_mode);
+        err = mkdir(rpath(fs_ctx, path, buffer), credp->fc_mode);
         if (err == -1) {
             return err;
         }
@@ -277,7 +289,7 @@
     return err;
 
 err_end:
-    remove(rpath(fs_ctx, path));
+    remove(rpath(fs_ctx, path, buffer));
     errno = serrno;
     return err;
 }
@@ -318,23 +330,24 @@
     int fd = -1;
     int err = -1;
     int serrno = 0;
+    char buffer[PATH_MAX];
 
     /* Determine the security model */
     if (fs_ctx->fs_sm == SM_MAPPED) {
-        fd = open(rpath(fs_ctx, path), flags, SM_LOCAL_MODE_BITS);
+        fd = open(rpath(fs_ctx, path, buffer), flags, SM_LOCAL_MODE_BITS);
         if (fd == -1) {
             return fd;
         }
         credp->fc_mode = credp->fc_mode|S_IFREG;
         /* Set cleint credentials in xattr */
-        err = local_set_xattr(rpath(fs_ctx, path), credp);
+        err = local_set_xattr(rpath(fs_ctx, path, buffer), credp);
         if (err == -1) {
             serrno = errno;
             goto err_end;
         }
     } else if ((fs_ctx->fs_sm == SM_PASSTHROUGH) ||
                (fs_ctx->fs_sm == SM_NONE)) {
-        fd = open(rpath(fs_ctx, path), flags, credp->fc_mode);
+        fd = open(rpath(fs_ctx, path, buffer), flags, credp->fc_mode);
         if (fd == -1) {
             return fd;
         }
@@ -348,7 +361,7 @@
 
 err_end:
     close(fd);
-    remove(rpath(fs_ctx, path));
+    remove(rpath(fs_ctx, path, buffer));
     errno = serrno;
     return err;
 }
@@ -359,12 +372,13 @@
 {
     int err = -1;
     int serrno = 0;
+    char buffer[PATH_MAX];
 
     /* Determine the security model */
     if (fs_ctx->fs_sm == SM_MAPPED) {
         int fd;
         ssize_t oldpath_size, write_size;
-        fd = open(rpath(fs_ctx, newpath), O_CREAT|O_EXCL|O_RDWR,
+        fd = open(rpath(fs_ctx, newpath, buffer), O_CREAT|O_EXCL|O_RDWR,
                 SM_LOCAL_MODE_BITS);
         if (fd == -1) {
             return fd;
@@ -384,18 +398,19 @@
         close(fd);
         /* Set cleint credentials in symlink's xattr */
         credp->fc_mode = credp->fc_mode|S_IFLNK;
-        err = local_set_xattr(rpath(fs_ctx, newpath), credp);
+        err = local_set_xattr(rpath(fs_ctx, newpath, buffer), credp);
         if (err == -1) {
             serrno = errno;
             goto err_end;
         }
     } else if ((fs_ctx->fs_sm == SM_PASSTHROUGH) ||
                (fs_ctx->fs_sm == SM_NONE)) {
-        err = symlink(oldpath, rpath(fs_ctx, newpath));
+        err = symlink(oldpath, rpath(fs_ctx, newpath, buffer));
         if (err) {
             return err;
         }
-        err = lchown(rpath(fs_ctx, newpath), credp->fc_uid, credp->fc_gid);
+        err = lchown(rpath(fs_ctx, newpath, buffer), credp->fc_uid,
+                credp->fc_gid);
         if (err == -1) {
             /*
              * If we fail to change ownership and if we are
@@ -411,70 +426,45 @@
     return err;
 
 err_end:
-    remove(rpath(fs_ctx, newpath));
+    remove(rpath(fs_ctx, newpath, buffer));
     errno = serrno;
     return err;
 }
 
 static int local_link(FsContext *ctx, const char *oldpath, const char *newpath)
 {
-    char *tmp = qemu_strdup(rpath(ctx, oldpath));
-    int err, serrno = 0;
+    char buffer[PATH_MAX], buffer1[PATH_MAX];
 
-    if (tmp == NULL) {
-        return -ENOMEM;
-    }
-
-    err = link(tmp, rpath(ctx, newpath));
-    if (err == -1) {
-        serrno = errno;
-    }
-
-    qemu_free(tmp);
-
-    if (err == -1) {
-        errno = serrno;
-    }
-
-    return err;
+    return link(rpath(ctx, oldpath, buffer), rpath(ctx, newpath, buffer1));
 }
 
 static int local_truncate(FsContext *ctx, const char *path, off_t size)
 {
-    return truncate(rpath(ctx, path), size);
+    char buffer[PATH_MAX];
+    return truncate(rpath(ctx, path, buffer), size);
 }
 
 static int local_rename(FsContext *ctx, const char *oldpath,
                         const char *newpath)
 {
-    char *tmp;
-    int err;
+    char buffer[PATH_MAX], buffer1[PATH_MAX];
 
-    tmp = qemu_strdup(rpath(ctx, oldpath));
-
-    err = rename(tmp, rpath(ctx, newpath));
-    if (err == -1) {
-        int serrno = errno;
-        qemu_free(tmp);
-        errno = serrno;
-    } else {
-        qemu_free(tmp);
-    }
-
-    return err;
-
+    return rename(rpath(ctx, oldpath, buffer), rpath(ctx, newpath, buffer1));
 }
 
 static int local_chown(FsContext *fs_ctx, const char *path, FsCred *credp)
 {
+    char buffer[PATH_MAX];
     if ((credp->fc_uid == -1 && credp->fc_gid == -1) ||
             (fs_ctx->fs_sm == SM_PASSTHROUGH)) {
-        return lchown(rpath(fs_ctx, path), credp->fc_uid, credp->fc_gid);
+        return lchown(rpath(fs_ctx, path, buffer), credp->fc_uid,
+                credp->fc_gid);
     } else if (fs_ctx->fs_sm == SM_MAPPED) {
-        return local_set_xattr(rpath(fs_ctx, path), credp);
+        return local_set_xattr(rpath(fs_ctx, path, buffer), credp);
     } else if ((fs_ctx->fs_sm == SM_PASSTHROUGH) ||
                (fs_ctx->fs_sm == SM_NONE)) {
-        return lchown(rpath(fs_ctx, path), credp->fc_uid, credp->fc_gid);
+        return lchown(rpath(fs_ctx, path, buffer), credp->fc_uid,
+                credp->fc_gid);
     }
     return -1;
 }
@@ -482,12 +472,15 @@
 static int local_utimensat(FsContext *s, const char *path,
                            const struct timespec *buf)
 {
-    return qemu_utimensat(AT_FDCWD, rpath(s, path), buf, AT_SYMLINK_NOFOLLOW);
+    char buffer[PATH_MAX];
+    return qemu_utimensat(AT_FDCWD, rpath(s, path, buffer), buf,
+            AT_SYMLINK_NOFOLLOW);
 }
 
 static int local_remove(FsContext *ctx, const char *path)
 {
-    return remove(rpath(ctx, path));
+    char buffer[PATH_MAX];
+    return remove(rpath(ctx, path, buffer));
 }
 
 static int local_fsync(FsContext *ctx, int fd, int datasync)
@@ -501,7 +494,8 @@
 
 static int local_statfs(FsContext *s, const char *path, struct statfs *stbuf)
 {
-   return statfs(rpath(s, path), stbuf);
+    char buffer[PATH_MAX];
+   return statfs(rpath(s, path, buffer), stbuf);
 }
 
 static ssize_t local_lgetxattr(FsContext *ctx, const char *path,
diff --git a/hw/9pfs/virtio-9p-posix-acl.c b/hw/9pfs/virtio-9p-posix-acl.c
index 575abe8..f5b392e 100644
--- a/hw/9pfs/virtio-9p-posix-acl.c
+++ b/hw/9pfs/virtio-9p-posix-acl.c
@@ -13,7 +13,7 @@
 
 #include <sys/types.h>
 #include <attr/xattr.h>
-#include "virtio.h"
+#include "hw/virtio.h"
 #include "virtio-9p.h"
 #include "fsdev/file-op-9p.h"
 #include "virtio-9p-xattr.h"
@@ -26,7 +26,8 @@
 static ssize_t mp_pacl_getxattr(FsContext *ctx, const char *path,
                                 const char *name, void *value, size_t size)
 {
-    return lgetxattr(rpath(ctx, path), MAP_ACL_ACCESS, value, size);
+    char buffer[PATH_MAX];
+    return lgetxattr(rpath(ctx, path, buffer), MAP_ACL_ACCESS, value, size);
 }
 
 static ssize_t mp_pacl_listxattr(FsContext *ctx, const char *path,
@@ -50,14 +51,17 @@
 static int mp_pacl_setxattr(FsContext *ctx, const char *path, const char *name,
                             void *value, size_t size, int flags)
 {
-    return lsetxattr(rpath(ctx, path), MAP_ACL_ACCESS, value, size, flags);
+    char buffer[PATH_MAX];
+    return lsetxattr(rpath(ctx, path, buffer), MAP_ACL_ACCESS, value,
+            size, flags);
 }
 
 static int mp_pacl_removexattr(FsContext *ctx,
                                const char *path, const char *name)
 {
     int ret;
-    ret  = lremovexattr(rpath(ctx, path), MAP_ACL_ACCESS);
+    char buffer[PATH_MAX];
+    ret  = lremovexattr(rpath(ctx, path, buffer), MAP_ACL_ACCESS);
     if (ret == -1 && errno == ENODATA) {
         /*
          * We don't get ENODATA error when trying to remove a
@@ -73,7 +77,8 @@
 static ssize_t mp_dacl_getxattr(FsContext *ctx, const char *path,
                                 const char *name, void *value, size_t size)
 {
-    return lgetxattr(rpath(ctx, path), MAP_ACL_DEFAULT, value, size);
+    char buffer[PATH_MAX];
+    return lgetxattr(rpath(ctx, path, buffer), MAP_ACL_DEFAULT, value, size);
 }
 
 static ssize_t mp_dacl_listxattr(FsContext *ctx, const char *path,
@@ -97,14 +102,17 @@
 static int mp_dacl_setxattr(FsContext *ctx, const char *path, const char *name,
                             void *value, size_t size, int flags)
 {
-    return lsetxattr(rpath(ctx, path), MAP_ACL_DEFAULT, value, size, flags);
+    char buffer[PATH_MAX];
+    return lsetxattr(rpath(ctx, path, buffer), MAP_ACL_DEFAULT, value,
+            size, flags);
 }
 
 static int mp_dacl_removexattr(FsContext *ctx,
                                const char *path, const char *name)
 {
     int ret;
-    ret  = lremovexattr(rpath(ctx, path), MAP_ACL_DEFAULT);
+    char buffer[PATH_MAX];
+    ret  = lremovexattr(rpath(ctx, path, buffer), MAP_ACL_DEFAULT);
     if (ret == -1 && errno == ENODATA) {
         /*
          * We don't get ENODATA error when trying to remove a
diff --git a/hw/9pfs/virtio-9p-xattr-user.c b/hw/9pfs/virtio-9p-xattr-user.c
index bba13ce..5044a3e 100644
--- a/hw/9pfs/virtio-9p-xattr-user.c
+++ b/hw/9pfs/virtio-9p-xattr-user.c
@@ -12,7 +12,7 @@
  */
 
 #include <sys/types.h>
-#include "virtio.h"
+#include "hw/virtio.h"
 #include "virtio-9p.h"
 #include "fsdev/file-op-9p.h"
 #include "virtio-9p-xattr.h"
@@ -21,6 +21,7 @@
 static ssize_t mp_user_getxattr(FsContext *ctx, const char *path,
                                 const char *name, void *value, size_t size)
 {
+    char buffer[PATH_MAX];
     if (strncmp(name, "user.virtfs.", 12) == 0) {
         /*
          * Don't allow fetch of user.virtfs namesapce
@@ -29,7 +30,7 @@
         errno = ENOATTR;
         return -1;
     }
-    return lgetxattr(rpath(ctx, path), name, value, size);
+    return lgetxattr(rpath(ctx, path, buffer), name, value, size);
 }
 
 static ssize_t mp_user_listxattr(FsContext *ctx, const char *path,
@@ -67,6 +68,7 @@
 static int mp_user_setxattr(FsContext *ctx, const char *path, const char *name,
                             void *value, size_t size, int flags)
 {
+    char buffer[PATH_MAX];
     if (strncmp(name, "user.virtfs.", 12) == 0) {
         /*
          * Don't allow fetch of user.virtfs namesapce
@@ -75,12 +77,13 @@
         errno = EACCES;
         return -1;
     }
-    return lsetxattr(rpath(ctx, path), name, value, size, flags);
+    return lsetxattr(rpath(ctx, path, buffer), name, value, size, flags);
 }
 
 static int mp_user_removexattr(FsContext *ctx,
                                const char *path, const char *name)
 {
+    char buffer[PATH_MAX];
     if (strncmp(name, "user.virtfs.", 12) == 0) {
         /*
          * Don't allow fetch of user.virtfs namesapce
@@ -89,7 +92,7 @@
         errno = EACCES;
         return -1;
     }
-    return lremovexattr(rpath(ctx, path), name);
+    return lremovexattr(rpath(ctx, path, buffer), name);
 }
 
 XattrOperations mapped_user_xattr = {
diff --git a/hw/9pfs/virtio-9p-xattr.c b/hw/9pfs/virtio-9p-xattr.c
index 03c3d3f..bde0b7f 100644
--- a/hw/9pfs/virtio-9p-xattr.c
+++ b/hw/9pfs/virtio-9p-xattr.c
@@ -11,7 +11,7 @@
  *
  */
 
-#include "virtio.h"
+#include "hw/virtio.h"
 #include "virtio-9p.h"
 #include "fsdev/file-op-9p.h"
 #include "virtio-9p-xattr.h"
@@ -66,20 +66,21 @@
                         void *value, size_t vsize)
 {
     ssize_t size = 0;
+    char buffer[PATH_MAX];
     void *ovalue = value;
     XattrOperations *xops;
     char *orig_value, *orig_value_start;
     ssize_t xattr_len, parsed_len = 0, attr_len;
 
     /* Get the actual len */
-    xattr_len = llistxattr(rpath(ctx, path), value, 0);
+    xattr_len = llistxattr(rpath(ctx, path, buffer), value, 0);
     if (xattr_len <= 0) {
         return xattr_len;
     }
 
     /* Now fetch the xattr and find the actual size */
     orig_value = qemu_malloc(xattr_len);
-    xattr_len = llistxattr(rpath(ctx, path), orig_value, xattr_len);
+    xattr_len = llistxattr(rpath(ctx, path, buffer), orig_value, xattr_len);
 
     /* store the orig pointer */
     orig_value_start = orig_value;
diff --git a/hw/9pfs/virtio-9p-xattr.h b/hw/9pfs/virtio-9p-xattr.h
index 2bbae2d..247e414 100644
--- a/hw/9pfs/virtio-9p-xattr.h
+++ b/hw/9pfs/virtio-9p-xattr.h
@@ -54,20 +54,23 @@
 static inline ssize_t pt_getxattr(FsContext *ctx, const char *path,
                                   const char *name, void *value, size_t size)
 {
-    return lgetxattr(rpath(ctx, path), name, value, size);
+    char buffer[PATH_MAX];
+    return lgetxattr(rpath(ctx, path, buffer), name, value, size);
 }
 
 static inline int pt_setxattr(FsContext *ctx, const char *path,
                               const char *name, void *value,
                               size_t size, int flags)
 {
-    return lsetxattr(rpath(ctx, path), name, value, size, flags);
+    char buffer[PATH_MAX];
+    return lsetxattr(rpath(ctx, path, buffer), name, value, size, flags);
 }
 
 static inline int pt_removexattr(FsContext *ctx,
                                  const char *path, const char *name)
 {
-    return lremovexattr(rpath(ctx, path), name);
+    char buffer[PATH_MAX];
+    return lremovexattr(rpath(ctx, path, buffer), name);
 }
 
 static inline ssize_t notsup_getxattr(FsContext *ctx, const char *path,
diff --git a/hw/9pfs/virtio-9p.c b/hw/9pfs/virtio-9p.c
index b5fc52b..4890df6 100644
--- a/hw/9pfs/virtio-9p.c
+++ b/hw/9pfs/virtio-9p.c
@@ -11,9 +11,10 @@
  *
  */
 
-#include "virtio.h"
-#include "pc.h"
+#include "hw/virtio.h"
+#include "hw/pc.h"
 #include "qemu_socket.h"
+#include "hw/virtio-pci.h"
 #include "virtio-9p.h"
 #include "fsdev/qemu-fsdev.h"
 #include "virtio-9p-debug.h"
@@ -194,7 +195,6 @@
     cred.fc_uid = uid;
     cred.fc_gid = gid;
     cred.fc_mode = mode & 07777;
-    flags = flags;
 
     return s->ops->open2(&s->ctx, fullname, flags, &cred);
 }
@@ -423,6 +423,22 @@
     v9fs_string_sprintf(lhs, "%s", rhs->data);
 }
 
+/*
+ * Return TRUE if s1 is an ancestor of s2.
+ *
+ * E.g. "a/b" is an ancestor of "a/b/c" but not of "a/bc/d".
+ * As a special case, We treat s1 as ancestor of s2 if they are same!
+ */
+static int v9fs_path_is_ancestor(V9fsString *s1, V9fsString *s2)
+{
+    if (!strncmp(s1->data, s2->data, s1->size)) {
+        if (s2->data[s1->size] == '\0' || s2->data[s1->size] == '/') {
+            return 1;
+        }
+    }
+    return 0;
+}
+
 static size_t v9fs_string_size(V9fsString *str)
 {
     return str->size;
@@ -2805,13 +2821,13 @@
             for (fidp = s->fid_list; fidp; fidp = fidp->next) {
                 if (vs->fidp == fidp) {
                     /*
-                    * we replace name of this fid towards the end
-                    * so that our below strcmp will work
+                    * we replace name of this fid towards the end so
+                    * that our below v9fs_path_is_ancestor check will
+                    * work
                     */
                     continue;
                 }
-                if (!strncmp(vs->fidp->path.data, fidp->path.data,
-                    strlen(vs->fidp->path.data))) {
+                if (v9fs_path_is_ancestor(&vs->fidp->path, &fidp->path)) {
                     /* replace the name */
                     v9fs_fix_path(&fidp->path, &vs->name,
                                   strlen(vs->fidp->path.data));
@@ -3589,6 +3605,11 @@
     [P9_TREMOVE] = v9fs_remove,
 };
 
+static void v9fs_op_not_supp(V9fsState *s, V9fsPDU *pdu)
+{
+    complete_pdu(s, pdu, -EOPNOTSUPP);
+}
+
 static void submit_pdu(V9fsState *s, V9fsPDU *pdu)
 {
     pdu_handler_t *handler;
@@ -3596,16 +3617,16 @@
     if (debug_9p_pdu) {
         pprint_pdu(pdu);
     }
-
-    BUG_ON(pdu->id >= ARRAY_SIZE(pdu_handlers));
-
-    handler = pdu_handlers[pdu->id];
-    BUG_ON(handler == NULL);
-
+    if (pdu->id >= ARRAY_SIZE(pdu_handlers) ||
+        (pdu_handlers[pdu->id] == NULL)) {
+        handler = v9fs_op_not_supp;
+    } else {
+        handler = pdu_handlers[pdu->id];
+    }
     handler(s, pdu);
 }
 
-static void handle_9p_output(VirtIODevice *vdev, VirtQueue *vq)
+void handle_9p_output(VirtIODevice *vdev, VirtQueue *vq)
 {
     V9fsState *s = (V9fsState *)vdev;
     V9fsPDU *pdu;
@@ -3629,119 +3650,3 @@
 
     free_pdu(s, pdu);
 }
-
-static uint32_t virtio_9p_get_features(VirtIODevice *vdev, uint32_t features)
-{
-    features |= 1 << VIRTIO_9P_MOUNT_TAG;
-    return features;
-}
-
-static V9fsState *to_virtio_9p(VirtIODevice *vdev)
-{
-    return (V9fsState *)vdev;
-}
-
-static void virtio_9p_get_config(VirtIODevice *vdev, uint8_t *config)
-{
-    struct virtio_9p_config *cfg;
-    V9fsState *s = to_virtio_9p(vdev);
-
-    cfg = qemu_mallocz(sizeof(struct virtio_9p_config) +
-                        s->tag_len);
-    stw_raw(&cfg->tag_len, s->tag_len);
-    memcpy(cfg->tag, s->tag, s->tag_len);
-    memcpy(config, cfg, s->config_size);
-    qemu_free(cfg);
-}
-
-VirtIODevice *virtio_9p_init(DeviceState *dev, V9fsConf *conf)
- {
-    V9fsState *s;
-    int i, len;
-    struct stat stat;
-    FsTypeEntry *fse;
-
-
-    s = (V9fsState *)virtio_common_init("virtio-9p",
-                                    VIRTIO_ID_9P,
-                                    sizeof(struct virtio_9p_config)+
-                                    MAX_TAG_LEN,
-                                    sizeof(V9fsState));
-
-    /* initialize pdu allocator */
-    QLIST_INIT(&s->free_list);
-    for (i = 0; i < (MAX_REQ - 1); i++) {
-	QLIST_INSERT_HEAD(&s->free_list, &s->pdus[i], next);
-    }
-
-    s->vq = virtio_add_queue(&s->vdev, MAX_REQ, handle_9p_output);
-
-    fse = get_fsdev_fsentry(conf->fsdev_id);
-
-    if (!fse) {
-        /* We don't have a fsdev identified by fsdev_id */
-        fprintf(stderr, "Virtio-9p device couldn't find fsdev with the "
-                "id = %s\n", conf->fsdev_id ? conf->fsdev_id : "NULL");
-        exit(1);
-    }
-
-    if (!fse->path || !conf->tag) {
-        /* we haven't specified a mount_tag or the path */
-        fprintf(stderr, "fsdev with id %s needs path "
-                "and Virtio-9p device needs mount_tag arguments\n",
-                conf->fsdev_id);
-        exit(1);
-    }
-
-    if (!strcmp(fse->security_model, "passthrough")) {
-        /* Files on the Fileserver set to client user credentials */
-        s->ctx.fs_sm = SM_PASSTHROUGH;
-        s->ctx.xops = passthrough_xattr_ops;
-    } else if (!strcmp(fse->security_model, "mapped")) {
-        /* Files on the fileserver are set to QEMU credentials.
-         * Client user credentials are saved in extended attributes.
-         */
-        s->ctx.fs_sm = SM_MAPPED;
-        s->ctx.xops = mapped_xattr_ops;
-    } else if (!strcmp(fse->security_model, "none")) {
-        /*
-         * Files on the fileserver are set to QEMU credentials.
-         */
-        s->ctx.fs_sm = SM_NONE;
-        s->ctx.xops = none_xattr_ops;
-    } else {
-        fprintf(stderr, "Default to security_model=none. You may want"
-                " enable advanced security model using "
-                "security option:\n\t security_model=passthrough \n\t "
-                "security_model=mapped\n");
-        s->ctx.fs_sm = SM_NONE;
-        s->ctx.xops = none_xattr_ops;
-    }
-
-    if (lstat(fse->path, &stat)) {
-        fprintf(stderr, "share path %s does not exist\n", fse->path);
-        exit(1);
-    } else if (!S_ISDIR(stat.st_mode)) {
-        fprintf(stderr, "share path %s is not a directory \n", fse->path);
-        exit(1);
-    }
-
-    s->ctx.fs_root = qemu_strdup(fse->path);
-    len = strlen(conf->tag);
-    if (len > MAX_TAG_LEN) {
-        len = MAX_TAG_LEN;
-    }
-    /* s->tag is non-NULL terminated string */
-    s->tag = qemu_malloc(len);
-    memcpy(s->tag, conf->tag, len);
-    s->tag_len = len;
-    s->ctx.uid = -1;
-
-    s->ops = fse->ops;
-    s->vdev.get_features = virtio_9p_get_features;
-    s->config_size = sizeof(struct virtio_9p_config) +
-                        s->tag_len;
-    s->vdev.get_config = virtio_9p_get_config;
-
-    return &s->vdev;
-}
diff --git a/hw/9pfs/virtio-9p.h b/hw/9pfs/virtio-9p.h
index 622928f..2bfbe62 100644
--- a/hw/9pfs/virtio-9p.h
+++ b/hw/9pfs/virtio-9p.h
@@ -101,6 +101,11 @@
 #define P9_NOTAG    (u16)(~0)
 #define P9_NOFID    (u32)(~0)
 #define P9_MAXWELEM 16
+static inline const char *rpath(FsContext *ctx, const char *path, char *buffer)
+{
+    snprintf(buffer, PATH_MAX, "%s/%s", ctx->fs_root, path);
+    return buffer;
+}
 
 /*
  * ample room for Twrite/Rread header
@@ -504,4 +509,6 @@
     return pdu_packunpack(dst, sg, sg_count, offset, size, 0);
 }
 
+extern void handle_9p_output(VirtIODevice *vdev, VirtQueue *vq);
+
 #endif
diff --git a/hw/alpha_palcode.c b/hw/alpha_palcode.c
deleted file mode 100644
index 033b542..0000000
--- a/hw/alpha_palcode.c
+++ /dev/null
@@ -1,1048 +0,0 @@
-/*
- *  Alpha emulation - PALcode emulation for qemu.
- *
- *  Copyright (c) 2007 Jocelyn Mayer
- *
- * This library is free software; you can redistribute it and/or
- * modify it under the terms of the GNU Lesser General Public
- * License as published by the Free Software Foundation; either
- * version 2 of the License, or (at your option) any later version.
- *
- * This library is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
- * Lesser General Public License for more details.
- *
- * You should have received a copy of the GNU Lesser General Public
- * License along with this library; if not, see <http://www.gnu.org/licenses/>.
- */
-
-#include <stdint.h>
-#include <stdlib.h>
-#include <stdio.h>
-
-#include "cpu.h"
-#include "exec-all.h"
-
-/* Shared handlers */
-static void pal_reset (CPUState *env);
-/* Console handlers */
-static void pal_console_call (CPUState *env, uint32_t palcode);
-/* OpenVMS handlers */
-static void pal_openvms_call (CPUState *env, uint32_t palcode);
-/* UNIX / Linux handlers */
-static void pal_unix_call (CPUState *env, uint32_t palcode);
-
-pal_handler_t pal_handlers[] = {
-    /* Console handler */
-    {
-        .reset = &pal_reset,
-        .call_pal = &pal_console_call,
-    },
-    /* OpenVMS handler */
-    {
-        .reset = &pal_reset,
-        .call_pal = &pal_openvms_call,
-    },
-    /* UNIX / Linux handler */
-    {
-        .reset = &pal_reset,
-        .call_pal = &pal_unix_call,
-    },
-};
-
-#if 0
-/* One must explicitly check that the TB is valid and the FOE bit is reset */
-static void update_itb (void)
-{
-    /* This writes into a temp register, not the actual one */
-    mtpr(TB_TAG);
-    mtpr(TB_CTL);
-    /* This commits the TB update */
-    mtpr(ITB_PTE);
-}
-
-static void update_dtb (void);
-{
-    mtpr(TB_CTL);
-    /* This write into a temp register, not the actual one */
-    mtpr(TB_TAG);
-    /* This commits the TB update */
-    mtpr(DTB_PTE);
-}
-#endif
-
-static void pal_reset (CPUState *env)
-{
-}
-
-static void do_swappal (CPUState *env, uint64_t palid)
-{
-    pal_handler_t *pal_handler;
-
-    switch (palid) {
-    case 0 ... 2:
-        pal_handler = &pal_handlers[palid];
-        env->pal_handler = pal_handler;
-        env->ipr[IPR_PAL_BASE] = -1ULL;
-        (*pal_handler->reset)(env);
-        break;
-    case 3 ... 255:
-        /* Unknown identifier */
-        env->ir[0] = 1;
-        return;
-    default:
-        /* We were given the entry point address */
-        env->pal_handler = NULL;
-        env->ipr[IPR_PAL_BASE] = palid;
-        env->pc = env->ipr[IPR_PAL_BASE];
-        cpu_loop_exit();
-    }
-}
-
-static void pal_console_call (CPUState *env, uint32_t palcode)
-{
-    uint64_t palid;
-
-    if (palcode < 0x00000080) {
-        /* Privileged palcodes */
-        if (!(env->ps >> 3)) {
-            /* TODO: generate privilege exception */
-        }
-    }
-    switch (palcode) {
-    case 0x00000000:
-        /* HALT */
-        /* REQUIRED */
-        break;
-    case 0x00000001:
-        /* CFLUSH */
-        break;
-    case 0x00000002:
-        /* DRAINA */
-        /* REQUIRED */
-        /* Implemented as no-op */
-        break;
-    case 0x00000009:
-        /* CSERVE */
-        /* REQUIRED */
-        break;
-    case 0x0000000A:
-        /* SWPPAL */
-        /* REQUIRED */
-        palid = env->ir[16];
-        do_swappal(env, palid);
-        break;
-    case 0x00000080:
-        /* BPT */
-        /* REQUIRED */
-        break;
-    case 0x00000081:
-        /* BUGCHK */
-        /* REQUIRED */
-        break;
-    case 0x00000086:
-        /* IMB */
-        /* REQUIRED */
-        /* Implemented as no-op */
-        break;
-    case 0x0000009E:
-        /* RDUNIQUE */
-        /* REQUIRED */
-        break;
-    case 0x0000009F:
-        /* WRUNIQUE */
-        /* REQUIRED */
-        break;
-    case 0x000000AA:
-        /* GENTRAP */
-        /* REQUIRED */
-        break;
-    default:
-        break;
-    }
-}
-
-static void pal_openvms_call (CPUState *env, uint32_t palcode)
-{
-    uint64_t palid, val, oldval;
-
-    if (palcode < 0x00000080) {
-        /* Privileged palcodes */
-        if (!(env->ps >> 3)) {
-            /* TODO: generate privilege exception */
-        }
-    }
-    switch (palcode) {
-    case 0x00000000:
-        /* HALT */
-        /* REQUIRED */
-        break;
-    case 0x00000001:
-        /* CFLUSH */
-        break;
-    case 0x00000002:
-        /* DRAINA */
-        /* REQUIRED */
-        /* Implemented as no-op */
-        break;
-    case 0x00000003:
-        /* LDQP */
-        break;
-    case 0x00000004:
-        /* STQP */
-        break;
-    case 0x00000005:
-        /* SWPCTX */
-        break;
-    case 0x00000006:
-        /* MFPR_ASN */
-        if (cpu_alpha_mfpr(env, IPR_ASN, &val) == 0)
-            env->ir[0] = val;
-        break;
-    case 0x00000007:
-        /* MTPR_ASTEN */
-        val = env->ir[16];
-        if (cpu_alpha_mtpr(env, IPR_ASTEN, val, &oldval) == 1)
-            env->ir[0] = val;
-        break;
-    case 0x00000008:
-        /* MTPR_ASTSR */
-        val = env->ir[16];
-        if (cpu_alpha_mtpr(env, IPR_ASTSR, val, &oldval) == 1)
-            env->ir[0] = val;
-        break;
-    case 0x00000009:
-        /* CSERVE */
-        /* REQUIRED */
-        break;
-    case 0x0000000A:
-        /* SWPPAL */
-        /* REQUIRED */
-        palid = env->ir[16];
-        do_swappal(env, palid);
-        break;
-    case 0x0000000B:
-        /* MFPR_FEN */
-        if (cpu_alpha_mfpr(env, IPR_FEN, &val) == 0)
-            env->ir[0] = val;
-        break;
-    case 0x0000000C:
-        /* MTPR_FEN */
-        val = env->ir[16];
-        if (cpu_alpha_mtpr(env, IPR_FEN, val, &oldval) == 1)
-            env->ir[0] = val;
-        break;
-    case 0x0000000D:
-        /* MTPR_IPIR */
-        val = env->ir[16];
-        if (cpu_alpha_mtpr(env, IPR_IPIR, val, &oldval) == 1)
-            env->ir[0] = val;
-        break;
-    case 0x0000000E:
-        /* MFPR_IPL */
-        if (cpu_alpha_mfpr(env, IPR_IPL, &val) == 0)
-            env->ir[0] = val;
-        break;
-    case 0x0000000F:
-        /* MTPR_IPL */
-        val = env->ir[16];
-        if (cpu_alpha_mtpr(env, IPR_IPL, val, &oldval) == 1)
-            env->ir[0] = val;
-        break;
-    case 0x00000010:
-        /* MFPR_MCES */
-        if (cpu_alpha_mfpr(env, IPR_MCES, &val) == 0)
-            env->ir[0] = val;
-        break;
-    case 0x00000011:
-        /* MTPR_MCES */
-        val = env->ir[16];
-        if (cpu_alpha_mtpr(env, IPR_MCES, val, &oldval) == 1)
-            env->ir[0] = val;
-        break;
-    case 0x00000012:
-        /* MFPR_PCBB */
-        if (cpu_alpha_mfpr(env, IPR_PCBB, &val) == 0)
-            env->ir[0] = val;
-        break;
-    case 0x00000013:
-        /* MFPR_PRBR */
-        if (cpu_alpha_mfpr(env, IPR_PRBR, &val) == 0)
-            env->ir[0] = val;
-        break;
-    case 0x00000014:
-        /* MTPR_PRBR */
-        val = env->ir[16];
-        if (cpu_alpha_mtpr(env, IPR_PRBR, val, &oldval) == 1)
-            env->ir[0] = val;
-        break;
-    case 0x00000015:
-        /* MFPR_PTBR */
-        if (cpu_alpha_mfpr(env, IPR_PTBR, &val) == 0)
-            env->ir[0] = val;
-        break;
-    case 0x00000016:
-        /* MFPR_SCBB */
-        if (cpu_alpha_mfpr(env, IPR_SCBB, &val) == 0)
-            env->ir[0] = val;
-        break;
-    case 0x00000017:
-        /* MTPR_SCBB */
-        val = env->ir[16];
-        if (cpu_alpha_mtpr(env, IPR_SCBB, val, &oldval) == 1)
-            env->ir[0] = val;
-        break;
-    case 0x00000018:
-        /* MTPR_SIRR */
-        val = env->ir[16];
-        if (cpu_alpha_mtpr(env, IPR_SIRR, val, &oldval) == 1)
-            env->ir[0] = val;
-        break;
-    case 0x00000019:
-        /* MFPR_SISR */
-        if (cpu_alpha_mfpr(env, IPR_SISR, &val) == 0)
-            env->ir[0] = val;
-        break;
-    case 0x0000001A:
-        /* MFPR_TBCHK */
-        if (cpu_alpha_mfpr(env, IPR_TBCHK, &val) == 0)
-            env->ir[0] = val;
-        break;
-    case 0x0000001B:
-        /* MTPR_TBIA */
-        val = env->ir[16];
-        if (cpu_alpha_mtpr(env, IPR_TBIA, val, &oldval) == 1)
-            env->ir[0] = val;
-        break;
-    case 0x0000001C:
-        /* MTPR_TBIAP */
-        val = env->ir[16];
-        if (cpu_alpha_mtpr(env, IPR_TBIAP, val, &oldval) == 1)
-            env->ir[0] = val;
-        break;
-    case 0x0000001D:
-        /* MTPR_TBIS */
-        val = env->ir[16];
-        if (cpu_alpha_mtpr(env, IPR_TBIS, val, &oldval) == 1)
-            env->ir[0] = val;
-        break;
-    case 0x0000001E:
-        /* MFPR_ESP */
-        if (cpu_alpha_mfpr(env, IPR_ESP, &val) == 0)
-            env->ir[0] = val;
-        break;
-    case 0x0000001F:
-        /* MTPR_ESP */
-        val = env->ir[16];
-        if (cpu_alpha_mtpr(env, IPR_ESP, val, &oldval) == 1)
-            env->ir[0] = val;
-        break;
-    case 0x00000020:
-        /* MFPR_SSP */
-        if (cpu_alpha_mfpr(env, IPR_SSP, &val) == 0)
-            env->ir[0] = val;
-        break;
-    case 0x00000021:
-        /* MTPR_SSP */
-        val = env->ir[16];
-        if (cpu_alpha_mtpr(env, IPR_SSP, val, &oldval) == 1)
-            env->ir[0] = val;
-        break;
-    case 0x00000022:
-        /* MFPR_USP */
-        if (cpu_alpha_mfpr(env, IPR_USP, &val) == 0)
-            env->ir[0] = val;
-        break;
-    case 0x00000023:
-        /* MTPR_USP */
-        val = env->ir[16];
-        if (cpu_alpha_mtpr(env, IPR_USP, val, &oldval) == 1)
-            env->ir[0] = val;
-        break;
-    case 0x00000024:
-        /* MTPR_TBISD */
-        val = env->ir[16];
-        if (cpu_alpha_mtpr(env, IPR_TBISD, val, &oldval) == 1)
-            env->ir[0] = val;
-        break;
-    case 0x00000025:
-        /* MTPR_TBISI */
-        val = env->ir[16];
-        if (cpu_alpha_mtpr(env, IPR_TBISI, val, &oldval) == 1)
-            env->ir[0] = val;
-        break;
-    case 0x00000026:
-        /* MFPR_ASTEN */
-        if (cpu_alpha_mfpr(env, IPR_ASTEN, &val) == 0)
-            env->ir[0] = val;
-        break;
-    case 0x00000027:
-        /* MFPR_ASTSR */
-        if (cpu_alpha_mfpr(env, IPR_ASTSR, &val) == 0)
-            env->ir[0] = val;
-        break;
-    case 0x00000029:
-        /* MFPR_VPTB */
-        if (cpu_alpha_mfpr(env, IPR_VPTB, &val) == 0)
-            env->ir[0] = val;
-        break;
-    case 0x0000002A:
-        /* MTPR_VPTB */
-        val = env->ir[16];
-        if (cpu_alpha_mtpr(env, IPR_VPTB, val, &oldval) == 1)
-            env->ir[0] = val;
-        break;
-    case 0x0000002B:
-        /* MTPR_PERFMON */
-        val = env->ir[16];
-        if (cpu_alpha_mtpr(env, IPR_PERFMON, val, &oldval) == 1)
-            env->ir[0] = val;
-        break;
-    case 0x0000002E:
-        /* MTPR_DATFX */
-        val = env->ir[16];
-        if (cpu_alpha_mtpr(env, IPR_DATFX, val, &oldval) == 1)
-            env->ir[0] = val;
-        break;
-    case 0x0000003E:
-        /* WTINT */
-        break;
-    case 0x0000003F:
-        /* MFPR_WHAMI */
-        if (cpu_alpha_mfpr(env, IPR_WHAMI, &val) == 0)
-            env->ir[0] = val;
-        break;
-    case 0x00000080:
-        /* BPT */
-        /* REQUIRED */
-        break;
-    case 0x00000081:
-        /* BUGCHK */
-        /* REQUIRED */
-        break;
-    case 0x00000082:
-        /* CHME */
-        break;
-    case 0x00000083:
-        /* CHMK */
-        break;
-    case 0x00000084:
-        /* CHMS */
-        break;
-    case 0x00000085:
-        /* CHMU */
-        break;
-    case 0x00000086:
-        /* IMB */
-        /* REQUIRED */
-        /* Implemented as no-op */
-        break;
-    case 0x00000087:
-        /* INSQHIL */
-        break;
-    case 0x00000088:
-        /* INSQTIL */
-        break;
-    case 0x00000089:
-        /* INSQHIQ */
-        break;
-    case 0x0000008A:
-        /* INSQTIQ */
-        break;
-    case 0x0000008B:
-        /* INSQUEL */
-        break;
-    case 0x0000008C:
-        /* INSQUEQ */
-        break;
-    case 0x0000008D:
-        /* INSQUEL/D */
-        break;
-    case 0x0000008E:
-        /* INSQUEQ/D */
-        break;
-    case 0x0000008F:
-        /* PROBER */
-        break;
-    case 0x00000090:
-        /* PROBEW */
-        break;
-    case 0x00000091:
-        /* RD_PS */
-        break;
-    case 0x00000092:
-        /* REI */
-        break;
-    case 0x00000093:
-        /* REMQHIL */
-        break;
-    case 0x00000094:
-        /* REMQTIL */
-        break;
-    case 0x00000095:
-        /* REMQHIQ */
-        break;
-    case 0x00000096:
-        /* REMQTIQ */
-        break;
-    case 0x00000097:
-        /* REMQUEL */
-        break;
-    case 0x00000098:
-        /* REMQUEQ */
-        break;
-    case 0x00000099:
-        /* REMQUEL/D */
-        break;
-    case 0x0000009A:
-        /* REMQUEQ/D */
-        break;
-    case 0x0000009B:
-        /* SWASTEN */
-        break;
-    case 0x0000009C:
-        /* WR_PS_SW */
-        break;
-    case 0x0000009D:
-        /* RSCC */
-        break;
-    case 0x0000009E:
-        /* READ_UNQ */
-        /* REQUIRED */
-        break;
-    case 0x0000009F:
-        /* WRITE_UNQ */
-        /* REQUIRED */
-        break;
-    case 0x000000A0:
-        /* AMOVRR */
-        break;
-    case 0x000000A1:
-        /* AMOVRM */
-        break;
-    case 0x000000A2:
-        /* INSQHILR */
-        break;
-    case 0x000000A3:
-        /* INSQTILR */
-        break;
-    case 0x000000A4:
-        /* INSQHIQR */
-        break;
-    case 0x000000A5:
-        /* INSQTIQR */
-        break;
-    case 0x000000A6:
-        /* REMQHILR */
-        break;
-    case 0x000000A7:
-        /* REMQTILR */
-        break;
-    case 0x000000A8:
-        /* REMQHIQR */
-        break;
-    case 0x000000A9:
-        /* REMQTIQR */
-        break;
-    case 0x000000AA:
-        /* GENTRAP */
-        /* REQUIRED */
-        break;
-    case 0x000000AE:
-        /* CLRFEN */
-        break;
-    default:
-        break;
-    }
-}
-
-static void pal_unix_call (CPUState *env, uint32_t palcode)
-{
-    uint64_t palid, val, oldval;
-
-    if (palcode < 0x00000080) {
-        /* Privileged palcodes */
-        if (!(env->ps >> 3)) {
-            /* TODO: generate privilege exception */
-        }
-    }
-    switch (palcode) {
-    case 0x00000000:
-        /* HALT */
-        /* REQUIRED */
-        break;
-    case 0x00000001:
-        /* CFLUSH */
-        break;
-    case 0x00000002:
-        /* DRAINA */
-        /* REQUIRED */
-        /* Implemented as no-op */
-        break;
-    case 0x00000009:
-        /* CSERVE */
-        /* REQUIRED */
-        break;
-    case 0x0000000A:
-        /* SWPPAL */
-        /* REQUIRED */
-        palid = env->ir[16];
-        do_swappal(env, palid);
-        break;
-    case 0x0000000D:
-        /* WRIPIR */
-        val = env->ir[16];
-        if (cpu_alpha_mtpr(env, IPR_IPIR, val, &oldval) == 1)
-            env->ir[0] = val;
-        break;
-    case 0x00000010:
-        /* RDMCES */
-        if (cpu_alpha_mfpr(env, IPR_MCES, &val) == 0)
-            env->ir[0] = val;
-        break;
-    case 0x00000011:
-        /* WRMCES */
-        val = env->ir[16];
-        if (cpu_alpha_mtpr(env, IPR_MCES, val, &oldval) == 1)
-            env->ir[0] = val;
-        break;
-    case 0x0000002B:
-        /* WRFEN */
-        val = env->ir[16];
-        if (cpu_alpha_mtpr(env, IPR_PERFMON, val, &oldval) == 1)
-            env->ir[0] = val;
-        break;
-    case 0x0000002D:
-        /* WRVPTPTR */
-        break;
-    case 0x00000030:
-        /* SWPCTX */
-        break;
-    case 0x00000031:
-        /* WRVAL */
-        break;
-    case 0x00000032:
-        /* RDVAL */
-        break;
-    case 0x00000033:
-        /* TBI */
-        val = env->ir[16];
-        if (cpu_alpha_mtpr(env, IPR_TBIS, val, &oldval) == 1)
-            env->ir[0] = val;
-        break;
-    case 0x00000034:
-        /* WRENT */
-        break;
-    case 0x00000035:
-        /* SWPIPL */
-        break;
-    case 0x00000036:
-        /* RDPS */
-        break;
-    case 0x00000037:
-        /* WRKGP */
-        break;
-    case 0x00000038:
-        /* WRUSP */
-        val = env->ir[16];
-        if (cpu_alpha_mtpr(env, IPR_USP, val, &oldval) == 1)
-            env->ir[0] = val;
-        break;
-    case 0x00000039:
-        /* WRPERFMON */
-        val = env->ir[16];
-        if (cpu_alpha_mtpr(env, IPR_PERFMON, val, &oldval) == 1)
-            env->ir[0] = val;
-        break;
-    case 0x0000003A:
-        /* RDUSP */
-        if (cpu_alpha_mfpr(env, IPR_USP, &val) == 0)
-            env->ir[0] = val;
-        break;
-    case 0x0000003C:
-        /* WHAMI */
-        if (cpu_alpha_mfpr(env, IPR_WHAMI, &val) == 0)
-            env->ir[0] = val;
-        break;
-    case 0x0000003D:
-        /* RETSYS */
-        break;
-    case 0x0000003E:
-        /* WTINT */
-        break;
-    case 0x0000003F:
-        /* RTI */
-        if (cpu_alpha_mfpr(env, IPR_WHAMI, &val) == 0)
-            env->ir[0] = val;
-        break;
-    case 0x00000080:
-        /* BPT */
-        /* REQUIRED */
-        break;
-    case 0x00000081:
-        /* BUGCHK */
-        /* REQUIRED */
-        break;
-    case 0x00000083:
-        /* CALLSYS */
-        break;
-    case 0x00000086:
-        /* IMB */
-        /* REQUIRED */
-        /* Implemented as no-op */
-        break;
-    case 0x00000092:
-        /* URTI */
-        break;
-    case 0x0000009E:
-        /* RDUNIQUE */
-        /* REQUIRED */
-        break;
-    case 0x0000009F:
-        /* WRUNIQUE */
-        /* REQUIRED */
-        break;
-    case 0x000000AA:
-        /* GENTRAP */
-        /* REQUIRED */
-        break;
-    case 0x000000AE:
-        /* CLRFEN */
-        break;
-    default:
-        break;
-    }
-}
-
-void call_pal (CPUState *env)
-{
-    pal_handler_t *pal_handler = env->pal_handler;
-
-    switch (env->exception_index) {
-    case EXCP_RESET:
-        (*pal_handler->reset)(env);
-        break;
-    case EXCP_MCHK:
-        (*pal_handler->machine_check)(env);
-        break;
-    case EXCP_ARITH:
-        (*pal_handler->arithmetic)(env);
-        break;
-    case EXCP_INTERRUPT:
-        (*pal_handler->interrupt)(env);
-        break;
-    case EXCP_DFAULT:
-        (*pal_handler->dfault)(env);
-        break;
-    case EXCP_DTB_MISS_PAL:
-        (*pal_handler->dtb_miss_pal)(env);
-        break;
-    case EXCP_DTB_MISS_NATIVE:
-        (*pal_handler->dtb_miss_native)(env);
-        break;
-    case EXCP_UNALIGN:
-        (*pal_handler->unalign)(env);
-        break;
-    case EXCP_ITB_MISS:
-        (*pal_handler->itb_miss)(env);
-        break;
-    case EXCP_ITB_ACV:
-        (*pal_handler->itb_acv)(env);
-        break;
-    case EXCP_OPCDEC:
-        (*pal_handler->opcdec)(env);
-        break;
-    case EXCP_FEN:
-        (*pal_handler->fen)(env);
-        break;
-    default:
-        if (env->exception_index >= EXCP_CALL_PAL &&
-            env->exception_index < EXCP_CALL_PALP) {
-            /* Unprivileged PAL call */
-            (*pal_handler->call_pal)
-                (env, (env->exception_index - EXCP_CALL_PAL) >> 6);
-        } else if (env->exception_index >= EXCP_CALL_PALP &&
-                   env->exception_index < EXCP_CALL_PALE) {
-            /* Privileged PAL call */
-            (*pal_handler->call_pal)
-                (env, ((env->exception_index - EXCP_CALL_PALP) >> 6) + 0x80);
-        } else {
-            /* Should never happen */
-        }
-        break;
-    }
-    env->ipr[IPR_EXC_ADDR] &= ~1;
-}
-
-void pal_init (CPUState *env)
-{
-    do_swappal(env, 0);
-}
-
-#if 0
-static uint64_t get_ptebase (CPUState *env, uint64_t vaddr)
-{
-    uint64_t virbnd, ptbr;
-
-    if ((env->features & FEATURE_VIRBND)) {
-        cpu_alpha_mfpr(env, IPR_VIRBND, &virbnd);
-        if (vaddr >= virbnd)
-            cpu_alpha_mfpr(env, IPR_SYSPTBR, &ptbr);
-        else
-            cpu_alpha_mfpr(env, IPR_PTBR, &ptbr);
-    } else {
-        cpu_alpha_mfpr(env, IPR_PTBR, &ptbr);
-    }
-
-    return ptbr;
-}
-
-static int get_page_bits (CPUState *env)
-{
-    /* XXX */
-    return 13;
-}
-
-static int get_pte (uint64_t *pfnp, int *zbitsp, int *protp,
-                    uint64_t ptebase, int page_bits, uint64_t level,
-                    int mmu_idx, int rw)
-{
-    uint64_t pteaddr, pte, pfn;
-    uint8_t gh;
-    int ure, uwe, kre, kwe, foE, foR, foW, v, ret, ar, is_user;
-
-    /* XXX: TOFIX */
-    is_user = mmu_idx == MMU_USER_IDX;
-    pteaddr = (ptebase << page_bits) + (8 * level);
-    pte = ldq_raw(pteaddr);
-    /* Decode all interresting PTE fields */
-    pfn = pte >> 32;
-    uwe = (pte >> 13) & 1;
-    kwe = (pte >> 12) & 1;
-    ure = (pte >> 9) & 1;
-    kre = (pte >> 8) & 1;
-    gh = (pte >> 5) & 3;
-    foE = (pte >> 3) & 1;
-    foW = (pte >> 2) & 1;
-    foR = (pte >> 1) & 1;
-    v = pte & 1;
-    ret = 0;
-    if (!v)
-        ret = 0x1;
-    /* Check access rights */
-    ar = 0;
-    if (is_user) {
-        if (ure)
-            ar |= PAGE_READ;
-        if (uwe)
-            ar |= PAGE_WRITE;
-        if (rw == 1 && !uwe)
-            ret |= 0x2;
-        if (rw != 1 && !ure)
-            ret |= 0x2;
-    } else {
-        if (kre)
-            ar |= PAGE_READ;
-        if (kwe)
-            ar |= PAGE_WRITE;
-        if (rw == 1 && !kwe)
-            ret |= 0x2;
-        if (rw != 1 && !kre)
-            ret |= 0x2;
-    }
-    if (rw == 0 && foR)
-        ret |= 0x4;
-    if (rw == 2 && foE)
-        ret |= 0x8;
-    if (rw == 1 && foW)
-        ret |= 0xC;
-    *pfnp = pfn;
-    if (zbitsp != NULL)
-        *zbitsp = page_bits + (3 * gh);
-    if (protp != NULL)
-        *protp = ar;
-
-    return ret;
-}
-
-static int paddr_from_pte (uint64_t *paddr, int *zbitsp, int *prot,
-                           uint64_t ptebase, int page_bits,
-                           uint64_t vaddr, int mmu_idx, int rw)
-{
-    uint64_t pfn, page_mask, lvl_mask, level1, level2, level3;
-    int lvl_bits, ret;
-
-    page_mask = (1ULL << page_bits) - 1ULL;
-    lvl_bits = page_bits - 3;
-    lvl_mask = (1ULL << lvl_bits) - 1ULL;
-    level3 = (vaddr >> page_bits) & lvl_mask;
-    level2 = (vaddr >> (page_bits + lvl_bits)) & lvl_mask;
-    level1 = (vaddr >> (page_bits + (2 * lvl_bits))) & lvl_mask;
-    /* Level 1 PTE */
-    ret = get_pte(&pfn, NULL, NULL, ptebase, page_bits, level1, 0, 0);
-    switch (ret) {
-    case 3:
-        /* Access violation */
-        return 2;
-    case 2:
-        /* translation not valid */
-        return 1;
-    default:
-        /* OK */
-        break;
-    }
-    /* Level 2 PTE */
-    ret = get_pte(&pfn, NULL, NULL, pfn, page_bits, level2, 0, 0);
-    switch (ret) {
-    case 3:
-        /* Access violation */
-        return 2;
-    case 2:
-        /* translation not valid */
-        return 1;
-    default:
-        /* OK */
-        break;
-    }
-    /* Level 3 PTE */
-    ret = get_pte(&pfn, zbitsp, prot, pfn, page_bits, level3, mmu_idx, rw);
-    if (ret & 0x1) {
-        /* Translation not valid */
-        ret = 1;
-    } else if (ret & 2) {
-        /* Access violation */
-        ret = 2;
-    } else {
-        switch (ret & 0xC) {
-        case 0:
-            /* OK */
-            ret = 0;
-            break;
-        case 0x4:
-            /* Fault on read */
-            ret = 3;
-            break;
-        case 0x8:
-            /* Fault on execute */
-            ret = 4;
-            break;
-        case 0xC:
-            /* Fault on write */
-            ret = 5;
-            break;
-        }
-    }
-    *paddr = (pfn << page_bits) | (vaddr & page_mask);
-
-    return 0;
-}
-
-static int virtual_to_physical (CPUState *env, uint64_t *physp,
-                                int *zbitsp, int *protp,
-                                uint64_t virtual, int mmu_idx, int rw)
-{
-    uint64_t sva, ptebase;
-    int seg, page_bits, ret;
-
-    sva = ((int64_t)(virtual << (64 - VA_BITS))) >> (64 - VA_BITS);
-    if (sva != virtual)
-        seg = -1;
-    else
-        seg = sva >> (VA_BITS - 2);
-    virtual &= ~(0xFFFFFC0000000000ULL << (VA_BITS - 43));
-    ptebase = get_ptebase(env, virtual);
-    page_bits = get_page_bits(env);
-    ret = 0;
-    switch (seg) {
-    case 0:
-        /* seg1: 3 levels of PTE */
-        ret = paddr_from_pte(physp, zbitsp, protp, ptebase, page_bits,
-                             virtual, mmu_idx, rw);
-        break;
-    case 1:
-        /* seg1: 2 levels of PTE */
-        ret = paddr_from_pte(physp, zbitsp, protp, ptebase, page_bits,
-                             virtual, mmu_idx, rw);
-        break;
-    case 2:
-        /* kernel segment */
-        if (mmu_idx != 0) {
-            ret = 2;
-        } else {
-            *physp = virtual;
-        }
-        break;
-    case 3:
-        /* seg1: TB mapped */
-        ret = paddr_from_pte(physp, zbitsp, protp, ptebase, page_bits,
-                             virtual, mmu_idx, rw);
-        break;
-    default:
-        ret = 1;
-        break;
-    }
-
-    return ret;
-}
-
-/* XXX: code provision */
-int cpu_ppc_handle_mmu_fault (CPUState *env, uint32_t address, int rw,
-                              int mmu_idx, int is_softmmu)
-{
-    uint64_t physical, page_size, end;
-    int prot, zbits, ret;
-
-    ret = virtual_to_physical(env, &physical, &zbits, &prot,
-                              address, mmu_idx, rw);
-
-    switch (ret) {
-    case 0:
-        /* No fault */
-        page_size = 1ULL << zbits;
-        address &= ~(page_size - 1);
-        /* FIXME: page_size should probably be passed to tlb_set_page,
-           and this loop removed.   */
-        for (end = physical + page_size; physical < end; physical += 0x1000) {
-            tlb_set_page(env, address, physical, prot, mmu_idx,
-                         TARGET_PAGE_SIZE);
-            address += 0x1000;
-        }
-        ret = 0;
-        break;
-#if 0
-    case 1:
-        env->exception_index = EXCP_DFAULT;
-        env->ipr[IPR_EXC_ADDR] = address;
-        ret = 1;
-        break;
-    case 2:
-        env->exception_index = EXCP_ACCESS_VIOLATION;
-        env->ipr[IPR_EXC_ADDR] = address;
-        ret = 1;
-        break;
-    case 3:
-        env->exception_index = EXCP_FAULT_ON_READ;
-        env->ipr[IPR_EXC_ADDR] = address;
-        ret = 1;
-        break;
-    case 4:
-        env->exception_index = EXCP_FAULT_ON_EXECUTE;
-        env->ipr[IPR_EXC_ADDR] = address;
-        ret = 1;
-    case 5:
-        env->exception_index = EXCP_FAULT_ON_WRITE;
-        env->ipr[IPR_EXC_ADDR] = address;
-        ret = 1;
-#endif
-    default:
-        /* Should never happen */
-        env->exception_index = EXCP_MCHK;
-        env->ipr[IPR_EXC_ADDR] = address;
-        ret = 1;
-        break;
-    }
-
-    return ret;
-}
-#endif
diff --git a/hw/ide/core.c b/hw/ide/core.c
index 45410e8..ca17a43 100644
--- a/hw/ide/core.c
+++ b/hw/ide/core.c
@@ -78,7 +78,7 @@
 {
     uint16_t *p;
     unsigned int oldsize;
-    IDEDevice *dev;
+    IDEDevice *dev = s->unit ? s->bus->slave : s->bus->master;
 
     if (s->identify_set) {
 	memcpy(s->io_buffer, s->identify_data, sizeof(s->identify_data));
@@ -124,6 +124,9 @@
     put_le16(p + 66, 120);
     put_le16(p + 67, 120);
     put_le16(p + 68, 120);
+    if (dev && dev->conf.discard_granularity) {
+        put_le16(p + 69, (1 << 14)); /* determinate TRIM behavior */
+    }
 
     if (s->ncq_queues) {
         put_le16(p + 75, s->ncq_queues - 1);
@@ -154,9 +157,12 @@
     put_le16(p + 101, s->nb_sectors >> 16);
     put_le16(p + 102, s->nb_sectors >> 32);
     put_le16(p + 103, s->nb_sectors >> 48);
-    dev = s->unit ? s->bus->slave : s->bus->master;
+
     if (dev && dev->conf.physical_block_size)
         put_le16(p + 106, 0x6000 | get_physical_block_exp(&dev->conf));
+    if (dev && dev->conf.discard_granularity) {
+        put_le16(p + 169, 1); /* TRIM support */
+    }
 
     memcpy(s->identify_data, p, sizeof(s->identify_data));
     s->identify_set = 1;
@@ -299,6 +305,74 @@
     }
 }
 
+typedef struct TrimAIOCB {
+    BlockDriverAIOCB common;
+    QEMUBH *bh;
+    int ret;
+} TrimAIOCB;
+
+static void trim_aio_cancel(BlockDriverAIOCB *acb)
+{
+    TrimAIOCB *iocb = container_of(acb, TrimAIOCB, common);
+
+    qemu_bh_delete(iocb->bh);
+    iocb->bh = NULL;
+    qemu_aio_release(iocb);
+}
+
+static AIOPool trim_aio_pool = {
+    .aiocb_size         = sizeof(TrimAIOCB),
+    .cancel             = trim_aio_cancel,
+};
+
+static void ide_trim_bh_cb(void *opaque)
+{
+    TrimAIOCB *iocb = opaque;
+
+    iocb->common.cb(iocb->common.opaque, iocb->ret);
+
+    qemu_bh_delete(iocb->bh);
+    iocb->bh = NULL;
+
+    qemu_aio_release(iocb);
+}
+
+BlockDriverAIOCB *ide_issue_trim(BlockDriverState *bs,
+        int64_t sector_num, QEMUIOVector *qiov, int nb_sectors,
+        BlockDriverCompletionFunc *cb, void *opaque)
+{
+    TrimAIOCB *iocb;
+    int i, j, ret;
+
+    iocb = qemu_aio_get(&trim_aio_pool, bs, cb, opaque);
+    iocb->bh = qemu_bh_new(ide_trim_bh_cb, iocb);
+    iocb->ret = 0;
+
+    for (j = 0; j < qiov->niov; j++) {
+        uint64_t *buffer = qiov->iov[j].iov_base;
+
+        for (i = 0; i < qiov->iov[j].iov_len / 8; i++) {
+            /* 6-byte LBA + 2-byte range per entry */
+            uint64_t entry = le64_to_cpu(buffer[i]);
+            uint64_t sector = entry & 0x0000ffffffffffffULL;
+            uint16_t count = entry >> 48;
+
+            if (count == 0) {
+                break;
+            }
+
+            ret = bdrv_discard(bs, sector, count);
+            if (!iocb->ret) {
+                iocb->ret = ret;
+            }
+        }
+    }
+
+    qemu_bh_schedule(iocb->bh);
+
+    return &iocb->common;
+}
+
 static inline void ide_abort_command(IDEState *s)
 {
     s->status = READY_STAT | ERR_STAT;
@@ -430,7 +504,6 @@
     s->error = ABRT_ERR;
     s->status = READY_STAT | ERR_STAT;
     ide_set_inactive(s);
-    s->bus->dma->ops->add_status(s->bus->dma, BM_STATUS_INT);
     ide_set_irq(s->bus);
 }
 
@@ -447,7 +520,7 @@
     if ((error == ENOSPC && action == BLOCK_ERR_STOP_ENOSPC)
             || action == BLOCK_ERR_STOP_ANY) {
         s->bus->dma->ops->set_unit(s->bus->dma, s->unit);
-        s->bus->dma->ops->add_status(s->bus->dma, op);
+        s->bus->error_status = op;
         bdrv_mon_event(s->bs, BDRV_ACTION_STOP, is_read);
         vm_stop(VMSTOP_DISKFULL);
     } else {
@@ -473,8 +546,11 @@
     if (ret < 0) {
         int op = BM_STATUS_DMA_RETRY;
 
-        if (s->is_read)
+        if (s->dma_cmd == IDE_DMA_READ)
             op |= BM_STATUS_RETRY_READ;
+        else if (s->dma_cmd == IDE_DMA_TRIM)
+            op |= BM_STATUS_RETRY_TRIM;
+
         if (ide_handle_rw_error(s, -ret, op)) {
             return;
         }
@@ -483,7 +559,7 @@
     n = s->io_buffer_size >> 9;
     sector_num = ide_get_sector(s);
     if (n > 0) {
-        dma_buf_commit(s, s->is_read);
+        dma_buf_commit(s, ide_cmd_is_read(s));
         sector_num += n;
         ide_set_sector(s, sector_num);
         s->nsector -= n;
@@ -500,20 +576,30 @@
     n = s->nsector;
     s->io_buffer_index = 0;
     s->io_buffer_size = n * 512;
-    if (s->bus->dma->ops->prepare_buf(s->bus->dma, s->is_read) == 0)
+    if (s->bus->dma->ops->prepare_buf(s->bus->dma, ide_cmd_is_read(s)) == 0) {
+        /* The PRDs were too short. Reset the Active bit, but don't raise an
+         * interrupt. */
         goto eot;
+    }
 
 #ifdef DEBUG_AIO
-    printf("ide_dma_cb: sector_num=%" PRId64 " n=%d, is_read=%d\n",
-           sector_num, n, s->is_read);
+    printf("ide_dma_cb: sector_num=%" PRId64 " n=%d, cmd_cmd=%d\n",
+           sector_num, n, s->dma_cmd);
 #endif
 
-    if (s->is_read) {
+    switch (s->dma_cmd) {
+    case IDE_DMA_READ:
         s->bus->dma->aiocb = dma_bdrv_read(s->bs, &s->sg, sector_num,
                                            ide_dma_cb, s);
-    } else {
+        break;
+    case IDE_DMA_WRITE:
         s->bus->dma->aiocb = dma_bdrv_write(s->bs, &s->sg, sector_num,
                                             ide_dma_cb, s);
+        break;
+    case IDE_DMA_TRIM:
+        s->bus->dma->aiocb = dma_bdrv_io(s->bs, &s->sg, sector_num,
+                                         ide_issue_trim, ide_dma_cb, s, 1);
+        break;
     }
 
     if (!s->bus->dma->aiocb) {
@@ -523,16 +609,15 @@
     return;
 
 eot:
-   s->bus->dma->ops->add_status(s->bus->dma, BM_STATUS_INT);
    ide_set_inactive(s);
 }
 
-static void ide_sector_start_dma(IDEState *s, int is_read)
+static void ide_sector_start_dma(IDEState *s, enum ide_dma_cmd dma_cmd)
 {
     s->status = READY_STAT | SEEK_STAT | DRQ_STAT | BUSY_STAT;
     s->io_buffer_index = 0;
     s->io_buffer_size = 0;
-    s->is_read = is_read;
+    s->dma_cmd = dma_cmd;
     s->bus->dma->ops->start_dma(s->bus->dma, s, ide_dma_cb);
 }
 
@@ -814,6 +899,18 @@
         return;
 
     switch(val) {
+    case WIN_DSM:
+        switch (s->feature) {
+        case DSM_TRIM:
+            if (!s->bs) {
+                goto abort_cmd;
+            }
+            ide_sector_start_dma(s, IDE_DMA_TRIM);
+            break;
+        default:
+            goto abort_cmd;
+        }
+        break;
     case WIN_IDENTIFY:
         if (s->bs && s->drive_kind != IDE_CD) {
             if (s->drive_kind != IDE_CFATA)
@@ -915,7 +1012,7 @@
         if (!s->bs)
             goto abort_cmd;
 	ide_cmd_lba48_transform(s, lba48);
-        ide_sector_start_dma(s, 1);
+        ide_sector_start_dma(s, IDE_DMA_READ);
         break;
 	case WIN_WRITEDMA_EXT:
 	lba48 = 1;
@@ -924,7 +1021,7 @@
         if (!s->bs)
             goto abort_cmd;
 	ide_cmd_lba48_transform(s, lba48);
-        ide_sector_start_dma(s, 0);
+        ide_sector_start_dma(s, IDE_DMA_WRITE);
         s->media_changed = 1;
         break;
     case WIN_READ_NATIVE_MAX_EXT:
@@ -1836,7 +1933,8 @@
 {
     IDEState *s = opaque;
 
-    return (s->status & DRQ_STAT) != 0;
+    return ((s->status & DRQ_STAT) != 0)
+        || (s->bus->error_status & BM_STATUS_PIO_RETRY);
 }
 
 static bool ide_atapi_gesn_needed(void *opaque)
@@ -1846,6 +1944,13 @@
     return s->events.new_media || s->events.eject_request;
 }
 
+static bool ide_error_needed(void *opaque)
+{
+    IDEBus *bus = opaque;
+
+    return (bus->error_status != 0);
+}
+
 /* Fields for GET_EVENT_STATUS_NOTIFICATION ATAPI command */
 const VMStateDescription vmstate_ide_atapi_gesn_state = {
     .name ="ide_drive/atapi/gesn_state",
@@ -1855,6 +1960,7 @@
     .fields = (VMStateField []) {
         VMSTATE_BOOL(events.new_media, IDEState),
         VMSTATE_BOOL(events.eject_request, IDEState),
+        VMSTATE_END_OF_LIST()
     }
 };
 
@@ -1920,6 +2026,17 @@
     }
 };
 
+const VMStateDescription vmstate_ide_error_status = {
+    .name ="ide_bus/error",
+    .version_id = 1,
+    .minimum_version_id = 1,
+    .minimum_version_id_old = 1,
+    .fields = (VMStateField []) {
+        VMSTATE_INT32(error_status, IDEBus),
+        VMSTATE_END_OF_LIST()
+    }
+};
+
 const VMStateDescription vmstate_ide_bus = {
     .name = "ide_bus",
     .version_id = 1,
@@ -1929,6 +2046,14 @@
         VMSTATE_UINT8(cmd, IDEBus),
         VMSTATE_UINT8(unit, IDEBus),
         VMSTATE_END_OF_LIST()
+    },
+    .subsections = (VMStateSubsection []) {
+        {
+            .vmsd = &vmstate_ide_error_status,
+            .needed = ide_error_needed,
+        }, {
+            /* empty */
+        }
     }
 };
 
diff --git a/hw/ide/internal.h b/hw/ide/internal.h
index c2b35ec..02e805f 100644
--- a/hw/ide/internal.h
+++ b/hw/ide/internal.h
@@ -62,7 +62,11 @@
  */
 #define CFA_REQ_EXT_ERROR_CODE		0x03 /* CFA Request Extended Error Code */
 /*
- *	0x04->0x07 Reserved
+ *      0x04->0x05 Reserved
+ */
+#define WIN_DSM                         0x06
+/*
+ *      0x07 Reserved
  */
 #define WIN_SRST			0x08 /* ATAPI soft reset command */
 #define WIN_DEVICE_RESET		0x08
@@ -190,6 +194,9 @@
 
 #define IDE_DMA_BUF_SECTORS 256
 
+/* feature values for Data Set Management */
+#define DSM_TRIM                        0x01
+
 #if (IDE_DMA_BUF_SECTORS < MAX_MULT_SECTORS)
 #error "IDE_DMA_BUF_SECTORS must be bigger or equal to MAX_MULT_SECTORS"
 #endif
@@ -379,6 +386,15 @@
     bool new_media;
 };
 
+enum ide_dma_cmd {
+    IDE_DMA_READ,
+    IDE_DMA_WRITE,
+    IDE_DMA_TRIM,
+};
+
+#define ide_cmd_is_read(s) \
+	((s)->dma_cmd == IDE_DMA_READ)
+
 /* NOTE: IDEState represents in fact one drive */
 struct IDEState {
     IDEBus *bus;
@@ -446,7 +462,7 @@
     uint32_t mdata_size;
     uint8_t *mdata_storage;
     int media_changed;
-    int is_read;
+    enum ide_dma_cmd dma_cmd;
     /* SMART */
     uint8_t smart_enabled;
     uint8_t smart_autosave;
@@ -486,6 +502,8 @@
     uint8_t unit;
     uint8_t cmd;
     qemu_irq irq;
+
+    int error_status;
 };
 
 struct IDEDevice {
@@ -505,10 +523,17 @@
 #define BM_STATUS_DMAING 0x01
 #define BM_STATUS_ERROR  0x02
 #define BM_STATUS_INT    0x04
+
+/* FIXME These are not status register bits */
 #define BM_STATUS_DMA_RETRY  0x08
 #define BM_STATUS_PIO_RETRY  0x10
 #define BM_STATUS_RETRY_READ  0x20
 #define BM_STATUS_RETRY_FLUSH 0x40
+#define BM_STATUS_RETRY_TRIM 0x80
+
+#define BM_MIGRATION_COMPAT_STATUS_BITS \
+        (BM_STATUS_DMA_RETRY | BM_STATUS_PIO_RETRY | \
+        BM_STATUS_RETRY_READ | BM_STATUS_RETRY_FLUSH)
 
 #define BM_CMD_START     0x01
 #define BM_CMD_READ      0x08
@@ -575,6 +600,9 @@
                         EndTransferFunc *end_transfer_func);
 void ide_transfer_stop(IDEState *s);
 void ide_set_inactive(IDEState *s);
+BlockDriverAIOCB *ide_issue_trim(BlockDriverState *bs,
+        int64_t sector_num, QEMUIOVector *qiov, int nb_sectors,
+        BlockDriverCompletionFunc *cb, void *opaque);
 
 /* hw/ide/atapi.c */
 void ide_atapi_cmd(IDEState *s);
diff --git a/hw/ide/macio.c b/hw/ide/macio.c
index 7107f6b..7daeb31 100644
--- a/hw/ide/macio.c
+++ b/hw/ide/macio.c
@@ -145,12 +145,21 @@
     io->addr += io->len;
     io->len = 0;
 
-    if (s->is_read)
+    switch (s->dma_cmd) {
+    case IDE_DMA_READ:
         m->aiocb = dma_bdrv_read(s->bs, &s->sg, sector_num,
 		                 pmac_ide_transfer_cb, io);
-    else
+        break;
+    case IDE_DMA_WRITE:
         m->aiocb = dma_bdrv_write(s->bs, &s->sg, sector_num,
 		                  pmac_ide_transfer_cb, io);
+        break;
+    case IDE_DMA_TRIM:
+        m->aiocb = dma_bdrv_io(s->bs, &s->sg, sector_num,
+                               ide_issue_trim, pmac_ide_transfer_cb, s, 1);
+        break;
+    }
+
     if (!m->aiocb)
         pmac_ide_transfer_cb(io, -1);
 }
diff --git a/hw/ide/pci.c b/hw/ide/pci.c
index a4726ad..9f3050a 100644
--- a/hw/ide/pci.c
+++ b/hw/ide/pci.c
@@ -169,7 +169,7 @@
     return 0;
 }
 
-static void bmdma_restart_dma(BMDMAState *bm, int is_read)
+static void bmdma_restart_dma(BMDMAState *bm, enum ide_dma_cmd dma_cmd)
 {
     IDEState *s = bmdma_active_if(bm);
 
@@ -177,33 +177,48 @@
     s->io_buffer_index = 0;
     s->io_buffer_size = 0;
     s->nsector = bm->nsector;
-    s->is_read = is_read;
+    s->dma_cmd = dma_cmd;
     bm->cur_addr = bm->addr;
     bm->dma_cb = ide_dma_cb;
     bmdma_start_dma(&bm->dma, s, bm->dma_cb);
 }
 
+/* TODO This should be common IDE code */
 static void bmdma_restart_bh(void *opaque)
 {
     BMDMAState *bm = opaque;
+    IDEBus *bus = bm->bus;
     int is_read;
+    int error_status;
 
     qemu_bh_delete(bm->bh);
     bm->bh = NULL;
 
-    is_read = !!(bm->status & BM_STATUS_RETRY_READ);
+    if (bm->unit == (uint8_t) -1) {
+        return;
+    }
 
-    if (bm->status & BM_STATUS_DMA_RETRY) {
-        bm->status &= ~(BM_STATUS_DMA_RETRY | BM_STATUS_RETRY_READ);
-        bmdma_restart_dma(bm, is_read);
-    } else if (bm->status & BM_STATUS_PIO_RETRY) {
-        bm->status &= ~(BM_STATUS_PIO_RETRY | BM_STATUS_RETRY_READ);
+    is_read = !!(bus->error_status & BM_STATUS_RETRY_READ);
+
+    /* The error status must be cleared before resubmitting the request: The
+     * request may fail again, and this case can only be distinguished if the
+     * called function can set a new error status. */
+    error_status = bus->error_status;
+    bus->error_status = 0;
+
+    if (error_status & BM_STATUS_DMA_RETRY) {
+        if (error_status & BM_STATUS_RETRY_TRIM) {
+            bmdma_restart_dma(bm, IDE_DMA_TRIM);
+        } else {
+            bmdma_restart_dma(bm, is_read ? IDE_DMA_READ : IDE_DMA_WRITE);
+        }
+    } else if (error_status & BM_STATUS_PIO_RETRY) {
         if (is_read) {
             ide_sector_read(bmdma_active_if(bm));
         } else {
             ide_sector_write(bmdma_active_if(bm));
         }
-    } else if (bm->status & BM_STATUS_RETRY_FLUSH) {
+    } else if (error_status & BM_STATUS_RETRY_FLUSH) {
         ide_flush_cache(bmdma_active_if(bm));
     }
 }
@@ -351,6 +366,43 @@
     return (bm->cur_prd_len != 0);
 }
 
+static bool ide_bmdma_status_needed(void *opaque)
+{
+    BMDMAState *bm = opaque;
+
+    /* Older versions abused some bits in the status register for internal
+     * error state. If any of these bits are set, we must add a subsection to
+     * transfer the real status register */
+    uint8_t abused_bits = BM_MIGRATION_COMPAT_STATUS_BITS;
+
+    return ((bm->status & abused_bits) != 0);
+}
+
+static void ide_bmdma_pre_save(void *opaque)
+{
+    BMDMAState *bm = opaque;
+    uint8_t abused_bits = BM_MIGRATION_COMPAT_STATUS_BITS;
+
+    bm->migration_compat_status =
+        (bm->status & ~abused_bits) | (bm->bus->error_status & abused_bits);
+}
+
+/* This function accesses bm->bus->error_status which is loaded only after
+ * BMDMA itself. This is why the function is called from ide_pci_post_load
+ * instead of being registered with VMState where it would run too early. */
+static int ide_bmdma_post_load(void *opaque, int version_id)
+{
+    BMDMAState *bm = opaque;
+    uint8_t abused_bits = BM_MIGRATION_COMPAT_STATUS_BITS;
+
+    if (bm->status == 0) {
+        bm->status = bm->migration_compat_status & ~abused_bits;
+        bm->bus->error_status |= bm->migration_compat_status & abused_bits;
+    }
+
+    return 0;
+}
+
 static const VMStateDescription vmstate_bmdma_current = {
     .name = "ide bmdma_current",
     .version_id = 1,
@@ -365,15 +417,26 @@
     }
 };
 
+const VMStateDescription vmstate_bmdma_status = {
+    .name ="ide bmdma/status",
+    .version_id = 1,
+    .minimum_version_id = 1,
+    .minimum_version_id_old = 1,
+    .fields = (VMStateField []) {
+        VMSTATE_UINT8(status, BMDMAState),
+        VMSTATE_END_OF_LIST()
+    }
+};
 
 static const VMStateDescription vmstate_bmdma = {
     .name = "ide bmdma",
     .version_id = 3,
     .minimum_version_id = 0,
     .minimum_version_id_old = 0,
+    .pre_save  = ide_bmdma_pre_save,
     .fields      = (VMStateField []) {
         VMSTATE_UINT8(cmd, BMDMAState),
-        VMSTATE_UINT8(status, BMDMAState),
+        VMSTATE_UINT8(migration_compat_status, BMDMAState),
         VMSTATE_UINT32(addr, BMDMAState),
         VMSTATE_INT64(sector_num, BMDMAState),
         VMSTATE_UINT32(nsector, BMDMAState),
@@ -385,6 +448,9 @@
             .vmsd = &vmstate_bmdma_current,
             .needed = ide_bmdma_current_needed,
         }, {
+            .vmsd = &vmstate_bmdma_status,
+            .needed = ide_bmdma_status_needed,
+        }, {
             /* empty */
         }
     }
@@ -399,7 +465,9 @@
         /* current versions always store 0/1, but older version
            stored bigger values. We only need last bit */
         d->bmdma[i].unit &= 1;
+        ide_bmdma_post_load(&d->bmdma[i], -1);
     }
+
     return 0;
 }
 
diff --git a/hw/ide/pci.h b/hw/ide/pci.h
index cd72cba..b4f3691 100644
--- a/hw/ide/pci.h
+++ b/hw/ide/pci.h
@@ -22,6 +22,10 @@
     IORange addr_ioport;
     QEMUBH *bh;
     qemu_irq irq;
+
+    /* Bit 0-2 and 7:   BM status register
+     * Bit 3-6:         bus->error_status */
+    uint8_t migration_compat_status;
 } BMDMAState;
 
 typedef struct PCIIDEState {
diff --git a/hw/ide/qdev.c b/hw/ide/qdev.c
index 3f9dc89..d9b8f24 100644
--- a/hw/ide/qdev.c
+++ b/hw/ide/qdev.c
@@ -125,6 +125,11 @@
     const char *serial;
     DriveInfo *dinfo;
 
+    if (dev->conf.discard_granularity && dev->conf.discard_granularity != 512) {
+        error_report("discard_granularity must be 512 for ide");
+        return -1;
+    }
+
     serial = dev->serial;
     if (!serial) {
         /* try to fall back to value set with legacy -drive serial=... */
diff --git a/hw/milkymist-softusb.c b/hw/milkymist-softusb.c
index 1565260..028f3b7 100644
--- a/hw/milkymist-softusb.c
+++ b/hw/milkymist-softusb.c
@@ -247,10 +247,18 @@
 {
 }
 
+static void softusb_device_destroy(USBBus *bus, USBDevice *dev)
+{
+}
+
 static USBPortOps softusb_ops = {
     .attach = softusb_attach,
 };
 
+static USBBusOps softusb_bus_ops = {
+    .device_destroy = softusb_device_destroy,
+};
+
 static void milkymist_softusb_reset(DeviceState *d)
 {
     MilkymistSoftUsbState *s =
@@ -294,7 +302,7 @@
     qemu_add_mouse_event_handler(softusb_mouse_event, s, 0, "Milkymist Mouse");
 
     /* create our usb bus */
-    usb_bus_new(&s->usbbus, NULL);
+    usb_bus_new(&s->usbbus, &softusb_bus_ops, NULL);
 
     /* our two ports */
     usb_register_port(&s->usbbus, &s->usbport[0], NULL, 0, &softusb_ops,
diff --git a/hw/multiboot.c b/hw/multiboot.c
index 394ed01..6e6cfb9 100644
--- a/hw/multiboot.c
+++ b/hw/multiboot.c
@@ -307,7 +307,7 @@
                                 | MULTIBOOT_FLAGS_MMAP);
     stl_p(bootinfo + MBI_MEM_LOWER,   640);
     stl_p(bootinfo + MBI_MEM_UPPER,   (ram_size / 1024) - 1024);
-    stl_p(bootinfo + MBI_BOOT_DEVICE, 0x8001ffff); /* XXX: use the -boot switch? */
+    stl_p(bootinfo + MBI_BOOT_DEVICE, 0x8000ffff); /* XXX: use the -boot switch? */
     stl_p(bootinfo + MBI_MMAP_ADDR,   ADDR_E820_MAP);
 
     mb_debug("multiboot: mh_entry_addr = %#x\n", mh_entry_addr);
diff --git a/hw/pc.c b/hw/pc.c
index 8106197..a3e8539 100644
--- a/hw/pc.c
+++ b/hw/pc.c
@@ -1070,6 +1070,15 @@
             isa_vga_init();
         }
     }
+
+    /*
+     * sga does not suppress normal vga output. So a machine can have both a
+     * vga card and sga manually enabled. Output will be seen on both.
+     * For nographic case, sga is enabled at all times
+     */
+    if (display_type == DT_NOGRAPHIC) {
+        isa_create_simple("sga");
+    }
 }
 
 static void cpu_request_exit(void *opaque, int irq, int level)
diff --git a/hw/ppce500_mpc8544ds.c b/hw/ppce500_mpc8544ds.c
index 17b0165..6b57fbf 100644
--- a/hw/ppce500_mpc8544ds.c
+++ b/hw/ppce500_mpc8544ds.c
@@ -275,7 +275,7 @@
                                 mpic[pci_irq_nrs[0]], mpic[pci_irq_nrs[1]],
                                 mpic[pci_irq_nrs[2]], mpic[pci_irq_nrs[3]],
                                 NULL);
-    pci_bus = (PCIBus *)qdev_get_child_bus(dev, "pci");
+    pci_bus = (PCIBus *)qdev_get_child_bus(dev, "pci.0");
     if (!pci_bus)
         printf("couldn't create PCI controller!\n");
 
diff --git a/hw/qxl.c b/hw/qxl.c
index 2bb36c6..1906e84 100644
--- a/hw/qxl.c
+++ b/hw/qxl.c
@@ -357,7 +357,9 @@
             ret = true;
         }
         qemu_mutex_unlock(&qxl->ssd.lock);
-        qxl_log_command(qxl, "vga", ext);
+        if (ret) {
+            qxl_log_command(qxl, "vga", ext);
+        }
         return ret;
     case QXL_MODE_COMPAT:
     case QXL_MODE_NATIVE:
diff --git a/hw/scsi-bus.c b/hw/scsi-bus.c
index 837f24e..ad6a730 100644
--- a/hw/scsi-bus.c
+++ b/hw/scsi-bus.c
@@ -413,7 +413,11 @@
     scsi_req_xfer_mode(req);
     req->cmd.lba = scsi_req_lba(req);
     trace_scsi_req_parsed(req->dev->id, req->lun, req->tag, buf[0],
-                          req->cmd.mode, req->cmd.xfer, req->cmd.lba);
+                          req->cmd.mode, req->cmd.xfer);
+    if (req->cmd.lba != -1) {
+        trace_scsi_req_parsed_lba(req->dev->id, req->lun, req->tag, buf[0],
+                              req->cmd.lba);
+    }
     return 0;
 }
 
diff --git a/hw/sga.c b/hw/sga.c
new file mode 100644
index 0000000..7ef750a
--- /dev/null
+++ b/hw/sga.c
@@ -0,0 +1,56 @@
+/*
+ * QEMU dummy ISA device for loading sgabios option rom.
+ *
+ * Copyright (c) 2011 Glauber Costa, Red Hat Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a copy
+ * of this software and associated documentation files (the "Software"), to deal
+ * in the Software without restriction, including without limitation the rights
+ * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+ * copies of the Software, and to permit persons to whom the Software is
+ * furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
+ * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+ * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
+ * THE SOFTWARE.
+ *
+ * sgabios code originally available at code.google.com/p/sgabios
+ *
+ */
+#include "pci.h"
+#include "pc.h"
+#include "loader.h"
+#include "sysemu.h"
+
+#define SGABIOS_FILENAME "sgabios.bin"
+
+typedef struct ISAGAState {
+    ISADevice dev;
+} ISASGAState;
+
+static int isa_cirrus_vga_initfn(ISADevice *dev)
+{
+    rom_add_vga(SGABIOS_FILENAME);
+    return 0;
+}
+
+static ISADeviceInfo sga_info = {
+    .qdev.name    = "sga",
+    .qdev.desc    = "Serial Graphics Adapter",
+    .qdev.size    = sizeof(ISASGAState),
+    .init         = isa_cirrus_vga_initfn,
+};
+
+static void sga_register(void)
+{
+      isa_qdev_register(&sga_info);
+}
+
+device_init(sga_register);
diff --git a/hw/usb-bus.c b/hw/usb-bus.c
index abc7e61..480956d 100644
--- a/hw/usb-bus.c
+++ b/hw/usb-bus.c
@@ -39,9 +39,10 @@
     }
 };
 
-void usb_bus_new(USBBus *bus, DeviceState *host)
+void usb_bus_new(USBBus *bus, USBBusOps *ops, DeviceState *host)
 {
     qbus_create_inplace(&bus->qbus, &usb_bus_info, host, NULL);
+    bus->ops = ops;
     bus->busnr = next_usb_bus++;
     bus->qbus.allow_hotplug = 1; /* Yes, we can */
     QTAILQ_INIT(&bus->free);
@@ -81,8 +82,12 @@
 static int usb_qdev_exit(DeviceState *qdev)
 {
     USBDevice *dev = DO_UPCAST(USBDevice, qdev, qdev);
+    USBBus *bus = usb_bus_from_device(dev);
 
-    usb_device_detach(dev);
+    if (dev->attached) {
+        usb_device_detach(dev);
+    }
+    bus->ops->device_destroy(bus, dev);
     if (dev->info->handle_destroy) {
         dev->info->handle_destroy(dev);
     }
@@ -270,6 +275,7 @@
         [ USB_SPEED_LOW  ] = "1.5",
         [ USB_SPEED_FULL ] = "12",
         [ USB_SPEED_HIGH ] = "480",
+        [ USB_SPEED_SUPER ] = "5000",
     };
     if (speed >= ARRAY_SIZE(txt))
         return "?";
diff --git a/hw/usb-ehci.c b/hw/usb-ehci.c
index f63519e..c909127 100644
--- a/hw/usb-ehci.c
+++ b/hw/usb-ehci.c
@@ -30,22 +30,16 @@
 #include "usb.h"
 #include "pci.h"
 #include "monitor.h"
+#include "trace.h"
 
 #define EHCI_DEBUG   0
-#define STATE_DEBUG  0       /* state transitions  */
 
-#if EHCI_DEBUG || STATE_DEBUG
+#if EHCI_DEBUG
 #define DPRINTF printf
 #else
 #define DPRINTF(...)
 #endif
 
-#if STATE_DEBUG
-#define DPRINTF_ST DPRINTF
-#else
-#define DPRINTF_ST(...)
-#endif
-
 /* internal processing - reset HC to try and recover */
 #define USB_RET_PROCERR   (-99)
 
@@ -204,6 +198,7 @@
 #define ITD_BUFPTR_MAXPKT_MASK   0x000007ff
 #define ITD_BUFPTR_MAXPKT_SH     0
 #define ITD_BUFPTR_MULT_MASK     0x00000003
+#define ITD_BUFPTR_MULT_SH       0
 } EHCIitd;
 
 /*  EHCI spec version 1.0 Section 3.4
@@ -340,8 +335,40 @@
     uint32_t backptr;                 // Standard next link pointer
 } EHCIfstn;
 
-typedef struct {
+typedef struct EHCIQueue EHCIQueue;
+typedef struct EHCIState EHCIState;
+
+enum async_state {
+    EHCI_ASYNC_NONE = 0,
+    EHCI_ASYNC_INFLIGHT,
+    EHCI_ASYNC_FINISHED,
+};
+
+struct EHCIQueue {
+    EHCIState *ehci;
+    QTAILQ_ENTRY(EHCIQueue) next;
+    bool async_schedule;
+    uint32_t seen, ts;
+
+    /* cached data from guest - needs to be flushed
+     * when guest removes an entry (doorbell, handshake sequence)
+     */
+    EHCIqh qh;             // copy of current QH (being worked on)
+    uint32_t qhaddr;       // address QH read from
+    EHCIqtd qtd;           // copy of current QTD (being worked on)
+    uint32_t qtdaddr;      // address QTD read from
+
+    USBPacket packet;
+    uint8_t buffer[BUFF_SIZE];
+    int pid;
+    uint32_t tbytes;
+    enum async_state async;
+    int usb_status;
+};
+
+struct EHCIState {
     PCIDevice dev;
+    USBBus bus;
     qemu_irq irq;
     target_phys_addr_t mem_base;
     int mem;
@@ -366,6 +393,7 @@
             uint32_t portsc[NB_PORTS];
         };
     };
+
     /*
      *  Internal states, shadow registers, etc
      */
@@ -375,32 +403,19 @@
     int astate;                        // Current state in asynchronous schedule
     int pstate;                        // Current state in periodic schedule
     USBPort ports[NB_PORTS];
-    uint8_t buffer[BUFF_SIZE];
     uint32_t usbsts_pending;
+    QTAILQ_HEAD(, EHCIQueue) queues;
 
-    /* cached data from guest - needs to be flushed
-     * when guest removes an entry (doorbell, handshake sequence)
-     */
-    EHCIqh qh;             // copy of current QH (being worked on)
-    uint32_t qhaddr;       // address QH read from
+    uint32_t a_fetch_addr;   // which address to look at next
+    uint32_t p_fetch_addr;   // which address to look at next
 
-    EHCIqtd qtd;           // copy of current QTD (being worked on)
-    uint32_t qtdaddr;      // address QTD read from
-
-    uint32_t itdaddr;      // current ITD
-
-    uint32_t fetch_addr;   // which address to look at next
-
-    USBBus bus;
-    USBPacket usb_packet;
-    int async_complete;
-    uint32_t tbytes;
-    int pid;
-    int exec_status;
+    USBPacket ipacket;
+    uint8_t ibuffer[BUFF_SIZE];
     int isoch_pause;
+
     uint32_t last_run_usec;
     uint32_t frame_end_usec;
-} EHCIState;
+};
 
 #define SET_LAST_RUN_CLOCK(s) \
     (s)->last_run_usec = qemu_get_clock_ns(vm_clock) / 1000;
@@ -416,35 +431,113 @@
     *data = val; \
     } while(0)
 
+static const char *ehci_state_names[] = {
+    [ EST_INACTIVE ]     = "INACTIVE",
+    [ EST_ACTIVE ]       = "ACTIVE",
+    [ EST_EXECUTING ]    = "EXECUTING",
+    [ EST_SLEEPING ]     = "SLEEPING",
+    [ EST_WAITLISTHEAD ] = "WAITLISTHEAD",
+    [ EST_FETCHENTRY ]   = "FETCH ENTRY",
+    [ EST_FETCHQH ]      = "FETCH QH",
+    [ EST_FETCHITD ]     = "FETCH ITD",
+    [ EST_ADVANCEQUEUE ] = "ADVANCEQUEUE",
+    [ EST_FETCHQTD ]     = "FETCH QTD",
+    [ EST_EXECUTE ]      = "EXECUTE",
+    [ EST_WRITEBACK ]    = "WRITEBACK",
+    [ EST_HORIZONTALQH ] = "HORIZONTALQH",
+};
 
-#if EHCI_DEBUG
-static const char *addr2str(unsigned addr)
+static const char *ehci_mmio_names[] = {
+    [ CAPLENGTH ]        = "CAPLENGTH",
+    [ HCIVERSION ]       = "HCIVERSION",
+    [ HCSPARAMS ]        = "HCSPARAMS",
+    [ HCCPARAMS ]        = "HCCPARAMS",
+    [ USBCMD ]           = "USBCMD",
+    [ USBSTS ]           = "USBSTS",
+    [ USBINTR ]          = "USBINTR",
+    [ FRINDEX ]          = "FRINDEX",
+    [ PERIODICLISTBASE ] = "P-LIST BASE",
+    [ ASYNCLISTADDR ]    = "A-LIST ADDR",
+    [ PORTSC_BEGIN ]     = "PORTSC #0",
+    [ PORTSC_BEGIN + 4]  = "PORTSC #1",
+    [ PORTSC_BEGIN + 8]  = "PORTSC #2",
+    [ PORTSC_BEGIN + 12] = "PORTSC #3",
+    [ CONFIGFLAG ]       = "CONFIGFLAG",
+};
+
+static const char *nr2str(const char **n, size_t len, uint32_t nr)
 {
-    const char *r            = "   unknown";
-    const char *n[] = {
-        [ CAPLENGTH ]        = " CAPLENGTH",
-        [ HCIVERSION ]       = "HCIVERSION",
-        [ HCSPARAMS ]        = " HCSPARAMS",
-        [ HCCPARAMS ]        = " HCCPARAMS",
-        [ USBCMD ]           = "   COMMAND",
-        [ USBSTS ]           = "    STATUS",
-        [ USBINTR ]          = " INTERRUPT",
-        [ FRINDEX ]          = " FRAME IDX",
-        [ PERIODICLISTBASE ] = "P-LIST BASE",
-        [ ASYNCLISTADDR ]    = "A-LIST ADDR",
-        [ PORTSC_BEGIN ...
-          PORTSC_END ]       = "PORT STATUS",
-        [ CONFIGFLAG ]       = "CONFIG FLAG",
-    };
-
-    if (addr < ARRAY_SIZE(n) && n[addr] != NULL) {
-        return n[addr];
+    if (nr < len && n[nr] != NULL) {
+        return n[nr];
     } else {
-        return r;
+        return "unknown";
     }
 }
-#endif
 
+static const char *state2str(uint32_t state)
+{
+    return nr2str(ehci_state_names, ARRAY_SIZE(ehci_state_names), state);
+}
+
+static const char *addr2str(target_phys_addr_t addr)
+{
+    return nr2str(ehci_mmio_names, ARRAY_SIZE(ehci_mmio_names), addr);
+}
+
+static void ehci_trace_usbsts(uint32_t mask, int state)
+{
+    /* interrupts */
+    if (mask & USBSTS_INT) {
+        trace_usb_ehci_usbsts("INT", state);
+    }
+    if (mask & USBSTS_ERRINT) {
+        trace_usb_ehci_usbsts("ERRINT", state);
+    }
+    if (mask & USBSTS_PCD) {
+        trace_usb_ehci_usbsts("PCD", state);
+    }
+    if (mask & USBSTS_FLR) {
+        trace_usb_ehci_usbsts("FLR", state);
+    }
+    if (mask & USBSTS_HSE) {
+        trace_usb_ehci_usbsts("HSE", state);
+    }
+    if (mask & USBSTS_IAA) {
+        trace_usb_ehci_usbsts("IAA", state);
+    }
+
+    /* status */
+    if (mask & USBSTS_HALT) {
+        trace_usb_ehci_usbsts("HALT", state);
+    }
+    if (mask & USBSTS_REC) {
+        trace_usb_ehci_usbsts("REC", state);
+    }
+    if (mask & USBSTS_PSS) {
+        trace_usb_ehci_usbsts("PSS", state);
+    }
+    if (mask & USBSTS_ASS) {
+        trace_usb_ehci_usbsts("ASS", state);
+    }
+}
+
+static inline void ehci_set_usbsts(EHCIState *s, int mask)
+{
+    if ((s->usbsts & mask) == mask) {
+        return;
+    }
+    ehci_trace_usbsts(mask, 1);
+    s->usbsts |= mask;
+}
+
+static inline void ehci_clear_usbsts(EHCIState *s, int mask)
+{
+    if ((s->usbsts & mask) == 0) {
+        return;
+    }
+    ehci_trace_usbsts(mask, 0);
+    s->usbsts &= ~mask;
+}
 
 static inline void ehci_set_interrupt(EHCIState *s, int intr)
 {
@@ -452,7 +545,7 @@
 
     // TODO honour interrupt threshold requests
 
-    s->usbsts |= intr;
+    ehci_set_usbsts(s, intr);
 
     if ((s->usbsts & USBINTR_MASK) & s->usbintr) {
         level = 1;
@@ -475,6 +568,155 @@
     s->usbsts_pending = 0;
 }
 
+static void ehci_set_state(EHCIState *s, int async, int state)
+{
+    if (async) {
+        trace_usb_ehci_state("async", state2str(state));
+        s->astate = state;
+    } else {
+        trace_usb_ehci_state("periodic", state2str(state));
+        s->pstate = state;
+    }
+}
+
+static int ehci_get_state(EHCIState *s, int async)
+{
+    return async ? s->astate : s->pstate;
+}
+
+static void ehci_set_fetch_addr(EHCIState *s, int async, uint32_t addr)
+{
+    if (async) {
+        s->a_fetch_addr = addr;
+    } else {
+        s->p_fetch_addr = addr;
+    }
+}
+
+static int ehci_get_fetch_addr(EHCIState *s, int async)
+{
+    return async ? s->a_fetch_addr : s->p_fetch_addr;
+}
+
+static void ehci_trace_qh(EHCIQueue *q, target_phys_addr_t addr, EHCIqh *qh)
+{
+    /* need three here due to argument count limits */
+    trace_usb_ehci_qh_ptrs(q, addr, qh->next,
+                           qh->current_qtd, qh->next_qtd, qh->altnext_qtd);
+    trace_usb_ehci_qh_fields(addr,
+                             get_field(qh->epchar, QH_EPCHAR_RL),
+                             get_field(qh->epchar, QH_EPCHAR_MPLEN),
+                             get_field(qh->epchar, QH_EPCHAR_EPS),
+                             get_field(qh->epchar, QH_EPCHAR_EP),
+                             get_field(qh->epchar, QH_EPCHAR_DEVADDR));
+    trace_usb_ehci_qh_bits(addr,
+                           (bool)(qh->epchar & QH_EPCHAR_C),
+                           (bool)(qh->epchar & QH_EPCHAR_H),
+                           (bool)(qh->epchar & QH_EPCHAR_DTC),
+                           (bool)(qh->epchar & QH_EPCHAR_I));
+}
+
+static void ehci_trace_qtd(EHCIQueue *q, target_phys_addr_t addr, EHCIqtd *qtd)
+{
+    /* need three here due to argument count limits */
+    trace_usb_ehci_qtd_ptrs(q, addr, qtd->next, qtd->altnext);
+    trace_usb_ehci_qtd_fields(addr,
+                              get_field(qtd->token, QTD_TOKEN_TBYTES),
+                              get_field(qtd->token, QTD_TOKEN_CPAGE),
+                              get_field(qtd->token, QTD_TOKEN_CERR),
+                              get_field(qtd->token, QTD_TOKEN_PID));
+    trace_usb_ehci_qtd_bits(addr,
+                            (bool)(qtd->token & QTD_TOKEN_IOC),
+                            (bool)(qtd->token & QTD_TOKEN_ACTIVE),
+                            (bool)(qtd->token & QTD_TOKEN_HALT),
+                            (bool)(qtd->token & QTD_TOKEN_BABBLE),
+                            (bool)(qtd->token & QTD_TOKEN_XACTERR));
+}
+
+static void ehci_trace_itd(EHCIState *s, target_phys_addr_t addr, EHCIitd *itd)
+{
+    trace_usb_ehci_itd(addr, itd->next,
+                       get_field(itd->bufptr[1], ITD_BUFPTR_MAXPKT),
+                       get_field(itd->bufptr[2], ITD_BUFPTR_MULT),
+                       get_field(itd->bufptr[0], ITD_BUFPTR_EP),
+                       get_field(itd->bufptr[0], ITD_BUFPTR_DEVADDR));
+}
+
+/* queue management */
+
+static EHCIQueue *ehci_alloc_queue(EHCIState *ehci, int async)
+{
+    EHCIQueue *q;
+
+    q = qemu_mallocz(sizeof(*q));
+    q->ehci = ehci;
+    q->async_schedule = async;
+    QTAILQ_INSERT_HEAD(&ehci->queues, q, next);
+    trace_usb_ehci_queue_action(q, "alloc");
+    return q;
+}
+
+static void ehci_free_queue(EHCIQueue *q)
+{
+    trace_usb_ehci_queue_action(q, "free");
+    if (q->async == EHCI_ASYNC_INFLIGHT) {
+        usb_cancel_packet(&q->packet);
+    }
+    QTAILQ_REMOVE(&q->ehci->queues, q, next);
+    qemu_free(q);
+}
+
+static EHCIQueue *ehci_find_queue_by_qh(EHCIState *ehci, uint32_t addr)
+{
+    EHCIQueue *q;
+
+    QTAILQ_FOREACH(q, &ehci->queues, next) {
+        if (addr == q->qhaddr) {
+            return q;
+        }
+    }
+    return NULL;
+}
+
+static void ehci_queues_rip_unused(EHCIState *ehci)
+{
+    EHCIQueue *q, *tmp;
+
+    QTAILQ_FOREACH_SAFE(q, &ehci->queues, next, tmp) {
+        if (q->seen) {
+            q->seen = 0;
+            q->ts = ehci->last_run_usec;
+            continue;
+        }
+        if (ehci->last_run_usec < q->ts + 250000) {
+            /* allow 0.25 sec idle */
+            continue;
+        }
+        ehci_free_queue(q);
+    }
+}
+
+static void ehci_queues_rip_device(EHCIState *ehci, USBDevice *dev)
+{
+    EHCIQueue *q, *tmp;
+
+    QTAILQ_FOREACH_SAFE(q, &ehci->queues, next, tmp) {
+        if (q->packet.owner != dev) {
+            continue;
+        }
+        ehci_free_queue(q);
+    }
+}
+
+static void ehci_queues_rip_all(EHCIState *ehci)
+{
+    EHCIQueue *q, *tmp;
+
+    QTAILQ_FOREACH_SAFE(q, &ehci->queues, next, tmp) {
+        ehci_free_queue(q);
+    }
+}
+
 /* Attach or detach a device on root hub */
 
 static void ehci_attach(USBPort *port)
@@ -482,8 +724,7 @@
     EHCIState *s = port->opaque;
     uint32_t *portsc = &s->portsc[port->index];
 
-    DPRINTF("ehci_attach invoked for index %d, portsc 0x%x, desc %s\n",
-           port->index, *portsc, port->dev->product_desc);
+    trace_usb_ehci_port_attach(port->index, port->dev->product_desc);
 
     *portsc |= PORTSC_CONNECT;
     *portsc |= PORTSC_CSC;
@@ -503,8 +744,7 @@
     EHCIState *s = port->opaque;
     uint32_t *portsc = &s->portsc[port->index];
 
-    DPRINTF("ehci_attach invoked for index %d, portsc 0x%x\n",
-           port->index, *portsc);
+    trace_usb_ehci_port_detach(port->index);
 
     *portsc &= ~PORTSC_CONNECT;
     *portsc |= PORTSC_CSC;
@@ -523,10 +763,9 @@
 static void ehci_reset(void *opaque)
 {
     EHCIState *s = opaque;
-    uint8_t *pci_conf;
     int i;
 
-    pci_conf = s->dev.config;
+    trace_usb_ehci_reset();
 
     memset(&s->mmio[OPREGBASE], 0x00, MMIO_SIZE - OPREGBASE);
 
@@ -535,7 +774,6 @@
 
     s->astate = EST_INACTIVE;
     s->pstate = EST_INACTIVE;
-    s->async_complete = 0;
     s->isoch_pause = -1;
     s->attach_poll_counter = 0;
 
@@ -546,6 +784,7 @@
             usb_attach(&s->ports[i], s->ports[i].dev);
         }
     }
+    ehci_queues_rip_all(s);
 }
 
 static uint32_t ehci_mem_readb(void *ptr, target_phys_addr_t addr)
@@ -576,6 +815,7 @@
     val = s->mmio[addr] | (s->mmio[addr+1] << 8) |
           (s->mmio[addr+2] << 16) | (s->mmio[addr+3] << 24);
 
+    trace_usb_ehci_mmio_readl(addr, addr2str(addr), val);
     return val;
 }
 
@@ -597,10 +837,6 @@
     int rwc;
     USBDevice *dev = s->ports[port].dev;
 
-    DPRINTF("port_status_write: "
-            "PORTSC (port %d) curr %08X new %08X rw-clear %08X rw %08X\n",
-            port, *portsc, val, (val & PORTSC_RWC_MASK), val & PORTSC_RO_MASK);
-
     rwc = val & PORTSC_RWC_MASK;
     val &= PORTSC_RO_MASK;
 
@@ -609,11 +845,11 @@
     *portsc &= ~rwc;
 
     if ((val & PORTSC_PRESET) && !(*portsc & PORTSC_PRESET)) {
-        DPRINTF("port_status_write: USBTRAN Port %d reset begin\n", port);
+        trace_usb_ehci_port_reset(port, 1);
     }
 
     if (!(val & PORTSC_PRESET) &&(*portsc & PORTSC_PRESET)) {
-        DPRINTF("port_status_write: USBTRAN Port %d reset done\n", port);
+        trace_usb_ehci_port_reset(port, 0);
         usb_attach(&s->ports[port], dev);
 
         // TODO how to handle reset of ports with no device
@@ -622,8 +858,6 @@
         }
 
         if (s->ports[port].dev) {
-            DPRINTF("port_status_write: "
-                    "Device was connected before reset, clearing CSC bit\n");
             *portsc &= ~PORTSC_CSC;
         }
 
@@ -638,16 +872,16 @@
 
     *portsc &= ~PORTSC_RO_MASK;
     *portsc |= val;
-    DPRINTF("port_status_write: Port %d status set to 0x%08x\n", port, *portsc);
 }
 
 static void ehci_mem_writel(void *ptr, target_phys_addr_t addr, uint32_t val)
 {
     EHCIState *s = ptr;
+    uint32_t *mmio = (uint32_t *)(&s->mmio[addr]);
+    uint32_t old = *mmio;
     int i;
-#if EHCI_DEBUG
-    const char *str;
-#endif
+
+    trace_usb_ehci_mmio_writel(addr, addr2str(addr), val);
 
     /* Only aligned reads are allowed on OHCI */
     if (addr & 3) {
@@ -658,6 +892,7 @@
 
     if (addr >= PORTSC && addr < PORTSC + 4 * NB_PORTS) {
         handle_port_status_write(s, (addr-PORTSC)/4, val);
+        trace_usb_ehci_mmio_change(addr, addr2str(addr), *mmio, old);
         return;
     }
 
@@ -669,30 +904,21 @@
 
 
     /* Do any register specific pre-write processing here.  */
-#if EHCI_DEBUG
-    str = addr2str((unsigned) addr);
-#endif
     switch(addr) {
     case USBCMD:
-        DPRINTF("ehci_mem_writel: USBCMD val=0x%08X, current cmd=0x%08X\n",
-                val, s->usbcmd);
-
         if ((val & USBCMD_RUNSTOP) && !(s->usbcmd & USBCMD_RUNSTOP)) {
-            DPRINTF("ehci_mem_writel: %s run, clear halt\n", str);
             qemu_mod_timer(s->frame_timer, qemu_get_clock_ns(vm_clock));
             SET_LAST_RUN_CLOCK(s);
-            s->usbsts &= ~USBSTS_HALT;
+            ehci_clear_usbsts(s, USBSTS_HALT);
         }
 
         if (!(val & USBCMD_RUNSTOP) && (s->usbcmd & USBCMD_RUNSTOP)) {
-            DPRINTF("                         ** STOP **\n");
             qemu_del_timer(s->frame_timer);
             // TODO - should finish out some stuff before setting halt
-            s->usbsts |= USBSTS_HALT;
+            ehci_set_usbsts(s, USBSTS_HALT);
         }
 
         if (val & USBCMD_HCRESET) {
-            DPRINTF("ehci_mem_writel: %s run, resetting\n", str);
             ehci_reset(s);
             val &= ~USBCMD_HCRESET;
         }
@@ -703,56 +929,24 @@
                     val & USBCMD_FLS);
             val &= ~USBCMD_FLS;
         }
-#if EHCI_DEBUG
-        if ((val & USBCMD_PSE) && !(s->usbcmd & USBCMD_PSE)) {
-            DPRINTF("periodic scheduling enabled\n");
-        }
-        if (!(val & USBCMD_PSE) && (s->usbcmd & USBCMD_PSE)) {
-            DPRINTF("periodic scheduling disabled\n");
-        }
-        if ((val & USBCMD_ASE) && !(s->usbcmd & USBCMD_ASE)) {
-            DPRINTF("asynchronous scheduling enabled\n");
-        }
-        if (!(val & USBCMD_ASE) && (s->usbcmd & USBCMD_ASE)) {
-            DPRINTF("asynchronous scheduling disabled\n");
-        }
-        if ((val & USBCMD_IAAD) && !(s->usbcmd & USBCMD_IAAD)) {
-            DPRINTF("doorbell request received\n");
-        }
-        if ((val & USBCMD_LHCR) && !(s->usbcmd & USBCMD_LHCR)) {
-            DPRINTF("light host controller reset received\n");
-        }
-        if ((val & USBCMD_ITC) != (s->usbcmd & USBCMD_ITC)) {
-            DPRINTF("interrupt threshold control set to %x\n",
-                    (val & USBCMD_ITC)>>USBCMD_ITC_SH);
-        }
-#endif
         break;
 
-
     case USBSTS:
         val &= USBSTS_RO_MASK;              // bits 6 thru 31 are RO
-        DPRINTF("ehci_mem_writel: %s RWC set to 0x%08X\n", str, val);
-
-        val = (s->usbsts &= ~val);         // bits 0 thru 5 are R/WC
-
-        DPRINTF("ehci_mem_writel: %s updating interrupt condition\n", str);
+        ehci_clear_usbsts(s, val);          // bits 0 thru 5 are R/WC
+        val = s->usbsts;
         ehci_set_interrupt(s, 0);
         break;
 
-
     case USBINTR:
         val &= USBINTR_MASK;
-        DPRINTF("ehci_mem_writel: %s set to 0x%08X\n", str, val);
         break;
 
     case FRINDEX:
         s->sofv = val >> 3;
-        DPRINTF("ehci_mem_writel: %s set to 0x%08X\n", str, val);
         break;
 
     case CONFIGFLAG:
-        DPRINTF("ehci_mem_writel: %s set to 0x%08X\n", str, val);
         val &= 0x1;
         if (val) {
             for(i = 0; i < NB_PORTS; i++)
@@ -766,7 +960,6 @@
               "ehci: PERIODIC list base register set while periodic schedule\n"
               "      is enabled and HC is enabled\n");
         }
-        DPRINTF("ehci_mem_writel: P-LIST BASE set to 0x%08X\n", val);
         break;
 
     case ASYNCLISTADDR:
@@ -775,11 +968,11 @@
               "ehci: ASYNC list address register set while async schedule\n"
               "      is enabled and HC is enabled\n");
         }
-        DPRINTF("ehci_mem_writel: A-LIST ADDR set to 0x%08X\n", val);
         break;
     }
 
-    *(uint32_t *)(&s->mmio[addr]) = val;
+    *mmio = val;
+    trace_usb_ehci_mmio_change(addr, addr2str(addr), *mmio, old);
 }
 
 
@@ -813,7 +1006,7 @@
 
 // 4.10.2
 
-static int ehci_qh_do_overlay(EHCIState *ehci, EHCIqh *qh, EHCIqtd *qtd)
+static int ehci_qh_do_overlay(EHCIQueue *q)
 {
     int i;
     int dtoggle;
@@ -823,45 +1016,43 @@
 
     // remember values in fields to preserve in qh after overlay
 
-    dtoggle = qh->token & QTD_TOKEN_DTOGGLE;
-    ping    = qh->token & QTD_TOKEN_PING;
+    dtoggle = q->qh.token & QTD_TOKEN_DTOGGLE;
+    ping    = q->qh.token & QTD_TOKEN_PING;
 
-    DPRINTF("setting qh.current from %08X to 0x%08X\n", qh->current_qtd,
-            ehci->qtdaddr);
-    qh->current_qtd = ehci->qtdaddr;
-    qh->next_qtd    = qtd->next;
-    qh->altnext_qtd = qtd->altnext;
-    qh->token       = qtd->token;
+    q->qh.current_qtd = q->qtdaddr;
+    q->qh.next_qtd    = q->qtd.next;
+    q->qh.altnext_qtd = q->qtd.altnext;
+    q->qh.token       = q->qtd.token;
 
 
-    eps = get_field(qh->epchar, QH_EPCHAR_EPS);
+    eps = get_field(q->qh.epchar, QH_EPCHAR_EPS);
     if (eps == EHCI_QH_EPS_HIGH) {
-        qh->token &= ~QTD_TOKEN_PING;
-        qh->token |= ping;
+        q->qh.token &= ~QTD_TOKEN_PING;
+        q->qh.token |= ping;
     }
 
-    reload = get_field(qh->epchar, QH_EPCHAR_RL);
-    set_field(&qh->altnext_qtd, reload, QH_ALTNEXT_NAKCNT);
+    reload = get_field(q->qh.epchar, QH_EPCHAR_RL);
+    set_field(&q->qh.altnext_qtd, reload, QH_ALTNEXT_NAKCNT);
 
     for (i = 0; i < 5; i++) {
-        qh->bufptr[i] = qtd->bufptr[i];
+        q->qh.bufptr[i] = q->qtd.bufptr[i];
     }
 
-    if (!(qh->epchar & QH_EPCHAR_DTC)) {
+    if (!(q->qh.epchar & QH_EPCHAR_DTC)) {
         // preserve QH DT bit
-        qh->token &= ~QTD_TOKEN_DTOGGLE;
-        qh->token |= dtoggle;
+        q->qh.token &= ~QTD_TOKEN_DTOGGLE;
+        q->qh.token |= dtoggle;
     }
 
-    qh->bufptr[1] &= ~BUFPTR_CPROGMASK_MASK;
-    qh->bufptr[2] &= ~BUFPTR_FRAMETAG_MASK;
+    q->qh.bufptr[1] &= ~BUFPTR_CPROGMASK_MASK;
+    q->qh.bufptr[2] &= ~BUFPTR_FRAMETAG_MASK;
 
-    put_dwords(NLPTR_GET(ehci->qhaddr), (uint32_t *) qh, sizeof(EHCIqh) >> 2);
+    put_dwords(NLPTR_GET(q->qhaddr), (uint32_t *) &q->qh, sizeof(EHCIqh) >> 2);
 
     return 0;
 }
 
-static int ehci_buffer_rw(uint8_t *buffer, EHCIqh *qh, int bytes, int rw)
+static int ehci_buffer_rw(EHCIQueue *q, int bytes, int rw)
 {
     int bufpos = 0;
     int cpage, offset;
@@ -873,19 +1064,17 @@
         return 0;
     }
 
-    cpage = get_field(qh->token, QTD_TOKEN_CPAGE);
+    cpage = get_field(q->qh.token, QTD_TOKEN_CPAGE);
     if (cpage > 4) {
         fprintf(stderr, "cpage out of range (%d)\n", cpage);
         return USB_RET_PROCERR;
     }
 
-    offset = qh->bufptr[0] & ~QTD_BUFPTR_MASK;
-    DPRINTF("ehci_buffer_rw: %sing %d bytes %08x cpage %d offset %d\n",
-           rw ? "writ" : "read", bytes, qh->bufptr[0], cpage, offset);
+    offset = q->qh.bufptr[0] & ~QTD_BUFPTR_MASK;
 
     do {
         /* start and end of this page */
-        head = qh->bufptr[cpage] & QTD_BUFPTR_MASK;
+        head = q->qh.bufptr[cpage] & QTD_BUFPTR_MASK;
         tail = head + ~QTD_BUFPTR_MASK + 1;
         /* add offset into page */
         head |= offset;
@@ -894,12 +1083,11 @@
             tail = head + bytes;
         }
 
-        DPRINTF("DATA %s cpage:%d head:%08X tail:%08X target:%08X\n",
-                rw ? "WRITE" : "READ ", cpage, head, tail, bufpos);
-
-        cpu_physical_memory_rw(head, &buffer[bufpos], tail - head, rw);
+        trace_usb_ehci_data(rw, cpage, offset, head, tail-head, bufpos);
+        cpu_physical_memory_rw(head, q->buffer + bufpos, tail - head, rw);
 
         bufpos += (tail - head);
+        offset += (tail - head);
         bytes -= (tail - head);
 
         if (bytes > 0) {
@@ -909,112 +1097,106 @@
     } while (bytes > 0);
 
     /* save cpage */
-    set_field(&qh->token, cpage, QTD_TOKEN_CPAGE);
+    set_field(&q->qh.token, cpage, QTD_TOKEN_CPAGE);
 
     /* save offset into cpage */
-    offset = tail - head;
-    qh->bufptr[0] &= ~QTD_BUFPTR_MASK;
-    qh->bufptr[0] |= offset;
+    q->qh.bufptr[0] &= QTD_BUFPTR_MASK;
+    q->qh.bufptr[0] |= offset;
 
     return 0;
 }
 
 static void ehci_async_complete_packet(USBDevice *dev, USBPacket *packet)
 {
-    EHCIState *ehci = container_of(packet, EHCIState, usb_packet);
+    EHCIQueue *q = container_of(packet, EHCIQueue, packet);
 
-    DPRINTF("Async packet complete\n");
-    ehci->async_complete = 1;
-    ehci->exec_status = packet->len;
+    trace_usb_ehci_queue_action(q, "wakeup");
+    assert(q->async == EHCI_ASYNC_INFLIGHT);
+    q->async = EHCI_ASYNC_FINISHED;
+    q->usb_status = packet->len;
 }
 
-static int ehci_execute_complete(EHCIState *ehci, EHCIqh *qh, int ret)
+static void ehci_execute_complete(EHCIQueue *q)
 {
     int c_err, reload;
 
-    if (ret == USB_RET_ASYNC && !ehci->async_complete) {
-        DPRINTF("not done yet\n");
-        return ret;
-    }
-
-    ehci->async_complete = 0;
+    assert(q->async != EHCI_ASYNC_INFLIGHT);
+    q->async = EHCI_ASYNC_NONE;
 
     DPRINTF("execute_complete: qhaddr 0x%x, next %x, qtdaddr 0x%x, status %d\n",
-            ehci->qhaddr, qh->next, ehci->qtdaddr, ret);
+            q->qhaddr, q->qh.next, q->qtdaddr, q->usb_status);
 
-    if (ret < 0) {
+    if (q->usb_status < 0) {
 err:
         /* TO-DO: put this is in a function that can be invoked below as well */
-        c_err = get_field(qh->token, QTD_TOKEN_CERR);
+        c_err = get_field(q->qh.token, QTD_TOKEN_CERR);
         c_err--;
-        set_field(&qh->token, c_err, QTD_TOKEN_CERR);
+        set_field(&q->qh.token, c_err, QTD_TOKEN_CERR);
 
-        switch(ret) {
+        switch(q->usb_status) {
         case USB_RET_NODEV:
-            fprintf(stderr, "USB no device\n");
+            q->qh.token |= (QTD_TOKEN_HALT | QTD_TOKEN_XACTERR);
+            ehci_record_interrupt(q->ehci, USBSTS_ERRINT);
             break;
         case USB_RET_STALL:
-            fprintf(stderr, "USB stall\n");
-            qh->token |= QTD_TOKEN_HALT;
-            ehci_record_interrupt(ehci, USBSTS_ERRINT);
+            q->qh.token |= QTD_TOKEN_HALT;
+            ehci_record_interrupt(q->ehci, USBSTS_ERRINT);
             break;
         case USB_RET_NAK:
             /* 4.10.3 */
-            reload = get_field(qh->epchar, QH_EPCHAR_RL);
-            if ((ehci->pid == USB_TOKEN_IN) && reload) {
-                int nakcnt = get_field(qh->altnext_qtd, QH_ALTNEXT_NAKCNT);
+            reload = get_field(q->qh.epchar, QH_EPCHAR_RL);
+            if ((q->pid == USB_TOKEN_IN) && reload) {
+                int nakcnt = get_field(q->qh.altnext_qtd, QH_ALTNEXT_NAKCNT);
                 nakcnt--;
-                set_field(&qh->altnext_qtd, nakcnt, QH_ALTNEXT_NAKCNT);
+                set_field(&q->qh.altnext_qtd, nakcnt, QH_ALTNEXT_NAKCNT);
             } else if (!reload) {
-                return USB_RET_NAK;
+                return;
             }
             break;
         case USB_RET_BABBLE:
-            fprintf(stderr, "USB babble TODO\n");
-            qh->token |= QTD_TOKEN_BABBLE;
-            ehci_record_interrupt(ehci, USBSTS_ERRINT);
+            q->qh.token |= (QTD_TOKEN_HALT | QTD_TOKEN_BABBLE);
+            ehci_record_interrupt(q->ehci, USBSTS_ERRINT);
             break;
         default:
-            fprintf(stderr, "USB invalid response %d to handle\n", ret);
-            /* TO-DO: transaction error */
-            ret = USB_RET_PROCERR;
+            /* should not be triggerable */
+            fprintf(stderr, "USB invalid response %d to handle\n", q->usb_status);
+            assert(0);
             break;
         }
     } else {
         // DPRINTF("Short packet condition\n");
         // TODO check 4.12 for splits
 
-        if ((ret > ehci->tbytes) && (ehci->pid == USB_TOKEN_IN)) {
-            ret = USB_RET_BABBLE;
+        if ((q->usb_status > q->tbytes) && (q->pid == USB_TOKEN_IN)) {
+            q->usb_status = USB_RET_BABBLE;
             goto err;
         }
 
-        if (ehci->tbytes && ehci->pid == USB_TOKEN_IN) {
-            if (ehci_buffer_rw(ehci->buffer, qh, ret, 1) != 0) {
-                return USB_RET_PROCERR;
+        if (q->tbytes && q->pid == USB_TOKEN_IN) {
+            if (ehci_buffer_rw(q, q->usb_status, 1) != 0) {
+                q->usb_status = USB_RET_PROCERR;
+                return;
             }
-            ehci->tbytes -= ret;
+            q->tbytes -= q->usb_status;
         } else {
-            ehci->tbytes = 0;
+            q->tbytes = 0;
         }
 
-        DPRINTF("updating tbytes to %d\n", ehci->tbytes);
-        set_field(&qh->token, ehci->tbytes, QTD_TOKEN_TBYTES);
+        DPRINTF("updating tbytes to %d\n", q->tbytes);
+        set_field(&q->qh.token, q->tbytes, QTD_TOKEN_TBYTES);
     }
 
-    qh->token ^= QTD_TOKEN_DTOGGLE;
-    qh->token &= ~QTD_TOKEN_ACTIVE;
+    q->qh.token ^= QTD_TOKEN_DTOGGLE;
+    q->qh.token &= ~QTD_TOKEN_ACTIVE;
 
-    if ((ret >= 0) && (qh->token & QTD_TOKEN_IOC)) {
-        ehci_record_interrupt(ehci, USBSTS_INT);
+    if ((q->usb_status >= 0) && (q->qh.token & QTD_TOKEN_IOC)) {
+        ehci_record_interrupt(q->ehci, USBSTS_INT);
     }
-
-    return ret;
 }
 
 // 4.10.3
 
-static int ehci_execute(EHCIState *ehci, EHCIqh *qh)
+static int ehci_execute(EHCIQueue *q)
 {
     USBPort *port;
     USBDevice *dev;
@@ -1023,59 +1205,59 @@
     int endp;
     int devadr;
 
-    if ( !(qh->token & QTD_TOKEN_ACTIVE)) {
+    if ( !(q->qh.token & QTD_TOKEN_ACTIVE)) {
         fprintf(stderr, "Attempting to execute inactive QH\n");
         return USB_RET_PROCERR;
     }
 
-    ehci->tbytes = (qh->token & QTD_TOKEN_TBYTES_MASK) >> QTD_TOKEN_TBYTES_SH;
-    if (ehci->tbytes > BUFF_SIZE) {
+    q->tbytes = (q->qh.token & QTD_TOKEN_TBYTES_MASK) >> QTD_TOKEN_TBYTES_SH;
+    if (q->tbytes > BUFF_SIZE) {
         fprintf(stderr, "Request for more bytes than allowed\n");
         return USB_RET_PROCERR;
     }
 
-    ehci->pid = (qh->token & QTD_TOKEN_PID_MASK) >> QTD_TOKEN_PID_SH;
-    switch(ehci->pid) {
-        case 0: ehci->pid = USB_TOKEN_OUT; break;
-        case 1: ehci->pid = USB_TOKEN_IN; break;
-        case 2: ehci->pid = USB_TOKEN_SETUP; break;
+    q->pid = (q->qh.token & QTD_TOKEN_PID_MASK) >> QTD_TOKEN_PID_SH;
+    switch(q->pid) {
+        case 0: q->pid = USB_TOKEN_OUT; break;
+        case 1: q->pid = USB_TOKEN_IN; break;
+        case 2: q->pid = USB_TOKEN_SETUP; break;
         default: fprintf(stderr, "bad token\n"); break;
     }
 
-    if ((ehci->tbytes && ehci->pid != USB_TOKEN_IN) &&
-        (ehci_buffer_rw(ehci->buffer, qh, ehci->tbytes, 0) != 0)) {
+    if ((q->tbytes && q->pid != USB_TOKEN_IN) &&
+        (ehci_buffer_rw(q, q->tbytes, 0) != 0)) {
         return USB_RET_PROCERR;
     }
 
-    endp = get_field(qh->epchar, QH_EPCHAR_EP);
-    devadr = get_field(qh->epchar, QH_EPCHAR_DEVADDR);
+    endp = get_field(q->qh.epchar, QH_EPCHAR_EP);
+    devadr = get_field(q->qh.epchar, QH_EPCHAR_DEVADDR);
 
     ret = USB_RET_NODEV;
 
     // TO-DO: associating device with ehci port
     for(i = 0; i < NB_PORTS; i++) {
-        port = &ehci->ports[i];
+        port = &q->ehci->ports[i];
         dev = port->dev;
 
         // TODO sometime we will also need to check if we are the port owner
 
-        if (!(ehci->portsc[i] &(PORTSC_CONNECT))) {
+        if (!(q->ehci->portsc[i] &(PORTSC_CONNECT))) {
             DPRINTF("Port %d, no exec, not connected(%08X)\n",
-                    i, ehci->portsc[i]);
+                    i, q->ehci->portsc[i]);
             continue;
         }
 
-        ehci->usb_packet.pid = ehci->pid;
-        ehci->usb_packet.devaddr = devadr;
-        ehci->usb_packet.devep = endp;
-        ehci->usb_packet.data = ehci->buffer;
-        ehci->usb_packet.len = ehci->tbytes;
+        q->packet.pid = q->pid;
+        q->packet.devaddr = devadr;
+        q->packet.devep = endp;
+        q->packet.data = q->buffer;
+        q->packet.len = q->tbytes;
 
-        ret = usb_handle_packet(dev, &ehci->usb_packet);
+        ret = usb_handle_packet(dev, &q->packet);
 
         DPRINTF("submit: qh %x next %x qtd %x pid %x len %d (total %d) endp %x ret %d\n",
-                ehci->qhaddr, qh->next, ehci->qtdaddr, ehci->pid,
-                ehci->usb_packet.len, ehci->tbytes, endp, ret);
+                q->qhaddr, q->qh.next, q->qtdaddr, q->pid,
+                q->packet.len, q->tbytes, endp, ret);
 
         if (ret != USB_RET_NODEV) {
             break;
@@ -1087,10 +1269,6 @@
         return USB_RET_PROCERR;
     }
 
-    if (ret == USB_RET_ASYNC) {
-        ehci->async_complete = 0;
-    }
-
     return ret;
 }
 
@@ -1103,42 +1281,51 @@
     USBPort *port;
     USBDevice *dev;
     int ret;
-    int i, j;
-    int ptr;
-    int pid;
-    int pg;
-    int len;
-    int dir;
-    int devadr;
-    int endp;
-    int maxpkt;
+    uint32_t i, j, len, len1, len2, pid, dir, devaddr, endp;
+    uint32_t pg, off, ptr1, ptr2, max, mult;
 
     dir =(itd->bufptr[1] & ITD_BUFPTR_DIRECTION);
-    devadr = get_field(itd->bufptr[0], ITD_BUFPTR_DEVADDR);
+    devaddr = get_field(itd->bufptr[0], ITD_BUFPTR_DEVADDR);
     endp = get_field(itd->bufptr[0], ITD_BUFPTR_EP);
-    maxpkt = get_field(itd->bufptr[1], ITD_BUFPTR_MAXPKT);
+    max = get_field(itd->bufptr[1], ITD_BUFPTR_MAXPKT);
+    mult = get_field(itd->bufptr[2], ITD_BUFPTR_MULT);
 
     for(i = 0; i < 8; i++) {
         if (itd->transact[i] & ITD_XACT_ACTIVE) {
-            DPRINTF("ISOCHRONOUS active for frame %d, interval %d\n",
-                    ehci->frindex >> 3, i);
+            pg   = get_field(itd->transact[i], ITD_XACT_PGSEL);
+            off  = itd->transact[i] & ITD_XACT_OFFSET_MASK;
+            ptr1 = (itd->bufptr[pg] & ITD_BUFPTR_MASK);
+            ptr2 = (itd->bufptr[pg+1] & ITD_BUFPTR_MASK);
+            len  = get_field(itd->transact[i], ITD_XACT_LENGTH);
 
-            pg = get_field(itd->transact[i], ITD_XACT_PGSEL);
-            ptr = (itd->bufptr[pg] & ITD_BUFPTR_MASK) |
-                (itd->transact[i] & ITD_XACT_OFFSET_MASK);
-            len = get_field(itd->transact[i], ITD_XACT_LENGTH);
+            if (len > max * mult) {
+                len = max * mult;
+            }
 
             if (len > BUFF_SIZE) {
                 return USB_RET_PROCERR;
             }
 
-            DPRINTF("ISOCH: buffer %08X len %d\n", ptr, len);
+            if (off + len > 4096) {
+                /* transfer crosses page border */
+                len2 = off + len - 4096;
+                len1 = len - len2;
+            } else {
+                len1 = len;
+                len2 = 0;
+            }
 
             if (!dir) {
-                cpu_physical_memory_rw(ptr, &ehci->buffer[0], len, 0);
                 pid = USB_TOKEN_OUT;
-            } else
+                trace_usb_ehci_data(0, pg, off, ptr1 + off, len1, 0);
+                cpu_physical_memory_rw(ptr1 + off, &ehci->ibuffer[0], len1, 0);
+                if (len2) {
+                    trace_usb_ehci_data(0, pg+1, 0, ptr2, len2, len1);
+                    cpu_physical_memory_rw(ptr2, &ehci->ibuffer[len1], len2, 0);
+                }
+            } else {
                 pid = USB_TOKEN_IN;
+            }
 
             ret = USB_RET_NODEV;
 
@@ -1149,25 +1336,23 @@
                 // TODO sometime we will also need to check if we are the port owner
 
                 if (!(ehci->portsc[j] &(PORTSC_CONNECT))) {
-                    DPRINTF("Port %d, no exec, not connected(%08X)\n",
-                            j, ehci->portsc[j]);
                     continue;
                 }
 
-                ehci->usb_packet.pid = ehci->pid;
-                ehci->usb_packet.devaddr = devadr;
-                ehci->usb_packet.devep = endp;
-                ehci->usb_packet.data = ehci->buffer;
-                ehci->usb_packet.len = len;
+                ehci->ipacket.pid = pid;
+                ehci->ipacket.devaddr = devaddr;
+                ehci->ipacket.devep = endp;
+                ehci->ipacket.data = ehci->ibuffer;
+                ehci->ipacket.len = len;
 
-                DPRINTF("calling usb_handle_packet\n");
-                ret = usb_handle_packet(dev, &ehci->usb_packet);
+                ret = usb_handle_packet(dev, &ehci->ipacket);
 
                 if (ret != USB_RET_NODEV) {
                     break;
                 }
             }
 
+#if 0
             /*  In isoch, there is no facility to indicate a NAK so let's
              *  instead just complete a zero-byte transaction.  Setting
              *  DBERR seems too draconian.
@@ -1192,24 +1377,40 @@
                 DPRINTF("ISOCH: received ACK, clearing pause\n");
                 ehci->isoch_pause = -1;
             }
+#else
+            if (ret == USB_RET_NAK) {
+                ret = 0;
+            }
+#endif
 
             if (ret >= 0) {
-                itd->transact[i] &= ~ITD_XACT_ACTIVE;
+                if (!dir) {
+                    /* OUT */
+                    set_field(&itd->transact[i], len - ret, ITD_XACT_LENGTH);
+                } else {
+                    /* IN */
+                    if (len1 > ret) {
+                        len1 = ret;
+                    }
+                    if (len2 > ret - len1) {
+                        len2 = ret - len1;
+                    }
+                    if (len1) {
+                        trace_usb_ehci_data(1, pg, off, ptr1 + off, len1, 0);
+                        cpu_physical_memory_rw(ptr1 + off, &ehci->ibuffer[0], len1, 1);
+                    }
+                    if (len2) {
+                        trace_usb_ehci_data(1, pg+1, 0, ptr2, len2, len1);
+                        cpu_physical_memory_rw(ptr2, &ehci->ibuffer[len1], len2, 1);
+                    }
+                    set_field(&itd->transact[i], ret, ITD_XACT_LENGTH);
+                }
 
                 if (itd->transact[i] & ITD_XACT_IOC) {
                     ehci_record_interrupt(ehci, USBSTS_INT);
                 }
             }
-
-            if (ret >= 0 && dir) {
-                cpu_physical_memory_rw(ptr, &ehci->buffer[0], len, 1);
-
-                if (ret != len) {
-                    DPRINTF("ISOCH IN expected %d, got %d\n",
-                            len, ret);
-                    set_field(&itd->transact[i], ret, ITD_XACT_LENGTH);
-                }
-            }
+            itd->transact[i] &= ~ITD_XACT_ACTIVE;
         }
     }
     return 0;
@@ -1218,47 +1419,45 @@
 /*  This state is the entry point for asynchronous schedule
  *  processing.  Entry here consitutes a EHCI start event state (4.8.5)
  */
-static int ehci_state_waitlisthead(EHCIState *ehci,  int async, int *state)
+static int ehci_state_waitlisthead(EHCIState *ehci,  int async)
 {
-    EHCIqh *qh = &ehci->qh;
+    EHCIqh qh;
     int i = 0;
     int again = 0;
     uint32_t entry = ehci->asynclistaddr;
 
     /* set reclamation flag at start event (4.8.6) */
     if (async) {
-        ehci->usbsts |= USBSTS_REC;
+        ehci_set_usbsts(ehci, USBSTS_REC);
     }
 
+    ehci_queues_rip_unused(ehci);
+
     /*  Find the head of the list (4.9.1.1) */
     for(i = 0; i < MAX_QH; i++) {
-        get_dwords(NLPTR_GET(entry), (uint32_t *) qh, sizeof(EHCIqh) >> 2);
+        get_dwords(NLPTR_GET(entry), (uint32_t *) &qh, sizeof(EHCIqh) >> 2);
+        ehci_trace_qh(NULL, NLPTR_GET(entry), &qh);
 
-        if (qh->epchar & QH_EPCHAR_H) {
-            DPRINTF_ST("WAITLISTHEAD: QH %08X is the HEAD of the list\n",
-                       entry);
+        if (qh.epchar & QH_EPCHAR_H) {
             if (async) {
                 entry |= (NLPTR_TYPE_QH << 1);
             }
 
-            ehci->fetch_addr = entry;
-            *state = EST_FETCHENTRY;
+            ehci_set_fetch_addr(ehci, async, entry);
+            ehci_set_state(ehci, async, EST_FETCHENTRY);
             again = 1;
             goto out;
         }
 
-        DPRINTF_ST("WAITLISTHEAD: QH %08X is NOT the HEAD of the list\n",
-                   entry);
-        entry = qh->next;
+        entry = qh.next;
         if (entry == ehci->asynclistaddr) {
-            DPRINTF("WAITLISTHEAD: reached beginning of QH list\n");
             break;
         }
     }
 
     /* no head found for list. */
 
-    *state = EST_ACTIVE;
+    ehci_set_state(ehci, async, EST_ACTIVE);
 
 out:
     return again;
@@ -1268,25 +1467,14 @@
 /*  This state is the entry point for periodic schedule processing as
  *  well as being a continuation state for async processing.
  */
-static int ehci_state_fetchentry(EHCIState *ehci, int async, int *state)
+static int ehci_state_fetchentry(EHCIState *ehci, int async)
 {
     int again = 0;
-    uint32_t entry = ehci->fetch_addr;
+    uint32_t entry = ehci_get_fetch_addr(ehci, async);
 
-#if EHCI_DEBUG == 0
-    if (qemu_get_clock_ns(vm_clock) / 1000 >= ehci->frame_end_usec) {
-        if (async) {
-            DPRINTF("FETCHENTRY: FRAME timer elapsed, exit state machine\n");
-            goto out;
-        } else {
-            DPRINTF("FETCHENTRY: WARNING "
-                    "- frame timer elapsed during periodic\n");
-        }
-    }
-#endif
     if (entry < 0x1000) {
         DPRINTF("fetchentry: entry invalid (0x%08x)\n", entry);
-        *state = EST_ACTIVE;
+        ehci_set_state(ehci, async, EST_ACTIVE);
         goto out;
     }
 
@@ -1298,16 +1486,12 @@
 
     switch (NLPTR_TYPE_GET(entry)) {
     case NLPTR_TYPE_QH:
-        DPRINTF_ST("FETCHENTRY: entry %X is a Queue Head\n", entry);
-        *state = EST_FETCHQH;
-        ehci->qhaddr = entry;
+        ehci_set_state(ehci, async, EST_FETCHQH);
         again = 1;
         break;
 
     case NLPTR_TYPE_ITD:
-        DPRINTF_ST("FETCHENTRY: entry %X is an ITD\n", entry);
-        *state = EST_FETCHITD;
-        ehci->itdaddr = entry;
+        ehci_set_state(ehci, async, EST_FETCHITD);
         again = 1;
         break;
 
@@ -1322,89 +1506,114 @@
     return again;
 }
 
-static int ehci_state_fetchqh(EHCIState *ehci, int async, int *state)
+static EHCIQueue *ehci_state_fetchqh(EHCIState *ehci, int async)
 {
-    EHCIqh *qh = &ehci->qh;
+    uint32_t entry;
+    EHCIQueue *q;
     int reload;
-    int again = 0;
 
-    get_dwords(NLPTR_GET(ehci->qhaddr), (uint32_t *) qh, sizeof(EHCIqh) >> 2);
+    entry = ehci_get_fetch_addr(ehci, async);
+    q = ehci_find_queue_by_qh(ehci, entry);
+    if (NULL == q) {
+        q = ehci_alloc_queue(ehci, async);
+    }
+    q->qhaddr = entry;
+    q->seen++;
 
-    if (async && (qh->epchar & QH_EPCHAR_H)) {
+    if (q->seen > 1) {
+        /* we are going in circles -- stop processing */
+        ehci_set_state(ehci, async, EST_ACTIVE);
+        q = NULL;
+        goto out;
+    }
+
+    get_dwords(NLPTR_GET(q->qhaddr), (uint32_t *) &q->qh, sizeof(EHCIqh) >> 2);
+    ehci_trace_qh(q, NLPTR_GET(q->qhaddr), &q->qh);
+
+    if (q->async == EHCI_ASYNC_INFLIGHT) {
+        /* I/O still in progress -- skip queue */
+        ehci_set_state(ehci, async, EST_HORIZONTALQH);
+        goto out;
+    }
+    if (q->async == EHCI_ASYNC_FINISHED) {
+        /* I/O finished -- continue processing queue */
+        trace_usb_ehci_queue_action(q, "resume");
+        ehci_set_state(ehci, async, EST_EXECUTING);
+        goto out;
+    }
+
+    if (async && (q->qh.epchar & QH_EPCHAR_H)) {
 
         /*  EHCI spec version 1.0 Section 4.8.3 & 4.10.1 */
         if (ehci->usbsts & USBSTS_REC) {
-            ehci->usbsts &= ~USBSTS_REC;
+            ehci_clear_usbsts(ehci, USBSTS_REC);
         } else {
             DPRINTF("FETCHQH:  QH 0x%08x. H-bit set, reclamation status reset"
-                       " - done processing\n", ehci->qhaddr);
-            *state = EST_ACTIVE;
+                       " - done processing\n", q->qhaddr);
+            ehci_set_state(ehci, async, EST_ACTIVE);
+            q = NULL;
             goto out;
         }
     }
 
 #if EHCI_DEBUG
-    if (ehci->qhaddr != qh->next) {
+    if (q->qhaddr != q->qh.next) {
     DPRINTF("FETCHQH:  QH 0x%08x (h %x halt %x active %x) next 0x%08x\n",
-               ehci->qhaddr,
-               qh->epchar & QH_EPCHAR_H,
-               qh->token & QTD_TOKEN_HALT,
-               qh->token & QTD_TOKEN_ACTIVE,
-               qh->next);
+               q->qhaddr,
+               q->qh.epchar & QH_EPCHAR_H,
+               q->qh.token & QTD_TOKEN_HALT,
+               q->qh.token & QTD_TOKEN_ACTIVE,
+               q->qh.next);
     }
 #endif
 
-    reload = get_field(qh->epchar, QH_EPCHAR_RL);
+    reload = get_field(q->qh.epchar, QH_EPCHAR_RL);
     if (reload) {
-        DPRINTF_ST("FETCHQH: reloading nakcnt to %d\n", reload);
-        set_field(&qh->altnext_qtd, reload, QH_ALTNEXT_NAKCNT);
+        set_field(&q->qh.altnext_qtd, reload, QH_ALTNEXT_NAKCNT);
     }
 
-    if (qh->token & QTD_TOKEN_HALT) {
-        DPRINTF_ST("FETCHQH: QH Halted, go horizontal\n");
-        *state = EST_HORIZONTALQH;
-        again = 1;
+    if (q->qh.token & QTD_TOKEN_HALT) {
+        ehci_set_state(ehci, async, EST_HORIZONTALQH);
 
-    } else if ((qh->token & QTD_TOKEN_ACTIVE) && (qh->current_qtd > 0x1000)) {
-        DPRINTF_ST("FETCHQH: Active, !Halt, execute - fetch qTD\n");
-        ehci->qtdaddr = qh->current_qtd;
-        *state = EST_FETCHQTD;
-        again = 1;
+    } else if ((q->qh.token & QTD_TOKEN_ACTIVE) && (q->qh.current_qtd > 0x1000)) {
+        q->qtdaddr = q->qh.current_qtd;
+        ehci_set_state(ehci, async, EST_FETCHQTD);
 
     } else {
         /*  EHCI spec version 1.0 Section 4.10.2 */
-        DPRINTF_ST("FETCHQH: !Active, !Halt, advance queue\n");
-        *state = EST_ADVANCEQUEUE;
-        again = 1;
+        ehci_set_state(ehci, async, EST_ADVANCEQUEUE);
     }
 
 out:
-    return again;
+    return q;
 }
 
-static int ehci_state_fetchitd(EHCIState *ehci, int async, int *state)
+static int ehci_state_fetchitd(EHCIState *ehci, int async)
 {
+    uint32_t entry;
     EHCIitd itd;
 
-    get_dwords(NLPTR_GET(ehci->itdaddr),(uint32_t *) &itd,
+    assert(!async);
+    entry = ehci_get_fetch_addr(ehci, async);
+
+    get_dwords(NLPTR_GET(entry),(uint32_t *) &itd,
                sizeof(EHCIitd) >> 2);
-    DPRINTF_ST("FETCHITD: Fetched ITD at address %08X " "(next is %08X)\n",
-               ehci->itdaddr, itd.next);
+    ehci_trace_itd(ehci, entry, &itd);
 
     if (ehci_process_itd(ehci, &itd) != 0) {
         return -1;
     }
 
-    put_dwords(NLPTR_GET(ehci->itdaddr), (uint32_t *) &itd,
+    put_dwords(NLPTR_GET(entry), (uint32_t *) &itd,
                 sizeof(EHCIitd) >> 2);
-    ehci->fetch_addr = itd.next;
-    *state = EST_FETCHENTRY;
+    ehci_set_fetch_addr(ehci, async, itd.next);
+    ehci_set_state(ehci, async, EST_FETCHENTRY);
 
     return 1;
 }
 
 /* Section 4.10.2 - paragraph 3 */
-static int ehci_state_advqueue(EHCIState *ehci, int async, int *state)
+static int ehci_state_advqueue(EHCIQueue *q, int async)
 {
 #if 0
     /* TO-DO: 4.10.2 - paragraph 2
@@ -1412,7 +1621,7 @@
      * go to horizontal QH
      */
     if (I-bit set) {
-        *state = EST_HORIZONTALQH;
+        ehci_set_state(ehci, async, EST_HORIZONTALQH);
         goto out;
     }
 #endif
@@ -1420,100 +1629,98 @@
     /*
      * want data and alt-next qTD is valid
      */
-    if (((ehci->qh.token & QTD_TOKEN_TBYTES_MASK) != 0) &&
-        (ehci->qh.altnext_qtd > 0x1000) &&
-        (NLPTR_TBIT(ehci->qh.altnext_qtd) == 0)) {
-        DPRINTF_ST("ADVQUEUE: goto alt next qTD. "
-                   "curr 0x%08x next 0x%08x alt 0x%08x (next qh %x)\n",
-                   ehci->qh.current_qtd, ehci->qh.altnext_qtd,
-                   ehci->qh.next_qtd, ehci->qh.next);
-        ehci->qtdaddr = ehci->qh.altnext_qtd;
-        *state = EST_FETCHQTD;
+    if (((q->qh.token & QTD_TOKEN_TBYTES_MASK) != 0) &&
+        (q->qh.altnext_qtd > 0x1000) &&
+        (NLPTR_TBIT(q->qh.altnext_qtd) == 0)) {
+        q->qtdaddr = q->qh.altnext_qtd;
+        ehci_set_state(q->ehci, async, EST_FETCHQTD);
 
     /*
      *  next qTD is valid
      */
-    } else if ((ehci->qh.next_qtd > 0x1000) &&
-               (NLPTR_TBIT(ehci->qh.next_qtd) == 0)) {
-        DPRINTF_ST("ADVQUEUE: next qTD. "
-                   "curr 0x%08x next 0x%08x alt 0x%08x (next qh %x)\n",
-                   ehci->qh.current_qtd, ehci->qh.altnext_qtd,
-                   ehci->qh.next_qtd, ehci->qh.next);
-        ehci->qtdaddr = ehci->qh.next_qtd;
-        *state = EST_FETCHQTD;
+    } else if ((q->qh.next_qtd > 0x1000) &&
+               (NLPTR_TBIT(q->qh.next_qtd) == 0)) {
+        q->qtdaddr = q->qh.next_qtd;
+        ehci_set_state(q->ehci, async, EST_FETCHQTD);
 
     /*
      *  no valid qTD, try next QH
      */
     } else {
-        DPRINTF_ST("ADVQUEUE: go to horizontal QH\n");
-        *state = EST_HORIZONTALQH;
+        ehci_set_state(q->ehci, async, EST_HORIZONTALQH);
     }
 
     return 1;
 }
 
 /* Section 4.10.2 - paragraph 4 */
-static int ehci_state_fetchqtd(EHCIState *ehci, int async, int *state)
+static int ehci_state_fetchqtd(EHCIQueue *q, int async)
 {
-    EHCIqtd *qtd = &ehci->qtd;
     int again = 0;
 
-    get_dwords(NLPTR_GET(ehci->qtdaddr),(uint32_t *) qtd, sizeof(EHCIqtd) >> 2);
+    get_dwords(NLPTR_GET(q->qtdaddr),(uint32_t *) &q->qtd, sizeof(EHCIqtd) >> 2);
+    ehci_trace_qtd(q, NLPTR_GET(q->qtdaddr), &q->qtd);
 
-    if (qtd->token & QTD_TOKEN_ACTIVE) {
-        *state = EST_EXECUTE;
+    if (q->qtd.token & QTD_TOKEN_ACTIVE) {
+        ehci_set_state(q->ehci, async, EST_EXECUTE);
         again = 1;
     } else {
-        *state = EST_HORIZONTALQH;
+        ehci_set_state(q->ehci, async, EST_HORIZONTALQH);
         again = 1;
     }
 
     return again;
 }
 
-static int ehci_state_horizqh(EHCIState *ehci, int async, int *state)
+static int ehci_state_horizqh(EHCIQueue *q, int async)
 {
     int again = 0;
 
-    if (ehci->fetch_addr != ehci->qh.next) {
-        ehci->fetch_addr = ehci->qh.next;
-        *state = EST_FETCHENTRY;
+    if (ehci_get_fetch_addr(q->ehci, async) != q->qh.next) {
+        ehci_set_fetch_addr(q->ehci, async, q->qh.next);
+        ehci_set_state(q->ehci, async, EST_FETCHENTRY);
         again = 1;
     } else {
-        *state = EST_ACTIVE;
+        ehci_set_state(q->ehci, async, EST_ACTIVE);
     }
 
     return again;
 }
 
-static int ehci_state_execute(EHCIState *ehci, int async, int *state)
+/*
+ *  Write the qh back to guest physical memory.  This step isn't
+ *  in the EHCI spec but we need to do it since we don't share
+ *  physical memory with our guest VM.
+ *
+ *  The first three dwords are read-only for the EHCI, so skip them
+ *  when writing back the qh.
+ */
+static void ehci_flush_qh(EHCIQueue *q)
 {
-    EHCIqh *qh = &ehci->qh;
-    EHCIqtd *qtd = &ehci->qtd;
+    uint32_t *qh = (uint32_t *) &q->qh;
+    uint32_t dwords = sizeof(EHCIqh) >> 2;
+    uint32_t addr = NLPTR_GET(q->qhaddr);
+
+    put_dwords(addr + 3 * sizeof(uint32_t), qh + 3, dwords - 3);
+}
+
+static int ehci_state_execute(EHCIQueue *q, int async)
+{
     int again = 0;
     int reload, nakcnt;
     int smask;
 
-    if (async) {
-        DPRINTF_ST(">>>>> ASYNC STATE MACHINE execute QH 0x%08x, QTD 0x%08x\n",
-                  ehci->qhaddr, ehci->qtdaddr);
-    } else {
-        DPRINTF_ST(">>>>> PERIODIC STATE MACHINE execute\n");
-    }
-
-    if (ehci_qh_do_overlay(ehci, qh, qtd) != 0) {
+    if (ehci_qh_do_overlay(q) != 0) {
         return -1;
     }
 
-    smask = get_field(qh->epcap, QH_EPCAP_SMASK);
+    smask = get_field(q->qh.epcap, QH_EPCAP_SMASK);
 
     if (!smask) {
-        reload = get_field(qh->epchar, QH_EPCHAR_RL);
-        nakcnt = get_field(qh->altnext_qtd, QH_ALTNEXT_NAKCNT);
+        reload = get_field(q->qh.epchar, QH_EPCHAR_RL);
+        nakcnt = get_field(q->qh.altnext_qtd, QH_ALTNEXT_NAKCNT);
         if (reload && !nakcnt) {
-            DPRINTF_ST("EXECUTE: RL != 0 but NakCnt == 0 -- no execute\n");
-            *state = EST_HORIZONTALQH;
+            ehci_set_state(q->ehci, async, EST_HORIZONTALQH);
             again = 1;
             goto out;
         }
@@ -1524,119 +1731,114 @@
     // TODO Windows does not seem to ever set the MULT field
 
     if (!async) {
-        int transactCtr = get_field(qh->epcap, QH_EPCAP_MULT);
+        int transactCtr = get_field(q->qh.epcap, QH_EPCAP_MULT);
         if (!transactCtr) {
-            DPRINTF("ZERO transactctr for int qh, go HORIZ\n");
-            *state = EST_HORIZONTALQH;
+            ehci_set_state(q->ehci, async, EST_HORIZONTALQH);
             again = 1;
             goto out;
         }
     }
 
     if (async) {
-        ehci->usbsts |= USBSTS_REC;
+        ehci_set_usbsts(q->ehci, USBSTS_REC);
     }
 
-    ehci->exec_status = ehci_execute(ehci, qh);
-    if (ehci->exec_status == USB_RET_PROCERR) {
+    q->usb_status = ehci_execute(q);
+    if (q->usb_status == USB_RET_PROCERR) {
         again = -1;
         goto out;
     }
-    *state = EST_EXECUTING;
-
-    if (ehci->exec_status != USB_RET_ASYNC) {
+    if (q->usb_status == USB_RET_ASYNC) {
+        ehci_flush_qh(q);
+        trace_usb_ehci_queue_action(q, "suspend");
+        q->async = EHCI_ASYNC_INFLIGHT;
+        ehci_set_state(q->ehci, async, EST_HORIZONTALQH);
         again = 1;
-    }
-
-out:
-    return again;
-}
-
-static int ehci_state_executing(EHCIState *ehci, int async, int *state)
-{
-    EHCIqh *qh = &ehci->qh;
-    int again = 0;
-    int reload, nakcnt;
-
-    ehci->exec_status = ehci_execute_complete(ehci, qh, ehci->exec_status);
-    if (ehci->exec_status == USB_RET_ASYNC) {
-        goto out;
-    }
-    if (ehci->exec_status == USB_RET_PROCERR) {
-        again = -1;
         goto out;
     }
 
-    // 4.10.3
-    if (!async) {
-        int transactCtr = get_field(qh->epcap, QH_EPCAP_MULT);
-        transactCtr--;
-        set_field(&qh->epcap, transactCtr, QH_EPCAP_MULT);
-        // 4.10.3, bottom of page 82, should exit this state when transaction
-        // counter decrements to 0
-    }
-
-
-    reload = get_field(qh->epchar, QH_EPCHAR_RL);
-    if (reload) {
-        nakcnt = get_field(qh->altnext_qtd, QH_ALTNEXT_NAKCNT);
-        if (ehci->exec_status == USB_RET_NAK) {
-            if (nakcnt) {
-                nakcnt--;
-            }
-            DPRINTF_ST("EXECUTING: Nak occured and RL != 0, dec NakCnt to %d\n",
-                    nakcnt);
-        } else {
-            nakcnt = reload;
-            DPRINTF_ST("EXECUTING: Nak didn't occur, reloading to %d\n",
-                       nakcnt);
-        }
-        set_field(&qh->altnext_qtd, nakcnt, QH_ALTNEXT_NAKCNT);
-    }
-
-    /*
-     *  Write the qh back to guest physical memory.  This step isn't
-     *  in the EHCI spec but we need to do it since we don't share
-     *  physical memory with our guest VM.
-     */
-
-    DPRINTF("EXECUTING: write QH to VM memory: qhaddr 0x%x, next 0x%x\n",
-              ehci->qhaddr, qh->next);
-    put_dwords(NLPTR_GET(ehci->qhaddr), (uint32_t *) qh, sizeof(EHCIqh) >> 2);
-
-    /* 4.10.5 */
-    if ((ehci->exec_status == USB_RET_NAK) || (qh->token & QTD_TOKEN_ACTIVE)) {
-        *state = EST_HORIZONTALQH;
-    } else {
-        *state = EST_WRITEBACK;
-    }
-
+    ehci_set_state(q->ehci, async, EST_EXECUTING);
     again = 1;
 
 out:
     return again;
 }
 
-
-static int ehci_state_writeback(EHCIState *ehci, int async, int *state)
+static int ehci_state_executing(EHCIQueue *q, int async)
 {
-    EHCIqh *qh = &ehci->qh;
+    int again = 0;
+    int reload, nakcnt;
+
+    ehci_execute_complete(q);
+    if (q->usb_status == USB_RET_ASYNC) {
+        goto out;
+    }
+    if (q->usb_status == USB_RET_PROCERR) {
+        again = -1;
+        goto out;
+    }
+
+    // 4.10.3
+    if (!async) {
+        int transactCtr = get_field(q->qh.epcap, QH_EPCAP_MULT);
+        transactCtr--;
+        set_field(&q->qh.epcap, transactCtr, QH_EPCAP_MULT);
+        // 4.10.3, bottom of page 82, should exit this state when transaction
+        // counter decrements to 0
+    }
+
+    reload = get_field(q->qh.epchar, QH_EPCHAR_RL);
+    if (reload) {
+        nakcnt = get_field(q->qh.altnext_qtd, QH_ALTNEXT_NAKCNT);
+        if (q->usb_status == USB_RET_NAK) {
+            if (nakcnt) {
+                nakcnt--;
+            }
+        } else {
+            nakcnt = reload;
+        }
+        set_field(&q->qh.altnext_qtd, nakcnt, QH_ALTNEXT_NAKCNT);
+    }
+
+    /* 4.10.5 */
+    if ((q->usb_status == USB_RET_NAK) || (q->qh.token & QTD_TOKEN_ACTIVE)) {
+        ehci_set_state(q->ehci, async, EST_HORIZONTALQH);
+    } else {
+        ehci_set_state(q->ehci, async, EST_WRITEBACK);
+    }
+
+    again = 1;
+
+out:
+    ehci_flush_qh(q);
+    return again;
+}
+
+
+static int ehci_state_writeback(EHCIQueue *q, int async)
+{
     int again = 0;
 
     /*  Write back the QTD from the QH area */
-    DPRINTF_ST("WRITEBACK: write QTD to VM memory\n");
-    put_dwords(NLPTR_GET(ehci->qtdaddr),(uint32_t *) &qh->next_qtd,
+    ehci_trace_qtd(q, NLPTR_GET(q->qtdaddr), (EHCIqtd*) &q->qh.next_qtd);
+    put_dwords(NLPTR_GET(q->qtdaddr),(uint32_t *) &q->qh.next_qtd,
                 sizeof(EHCIqtd) >> 2);
 
-    /* TODO confirm next state.  For now, keep going if async
-     * but stop after one qtd if periodic
+    /*
+     * EHCI specs say go horizontal here.
+     *
+     * We can also advance the queue here for performance reasons.  We
+     * need to take care to only take that shortcut in case we've
+     * processed the qtd just written back without errors, i.e. halt
+     * bit is clear.
      */
-    //if (async) {
-        *state = EST_ADVANCEQUEUE;
+    if (q->qh.token & QTD_TOKEN_HALT) {
+        ehci_set_state(q->ehci, async, EST_HORIZONTALQH);
         again = 1;
-    //} else {
-    //    *state = EST_ACTIVE;
-    //}
+    } else {
+        ehci_set_state(q->ehci, async, EST_ADVANCEQUEUE);
+        again = 1;
+    }
     return again;
 }
 
@@ -1644,71 +1846,77 @@
  * This is the state machine that is common to both async and periodic
  */
 
-static int ehci_advance_state(EHCIState *ehci,
-                              int async,
-                              int state)
+static void ehci_advance_state(EHCIState *ehci,
+                               int async)
 {
+    EHCIQueue *q = NULL;
     int again;
     int iter = 0;
 
     do {
-        if (state == EST_FETCHQH) {
+        if (ehci_get_state(ehci, async) == EST_FETCHQH) {
             iter++;
             /* if we are roaming a lot of QH without executing a qTD
              * something is wrong with the linked list. TO-DO: why is
              * this hack needed?
              */
+            assert(iter < MAX_ITERATIONS);
+#if 0
             if (iter > MAX_ITERATIONS) {
                 DPRINTF("\n*** advance_state: bailing on MAX ITERATIONS***\n");
-                state = EST_ACTIVE;
+                ehci_set_state(ehci, async, EST_ACTIVE);
                 break;
             }
+#endif
         }
-        switch(state) {
+        switch(ehci_get_state(ehci, async)) {
         case EST_WAITLISTHEAD:
-            again = ehci_state_waitlisthead(ehci, async, &state);
+            again = ehci_state_waitlisthead(ehci, async);
             break;
 
         case EST_FETCHENTRY:
-            again = ehci_state_fetchentry(ehci, async, &state);
+            again = ehci_state_fetchentry(ehci, async);
             break;
 
         case EST_FETCHQH:
-            again = ehci_state_fetchqh(ehci, async, &state);
+            q = ehci_state_fetchqh(ehci, async);
+            again = q ? 1 : 0;
             break;
 
         case EST_FETCHITD:
-            again = ehci_state_fetchitd(ehci, async, &state);
+            again = ehci_state_fetchitd(ehci, async);
             break;
 
         case EST_ADVANCEQUEUE:
-            again = ehci_state_advqueue(ehci, async, &state);
+            again = ehci_state_advqueue(q, async);
             break;
 
         case EST_FETCHQTD:
-            again = ehci_state_fetchqtd(ehci, async, &state);
+            again = ehci_state_fetchqtd(q, async);
             break;
 
         case EST_HORIZONTALQH:
-            again = ehci_state_horizqh(ehci, async, &state);
+            again = ehci_state_horizqh(q, async);
             break;
 
         case EST_EXECUTE:
             iter = 0;
-            again = ehci_state_execute(ehci, async, &state);
+            again = ehci_state_execute(q, async);
             break;
 
         case EST_EXECUTING:
-            again = ehci_state_executing(ehci, async, &state);
+            assert(q != NULL);
+            again = ehci_state_executing(q, async);
             break;
 
         case EST_WRITEBACK:
-            again = ehci_state_writeback(ehci, async, &state);
+            again = ehci_state_writeback(q, async);
             break;
 
         default:
             fprintf(stderr, "Bad state!\n");
             again = -1;
+            assert(0);
             break;
         }
 
@@ -1716,32 +1924,31 @@
             fprintf(stderr, "processing error - resetting ehci HC\n");
             ehci_reset(ehci);
             again = 0;
+            assert(0);
         }
     }
     while (again);
 
     ehci_commit_interrupt(ehci);
-    return state;
 }
 
 static void ehci_advance_async_state(EHCIState *ehci)
 {
-    EHCIqh qh;
-    int state = ehci->astate;
+    int async = 1;
 
-    switch(state) {
+    switch(ehci_get_state(ehci, async)) {
     case EST_INACTIVE:
         if (!(ehci->usbcmd & USBCMD_ASE)) {
             break;
         }
-        ehci->usbsts |= USBSTS_ASS;
-        ehci->astate = EST_ACTIVE;
+        ehci_set_usbsts(ehci, USBSTS_ASS);
+        ehci_set_state(ehci, async, EST_ACTIVE);
         // No break, fall through to ACTIVE
 
     case EST_ACTIVE:
         if ( !(ehci->usbcmd & USBCMD_ASE)) {
-            ehci->usbsts &= ~USBSTS_ASS;
-            ehci->astate = EST_INACTIVE;
+            ehci_clear_usbsts(ehci, USBSTS_ASS);
+            ehci_set_state(ehci, async, EST_INACTIVE);
             break;
         }
 
@@ -1763,30 +1970,20 @@
             break;
         }
 
-        DPRINTF_ST("ASYNC: waiting for listhead, starting at %08x\n",
-                ehci->asynclistaddr);
         /* check that address register has been set */
         if (ehci->asynclistaddr == 0) {
             break;
         }
 
-        state = EST_WAITLISTHEAD;
-        /* fall through */
-
-    case EST_FETCHENTRY:
-        /* fall through */
-
-    case EST_EXECUTING:
-        get_dwords(NLPTR_GET(ehci->qhaddr), (uint32_t *) &qh,
-                   sizeof(EHCIqh) >> 2);
-        ehci->astate = ehci_advance_state(ehci, 1, state);
+        ehci_set_state(ehci, async, EST_WAITLISTHEAD);
+        ehci_advance_state(ehci, async);
         break;
 
     default:
         /* this should only be due to a developer mistake */
         fprintf(stderr, "ehci: Bad asynchronous state %d. "
                 "Resetting to active\n", ehci->astate);
-        ehci->astate = EST_ACTIVE;
+        assert(0);
     }
 }
 
@@ -1794,24 +1991,23 @@
 {
     uint32_t entry;
     uint32_t list;
+    int async = 0;
 
     // 4.6
 
-    switch(ehci->pstate) {
+    switch(ehci_get_state(ehci, async)) {
     case EST_INACTIVE:
         if ( !(ehci->frindex & 7) && (ehci->usbcmd & USBCMD_PSE)) {
-            DPRINTF("PERIODIC going active\n");
-            ehci->usbsts |= USBSTS_PSS;
-            ehci->pstate = EST_ACTIVE;
+            ehci_set_usbsts(ehci, USBSTS_PSS);
+            ehci_set_state(ehci, async, EST_ACTIVE);
             // No break, fall through to ACTIVE
         } else
             break;
 
     case EST_ACTIVE:
         if ( !(ehci->frindex & 7) && !(ehci->usbcmd & USBCMD_PSE)) {
-            DPRINTF("PERIODIC going inactive\n");
-            ehci->usbsts &= ~USBSTS_PSS;
-            ehci->pstate = EST_INACTIVE;
+            ehci_clear_usbsts(ehci, USBSTS_PSS);
+            ehci_set_state(ehci, async, EST_INACTIVE);
             break;
         }
 
@@ -1827,20 +2023,16 @@
 
         DPRINTF("PERIODIC state adv fr=%d.  [%08X] -> %08X\n",
                 ehci->frindex / 8, list, entry);
-        ehci->fetch_addr = entry;
-        ehci->pstate = ehci_advance_state(ehci, 0, EST_FETCHENTRY);
-        break;
-
-    case EST_EXECUTING:
-        DPRINTF("PERIODIC state adv for executing\n");
-        ehci->pstate = ehci_advance_state(ehci, 0, EST_EXECUTING);
+        ehci_set_fetch_addr(ehci, async,entry);
+        ehci_set_state(ehci, async, EST_FETCHENTRY);
+        ehci_advance_state(ehci, async);
         break;
 
     default:
         /* this should only be due to a developer mistake */
         fprintf(stderr, "ehci: Bad periodic state %d. "
                 "Resetting to active\n", ehci->pstate);
-        ehci->pstate = EST_ACTIVE;
+        assert(0);
     }
 }
 
@@ -1884,11 +2076,7 @@
         if (frames - i > 10) {
             skipped_frames++;
         } else {
-            // TODO could this cause periodic frames to get skipped if async
-            // active?
-            if (ehci->astate != EST_EXECUTING) {
-                ehci_advance_periodic_state(ehci);
-            }
+            ehci_advance_periodic_state(ehci);
         }
 
         ehci->last_run_usec += FRAME_TIMER_USEC;
@@ -1903,9 +2091,7 @@
     /*  Async is not inside loop since it executes everything it can once
      *  called
      */
-    if (ehci->pstate != EST_EXECUTING) {
-        ehci_advance_async_state(ehci);
-    }
+    ehci_advance_async_state(ehci);
 
     qemu_mod_timer(ehci->frame_timer, expire_time);
 }
@@ -1933,6 +2119,13 @@
     cpu_register_physical_memory(addr, size, s->mem);
 }
 
+static void ehci_device_destroy(USBBus *bus, USBDevice *dev)
+{
+    EHCIState *s = container_of(bus, EHCIState, bus);
+
+    ehci_queues_rip_device(s, dev);
+}
+
 static int usb_ehci_initfn(PCIDevice *dev);
 
 static USBPortOps ehci_port_ops = {
@@ -1941,6 +2134,10 @@
     .complete = ehci_async_complete_packet,
 };
 
+static USBBusOps ehci_bus_ops = {
+    .device_destroy = ehci_device_destroy,
+};
+
 static PCIDeviceInfo ehci_info = {
     .qdev.name    = "usb-ehci",
     .qdev.size    = sizeof(EHCIState),
@@ -1970,7 +2167,7 @@
 
     // pci_conf[0x50] = 0x01; // power management caps
 
-    pci_set_byte(&pci_conf[0x60], 0x20);  // spec release number (2.1.4)
+    pci_set_byte(&pci_conf[USB_SBRN], USB_RELEASE_2); // release number (2.1.4)
     pci_set_byte(&pci_conf[0x61], 0x20);  // frame length adjustment (2.1.5)
     pci_set_word(&pci_conf[0x62], 0x00);  // port wake up capability (2.1.6)
 
@@ -2003,7 +2200,7 @@
 
     s->irq = s->dev.irq[3];
 
-    usb_bus_new(&s->bus, &s->dev.qdev);
+    usb_bus_new(&s->bus, &ehci_bus_ops, &s->dev.qdev);
     for(i = 0; i < NB_PORTS; i++) {
         usb_register_port(&s->bus, &s->ports[i], s, i, &ehci_port_ops,
                           USB_SPEED_MASK_HIGH);
@@ -2012,6 +2209,7 @@
     }
 
     s->frame_timer = qemu_new_timer_ns(vm_clock, ehci_frame_timer, s);
+    QTAILQ_INIT(&s->queues);
 
     qemu_register_reset(ehci_reset, s);
 
diff --git a/hw/usb-hid.c b/hw/usb-hid.c
index 53b261c..d711b5c 100644
--- a/hw/usb-hid.c
+++ b/hw/usb-hid.c
@@ -142,7 +142,6 @@
     .bInterfaceNumber              = 0,
     .bNumEndpoints                 = 1,
     .bInterfaceClass               = USB_CLASS_HID,
-    .bInterfaceSubClass            = 0x01, /* boot */
     .bInterfaceProtocol            = 0x02,
     .ndesc                         = 1,
     .descs = (USBDescOther[]) {
@@ -782,13 +781,13 @@
             goto fail;
         break;
     case GET_PROTOCOL:
-        if (s->kind != USB_KEYBOARD)
+        if (s->kind != USB_KEYBOARD && s->kind != USB_MOUSE)
             goto fail;
         ret = 1;
         data[0] = s->protocol;
         break;
     case SET_PROTOCOL:
-        if (s->kind != USB_KEYBOARD)
+        if (s->kind != USB_KEYBOARD && s->kind != USB_MOUSE)
             goto fail;
         ret = 0;
         s->protocol = value;
diff --git a/hw/usb-musb.c b/hw/usb-musb.c
index 6037193..21f35afa 100644
--- a/hw/usb-musb.c
+++ b/hw/usb-musb.c
@@ -262,6 +262,7 @@
 static void musb_attach(USBPort *port);
 static void musb_detach(USBPort *port);
 static void musb_schedule_cb(USBDevice *dev, USBPacket *p);
+static void musb_device_destroy(USBBus *bus, USBDevice *dev);
 
 static USBPortOps musb_port_ops = {
     .attach = musb_attach,
@@ -269,6 +270,10 @@
     .complete = musb_schedule_cb,
 };
 
+static USBBusOps musb_bus_ops = {
+    .device_destroy = musb_device_destroy,
+};
+
 typedef struct MUSBPacket MUSBPacket;
 typedef struct MUSBEndPoint MUSBEndPoint;
 
@@ -361,7 +366,7 @@
         s->ep[i].epnum = i;
     }
 
-    usb_bus_new(&s->bus, NULL /* FIXME */);
+    usb_bus_new(&s->bus, &musb_bus_ops, NULL /* FIXME */);
     usb_register_port(&s->bus, &s->port, s, 0, &musb_port_ops,
                       USB_SPEED_MASK_LOW | USB_SPEED_MASK_FULL);
     usb_port_location(&s->port, NULL, 1);
@@ -778,6 +783,22 @@
     musb_rx_intr_set(s, epnum, 1);
 }
 
+static void musb_device_destroy(USBBus *bus, USBDevice *dev)
+{
+    MUSBState *s = container_of(bus, MUSBState, bus);
+    int ep, dir;
+
+    for (ep = 0; ep < 16; ep++) {
+        for (dir = 0; dir < 2; dir++) {
+            if (s->ep[ep].packey[dir].p.owner != dev) {
+                continue;
+            }
+            usb_cancel_packet(&s->ep[ep].packey[dir].p);
+            /* status updates needed here? */
+        }
+    }
+}
+
 static void musb_tx_rdy(MUSBState *s, int epnum)
 {
     MUSBEndPoint *ep = s->ep + epnum;
diff --git a/hw/usb-ohci.c b/hw/usb-ohci.c
index 8b966f7..832dcd6 100644
--- a/hw/usb-ohci.c
+++ b/hw/usb-ohci.c
@@ -367,6 +367,22 @@
         ohci_set_interrupt(s, OHCI_INTR_RHSC);
 }
 
+static void ohci_wakeup(USBDevice *dev)
+{
+    USBBus *bus = usb_bus_from_device(dev);
+    OHCIState *s = container_of(bus, OHCIState, bus);
+    int portnum = dev->port->index;
+    OHCIPort *port = &s->rhport[portnum];
+    if (port->ctrl & OHCI_PORT_PSS) {
+        DPRINTF("usb-ohci: port %d: wakeup\n", portnum);
+        port->ctrl |= OHCI_PORT_PSSC;
+        port->ctrl &= ~OHCI_PORT_PSS;
+        if ((s->ctl & OHCI_CTL_HCFS) == OHCI_USB_SUSPEND) {
+            ohci_set_interrupt(s, OHCI_INTR_RD);
+        }
+    }
+}
+
 /* Reset the controller */
 static void ohci_reset(void *opaque)
 {
@@ -1575,6 +1591,10 @@
         ohci->hcca = val & OHCI_HCCA_MASK;
         break;
 
+    case 7: /* HcPeriodCurrentED */
+        /* Ignore writes to this read-only register, Linux does them */
+        break;
+
     case 8: /* HcControlHeadED */
         ohci->ctrl_head = val & OHCI_EDPTR_MASK;
         break;
@@ -1644,6 +1664,16 @@
     }
 }
 
+static void ohci_device_destroy(USBBus *bus, USBDevice *dev)
+{
+    OHCIState *ohci = container_of(bus, OHCIState, bus);
+
+    if (ohci->async_td && ohci->usb_packet.owner == dev) {
+        usb_cancel_packet(&ohci->usb_packet);
+        ohci->async_td = 0;
+    }
+}
+
 /* Only dword reads are defined on OHCI register space */
 static CPUReadMemoryFunc * const ohci_readfn[3]={
     ohci_mem_read,
@@ -1661,9 +1691,14 @@
 static USBPortOps ohci_port_ops = {
     .attach = ohci_attach,
     .detach = ohci_detach,
+    .wakeup = ohci_wakeup,
     .complete = ohci_async_complete_packet,
 };
 
+static USBBusOps ohci_bus_ops = {
+    .device_destroy = ohci_device_destroy,
+};
+
 static void usb_ohci_init(OHCIState *ohci, DeviceState *dev,
                           int num_ports, uint32_t localmem_base)
 {
@@ -1691,7 +1726,7 @@
 
     ohci->name = dev->info->name;
 
-    usb_bus_new(&ohci->bus, dev);
+    usb_bus_new(&ohci->bus, &ohci_bus_ops, dev);
     ohci->num_ports = num_ports;
     for (i = 0; i < num_ports; i++) {
         usb_register_port(&ohci->bus, &ohci->rhport[i].port, ohci, i, &ohci_port_ops,
diff --git a/hw/usb-uhci.c b/hw/usb-uhci.c
index c0de05b..75cd231 100644
--- a/hw/usb-uhci.c
+++ b/hw/usb-uhci.c
@@ -234,6 +234,19 @@
     }
 }
 
+static void uhci_async_cancel_device(UHCIState *s, USBDevice *dev)
+{
+    UHCIAsync *curr, *n;
+
+    QTAILQ_FOREACH_SAFE(curr, &s->async_pending, next, n) {
+        if (curr->packet.owner != dev) {
+            continue;
+        }
+        uhci_async_unlink(s, curr);
+        uhci_async_cancel(s, curr);
+    }
+}
+
 static void uhci_async_cancel_all(UHCIState *s)
 {
     UHCIAsync *curr, *n;
@@ -411,6 +424,8 @@
     case 0x00:
         if ((val & UHCI_CMD_RS) && !(s->cmd & UHCI_CMD_RS)) {
             /* start frame processing */
+            s->expire_time = qemu_get_clock_ns(vm_clock) +
+                (get_ticks_per_sec() / FRAME_TIMER_FREQ);
             qemu_mod_timer(s->frame_timer, qemu_get_clock_ns(vm_clock));
             s->status &= ~UHCI_STS_HCHALTED;
         } else if (!(val & UHCI_CMD_RS)) {
@@ -1081,6 +1096,13 @@
     register_ioport_read(addr, 32, 1, uhci_ioport_readb, s);
 }
 
+static void uhci_device_destroy(USBBus *bus, USBDevice *dev)
+{
+    UHCIState *s = container_of(bus, UHCIState, bus);
+
+    uhci_async_cancel_device(s, dev);
+}
+
 static USBPortOps uhci_port_ops = {
     .attach = uhci_attach,
     .detach = uhci_detach,
@@ -1088,6 +1110,10 @@
     .complete = uhci_async_complete,
 };
 
+static USBBusOps uhci_bus_ops = {
+    .device_destroy = uhci_device_destroy,
+};
+
 static int usb_uhci_common_initfn(UHCIState *s)
 {
     uint8_t *pci_conf = s->dev.config;
@@ -1098,17 +1124,15 @@
     pci_config_set_class(pci_conf, PCI_CLASS_SERIAL_USB);
     /* TODO: reset value should be 0. */
     pci_conf[PCI_INTERRUPT_PIN] = 4; // interrupt pin 3
-    pci_conf[0x60] = 0x10; // release number
+    pci_conf[USB_SBRN] = USB_RELEASE_1; // release number
 
-    usb_bus_new(&s->bus, &s->dev.qdev);
+    usb_bus_new(&s->bus, &uhci_bus_ops, &s->dev.qdev);
     for(i = 0; i < NB_PORTS; i++) {
         usb_register_port(&s->bus, &s->ports[i].port, s, i, &uhci_port_ops,
                           USB_SPEED_MASK_LOW | USB_SPEED_MASK_FULL);
         usb_port_location(&s->ports[i].port, NULL, i+1);
     }
     s->frame_timer = qemu_new_timer_ns(vm_clock, uhci_frame_timer, s);
-    s->expire_time = qemu_get_clock_ns(vm_clock) +
-        (get_ticks_per_sec() / FRAME_TIMER_FREQ);
     s->num_ports_vmstate = NB_PORTS;
     QTAILQ_INIT(&s->async_pending);
 
diff --git a/hw/usb.h b/hw/usb.h
index 9882400..06ce058 100644
--- a/hw/usb.h
+++ b/hw/usb.h
@@ -26,6 +26,12 @@
 #include "qdev.h"
 #include "qemu-queue.h"
 
+/* Constants related to the USB / PCI interaction */
+#define USB_SBRN    0x60 /* Serial Bus Release Number Register */
+#define USB_RELEASE_1  0x10 /* USB 1.0 */
+#define USB_RELEASE_2  0x20 /* USB 2.0 */
+#define USB_RELEASE_3  0x30 /* USB 3.0 */
+
 #define USB_TOKEN_SETUP 0x2d
 #define USB_TOKEN_IN    0x69 /* device -> host */
 #define USB_TOKEN_OUT   0xe1 /* host -> device */
@@ -132,6 +138,7 @@
 #define USB_ENDPOINT_XFER_INT		3
 
 typedef struct USBBus USBBus;
+typedef struct USBBusOps USBBusOps;
 typedef struct USBPort USBPort;
 typedef struct USBDevice USBDevice;
 typedef struct USBDeviceInfo USBDeviceInfo;
@@ -323,6 +330,7 @@
 
 struct USBBus {
     BusState qbus;
+    USBBusOps *ops;
     int busnr;
     int nfree;
     int nused;
@@ -331,7 +339,11 @@
     QTAILQ_ENTRY(USBBus) next;
 };
 
-void usb_bus_new(USBBus *bus, DeviceState *host);
+struct USBBusOps {
+    void (*device_destroy)(USBBus *bus, USBDevice *dev);
+};
+
+void usb_bus_new(USBBus *bus, USBBusOps *ops, DeviceState *host);
 USBBus *usb_bus_find(int busnr);
 void usb_qdev_register(USBDeviceInfo *info);
 void usb_qdev_register_many(USBDeviceInfo *info);
diff --git a/hw/vga-isa.c b/hw/vga-isa.c
index fde0d56..245841f 100644
--- a/hw/vga-isa.c
+++ b/hw/vga-isa.c
@@ -77,7 +77,6 @@
     .qdev.size     = sizeof(ISAVGAState),
     .qdev.vmsd     = &vmstate_vga_common,
     .qdev.reset     = vga_reset_isa,
-    .qdev.no_user  = 1,
     .init          = vga_initfn,
 };
 
diff --git a/hw/virtio-pci.c b/hw/virtio-pci.c
index c19629d..c018351 100644
--- a/hw/virtio-pci.c
+++ b/hw/virtio-pci.c
@@ -26,6 +26,7 @@
 #include "loader.h"
 #include "kvm.h"
 #include "blockdev.h"
+#include "virtio-pci.h"
 
 /* from Linux's linux/virtio_pci.h */
 
@@ -95,27 +96,6 @@
  */
 #define wmb() do { } while (0)
 
-/* PCI bindings.  */
-
-typedef struct {
-    PCIDevice pci_dev;
-    VirtIODevice *vdev;
-    uint32_t flags;
-    uint32_t addr;
-    uint32_t class_code;
-    uint32_t nvectors;
-    BlockConf block;
-    NICConf nic;
-    uint32_t host_features;
-#ifdef CONFIG_LINUX
-    V9fsConf fsconf;
-#endif
-    virtio_serial_conf serial;
-    virtio_net_conf net;
-    bool ioeventfd_disabled;
-    bool ioeventfd_started;
-} VirtIOPCIProxy;
-
 /* virtio device */
 
 static void virtio_pci_notify(void *opaque, uint16_t vector)
@@ -669,7 +649,7 @@
     .vmstate_change = virtio_pci_vmstate_change,
 };
 
-static void virtio_init_pci(VirtIOPCIProxy *proxy, VirtIODevice *vdev,
+void virtio_init_pci(VirtIOPCIProxy *proxy, VirtIODevice *vdev,
                             uint16_t vendor, uint16_t device,
                             uint16_t class_code, uint8_t pif)
 {
@@ -835,25 +815,6 @@
     return 0;
 }
 
-#ifdef CONFIG_VIRTFS
-static int virtio_9p_init_pci(PCIDevice *pci_dev)
-{
-    VirtIOPCIProxy *proxy = DO_UPCAST(VirtIOPCIProxy, pci_dev, pci_dev);
-    VirtIODevice *vdev;
-
-    vdev = virtio_9p_init(&pci_dev->qdev, &proxy->fsconf);
-    vdev->nvectors = proxy->nvectors;
-    virtio_init_pci(proxy, vdev,
-                    PCI_VENDOR_ID_REDHAT_QUMRANET,
-                    0x1009,
-                    0x2,
-                    0x00);
-    /* make the actual value visible */
-    proxy->nvectors = vdev->nvectors;
-    return 0;
-}
-#endif
-
 static PCIDeviceInfo virtio_info[] = {
     {
         .qdev.name = "virtio-blk-pci",
@@ -922,20 +883,6 @@
         },
         .qdev.reset = virtio_pci_reset,
     },{
-#ifdef CONFIG_VIRTFS
-        .qdev.name = "virtio-9p-pci",
-        .qdev.alias = "virtio-9p",
-        .qdev.size = sizeof(VirtIOPCIProxy),
-        .init      = virtio_9p_init_pci,
-        .qdev.props = (Property[]) {
-            DEFINE_PROP_UINT32("vectors", VirtIOPCIProxy, nvectors, 2),
-            DEFINE_VIRTIO_COMMON_FEATURES(VirtIOPCIProxy, host_features),
-            DEFINE_PROP_STRING("mount_tag", VirtIOPCIProxy, fsconf.tag),
-            DEFINE_PROP_STRING("fsdev", VirtIOPCIProxy, fsconf.fsdev_id),
-            DEFINE_PROP_END_OF_LIST(),
-        },
-    }, {
-#endif
         /* end of list */
     }
 };
diff --git a/hw/virtio-pci.h b/hw/virtio-pci.h
new file mode 100644
index 0000000..a4b5fd3
--- /dev/null
+++ b/hw/virtio-pci.h
@@ -0,0 +1,43 @@
+/*
+ * Virtio PCI Bindings
+ *
+ * Copyright IBM, Corp. 2007
+ * Copyright (c) 2009 CodeSourcery
+ *
+ * Authors:
+ *  Anthony Liguori   <aliguori@us.ibm.com>
+ *  Paul Brook        <paul@codesourcery.com>
+ *
+ * This work is licensed under the terms of the GNU GPL, version 2.  See
+ * the COPYING file in the top-level directory.
+ */
+
+#ifndef QEMU_VIRTIO_PCI_H
+#define QEMU_VIRTIO_PCI_H
+
+#include "virtio-net.h"
+#include "virtio-serial.h"
+
+typedef struct {
+    PCIDevice pci_dev;
+    VirtIODevice *vdev;
+    uint32_t flags;
+    uint32_t addr;
+    uint32_t class_code;
+    uint32_t nvectors;
+    BlockConf block;
+    NICConf nic;
+    uint32_t host_features;
+#ifdef CONFIG_LINUX
+    V9fsConf fsconf;
+#endif
+    virtio_serial_conf serial;
+    virtio_net_conf net;
+    bool ioeventfd_disabled;
+    bool ioeventfd_started;
+} VirtIOPCIProxy;
+
+extern void virtio_init_pci(VirtIOPCIProxy *proxy, VirtIODevice *vdev,
+                            uint16_t vendor, uint16_t device,
+                            uint16_t class_code, uint8_t pif);
+#endif
diff --git a/json-lexer.c b/json-lexer.c
index 65c9720..c21338f 100644
--- a/json-lexer.c
+++ b/json-lexer.c
@@ -18,6 +18,8 @@
 #include "qemu-common.h"
 #include "json-lexer.h"
 
+#define MAX_TOKEN_SIZE (64ULL << 20)
+
 /*
  * \"([^\\\"]|(\\\"\\'\\\\\\/\\b\\f\\n\\r\\t\\u[0-9a-fA-F][0-9a-fA-F][0-9a-fA-F][0-9a-fA-F]))*\"
  * '([^\\']|(\\\"\\'\\\\\\/\\b\\f\\n\\r\\t\\u[0-9a-fA-F][0-9a-fA-F][0-9a-fA-F][0-9a-fA-F]))*'
@@ -103,7 +105,8 @@
         ['u'] = IN_DQ_UCODE0,
     },
     [IN_DQ_STRING] = {
-        [1 ... 0xFF] = IN_DQ_STRING,
+        [1 ... 0xBF] = IN_DQ_STRING,
+        [0xC2 ... 0xF4] = IN_DQ_STRING,
         ['\\'] = IN_DQ_STRING_ESCAPE,
         ['"'] = JSON_STRING,
     },
@@ -142,7 +145,8 @@
         ['u'] = IN_SQ_UCODE0,
     },
     [IN_SQ_STRING] = {
-        [1 ... 0xFF] = IN_SQ_STRING,
+        [1 ... 0xBF] = IN_SQ_STRING,
+        [0xC2 ... 0xF4] = IN_SQ_STRING,
         ['\\'] = IN_SQ_STRING_ESCAPE,
         ['\''] = JSON_STRING,
     },
@@ -272,7 +276,7 @@
     lexer->x = lexer->y = 0;
 }
 
-static int json_lexer_feed_char(JSONLexer *lexer, char ch)
+static int json_lexer_feed_char(JSONLexer *lexer, char ch, bool flush)
 {
     int char_consumed, new_state;
 
@@ -303,12 +307,41 @@
             new_state = IN_START;
             break;
         case IN_ERROR:
-            return -EINVAL;
+            /* XXX: To avoid having previous bad input leaving the parser in an
+             * unresponsive state where we consume unpredictable amounts of
+             * subsequent "good" input, percolate this error state up to the
+             * tokenizer/parser by forcing a NULL object to be emitted, then
+             * reset state.
+             *
+             * Also note that this handling is required for reliable channel
+             * negotiation between QMP and the guest agent, since chr(0xFF)
+             * is placed at the beginning of certain events to ensure proper
+             * delivery when the channel is in an unknown state. chr(0xFF) is
+             * never a valid ASCII/UTF-8 sequence, so this should reliably
+             * induce an error/flush state.
+             */
+            lexer->emit(lexer, lexer->token, JSON_ERROR, lexer->x, lexer->y);
+            QDECREF(lexer->token);
+            lexer->token = qstring_new();
+            new_state = IN_START;
+            lexer->state = new_state;
+            return 0;
         default:
             break;
         }
         lexer->state = new_state;
-    } while (!char_consumed);
+    } while (!char_consumed && !flush);
+
+    /* Do not let a single token grow to an arbitrarily large size,
+     * this is a security consideration.
+     */
+    if (lexer->token->length > MAX_TOKEN_SIZE) {
+        lexer->emit(lexer, lexer->token, lexer->state, lexer->x, lexer->y);
+        QDECREF(lexer->token);
+        lexer->token = qstring_new();
+        lexer->state = IN_START;
+    }
+
     return 0;
 }
 
@@ -319,7 +352,7 @@
     for (i = 0; i < size; i++) {
         int err;
 
-        err = json_lexer_feed_char(lexer, buffer[i]);
+        err = json_lexer_feed_char(lexer, buffer[i], false);
         if (err < 0) {
             return err;
         }
@@ -330,7 +363,7 @@
 
 int json_lexer_flush(JSONLexer *lexer)
 {
-    return lexer->state == IN_START ? 0 : json_lexer_feed_char(lexer, 0);
+    return lexer->state == IN_START ? 0 : json_lexer_feed_char(lexer, 0, true);
 }
 
 void json_lexer_destroy(JSONLexer *lexer)
diff --git a/json-lexer.h b/json-lexer.h
index 3b50c46..10bc0a7 100644
--- a/json-lexer.h
+++ b/json-lexer.h
@@ -25,6 +25,7 @@
     JSON_STRING,
     JSON_ESCAPE,
     JSON_SKIP,
+    JSON_ERROR,
 } JSONTokenType;
 
 typedef struct JSONLexer JSONLexer;
diff --git a/json-parser.c b/json-parser.c
index 6c06ef9..849e215 100644
--- a/json-parser.c
+++ b/json-parser.c
@@ -22,9 +22,11 @@
 #include "qbool.h"
 #include "json-parser.h"
 #include "json-lexer.h"
+#include "qerror.h"
 
 typedef struct JSONParserContext
 {
+    Error *err;
 } JSONParserContext;
 
 #define BUG_ON(cond) assert(!(cond))
@@ -95,11 +97,15 @@
                                            QObject *token, const char *msg, ...)
 {
     va_list ap;
+    char message[1024];
     va_start(ap, msg);
-    fprintf(stderr, "parse error: ");
-    vfprintf(stderr, msg, ap);
-    fprintf(stderr, "\n");
+    vsnprintf(message, sizeof(message), msg, ap);
     va_end(ap);
+    if (ctxt->err) {
+        error_free(ctxt->err);
+        ctxt->err = NULL;
+    }
+    error_set(&ctxt->err, QERR_JSON_PARSE_ERROR, message);
 }
 
 /**
@@ -269,10 +275,15 @@
  */
 static int parse_pair(JSONParserContext *ctxt, QDict *dict, QList **tokens, va_list *ap)
 {
-    QObject *key, *token = NULL, *value, *peek;
+    QObject *key = NULL, *token = NULL, *value, *peek;
     QList *working = qlist_copy(*tokens);
 
     peek = qlist_peek(working);
+    if (peek == NULL) {
+        parse_error(ctxt, NULL, "premature EOI");
+        goto out;
+    }
+
     key = parse_value(ctxt, &working, ap);
     if (!key || qobject_type(key) != QTYPE_QSTRING) {
         parse_error(ctxt, peek, "key is not a string in object");
@@ -280,6 +291,11 @@
     }
 
     token = qlist_pop(working);
+    if (token == NULL) {
+        parse_error(ctxt, NULL, "premature EOI");
+        goto out;
+    }
+
     if (!token_is_operator(token, ':')) {
         parse_error(ctxt, token, "missing : in object pair");
         goto out;
@@ -315,6 +331,10 @@
     QList *working = qlist_copy(*tokens);
 
     token = qlist_pop(working);
+    if (token == NULL) {
+        goto out;
+    }
+
     if (!token_is_operator(token, '{')) {
         goto out;
     }
@@ -324,12 +344,22 @@
     dict = qdict_new();
 
     peek = qlist_peek(working);
+    if (peek == NULL) {
+        parse_error(ctxt, NULL, "premature EOI");
+        goto out;
+    }
+
     if (!token_is_operator(peek, '}')) {
         if (parse_pair(ctxt, dict, &working, ap) == -1) {
             goto out;
         }
 
         token = qlist_pop(working);
+        if (token == NULL) {
+            parse_error(ctxt, NULL, "premature EOI");
+            goto out;
+        }
+
         while (!token_is_operator(token, '}')) {
             if (!token_is_operator(token, ',')) {
                 parse_error(ctxt, token, "expected separator in dict");
@@ -343,6 +373,10 @@
             }
 
             token = qlist_pop(working);
+            if (token == NULL) {
+                parse_error(ctxt, NULL, "premature EOI");
+                goto out;
+            }
         }
         qobject_decref(token);
         token = NULL;
@@ -371,6 +405,10 @@
     QList *working = qlist_copy(*tokens);
 
     token = qlist_pop(working);
+    if (token == NULL) {
+        goto out;
+    }
+
     if (!token_is_operator(token, '[')) {
         goto out;
     }
@@ -380,6 +418,11 @@
     list = qlist_new();
 
     peek = qlist_peek(working);
+    if (peek == NULL) {
+        parse_error(ctxt, NULL, "premature EOI");
+        goto out;
+    }
+
     if (!token_is_operator(peek, ']')) {
         QObject *obj;
 
@@ -392,6 +435,11 @@
         qlist_append_obj(list, obj);
 
         token = qlist_pop(working);
+        if (token == NULL) {
+            parse_error(ctxt, NULL, "premature EOI");
+            goto out;
+        }
+
         while (!token_is_operator(token, ']')) {
             if (!token_is_operator(token, ',')) {
                 parse_error(ctxt, token, "expected separator in list");
@@ -410,6 +458,10 @@
             qlist_append_obj(list, obj);
 
             token = qlist_pop(working);
+            if (token == NULL) {
+                parse_error(ctxt, NULL, "premature EOI");
+                goto out;
+            }
         }
 
         qobject_decref(token);
@@ -438,6 +490,9 @@
     QList *working = qlist_copy(*tokens);
 
     token = qlist_pop(working);
+    if (token == NULL) {
+        goto out;
+    }
 
     if (token_get_type(token) != JSON_KEYWORD) {
         goto out;
@@ -475,6 +530,9 @@
     }
 
     token = qlist_pop(working);
+    if (token == NULL) {
+        goto out;
+    }
 
     if (token_is_escape(token, "%p")) {
         obj = va_arg(*ap, QObject *);
@@ -514,6 +572,10 @@
     QList *working = qlist_copy(*tokens);
 
     token = qlist_pop(working);
+    if (token == NULL) {
+        goto out;
+    }
+
     switch (token_get_type(token)) {
     case JSON_STRING:
         obj = QOBJECT(qstring_from_escaped_str(ctxt, token));
@@ -565,13 +627,24 @@
 
 QObject *json_parser_parse(QList *tokens, va_list *ap)
 {
+    return json_parser_parse_err(tokens, ap, NULL);
+}
+
+QObject *json_parser_parse_err(QList *tokens, va_list *ap, Error **errp)
+{
     JSONParserContext ctxt = {};
-    QList *working = qlist_copy(tokens);
+    QList *working;
     QObject *result;
 
+    if (!tokens) {
+        return NULL;
+    }
+    working = qlist_copy(tokens);
     result = parse_value(&ctxt, &working, ap);
 
     QDECREF(working);
 
+    error_propagate(errp, ctxt.err);
+
     return result;
 }
diff --git a/json-parser.h b/json-parser.h
index 97f43f6..8f2b5ec 100644
--- a/json-parser.h
+++ b/json-parser.h
@@ -16,7 +16,9 @@
 
 #include "qemu-common.h"
 #include "qlist.h"
+#include "error.h"
 
 QObject *json_parser_parse(QList *tokens, va_list *ap);
+QObject *json_parser_parse_err(QList *tokens, va_list *ap, Error **errp);
 
 #endif
diff --git a/json-streamer.c b/json-streamer.c
index f7e7a68..c255c78 100644
--- a/json-streamer.c
+++ b/json-streamer.c
@@ -18,6 +18,9 @@
 #include "json-lexer.h"
 #include "json-streamer.h"
 
+#define MAX_TOKEN_SIZE (64ULL << 20)
+#define MAX_NESTING (1ULL << 10)
+
 static void json_message_process_token(JSONLexer *lexer, QString *token, JSONTokenType type, int x, int y)
 {
     JSONMessageParser *parser = container_of(lexer, JSONMessageParser, lexer);
@@ -49,14 +52,44 @@
     qdict_put(dict, "x", qint_from_int(x));
     qdict_put(dict, "y", qint_from_int(y));
 
+    parser->token_size += token->length;
+
     qlist_append(parser->tokens, dict);
 
-    if (parser->brace_count == 0 &&
-        parser->bracket_count == 0) {
-        parser->emit(parser, parser->tokens);
-        QDECREF(parser->tokens);
-        parser->tokens = qlist_new();
+    if (type == JSON_ERROR) {
+        goto out_emit_bad;
+    } else if (parser->brace_count < 0 ||
+        parser->bracket_count < 0 ||
+        (parser->brace_count == 0 &&
+         parser->bracket_count == 0)) {
+        goto out_emit;
+    } else if (parser->token_size > MAX_TOKEN_SIZE ||
+               parser->bracket_count > MAX_NESTING ||
+               parser->brace_count > MAX_NESTING) {
+        /* Security consideration, we limit total memory allocated per object
+         * and the maximum recursion depth that a message can force.
+         */
+        goto out_emit;
     }
+
+    return;
+
+out_emit_bad:
+    /* clear out token list and tell the parser to emit and error
+     * indication by passing it a NULL list
+     */
+    QDECREF(parser->tokens);
+    parser->tokens = NULL;
+out_emit:
+    /* send current list of tokens to parser and reset tokenizer */
+    parser->brace_count = 0;
+    parser->bracket_count = 0;
+    parser->emit(parser, parser->tokens);
+    if (parser->tokens) {
+        QDECREF(parser->tokens);
+    }
+    parser->tokens = qlist_new();
+    parser->token_size = 0;
 }
 
 void json_message_parser_init(JSONMessageParser *parser,
@@ -66,6 +99,7 @@
     parser->brace_count = 0;
     parser->bracket_count = 0;
     parser->tokens = qlist_new();
+    parser->token_size = 0;
 
     json_lexer_init(&parser->lexer, json_message_process_token);
 }
diff --git a/json-streamer.h b/json-streamer.h
index 09f3bd7..f09bc4d 100644
--- a/json-streamer.h
+++ b/json-streamer.h
@@ -24,6 +24,7 @@
     int brace_count;
     int bracket_count;
     QList *tokens;
+    uint64_t token_size;
 } JSONMessageParser;
 
 void json_message_parser_init(JSONMessageParser *parser,
diff --git a/linux-user/main.c b/linux-user/main.c
index 088def3..04da0a4 100644
--- a/linux-user/main.c
+++ b/linux-user/main.c
@@ -2508,49 +2508,27 @@
             fprintf(stderr, "Machine check exception. Exit\n");
             exit(1);
             break;
-        case EXCP_ARITH:
-            env->lock_addr = -1;
-            info.si_signo = TARGET_SIGFPE;
-            info.si_errno = 0;
-            info.si_code = TARGET_FPE_FLTINV;
-            info._sifields._sigfault._addr = env->pc;
-            queue_signal(env, info.si_signo, &info);
-            break;
-        case EXCP_HW_INTERRUPT:
+        case EXCP_SMP_INTERRUPT:
+        case EXCP_CLK_INTERRUPT:
+        case EXCP_DEV_INTERRUPT:
             fprintf(stderr, "External interrupt. Exit\n");
             exit(1);
             break;
-        case EXCP_DFAULT:
+        case EXCP_MMFAULT:
             env->lock_addr = -1;
             info.si_signo = TARGET_SIGSEGV;
             info.si_errno = 0;
-            info.si_code = (page_get_flags(env->ipr[IPR_EXC_ADDR]) & PAGE_VALID
+            info.si_code = (page_get_flags(env->trap_arg0) & PAGE_VALID
                             ? TARGET_SEGV_ACCERR : TARGET_SEGV_MAPERR);
-            info._sifields._sigfault._addr = env->ipr[IPR_EXC_ADDR];
+            info._sifields._sigfault._addr = env->trap_arg0;
             queue_signal(env, info.si_signo, &info);
             break;
-        case EXCP_DTB_MISS_PAL:
-            fprintf(stderr, "MMU data TLB miss in PALcode\n");
-            exit(1);
-            break;
-        case EXCP_ITB_MISS:
-            fprintf(stderr, "MMU instruction TLB miss\n");
-            exit(1);
-            break;
-        case EXCP_ITB_ACV:
-            fprintf(stderr, "MMU instruction access violation\n");
-            exit(1);
-            break;
-        case EXCP_DTB_MISS_NATIVE:
-            fprintf(stderr, "MMU data TLB miss\n");
-            exit(1);
-            break;
         case EXCP_UNALIGN:
             env->lock_addr = -1;
             info.si_signo = TARGET_SIGBUS;
             info.si_errno = 0;
             info.si_code = TARGET_BUS_ADRALN;
-            info._sifields._sigfault._addr = env->ipr[IPR_EXC_ADDR];
+            info._sifields._sigfault._addr = env->trap_arg0;
             queue_signal(env, info.si_signo, &info);
             break;
         case EXCP_OPCDEC:
@@ -2562,12 +2540,20 @@
             info._sifields._sigfault._addr = env->pc;
             queue_signal(env, info.si_signo, &info);
             break;
+        case EXCP_ARITH:
+            env->lock_addr = -1;
+            info.si_signo = TARGET_SIGFPE;
+            info.si_errno = 0;
+            info.si_code = TARGET_FPE_FLTINV;
+            info._sifields._sigfault._addr = env->pc;
+            queue_signal(env, info.si_signo, &info);
+            break;
         case EXCP_FEN:
             /* No-op.  Linux simply re-enables the FPU.  */
             break;
-        case EXCP_CALL_PAL ... (EXCP_CALL_PALP - 1):
+        case EXCP_CALL_PAL:
             env->lock_addr = -1;
-            switch ((trapnr >> 6) | 0x80) {
+            switch (env->error_code) {
             case 0x80:
                 /* BPT */
                 info.si_signo = TARGET_SIGTRAP;
@@ -2658,8 +2644,6 @@
                 goto do_sigill;
             }
             break;
-        case EXCP_CALL_PALP ... (EXCP_CALL_PALE - 1):
-            goto do_sigill;
         case EXCP_DEBUG:
             info.si_signo = gdb_handlesig (env, TARGET_SIGTRAP);
             if (info.si_signo) {
diff --git a/linux-user/signal.c b/linux-user/signal.c
index c7a375f..11b25be 100644
--- a/linux-user/signal.c
+++ b/linux-user/signal.c
@@ -21,7 +21,6 @@
 #include <string.h>
 #include <stdarg.h>
 #include <unistd.h>
-#include <signal.h>
 #include <errno.h>
 #include <assert.h>
 #include <sys/ucontext.h>
diff --git a/monitor.c b/monitor.c
index f63cce0..6af6a4d 100644
--- a/monitor.c
+++ b/monitor.c
@@ -2544,16 +2544,21 @@
 #endif
 
 #if defined(TARGET_I386)
-static void do_inject_nmi(Monitor *mon, const QDict *qdict)
+static int do_inject_nmi(Monitor *mon, const QDict *qdict, QObject **ret_data)
 {
     CPUState *env;
-    int cpu_index = qdict_get_int(qdict, "cpu_index");
 
-    for (env = first_cpu; env != NULL; env = env->next_cpu)
-        if (env->cpu_index == cpu_index) {
-            cpu_interrupt(env, CPU_INTERRUPT_NMI);
-            break;
-        }
+    for (env = first_cpu; env != NULL; env = env->next_cpu) {
+        cpu_interrupt(env, CPU_INTERRUPT_NMI);
+    }
+
+    return 0;
+}
+#else
+static int do_inject_nmi(Monitor *mon, const QDict *qdict, QObject **ret_data)
+{
+    qerror_report(QERR_UNSUPPORTED);
+    return -1;
 }
 #endif
 
diff --git a/net/slirp.c b/net/slirp.c
index e387a11..e057a14 100644
--- a/net/slirp.c
+++ b/net/slirp.c
@@ -614,7 +614,7 @@
     }
 
     fwd = qemu_malloc(sizeof(struct GuestFwd));
-    snprintf(buf, sizeof(buf), "guestfwd.tcp:%d", port);
+    snprintf(buf, sizeof(buf), "guestfwd.tcp.%d", port);
     fwd->hd = qemu_chr_open(buf, p, NULL);
     if (!fwd->hd) {
         error_report("could not open guest forwarding device '%s'", buf);
diff --git a/net/tap.c b/net/tap.c
index b8cd252..1f26dc9 100644
--- a/net/tap.c
+++ b/net/tap.c
@@ -27,7 +27,6 @@
 
 #include "config-host.h"
 
-#include <signal.h>
 #include <sys/ioctl.h>
 #include <sys/stat.h>
 #include <sys/wait.h>
diff --git a/pc-bios/s390-zipl.rom b/pc-bios/s390-zipl.rom
index f7af9b1..3115128 100644
--- a/pc-bios/s390-zipl.rom
+++ b/pc-bios/s390-zipl.rom
Binary files differ
diff --git a/posix-aio-compat.c b/posix-aio-compat.c
index f3cc868..c4116e3 100644
--- a/posix-aio-compat.c
+++ b/posix-aio-compat.c
@@ -17,7 +17,6 @@
 #include <unistd.h>
 #include <errno.h>
 #include <time.h>
-#include <signal.h>
 #include <string.h>
 #include <stdlib.h>
 #include <stdio.h>
diff --git a/qemu-char.c b/qemu-char.c
index 5e04a20..fb13b28 100644
--- a/qemu-char.c
+++ b/qemu-char.c
@@ -35,7 +35,6 @@
 
 #include <unistd.h>
 #include <fcntl.h>
-#include <signal.h>
 #include <time.h>
 #include <errno.h>
 #include <sys/time.h>
diff --git a/qemu-common.h b/qemu-common.h
index b851b20..39fabc9 100644
--- a/qemu-common.h
+++ b/qemu-common.h
@@ -39,6 +39,7 @@
 #include <sys/stat.h>
 #include <sys/time.h>
 #include <assert.h>
+#include <signal.h>
 
 #ifdef _WIN32
 #include "qemu-os-win32.h"
diff --git a/qemu-config.c b/qemu-config.c
index 5d7ffa2..c63741c 100644
--- a/qemu-config.c
+++ b/qemu-config.c
@@ -306,7 +306,7 @@
             .name = "file",
             .type = QEMU_OPT_STRING,
         },
-        { /* end if list */ }
+        { /* end of list */ }
     },
 };
 #endif
@@ -385,6 +385,12 @@
             .name = "disable-ticketing",
             .type = QEMU_OPT_BOOL,
         },{
+            .name = "disable-copy-paste",
+            .type = QEMU_OPT_BOOL,
+        },{
+            .name = "sasl",
+            .type = QEMU_OPT_BOOL,
+        },{
             .name = "x509-dir",
             .type = QEMU_OPT_STRING,
         },{
@@ -430,7 +436,7 @@
             .name = "playback-compression",
             .type = QEMU_OPT_BOOL,
         },
-        { /* end if list */ }
+        { /* end of list */ }
     },
 };
 
@@ -446,7 +452,7 @@
             .name = "romfile",
             .type = QEMU_OPT_STRING,
         },
-        { /* end if list */ }
+        { /* end of list */ }
     },
 };
 
diff --git a/qemu-io.c b/qemu-io.c
index 4470e49..dd4ebf5 100644
--- a/qemu-io.c
+++ b/qemu-io.c
@@ -1655,7 +1655,7 @@
 			flags |= BDRV_O_SNAPSHOT;
 			break;
 		case 'n':
-			flags |= BDRV_O_NOCACHE;
+			flags |= BDRV_O_NOCACHE | BDRV_O_CACHE_WB;
 			break;
 		case 'r':
 			readonly = 1;
@@ -1751,7 +1751,7 @@
 			flags |= BDRV_O_SNAPSHOT;
 			break;
 		case 'n':
-			flags |= BDRV_O_NOCACHE;
+			flags |= BDRV_O_NOCACHE | BDRV_O_CACHE_WB;
 			break;
 		case 'c':
 			add_user_command(optarg);
diff --git a/qemu-nbd.c b/qemu-nbd.c
index e858033..110d78e 100644
--- a/qemu-nbd.c
+++ b/qemu-nbd.c
@@ -238,7 +238,7 @@
             flags |= BDRV_O_SNAPSHOT;
             break;
         case 'n':
-            flags |= BDRV_O_NOCACHE;
+            flags |= BDRV_O_NOCACHE | BDRV_O_CACHE_WB;
             break;
         case 'b':
             bindto = optarg;
diff --git a/qemu-options.hx b/qemu-options.hx
index 82e085a..f2ef9a1 100644
--- a/qemu-options.hx
+++ b/qemu-options.hx
@@ -714,9 +714,25 @@
 @item password=<secret>
 Set the password you need to authenticate.
 
+@item sasl
+Require that the client use SASL to authenticate with the spice.
+The exact choice of authentication method used is controlled from the
+system / user's SASL configuration file for the 'qemu' service. This
+is typically found in /etc/sasl2/qemu.conf. If running QEMU as an
+unprivileged user, an environment variable SASL_CONF_PATH can be used
+to make it search alternate locations for the service config.
+While some SASL auth methods can also provide data encryption (eg GSSAPI),
+it is recommended that SASL always be combined with the 'tls' and
+'x509' settings to enable use of SSL and server certificates. This
+ensures a data encryption preventing compromise of authentication
+credentials.
+
 @item disable-ticketing
 Allow client connects without authentication.
 
+@item disable-copy-paste
+Disable copy paste between the client and the guest.
+
 @item tls-port=<nr>
 Set the TCP port spice is listening on for encrypted channels.
 
@@ -1161,9 +1177,9 @@
 guest network, i.e. x.x.x.2.
 
 @item restrict=y|yes|n|no
-If this options is enabled, the guest will be isolated, i.e. it will not be
+If this option is enabled, the guest will be isolated, i.e. it will not be
 able to contact the host and no guest IP packets will be routed over the host
-to the outside. This option does not affect explicitly set forwarding rule.
+to the outside. This option does not affect any explicitly set forwarding rules.
 
 @item hostname=@var{name}
 Specifies the client hostname reported by the builtin DHCP server.
diff --git a/qemu-progress.c b/qemu-progress.c
index 8ebe8ef..5f1b8df 100644
--- a/qemu-progress.c
+++ b/qemu-progress.c
@@ -26,7 +26,6 @@
 #include "osdep.h"
 #include "sysemu.h"
 #include <stdio.h>
-#include <signal.h>
 
 struct progress_state {
     float current;
diff --git a/qemu-timer.c b/qemu-timer.c
index 4141b6e..72066c7 100644
--- a/qemu-timer.c
+++ b/qemu-timer.c
@@ -39,15 +39,6 @@
 #include <sys/param.h>
 #endif
 
-#ifdef __linux__
-#include <sys/ioctl.h>
-#include <linux/rtc.h>
-/* For the benefit of older linux systems which don't supply it,
-   we use a local copy of hpet.h. */
-/* #include <linux/hpet.h> */
-#include "hpet.h"
-#endif
-
 #ifdef _WIN32
 #include <windows.h>
 #include <mmsystem.h>
@@ -234,12 +225,6 @@
 static void dynticks_stop_timer(struct qemu_alarm_timer *t);
 static void dynticks_rearm_timer(struct qemu_alarm_timer *t);
 
-static int hpet_start_timer(struct qemu_alarm_timer *t);
-static void hpet_stop_timer(struct qemu_alarm_timer *t);
-
-static int rtc_start_timer(struct qemu_alarm_timer *t);
-static void rtc_stop_timer(struct qemu_alarm_timer *t);
-
 #endif /* __linux__ */
 
 #endif /* _WIN32 */
@@ -304,10 +289,6 @@
 #ifdef __linux__
     {"dynticks", dynticks_start_timer,
      dynticks_stop_timer, dynticks_rearm_timer},
-    /* HPET - if available - is preferred */
-    {"hpet", hpet_start_timer, hpet_stop_timer, NULL},
-    /* ...otherwise try RTC */
-    {"rtc", rtc_start_timer, rtc_stop_timer, NULL},
 #endif
     {"unix", unix_start_timer, unix_stop_timer, NULL},
 #else
@@ -822,107 +803,6 @@
 
 #if defined(__linux__)
 
-#define RTC_FREQ 1024
-
-static void enable_sigio_timer(int fd)
-{
-    struct sigaction act;
-
-    /* timer signal */
-    sigfillset(&act.sa_mask);
-    act.sa_flags = 0;
-    act.sa_handler = host_alarm_handler;
-
-    sigaction(SIGIO, &act, NULL);
-    fcntl_setfl(fd, O_ASYNC);
-    fcntl(fd, F_SETOWN, getpid());
-}
-
-static int hpet_start_timer(struct qemu_alarm_timer *t)
-{
-    struct hpet_info info;
-    int r, fd;
-
-    fd = qemu_open("/dev/hpet", O_RDONLY);
-    if (fd < 0)
-        return -1;
-
-    /* Set frequency */
-    r = ioctl(fd, HPET_IRQFREQ, RTC_FREQ);
-    if (r < 0) {
-        fprintf(stderr, "Could not configure '/dev/hpet' to have a 1024Hz timer. This is not a fatal\n"
-                "error, but for better emulation accuracy type:\n"
-                "'echo 1024 > /proc/sys/dev/hpet/max-user-freq' as root.\n");
-        goto fail;
-    }
-
-    /* Check capabilities */
-    r = ioctl(fd, HPET_INFO, &info);
-    if (r < 0)
-        goto fail;
-
-    /* Enable periodic mode */
-    r = ioctl(fd, HPET_EPI, 0);
-    if (info.hi_flags && (r < 0))
-        goto fail;
-
-    /* Enable interrupt */
-    r = ioctl(fd, HPET_IE_ON, 0);
-    if (r < 0)
-        goto fail;
-
-    enable_sigio_timer(fd);
-    t->fd = fd;
-
-    return 0;
-fail:
-    close(fd);
-    return -1;
-}
-
-static void hpet_stop_timer(struct qemu_alarm_timer *t)
-{
-    int fd = t->fd;
-
-    close(fd);
-}
-
-static int rtc_start_timer(struct qemu_alarm_timer *t)
-{
-    int rtc_fd;
-    unsigned long current_rtc_freq = 0;
-
-    TFR(rtc_fd = qemu_open("/dev/rtc", O_RDONLY));
-    if (rtc_fd < 0)
-        return -1;
-    ioctl(rtc_fd, RTC_IRQP_READ, &current_rtc_freq);
-    if (current_rtc_freq != RTC_FREQ &&
-        ioctl(rtc_fd, RTC_IRQP_SET, RTC_FREQ) < 0) {
-        fprintf(stderr, "Could not configure '/dev/rtc' to have a 1024 Hz timer. This is not a fatal\n"
-                "error, but for better emulation accuracy either use a 2.6 host Linux kernel or\n"
-                "type 'echo 1024 > /proc/sys/dev/rtc/max-user-freq' as root.\n");
-        goto fail;
-    }
-    if (ioctl(rtc_fd, RTC_PIE_ON, 0) < 0) {
-    fail:
-        close(rtc_fd);
-        return -1;
-    }
-
-    enable_sigio_timer(rtc_fd);
-
-    t->fd = rtc_fd;
-
-    return 0;
-}
-
-static void rtc_stop_timer(struct qemu_alarm_timer *t)
-{
-    int rtc_fd = t->fd;
-
-    close(rtc_fd);
-}
-
 static int dynticks_start_timer(struct qemu_alarm_timer *t)
 {
     struct sigevent ev;
diff --git a/qerror.c b/qerror.c
index 4855604..d7fcd93 100644
--- a/qerror.c
+++ b/qerror.c
@@ -201,6 +201,10 @@
         .desc      = "An undefined error has ocurred",
     },
     {
+        .error_fmt = QERR_UNSUPPORTED,
+        .desc      = "this feature or command is not currently supported",
+    },
+    {
         .error_fmt = QERR_UNKNOWN_BLOCK_FORMAT_FEATURE,
         .desc      = "'%(device)' uses a %(format) feature which is not "
                      "supported by this qemu version: %(feature)",
@@ -326,12 +330,14 @@
     return qerr;
 }
 
-static void parse_error(const QError *qerror, int c)
+static void parse_error(const QErrorStringTable *entry, int c)
 {
-    qerror_abort(qerror, "expected '%c' in '%s'", c, qerror->entry->desc);
+    fprintf(stderr, "expected '%c' in '%s'", c, entry->desc);
+    abort();
 }
 
-static const char *append_field(QString *outstr, const QError *qerror,
+static const char *append_field(QDict *error, QString *outstr,
+                                const QErrorStringTable *entry,
                                 const char *start)
 {
     QObject *obj;
@@ -340,23 +346,23 @@
     const char *end, *key;
 
     if (*start != '%')
-        parse_error(qerror, '%');
+        parse_error(entry, '%');
     start++;
     if (*start != '(')
-        parse_error(qerror, '(');
+        parse_error(entry, '(');
     start++;
 
     end = strchr(start, ')');
     if (!end)
-        parse_error(qerror, ')');
+        parse_error(entry, ')');
 
     key_qs = qstring_from_substr(start, 0, end - start - 1);
     key = qstring_get_str(key_qs);
 
-    qdict = qobject_to_qdict(qdict_get(qerror->error, "data"));
+    qdict = qobject_to_qdict(qdict_get(error, "data"));
     obj = qdict_get(qdict, key);
     if (!obj) {
-        qerror_abort(qerror, "key '%s' not found in QDict", key);
+        abort();
     }
 
     switch (qobject_type(obj)) {
@@ -367,41 +373,60 @@
             qstring_append_int(outstr, qdict_get_int(qdict, key));
             break;
         default:
-            qerror_abort(qerror, "invalid type '%c'", qobject_type(obj));
+            abort();
     }
 
     QDECREF(key_qs);
     return ++end;
 }
 
-/**
- * qerror_human(): Format QError data into human-readable string.
- *
- * Formats according to member 'desc' of the specified QError object.
- */
-QString *qerror_human(const QError *qerror)
+static QString *qerror_format_desc(QDict *error,
+                                   const QErrorStringTable *entry)
 {
-    const char *p;
     QString *qstring;
+    const char *p;
 
-    assert(qerror->entry != NULL);
+    assert(entry != NULL);
 
     qstring = qstring_new();
 
-    for (p = qerror->entry->desc; *p != '\0';) {
+    for (p = entry->desc; *p != '\0';) {
         if (*p != '%') {
             qstring_append_chr(qstring, *p++);
         } else if (*(p + 1) == '%') {
             qstring_append_chr(qstring, '%');
             p += 2;
         } else {
-            p = append_field(qstring, qerror, p);
+            p = append_field(error, qstring, entry, p);
         }
     }
 
     return qstring;
 }
 
+QString *qerror_format(const char *fmt, QDict *error)
+{
+    const QErrorStringTable *entry = NULL;
+    int i;
+
+    for (i = 0; qerror_table[i].error_fmt; i++) {
+        if (strcmp(qerror_table[i].error_fmt, fmt) == 0) {
+            entry = &qerror_table[i];
+            break;
+        }
+    }
+
+    return qerror_format_desc(error, entry);
+}
+
+/**
+ * qerror_human(): Format QError data into human-readable string.
+ */
+QString *qerror_human(const QError *qerror)
+{
+    return qerror_format_desc(qerror->error, qerror->entry);
+}
+
 /**
  * qerror_print(): Print QError data
  *
diff --git a/qerror.h b/qerror.h
index df61d2c..16c830d 100644
--- a/qerror.h
+++ b/qerror.h
@@ -39,6 +39,7 @@
 void qerror_print(QError *qerror);
 void qerror_report_internal(const char *file, int linenr, const char *func,
                             const char *fmt, ...) GCC_FMT_ATTR(4, 5);
+QString *qerror_format(const char *fmt, QDict *error);
 #define qerror_report(fmt, ...) \
     qerror_report_internal(__FILE__, __LINE__, __func__, fmt, ## __VA_ARGS__)
 QError *qobject_to_qerror(const QObject *obj);
@@ -120,6 +121,9 @@
 #define QERR_JSON_PARSING \
     "{ 'class': 'JSONParsing', 'data': {} }"
 
+#define QERR_JSON_PARSE_ERROR \
+    "{ 'class': 'JSONParseError', 'data': { 'message': %s } }"
+
 #define QERR_KVM_MISSING_CAP \
     "{ 'class': 'KVMMissingCap', 'data': { 'capability': %s, 'feature': %s } }"
 
@@ -165,6 +169,9 @@
 #define QERR_UNDEFINED_ERROR \
     "{ 'class': 'UndefinedError', 'data': {} }"
 
+#define QERR_UNSUPPORTED \
+    "{ 'class': 'Unsupported', 'data': {} }"
+
 #define QERR_UNKNOWN_BLOCK_FORMAT_FEATURE \
     "{ 'class': 'UnknownBlockFormatFeature', 'data': { 'device': %s, 'format': %s, 'feature': %s } }"
 
diff --git a/qmp-commands.hx b/qmp-commands.hx
index a9f109a..92c5c3a 100644
--- a/qmp-commands.hx
+++ b/qmp-commands.hx
@@ -430,6 +430,33 @@
 EQMP
 
     {
+        .name       = "inject-nmi",
+        .args_type  = "",
+        .params     = "",
+        .help       = "",
+        .user_print = monitor_user_noop,
+        .mhandler.cmd_new = do_inject_nmi,
+    },
+
+SQMP
+inject-nmi
+----------
+
+Inject an NMI on guest's CPUs.
+
+Arguments: None.
+
+Example:
+
+-> { "execute": "inject-nmi" }
+<- { "return": {} }
+
+Note: inject-nmi is only supported for x86 guest currently, it will
+      returns "Unsupported" error for non-x86 guest.
+
+EQMP
+
+    {
         .name       = "migrate",
         .args_type  = "detach:-d,blk:-b,inc:-i,uri:s",
         .params     = "[-d] [-b] [-i] uri",
diff --git a/savevm.c b/savevm.c
index f4ff1a1..939845c 100644
--- a/savevm.c
+++ b/savevm.c
@@ -23,7 +23,6 @@
  */
 #include <unistd.h>
 #include <fcntl.h>
-#include <signal.h>
 #include <time.h>
 #include <errno.h>
 #include <sys/time.h>
diff --git a/spice-qemu-char.c b/spice-qemu-char.c
index fa15a71..605c241 100644
--- a/spice-qemu-char.c
+++ b/spice-qemu-char.c
@@ -36,14 +36,13 @@
 
     while (len > 0) {
         last_out = MIN(len, VMC_MAX_HOST_WRITE);
-        qemu_chr_read(scd->chr, p, last_out);
-        if (last_out > 0) {
-            out += last_out;
-            len -= last_out;
-            p += last_out;
-        } else {
+        if (qemu_chr_can_read(scd->chr) < last_out) {
             break;
         }
+        qemu_chr_read(scd->chr, p, last_out);
+        out += last_out;
+        len -= last_out;
+        p += last_out;
     }
 
     dprintf(scd, 3, "%s: %lu/%zd\n", __func__, out, len + out);
diff --git a/target-alpha/cpu.h b/target-alpha/cpu.h
index 686fb4a..e98b325 100644
--- a/target-alpha/cpu.h
+++ b/target-alpha/cpu.h
@@ -192,171 +192,39 @@
 
 #define SWCR_MASK  (SWCR_TRAP_ENABLE_MASK | SWCR_MAP_MASK | SWCR_STATUS_MASK)
 
-/* Internal processor registers */
-/* XXX: TOFIX: most of those registers are implementation dependant */
-enum {
-#if defined(CONFIG_USER_ONLY)
-    IPR_EXC_ADDR,
-    IPR_EXC_SUM,
-    IPR_EXC_MASK,
-#else
-    /* Ebox IPRs */
-    IPR_CC           = 0xC0,            /* 21264 */
-    IPR_CC_CTL       = 0xC1,            /* 21264 */
-#define IPR_CC_CTL_ENA_SHIFT 32
-#define IPR_CC_CTL_COUNTER_MASK 0xfffffff0UL
-    IPR_VA           = 0xC2,            /* 21264 */
-    IPR_VA_CTL       = 0xC4,            /* 21264 */
-#define IPR_VA_CTL_VA_48_SHIFT 1
-#define IPR_VA_CTL_VPTB_SHIFT 30
-    IPR_VA_FORM      = 0xC3,            /* 21264 */
-    /* Ibox IPRs */
-    IPR_ITB_TAG      = 0x00,            /* 21264 */
-    IPR_ITB_PTE      = 0x01,            /* 21264 */
-    IPR_ITB_IAP      = 0x02,
-    IPR_ITB_IA       = 0x03,            /* 21264 */
-    IPR_ITB_IS       = 0x04,            /* 21264 */
-    IPR_PMPC         = 0x05,
-    IPR_EXC_ADDR     = 0x06,            /* 21264 */
-    IPR_IVA_FORM     = 0x07,            /* 21264 */
-    IPR_CM           = 0x09,            /* 21264 */
-#define IPR_CM_SHIFT 3
-#define IPR_CM_MASK (3ULL << IPR_CM_SHIFT)      /* 21264 */
-    IPR_IER          = 0x0A,            /* 21264 */
-#define IPR_IER_MASK 0x0000007fffffe000ULL
-    IPR_IER_CM       = 0x0B,            /* 21264: = CM | IER */
-    IPR_SIRR         = 0x0C,            /* 21264 */
-#define IPR_SIRR_SHIFT 14
-#define IPR_SIRR_MASK 0x7fff
-    IPR_ISUM         = 0x0D,            /* 21264 */
-    IPR_HW_INT_CLR   = 0x0E,            /* 21264 */
-    IPR_EXC_SUM      = 0x0F,
-    IPR_PAL_BASE     = 0x10,
-    IPR_I_CTL        = 0x11,
-#define IPR_I_CTL_CHIP_ID_SHIFT 24      /* 21264 */
-#define IPR_I_CTL_BIST_FAIL (1 << 23)   /* 21264 */
-#define IPR_I_CTL_IC_EN_SHIFT 2         /* 21264 */
-#define IPR_I_CTL_SDE1_SHIFT 7          /* 21264 */
-#define IPR_I_CTL_HWE_SHIFT 12          /* 21264 */
-#define IPR_I_CTL_VA_48_SHIFT 15        /* 21264 */
-#define IPR_I_CTL_SPE_SHIFT 3           /* 21264 */
-#define IPR_I_CTL_CALL_PAL_R23_SHIFT 20 /* 21264 */
-    IPR_I_STAT       = 0x16,            /* 21264 */
-    IPR_IC_FLUSH     = 0x13,            /* 21264 */
-    IPR_IC_FLUSH_ASM = 0x12,            /* 21264 */
-    IPR_CLR_MAP      = 0x15,
-    IPR_SLEEP        = 0x17,
-    IPR_PCTX         = 0x40,
-    IPR_PCTX_ASN       = 0x01,  /* field */
-#define IPR_PCTX_ASN_SHIFT 39
-    IPR_PCTX_ASTER     = 0x02,  /* field */
-#define IPR_PCTX_ASTER_SHIFT 5
-    IPR_PCTX_ASTRR     = 0x04,  /* field */
-#define IPR_PCTX_ASTRR_SHIFT 9
-    IPR_PCTX_PPCE      = 0x08,  /* field */
-#define IPR_PCTX_PPCE_SHIFT 1
-    IPR_PCTX_FPE       = 0x10,  /* field */
-#define IPR_PCTX_FPE_SHIFT 2
-    IPR_PCTX_ALL       = 0x5f,  /* all fields */
-    IPR_PCTR_CTL     = 0x14,            /* 21264 */
-    /* Mbox IPRs */
-    IPR_DTB_TAG0     = 0x20,            /* 21264 */
-    IPR_DTB_TAG1     = 0xA0,            /* 21264 */
-    IPR_DTB_PTE0     = 0x21,            /* 21264 */
-    IPR_DTB_PTE1     = 0xA1,            /* 21264 */
-    IPR_DTB_ALTMODE  = 0xA6,
-    IPR_DTB_ALTMODE0 = 0x26,            /* 21264 */
-#define IPR_DTB_ALTMODE_MASK 3
-    IPR_DTB_IAP      = 0xA2,
-    IPR_DTB_IA       = 0xA3,            /* 21264 */
-    IPR_DTB_IS0      = 0x24,
-    IPR_DTB_IS1      = 0xA4,
-    IPR_DTB_ASN0     = 0x25,            /* 21264 */
-    IPR_DTB_ASN1     = 0xA5,            /* 21264 */
-#define IPR_DTB_ASN_SHIFT 56
-    IPR_MM_STAT      = 0x27,            /* 21264 */
-    IPR_M_CTL        = 0x28,            /* 21264 */
-#define IPR_M_CTL_SPE_SHIFT 1
-#define IPR_M_CTL_SPE_MASK 7
-    IPR_DC_CTL       = 0x29,            /* 21264 */
-    IPR_DC_STAT      = 0x2A,            /* 21264 */
-    /* Cbox IPRs */
-    IPR_C_DATA       = 0x2B,
-    IPR_C_SHIFT      = 0x2C,
+/* MMU modes definitions */
 
-    IPR_ASN,
-    IPR_ASTEN,
-    IPR_ASTSR,
-    IPR_DATFX,
-    IPR_ESP,
-    IPR_FEN,
-    IPR_IPIR,
-    IPR_IPL,
-    IPR_KSP,
-    IPR_MCES,
-    IPR_PERFMON,
-    IPR_PCBB,
-    IPR_PRBR,
-    IPR_PTBR,
-    IPR_SCBB,
-    IPR_SISR,
-    IPR_SSP,
-    IPR_SYSPTBR,
-    IPR_TBCHK,
-    IPR_TBIA,
-    IPR_TBIAP,
-    IPR_TBIS,
-    IPR_TBISD,
-    IPR_TBISI,
-    IPR_USP,
-    IPR_VIRBND,
-    IPR_VPTB,
-    IPR_WHAMI,
-    IPR_ALT_MODE,
-#endif
-    IPR_LAST,
-};
+/* Alpha has 5 MMU modes: PALcode, kernel, executive, supervisor, and user.
+   The Unix PALcode only exposes the kernel and user modes; presumably
+   executive and supervisor are used by VMS.
+
+   PALcode itself uses physical mode for code and kernel mode for data;
+   there are PALmode instructions that can access data via physical mode
+   or via an os-installed "alternate mode", which is one of the 4 above.
+
+   QEMU does not currently properly distinguish between code/data when
+   looking up addresses.  To avoid having to address this issue, our
+   emulated PALcode will cheat and use the KSEG mapping for its code+data
+   rather than physical addresses.
+
+   Moreover, we're only emulating Unix PALcode, and not attempting VMS.
+
+   All of which allows us to drop all but kernel and user modes.
+   Elide the unused MMU modes to save space.  */
+
+#define NB_MMU_MODES 2
+
+#define MMU_MODE0_SUFFIX _kernel
+#define MMU_MODE1_SUFFIX _user
+#define MMU_KERNEL_IDX   0
+#define MMU_USER_IDX     1
 
 typedef struct CPUAlphaState CPUAlphaState;
 
-typedef struct pal_handler_t pal_handler_t;
-struct pal_handler_t {
-    /* Reset */
-    void (*reset)(CPUAlphaState *env);
-    /* Uncorrectable hardware error */
-    void (*machine_check)(CPUAlphaState *env);
-    /* Arithmetic exception */
-    void (*arithmetic)(CPUAlphaState *env);
-    /* Interrupt / correctable hardware error */
-    void (*interrupt)(CPUAlphaState *env);
-    /* Data fault */
-    void (*dfault)(CPUAlphaState *env);
-    /* DTB miss pal */
-    void (*dtb_miss_pal)(CPUAlphaState *env);
-    /* DTB miss native */
-    void (*dtb_miss_native)(CPUAlphaState *env);
-    /* Unaligned access */
-    void (*unalign)(CPUAlphaState *env);
-    /* ITB miss */
-    void (*itb_miss)(CPUAlphaState *env);
-    /* Instruction stream access violation */
-    void (*itb_acv)(CPUAlphaState *env);
-    /* Reserved or privileged opcode */
-    void (*opcdec)(CPUAlphaState *env);
-    /* Floating point exception */
-    void (*fen)(CPUAlphaState *env);
-    /* Call pal instruction */
-    void (*call_pal)(CPUAlphaState *env, uint32_t palcode);
-};
-
-#define NB_MMU_MODES 4
-
 struct CPUAlphaState {
     uint64_t ir[31];
     float64 fir[31];
     uint64_t pc;
-    uint64_t ipr[IPR_LAST];
-    uint64_t ps;
     uint64_t unique;
     uint64_t lock_addr;
     uint64_t lock_st_addr;
@@ -371,10 +239,33 @@
     uint8_t fpcr_dnod;
     uint8_t fpcr_undz;
 
-    /* Used for HW_LD / HW_ST */
-    uint8_t saved_mode;
-    /* For RC and RS */
+    /* The Internal Processor Registers.  Some of these we assume always
+       exist for use in user-mode.  */
+    uint8_t ps;
     uint8_t intr_flag;
+    uint8_t pal_mode;
+    uint8_t fen;
+
+    uint32_t pcc_ofs;
+
+    /* These pass data from the exception logic in the translator and
+       helpers to the OS entry point.  This is used for both system
+       emulation and user-mode.  */
+    uint64_t trap_arg0;
+    uint64_t trap_arg1;
+    uint64_t trap_arg2;
+
+#if !defined(CONFIG_USER_ONLY)
+    /* The internal data required by our emulation of the Unix PALcode.  */
+    uint64_t exc_addr;
+    uint64_t palbr;
+    uint64_t ptbr;
+    uint64_t vptptr;
+    uint64_t sysval;
+    uint64_t usp;
+    uint64_t shadow[8];
+    uint64_t scratch[24];
+#endif
 
 #if TARGET_LONG_BITS > HOST_LONG_BITS
     /* temporary fixed-point registers
@@ -386,14 +277,11 @@
     /* Those resources are used only in Qemu core */
     CPU_COMMON
 
-    uint32_t hflags;
-
     int error_code;
 
     uint32_t features;
     uint32_t amask;
     int implver;
-    pal_handler_t *pal_handler;
 };
 
 #define cpu_init cpu_alpha_init
@@ -401,17 +289,6 @@
 #define cpu_gen_code cpu_alpha_gen_code
 #define cpu_signal_handler cpu_alpha_signal_handler
 
-/* MMU modes definitions */
-#define MMU_MODE0_SUFFIX _kernel
-#define MMU_MODE1_SUFFIX _executive
-#define MMU_MODE2_SUFFIX _supervisor
-#define MMU_MODE3_SUFFIX _user
-#define MMU_USER_IDX 3
-static inline int cpu_mmu_index (CPUState *env)
-{
-    return (env->ps >> 3) & 3;
-}
-
 #include "cpu-all.h"
 
 enum {
@@ -422,36 +299,89 @@
 };
 
 enum {
-    EXCP_RESET            = 0x0000,
-    EXCP_MCHK             = 0x0020,
-    EXCP_ARITH            = 0x0060,
-    EXCP_HW_INTERRUPT     = 0x00E0,
-    EXCP_DFAULT           = 0x01E0,
-    EXCP_DTB_MISS_PAL     = 0x09E0,
-    EXCP_ITB_MISS         = 0x03E0,
-    EXCP_ITB_ACV          = 0x07E0,
-    EXCP_DTB_MISS_NATIVE  = 0x08E0,
-    EXCP_UNALIGN          = 0x11E0,
-    EXCP_OPCDEC           = 0x13E0,
-    EXCP_FEN              = 0x17E0,
-    EXCP_CALL_PAL         = 0x2000,
-    EXCP_CALL_PALP        = 0x3000,
-    EXCP_CALL_PALE        = 0x4000,
-    /* Pseudo exception for console */
-    EXCP_CONSOLE_DISPATCH = 0x4001,
-    EXCP_CONSOLE_FIXUP    = 0x4002,
-    EXCP_STL_C            = 0x4003,
-    EXCP_STQ_C            = 0x4004,
+    EXCP_RESET,
+    EXCP_MCHK,
+    EXCP_SMP_INTERRUPT,
+    EXCP_CLK_INTERRUPT,
+    EXCP_DEV_INTERRUPT,
+    EXCP_MMFAULT,
+    EXCP_UNALIGN,
+    EXCP_OPCDEC,
+    EXCP_ARITH,
+    EXCP_FEN,
+    EXCP_CALL_PAL,
+    /* For Usermode emulation.  */
+    EXCP_STL_C,
+    EXCP_STQ_C,
 };
 
-/* Arithmetic exception */
-#define EXC_M_IOV       (1<<16)         /* Integer Overflow */
-#define EXC_M_INE       (1<<15)         /* Inexact result */
-#define EXC_M_UNF       (1<<14)         /* Underflow */
-#define EXC_M_FOV       (1<<13)         /* Overflow */
-#define EXC_M_DZE       (1<<12)         /* Division by zero */
-#define EXC_M_INV       (1<<11)         /* Invalid operation */
-#define EXC_M_SWC       (1<<10)         /* Software completion */
+/* Alpha-specific interrupt pending bits.  */
+#define CPU_INTERRUPT_TIMER	CPU_INTERRUPT_TGT_EXT_0
+#define CPU_INTERRUPT_SMP	CPU_INTERRUPT_TGT_EXT_1
+#define CPU_INTERRUPT_MCHK	CPU_INTERRUPT_TGT_EXT_2
+
+/* OSF/1 Page table bits.  */
+enum {
+    PTE_VALID = 0x0001,
+    PTE_FOR   = 0x0002,  /* used for page protection (fault on read) */
+    PTE_FOW   = 0x0004,  /* used for page protection (fault on write) */
+    PTE_FOE   = 0x0008,  /* used for page protection (fault on exec) */
+    PTE_ASM   = 0x0010,
+    PTE_KRE   = 0x0100,
+    PTE_URE   = 0x0200,
+    PTE_KWE   = 0x1000,
+    PTE_UWE   = 0x2000
+};
+
+/* Hardware interrupt (entInt) constants.  */
+enum {
+    INT_K_IP,
+    INT_K_CLK,
+    INT_K_MCHK,
+    INT_K_DEV,
+    INT_K_PERF,
+};
+
+/* Memory management (entMM) constants.  */
+enum {
+    MM_K_TNV,
+    MM_K_ACV,
+    MM_K_FOR,
+    MM_K_FOE,
+    MM_K_FOW
+};
+
+/* Arithmetic exception (entArith) constants.  */
+enum {
+    EXC_M_SWC = 1,      /* Software completion */
+    EXC_M_INV = 2,      /* Invalid operation */
+    EXC_M_DZE = 4,      /* Division by zero */
+    EXC_M_FOV = 8,      /* Overflow */
+    EXC_M_UNF = 16,     /* Underflow */
+    EXC_M_INE = 32,     /* Inexact result */
+    EXC_M_IOV = 64      /* Integer Overflow */
+};
+
+/* Processor status constants.  */
+enum {
+    /* Low 3 bits are interrupt mask level.  */
+    PS_INT_MASK = 7,
+
+    /* Bits 4 and 5 are the mmu mode.  The VMS PALcode uses all 4 modes;
+       The Unix PALcode only uses bit 4.  */
+    PS_USER_MODE = 8
+};
+
+static inline int cpu_mmu_index(CPUState *env)
+{
+    if (env->pal_mode) {
+        return MMU_KERNEL_IDX;
+    } else if (env->ps & PS_USER_MODE) {
+        return MMU_USER_IDX;
+    } else {
+        return MMU_KERNEL_IDX;
+    }
+}
 
 enum {
     IR_V0   = 0,
@@ -504,19 +434,46 @@
 
 uint64_t cpu_alpha_load_fpcr (CPUState *env);
 void cpu_alpha_store_fpcr (CPUState *env, uint64_t val);
-int cpu_alpha_mfpr (CPUState *env, int iprn, uint64_t *valp);
-int cpu_alpha_mtpr (CPUState *env, int iprn, uint64_t val, uint64_t *oldvalp);
-#if !defined (CONFIG_USER_ONLY)
-void pal_init (CPUState *env);
-void call_pal (CPUState *env);
+#ifndef CONFIG_USER_ONLY
+void swap_shadow_regs(CPUState *env);
+extern QEMU_NORETURN void do_unassigned_access(target_phys_addr_t addr,
+                                               int, int, int, int);
 #endif
 
+/* Bits in TB->FLAGS that control how translation is processed.  */
+enum {
+    TB_FLAGS_PAL_MODE = 1,
+    TB_FLAGS_FEN = 2,
+    TB_FLAGS_USER_MODE = 8,
+
+    TB_FLAGS_AMASK_SHIFT = 4,
+    TB_FLAGS_AMASK_BWX = AMASK_BWX << TB_FLAGS_AMASK_SHIFT,
+    TB_FLAGS_AMASK_FIX = AMASK_FIX << TB_FLAGS_AMASK_SHIFT,
+    TB_FLAGS_AMASK_CIX = AMASK_CIX << TB_FLAGS_AMASK_SHIFT,
+    TB_FLAGS_AMASK_MVI = AMASK_MVI << TB_FLAGS_AMASK_SHIFT,
+    TB_FLAGS_AMASK_TRAP = AMASK_TRAP << TB_FLAGS_AMASK_SHIFT,
+    TB_FLAGS_AMASK_PREFETCH = AMASK_PREFETCH << TB_FLAGS_AMASK_SHIFT,
+};
+
 static inline void cpu_get_tb_cpu_state(CPUState *env, target_ulong *pc,
-                                        target_ulong *cs_base, int *flags)
+                                        target_ulong *cs_base, int *pflags)
 {
+    int flags = 0;
+
     *pc = env->pc;
     *cs_base = 0;
-    *flags = env->ps;
+
+    if (env->pal_mode) {
+        flags = TB_FLAGS_PAL_MODE;
+    } else {
+        flags = env->ps & PS_USER_MODE;
+    }
+    if (env->fen) {
+        flags |= TB_FLAGS_FEN;
+    }
+    flags |= env->amask << TB_FLAGS_AMASK_SHIFT;
+
+    *pflags = flags;
 }
 
 #if defined(CONFIG_USER_ONLY)
diff --git a/target-alpha/exec.h b/target-alpha/exec.h
index 6ae96d1..7a325e7 100644
--- a/target-alpha/exec.h
+++ b/target-alpha/exec.h
@@ -39,7 +39,17 @@
 
 static inline int cpu_has_work(CPUState *env)
 {
-    return (env->interrupt_request & CPU_INTERRUPT_HARD);
+    /* Here we are checking to see if the CPU should wake up from HALT.
+       We will have gotten into this state only for WTINT from PALmode.  */
+    /* ??? I'm not sure how the IPL state works with WTINT to keep a CPU
+       asleep even if (some) interrupts have been asserted.  For now, 
+       assume that if a CPU really wants to stay asleep, it will mask
+       interrupts at the chipset level, which will prevent these bits
+       from being set in the first place.  */
+    return env->interrupt_request & (CPU_INTERRUPT_HARD
+                                     | CPU_INTERRUPT_TIMER
+                                     | CPU_INTERRUPT_SMP
+                                     | CPU_INTERRUPT_MCHK);
 }
 
 static inline void cpu_pc_from_tb(CPUState *env, TranslationBlock *tb)
diff --git a/target-alpha/helper.c b/target-alpha/helper.c
index 3ba4478..32c2cf9 100644
--- a/target-alpha/helper.c
+++ b/target-alpha/helper.c
@@ -160,382 +160,299 @@
 }
 
 #if defined(CONFIG_USER_ONLY)
-
 int cpu_alpha_handle_mmu_fault (CPUState *env, target_ulong address, int rw,
                                 int mmu_idx, int is_softmmu)
 {
-    if (rw == 2)
-        env->exception_index = EXCP_ITB_MISS;
-    else
-        env->exception_index = EXCP_DFAULT;
-    env->ipr[IPR_EXC_ADDR] = address;
-
+    env->exception_index = EXCP_MMFAULT;
+    env->trap_arg0 = address;
     return 1;
 }
-
-void do_interrupt (CPUState *env)
-{
-    env->exception_index = -1;
-}
-
 #else
-
-target_phys_addr_t cpu_get_phys_page_debug (CPUState *env, target_ulong addr)
+void swap_shadow_regs(CPUState *env)
 {
-    return -1;
+    uint64_t i0, i1, i2, i3, i4, i5, i6, i7;
+
+    i0 = env->ir[8];
+    i1 = env->ir[9];
+    i2 = env->ir[10];
+    i3 = env->ir[11];
+    i4 = env->ir[12];
+    i5 = env->ir[13];
+    i6 = env->ir[14];
+    i7 = env->ir[25];
+
+    env->ir[8]  = env->shadow[0];
+    env->ir[9]  = env->shadow[1];
+    env->ir[10] = env->shadow[2];
+    env->ir[11] = env->shadow[3];
+    env->ir[12] = env->shadow[4];
+    env->ir[13] = env->shadow[5];
+    env->ir[14] = env->shadow[6];
+    env->ir[25] = env->shadow[7];
+
+    env->shadow[0] = i0;
+    env->shadow[1] = i1;
+    env->shadow[2] = i2;
+    env->shadow[3] = i3;
+    env->shadow[4] = i4;
+    env->shadow[5] = i5;
+    env->shadow[6] = i6;
+    env->shadow[7] = i7;
 }
 
-int cpu_alpha_handle_mmu_fault (CPUState *env, target_ulong address, int rw,
-                                int mmu_idx, int is_softmmu)
+/* Returns the OSF/1 entMM failure indication, or -1 on success.  */
+static int get_physical_address(CPUState *env, target_ulong addr,
+                                int prot_need, int mmu_idx,
+                                target_ulong *pphys, int *pprot)
 {
-    uint32_t opc;
+    target_long saddr = addr;
+    target_ulong phys = 0;
+    target_ulong L1pte, L2pte, L3pte;
+    target_ulong pt, index;
+    int prot = 0;
+    int ret = MM_K_ACV;
 
-    if (rw == 2) {
-        /* Instruction translation buffer miss */
-        env->exception_index = EXCP_ITB_MISS;
-    } else {
-        if (env->ipr[IPR_EXC_ADDR] & 1)
-            env->exception_index = EXCP_DTB_MISS_PAL;
-        else
-            env->exception_index = EXCP_DTB_MISS_NATIVE;
-        opc = (ldl_code(env->pc) >> 21) << 4;
-        if (rw) {
-            opc |= 0x9;
-        } else {
-            opc |= 0x4;
-        }
-        env->ipr[IPR_MM_STAT] = opc;
+    /* Ensure that the virtual address is properly sign-extended from
+       the last implemented virtual address bit.  */
+    if (saddr >> TARGET_VIRT_ADDR_SPACE_BITS != saddr >> 63) {
+        goto exit;
     }
 
-    return 1;
-}
-
-int cpu_alpha_mfpr (CPUState *env, int iprn, uint64_t *valp)
-{
-    uint64_t hwpcb;
-    int ret = 0;
-
-    hwpcb = env->ipr[IPR_PCBB];
-    switch (iprn) {
-    case IPR_ASN:
-        if (env->features & FEATURE_ASN)
-            *valp = env->ipr[IPR_ASN];
-        else
-            *valp = 0;
-        break;
-    case IPR_ASTEN:
-        *valp = ((int64_t)(env->ipr[IPR_ASTEN] << 60)) >> 60;
-        break;
-    case IPR_ASTSR:
-        *valp = ((int64_t)(env->ipr[IPR_ASTSR] << 60)) >> 60;
-        break;
-    case IPR_DATFX:
-        /* Write only */
-        ret = -1;
-        break;
-    case IPR_ESP:
-        if (env->features & FEATURE_SPS)
-            *valp = env->ipr[IPR_ESP];
-        else
-            *valp = ldq_raw(hwpcb + 8);
-        break;
-    case IPR_FEN:
-        *valp = ((int64_t)(env->ipr[IPR_FEN] << 63)) >> 63;
-        break;
-    case IPR_IPIR:
-        /* Write-only */
-        ret = -1;
-        break;
-    case IPR_IPL:
-        *valp = ((int64_t)(env->ipr[IPR_IPL] << 59)) >> 59;
-        break;
-    case IPR_KSP:
-        if (!(env->ipr[IPR_EXC_ADDR] & 1)) {
-            ret = -1;
-        } else {
-            if (env->features & FEATURE_SPS)
-                *valp = env->ipr[IPR_KSP];
-            else
-                *valp = ldq_raw(hwpcb + 0);
+    /* Translate the superpage.  */
+    /* ??? When we do more than emulate Unix PALcode, we'll need to
+       determine which KSEG is actually active.  */
+    if (saddr < 0 && ((saddr >> 41) & 3) == 2) {
+        /* User-space cannot access KSEG addresses.  */
+        if (mmu_idx != MMU_KERNEL_IDX) {
+            goto exit;
         }
-        break;
-    case IPR_MCES:
-        *valp = ((int64_t)(env->ipr[IPR_MCES] << 59)) >> 59;
-        break;
-    case IPR_PERFMON:
-        /* Implementation specific */
-        *valp = 0;
-        break;
-    case IPR_PCBB:
-        *valp = ((int64_t)env->ipr[IPR_PCBB] << 16) >> 16;
-        break;
-    case IPR_PRBR:
-        *valp = env->ipr[IPR_PRBR];
-        break;
-    case IPR_PTBR:
-        *valp = env->ipr[IPR_PTBR];
-        break;
-    case IPR_SCBB:
-        *valp = (int64_t)((int32_t)env->ipr[IPR_SCBB]);
-        break;
-    case IPR_SIRR:
-        /* Write-only */
+
+        /* For the benefit of the Typhoon chipset, move bit 40 to bit 43.
+           We would not do this if the 48-bit KSEG is enabled.  */
+        phys = saddr & ((1ull << 40) - 1);
+        phys |= (saddr & (1ull << 40)) << 3;
+
+        prot = PAGE_READ | PAGE_WRITE | PAGE_EXEC;
         ret = -1;
-        break;
-    case IPR_SISR:
-        *valp = (int64_t)((int16_t)env->ipr[IPR_SISR]);
-    case IPR_SSP:
-        if (env->features & FEATURE_SPS)
-            *valp = env->ipr[IPR_SSP];
-        else
-            *valp = ldq_raw(hwpcb + 16);
-        break;
-    case IPR_SYSPTBR:
-        if (env->features & FEATURE_VIRBND)
-            *valp = env->ipr[IPR_SYSPTBR];
-        else
-            ret = -1;
-        break;
-    case IPR_TBCHK:
-        if ((env->features & FEATURE_TBCHK)) {
-            /* XXX: TODO */
-            *valp = 0;
-            ret = -1;
-        } else {
-            ret = -1;
-        }
-        break;
-    case IPR_TBIA:
-        /* Write-only */
-        ret = -1;
-        break;
-    case IPR_TBIAP:
-        /* Write-only */
-        ret = -1;
-        break;
-    case IPR_TBIS:
-        /* Write-only */
-        ret = -1;
-        break;
-    case IPR_TBISD:
-        /* Write-only */
-        ret = -1;
-        break;
-    case IPR_TBISI:
-        /* Write-only */
-        ret = -1;
-        break;
-    case IPR_USP:
-        if (env->features & FEATURE_SPS)
-            *valp = env->ipr[IPR_USP];
-        else
-            *valp = ldq_raw(hwpcb + 24);
-        break;
-    case IPR_VIRBND:
-        if (env->features & FEATURE_VIRBND)
-            *valp = env->ipr[IPR_VIRBND];
-        else
-            ret = -1;
-        break;
-    case IPR_VPTB:
-        *valp = env->ipr[IPR_VPTB];
-        break;
-    case IPR_WHAMI:
-        *valp = env->ipr[IPR_WHAMI];
-        break;
-    default:
-        /* Invalid */
-        ret = -1;
-        break;
+        goto exit;
     }
 
+    /* Interpret the page table exactly like PALcode does.  */
+
+    pt = env->ptbr;
+
+    /* L1 page table read.  */
+    index = (addr >> (TARGET_PAGE_BITS + 20)) & 0x3ff;
+    L1pte = ldq_phys(pt + index*8);
+
+    if (unlikely((L1pte & PTE_VALID) == 0)) {
+        ret = MM_K_TNV;
+        goto exit;
+    }
+    if (unlikely((L1pte & PTE_KRE) == 0)) {
+        goto exit;
+    }
+    pt = L1pte >> 32 << TARGET_PAGE_BITS;
+
+    /* L2 page table read.  */
+    index = (addr >> (TARGET_PAGE_BITS + 10)) & 0x3ff;
+    L2pte = ldq_phys(pt + index*8);
+
+    if (unlikely((L2pte & PTE_VALID) == 0)) {
+        ret = MM_K_TNV;
+        goto exit;
+    }
+    if (unlikely((L2pte & PTE_KRE) == 0)) {
+        goto exit;
+    }
+    pt = L2pte >> 32 << TARGET_PAGE_BITS;
+
+    /* L3 page table read.  */
+    index = (addr >> TARGET_PAGE_BITS) & 0x3ff;
+    L3pte = ldq_phys(pt + index*8);
+
+    phys = L3pte >> 32 << TARGET_PAGE_BITS;
+    if (unlikely((L3pte & PTE_VALID) == 0)) {
+        ret = MM_K_TNV;
+        goto exit;
+    }
+
+#if PAGE_READ != 1 || PAGE_WRITE != 2 || PAGE_EXEC != 4
+# error page bits out of date
+#endif
+
+    /* Check access violations.  */
+    if (L3pte & (PTE_KRE << mmu_idx)) {
+        prot |= PAGE_READ | PAGE_EXEC;
+    }
+    if (L3pte & (PTE_KWE << mmu_idx)) {
+        prot |= PAGE_WRITE;
+    }
+    if (unlikely((prot & prot_need) == 0 && prot_need)) {
+        goto exit;
+    }
+
+    /* Check fault-on-operation violations.  */
+    prot &= ~(L3pte >> 1);
+    ret = -1;
+    if (unlikely((prot & prot_need) == 0)) {
+        ret = (prot_need & PAGE_EXEC ? MM_K_FOE :
+               prot_need & PAGE_WRITE ? MM_K_FOW :
+               prot_need & PAGE_READ ? MM_K_FOR : -1);
+    }
+
+ exit:
+    *pphys = phys;
+    *pprot = prot;
     return ret;
 }
 
-int cpu_alpha_mtpr (CPUState *env, int iprn, uint64_t val, uint64_t *oldvalp)
+target_phys_addr_t cpu_get_phys_page_debug(CPUState *env, target_ulong addr)
 {
-    uint64_t hwpcb, tmp64;
-    uint8_t tmp8;
-    int ret = 0;
+    target_ulong phys;
+    int prot, fail;
 
-    hwpcb = env->ipr[IPR_PCBB];
-    switch (iprn) {
-    case IPR_ASN:
-        /* Read-only */
-        ret = -1;
-        break;
-    case IPR_ASTEN:
-        tmp8 = ((int8_t)(env->ipr[IPR_ASTEN] << 4)) >> 4;
-        *oldvalp = tmp8;
-        tmp8 &= val & 0xF;
-        tmp8 |= (val >> 4) & 0xF;
-        env->ipr[IPR_ASTEN] &= ~0xF;
-        env->ipr[IPR_ASTEN] |= tmp8;
-        ret = 1;
-        break;
-    case IPR_ASTSR:
-        tmp8 = ((int8_t)(env->ipr[IPR_ASTSR] << 4)) >> 4;
-        *oldvalp = tmp8;
-        tmp8 &= val & 0xF;
-        tmp8 |= (val >> 4) & 0xF;
-        env->ipr[IPR_ASTSR] &= ~0xF;
-        env->ipr[IPR_ASTSR] |= tmp8;
-        ret = 1;
-    case IPR_DATFX:
-        env->ipr[IPR_DATFX] &= ~0x1;
-        env->ipr[IPR_DATFX] |= val & 1;
-        tmp64 = ldq_raw(hwpcb + 56);
-        tmp64 &= ~0x8000000000000000ULL;
-        tmp64 |= (val & 1) << 63;
-        stq_raw(hwpcb + 56, tmp64);
-        break;
-    case IPR_ESP:
-        if (env->features & FEATURE_SPS)
-            env->ipr[IPR_ESP] = val;
-        else
-            stq_raw(hwpcb + 8, val);
-        break;
-    case IPR_FEN:
-        env->ipr[IPR_FEN] = val & 1;
-        tmp64 = ldq_raw(hwpcb + 56);
-        tmp64 &= ~1;
-        tmp64 |= val & 1;
-        stq_raw(hwpcb + 56, tmp64);
-        break;
-    case IPR_IPIR:
-        /* XXX: TODO: Send IRQ to CPU #ir[16] */
-        break;
-    case IPR_IPL:
-        *oldvalp = ((int64_t)(env->ipr[IPR_IPL] << 59)) >> 59;
-        env->ipr[IPR_IPL] &= ~0x1F;
-        env->ipr[IPR_IPL] |= val & 0x1F;
-        /* XXX: may issue an interrupt or ASR _now_ */
-        ret = 1;
-        break;
-    case IPR_KSP:
-        if (!(env->ipr[IPR_EXC_ADDR] & 1)) {
-            ret = -1;
-        } else {
-            if (env->features & FEATURE_SPS)
-                env->ipr[IPR_KSP] = val;
-            else
-                stq_raw(hwpcb + 0, val);
-        }
-        break;
-    case IPR_MCES:
-        env->ipr[IPR_MCES] &= ~((val & 0x7) | 0x18);
-        env->ipr[IPR_MCES] |= val & 0x18;
-        break;
-    case IPR_PERFMON:
-        /* Implementation specific */
-        *oldvalp = 0;
-        ret = 1;
-        break;
-    case IPR_PCBB:
-        /* Read-only */
-        ret = -1;
-        break;
-    case IPR_PRBR:
-        env->ipr[IPR_PRBR] = val;
-        break;
-    case IPR_PTBR:
-        /* Read-only */
-        ret = -1;
-        break;
-    case IPR_SCBB:
-        env->ipr[IPR_SCBB] = (uint32_t)val;
-        break;
-    case IPR_SIRR:
-        if (val & 0xF) {
-            env->ipr[IPR_SISR] |= 1 << (val & 0xF);
-            /* XXX: request a software interrupt _now_ */
-        }
-        break;
-    case IPR_SISR:
-        /* Read-only */
-        ret = -1;
-        break;
-    case IPR_SSP:
-        if (env->features & FEATURE_SPS)
-            env->ipr[IPR_SSP] = val;
-        else
-            stq_raw(hwpcb + 16, val);
-        break;
-    case IPR_SYSPTBR:
-        if (env->features & FEATURE_VIRBND)
-            env->ipr[IPR_SYSPTBR] = val;
-        else
-            ret = -1;
-        break;
-    case IPR_TBCHK:
-        /* Read-only */
-        ret = -1;
-        break;
-    case IPR_TBIA:
-        tlb_flush(env, 1);
-        break;
-    case IPR_TBIAP:
-        tlb_flush(env, 1);
-        break;
-    case IPR_TBIS:
-        tlb_flush_page(env, val);
-        break;
-    case IPR_TBISD:
-        tlb_flush_page(env, val);
-        break;
-    case IPR_TBISI:
-        tlb_flush_page(env, val);
-        break;
-    case IPR_USP:
-        if (env->features & FEATURE_SPS)
-            env->ipr[IPR_USP] = val;
-        else
-            stq_raw(hwpcb + 24, val);
-        break;
-    case IPR_VIRBND:
-        if (env->features & FEATURE_VIRBND)
-            env->ipr[IPR_VIRBND] = val;
-        else
-            ret = -1;
-        break;
-    case IPR_VPTB:
-        env->ipr[IPR_VPTB] = val;
-        break;
-    case IPR_WHAMI:
-        /* Read-only */
-        ret = -1;
-        break;
-    default:
-        /* Invalid */
-        ret = -1;
-        break;
+    fail = get_physical_address(env, addr, 0, 0, &phys, &prot);
+    return (fail >= 0 ? -1 : phys);
+}
+
+int cpu_alpha_handle_mmu_fault(CPUState *env, target_ulong addr, int rw,
+                               int mmu_idx, int is_softmmu)
+{
+    target_ulong phys;
+    int prot, fail;
+
+    fail = get_physical_address(env, addr, 1 << rw, mmu_idx, &phys, &prot);
+    if (unlikely(fail >= 0)) {
+        env->exception_index = EXCP_MMFAULT;
+        env->trap_arg0 = addr;
+        env->trap_arg1 = fail;
+        env->trap_arg2 = (rw == 2 ? -1 : rw);
+        return 1;
     }
 
-    return ret;
+    tlb_set_page(env, addr & TARGET_PAGE_MASK, phys & TARGET_PAGE_MASK,
+                 prot, mmu_idx, TARGET_PAGE_SIZE);
+    return 0;
 }
+#endif /* USER_ONLY */
 
 void do_interrupt (CPUState *env)
 {
-    int excp;
+    int i = env->exception_index;
 
-    env->ipr[IPR_EXC_ADDR] = env->pc | 1;
-    excp = env->exception_index;
-    env->exception_index = -1;
-    env->error_code = 0;
-    /* XXX: disable interrupts and memory mapping */
-    if (env->ipr[IPR_PAL_BASE] != -1ULL) {
-        /* We use native PALcode */
-        env->pc = env->ipr[IPR_PAL_BASE] + excp;
-    } else {
-        /* We use emulated PALcode */
-        call_pal(env);
-        /* Emulate REI */
-        env->pc = env->ipr[IPR_EXC_ADDR] & ~7;
-        env->ipr[IPR_EXC_ADDR] = env->ipr[IPR_EXC_ADDR] & 1;
-        /* XXX: re-enable interrupts and memory mapping */
+    if (qemu_loglevel_mask(CPU_LOG_INT)) {
+        static int count;
+        const char *name = "<unknown>";
+
+        switch (i) {
+        case EXCP_RESET:
+            name = "reset";
+            break;
+        case EXCP_MCHK:
+            name = "mchk";
+            break;
+        case EXCP_SMP_INTERRUPT:
+            name = "smp_interrupt";
+            break;
+        case EXCP_CLK_INTERRUPT:
+            name = "clk_interrupt";
+            break;
+        case EXCP_DEV_INTERRUPT:
+            name = "dev_interrupt";
+            break;
+        case EXCP_MMFAULT:
+            name = "mmfault";
+            break;
+        case EXCP_UNALIGN:
+            name = "unalign";
+            break;
+        case EXCP_OPCDEC:
+            name = "opcdec";
+            break;
+        case EXCP_ARITH:
+            name = "arith";
+            break;
+        case EXCP_FEN:
+            name = "fen";
+            break;
+        case EXCP_CALL_PAL:
+            name = "call_pal";
+            break;
+        case EXCP_STL_C:
+            name = "stl_c";
+            break;
+        case EXCP_STQ_C:
+            name = "stq_c";
+            break;
+        }
+        qemu_log("INT %6d: %s(%#x) pc=%016" PRIx64 " sp=%016" PRIx64 "\n",
+                 ++count, name, env->error_code, env->pc, env->ir[IR_SP]);
     }
+
+    env->exception_index = -1;
+
+#if !defined(CONFIG_USER_ONLY)
+    switch (i) {
+    case EXCP_RESET:
+        i = 0x0000;
+        break;
+    case EXCP_MCHK:
+        i = 0x0080;
+        break;
+    case EXCP_SMP_INTERRUPT:
+        i = 0x0100;
+        break;
+    case EXCP_CLK_INTERRUPT:
+        i = 0x0180;
+        break;
+    case EXCP_DEV_INTERRUPT:
+        i = 0x0200;
+        break;
+    case EXCP_MMFAULT:
+        i = 0x0280;
+        break;
+    case EXCP_UNALIGN:
+        i = 0x0300;
+        break;
+    case EXCP_OPCDEC:
+        i = 0x0380;
+        break;
+    case EXCP_ARITH:
+        i = 0x0400;
+        break;
+    case EXCP_FEN:
+        i = 0x0480;
+        break;
+    case EXCP_CALL_PAL:
+        i = env->error_code;
+        /* There are 64 entry points for both privileged and unprivileged,
+           with bit 0x80 indicating unprivileged.  Each entry point gets
+           64 bytes to do its job.  */
+        if (i & 0x80) {
+            i = 0x2000 + (i - 0x80) * 64;
+        } else {
+            i = 0x1000 + i * 64;
+        }
+        break;
+    default:
+        cpu_abort(env, "Unhandled CPU exception");
+    }
+
+    /* Remember where the exception happened.  Emulate real hardware in
+       that the low bit of the PC indicates PALmode.  */
+    env->exc_addr = env->pc | env->pal_mode;
+
+    /* Continue execution at the PALcode entry point.  */
+    env->pc = env->palbr + i;
+
+    /* Switch to PALmode.  */
+    if (!env->pal_mode) {
+        env->pal_mode = 1;
+        swap_shadow_regs(env);
+    }
+#endif /* !USER_ONLY */
 }
-#endif
 
 void cpu_dump_state (CPUState *env, FILE *f, fprintf_function cpu_fprintf,
                      int flags)
@@ -548,7 +465,7 @@
     };
     int i;
 
-    cpu_fprintf(f, "     PC  " TARGET_FMT_lx "      PS  " TARGET_FMT_lx "\n",
+    cpu_fprintf(f, "     PC  " TARGET_FMT_lx "      PS  %02x\n",
                 env->pc, env->ps);
     for (i = 0; i < 31; i++) {
         cpu_fprintf(f, "IR%02d %s " TARGET_FMT_lx " ", i,
diff --git a/target-alpha/helper.h b/target-alpha/helper.h
index ccf6a2a..2dec57e 100644
--- a/target-alpha/helper.h
+++ b/target-alpha/helper.h
@@ -100,27 +100,19 @@
 DEF_HELPER_1(ieee_input_s, i64, i64)
 
 #if !defined (CONFIG_USER_ONLY)
-DEF_HELPER_0(hw_rei, void)
 DEF_HELPER_1(hw_ret, void, i64)
-DEF_HELPER_2(mfpr, i64, int, i64)
-DEF_HELPER_2(mtpr, void, int, i64)
-DEF_HELPER_0(set_alt_mode, void)
-DEF_HELPER_0(restore_mode, void)
 
-DEF_HELPER_1(ld_virt_to_phys, i64, i64)
-DEF_HELPER_1(st_virt_to_phys, i64, i64)
-DEF_HELPER_2(ldl_raw, void, i64, i64)
-DEF_HELPER_2(ldq_raw, void, i64, i64)
-DEF_HELPER_2(ldl_l_raw, void, i64, i64)
-DEF_HELPER_2(ldq_l_raw, void, i64, i64)
-DEF_HELPER_2(ldl_kernel, void, i64, i64)
-DEF_HELPER_2(ldq_kernel, void, i64, i64)
-DEF_HELPER_2(ldl_data, void, i64, i64)
-DEF_HELPER_2(ldq_data, void, i64, i64)
-DEF_HELPER_2(stl_raw, void, i64, i64)
-DEF_HELPER_2(stq_raw, void, i64, i64)
-DEF_HELPER_2(stl_c_raw, i64, i64, i64)
-DEF_HELPER_2(stq_c_raw, i64, i64, i64)
+DEF_HELPER_1(ldl_phys, i64, i64)
+DEF_HELPER_1(ldq_phys, i64, i64)
+DEF_HELPER_1(ldl_l_phys, i64, i64)
+DEF_HELPER_1(ldq_l_phys, i64, i64)
+DEF_HELPER_2(stl_phys, void, i64, i64)
+DEF_HELPER_2(stq_phys, void, i64, i64)
+DEF_HELPER_2(stl_c_phys, i64, i64, i64)
+DEF_HELPER_2(stq_c_phys, i64, i64, i64)
+
+DEF_HELPER_FLAGS_0(tbia, TCG_CALL_CONST, void)
+DEF_HELPER_FLAGS_1(tbis, TCG_CALL_CONST, void, i64)
 #endif
 
 #include "def-helper.h"
diff --git a/target-alpha/machine.c b/target-alpha/machine.c
new file mode 100644
index 0000000..76d70d9
--- /dev/null
+++ b/target-alpha/machine.c
@@ -0,0 +1,87 @@
+#include "hw/hw.h"
+#include "hw/boards.h"
+
+static int get_fpcr(QEMUFile *f, void *opaque, size_t size)
+{
+    CPUAlphaState *env = opaque;
+    cpu_alpha_store_fpcr(env, qemu_get_be64(f));
+    return 0;
+}
+
+static void put_fpcr(QEMUFile *f, void *opaque, size_t size)
+{
+    CPUAlphaState *env = opaque;
+    qemu_put_be64(f, cpu_alpha_load_fpcr(env));
+}
+
+static const VMStateInfo vmstate_fpcr = {
+    .name = "fpcr",
+    .get = get_fpcr,
+    .put = put_fpcr,
+};
+
+static VMStateField vmstate_cpu_fields[] = {
+    VMSTATE_UINTTL_ARRAY(ir, CPUState, 31),
+    VMSTATE_UINTTL_ARRAY(fir, CPUState, 31),
+    /* Save the architecture value of the fpcr, not the internally
+       expanded version.  Since this architecture value does not
+       exist in memory to be stored, this requires a but of hoop
+       jumping.  We want OFFSET=0 so that we effectively pass ENV
+       to the helper functions, and we need to fill in the name by
+       hand since there's no field of that name.  */
+    {
+        .name = "fpcr",
+        .version_id = 0,
+        .size = sizeof(uint64_t),
+        .info = &vmstate_fpcr,
+        .flags = VMS_SINGLE,
+        .offset = 0
+    },
+    VMSTATE_UINTTL(pc, CPUState),
+    VMSTATE_UINTTL(unique, CPUState),
+    VMSTATE_UINTTL(lock_addr, CPUState),
+    VMSTATE_UINTTL(lock_value, CPUState),
+    /* Note that lock_st_addr is not saved; it is a temporary
+       used during the execution of the st[lq]_c insns.  */
+
+    VMSTATE_UINT8(ps, CPUState),
+    VMSTATE_UINT8(intr_flag, CPUState),
+    VMSTATE_UINT8(pal_mode, CPUState),
+    VMSTATE_UINT8(fen, CPUState),
+
+    VMSTATE_UINT32(pcc_ofs, CPUState),
+
+    VMSTATE_UINTTL(trap_arg0, CPUState),
+    VMSTATE_UINTTL(trap_arg1, CPUState),
+    VMSTATE_UINTTL(trap_arg2, CPUState),
+
+    VMSTATE_UINTTL(exc_addr, CPUState),
+    VMSTATE_UINTTL(palbr, CPUState),
+    VMSTATE_UINTTL(ptbr, CPUState),
+    VMSTATE_UINTTL(vptptr, CPUState),
+    VMSTATE_UINTTL(sysval, CPUState),
+    VMSTATE_UINTTL(usp, CPUState),
+
+    VMSTATE_UINTTL_ARRAY(shadow, CPUState, 8),
+    VMSTATE_UINTTL_ARRAY(scratch, CPUState, 24),
+
+    VMSTATE_END_OF_LIST()
+};
+
+static const VMStateDescription vmstate_cpu = {
+    .name = "cpu",
+    .version_id = 1,
+    .minimum_version_id = 1,
+    .minimum_version_id_old = 1,
+    .fields = vmstate_cpu_fields,
+};
+
+void cpu_save(QEMUFile *f, void *opaque)
+{
+    vmstate_save_state(f, &vmstate_cpu, opaque);
+}
+
+int cpu_load(QEMUFile *f, void *opaque, int version_id)
+{
+    return vmstate_load_state(f, &vmstate_cpu, opaque, version_id);
+}
diff --git a/target-alpha/op_helper.c b/target-alpha/op_helper.c
index 4ccb10b..d332719 100644
--- a/target-alpha/op_helper.c
+++ b/target-alpha/op_helper.c
@@ -25,17 +25,57 @@
 
 /*****************************************************************************/
 /* Exceptions processing helpers */
-void QEMU_NORETURN helper_excp (int excp, int error)
+
+/* This should only be called from translate, via gen_excp.
+   We expect that ENV->PC has already been updated.  */
+void QEMU_NORETURN helper_excp(int excp, int error)
 {
     env->exception_index = excp;
     env->error_code = error;
     cpu_loop_exit();
 }
 
+static void do_restore_state(void *retaddr)
+{
+    unsigned long pc = (unsigned long)retaddr;
+
+    if (pc) {
+        TranslationBlock *tb = tb_find_pc(pc);
+        if (tb) {
+            cpu_restore_state(tb, env, pc);
+        }
+    }
+}
+
+/* This may be called from any of the helpers to set up EXCEPTION_INDEX.  */
+static void QEMU_NORETURN dynamic_excp(int excp, int error)
+{
+    env->exception_index = excp;
+    env->error_code = error;
+    do_restore_state(GETPC());
+    cpu_loop_exit();
+}
+
+static void QEMU_NORETURN arith_excp(int exc, uint64_t mask)
+{
+    env->trap_arg0 = exc;
+    env->trap_arg1 = mask;
+    dynamic_excp(EXCP_ARITH, 0);
+}
+
 uint64_t helper_load_pcc (void)
 {
-    /* ??? This isn't a timer for which we have any rate info.  */
+#ifndef CONFIG_USER_ONLY
+    /* In system mode we have access to a decent high-resolution clock.
+       In order to make OS-level time accounting work with the RPCC,
+       present it with a well-timed clock fixed at 250MHz.  */
+    return (((uint64_t)env->pcc_ofs << 32)
+            | (uint32_t)(qemu_get_clock_ns(vm_clock) >> 2));
+#else
+    /* In user-mode, vm_clock doesn't exist.  Just pass through the host cpu
+       clock ticks.  Also, don't bother taking PCC_OFS into account.  */
     return (uint32_t)cpu_get_real_ticks();
+#endif
 }
 
 uint64_t helper_load_fpcr (void)
@@ -53,7 +93,7 @@
     uint64_t tmp = op1;
     op1 += op2;
     if (unlikely((tmp ^ op2 ^ (-1ULL)) & (tmp ^ op1) & (1ULL << 63))) {
-        helper_excp(EXCP_ARITH, EXC_M_IOV);
+        arith_excp(EXC_M_IOV, 0);
     }
     return op1;
 }
@@ -63,7 +103,7 @@
     uint64_t tmp = op1;
     op1 = (uint32_t)(op1 + op2);
     if (unlikely((tmp ^ op2 ^ (-1UL)) & (tmp ^ op1) & (1UL << 31))) {
-        helper_excp(EXCP_ARITH, EXC_M_IOV);
+        arith_excp(EXC_M_IOV, 0);
     }
     return op1;
 }
@@ -73,7 +113,7 @@
     uint64_t res;
     res = op1 - op2;
     if (unlikely((op1 ^ op2) & (res ^ op1) & (1ULL << 63))) {
-        helper_excp(EXCP_ARITH, EXC_M_IOV);
+        arith_excp(EXC_M_IOV, 0);
     }
     return res;
 }
@@ -83,7 +123,7 @@
     uint32_t res;
     res = op1 - op2;
     if (unlikely((op1 ^ op2) & (res ^ op1) & (1UL << 31))) {
-        helper_excp(EXCP_ARITH, EXC_M_IOV);
+        arith_excp(EXC_M_IOV, 0);
     }
     return res;
 }
@@ -93,7 +133,7 @@
     int64_t res = (int64_t)op1 * (int64_t)op2;
 
     if (unlikely((int32_t)res != res)) {
-        helper_excp(EXCP_ARITH, EXC_M_IOV);
+        arith_excp(EXC_M_IOV, 0);
     }
     return (int64_t)((int32_t)res);
 }
@@ -105,7 +145,7 @@
     muls64(&tl, &th, op1, op2);
     /* If th != 0 && th != -1, then we had an overflow */
     if (unlikely((th + 1) > 1)) {
-        helper_excp(EXCP_ARITH, EXC_M_IOV);
+        arith_excp(EXC_M_IOV, 0);
     }
     return tl;
 }
@@ -373,8 +413,6 @@
     if (exc) {
         uint32_t hw_exc = 0;
 
-        env->ipr[IPR_EXC_MASK] |= 1ull << regno;
-
         if (exc & float_flag_invalid) {
             hw_exc |= EXC_M_INV;
         }
@@ -390,7 +428,8 @@
         if (exc & float_flag_inexact) {
             hw_exc |= EXC_M_INE;
         }
-        helper_excp(EXCP_ARITH, hw_exc);
+
+        arith_excp(hw_exc, 1ull << regno);
     }
 }
 
@@ -420,7 +459,7 @@
             if (env->fpcr_dnz) {
                 val &= 1ull << 63;
             } else {
-                helper_excp(EXCP_ARITH, EXC_M_UNF);
+                arith_excp(EXC_M_UNF, 0);
             }
         }
     } else if (exp == 0x7ff) {
@@ -428,7 +467,7 @@
         /* ??? I'm not sure these exception bit flags are correct.  I do
            know that the Linux kernel, at least, doesn't rely on them and
            just emulates the insn to figure out what exception to use.  */
-        helper_excp(EXCP_ARITH, frac ? EXC_M_INV : EXC_M_FOV);
+        arith_excp(frac ? EXC_M_INV : EXC_M_FOV, 0);
     }
     return val;
 }
@@ -445,12 +484,12 @@
             if (env->fpcr_dnz) {
                 val &= 1ull << 63;
             } else {
-                helper_excp(EXCP_ARITH, EXC_M_UNF);
+                arith_excp(EXC_M_UNF, 0);
             }
         }
     } else if (exp == 0x7ff && frac) {
         /* NaN.  */
-        helper_excp(EXCP_ARITH, EXC_M_INV);
+        arith_excp(EXC_M_INV, 0);
     }
     return val;
 }
@@ -513,7 +552,7 @@
 
     if (unlikely(!exp && mant_sig)) {
         /* Reserved operands / Dirty zero */
-        helper_excp(EXCP_OPCDEC, 0);
+        dynamic_excp(EXCP_OPCDEC, 0);
     }
 
     if (exp < 3) {
@@ -643,7 +682,7 @@
 
     if (!exp && mant_sig) {
         /* Reserved operands / Dirty zero */
-        helper_excp(EXCP_OPCDEC, 0);
+        dynamic_excp(EXCP_OPCDEC, 0);
     }
 
     if (exp < 3) {
@@ -1156,187 +1195,122 @@
 
 /* PALcode support special instructions */
 #if !defined (CONFIG_USER_ONLY)
-void helper_hw_rei (void)
-{
-    env->pc = env->ipr[IPR_EXC_ADDR] & ~3;
-    env->ipr[IPR_EXC_ADDR] = env->ipr[IPR_EXC_ADDR] & 1;
-    env->intr_flag = 0;
-    env->lock_addr = -1;
-    /* XXX: re-enable interrupts and memory mapping */
-}
-
 void helper_hw_ret (uint64_t a)
 {
     env->pc = a & ~3;
-    env->ipr[IPR_EXC_ADDR] = a & 1;
     env->intr_flag = 0;
     env->lock_addr = -1;
-    /* XXX: re-enable interrupts and memory mapping */
+    if ((a & 1) == 0) {
+        env->pal_mode = 0;
+        swap_shadow_regs(env);
+    }
 }
 
-uint64_t helper_mfpr (int iprn, uint64_t val)
+void helper_tbia(void)
 {
-    uint64_t tmp;
-
-    if (cpu_alpha_mfpr(env, iprn, &tmp) == 0)
-        val = tmp;
-
-    return val;
+    tlb_flush(env, 1);
 }
 
-void helper_mtpr (int iprn, uint64_t val)
+void helper_tbis(uint64_t p)
 {
-    cpu_alpha_mtpr(env, iprn, val, NULL);
+    tlb_flush_page(env, p);
 }
-
-void helper_set_alt_mode (void)
-{
-    env->saved_mode = env->ps & 0xC;
-    env->ps = (env->ps & ~0xC) | (env->ipr[IPR_ALT_MODE] & 0xC);
-}
-
-void helper_restore_mode (void)
-{
-    env->ps = (env->ps & ~0xC) | env->saved_mode;
-}
-
 #endif
 
 /*****************************************************************************/
 /* Softmmu support */
 #if !defined (CONFIG_USER_ONLY)
-
-/* XXX: the two following helpers are pure hacks.
- *      Hopefully, we emulate the PALcode, then we should never see
- *      HW_LD / HW_ST instructions.
- */
-uint64_t helper_ld_virt_to_phys (uint64_t virtaddr)
+uint64_t helper_ldl_phys(uint64_t p)
 {
-    uint64_t tlb_addr, physaddr;
-    int index, mmu_idx;
-    void *retaddr;
+    return (int32_t)ldl_phys(p);
+}
 
-    mmu_idx = cpu_mmu_index(env);
-    index = (virtaddr >> TARGET_PAGE_BITS) & (CPU_TLB_SIZE - 1);
- redo:
-    tlb_addr = env->tlb_table[mmu_idx][index].addr_read;
-    if ((virtaddr & TARGET_PAGE_MASK) ==
-        (tlb_addr & (TARGET_PAGE_MASK | TLB_INVALID_MASK))) {
-        physaddr = virtaddr + env->tlb_table[mmu_idx][index].addend;
-    } else {
-        /* the page is not in the TLB : fill it */
-        retaddr = GETPC();
-        tlb_fill(virtaddr, 0, mmu_idx, retaddr);
-        goto redo;
+uint64_t helper_ldq_phys(uint64_t p)
+{
+    return ldq_phys(p);
+}
+
+uint64_t helper_ldl_l_phys(uint64_t p)
+{
+    env->lock_addr = p;
+    return env->lock_value = (int32_t)ldl_phys(p);
+}
+
+uint64_t helper_ldq_l_phys(uint64_t p)
+{
+    env->lock_addr = p;
+    return env->lock_value = ldl_phys(p);
+}
+
+void helper_stl_phys(uint64_t p, uint64_t v)
+{
+    stl_phys(p, v);
+}
+
+void helper_stq_phys(uint64_t p, uint64_t v)
+{
+    stq_phys(p, v);
+}
+
+uint64_t helper_stl_c_phys(uint64_t p, uint64_t v)
+{
+    uint64_t ret = 0;
+
+    if (p == env->lock_addr) {
+        int32_t old = ldl_phys(p);
+        if (old == (int32_t)env->lock_value) {
+            stl_phys(p, v);
+            ret = 1;
+        }
     }
-    return physaddr;
-}
-
-uint64_t helper_st_virt_to_phys (uint64_t virtaddr)
-{
-    uint64_t tlb_addr, physaddr;
-    int index, mmu_idx;
-    void *retaddr;
-
-    mmu_idx = cpu_mmu_index(env);
-    index = (virtaddr >> TARGET_PAGE_BITS) & (CPU_TLB_SIZE - 1);
- redo:
-    tlb_addr = env->tlb_table[mmu_idx][index].addr_write;
-    if ((virtaddr & TARGET_PAGE_MASK) ==
-        (tlb_addr & (TARGET_PAGE_MASK | TLB_INVALID_MASK))) {
-        physaddr = virtaddr + env->tlb_table[mmu_idx][index].addend;
-    } else {
-        /* the page is not in the TLB : fill it */
-        retaddr = GETPC();
-        tlb_fill(virtaddr, 1, mmu_idx, retaddr);
-        goto redo;
-    }
-    return physaddr;
-}
-
-void helper_ldl_raw(uint64_t t0, uint64_t t1)
-{
-    ldl_raw(t1, t0);
-}
-
-void helper_ldq_raw(uint64_t t0, uint64_t t1)
-{
-    ldq_raw(t1, t0);
-}
-
-void helper_ldl_l_raw(uint64_t t0, uint64_t t1)
-{
-    env->lock = t1;
-    ldl_raw(t1, t0);
-}
-
-void helper_ldq_l_raw(uint64_t t0, uint64_t t1)
-{
-    env->lock = t1;
-    ldl_raw(t1, t0);
-}
-
-void helper_ldl_kernel(uint64_t t0, uint64_t t1)
-{
-    ldl_kernel(t1, t0);
-}
-
-void helper_ldq_kernel(uint64_t t0, uint64_t t1)
-{
-    ldq_kernel(t1, t0);
-}
-
-void helper_ldl_data(uint64_t t0, uint64_t t1)
-{
-    ldl_data(t1, t0);
-}
-
-void helper_ldq_data(uint64_t t0, uint64_t t1)
-{
-    ldq_data(t1, t0);
-}
-
-void helper_stl_raw(uint64_t t0, uint64_t t1)
-{
-    stl_raw(t1, t0);
-}
-
-void helper_stq_raw(uint64_t t0, uint64_t t1)
-{
-    stq_raw(t1, t0);
-}
-
-uint64_t helper_stl_c_raw(uint64_t t0, uint64_t t1)
-{
-    uint64_t ret;
-
-    if (t1 == env->lock) {
-        stl_raw(t1, t0);
-        ret = 0;
-    } else
-        ret = 1;
-
-    env->lock = 1;
+    env->lock_addr = -1;
 
     return ret;
 }
 
-uint64_t helper_stq_c_raw(uint64_t t0, uint64_t t1)
+uint64_t helper_stq_c_phys(uint64_t p, uint64_t v)
 {
-    uint64_t ret;
+    uint64_t ret = 0;
 
-    if (t1 == env->lock) {
-        stq_raw(t1, t0);
-        ret = 0;
-    } else
-        ret = 1;
-
-    env->lock = 1;
+    if (p == env->lock_addr) {
+        uint64_t old = ldq_phys(p);
+        if (old == env->lock_value) {
+            stq_phys(p, v);
+            ret = 1;
+        }
+    }
+    env->lock_addr = -1;
 
     return ret;
 }
 
+static void QEMU_NORETURN do_unaligned_access(target_ulong addr, int is_write,
+                                              int is_user, void *retaddr)
+{
+    uint64_t pc;
+    uint32_t insn;
+
+    do_restore_state(retaddr);
+
+    pc = env->pc;
+    insn = ldl_code(pc);
+
+    env->trap_arg0 = addr;
+    env->trap_arg1 = insn >> 26;                /* opcode */
+    env->trap_arg2 = (insn >> 21) & 31;         /* dest regno */
+    helper_excp(EXCP_UNALIGN, 0);
+}
+
+void QEMU_NORETURN do_unassigned_access(target_phys_addr_t addr, int is_write,
+                                        int is_exec, int unused, int size)
+{
+    env->trap_arg0 = addr;
+    env->trap_arg1 = is_write;
+    dynamic_excp(EXCP_MCHK, 0);
+}
+
 #define MMUSUFFIX _mmu
+#define ALIGNED_ONLY
 
 #define SHIFT 0
 #include "softmmu_template.h"
@@ -1356,9 +1330,7 @@
 /* XXX: fix it to restore all registers */
 void tlb_fill (target_ulong addr, int is_write, int mmu_idx, void *retaddr)
 {
-    TranslationBlock *tb;
     CPUState *saved_env;
-    unsigned long pc;
     int ret;
 
     /* XXX: hack to restore env in all cases, even if not called from
@@ -1366,21 +1338,11 @@
     saved_env = env;
     env = cpu_single_env;
     ret = cpu_alpha_handle_mmu_fault(env, addr, is_write, mmu_idx, 1);
-    if (!likely(ret == 0)) {
-        if (likely(retaddr)) {
-            /* now we have a real cpu fault */
-            pc = (unsigned long)retaddr;
-            tb = tb_find_pc(pc);
-            if (likely(tb)) {
-                /* the PC is inside the translated code. It means that we have
-                   a virtual CPU fault */
-                cpu_restore_state(tb, env, pc);
-            }
-        }
+    if (unlikely(ret != 0)) {
+        do_restore_state(retaddr);
         /* Exception index and error code are already set */
         cpu_loop_exit();
     }
     env = saved_env;
 }
-
 #endif
diff --git a/target-alpha/translate.c b/target-alpha/translate.c
index 456ba51..ad6c2ca 100644
--- a/target-alpha/translate.c
+++ b/target-alpha/translate.c
@@ -47,10 +47,6 @@
     CPUAlphaState *env;
     uint64_t pc;
     int mem_idx;
-#if !defined (CONFIG_USER_ONLY)
-    int pal_mode;
-#endif
-    uint32_t amask;
 
     /* Current rounding mode for this TB.  */
     int tb_rm;
@@ -89,8 +85,10 @@
 static TCGv cpu_lock_addr;
 static TCGv cpu_lock_st_addr;
 static TCGv cpu_lock_value;
-#ifdef CONFIG_USER_ONLY
-static TCGv cpu_uniq;
+static TCGv cpu_unique;
+#ifndef CONFIG_USER_ONLY
+static TCGv cpu_sysval;
+static TCGv cpu_usp;
 #endif
 
 /* register names */
@@ -135,9 +133,13 @@
 					    offsetof(CPUState, lock_value),
 					    "lock_value");
 
-#ifdef CONFIG_USER_ONLY
-    cpu_uniq = tcg_global_mem_new_i64(TCG_AREG0,
-                                      offsetof(CPUState, unique), "uniq");
+    cpu_unique = tcg_global_mem_new_i64(TCG_AREG0,
+                                        offsetof(CPUState, unique), "unique");
+#ifndef CONFIG_USER_ONLY
+    cpu_sysval = tcg_global_mem_new_i64(TCG_AREG0,
+                                        offsetof(CPUState, sysval), "sysval");
+    cpu_usp = tcg_global_mem_new_i64(TCG_AREG0,
+                                     offsetof(CPUState, usp), "usp");
 #endif
 
     /* register helpers */
@@ -147,17 +149,21 @@
     done_init = 1;
 }
 
-static ExitStatus gen_excp(DisasContext *ctx, int exception, int error_code)
+static void gen_excp_1(int exception, int error_code)
 {
     TCGv_i32 tmp1, tmp2;
 
-    tcg_gen_movi_i64(cpu_pc, ctx->pc);
     tmp1 = tcg_const_i32(exception);
     tmp2 = tcg_const_i32(error_code);
     gen_helper_excp(tmp1, tmp2);
     tcg_temp_free_i32(tmp2);
     tcg_temp_free_i32(tmp1);
+}
 
+static ExitStatus gen_excp(DisasContext *ctx, int exception, int error_code)
+{
+    tcg_gen_movi_i64(cpu_pc, ctx->pc);
+    gen_excp_1(exception, error_code);
     return EXIT_NORETURN;
 }
 
@@ -322,7 +328,7 @@
 #if defined(CONFIG_USER_ONLY)
     addr = cpu_lock_st_addr;
 #else
-    addr = tcg_local_new();
+    addr = tcg_temp_local_new();
 #endif
 
     if (rb != 31) {
@@ -345,7 +351,7 @@
 
         lab_fail = gen_new_label();
         lab_done = gen_new_label();
-        tcg_gen_brcond(TCG_COND_NE, addr, cpu_lock_addr, lab_fail);
+        tcg_gen_brcond_i64(TCG_COND_NE, addr, cpu_lock_addr, lab_fail);
 
         val = tcg_temp_new();
         if (quad) {
@@ -353,7 +359,7 @@
         } else {
             tcg_gen_qemu_ld32s(val, addr, ctx->mem_idx);
         }
-        tcg_gen_brcond(TCG_COND_NE, val, cpu_lock_value, lab_fail);
+        tcg_gen_brcond_i64(TCG_COND_NE, val, cpu_lock_value, lab_fail);
 
         if (quad) {
             tcg_gen_qemu_st64(cpu_ir[ra], addr, ctx->mem_idx);
@@ -1464,6 +1470,194 @@
     tcg_temp_free_i32(tmp);
 }
 
+static ExitStatus gen_call_pal(DisasContext *ctx, int palcode)
+{
+    /* We're emulating OSF/1 PALcode.  Many of these are trivial access
+       to internal cpu registers.  */
+
+    /* Unprivileged PAL call */
+    if (palcode >= 0x80 && palcode < 0xC0) {
+        switch (palcode) {
+        case 0x86:
+            /* IMB */
+            /* No-op inside QEMU.  */
+            break;
+        case 0x9E:
+            /* RDUNIQUE */
+            tcg_gen_mov_i64(cpu_ir[IR_V0], cpu_unique);
+            break;
+        case 0x9F:
+            /* WRUNIQUE */
+            tcg_gen_mov_i64(cpu_unique, cpu_ir[IR_A0]);
+            break;
+        default:
+            return gen_excp(ctx, EXCP_CALL_PAL, palcode & 0xbf);
+        }
+        return NO_EXIT;
+    }
+
+#ifndef CONFIG_USER_ONLY
+    /* Privileged PAL code */
+    if (palcode < 0x40 && (ctx->tb->flags & TB_FLAGS_USER_MODE) == 0) {
+        switch (palcode) {
+        case 0x01:
+            /* CFLUSH */
+            /* No-op inside QEMU.  */
+            break;
+        case 0x02:
+            /* DRAINA */
+            /* No-op inside QEMU.  */
+            break;
+        case 0x2D:
+            /* WRVPTPTR */
+            tcg_gen_st_i64(cpu_ir[IR_A0], cpu_env, offsetof(CPUState, vptptr));
+            break;
+        case 0x31:
+            /* WRVAL */
+            tcg_gen_mov_i64(cpu_sysval, cpu_ir[IR_A0]);
+            break;
+        case 0x32:
+            /* RDVAL */
+            tcg_gen_mov_i64(cpu_ir[IR_V0], cpu_sysval);
+            break;
+
+        case 0x35: {
+            /* SWPIPL */
+            TCGv tmp;
+
+            /* Note that we already know we're in kernel mode, so we know
+               that PS only contains the 3 IPL bits.  */
+            tcg_gen_ld8u_i64(cpu_ir[IR_V0], cpu_env, offsetof(CPUState, ps));
+
+            /* But make sure and store only the 3 IPL bits from the user.  */
+            tmp = tcg_temp_new();
+            tcg_gen_andi_i64(tmp, cpu_ir[IR_A0], PS_INT_MASK);
+            tcg_gen_st8_i64(tmp, cpu_env, offsetof(CPUState, ps));
+            tcg_temp_free(tmp);
+            break;
+        }
+
+        case 0x36:
+            /* RDPS */
+            tcg_gen_ld8u_i64(cpu_ir[IR_V0], cpu_env, offsetof(CPUState, ps));
+            break;
+        case 0x38:
+            /* WRUSP */
+            tcg_gen_mov_i64(cpu_usp, cpu_ir[IR_A0]);
+            break;
+        case 0x3A:
+            /* RDUSP */
+            tcg_gen_mov_i64(cpu_ir[IR_V0], cpu_usp);
+            break;
+        case 0x3C:
+            /* WHAMI */
+            tcg_gen_ld32s_i64(cpu_ir[IR_V0], cpu_env,
+                              offsetof(CPUState, cpu_index));
+            break;
+
+        default:
+            return gen_excp(ctx, EXCP_CALL_PAL, palcode & 0x3f);
+        }
+        return NO_EXIT;
+    }
+#endif
+
+    return gen_invalid(ctx);
+}
+
+#ifndef CONFIG_USER_ONLY
+
+#define PR_BYTE         0x100000
+#define PR_LONG         0x200000
+
+static int cpu_pr_data(int pr)
+{
+    switch (pr) {
+    case  0: return offsetof(CPUAlphaState, ps) | PR_BYTE;
+    case  1: return offsetof(CPUAlphaState, fen) | PR_BYTE;
+    case  2: return offsetof(CPUAlphaState, pcc_ofs) | PR_LONG;
+    case  3: return offsetof(CPUAlphaState, trap_arg0);
+    case  4: return offsetof(CPUAlphaState, trap_arg1);
+    case  5: return offsetof(CPUAlphaState, trap_arg2);
+    case  6: return offsetof(CPUAlphaState, exc_addr);
+    case  7: return offsetof(CPUAlphaState, palbr);
+    case  8: return offsetof(CPUAlphaState, ptbr);
+    case  9: return offsetof(CPUAlphaState, vptptr);
+    case 10: return offsetof(CPUAlphaState, unique);
+    case 11: return offsetof(CPUAlphaState, sysval);
+    case 12: return offsetof(CPUAlphaState, usp);
+
+    case 32 ... 39:
+        return offsetof(CPUAlphaState, shadow[pr - 32]);
+    case 40 ... 63:
+        return offsetof(CPUAlphaState, scratch[pr - 40]);
+    }
+    return 0;
+}
+
+static void gen_mfpr(int ra, int regno)
+{
+    int data = cpu_pr_data(regno);
+
+    /* In our emulated PALcode, these processor registers have no
+       side effects from reading.  */
+    if (ra == 31) {
+        return;
+    }
+
+    /* The basic registers are data only, and unknown registers
+       are read-zero, write-ignore.  */
+    if (data == 0) {
+        tcg_gen_movi_i64(cpu_ir[ra], 0);
+    } else if (data & PR_BYTE) {
+        tcg_gen_ld8u_i64(cpu_ir[ra], cpu_env, data & ~PR_BYTE);
+    } else if (data & PR_LONG) {
+        tcg_gen_ld32s_i64(cpu_ir[ra], cpu_env, data & ~PR_LONG);
+    } else {
+        tcg_gen_ld_i64(cpu_ir[ra], cpu_env, data);
+    }
+}
+
+static void gen_mtpr(int rb, int regno)
+{
+    TCGv tmp;
+
+    if (rb == 31) {
+        tmp = tcg_const_i64(0);
+    } else {
+        tmp = cpu_ir[rb];
+    }
+
+    /* These two register numbers perform a TLB cache flush.  Thankfully we
+       can only do this inside PALmode, which means that the current basic
+       block cannot be affected by the change in mappings.  */
+    if (regno == 255) {
+        /* TBIA */
+        gen_helper_tbia();
+    } else if (regno == 254) {
+        /* TBIS */
+        gen_helper_tbis(tmp);
+    } else {
+        /* The basic registers are data only, and unknown registers
+           are read-zero, write-ignore.  */
+        int data = cpu_pr_data(regno);
+        if (data != 0) {
+            if (data & PR_BYTE) {
+                tcg_gen_st8_i64(tmp, cpu_env, data & ~PR_BYTE);
+            } else if (data & PR_LONG) {
+                tcg_gen_st32_i64(tmp, cpu_env, data & ~PR_LONG);
+            } else {
+                tcg_gen_st_i64(tmp, cpu_env, data);
+            }
+        }
+    }
+
+    if (rb == 31) {
+        tcg_temp_free(tmp);
+    }
+}
+#endif /* !USER_ONLY*/
+
 static ExitStatus translate_one(DisasContext *ctx, uint32_t insn)
 {
     uint32_t palcode;
@@ -1499,32 +1693,8 @@
     switch (opc) {
     case 0x00:
         /* CALL_PAL */
-#ifdef CONFIG_USER_ONLY
-        if (palcode == 0x9E) {
-            /* RDUNIQUE */
-            tcg_gen_mov_i64(cpu_ir[IR_V0], cpu_uniq);
-            break;
-        } else if (palcode == 0x9F) {
-            /* WRUNIQUE */
-            tcg_gen_mov_i64(cpu_uniq, cpu_ir[IR_A0]);
-            break;
-        }
-#endif
-        if (palcode >= 0x80 && palcode < 0xC0) {
-            /* Unprivileged PAL call */
-            ret = gen_excp(ctx, EXCP_CALL_PAL + ((palcode & 0x3F) << 6), 0);
-            break;
-        }
-#ifndef CONFIG_USER_ONLY
-        if (palcode < 0x40) {
-            /* Privileged PAL code */
-            if (ctx->mem_idx & 1)
-                goto invalid_opc;
-            ret = gen_excp(ctx, EXCP_CALL_PALP + ((palcode & 0x3F) << 6), 0);
-        }
-#endif
-        /* Invalid PAL call */
-        goto invalid_opc;
+        ret = gen_call_pal(ctx, palcode);
+        break;
     case 0x01:
         /* OPC01 */
         goto invalid_opc;
@@ -1566,20 +1736,22 @@
         break;
     case 0x0A:
         /* LDBU */
-        if (!(ctx->amask & AMASK_BWX))
-            goto invalid_opc;
-        gen_load_mem(ctx, &tcg_gen_qemu_ld8u, ra, rb, disp16, 0, 0);
-        break;
+        if (ctx->tb->flags & TB_FLAGS_AMASK_BWX) {
+            gen_load_mem(ctx, &tcg_gen_qemu_ld8u, ra, rb, disp16, 0, 0);
+            break;
+        }
+        goto invalid_opc;
     case 0x0B:
         /* LDQ_U */
         gen_load_mem(ctx, &tcg_gen_qemu_ld64, ra, rb, disp16, 0, 1);
         break;
     case 0x0C:
         /* LDWU */
-        if (!(ctx->amask & AMASK_BWX))
-            goto invalid_opc;
-        gen_load_mem(ctx, &tcg_gen_qemu_ld16u, ra, rb, disp16, 0, 0);
-        break;
+        if (ctx->tb->flags & TB_FLAGS_AMASK_BWX) {
+            gen_load_mem(ctx, &tcg_gen_qemu_ld16u, ra, rb, disp16, 0, 0);
+            break;
+        }
+        goto invalid_opc;
     case 0x0D:
         /* STW */
         gen_store_mem(ctx, &tcg_gen_qemu_st16, ra, rb, disp16, 0, 0);
@@ -1983,20 +2155,12 @@
         case 0x61:
             /* AMASK */
             if (likely(rc != 31)) {
-                if (islit)
-                    tcg_gen_movi_i64(cpu_ir[rc], lit);
-                else
-                    tcg_gen_mov_i64(cpu_ir[rc], cpu_ir[rb]);
-                switch (ctx->env->implver) {
-                case IMPLVER_2106x:
-                    /* EV4, EV45, LCA, LCA45 & EV5 */
-                    break;
-                case IMPLVER_21164:
-                case IMPLVER_21264:
-                case IMPLVER_21364:
-                    tcg_gen_andi_i64(cpu_ir[rc], cpu_ir[rc],
-                                     ~(uint64_t)ctx->amask);
-                    break;
+                uint64_t amask = ctx->tb->flags >> TB_FLAGS_AMASK_SHIFT;
+
+                if (islit) {
+                    tcg_gen_movi_i64(cpu_ir[rc], lit & ~amask);
+                } else {
+                    tcg_gen_andi_i64(cpu_ir[rc], cpu_ir[rb], ~amask);
                 }
             }
             break;
@@ -2210,8 +2374,9 @@
         switch (fpfn) { /* fn11 & 0x3F */
         case 0x04:
             /* ITOFS */
-            if (!(ctx->amask & AMASK_FIX))
+            if ((ctx->tb->flags & TB_FLAGS_AMASK_FIX) == 0) {
                 goto invalid_opc;
+            }
             if (likely(rc != 31)) {
                 if (ra != 31) {
                     TCGv_i32 tmp = tcg_temp_new_i32();
@@ -2224,20 +2389,23 @@
             break;
         case 0x0A:
             /* SQRTF */
-            if (!(ctx->amask & AMASK_FIX))
-                goto invalid_opc;
-            gen_fsqrtf(rb, rc);
-            break;
+            if (ctx->tb->flags & TB_FLAGS_AMASK_FIX) {
+                gen_fsqrtf(rb, rc);
+                break;
+            }
+            goto invalid_opc;
         case 0x0B:
             /* SQRTS */
-            if (!(ctx->amask & AMASK_FIX))
-                goto invalid_opc;
-            gen_fsqrts(ctx, rb, rc, fn11);
-            break;
+            if (ctx->tb->flags & TB_FLAGS_AMASK_FIX) {
+                gen_fsqrts(ctx, rb, rc, fn11);
+                break;
+            }
+            goto invalid_opc;
         case 0x14:
             /* ITOFF */
-            if (!(ctx->amask & AMASK_FIX))
+            if ((ctx->tb->flags & TB_FLAGS_AMASK_FIX) == 0) {
                 goto invalid_opc;
+            }
             if (likely(rc != 31)) {
                 if (ra != 31) {
                     TCGv_i32 tmp = tcg_temp_new_i32();
@@ -2250,8 +2418,9 @@
             break;
         case 0x24:
             /* ITOFT */
-            if (!(ctx->amask & AMASK_FIX))
+            if ((ctx->tb->flags & TB_FLAGS_AMASK_FIX) == 0) {
                 goto invalid_opc;
+            }
             if (likely(rc != 31)) {
                 if (ra != 31)
                     tcg_gen_mov_i64(cpu_fir[rc], cpu_ir[ra]);
@@ -2261,16 +2430,18 @@
             break;
         case 0x2A:
             /* SQRTG */
-            if (!(ctx->amask & AMASK_FIX))
-                goto invalid_opc;
-            gen_fsqrtg(rb, rc);
-            break;
+            if (ctx->tb->flags & TB_FLAGS_AMASK_FIX) {
+                gen_fsqrtg(rb, rc);
+                break;
+            }
+            goto invalid_opc;
         case 0x02B:
             /* SQRTT */
-            if (!(ctx->amask & AMASK_FIX))
-                goto invalid_opc;
-            gen_fsqrtt(ctx, rb, rc, fn11);
-            break;
+            if (ctx->tb->flags & TB_FLAGS_AMASK_FIX) {
+                gen_fsqrtt(ctx, rb, rc, fn11);
+                break;
+            }
+            goto invalid_opc;
         default:
             goto invalid_opc;
         }
@@ -2571,18 +2742,13 @@
         break;
     case 0x19:
         /* HW_MFPR (PALcode) */
-#if defined (CONFIG_USER_ONLY)
-        goto invalid_opc;
-#else
-        if (!ctx->pal_mode)
-            goto invalid_opc;
-        if (ra != 31) {
-            TCGv tmp = tcg_const_i32(insn & 0xFF);
-            gen_helper_mfpr(cpu_ir[ra], tmp, cpu_ir[ra]);
-            tcg_temp_free(tmp);
+#ifndef CONFIG_USER_ONLY
+        if (ctx->tb->flags & TB_FLAGS_PAL_MODE) {
+            gen_mfpr(ra, insn & 0xffff);
+            break;
         }
-        break;
 #endif
+        goto invalid_opc;
     case 0x1A:
         /* JMP, JSR, RET, JSR_COROUTINE.  These only differ by the branch
            prediction stack action, which of course we don't implement.  */
@@ -2598,13 +2764,15 @@
         break;
     case 0x1B:
         /* HW_LD (PALcode) */
-#if defined (CONFIG_USER_ONLY)
-        goto invalid_opc;
-#else
-        if (!ctx->pal_mode)
-            goto invalid_opc;
-        if (ra != 31) {
-            TCGv addr = tcg_temp_new();
+#ifndef CONFIG_USER_ONLY
+        if (ctx->tb->flags & TB_FLAGS_PAL_MODE) {
+            TCGv addr;
+
+            if (ra == 31) {
+                break;
+            }
+
+            addr = tcg_temp_new();
             if (rb != 31)
                 tcg_gen_addi_i64(addr, cpu_ir[rb], disp12);
             else
@@ -2612,27 +2780,26 @@
             switch ((insn >> 12) & 0xF) {
             case 0x0:
                 /* Longword physical access (hw_ldl/p) */
-                gen_helper_ldl_raw(cpu_ir[ra], addr);
+                gen_helper_ldl_phys(cpu_ir[ra], addr);
                 break;
             case 0x1:
                 /* Quadword physical access (hw_ldq/p) */
-                gen_helper_ldq_raw(cpu_ir[ra], addr);
+                gen_helper_ldq_phys(cpu_ir[ra], addr);
                 break;
             case 0x2:
                 /* Longword physical access with lock (hw_ldl_l/p) */
-                gen_helper_ldl_l_raw(cpu_ir[ra], addr);
+                gen_helper_ldl_l_phys(cpu_ir[ra], addr);
                 break;
             case 0x3:
                 /* Quadword physical access with lock (hw_ldq_l/p) */
-                gen_helper_ldq_l_raw(cpu_ir[ra], addr);
+                gen_helper_ldq_l_phys(cpu_ir[ra], addr);
                 break;
             case 0x4:
                 /* Longword virtual PTE fetch (hw_ldl/v) */
-                tcg_gen_qemu_ld32s(cpu_ir[ra], addr, 0);
-                break;
+                goto invalid_opc;
             case 0x5:
                 /* Quadword virtual PTE fetch (hw_ldq/v) */
-                tcg_gen_qemu_ld64(cpu_ir[ra], addr, 0);
+                goto invalid_opc;
                 break;
             case 0x6:
                 /* Incpu_ir[ra]id */
@@ -2642,63 +2809,47 @@
                 goto invalid_opc;
             case 0x8:
                 /* Longword virtual access (hw_ldl) */
-                gen_helper_st_virt_to_phys(addr, addr);
-                gen_helper_ldl_raw(cpu_ir[ra], addr);
-                break;
+                goto invalid_opc;
             case 0x9:
                 /* Quadword virtual access (hw_ldq) */
-                gen_helper_st_virt_to_phys(addr, addr);
-                gen_helper_ldq_raw(cpu_ir[ra], addr);
-                break;
+                goto invalid_opc;
             case 0xA:
                 /* Longword virtual access with protection check (hw_ldl/w) */
-                tcg_gen_qemu_ld32s(cpu_ir[ra], addr, 0);
+                tcg_gen_qemu_ld32s(cpu_ir[ra], addr, MMU_KERNEL_IDX);
                 break;
             case 0xB:
                 /* Quadword virtual access with protection check (hw_ldq/w) */
-                tcg_gen_qemu_ld64(cpu_ir[ra], addr, 0);
+                tcg_gen_qemu_ld64(cpu_ir[ra], addr, MMU_KERNEL_IDX);
                 break;
             case 0xC:
                 /* Longword virtual access with alt access mode (hw_ldl/a)*/
-                gen_helper_set_alt_mode();
-                gen_helper_st_virt_to_phys(addr, addr);
-                gen_helper_ldl_raw(cpu_ir[ra], addr);
-                gen_helper_restore_mode();
-                break;
+                goto invalid_opc;
             case 0xD:
                 /* Quadword virtual access with alt access mode (hw_ldq/a) */
-                gen_helper_set_alt_mode();
-                gen_helper_st_virt_to_phys(addr, addr);
-                gen_helper_ldq_raw(cpu_ir[ra], addr);
-                gen_helper_restore_mode();
-                break;
+                goto invalid_opc;
             case 0xE:
                 /* Longword virtual access with alternate access mode and
-                 * protection checks (hw_ldl/wa)
-                 */
-                gen_helper_set_alt_mode();
-                gen_helper_ldl_data(cpu_ir[ra], addr);
-                gen_helper_restore_mode();
+                   protection checks (hw_ldl/wa) */
+                tcg_gen_qemu_ld32s(cpu_ir[ra], addr, MMU_USER_IDX);
                 break;
             case 0xF:
                 /* Quadword virtual access with alternate access mode and
-                 * protection checks (hw_ldq/wa)
-                 */
-                gen_helper_set_alt_mode();
-                gen_helper_ldq_data(cpu_ir[ra], addr);
-                gen_helper_restore_mode();
+                   protection checks (hw_ldq/wa) */
+                tcg_gen_qemu_ld64(cpu_ir[ra], addr, MMU_USER_IDX);
                 break;
             }
             tcg_temp_free(addr);
+            break;
         }
-        break;
 #endif
+        goto invalid_opc;
     case 0x1C:
         switch (fn7) {
         case 0x00:
             /* SEXTB */
-            if (!(ctx->amask & AMASK_BWX))
+            if ((ctx->tb->flags & TB_FLAGS_AMASK_BWX) == 0) {
                 goto invalid_opc;
+            }
             if (likely(rc != 31)) {
                 if (islit)
                     tcg_gen_movi_i64(cpu_ir[rc], (int64_t)((int8_t)lit));
@@ -2708,138 +2859,164 @@
             break;
         case 0x01:
             /* SEXTW */
-            if (!(ctx->amask & AMASK_BWX))
-                goto invalid_opc;
-            if (likely(rc != 31)) {
-                if (islit)
-                    tcg_gen_movi_i64(cpu_ir[rc], (int64_t)((int16_t)lit));
-                else
-                    tcg_gen_ext16s_i64(cpu_ir[rc], cpu_ir[rb]);
+            if (ctx->tb->flags & TB_FLAGS_AMASK_BWX) {
+                if (likely(rc != 31)) {
+                    if (islit) {
+                        tcg_gen_movi_i64(cpu_ir[rc], (int64_t)((int16_t)lit));
+                    } else {
+                        tcg_gen_ext16s_i64(cpu_ir[rc], cpu_ir[rb]);
+                    }
+                }
+                break;
             }
-            break;
+            goto invalid_opc;
         case 0x30:
             /* CTPOP */
-            if (!(ctx->amask & AMASK_CIX))
-                goto invalid_opc;
-            if (likely(rc != 31)) {
-                if (islit)
-                    tcg_gen_movi_i64(cpu_ir[rc], ctpop64(lit));
-                else
-                    gen_helper_ctpop(cpu_ir[rc], cpu_ir[rb]);
+            if (ctx->tb->flags & TB_FLAGS_AMASK_CIX) {
+                if (likely(rc != 31)) {
+                    if (islit) {
+                        tcg_gen_movi_i64(cpu_ir[rc], ctpop64(lit));
+                    } else {
+                        gen_helper_ctpop(cpu_ir[rc], cpu_ir[rb]);
+                    }
+                }
+                break;
             }
-            break;
+            goto invalid_opc;
         case 0x31:
             /* PERR */
-            if (!(ctx->amask & AMASK_MVI))
-                goto invalid_opc;
-            gen_perr(ra, rb, rc, islit, lit);
-            break;
+            if (ctx->tb->flags & TB_FLAGS_AMASK_MVI) {
+                gen_perr(ra, rb, rc, islit, lit);
+                break;
+            }
+            goto invalid_opc;
         case 0x32:
             /* CTLZ */
-            if (!(ctx->amask & AMASK_CIX))
-                goto invalid_opc;
-            if (likely(rc != 31)) {
-                if (islit)
-                    tcg_gen_movi_i64(cpu_ir[rc], clz64(lit));
-                else
-                    gen_helper_ctlz(cpu_ir[rc], cpu_ir[rb]);
+            if (ctx->tb->flags & TB_FLAGS_AMASK_CIX) {
+                if (likely(rc != 31)) {
+                    if (islit) {
+                        tcg_gen_movi_i64(cpu_ir[rc], clz64(lit));
+                    } else {
+                        gen_helper_ctlz(cpu_ir[rc], cpu_ir[rb]);
+                    }
+                }
+                break;
             }
-            break;
+            goto invalid_opc;
         case 0x33:
             /* CTTZ */
-            if (!(ctx->amask & AMASK_CIX))
-                goto invalid_opc;
-            if (likely(rc != 31)) {
-                if (islit)
-                    tcg_gen_movi_i64(cpu_ir[rc], ctz64(lit));
-                else
-                    gen_helper_cttz(cpu_ir[rc], cpu_ir[rb]);
+            if (ctx->tb->flags & TB_FLAGS_AMASK_CIX) {
+                if (likely(rc != 31)) {
+                    if (islit) {
+                        tcg_gen_movi_i64(cpu_ir[rc], ctz64(lit));
+                    } else {
+                        gen_helper_cttz(cpu_ir[rc], cpu_ir[rb]);
+                    }
+                }
+                break;
             }
-            break;
+            goto invalid_opc;
         case 0x34:
             /* UNPKBW */
-            if (!(ctx->amask & AMASK_MVI))
-                goto invalid_opc;
-            if (real_islit || ra != 31)
-                goto invalid_opc;
-            gen_unpkbw (rb, rc);
-            break;
+            if (ctx->tb->flags & TB_FLAGS_AMASK_MVI) {
+                if (real_islit || ra != 31) {
+                    goto invalid_opc;
+                }
+                gen_unpkbw(rb, rc);
+                break;
+            }
+            goto invalid_opc;
         case 0x35:
             /* UNPKBL */
-            if (!(ctx->amask & AMASK_MVI))
-                goto invalid_opc;
-            if (real_islit || ra != 31)
-                goto invalid_opc;
-            gen_unpkbl (rb, rc);
-            break;
+            if (ctx->tb->flags & TB_FLAGS_AMASK_MVI) {
+                if (real_islit || ra != 31) {
+                    goto invalid_opc;
+                }
+                gen_unpkbl(rb, rc);
+                break;
+            }
+            goto invalid_opc;
         case 0x36:
             /* PKWB */
-            if (!(ctx->amask & AMASK_MVI))
-                goto invalid_opc;
-            if (real_islit || ra != 31)
-                goto invalid_opc;
-            gen_pkwb (rb, rc);
-            break;
+            if (ctx->tb->flags & TB_FLAGS_AMASK_MVI) {
+                if (real_islit || ra != 31) {
+                    goto invalid_opc;
+                }
+                gen_pkwb(rb, rc);
+                break;
+            }
+            goto invalid_opc;
         case 0x37:
             /* PKLB */
-            if (!(ctx->amask & AMASK_MVI))
-                goto invalid_opc;
-            if (real_islit || ra != 31)
-                goto invalid_opc;
-            gen_pklb (rb, rc);
-            break;
+            if (ctx->tb->flags & TB_FLAGS_AMASK_MVI) {
+                if (real_islit || ra != 31) {
+                    goto invalid_opc;
+                }
+                gen_pklb(rb, rc);
+                break;
+            }
+            goto invalid_opc;
         case 0x38:
             /* MINSB8 */
-            if (!(ctx->amask & AMASK_MVI))
-                goto invalid_opc;
-            gen_minsb8 (ra, rb, rc, islit, lit);
-            break;
+            if (ctx->tb->flags & TB_FLAGS_AMASK_MVI) {
+                gen_minsb8(ra, rb, rc, islit, lit);
+                break;
+            }
+            goto invalid_opc;
         case 0x39:
             /* MINSW4 */
-            if (!(ctx->amask & AMASK_MVI))
-                goto invalid_opc;
-            gen_minsw4 (ra, rb, rc, islit, lit);
-            break;
+            if (ctx->tb->flags & TB_FLAGS_AMASK_MVI) {
+                gen_minsw4(ra, rb, rc, islit, lit);
+                break;
+            }
+            goto invalid_opc;
         case 0x3A:
             /* MINUB8 */
-            if (!(ctx->amask & AMASK_MVI))
-                goto invalid_opc;
-            gen_minub8 (ra, rb, rc, islit, lit);
-            break;
+            if (ctx->tb->flags & TB_FLAGS_AMASK_MVI) {
+                gen_minub8(ra, rb, rc, islit, lit);
+                break;
+            }
+            goto invalid_opc;
         case 0x3B:
             /* MINUW4 */
-            if (!(ctx->amask & AMASK_MVI))
-                goto invalid_opc;
-            gen_minuw4 (ra, rb, rc, islit, lit);
-            break;
+            if (ctx->tb->flags & TB_FLAGS_AMASK_MVI) {
+                gen_minuw4(ra, rb, rc, islit, lit);
+                break;
+            }
+            goto invalid_opc;
         case 0x3C:
             /* MAXUB8 */
-            if (!(ctx->amask & AMASK_MVI))
-                goto invalid_opc;
-            gen_maxub8 (ra, rb, rc, islit, lit);
-            break;
+            if (ctx->tb->flags & TB_FLAGS_AMASK_MVI) {
+                gen_maxub8(ra, rb, rc, islit, lit);
+                break;
+            }
+            goto invalid_opc;
         case 0x3D:
             /* MAXUW4 */
-            if (!(ctx->amask & AMASK_MVI))
-                goto invalid_opc;
-            gen_maxuw4 (ra, rb, rc, islit, lit);
-            break;
+            if (ctx->tb->flags & TB_FLAGS_AMASK_MVI) {
+                gen_maxuw4(ra, rb, rc, islit, lit);
+                break;
+            }
+            goto invalid_opc;
         case 0x3E:
             /* MAXSB8 */
-            if (!(ctx->amask & AMASK_MVI))
-                goto invalid_opc;
-            gen_maxsb8 (ra, rb, rc, islit, lit);
-            break;
+            if (ctx->tb->flags & TB_FLAGS_AMASK_MVI) {
+                gen_maxsb8(ra, rb, rc, islit, lit);
+                break;
+            }
+            goto invalid_opc;
         case 0x3F:
             /* MAXSW4 */
-            if (!(ctx->amask & AMASK_MVI))
-                goto invalid_opc;
-            gen_maxsw4 (ra, rb, rc, islit, lit);
-            break;
+            if (ctx->tb->flags & TB_FLAGS_AMASK_MVI) {
+                gen_maxsw4(ra, rb, rc, islit, lit);
+                break;
+            }
+            goto invalid_opc;
         case 0x70:
             /* FTOIT */
-            if (!(ctx->amask & AMASK_FIX))
+            if ((ctx->tb->flags & TB_FLAGS_AMASK_FIX) == 0) {
                 goto invalid_opc;
+            }
             if (likely(rc != 31)) {
                 if (ra != 31)
                     tcg_gen_mov_i64(cpu_ir[rc], cpu_fir[ra]);
@@ -2849,8 +3026,9 @@
             break;
         case 0x78:
             /* FTOIS */
-            if (!(ctx->amask & AMASK_FIX))
+            if ((ctx->tb->flags & TB_FLAGS_AMASK_FIX) == 0) {
                 goto invalid_opc;
+            }
             if (rc != 31) {
                 TCGv_i32 tmp1 = tcg_temp_new_i32();
                 if (ra != 31)
@@ -2870,57 +3048,37 @@
         break;
     case 0x1D:
         /* HW_MTPR (PALcode) */
-#if defined (CONFIG_USER_ONLY)
-        goto invalid_opc;
-#else
-        if (!ctx->pal_mode)
-            goto invalid_opc;
-        else {
-            TCGv tmp1 = tcg_const_i32(insn & 0xFF);
-            if (ra != 31)
-                gen_helper_mtpr(tmp1, cpu_ir[ra]);
-            else {
-                TCGv tmp2 = tcg_const_i64(0);
-                gen_helper_mtpr(tmp1, tmp2);
-                tcg_temp_free(tmp2);
-            }
-            tcg_temp_free(tmp1);
-            ret = EXIT_PC_STALE;
+#ifndef CONFIG_USER_ONLY
+        if (ctx->tb->flags & TB_FLAGS_PAL_MODE) {
+            gen_mtpr(rb, insn & 0xffff);
+            break;
         }
-        break;
 #endif
+        goto invalid_opc;
     case 0x1E:
-        /* HW_REI (PALcode) */
-#if defined (CONFIG_USER_ONLY)
-        goto invalid_opc;
-#else
-        if (!ctx->pal_mode)
-            goto invalid_opc;
-        if (rb == 31) {
-            /* "Old" alpha */
-            gen_helper_hw_rei();
-        } else {
-            TCGv tmp;
-
-            if (ra != 31) {
-                tmp = tcg_temp_new();
-                tcg_gen_addi_i64(tmp, cpu_ir[rb], (((int64_t)insn << 51) >> 51));
-            } else
-                tmp = tcg_const_i64(((int64_t)insn << 51) >> 51);
-            gen_helper_hw_ret(tmp);
-            tcg_temp_free(tmp);
+        /* HW_RET (PALcode) */
+#ifndef CONFIG_USER_ONLY
+        if (ctx->tb->flags & TB_FLAGS_PAL_MODE) {
+            if (rb == 31) {
+                /* Pre-EV6 CPUs interpreted this as HW_REI, loading the return
+                   address from EXC_ADDR.  This turns out to be useful for our
+                   emulation PALcode, so continue to accept it.  */
+                TCGv tmp = tcg_temp_new();
+                tcg_gen_ld_i64(tmp, cpu_env, offsetof(CPUState, exc_addr));
+                gen_helper_hw_ret(tmp);
+                tcg_temp_free(tmp);
+            } else {
+                gen_helper_hw_ret(cpu_ir[rb]);
+            }
+            ret = EXIT_PC_UPDATED;
+            break;
         }
-        ret = EXIT_PC_UPDATED;
-        break;
 #endif
+        goto invalid_opc;
     case 0x1F:
         /* HW_ST (PALcode) */
-#if defined (CONFIG_USER_ONLY)
-        goto invalid_opc;
-#else
-        if (!ctx->pal_mode)
-            goto invalid_opc;
-        else {
+#ifndef CONFIG_USER_ONLY
+        if (ctx->tb->flags & TB_FLAGS_PAL_MODE) {
             TCGv addr, val;
             addr = tcg_temp_new();
             if (rb != 31)
@@ -2936,30 +3094,26 @@
             switch ((insn >> 12) & 0xF) {
             case 0x0:
                 /* Longword physical access */
-                gen_helper_stl_raw(val, addr);
+                gen_helper_stl_phys(addr, val);
                 break;
             case 0x1:
                 /* Quadword physical access */
-                gen_helper_stq_raw(val, addr);
+                gen_helper_stq_phys(addr, val);
                 break;
             case 0x2:
                 /* Longword physical access with lock */
-                gen_helper_stl_c_raw(val, val, addr);
+                gen_helper_stl_c_phys(val, addr, val);
                 break;
             case 0x3:
                 /* Quadword physical access with lock */
-                gen_helper_stq_c_raw(val, val, addr);
+                gen_helper_stq_c_phys(val, addr, val);
                 break;
             case 0x4:
                 /* Longword virtual access */
-                gen_helper_st_virt_to_phys(addr, addr);
-                gen_helper_stl_raw(val, addr);
-                break;
+                goto invalid_opc;
             case 0x5:
                 /* Quadword virtual access */
-                gen_helper_st_virt_to_phys(addr, addr);
-                gen_helper_stq_raw(val, addr);
-                break;
+                goto invalid_opc;
             case 0x6:
                 /* Invalid */
                 goto invalid_opc;
@@ -2980,18 +3134,10 @@
                 goto invalid_opc;
             case 0xC:
                 /* Longword virtual access with alternate access mode */
-                gen_helper_set_alt_mode();
-                gen_helper_st_virt_to_phys(addr, addr);
-                gen_helper_stl_raw(val, addr);
-                gen_helper_restore_mode();
-                break;
+                goto invalid_opc;
             case 0xD:
                 /* Quadword virtual access with alternate access mode */
-                gen_helper_set_alt_mode();
-                gen_helper_st_virt_to_phys(addr, addr);
-                gen_helper_stl_raw(val, addr);
-                gen_helper_restore_mode();
-                break;
+                goto invalid_opc;
             case 0xE:
                 /* Invalid */
                 goto invalid_opc;
@@ -3002,9 +3148,10 @@
             if (ra == 31)
                 tcg_temp_free(val);
             tcg_temp_free(addr);
+            break;
         }
-        break;
 #endif
+        goto invalid_opc;
     case 0x20:
         /* LDF */
         gen_load_mem(ctx, &gen_qemu_ldf, ra, rb, disp16, 1, 0);
@@ -3155,13 +3302,7 @@
     ctx.tb = tb;
     ctx.env = env;
     ctx.pc = pc_start;
-    ctx.amask = env->amask;
-#if defined (CONFIG_USER_ONLY)
-    ctx.mem_idx = 0;
-#else
-    ctx.mem_idx = ((env->ps >> 3) & 3);
-    ctx.pal_mode = env->ipr[IPR_EXC_ADDR] & 1;
-#endif
+    ctx.mem_idx = cpu_mmu_index(env);
 
     /* ??? Every TB begins with unset rounding mode, to be initialized on
        the first fp insn of the TB.  Alternately we could define a proper
@@ -3211,18 +3352,15 @@
         ctx.pc += 4;
         ret = translate_one(ctxp, insn);
 
-        if (ret == NO_EXIT) {
-            /* If we reach a page boundary, are single stepping,
-               or exhaust instruction count, stop generation.  */
-            if (env->singlestep_enabled) {
-                gen_excp(&ctx, EXCP_DEBUG, 0);
-                ret = EXIT_PC_UPDATED;
-            } else if ((ctx.pc & (TARGET_PAGE_SIZE - 1)) == 0
-                       || gen_opc_ptr >= gen_opc_end
-                       || num_insns >= max_insns
-                       || singlestep) {
-                ret = EXIT_PC_STALE;
-            }
+        /* If we reach a page boundary, are single stepping,
+           or exhaust instruction count, stop generation.  */
+        if (ret == NO_EXIT
+            && ((ctx.pc & (TARGET_PAGE_SIZE - 1)) == 0
+                || gen_opc_ptr >= gen_opc_end
+                || num_insns >= max_insns
+                || singlestep
+                || env->singlestep_enabled)) {
+            ret = EXIT_PC_STALE;
         }
     } while (ret == NO_EXIT);
 
@@ -3238,7 +3376,11 @@
         tcg_gen_movi_i64(cpu_pc, ctx.pc);
         /* FALLTHRU */
     case EXIT_PC_UPDATED:
-        tcg_gen_exit_tb(0);
+        if (env->singlestep_enabled) {
+            gen_excp_1(EXCP_DEBUG, 0);
+        } else {
+            tcg_gen_exit_tb(0);
+        }
         break;
     default:
         abort();
@@ -3325,43 +3467,13 @@
     env->implver = implver;
     env->amask = amask;
 
-    env->ps = 0x1F00;
 #if defined (CONFIG_USER_ONLY)
-    env->ps |= 1 << 3;
+    env->ps = PS_USER_MODE;
     cpu_alpha_store_fpcr(env, (FPCR_INVD | FPCR_DZED | FPCR_OVFD
                                | FPCR_UNFD | FPCR_INED | FPCR_DNOD));
-#else
-    pal_init(env);
 #endif
     env->lock_addr = -1;
-
-    /* Initialize IPR */
-#if defined (CONFIG_USER_ONLY)
-    env->ipr[IPR_EXC_ADDR] = 0;
-    env->ipr[IPR_EXC_SUM] = 0;
-    env->ipr[IPR_EXC_MASK] = 0;
-#else
-    {
-        // uint64_t hwpcb;
-        // hwpcb = env->ipr[IPR_PCBB];
-        env->ipr[IPR_ASN] = 0;
-        env->ipr[IPR_ASTEN] = 0;
-        env->ipr[IPR_ASTSR] = 0;
-        env->ipr[IPR_DATFX] = 0;
-        /* XXX: fix this */
-        //    env->ipr[IPR_ESP] = ldq_raw(hwpcb + 8);
-        //    env->ipr[IPR_KSP] = ldq_raw(hwpcb + 0);
-        //    env->ipr[IPR_SSP] = ldq_raw(hwpcb + 16);
-        //    env->ipr[IPR_USP] = ldq_raw(hwpcb + 24);
-        env->ipr[IPR_FEN] = 0;
-        env->ipr[IPR_IPL] = 31;
-        env->ipr[IPR_MCES] = 0;
-        env->ipr[IPR_PERFMON] = 0; /* Implementation specific */
-        //    env->ipr[IPR_PTBR] = ldq_raw(hwpcb + 32);
-        env->ipr[IPR_SISR] = 0;
-        env->ipr[IPR_VIRBND] = -1ULL;
-    }
-#endif
+    env->fen = 1;
 
     qemu_init_vcpu(env);
     return env;
diff --git a/target-arm/helper.c b/target-arm/helper.c
index 1cc492d..1208416 100644
--- a/target-arm/helper.c
+++ b/target-arm/helper.c
@@ -848,6 +848,7 @@
                 return;
             }
         }
+        env->cp15.c5_insn = 2;
         /* Fall through to prefetch abort.  */
     case EXCP_PREFETCH_ABORT:
         new_mode = ARM_CPU_MODE_ABT;
diff --git a/target-arm/translate.c b/target-arm/translate.c
index 1501db1..f5507ec 100644
--- a/target-arm/translate.c
+++ b/target-arm/translate.c
@@ -980,20 +980,20 @@
 #define VFP_GEN_ITOF(name) \
 static inline void gen_vfp_##name(int dp, int neon) \
 { \
-    TCGv statusptr = tcg_temp_new_i32(); \
+    TCGv_ptr statusptr = tcg_temp_new_ptr(); \
     int offset; \
     if (neon) { \
         offset = offsetof(CPUState, vfp.standard_fp_status); \
     } else { \
         offset = offsetof(CPUState, vfp.fp_status); \
     } \
-    tcg_gen_addi_i32(statusptr, cpu_env, offset); \
+    tcg_gen_addi_ptr(statusptr, cpu_env, offset); \
     if (dp) { \
         gen_helper_vfp_##name##d(cpu_F0d, cpu_F0s, statusptr); \
     } else { \
         gen_helper_vfp_##name##s(cpu_F0s, cpu_F0s, statusptr); \
     } \
-    tcg_temp_free_i32(statusptr); \
+    tcg_temp_free_ptr(statusptr); \
 }
 
 VFP_GEN_ITOF(uito)
@@ -1003,20 +1003,20 @@
 #define VFP_GEN_FTOI(name) \
 static inline void gen_vfp_##name(int dp, int neon) \
 { \
-    TCGv statusptr = tcg_temp_new_i32(); \
+    TCGv_ptr statusptr = tcg_temp_new_ptr(); \
     int offset; \
     if (neon) { \
         offset = offsetof(CPUState, vfp.standard_fp_status); \
     } else { \
         offset = offsetof(CPUState, vfp.fp_status); \
     } \
-    tcg_gen_addi_i32(statusptr, cpu_env, offset); \
+    tcg_gen_addi_ptr(statusptr, cpu_env, offset); \
     if (dp) { \
         gen_helper_vfp_##name##d(cpu_F0s, cpu_F0d, statusptr); \
     } else { \
         gen_helper_vfp_##name##s(cpu_F0s, cpu_F0s, statusptr); \
     } \
-    tcg_temp_free_i32(statusptr); \
+    tcg_temp_free_ptr(statusptr); \
 }
 
 VFP_GEN_FTOI(toui)
@@ -1029,21 +1029,21 @@
 static inline void gen_vfp_##name(int dp, int shift, int neon) \
 { \
     TCGv tmp_shift = tcg_const_i32(shift); \
-    TCGv statusptr = tcg_temp_new_i32(); \
+    TCGv_ptr statusptr = tcg_temp_new_ptr(); \
     int offset; \
     if (neon) { \
         offset = offsetof(CPUState, vfp.standard_fp_status); \
     } else { \
         offset = offsetof(CPUState, vfp.fp_status); \
     } \
-    tcg_gen_addi_i32(statusptr, cpu_env, offset); \
+    tcg_gen_addi_ptr(statusptr, cpu_env, offset); \
     if (dp) { \
         gen_helper_vfp_##name##d(cpu_F0d, cpu_F0d, tmp_shift, statusptr); \
     } else { \
         gen_helper_vfp_##name##s(cpu_F0s, cpu_F0s, tmp_shift, statusptr); \
     } \
     tcg_temp_free_i32(tmp_shift); \
-    tcg_temp_free_i32(statusptr); \
+    tcg_temp_free_ptr(statusptr); \
 }
 VFP_GEN_FIX(tosh)
 VFP_GEN_FIX(tosl)
diff --git a/target-i386/cpu.h b/target-i386/cpu.h
index 715828f..9c3340d 100644
--- a/target-i386/cpu.h
+++ b/target-i386/cpu.h
@@ -438,9 +438,13 @@
 #define CPUID_VENDOR_INTEL_3 0x6c65746e /* "ntel" */
 
 #define CPUID_VENDOR_AMD_1   0x68747541 /* "Auth" */
-#define CPUID_VENDOR_AMD_2   0x69746e65 /* "enti" */ 
+#define CPUID_VENDOR_AMD_2   0x69746e65 /* "enti" */
 #define CPUID_VENDOR_AMD_3   0x444d4163 /* "cAMD" */
 
+#define CPUID_VENDOR_VIA_1   0x746e6543 /* "Cent" */
+#define CPUID_VENDOR_VIA_2   0x48727561 /* "aurH" */
+#define CPUID_VENDOR_VIA_3   0x736c7561 /* "auls" */
+
 #define CPUID_MWAIT_IBE     (1 << 1) /* Interrupts can exit capability */
 #define CPUID_MWAIT_EMX     (1 << 0) /* enumeration supported */
 
@@ -532,16 +536,6 @@
     CC_OP_NB,
 };
 
-#ifdef FLOATX80
-#define USE_X86LDOUBLE
-#endif
-
-#ifdef USE_X86LDOUBLE
-typedef floatx80 CPU86_LDouble;
-#else
-typedef float64 CPU86_LDouble;
-#endif
-
 typedef struct SegmentCache {
     uint32_t selector;
     target_ulong base;
@@ -594,11 +588,7 @@
 #define MMX_Q(n) q
 
 typedef union {
-#ifdef USE_X86LDOUBLE
-    CPU86_LDouble d __attribute__((aligned(16)));
-#else
-    CPU86_LDouble d;
-#endif
+    floatx80 d __attribute__((aligned(16)));
     MMXReg mmx;
 } FPReg;
 
@@ -654,7 +644,7 @@
 
     /* emulator internal variables */
     float_status fp_status;
-    CPU86_LDouble ft0;
+    floatx80 ft0;
 
     float_status mmx_status; /* for 3DNow! float ops */
     float_status sse_status;
@@ -730,6 +720,9 @@
     uint32_t cpuid_ext3_features;
     uint32_t cpuid_apic_id;
     int cpuid_vendor_override;
+    /* Store the results of Centaur's CPUID instructions */
+    uint32_t cpuid_xlevel2;
+    uint32_t cpuid_ext4_features;
 
     /* MTRRs */
     uint64_t mtrr_fixed[11];
@@ -865,8 +858,8 @@
 
 /* op_helper.c */
 /* used for debug or cpu save/restore */
-void cpu_get_fp80(uint64_t *pmant, uint16_t *pexp, CPU86_LDouble f);
-CPU86_LDouble cpu_set_fp80(uint64_t mant, uint16_t upper);
+void cpu_get_fp80(uint64_t *pmant, uint16_t *pexp, floatx80 f);
+floatx80 cpu_set_fp80(uint64_t mant, uint16_t upper);
 
 /* cpu-exec.c */
 /* the following helpers are only usable in user mode simulation as
diff --git a/target-i386/cpuid.c b/target-i386/cpuid.c
index e479a4d..79e7580 100644
--- a/target-i386/cpuid.c
+++ b/target-i386/cpuid.c
@@ -230,6 +230,9 @@
     char model_id[48];
     int vendor_override;
     uint32_t flags;
+    /* Store the results of Centaur's CPUID instructions */
+    uint32_t ext4_features;
+    uint32_t xlevel2;
 } x86_def_t;
 
 #define I486_FEATURES (CPUID_FP87 | CPUID_VME | CPUID_PSE)
@@ -522,6 +525,18 @@
     cpu_x86_fill_model_id(x86_cpu_def->model_id);
     x86_cpu_def->vendor_override = 0;
 
+    /* Call Centaur's CPUID instruction. */
+    if (x86_cpu_def->vendor1 == CPUID_VENDOR_VIA_1 &&
+        x86_cpu_def->vendor2 == CPUID_VENDOR_VIA_2 &&
+        x86_cpu_def->vendor3 == CPUID_VENDOR_VIA_3) {
+        host_cpuid(0xC0000000, 0, &eax, &ebx, &ecx, &edx);
+        if (eax >= 0xC0000001) {
+            /* Support VIA max extended level */
+            x86_cpu_def->xlevel2 = eax;
+            host_cpuid(0xC0000001, 0, &eax, &ebx, &ecx, &edx);
+            x86_cpu_def->ext4_features = edx;
+        }
+    }
 
     /*
      * Every SVM feature requires emulation support in KVM - so we can't just
@@ -855,6 +870,8 @@
     env->cpuid_xlevel = def->xlevel;
     env->cpuid_kvm_features = def->kvm_features;
     env->cpuid_svm_features = def->svm_features;
+    env->cpuid_ext4_features = def->ext4_features;
+    env->cpuid_xlevel2 = def->xlevel2;
     if (!kvm_enabled()) {
         env->cpuid_features &= TCG_FEATURES;
         env->cpuid_ext_features &= TCG_EXT_FEATURES;
@@ -1035,8 +1052,18 @@
 {
     /* test if maximum index reached */
     if (index & 0x80000000) {
-        if (index > env->cpuid_xlevel)
-            index = env->cpuid_level;
+        if (index > env->cpuid_xlevel) {
+            if (env->cpuid_xlevel2 > 0) {
+                /* Handle the Centaur's CPUID instruction. */
+                if (index > env->cpuid_xlevel2) {
+                    index = env->cpuid_xlevel2;
+                } else if (index < 0xC0000000) {
+                    index = env->cpuid_xlevel;
+                }
+            } else {
+                index =  env->cpuid_xlevel;
+            }
+        }
     } else {
         if (index > env->cpuid_level)
             index = env->cpuid_level;
@@ -1115,6 +1142,19 @@
         *ecx = 0;
         *edx = 0;
         break;
+    case 7:
+        if (kvm_enabled()) {
+            *eax = kvm_arch_get_supported_cpuid(env, 0x7, count, R_EAX);
+            *ebx = kvm_arch_get_supported_cpuid(env, 0x7, count, R_EBX);
+            *ecx = kvm_arch_get_supported_cpuid(env, 0x7, count, R_ECX);
+            *edx = kvm_arch_get_supported_cpuid(env, 0x7, count, R_EDX);
+        } else {
+            *eax = 0;
+            *ebx = 0;
+            *ecx = 0;
+            *edx = 0;
+        }
+        break;
     case 9:
         /* Direct Cache Access Information Leaf */
         *eax = 0; /* Bits 0-31 in DCA_CAP MSR */
@@ -1231,6 +1271,28 @@
 		*edx = 0;
 	}
         break;
+    case 0xC0000000:
+        *eax = env->cpuid_xlevel2;
+        *ebx = 0;
+        *ecx = 0;
+        *edx = 0;
+        break;
+    case 0xC0000001:
+        /* Support for VIA CPU's CPUID instruction */
+        *eax = env->cpuid_version;
+        *ebx = 0;
+        *ecx = 0;
+        *edx = env->cpuid_ext4_features;
+        break;
+    case 0xC0000002:
+    case 0xC0000003:
+    case 0xC0000004:
+        /* Reserved for the future, and now filled with zero */
+        *eax = 0;
+        *ebx = 0;
+        *ecx = 0;
+        *edx = 0;
+        break;
     default:
         /* reserved values: zero */
         *eax = 0;
diff --git a/target-i386/exec.h b/target-i386/exec.h
index ee36a71..9bd080e 100644
--- a/target-i386/exec.h
+++ b/target-i386/exec.h
@@ -98,67 +98,6 @@
 
 #endif /* !defined(CONFIG_USER_ONLY) */
 
-#ifdef USE_X86LDOUBLE
-/* use long double functions */
-#define floatx_to_int32 floatx80_to_int32
-#define floatx_to_int64 floatx80_to_int64
-#define floatx_to_int32_round_to_zero floatx80_to_int32_round_to_zero
-#define floatx_to_int64_round_to_zero floatx80_to_int64_round_to_zero
-#define int32_to_floatx int32_to_floatx80
-#define int64_to_floatx int64_to_floatx80
-#define float32_to_floatx float32_to_floatx80
-#define float64_to_floatx float64_to_floatx80
-#define floatx_to_float32 floatx80_to_float32
-#define floatx_to_float64 floatx80_to_float64
-#define floatx_add floatx80_add
-#define floatx_div floatx80_div
-#define floatx_mul floatx80_mul
-#define floatx_sub floatx80_sub
-#define floatx_sqrt floatx80_sqrt
-#define floatx_abs floatx80_abs
-#define floatx_chs floatx80_chs
-#define floatx_scalbn floatx80_scalbn
-#define floatx_round_to_int floatx80_round_to_int
-#define floatx_compare floatx80_compare
-#define floatx_compare_quiet floatx80_compare_quiet
-#define floatx_is_any_nan floatx80_is_any_nan
-#define floatx_is_neg floatx80_is_neg
-#define floatx_is_zero floatx80_is_zero
-#define floatx_zero floatx80_zero
-#define floatx_one floatx80_one
-#define floatx_ln2 floatx80_ln2
-#define floatx_pi floatx80_pi
-#else
-#define floatx_to_int32 float64_to_int32
-#define floatx_to_int64 float64_to_int64
-#define floatx_to_int32_round_to_zero float64_to_int32_round_to_zero
-#define floatx_to_int64_round_to_zero float64_to_int64_round_to_zero
-#define int32_to_floatx int32_to_float64
-#define int64_to_floatx int64_to_float64
-#define float32_to_floatx float32_to_float64
-#define float64_to_floatx(x, e) (x)
-#define floatx_to_float32 float64_to_float32
-#define floatx_to_float64(x, e) (x)
-#define floatx_add float64_add
-#define floatx_div float64_div
-#define floatx_mul float64_mul
-#define floatx_sub float64_sub
-#define floatx_sqrt float64_sqrt
-#define floatx_abs float64_abs
-#define floatx_chs float64_chs
-#define floatx_scalbn float64_scalbn
-#define floatx_round_to_int float64_round_to_int
-#define floatx_compare float64_compare
-#define floatx_compare_quiet float64_compare_quiet
-#define floatx_is_any_nan float64_is_any_nan
-#define floatx_is_neg float64_is_neg
-#define floatx_is_zero float64_is_zero
-#define floatx_zero float64_zero
-#define floatx_one float64_one
-#define floatx_ln2 float64_ln2
-#define floatx_pi float64_pi
-#endif
-
 #define RC_MASK         0xc00
 #define RC_NEAR		0x000
 #define RC_DOWN		0x400
@@ -167,11 +106,6 @@
 
 #define MAXTAN 9223372036854775808.0
 
-#ifdef USE_X86LDOUBLE
-
-/* only for x86 */
-typedef CPU_LDoubleU CPU86_LDoubleU;
-
 /* the following deal with x86 long double-precision numbers */
 #define MAXEXPD 0x7fff
 #define EXPBIAS 16383
@@ -180,23 +114,6 @@
 #define MANTD(fp)       (fp.l.lower)
 #define BIASEXPONENT(fp) fp.l.upper = (fp.l.upper & ~(0x7fff)) | EXPBIAS
 
-#else
-
-typedef CPU_DoubleU CPU86_LDoubleU;
-
-/* the following deal with IEEE double-precision numbers */
-#define MAXEXPD 0x7ff
-#define EXPBIAS 1023
-#define EXPD(fp)	(((fp.l.upper) >> 20) & 0x7FF)
-#define SIGND(fp)	((fp.l.upper) & 0x80000000)
-#ifdef __arm__
-#define MANTD(fp)	(fp.l.lower | ((uint64_t)(fp.l.upper & ((1 << 20) - 1)) << 32))
-#else
-#define MANTD(fp)	(fp.ll & ((1LL << 52) - 1))
-#endif
-#define BIASEXPONENT(fp) fp.l.upper = (fp.l.upper & ~(0x7ff << 20)) | (EXPBIAS << 20)
-#endif
-
 static inline void fpush(void)
 {
     env->fpstt = (env->fpstt - 1) & 7;
@@ -209,65 +126,24 @@
     env->fpstt = (env->fpstt + 1) & 7;
 }
 
-#ifndef USE_X86LDOUBLE
-static inline CPU86_LDouble helper_fldt(target_ulong ptr)
+static inline floatx80 helper_fldt(target_ulong ptr)
 {
-    CPU86_LDoubleU temp;
-    int upper, e;
-    uint64_t ll;
-
-    /* mantissa */
-    upper = lduw(ptr + 8);
-    /* XXX: handle overflow ? */
-    e = (upper & 0x7fff) - 16383 + EXPBIAS; /* exponent */
-    e |= (upper >> 4) & 0x800; /* sign */
-    ll = (ldq(ptr) >> 11) & ((1LL << 52) - 1);
-#ifdef __arm__
-    temp.l.upper = (e << 20) | (ll >> 32);
-    temp.l.lower = ll;
-#else
-    temp.ll = ll | ((uint64_t)e << 52);
-#endif
-    return temp.d;
-}
-
-static inline void helper_fstt(CPU86_LDouble f, target_ulong ptr)
-{
-    CPU86_LDoubleU temp;
-    int e;
-
-    temp.d = f;
-    /* mantissa */
-    stq(ptr, (MANTD(temp) << 11) | (1LL << 63));
-    /* exponent + sign */
-    e = EXPD(temp) - EXPBIAS + 16383;
-    e |= SIGND(temp) >> 16;
-    stw(ptr + 8, e);
-}
-#else
-
-/* we use memory access macros */
-
-static inline CPU86_LDouble helper_fldt(target_ulong ptr)
-{
-    CPU86_LDoubleU temp;
+    CPU_LDoubleU temp;
 
     temp.l.lower = ldq(ptr);
     temp.l.upper = lduw(ptr + 8);
     return temp.d;
 }
 
-static inline void helper_fstt(CPU86_LDouble f, target_ulong ptr)
+static inline void helper_fstt(floatx80 f, target_ulong ptr)
 {
-    CPU86_LDoubleU temp;
+    CPU_LDoubleU temp;
 
     temp.d = f;
     stq(ptr, temp.l.lower);
     stw(ptr + 8, temp.l.upper);
 }
 
-#endif /* USE_X86LDOUBLE */
-
 #define FPUS_IE (1 << 0)
 #define FPUS_DE (1 << 1)
 #define FPUS_ZE (1 << 2)
diff --git a/target-i386/helper.c b/target-i386/helper.c
index 89df997..509d68c 100644
--- a/target-i386/helper.c
+++ b/target-i386/helper.c
@@ -21,7 +21,6 @@
 #include <stdio.h>
 #include <string.h>
 #include <inttypes.h>
-#include <signal.h>
 
 #include "cpu.h"
 #include "exec-all.h"
@@ -403,15 +402,10 @@
                     fptag,
                     env->mxcsr);
         for(i=0;i<8;i++) {
-#if defined(USE_X86LDOUBLE)
             CPU_LDoubleU u;
             u.d = env->fpregs[i].d;
             cpu_fprintf(f, "FPR%d=%016" PRIx64 " %04x",
                         i, u.l.lower, u.l.upper);
-#else
-            cpu_fprintf(f, "FPR%d=%016" PRIx64,
-                        i, env->fpregs[i].mmx.q);
-#endif
             if ((i & 1) == 1)
                 cpu_fprintf(f, "\n");
             else
diff --git a/target-i386/kvm.c b/target-i386/kvm.c
index faedc6c..1ae2d61 100644
--- a/target-i386/kvm.c
+++ b/target-i386/kvm.c
@@ -482,6 +482,21 @@
         cpu_x86_cpuid(env, i, 0, &c->eax, &c->ebx, &c->ecx, &c->edx);
     }
 
+    /* Call Centaur's CPUID instructions they are supported. */
+    if (env->cpuid_xlevel2 > 0) {
+        env->cpuid_ext4_features &=
+            kvm_arch_get_supported_cpuid(env, 0xC0000001, 0, R_EDX);
+        cpu_x86_cpuid(env, 0xC0000000, 0, &limit, &unused, &unused, &unused);
+
+        for (i = 0xC0000000; i <= limit; i++) {
+            c = &cpuid_data.entries[cpuid_i++];
+
+            c->function = i;
+            c->flags = 0;
+            cpu_x86_cpuid(env, i, 0, &c->eax, &c->ebx, &c->ecx, &c->edx);
+        }
+    }
+
     cpuid_data.cpuid.nent = cpuid_i;
 
 #ifdef KVM_CAP_MCE
diff --git a/target-i386/machine.c b/target-i386/machine.c
index d78eceb..bbeae88 100644
--- a/target-i386/machine.c
+++ b/target-i386/machine.c
@@ -84,7 +84,6 @@
     exit(0);
 }
 
-#ifdef USE_X86LDOUBLE
 /* XXX: add that in a FPU generic layer */
 union x86_longdouble {
     uint64_t mant;
@@ -202,102 +201,6 @@
     VMSTATE_ARRAY_TEST(_field, _state, _n, fpregs_is_1_mmx, vmstate_fpreg_1_mmx, FPReg), \
     VMSTATE_ARRAY_TEST(_field, _state, _n, fpregs_is_1_no_mmx, vmstate_fpreg_1_no_mmx, FPReg)
 
-#else
-static int get_fpreg(QEMUFile *f, void *opaque, size_t size)
-{
-    FPReg *fp_reg = opaque;
-
-    qemu_get_be64s(f, &fp_reg->mmx.MMX_Q(0));
-    return 0;
-}
-
-static void put_fpreg(QEMUFile *f, void *opaque, size_t size)
-{
-    FPReg *fp_reg = opaque;
-    /* if we use doubles for float emulation, we save the doubles to
-       avoid losing information in case of MMX usage. It can give
-       problems if the image is restored on a CPU where long
-       doubles are used instead. */
-    qemu_put_be64s(f, &fp_reg->mmx.MMX_Q(0));
-}
-
-const VMStateInfo vmstate_fpreg = {
-    .name = "fpreg",
-    .get  = get_fpreg,
-    .put  = put_fpreg,
-};
-
-static int get_fpreg_0_mmx(QEMUFile *f, void *opaque, size_t size)
-{
-    FPReg *fp_reg = opaque;
-    uint64_t mant;
-    uint16_t exp;
-
-    qemu_get_be64s(f, &mant);
-    qemu_get_be16s(f, &exp);
-    fp_reg->mmx.MMX_Q(0) = mant;
-    return 0;
-}
-
-const VMStateInfo vmstate_fpreg_0_mmx = {
-    .name = "fpreg_0_mmx",
-    .get  = get_fpreg_0_mmx,
-    .put  = put_fpreg_error,
-};
-
-static int get_fpreg_0_no_mmx(QEMUFile *f, void *opaque, size_t size)
-{
-    FPReg *fp_reg = opaque;
-    uint64_t mant;
-    uint16_t exp;
-
-    qemu_get_be64s(f, &mant);
-    qemu_get_be16s(f, &exp);
-
-    fp_reg->d = cpu_set_fp80(mant, exp);
-    return 0;
-}
-
-const VMStateInfo vmstate_fpreg_0_no_mmx = {
-    .name = "fpreg_0_no_mmx",
-    .get  = get_fpreg_0_no_mmx,
-    .put  = put_fpreg_error,
-};
-
-static bool fpregs_is_1(void *opaque, int version_id)
-{
-    CPUState *env = opaque;
-
-    return env->fpregs_format_vmstate == 1;
-}
-
-static bool fpregs_is_0_mmx(void *opaque, int version_id)
-{
-    CPUState *env = opaque;
-    int guess_mmx;
-
-    guess_mmx = ((env->fptag_vmstate == 0xff) &&
-                 (env->fpus_vmstate & 0x3800) == 0);
-    return guess_mmx && env->fpregs_format_vmstate == 0;
-}
-
-static bool fpregs_is_0_no_mmx(void *opaque, int version_id)
-{
-    CPUState *env = opaque;
-    int guess_mmx;
-
-    guess_mmx = ((env->fptag_vmstate == 0xff) &&
-                 (env->fpus_vmstate & 0x3800) == 0);
-    return !guess_mmx && env->fpregs_format_vmstate == 0;
-}
-
-#define VMSTATE_FP_REGS(_field, _state, _n)                               \
-    VMSTATE_ARRAY_TEST(_field, _state, _n, fpregs_is_1, vmstate_fpreg, FPReg), \
-    VMSTATE_ARRAY_TEST(_field, _state, _n, fpregs_is_0_mmx, vmstate_fpreg_0_mmx, FPReg), \
-    VMSTATE_ARRAY_TEST(_field, _state, _n, fpregs_is_0_no_mmx, vmstate_fpreg_0_no_mmx, FPReg)
-
-#endif /* USE_X86LDOUBLE */
-
 static bool version_is_5(void *opaque, int version_id)
 {
     return version_id == 5;
@@ -344,11 +247,7 @@
         env->fptag_vmstate |= ((!env->fptags[i]) << i);
     }
 
-#ifdef USE_X86LDOUBLE
     env->fpregs_format_vmstate = 0;
-#else
-    env->fpregs_format_vmstate = 1;
-#endif
 }
 
 static int cpu_post_load(void *opaque, int version_id)
diff --git a/target-i386/op_helper.c b/target-i386/op_helper.c
index 3c539f3..cec0c76 100644
--- a/target-i386/op_helper.c
+++ b/target-i386/op_helper.c
@@ -95,26 +95,9 @@
     6, 7, 8, 0, 1, 2, 3, 4,
 };
 
-#if defined(CONFIG_SOFTFLOAT)
-# define floatx_lg2 make_floatx80( 0x3ffd, 0x9a209a84fbcff799LL )
-# define floatx_l2e make_floatx80( 0x3fff, 0xb8aa3b295c17f0bcLL )
-# define floatx_l2t make_floatx80( 0x4000, 0xd49a784bcd1b8afeLL )
-#else
-# define floatx_lg2 (0.30102999566398119523L)
-# define floatx_l2e (1.44269504088896340739L)
-# define floatx_l2t (3.32192809488736234781L)
-#endif
-
-static const CPU86_LDouble f15rk[7] =
-{
-    floatx_zero,
-    floatx_one,
-    floatx_pi,
-    floatx_lg2,
-    floatx_ln2,
-    floatx_l2e,
-    floatx_l2t,
-};
+#define floatx80_lg2 make_floatx80( 0x3ffd, 0x9a209a84fbcff799LL )
+#define floatx80_l2e make_floatx80( 0x3fff, 0xb8aa3b295c17f0bcLL )
+#define floatx80_l2t make_floatx80( 0x4000, 0xd49a784bcd1b8afeLL )
 
 /* broken thread support */
 
@@ -3442,18 +3425,18 @@
 
 /* x87 FPU helpers */
 
-static inline double CPU86_LDouble_to_double(CPU86_LDouble a)
+static inline double floatx80_to_double(floatx80 a)
 {
     union {
         float64 f64;
         double d;
     } u;
 
-    u.f64 = floatx_to_float64(a, &env->fp_status);
+    u.f64 = floatx80_to_float64(a, &env->fp_status);
     return u.d;
 }
 
-static inline CPU86_LDouble double_to_CPU86_LDouble(double a)
+static inline floatx80 double_to_floatx80(double a)
 {
     union {
         float64 f64;
@@ -3461,7 +3444,7 @@
     } u;
 
     u.d = a;
-    return float64_to_floatx(u.f64, &env->fp_status);
+    return float64_to_floatx80(u.f64, &env->fp_status);
 }
 
 static void fpu_set_exception(int mask)
@@ -3471,12 +3454,12 @@
         env->fpus |= FPUS_SE | FPUS_B;
 }
 
-static inline CPU86_LDouble helper_fdiv(CPU86_LDouble a, CPU86_LDouble b)
+static inline floatx80 helper_fdiv(floatx80 a, floatx80 b)
 {
-    if (floatx_is_zero(b)) {
+    if (floatx80_is_zero(b)) {
         fpu_set_exception(FPUS_ZE);
     }
-    return floatx_div(a, b, &env->fp_status);
+    return floatx80_div(a, b, &env->fp_status);
 }
 
 static void fpu_raise_exception(void)
@@ -3498,7 +3481,7 @@
         uint32_t i;
     } u;
     u.i = val;
-    FT0 = float32_to_floatx(u.f, &env->fp_status);
+    FT0 = float32_to_floatx80(u.f, &env->fp_status);
 }
 
 void helper_fldl_FT0(uint64_t val)
@@ -3508,12 +3491,12 @@
         uint64_t i;
     } u;
     u.i = val;
-    FT0 = float64_to_floatx(u.f, &env->fp_status);
+    FT0 = float64_to_floatx80(u.f, &env->fp_status);
 }
 
 void helper_fildl_FT0(int32_t val)
 {
-    FT0 = int32_to_floatx(val, &env->fp_status);
+    FT0 = int32_to_floatx80(val, &env->fp_status);
 }
 
 void helper_flds_ST0(uint32_t val)
@@ -3525,7 +3508,7 @@
     } u;
     new_fpstt = (env->fpstt - 1) & 7;
     u.i = val;
-    env->fpregs[new_fpstt].d = float32_to_floatx(u.f, &env->fp_status);
+    env->fpregs[new_fpstt].d = float32_to_floatx80(u.f, &env->fp_status);
     env->fpstt = new_fpstt;
     env->fptags[new_fpstt] = 0; /* validate stack entry */
 }
@@ -3539,7 +3522,7 @@
     } u;
     new_fpstt = (env->fpstt - 1) & 7;
     u.i = val;
-    env->fpregs[new_fpstt].d = float64_to_floatx(u.f, &env->fp_status);
+    env->fpregs[new_fpstt].d = float64_to_floatx80(u.f, &env->fp_status);
     env->fpstt = new_fpstt;
     env->fptags[new_fpstt] = 0; /* validate stack entry */
 }
@@ -3548,7 +3531,7 @@
 {
     int new_fpstt;
     new_fpstt = (env->fpstt - 1) & 7;
-    env->fpregs[new_fpstt].d = int32_to_floatx(val, &env->fp_status);
+    env->fpregs[new_fpstt].d = int32_to_floatx80(val, &env->fp_status);
     env->fpstt = new_fpstt;
     env->fptags[new_fpstt] = 0; /* validate stack entry */
 }
@@ -3557,7 +3540,7 @@
 {
     int new_fpstt;
     new_fpstt = (env->fpstt - 1) & 7;
-    env->fpregs[new_fpstt].d = int64_to_floatx(val, &env->fp_status);
+    env->fpregs[new_fpstt].d = int64_to_floatx80(val, &env->fp_status);
     env->fpstt = new_fpstt;
     env->fptags[new_fpstt] = 0; /* validate stack entry */
 }
@@ -3568,7 +3551,7 @@
         float32 f;
         uint32_t i;
     } u;
-    u.f = floatx_to_float32(ST0, &env->fp_status);
+    u.f = floatx80_to_float32(ST0, &env->fp_status);
     return u.i;
 }
 
@@ -3578,14 +3561,14 @@
         float64 f;
         uint64_t i;
     } u;
-    u.f = floatx_to_float64(ST0, &env->fp_status);
+    u.f = floatx80_to_float64(ST0, &env->fp_status);
     return u.i;
 }
 
 int32_t helper_fist_ST0(void)
 {
     int32_t val;
-    val = floatx_to_int32(ST0, &env->fp_status);
+    val = floatx80_to_int32(ST0, &env->fp_status);
     if (val != (int16_t)val)
         val = -32768;
     return val;
@@ -3594,21 +3577,21 @@
 int32_t helper_fistl_ST0(void)
 {
     int32_t val;
-    val = floatx_to_int32(ST0, &env->fp_status);
+    val = floatx80_to_int32(ST0, &env->fp_status);
     return val;
 }
 
 int64_t helper_fistll_ST0(void)
 {
     int64_t val;
-    val = floatx_to_int64(ST0, &env->fp_status);
+    val = floatx80_to_int64(ST0, &env->fp_status);
     return val;
 }
 
 int32_t helper_fistt_ST0(void)
 {
     int32_t val;
-    val = floatx_to_int32_round_to_zero(ST0, &env->fp_status);
+    val = floatx80_to_int32_round_to_zero(ST0, &env->fp_status);
     if (val != (int16_t)val)
         val = -32768;
     return val;
@@ -3617,14 +3600,14 @@
 int32_t helper_fisttl_ST0(void)
 {
     int32_t val;
-    val = floatx_to_int32_round_to_zero(ST0, &env->fp_status);
+    val = floatx80_to_int32_round_to_zero(ST0, &env->fp_status);
     return val;
 }
 
 int64_t helper_fisttll_ST0(void)
 {
     int64_t val;
-    val = floatx_to_int64_round_to_zero(ST0, &env->fp_status);
+    val = floatx80_to_int64_round_to_zero(ST0, &env->fp_status);
     return val;
 }
 
@@ -3693,7 +3676,7 @@
 
 void helper_fxchg_ST0_STN(int st_index)
 {
-    CPU86_LDouble tmp;
+    floatx80 tmp;
     tmp = ST(st_index);
     ST(st_index) = ST0;
     ST0 = tmp;
@@ -3707,7 +3690,7 @@
 {
     int ret;
 
-    ret = floatx_compare(ST0, FT0, &env->fp_status);
+    ret = floatx80_compare(ST0, FT0, &env->fp_status);
     env->fpus = (env->fpus & ~0x4500) | fcom_ccval[ret + 1];
 }
 
@@ -3715,7 +3698,7 @@
 {
     int ret;
 
-    ret = floatx_compare_quiet(ST0, FT0, &env->fp_status);
+    ret = floatx80_compare_quiet(ST0, FT0, &env->fp_status);
     env->fpus = (env->fpus & ~0x4500) | fcom_ccval[ret+ 1];
 }
 
@@ -3726,7 +3709,7 @@
     int eflags;
     int ret;
 
-    ret = floatx_compare(ST0, FT0, &env->fp_status);
+    ret = floatx80_compare(ST0, FT0, &env->fp_status);
     eflags = helper_cc_compute_all(CC_OP);
     eflags = (eflags & ~(CC_Z | CC_P | CC_C)) | fcomi_ccval[ret + 1];
     CC_SRC = eflags;
@@ -3737,7 +3720,7 @@
     int eflags;
     int ret;
 
-    ret = floatx_compare_quiet(ST0, FT0, &env->fp_status);
+    ret = floatx80_compare_quiet(ST0, FT0, &env->fp_status);
     eflags = helper_cc_compute_all(CC_OP);
     eflags = (eflags & ~(CC_Z | CC_P | CC_C)) | fcomi_ccval[ret + 1];
     CC_SRC = eflags;
@@ -3745,22 +3728,22 @@
 
 void helper_fadd_ST0_FT0(void)
 {
-    ST0 = floatx_add(ST0, FT0, &env->fp_status);
+    ST0 = floatx80_add(ST0, FT0, &env->fp_status);
 }
 
 void helper_fmul_ST0_FT0(void)
 {
-    ST0 = floatx_mul(ST0, FT0, &env->fp_status);
+    ST0 = floatx80_mul(ST0, FT0, &env->fp_status);
 }
 
 void helper_fsub_ST0_FT0(void)
 {
-    ST0 = floatx_sub(ST0, FT0, &env->fp_status);
+    ST0 = floatx80_sub(ST0, FT0, &env->fp_status);
 }
 
 void helper_fsubr_ST0_FT0(void)
 {
-    ST0 = floatx_sub(FT0, ST0, &env->fp_status);
+    ST0 = floatx80_sub(FT0, ST0, &env->fp_status);
 }
 
 void helper_fdiv_ST0_FT0(void)
@@ -3777,34 +3760,34 @@
 
 void helper_fadd_STN_ST0(int st_index)
 {
-    ST(st_index) = floatx_add(ST(st_index), ST0, &env->fp_status);
+    ST(st_index) = floatx80_add(ST(st_index), ST0, &env->fp_status);
 }
 
 void helper_fmul_STN_ST0(int st_index)
 {
-    ST(st_index) = floatx_mul(ST(st_index), ST0, &env->fp_status);
+    ST(st_index) = floatx80_mul(ST(st_index), ST0, &env->fp_status);
 }
 
 void helper_fsub_STN_ST0(int st_index)
 {
-    ST(st_index) = floatx_sub(ST(st_index), ST0, &env->fp_status);
+    ST(st_index) = floatx80_sub(ST(st_index), ST0, &env->fp_status);
 }
 
 void helper_fsubr_STN_ST0(int st_index)
 {
-    ST(st_index) = floatx_sub(ST0, ST(st_index), &env->fp_status);
+    ST(st_index) = floatx80_sub(ST0, ST(st_index), &env->fp_status);
 }
 
 void helper_fdiv_STN_ST0(int st_index)
 {
-    CPU86_LDouble *p;
+    floatx80 *p;
     p = &ST(st_index);
     *p = helper_fdiv(*p, ST0);
 }
 
 void helper_fdivr_STN_ST0(int st_index)
 {
-    CPU86_LDouble *p;
+    floatx80 *p;
     p = &ST(st_index);
     *p = helper_fdiv(ST0, *p);
 }
@@ -3812,52 +3795,52 @@
 /* misc FPU operations */
 void helper_fchs_ST0(void)
 {
-    ST0 = floatx_chs(ST0);
+    ST0 = floatx80_chs(ST0);
 }
 
 void helper_fabs_ST0(void)
 {
-    ST0 = floatx_abs(ST0);
+    ST0 = floatx80_abs(ST0);
 }
 
 void helper_fld1_ST0(void)
 {
-    ST0 = f15rk[1];
+    ST0 = floatx80_one;
 }
 
 void helper_fldl2t_ST0(void)
 {
-    ST0 = f15rk[6];
+    ST0 = floatx80_l2t;
 }
 
 void helper_fldl2e_ST0(void)
 {
-    ST0 = f15rk[5];
+    ST0 = floatx80_l2e;
 }
 
 void helper_fldpi_ST0(void)
 {
-    ST0 = f15rk[2];
+    ST0 = floatx80_pi;
 }
 
 void helper_fldlg2_ST0(void)
 {
-    ST0 = f15rk[3];
+    ST0 = floatx80_lg2;
 }
 
 void helper_fldln2_ST0(void)
 {
-    ST0 = f15rk[4];
+    ST0 = floatx80_ln2;
 }
 
 void helper_fldz_ST0(void)
 {
-    ST0 = f15rk[0];
+    ST0 = floatx80_zero;
 }
 
 void helper_fldz_FT0(void)
 {
-    FT0 = f15rk[0];
+    FT0 = floatx80_zero;
 }
 
 uint32_t helper_fnstsw(void)
@@ -3891,7 +3874,6 @@
         break;
     }
     set_float_rounding_mode(rnd_type, &env->fp_status);
-#ifdef FLOATX80
     switch((env->fpuc >> 8) & 3) {
     case 0:
         rnd_type = 32;
@@ -3905,7 +3887,6 @@
         break;
     }
     set_floatx80_rounding_precision(rnd_type, &env->fp_status);
-#endif
 }
 
 void helper_fldcw(uint32_t val)
@@ -3944,7 +3925,7 @@
 
 void helper_fbld_ST0(target_ulong ptr)
 {
-    CPU86_LDouble tmp;
+    floatx80 tmp;
     uint64_t val;
     unsigned int v;
     int i;
@@ -3954,9 +3935,9 @@
         v = ldub(ptr + i);
         val = (val * 100) + ((v >> 4) * 10) + (v & 0xf);
     }
-    tmp = int64_to_floatx(val, &env->fp_status);
+    tmp = int64_to_floatx80(val, &env->fp_status);
     if (ldub(ptr + 9) & 0x80) {
-        floatx_chs(tmp);
+        floatx80_chs(tmp);
     }
     fpush();
     ST0 = tmp;
@@ -3968,7 +3949,7 @@
     target_ulong mem_ref, mem_end;
     int64_t val;
 
-    val = floatx_to_int64(ST0, &env->fp_status);
+    val = floatx80_to_int64(ST0, &env->fp_status);
     mem_ref = ptr;
     mem_end = mem_ref + 9;
     if (val < 0) {
@@ -3992,19 +3973,19 @@
 
 void helper_f2xm1(void)
 {
-    double val = CPU86_LDouble_to_double(ST0);
+    double val = floatx80_to_double(ST0);
     val = pow(2.0, val) - 1.0;
-    ST0 = double_to_CPU86_LDouble(val);
+    ST0 = double_to_floatx80(val);
 }
 
 void helper_fyl2x(void)
 {
-    double fptemp = CPU86_LDouble_to_double(ST0);
+    double fptemp = floatx80_to_double(ST0);
 
     if (fptemp>0.0){
         fptemp = log(fptemp)/log(2.0);    /* log2(ST) */
-        fptemp *= CPU86_LDouble_to_double(ST1);
-        ST1 = double_to_CPU86_LDouble(fptemp);
+        fptemp *= floatx80_to_double(ST1);
+        ST1 = double_to_floatx80(fptemp);
         fpop();
     } else {
         env->fpus &= (~0x4700);
@@ -4014,15 +3995,15 @@
 
 void helper_fptan(void)
 {
-    double fptemp = CPU86_LDouble_to_double(ST0);
+    double fptemp = floatx80_to_double(ST0);
 
     if((fptemp > MAXTAN)||(fptemp < -MAXTAN)) {
         env->fpus |= 0x400;
     } else {
         fptemp = tan(fptemp);
-        ST0 = double_to_CPU86_LDouble(fptemp);
+        ST0 = double_to_floatx80(fptemp);
         fpush();
-        ST0 = floatx_one;
+        ST0 = floatx80_one;
         env->fpus &= (~0x400);  /* C2 <-- 0 */
         /* the above code is for  |arg| < 2**52 only */
     }
@@ -4032,21 +4013,21 @@
 {
     double fptemp, fpsrcop;
 
-    fpsrcop = CPU86_LDouble_to_double(ST1);
-    fptemp = CPU86_LDouble_to_double(ST0);
-    ST1 = double_to_CPU86_LDouble(atan2(fpsrcop, fptemp));
+    fpsrcop = floatx80_to_double(ST1);
+    fptemp = floatx80_to_double(ST0);
+    ST1 = double_to_floatx80(atan2(fpsrcop, fptemp));
     fpop();
 }
 
 void helper_fxtract(void)
 {
-    CPU86_LDoubleU temp;
+    CPU_LDoubleU temp;
 
     temp.d = ST0;
 
-    if (floatx_is_zero(ST0)) {
+    if (floatx80_is_zero(ST0)) {
         /* Easy way to generate -inf and raising division by 0 exception */
-        ST0 = floatx_div(floatx_chs(floatx_one), floatx_zero, &env->fp_status);
+        ST0 = floatx80_div(floatx80_chs(floatx80_one), floatx80_zero, &env->fp_status);
         fpush();
         ST0 = temp.d;
     } else {
@@ -4054,7 +4035,7 @@
 
         expdif = EXPD(temp) - EXPBIAS;
         /*DP exponent bias*/
-        ST0 = int32_to_floatx(expdif, &env->fp_status);
+        ST0 = int32_to_floatx80(expdif, &env->fp_status);
         fpush();
         BIASEXPONENT(temp);
         ST0 = temp.d;
@@ -4064,15 +4045,15 @@
 void helper_fprem1(void)
 {
     double st0, st1, dblq, fpsrcop, fptemp;
-    CPU86_LDoubleU fpsrcop1, fptemp1;
+    CPU_LDoubleU fpsrcop1, fptemp1;
     int expdif;
     signed long long int q;
 
-    st0 = CPU86_LDouble_to_double(ST0);
-    st1 = CPU86_LDouble_to_double(ST1);
+    st0 = floatx80_to_double(ST0);
+    st1 = floatx80_to_double(ST1);
 
     if (isinf(st0) || isnan(st0) || isnan(st1) || (st1 == 0.0)) {
-        ST0 = double_to_CPU86_LDouble(0.0 / 0.0); /* NaN */
+        ST0 = double_to_floatx80(0.0 / 0.0); /* NaN */
         env->fpus &= (~0x4700); /* (C3,C2,C1,C0) <-- 0000 */
         return;
     }
@@ -4116,21 +4097,21 @@
                   -(floor(fabs(fpsrcop))) : floor(fpsrcop);
         st0 -= (st1 * fpsrcop * fptemp);
     }
-    ST0 = double_to_CPU86_LDouble(st0);
+    ST0 = double_to_floatx80(st0);
 }
 
 void helper_fprem(void)
 {
     double st0, st1, dblq, fpsrcop, fptemp;
-    CPU86_LDoubleU fpsrcop1, fptemp1;
+    CPU_LDoubleU fpsrcop1, fptemp1;
     int expdif;
     signed long long int q;
 
-    st0 = CPU86_LDouble_to_double(ST0);
-    st1 = CPU86_LDouble_to_double(ST1);
+    st0 = floatx80_to_double(ST0);
+    st1 = floatx80_to_double(ST1);
 
     if (isinf(st0) || isnan(st0) || isnan(st1) || (st1 == 0.0)) {
-       ST0 = double_to_CPU86_LDouble(0.0 / 0.0); /* NaN */
+       ST0 = double_to_floatx80(0.0 / 0.0); /* NaN */
        env->fpus &= (~0x4700); /* (C3,C2,C1,C0) <-- 0000 */
        return;
     }
@@ -4175,17 +4156,17 @@
                   -(floor(fabs(fpsrcop))) : floor(fpsrcop);
         st0 -= (st1 * fpsrcop * fptemp);
     }
-    ST0 = double_to_CPU86_LDouble(st0);
+    ST0 = double_to_floatx80(st0);
 }
 
 void helper_fyl2xp1(void)
 {
-    double fptemp = CPU86_LDouble_to_double(ST0);
+    double fptemp = floatx80_to_double(ST0);
 
     if ((fptemp+1.0)>0.0) {
         fptemp = log(fptemp+1.0) / log(2.0); /* log2(ST+1.0) */
-        fptemp *= CPU86_LDouble_to_double(ST1);
-        ST1 = double_to_CPU86_LDouble(fptemp);
+        fptemp *= floatx80_to_double(ST1);
+        ST1 = double_to_floatx80(fptemp);
         fpop();
     } else {
         env->fpus &= (~0x4700);
@@ -4195,23 +4176,23 @@
 
 void helper_fsqrt(void)
 {
-    if (floatx_is_neg(ST0)) {
+    if (floatx80_is_neg(ST0)) {
         env->fpus &= (~0x4700);  /* (C3,C2,C1,C0) <-- 0000 */
         env->fpus |= 0x400;
     }
-    ST0 = floatx_sqrt(ST0, &env->fp_status);
+    ST0 = floatx80_sqrt(ST0, &env->fp_status);
 }
 
 void helper_fsincos(void)
 {
-    double fptemp = CPU86_LDouble_to_double(ST0);
+    double fptemp = floatx80_to_double(ST0);
 
     if ((fptemp > MAXTAN)||(fptemp < -MAXTAN)) {
         env->fpus |= 0x400;
     } else {
-        ST0 = double_to_CPU86_LDouble(sin(fptemp));
+        ST0 = double_to_floatx80(sin(fptemp));
         fpush();
-        ST0 = double_to_CPU86_LDouble(cos(fptemp));
+        ST0 = double_to_floatx80(cos(fptemp));
         env->fpus &= (~0x400);  /* C2 <-- 0 */
         /* the above code is for  |arg| < 2**63 only */
     }
@@ -4219,27 +4200,27 @@
 
 void helper_frndint(void)
 {
-    ST0 = floatx_round_to_int(ST0, &env->fp_status);
+    ST0 = floatx80_round_to_int(ST0, &env->fp_status);
 }
 
 void helper_fscale(void)
 {
-    if (floatx_is_any_nan(ST1)) {
+    if (floatx80_is_any_nan(ST1)) {
         ST0 = ST1;
     } else {
-        int n = floatx_to_int32_round_to_zero(ST1, &env->fp_status);
-        ST0 = floatx_scalbn(ST0, n, &env->fp_status);
+        int n = floatx80_to_int32_round_to_zero(ST1, &env->fp_status);
+        ST0 = floatx80_scalbn(ST0, n, &env->fp_status);
     }
 }
 
 void helper_fsin(void)
 {
-    double fptemp = CPU86_LDouble_to_double(ST0);
+    double fptemp = floatx80_to_double(ST0);
 
     if ((fptemp > MAXTAN)||(fptemp < -MAXTAN)) {
         env->fpus |= 0x400;
     } else {
-        ST0 = double_to_CPU86_LDouble(sin(fptemp));
+        ST0 = double_to_floatx80(sin(fptemp));
         env->fpus &= (~0x400);  /* C2 <-- 0 */
         /* the above code is for  |arg| < 2**53 only */
     }
@@ -4247,12 +4228,12 @@
 
 void helper_fcos(void)
 {
-    double fptemp = CPU86_LDouble_to_double(ST0);
+    double fptemp = floatx80_to_double(ST0);
 
     if((fptemp > MAXTAN)||(fptemp < -MAXTAN)) {
         env->fpus |= 0x400;
     } else {
-        ST0 = double_to_CPU86_LDouble(cos(fptemp));
+        ST0 = double_to_floatx80(cos(fptemp));
         env->fpus &= (~0x400);  /* C2 <-- 0 */
         /* the above code is for  |arg5 < 2**63 only */
     }
@@ -4260,7 +4241,7 @@
 
 void helper_fxam_ST0(void)
 {
-    CPU86_LDoubleU temp;
+    CPU_LDoubleU temp;
     int expdif;
 
     temp.d = ST0;
@@ -4272,11 +4253,7 @@
     /* XXX: test fptags too */
     expdif = EXPD(temp);
     if (expdif == MAXEXPD) {
-#ifdef USE_X86LDOUBLE
         if (MANTD(temp) == 0x8000000000000000ULL)
-#else
-        if (MANTD(temp) == 0)
-#endif
             env->fpus |=  0x500 /*Infinity*/;
         else
             env->fpus |=  0x100 /*NaN*/;
@@ -4294,7 +4271,7 @@
 {
     int fpus, fptag, exp, i;
     uint64_t mant;
-    CPU86_LDoubleU tmp;
+    CPU_LDoubleU tmp;
 
     fpus = (env->fpus & ~0x3800) | (env->fpstt & 0x7) << 11;
     fptag = 0;
@@ -4310,9 +4287,7 @@
                 /* zero */
 	        fptag |= 1;
 	    } else if (exp == 0 || exp == MAXEXPD
-#ifdef USE_X86LDOUBLE
                        || (mant & (1LL << 63)) == 0
-#endif
                        ) {
                 /* NaNs, infinity, denormal */
                 fptag |= 2;
@@ -4364,7 +4339,7 @@
 
 void helper_fsave(target_ulong ptr, int data32)
 {
-    CPU86_LDouble tmp;
+    floatx80 tmp;
     int i;
 
     helper_fstenv(ptr, data32);
@@ -4392,7 +4367,7 @@
 
 void helper_frstor(target_ulong ptr, int data32)
 {
-    CPU86_LDouble tmp;
+    floatx80 tmp;
     int i;
 
     helper_fldenv(ptr, data32);
@@ -4408,7 +4383,7 @@
 void helper_fxsave(target_ulong ptr, int data64)
 {
     int fpus, fptag, i, nb_xmm_regs;
-    CPU86_LDouble tmp;
+    floatx80 tmp;
     target_ulong addr;
 
     /* The operand must be 16 byte aligned */
@@ -4469,7 +4444,7 @@
 void helper_fxrstor(target_ulong ptr, int data64)
 {
     int i, fpus, fptag, nb_xmm_regs;
-    CPU86_LDouble tmp;
+    floatx80 tmp;
     target_ulong addr;
 
     /* The operand must be 16 byte aligned */
@@ -4516,61 +4491,23 @@
     }
 }
 
-#ifndef USE_X86LDOUBLE
-
-void cpu_get_fp80(uint64_t *pmant, uint16_t *pexp, CPU86_LDouble f)
+void cpu_get_fp80(uint64_t *pmant, uint16_t *pexp, floatx80 f)
 {
-    CPU86_LDoubleU temp;
-    int e;
-
-    temp.d = f;
-    /* mantissa */
-    *pmant = (MANTD(temp) << 11) | (1LL << 63);
-    /* exponent + sign */
-    e = EXPD(temp) - EXPBIAS + 16383;
-    e |= SIGND(temp) >> 16;
-    *pexp = e;
-}
-
-CPU86_LDouble cpu_set_fp80(uint64_t mant, uint16_t upper)
-{
-    CPU86_LDoubleU temp;
-    int e;
-    uint64_t ll;
-
-    /* XXX: handle overflow ? */
-    e = (upper & 0x7fff) - 16383 + EXPBIAS; /* exponent */
-    e |= (upper >> 4) & 0x800; /* sign */
-    ll = (mant >> 11) & ((1LL << 52) - 1);
-#ifdef __arm__
-    temp.l.upper = (e << 20) | (ll >> 32);
-    temp.l.lower = ll;
-#else
-    temp.ll = ll | ((uint64_t)e << 52);
-#endif
-    return temp.d;
-}
-
-#else
-
-void cpu_get_fp80(uint64_t *pmant, uint16_t *pexp, CPU86_LDouble f)
-{
-    CPU86_LDoubleU temp;
+    CPU_LDoubleU temp;
 
     temp.d = f;
     *pmant = temp.l.lower;
     *pexp = temp.l.upper;
 }
 
-CPU86_LDouble cpu_set_fp80(uint64_t mant, uint16_t upper)
+floatx80 cpu_set_fp80(uint64_t mant, uint16_t upper)
 {
-    CPU86_LDoubleU temp;
+    CPU_LDoubleU temp;
 
     temp.l.upper = upper;
     temp.l.lower = mant;
     return temp.d;
 }
-#endif
 
 #ifdef TARGET_X86_64
 
diff --git a/target-i386/translate.c b/target-i386/translate.c
index 199302e..10bd72a 100644
--- a/target-i386/translate.c
+++ b/target-i386/translate.c
@@ -7538,7 +7538,7 @@
             break;
         case 5: /* lfence */
         case 6: /* mfence */
-            if ((modrm & 0xc7) != 0xc0 || !(s->cpuid_features & CPUID_SSE))
+            if ((modrm & 0xc7) != 0xc0 || !(s->cpuid_features & CPUID_SSE2))
                 goto illegal_op;
             break;
         case 7: /* sfence / clflush */
diff --git a/target-m68k/translate.c b/target-m68k/translate.c
index 9e5578d..26f0ee4 100644
--- a/target-m68k/translate.c
+++ b/target-m68k/translate.c
@@ -171,9 +171,6 @@
   static void disas_##name (DisasContext *s, uint16_t insn)
 #endif
 
-/* FIXME: Remove this.  */
-#define gen_im32(val) tcg_const_i32(val)
-
 /* Generate a load from the specified address.  Narrow values are
    sign extended to full register width.  */
 static inline TCGv gen_load(DisasContext * s, int opsize, TCGv addr, int sign)
@@ -339,7 +336,7 @@
         if ((ext & 0x80) == 0) {
             /* base not suppressed */
             if (IS_NULL_QREG(base)) {
-                base = gen_im32(offset + bd);
+                base = tcg_const_i32(offset + bd);
                 bd = 0;
             }
             if (!IS_NULL_QREG(add)) {
@@ -355,7 +352,7 @@
                 add = tmp;
             }
         } else {
-            add = gen_im32(bd);
+            add = tcg_const_i32(bd);
         }
         if ((ext & 3) != 0) {
             /* memory indirect */
@@ -536,15 +533,15 @@
         case 0: /* Absolute short.  */
             offset = ldsw_code(s->pc);
             s->pc += 2;
-            return gen_im32(offset);
+            return tcg_const_i32(offset);
         case 1: /* Absolute long.  */
             offset = read_im32(s);
-            return gen_im32(offset);
+            return tcg_const_i32(offset);
         case 2: /* pc displacement  */
             offset = s->pc;
             offset += ldsw_code(s->pc);
             s->pc += 2;
-            return gen_im32(offset);
+            return tcg_const_i32(offset);
         case 3: /* pc index+displacement.  */
             return gen_lea_indexed(s, opsize, NULL_QREG);
         case 4: /* Immediate.  */
@@ -1209,16 +1206,16 @@
         break;
     case 2: /* subi */
         tcg_gen_mov_i32(dest, src1);
-        gen_helper_xflag_lt(QREG_CC_X, dest, gen_im32(im));
+        gen_helper_xflag_lt(QREG_CC_X, dest, tcg_const_i32(im));
         tcg_gen_subi_i32(dest, dest, im);
-        gen_update_cc_add(dest, gen_im32(im));
+        gen_update_cc_add(dest, tcg_const_i32(im));
         s->cc_op = CC_OP_SUB;
         break;
     case 3: /* addi */
         tcg_gen_mov_i32(dest, src1);
         tcg_gen_addi_i32(dest, dest, im);
-        gen_update_cc_add(dest, gen_im32(im));
-        gen_helper_xflag_lt(QREG_CC_X, dest, gen_im32(im));
+        gen_update_cc_add(dest, tcg_const_i32(im));
+        gen_helper_xflag_lt(QREG_CC_X, dest, tcg_const_i32(im));
         s->cc_op = CC_OP_ADD;
         break;
     case 5: /* eori */
@@ -1228,7 +1225,7 @@
     case 6: /* cmpi */
         tcg_gen_mov_i32(dest, src1);
         tcg_gen_subi_i32(dest, dest, im);
-        gen_update_cc_add(dest, gen_im32(im));
+        gen_update_cc_add(dest, tcg_const_i32(im));
         s->cc_op = CC_OP_SUB;
         break;
     default:
@@ -1324,8 +1321,8 @@
     default:
         abort();
     }
-    DEST_EA(insn, opsize, gen_im32(0), NULL);
-    gen_logic_cc(s, gen_im32(0));
+    DEST_EA(insn, opsize, tcg_const_i32(0), NULL);
+    gen_logic_cc(s, tcg_const_i32(0));
 }
 
 static TCGv gen_get_ccr(DisasContext *s)
@@ -1589,7 +1586,7 @@
     }
     if ((insn & 0x40) == 0) {
         /* jsr */
-        gen_push(s, gen_im32(s->pc));
+        gen_push(s, tcg_const_i32(s->pc));
     }
     gen_jmp(s, tmp);
 }
@@ -1617,7 +1614,7 @@
             tcg_gen_addi_i32(dest, dest, val);
         }
     } else {
-        src2 = gen_im32(val);
+        src2 = tcg_const_i32(val);
         if (insn & 0x0100) {
             gen_helper_xflag_lt(QREG_CC_X, dest, src2);
             tcg_gen_subi_i32(dest, dest, val);
@@ -1666,7 +1663,7 @@
     }
     if (op == 1) {
         /* bsr */
-        gen_push(s, gen_im32(s->pc));
+        gen_push(s, tcg_const_i32(s->pc));
     }
     gen_flush_cc_op(s);
     if (op > 1) {
@@ -1757,7 +1754,7 @@
     val = (insn >> 9) & 7;
     if (val == 0)
         val = -1;
-    src = gen_im32(val);
+    src = tcg_const_i32(val);
     gen_logic_cc(s, src);
     DEST_EA(insn, OS_LONG, src, NULL);
 }
@@ -1883,7 +1880,7 @@
     tmp = (insn >> 9) & 7;
     if (tmp == 0)
         tmp = 8;
-    shift = gen_im32(tmp);
+    shift = tcg_const_i32(tmp);
     /* No need to flush flags becuse we know we will set C flag.  */
     if (insn & 0x100) {
         gen_helper_shl_cc(reg, cpu_env, reg, shift);
@@ -2191,7 +2188,7 @@
         switch ((ext >> 10) & 7) {
         case 4: /* FPCR */
             /* Not implemented.  Always return zero.  */
-            tmp32 = gen_im32(0);
+            tmp32 = tcg_const_i32(0);
             break;
         case 1: /* FPIAR */
         case 2: /* FPSR */
@@ -2592,7 +2589,7 @@
         /* Skip the accumulate if the value is already saturated.  */
         l1 = gen_new_label();
         tmp = tcg_temp_new();
-        gen_op_and32(tmp, QREG_MACSR, gen_im32(MACSR_PAV0 << acc));
+        gen_op_and32(tmp, QREG_MACSR, tcg_const_i32(MACSR_PAV0 << acc));
         gen_op_jmp_nz32(tmp, l1);
     }
 #endif
@@ -2626,7 +2623,7 @@
             /* Skip the accumulate if the value is already saturated.  */
             l1 = gen_new_label();
             tmp = tcg_temp_new();
-            gen_op_and32(tmp, QREG_MACSR, gen_im32(MACSR_PAV0 << acc));
+            gen_op_and32(tmp, QREG_MACSR, tcg_const_i32(MACSR_PAV0 << acc));
             gen_op_jmp_nz32(tmp, l1);
         }
 #endif
diff --git a/target-ppc/helper.c b/target-ppc/helper.c
index 4700632..cf2a368 100644
--- a/target-ppc/helper.c
+++ b/target-ppc/helper.c
@@ -21,7 +21,6 @@
 #include <stdio.h>
 #include <string.h>
 #include <inttypes.h>
-#include <signal.h>
 
 #include "cpu.h"
 #include "exec-all.h"
diff --git a/target-ppc/helper.h b/target-ppc/helper.h
index 51c99c8..470e42f 100644
--- a/target-ppc/helper.h
+++ b/target-ppc/helper.h
@@ -51,9 +51,7 @@
 DEF_HELPER_FLAGS_2(brinc, TCG_CALL_CONST | TCG_CALL_PURE, tl, tl, tl)
 
 DEF_HELPER_0(float_check_status, void)
-#ifdef CONFIG_SOFTFLOAT
 DEF_HELPER_0(reset_fpstatus, void)
-#endif
 DEF_HELPER_2(compute_fprf, i32, i64, i32)
 DEF_HELPER_2(store_fpscr, void, i64, i32)
 DEF_HELPER_1(fpscr_clrbit, void, i32)
diff --git a/target-ppc/kvm.c b/target-ppc/kvm.c
index ccf4668..e7b1b10 100644
--- a/target-ppc/kvm.c
+++ b/target-ppc/kvm.c
@@ -45,9 +45,7 @@
 static int cap_interrupt_unset = false;
 static int cap_interrupt_level = false;
 static int cap_segstate;
-#ifdef KVM_CAP_PPC_BOOKE_SREGS
 static int cap_booke_sregs;
-#endif
 
 /* XXX We have a race condition where we actually have a level triggered
  *     interrupt, but the infrastructure can't expose that yet, so the guest
@@ -222,13 +220,13 @@
     for (i = 0;i < 32; i++)
         env->gpr[i] = regs.gpr[i];
 
-#ifdef KVM_CAP_PPC_BOOKE_SREGS
     if (cap_booke_sregs) {
         ret = kvm_vcpu_ioctl(env, KVM_GET_SREGS, &sregs);
         if (ret < 0) {
             return ret;
         }
 
+#ifdef KVM_CAP_PPC_BOOKE_SREGS
         if (sregs.u.e.features & KVM_SREGS_E_BASE) {
             env->spr[SPR_BOOKE_CSRR0] = sregs.u.e.csrr0;
             env->spr[SPR_BOOKE_CSRR1] = sregs.u.e.csrr1;
@@ -325,16 +323,16 @@
                 env->spr[SPR_BOOKE_PID2] = sregs.u.e.impl.fsl.pid2;
             }
         }
-    }
 #endif
+    }
 
-#ifdef KVM_CAP_PPC_SEGSTATE
     if (cap_segstate) {
         ret = kvm_vcpu_ioctl(env, KVM_GET_SREGS, &sregs);
         if (ret < 0) {
             return ret;
         }
 
+#ifdef KVM_CAP_PPC_SEGSTATE
         ppc_store_sdr1(env, sregs.u.s.sdr1);
 
         /* Sync SLB */
@@ -357,8 +355,8 @@
             env->IBAT[0][i] = sregs.u.s.ppc32.ibat[i] & 0xffffffff;
             env->IBAT[1][i] = sregs.u.s.ppc32.ibat[i] >> 32;
         }
-    }
 #endif
+    }
 
     return 0;
 }
diff --git a/target-ppc/op_helper.c b/target-ppc/op_helper.c
index c52a371..15d9222 100644
--- a/target-ppc/op_helper.c
+++ b/target-ppc/op_helper.c
@@ -971,7 +971,6 @@
 
 void helper_float_check_status (void)
 {
-#ifdef CONFIG_SOFTFLOAT
     if (env->exception_index == POWERPC_EXCP_PROGRAM &&
         (env->error_code & POWERPC_EXCP_FP)) {
         /* Differred floating-point exception after target FPR update */
@@ -989,22 +988,12 @@
             float_inexact_excp();
         }
     }
-#else
-    if (env->exception_index == POWERPC_EXCP_PROGRAM &&
-        (env->error_code & POWERPC_EXCP_FP)) {
-        /* Differred floating-point exception after target FPR update */
-        if (msr_fe0 != 0 || msr_fe1 != 0)
-            helper_raise_exception_err(env->exception_index, env->error_code);
-    }
-#endif
 }
 
-#ifdef CONFIG_SOFTFLOAT
 void helper_reset_fpstatus (void)
 {
     set_float_exception_flags(0, &env->fp_status);
 }
-#endif
 
 /* fadd - fadd. */
 uint64_t helper_fadd (uint64_t arg1, uint64_t arg2)
diff --git a/target-ppc/translate.c b/target-ppc/translate.c
index 9b3f90c..59aef85 100644
--- a/target-ppc/translate.c
+++ b/target-ppc/translate.c
@@ -215,9 +215,7 @@
 
 static inline void gen_reset_fpstatus(void)
 {
-#ifdef CONFIG_SOFTFLOAT
     gen_helper_reset_fpstatus();
-#endif
 }
 
 static inline void gen_compute_fprf(TCGv_i64 arg, int set_fprf, int set_rc)
diff --git a/target-ppc/translate_init.c b/target-ppc/translate_init.c
index b511afa..fc50ae3 100644
--- a/target-ppc/translate_init.c
+++ b/target-ppc/translate_init.c
@@ -73,7 +73,7 @@
     gen_load_spr(cpu_gpr[gprn], sprn);
 #ifdef PPC_DUMP_SPR_ACCESSES
     {
-        TCGv t0 = tcg_const_i32(sprn);
+        TCGv_i32 t0 = tcg_const_i32(sprn);
         gen_helper_load_dump_spr(t0);
         tcg_temp_free_i32(t0);
     }
@@ -85,7 +85,7 @@
     gen_store_spr(sprn, cpu_gpr[gprn]);
 #ifdef PPC_DUMP_SPR_ACCESSES
     {
-        TCGv t0 = tcg_const_i32(sprn);
+        TCGv_i32 t0 = tcg_const_i32(sprn);
         gen_helper_store_dump_spr(t0);
         tcg_temp_free_i32(t0);
     }
@@ -1367,16 +1367,16 @@
 
 static void spr_write_booke206_mmucsr0 (void *opaque, int sprn, int gprn)
 {
-    TCGv t0 = tcg_const_i32(sprn);
+    TCGv_i32 t0 = tcg_const_i32(sprn);
     gen_helper_booke206_tlbflush(t0);
-    tcg_temp_free(t0);
+    tcg_temp_free_i32(t0);
 }
 
 static void spr_write_booke_pid (void *opaque, int sprn, int gprn)
 {
-    TCGv t0 = tcg_const_i32(sprn);
+    TCGv_i32 t0 = tcg_const_i32(sprn);
     gen_helper_booke_setpid(t0, cpu_gpr[gprn]);
-    tcg_temp_free(t0);
+    tcg_temp_free_i32(t0);
 }
 #endif
 
diff --git a/target-s390x/helper.c b/target-s390x/helper.c
index c79af46..745d8c5 100644
--- a/target-s390x/helper.c
+++ b/target-s390x/helper.c
@@ -28,11 +28,6 @@
 #include "qemu-common.h"
 #include "qemu-timer.h"
 
-#if !defined(CONFIG_USER_ONLY)
-#include <linux/kvm.h>
-#include "kvm.h"
-#endif
-
 //#define DEBUG_S390
 //#define DEBUG_S390_PTE
 //#define DEBUG_S390_STDOUT
diff --git a/target-s390x/op_helper.c b/target-s390x/op_helper.c
index 49760a4..db03a79 100644
--- a/target-s390x/op_helper.c
+++ b/target-s390x/op_helper.c
@@ -1731,25 +1731,15 @@
     env->fregs[f1].d = float64_sqrt(env->fregs[f2].d, &env->fpu_status);
 }
 
-static inline uint64_t cksm_overflow(uint64_t cksm)
-{
-    if (cksm > 0xffffffffULL) {
-        cksm &= 0xffffffffULL;
-        cksm++;
-    }
-    return cksm;
-}
-
 /* checksum */
 void HELPER(cksm)(uint32_t r1, uint32_t r2)
 {
     uint64_t src = get_address_31fix(r2);
     uint64_t src_len = env->regs[(r2 + 1) & 15];
-    uint64_t cksm = 0;
+    uint64_t cksm = (uint32_t)env->regs[r1];
 
     while (src_len >= 4) {
         cksm += ldl(src);
-        cksm = cksm_overflow(cksm);
 
         /* move to next word */
         src_len -= 4;
@@ -1760,26 +1750,24 @@
     case 0:
         break;
     case 1:
-        cksm += ldub(src);
-        cksm = cksm_overflow(cksm);
+        cksm += ldub(src) << 24;
         break;
     case 2:
-        cksm += lduw(src);
-        cksm = cksm_overflow(cksm);
+        cksm += lduw(src) << 16;
         break;
     case 3:
-        /* XXX check if this really is correct */
-        cksm += lduw(src) << 8;
-        cksm += ldub(src + 2);
-        cksm = cksm_overflow(cksm);
+        cksm += lduw(src) << 16;
+        cksm += ldub(src + 2) << 8;
         break;
     }
 
     /* indicate we've processed everything */
+    env->regs[r2] = src + src_len;
     env->regs[(r2 + 1) & 15] = 0;
 
     /* store result */
-    env->regs[r1] = (env->regs[r1] & 0xffffffff00000000ULL) | (uint32_t)cksm;
+    env->regs[r1] = (env->regs[r1] & 0xffffffff00000000ULL) |
+                    ((uint32_t)cksm + (cksm >> 32));
 }
 
 static inline uint32_t cc_calc_ltgt_32(CPUState *env, int32_t src,
diff --git a/target-s390x/translate.c b/target-s390x/translate.c
index 8e71df3..eda4624 100644
--- a/target-s390x/translate.c
+++ b/target-s390x/translate.c
@@ -1078,9 +1078,12 @@
             tcg_gen_brcondi_i32(TCG_COND_EQ, tmp, 0, skip);
             break;
         default:
+            tcg_temp_free_i32(tmp);
+            tcg_temp_free_i32(tmp2);
             goto do_dynamic;
         }
         tcg_temp_free_i32(tmp);
+        tcg_temp_free_i32(tmp2);
         account_inline_branch(s);
         break;
     case CC_OP_TM_64:
@@ -1095,6 +1098,7 @@
             tcg_gen_brcondi_i64(TCG_COND_EQ, tmp64, 0, skip);
             break;
         default:
+            tcg_temp_free_i64(tmp64);
             goto do_dynamic;
         }
         tcg_temp_free_i64(tmp64);
@@ -2056,7 +2060,7 @@
            even for very long ones... */
         tmp = get_address(s, 0, b2, d2);
         tmp3 = tcg_const_i64(stm_len);
-        tmp4 = tcg_const_i64(32);
+        tmp4 = tcg_const_i64(op == 0x26 ? 32 : 4);
         for (i = r1;; i = (i + 1) % 16) {
             switch (op) {
             case 0x4:
@@ -2068,9 +2072,8 @@
                 tcg_gen_qemu_ld32u(tmp2, tmp, get_mem_index(s));
                 tcg_gen_trunc_i64_i32(TCGV_HIGH(regs[i]), tmp2);
 #else
-                tmp2 = tcg_temp_new_i64();
                 tcg_gen_qemu_ld32u(tmp2, tmp, get_mem_index(s));
-                tcg_gen_shl_i64(tmp2, tmp2, 4);
+                tcg_gen_shl_i64(tmp2, tmp2, tmp4);
                 tcg_gen_ext32u_i64(regs[i], regs[i]);
                 tcg_gen_or_i64(regs[i], regs[i], tmp2);
 #endif
@@ -2094,6 +2097,7 @@
             tcg_gen_add_i64(tmp, tmp, tmp3);
         }
         tcg_temp_free_i64(tmp);
+        tcg_temp_free_i64(tmp3);
         tcg_temp_free_i64(tmp4);
         break;
     case 0x2c: /* STCMH R1,M3,D2(B2) [RSY] */
@@ -2330,18 +2334,22 @@
     case 0x0: /* IIHH     R1,I2     [RI] */
         tmp = tcg_const_i64(i2);
         tcg_gen_deposit_i64(regs[r1], regs[r1], tmp, 48, 16);
+        tcg_temp_free_i64(tmp);
         break;
     case 0x1: /* IIHL     R1,I2     [RI] */
         tmp = tcg_const_i64(i2);
         tcg_gen_deposit_i64(regs[r1], regs[r1], tmp, 32, 16);
+        tcg_temp_free_i64(tmp);
         break;
     case 0x2: /* IILH     R1,I2     [RI] */
         tmp = tcg_const_i64(i2);
         tcg_gen_deposit_i64(regs[r1], regs[r1], tmp, 16, 16);
+        tcg_temp_free_i64(tmp);
         break;
     case 0x3: /* IILL     R1,I2     [RI] */
         tmp = tcg_const_i64(i2);
         tcg_gen_deposit_i64(regs[r1], regs[r1], tmp, 0, 16);
+        tcg_temp_free_i64(tmp);
         break;
     case 0x4: /* NIHH     R1,I2     [RI] */
     case 0x8: /* OIHH     R1,I2     [RI] */
@@ -2366,6 +2374,7 @@
         set_cc_nz_u32(s, tmp32);
         tcg_temp_free_i64(tmp2);
         tcg_temp_free_i32(tmp32);
+        tcg_temp_free_i64(tmp);
         break;
     case 0x5: /* NIHL     R1,I2     [RI] */
     case 0x9: /* OIHL     R1,I2     [RI] */
@@ -2391,6 +2400,7 @@
         set_cc_nz_u32(s, tmp32);
         tcg_temp_free_i64(tmp2);
         tcg_temp_free_i32(tmp32);
+        tcg_temp_free_i64(tmp);
         break;
     case 0x6: /* NILH     R1,I2     [RI] */
     case 0xa: /* OILH     R1,I2     [RI] */
@@ -2416,6 +2426,7 @@
         set_cc_nz_u32(s, tmp32);
         tcg_temp_free_i64(tmp2);
         tcg_temp_free_i32(tmp32);
+        tcg_temp_free_i64(tmp);
         break;
     case 0x7: /* NILL     R1,I2     [RI] */
     case 0xb: /* OILL     R1,I2     [RI] */
@@ -2439,29 +2450,33 @@
         set_cc_nz_u32(s, tmp32);        /* signedness should not matter here */
         tcg_temp_free_i64(tmp2);
         tcg_temp_free_i32(tmp32);
+        tcg_temp_free_i64(tmp);
         break;
     case 0xc: /* LLIHH     R1,I2     [RI] */
         tmp = tcg_const_i64( ((uint64_t)i2) << 48 );
         store_reg(r1, tmp);
+        tcg_temp_free_i64(tmp);
         break;
     case 0xd: /* LLIHL     R1,I2     [RI] */
         tmp = tcg_const_i64( ((uint64_t)i2) << 32 );
         store_reg(r1, tmp);
+        tcg_temp_free_i64(tmp);
         break;
     case 0xe: /* LLILH     R1,I2     [RI] */
         tmp = tcg_const_i64( ((uint64_t)i2) << 16 );
         store_reg(r1, tmp);
+        tcg_temp_free_i64(tmp);
         break;
     case 0xf: /* LLILL     R1,I2     [RI] */
         tmp = tcg_const_i64(i2);
         store_reg(r1, tmp);
+        tcg_temp_free_i64(tmp);
         break;
     default:
         LOG_DISAS("illegal a5 operation 0x%x\n", op);
         gen_illegal_opcode(s, 2);
         return;
     }
-    tcg_temp_free_i64(tmp);
 }
 
 static void disas_a7(DisasContext *s, int op, int r1, int i2)
@@ -2963,6 +2978,8 @@
         /* we need to keep cc_op intact */
         s->is_jmp = DISAS_JUMP;
         tcg_temp_free_i64(tmp);
+        tcg_temp_free_i64(tmp2);
+        tcg_temp_free_i64(tmp3);
         break;
     case 0x20: /* SERVC     R1,R2     [RRE] */
         /* SCLP Service call (PV hypercall) */
@@ -3456,6 +3473,9 @@
         tcg_temp_free_i64(tmp2);
         tcg_temp_free_i64(tmp3);
         break;
+    case 0x0f: /* LRVGR    R1,R2     [RRE] */
+        tcg_gen_bswap64_i64(regs[r1], regs[r2]);
+        break;
     case 0x1f: /* LRVR     R1,R2     [RRE] */
         tmp32_1 = load_reg32(r2);
         tcg_gen_bswap32_i32(tmp32_1, tmp32_1);
@@ -4593,6 +4613,8 @@
         store_reg32(r1, tmp32_1);
         tcg_gen_trunc_i64_i32(tmp32_2, tmp2);
         store_reg32(r1 + 1, tmp32_2);
+        tcg_temp_free_i64(tmp);
+        tcg_temp_free_i64(tmp2);
         break;
     case 0x98: /* LM     R1,R3,D2(B2)     [RS] */
     case 0x90: /* STM    R1,R3,D2(B2)     [RS] */
@@ -4616,6 +4638,7 @@
             }
             tcg_gen_add_i64(tmp, tmp, tmp3);
         }
+        tcg_temp_free_i64(tmp);
         tcg_temp_free_i64(tmp2);
         tcg_temp_free_i64(tmp3);
         tcg_temp_free_i64(tmp4);
diff --git a/target-sparc/helper.c b/target-sparc/helper.c
index b2d4d70..e9b42d0 100644
--- a/target-sparc/helper.c
+++ b/target-sparc/helper.c
@@ -21,7 +21,6 @@
 #include <stdio.h>
 #include <string.h>
 #include <inttypes.h>
-#include <signal.h>
 
 #include "cpu.h"
 #include "exec-all.h"
diff --git a/tcg/mips/tcg-target.h b/tcg/mips/tcg-target.h
index 0028bfa..8cb7d88 100644
--- a/tcg/mips/tcg-target.h
+++ b/tcg/mips/tcg-target.h
@@ -102,7 +102,11 @@
 /* guest base is supported */
 #define TCG_TARGET_HAS_GUEST_BASE
 
+#ifdef __OpenBSD__
+#include <machine/sysarch.h>
+#else
 #include <sys/cachectl.h>
+#endif
 
 static inline void flush_icache_range(unsigned long start, unsigned long stop)
 {
diff --git a/tcg/tcg-op.h b/tcg/tcg-op.h
index 207a89f..ebf5e13 100644
--- a/tcg/tcg-op.h
+++ b/tcg/tcg-op.h
@@ -1063,66 +1063,66 @@
     tcg_gen_op2i_i64(INDEX_op_movi_i64, ret, arg);
 }
 
-static inline void tcg_gen_ld8u_i64(TCGv_i64 ret, TCGv_i64 arg2,
+static inline void tcg_gen_ld8u_i64(TCGv_i64 ret, TCGv_ptr arg2,
                                     tcg_target_long offset)
 {
     tcg_gen_ldst_op_i64(INDEX_op_ld8u_i64, ret, arg2, offset);
 }
 
-static inline void tcg_gen_ld8s_i64(TCGv_i64 ret, TCGv_i64 arg2,
+static inline void tcg_gen_ld8s_i64(TCGv_i64 ret, TCGv_ptr arg2,
                                     tcg_target_long offset)
 {
     tcg_gen_ldst_op_i64(INDEX_op_ld8s_i64, ret, arg2, offset);
 }
 
-static inline void tcg_gen_ld16u_i64(TCGv_i64 ret, TCGv_i64 arg2,
+static inline void tcg_gen_ld16u_i64(TCGv_i64 ret, TCGv_ptr arg2,
                                      tcg_target_long offset)
 {
     tcg_gen_ldst_op_i64(INDEX_op_ld16u_i64, ret, arg2, offset);
 }
 
-static inline void tcg_gen_ld16s_i64(TCGv_i64 ret, TCGv_i64 arg2,
+static inline void tcg_gen_ld16s_i64(TCGv_i64 ret, TCGv_ptr arg2,
                                      tcg_target_long offset)
 {
     tcg_gen_ldst_op_i64(INDEX_op_ld16s_i64, ret, arg2, offset);
 }
 
-static inline void tcg_gen_ld32u_i64(TCGv_i64 ret, TCGv_i64 arg2,
+static inline void tcg_gen_ld32u_i64(TCGv_i64 ret, TCGv_ptr arg2,
                                      tcg_target_long offset)
 {
     tcg_gen_ldst_op_i64(INDEX_op_ld32u_i64, ret, arg2, offset);
 }
 
-static inline void tcg_gen_ld32s_i64(TCGv_i64 ret, TCGv_i64 arg2,
+static inline void tcg_gen_ld32s_i64(TCGv_i64 ret, TCGv_ptr arg2,
                                      tcg_target_long offset)
 {
     tcg_gen_ldst_op_i64(INDEX_op_ld32s_i64, ret, arg2, offset);
 }
 
-static inline void tcg_gen_ld_i64(TCGv_i64 ret, TCGv_i64 arg2, tcg_target_long offset)
+static inline void tcg_gen_ld_i64(TCGv_i64 ret, TCGv_ptr arg2, tcg_target_long offset)
 {
     tcg_gen_ldst_op_i64(INDEX_op_ld_i64, ret, arg2, offset);
 }
 
-static inline void tcg_gen_st8_i64(TCGv_i64 arg1, TCGv_i64 arg2,
+static inline void tcg_gen_st8_i64(TCGv_i64 arg1, TCGv_ptr arg2,
                                    tcg_target_long offset)
 {
     tcg_gen_ldst_op_i64(INDEX_op_st8_i64, arg1, arg2, offset);
 }
 
-static inline void tcg_gen_st16_i64(TCGv_i64 arg1, TCGv_i64 arg2,
+static inline void tcg_gen_st16_i64(TCGv_i64 arg1, TCGv_ptr arg2,
                                     tcg_target_long offset)
 {
     tcg_gen_ldst_op_i64(INDEX_op_st16_i64, arg1, arg2, offset);
 }
 
-static inline void tcg_gen_st32_i64(TCGv_i64 arg1, TCGv_i64 arg2,
+static inline void tcg_gen_st32_i64(TCGv_i64 arg1, TCGv_ptr arg2,
                                     tcg_target_long offset)
 {
     tcg_gen_ldst_op_i64(INDEX_op_st32_i64, arg1, arg2, offset);
 }
 
-static inline void tcg_gen_st_i64(TCGv_i64 arg1, TCGv_i64 arg2, tcg_target_long offset)
+static inline void tcg_gen_st_i64(TCGv_i64 arg1, TCGv_ptr arg2, tcg_target_long offset)
 {
     tcg_gen_ldst_op_i64(INDEX_op_st_i64, arg1, arg2, offset);
 }
@@ -2304,8 +2304,8 @@
 #endif
 }
 
-#define tcg_gen_ld_ptr tcg_gen_ld_i32
-#define tcg_gen_discard_ptr tcg_gen_discard_i32
+#define tcg_gen_ld_ptr(R, A, O) tcg_gen_ld_i32(TCGV_PTR_TO_NAT(R), (A), (O))
+#define tcg_gen_discard_ptr(A) tcg_gen_discard_i32(TCGV_PTR_TO_NAT(A))
 
 #else /* TCG_TARGET_REG_BITS == 32 */
 
@@ -2372,8 +2372,8 @@
     tcg_gen_qemu_ldst_op_i64(INDEX_op_qemu_st64, arg, addr, mem_index);
 }
 
-#define tcg_gen_ld_ptr tcg_gen_ld_i64
-#define tcg_gen_discard_ptr tcg_gen_discard_i64
+#define tcg_gen_ld_ptr(R, A, O) tcg_gen_ld_i64(TCGV_PTR_TO_NAT(R), (A), (O))
+#define tcg_gen_discard_ptr(A) tcg_gen_discard_i64(TCGV_PTR_TO_NAT(A))
 
 #endif /* TCG_TARGET_REG_BITS != 32 */
 
@@ -2523,11 +2523,17 @@
 #endif
 
 #if TCG_TARGET_REG_BITS == 32
-#define tcg_gen_add_ptr tcg_gen_add_i32
-#define tcg_gen_addi_ptr tcg_gen_addi_i32
-#define tcg_gen_ext_i32_ptr tcg_gen_mov_i32
+#define tcg_gen_add_ptr(R, A, B) tcg_gen_add_i32(TCGV_PTR_TO_NAT(R), \
+                                               TCGV_PTR_TO_NAT(A), \
+                                               TCGV_PTR_TO_NAT(B))
+#define tcg_gen_addi_ptr(R, A, B) tcg_gen_addi_i32(TCGV_PTR_TO_NAT(R), \
+                                                 TCGV_PTR_TO_NAT(A), (B))
+#define tcg_gen_ext_i32_ptr(R, A) tcg_gen_mov_i32(TCGV_PTR_TO_NAT(R), (A))
 #else /* TCG_TARGET_REG_BITS == 32 */
-#define tcg_gen_add_ptr tcg_gen_add_i64
-#define tcg_gen_addi_ptr tcg_gen_addi_i64
-#define tcg_gen_ext_i32_ptr tcg_gen_ext_i32_i64
+#define tcg_gen_add_ptr(R, A, B) tcg_gen_add_i64(TCGV_PTR_TO_NAT(R), \
+                                               TCGV_PTR_TO_NAT(A), \
+                                               TCGV_PTR_TO_NAT(B))
+#define tcg_gen_addi_ptr(R, A, B) tcg_gen_addi_i64(TCGV_PTR_TO_NAT(R),   \
+                                                 TCGV_PTR_TO_NAT(A), (B))
+#define tcg_gen_ext_i32_ptr(R, A) tcg_gen_ext_i32_i64(TCGV_PTR_TO_NAT(R), (A))
 #endif /* TCG_TARGET_REG_BITS != 32 */
diff --git a/tcg/tcg.c b/tcg/tcg.c
index fad92f9..184c208 100644
--- a/tcg/tcg.c
+++ b/tcg/tcg.c
@@ -585,7 +585,7 @@
 void tcg_gen_callN(TCGContext *s, TCGv_ptr func, unsigned int flags,
                    int sizemask, TCGArg ret, int nargs, TCGArg *args)
 {
-#ifdef TCG_TARGET_I386
+#if defined(TCG_TARGET_I386) && TCG_TARGET_REG_BITS < 64
     int call_type;
 #endif
     int i;
@@ -612,7 +612,7 @@
 
     *gen_opc_ptr++ = INDEX_op_call;
     nparam = gen_opparam_ptr++;
-#ifdef TCG_TARGET_I386
+#if defined(TCG_TARGET_I386) && TCG_TARGET_REG_BITS < 64
     call_type = (flags & TCG_CALL_TYPE_MASK);
 #endif
     if (ret != TCG_CALL_DUMMY_ARG) {
diff --git a/tcg/tcg.h b/tcg/tcg.h
index 746378a..3647390 100644
--- a/tcg/tcg.h
+++ b/tcg/tcg.h
@@ -150,12 +150,19 @@
     int i64;
 } TCGv_i64;
 
+typedef struct {
+    int iptr;
+} TCGv_ptr;
+
 #define MAKE_TCGV_I32(i) __extension__                  \
     ({ TCGv_i32 make_tcgv_tmp = {i}; make_tcgv_tmp;})
 #define MAKE_TCGV_I64(i) __extension__                  \
     ({ TCGv_i64 make_tcgv_tmp = {i}; make_tcgv_tmp;})
+#define MAKE_TCGV_PTR(i) __extension__                  \
+    ({ TCGv_ptr make_tcgv_tmp = {i}; make_tcgv_tmp; })
 #define GET_TCGV_I32(t) ((t).i32)
 #define GET_TCGV_I64(t) ((t).i64)
+#define GET_TCGV_PTR(t) ((t).iptr)
 #if TCG_TARGET_REG_BITS == 32
 #define TCGV_LOW(t) MAKE_TCGV_I32(GET_TCGV_I64(t))
 #define TCGV_HIGH(t) MAKE_TCGV_I32(GET_TCGV_I64(t) + 1)
@@ -165,10 +172,17 @@
 
 typedef int TCGv_i32;
 typedef int TCGv_i64;
+#if TCG_TARGET_REG_BITS == 32
+#define TCGv_ptr TCGv_i32
+#else
+#define TCGv_ptr TCGv_i64
+#endif
 #define MAKE_TCGV_I32(x) (x)
 #define MAKE_TCGV_I64(x) (x)
+#define MAKE_TCGV_PTR(x) (x)
 #define GET_TCGV_I32(t) (t)
 #define GET_TCGV_I64(t) (t)
+#define GET_TCGV_PTR(t) (t)
 
 #if TCG_TARGET_REG_BITS == 32
 #define TCGV_LOW(t) (t)
@@ -459,25 +473,27 @@
 void tcg_add_target_add_op_defs(const TCGTargetOpDef *tdefs);
 
 #if TCG_TARGET_REG_BITS == 32
-#define tcg_const_ptr tcg_const_i32
-#define tcg_add_ptr tcg_add_i32
-#define tcg_sub_ptr tcg_sub_i32
-#define TCGv_ptr TCGv_i32
-#define GET_TCGV_PTR GET_TCGV_I32
-#define tcg_global_reg_new_ptr tcg_global_reg_new_i32
-#define tcg_global_mem_new_ptr tcg_global_mem_new_i32
-#define tcg_temp_new_ptr tcg_temp_new_i32
-#define tcg_temp_free_ptr tcg_temp_free_i32
+#define TCGV_NAT_TO_PTR(n) MAKE_TCGV_PTR(GET_TCGV_I32(n))
+#define TCGV_PTR_TO_NAT(n) MAKE_TCGV_I32(GET_TCGV_PTR(n))
+
+#define tcg_const_ptr(V) TCGV_NAT_TO_PTR(tcg_const_i32(V))
+#define tcg_global_reg_new_ptr(R, N) \
+    TCGV_NAT_TO_PTR(tcg_global_reg_new_i32((R), (N)))
+#define tcg_global_mem_new_ptr(R, O, N) \
+    TCGV_NAT_TO_PTR(tcg_global_mem_new_i32((R), (O), (N)))
+#define tcg_temp_new_ptr() TCGV_NAT_TO_PTR(tcg_temp_new_i32())
+#define tcg_temp_free_ptr(T) tcg_temp_free_i32(TCGV_PTR_TO_NAT(T))
 #else
-#define tcg_const_ptr tcg_const_i64
-#define tcg_add_ptr tcg_add_i64
-#define tcg_sub_ptr tcg_sub_i64
-#define TCGv_ptr TCGv_i64
-#define GET_TCGV_PTR GET_TCGV_I64
-#define tcg_global_reg_new_ptr tcg_global_reg_new_i64
-#define tcg_global_mem_new_ptr tcg_global_mem_new_i64
-#define tcg_temp_new_ptr tcg_temp_new_i64
-#define tcg_temp_free_ptr tcg_temp_free_i64
+#define TCGV_NAT_TO_PTR(n) MAKE_TCGV_PTR(GET_TCGV_I64(n))
+#define TCGV_PTR_TO_NAT(n) MAKE_TCGV_I64(GET_TCGV_PTR(n))
+
+#define tcg_const_ptr(V) TCGV_NAT_TO_PTR(tcg_const_i64(V))
+#define tcg_global_reg_new_ptr(R, N) \
+    TCGV_NAT_TO_PTR(tcg_global_reg_new_i64((R), (N)))
+#define tcg_global_mem_new_ptr(R, O, N) \
+    TCGV_NAT_TO_PTR(tcg_global_mem_new_i64((R), (O), (N)))
+#define tcg_temp_new_ptr() TCGV_NAT_TO_PTR(tcg_temp_new_i64())
+#define tcg_temp_free_ptr(T) tcg_temp_free_i64(TCGV_PTR_TO_NAT(T))
 #endif
 
 void tcg_gen_callN(TCGContext *s, TCGv_ptr func, unsigned int flags,
diff --git a/trace-events b/trace-events
index 3137a15..f1230f1 100644
--- a/trace-events
+++ b/trace-events
@@ -194,6 +194,26 @@
 disable sun4m_iommu_translate_pa(uint64_t addr, uint64_t pa, uint32_t iopte) "xlate dva %"PRIx64" => pa %"PRIx64" iopte = %x"
 disable sun4m_iommu_bad_addr(uint64_t addr) "bad addr %"PRIx64""
 
+# hw/usb-ehci.c
+disable usb_ehci_reset(void) "=== RESET ==="
+disable usb_ehci_mmio_readl(uint32_t addr, const char *str, uint32_t val) "rd mmio %04x [%s] = %x"
+disable usb_ehci_mmio_writel(uint32_t addr, const char *str, uint32_t val) "wr mmio %04x [%s] = %x"
+disable usb_ehci_mmio_change(uint32_t addr, const char *str, uint32_t new, uint32_t old) "ch mmio %04x [%s] = %x (old: %x)"
+disable usb_ehci_usbsts(const char *sts, int state) "usbsts %s %d"
+disable usb_ehci_state(const char *schedule, const char *state) "%s schedule %s"
+disable usb_ehci_qh_ptrs(void *q, uint32_t addr, uint32_t next, uint32_t c_qtd, uint32_t n_qtd, uint32_t a_qtd) "q %p - QH @ %08x: next %08x qtds %08x,%08x,%08x"
+disable usb_ehci_qh_fields(uint32_t addr, int rl, int mplen, int eps, int ep, int devaddr) "QH @ %08x - rl %d, mplen %d, eps %d, ep %d, dev %d"
+disable usb_ehci_qh_bits(uint32_t addr, int c, int h, int dtc, int i) "QH @ %08x - c %d, h %d, dtc %d, i %d"
+disable usb_ehci_qtd_ptrs(void *q, uint32_t addr, uint32_t next, uint32_t altnext) "q %p - QTD @ %08x: next %08x altnext %08x"
+disable usb_ehci_qtd_fields(uint32_t addr, int tbytes, int cpage, int cerr, int pid) "QTD @ %08x - tbytes %d, cpage %d, cerr %d, pid %d"
+disable usb_ehci_qtd_bits(uint32_t addr, int ioc, int active, int halt, int babble, int xacterr) "QTD @ %08x - ioc %d, active %d, halt %d, babble %d, xacterr %d"
+disable usb_ehci_itd(uint32_t addr, uint32_t next, uint32_t mplen, uint32_t mult, uint32_t ep, uint32_t devaddr) "ITD @ %08x: next %08x - mplen %d, mult %d, ep %d, dev %d"
+disable usb_ehci_port_attach(uint32_t port, const char *device) "attach port #%d - %s"
+disable usb_ehci_port_detach(uint32_t port) "detach port #%d"
+disable usb_ehci_port_reset(uint32_t port, int enable) "reset port #%d - %d"
+disable usb_ehci_data(int rw, uint32_t cpage, uint32_t offset, uint32_t addr, uint32_t len, uint32_t bufpos) "write %d, cpage %d, offset 0x%03x, addr 0x%08x, len %d, bufpos %d"
+disable usb_ehci_queue_action(void *q, const char *action) "q %p: %s"
+
 # hw/usb-desc.c
 disable usb_desc_device(int addr, int len, int ret) "dev %d query device, len %d, ret %d"
 disable usb_desc_device_qualifier(int addr, int len, int ret) "dev %d query device qualifier, len %d, ret %d"
@@ -210,7 +230,8 @@
 disable scsi_req_data(int target, int lun, int tag, int len) "target %d lun %d tag %d len %d"
 disable scsi_req_dequeue(int target, int lun, int tag) "target %d lun %d tag %d"
 disable scsi_req_continue(int target, int lun, int tag) "target %d lun %d tag %d"
-disable scsi_req_parsed(int target, int lun, int tag, int cmd, int mode, int xfer, uint64_t lba) "target %d lun %d tag %d command %d dir %d length %d lba %"PRIu64""
+disable scsi_req_parsed(int target, int lun, int tag, int cmd, int mode, int xfer) "target %d lun %d tag %d command %d dir %d length %d"
+disable scsi_req_parsed_lba(int target, int lun, int tag, int cmd, uint64_t lba) "target %d lun %d tag %d command %d lba %"PRIu64""
 disable scsi_req_parse_bad(int target, int lun, int tag, int cmd) "target %d lun %d tag %d command %d"
 
 # vl.c
diff --git a/ui/curses.c b/ui/curses.c
index 82bc614..d29b6cf 100644
--- a/ui/curses.c
+++ b/ui/curses.c
@@ -24,7 +24,6 @@
 #include <curses.h>
 
 #ifndef _WIN32
-#include <signal.h>
 #include <sys/ioctl.h>
 #include <termios.h>
 #endif
diff --git a/ui/sdl.c b/ui/sdl.c
index 14a62d9..f2bd4a0 100644
--- a/ui/sdl.c
+++ b/ui/sdl.c
@@ -28,10 +28,6 @@
 #include <SDL.h>
 #include <SDL_syswm.h>
 
-#ifndef _WIN32
-#include <signal.h>
-#endif
-
 #include "qemu-common.h"
 #include "console.h"
 #include "sysemu.h"
diff --git a/ui/spice-core.c b/ui/spice-core.c
index ef56ed6..dd9905b 100644
--- a/ui/spice-core.c
+++ b/ui/spice-core.c
@@ -299,8 +299,6 @@
     exit(1);
 }
 
-#if SPICE_SERVER_VERSION >= 0x000600 /* 0.6.0 */
-
 static const char *stream_video_names[] = {
     [ SPICE_STREAM_VIDEO_OFF ]    = "off",
     [ SPICE_STREAM_VIDEO_ALL ]    = "all",
@@ -309,8 +307,6 @@
 #define parse_stream_video(_name) \
     name2enum(_name, stream_video_names, ARRAY_SIZE(stream_video_names))
 
-#endif /* >= 0.6.0 */
-
 static const char *compression_names[] = {
     [ SPICE_IMAGE_COMPRESS_OFF ]      = "off",
     [ SPICE_IMAGE_COMPRESS_AUTO_GLZ ] = "auto_glz",
@@ -549,11 +545,29 @@
     if (password) {
         spice_server_set_ticket(spice_server, password, 0, 0, 0);
     }
+    if (qemu_opt_get_bool(opts, "sasl", 0)) {
+#if SPICE_SERVER_VERSION >= 0x000900 /* 0.9.0 */
+        if (spice_server_set_sasl_appname(spice_server, "qemu") == -1 ||
+            spice_server_set_sasl(spice_server, 1) == -1) {
+            fprintf(stderr, "spice: failed to enable sasl\n");
+            exit(1);
+        }
+#else
+        fprintf(stderr, "spice: sasl is not available (spice >= 0.9 required)\n");
+        exit(1);
+#endif
+    }
     if (qemu_opt_get_bool(opts, "disable-ticketing", 0)) {
         auth = "none";
         spice_server_set_noauth(spice_server);
     }
 
+#if SPICE_SERVER_VERSION >= 0x000801
+    if (qemu_opt_get_bool(opts, "disable-copy-paste", 0)) {
+        spice_server_set_agent_copypaste(spice_server, false);
+    }
+#endif
+
     compression = SPICE_IMAGE_COMPRESS_AUTO_GLZ;
     str = qemu_opt_get(opts, "image-compression");
     if (str) {
@@ -575,8 +589,6 @@
     }
     spice_server_set_zlib_glz_compression(spice_server, wan_compr);
 
-#if SPICE_SERVER_VERSION >= 0x000600 /* 0.6.0 */
-
     str = qemu_opt_get(opts, "streaming-video");
     if (str) {
         int streaming_video = parse_stream_video(str);
@@ -588,8 +600,6 @@
     spice_server_set_playback_compression
         (spice_server, qemu_opt_get_bool(opts, "playback-compression", 1));
 
-#endif /* >= 0.6.0 */
-
     qemu_opt_foreach(opts, add_channel, NULL, 0);
 
     spice_server_init(spice_server, &core_interface);
diff --git a/usb-bsd.c b/usb-bsd.c
index 9bab6e3..c1bcc4a 100644
--- a/usb-bsd.c
+++ b/usb-bsd.c
@@ -39,7 +39,6 @@
 #else
 #include <bus/usb/usb.h>
 #endif
-#include <signal.h>
 
 /* This value has maximum potential at 16.
  * You should also set hw.usb.debug to gain
diff --git a/usb-linux.c b/usb-linux.c
index baa6574..5d2ec5c 100644
--- a/usb-linux.c
+++ b/usb-linux.c
@@ -37,7 +37,6 @@
 
 #include <dirent.h>
 #include <sys/ioctl.h>
-#include <signal.h>
 
 #include <linux/usbdevice_fs.h>
 #include <linux/version.h>
@@ -116,7 +115,7 @@
     USBDevice dev;
     int       fd;
 
-    uint8_t   descr[1024];
+    uint8_t   descr[8192];
     int       descr_len;
     int       configuration;
     int       ninterfaces;
@@ -268,6 +267,14 @@
     qemu_free(aurb);
 }
 
+static void do_disconnect(USBHostDevice *s)
+{
+    printf("husb: device %d.%d disconnected\n",
+           s->bus_num, s->addr);
+    usb_host_close(s);
+    usb_host_auto_check(NULL);
+}
+
 static void async_complete(void *opaque)
 {
     USBHostDevice *s = opaque;
@@ -282,10 +289,7 @@
                 return;
             }
             if (errno == ENODEV && !s->closing) {
-                printf("husb: device %d.%d disconnected\n",
-                       s->bus_num, s->addr);
-                usb_host_close(s);
-                usb_host_auto_check(NULL);
+                do_disconnect(s);
                 return;
             }
 
@@ -359,6 +363,7 @@
 
 static int usb_host_claim_interfaces(USBHostDevice *dev, int configuration)
 {
+    const char *op = NULL;
     int dev_descr_len, config_descr_len;
     int interface, nb_interfaces;
     int ret, i;
@@ -371,7 +376,8 @@
     i = 0;
     dev_descr_len = dev->descr[0];
     if (dev_descr_len > dev->descr_len) {
-        goto fail;
+        fprintf(stderr, "husb: update iface failed. descr too short\n");
+        return 0;
     }
 
     i += dev_descr_len;
@@ -399,7 +405,7 @@
     if (i >= dev->descr_len) {
         fprintf(stderr,
                 "husb: update iface failed. no matching configuration\n");
-        goto fail;
+        return 0;
     }
     nb_interfaces = dev->descr[i + 4];
 
@@ -411,9 +417,9 @@
             ctrl.ioctl_code = USBDEVFS_DISCONNECT;
             ctrl.ifno = interface;
             ctrl.data = 0;
+            op = "USBDEVFS_DISCONNECT";
             ret = ioctl(dev->fd, USBDEVFS_IOCTL, &ctrl);
             if (ret < 0 && errno != ENODATA) {
-                perror("USBDEVFS_DISCONNECT");
                 goto fail;
             }
         }
@@ -422,6 +428,7 @@
 
     /* XXX: only grab if all interfaces are free */
     for (interface = 0; interface < nb_interfaces; interface++) {
+        op = "USBDEVFS_CLAIMINTERFACE";
         ret = ioctl(dev->fd, USBDEVFS_CLAIMINTERFACE, &interface);
         if (ret < 0) {
             if (errno == EBUSY) {
@@ -429,8 +436,7 @@
             } else {
                 perror("husb: failed to claim interface");
             }
-        fail:
-            return 0;
+            goto fail;
         }
     }
 
@@ -440,6 +446,13 @@
     dev->ninterfaces   = nb_interfaces;
     dev->configuration = configuration;
     return 1;
+
+fail:
+    if (errno == ENODEV) {
+        do_disconnect(dev);
+    }
+    perror(op);
+    return 0;
 }
 
 static int usb_host_release_interfaces(USBHostDevice *s)
@@ -1016,6 +1029,11 @@
             }
 
             devep = descriptors[i + 2];
+            if ((devep & 0x0f) == 0) {
+                fprintf(stderr, "usb-linux: invalid ep descriptor, ep == 0\n");
+                return 1;
+            }
+
             switch (descriptors[i + 3] & 0x3) {
             case 0x00:
                 type = USBDEVFS_URB_TYPE_CONTROL;
@@ -1044,10 +1062,9 @@
 }
 
 static int usb_host_open(USBHostDevice *dev, int bus_num,
-                         int addr, char *port, const char *prod_name)
+                        int addr, char *port, const char *prod_name, int speed)
 {
     int fd = -1, ret;
-    struct usbdevfs_connectinfo ci;
     char buf[1024];
 
     if (dev->fd != -1) {
@@ -1102,24 +1119,29 @@
         goto fail;
     }
 
-    ret = ioctl(fd, USBDEVFS_CONNECTINFO, &ci);
-    if (ret < 0) {
-        perror("usb_host_device_open: USBDEVFS_CONNECTINFO");
-        goto fail;
-    }
-
-    printf("husb: grabbed usb device %d.%d\n", bus_num, addr);
-
     ret = usb_linux_update_endp_table(dev);
     if (ret) {
         goto fail;
     }
 
-    if (ci.slow) {
-        dev->dev.speed = USB_SPEED_LOW;
-    } else {
-        dev->dev.speed = USB_SPEED_HIGH;
+    if (speed == -1) {
+        struct usbdevfs_connectinfo ci;
+
+        ret = ioctl(fd, USBDEVFS_CONNECTINFO, &ci);
+        if (ret < 0) {
+            perror("usb_host_device_open: USBDEVFS_CONNECTINFO");
+            goto fail;
+        }
+
+        if (ci.slow) {
+            speed = USB_SPEED_LOW;
+        } else {
+            speed = USB_SPEED_HIGH;
+        }
     }
+    dev->dev.speed = speed;
+
+    printf("husb: grabbed usb device %d.%d\n", bus_num, addr);
 
     if (!prod_name || prod_name[0] == '\0') {
         snprintf(dev->dev.product_desc, sizeof(dev->dev.product_desc),
@@ -1136,9 +1158,9 @@
     return 0;
 
 fail:
-    dev->fd = -1;
-    if (fd != -1) {
-        close(fd);
+    if (dev->fd != -1) {
+        close(dev->fd);
+        dev->fd = -1;
     }
     return -1;
 }
@@ -1147,7 +1169,7 @@
 {
     int i;
 
-    if (dev->fd == -1) {
+    if (dev->fd == -1 || !dev->dev.attached) {
         return -1;
     }
 
@@ -1333,7 +1355,8 @@
     }
 
     device_count = 0;
-    bus_num = addr = speed = class_id = product_id = vendor_id = 0;
+    bus_num = addr = class_id = product_id = vendor_id = 0;
+    speed = -1; /* Can't get the speed from /[proc|dev]/bus/usb/devices */
     for(;;) {
         if (fgets(line, sizeof(line), f) == NULL) {
             break;
@@ -1361,7 +1384,9 @@
             if (get_tag_value(buf, sizeof(buf), line, "Spd=", " ") < 0) {
                 goto fail;
             }
-            if (!strcmp(buf, "480")) {
+            if (!strcmp(buf, "5000")) {
+                speed = USB_SPEED_SUPER;
+            } else if (!strcmp(buf, "480")) {
                 speed = USB_SPEED_HIGH;
             } else if (!strcmp(buf, "1.5")) {
                 speed = USB_SPEED_LOW;
@@ -1505,7 +1530,9 @@
             if (!usb_host_read_file(line, sizeof(line), "speed", de->d_name)) {
                 goto the_end;
             }
-            if (!strcmp(line, "480\n")) {
+            if (!strcmp(line, "5000\n")) {
+                speed = USB_SPEED_SUPER;
+            } else if (!strcmp(line, "480\n")) {
                 speed = USB_SPEED_HIGH;
             } else if (!strcmp(line, "1.5\n")) {
                 speed = USB_SPEED_LOW;
@@ -1643,7 +1670,8 @@
         }
         DPRINTF("husb: auto open: bus_num %d addr %d\n", bus_num, addr);
 
-        usb_host_open(s, bus_num, addr, port, product_name);
+        usb_host_open(s, bus_num, addr, port, product_name, speed);
+        break;
     }
 
     return 0;
@@ -1782,6 +1810,9 @@
     case USB_SPEED_HIGH:
         speed_str = "480";
         break;
+    case USB_SPEED_SUPER:
+        speed_str = "5000";
+        break;
     default:
         speed_str = "?";
         break;
diff --git a/vl.c b/vl.c
index b362871..c1cc614 100644
--- a/vl.c
+++ b/vl.c
@@ -289,6 +289,8 @@
     { .driver = "VGA",                  .flag = &default_vga       },
     { .driver = "cirrus-vga",           .flag = &default_vga       },
     { .driver = "vmware-svga",          .flag = &default_vga       },
+    { .driver = "isa-vga",              .flag = &default_vga       },
+    { .driver = "qxl-vga",              .flag = &default_vga       },
 };
 
 static int default_driver_check(QemuOpts *opts, void *opaque)
@@ -923,9 +925,13 @@
         goto done;
 
     /* the other ones */
+#ifndef CONFIG_LINUX
+    /* only the linux version is qdev-ified, usb-bsd still needs this */
     if (strstart(devname, "host:", &p)) {
         dev = usb_host_device_open(p);
-    } else if (!strcmp(devname, "bt") || strstart(devname, "bt:", &p)) {
+    } else
+#endif
+    if (!strcmp(devname, "bt") || strstart(devname, "bt:", &p)) {
         dev = usb_bt_init(devname[2] ? hci_init(p) :
                         bt_new_hci(qemu_find_bt_vlan(0)));
     } else {
@@ -1933,6 +1939,7 @@
         p = get_opt_name(buf, sizeof (buf), p, ':');
         for (i = 0; i < ARRAY_SIZE(accel_list); i++) {
             if (strcmp(accel_list[i].opt_name, buf) == 0) {
+                *(accel_list[i].allowed) = 1;
                 ret = accel_list[i].init();
                 if (ret < 0) {
                     init_failed = 1;
@@ -1944,9 +1951,9 @@
                                 accel_list[i].name,
                                 strerror(-ret));
                     }
+                    *(accel_list[i].allowed) = 0;
                 } else {
                     accel_initalised = 1;
-                    *(accel_list[i].allowed) = 1;
                 }
                 break;
             }