usb-hid: add microsoft os descriptor support

Set SelectiveSuspendEnabled registy entry to one.
This makes Windows use remote suspend by default,
without manual registry fiddeling.

Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>

.gitignore

@@ -3,6 +3,7 @@ config-all-devices.*
 config-all-disas.*
 config-host.*
 config-target.*
+config.status
 trace/generated-tracers.h
 trace/generated-tracers.c
 trace/generated-tracers-dtrace.h

.travis.yml

@@ -16,6 +16,7 @@ env:
   matrix:
   - TARGETS=alpha-softmmu,alpha-linux-user
   - TARGETS=arm-softmmu,arm-linux-user
+  - TARGETS=aarch64-softmmu,aarch64-linux-user
   - TARGETS=cris-softmmu
   - TARGETS=i386-softmmu,x86_64-softmmu
   - TARGETS=lm32-softmmu

MAINTAINERS

@@ -219,6 +219,13 @@ F: *win32*
 
 ARM Machines
 ------------
+Allwinner-a10
+M: Li Guang <lig.fnst@cn.fujitsu.com>
+S: Maintained
+F: hw/*/allwinner-a10*
+F: include/hw/*/allwinner-a10*
+F: hw/arm/cubieboard.c
+
 Exynos
 M: Evgeny Voevodin <e.voevodin@samsung.com>
 M: Maksim Kozlov <m.kozlov@samsung.com>
@@ -233,6 +240,12 @@ S: Supported
 F: hw/arm/highbank.c
 F: hw/net/xgmac.c
 
+Canon DIGIC
+M: Antony Pavlov <antonynpavlov@gmail.com>
+S: Maintained
+F: include/hw/arm/digic.h
+F: hw/*/digic*
+
 Gumstix
 M: qemu-devel@nongnu.org
 S: Orphan
@@ -500,9 +513,23 @@ X86 Machines
 ------------
 PC
 M: Anthony Liguori <aliguori@amazon.com>
+M: Michael S. Tsirkin <mst@redhat.com>
 S: Supported
-F: hw/i386/pc.[ch]
-F: hw/i386/pc_piix.c
+F: include/hw/i386/
+F: hw/i386/
+F: hw/pci-host/piix.c
+F: hw/pci-host/q35.c
+F: hw/pci-host/pam.c
+F: include/hw/pci-host/q35.h
+F: include/hw/pci-host/pam.h
+F: hw/isa/piix4.c
+F: hw/isa/lpc_ich9.c
+F: hw/i2c/smbus_ich9.c
+F: hw/acpi/piix4.c
+F: hw/acpi/ich9.c
+F: include/hw/acpi/ich9.h
+F: include/hw/acpi/piix.h
+
 
 Xtensa Machines
 ---------------
@@ -687,6 +714,7 @@ F: ui/
 
 Cocoa graphics
 M: Andreas Färber <andreas.faerber@web.de>
+M: Peter Maydell <peter.maydell@linaro.org>
 S: Odd Fixes
 F: ui/cocoa.m
 
@@ -710,6 +738,14 @@ S: Maintained
 F: net/
 T: git git://github.com/stefanha/qemu.git net
 
+Netmap network backend
+M: Luigi Rizzo <rizzo@iet.unipi.it>
+M: Giuseppe Lettieri <g.lettieri@iet.unipi.it>
+M: Vincenzo Maffione <v.maffione@gmail.com>
+W: http://info.iet.unipi.it/~luigi/netmap/
+S: Maintained
+F: net/netmap.c
+
 Network Block Device (NBD)
 M: Paolo Bonzini <pbonzini@redhat.com>
 S: Odd Fixes
@@ -879,6 +915,7 @@ F: block/rbd.c
 Sheepdog
 M: MORITA Kazutaka <morita.kazutaka@lab.ntt.co.jp>
 M: Liu Yuan <namei.unix@gmail.com>
+L: sheepdog@lists.wpkg.org
 S: Supported
 F: block/sheepdog.c
 

Makefile.objs

@@ -51,6 +51,8 @@ common-obj-$(CONFIG_POSIX) += os-posix.o
 common-obj-$(CONFIG_LINUX) += fsdev/
 
 common-obj-y += migration.o migration-tcp.o
+common-obj-y += vmstate.o
+common-obj-y += qemu-file.o
 common-obj-$(CONFIG_RDMA) += migration-rdma.o
 common-obj-y += qemu-char.o #aio.o
 common-obj-y += block-migration.o

VERSION

@@ -1 +1 @@
-1.7.0
+1.7.50

aio-posix.c

@@ -217,11 +217,6 @@ bool aio_poll(AioContext *ctx, bool blocking)
 
     ctx->walking_handlers--;
 
-    /* early return if we only have the aio_notify() fd */
-    if (ctx->pollfds->len == 1) {
-        return progress;
-    }
-
     /* wait until next event */
     ret = qemu_poll_ns((GPollFD *)ctx->pollfds->data,
                          ctx->pollfds->len,

aio-win32.c

@@ -161,11 +161,6 @@ bool aio_poll(AioContext *ctx, bool blocking)
 
     ctx->walking_handlers--;
 
-    /* early return if we only have the aio_notify() fd */
-    if (count == 1) {
-        return progress;
-    }
-
     /* wait until next event */
     while (count > 0) {
         int ret;

arch_init.c

@@ -48,7 +48,9 @@
 #include "qmp-commands.h"
 #include "trace.h"
 #include "exec/cpu-all.h"
+#include "exec/ram_addr.h"
 #include "hw/acpi/acpi.h"
+#include "qemu/host-utils.h"
 
 #ifdef DEBUG_ARCH_INIT
 #define DPRINTF(fmt, ...) \
@@ -359,11 +361,10 @@ ram_addr_t migration_bitmap_find_and_reset_dirty(MemoryRegion *mr,
     return (next - base) << TARGET_PAGE_BITS;
 }
 
-static inline bool migration_bitmap_set_dirty(MemoryRegion *mr,
-                                              ram_addr_t offset)
+static inline bool migration_bitmap_set_dirty(ram_addr_t addr)
 {
     bool ret;
-    int nr = (mr->ram_addr + offset) >> TARGET_PAGE_BITS;
+    int nr = addr >> TARGET_PAGE_BITS;
 
     ret = test_and_set_bit(nr, migration_bitmap);
 
@@ -373,12 +374,47 @@ static inline bool migration_bitmap_set_dirty(MemoryRegion *mr,
     return ret;
 }
 
+static void migration_bitmap_sync_range(ram_addr_t start, ram_addr_t length)
+{
+    ram_addr_t addr;
+    unsigned long page = BIT_WORD(start >> TARGET_PAGE_BITS);
+
+    /* start address is aligned at the start of a word? */
+    if (((page * BITS_PER_LONG) << TARGET_PAGE_BITS) == start) {
+        int k;
+        int nr = BITS_TO_LONGS(length >> TARGET_PAGE_BITS);
+        unsigned long *src = ram_list.dirty_memory[DIRTY_MEMORY_MIGRATION];
+
+        for (k = page; k < page + nr; k++) {
+            if (src[k]) {
+                unsigned long new_dirty;
+                new_dirty = ~migration_bitmap[k];
+                migration_bitmap[k] |= src[k];
+                new_dirty &= src[k];
+                migration_dirty_pages += ctpopl(new_dirty);
+                src[k] = 0;
+            }
+        }
+    } else {
+        for (addr = 0; addr < length; addr += TARGET_PAGE_SIZE) {
+            if (cpu_physical_memory_get_dirty(start + addr,
+                                              TARGET_PAGE_SIZE,
+                                              DIRTY_MEMORY_MIGRATION)) {
+                cpu_physical_memory_reset_dirty(start + addr,
+                                                TARGET_PAGE_SIZE,
+                                                DIRTY_MEMORY_MIGRATION);
+                migration_bitmap_set_dirty(start + addr);
+            }
+        }
+    }
+}
+
+
 /* Needs iothread lock! */
 
 static void migration_bitmap_sync(void)
 {
     RAMBlock *block;
-    ram_addr_t addr;
     uint64_t num_dirty_pages_init = migration_dirty_pages;
     MigrationState *s = migrate_get_current();
     static int64_t start_time;
@@ -399,13 +435,7 @@ static void migration_bitmap_sync(void)
     address_space_sync_dirty_bitmap(&address_space_memory);
 
     QTAILQ_FOREACH(block, &ram_list.blocks, next) {
-        for (addr = 0; addr < block->length; addr += TARGET_PAGE_SIZE) {
-            if (memory_region_test_and_clear_dirty(block->mr,
-                                                   addr, TARGET_PAGE_SIZE,
-                                                   DIRTY_MEMORY_MIGRATION)) {
-                migration_bitmap_set_dirty(block->mr, addr);
-            }
-        }
+        migration_bitmap_sync_range(block->mr->ram_addr, block->length);
     }
     trace_migration_bitmap_sync_end(migration_dirty_pages
                                     - num_dirty_pages_init);

audio/audio.c

@@ -95,7 +95,7 @@ static struct {
         }
     },
 
-    .period = { .hertz = 250 },
+    .period = { .hertz = 100 },
     .plive = 0,
     .log_to_monitor = 0,
     .try_poll_in = 1,

audio/paaudio.c

@@ -547,11 +547,11 @@ static int qpa_init_out (HWVoiceOut *hw, struct audsettings *as)
     ss.rate = as->freq;
 
     /*
-     * qemu audio tick runs at 250 Hz (by default), so processing
-     * data chunks worth 4 ms of sound should be a good fit.
+     * qemu audio tick runs at 100 Hz (by default), so processing
+     * data chunks worth 10 ms of sound should be a good fit.
      */
-    ba.tlength = pa_usec_to_bytes (4 * 1000, &ss);
-    ba.minreq = pa_usec_to_bytes (2 * 1000, &ss);
+    ba.tlength = pa_usec_to_bytes (10 * 1000, &ss);
+    ba.minreq = pa_usec_to_bytes (5 * 1000, &ss);
     ba.maxlength = -1;
     ba.prebuf = -1;
 

backends/rng-random.c

@@ -123,15 +123,15 @@ static void rng_random_init(Object *obj)
                             NULL);
 
     s->filename = g_strdup("/dev/random");
+    s->fd = -1;
 }
 
 static void rng_random_finalize(Object *obj)
 {
     RndRandom *s = RNG_RANDOM(obj);
 
-    qemu_set_fd_handler(s->fd, NULL, NULL, NULL);
-
     if (s->fd != -1) {
+        qemu_set_fd_handler(s->fd, NULL, NULL, NULL);
         qemu_close(s->fd);
     }
 

block-migration.c

@@ -58,6 +58,7 @@ typedef struct BlkMigDevState {
     /* Protected by block migration lock.  */
     unsigned long *aio_bitmap;
     int64_t completed_sectors;
+    BdrvDirtyBitmap *dirty_bitmap;
 } BlkMigDevState;
 
 typedef struct BlkMigBlock {
@@ -309,12 +310,21 @@ static int mig_save_device_bulk(QEMUFile *f, BlkMigDevState *bmds)
 
 /* Called with iothread lock taken.  */
 
-static void set_dirty_tracking(int enable)
+static void set_dirty_tracking(void)
 {
     BlkMigDevState *bmds;
 
     QSIMPLEQ_FOREACH(bmds, &block_mig_state.bmds_list, entry) {
-        bdrv_set_dirty_tracking(bmds->bs, enable ? BLOCK_SIZE : 0);
+        bmds->dirty_bitmap = bdrv_create_dirty_bitmap(bmds->bs, BLOCK_SIZE);
+    }
+}
+
+static void unset_dirty_tracking(void)
+{
+    BlkMigDevState *bmds;
+
+    QSIMPLEQ_FOREACH(bmds, &block_mig_state.bmds_list, entry) {
+        bdrv_release_dirty_bitmap(bmds->bs, bmds->dirty_bitmap);
     }
 }
 
@@ -432,7 +442,7 @@ static int mig_save_device_dirty(QEMUFile *f, BlkMigDevState *bmds,
         } else {
             blk_mig_unlock();
         }
-        if (bdrv_get_dirty(bmds->bs, sector)) {
+        if (bdrv_get_dirty(bmds->bs, bmds->dirty_bitmap, sector)) {
 
             if (total_sectors - sector < BDRV_SECTORS_PER_DIRTY_CHUNK) {
                 nr_sectors = total_sectors - sector;
@@ -554,7 +564,7 @@ static int64_t get_remaining_dirty(void)
     int64_t dirty = 0;
 
     QSIMPLEQ_FOREACH(bmds, &block_mig_state.bmds_list, entry) {
-        dirty += bdrv_get_dirty_count(bmds->bs);
+        dirty += bdrv_get_dirty_count(bmds->bs, bmds->dirty_bitmap);
     }
 
     return dirty << BDRV_SECTOR_BITS;
@@ -569,7 +579,7 @@ static void blk_mig_cleanup(void)
 
     bdrv_drain_all();
 
-    set_dirty_tracking(0);
+    unset_dirty_tracking();
 
     blk_mig_lock();
     while ((bmds = QSIMPLEQ_FIRST(&block_mig_state.bmds_list)) != NULL) {
@@ -604,7 +614,7 @@ static int block_save_setup(QEMUFile *f, void *opaque)
     init_blk_migration(f);
 
     /* start track dirty blocks */
-    set_dirty_tracking(1);
+    set_dirty_tracking();
     qemu_mutex_unlock_iothread();
 
     ret = flush_blks(f);
@@ -780,7 +790,8 @@ static int block_load(QEMUFile *f, void *opaque, int version_id)
             }
 
             if (flags & BLK_MIG_FLAG_ZERO_BLOCK) {
-                ret = bdrv_write_zeroes(bs, addr, nr_sectors);
+                ret = bdrv_write_zeroes(bs, addr, nr_sectors,
+                                        BDRV_REQ_MAY_UNMAP);
             } else {
                 buf = g_malloc(BLOCK_SIZE);
                 qemu_get_buffer(f, buf, BLOCK_SIZE);

block.c

@@ -49,12 +49,12 @@
 #include <windows.h>
 #endif
 
-#define NOT_DONE 0x7fffffff /* used while emulated sync operation in progress */
+struct BdrvDirtyBitmap {
+    HBitmap *bitmap;
+    QLIST_ENTRY(BdrvDirtyBitmap) list;
+};
 
-typedef enum {
-    BDRV_REQ_COPY_ON_READ = 0x1,
-    BDRV_REQ_ZERO_WRITE   = 0x2,
-} BdrvRequestFlags;
+#define NOT_DONE 0x7fffffff /* used while emulated sync operation in progress */
 
 static void bdrv_dev_change_media_cb(BlockDriverState *bs, bool load);
 static BlockDriverAIOCB *bdrv_aio_readv_em(BlockDriverState *bs,
@@ -79,12 +79,13 @@ static BlockDriverAIOCB *bdrv_co_aio_rw_vector(BlockDriverState *bs,
                                                int64_t sector_num,
                                                QEMUIOVector *qiov,
                                                int nb_sectors,
+                                               BdrvRequestFlags flags,
                                                BlockDriverCompletionFunc *cb,
                                                void *opaque,
                                                bool is_write);
 static void coroutine_fn bdrv_co_do_rw(void *opaque);
 static int coroutine_fn bdrv_co_do_write_zeroes(BlockDriverState *bs,
-    int64_t sector_num, int nb_sectors);
+    int64_t sector_num, int nb_sectors, BdrvRequestFlags flags);
 
 static QTAILQ_HEAD(, BlockDriverState) bdrv_states =
     QTAILQ_HEAD_INITIALIZER(bdrv_states);
@@ -323,6 +324,7 @@ BlockDriverState *bdrv_new(const char *device_name)
     BlockDriverState *bs;
 
     bs = g_malloc0(sizeof(BlockDriverState));
+    QLIST_INIT(&bs->dirty_bitmaps);
     pstrcpy(bs->device_name, sizeof(bs->device_name), device_name);
     if (device_name[0] != '\0') {
         QTAILQ_INSERT_TAIL(&bdrv_states, bs, list);
@@ -1052,21 +1054,16 @@ int bdrv_open(BlockDriverState *bs, const char *filename, QDict *options,
         int64_t total_size;
         BlockDriver *bdrv_qcow2;
         QEMUOptionParameter *create_options;
-        char backing_filename[PATH_MAX];
-
-        if (qdict_size(options) != 0) {
-            error_setg(errp, "Can't use snapshot=on with driver-specific options");
-            ret = -EINVAL;
-            goto fail;
-        }
-        assert(filename != NULL);
+        QDict *snapshot_options;
 
         /* if snapshot, we create a temporary backing file and open it
            instead of opening 'filename' directly */
 
-        /* if there is a backing file, use it */
+        /* Get the required size from the image */
         bs1 = bdrv_new("");
-        ret = bdrv_open(bs1, filename, NULL, 0, drv, &local_err);
+        QINCREF(options);
+        ret = bdrv_open(bs1, filename, options, BDRV_O_NO_BACKING,
+                        drv, &local_err);
         if (ret < 0) {
             bdrv_unref(bs1);
             goto fail;
@@ -1075,33 +1072,18 @@ int bdrv_open(BlockDriverState *bs, const char *filename, QDict *options,
 
         bdrv_unref(bs1);
 
+        /* Create the temporary image */
         ret = get_tmp_filename(tmp_filename, sizeof(tmp_filename));
         if (ret < 0) {
             error_setg_errno(errp, -ret, "Could not get temporary filename");
             goto fail;
         }
 
-        /* Real path is meaningless for protocols */
-        if (path_has_protocol(filename)) {
-            snprintf(backing_filename, sizeof(backing_filename),
-                     "%s", filename);
-        } else if (!realpath(filename, backing_filename)) {
-            ret = -errno;
-            error_setg_errno(errp, errno, "Could not resolve path '%s'", filename);
-            goto fail;
-        }
-
         bdrv_qcow2 = bdrv_find_format("qcow2");
         create_options = parse_option_parameters("", bdrv_qcow2->create_options,
                                                  NULL);
 
         set_option_parameter_int(create_options, BLOCK_OPT_SIZE, total_size);
-        set_option_parameter(create_options, BLOCK_OPT_BACKING_FILE,
-                             backing_filename);
-        if (drv) {
-            set_option_parameter(create_options, BLOCK_OPT_BACKING_FMT,
-                drv->format_name);
-        }
 
         ret = bdrv_create(bdrv_qcow2, tmp_filename, create_options, &local_err);
         free_option_parameters(create_options);
@@ -1114,6 +1096,22 @@ int bdrv_open(BlockDriverState *bs, const char *filename, QDict *options,
             goto fail;
         }
 
+        /* Prepare a new options QDict for the temporary file, where user
+         * options refer to the backing file */
+        if (filename) {
+            qdict_put(options, "file.filename", qstring_from_str(filename));
+        }
+        if (drv) {
+            qdict_put(options, "driver", qstring_from_str(drv->format_name));
+        }
+
+        snapshot_options = qdict_new();
+        qdict_put(snapshot_options, "backing", options);
+        qdict_flatten(snapshot_options);
+
+        bs->options = snapshot_options;
+        options = qdict_clone_shallow(bs->options);
+
         filename = tmp_filename;
         drv = bdrv_qcow2;
         bs->is_temporary = 1;
@@ -1559,13 +1557,8 @@ void bdrv_drain_all(void)
     BlockDriverState *bs;
 
     while (busy) {
-        /* FIXME: We do not have timer support here, so this is effectively
-         * a busy wait.
-         */
         QTAILQ_FOREACH(bs, &bdrv_states, list) {
-            if (bdrv_start_throttled_reqs(bs)) {
-                busy = true;
-            }
+            bdrv_start_throttled_reqs(bs);
         }
 
         busy = bdrv_requests_pending_all();
@@ -1622,7 +1615,7 @@ static void bdrv_move_feature_fields(BlockDriverState *bs_dest,
     bs_dest->iostatus           = bs_src->iostatus;
 
     /* dirty bitmap */
-    bs_dest->dirty_bitmap       = bs_src->dirty_bitmap;
+    bs_dest->dirty_bitmaps      = bs_src->dirty_bitmaps;
 
     /* reference count */
     bs_dest->refcnt             = bs_src->refcnt;
@@ -1655,7 +1648,7 @@ void bdrv_swap(BlockDriverState *bs_new, BlockDriverState *bs_old)
 
     /* bs_new must be anonymous and shouldn't have anything fancy enabled */
     assert(bs_new->device_name[0] == '\0');
-    assert(bs_new->dirty_bitmap == NULL);
+    assert(QLIST_EMPTY(&bs_new->dirty_bitmaps));
     assert(bs_new->job == NULL);
     assert(bs_new->dev == NULL);
     assert(bs_new->in_use == 0);
@@ -1716,6 +1709,7 @@ static void bdrv_delete(BlockDriverState *bs)
     assert(!bs->job);
     assert(!bs->in_use);
     assert(!bs->refcnt);
+    assert(QLIST_EMPTY(&bs->dirty_bitmaps));
 
     bdrv_close(bs);
 
@@ -2397,10 +2391,53 @@ int bdrv_writev(BlockDriverState *bs, int64_t sector_num, QEMUIOVector *qiov)
     return bdrv_rwv_co(bs, sector_num, qiov, true, 0);
 }
 
-int bdrv_write_zeroes(BlockDriverState *bs, int64_t sector_num, int nb_sectors)
+int bdrv_write_zeroes(BlockDriverState *bs, int64_t sector_num,
+                      int nb_sectors, BdrvRequestFlags flags)
 {
     return bdrv_rw_co(bs, sector_num, NULL, nb_sectors, true,
-                      BDRV_REQ_ZERO_WRITE);
+                      BDRV_REQ_ZERO_WRITE | flags);
+}
+
+/*
+ * Completely zero out a block device with the help of bdrv_write_zeroes.
+ * The operation is sped up by checking the block status and only writing
+ * zeroes to the device if they currently do not return zeroes. Optional
+ * flags are passed through to bdrv_write_zeroes (e.g. BDRV_REQ_MAY_UNMAP).
+ *
+ * Returns < 0 on error, 0 on success. For error codes see bdrv_write().
+ */
+int bdrv_make_zero(BlockDriverState *bs, BdrvRequestFlags flags)
+{
+    int64_t target_size = bdrv_getlength(bs) / BDRV_SECTOR_SIZE;
+    int64_t ret, nb_sectors, sector_num = 0;
+    int n;
+
+    for (;;) {
+        nb_sectors = target_size - sector_num;
+        if (nb_sectors <= 0) {
+            return 0;
+        }
+        if (nb_sectors > INT_MAX) {
+            nb_sectors = INT_MAX;
+        }
+        ret = bdrv_get_block_status(bs, sector_num, nb_sectors, &n);
+        if (ret < 0) {
+            error_report("error getting block status at sector %" PRId64 ": %s",
+                         sector_num, strerror(-ret));
+            return ret;
+        }
+        if (ret & BDRV_BLOCK_ZERO) {
+            sector_num += n;
+            continue;
+        }
+        ret = bdrv_write_zeroes(bs, sector_num, n, flags);
+        if (ret < 0) {
+            error_report("error writing zeroes at sector %" PRId64 ": %s",
+                         sector_num, strerror(-ret));
+            return ret;
+        }
+        sector_num += n;
+    }
 }
 
 int bdrv_pread(BlockDriverState *bs, int64_t offset,
@@ -2582,7 +2619,7 @@ static int coroutine_fn bdrv_co_do_copy_on_readv(BlockDriverState *bs,
     if (drv->bdrv_co_write_zeroes &&
         buffer_is_zero(bounce_buffer, iov.iov_len)) {
         ret = bdrv_co_do_write_zeroes(bs, cluster_sector_num,
-                                      cluster_nb_sectors);
+                                      cluster_nb_sectors, 0);
     } else {
         /* This does not change the data on the disk, it is not necessary
          * to flush even in cache=writethrough mode.
@@ -2715,33 +2752,77 @@ int coroutine_fn bdrv_co_copy_on_readv(BlockDriverState *bs,
                             BDRV_REQ_COPY_ON_READ);
 }
 
+/* if no limit is specified in the BlockLimits use a default
+ * of 32768 512-byte sectors (16 MiB) per request.
+ */
+#define MAX_WRITE_ZEROES_DEFAULT 32768
+
 static int coroutine_fn bdrv_co_do_write_zeroes(BlockDriverState *bs,
-    int64_t sector_num, int nb_sectors)
+    int64_t sector_num, int nb_sectors, BdrvRequestFlags flags)
 {
     BlockDriver *drv = bs->drv;
     QEMUIOVector qiov;
-    struct iovec iov;
-    int ret;
+    struct iovec iov = {0};
+    int ret = 0;
 
-    /* TODO Emulate only part of misaligned requests instead of letting block
-     * drivers return -ENOTSUP and emulate everything */
+    int max_write_zeroes = bs->bl.max_write_zeroes ?
+                           bs->bl.max_write_zeroes : MAX_WRITE_ZEROES_DEFAULT;
 
-    /* First try the efficient write zeroes operation */
-    if (drv->bdrv_co_write_zeroes) {
-        ret = drv->bdrv_co_write_zeroes(bs, sector_num, nb_sectors);
-        if (ret != -ENOTSUP) {
-            return ret;
+    while (nb_sectors > 0 && !ret) {
+        int num = nb_sectors;
+
+        /* Align request.  Block drivers can expect the "bulk" of the request
+         * to be aligned.
+         */
+        if (bs->bl.write_zeroes_alignment
+            && num > bs->bl.write_zeroes_alignment) {
+            if (sector_num % bs->bl.write_zeroes_alignment != 0) {
+                /* Make a small request up to the first aligned sector.  */
+                num = bs->bl.write_zeroes_alignment;
+                num -= sector_num % bs->bl.write_zeroes_alignment;
+            } else if ((sector_num + num) % bs->bl.write_zeroes_alignment != 0) {
+                /* Shorten the request to the last aligned sector.  num cannot
+                 * underflow because num > bs->bl.write_zeroes_alignment.
+                 */
+                num -= (sector_num + num) % bs->bl.write_zeroes_alignment;
+            }
         }
+
+        /* limit request size */
+        if (num > max_write_zeroes) {
+            num = max_write_zeroes;
+        }
+
+        ret = -ENOTSUP;
+        /* First try the efficient write zeroes operation */
+        if (drv->bdrv_co_write_zeroes) {
+            ret = drv->bdrv_co_write_zeroes(bs, sector_num, num, flags);
+        }
+
+        if (ret == -ENOTSUP) {
+            /* Fall back to bounce buffer if write zeroes is unsupported */
+            iov.iov_len = num * BDRV_SECTOR_SIZE;
+            if (iov.iov_base == NULL) {
+                iov.iov_base = qemu_blockalign(bs, num * BDRV_SECTOR_SIZE);
+                memset(iov.iov_base, 0, num * BDRV_SECTOR_SIZE);
+            }
+            qemu_iovec_init_external(&qiov, &iov, 1);
+
+            ret = drv->bdrv_co_writev(bs, sector_num, num, &qiov);
+
+            /* Keep bounce buffer around if it is big enough for all
+             * all future requests.
+             */
+            if (num < max_write_zeroes) {
+                qemu_vfree(iov.iov_base);
+                iov.iov_base = NULL;
+            }
+        }
+
+        sector_num += num;
+        nb_sectors -= num;
     }
 
-    /* Fall back to bounce buffer if write zeroes is unsupported */
-    iov.iov_len  = nb_sectors * BDRV_SECTOR_SIZE;
-    iov.iov_base = qemu_blockalign(bs, iov.iov_len);
-    memset(iov.iov_base, 0, iov.iov_len);
-    qemu_iovec_init_external(&qiov, &iov, 1);
-
-    ret = drv->bdrv_co_writev(bs, sector_num, nb_sectors, &qiov);
-
     qemu_vfree(iov.iov_base);
     return ret;
 }
@@ -2783,7 +2864,7 @@ static int coroutine_fn bdrv_co_do_writev(BlockDriverState *bs,
     if (ret < 0) {
         /* Do nothing, write notifier decided to fail this request */
     } else if (flags & BDRV_REQ_ZERO_WRITE) {
-        ret = bdrv_co_do_write_zeroes(bs, sector_num, nb_sectors);
+        ret = bdrv_co_do_write_zeroes(bs, sector_num, nb_sectors, flags);
     } else {
         ret = drv->bdrv_co_writev(bs, sector_num, nb_sectors, qiov);
     }
@@ -2792,9 +2873,7 @@ static int coroutine_fn bdrv_co_do_writev(BlockDriverState *bs,
         ret = bdrv_co_flush(bs);
     }
 
-    if (bs->dirty_bitmap) {
-        bdrv_set_dirty(bs, sector_num, nb_sectors);
-    }
+    bdrv_set_dirty(bs, sector_num, nb_sectors);
 
     if (bs->wr_highest_sector < sector_num + nb_sectors - 1) {
         bs->wr_highest_sector = sector_num + nb_sectors - 1;
@@ -2817,12 +2896,17 @@ int coroutine_fn bdrv_co_writev(BlockDriverState *bs, int64_t sector_num,
 }
 
 int coroutine_fn bdrv_co_write_zeroes(BlockDriverState *bs,
-                                      int64_t sector_num, int nb_sectors)
+                                      int64_t sector_num, int nb_sectors,
+                                      BdrvRequestFlags flags)
 {
-    trace_bdrv_co_write_zeroes(bs, sector_num, nb_sectors);
+    trace_bdrv_co_write_zeroes(bs, sector_num, nb_sectors, flags);
+
+    if (!(bs->open_flags & BDRV_O_UNMAP)) {
+        flags &= ~BDRV_REQ_MAY_UNMAP;
+    }
 
     return bdrv_co_do_writev(bs, sector_num, nb_sectors, NULL,
-                             BDRV_REQ_ZERO_WRITE);
+                             BDRV_REQ_ZERO_WRITE | flags);
 }
 
 /**
@@ -3102,6 +3186,36 @@ int bdrv_has_zero_init(BlockDriverState *bs)
     return 0;
 }
 
+bool bdrv_unallocated_blocks_are_zero(BlockDriverState *bs)
+{
+    BlockDriverInfo bdi;
+
+    if (bs->backing_hd) {
+        return false;
+    }
+
+    if (bdrv_get_info(bs, &bdi) == 0) {
+        return bdi.unallocated_blocks_are_zero;
+    }
+
+    return false;
+}
+
+bool bdrv_can_write_zeroes_with_unmap(BlockDriverState *bs)
+{
+    BlockDriverInfo bdi;
+
+    if (bs->backing_hd || !(bs->open_flags & BDRV_O_UNMAP)) {
+        return false;
+    }
+
+    if (bdrv_get_info(bs, &bdi) == 0) {
+        return bdi.can_write_zeroes_with_unmap;
+    }
+
+    return false;
+}
+
 typedef struct BdrvCoGetBlockStatusData {
     BlockDriverState *bs;
     BlockDriverState *base;
@@ -3171,8 +3285,8 @@ static int64_t coroutine_fn bdrv_co_get_block_status(BlockDriverState *bs,
                                      *pnum, pnum);
     }
 
-    if (!(ret & BDRV_BLOCK_DATA)) {
-        if (bdrv_has_zero_init(bs)) {
+    if (!(ret & BDRV_BLOCK_DATA) && !(ret & BDRV_BLOCK_ZERO)) {
+        if (bdrv_unallocated_blocks_are_zero(bs)) {
             ret |= BDRV_BLOCK_ZERO;
         } else if (bs->backing_hd) {
             BlockDriverState *bs2 = bs->backing_hd;
@@ -3330,7 +3444,7 @@ int bdrv_write_compressed(BlockDriverState *bs, int64_t sector_num,
     if (bdrv_check_request(bs, sector_num, nb_sectors))
         return -EIO;
 
-    assert(!bs->dirty_bitmap);
+    assert(QLIST_EMPTY(&bs->dirty_bitmaps));
 
     return drv->bdrv_write_compressed(bs, sector_num, buf, nb_sectors);
 }
@@ -3419,6 +3533,19 @@ int bdrv_debug_breakpoint(BlockDriverState *bs, const char *event,
     return -ENOTSUP;
 }
 
+int bdrv_debug_remove_breakpoint(BlockDriverState *bs, const char *tag)
+{
+    while (bs && bs->drv && !bs->drv->bdrv_debug_remove_breakpoint) {
+        bs = bs->file;
+    }
+
+    if (bs && bs->drv && bs->drv->bdrv_debug_remove_breakpoint) {
+        return bs->drv->bdrv_debug_remove_breakpoint(bs, tag);
+    }
+
+    return -ENOTSUP;
+}
+
 int bdrv_debug_resume(BlockDriverState *bs, const char *tag)
 {
     while (bs && bs->drv && !bs->drv->bdrv_debug_resume) {
@@ -3554,7 +3681,7 @@ BlockDriverAIOCB *bdrv_aio_readv(BlockDriverState *bs, int64_t sector_num,
 {
     trace_bdrv_aio_readv(bs, sector_num, nb_sectors, opaque);
 
-    return bdrv_co_aio_rw_vector(bs, sector_num, qiov, nb_sectors,
+    return bdrv_co_aio_rw_vector(bs, sector_num, qiov, nb_sectors, 0,
                                  cb, opaque, false);
 }
 
@@ -3564,7 +3691,18 @@ BlockDriverAIOCB *bdrv_aio_writev(BlockDriverState *bs, int64_t sector_num,
 {
     trace_bdrv_aio_writev(bs, sector_num, nb_sectors, opaque);
 
-    return bdrv_co_aio_rw_vector(bs, sector_num, qiov, nb_sectors,
+    return bdrv_co_aio_rw_vector(bs, sector_num, qiov, nb_sectors, 0,
+                                 cb, opaque, true);
+}
+
+BlockDriverAIOCB *bdrv_aio_write_zeroes(BlockDriverState *bs,
+        int64_t sector_num, int nb_sectors, BdrvRequestFlags flags,
+        BlockDriverCompletionFunc *cb, void *opaque)
+{
+    trace_bdrv_aio_write_zeroes(bs, sector_num, nb_sectors, flags, opaque);
+
+    return bdrv_co_aio_rw_vector(bs, sector_num, NULL, nb_sectors,
+                                 BDRV_REQ_ZERO_WRITE | flags,
                                  cb, opaque, true);
 }
 
@@ -3736,8 +3874,10 @@ int bdrv_aio_multiwrite(BlockDriverState *bs, BlockRequest *reqs, int num_reqs)
     /* Run the aio requests. */
     mcb->num_requests = num_reqs;
     for (i = 0; i < num_reqs; i++) {
-        bdrv_aio_writev(bs, reqs[i].sector, reqs[i].qiov,
-            reqs[i].nb_sectors, multiwrite_cb, mcb);
+        bdrv_co_aio_rw_vector(bs, reqs[i].sector, reqs[i].qiov,
+                              reqs[i].nb_sectors, reqs[i].flags,
+                              multiwrite_cb, mcb,
+                              true);
     }
 
     return 0;
@@ -3879,10 +4019,10 @@ static void coroutine_fn bdrv_co_do_rw(void *opaque)
 
     if (!acb->is_write) {
         acb->req.error = bdrv_co_do_readv(bs, acb->req.sector,
-            acb->req.nb_sectors, acb->req.qiov, 0);
+            acb->req.nb_sectors, acb->req.qiov, acb->req.flags);
     } else {
         acb->req.error = bdrv_co_do_writev(bs, acb->req.sector,
-            acb->req.nb_sectors, acb->req.qiov, 0);
+            acb->req.nb_sectors, acb->req.qiov, acb->req.flags);
     }
 
     acb->bh = qemu_bh_new(bdrv_co_em_bh, acb);
@@ -3893,6 +4033,7 @@ static BlockDriverAIOCB *bdrv_co_aio_rw_vector(BlockDriverState *bs,
                                                int64_t sector_num,
                                                QEMUIOVector *qiov,
                                                int nb_sectors,
+                                               BdrvRequestFlags flags,
                                                BlockDriverCompletionFunc *cb,
                                                void *opaque,
                                                bool is_write)
@@ -3904,6 +4045,7 @@ static BlockDriverAIOCB *bdrv_co_aio_rw_vector(BlockDriverState *bs,
     acb->req.sector = sector_num;
     acb->req.nb_sectors = nb_sectors;
     acb->req.qiov = qiov;
+    acb->req.flags = flags;
     acb->is_write = is_write;
     acb->done = NULL;
 
@@ -4179,9 +4321,16 @@ static void coroutine_fn bdrv_discard_co_entry(void *opaque)
     rwco->ret = bdrv_co_discard(rwco->bs, rwco->sector_num, rwco->nb_sectors);
 }
 
+/* if no limit is specified in the BlockLimits use a default
+ * of 32768 512-byte sectors (16 MiB) per request.
+ */
+#define MAX_DISCARD_DEFAULT 32768
+
 int coroutine_fn bdrv_co_discard(BlockDriverState *bs, int64_t sector_num,
                                  int nb_sectors)
 {
+    int max_discard;
+
     if (!bs->drv) {
         return -ENOMEDIUM;
     } else if (bdrv_check_request(bs, sector_num, nb_sectors)) {
@@ -4190,34 +4339,62 @@ int coroutine_fn bdrv_co_discard(BlockDriverState *bs, int64_t sector_num,
         return -EROFS;
     }
 
-    if (bs->dirty_bitmap) {
-        bdrv_reset_dirty(bs, sector_num, nb_sectors);
-    }
+    bdrv_reset_dirty(bs, sector_num, nb_sectors);
 
     /* Do nothing if disabled.  */
     if (!(bs->open_flags & BDRV_O_UNMAP)) {
         return 0;
     }
 
-    if (bs->drv->bdrv_co_discard) {
-        return bs->drv->bdrv_co_discard(bs, sector_num, nb_sectors);
-    } else if (bs->drv->bdrv_aio_discard) {
-        BlockDriverAIOCB *acb;
-        CoroutineIOCompletion co = {
-            .coroutine = qemu_coroutine_self(),
-        };
-
-        acb = bs->drv->bdrv_aio_discard(bs, sector_num, nb_sectors,
-                                        bdrv_co_io_em_complete, &co);
-        if (acb == NULL) {
-            return -EIO;
-        } else {
-            qemu_coroutine_yield();
-            return co.ret;
-        }
-    } else {
+    if (!bs->drv->bdrv_co_discard && !bs->drv->bdrv_aio_discard) {
         return 0;
     }
+
+    max_discard = bs->bl.max_discard ?  bs->bl.max_discard : MAX_DISCARD_DEFAULT;
+    while (nb_sectors > 0) {
+        int ret;
+        int num = nb_sectors;
+
+        /* align request */
+        if (bs->bl.discard_alignment &&
+            num >= bs->bl.discard_alignment &&
+            sector_num % bs->bl.discard_alignment) {
+            if (num > bs->bl.discard_alignment) {
+                num = bs->bl.discard_alignment;
+            }
+            num -= sector_num % bs->bl.discard_alignment;
+        }
+
+        /* limit request size */
+        if (num > max_discard) {
+            num = max_discard;
+        }
+
+        if (bs->drv->bdrv_co_discard) {
+            ret = bs->drv->bdrv_co_discard(bs, sector_num, num);
+        } else {
+            BlockDriverAIOCB *acb;
+            CoroutineIOCompletion co = {
+                .coroutine = qemu_coroutine_self(),
+            };
+
+            acb = bs->drv->bdrv_aio_discard(bs, sector_num, nb_sectors,
+                                            bdrv_co_io_em_complete, &co);
+            if (acb == NULL) {
+                return -EIO;
+            } else {
+                qemu_coroutine_yield();
+                ret = co.ret;
+            }
+        }
+        if (ret && ret != -ENOTSUP) {
+            return ret;
+        }
+
+        sector_num += num;
+        nb_sectors -= num;
+    }
+    return 0;
 }
 
 int bdrv_discard(BlockDriverState *bs, int64_t sector_num, int nb_sectors)
@@ -4354,60 +4531,92 @@ bool bdrv_qiov_is_aligned(BlockDriverState *bs, QEMUIOVector *qiov)
     return true;
 }
 
-void bdrv_set_dirty_tracking(BlockDriverState *bs, int granularity)
+BdrvDirtyBitmap *bdrv_create_dirty_bitmap(BlockDriverState *bs, int granularity)
 {
     int64_t bitmap_size;
+    BdrvDirtyBitmap *bitmap;
 
     assert((granularity & (granularity - 1)) == 0);
 
-    if (granularity) {
-        granularity >>= BDRV_SECTOR_BITS;
-        assert(!bs->dirty_bitmap);
-        bitmap_size = (bdrv_getlength(bs) >> BDRV_SECTOR_BITS);
-        bs->dirty_bitmap = hbitmap_alloc(bitmap_size, ffs(granularity) - 1);
-    } else {
-        if (bs->dirty_bitmap) {
-            hbitmap_free(bs->dirty_bitmap);
-            bs->dirty_bitmap = NULL;
+    granularity >>= BDRV_SECTOR_BITS;
+    assert(granularity);
+    bitmap_size = (bdrv_getlength(bs) >> BDRV_SECTOR_BITS);
+    bitmap = g_malloc0(sizeof(BdrvDirtyBitmap));
+    bitmap->bitmap = hbitmap_alloc(bitmap_size, ffs(granularity) - 1);
+    QLIST_INSERT_HEAD(&bs->dirty_bitmaps, bitmap, list);
+    return bitmap;
+}
+
+void bdrv_release_dirty_bitmap(BlockDriverState *bs, BdrvDirtyBitmap *bitmap)
+{
+    BdrvDirtyBitmap *bm, *next;
+    QLIST_FOREACH_SAFE(bm, &bs->dirty_bitmaps, list, next) {
+        if (bm == bitmap) {
+            QLIST_REMOVE(bitmap, list);
+            hbitmap_free(bitmap->bitmap);
+            g_free(bitmap);
+            return;
         }
     }
 }
 
-int bdrv_get_dirty(BlockDriverState *bs, int64_t sector)
+BlockDirtyInfoList *bdrv_query_dirty_bitmaps(BlockDriverState *bs)
 {
-    if (bs->dirty_bitmap) {
-        return hbitmap_get(bs->dirty_bitmap, sector);
+    BdrvDirtyBitmap *bm;
+    BlockDirtyInfoList *list = NULL;
+    BlockDirtyInfoList **plist = &list;
+
+    QLIST_FOREACH(bm, &bs->dirty_bitmaps, list) {
+        BlockDirtyInfo *info = g_malloc0(sizeof(BlockDirtyInfo));
+        BlockDirtyInfoList *entry = g_malloc0(sizeof(BlockDirtyInfoList));
+        info->count = bdrv_get_dirty_count(bs, bm);
+        info->granularity =
+            ((int64_t) BDRV_SECTOR_SIZE << hbitmap_granularity(bm->bitmap));
+        entry->value = info;
+        *plist = entry;
+        plist = &entry->next;
+    }
+
+    return list;
+}
+
+int bdrv_get_dirty(BlockDriverState *bs, BdrvDirtyBitmap *bitmap, int64_t sector)
+{
+    if (bitmap) {
+        return hbitmap_get(bitmap->bitmap, sector);
     } else {
         return 0;
     }
 }
 
-void bdrv_dirty_iter_init(BlockDriverState *bs, HBitmapIter *hbi)
+void bdrv_dirty_iter_init(BlockDriverState *bs,
+                          BdrvDirtyBitmap *bitmap, HBitmapIter *hbi)
 {
-    hbitmap_iter_init(hbi, bs->dirty_bitmap, 0);
+    hbitmap_iter_init(hbi, bitmap->bitmap, 0);
 }
 
 void bdrv_set_dirty(BlockDriverState *bs, int64_t cur_sector,
                     int nr_sectors)
 {
-    hbitmap_set(bs->dirty_bitmap, cur_sector, nr_sectors);
-}
-
-void bdrv_reset_dirty(BlockDriverState *bs, int64_t cur_sector,
-                      int nr_sectors)
-{
-    hbitmap_reset(bs->dirty_bitmap, cur_sector, nr_sectors);
-}
-
-int64_t bdrv_get_dirty_count(BlockDriverState *bs)
-{
-    if (bs->dirty_bitmap) {
-        return hbitmap_count(bs->dirty_bitmap);
-    } else {
-        return 0;
+    BdrvDirtyBitmap *bitmap;
+    QLIST_FOREACH(bitmap, &bs->dirty_bitmaps, list) {
+        hbitmap_set(bitmap->bitmap, cur_sector, nr_sectors);
     }
 }
 
+void bdrv_reset_dirty(BlockDriverState *bs, int64_t cur_sector, int nr_sectors)
+{
+    BdrvDirtyBitmap *bitmap;
+    QLIST_FOREACH(bitmap, &bs->dirty_bitmaps, list) {
+        hbitmap_reset(bitmap->bitmap, cur_sector, nr_sectors);
+    }
+}
+
+int64_t bdrv_get_dirty_count(BlockDriverState *bs, BdrvDirtyBitmap *bitmap)
+{
+    return hbitmap_count(bitmap->bitmap);
+}
+
 /* Get a reference to bs */
 void bdrv_ref(BlockDriverState *bs)
 {
@@ -4504,7 +4713,6 @@ void bdrv_img_create(const char *filename, const char *fmt,
 {
     QEMUOptionParameter *param = NULL, *create_options = NULL;
     QEMUOptionParameter *backing_fmt, *backing_file, *size;
-    BlockDriverState *bs = NULL;
     BlockDriver *drv, *proto_drv;
     BlockDriver *backing_drv = NULL;
     Error *local_err = NULL;
@@ -4583,6 +4791,7 @@ void bdrv_img_create(const char *filename, const char *fmt,
     size = get_option_parameter(param, BLOCK_OPT_SIZE);
     if (size && size->value.n == -1) {
         if (backing_file && backing_file->value.s) {
+            BlockDriverState *bs;
             uint64_t size;
             char buf[32];
             int back_flags;
@@ -4601,6 +4810,7 @@ void bdrv_img_create(const char *filename, const char *fmt,
                                  error_get_pretty(local_err));
                 error_free(local_err);
                 local_err = NULL;
+                bdrv_unref(bs);
                 goto out;
             }
             bdrv_get_geometry(bs, &size);
@@ -4608,6 +4818,8 @@ void bdrv_img_create(const char *filename, const char *fmt,
 
             snprintf(buf, sizeof(buf), "%" PRId64, size);
             set_option_parameter(param, BLOCK_OPT_SIZE, buf);
+
+            bdrv_unref(bs);
         } else {
             error_setg(errp, "Image creation needs a size parameter");
             goto out;
@@ -4638,9 +4850,6 @@ out:
     free_option_parameters(create_options);
     free_option_parameters(param);
 
-    if (bs) {
-        bdrv_unref(bs);
-    }
     if (error_is_set(&local_err)) {
         error_propagate(errp, local_err);
     }

block/Makefile.objs

@@ -10,7 +10,7 @@ block-obj-$(CONFIG_POSIX) += raw-posix.o
 block-obj-$(CONFIG_LINUX_AIO) += linux-aio.o
 
 ifeq ($(CONFIG_POSIX),y)
-block-obj-y += nbd.o sheepdog.o
+block-obj-y += nbd.o nbd-client.o sheepdog.o
 block-obj-$(CONFIG_LIBISCSI) += iscsi.o
 block-obj-$(CONFIG_CURL) += curl.o
 block-obj-$(CONFIG_RBD) += rbd.o

block/backup.c

@@ -138,7 +138,8 @@ static int coroutine_fn backup_do_cow(BlockDriverState *bs,
 
         if (buffer_is_zero(iov.iov_base, iov.iov_len)) {
             ret = bdrv_co_write_zeroes(job->target,
-                                       start * BACKUP_SECTORS_PER_CLUSTER, n);
+                                       start * BACKUP_SECTORS_PER_CLUSTER,
+                                       n, BDRV_REQ_MAY_UNMAP);
         } else {
             ret = bdrv_co_writev(job->target,
                                  start * BACKUP_SECTORS_PER_CLUSTER, n,

block/blkdebug.c

@@ -359,7 +359,7 @@ static int blkdebug_open(BlockDriverState *bs, QDict *options, int flags,
     const char *filename, *config;
     int ret;
 
-    opts = qemu_opts_create_nofail(&runtime_opts);
+    opts = qemu_opts_create(&runtime_opts, NULL, 0, &error_abort);
     qemu_opts_absorb_qdict(opts, options, &local_err);
     if (error_is_set(&local_err)) {
         error_propagate(errp, local_err);
@@ -594,9 +594,9 @@ static int blkdebug_debug_breakpoint(BlockDriverState *bs, const char *event,
 static int blkdebug_debug_resume(BlockDriverState *bs, const char *tag)
 {
     BDRVBlkdebugState *s = bs->opaque;
-    BlkdebugSuspendedReq *r;
+    BlkdebugSuspendedReq *r, *next;
 
-    QLIST_FOREACH(r, &s->suspended_reqs, next) {
+    QLIST_FOREACH_SAFE(r, &s->suspended_reqs, next, next) {
         if (!strcmp(r->tag, tag)) {
             qemu_coroutine_enter(r->co, NULL);
             return 0;
@@ -605,6 +605,31 @@ static int blkdebug_debug_resume(BlockDriverState *bs, const char *tag)
     return -ENOENT;
 }
 
+static int blkdebug_debug_remove_breakpoint(BlockDriverState *bs,
+                                            const char *tag)
+{
+    BDRVBlkdebugState *s = bs->opaque;
+    BlkdebugSuspendedReq *r, *r_next;
+    BlkdebugRule *rule, *next;
+    int i, ret = -ENOENT;
+
+    for (i = 0; i < BLKDBG_EVENT_MAX; i++) {
+        QLIST_FOREACH_SAFE(rule, &s->rules[i], next, next) {
+            if (rule->action == ACTION_SUSPEND &&
+                !strcmp(rule->options.suspend.tag, tag)) {
+                remove_rule(rule);
+                ret = 0;
+            }
+        }
+    }
+    QLIST_FOREACH_SAFE(r, &s->suspended_reqs, next, r_next) {
+        if (!strcmp(r->tag, tag)) {
+            qemu_coroutine_enter(r->co, NULL);
+            ret = 0;
+        }
+    }
+    return ret;
+}
 
 static bool blkdebug_debug_is_suspended(BlockDriverState *bs, const char *tag)
 {
@@ -639,6 +664,8 @@ static BlockDriver bdrv_blkdebug = {
 
     .bdrv_debug_event           = blkdebug_debug_event,
     .bdrv_debug_breakpoint      = blkdebug_debug_breakpoint,
+    .bdrv_debug_remove_breakpoint
+                                = blkdebug_debug_remove_breakpoint,
     .bdrv_debug_resume          = blkdebug_debug_resume,
     .bdrv_debug_is_suspended    = blkdebug_debug_is_suspended,
 };

block/blkverify.c

@@ -125,7 +125,7 @@ static int blkverify_open(BlockDriverState *bs, QDict *options, int flags,
     const char *filename, *raw;
     int ret;
 
-    opts = qemu_opts_create_nofail(&runtime_opts);
+    opts = qemu_opts_create(&runtime_opts, NULL, 0, &error_abort);
     qemu_opts_absorb_qdict(opts, options, &local_err);
     if (error_is_set(&local_err)) {
         error_propagate(errp, local_err);

block/commit.c

@@ -198,13 +198,7 @@ void commit_start(BlockDriverState *bs, BlockDriverState *base,
         return;
     }
 
-    /* Once we support top == active layer, remove this check */
-    if (top == bs) {
-        error_setg(errp,
-                   "Top image as the active layer is currently unsupported");
-        return;
-    }
-
+    assert(top != bs);
     if (top == base) {
         error_setg(errp, "Invalid files for merge: top and base are the same");
         return;

block/cow.c

@@ -103,40 +103,18 @@ static int cow_open(BlockDriverState *bs, QDict *options, int flags,
     return ret;
 }
 
-/*
- * XXX(hch): right now these functions are extremely inefficient.
- * We should just read the whole bitmap we'll need in one go instead.
- */
-static inline int cow_set_bit(BlockDriverState *bs, int64_t bitnum, bool *first)
+static inline void cow_set_bits(uint8_t *bitmap, int start, int64_t nb_sectors)
 {
-    uint64_t offset = sizeof(struct cow_header_v2) + bitnum / 8;
-    uint8_t bitmap;
-    int ret;
-
-    ret = bdrv_pread(bs->file, offset, &bitmap, sizeof(bitmap));
-    if (ret < 0) {
-       return ret;
-    }
-
-    if (bitmap & (1 << (bitnum % 8))) {
-        return 0;
-    }
-
-    if (*first) {
-        ret = bdrv_flush(bs->file);
-        if (ret < 0) {
-            return ret;
+    int64_t bitnum = start, last = start + nb_sectors;
+    while (bitnum < last) {
+        if ((bitnum & 7) == 0 && bitnum + 8 <= last) {
+            bitmap[bitnum / 8] = 0xFF;
+            bitnum += 8;
+            continue;
         }
-        *first = false;
+        bitmap[bitnum/8] |= (1 << (bitnum % 8));
+        bitnum++;
     }
-
-    bitmap |= (1 << (bitnum % 8));
-
-    ret = bdrv_pwrite(bs->file, offset, &bitmap, sizeof(bitmap));
-    if (ret < 0) {
-       return ret;
-    }
-    return 0;
 }
 
 #define BITS_PER_BITMAP_SECTOR (512 * 8)
@@ -174,18 +152,34 @@ static int coroutine_fn cow_co_is_allocated(BlockDriverState *bs,
 {
     int64_t bitnum = sector_num + sizeof(struct cow_header_v2) * 8;
     uint64_t offset = (bitnum / 8) & -BDRV_SECTOR_SIZE;
-    uint8_t bitmap[BDRV_SECTOR_SIZE];
-    int ret;
-    int changed;
+    bool first = true;
+    int changed = 0, same = 0;
 
-    ret = bdrv_pread(bs->file, offset, &bitmap, sizeof(bitmap));
-    if (ret < 0) {
-        return ret;
-    }
+    do {
+        int ret;
+        uint8_t bitmap[BDRV_SECTOR_SIZE];
 
-    bitnum &= BITS_PER_BITMAP_SECTOR - 1;
-    changed = cow_test_bit(bitnum, bitmap);
-    *num_same = cow_find_streak(bitmap, changed, bitnum, nb_sectors);
+        bitnum &= BITS_PER_BITMAP_SECTOR - 1;
+        int sector_bits = MIN(nb_sectors, BITS_PER_BITMAP_SECTOR - bitnum);
+
+        ret = bdrv_pread(bs->file, offset, &bitmap, sizeof(bitmap));
+        if (ret < 0) {
+            return ret;
+        }
+
+        if (first) {
+            changed = cow_test_bit(bitnum, bitmap);
+            first = false;
+        }
+
+        same += cow_find_streak(bitmap, changed, bitnum, nb_sectors);
+
+        bitnum += sector_bits;
+        nb_sectors -= sector_bits;
+        offset += BDRV_SECTOR_SIZE;
+    } while (nb_sectors);
+
+    *num_same = same;
     return changed;
 }
 
@@ -204,18 +198,52 @@ static int64_t coroutine_fn cow_co_get_block_status(BlockDriverState *bs,
 static int cow_update_bitmap(BlockDriverState *bs, int64_t sector_num,
         int nb_sectors)
 {
-    int error = 0;
-    int i;
+    int64_t bitnum = sector_num + sizeof(struct cow_header_v2) * 8;
+    uint64_t offset = (bitnum / 8) & -BDRV_SECTOR_SIZE;
     bool first = true;
+    int sector_bits;
 
-    for (i = 0; i < nb_sectors; i++) {
-        error = cow_set_bit(bs, sector_num + i, &first);
-        if (error) {
-            break;
+    for ( ; nb_sectors;
+            bitnum += sector_bits,
+            nb_sectors -= sector_bits,
+            offset += BDRV_SECTOR_SIZE) {
+        int ret, set;
+        uint8_t bitmap[BDRV_SECTOR_SIZE];
+
+        bitnum &= BITS_PER_BITMAP_SECTOR - 1;
+        sector_bits = MIN(nb_sectors, BITS_PER_BITMAP_SECTOR - bitnum);
+
+        ret = bdrv_pread(bs->file, offset, &bitmap, sizeof(bitmap));
+        if (ret < 0) {
+            return ret;
+        }
+
+        /* Skip over any already set bits */
+        set = cow_find_streak(bitmap, 1, bitnum, sector_bits);
+        bitnum += set;
+        sector_bits -= set;
+        nb_sectors -= set;
+        if (!sector_bits) {
+            continue;
+        }
+
+        if (first) {
+            ret = bdrv_flush(bs->file);
+            if (ret < 0) {
+                return ret;
+            }
+            first = false;
+        }
+
+        cow_set_bits(bitmap, bitnum, sector_bits);
+
+        ret = bdrv_pwrite(bs->file, offset, &bitmap, sizeof(bitmap));
+        if (ret < 0) {
+            return ret;
         }
     }
 
-    return error;
+    return 0;
 }
 
 static int coroutine_fn cow_read(BlockDriverState *bs, int64_t sector_num,

block/curl.c

@@ -413,7 +413,7 @@ static int curl_open(BlockDriverState *bs, QDict *options, int flags,
         return -EROFS;
     }
 
-    opts = qemu_opts_create_nofail(&runtime_opts);
+    opts = qemu_opts_create(&runtime_opts, NULL, 0, &error_abort);
     qemu_opts_absorb_qdict(opts, options, &local_err);
     if (error_is_set(&local_err)) {
         qerror_report_err(local_err);

block/gluster.c

@@ -298,7 +298,7 @@ static int qemu_gluster_open(BlockDriverState *bs,  QDict *options,
     Error *local_err = NULL;
     const char *filename;
 
-    opts = qemu_opts_create_nofail(&runtime_opts);
+    opts = qemu_opts_create(&runtime_opts, NULL, 0, &error_abort);
     qemu_opts_absorb_qdict(opts, options, &local_err);
     if (error_is_set(&local_err)) {
         qerror_report_err(local_err);

block/iscsi.c

@@ -2,6 +2,7 @@
  * QEMU Block driver for iSCSI images
  *
  * Copyright (c) 2010-2011 Ronnie Sahlberg <ronniesahlberg@gmail.com>
+ * Copyright (c) 2012-2013 Peter Lieven <pl@kamp.de>
  *
  * Permission is hereby granted, free of charge, to any person obtaining a copy
  * of this software and associated documentation files (the "Software"), to deal
@@ -54,8 +55,10 @@ typedef struct IscsiLun {
     QEMUTimer *nop_timer;
     uint8_t lbpme;
     uint8_t lbprz;
+    uint8_t has_write_same;
     struct scsi_inquiry_logical_block_provisioning lbp;
     struct scsi_inquiry_block_limits bl;
+    unsigned char *zeroblock;
 } IscsiLun;
 
 typedef struct IscsiTask {
@@ -65,6 +68,7 @@ typedef struct IscsiTask {
     int do_retry;
     struct scsi_task *task;
     Coroutine *co;
+    QEMUBH *bh;
 } IscsiTask;
 
 typedef struct IscsiAIOCB {
@@ -87,7 +91,6 @@ typedef struct IscsiAIOCB {
 #define NOP_INTERVAL 5000
 #define MAX_NOP_FAILURES 3
 #define ISCSI_CMD_RETRIES 5
-#define ISCSI_MAX_UNMAP 131072
 
 static void
 iscsi_bh_cb(void *p)
@@ -121,6 +124,13 @@ iscsi_schedule_bh(IscsiAIOCB *acb)
     qemu_bh_schedule(acb->bh);
 }
 
+static void iscsi_co_generic_bh_cb(void *opaque)
+{
+    struct IscsiTask *iTask = opaque;
+    qemu_bh_delete(iTask->bh);
+    qemu_coroutine_enter(iTask->co, NULL);
+}
+
 static void
 iscsi_co_generic_cb(struct iscsi_context *iscsi, int status,
                         void *command_data, void *opaque)
@@ -145,7 +155,8 @@ iscsi_co_generic_cb(struct iscsi_context *iscsi, int status,
 
 out:
     if (iTask->co) {
-        qemu_coroutine_enter(iTask->co, NULL);
+        iTask->bh = qemu_bh_new(iscsi_co_generic_bh_cb, iTask);
+        qemu_bh_schedule(iTask->bh);
     }
 }
 
@@ -237,44 +248,6 @@ iscsi_process_write(void *arg)
     iscsi_set_events(iscsilun);
 }
 
-static int
-iscsi_aio_writev_acb(IscsiAIOCB *acb);
-
-static void
-iscsi_aio_write16_cb(struct iscsi_context *iscsi, int status,
-                     void *command_data, void *opaque)
-{
-    IscsiAIOCB *acb = opaque;
-
-    trace_iscsi_aio_write16_cb(iscsi, status, acb, acb->canceled);
-
-    g_free(acb->buf);
-    acb->buf = NULL;
-
-    if (acb->canceled != 0) {
-        return;
-    }
-
-    acb->status = 0;
-    if (status != 0) {
-        if (status == SCSI_STATUS_CHECK_CONDITION
-            && acb->task->sense.key == SCSI_SENSE_UNIT_ATTENTION
-            && acb->retries-- > 0) {
-            scsi_free_scsi_task(acb->task);
-            acb->task = NULL;
-            if (iscsi_aio_writev_acb(acb) == 0) {
-                iscsi_set_events(acb->iscsilun);
-                return;
-            }
-        }
-        error_report("Failed to write16 data to iSCSI lun. %s",
-                     iscsi_get_error(iscsi));
-        acb->status = -EIO;
-    }
-
-    iscsi_schedule_bh(acb);
-}
-
 static int64_t sector_lun2qemu(int64_t sector, IscsiLun *iscsilun)
 {
     return sector * iscsilun->block_size / BDRV_SECTOR_SIZE;
@@ -299,324 +272,175 @@ static bool is_request_lun_aligned(int64_t sector_num, int nb_sectors,
     return 1;
 }
 
-static int
-iscsi_aio_writev_acb(IscsiAIOCB *acb)
+static int coroutine_fn iscsi_co_writev(BlockDriverState *bs,
+                                        int64_t sector_num, int nb_sectors,
+                                        QEMUIOVector *iov)
 {
-    struct iscsi_context *iscsi = acb->iscsilun->iscsi;
-    size_t size;
-    uint32_t num_sectors;
+    IscsiLun *iscsilun = bs->opaque;
+    struct IscsiTask iTask;
     uint64_t lba;
+    uint32_t num_sectors;
+    uint8_t *data = NULL;
+    uint8_t *buf = NULL;
+
+    if (!is_request_lun_aligned(sector_num, nb_sectors, iscsilun)) {
+        return -EINVAL;
+    }
+
+    lba = sector_qemu2lun(sector_num, iscsilun);
+    num_sectors = sector_qemu2lun(nb_sectors, iscsilun);
 #if !defined(LIBISCSI_FEATURE_IOVECTOR)
-    struct iscsi_data data;
-#endif
-    int ret;
-
-    acb->canceled   = 0;
-    acb->bh         = NULL;
-    acb->status     = -EINPROGRESS;
-    acb->buf        = NULL;
-
-    /* this will allow us to get rid of 'buf' completely */
-    size = acb->nb_sectors * BDRV_SECTOR_SIZE;
-
-#if !defined(LIBISCSI_FEATURE_IOVECTOR)
-    data.size = MIN(size, acb->qiov->size);
-
     /* if the iovec only contains one buffer we can pass it directly */
-    if (acb->qiov->niov == 1) {
-        data.data = acb->qiov->iov[0].iov_base;
+    if (iov->niov == 1) {
+        data = iov->iov[0].iov_base;
     } else {
-        acb->buf = g_malloc(data.size);
-        qemu_iovec_to_buf(acb->qiov, 0, acb->buf, data.size);
-        data.data = acb->buf;
+        size_t size = MIN(nb_sectors * BDRV_SECTOR_SIZE, iov->size);
+        buf = g_malloc(size);
+        qemu_iovec_to_buf(iov, 0, buf, size);
+        data = buf;
     }
 #endif
-
-    acb->task = malloc(sizeof(struct scsi_task));
-    if (acb->task == NULL) {
-        error_report("iSCSI: Failed to allocate task for scsi WRITE16 "
-                     "command. %s", iscsi_get_error(iscsi));
-        return -1;
+    iscsi_co_init_iscsitask(iscsilun, &iTask);
+retry:
+    iTask.task = iscsi_write16_task(iscsilun->iscsi, iscsilun->lun, lba,
+                                    data, num_sectors * iscsilun->block_size,
+                                    iscsilun->block_size, 0, 0, 0, 0, 0,
+                                    iscsi_co_generic_cb, &iTask);
+    if (iTask.task == NULL) {
+        g_free(buf);
+        return -EIO;
     }
-    memset(acb->task, 0, sizeof(struct scsi_task));
-
-    acb->task->xfer_dir = SCSI_XFER_WRITE;
-    acb->task->cdb_size = 16;
-    acb->task->cdb[0] = 0x8a;
-    lba = sector_qemu2lun(acb->sector_num, acb->iscsilun);
-    *(uint32_t *)&acb->task->cdb[2]  = htonl(lba >> 32);
-    *(uint32_t *)&acb->task->cdb[6]  = htonl(lba & 0xffffffff);
-    num_sectors = sector_qemu2lun(acb->nb_sectors, acb->iscsilun);
-    *(uint32_t *)&acb->task->cdb[10] = htonl(num_sectors);
-    acb->task->expxferlen = size;
-
 #if defined(LIBISCSI_FEATURE_IOVECTOR)
-    ret = iscsi_scsi_command_async(iscsi, acb->iscsilun->lun, acb->task,
-                                   iscsi_aio_write16_cb,
-                                   NULL,
-                                   acb);
-#else
-    ret = iscsi_scsi_command_async(iscsi, acb->iscsilun->lun, acb->task,
-                                   iscsi_aio_write16_cb,
-                                   &data,
-                                   acb);
+    scsi_task_set_iov_out(iTask.task, (struct scsi_iovec *) iov->iov,
+                          iov->niov);
 #endif
-    if (ret != 0) {
-        scsi_free_scsi_task(acb->task);
-        g_free(acb->buf);
-        return -1;
+    while (!iTask.complete) {
+        iscsi_set_events(iscsilun);
+        qemu_coroutine_yield();
     }
 
-#if defined(LIBISCSI_FEATURE_IOVECTOR)
-    scsi_task_set_iov_out(acb->task, (struct scsi_iovec*) acb->qiov->iov, acb->qiov->niov);
-#endif
+    if (iTask.task != NULL) {
+        scsi_free_scsi_task(iTask.task);
+        iTask.task = NULL;
+    }
+
+    if (iTask.do_retry) {
+        goto retry;
+    }
+
+    g_free(buf);
+
+    if (iTask.status != SCSI_STATUS_GOOD) {
+        return -EIO;
+    }
 
     return 0;
 }
 
-static BlockDriverAIOCB *
-iscsi_aio_writev(BlockDriverState *bs, int64_t sector_num,
-                 QEMUIOVector *qiov, int nb_sectors,
-                 BlockDriverCompletionFunc *cb,
-                 void *opaque)
+static int coroutine_fn iscsi_co_readv(BlockDriverState *bs,
+                                       int64_t sector_num, int nb_sectors,
+                                       QEMUIOVector *iov)
 {
     IscsiLun *iscsilun = bs->opaque;
-    IscsiAIOCB *acb;
-
-    if (!is_request_lun_aligned(sector_num, nb_sectors, iscsilun)) {
-        return NULL;
-    }
-
-    acb = qemu_aio_get(&iscsi_aiocb_info, bs, cb, opaque);
-    trace_iscsi_aio_writev(iscsilun->iscsi, sector_num, nb_sectors, opaque, acb);
-
-    acb->iscsilun    = iscsilun;
-    acb->qiov        = qiov;
-    acb->nb_sectors  = nb_sectors;
-    acb->sector_num  = sector_num;
-    acb->retries     = ISCSI_CMD_RETRIES;
-
-    if (iscsi_aio_writev_acb(acb) != 0) {
-        qemu_aio_release(acb);
-        return NULL;
-    }
-
-    iscsi_set_events(iscsilun);
-    return &acb->common;
-}
-
-static int
-iscsi_aio_readv_acb(IscsiAIOCB *acb);
-
-static void
-iscsi_aio_read16_cb(struct iscsi_context *iscsi, int status,
-                    void *command_data, void *opaque)
-{
-    IscsiAIOCB *acb = opaque;
-
-    trace_iscsi_aio_read16_cb(iscsi, status, acb, acb->canceled);
-
-    if (acb->canceled != 0) {
-        return;
-    }
-
-    acb->status = 0;
-    if (status != 0) {
-        if (status == SCSI_STATUS_CHECK_CONDITION
-            && acb->task->sense.key == SCSI_SENSE_UNIT_ATTENTION
-            && acb->retries-- > 0) {
-            scsi_free_scsi_task(acb->task);
-            acb->task = NULL;
-            if (iscsi_aio_readv_acb(acb) == 0) {
-                iscsi_set_events(acb->iscsilun);
-                return;
-            }
-        }
-        error_report("Failed to read16 data from iSCSI lun. %s",
-                     iscsi_get_error(iscsi));
-        acb->status = -EIO;
-    }
-
-    iscsi_schedule_bh(acb);
-}
-
-static int
-iscsi_aio_readv_acb(IscsiAIOCB *acb)
-{
-    struct iscsi_context *iscsi = acb->iscsilun->iscsi;
-    size_t size;
+    struct IscsiTask iTask;
     uint64_t lba;
     uint32_t num_sectors;
-    int ret;
 #if !defined(LIBISCSI_FEATURE_IOVECTOR)
     int i;
 #endif
 
-    acb->canceled    = 0;
-    acb->bh          = NULL;
-    acb->status      = -EINPROGRESS;
-    acb->buf         = NULL;
-
-    size = acb->nb_sectors * BDRV_SECTOR_SIZE;
-
-    acb->task = malloc(sizeof(struct scsi_task));
-    if (acb->task == NULL) {
-        error_report("iSCSI: Failed to allocate task for scsi READ16 "
-                     "command. %s", iscsi_get_error(iscsi));
-        return -1;
+    if (!is_request_lun_aligned(sector_num, nb_sectors, iscsilun)) {
+        return -EINVAL;
     }
-    memset(acb->task, 0, sizeof(struct scsi_task));
 
-    acb->task->xfer_dir = SCSI_XFER_READ;
-    acb->task->expxferlen = size;
-    lba = sector_qemu2lun(acb->sector_num, acb->iscsilun);
-    num_sectors = sector_qemu2lun(acb->nb_sectors, acb->iscsilun);
+    lba = sector_qemu2lun(sector_num, iscsilun);
+    num_sectors = sector_qemu2lun(nb_sectors, iscsilun);
 
-    switch (acb->iscsilun->type) {
+    iscsi_co_init_iscsitask(iscsilun, &iTask);
+retry:
+    switch (iscsilun->type) {
     case TYPE_DISK:
-        acb->task->cdb_size = 16;
-        acb->task->cdb[0]  = 0x88;
-        *(uint32_t *)&acb->task->cdb[2]  = htonl(lba >> 32);
-        *(uint32_t *)&acb->task->cdb[6]  = htonl(lba & 0xffffffff);
-        *(uint32_t *)&acb->task->cdb[10] = htonl(num_sectors);
+        iTask.task = iscsi_read16_task(iscsilun->iscsi, iscsilun->lun, lba,
+                                       num_sectors * iscsilun->block_size,
+                                       iscsilun->block_size, 0, 0, 0, 0, 0,
+                                       iscsi_co_generic_cb, &iTask);
         break;
     default:
-        acb->task->cdb_size = 10;
-        acb->task->cdb[0]  = 0x28;
-        *(uint32_t *)&acb->task->cdb[2] = htonl(lba);
-        *(uint16_t *)&acb->task->cdb[7] = htons(num_sectors);
+        iTask.task = iscsi_read10_task(iscsilun->iscsi, iscsilun->lun, lba,
+                                       num_sectors * iscsilun->block_size,
+                                       iscsilun->block_size,
+#if !defined(CONFIG_LIBISCSI_1_4) /* API change from 1.4.0 to 1.5.0 */
+                                       0, 0, 0, 0, 0,
+#endif
+                                       iscsi_co_generic_cb, &iTask);
         break;
     }
-
-    ret = iscsi_scsi_command_async(iscsi, acb->iscsilun->lun, acb->task,
-                                   iscsi_aio_read16_cb,
-                                   NULL,
-                                   acb);
-    if (ret != 0) {
-        scsi_free_scsi_task(acb->task);
-        return -1;
+    if (iTask.task == NULL) {
+        return -EIO;
     }
-
 #if defined(LIBISCSI_FEATURE_IOVECTOR)
-    scsi_task_set_iov_in(acb->task, (struct scsi_iovec*) acb->qiov->iov, acb->qiov->niov);
+    scsi_task_set_iov_in(iTask.task, (struct scsi_iovec *) iov->iov, iov->niov);
 #else
-    for (i = 0; i < acb->qiov->niov; i++) {
-        scsi_task_add_data_in_buffer(acb->task,
-                acb->qiov->iov[i].iov_len,
-                acb->qiov->iov[i].iov_base);
+    for (i = 0; i < iov->niov; i++) {
+        scsi_task_add_data_in_buffer(iTask.task,
+                                     iov->iov[i].iov_len,
+                                     iov->iov[i].iov_base);
     }
 #endif
-    return 0;
-}
 
-static BlockDriverAIOCB *
-iscsi_aio_readv(BlockDriverState *bs, int64_t sector_num,
-                QEMUIOVector *qiov, int nb_sectors,
-                BlockDriverCompletionFunc *cb,
-                void *opaque)
-{
-    IscsiLun *iscsilun = bs->opaque;
-    IscsiAIOCB *acb;
-
-    if (!is_request_lun_aligned(sector_num, nb_sectors, iscsilun)) {
-        return NULL;
+    while (!iTask.complete) {
+        iscsi_set_events(iscsilun);
+        qemu_coroutine_yield();
     }
 
-    acb = qemu_aio_get(&iscsi_aiocb_info, bs, cb, opaque);
-    trace_iscsi_aio_readv(iscsilun->iscsi, sector_num, nb_sectors, opaque, acb);
-
-    acb->nb_sectors  = nb_sectors;
-    acb->sector_num  = sector_num;
-    acb->iscsilun    = iscsilun;
-    acb->qiov        = qiov;
-    acb->retries     = ISCSI_CMD_RETRIES;
-
-    if (iscsi_aio_readv_acb(acb) != 0) {
-        qemu_aio_release(acb);
-        return NULL;
+    if (iTask.task != NULL) {
+        scsi_free_scsi_task(iTask.task);
+        iTask.task = NULL;
     }
 
-    iscsi_set_events(iscsilun);
-    return &acb->common;
-}
-
-static int
-iscsi_aio_flush_acb(IscsiAIOCB *acb);
-
-static void
-iscsi_synccache10_cb(struct iscsi_context *iscsi, int status,
-                     void *command_data, void *opaque)
-{
-    IscsiAIOCB *acb = opaque;
-
-    if (acb->canceled != 0) {
-        return;
+    if (iTask.do_retry) {
+        goto retry;
     }
 
-    acb->status = 0;
-    if (status != 0) {
-        if (status == SCSI_STATUS_CHECK_CONDITION
-            && acb->task->sense.key == SCSI_SENSE_UNIT_ATTENTION
-            && acb->retries-- > 0) {
-            scsi_free_scsi_task(acb->task);
-            acb->task = NULL;
-            if (iscsi_aio_flush_acb(acb) == 0) {
-                iscsi_set_events(acb->iscsilun);
-                return;
-            }
-        }
-        error_report("Failed to sync10 data on iSCSI lun. %s",
-                     iscsi_get_error(iscsi));
-        acb->status = -EIO;
-    }
-
-    iscsi_schedule_bh(acb);
-}
-
-static int
-iscsi_aio_flush_acb(IscsiAIOCB *acb)
-{
-    struct iscsi_context *iscsi = acb->iscsilun->iscsi;
-
-    acb->canceled   = 0;
-    acb->bh         = NULL;
-    acb->status     = -EINPROGRESS;
-    acb->buf        = NULL;
-
-    acb->task = iscsi_synchronizecache10_task(iscsi, acb->iscsilun->lun,
-                                         0, 0, 0, 0,
-                                         iscsi_synccache10_cb,
-                                         acb);
-    if (acb->task == NULL) {
-        error_report("iSCSI: Failed to send synchronizecache10 command. %s",
-                     iscsi_get_error(iscsi));
-        return -1;
+    if (iTask.status != SCSI_STATUS_GOOD) {
+        return -EIO;
     }
 
     return 0;
 }
 
-static BlockDriverAIOCB *
-iscsi_aio_flush(BlockDriverState *bs,
-                BlockDriverCompletionFunc *cb, void *opaque)
+static int coroutine_fn iscsi_co_flush(BlockDriverState *bs)
 {
     IscsiLun *iscsilun = bs->opaque;
+    struct IscsiTask iTask;
 
-    IscsiAIOCB *acb;
+    iscsi_co_init_iscsitask(iscsilun, &iTask);
 
-    acb = qemu_aio_get(&iscsi_aiocb_info, bs, cb, opaque);
-
-    acb->iscsilun    = iscsilun;
-    acb->retries     = ISCSI_CMD_RETRIES;
-
-    if (iscsi_aio_flush_acb(acb) != 0) {
-        qemu_aio_release(acb);
-        return NULL;
+retry:
+    if (iscsi_synchronizecache10_task(iscsilun->iscsi, iscsilun->lun, 0, 0, 0,
+                                      0, iscsi_co_generic_cb, &iTask) == NULL) {
+        return -EIO;
     }
 
-    iscsi_set_events(iscsilun);
+    while (!iTask.complete) {
+        iscsi_set_events(iscsilun);
+        qemu_coroutine_yield();
+    }
 
-    return &acb->common;
+    if (iTask.task != NULL) {
+        scsi_free_scsi_task(iTask.task);
+        iTask.task = NULL;
+    }
+
+    if (iTask.do_retry) {
+        goto retry;
+    }
+
+    if (iTask.status != SCSI_STATUS_GOOD) {
+        return -EIO;
+    }
+
+    return 0;
 }
 
 #ifdef __linux__
@@ -912,8 +736,6 @@ coroutine_fn iscsi_co_discard(BlockDriverState *bs, int64_t sector_num,
     IscsiLun *iscsilun = bs->opaque;
     struct IscsiTask iTask;
     struct unmap_list list;
-    uint32_t nb_blocks;
-    uint32_t max_unmap;
 
     if (!is_request_lun_aligned(sector_num, nb_sectors, iscsilun)) {
         return -EINVAL;
@@ -925,57 +747,115 @@ coroutine_fn iscsi_co_discard(BlockDriverState *bs, int64_t sector_num,
     }
 
     list.lba = sector_qemu2lun(sector_num, iscsilun);
-    nb_blocks = sector_qemu2lun(nb_sectors, iscsilun);
+    list.num = sector_qemu2lun(nb_sectors, iscsilun);
 
-    max_unmap = iscsilun->bl.max_unmap;
-    if (max_unmap == 0xffffffff) {
-        max_unmap = ISCSI_MAX_UNMAP;
+    iscsi_co_init_iscsitask(iscsilun, &iTask);
+retry:
+    if (iscsi_unmap_task(iscsilun->iscsi, iscsilun->lun, 0, 0, &list, 1,
+                     iscsi_co_generic_cb, &iTask) == NULL) {
+        return -EIO;
     }
 
-    while (nb_blocks > 0) {
-        iscsi_co_init_iscsitask(iscsilun, &iTask);
-        list.num = nb_blocks;
-        if (list.num > max_unmap) {
-            list.num = max_unmap;
-        }
-retry:
-        if (iscsi_unmap_task(iscsilun->iscsi, iscsilun->lun, 0, 0, &list, 1,
-                         iscsi_co_generic_cb, &iTask) == NULL) {
-            return -EIO;
-        }
+    while (!iTask.complete) {
+        iscsi_set_events(iscsilun);
+        qemu_coroutine_yield();
+    }
 
-        while (!iTask.complete) {
-            iscsi_set_events(iscsilun);
-            qemu_coroutine_yield();
-        }
+    if (iTask.task != NULL) {
+        scsi_free_scsi_task(iTask.task);
+        iTask.task = NULL;
+    }
 
-        if (iTask.task != NULL) {
-            scsi_free_scsi_task(iTask.task);
-            iTask.task = NULL;
-        }
+    if (iTask.do_retry) {
+        goto retry;
+    }
 
-        if (iTask.do_retry) {
-            goto retry;
-        }
+    if (iTask.status == SCSI_STATUS_CHECK_CONDITION) {
+        /* the target might fail with a check condition if it
+           is not happy with the alignment of the UNMAP request
+           we silently fail in this case */
+        return 0;
+    }
 
-        if (iTask.status == SCSI_STATUS_CHECK_CONDITION) {
-            /* the target might fail with a check condition if it
-               is not happy with the alignment of the UNMAP request
-               we silently fail in this case */
-            return 0;
-        }
-
-        if (iTask.status != SCSI_STATUS_GOOD) {
-            return -EIO;
-        }
-
-        list.lba += list.num;
-        nb_blocks -= list.num;
+    if (iTask.status != SCSI_STATUS_GOOD) {
+        return -EIO;
     }
 
     return 0;
 }
 
+#if defined(SCSI_SENSE_ASCQ_CAPACITY_DATA_HAS_CHANGED)
+
+static int
+coroutine_fn iscsi_co_write_zeroes(BlockDriverState *bs, int64_t sector_num,
+                                   int nb_sectors, BdrvRequestFlags flags)
+{
+    IscsiLun *iscsilun = bs->opaque;
+    struct IscsiTask iTask;
+    uint64_t lba;
+    uint32_t nb_blocks;
+
+    if (!is_request_lun_aligned(sector_num, nb_sectors, iscsilun)) {
+        return -EINVAL;
+    }
+
+    if (!(flags & BDRV_REQ_MAY_UNMAP) && !iscsilun->has_write_same) {
+        /* WRITE SAME without UNMAP is not supported by the target */
+        return -ENOTSUP;
+    }
+
+    if ((flags & BDRV_REQ_MAY_UNMAP) && !iscsilun->lbp.lbpws) {
+        /* WRITE SAME with UNMAP is not supported by the target */
+        return -ENOTSUP;
+    }
+
+    lba = sector_qemu2lun(sector_num, iscsilun);
+    nb_blocks = sector_qemu2lun(nb_sectors, iscsilun);
+
+    if (iscsilun->zeroblock == NULL) {
+        iscsilun->zeroblock = g_malloc0(iscsilun->block_size);
+    }
+
+    iscsi_co_init_iscsitask(iscsilun, &iTask);
+retry:
+    if (iscsi_writesame16_task(iscsilun->iscsi, iscsilun->lun, lba,
+                               iscsilun->zeroblock, iscsilun->block_size,
+                               nb_blocks, 0, !!(flags & BDRV_REQ_MAY_UNMAP),
+                               0, 0, iscsi_co_generic_cb, &iTask) == NULL) {
+        return -EIO;
+    }
+
+    while (!iTask.complete) {
+        iscsi_set_events(iscsilun);
+        qemu_coroutine_yield();
+    }
+
+    if (iTask.task != NULL) {
+        scsi_free_scsi_task(iTask.task);
+        iTask.task = NULL;
+    }
+
+    if (iTask.do_retry) {
+        goto retry;
+    }
+
+    if (iTask.status != SCSI_STATUS_GOOD) {
+        if (iTask.status == SCSI_STATUS_CHECK_CONDITION &&
+            iTask.task->sense.key == SCSI_SENSE_ILLEGAL_REQUEST &&
+            iTask.task->sense.ascq == SCSI_SENSE_ASCQ_INVALID_OPERATION_CODE) {
+            /* WRITE SAME is not supported by the target */
+            iscsilun->has_write_same = false;
+            return -ENOTSUP;
+        }
+
+        return -EIO;
+    }
+
+    return 0;
+}
+
+#endif /* SCSI_SENSE_ASCQ_CAPACITY_DATA_HAS_CHANGED */
+
 static int parse_chap(struct iscsi_context *iscsi, const char *target)
 {
     QemuOptsList *list;
@@ -1241,7 +1121,7 @@ static int iscsi_open(BlockDriverState *bs, QDict *options, int flags,
         return -EINVAL;
     }
 
-    opts = qemu_opts_create_nofail(&runtime_opts);
+    opts = qemu_opts_create(&runtime_opts, NULL, 0, &error_abort);
     qemu_opts_absorb_qdict(opts, options, &local_err);
     if (error_is_set(&local_err)) {
         qerror_report_err(local_err);
@@ -1331,6 +1211,7 @@ static int iscsi_open(BlockDriverState *bs, QDict *options, int flags,
     }
 
     iscsilun->type = inq->periperal_device_type;
+    iscsilun->has_write_same = true;
 
     if ((ret = iscsi_readcapacity_sync(iscsilun)) != 0) {
         goto out;
@@ -1384,6 +1265,23 @@ static int iscsi_open(BlockDriverState *bs, QDict *options, int flags,
                sizeof(struct scsi_inquiry_block_limits));
         scsi_free_scsi_task(task);
         task = NULL;
+
+        if (iscsilun->bl.max_unmap < 0xffffffff) {
+            bs->bl.max_discard = sector_lun2qemu(iscsilun->bl.max_unmap,
+                                                 iscsilun);
+        }
+        bs->bl.discard_alignment = sector_lun2qemu(iscsilun->bl.opt_unmap_gran,
+                                                   iscsilun);
+
+        if (iscsilun->bl.max_ws_len < 0xffffffff) {
+            bs->bl.max_write_zeroes = sector_lun2qemu(iscsilun->bl.max_ws_len,
+                                                      iscsilun);
+        }
+        bs->bl.write_zeroes_alignment = sector_lun2qemu(iscsilun->bl.opt_unmap_gran,
+                                                        iscsilun);
+
+        bs->bl.opt_transfer_length = sector_lun2qemu(iscsilun->bl.opt_xfer_len,
+                                                     iscsilun);
     }
 
 #if defined(LIBISCSI_FEATURE_NOP_COUNTER)
@@ -1424,6 +1322,7 @@ static void iscsi_close(BlockDriverState *bs)
     }
     qemu_aio_set_fd_handler(iscsi_get_fd(iscsi), NULL, NULL, NULL);
     iscsi_destroy_context(iscsi);
+    g_free(iscsilun->zeroblock);
     memset(iscsilun, 0, sizeof(IscsiLun));
 }
 
@@ -1447,11 +1346,6 @@ static int iscsi_truncate(BlockDriverState *bs, int64_t offset)
     return 0;
 }
 
-static int iscsi_has_zero_init(BlockDriverState *bs)
-{
-    return 0;
-}
-
 static int iscsi_create(const char *filename, QEMUOptionParameter *options,
                         Error **errp)
 {
@@ -1506,6 +1400,21 @@ out:
     return ret;
 }
 
+static int iscsi_get_info(BlockDriverState *bs, BlockDriverInfo *bdi)
+{
+    IscsiLun *iscsilun = bs->opaque;
+    bdi->unallocated_blocks_are_zero = !!iscsilun->lbprz;
+    bdi->can_write_zeroes_with_unmap = iscsilun->lbprz && iscsilun->lbp.lbpws;
+    /* Guess the internal cluster (page) size of the iscsi target by the means
+     * of opt_unmap_gran. Transfer the unmap granularity only if it has a
+     * reasonable size for bdi->cluster_size */
+    if (iscsilun->bl.opt_unmap_gran * iscsilun->block_size >= 64 * 1024 &&
+        iscsilun->bl.opt_unmap_gran * iscsilun->block_size <= 16 * 1024 * 1024) {
+        bdi->cluster_size = iscsilun->bl.opt_unmap_gran * iscsilun->block_size;
+    }
+    return 0;
+}
+
 static QEMUOptionParameter iscsi_create_options[] = {
     {
         .name = BLOCK_OPT_SIZE,
@@ -1527,18 +1436,19 @@ static BlockDriver bdrv_iscsi = {
     .create_options  = iscsi_create_options,
 
     .bdrv_getlength  = iscsi_getlength,
+    .bdrv_get_info   = iscsi_get_info,
     .bdrv_truncate   = iscsi_truncate,
 
 #if defined(LIBISCSI_FEATURE_IOVECTOR)
     .bdrv_co_get_block_status = iscsi_co_get_block_status,
 #endif
     .bdrv_co_discard      = iscsi_co_discard,
-
-    .bdrv_aio_readv  = iscsi_aio_readv,
-    .bdrv_aio_writev = iscsi_aio_writev,
-    .bdrv_aio_flush  = iscsi_aio_flush,
-
-    .bdrv_has_zero_init = iscsi_has_zero_init,
+#if defined(SCSI_SENSE_ASCQ_CAPACITY_DATA_HAS_CHANGED)
+    .bdrv_co_write_zeroes = iscsi_co_write_zeroes,
+#endif
+    .bdrv_co_readv         = iscsi_co_readv,
+    .bdrv_co_writev        = iscsi_co_writev,
+    .bdrv_co_flush_to_disk = iscsi_co_flush,
 
 #ifdef __linux__
     .bdrv_ioctl       = iscsi_ioctl,

block/mirror.c

@@ -31,7 +31,8 @@ typedef struct MirrorBlockJob {
     BlockJob common;
     RateLimit limit;
     BlockDriverState *target;
-    MirrorSyncMode mode;
+    BlockDriverState *base;
+    bool is_none_mode;
     BlockdevOnError on_source_error, on_target_error;
     bool synced;
     bool should_complete;
@@ -39,6 +40,7 @@ typedef struct MirrorBlockJob {
     int64_t granularity;
     size_t buf_size;
     unsigned long *cow_bitmap;
+    BdrvDirtyBitmap *dirty_bitmap;
     HBitmapIter hbi;
     uint8_t *buf;
     QSIMPLEQ_HEAD(, MirrorBuffer) buf_free;
@@ -145,9 +147,10 @@ static void coroutine_fn mirror_iteration(MirrorBlockJob *s)
 
     s->sector_num = hbitmap_iter_next(&s->hbi);
     if (s->sector_num < 0) {
-        bdrv_dirty_iter_init(source, &s->hbi);
+        bdrv_dirty_iter_init(source, s->dirty_bitmap, &s->hbi);
         s->sector_num = hbitmap_iter_next(&s->hbi);
-        trace_mirror_restart_iter(s, bdrv_get_dirty_count(source));
+        trace_mirror_restart_iter(s,
+                                  bdrv_get_dirty_count(source, s->dirty_bitmap));
         assert(s->sector_num >= 0);
     }
 
@@ -183,7 +186,7 @@ static void coroutine_fn mirror_iteration(MirrorBlockJob *s)
     do {
         int added_sectors, added_chunks;
 
-        if (!bdrv_get_dirty(source, next_sector) ||
+        if (!bdrv_get_dirty(source, s->dirty_bitmap, next_sector) ||
             test_bit(next_chunk, s->in_flight_bitmap)) {
             assert(nb_sectors > 0);
             break;
@@ -249,7 +252,8 @@ static void coroutine_fn mirror_iteration(MirrorBlockJob *s)
         /* Advance the HBitmapIter in parallel, so that we do not examine
          * the same sector twice.
          */
-        if (next_sector > hbitmap_next_sector && bdrv_get_dirty(source, next_sector)) {
+        if (next_sector > hbitmap_next_sector
+            && bdrv_get_dirty(source, s->dirty_bitmap, next_sector)) {
             hbitmap_next_sector = hbitmap_iter_next(&s->hbi);
         }
 
@@ -332,10 +336,9 @@ static void coroutine_fn mirror_run(void *opaque)
     sectors_per_chunk = s->granularity >> BDRV_SECTOR_BITS;
     mirror_free_init(s);
 
-    if (s->mode != MIRROR_SYNC_MODE_NONE) {
+    if (!s->is_none_mode) {
         /* First part, loop on the sectors and initialize the dirty bitmap.  */
-        BlockDriverState *base;
-        base = s->mode == MIRROR_SYNC_MODE_FULL ? NULL : bs->backing_hd;
+        BlockDriverState *base = s->base;
         for (sector_num = 0; sector_num < end; ) {
             int64_t next = (sector_num | (sectors_per_chunk - 1)) + 1;
             ret = bdrv_is_allocated_above(bs, base,
@@ -355,7 +358,7 @@ static void coroutine_fn mirror_run(void *opaque)
         }
     }
 
-    bdrv_dirty_iter_init(bs, &s->hbi);
+    bdrv_dirty_iter_init(bs, s->dirty_bitmap, &s->hbi);
     last_pause_ns = qemu_clock_get_ns(QEMU_CLOCK_REALTIME);
     for (;;) {
         uint64_t delay_ns;
@@ -367,7 +370,7 @@ static void coroutine_fn mirror_run(void *opaque)
             goto immediate_exit;
         }
 
-        cnt = bdrv_get_dirty_count(bs);
+        cnt = bdrv_get_dirty_count(bs, s->dirty_bitmap);
 
         /* Note that even when no rate limit is applied we need to yield
          * periodically with no pending I/O so that qemu_aio_flush() returns.
@@ -409,7 +412,7 @@ static void coroutine_fn mirror_run(void *opaque)
 
                 should_complete = s->should_complete ||
                     block_job_is_cancelled(&s->common);
-                cnt = bdrv_get_dirty_count(bs);
+                cnt = bdrv_get_dirty_count(bs, s->dirty_bitmap);
             }
         }
 
@@ -424,7 +427,7 @@ static void coroutine_fn mirror_run(void *opaque)
              */
             trace_mirror_before_drain(s, cnt);
             bdrv_drain_all();
-            cnt = bdrv_get_dirty_count(bs);
+            cnt = bdrv_get_dirty_count(bs, s->dirty_bitmap);
         }
 
         ret = 0;
@@ -471,15 +474,21 @@ immediate_exit:
     qemu_vfree(s->buf);
     g_free(s->cow_bitmap);
     g_free(s->in_flight_bitmap);
-    bdrv_set_dirty_tracking(bs, 0);
+    bdrv_release_dirty_bitmap(bs, s->dirty_bitmap);
     bdrv_iostatus_disable(s->target);
     if (s->should_complete && ret == 0) {
         if (bdrv_get_flags(s->target) != bdrv_get_flags(s->common.bs)) {
             bdrv_reopen(s->target, bdrv_get_flags(s->common.bs), NULL);
         }
         bdrv_swap(s->target, s->common.bs);
+        if (s->common.driver->job_type == BLOCK_JOB_TYPE_COMMIT) {
+            /* drop the bs loop chain formed by the swap: break the loop then
+             * trigger the unref from the top one */
+            BlockDriverState *p = s->base->backing_hd;
+            s->base->backing_hd = NULL;
+            bdrv_unref(p);
+        }
     }
-    bdrv_close(s->target);
     bdrv_unref(s->target);
     block_job_completed(&s->common, ret);
 }
@@ -533,12 +542,24 @@ static const BlockJobDriver mirror_job_driver = {
     .complete      = mirror_complete,
 };
 
-void mirror_start(BlockDriverState *bs, BlockDriverState *target,
-                  int64_t speed, int64_t granularity, int64_t buf_size,
-                  MirrorSyncMode mode, BlockdevOnError on_source_error,
-                  BlockdevOnError on_target_error,
-                  BlockDriverCompletionFunc *cb,
-                  void *opaque, Error **errp)
+static const BlockJobDriver commit_active_job_driver = {
+    .instance_size = sizeof(MirrorBlockJob),
+    .job_type      = BLOCK_JOB_TYPE_COMMIT,
+    .set_speed     = mirror_set_speed,
+    .iostatus_reset
+                   = mirror_iostatus_reset,
+    .complete      = mirror_complete,
+};
+
+static void mirror_start_job(BlockDriverState *bs, BlockDriverState *target,
+                            int64_t speed, int64_t granularity,
+                            int64_t buf_size,
+                            BlockdevOnError on_source_error,
+                            BlockdevOnError on_target_error,
+                            BlockDriverCompletionFunc *cb,
+                            void *opaque, Error **errp,
+                            const BlockJobDriver *driver,
+                            bool is_none_mode, BlockDriverState *base)
 {
     MirrorBlockJob *s;
 
@@ -563,7 +584,8 @@ void mirror_start(BlockDriverState *bs, BlockDriverState *target,
         return;
     }
 
-    s = block_job_create(&mirror_job_driver, bs, speed, cb, opaque, errp);
+
+    s = block_job_create(driver, bs, speed, cb, opaque, errp);
     if (!s) {
         return;
     }
@@ -571,11 +593,12 @@ void mirror_start(BlockDriverState *bs, BlockDriverState *target,
     s->on_source_error = on_source_error;
     s->on_target_error = on_target_error;
     s->target = target;
-    s->mode = mode;
+    s->is_none_mode = is_none_mode;
+    s->base = base;
     s->granularity = granularity;
     s->buf_size = MAX(buf_size, granularity);
 
-    bdrv_set_dirty_tracking(bs, granularity);
+    s->dirty_bitmap = bdrv_create_dirty_bitmap(bs, granularity);
     bdrv_set_enable_write_cache(s->target, true);
     bdrv_set_on_error(s->target, on_target_error, on_target_error);
     bdrv_iostatus_enable(s->target);
@@ -583,3 +606,35 @@ void mirror_start(BlockDriverState *bs, BlockDriverState *target,
     trace_mirror_start(bs, s, s->common.co, opaque);
     qemu_coroutine_enter(s->common.co, s);
 }
+
+void mirror_start(BlockDriverState *bs, BlockDriverState *target,
+                  int64_t speed, int64_t granularity, int64_t buf_size,
+                  MirrorSyncMode mode, BlockdevOnError on_source_error,
+                  BlockdevOnError on_target_error,
+                  BlockDriverCompletionFunc *cb,
+                  void *opaque, Error **errp)
+{
+    bool is_none_mode;
+    BlockDriverState *base;
+
+    is_none_mode = mode == MIRROR_SYNC_MODE_NONE;
+    base = mode == MIRROR_SYNC_MODE_TOP ? bs->backing_hd : NULL;
+    mirror_start_job(bs, target, speed, granularity, buf_size,
+                     on_source_error, on_target_error, cb, opaque, errp,
+                     &mirror_job_driver, is_none_mode, base);
+}
+
+void commit_active_start(BlockDriverState *bs, BlockDriverState *base,
+                         int64_t speed,
+                         BlockdevOnError on_error,
+                         BlockDriverCompletionFunc *cb,
+                         void *opaque, Error **errp)
+{
+    if (bdrv_reopen(base, bs->open_flags, errp)) {
+        return;
+    }
+    bdrv_ref(base);
+    mirror_start_job(bs, base, speed, 0, 0,
+                     on_error, on_error, cb, opaque, errp,
+                     &commit_active_job_driver, false, base);
+}

block/nbd-client.c

@@ -0,0 +1,385 @@
+/*
+ * QEMU Block driver for  NBD
+ *
+ * Copyright (C) 2008 Bull S.A.S.
+ *     Author: Laurent Vivier <Laurent.Vivier@bull.net>
+ *
+ * Some parts:
+ *    Copyright (C) 2007 Anthony Liguori <anthony@codemonkey.ws>
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a copy
+ * of this software and associated documentation files (the "Software"), to deal
+ * in the Software without restriction, including without limitation the rights
+ * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+ * copies of the Software, and to permit persons to whom the Software is
+ * furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
+ * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+ * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
+ * THE SOFTWARE.
+ */
+
+#include "nbd-client.h"
+#include "qemu/sockets.h"
+
+#define HANDLE_TO_INDEX(bs, handle) ((handle) ^ ((uint64_t)(intptr_t)bs))
+#define INDEX_TO_HANDLE(bs, index)  ((index)  ^ ((uint64_t)(intptr_t)bs))
+
+static void nbd_recv_coroutines_enter_all(NbdClientSession *s)
+{
+    int i;
+
+    for (i = 0; i < MAX_NBD_REQUESTS; i++) {
+        if (s->recv_coroutine[i]) {
+            qemu_coroutine_enter(s->recv_coroutine[i], NULL);
+        }
+    }
+}
+
+static void nbd_reply_ready(void *opaque)
+{
+    NbdClientSession *s = opaque;
+    uint64_t i;
+    int ret;
+
+    if (s->reply.handle == 0) {
+        /* No reply already in flight.  Fetch a header.  It is possible
+         * that another thread has done the same thing in parallel, so
+         * the socket is not readable anymore.
+         */
+        ret = nbd_receive_reply(s->sock, &s->reply);
+        if (ret == -EAGAIN) {
+            return;
+        }
+        if (ret < 0) {
+            s->reply.handle = 0;
+            goto fail;
+        }
+    }
+
+    /* There's no need for a mutex on the receive side, because the
+     * handler acts as a synchronization point and ensures that only
+     * one coroutine is called until the reply finishes.  */
+    i = HANDLE_TO_INDEX(s, s->reply.handle);
+    if (i >= MAX_NBD_REQUESTS) {
+        goto fail;
+    }
+
+    if (s->recv_coroutine[i]) {
+        qemu_coroutine_enter(s->recv_coroutine[i], NULL);
+        return;
+    }
+
+fail:
+    nbd_recv_coroutines_enter_all(s);
+}
+
+static void nbd_restart_write(void *opaque)
+{
+    NbdClientSession *s = opaque;
+
+    qemu_coroutine_enter(s->send_coroutine, NULL);
+}
+
+static int nbd_co_send_request(NbdClientSession *s,
+    struct nbd_request *request,
+    QEMUIOVector *qiov, int offset)
+{
+    int rc, ret;
+
+    qemu_co_mutex_lock(&s->send_mutex);
+    s->send_coroutine = qemu_coroutine_self();
+    qemu_aio_set_fd_handler(s->sock, nbd_reply_ready, nbd_restart_write, s);
+    if (qiov) {
+        if (!s->is_unix) {
+            socket_set_cork(s->sock, 1);
+        }
+        rc = nbd_send_request(s->sock, request);
+        if (rc >= 0) {
+            ret = qemu_co_sendv(s->sock, qiov->iov, qiov->niov,
+                                offset, request->len);
+            if (ret != request->len) {
+                rc = -EIO;
+            }
+        }
+        if (!s->is_unix) {
+            socket_set_cork(s->sock, 0);
+        }
+    } else {
+        rc = nbd_send_request(s->sock, request);
+    }
+    qemu_aio_set_fd_handler(s->sock, nbd_reply_ready, NULL, s);
+    s->send_coroutine = NULL;
+    qemu_co_mutex_unlock(&s->send_mutex);
+    return rc;
+}
+
+static void nbd_co_receive_reply(NbdClientSession *s,
+    struct nbd_request *request, struct nbd_reply *reply,
+    QEMUIOVector *qiov, int offset)
+{
+    int ret;
+
+    /* Wait until we're woken up by the read handler.  TODO: perhaps
+     * peek at the next reply and avoid yielding if it's ours?  */
+    qemu_coroutine_yield();
+    *reply = s->reply;
+    if (reply->handle != request->handle) {
+        reply->error = EIO;
+    } else {
+        if (qiov && reply->error == 0) {
+            ret = qemu_co_recvv(s->sock, qiov->iov, qiov->niov,
+                                offset, request->len);
+            if (ret != request->len) {
+                reply->error = EIO;
+            }
+        }
+
+        /* Tell the read handler to read another header.  */
+        s->reply.handle = 0;
+    }
+}
+
+static void nbd_coroutine_start(NbdClientSession *s,
+   struct nbd_request *request)
+{
+    int i;
+
+    /* Poor man semaphore.  The free_sema is locked when no other request
+     * can be accepted, and unlocked after receiving one reply.  */
+    if (s->in_flight >= MAX_NBD_REQUESTS - 1) {
+        qemu_co_mutex_lock(&s->free_sema);
+        assert(s->in_flight < MAX_NBD_REQUESTS);
+    }
+    s->in_flight++;
+
+    for (i = 0; i < MAX_NBD_REQUESTS; i++) {
+        if (s->recv_coroutine[i] == NULL) {
+            s->recv_coroutine[i] = qemu_coroutine_self();
+            break;
+        }
+    }
+
+    assert(i < MAX_NBD_REQUESTS);
+    request->handle = INDEX_TO_HANDLE(s, i);
+}
+
+static void nbd_coroutine_end(NbdClientSession *s,
+    struct nbd_request *request)
+{
+    int i = HANDLE_TO_INDEX(s, request->handle);
+    s->recv_coroutine[i] = NULL;
+    if (s->in_flight-- == MAX_NBD_REQUESTS) {
+        qemu_co_mutex_unlock(&s->free_sema);
+    }
+}
+
+static int nbd_co_readv_1(NbdClientSession *client, int64_t sector_num,
+                          int nb_sectors, QEMUIOVector *qiov,
+                          int offset)
+{
+    struct nbd_request request = { .type = NBD_CMD_READ };
+    struct nbd_reply reply;
+    ssize_t ret;
+
+    request.from = sector_num * 512;
+    request.len = nb_sectors * 512;
+
+    nbd_coroutine_start(client, &request);
+    ret = nbd_co_send_request(client, &request, NULL, 0);
+    if (ret < 0) {
+        reply.error = -ret;
+    } else {
+        nbd_co_receive_reply(client, &request, &reply, qiov, offset);
+    }
+    nbd_coroutine_end(client, &request);
+    return -reply.error;
+
+}
+
+static int nbd_co_writev_1(NbdClientSession *client, int64_t sector_num,
+                           int nb_sectors, QEMUIOVector *qiov,
+                           int offset)
+{
+    struct nbd_request request = { .type = NBD_CMD_WRITE };
+    struct nbd_reply reply;
+    ssize_t ret;
+
+    if (!bdrv_enable_write_cache(client->bs) &&
+        (client->nbdflags & NBD_FLAG_SEND_FUA)) {
+        request.type |= NBD_CMD_FLAG_FUA;
+    }
+
+    request.from = sector_num * 512;
+    request.len = nb_sectors * 512;
+
+    nbd_coroutine_start(client, &request);
+    ret = nbd_co_send_request(client, &request, qiov, offset);
+    if (ret < 0) {
+        reply.error = -ret;
+    } else {
+        nbd_co_receive_reply(client, &request, &reply, NULL, 0);
+    }
+    nbd_coroutine_end(client, &request);
+    return -reply.error;
+}
+
+/* qemu-nbd has a limit of slightly less than 1M per request.  Try to
+ * remain aligned to 4K. */
+#define NBD_MAX_SECTORS 2040
+
+int nbd_client_session_co_readv(NbdClientSession *client, int64_t sector_num,
+    int nb_sectors, QEMUIOVector *qiov)
+{
+    int offset = 0;
+    int ret;
+    while (nb_sectors > NBD_MAX_SECTORS) {
+        ret = nbd_co_readv_1(client, sector_num,
+                             NBD_MAX_SECTORS, qiov, offset);
+        if (ret < 0) {
+            return ret;
+        }
+        offset += NBD_MAX_SECTORS * 512;
+        sector_num += NBD_MAX_SECTORS;
+        nb_sectors -= NBD_MAX_SECTORS;
+    }
+    return nbd_co_readv_1(client, sector_num, nb_sectors, qiov, offset);
+}
+
+int nbd_client_session_co_writev(NbdClientSession *client, int64_t sector_num,
+                                 int nb_sectors, QEMUIOVector *qiov)
+{
+    int offset = 0;
+    int ret;
+    while (nb_sectors > NBD_MAX_SECTORS) {
+        ret = nbd_co_writev_1(client, sector_num,
+                              NBD_MAX_SECTORS, qiov, offset);
+        if (ret < 0) {
+            return ret;
+        }
+        offset += NBD_MAX_SECTORS * 512;
+        sector_num += NBD_MAX_SECTORS;
+        nb_sectors -= NBD_MAX_SECTORS;
+    }
+    return nbd_co_writev_1(client, sector_num, nb_sectors, qiov, offset);
+}
+
+int nbd_client_session_co_flush(NbdClientSession *client)
+{
+    struct nbd_request request = { .type = NBD_CMD_FLUSH };
+    struct nbd_reply reply;
+    ssize_t ret;
+
+    if (!(client->nbdflags & NBD_FLAG_SEND_FLUSH)) {
+        return 0;
+    }
+
+    if (client->nbdflags & NBD_FLAG_SEND_FUA) {
+        request.type |= NBD_CMD_FLAG_FUA;
+    }
+
+    request.from = 0;
+    request.len = 0;
+
+    nbd_coroutine_start(client, &request);
+    ret = nbd_co_send_request(client, &request, NULL, 0);
+    if (ret < 0) {
+        reply.error = -ret;
+    } else {
+        nbd_co_receive_reply(client, &request, &reply, NULL, 0);
+    }
+    nbd_coroutine_end(client, &request);
+    return -reply.error;
+}
+
+int nbd_client_session_co_discard(NbdClientSession *client, int64_t sector_num,
+    int nb_sectors)
+{
+    struct nbd_request request = { .type = NBD_CMD_TRIM };
+    struct nbd_reply reply;
+    ssize_t ret;
+
+    if (!(client->nbdflags & NBD_FLAG_SEND_TRIM)) {
+        return 0;
+    }
+    request.from = sector_num * 512;
+    request.len = nb_sectors * 512;
+
+    nbd_coroutine_start(client, &request);
+    ret = nbd_co_send_request(client, &request, NULL, 0);
+    if (ret < 0) {
+        reply.error = -ret;
+    } else {
+        nbd_co_receive_reply(client, &request, &reply, NULL, 0);
+    }
+    nbd_coroutine_end(client, &request);
+    return -reply.error;
+
+}
+
+static void nbd_teardown_connection(NbdClientSession *client)
+{
+    struct nbd_request request = {
+        .type = NBD_CMD_DISC,
+        .from = 0,
+        .len = 0
+    };
+
+    nbd_send_request(client->sock, &request);
+
+    /* finish any pending coroutines */
+    shutdown(client->sock, 2);
+    nbd_recv_coroutines_enter_all(client);
+
+    qemu_aio_set_fd_handler(client->sock, NULL, NULL, NULL);
+    closesocket(client->sock);
+    client->sock = -1;
+}
+
+void nbd_client_session_close(NbdClientSession *client)
+{
+    if (!client->bs) {
+        return;
+    }
+
+    nbd_teardown_connection(client);
+    client->bs = NULL;
+}
+
+int nbd_client_session_init(NbdClientSession *client, BlockDriverState *bs,
+    int sock, const char *export)
+{
+    int ret;
+
+    /* NBD handshake */
+    logout("session init %s\n", export);
+    qemu_set_block(sock);
+    ret = nbd_receive_negotiate(sock, export,
+                                &client->nbdflags, &client->size,
+                                &client->blocksize);
+    if (ret < 0) {
+        logout("Failed to negotiate with the NBD server\n");
+        closesocket(sock);
+        return ret;
+    }
+
+    qemu_co_mutex_init(&client->send_mutex);
+    qemu_co_mutex_init(&client->free_sema);
+    client->bs = bs;
+    client->sock = sock;
+
+    /* Now that we're connected, set the socket to be non-blocking and
+     * kick the reply mechanism.  */
+    qemu_set_nonblock(sock);
+    qemu_aio_set_fd_handler(sock, nbd_reply_ready, NULL, client);
+
+    logout("Established connection with NBD server\n");
+    return 0;
+}

block/nbd-client.h

@@ -0,0 +1,50 @@
+#ifndef NBD_CLIENT_H
+#define NBD_CLIENT_H
+
+#include "qemu-common.h"
+#include "block/nbd.h"
+#include "block/block_int.h"
+
+/* #define DEBUG_NBD */
+
+#if defined(DEBUG_NBD)
+#define logout(fmt, ...) \
+    fprintf(stderr, "nbd\t%-24s" fmt, __func__, ##__VA_ARGS__)
+#else
+#define logout(fmt, ...) ((void)0)
+#endif
+
+#define MAX_NBD_REQUESTS    16
+
+typedef struct NbdClientSession {
+    int sock;
+    uint32_t nbdflags;
+    off_t size;
+    size_t blocksize;
+
+    CoMutex send_mutex;
+    CoMutex free_sema;
+    Coroutine *send_coroutine;
+    int in_flight;
+
+    Coroutine *recv_coroutine[MAX_NBD_REQUESTS];
+    struct nbd_reply reply;
+
+    bool is_unix;
+
+    BlockDriverState *bs;
+} NbdClientSession;
+
+int nbd_client_session_init(NbdClientSession *client, BlockDriverState *bs,
+                            int sock, const char *export_name);
+void nbd_client_session_close(NbdClientSession *client);
+
+int nbd_client_session_co_discard(NbdClientSession *client, int64_t sector_num,
+                                  int nb_sectors);
+int nbd_client_session_co_flush(NbdClientSession *client);
+int nbd_client_session_co_writev(NbdClientSession *client, int64_t sector_num,
+                                 int nb_sectors, QEMUIOVector *qiov);
+int nbd_client_session_co_readv(NbdClientSession *client, int64_t sector_num,
+                                int nb_sectors, QEMUIOVector *qiov);
+
+#endif /* NBD_CLIENT_H */

block/nbd.c

@@ -26,8 +26,7 @@
  * THE SOFTWARE.
  */
 
-#include "qemu-common.h"
-#include "block/nbd.h"
+#include "block/nbd-client.h"
 #include "qemu/uri.h"
 #include "block/block_int.h"
 #include "qemu/module.h"
@@ -40,37 +39,9 @@
 
 #define EN_OPTSTR ":exportname="
 
-/* #define DEBUG_NBD */
-
-#if defined(DEBUG_NBD)
-#define logout(fmt, ...) \
-                fprintf(stderr, "nbd\t%-24s" fmt, __func__, ##__VA_ARGS__)
-#else
-#define logout(fmt, ...) ((void)0)
-#endif
-
-#define MAX_NBD_REQUESTS	16
-#define HANDLE_TO_INDEX(bs, handle) ((handle) ^ ((uint64_t)(intptr_t)bs))
-#define INDEX_TO_HANDLE(bs, index)  ((index)  ^ ((uint64_t)(intptr_t)bs))
-
 typedef struct BDRVNBDState {
-    int sock;
-    uint32_t nbdflags;
-    off_t size;
-    size_t blocksize;
-
-    CoMutex send_mutex;
-    CoMutex free_sema;
-    Coroutine *send_coroutine;
-    int in_flight;
-
-    Coroutine *recv_coroutine[MAX_NBD_REQUESTS];
-    struct nbd_reply reply;
-
-    bool is_unix;
+    NbdClientSession client;
     QemuOpts *socket_opts;
-
-    char *export_name; /* An NBD server may export several devices */
 } BDRVNBDState;
 
 static int nbd_parse_uri(const char *filename, QDict *options)
@@ -217,7 +188,7 @@ out:
     g_free(file);
 }
 
-static int nbd_config(BDRVNBDState *s, QDict *options)
+static int nbd_config(BDRVNBDState *s, QDict *options, char **export)
 {
     Error *local_err = NULL;
 
@@ -227,14 +198,15 @@ static int nbd_config(BDRVNBDState *s, QDict *options)
                           "be used at the same time.");
             return -EINVAL;
         }
-        s->is_unix = true;
+        s->client.is_unix = true;
     } else if (qdict_haskey(options, "host")) {
-        s->is_unix = false;
+        s->client.is_unix = false;
     } else {
         return -EINVAL;
     }
 
-    s->socket_opts = qemu_opts_create_nofail(&socket_optslist);
+    s->socket_opts = qemu_opts_create(&socket_optslist, NULL, 0,
+                                      &error_abort);
 
     qemu_opts_absorb_qdict(s->socket_opts, options, &local_err);
     if (error_is_set(&local_err)) {
@@ -247,162 +219,20 @@ static int nbd_config(BDRVNBDState *s, QDict *options)
         qemu_opt_set_number(s->socket_opts, "port", NBD_DEFAULT_PORT);
     }
 
-    s->export_name = g_strdup(qdict_get_try_str(options, "export"));
-    if (s->export_name) {
+    *export = g_strdup(qdict_get_try_str(options, "export"));
+    if (*export) {
         qdict_del(options, "export");
     }
 
     return 0;
 }
 
-
-static void nbd_coroutine_start(BDRVNBDState *s, struct nbd_request *request)
-{
-    int i;
-
-    /* Poor man semaphore.  The free_sema is locked when no other request
-     * can be accepted, and unlocked after receiving one reply.  */
-    if (s->in_flight >= MAX_NBD_REQUESTS - 1) {
-        qemu_co_mutex_lock(&s->free_sema);
-        assert(s->in_flight < MAX_NBD_REQUESTS);
-    }
-    s->in_flight++;
-
-    for (i = 0; i < MAX_NBD_REQUESTS; i++) {
-        if (s->recv_coroutine[i] == NULL) {
-            s->recv_coroutine[i] = qemu_coroutine_self();
-            break;
-        }
-    }
-
-    assert(i < MAX_NBD_REQUESTS);
-    request->handle = INDEX_TO_HANDLE(s, i);
-}
-
-static void nbd_reply_ready(void *opaque)
-{
-    BDRVNBDState *s = opaque;
-    uint64_t i;
-    int ret;
-
-    if (s->reply.handle == 0) {
-        /* No reply already in flight.  Fetch a header.  It is possible
-         * that another thread has done the same thing in parallel, so
-         * the socket is not readable anymore.
-         */
-        ret = nbd_receive_reply(s->sock, &s->reply);
-        if (ret == -EAGAIN) {
-            return;
-        }
-        if (ret < 0) {
-            s->reply.handle = 0;
-            goto fail;
-        }
-    }
-
-    /* There's no need for a mutex on the receive side, because the
-     * handler acts as a synchronization point and ensures that only
-     * one coroutine is called until the reply finishes.  */
-    i = HANDLE_TO_INDEX(s, s->reply.handle);
-    if (i >= MAX_NBD_REQUESTS) {
-        goto fail;
-    }
-
-    if (s->recv_coroutine[i]) {
-        qemu_coroutine_enter(s->recv_coroutine[i], NULL);
-        return;
-    }
-
-fail:
-    for (i = 0; i < MAX_NBD_REQUESTS; i++) {
-        if (s->recv_coroutine[i]) {
-            qemu_coroutine_enter(s->recv_coroutine[i], NULL);
-        }
-    }
-}
-
-static void nbd_restart_write(void *opaque)
-{
-    BDRVNBDState *s = opaque;
-    qemu_coroutine_enter(s->send_coroutine, NULL);
-}
-
-static int nbd_co_send_request(BDRVNBDState *s, struct nbd_request *request,
-                               QEMUIOVector *qiov, int offset)
-{
-    int rc, ret;
-
-    qemu_co_mutex_lock(&s->send_mutex);
-    s->send_coroutine = qemu_coroutine_self();
-    qemu_aio_set_fd_handler(s->sock, nbd_reply_ready, nbd_restart_write, s);
-    if (qiov) {
-        if (!s->is_unix) {
-            socket_set_cork(s->sock, 1);
-        }
-        rc = nbd_send_request(s->sock, request);
-        if (rc >= 0) {
-            ret = qemu_co_sendv(s->sock, qiov->iov, qiov->niov,
-                                offset, request->len);
-            if (ret != request->len) {
-                rc = -EIO;
-            }
-        }
-        if (!s->is_unix) {
-            socket_set_cork(s->sock, 0);
-        }
-    } else {
-        rc = nbd_send_request(s->sock, request);
-    }
-    qemu_aio_set_fd_handler(s->sock, nbd_reply_ready, NULL, s);
-    s->send_coroutine = NULL;
-    qemu_co_mutex_unlock(&s->send_mutex);
-    return rc;
-}
-
-static void nbd_co_receive_reply(BDRVNBDState *s, struct nbd_request *request,
-                                 struct nbd_reply *reply,
-                                 QEMUIOVector *qiov, int offset)
-{
-    int ret;
-
-    /* Wait until we're woken up by the read handler.  TODO: perhaps
-     * peek at the next reply and avoid yielding if it's ours?  */
-    qemu_coroutine_yield();
-    *reply = s->reply;
-    if (reply->handle != request->handle) {
-        reply->error = EIO;
-    } else {
-        if (qiov && reply->error == 0) {
-            ret = qemu_co_recvv(s->sock, qiov->iov, qiov->niov,
-                                offset, request->len);
-            if (ret != request->len) {
-                reply->error = EIO;
-            }
-        }
-
-        /* Tell the read handler to read another header.  */
-        s->reply.handle = 0;
-    }
-}
-
-static void nbd_coroutine_end(BDRVNBDState *s, struct nbd_request *request)
-{
-    int i = HANDLE_TO_INDEX(s, request->handle);
-    s->recv_coroutine[i] = NULL;
-    if (s->in_flight-- == MAX_NBD_REQUESTS) {
-        qemu_co_mutex_unlock(&s->free_sema);
-    }
-}
-
 static int nbd_establish_connection(BlockDriverState *bs)
 {
     BDRVNBDState *s = bs->opaque;
     int sock;
-    int ret;
-    off_t size;
-    size_t blocksize;
 
-    if (s->is_unix) {
+    if (s->client.is_unix) {
         sock = unix_socket_outgoing(qemu_opt_get(s->socket_opts, "path"));
     } else {
         sock = tcp_socket_outgoing_opts(s->socket_opts);
@@ -417,53 +247,18 @@ static int nbd_establish_connection(BlockDriverState *bs)
         return -errno;
     }
 
-    /* NBD handshake */
-    ret = nbd_receive_negotiate(sock, s->export_name, &s->nbdflags, &size,
-                                &blocksize);
-    if (ret < 0) {
-        logout("Failed to negotiate with the NBD server\n");
-        closesocket(sock);
-        return ret;
-    }
-
-    /* Now that we're connected, set the socket to be non-blocking and
-     * kick the reply mechanism.  */
-    qemu_set_nonblock(sock);
-    qemu_aio_set_fd_handler(sock, nbd_reply_ready, NULL, s);
-
-    s->sock = sock;
-    s->size = size;
-    s->blocksize = blocksize;
-
-    logout("Established connection with NBD server\n");
-    return 0;
-}
-
-static void nbd_teardown_connection(BlockDriverState *bs)
-{
-    BDRVNBDState *s = bs->opaque;
-    struct nbd_request request;
-
-    request.type = NBD_CMD_DISC;
-    request.from = 0;
-    request.len = 0;
-    nbd_send_request(s->sock, &request);
-
-    qemu_aio_set_fd_handler(s->sock, NULL, NULL, NULL);
-    closesocket(s->sock);
+    return sock;
 }
 
 static int nbd_open(BlockDriverState *bs, QDict *options, int flags,
                     Error **errp)
 {
     BDRVNBDState *s = bs->opaque;
-    int result;
-
-    qemu_co_mutex_init(&s->send_mutex);
-    qemu_co_mutex_init(&s->free_sema);
+    char *export = NULL;
+    int result, sock;
 
     /* Pop the config into our state object. Exit if invalid. */
-    result = nbd_config(s, options);
+    result = nbd_config(s, options, &export);
     if (result != 0) {
         return result;
     }
@@ -471,172 +266,64 @@ static int nbd_open(BlockDriverState *bs, QDict *options, int flags,
     /* establish TCP connection, return error if it fails
      * TODO: Configurable retry-until-timeout behaviour.
      */
-    result = nbd_establish_connection(bs);
+    sock = nbd_establish_connection(bs);
+    if (sock < 0) {
+        return sock;
+    }
 
+    /* NBD handshake */
+    result = nbd_client_session_init(&s->client, bs, sock, export);
+    g_free(export);
     return result;
 }
 
-static int nbd_co_readv_1(BlockDriverState *bs, int64_t sector_num,
-                          int nb_sectors, QEMUIOVector *qiov,
-                          int offset)
-{
-    BDRVNBDState *s = bs->opaque;
-    struct nbd_request request;
-    struct nbd_reply reply;
-    ssize_t ret;
-
-    request.type = NBD_CMD_READ;
-    request.from = sector_num * 512;
-    request.len = nb_sectors * 512;
-
-    nbd_coroutine_start(s, &request);
-    ret = nbd_co_send_request(s, &request, NULL, 0);
-    if (ret < 0) {
-        reply.error = -ret;
-    } else {
-        nbd_co_receive_reply(s, &request, &reply, qiov, offset);
-    }
-    nbd_coroutine_end(s, &request);
-    return -reply.error;
-
-}
-
-static int nbd_co_writev_1(BlockDriverState *bs, int64_t sector_num,
-                           int nb_sectors, QEMUIOVector *qiov,
-                           int offset)
-{
-    BDRVNBDState *s = bs->opaque;
-    struct nbd_request request;
-    struct nbd_reply reply;
-    ssize_t ret;
-
-    request.type = NBD_CMD_WRITE;
-    if (!bdrv_enable_write_cache(bs) && (s->nbdflags & NBD_FLAG_SEND_FUA)) {
-        request.type |= NBD_CMD_FLAG_FUA;
-    }
-
-    request.from = sector_num * 512;
-    request.len = nb_sectors * 512;
-
-    nbd_coroutine_start(s, &request);
-    ret = nbd_co_send_request(s, &request, qiov, offset);
-    if (ret < 0) {
-        reply.error = -ret;
-    } else {
-        nbd_co_receive_reply(s, &request, &reply, NULL, 0);
-    }
-    nbd_coroutine_end(s, &request);
-    return -reply.error;
-}
-
-/* qemu-nbd has a limit of slightly less than 1M per request.  Try to
- * remain aligned to 4K. */
-#define NBD_MAX_SECTORS 2040
-
 static int nbd_co_readv(BlockDriverState *bs, int64_t sector_num,
                         int nb_sectors, QEMUIOVector *qiov)
 {
-    int offset = 0;
-    int ret;
-    while (nb_sectors > NBD_MAX_SECTORS) {
-        ret = nbd_co_readv_1(bs, sector_num, NBD_MAX_SECTORS, qiov, offset);
-        if (ret < 0) {
-            return ret;
-        }
-        offset += NBD_MAX_SECTORS * 512;
-        sector_num += NBD_MAX_SECTORS;
-        nb_sectors -= NBD_MAX_SECTORS;
-    }
-    return nbd_co_readv_1(bs, sector_num, nb_sectors, qiov, offset);
+    BDRVNBDState *s = bs->opaque;
+
+    return nbd_client_session_co_readv(&s->client, sector_num,
+                                       nb_sectors, qiov);
 }
 
 static int nbd_co_writev(BlockDriverState *bs, int64_t sector_num,
                          int nb_sectors, QEMUIOVector *qiov)
 {
-    int offset = 0;
-    int ret;
-    while (nb_sectors > NBD_MAX_SECTORS) {
-        ret = nbd_co_writev_1(bs, sector_num, NBD_MAX_SECTORS, qiov, offset);
-        if (ret < 0) {
-            return ret;
-        }
-        offset += NBD_MAX_SECTORS * 512;
-        sector_num += NBD_MAX_SECTORS;
-        nb_sectors -= NBD_MAX_SECTORS;
-    }
-    return nbd_co_writev_1(bs, sector_num, nb_sectors, qiov, offset);
+    BDRVNBDState *s = bs->opaque;
+
+    return nbd_client_session_co_writev(&s->client, sector_num,
+                                        nb_sectors, qiov);
 }
 
 static int nbd_co_flush(BlockDriverState *bs)
 {
     BDRVNBDState *s = bs->opaque;
-    struct nbd_request request;
-    struct nbd_reply reply;
-    ssize_t ret;
 
-    if (!(s->nbdflags & NBD_FLAG_SEND_FLUSH)) {
-        return 0;
-    }
-
-    request.type = NBD_CMD_FLUSH;
-    if (s->nbdflags & NBD_FLAG_SEND_FUA) {
-        request.type |= NBD_CMD_FLAG_FUA;
-    }
-
-    request.from = 0;
-    request.len = 0;
-
-    nbd_coroutine_start(s, &request);
-    ret = nbd_co_send_request(s, &request, NULL, 0);
-    if (ret < 0) {
-        reply.error = -ret;
-    } else {
-        nbd_co_receive_reply(s, &request, &reply, NULL, 0);
-    }
-    nbd_coroutine_end(s, &request);
-    return -reply.error;
+    return nbd_client_session_co_flush(&s->client);
 }
 
 static int nbd_co_discard(BlockDriverState *bs, int64_t sector_num,
                           int nb_sectors)
 {
     BDRVNBDState *s = bs->opaque;
-    struct nbd_request request;
-    struct nbd_reply reply;
-    ssize_t ret;
 
-    if (!(s->nbdflags & NBD_FLAG_SEND_TRIM)) {
-        return 0;
-    }
-    request.type = NBD_CMD_TRIM;
-    request.from = sector_num * 512;
-    request.len = nb_sectors * 512;
-
-    nbd_coroutine_start(s, &request);
-    ret = nbd_co_send_request(s, &request, NULL, 0);
-    if (ret < 0) {
-        reply.error = -ret;
-    } else {
-        nbd_co_receive_reply(s, &request, &reply, NULL, 0);
-    }
-    nbd_coroutine_end(s, &request);
-    return -reply.error;
+    return nbd_client_session_co_discard(&s->client, sector_num,
+                                         nb_sectors);
 }
 
 static void nbd_close(BlockDriverState *bs)
 {
     BDRVNBDState *s = bs->opaque;
-    g_free(s->export_name);
-    qemu_opts_del(s->socket_opts);
 
-    nbd_teardown_connection(bs);
+    qemu_opts_del(s->socket_opts);
+    nbd_client_session_close(&s->client);
 }
 
 static int64_t nbd_getlength(BlockDriverState *bs)
 {
     BDRVNBDState *s = bs->opaque;
 
-    return s->size;
+    return s->client.size;
 }
 
 static BlockDriver bdrv_nbd = {

block/qapi.c

@@ -204,12 +204,9 @@ void bdrv_query_info(BlockDriverState *bs,
         info->io_status = bs->iostatus;
     }
 
-    if (bs->dirty_bitmap) {
-        info->has_dirty = true;
-        info->dirty = g_malloc0(sizeof(*info->dirty));
-        info->dirty->count = bdrv_get_dirty_count(bs) * BDRV_SECTOR_SIZE;
-        info->dirty->granularity =
-         ((int64_t) BDRV_SECTOR_SIZE << hbitmap_granularity(bs->dirty_bitmap));
+    if (!QLIST_EMPTY(&bs->dirty_bitmaps)) {
+        info->has_dirty_bitmaps = true;
+        info->dirty_bitmaps = bdrv_query_dirty_bitmaps(bs);
     }
 
     if (bs->drv) {

block/qcow2-cluster.c

@@ -1401,7 +1401,7 @@ int qcow2_discard_clusters(BlockDriverState *bs, uint64_t offset,
 
     /* Round start up and end down */
     offset = align_offset(offset, s->cluster_size);
-    end_offset &= ~(s->cluster_size - 1);
+    end_offset = start_of_cluster(s, end_offset);
 
     if (offset > end_offset) {
         return 0;
@@ -1613,7 +1613,7 @@ static int expand_zero_clusters_in_l1(BlockDriverState *bs, uint64_t *l1_table,
             }
 
             ret = bdrv_write_zeroes(bs->file, offset / BDRV_SECTOR_SIZE,
-                                    s->cluster_sectors);
+                                    s->cluster_sectors, 0);
             if (ret < 0) {
                 if (!preallocated) {
                     qcow2_free_clusters(bs, offset, s->cluster_size,

block/qcow2-refcount.c

@@ -515,8 +515,8 @@ static int QEMU_WARN_UNUSED_RESULT update_refcount(BlockDriverState *bs,
             s->l2_table_cache);
     }
 
-    start = offset & ~(s->cluster_size - 1);
-    last = (offset + length - 1) & ~(s->cluster_size - 1);
+    start = start_of_cluster(s, offset);
+    last = start_of_cluster(s, offset + length - 1);
     for(cluster_offset = start; cluster_offset <= last;
         cluster_offset += s->cluster_size)
     {
@@ -724,7 +724,7 @@ int64_t qcow2_alloc_bytes(BlockDriverState *bs, int size)
     }
  redo:
     free_in_cluster = s->cluster_size -
-        (s->free_byte_offset & (s->cluster_size - 1));
+        offset_into_cluster(s, s->free_byte_offset);
     if (size <= free_in_cluster) {
         /* enough space in current cluster */
         offset = s->free_byte_offset;
@@ -732,7 +732,7 @@ int64_t qcow2_alloc_bytes(BlockDriverState *bs, int size)
         free_in_cluster -= size;
         if (free_in_cluster == 0)
             s->free_byte_offset = 0;
-        if ((offset & (s->cluster_size - 1)) != 0)
+        if (offset_into_cluster(s, offset) != 0)
             qcow2_update_cluster_refcount(bs, offset >> s->cluster_bits, 1,
                                           QCOW2_DISCARD_NEVER);
     } else {
@@ -740,7 +740,7 @@ int64_t qcow2_alloc_bytes(BlockDriverState *bs, int size)
         if (offset < 0) {
             return offset;
         }
-        cluster_offset = s->free_byte_offset & ~(s->cluster_size - 1);
+        cluster_offset = start_of_cluster(s, s->free_byte_offset);
         if ((cluster_offset + s->cluster_size) == offset) {
             /* we are lucky: contiguous data */
             offset = s->free_byte_offset;
@@ -1010,8 +1010,8 @@ static void inc_refcounts(BlockDriverState *bs,
     if (size <= 0)
         return;
 
-    start = offset & ~(s->cluster_size - 1);
-    last = (offset + size - 1) & ~(s->cluster_size - 1);
+    start = start_of_cluster(s, offset);
+    last = start_of_cluster(s, offset + size - 1);
     for(cluster_offset = start; cluster_offset <= last;
         cluster_offset += s->cluster_size) {
         k = cluster_offset >> s->cluster_bits;
@@ -1122,7 +1122,7 @@ static int check_refcounts_l2(BlockDriverState *bs, BdrvCheckResult *res,
                 offset, s->cluster_size);
 
             /* Correct offsets are cluster aligned */
-            if (offset & (s->cluster_size - 1)) {
+            if (offset_into_cluster(s, offset)) {
                 fprintf(stderr, "ERROR offset=%" PRIx64 ": Cluster is not "
                     "properly aligned; L2 entry corrupted.\n", offset);
                 res->corruptions++;
@@ -1194,7 +1194,7 @@ static int check_refcounts_l1(BlockDriverState *bs,
                 l2_offset, s->cluster_size);
 
             /* L2 tables are cluster aligned */
-            if (l2_offset & (s->cluster_size - 1)) {
+            if (offset_into_cluster(s, l2_offset)) {
                 fprintf(stderr, "ERROR l2_offset=%" PRIx64 ": Table is not "
                     "cluster aligned; L1 entry corrupted\n", l2_offset);
                 res->corruptions++;
@@ -1423,7 +1423,7 @@ static int64_t realloc_refcount_block(BlockDriverState *bs, int reftable_index,
     }
 
     /* update refcount table */
-    assert(!(new_offset & (s->cluster_size - 1)));
+    assert(!offset_into_cluster(s, new_offset));
     s->refcount_table[reftable_index] = new_offset;
     ret = write_reftable_entry(bs, reftable_index);
     if (ret < 0) {
@@ -1507,7 +1507,7 @@ int qcow2_check_refcounts(BlockDriverState *bs, BdrvCheckResult *res,
         cluster = offset >> s->cluster_bits;
 
         /* Refcount blocks are cluster aligned */
-        if (offset & (s->cluster_size - 1)) {
+        if (offset_into_cluster(s, offset)) {
             fprintf(stderr, "ERROR refcount block %" PRId64 " is not "
                 "cluster aligned; refcount table entry corrupted\n", i);
             res->corruptions++;

block/qcow2-snapshot.c

@@ -675,7 +675,10 @@ int qcow2_snapshot_list(BlockDriverState *bs, QEMUSnapshotInfo **psn_tab)
     return s->nb_snapshots;
 }
 
-int qcow2_snapshot_load_tmp(BlockDriverState *bs, const char *snapshot_name)
+int qcow2_snapshot_load_tmp(BlockDriverState *bs,
+                            const char *snapshot_id,
+                            const char *name,
+                            Error **errp)
 {
     int i, snapshot_index;
     BDRVQcowState *s = bs->opaque;
@@ -687,8 +690,10 @@ int qcow2_snapshot_load_tmp(BlockDriverState *bs, const char *snapshot_name)
     assert(bs->read_only);
 
     /* Search the snapshot */
-    snapshot_index = find_snapshot_by_id_or_name(bs, snapshot_name);
+    snapshot_index = find_snapshot_by_id_and_name(bs, snapshot_id, name);
     if (snapshot_index < 0) {
+        error_setg(errp,
+                   "Can't find snapshot");
         return -ENOENT;
     }
     sn = &s->snapshots[snapshot_index];
@@ -699,6 +704,7 @@ int qcow2_snapshot_load_tmp(BlockDriverState *bs, const char *snapshot_name)
 
     ret = bdrv_pread(bs->file, sn->l1_table_offset, new_l1_table, new_l1_bytes);
     if (ret < 0) {
+        error_setg(errp, "Failed to read l1 table for snapshot");
         g_free(new_l1_table);
         return ret;
     }

block/qcow2.c

@@ -669,7 +669,7 @@ static int qcow2_open(BlockDriverState *bs, QDict *options, int flags,
     }
 
     /* Enable lazy_refcounts according to image and command line options */
-    opts = qemu_opts_create_nofail(&qcow2_runtime_opts);
+    opts = qemu_opts_create(&qcow2_runtime_opts, NULL, 0, &error_abort);
     qemu_opts_absorb_qdict(opts, options, &local_err);
     if (error_is_set(&local_err)) {
         error_propagate(errp, local_err);
@@ -718,6 +718,7 @@ static int qcow2_open(BlockDriverState *bs, QDict *options, int flags,
     }
 
     qemu_opts_del(opts);
+    bs->bl.write_zeroes_alignment = s->cluster_sectors;
 
     if (s->use_lazy_refcounts && s->qcow_version < 3) {
         error_setg(errp, "Lazy refcounts require a qcow2 image with at least "
@@ -1471,7 +1472,7 @@ static int qcow2_create2(const char *filename, int64_t total_size,
      * size for any qcow2 image.
      */
     BlockDriverState* bs;
-    QCowHeader header;
+    QCowHeader *header;
     uint8_t* refcount_table;
     Error *local_err = NULL;
     int ret;
@@ -1489,30 +1490,34 @@ static int qcow2_create2(const char *filename, int64_t total_size,
     }
 
     /* Write the header */
-    memset(&header, 0, sizeof(header));
-    header.magic = cpu_to_be32(QCOW_MAGIC);
-    header.version = cpu_to_be32(version);
-    header.cluster_bits = cpu_to_be32(cluster_bits);
-    header.size = cpu_to_be64(0);
-    header.l1_table_offset = cpu_to_be64(0);
-    header.l1_size = cpu_to_be32(0);
-    header.refcount_table_offset = cpu_to_be64(cluster_size);
-    header.refcount_table_clusters = cpu_to_be32(1);
-    header.refcount_order = cpu_to_be32(3 + REFCOUNT_SHIFT);
-    header.header_length = cpu_to_be32(sizeof(header));
+    QEMU_BUILD_BUG_ON((1 << MIN_CLUSTER_BITS) < sizeof(*header));
+    header = g_malloc0(cluster_size);
+    *header = (QCowHeader) {
+        .magic                      = cpu_to_be32(QCOW_MAGIC),
+        .version                    = cpu_to_be32(version),
+        .cluster_bits               = cpu_to_be32(cluster_bits),
+        .size                       = cpu_to_be64(0),
+        .l1_table_offset            = cpu_to_be64(0),
+        .l1_size                    = cpu_to_be32(0),
+        .refcount_table_offset      = cpu_to_be64(cluster_size),
+        .refcount_table_clusters    = cpu_to_be32(1),
+        .refcount_order             = cpu_to_be32(3 + REFCOUNT_SHIFT),
+        .header_length              = cpu_to_be32(sizeof(*header)),
+    };
 
     if (flags & BLOCK_FLAG_ENCRYPT) {
-        header.crypt_method = cpu_to_be32(QCOW_CRYPT_AES);
+        header->crypt_method = cpu_to_be32(QCOW_CRYPT_AES);
     } else {
-        header.crypt_method = cpu_to_be32(QCOW_CRYPT_NONE);
+        header->crypt_method = cpu_to_be32(QCOW_CRYPT_NONE);
     }
 
     if (flags & BLOCK_FLAG_LAZY_REFCOUNTS) {
-        header.compatible_features |=
+        header->compatible_features |=
             cpu_to_be64(QCOW2_COMPAT_LAZY_REFCOUNTS);
     }
 
-    ret = bdrv_pwrite(bs, 0, &header, sizeof(header));
+    ret = bdrv_pwrite(bs, 0, header, cluster_size);
+    g_free(header);
     if (ret < 0) {
         error_setg_errno(errp, -ret, "Could not write qcow2 header");
         goto out;
@@ -1588,7 +1593,8 @@ static int qcow2_create2(const char *filename, int64_t total_size,
 
     /* Reopen the image without BDRV_O_NO_FLUSH to flush it before returning */
     ret = bdrv_open(bs, filename, NULL,
-                    BDRV_O_RDWR | BDRV_O_CACHE_WB, drv, &local_err);
+                    BDRV_O_RDWR | BDRV_O_CACHE_WB | BDRV_O_NO_BACKING,
+                    drv, &local_err);
     if (error_is_set(&local_err)) {
         error_propagate(errp, local_err);
         goto out;
@@ -1696,7 +1702,7 @@ static int qcow2_make_empty(BlockDriverState *bs)
 }
 
 static coroutine_fn int qcow2_co_write_zeroes(BlockDriverState *bs,
-    int64_t sector_num, int nb_sectors)
+    int64_t sector_num, int nb_sectors, BdrvRequestFlags flags)
 {
     int ret;
     BDRVQcowState *s = bs->opaque;
@@ -1892,6 +1898,8 @@ static coroutine_fn int qcow2_co_flush_to_os(BlockDriverState *bs)
 static int qcow2_get_info(BlockDriverState *bs, BlockDriverInfo *bdi)
 {
     BDRVQcowState *s = bs->opaque;
+    bdi->unallocated_blocks_are_zero = true;
+    bdi->can_write_zeroes_with_unmap = (s->qcow_version >= 3);
     bdi->cluster_size = s->cluster_size;
     bdi->vm_state_offset = qcow2_vm_state_offset(s);
     return 0;

block/qcow2.h

@@ -488,7 +488,10 @@ int qcow2_snapshot_delete(BlockDriverState *bs,
                           const char *name,
                           Error **errp);
 int qcow2_snapshot_list(BlockDriverState *bs, QEMUSnapshotInfo **psn_tab);
-int qcow2_snapshot_load_tmp(BlockDriverState *bs, const char *snapshot_name);
+int qcow2_snapshot_load_tmp(BlockDriverState *bs,
+                            const char *snapshot_id,
+                            const char *name,
+                            Error **errp);
 
 void qcow2_free_snapshots(BlockDriverState *bs);
 int qcow2_read_snapshots(BlockDriverState *bs);

block/qed.c

@@ -495,6 +495,7 @@ static int bdrv_qed_open(BlockDriverState *bs, QDict *options, int flags,
         }
     }
 
+    bs->bl.write_zeroes_alignment = s->header.cluster_size >> BDRV_SECTOR_BITS;
     s->need_check_timer = timer_new_ns(QEMU_CLOCK_VIRTUAL,
                                             qed_need_check_timer_cb, s);
 
@@ -1397,7 +1398,8 @@ static void coroutine_fn qed_co_write_zeroes_cb(void *opaque, int ret)
 
 static int coroutine_fn bdrv_qed_co_write_zeroes(BlockDriverState *bs,
                                                  int64_t sector_num,
-                                                 int nb_sectors)
+                                                 int nb_sectors,
+                                                 BdrvRequestFlags flags)
 {
     BlockDriverAIOCB *blockacb;
     BDRVQEDState *s = bs->opaque;
@@ -1474,6 +1476,8 @@ static int bdrv_qed_get_info(BlockDriverState *bs, BlockDriverInfo *bdi)
     memset(bdi, 0, sizeof(*bdi));
     bdi->cluster_size = s->header.cluster_size;
     bdi->is_dirty = s->header.features & QED_F_NEED_CHECK;
+    bdi->unallocated_blocks_are_zero = true;
+    bdi->can_write_zeroes_with_unmap = true;
     return 0;
 }
 

block/raw-aio.h

@@ -21,9 +21,10 @@
 #define QEMU_AIO_IOCTL        0x0004
 #define QEMU_AIO_FLUSH        0x0008
 #define QEMU_AIO_DISCARD      0x0010
+#define QEMU_AIO_WRITE_ZEROES 0x0020
 #define QEMU_AIO_TYPE_MASK \
         (QEMU_AIO_READ|QEMU_AIO_WRITE|QEMU_AIO_IOCTL|QEMU_AIO_FLUSH| \
-         QEMU_AIO_DISCARD)
+         QEMU_AIO_DISCARD|QEMU_AIO_WRITE_ZEROES)
 
 /* AIO flags */
 #define QEMU_AIO_MISALIGNED   0x1000

block/raw-posix.c

@@ -139,9 +139,11 @@ typedef struct BDRVRawState {
     void *aio_ctx;
 #endif
 #ifdef CONFIG_XFS
-    bool is_xfs : 1;
+    bool is_xfs:1;
 #endif
-    bool has_discard : 1;
+    bool has_discard:1;
+    bool has_write_zeroes:1;
+    bool discard_zeroes:1;
 } BDRVRawState;
 
 typedef struct BDRVRawReopenState {
@@ -283,8 +285,9 @@ static int raw_open_common(BlockDriverState *bs, QDict *options,
     Error *local_err = NULL;
     const char *filename;
     int fd, ret;
+    struct stat st;
 
-    opts = qemu_opts_create_nofail(&raw_runtime_opts);
+    opts = qemu_opts_create(&raw_runtime_opts, NULL, 0, &error_abort);
     qemu_opts_absorb_qdict(opts, options, &local_err);
     if (error_is_set(&local_err)) {
         error_propagate(errp, local_err);
@@ -323,10 +326,38 @@ static int raw_open_common(BlockDriverState *bs, QDict *options,
     }
 #endif
 
-    s->has_discard = 1;
+    s->has_discard = true;
+    s->has_write_zeroes = true;
+
+    if (fstat(s->fd, &st) < 0) {
+        error_setg_errno(errp, errno, "Could not stat file");
+        goto fail;
+    }
+    if (S_ISREG(st.st_mode)) {
+        s->discard_zeroes = true;
+    }
+    if (S_ISBLK(st.st_mode)) {
+#ifdef BLKDISCARDZEROES
+        unsigned int arg;
+        if (ioctl(s->fd, BLKDISCARDZEROES, &arg) == 0 && arg) {
+            s->discard_zeroes = true;
+        }
+#endif
+#ifdef __linux__
+        /* On Linux 3.10, BLKDISCARD leaves stale data in the page cache.  Do
+         * not rely on the contents of discarded blocks unless using O_DIRECT.
+         * Same for BLKZEROOUT.
+         */
+        if (!(bs->open_flags & BDRV_O_NOCACHE)) {
+            s->discard_zeroes = false;
+            s->has_write_zeroes = false;
+        }
+#endif
+    }
+
 #ifdef CONFIG_XFS
     if (platform_test_xfs_fd(s->fd)) {
-        s->is_xfs = 1;
+        s->is_xfs = true;
     }
 #endif
 
@@ -675,6 +706,23 @@ static ssize_t handle_aiocb_rw(RawPosixAIOData *aiocb)
 }
 
 #ifdef CONFIG_XFS
+static int xfs_write_zeroes(BDRVRawState *s, int64_t offset, uint64_t bytes)
+{
+    struct xfs_flock64 fl;
+
+    memset(&fl, 0, sizeof(fl));
+    fl.l_whence = SEEK_SET;
+    fl.l_start = offset;
+    fl.l_len = bytes;
+
+    if (xfsctl(NULL, s->fd, XFS_IOC_ZERO_RANGE, &fl) < 0) {
+        DEBUG_BLOCK_PRINT("cannot write zero range (%s)\n", strerror(errno));
+        return -errno;
+    }
+
+    return 0;
+}
+
 static int xfs_discard(BDRVRawState *s, int64_t offset, uint64_t bytes)
 {
     struct xfs_flock64 fl;
@@ -693,13 +741,49 @@ static int xfs_discard(BDRVRawState *s, int64_t offset, uint64_t bytes)
 }
 #endif
 
+static ssize_t handle_aiocb_write_zeroes(RawPosixAIOData *aiocb)
+{
+    int ret = -EOPNOTSUPP;
+    BDRVRawState *s = aiocb->bs->opaque;
+
+    if (s->has_write_zeroes == 0) {
+        return -ENOTSUP;
+    }
+
+    if (aiocb->aio_type & QEMU_AIO_BLKDEV) {
+#ifdef BLKZEROOUT
+        do {
+            uint64_t range[2] = { aiocb->aio_offset, aiocb->aio_nbytes };
+            if (ioctl(aiocb->aio_fildes, BLKZEROOUT, range) == 0) {
+                return 0;
+            }
+        } while (errno == EINTR);
+
+        ret = -errno;
+#endif
+    } else {
+#ifdef CONFIG_XFS
+        if (s->is_xfs) {
+            return xfs_write_zeroes(s, aiocb->aio_offset, aiocb->aio_nbytes);
+        }
+#endif
+    }
+
+    if (ret == -ENODEV || ret == -ENOSYS || ret == -EOPNOTSUPP ||
+        ret == -ENOTTY) {
+        s->has_write_zeroes = false;
+        ret = -ENOTSUP;
+    }
+    return ret;
+}
+
 static ssize_t handle_aiocb_discard(RawPosixAIOData *aiocb)
 {
     int ret = -EOPNOTSUPP;
     BDRVRawState *s = aiocb->bs->opaque;
 
-    if (s->has_discard == 0) {
-        return 0;
+    if (!s->has_discard) {
+        return -ENOTSUP;
     }
 
     if (aiocb->aio_type & QEMU_AIO_BLKDEV) {
@@ -734,8 +818,8 @@ static ssize_t handle_aiocb_discard(RawPosixAIOData *aiocb)
 
     if (ret == -ENODEV || ret == -ENOSYS || ret == -EOPNOTSUPP ||
         ret == -ENOTTY) {
-        s->has_discard = 0;
-        ret = 0;
+        s->has_discard = false;
+        ret = -ENOTSUP;
     }
     return ret;
 }
@@ -777,6 +861,9 @@ static int aio_worker(void *arg)
     case QEMU_AIO_DISCARD:
         ret = handle_aiocb_discard(aiocb);
         break;
+    case QEMU_AIO_WRITE_ZEROES:
+        ret = handle_aiocb_write_zeroes(aiocb);
+        break;
     default:
         fprintf(stderr, "invalid aio request (0x%x)\n", aiocb->aio_type);
         ret = -EINVAL;
@@ -787,6 +874,29 @@ static int aio_worker(void *arg)
     return ret;
 }
 
+static int paio_submit_co(BlockDriverState *bs, int fd,
+        int64_t sector_num, QEMUIOVector *qiov, int nb_sectors,
+        int type)
+{
+    RawPosixAIOData *acb = g_slice_new(RawPosixAIOData);
+    ThreadPool *pool;
+
+    acb->bs = bs;
+    acb->aio_type = type;
+    acb->aio_fildes = fd;
+
+    if (qiov) {
+        acb->aio_iov = qiov->iov;
+        acb->aio_niov = qiov->niov;
+    }
+    acb->aio_nbytes = nb_sectors * 512;
+    acb->aio_offset = sector_num * 512;
+
+    trace_paio_submit_co(sector_num, nb_sectors, type);
+    pool = aio_get_thread_pool(bdrv_get_aio_context(bs));
+    return thread_pool_submit_co(pool, aio_worker, acb);
+}
+
 static BlockDriverAIOCB *paio_submit(BlockDriverState *bs, int fd,
         int64_t sector_num, QEMUIOVector *qiov, int nb_sectors,
         BlockDriverCompletionFunc *cb, void *opaque, int type)
@@ -1199,6 +1309,31 @@ static coroutine_fn BlockDriverAIOCB *raw_aio_discard(BlockDriverState *bs,
                        cb, opaque, QEMU_AIO_DISCARD);
 }
 
+static int coroutine_fn raw_co_write_zeroes(
+    BlockDriverState *bs, int64_t sector_num,
+    int nb_sectors, BdrvRequestFlags flags)
+{
+    BDRVRawState *s = bs->opaque;
+
+    if (!(flags & BDRV_REQ_MAY_UNMAP)) {
+        return paio_submit_co(bs, s->fd, sector_num, NULL, nb_sectors,
+                              QEMU_AIO_WRITE_ZEROES);
+    } else if (s->discard_zeroes) {
+        return paio_submit_co(bs, s->fd, sector_num, NULL, nb_sectors,
+                              QEMU_AIO_DISCARD);
+    }
+    return -ENOTSUP;
+}
+
+static int raw_get_info(BlockDriverState *bs, BlockDriverInfo *bdi)
+{
+    BDRVRawState *s = bs->opaque;
+
+    bdi->unallocated_blocks_are_zero = s->discard_zeroes;
+    bdi->can_write_zeroes_with_unmap = s->discard_zeroes;
+    return 0;
+}
+
 static QEMUOptionParameter raw_create_options[] = {
     {
         .name = BLOCK_OPT_SIZE,
@@ -1222,6 +1357,7 @@ static BlockDriver bdrv_file = {
     .bdrv_create = raw_create,
     .bdrv_has_zero_init = bdrv_has_zero_init_1,
     .bdrv_co_get_block_status = raw_co_get_block_status,
+    .bdrv_co_write_zeroes = raw_co_write_zeroes,
 
     .bdrv_aio_readv = raw_aio_readv,
     .bdrv_aio_writev = raw_aio_writev,
@@ -1230,6 +1366,7 @@ static BlockDriver bdrv_file = {
 
     .bdrv_truncate = raw_truncate,
     .bdrv_getlength = raw_getlength,
+    .bdrv_get_info = raw_get_info,
     .bdrv_get_allocated_file_size
                         = raw_get_allocated_file_size,
 
@@ -1525,6 +1662,26 @@ static coroutine_fn BlockDriverAIOCB *hdev_aio_discard(BlockDriverState *bs,
                        cb, opaque, QEMU_AIO_DISCARD|QEMU_AIO_BLKDEV);
 }
 
+static coroutine_fn int hdev_co_write_zeroes(BlockDriverState *bs,
+    int64_t sector_num, int nb_sectors, BdrvRequestFlags flags)
+{
+    BDRVRawState *s = bs->opaque;
+    int rc;
+
+    rc = fd_open(bs);
+    if (rc < 0) {
+        return rc;
+    }
+    if (!(flags & BDRV_REQ_MAY_UNMAP)) {
+        return paio_submit_co(bs, s->fd, sector_num, NULL, nb_sectors,
+                              QEMU_AIO_WRITE_ZEROES|QEMU_AIO_BLKDEV);
+    } else if (s->discard_zeroes) {
+        return paio_submit_co(bs, s->fd, sector_num, NULL, nb_sectors,
+                              QEMU_AIO_DISCARD|QEMU_AIO_BLKDEV);
+    }
+    return -ENOTSUP;
+}
+
 static int hdev_create(const char *filename, QEMUOptionParameter *options,
                        Error **errp)
 {
@@ -1577,6 +1734,7 @@ static BlockDriver bdrv_host_device = {
     .bdrv_reopen_abort   = raw_reopen_abort,
     .bdrv_create        = hdev_create,
     .create_options     = raw_create_options,
+    .bdrv_co_write_zeroes = hdev_co_write_zeroes,
 
     .bdrv_aio_readv	= raw_aio_readv,
     .bdrv_aio_writev	= raw_aio_writev,
@@ -1585,6 +1743,7 @@ static BlockDriver bdrv_host_device = {
 
     .bdrv_truncate      = raw_truncate,
     .bdrv_getlength	= raw_getlength,
+    .bdrv_get_info = raw_get_info,
     .bdrv_get_allocated_file_size
                         = raw_get_allocated_file_size,
 

block/raw-win32.c

@@ -248,7 +248,7 @@ static int raw_open(BlockDriverState *bs, QDict *options, int flags,
 
     s->type = FTYPE_FILE;
 
-    opts = qemu_opts_create_nofail(&raw_runtime_opts);
+    opts = qemu_opts_create(&raw_runtime_opts, NULL, 0, &error_abort);
     qemu_opts_absorb_qdict(opts, options, &local_err);
     if (error_is_set(&local_err)) {
         error_propagate(errp, local_err);
@@ -550,7 +550,8 @@ static int hdev_open(BlockDriverState *bs, QDict *options, int flags,
     Error *local_err = NULL;
     const char *filename;
 
-    QemuOpts *opts = qemu_opts_create_nofail(&raw_runtime_opts);
+    QemuOpts *opts = qemu_opts_create(&raw_runtime_opts, NULL, 0,
+                                      &error_abort);
     qemu_opts_absorb_qdict(opts, options, &local_err);
     if (error_is_set(&local_err)) {
         error_propagate(errp, local_err);

block/raw_bsd.c

@@ -68,9 +68,10 @@ static int64_t coroutine_fn raw_co_get_block_status(BlockDriverState *bs,
 }
 
 static int coroutine_fn raw_co_write_zeroes(BlockDriverState *bs,
-                                            int64_t sector_num, int nb_sectors)
+                                            int64_t sector_num, int nb_sectors,
+                                            BdrvRequestFlags flags)
 {
-    return bdrv_co_write_zeroes(bs->file, sector_num, nb_sectors);
+    return bdrv_co_write_zeroes(bs->file, sector_num, nb_sectors, flags);
 }
 
 static int coroutine_fn raw_co_discard(BlockDriverState *bs,
@@ -149,6 +150,7 @@ static int raw_open(BlockDriverState *bs, QDict *options, int flags,
                     Error **errp)
 {
     bs->sg = bs->file->sg;
+    bs->bl = bs->file->bl;
     return 0;
 }
 

block/rbd.c

@@ -461,7 +461,7 @@ static int qemu_rbd_open(BlockDriverState *bs, QDict *options, int flags,
     const char *filename;
     int r;
 
-    opts = qemu_opts_create_nofail(&runtime_opts);
+    opts = qemu_opts_create(&runtime_opts, NULL, 0, &error_abort);
     qemu_opts_absorb_qdict(opts, options, &local_err);
     if (error_is_set(&local_err)) {
         qerror_report_err(local_err);

block/sheepdog.c

@@ -91,6 +91,14 @@
 #define SD_NR_VDIS   (1U << 24)
 #define SD_DATA_OBJ_SIZE (UINT64_C(1) << 22)
 #define SD_MAX_VDI_SIZE (SD_DATA_OBJ_SIZE * MAX_DATA_OBJS)
+/*
+ * For erasure coding, we use at most SD_EC_MAX_STRIP for data strips and
+ * (SD_EC_MAX_STRIP - 1) for parity strips
+ *
+ * SD_MAX_COPIES is sum of number of data strips and parity strips.
+ */
+#define SD_EC_MAX_STRIP 16
+#define SD_MAX_COPIES (SD_EC_MAX_STRIP * 2 - 1)
 
 #define SD_INODE_SIZE (sizeof(SheepdogInode))
 #define CURRENT_VDI_ID 0
@@ -1375,7 +1383,7 @@ static int sd_open(BlockDriverState *bs, QDict *options, int flags,
 
     s->bs = bs;
 
-    opts = qemu_opts_create_nofail(&runtime_opts);
+    opts = qemu_opts_create(&runtime_opts, NULL, 0, &error_abort);
     qemu_opts_absorb_qdict(opts, options, &local_err);
     if (error_is_set(&local_err)) {
         qerror_report_err(local_err);
@@ -1464,9 +1472,7 @@ out:
     return ret;
 }
 
-static int do_sd_create(BDRVSheepdogState *s, char *filename, int64_t vdi_size,
-                        uint32_t base_vid, uint32_t *vdi_id, int snapshot,
-                        uint8_t copy_policy)
+static int do_sd_create(BDRVSheepdogState *s, uint32_t *vdi_id, int snapshot)
 {
     SheepdogVdiReq hdr;
     SheepdogVdiRsp *rsp = (SheepdogVdiRsp *)&hdr;
@@ -1483,11 +1489,11 @@ static int do_sd_create(BDRVSheepdogState *s, char *filename, int64_t vdi_size,
      * does not fit in buf?  For now, just truncate and avoid buffer overrun.
      */
     memset(buf, 0, sizeof(buf));
-    pstrcpy(buf, sizeof(buf), filename);
+    pstrcpy(buf, sizeof(buf), s->name);
 
     memset(&hdr, 0, sizeof(hdr));
     hdr.opcode = SD_OP_NEW_VDI;
-    hdr.vdi_id = base_vid;
+    hdr.vdi_id = s->inode.vdi_id;
 
     wlen = SD_MAX_VDI_LEN;
 
@@ -1495,8 +1501,9 @@ static int do_sd_create(BDRVSheepdogState *s, char *filename, int64_t vdi_size,
     hdr.snapid = snapshot;
 
     hdr.data_length = wlen;
-    hdr.vdi_size = vdi_size;
-    hdr.copy_policy = copy_policy;
+    hdr.vdi_size = s->inode.vdi_size;
+    hdr.copy_policy = s->inode.copy_policy;
+    hdr.copies = s->inode.nr_copies;
 
     ret = do_req(fd, (SheepdogReq *)&hdr, buf, &wlen, &rlen);
 
@@ -1507,7 +1514,7 @@ static int do_sd_create(BDRVSheepdogState *s, char *filename, int64_t vdi_size,
     }
 
     if (rsp->result != SD_RES_SUCCESS) {
-        error_report("%s, %s", sd_strerror(rsp->result), filename);
+        error_report("%s, %s", sd_strerror(rsp->result), s->inode.name);
         return -EIO;
     }
 
@@ -1564,27 +1571,79 @@ out:
     return ret;
 }
 
+/*
+ * Sheepdog support two kinds of redundancy, full replication and erasure
+ * coding.
+ *
+ * # create a fully replicated vdi with x copies
+ * -o redundancy=x (1 <= x <= SD_MAX_COPIES)
+ *
+ * # create a erasure coded vdi with x data strips and y parity strips
+ * -o redundancy=x:y (x must be one of {2,4,8,16} and 1 <= y < SD_EC_MAX_STRIP)
+ */
+static int parse_redundancy(BDRVSheepdogState *s, const char *opt)
+{
+    struct SheepdogInode *inode = &s->inode;
+    const char *n1, *n2;
+    long copy, parity;
+    char p[10];
+
+    pstrcpy(p, sizeof(p), opt);
+    n1 = strtok(p, ":");
+    n2 = strtok(NULL, ":");
+
+    if (!n1) {
+        return -EINVAL;
+    }
+
+    copy = strtol(n1, NULL, 10);
+    if (copy > SD_MAX_COPIES || copy < 1) {
+        return -EINVAL;
+    }
+    if (!n2) {
+        inode->copy_policy = 0;
+        inode->nr_copies = copy;
+        return 0;
+    }
+
+    if (copy != 2 && copy != 4 && copy != 8 && copy != 16) {
+        return -EINVAL;
+    }
+
+    parity = strtol(n2, NULL, 10);
+    if (parity >= SD_EC_MAX_STRIP || parity < 1) {
+        return -EINVAL;
+    }
+
+    /*
+     * 4 bits for parity and 4 bits for data.
+     * We have to compress upper data bits because it can't represent 16
+     */
+    inode->copy_policy = ((copy / 2) << 4) + parity;
+    inode->nr_copies = copy + parity;
+
+    return 0;
+}
+
 static int sd_create(const char *filename, QEMUOptionParameter *options,
                      Error **errp)
 {
     int ret = 0;
-    uint32_t vid = 0, base_vid = 0;
-    int64_t vdi_size = 0;
+    uint32_t vid = 0;
     char *backing_file = NULL;
     BDRVSheepdogState *s;
-    char vdi[SD_MAX_VDI_LEN], tag[SD_MAX_VDI_TAG_LEN];
+    char tag[SD_MAX_VDI_TAG_LEN];
     uint32_t snapid;
     bool prealloc = false;
     Error *local_err = NULL;
 
     s = g_malloc0(sizeof(BDRVSheepdogState));
 
-    memset(vdi, 0, sizeof(vdi));
     memset(tag, 0, sizeof(tag));
     if (strstr(filename, "://")) {
-        ret = sd_parse_uri(s, filename, vdi, &snapid, tag);
+        ret = sd_parse_uri(s, filename, s->name, &snapid, tag);
     } else {
-        ret = parse_vdiname(s, filename, vdi, &snapid, tag);
+        ret = parse_vdiname(s, filename, s->name, &snapid, tag);
     }
     if (ret < 0) {
         goto out;
@@ -1592,7 +1651,7 @@ static int sd_create(const char *filename, QEMUOptionParameter *options,
 
     while (options && options->name) {
         if (!strcmp(options->name, BLOCK_OPT_SIZE)) {
-            vdi_size = options->value.n;
+            s->inode.vdi_size = options->value.n;
         } else if (!strcmp(options->name, BLOCK_OPT_BACKING_FILE)) {
             backing_file = options->value.s;
         } else if (!strcmp(options->name, BLOCK_OPT_PREALLOC)) {
@@ -1606,11 +1665,18 @@ static int sd_create(const char *filename, QEMUOptionParameter *options,
                 ret = -EINVAL;
                 goto out;
             }
+        } else if (!strcmp(options->name, BLOCK_OPT_REDUNDANCY)) {
+            if (options->value.s) {
+                ret = parse_redundancy(s, options->value.s);
+                if (ret < 0) {
+                    goto out;
+                }
+            }
         }
         options++;
     }
 
-    if (vdi_size > SD_MAX_VDI_SIZE) {
+    if (s->inode.vdi_size > SD_MAX_VDI_SIZE) {
         error_report("too big image size");
         ret = -EINVAL;
         goto out;
@@ -1645,12 +1711,10 @@ static int sd_create(const char *filename, QEMUOptionParameter *options,
             goto out;
         }
 
-        base_vid = s->inode.vdi_id;
         bdrv_unref(bs);
     }
 
-    /* TODO: allow users to specify copy number */
-    ret = do_sd_create(s, vdi, vdi_size, base_vid, &vid, 0, 0);
+    ret = do_sd_create(s, &vid, 0);
     if (!prealloc || ret) {
         goto out;
     }
@@ -1833,8 +1897,7 @@ static int sd_create_branch(BDRVSheepdogState *s)
      * false bail out.
      */
     deleted = sd_delete(s);
-    ret = do_sd_create(s, s->name, s->inode.vdi_size, s->inode.vdi_id, &vid,
-                       !deleted, s->inode.copy_policy);
+    ret = do_sd_create(s, &vid, !deleted);
     if (ret) {
         goto out;
     }
@@ -1985,13 +2048,14 @@ static coroutine_fn int sd_co_writev(BlockDriverState *bs, int64_t sector_num,
 {
     SheepdogAIOCB *acb;
     int ret;
+    int64_t offset = (sector_num + nb_sectors) * BDRV_SECTOR_SIZE;
+    BDRVSheepdogState *s = bs->opaque;
 
-    if (bs->growable && sector_num + nb_sectors > bs->total_sectors) {
-        ret = sd_truncate(bs, (sector_num + nb_sectors) * BDRV_SECTOR_SIZE);
+    if (bs->growable && offset > s->inode.vdi_size) {
+        ret = sd_truncate(bs, offset);
         if (ret < 0) {
             return ret;
         }
-        bs->total_sectors = sector_num + nb_sectors;
     }
 
     acb = sd_aio_setup(bs, qiov, sector_num, nb_sectors);
@@ -2097,8 +2161,7 @@ static int sd_snapshot_create(BlockDriverState *bs, QEMUSnapshotInfo *sn_info)
         goto cleanup;
     }
 
-    ret = do_sd_create(s, s->name, s->inode.vdi_size, s->inode.vdi_id, &new_vid,
-                       1, s->inode.copy_policy);
+    ret = do_sd_create(s, &new_vid, 1);
     if (ret < 0) {
         error_report("failed to create inode for snapshot. %s",
                      strerror(errno));
@@ -2407,6 +2470,22 @@ sd_co_get_block_status(BlockDriverState *bs, int64_t sector_num, int nb_sectors,
     return ret;
 }
 
+static int64_t sd_get_allocated_file_size(BlockDriverState *bs)
+{
+    BDRVSheepdogState *s = bs->opaque;
+    SheepdogInode *inode = &s->inode;
+    unsigned long i, last = DIV_ROUND_UP(inode->vdi_size, SD_DATA_OBJ_SIZE);
+    uint64_t size = 0;
+
+    for (i = 0; i < last; i++) {
+        if (inode->data_vdi_id[i] == 0) {
+            continue;
+        }
+        size += SD_DATA_OBJ_SIZE;
+    }
+    return size;
+}
+
 static QEMUOptionParameter sd_create_options[] = {
     {
         .name = BLOCK_OPT_SIZE,
@@ -2423,6 +2502,11 @@ static QEMUOptionParameter sd_create_options[] = {
         .type = OPT_STRING,
         .help = "Preallocation mode (allowed values: off, full)"
     },
+    {
+        .name = BLOCK_OPT_REDUNDANCY,
+        .type = OPT_STRING,
+        .help = "Redundancy of the image"
+    },
     { NULL }
 };
 
@@ -2436,6 +2520,7 @@ static BlockDriver bdrv_sheepdog = {
     .bdrv_create    = sd_create,
     .bdrv_has_zero_init = bdrv_has_zero_init_1,
     .bdrv_getlength = sd_getlength,
+    .bdrv_get_allocated_file_size = sd_get_allocated_file_size,
     .bdrv_truncate  = sd_truncate,
 
     .bdrv_co_readv  = sd_co_readv,
@@ -2465,6 +2550,7 @@ static BlockDriver bdrv_sheepdog_tcp = {
     .bdrv_create    = sd_create,
     .bdrv_has_zero_init = bdrv_has_zero_init_1,
     .bdrv_getlength = sd_getlength,
+    .bdrv_get_allocated_file_size = sd_get_allocated_file_size,
     .bdrv_truncate  = sd_truncate,
 
     .bdrv_co_readv  = sd_co_readv,
@@ -2494,6 +2580,7 @@ static BlockDriver bdrv_sheepdog_unix = {
     .bdrv_create    = sd_create,
     .bdrv_has_zero_init = bdrv_has_zero_init_1,
     .bdrv_getlength = sd_getlength,
+    .bdrv_get_allocated_file_size = sd_get_allocated_file_size,
     .bdrv_truncate  = sd_truncate,
 
     .bdrv_co_readv  = sd_co_readv,

block/snapshot.c

@@ -25,6 +25,24 @@
 #include "block/snapshot.h"
 #include "block/block_int.h"
 
+QemuOptsList internal_snapshot_opts = {
+    .name = "snapshot",
+    .head = QTAILQ_HEAD_INITIALIZER(internal_snapshot_opts.head),
+    .desc = {
+        {
+            .name = SNAPSHOT_OPT_ID,
+            .type = QEMU_OPT_STRING,
+            .help = "snapshot id"
+        },{
+            .name = SNAPSHOT_OPT_NAME,
+            .type = QEMU_OPT_STRING,
+            .help = "snapshot name"
+        },{
+            /* end of list */
+        }
+    },
+};
+
 int bdrv_snapshot_find(BlockDriverState *bs, QEMUSnapshotInfo *sn_info,
                        const char *name)
 {
@@ -194,7 +212,7 @@ int bdrv_snapshot_goto(BlockDriverState *bs,
  * If only @snapshot_id is specified, delete the first one with id
  * @snapshot_id.
  * If only @name is specified, delete the first one with name @name.
- * if none is specified, return -ENINVAL.
+ * if none is specified, return -EINVAL.
  *
  * Returns: 0 on success, -errno on failure. If @bs is not inserted, return
  * -ENOMEDIUM. If @snapshot_id and @name are both NULL, return -EINVAL. If @bs
@@ -265,18 +283,71 @@ int bdrv_snapshot_list(BlockDriverState *bs,
     return -ENOTSUP;
 }
 
+/**
+ * Temporarily load an internal snapshot by @snapshot_id and @name.
+ * @bs: block device used in the operation
+ * @snapshot_id: unique snapshot ID, or NULL
+ * @name: snapshot name, or NULL
+ * @errp: location to store error
+ *
+ * If both @snapshot_id and @name are specified, load the first one with
+ * id @snapshot_id and name @name.
+ * If only @snapshot_id is specified, load the first one with id
+ * @snapshot_id.
+ * If only @name is specified, load the first one with name @name.
+ * if none is specified, return -EINVAL.
+ *
+ * Returns: 0 on success, -errno on fail. If @bs is not inserted, return
+ * -ENOMEDIUM. If @bs is not readonly, return -EINVAL. If @bs did not support
+ * internal snapshot, return -ENOTSUP. If qemu can't find a matching @id and
+ * @name, return -ENOENT. If @errp != NULL, it will always be filled on
+ * failure.
+ */
 int bdrv_snapshot_load_tmp(BlockDriverState *bs,
-        const char *snapshot_name)
+                           const char *snapshot_id,
+                           const char *name,
+                           Error **errp)
 {
     BlockDriver *drv = bs->drv;
+
     if (!drv) {
+        error_set(errp, QERR_DEVICE_HAS_NO_MEDIUM, bdrv_get_device_name(bs));
         return -ENOMEDIUM;
     }
+    if (!snapshot_id && !name) {
+        error_setg(errp, "snapshot_id and name are both NULL");
+        return -EINVAL;
+    }
     if (!bs->read_only) {
+        error_setg(errp, "Device is not readonly");
         return -EINVAL;
     }
     if (drv->bdrv_snapshot_load_tmp) {
-        return drv->bdrv_snapshot_load_tmp(bs, snapshot_name);
+        return drv->bdrv_snapshot_load_tmp(bs, snapshot_id, name, errp);
     }
+    error_set(errp, QERR_BLOCK_FORMAT_FEATURE_NOT_SUPPORTED,
+              drv->format_name, bdrv_get_device_name(bs),
+              "temporarily load internal snapshot");
     return -ENOTSUP;
 }
+
+int bdrv_snapshot_load_tmp_by_id_or_name(BlockDriverState *bs,
+                                         const char *id_or_name,
+                                         Error **errp)
+{
+    int ret;
+    Error *local_err = NULL;
+
+    ret = bdrv_snapshot_load_tmp(bs, id_or_name, NULL, &local_err);
+    if (ret == -ENOENT || ret == -EINVAL) {
+        error_free(local_err);
+        local_err = NULL;
+        ret = bdrv_snapshot_load_tmp(bs, NULL, id_or_name, &local_err);
+    }
+
+    if (error_is_set(&local_err)) {
+        error_propagate(errp, local_err);
+    }
+
+    return ret;
+}

block/stream.c

@@ -88,6 +88,11 @@ static void coroutine_fn stream_run(void *opaque)
     int n = 0;
     void *buf;
 
+    if (!bs->backing_hd) {
+        block_job_completed(&s->common, 0);
+        return;
+    }
+
     s->common.len = bdrv_getlength(bs);
     if (s->common.len < 0) {
         block_job_completed(&s->common, s->common.len);

block/vdi.c

@@ -331,6 +331,7 @@ static int vdi_get_info(BlockDriverState *bs, BlockDriverInfo *bdi)
     logout("\n");
     bdi->cluster_size = s->block_size;
     bdi->vm_state_offset = 0;
+    bdi->unallocated_blocks_are_zero = true;
     return 0;
 }
 

block/vhdx-log.c

@@ -706,7 +706,8 @@ exit:
  *
  * If read-only, we must replay the log in RAM (or refuse to open
  * a dirty VHDX file read-only) */
-int vhdx_parse_log(BlockDriverState *bs, BDRVVHDXState *s, bool *flushed)
+int vhdx_parse_log(BlockDriverState *bs, BDRVVHDXState *s, bool *flushed,
+                   Error **errp)
 {
     int ret = 0;
     VHDXHeader *hdr;
@@ -761,6 +762,16 @@ int vhdx_parse_log(BlockDriverState *bs, BDRVVHDXState *s, bool *flushed)
     }
 
     if (logs.valid) {
+        if (bs->read_only) {
+            ret = -EPERM;
+            error_setg_errno(errp, EPERM,
+                             "VHDX image file '%s' opened read-only, but "
+                             "contains a log that needs to be replayed.  To "
+                             "replay the log, execute:\n qemu-img check -r "
+                             "all '%s'",
+                             bs->filename, bs->filename);
+            goto exit;
+        }
         /* now flush the log */
         ret = vhdx_log_flush(bs, s, &logs);
         if (ret < 0) {

block/vhdx.c

@@ -878,7 +878,6 @@ static int vhdx_open(BlockDriverState *bs, QDict *options, int flags,
     int ret = 0;
     uint32_t i;
     uint64_t signature;
-    bool log_flushed = false;
 
 
     s->bat = NULL;
@@ -907,7 +906,7 @@ static int vhdx_open(BlockDriverState *bs, QDict *options, int flags,
         goto fail;
     }
 
-    ret = vhdx_parse_log(bs, s, &log_flushed);
+    ret = vhdx_parse_log(bs, s, &s->log_replayed_on_open, errp);
     if (ret < 0) {
         goto fail;
     }
@@ -1043,6 +1042,18 @@ static void vhdx_block_translate(BDRVVHDXState *s, int64_t sector_num,
 }
 
 
+static int vhdx_get_info(BlockDriverState *bs, BlockDriverInfo *bdi)
+{
+    BDRVVHDXState *s = bs->opaque;
+
+    bdi->cluster_size = s->block_size;
+
+    bdi->unallocated_blocks_are_zero =
+        (s->params.data_bits & VHDX_PARAMS_HAS_PARENT) == 0;
+
+    return 0;
+}
+
 
 static coroutine_fn int vhdx_co_readv(BlockDriverState *bs, int64_t sector_num,
                                       int nb_sectors, QEMUIOVector *qiov)
@@ -1842,6 +1853,24 @@ exit:
     return ret;
 }
 
+/* If opened r/w, the VHDX driver will automatically replay the log,
+ * if one is present, inside the vhdx_open() call.
+ *
+ * If qemu-img check -r all is called, the image is automatically opened
+ * r/w and any log has already been replayed, so there is nothing (currently)
+ * for us to do here
+ */
+static int vhdx_check(BlockDriverState *bs, BdrvCheckResult *result,
+                       BdrvCheckMode fix)
+{
+    BDRVVHDXState *s = bs->opaque;
+
+    if (s->log_replayed_on_open) {
+        result->corruptions_fixed++;
+    }
+    return 0;
+}
+
 static QEMUOptionParameter vhdx_create_options[] = {
     {
         .name = BLOCK_OPT_SIZE,
@@ -1885,6 +1914,8 @@ static BlockDriver bdrv_vhdx = {
     .bdrv_co_readv          = vhdx_co_readv,
     .bdrv_co_writev         = vhdx_co_writev,
     .bdrv_create            = vhdx_create,
+    .bdrv_get_info          = vhdx_get_info,
+    .bdrv_check             = vhdx_check,
 
     .create_options         = vhdx_create_options,
 };

block/vhdx.h

@@ -394,6 +394,8 @@ typedef struct BDRVVHDXState {
 
     Error *migration_blocker;
 
+    bool log_replayed_on_open;
+
     QLIST_HEAD(VHDXRegionHead, VHDXRegionEntry) regions;
 } BDRVVHDXState;
 
@@ -408,7 +410,8 @@ uint32_t vhdx_checksum_calc(uint32_t crc, uint8_t *buf, size_t size,
 
 bool vhdx_checksum_is_valid(uint8_t *buf, size_t size, int crc_offset);
 
-int vhdx_parse_log(BlockDriverState *bs, BDRVVHDXState *s, bool *flushed);
+int vhdx_parse_log(BlockDriverState *bs, BDRVVHDXState *s, bool *flushed,
+                   Error **errp);
 
 int vhdx_log_write_and_flush(BlockDriverState *bs, BDRVVHDXState *s,
                              void *data, uint32_t length, uint64_t offset);

block/vmdk.c

@@ -428,6 +428,10 @@ static int vmdk_add_extent(BlockDriverState *bs,
     extent->l2_size = l2_size;
     extent->cluster_sectors = flat ? sectors : cluster_sectors;
 
+    if (!flat) {
+        bs->bl.write_zeroes_alignment =
+            MAX(bs->bl.write_zeroes_alignment, cluster_sectors);
+    }
     if (s->num_extents > 1) {
         extent->end_sector = (*(extent - 1)).end_sector + extent->sectors;
     } else {
@@ -605,13 +609,20 @@ static int vmdk_open_vmdk4(BlockDriverState *bs,
         header = footer.header;
     }
 
-    if (le32_to_cpu(header.version) >= 3) {
+    if (le32_to_cpu(header.version) > 3) {
         char buf[64];
         snprintf(buf, sizeof(buf), "VMDK version %d",
                  le32_to_cpu(header.version));
         qerror_report(QERR_UNKNOWN_BLOCK_FORMAT_FEATURE,
                 bs->device_name, "vmdk", buf);
         return -ENOTSUP;
+    } else if (le32_to_cpu(header.version) == 3 && (flags & BDRV_O_RDWR)) {
+        /* VMware KB 2064959 explains that version 3 added support for
+         * persistent changed block tracking (CBT), and backup software can
+         * read it as version=1 if it doesn't care about the changed area
+         * information. So we are safe to enable read only. */
+        error_setg(errp, "VMDK version 3 must be read only");
+        return -EINVAL;
     }
 
     if (le32_to_cpu(header.num_gtes_per_gt) > 512) {
@@ -738,9 +749,14 @@ static int vmdk_parse_extents(const char *desc, BlockDriverState *bs,
                 return -EINVAL;
             }
         } else if (!strcmp(type, "VMFS")) {
-            flat_offset = 0;
+            if (ret == 4) {
+                flat_offset = 0;
+            } else {
+                error_setg(errp, "Invalid extent lines:\n%s", p);
+                return -EINVAL;
+            }
         } else if (ret != 4) {
-            error_setg(errp, "Invalid extent lines: \n%s", p);
+            error_setg(errp, "Invalid extent lines:\n%s", p);
             return -EINVAL;
         }
 
@@ -1419,7 +1435,8 @@ static coroutine_fn int vmdk_co_write(BlockDriverState *bs, int64_t sector_num,
 
 static int coroutine_fn vmdk_co_write_zeroes(BlockDriverState *bs,
                                              int64_t sector_num,
-                                             int nb_sectors)
+                                             int nb_sectors,
+                                             BdrvRequestFlags flags)
 {
     int ret;
     BDRVVmdkState *s = bs->opaque;
@@ -1435,23 +1452,33 @@ static int coroutine_fn vmdk_co_write_zeroes(BlockDriverState *bs,
 }
 
 static int vmdk_create_extent(const char *filename, int64_t filesize,
-                              bool flat, bool compress, bool zeroed_grain)
+                              bool flat, bool compress, bool zeroed_grain,
+                              Error **errp)
 {
     int ret, i;
-    int fd = 0;
+    BlockDriverState *bs = NULL;
     VMDK4Header header;
-    uint32_t tmp, magic, grains, gd_size, gt_size, gt_count;
+    Error *local_err;
+    uint32_t tmp, magic, grains, gd_sectors, gt_size, gt_count;
+    uint32_t *gd_buf = NULL;
+    int gd_buf_size;
 
-    fd = qemu_open(filename,
-                   O_WRONLY | O_CREAT | O_TRUNC | O_BINARY | O_LARGEFILE,
-                   0644);
-    if (fd < 0) {
-        return -errno;
+    ret = bdrv_create_file(filename, NULL, &local_err);
+    if (ret < 0) {
+        error_propagate(errp, local_err);
+        goto exit;
     }
+
+    ret = bdrv_file_open(&bs, filename, NULL, BDRV_O_RDWR, &local_err);
+    if (ret < 0) {
+        error_propagate(errp, local_err);
+        goto exit;
+    }
+
     if (flat) {
-        ret = ftruncate(fd, filesize);
+        ret = bdrv_truncate(bs, filesize);
         if (ret < 0) {
-            ret = -errno;
+            error_setg(errp, "Could not truncate file");
         }
         goto exit;
     }
@@ -1462,24 +1489,23 @@ static int vmdk_create_extent(const char *filename, int64_t filesize,
                    | (compress ? VMDK4_FLAG_COMPRESS | VMDK4_FLAG_MARKER : 0)
                    | (zeroed_grain ? VMDK4_FLAG_ZERO_GRAIN : 0);
     header.compressAlgorithm = compress ? VMDK4_COMPRESSION_DEFLATE : 0;
-    header.capacity = filesize / 512;
+    header.capacity = filesize / BDRV_SECTOR_SIZE;
     header.granularity = 128;
-    header.num_gtes_per_gt = 512;
+    header.num_gtes_per_gt = BDRV_SECTOR_SIZE;
 
-    grains = (filesize / 512 + header.granularity - 1) / header.granularity;
-    gt_size = ((header.num_gtes_per_gt * sizeof(uint32_t)) + 511) >> 9;
-    gt_count =
-        (grains + header.num_gtes_per_gt - 1) / header.num_gtes_per_gt;
-    gd_size = (gt_count * sizeof(uint32_t) + 511) >> 9;
+    grains = DIV_ROUND_UP(filesize / BDRV_SECTOR_SIZE, header.granularity);
+    gt_size = DIV_ROUND_UP(header.num_gtes_per_gt * sizeof(uint32_t),
+                           BDRV_SECTOR_SIZE);
+    gt_count = DIV_ROUND_UP(grains, header.num_gtes_per_gt);
+    gd_sectors = DIV_ROUND_UP(gt_count * sizeof(uint32_t), BDRV_SECTOR_SIZE);
 
     header.desc_offset = 1;
     header.desc_size = 20;
     header.rgd_offset = header.desc_offset + header.desc_size;
-    header.gd_offset = header.rgd_offset + gd_size + (gt_size * gt_count);
+    header.gd_offset = header.rgd_offset + gd_sectors + (gt_size * gt_count);
     header.grain_offset =
-       ((header.gd_offset + gd_size + (gt_size * gt_count) +
-         header.granularity - 1) / header.granularity) *
-        header.granularity;
+        ROUND_UP(header.gd_offset + gd_sectors + (gt_size * gt_count),
+                 header.granularity);
     /* swap endianness for all header fields */
     header.version = cpu_to_le32(header.version);
     header.flags = cpu_to_le32(header.flags);
@@ -1499,48 +1525,55 @@ static int vmdk_create_extent(const char *filename, int64_t filesize,
     header.check_bytes[3] = 0xa;
 
     /* write all the data */
-    ret = qemu_write_full(fd, &magic, sizeof(magic));
-    if (ret != sizeof(magic)) {
-        ret = -errno;
+    ret = bdrv_pwrite(bs, 0, &magic, sizeof(magic));
+    if (ret < 0) {
+        error_set(errp, QERR_IO_ERROR);
         goto exit;
     }
-    ret = qemu_write_full(fd, &header, sizeof(header));
-    if (ret != sizeof(header)) {
-        ret = -errno;
+    ret = bdrv_pwrite(bs, sizeof(magic), &header, sizeof(header));
+    if (ret < 0) {
+        error_set(errp, QERR_IO_ERROR);
         goto exit;
     }
 
-    ret = ftruncate(fd, le64_to_cpu(header.grain_offset) << 9);
+    ret = bdrv_truncate(bs, le64_to_cpu(header.grain_offset) << 9);
     if (ret < 0) {
-        ret = -errno;
+        error_setg(errp, "Could not truncate file");
         goto exit;
     }
 
     /* write grain directory */
-    lseek(fd, le64_to_cpu(header.rgd_offset) << 9, SEEK_SET);
-    for (i = 0, tmp = le64_to_cpu(header.rgd_offset) + gd_size;
+    gd_buf_size = gd_sectors * BDRV_SECTOR_SIZE;
+    gd_buf = g_malloc0(gd_buf_size);
+    for (i = 0, tmp = le64_to_cpu(header.rgd_offset) + gd_sectors;
          i < gt_count; i++, tmp += gt_size) {
-        ret = qemu_write_full(fd, &tmp, sizeof(tmp));
-        if (ret != sizeof(tmp)) {
-            ret = -errno;
-            goto exit;
-        }
+        gd_buf[i] = cpu_to_le32(tmp);
+    }
+    ret = bdrv_pwrite(bs, le64_to_cpu(header.rgd_offset) * BDRV_SECTOR_SIZE,
+                      gd_buf, gd_buf_size);
+    if (ret < 0) {
+        error_set(errp, QERR_IO_ERROR);
+        goto exit;
     }
 
     /* write backup grain directory */
-    lseek(fd, le64_to_cpu(header.gd_offset) << 9, SEEK_SET);
-    for (i = 0, tmp = le64_to_cpu(header.gd_offset) + gd_size;
+    for (i = 0, tmp = le64_to_cpu(header.gd_offset) + gd_sectors;
          i < gt_count; i++, tmp += gt_size) {
-        ret = qemu_write_full(fd, &tmp, sizeof(tmp));
-        if (ret != sizeof(tmp)) {
-            ret = -errno;
-            goto exit;
-        }
+        gd_buf[i] = cpu_to_le32(tmp);
+    }
+    ret = bdrv_pwrite(bs, le64_to_cpu(header.gd_offset) * BDRV_SECTOR_SIZE,
+                      gd_buf, gd_buf_size);
+    if (ret < 0) {
+        error_set(errp, QERR_IO_ERROR);
+        goto exit;
     }
 
     ret = 0;
- exit:
-    qemu_close(fd);
+exit:
+    if (bs) {
+        bdrv_unref(bs);
+    }
+    g_free(gd_buf);
     return ret;
 }
 
@@ -1587,8 +1620,10 @@ static int filename_decompose(const char *filename, char *path, char *prefix,
 static int vmdk_create(const char *filename, QEMUOptionParameter *options,
                        Error **errp)
 {
-    int fd, idx = 0;
-    char desc[BUF_SIZE];
+    int idx = 0;
+    BlockDriverState *new_bs = NULL;
+    Error *local_err;
+    char *desc = NULL;
     int64_t total_size = 0, filesize;
     const char *adapter_type = NULL;
     const char *backing_file = NULL;
@@ -1596,7 +1631,7 @@ static int vmdk_create(const char *filename, QEMUOptionParameter *options,
     int flags = 0;
     int ret = 0;
     bool flat, split, compress;
-    char ext_desc_lines[BUF_SIZE] = "";
+    GString *ext_desc_lines;
     char path[PATH_MAX], prefix[PATH_MAX], postfix[PATH_MAX];
     const int64_t split_size = 0x80000000;  /* VMDK has constant split size */
     const char *desc_extent_line;
@@ -1604,6 +1639,7 @@ static int vmdk_create(const char *filename, QEMUOptionParameter *options,
     uint32_t parent_cid = 0xffffffff;
     uint32_t number_heads = 16;
     bool zeroed_grain = false;
+    uint32_t desc_offset = 0, desc_len;
     const char desc_template[] =
         "# Disk DescriptorFile\n"
         "version=1\n"
@@ -1624,8 +1660,11 @@ static int vmdk_create(const char *filename, QEMUOptionParameter *options,
         "ddb.geometry.sectors = \"63\"\n"
         "ddb.adapterType = \"%s\"\n";
 
+    ext_desc_lines = g_string_new(NULL);
+
     if (filename_decompose(filename, path, prefix, postfix, PATH_MAX, errp)) {
-        return -EINVAL;
+        ret = -EINVAL;
+        goto exit;
     }
     /* Read out options */
     while (options && options->name) {
@@ -1651,7 +1690,8 @@ static int vmdk_create(const char *filename, QEMUOptionParameter *options,
                strcmp(adapter_type, "lsilogic") &&
                strcmp(adapter_type, "legacyESX")) {
         error_setg(errp, "Unknown adapter type: '%s'", adapter_type);
-        return -EINVAL;
+        ret = -EINVAL;
+        goto exit;
     }
     if (strcmp(adapter_type, "ide") != 0) {
         /* that's the number of heads with which vmware operates when
@@ -1667,7 +1707,8 @@ static int vmdk_create(const char *filename, QEMUOptionParameter *options,
                strcmp(fmt, "twoGbMaxExtentFlat") &&
                strcmp(fmt, "streamOptimized")) {
         error_setg(errp, "Unknown subformat: '%s'", fmt);
-        return -EINVAL;
+        ret = -EINVAL;
+        goto exit;
     }
     split = !(strcmp(fmt, "twoGbMaxExtentFlat") &&
               strcmp(fmt, "twoGbMaxExtentSparse"));
@@ -1681,22 +1722,25 @@ static int vmdk_create(const char *filename, QEMUOptionParameter *options,
     }
     if (flat && backing_file) {
         error_setg(errp, "Flat image can't have backing file");
-        return -ENOTSUP;
+        ret = -ENOTSUP;
+        goto exit;
     }
     if (flat && zeroed_grain) {
         error_setg(errp, "Flat image can't enable zeroed grain");
-        return -ENOTSUP;
+        ret = -ENOTSUP;
+        goto exit;
     }
     if (backing_file) {
         BlockDriverState *bs = bdrv_new("");
-        ret = bdrv_open(bs, backing_file, NULL, 0, NULL, errp);
+        ret = bdrv_open(bs, backing_file, NULL, BDRV_O_NO_BACKING, NULL, errp);
         if (ret != 0) {
             bdrv_unref(bs);
-            return ret;
+            goto exit;
         }
         if (strcmp(bs->drv->format_name, "vmdk")) {
             bdrv_unref(bs);
-            return -EINVAL;
+            ret = -EINVAL;
+            goto exit;
         }
         parent_cid = vmdk_read_cid(bs, 0);
         bdrv_unref(bs);
@@ -1729,51 +1773,64 @@ static int vmdk_create(const char *filename, QEMUOptionParameter *options,
                 path, desc_filename);
 
         if (vmdk_create_extent(ext_filename, size,
-                               flat, compress, zeroed_grain)) {
-            return -EINVAL;
+                               flat, compress, zeroed_grain, errp)) {
+            ret = -EINVAL;
+            goto exit;
         }
         filesize -= size;
 
         /* Format description line */
         snprintf(desc_line, sizeof(desc_line),
-                    desc_extent_line, size / 512, desc_filename);
-        pstrcat(ext_desc_lines, sizeof(ext_desc_lines), desc_line);
+                    desc_extent_line, size / BDRV_SECTOR_SIZE, desc_filename);
+        g_string_append(ext_desc_lines, desc_line);
     }
     /* generate descriptor file */
-    snprintf(desc, sizeof(desc), desc_template,
-            (unsigned int)time(NULL),
-            parent_cid,
-            fmt,
-            parent_desc_line,
-            ext_desc_lines,
-            (flags & BLOCK_FLAG_COMPAT6 ? 6 : 4),
-            total_size / (int64_t)(63 * number_heads * 512), number_heads,
-                adapter_type);
-    if (split || flat) {
-        fd = qemu_open(filename,
-                       O_WRONLY | O_CREAT | O_TRUNC | O_BINARY | O_LARGEFILE,
-                       0644);
-    } else {
-        fd = qemu_open(filename,
-                       O_WRONLY | O_BINARY | O_LARGEFILE,
-                       0644);
-    }
-    if (fd < 0) {
-        return -errno;
-    }
+    desc = g_strdup_printf(desc_template,
+                           (unsigned int)time(NULL),
+                           parent_cid,
+                           fmt,
+                           parent_desc_line,
+                           ext_desc_lines->str,
+                           (flags & BLOCK_FLAG_COMPAT6 ? 6 : 4),
+                           total_size /
+                               (int64_t)(63 * number_heads * BDRV_SECTOR_SIZE),
+                           number_heads,
+                           adapter_type);
+    desc_len = strlen(desc);
     /* the descriptor offset = 0x200 */
-    if (!split && !flat && 0x200 != lseek(fd, 0x200, SEEK_SET)) {
-        ret = -errno;
+    if (!split && !flat) {
+        desc_offset = 0x200;
+    } else {
+        ret = bdrv_create_file(filename, options, &local_err);
+        if (ret < 0) {
+            error_setg_errno(errp, -ret, "Could not create image file");
+            goto exit;
+        }
+    }
+    ret = bdrv_file_open(&new_bs, filename, NULL, BDRV_O_RDWR, &local_err);
+    if (ret < 0) {
+        error_setg_errno(errp, -ret, "Could not write description");
         goto exit;
     }
-    ret = qemu_write_full(fd, desc, strlen(desc));
-    if (ret != strlen(desc)) {
-        ret = -errno;
+    ret = bdrv_pwrite(new_bs, desc_offset, desc, desc_len);
+    if (ret < 0) {
+        error_setg_errno(errp, -ret, "Could not write description");
         goto exit;
     }
-    ret = 0;
+    /* bdrv_pwrite write padding zeros to align to sector, we don't need that
+     * for description file */
+    if (desc_offset == 0) {
+        ret = bdrv_truncate(new_bs, desc_len);
+        if (ret < 0) {
+            error_setg(errp, "Could not truncate file");
+        }
+    }
 exit:
-    qemu_close(fd);
+    if (new_bs) {
+        bdrv_unref(new_bs);
+    }
+    g_free(desc);
+    g_string_free(ext_desc_lines, true);
     return ret;
 }
 

block/vpc.c

@@ -455,6 +455,19 @@ fail:
     return -1;
 }
 
+static int vpc_get_info(BlockDriverState *bs, BlockDriverInfo *bdi)
+{
+    BDRVVPCState *s = (BDRVVPCState *)bs->opaque;
+    VHDFooter *footer = (VHDFooter *) s->footer_buf;
+
+    if (cpu_to_be32(footer->type) != VHD_FIXED) {
+        bdi->cluster_size = s->block_size;
+    }
+
+    bdi->unallocated_blocks_are_zero = true;
+    return 0;
+}
+
 static int vpc_read(BlockDriverState *bs, int64_t sector_num,
                     uint8_t *buf, int nb_sectors)
 {
@@ -857,6 +870,8 @@ static BlockDriver bdrv_vpc = {
     .bdrv_read              = vpc_co_read,
     .bdrv_write             = vpc_co_write,
 
+    .bdrv_get_info          = vpc_get_info,
+
     .create_options         = vpc_create_options,
     .bdrv_has_zero_init     = vpc_has_zero_init,
 };

block/vvfat.c

@@ -266,8 +266,7 @@ typedef struct mbr_t {
 } QEMU_PACKED mbr_t;
 
 typedef struct direntry_t {
-    uint8_t name[8];
-    uint8_t extension[3];
+    uint8_t name[8 + 3];
     uint8_t attributes;
     uint8_t reserved[2];
     uint16_t ctime;
@@ -518,11 +517,9 @@ static inline uint8_t fat_chksum(const direntry_t* entry)
     uint8_t chksum=0;
     int i;
 
-    for(i=0;i<11;i++) {
-        unsigned char c;
-
-        c = (i < 8) ? entry->name[i] : entry->extension[i-8];
-        chksum=(((chksum&0xfe)>>1)|((chksum&0x01)?0x80:0)) + c;
+    for (i = 0; i < ARRAY_SIZE(entry->name); i++) {
+        chksum = (((chksum & 0xfe) >> 1) |
+                  ((chksum & 0x01) ? 0x80 : 0)) + entry->name[i];
     }
 
     return chksum;
@@ -617,7 +614,7 @@ static inline direntry_t* create_short_and_long_name(BDRVVVFATState* s,
 
     if(is_dot) {
 	entry=array_get_next(&(s->directory));
-	memset(entry->name,0x20,11);
+        memset(entry->name, 0x20, sizeof(entry->name));
 	memcpy(entry->name,filename,strlen(filename));
 	return entry;
     }
@@ -632,12 +629,14 @@ static inline direntry_t* create_short_and_long_name(BDRVVVFATState* s,
 	i = 8;
 
     entry=array_get_next(&(s->directory));
-    memset(entry->name,0x20,11);
+    memset(entry->name, 0x20, sizeof(entry->name));
     memcpy(entry->name, filename, i);
 
-    if(j > 0)
-	for (i = 0; i < 3 && filename[j+1+i]; i++)
-	    entry->extension[i] = filename[j+1+i];
+    if (j > 0) {
+        for (i = 0; i < 3 && filename[j + 1 + i]; i++) {
+            entry->name[8 + i] = filename[j + 1 + i];
+        }
+    }
 
     /* upcase & remove unwanted characters */
     for(i=10;i>=0;i--) {
@@ -861,8 +860,7 @@ static int init_directories(BDRVVVFATState* s,
     {
 	direntry_t* entry=array_get_next(&(s->directory));
 	entry->attributes=0x28; /* archive | volume label */
-	memcpy(entry->name,"QEMU VVF",8);
-	memcpy(entry->extension,"AT ",3);
+        memcpy(entry->name, "QEMU VVFAT ", sizeof(entry->name));
     }
 
     /* Now build FAT, and write back information into directory */
@@ -1085,7 +1083,7 @@ DLOG(if (stderr == NULL) {
     setbuf(stderr, NULL);
 })
 
-    opts = qemu_opts_create_nofail(&runtime_opts);
+    opts = qemu_opts_create(&runtime_opts, NULL, 0, &error_abort);
     qemu_opts_absorb_qdict(opts, options, &local_err);
     if (error_is_set(&local_err)) {
         qerror_report_err(local_err);
@@ -1591,17 +1589,20 @@ static int parse_short_name(BDRVVVFATState* s,
 	    lfn->name[i] = direntry->name[i];
     }
 
-    for (j = 2; j >= 0 && direntry->extension[j] == ' '; j--);
+    for (j = 2; j >= 0 && direntry->name[8 + j] == ' '; j--) {
+    }
     if (j >= 0) {
 	lfn->name[i++] = '.';
 	lfn->name[i + j + 1] = '\0';
 	for (;j >= 0; j--) {
-	    if (direntry->extension[j] <= ' ' || direntry->extension[j] > 0x7f)
-		return -2;
-	    else if (s->downcase_short_names)
-		lfn->name[i + j] = qemu_tolower(direntry->extension[j]);
-	    else
-		lfn->name[i + j] = direntry->extension[j];
+            uint8_t c = direntry->name[8 + j];
+            if (c <= ' ' || c > 0x7f) {
+                return -2;
+            } else if (s->downcase_short_names) {
+                lfn->name[i + j] = qemu_tolower(c);
+            } else {
+                lfn->name[i + j] = c;
+            }
 	}
     } else
 	lfn->name[i + j + 1] = '\0';

blockdev.c

@@ -682,7 +682,8 @@ DriveInfo *drive_init(QemuOpts *all_opts, BlockInterfaceType block_default_type)
     bs_opts = qdict_new();
     qemu_opts_to_qdict(all_opts, bs_opts);
 
-    legacy_opts = qemu_opts_create_nofail(&qemu_legacy_drive_opts);
+    legacy_opts = qemu_opts_create(&qemu_legacy_drive_opts, NULL, 0,
+                                   &error_abort);
     qemu_opts_absorb_qdict(legacy_opts, bs_opts, &local_err);
     if (error_is_set(&local_err)) {
         qerror_report_err(local_err);
@@ -853,7 +854,8 @@ DriveInfo *drive_init(QemuOpts *all_opts, BlockInterfaceType block_default_type)
 
     if (type == IF_VIRTIO) {
         QemuOpts *devopts;
-        devopts = qemu_opts_create_nofail(qemu_find_opts("device"));
+        devopts = qemu_opts_create(qemu_find_opts("device"), NULL, 0,
+                                   &error_abort);
         if (arch_type == QEMU_ARCH_S390X) {
             qemu_opt_set(devopts, "driver", "virtio-blk-s390");
         } else {
@@ -1523,7 +1525,7 @@ static void qmp_bdrv_open_encrypted(BlockDriverState *bs, const char *filename,
 }
 
 void qmp_change_blockdev(const char *device, const char *filename,
-                         bool has_format, const char *format, Error **errp)
+                         const char *format, Error **errp)
 {
     BlockDriverState *bs;
     BlockDriver *drv = NULL;
@@ -1820,8 +1822,13 @@ void qmp_block_commit(const char *device,
         return;
     }
 
-    commit_start(bs, base_bs, top_bs, speed, on_error, block_job_cb, bs,
-                &local_err);
+    if (top_bs == bs) {
+        commit_active_start(bs, base_bs, speed, on_error, block_job_cb,
+                            bs, &local_err);
+    } else {
+        commit_start(bs, base_bs, top_bs, speed, on_error, block_job_cb, bs,
+                    &local_err);
+    }
     if (local_err != NULL) {
         error_propagate(errp, local_err);
         return;

configure

@@ -169,6 +169,7 @@ curl=""
 curses=""
 docs=""
 fdt=""
+netmap="no"
 pixman=""
 sdl=""
 virtfs=""
@@ -488,6 +489,7 @@ FreeBSD)
   audio_possible_drivers="oss sdl esd pa"
   # needed for kinfo_getvmmap(3) in libutil.h
   LIBS="-lutil $LIBS"
+  netmap=""  # enable netmap autodetect
 ;;
 DragonFly)
   bsd="yes"
@@ -797,6 +799,10 @@ for opt do
   ;;
   --enable-vde) vde="yes"
   ;;
+  --disable-netmap) netmap="no"
+  ;;
+  --enable-netmap) netmap="yes"
+  ;;
   --disable-xen) xen="no"
   ;;
   --enable-xen) xen="yes"
@@ -998,6 +1004,25 @@ for opt do
   esac
 done
 
+if ! has $python; then
+  error_exit "Python not found. Use --python=/path/to/python"
+fi
+
+# Note that if the Python conditional here evaluates True we will exit
+# with status 1 which is a shell 'false' value.
+if ! $python -c 'import sys; sys.exit(sys.version_info < (2,4) or sys.version_info >= (3,))'; then
+  error_exit "Cannot use '$python', Python 2.4 or later is required." \
+      "Note that Python 3 or later is not yet supported." \
+      "Use --python=/path/to/python to specify a supported Python."
+fi
+
+# The -B switch was added in Python 2.6.
+# If it is supplied, compiled files are not written.
+# Use it for Python versions which support it.
+if $python -B -c 'import sys; sys.exit(0)' 2>/dev/null; then
+  python="$python -B"
+fi
+
 case "$cpu" in
     ppc)
            CPU_CFLAGS="-m32"
@@ -1068,167 +1093,169 @@ cat << EOF
 Usage: configure [options]
 Options: [defaults in brackets after descriptions]
 
+Standard options:
+  --help                   print this message
+  --prefix=PREFIX          install in PREFIX [$prefix]
+  --interp-prefix=PREFIX   where to find shared libraries, etc.
+                           use %M for cpu name [$interp_prefix]
+  --target-list=LIST       set target list (default: build everything)
+$(echo Available targets: $default_target_list | \
+  fold -s -w 53 | sed -e 's/^/                           /')
+
+Advanced options (experts only):
+  --source-path=PATH       path of source code [$source_path]
+  --cross-prefix=PREFIX    use PREFIX for compile tools [$cross_prefix]
+  --cc=CC                  use C compiler CC [$cc]
+  --iasl=IASL              use ACPI compiler IASL [$iasl]
+  --host-cc=CC             use C compiler CC [$host_cc] for code run at
+                           build time
+  --cxx=CXX                use C++ compiler CXX [$cxx]
+  --objcc=OBJCC            use Objective-C compiler OBJCC [$objcc]
+  --extra-cflags=CFLAGS    append extra C compiler flags QEMU_CFLAGS
+  --extra-ldflags=LDFLAGS  append extra linker flags LDFLAGS
+  --make=MAKE              use specified make [$make]
+  --install=INSTALL        use specified install [$install]
+  --python=PYTHON          use specified python [$python]
+  --smbd=SMBD              use specified smbd [$smbd]
+  --static                 enable static build [$static]
+  --mandir=PATH            install man pages in PATH
+  --datadir=PATH           install firmware in PATH$confsuffix
+  --docdir=PATH            install documentation in PATH$confsuffix
+  --bindir=PATH            install binaries in PATH
+  --libdir=PATH            install libraries in PATH
+  --sysconfdir=PATH        install config in PATH$confsuffix
+  --localstatedir=PATH     install local state in PATH (set at runtime on win32)
+  --with-confsuffix=SUFFIX suffix for QEMU data inside datadir and sysconfdir [$confsuffix]
+  --enable-debug-tcg       enable TCG debugging
+  --disable-debug-tcg      disable TCG debugging (default)
+  --enable-debug-info       enable debugging information (default)
+  --disable-debug-info      disable debugging information
+  --enable-debug           enable common debug build options
+  --enable-sparse          enable sparse checker
+  --disable-sparse         disable sparse checker (default)
+  --disable-strip          disable stripping binaries
+  --disable-werror         disable compilation abort on warning
+  --disable-sdl            disable SDL
+  --enable-sdl             enable SDL
+  --disable-gtk            disable gtk UI
+  --enable-gtk             enable gtk UI
+  --disable-virtfs         disable VirtFS
+  --enable-virtfs          enable VirtFS
+  --disable-vnc            disable VNC
+  --enable-vnc             enable VNC
+  --disable-cocoa          disable Cocoa (Mac OS X only)
+  --enable-cocoa           enable Cocoa (default on Mac OS X)
+  --audio-drv-list=LIST    set audio drivers list:
+                           Available drivers: $audio_possible_drivers
+  --block-drv-whitelist=L  Same as --block-drv-rw-whitelist=L
+  --block-drv-rw-whitelist=L
+                           set block driver read-write whitelist
+                           (affects only QEMU, not qemu-img)
+  --block-drv-ro-whitelist=L
+                           set block driver read-only whitelist
+                           (affects only QEMU, not qemu-img)
+  --disable-xen            disable xen backend driver support
+  --enable-xen             enable xen backend driver support
+  --disable-xen-pci-passthrough
+  --enable-xen-pci-passthrough
+  --disable-brlapi         disable BrlAPI
+  --enable-brlapi          enable BrlAPI
+  --disable-vnc-tls        disable TLS encryption for VNC server
+  --enable-vnc-tls         enable TLS encryption for VNC server
+  --disable-vnc-sasl       disable SASL encryption for VNC server
+  --enable-vnc-sasl        enable SASL encryption for VNC server
+  --disable-vnc-jpeg       disable JPEG lossy compression for VNC server
+  --enable-vnc-jpeg        enable JPEG lossy compression for VNC server
+  --disable-vnc-png        disable PNG compression for VNC server (default)
+  --enable-vnc-png         enable PNG compression for VNC server
+  --disable-vnc-ws         disable Websockets support for VNC server
+  --enable-vnc-ws          enable Websockets support for VNC server
+  --disable-curses         disable curses output
+  --enable-curses          enable curses output
+  --disable-curl           disable curl connectivity
+  --enable-curl            enable curl connectivity
+  --disable-fdt            disable fdt device tree
+  --enable-fdt             enable fdt device tree
+  --disable-bluez          disable bluez stack connectivity
+  --enable-bluez           enable bluez stack connectivity
+  --disable-slirp          disable SLIRP userspace network connectivity
+  --disable-kvm            disable KVM acceleration support
+  --enable-kvm             enable KVM acceleration support
+  --disable-rdma           disable RDMA-based migration support
+  --enable-rdma            enable RDMA-based migration support
+  --enable-tcg-interpreter enable TCG with bytecode interpreter (TCI)
+  --enable-system          enable all system emulation targets
+  --disable-system         disable all system emulation targets
+  --enable-user            enable supported user emulation targets
+  --disable-user           disable all user emulation targets
+  --enable-linux-user      enable all linux usermode emulation targets
+  --disable-linux-user     disable all linux usermode emulation targets
+  --enable-bsd-user        enable all BSD usermode emulation targets
+  --disable-bsd-user       disable all BSD usermode emulation targets
+  --enable-guest-base      enable GUEST_BASE support for usermode
+                           emulation targets
+  --disable-guest-base     disable GUEST_BASE support
+  --enable-pie             build Position Independent Executables
+  --disable-pie            do not build Position Independent Executables
+  --fmod-lib               path to FMOD library
+  --fmod-inc               path to FMOD includes
+  --oss-lib                path to OSS library
+  --enable-uname-release=R Return R for uname -r in usermode emulation
+  --cpu=CPU                Build for host CPU [$cpu]
+  --disable-uuid           disable uuid support
+  --enable-uuid            enable uuid support
+  --disable-vde            disable support for vde network
+  --enable-vde             enable support for vde network
+  --disable-netmap         disable support for netmap network
+  --enable-netmap          enable support for netmap network
+  --disable-linux-aio      disable Linux AIO support
+  --enable-linux-aio       enable Linux AIO support
+  --disable-cap-ng         disable libcap-ng support
+  --enable-cap-ng          enable libcap-ng support
+  --disable-attr           disables attr and xattr support
+  --enable-attr            enable attr and xattr support
+  --disable-blobs          disable installing provided firmware blobs
+  --enable-docs            enable documentation build
+  --disable-docs           disable documentation build
+  --disable-vhost-net      disable vhost-net acceleration support
+  --enable-vhost-net       enable vhost-net acceleration support
+  --enable-trace-backend=B Set trace backend
+                           Available backends: $($python $source_path/scripts/tracetool.py --list-backends)
+  --with-trace-file=NAME   Full PATH,NAME of file to store traces
+                           Default:trace-<pid>
+  --disable-spice          disable spice
+  --enable-spice           enable spice
+  --enable-rbd             enable building the rados block device (rbd)
+  --disable-libiscsi       disable iscsi support
+  --enable-libiscsi        enable iscsi support
+  --disable-smartcard-nss  disable smartcard nss support
+  --enable-smartcard-nss   enable smartcard nss support
+  --disable-libusb         disable libusb (for usb passthrough)
+  --enable-libusb          enable libusb (for usb passthrough)
+  --disable-usb-redir      disable usb network redirection support
+  --enable-usb-redir       enable usb network redirection support
+  --disable-guest-agent    disable building of the QEMU Guest Agent
+  --enable-guest-agent     enable building of the QEMU Guest Agent
+  --with-vss-sdk=SDK-path  enable Windows VSS support in QEMU Guest Agent
+  --with-win-sdk=SDK-path  path to Windows Platform SDK (to build VSS .tlb)
+  --disable-seccomp        disable seccomp support
+  --enable-seccomp         enables seccomp support
+  --with-coroutine=BACKEND coroutine backend. Supported options:
+                           gthread, ucontext, sigaltstack, windows
+  --disable-coroutine-pool disable coroutine freelist (worse performance)
+  --enable-coroutine-pool  enable coroutine freelist (better performance)
+  --enable-glusterfs       enable GlusterFS backend
+  --disable-glusterfs      disable GlusterFS backend
+  --enable-gcov            enable test coverage analysis with gcov
+  --gcov=GCOV              use specified gcov [$gcov_tool]
+  --enable-tpm             enable TPM support
+  --disable-libssh2        disable ssh block device support
+  --enable-libssh2         enable ssh block device support
+  --disable-vhdx           disables support for the Microsoft VHDX image format
+  --enable-vhdx            enable support for the Microsoft VHDX image format
+
+NOTE: The object files are built at the place where configure is launched
 EOF
-echo "Standard options:"
-echo "  --help                   print this message"
-echo "  --prefix=PREFIX          install in PREFIX [$prefix]"
-echo "  --interp-prefix=PREFIX   where to find shared libraries, etc."
-echo "                           use %M for cpu name [$interp_prefix]"
-echo "  --target-list=LIST       set target list (default: build everything)"
-echo "Available targets: $default_target_list" | \
-    fold -s -w 53 | sed -e 's/^/                           /'
-echo ""
-echo "Advanced options (experts only):"
-echo "  --source-path=PATH       path of source code [$source_path]"
-echo "  --cross-prefix=PREFIX    use PREFIX for compile tools [$cross_prefix]"
-echo "  --cc=CC                  use C compiler CC [$cc]"
-echo "  --iasl=IASL              use ACPI compiler IASL [$iasl]"
-echo "  --host-cc=CC             use C compiler CC [$host_cc] for code run at"
-echo "                           build time"
-echo "  --cxx=CXX                use C++ compiler CXX [$cxx]"
-echo "  --objcc=OBJCC            use Objective-C compiler OBJCC [$objcc]"
-echo "  --extra-cflags=CFLAGS    append extra C compiler flags QEMU_CFLAGS"
-echo "  --extra-ldflags=LDFLAGS  append extra linker flags LDFLAGS"
-echo "  --make=MAKE              use specified make [$make]"
-echo "  --install=INSTALL        use specified install [$install]"
-echo "  --python=PYTHON          use specified python [$python]"
-echo "  --smbd=SMBD              use specified smbd [$smbd]"
-echo "  --static                 enable static build [$static]"
-echo "  --mandir=PATH            install man pages in PATH"
-echo "  --datadir=PATH           install firmware in PATH$confsuffix"
-echo "  --docdir=PATH            install documentation in PATH$confsuffix"
-echo "  --bindir=PATH            install binaries in PATH"
-echo "  --libdir=PATH            install libraries in PATH"
-echo "  --sysconfdir=PATH        install config in PATH$confsuffix"
-echo "  --localstatedir=PATH     install local state in PATH (set at runtime on win32)"
-echo "  --with-confsuffix=SUFFIX suffix for QEMU data inside datadir and sysconfdir [$confsuffix]"
-echo "  --enable-debug-tcg       enable TCG debugging"
-echo "  --disable-debug-tcg      disable TCG debugging (default)"
-echo "  --enable-debug-info       enable debugging information (default)"
-echo "  --disable-debug-info      disable debugging information"
-echo "  --enable-debug           enable common debug build options"
-echo "  --enable-sparse          enable sparse checker"
-echo "  --disable-sparse         disable sparse checker (default)"
-echo "  --disable-strip          disable stripping binaries"
-echo "  --disable-werror         disable compilation abort on warning"
-echo "  --disable-sdl            disable SDL"
-echo "  --enable-sdl             enable SDL"
-echo "  --disable-gtk            disable gtk UI"
-echo "  --enable-gtk             enable gtk UI"
-echo "  --disable-virtfs         disable VirtFS"
-echo "  --enable-virtfs          enable VirtFS"
-echo "  --disable-vnc            disable VNC"
-echo "  --enable-vnc             enable VNC"
-echo "  --disable-cocoa          disable Cocoa (Mac OS X only)"
-echo "  --enable-cocoa           enable Cocoa (default on Mac OS X)"
-echo "  --audio-drv-list=LIST    set audio drivers list:"
-echo "                           Available drivers: $audio_possible_drivers"
-echo "  --block-drv-whitelist=L  Same as --block-drv-rw-whitelist=L"
-echo "  --block-drv-rw-whitelist=L"
-echo "                           set block driver read-write whitelist"
-echo "                           (affects only QEMU, not qemu-img)"
-echo "  --block-drv-ro-whitelist=L"
-echo "                           set block driver read-only whitelist"
-echo "                           (affects only QEMU, not qemu-img)"
-echo "  --disable-xen            disable xen backend driver support"
-echo "  --enable-xen             enable xen backend driver support"
-echo "  --disable-xen-pci-passthrough"
-echo "  --enable-xen-pci-passthrough"
-echo "  --disable-brlapi         disable BrlAPI"
-echo "  --enable-brlapi          enable BrlAPI"
-echo "  --disable-vnc-tls        disable TLS encryption for VNC server"
-echo "  --enable-vnc-tls         enable TLS encryption for VNC server"
-echo "  --disable-vnc-sasl       disable SASL encryption for VNC server"
-echo "  --enable-vnc-sasl        enable SASL encryption for VNC server"
-echo "  --disable-vnc-jpeg       disable JPEG lossy compression for VNC server"
-echo "  --enable-vnc-jpeg        enable JPEG lossy compression for VNC server"
-echo "  --disable-vnc-png        disable PNG compression for VNC server (default)"
-echo "  --enable-vnc-png         enable PNG compression for VNC server"
-echo "  --disable-vnc-ws         disable Websockets support for VNC server"
-echo "  --enable-vnc-ws          enable Websockets support for VNC server"
-echo "  --disable-curses         disable curses output"
-echo "  --enable-curses          enable curses output"
-echo "  --disable-curl           disable curl connectivity"
-echo "  --enable-curl            enable curl connectivity"
-echo "  --disable-fdt            disable fdt device tree"
-echo "  --enable-fdt             enable fdt device tree"
-echo "  --disable-bluez          disable bluez stack connectivity"
-echo "  --enable-bluez           enable bluez stack connectivity"
-echo "  --disable-slirp          disable SLIRP userspace network connectivity"
-echo "  --disable-kvm            disable KVM acceleration support"
-echo "  --enable-kvm             enable KVM acceleration support"
-echo "  --disable-rdma           disable RDMA-based migration support"
-echo "  --enable-rdma            enable RDMA-based migration support"
-echo "  --enable-tcg-interpreter enable TCG with bytecode interpreter (TCI)"
-echo "  --enable-system          enable all system emulation targets"
-echo "  --disable-system         disable all system emulation targets"
-echo "  --enable-user            enable supported user emulation targets"
-echo "  --disable-user           disable all user emulation targets"
-echo "  --enable-linux-user      enable all linux usermode emulation targets"
-echo "  --disable-linux-user     disable all linux usermode emulation targets"
-echo "  --enable-bsd-user        enable all BSD usermode emulation targets"
-echo "  --disable-bsd-user       disable all BSD usermode emulation targets"
-echo "  --enable-guest-base      enable GUEST_BASE support for usermode"
-echo "                           emulation targets"
-echo "  --disable-guest-base     disable GUEST_BASE support"
-echo "  --enable-pie             build Position Independent Executables"
-echo "  --disable-pie            do not build Position Independent Executables"
-echo "  --fmod-lib               path to FMOD library"
-echo "  --fmod-inc               path to FMOD includes"
-echo "  --oss-lib                path to OSS library"
-echo "  --enable-uname-release=R Return R for uname -r in usermode emulation"
-echo "  --cpu=CPU                Build for host CPU [$cpu]"
-echo "  --disable-uuid           disable uuid support"
-echo "  --enable-uuid            enable uuid support"
-echo "  --disable-vde            disable support for vde network"
-echo "  --enable-vde             enable support for vde network"
-echo "  --disable-linux-aio      disable Linux AIO support"
-echo "  --enable-linux-aio       enable Linux AIO support"
-echo "  --disable-cap-ng         disable libcap-ng support"
-echo "  --enable-cap-ng          enable libcap-ng support"
-echo "  --disable-attr           disables attr and xattr support"
-echo "  --enable-attr            enable attr and xattr support"
-echo "  --disable-blobs          disable installing provided firmware blobs"
-echo "  --enable-docs            enable documentation build"
-echo "  --disable-docs           disable documentation build"
-echo "  --disable-vhost-net      disable vhost-net acceleration support"
-echo "  --enable-vhost-net       enable vhost-net acceleration support"
-echo "  --enable-trace-backend=B Set trace backend"
-echo "                           Available backends:" $($python "$source_path"/scripts/tracetool.py --list-backends)
-echo "  --with-trace-file=NAME   Full PATH,NAME of file to store traces"
-echo "                           Default:trace-<pid>"
-echo "  --disable-spice          disable spice"
-echo "  --enable-spice           enable spice"
-echo "  --enable-rbd             enable building the rados block device (rbd)"
-echo "  --disable-libiscsi       disable iscsi support"
-echo "  --enable-libiscsi        enable iscsi support"
-echo "  --disable-smartcard-nss  disable smartcard nss support"
-echo "  --enable-smartcard-nss   enable smartcard nss support"
-echo "  --disable-libusb         disable libusb (for usb passthrough)"
-echo "  --enable-libusb          enable libusb (for usb passthrough)"
-echo "  --disable-usb-redir      disable usb network redirection support"
-echo "  --enable-usb-redir       enable usb network redirection support"
-echo "  --disable-guest-agent    disable building of the QEMU Guest Agent"
-echo "  --enable-guest-agent     enable building of the QEMU Guest Agent"
-echo "  --with-vss-sdk=SDK-path  enable Windows VSS support in QEMU Guest Agent"
-echo "  --with-win-sdk=SDK-path  path to Windows Platform SDK (to build VSS .tlb)"
-echo "  --disable-seccomp        disable seccomp support"
-echo "  --enable-seccomp         enables seccomp support"
-echo "  --with-coroutine=BACKEND coroutine backend. Supported options:"
-echo "                           gthread, ucontext, sigaltstack, windows"
-echo "  --disable-coroutine-pool disable coroutine freelist (worse performance)"
-echo "  --enable-coroutine-pool  enable coroutine freelist (better performance)"
-echo "  --enable-glusterfs       enable GlusterFS backend"
-echo "  --disable-glusterfs      disable GlusterFS backend"
-echo "  --enable-gcov            enable test coverage analysis with gcov"
-echo "  --gcov=GCOV              use specified gcov [$gcov_tool]"
-echo "  --enable-tpm             enable TPM support"
-echo "  --disable-libssh2        disable ssh block device support"
-echo "  --enable-libssh2         enable ssh block device support"
-echo "  --disable-vhdx           disables support for the Microsoft VHDX image format"
-echo "  --enable-vhdx            enable support for the Microsoft VHDX image format"
-echo ""
-echo "NOTE: The object files are built at the place where configure is launched"
 exit 1
 fi
 
@@ -1411,25 +1438,6 @@ if test "$solaris" = "yes" ; then
   fi
 fi
 
-if ! has $python; then
-  error_exit "Python not found. Use --python=/path/to/python"
-fi
-
-# Note that if the Python conditional here evaluates True we will exit
-# with status 1 which is a shell 'false' value.
-if ! $python -c 'import sys; sys.exit(sys.version_info < (2,4) or sys.version_info >= (3,))'; then
-  error_exit "Cannot use '$python', Python 2.4 or later is required." \
-      "Note that Python 3 or later is not yet supported." \
-      "Use --python=/path/to/python to specify a supported Python."
-fi
-
-# The -B switch was added in Python 2.6.
-# If it is supplied, compiled files are not written.
-# Use it for Python versions which support it.
-if $python -B -c 'import sys; sys.exit(0)' 2>/dev/null; then
-  python="$python -B"
-fi
-
 if test -z "${target_list+xxx}" ; then
     target_list="$default_target_list"
 else
@@ -2094,6 +2102,26 @@ EOF
   fi
 fi
 
+##########################################
+# netmap headers probe
+if test "$netmap" != "no" ; then
+  cat > $TMPC << EOF
+#include <inttypes.h>
+#include <net/if.h>
+#include <net/netmap.h>
+#include <net/netmap_user.h>
+int main(void) { return 0; }
+EOF
+  if compile_prog "" "" ; then
+    netmap=yes
+  else
+    if test "$netmap" = "yes" ; then
+      feature_not_found "netmap"
+    fi
+    netmap=no
+  fi
+fi
+
 ##########################################
 # libcap-ng library probe
 if test "$cap_ng" != "no" ; then
@@ -3025,13 +3053,13 @@ fi
 
 ##########################################
 # Do we have libiscsi
-# We check for iscsi_unmap_sync() to make sure we have a
-# recent enough version of libiscsi.
+# We check for iscsi_write16_sync() to make sure we have a
+# at least version 1.4.0 of libiscsi.
 if test "$libiscsi" != "no" ; then
   cat > $TMPC << EOF
 #include <stdio.h>
 #include <iscsi/iscsi.h>
-int main(void) { iscsi_unmap_sync(NULL,0,0,0,NULL,0); return 0; }
+int main(void) { iscsi_write16_sync(NULL,0,0,NULL,0,0,0,0,0,0,0); return 0; }
 EOF
   if $pkg_config --atleast-version=1.7.0 libiscsi; then
     libiscsi="yes"
@@ -3050,6 +3078,21 @@ EOF
   fi
 fi
 
+# We also need to know the API version because there was an
+# API change from 1.4.0 to 1.5.0.
+if test "$libiscsi" = "yes"; then
+  cat >$TMPC <<EOF
+#include <iscsi/iscsi.h>
+int main(void)
+{
+  iscsi_read10_task(0, 0, 0, 0, 0, 0, 0);
+  return 0;
+}
+EOF
+  if compile_prog "" "-liscsi"; then
+    libiscsi_version="1.4.0"
+  fi
+fi
 
 ##########################################
 # Do we need libm
@@ -3751,6 +3794,7 @@ echo "uname -r          $uname_release"
 echo "GUEST_BASE        $guest_base"
 echo "PIE               $pie"
 echo "vde support       $vde"
+echo "netmap support    $netmap"
 echo "Linux AIO support $linux_aio"
 echo "ATTR/XATTR support $attr"
 echo "Install blobs     $blobs"
@@ -3776,7 +3820,11 @@ echo "nss used          $smartcard_nss"
 echo "libusb            $libusb"
 echo "usb net redir     $usb_redir"
 echo "GLX support       $glx"
+if test "$libiscsi_version" = "1.4.0"; then
+echo "libiscsi support  $libiscsi (1.4.0)"
+else
 echo "libiscsi support  $libiscsi"
+fi
 echo "build guest agent $guest_agent"
 echo "QGA VSS support   $guest_agent_with_vss"
 echo "seccomp support   $seccomp"
@@ -3888,6 +3936,9 @@ fi
 if test "$vde" = "yes" ; then
   echo "CONFIG_VDE=y" >> $config_host_mak
 fi
+if test "$netmap" = "yes" ; then
+  echo "CONFIG_NETMAP=y" >> $config_host_mak
+fi
 if test "$cap_ng" = "yes" ; then
   echo "CONFIG_LIBCAP=y" >> $config_host_mak
 fi
@@ -4105,6 +4156,9 @@ fi
 
 if test "$libiscsi" = "yes" ; then
   echo "CONFIG_LIBISCSI=y" >> $config_host_mak
+  if test "$libiscsi_version" = "1.4.0"; then
+    echo "CONFIG_LIBISCSI_1_4=y" >> $config_host_mak
+  fi
 fi
 
 if test "$seccomp" = "yes"; then
@@ -4406,7 +4460,7 @@ case "$target_name" in
   aarch64)
     TARGET_BASE_ARCH=arm
     bflt="yes"
-    gdb_xml_files="aarch64-core.xml"
+    gdb_xml_files="aarch64-core.xml aarch64-fpu.xml"
   ;;
   cris)
   ;;
@@ -4518,7 +4572,7 @@ case "$target_name" in
   *)
 esac
 case "$target_name" in
-  arm|i386|x86_64|ppcemb|ppc|ppc64|s390x)
+  aarch64|arm|i386|x86_64|ppcemb|ppc|ppc64|s390x)
     # Make sure the target and host cpus are compatible
     if test "$kvm" = "yes" -a "$target_softmmu" = "yes" -a \
       \( "$target_name" = "$cpu" -o \

cpu-exec.c

@@ -53,7 +53,25 @@ void cpu_resume_from_signal(CPUArchState *env, void *puc)
 static inline tcg_target_ulong cpu_tb_exec(CPUState *cpu, uint8_t *tb_ptr)
 {
     CPUArchState *env = cpu->env_ptr;
-    uintptr_t next_tb = tcg_qemu_tb_exec(env, tb_ptr);
+    uintptr_t next_tb;
+
+#if defined(DEBUG_DISAS)
+    if (qemu_loglevel_mask(CPU_LOG_TB_CPU)) {
+#if defined(TARGET_I386)
+        log_cpu_state(cpu, CPU_DUMP_CCOP);
+#elif defined(TARGET_M68K)
+        /* ??? Should not modify env state for dumping.  */
+        cpu_m68k_flush_flags(env, env->cc_op);
+        env->cc_op = CC_OP_FLAGS;
+        env->sr = (env->sr & 0xffe0) | env->cc_dest | (env->cc_x << 4);
+        log_cpu_state(cpu, 0);
+#else
+        log_cpu_state(cpu, 0);
+#endif
+    }
+#endif /* DEBUG_DISAS */
+
+    next_tb = tcg_qemu_tb_exec(env, tb_ptr);
     if ((next_tb & TB_EXIT_MASK) > TB_EXIT_IDX1) {
         /* We didn't start executing this TB (eg because the instruction
          * counter hit zero); we must restore the guest PC to the address
@@ -205,6 +223,9 @@ int cpu_exec(CPUArchState *env)
 #if !(defined(CONFIG_USER_ONLY) && \
       (defined(TARGET_M68K) || defined(TARGET_PPC) || defined(TARGET_S390X)))
     CPUClass *cc = CPU_GET_CLASS(cpu);
+#endif
+#ifdef TARGET_I386
+    X86CPU *x86_cpu = X86_CPU(cpu);
 #endif
     int ret, interrupt_request;
     TranslationBlock *tb;
@@ -320,24 +341,24 @@ int cpu_exec(CPUArchState *env)
 #if !defined(CONFIG_USER_ONLY)
                     if (interrupt_request & CPU_INTERRUPT_POLL) {
                         cpu->interrupt_request &= ~CPU_INTERRUPT_POLL;
-                        apic_poll_irq(env->apic_state);
+                        apic_poll_irq(x86_cpu->apic_state);
                     }
 #endif
                     if (interrupt_request & CPU_INTERRUPT_INIT) {
                             cpu_svm_check_intercept_param(env, SVM_EXIT_INIT,
                                                           0);
-                            do_cpu_init(x86_env_get_cpu(env));
+                            do_cpu_init(x86_cpu);
                             env->exception_index = EXCP_HALTED;
                             cpu_loop_exit(env);
                     } else if (interrupt_request & CPU_INTERRUPT_SIPI) {
-                            do_cpu_sipi(x86_env_get_cpu(env));
+                            do_cpu_sipi(x86_cpu);
                     } else if (env->hflags2 & HF2_GIF_MASK) {
                         if ((interrupt_request & CPU_INTERRUPT_SMI) &&
                             !(env->hflags & HF_SMM_MASK)) {
                             cpu_svm_check_intercept_param(env, SVM_EXIT_SMI,
                                                           0);
                             cpu->interrupt_request &= ~CPU_INTERRUPT_SMI;
-                            do_smm_enter(x86_env_get_cpu(env));
+                            do_smm_enter(x86_cpu);
                             next_tb = 0;
                         } else if ((interrupt_request & CPU_INTERRUPT_NMI) &&
                                    !(env->hflags2 & HF2_NMI_MASK)) {
@@ -579,22 +600,6 @@ int cpu_exec(CPUArchState *env)
                     env->exception_index = EXCP_INTERRUPT;
                     cpu_loop_exit(env);
                 }
-#if defined(DEBUG_DISAS)
-                if (qemu_loglevel_mask(CPU_LOG_TB_CPU)) {
-                    /* restore flags in standard format */
-#if defined(TARGET_I386)
-                    log_cpu_state(cpu, CPU_DUMP_CCOP);
-#elif defined(TARGET_M68K)
-                    cpu_m68k_flush_flags(env, env->cc_op);
-                    env->cc_op = CC_OP_FLAGS;
-                    env->sr = (env->sr & 0xffe0)
-                              | env->cc_dest | (env->cc_x << 4);
-                    log_cpu_state(cpu, 0);
-#else
-                    log_cpu_state(cpu, 0);
-#endif
-                }
-#endif /* DEBUG_DISAS */
                 spin_lock(&tcg_ctx.tb_ctx.tb_lock);
                 tb = tb_find_fast(env);
                 /* Note: we do it here to avoid a gcc bug on Mac OS X when
@@ -684,6 +689,9 @@ int cpu_exec(CPUArchState *env)
 #if !(defined(CONFIG_USER_ONLY) && \
       (defined(TARGET_M68K) || defined(TARGET_PPC) || defined(TARGET_S390X)))
             cc = CPU_GET_CLASS(cpu);
+#endif
+#ifdef TARGET_I386
+            x86_cpu = X86_CPU(cpu);
 #endif
         }
     } /* for(;;) */

cpus.c

@@ -1458,12 +1458,11 @@ void qmp_inject_nmi(Error **errp)
 
     CPU_FOREACH(cs) {
         X86CPU *cpu = X86_CPU(cs);
-        CPUX86State *env = &cpu->env;
 
-        if (!env->apic_state) {
+        if (!cpu->apic_state) {
             cpu_interrupt(cs, CPU_INTERRUPT_NMI);
         } else {
-            apic_deliver_nmi(env->apic_state);
+            apic_deliver_nmi(cpu->apic_state);
         }
     }
 #elif defined(TARGET_S390X)

cputlb.c

@@ -26,6 +26,7 @@
 #include "exec/cputlb.h"
 
 #include "exec/memory-internal.h"
+#include "exec/ram_addr.h"
 
 //#define DEBUG_TLB
 //#define DEBUG_TLB_CHECK
@@ -33,13 +34,6 @@
 /* statistics */
 int tlb_flush_count;
 
-static const CPUTLBEntry s_cputlb_empty_entry = {
-    .addr_read  = -1,
-    .addr_write = -1,
-    .addr_code  = -1,
-    .addend     = -1,
-};
-
 /* NOTE:
  * If flush_global is true (the usual case), flush all tlb entries.
  * If flush_global is false, flush (at least) all tlb entries not
@@ -55,7 +49,6 @@ static const CPUTLBEntry s_cputlb_empty_entry = {
 void tlb_flush(CPUArchState *env, int flush_global)
 {
     CPUState *cpu = ENV_GET_CPU(env);
-    int i;
 
 #if defined(DEBUG_TLB)
     printf("tlb_flush:\n");
@@ -64,15 +57,8 @@ void tlb_flush(CPUArchState *env, int flush_global)
        links while we are modifying them */
     cpu->current_tb = NULL;
 
-    for (i = 0; i < CPU_TLB_SIZE; i++) {
-        int mmu_idx;
-
-        for (mmu_idx = 0; mmu_idx < NB_MMU_MODES; mmu_idx++) {
-            env->tlb_table[mmu_idx][i] = s_cputlb_empty_entry;
-        }
-    }
-
-    memset(env->tb_jmp_cache, 0, TB_JMP_CACHE_SIZE * sizeof (void *));
+    memset(env->tlb_table, -1, sizeof(env->tlb_table));
+    memset(env->tb_jmp_cache, 0, sizeof(env->tb_jmp_cache));
 
     env->tlb_flush_addr = -1;
     env->tlb_flush_mask = 0;
@@ -87,7 +73,7 @@ static inline void tlb_flush_entry(CPUTLBEntry *tlb_entry, target_ulong addr)
                  (TARGET_PAGE_MASK | TLB_INVALID_MASK)) ||
         addr == (tlb_entry->addr_code &
                  (TARGET_PAGE_MASK | TLB_INVALID_MASK))) {
-        *tlb_entry = s_cputlb_empty_entry;
+        memset(tlb_entry, -1, sizeof(*tlb_entry));
     }
 }
 
@@ -127,9 +113,8 @@ void tlb_flush_page(CPUArchState *env, target_ulong addr)
    can be detected */
 void tlb_protect_code(ram_addr_t ram_addr)
 {
-    cpu_physical_memory_reset_dirty(ram_addr,
-                                    ram_addr + TARGET_PAGE_SIZE,
-                                    CODE_DIRTY_FLAG);
+    cpu_physical_memory_reset_dirty(ram_addr, TARGET_PAGE_SIZE,
+                                    DIRTY_MEMORY_CODE);
 }
 
 /* update the TLB so that writes in physical page 'phys_addr' are no longer
@@ -137,7 +122,7 @@ void tlb_protect_code(ram_addr_t ram_addr)
 void tlb_unprotect_code_phys(CPUArchState *env, ram_addr_t ram_addr,
                              target_ulong vaddr)
 {
-    cpu_physical_memory_set_dirty_flags(ram_addr, CODE_DIRTY_FLAG);
+    cpu_physical_memory_set_dirty_flag(ram_addr, DIRTY_MEMORY_CODE);
 }
 
 static bool tlb_is_dirty_ram(CPUTLBEntry *tlbe)
@@ -299,7 +284,8 @@ void tlb_set_page(CPUArchState *env, target_ulong vaddr,
             /* Write access calls the I/O callback.  */
             te->addr_write = address | TLB_MMIO;
         } else if (memory_region_is_ram(section->mr)
-                   && !cpu_physical_memory_is_dirty(section->mr->ram_addr + xlat)) {
+                   && cpu_physical_memory_is_clean(section->mr->ram_addr
+                                                   + xlat)) {
             te->addr_write = address | TLB_NOTDIRTY;
         } else {
             te->addr_write = address;

default-configs/aarch64-linux-user.mak

@@ -0,0 +1,3 @@
+# Default configuration for aarch64-linux-user
+
+CONFIG_GDBSTUB_XML=y

default-configs/aarch64-softmmu.mak

@@ -0,0 +1,6 @@
+# Default configuration for aarch64-softmmu
+
+# We support all the 32 bit boards so need all their config
+include arm-softmmu.mak
+
+# Currently no 64-bit specific config requirements

default-configs/arm-softmmu.mak

@@ -41,6 +41,7 @@ CONFIG_ARM_GIC=y
 CONFIG_ARM_GIC_KVM=$(CONFIG_KVM)
 CONFIG_ARM_TIMER=y
 CONFIG_ARM_MPTIMER=y
+CONFIG_A9_GTIMER=y
 CONFIG_PL011=y
 CONFIG_PL022=y
 CONFIG_PL031=y
@@ -63,6 +64,7 @@ CONFIG_XILINX_SPIPS=y
 
 CONFIG_ARM11SCU=y
 CONFIG_A9SCU=y
+CONFIG_DIGIC=y
 CONFIG_MARVELL_88W8618=y
 CONFIG_OMAP=y
 CONFIG_TSC210X=y
@@ -81,3 +83,7 @@ CONFIG_VERSATILE_I2C=y
 
 CONFIG_SDHCI=y
 CONFIG_INTEGRATOR_DEBUG=y
+
+CONFIG_ALLWINNER_A10_PIT=y
+CONFIG_ALLWINNER_A10_PIC=y
+CONFIG_ALLWINNER_A10=y

device_tree.c

@@ -41,6 +41,10 @@ void *create_device_tree(int *sizep)
     if (ret < 0) {
         goto fail;
     }
+    ret = fdt_finish_reservemap(fdt);
+    if (ret < 0) {
+        goto fail;
+    }
     ret = fdt_begin_node(fdt, "");
     if (ret < 0) {
         goto fail;
@@ -127,12 +131,12 @@ static int findnode_nofail(void *fdt, const char *node_path)
     return offset;
 }
 
-int qemu_devtree_setprop(void *fdt, const char *node_path,
-                         const char *property, const void *val_array, int size)
+int qemu_fdt_setprop(void *fdt, const char *node_path,
+                     const char *property, const void *val, int size)
 {
     int r;
 
-    r = fdt_setprop(fdt, findnode_nofail(fdt, node_path), property, val_array, size);
+    r = fdt_setprop(fdt, findnode_nofail(fdt, node_path), property, val, size);
     if (r < 0) {
         fprintf(stderr, "%s: Couldn't set %s/%s: %s\n", __func__, node_path,
                 property, fdt_strerror(r));
@@ -142,8 +146,8 @@ int qemu_devtree_setprop(void *fdt, const char *node_path,
     return r;
 }
 
-int qemu_devtree_setprop_cell(void *fdt, const char *node_path,
-                              const char *property, uint32_t val)
+int qemu_fdt_setprop_cell(void *fdt, const char *node_path,
+                          const char *property, uint32_t val)
 {
     int r;
 
@@ -157,15 +161,15 @@ int qemu_devtree_setprop_cell(void *fdt, const char *node_path,
     return r;
 }
 
-int qemu_devtree_setprop_u64(void *fdt, const char *node_path,
-                             const char *property, uint64_t val)
+int qemu_fdt_setprop_u64(void *fdt, const char *node_path,
+                         const char *property, uint64_t val)
 {
     val = cpu_to_be64(val);
-    return qemu_devtree_setprop(fdt, node_path, property, &val, sizeof(val));
+    return qemu_fdt_setprop(fdt, node_path, property, &val, sizeof(val));
 }
 
-int qemu_devtree_setprop_string(void *fdt, const char *node_path,
-                                const char *property, const char *string)
+int qemu_fdt_setprop_string(void *fdt, const char *node_path,
+                            const char *property, const char *string)
 {
     int r;
 
@@ -179,8 +183,8 @@ int qemu_devtree_setprop_string(void *fdt, const char *node_path,
     return r;
 }
 
-const void *qemu_devtree_getprop(void *fdt, const char *node_path,
-                                 const char *property, int *lenp)
+const void *qemu_fdt_getprop(void *fdt, const char *node_path,
+                             const char *property, int *lenp)
 {
     int len;
     const void *r;
@@ -196,11 +200,11 @@ const void *qemu_devtree_getprop(void *fdt, const char *node_path,
     return r;
 }
 
-uint32_t qemu_devtree_getprop_cell(void *fdt, const char *node_path,
-                                   const char *property)
+uint32_t qemu_fdt_getprop_cell(void *fdt, const char *node_path,
+                               const char *property)
 {
     int len;
-    const uint32_t *p = qemu_devtree_getprop(fdt, node_path, property, &len);
+    const uint32_t *p = qemu_fdt_getprop(fdt, node_path, property, &len);
     if (len != 4) {
         fprintf(stderr, "%s: %s/%s not 4 bytes long (not a cell?)\n",
                 __func__, node_path, property);
@@ -209,7 +213,7 @@ uint32_t qemu_devtree_getprop_cell(void *fdt, const char *node_path,
     return be32_to_cpu(*p);
 }
 
-uint32_t qemu_devtree_get_phandle(void *fdt, const char *path)
+uint32_t qemu_fdt_get_phandle(void *fdt, const char *path)
 {
     uint32_t r;
 
@@ -223,15 +227,15 @@ uint32_t qemu_devtree_get_phandle(void *fdt, const char *path)
     return r;
 }
 
-int qemu_devtree_setprop_phandle(void *fdt, const char *node_path,
-                                 const char *property,
-                                 const char *target_node_path)
+int qemu_fdt_setprop_phandle(void *fdt, const char *node_path,
+                             const char *property,
+                             const char *target_node_path)
 {
-    uint32_t phandle = qemu_devtree_get_phandle(fdt, target_node_path);
-    return qemu_devtree_setprop_cell(fdt, node_path, property, phandle);
+    uint32_t phandle = qemu_fdt_get_phandle(fdt, target_node_path);
+    return qemu_fdt_setprop_cell(fdt, node_path, property, phandle);
 }
 
-uint32_t qemu_devtree_alloc_phandle(void *fdt)
+uint32_t qemu_fdt_alloc_phandle(void *fdt)
 {
     static int phandle = 0x0;
 
@@ -255,7 +259,7 @@ uint32_t qemu_devtree_alloc_phandle(void *fdt)
     return phandle++;
 }
 
-int qemu_devtree_nop_node(void *fdt, const char *node_path)
+int qemu_fdt_nop_node(void *fdt, const char *node_path)
 {
     int r;
 
@@ -269,7 +273,7 @@ int qemu_devtree_nop_node(void *fdt, const char *node_path)
     return r;
 }
 
-int qemu_devtree_add_subnode(void *fdt, const char *name)
+int qemu_fdt_add_subnode(void *fdt, const char *name)
 {
     char *dupname = g_strdup(name);
     char *basename = strrchr(dupname, '/');
@@ -299,7 +303,7 @@ int qemu_devtree_add_subnode(void *fdt, const char *name)
     return retval;
 }
 
-void qemu_devtree_dumpdtb(void *fdt, int size)
+void qemu_fdt_dumpdtb(void *fdt, int size)
 {
     const char *dumpdtb = qemu_opt_get(qemu_get_machine_opts(), "dumpdtb");
 
@@ -309,11 +313,11 @@ void qemu_devtree_dumpdtb(void *fdt, int size)
     }
 }
 
-int qemu_devtree_setprop_sized_cells_from_array(void *fdt,
-                                                const char *node_path,
-                                                const char *property,
-                                                int numvalues,
-                                                uint64_t *values)
+int qemu_fdt_setprop_sized_cells_from_array(void *fdt,
+                                            const char *node_path,
+                                            const char *property,
+                                            int numvalues,
+                                            uint64_t *values)
 {
     uint32_t *propcells;
     uint64_t value;
@@ -338,6 +342,6 @@ int qemu_devtree_setprop_sized_cells_from_array(void *fdt,
         propcells[cellnum++] = cpu_to_be32(value);
     }
 
-    return qemu_devtree_setprop(fdt, node_path, property, propcells,
-                                cellnum * sizeof(uint32_t));
+    return qemu_fdt_setprop(fdt, node_path, property, propcells,
+                            cellnum * sizeof(uint32_t));
 }

exec.c

@@ -50,11 +50,15 @@
 #include "translate-all.h"
 
 #include "exec/memory-internal.h"
+#include "exec/ram_addr.h"
+#include "qemu/cache-utils.h"
+
+#include "qemu/range.h"
 
 //#define DEBUG_SUBPAGE
 
 #if !defined(CONFIG_USER_ONLY)
-static int in_migration;
+static bool in_migration;
 
 RAMList ram_list = { .blocks = QTAILQ_HEAD_INITIALIZER(ram_list.blocks) };
 
@@ -83,20 +87,39 @@ int use_icount;
 typedef struct PhysPageEntry PhysPageEntry;
 
 struct PhysPageEntry {
-    uint16_t is_leaf : 1;
-     /* index into phys_sections (is_leaf) or phys_map_nodes (!is_leaf) */
-    uint16_t ptr : 15;
+    /* How many bits skip to next level (in units of L2_SIZE). 0 for a leaf. */
+    uint32_t skip : 6;
+     /* index into phys_sections (!skip) or phys_map_nodes (skip) */
+    uint32_t ptr : 26;
 };
 
-typedef PhysPageEntry Node[L2_SIZE];
+#define PHYS_MAP_NODE_NIL (((uint32_t)~0) >> 6)
+
+/* Size of the L2 (and L3, etc) page tables.  */
+#define ADDR_SPACE_BITS 64
+
+#define P_L2_BITS 9
+#define P_L2_SIZE (1 << P_L2_BITS)
+
+#define P_L2_LEVELS (((ADDR_SPACE_BITS - TARGET_PAGE_BITS - 1) / P_L2_BITS) + 1)
+
+typedef PhysPageEntry Node[P_L2_SIZE];
+
+typedef struct PhysPageMap {
+    unsigned sections_nb;
+    unsigned sections_nb_alloc;
+    unsigned nodes_nb;
+    unsigned nodes_nb_alloc;
+    Node *nodes;
+    MemoryRegionSection *sections;
+} PhysPageMap;
 
 struct AddressSpaceDispatch {
     /* This is a multi-level map on the physical address space.
      * The bottom level has pointers to MemoryRegionSections.
      */
     PhysPageEntry phys_map;
-    Node *nodes;
-    MemoryRegionSection *sections;
+    PhysPageMap map;
     AddressSpace *as;
 };
 
@@ -113,20 +136,6 @@ typedef struct subpage_t {
 #define PHYS_SECTION_ROM 2
 #define PHYS_SECTION_WATCH 3
 
-typedef struct PhysPageMap {
-    unsigned sections_nb;
-    unsigned sections_nb_alloc;
-    unsigned nodes_nb;
-    unsigned nodes_nb_alloc;
-    Node *nodes;
-    MemoryRegionSection *sections;
-} PhysPageMap;
-
-static PhysPageMap *prev_map;
-static PhysPageMap next_map;
-
-#define PHYS_MAP_NODE_NIL (((uint16_t)~0) >> 1)
-
 static void io_mem_init(void);
 static void memory_map_init(void);
 
@@ -135,63 +144,60 @@ static MemoryRegion io_mem_watch;
 
 #if !defined(CONFIG_USER_ONLY)
 
-static void phys_map_node_reserve(unsigned nodes)
+static void phys_map_node_reserve(PhysPageMap *map, unsigned nodes)
 {
-    if (next_map.nodes_nb + nodes > next_map.nodes_nb_alloc) {
-        next_map.nodes_nb_alloc = MAX(next_map.nodes_nb_alloc * 2,
-                                            16);
-        next_map.nodes_nb_alloc = MAX(next_map.nodes_nb_alloc,
-                                      next_map.nodes_nb + nodes);
-        next_map.nodes = g_renew(Node, next_map.nodes,
-                                 next_map.nodes_nb_alloc);
+    if (map->nodes_nb + nodes > map->nodes_nb_alloc) {
+        map->nodes_nb_alloc = MAX(map->nodes_nb_alloc * 2, 16);
+        map->nodes_nb_alloc = MAX(map->nodes_nb_alloc, map->nodes_nb + nodes);
+        map->nodes = g_renew(Node, map->nodes, map->nodes_nb_alloc);
     }
 }
 
-static uint16_t phys_map_node_alloc(void)
+static uint32_t phys_map_node_alloc(PhysPageMap *map)
 {
     unsigned i;
-    uint16_t ret;
+    uint32_t ret;
 
-    ret = next_map.nodes_nb++;
+    ret = map->nodes_nb++;
     assert(ret != PHYS_MAP_NODE_NIL);
-    assert(ret != next_map.nodes_nb_alloc);
-    for (i = 0; i < L2_SIZE; ++i) {
-        next_map.nodes[ret][i].is_leaf = 0;
-        next_map.nodes[ret][i].ptr = PHYS_MAP_NODE_NIL;
+    assert(ret != map->nodes_nb_alloc);
+    for (i = 0; i < P_L2_SIZE; ++i) {
+        map->nodes[ret][i].skip = 1;
+        map->nodes[ret][i].ptr = PHYS_MAP_NODE_NIL;
     }
     return ret;
 }
 
-static void phys_page_set_level(PhysPageEntry *lp, hwaddr *index,
-                                hwaddr *nb, uint16_t leaf,
+static void phys_page_set_level(PhysPageMap *map, PhysPageEntry *lp,
+                                hwaddr *index, hwaddr *nb, uint16_t leaf,
                                 int level)
 {
     PhysPageEntry *p;
     int i;
-    hwaddr step = (hwaddr)1 << (level * L2_BITS);
+    hwaddr step = (hwaddr)1 << (level * P_L2_BITS);
 
-    if (!lp->is_leaf && lp->ptr == PHYS_MAP_NODE_NIL) {
-        lp->ptr = phys_map_node_alloc();
-        p = next_map.nodes[lp->ptr];
+    if (lp->skip && lp->ptr == PHYS_MAP_NODE_NIL) {
+        lp->ptr = phys_map_node_alloc(map);
+        p = map->nodes[lp->ptr];
         if (level == 0) {
-            for (i = 0; i < L2_SIZE; i++) {
-                p[i].is_leaf = 1;
+            for (i = 0; i < P_L2_SIZE; i++) {
+                p[i].skip = 0;
                 p[i].ptr = PHYS_SECTION_UNASSIGNED;
             }
         }
     } else {
-        p = next_map.nodes[lp->ptr];
+        p = map->nodes[lp->ptr];
     }
-    lp = &p[(*index >> (level * L2_BITS)) & (L2_SIZE - 1)];
+    lp = &p[(*index >> (level * P_L2_BITS)) & (P_L2_SIZE - 1)];
 
-    while (*nb && lp < &p[L2_SIZE]) {
+    while (*nb && lp < &p[P_L2_SIZE]) {
         if ((*index & (step - 1)) == 0 && *nb >= step) {
-            lp->is_leaf = true;
+            lp->skip = 0;
             lp->ptr = leaf;
             *index += step;
             *nb -= step;
         } else {
-            phys_page_set_level(lp, index, nb, leaf, level - 1);
+            phys_page_set_level(map, lp, index, nb, leaf, level - 1);
         }
         ++lp;
     }
@@ -202,25 +208,95 @@ static void phys_page_set(AddressSpaceDispatch *d,
                           uint16_t leaf)
 {
     /* Wildly overreserve - it doesn't matter much. */
-    phys_map_node_reserve(3 * P_L2_LEVELS);
+    phys_map_node_reserve(&d->map, 3 * P_L2_LEVELS);
 
-    phys_page_set_level(&d->phys_map, &index, &nb, leaf, P_L2_LEVELS - 1);
+    phys_page_set_level(&d->map, &d->phys_map, &index, &nb, leaf, P_L2_LEVELS - 1);
 }
 
-static MemoryRegionSection *phys_page_find(PhysPageEntry lp, hwaddr index,
-                                           Node *nodes, MemoryRegionSection *sections)
+/* Compact a non leaf page entry. Simply detect that the entry has a single child,
+ * and update our entry so we can skip it and go directly to the destination.
+ */
+static void phys_page_compact(PhysPageEntry *lp, Node *nodes, unsigned long *compacted)
 {
+    unsigned valid_ptr = P_L2_SIZE;
+    int valid = 0;
     PhysPageEntry *p;
     int i;
 
-    for (i = P_L2_LEVELS - 1; i >= 0 && !lp.is_leaf; i--) {
+    if (lp->ptr == PHYS_MAP_NODE_NIL) {
+        return;
+    }
+
+    p = nodes[lp->ptr];
+    for (i = 0; i < P_L2_SIZE; i++) {
+        if (p[i].ptr == PHYS_MAP_NODE_NIL) {
+            continue;
+        }
+
+        valid_ptr = i;
+        valid++;
+        if (p[i].skip) {
+            phys_page_compact(&p[i], nodes, compacted);
+        }
+    }
+
+    /* We can only compress if there's only one child. */
+    if (valid != 1) {
+        return;
+    }
+
+    assert(valid_ptr < P_L2_SIZE);
+
+    /* Don't compress if it won't fit in the # of bits we have. */
+    if (lp->skip + p[valid_ptr].skip >= (1 << 3)) {
+        return;
+    }
+
+    lp->ptr = p[valid_ptr].ptr;
+    if (!p[valid_ptr].skip) {
+        /* If our only child is a leaf, make this a leaf. */
+        /* By design, we should have made this node a leaf to begin with so we
+         * should never reach here.
+         * But since it's so simple to handle this, let's do it just in case we
+         * change this rule.
+         */
+        lp->skip = 0;
+    } else {
+        lp->skip += p[valid_ptr].skip;
+    }
+}
+
+static void phys_page_compact_all(AddressSpaceDispatch *d, int nodes_nb)
+{
+    DECLARE_BITMAP(compacted, nodes_nb);
+
+    if (d->phys_map.skip) {
+        phys_page_compact(&d->phys_map, d->map.nodes, compacted);
+    }
+}
+
+static MemoryRegionSection *phys_page_find(PhysPageEntry lp, hwaddr addr,
+                                           Node *nodes, MemoryRegionSection *sections)
+{
+    PhysPageEntry *p;
+    hwaddr index = addr >> TARGET_PAGE_BITS;
+    int i;
+
+    for (i = P_L2_LEVELS; lp.skip && (i -= lp.skip) >= 0;) {
         if (lp.ptr == PHYS_MAP_NODE_NIL) {
             return &sections[PHYS_SECTION_UNASSIGNED];
         }
         p = nodes[lp.ptr];
-        lp = p[(index >> (i * L2_BITS)) & (L2_SIZE - 1)];
+        lp = p[(index >> (i * P_L2_BITS)) & (P_L2_SIZE - 1)];
+    }
+
+    if (sections[lp.ptr].size.hi ||
+        range_covers_byte(sections[lp.ptr].offset_within_address_space,
+                          sections[lp.ptr].size.lo, addr)) {
+        return &sections[lp.ptr];
+    } else {
+        return &sections[PHYS_SECTION_UNASSIGNED];
     }
-    return &sections[lp.ptr];
 }
 
 bool memory_region_is_unassigned(MemoryRegion *mr)
@@ -236,11 +312,10 @@ static MemoryRegionSection *address_space_lookup_region(AddressSpaceDispatch *d,
     MemoryRegionSection *section;
     subpage_t *subpage;
 
-    section = phys_page_find(d->phys_map, addr >> TARGET_PAGE_BITS,
-                             d->nodes, d->sections);
+    section = phys_page_find(d->phys_map, addr, d->map.nodes, d->map.sections);
     if (resolve_subpage && section->mr->subpage) {
         subpage = container_of(section->mr, subpage_t, iomem);
-        section = &d->sections[subpage->sub_section[SUBPAGE_IDX(addr)]];
+        section = &d->map.sections[subpage->sub_section[SUBPAGE_IDX(addr)]];
     }
     return section;
 }
@@ -650,11 +725,14 @@ found:
     return block;
 }
 
-static void tlb_reset_dirty_range_all(ram_addr_t start, ram_addr_t end,
-                                      uintptr_t length)
+static void tlb_reset_dirty_range_all(ram_addr_t start, ram_addr_t length)
 {
-    RAMBlock *block;
     ram_addr_t start1;
+    RAMBlock *block;
+    ram_addr_t end;
+
+    end = TARGET_PAGE_ALIGN(start + length);
+    start &= TARGET_PAGE_MASK;
 
     block = qemu_get_ram_block(start);
     assert(block == qemu_get_ram_block(end - 1));
@@ -663,29 +741,21 @@ static void tlb_reset_dirty_range_all(ram_addr_t start, ram_addr_t end,
 }
 
 /* Note: start and end must be within the same ram block.  */
-void cpu_physical_memory_reset_dirty(ram_addr_t start, ram_addr_t end,
-                                     int dirty_flags)
+void cpu_physical_memory_reset_dirty(ram_addr_t start, ram_addr_t length,
+                                     unsigned client)
 {
-    uintptr_t length;
-
-    start &= TARGET_PAGE_MASK;
-    end = TARGET_PAGE_ALIGN(end);
-
-    length = end - start;
     if (length == 0)
         return;
-    cpu_physical_memory_mask_dirty_range(start, length, dirty_flags);
+    cpu_physical_memory_clear_dirty_range(start, length, client);
 
     if (tcg_enabled()) {
-        tlb_reset_dirty_range_all(start, end, length);
+        tlb_reset_dirty_range_all(start, length);
     }
 }
 
-static int cpu_physical_memory_set_dirty_tracking(int enable)
+static void cpu_physical_memory_set_dirty_tracking(bool enable)
 {
-    int ret = 0;
     in_migration = enable;
-    return ret;
 }
 
 hwaddr memory_region_section_get_iotlb(CPUArchState *env,
@@ -708,7 +778,7 @@ hwaddr memory_region_section_get_iotlb(CPUArchState *env,
             iotlb |= PHYS_SECTION_ROM;
         }
     } else {
-        iotlb = section - address_space_memory.dispatch->sections;
+        iotlb = section - address_space_memory.dispatch->map.sections;
         iotlb += xlat;
     }
 
@@ -747,23 +817,23 @@ void phys_mem_set_alloc(void *(*alloc)(size_t))
     phys_mem_alloc = alloc;
 }
 
-static uint16_t phys_section_add(MemoryRegionSection *section)
+static uint16_t phys_section_add(PhysPageMap *map,
+                                 MemoryRegionSection *section)
 {
     /* The physical section number is ORed with a page-aligned
      * pointer to produce the iotlb entries.  Thus it should
      * never overflow into the page-aligned value.
      */
-    assert(next_map.sections_nb < TARGET_PAGE_SIZE);
+    assert(map->sections_nb < TARGET_PAGE_SIZE);
 
-    if (next_map.sections_nb == next_map.sections_nb_alloc) {
-        next_map.sections_nb_alloc = MAX(next_map.sections_nb_alloc * 2,
-                                         16);
-        next_map.sections = g_renew(MemoryRegionSection, next_map.sections,
-                                    next_map.sections_nb_alloc);
+    if (map->sections_nb == map->sections_nb_alloc) {
+        map->sections_nb_alloc = MAX(map->sections_nb_alloc * 2, 16);
+        map->sections = g_renew(MemoryRegionSection, map->sections,
+                                map->sections_nb_alloc);
     }
-    next_map.sections[next_map.sections_nb] = *section;
+    map->sections[map->sections_nb] = *section;
     memory_region_ref(section->mr);
-    return next_map.sections_nb++;
+    return map->sections_nb++;
 }
 
 static void phys_section_destroy(MemoryRegion *mr)
@@ -785,7 +855,6 @@ static void phys_sections_free(PhysPageMap *map)
     }
     g_free(map->sections);
     g_free(map->nodes);
-    g_free(map);
 }
 
 static void register_subpage(AddressSpaceDispatch *d, MemoryRegionSection *section)
@@ -793,8 +862,8 @@ static void register_subpage(AddressSpaceDispatch *d, MemoryRegionSection *secti
     subpage_t *subpage;
     hwaddr base = section->offset_within_address_space
         & TARGET_PAGE_MASK;
-    MemoryRegionSection *existing = phys_page_find(d->phys_map, base >> TARGET_PAGE_BITS,
-                                                   next_map.nodes, next_map.sections);
+    MemoryRegionSection *existing = phys_page_find(d->phys_map, base,
+                                                   d->map.nodes, d->map.sections);
     MemoryRegionSection subsection = {
         .offset_within_address_space = base,
         .size = int128_make64(TARGET_PAGE_SIZE),
@@ -807,13 +876,14 @@ static void register_subpage(AddressSpaceDispatch *d, MemoryRegionSection *secti
         subpage = subpage_init(d->as, base);
         subsection.mr = &subpage->iomem;
         phys_page_set(d, base >> TARGET_PAGE_BITS, 1,
-                      phys_section_add(&subsection));
+                      phys_section_add(&d->map, &subsection));
     } else {
         subpage = container_of(existing->mr, subpage_t, iomem);
     }
     start = section->offset_within_address_space & ~TARGET_PAGE_MASK;
     end = start + int128_get64(section->size) - 1;
-    subpage_register(subpage, start, end, phys_section_add(section));
+    subpage_register(subpage, start, end,
+                     phys_section_add(&d->map, section));
 }
 
 
@@ -821,7 +891,7 @@ static void register_multipage(AddressSpaceDispatch *d,
                                MemoryRegionSection *section)
 {
     hwaddr start_addr = section->offset_within_address_space;
-    uint16_t section_index = phys_section_add(section);
+    uint16_t section_index = phys_section_add(&d->map, section);
     uint64_t num_pages = int128_get64(int128_rshift(section->size,
                                                     TARGET_PAGE_BITS));
 
@@ -904,6 +974,13 @@ static long gethugepagesize(const char *path)
     return fs.f_bsize;
 }
 
+static sigjmp_buf sigjump;
+
+static void sigbus_handler(int signal)
+{
+    siglongjmp(sigjump, 1);
+}
+
 static void *file_ram_alloc(RAMBlock *block,
                             ram_addr_t memory,
                             const char *path)
@@ -913,9 +990,6 @@ static void *file_ram_alloc(RAMBlock *block,
     char *c;
     void *area;
     int fd;
-#ifdef MAP_POPULATE
-    int flags;
-#endif
     unsigned long hpagesize;
 
     hpagesize = gethugepagesize(path);
@@ -963,21 +1037,52 @@ static void *file_ram_alloc(RAMBlock *block,
     if (ftruncate(fd, memory))
         perror("ftruncate");
 
-#ifdef MAP_POPULATE
-    /* NB: MAP_POPULATE won't exhaustively alloc all phys pages in the case
-     * MAP_PRIVATE is requested.  For mem_prealloc we mmap as MAP_SHARED
-     * to sidestep this quirk.
-     */
-    flags = mem_prealloc ? MAP_POPULATE | MAP_SHARED : MAP_PRIVATE;
-    area = mmap(0, memory, PROT_READ | PROT_WRITE, flags, fd, 0);
-#else
     area = mmap(0, memory, PROT_READ | PROT_WRITE, MAP_PRIVATE, fd, 0);
-#endif
     if (area == MAP_FAILED) {
         perror("file_ram_alloc: can't mmap RAM pages");
         close(fd);
         return (NULL);
     }
+
+    if (mem_prealloc) {
+        int ret, i;
+        struct sigaction act, oldact;
+        sigset_t set, oldset;
+
+        memset(&act, 0, sizeof(act));
+        act.sa_handler = &sigbus_handler;
+        act.sa_flags = 0;
+
+        ret = sigaction(SIGBUS, &act, &oldact);
+        if (ret) {
+            perror("file_ram_alloc: failed to install signal handler");
+            exit(1);
+        }
+
+        /* unblock SIGBUS */
+        sigemptyset(&set);
+        sigaddset(&set, SIGBUS);
+        pthread_sigmask(SIG_UNBLOCK, &set, &oldset);
+
+        if (sigsetjmp(sigjump, 1)) {
+            fprintf(stderr, "file_ram_alloc: failed to preallocate pages\n");
+            exit(1);
+        }
+
+        /* MAP_POPULATE silently ignores failures */
+        for (i = 0; i < (memory/hpagesize)-1; i++) {
+            memset(area + (hpagesize*i), 0, 1);
+        }
+
+        ret = sigaction(SIGBUS, &oldact, NULL);
+        if (ret) {
+            perror("file_ram_alloc: failed to reinstall signal handler");
+            exit(1);
+        }
+
+        pthread_sigmask(SIG_SETMASK, &oldset, NULL);
+    }
+
     block->fd = fd;
     return area;
 }
@@ -1102,6 +1207,9 @@ ram_addr_t qemu_ram_alloc_from_ptr(ram_addr_t size, void *host,
                                    MemoryRegion *mr)
 {
     RAMBlock *block, *new_block;
+    ram_addr_t old_ram_size, new_ram_size;
+
+    old_ram_size = last_ram_offset() >> TARGET_PAGE_BITS;
 
     size = TARGET_PAGE_ALIGN(size);
     new_block = g_malloc0(sizeof(*new_block));
@@ -1162,11 +1270,17 @@ ram_addr_t qemu_ram_alloc_from_ptr(ram_addr_t size, void *host,
     ram_list.version++;
     qemu_mutex_unlock_ramlist();
 
-    ram_list.phys_dirty = g_realloc(ram_list.phys_dirty,
-                                       last_ram_offset() >> TARGET_PAGE_BITS);
-    memset(ram_list.phys_dirty + (new_block->offset >> TARGET_PAGE_BITS),
-           0, size >> TARGET_PAGE_BITS);
-    cpu_physical_memory_set_dirty_range(new_block->offset, size, 0xff);
+    new_ram_size = last_ram_offset() >> TARGET_PAGE_BITS;
+
+    if (new_ram_size > old_ram_size) {
+        int i;
+        for (i = 0; i < DIRTY_MEMORY_NUM; i++) {
+            ram_list.dirty_memory[i] =
+                bitmap_zero_extend(ram_list.dirty_memory[i],
+                                   old_ram_size, new_ram_size);
+       }
+    }
+    cpu_physical_memory_set_dirty_range(new_block->offset, size);
 
     qemu_ram_setup_dump(new_block->host, size);
     qemu_madvise(new_block->host, size, QEMU_MADV_HUGEPAGE);
@@ -1376,11 +1490,8 @@ found:
 static void notdirty_mem_write(void *opaque, hwaddr ram_addr,
                                uint64_t val, unsigned size)
 {
-    int dirty_flags;
-    dirty_flags = cpu_physical_memory_get_dirty_flags(ram_addr);
-    if (!(dirty_flags & CODE_DIRTY_FLAG)) {
+    if (!cpu_physical_memory_get_dirty_flag(ram_addr, DIRTY_MEMORY_CODE)) {
         tb_invalidate_phys_page_fast(ram_addr, size);
-        dirty_flags = cpu_physical_memory_get_dirty_flags(ram_addr);
     }
     switch (size) {
     case 1:
@@ -1395,11 +1506,11 @@ static void notdirty_mem_write(void *opaque, hwaddr ram_addr,
     default:
         abort();
     }
-    dirty_flags |= (0xff & ~CODE_DIRTY_FLAG);
-    cpu_physical_memory_set_dirty_flags(ram_addr, dirty_flags);
+    cpu_physical_memory_set_dirty_flag(ram_addr, DIRTY_MEMORY_MIGRATION);
+    cpu_physical_memory_set_dirty_flag(ram_addr, DIRTY_MEMORY_VGA);
     /* we remove the notdirty callback only if the code has been
        flushed */
-    if (dirty_flags == 0xff) {
+    if (!cpu_physical_memory_is_clean(ram_addr)) {
         CPUArchState *env = current_cpu->env_ptr;
         tlb_set_dirty(env, env->mem_io_vaddr);
     }
@@ -1605,7 +1716,7 @@ static subpage_t *subpage_init(AddressSpace *as, hwaddr base)
     return mmio;
 }
 
-static uint16_t dummy_section(MemoryRegion *mr)
+static uint16_t dummy_section(PhysPageMap *map, MemoryRegion *mr)
 {
     MemoryRegionSection section = {
         .mr = mr,
@@ -1614,12 +1725,13 @@ static uint16_t dummy_section(MemoryRegion *mr)
         .size = int128_2_64(),
     };
 
-    return phys_section_add(&section);
+    return phys_section_add(map, &section);
 }
 
 MemoryRegion *iotlb_to_region(hwaddr index)
 {
-    return address_space_memory.dispatch->sections[index & ~TARGET_PAGE_MASK].mr;
+    return address_space_memory.dispatch->map.sections[
+           index & ~TARGET_PAGE_MASK].mr;
 }
 
 static void io_mem_init(void)
@@ -1636,9 +1748,19 @@ static void io_mem_init(void)
 static void mem_begin(MemoryListener *listener)
 {
     AddressSpace *as = container_of(listener, AddressSpace, dispatch_listener);
-    AddressSpaceDispatch *d = g_new(AddressSpaceDispatch, 1);
+    AddressSpaceDispatch *d = g_new0(AddressSpaceDispatch, 1);
+    uint16_t n;
 
-    d->phys_map  = (PhysPageEntry) { .ptr = PHYS_MAP_NODE_NIL, .is_leaf = 0 };
+    n = dummy_section(&d->map, &io_mem_unassigned);
+    assert(n == PHYS_SECTION_UNASSIGNED);
+    n = dummy_section(&d->map, &io_mem_notdirty);
+    assert(n == PHYS_SECTION_NOTDIRTY);
+    n = dummy_section(&d->map, &io_mem_rom);
+    assert(n == PHYS_SECTION_ROM);
+    n = dummy_section(&d->map, &io_mem_watch);
+    assert(n == PHYS_SECTION_WATCH);
+
+    d->phys_map  = (PhysPageEntry) { .ptr = PHYS_MAP_NODE_NIL, .skip = 1 };
     d->as = as;
     as->next_dispatch = d;
 }
@@ -1649,37 +1771,14 @@ static void mem_commit(MemoryListener *listener)
     AddressSpaceDispatch *cur = as->dispatch;
     AddressSpaceDispatch *next = as->next_dispatch;
 
-    next->nodes = next_map.nodes;
-    next->sections = next_map.sections;
+    phys_page_compact_all(next, next->map.nodes_nb);
 
     as->dispatch = next;
-    g_free(cur);
-}
 
-static void core_begin(MemoryListener *listener)
-{
-    uint16_t n;
-
-    prev_map = g_new(PhysPageMap, 1);
-    *prev_map = next_map;
-
-    memset(&next_map, 0, sizeof(next_map));
-    n = dummy_section(&io_mem_unassigned);
-    assert(n == PHYS_SECTION_UNASSIGNED);
-    n = dummy_section(&io_mem_notdirty);
-    assert(n == PHYS_SECTION_NOTDIRTY);
-    n = dummy_section(&io_mem_rom);
-    assert(n == PHYS_SECTION_ROM);
-    n = dummy_section(&io_mem_watch);
-    assert(n == PHYS_SECTION_WATCH);
-}
-
-/* This listener's commit run after the other AddressSpaceDispatch listeners'.
- * All AddressSpaceDispatch instances have switched to the next map.
- */
-static void core_commit(MemoryListener *listener)
-{
-    phys_sections_free(prev_map);
+    if (cur) {
+        phys_sections_free(&cur->map);
+        g_free(cur);
+    }
 }
 
 static void tcg_commit(MemoryListener *listener)
@@ -1698,17 +1797,15 @@ static void tcg_commit(MemoryListener *listener)
 
 static void core_log_global_start(MemoryListener *listener)
 {
-    cpu_physical_memory_set_dirty_tracking(1);
+    cpu_physical_memory_set_dirty_tracking(true);
 }
 
 static void core_log_global_stop(MemoryListener *listener)
 {
-    cpu_physical_memory_set_dirty_tracking(0);
+    cpu_physical_memory_set_dirty_tracking(false);
 }
 
 static MemoryListener core_memory_listener = {
-    .begin = core_begin,
-    .commit = core_commit,
     .log_global_start = core_log_global_start,
     .log_global_stop = core_log_global_stop,
     .priority = 1,
@@ -1743,7 +1840,8 @@ void address_space_destroy_dispatch(AddressSpace *as)
 static void memory_map_init(void)
 {
     system_memory = g_malloc(sizeof(*system_memory));
-    memory_region_init(system_memory, NULL, "system", INT64_MAX);
+
+    memory_region_init(system_memory, NULL, "system", UINT64_MAX);
     address_space_init(&address_space_memory, system_memory, "memory");
 
     system_io = g_malloc(sizeof(*system_io));
@@ -1815,11 +1913,12 @@ int cpu_memory_rw_debug(CPUState *cpu, target_ulong addr,
 static void invalidate_and_set_dirty(hwaddr addr,
                                      hwaddr length)
 {
-    if (!cpu_physical_memory_is_dirty(addr)) {
+    if (cpu_physical_memory_is_clean(addr)) {
         /* invalidate code */
         tb_invalidate_phys_page_range(addr, addr + length, 0);
         /* set dirty bit */
-        cpu_physical_memory_set_dirty_flags(addr, (0xff & ~CODE_DIRTY_FLAG));
+        cpu_physical_memory_set_dirty_flag(addr, DIRTY_MEMORY_VGA);
+        cpu_physical_memory_set_dirty_flag(addr, DIRTY_MEMORY_MIGRATION);
     }
     xen_modified_memory(addr, length);
 }
@@ -1975,9 +2074,13 @@ void cpu_physical_memory_rw(hwaddr addr, uint8_t *buf,
     address_space_rw(&address_space_memory, addr, buf, len, is_write);
 }
 
-/* used for ROM loading : can write in RAM and ROM */
-void cpu_physical_memory_write_rom(hwaddr addr,
-                                   const uint8_t *buf, int len)
+enum write_rom_type {
+    WRITE_DATA,
+    FLUSH_CACHE,
+};
+
+static inline void cpu_physical_memory_write_rom_internal(
+    hwaddr addr, const uint8_t *buf, int len, enum write_rom_type type)
 {
     hwaddr l;
     uint8_t *ptr;
@@ -1996,8 +2099,15 @@ void cpu_physical_memory_write_rom(hwaddr addr,
             addr1 += memory_region_get_ram_addr(mr);
             /* ROM/RAM case */
             ptr = qemu_get_ram_ptr(addr1);
-            memcpy(ptr, buf, l);
-            invalidate_and_set_dirty(addr1, l);
+            switch (type) {
+            case WRITE_DATA:
+                memcpy(ptr, buf, l);
+                invalidate_and_set_dirty(addr1, l);
+                break;
+            case FLUSH_CACHE:
+                flush_icache_range((uintptr_t)ptr, (uintptr_t)ptr + l);
+                break;
+            }
         }
         len -= l;
         buf += l;
@@ -2005,6 +2115,28 @@ void cpu_physical_memory_write_rom(hwaddr addr,
     }
 }
 
+/* used for ROM loading : can write in RAM and ROM */
+void cpu_physical_memory_write_rom(hwaddr addr,
+                                   const uint8_t *buf, int len)
+{
+    cpu_physical_memory_write_rom_internal(addr, buf, len, WRITE_DATA);
+}
+
+void cpu_flush_icache_range(hwaddr start, int len)
+{
+    /*
+     * This function should do the same thing as an icache flush that was
+     * triggered from within the guest. For TCG we are always cache coherent,
+     * so there is no need to flush anything. For KVM / Xen we need to flush
+     * the host's instruction cache at least.
+     */
+    if (tcg_enabled()) {
+        return;
+    }
+
+    cpu_physical_memory_write_rom_internal(start, NULL, len, FLUSH_CACHE);
+}
+
 typedef struct {
     MemoryRegion *mr;
     void *buffer;
@@ -2397,12 +2529,13 @@ void stl_phys_notdirty(hwaddr addr, uint32_t val)
         stl_p(ptr, val);
 
         if (unlikely(in_migration)) {
-            if (!cpu_physical_memory_is_dirty(addr1)) {
+            if (cpu_physical_memory_is_clean(addr1)) {
                 /* invalidate code */
                 tb_invalidate_phys_page_range(addr1, addr1 + 4, 0);
                 /* set dirty bit */
-                cpu_physical_memory_set_dirty_flags(
-                    addr1, (0xff & ~CODE_DIRTY_FLAG));
+                cpu_physical_memory_set_dirty_flag(addr1,
+                                                   DIRTY_MEMORY_MIGRATION);
+                cpu_physical_memory_set_dirty_flag(addr1, DIRTY_MEMORY_VGA);
             }
         }
     }

gdb-xml/aarch64-fpu.xml

@@ -0,0 +1,86 @@
+<?xml version="1.0"?>
+<!-- Copyright (C) 2009-2012 Free Software Foundation, Inc.
+     Contributed by ARM Ltd.
+
+     Copying and distribution of this file, with or without modification,
+     are permitted in any medium without royalty provided the copyright
+     notice and this notice are preserved.  -->
+
+<!DOCTYPE feature SYSTEM "gdb-target.dtd">
+<feature name="org.gnu.gdb.aarch64.fpu">
+  <vector id="v2d" type="ieee_double" count="2"/>
+  <vector id="v2u" type="uint64" count="2"/>
+  <vector id="v2i" type="int64" count="2"/>
+  <vector id="v4f" type="ieee_single" count="4"/>
+  <vector id="v4u" type="uint32" count="4"/>
+  <vector id="v4i" type="int32" count="4"/>
+  <vector id="v8u" type="uint16" count="8"/>
+  <vector id="v8i" type="int16" count="8"/>
+  <vector id="v16u" type="uint8" count="16"/>
+  <vector id="v16i" type="int8" count="16"/>
+  <vector id="v1u" type="uint128" count="1"/>
+  <vector id="v1i" type="int128" count="1"/>
+  <union id="vnd">
+    <field name="f" type="v2d"/>
+    <field name="u" type="v2u"/>
+    <field name="s" type="v2i"/>
+  </union>
+  <union id="vns">
+    <field name="f" type="v4f"/>
+    <field name="u" type="v4u"/>
+    <field name="s" type="v4i"/>
+  </union>
+  <union id="vnh">
+    <field name="u" type="v8u"/>
+    <field name="s" type="v8i"/>
+  </union>
+  <union id="vnb">
+    <field name="u" type="v16u"/>
+    <field name="s" type="v16i"/>
+  </union>
+  <union id="vnq">
+    <field name="u" type="v1u"/>
+    <field name="s" type="v1i"/>
+  </union>
+  <union id="aarch64v">
+    <field name="d" type="vnd"/>
+    <field name="s" type="vns"/>
+    <field name="h" type="vnh"/>
+    <field name="b" type="vnb"/>
+    <field name="q" type="vnq"/>
+  </union>
+  <reg name="v0" bitsize="128" type="aarch64v" regnum="34"/>
+  <reg name="v1" bitsize="128" type="aarch64v" />
+  <reg name="v2" bitsize="128" type="aarch64v" />
+  <reg name="v3" bitsize="128" type="aarch64v" />
+  <reg name="v4" bitsize="128" type="aarch64v" />
+  <reg name="v5" bitsize="128" type="aarch64v" />
+  <reg name="v6" bitsize="128" type="aarch64v" />
+  <reg name="v7" bitsize="128" type="aarch64v" />
+  <reg name="v8" bitsize="128" type="aarch64v" />
+  <reg name="v9" bitsize="128" type="aarch64v" />
+  <reg name="v10" bitsize="128" type="aarch64v"/>
+  <reg name="v11" bitsize="128" type="aarch64v"/>
+  <reg name="v12" bitsize="128" type="aarch64v"/>
+  <reg name="v13" bitsize="128" type="aarch64v"/>
+  <reg name="v14" bitsize="128" type="aarch64v"/>
+  <reg name="v15" bitsize="128" type="aarch64v"/>
+  <reg name="v16" bitsize="128" type="aarch64v"/>
+  <reg name="v17" bitsize="128" type="aarch64v"/>
+  <reg name="v18" bitsize="128" type="aarch64v"/>
+  <reg name="v19" bitsize="128" type="aarch64v"/>
+  <reg name="v20" bitsize="128" type="aarch64v"/>
+  <reg name="v21" bitsize="128" type="aarch64v"/>
+  <reg name="v22" bitsize="128" type="aarch64v"/>
+  <reg name="v23" bitsize="128" type="aarch64v"/>
+  <reg name="v24" bitsize="128" type="aarch64v"/>
+  <reg name="v25" bitsize="128" type="aarch64v"/>
+  <reg name="v26" bitsize="128" type="aarch64v"/>
+  <reg name="v27" bitsize="128" type="aarch64v"/>
+  <reg name="v28" bitsize="128" type="aarch64v"/>
+  <reg name="v29" bitsize="128" type="aarch64v"/>
+  <reg name="v30" bitsize="128" type="aarch64v"/>
+  <reg name="v31" bitsize="128" type="aarch64v"/>
+  <reg name="fpsr" bitsize="32"/>
+  <reg name="fpcr" bitsize="32"/>
+</feature>

hmp-commands.hx

@@ -1190,7 +1190,7 @@ ETEXI
     {
         .name       = "host_net_add",
         .args_type  = "device:s,opts:s?",
-        .params     = "tap|user|socket|vde|dump [options]",
+        .params     = "tap|user|socket|vde|netmap|dump [options]",
         .help       = "add host VLAN client",
         .mhandler.cmd = net_host_device_add,
     },
@@ -1218,7 +1218,7 @@ ETEXI
     {
         .name       = "netdev_add",
         .args_type  = "netdev:O",
-        .params     = "[user|tap|socket|hubport],id=str[,prop=value][,...]",
+        .params     = "[user|tap|socket|hubport|netmap],id=str[,prop=value][,...]",
         .help       = "add host network device",
         .mhandler.cmd = hmp_netdev_add,
     },
@@ -1241,6 +1241,34 @@ STEXI
 @item netdev_del
 @findex netdev_del
 Remove host network device.
+ETEXI
+
+    {
+        .name       = "object_add",
+        .args_type  = "object:O",
+        .params     = "[qom-type=]type,id=str[,prop=value][,...]",
+        .help       = "create QOM object",
+        .mhandler.cmd = hmp_object_add,
+    },
+
+STEXI
+@item object_add
+@findex object_add
+Create QOM object.
+ETEXI
+
+    {
+        .name       = "object_del",
+        .args_type  = "id:s",
+        .params     = "id",
+        .help       = "destroy QOM object",
+        .mhandler.cmd = hmp_object_del,
+    },
+
+STEXI
+@item object_del
+@findex object_del
+Destroy QOM object.
 ETEXI
 
 #ifdef CONFIG_SLIRP
@@ -1617,6 +1645,19 @@ STEXI
 
 Executes a qemu-io command on the given block device.
 
+ETEXI
+
+    {
+        .name       = "cpu-add",
+        .args_type  = "id:i",
+        .params     = "id",
+        .help       = "add cpu",
+        .mhandler.cmd  = hmp_cpu_add,
+    },
+
+STEXI
+@item cpu-add @var{id}
+Add CPU with id @var{id}
 ETEXI
 
     {

hmp.c

@@ -21,6 +21,7 @@
 #include "qmp-commands.h"
 #include "qemu/sockets.h"
 #include "monitor/monitor.h"
+#include "qapi/opts-visitor.h"
 #include "ui/console.h"
 #include "block/qapi.h"
 #include "qemu-io.h"
@@ -1354,6 +1355,63 @@ void hmp_netdev_del(Monitor *mon, const QDict *qdict)
     hmp_handle_error(mon, &err);
 }
 
+void hmp_object_add(Monitor *mon, const QDict *qdict)
+{
+    Error *err = NULL;
+    QemuOpts *opts;
+    char *type = NULL;
+    char *id = NULL;
+    void *dummy = NULL;
+    OptsVisitor *ov;
+    QDict *pdict;
+
+    opts = qemu_opts_from_qdict(qemu_find_opts("object"), qdict, &err);
+    if (err) {
+        goto out;
+    }
+
+    ov = opts_visitor_new(opts);
+    pdict = qdict_clone_shallow(qdict);
+
+    visit_start_struct(opts_get_visitor(ov), &dummy, NULL, NULL, 0, &err);
+    if (err) {
+        goto out_clean;
+    }
+
+    qdict_del(pdict, "qom-type");
+    visit_type_str(opts_get_visitor(ov), &type, "qom-type", &err);
+    if (err) {
+        goto out_clean;
+    }
+
+    qdict_del(pdict, "id");
+    visit_type_str(opts_get_visitor(ov), &id, "id", &err);
+    if (err) {
+        goto out_clean;
+    }
+
+    object_add(type, id, pdict, opts_get_visitor(ov), &err);
+    if (err) {
+        goto out_clean;
+    }
+    visit_end_struct(opts_get_visitor(ov), &err);
+    if (err) {
+        qmp_object_del(id, NULL);
+    }
+
+out_clean:
+    opts_visitor_cleanup(ov);
+
+    QDECREF(pdict);
+    qemu_opts_del(opts);
+    g_free(id);
+    g_free(type);
+    g_free(dummy);
+
+out:
+    hmp_handle_error(mon, &err);
+}
+
 void hmp_getfd(Monitor *mon, const QDict *qdict)
 {
     const char *fdname = qdict_get_str(qdict, "fdname");
@@ -1525,6 +1583,16 @@ void hmp_nbd_server_stop(Monitor *mon, const QDict *qdict)
     hmp_handle_error(mon, &errp);
 }
 
+void hmp_cpu_add(Monitor *mon, const QDict *qdict)
+{
+    int cpuid;
+    Error *err = NULL;
+
+    cpuid = qdict_get_int(qdict, "id");
+    qmp_cpu_add(cpuid, &err);
+    hmp_handle_error(mon, &err);
+}
+
 void hmp_chardev_add(Monitor *mon, const QDict *qdict)
 {
     const char *args = qdict_get_str(qdict, "args");
@@ -1564,3 +1632,12 @@ void hmp_qemu_io(Monitor *mon, const QDict *qdict)
 
     hmp_handle_error(mon, &err);
 }
+
+void hmp_object_del(Monitor *mon, const QDict *qdict)
+{
+    const char *id = qdict_get_str(qdict, "id");
+    Error *err = NULL;
+
+    qmp_object_del(id, &err);
+    hmp_handle_error(mon, &err);
+}

hmp.h

@@ -89,5 +89,8 @@ void hmp_nbd_server_stop(Monitor *mon, const QDict *qdict);
 void hmp_chardev_add(Monitor *mon, const QDict *qdict);
 void hmp_chardev_remove(Monitor *mon, const QDict *qdict);
 void hmp_qemu_io(Monitor *mon, const QDict *qdict);
+void hmp_cpu_add(Monitor *mon, const QDict *qdict);
+void hmp_object_add(Monitor *mon, const QDict *qdict);
+void hmp_object_del(Monitor *mon, const QDict *qdict);
 
 #endif

hw/9pfs/virtio-9p-device.c

@@ -41,15 +41,16 @@ static void virtio_9p_get_config(VirtIODevice *vdev, uint8_t *config)
     g_free(cfg);
 }
 
-static int virtio_9p_device_init(VirtIODevice *vdev)
+static void virtio_9p_device_realize(DeviceState *dev, Error **errp)
 {
-    V9fsState *s = VIRTIO_9P(vdev);
+    VirtIODevice *vdev = VIRTIO_DEVICE(dev);
+    V9fsState *s = VIRTIO_9P(dev);
     int i, len;
     struct stat stat;
     FsDriverEntry *fse;
     V9fsPath path;
 
-    virtio_init(VIRTIO_DEVICE(s), "virtio-9p", VIRTIO_ID_9P,
+    virtio_init(vdev, "virtio-9p", VIRTIO_ID_9P,
                 sizeof(struct virtio_9p_config) + MAX_TAG_LEN);
 
     /* initialize pdu allocator */
@@ -67,16 +68,16 @@ static int virtio_9p_device_init(VirtIODevice *vdev)
 
     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",
-                s->fsconf.fsdev_id ? s->fsconf.fsdev_id : "NULL");
+        error_setg(errp, "Virtio-9p device couldn't find fsdev with the "
+                   "id = %s",
+                   s->fsconf.fsdev_id ? s->fsconf.fsdev_id : "NULL");
         goto out;
     }
 
     if (!s->fsconf.tag) {
         /* we haven't specified a mount_tag */
-        fprintf(stderr, "fsdev with id %s needs mount_tag arguments\n",
-                s->fsconf.fsdev_id);
+        error_setg(errp, "fsdev with id %s needs mount_tag arguments",
+                   s->fsconf.fsdev_id);
         goto out;
     }
 
@@ -85,8 +86,8 @@ static int virtio_9p_device_init(VirtIODevice *vdev)
     s->ctx.exops.get_st_gen = NULL;
     len = strlen(s->fsconf.tag);
     if (len > MAX_TAG_LEN - 1) {
-        fprintf(stderr, "mount tag '%s' (%d bytes) is longer than "
-                "maximum (%d bytes)", s->fsconf.tag, len, MAX_TAG_LEN - 1);
+        error_setg(errp, "mount tag '%s' (%d bytes) is longer than "
+                   "maximum (%d bytes)", s->fsconf.tag, len, MAX_TAG_LEN - 1);
         goto out;
     }
 
@@ -99,12 +100,12 @@ static int virtio_9p_device_init(VirtIODevice *vdev)
     qemu_co_rwlock_init(&s->rename_lock);
 
     if (s->ops->init(&s->ctx) < 0) {
-        fprintf(stderr, "Virtio-9p Failed to initialize fs-driver with id:%s"
-                " and export path:%s\n", s->fsconf.fsdev_id, s->ctx.fs_root);
+        error_setg(errp, "Virtio-9p Failed to initialize fs-driver with id:%s"
+                   " and export path:%s", s->fsconf.fsdev_id, s->ctx.fs_root);
         goto out;
     }
     if (v9fs_init_worker_threads() < 0) {
-        fprintf(stderr, "worker thread initialization failed\n");
+        error_setg(errp, "worker thread initialization failed");
         goto out;
     }
 
@@ -114,28 +115,25 @@ static int virtio_9p_device_init(VirtIODevice *vdev)
      * use co-routines here.
      */
     if (s->ops->name_to_path(&s->ctx, NULL, "/", &path) < 0) {
-        fprintf(stderr,
-                "error in converting name to path %s", strerror(errno));
+        error_setg(errp,
+                   "error in converting name to path %s", strerror(errno));
         goto out;
     }
     if (s->ops->lstat(&s->ctx, &path, &stat)) {
-        fprintf(stderr, "share path %s does not exist\n", fse->path);
+        error_setg(errp, "share path %s does not exist", fse->path);
         goto out;
     } else if (!S_ISDIR(stat.st_mode)) {
-        fprintf(stderr, "share path %s is not a directory\n", fse->path);
+        error_setg(errp, "share path %s is not a directory", fse->path);
         goto out;
     }
     v9fs_path_free(&path);
 
-    return 0;
+    return;
 out:
     g_free(s->ctx.fs_root);
     g_free(s->tag);
     virtio_cleanup(vdev);
     v9fs_path_free(&path);
-
-    return -1;
-
 }
 
 /* virtio-9p device */
@@ -149,9 +147,10 @@ static void virtio_9p_class_init(ObjectClass *klass, void *data)
 {
     DeviceClass *dc = DEVICE_CLASS(klass);
     VirtioDeviceClass *vdc = VIRTIO_DEVICE_CLASS(klass);
+
     dc->props = virtio_9p_properties;
     set_bit(DEVICE_CATEGORY_STORAGE, dc->categories);
-    vdc->init = virtio_9p_device_init;
+    vdc->realize = virtio_9p_device_realize;
     vdc->get_features = virtio_9p_get_features;
     vdc->get_config = virtio_9p_get_config;
 }

hw/acpi/core.c

@@ -662,3 +662,21 @@ uint32_t acpi_gpe_ioport_readb(ACPIREGS *ar, uint32_t addr)
 
     return val;
 }
+
+void acpi_update_sci(ACPIREGS *regs, qemu_irq irq)
+{
+    int sci_level, pm1a_sts;
+
+    pm1a_sts = acpi_pm1_evt_get_sts(regs);
+
+    sci_level = ((pm1a_sts &
+                  regs->pm1.evt.en & ACPI_BITMASK_PM1_COMMON_ENABLED) != 0) ||
+                ((regs->gpe.sts[0] & regs->gpe.en[0]) != 0);
+
+    qemu_set_irq(irq, sci_level);
+
+    /* schedule a timer interruption if needed */
+    acpi_pm_tmr_update(regs,
+                       (regs->pm1.evt.en & ACPI_BITMASK_TIMER_ENABLE) &&
+                       !(pm1a_sts & ACPI_BITMASK_TIMER_STATUS));
+}

hw/acpi/ich9.c

@@ -44,29 +44,10 @@ do { printf("%s "fmt, __func__, ## __VA_ARGS__); } while (0)
 #define ICH9_DEBUG(fmt, ...)    do { } while (0)
 #endif
 
-static void pm_update_sci(ICH9LPCPMRegs *pm)
-{
-    int sci_level, pm1a_sts;
-
-    pm1a_sts = acpi_pm1_evt_get_sts(&pm->acpi_regs);
-
-    sci_level = (((pm1a_sts & pm->acpi_regs.pm1.evt.en) &
-                  (ACPI_BITMASK_RT_CLOCK_ENABLE |
-                   ACPI_BITMASK_POWER_BUTTON_ENABLE |
-                   ACPI_BITMASK_GLOBAL_LOCK_ENABLE |
-                   ACPI_BITMASK_TIMER_ENABLE)) != 0);
-    qemu_set_irq(pm->irq, sci_level);
-
-    /* schedule a timer interruption if needed */
-    acpi_pm_tmr_update(&pm->acpi_regs,
-                       (pm->acpi_regs.pm1.evt.en & ACPI_BITMASK_TIMER_ENABLE) &&
-                       !(pm1a_sts & ACPI_BITMASK_TIMER_STATUS));
-}
-
 static void ich9_pm_update_sci_fn(ACPIREGS *regs)
 {
     ICH9LPCPMRegs *pm = container_of(regs, ICH9LPCPMRegs, acpi_regs);
-    pm_update_sci(pm);
+    acpi_update_sci(&pm->acpi_regs, pm->irq);
 }
 
 static uint64_t ich9_gpe_readb(void *opaque, hwaddr addr, unsigned width)
@@ -80,6 +61,7 @@ static void ich9_gpe_writeb(void *opaque, hwaddr addr, uint64_t val,
 {
     ICH9LPCPMRegs *pm = opaque;
     acpi_gpe_ioport_writeb(&pm->acpi_regs, addr, val);
+    acpi_update_sci(&pm->acpi_regs, pm->irq);
 }
 
 static const MemoryRegionOps ich9_gpe_ops = {
@@ -193,7 +175,7 @@ static void pm_reset(void *opaque)
         pm->smi_en |= ICH9_PMIO_SMI_EN_APMC_EN;
     }
 
-    pm_update_sci(pm);
+    acpi_update_sci(&pm->acpi_regs, pm->irq);
 }
 
 static void pm_powerdown_req(Notifier *n, void *opaque)

hw/acpi/piix4.c

@@ -112,29 +112,10 @@ static void piix4_acpi_system_hot_add_init(MemoryRegion *parent,
 #define ACPI_ENABLE 0xf1
 #define ACPI_DISABLE 0xf0
 
-static void pm_update_sci(PIIX4PMState *s)
-{
-    int sci_level, pmsts;
-
-    pmsts = acpi_pm1_evt_get_sts(&s->ar);
-    sci_level = (((pmsts & s->ar.pm1.evt.en) &
-                  (ACPI_BITMASK_RT_CLOCK_ENABLE |
-                   ACPI_BITMASK_POWER_BUTTON_ENABLE |
-                   ACPI_BITMASK_GLOBAL_LOCK_ENABLE |
-                   ACPI_BITMASK_TIMER_ENABLE)) != 0) ||
-        (((s->ar.gpe.sts[0] & s->ar.gpe.en[0]) &
-          (PIIX4_PCI_HOTPLUG_STATUS | PIIX4_CPU_HOTPLUG_STATUS)) != 0);
-
-    qemu_set_irq(s->irq, sci_level);
-    /* schedule a timer interruption if needed */
-    acpi_pm_tmr_update(&s->ar, (s->ar.pm1.evt.en & ACPI_BITMASK_TIMER_ENABLE) &&
-                       !(pmsts & ACPI_BITMASK_TIMER_STATUS));
-}
-
 static void pm_tmr_timer(ACPIREGS *ar)
 {
     PIIX4PMState *s = container_of(ar, PIIX4PMState, ar);
-    pm_update_sci(s);
+    acpi_update_sci(&s->ar, s->irq);
 }
 
 static void apm_ctrl_changed(uint32_t val, void *arg)
@@ -544,9 +525,13 @@ static void piix4_pm_class_init(ObjectClass *klass, void *data)
     k->revision = 0x03;
     k->class_id = PCI_CLASS_BRIDGE_OTHER;
     dc->desc = "PM";
-    dc->no_user = 1;
     dc->vmsd = &vmstate_acpi;
     dc->props = piix4_pm_properties;
+    /*
+     * Reason: part of PIIX4 southbridge, needs to be wired up,
+     * e.g. by mips_malta_init()
+     */
+    dc->cannot_instantiate_with_device_add_yet = true;
 }
 
 static const TypeInfo piix4_pm_info = {
@@ -578,7 +563,7 @@ static void gpe_writeb(void *opaque, hwaddr addr, uint64_t val,
     PIIX4PMState *s = opaque;
 
     acpi_gpe_ioport_writeb(&s->ar, addr, val);
-    pm_update_sci(s);
+    acpi_update_sci(&s->ar, s->irq);
 
     PIIX4_DPRINTF("gpe write %" HWADDR_PRIx " <== %" PRIu64 "\n", addr, val);
 }
@@ -694,7 +679,7 @@ static void piix4_cpu_hotplug_req(PIIX4PMState *s, CPUState *cpu,
     } else {
         g->sts[cpu_id / 8] &= ~(1 << (cpu_id % 8));
     }
-    pm_update_sci(s);
+    acpi_update_sci(&s->ar, s->irq);
 }
 
 static void piix4_cpu_added_req(Notifier *n, void *opaque)
@@ -768,7 +753,7 @@ static int piix4_device_hotplug(DeviceState *qdev, PCIDevice *dev,
         disable_device(s, slot);
     }
 
-    pm_update_sci(s);
+    acpi_update_sci(&s->ar, s->irq);
 
     return 0;
 }

hw/alpha/typhoon.c

@@ -934,11 +934,9 @@ static int typhoon_pcihost_init(SysBusDevice *dev)
 
 static void typhoon_pcihost_class_init(ObjectClass *klass, void *data)
 {
-    DeviceClass *dc = DEVICE_CLASS(klass);
     SysBusDeviceClass *k = SYS_BUS_DEVICE_CLASS(klass);
 
     k->init = typhoon_pcihost_init;
-    dc->no_user = 1;
 }
 
 static const TypeInfo typhoon_pcihost_info = {

hw/arm/Makefile.objs

@@ -1,7 +1,10 @@
 obj-y += boot.o collie.o exynos4_boards.o gumstix.o highbank.o
+obj-$(CONFIG_DIGIC) += digic_boards.o
 obj-y += integratorcp.o kzm.o mainstone.o musicpal.o nseries.o
 obj-y += omap_sx1.o palm.o realview.o spitz.o stellaris.o
-obj-y += tosa.o versatilepb.o vexpress.o xilinx_zynq.o z2.o
+obj-y += tosa.o versatilepb.o vexpress.o virt.o xilinx_zynq.o z2.o
 
 obj-y += armv7m.o exynos4210.o pxa2xx.o pxa2xx_gpio.o pxa2xx_pic.o
+obj-$(CONFIG_DIGIC) += digic.o
 obj-y += omap1.o omap2.o strongarm.o
+obj-$(CONFIG_ALLWINNER_A10) += allwinner-a10.o cubieboard.o

hw/arm/allwinner-a10.c

@@ -0,0 +1,103 @@
+/*
+ * Allwinner A10 SoC emulation
+ *
+ * Copyright (C) 2013 Li Guang
+ * Written by Li Guang <lig.fnst@cn.fujitsu.com>
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License as published by the
+ * Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License
+ * for more details.
+ */
+
+#include "hw/sysbus.h"
+#include "hw/devices.h"
+#include "hw/arm/allwinner-a10.h"
+
+static void aw_a10_init(Object *obj)
+{
+    AwA10State *s = AW_A10(obj);
+
+    object_initialize(&s->cpu, sizeof(s->cpu), "cortex-a8-" TYPE_ARM_CPU);
+    object_property_add_child(obj, "cpu", OBJECT(&s->cpu), NULL);
+
+    object_initialize(&s->intc, sizeof(s->intc), TYPE_AW_A10_PIC);
+    qdev_set_parent_bus(DEVICE(&s->intc), sysbus_get_default());
+
+    object_initialize(&s->timer, sizeof(s->timer), TYPE_AW_A10_PIT);
+    qdev_set_parent_bus(DEVICE(&s->timer), sysbus_get_default());
+}
+
+static void aw_a10_realize(DeviceState *dev, Error **errp)
+{
+    AwA10State *s = AW_A10(dev);
+    SysBusDevice *sysbusdev;
+    uint8_t i;
+    qemu_irq fiq, irq;
+    Error *err = NULL;
+
+    object_property_set_bool(OBJECT(&s->cpu), true, "realized", &err);
+    if (err != NULL) {
+        error_propagate(errp, err);
+        return;
+    }
+    irq = qdev_get_gpio_in(DEVICE(&s->cpu), ARM_CPU_IRQ);
+    fiq = qdev_get_gpio_in(DEVICE(&s->cpu), ARM_CPU_FIQ);
+
+    object_property_set_bool(OBJECT(&s->intc), true, "realized", &err);
+    if (err != NULL) {
+        error_propagate(errp, err);
+        return;
+    }
+    sysbusdev = SYS_BUS_DEVICE(&s->intc);
+    sysbus_mmio_map(sysbusdev, 0, AW_A10_PIC_REG_BASE);
+    sysbus_connect_irq(sysbusdev, 0, irq);
+    sysbus_connect_irq(sysbusdev, 1, fiq);
+    for (i = 0; i < AW_A10_PIC_INT_NR; i++) {
+        s->irq[i] = qdev_get_gpio_in(DEVICE(&s->intc), i);
+    }
+
+    object_property_set_bool(OBJECT(&s->timer), true, "realized", &err);
+    if (err != NULL) {
+        error_propagate(errp, err);
+        return;
+    }
+    sysbusdev = SYS_BUS_DEVICE(&s->timer);
+    sysbus_mmio_map(sysbusdev, 0, AW_A10_PIT_REG_BASE);
+    sysbus_connect_irq(sysbusdev, 0, s->irq[22]);
+    sysbus_connect_irq(sysbusdev, 1, s->irq[23]);
+    sysbus_connect_irq(sysbusdev, 2, s->irq[24]);
+    sysbus_connect_irq(sysbusdev, 3, s->irq[25]);
+    sysbus_connect_irq(sysbusdev, 4, s->irq[67]);
+    sysbus_connect_irq(sysbusdev, 5, s->irq[68]);
+
+    serial_mm_init(get_system_memory(), AW_A10_UART0_REG_BASE, 2, s->irq[1],
+                   115200, serial_hds[0], DEVICE_NATIVE_ENDIAN);
+}
+
+static void aw_a10_class_init(ObjectClass *oc, void *data)
+{
+    DeviceClass *dc = DEVICE_CLASS(oc);
+
+    dc->realize = aw_a10_realize;
+}
+
+static const TypeInfo aw_a10_type_info = {
+    .name = TYPE_AW_A10,
+    .parent = TYPE_DEVICE,
+    .instance_size = sizeof(AwA10State),
+    .instance_init = aw_a10_init,
+    .class_init = aw_a10_class_init,
+};
+
+static void aw_a10_register_types(void)
+{
+    type_register_static(&aw_a10_type_info);
+}
+
+type_init(aw_a10_register_types)

hw/arm/boot.c

@@ -17,18 +17,55 @@
 #include "sysemu/device_tree.h"
 #include "qemu/config-file.h"
 
+/* Kernel boot protocol is specified in the kernel docs
+ * Documentation/arm/Booting and Documentation/arm64/booting.txt
+ * They have different preferred image load offsets from system RAM base.
+ */
 #define KERNEL_ARGS_ADDR 0x100
 #define KERNEL_LOAD_ADDR 0x00010000
+#define KERNEL64_LOAD_ADDR 0x00080000
+
+typedef enum {
+    FIXUP_NONE = 0,   /* do nothing */
+    FIXUP_TERMINATOR, /* end of insns */
+    FIXUP_BOARDID,    /* overwrite with board ID number */
+    FIXUP_ARGPTR,     /* overwrite with pointer to kernel args */
+    FIXUP_ENTRYPOINT, /* overwrite with kernel entry point */
+    FIXUP_GIC_CPU_IF, /* overwrite with GIC CPU interface address */
+    FIXUP_BOOTREG,    /* overwrite with boot register address */
+    FIXUP_DSB,        /* overwrite with correct DSB insn for cpu */
+    FIXUP_MAX,
+} FixupType;
+
+typedef struct ARMInsnFixup {
+    uint32_t insn;
+    FixupType fixup;
+} ARMInsnFixup;
+
+static const ARMInsnFixup bootloader_aarch64[] = {
+    { 0x580000c0 }, /* ldr x0, arg ; Load the lower 32-bits of DTB */
+    { 0xaa1f03e1 }, /* mov x1, xzr */
+    { 0xaa1f03e2 }, /* mov x2, xzr */
+    { 0xaa1f03e3 }, /* mov x3, xzr */
+    { 0x58000084 }, /* ldr x4, entry ; Load the lower 32-bits of kernel entry */
+    { 0xd61f0080 }, /* br x4      ; Jump to the kernel entry point */
+    { 0, FIXUP_ARGPTR }, /* arg: .word @DTB Lower 32-bits */
+    { 0 }, /* .word @DTB Higher 32-bits */
+    { 0, FIXUP_ENTRYPOINT }, /* entry: .word @Kernel Entry Lower 32-bits */
+    { 0 }, /* .word @Kernel Entry Higher 32-bits */
+    { 0, FIXUP_TERMINATOR }
+};
 
 /* The worlds second smallest bootloader.  Set r0-r2, then jump to kernel.  */
-static uint32_t bootloader[] = {
-  0xe3a00000, /* mov     r0, #0 */
-  0xe59f1004, /* ldr     r1, [pc, #4] */
-  0xe59f2004, /* ldr     r2, [pc, #4] */
-  0xe59ff004, /* ldr     pc, [pc, #4] */
-  0, /* Board ID */
-  0, /* Address of kernel args.  Set by integratorcp_init.  */
-  0  /* Kernel entry point.  Set by integratorcp_init.  */
+static const ARMInsnFixup bootloader[] = {
+    { 0xe3a00000 }, /* mov     r0, #0 */
+    { 0xe59f1004 }, /* ldr     r1, [pc, #4] */
+    { 0xe59f2004 }, /* ldr     r2, [pc, #4] */
+    { 0xe59ff004 }, /* ldr     pc, [pc, #4] */
+    { 0, FIXUP_BOARDID },
+    { 0, FIXUP_ARGPTR },
+    { 0, FIXUP_ENTRYPOINT },
+    { 0, FIXUP_TERMINATOR }
 };
 
 /* Handling for secondary CPU boot in a multicore system.
@@ -48,39 +85,83 @@ static uint32_t bootloader[] = {
 #define DSB_INSN 0xf57ff04f
 #define CP15_DSB_INSN 0xee070f9a /* mcr cp15, 0, r0, c7, c10, 4 */
 
-static uint32_t smpboot[] = {
-  0xe59f2028, /* ldr r2, gic_cpu_if */
-  0xe59f0028, /* ldr r0, startaddr */
-  0xe3a01001, /* mov r1, #1 */
-  0xe5821000, /* str r1, [r2] - set GICC_CTLR.Enable */
-  0xe3a010ff, /* mov r1, #0xff */
-  0xe5821004, /* str r1, [r2, 4] - set GIC_PMR.Priority to 0xff */
-  DSB_INSN,   /* dsb */
-  0xe320f003, /* wfi */
-  0xe5901000, /* ldr     r1, [r0] */
-  0xe1110001, /* tst     r1, r1 */
-  0x0afffffb, /* beq     <wfi> */
-  0xe12fff11, /* bx      r1 */
-  0,          /* gic_cpu_if: base address of GIC CPU interface */
-  0           /* bootreg: Boot register address is held here */
+static const ARMInsnFixup smpboot[] = {
+    { 0xe59f2028 }, /* ldr r2, gic_cpu_if */
+    { 0xe59f0028 }, /* ldr r0, bootreg_addr */
+    { 0xe3a01001 }, /* mov r1, #1 */
+    { 0xe5821000 }, /* str r1, [r2] - set GICC_CTLR.Enable */
+    { 0xe3a010ff }, /* mov r1, #0xff */
+    { 0xe5821004 }, /* str r1, [r2, 4] - set GIC_PMR.Priority to 0xff */
+    { 0, FIXUP_DSB },   /* dsb */
+    { 0xe320f003 }, /* wfi */
+    { 0xe5901000 }, /* ldr     r1, [r0] */
+    { 0xe1110001 }, /* tst     r1, r1 */
+    { 0x0afffffb }, /* beq     <wfi> */
+    { 0xe12fff11 }, /* bx      r1 */
+    { 0, FIXUP_GIC_CPU_IF }, /* gic_cpu_if: .word 0x.... */
+    { 0, FIXUP_BOOTREG }, /* bootreg_addr: .word 0x.... */
+    { 0, FIXUP_TERMINATOR }
 };
 
+static void write_bootloader(const char *name, hwaddr addr,
+                             const ARMInsnFixup *insns, uint32_t *fixupcontext)
+{
+    /* Fix up the specified bootloader fragment and write it into
+     * guest memory using rom_add_blob_fixed(). fixupcontext is
+     * an array giving the values to write in for the fixup types
+     * which write a value into the code array.
+     */
+    int i, len;
+    uint32_t *code;
+
+    len = 0;
+    while (insns[len].fixup != FIXUP_TERMINATOR) {
+        len++;
+    }
+
+    code = g_new0(uint32_t, len);
+
+    for (i = 0; i < len; i++) {
+        uint32_t insn = insns[i].insn;
+        FixupType fixup = insns[i].fixup;
+
+        switch (fixup) {
+        case FIXUP_NONE:
+            break;
+        case FIXUP_BOARDID:
+        case FIXUP_ARGPTR:
+        case FIXUP_ENTRYPOINT:
+        case FIXUP_GIC_CPU_IF:
+        case FIXUP_BOOTREG:
+        case FIXUP_DSB:
+            insn = fixupcontext[fixup];
+            break;
+        default:
+            abort();
+        }
+        code[i] = tswap32(insn);
+    }
+
+    rom_add_blob_fixed(name, code, len * sizeof(uint32_t), addr);
+
+    g_free(code);
+}
+
 static void default_write_secondary(ARMCPU *cpu,
                                     const struct arm_boot_info *info)
 {
-    int n;
-    smpboot[ARRAY_SIZE(smpboot) - 1] = info->smp_bootreg_addr;
-    smpboot[ARRAY_SIZE(smpboot) - 2] = info->gic_cpu_if_addr;
-    for (n = 0; n < ARRAY_SIZE(smpboot); n++) {
-        /* Replace DSB with the pre-v7 DSB if necessary. */
-        if (!arm_feature(&cpu->env, ARM_FEATURE_V7) &&
-            smpboot[n] == DSB_INSN) {
-            smpboot[n] = CP15_DSB_INSN;
-        }
-        smpboot[n] = tswap32(smpboot[n]);
+    uint32_t fixupcontext[FIXUP_MAX];
+
+    fixupcontext[FIXUP_GIC_CPU_IF] = info->gic_cpu_if_addr;
+    fixupcontext[FIXUP_BOOTREG] = info->smp_bootreg_addr;
+    if (arm_feature(&cpu->env, ARM_FEATURE_V7)) {
+        fixupcontext[FIXUP_DSB] = DSB_INSN;
+    } else {
+        fixupcontext[FIXUP_DSB] = CP15_DSB_INSN;
     }
-    rom_add_blob_fixed("smpboot", smpboot, sizeof(smpboot),
-                       info->smp_loader_start);
+
+    write_bootloader("smpboot", info->smp_loader_start,
+                     smpboot, fixupcontext);
 }
 
 static void default_reset_secondary(ARMCPU *cpu,
@@ -228,26 +309,34 @@ static void set_kernel_args_old(const struct arm_boot_info *info)
 static int load_dtb(hwaddr addr, const struct arm_boot_info *binfo)
 {
     void *fdt = NULL;
-    char *filename;
     int size, rc;
     uint32_t acells, scells;
 
-    filename = qemu_find_file(QEMU_FILE_TYPE_BIOS, binfo->dtb_filename);
-    if (!filename) {
-        fprintf(stderr, "Couldn't open dtb file %s\n", binfo->dtb_filename);
-        goto fail;
-    }
+    if (binfo->dtb_filename) {
+        char *filename;
+        filename = qemu_find_file(QEMU_FILE_TYPE_BIOS, binfo->dtb_filename);
+        if (!filename) {
+            fprintf(stderr, "Couldn't open dtb file %s\n", binfo->dtb_filename);
+            goto fail;
+        }
 
-    fdt = load_device_tree(filename, &size);
-    if (!fdt) {
-        fprintf(stderr, "Couldn't open dtb file %s\n", filename);
+        fdt = load_device_tree(filename, &size);
+        if (!fdt) {
+            fprintf(stderr, "Couldn't open dtb file %s\n", filename);
+            g_free(filename);
+            goto fail;
+        }
         g_free(filename);
-        goto fail;
+    } else if (binfo->get_dtb) {
+        fdt = binfo->get_dtb(binfo, &size);
+        if (!fdt) {
+            fprintf(stderr, "Board was unable to create a dtb blob\n");
+            goto fail;
+        }
     }
-    g_free(filename);
 
-    acells = qemu_devtree_getprop_cell(fdt, "/", "#address-cells");
-    scells = qemu_devtree_getprop_cell(fdt, "/", "#size-cells");
+    acells = qemu_fdt_getprop_cell(fdt, "/", "#address-cells");
+    scells = qemu_fdt_getprop_cell(fdt, "/", "#size-cells");
     if (acells == 0 || scells == 0) {
         fprintf(stderr, "dtb file invalid (#address-cells or #size-cells 0)\n");
         goto fail;
@@ -262,17 +351,17 @@ static int load_dtb(hwaddr addr, const struct arm_boot_info *binfo)
         goto fail;
     }
 
-    rc = qemu_devtree_setprop_sized_cells(fdt, "/memory", "reg",
-                                          acells, binfo->loader_start,
-                                          scells, binfo->ram_size);
+    rc = qemu_fdt_setprop_sized_cells(fdt, "/memory", "reg",
+                                      acells, binfo->loader_start,
+                                      scells, binfo->ram_size);
     if (rc < 0) {
         fprintf(stderr, "couldn't set /memory/reg\n");
         goto fail;
     }
 
     if (binfo->kernel_cmdline && *binfo->kernel_cmdline) {
-        rc = qemu_devtree_setprop_string(fdt, "/chosen", "bootargs",
-                                          binfo->kernel_cmdline);
+        rc = qemu_fdt_setprop_string(fdt, "/chosen", "bootargs",
+                                     binfo->kernel_cmdline);
         if (rc < 0) {
             fprintf(stderr, "couldn't set /chosen/bootargs\n");
             goto fail;
@@ -280,15 +369,15 @@ static int load_dtb(hwaddr addr, const struct arm_boot_info *binfo)
     }
 
     if (binfo->initrd_size) {
-        rc = qemu_devtree_setprop_cell(fdt, "/chosen", "linux,initrd-start",
-                binfo->initrd_start);
+        rc = qemu_fdt_setprop_cell(fdt, "/chosen", "linux,initrd-start",
+                                   binfo->initrd_start);
         if (rc < 0) {
             fprintf(stderr, "couldn't set /chosen/linux,initrd-start\n");
             goto fail;
         }
 
-        rc = qemu_devtree_setprop_cell(fdt, "/chosen", "linux,initrd-end",
-                    binfo->initrd_start + binfo->initrd_size);
+        rc = qemu_fdt_setprop_cell(fdt, "/chosen", "linux,initrd-end",
+                                   binfo->initrd_start + binfo->initrd_size);
         if (rc < 0) {
             fprintf(stderr, "couldn't set /chosen/linux,initrd-end\n");
             goto fail;
@@ -299,7 +388,7 @@ static int load_dtb(hwaddr addr, const struct arm_boot_info *binfo)
         binfo->modify_dtb(binfo, fdt);
     }
 
-    qemu_devtree_dumpdtb(fdt, size);
+    qemu_fdt_dumpdtb(fdt, size);
 
     cpu_physical_memory_write(addr, fdt, size);
 
@@ -326,7 +415,12 @@ static void do_cpu_reset(void *opaque)
             env->thumb = info->entry & 1;
         } else {
             if (CPU(cpu) == first_cpu) {
-                env->regs[15] = info->loader_start;
+                if (env->aarch64) {
+                    env->pc = info->loader_start;
+                } else {
+                    env->regs[15] = info->loader_start;
+                }
+
                 if (!info->dtb_filename) {
                     if (old_param) {
                         set_kernel_args_old(info);
@@ -346,11 +440,11 @@ void arm_load_kernel(ARMCPU *cpu, struct arm_boot_info *info)
     CPUState *cs = CPU(cpu);
     int kernel_size;
     int initrd_size;
-    int n;
     int is_linux = 0;
     uint64_t elf_entry;
-    hwaddr entry;
+    hwaddr entry, kernel_load_offset;
     int big_endian;
+    static const ARMInsnFixup *primary_loader;
 
     /* Load the kernel.  */
     if (!info->kernel_filename) {
@@ -360,6 +454,14 @@ void arm_load_kernel(ARMCPU *cpu, struct arm_boot_info *info)
         return;
     }
 
+    if (arm_feature(&cpu->env, ARM_FEATURE_AARCH64)) {
+        primary_loader = bootloader_aarch64;
+        kernel_load_offset = KERNEL64_LOAD_ADDR;
+    } else {
+        primary_loader = bootloader;
+        kernel_load_offset = KERNEL_LOAD_ADDR;
+    }
+
     info->dtb_filename = qemu_opt_get(qemu_get_machine_opts(), "dtb");
 
     if (!info->secondary_cpu_reset_hook) {
@@ -400,9 +502,9 @@ void arm_load_kernel(ARMCPU *cpu, struct arm_boot_info *info)
                                   &is_linux);
     }
     if (kernel_size < 0) {
-        entry = info->loader_start + KERNEL_LOAD_ADDR;
+        entry = info->loader_start + kernel_load_offset;
         kernel_size = load_image_targphys(info->kernel_filename, entry,
-                                          info->ram_size - KERNEL_LOAD_ADDR);
+                                          info->ram_size - kernel_load_offset);
         is_linux = 1;
     }
     if (kernel_size < 0) {
@@ -412,6 +514,8 @@ void arm_load_kernel(ARMCPU *cpu, struct arm_boot_info *info)
     }
     info->entry = entry;
     if (is_linux) {
+        uint32_t fixupcontext[FIXUP_MAX];
+
         if (info->initrd_filename) {
             initrd_size = load_ramdisk(info->initrd_filename,
                                        info->initrd_start,
@@ -433,12 +537,12 @@ void arm_load_kernel(ARMCPU *cpu, struct arm_boot_info *info)
         }
         info->initrd_size = initrd_size;
 
-        bootloader[4] = info->board_id;
+        fixupcontext[FIXUP_BOARDID] = info->board_id;
 
         /* for device tree boot, we pass the DTB directly in r2. Otherwise
          * we point to the kernel args.
          */
-        if (info->dtb_filename) {
+        if (info->dtb_filename || info->get_dtb) {
             /* Place the DTB after the initrd in memory. Note that some
              * kernels will trash anything in the 4K page the initrd
              * ends in, so make sure the DTB isn't caught up in that.
@@ -448,9 +552,9 @@ void arm_load_kernel(ARMCPU *cpu, struct arm_boot_info *info)
             if (load_dtb(dtb_start, info)) {
                 exit(1);
             }
-            bootloader[5] = dtb_start;
+            fixupcontext[FIXUP_ARGPTR] = dtb_start;
         } else {
-            bootloader[5] = info->loader_start + KERNEL_ARGS_ADDR;
+            fixupcontext[FIXUP_ARGPTR] = info->loader_start + KERNEL_ARGS_ADDR;
             if (info->ram_size >= (1ULL << 32)) {
                 fprintf(stderr, "qemu: RAM size must be less than 4GB to boot"
                         " Linux kernel using ATAGS (try passing a device tree"
@@ -458,12 +562,11 @@ void arm_load_kernel(ARMCPU *cpu, struct arm_boot_info *info)
                 exit(1);
             }
         }
-        bootloader[6] = entry;
-        for (n = 0; n < sizeof(bootloader) / 4; n++) {
-            bootloader[n] = tswap32(bootloader[n]);
-        }
-        rom_add_blob_fixed("bootloader", bootloader, sizeof(bootloader),
-                           info->loader_start);
+        fixupcontext[FIXUP_ENTRYPOINT] = entry;
+
+        write_bootloader("bootloader", info->loader_start,
+                         primary_loader, fixupcontext);
+
         if (info->nb_cpus > 1) {
             info->write_secondary_boot(cpu, info);
         }

hw/arm/cubieboard.c

@@ -0,0 +1,69 @@
+/*
+ * cubieboard emulation
+ *
+ * Copyright (C) 2013 Li Guang
+ * Written by Li Guang <lig.fnst@cn.fujitsu.com>
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License as published by the
+ * Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License
+ * for more details.
+ */
+
+#include "hw/sysbus.h"
+#include "hw/devices.h"
+#include "hw/boards.h"
+#include "hw/arm/allwinner-a10.h"
+
+static struct arm_boot_info cubieboard_binfo = {
+    .loader_start = AW_A10_SDRAM_BASE,
+    .board_id = 0x1008,
+};
+
+typedef struct CubieBoardState {
+    AwA10State *a10;
+    MemoryRegion sdram;
+} CubieBoardState;
+
+static void cubieboard_init(QEMUMachineInitArgs *args)
+{
+    CubieBoardState *s = g_new(CubieBoardState, 1);
+    Error *err = NULL;
+
+    s->a10 = AW_A10(object_new(TYPE_AW_A10));
+    object_property_set_bool(OBJECT(s->a10), true, "realized", &err);
+    if (err != NULL) {
+        error_report("Couldn't realize Allwinner A10: %s\n",
+                error_get_pretty(err));
+        exit(1);
+    }
+
+    memory_region_init_ram(&s->sdram, NULL, "cubieboard.ram", args->ram_size);
+    vmstate_register_ram_global(&s->sdram);
+    memory_region_add_subregion(get_system_memory(), AW_A10_SDRAM_BASE,
+                                &s->sdram);
+
+    cubieboard_binfo.ram_size = args->ram_size;
+    cubieboard_binfo.kernel_filename = args->kernel_filename;
+    cubieboard_binfo.kernel_cmdline = args->kernel_cmdline;
+    arm_load_kernel(&s->a10->cpu, &cubieboard_binfo);
+}
+
+static QEMUMachine cubieboard_machine = {
+    .name = "cubieboard",
+    .desc = "cubietech cubieboard",
+    .init = cubieboard_init,
+};
+
+
+static void cubieboard_machine_init(void)
+{
+    qemu_register_machine(&cubieboard_machine);
+}
+
+machine_init(cubieboard_machine_init)

hw/arm/digic.c

@@ -0,0 +1,115 @@
+/*
+ * QEMU model of the Canon DIGIC SoC.
+ *
+ * Copyright (C) 2013 Antony Pavlov <antonynpavlov@gmail.com>
+ *
+ * This model is based on reverse engineering efforts
+ * made by CHDK (http://chdk.wikia.com) and
+ * Magic Lantern (http://www.magiclantern.fm) projects
+ * contributors.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ */
+
+#include "hw/arm/digic.h"
+
+#define DIGIC4_TIMER_BASE(n)    (0xc0210000 + (n) * 0x100)
+
+#define DIGIC_UART_BASE          0xc0800000
+
+static void digic_init(Object *obj)
+{
+    DigicState *s = DIGIC(obj);
+    DeviceState *dev;
+    int i;
+
+    object_initialize(&s->cpu, sizeof(s->cpu), "arm946-" TYPE_ARM_CPU);
+    object_property_add_child(obj, "cpu", OBJECT(&s->cpu), NULL);
+
+    for (i = 0; i < DIGIC4_NB_TIMERS; i++) {
+#define DIGIC_TIMER_NAME_MLEN    11
+        char name[DIGIC_TIMER_NAME_MLEN];
+
+        object_initialize(&s->timer[i], sizeof(s->timer[i]), TYPE_DIGIC_TIMER);
+        dev = DEVICE(&s->timer[i]);
+        qdev_set_parent_bus(dev, sysbus_get_default());
+        snprintf(name, DIGIC_TIMER_NAME_MLEN, "timer[%d]", i);
+        object_property_add_child(obj, name, OBJECT(&s->timer[i]), NULL);
+    }
+
+    object_initialize(&s->uart, sizeof(s->uart), TYPE_DIGIC_UART);
+    dev = DEVICE(&s->uart);
+    qdev_set_parent_bus(dev, sysbus_get_default());
+    object_property_add_child(obj, "uart", OBJECT(&s->uart), NULL);
+}
+
+static void digic_realize(DeviceState *dev, Error **errp)
+{
+    DigicState *s = DIGIC(dev);
+    Error *err = NULL;
+    SysBusDevice *sbd;
+    int i;
+
+    object_property_set_bool(OBJECT(&s->cpu), true, "reset-hivecs", &err);
+    if (err != NULL) {
+        error_propagate(errp, err);
+        return;
+    }
+
+    object_property_set_bool(OBJECT(&s->cpu), true, "realized", &err);
+    if (err != NULL) {
+        error_propagate(errp, err);
+        return;
+    }
+
+    for (i = 0; i < DIGIC4_NB_TIMERS; i++) {
+        object_property_set_bool(OBJECT(&s->timer[i]), true, "realized", &err);
+        if (err != NULL) {
+            error_propagate(errp, err);
+            return;
+        }
+
+        sbd = SYS_BUS_DEVICE(&s->timer[i]);
+        sysbus_mmio_map(sbd, 0, DIGIC4_TIMER_BASE(i));
+    }
+
+    object_property_set_bool(OBJECT(&s->uart), true, "realized", &err);
+    if (err != NULL) {
+        error_propagate(errp, err);
+        return;
+    }
+
+    sbd = SYS_BUS_DEVICE(&s->uart);
+    sysbus_mmio_map(sbd, 0, DIGIC_UART_BASE);
+}
+
+static void digic_class_init(ObjectClass *oc, void *data)
+{
+    DeviceClass *dc = DEVICE_CLASS(oc);
+
+    dc->realize = digic_realize;
+}
+
+static const TypeInfo digic_type_info = {
+    .name = TYPE_DIGIC,
+    .parent = TYPE_DEVICE,
+    .instance_size = sizeof(DigicState),
+    .instance_init = digic_init,
+    .class_init = digic_class_init,
+};
+
+static void digic_register_types(void)
+{
+    type_register_static(&digic_type_info);
+}
+
+type_init(digic_register_types)

hw/arm/digic_boards.c

@@ -0,0 +1,162 @@
+/*
+ * QEMU model of the Canon DIGIC boards (cameras indeed :).
+ *
+ * Copyright (C) 2013 Antony Pavlov <antonynpavlov@gmail.com>
+ *
+ * This model is based on reverse engineering efforts
+ * made by CHDK (http://chdk.wikia.com) and
+ * Magic Lantern (http://www.magiclantern.fm) projects
+ * contributors.
+ *
+ * See docs here:
+ *   http://magiclantern.wikia.com/wiki/Register_Map
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ */
+
+#include "hw/boards.h"
+#include "exec/address-spaces.h"
+#include "qemu/error-report.h"
+#include "hw/arm/digic.h"
+#include "hw/block/flash.h"
+#include "hw/loader.h"
+#include "sysemu/sysemu.h"
+#include "sysemu/qtest.h"
+
+#define DIGIC4_ROM0_BASE      0xf0000000
+#define DIGIC4_ROM1_BASE      0xf8000000
+#define DIGIC4_ROM_MAX_SIZE   0x08000000
+
+typedef struct DigicBoardState {
+    DigicState *digic;
+    MemoryRegion ram;
+} DigicBoardState;
+
+typedef struct DigicBoard {
+    hwaddr ram_size;
+    void (*add_rom0)(DigicBoardState *, hwaddr, const char *);
+    const char *rom0_def_filename;
+    void (*add_rom1)(DigicBoardState *, hwaddr, const char *);
+    const char *rom1_def_filename;
+} DigicBoard;
+
+static void digic4_board_setup_ram(DigicBoardState *s, hwaddr ram_size)
+{
+    memory_region_init_ram(&s->ram, NULL, "ram", ram_size);
+    memory_region_add_subregion(get_system_memory(), 0, &s->ram);
+    vmstate_register_ram_global(&s->ram);
+}
+
+static void digic4_board_init(DigicBoard *board)
+{
+    Error *err = NULL;
+
+    DigicBoardState *s = g_new(DigicBoardState, 1);
+
+    s->digic = DIGIC(object_new(TYPE_DIGIC));
+    object_property_set_bool(OBJECT(s->digic), true, "realized", &err);
+    if (err != NULL) {
+        error_report("Couldn't realize DIGIC SoC: %s\n",
+                     error_get_pretty(err));
+        exit(1);
+    }
+
+    digic4_board_setup_ram(s, board->ram_size);
+
+    if (board->add_rom0) {
+        board->add_rom0(s, DIGIC4_ROM0_BASE, board->rom0_def_filename);
+    }
+
+    if (board->add_rom1) {
+        board->add_rom1(s, DIGIC4_ROM1_BASE, board->rom1_def_filename);
+    }
+}
+
+static void digic_load_rom(DigicBoardState *s, hwaddr addr,
+                           hwaddr max_size, const char *def_filename)
+{
+    target_long rom_size;
+    const char *filename;
+
+    if (qtest_enabled()) {
+        /* qtest runs no code so don't attempt a ROM load which
+         * could fail and result in a spurious test failure.
+         */
+        return;
+    }
+
+    if (bios_name) {
+        filename = bios_name;
+    } else {
+        filename = def_filename;
+    }
+
+    if (filename) {
+        char *fn = qemu_find_file(QEMU_FILE_TYPE_BIOS, filename);
+
+        if (!fn) {
+            error_report("Couldn't find rom image '%s'.\n", filename);
+            exit(1);
+        }
+
+        rom_size = load_image_targphys(fn, addr, max_size);
+        if (rom_size < 0 || rom_size > max_size) {
+            error_report("Couldn't load rom image '%s'.\n", filename);
+            exit(1);
+        }
+    }
+}
+
+/*
+ * Samsung K8P3215UQB
+ * 64M Bit (4Mx16) Page Mode / Multi-Bank NOR Flash Memory
+ */
+static void digic4_add_k8p3215uqb_rom(DigicBoardState *s, hwaddr addr,
+                                      const char *def_filename)
+{
+#define FLASH_K8P3215UQB_SIZE (4 * 1024 * 1024)
+#define FLASH_K8P3215UQB_SECTOR_SIZE (64 * 1024)
+
+    pflash_cfi02_register(addr, NULL, "pflash", FLASH_K8P3215UQB_SIZE,
+                          NULL, FLASH_K8P3215UQB_SECTOR_SIZE,
+                          FLASH_K8P3215UQB_SIZE / FLASH_K8P3215UQB_SECTOR_SIZE,
+                          DIGIC4_ROM_MAX_SIZE / FLASH_K8P3215UQB_SIZE,
+                          4,
+                          0x00EC, 0x007E, 0x0003, 0x0001,
+                          0x0555, 0x2aa, 0);
+
+    digic_load_rom(s, addr, FLASH_K8P3215UQB_SIZE, def_filename);
+}
+
+static DigicBoard digic4_board_canon_a1100 = {
+    .ram_size = 64 * 1024 * 1024,
+    .add_rom1 = digic4_add_k8p3215uqb_rom,
+    .rom1_def_filename = "canon-a1100-rom1.bin",
+};
+
+static void canon_a1100_init(QEMUMachineInitArgs *args)
+{
+    digic4_board_init(&digic4_board_canon_a1100);
+}
+
+static QEMUMachine canon_a1100 = {
+    .name = "canon-a1100",
+    .desc = "Canon PowerShot A1100 IS",
+    .init = &canon_a1100_init,
+};
+
+static void digic_register_machines(void)
+{
+    qemu_register_machine(&canon_a1100);
+}
+
+machine_init(digic_register_machines)

hw/arm/highbank.c

@@ -26,12 +26,13 @@
 #include "hw/boards.h"
 #include "sysemu/blockdev.h"
 #include "exec/address-spaces.h"
+#include "qemu/error-report.h"
 
-#define SMP_BOOT_ADDR 0x100
-#define SMP_BOOT_REG  0x40
-#define GIC_BASE_ADDR 0xfff10000
+#define SMP_BOOT_ADDR           0x100
+#define SMP_BOOT_REG            0x40
+#define MPCORE_PERIPHBASE       0xfff10000
 
-#define NIRQ_GIC      160
+#define NIRQ_GIC                160
 
 /* Board init.  */
 
@@ -54,7 +55,7 @@ static void hb_write_secondary(ARMCPU *cpu, const struct arm_boot_info *info)
         0xe1110001, /* tst     r1, r1 */
         0x0afffffb, /* beq     <wfi> */
         0xe12fff11, /* bx      r1 */
-        GIC_BASE_ADDR      /* privbase: gic address.  */
+        MPCORE_PERIPHBASE   /* privbase: MPCore peripheral base address.  */
     };
     for (n = 0; n < ARRAY_SIZE(smpboot); n++) {
         smpboot[n] = tswap32(smpboot[n]);
@@ -125,7 +126,7 @@ typedef struct {
     SysBusDevice parent_obj;
     /*< public >*/
 
-    MemoryRegion *iomem;
+    MemoryRegion iomem;
     uint32_t regs[NUM_REGS];
 } HighbankRegsState;
 
@@ -154,10 +155,9 @@ static int highbank_regs_init(SysBusDevice *dev)
 {
     HighbankRegsState *s = HIGHBANK_REGISTERS(dev);
 
-    s->iomem = g_new(MemoryRegion, 1);
-    memory_region_init_io(s->iomem, OBJECT(s), &hb_mem_ops, s->regs,
+    memory_region_init_io(&s->iomem, OBJECT(s), &hb_mem_ops, s->regs,
                           "highbank_regs", 0x1000);
-    sysbus_init_mmio(dev, s->iomem);
+    sysbus_init_mmio(dev, &s->iomem);
 
     return 0;
 }
@@ -229,15 +229,23 @@ static void calxeda_init(QEMUMachineInitArgs *args, enum cxmachines machine)
     }
 
     for (n = 0; n < smp_cpus; n++) {
+        ObjectClass *oc = cpu_class_by_name(TYPE_ARM_CPU, cpu_model);
         ARMCPU *cpu;
-        cpu = cpu_arm_init(cpu_model);
-        if (cpu == NULL) {
-            fprintf(stderr, "Unable to find CPU definition\n");
+        Error *err = NULL;
+
+        cpu = ARM_CPU(object_new(object_class_get_name(oc)));
+
+        object_property_set_int(OBJECT(cpu), MPCORE_PERIPHBASE, "reset-cbar",
+                                &err);
+        if (err) {
+            error_report("%s", error_get_pretty(err));
+            exit(1);
+        }
+        object_property_set_bool(OBJECT(cpu), true, "realized", &err);
+        if (err) {
+            error_report("%s", error_get_pretty(err));
             exit(1);
         }
-
-        /* This will become a QOM property eventually */
-        cpu->reset_cbar = GIC_BASE_ADDR;
         cpu_irq[n] = qdev_get_gpio_in(DEVICE(cpu), ARM_CPU_IRQ);
     }
 
@@ -279,7 +287,7 @@ static void calxeda_init(QEMUMachineInitArgs *args, enum cxmachines machine)
     qdev_prop_set_uint32(dev, "num-irq", NIRQ_GIC);
     qdev_init_nofail(dev);
     busdev = SYS_BUS_DEVICE(dev);
-    sysbus_mmio_map(busdev, 0, GIC_BASE_ADDR);
+    sysbus_mmio_map(busdev, 0, MPCORE_PERIPHBASE);
     for (n = 0; n < smp_cpus; n++) {
         sysbus_connect_irq(busdev, n, cpu_irq[n]);
     }

hw/arm/integratorcp.c

@@ -36,6 +36,7 @@ typedef struct IntegratorCMState {
     uint32_t cm_init;
     uint32_t cm_flags;
     uint32_t cm_nvflags;
+    uint32_t cm_refcnt_offset;
     uint32_t int_level;
     uint32_t irq_enabled;
     uint32_t fiq_enabled;
@@ -82,9 +83,13 @@ static uint64_t integratorcm_read(void *opaque, hwaddr offset,
         return s->cm_sdram;
     case 9: /* CM_INIT */
         return s->cm_init;
-    case 10: /* CM_REFCT */
-        /* ??? High frequency timer.  */
-        hw_error("integratorcm_read: CM_REFCT");
+    case 10: /* CM_REFCNT */
+        /* This register, CM_REFCNT, provides a 32-bit count value.
+         * The count increments at the fixed reference clock frequency of 24MHz
+         * and can be used as a real-time counter.
+         */
+        return (uint32_t)muldiv64(qemu_clock_get_ns(QEMU_CLOCK_VIRTUAL), 24,
+                                  1000) - s->cm_refcnt_offset;
     case 12: /* CM_FLAGS */
         return s->cm_flags;
     case 14: /* CM_NVFLAGS */
@@ -257,6 +262,8 @@ static int integratorcm_init(SysBusDevice *dev)
     }
     memcpy(integrator_spd + 73, "QEMU-MEMORY", 11);
     s->cm_init = 0x00000112;
+    s->cm_refcnt_offset = muldiv64(qemu_clock_get_ns(QEMU_CLOCK_VIRTUAL), 24,
+                                   1000);
     memory_region_init_ram(&s->flash, OBJECT(s), "integrator.flash", 0x100000);
     vmstate_register_ram_global(&s->flash);
 

hw/arm/mainstone.c

@@ -45,7 +45,7 @@
 #define S1_STSCHG_IRQ 14
 #define S1_IRQ        15
 
-static struct keymap map[0xE0] = {
+static const struct keymap map[0xE0] = {
     [0 ... 0xDF] = { -1, -1 },
     [0x1e] = {0,0}, /* a */
     [0x30] = {0,1}, /* b */
@@ -75,9 +75,18 @@ static struct keymap map[0xE0] = {
     [0x2c] = {4,3}, /* z */
     [0xc7] = {5,0}, /* Home */
     [0x2a] = {5,1}, /* shift */
-    [0x39] = {5,2}, /* space */
+    /*
+     * There are two matrix positions which map to space,
+     * but QEMU can only use one of them for the reverse
+     * mapping, so simply use the second one.
+     */
+    /* [0x39] = {5,2}, space */
     [0x39] = {5,3}, /* space */
-    [0x1c] = {5,5}, /*  enter */
+    /*
+     * Matrix position {5,4} and other keys are missing here.
+     * TODO: Compare with Linux code and test real hardware.
+     */
+    [0x1c] = {5,5}, /* enter (TODO: might be wrong) */
     [0xc8] = {6,0}, /* up */
     [0xd0] = {6,1}, /* down */
     [0xcb] = {6,2}, /* left */

hw/arm/versatilepb.c

@@ -390,7 +390,6 @@ static void vpb_sic_class_init(ObjectClass *klass, void *data)
     SysBusDeviceClass *k = SYS_BUS_DEVICE_CLASS(klass);
 
     k->init = vpb_sic_init;
-    dc->no_user = 1;
     dc->vmsd = &vmstate_vpb_sic;
 }
 

hw/arm/vexpress.c

@@ -419,13 +419,13 @@ static int add_virtio_mmio_node(void *fdt, uint32_t acells, uint32_t scells,
     int rc;
     char *nodename = g_strdup_printf("/virtio_mmio@%" PRIx64, addr);
 
-    rc = qemu_devtree_add_subnode(fdt, nodename);
-    rc |= qemu_devtree_setprop_string(fdt, nodename,
-                                      "compatible", "virtio,mmio");
-    rc |= qemu_devtree_setprop_sized_cells(fdt, nodename, "reg",
-                                           acells, addr, scells, size);
-    qemu_devtree_setprop_cells(fdt, nodename, "interrupt-parent", intc);
-    qemu_devtree_setprop_cells(fdt, nodename, "interrupts", 0, irq, 1);
+    rc = qemu_fdt_add_subnode(fdt, nodename);
+    rc |= qemu_fdt_setprop_string(fdt, nodename,
+                                  "compatible", "virtio,mmio");
+    rc |= qemu_fdt_setprop_sized_cells(fdt, nodename, "reg",
+                                       acells, addr, scells, size);
+    qemu_fdt_setprop_cells(fdt, nodename, "interrupt-parent", intc);
+    qemu_fdt_setprop_cells(fdt, nodename, "interrupts", 0, irq, 1);
     g_free(nodename);
     if (rc) {
         return -1;
@@ -456,8 +456,8 @@ static void vexpress_modify_dtb(const struct arm_boot_info *info, void *fdt)
     uint32_t acells, scells, intc;
     const VEDBoardInfo *daughterboard = (const VEDBoardInfo *)info;
 
-    acells = qemu_devtree_getprop_cell(fdt, "/", "#address-cells");
-    scells = qemu_devtree_getprop_cell(fdt, "/", "#size-cells");
+    acells = qemu_fdt_getprop_cell(fdt, "/", "#address-cells");
+    scells = qemu_fdt_getprop_cell(fdt, "/", "#size-cells");
     intc = find_int_controller(fdt);
     if (!intc) {
         /* Not fatal, we just won't provide virtio. This will
@@ -480,6 +480,36 @@ static void vexpress_modify_dtb(const struct arm_boot_info *info, void *fdt)
     }
 }
 
+
+/* Open code a private version of pflash registration since we
+ * need to set non-default device width for VExpress platform.
+ */
+static pflash_t *ve_pflash_cfi01_register(hwaddr base, const char *name,
+                                          DriveInfo *di)
+{
+    DeviceState *dev = qdev_create(NULL, "cfi.pflash01");
+
+    if (di && qdev_prop_set_drive(dev, "drive", di->bdrv)) {
+        abort();
+    }
+
+    qdev_prop_set_uint32(dev, "num-blocks",
+                         VEXPRESS_FLASH_SIZE / VEXPRESS_FLASH_SECT_SIZE);
+    qdev_prop_set_uint64(dev, "sector-length", VEXPRESS_FLASH_SECT_SIZE);
+    qdev_prop_set_uint8(dev, "width", 4);
+    qdev_prop_set_uint8(dev, "device-width", 2);
+    qdev_prop_set_uint8(dev, "big-endian", 0);
+    qdev_prop_set_uint16(dev, "id0", 0x89);
+    qdev_prop_set_uint16(dev, "id1", 0x18);
+    qdev_prop_set_uint16(dev, "id2", 0x00);
+    qdev_prop_set_uint16(dev, "id3", 0x00);
+    qdev_prop_set_string(dev, "name", name);
+    qdev_init_nofail(dev);
+
+    sysbus_mmio_map(SYS_BUS_DEVICE(dev), 0, base);
+    return OBJECT_CHECK(pflash_t, (dev), "cfi.pflash01");
+}
+
 static void vexpress_common_init(VEDBoardInfo *daughterboard,
                                  QEMUMachineInitArgs *args)
 {
@@ -561,11 +591,8 @@ static void vexpress_common_init(VEDBoardInfo *daughterboard,
     sysbus_create_simple("pl111", map[VE_CLCD], pic[14]);
 
     dinfo = drive_get_next(IF_PFLASH);
-    pflash0 = pflash_cfi01_register(map[VE_NORFLASH0], NULL, "vexpress.flash0",
-            VEXPRESS_FLASH_SIZE, dinfo ? dinfo->bdrv : NULL,
-            VEXPRESS_FLASH_SECT_SIZE,
-            VEXPRESS_FLASH_SIZE / VEXPRESS_FLASH_SECT_SIZE, 4,
-            0x00, 0x89, 0x00, 0x18, 0);
+    pflash0 = ve_pflash_cfi01_register(map[VE_NORFLASH0], "vexpress.flash0",
+                                       dinfo);
     if (!pflash0) {
         fprintf(stderr, "vexpress: error registering flash 0.\n");
         exit(1);
@@ -580,11 +607,8 @@ static void vexpress_common_init(VEDBoardInfo *daughterboard,
     }
 
     dinfo = drive_get_next(IF_PFLASH);
-    if (!pflash_cfi01_register(map[VE_NORFLASH1], NULL, "vexpress.flash1",
-            VEXPRESS_FLASH_SIZE, dinfo ? dinfo->bdrv : NULL,
-            VEXPRESS_FLASH_SECT_SIZE,
-            VEXPRESS_FLASH_SIZE / VEXPRESS_FLASH_SECT_SIZE, 4,
-            0x00, 0x89, 0x00, 0x18, 0)) {
+    if (!ve_pflash_cfi01_register(map[VE_NORFLASH1], "vexpress.flash1",
+                                  dinfo)) {
         fprintf(stderr, "vexpress: error registering flash 1.\n");
         exit(1);
     }

hw/arm/virt.c

@@ -0,0 +1,452 @@
+/*
+ * ARM mach-virt emulation
+ *
+ * Copyright (c) 2013 Linaro Limited
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms and conditions of the GNU General Public License,
+ * version 2 or later, as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
+ * more details.
+ *
+ * You should have received a copy of the GNU General Public License along with
+ * this program.  If not, see <http://www.gnu.org/licenses/>.
+ *
+ * Emulate a virtual board which works by passing Linux all the information
+ * it needs about what devices are present via the device tree.
+ * There are some restrictions about what we can do here:
+ *  + we can only present devices whose Linux drivers will work based
+ *    purely on the device tree with no platform data at all
+ *  + we want to present a very stripped-down minimalist platform,
+ *    both because this reduces the security attack surface from the guest
+ *    and also because it reduces our exposure to being broken when
+ *    the kernel updates its device tree bindings and requires further
+ *    information in a device binding that we aren't providing.
+ * This is essentially the same approach kvmtool uses.
+ */
+
+#include "hw/sysbus.h"
+#include "hw/arm/arm.h"
+#include "hw/arm/primecell.h"
+#include "hw/devices.h"
+#include "net/net.h"
+#include "sysemu/device_tree.h"
+#include "sysemu/sysemu.h"
+#include "sysemu/kvm.h"
+#include "hw/boards.h"
+#include "exec/address-spaces.h"
+#include "qemu/bitops.h"
+#include "qemu/error-report.h"
+
+#define NUM_VIRTIO_TRANSPORTS 32
+
+/* Number of external interrupt lines to configure the GIC with */
+#define NUM_IRQS 128
+
+#define GIC_FDT_IRQ_TYPE_SPI 0
+#define GIC_FDT_IRQ_TYPE_PPI 1
+
+#define GIC_FDT_IRQ_FLAGS_EDGE_LO_HI 1
+#define GIC_FDT_IRQ_FLAGS_EDGE_HI_LO 2
+#define GIC_FDT_IRQ_FLAGS_LEVEL_HI 4
+#define GIC_FDT_IRQ_FLAGS_LEVEL_LO 8
+
+#define GIC_FDT_IRQ_PPI_CPU_START 8
+#define GIC_FDT_IRQ_PPI_CPU_WIDTH 8
+
+enum {
+    VIRT_FLASH,
+    VIRT_MEM,
+    VIRT_CPUPERIPHS,
+    VIRT_GIC_DIST,
+    VIRT_GIC_CPU,
+    VIRT_UART,
+    VIRT_MMIO,
+};
+
+typedef struct MemMapEntry {
+    hwaddr base;
+    hwaddr size;
+} MemMapEntry;
+
+typedef struct VirtBoardInfo {
+    struct arm_boot_info bootinfo;
+    const char *cpu_model;
+    const char *qdevname;
+    const char *gic_compatible;
+    const MemMapEntry *memmap;
+    const int *irqmap;
+    int smp_cpus;
+    void *fdt;
+    int fdt_size;
+    uint32_t clock_phandle;
+} VirtBoardInfo;
+
+/* Addresses and sizes of our components.
+ * 0..128MB is space for a flash device so we can run bootrom code such as UEFI.
+ * 128MB..256MB is used for miscellaneous device I/O.
+ * 256MB..1GB is reserved for possible future PCI support (ie where the
+ * PCI memory window will go if we add a PCI host controller).
+ * 1GB and up is RAM (which may happily spill over into the
+ * high memory region beyond 4GB).
+ * This represents a compromise between how much RAM can be given to
+ * a 32 bit VM and leaving space for expansion and in particular for PCI.
+ */
+static const MemMapEntry a15memmap[] = {
+    /* Space up to 0x8000000 is reserved for a boot ROM */
+    [VIRT_FLASH] = { 0, 0x8000000 },
+    [VIRT_CPUPERIPHS] = { 0x8000000, 0x8000 },
+    /* GIC distributor and CPU interfaces sit inside the CPU peripheral space */
+    [VIRT_GIC_DIST] = { 0x8001000, 0x1000 },
+    [VIRT_GIC_CPU] = { 0x8002000, 0x1000 },
+    [VIRT_UART] = { 0x9000000, 0x1000 },
+    [VIRT_MMIO] = { 0xa000000, 0x200 },
+    /* ...repeating for a total of NUM_VIRTIO_TRANSPORTS, each of that size */
+    /* 0x10000000 .. 0x40000000 reserved for PCI */
+    [VIRT_MEM] = { 0x40000000, 30ULL * 1024 * 1024 * 1024 },
+};
+
+static const int a15irqmap[] = {
+    [VIRT_UART] = 1,
+    [VIRT_MMIO] = 16, /* ...to 16 + NUM_VIRTIO_TRANSPORTS - 1 */
+};
+
+static VirtBoardInfo machines[] = {
+    {
+        .cpu_model = "cortex-a15",
+        .qdevname = "a15mpcore_priv",
+        .gic_compatible = "arm,cortex-a15-gic",
+        .memmap = a15memmap,
+        .irqmap = a15irqmap,
+    },
+    {
+        .cpu_model = "host",
+        /* We use the A15 private peripheral model to get a V2 GIC */
+        .qdevname = "a15mpcore_priv",
+        .gic_compatible = "arm,cortex-a15-gic",
+        .memmap = a15memmap,
+        .irqmap = a15irqmap,
+    },
+};
+
+static VirtBoardInfo *find_machine_info(const char *cpu)
+{
+    int i;
+
+    for (i = 0; i < ARRAY_SIZE(machines); i++) {
+        if (strcmp(cpu, machines[i].cpu_model) == 0) {
+            return &machines[i];
+        }
+    }
+    return NULL;
+}
+
+static void create_fdt(VirtBoardInfo *vbi)
+{
+    void *fdt = create_device_tree(&vbi->fdt_size);
+
+    if (!fdt) {
+        error_report("create_device_tree() failed");
+        exit(1);
+    }
+
+    vbi->fdt = fdt;
+
+    /* Header */
+    qemu_fdt_setprop_string(fdt, "/", "compatible", "linux,dummy-virt");
+    qemu_fdt_setprop_cell(fdt, "/", "#address-cells", 0x2);
+    qemu_fdt_setprop_cell(fdt, "/", "#size-cells", 0x2);
+
+    /*
+     * /chosen and /memory nodes must exist for load_dtb
+     * to fill in necessary properties later
+     */
+    qemu_fdt_add_subnode(fdt, "/chosen");
+    qemu_fdt_add_subnode(fdt, "/memory");
+    qemu_fdt_setprop_string(fdt, "/memory", "device_type", "memory");
+
+    /* Clock node, for the benefit of the UART. The kernel device tree
+     * binding documentation claims the PL011 node clock properties are
+     * optional but in practice if you omit them the kernel refuses to
+     * probe for the device.
+     */
+    vbi->clock_phandle = qemu_fdt_alloc_phandle(fdt);
+    qemu_fdt_add_subnode(fdt, "/apb-pclk");
+    qemu_fdt_setprop_string(fdt, "/apb-pclk", "compatible", "fixed-clock");
+    qemu_fdt_setprop_cell(fdt, "/apb-pclk", "#clock-cells", 0x0);
+    qemu_fdt_setprop_cell(fdt, "/apb-pclk", "clock-frequency", 24000000);
+    qemu_fdt_setprop_string(fdt, "/apb-pclk", "clock-output-names",
+                                "clk24mhz");
+    qemu_fdt_setprop_cell(fdt, "/apb-pclk", "phandle", vbi->clock_phandle);
+
+    /* No PSCI for TCG yet */
+    if (kvm_enabled()) {
+        qemu_fdt_add_subnode(fdt, "/psci");
+        qemu_fdt_setprop_string(fdt, "/psci", "compatible", "arm,psci");
+        qemu_fdt_setprop_string(fdt, "/psci", "method", "hvc");
+        qemu_fdt_setprop_cell(fdt, "/psci", "cpu_suspend",
+                                  PSCI_FN_CPU_SUSPEND);
+        qemu_fdt_setprop_cell(fdt, "/psci", "cpu_off", PSCI_FN_CPU_OFF);
+        qemu_fdt_setprop_cell(fdt, "/psci", "cpu_on", PSCI_FN_CPU_ON);
+        qemu_fdt_setprop_cell(fdt, "/psci", "migrate", PSCI_FN_MIGRATE);
+    }
+}
+
+static void fdt_add_timer_nodes(const VirtBoardInfo *vbi)
+{
+    /* Note that on A15 h/w these interrupts are level-triggered,
+     * but for the GIC implementation provided by both QEMU and KVM
+     * they are edge-triggered.
+     */
+    uint32_t irqflags = GIC_FDT_IRQ_FLAGS_EDGE_LO_HI;
+
+    irqflags = deposit32(irqflags, GIC_FDT_IRQ_PPI_CPU_START,
+                         GIC_FDT_IRQ_PPI_CPU_WIDTH, (1 << vbi->smp_cpus) - 1);
+
+    qemu_fdt_add_subnode(vbi->fdt, "/timer");
+    qemu_fdt_setprop_string(vbi->fdt, "/timer",
+                                "compatible", "arm,armv7-timer");
+    qemu_fdt_setprop_cells(vbi->fdt, "/timer", "interrupts",
+                               GIC_FDT_IRQ_TYPE_PPI, 13, irqflags,
+                               GIC_FDT_IRQ_TYPE_PPI, 14, irqflags,
+                               GIC_FDT_IRQ_TYPE_PPI, 11, irqflags,
+                               GIC_FDT_IRQ_TYPE_PPI, 10, irqflags);
+}
+
+static void fdt_add_cpu_nodes(const VirtBoardInfo *vbi)
+{
+    int cpu;
+
+    qemu_fdt_add_subnode(vbi->fdt, "/cpus");
+    qemu_fdt_setprop_cell(vbi->fdt, "/cpus", "#address-cells", 0x1);
+    qemu_fdt_setprop_cell(vbi->fdt, "/cpus", "#size-cells", 0x0);
+
+    for (cpu = vbi->smp_cpus - 1; cpu >= 0; cpu--) {
+        char *nodename = g_strdup_printf("/cpus/cpu@%d", cpu);
+        ARMCPU *armcpu = ARM_CPU(qemu_get_cpu(cpu));
+
+        qemu_fdt_add_subnode(vbi->fdt, nodename);
+        qemu_fdt_setprop_string(vbi->fdt, nodename, "device_type", "cpu");
+        qemu_fdt_setprop_string(vbi->fdt, nodename, "compatible",
+                                    armcpu->dtb_compatible);
+
+        if (vbi->smp_cpus > 1) {
+            qemu_fdt_setprop_string(vbi->fdt, nodename,
+                                        "enable-method", "psci");
+        }
+
+        qemu_fdt_setprop_cell(vbi->fdt, nodename, "reg", cpu);
+        g_free(nodename);
+    }
+}
+
+static void fdt_add_gic_node(const VirtBoardInfo *vbi)
+{
+    uint32_t gic_phandle;
+
+    gic_phandle = qemu_fdt_alloc_phandle(vbi->fdt);
+    qemu_fdt_setprop_cell(vbi->fdt, "/", "interrupt-parent", gic_phandle);
+
+    qemu_fdt_add_subnode(vbi->fdt, "/intc");
+    qemu_fdt_setprop_string(vbi->fdt, "/intc", "compatible",
+                                vbi->gic_compatible);
+    qemu_fdt_setprop_cell(vbi->fdt, "/intc", "#interrupt-cells", 3);
+    qemu_fdt_setprop(vbi->fdt, "/intc", "interrupt-controller", NULL, 0);
+    qemu_fdt_setprop_sized_cells(vbi->fdt, "/intc", "reg",
+                                     2, vbi->memmap[VIRT_GIC_DIST].base,
+                                     2, vbi->memmap[VIRT_GIC_DIST].size,
+                                     2, vbi->memmap[VIRT_GIC_CPU].base,
+                                     2, vbi->memmap[VIRT_GIC_CPU].size);
+    qemu_fdt_setprop_cell(vbi->fdt, "/intc", "phandle", gic_phandle);
+}
+
+static void create_uart(const VirtBoardInfo *vbi, qemu_irq *pic)
+{
+    char *nodename;
+    hwaddr base = vbi->memmap[VIRT_UART].base;
+    hwaddr size = vbi->memmap[VIRT_UART].size;
+    int irq = vbi->irqmap[VIRT_UART];
+    const char compat[] = "arm,pl011\0arm,primecell";
+    const char clocknames[] = "uartclk\0apb_pclk";
+
+    sysbus_create_simple("pl011", base, pic[irq]);
+
+    nodename = g_strdup_printf("/pl011@%" PRIx64, base);
+    qemu_fdt_add_subnode(vbi->fdt, nodename);
+    /* Note that we can't use setprop_string because of the embedded NUL */
+    qemu_fdt_setprop(vbi->fdt, nodename, "compatible",
+                         compat, sizeof(compat));
+    qemu_fdt_setprop_sized_cells(vbi->fdt, nodename, "reg",
+                                     2, base, 2, size);
+    qemu_fdt_setprop_cells(vbi->fdt, nodename, "interrupts",
+                               GIC_FDT_IRQ_TYPE_SPI, irq,
+                               GIC_FDT_IRQ_FLAGS_EDGE_LO_HI);
+    qemu_fdt_setprop_cells(vbi->fdt, nodename, "clocks",
+                               vbi->clock_phandle, vbi->clock_phandle);
+    qemu_fdt_setprop(vbi->fdt, nodename, "clock-names",
+                         clocknames, sizeof(clocknames));
+    g_free(nodename);
+}
+
+static void create_virtio_devices(const VirtBoardInfo *vbi, qemu_irq *pic)
+{
+    int i;
+    hwaddr size = vbi->memmap[VIRT_MMIO].size;
+
+    /* Note that we have to create the transports in forwards order
+     * so that command line devices are inserted lowest address first,
+     * and then add dtb nodes in reverse order so that they appear in
+     * the finished device tree lowest address first.
+     */
+    for (i = 0; i < NUM_VIRTIO_TRANSPORTS; i++) {
+        int irq = vbi->irqmap[VIRT_MMIO] + i;
+        hwaddr base = vbi->memmap[VIRT_MMIO].base + i * size;
+
+        sysbus_create_simple("virtio-mmio", base, pic[irq]);
+    }
+
+    for (i = NUM_VIRTIO_TRANSPORTS - 1; i >= 0; i--) {
+        char *nodename;
+        int irq = vbi->irqmap[VIRT_MMIO] + i;
+        hwaddr base = vbi->memmap[VIRT_MMIO].base + i * size;
+
+        nodename = g_strdup_printf("/virtio_mmio@%" PRIx64, base);
+        qemu_fdt_add_subnode(vbi->fdt, nodename);
+        qemu_fdt_setprop_string(vbi->fdt, nodename,
+                                "compatible", "virtio,mmio");
+        qemu_fdt_setprop_sized_cells(vbi->fdt, nodename, "reg",
+                                     2, base, 2, size);
+        qemu_fdt_setprop_cells(vbi->fdt, nodename, "interrupts",
+                               GIC_FDT_IRQ_TYPE_SPI, irq,
+                               GIC_FDT_IRQ_FLAGS_EDGE_LO_HI);
+        g_free(nodename);
+    }
+}
+
+static void *machvirt_dtb(const struct arm_boot_info *binfo, int *fdt_size)
+{
+    const VirtBoardInfo *board = (const VirtBoardInfo *)binfo;
+
+    *fdt_size = board->fdt_size;
+    return board->fdt;
+}
+
+static void machvirt_init(QEMUMachineInitArgs *args)
+{
+    qemu_irq pic[NUM_IRQS];
+    MemoryRegion *sysmem = get_system_memory();
+    int n;
+    MemoryRegion *ram = g_new(MemoryRegion, 1);
+    DeviceState *dev;
+    SysBusDevice *busdev;
+    const char *cpu_model = args->cpu_model;
+    VirtBoardInfo *vbi;
+
+    if (!cpu_model) {
+        cpu_model = "cortex-a15";
+    }
+
+    vbi = find_machine_info(cpu_model);
+
+    if (!vbi) {
+        error_report("mach-virt: CPU %s not supported", cpu_model);
+        exit(1);
+    }
+
+    vbi->smp_cpus = smp_cpus;
+
+    /*
+     * Only supported method of starting secondary CPUs is PSCI and
+     * PSCI is not yet supported with TCG, so limit smp_cpus to 1
+     * if we're not using KVM.
+     */
+    if (!kvm_enabled() && smp_cpus > 1) {
+        error_report("mach-virt: must enable KVM to use multiple CPUs");
+        exit(1);
+    }
+
+    if (args->ram_size > vbi->memmap[VIRT_MEM].size) {
+        error_report("mach-virt: cannot model more than 30GB RAM");
+        exit(1);
+    }
+
+    create_fdt(vbi);
+    fdt_add_timer_nodes(vbi);
+
+    for (n = 0; n < smp_cpus; n++) {
+        ObjectClass *oc = cpu_class_by_name(TYPE_ARM_CPU, cpu_model);
+        Object *cpuobj;
+
+        if (!oc) {
+            fprintf(stderr, "Unable to find CPU definition\n");
+            exit(1);
+        }
+        cpuobj = object_new(object_class_get_name(oc));
+
+        /* Secondary CPUs start in PSCI powered-down state */
+        if (n > 0) {
+            object_property_set_bool(cpuobj, true, "start-powered-off", NULL);
+        }
+        object_property_set_bool(cpuobj, true, "realized", NULL);
+    }
+    fdt_add_cpu_nodes(vbi);
+
+    memory_region_init_ram(ram, NULL, "mach-virt.ram", args->ram_size);
+    vmstate_register_ram_global(ram);
+    memory_region_add_subregion(sysmem, vbi->memmap[VIRT_MEM].base, ram);
+
+    dev = qdev_create(NULL, vbi->qdevname);
+    qdev_prop_set_uint32(dev, "num-cpu", smp_cpus);
+    /* Note that the num-irq property counts both internal and external
+     * interrupts; there are always 32 of the former (mandated by GIC spec).
+     */
+    qdev_prop_set_uint32(dev, "num-irq", NUM_IRQS + 32);
+    qdev_init_nofail(dev);
+    busdev = SYS_BUS_DEVICE(dev);
+    sysbus_mmio_map(busdev, 0, vbi->memmap[VIRT_CPUPERIPHS].base);
+    fdt_add_gic_node(vbi);
+    for (n = 0; n < smp_cpus; n++) {
+        DeviceState *cpudev = DEVICE(qemu_get_cpu(n));
+
+        sysbus_connect_irq(busdev, n, qdev_get_gpio_in(cpudev, ARM_CPU_IRQ));
+    }
+
+    for (n = 0; n < NUM_IRQS; n++) {
+        pic[n] = qdev_get_gpio_in(dev, n);
+    }
+
+    create_uart(vbi, pic);
+
+    /* Create mmio transports, so the user can create virtio backends
+     * (which will be automatically plugged in to the transports). If
+     * no backend is created the transport will just sit harmlessly idle.
+     */
+    create_virtio_devices(vbi, pic);
+
+    vbi->bootinfo.ram_size = args->ram_size;
+    vbi->bootinfo.kernel_filename = args->kernel_filename;
+    vbi->bootinfo.kernel_cmdline = args->kernel_cmdline;
+    vbi->bootinfo.initrd_filename = args->initrd_filename;
+    vbi->bootinfo.nb_cpus = smp_cpus;
+    vbi->bootinfo.board_id = -1;
+    vbi->bootinfo.loader_start = vbi->memmap[VIRT_MEM].base;
+    vbi->bootinfo.get_dtb = machvirt_dtb;
+    arm_load_kernel(ARM_CPU(first_cpu), &vbi->bootinfo);
+}
+
+static QEMUMachine machvirt_a15_machine = {
+    .name = "virt",
+    .desc = "ARM Virtual Machine",
+    .init = machvirt_init,
+    .max_cpus = 4,
+};
+
+static void machvirt_machine_init(void)
+{
+    qemu_register_machine(&machvirt_a15_machine);
+}
+
+machine_init(machvirt_machine_init);

hw/arm/xilinx_zynq.c

@@ -25,6 +25,7 @@
 #include "sysemu/blockdev.h"
 #include "hw/loader.h"
 #include "hw/ssi.h"
+#include "qemu/error-report.h"
 
 #define NUM_SPI_FLASHES 4
 #define NUM_QSPI_FLASHES 2
@@ -35,6 +36,8 @@
 
 #define IRQ_OFFSET 32 /* pic interrupts start from index 32 */
 
+#define MPCORE_PERIPHBASE 0xF8F00000
+
 static const int dma_irqs[8] = {
     46, 47, 48, 49, 72, 73, 74, 75
 };
@@ -46,9 +49,11 @@ static void gem_init(NICInfo *nd, uint32_t base, qemu_irq irq)
     DeviceState *dev;
     SysBusDevice *s;
 
-    qemu_check_nic_model(nd, "cadence_gem");
     dev = qdev_create(NULL, "cadence_gem");
-    qdev_set_nic_properties(dev, nd);
+    if (nd->used) {
+        qemu_check_nic_model(nd, "cadence_gem");
+        qdev_set_nic_properties(dev, nd);
+    }
     qdev_init_nofail(dev);
     s = SYS_BUS_DEVICE(dev);
     sysbus_mmio_map(s, 0, base);
@@ -102,6 +107,7 @@ static void zynq_init(QEMUMachineInitArgs *args)
     const char *kernel_filename = args->kernel_filename;
     const char *kernel_cmdline = args->kernel_cmdline;
     const char *initrd_filename = args->initrd_filename;
+    ObjectClass *cpu_oc;
     ARMCPU *cpu;
     MemoryRegion *address_space_mem = get_system_memory();
     MemoryRegion *ext_ram = g_new(MemoryRegion, 1);
@@ -109,16 +115,24 @@ static void zynq_init(QEMUMachineInitArgs *args)
     DeviceState *dev;
     SysBusDevice *busdev;
     qemu_irq pic[64];
-    NICInfo *nd;
+    Error *err = NULL;
     int n;
 
     if (!cpu_model) {
         cpu_model = "cortex-a9";
     }
+    cpu_oc = cpu_class_by_name(TYPE_ARM_CPU, cpu_model);
 
-    cpu = cpu_arm_init(cpu_model);
-    if (!cpu) {
-        fprintf(stderr, "Unable to find CPU definition\n");
+    cpu = ARM_CPU(object_new(object_class_get_name(cpu_oc)));
+
+    object_property_set_int(OBJECT(cpu), MPCORE_PERIPHBASE, "reset-cbar", &err);
+    if (err) {
+        error_report("%s", error_get_pretty(err));
+        exit(1);
+    }
+    object_property_set_bool(OBJECT(cpu), true, "realized", &err);
+    if (err) {
+        error_report("%s", error_get_pretty(err));
         exit(1);
     }
 
@@ -154,7 +168,7 @@ static void zynq_init(QEMUMachineInitArgs *args)
     qdev_prop_set_uint32(dev, "num-cpu", 1);
     qdev_init_nofail(dev);
     busdev = SYS_BUS_DEVICE(dev);
-    sysbus_mmio_map(busdev, 0, 0xF8F00000);
+    sysbus_mmio_map(busdev, 0, MPCORE_PERIPHBASE);
     sysbus_connect_irq(busdev, 0,
                        qdev_get_gpio_in(DEVICE(cpu), ARM_CPU_IRQ));
 
@@ -177,14 +191,8 @@ static void zynq_init(QEMUMachineInitArgs *args)
     sysbus_create_varargs("cadence_ttc", 0xF8002000,
             pic[69-IRQ_OFFSET], pic[70-IRQ_OFFSET], pic[71-IRQ_OFFSET], NULL);
 
-    for (n = 0; n < nb_nics; n++) {
-        nd = &nd_table[n];
-        if (n == 0) {
-            gem_init(nd, 0xE000B000, pic[54-IRQ_OFFSET]);
-        } else if (n == 1) {
-            gem_init(nd, 0xE000C000, pic[77-IRQ_OFFSET]);
-        }
-    }
+    gem_init(&nd_table[0], 0xE000B000, pic[54-IRQ_OFFSET]);
+    gem_init(&nd_table[1], 0xE000C000, pic[77-IRQ_OFFSET]);
 
     dev = qdev_create(NULL, "generic-sdhci");
     qdev_init_nofail(dev);

hw/arm/z2.c

@@ -33,7 +33,7 @@
 #define DPRINTF(fmt, ...)
 #endif
 
-static struct keymap map[0x100] = {
+static const struct keymap map[0x100] = {
     [0 ... 0xff] = { -1, -1 },
     [0x3b] = {0, 0}, /* Option = F1 */
     [0xc8] = {0, 1}, /* Up */

hw/audio/adlib.c

@@ -347,8 +347,8 @@ static void adlib_realizefn (DeviceState *dev, Error **errp)
     s->samples = AUD_get_buffer_size_out (s->voice) >> SHIFT;
     s->mixbuf = g_malloc0 (s->samples << SHIFT);
 
-    adlib_portio_list[1].offset = s->port;
-    adlib_portio_list[2].offset = s->port + 8;
+    adlib_portio_list[0].offset = s->port;
+    adlib_portio_list[1].offset = s->port + 8;
     portio_list_init (port_list, OBJECT(s), adlib_portio_list, s, "adlib");
     portio_list_add (port_list, isa_address_space_io(&s->parent_obj), 0);
 }

hw/audio/intel-hda.c

@@ -444,6 +444,7 @@ static bool intel_hda_xfer(HDACodecDevice *dev, uint32_t stnr, bool output,
         }
     }
     if (d->dp_lbase & 0x01) {
+        s = st - d->st;
         addr = intel_hda_addr(d->dp_lbase & ~0x01, d->dp_ubase);
         stl_le_pci_dma(&d->pci, addr + 8*s, st->lpib);
     }
@@ -899,7 +900,7 @@ static const IntelHDAReg *intel_hda_reg_find(IntelHDAState *d, hwaddr addr)
 {
     const IntelHDAReg *reg;
 
-    if (addr >= sizeof(regtab)/sizeof(regtab[0])) {
+    if (addr >= ARRAY_SIZE(regtab)) {
         goto noreg;
     }
     reg = regtab+addr;
@@ -1024,7 +1025,7 @@ static void intel_hda_regs_reset(IntelHDAState *d)
     uint32_t *addr;
     int i;
 
-    for (i = 0; i < sizeof(regtab)/sizeof(regtab[0]); i++) {
+    for (i = 0; i < ARRAY_SIZE(regtab); i++) {
         if (regtab[i].name == NULL) {
             continue;
         }

hw/audio/marvell_88w8618.c

@@ -288,6 +288,8 @@ static void mv88w8618_audio_class_init(ObjectClass *klass, void *data)
     dc->reset = mv88w8618_audio_reset;
     dc->vmsd = &mv88w8618_audio_vmsd;
     dc->props = mv88w8618_audio_properties;
+    /* Reason: pointer property "wm8750" */
+    dc->cannot_instantiate_with_device_add_yet = true;
 }
 
 static const TypeInfo mv88w8618_audio_info = {

hw/audio/pcspk.c

@@ -192,8 +192,9 @@ static void pcspk_class_initfn(ObjectClass *klass, void *data)
 
     dc->realize = pcspk_realizefn;
     set_bit(DEVICE_CATEGORY_SOUND, dc->categories);
-    dc->no_user = 1;
     dc->props = pcspk_properties;
+    /* Reason: pointer property "pit", realize sets global pcspk_state */
+    dc->cannot_instantiate_with_device_add_yet = true;
 }
 
 static const TypeInfo pcspk_info = {

hw/audio/pl041.c

@@ -632,7 +632,6 @@ static void pl041_device_class_init(ObjectClass *klass, void *data)
 
     k->init = pl041_init;
     set_bit(DEVICE_CATEGORY_SOUND, dc->categories);
-    dc->no_user = 1;
     dc->reset = pl041_device_reset;
     dc->vmsd = &vmstate_pl041;
     dc->props = pl041_device_properties;

hw/block/dataplane/virtio-blk.c

@@ -35,7 +35,7 @@ enum {
 typedef struct {
     struct iocb iocb;               /* Linux AIO control block */
     QEMUIOVector *inhdr;            /* iovecs for virtio_blk_inhdr */
-    unsigned int head;              /* vring descriptor index */
+    VirtQueueElement *elem;         /* saved data from the virtqueue */
     struct iovec *bounce_iov;       /* used if guest buffers are unaligned */
     QEMUIOVector *read_qiov;        /* for read completion /w bounce buffer */
 } VirtIOBlockRequest;
@@ -96,7 +96,7 @@ static void complete_request(struct iocb *iocb, ssize_t ret, void *opaque)
         len = 0;
     }
 
-    trace_virtio_blk_data_plane_complete_request(s, req->head, ret);
+    trace_virtio_blk_data_plane_complete_request(s, req->elem->index, ret);
 
     if (req->read_qiov) {
         assert(req->bounce_iov);
@@ -118,12 +118,12 @@ static void complete_request(struct iocb *iocb, ssize_t ret, void *opaque)
      * written to, but for virtio-blk it seems to be the number of bytes
      * transferred plus the status bytes.
      */
-    vring_push(&s->vring, req->head, len + sizeof(hdr));
-
+    vring_push(&s->vring, req->elem, len + sizeof(hdr));
+    req->elem = NULL;
     s->num_reqs--;
 }
 
-static void complete_request_early(VirtIOBlockDataPlane *s, unsigned int head,
+static void complete_request_early(VirtIOBlockDataPlane *s, VirtQueueElement *elem,
                                    QEMUIOVector *inhdr, unsigned char status)
 {
     struct virtio_blk_inhdr hdr = {
@@ -134,26 +134,26 @@ static void complete_request_early(VirtIOBlockDataPlane *s, unsigned int head,
     qemu_iovec_destroy(inhdr);
     g_slice_free(QEMUIOVector, inhdr);
 
-    vring_push(&s->vring, head, sizeof(hdr));
+    vring_push(&s->vring, elem, sizeof(hdr));
     notify_guest(s);
 }
 
 /* Get disk serial number */
 static void do_get_id_cmd(VirtIOBlockDataPlane *s,
                           struct iovec *iov, unsigned int iov_cnt,
-                          unsigned int head, QEMUIOVector *inhdr)
+                          VirtQueueElement *elem, QEMUIOVector *inhdr)
 {
     char id[VIRTIO_BLK_ID_BYTES];
 
     /* Serial number not NUL-terminated when shorter than buffer */
     strncpy(id, s->blk->serial ? s->blk->serial : "", sizeof(id));
     iov_from_buf(iov, iov_cnt, 0, id, sizeof(id));
-    complete_request_early(s, head, inhdr, VIRTIO_BLK_S_OK);
+    complete_request_early(s, elem, inhdr, VIRTIO_BLK_S_OK);
 }
 
 static int do_rdwr_cmd(VirtIOBlockDataPlane *s, bool read,
-                       struct iovec *iov, unsigned int iov_cnt,
-                       long long offset, unsigned int head,
+                       struct iovec *iov, unsigned iov_cnt,
+                       long long offset, VirtQueueElement *elem,
                        QEMUIOVector *inhdr)
 {
     struct iocb *iocb;
@@ -186,19 +186,20 @@ static int do_rdwr_cmd(VirtIOBlockDataPlane *s, bool read,
 
     /* Fill in virtio block metadata needed for completion */
     VirtIOBlockRequest *req = container_of(iocb, VirtIOBlockRequest, iocb);
-    req->head = head;
+    req->elem = elem;
     req->inhdr = inhdr;
     req->bounce_iov = bounce_iov;
     req->read_qiov = read_qiov;
     return 0;
 }
 
-static int process_request(IOQueue *ioq, struct iovec iov[],
-                           unsigned int out_num, unsigned int in_num,
-                           unsigned int head)
+static int process_request(IOQueue *ioq, VirtQueueElement *elem)
 {
     VirtIOBlockDataPlane *s = container_of(ioq, VirtIOBlockDataPlane, ioqueue);
-    struct iovec *in_iov = &iov[out_num];
+    struct iovec *iov = elem->out_sg;
+    struct iovec *in_iov = elem->in_sg;
+    unsigned out_num = elem->out_num;
+    unsigned in_num = elem->in_num;
     struct virtio_blk_outhdr outhdr;
     QEMUIOVector *inhdr;
     size_t in_size;
@@ -229,29 +230,29 @@ static int process_request(IOQueue *ioq, struct iovec iov[],
 
     switch (outhdr.type) {
     case VIRTIO_BLK_T_IN:
-        do_rdwr_cmd(s, true, in_iov, in_num, outhdr.sector * 512, head, inhdr);
+        do_rdwr_cmd(s, true, in_iov, in_num, outhdr.sector * 512, elem, inhdr);
         return 0;
 
     case VIRTIO_BLK_T_OUT:
-        do_rdwr_cmd(s, false, iov, out_num, outhdr.sector * 512, head, inhdr);
+        do_rdwr_cmd(s, false, iov, out_num, outhdr.sector * 512, elem, inhdr);
         return 0;
 
     case VIRTIO_BLK_T_SCSI_CMD:
         /* TODO support SCSI commands */
-        complete_request_early(s, head, inhdr, VIRTIO_BLK_S_UNSUPP);
+        complete_request_early(s, elem, inhdr, VIRTIO_BLK_S_UNSUPP);
         return 0;
 
     case VIRTIO_BLK_T_FLUSH:
         /* TODO fdsync not supported by Linux AIO, do it synchronously here! */
         if (qemu_fdatasync(s->fd) < 0) {
-            complete_request_early(s, head, inhdr, VIRTIO_BLK_S_IOERR);
+            complete_request_early(s, elem, inhdr, VIRTIO_BLK_S_IOERR);
         } else {
-            complete_request_early(s, head, inhdr, VIRTIO_BLK_S_OK);
+            complete_request_early(s, elem, inhdr, VIRTIO_BLK_S_OK);
         }
         return 0;
 
     case VIRTIO_BLK_T_GET_ID:
-        do_get_id_cmd(s, in_iov, in_num, head, inhdr);
+        do_get_id_cmd(s, in_iov, in_num, elem, inhdr);
         return 0;
 
     default:
@@ -267,29 +268,8 @@ static void handle_notify(EventNotifier *e)
     VirtIOBlockDataPlane *s = container_of(e, VirtIOBlockDataPlane,
                                            host_notifier);
 
-    /* There is one array of iovecs into which all new requests are extracted
-     * from the vring.  Requests are read from the vring and the translated
-     * descriptors are written to the iovecs array.  The iovecs do not have to
-     * persist across handle_notify() calls because the kernel copies the
-     * iovecs on io_submit().
-     *
-     * Handling io_submit() EAGAIN may require storing the requests across
-     * handle_notify() calls until the kernel has sufficient resources to
-     * accept more I/O.  This is not implemented yet.
-     */
-    struct iovec iovec[VRING_MAX];
-    struct iovec *end = &iovec[VRING_MAX];
-    struct iovec *iov = iovec;
-
-    /* When a request is read from the vring, the index of the first descriptor
-     * (aka head) is returned so that the completed request can be pushed onto
-     * the vring later.
-     *
-     * The number of hypervisor read-only iovecs is out_num.  The number of
-     * hypervisor write-only iovecs is in_num.
-     */
-    int head;
-    unsigned int out_num = 0, in_num = 0;
+    VirtQueueElement *elem;
+    int ret;
     unsigned int num_queued;
 
     event_notifier_test_and_clear(&s->host_notifier);
@@ -298,29 +278,31 @@ static void handle_notify(EventNotifier *e)
         vring_disable_notification(s->vdev, &s->vring);
 
         for (;;) {
-            head = vring_pop(s->vdev, &s->vring, iov, end, &out_num, &in_num);
-            if (head < 0) {
+            ret = vring_pop(s->vdev, &s->vring, &elem);
+            if (ret < 0) {
+                assert(elem == NULL);
                 break; /* no more requests */
             }
 
-            trace_virtio_blk_data_plane_process_request(s, out_num, in_num,
-                                                        head);
+            trace_virtio_blk_data_plane_process_request(s, elem->out_num,
+                                                        elem->in_num, elem->index);
 
-            if (process_request(&s->ioqueue, iov, out_num, in_num, head) < 0) {
+            if (process_request(&s->ioqueue, elem) < 0) {
                 vring_set_broken(&s->vring);
+                vring_free_element(elem);
+                ret = -EFAULT;
                 break;
             }
-            iov += out_num + in_num;
         }
 
-        if (likely(head == -EAGAIN)) { /* vring emptied */
+        if (likely(ret == -EAGAIN)) { /* vring emptied */
             /* Re-enable guest->host notifies and stop processing the vring.
              * But if the guest has snuck in more descriptors, keep processing.
              */
             if (vring_enable_notification(s->vdev, &s->vring)) {
                 break;
             }
-        } else { /* head == -ENOBUFS or fatal error, iovecs[] is depleted */
+        } else { /* ret == -ENOBUFS or fatal error, iovecs[] is depleted */
             /* Since there are no iovecs[] left, stop processing for now.  Do
              * not re-enable guest->host notifies since the I/O completion
              * handler knows to check for more vring descriptors anyway.
@@ -380,8 +362,9 @@ static void start_data_plane_bh(void *opaque)
                        s, QEMU_THREAD_JOINABLE);
 }
 
-bool virtio_blk_data_plane_create(VirtIODevice *vdev, VirtIOBlkConf *blk,
-                                  VirtIOBlockDataPlane **dataplane)
+void virtio_blk_data_plane_create(VirtIODevice *vdev, VirtIOBlkConf *blk,
+                                  VirtIOBlockDataPlane **dataplane,
+                                  Error **errp)
 {
     VirtIOBlockDataPlane *s;
     int fd;
@@ -389,33 +372,35 @@ bool virtio_blk_data_plane_create(VirtIODevice *vdev, VirtIOBlkConf *blk,
     *dataplane = NULL;
 
     if (!blk->data_plane) {
-        return true;
+        return;
     }
 
     if (blk->scsi) {
-        error_report("device is incompatible with x-data-plane, use scsi=off");
-        return false;
+        error_setg(errp,
+                   "device is incompatible with x-data-plane, use scsi=off");
+        return;
     }
 
     if (blk->config_wce) {
-        error_report("device is incompatible with x-data-plane, "
-                     "use config-wce=off");
-        return false;
+        error_setg(errp, "device is incompatible with x-data-plane, "
+                         "use config-wce=off");
+        return;
     }
 
     /* If dataplane is (re-)enabled while the guest is running there could be
      * block jobs that can conflict.
      */
     if (bdrv_in_use(blk->conf.bs)) {
-        error_report("cannot start dataplane thread while device is in use");
-        return false;
+        error_setg(errp,
+                   "cannot start dataplane thread while device is in use");
+        return;
     }
 
     fd = raw_get_aio_fd(blk->conf.bs);
     if (fd < 0) {
-        error_report("drive is incompatible with x-data-plane, "
-                     "use format=raw,cache=none,aio=native");
-        return false;
+        error_setg(errp, "drive is incompatible with x-data-plane, "
+                         "use format=raw,cache=none,aio=native");
+        return;
     }
 
     s = g_new0(VirtIOBlockDataPlane, 1);
@@ -427,7 +412,6 @@ bool virtio_blk_data_plane_create(VirtIODevice *vdev, VirtIOBlkConf *blk,
     bdrv_set_in_use(blk->conf.bs, 1);
 
     *dataplane = s;
-    return true;
 }
 
 void virtio_blk_data_plane_destroy(VirtIOBlockDataPlane *s)

hw/block/dataplane/virtio-blk.h

@@ -19,8 +19,9 @@
 
 typedef struct VirtIOBlockDataPlane VirtIOBlockDataPlane;
 
-bool virtio_blk_data_plane_create(VirtIODevice *vdev, VirtIOBlkConf *blk,
-                                  VirtIOBlockDataPlane **dataplane);
+void virtio_blk_data_plane_create(VirtIODevice *vdev, VirtIOBlkConf *blk,
+                                  VirtIOBlockDataPlane **dataplane,
+                                  Error **errp);
 void virtio_blk_data_plane_destroy(VirtIOBlockDataPlane *s);
 void virtio_blk_data_plane_start(VirtIOBlockDataPlane *s);
 void virtio_blk_data_plane_stop(VirtIOBlockDataPlane *s);

hw/block/fdc.c

@@ -2234,7 +2234,6 @@ static void isabus_fdc_class_init(ObjectClass *klass, void *data)
 
     dc->realize = isabus_fdc_realize;
     dc->fw_name = "fdc";
-    dc->no_user = 1;
     dc->reset = fdctrl_external_reset_isa;
     dc->vmsd = &vmstate_isa_fdc;
     dc->props = isa_fdc_properties;

hw/block/pflash_cfi01.c

@@ -40,6 +40,7 @@
 #include "hw/block/flash.h"
 #include "block/block.h"
 #include "qemu/timer.h"
+#include "qemu/bitops.h"
 #include "exec/address-spaces.h"
 #include "qemu/host-utils.h"
 #include "hw/sysbus.h"
@@ -71,7 +72,9 @@ struct pflash_t {
     BlockDriverState *bs;
     uint32_t nb_blocs;
     uint64_t sector_len;
-    uint8_t width;
+    uint8_t bank_width;
+    uint8_t device_width; /* If 0, device width not specified. */
+    uint8_t max_device_width;  /* max device width in bytes */
     uint8_t be;
     uint8_t wcycle; /* if 0, the flash is read normally */
     int ro;
@@ -116,6 +119,119 @@ static void pflash_timer (void *opaque)
     pfl->cmd = 0;
 }
 
+/* Perform a CFI query based on the bank width of the flash.
+ * If this code is called we know we have a device_width set for
+ * this flash.
+ */
+static uint32_t pflash_cfi_query(pflash_t *pfl, hwaddr offset)
+{
+    int i;
+    uint32_t resp = 0;
+    hwaddr boff;
+
+    /* Adjust incoming offset to match expected device-width
+     * addressing. CFI query addresses are always specified in terms of
+     * the maximum supported width of the device.  This means that x8
+     * devices and x8/x16 devices in x8 mode behave differently.  For
+     * devices that are not used at their max width, we will be
+     * provided with addresses that use higher address bits than
+     * expected (based on the max width), so we will shift them lower
+     * so that they will match the addresses used when
+     * device_width==max_device_width.
+     */
+    boff = offset >> (ctz32(pfl->bank_width) +
+                      ctz32(pfl->max_device_width) - ctz32(pfl->device_width));
+
+    if (boff > pfl->cfi_len) {
+        return 0;
+    }
+    /* Now we will construct the CFI response generated by a single
+     * device, then replicate that for all devices that make up the
+     * bus.  For wide parts used in x8 mode, CFI query responses
+     * are different than native byte-wide parts.
+     */
+    resp = pfl->cfi_table[boff];
+    if (pfl->device_width != pfl->max_device_width) {
+        /* The only case currently supported is x8 mode for a
+         * wider part.
+         */
+        if (pfl->device_width != 1 || pfl->bank_width > 4) {
+            DPRINTF("%s: Unsupported device configuration: "
+                    "device_width=%d, max_device_width=%d\n",
+                    __func__, pfl->device_width,
+                    pfl->max_device_width);
+            return 0;
+        }
+        /* CFI query data is repeated, rather than zero padded for
+         * wide devices used in x8 mode.
+         */
+        for (i = 1; i < pfl->max_device_width; i++) {
+            resp = deposit32(resp, 8 * i, 8, pfl->cfi_table[boff]);
+        }
+    }
+    /* Replicate responses for each device in bank. */
+    if (pfl->device_width < pfl->bank_width) {
+        for (i = pfl->device_width;
+             i < pfl->bank_width; i += pfl->device_width) {
+            resp = deposit32(resp, 8 * i, 8 * pfl->device_width, resp);
+        }
+    }
+
+    return resp;
+}
+
+
+
+/* Perform a device id query based on the bank width of the flash. */
+static uint32_t pflash_devid_query(pflash_t *pfl, hwaddr offset)
+{
+    int i;
+    uint32_t resp;
+    hwaddr boff;
+
+    /* Adjust incoming offset to match expected device-width
+     * addressing. Device ID read addresses are always specified in
+     * terms of the maximum supported width of the device.  This means
+     * that x8 devices and x8/x16 devices in x8 mode behave
+     * differently. For devices that are not used at their max width,
+     * we will be provided with addresses that use higher address bits
+     * than expected (based on the max width), so we will shift them
+     * lower so that they will match the addresses used when
+     * device_width==max_device_width.
+     */
+    boff = offset >> (ctz32(pfl->bank_width) +
+                      ctz32(pfl->max_device_width) - ctz32(pfl->device_width));
+
+    /* Mask off upper bits which may be used in to query block
+     * or sector lock status at other addresses.
+     * Offsets 2/3 are block lock status, is not emulated.
+     */
+    switch (boff & 0xFF) {
+    case 0:
+        resp = pfl->ident0;
+        DPRINTF("%s: Manufacturer Code %04x\n", __func__, ret);
+        break;
+    case 1:
+        resp = pfl->ident1;
+        DPRINTF("%s: Device ID Code %04x\n", __func__, ret);
+        break;
+    default:
+        DPRINTF("%s: Read Device Information offset=%x\n", __func__,
+                (unsigned)offset);
+        return 0;
+        break;
+    }
+    /* Replicate responses for each device in bank. */
+    if (pfl->device_width < pfl->bank_width) {
+        for (i = pfl->device_width;
+              i < pfl->bank_width; i += pfl->device_width) {
+            resp = deposit32(resp, 8 * i, 8 * pfl->device_width, resp);
+        }
+    }
+
+    return resp;
+}
+
 static uint32_t pflash_read (pflash_t *pfl, hwaddr offset,
                              int width, int be)
 {
@@ -124,12 +240,6 @@ static uint32_t pflash_read (pflash_t *pfl, hwaddr offset,
     uint8_t *p;
 
     ret = -1;
-    boff = offset & 0xFF; /* why this here ?? */
-
-    if (pfl->width == 2)
-        boff = boff >> 1;
-    else if (pfl->width == 4)
-        boff = boff >> 2;
 
 #if 0
     DPRINTF("%s: reading offset " TARGET_FMT_plx " under cmd %02x width %d\n",
@@ -190,35 +300,88 @@ static uint32_t pflash_read (pflash_t *pfl, hwaddr offset,
     case 0x60: /* Block /un)lock */
     case 0x70: /* Status Register */
     case 0xe8: /* Write block */
-        /* Status register read */
+        /* Status register read.  Return status from each device in
+         * bank.
+         */
         ret = pfl->status;
-        if (width > 2) {
+        if (pfl->device_width && width > pfl->device_width) {
+            int shift = pfl->device_width * 8;
+            while (shift + pfl->device_width * 8 <= width * 8) {
+                ret |= pfl->status << shift;
+                shift += pfl->device_width * 8;
+            }
+        } else if (!pfl->device_width && width > 2) {
+            /* Handle 32 bit flash cases where device width is not
+             * set. (Existing behavior before device width added.)
+             */
             ret |= pfl->status << 16;
         }
         DPRINTF("%s: status %x\n", __func__, ret);
         break;
     case 0x90:
-        switch (boff) {
-        case 0:
-            ret = pfl->ident0 << 8 | pfl->ident1;
-            DPRINTF("%s: Manufacturer Code %04x\n", __func__, ret);
-            break;
-        case 1:
-            ret = pfl->ident2 << 8 | pfl->ident3;
-            DPRINTF("%s: Device ID Code %04x\n", __func__, ret);
-            break;
-        default:
-            DPRINTF("%s: Read Device Information boff=%x\n", __func__,
-                    (unsigned)boff);
-            ret = 0;
-            break;
+        if (!pfl->device_width) {
+            /* Preserve old behavior if device width not specified */
+            boff = offset & 0xFF;
+            if (pfl->bank_width == 2) {
+                boff = boff >> 1;
+            } else if (pfl->bank_width == 4) {
+                boff = boff >> 2;
+            }
+
+            switch (boff) {
+            case 0:
+                ret = pfl->ident0 << 8 | pfl->ident1;
+                DPRINTF("%s: Manufacturer Code %04x\n", __func__, ret);
+                break;
+            case 1:
+                ret = pfl->ident2 << 8 | pfl->ident3;
+                DPRINTF("%s: Device ID Code %04x\n", __func__, ret);
+                break;
+            default:
+                DPRINTF("%s: Read Device Information boff=%x\n", __func__,
+                        (unsigned)boff);
+                ret = 0;
+                break;
+            }
+        } else {
+            /* If we have a read larger than the bank_width, combine multiple
+             * manufacturer/device ID queries into a single response.
+             */
+            int i;
+            for (i = 0; i < width; i += pfl->bank_width) {
+                ret = deposit32(ret, i * 8, pfl->bank_width * 8,
+                                pflash_devid_query(pfl,
+                                                 offset + i * pfl->bank_width));
+            }
         }
         break;
     case 0x98: /* Query mode */
-        if (boff > pfl->cfi_len)
-            ret = 0;
-        else
-            ret = pfl->cfi_table[boff];
+        if (!pfl->device_width) {
+            /* Preserve old behavior if device width not specified */
+            boff = offset & 0xFF;
+            if (pfl->bank_width == 2) {
+                boff = boff >> 1;
+            } else if (pfl->bank_width == 4) {
+                boff = boff >> 2;
+            }
+
+            if (boff > pfl->cfi_len) {
+                ret = 0;
+            } else {
+                ret = pfl->cfi_table[boff];
+            }
+        } else {
+            /* If we have a read larger than the bank_width, combine multiple
+             * CFI queries into a single response.
+             */
+            int i;
+            for (i = 0; i < width; i += pfl->bank_width) {
+                ret = deposit32(ret, i * 8, pfl->bank_width * 8,
+                                pflash_cfi_query(pfl,
+                                                 offset + i * pfl->bank_width));
+            }
+        }
+
         break;
     }
     return ret;
@@ -378,6 +541,14 @@ static void pflash_write(pflash_t *pfl, hwaddr offset,
 
             break;
         case 0xe8:
+            /* Mask writeblock size based on device width, or bank width if
+             * device width not specified.
+             */
+            if (pfl->device_width) {
+                value = extract32(value, 0, pfl->device_width * 8);
+            } else {
+                value = extract32(value, 0, pfl->bank_width * 8);
+            }
             DPRINTF("%s: block write of %x bytes\n", __func__, value);
             pfl->counter = value;
             pfl->wcycle++;
@@ -613,6 +784,13 @@ static void pflash_cfi01_realize(DeviceState *dev, Error **errp)
         pfl->ro = 0;
     }
 
+    /* Default to devices being used at their maximum device width. This was
+     * assumed before the device_width support was added.
+     */
+    if (!pfl->max_device_width) {
+        pfl->max_device_width = pfl->device_width;
+    }
+
     pfl->timer = timer_new_ns(QEMU_CLOCK_VIRTUAL, pflash_timer, pfl);
     pfl->wcycle = 0;
     pfl->cmd = 0;
@@ -665,7 +843,7 @@ static void pflash_cfi01_realize(DeviceState *dev, Error **errp)
     pfl->cfi_table[0x28] = 0x02;
     pfl->cfi_table[0x29] = 0x00;
     /* Max number of bytes in multi-bytes write */
-    if (pfl->width == 1) {
+    if (pfl->bank_width == 1) {
         pfl->cfi_table[0x2A] = 0x08;
     } else {
         pfl->cfi_table[0x2A] = 0x0B;
@@ -706,7 +884,25 @@ static Property pflash_cfi01_properties[] = {
     DEFINE_PROP_DRIVE("drive", struct pflash_t, bs),
     DEFINE_PROP_UINT32("num-blocks", struct pflash_t, nb_blocs, 0),
     DEFINE_PROP_UINT64("sector-length", struct pflash_t, sector_len, 0),
-    DEFINE_PROP_UINT8("width", struct pflash_t, width, 0),
+    /* width here is the overall width of this QEMU device in bytes.
+     * The QEMU device may be emulating a number of flash devices
+     * wired up in parallel; the width of each individual flash
+     * device should be specified via device-width. If the individual
+     * devices have a maximum width which is greater than the width
+     * they are being used for, this maximum width should be set via
+     * max-device-width (which otherwise defaults to device-width).
+     * So for instance a 32-bit wide QEMU flash device made from four
+     * 16-bit flash devices used in 8-bit wide mode would be configured
+     * with width = 4, device-width = 1, max-device-width = 2.
+     *
+     * If device-width is not specified we default to backwards
+     * compatible behaviour which is a bad emulation of two
+     * 16 bit devices making up a 32 bit wide QEMU device. This
+     * is deprecated for new uses of this device.
+     */
+    DEFINE_PROP_UINT8("width", struct pflash_t, bank_width, 0),
+    DEFINE_PROP_UINT8("device-width", struct pflash_t, device_width, 0),
+    DEFINE_PROP_UINT8("max-device-width", struct pflash_t, max_device_width, 0),
     DEFINE_PROP_UINT8("big-endian", struct pflash_t, be, 0),
     DEFINE_PROP_UINT16("id0", struct pflash_t, ident0, 0),
     DEFINE_PROP_UINT16("id1", struct pflash_t, ident1, 0),
@@ -745,8 +941,8 @@ pflash_t *pflash_cfi01_register(hwaddr base,
                                 DeviceState *qdev, const char *name,
                                 hwaddr size,
                                 BlockDriverState *bs,
-                                uint32_t sector_len, int nb_blocs, int width,
-                                uint16_t id0, uint16_t id1,
+                                uint32_t sector_len, int nb_blocs,
+                                int bank_width, uint16_t id0, uint16_t id1,
                                 uint16_t id2, uint16_t id3, int be)
 {
     DeviceState *dev = qdev_create(NULL, TYPE_CFI_PFLASH01);
@@ -756,7 +952,7 @@ pflash_t *pflash_cfi01_register(hwaddr base,
     }
     qdev_prop_set_uint32(dev, "num-blocks", nb_blocs);
     qdev_prop_set_uint64(dev, "sector-length", sector_len);
-    qdev_prop_set_uint8(dev, "width", width);
+    qdev_prop_set_uint8(dev, "width", bank_width);
     qdev_prop_set_uint8(dev, "big-endian", !!be);
     qdev_prop_set_uint16(dev, "id0", id0);
     qdev_prop_set_uint16(dev, "id1", id1);

hw/block/virtio-blk.c

@@ -657,6 +657,7 @@ static void virtio_blk_migration_state_changed(Notifier *notifier, void *data)
     VirtIOBlock *s = container_of(notifier, VirtIOBlock,
                                   migration_state_notifier);
     MigrationState *mig = data;
+    Error *err = NULL;
 
     if (migration_in_setup(mig)) {
         if (!s->dataplane) {
@@ -671,31 +672,39 @@ static void virtio_blk_migration_state_changed(Notifier *notifier, void *data)
         }
         bdrv_drain_all(); /* complete in-flight non-dataplane requests */
         virtio_blk_data_plane_create(VIRTIO_DEVICE(s), &s->blk,
-                                     &s->dataplane);
+                                     &s->dataplane, &err);
+        if (err != NULL) {
+            error_report("%s", error_get_pretty(err));
+            error_free(err);
+        }
     }
 }
 #endif /* CONFIG_VIRTIO_BLK_DATA_PLANE */
 
-static int virtio_blk_device_init(VirtIODevice *vdev)
+static void virtio_blk_device_realize(DeviceState *dev, Error **errp)
 {
-    DeviceState *qdev = DEVICE(vdev);
-    VirtIOBlock *s = VIRTIO_BLK(vdev);
+    VirtIODevice *vdev = VIRTIO_DEVICE(dev);
+    VirtIOBlock *s = VIRTIO_BLK(dev);
     VirtIOBlkConf *blk = &(s->blk);
+#ifdef CONFIG_VIRTIO_BLK_DATA_PLANE
+    Error *err = NULL;
+#endif
     static int virtio_blk_id;
 
     if (!blk->conf.bs) {
-        error_report("drive property not set");
-        return -1;
+        error_setg(errp, "drive property not set");
+        return;
     }
     if (!bdrv_is_inserted(blk->conf.bs)) {
-        error_report("Device needs media, but drive is empty");
-        return -1;
+        error_setg(errp, "Device needs media, but drive is empty");
+        return;
     }
 
     blkconf_serial(&blk->conf, &blk->serial);
     s->original_wce = bdrv_enable_write_cache(blk->conf.bs);
     if (blkconf_geometry(&blk->conf, NULL, 65535, 255, 255) < 0) {
-        return -1;
+        error_setg(errp, "Error setting geometry");
+        return;
     }
 
     virtio_init(vdev, "virtio-blk", VIRTIO_ID_BLOCK,
@@ -708,30 +717,32 @@ static int virtio_blk_device_init(VirtIODevice *vdev)
 
     s->vq = virtio_add_queue(vdev, 128, virtio_blk_handle_output);
 #ifdef CONFIG_VIRTIO_BLK_DATA_PLANE
-    if (!virtio_blk_data_plane_create(vdev, blk, &s->dataplane)) {
+    virtio_blk_data_plane_create(vdev, blk, &s->dataplane, &err);
+    if (err != NULL) {
+        error_propagate(errp, err);
         virtio_cleanup(vdev);
-        return -1;
+        return;
     }
     s->migration_state_notifier.notify = virtio_blk_migration_state_changed;
     add_migration_state_change_notifier(&s->migration_state_notifier);
 #endif
 
     s->change = qemu_add_vm_change_state_handler(virtio_blk_dma_restart_cb, s);
-    register_savevm(qdev, "virtio-blk", virtio_blk_id++, 2,
+    register_savevm(dev, "virtio-blk", virtio_blk_id++, 2,
                     virtio_blk_save, virtio_blk_load, s);
     bdrv_set_dev_ops(s->bs, &virtio_block_ops, s);
     bdrv_set_buffer_alignment(s->bs, s->conf->logical_block_size);
 
     bdrv_iostatus_enable(s->bs);
 
-    add_boot_device_path(s->conf->bootindex, qdev, "/disk@0,0");
-    return 0;
+    add_boot_device_path(s->conf->bootindex, dev, "/disk@0,0");
 }
 
-static int virtio_blk_device_exit(DeviceState *dev)
+static void virtio_blk_device_unrealize(DeviceState *dev, Error **errp)
 {
     VirtIODevice *vdev = VIRTIO_DEVICE(dev);
     VirtIOBlock *s = VIRTIO_BLK(dev);
+
 #ifdef CONFIG_VIRTIO_BLK_DATA_PLANE
     remove_migration_state_change_notifier(&s->migration_state_notifier);
     virtio_blk_data_plane_destroy(s->dataplane);
@@ -741,7 +752,6 @@ static int virtio_blk_device_exit(DeviceState *dev)
     unregister_savevm(dev, "virtio-blk", s);
     blockdev_mark_auto_del(s->bs);
     virtio_cleanup(vdev);
-    return 0;
 }
 
 static Property virtio_blk_properties[] = {
@@ -753,10 +763,11 @@ static void virtio_blk_class_init(ObjectClass *klass, void *data)
 {
     DeviceClass *dc = DEVICE_CLASS(klass);
     VirtioDeviceClass *vdc = VIRTIO_DEVICE_CLASS(klass);
-    dc->exit = virtio_blk_device_exit;
+
     dc->props = virtio_blk_properties;
     set_bit(DEVICE_CATEGORY_STORAGE, dc->categories);
-    vdc->init = virtio_blk_device_init;
+    vdc->realize = virtio_blk_device_realize;
+    vdc->unrealize = virtio_blk_device_unrealize;
     vdc->get_config = virtio_blk_update_config;
     vdc->set_config = virtio_blk_set_config;
     vdc->get_features = virtio_blk_get_features;

hw/char/Makefile.objs

@@ -14,6 +14,7 @@ obj-$(CONFIG_COLDFIRE) += mcf_uart.o
 obj-$(CONFIG_OMAP) += omap_uart.o
 obj-$(CONFIG_SH4) += sh_serial.o
 obj-$(CONFIG_PSERIES) += spapr_vty.o
+obj-$(CONFIG_DIGIC) += digic-uart.o
 
 common-obj-$(CONFIG_ETRAXFS) += etraxfs_ser.o
 common-obj-$(CONFIG_ISA_DEBUG) += debugcon.o

hw/char/cadence_uart.c

@@ -34,6 +34,9 @@
 #define UART_SR_INTR_RFUL      0x00000004
 #define UART_SR_INTR_TEMPTY    0x00000008
 #define UART_SR_INTR_TFUL      0x00000010
+/* somewhat awkwardly, TTRIG is misaligned between SR and ISR */
+#define UART_SR_TTRIG          0x00002000
+#define UART_INTR_TTRIG        0x00000400
 /* bits fields in CSR that correlate to CISR. If any of these bits are set in
  * SR, then the same bit in CISR is set high too */
 #define UART_SR_TO_CISR_MASK   0x0000001F
@@ -43,6 +46,7 @@
 #define UART_INTR_PARE         0x00000080
 #define UART_INTR_TIMEOUT      0x00000100
 #define UART_INTR_DMSI         0x00000200
+#define UART_INTR_TOVR         0x00001000
 
 #define UART_SR_RACTIVE    0x00000400
 #define UART_SR_TACTIVE    0x00000800
@@ -110,23 +114,37 @@
 #define CADENCE_UART(obj) OBJECT_CHECK(UartState, (obj), TYPE_CADENCE_UART)
 
 typedef struct {
+    /*< private >*/
     SysBusDevice parent_obj;
+    /*< public >*/
 
     MemoryRegion iomem;
     uint32_t r[R_MAX];
-    uint8_t r_fifo[RX_FIFO_SIZE];
+    uint8_t rx_fifo[RX_FIFO_SIZE];
+    uint8_t tx_fifo[TX_FIFO_SIZE];
     uint32_t rx_wpos;
     uint32_t rx_count;
+    uint32_t tx_count;
     uint64_t char_tx_time;
     CharDriverState *chr;
     qemu_irq irq;
-    struct QEMUTimer *fifo_trigger_handle;
-    struct QEMUTimer *tx_time_handle;
+    QEMUTimer *fifo_trigger_handle;
 } UartState;
 
 static void uart_update_status(UartState *s)
 {
+    s->r[R_SR] = 0;
+
+    s->r[R_SR] |= s->rx_count == RX_FIFO_SIZE ? UART_SR_INTR_RFUL : 0;
+    s->r[R_SR] |= !s->rx_count ? UART_SR_INTR_REMPTY : 0;
+    s->r[R_SR] |= s->rx_count >= s->r[R_RTRIG] ? UART_SR_INTR_RTRIG : 0;
+
+    s->r[R_SR] |= s->tx_count == TX_FIFO_SIZE ? UART_SR_INTR_TFUL : 0;
+    s->r[R_SR] |= !s->tx_count ? UART_SR_INTR_TEMPTY : 0;
+    s->r[R_SR] |= s->tx_count >= s->r[R_TTRIG] ? UART_SR_TTRIG : 0;
+
     s->r[R_CISR] |= s->r[R_SR] & UART_SR_TO_CISR_MASK;
+    s->r[R_CISR] |= s->r[R_SR] & UART_SR_TTRIG ? UART_INTR_TTRIG : 0;
     qemu_set_irq(s->irq, !!(s->r[R_IMR] & s->r[R_CISR]));
 }
 
@@ -139,24 +157,6 @@ static void fifo_trigger_update(void *opaque)
     uart_update_status(s);
 }
 
-static void uart_tx_redo(UartState *s)
-{
-    uint64_t new_tx_time = qemu_clock_get_ns(QEMU_CLOCK_VIRTUAL);
-
-    timer_mod(s->tx_time_handle, new_tx_time + s->char_tx_time);
-
-    s->r[R_SR] |= UART_SR_INTR_TEMPTY;
-
-    uart_update_status(s);
-}
-
-static void uart_tx_write(void *opaque)
-{
-    UartState *s = (UartState *)opaque;
-
-    uart_tx_redo(s);
-}
-
 static void uart_rx_reset(UartState *s)
 {
     s->rx_wpos = 0;
@@ -164,15 +164,11 @@ static void uart_rx_reset(UartState *s)
     if (s->chr) {
         qemu_chr_accept_input(s->chr);
     }
-
-    s->r[R_SR] |= UART_SR_INTR_REMPTY;
-    s->r[R_SR] &= ~UART_SR_INTR_RFUL;
 }
 
 static void uart_tx_reset(UartState *s)
 {
-    s->r[R_SR] |= UART_SR_INTR_TEMPTY;
-    s->r[R_SR] &= ~UART_SR_INTR_TFUL;
+    s->tx_count = 0;
 }
 
 static void uart_send_breaks(UartState *s)
@@ -237,8 +233,16 @@ static void uart_parameters_setup(UartState *s)
 static int uart_can_receive(void *opaque)
 {
     UartState *s = (UartState *)opaque;
+    int ret = MAX(RX_FIFO_SIZE, TX_FIFO_SIZE);
+    uint32_t ch_mode = s->r[R_MR] & UART_MR_CHMODE;
 
-    return RX_FIFO_SIZE - s->rx_count;
+    if (ch_mode == NORMAL_MODE || ch_mode == ECHO_MODE) {
+        ret = MIN(ret, RX_FIFO_SIZE - s->rx_count);
+    }
+    if (ch_mode == REMOTE_LOOPBACK || ch_mode == ECHO_MODE) {
+        ret = MIN(ret, TX_FIFO_SIZE - s->tx_count);
+    }
+    return ret;
 }
 
 static void uart_ctrl_update(UartState *s)
@@ -253,10 +257,6 @@ static void uart_ctrl_update(UartState *s)
 
     s->r[R_CR] &= ~(UART_CR_TXRST | UART_CR_RXRST);
 
-    if ((s->r[R_CR] & UART_CR_TX_EN) && !(s->r[R_CR] & UART_CR_TX_DIS)) {
-            uart_tx_redo(s);
-    }
-
     if (s->r[R_CR] & UART_CR_STARTBRK && !(s->r[R_CR] & UART_CR_STOPBRK)) {
         uart_send_breaks(s);
     }
@@ -272,24 +272,13 @@ static void uart_write_rx_fifo(void *opaque, const uint8_t *buf, int size)
         return;
     }
 
-    s->r[R_SR] &= ~UART_SR_INTR_REMPTY;
-
     if (s->rx_count == RX_FIFO_SIZE) {
         s->r[R_CISR] |= UART_INTR_ROVR;
     } else {
         for (i = 0; i < size; i++) {
-            s->r_fifo[s->rx_wpos] = buf[i];
+            s->rx_fifo[s->rx_wpos] = buf[i];
             s->rx_wpos = (s->rx_wpos + 1) % RX_FIFO_SIZE;
             s->rx_count++;
-
-            if (s->rx_count == RX_FIFO_SIZE) {
-                s->r[R_SR] |= UART_SR_INTR_RFUL;
-                break;
-            }
-
-            if (s->rx_count >= s->r[R_RTRIG]) {
-                s->r[R_SR] |= UART_SR_INTR_RTRIG;
-            }
         }
         timer_mod(s->fifo_trigger_handle, new_rx_time +
                                                 (s->char_tx_time * 4));
@@ -297,13 +286,55 @@ static void uart_write_rx_fifo(void *opaque, const uint8_t *buf, int size)
     uart_update_status(s);
 }
 
+static gboolean cadence_uart_xmit(GIOChannel *chan, GIOCondition cond,
+                                  void *opaque)
+{
+    UartState *s = opaque;
+    int ret;
+
+    /* instant drain the fifo when there's no back-end */
+    if (!s->chr) {
+        s->tx_count = 0;
+    }
+
+    if (!s->tx_count) {
+        return FALSE;
+    }
+
+    ret = qemu_chr_fe_write(s->chr, s->tx_fifo, s->tx_count);
+    s->tx_count -= ret;
+    memmove(s->tx_fifo, s->tx_fifo + ret, s->tx_count);
+
+    if (s->tx_count) {
+        int r = qemu_chr_fe_add_watch(s->chr, G_IO_OUT, cadence_uart_xmit, s);
+        assert(r);
+    }
+
+    uart_update_status(s);
+    return FALSE;
+}
+
 static void uart_write_tx_fifo(UartState *s, const uint8_t *buf, int size)
 {
     if ((s->r[R_CR] & UART_CR_TX_DIS) || !(s->r[R_CR] & UART_CR_TX_EN)) {
         return;
     }
 
-    qemu_chr_fe_write_all(s->chr, buf, size);
+    if (size > TX_FIFO_SIZE - s->tx_count) {
+        size = TX_FIFO_SIZE - s->tx_count;
+        /*
+         * This can only be a guest error via a bad tx fifo register push,
+         * as can_receive() should stop remote loop and echo modes ever getting
+         * us to here.
+         */
+        qemu_log_mask(LOG_GUEST_ERROR, "cadence_uart: TxFIFO overflow");
+        s->r[R_CISR] |= UART_INTR_ROVR;
+    }
+
+    memcpy(s->tx_fifo + s->tx_count, buf, size);
+    s->tx_count += size;
+
+    cadence_uart_xmit(NULL, G_IO_OUT, s);
 }
 
 static void uart_receive(void *opaque, const uint8_t *buf, int size)
@@ -337,26 +368,17 @@ static void uart_read_rx_fifo(UartState *s, uint32_t *c)
         return;
     }
 
-    s->r[R_SR] &= ~UART_SR_INTR_RFUL;
-
     if (s->rx_count) {
         uint32_t rx_rpos =
                 (RX_FIFO_SIZE + s->rx_wpos - s->rx_count) % RX_FIFO_SIZE;
-        *c = s->r_fifo[rx_rpos];
+        *c = s->rx_fifo[rx_rpos];
         s->rx_count--;
 
-        if (!s->rx_count) {
-            s->r[R_SR] |= UART_SR_INTR_REMPTY;
-        }
         qemu_chr_accept_input(s->chr);
     } else {
         *c = 0;
-        s->r[R_SR] |= UART_SR_INTR_REMPTY;
     }
 
-    if (s->rx_count < s->r[R_RTRIG]) {
-        s->r[R_SR] &= ~UART_SR_INTR_RTRIG;
-    }
     uart_update_status(s);
 }
 
@@ -401,6 +423,7 @@ static void uart_write(void *opaque, hwaddr offset,
         uart_parameters_setup(s);
         break;
     }
+    uart_update_status(s);
 }
 
 static uint64_t uart_read(void *opaque, hwaddr offset,
@@ -428,8 +451,10 @@ static const MemoryRegionOps uart_ops = {
     .endianness = DEVICE_NATIVE_ENDIAN,
 };
 
-static void cadence_uart_reset(UartState *s)
+static void cadence_uart_reset(DeviceState *dev)
 {
+    UartState *s = CADENCE_UART(dev);
+
     s->r[R_CR] = 0x00000128;
     s->r[R_IMR] = 0;
     s->r[R_CISR] = 0;
@@ -440,8 +465,7 @@ static void cadence_uart_reset(UartState *s)
     uart_rx_reset(s);
     uart_tx_reset(s);
 
-    s->rx_count = 0;
-    s->rx_wpos = 0;
+    uart_update_status(s);
 }
 
 static int cadence_uart_init(SysBusDevice *dev)
@@ -455,15 +479,10 @@ static int cadence_uart_init(SysBusDevice *dev)
     s->fifo_trigger_handle = timer_new_ns(QEMU_CLOCK_VIRTUAL,
             (QEMUTimerCB *)fifo_trigger_update, s);
 
-    s->tx_time_handle = timer_new_ns(QEMU_CLOCK_VIRTUAL,
-            (QEMUTimerCB *)uart_tx_write, s);
-
     s->char_tx_time = (get_ticks_per_sec() / 9600) * 10;
 
     s->chr = qemu_char_get_next_serial();
 
-    cadence_uart_reset(s);
-
     if (s->chr) {
         qemu_chr_add_handlers(s->chr, uart_can_receive, uart_receive,
                               uart_event, s);
@@ -483,17 +502,18 @@ static int cadence_uart_post_load(void *opaque, int version_id)
 
 static const VMStateDescription vmstate_cadence_uart = {
     .name = "cadence_uart",
-    .version_id = 1,
-    .minimum_version_id = 1,
-    .minimum_version_id_old = 1,
+    .version_id = 2,
+    .minimum_version_id = 2,
+    .minimum_version_id_old = 2,
     .post_load = cadence_uart_post_load,
     .fields = (VMStateField[]) {
         VMSTATE_UINT32_ARRAY(r, UartState, R_MAX),
-        VMSTATE_UINT8_ARRAY(r_fifo, UartState, RX_FIFO_SIZE),
+        VMSTATE_UINT8_ARRAY(rx_fifo, UartState, RX_FIFO_SIZE),
+        VMSTATE_UINT8_ARRAY(tx_fifo, UartState, RX_FIFO_SIZE),
         VMSTATE_UINT32(rx_count, UartState),
+        VMSTATE_UINT32(tx_count, UartState),
         VMSTATE_UINT32(rx_wpos, UartState),
         VMSTATE_TIMER(fifo_trigger_handle, UartState),
-        VMSTATE_TIMER(tx_time_handle, UartState),
         VMSTATE_END_OF_LIST()
     }
 };
@@ -505,6 +525,7 @@ static void cadence_uart_class_init(ObjectClass *klass, void *data)
 
     sdc->init = cadence_uart_init;
     dc->vmsd = &vmstate_cadence_uart;
+    dc->reset = cadence_uart_reset;
 }
 
 static const TypeInfo cadence_uart_info = {

hw/char/digic-uart.c

@@ -0,0 +1,195 @@
+/*
+ * QEMU model of the Canon DIGIC UART block.
+ *
+ * Copyright (C) 2013 Antony Pavlov <antonynpavlov@gmail.com>
+ *
+ * This model is based on reverse engineering efforts
+ * made by CHDK (http://chdk.wikia.com) and
+ * Magic Lantern (http://www.magiclantern.fm) projects
+ * contributors.
+ *
+ * See "Serial terminal" docs here:
+ *   http://magiclantern.wikia.com/wiki/Register_Map#Misc_Registers
+ *
+ * The QEMU model of the Milkymist UART block by Michael Walle
+ * is used as a template.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ */
+
+#include "hw/hw.h"
+#include "hw/sysbus.h"
+#include "sysemu/char.h"
+
+#include "hw/char/digic-uart.h"
+
+enum {
+    ST_RX_RDY = (1 << 0),
+    ST_TX_RDY = (1 << 1),
+};
+
+static uint64_t digic_uart_read(void *opaque, hwaddr addr,
+                                unsigned size)
+{
+    DigicUartState *s = opaque;
+    uint64_t ret = 0;
+
+    addr >>= 2;
+
+    switch (addr) {
+    case R_RX:
+        s->reg_st &= ~(ST_RX_RDY);
+        ret = s->reg_rx;
+        break;
+
+    case R_ST:
+        ret = s->reg_st;
+        break;
+
+    default:
+        qemu_log_mask(LOG_UNIMP,
+                      "digic-uart: read access to unknown register 0x"
+                      TARGET_FMT_plx, addr << 2);
+    }
+
+    return ret;
+}
+
+static void digic_uart_write(void *opaque, hwaddr addr, uint64_t value,
+                             unsigned size)
+{
+    DigicUartState *s = opaque;
+    unsigned char ch = value;
+
+    addr >>= 2;
+
+    switch (addr) {
+    case R_TX:
+        if (s->chr) {
+            qemu_chr_fe_write_all(s->chr, &ch, 1);
+        }
+        break;
+
+    case R_ST:
+        /*
+         * Ignore write to R_ST.
+         *
+         * The point is that this register is actively used
+         * during receiving and transmitting symbols,
+         * but we don't know the function of most of bits.
+         *
+         * Ignoring writes to R_ST is only a simplification
+         * of the model. It has no perceptible side effects
+         * for existing guests.
+         */
+        break;
+
+    default:
+        qemu_log_mask(LOG_UNIMP,
+                      "digic-uart: write access to unknown register 0x"
+                      TARGET_FMT_plx, addr << 2);
+    }
+}
+
+static const MemoryRegionOps uart_mmio_ops = {
+    .read = digic_uart_read,
+    .write = digic_uart_write,
+    .valid = {
+        .min_access_size = 4,
+        .max_access_size = 4,
+    },
+    .endianness = DEVICE_NATIVE_ENDIAN,
+};
+
+static int uart_can_rx(void *opaque)
+{
+    DigicUartState *s = opaque;
+
+    return !(s->reg_st & ST_RX_RDY);
+}
+
+static void uart_rx(void *opaque, const uint8_t *buf, int size)
+{
+    DigicUartState *s = opaque;
+
+    assert(uart_can_rx(opaque));
+
+    s->reg_st |= ST_RX_RDY;
+    s->reg_rx = *buf;
+}
+
+static void uart_event(void *opaque, int event)
+{
+}
+
+static void digic_uart_reset(DeviceState *d)
+{
+    DigicUartState *s = DIGIC_UART(d);
+
+    s->reg_rx = 0;
+    s->reg_st = ST_TX_RDY;
+}
+
+static void digic_uart_realize(DeviceState *dev, Error **errp)
+{
+    DigicUartState *s = DIGIC_UART(dev);
+
+    s->chr = qemu_char_get_next_serial();
+    if (s->chr) {
+        qemu_chr_add_handlers(s->chr, uart_can_rx, uart_rx, uart_event, s);
+    }
+}
+
+static void digic_uart_init(Object *obj)
+{
+    DigicUartState *s = DIGIC_UART(obj);
+
+    memory_region_init_io(&s->regs_region, OBJECT(s), &uart_mmio_ops, s,
+                          TYPE_DIGIC_UART, 0x18);
+    sysbus_init_mmio(SYS_BUS_DEVICE(obj), &s->regs_region);
+}
+
+static const VMStateDescription vmstate_digic_uart = {
+    .name = "digic-uart",
+    .version_id = 1,
+    .minimum_version_id = 1,
+    .minimum_version_id_old = 1,
+    .fields = (VMStateField[]) {
+        VMSTATE_UINT32(reg_rx, DigicUartState),
+        VMSTATE_UINT32(reg_st, DigicUartState),
+        VMSTATE_END_OF_LIST()
+    }
+};
+
+static void digic_uart_class_init(ObjectClass *klass, void *data)
+{
+    DeviceClass *dc = DEVICE_CLASS(klass);
+
+    dc->realize = digic_uart_realize;
+    dc->reset = digic_uart_reset;
+    dc->vmsd = &vmstate_digic_uart;
+}
+
+static const TypeInfo digic_uart_info = {
+    .name = TYPE_DIGIC_UART,
+    .parent = TYPE_SYS_BUS_DEVICE,
+    .instance_size = sizeof(DigicUartState),
+    .instance_init = digic_uart_init,
+    .class_init = digic_uart_class_init,
+};
+
+static void digic_uart_register_types(void)
+{
+    type_register_static(&digic_uart_info);
+}
+
+type_init(digic_uart_register_types)

hw/char/exynos4210_uart.c

@@ -192,10 +192,9 @@ typedef struct Exynos4210UartState {
 static const char *exynos4210_uart_regname(hwaddr  offset)
 {
 
-    int regs_number = sizeof(exynos4210_uart_regs) / sizeof(Exynos4210UartReg);
     int i;
 
-    for (i = 0; i < regs_number; i++) {
+    for (i = 0; i < ARRAY_SIZE(exynos4210_uart_regs); i++) {
         if (offset == exynos4210_uart_regs[i].offset) {
             return exynos4210_uart_regs[i].name;
         }
@@ -544,10 +543,9 @@ static void exynos4210_uart_event(void *opaque, int event)
 static void exynos4210_uart_reset(DeviceState *dev)
 {
     Exynos4210UartState *s = EXYNOS4210_UART(dev);
-    int regs_number = sizeof(exynos4210_uart_regs)/sizeof(Exynos4210UartReg);
     int i;
 
-    for (i = 0; i < regs_number; i++) {
+    for (i = 0; i < ARRAY_SIZE(exynos4210_uart_regs); i++) {
         s->reg[I_(exynos4210_uart_regs[i].offset)] =
                 exynos4210_uart_regs[i].reset_value;
     }

hw/char/virtio-serial-bus.c

@@ -889,22 +889,24 @@ static int virtser_port_qdev_exit(DeviceState *qdev)
     return 0;
 }
 
-static int virtio_serial_device_init(VirtIODevice *vdev)
+static void virtio_serial_device_realize(DeviceState *dev, Error **errp)
 {
-    DeviceState *qdev = DEVICE(vdev);
-    VirtIOSerial *vser = VIRTIO_SERIAL(vdev);
+    VirtIODevice *vdev = VIRTIO_DEVICE(dev);
+    VirtIOSerial *vser = VIRTIO_SERIAL(dev);
+    BusState *bus;
     uint32_t i, max_supported_ports;
 
     if (!vser->serial.max_virtserial_ports) {
-        return -1;
+        error_setg(errp, "Maximum number of serial ports not specified");
+        return;
     }
 
     /* Each port takes 2 queues, and one pair is for the control queue */
     max_supported_ports = VIRTIO_PCI_QUEUE_MAX / 2 - 1;
 
     if (vser->serial.max_virtserial_ports > max_supported_ports) {
-        error_report("maximum ports supported: %u", max_supported_ports);
-        return -1;
+        error_setg(errp, "maximum ports supported: %u", max_supported_ports);
+        return;
     }
 
     virtio_init(vdev, "virtio-serial", VIRTIO_ID_CONSOLE,
@@ -912,8 +914,9 @@ static int virtio_serial_device_init(VirtIODevice *vdev)
 
     /* Spawn a new virtio-serial bus on which the ports will ride as devices */
     qbus_create_inplace(&vser->bus, sizeof(vser->bus), TYPE_VIRTIO_SERIAL_BUS,
-                        qdev, vdev->bus_name);
-    vser->bus.qbus.allow_hotplug = 1;
+                        dev, vdev->bus_name);
+    bus = BUS(&vser->bus);
+    bus->allow_hotplug = 1;
     vser->bus.vser = vser;
     QTAILQ_INIT(&vser->ports);
 
@@ -961,10 +964,8 @@ static int virtio_serial_device_init(VirtIODevice *vdev)
      * Register for the savevm section with the virtio-console name
      * to preserve backward compat
      */
-    register_savevm(qdev, "virtio-console", -1, 3, virtio_serial_save,
+    register_savevm(dev, "virtio-console", -1, 3, virtio_serial_save,
                     virtio_serial_load, vser);
-
-    return 0;
 }
 
 static void virtio_serial_port_class_init(ObjectClass *klass, void *data)
@@ -987,10 +988,10 @@ static const TypeInfo virtio_serial_port_type_info = {
     .class_init = virtio_serial_port_class_init,
 };
 
-static int virtio_serial_device_exit(DeviceState *dev)
+static void virtio_serial_device_unrealize(DeviceState *dev, Error **errp)
 {
-    VirtIOSerial *vser = VIRTIO_SERIAL(dev);
     VirtIODevice *vdev = VIRTIO_DEVICE(dev);
+    VirtIOSerial *vser = VIRTIO_SERIAL(dev);
 
     unregister_savevm(dev, "virtio-console", vser);
 
@@ -1004,7 +1005,6 @@ static int virtio_serial_device_exit(DeviceState *dev)
         g_free(vser->post_load);
     }
     virtio_cleanup(vdev);
-    return 0;
 }
 
 static Property virtio_serial_properties[] = {
@@ -1016,10 +1016,11 @@ static void virtio_serial_class_init(ObjectClass *klass, void *data)
 {
     DeviceClass *dc = DEVICE_CLASS(klass);
     VirtioDeviceClass *vdc = VIRTIO_DEVICE_CLASS(klass);
-    dc->exit = virtio_serial_device_exit;
+
     dc->props = virtio_serial_properties;
     set_bit(DEVICE_CATEGORY_INPUT, dc->categories);
-    vdc->init = virtio_serial_device_init;
+    vdc->realize = virtio_serial_device_realize;
+    vdc->unrealize = virtio_serial_device_unrealize;
     vdc->get_features = get_features;
     vdc->get_config = get_config;
     vdc->set_config = set_config;

hw/core/loader.c

@@ -785,6 +785,13 @@ static void rom_reset(void *unused)
             g_free(rom->data);
             rom->data = NULL;
         }
+        /*
+         * The rom loader is really on the same level as firmware in the guest
+         * shadowing a ROM into RAM. Such a shadowing mechanism needs to ensure
+         * that the instruction cache for that new region is clear, so that the
+         * CPU definitely fetches its instructions from the just written data.
+         */
+        cpu_flush_icache_range(rom->addr, rom->datasize);
     }
 }
 

hw/core/qdev-properties-system.c

@@ -352,21 +352,17 @@ void qdev_prop_set_drive_nofail(DeviceState *dev, const char *name,
 void qdev_prop_set_chr(DeviceState *dev, const char *name,
                        CharDriverState *value)
 {
-    Error *errp = NULL;
     assert(!value || value->label);
     object_property_set_str(OBJECT(dev),
-                            value ? value->label : "", name, &errp);
-    assert_no_error(errp);
+                            value ? value->label : "", name, &error_abort);
 }
 
 void qdev_prop_set_netdev(DeviceState *dev, const char *name,
                           NetClientState *value)
 {
-    Error *errp = NULL;
     assert(!value || value->name);
     object_property_set_str(OBJECT(dev),
-                            value ? value->name : "", name, &errp);
-    assert_no_error(errp);
+                            value ? value->name : "", name, &error_abort);
 }
 
 void qdev_set_nic_properties(DeviceState *dev, NICInfo *nd)

hw/core/qdev-properties.c

@@ -1003,73 +1003,55 @@ void qdev_prop_parse(DeviceState *dev, const char *name, const char *value,
 
 void qdev_prop_set_bit(DeviceState *dev, const char *name, bool value)
 {
-    Error *errp = NULL;
-    object_property_set_bool(OBJECT(dev), value, name, &errp);
-    assert_no_error(errp);
+    object_property_set_bool(OBJECT(dev), value, name, &error_abort);
 }
 
 void qdev_prop_set_uint8(DeviceState *dev, const char *name, uint8_t value)
 {
-    Error *errp = NULL;
-    object_property_set_int(OBJECT(dev), value, name, &errp);
-    assert_no_error(errp);
+    object_property_set_int(OBJECT(dev), value, name, &error_abort);
 }
 
 void qdev_prop_set_uint16(DeviceState *dev, const char *name, uint16_t value)
 {
-    Error *errp = NULL;
-    object_property_set_int(OBJECT(dev), value, name, &errp);
-    assert_no_error(errp);
+    object_property_set_int(OBJECT(dev), value, name, &error_abort);
 }
 
 void qdev_prop_set_uint32(DeviceState *dev, const char *name, uint32_t value)
 {
-    Error *errp = NULL;
-    object_property_set_int(OBJECT(dev), value, name, &errp);
-    assert_no_error(errp);
+    object_property_set_int(OBJECT(dev), value, name, &error_abort);
 }
 
 void qdev_prop_set_int32(DeviceState *dev, const char *name, int32_t value)
 {
-    Error *errp = NULL;
-    object_property_set_int(OBJECT(dev), value, name, &errp);
-    assert_no_error(errp);
+    object_property_set_int(OBJECT(dev), value, name, &error_abort);
 }
 
 void qdev_prop_set_uint64(DeviceState *dev, const char *name, uint64_t value)
 {
-    Error *errp = NULL;
-    object_property_set_int(OBJECT(dev), value, name, &errp);
-    assert_no_error(errp);
+    object_property_set_int(OBJECT(dev), value, name, &error_abort);
 }
 
 void qdev_prop_set_string(DeviceState *dev, const char *name, const char *value)
 {
-    Error *errp = NULL;
-    object_property_set_str(OBJECT(dev), value, name, &errp);
-    assert_no_error(errp);
+    object_property_set_str(OBJECT(dev), value, name, &error_abort);
 }
 
 void qdev_prop_set_macaddr(DeviceState *dev, const char *name, uint8_t *value)
 {
-    Error *errp = NULL;
     char str[2 * 6 + 5 + 1];
     snprintf(str, sizeof(str), "%02x:%02x:%02x:%02x:%02x:%02x",
              value[0], value[1], value[2], value[3], value[4], value[5]);
 
-    object_property_set_str(OBJECT(dev), str, name, &errp);
-    assert_no_error(errp);
+    object_property_set_str(OBJECT(dev), str, name, &error_abort);
 }
 
 void qdev_prop_set_enum(DeviceState *dev, const char *name, int value)
 {
     Property *prop;
-    Error *errp = NULL;
 
     prop = qdev_prop_find(dev, name);
     object_property_set_str(OBJECT(dev), prop->info->enum_table[value],
-                            name, &errp);
-    assert_no_error(errp);
+                            name, &error_abort);
 }
 
 void qdev_prop_set_ptr(DeviceState *dev, const char *name, void *value)
@@ -1161,12 +1143,10 @@ static void set_size(Object *obj, Visitor *v, void *opaque,
 static int parse_size(DeviceState *dev, Property *prop, const char *str)
 {
     uint64_t *ptr = qdev_get_prop_ptr(dev, prop);
-    Error *errp = NULL;
 
     if (str != NULL) {
-        parse_option_size(prop->name, str, ptr, &errp);
+        parse_option_size(prop->name, str, ptr, &error_abort);
     }
-    assert_no_error(errp);
     return 0;
 }