Merge remote-tracking branch 'kwolf/for-anthony' into staging

* kwolf/for-anthony: (24 commits)
  block: Factor bdrv_read_unthrottled() out of guess_disk_lchs()
  qtest: Tidy up temporary files properly
  fdc: Drop broken code for user-defined floppy geometry
  fdc_test: introduce test_sense_interrupt
  fdc_test: update media_change test
  fdc: fix interrupt handling
  fdc: rewrite seek and DSKCHG bit handling
  block: introduce bdrv_swap, implement bdrv_append on top of it
  block: copy over job and dirty bitmap fields in bdrv_append
  raw: hook into blkdebug
  blkdebug: optionally tie errors to a specific sector
  blkdebug: store list of active rules
  blkdebug: pass getlength to underlying file
  blkdebug: tiny cleanup
  blkdebug: remove sync i/o events
  sheepdog: traverse pending_list from the first for each time
  sheepdog: split outstanding list into inflight and pending
  sheepdog: make sure we don't free aiocb before sending all requests
  sheepdog: use coroutine based socket functions in coroutine context
  sheepdog: restart I/O when socket becomes ready in do_co_req()
  ...

block.c

@@ -971,101 +971,130 @@ static void bdrv_rebind(BlockDriverState *bs)
     }
 }
 
+static void bdrv_move_feature_fields(BlockDriverState *bs_dest,
+                                     BlockDriverState *bs_src)
+{
+    /* move some fields that need to stay attached to the device */
+    bs_dest->open_flags         = bs_src->open_flags;
+
+    /* dev info */
+    bs_dest->dev_ops            = bs_src->dev_ops;
+    bs_dest->dev_opaque         = bs_src->dev_opaque;
+    bs_dest->dev                = bs_src->dev;
+    bs_dest->buffer_alignment   = bs_src->buffer_alignment;
+    bs_dest->copy_on_read       = bs_src->copy_on_read;
+
+    bs_dest->enable_write_cache = bs_src->enable_write_cache;
+
+    /* i/o timing parameters */
+    bs_dest->slice_time         = bs_src->slice_time;
+    bs_dest->slice_start        = bs_src->slice_start;
+    bs_dest->slice_end          = bs_src->slice_end;
+    bs_dest->io_limits          = bs_src->io_limits;
+    bs_dest->io_base            = bs_src->io_base;
+    bs_dest->throttled_reqs     = bs_src->throttled_reqs;
+    bs_dest->block_timer        = bs_src->block_timer;
+    bs_dest->io_limits_enabled  = bs_src->io_limits_enabled;
+
+    /* geometry */
+    bs_dest->cyls               = bs_src->cyls;
+    bs_dest->heads              = bs_src->heads;
+    bs_dest->secs               = bs_src->secs;
+    bs_dest->translation        = bs_src->translation;
+
+    /* r/w error */
+    bs_dest->on_read_error      = bs_src->on_read_error;
+    bs_dest->on_write_error     = bs_src->on_write_error;
+
+    /* i/o status */
+    bs_dest->iostatus_enabled   = bs_src->iostatus_enabled;
+    bs_dest->iostatus           = bs_src->iostatus;
+
+    /* dirty bitmap */
+    bs_dest->dirty_count        = bs_src->dirty_count;
+    bs_dest->dirty_bitmap       = bs_src->dirty_bitmap;
+
+    /* job */
+    bs_dest->in_use             = bs_src->in_use;
+    bs_dest->job                = bs_src->job;
+
+    /* keep the same entry in bdrv_states */
+    pstrcpy(bs_dest->device_name, sizeof(bs_dest->device_name),
+            bs_src->device_name);
+    bs_dest->list = bs_src->list;
+}
+
 /*
- * Add new bs contents at the top of an image chain while the chain is
- * live, while keeping required fields on the top layer.
+ * Swap bs contents for two image chains while they are live,
+ * while keeping required fields on the BlockDriverState that is
+ * actually attached to a device.
  *
  * This will modify the BlockDriverState fields, and swap contents
- * between bs_new and bs_top. Both bs_new and bs_top are modified.
+ * between bs_new and bs_old. Both bs_new and bs_old are modified.
  *
  * bs_new is required to be anonymous.
  *
  * This function does not create any image files.
  */
-void bdrv_append(BlockDriverState *bs_new, BlockDriverState *bs_top)
+void bdrv_swap(BlockDriverState *bs_new, BlockDriverState *bs_old)
 {
     BlockDriverState tmp;
 
-    /* bs_new must be anonymous */
+    /* 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(bs_new->job == NULL);
+    assert(bs_new->dev == NULL);
+    assert(bs_new->in_use == 0);
+    assert(bs_new->io_limits_enabled == false);
+    assert(bs_new->block_timer == NULL);
 
     tmp = *bs_new;
+    *bs_new = *bs_old;
+    *bs_old = tmp;
 
-    /* there are some fields that need to stay on the top layer: */
-    tmp.open_flags        = bs_top->open_flags;
-
-    /* dev info */
-    tmp.dev_ops           = bs_top->dev_ops;
-    tmp.dev_opaque        = bs_top->dev_opaque;
-    tmp.dev               = bs_top->dev;
-    tmp.buffer_alignment  = bs_top->buffer_alignment;
-    tmp.copy_on_read      = bs_top->copy_on_read;
-
-    tmp.enable_write_cache = bs_top->enable_write_cache;
-
-    /* i/o timing parameters */
-    tmp.slice_time        = bs_top->slice_time;
-    tmp.slice_start       = bs_top->slice_start;
-    tmp.slice_end         = bs_top->slice_end;
-    tmp.io_limits         = bs_top->io_limits;
-    tmp.io_base           = bs_top->io_base;
-    tmp.throttled_reqs    = bs_top->throttled_reqs;
-    tmp.block_timer       = bs_top->block_timer;
-    tmp.io_limits_enabled = bs_top->io_limits_enabled;
+    /* there are some fields that should not be swapped, move them back */
+    bdrv_move_feature_fields(&tmp, bs_old);
+    bdrv_move_feature_fields(bs_old, bs_new);
+    bdrv_move_feature_fields(bs_new, &tmp);
 
-    /* geometry */
-    tmp.cyls              = bs_top->cyls;
-    tmp.heads             = bs_top->heads;
-    tmp.secs              = bs_top->secs;
-    tmp.translation       = bs_top->translation;
+    /* bs_new shouldn't be in bdrv_states even after the swap!  */
+    assert(bs_new->device_name[0] == '\0');
 
-    /* r/w error */
-    tmp.on_read_error     = bs_top->on_read_error;
-    tmp.on_write_error    = bs_top->on_write_error;
+    /* Check a few fields that should remain attached to the device */
+    assert(bs_new->dev == NULL);
+    assert(bs_new->job == NULL);
+    assert(bs_new->in_use == 0);
+    assert(bs_new->io_limits_enabled == false);
+    assert(bs_new->block_timer == NULL);
 
-    /* i/o status */
-    tmp.iostatus_enabled  = bs_top->iostatus_enabled;
-    tmp.iostatus          = bs_top->iostatus;
+    bdrv_rebind(bs_new);
+    bdrv_rebind(bs_old);
+}
 
-    /* keep the same entry in bdrv_states */
-    pstrcpy(tmp.device_name, sizeof(tmp.device_name), bs_top->device_name);
-    tmp.list = bs_top->list;
+/*
+ * Add new bs contents at the top of an image chain while the chain is
+ * live, while keeping required fields on the top layer.
+ *
+ * This will modify the BlockDriverState fields, and swap contents
+ * between bs_new and bs_top. Both bs_new and bs_top are modified.
+ *
+ * bs_new is required to be anonymous.
+ *
+ * This function does not create any image files.
+ */
+void bdrv_append(BlockDriverState *bs_new, BlockDriverState *bs_top)
+{
+    bdrv_swap(bs_new, bs_top);
 
     /* The contents of 'tmp' will become bs_top, as we are
      * swapping bs_new and bs_top contents. */
-    tmp.backing_hd = bs_new;
-    pstrcpy(tmp.backing_file, sizeof(tmp.backing_file), bs_top->filename);
-    pstrcpy(tmp.backing_format, sizeof(tmp.backing_format),
-            bs_top->drv ? bs_top->drv->format_name : "");
-
-    /* swap contents of the fixed new bs and the current top */
-    *bs_new = *bs_top;
-    *bs_top = tmp;
-
-    /* device_name[] was carried over from the old bs_top.  bs_new
-     * shouldn't be in bdrv_states, so we need to make device_name[]
-     * reflect the anonymity of bs_new
-     */
-    bs_new->device_name[0] = '\0';
-
-    /* clear the copied fields in the new backing file */
-    bdrv_detach_dev(bs_new, bs_new->dev);
-
-    qemu_co_queue_init(&bs_new->throttled_reqs);
-    memset(&bs_new->io_base,   0, sizeof(bs_new->io_base));
-    memset(&bs_new->io_limits, 0, sizeof(bs_new->io_limits));
-    bdrv_iostatus_disable(bs_new);
-
-    /* we don't use bdrv_io_limits_disable() for this, because we don't want
-     * to affect or delete the block_timer, as it has been moved to bs_top */
-    bs_new->io_limits_enabled = false;
-    bs_new->block_timer       = NULL;
-    bs_new->slice_time        = 0;
-    bs_new->slice_start       = 0;
-    bs_new->slice_end         = 0;
-
-    bdrv_rebind(bs_new);
-    bdrv_rebind(bs_top);
+    bs_top->backing_hd = bs_new;
+    bs_top->open_flags &= ~BDRV_O_NO_BACKING;
+    pstrcpy(bs_top->backing_file, sizeof(bs_top->backing_file),
+            bs_new->filename);
+    pstrcpy(bs_top->backing_format, sizeof(bs_top->backing_format),
+            bs_new->drv ? bs_new->drv->format_name : "");
 }
 
 void bdrv_delete(BlockDriverState *bs)
@@ -1610,6 +1639,20 @@ int bdrv_read(BlockDriverState *bs, int64_t sector_num,
     return bdrv_rw_co(bs, sector_num, buf, nb_sectors, false);
 }
 
+/* Just like bdrv_read(), but with I/O throttling temporarily disabled */
+int bdrv_read_unthrottled(BlockDriverState *bs, int64_t sector_num,
+                          uint8_t *buf, int nb_sectors)
+{
+    bool enabled;
+    int ret;
+
+    enabled = bs->io_limits_enabled;
+    bs->io_limits_enabled = false;
+    ret = bdrv_read(bs, 0, buf, 1);
+    bs->io_limits_enabled = enabled;
+    return ret;
+}
+
 #define BITS_PER_LONG  (sizeof(unsigned long) * 8)
 
 static void set_dirty_bitmap(BlockDriverState *bs, int64_t sector_num,
@@ -2107,11 +2150,10 @@ static int guess_disk_lchs(BlockDriverState *bs,
                            int *pcylinders, int *pheads, int *psectors)
 {
     uint8_t buf[BDRV_SECTOR_SIZE];
-    int ret, i, heads, sectors, cylinders;
+    int i, heads, sectors, cylinders;
     struct partition *p;
     uint32_t nr_sects;
     uint64_t nb_sectors;
-    bool enabled;
 
     bdrv_get_geometry(bs, &nb_sectors);
 
@@ -2120,12 +2162,9 @@ static int guess_disk_lchs(BlockDriverState *bs,
      * but also in async I/O mode. So the I/O throttling function has to
      * be disabled temporarily here, not permanently.
      */
-    enabled = bs->io_limits_enabled;
-    bs->io_limits_enabled = false;
-    ret = bdrv_read(bs, 0, buf, 1);
-    bs->io_limits_enabled = enabled;
-    if (ret < 0)
+    if (bdrv_read_unthrottled(bs, 0, buf, 1) < 0) {
         return -1;
+    }
     /* test msdos magic */
     if (buf[510] != 0x55 || buf[511] != 0xaa)
         return -1;
@@ -2308,46 +2347,40 @@ void bdrv_get_floppy_geometry_hint(BlockDriverState *bs, int *nb_heads,
     uint64_t nb_sectors, size;
     int i, first_match, match;
 
-    bdrv_get_geometry_hint(bs, nb_heads, max_track, last_sect);
-    if (*nb_heads != 0 && *max_track != 0 && *last_sect != 0) {
-        /* User defined disk */
-        *rate = FDRIVE_RATE_500K;
-    } else {
-        bdrv_get_geometry(bs, &nb_sectors);
-        match = -1;
-        first_match = -1;
-        for (i = 0; ; i++) {
-            parse = &fd_formats[i];
-            if (parse->drive == FDRIVE_DRV_NONE) {
+    bdrv_get_geometry(bs, &nb_sectors);
+    match = -1;
+    first_match = -1;
+    for (i = 0; ; i++) {
+        parse = &fd_formats[i];
+        if (parse->drive == FDRIVE_DRV_NONE) {
+            break;
+        }
+        if (drive_in == parse->drive ||
+            drive_in == FDRIVE_DRV_NONE) {
+            size = (parse->max_head + 1) * parse->max_track *
+                parse->last_sect;
+            if (nb_sectors == size) {
+                match = i;
                 break;
             }
-            if (drive_in == parse->drive ||
-                drive_in == FDRIVE_DRV_NONE) {
-                size = (parse->max_head + 1) * parse->max_track *
-                    parse->last_sect;
-                if (nb_sectors == size) {
-                    match = i;
-                    break;
-                }
-                if (first_match == -1) {
-                    first_match = i;
-                }
-            }
-        }
-        if (match == -1) {
             if (first_match == -1) {
-                match = 1;
-            } else {
-                match = first_match;
+                first_match = i;
             }
-            parse = &fd_formats[match];
         }
-        *nb_heads = parse->max_head + 1;
-        *max_track = parse->max_track;
-        *last_sect = parse->last_sect;
-        *drive = parse->drive;
-        *rate = parse->rate;
     }
+    if (match == -1) {
+        if (first_match == -1) {
+            match = 1;
+        } else {
+            match = first_match;
+        }
+        parse = &fd_formats[match];
+    }
+    *nb_heads = parse->max_head + 1;
+    *max_track = parse->max_track;
+    *last_sect = parse->last_sect;
+    *drive = parse->drive;
+    *rate = parse->rate;
 }
 
 int bdrv_get_translation_hint(BlockDriverState *bs)

block.h

@@ -122,6 +122,7 @@ int bdrv_create(BlockDriver *drv, const char* filename,
 int bdrv_create_file(const char* filename, QEMUOptionParameter *options);
 BlockDriverState *bdrv_new(const char *device_name);
 void bdrv_make_anon(BlockDriverState *bs);
+void bdrv_swap(BlockDriverState *bs_new, BlockDriverState *bs_old);
 void bdrv_append(BlockDriverState *bs_new, BlockDriverState *bs_top);
 void bdrv_delete(BlockDriverState *bs);
 int bdrv_parse_cache_flags(const char *mode, int *flags);
@@ -141,6 +142,8 @@ bool bdrv_dev_is_tray_open(BlockDriverState *bs);
 bool bdrv_dev_is_medium_locked(BlockDriverState *bs);
 int bdrv_read(BlockDriverState *bs, int64_t sector_num,
               uint8_t *buf, int nb_sectors);
+int bdrv_read_unthrottled(BlockDriverState *bs, int64_t sector_num,
+                          uint8_t *buf, int nb_sectors);
 int bdrv_write(BlockDriverState *bs, int64_t sector_num,
                const uint8_t *buf, int nb_sectors);
 int bdrv_pread(BlockDriverState *bs, int64_t offset,
@@ -395,9 +398,7 @@ typedef enum {
     BLKDBG_L2_ALLOC_COW_READ,
     BLKDBG_L2_ALLOC_WRITE,
 
-    BLKDBG_READ,
     BLKDBG_READ_AIO,
-    BLKDBG_READ_BACKING,
     BLKDBG_READ_BACKING_AIO,
     BLKDBG_READ_COMPRESSED,
 

block/blkdebug.c

@@ -26,24 +26,10 @@
 #include "block_int.h"
 #include "module.h"
 
-typedef struct BlkdebugVars {
-    int state;
-
-    /* If inject_errno != 0, an error is injected for requests */
-    int inject_errno;
-
-    /* Decides if all future requests fail (false) or only the next one and
-     * after the next request inject_errno is reset to 0 (true) */
-    bool inject_once;
-
-    /* Decides if aio_readv/writev fails right away (true) or returns an error
-     * return value only in the callback (false) */
-    bool inject_immediately;
-} BlkdebugVars;
-
 typedef struct BDRVBlkdebugState {
-    BlkdebugVars vars;
-    QLIST_HEAD(list, BlkdebugRule) rules[BLKDBG_EVENT_MAX];
+    int state;
+    QLIST_HEAD(, BlkdebugRule) rules[BLKDBG_EVENT_MAX];
+    QSIMPLEQ_HEAD(, BlkdebugRule) active_rules;
 } BDRVBlkdebugState;
 
 typedef struct BlkdebugAIOCB {
@@ -73,12 +59,14 @@ typedef struct BlkdebugRule {
             int error;
             int immediately;
             int once;
+            int64_t sector;
         } inject;
         struct {
             int new_state;
         } set_state;
     } options;
     QLIST_ENTRY(BlkdebugRule) next;
+    QSIMPLEQ_ENTRY(BlkdebugRule) active_next;
 } BlkdebugRule;
 
 static QemuOptsList inject_error_opts = {
@@ -97,6 +85,10 @@ static QemuOptsList inject_error_opts = {
             .name = "errno",
             .type = QEMU_OPT_NUMBER,
         },
+        {
+            .name = "sector",
+            .type = QEMU_OPT_NUMBER,
+        },
         {
             .name = "once",
             .type = QEMU_OPT_BOOL,
@@ -147,9 +139,7 @@ static const char *event_names[BLKDBG_EVENT_MAX] = {
     [BLKDBG_L2_ALLOC_COW_READ]              = "l2_alloc.cow_read",
     [BLKDBG_L2_ALLOC_WRITE]                 = "l2_alloc.write",
 
-    [BLKDBG_READ]                           = "read",
     [BLKDBG_READ_AIO]                       = "read_aio",
-    [BLKDBG_READ_BACKING]                   = "read_backing",
     [BLKDBG_READ_BACKING_AIO]               = "read_backing_aio",
     [BLKDBG_READ_COMPRESSED]                = "read_compressed",
 
@@ -228,6 +218,7 @@ static int add_rule(QemuOpts *opts, void *opaque)
         rule->options.inject.once  = qemu_opt_get_bool(opts, "once", 0);
         rule->options.inject.immediately =
             qemu_opt_get_bool(opts, "immediately", 0);
+        rule->options.inject.sector = qemu_opt_get_number(opts, "sector", -1);
         break;
 
     case ACTION_SET_STATE:
@@ -302,7 +293,7 @@ static int blkdebug_open(BlockDriverState *bs, const char *filename, int flags)
     filename = c + 1;
 
     /* Set initial state */
-    s->vars.state = 1;
+    s->state = 1;
 
     /* Open the backing file */
     ret = bdrv_file_open(&bs->file, filename, flags);
@@ -328,18 +319,18 @@ static void blkdebug_aio_cancel(BlockDriverAIOCB *blockacb)
 }
 
 static BlockDriverAIOCB *inject_error(BlockDriverState *bs,
-    BlockDriverCompletionFunc *cb, void *opaque)
+    BlockDriverCompletionFunc *cb, void *opaque, BlkdebugRule *rule)
 {
     BDRVBlkdebugState *s = bs->opaque;
-    int error = s->vars.inject_errno;
+    int error = rule->options.inject.error;
     struct BlkdebugAIOCB *acb;
     QEMUBH *bh;
 
-    if (s->vars.inject_once) {
-        s->vars.inject_errno = 0;
+    if (rule->options.inject.once) {
+        QSIMPLEQ_INIT(&s->active_rules);
     }
 
-    if (s->vars.inject_immediately) {
+    if (rule->options.inject.immediately) {
         return NULL;
     }
 
@@ -358,14 +349,21 @@ static BlockDriverAIOCB *blkdebug_aio_readv(BlockDriverState *bs,
     BlockDriverCompletionFunc *cb, void *opaque)
 {
     BDRVBlkdebugState *s = bs->opaque;
+    BlkdebugRule *rule = NULL;
 
-    if (s->vars.inject_errno) {
-        return inject_error(bs, cb, opaque);
+    QSIMPLEQ_FOREACH(rule, &s->active_rules, active_next) {
+        if (rule->options.inject.sector == -1 ||
+            (rule->options.inject.sector >= sector_num &&
+             rule->options.inject.sector < sector_num + nb_sectors)) {
+            break;
+        }
+    }
+
+    if (rule && rule->options.inject.error) {
+        return inject_error(bs, cb, opaque, rule);
     }
 
-    BlockDriverAIOCB *acb =
-        bdrv_aio_readv(bs->file, sector_num, qiov, nb_sectors, cb, opaque);
-    return acb;
+    return bdrv_aio_readv(bs->file, sector_num, qiov, nb_sectors, cb, opaque);
 }
 
 static BlockDriverAIOCB *blkdebug_aio_writev(BlockDriverState *bs,
@@ -373,14 +371,21 @@ static BlockDriverAIOCB *blkdebug_aio_writev(BlockDriverState *bs,
     BlockDriverCompletionFunc *cb, void *opaque)
 {
     BDRVBlkdebugState *s = bs->opaque;
+    BlkdebugRule *rule = NULL;
+
+    QSIMPLEQ_FOREACH(rule, &s->active_rules, active_next) {
+        if (rule->options.inject.sector == -1 ||
+            (rule->options.inject.sector >= sector_num &&
+             rule->options.inject.sector < sector_num + nb_sectors)) {
+            break;
+        }
+    }
 
-    if (s->vars.inject_errno) {
-        return inject_error(bs, cb, opaque);
+    if (rule && rule->options.inject.error) {
+        return inject_error(bs, cb, opaque, rule);
     }
 
-    BlockDriverAIOCB *acb =
-        bdrv_aio_writev(bs->file, sector_num, qiov, nb_sectors, cb, opaque);
-    return acb;
+    return bdrv_aio_writev(bs->file, sector_num, qiov, nb_sectors, cb, opaque);
 }
 
 static void blkdebug_close(BlockDriverState *bs)
@@ -397,44 +402,53 @@ static void blkdebug_close(BlockDriverState *bs)
     }
 }
 
-static void process_rule(BlockDriverState *bs, struct BlkdebugRule *rule,
-    BlkdebugVars *old_vars)
+static bool process_rule(BlockDriverState *bs, struct BlkdebugRule *rule,
+    int old_state, bool injected)
 {
     BDRVBlkdebugState *s = bs->opaque;
-    BlkdebugVars *vars = &s->vars;
 
     /* Only process rules for the current state */
-    if (rule->state && rule->state != old_vars->state) {
-        return;
+    if (rule->state && rule->state != old_state) {
+        return injected;
     }
 
     /* Take the action */
     switch (rule->action) {
     case ACTION_INJECT_ERROR:
-        vars->inject_errno       = rule->options.inject.error;
-        vars->inject_once        = rule->options.inject.once;
-        vars->inject_immediately = rule->options.inject.immediately;
+        if (!injected) {
+            QSIMPLEQ_INIT(&s->active_rules);
+            injected = true;
+        }
+        QSIMPLEQ_INSERT_HEAD(&s->active_rules, rule, active_next);
         break;
 
     case ACTION_SET_STATE:
-        vars->state              = rule->options.set_state.new_state;
+        s->state = rule->options.set_state.new_state;
         break;
     }
+    return injected;
 }
 
 static void blkdebug_debug_event(BlockDriverState *bs, BlkDebugEvent event)
 {
     BDRVBlkdebugState *s = bs->opaque;
     struct BlkdebugRule *rule;
-    BlkdebugVars old_vars = s->vars;
+    int old_state = s->state;
+    bool injected;
 
     assert((int)event >= 0 && event < BLKDBG_EVENT_MAX);
 
+    injected = false;
     QLIST_FOREACH(rule, &s->rules[event], next) {
-        process_rule(bs, rule, &old_vars);
+        injected = process_rule(bs, rule, old_state, injected);
     }
 }
 
+static int64_t blkdebug_getlength(BlockDriverState *bs)
+{
+    return bdrv_getlength(bs->file);
+}
+
 static BlockDriver bdrv_blkdebug = {
     .format_name        = "blkdebug",
     .protocol_name      = "blkdebug",
@@ -443,6 +457,7 @@ static BlockDriver bdrv_blkdebug = {
 
     .bdrv_file_open     = blkdebug_open,
     .bdrv_close         = blkdebug_close,
+    .bdrv_getlength     = blkdebug_getlength,
 
     .bdrv_aio_readv     = blkdebug_aio_readv,
     .bdrv_aio_writev    = blkdebug_aio_writev,

block/qcow2-refcount.c

@@ -627,10 +627,11 @@ int64_t qcow2_alloc_bytes(BlockDriverState *bs, int size)
     BLKDBG_EVENT(bs->file, BLKDBG_CLUSTER_ALLOC_BYTES);
     assert(size > 0 && size <= s->cluster_size);
     if (s->free_byte_offset == 0) {
-        s->free_byte_offset = qcow2_alloc_clusters(bs, s->cluster_size);
-        if (s->free_byte_offset < 0) {
-            return s->free_byte_offset;
+        offset = qcow2_alloc_clusters(bs, s->cluster_size);
+        if (offset < 0) {
+            return offset;
         }
+        s->free_byte_offset = offset;
     }
  redo:
     free_in_cluster = s->cluster_size -

block/qcow2-snapshot.c

@@ -405,7 +405,7 @@ int qcow2_snapshot_create(BlockDriverState *bs, QEMUSnapshotInfo *sn_info)
 #ifdef DEBUG_ALLOC
     {
       BdrvCheckResult result = {0};
-      qcow2_check_refcounts(bs, &result);
+      qcow2_check_refcounts(bs, &result, 0);
     }
 #endif
     return 0;
@@ -522,7 +522,7 @@ int qcow2_snapshot_goto(BlockDriverState *bs, const char *snapshot_id)
 #ifdef DEBUG_ALLOC
     {
         BdrvCheckResult result = {0};
-        qcow2_check_refcounts(bs, &result);
+        qcow2_check_refcounts(bs, &result, 0);
     }
 #endif
     return 0;
@@ -582,7 +582,7 @@ int qcow2_snapshot_delete(BlockDriverState *bs, const char *snapshot_id)
 #ifdef DEBUG_ALLOC
     {
         BdrvCheckResult result = {0};
-        qcow2_check_refcounts(bs, &result);
+        qcow2_check_refcounts(bs, &result, 0);
     }
 #endif
     return 0;

block/qcow2.c

@@ -415,7 +415,7 @@ static int qcow2_open(BlockDriverState *bs, int flags)
 #ifdef DEBUG_ALLOC
     {
         BdrvCheckResult result = {0};
-        qcow2_check_refcounts(bs, &result);
+        qcow2_check_refcounts(bs, &result, 0);
     }
 #endif
     return ret;

block/qed.c

@@ -748,7 +748,7 @@ static void qed_read_backing_file(BDRVQEDState *s, uint64_t pos,
     /* If the read straddles the end of the backing file, shorten it */
     size = MIN((uint64_t)backing_length - pos, qiov->size);
 
-    BLKDBG_EVENT(s->bs->file, BLKDBG_READ_BACKING);
+    BLKDBG_EVENT(s->bs->file, BLKDBG_READ_BACKING_AIO);
     bdrv_aio_readv(s->bs->backing_hd, pos / BDRV_SECTOR_SIZE,
                    qiov, size / BDRV_SECTOR_SIZE, cb, opaque);
 }

block/raw.c

@@ -12,12 +12,14 @@ static int raw_open(BlockDriverState *bs, int flags)
 static int coroutine_fn raw_co_readv(BlockDriverState *bs, int64_t sector_num,
                                      int nb_sectors, QEMUIOVector *qiov)
 {
+    BLKDBG_EVENT(bs->file, BLKDBG_READ_AIO);
     return bdrv_co_readv(bs->file, sector_num, nb_sectors, qiov);
 }
 
 static int coroutine_fn raw_co_writev(BlockDriverState *bs, int64_t sector_num,
                                       int nb_sectors, QEMUIOVector *qiov)
 {
+    BLKDBG_EVENT(bs->file, BLKDBG_WRITE_AIO);
     return bdrv_co_writev(bs->file, sector_num, nb_sectors, qiov);
 }
 

block/sheepdog.c

@@ -259,8 +259,7 @@ typedef struct AIOReq {
     uint8_t flags;
     uint32_t id;
 
-    QLIST_ENTRY(AIOReq) outstanding_aio_siblings;
-    QLIST_ENTRY(AIOReq) aioreq_siblings;
+    QLIST_ENTRY(AIOReq) aio_siblings;
 } AIOReq;
 
 enum AIOCBState {
@@ -283,8 +282,7 @@ struct SheepdogAIOCB {
     void (*aio_done_func)(SheepdogAIOCB *);
 
     int canceled;
-
-    QLIST_HEAD(aioreq_head, AIOReq) aioreq_head;
+    int nr_pending;
 };
 
 typedef struct BDRVSheepdogState {
@@ -307,7 +305,8 @@ typedef struct BDRVSheepdogState {
     Coroutine *co_recv;
 
     uint32_t aioreq_seq_num;
-    QLIST_HEAD(outstanding_aio_head, AIOReq) outstanding_aio_head;
+    QLIST_HEAD(inflight_aio_head, AIOReq) inflight_aio_head;
+    QLIST_HEAD(pending_aio_head, AIOReq) pending_aio_head;
 } BDRVSheepdogState;
 
 static const char * sd_strerror(int err)
@@ -358,7 +357,7 @@ static const char * sd_strerror(int err)
  * Sheepdog I/O handling:
  *
  * 1. In sd_co_rw_vector, we send the I/O requests to the server and
- *    link the requests to the outstanding_list in the
+ *    link the requests to the inflight_list in the
  *    BDRVSheepdogState.  The function exits without waiting for
  *    receiving the response.
  *
@@ -386,21 +385,18 @@ static inline AIOReq *alloc_aio_req(BDRVSheepdogState *s, SheepdogAIOCB *acb,
     aio_req->flags = flags;
     aio_req->id = s->aioreq_seq_num++;
 
-    QLIST_INSERT_HEAD(&s->outstanding_aio_head, aio_req,
-                      outstanding_aio_siblings);
-    QLIST_INSERT_HEAD(&acb->aioreq_head, aio_req, aioreq_siblings);
-
+    acb->nr_pending++;
     return aio_req;
 }
 
-static inline int free_aio_req(BDRVSheepdogState *s, AIOReq *aio_req)
+static inline void free_aio_req(BDRVSheepdogState *s, AIOReq *aio_req)
 {
     SheepdogAIOCB *acb = aio_req->aiocb;
-    QLIST_REMOVE(aio_req, outstanding_aio_siblings);
-    QLIST_REMOVE(aio_req, aioreq_siblings);
+
+    QLIST_REMOVE(aio_req, aio_siblings);
     g_free(aio_req);
 
-    return !QLIST_EMPTY(&acb->aioreq_head);
+    acb->nr_pending--;
 }
 
 static void coroutine_fn sd_finish_aiocb(SheepdogAIOCB *acb)
@@ -446,7 +442,7 @@ static SheepdogAIOCB *sd_aio_setup(BlockDriverState *bs, QEMUIOVector *qiov,
     acb->canceled = 0;
     acb->coroutine = qemu_coroutine_self();
     acb->ret = 0;
-    QLIST_INIT(&acb->aioreq_head);
+    acb->nr_pending = 0;
     return acb;
 }
 
@@ -541,11 +537,18 @@ static coroutine_fn int send_co_req(int sockfd, SheepdogReq *hdr, void *data,
     return ret;
 }
 
+static coroutine_fn int do_co_req(int sockfd, SheepdogReq *hdr, void *data,
+                                  unsigned int *wlen, unsigned int *rlen);
+
 static int do_req(int sockfd, SheepdogReq *hdr, void *data,
                   unsigned int *wlen, unsigned int *rlen)
 {
     int ret;
 
+    if (qemu_in_coroutine()) {
+        return do_co_req(sockfd, hdr, data, wlen, rlen);
+    }
+
     socket_set_block(sockfd);
     ret = send_req(sockfd, hdr, data, wlen);
     if (ret < 0) {
@@ -577,10 +580,21 @@ out:
     return ret;
 }
 
+static void restart_co_req(void *opaque)
+{
+    Coroutine *co = opaque;
+
+    qemu_coroutine_enter(co, NULL);
+}
+
 static coroutine_fn int do_co_req(int sockfd, SheepdogReq *hdr, void *data,
                                   unsigned int *wlen, unsigned int *rlen)
 {
     int ret;
+    Coroutine *co;
+
+    co = qemu_coroutine_self();
+    qemu_aio_set_fd_handler(sockfd, NULL, restart_co_req, NULL, co);
 
     socket_set_block(sockfd);
     ret = send_co_req(sockfd, hdr, data, wlen);
@@ -588,6 +602,8 @@ static coroutine_fn int do_co_req(int sockfd, SheepdogReq *hdr, void *data,
         goto out;
     }
 
+    qemu_aio_set_fd_handler(sockfd, restart_co_req, NULL, NULL, co);
+
     ret = qemu_co_recv(sockfd, hdr, sizeof(*hdr));
     if (ret < sizeof(*hdr)) {
         error_report("failed to get a rsp, %s", strerror(errno));
@@ -609,6 +625,7 @@ static coroutine_fn int do_co_req(int sockfd, SheepdogReq *hdr, void *data,
     }
     ret = 0;
 out:
+    qemu_aio_set_fd_handler(sockfd, NULL, NULL, NULL, NULL);
     socket_set_nonblock(sockfd);
     return ret;
 }
@@ -617,32 +634,41 @@ static int coroutine_fn add_aio_request(BDRVSheepdogState *s, AIOReq *aio_req,
                            struct iovec *iov, int niov, int create,
                            enum AIOCBState aiocb_type);
 
+
+static AIOReq *find_pending_req(BDRVSheepdogState *s, uint64_t oid)
+{
+    AIOReq *aio_req;
+
+    QLIST_FOREACH(aio_req, &s->pending_aio_head, aio_siblings) {
+        if (aio_req->oid == oid) {
+            return aio_req;
+        }
+    }
+
+    return NULL;
+}
+
 /*
  * This function searchs pending requests to the object `oid', and
  * sends them.
  */
-static void coroutine_fn send_pending_req(BDRVSheepdogState *s, uint64_t oid, uint32_t id)
+static void coroutine_fn send_pending_req(BDRVSheepdogState *s, uint64_t oid)
 {
-    AIOReq *aio_req, *next;
+    AIOReq *aio_req;
     SheepdogAIOCB *acb;
     int ret;
 
-    QLIST_FOREACH_SAFE(aio_req, &s->outstanding_aio_head,
-                       outstanding_aio_siblings, next) {
-        if (id == aio_req->id) {
-            continue;
-        }
-        if (aio_req->oid != oid) {
-            continue;
-        }
-
+    while ((aio_req = find_pending_req(s, oid)) != NULL) {
         acb = aio_req->aiocb;
+        /* move aio_req from pending list to inflight one */
+        QLIST_REMOVE(aio_req, aio_siblings);
+        QLIST_INSERT_HEAD(&s->inflight_aio_head, aio_req, aio_siblings);
         ret = add_aio_request(s, aio_req, acb->qiov->iov,
                               acb->qiov->niov, 0, acb->aiocb_type);
         if (ret < 0) {
             error_report("add_aio_request is failed");
             free_aio_req(s, aio_req);
-            if (QLIST_EMPTY(&acb->aioreq_head)) {
+            if (!acb->nr_pending) {
                 sd_finish_aiocb(acb);
             }
         }
@@ -663,10 +689,9 @@ static void coroutine_fn aio_read_response(void *opaque)
     int ret;
     AIOReq *aio_req = NULL;
     SheepdogAIOCB *acb;
-    int rest;
     unsigned long idx;
 
-    if (QLIST_EMPTY(&s->outstanding_aio_head)) {
+    if (QLIST_EMPTY(&s->inflight_aio_head)) {
         goto out;
     }
 
@@ -677,8 +702,8 @@ static void coroutine_fn aio_read_response(void *opaque)
         goto out;
     }
 
-    /* find the right aio_req from the outstanding_aio list */
-    QLIST_FOREACH(aio_req, &s->outstanding_aio_head, outstanding_aio_siblings) {
+    /* find the right aio_req from the inflight aio list */
+    QLIST_FOREACH(aio_req, &s->inflight_aio_head, aio_siblings) {
         if (aio_req->id == rsp.id) {
             break;
         }
@@ -716,7 +741,7 @@ static void coroutine_fn aio_read_response(void *opaque)
              * create requests are not allowed, so we search the
              * pending requests here.
              */
-            send_pending_req(s, vid_to_data_oid(s->inode.vdi_id, idx), rsp.id);
+            send_pending_req(s, vid_to_data_oid(s->inode.vdi_id, idx));
         }
         break;
     case AIOCB_READ_UDATA:
@@ -734,8 +759,8 @@ static void coroutine_fn aio_read_response(void *opaque)
         error_report("%s", sd_strerror(rsp.result));
     }
 
-    rest = free_aio_req(s, aio_req);
-    if (!rest) {
+    free_aio_req(s, aio_req);
+    if (!acb->nr_pending) {
         /*
          * We've finished all requests which belong to the AIOCB, so
          * we can switch back to sd_co_readv/writev now.
@@ -768,7 +793,8 @@ static int aio_flush_request(void *opaque)
 {
     BDRVSheepdogState *s = opaque;
 
-    return !QLIST_EMPTY(&s->outstanding_aio_head);
+    return !QLIST_EMPTY(&s->inflight_aio_head) ||
+        !QLIST_EMPTY(&s->pending_aio_head);
 }
 
 static int set_nodelay(int fd)
@@ -1085,7 +1111,8 @@ static int sd_open(BlockDriverState *bs, const char *filename, int flags)
 
     strstart(filename, "sheepdog:", (const char **)&filename);
 
-    QLIST_INIT(&s->outstanding_aio_head);
+    QLIST_INIT(&s->inflight_aio_head);
+    QLIST_INIT(&s->pending_aio_head);
     s->fd = -1;
 
     memset(vdi, 0, sizeof(vdi));
@@ -1447,6 +1474,7 @@ static void coroutine_fn sd_write_done(SheepdogAIOCB *acb)
         iov.iov_len = sizeof(s->inode);
         aio_req = alloc_aio_req(s, acb, vid_to_vdi_oid(s->inode.vdi_id),
                                 data_len, offset, 0, 0, offset);
+        QLIST_INSERT_HEAD(&s->inflight_aio_head, aio_req, aio_siblings);
         ret = add_aio_request(s, aio_req, &iov, 1, 0, AIOCB_WRITE_UDATA);
         if (ret) {
             free_aio_req(s, aio_req);
@@ -1515,7 +1543,7 @@ out:
  * Send I/O requests to the server.
  *
  * This function sends requests to the server, links the requests to
- * the outstanding_list in BDRVSheepdogState, and exits without
+ * the inflight_list in BDRVSheepdogState, and exits without
  * waiting the response.  The responses are received in the
  * `aio_read_response' function which is called from the main loop as
  * a fd handler.
@@ -1547,6 +1575,12 @@ static int coroutine_fn sd_co_rw_vector(void *p)
         }
     }
 
+    /*
+     * Make sure we don't free the aiocb before we are done with all requests.
+     * This additional reference is dropped at the end of this function.
+     */
+    acb->nr_pending++;
+
     while (done != total) {
         uint8_t flags = 0;
         uint64_t old_oid = 0;
@@ -1571,22 +1605,18 @@ static int coroutine_fn sd_co_rw_vector(void *p)
         }
 
         if (create) {
-            dprintf("update ino (%" PRIu32") %" PRIu64 " %" PRIu64
-                    " %" PRIu64 "\n", inode->vdi_id, oid,
+            dprintf("update ino (%" PRIu32 ") %" PRIu64 " %" PRIu64 " %ld\n",
+                    inode->vdi_id, oid,
                     vid_to_data_oid(inode->data_vdi_id[idx], idx), idx);
             oid = vid_to_data_oid(inode->vdi_id, idx);
-            dprintf("new oid %lx\n", oid);
+            dprintf("new oid %" PRIx64 "\n", oid);
         }
 
         aio_req = alloc_aio_req(s, acb, oid, len, offset, flags, old_oid, done);
 
         if (create) {
             AIOReq *areq;
-            QLIST_FOREACH(areq, &s->outstanding_aio_head,
-                          outstanding_aio_siblings) {
-                if (areq == aio_req) {
-                    continue;
-                }
+            QLIST_FOREACH(areq, &s->inflight_aio_head, aio_siblings) {
                 if (areq->oid == oid) {
                     /*
                      * Sheepdog cannot handle simultaneous create
@@ -1596,11 +1626,14 @@ static int coroutine_fn sd_co_rw_vector(void *p)
                      */
                     aio_req->flags = 0;
                     aio_req->base_oid = 0;
+                    QLIST_INSERT_HEAD(&s->pending_aio_head, aio_req,
+                                      aio_siblings);
                     goto done;
                 }
             }
         }
 
+        QLIST_INSERT_HEAD(&s->inflight_aio_head, aio_req, aio_siblings);
         ret = add_aio_request(s, aio_req, acb->qiov->iov, acb->qiov->niov,
                               create, acb->aiocb_type);
         if (ret < 0) {
@@ -1615,7 +1648,7 @@ static int coroutine_fn sd_co_rw_vector(void *p)
         done += len;
     }
 out:
-    if (QLIST_EMPTY(&acb->aioreq_head)) {
+    if (!--acb->nr_pending) {
         return acb->ret;
     }
     return 1;
@@ -1628,7 +1661,6 @@ static coroutine_fn int sd_co_writev(BlockDriverState *bs, int64_t sector_num,
     int ret;
 
     if (bs->growable && sector_num + nb_sectors > bs->total_sectors) {
-        /* TODO: shouldn't block here */
         ret = sd_truncate(bs, (sector_num + nb_sectors) * SECTOR_SIZE);
         if (ret < 0) {
             return ret;
@@ -1696,7 +1728,7 @@ static int coroutine_fn sd_co_flush_to_disk(BlockDriverState *bs)
     hdr.opcode = SD_OP_FLUSH_VDI;
     hdr.oid = vid_to_vdi_oid(inode->vdi_id);
 
-    ret = do_co_req(s->flush_fd, (SheepdogReq *)&hdr, NULL, &wlen, &rlen);
+    ret = do_req(s->flush_fd, (SheepdogReq *)&hdr, NULL, &wlen, &rlen);
     if (ret) {
         error_report("failed to send a request to the sheep");
         return ret;
@@ -1726,7 +1758,7 @@ static int sd_snapshot_create(BlockDriverState *bs, QEMUSnapshotInfo *sn_info)
     SheepdogInode *inode;
     unsigned int datalen;
 
-    dprintf("sn_info: name %s id_str %s s: name %s vm_state_size %d "
+    dprintf("sn_info: name %s id_str %s s: name %s vm_state_size %" PRId64 " "
             "is_snapshot %d\n", sn_info->name, sn_info->id_str,
             s->name, sn_info->vm_state_size, s->is_snapshot);
 

blockdev.c

@@ -609,6 +609,10 @@ DriveInfo *drive_init(QemuOpts *opts, int default_to_scsi)
 
     bdrv_flags |= ro ? 0 : BDRV_O_RDWR;
 
+    if (ro && copy_on_read) {
+        error_report("warning: disabling copy_on_read on readonly drive");
+    }
+
     ret = bdrv_open(dinfo->bdrv, file, bdrv_flags, drv);
     if (ret < 0) {
         error_report("could not open disk image %s: %s",

hw/fdc.c

@@ -153,8 +153,12 @@ static int fd_seek(FDrive *drv, uint8_t head, uint8_t track, uint8_t sect,
         }
 #endif
         drv->head = head;
-        if (drv->track != track)
+        if (drv->track != track) {
+            if (drv->bs != NULL && bdrv_is_inserted(drv->bs)) {
+                drv->media_changed = 0;
+            }
             ret = 1;
+        }
         drv->track = track;
         drv->sect = sect;
     }
@@ -170,9 +174,7 @@ static int fd_seek(FDrive *drv, uint8_t head, uint8_t track, uint8_t sect,
 static void fd_recalibrate(FDrive *drv)
 {
     FLOPPY_DPRINTF("recalibrate\n");
-    drv->head = 0;
-    drv->track = 0;
-    drv->sect = 1;
+    fd_seek(drv, 0, 0, 1, 1);
 }
 
 /* Revalidate a disk drive after a disk change */
@@ -189,9 +191,6 @@ static void fd_revalidate(FDrive *drv)
                                       &last_sect, drv->drive, &drive, &rate);
         if (!bdrv_is_inserted(drv->bs)) {
             FLOPPY_DPRINTF("No disk in drive\n");
-        } else if (nb_heads != 0 && max_track != 0 && last_sect != 0) {
-            FLOPPY_DPRINTF("User defined disk (%d %d %d)\n",
-                           nb_heads - 1, max_track, last_sect);
         } else {
             FLOPPY_DPRINTF("Floppy disk (%d h %d t %d s) %s\n", nb_heads,
                            max_track, last_sect, ro ? "ro" : "rw");
@@ -305,6 +304,9 @@ enum {
 };
 
 enum {
+    FD_SR0_DS0      = 0x01,
+    FD_SR0_DS1      = 0x02,
+    FD_SR0_HEAD     = 0x04,
     FD_SR0_EQPMT    = 0x10,
     FD_SR0_SEEK     = 0x20,
     FD_SR0_ABNTERM  = 0x40,
@@ -711,14 +713,6 @@ static void fdctrl_raise_irq(FDCtrl *fdctrl, uint8_t status0)
         qemu_set_irq(fdctrl->irq, 1);
         fdctrl->sra |= FD_SRA_INTPEND;
     }
-    if (status0 & FD_SR0_SEEK) {
-        FDrive *cur_drv;
-        /* A seek clears the disk change line (if a disk is inserted) */
-        cur_drv = get_cur_drv(fdctrl);
-        if (cur_drv->bs != NULL && bdrv_is_inserted(cur_drv->bs)) {
-            cur_drv->media_changed = 0;
-        }
-    }
 
     fdctrl->reset_sensei = 0;
     fdctrl->status0 = status0;
@@ -978,14 +972,15 @@ static void fdctrl_reset_fifo(FDCtrl *fdctrl)
 }
 
 /* Set FIFO status for the host to read */
-static void fdctrl_set_fifo(FDCtrl *fdctrl, int fifo_len, int do_irq)
+static void fdctrl_set_fifo(FDCtrl *fdctrl, int fifo_len, uint8_t status0)
 {
     fdctrl->data_dir = FD_DIR_READ;
     fdctrl->data_len = fifo_len;
     fdctrl->data_pos = 0;
     fdctrl->msr |= FD_MSR_CMDBUSY | FD_MSR_RQM | FD_MSR_DIO;
-    if (do_irq)
-        fdctrl_raise_irq(fdctrl, 0x00);
+    if (status0) {
+        fdctrl_raise_irq(fdctrl, status0);
+    }
 }
 
 /* Set an error: unimplemented/unknown command */
@@ -997,7 +992,10 @@ static void fdctrl_unimplemented(FDCtrl *fdctrl, int direction)
     fdctrl_set_fifo(fdctrl, 1, 0);
 }
 
-/* Seek to next sector */
+/* Seek to next sector
+ * returns 0 when end of track reached (for DBL_SIDES on head 1)
+ * otherwise returns 1
+ */
 static int fdctrl_seek_to_next_sect(FDCtrl *fdctrl, FDrive *cur_drv)
 {
     FLOPPY_DPRINTF("seek to next sector (%d %02x %02x => %d)\n",
@@ -1005,30 +1003,39 @@ static int fdctrl_seek_to_next_sect(FDCtrl *fdctrl, FDrive *cur_drv)
                    fd_sector(cur_drv));
     /* XXX: cur_drv->sect >= cur_drv->last_sect should be an
        error in fact */
-    if (cur_drv->sect >= cur_drv->last_sect ||
-        cur_drv->sect == fdctrl->eot) {
-        cur_drv->sect = 1;
+    uint8_t new_head = cur_drv->head;
+    uint8_t new_track = cur_drv->track;
+    uint8_t new_sect = cur_drv->sect;
+
+    int ret = 1;
+
+    if (new_sect >= cur_drv->last_sect ||
+        new_sect == fdctrl->eot) {
+        new_sect = 1;
         if (FD_MULTI_TRACK(fdctrl->data_state)) {
-            if (cur_drv->head == 0 &&
+            if (new_head == 0 &&
                 (cur_drv->flags & FDISK_DBL_SIDES) != 0) {
-                cur_drv->head = 1;
+                new_head = 1;
             } else {
-                cur_drv->head = 0;
-                cur_drv->track++;
-                if ((cur_drv->flags & FDISK_DBL_SIDES) == 0)
-                    return 0;
+                new_head = 0;
+                new_track++;
+                if ((cur_drv->flags & FDISK_DBL_SIDES) == 0) {
+                    ret = 0;
+                }
             }
         } else {
-            cur_drv->track++;
-            return 0;
+            new_track++;
+            ret = 0;
+        }
+        if (ret == 1) {
+            FLOPPY_DPRINTF("seek to next track (%d %02x %02x => %d)\n",
+                    new_head, new_track, new_sect, fd_sector(cur_drv));
         }
-        FLOPPY_DPRINTF("seek to next track (%d %02x %02x => %d)\n",
-                       cur_drv->head, cur_drv->track,
-                       cur_drv->sect, fd_sector(cur_drv));
     } else {
-        cur_drv->sect++;
+        new_sect++;
     }
-    return 1;
+    fd_seek(cur_drv, new_head, new_track, new_sect, 1);
+    return ret;
 }
 
 /* Callback for transfer end (stop or abort) */
@@ -1038,10 +1045,12 @@ static void fdctrl_stop_transfer(FDCtrl *fdctrl, uint8_t status0,
     FDrive *cur_drv;
 
     cur_drv = get_cur_drv(fdctrl);
+    fdctrl->status0 = status0 | FD_SR0_SEEK | (cur_drv->head << 2) |
+                      GET_CUR_DRV(fdctrl);
+
     FLOPPY_DPRINTF("transfer status: %02x %02x %02x (%02x)\n",
-                   status0, status1, status2,
-                   status0 | (cur_drv->head << 2) | GET_CUR_DRV(fdctrl));
-    fdctrl->fifo[0] = status0 | (cur_drv->head << 2) | GET_CUR_DRV(fdctrl);
+                   status0, status1, status2, fdctrl->status0);
+    fdctrl->fifo[0] = fdctrl->status0;
     fdctrl->fifo[1] = status1;
     fdctrl->fifo[2] = status2;
     fdctrl->fifo[3] = cur_drv->track;
@@ -1054,7 +1063,7 @@ static void fdctrl_stop_transfer(FDCtrl *fdctrl, uint8_t status0,
     }
     fdctrl->msr |= FD_MSR_RQM | FD_MSR_DIO;
     fdctrl->msr &= ~FD_MSR_NONDMA;
-    fdctrl_set_fifo(fdctrl, 7, 1);
+    fdctrl_set_fifo(fdctrl, 7, fdctrl->status0);
 }
 
 /* Prepare a data transfer (either DMA or FIFO) */
@@ -1169,7 +1178,7 @@ static void fdctrl_start_transfer(FDCtrl *fdctrl, int direction)
     if (direction != FD_DIR_WRITE)
         fdctrl->msr |= FD_MSR_DIO;
     /* IO based transfer: calculate len */
-    fdctrl_raise_irq(fdctrl, 0x00);
+    fdctrl_raise_irq(fdctrl, FD_SR0_SEEK);
 
     return;
 }
@@ -1598,16 +1607,18 @@ static void fdctrl_handle_sense_interrupt_status(FDCtrl *fdctrl, int direction)
 {
     FDrive *cur_drv = get_cur_drv(fdctrl);
 
-    if(fdctrl->reset_sensei > 0) {
+    if (fdctrl->reset_sensei > 0) {
         fdctrl->fifo[0] =
             FD_SR0_RDYCHG + FD_RESET_SENSEI_COUNT - fdctrl->reset_sensei;
         fdctrl->reset_sensei--;
+    } else if (!(fdctrl->sra & FD_SRA_INTPEND)) {
+        fdctrl->fifo[0] = FD_SR0_INVCMD;
+        fdctrl_set_fifo(fdctrl, 1, 0);
+        return;
     } else {
-        /* XXX: status0 handling is broken for read/write
-           commands, so we do this hack. It should be suppressed
-           ASAP */
         fdctrl->fifo[0] =
-            FD_SR0_SEEK | (cur_drv->head << 2) | GET_CUR_DRV(fdctrl);
+                (fdctrl->status0 & ~(FD_SR0_HEAD | FD_SR0_DS1 | FD_SR0_DS0))
+                | GET_CUR_DRV(fdctrl);
     }
 
     fdctrl->fifo[1] = cur_drv->track;
@@ -1626,11 +1637,7 @@ static void fdctrl_handle_seek(FDCtrl *fdctrl, int direction)
     /* The seek command just sends step pulses to the drive and doesn't care if
      * there is a medium inserted of if it's banging the head against the drive.
      */
-    if (fdctrl->fifo[2] > cur_drv->max_track) {
-        cur_drv->track = cur_drv->max_track;
-    } else {
-        cur_drv->track = fdctrl->fifo[2];
-    }
+    fd_seek(cur_drv, cur_drv->head, fdctrl->fifo[2], cur_drv->sect, 1);
     /* Raise Interrupt */
     fdctrl_raise_irq(fdctrl, FD_SR0_SEEK);
 }
@@ -1695,9 +1702,10 @@ static void fdctrl_handle_relative_seek_out(FDCtrl *fdctrl, int direction)
     SET_CUR_DRV(fdctrl, fdctrl->fifo[1] & FD_DOR_SELMASK);
     cur_drv = get_cur_drv(fdctrl);
     if (fdctrl->fifo[2] + cur_drv->track >= cur_drv->max_track) {
-        cur_drv->track = cur_drv->max_track - 1;
+        fd_seek(cur_drv, cur_drv->head, cur_drv->max_track - 1,
+                cur_drv->sect, 1);
     } else {
-        cur_drv->track += fdctrl->fifo[2];
+        fd_seek(cur_drv, cur_drv->head, fdctrl->fifo[2], cur_drv->sect, 1);
     }
     fdctrl_reset_fifo(fdctrl);
     /* Raise Interrupt */
@@ -1711,9 +1719,9 @@ static void fdctrl_handle_relative_seek_in(FDCtrl *fdctrl, int direction)
     SET_CUR_DRV(fdctrl, fdctrl->fifo[1] & FD_DOR_SELMASK);
     cur_drv = get_cur_drv(fdctrl);
     if (fdctrl->fifo[2] > cur_drv->track) {
-        cur_drv->track = 0;
+        fd_seek(cur_drv, cur_drv->head, 0, cur_drv->sect, 1);
     } else {
-        cur_drv->track -= fdctrl->fifo[2];
+        fd_seek(cur_drv, cur_drv->head, fdctrl->fifo[2], cur_drv->sect, 1);
     }
     fdctrl_reset_fifo(fdctrl);
     /* Raise Interrupt */

tests/fdc-test.c

@@ -142,7 +142,7 @@ static uint8_t send_read_command(void)
     }
 
     st0 = floppy_recv();
-    if (st0 != 0x40) {
+    if (st0 != 0x60) {
         ret = 1;
     }
 
@@ -156,19 +156,16 @@ static uint8_t send_read_command(void)
     return ret;
 }
 
-static void send_step_pulse(void)
+static void send_step_pulse(int cyl)
 {
     int drive = 0;
     int head = 0;
-    static int cyl = 0;
 
     floppy_send(CMD_SEEK);
     floppy_send(head << 2 | drive);
     g_assert(!get_irq(FLOPPY_IRQ));
     floppy_send(cyl);
     ack_irq();
-
-    cyl = (cyl + 1) % 4;
 }
 
 static uint8_t cmos_read(uint8_t reg)
@@ -195,8 +192,7 @@ static void test_no_media_on_start(void)
     assert_bit_set(dir, DSKCHG);
     dir = inb(FLOPPY_BASE + reg_dir);
     assert_bit_set(dir, DSKCHG);
-    send_step_pulse();
-    send_step_pulse();
+    send_step_pulse(1);
     dir = inb(FLOPPY_BASE + reg_dir);
     assert_bit_set(dir, DSKCHG);
     dir = inb(FLOPPY_BASE + reg_dir);
@@ -227,7 +223,14 @@ static void test_media_change(void)
     dir = inb(FLOPPY_BASE + reg_dir);
     assert_bit_set(dir, DSKCHG);
 
-    send_step_pulse();
+    send_step_pulse(0);
+    dir = inb(FLOPPY_BASE + reg_dir);
+    assert_bit_set(dir, DSKCHG);
+    dir = inb(FLOPPY_BASE + reg_dir);
+    assert_bit_set(dir, DSKCHG);
+
+    /* Step to next track should clear DSKCHG bit. */
+    send_step_pulse(1);
     dir = inb(FLOPPY_BASE + reg_dir);
     assert_bit_clear(dir, DSKCHG);
     dir = inb(FLOPPY_BASE + reg_dir);
@@ -243,11 +246,39 @@ static void test_media_change(void)
     dir = inb(FLOPPY_BASE + reg_dir);
     assert_bit_set(dir, DSKCHG);
 
-    send_step_pulse();
+    send_step_pulse(0);
     dir = inb(FLOPPY_BASE + reg_dir);
     assert_bit_set(dir, DSKCHG);
     dir = inb(FLOPPY_BASE + reg_dir);
     assert_bit_set(dir, DSKCHG);
+
+    send_step_pulse(1);
+    dir = inb(FLOPPY_BASE + reg_dir);
+    assert_bit_set(dir, DSKCHG);
+    dir = inb(FLOPPY_BASE + reg_dir);
+    assert_bit_set(dir, DSKCHG);
+}
+
+static void test_sense_interrupt(void)
+{
+    int drive = 0;
+    int head = 0;
+    int cyl = 0;
+    int ret = 0;
+
+    floppy_send(CMD_SENSE_INT);
+    ret = floppy_recv();
+    g_assert(ret == 0x80);
+
+    floppy_send(CMD_SEEK);
+    floppy_send(head << 2 | drive);
+    g_assert(!get_irq(FLOPPY_IRQ));
+    floppy_send(cyl);
+
+    floppy_send(CMD_SENSE_INT);
+    ret = floppy_recv();
+    g_assert(ret == 0x20);
+    floppy_recv();
 }
 
 /* success if no crash or abort */
@@ -297,6 +328,7 @@ int main(int argc, char **argv)
     qtest_add_func("/fdc/no_media_on_start", test_no_media_on_start);
     qtest_add_func("/fdc/read_without_media", test_read_without_media);
     qtest_add_func("/fdc/media_change", test_media_change);
+    qtest_add_func("/fdc/sense_interrupt", test_sense_interrupt);
     qtest_add_func("/fdc/fuzz-registers", fuzz_registers);
 
     ret = g_test_run();

tests/libqtest.c

@@ -40,6 +40,7 @@ struct QTestState
     bool irq_level[MAX_IRQ];
     GString *rx;
     gchar *pid_file;
+    char *socket_path, *qmp_socket_path;
 };
 
 #define g_assert_no_errno(ret) do { \
@@ -88,8 +89,6 @@ QTestState *qtest_init(const char *extra_args)
 {
     QTestState *s;
     int sock, qmpsock, ret, i;
-    gchar *socket_path;
-    gchar *qmp_socket_path;
     gchar *pid_file;
     gchar *command;
     const char *qemu_binary;
@@ -98,14 +97,14 @@ QTestState *qtest_init(const char *extra_args)
     qemu_binary = getenv("QTEST_QEMU_BINARY");
     g_assert(qemu_binary != NULL);
 
-    socket_path = g_strdup_printf("/tmp/qtest-%d.sock", getpid());
-    qmp_socket_path = g_strdup_printf("/tmp/qtest-%d.qmp", getpid());
-    pid_file = g_strdup_printf("/tmp/qtest-%d.pid", getpid());
-
     s = g_malloc(sizeof(*s));
 
-    sock = init_socket(socket_path);
-    qmpsock = init_socket(qmp_socket_path);
+    s->socket_path = g_strdup_printf("/tmp/qtest-%d.sock", getpid());
+    s->qmp_socket_path = g_strdup_printf("/tmp/qtest-%d.qmp", getpid());
+    pid_file = g_strdup_printf("/tmp/qtest-%d.pid", getpid());
+
+    sock = init_socket(s->socket_path);
+    qmpsock = init_socket(s->qmp_socket_path);
 
     pid = fork();
     if (pid == 0) {
@@ -115,8 +114,8 @@ QTestState *qtest_init(const char *extra_args)
                                   "-qmp unix:%s,nowait "
                                   "-pidfile %s "
                                   "-machine accel=qtest "
-                                  "%s", qemu_binary, socket_path,
-                                  qmp_socket_path, pid_file,
+                                  "%s", qemu_binary, s->socket_path,
+                                  s->qmp_socket_path, pid_file,
                                   extra_args ?: "");
 
         ret = system(command);
@@ -133,9 +132,6 @@ QTestState *qtest_init(const char *extra_args)
         s->irq_level[i] = false;
     }
 
-    g_free(socket_path);
-    g_free(qmp_socket_path);
-
     /* Read the QMP greeting and then do the handshake */
     qtest_qmp(s, "");
     qtest_qmp(s, "{ 'execute': 'qmp_capabilities' }");
@@ -160,6 +156,13 @@ void qtest_quit(QTestState *s)
 
         fclose(f);
     }
+
+    unlink(s->pid_file);
+    unlink(s->socket_path);
+    unlink(s->qmp_socket_path);
+    g_free(s->pid_file);
+    g_free(s->socket_path);
+    g_free(s->qmp_socket_path);
 }
 
 static void socket_sendf(int fd, const char *fmt, va_list ap)