From: Stefan Hajnoczi <stefa...@redhat.com>
The block.c file has grown to over 6000 lines. It is time to split this
file so there are fewer conflicts and the code is easier to maintain.
Extract I/O request processing code:
* Read
* Write
* Flush
* Discard
* ioctl
* Tracked requests and queuing
* Throttling and copy-on-read
The patch simply moves code from block.c into block/io.c.
No code changes are made except adding the following block_int.h
functions so they can be called across block.c and block/io.c:
bdrv_drain_one(), bdrv_set_dirty(), bdrv_reset_dirty().
I/O request processing needs to set up BlockDriver coroutine and AIO
emulation function pointers, so add bdrv_setup_io_funcs(bdrv) interface
that block.c calls.
Signed-off-by: Stefan Hajnoczi <stefa...@redhat.com>
---
This patch produces identical results to Stefan's email, but is
MUCH more readable (hint: git config diff.algorithm patience)
block.c | 1980 +-------------------------------------------
block/Makefile.objs | 1 +
block/io.c | 1997 +++++++++++++++++++++++++++++++++++++++++++++
include/block/block_int.h | 14 +
4 files changed, 2015 insertions(+), 1977 deletions(-)
create mode 100644 block/io.c
diff --git a/block.c b/block.c
index d45e4dd..939a04a 100644
--- a/block.c
+++ b/block.c
@@ -58,36 +58,6 @@ struct BdrvDirtyBitmap {
#define NOT_DONE 0x7fffffff /* used while emulated sync operation in progress
*/
-static BlockAIOCB *bdrv_aio_readv_em(BlockDriverState *bs,
- int64_t sector_num, QEMUIOVector *qiov, int nb_sectors,
- BlockCompletionFunc *cb, void *opaque);
-static BlockAIOCB *bdrv_aio_writev_em(BlockDriverState *bs,
- int64_t sector_num, QEMUIOVector *qiov, int nb_sectors,
- BlockCompletionFunc *cb, void *opaque);
-static int coroutine_fn bdrv_co_readv_em(BlockDriverState *bs,
- int64_t sector_num, int nb_sectors,
- QEMUIOVector *iov);
-static int coroutine_fn bdrv_co_writev_em(BlockDriverState *bs,
- int64_t sector_num, int nb_sectors,
- QEMUIOVector *iov);
-static int coroutine_fn bdrv_co_do_preadv(BlockDriverState *bs,
- int64_t offset, unsigned int bytes, QEMUIOVector *qiov,
- BdrvRequestFlags flags);
-static int coroutine_fn bdrv_co_do_pwritev(BlockDriverState *bs,
- int64_t offset, unsigned int bytes, QEMUIOVector *qiov,
- BdrvRequestFlags flags);
-static BlockAIOCB *bdrv_co_aio_rw_vector(BlockDriverState *bs,
- int64_t sector_num,
- QEMUIOVector *qiov,
- int nb_sectors,
- BdrvRequestFlags flags,
- BlockCompletionFunc *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, BdrvRequestFlags flags);
-
static QTAILQ_HEAD(, BlockDriverState) bdrv_states =
QTAILQ_HEAD_INITIALIZER(bdrv_states);
@@ -97,10 +67,6 @@ static QTAILQ_HEAD(, BlockDriverState) graph_bdrv_states =
static QLIST_HEAD(, BlockDriver) bdrv_drivers =
QLIST_HEAD_INITIALIZER(bdrv_drivers);
-static void bdrv_set_dirty(BlockDriverState *bs, int64_t cur_sector,
- int nr_sectors);
-static void bdrv_reset_dirty(BlockDriverState *bs, int64_t cur_sector,
- int nr_sectors);
/* If non-zero, use only whitelisted block drivers */
static int use_bdrv_whitelist;
@@ -124,104 +90,6 @@ int is_windows_drive(const char *filename)
}
#endif
-/* throttling disk I/O limits */
-void bdrv_set_io_limits(BlockDriverState *bs,
- ThrottleConfig *cfg)
-{
- int i;
-
- throttle_config(&bs->throttle_state, cfg);
-
- for (i = 0; i < 2; i++) {
- qemu_co_enter_next(&bs->throttled_reqs[i]);
- }
-}
-
-/* this function drain all the throttled IOs */
-static bool bdrv_start_throttled_reqs(BlockDriverState *bs)
-{
- bool drained = false;
- bool enabled = bs->io_limits_enabled;
- int i;
-
- bs->io_limits_enabled = false;
-
- for (i = 0; i < 2; i++) {
- while (qemu_co_enter_next(&bs->throttled_reqs[i])) {
- drained = true;
- }
- }
-
- bs->io_limits_enabled = enabled;
-
- return drained;
-}
-
-void bdrv_io_limits_disable(BlockDriverState *bs)
-{
- bs->io_limits_enabled = false;
-
- bdrv_start_throttled_reqs(bs);
-
- throttle_destroy(&bs->throttle_state);
-}
-
-static void bdrv_throttle_read_timer_cb(void *opaque)
-{
- BlockDriverState *bs = opaque;
- qemu_co_enter_next(&bs->throttled_reqs[0]);
-}
-
-static void bdrv_throttle_write_timer_cb(void *opaque)
-{
- BlockDriverState *bs = opaque;
- qemu_co_enter_next(&bs->throttled_reqs[1]);
-}
-
-/* should be called before bdrv_set_io_limits if a limit is set */
-void bdrv_io_limits_enable(BlockDriverState *bs)
-{
- assert(!bs->io_limits_enabled);
- throttle_init(&bs->throttle_state,
- bdrv_get_aio_context(bs),
- QEMU_CLOCK_VIRTUAL,
- bdrv_throttle_read_timer_cb,
- bdrv_throttle_write_timer_cb,
- bs);
- bs->io_limits_enabled = true;
-}
-
-/* This function makes an IO wait if needed
- *
- * @nb_sectors: the number of sectors of the IO
- * @is_write: is the IO a write
- */
-static void bdrv_io_limits_intercept(BlockDriverState *bs,
- unsigned int bytes,
- bool is_write)
-{
- /* does this io must wait */
- bool must_wait = throttle_schedule_timer(&bs->throttle_state, is_write);
-
- /* if must wait or any request of this type throttled queue the IO */
- if (must_wait ||
- !qemu_co_queue_empty(&bs->throttled_reqs[is_write])) {
- qemu_co_queue_wait(&bs->throttled_reqs[is_write]);
- }
-
- /* the IO will be executed, do the accounting */
- throttle_account(&bs->throttle_state, is_write, bytes);
-
-
- /* if the next request must wait -> do nothing */
- if (throttle_schedule_timer(&bs->throttle_state, is_write)) {
- return;
- }
-
- /* else queue next request for execution */
- qemu_co_queue_next(&bs->throttled_reqs[is_write]);
-}
-
size_t bdrv_opt_mem_align(BlockDriverState *bs)
{
if (!bs || !bs->drv) {
@@ -335,20 +203,7 @@ void bdrv_get_full_backing_filename(BlockDriverState *bs,
char *dest, size_t sz,
void bdrv_register(BlockDriver *bdrv)
{
- /* Block drivers without coroutine functions need emulation */
- if (!bdrv->bdrv_co_readv) {
- bdrv->bdrv_co_readv = bdrv_co_readv_em;
- bdrv->bdrv_co_writev = bdrv_co_writev_em;
-
- /* bdrv_co_readv_em()/brdv_co_writev_em() work in terms of aio, so if
- * the block driver lacks aio we need to emulate that too.
- */
- if (!bdrv->bdrv_aio_readv) {
- /* add AIO emulation layer */
- bdrv->bdrv_aio_readv = bdrv_aio_readv_em;
- bdrv->bdrv_aio_writev = bdrv_aio_writev_em;
- }
- }
+ bdrv_setup_io_funcs(bdrv);
QLIST_INSERT_HEAD(&bdrv_drivers, bdrv, list);
}
@@ -805,22 +660,6 @@ int bdrv_parse_cache_flags(const char *mode, int *flags)
return 0;
}
-/**
- * The copy-on-read flag is actually a reference count so multiple users may
- * use the feature without worrying about clobbering its previous state.
- * Copy-on-read stays enabled until all users have called to disable it.
- */
-void bdrv_enable_copy_on_read(BlockDriverState *bs)
-{
- bs->copy_on_read++;
-}
-
-void bdrv_disable_copy_on_read(BlockDriverState *bs)
-{
- assert(bs->copy_on_read > 0);
- bs->copy_on_read--;
-}
-
/*
* Returns the flags that a temporary snapshot should get, based on the
* originally requested flags (the originally requested image will have flags
@@ -1926,55 +1765,6 @@ void bdrv_close_all(void)
}
}
-/* Check if any requests are in-flight (including throttled requests) */
-static bool bdrv_requests_pending(BlockDriverState *bs)
-{
- if (!QLIST_EMPTY(&bs->tracked_requests)) {
- return true;
- }
- if (!qemu_co_queue_empty(&bs->throttled_reqs[0])) {
- return true;
- }
- if (!qemu_co_queue_empty(&bs->throttled_reqs[1])) {
- return true;
- }
- if (bs->file && bdrv_requests_pending(bs->file)) {
- return true;
- }
- if (bs->backing_hd && bdrv_requests_pending(bs->backing_hd)) {
- return true;
- }
- return false;
-}
-
-static bool bdrv_drain_one(BlockDriverState *bs)
-{
- bool bs_busy;
-
- bdrv_flush_io_queue(bs);
- bdrv_start_throttled_reqs(bs);
- bs_busy = bdrv_requests_pending(bs);
- bs_busy |= aio_poll(bdrv_get_aio_context(bs), bs_busy);
- return bs_busy;
-}
-
-/*
- * Wait for pending requests to complete on a single BlockDriverState subtree
- *
- * See the warning in bdrv_drain_all(). This function can only be called if
- * you are sure nothing can generate I/O because you have op blockers
- * installed.
- *
- * Note that unlike bdrv_drain_all(), the caller must hold the BlockDriverState
- * AioContext.
- */
-void bdrv_drain(BlockDriverState *bs)
-{
- while (bdrv_drain_one(bs)) {
- /* Keep iterating */
- }
-}
-
/*
* Wait for pending requests to complete across all BlockDriverStates
*
@@ -2327,152 +2117,6 @@ int bdrv_commit_all(void)
return 0;
}
-/**
- * Remove an active request from the tracked requests list
- *
- * This function should be called when a tracked request is completing.
- */
-static void tracked_request_end(BdrvTrackedRequest *req)
-{
- if (req->serialising) {
- req->bs->serialising_in_flight--;
- }
-
- QLIST_REMOVE(req, list);
- qemu_co_queue_restart_all(&req->wait_queue);
-}
-
-/**
- * Add an active request to the tracked requests list
- */
-static void tracked_request_begin(BdrvTrackedRequest *req,
- BlockDriverState *bs,
- int64_t offset,
- unsigned int bytes, bool is_write)
-{
- *req = (BdrvTrackedRequest){
- .bs = bs,
- .offset = offset,
- .bytes = bytes,
- .is_write = is_write,
- .co = qemu_coroutine_self(),
- .serialising = false,
- .overlap_offset = offset,
- .overlap_bytes = bytes,
- };
-
- qemu_co_queue_init(&req->wait_queue);
-
- QLIST_INSERT_HEAD(&bs->tracked_requests, req, list);
-}
-
-static void mark_request_serialising(BdrvTrackedRequest *req, uint64_t align)
-{
- int64_t overlap_offset = req->offset & ~(align - 1);
- unsigned int overlap_bytes = ROUND_UP(req->offset + req->bytes, align)
- - overlap_offset;
-
- if (!req->serialising) {
- req->bs->serialising_in_flight++;
- req->serialising = true;
- }
-
- req->overlap_offset = MIN(req->overlap_offset, overlap_offset);
- req->overlap_bytes = MAX(req->overlap_bytes, overlap_bytes);
-}
-
-/**
- * Round a region to cluster boundaries
- */
-void bdrv_round_to_clusters(BlockDriverState *bs,
- int64_t sector_num, int nb_sectors,
- int64_t *cluster_sector_num,
- int *cluster_nb_sectors)
-{
- BlockDriverInfo bdi;
-
- if (bdrv_get_info(bs, &bdi) < 0 || bdi.cluster_size == 0) {
- *cluster_sector_num = sector_num;
- *cluster_nb_sectors = nb_sectors;
- } else {
- int64_t c = bdi.cluster_size / BDRV_SECTOR_SIZE;
- *cluster_sector_num = QEMU_ALIGN_DOWN(sector_num, c);
- *cluster_nb_sectors = QEMU_ALIGN_UP(sector_num - *cluster_sector_num +
- nb_sectors, c);
- }
-}
-
-static int bdrv_get_cluster_size(BlockDriverState *bs)
-{
- BlockDriverInfo bdi;
- int ret;
-
- ret = bdrv_get_info(bs, &bdi);
- if (ret < 0 || bdi.cluster_size == 0) {
- return bs->request_alignment;
- } else {
- return bdi.cluster_size;
- }
-}
-
-static bool tracked_request_overlaps(BdrvTrackedRequest *req,
- int64_t offset, unsigned int bytes)
-{
- /* aaaa bbbb */
- if (offset >= req->overlap_offset + req->overlap_bytes) {
- return false;
- }
- /* bbbb aaaa */
- if (req->overlap_offset >= offset + bytes) {
- return false;
- }
- return true;
-}
-
-static bool coroutine_fn wait_serialising_requests(BdrvTrackedRequest *self)
-{
- BlockDriverState *bs = self->bs;
- BdrvTrackedRequest *req;
- bool retry;
- bool waited = false;
-
- if (!bs->serialising_in_flight) {
- return false;
- }
-
- do {
- retry = false;
- QLIST_FOREACH(req, &bs->tracked_requests, list) {
- if (req == self || (!req->serialising && !self->serialising)) {
- continue;
- }
- if (tracked_request_overlaps(req, self->overlap_offset,
- self->overlap_bytes))
- {
- /* Hitting this means there was a reentrant request, for
- * example, a block driver issuing nested requests. This must
- * never happen since it means deadlock.
- */
- assert(qemu_coroutine_self() != req->co);
-
- /* If the request is already (indirectly) waiting for us, or
- * will wait for us as soon as it wakes up, then just go on
- * (instead of producing a deadlock in the former case). */
- if (!req->waiting_for) {
- self->waiting_for = req;
- qemu_co_queue_wait(&req->wait_queue);
- self->waiting_for = NULL;
- retry = true;
- waited = true;
- break;
- }
- }
- }
- } while (retry);
-
- return waited;
-}
-
/*
* Return values:
* 0 - success
@@ -2641,172 +2285,6 @@ exit:
return ret;
}
-
-static int bdrv_check_byte_request(BlockDriverState *bs, int64_t offset,
- size_t size)
-{
- int64_t len;
-
- if (size > INT_MAX) {
- return -EIO;
- }
-
- if (!bdrv_is_inserted(bs))
- return -ENOMEDIUM;
-
- if (bs->growable)
- return 0;
-
- len = bdrv_getlength(bs);
-
- if (offset < 0)
- return -EIO;
-
- if ((offset > len) || (len - offset < size))
- return -EIO;
-
- return 0;
-}
-
-static int bdrv_check_request(BlockDriverState *bs, int64_t sector_num,
- int nb_sectors)
-{
- if (nb_sectors < 0 || nb_sectors > INT_MAX / BDRV_SECTOR_SIZE) {
- return -EIO;
- }
-
- return bdrv_check_byte_request(bs, sector_num * BDRV_SECTOR_SIZE,
- nb_sectors * BDRV_SECTOR_SIZE);
-}
-
-typedef struct RwCo {
- BlockDriverState *bs;
- int64_t offset;
- QEMUIOVector *qiov;
- bool is_write;
- int ret;
- BdrvRequestFlags flags;
-} RwCo;
-
-static void coroutine_fn bdrv_rw_co_entry(void *opaque)
-{
- RwCo *rwco = opaque;
-
- if (!rwco->is_write) {
- rwco->ret = bdrv_co_do_preadv(rwco->bs, rwco->offset,
- rwco->qiov->size, rwco->qiov,
- rwco->flags);
- } else {
- rwco->ret = bdrv_co_do_pwritev(rwco->bs, rwco->offset,
- rwco->qiov->size, rwco->qiov,
- rwco->flags);
- }
-}
-
-/*
- * Process a vectored synchronous request using coroutines
- */
-static int bdrv_prwv_co(BlockDriverState *bs, int64_t offset,
- QEMUIOVector *qiov, bool is_write,
- BdrvRequestFlags flags)
-{
- Coroutine *co;
- RwCo rwco = {
- .bs = bs,
- .offset = offset,
- .qiov = qiov,
- .is_write = is_write,
- .ret = NOT_DONE,
- .flags = flags,
- };
-
- /**
- * In sync call context, when the vcpu is blocked, this throttling timer
- * will not fire; so the I/O throttling function has to be disabled here
- * if it has been enabled.
- */
- if (bs->io_limits_enabled) {
- fprintf(stderr, "Disabling I/O throttling on '%s' due "
- "to synchronous I/O.\n", bdrv_get_device_name(bs));
- bdrv_io_limits_disable(bs);
- }
-
- if (qemu_in_coroutine()) {
- /* Fast-path if already in coroutine context */
- bdrv_rw_co_entry(&rwco);
- } else {
- AioContext *aio_context = bdrv_get_aio_context(bs);
-
- co = qemu_coroutine_create(bdrv_rw_co_entry);
- qemu_coroutine_enter(co, &rwco);
- while (rwco.ret == NOT_DONE) {
- aio_poll(aio_context, true);
- }
- }
- return rwco.ret;
-}
-
-/*
- * Process a synchronous request using coroutines
- */
-static int bdrv_rw_co(BlockDriverState *bs, int64_t sector_num, uint8_t *buf,
- int nb_sectors, bool is_write, BdrvRequestFlags flags)
-{
- QEMUIOVector qiov;
- struct iovec iov = {
- .iov_base = (void *)buf,
- .iov_len = nb_sectors * BDRV_SECTOR_SIZE,
- };
-
- if (nb_sectors < 0 || nb_sectors > INT_MAX / BDRV_SECTOR_SIZE) {
- return -EINVAL;
- }
-
- qemu_iovec_init_external(&qiov, &iov, 1);
- return bdrv_prwv_co(bs, sector_num << BDRV_SECTOR_BITS,
- &qiov, is_write, flags);
-}
-
-/* return < 0 if error. See bdrv_write() for the return codes */
-int bdrv_read(BlockDriverState *bs, int64_t sector_num,
- uint8_t *buf, int nb_sectors)
-{
- return bdrv_rw_co(bs, sector_num, buf, nb_sectors, false, 0);
-}
-
-/* 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, sector_num, buf, nb_sectors);
- bs->io_limits_enabled = enabled;
- return ret;
-}
-
-/* Return < 0 if error. Important errors are:
- -EIO generic I/O error (may happen for all errors)
- -ENOMEDIUM No media inserted.
- -EINVAL Invalid sector number or nb_sectors
- -EACCES Trying to write a read-only device
-*/
-int bdrv_write(BlockDriverState *bs, int64_t sector_num,
- const uint8_t *buf, int nb_sectors)
-{
- return bdrv_rw_co(bs, sector_num, (uint8_t *)buf, nb_sectors, true, 0);
-}
-
-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 | 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
@@ -2853,644 +2331,6 @@ int bdrv_make_zero(BlockDriverState *bs,
BdrvRequestFlags flags)
}
}
-int bdrv_pread(BlockDriverState *bs, int64_t offset, void *buf, int bytes)
-{
- QEMUIOVector qiov;
- struct iovec iov = {
- .iov_base = (void *)buf,
- .iov_len = bytes,
- };
- int ret;
-
- if (bytes < 0) {
- return -EINVAL;
- }
-
- qemu_iovec_init_external(&qiov, &iov, 1);
- ret = bdrv_prwv_co(bs, offset, &qiov, false, 0);
- if (ret < 0) {
- return ret;
- }
-
- return bytes;
-}
-
-int bdrv_pwritev(BlockDriverState *bs, int64_t offset, QEMUIOVector *qiov)
-{
- int ret;
-
- ret = bdrv_prwv_co(bs, offset, qiov, true, 0);
- if (ret < 0) {
- return ret;
- }
-
- return qiov->size;
-}
-
-int bdrv_pwrite(BlockDriverState *bs, int64_t offset,
- const void *buf, int bytes)
-{
- QEMUIOVector qiov;
- struct iovec iov = {
- .iov_base = (void *) buf,
- .iov_len = bytes,
- };
-
- if (bytes < 0) {
- return -EINVAL;
- }
-
- qemu_iovec_init_external(&qiov, &iov, 1);
- return bdrv_pwritev(bs, offset, &qiov);
-}
-
-/*
- * Writes to the file and ensures that no writes are reordered across this
- * request (acts as a barrier)
- *
- * Returns 0 on success, -errno in error cases.
- */
-int bdrv_pwrite_sync(BlockDriverState *bs, int64_t offset,
- const void *buf, int count)
-{
- int ret;
-
- ret = bdrv_pwrite(bs, offset, buf, count);
- if (ret < 0) {
- return ret;
- }
-
- /* No flush needed for cache modes that already do it */
- if (bs->enable_write_cache) {
- bdrv_flush(bs);
- }
-
- return 0;
-}
-
-static int coroutine_fn bdrv_co_do_copy_on_readv(BlockDriverState *bs,
- int64_t sector_num, int nb_sectors, QEMUIOVector *qiov)
-{
- /* Perform I/O through a temporary buffer so that users who scribble over
- * their read buffer while the operation is in progress do not end up
- * modifying the image file. This is critical for zero-copy guest I/O
- * where anything might happen inside guest memory.
- */
- void *bounce_buffer;
-
- BlockDriver *drv = bs->drv;
- struct iovec iov;
- QEMUIOVector bounce_qiov;
- int64_t cluster_sector_num;
- int cluster_nb_sectors;
- size_t skip_bytes;
- int ret;
-
- /* Cover entire cluster so no additional backing file I/O is required when
- * allocating cluster in the image file.
- */
- bdrv_round_to_clusters(bs, sector_num, nb_sectors,
- &cluster_sector_num, &cluster_nb_sectors);
-
- trace_bdrv_co_do_copy_on_readv(bs, sector_num, nb_sectors,
- cluster_sector_num, cluster_nb_sectors);
-
- iov.iov_len = cluster_nb_sectors * BDRV_SECTOR_SIZE;
- iov.iov_base = bounce_buffer = qemu_try_blockalign(bs, iov.iov_len);
- if (bounce_buffer == NULL) {
- ret = -ENOMEM;
- goto err;
- }
-
- qemu_iovec_init_external(&bounce_qiov, &iov, 1);
-
- ret = drv->bdrv_co_readv(bs, cluster_sector_num, cluster_nb_sectors,
- &bounce_qiov);
- if (ret < 0) {
- goto err;
- }
-
- 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, 0);
- } else {
- /* This does not change the data on the disk, it is not necessary
- * to flush even in cache=writethrough mode.
- */
- ret = drv->bdrv_co_writev(bs, cluster_sector_num, cluster_nb_sectors,
- &bounce_qiov);
- }
-
- if (ret < 0) {
- /* It might be okay to ignore write errors for guest requests. If this
- * is a deliberate copy-on-read then we don't want to ignore the error.
- * Simply report it in all cases.
- */
- goto err;
- }
-
- skip_bytes = (sector_num - cluster_sector_num) * BDRV_SECTOR_SIZE;
- qemu_iovec_from_buf(qiov, 0, bounce_buffer + skip_bytes,
- nb_sectors * BDRV_SECTOR_SIZE);
-
-err:
- qemu_vfree(bounce_buffer);
- return ret;
-}
-
-/*
- * Forwards an already correctly aligned request to the BlockDriver. This
- * handles copy on read and zeroing after EOF; any other features must be
- * implemented by the caller.
- */
-static int coroutine_fn bdrv_aligned_preadv(BlockDriverState *bs,
- BdrvTrackedRequest *req, int64_t offset, unsigned int bytes,
- int64_t align, QEMUIOVector *qiov, int flags)
-{
- BlockDriver *drv = bs->drv;
- int ret;
-
- int64_t sector_num = offset >> BDRV_SECTOR_BITS;
- unsigned int nb_sectors = bytes >> BDRV_SECTOR_BITS;
-
- assert((offset & (BDRV_SECTOR_SIZE - 1)) == 0);
- assert((bytes & (BDRV_SECTOR_SIZE - 1)) == 0);
- assert(!qiov || bytes == qiov->size);
-
- /* Handle Copy on Read and associated serialisation */
- if (flags & BDRV_REQ_COPY_ON_READ) {
- /* If we touch the same cluster it counts as an overlap. This
- * guarantees that allocating writes will be serialized and not race
- * with each other for the same cluster. For example, in copy-on-read
- * it ensures that the CoR read and write operations are atomic and
- * guest writes cannot interleave between them. */
- mark_request_serialising(req, bdrv_get_cluster_size(bs));
- }
-
- wait_serialising_requests(req);
-
- if (flags & BDRV_REQ_COPY_ON_READ) {
- int pnum;
-
- ret = bdrv_is_allocated(bs, sector_num, nb_sectors, &pnum);
- if (ret < 0) {
- goto out;
- }
-
- if (!ret || pnum != nb_sectors) {
- ret = bdrv_co_do_copy_on_readv(bs, sector_num, nb_sectors, qiov);
- goto out;
- }
- }
-
- /* Forward the request to the BlockDriver */
- if (!(bs->zero_beyond_eof && bs->growable)) {
- ret = drv->bdrv_co_readv(bs, sector_num, nb_sectors, qiov);
- } else {
- /* Read zeros after EOF of growable BDSes */
- int64_t total_sectors, max_nb_sectors;
-
- total_sectors = bdrv_nb_sectors(bs);
- if (total_sectors < 0) {
- ret = total_sectors;
- goto out;
- }
-
- max_nb_sectors = ROUND_UP(MAX(0, total_sectors - sector_num),
- align >> BDRV_SECTOR_BITS);
- if (nb_sectors < max_nb_sectors) {
- ret = drv->bdrv_co_readv(bs, sector_num, nb_sectors, qiov);
- } else if (max_nb_sectors > 0) {
- QEMUIOVector local_qiov;
-
- qemu_iovec_init(&local_qiov, qiov->niov);
- qemu_iovec_concat(&local_qiov, qiov, 0,
- max_nb_sectors * BDRV_SECTOR_SIZE);
-
- ret = drv->bdrv_co_readv(bs, sector_num, max_nb_sectors,
- &local_qiov);
-
- qemu_iovec_destroy(&local_qiov);
- } else {
- ret = 0;
- }
-
- /* Reading beyond end of file is supposed to produce zeroes */
- if (ret == 0 && total_sectors < sector_num + nb_sectors) {
- uint64_t offset = MAX(0, total_sectors - sector_num);
- uint64_t bytes = (sector_num + nb_sectors - offset) *
- BDRV_SECTOR_SIZE;
- qemu_iovec_memset(qiov, offset * BDRV_SECTOR_SIZE, 0, bytes);
- }
- }
-
-out:
- return ret;
-}
-
-/*
- * Handle a read request in coroutine context
- */
-static int coroutine_fn bdrv_co_do_preadv(BlockDriverState *bs,
- int64_t offset, unsigned int bytes, QEMUIOVector *qiov,
- BdrvRequestFlags flags)
-{
- BlockDriver *drv = bs->drv;
- BdrvTrackedRequest req;
-
- /* TODO Lift BDRV_SECTOR_SIZE restriction in BlockDriver interface */
- uint64_t align = MAX(BDRV_SECTOR_SIZE, bs->request_alignment);
- uint8_t *head_buf = NULL;
- uint8_t *tail_buf = NULL;
- QEMUIOVector local_qiov;
- bool use_local_qiov = false;
- int ret;
-
- if (!drv) {
- return -ENOMEDIUM;
- }
- if (bdrv_check_byte_request(bs, offset, bytes)) {
- return -EIO;
- }
-
- if (bs->copy_on_read) {
- flags |= BDRV_REQ_COPY_ON_READ;
- }
-
- /* throttling disk I/O */
- if (bs->io_limits_enabled) {
- bdrv_io_limits_intercept(bs, bytes, false);
- }
-
- /* Align read if necessary by padding qiov */
- if (offset & (align - 1)) {
- head_buf = qemu_blockalign(bs, align);
- qemu_iovec_init(&local_qiov, qiov->niov + 2);
- qemu_iovec_add(&local_qiov, head_buf, offset & (align - 1));
- qemu_iovec_concat(&local_qiov, qiov, 0, qiov->size);
- use_local_qiov = true;
-
- bytes += offset & (align - 1);
- offset = offset & ~(align - 1);
- }
-
- if ((offset + bytes) & (align - 1)) {
- if (!use_local_qiov) {
- qemu_iovec_init(&local_qiov, qiov->niov + 1);
- qemu_iovec_concat(&local_qiov, qiov, 0, qiov->size);
- use_local_qiov = true;
- }
- tail_buf = qemu_blockalign(bs, align);
- qemu_iovec_add(&local_qiov, tail_buf,
- align - ((offset + bytes) & (align - 1)));
-
- bytes = ROUND_UP(bytes, align);
- }
-
- tracked_request_begin(&req, bs, offset, bytes, false);
- ret = bdrv_aligned_preadv(bs, &req, offset, bytes, align,
- use_local_qiov ? &local_qiov : qiov,
- flags);
- tracked_request_end(&req);
-
- if (use_local_qiov) {
- qemu_iovec_destroy(&local_qiov);
- qemu_vfree(head_buf);
- qemu_vfree(tail_buf);
- }
-
- return ret;
-}
-
-static int coroutine_fn bdrv_co_do_readv(BlockDriverState *bs,
- int64_t sector_num, int nb_sectors, QEMUIOVector *qiov,
- BdrvRequestFlags flags)
-{
- if (nb_sectors < 0 || nb_sectors > (UINT_MAX >> BDRV_SECTOR_BITS)) {
- return -EINVAL;
- }
-
- return bdrv_co_do_preadv(bs, sector_num << BDRV_SECTOR_BITS,
- nb_sectors << BDRV_SECTOR_BITS, qiov, flags);
-}
-
-int coroutine_fn bdrv_co_readv(BlockDriverState *bs, int64_t sector_num,
- int nb_sectors, QEMUIOVector *qiov)
-{
- trace_bdrv_co_readv(bs, sector_num, nb_sectors);
-
- return bdrv_co_do_readv(bs, sector_num, nb_sectors, qiov, 0);
-}
-
-int coroutine_fn bdrv_co_copy_on_readv(BlockDriverState *bs,
- int64_t sector_num, int nb_sectors, QEMUIOVector *qiov)
-{
- trace_bdrv_co_copy_on_readv(bs, sector_num, nb_sectors);
-
- return bdrv_co_do_readv(bs, sector_num, nb_sectors, qiov,
- 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, BdrvRequestFlags flags)
-{
- BlockDriver *drv = bs->drv;
- QEMUIOVector qiov;
- struct iovec iov = {0};
- int ret = 0;
-
- int max_write_zeroes = bs->bl.max_write_zeroes ?
- bs->bl.max_write_zeroes : MAX_WRITE_ZEROES_DEFAULT;
-
- 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 */
- int max_xfer_len = MIN_NON_ZERO(bs->bl.max_transfer_length,
- MAX_WRITE_ZEROES_DEFAULT);
- num = MIN(num, max_xfer_len);
- iov.iov_len = num * BDRV_SECTOR_SIZE;
- if (iov.iov_base == NULL) {
- iov.iov_base = qemu_try_blockalign(bs, num * BDRV_SECTOR_SIZE);
- if (iov.iov_base == NULL) {
- ret = -ENOMEM;
- goto fail;
- }
- 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_xfer_len) {
- qemu_vfree(iov.iov_base);
- iov.iov_base = NULL;
- }
- }
-
- sector_num += num;
- nb_sectors -= num;
- }
-
-fail:
- qemu_vfree(iov.iov_base);
- return ret;
-}
-
-/*
- * Forwards an already correctly aligned write request to the BlockDriver.
- */
-static int coroutine_fn bdrv_aligned_pwritev(BlockDriverState *bs,
- BdrvTrackedRequest *req, int64_t offset, unsigned int bytes,
- QEMUIOVector *qiov, int flags)
-{
- BlockDriver *drv = bs->drv;
- bool waited;
- int ret;
-
- int64_t sector_num = offset >> BDRV_SECTOR_BITS;
- unsigned int nb_sectors = bytes >> BDRV_SECTOR_BITS;
-
- assert((offset & (BDRV_SECTOR_SIZE - 1)) == 0);
- assert((bytes & (BDRV_SECTOR_SIZE - 1)) == 0);
- assert(!qiov || bytes == qiov->size);
-
- waited = wait_serialising_requests(req);
- assert(!waited || !req->serialising);
- assert(req->overlap_offset <= offset);
- assert(offset + bytes <= req->overlap_offset + req->overlap_bytes);
-
- ret = notifier_with_return_list_notify(&bs->before_write_notifiers, req);
-
- if (!ret && bs->detect_zeroes != BLOCKDEV_DETECT_ZEROES_OPTIONS_OFF &&
- !(flags & BDRV_REQ_ZERO_WRITE) && drv->bdrv_co_write_zeroes &&
- qemu_iovec_is_zero(qiov)) {
- flags |= BDRV_REQ_ZERO_WRITE;
- if (bs->detect_zeroes == BLOCKDEV_DETECT_ZEROES_OPTIONS_UNMAP) {
- flags |= BDRV_REQ_MAY_UNMAP;
- }
- }
-
- if (ret < 0) {
- /* Do nothing, write notifier decided to fail this request */
- } else if (flags & BDRV_REQ_ZERO_WRITE) {
- BLKDBG_EVENT(bs, BLKDBG_PWRITEV_ZERO);
- ret = bdrv_co_do_write_zeroes(bs, sector_num, nb_sectors, flags);
- } else {
- BLKDBG_EVENT(bs, BLKDBG_PWRITEV);
- ret = drv->bdrv_co_writev(bs, sector_num, nb_sectors, qiov);
- }
- BLKDBG_EVENT(bs, BLKDBG_PWRITEV_DONE);
-
- if (ret == 0 && !bs->enable_write_cache) {
- ret = bdrv_co_flush(bs);
- }
-
- bdrv_set_dirty(bs, sector_num, nb_sectors);
-
- block_acct_highest_sector(&bs->stats, sector_num, nb_sectors);
-
- if (bs->growable && ret >= 0) {
- bs->total_sectors = MAX(bs->total_sectors, sector_num + nb_sectors);
- }
-
- return ret;
-}
-
-/*
- * Handle a write request in coroutine context
- */
-static int coroutine_fn bdrv_co_do_pwritev(BlockDriverState *bs,
- int64_t offset, unsigned int bytes, QEMUIOVector *qiov,
- BdrvRequestFlags flags)
-{
- BdrvTrackedRequest req;
- /* TODO Lift BDRV_SECTOR_SIZE restriction in BlockDriver interface */
- uint64_t align = MAX(BDRV_SECTOR_SIZE, bs->request_alignment);
- uint8_t *head_buf = NULL;
- uint8_t *tail_buf = NULL;
- QEMUIOVector local_qiov;
- bool use_local_qiov = false;
- int ret;
-
- if (!bs->drv) {
- return -ENOMEDIUM;
- }
- if (bs->read_only) {
- return -EACCES;
- }
- if (bdrv_check_byte_request(bs, offset, bytes)) {
- return -EIO;
- }
-
- /* throttling disk I/O */
- if (bs->io_limits_enabled) {
- bdrv_io_limits_intercept(bs, bytes, true);
- }
-
- /*
- * Align write if necessary by performing a read-modify-write cycle.
- * Pad qiov with the read parts and be sure to have a tracked request not
- * only for bdrv_aligned_pwritev, but also for the reads of the RMW cycle.
- */
- tracked_request_begin(&req, bs, offset, bytes, true);
-
- if (offset & (align - 1)) {
- QEMUIOVector head_qiov;
- struct iovec head_iov;
-
- mark_request_serialising(&req, align);
- wait_serialising_requests(&req);
-
- head_buf = qemu_blockalign(bs, align);
- head_iov = (struct iovec) {
- .iov_base = head_buf,
- .iov_len = align,
- };
- qemu_iovec_init_external(&head_qiov, &head_iov, 1);
-
- BLKDBG_EVENT(bs, BLKDBG_PWRITEV_RMW_HEAD);
- ret = bdrv_aligned_preadv(bs, &req, offset & ~(align - 1), align,
- align, &head_qiov, 0);
- if (ret < 0) {
- goto fail;
- }
- BLKDBG_EVENT(bs, BLKDBG_PWRITEV_RMW_AFTER_HEAD);
-
- qemu_iovec_init(&local_qiov, qiov->niov + 2);
- qemu_iovec_add(&local_qiov, head_buf, offset & (align - 1));
- qemu_iovec_concat(&local_qiov, qiov, 0, qiov->size);
- use_local_qiov = true;
-
- bytes += offset & (align - 1);
- offset = offset & ~(align - 1);
- }
-
- if ((offset + bytes) & (align - 1)) {
- QEMUIOVector tail_qiov;
- struct iovec tail_iov;
- size_t tail_bytes;
- bool waited;
-
- mark_request_serialising(&req, align);
- waited = wait_serialising_requests(&req);
- assert(!waited || !use_local_qiov);
-
- tail_buf = qemu_blockalign(bs, align);
- tail_iov = (struct iovec) {
- .iov_base = tail_buf,
- .iov_len = align,
- };
- qemu_iovec_init_external(&tail_qiov, &tail_iov, 1);
-
- BLKDBG_EVENT(bs, BLKDBG_PWRITEV_RMW_TAIL);
- ret = bdrv_aligned_preadv(bs, &req, (offset + bytes) & ~(align - 1),
align,
- align, &tail_qiov, 0);
- if (ret < 0) {
- goto fail;
- }
- BLKDBG_EVENT(bs, BLKDBG_PWRITEV_RMW_AFTER_TAIL);
-
- if (!use_local_qiov) {
- qemu_iovec_init(&local_qiov, qiov->niov + 1);
- qemu_iovec_concat(&local_qiov, qiov, 0, qiov->size);
- use_local_qiov = true;
- }
-
- tail_bytes = (offset + bytes) & (align - 1);
- qemu_iovec_add(&local_qiov, tail_buf + tail_bytes, align - tail_bytes);
-
- bytes = ROUND_UP(bytes, align);
- }
-
- ret = bdrv_aligned_pwritev(bs, &req, offset, bytes,
- use_local_qiov ? &local_qiov : qiov,
- flags);
-
-fail:
- tracked_request_end(&req);
-
- if (use_local_qiov) {
- qemu_iovec_destroy(&local_qiov);
- }
- qemu_vfree(head_buf);
- qemu_vfree(tail_buf);
-
- return ret;
-}
-
-static int coroutine_fn bdrv_co_do_writev(BlockDriverState *bs,
- int64_t sector_num, int nb_sectors, QEMUIOVector *qiov,
- BdrvRequestFlags flags)
-{
- if (nb_sectors < 0 || nb_sectors > (INT_MAX >> BDRV_SECTOR_BITS)) {
- return -EINVAL;
- }
-
- return bdrv_co_do_pwritev(bs, sector_num << BDRV_SECTOR_BITS,
- nb_sectors << BDRV_SECTOR_BITS, qiov, flags);
-}
-
-int coroutine_fn bdrv_co_writev(BlockDriverState *bs, int64_t sector_num,
- int nb_sectors, QEMUIOVector *qiov)
-{
- trace_bdrv_co_writev(bs, sector_num, nb_sectors);
-
- return bdrv_co_do_writev(bs, sector_num, nb_sectors, qiov, 0);
-}
-
-int coroutine_fn bdrv_co_write_zeroes(BlockDriverState *bs,
- int64_t sector_num, int nb_sectors,
- BdrvRequestFlags flags)
-{
- 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 | flags);
-}
-
/**
* Truncate file to 'offset' bytes (needed only for file protocols)
*/
@@ -4178,22 +3018,6 @@ void bdrv_get_backing_filename(BlockDriverState *bs,
pstrcpy(filename, filename_size, bs->backing_file);
}
-int bdrv_write_compressed(BlockDriverState *bs, int64_t sector_num,
- const uint8_t *buf, int nb_sectors)
-{
- BlockDriver *drv = bs->drv;
- if (!drv)
- return -ENOMEDIUM;
- if (!drv->bdrv_write_compressed)
- return -ENOTSUP;
- if (bdrv_check_request(bs, sector_num, nb_sectors))
- return -EIO;
-
- assert(QLIST_EMPTY(&bs->dirty_bitmaps));
-
- return drv->bdrv_write_compressed(bs, sector_num, buf, nb_sectors);
-}
-
int bdrv_get_info(BlockDriverState *bs, BlockDriverInfo *bdi)
{
BlockDriver *drv = bs->drv;
@@ -4401,450 +3225,6 @@ int bdrv_get_backing_file_depth(BlockDriverState *bs)
return 1 + bdrv_get_backing_file_depth(bs->backing_hd);
}
-/**************************************************************/
-/* async I/Os */
-
-BlockAIOCB *bdrv_aio_readv(BlockDriverState *bs, int64_t sector_num,
- QEMUIOVector *qiov, int nb_sectors,
- BlockCompletionFunc *cb, void *opaque)
-{
- trace_bdrv_aio_readv(bs, sector_num, nb_sectors, opaque);
-
- return bdrv_co_aio_rw_vector(bs, sector_num, qiov, nb_sectors, 0,
- cb, opaque, false);
-}
-
-BlockAIOCB *bdrv_aio_writev(BlockDriverState *bs, int64_t sector_num,
- QEMUIOVector *qiov, int nb_sectors,
- BlockCompletionFunc *cb, void *opaque)
-{
- trace_bdrv_aio_writev(bs, sector_num, nb_sectors, opaque);
-
- return bdrv_co_aio_rw_vector(bs, sector_num, qiov, nb_sectors, 0,
- cb, opaque, true);
-}
-
-BlockAIOCB *bdrv_aio_write_zeroes(BlockDriverState *bs,
- int64_t sector_num, int nb_sectors, BdrvRequestFlags flags,
- BlockCompletionFunc *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);
-}
-
-
-typedef struct MultiwriteCB {
- int error;
- int num_requests;
- int num_callbacks;
- struct {
- BlockCompletionFunc *cb;
- void *opaque;
- QEMUIOVector *free_qiov;
- } callbacks[];
-} MultiwriteCB;
-
-static void multiwrite_user_cb(MultiwriteCB *mcb)
-{
- int i;
-
- for (i = 0; i < mcb->num_callbacks; i++) {
- mcb->callbacks[i].cb(mcb->callbacks[i].opaque, mcb->error);
- if (mcb->callbacks[i].free_qiov) {
- qemu_iovec_destroy(mcb->callbacks[i].free_qiov);
- }
- g_free(mcb->callbacks[i].free_qiov);
- }
-}
-
-static void multiwrite_cb(void *opaque, int ret)
-{
- MultiwriteCB *mcb = opaque;
-
- trace_multiwrite_cb(mcb, ret);
-
- if (ret < 0 && !mcb->error) {
- mcb->error = ret;
- }
-
- mcb->num_requests--;
- if (mcb->num_requests == 0) {
- multiwrite_user_cb(mcb);
- g_free(mcb);
- }
-}
-
-static int multiwrite_req_compare(const void *a, const void *b)
-{
- const BlockRequest *req1 = a, *req2 = b;
-
- /*
- * Note that we can't simply subtract req2->sector from req1->sector
- * here as that could overflow the return value.
- */
- if (req1->sector > req2->sector) {
- return 1;
- } else if (req1->sector < req2->sector) {
- return -1;
- } else {
- return 0;
- }
-}
-
-/*
- * Takes a bunch of requests and tries to merge them. Returns the number of
- * requests that remain after merging.
- */
-static int multiwrite_merge(BlockDriverState *bs, BlockRequest *reqs,
- int num_reqs, MultiwriteCB *mcb)
-{
- int i, outidx;
-
- // Sort requests by start sector
- qsort(reqs, num_reqs, sizeof(*reqs), &multiwrite_req_compare);
-
- // Check if adjacent requests touch the same clusters. If so, combine them,
- // filling up gaps with zero sectors.
- outidx = 0;
- for (i = 1; i < num_reqs; i++) {
- int merge = 0;
- int64_t oldreq_last = reqs[outidx].sector + reqs[outidx].nb_sectors;
-
- // Handle exactly sequential writes and overlapping writes.
- if (reqs[i].sector <= oldreq_last) {
- merge = 1;
- }
-
- if (reqs[outidx].qiov->niov + reqs[i].qiov->niov + 1 > IOV_MAX) {
- merge = 0;
- }
-
- if (bs->bl.max_transfer_length && reqs[outidx].nb_sectors +
- reqs[i].nb_sectors > bs->bl.max_transfer_length) {
- merge = 0;
- }
-
- if (merge) {
- size_t size;
- QEMUIOVector *qiov = g_malloc0(sizeof(*qiov));
- qemu_iovec_init(qiov,
- reqs[outidx].qiov->niov + reqs[i].qiov->niov + 1);
-
- // Add the first request to the merged one. If the requests are
- // overlapping, drop the last sectors of the first request.
- size = (reqs[i].sector - reqs[outidx].sector) << 9;
- qemu_iovec_concat(qiov, reqs[outidx].qiov, 0, size);
-
- // We should need to add any zeros between the two requests
- assert (reqs[i].sector <= oldreq_last);
-
- // Add the second request
- qemu_iovec_concat(qiov, reqs[i].qiov, 0, reqs[i].qiov->size);
-
- // Add tail of first request, if necessary
- if (qiov->size < reqs[outidx].qiov->size) {
- qemu_iovec_concat(qiov, reqs[outidx].qiov, qiov->size,
- reqs[outidx].qiov->size - qiov->size);
- }
-
- reqs[outidx].nb_sectors = qiov->size >> 9;
- reqs[outidx].qiov = qiov;
-
- mcb->callbacks[i].free_qiov = reqs[outidx].qiov;
- } else {
- outidx++;
- reqs[outidx].sector = reqs[i].sector;
- reqs[outidx].nb_sectors = reqs[i].nb_sectors;
- reqs[outidx].qiov = reqs[i].qiov;
- }
- }
-
- return outidx + 1;
-}
-
-/*
- * Submit multiple AIO write requests at once.
- *
- * On success, the function returns 0 and all requests in the reqs array have
- * been submitted. In error case this function returns -1, and any of the
- * requests may or may not be submitted yet. In particular, this means that the
- * callback will be called for some of the requests, for others it won't. The
- * caller must check the error field of the BlockRequest to wait for the right
- * callbacks (if error != 0, no callback will be called).
- *
- * The implementation may modify the contents of the reqs array, e.g. to merge
- * requests. However, the fields opaque and error are left unmodified as they
- * are used to signal failure for a single request to the caller.
- */
-int bdrv_aio_multiwrite(BlockDriverState *bs, BlockRequest *reqs, int num_reqs)
-{
- MultiwriteCB *mcb;
- int i;
-
- /* don't submit writes if we don't have a medium */
- if (bs->drv == NULL) {
- for (i = 0; i < num_reqs; i++) {
- reqs[i].error = -ENOMEDIUM;
- }
- return -1;
- }
-
- if (num_reqs == 0) {
- return 0;
- }
-
- // Create MultiwriteCB structure
- mcb = g_malloc0(sizeof(*mcb) + num_reqs * sizeof(*mcb->callbacks));
- mcb->num_requests = 0;
- mcb->num_callbacks = num_reqs;
-
- for (i = 0; i < num_reqs; i++) {
- mcb->callbacks[i].cb = reqs[i].cb;
- mcb->callbacks[i].opaque = reqs[i].opaque;
- }
-
- // Check for mergable requests
- num_reqs = multiwrite_merge(bs, reqs, num_reqs, mcb);
-
- trace_bdrv_aio_multiwrite(mcb, mcb->num_callbacks, num_reqs);
-
- /* Run the aio requests. */
- mcb->num_requests = num_reqs;
- for (i = 0; i < num_reqs; i++) {
- 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;
-}
-
-void bdrv_aio_cancel(BlockAIOCB *acb)
-{
- qemu_aio_ref(acb);
- bdrv_aio_cancel_async(acb);
- while (acb->refcnt > 1) {
- if (acb->aiocb_info->get_aio_context) {
- aio_poll(acb->aiocb_info->get_aio_context(acb), true);
- } else if (acb->bs) {
- aio_poll(bdrv_get_aio_context(acb->bs), true);
- } else {
- abort();
- }
- }
- qemu_aio_unref(acb);
-}
-
-/* Async version of aio cancel. The caller is not blocked if the acb implements
- * cancel_async, otherwise we do nothing and let the request normally complete.
- * In either case the completion callback must be called. */
-void bdrv_aio_cancel_async(BlockAIOCB *acb)
-{
- if (acb->aiocb_info->cancel_async) {
- acb->aiocb_info->cancel_async(acb);
- }
-}
-
-/**************************************************************/
-/* async block device emulation */
-
-typedef struct BlockAIOCBSync {
- BlockAIOCB common;
- QEMUBH *bh;
- int ret;
- /* vector translation state */
- QEMUIOVector *qiov;
- uint8_t *bounce;
- int is_write;
-} BlockAIOCBSync;
-
-static const AIOCBInfo bdrv_em_aiocb_info = {
- .aiocb_size = sizeof(BlockAIOCBSync),
-};
-
-static void bdrv_aio_bh_cb(void *opaque)
-{
- BlockAIOCBSync *acb = opaque;
-
- if (!acb->is_write && acb->ret >= 0) {
- qemu_iovec_from_buf(acb->qiov, 0, acb->bounce, acb->qiov->size);
- }
- qemu_vfree(acb->bounce);
- acb->common.cb(acb->common.opaque, acb->ret);
- qemu_bh_delete(acb->bh);
- acb->bh = NULL;
- qemu_aio_unref(acb);
-}
-
-static BlockAIOCB *bdrv_aio_rw_vector(BlockDriverState *bs,
- int64_t sector_num,
- QEMUIOVector *qiov,
- int nb_sectors,
- BlockCompletionFunc *cb,
- void *opaque,
- int is_write)
-
-{
- BlockAIOCBSync *acb;
-
- acb = qemu_aio_get(&bdrv_em_aiocb_info, bs, cb, opaque);
- acb->is_write = is_write;
- acb->qiov = qiov;
- acb->bounce = qemu_try_blockalign(bs, qiov->size);
- acb->bh = aio_bh_new(bdrv_get_aio_context(bs), bdrv_aio_bh_cb, acb);
-
- if (acb->bounce == NULL) {
- acb->ret = -ENOMEM;
- } else if (is_write) {
- qemu_iovec_to_buf(acb->qiov, 0, acb->bounce, qiov->size);
- acb->ret = bs->drv->bdrv_write(bs, sector_num, acb->bounce,
nb_sectors);
- } else {
- acb->ret = bs->drv->bdrv_read(bs, sector_num, acb->bounce, nb_sectors);
- }
-
- qemu_bh_schedule(acb->bh);
-
- return &acb->common;
-}
-
-static BlockAIOCB *bdrv_aio_readv_em(BlockDriverState *bs,
- int64_t sector_num, QEMUIOVector *qiov, int nb_sectors,
- BlockCompletionFunc *cb, void *opaque)
-{
- return bdrv_aio_rw_vector(bs, sector_num, qiov, nb_sectors, cb, opaque, 0);
-}
-
-static BlockAIOCB *bdrv_aio_writev_em(BlockDriverState *bs,
- int64_t sector_num, QEMUIOVector *qiov, int nb_sectors,
- BlockCompletionFunc *cb, void *opaque)
-{
- return bdrv_aio_rw_vector(bs, sector_num, qiov, nb_sectors, cb, opaque, 1);
-}
-
-
-typedef struct BlockAIOCBCoroutine {
- BlockAIOCB common;
- BlockRequest req;
- bool is_write;
- bool *done;
- QEMUBH* bh;
-} BlockAIOCBCoroutine;
-
-static const AIOCBInfo bdrv_em_co_aiocb_info = {
- .aiocb_size = sizeof(BlockAIOCBCoroutine),
-};
-
-static void bdrv_co_em_bh(void *opaque)
-{
- BlockAIOCBCoroutine *acb = opaque;
-
- acb->common.cb(acb->common.opaque, acb->req.error);
-
- qemu_bh_delete(acb->bh);
- qemu_aio_unref(acb);
-}
-
-/* Invoke bdrv_co_do_readv/bdrv_co_do_writev */
-static void coroutine_fn bdrv_co_do_rw(void *opaque)
-{
- BlockAIOCBCoroutine *acb = opaque;
- BlockDriverState *bs = acb->common.bs;
-
- if (!acb->is_write) {
- acb->req.error = bdrv_co_do_readv(bs, acb->req.sector,
- 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, acb->req.flags);
- }
-
- acb->bh = aio_bh_new(bdrv_get_aio_context(bs), bdrv_co_em_bh, acb);
- qemu_bh_schedule(acb->bh);
-}
-
-static BlockAIOCB *bdrv_co_aio_rw_vector(BlockDriverState *bs,
- int64_t sector_num,
- QEMUIOVector *qiov,
- int nb_sectors,
- BdrvRequestFlags flags,
- BlockCompletionFunc *cb,
- void *opaque,
- bool is_write)
-{
- Coroutine *co;
- BlockAIOCBCoroutine *acb;
-
- acb = qemu_aio_get(&bdrv_em_co_aiocb_info, bs, cb, opaque);
- acb->req.sector = sector_num;
- acb->req.nb_sectors = nb_sectors;
- acb->req.qiov = qiov;
- acb->req.flags = flags;
- acb->is_write = is_write;
-
- co = qemu_coroutine_create(bdrv_co_do_rw);
- qemu_coroutine_enter(co, acb);
-
- return &acb->common;
-}
-
-static void coroutine_fn bdrv_aio_flush_co_entry(void *opaque)
-{
- BlockAIOCBCoroutine *acb = opaque;
- BlockDriverState *bs = acb->common.bs;
-
- acb->req.error = bdrv_co_flush(bs);
- acb->bh = aio_bh_new(bdrv_get_aio_context(bs), bdrv_co_em_bh, acb);
- qemu_bh_schedule(acb->bh);
-}
-
-BlockAIOCB *bdrv_aio_flush(BlockDriverState *bs,
- BlockCompletionFunc *cb, void *opaque)
-{
- trace_bdrv_aio_flush(bs, opaque);
-
- Coroutine *co;
- BlockAIOCBCoroutine *acb;
-
- acb = qemu_aio_get(&bdrv_em_co_aiocb_info, bs, cb, opaque);
-
- co = qemu_coroutine_create(bdrv_aio_flush_co_entry);
- qemu_coroutine_enter(co, acb);
-
- return &acb->common;
-}
-
-static void coroutine_fn bdrv_aio_discard_co_entry(void *opaque)
-{
- BlockAIOCBCoroutine *acb = opaque;
- BlockDriverState *bs = acb->common.bs;
-
- acb->req.error = bdrv_co_discard(bs, acb->req.sector, acb->req.nb_sectors);
- acb->bh = aio_bh_new(bdrv_get_aio_context(bs), bdrv_co_em_bh, acb);
- qemu_bh_schedule(acb->bh);
-}
-
-BlockAIOCB *bdrv_aio_discard(BlockDriverState *bs,
- int64_t sector_num, int nb_sectors,
- BlockCompletionFunc *cb, void *opaque)
-{
- Coroutine *co;
- BlockAIOCBCoroutine *acb;
-
- trace_bdrv_aio_discard(bs, sector_num, nb_sectors, opaque);
-
- acb = qemu_aio_get(&bdrv_em_co_aiocb_info, bs, cb, opaque);
- acb->req.sector = sector_num;
- acb->req.nb_sectors = nb_sectors;
- co = qemu_coroutine_create(bdrv_aio_discard_co_entry);
- qemu_coroutine_enter(co, acb);
-
- return &acb->common;
-}
-
void bdrv_init(void)
{
module_call_init(MODULE_INIT_BLOCK);
@@ -4856,161 +3236,6 @@ void bdrv_init_with_whitelist(void)
bdrv_init();
}
-void *qemu_aio_get(const AIOCBInfo *aiocb_info, BlockDriverState *bs,
- BlockCompletionFunc *cb, void *opaque)
-{
- BlockAIOCB *acb;
-
- acb = g_slice_alloc(aiocb_info->aiocb_size);
- acb->aiocb_info = aiocb_info;
- acb->bs = bs;
- acb->cb = cb;
- acb->opaque = opaque;
- acb->refcnt = 1;
- return acb;
-}
-
-void qemu_aio_ref(void *p)
-{
- BlockAIOCB *acb = p;
- acb->refcnt++;
-}
-
-void qemu_aio_unref(void *p)
-{
- BlockAIOCB *acb = p;
- assert(acb->refcnt > 0);
- if (--acb->refcnt == 0) {
- g_slice_free1(acb->aiocb_info->aiocb_size, acb);
- }
-}
-
-/**************************************************************/
-/* Coroutine block device emulation */
-
-typedef struct CoroutineIOCompletion {
- Coroutine *coroutine;
- int ret;
-} CoroutineIOCompletion;
-
-static void bdrv_co_io_em_complete(void *opaque, int ret)
-{
- CoroutineIOCompletion *co = opaque;
-
- co->ret = ret;
- qemu_coroutine_enter(co->coroutine, NULL);
-}
-
-static int coroutine_fn bdrv_co_io_em(BlockDriverState *bs, int64_t sector_num,
- int nb_sectors, QEMUIOVector *iov,
- bool is_write)
-{
- CoroutineIOCompletion co = {
- .coroutine = qemu_coroutine_self(),
- };
- BlockAIOCB *acb;
-
- if (is_write) {
- acb = bs->drv->bdrv_aio_writev(bs, sector_num, iov, nb_sectors,
- bdrv_co_io_em_complete, &co);
- } else {
- acb = bs->drv->bdrv_aio_readv(bs, sector_num, iov, nb_sectors,
- bdrv_co_io_em_complete, &co);
- }
-
- trace_bdrv_co_io_em(bs, sector_num, nb_sectors, is_write, acb);
- if (!acb) {
- return -EIO;
- }
- qemu_coroutine_yield();
-
- return co.ret;
-}
-
-static int coroutine_fn bdrv_co_readv_em(BlockDriverState *bs,
- int64_t sector_num, int nb_sectors,
- QEMUIOVector *iov)
-{
- return bdrv_co_io_em(bs, sector_num, nb_sectors, iov, false);
-}
-
-static int coroutine_fn bdrv_co_writev_em(BlockDriverState *bs,
- int64_t sector_num, int nb_sectors,
- QEMUIOVector *iov)
-{
- return bdrv_co_io_em(bs, sector_num, nb_sectors, iov, true);
-}
-
-static void coroutine_fn bdrv_flush_co_entry(void *opaque)
-{
- RwCo *rwco = opaque;
-
- rwco->ret = bdrv_co_flush(rwco->bs);
-}
-
-int coroutine_fn bdrv_co_flush(BlockDriverState *bs)
-{
- int ret;
-
- if (!bs || !bdrv_is_inserted(bs) || bdrv_is_read_only(bs)) {
- return 0;
- }
-
- /* Write back cached data to the OS even with cache=unsafe */
- BLKDBG_EVENT(bs->file, BLKDBG_FLUSH_TO_OS);
- if (bs->drv->bdrv_co_flush_to_os) {
- ret = bs->drv->bdrv_co_flush_to_os(bs);
- if (ret < 0) {
- return ret;
- }
- }
-
- /* But don't actually force it to the disk with cache=unsafe */
- if (bs->open_flags & BDRV_O_NO_FLUSH) {
- goto flush_parent;
- }
-
- BLKDBG_EVENT(bs->file, BLKDBG_FLUSH_TO_DISK);
- if (bs->drv->bdrv_co_flush_to_disk) {
- ret = bs->drv->bdrv_co_flush_to_disk(bs);
- } else if (bs->drv->bdrv_aio_flush) {
- BlockAIOCB *acb;
- CoroutineIOCompletion co = {
- .coroutine = qemu_coroutine_self(),
- };
-
- acb = bs->drv->bdrv_aio_flush(bs, bdrv_co_io_em_complete, &co);
- if (acb == NULL) {
- ret = -EIO;
- } else {
- qemu_coroutine_yield();
- ret = co.ret;
- }
- } else {
- /*
- * Some block drivers always operate in either writethrough or unsafe
- * mode and don't support bdrv_flush therefore. Usually qemu doesn't
- * know how the server works (because the behaviour is hardcoded or
- * depends on server-side configuration), so we can't ensure that
- * everything is safe on disk. Returning an error doesn't work because
- * that would break guests even if the server operates in writethrough
- * mode.
- *
- * Let's hope the user knows what he's doing.
- */
- ret = 0;
- }
- if (ret < 0) {
- return ret;
- }
-
- /* Now flush the underlying protocol. It will also have BDRV_O_NO_FLUSH
- * in the case of cache=unsafe, so there are no useless flushes.
- */
-flush_parent:
- return bdrv_co_flush(bs->file);
-}
-
void bdrv_invalidate_cache(BlockDriverState *bs, Error **errp)
{
Error *local_err = NULL;
@@ -5060,145 +3285,6 @@ void bdrv_invalidate_cache_all(Error **errp)
}
}
-int bdrv_flush(BlockDriverState *bs)
-{
- Coroutine *co;
- RwCo rwco = {
- .bs = bs,
- .ret = NOT_DONE,
- };
-
- if (qemu_in_coroutine()) {
- /* Fast-path if already in coroutine context */
- bdrv_flush_co_entry(&rwco);
- } else {
- AioContext *aio_context = bdrv_get_aio_context(bs);
-
- co = qemu_coroutine_create(bdrv_flush_co_entry);
- qemu_coroutine_enter(co, &rwco);
- while (rwco.ret == NOT_DONE) {
- aio_poll(aio_context, true);
- }
- }
-
- return rwco.ret;
-}
-
-typedef struct DiscardCo {
- BlockDriverState *bs;
- int64_t sector_num;
- int nb_sectors;
- int ret;
-} DiscardCo;
-static void coroutine_fn bdrv_discard_co_entry(void *opaque)
-{
- DiscardCo *rwco = 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)) {
- return -EIO;
- } else if (bs->read_only) {
- return -EROFS;
- }
-
- 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 && !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 {
- BlockAIOCB *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)
-{
- Coroutine *co;
- DiscardCo rwco = {
- .bs = bs,
- .sector_num = sector_num,
- .nb_sectors = nb_sectors,
- .ret = NOT_DONE,
- };
-
- if (qemu_in_coroutine()) {
- /* Fast-path if already in coroutine context */
- bdrv_discard_co_entry(&rwco);
- } else {
- AioContext *aio_context = bdrv_get_aio_context(bs);
-
- co = qemu_coroutine_create(bdrv_discard_co_entry);
- qemu_coroutine_enter(co, &rwco);
- while (rwco.ret == NOT_DONE) {
- aio_poll(aio_context, true);
- }
- }
-
- return rwco.ret;
-}
-
/**************************************************************/
/* removable device support */
@@ -5264,28 +3350,6 @@ void bdrv_lock_medium(BlockDriverState *bs, bool locked)
}
}
-/* needed for generic scsi interface */
-
-int bdrv_ioctl(BlockDriverState *bs, unsigned long int req, void *buf)
-{
- BlockDriver *drv = bs->drv;
-
- if (drv && drv->bdrv_ioctl)
- return drv->bdrv_ioctl(bs, req, buf);
- return -ENOTSUP;
-}
-
-BlockAIOCB *bdrv_aio_ioctl(BlockDriverState *bs,
- unsigned long int req, void *buf,
- BlockCompletionFunc *cb, void *opaque)
-{
- BlockDriver *drv = bs->drv;
-
- if (drv && drv->bdrv_aio_ioctl)
- return drv->bdrv_aio_ioctl(bs, req, buf, cb, opaque);
- return NULL;
-}
-
void bdrv_set_guest_block_size(BlockDriverState *bs, int align)
{
bs->guest_block_size = align;
@@ -5427,8 +3491,7 @@ void bdrv_reset_dirty_bitmap(BlockDriverState *bs,
BdrvDirtyBitmap *bitmap,
hbitmap_reset(bitmap->bitmap, cur_sector, nr_sectors);
}
-static void bdrv_set_dirty(BlockDriverState *bs, int64_t cur_sector,
- int nr_sectors)
+void bdrv_set_dirty(BlockDriverState *bs, int64_t cur_sector, int nr_sectors)
{
BdrvDirtyBitmap *bitmap;
QLIST_FOREACH(bitmap, &bs->dirty_bitmaps, list) {
@@ -5436,8 +3499,7 @@ static void bdrv_set_dirty(BlockDriverState *bs, int64_t
cur_sector,
}
}
-static void bdrv_reset_dirty(BlockDriverState *bs, int64_t cur_sector,
- int nr_sectors)
+void bdrv_reset_dirty(BlockDriverState *bs, int64_t cur_sector, int nr_sectors)
{
BdrvDirtyBitmap *bitmap;
QLIST_FOREACH(bitmap, &bs->dirty_bitmaps, list) {
@@ -5859,12 +3921,6 @@ void bdrv_remove_aio_context_notifier(BlockDriverState
*bs,
abort();
}
-void bdrv_add_before_write_notifier(BlockDriverState *bs,
- NotifierWithReturn *notifier)
-{
- notifier_with_return_list_add(&bs->before_write_notifiers, notifier);
-}
-
int bdrv_amend_options(BlockDriverState *bs, QemuOpts *opts,
BlockDriverAmendStatusCB *status_cb)
{
@@ -5965,36 +4021,6 @@ out:
return to_replace_bs;
}
-void bdrv_io_plug(BlockDriverState *bs)
-{
- BlockDriver *drv = bs->drv;
- if (drv && drv->bdrv_io_plug) {
- drv->bdrv_io_plug(bs);
- } else if (bs->file) {
- bdrv_io_plug(bs->file);
- }
-}
-
-void bdrv_io_unplug(BlockDriverState *bs)
-{
- BlockDriver *drv = bs->drv;
- if (drv && drv->bdrv_io_unplug) {
- drv->bdrv_io_unplug(bs);
- } else if (bs->file) {
- bdrv_io_unplug(bs->file);
- }
-}
-
-void bdrv_flush_io_queue(BlockDriverState *bs)
-{
- BlockDriver *drv = bs->drv;
- if (drv && drv->bdrv_flush_io_queue) {
- drv->bdrv_flush_io_queue(bs);
- } else if (bs->file) {
- bdrv_flush_io_queue(bs->file);
- }
-}
-
static bool append_open_options(QDict *d, BlockDriverState *bs)
{
const QDictEntry *entry;
diff --git a/block/Makefile.objs b/block/Makefile.objs
index 04b0e43..ba8de70 100644
--- a/block/Makefile.objs
+++ b/block/Makefile.objs
@@ -1,3 +1,4 @@
+block-obj-y += io.o
block-obj-y += raw_bsd.o qcow.o vdi.o vmdk.o cloop.o dmg.o bochs.o vpc.o
vvfat.o
block-obj-y += qcow2.o qcow2-refcount.o qcow2-cluster.o qcow2-snapshot.o
qcow2-cache.o
block-obj-y += qed.o qed-gencb.o qed-l2-cache.o qed-table.o qed-cluster.o
diff --git a/block/io.c b/block/io.c
new file mode 100644
index 0000000..57176ba
--- /dev/null
+++ b/block/io.c
@@ -0,0 +1,1997 @@
+/*
+ * QEMU System Emulator block driver
+ *
+ * Copyright (c) 2003 Fabrice Bellard
+ *
+ * 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 "trace.h"
+#include "block/block_int.h"
+
+#define NOT_DONE 0x7fffffff /* used while emulated sync operation in progress
*/
+
+static BlockAIOCB *bdrv_aio_readv_em(BlockDriverState *bs,
+ int64_t sector_num, QEMUIOVector *qiov, int nb_sectors,
+ BlockCompletionFunc *cb, void *opaque);
+static BlockAIOCB *bdrv_aio_writev_em(BlockDriverState *bs,
+ int64_t sector_num, QEMUIOVector *qiov, int nb_sectors,
+ BlockCompletionFunc *cb, void *opaque);
+static int coroutine_fn bdrv_co_readv_em(BlockDriverState *bs,
+ int64_t sector_num, int nb_sectors,
+ QEMUIOVector *iov);
+static int coroutine_fn bdrv_co_writev_em(BlockDriverState *bs,
+ int64_t sector_num, int nb_sectors,
+ QEMUIOVector *iov);
+static int coroutine_fn bdrv_co_do_preadv(BlockDriverState *bs,
+ int64_t offset, unsigned int bytes, QEMUIOVector *qiov,
+ BdrvRequestFlags flags);
+static int coroutine_fn bdrv_co_do_pwritev(BlockDriverState *bs,
+ int64_t offset, unsigned int bytes, QEMUIOVector *qiov,
+ BdrvRequestFlags flags);
+static BlockAIOCB *bdrv_co_aio_rw_vector(BlockDriverState *bs,
+ int64_t sector_num,
+ QEMUIOVector *qiov,
+ int nb_sectors,
+ BdrvRequestFlags flags,
+ BlockCompletionFunc *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, BdrvRequestFlags flags);
+
+void bdrv_setup_io_funcs(BlockDriver *bdrv)
+{
+ /* Block drivers without coroutine functions need emulation */
+ if (!bdrv->bdrv_co_readv) {
+ bdrv->bdrv_co_readv = bdrv_co_readv_em;
+ bdrv->bdrv_co_writev = bdrv_co_writev_em;
+
+ /* bdrv_co_readv_em()/brdv_co_writev_em() work in terms of aio, so if
+ * the block driver lacks aio we need to emulate that too.
+ */
+ if (!bdrv->bdrv_aio_readv) {
+ /* add AIO emulation layer */
+ bdrv->bdrv_aio_readv = bdrv_aio_readv_em;
+ bdrv->bdrv_aio_writev = bdrv_aio_writev_em;
+ }
+ }
+}
+
+/* throttling disk I/O limits */
+void bdrv_set_io_limits(BlockDriverState *bs,
+ ThrottleConfig *cfg)
+{
+ int i;
+
+ throttle_config(&bs->throttle_state, cfg);
+
+ for (i = 0; i < 2; i++) {
+ qemu_co_enter_next(&bs->throttled_reqs[i]);
+ }
+}
+
+/* this function drain all the throttled IOs */
+static bool bdrv_start_throttled_reqs(BlockDriverState *bs)
+{
+ bool drained = false;
+ bool enabled = bs->io_limits_enabled;
+ int i;
+
+ bs->io_limits_enabled = false;
+
+ for (i = 0; i < 2; i++) {
+ while (qemu_co_enter_next(&bs->throttled_reqs[i])) {
+ drained = true;
+ }
+ }
+
+ bs->io_limits_enabled = enabled;
+
+ return drained;
+}
+
+void bdrv_io_limits_disable(BlockDriverState *bs)
+{
+ bs->io_limits_enabled = false;
+
+ bdrv_start_throttled_reqs(bs);
+
+ throttle_destroy(&bs->throttle_state);
+}
+
+static void bdrv_throttle_read_timer_cb(void *opaque)
+{
+ BlockDriverState *bs = opaque;
+ qemu_co_enter_next(&bs->throttled_reqs[0]);
+}
+
+static void bdrv_throttle_write_timer_cb(void *opaque)
+{
+ BlockDriverState *bs = opaque;
+ qemu_co_enter_next(&bs->throttled_reqs[1]);
+}
+
+/* should be called before bdrv_set_io_limits if a limit is set */
+void bdrv_io_limits_enable(BlockDriverState *bs)
+{
+ assert(!bs->io_limits_enabled);
+ throttle_init(&bs->throttle_state,
+ bdrv_get_aio_context(bs),
+ QEMU_CLOCK_VIRTUAL,
+ bdrv_throttle_read_timer_cb,
+ bdrv_throttle_write_timer_cb,
+ bs);
+ bs->io_limits_enabled = true;
+}
+
+/* This function makes an IO wait if needed
+ *
+ * @nb_sectors: the number of sectors of the IO
+ * @is_write: is the IO a write
+ */
+static void bdrv_io_limits_intercept(BlockDriverState *bs,
+ unsigned int bytes,
+ bool is_write)
+{
+ /* does this io must wait */
+ bool must_wait = throttle_schedule_timer(&bs->throttle_state, is_write);
+
+ /* if must wait or any request of this type throttled queue the IO */
+ if (must_wait ||
+ !qemu_co_queue_empty(&bs->throttled_reqs[is_write])) {
+ qemu_co_queue_wait(&bs->throttled_reqs[is_write]);
+ }
+
+ /* the IO will be executed, do the accounting */
+ throttle_account(&bs->throttle_state, is_write, bytes);
+
+
+ /* if the next request must wait -> do nothing */
+ if (throttle_schedule_timer(&bs->throttle_state, is_write)) {
+ return;
+ }
+
+ /* else queue next request for execution */
+ qemu_co_queue_next(&bs->throttled_reqs[is_write]);
+}
+
+/**
+ * The copy-on-read flag is actually a reference count so multiple users may
+ * use the feature without worrying about clobbering its previous state.
+ * Copy-on-read stays enabled until all users have called to disable it.
+ */
+void bdrv_enable_copy_on_read(BlockDriverState *bs)
+{
+ bs->copy_on_read++;
+}
+
+void bdrv_disable_copy_on_read(BlockDriverState *bs)
+{
+ assert(bs->copy_on_read > 0);
+ bs->copy_on_read--;
+}
+
+/* Check if any requests are in-flight (including throttled requests) */
+static bool bdrv_requests_pending(BlockDriverState *bs)
+{
+ if (!QLIST_EMPTY(&bs->tracked_requests)) {
+ return true;
+ }
+ if (!qemu_co_queue_empty(&bs->throttled_reqs[0])) {
+ return true;
+ }
+ if (!qemu_co_queue_empty(&bs->throttled_reqs[1])) {
+ return true;
+ }
+ if (bs->file && bdrv_requests_pending(bs->file)) {
+ return true;
+ }
+ if (bs->backing_hd && bdrv_requests_pending(bs->backing_hd)) {
+ return true;
+ }
+ return false;
+}
+
+bool bdrv_drain_one(BlockDriverState *bs)
+{
+ bool bs_busy;
+
+ bdrv_flush_io_queue(bs);
+ bdrv_start_throttled_reqs(bs);
+ bs_busy = bdrv_requests_pending(bs);
+ bs_busy |= aio_poll(bdrv_get_aio_context(bs), bs_busy);
+ return bs_busy;
+}
+
+/*
+ * Wait for pending requests to complete on a single BlockDriverState subtree
+ *
+ * See the warning in bdrv_drain_all(). This function can only be called if
+ * you are sure nothing can generate I/O because you have op blockers
+ * installed.
+ *
+ * Note that unlike bdrv_drain_all(), the caller must hold the BlockDriverState
+ * AioContext.
+ */
+void bdrv_drain(BlockDriverState *bs)
+{
+ while (bdrv_drain_one(bs)) {
+ /* Keep iterating */
+ }
+}
+
+/**
+ * Remove an active request from the tracked requests list
+ *
+ * This function should be called when a tracked request is completing.
+ */
+static void tracked_request_end(BdrvTrackedRequest *req)
+{
+ if (req->serialising) {
+ req->bs->serialising_in_flight--;
+ }
+
+ QLIST_REMOVE(req, list);
+ qemu_co_queue_restart_all(&req->wait_queue);
+}
+
+/**
+ * Add an active request to the tracked requests list
+ */
+static void tracked_request_begin(BdrvTrackedRequest *req,
+ BlockDriverState *bs,
+ int64_t offset,
+ unsigned int bytes, bool is_write)
+{
+ *req = (BdrvTrackedRequest){
+ .bs = bs,
+ .offset = offset,
+ .bytes = bytes,
+ .is_write = is_write,
+ .co = qemu_coroutine_self(),
+ .serialising = false,
+ .overlap_offset = offset,
+ .overlap_bytes = bytes,
+ };
+
+ qemu_co_queue_init(&req->wait_queue);
+
+ QLIST_INSERT_HEAD(&bs->tracked_requests, req, list);
+}
+
+static void mark_request_serialising(BdrvTrackedRequest *req, uint64_t align)
+{
+ int64_t overlap_offset = req->offset & ~(align - 1);
+ unsigned int overlap_bytes = ROUND_UP(req->offset + req->bytes, align)
+ - overlap_offset;
+
+ if (!req->serialising) {
+ req->bs->serialising_in_flight++;
+ req->serialising = true;
+ }
+
+ req->overlap_offset = MIN(req->overlap_offset, overlap_offset);
+ req->overlap_bytes = MAX(req->overlap_bytes, overlap_bytes);
+}
+
+/**
+ * Round a region to cluster boundaries
+ */
+void bdrv_round_to_clusters(BlockDriverState *bs,
+ int64_t sector_num, int nb_sectors,
+ int64_t *cluster_sector_num,
+ int *cluster_nb_sectors)
+{
+ BlockDriverInfo bdi;
+
+ if (bdrv_get_info(bs, &bdi) < 0 || bdi.cluster_size == 0) {
+ *cluster_sector_num = sector_num;
+ *cluster_nb_sectors = nb_sectors;
+ } else {
+ int64_t c = bdi.cluster_size / BDRV_SECTOR_SIZE;
+ *cluster_sector_num = QEMU_ALIGN_DOWN(sector_num, c);
+ *cluster_nb_sectors = QEMU_ALIGN_UP(sector_num - *cluster_sector_num +
+ nb_sectors, c);
+ }
+}
+
+static bool tracked_request_overlaps(BdrvTrackedRequest *req,
+ int64_t offset, unsigned int bytes)
+{
+ /* aaaa bbbb */
+ if (offset >= req->overlap_offset + req->overlap_bytes) {
+ return false;
+ }
+ /* bbbb aaaa */
+ if (req->overlap_offset >= offset + bytes) {
+ return false;
+ }
+ return true;
+}
+
+static bool coroutine_fn wait_serialising_requests(BdrvTrackedRequest *self)
+{
+ BlockDriverState *bs = self->bs;
+ BdrvTrackedRequest *req;
+ bool retry;
+ bool waited = false;
+
+ if (!bs->serialising_in_flight) {
+ return false;
+ }
+
+ do {
+ retry = false;
+ QLIST_FOREACH(req, &bs->tracked_requests, list) {
+ if (req == self || (!req->serialising && !self->serialising)) {
+ continue;
+ }
+ if (tracked_request_overlaps(req, self->overlap_offset,
+ self->overlap_bytes))
+ {
+ /* Hitting this means there was a reentrant request, for
+ * example, a block driver issuing nested requests. This must
+ * never happen since it means deadlock.
+ */
+ assert(qemu_coroutine_self() != req->co);
+
+ /* If the request is already (indirectly) waiting for us, or
+ * will wait for us as soon as it wakes up, then just go on
+ * (instead of producing a deadlock in the former case). */
+ if (!req->waiting_for) {
+ self->waiting_for = req;
+ qemu_co_queue_wait(&req->wait_queue);
+ self->waiting_for = NULL;
+ retry = true;
+ waited = true;
+ break;
+ }
+ }
+ }
+ } while (retry);
+
+ return waited;
+}
+
+static int bdrv_check_byte_request(BlockDriverState *bs, int64_t offset,
+ size_t size)
+{
+ int64_t len;
+
+ if (size > INT_MAX) {
+ return -EIO;
+ }
+
+ if (!bdrv_is_inserted(bs))
+ return -ENOMEDIUM;
+
+ if (bs->growable)
+ return 0;
+
+ len = bdrv_getlength(bs);
+
+ if (offset < 0)
+ return -EIO;
+
+ if ((offset > len) || (len - offset < size))
+ return -EIO;
+
+ return 0;
+}
+
+static int bdrv_check_request(BlockDriverState *bs, int64_t sector_num,
+ int nb_sectors)
+{
+ if (nb_sectors < 0 || nb_sectors > INT_MAX / BDRV_SECTOR_SIZE) {
+ return -EIO;
+ }
+
+ return bdrv_check_byte_request(bs, sector_num * BDRV_SECTOR_SIZE,
+ nb_sectors * BDRV_SECTOR_SIZE);
+}
+
+typedef struct RwCo {
+ BlockDriverState *bs;
+ int64_t offset;
+ QEMUIOVector *qiov;
+ bool is_write;
+ int ret;
+ BdrvRequestFlags flags;
+} RwCo;
+
+static void coroutine_fn bdrv_rw_co_entry(void *opaque)
+{
+ RwCo *rwco = opaque;
+
+ if (!rwco->is_write) {
+ rwco->ret = bdrv_co_do_preadv(rwco->bs, rwco->offset,
+ rwco->qiov->size, rwco->qiov,
+ rwco->flags);
+ } else {
+ rwco->ret = bdrv_co_do_pwritev(rwco->bs, rwco->offset,
+ rwco->qiov->size, rwco->qiov,
+ rwco->flags);
+ }
+}
+
+/*
+ * Process a vectored synchronous request using coroutines
+ */
+static int bdrv_prwv_co(BlockDriverState *bs, int64_t offset,
+ QEMUIOVector *qiov, bool is_write,
+ BdrvRequestFlags flags)
+{
+ Coroutine *co;
+ RwCo rwco = {
+ .bs = bs,
+ .offset = offset,
+ .qiov = qiov,
+ .is_write = is_write,
+ .ret = NOT_DONE,
+ .flags = flags,
+ };
+
+ /**
+ * In sync call context, when the vcpu is blocked, this throttling timer
+ * will not fire; so the I/O throttling function has to be disabled here
+ * if it has been enabled.
+ */
+ if (bs->io_limits_enabled) {
+ fprintf(stderr, "Disabling I/O throttling on '%s' due "
+ "to synchronous I/O.\n", bdrv_get_device_name(bs));
+ bdrv_io_limits_disable(bs);
+ }
+
+ if (qemu_in_coroutine()) {
+ /* Fast-path if already in coroutine context */
+ bdrv_rw_co_entry(&rwco);
+ } else {
+ AioContext *aio_context = bdrv_get_aio_context(bs);
+
+ co = qemu_coroutine_create(bdrv_rw_co_entry);
+ qemu_coroutine_enter(co, &rwco);
+ while (rwco.ret == NOT_DONE) {
+ aio_poll(aio_context, true);
+ }
+ }
+ return rwco.ret;
+}
+
+/*
+ * Process a synchronous request using coroutines
+ */
+static int bdrv_rw_co(BlockDriverState *bs, int64_t sector_num, uint8_t *buf,
+ int nb_sectors, bool is_write, BdrvRequestFlags flags)
+{
+ QEMUIOVector qiov;
+ struct iovec iov = {
+ .iov_base = (void *)buf,
+ .iov_len = nb_sectors * BDRV_SECTOR_SIZE,
+ };
+
+ if (nb_sectors < 0 || nb_sectors > INT_MAX / BDRV_SECTOR_SIZE) {
+ return -EINVAL;
+ }
+
+ qemu_iovec_init_external(&qiov, &iov, 1);
+ return bdrv_prwv_co(bs, sector_num << BDRV_SECTOR_BITS,
+ &qiov, is_write, flags);
+}
+
+/* return < 0 if error. See bdrv_write() for the return codes */
+int bdrv_read(BlockDriverState *bs, int64_t sector_num,
+ uint8_t *buf, int nb_sectors)
+{
+ return bdrv_rw_co(bs, sector_num, buf, nb_sectors, false, 0);
+}
+
+/* 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, sector_num, buf, nb_sectors);
+ bs->io_limits_enabled = enabled;
+ return ret;
+}
+
+/* Return < 0 if error. Important errors are:
+ -EIO generic I/O error (may happen for all errors)
+ -ENOMEDIUM No media inserted.
+ -EINVAL Invalid sector number or nb_sectors
+ -EACCES Trying to write a read-only device
+*/
+int bdrv_write(BlockDriverState *bs, int64_t sector_num,
+ const uint8_t *buf, int nb_sectors)
+{
+ return bdrv_rw_co(bs, sector_num, (uint8_t *)buf, nb_sectors, true, 0);
+}
+
+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 | flags);
+}
+
+int bdrv_pread(BlockDriverState *bs, int64_t offset, void *buf, int bytes)
+{
+ QEMUIOVector qiov;
+ struct iovec iov = {
+ .iov_base = (void *)buf,
+ .iov_len = bytes,
+ };
+ int ret;
+
+ if (bytes < 0) {
+ return -EINVAL;
+ }
+
+ qemu_iovec_init_external(&qiov, &iov, 1);
+ ret = bdrv_prwv_co(bs, offset, &qiov, false, 0);
+ if (ret < 0) {
+ return ret;
+ }
+
+ return bytes;
+}
+
+int bdrv_pwritev(BlockDriverState *bs, int64_t offset, QEMUIOVector *qiov)
+{
+ int ret;
+
+ ret = bdrv_prwv_co(bs, offset, qiov, true, 0);
+ if (ret < 0) {
+ return ret;
+ }
+
+ return qiov->size;
+}
+
+int bdrv_pwrite(BlockDriverState *bs, int64_t offset,
+ const void *buf, int bytes)
+{
+ QEMUIOVector qiov;
+ struct iovec iov = {
+ .iov_base = (void *) buf,
+ .iov_len = bytes,
+ };
+
+ if (bytes < 0) {
+ return -EINVAL;
+ }
+
+ qemu_iovec_init_external(&qiov, &iov, 1);
+ return bdrv_pwritev(bs, offset, &qiov);
+}
+
+/*
+ * Writes to the file and ensures that no writes are reordered across this
+ * request (acts as a barrier)
+ *
+ * Returns 0 on success, -errno in error cases.
+ */
+int bdrv_pwrite_sync(BlockDriverState *bs, int64_t offset,
+ const void *buf, int count)
+{
+ int ret;
+
+ ret = bdrv_pwrite(bs, offset, buf, count);
+ if (ret < 0) {
+ return ret;
+ }
+
+ /* No flush needed for cache modes that already do it */
+ if (bs->enable_write_cache) {
+ bdrv_flush(bs);
+ }
+
+ return 0;
+}
+
+int bdrv_write_compressed(BlockDriverState *bs, int64_t sector_num,
+ const uint8_t *buf, int nb_sectors)
+{
+ BlockDriver *drv = bs->drv;
+ if (!drv)
+ return -ENOMEDIUM;
+ if (!drv->bdrv_write_compressed)
+ return -ENOTSUP;
+ if (bdrv_check_request(bs, sector_num, nb_sectors))
+ return -EIO;
+
+ assert(QLIST_EMPTY(&bs->dirty_bitmaps));
+
+ return drv->bdrv_write_compressed(bs, sector_num, buf, nb_sectors);
+}
+
+static int coroutine_fn bdrv_co_do_copy_on_readv(BlockDriverState *bs,
+ int64_t sector_num, int nb_sectors, QEMUIOVector *qiov)
+{
+ /* Perform I/O through a temporary buffer so that users who scribble over
+ * their read buffer while the operation is in progress do not end up
+ * modifying the image file. This is critical for zero-copy guest I/O
+ * where anything might happen inside guest memory.
+ */
+ void *bounce_buffer;
+
+ BlockDriver *drv = bs->drv;
+ struct iovec iov;
+ QEMUIOVector bounce_qiov;
+ int64_t cluster_sector_num;
+ int cluster_nb_sectors;
+ size_t skip_bytes;
+ int ret;
+
+ /* Cover entire cluster so no additional backing file I/O is required when
+ * allocating cluster in the image file.
+ */
+ bdrv_round_to_clusters(bs, sector_num, nb_sectors,
+ &cluster_sector_num, &cluster_nb_sectors);
+
+ trace_bdrv_co_do_copy_on_readv(bs, sector_num, nb_sectors,
+ cluster_sector_num, cluster_nb_sectors);
+
+ iov.iov_len = cluster_nb_sectors * BDRV_SECTOR_SIZE;
+ iov.iov_base = bounce_buffer = qemu_try_blockalign(bs, iov.iov_len);
+ if (bounce_buffer == NULL) {
+ ret = -ENOMEM;
+ goto err;
+ }
+
+ qemu_iovec_init_external(&bounce_qiov, &iov, 1);
+
+ ret = drv->bdrv_co_readv(bs, cluster_sector_num, cluster_nb_sectors,
+ &bounce_qiov);
+ if (ret < 0) {
+ goto err;
+ }
+
+ 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, 0);
+ } else {
+ /* This does not change the data on the disk, it is not necessary
+ * to flush even in cache=writethrough mode.
+ */
+ ret = drv->bdrv_co_writev(bs, cluster_sector_num, cluster_nb_sectors,
+ &bounce_qiov);
+ }
+
+ if (ret < 0) {
+ /* It might be okay to ignore write errors for guest requests. If this
+ * is a deliberate copy-on-read then we don't want to ignore the error.
+ * Simply report it in all cases.
+ */
+ goto err;
+ }
+
+ skip_bytes = (sector_num - cluster_sector_num) * BDRV_SECTOR_SIZE;
+ qemu_iovec_from_buf(qiov, 0, bounce_buffer + skip_bytes,
+ nb_sectors * BDRV_SECTOR_SIZE);
+
+err:
+ qemu_vfree(bounce_buffer);
+ return ret;
+}
+
+static int bdrv_get_cluster_size(BlockDriverState *bs)
+{
+ BlockDriverInfo bdi;
+ int ret;
+
+ ret = bdrv_get_info(bs, &bdi);
+ if (ret < 0 || bdi.cluster_size == 0) {
+ return bs->request_alignment;
+ } else {
+ return bdi.cluster_size;
+ }
+}
+
+/*
+ * Forwards an already correctly aligned request to the BlockDriver. This
+ * handles copy on read and zeroing after EOF; any other features must be
+ * implemented by the caller.
+ */
+static int coroutine_fn bdrv_aligned_preadv(BlockDriverState *bs,
+ BdrvTrackedRequest *req, int64_t offset, unsigned int bytes,
+ int64_t align, QEMUIOVector *qiov, int flags)
+{
+ BlockDriver *drv = bs->drv;
+ int ret;
+
+ int64_t sector_num = offset >> BDRV_SECTOR_BITS;
+ unsigned int nb_sectors = bytes >> BDRV_SECTOR_BITS;
+
+ assert((offset & (BDRV_SECTOR_SIZE - 1)) == 0);
+ assert((bytes & (BDRV_SECTOR_SIZE - 1)) == 0);
+ assert(!qiov || bytes == qiov->size);
+
+ /* Handle Copy on Read and associated serialisation */
+ if (flags & BDRV_REQ_COPY_ON_READ) {
+ /* If we touch the same cluster it counts as an overlap. This
+ * guarantees that allocating writes will be serialized and not race
+ * with each other for the same cluster. For example, in copy-on-read
+ * it ensures that the CoR read and write operations are atomic and
+ * guest writes cannot interleave between them. */
+ mark_request_serialising(req, bdrv_get_cluster_size(bs));
+ }
+
+ wait_serialising_requests(req);
+
+ if (flags & BDRV_REQ_COPY_ON_READ) {
+ int pnum;
+
+ ret = bdrv_is_allocated(bs, sector_num, nb_sectors, &pnum);
+ if (ret < 0) {
+ goto out;
+ }
+
+ if (!ret || pnum != nb_sectors) {
+ ret = bdrv_co_do_copy_on_readv(bs, sector_num, nb_sectors, qiov);
+ goto out;
+ }
+ }
+
+ /* Forward the request to the BlockDriver */
+ if (!(bs->zero_beyond_eof && bs->growable)) {
+ ret = drv->bdrv_co_readv(bs, sector_num, nb_sectors, qiov);
+ } else {
+ /* Read zeros after EOF of growable BDSes */
+ int64_t total_sectors, max_nb_sectors;
+
+ total_sectors = bdrv_nb_sectors(bs);
+ if (total_sectors < 0) {
+ ret = total_sectors;
+ goto out;
+ }
+
+ max_nb_sectors = ROUND_UP(MAX(0, total_sectors - sector_num),
+ align >> BDRV_SECTOR_BITS);
+ if (nb_sectors < max_nb_sectors) {
+ ret = drv->bdrv_co_readv(bs, sector_num, nb_sectors, qiov);
+ } else if (max_nb_sectors > 0) {
+ QEMUIOVector local_qiov;
+
+ qemu_iovec_init(&local_qiov, qiov->niov);
+ qemu_iovec_concat(&local_qiov, qiov, 0,
+ max_nb_sectors * BDRV_SECTOR_SIZE);
+
+ ret = drv->bdrv_co_readv(bs, sector_num, max_nb_sectors,
+ &local_qiov);
+
+ qemu_iovec_destroy(&local_qiov);
+ } else {
+ ret = 0;
+ }
+
+ /* Reading beyond end of file is supposed to produce zeroes */
+ if (ret == 0 && total_sectors < sector_num + nb_sectors) {
+ uint64_t offset = MAX(0, total_sectors - sector_num);
+ uint64_t bytes = (sector_num + nb_sectors - offset) *
+ BDRV_SECTOR_SIZE;
+ qemu_iovec_memset(qiov, offset * BDRV_SECTOR_SIZE, 0, bytes);
+ }
+ }
+
+out:
+ return ret;
+}
+
+/*
+ * Handle a read request in coroutine context
+ */
+static int coroutine_fn bdrv_co_do_preadv(BlockDriverState *bs,
+ int64_t offset, unsigned int bytes, QEMUIOVector *qiov,
+ BdrvRequestFlags flags)
+{
+ BlockDriver *drv = bs->drv;
+ BdrvTrackedRequest req;
+
+ /* TODO Lift BDRV_SECTOR_SIZE restriction in BlockDriver interface */
+ uint64_t align = MAX(BDRV_SECTOR_SIZE, bs->request_alignment);
+ uint8_t *head_buf = NULL;
+ uint8_t *tail_buf = NULL;
+ QEMUIOVector local_qiov;
+ bool use_local_qiov = false;
+ int ret;
+
+ if (!drv) {
+ return -ENOMEDIUM;
+ }
+ if (bdrv_check_byte_request(bs, offset, bytes)) {
+ return -EIO;
+ }
+
+ if (bs->copy_on_read) {
+ flags |= BDRV_REQ_COPY_ON_READ;
+ }
+
+ /* throttling disk I/O */
+ if (bs->io_limits_enabled) {
+ bdrv_io_limits_intercept(bs, bytes, false);
+ }
+
+ /* Align read if necessary by padding qiov */
+ if (offset & (align - 1)) {
+ head_buf = qemu_blockalign(bs, align);
+ qemu_iovec_init(&local_qiov, qiov->niov + 2);
+ qemu_iovec_add(&local_qiov, head_buf, offset & (align - 1));
+ qemu_iovec_concat(&local_qiov, qiov, 0, qiov->size);
+ use_local_qiov = true;
+
+ bytes += offset & (align - 1);
+ offset = offset & ~(align - 1);
+ }
+
+ if ((offset + bytes) & (align - 1)) {
+ if (!use_local_qiov) {
+ qemu_iovec_init(&local_qiov, qiov->niov + 1);
+ qemu_iovec_concat(&local_qiov, qiov, 0, qiov->size);
+ use_local_qiov = true;
+ }
+ tail_buf = qemu_blockalign(bs, align);
+ qemu_iovec_add(&local_qiov, tail_buf,
+ align - ((offset + bytes) & (align - 1)));
+
+ bytes = ROUND_UP(bytes, align);
+ }
+
+ tracked_request_begin(&req, bs, offset, bytes, false);
+ ret = bdrv_aligned_preadv(bs, &req, offset, bytes, align,
+ use_local_qiov ? &local_qiov : qiov,
+ flags);
+ tracked_request_end(&req);
+
+ if (use_local_qiov) {
+ qemu_iovec_destroy(&local_qiov);
+ qemu_vfree(head_buf);
+ qemu_vfree(tail_buf);
+ }
+
+ return ret;
+}
+
+static int coroutine_fn bdrv_co_do_readv(BlockDriverState *bs,
+ int64_t sector_num, int nb_sectors, QEMUIOVector *qiov,
+ BdrvRequestFlags flags)
+{
+ if (nb_sectors < 0 || nb_sectors > (UINT_MAX >> BDRV_SECTOR_BITS)) {
+ return -EINVAL;
+ }
+
+ return bdrv_co_do_preadv(bs, sector_num << BDRV_SECTOR_BITS,
+ nb_sectors << BDRV_SECTOR_BITS, qiov, flags);
+}
+
+int coroutine_fn bdrv_co_readv(BlockDriverState *bs, int64_t sector_num,
+ int nb_sectors, QEMUIOVector *qiov)
+{
+ trace_bdrv_co_readv(bs, sector_num, nb_sectors);
+
+ return bdrv_co_do_readv(bs, sector_num, nb_sectors, qiov, 0);
+}
+
+int coroutine_fn bdrv_co_copy_on_readv(BlockDriverState *bs,
+ int64_t sector_num, int nb_sectors, QEMUIOVector *qiov)
+{
+ trace_bdrv_co_copy_on_readv(bs, sector_num, nb_sectors);
+
+ return bdrv_co_do_readv(bs, sector_num, nb_sectors, qiov,
+ 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, BdrvRequestFlags flags)
+{
+ BlockDriver *drv = bs->drv;
+ QEMUIOVector qiov;
+ struct iovec iov = {0};
+ int ret = 0;
+
+ int max_write_zeroes = bs->bl.max_write_zeroes ?
+ bs->bl.max_write_zeroes : MAX_WRITE_ZEROES_DEFAULT;
+
+ 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 */
+ int max_xfer_len = MIN_NON_ZERO(bs->bl.max_transfer_length,
+ MAX_WRITE_ZEROES_DEFAULT);
+ num = MIN(num, max_xfer_len);
+ iov.iov_len = num * BDRV_SECTOR_SIZE;
+ if (iov.iov_base == NULL) {
+ iov.iov_base = qemu_try_blockalign(bs, num * BDRV_SECTOR_SIZE);
+ if (iov.iov_base == NULL) {
+ ret = -ENOMEM;
+ goto fail;
+ }
+ 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_xfer_len) {
+ qemu_vfree(iov.iov_base);
+ iov.iov_base = NULL;
+ }
+ }
+
+ sector_num += num;
+ nb_sectors -= num;
+ }
+
+fail:
+ qemu_vfree(iov.iov_base);
+ return ret;
+}
+
+/*
+ * Forwards an already correctly aligned write request to the BlockDriver.
+ */
+static int coroutine_fn bdrv_aligned_pwritev(BlockDriverState *bs,
+ BdrvTrackedRequest *req, int64_t offset, unsigned int bytes,
+ QEMUIOVector *qiov, int flags)
+{
+ BlockDriver *drv = bs->drv;
+ bool waited;
+ int ret;
+
+ int64_t sector_num = offset >> BDRV_SECTOR_BITS;
+ unsigned int nb_sectors = bytes >> BDRV_SECTOR_BITS;
+
+ assert((offset & (BDRV_SECTOR_SIZE - 1)) == 0);
+ assert((bytes & (BDRV_SECTOR_SIZE - 1)) == 0);
+ assert(!qiov || bytes == qiov->size);
+
+ waited = wait_serialising_requests(req);
+ assert(!waited || !req->serialising);
+ assert(req->overlap_offset <= offset);
+ assert(offset + bytes <= req->overlap_offset + req->overlap_bytes);
+
+ ret = notifier_with_return_list_notify(&bs->before_write_notifiers, req);
+
+ if (!ret && bs->detect_zeroes != BLOCKDEV_DETECT_ZEROES_OPTIONS_OFF &&
+ !(flags & BDRV_REQ_ZERO_WRITE) && drv->bdrv_co_write_zeroes &&
+ qemu_iovec_is_zero(qiov)) {
+ flags |= BDRV_REQ_ZERO_WRITE;
+ if (bs->detect_zeroes == BLOCKDEV_DETECT_ZEROES_OPTIONS_UNMAP) {
+ flags |= BDRV_REQ_MAY_UNMAP;
+ }
+ }
+
+ if (ret < 0) {
+ /* Do nothing, write notifier decided to fail this request */
+ } else if (flags & BDRV_REQ_ZERO_WRITE) {
+ BLKDBG_EVENT(bs, BLKDBG_PWRITEV_ZERO);
+ ret = bdrv_co_do_write_zeroes(bs, sector_num, nb_sectors, flags);
+ } else {
+ BLKDBG_EVENT(bs, BLKDBG_PWRITEV);
+ ret = drv->bdrv_co_writev(bs, sector_num, nb_sectors, qiov);
+ }
+ BLKDBG_EVENT(bs, BLKDBG_PWRITEV_DONE);
+
+ if (ret == 0 && !bs->enable_write_cache) {
+ ret = bdrv_co_flush(bs);
+ }
+
+ bdrv_set_dirty(bs, sector_num, nb_sectors);
+
+ block_acct_highest_sector(&bs->stats, sector_num, nb_sectors);
+
+ if (bs->growable && ret >= 0) {
+ bs->total_sectors = MAX(bs->total_sectors, sector_num + nb_sectors);
+ }
+
+ return ret;
+}
+
+/*
+ * Handle a write request in coroutine context
+ */
+static int coroutine_fn bdrv_co_do_pwritev(BlockDriverState *bs,
+ int64_t offset, unsigned int bytes, QEMUIOVector *qiov,
+ BdrvRequestFlags flags)
+{
+ BdrvTrackedRequest req;
+ /* TODO Lift BDRV_SECTOR_SIZE restriction in BlockDriver interface */
+ uint64_t align = MAX(BDRV_SECTOR_SIZE, bs->request_alignment);
+ uint8_t *head_buf = NULL;
+ uint8_t *tail_buf = NULL;
+ QEMUIOVector local_qiov;
+ bool use_local_qiov = false;
+ int ret;
+
+ if (!bs->drv) {
+ return -ENOMEDIUM;
+ }
+ if (bs->read_only) {
+ return -EACCES;
+ }
+ if (bdrv_check_byte_request(bs, offset, bytes)) {
+ return -EIO;
+ }
+
+ /* throttling disk I/O */
+ if (bs->io_limits_enabled) {
+ bdrv_io_limits_intercept(bs, bytes, true);
+ }
+
+ /*
+ * Align write if necessary by performing a read-modify-write cycle.
+ * Pad qiov with the read parts and be sure to have a tracked request not
+ * only for bdrv_aligned_pwritev, but also for the reads of the RMW cycle.
+ */
+ tracked_request_begin(&req, bs, offset, bytes, true);
+
+ if (offset & (align - 1)) {
+ QEMUIOVector head_qiov;
+ struct iovec head_iov;
+
+ mark_request_serialising(&req, align);
+ wait_serialising_requests(&req);
+
+ head_buf = qemu_blockalign(bs, align);
+ head_iov = (struct iovec) {
+ .iov_base = head_buf,
+ .iov_len = align,
+ };
+ qemu_iovec_init_external(&head_qiov, &head_iov, 1);
+
+ BLKDBG_EVENT(bs, BLKDBG_PWRITEV_RMW_HEAD);
+ ret = bdrv_aligned_preadv(bs, &req, offset & ~(align - 1), align,
+ align, &head_qiov, 0);
+ if (ret < 0) {
+ goto fail;
+ }
+ BLKDBG_EVENT(bs, BLKDBG_PWRITEV_RMW_AFTER_HEAD);
+
+ qemu_iovec_init(&local_qiov, qiov->niov + 2);
+ qemu_iovec_add(&local_qiov, head_buf, offset & (align - 1));
+ qemu_iovec_concat(&local_qiov, qiov, 0, qiov->size);
+ use_local_qiov = true;
+
+ bytes += offset & (align - 1);
+ offset = offset & ~(align - 1);
+ }
+
+ if ((offset + bytes) & (align - 1)) {
+ QEMUIOVector tail_qiov;
+ struct iovec tail_iov;
+ size_t tail_bytes;
+ bool waited;
+
+ mark_request_serialising(&req, align);
+ waited = wait_serialising_requests(&req);
+ assert(!waited || !use_local_qiov);
+
+ tail_buf = qemu_blockalign(bs, align);
+ tail_iov = (struct iovec) {
+ .iov_base = tail_buf,
+ .iov_len = align,
+ };
+ qemu_iovec_init_external(&tail_qiov, &tail_iov, 1);
+
+ BLKDBG_EVENT(bs, BLKDBG_PWRITEV_RMW_TAIL);
+ ret = bdrv_aligned_preadv(bs, &req, (offset + bytes) & ~(align - 1),
align,
+ align, &tail_qiov, 0);
+ if (ret < 0) {
+ goto fail;
+ }
+ BLKDBG_EVENT(bs, BLKDBG_PWRITEV_RMW_AFTER_TAIL);
+
+ if (!use_local_qiov) {
+ qemu_iovec_init(&local_qiov, qiov->niov + 1);
+ qemu_iovec_concat(&local_qiov, qiov, 0, qiov->size);
+ use_local_qiov = true;
+ }
+
+ tail_bytes = (offset + bytes) & (align - 1);
+ qemu_iovec_add(&local_qiov, tail_buf + tail_bytes, align - tail_bytes);
+
+ bytes = ROUND_UP(bytes, align);
+ }
+
+ ret = bdrv_aligned_pwritev(bs, &req, offset, bytes,
+ use_local_qiov ? &local_qiov : qiov,
+ flags);
+
+fail:
+ tracked_request_end(&req);
+
+ if (use_local_qiov) {
+ qemu_iovec_destroy(&local_qiov);
+ }
+ qemu_vfree(head_buf);
+ qemu_vfree(tail_buf);
+
+ return ret;
+}
+
+static int coroutine_fn bdrv_co_do_writev(BlockDriverState *bs,
+ int64_t sector_num, int nb_sectors, QEMUIOVector *qiov,
+ BdrvRequestFlags flags)
+{
+ if (nb_sectors < 0 || nb_sectors > (INT_MAX >> BDRV_SECTOR_BITS)) {
+ return -EINVAL;
+ }
+
+ return bdrv_co_do_pwritev(bs, sector_num << BDRV_SECTOR_BITS,
+ nb_sectors << BDRV_SECTOR_BITS, qiov, flags);
+}
+
+int coroutine_fn bdrv_co_writev(BlockDriverState *bs, int64_t sector_num,
+ int nb_sectors, QEMUIOVector *qiov)
+{
+ trace_bdrv_co_writev(bs, sector_num, nb_sectors);
+
+ return bdrv_co_do_writev(bs, sector_num, nb_sectors, qiov, 0);
+}
+
+int coroutine_fn bdrv_co_write_zeroes(BlockDriverState *bs,
+ int64_t sector_num, int nb_sectors,
+ BdrvRequestFlags flags)
+{
+ 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 | flags);
+}
+
+/**************************************************************/
+/* async I/Os */
+
+BlockAIOCB *bdrv_aio_readv(BlockDriverState *bs, int64_t sector_num,
+ QEMUIOVector *qiov, int nb_sectors,
+ BlockCompletionFunc *cb, void *opaque)
+{
+ trace_bdrv_aio_readv(bs, sector_num, nb_sectors, opaque);
+
+ return bdrv_co_aio_rw_vector(bs, sector_num, qiov, nb_sectors, 0,
+ cb, opaque, false);
+}
+
+BlockAIOCB *bdrv_aio_writev(BlockDriverState *bs, int64_t sector_num,
+ QEMUIOVector *qiov, int nb_sectors,
+ BlockCompletionFunc *cb, void *opaque)
+{
+ trace_bdrv_aio_writev(bs, sector_num, nb_sectors, opaque);
+
+ return bdrv_co_aio_rw_vector(bs, sector_num, qiov, nb_sectors, 0,
+ cb, opaque, true);
+}
+
+BlockAIOCB *bdrv_aio_write_zeroes(BlockDriverState *bs,
+ int64_t sector_num, int nb_sectors, BdrvRequestFlags flags,
+ BlockCompletionFunc *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);
+}
+
+typedef struct MultiwriteCB {
+ int error;
+ int num_requests;
+ int num_callbacks;
+ struct {
+ BlockCompletionFunc *cb;
+ void *opaque;
+ QEMUIOVector *free_qiov;
+ } callbacks[];
+} MultiwriteCB;
+
+static void multiwrite_user_cb(MultiwriteCB *mcb)
+{
+ int i;
+
+ for (i = 0; i < mcb->num_callbacks; i++) {
+ mcb->callbacks[i].cb(mcb->callbacks[i].opaque, mcb->error);
+ if (mcb->callbacks[i].free_qiov) {
+ qemu_iovec_destroy(mcb->callbacks[i].free_qiov);
+ }
+ g_free(mcb->callbacks[i].free_qiov);
+ }
+}
+
+static void multiwrite_cb(void *opaque, int ret)
+{
+ MultiwriteCB *mcb = opaque;
+
+ trace_multiwrite_cb(mcb, ret);
+
+ if (ret < 0 && !mcb->error) {
+ mcb->error = ret;
+ }
+
+ mcb->num_requests--;
+ if (mcb->num_requests == 0) {
+ multiwrite_user_cb(mcb);
+ g_free(mcb);
+ }
+}
+
+static int multiwrite_req_compare(const void *a, const void *b)
+{
+ const BlockRequest *req1 = a, *req2 = b;
+
+ /*
+ * Note that we can't simply subtract req2->sector from req1->sector
+ * here as that could overflow the return value.
+ */
+ if (req1->sector > req2->sector) {
+ return 1;
+ } else if (req1->sector < req2->sector) {
+ return -1;
+ } else {
+ return 0;
+ }
+}
+
+/*
+ * Takes a bunch of requests and tries to merge them. Returns the number of
+ * requests that remain after merging.
+ */
+static int multiwrite_merge(BlockDriverState *bs, BlockRequest *reqs,
+ int num_reqs, MultiwriteCB *mcb)
+{
+ int i, outidx;
+
+ // Sort requests by start sector
+ qsort(reqs, num_reqs, sizeof(*reqs), &multiwrite_req_compare);
+
+ // Check if adjacent requests touch the same clusters. If so, combine them,
+ // filling up gaps with zero sectors.
+ outidx = 0;
+ for (i = 1; i < num_reqs; i++) {
+ int merge = 0;
+ int64_t oldreq_last = reqs[outidx].sector + reqs[outidx].nb_sectors;
+
+ // Handle exactly sequential writes and overlapping writes.
+ if (reqs[i].sector <= oldreq_last) {
+ merge = 1;
+ }
+
+ if (reqs[outidx].qiov->niov + reqs[i].qiov->niov + 1 > IOV_MAX) {
+ merge = 0;
+ }
+
+ if (bs->bl.max_transfer_length && reqs[outidx].nb_sectors +
+ reqs[i].nb_sectors > bs->bl.max_transfer_length) {
+ merge = 0;
+ }
+
+ if (merge) {
+ size_t size;
+ QEMUIOVector *qiov = g_malloc0(sizeof(*qiov));
+ qemu_iovec_init(qiov,
+ reqs[outidx].qiov->niov + reqs[i].qiov->niov + 1);
+
+ // Add the first request to the merged one. If the requests are
+ // overlapping, drop the last sectors of the first request.
+ size = (reqs[i].sector - reqs[outidx].sector) << 9;
+ qemu_iovec_concat(qiov, reqs[outidx].qiov, 0, size);
+
+ // We should need to add any zeros between the two requests
+ assert (reqs[i].sector <= oldreq_last);
+
+ // Add the second request
+ qemu_iovec_concat(qiov, reqs[i].qiov, 0, reqs[i].qiov->size);
+
+ // Add tail of first request, if necessary
+ if (qiov->size < reqs[outidx].qiov->size) {
+ qemu_iovec_concat(qiov, reqs[outidx].qiov, qiov->size,
+ reqs[outidx].qiov->size - qiov->size);
+ }
+
+ reqs[outidx].nb_sectors = qiov->size >> 9;
+ reqs[outidx].qiov = qiov;
+
+ mcb->callbacks[i].free_qiov = reqs[outidx].qiov;
+ } else {
+ outidx++;
+ reqs[outidx].sector = reqs[i].sector;
+ reqs[outidx].nb_sectors = reqs[i].nb_sectors;
+ reqs[outidx].qiov = reqs[i].qiov;
+ }
+ }
+
+ return outidx + 1;
+}
+
+/*
+ * Submit multiple AIO write requests at once.
+ *
+ * On success, the function returns 0 and all requests in the reqs array have
+ * been submitted. In error case this function returns -1, and any of the
+ * requests may or may not be submitted yet. In particular, this means that the
+ * callback will be called for some of the requests, for others it won't. The
+ * caller must check the error field of the BlockRequest to wait for the right
+ * callbacks (if error != 0, no callback will be called).
+ *
+ * The implementation may modify the contents of the reqs array, e.g. to merge
+ * requests. However, the fields opaque and error are left unmodified as they
+ * are used to signal failure for a single request to the caller.
+ */
+int bdrv_aio_multiwrite(BlockDriverState *bs, BlockRequest *reqs, int num_reqs)
+{
+ MultiwriteCB *mcb;
+ int i;
+
+ /* don't submit writes if we don't have a medium */
+ if (bs->drv == NULL) {
+ for (i = 0; i < num_reqs; i++) {
+ reqs[i].error = -ENOMEDIUM;
+ }
+ return -1;
+ }
+
+ if (num_reqs == 0) {
+ return 0;
+ }
+
+ // Create MultiwriteCB structure
+ mcb = g_malloc0(sizeof(*mcb) + num_reqs * sizeof(*mcb->callbacks));
+ mcb->num_requests = 0;
+ mcb->num_callbacks = num_reqs;
+
+ for (i = 0; i < num_reqs; i++) {
+ mcb->callbacks[i].cb = reqs[i].cb;
+ mcb->callbacks[i].opaque = reqs[i].opaque;
+ }
+
+ // Check for mergable requests
+ num_reqs = multiwrite_merge(bs, reqs, num_reqs, mcb);
+
+ trace_bdrv_aio_multiwrite(mcb, mcb->num_callbacks, num_reqs);
+
+ /* Run the aio requests. */
+ mcb->num_requests = num_reqs;
+ for (i = 0; i < num_reqs; i++) {
+ 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;
+}
+
+void bdrv_aio_cancel(BlockAIOCB *acb)
+{
+ qemu_aio_ref(acb);
+ bdrv_aio_cancel_async(acb);
+ while (acb->refcnt > 1) {
+ if (acb->aiocb_info->get_aio_context) {
+ aio_poll(acb->aiocb_info->get_aio_context(acb), true);
+ } else if (acb->bs) {
+ aio_poll(bdrv_get_aio_context(acb->bs), true);
+ } else {
+ abort();
+ }
+ }
+ qemu_aio_unref(acb);
+}
+
+/* Async version of aio cancel. The caller is not blocked if the acb implements
+ * cancel_async, otherwise we do nothing and let the request normally complete.
+ * In either case the completion callback must be called. */
+void bdrv_aio_cancel_async(BlockAIOCB *acb)
+{
+ if (acb->aiocb_info->cancel_async) {
+ acb->aiocb_info->cancel_async(acb);
+ }
+}
+
+/**************************************************************/
+/* async block device emulation */
+
+typedef struct BlockAIOCBSync {
+ BlockAIOCB common;
+ QEMUBH *bh;
+ int ret;
+ /* vector translation state */
+ QEMUIOVector *qiov;
+ uint8_t *bounce;
+ int is_write;
+} BlockAIOCBSync;
+
+static const AIOCBInfo bdrv_em_aiocb_info = {
+ .aiocb_size = sizeof(BlockAIOCBSync),
+};
+
+static void bdrv_aio_bh_cb(void *opaque)
+{
+ BlockAIOCBSync *acb = opaque;
+
+ if (!acb->is_write && acb->ret >= 0) {
+ qemu_iovec_from_buf(acb->qiov, 0, acb->bounce, acb->qiov->size);
+ }
+ qemu_vfree(acb->bounce);
+ acb->common.cb(acb->common.opaque, acb->ret);
+ qemu_bh_delete(acb->bh);
+ acb->bh = NULL;
+ qemu_aio_unref(acb);
+}
+
+static BlockAIOCB *bdrv_aio_rw_vector(BlockDriverState *bs,
+ int64_t sector_num,
+ QEMUIOVector *qiov,
+ int nb_sectors,
+ BlockCompletionFunc *cb,
+ void *opaque,
+ int is_write)
+
+{
+ BlockAIOCBSync *acb;
+
+ acb = qemu_aio_get(&bdrv_em_aiocb_info, bs, cb, opaque);
+ acb->is_write = is_write;
+ acb->qiov = qiov;
+ acb->bounce = qemu_try_blockalign(bs, qiov->size);
+ acb->bh = aio_bh_new(bdrv_get_aio_context(bs), bdrv_aio_bh_cb, acb);
+
+ if (acb->bounce == NULL) {
+ acb->ret = -ENOMEM;
+ } else if (is_write) {
+ qemu_iovec_to_buf(acb->qiov, 0, acb->bounce, qiov->size);
+ acb->ret = bs->drv->bdrv_write(bs, sector_num, acb->bounce,
nb_sectors);
+ } else {
+ acb->ret = bs->drv->bdrv_read(bs, sector_num, acb->bounce, nb_sectors);
+ }
+
+ qemu_bh_schedule(acb->bh);
+
+ return &acb->common;
+}
+
+static BlockAIOCB *bdrv_aio_readv_em(BlockDriverState *bs,
+ int64_t sector_num, QEMUIOVector *qiov, int nb_sectors,
+ BlockCompletionFunc *cb, void *opaque)
+{
+ return bdrv_aio_rw_vector(bs, sector_num, qiov, nb_sectors, cb, opaque, 0);
+}
+
+static BlockAIOCB *bdrv_aio_writev_em(BlockDriverState *bs,
+ int64_t sector_num, QEMUIOVector *qiov, int nb_sectors,
+ BlockCompletionFunc *cb, void *opaque)
+{
+ return bdrv_aio_rw_vector(bs, sector_num, qiov, nb_sectors, cb, opaque, 1);
+}
+
+typedef struct BlockAIOCBCoroutine {
+ BlockAIOCB common;
+ BlockRequest req;
+ bool is_write;
+ bool *done;
+ QEMUBH* bh;
+} BlockAIOCBCoroutine;
+
+static const AIOCBInfo bdrv_em_co_aiocb_info = {
+ .aiocb_size = sizeof(BlockAIOCBCoroutine),
+};
+
+static void bdrv_co_em_bh(void *opaque)
+{
+ BlockAIOCBCoroutine *acb = opaque;
+
+ acb->common.cb(acb->common.opaque, acb->req.error);
+
+ qemu_bh_delete(acb->bh);
+ qemu_aio_unref(acb);
+}
+
+/* Invoke bdrv_co_do_readv/bdrv_co_do_writev */
+static void coroutine_fn bdrv_co_do_rw(void *opaque)
+{
+ BlockAIOCBCoroutine *acb = opaque;
+ BlockDriverState *bs = acb->common.bs;
+
+ if (!acb->is_write) {
+ acb->req.error = bdrv_co_do_readv(bs, acb->req.sector,
+ 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, acb->req.flags);
+ }
+
+ acb->bh = aio_bh_new(bdrv_get_aio_context(bs), bdrv_co_em_bh, acb);
+ qemu_bh_schedule(acb->bh);
+}
+
+static BlockAIOCB *bdrv_co_aio_rw_vector(BlockDriverState *bs,
+ int64_t sector_num,
+ QEMUIOVector *qiov,
+ int nb_sectors,
+ BdrvRequestFlags flags,
+ BlockCompletionFunc *cb,
+ void *opaque,
+ bool is_write)
+{
+ Coroutine *co;
+ BlockAIOCBCoroutine *acb;
+
+ acb = qemu_aio_get(&bdrv_em_co_aiocb_info, bs, cb, opaque);
+ acb->req.sector = sector_num;
+ acb->req.nb_sectors = nb_sectors;
+ acb->req.qiov = qiov;
+ acb->req.flags = flags;
+ acb->is_write = is_write;
+
+ co = qemu_coroutine_create(bdrv_co_do_rw);
+ qemu_coroutine_enter(co, acb);
+
+ return &acb->common;
+}
+
+static void coroutine_fn bdrv_aio_flush_co_entry(void *opaque)
+{
+ BlockAIOCBCoroutine *acb = opaque;
+ BlockDriverState *bs = acb->common.bs;
+
+ acb->req.error = bdrv_co_flush(bs);
+ acb->bh = aio_bh_new(bdrv_get_aio_context(bs), bdrv_co_em_bh, acb);
+ qemu_bh_schedule(acb->bh);
+}
+
+BlockAIOCB *bdrv_aio_flush(BlockDriverState *bs,
+ BlockCompletionFunc *cb, void *opaque)
+{
+ trace_bdrv_aio_flush(bs, opaque);
+
+ Coroutine *co;
+ BlockAIOCBCoroutine *acb;
+
+ acb = qemu_aio_get(&bdrv_em_co_aiocb_info, bs, cb, opaque);
+
+ co = qemu_coroutine_create(bdrv_aio_flush_co_entry);
+ qemu_coroutine_enter(co, acb);
+
+ return &acb->common;
+}
+
+static void coroutine_fn bdrv_aio_discard_co_entry(void *opaque)
+{
+ BlockAIOCBCoroutine *acb = opaque;
+ BlockDriverState *bs = acb->common.bs;
+
+ acb->req.error = bdrv_co_discard(bs, acb->req.sector, acb->req.nb_sectors);
+ acb->bh = aio_bh_new(bdrv_get_aio_context(bs), bdrv_co_em_bh, acb);
+ qemu_bh_schedule(acb->bh);
+}
+
+BlockAIOCB *bdrv_aio_discard(BlockDriverState *bs,
+ int64_t sector_num, int nb_sectors,
+ BlockCompletionFunc *cb, void *opaque)
+{
+ Coroutine *co;
+ BlockAIOCBCoroutine *acb;
+
+ trace_bdrv_aio_discard(bs, sector_num, nb_sectors, opaque);
+
+ acb = qemu_aio_get(&bdrv_em_co_aiocb_info, bs, cb, opaque);
+ acb->req.sector = sector_num;
+ acb->req.nb_sectors = nb_sectors;
+ co = qemu_coroutine_create(bdrv_aio_discard_co_entry);
+ qemu_coroutine_enter(co, acb);
+
+ return &acb->common;
+}
+void *qemu_aio_get(const AIOCBInfo *aiocb_info, BlockDriverState *bs,
+ BlockCompletionFunc *cb, void *opaque)
+{
+ BlockAIOCB *acb;
+
+ acb = g_slice_alloc(aiocb_info->aiocb_size);
+ acb->aiocb_info = aiocb_info;
+ acb->bs = bs;
+ acb->cb = cb;
+ acb->opaque = opaque;
+ acb->refcnt = 1;
+ return acb;
+}
+
+void qemu_aio_ref(void *p)
+{
+ BlockAIOCB *acb = p;
+ acb->refcnt++;
+}
+
+void qemu_aio_unref(void *p)
+{
+ BlockAIOCB *acb = p;
+ assert(acb->refcnt > 0);
+ if (--acb->refcnt == 0) {
+ g_slice_free1(acb->aiocb_info->aiocb_size, acb);
+ }
+}
+
+/**************************************************************/
+/* Coroutine block device emulation */
+
+typedef struct CoroutineIOCompletion {
+ Coroutine *coroutine;
+ int ret;
+} CoroutineIOCompletion;
+
+static void bdrv_co_io_em_complete(void *opaque, int ret)
+{
+ CoroutineIOCompletion *co = opaque;
+
+ co->ret = ret;
+ qemu_coroutine_enter(co->coroutine, NULL);
+}
+
+static int coroutine_fn bdrv_co_io_em(BlockDriverState *bs, int64_t sector_num,
+ int nb_sectors, QEMUIOVector *iov,
+ bool is_write)
+{
+ CoroutineIOCompletion co = {
+ .coroutine = qemu_coroutine_self(),
+ };
+ BlockAIOCB *acb;
+
+ if (is_write) {
+ acb = bs->drv->bdrv_aio_writev(bs, sector_num, iov, nb_sectors,
+ bdrv_co_io_em_complete, &co);
+ } else {
+ acb = bs->drv->bdrv_aio_readv(bs, sector_num, iov, nb_sectors,
+ bdrv_co_io_em_complete, &co);
+ }
+
+ trace_bdrv_co_io_em(bs, sector_num, nb_sectors, is_write, acb);
+ if (!acb) {
+ return -EIO;
+ }
+ qemu_coroutine_yield();
+
+ return co.ret;
+}
+
+static int coroutine_fn bdrv_co_readv_em(BlockDriverState *bs,
+ int64_t sector_num, int nb_sectors,
+ QEMUIOVector *iov)
+{
+ return bdrv_co_io_em(bs, sector_num, nb_sectors, iov, false);
+}
+
+static int coroutine_fn bdrv_co_writev_em(BlockDriverState *bs,
+ int64_t sector_num, int nb_sectors,
+ QEMUIOVector *iov)
+{
+ return bdrv_co_io_em(bs, sector_num, nb_sectors, iov, true);
+}
+
+static void coroutine_fn bdrv_flush_co_entry(void *opaque)
+{
+ RwCo *rwco = opaque;
+
+ rwco->ret = bdrv_co_flush(rwco->bs);
+}
+
+int coroutine_fn bdrv_co_flush(BlockDriverState *bs)
+{
+ int ret;
+
+ if (!bs || !bdrv_is_inserted(bs) || bdrv_is_read_only(bs)) {
+ return 0;
+ }
+
+ /* Write back cached data to the OS even with cache=unsafe */
+ BLKDBG_EVENT(bs->file, BLKDBG_FLUSH_TO_OS);
+ if (bs->drv->bdrv_co_flush_to_os) {
+ ret = bs->drv->bdrv_co_flush_to_os(bs);
+ if (ret < 0) {
+ return ret;
+ }
+ }
+
+ /* But don't actually force it to the disk with cache=unsafe */
+ if (bs->open_flags & BDRV_O_NO_FLUSH) {
+ goto flush_parent;
+ }
+
+ BLKDBG_EVENT(bs->file, BLKDBG_FLUSH_TO_DISK);
+ if (bs->drv->bdrv_co_flush_to_disk) {
+ ret = bs->drv->bdrv_co_flush_to_disk(bs);
+ } else if (bs->drv->bdrv_aio_flush) {
+ BlockAIOCB *acb;
+ CoroutineIOCompletion co = {
+ .coroutine = qemu_coroutine_self(),
+ };
+
+ acb = bs->drv->bdrv_aio_flush(bs, bdrv_co_io_em_complete, &co);
+ if (acb == NULL) {
+ ret = -EIO;
+ } else {
+ qemu_coroutine_yield();
+ ret = co.ret;
+ }
+ } else {
+ /*
+ * Some block drivers always operate in either writethrough or unsafe
+ * mode and don't support bdrv_flush therefore. Usually qemu doesn't
+ * know how the server works (because the behaviour is hardcoded or
+ * depends on server-side configuration), so we can't ensure that
+ * everything is safe on disk. Returning an error doesn't work because
+ * that would break guests even if the server operates in writethrough
+ * mode.
+ *
+ * Let's hope the user knows what he's doing.
+ */
+ ret = 0;
+ }
+ if (ret < 0) {
+ return ret;
+ }
+
+ /* Now flush the underlying protocol. It will also have BDRV_O_NO_FLUSH
+ * in the case of cache=unsafe, so there are no useless flushes.
+ */
+flush_parent:
+ return bdrv_co_flush(bs->file);
+}
+
+int bdrv_flush(BlockDriverState *bs)
+{
+ Coroutine *co;
+ RwCo rwco = {
+ .bs = bs,
+ .ret = NOT_DONE,
+ };
+
+ if (qemu_in_coroutine()) {
+ /* Fast-path if already in coroutine context */
+ bdrv_flush_co_entry(&rwco);
+ } else {
+ AioContext *aio_context = bdrv_get_aio_context(bs);
+
+ co = qemu_coroutine_create(bdrv_flush_co_entry);
+ qemu_coroutine_enter(co, &rwco);
+ while (rwco.ret == NOT_DONE) {
+ aio_poll(aio_context, true);
+ }
+ }
+
+ return rwco.ret;
+}
+
+typedef struct DiscardCo {
+ BlockDriverState *bs;
+ int64_t sector_num;
+ int nb_sectors;
+ int ret;
+} DiscardCo;
+static void coroutine_fn bdrv_discard_co_entry(void *opaque)
+{
+ DiscardCo *rwco = 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)) {
+ return -EIO;
+ } else if (bs->read_only) {
+ return -EROFS;
+ }
+
+ 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 && !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 {
+ BlockAIOCB *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)
+{
+ Coroutine *co;
+ DiscardCo rwco = {
+ .bs = bs,
+ .sector_num = sector_num,
+ .nb_sectors = nb_sectors,
+ .ret = NOT_DONE,
+ };
+
+ if (qemu_in_coroutine()) {
+ /* Fast-path if already in coroutine context */
+ bdrv_discard_co_entry(&rwco);
+ } else {
+ AioContext *aio_context = bdrv_get_aio_context(bs);
+
+ co = qemu_coroutine_create(bdrv_discard_co_entry);
+ qemu_coroutine_enter(co, &rwco);
+ while (rwco.ret == NOT_DONE) {
+ aio_poll(aio_context, true);
+ }
+ }
+
+ return rwco.ret;
+}
+
+/* needed for generic scsi interface */
+
+int bdrv_ioctl(BlockDriverState *bs, unsigned long int req, void *buf)
+{
+ BlockDriver *drv = bs->drv;
+
+ if (drv && drv->bdrv_ioctl)
+ return drv->bdrv_ioctl(bs, req, buf);
+ return -ENOTSUP;
+}
+
+BlockAIOCB *bdrv_aio_ioctl(BlockDriverState *bs,
+ unsigned long int req, void *buf,
+ BlockCompletionFunc *cb, void *opaque)
+{
+ BlockDriver *drv = bs->drv;
+
+ if (drv && drv->bdrv_aio_ioctl)
+ return drv->bdrv_aio_ioctl(bs, req, buf, cb, opaque);
+ return NULL;
+}
+
+void bdrv_add_before_write_notifier(BlockDriverState *bs,
+ NotifierWithReturn *notifier)
+{
+ notifier_with_return_list_add(&bs->before_write_notifiers, notifier);
+}
+
+void bdrv_io_plug(BlockDriverState *bs)
+{
+ BlockDriver *drv = bs->drv;
+ if (drv && drv->bdrv_io_plug) {
+ drv->bdrv_io_plug(bs);
+ } else if (bs->file) {
+ bdrv_io_plug(bs->file);
+ }
+}
+
+void bdrv_io_unplug(BlockDriverState *bs)
+{
+ BlockDriver *drv = bs->drv;
+ if (drv && drv->bdrv_io_unplug) {
+ drv->bdrv_io_unplug(bs);
+ } else if (bs->file) {
+ bdrv_io_unplug(bs->file);
+ }
+}
+
+void bdrv_flush_io_queue(BlockDriverState *bs)
+{
+ BlockDriver *drv = bs->drv;
+ if (drv && drv->bdrv_flush_io_queue) {
+ drv->bdrv_flush_io_queue(bs);
+ } else if (bs->file) {
+ bdrv_flush_io_queue(bs->file);
+ }
+}
diff --git a/include/block/block_int.h b/include/block/block_int.h
index e264be9..dd9a376 100644
--- a/include/block/block_int.h
+++ b/include/block/block_int.h
@@ -421,6 +421,14 @@ extern BlockDriver bdrv_file;
extern BlockDriver bdrv_raw;
extern BlockDriver bdrv_qcow2;
+/**
+ * bdrv_setup_io_funcs:
+ *
+ * Prepare a #BlockDriver for I/O request processing by populating
+ * unimplemented coroutine and AIO interfaces with generic wrapper functions
+ * that fall back to implemented interfaces.
+ */
+void bdrv_setup_io_funcs(BlockDriver *bdrv);
int get_tmp_filename(char *filename, int size);
BlockDriver *bdrv_probe_all(const uint8_t *buf, int buf_size,
@@ -606,4 +614,10 @@ bool blk_dev_is_tray_open(BlockBackend *blk);
bool blk_dev_is_medium_locked(BlockBackend *blk);
void blk_dev_resize_cb(BlockBackend *blk);
+bool bdrv_drain_one(BlockDriverState *bs);
+
+void bdrv_set_dirty(BlockDriverState *bs, int64_t cur_sector, int nr_sectors);
+void bdrv_reset_dirty(BlockDriverState *bs, int64_t cur_sector,
+ int nr_sectors);
+
#endif /* BLOCK_INT_H */
--
2.1.0
