void *aio_ctx;
#endif
#ifdef CONFIG_XFS
- bool is_xfs : 1;
+ bool is_xfs:1;
#endif
- bool has_discard : 1;
+ bool has_discard:1;
+ bool has_write_zeroes:1;
+ bool discard_zeroes:1;
} BDRVRawState;
typedef struct BDRVRawReopenState {
Error *local_err = NULL;
const char *filename;
int fd, ret;
+ struct stat st;
opts = qemu_opts_create_nofail(&raw_runtime_opts);
qemu_opts_absorb_qdict(opts, options, &local_err);
#endif
s->has_discard = true;
+ s->has_write_zeroes = true;
+
+ if (fstat(s->fd, &st) < 0) {
+ error_setg_errno(errp, errno, "Could not stat file");
+ goto fail;
+ }
+ if (S_ISREG(st.st_mode)) {
+ s->discard_zeroes = true;
+ }
+ if (S_ISBLK(st.st_mode)) {
+#ifdef BLKDISCARDZEROES
+ unsigned int arg;
+ if (ioctl(s->fd, BLKDISCARDZEROES, &arg) == 0 && arg) {
+ s->discard_zeroes = true;
+ }
+#endif
+#ifdef __linux__
+ /* On Linux 3.10, BLKDISCARD leaves stale data in the page cache. Do
+ * not rely on the contents of discarded blocks unless using O_DIRECT.
+ * Same for BLKZEROOUT.
+ */
+ if (!(bs->open_flags & BDRV_O_NOCACHE)) {
+ s->discard_zeroes = false;
+ s->has_write_zeroes = false;
+ }
+#endif
+ }
+
#ifdef CONFIG_XFS
if (platform_test_xfs_fd(s->fd)) {
s->is_xfs = true;
}
#ifdef CONFIG_XFS
+static int xfs_write_zeroes(BDRVRawState *s, int64_t offset, uint64_t bytes)
+{
+ struct xfs_flock64 fl;
+
+ memset(&fl, 0, sizeof(fl));
+ fl.l_whence = SEEK_SET;
+ fl.l_start = offset;
+ fl.l_len = bytes;
+
+ if (xfsctl(NULL, s->fd, XFS_IOC_ZERO_RANGE, &fl) < 0) {
+ DEBUG_BLOCK_PRINT("cannot write zero range (%s)\n", strerror(errno));
+ return -errno;
+ }
+
+ return 0;
+}
+
static int xfs_discard(BDRVRawState *s, int64_t offset, uint64_t bytes)
{
struct xfs_flock64 fl;
}
#endif
+static ssize_t handle_aiocb_write_zeroes(RawPosixAIOData *aiocb)
+{
+ int ret = -EOPNOTSUPP;
+ BDRVRawState *s = aiocb->bs->opaque;
+
+ if (s->has_write_zeroes == 0) {
+ return -ENOTSUP;
+ }
+
+ if (aiocb->aio_type & QEMU_AIO_BLKDEV) {
+#ifdef BLKZEROOUT
+ do {
+ uint64_t range[2] = { aiocb->aio_offset, aiocb->aio_nbytes };
+ if (ioctl(aiocb->aio_fildes, BLKZEROOUT, range) == 0) {
+ return 0;
+ }
+ } while (errno == EINTR);
+
+ ret = -errno;
+#endif
+ } else {
+#ifdef CONFIG_XFS
+ if (s->is_xfs) {
+ return xfs_write_zeroes(s, aiocb->aio_offset, aiocb->aio_nbytes);
+ }
+#endif
+ }
+
+ if (ret == -ENODEV || ret == -ENOSYS || ret == -EOPNOTSUPP ||
+ ret == -ENOTTY) {
+ s->has_write_zeroes = false;
+ ret = -ENOTSUP;
+ }
+ return ret;
+}
+
static ssize_t handle_aiocb_discard(RawPosixAIOData *aiocb)
{
int ret = -EOPNOTSUPP;
case QEMU_AIO_DISCARD:
ret = handle_aiocb_discard(aiocb);
break;
+ case QEMU_AIO_WRITE_ZEROES:
+ ret = handle_aiocb_write_zeroes(aiocb);
+ break;
default:
fprintf(stderr, "invalid aio request (0x%x)\n", aiocb->aio_type);
ret = -EINVAL;
return ret;
}
+static int paio_submit_co(BlockDriverState *bs, int fd,
+ int64_t sector_num, QEMUIOVector *qiov, int nb_sectors,
+ int type)
+{
+ RawPosixAIOData *acb = g_slice_new(RawPosixAIOData);
+ ThreadPool *pool;
+
+ acb->bs = bs;
+ acb->aio_type = type;
+ acb->aio_fildes = fd;
+
+ if (qiov) {
+ acb->aio_iov = qiov->iov;
+ acb->aio_niov = qiov->niov;
+ }
+ acb->aio_nbytes = nb_sectors * 512;
+ acb->aio_offset = sector_num * 512;
+
+ trace_paio_submit_co(sector_num, nb_sectors, type);
+ pool = aio_get_thread_pool(bdrv_get_aio_context(bs));
+ return thread_pool_submit_co(pool, aio_worker, acb);
+}
+
static BlockDriverAIOCB *paio_submit(BlockDriverState *bs, int fd,
int64_t sector_num, QEMUIOVector *qiov, int nb_sectors,
BlockDriverCompletionFunc *cb, void *opaque, int type)
cb, opaque, QEMU_AIO_DISCARD);
}
+static int coroutine_fn raw_co_write_zeroes(
+ BlockDriverState *bs, int64_t sector_num,
+ int nb_sectors, BdrvRequestFlags flags)
+{
+ BDRVRawState *s = bs->opaque;
+
+ if (!(flags & BDRV_REQ_MAY_UNMAP)) {
+ return paio_submit_co(bs, s->fd, sector_num, NULL, nb_sectors,
+ QEMU_AIO_WRITE_ZEROES);
+ } else if (s->discard_zeroes) {
+ return paio_submit_co(bs, s->fd, sector_num, NULL, nb_sectors,
+ QEMU_AIO_DISCARD);
+ }
+ return -ENOTSUP;
+}
+
+static int raw_get_info(BlockDriverState *bs, BlockDriverInfo *bdi)
+{
+ BDRVRawState *s = bs->opaque;
+
+ bdi->unallocated_blocks_are_zero = s->discard_zeroes;
+ bdi->can_write_zeroes_with_unmap = s->discard_zeroes;
+ return 0;
+}
+
static QEMUOptionParameter raw_create_options[] = {
{
.name = BLOCK_OPT_SIZE,
.bdrv_create = raw_create,
.bdrv_has_zero_init = bdrv_has_zero_init_1,
.bdrv_co_get_block_status = raw_co_get_block_status,
+ .bdrv_co_write_zeroes = raw_co_write_zeroes,
.bdrv_aio_readv = raw_aio_readv,
.bdrv_aio_writev = raw_aio_writev,
.bdrv_truncate = raw_truncate,
.bdrv_getlength = raw_getlength,
+ .bdrv_get_info = raw_get_info,
.bdrv_get_allocated_file_size
= raw_get_allocated_file_size,
cb, opaque, QEMU_AIO_DISCARD|QEMU_AIO_BLKDEV);
}
+static coroutine_fn int hdev_co_write_zeroes(BlockDriverState *bs,
+ int64_t sector_num, int nb_sectors, BdrvRequestFlags flags)
+{
+ BDRVRawState *s = bs->opaque;
+ int rc;
+
+ rc = fd_open(bs);
+ if (rc < 0) {
+ return rc;
+ }
+ if (!(flags & BDRV_REQ_MAY_UNMAP)) {
+ return paio_submit_co(bs, s->fd, sector_num, NULL, nb_sectors,
+ QEMU_AIO_WRITE_ZEROES|QEMU_AIO_BLKDEV);
+ } else if (s->discard_zeroes) {
+ return paio_submit_co(bs, s->fd, sector_num, NULL, nb_sectors,
+ QEMU_AIO_DISCARD|QEMU_AIO_BLKDEV);
+ }
+ return -ENOTSUP;
+}
+
static int hdev_create(const char *filename, QEMUOptionParameter *options,
Error **errp)
{
.bdrv_reopen_abort = raw_reopen_abort,
.bdrv_create = hdev_create,
.create_options = raw_create_options,
+ .bdrv_co_write_zeroes = hdev_co_write_zeroes,
.bdrv_aio_readv = raw_aio_readv,
.bdrv_aio_writev = raw_aio_writev,
.bdrv_truncate = raw_truncate,
.bdrv_getlength = raw_getlength,
+ .bdrv_get_info = raw_get_info,
.bdrv_get_allocated_file_size
= raw_get_allocated_file_size,
projects / qemu.git / blobdiff