If the guest tries to write data that results on the allocation of a new cluster, instead of writing the guest data first and then the data from the COW regions, write everything together using one single I/O operation.This can improve the write performance by 25% or more, depending on several factors such as the media type, the cluster size and the I/O request size. Signed-off-by: Alberto Garcia <be...@igalia.com> --- block/qcow2-cluster.c | 34 ++++++++++++++++++++++-------- block/qcow2.c | 58 ++++++++++++++++++++++++++++++++++++++++++--------- block/qcow2.h | 7 +++++++ 3 files changed, 80 insertions(+), 19 deletions(-) - /* And now we can write everything */ - qemu_iovec_reset(&qiov); - qemu_iovec_add(&qiov, start_buffer, start->nb_bytes); - ret = do_perform_cow_write(bs, m->alloc_offset, start->offset, &qiov); - if (ret < 0) { - goto fail; + /* And now we can write everything. If we have the guest data we + * can write everything in one single operation */ + if (m->data_qiov) { + qemu_iovec_reset(&qiov); + qemu_iovec_add(&qiov, start_buffer, start->nb_bytes); + qemu_iovec_concat(&qiov, m->data_qiov, 0, data_bytes); + qemu_iovec_add(&qiov, end_buffer, end->nb_bytes);
Can it be a problem if (m->data_qiov->niov == IOV_MAX)? We had to add merge-iovecs code for the case (maybe there's better solution?) Also, will this work if allocation is split into several l2metas? e.g. one has cow_start.nb_bytes and another has cow_end.nb_bytes /Anton
