I guess the main block code is not as defensive as I thought it was. This patch uses qemu_memalign to allocate the buffers for IO so that you don't get errors when using O_DIRECT.
It applies on top of my previous patch to introduce copies in virtio-blk. Signed-off-by: Anthony Liguori <[EMAIL PROTECTED]> diff --git a/qemu/hw/virtio-blk.c b/qemu/hw/virtio-blk.c index 2ea5669..669e55f 100644 --- a/qemu/hw/virtio-blk.c +++ b/qemu/hw/virtio-blk.c @@ -174,7 +174,7 @@ static void virtio_blk_handle_output(VirtIODevice *vdev, VirtQueue *vq) for (i = 1; i < req->elem.out_num; i++) req->size += req->elem.out_sg[i].iov_len; - req->buffer = qemu_malloc(req->size); + req->buffer = qemu_memalign(512, req->size); if (req->buffer == NULL) { qemu_free(req); break; @@ -203,7 +203,7 @@ static void virtio_blk_handle_output(VirtIODevice *vdev, VirtQueue *vq) for (i = 0; i < req->elem.in_num - 1; i++) req->size += req->elem.in_sg[i].iov_len; - req->buffer = qemu_malloc(req->size); + req->buffer = qemu_memalign(512, req->size); if (req->buffer == NULL) { qemu_free(req); break; -- To unsubscribe from this list: send the line "unsubscribe kvm" in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html