Anthony PERARD writes ("[Xen-devel] [PATCH V6 3/3] libxl, Introduce a QMP
client"):
> QMP stands for QEMU Monitor Protocol and it is used to query information
> from QEMU or to control QEMU.
> if (dm_starting) {
> + if (dm_info->device_model_version == LIBXL_DEVICE_MODEL_VERSION_QEMU
_XEN) {
> + libxl__qmp_initializations(ctx, domid);
> + }
...
> + flexarray_append(dm_args,
> + libxl__sprintf(gc, "socket,id=libxl-cmd,path=%s/qmp-%d,
server,nowait",
> + libxl_run_dir_path(),
Lines too long, please wrap
> +static void qmp_handle_error_response(libxl__qmp_handler *qmp,
> + const libxl__json_object *resp)
> +{
> + callback_id_pair *pp = qmp_get_callback_from_id(qmp, resp);
...
> + if (pp) {
> + if (pp->id == qmp->wait_for_id) {
> + /* tell that the id have been processed */
> + qmp->wait_for_id = 0;
> + }
> + SIMPLEQ_REMOVE(&qmp->callback_list, pp, callback_id_pair, next);
> + free(pp);
I think this needs to call the callback, or the code which set up the
callback will surely just continue to wait forever.
...
Later: I see that you have a single wait_for_id. What if multiple
callers want to use a single qmp for multiple things ? Or is
"wait_for_id" simply the one that you're waiting on synchronously ?
> +static int qmp_next(libxl__qmp_handler *qmp)
> +{
...
> + ret = select(qmp->qmp_fd + 1, &rfds, NULL, NULL, &timeout);
> + if (ret > 0) {
> + rd = read(qmp->qmp_fd, qmp->buffer, QMP_RECEIVE_BUFFER_SIZE);
> + if (rd > 0) {
> + break;
...
> + s = qmp->buffer;
> + s_end = qmp->buffer + rd;
> + while (s < s_end) {
> + libxl__json_object *o = libxl__json_parse(qmp->ctx, &qmp->yajl_ctx,
> + &s, s_end - s);
This assumes that the response will be received in a single read().
This is not correct. read() may return partial results; it may also
aggregate multiple writes from the sending qemu into a single read()
result.
You need to pull data into the buffer and then test the buffer for
completeness (eg by looking for the cr-lf), and split the buffer up
into packets yourself, and if there are partial packets left over
go round and read again.
Thanks,
Ian.
_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxxxxxxxx
http://lists.xensource.com/xen-devel
|
Home