[Spice-devel] [RFC v4 57/62] server/reds: add RedsState.allow_multiple_clients (temp - add accessors too)

[Marc-André Lureau]

On Tue, Apr 26, 2011 at 12:55 PM, Alon Levy <alevy at redhat.com> wrote:
> ---
>  server/reds.c |    6 +++++-
>  1 files changed, 5 insertions(+), 1 deletions(-)
>
> diff --git a/server/reds.c b/server/reds.c
> index 0ce6f1c..dc73202 100644
> --- a/server/reds.c
> +++ b/server/reds.c
> @@ -228,6 +228,7 @@ typedef struct RedsState {
>     RedsStatValue roundtrip_stat;
>  #endif
>     int peer_minor_version;
> +    int allow_multiple_clients;
>  } RedsState;
>
>  static RedsState *reds = NULL;
> @@ -1498,7 +1499,9 @@ static void reds_handle_main_link(RedLinkInfo *link)
>
>     red_printf("");
>     link_mess = link->link_mess;
> -    reds_disconnect();
> +    if (!reds->allow_multiple_clients) {
> +        reds_disconnect();
> +    }
>
>     if (link_mess->connection_id == 0) {
>         reds_send_link_result(link, SPICE_LINK_ERR_OK);
> @@ -3519,6 +3522,7 @@ static int do_spice_init(SpiceCoreInterface *core_interface)
>  #endif
>
>     reds->mouse_mode = SPICE_MOUSE_MODE_SERVER;
> +    reds->allow_multiple_clients = 1;

Without accessors (and qemu option), having an environment variable
would be helpful. Something like:

reds->allow_multiple_clients = getenv("SPICE_ALLOW_MC") != NULL.

I am also wondering if applying this patch earlier in the patch series
would help to exercise iteratively the MC code before all the changes
are applied (do you see what I mean?).

>     atexit(reds_exit);
>     return 0;
>
> --
> 1.7.4.4
>
> _______________________________________________
> Spice-devel mailing list
> Spice-devel at lists.freedesktop.org
> http://lists.freedesktop.org/mailman/listinfo/spice-devel
>



-- 
Marc-André Lureau