[Spice-devel] [PATCH] cursor: fix wrong logic when initializing the channel

[Frediano Ziglio]

> 
> It's a regression introduced by commit e601e920bd5. The logic error was
> introduced when trying to achieve the following code[0]. but rewritten
> to prefer an early return.
> 
> [0]:
> if (cursor_is_connected(worker)
>     && !COMMON_CHANNEL(worker->cursor_channel)->during_target_migrate) {
>         red_channel_pipes_add_type(RED_CHANNEL(worker->cursor_channel),
>                                    PIPE_ITEM_TYPE_CURSOR_INIT);
> }
> 
> Signed-off-by: Fabiano Fidêncio <fidencio at redhat.com>
> ---
>  server/cursor-channel.c | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/server/cursor-channel.c b/server/cursor-channel.c
> index ce360e1..23d5b5a 100644
> --- a/server/cursor-channel.c
> +++ b/server/cursor-channel.c
> @@ -545,7 +545,7 @@ void cursor_channel_init(CursorChannel *cursor,
> CursorChannelClient *client)
>      spice_return_if_fail(cursor);
>  
>      if (red_channel_is_connected(&cursor->common.base)
> -        || COMMON_CHANNEL(cursor)->during_target_migrate) {
> +        && COMMON_CHANNEL(cursor)->during_target_migrate) {
>          spice_debug("during_target_migrate: skip init");
>          return;
>      }
> --
> 2.5.0

The original code was

    if (cursor_is_connected(worker)
        && !COMMON_CHANNEL(worker->cursor_channel)->during_target_migrate) {
        red_channel_pipes_add_type(RED_CHANNEL(worker->cursor_channel),
                                   PIPE_ITEM_TYPE_CURSOR_INIT);
    }

which was wrongly negated as

    if (red_channel_is_connected(&cursor->common.base)
        || COMMON_CHANNEL(cursor)->during_target_migrate) {
        spice_debug("during_target_migrate: skip init");
        return;
    }

I think the right code should be

    if (!red_channel_is_connected(&cursor->common.base)
        || COMMON_CHANNEL(cursor)->during_target_migrate) {
        spice_debug("during_target_migrate: skip init");
        return;
    }

Frediano