Rémi Bernon rbernon@codeweavers.com writes:
On 5/14/21 9:38 PM, Alexandre Julliard wrote:
Rémi Bernon rbernon@codeweavers.com writes:
@@ -302,6 +302,9 @@ union rawinput unsigned int param; /* rawinput message param */ unsigned short usage_page;/* HID usage page */ unsigned short usage; /* HID usage */
unsigned int length; /* HID report length */int __pad;/* followed by length bytes of HID report data */That's not true in server requests, I don't think we should pretend that it is. The report data should be declared as a separate VARARG field in the request, and use a separate pointer where necessary. The length can probably be stored separately as well.
Yes, I think I had it like that at some point, it probably got lost somewhere.
I understand that this is only for the send_hardware_message request? I don't see how to do that for get_message request or get_rawinput_buffer, where the replied message data (or message buffer) is already a VARARG, and for which we reply a struct hardware_msg_data, followed by extra bytes for the hid report?
Yes, in the message data it would need to be packed in with the rest. But even there it would follow the hardware_msg_data struct, not the rawinput.hid field (right now it's the same, but that could change if other fields are added).