On 9/18/20 4:31 AM, Alexandre Julliard wrote:
Zebediah Figura <z.figura12(a)gmail.com> writes:
+#include <winioctl.h> + +#define IOCTL_AFD_CREATE CTL_CODE(FILE_DEVICE_NETWORK, 0, METHOD_BUFFERED, FILE_WRITE_ACCESS) + +struct afd_create_params +{ + int family, type, protocol; + unsigned int flags; +};
Do we really need yet another set of Wine-specific ioctls?
I assumed that at least it'd be preferable to a set of Wine-specific server calls; is this not the case? In particular it lets us use the ntdll infrastructure for asynchronous I/O, instead of half-duplicating it in ws2_32. I don't know what would be better here. Unless the proposal is to use native ioctls, but I couldn't easily find them and I don't know how to safely reverse-engineer them.
Note that (FILE_DEVICE_NETWORK,0) is already used for the documented IOCTL_TCP_QUERY_INFORMATION_EX, it may be better to avoid it.