[PATCH 1/5] d3d9/tests: Test GetData behavior with various size values. (try 2)

[Henri Verbeet]

On 2 May 2014 16:30, Matteo Bruni <mbruni at codeweavers.com> wrote:
> +    memset(&data, 0xff, sizeof(data));
This is a bit of an odd way to write "data = ~0u;".

> +    ok(*(WORD *)&data == 1 || broken(*(WORD *)&data != 1),
> +            "Occlusion query returned an unexpected result (0x%.8x).\n", data);
> +    todo_wine ok(*((WORD *)&data + 1) == 0xffff,
> +            "data was modified outside of the expected size (0x%.8x).\n", data);
I think this is a bit awkward, perhaps it makes more sense the declare
"data" as something like "WORD data[4];". Also, can this really ever
return 0 for the query data? Would it help to draw a different
primitive type, instead of a tiny point? (Because, well, point
rasterization is complicated.)