-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Reimar Döffinger wrote:
On Mon, Dec 08, 2008 at 09:14:59AM +0200, Georgi Petrov wrote:
I vote for inclusion now. If there are some indent / spacing issues, I'll fix them after the patch is commited. We need the OSD.
Have you tested it? Also there is still almost the same buffer flow problem: the allocated texture may be smaller than the arguments to vo_draw_text, allowing for a buffer overflow.
Yeh, I see the issue. I'll fix this.
There are also minor issues like the expression to select non-power-of-two texture is extremely overcomplicated (e.g. !(a ^ b) is the same as a == b) and also wrong. I guess I can go over it myself and fix it when I got a bit of time.
a = D3DPTEXTURECAPS_POW2 b = D3DPTEXTURECAPS_NONPOW2CONDITIONAL a b a ^b !(a^b) (a==b) 0 0 0 1 1 0 1 1 0 0 1 0 1 0 0 1 1 0 1 1 The D3D docs say the if a & b are neither defined then you assume the device can support non power2 textures. If they are both set then also this is true. If a then power2 dimensions, but possibly not square. The other case is invalid. I think the logic is correct, but as you said it could be simplified by the use of (a==b) Jim -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (MingW32) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkk8/4IACgkQhrNWoHjgI1DfcQCgqmJITjU3wpJVnMZkjz/0qsyJ JeYAnRG1RnDcZOSwmjhaFbnZFBrd+kdx =lOj3 -----END PGP SIGNATURE-----