On Mon, 7 Dec 2020 12:35:14 +0200 Pekka Paalanen ppaalanen@gmail.com wrote:
On Mon, 7 Dec 2020 01:08:58 -0800 James Park james.park@lagfreegames.com wrote:
I'm not completely sure what you're saying, but doesn't drm_base_types.h (now drm_basic_types.h) make things robust to header order? The patch is in another email. You can also see it here: https://github.com/jpark37/linux/commit/0cc8ae750bfd9eab7e31c7de6aa84f24682f...
If that is robust (I don't know if it is, I haven't checked), then why do you have #ifdef DRM_FOURCC_STANDALONE in it at all?
Like Simon said:
On Mon, 07 Dec 2020 10:02:36 +0000 Simon Ser contact@emersion.fr wrote:
In my compositors I'd like to globally define DRM_FOURCC_STANDALONE (to make sure I don't miss any #define) but I still may include drm.h in the same files as well.
If any project #defines it globally, then what good does checking for it do? Why not assume that everyone will always want what DRM_FOURCC_STANDALONE would bring?
Sorry! Now I got it.
DRM_FOURCC_STANDALONE is a promise that the user is not relying on drm_foucc.h to pull in drm.h. Nothing else. That's fine.
But then, the code in the header should be literally
#ifndef DRM_FOURCC_STANDALONE #include "drm.h" #endif
without an #else branch.
Thanks, pq