On Don, 2012-07-19 at 10:38 -0400, Kristian Høgsberg wrote: > On Thu, Jul 19, 2012 at 8:39 AM, Michel Dänzer <mic...@daenzer.net> wrote: > > On Don, 2012-07-19 at 01:15 -0600, Scott Moreau wrote: > >> gbm_bo_get_pitch was renamed to gbm_bo_get_stride. eglkms fails > >> to build against mesa master without this. > >> --- > >> src/egl/opengl/eglkms.c | 2 +- > >> 1 file changed, 1 insertion(+), 1 deletion(-) > >> > >> diff --git a/src/egl/opengl/eglkms.c b/src/egl/opengl/eglkms.c > >> index 1006317..6ba0f12 100644 > >> --- a/src/egl/opengl/eglkms.c > >> +++ b/src/egl/opengl/eglkms.c > >> @@ -239,7 +239,7 @@ int main(int argc, char *argv[]) > >> > >> bo = gbm_surface_lock_front_buffer(gs); > >> handle = gbm_bo_get_handle(bo).u32; > >> - stride = gbm_bo_get_pitch(bo); > >> + stride = gbm_bo_get_stride(bo); > > > > This is a libgbm ABI break, which requires one of the following: > > > > 1. Bumping the libgbm ABI major version. > > 2. Keeping gbm_bo_get_pitch as an alias of gbm_bo_get_stride (at > > least until 1. happens for a better reason). > > 3. Reverting the symbol rename. > > We have not committed to a stable API for GBM yet. It's unfortunate > and confusing that an experimental library is released alongside > components with very stable and mature APIs, but I don't see any other > way to develop this. We're expecting to release Wayland and Weston > 1.0 this year (in a couple of months) and at that point we'll lock the > API. > > I can bump the libgbm .so version.
I think it's necessary[0]. Otherwise, it won't be possible to use drivers from Mesa 8.1 and 8.0 in parallel, will it? [0] If the other options aren't feasible — it seems rather annoying that Mesa demos can't (trivially) build against Mesa 8.1 and older at the same time, for such a minor change. -- Earthling Michel Dänzer | http://www.amd.com Libre software enthusiast | Debian, X and DRI developer _______________________________________________ mesa-dev mailing list mesa-dev@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/mesa-dev
