You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I pulled Rao's work on the dynamically allocated buffer.
But given all the discussion around a smart build system, would it make sense to have CMake control not only the value of R3D_MAX_VERTS, but also whether the code uses static vs dynamic allocation? We could make the API consistent for both cases, so that one could switch at will (i.e., r3d_init_poly() would really do nothing in the static case, but could be left in for consistency)
The text was updated successfully, but these errors were encountered:
If the language were C++, that could be controlled by template parameters as well. Not saying that's the best idea, depends on everyone's willingness to use C++.
That would take some work (for which I don't have time right now). Unless others see a need for actually having this be a thing, I am fine with letting it die. It was an instructive experiment.
@ibaned @shevitz @raovgarimella @MackKenamond
I am opening this for a new discussion thread.
I pulled Rao's work on the dynamically allocated buffer.
But given all the discussion around a smart build system, would it make sense to have CMake control not only the value of R3D_MAX_VERTS, but also whether the code uses static vs dynamic allocation? We could make the API consistent for both cases, so that one could switch at will (i.e., r3d_init_poly() would really do nothing in the static case, but could be left in for consistency)
The text was updated successfully, but these errors were encountered: