Previously, we allocated space in brw_vs_prog_data's params and pull_params arrays for MAX_CLIP_PLANES vec4s---even when it wasn't necessary.
On a 64-bit architecture, this used 0.5 kB of space (8 clip planes * 4 floats per plane * 8 bytes per float pointer * 2 arrays of pointers = 512 bytes). Since this cost is per-vertex shader, it added up. Conveniently, we already store the number of clip plane constants in the program key. By using that, we can allocate the exact amount of space needed. For the common case where user clipping is disabled, this means 0 bytes. While we're here, mention exactly what code requires this extra space, since it wasn't obvious. Signed-off-by: Kenneth Graunke <kenn...@whitecape.org> --- src/mesa/drivers/dri/i965/brw_vs.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) No Piglit regressions on Ivybridge. Portal still works as well (it uses a lot of user clipping). diff --git a/src/mesa/drivers/dri/i965/brw_vs.c b/src/mesa/drivers/dri/i965/brw_vs.c index 6a67bc4..a5a9fb1 100644 --- a/src/mesa/drivers/dri/i965/brw_vs.c +++ b/src/mesa/drivers/dri/i965/brw_vs.c @@ -234,8 +234,10 @@ do_vs_prog(struct brw_context *brw, } else { param_count = vp->program.Base.Parameters->NumParameters * 4; } - /* We also upload clip plane data as uniforms */ - param_count += MAX_CLIP_PLANES * 4; + /* vec4_visitor::setup_uniform_clipplane_values() also uploads user clip + * planes as uniforms. + */ + param_count += c.key.base.nr_userclip_plane_consts * 4; prog_data.base.param = rzalloc_array(NULL, const float *, param_count); prog_data.base.pull_param = rzalloc_array(NULL, const float *, param_count); -- 1.8.3.4 _______________________________________________ mesa-dev mailing list mesa-dev@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/mesa-dev