The ir_tex opcode turns into a sample or sample_c message, which will try to compute derivatives to determine the lod. This produces garbage for non-fragment shaders where the sample coordinates don't correspond to subspans.
We fix this by rewriting the opcode from ir_tex to ir_txl and setting the lod to 0. https://bugs.freedesktop.org/show_bug.cgi?id=89457 Signed-off-by: Kristian Høgsberg <k...@bitplanet.net> --- src/mesa/drivers/dri/i965/brw_fs_visitor.cpp | 9 +++++++++ 1 file changed, 9 insertions(+) diff --git a/src/mesa/drivers/dri/i965/brw_fs_visitor.cpp b/src/mesa/drivers/dri/i965/brw_fs_visitor.cpp index 0049b2d..4e99366 100644 --- a/src/mesa/drivers/dri/i965/brw_fs_visitor.cpp +++ b/src/mesa/drivers/dri/i965/brw_fs_visitor.cpp @@ -1839,6 +1839,15 @@ fs_visitor::emit_texture_gen7(ir_texture_opcode op, fs_reg dst, offset_value.file != BAD_FILE && offset_value.file != IMM; bool coordinate_done = false; + /* The sampler can only meaningfully compute LOD for fragment shader + * messages. For all other stages, we change the opcode to ir_txl and + * hardcode the LOD to 0. + */ + if (stage != MESA_SHADER_FRAGMENT && op == ir_tex) { + op = ir_txl; + lod = fs_reg(0.0f); + } + /* Set up the LOD info */ switch (op) { case ir_tex: -- 2.1.0 _______________________________________________ mesa-dev mailing list mesa-dev@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/mesa-dev
