I had my understanding of this bit flipped. We're using the full register
space, so we need to say so.
exec->shader_rec_v += roundup(packet_size, 16);
exec->shader_rec_size -= packet_size;
+ if (!(*(uint16_t *)pkt_u & VC4_SHADER_FLAG_FS_SINGLE_THREAD)) {
+ DRM_ERROR("Multi-threaded fragment shaders not supported.\n");
+ return -EINVAL;
+ }
+
for (i = 0; i < shader_reloc_count; i++) {
if (src_handles[i] > exec->bo_count) {
DRM_ERROR("Shader handle %d too big\n", src_handles[i]);
/* VC4_DIRTY_PRIM_MODE | VC4_DIRTY_RASTERIZER */
cl_u16(&shader_rec,
VC4_SHADER_FLAG_ENABLE_CLIPPING |
+ VC4_SHADER_FLAG_FS_SINGLE_THREAD |
((info->mode == PIPE_PRIM_POINTS &&
vc4->rasterizer->base.point_size_per_vertex) ?
VC4_SHADER_FLAG_VS_POINT_SIZE : 0));