shader: introduce a crude setting for adjusting the maximum workgroup size

Both the Vulkan and OpenGL ES spec allow implementations to limit workgroups to
128 threads. Add a LG_WG_FACTOR setting for easy switching between 128 and 256
threads, with 256 being kept as the default setting.

Manually tested that LG_WG_FACTOR = 0 (128 threads) works as expected.

Signed-off-by: Elias Naur <mail@eliasnaur.com>

piet-gpu/shader/backdrop.comp

@@ -15,7 +15,7 @@
 
 #include "setup.h"
 
-#define LG_BACKDROP_WG 8
+#define LG_BACKDROP_WG (7 + LG_WG_FACTOR)
 #define BACKDROP_WG (1 << LG_BACKDROP_WG)
 
 layout(local_size_x = BACKDROP_WG, local_size_y = 1) in;

piet-gpu/shader/coarse.comp

@@ -41,7 +41,7 @@ layout(set = 0, binding = 4) buffer PtclBuf {
 #include "tile.h"
 #include "ptcl.h"
 
-#define LG_N_PART_READ 8
+#define LG_N_PART_READ (7 + LG_WG_FACTOR)
 #define N_PART_READ (1 << LG_N_PART_READ)
 
 shared uint sh_elements[N_TILE];

piet-gpu/shader/setup.h

@@ -3,6 +3,11 @@
 // Much of this will be made dynamic in various ways, but for now it's easiest
 // to hardcode and keep all in one place.
 
+// A LG_WG_FACTOR of n scales workgroup sizes by 2^n. Use 0 for a
+// maximum workgroup size of 128, or 1 for a maximum size of 256.
+#define LG_WG_FACTOR 1
+#define WG_FACTOR (1<<LG_WG_FACTOR)
+
 // TODO: compute all these
 
 #define WIDTH_IN_TILES 128
@@ -17,7 +22,7 @@
 // of that kernel is equal to the number of bins, but should probably
 // be more flexible (it's 512 in the K&L paper).
 #define N_TILE_X 16
-#define N_TILE_Y 16
+#define N_TILE_Y (8 * WG_FACTOR)
 #define N_TILE (N_TILE_X * N_TILE_Y)
-#define LG_N_TILE 8
+#define LG_N_TILE (7 + LG_WG_FACTOR)
 #define N_SLICE (N_TILE / 32)

piet-gpu/shader/tile_alloc.comp

@@ -5,7 +5,7 @@
 
 #include "setup.h"
 
-#define LG_TILE_ALLOC_WG 8
+#define LG_TILE_ALLOC_WG (7 + LG_WG_FACTOR)
 #define TILE_ALLOC_WG (1 << LG_TILE_ALLOC_WG)
 
 layout(local_size_x = TILE_ALLOC_WG, local_size_y = 1) in;