freedreno/ir3: convert scheduler back to recursive algo

I've played with a few different approaches to tweak instruction
priority according to how much they increase/decrease register pressure,
etc.  But nothing seems to change the fact that compared to original
(pre-multiple-block-support) scheduler, in some edge cases we are
generating shaders w/ 5-6x higher register usage.

The problem is that the priority queue approach completely looses the
dependency between instructions, and ends up scheduling all paths at the
same time.

Original reason for switching was that recursive approach relied on
starting from the shader outputs array.  But we can achieve more or less
the same thing by starting from the depth-sorted list.

shader-db results:

total instructions in shared programs:          113350 -> 105183 (-7.21%)
total dwords in shared programs:                219328 -> 211168 (-3.72%)
total full registers used in shared programs:   7911 -> 7383 (-6.67%)
total half registers used in shader programs:   109 -> 109 (0.00%)
total const registers used in shared programs:  21294 -> 21294 (0.00%)

                 half       full      const      instr     dwords
    helped           0         322           0         711         215
      hurt           0         163           0          38           4

The shaders hurt tend to gain a register or two.  While there are also a
lot of helped shaders that only loose a register or two, the more
complex ones tend to loose significanly more registers used.  In some
more extreme cases, like glsl-fs-convolution-1.shader_test it is more
like 7 vs 34 registers!

Signed-off-by: Rob Clark <[email protected]>

1 files changed, 4 insertions, 0 deletions

diff --git a/src/gallium/drivers/freedreno/ir3/ir3.h b/src/gallium/drivers/freedreno/ir3/ir3.h
index 89b93105cbc..62d14a0ae37 100644
--- a/src/gallium/drivers/freedreno/ir3/ir3.h
+++ b/src/gallium/drivers/freedreno/ir3/ir3.h
@@ -255,6 +255,10 @@ struct ir3_instruction {
 		};
 	};
 
+	/* used for per-pass extra instruction data.
+	 */
+	void *data;
+
 	/* Used during CP and RA stages.  For fanin and shader inputs/
 	 * outputs where we need a sequence of consecutive registers,
 	 * keep track of each src instructions left (ie 'n-1') and right