nir: Implement optional b2f->iand lowering

This pass is required by the Midgard compiler; our instruction set uses
NIR-style booleans (~0 for true) but lacks a dedicated b2f instruction.
Normally, this lowering pass would be implemented in a backend-specific
algebraic pass, but this conflicts with the existing iand->b2f pass in
nir_opt_algebraic.py, hanging the compiler. This patch thus makes the
existing pass optional (default on -- all other backends should remain
unaffected), adding an optional pass for lowering the opposite
direction.

v2: Defer lowering until late algebraic optimisations to allow
optimising the b2f instruction itself.

Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Reviewed-by: Matt Turner <mattst88@gmail.com>
Reviewed-by: Christian Gmeiner <christian.gmeiner@gmail.com>

diff --git a/src/compiler/nir/nir.h b/src/compiler/nir/nir.h
index a379928cdcd9cecadde9f00a6d954a75ccce2c4d..f6086bd6c058a40e07ec00839a5491493dbc6459 100644 (file)
--- a/src/compiler/nir/nir.h
+++ b/src/compiler/nir/nir.h
@@ -1894,6 +1894,9 @@ typedef struct nir_shader_compiler_options {
    /** enables rules to lower idiv by power-of-two: */
    bool lower_idiv;
 
+   /* lower b2f to iand */
+   bool lower_b2f;
+
    /* Does the native fdot instruction replicate its result for four
     * components?  If so, then opt_algebraic_late will turn all fdotN
     * instructions into fdot_replicatedN instructions.

diff --git a/src/compiler/nir/nir_opt_algebraic.py b/src/compiler/nir/nir_opt_algebraic.py
index 96232f0e549c5707b5c4072ac4464cff5cc676be..909ea3daf428c362357e0d0fe7be14f3359b0d5f 100644 (file)
--- a/src/compiler/nir/nir_opt_algebraic.py
+++ b/src/compiler/nir/nir_opt_algebraic.py
@@ -321,7 +321,7 @@ optimizations = [
    (('imul', ('b2i', a), ('b2i', b)), ('b2i', ('iand', a, b))),
    (('fmul', ('b2f', a), ('b2f', b)), ('b2f', ('iand', a, b))),
    (('fsat', ('fadd', ('b2f', a), ('b2f', b))), ('b2f', ('ior', a, b))),
-   (('iand', 'a@bool', 1.0), ('b2f', a)),
+   (('iand', 'a@bool', 1.0), ('b2f', a), '!options->lower_b2f'),
    # True/False are ~0 and 0 in NIR.  b2i of True is 1, and -1 is ~0 (True).
    (('ineg', ('b2i@32', a)), a),
    (('flt', ('fneg', ('b2f', a)), 0), a), # Generated by TGSI KILL_IF.
@@ -724,6 +724,9 @@ late_optimizations = [
    # we do these late so that we don't get in the way of creating ffmas
    (('fmin', ('fadd(is_used_once)', '#c', a), ('fadd(is_used_once)', '#c', b)), ('fadd', c, ('fmin', a, b))),
    (('fmax', ('fadd(is_used_once)', '#c', a), ('fadd(is_used_once)', '#c', b)), ('fadd', c, ('fmax', a, b))),
+
+   # Lowered for backends without a dedicated b2f instruction
+   (('b2f@32', a), ('iand', a, 1.0), 'options->lower_b2f'),
 ]
 
 print nir_algebraic.AlgebraicPass("nir_opt_algebraic", optimizations).render()