From 8e9dbfff17f29533ad03f3afd25ada8cb6bf1e2b Mon Sep 17 00:00:00 2001 From: Eric Anholt Date: Tue, 27 Jan 2015 16:22:54 -0800 Subject: [PATCH] nir: Conditionalize the POW reconstruction on shader compiler options. Mesa has a shader compiler struct flagging whether GLSL IR's opt_algebraic and other passes should try and generate certain types of opcodes or patterns. Extend that to NIR by defining our own struct, which is automatically generated from the Mesa struct in glsl_to_nir and provided directly by the driver in TGSI-to-NIR. v2: Split out the previous two prep patches. v3: Rebase to master (no TGSI->NIR present) Reviewed-by: Kenneth Graunke (v2) --- src/glsl/nir/glsl_to_nir.cpp | 3 +++ src/glsl/nir/nir.h | 1 + src/glsl/nir/nir_opt_algebraic.py | 4 ++-- 3 files changed, 6 insertions(+), 2 deletions(-) diff --git a/src/glsl/nir/glsl_to_nir.cpp b/src/glsl/nir/glsl_to_nir.cpp index bc43a75..544d0d9 100644 --- a/src/glsl/nir/glsl_to_nir.cpp +++ b/src/glsl/nir/glsl_to_nir.cpp @@ -142,6 +142,9 @@ glsl_to_nir(exec_list *ir, _mesa_glsl_parse_state *state, nir_shader_compiler_options *new_options = rzalloc(ctx, nir_shader_compiler_options); options = gl_options->NirOptions = new_options; + + if (gl_options->EmitNoPow) + new_options->lower_fpow = true; } else { options = gl_options->NirOptions; } diff --git a/src/glsl/nir/nir.h b/src/glsl/nir/nir.h index d5253c4..bcc51e8 100644 --- a/src/glsl/nir/nir.h +++ b/src/glsl/nir/nir.h @@ -1327,6 +1327,7 @@ typedef struct nir_function { exec_list_get_head(&(func)->overload_list), node) typedef struct nir_shader_compiler_options { + bool lower_fpow; } nir_shader_compiler_options; typedef struct nir_shader { diff --git a/src/glsl/nir/nir_opt_algebraic.py b/src/glsl/nir/nir_opt_algebraic.py index 83223f7..76917f5 100644 --- a/src/glsl/nir/nir_opt_algebraic.py +++ b/src/glsl/nir/nir_opt_algebraic.py @@ -118,8 +118,8 @@ optimizations = [ (('fexp', ('flog', a)), a), # e^ln(a) = a (('flog2', ('fexp2', a)), a), # lg2(2^a) = a (('flog', ('fexp', a)), a), # ln(e^a) = a - (('fexp2', ('fmul', ('flog2', a), b)), ('fpow', a, b)), # 2^(lg2(a)*b) = a^b - (('fexp', ('fmul', ('flog', a), b)), ('fpow', a, b)), # e^(ln(a)*b) = a^b + (('fexp2', ('fmul', ('flog2', a), b)), ('fpow', a, b), '!options->lower_fpow'), # 2^(lg2(a)*b) = a^b + (('fexp', ('fmul', ('flog', a), b)), ('fpow', a, b), '!options->lower_fpow'), # e^(ln(a)*b) = a^b (('fpow', a, 1.0), a), (('fpow', a, 2.0), ('fmul', a, a)), (('fpow', 2.0, a), ('fexp2', a)), -- 2.7.4