From: Iago Toral Quiroga <itoral@igalia.com>
Date: Wed, 18 Apr 2018 07:36:41 +0000 (+0200)
Subject: compiler/spirv: implement 16-bit atan
X-Git-Url: https://git.libre-soc.org/?a=commitdiff_plain;h=1c8de08ec92bff7c44f9e4732c591b05ba1f2d76;p=mesa.git

compiler/spirv: implement 16-bit atan

v2:
 - use nir_fadd_imm and nir_fmul_imm helpers (Jason)
 - rebased on top of new sized boolean opcodes
 - use nir_b2f helper

Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
---

diff --git a/src/compiler/spirv/vtn_glsl450.c b/src/compiler/spirv/vtn_glsl450.c
index 9211bca301c..1548c52dbe1 100644
--- a/src/compiler/spirv/vtn_glsl450.c
+++ b/src/compiler/spirv/vtn_glsl450.c
@@ -237,8 +237,10 @@ build_fsum(nir_builder *b, nir_ssa_def **xs, int terms)
 static nir_ssa_def *
 build_atan(nir_builder *b, nir_ssa_def *y_over_x)
 {
+   const uint32_t bit_size = y_over_x->bit_size;
+
    nir_ssa_def *abs_y_over_x = nir_fabs(b, y_over_x);
-   nir_ssa_def *one = nir_imm_float(b, 1.0f);
+   nir_ssa_def *one = nir_imm_floatN_t(b, 1.0f, bit_size);
 
    /*
     * range-reduction, first step:
@@ -265,12 +267,12 @@ build_atan(nir_builder *b, nir_ssa_def *y_over_x)
    nir_ssa_def *x_11 = nir_fmul(b, x_9, x_2);
 
    nir_ssa_def *polynomial_terms[] = {
-      nir_fmul(b, x,    nir_imm_float(b,  0.9999793128310355f)),
-      nir_fmul(b, x_3,  nir_imm_float(b, -0.3326756418091246f)),
-      nir_fmul(b, x_5,  nir_imm_float(b,  0.1938924977115610f)),
-      nir_fmul(b, x_7,  nir_imm_float(b, -0.1173503194786851f)),
-      nir_fmul(b, x_9,  nir_imm_float(b,  0.0536813784310406f)),
-      nir_fmul(b, x_11, nir_imm_float(b, -0.0121323213173444f)),
+      nir_fmul_imm(b, x,     0.9999793128310355f),
+      nir_fmul_imm(b, x_3,  -0.3326756418091246f),
+      nir_fmul_imm(b, x_5,   0.1938924977115610f),
+      nir_fmul_imm(b, x_7,  -0.1173503194786851f),
+      nir_fmul_imm(b, x_9,   0.0536813784310406f),
+      nir_fmul_imm(b, x_11, -0.0121323213173444f),
    };
 
    nir_ssa_def *tmp =
@@ -278,11 +280,8 @@ build_atan(nir_builder *b, nir_ssa_def *y_over_x)
 
    /* range-reduction fixup */
    tmp = nir_fadd(b, tmp,
-                  nir_fmul(b,
-                           nir_b2f32(b, nir_flt(b, one, abs_y_over_x)),
-                           nir_fadd(b, nir_fmul(b, tmp,
-                                                nir_imm_float(b, -2.0f)),
-                                       nir_imm_float(b, M_PI_2f))));
+                  nir_fmul(b, nir_b2f(b, nir_flt(b, one, abs_y_over_x), bit_size),
+                           nir_fadd_imm(b, nir_fmul_imm(b, tmp, -2.0f), M_PI_2f)));
 
    /* sign fixup */
    return nir_fmul(b, tmp, nir_fsign(b, y_over_x));