From 65765c9f2c5d3568608bde57db0bf44d6b724755 Mon Sep 17 00:00:00 2001 From: Brian Paul Date: Tue, 29 Sep 2009 18:57:13 -0600 Subject: glsl: rewrite sqrt(x) intrinsic to handle x=0 Since sqrt() is basically implemented in terms of RSQ/RCP we'll do a divide by zero if x=0 and wind up with unpredictable results. Now use CMP instruction to test for x<=0 and return zero in that case. --- .../shader/slang/library/slang_common_builtin.gc | 76 ++++++++++++---------- 1 file changed, 42 insertions(+), 34 deletions(-) (limited to 'src/mesa/shader/slang') diff --git a/src/mesa/shader/slang/library/slang_common_builtin.gc b/src/mesa/shader/slang/library/slang_common_builtin.gc index 9764fc25b0..56de47ee8d 100644 --- a/src/mesa/shader/slang/library/slang_common_builtin.gc +++ b/src/mesa/shader/slang/library/slang_common_builtin.gc @@ -602,42 +602,50 @@ vec4 exp2(const vec4 a) float sqrt(const float x) { + const float nx = -x; float r; __asm float_rsq r, x; - __asm float_rcp __retVal, r; -} - -vec2 sqrt(const vec2 v) -{ - float r; - __asm float_rsq r, v.x; - __asm float_rcp __retVal.x, r; - __asm float_rsq r, v.y; - __asm float_rcp __retVal.y, r; -} - -vec3 sqrt(const vec3 v) -{ - float r; - __asm float_rsq r, v.x; - __asm float_rcp __retVal.x, r; - __asm float_rsq r, v.y; - __asm float_rcp __retVal.y, r; - __asm float_rsq r, v.z; - __asm float_rcp __retVal.z, r; -} - -vec4 sqrt(const vec4 v) -{ - float r; - __asm float_rsq r, v.x; - __asm float_rcp __retVal.x, r; - __asm float_rsq r, v.y; - __asm float_rcp __retVal.y, r; - __asm float_rsq r, v.z; - __asm float_rcp __retVal.z, r; - __asm float_rsq r, v.w; - __asm float_rcp __retVal.w, r; + __asm float_rcp r, r; + __asm vec4_cmp __retVal, nx, r, 0.0; +} + +vec2 sqrt(const vec2 x) +{ + const vec2 nx = -x, zero = vec2(0.0); + vec2 r; + __asm float_rsq r.x, x.x; + __asm float_rsq r.y, x.y; + __asm float_rcp r.x, r.x; + __asm float_rcp r.y, r.y; + __asm vec4_cmp __retVal, nx, r, zero; +} + +vec3 sqrt(const vec3 x) +{ + const vec3 nx = -x, zero = vec3(0.0); + vec3 r; + __asm float_rsq r.x, x.x; + __asm float_rsq r.y, x.y; + __asm float_rsq r.z, x.z; + __asm float_rcp r.x, r.x; + __asm float_rcp r.y, r.y; + __asm float_rcp r.z, r.z; + __asm vec4_cmp __retVal, nx, r, zero; +} + +vec4 sqrt(const vec4 x) +{ + const vec4 nx = -x, zero = vec4(0.0); + vec4 r; + __asm float_rsq r.x, x.x; + __asm float_rsq r.y, x.y; + __asm float_rsq r.z, x.z; + __asm float_rsq r.w, x.w; + __asm float_rcp r.x, r.x; + __asm float_rcp r.y, r.y; + __asm float_rcp r.z, r.z; + __asm float_rcp r.w, r.w; + __asm vec4_cmp __retVal, nx, r, zero; } -- cgit v1.2.3