Attachment 78390 Details for Bug 49713 – glsl: Improve the accuracy of the asin() builtin function.

[patch] glsl: Improve the accuracy of the asin() builtin function.

0001-glsl-Improve-the-accuracy-of-the-asin-builtin-functi.patch (text/plain), 4.30 KB, created by Fabian Bieler on 2013-04-23 17:21:41 UTC

(hide)

Description:

Filename:

MIME Type:

Creator: Fabian Bieler

Created: 2013-04-23 17:21:41 UTC

Size: 4.30 KB

patch

obsolete

>From c62804ed7d17d0c254308a8a9a0ab0f0bc83818e Mon Sep 17 00:00:00 2001
>From: Fabian Bieler <fabianbieler@fastmail.fm>
>Date: Mon, 22 Apr 2013 19:50:51 +0200
>Subject: [PATCH] glsl: Improve the accuracy of the asin() builtin function.
>To: mesa-dev@lists.freedesktop.org
>
>Extended the 3rd degree Series by two elements to 5th degree.
>
>The function is now approximated by:
>asin(x) = sign(x) * (pi/2 - pi/2 * sqrt(1 - |x|) * (1 + |x| * ((1/2 - 2/pi) + |x| * (A + |x| * (B + |x| * (C + |x| * D))))))
>with constants A, B, C, D found by numerical fitting.
>---
> src/glsl/builtins/ir/asin.ir |   80 +++++++++++++++++++++++++++++-------------
> 1 file changed, 56 insertions(+), 24 deletions(-)
>
>diff --git a/src/glsl/builtins/ir/asin.ir b/src/glsl/builtins/ir/asin.ir
>index 45d9e67..4259447 100644
>--- a/src/glsl/builtins/ir/asin.ir
>+++ b/src/glsl/builtins/ir/asin.ir
>@@ -19,12 +19,20 @@
> 		    (constant float (-0.21460183))
> 		    (expression float *
> 		     (expression float abs (var_ref x))
>-                     (expression float +
>-                      (constant float (0.086566724))
>-                      (expression float *
>-                       (expression float abs (var_ref x))
>-                       (constant float (-0.03102955))
>-		     ))))))))))))
>+		     (expression float +
>+		      (constant float (0.08873435))
>+		      (expression float *
>+		       (expression float abs (var_ref x))
>+		       (expression float +
>+			(constant float (-0.047800794))
>+			(expression float *
>+			 (expression float abs (var_ref x))
>+			 (expression float +
>+			  (constant float (0.023164911))
>+			  (expression float *
>+			   (expression float abs (var_ref x))
>+			   (constant float (-0.006116644))
>+   ))))))))))))))))
> 
>    (signature vec2
>      (parameters
>@@ -46,12 +54,20 @@
> 		    (constant float (-0.21460183))
> 		    (expression vec2 *
> 		     (expression vec2 abs (var_ref x))
>-                     (expression vec2 +
>-                      (constant float (0.086566724))
>-                      (expression vec2 *
>-                       (expression vec2 abs (var_ref x))
>-                       (constant float (-0.03102955))
>-		     ))))))))))))
>+		     (expression vec2 +
>+		      (constant float (0.08873435))
>+		      (expression vec2 *
>+		       (expression vec2 abs (var_ref x))
>+		       (expression vec2 +
>+			(constant float (-0.047800794))
>+			(expression vec2 *
>+			 (expression vec2 abs (var_ref x))
>+			 (expression vec2 +
>+			  (constant float (0.023164911))
>+			  (expression vec2 *
>+			   (expression vec2 abs (var_ref x))
>+			   (constant float (-0.006116644))
>+   ))))))))))))))))
> 
>    (signature vec3
>      (parameters
>@@ -73,12 +89,20 @@
> 		    (constant float (-0.21460183))
> 		    (expression vec3 *
> 		     (expression vec3 abs (var_ref x))
>-                     (expression vec3 +
>-                      (constant float (0.086566724))
>-                      (expression vec3 *
>-                       (expression vec3 abs (var_ref x))
>-                       (constant float (-0.03102955))
>-		     ))))))))))))
>+		     (expression vec3 +
>+		      (constant float (0.08873435))
>+		      (expression vec3 *
>+		       (expression vec3 abs (var_ref x))
>+		       (expression vec3 +
>+			(constant float (-0.047800794))
>+			(expression vec3 *
>+			 (expression vec3 abs (var_ref x))
>+			 (expression vec3 +
>+			  (constant float (0.023164911))
>+			  (expression vec3 *
>+			   (expression vec3 abs (var_ref x))
>+			   (constant float (-0.006116644))
>+   ))))))))))))))))
> 
>    (signature vec4
>      (parameters
>@@ -100,10 +124,18 @@
> 		    (constant float (-0.21460183))
> 		    (expression vec4 *
> 		     (expression vec4 abs (var_ref x))
>-                     (expression vec4 +
>-                      (constant float (0.086566724))
>-                      (expression vec4 *
>-                       (expression vec4 abs (var_ref x))
>-                       (constant float (-0.03102955))
>-		     ))))))))))))
>+		     (expression vec4 +
>+		      (constant float (0.08873435))
>+		      (expression vec4 *
>+		       (expression vec4 abs (var_ref x))
>+		       (expression vec4 +
>+			(constant float (-0.047800794))
>+			(expression vec4 *
>+			 (expression vec4 abs (var_ref x))
>+			 (expression vec4 +
>+			  (constant float (0.023164911))
>+			  (expression vec4 *
>+			   (expression vec4 abs (var_ref x))
>+			   (constant float (-0.006116644))
>+   ))))))))))))))))
> ))
>-- 
>1.7.10.4
>

Actions: View | Splinter Review

Attachments on bug 49713: 78390

Use of freedesktop.org services, including Bugzilla, is subject to our Code of Conduct. How we collect and use information is described in our Privacy Policy.