Fix regression of commit de2b129a56335262a44a05541a3ab2e35668cc6e: ProjectFloat Matrix Multiplication of gluUnProject(..) impl.

ProjectFloat's previous gluMultMatricesf(..) used row-major order,
but the replacement multMatrixf(..) uses column-major order (like OpenGL, ..).

Note: The replaced 'gluMultMatrixVecf' by multMatrixVecf() already
      used column-major order.

Fix: Reverse the arguments of matrix multiplication
    m1 x m2 -> m2 x m1

Added proper API documentation in FloatUtil -> Column Major Order of Linear Matrix Layout

4 files changed, 161 insertions, 41 deletions

diff --git a/make/scripts/tests.sh b/make/scripts/tests.sh
index 7ffbfddb7..b0a55ccf1 100755
--- a/make/scripts/tests.sh
+++ b/make/scripts/tests.sh
@@ -208,6 +208,8 @@ function testawtswt() {
 #testnoawt com.jogamp.newt.opengl.GLWindow $*
 #testnoawt com.jogamp.opengl.test.junit.jogl.offscreen.TestOffscreen01GLPBufferNEWT $*
 #testnoawt com.jogamp.opengl.test.junit.jogl.offscreen.TestOffscreen02BitmapNEWT $*
+#testnoawt com.jogamp.opengl.test.junit.jogl.demos.es2.newt.TestGearsES2NEWT $*
+testnoawt com.jogamp.opengl.test.junit.jogl.acore.TestFloatUtil01MatrixMatrixMultNOUI $*
 #testnoawt com.jogamp.opengl.test.junit.jogl.acore.TestNEWTCloseX11DisplayBug565 $*
 #testnoawt com.jogamp.opengl.test.junit.jogl.acore.TestMainVersionGLWindowNEWT $*
 #testnoawt com.jogamp.opengl.test.junit.jogl.acore.TestGLProfile01NEWT $*
@@ -226,7 +228,7 @@ function testawtswt() {
 #testnoawt com.jogamp.opengl.test.junit.jogl.demos.gl2.newt.TestGearsNEWT $*
 #testnoawt com.jogamp.opengl.test.junit.jogl.demos.es1.newt.TestGearsES1NEWT $*
 #testawt com.jogamp.opengl.test.junit.jogl.demos.es2.newt.TestGearsES2NEWT $*
-testnoawt com.jogamp.opengl.test.junit.jogl.demos.es2.newt.TestGearsES2NEWT $*
+#testnoawt com.jogamp.opengl.test.junit.jogl.demos.es2.newt.TestGearsES2NEWT $*
 #testnoawt com.jogamp.opengl.test.junit.jogl.demos.es1.newt.TestRedSquareES1NEWT $*
 #testnoawt com.jogamp.opengl.test.junit.jogl.demos.es2.newt.TestRedSquareES2NEWT $*
 #testnoawt com.jogamp.opengl.test.junit.newt.TestWindows01NEWT $*
diff --git a/src/jogl/classes/com/jogamp/opengl/FloatUtil.java b/src/jogl/classes/com/jogamp/opengl/FloatUtil.java
index 54fac46e6..93543eaf6 100644
--- a/src/jogl/classes/com/jogamp/opengl/FloatUtil.java
+++ b/src/jogl/classes/com/jogamp/opengl/FloatUtil.java
@@ -32,6 +32,10 @@ import java.nio.FloatBuffer;
 /**
  * Basic Float math utility functions.
  * <p>
+ * Implementation assumes linear matrix layout in column-major order
+ * matching OpenGL's implementation.
+ * </p>
+ * <p>
  * Derived from ProjectFloat.java - Created 11-jan-2004
  * </p>
  * 
@@ -91,29 +95,31 @@ public class FloatUtil {
   }
   
   /**
-   * @param a
-   * @param b
-   * @param d result a*b
+   * @param a 4x4 matrix in column-major order
+   * @param b 4x4 matrix in column-major order
+   * @param d result a*b in column-major order
    */
   public static final void multMatrixf(final float[] a, int a_off, final float[] b, int b_off, float[] d, int d_off) {
      for (int i = 0; i < 4; i++) {
-        final float ai0=a[a_off+i+0*4],  ai1=a[a_off+i+1*4],  ai2=a[a_off+i+2*4],  ai3=a[a_off+i+3*4];
+        // one row in column-major order
+        final float ai0=a[a_off+i+0*4],  ai1=a[a_off+i+1*4],  ai2=a[a_off+i+2*4],  ai3=a[a_off+i+3*4]; // row-i of a
         d[d_off+i+0*4] = ai0 * b[b_off+0+0*4] + ai1 * b[b_off+1+0*4] + ai2 * b[b_off+2+0*4] + ai3 * b[b_off+3+0*4] ;
         d[d_off+i+1*4] = ai0 * b[b_off+0+1*4] + ai1 * b[b_off+1+1*4] + ai2 * b[b_off+2+1*4] + ai3 * b[b_off+3+1*4] ;
         d[d_off+i+2*4] = ai0 * b[b_off+0+2*4] + ai1 * b[b_off+1+2*4] + ai2 * b[b_off+2+2*4] + ai3 * b[b_off+3+2*4] ;
         d[d_off+i+3*4] = ai0 * b[b_off+0+3*4] + ai1 * b[b_off+1+3*4] + ai2 * b[b_off+2+3*4] + ai3 * b[b_off+3+3*4] ;
      }
   }
-
+  
   /**
-   * @param a
-   * @param b
-   * @param d result a*b
+   * @param a 4x4 matrix in column-major order
+   * @param b 4x4 matrix in column-major order
+   * @param d result a*b in column-major order
    */
   public static final void multMatrixf(final float[] a, int a_off, final float[] b, int b_off, FloatBuffer d) {
      final int dP = d.position();
      for (int i = 0; i < 4; i++) {
-        final float ai0=a[a_off+i+0*4],  ai1=a[a_off+i+1*4],  ai2=a[a_off+i+2*4],  ai3=a[a_off+i+3*4];
+        // one row in column-major order
+        final float ai0=a[a_off+i+0*4],  ai1=a[a_off+i+1*4],  ai2=a[a_off+i+2*4],  ai3=a[a_off+i+3*4]; // row-i of a
         d.put(dP+i+0*4 , ai0 * b[b_off+0+0*4] + ai1 * b[b_off+1+0*4] + ai2 * b[b_off+2+0*4] + ai3 * b[b_off+3+0*4] );
         d.put(dP+i+1*4 , ai0 * b[b_off+0+1*4] + ai1 * b[b_off+1+1*4] + ai2 * b[b_off+2+1*4] + ai3 * b[b_off+3+1*4] );
         d.put(dP+i+2*4 , ai0 * b[b_off+0+2*4] + ai1 * b[b_off+1+2*4] + ai2 * b[b_off+2+2*4] + ai3 * b[b_off+3+2*4] );
@@ -122,15 +128,16 @@ public class FloatUtil {
   }
 
   /**
-   * @param a
-   * @param b
-   * @param d result a*b
+   * @param a 4x4 matrix in column-major order
+   * @param b 4x4 matrix in column-major order
+   * @param d result a*b in column-major order
    */
   public static final void multMatrixf(final FloatBuffer a, final float[] b, int b_off, FloatBuffer d) {
      final int aP = a.position(); 
      final int dP = d.position();
      for (int i = 0; i < 4; i++) {
-        final float ai0=a.get(aP+i+0*4),  ai1=a.get(aP+i+1*4),  ai2=a.get(aP+i+2*4),  ai3=a.get(aP+i+3*4);
+        // one row in column-major order
+        final float ai0=a.get(aP+i+0*4),  ai1=a.get(aP+i+1*4),  ai2=a.get(aP+i+2*4),  ai3=a.get(aP+i+3*4); // row-i of a
         d.put(dP+i+0*4 , ai0 * b[b_off+0+0*4] + ai1 * b[b_off+1+0*4] + ai2 * b[b_off+2+0*4] + ai3 * b[b_off+3+0*4] );
         d.put(dP+i+1*4 , ai0 * b[b_off+0+1*4] + ai1 * b[b_off+1+1*4] + ai2 * b[b_off+2+1*4] + ai3 * b[b_off+3+1*4] );
         d.put(dP+i+2*4 , ai0 * b[b_off+0+2*4] + ai1 * b[b_off+1+2*4] + ai2 * b[b_off+2+2*4] + ai3 * b[b_off+3+2*4] );
@@ -139,16 +146,17 @@ public class FloatUtil {
   }
 
   /**
-   * @param a
-   * @param b
-   * @param d result a*b
+   * @param a 4x4 matrix in column-major order
+   * @param b 4x4 matrix in column-major order
+   * @param d result a*b in column-major order
    */
   public static final void multMatrixf(final FloatBuffer a, final FloatBuffer b, FloatBuffer d) {
      final int aP = a.position(); 
      final int bP = b.position();
      final int dP = d.position();
      for (int i = 0; i < 4; i++) {
-        final float ai0=a.get(aP+i+0*4),  ai1=a.get(aP+i+1*4),  ai2=a.get(aP+i+2*4),  ai3=a.get(aP+i+3*4);
+        // one row in column-major order
+        final float ai0=a.get(aP+i+0*4),  ai1=a.get(aP+i+1*4),  ai2=a.get(aP+i+2*4),  ai3=a.get(aP+i+3*4); // row-i of a
         d.put(dP+i+0*4 , ai0 * b.get(bP+0+0*4) + ai1 * b.get(bP+1+0*4) + ai2 * b.get(bP+2+0*4) + ai3 * b.get(bP+3+0*4) );
         d.put(dP+i+1*4 , ai0 * b.get(bP+0+1*4) + ai1 * b.get(bP+1+1*4) + ai2 * b.get(bP+2+1*4) + ai3 * b.get(bP+3+1*4) );
         d.put(dP+i+2*4 , ai0 * b.get(bP+0+2*4) + ai1 * b.get(bP+1+2*4) + ai2 * b.get(bP+2+2*4) + ai3 * b.get(bP+3+2*4) );
@@ -201,10 +209,10 @@ public class FloatUtil {
 
 
   /**
-   * Calculate cross-product
+   * Calculate cross-product of 2 vector
    *
-   * @param v1
-   * @param v2
+   * @param v1 3-component vector
+   * @param v2 3-component vector
    * @param result v1 X v2
    */
   public static final void cross(float[] v1, float[] v2, float[] result) {
@@ -214,10 +222,10 @@ public class FloatUtil {
   }
 
   /**
-   * Calculate cross-product
+   * Calculate cross-product of 2 vector
    *
-   * @param v1
-   * @param v2
+   * @param v1 3-component vector
+   * @param v2 3-component vector
    * @param result v1 X v2
    */
   public static final void cross(FloatBuffer v1, FloatBuffer v2, FloatBuffer result) {
@@ -231,15 +239,14 @@ public class FloatUtil {
   }
 
   /**
-   * Method __gluMultMatrixVecf
-   * 
-   * @param m_in
+   * @param m_in 4x4 matrix in column-major order
    * @param m_in_off
-   * @param v_in
+   * @param v_in 4-component column-vector
    * @param v_out m_in * v_in
    */
   public static final void multMatrixVecf(float[] m_in, int m_in_off, float[] v_in, int v_in_off, float[] v_out) {
     for (int i = 0; i < 4; i++) {
+      // (one matrix row in column-major order) X (column vector)
       v_out[i] =
         v_in[0+v_in_off] * m_in[0*4+i+m_in_off] +
         v_in[1+v_in_off] * m_in[1*4+i+m_in_off] +
@@ -249,15 +256,14 @@ public class FloatUtil {
   }
 
   /**
-   * Method __gluMultMatrixVecf
-   * 
-   * @param m_in
+   * @param m_in 4x4 matrix in column-major order
    * @param m_in_off
-   * @param v_in
+   * @param v_in 4-component column-vector
    * @param v_out m_in * v_in
    */
   public static final void multMatrixVecf(float[] m_in, float[] v_in, float[] v_out) {
     for (int i = 0; i < 4; i++) {
+      // (one matrix row in column-major order) X (column vector)
       v_out[i] =
         v_in[0] * m_in[0*4+i] +
         v_in[1] * m_in[1*4+i] +
@@ -267,10 +273,8 @@ public class FloatUtil {
   }
   
   /**
-   * Method __gluMultMatrixVecf
-   * 
-   * @param m_in
-   * @param v_in
+   * @param m_in 4x4 matrix in column-major order
+   * @param v_in 4-component column-vector
    * @param v_out m_in * v_in
    */
   public static final void multMatrixVecf(FloatBuffer m_in, FloatBuffer v_in, FloatBuffer v_out) {
@@ -278,6 +282,7 @@ public class FloatUtil {
     int outPos = v_out.position();
     int matrixPos = m_in.position();
     for (int i = 0; i < 4; i++) {
+      // (one matrix row in column-major order) X (column vector)
       v_out.put(i + outPos,
               v_in.get(0+inPos) * m_in.get(0*4+i+matrixPos) +
               v_in.get(1+inPos) * m_in.get(1*4+i+matrixPos) +
diff --git a/src/jogl/classes/jogamp/opengl/ProjectFloat.java b/src/jogl/classes/jogamp/opengl/ProjectFloat.java
index bce3f3cfc..ce8405f74 100644
--- a/src/jogl/classes/jogamp/opengl/ProjectFloat.java
+++ b/src/jogl/classes/jogamp/opengl/ProjectFloat.java
@@ -615,7 +615,7 @@ public class ProjectFloat {
     float[] in = this.in;
     float[] out = this.out;
 
-    FloatUtil.multMatrixf(modelMatrix, modelMatrix_offset, projMatrix, projMatrix_offset, matrix, 0);
+    FloatUtil.multMatrixf(projMatrix, projMatrix_offset, modelMatrix, modelMatrix_offset, matrix, 0);
 
     if (!gluInvertMatrixf(matrix, 0, matrix, 0)) {
       return false;
@@ -659,7 +659,7 @@ public class ProjectFloat {
     FloatBuffer in = this.inBuf;
     FloatBuffer out = this.outBuf;
 
-    FloatUtil.multMatrixf(modelMatrix, projMatrix, matrixBuf);
+    FloatUtil.multMatrixf(projMatrix, modelMatrix, matrixBuf);
 
     if (!gluInvertMatrixf(matrixBuf, matrixBuf)) {
       return false;
@@ -715,7 +715,7 @@ public class ProjectFloat {
     FloatBuffer in = this.inBuf;
     FloatBuffer out = this.outBuf;
 
-    FloatUtil.multMatrixf(modelMatrix, projMatrix, matrixBuf);
+    FloatUtil.multMatrixf(projMatrix, modelMatrix, matrixBuf);
 
     if (!gluInvertMatrixf(matrixBuf, matrixBuf)) {
       return false;
@@ -786,7 +786,7 @@ public class ProjectFloat {
     float[] in = this.in;
     float[] out = this.out;
 
-    FloatUtil.multMatrixf(modelMatrix, modelMatrix_offset, projMatrix, projMatrix_offset, matrix, 0);
+    FloatUtil.multMatrixf(projMatrix, projMatrix_offset, modelMatrix, modelMatrix_offset, matrix, 0);
 
     if (!gluInvertMatrixf(matrix, 0, matrix, 0))
       return false;
@@ -847,7 +847,7 @@ public class ProjectFloat {
     FloatBuffer in = this.inBuf;
     FloatBuffer out = this.outBuf;
 
-    FloatUtil.multMatrixf(modelMatrix, projMatrix, matrixBuf);
+    FloatUtil.multMatrixf(projMatrix, modelMatrix, matrixBuf);
 
     if (!gluInvertMatrixf(matrixBuf, matrixBuf))
       return false;
diff --git a/src/test/com/jogamp/opengl/test/junit/jogl/acore/TestFloatUtil01MatrixMatrixMultNOUI.java b/src/test/com/jogamp/opengl/test/junit/jogl/acore/TestFloatUtil01MatrixMatrixMultNOUI.java
new file mode 100755
index 000000000..42b5972bb
--- /dev/null
+++ b/src/test/com/jogamp/opengl/test/junit/jogl/acore/TestFloatUtil01MatrixMatrixMultNOUI.java
@@ -0,0 +1,113 @@
+/**
+ * Copyright 2012 JogAmp Community. All rights reserved.
+ *
+ * Redistribution and use in source and binary forms, with or without modification, are
+ * permitted provided that the following conditions are met:
+ * 
+ *    1. Redistributions of source code must retain the above copyright notice, this list of
+ *       conditions and the following disclaimer.
+ * 
+ *    2. Redistributions in binary form must reproduce the above copyright notice, this list
+ *       of conditions and the following disclaimer in the documentation and/or other materials
+ *       provided with the distribution.
+ * 
+ * THIS SOFTWARE IS PROVIDED BY JogAmp Community ``AS IS'' AND ANY EXPRESS OR IMPLIED
+ * WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND
+ * FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL JogAmp Community OR
+ * CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+ * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR
+ * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON
+ * ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING
+ * NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF
+ * ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+ * 
+ * The views and conclusions contained in the software and documentation are those of the
+ * authors and should not be interpreted as representing official policies, either expressed
+ * or implied, of JogAmp Community.
+ */
+ 
+package com.jogamp.opengl.test.junit.jogl.acore;
+
+import org.junit.Assert;
+import org.junit.Test;
+
+import com.jogamp.opengl.FloatUtil;
+
+public class TestFloatUtil01MatrixMatrixMultNOUI {
+
+    final float[] m1 = new float[]{    1,    3,    4,    0,
+                                       6,    7,    8,    5,
+                                      98,    7,    6,    9,
+                                      54,    3,    2,    5 };
+
+    final float[] m2 = new float[]{    1,    6,   98,   54,
+                                       3,    7,    7,    3,
+                                       4,    8,    6,    2,
+                                       0,    5,    9,    5 };
+
+    final float[] m1xm2_RM = // m2xm1_CM 
+                       new float[]{   26,   59,  143,   71,
+                                      59,  174,  730,  386,
+                                     143,  730, 9770, 5370,
+                                      71,  386, 5370, 2954 };
+
+    final float[] m2xm1_RM = // m1xm2_CM
+                       new float[]{12557,  893,  748, 1182,
+                                     893,  116,  116,  113,
+                                     748,  116,  120,  104,
+                                    1182,  113,  104,  131 };
+    
+    public static final void multMatrixf_RM(final float[] a, int a_off, final float[] b, int b_off, float[] d, int d_off) {
+     for (int i = 0; i < 4; i++) {
+        final float ai0=a[a_off+i*4+0],  ai1=a[a_off+i*4+1],  ai2=a[a_off+i*4+2],  ai3=a[a_off+i*4+3];
+        d[d_off+i*4+0] = ai0 * b[b_off+0*4+0] + ai1 * b[b_off+1*4+0] + ai2 * b[b_off+2*4+0] + ai3 * b[b_off+3*4+0] ;
+        d[d_off+i*4+1] = ai0 * b[b_off+0*4+1] + ai1 * b[b_off+1*4+1] + ai2 * b[b_off+2*4+1] + ai3 * b[b_off+3*4+1] ;
+        d[d_off+i*4+2] = ai0 * b[b_off+0*4+2] + ai1 * b[b_off+1*4+2] + ai2 * b[b_off+2*4+2] + ai3 * b[b_off+3*4+2] ;
+        d[d_off+i*4+3] = ai0 * b[b_off+0*4+3] + ai1 * b[b_off+1*4+3] + ai2 * b[b_off+2*4+3] + ai3 * b[b_off+3*4+3] ;
+     }
+    }
+
+    @Test
+    public void testCM_m1xm2(){
+        
+        float[] r = new float[16];
+        
+        FloatUtil.multMatrixf(m1, 0, m2, 0, r, 0);
+
+        Assert.assertArrayEquals(m2xm1_RM, r, 0f);
+    }
+    
+    @Test
+    public void testCM_m2xm1(){
+        
+        float[] r = new float[16];
+        
+        FloatUtil.multMatrixf(m2, 0, m1, 0, r, 0);
+
+        Assert.assertArrayEquals(m1xm2_RM, r, 0f);
+    }
+
+    @Test
+    public void testRM_m1xm2(){
+        
+        float[] r = new float[16];
+        
+        multMatrixf_RM(m1, 0, m2, 0, r, 0);
+
+        Assert.assertArrayEquals(m1xm2_RM, r, 0f);
+    }
+    
+    @Test
+    public void testRM_m2xm1(){
+        
+        float[] r = new float[16];
+        
+        multMatrixf_RM(m2, 0, m1, 0, r, 0);
+
+        Assert.assertArrayEquals(m2xm1_RM, r, 0f);
+    }
+    
+    public static void main(String args[]) {
+        org.junit.runner.JUnitCore.main(TestFloatUtil01MatrixMatrixMultNOUI.class.getName());
+    }
+}