Merge remote-tracking branch 'origin/released'

42 files changed, 3678 insertions, 6762 deletions

diff --git a/mesalib/include/GL/internal/dri_interface.h b/mesalib/include/GL/internal/dri_interface.h
index 5561fcb0d..6aa6db74a 100644
--- a/mesalib/include/GL/internal/dri_interface.h
+++ b/mesalib/include/GL/internal/dri_interface.h
@@ -681,7 +681,7 @@ struct __DRIlegacyExtensionRec {
  * conjunction with the core extension.
  */
 #define __DRI_SWRAST "DRI_SWRast"
-#define __DRI_SWRAST_VERSION 2
+#define __DRI_SWRAST_VERSION 3
 
 struct __DRIswrastExtensionRec {
     __DRIextension base;
@@ -701,6 +701,22 @@ struct __DRIswrastExtensionRec {
                                            const __DRIconfig *config,
                                            __DRIcontext *shared,
                                            void *data);
+
+   /**
+    * Create a context for a particular API with a set of attributes
+    *
+    * \since version 3
+    *
+    * \sa __DRIdri2ExtensionRec::createContextAttribs
+    */
+   __DRIcontext *(*createContextAttribs)(__DRIscreen *screen,
+					 int api,
+					 const __DRIconfig *config,
+					 __DRIcontext *shared,
+					 unsigned num_attribs,
+					 const uint32_t *attribs,
+					 unsigned *error,
+					 void *loaderPrivate);
 };
 
 /**
@@ -783,11 +799,45 @@ struct __DRIdri2LoaderExtensionRec {
  * constructors for DRI2.
  */
 #define __DRI_DRI2 "DRI_DRI2"
-#define __DRI_DRI2_VERSION 2
+#define __DRI_DRI2_VERSION 3
 
-#define __DRI_API_OPENGL	0
+#define __DRI_API_OPENGL	0	/**< OpenGL compatibility profile */
 #define __DRI_API_GLES		1
 #define __DRI_API_GLES2		2
+#define __DRI_API_OPENGL_CORE	3	/**< OpenGL 3.2+ core profile */
+
+#define __DRI_CTX_ATTRIB_MAJOR_VERSION		0
+#define __DRI_CTX_ATTRIB_MINOR_VERSION		1
+#define __DRI_CTX_ATTRIB_FLAGS			2
+
+#define __DRI_CTX_FLAG_DEBUG			0x00000001
+#define __DRI_CTX_FLAG_FORWARD_COMPATIBLE	0x00000002
+
+/**
+ * \name Reasons that __DRIdri2Extension::createContextAttribs might fail
+ */
+/*@{*/
+/** Success! */
+#define __DRI_CTX_ERROR_SUCCESS			0
+
+/** Memory allocation failure */
+#define __DRI_CTX_ERROR_NO_MEMORY		1
+
+/** Client requested an API (e.g., OpenGL ES 2.0) that the driver can't do. */
+#define __DRI_CTX_ERROR_BAD_API			2
+
+/** Client requested an API version that the driver can't do. */
+#define __DRI_CTX_ERROR_BAD_VERSION		3
+
+/** Client requested a flag or combination of flags the driver can't do. */
+#define __DRI_CTX_ERROR_BAD_FLAG		4
+
+/** Client requested an attribute the driver doesn't understand. */
+#define __DRI_CTX_ERROR_UNKNOWN_ATTRIBUTE	5
+
+/** Client requested a flag the driver doesn't understand. */
+#define __DRI_CTX_ERROR_UNKNOWN_FLAG		6
+/*@}*/
 
 struct __DRIdri2ExtensionRec {
     __DRIextension base;
@@ -822,6 +872,22 @@ struct __DRIdri2ExtensionRec {
 				  int height);
    void (*releaseBuffer)(__DRIscreen *screen,
 			 __DRIbuffer *buffer);
+
+   /**
+    * Create a context for a particular API with a set of attributes
+    *
+    * \since version 3
+    *
+    * \sa __DRIswrastExtensionRec::createContextAttribs
+    */
+   __DRIcontext *(*createContextAttribs)(__DRIscreen *screen,
+					 int api,
+					 const __DRIconfig *config,
+					 __DRIcontext *shared,
+					 unsigned num_attribs,
+					 const uint32_t *attribs,
+					 unsigned *error,
+					 void *loaderPrivate);
 };
 
 
diff --git a/mesalib/src/glsl/autogen.sh b/mesalib/src/glsl/autogen.sh
deleted file mode 100644
index 6fcae015c..000000000
--- a/mesalib/src/glsl/autogen.sh
+++ /dev/null
@@ -1,12 +0,0 @@
-#! /bin/sh

-

-srcdir=`dirname $0`

-test -z "$srcdir" && srcdir=.

-

-ORIGDIR=`pwd`

-cd $srcdir

-

-autoreconf -v --install || exit 1

-cd $ORIGDIR || exit $?

-

-$srcdir/configure --enable-maintainer-mode "$@"

diff --git a/mesalib/src/mesa/drivers/common/meta.c b/mesalib/src/mesa/drivers/common/meta.c
index c5c59ebee..e62267394 100644
--- a/mesalib/src/mesa/drivers/common/meta.c
+++ b/mesalib/src/mesa/drivers/common/meta.c
@@ -65,6 +65,7 @@
 #include "main/teximage.h"
 #include "main/texparam.h"
 #include "main/texstate.h"
+#include "main/transformfeedback.h"
 #include "main/uniforms.h"
 #include "main/varray.h"
 #include "main/viewport.h"
@@ -180,6 +181,8 @@ struct save_state
 
    /** Miscellaneous (always disabled) */
    GLboolean Lighting;
+   GLboolean RasterDiscard;
+   GLboolean TransformFeedbackNeedsResume;
 };
 
 /**
@@ -422,6 +425,15 @@ _mesa_meta_begin(struct gl_context *ctx, GLbitfield state)
    memset(save, 0, sizeof(*save));
    save->SavedState = state;
 
+   /* Pausing transform feedback needs to be done early, or else we won't be
+    * able to change other state.
+    */
+   save->TransformFeedbackNeedsResume =
+      ctx->TransformFeedback.CurrentObject->Active &&
+      !ctx->TransformFeedback.CurrentObject->Paused;
+   if (save->TransformFeedbackNeedsResume)
+      _mesa_PauseTransformFeedback();
+
    if (state & MESA_META_ALPHA_TEST) {
       save->AlphaEnabled = ctx->Color.AlphaEnabled;
       save->AlphaFunc = ctx->Color.AlphaFunc;
@@ -702,6 +714,9 @@ _mesa_meta_begin(struct gl_context *ctx, GLbitfield state)
       save->Lighting = ctx->Light.Enabled;
       if (ctx->Light.Enabled)
          _mesa_set_enable(ctx, GL_LIGHTING, GL_FALSE);
+      save->RasterDiscard = ctx->RasterDiscard;
+      if (ctx->RasterDiscard)
+         _mesa_set_enable(ctx, GL_RASTERIZER_DISCARD, GL_FALSE);
    }
 }
 
@@ -981,6 +996,11 @@ _mesa_meta_end(struct gl_context *ctx)
    if (save->Lighting) {
       _mesa_set_enable(ctx, GL_LIGHTING, GL_TRUE);
    }
+   if (save->RasterDiscard) {
+      _mesa_set_enable(ctx, GL_RASTERIZER_DISCARD, GL_TRUE);
+   }
+   if (save->TransformFeedbackNeedsResume)
+      _mesa_ResumeTransformFeedback();
 }
 
 
diff --git a/mesalib/src/mesa/drivers/dri/common/depthtmp.h b/mesalib/src/mesa/drivers/dri/common/depthtmp.h
index c6e2bb75c..d8f259f67 100644
--- a/mesalib/src/mesa/drivers/dri/common/depthtmp.h
+++ b/mesalib/src/mesa/drivers/dri/common/depthtmp.h
@@ -71,63 +71,6 @@ static void TAG(WriteDepthSpan)( struct gl_context *ctx,
 }
 
 
-#if HAVE_HW_DEPTH_SPANS
-/* implement MonoWriteDepthSpan() in terms of WriteDepthSpan() */
-static void
-TAG(WriteMonoDepthSpan)( struct gl_context *ctx, struct gl_renderbuffer *rb,
-                         GLuint n, GLint x, GLint y,
-                         const void *value, const GLubyte mask[] )
-{
-   const GLuint depthVal = *((GLuint *) value);
-   GLuint depths[MAX_WIDTH];
-   GLuint i;
-   for (i = 0; i < n; i++)
-      depths[i] = depthVal;
-   TAG(WriteDepthSpan)(ctx, rb, n, x, y, depths, mask);
-}
-#else
-static void TAG(WriteMonoDepthSpan)( struct gl_context *ctx,
-                                     struct gl_renderbuffer *rb,
-                                     GLuint n, GLint x, GLint y,
-                                     const void *value,
-                                     const GLubyte mask[] )
-{
-   HW_WRITE_LOCK()
-      {
-         const GLuint depth = *((GLuint *) value);
-	 GLint x1;
-	 GLint n1;
-	 LOCAL_DEPTH_VARS;
-
-	 y = Y_FLIP( y );
-
-	 HW_CLIPLOOP()
-	    {
-	       GLint i = 0;
-	       CLIPSPAN( x, y, n, x1, n1, i );
-
-	       if ( DBG ) fprintf( stderr, "%s %d..%d (x1 %d) = %u\n",
-				   __FUNCTION__, (int)i, (int)n1, (int)x1, (GLuint)depth );
-
-	       if ( mask ) {
-		  for ( ; n1>0 ; i++, x1++, n1-- ) {
-		     if ( mask[i] ) WRITE_DEPTH( x1, y, depth );
-		  }
-	       } else {
-		  for ( ; n1>0 ; x1++, n1-- ) {
-		     WRITE_DEPTH( x1, y, depth );
-		  }
-	       }
-	    }
-	 HW_ENDCLIPLOOP();
-      }
-   HW_WRITE_UNLOCK();
-
-   (void) ctx;
-}
-#endif
-
-
 static void TAG(WriteDepthPixels)( struct gl_context *ctx,
                                    struct gl_renderbuffer *rb,
 				   GLuint n,
@@ -258,10 +201,7 @@ static void TAG(InitDepthPointers)(struct gl_renderbuffer *rb)
    rb->GetRow = TAG(ReadDepthSpan);
    rb->GetValues = TAG(ReadDepthPixels);
    rb->PutRow = TAG(WriteDepthSpan);
-   rb->PutRowRGB = NULL;
-   rb->PutMonoRow = TAG(WriteMonoDepthSpan);
    rb->PutValues = TAG(WriteDepthPixels);
-   rb->PutMonoValues = NULL;
 }
 
 
diff --git a/mesalib/src/mesa/drivers/dri/common/dri_util.c b/mesalib/src/mesa/drivers/dri/common/dri_util.c
index 1640c1448..a153d525d 100644
--- a/mesalib/src/mesa/drivers/dri/common/dri_util.c
+++ b/mesalib/src/mesa/drivers/dri/common/dri_util.c
@@ -454,14 +454,18 @@ const __DRIcoreExtension driCoreExtension = {
 
 /** DRI2 interface */
 const __DRIdri2Extension driDRI2Extension = {
-    { __DRI_DRI2, __DRI_DRI2_VERSION },
+    /* Force the version to 2 because the underlying drivers don't (can't!)
+     * support the extra requirements of CreateContextAttribs.
+     */
+    { __DRI_DRI2, 2 },
     dri2CreateNewScreen,
     dri2CreateNewDrawable,
     dri2CreateNewContext,
     dri2GetAPIMask,
     dri2CreateNewContextForAPI,
     dri2AllocateBuffer,
-    dri2ReleaseBuffer
+    dri2ReleaseBuffer,
+    NULL
 };
 
 const __DRI2configQueryExtension dri2ConfigQueryExtension = {
diff --git a/mesalib/src/mesa/drivers/dri/common/drisw_util.c b/mesalib/src/mesa/drivers/dri/common/drisw_util.c
index 614339eeb..a19123f70 100644
--- a/mesalib/src/mesa/drivers/dri/common/drisw_util.c
+++ b/mesalib/src/mesa/drivers/dri/common/drisw_util.c
@@ -288,8 +288,12 @@ const __DRIcoreExtension driCoreExtension = {
 };
 
 const __DRIswrastExtension driSWRastExtension = {
-    { __DRI_SWRAST, __DRI_SWRAST_VERSION },
+    /* Force the version to 2 because the underlying driver don't (can't!)
+     * support the extra requirements of CreateContextAttribs.
+     */
+    { __DRI_SWRAST, 2 },
     driCreateNewScreen,
     driCreateNewDrawable,
-    driCreateNewContextForAPI
+    driCreateNewContextForAPI,
+    NULL
 };
diff --git a/mesalib/src/mesa/drivers/dri/common/spantmp2.h b/mesalib/src/mesa/drivers/dri/common/spantmp2.h
index 83cfbb164..744dfcd3b 100644
--- a/mesalib/src/mesa/drivers/dri/common/spantmp2.h
+++ b/mesalib/src/mesa/drivers/dri/common/spantmp2.h
@@ -63,9 +63,6 @@
 #define PUT_VALUE(_x, _y, _v) *(volatile GLushort *)(GET_PTR(_x, _y)) = (_v)
 #endif /* GET_VALUE */
 
-#define INIT_MONO_PIXEL(p, color) \
-  p = PACK_COLOR_565( color[0], color[1], color[2] )
-
 #define WRITE_RGBA( _x, _y, r, g, b, a )				\
    PUT_VALUE(_x, _y, ((((int)r & 0xf8) << 8) |				\
 		      (((int)g & 0xfc) << 3) |				\
@@ -97,9 +94,6 @@
 #define PUT_VALUE(_x, _y, _v) *(volatile GLushort *)(GET_PTR(_x, _y)) = (_v)
 #endif /* GET_VALUE */
 
-#define INIT_MONO_PIXEL(p, color) \
-  p = PACK_COLOR_565_REV( color[0], color[1], color[2] )
-
 #define WRITE_RGBA( _x, _y, r, g, b, a )				\
    PUT_VALUE(_x, _y, PACK_COLOR_565_REV( r, g, b ))
 
@@ -130,9 +124,6 @@
 #define PUT_VALUE(_x, _y, _v) *(volatile GLushort *)(GET_PTR(_x, _y)) = (_v)
 #endif /* GET_VALUE */
 
-#define INIT_MONO_PIXEL(p, color) \
-   p = PACK_COLOR_4444_REV(color[3], color[0], color[1], color[2])
-
 #define WRITE_RGBA( _x, _y, r, g, b, a )				\
    PUT_VALUE(_x, _y, PACK_COLOR_4444_REV(a, r, g, b))			\
 
@@ -163,9 +154,6 @@
 #define PUT_VALUE(_x, _y, _v) *(volatile GLushort *)(GET_PTR(_x, _y)) = (_v)
 #endif /* GET_VALUE */
 
-#define INIT_MONO_PIXEL(p, color) \
-   p = PACK_COLOR_4444(color[3], color[0], color[1], color[2])
-
 #define WRITE_RGBA( _x, _y, r, g, b, a )				\
    PUT_VALUE(_x, _y, PACK_COLOR_4444(a, r, g, b))			\
 
@@ -196,9 +184,6 @@
 #define PUT_VALUE(_x, _y, _v) *(volatile GLushort *)(GET_PTR(_x, _y)) = (_v)
 #endif /* GET_VALUE */
 
-#define INIT_MONO_PIXEL(p, color) \
-   p = PACK_COLOR_1555(color[3], color[0], color[1], color[2])
-
 #define WRITE_RGBA( _x, _y, r, g, b, a )				\
    PUT_VALUE(_x, _y, PACK_COLOR_1555(a, r, g, b))			\
 
@@ -228,9 +213,6 @@
 #define PUT_VALUE(_x, _y, _v) *(volatile GLushort *)(GET_PTR(_x, _y)) = (_v)
 #endif /* GET_VALUE */
 
-#define INIT_MONO_PIXEL(p, color) \
-   p = PACK_COLOR_1555_REV(color[3], color[0], color[1], color[2])
-
 #define WRITE_RGBA( _x, _y, r, g, b, a )				\
    PUT_VALUE(_x, _y, PACK_COLOR_1555_REV(a, r, g, b))			\
 
@@ -261,9 +243,6 @@
 #define PUT_VALUE(_x, _y, _v) *(volatile GLuint *)(GET_PTR(_x, _y)) = (_v)
 #endif /* GET_VALUE */
 
-# define INIT_MONO_PIXEL(p, color)                       \
-     p = PACK_COLOR_8888(color[3], color[0], color[1], color[2]) 
-
 # define WRITE_RGBA(_x, _y, r, g, b, a)                                 \
    PUT_VALUE(_x, _y, ((r << 16) |					\
 		      (g << 8) |					\
@@ -316,9 +295,6 @@
 #define PUT_VALUE(_x, _y, _v) *(volatile GLuint *)(GET_PTR(_x, _y)) = (_v)
 #endif /* GET_VALUE */
 
-# define INIT_MONO_PIXEL(p, color)                       \
-     p = PACK_COLOR_8888(color[2], color[1], color[0], color[3]) 
-
 # define WRITE_RGBA(_x, _y, r, g, b, a)                                 \
    PUT_VALUE(_x, _y, ((r << 8) |					\
 		      (g << 16) |					\
@@ -374,9 +350,6 @@
 #define PUT_VALUE(_x, _y, _v) *(volatile GLuint *)(GET_PTR(_x, _y)) = (_v)
 #endif /* GET_VALUE */
 
-# define INIT_MONO_PIXEL(p, color)                       \
-     p = PACK_COLOR_8888(0xff, color[0], color[1], color[2])
-
 # define WRITE_RGBA(_x, _y, r, g, b, a)					\
    PUT_VALUE(_x, _y, ((r << 16) |					\
 		      (g << 8) |					\
@@ -428,9 +401,6 @@
 #define PUT_VALUE(_x, _y, _v) *(volatile GLubyte *)(GET_PTR(_x, _y)) = (_v)
 #endif /* GET_VALUE */
 
-# define INIT_MONO_PIXEL(p, color)                       \
-     p = color[3]
-
 # define WRITE_RGBA(_x, _y, r, g, b, a)                                 \
    PUT_VALUE(_x, _y, a | (r & 0 /* quiet warnings */))
 
@@ -505,46 +475,6 @@ static void TAG(WriteRGBASpan)( struct gl_context *ctx,
    HW_WRITE_UNLOCK();
 }
 
-static void TAG(WriteRGBSpan)( struct gl_context *ctx,
-                               struct gl_renderbuffer *rb,
-			       GLuint n, GLint x, GLint y,
-			       const void *values, const GLubyte mask[] )
-{
-   (void) ctx;
-
-   HW_WRITE_LOCK()
-      {
-         const GLubyte (*rgb)[3] = (const GLubyte (*)[3]) values;
-	 GLint x1;
-	 GLint n1;
-	 LOCAL_VARS;
-
-	 y = Y_FLIP(y);
-
-	 HW_WRITE_CLIPLOOP()
-	    {
-	       GLint i = 0;
-	       CLIPSPAN(x,y,n,x1,n1,i);
-
-	       if (DBG) fprintf(stderr, "WriteRGBSpan %d..%d (x1 %d)\n",
-				(int)i, (int)n1, (int)x1);
-
-	       if (mask)
-	       {
-		  for (;n1>0;i++,x1++,n1--)
-		     if (mask[i])
-			WRITE_RGBA( x1, y, rgb[i][0], rgb[i][1], rgb[i][2], 255 );
-	       }
-	       else
-	       {
-		  for (;n1>0;i++,x1++,n1--)
-		     WRITE_RGBA( x1, y, rgb[i][0], rgb[i][1], rgb[i][2], 255 );
-	       }
-	    }
-	 HW_ENDCLIPLOOP();
-      }
-   HW_WRITE_UNLOCK();
-}
 
 static void TAG(WriteRGBAPixels)( struct gl_context *ctx,
                                   struct gl_renderbuffer *rb,
@@ -594,91 +524,6 @@ static void TAG(WriteRGBAPixels)( struct gl_context *ctx,
 }
 
 
-static void TAG(WriteMonoRGBASpan)( struct gl_context *ctx,	
-                                    struct gl_renderbuffer *rb,
-				    GLuint n, GLint x, GLint y, 
-				    const void *value, const GLubyte mask[] )
-{
-   (void) ctx;
-
-   HW_WRITE_LOCK()
-      {
-         const GLubyte *color = (const GLubyte *) value;
-	 GLint x1;
-	 GLint n1;
-	 LOCAL_VARS;
-	 INIT_MONO_PIXEL(p, color);
-
-	 y = Y_FLIP( y );
-
-	 if (DBG) fprintf(stderr, "WriteMonoRGBASpan\n");
-
-	 HW_WRITE_CLIPLOOP()
-	    {
-	       GLint i = 0;
-	       CLIPSPAN(x,y,n,x1,n1,i);
-	       if (mask)
-	       {
-	          for (;n1>0;i++,x1++,n1--)
-		     if (mask[i])
-		        WRITE_PIXEL( x1, y, p );
-	       }
-	       else
-	       {
-	          for (;n1>0;i++,x1++,n1--)
-		     WRITE_PIXEL( x1, y, p );
-	       }
-	    }
-	 HW_ENDCLIPLOOP();
-      }
-   HW_WRITE_UNLOCK();
-}
-
-
-static void TAG(WriteMonoRGBAPixels)( struct gl_context *ctx,
-                                      struct gl_renderbuffer *rb,
-				      GLuint n,
-				      const GLint x[], const GLint y[],
-				      const void *value,
-				      const GLubyte mask[] ) 
-{
-   (void) ctx;
-
-   HW_WRITE_LOCK()
-      {
-         const GLubyte *color = (const GLubyte *) value;
-	 GLint i;
-	 LOCAL_VARS;
-	 INIT_MONO_PIXEL(p, color);
-
-	 if (DBG) fprintf(stderr, "WriteMonoRGBAPixels\n");
-
-	 HW_WRITE_CLIPLOOP()
-	    {
-	       if (mask)
-	       {
-		  for (i=0;i<n;i++)
-		     if (mask[i]) {
-			int fy = Y_FLIP(y[i]);
-			if (CLIPPIXEL( x[i], fy ))
-			   WRITE_PIXEL( x[i], fy, p );
-		     }
-	       }
-	       else
-	       {
-		  for (i=0;i<n;i++) {
-		     int fy = Y_FLIP(y[i]);
-		     if (CLIPPIXEL( x[i], fy ))
-			WRITE_PIXEL( x[i], fy, p );
-		  }
-	       }
-	    }
-	 HW_ENDCLIPLOOP();
-      }
-   HW_WRITE_UNLOCK();
-}
-
-
 static void TAG(ReadRGBASpan)( struct gl_context *ctx,
                                struct gl_renderbuffer *rb,
 			       GLuint n, GLint x, GLint y, void *values)
@@ -878,10 +723,7 @@ static void TAG(ReadRGBAPixels)( struct gl_context *ctx,
 static void TAG(InitPointers)(struct gl_renderbuffer *rb)
 {
    rb->PutRow = TAG(WriteRGBASpan);
-   rb->PutRowRGB = TAG(WriteRGBSpan);
-   rb->PutMonoRow = TAG(WriteMonoRGBASpan);
    rb->PutValues = TAG(WriteRGBAPixels);
-   rb->PutMonoValues = TAG(WriteMonoRGBAPixels);
    rb->GetValues = TAG(ReadRGBAPixels);
 
 #if defined(GET_PTR)
@@ -923,7 +765,6 @@ static void TAG(InitPointers)(struct gl_renderbuffer *rb)
 }
 
 
-#undef INIT_MONO_PIXEL
 #undef WRITE_PIXEL
 #undef WRITE_RGBA
 #undef READ_RGBA
diff --git a/mesalib/src/mesa/drivers/dri/common/stenciltmp.h b/mesalib/src/mesa/drivers/dri/common/stenciltmp.h
index 2ea2e207a..950d3c4df 100644
--- a/mesalib/src/mesa/drivers/dri/common/stenciltmp.h
+++ b/mesalib/src/mesa/drivers/dri/common/stenciltmp.h
@@ -1,245 +1,186 @@
-

-#include "spantmp_common.h"

-

-#ifndef DBG

-#define DBG 0

-#endif

-

-#ifndef HAVE_HW_STENCIL_SPANS

-#define HAVE_HW_STENCIL_SPANS 0

-#endif

-

-#ifndef HAVE_HW_STENCIL_PIXELS

-#define HAVE_HW_STENCIL_PIXELS 0

-#endif

-

-static void TAG(WriteStencilSpan)( struct gl_context *ctx,

-                                   struct gl_renderbuffer *rb,

-				   GLuint n, GLint x, GLint y,

-				   const void *values, const GLubyte mask[] )

-{

-   HW_WRITE_LOCK()

-      {

-         const GLubyte *stencil = (const GLubyte *) values;

-	 GLint x1;

-	 GLint n1;

-	 LOCAL_STENCIL_VARS;

-

-	 y = Y_FLIP(y);

-

-#if HAVE_HW_STENCIL_SPANS

-	 (void) x1; (void) n1;

-

-	 if (DBG) fprintf(stderr, "WriteStencilSpan 0..%d (x1 %d)\n",

-			  (int)n1, (int)x1);

-

-	 WRITE_STENCIL_SPAN();

-#else /* HAVE_HW_STENCIL_SPANS */

-	 HW_CLIPLOOP() 

-	    {

-	       GLint i = 0;

-	       CLIPSPAN(x,y,n,x1,n1,i);

-

-	       if (DBG) fprintf(stderr, "WriteStencilSpan %d..%d (x1 %d)\n",

-				(int)i, (int)n1, (int)x1);

-

-	       if (mask)

-	       {

-		  for (;n1>0;i++,x1++,n1--)

-		     if (mask[i])

-			WRITE_STENCIL( x1, y, stencil[i] );

-	       }

-	       else

-	       {

-		  for (;n1>0;i++,x1++,n1--)

-		     WRITE_STENCIL( x1, y, stencil[i] );

-	       }

-	    }

-	 HW_ENDCLIPLOOP();

-#endif /* !HAVE_HW_STENCIL_SPANS */

-      }

-   HW_WRITE_UNLOCK();

-}

-

-#if HAVE_HW_STENCIL_SPANS

-/* implement MonoWriteDepthSpan() in terms of WriteDepthSpan() */

-static void

-TAG(WriteMonoStencilSpan)( struct gl_context *ctx, struct gl_renderbuffer *rb,

-                           GLuint n, GLint x, GLint y,

-                           const void *value, const GLubyte mask[] )

-{

-   const GLuint stenVal = *((GLuint *) value);

-   GLuint stens[MAX_WIDTH];

-   GLuint i;

-   for (i = 0; i < n; i++)

-      stens[i] = stenVal;

-   TAG(WriteStencilSpan)(ctx, rb, n, x, y, stens, mask);

-}

-#else /* HAVE_HW_STENCIL_SPANS */

-static void TAG(WriteMonoStencilSpan)( struct gl_context *ctx,

-                                       struct gl_renderbuffer *rb,

-                                       GLuint n, GLint x, GLint y,

-                                       const void *value,

-                                       const GLubyte mask[] )

-{

-   HW_WRITE_LOCK()

-      {

-         const GLubyte stencil = *((const GLubyte *) value);

-	 GLint x1;

-	 GLint n1;

-	 LOCAL_STENCIL_VARS;

-

-	 y = Y_FLIP(y);

-

-	 HW_CLIPLOOP() 

-	    {

-	       GLint i = 0;

-	       CLIPSPAN(x,y,n,x1,n1,i);

-

-	       if (DBG) fprintf(stderr, "WriteStencilSpan %d..%d (x1 %d)\n",

-				(int)i, (int)n1, (int)x1);

-

-	       if (mask)

-	       {

-		  for (;n1>0;i++,x1++,n1--)

-		     if (mask[i])

-			WRITE_STENCIL( x1, y, stencil );

-	       }

-	       else

-	       {

-		  for (;n1>0;i++,x1++,n1--)

-		     WRITE_STENCIL( x1, y, stencil );

-	       }

-	    }

-	 HW_ENDCLIPLOOP();

-      }

-   HW_WRITE_UNLOCK();

-}

-#endif /* !HAVE_HW_STENCIL_SPANS */

-

-

-static void TAG(WriteStencilPixels)( struct gl_context *ctx,

-                                     struct gl_renderbuffer *rb,

-				     GLuint n,

-				     const GLint x[], const GLint y[],

-				     const void *values, const GLubyte mask[] )

-{

-   HW_WRITE_LOCK()

-      {

-         const GLubyte *stencil = (const GLubyte *) values;

-	 GLuint i;

-	 LOCAL_STENCIL_VARS;

-

-	 if (DBG) fprintf(stderr, "WriteStencilPixels\n");

-

-#if HAVE_HW_STENCIL_PIXELS

-	 (void) i;

-

-	 WRITE_STENCIL_PIXELS();

-#else /* HAVE_HW_STENCIL_PIXELS */

-	 HW_CLIPLOOP()

-	    {

-	       for (i=0;i<n;i++)

-	       {

-		  if (mask[i]) {

-		     const int fy = Y_FLIP(y[i]);

-		     if (CLIPPIXEL(x[i],fy))

-			WRITE_STENCIL( x[i], fy, stencil[i] );

-		  }

-	       }

-	    }

-	 HW_ENDCLIPLOOP();

-#endif /* !HAVE_HW_STENCIL_PIXELS */

-      }

-   HW_WRITE_UNLOCK();

-}

-

-

-/* Read stencil spans and pixels

- */

-static void TAG(ReadStencilSpan)( struct gl_context *ctx,

-                                  struct gl_renderbuffer *rb,

-				  GLuint n, GLint x, GLint y,

-				  void *values)

-{

-   HW_READ_LOCK()

-      {

-         GLubyte *stencil = (GLubyte *) values;

-	 GLint x1,n1;

-	 LOCAL_STENCIL_VARS;

-

-	 y = Y_FLIP(y);

-

-	 if (DBG) fprintf(stderr, "ReadStencilSpan\n");

-

-#if HAVE_HW_STENCIL_SPANS

-	 (void) x1; (void) n1;

-

-	 READ_STENCIL_SPAN();

-#else /* HAVE_HW_STENCIL_SPANS */

-	 HW_CLIPLOOP() 

-	    {

-	       GLint i = 0;

-	       CLIPSPAN(x,y,n,x1,n1,i);

-	       for (;n1>0;i++,n1--)

-		  READ_STENCIL( stencil[i], (x+i), y );

-	    }

-	 HW_ENDCLIPLOOP();

-#endif /* !HAVE_HW_STENCIL_SPANS */

-      }

-   HW_READ_UNLOCK();

-}

-

-static void TAG(ReadStencilPixels)( struct gl_context *ctx,

-                                    struct gl_renderbuffer *rb,

-                                    GLuint n, const GLint x[], const GLint y[],

-				    void *values )

-{

-   HW_READ_LOCK()

-      {

-         GLubyte *stencil = (GLubyte *) values;

-	 GLuint i;

-	 LOCAL_STENCIL_VARS;

-

-	 if (DBG) fprintf(stderr, "ReadStencilPixels\n");

- 

-#if HAVE_HW_STENCIL_PIXELS

-	 (void) i;

-

-	 READ_STENCIL_PIXELS();

-#else /* HAVE_HW_STENCIL_PIXELS */

-	 HW_CLIPLOOP()

-	    {

-	       for (i=0;i<n;i++) {

-		  int fy = Y_FLIP( y[i] );

-		  if (CLIPPIXEL( x[i], fy ))

-		     READ_STENCIL( stencil[i], x[i], fy );

-	       }

-	    }

-	 HW_ENDCLIPLOOP();

-#endif /* !HAVE_HW_STENCIL_PIXELS */

-      }

-   HW_READ_UNLOCK();

-}

-

-

-

-/**

- * Initialize the given renderbuffer's span routines to point to

- * the stencil functions we generated above.

- */

-static void TAG(InitStencilPointers)(struct gl_renderbuffer *rb)

-{

-   rb->GetRow = TAG(ReadStencilSpan);

-   rb->GetValues = TAG(ReadStencilPixels);

-   rb->PutRow = TAG(WriteStencilSpan);

-   rb->PutRowRGB = NULL;

-   rb->PutMonoRow = TAG(WriteMonoStencilSpan);

-   rb->PutValues = TAG(WriteStencilPixels);

-   rb->PutMonoValues = NULL;

-}

-

-

-#undef WRITE_STENCIL

-#undef READ_STENCIL

-#undef TAG

+
+#include "spantmp_common.h"
+
+#ifndef DBG
+#define DBG 0
+#endif
+
+#ifndef HAVE_HW_STENCIL_SPANS
+#define HAVE_HW_STENCIL_SPANS 0
+#endif
+
+#ifndef HAVE_HW_STENCIL_PIXELS
+#define HAVE_HW_STENCIL_PIXELS 0
+#endif
+
+static void TAG(WriteStencilSpan)( struct gl_context *ctx,
+                                   struct gl_renderbuffer *rb,
+				   GLuint n, GLint x, GLint y,
+				   const void *values, const GLubyte mask[] )
+{
+   HW_WRITE_LOCK()
+      {
+         const GLubyte *stencil = (const GLubyte *) values;
+	 GLint x1;
+	 GLint n1;
+	 LOCAL_STENCIL_VARS;
+
+	 y = Y_FLIP(y);
+
+#if HAVE_HW_STENCIL_SPANS
+	 (void) x1; (void) n1;
+
+	 if (DBG) fprintf(stderr, "WriteStencilSpan 0..%d (x1 %d)\n",
+			  (int)n1, (int)x1);
+
+	 WRITE_STENCIL_SPAN();
+#else /* HAVE_HW_STENCIL_SPANS */
+	 HW_CLIPLOOP() 
+	    {
+	       GLint i = 0;
+	       CLIPSPAN(x,y,n,x1,n1,i);
+
+	       if (DBG) fprintf(stderr, "WriteStencilSpan %d..%d (x1 %d)\n",
+				(int)i, (int)n1, (int)x1);
+
+	       if (mask)
+	       {
+		  for (;n1>0;i++,x1++,n1--)
+		     if (mask[i])
+			WRITE_STENCIL( x1, y, stencil[i] );
+	       }
+	       else
+	       {
+		  for (;n1>0;i++,x1++,n1--)
+		     WRITE_STENCIL( x1, y, stencil[i] );
+	       }
+	    }
+	 HW_ENDCLIPLOOP();
+#endif /* !HAVE_HW_STENCIL_SPANS */
+      }
+   HW_WRITE_UNLOCK();
+}
+
+
+static void TAG(WriteStencilPixels)( struct gl_context *ctx,
+                                     struct gl_renderbuffer *rb,
+				     GLuint n,
+				     const GLint x[], const GLint y[],
+				     const void *values, const GLubyte mask[] )
+{
+   HW_WRITE_LOCK()
+      {
+         const GLubyte *stencil = (const GLubyte *) values;
+	 GLuint i;
+	 LOCAL_STENCIL_VARS;
+
+	 if (DBG) fprintf(stderr, "WriteStencilPixels\n");
+
+#if HAVE_HW_STENCIL_PIXELS
+	 (void) i;
+
+	 WRITE_STENCIL_PIXELS();
+#else /* HAVE_HW_STENCIL_PIXELS */
+	 HW_CLIPLOOP()
+	    {
+	       for (i=0;i<n;i++)
+	       {
+		  if (mask[i]) {
+		     const int fy = Y_FLIP(y[i]);
+		     if (CLIPPIXEL(x[i],fy))
+			WRITE_STENCIL( x[i], fy, stencil[i] );
+		  }
+	       }
+	    }
+	 HW_ENDCLIPLOOP();
+#endif /* !HAVE_HW_STENCIL_PIXELS */
+      }
+   HW_WRITE_UNLOCK();
+}
+
+
+/* Read stencil spans and pixels
+ */
+static void TAG(ReadStencilSpan)( struct gl_context *ctx,
+                                  struct gl_renderbuffer *rb,
+				  GLuint n, GLint x, GLint y,
+				  void *values)
+{
+   HW_READ_LOCK()
+      {
+         GLubyte *stencil = (GLubyte *) values;
+	 GLint x1,n1;
+	 LOCAL_STENCIL_VARS;
+
+	 y = Y_FLIP(y);
+
+	 if (DBG) fprintf(stderr, "ReadStencilSpan\n");
+
+#if HAVE_HW_STENCIL_SPANS
+	 (void) x1; (void) n1;
+
+	 READ_STENCIL_SPAN();
+#else /* HAVE_HW_STENCIL_SPANS */
+	 HW_CLIPLOOP() 
+	    {
+	       GLint i = 0;
+	       CLIPSPAN(x,y,n,x1,n1,i);
+	       for (;n1>0;i++,n1--)
+		  READ_STENCIL( stencil[i], (x+i), y );
+	    }
+	 HW_ENDCLIPLOOP();
+#endif /* !HAVE_HW_STENCIL_SPANS */
+      }
+   HW_READ_UNLOCK();
+}
+
+static void TAG(ReadStencilPixels)( struct gl_context *ctx,
+                                    struct gl_renderbuffer *rb,
+                                    GLuint n, const GLint x[], const GLint y[],
+				    void *values )
+{
+   HW_READ_LOCK()
+      {
+         GLubyte *stencil = (GLubyte *) values;
+	 GLuint i;
+	 LOCAL_STENCIL_VARS;
+
+	 if (DBG) fprintf(stderr, "ReadStencilPixels\n");
+ 
+#if HAVE_HW_STENCIL_PIXELS
+	 (void) i;
+
+	 READ_STENCIL_PIXELS();
+#else /* HAVE_HW_STENCIL_PIXELS */
+	 HW_CLIPLOOP()
+	    {
+	       for (i=0;i<n;i++) {
+		  int fy = Y_FLIP( y[i] );
+		  if (CLIPPIXEL( x[i], fy ))
+		     READ_STENCIL( stencil[i], x[i], fy );
+	       }
+	    }
+	 HW_ENDCLIPLOOP();
+#endif /* !HAVE_HW_STENCIL_PIXELS */
+      }
+   HW_READ_UNLOCK();
+}
+
+
+
+/**
+ * Initialize the given renderbuffer's span routines to point to
+ * the stencil functions we generated above.
+ */
+static void TAG(InitStencilPointers)(struct gl_renderbuffer *rb)
+{
+   rb->GetRow = TAG(ReadStencilSpan);
+   rb->GetValues = TAG(ReadStencilPixels);
+   rb->PutRow = TAG(WriteStencilSpan);
+   rb->PutValues = TAG(WriteStencilPixels);
+}
+
+
+#undef WRITE_STENCIL
+#undef READ_STENCIL
+#undef TAG
diff --git a/mesalib/src/mesa/drivers/dri/swrast/swrast_span.c b/mesalib/src/mesa/drivers/dri/swrast/swrast_span.c
index 772d09f5a..ba6174f9d 100644
--- a/mesalib/src/mesa/drivers/dri/swrast/swrast_span.c
+++ b/mesalib/src/mesa/drivers/dri/swrast/swrast_span.c
@@ -276,37 +276,25 @@ swrast_set_span_funcs_back(struct swrast_renderbuffer *xrb,
 	xrb->Base.GetRow = get_row_A8R8G8B8;
 	xrb->Base.GetValues = get_values_A8R8G8B8;
 	xrb->Base.PutRow = put_row_A8R8G8B8;
-	xrb->Base.PutRowRGB = put_row_rgb_A8R8G8B8;
-	xrb->Base.PutMonoRow = put_mono_row_A8R8G8B8;
 	xrb->Base.PutValues = put_values_A8R8G8B8;
-	xrb->Base.PutMonoValues = put_mono_values_A8R8G8B8;
 	break;
     case PF_X8R8G8B8:
 	xrb->Base.GetRow = get_row_X8R8G8B8;
 	xrb->Base.GetValues = get_values_X8R8G8B8;
 	xrb->Base.PutRow = put_row_X8R8G8B8;
-	xrb->Base.PutRowRGB = put_row_rgb_X8R8G8B8;
-	xrb->Base.PutMonoRow = put_mono_row_X8R8G8B8;
 	xrb->Base.PutValues = put_values_X8R8G8B8;
-	xrb->Base.PutMonoValues = put_mono_values_X8R8G8B8;
 	break;
     case PF_R5G6B5:
 	xrb->Base.GetRow = get_row_R5G6B5;
 	xrb->Base.GetValues = get_values_R5G6B5;
 	xrb->Base.PutRow = put_row_R5G6B5;
-	xrb->Base.PutRowRGB = put_row_rgb_R5G6B5;
-	xrb->Base.PutMonoRow = put_mono_row_R5G6B5;
 	xrb->Base.PutValues = put_values_R5G6B5;
-	xrb->Base.PutMonoValues = put_mono_values_R5G6B5;
 	break;
     case PF_R3G3B2:
 	xrb->Base.GetRow = get_row_R3G3B2;
 	xrb->Base.GetValues = get_values_R3G3B2;
 	xrb->Base.PutRow = put_row_R3G3B2;
-	xrb->Base.PutRowRGB = put_row_rgb_R3G3B2;
-	xrb->Base.PutMonoRow = put_mono_row_R3G3B2;
 	xrb->Base.PutValues = put_values_R3G3B2;
-	xrb->Base.PutMonoValues = put_mono_values_R3G3B2;
 	break;
     default:
 	assert(0);
@@ -332,37 +320,25 @@ swrast_set_span_funcs_front(struct swrast_renderbuffer *xrb,
 	xrb->Base.GetRow = get_row_A8R8G8B8_front;
 	xrb->Base.GetValues = get_values_A8R8G8B8_front;
 	xrb->Base.PutRow = put_row_A8R8G8B8_front;
-	xrb->Base.PutRowRGB = put_row_rgb_A8R8G8B8_front;
-	xrb->Base.PutMonoRow = put_mono_row_A8R8G8B8_front;
 	xrb->Base.PutValues = put_values_A8R8G8B8_front;
-	xrb->Base.PutMonoValues = put_mono_values_A8R8G8B8_front;
 	break;
     case PF_X8R8G8B8:
 	xrb->Base.GetRow = get_row_X8R8G8B8_front;
 	xrb->Base.GetValues = get_values_X8R8G8B8_front;
 	xrb->Base.PutRow = put_row_X8R8G8B8_front;
-	xrb->Base.PutRowRGB = put_row_rgb_X8R8G8B8_front;
-	xrb->Base.PutMonoRow = put_mono_row_X8R8G8B8_front;
 	xrb->Base.PutValues = put_values_X8R8G8B8_front;
-	xrb->Base.PutMonoValues = put_mono_values_X8R8G8B8_front;
 	break;
     case PF_R5G6B5:
 	xrb->Base.GetRow = get_row_R5G6B5_front;
 	xrb->Base.GetValues = get_values_R5G6B5_front;
 	xrb->Base.PutRow = put_row_R5G6B5_front;
-	xrb->Base.PutRowRGB = put_row_rgb_R5G6B5_front;
-	xrb->Base.PutMonoRow = put_mono_row_R5G6B5_front;
 	xrb->Base.PutValues = put_values_R5G6B5_front;
-	xrb->Base.PutMonoValues = put_mono_values_R5G6B5_front;
 	break;
     case PF_R3G3B2:
 	xrb->Base.GetRow = get_row_R3G3B2_front;
 	xrb->Base.GetValues = get_values_R3G3B2_front;
 	xrb->Base.PutRow = put_row_R3G3B2_front;
-	xrb->Base.PutRowRGB = put_row_rgb_R3G3B2_front;
-	xrb->Base.PutMonoRow = put_mono_row_R3G3B2_front;
 	xrb->Base.PutValues = put_values_R3G3B2_front;
-	xrb->Base.PutMonoValues = put_mono_values_R3G3B2_front;
 	break;
     default:
 	assert(0);
diff --git a/mesalib/src/mesa/drivers/dri/swrast/swrast_spantemp.h b/mesalib/src/mesa/drivers/dri/swrast/swrast_spantemp.h
index 0d2703ca5..50671193e 100644
--- a/mesalib/src/mesa/drivers/dri/swrast/swrast_spantemp.h
+++ b/mesalib/src/mesa/drivers/dri/swrast/swrast_spantemp.h
@@ -1,319 +1,223 @@
-/*

- * Mesa 3-D graphics library

- * Version:  6.5.1

- *

- * Copyright (C) 1999-2006  Brian Paul   All Rights Reserved.

- *

- * Permission is hereby granted, free of charge, to any person obtaining a

- * copy of this software and associated documentation files (the "Software"),

- * to deal in the Software without restriction, including without limitation

- * the rights to use, copy, modify, merge, publish, distribute, sublicense,

- * and/or sell copies of the Software, and to permit persons to whom the

- * Software is furnished to do so, subject to the following conditions:

- *

- * The above copyright notice and this permission notice shall be included

- * in all copies or substantial portions of the Software.

- *

- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS

- * OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,

- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL

- * BRIAN PAUL BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN

- * AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN

- * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.

- */

-

-

-/*

- * Modified version of swrast/s_spantemp.h for front-buffer rendering. The

- * no-mask paths use a scratch row to avoid repeated calls to the loader.

- *

- * For the mask paths we always use an array of 4 elements of RB_TYPE. This is

- * to satisfy the xorg loader requirement of an image pitch of 32 bits and

- * should be ok for other loaders also.

- */

-

-

-#ifndef _SWRAST_SPANTEMP_ONCE

-#define _SWRAST_SPANTEMP_ONCE

-

-static INLINE void

-PUT_PIXEL( struct gl_context *glCtx, GLint x, GLint y, GLvoid *p )

-{

-    __DRIcontext *ctx = swrast_context(glCtx)->cPriv;

-    __DRIdrawable *draw = swrast_drawable(glCtx->DrawBuffer)->dPriv;

-

-    __DRIscreen *screen = ctx->driScreenPriv;

-

-    screen->swrast_loader->putImage(draw, __DRI_SWRAST_IMAGE_OP_DRAW,

-				    x, y, 1, 1, (char *)p,

-				    draw->loaderPrivate);

-}

-

-

-static INLINE void

-GET_PIXEL( struct gl_context *glCtx, GLint x, GLint y, GLubyte *p )

-{

-    __DRIcontext *ctx = swrast_context(glCtx)->cPriv;

-    __DRIdrawable *read = swrast_drawable(glCtx->ReadBuffer)->dPriv;

-

-    __DRIscreen *screen = ctx->driScreenPriv;

-

-    screen->swrast_loader->getImage(read, x, y, 1, 1, (char *)p,

-				    read->loaderPrivate);

-}

-

-static INLINE void

-PUT_ROW( struct gl_context *glCtx, GLint x, GLint y, GLuint n, char *row )

-{

-    __DRIcontext *ctx = swrast_context(glCtx)->cPriv;

-    __DRIdrawable *draw = swrast_drawable(glCtx->DrawBuffer)->dPriv;

-

-    __DRIscreen *screen = ctx->driScreenPriv;

-

-    screen->swrast_loader->putImage(draw, __DRI_SWRAST_IMAGE_OP_DRAW,

-				    x, y, n, 1, row,

-				    draw->loaderPrivate);

-}

-

-static INLINE void

-GET_ROW( struct gl_context *glCtx, GLint x, GLint y, GLuint n, char *row )

-{

-    __DRIcontext *ctx = swrast_context(glCtx)->cPriv;

-    __DRIdrawable *read = swrast_drawable(glCtx->ReadBuffer)->dPriv;

-

-    __DRIscreen *screen = ctx->driScreenPriv;

-

-    screen->swrast_loader->getImage(read, x, y, n, 1, row,

-				    read->loaderPrivate);

-}

-

-#endif /* _SWRAST_SPANTEMP_ONCE */

-

-

-/*

- * Templates for the span/pixel-array write/read functions called via

- * the gl_renderbuffer's GetRow, GetValues, PutRow, PutMonoRow, PutValues

- * and PutMonoValues functions.

- *

- * Define the following macros before including this file:

- *   NAME(BASE)  to generate the function name (i.e. add prefix or suffix)

- *   RB_TYPE  the renderbuffer DataType

- *   SPAN_VARS  to declare any local variables

- *   INIT_PIXEL_PTR(P, X, Y)  to initialize a pointer to a pixel

- *   INC_PIXEL_PTR(P)  to increment a pixel pointer by one pixel

- *   STORE_PIXEL(DST, X, Y, VALUE)  to store pixel values in buffer

- *   FETCH_PIXEL(DST, SRC)  to fetch pixel values from buffer

- *

- * Note that in the STORE_PIXEL macros, we also pass in the (X,Y) coordinates

- * for the pixels to be stored.  This is useful when dithering and probably

- * ignored otherwise.

- */

-

-#include "main/macros.h"

-

-

-#if !defined(RB_COMPONENTS)

-#define RB_COMPONENTS 4

-#endif

-

-

-static void

-NAME(get_row)( struct gl_context *ctx, struct gl_renderbuffer *rb,

-               GLuint count, GLint x, GLint y, void *values )

-{

-#ifdef SPAN_VARS

-   SPAN_VARS

-#endif

-   RB_TYPE (*dest)[RB_COMPONENTS] = (RB_TYPE (*)[RB_COMPONENTS]) values;

-   GLuint i;

-   char *row = swrast_drawable(ctx->ReadBuffer)->row;

-   INIT_PIXEL_PTR(pixel, x, y);

-   GET_ROW( ctx, x, YFLIP(xrb, y), count, row );

-   for (i = 0; i < count; i++) {

-      FETCH_PIXEL(dest[i], pixel);

-      INC_PIXEL_PTR(pixel);

-   }

-   (void) rb;

-}

-

-

-static void

-NAME(get_values)( struct gl_context *ctx, struct gl_renderbuffer *rb,

-                  GLuint count, const GLint x[], const GLint y[], void *values )

-{

-#ifdef SPAN_VARS

-   SPAN_VARS

-#endif

-   RB_TYPE (*dest)[RB_COMPONENTS] = (RB_TYPE (*)[RB_COMPONENTS]) values;

-   GLuint i;

-   for (i = 0; i < count; i++) {

-      RB_TYPE pixel[4];

-      GET_PIXEL(ctx, x[i], YFLIP(xrb, y[i]), pixel);

-      FETCH_PIXEL(dest[i], pixel);

-   }

-   (void) rb;

-}

-

-

-static void

-NAME(put_row)( struct gl_context *ctx, struct gl_renderbuffer *rb,

-               GLuint count, GLint x, GLint y,

-               const void *values, const GLubyte mask[] )

-{

-#ifdef SPAN_VARS

-   SPAN_VARS

-#endif

-   const RB_TYPE (*src)[RB_COMPONENTS] = (const RB_TYPE (*)[RB_COMPONENTS]) values;

-   GLuint i;

-   if (mask) {

-      for (i = 0; i < count; i++) {

-         if (mask[i]) {

-            RB_TYPE row[4];

-            INIT_PIXEL_PTR(pixel, x, y);

-            STORE_PIXEL(pixel, x + i, y, src[i]);

-            PUT_PIXEL(ctx, x + i, YFLIP(xrb, y), pixel);

-         }

-      }

-   }

-   else {

-      char *row = swrast_drawable(ctx->DrawBuffer)->row;

-      INIT_PIXEL_PTR(pixel, x, y);

-      for (i = 0; i < count; i++) {

-         STORE_PIXEL(pixel, x + i, y, src[i]);

-         INC_PIXEL_PTR(pixel);

-      }

-      PUT_ROW( ctx, x, YFLIP(xrb, y), count, row );

-   }

-   (void) rb;

-}

-

-

-static void

-NAME(put_row_rgb)( struct gl_context *ctx, struct gl_renderbuffer *rb,

-                   GLuint count, GLint x, GLint y,

-                   const void *values, const GLubyte mask[] )

-{

-#ifdef SPAN_VARS

-   SPAN_VARS

-#endif

-   const RB_TYPE (*src)[3] = (const RB_TYPE (*)[3]) values;

-   GLuint i;

-   if (mask) {

-      for (i = 0; i < count; i++) {

-         if (mask[i]) {

-            RB_TYPE row[4];

-            INIT_PIXEL_PTR(pixel, x, y);

-#ifdef STORE_PIXEL_RGB

-            STORE_PIXEL_RGB(pixel, x + i, y, src[i]);

-#else

-            STORE_PIXEL(pixel, x + i, y, src[i]);

-#endif

-            PUT_PIXEL(ctx, x + i, YFLIP(xrb, y), pixel);

-         }

-      }

-   }

-   else {

-      char *row = swrast_drawable(ctx->DrawBuffer)->row;

-      INIT_PIXEL_PTR(pixel, x, y);

-      for (i = 0; i < count; i++) {

-#ifdef STORE_PIXEL_RGB

-         STORE_PIXEL_RGB(pixel, x + i, y, src[i]);

-#else

-         STORE_PIXEL(pixel, x + i, y, src[i]);

-#endif

-         INC_PIXEL_PTR(pixel);

-      }

-      PUT_ROW( ctx, x, YFLIP(xrb, y), count, row );

-   }

-   (void) rb;

-}

-

-

-static void

-NAME(put_mono_row)( struct gl_context *ctx, struct gl_renderbuffer *rb,

-                    GLuint count, GLint x, GLint y,

-                    const void *value, const GLubyte mask[] )

-{

-#ifdef SPAN_VARS

-   SPAN_VARS

-#endif

-   const RB_TYPE *src = (const RB_TYPE *) value;

-   GLuint i;

-   if (mask) {

-      for (i = 0; i < count; i++) {

-         if (mask[i]) {

-            RB_TYPE row[4];

-            INIT_PIXEL_PTR(pixel, x, y);

-            STORE_PIXEL(pixel, x + i, y, src);

-            PUT_PIXEL(ctx, x + i, YFLIP(xrb, y), pixel);

-         }

-      }

-   }

-   else {

-      char *row = swrast_drawable(ctx->DrawBuffer)->row;

-      INIT_PIXEL_PTR(pixel, x, y);

-      for (i = 0; i < count; i++) {

-         STORE_PIXEL(pixel, x + i, y, src);

-         INC_PIXEL_PTR(pixel);

-      }

-      PUT_ROW( ctx, x, YFLIP(xrb, y), count, row );

-   }

-   (void) rb;

-}

-

-

-static void

-NAME(put_values)( struct gl_context *ctx, struct gl_renderbuffer *rb,

-                  GLuint count, const GLint x[], const GLint y[],

-                  const void *values, const GLubyte mask[] )

-{

-#ifdef SPAN_VARS

-   SPAN_VARS

-#endif

-   const RB_TYPE (*src)[RB_COMPONENTS] = (const RB_TYPE (*)[RB_COMPONENTS]) values;

-   GLuint i;

-   ASSERT(mask);

-   for (i = 0; i < count; i++) {

-      if (mask[i]) {

-         RB_TYPE row[4];

-         INIT_PIXEL_PTR(pixel, x, y);

-         STORE_PIXEL(pixel, x[i], y[i], src[i]);

-         PUT_PIXEL(ctx, x[i], YFLIP(xrb, y[i]), pixel);

-      }

-   }

-   (void) rb;

-}

-

-

-static void

-NAME(put_mono_values)( struct gl_context *ctx, struct gl_renderbuffer *rb,

-                       GLuint count, const GLint x[], const GLint y[],

-                       const void *value, const GLubyte mask[] )

-{

-#ifdef SPAN_VARS

-   SPAN_VARS

-#endif

-   const RB_TYPE *src = (const RB_TYPE *) value;

-   GLuint i;

-   ASSERT(mask);

-   for (i = 0; i < count; i++) {

-      if (mask[i]) {

-         RB_TYPE row[4];

-         INIT_PIXEL_PTR(pixel, x, y);

-         STORE_PIXEL(pixel, x[i], y[i], src);

-         PUT_PIXEL(ctx, x[i], YFLIP(xrb, y[i]), pixel);

-      }

-   }

-   (void) rb;

-}

-

-

-#undef NAME

-#undef RB_TYPE

-#undef RB_COMPONENTS

-#undef SPAN_VARS

-#undef INIT_PIXEL_PTR

-#undef INC_PIXEL_PTR

-#undef STORE_PIXEL

-#undef STORE_PIXEL_RGB

-#undef FETCH_PIXEL

+/*
+ * Mesa 3-D graphics library
+ * Version:  6.5.1
+ *
+ * Copyright (C) 1999-2006  Brian Paul   All Rights Reserved.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included
+ * in all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS
+ * OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * BRIAN PAUL BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN
+ * AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
+ */
+
+
+/*
+ * Modified version of swrast/s_spantemp.h for front-buffer rendering. The
+ * no-mask paths use a scratch row to avoid repeated calls to the loader.
+ *
+ * For the mask paths we always use an array of 4 elements of RB_TYPE. This is
+ * to satisfy the xorg loader requirement of an image pitch of 32 bits and
+ * should be ok for other loaders also.
+ */
+
+
+#ifndef _SWRAST_SPANTEMP_ONCE
+#define _SWRAST_SPANTEMP_ONCE
+
+static INLINE void
+PUT_PIXEL( struct gl_context *glCtx, GLint x, GLint y, GLvoid *p )
+{
+    __DRIcontext *ctx = swrast_context(glCtx)->cPriv;
+    __DRIdrawable *draw = swrast_drawable(glCtx->DrawBuffer)->dPriv;
+
+    __DRIscreen *screen = ctx->driScreenPriv;
+
+    screen->swrast_loader->putImage(draw, __DRI_SWRAST_IMAGE_OP_DRAW,
+				    x, y, 1, 1, (char *)p,
+				    draw->loaderPrivate);
+}
+
+
+static INLINE void
+GET_PIXEL( struct gl_context *glCtx, GLint x, GLint y, GLubyte *p )
+{
+    __DRIcontext *ctx = swrast_context(glCtx)->cPriv;
+    __DRIdrawable *read = swrast_drawable(glCtx->ReadBuffer)->dPriv;
+
+    __DRIscreen *screen = ctx->driScreenPriv;
+
+    screen->swrast_loader->getImage(read, x, y, 1, 1, (char *)p,
+				    read->loaderPrivate);
+}
+
+static INLINE void
+PUT_ROW( struct gl_context *glCtx, GLint x, GLint y, GLuint n, char *row )
+{
+    __DRIcontext *ctx = swrast_context(glCtx)->cPriv;
+    __DRIdrawable *draw = swrast_drawable(glCtx->DrawBuffer)->dPriv;
+
+    __DRIscreen *screen = ctx->driScreenPriv;
+
+    screen->swrast_loader->putImage(draw, __DRI_SWRAST_IMAGE_OP_DRAW,
+				    x, y, n, 1, row,
+				    draw->loaderPrivate);
+}
+
+static INLINE void
+GET_ROW( struct gl_context *glCtx, GLint x, GLint y, GLuint n, char *row )
+{
+    __DRIcontext *ctx = swrast_context(glCtx)->cPriv;
+    __DRIdrawable *read = swrast_drawable(glCtx->ReadBuffer)->dPriv;
+
+    __DRIscreen *screen = ctx->driScreenPriv;
+
+    screen->swrast_loader->getImage(read, x, y, n, 1, row,
+				    read->loaderPrivate);
+}
+
+#endif /* _SWRAST_SPANTEMP_ONCE */
+
+
+/*
+ * Templates for the span/pixel-array write/read functions called via
+ * the gl_renderbuffer's GetRow, GetValues, PutRow and PutValues.
+ *
+ * Define the following macros before including this file:
+ *   NAME(BASE)  to generate the function name (i.e. add prefix or suffix)
+ *   RB_TYPE  the renderbuffer DataType
+ *   SPAN_VARS  to declare any local variables
+ *   INIT_PIXEL_PTR(P, X, Y)  to initialize a pointer to a pixel
+ *   INC_PIXEL_PTR(P)  to increment a pixel pointer by one pixel
+ *   STORE_PIXEL(DST, X, Y, VALUE)  to store pixel values in buffer
+ *   FETCH_PIXEL(DST, SRC)  to fetch pixel values from buffer
+ *
+ * Note that in the STORE_PIXEL macros, we also pass in the (X,Y) coordinates
+ * for the pixels to be stored.  This is useful when dithering and probably
+ * ignored otherwise.
+ */
+
+#include "main/macros.h"
+
+
+#if !defined(RB_COMPONENTS)
+#define RB_COMPONENTS 4
+#endif
+
+
+static void
+NAME(get_row)( struct gl_context *ctx, struct gl_renderbuffer *rb,
+               GLuint count, GLint x, GLint y, void *values )
+{
+#ifdef SPAN_VARS
+   SPAN_VARS
+#endif
+   RB_TYPE (*dest)[RB_COMPONENTS] = (RB_TYPE (*)[RB_COMPONENTS]) values;
+   GLuint i;
+   char *row = swrast_drawable(ctx->ReadBuffer)->row;
+   INIT_PIXEL_PTR(pixel, x, y);
+   GET_ROW( ctx, x, YFLIP(xrb, y), count, row );
+   for (i = 0; i < count; i++) {
+      FETCH_PIXEL(dest[i], pixel);
+      INC_PIXEL_PTR(pixel);
+   }
+   (void) rb;
+}
+
+
+static void
+NAME(get_values)( struct gl_context *ctx, struct gl_renderbuffer *rb,
+                  GLuint count, const GLint x[], const GLint y[], void *values )
+{
+#ifdef SPAN_VARS
+   SPAN_VARS
+#endif
+   RB_TYPE (*dest)[RB_COMPONENTS] = (RB_TYPE (*)[RB_COMPONENTS]) values;
+   GLuint i;
+   for (i = 0; i < count; i++) {
+      RB_TYPE pixel[4];
+      GET_PIXEL(ctx, x[i], YFLIP(xrb, y[i]), pixel);
+      FETCH_PIXEL(dest[i], pixel);
+   }
+   (void) rb;
+}
+
+
+static void
+NAME(put_row)( struct gl_context *ctx, struct gl_renderbuffer *rb,
+               GLuint count, GLint x, GLint y,
+               const void *values, const GLubyte mask[] )
+{
+#ifdef SPAN_VARS
+   SPAN_VARS
+#endif
+   const RB_TYPE (*src)[RB_COMPONENTS] = (const RB_TYPE (*)[RB_COMPONENTS]) values;
+   GLuint i;
+   if (mask) {
+      for (i = 0; i < count; i++) {
+         if (mask[i]) {
+            RB_TYPE row[4];
+            INIT_PIXEL_PTR(pixel, x, y);
+            STORE_PIXEL(pixel, x + i, y, src[i]);
+            PUT_PIXEL(ctx, x + i, YFLIP(xrb, y), pixel);
+         }
+      }
+   }
+   else {
+      char *row = swrast_drawable(ctx->DrawBuffer)->row;
+      INIT_PIXEL_PTR(pixel, x, y);
+      for (i = 0; i < count; i++) {
+         STORE_PIXEL(pixel, x + i, y, src[i]);
+         INC_PIXEL_PTR(pixel);
+      }
+      PUT_ROW( ctx, x, YFLIP(xrb, y), count, row );
+   }
+   (void) rb;
+}
+
+
+static void
+NAME(put_values)( struct gl_context *ctx, struct gl_renderbuffer *rb,
+                  GLuint count, const GLint x[], const GLint y[],
+                  const void *values, const GLubyte mask[] )
+{
+#ifdef SPAN_VARS
+   SPAN_VARS
+#endif
+   const RB_TYPE (*src)[RB_COMPONENTS] = (const RB_TYPE (*)[RB_COMPONENTS]) values;
+   GLuint i;
+   ASSERT(mask);
+   for (i = 0; i < count; i++) {
+      if (mask[i]) {
+         RB_TYPE row[4];
+         INIT_PIXEL_PTR(pixel, x, y);
+         STORE_PIXEL(pixel, x[i], y[i], src[i]);
+         PUT_PIXEL(ctx, x[i], YFLIP(xrb, y[i]), pixel);
+      }
+   }
+   (void) rb;
+}
+
+
+
+
+#undef NAME
+#undef RB_TYPE
+#undef RB_COMPONENTS
+#undef SPAN_VARS
+#undef INIT_PIXEL_PTR
+#undef INC_PIXEL_PTR
+#undef STORE_PIXEL
+#undef STORE_PIXEL_RGB
+#undef FETCH_PIXEL
diff --git a/mesalib/src/mesa/drivers/windows/gdi/wmesa.c b/mesalib/src/mesa/drivers/windows/gdi/wmesa.c
index 675ffdaaa..255d811ca 100644
--- a/mesalib/src/mesa/drivers/windows/gdi/wmesa.c
+++ b/mesalib/src/mesa/drivers/windows/gdi/wmesa.c
@@ -526,61 +526,6 @@ static void write_rgba_span_front(struct gl_context *ctx,
    }
 }
 
-/* Write a horizontal span of RGB color pixels with a boolean mask. */
-static void write_rgb_span_front(struct gl_context *ctx, 
-				 struct gl_renderbuffer *rb, 
-				 GLuint n, GLint x, GLint y,
-				 const void *values, 
-				 const GLubyte *mask)
-{
-    const GLubyte (*rgb)[3] = (const GLubyte (*)[3])values;
-    WMesaContext pwc = wmesa_context(ctx);
-    GLuint i;
-    
-    (void) ctx;
-    y=FLIP(y);
-    if (mask) {
-	for (i=0; i<n; i++)
-	    if (mask[i])
-		SetPixel(pwc->hDC, x+i, y, RGB(rgb[i][RCOMP], rgb[i][GCOMP], 
-					       rgb[i][BCOMP]));
-    }
-    else {
-	for (i=0; i<n; i++)
-	    SetPixel(pwc->hDC, x+i, y, RGB(rgb[i][RCOMP], rgb[i][GCOMP], 
-					   rgb[i][BCOMP]));
-    }
-    
-}
-
-/*
- * Write a horizontal span of pixels with a boolean mask.  The current color
- * is used for all pixels.
- */
-static void write_mono_rgba_span_front(struct gl_context *ctx, 
-                                       struct gl_renderbuffer *rb,
-                                       GLuint n, GLint x, GLint y,
-                                       const void *value, 
-                                       const GLubyte *mask)
-{
-    const GLchan *color = (const GLchan *)value;
-    GLuint i;
-    WMesaContext pwc = wmesa_context(ctx);
-    COLORREF colorref;
-
-    (void) ctx;
-    colorref = RGB(color[RCOMP], color[GCOMP], color[BCOMP]);
-    y=FLIP(y);
-    if (mask) {
-	for (i=0; i<n; i++)
-	    if (mask[i])
-		SetPixel(pwc->hDC, x+i, y, colorref);
-    }
-    else
-	for (i=0; i<n; i++)
-	    SetPixel(pwc->hDC, x+i, y, colorref);
-
-}
 
 /* Write an array of RGBA pixels with a boolean mask. */
 static void write_rgba_pixels_front(struct gl_context *ctx, 
@@ -603,28 +548,6 @@ static void write_rgba_pixels_front(struct gl_context *ctx,
 
 
 
-/*
- * Write an array of pixels with a boolean mask.  The current color
- * is used for all pixels.
- */
-static void write_mono_rgba_pixels_front(struct gl_context *ctx, 
-                                         struct gl_renderbuffer *rb,
-                                         GLuint n,
-                                         const GLint x[], const GLint y[],
-                                         const void *value,
-                                         const GLubyte *mask)
-{
-    const GLchan *color = (const GLchan *)value;
-    GLuint i;
-    WMesaContext pwc = wmesa_context(ctx);
-    COLORREF colorref;
-    (void) ctx;
-    colorref = RGB(color[RCOMP], color[GCOMP], color[BCOMP]);
-    for (i=0; i<n; i++)
-	if (mask[i])
-	    SetPixel(pwc->hDC, x[i], FLIP(y[i]), colorref);
-}
-
 /* Read a horizontal span of color pixels. */
 static void read_rgba_span_front(struct gl_context *ctx, 
                                  struct gl_renderbuffer *rb,
@@ -706,64 +629,6 @@ static void write_rgba_span_32(struct gl_context *ctx,
 }
 
 
-/* Write a horizontal span of RGB color pixels with a boolean mask. */
-static void write_rgb_span_32(struct gl_context *ctx, 
-			      struct gl_renderbuffer *rb, 
-			      GLuint n, GLint x, GLint y,
-			      const void *values, 
-			      const GLubyte *mask)
-{
-    const GLubyte (*rgb)[3] = (const GLubyte (*)[3])values;
-    WMesaFramebuffer pwfb = wmesa_framebuffer(ctx->DrawBuffer);
-    GLuint i;
-    LPDWORD lpdw;
-
-    (void) ctx;
-    
-    y=FLIP(y);
-    lpdw = ((LPDWORD)(pwfb->pbPixels + pwfb->ScanWidth * y)) + x;
-    if (mask) {
-	for (i=0; i<n; i++)
-	    if (mask[i])
-                lpdw[i] = BGR32(rgb[i][RCOMP], rgb[i][GCOMP], 
-				rgb[i][BCOMP]);
-    }
-    else {
-	for (i=0; i<n; i++)
-                *lpdw++ = BGR32(rgb[i][RCOMP], rgb[i][GCOMP], 
-				rgb[i][BCOMP]);
-    }
-}
-
-/*
- * Write a horizontal span of pixels with a boolean mask.  The current color
- * is used for all pixels.
- */
-static void write_mono_rgba_span_32(struct gl_context *ctx, 
-				    struct gl_renderbuffer *rb,
-				    GLuint n, GLint x, GLint y,
-				    const void *value, 
-				    const GLubyte *mask)
-{
-    const GLchan *color = (const GLchan *)value;
-    LPDWORD lpdw;
-    DWORD pixel;
-    GLuint i;
-    WMesaFramebuffer pwfb = wmesa_framebuffer(ctx->DrawBuffer);
-    lpdw = ((LPDWORD)(pwfb->pbPixels + pwfb->ScanWidth * y)) + x;
-    y=FLIP(y);
-    pixel = BGR32(color[RCOMP], color[GCOMP], color[BCOMP]);
-    if (mask) {
-	for (i=0; i<n; i++)
-	    if (mask[i])
-                lpdw[i] = pixel;
-    }
-    else
-	for (i=0; i<n; i++)
-                *lpdw++ = pixel;
-
-}
-
 /* Write an array of RGBA pixels with a boolean mask. */
 static void write_rgba_pixels_32(struct gl_context *ctx, 
 				 struct gl_renderbuffer *rb,
@@ -780,25 +645,6 @@ static void write_rgba_pixels_32(struct gl_context *ctx,
 			 rgba[i][RCOMP], rgba[i][GCOMP], rgba[i][BCOMP]);
 }
 
-/*
- * Write an array of pixels with a boolean mask.  The current color
- * is used for all pixels.
- */
-static void write_mono_rgba_pixels_32(struct gl_context *ctx, 
-				      struct gl_renderbuffer *rb,
-				      GLuint n,
-				      const GLint x[], const GLint y[],
-				      const void *value,
-				      const GLubyte *mask)
-{
-    const GLchan *color = (const GLchan *)value;
-    GLuint i;
-    WMesaFramebuffer pwfb = wmesa_framebuffer(ctx->DrawBuffer);
-    for (i=0; i<n; i++)
-	if (mask[i])
-	    WMSETPIXEL32(pwfb, FLIP(y[i]),x[i],color[RCOMP],
-			 color[GCOMP], color[BCOMP]);
-}
 
 /* Read a horizontal span of color pixels. */
 static void read_rgba_span_32(struct gl_context *ctx, 
@@ -892,71 +738,6 @@ static void write_rgba_span_24(struct gl_context *ctx,
 }
 
 
-/* Write a horizontal span of RGB color pixels with a boolean mask. */
-static void write_rgb_span_24(struct gl_context *ctx, 
-			      struct gl_renderbuffer *rb, 
-			      GLuint n, GLint x, GLint y,
-			      const void *values, 
-			      const GLubyte *mask)
-{
-    const GLubyte (*rgb)[3] = (const GLubyte (*)[3])values;
-    WMesaFramebuffer pwfb = wmesa_framebuffer(ctx->DrawBuffer);
-    GLuint i;
-    LPBYTE lpb;
-
-    (void) ctx;
-    
-    y=FLIP(y);
-    lpb = ((LPBYTE)(pwfb->pbPixels + pwfb->ScanWidth * y)) + (3 * x);
-    if (mask) {
-	for (i=0; i<n; i++)
-	    if (mask[i]) {
-            lpb[3*i] = rgb[i][BCOMP];
-            lpb[3*i+1] = rgb[i][GCOMP];
-            lpb[3*i+2] = rgb[i][RCOMP];
-	    }
-    }
-    else {
-    	for (i=0; i<n; i++) {
-    		*lpb++ = rgb[i][BCOMP];
-    		*lpb++ = rgb[i][GCOMP];
-    		*lpb++ = rgb[i][RCOMP];
-    	}
-    }
-}
-
-/*
- * Write a horizontal span of pixels with a boolean mask.  The current color
- * is used for all pixels.
- */
-static void write_mono_rgba_span_24(struct gl_context *ctx, 
-				    struct gl_renderbuffer *rb,
-				    GLuint n, GLint x, GLint y,
-				    const void *value, 
-				    const GLubyte *mask)
-{
-    const GLchan *color = (const GLchan *)value;
-    LPBYTE lpb;
-    GLuint i;
-    WMesaFramebuffer pwfb = wmesa_framebuffer(ctx->DrawBuffer);
-    lpb = ((LPBYTE)(pwfb->pbPixels + pwfb->ScanWidth * y)) + (3 * x);
-    y=FLIP(y);
-    if (mask) {
-	for (i=0; i<n; i++)
-	    if (mask[i]) {
-	    	lpb[3*i] = color[BCOMP];
-	    	lpb[3*i+1] = color[GCOMP];
-	    	lpb[3*i+2] = color[RCOMP];
-	    }
-    }
-    else
-	for (i=0; i<n; i++) {
-		*lpb++ = color[BCOMP];
-		*lpb++ = color[GCOMP];
-		*lpb++ = color[RCOMP];		
-	}
-}
-
 /* Write an array of RGBA pixels with a boolean mask. */
 static void write_rgba_pixels_24(struct gl_context *ctx, 
 				 struct gl_renderbuffer *rb,
@@ -973,25 +754,6 @@ static void write_rgba_pixels_24(struct gl_context *ctx,
 			 rgba[i][RCOMP], rgba[i][GCOMP], rgba[i][BCOMP]);
 }
 
-/*
- * Write an array of pixels with a boolean mask.  The current color
- * is used for all pixels.
- */
-static void write_mono_rgba_pixels_24(struct gl_context *ctx, 
-				      struct gl_renderbuffer *rb,
-				      GLuint n,
-				      const GLint x[], const GLint y[],
-				      const void *value,
-				      const GLubyte *mask)
-{
-    const GLchan *color = (const GLchan *)value;
-    GLuint i;
-    WMesaFramebuffer pwfb = wmesa_framebuffer(ctx->DrawBuffer);
-    for (i=0; i<n; i++)
-	if (mask[i])
-	    WMSETPIXEL24(pwfb, FLIP(y[i]),x[i],color[RCOMP],
-			 color[GCOMP], color[BCOMP]);
-}
 
 /* Read a horizontal span of color pixels. */
 static void read_rgba_span_24(struct gl_context *ctx, 
@@ -1077,64 +839,6 @@ static void write_rgba_span_16(struct gl_context *ctx,
 }
 
 
-/* Write a horizontal span of RGB color pixels with a boolean mask. */
-static void write_rgb_span_16(struct gl_context *ctx, 
-			      struct gl_renderbuffer *rb, 
-			      GLuint n, GLint x, GLint y,
-			      const void *values, 
-			      const GLubyte *mask)
-{
-    const GLubyte (*rgb)[3] = (const GLubyte (*)[3])values;
-    WMesaFramebuffer pwfb = wmesa_framebuffer(ctx->DrawBuffer);
-    GLuint i;
-    LPWORD lpw;
-
-    (void) ctx;
-    
-    y=FLIP(y);
-    lpw = ((LPWORD)(pwfb->pbPixels + pwfb->ScanWidth * y)) + x;
-    if (mask) {
-	for (i=0; i<n; i++)
-	    if (mask[i])
-                lpw[i] = BGR16(rgb[i][RCOMP], rgb[i][GCOMP], 
-			       rgb[i][BCOMP]);
-    }
-    else {
-	for (i=0; i<n; i++)
-                *lpw++ = BGR16(rgb[i][RCOMP], rgb[i][GCOMP], 
-			       rgb[i][BCOMP]);
-    }
-}
-
-/*
- * Write a horizontal span of pixels with a boolean mask.  The current color
- * is used for all pixels.
- */
-static void write_mono_rgba_span_16(struct gl_context *ctx, 
-				    struct gl_renderbuffer *rb,
-				    GLuint n, GLint x, GLint y,
-				    const void *value, 
-				    const GLubyte *mask)
-{
-    const GLchan *color = (const GLchan *)value;
-    LPWORD lpw;
-    WORD pixel;
-    GLuint i;
-    WMesaFramebuffer pwfb = wmesa_framebuffer(ctx->DrawBuffer);
-    (void) ctx;
-    lpw = ((LPWORD)(pwfb->pbPixels + pwfb->ScanWidth * y)) + x;
-    y=FLIP(y);
-    pixel = BGR16(color[RCOMP], color[GCOMP], color[BCOMP]);
-    if (mask) {
-	for (i=0; i<n; i++)
-	    if (mask[i])
-                lpw[i] = pixel;
-    }
-    else
-	for (i=0; i<n; i++)
-                *lpw++ = pixel;
-
-}
 
 /* Write an array of RGBA pixels with a boolean mask. */
 static void write_rgba_pixels_16(struct gl_context *ctx, 
@@ -1153,26 +857,6 @@ static void write_rgba_pixels_16(struct gl_context *ctx,
 			 rgba[i][RCOMP], rgba[i][GCOMP], rgba[i][BCOMP]);
 }
 
-/*
- * Write an array of pixels with a boolean mask.  The current color
- * is used for all pixels.
- */
-static void write_mono_rgba_pixels_16(struct gl_context *ctx, 
-				      struct gl_renderbuffer *rb,
-				      GLuint n,
-				      const GLint x[], const GLint y[],
-				      const void *value,
-				      const GLubyte *mask)
-{
-    const GLchan *color = (const GLchan *)value;
-    GLuint i;
-    WMesaFramebuffer pwfb = wmesa_framebuffer(ctx->DrawBuffer);
-    (void) ctx;
-    for (i=0; i<n; i++)
-	if (mask[i])
-	    WMSETPIXEL16(pwfb, FLIP(y[i]),x[i],color[RCOMP],
-			 color[GCOMP], color[BCOMP]);
-}
 
 /* Read a horizontal span of color pixels. */
 static void read_rgba_span_16(struct gl_context *ctx, 
@@ -1270,10 +954,7 @@ wmesa_set_renderbuffer_funcs(struct gl_renderbuffer *rb, int pixelformat,
 	switch(pixelformat) {
 	case PF_5R6G5B:
 	    rb->PutRow = write_rgba_span_16;
-	    rb->PutRowRGB = write_rgb_span_16;
-	    rb->PutMonoRow = write_mono_rgba_span_16;
 	    rb->PutValues = write_rgba_pixels_16;
-	    rb->PutMonoValues = write_mono_rgba_pixels_16;
 	    rb->GetRow = read_rgba_span_16;
 	    rb->GetValues = read_rgba_pixels_16;
 	    break;
@@ -1281,22 +962,16 @@ wmesa_set_renderbuffer_funcs(struct gl_renderbuffer *rb, int pixelformat,
 		if (cColorBits == 24)
 		{
 		    rb->PutRow = write_rgba_span_24;
-		    rb->PutRowRGB = write_rgb_span_24;
-		    rb->PutMonoRow = write_mono_rgba_span_24;
 		    rb->PutValues = write_rgba_pixels_24;
-		    rb->PutMonoValues = write_mono_rgba_pixels_24;
 		    rb->GetRow = read_rgba_span_24;
 		    rb->GetValues = read_rgba_pixels_24;
 		}
 		else
 		{
-	        rb->PutRow = write_rgba_span_32;
-	        rb->PutRowRGB = write_rgb_span_32;
-	        rb->PutMonoRow = write_mono_rgba_span_32;
-	        rb->PutValues = write_rgba_pixels_32;
-	        rb->PutMonoValues = write_mono_rgba_pixels_32;
-	        rb->GetRow = read_rgba_span_32;
-	        rb->GetValues = read_rgba_pixels_32;
+                    rb->PutRow = write_rgba_span_32;
+                    rb->PutValues = write_rgba_pixels_32;
+                    rb->GetRow = read_rgba_span_32;
+                    rb->GetValues = read_rgba_pixels_32;
 		}
 	    break;
 	default:
@@ -1306,10 +981,7 @@ wmesa_set_renderbuffer_funcs(struct gl_renderbuffer *rb, int pixelformat,
     else {
         /* front buffer (actual Windows window) */
 	rb->PutRow = write_rgba_span_front;
-	rb->PutRowRGB = write_rgb_span_front;
-	rb->PutMonoRow = write_mono_rgba_span_front;
 	rb->PutValues = write_rgba_pixels_front;
-	rb->PutMonoValues = write_mono_rgba_pixels_front;
 	rb->GetRow = read_rgba_span_front;
 	rb->GetValues = read_rgba_pixels_front;
     }
diff --git a/mesalib/src/mesa/main/dd.h b/mesalib/src/mesa/main/dd.h
index 01cfff8ab..5816faa78 100644
--- a/mesalib/src/mesa/main/dd.h
+++ b/mesalib/src/mesa/main/dd.h
@@ -491,6 +491,8 @@ struct dd_function_table {
    void (*FreeTextureImageBuffer)( struct gl_context *ctx, struct gl_texture_image *tImage );
 
    /** Map a slice of a texture image into user space.
+    * Note: for GL_TEXTURE_1D_ARRAY, height must be 1, y must be 0 and slice
+    * indicates the 1D array index.
     * \param texImage  the texture image
     * \param slice  the 3D image slice or array texture slice
     * \param x, y, w, h  region of interest
diff --git a/mesalib/src/mesa/main/format_pack.c b/mesalib/src/mesa/main/format_pack.c
index 390b494c0..840559bac 100644
--- a/mesalib/src/mesa/main/format_pack.c
+++ b/mesalib/src/mesa/main/format_pack.c
@@ -2494,3 +2494,78 @@ _mesa_pack_uint_24_8_depth_stencil_row(gl_format format, GLuint n,
       return;
    }
 }
+
+
+
+/**
+ * Convert a boolean color mask to a packed color where each channel of
+ * the packed value at dst will be 0 or ~0 depending on the colorMask.
+ */
+void
+_mesa_pack_colormask(gl_format format, const GLubyte colorMask[4], void *dst)
+{
+   GLfloat maskColor[4];
+
+   switch (_mesa_get_format_datatype(format)) {
+   case GL_UNSIGNED_NORMALIZED:
+      /* simple: 1.0 will convert to ~0 in the right bit positions */
+      maskColor[0] = colorMask[0] ? 1.0 : 0.0;
+      maskColor[1] = colorMask[1] ? 1.0 : 0.0;
+      maskColor[2] = colorMask[2] ? 1.0 : 0.0;
+      maskColor[3] = colorMask[3] ? 1.0 : 0.0;
+      _mesa_pack_float_rgba_row(format, 1,
+                                (const GLfloat (*)[4]) maskColor, dst);
+      break;
+   case GL_SIGNED_NORMALIZED:
+   case GL_FLOAT:
+      /* These formats are harder because it's hard to know the floating
+       * point values that will convert to ~0 for each color channel's bits.
+       * This solution just generates a non-zero value for each color channel
+       * then fixes up the non-zero values to be ~0.
+       * Note: we'll need to add special case code if we ever have to deal
+       * with formats with unequal color channel sizes, like R11_G11_B10.
+       * We issue a warning below for channel sizes other than 8,16,32.
+       */
+      {
+         GLuint bits = _mesa_get_format_max_bits(format); /* bits per chan */
+         GLuint bytes = _mesa_get_format_bytes(format);
+         GLuint i;
+
+         /* this should put non-zero values into the channels of dst */
+         maskColor[0] = colorMask[0] ? -1.0 : 0.0;
+         maskColor[1] = colorMask[1] ? -1.0 : 0.0;
+         maskColor[2] = colorMask[2] ? -1.0 : 0.0;
+         maskColor[3] = colorMask[3] ? -1.0 : 0.0;
+         _mesa_pack_float_rgba_row(format, 1,
+                                   (const GLfloat (*)[4]) maskColor, dst);
+
+         /* fix-up the dst channels by converting non-zero values to ~0 */
+         if (bits == 8) {
+            GLubyte *d = (GLubyte *) dst;
+            for (i = 0; i < bytes; i++) {
+               d[i] = d[i] ? 0xffff : 0x0;
+            }
+         }
+         else if (bits == 16) {
+            GLushort *d = (GLushort *) dst;
+            for (i = 0; i < bytes / 2; i++) {
+               d[i] = d[i] ? 0xffff : 0x0;
+            }
+         }
+         else if (bits == 32) {
+            GLuint *d = (GLuint *) dst;
+            for (i = 0; i < bytes / 4; i++) {
+               d[i] = d[i] ? 0xffffffffU : 0x0;
+            }
+         }
+         else {
+            _mesa_problem(NULL, "unexpected size in _mesa_pack_colormask()");
+            return;
+         }
+      }
+      break;
+   default:
+      _mesa_problem(NULL, "unexpected format data type in gen_color_mask()");
+      return;
+   }
+}
diff --git a/mesalib/src/mesa/main/format_pack.h b/mesalib/src/mesa/main/format_pack.h
index 7df135632..f1b480510 100644
--- a/mesalib/src/mesa/main/format_pack.h
+++ b/mesalib/src/mesa/main/format_pack.h
@@ -95,4 +95,7 @@ _mesa_pack_uint_24_8_depth_stencil_row(gl_format format, GLuint n,
                                        const GLuint *src, void *dst);
 
 
+extern void
+_mesa_pack_colormask(gl_format format, const GLubyte colorMask[4], void *dst);
+
 #endif
diff --git a/mesalib/src/mesa/main/formats.c b/mesalib/src/mesa/main/formats.c
index 1f83a5368..cca0014b1 100644
--- a/mesalib/src/mesa/main/formats.c
+++ b/mesalib/src/mesa/main/formats.c
@@ -27,6 +27,7 @@
 #include "imports.h"
 #include "formats.h"
 #include "mfeatures.h"
+#include "macros.h"
 
 
 /**
@@ -1613,6 +1614,21 @@ _mesa_get_format_bits(gl_format format, GLenum pname)
 }
 
 
+GLuint
+_mesa_get_format_max_bits(gl_format format)
+{
+   const struct gl_format_info *info = _mesa_get_format_info(format);
+   GLuint max = MAX2(info->RedBits, info->GreenBits);
+   max = MAX2(max, info->BlueBits);
+   max = MAX2(max, info->AlphaBits);
+   max = MAX2(max, info->LuminanceBits);
+   max = MAX2(max, info->IntensityBits);
+   max = MAX2(max, info->DepthBits);
+   max = MAX2(max, info->StencilBits);
+   return max;
+}
+
+
 /**
  * Return the data type (or more specifically, the data representation)
  * for the given format.
diff --git a/mesalib/src/mesa/main/formats.h b/mesalib/src/mesa/main/formats.h
index e6b429d5f..960934338 100644
--- a/mesalib/src/mesa/main/formats.h
+++ b/mesalib/src/mesa/main/formats.h
@@ -290,6 +290,9 @@ _mesa_get_format_bytes(gl_format format);
 extern GLint
 _mesa_get_format_bits(gl_format format, GLenum pname);
 
+extern GLuint
+_mesa_get_format_max_bits(gl_format format);
+
 extern GLenum
 _mesa_get_format_datatype(gl_format format);
 
diff --git a/mesalib/src/mesa/main/mtypes.h b/mesalib/src/mesa/main/mtypes.h
index 3d3b006d5..02452aa68 100644
--- a/mesalib/src/mesa/main/mtypes.h
+++ b/mesalib/src/mesa/main/mtypes.h
@@ -2538,33 +2538,12 @@ struct gl_renderbuffer
    void (*PutRow)(struct gl_context *ctx, struct gl_renderbuffer *rb, GLuint count,
                   GLint x, GLint y, const void *values, const GLubyte *mask);
 
-   /* Put/Write a row of RGB values.  This is a special-case routine that's
-    * only used for RGBA renderbuffers when the source data is GL_RGB. That's
-    * a common case for glDrawPixels and some triangle routines.
-    * The values will be of format GL_RGB and type DataType.
-    */
-   void (*PutRowRGB)(struct gl_context *ctx, struct gl_renderbuffer *rb, GLuint count,
-                    GLint x, GLint y, const void *values, const GLubyte *mask);
-
-
-   /* Put/Write a row of identical values.
-    * The values will be of format _BaseFormat and type DataType.
-    */
-   void (*PutMonoRow)(struct gl_context *ctx, struct gl_renderbuffer *rb, GLuint count,
-                     GLint x, GLint y, const void *value, const GLubyte *mask);
-
    /* Put/Write values at arbitrary locations.
     * The values will be of format _BaseFormat and type DataType.
     */
    void (*PutValues)(struct gl_context *ctx, struct gl_renderbuffer *rb, GLuint count,
                      const GLint x[], const GLint y[], const void *values,
                      const GLubyte *mask);
-   /* Put/Write identical values at arbitrary locations.
-    * The values will be of format _BaseFormat and type DataType.
-    */
-   void (*PutMonoValues)(struct gl_context *ctx, struct gl_renderbuffer *rb,
-                         GLuint count, const GLint x[], const GLint y[],
-                         const void *value, const GLubyte *mask);
 };
 
 
diff --git a/mesalib/src/mesa/main/renderbuffer.c b/mesalib/src/mesa/main/renderbuffer.c
index d82b19ad5..bb8f46d11 100644
--- a/mesalib/src/mesa/main/renderbuffer.c
+++ b/mesalib/src/mesa/main/renderbuffer.c
@@ -78,10 +78,7 @@ _mesa_init_renderbuffer(struct gl_renderbuffer *rb, GLuint name)
    rb->GetRow = NULL;
    rb->GetValues = NULL;
    rb->PutRow = NULL;
-   rb->PutRowRGB = NULL;
-   rb->PutMonoRow = NULL;
    rb->PutValues = NULL;
-   rb->PutMonoValues = NULL;
 }
 
 
diff --git a/mesalib/src/mesa/main/shaderapi.c b/mesalib/src/mesa/main/shaderapi.c
index c4d01abc7..b71b44b70 100644
--- a/mesalib/src/mesa/main/shaderapi.c
+++ b/mesalib/src/mesa/main/shaderapi.c
@@ -1436,7 +1436,7 @@ _mesa_UseProgramObjectARB(GLhandleARB program)
 
    ASSERT_OUTSIDE_BEGIN_END(ctx);
 
-   if (obj->Active) {
+   if (obj->Active && !obj->Paused) {
       _mesa_error(ctx, GL_INVALID_OPERATION,
                   "glUseProgram(transform feedback active)");
       return;
@@ -1638,7 +1638,8 @@ _mesa_UseShaderProgramEXT(GLenum type, GLuint program)
       return;
    }
 
-   if (ctx->TransformFeedback.CurrentObject->Active) {
+   if (ctx->TransformFeedback.CurrentObject->Active &&
+       !ctx->TransformFeedback.CurrentObject->Paused) {
       _mesa_error(ctx, GL_INVALID_OPERATION,
                   "glUseShaderProgramEXT(transform feedback is active)");
       return;
diff --git a/mesalib/src/mesa/main/texgetimage.c b/mesalib/src/mesa/main/texgetimage.c
index 3f2418729..738c18112 100644
--- a/mesalib/src/mesa/main/texgetimage.c
+++ b/mesalib/src/mesa/main/texgetimage.c
@@ -216,183 +216,213 @@ get_tex_ycbcr(struct gl_context *ctx, GLuint dimensions,
 
 
 /**
- * glGetTexImage for color formats (RGBA, RGB, alpha, LA, etc).
- * Compressed textures are handled here as well.
+ * Get a color texture image with decompression.
  */
 static void
-get_tex_rgba(struct gl_context *ctx, GLuint dimensions,
-             GLenum format, GLenum type, GLvoid *pixels,
-             struct gl_texture_image *texImage)
+get_tex_rgba_compressed(struct gl_context *ctx, GLuint dimensions,
+                        GLenum format, GLenum type, GLvoid *pixels,
+                        struct gl_texture_image *texImage,
+                        GLbitfield transferOps)
 {
    /* don't want to apply sRGB -> RGB conversion here so override the format */
-   const gl_format texFormat = _mesa_get_srgb_format_linear(texImage->TexFormat);
+   const gl_format texFormat =
+      _mesa_get_srgb_format_linear(texImage->TexFormat);
+   const GLenum baseFormat = _mesa_get_format_base_format(texFormat);
    const GLuint width = texImage->Width;
    const GLuint height = texImage->Height;
    const GLuint depth = texImage->Depth;
-   const GLenum dataType = _mesa_get_format_datatype(texFormat);
-   const GLenum baseFormat = _mesa_get_format_base_format(texFormat);
-   /* Normally, no pixel transfer ops are performed during glGetTexImage.
-    * The only possible exception is component clamping to [0,1].
-    */
-   GLbitfield transferOps = 0x0;
-
-   /* In general, clamping does not apply to glGetTexImage, except when
-    * the returned type of the image can't hold negative values.
-    */
-   if (type_needs_clamping(type)) {
-      /* the returned image type can't have negative values */
-      if (dataType == GL_FLOAT ||
-          dataType == GL_SIGNED_NORMALIZED ||
-          format == GL_LUMINANCE ||
-          format == GL_LUMINANCE_ALPHA) {
-         transferOps |= IMAGE_CLAMP_BIT;
-      }
+   GLfloat *tempImage, *srcRow;
+   GLuint row;
+
+   /* Decompress into temp float buffer, then pack into user buffer */
+   tempImage = (GLfloat *) malloc(width * height * depth
+                                  * 4 * sizeof(GLfloat));
+   if (!tempImage) {
+      _mesa_error(ctx, GL_OUT_OF_MEMORY, "glGetTexImage()");
+      return;
    }
 
-   if (_mesa_is_format_compressed(texFormat)) {
-      /* Decompress into temp buffer, then pack into user buffer */
-      GLfloat *tempImage, *srcRow;
-      GLuint row;
+   /* Decompress the texture image - results in 'tempImage' */
+   {
+      GLubyte *srcMap;
+      GLint srcRowStride;
+      GLuint bytes, bw, bh;
 
-      tempImage = (GLfloat *) malloc(texImage->Width * texImage->Height *
-                                     texImage->Depth * 4 * sizeof(GLfloat));
-      if (!tempImage) {
-         _mesa_error(ctx, GL_OUT_OF_MEMORY, "glGetTexImage()");
-         return;
-      }
+      bytes = _mesa_get_format_bytes(texFormat);
+      _mesa_get_format_block_size(texFormat, &bw, &bh);
 
-      /* Decompress the texture image - results in 'tempImage' */
-      {
-         GLubyte *srcMap;
-         GLint srcRowStride;
-         GLuint bytes, bw, bh;
+      ctx->Driver.MapTextureImage(ctx, texImage, 0,
+                                  0, 0, width, height,
+                                  GL_MAP_READ_BIT,
+                                  &srcMap, &srcRowStride);
+      if (srcMap) {
+         /* XXX This line is a bit of a hack to work around the
+          * mismatch of compressed row strides as returned by
+          * MapTextureImage() vs. what the texture decompression code
+          * uses.  This will be fixed in the future.
+          */
+         srcRowStride = srcRowStride * bh / bytes;
 
-         bytes = _mesa_get_format_bytes(texImage->TexFormat);
-         _mesa_get_format_block_size(texImage->TexFormat, &bw, &bh);
+         _mesa_decompress_image(texFormat, width, height,
+                                srcMap, srcRowStride, tempImage);
 
-         ctx->Driver.MapTextureImage(ctx, texImage, 0,
-                                     0, 0, width, height,
-                                     GL_MAP_READ_BIT,
-                                     &srcMap, &srcRowStride);
+         ctx->Driver.UnmapTextureImage(ctx, texImage, 0);
+      }
+      else {
+         _mesa_error(ctx, GL_OUT_OF_MEMORY, "glGetTexImage");
+      }
+   }
 
-         if (srcMap) {
-            /* XXX This line is a bit of a hack to work around the
-             * mismatch of compressed row strides as returned by
-             * MapTextureImage() vs. what the texture decompression code
-             * uses.  This will be fixed in the future.
-             */
-            srcRowStride = srcRowStride * bh / bytes;
+   if (baseFormat == GL_LUMINANCE ||
+       baseFormat == GL_LUMINANCE_ALPHA) {
+      /* Set green and blue to zero since the pack function here will
+       * compute L=R+G+B.
+       */
+      GLuint i;
+      for (i = 0; i < width * height; i++) {
+         tempImage[i * 4 + GCOMP] = tempImage[i * 4 + BCOMP] = 0.0f;
+      }
+   }
 
-            _mesa_decompress_image(texFormat, width, height,
-                                   srcMap, srcRowStride, tempImage);
+   srcRow = tempImage;
+   for (row = 0; row < height; row++) {
+      void *dest = _mesa_image_address(dimensions, &ctx->Pack, pixels,
+                                       width, height, format, type,
+                                       0, row, 0);
 
-            ctx->Driver.UnmapTextureImage(ctx, texImage, 0);
-         }
-         else {
-            _mesa_error(ctx, GL_OUT_OF_MEMORY, "glGetTexImage");
-         }
-      }
+      _mesa_pack_rgba_span_float(ctx, width, (GLfloat (*)[4]) srcRow,
+                                 format, type, dest, &ctx->Pack, transferOps);
+      srcRow += width * 4;
+   }
 
-      if (baseFormat == GL_LUMINANCE ||
-          baseFormat == GL_LUMINANCE_ALPHA) {
-         /* Set green and blue to zero since the pack function here will
-          * compute L=R+G+B.
-          */
-         GLuint i;
-         for (i = 0; i < width * height; i++) {
-            tempImage[i * 4 + GCOMP] = tempImage[i * 4 + BCOMP] = 0.0f;
-         }
-      }
+   free(tempImage);
+}
 
-      srcRow = tempImage;
-      for (row = 0; row < height; row++) {
-         void *dest = _mesa_image_address(dimensions, &ctx->Pack, pixels,
-                                          width, height, format, type,
-                                          0, row, 0);
 
-         _mesa_pack_rgba_span_float(ctx, width, (GLfloat (*)[4]) srcRow,
-                                    format, type, dest, &ctx->Pack, transferOps);
-         srcRow += width * 4;
-      }
+/**
+ * Get an uncompressed color texture image.
+ */
+static void
+get_tex_rgba_uncompressed(struct gl_context *ctx, GLuint dimensions,
+                          GLenum format, GLenum type, GLvoid *pixels,
+                          struct gl_texture_image *texImage,
+                          GLbitfield transferOps)
+{
+   /* don't want to apply sRGB -> RGB conversion here so override the format */
+   const gl_format texFormat =
+      _mesa_get_srgb_format_linear(texImage->TexFormat);
+   const GLuint width = texImage->Width;
+   const GLuint height = texImage->Height;
+   const GLuint depth = texImage->Depth;
+   GLuint img, row;
+   GLfloat (*rgba)[4];
 
-      free(tempImage);
+   /* Allocate buffer for one row of texels */
+   rgba = (GLfloat (*)[4]) malloc(4 * width * sizeof(GLfloat));
+   if (!rgba) {
+      _mesa_error(ctx, GL_OUT_OF_MEMORY, "glGetTexImage()");
+      return;
    }
-   else {
-      /* No decompression needed */
-      GLuint img, row;
-      GLfloat (*rgba)[4];
 
-      rgba = (GLfloat (*)[4]) malloc(4 * width * sizeof(GLfloat));
-      if (!rgba) {
-         _mesa_error(ctx, GL_OUT_OF_MEMORY, "glGetTexImage()");
-         return;
-      }
+   for (img = 0; img < depth; img++) {
+      GLubyte *srcMap;
+      GLint rowstride;
 
-      for (img = 0; img < depth; img++) {
-	 GLubyte *srcMap;
-	 GLint rowstride;
-
-         /* map src texture buffer */
-         ctx->Driver.MapTextureImage(ctx, texImage, img,
-                                     0, 0, width, height, GL_MAP_READ_BIT,
-                                     &srcMap, &rowstride);
-
-         if (srcMap) {
-            for (row = 0; row < height; row++) {
-               const GLubyte *src = srcMap + row * rowstride;
-               void *dest = _mesa_image_address(dimensions, &ctx->Pack, pixels,
-                                                width, height, format, type,
-                                                img, row, 0);
-
-               _mesa_unpack_rgba_row(texFormat, width, src, rgba);
-
-               if (texImage->_BaseFormat == GL_ALPHA) {
-                  GLint col;
-                  for (col = 0; col < width; col++) {
-                     rgba[col][RCOMP] = 0.0F;
-                     rgba[col][GCOMP] = 0.0F;
-                     rgba[col][BCOMP] = 0.0F;
-                  }
+      /* map src texture buffer */
+      ctx->Driver.MapTextureImage(ctx, texImage, img,
+                                  0, 0, width, height, GL_MAP_READ_BIT,
+                                  &srcMap, &rowstride);
+      if (srcMap) {
+         for (row = 0; row < height; row++) {
+            const GLubyte *src = srcMap + row * rowstride;
+            void *dest = _mesa_image_address(dimensions, &ctx->Pack, pixels,
+                                             width, height, format, type,
+                                             img, row, 0);
+
+            _mesa_unpack_rgba_row(texFormat, width, src, rgba);
+
+            if (texImage->_BaseFormat == GL_ALPHA) {
+               GLint col;
+               for (col = 0; col < width; col++) {
+                  rgba[col][RCOMP] = 0.0F;
+                  rgba[col][GCOMP] = 0.0F;
+                  rgba[col][BCOMP] = 0.0F;
                }
-               else if (texImage->_BaseFormat == GL_LUMINANCE) {
-                  GLint col;
-                  for (col = 0; col < width; col++) {
-                     rgba[col][GCOMP] = 0.0F;
-                     rgba[col][BCOMP] = 0.0F;
-                     rgba[col][ACOMP] = 1.0F;
-                  }
+            }
+            else if (texImage->_BaseFormat == GL_LUMINANCE) {
+               GLint col;
+               for (col = 0; col < width; col++) {
+                  rgba[col][GCOMP] = 0.0F;
+                  rgba[col][BCOMP] = 0.0F;
+                  rgba[col][ACOMP] = 1.0F;
                }
-               else if (texImage->_BaseFormat == GL_LUMINANCE_ALPHA) {
-                  GLint col;
-                  for (col = 0; col < width; col++) {
-                     rgba[col][GCOMP] = 0.0F;
-                     rgba[col][BCOMP] = 0.0F;
-                  }
+            }
+            else if (texImage->_BaseFormat == GL_LUMINANCE_ALPHA) {
+               GLint col;
+               for (col = 0; col < width; col++) {
+                  rgba[col][GCOMP] = 0.0F;
+                  rgba[col][BCOMP] = 0.0F;
                }
-               else if (texImage->_BaseFormat == GL_INTENSITY) {
-                  GLint col;
-                  for (col = 0; col < width; col++) {
-                     rgba[col][GCOMP] = 0.0F;
-                     rgba[col][BCOMP] = 0.0F;
-                     rgba[col][ACOMP] = 1.0F;
-                  }
+            }
+            else if (texImage->_BaseFormat == GL_INTENSITY) {
+               GLint col;
+               for (col = 0; col < width; col++) {
+                  rgba[col][GCOMP] = 0.0F;
+                  rgba[col][BCOMP] = 0.0F;
+                  rgba[col][ACOMP] = 1.0F;
                }
-
-               _mesa_pack_rgba_span_float(ctx, width, (GLfloat (*)[4]) rgba,
-                                          format, type, dest,
-                                          &ctx->Pack, transferOps);
             }
 
-            /* Unmap the src texture buffer */
-            ctx->Driver.UnmapTextureImage(ctx, texImage, img);
-         }
-         else {
-            _mesa_error(ctx, GL_OUT_OF_MEMORY, "glGetTexImage");
-            break;
+            _mesa_pack_rgba_span_float(ctx, width, (GLfloat (*)[4]) rgba,
+                                       format, type, dest,
+                                       &ctx->Pack, transferOps);
          }
+
+         /* Unmap the src texture buffer */
+         ctx->Driver.UnmapTextureImage(ctx, texImage, img);
+      }
+      else {
+         _mesa_error(ctx, GL_OUT_OF_MEMORY, "glGetTexImage");
+         break;
       }
+   }
+
+   free(rgba);
+}
+
 
-      free(rgba);
+/**
+ * glGetTexImage for color formats (RGBA, RGB, alpha, LA, etc).
+ * Compressed textures are handled here as well.
+ */
+static void
+get_tex_rgba(struct gl_context *ctx, GLuint dimensions,
+             GLenum format, GLenum type, GLvoid *pixels,
+             struct gl_texture_image *texImage)
+{
+   const GLenum dataType = _mesa_get_format_datatype(texImage->TexFormat);
+   GLbitfield transferOps = 0x0;
+
+   /* In general, clamping does not apply to glGetTexImage, except when
+    * the returned type of the image can't hold negative values.
+    */
+   if (type_needs_clamping(type)) {
+      /* the returned image type can't have negative values */
+      if (dataType == GL_FLOAT ||
+          dataType == GL_SIGNED_NORMALIZED ||
+          format == GL_LUMINANCE ||
+          format == GL_LUMINANCE_ALPHA) {
+         transferOps |= IMAGE_CLAMP_BIT;
+      }
+   }
+
+   if (_mesa_is_format_compressed(texImage->TexFormat)) {
+      get_tex_rgba_compressed(ctx, dimensions, format, type,
+                              pixels, texImage, transferOps);
+   }
+   else {
+      get_tex_rgba_uncompressed(ctx, dimensions, format, type,
+                                pixels, texImage, transferOps);
    }
 }
 
@@ -702,20 +732,20 @@ getteximage_error_check(struct gl_context *ctx, GLenum target, GLint level,
       return GL_TRUE;
    }
 
-   texImage = _mesa_select_tex_image(ctx, texObj, target, level);
-   if (!texImage) {
-      /* out of memory */
-      return GL_TRUE;
-   }
-
    if (!_mesa_is_legal_format_and_type(ctx, format, type)) {
-      /*GL_INVALID_OPERATION is generated by a format/type
+      /* GL_INVALID_OPERATION is generated by a format/type
        * mismatch (see the 1.2 spec page 94, sec 3.6.4.)
        */
       _mesa_error(ctx, GL_INVALID_OPERATION, "glGetTexImage(target)");
       return GL_TRUE;
    }
 
+   texImage = _mesa_select_tex_image(ctx, texObj, target, level);
+   if (!texImage) {
+      /* non-existant texture image */
+      return GL_TRUE;
+   }
+
    baseFormat = _mesa_get_format_base_format(texImage->TexFormat);
       
    /* Make sure the requested image format is compatible with the
diff --git a/mesalib/src/mesa/main/texstore.c b/mesalib/src/mesa/main/texstore.c
index a57a38e12..86c35d38f 100644
--- a/mesalib/src/mesa/main/texstore.c
+++ b/mesalib/src/mesa/main/texstore.c
@@ -4565,9 +4565,137 @@ get_read_write_mode(GLenum userFormat, gl_format texFormat)
       return GL_MAP_WRITE_BIT;
 }
 
+
+/**
+ * Helper function for storing 1D, 2D, 3D whole and subimages into texture
+ * memory.
+ * The source of the image data may be user memory or a PBO.  In the later
+ * case, we'll map the PBO, copy from it, then unmap it.
+ */
+static void
+store_texsubimage(struct gl_context *ctx,
+                  struct gl_texture_image *texImage,
+                  GLint xoffset, GLint yoffset, GLint zoffset,
+                  GLint width, GLint height, GLint depth,
+                  GLenum format, GLenum type, const GLvoid *pixels,
+                  const struct gl_pixelstore_attrib *packing,
+                  const char *caller)
+
+{
+   const GLbitfield mapMode = get_read_write_mode(format, texImage->TexFormat);
+   const GLenum target = texImage->TexObject->Target;
+   GLboolean success = GL_FALSE;
+   GLuint dims, slice, numSlices = 1, sliceOffset = 0;
+   GLint srcImageStride = 0;
+   const GLubyte *src;
+
+   assert(xoffset + width <= texImage->Width);
+   assert(yoffset + height <= texImage->Height);
+   assert(zoffset + depth <= texImage->Depth);
+
+   switch (target) {
+   case GL_TEXTURE_1D:
+      dims = 1;
+      break;
+   case GL_TEXTURE_2D_ARRAY:
+   case GL_TEXTURE_3D:
+      dims = 3;
+      break;
+   default:
+      dims = 2;
+   }
+
+   /* get pointer to src pixels (may be in a pbo which we'll map here) */
+   src = (const GLubyte *)
+      _mesa_validate_pbo_teximage(ctx, dims, width, height, depth,
+                                  format, type, pixels, packing, caller);
+   if (!src)
+      return;
+
+   /* compute slice info (and do some sanity checks) */
+   switch (target) {
+   case GL_TEXTURE_2D:
+   case GL_TEXTURE_RECTANGLE:
+   case GL_TEXTURE_CUBE_MAP:
+      /* one image slice, nothing special needs to be done */
+      break;
+   case GL_TEXTURE_1D:
+      assert(height == 1);
+      assert(depth == 1);
+      assert(yoffset == 0);
+      assert(zoffset == 0);
+      break;
+   case GL_TEXTURE_1D_ARRAY:
+      assert(depth == 1);
+      assert(zoffset == 0);
+      numSlices = height;
+      sliceOffset = yoffset;
+      height = 1;
+      yoffset = 0;
+      srcImageStride = _mesa_image_row_stride(packing, width, format, type);
+      break;
+   case GL_TEXTURE_2D_ARRAY:
+      numSlices = depth;
+      sliceOffset = zoffset;
+      depth = 1;
+      zoffset = 0;
+      srcImageStride = _mesa_image_image_stride(packing, width, height,
+                                                format, type);
+      break;
+   case GL_TEXTURE_3D:
+      /* we'll store 3D images as a series of slices */
+      numSlices = depth;
+      sliceOffset = zoffset;
+      srcImageStride = _mesa_image_image_stride(packing, width, height,
+                                                format, type);
+      break;
+   default:
+      _mesa_warning(ctx, "Unexpected target 0x%x in store_texsubimage()", target);
+      return;
+   }
+
+   assert(numSlices == 1 || srcImageStride != 0);
+
+   for (slice = 0; slice < numSlices; slice++) {
+      GLubyte *dstMap;
+      GLint dstRowStride;
+
+      ctx->Driver.MapTextureImage(ctx, texImage,
+                                  slice + sliceOffset,
+                                  xoffset, yoffset, width, height,
+                                  mapMode, &dstMap, &dstRowStride);
+      if (dstMap) {
+         /* Note: we're only storing a 2D (or 1D) slice at a time but we need
+          * to pass the right 'dims' value so that GL_UNPACK_SKIP_IMAGES is
+          * used for 3D images.
+          */
+         success = _mesa_texstore(ctx, dims, texImage->_BaseFormat,
+                                  texImage->TexFormat,
+                                  0, 0, 0,  /* dstX/Y/Zoffset */
+                                  dstRowStride,
+                                  &dstMap,
+                                  width, height, 1,  /* w, h, d */
+                                  format, type, src, packing);
+
+         ctx->Driver.UnmapTextureImage(ctx, texImage, slice + sliceOffset);
+      }
+
+      src += srcImageStride;
+
+      if (!success)
+         break;
+   }
+
+   if (!success)
+      _mesa_error(ctx, GL_OUT_OF_MEMORY, "%s", caller);
+
+   _mesa_unmap_teximage_pbo(ctx, packing);
+}
+
+
+
 /**
- * This is the software fallback for Driver.TexImage1D().
- * \sa _mesa_store_teximage2d()
+ * This is the fallback for Driver.TexImage1D().
  */
 void
 _mesa_store_teximage1d(struct gl_context *ctx, GLenum target, GLint level,
@@ -4578,13 +4706,6 @@ _mesa_store_teximage1d(struct gl_context *ctx, GLenum target, GLint level,
                        struct gl_texture_object *texObj,
                        struct gl_texture_image *texImage)
 {
-   const GLbitfield rwMode = get_read_write_mode(format, texImage->TexFormat);
-   GLubyte *dstMap;
-   GLint dstRowStride;
-   GLboolean success;
-
-   (void) border;
-
    if (width == 0)
       return;
 
@@ -4595,47 +4716,14 @@ _mesa_store_teximage1d(struct gl_context *ctx, GLenum target, GLint level,
       return;
    }
 
-   pixels = _mesa_validate_pbo_teximage(ctx, 1, width, 1, 1, format, type,
-                                        pixels, packing, "glTexImage1D");
-   if (!pixels) {
-      /* Note: we check for a NULL image pointer here, _after_ we allocated
-       * memory for the texture.  That's what the GL spec calls for.
-       */
-      return;
-   }
-
-   /* Map dest texture buffer (write to whole region) */
-   ctx->Driver.MapTextureImage(ctx, texImage, 0,
-                               0, 0, width, 1,
-                               rwMode,
-                               &dstMap, &dstRowStride);
-   if (dstMap) {
-      success = _mesa_texstore(ctx, 1, texImage->_BaseFormat,
-                               texImage->TexFormat,
-                               0, 0, 0,  /* dstX/Y/Zoffset */
-                               0, /* dstRowStride */
-                               &dstMap,
-                               width, 1, 1,
-                               format, type, pixels, packing);
-
-      ctx->Driver.UnmapTextureImage(ctx, texImage, 0);
-   }
-   else {
-      success = GL_FALSE;
-   }
-
-   if (!success)
-      _mesa_error(ctx, GL_OUT_OF_MEMORY, "glTexImage1D");
-
-   _mesa_unmap_teximage_pbo(ctx, packing);
+   store_texsubimage(ctx, texImage,
+                     0, 0, 0, width, 1, 1,
+                     format, type, pixels, packing, "glTexImage1D");
 }
 
 
 /**
- * This is the software fallback for Driver.TexImage2D().
- *
- * This function is oriented toward storing images in main memory, rather
- * than VRAM.  Device driver's can easily plug in their own replacement.
+ * This is the fallback for Driver.TexImage2D().
  */
 void
 _mesa_store_teximage2d(struct gl_context *ctx, GLenum target, GLint level,
@@ -4646,13 +4734,6 @@ _mesa_store_teximage2d(struct gl_context *ctx, GLenum target, GLint level,
                        struct gl_texture_object *texObj,
                        struct gl_texture_image *texImage)
 {
-   const GLbitfield rwMode = get_read_write_mode(format, texImage->TexFormat);
-   GLubyte *dstMap;
-   GLint dstRowStride;
-   GLboolean success;
-
-   (void) border;
-
    if (width == 0 || height == 0)
       return;
 
@@ -4663,80 +4744,15 @@ _mesa_store_teximage2d(struct gl_context *ctx, GLenum target, GLint level,
       return;
    }
 
-   pixels = _mesa_validate_pbo_teximage(ctx, 2, width, height, 1, format, type,
-                                        pixels, packing, "glTexImage2D");
-   if (!pixels) {
-      /* Note: we check for a NULL image pointer here, _after_ we allocated
-       * memory for the texture.  That's what the GL spec calls for.
-       */
-      return;
-   }
-
-   if (target == GL_TEXTURE_1D_ARRAY) {
-      const GLint srcStride =
-         _mesa_image_row_stride(packing, width, format, type);
-      int y;
-
-      success = GL_TRUE;
-
-      for (y = 0; y < height; y++) {
-         /* Map dest texture buffer (write to whole region) */
-         ctx->Driver.MapTextureImage(ctx, texImage, y,
-                                     0, 0, width, 1,
-                                     rwMode,
-                                     &dstMap, &dstRowStride);
-         if (dstMap) {
-            success = _mesa_texstore(ctx, 2, texImage->_BaseFormat,
-                                     texImage->TexFormat,
-                                     0, 0, 0,  /* dstX/Y/Zoffset */
-                                     dstRowStride,
-                                     &dstMap,
-                                     width, 1, 1,
-                                     format, type, pixels, packing);
-            ctx->Driver.UnmapTextureImage(ctx, texImage, y);
-         }
-         else {
-            success = GL_FALSE;
-         }
-
-         if (!success)
-            break;
-
-         pixels = (const GLubyte *) pixels + srcStride;
-      }
-   } else {
-      /* Map dest texture buffer (write to whole region) */
-      ctx->Driver.MapTextureImage(ctx, texImage, 0,
-                                  0, 0, width, height,
-                                  rwMode,
-                                  &dstMap, &dstRowStride);
-      if (dstMap) {
-         success = _mesa_texstore(ctx, 2, texImage->_BaseFormat,
-                                  texImage->TexFormat,
-                                  0, 0, 0,  /* dstX/Y/Zoffset */
-                                  dstRowStride,
-                                  &dstMap,
-                                  width, height, 1,
-                                  format, type, pixels, packing);
-
-         ctx->Driver.UnmapTextureImage(ctx, texImage, 0);
-      }
-      else {
-         success = GL_FALSE;
-      }
-   }
-
-   if (!success)
-      _mesa_error(ctx, GL_OUT_OF_MEMORY, "glTexImage2D");
-
-   _mesa_unmap_teximage_pbo(ctx, packing);
+   store_texsubimage(ctx, texImage,
+                     0, 0, 0, width, height, 1,
+                     format, type, pixels, packing, "glTexImage2D");
 }
 
 
 
 /**
- * This is the software fallback for Driver.TexImage3D().
- * \sa _mesa_store_teximage2d()
+ * This is the fallback for Driver.TexImage3D().
  */
 void
 _mesa_store_teximage3d(struct gl_context *ctx, GLenum target, GLint level,
@@ -4747,14 +4763,6 @@ _mesa_store_teximage3d(struct gl_context *ctx, GLenum target, GLint level,
                        struct gl_texture_object *texObj,
                        struct gl_texture_image *texImage)
 {
-   const GLbitfield rwMode = get_read_write_mode(format, texImage->TexFormat);
-   GLboolean success = GL_TRUE;
-   GLint slice;
-   GLubyte **sliceMaps;
-   GLint dstRowStride;
-
-   (void) border;
-
    if (width == 0 || height == 0 || depth == 0)
       return;
 
@@ -4765,66 +4773,16 @@ _mesa_store_teximage3d(struct gl_context *ctx, GLenum target, GLint level,
       return;
    }
 
-   pixels = _mesa_validate_pbo_teximage(ctx, 3, width, height, depth,
-                                        format, type,
-                                        pixels, packing, "glTexImage3D");
-   if (!pixels) {
-      /* Note: we check for a NULL image pointer here, _after_ we allocated
-       * memory for the texture.  That's what the GL spec calls for.
-       */
-      return;
-   }
-
-   if (target == GL_TEXTURE_1D_ARRAY) {
-      depth = height;
-      height = 1;
-   }
-
-   sliceMaps = (GLubyte **) calloc(depth, sizeof(GLubyte *));
-
-   /* Map dest texture buffer slices */
-   for (slice = 0; slice < depth; slice++) {
-      ctx->Driver.MapTextureImage(ctx, texImage, slice,
-                                  0, 0, width, height,
-                                  rwMode,
-                                  &sliceMaps[slice], &dstRowStride);
-      if (!sliceMaps[slice]) {
-         success = GL_FALSE;
-         break;
-      }
-   }
-
-   if (success) {
-      success = _mesa_texstore(ctx, 3, texImage->_BaseFormat,
-                               texImage->TexFormat,
-                               0, 0, 0,  /* dstX/Y/Zoffset */
-                               dstRowStride,
-                               sliceMaps,
-                               width, height, depth,
-                               format, type, pixels, packing);
-   }
-
-   /* Unmap dest texture buffer slices */
-   for (slice = 0; slice < depth; slice++) {
-      if (sliceMaps[slice]) {
-         ctx->Driver.UnmapTextureImage(ctx, texImage, slice);
-      }
-   }
-
-   if (!success)
-      _mesa_error(ctx, GL_OUT_OF_MEMORY, "glTexImage3D");
-
-   _mesa_unmap_teximage_pbo(ctx, packing);
-
-   free(sliceMaps);
+   store_texsubimage(ctx, texImage,
+                     0, 0, 0, width, height, depth,
+                     format, type, pixels, packing, "glTexImage3D");
 }
 
 
 
 
 /*
- * This is the software fallback for Driver.TexSubImage1D()
- * and Driver.CopyTexSubImage1D().
+ * This is the fallback for Driver.TexSubImage1D().
  */
 void
 _mesa_store_texsubimage1d(struct gl_context *ctx, GLenum target, GLint level,
@@ -4834,49 +4792,15 @@ _mesa_store_texsubimage1d(struct gl_context *ctx, GLenum target, GLint level,
                           struct gl_texture_object *texObj,
                           struct gl_texture_image *texImage)
 {
-   const GLbitfield rwMode = get_read_write_mode(format, texImage->TexFormat);
-   GLubyte *dstMap;
-   GLint dstRowStride;
-   GLboolean success;
-
-   /* get pointer to src pixels (may be in a pbo which we'll map here) */
-   pixels = _mesa_validate_pbo_teximage(ctx, 1, width, 1, 1, format, type,
-                                        pixels, packing, "glTexSubImage1D");
-   if (!pixels)
-      return;
-
-   /* Map dest texture buffer */
-   ctx->Driver.MapTextureImage(ctx, texImage, 0,
-                               xoffset, 0, width, 1,
-                               rwMode,
-                               &dstMap, &dstRowStride);
-
-   if (dstMap) {
-      success = _mesa_texstore(ctx, 1, texImage->_BaseFormat,
-                               texImage->TexFormat,
-                               0, 0, 0,  /* dstX/Y/Zoffset */
-                               dstRowStride,
-                               &dstMap,
-                               width, 1, 1,
-                               format, type, pixels, packing);
-
-      ctx->Driver.UnmapTextureImage(ctx, texImage, 0);
-   }
-   else {
-      success = GL_FALSE;
-   }
-
-   if (!success)
-      _mesa_error(ctx, GL_OUT_OF_MEMORY, "glTexSubImage1D");
-
-   _mesa_unmap_teximage_pbo(ctx, packing);
+   store_texsubimage(ctx, texImage,
+                     xoffset, 0, 0, width, 1, 1,
+                     format, type, pixels, packing, "glTexSubImage1D");
 }
 
 
 
 /**
- * This is the software fallback for Driver.TexSubImage2D()
- * and Driver.CopyTexSubImage2D().
+ * This is the fallback for Driver.TexSubImage2D().
  */
 void
 _mesa_store_texsubimage2d(struct gl_context *ctx, GLenum target, GLint level,
@@ -4887,48 +4811,14 @@ _mesa_store_texsubimage2d(struct gl_context *ctx, GLenum target, GLint level,
                           struct gl_texture_object *texObj,
                           struct gl_texture_image *texImage)
 {
-   const GLbitfield rwMode = get_read_write_mode(format, texImage->TexFormat);
-   GLubyte *dstMap;
-   GLint dstRowStride;
-   GLboolean success;
-
-   /* get pointer to src pixels (may be in a pbo which we'll map here) */
-   pixels = _mesa_validate_pbo_teximage(ctx, 2, width, height, 1, format, type,
-                                        pixels, packing, "glTexSubImage2D");
-   if (!pixels)
-      return;
-
-   /* Map dest texture buffer */
-   ctx->Driver.MapTextureImage(ctx, texImage, 0,
-                               xoffset, yoffset, width, height,
-                               rwMode,
-                               &dstMap, &dstRowStride);
-
-   if (dstMap) {
-      success = _mesa_texstore(ctx, 2, texImage->_BaseFormat,
-                               texImage->TexFormat,
-                               0, 0, 0,  /* dstX/Y/Zoffset */
-                               dstRowStride,
-                               &dstMap,
-                               width, height, 1,
-                               format, type, pixels, packing);
-
-      ctx->Driver.UnmapTextureImage(ctx, texImage, 0);
-   }
-   else {
-      success = GL_FALSE;
-   }
-
-   if (!success)
-      _mesa_error(ctx, GL_OUT_OF_MEMORY, "glTexSubImage2D");
-
-   _mesa_unmap_teximage_pbo(ctx, packing);
+   store_texsubimage(ctx, texImage,
+                     xoffset, yoffset, 0, width, height, 1,
+                     format, type, pixels, packing, "glTexSubImage2D");
 }
 
 
 /*
- * This is the software fallback for Driver.TexSubImage3D().
- * and Driver.CopyTexSubImage3D().
+ * This is the fallback for Driver.TexSubImage3D().
  */
 void
 _mesa_store_texsubimage3d(struct gl_context *ctx, GLenum target, GLint level,
@@ -4939,56 +4829,9 @@ _mesa_store_texsubimage3d(struct gl_context *ctx, GLenum target, GLint level,
                           struct gl_texture_object *texObj,
                           struct gl_texture_image *texImage)
 {
-   const GLbitfield rwMode = get_read_write_mode(format, texImage->TexFormat);
-   GLboolean success = GL_TRUE;
-   GLint slice;
-   GLubyte **sliceMaps;
-   GLint dstRowStride;
-
-   /* get pointer to src pixels (may be in a pbo which we'll map here) */
-   pixels = _mesa_validate_pbo_teximage(ctx, 3, width, height, depth, format,
-                                        type, pixels, packing,
-                                        "glTexSubImage3D");
-   if (!pixels)
-      return;
-
-   sliceMaps = (GLubyte **) calloc(depth, sizeof(GLubyte *));
-
-   /* Map dest texture buffer slices */
-   for (slice = 0; slice < depth; slice++) {
-      ctx->Driver.MapTextureImage(ctx, texImage, zoffset + slice,
-                                  xoffset, yoffset, width, height,
-                                  rwMode,
-                                  &sliceMaps[slice], &dstRowStride);
-      if (!sliceMaps[slice]) {
-         success = GL_FALSE;
-         break;
-      }
-   }
-
-   if (success) {
-      success = _mesa_texstore(ctx, 3, texImage->_BaseFormat,
-                               texImage->TexFormat,
-                               0, 0, 0,
-                               dstRowStride,
-                               sliceMaps,
-                               width, height, depth,
-                               format, type, pixels, packing);
-   }
-
-   /* Unmap dest texture buffer slices */
-   for (slice = 0; slice < depth; slice++) {
-      if (sliceMaps[slice]) {
-         ctx->Driver.UnmapTextureImage(ctx, texImage, zoffset + slice);
-      }
-   }
-
-   if (!success)
-      _mesa_error(ctx, GL_OUT_OF_MEMORY, "glTexSubImage3D");
-
-   _mesa_unmap_teximage_pbo(ctx, packing);
-
-   free(sliceMaps);
+   store_texsubimage(ctx, texImage,
+                     xoffset, yoffset, zoffset, width, height, depth,
+                     format, type, pixels, packing, "glTexSubImage3D");
 }
 
 
diff --git a/mesalib/src/mesa/main/transformfeedback.c b/mesalib/src/mesa/main/transformfeedback.c
index 53c09e265..be0d0ff19 100644
--- a/mesalib/src/mesa/main/transformfeedback.c
+++ b/mesalib/src/mesa/main/transformfeedback.c
@@ -98,7 +98,8 @@ reference_transform_feedback_object(struct gl_transform_feedback_object **ptr,
 GLboolean
 _mesa_validate_primitive_mode(struct gl_context *ctx, GLenum mode)
 {
-   if (ctx->TransformFeedback.CurrentObject->Active) {
+   if (ctx->TransformFeedback.CurrentObject->Active &&
+       !ctx->TransformFeedback.CurrentObject->Paused) {
       switch (mode) {
       case GL_POINTS:
          return ctx->TransformFeedback.Mode == GL_POINTS;
@@ -387,6 +388,7 @@ _mesa_EndTransformFeedback(void)
 
    FLUSH_VERTICES(ctx, _NEW_TRANSFORM_FEEDBACK);
    ctx->TransformFeedback.CurrentObject->Active = GL_FALSE;
+   ctx->TransformFeedback.CurrentObject->Paused = GL_FALSE;
    ctx->TransformFeedback.CurrentObject->EndedAnytime = GL_TRUE;
 
    assert(ctx->Driver.EndTransformFeedback);
diff --git a/mesalib/src/mesa/state_tracker/st_cb_xformfb.c b/mesalib/src/mesa/state_tracker/st_cb_xformfb.c
index 2fc28dc24..b8534855b 100644
--- a/mesalib/src/mesa/state_tracker/st_cb_xformfb.c
+++ b/mesalib/src/mesa/state_tracker/st_cb_xformfb.c
@@ -55,8 +55,18 @@ struct st_transform_feedback_object {
 
    unsigned num_targets;
    struct pipe_stream_output_target *targets[PIPE_MAX_SO_BUFFERS];
+
+   /* This encapsulates the count that can be used as a source for draw_vbo.
+    * It contains a stream output target from the last call of
+    * EndTransformFeedback. */
+   struct pipe_stream_output_target *draw_count;
 };
 
+static INLINE struct st_transform_feedback_object *
+st_transform_feedback_object(struct gl_transform_feedback_object *obj)
+{
+   return (struct st_transform_feedback_object *) obj;
+}
 
 static struct gl_transform_feedback_object *
 st_new_transform_feedback(struct gl_context *ctx, GLuint name)
@@ -78,9 +88,11 @@ st_delete_transform_feedback(struct gl_context *ctx,
                              struct gl_transform_feedback_object *obj)
 {
    struct st_transform_feedback_object *sobj =
-         (struct st_transform_feedback_object*)obj;
+         st_transform_feedback_object(obj);
    unsigned i;
 
+   pipe_so_target_reference(&sobj->draw_count, NULL);
+
    /* Unreference targets. */
    for (i = 0; i < sobj->num_targets; i++) {
       pipe_so_target_reference(&sobj->targets[i], NULL);
@@ -102,7 +114,7 @@ st_begin_transform_feedback(struct gl_context *ctx, GLenum mode,
    struct st_context *st = st_context(ctx);
    struct pipe_context *pipe = st->pipe;
    struct st_transform_feedback_object *sobj =
-         (struct st_transform_feedback_object*)obj;
+         st_transform_feedback_object(obj);
    unsigned i, max_num_targets;
 
    max_num_targets = MIN2(Elements(sobj->base.Buffers),
@@ -115,6 +127,7 @@ st_begin_transform_feedback(struct gl_context *ctx, GLenum mode,
       if (bo) {
          /* Check whether we need to recreate the target. */
          if (!sobj->targets[i] ||
+             sobj->targets[i] == sobj->draw_count ||
              sobj->targets[i]->buffer != bo->buffer ||
              sobj->targets[i]->buffer_offset != sobj->base.Offset[i] ||
              sobj->targets[i]->buffer_size != sobj->base.Size[i]) {
@@ -141,7 +154,7 @@ st_begin_transform_feedback(struct gl_context *ctx, GLenum mode,
 
 
 static void
-st_stop_transform_feedback(struct gl_context *ctx,
+st_pause_transform_feedback(struct gl_context *ctx,
                            struct gl_transform_feedback_object *obj)
 {
    struct st_context *st = st_context(ctx);
@@ -155,31 +168,54 @@ st_resume_transform_feedback(struct gl_context *ctx,
 {
    struct st_context *st = st_context(ctx);
    struct st_transform_feedback_object *sobj =
-         (struct st_transform_feedback_object*)obj;
+         st_transform_feedback_object(obj);
 
    cso_set_stream_outputs(st->cso_context, sobj->num_targets, sobj->targets,
                           ~0);
 }
 
-/* Set count_from_stream_output to any stream output target
- * from the transform feedback object. */
-void
-st_transform_feedback_draw_init(struct gl_transform_feedback_object *obj,
-                                struct pipe_draw_info *out)
+
+static struct pipe_stream_output_target *
+st_transform_feedback_get_draw_target(struct gl_transform_feedback_object *obj)
 {
    struct st_transform_feedback_object *sobj =
-         (struct st_transform_feedback_object*)obj;
+         st_transform_feedback_object(obj);
    unsigned i;
 
    for (i = 0; i < Elements(sobj->targets); i++) {
       if (sobj->targets[i]) {
-         out->count_from_stream_output = sobj->targets[i];
-         return;
+         return sobj->targets[i];
       }
    }
 
    assert(0);
-   out->count_from_stream_output = NULL;
+   return NULL;
+}
+
+
+static void
+st_end_transform_feedback(struct gl_context *ctx,
+                          struct gl_transform_feedback_object *obj)
+{
+   struct st_context *st = st_context(ctx);
+   struct st_transform_feedback_object *sobj =
+         st_transform_feedback_object(obj);
+
+   cso_set_stream_outputs(st->cso_context, 0, NULL, 0);
+
+   pipe_so_target_reference(&sobj->draw_count,
+                            st_transform_feedback_get_draw_target(obj));
+}
+
+
+void
+st_transform_feedback_draw_init(struct gl_transform_feedback_object *obj,
+                                struct pipe_draw_info *out)
+{
+   struct st_transform_feedback_object *sobj =
+         st_transform_feedback_object(obj);
+
+   out->count_from_stream_output = sobj->draw_count;
 }
 
 
@@ -189,8 +225,8 @@ st_init_xformfb_functions(struct dd_function_table *functions)
    functions->NewTransformFeedback = st_new_transform_feedback;
    functions->DeleteTransformFeedback = st_delete_transform_feedback;
    functions->BeginTransformFeedback = st_begin_transform_feedback;
-   functions->EndTransformFeedback = st_stop_transform_feedback;
-   functions->PauseTransformFeedback = st_stop_transform_feedback;
+   functions->EndTransformFeedback = st_end_transform_feedback;
+   functions->PauseTransformFeedback = st_pause_transform_feedback;
    functions->ResumeTransformFeedback = st_resume_transform_feedback;
 }
 
diff --git a/mesalib/src/mesa/state_tracker/st_extensions.c b/mesalib/src/mesa/state_tracker/st_extensions.c
index dc17d76a5..4c8c67fca 100644
--- a/mesalib/src/mesa/state_tracker/st_extensions.c
+++ b/mesalib/src/mesa/state_tracker/st_extensions.c
@@ -222,7 +222,7 @@ void st_init_limits(struct st_context *st)
    c->UniformBooleanTrue = ~0;
 
    c->MaxTransformFeedbackSeparateAttribs =
-      screen->get_param(screen, PIPE_CAP_MAX_STREAM_OUTPUT_SEPARATE_ATTRIBS);
+      screen->get_param(screen, PIPE_CAP_MAX_STREAM_OUTPUT_BUFFERS);
    c->MaxTransformFeedbackSeparateComponents =
       screen->get_param(screen, PIPE_CAP_MAX_STREAM_OUTPUT_SEPARATE_COMPONENTS);
    c->MaxTransformFeedbackInterleavedComponents =
diff --git a/mesalib/src/mesa/state_tracker/st_manager.c b/mesalib/src/mesa/state_tracker/st_manager.c
index 55699e721..828f0d81f 100644
--- a/mesalib/src/mesa/state_tracker/st_manager.c
+++ b/mesalib/src/mesa/state_tracker/st_manager.c
@@ -600,6 +600,7 @@ st_context_destroy(struct st_context_iface *stctxi)
 static struct st_context_iface *
 st_api_create_context(struct st_api *stapi, struct st_manager *smapi,
                       const struct st_context_attribs *attribs,
+                      enum st_context_error *error,
                       struct st_context_iface *shared_stctxi)
 {
    struct st_context *shared_ctx = (struct st_context *) shared_stctxi;
@@ -623,17 +624,21 @@ st_api_create_context(struct st_api *stapi, struct st_manager *smapi,
       break;
    case ST_PROFILE_OPENGL_CORE:
    default:
+      *error = ST_CONTEXT_ERROR_BAD_API;
       return NULL;
       break;
    }
 
    pipe = smapi->screen->context_create(smapi->screen, NULL);
-   if (!pipe)
+   if (!pipe) {
+      *error = ST_CONTEXT_ERROR_NO_MEMORY;
       return NULL;
+   }
 
    st_visual_to_context_mode(&attribs->visual, &mode);
    st = st_create_context(api, pipe, &mode, shared_ctx);
    if (!st) {
+      *error = ST_CONTEXT_ERROR_NO_MEMORY;
       pipe->destroy(pipe);
       return NULL;
    }
@@ -645,6 +650,7 @@ st_api_create_context(struct st_api *stapi, struct st_manager *smapi,
       /* is the actual version less than the requested version? */
       if (st->ctx->VersionMajor * 10 + st->ctx->VersionMinor <
           attribs->major * 10 + attribs->minor) {
+	 *error = ST_CONTEXT_ERROR_BAD_VERSION;
          st_destroy_context(st);
          return NULL;
       }
@@ -660,6 +666,7 @@ st_api_create_context(struct st_api *stapi, struct st_manager *smapi,
    st->iface.share = st_context_share;
    st->iface.st_context_private = (void *) smapi;
 
+   *error = ST_CONTEXT_SUCCESS;
    return &st->iface;
 }
 
diff --git a/mesalib/src/mesa/swrast/s_clear.c b/mesalib/src/mesa/swrast/s_clear.c
index 851f6d165..d942e6e63 100644
--- a/mesalib/src/mesa/swrast/s_clear.c
+++ b/mesalib/src/mesa/swrast/s_clear.c
@@ -24,131 +24,146 @@
 
 #include "main/glheader.h"
 #include "main/accum.h"
-#include "main/colormac.h"
 #include "main/condrender.h"
+#include "main/format_pack.h"
 #include "main/macros.h"
 #include "main/imports.h"
 #include "main/mtypes.h"
 
 #include "s_context.h"
 #include "s_depth.h"
-#include "s_masking.h"
 #include "s_stencil.h"
 
 
+
 /**
- * Clear the color buffer when glColorMask is in effect.
+ * Clear an rgba color buffer with masking if needed.
  */
 static void
-clear_rgba_buffer_with_masking(struct gl_context *ctx, struct gl_renderbuffer *rb,
-                               GLuint buf)
+clear_rgba_buffer(struct gl_context *ctx, struct gl_renderbuffer *rb,
+                  const GLubyte colorMask[4])
 {
    const GLint x = ctx->DrawBuffer->_Xmin;
    const GLint y = ctx->DrawBuffer->_Ymin;
    const GLint height = ctx->DrawBuffer->_Ymax - ctx->DrawBuffer->_Ymin;
    const GLint width  = ctx->DrawBuffer->_Xmax - ctx->DrawBuffer->_Xmin;
-   SWspan span;
-   GLint i;
-
-   ASSERT(rb->PutRow);
-
-   /* Initialize color span with clear color */
-   /* XXX optimize for clearcolor == black/zero (bzero) */
-   INIT_SPAN(span, GL_BITMAP);
-   span.end = width;
-   span.arrayMask = SPAN_RGBA;
-   span.array->ChanType = rb->DataType;
-   if (span.array->ChanType == GL_UNSIGNED_BYTE) {
-      GLubyte clearColor[4];
-      _mesa_unclamped_float_rgba_to_ubyte(clearColor, ctx->Color.ClearColor.f);
-      for (i = 0; i < width; i++) {
-         COPY_4UBV(span.array->rgba[i], clearColor);
-      }
-   }
-   else if (span.array->ChanType == GL_UNSIGNED_SHORT) {
-      GLushort clearColor[4];
-      UNCLAMPED_FLOAT_TO_USHORT(clearColor[RCOMP], ctx->Color.ClearColor.f[0]);
-      UNCLAMPED_FLOAT_TO_USHORT(clearColor[GCOMP], ctx->Color.ClearColor.f[1]);
-      UNCLAMPED_FLOAT_TO_USHORT(clearColor[BCOMP], ctx->Color.ClearColor.f[2]);
-      UNCLAMPED_FLOAT_TO_USHORT(clearColor[ACOMP], ctx->Color.ClearColor.f[3]);
-      for (i = 0; i < width; i++) {
-         COPY_4V_CAST(span.array->rgba[i], clearColor, GLchan);
-      }
-   }
-   else {
-      ASSERT(span.array->ChanType == GL_FLOAT);
-      for (i = 0; i < width; i++) {
-         UNCLAMPED_FLOAT_TO_CHAN(span.array->rgba[i][0], ctx->Color.ClearColor.f[0]);
-         UNCLAMPED_FLOAT_TO_CHAN(span.array->rgba[i][1], ctx->Color.ClearColor.f[1]);
-         UNCLAMPED_FLOAT_TO_CHAN(span.array->rgba[i][2], ctx->Color.ClearColor.f[2]);
-         UNCLAMPED_FLOAT_TO_CHAN(span.array->rgba[i][3], ctx->Color.ClearColor.f[3]);
-      }
+   const GLuint pixelSize = _mesa_get_format_bytes(rb->Format);
+   const GLboolean doMasking = (colorMask[0] == 0 ||
+                                colorMask[1] == 0 ||
+                                colorMask[2] == 0 ||
+                                colorMask[3] == 0);
+   const GLfloat (*clearColor)[4] =
+      (const GLfloat (*)[4]) ctx->Color.ClearColor.f;
+   GLbitfield mapMode = GL_MAP_WRITE_BIT;
+   GLubyte *map;
+   GLint rowStride;
+   GLint i, j;
+
+   if (doMasking) {
+      /* we'll need to read buffer values too */
+      mapMode |= GL_MAP_READ_BIT;
    }
 
-   /* Note that masking will change the color values, but only the
-    * channels for which the write mask is GL_FALSE.  The channels
-    * which which are write-enabled won't get modified.
-    */
-   for (i = 0; i < height; i++) {
-      span.x = x;
-      span.y = y + i;
-      _swrast_mask_rgba_span(ctx, rb, &span, buf);
-      /* write masked row */
-      rb->PutRow(ctx, rb, width, x, y + i, span.array->rgba, NULL);
+   /* map dest buffer */
+   ctx->Driver.MapRenderbuffer(ctx, rb, x, y, width, height,
+                               mapMode, &map, &rowStride);
+   if (!map) {
+      _mesa_error(ctx, GL_OUT_OF_MEMORY, "glClear(color)");
+      return;
    }
-}
 
+   /* for 1, 2, 4-byte clearing */
+#define SIMPLE_TYPE_CLEAR(TYPE)                                         \
+   do {                                                                 \
+      TYPE pixel, pixelMask;                                            \
+      _mesa_pack_float_rgba_row(rb->Format, 1, clearColor, &pixel);     \
+      if (doMasking) {                                                  \
+         _mesa_pack_colormask(rb->Format, colorMask, &pixelMask);       \
+         pixel &= pixelMask;                                            \
+         pixelMask = ~pixelMask;                                        \
+      }                                                                 \
+      for (i = 0; i < height; i++) {                                    \
+         TYPE *row = (TYPE *) map;                                      \
+         if (doMasking) {                                               \
+            for (j = 0; j < width; j++) {                               \
+               row[j] = (row[j] & pixelMask) | pixel;                   \
+            }                                                           \
+         }                                                              \
+         else {                                                         \
+            for (j = 0; j < width; j++) {                               \
+               row[j] = pixel;                                          \
+            }                                                           \
+         }                                                              \
+         map += rowStride;                                              \
+      }                                                                 \
+   } while (0)
 
-/**
- * Clear an rgba color buffer without channel masking.
- */
-static void
-clear_rgba_buffer(struct gl_context *ctx, struct gl_renderbuffer *rb, GLuint buf)
-{
-   const GLint x = ctx->DrawBuffer->_Xmin;
-   const GLint y = ctx->DrawBuffer->_Ymin;
-   const GLint height = ctx->DrawBuffer->_Ymax - ctx->DrawBuffer->_Ymin;
-   const GLint width  = ctx->DrawBuffer->_Xmax - ctx->DrawBuffer->_Xmin;
-   GLubyte clear8[4];
-   GLushort clear16[4];
-   GLvoid *clearVal;
-   GLfloat clearFloat[4];
-   GLint i;
-
-   ASSERT(ctx->Color.ColorMask[buf][0] &&
-          ctx->Color.ColorMask[buf][1] &&
-          ctx->Color.ColorMask[buf][2] &&
-          ctx->Color.ColorMask[buf][3]);             
-
-   ASSERT(rb->PutMonoRow);
-
-   switch (rb->DataType) {
-      case GL_UNSIGNED_BYTE:
-         _mesa_unclamped_float_rgba_to_ubyte(clear8, ctx->Color.ClearColor.f);
-         clearVal = clear8;
-         break;
-      case GL_UNSIGNED_SHORT:
-         UNCLAMPED_FLOAT_TO_USHORT(clear16[0], ctx->Color.ClearColor.f[0]);
-         UNCLAMPED_FLOAT_TO_USHORT(clear16[1], ctx->Color.ClearColor.f[1]);
-         UNCLAMPED_FLOAT_TO_USHORT(clear16[2], ctx->Color.ClearColor.f[2]);
-         UNCLAMPED_FLOAT_TO_USHORT(clear16[3], ctx->Color.ClearColor.f[3]);
-         clearVal = clear16;
-         break;
-      case GL_FLOAT:
-         clearFloat[0] = CLAMP(ctx->Color.ClearColor.f[0], 0.0F, 1.0F);
-         clearFloat[1] = CLAMP(ctx->Color.ClearColor.f[1], 0.0F, 1.0F);
-         clearFloat[2] = CLAMP(ctx->Color.ClearColor.f[2], 0.0F, 1.0F);
-         clearFloat[3] = CLAMP(ctx->Color.ClearColor.f[3], 0.0F, 1.0F);
-         clearVal = clearFloat;
-         break;
-      default:
-         _mesa_problem(ctx, "Bad rb DataType in clear_color_buffer");
-         return;
-   }
 
-   for (i = 0; i < height; i++) {
-      rb->PutMonoRow(ctx, rb, width, x, y + i, clearVal, NULL);
+   /* for 3, 6, 8, 12, 16-byte clearing */
+#define MULTI_WORD_CLEAR(TYPE, N)                                       \
+   do {                                                                 \
+      TYPE pixel[N], pixelMask[N];                                      \
+      GLuint k;                                                         \
+      _mesa_pack_float_rgba_row(rb->Format, 1, clearColor, pixel);      \
+      if (doMasking) {                                                  \
+         _mesa_pack_colormask(rb->Format, colorMask, pixelMask);        \
+         for (k = 0; k < N; k++) {                                      \
+            pixel[k] &= pixelMask[k];                                   \
+            pixelMask[k] = ~pixelMask[k];                               \
+         }                                                              \
+      }                                                                 \
+      for (i = 0; i < height; i++) {                                    \
+         TYPE *row = (TYPE *) map;                                      \
+         if (doMasking) {                                               \
+            for (j = 0; j < width; j++) {                               \
+               for (k = 0; k < N; k++) {                                \
+                  row[j * N + k] =                                      \
+                     (row[j * N + k] & pixelMask[k]) | pixel[k];        \
+               }                                                        \
+            }                                                           \
+         }                                                              \
+         else {                                                         \
+            for (j = 0; j < width; j++) {                               \
+               for (k = 0; k < N; k++) {                                \
+                  row[j * N + k] = pixel[k];                            \
+               }                                                        \
+            }                                                           \
+         }                                                              \
+         map += rowStride;                                              \
+      }                                                                 \
+   } while(0)
+
+   switch (pixelSize) {
+   case 1:
+      SIMPLE_TYPE_CLEAR(GLubyte);
+      break;
+   case 2:
+      SIMPLE_TYPE_CLEAR(GLushort);
+      break;
+   case 3:
+      MULTI_WORD_CLEAR(GLubyte, 3);
+      break;
+   case 4:
+      SIMPLE_TYPE_CLEAR(GLuint);
+      break;
+   case 6:
+      MULTI_WORD_CLEAR(GLushort, 3);
+      break;
+   case 8:
+      MULTI_WORD_CLEAR(GLuint, 2);
+      break;
+   case 12:
+      MULTI_WORD_CLEAR(GLuint, 3);
+      break;
+   case 16:
+      MULTI_WORD_CLEAR(GLuint, 4);
+      break;
+   default:
+      _mesa_problem(ctx, "bad pixel size in clear_rgba_buffer()");
    }
+
+   /* unmap buffer */
+   ctx->Driver.UnmapRenderbuffer(ctx, rb);
 }
 
 
@@ -172,15 +187,7 @@ clear_color_buffers(struct gl_context *ctx)
       if (rb == NULL)
 	 continue;
 
-      if (ctx->Color.ColorMask[buf][0] == 0 ||
-          ctx->Color.ColorMask[buf][1] == 0 ||
-          ctx->Color.ColorMask[buf][2] == 0 ||
-          ctx->Color.ColorMask[buf][3] == 0) {
-         clear_rgba_buffer_with_masking(ctx, rb, buf);
-      }
-      else {
-         clear_rgba_buffer(ctx, rb, buf);
-      }
+      clear_rgba_buffer(ctx, rb, ctx->Color.ColorMask[buf]);
    }
 }
 
@@ -195,6 +202,8 @@ clear_color_buffers(struct gl_context *ctx)
 void
 _swrast_Clear(struct gl_context *ctx, GLbitfield buffers)
 {
+   const GLbitfield BUFFER_DS = BUFFER_BIT_DEPTH | BUFFER_BIT_STENCIL;
+
 #ifdef DEBUG_FOO
    {
       const GLbitfield legalBits =
@@ -216,24 +225,33 @@ _swrast_Clear(struct gl_context *ctx, GLbitfield buffers)
    if (SWRAST_CONTEXT(ctx)->NewState)
       _swrast_validate_derived(ctx);
 
-   swrast_render_start(ctx);
+   if ((buffers & BUFFER_BITS_COLOR)
+       && (ctx->DrawBuffer->_NumColorDrawBuffers > 0)) {
+      clear_color_buffers(ctx);
+   }
 
-   /* do software clearing here */
-   if (buffers) {
-      if ((buffers & BUFFER_BITS_COLOR)
-          && (ctx->DrawBuffer->_NumColorDrawBuffers > 0)) {
-         clear_color_buffers(ctx);
-      }
-      if (buffers & BUFFER_BIT_DEPTH) {
-         _swrast_clear_depth_buffer(ctx, ctx->DrawBuffer->_DepthBuffer);
-      }
-      if (buffers & BUFFER_BIT_ACCUM) {
-         _mesa_clear_accum_buffer(ctx);
+   if (buffers & BUFFER_BIT_ACCUM) {
+      _mesa_clear_accum_buffer(ctx);
+   }
+
+   if (buffers & BUFFER_DS) {
+      struct gl_renderbuffer *depthRb =
+         ctx->DrawBuffer->Attachment[BUFFER_DEPTH].Renderbuffer;
+      struct gl_renderbuffer *stencilRb =
+         ctx->DrawBuffer->Attachment[BUFFER_STENCIL].Renderbuffer;
+
+      if ((buffers & BUFFER_DS) == BUFFER_DS && depthRb == stencilRb) {
+         /* clear depth and stencil together */
+         _swrast_clear_depth_stencil_buffer(ctx);
       }
-      if (buffers & BUFFER_BIT_STENCIL) {
-         _swrast_clear_stencil_buffer(ctx, ctx->DrawBuffer->_StencilBuffer);
+      else {
+         /* clear depth, stencil separately */
+         if (buffers & BUFFER_BIT_DEPTH) {
+            _swrast_clear_depth_buffer(ctx);
+         }
+         if (buffers & BUFFER_BIT_STENCIL) {
+            _swrast_clear_stencil_buffer(ctx);
+         }
       }
    }
-
-   swrast_render_finish(ctx);
 }
diff --git a/mesalib/src/mesa/swrast/s_context.h b/mesalib/src/mesa/swrast/s_context.h
index 446b990d9..af9e49ecb 100644
--- a/mesalib/src/mesa/swrast/s_context.h
+++ b/mesalib/src/mesa/swrast/s_context.h
@@ -422,5 +422,17 @@ _swrast_unmap_renderbuffers(struct gl_context *ctx);
 #define ATTRIB_LOOP_END } }
 
 
+/**
+ * Return the address of a pixel value in a mapped renderbuffer.
+ */
+static inline GLubyte *
+_swrast_pixel_address(struct gl_renderbuffer *rb, GLint x, GLint y)
+{
+   const GLint bpp = _mesa_get_format_bytes(rb->Format);
+   const GLint rowStride = rb->RowStride * bpp;
+   return (GLubyte *) rb->Data + y * rowStride + x * bpp;
+}
+
+
 
 #endif
diff --git a/mesalib/src/mesa/swrast/s_copypix.c b/mesalib/src/mesa/swrast/s_copypix.c
index 3bdf48b25..1dc4286a3 100644
--- a/mesalib/src/mesa/swrast/s_copypix.c
+++ b/mesalib/src/mesa/swrast/s_copypix.c
@@ -245,7 +245,7 @@ copy_depth_pixels( struct gl_context *ctx, GLint srcx, GLint srcy,
                    GLint destx, GLint desty )
 {
    struct gl_framebuffer *fb = ctx->ReadBuffer;
-   struct gl_renderbuffer *readRb = fb->_DepthBuffer;
+   struct gl_renderbuffer *readRb = fb->Attachment[BUFFER_DEPTH].Renderbuffer;
    GLfloat *p, *tmpImage;
    GLint sy, dy, stepy;
    GLint j;
@@ -339,7 +339,7 @@ copy_stencil_pixels( struct gl_context *ctx, GLint srcx, GLint srcy,
                      GLint destx, GLint desty )
 {
    struct gl_framebuffer *fb = ctx->ReadBuffer;
-   struct gl_renderbuffer *rb = fb->_StencilBuffer;
+   struct gl_renderbuffer *rb = fb->Attachment[BUFFER_STENCIL].Renderbuffer;
    GLint sy, dy, stepy;
    GLint j;
    GLubyte *p, *tmpImage;
@@ -422,184 +422,8 @@ copy_stencil_pixels( struct gl_context *ctx, GLint srcx, GLint srcy,
 
 
 /**
- * This isn't terribly efficient.  If a driver really has combined
- * depth/stencil buffers the driver should implement an optimized
- * CopyPixels function.
- */
-static void
-copy_depth_stencil_pixels(struct gl_context *ctx,
-                          const GLint srcX, const GLint srcY,
-                          const GLint width, const GLint height,
-                          const GLint destX, const GLint destY)
-{
-   struct gl_renderbuffer *stencilReadRb, *depthReadRb, *depthDrawRb;
-   GLint sy, dy, stepy;
-   GLint j;
-   GLubyte *tempStencilImage = NULL, *stencilPtr = NULL;
-   GLfloat *tempDepthImage = NULL, *depthPtr = NULL;
-   const GLfloat depthScale = ctx->DrawBuffer->_DepthMaxF;
-   const GLuint stencilMask = ctx->Stencil.WriteMask[0];
-   const GLboolean zoom = ctx->Pixel.ZoomX != 1.0F || ctx->Pixel.ZoomY != 1.0F;
-   const GLboolean scaleOrBias
-      = ctx->Pixel.DepthScale != 1.0 || ctx->Pixel.DepthBias != 0.0;
-   GLint overlapping;
-
-   depthDrawRb = ctx->DrawBuffer->_DepthBuffer;
-   depthReadRb = ctx->ReadBuffer->_DepthBuffer;
-   stencilReadRb = ctx->ReadBuffer->_StencilBuffer;
-
-   ASSERT(depthDrawRb);
-   ASSERT(depthReadRb);
-   ASSERT(stencilReadRb);
-
-   if (ctx->DrawBuffer == ctx->ReadBuffer) {
-      overlapping = regions_overlap(srcX, srcY, destX, destY, width, height,
-                                    ctx->Pixel.ZoomX, ctx->Pixel.ZoomY);
-   }
-   else {
-      overlapping = GL_FALSE;
-   }
-
-   /* Determine if copy should be bottom-to-top or top-to-bottom */
-   if (!overlapping && srcY < destY) {
-      /* top-down  max-to-min */
-      sy = srcY + height - 1;
-      dy = destY + height - 1;
-      stepy = -1;
-   }
-   else {
-      /* bottom-up  min-to-max */
-      sy = srcY;
-      dy = destY;
-      stepy = 1;
-   }
-
-   if (overlapping) {
-      GLint ssy = sy;
-
-      if (stencilMask != 0x0) {
-         tempStencilImage
-            = (GLubyte *) malloc(width * height * sizeof(GLubyte));
-         if (!tempStencilImage) {
-            _mesa_error(ctx, GL_OUT_OF_MEMORY, "glCopyPixels");
-            return;
-         }
-
-         /* get copy of stencil pixels */
-         stencilPtr = tempStencilImage;
-         for (j = 0; j < height; j++, ssy += stepy) {
-            _swrast_read_stencil_span(ctx, stencilReadRb,
-                                      width, srcX, ssy, stencilPtr);
-            stencilPtr += width;
-         }
-         stencilPtr = tempStencilImage;
-      }
-
-      if (ctx->Depth.Mask) {
-         tempDepthImage
-            = (GLfloat *) malloc(width * height * sizeof(GLfloat));
-         if (!tempDepthImage) {
-            _mesa_error(ctx, GL_OUT_OF_MEMORY, "glCopyPixels");
-            free(tempStencilImage);
-            return;
-         }
-
-         /* get copy of depth pixels */
-         depthPtr = tempDepthImage;
-         for (j = 0; j < height; j++, ssy += stepy) {
-            _swrast_read_depth_span_float(ctx, depthReadRb,
-                                          width, srcX, ssy, depthPtr);
-            depthPtr += width;
-         }
-         depthPtr = tempDepthImage;
-      }
-   }
-
-   for (j = 0; j < height; j++, sy += stepy, dy += stepy) {
-      if (stencilMask != 0x0) {
-         GLubyte stencil[MAX_WIDTH];
-
-         /* Get stencil values */
-         if (overlapping) {
-            memcpy(stencil, stencilPtr, width * sizeof(GLubyte));
-            stencilPtr += width;
-         }
-         else {
-            _swrast_read_stencil_span(ctx, stencilReadRb,
-                                      width, srcX, sy, stencil);
-         }
-
-         _mesa_apply_stencil_transfer_ops(ctx, width, stencil);
-
-         /* Write values */
-         if (zoom) {
-            _swrast_write_zoomed_stencil_span(ctx, destX, destY, width,
-                                              destX, dy, stencil);
-         }
-         else {
-            _swrast_write_stencil_span( ctx, width, destX, dy, stencil );
-         }
-      }
-
-      if (ctx->Depth.Mask) {
-         GLfloat depth[MAX_WIDTH];
-         GLuint zVals32[MAX_WIDTH];
-         GLushort zVals16[MAX_WIDTH];
-         GLvoid *zVals;
-         GLuint zBytes;
-
-         /* get depth values */
-         if (overlapping) {
-            memcpy(depth, depthPtr, width * sizeof(GLfloat));
-            depthPtr += width;
-         }
-         else {
-            _swrast_read_depth_span_float(ctx, depthReadRb,
-                                          width, srcX, sy, depth);
-         }
-
-         /* scale & bias */
-         if (scaleOrBias) {
-            _mesa_scale_and_bias_depth(ctx, width, depth);
-         }
-         /* convert to integer Z values */
-         if (depthDrawRb->DataType == GL_UNSIGNED_SHORT) {
-            GLint k;
-            for (k = 0; k < width; k++)
-               zVals16[k] = (GLushort) (depth[k] * depthScale);
-            zVals = zVals16;
-            zBytes = 2;
-         }
-         else {
-            GLint k;
-            for (k = 0; k < width; k++)
-               zVals32[k] = (GLuint) (depth[k] * depthScale);
-            zVals = zVals32;
-            zBytes = 4;
-         }
-
-         /* Write values */
-         if (zoom) {
-            _swrast_write_zoomed_z_span(ctx, destX, destY, width,
-                                        destX, dy, zVals);
-         }
-         else {
-            _swrast_put_row(ctx, depthDrawRb, width, destX, dy, zVals, zBytes);
-         }
-      }
-   }
-
-   if (tempStencilImage)
-      free(tempStencilImage);
-
-   if (tempDepthImage)
-      free(tempDepthImage);
-}
-
-
-
-/**
- * Try to do a fast copy pixels.
+ * Try to do a fast copy pixels with memcpy.
+ * \return GL_TRUE if successful, GL_FALSE otherwise.
  */
 static GLboolean
 fast_copy_pixels(struct gl_context *ctx,
@@ -609,8 +433,10 @@ fast_copy_pixels(struct gl_context *ctx,
    struct gl_framebuffer *srcFb = ctx->ReadBuffer;
    struct gl_framebuffer *dstFb = ctx->DrawBuffer;
    struct gl_renderbuffer *srcRb, *dstRb;
-   GLint row, yStep;
-   void *temp;
+   GLint row;
+   GLuint pixelBytes, widthInBytes;
+   GLubyte *srcMap, *dstMap;
+   GLint srcRowStride, dstRowStride;
 
    if (SWRAST_CONTEXT(ctx)->_RasterMask != 0x0 ||
        ctx->Pixel.ZoomX != 1.0F ||
@@ -627,12 +453,12 @@ fast_copy_pixels(struct gl_context *ctx,
       dstRb = dstFb->_ColorDrawBuffers[0];
    }
    else if (type == GL_STENCIL) {
-      srcRb = srcFb->_StencilBuffer;
-      dstRb = dstFb->_StencilBuffer;
+      srcRb = srcFb->Attachment[BUFFER_STENCIL].Renderbuffer;
+      dstRb = dstFb->Attachment[BUFFER_STENCIL].Renderbuffer;
    }
    else if (type == GL_DEPTH) {
-      srcRb = srcFb->_DepthBuffer;
-      dstRb = dstFb->_DepthBuffer;
+      srcRb = srcFb->Attachment[BUFFER_DEPTH].Renderbuffer;
+      dstRb = dstFb->Attachment[BUFFER_DEPTH].Renderbuffer;
    }
    else {
       ASSERT(type == GL_DEPTH_STENCIL_EXT);
@@ -641,13 +467,24 @@ fast_copy_pixels(struct gl_context *ctx,
       dstRb = dstFb->Attachment[BUFFER_DEPTH].Renderbuffer;
    }
 
-   /* src and dst renderbuffers must be same format and type */
-   if (!srcRb || !dstRb ||
-       srcRb->DataType != dstRb->DataType ||
-       srcRb->_BaseFormat != dstRb->_BaseFormat) {
+   /* src and dst renderbuffers must be same format */
+   if (!srcRb || !dstRb || srcRb->Format != dstRb->Format) {
       return GL_FALSE;
    }
 
+   if (type == GL_STENCIL || type == GL_DEPTH_COMPONENT) {
+      /* can't handle packed depth+stencil here */
+      if (_mesa_is_format_packed_depth_stencil(srcRb->Format) ||
+          _mesa_is_format_packed_depth_stencil(dstRb->Format))
+         return GL_FALSE;
+   }
+   else if (type == GL_DEPTH_STENCIL) {
+      /* can't handle separate depth/stencil buffers */
+      if (srcRb != srcFb->Attachment[BUFFER_STENCIL].Renderbuffer ||
+          dstRb != dstFb->Attachment[BUFFER_STENCIL].Renderbuffer)
+         return GL_FALSE;
+   }
+
    /* clipping not supported */
    if (srcX < 0 || srcX + width > (GLint) srcFb->Width ||
        srcY < 0 || srcY + height > (GLint) srcFb->Height ||
@@ -656,32 +493,73 @@ fast_copy_pixels(struct gl_context *ctx,
       return GL_FALSE;
    }
 
-   /* overlapping src/dst doesn't matter, just determine Y direction */
-   if (srcY < dstY) {
-      /* top-down  max-to-min */
-      srcY = srcY + height - 1;
-      dstY = dstY + height - 1;
-      yStep = -1;
+   pixelBytes = _mesa_get_format_bytes(srcRb->Format);
+   widthInBytes = width * pixelBytes;
+
+   if (srcRb == dstRb) {
+      /* map whole buffer for read/write */
+      /* XXX we could be clever and just map the union region of the
+       * source and dest rects.
+       */
+      GLubyte *map;
+      GLint rowStride;
+
+      ctx->Driver.MapRenderbuffer(ctx, srcRb, 0, 0,
+                                  srcRb->Width, srcRb->Height,
+                                  GL_MAP_READ_BIT | GL_MAP_WRITE_BIT,
+                                  &map, &rowStride);
+      if (!srcMap) {
+         _mesa_error(ctx, GL_OUT_OF_MEMORY, "glCopyPixels");
+         return GL_TRUE; /* don't retry with slow path */
+      }
+
+      srcMap = map + srcY * rowStride + srcX * pixelBytes;
+      dstMap = map + dstY * rowStride + dstX * pixelBytes;
+
+      /* this handles overlapping copies */
+      if (srcY < dstY) {
+         /* copy in reverse (top->down) order */
+         srcMap += rowStride * (height - 1);
+         dstMap += rowStride * (height - 1);
+         srcRowStride = -rowStride;
+         dstRowStride = -rowStride;
+      }
+      else {
+         /* copy in normal (bottom->up) order */
+         srcRowStride = rowStride;
+         dstRowStride = rowStride;
+      }
    }
    else {
-      /* bottom-up  min-to-max */
-      yStep = 1;
-   }
-
-   temp = malloc(width * MAX_PIXEL_BYTES);
-   if (!temp) {
-      _mesa_error(ctx, GL_OUT_OF_MEMORY, "glCopyPixels");
-      return GL_FALSE;
+      /* different src/dst buffers */
+      ctx->Driver.MapRenderbuffer(ctx, srcRb, srcX, srcY,
+                                  width, height,
+                                  GL_MAP_READ_BIT, &srcMap, &srcRowStride);
+      if (!srcMap) {
+         _mesa_error(ctx, GL_OUT_OF_MEMORY, "glCopyPixels");
+         return GL_TRUE; /* don't retry with slow path */
+      }
+      ctx->Driver.MapRenderbuffer(ctx, dstRb, dstX, dstY,
+                                  width, height,
+                                  GL_MAP_WRITE_BIT, &dstMap, &dstRowStride);
+      if (!dstMap) {
+         ctx->Driver.UnmapRenderbuffer(ctx, srcRb);
+         _mesa_error(ctx, GL_OUT_OF_MEMORY, "glCopyPixels");
+         return GL_TRUE; /* don't retry with slow path */
+      }
    }
 
    for (row = 0; row < height; row++) {
-      srcRb->GetRow(ctx, srcRb, width, srcX, srcY, temp);
-      dstRb->PutRow(ctx, dstRb, width, dstX, dstY, temp, NULL);
-      srcY += yStep;
-      dstY += yStep;
+      /* memmove() in case of overlap */
+      memmove(dstMap, srcMap, widthInBytes);
+      dstMap += dstRowStride;
+      srcMap += srcRowStride;
    }
 
-   free(temp);
+   ctx->Driver.UnmapRenderbuffer(ctx, srcRb);
+   if (dstRb != srcRb) {
+      ctx->Driver.UnmapRenderbuffer(ctx, dstRb);
+   }
 
    return GL_TRUE;
 }
@@ -697,7 +575,6 @@ _swrast_CopyPixels( struct gl_context *ctx,
 		    GLint destx, GLint desty, GLenum type )
 {
    SWcontext *swrast = SWRAST_CONTEXT(ctx);
-   swrast_render_start(ctx);
       
    if (!_mesa_check_conditional_render(ctx))
       return; /* don't copy */
@@ -705,23 +582,30 @@ _swrast_CopyPixels( struct gl_context *ctx,
    if (swrast->NewState)
       _swrast_validate_derived( ctx );
 
-   if (!fast_copy_pixels(ctx, srcx, srcy, width, height, destx, desty, type)) {
-      switch (type) {
-      case GL_COLOR:
-         copy_rgba_pixels( ctx, srcx, srcy, width, height, destx, desty );
-         break;
-      case GL_DEPTH:
-         copy_depth_pixels( ctx, srcx, srcy, width, height, destx, desty );
-         break;
-      case GL_STENCIL:
-         copy_stencil_pixels( ctx, srcx, srcy, width, height, destx, desty );
-         break;
-      case GL_DEPTH_STENCIL_EXT:
-         copy_depth_stencil_pixels(ctx, srcx, srcy, width, height, destx, desty);
-         break;
-      default:
-         _mesa_problem(ctx, "unexpected type in _swrast_CopyPixels");
-      }
+   if (fast_copy_pixels(ctx, srcx, srcy, width, height, destx, desty, type)) {
+      /* all done */
+      return;
+   }
+
+   swrast_render_start(ctx);
+
+   switch (type) {
+   case GL_COLOR:
+      copy_rgba_pixels( ctx, srcx, srcy, width, height, destx, desty );
+      break;
+   case GL_DEPTH:
+      copy_depth_pixels( ctx, srcx, srcy, width, height, destx, desty );
+      break;
+   case GL_STENCIL:
+      copy_stencil_pixels( ctx, srcx, srcy, width, height, destx, desty );
+      break;
+   case GL_DEPTH_STENCIL_EXT:
+      /* Copy buffers separately (if the fast copy path wasn't taken) */
+      copy_depth_pixels(ctx, srcx, srcy, width, height, destx, desty);
+      copy_stencil_pixels(ctx, srcx, srcy, width, height, destx, desty);
+      break;
+   default:
+      _mesa_problem(ctx, "unexpected type in _swrast_CopyPixels");
    }
 
    swrast_render_finish(ctx);
diff --git a/mesalib/src/mesa/swrast/s_depth.c b/mesalib/src/mesa/swrast/s_depth.c
index 8d3ad105b..f87adaa84 100644
--- a/mesalib/src/mesa/swrast/s_depth.c
+++ b/mesalib/src/mesa/swrast/s_depth.c
@@ -26,478 +26,134 @@
 #include "main/glheader.h"
 #include "main/context.h"
 #include "main/formats.h"
+#include "main/format_unpack.h"
+#include "main/format_pack.h"
 #include "main/macros.h"
 #include "main/imports.h"
 
+#include "s_context.h"
 #include "s_depth.h"
 #include "s_span.h"
 
 
+
+#define Z_TEST(COMPARE)                      \
+   do {                                      \
+      GLuint i;                              \
+      for (i = 0; i < n; i++) {              \
+         if (mask[i]) {                      \
+            if (COMPARE) {                   \
+               /* pass */                    \
+               if (write) {                  \
+                  zbuffer[i] = zfrag[i];     \
+               }                             \
+               passed++;                     \
+            }                                \
+            else {                           \
+               /* fail */                    \
+               mask[i] = 0;                  \
+            }                                \
+         }                                   \
+      }                                      \
+   } while (0)
+
+
 /**
- * Do depth test for a horizontal span of fragments.
- * Input:  zbuffer - array of z values in the zbuffer
- *         z - array of fragment z values
- * Return:  number of fragments which pass the test.
+ * Do depth test for an array of 16-bit Z values.
+ * @param zbuffer  array of Z buffer values (16-bit)
+ * @param zfrag  array of fragment Z values (use 16-bit in 32-bit uint)
+ * @param mask  which fragments are alive, killed afterward
+ * @return  number of fragments which pass the test.
  */
 static GLuint
 depth_test_span16( struct gl_context *ctx, GLuint n,
-                   GLushort zbuffer[], const GLuint z[], GLubyte mask[] )
+                   GLushort zbuffer[], const GLuint zfrag[], GLubyte mask[] )
 {
+   const GLboolean write = ctx->Depth.Mask;
    GLuint passed = 0;
 
    /* switch cases ordered from most frequent to less frequent */
    switch (ctx->Depth.Func) {
-      case GL_LESS:
-         if (ctx->Depth.Mask) {
-	    /* Update Z buffer */
-            GLuint i;
-	    for (i=0; i<n; i++) {
-	       if (mask[i]) {
-		  if (z[i] < zbuffer[i]) {
-		     /* pass */
-		     zbuffer[i] = z[i];
-		     passed++;
-		  }
-		  else {
-		     /* fail */
-		     mask[i] = 0;
-		  }
-	       }
-	    }
-	 }
-	 else {
-	    /* Don't update Z buffer */
-            GLuint i;
-	    for (i=0; i<n; i++) {
-	       if (mask[i]) {
-		  if (z[i] < zbuffer[i]) {
-		     /* pass */
-		     passed++;
-		  }
-		  else {
-		     mask[i] = 0;
-		  }
-	       }
-	    }
-	 }
-	 break;
-      case GL_LEQUAL:
-	 if (ctx->Depth.Mask) {
-	    /* Update Z buffer */
-            GLuint i;
-	    for (i=0;i<n;i++) {
-	       if (mask[i]) {
-		  if (z[i] <= zbuffer[i]) {
-		     zbuffer[i] = z[i];
-		     passed++;
-		  }
-		  else {
-		     mask[i] = 0;
-		  }
-	       }
-	    }
-	 }
-	 else {
-	    /* Don't update Z buffer */
-            GLuint i;
-	    for (i=0;i<n;i++) {
-	       if (mask[i]) {
-		  if (z[i] <= zbuffer[i]) {
-		     /* pass */
-		     passed++;
-		  }
-		  else {
-		     mask[i] = 0;
-		  }
-	       }
-	    }
-	 }
-	 break;
-      case GL_GEQUAL:
-	 if (ctx->Depth.Mask) {
-	    /* Update Z buffer */
-            GLuint i;
-	    for (i=0;i<n;i++) {
-	       if (mask[i]) {
-		  if (z[i] >= zbuffer[i]) {
-		     zbuffer[i] = z[i];
-		     passed++;
-		  }
-		  else {
-		     mask[i] = 0;
-		  }
-	       }
-	    }
-	 }
-	 else {
-	    /* Don't update Z buffer */
-            GLuint i;
-	    for (i=0;i<n;i++) {
-	       if (mask[i]) {
-		  if (z[i] >= zbuffer[i]) {
-		     /* pass */
-		     passed++;
-		  }
-		  else {
-		     mask[i] = 0;
-		  }
-	       }
-	    }
-	 }
-	 break;
-      case GL_GREATER:
-	 if (ctx->Depth.Mask) {
-	    /* Update Z buffer */
-            GLuint i;
-	    for (i=0;i<n;i++) {
-	       if (mask[i]) {
-		  if (z[i] > zbuffer[i]) {
-		     zbuffer[i] = z[i];
-		     passed++;
-		  }
-		  else {
-		     mask[i] = 0;
-		  }
-	       }
-	    }
-	 }
-	 else {
-	    /* Don't update Z buffer */
-            GLuint i;
-	    for (i=0;i<n;i++) {
-	       if (mask[i]) {
-		  if (z[i] > zbuffer[i]) {
-		     /* pass */
-		     passed++;
-		  }
-		  else {
-		     mask[i] = 0;
-		  }
-	       }
-	    }
-	 }
-	 break;
-      case GL_NOTEQUAL:
-	 if (ctx->Depth.Mask) {
-	    /* Update Z buffer */
-            GLuint i;
-	    for (i=0;i<n;i++) {
-	       if (mask[i]) {
-		  if (z[i] != zbuffer[i]) {
-		     zbuffer[i] = z[i];
-		     passed++;
-		  }
-		  else {
-		     mask[i] = 0;
-		  }
-	       }
-	    }
-	 }
-	 else {
-	    /* Don't update Z buffer */
-            GLuint i;
-	    for (i=0;i<n;i++) {
-	       if (mask[i]) {
-		  if (z[i] != zbuffer[i]) {
-		     /* pass */
-		     passed++;
-		  }
-		  else {
-		     mask[i] = 0;
-		  }
-	       }
-	    }
-	 }
-	 break;
-      case GL_EQUAL:
-	 if (ctx->Depth.Mask) {
-	    /* Update Z buffer */
-            GLuint i;
-	    for (i=0;i<n;i++) {
-	       if (mask[i]) {
-		  if (z[i] == zbuffer[i]) {
-		     zbuffer[i] = z[i];
-		     passed++;
-		  }
-		  else {
-		     mask[i] = 0;
-		  }
-	       }
-	    }
-	 }
-	 else {
-	    /* Don't update Z buffer */
-            GLuint i;
-	    for (i=0;i<n;i++) {
-	       if (mask[i]) {
-		  if (z[i] == zbuffer[i]) {
-		     /* pass */
-		     passed++;
-		  }
-		  else {
-		     mask[i] = 0;
-		  }
-	       }
-	    }
-	 }
-	 break;
-      case GL_ALWAYS:
-	 if (ctx->Depth.Mask) {
-	    /* Update Z buffer */
-            GLuint i;
-	    for (i=0;i<n;i++) {
-	       if (mask[i]) {
-		  zbuffer[i] = z[i];
-		  passed++;
-	       }
-	    }
-	 }
-	 else {
-	    /* Don't update Z buffer or mask */
-	    passed = n;
-	 }
-	 break;
-      case GL_NEVER:
-         memset(mask, 0, n * sizeof(GLubyte));
-	 break;
-      default:
-         _mesa_problem(ctx, "Bad depth func in depth_test_span16");
+   case GL_LESS:
+      Z_TEST(zfrag[i] < zbuffer[i]);
+      break;
+   case GL_LEQUAL:
+      Z_TEST(zfrag[i] <= zbuffer[i]);
+      break;
+   case GL_GEQUAL:
+      Z_TEST(zfrag[i] >= zbuffer[i]);
+      break;
+   case GL_GREATER:
+      Z_TEST(zfrag[i] > zbuffer[i]);
+      break;
+   case GL_NOTEQUAL:
+      Z_TEST(zfrag[i] != zbuffer[i]);
+      break;
+   case GL_EQUAL:
+      Z_TEST(zfrag[i] == zbuffer[i]);
+      break;
+   case GL_ALWAYS:
+      Z_TEST(1);
+      break;
+   case GL_NEVER:
+      memset(mask, 0, n * sizeof(GLubyte));
+      break;
+   default:
+      _mesa_problem(ctx, "Bad depth func in depth_test_span16");
    }
 
    return passed;
 }
 
 
+/**
+ * Do depth test for an array of 32-bit Z values.
+ * @param zbuffer  array of Z buffer values (32-bit)
+ * @param zfrag  array of fragment Z values (use 32-bits in 32-bit uint)
+ * @param mask  which fragments are alive, killed afterward
+ * @return  number of fragments which pass the test.
+ */
 static GLuint
 depth_test_span32( struct gl_context *ctx, GLuint n,
-                   GLuint zbuffer[], const GLuint z[], GLubyte mask[] )
+                   GLuint zbuffer[], const GLuint zfrag[], GLubyte mask[])
 {
+   const GLboolean write = ctx->Depth.Mask;
    GLuint passed = 0;
 
    /* switch cases ordered from most frequent to less frequent */
    switch (ctx->Depth.Func) {
-      case GL_LESS:
-         if (ctx->Depth.Mask) {
-	    /* Update Z buffer */
-            GLuint i;
-	    for (i=0; i<n; i++) {
-	       if (mask[i]) {
-		  if (z[i] < zbuffer[i]) {
-		     /* pass */
-		     zbuffer[i] = z[i];
-		     passed++;
-		  }
-		  else {
-		     /* fail */
-		     mask[i] = 0;
-		  }
-	       }
-	    }
-	 }
-	 else {
-	    /* Don't update Z buffer */
-            GLuint i;
-	    for (i=0; i<n; i++) {
-	       if (mask[i]) {
-		  if (z[i] < zbuffer[i]) {
-		     /* pass */
-		     passed++;
-		  }
-		  else {
-		     mask[i] = 0;
-		  }
-	       }
-	    }
-	 }
-	 break;
-      case GL_LEQUAL:
-	 if (ctx->Depth.Mask) {
-	    /* Update Z buffer */
-            GLuint i;
-	    for (i=0;i<n;i++) {
-	       if (mask[i]) {
-		  if (z[i] <= zbuffer[i]) {
-		     zbuffer[i] = z[i];
-		     passed++;
-		  }
-		  else {
-		     mask[i] = 0;
-		  }
-	       }
-	    }
-	 }
-	 else {
-	    /* Don't update Z buffer */
-            GLuint i;
-	    for (i=0;i<n;i++) {
-	       if (mask[i]) {
-		  if (z[i] <= zbuffer[i]) {
-		     /* pass */
-		     passed++;
-		  }
-		  else {
-		     mask[i] = 0;
-		  }
-	       }
-	    }
-	 }
-	 break;
-      case GL_GEQUAL:
-	 if (ctx->Depth.Mask) {
-	    /* Update Z buffer */
-            GLuint i;
-	    for (i=0;i<n;i++) {
-	       if (mask[i]) {
-		  if (z[i] >= zbuffer[i]) {
-		     zbuffer[i] = z[i];
-		     passed++;
-		  }
-		  else {
-		     mask[i] = 0;
-		  }
-	       }
-	    }
-	 }
-	 else {
-	    /* Don't update Z buffer */
-            GLuint i;
-	    for (i=0;i<n;i++) {
-	       if (mask[i]) {
-		  if (z[i] >= zbuffer[i]) {
-		     /* pass */
-		     passed++;
-		  }
-		  else {
-		     mask[i] = 0;
-		  }
-	       }
-	    }
-	 }
-	 break;
-      case GL_GREATER:
-	 if (ctx->Depth.Mask) {
-	    /* Update Z buffer */
-            GLuint i;
-	    for (i=0;i<n;i++) {
-	       if (mask[i]) {
-		  if (z[i] > zbuffer[i]) {
-		     zbuffer[i] = z[i];
-		     passed++;
-		  }
-		  else {
-		     mask[i] = 0;
-		  }
-	       }
-	    }
-	 }
-	 else {
-	    /* Don't update Z buffer */
-            GLuint i;
-	    for (i=0;i<n;i++) {
-	       if (mask[i]) {
-		  if (z[i] > zbuffer[i]) {
-		     /* pass */
-		     passed++;
-		  }
-		  else {
-		     mask[i] = 0;
-		  }
-	       }
-	    }
-	 }
-	 break;
-      case GL_NOTEQUAL:
-	 if (ctx->Depth.Mask) {
-	    /* Update Z buffer */
-            GLuint i;
-	    for (i=0;i<n;i++) {
-	       if (mask[i]) {
-		  if (z[i] != zbuffer[i]) {
-		     zbuffer[i] = z[i];
-		     passed++;
-		  }
-		  else {
-		     mask[i] = 0;
-		  }
-	       }
-	    }
-	 }
-	 else {
-	    /* Don't update Z buffer */
-            GLuint i;
-	    for (i=0;i<n;i++) {
-	       if (mask[i]) {
-		  if (z[i] != zbuffer[i]) {
-		     /* pass */
-		     passed++;
-		  }
-		  else {
-		     mask[i] = 0;
-		  }
-	       }
-	    }
-	 }
-	 break;
-      case GL_EQUAL:
-	 if (ctx->Depth.Mask) {
-	    /* Update Z buffer */
-            GLuint i;
-	    for (i=0;i<n;i++) {
-	       if (mask[i]) {
-		  if (z[i] == zbuffer[i]) {
-		     zbuffer[i] = z[i];
-		     passed++;
-		  }
-		  else {
-		     mask[i] = 0;
-		  }
-	       }
-	    }
-	 }
-	 else {
-	    /* Don't update Z buffer */
-            GLuint i;
-	    for (i=0;i<n;i++) {
-	       if (mask[i]) {
-		  if (z[i] == zbuffer[i]) {
-		     /* pass */
-		     passed++;
-		  }
-		  else {
-		     mask[i] = 0;
-		  }
-	       }
-	    }
-	 }
-	 break;
-      case GL_ALWAYS:
-	 if (ctx->Depth.Mask) {
-	    /* Update Z buffer */
-            GLuint i;
-	    for (i=0;i<n;i++) {
-	       if (mask[i]) {
-		  zbuffer[i] = z[i];
-		  passed++;
-	       }
-	    }
-	 }
-	 else {
-	    /* Don't update Z buffer or mask */
-	    passed = n;
-	 }
-	 break;
-      case GL_NEVER:
-         memset(mask, 0, n * sizeof(GLubyte));
-	 break;
-      default:
-         _mesa_problem(ctx, "Bad depth func in depth_test_span32");
+   case GL_LESS:
+      Z_TEST(zfrag[i] < zbuffer[i]);
+      break;
+   case GL_LEQUAL:
+      Z_TEST(zfrag[i] <= zbuffer[i]);
+      break;
+   case GL_GEQUAL:
+      Z_TEST(zfrag[i] >= zbuffer[i]);
+      break;
+   case GL_GREATER:
+      Z_TEST(zfrag[i] > zbuffer[i]);
+      break;
+   case GL_NOTEQUAL:
+      Z_TEST(zfrag[i] != zbuffer[i]);
+      break;
+   case GL_EQUAL:
+      Z_TEST(zfrag[i] == zbuffer[i]);
+      break;
+   case GL_ALWAYS:
+      Z_TEST(1);
+      break;
+   case GL_NEVER:
+      memset(mask, 0, n * sizeof(GLubyte));
+      break;
+   default:
+      _mesa_problem(ctx, "Bad depth func in depth_test_span32");
    }
 
    return passed;
 }
 
 
-
 /**
  * Clamp fragment Z values to the depth near/far range (glDepthRange()).
  * This is used when GL_ARB_depth_clamp/GL_DEPTH_CLAMP is turned on.
@@ -547,615 +203,171 @@ _swrast_depth_clamp_span( struct gl_context *ctx, SWspan *span )
 }
 
 
-
-/*
- * Apply depth test to span of fragments.
+/**
+ * Get array of 32-bit z values from the depth buffer.  With clipping.
  */
-static GLuint
-depth_test_span( struct gl_context *ctx, SWspan *span)
+static void
+get_z32_values(struct gl_context *ctx, struct gl_renderbuffer *rb,
+               GLuint count, const GLint x[], const GLint y[],
+               GLuint zbuffer[])
 {
-   struct gl_framebuffer *fb = ctx->DrawBuffer;
-   struct gl_renderbuffer *rb = fb->_DepthBuffer;
-   const GLint x = span->x;
-   const GLint y = span->y;
-   const GLuint count = span->end;
-   const GLuint *zValues = span->array->z;
-   GLubyte *mask = span->array->mask;
-   GLuint passed;
+   const GLint w = rb->Width, h = rb->Height;
+   const GLubyte *map = (const GLubyte *) rb->Data;
+   GLuint i;
 
-   ASSERT((span->arrayMask & SPAN_XY) == 0);
-   ASSERT(span->arrayMask & SPAN_Z);
-   
-   if (rb->GetPointer(ctx, rb, 0, 0)) {
-      /* Directly access buffer */
-      if (rb->DataType == GL_UNSIGNED_SHORT) {
-         GLushort *zbuffer = (GLushort *) rb->GetPointer(ctx, rb, x, y);
-         passed = depth_test_span16(ctx, count, zbuffer, zValues, mask);
-      }
-      else {
-         GLuint *zbuffer = (GLuint *) rb->GetPointer(ctx, rb, x, y);
-         ASSERT(rb->DataType == GL_UNSIGNED_INT);
-         passed = depth_test_span32(ctx, count, zbuffer, zValues, mask);
+   if (rb->Format == MESA_FORMAT_Z32) {
+      const GLint rowStride = rb->RowStride * 4;
+      for (i = 0; i < count; i++) {
+         if (x[i] >= 0 && y[i] >= 0 && x[i] < w && y[i] < h) {
+            zbuffer[i] = *((GLuint *) (map + y[i] * rowStride + x[i] * 4));
+         }
       }
    }
    else {
-      /* read depth values from buffer, test, write back */
-      if (rb->DataType == GL_UNSIGNED_SHORT) {
-         GLushort zbuffer[MAX_WIDTH];
-         rb->GetRow(ctx, rb, count, x, y, zbuffer);
-         passed = depth_test_span16(ctx, count, zbuffer, zValues, mask);
-         rb->PutRow(ctx, rb, count, x, y, zbuffer, mask);
-      }
-      else {
-         GLuint zbuffer[MAX_WIDTH];
-         ASSERT(rb->DataType == GL_UNSIGNED_INT);
-         rb->GetRow(ctx, rb, count, x, y, zbuffer);
-         passed = depth_test_span32(ctx, count, zbuffer, zValues, mask);
-         rb->PutRow(ctx, rb, count, x, y, zbuffer, mask);
+      const GLint bpp = _mesa_get_format_bytes(rb->Format);
+      const GLint rowStride = rb->RowStride * bpp;
+      for (i = 0; i < count; i++) {
+         if (x[i] >= 0 && y[i] >= 0 && x[i] < w && y[i] < h) {
+            const GLubyte *src = map + y[i] * rowStride+ x[i] * bpp;
+            _mesa_unpack_uint_z_row(rb->Format, 1, src, &zbuffer[i]);
+         }
       }
    }
-
-   if (passed < count) {
-      span->writeAll = GL_FALSE;
-   }
-   return passed;
 }
 
-
-
-#define Z_ADDRESS(X, Y)   (zStart + (Y) * stride + (X))
-
-
-/*
- * Do depth testing for an array of fragments at assorted locations.
- */
 static void
-direct_depth_test_pixels16(struct gl_context *ctx, GLushort *zStart, GLuint stride,
-                           GLuint n, const GLint x[], const GLint y[],
-                           const GLuint z[], GLubyte mask[] )
+put_z32_values(struct gl_context *ctx, struct gl_renderbuffer *rb,
+               GLuint count, const GLint x[], const GLint y[],
+               const GLuint zvalues[], const GLubyte mask[])
 {
-   /* switch cases ordered from most frequent to less frequent */
-   switch (ctx->Depth.Func) {
-      case GL_LESS:
-         if (ctx->Depth.Mask) {
-	    /* Update Z buffer */
-            GLuint i;
-	    for (i=0; i<n; i++) {
-	       if (mask[i]) {
-		  GLushort *zptr = Z_ADDRESS(x[i], y[i]);
-		  if (z[i] < *zptr) {
-		     /* pass */
-		     *zptr = z[i];
-		  }
-		  else {
-		     /* fail */
-		     mask[i] = 0;
-		  }
-	       }
-	    }
-	 }
-	 else {
-	    /* Don't update Z buffer */
-            GLuint i;
-	    for (i=0; i<n; i++) {
-	       if (mask[i]) {
-		  GLushort *zptr = Z_ADDRESS(x[i], y[i]);
-		  if (z[i] < *zptr) {
-		     /* pass */
-		  }
-		  else {
-		     /* fail */
-		     mask[i] = 0;
-		  }
-	       }
-	    }
-	 }
-	 break;
-      case GL_LEQUAL:
-         if (ctx->Depth.Mask) {
-	    /* Update Z buffer */
-            GLuint i;
-	    for (i=0; i<n; i++) {
-	       if (mask[i]) {
-		  GLushort *zptr = Z_ADDRESS(x[i], y[i]);
-		  if (z[i] <= *zptr) {
-		     /* pass */
-		     *zptr = z[i];
-		  }
-		  else {
-		     /* fail */
-		     mask[i] = 0;
-		  }
-	       }
-	    }
-	 }
-	 else {
-	    /* Don't update Z buffer */
-            GLuint i;
-	    for (i=0; i<n; i++) {
-	       if (mask[i]) {
-		  GLushort *zptr = Z_ADDRESS(x[i], y[i]);
-		  if (z[i] <= *zptr) {
-		     /* pass */
-		  }
-		  else {
-		     /* fail */
-		     mask[i] = 0;
-		  }
-	       }
-	    }
-	 }
-	 break;
-      case GL_GEQUAL:
-         if (ctx->Depth.Mask) {
-	    /* Update Z buffer */
-            GLuint i;
-	    for (i=0; i<n; i++) {
-	       if (mask[i]) {
-		  GLushort *zptr = Z_ADDRESS(x[i], y[i]);
-		  if (z[i] >= *zptr) {
-		     /* pass */
-		     *zptr = z[i];
-		  }
-		  else {
-		     /* fail */
-		     mask[i] = 0;
-		  }
-	       }
-	    }
-	 }
-	 else {
-	    /* Don't update Z buffer */
-            GLuint i;
-	    for (i=0; i<n; i++) {
-	       if (mask[i]) {
-		  GLushort *zptr = Z_ADDRESS(x[i], y[i]);
-		  if (z[i] >= *zptr) {
-		     /* pass */
-		  }
-		  else {
-		     /* fail */
-		     mask[i] = 0;
-		  }
-	       }
-	    }
-	 }
-	 break;
-      case GL_GREATER:
-         if (ctx->Depth.Mask) {
-	    /* Update Z buffer */
-            GLuint i;
-	    for (i=0; i<n; i++) {
-	       if (mask[i]) {
-		  GLushort *zptr = Z_ADDRESS(x[i], y[i]);
-		  if (z[i] > *zptr) {
-		     /* pass */
-		     *zptr = z[i];
-		  }
-		  else {
-		     /* fail */
-		     mask[i] = 0;
-		  }
-	       }
-	    }
-	 }
-	 else {
-	    /* Don't update Z buffer */
-            GLuint i;
-	    for (i=0; i<n; i++) {
-	       if (mask[i]) {
-		  GLushort *zptr = Z_ADDRESS(x[i], y[i]);
-		  if (z[i] > *zptr) {
-		     /* pass */
-		  }
-		  else {
-		     /* fail */
-		     mask[i] = 0;
-		  }
-	       }
-	    }
-	 }
-	 break;
-      case GL_NOTEQUAL:
-         if (ctx->Depth.Mask) {
-	    /* Update Z buffer */
-            GLuint i;
-	    for (i=0; i<n; i++) {
-	       if (mask[i]) {
-		  GLushort *zptr = Z_ADDRESS(x[i], y[i]);
-		  if (z[i] != *zptr) {
-		     /* pass */
-		     *zptr = z[i];
-		  }
-		  else {
-		     /* fail */
-		     mask[i] = 0;
-		  }
-	       }
-	    }
-	 }
-	 else {
-	    /* Don't update Z buffer */
-            GLuint i;
-	    for (i=0; i<n; i++) {
-	       if (mask[i]) {
-		  GLushort *zptr = Z_ADDRESS(x[i], y[i]);
-		  if (z[i] != *zptr) {
-		     /* pass */
-		  }
-		  else {
-		     /* fail */
-		     mask[i] = 0;
-		  }
-	       }
-	    }
-	 }
-	 break;
-      case GL_EQUAL:
-         if (ctx->Depth.Mask) {
-	    /* Update Z buffer */
-            GLuint i;
-	    for (i=0; i<n; i++) {
-	       if (mask[i]) {
-		  GLushort *zptr = Z_ADDRESS(x[i], y[i]);
-		  if (z[i] == *zptr) {
-		     /* pass */
-		     *zptr = z[i];
-		  }
-		  else {
-		     /* fail */
-		     mask[i] = 0;
-		  }
-	       }
-	    }
-	 }
-	 else {
-	    /* Don't update Z buffer */
-            GLuint i;
-	    for (i=0; i<n; i++) {
-	       if (mask[i]) {
-		  GLushort *zptr = Z_ADDRESS(x[i], y[i]);
-		  if (z[i] == *zptr) {
-		     /* pass */
-		  }
-		  else {
-		     /* fail */
-		     mask[i] = 0;
-		  }
-	       }
-	    }
-	 }
-	 break;
-      case GL_ALWAYS:
-	 if (ctx->Depth.Mask) {
-	    /* Update Z buffer */
-            GLuint i;
-	    for (i=0; i<n; i++) {
-	       if (mask[i]) {
-		  GLushort *zptr = Z_ADDRESS(x[i], y[i]);
-		  *zptr = z[i];
-	       }
-	    }
-	 }
-	 else {
-	    /* Don't update Z buffer or mask */
-	 }
-	 break;
-      case GL_NEVER:
-	 /* depth test never passes */
-         memset(mask, 0, n * sizeof(GLubyte));
-	 break;
-      default:
-         _mesa_problem(ctx, "Bad depth func in direct_depth_test_pixels");
-   }
-}
-
-
+   const GLint w = rb->Width, h = rb->Height;
+   GLubyte *map = (GLubyte *) rb->Data;
+   GLuint i;
 
-/*
- * Do depth testing for an array of fragments with direct access to zbuffer.
- */
-static void
-direct_depth_test_pixels32(struct gl_context *ctx, GLuint *zStart, GLuint stride,
-                           GLuint n, const GLint x[], const GLint y[],
-                           const GLuint z[], GLubyte mask[] )
-{
-   /* switch cases ordered from most frequent to less frequent */
-   switch (ctx->Depth.Func) {
-      case GL_LESS:
-         if (ctx->Depth.Mask) {
-	    /* Update Z buffer */
-            GLuint i;
-	    for (i=0; i<n; i++) {
-	       if (mask[i]) {
-		  GLuint *zptr = Z_ADDRESS(x[i], y[i]);
-		  if (z[i] < *zptr) {
-		     /* pass */
-		     *zptr = z[i];
-		  }
-		  else {
-		     /* fail */
-		     mask[i] = 0;
-		  }
-	       }
-	    }
-	 }
-	 else {
-	    /* Don't update Z buffer */
-            GLuint i;
-	    for (i=0; i<n; i++) {
-	       if (mask[i]) {
-		  GLuint *zptr = Z_ADDRESS(x[i], y[i]);
-		  if (z[i] < *zptr) {
-		     /* pass */
-		  }
-		  else {
-		     /* fail */
-		     mask[i] = 0;
-		  }
-	       }
-	    }
-	 }
-	 break;
-      case GL_LEQUAL:
-         if (ctx->Depth.Mask) {
-	    /* Update Z buffer */
-            GLuint i;
-	    for (i=0; i<n; i++) {
-	       if (mask[i]) {
-		  GLuint *zptr = Z_ADDRESS(x[i], y[i]);
-		  if (z[i] <= *zptr) {
-		     /* pass */
-		     *zptr = z[i];
-		  }
-		  else {
-		     /* fail */
-		     mask[i] = 0;
-		  }
-	       }
-	    }
-	 }
-	 else {
-	    /* Don't update Z buffer */
-            GLuint i;
-	    for (i=0; i<n; i++) {
-	       if (mask[i]) {
-		  GLuint *zptr = Z_ADDRESS(x[i], y[i]);
-		  if (z[i] <= *zptr) {
-		     /* pass */
-		  }
-		  else {
-		     /* fail */
-		     mask[i] = 0;
-		  }
-	       }
-	    }
-	 }
-	 break;
-      case GL_GEQUAL:
-         if (ctx->Depth.Mask) {
-	    /* Update Z buffer */
-            GLuint i;
-	    for (i=0; i<n; i++) {
-	       if (mask[i]) {
-		  GLuint *zptr = Z_ADDRESS(x[i], y[i]);
-		  if (z[i] >= *zptr) {
-		     /* pass */
-		     *zptr = z[i];
-		  }
-		  else {
-		     /* fail */
-		     mask[i] = 0;
-		  }
-	       }
-	    }
-	 }
-	 else {
-	    /* Don't update Z buffer */
-            GLuint i;
-	    for (i=0; i<n; i++) {
-	       if (mask[i]) {
-		  GLuint *zptr = Z_ADDRESS(x[i], y[i]);
-		  if (z[i] >= *zptr) {
-		     /* pass */
-		  }
-		  else {
-		     /* fail */
-		     mask[i] = 0;
-		  }
-	       }
-	    }
-	 }
-	 break;
-      case GL_GREATER:
-         if (ctx->Depth.Mask) {
-	    /* Update Z buffer */
-            GLuint i;
-	    for (i=0; i<n; i++) {
-	       if (mask[i]) {
-		  GLuint *zptr = Z_ADDRESS(x[i], y[i]);
-		  if (z[i] > *zptr) {
-		     /* pass */
-		     *zptr = z[i];
-		  }
-		  else {
-		     /* fail */
-		     mask[i] = 0;
-		  }
-	       }
-	    }
-	 }
-	 else {
-	    /* Don't update Z buffer */
-            GLuint i;
-	    for (i=0; i<n; i++) {
-	       if (mask[i]) {
-		  GLuint *zptr = Z_ADDRESS(x[i], y[i]);
-		  if (z[i] > *zptr) {
-		     /* pass */
-		  }
-		  else {
-		     /* fail */
-		     mask[i] = 0;
-		  }
-	       }
-	    }
-	 }
-	 break;
-      case GL_NOTEQUAL:
-         if (ctx->Depth.Mask) {
-	    /* Update Z buffer */
-            GLuint i;
-	    for (i=0; i<n; i++) {
-	       if (mask[i]) {
-		  GLuint *zptr = Z_ADDRESS(x[i], y[i]);
-		  if (z[i] != *zptr) {
-		     /* pass */
-		     *zptr = z[i];
-		  }
-		  else {
-		     /* fail */
-		     mask[i] = 0;
-		  }
-	       }
-	    }
-	 }
-	 else {
-	    /* Don't update Z buffer */
-            GLuint i;
-	    for (i=0; i<n; i++) {
-	       if (mask[i]) {
-		  GLuint *zptr = Z_ADDRESS(x[i], y[i]);
-		  if (z[i] != *zptr) {
-		     /* pass */
-		  }
-		  else {
-		     /* fail */
-		     mask[i] = 0;
-		  }
-	       }
-	    }
-	 }
-	 break;
-      case GL_EQUAL:
-         if (ctx->Depth.Mask) {
-	    /* Update Z buffer */
-            GLuint i;
-	    for (i=0; i<n; i++) {
-	       if (mask[i]) {
-		  GLuint *zptr = Z_ADDRESS(x[i], y[i]);
-		  if (z[i] == *zptr) {
-		     /* pass */
-		     *zptr = z[i];
-		  }
-		  else {
-		     /* fail */
-		     mask[i] = 0;
-		  }
-	       }
-	    }
-	 }
-	 else {
-	    /* Don't update Z buffer */
-            GLuint i;
-	    for (i=0; i<n; i++) {
-	       if (mask[i]) {
-		  GLuint *zptr = Z_ADDRESS(x[i], y[i]);
-		  if (z[i] == *zptr) {
-		     /* pass */
-		  }
-		  else {
-		     /* fail */
-		     mask[i] = 0;
-		  }
-	       }
-	    }
-	 }
-	 break;
-      case GL_ALWAYS:
-	 if (ctx->Depth.Mask) {
-	    /* Update Z buffer */
-            GLuint i;
-	    for (i=0; i<n; i++) {
-	       if (mask[i]) {
-		  GLuint *zptr = Z_ADDRESS(x[i], y[i]);
-		  *zptr = z[i];
-	       }
-	    }
-	 }
-	 else {
-	    /* Don't update Z buffer or mask */
-	 }
-	 break;
-      case GL_NEVER:
-	 /* depth test never passes */
-         memset(mask, 0, n * sizeof(GLubyte));
-	 break;
-      default:
-         _mesa_problem(ctx, "Bad depth func in direct_depth_test_pixels");
+   if (rb->Format == MESA_FORMAT_Z32) {
+      const GLuint rowStride = rb->RowStride * 4;
+      for (i = 0; i < count; i++) {
+         if (mask[i] && x[i] >= 0 && y[i] >= 0 && x[i] < w && y[i] < h) {
+            GLuint *dst = (GLuint *) (map + y[i] * rowStride + x[i] * 4);
+            *dst = zvalues[i];
+         }
+      }
+   }
+   else {
+      gl_pack_uint_z_func packZ = _mesa_get_pack_uint_z_func(rb->Format);
+      const GLint bpp = _mesa_get_format_bytes(rb->Format);
+      const GLint rowStride = rb->RowStride * bpp;
+      for (i = 0; i < count; i++) {
+         if (mask[i] && x[i] >= 0 && y[i] >= 0 && x[i] < w && y[i] < h) {
+            void *dst = map + y[i] * rowStride + x[i] * bpp;
+            packZ(zvalues + i, dst);
+         }
+      }
    }
 }
 
 
-
-
-static GLuint
-depth_test_pixels( struct gl_context *ctx, SWspan *span )
+/**
+ * Apply depth (Z) buffer testing to the span.
+ * \return approx number of pixels that passed (only zero is reliable)
+ */
+GLuint
+_swrast_depth_test_span(struct gl_context *ctx, SWspan *span)
 {
    struct gl_framebuffer *fb = ctx->DrawBuffer;
-   struct gl_renderbuffer *rb = fb->_DepthBuffer;
+   struct gl_renderbuffer *rb = fb->Attachment[BUFFER_DEPTH].Renderbuffer;
+   const GLint bpp = _mesa_get_format_bytes(rb->Format);
+   void *zStart = _swrast_pixel_address(rb, span->x, span->y);
    const GLuint count = span->end;
-   const GLint *x = span->array->x;
-   const GLint *y = span->array->y;
-   const GLuint *z = span->array->z;
+   const GLuint *fragZ = span->array->z;
    GLubyte *mask = span->array->mask;
+   void *zBufferVals;
+   GLuint *zBufferTemp = NULL;
+   GLuint passed;
+   GLboolean ztest16 = GL_FALSE;
+   GLboolean ztest24 = _mesa_get_format_bits(rb->Format, GL_DEPTH_BITS) == 24;
 
-   if (rb->GetPointer(ctx, rb, 0, 0)) {
-      /* Directly access values */
-      if (rb->DataType == GL_UNSIGNED_SHORT) {
-         GLushort *zStart = (GLushort *) rb->Data;
-         GLuint stride = rb->Width;
-         direct_depth_test_pixels16(ctx, zStart, stride, count, x, y, z, mask);
-      }
-      else {
-         GLuint *zStart = (GLuint *) rb->Data;
-         GLuint stride = rb->Width;
-         ASSERT(rb->DataType == GL_UNSIGNED_INT);
-         direct_depth_test_pixels32(ctx, zStart, stride, count, x, y, z, mask);
-      }
+   if (rb->Format == MESA_FORMAT_Z16 && !(span->arrayMask & SPAN_XY)) {
+      /* directly read/write row of 16-bit Z values */
+      zBufferVals = zStart;
+      ztest16 = GL_TRUE;
+   }
+   else if (rb->Format == MESA_FORMAT_Z32 && !(span->arrayMask & SPAN_XY)) {
+      /* directly read/write row of 32-bit Z values */
+      zBufferVals = zStart;
    }
    else {
-      /* read depth values from buffer, test, write back */
-      if (rb->DataType == GL_UNSIGNED_SHORT) {
-         GLushort zbuffer[MAX_WIDTH];
-         _swrast_get_values(ctx, rb, count, x, y, zbuffer, sizeof(GLushort));
-         depth_test_span16(ctx, count, zbuffer, z, mask);
-         rb->PutValues(ctx, rb, count, x, y, zbuffer, mask);
+      /* copy Z buffer values into temp buffer (32-bit Z values) */
+      zBufferTemp = (GLuint *) malloc(count * sizeof(GLuint));
+      if (!zBufferTemp)
+         return 0;
+
+      if (span->arrayMask & SPAN_XY) {
+         get_z32_values(ctx, rb, count,
+                        span->array->x, span->array->y, zBufferTemp);
       }
       else {
-         GLuint zbuffer[MAX_WIDTH];
-         ASSERT(rb->DataType == GL_UNSIGNED_INT);
-         _swrast_get_values(ctx, rb, count, x, y, zbuffer, sizeof(GLuint));
-         depth_test_span32(ctx, count, zbuffer, z, mask);
-         rb->PutValues(ctx, rb, count, x, y, zbuffer, mask);
+         _mesa_unpack_uint_z_row(rb->Format, count, zStart, zBufferTemp);
       }
-   }
 
-   return count; /* not really correct, but OK */
-}
+      if (ztest24) {
+         GLuint i;
+         /* Convert depth buffer values from 32 to 24 bits to match the
+          * fragment Z values generated by rasterization.
+          */
+         for (i = 0; i < count; i++) {
+            zBufferTemp[i] >>= 8;
+         }
+      }
 
+      zBufferVals = zBufferTemp;
+   }
 
-/**
- * Apply depth (Z) buffer testing to the span.
- * \return approx number of pixels that passed (only zero is reliable)
- */
-GLuint
-_swrast_depth_test_span( struct gl_context *ctx, SWspan *span)
-{
-   if (span->arrayMask & SPAN_XY)
-      return depth_test_pixels(ctx, span);
+   /* do the depth test either with 16 or 32-bit values */
+   if (ztest16)
+      passed = depth_test_span16(ctx, count, zBufferVals, fragZ, mask);
    else
-      return depth_test_span(ctx, span);
+      passed = depth_test_span32(ctx, count, zBufferVals, fragZ, mask);
+
+   if (zBufferTemp) {
+      /* need to write temp Z values back into the buffer */
+
+      if (ztest24) {
+         GLuint i;
+         /* Convert depth buffer values back to 32-bit values.  The least
+          * significant bits don't matter since they'll get dropped when
+          * they're packed back into the depth buffer.
+          */
+         for (i = 0; i < count; i++) {
+            zBufferTemp[i] = (zBufferTemp[i] << 8);
+         }
+      }
+
+      if (span->arrayMask & SPAN_XY) {
+         /* random locations */
+         put_z32_values(ctx, rb, count, span->array->x, span->array->y,
+                        zBufferTemp, mask);
+      }
+      else {
+         /* horizontal row */
+         gl_pack_uint_z_func packZ = _mesa_get_pack_uint_z_func(rb->Format);
+         GLubyte *dst = zStart;
+         GLuint i;
+         for (i = 0; i < count; i++) {
+            if (mask[i]) {
+               packZ(&zBufferTemp[i], dst);
+            }
+            dst += bpp;
+         }
+      }
+
+      free(zBufferTemp);
+   }
+
+   if (passed < count) {
+      span->writeAll = GL_FALSE;
+   }
+   return passed;
 }
 
 
@@ -1170,67 +382,42 @@ GLboolean
 _swrast_depth_bounds_test( struct gl_context *ctx, SWspan *span )
 {
    struct gl_framebuffer *fb = ctx->DrawBuffer;
-   struct gl_renderbuffer *rb = fb->_DepthBuffer;
+   struct gl_renderbuffer *rb = fb->Attachment[BUFFER_DEPTH].Renderbuffer;
+   const GLint bpp = _mesa_get_format_bytes(rb->Format);
+   const GLint rowStride = rb->RowStride * bpp;
+   GLubyte *zStart = (GLubyte*) rb->Data + span->y * rowStride + span->x * bpp;
    GLuint zMin = (GLuint) (ctx->Depth.BoundsMin * fb->_DepthMaxF + 0.5F);
    GLuint zMax = (GLuint) (ctx->Depth.BoundsMax * fb->_DepthMaxF + 0.5F);
    GLubyte *mask = span->array->mask;
    const GLuint count = span->end;
    GLuint i;
    GLboolean anyPass = GL_FALSE;
+   GLuint zBufferTemp[MAX_WIDTH];
+   const GLuint *zBufferVals;
 
-   if (rb->DataType == GL_UNSIGNED_SHORT) {
-      /* get 16-bit values */
-      GLushort zbuffer16[MAX_WIDTH], *zbuffer;
-      if (span->arrayMask & SPAN_XY) {
-         _swrast_get_values(ctx, rb, count, span->array->x, span->array->y,
-                            zbuffer16, sizeof(GLushort));
-         zbuffer = zbuffer16;
-      }
-      else {
-         zbuffer = (GLushort*) rb->GetPointer(ctx, rb, span->x, span->y);
-         if (!zbuffer) {
-            rb->GetRow(ctx, rb, count, span->x, span->y, zbuffer16);
-            zbuffer = zbuffer16;
-         }
-      }
-      assert(zbuffer);
-
-      /* Now do the tests */
-      for (i = 0; i < count; i++) {
-         if (mask[i]) {
-            if (zbuffer[i] < zMin || zbuffer[i] > zMax)
-               mask[i] = GL_FALSE;
-            else
-               anyPass = GL_TRUE;
-         }
-      }
+   if (rb->Format == MESA_FORMAT_Z32 && !(span->arrayMask & SPAN_XY)) {
+      /* directly access 32-bit values in the depth buffer */
+      zBufferVals = (const GLuint *) zStart;
    }
    else {
-      /* get 32-bit values */
-      GLuint zbuffer32[MAX_WIDTH], *zbuffer;
-      ASSERT(rb->DataType == GL_UNSIGNED_INT);
+      /* unpack Z values into a temporary array */
       if (span->arrayMask & SPAN_XY) {
-         _swrast_get_values(ctx, rb, count, span->array->x, span->array->y,
-                            zbuffer32, sizeof(GLuint));
-         zbuffer = zbuffer32;
+         get_z32_values(ctx, rb, count, span->array->x, span->array->y,
+                        zBufferTemp);
       }
       else {
-         zbuffer = (GLuint*) rb->GetPointer(ctx, rb, span->x, span->y);
-         if (!zbuffer) {
-            rb->GetRow(ctx, rb, count, span->x, span->y, zbuffer32);
-            zbuffer = zbuffer32;
-         }
+         _mesa_unpack_uint_z_row(rb->Format, count, zStart, zBufferTemp);
       }
-      assert(zbuffer);
+      zBufferVals = zBufferTemp;
+   }
 
-      /* Now do the tests */
-      for (i = 0; i < count; i++) {
-         if (mask[i]) {
-            if (zbuffer[i] < zMin || zbuffer[i] > zMax)
-               mask[i] = GL_FALSE;
-            else
-               anyPass = GL_TRUE;
-         }
+   /* Now do the tests */
+   for (i = 0; i < count; i++) {
+      if (mask[i]) {
+         if (zBufferVals[i] < zMin || zBufferVals[i] > zMax)
+            mask[i] = GL_FALSE;
+         else
+            anyPass = GL_TRUE;
       }
    }
 
@@ -1251,19 +438,16 @@ _swrast_depth_bounds_test( struct gl_context *ctx, SWspan *span )
  * bounds.
  */
 void
-_swrast_read_depth_span_float( struct gl_context *ctx, struct gl_renderbuffer *rb,
-                               GLint n, GLint x, GLint y, GLfloat depth[] )
+_swrast_read_depth_span_float(struct gl_context *ctx,
+                              struct gl_renderbuffer *rb,
+                              GLint n, GLint x, GLint y, GLfloat depth[])
 {
-   const GLfloat scale = 1.0F / ctx->DrawBuffer->_DepthMaxF;
-
    if (!rb) {
       /* really only doing this to prevent FP exceptions later */
       memset(depth, 0, n * sizeof(GLfloat));
       return;
    }
 
-   ASSERT(rb->_BaseFormat == GL_DEPTH_COMPONENT);
-
    if (y < 0 || y >= (GLint) rb->Height ||
        x + n <= 0 || x >= (GLint) rb->Width) {
       /* span is completely outside framebuffer */
@@ -1291,35 +475,25 @@ _swrast_read_depth_span_float( struct gl_context *ctx, struct gl_renderbuffer *r
       return;
    }
 
-   if (rb->DataType == GL_UNSIGNED_INT) {
-      GLuint temp[MAX_WIDTH];
-      GLint i;
-      rb->GetRow(ctx, rb, n, x, y, temp);
-      for (i = 0; i < n; i++) {
-         depth[i] = temp[i] * scale;
-      }
-   }
-   else if (rb->DataType == GL_UNSIGNED_SHORT) {
-      GLushort temp[MAX_WIDTH];
-      GLint i;
-      rb->GetRow(ctx, rb, n, x, y, temp);
-      for (i = 0; i < n; i++) {
-         depth[i] = temp[i] * scale;
-      }
-   }
-   else {
-      _mesa_problem(ctx, "Invalid depth renderbuffer data type");
-   }
+   _mesa_unpack_float_z_row(rb->Format, n, _swrast_pixel_address(rb, x, y),
+                            depth);
 }
 
+
 /**
- * Clear the given z/depth renderbuffer.
+ * Clear the given z/depth renderbuffer.  If the buffer is a combined
+ * depth+stencil buffer, only the Z bits will be touched.
  */
 void
-_swrast_clear_depth_buffer( struct gl_context *ctx, struct gl_renderbuffer *rb )
+_swrast_clear_depth_buffer(struct gl_context *ctx)
 {
+   struct gl_renderbuffer *rb =
+      ctx->DrawBuffer->Attachment[BUFFER_DEPTH].Renderbuffer;
    GLuint clearValue;
    GLint x, y, width, height;
+   GLubyte *map;
+   GLint rowStride, i, j;
+   GLbitfield mapMode;
 
    if (!rb || !ctx->Depth.Mask) {
       /* no depth buffer, or writing to it is disabled */
@@ -1334,67 +508,218 @@ _swrast_clear_depth_buffer( struct gl_context *ctx, struct gl_renderbuffer *rb )
       clearValue = (GLuint) (ctx->Depth.Clear * ctx->DrawBuffer->_DepthMaxF);
    }
 
-   assert(rb->_BaseFormat == GL_DEPTH_COMPONENT);
-
    /* compute region to clear */
    x = ctx->DrawBuffer->_Xmin;
    y = ctx->DrawBuffer->_Ymin;
    width  = ctx->DrawBuffer->_Xmax - ctx->DrawBuffer->_Xmin;
    height = ctx->DrawBuffer->_Ymax - ctx->DrawBuffer->_Ymin;
 
-   if (rb->GetPointer(ctx, rb, 0, 0)) {
-      /* Direct buffer access is possible.  Either this is just malloc'd
-       * memory, or perhaps the driver mmap'd the zbuffer memory.
-       */
-      if (rb->DataType == GL_UNSIGNED_SHORT) {
-         if ((clearValue & 0xff) == ((clearValue >> 8) & 0xff) &&
-             ((GLushort *) rb->GetPointer(ctx, rb, 0, 0) + width ==
-              (GLushort *) rb->GetPointer(ctx, rb, 0, 1))) {
-            /* optimized case */
-            GLushort *dst = (GLushort *) rb->GetPointer(ctx, rb, x, y);
-            GLuint len = width * height * sizeof(GLushort);
-            memset(dst, (clearValue & 0xff), len);
+   mapMode = GL_MAP_WRITE_BIT;
+   if (rb->Format == MESA_FORMAT_S8_Z24 ||
+       rb->Format == MESA_FORMAT_X8_Z24 ||
+       rb->Format == MESA_FORMAT_Z24_S8 ||
+       rb->Format == MESA_FORMAT_Z24_X8) {
+      mapMode |= GL_MAP_READ_BIT;
+   }
+
+   ctx->Driver.MapRenderbuffer(ctx, rb, x, y, width, height,
+                               mapMode, &map, &rowStride);
+   if (!map) {
+      _mesa_error(ctx, GL_OUT_OF_MEMORY, "glClear(depth)");
+      return;
+   }
+
+   switch (rb->Format) {
+   case MESA_FORMAT_Z16:
+      {
+         GLfloat clear = (GLfloat) ctx->Depth.Clear;
+         GLushort clearVal = 0;
+         _mesa_pack_float_z_row(rb->Format, 1, &clear, &clearVal);
+         if (clearVal == 0xffff && width * 2 == rowStride) {
+            /* common case */
+            memset(map, 0xff, width * height * 2);
          }
          else {
-            /* general case */
-            GLint i, j;
             for (i = 0; i < height; i++) {
-               GLushort *dst = (GLushort *) rb->GetPointer(ctx, rb, x, y + i);
+               GLushort *row = (GLushort *) map;
                for (j = 0; j < width; j++) {
-                  dst[j] = clearValue;
+                  row[j] = clearVal;
                }
+               map += rowStride;
             }
          }
       }
-      else {
-         GLint i, j;
-         ASSERT(rb->DataType == GL_UNSIGNED_INT);
+      break;
+   case MESA_FORMAT_Z32:
+   case MESA_FORMAT_Z32_FLOAT:
+      {
+         GLfloat clear = (GLfloat) ctx->Depth.Clear;
+         GLuint clearVal = 0;
+         _mesa_pack_float_z_row(rb->Format, 1, &clear, &clearVal);
          for (i = 0; i < height; i++) {
-            GLuint *dst = (GLuint *) rb->GetPointer(ctx, rb, x, y + i);
+            GLuint *row = (GLuint *) map;
             for (j = 0; j < width; j++) {
-               dst[j] = clearValue;
+               row[j] = clearVal;
             }
+            map += rowStride;
          }
       }
-   }
-   else {
-      /* Direct access not possible.  Use PutRow to write new values. */
-      if (rb->DataType == GL_UNSIGNED_SHORT) {
-         GLushort clearVal16 = (GLushort) (clearValue & 0xffff);
-         GLint i;
+      break;
+   case MESA_FORMAT_S8_Z24:
+   case MESA_FORMAT_X8_Z24:
+   case MESA_FORMAT_Z24_S8:
+   case MESA_FORMAT_Z24_X8:
+      {
+         GLfloat clear = (GLfloat) ctx->Depth.Clear;
+         GLuint clearVal = 0;
+         GLuint mask;
+
+         if (rb->Format == MESA_FORMAT_S8_Z24 ||
+             rb->Format == MESA_FORMAT_X8_Z24)
+            mask = 0xff000000;
+         else
+            mask = 0xff;
+
+         _mesa_pack_float_z_row(rb->Format, 1, &clear, &clearVal);
          for (i = 0; i < height; i++) {
-            rb->PutMonoRow(ctx, rb, width, x, y + i, &clearVal16, NULL);
+            GLuint *row = (GLuint *) map;
+            for (j = 0; j < width; j++) {
+               row[j] = (row[j] & mask) | clearVal;
+            }
+            map += rowStride;
          }
+
       }
-      else if (rb->DataType == GL_UNSIGNED_INT) {
-         GLint i;
-         ASSERT(sizeof(clearValue) == sizeof(GLuint));
+      break;
+   case MESA_FORMAT_Z32_FLOAT_X24S8:
+      /* XXX untested */
+      {
+         GLfloat clearVal = (GLfloat) ctx->Depth.Clear;
          for (i = 0; i < height; i++) {
-            rb->PutMonoRow(ctx, rb, width, x, y + i, &clearValue, NULL);
+            GLfloat *row = (GLfloat *) map;
+            for (j = 0; j < width; j++) {
+               row[j * 2] = clearVal;
+            }
+            map += rowStride;
          }
       }
-      else {
-         _mesa_problem(ctx, "bad depth renderbuffer DataType");
+      break;
+   default:
+      _mesa_problem(ctx, "Unexpected depth buffer format %s"
+                    " in _swrast_clear_depth_buffer()",
+                    _mesa_get_format_name(rb->Format));
+   }
+
+   ctx->Driver.UnmapRenderbuffer(ctx, rb);
+}
+
+
+
+
+/**
+ * Clear both depth and stencil values in a combined depth+stencil buffer.
+ */
+void
+_swrast_clear_depth_stencil_buffer(struct gl_context *ctx)
+{
+   const GLubyte stencilBits = ctx->DrawBuffer->Visual.stencilBits;
+   const GLuint writeMask = ctx->Stencil.WriteMask[0];
+   const GLuint stencilMax = (1 << stencilBits) - 1;
+   struct gl_renderbuffer *rb =
+      ctx->DrawBuffer->Attachment[BUFFER_DEPTH].Renderbuffer;
+   GLint x, y, width, height;
+   GLbitfield mapMode;
+   GLubyte *map;
+   GLint rowStride, i, j;
+
+   /* check that we really have a combined depth+stencil buffer */
+   assert(rb == ctx->DrawBuffer->Attachment[BUFFER_STENCIL].Renderbuffer);
+
+   /* compute region to clear */
+   x = ctx->DrawBuffer->_Xmin;
+   y = ctx->DrawBuffer->_Ymin;
+   width  = ctx->DrawBuffer->_Xmax - ctx->DrawBuffer->_Xmin;
+   height = ctx->DrawBuffer->_Ymax - ctx->DrawBuffer->_Ymin;
+
+   mapMode = GL_MAP_WRITE_BIT;
+   if ((writeMask & stencilMax) != stencilMax) {
+      /* need to mask stencil values */
+      mapMode |= GL_MAP_READ_BIT;
+   }
+
+   ctx->Driver.MapRenderbuffer(ctx, rb, x, y, width, height,
+                               mapMode, &map, &rowStride);
+   if (!map) {
+      _mesa_error(ctx, GL_OUT_OF_MEMORY, "glClear(depth+stencil)");
+      return;
+   }
+
+   switch (rb->Format) {
+   case MESA_FORMAT_S8_Z24:
+   case MESA_FORMAT_Z24_S8:
+      {
+         GLfloat zClear = (GLfloat) ctx->Depth.Clear;
+         GLuint clear = 0, mask;
+
+         _mesa_pack_float_z_row(rb->Format, 1, &zClear, &clear);
+
+         if (rb->Format == MESA_FORMAT_S8_Z24) {
+            mask = ((~writeMask) & 0xff) << 24;
+            clear |= (ctx->Stencil.Clear & writeMask & 0xff) << 24;
+         }
+         else {
+            mask = ((~writeMask) & 0xff);
+            clear |= (ctx->Stencil.Clear & writeMask & 0xff);
+         }
+
+         for (i = 0; i < height; i++) {
+            GLuint *row = (GLuint *) map;
+            if (mask != 0x0) {
+               for (j = 0; j < width; j++) {
+                  row[j] = (row[j] & mask) | clear;
+               }
+            }
+            else {
+               for (j = 0; j < width; j++) {
+                  row[j] = clear;
+               }
+            }
+            map += rowStride;
+         }
       }
+      break;
+   case MESA_FORMAT_Z32_FLOAT_X24S8:
+      /* XXX untested */
+      {
+         const GLfloat zClear = (GLfloat) ctx->Depth.Clear;
+         const GLuint sClear = ctx->Stencil.Clear & writeMask;
+         const GLuint sMask = (~writeMask) & 0xff;
+         for (i = 0; i < height; i++) {
+            GLfloat *zRow = (GLfloat *) map;
+            GLuint *sRow = (GLuint *) map;
+            for (j = 0; j < width; j++) {
+               zRow[j * 2 + 0] = zClear;
+            }
+            if (sMask != 0) {
+               for (j = 0; j < width; j++) {
+                  sRow[j * 2 + 1] = (sRow[j * 2 + 1] & sMask) | sClear;
+               }
+            }
+            else {
+               for (j = 0; j < width; j++) {
+                  sRow[j * 2 + 1] = sClear;
+               }
+            }
+            map += rowStride;
+         }
+      }
+      break;
+   default:
+      _mesa_problem(ctx, "Unexpected depth buffer format %s"
+                    " in _swrast_clear_depth_buffer()",
+                    _mesa_get_format_name(rb->Format));
    }
+
+   ctx->Driver.UnmapRenderbuffer(ctx, rb);
+
 }
diff --git a/mesalib/src/mesa/swrast/s_depth.h b/mesalib/src/mesa/swrast/s_depth.h
index 2caea82e0..25a7a0bbf 100644
--- a/mesalib/src/mesa/swrast/s_depth.h
+++ b/mesalib/src/mesa/swrast/s_depth.h
@@ -49,7 +49,10 @@ _swrast_read_depth_span_float( struct gl_context *ctx, struct gl_renderbuffer *r
                                GLint n, GLint x, GLint y, GLfloat depth[] );
 
 extern void
-_swrast_clear_depth_buffer( struct gl_context *ctx, struct gl_renderbuffer *rb );
+_swrast_clear_depth_buffer(struct gl_context *ctx);
+
+extern void
+_swrast_clear_depth_stencil_buffer(struct gl_context *ctx);
 
 
 #endif
diff --git a/mesalib/src/mesa/swrast/s_depthstencil.c b/mesalib/src/mesa/swrast/s_depthstencil.c
index dc768e216..e48177ec1 100644
--- a/mesalib/src/mesa/swrast/s_depthstencil.c
+++ b/mesalib/src/mesa/swrast/s_depthstencil.c
@@ -207,59 +207,6 @@ put_row_z24(struct gl_context *ctx, struct gl_renderbuffer *z24rb, GLuint count,
    }
 }
 
-static void
-put_mono_row_z24(struct gl_context *ctx, struct gl_renderbuffer *z24rb, GLuint count,
-                 GLint x, GLint y, const void *value, const GLubyte *mask)
-{
-   struct gl_renderbuffer *dsrb = z24rb->Wrapped;
-   GLuint *dst = (GLuint *) dsrb->GetPointer(ctx, dsrb, x, y);
-   ASSERT(z24rb->DataType == GL_UNSIGNED_INT);
-   ASSERT(dsrb->DataType == GL_UNSIGNED_INT_24_8_EXT);
-   if (dst) {
-      /* direct access */
-      GLuint i;
-      if (dsrb->Format == MESA_FORMAT_Z24_S8) {
-         const GLuint shiftedVal = *((GLuint *) value) << 8;
-         for (i = 0; i < count; i++) {
-            if (!mask || mask[i]) {
-               dst[i] = shiftedVal | (dst[i] & 0xff);
-            }
-         }
-      }
-      else {
-         const GLuint shiftedVal = *((GLuint *) value);
-         assert(dsrb->Format == MESA_FORMAT_S8_Z24);
-         for (i = 0; i < count; i++) {
-            if (!mask || mask[i]) {
-               dst[i] = shiftedVal | (dst[i] & 0xff000000);
-            }
-         }
-      }
-   }
-   else {
-      /* get, modify, put */
-      GLuint temp[MAX_WIDTH], i;
-      dsrb->GetRow(ctx, dsrb, count, x, y, temp);
-      if (dsrb->Format == MESA_FORMAT_Z24_S8) {
-         const GLuint shiftedVal = *((GLuint *) value) << 8;
-         for (i = 0; i < count; i++) {
-            if (!mask || mask[i]) {
-               temp[i] = shiftedVal | (temp[i] & 0xff);
-            }
-         }
-      }
-      else {
-         const GLuint shiftedVal = *((GLuint *) value);
-         assert(dsrb->Format == MESA_FORMAT_S8_Z24);
-         for (i = 0; i < count; i++) {
-            if (!mask || mask[i]) {
-               temp[i] = shiftedVal | (temp[i] & 0xff000000);
-            }
-         }
-      }
-      dsrb->PutRow(ctx, dsrb, count, x, y, temp, mask);
-   }
-}
 
 static void
 put_values_z24(struct gl_context *ctx, struct gl_renderbuffer *z24rb, GLuint count,
@@ -314,35 +261,6 @@ put_values_z24(struct gl_context *ctx, struct gl_renderbuffer *z24rb, GLuint cou
    }
 }
 
-static void
-put_mono_values_z24(struct gl_context *ctx, struct gl_renderbuffer *z24rb,
-                    GLuint count, const GLint x[], const GLint y[],
-                    const void *value, const GLubyte *mask)
-{
-   struct gl_renderbuffer *dsrb = z24rb->Wrapped;
-   GLuint temp[MAX_WIDTH], i;
-   /* get, modify, put */
-   dsrb->GetValues(ctx, dsrb, count, x, y, temp);
-   if (dsrb->Format == MESA_FORMAT_Z24_S8) {
-      const GLuint shiftedVal = *((GLuint *) value) << 8;
-      for (i = 0; i < count; i++) {
-         if (!mask || mask[i]) {
-            temp[i] = shiftedVal | (temp[i] & 0xff);
-         }
-      }
-   }
-   else {
-      const GLuint shiftedVal = *((GLuint *) value);
-      assert(dsrb->Format == MESA_FORMAT_S8_Z24);
-      for (i = 0; i < count; i++) {
-         if (!mask || mask[i]) {
-            temp[i] = shiftedVal | (temp[i] & 0xff000000);
-         }
-      }
-   }
-   dsrb->PutValues(ctx, dsrb, count, x, y, temp, mask);
-}
-
 
 /**
  * Wrap the given GL_DEPTH_STENCIL renderbuffer so that it acts like
@@ -385,10 +303,7 @@ new_z24_renderbuffer_wrapper(struct gl_context *ctx,
    z24rb->GetRow = get_row_z24;
    z24rb->GetValues = get_values_z24;
    z24rb->PutRow = put_row_z24;
-   z24rb->PutRowRGB = NULL;
-   z24rb->PutMonoRow = put_mono_row_z24;
    z24rb->PutValues = put_values_z24;
-   z24rb->PutMonoValues = put_mono_values_z24;
 
    return z24rb;
 }
@@ -467,39 +382,6 @@ put_row_z32f(struct gl_context *ctx, struct gl_renderbuffer *z32frb, GLuint coun
    }
 }
 
-static void
-put_mono_row_z32f(struct gl_context *ctx, struct gl_renderbuffer *z32frb, GLuint count,
-                  GLint x, GLint y, const void *value, const GLubyte *mask)
-{
-   struct gl_renderbuffer *dsrb = z32frb->Wrapped;
-   GLfloat *dst = (GLfloat *) dsrb->GetPointer(ctx, dsrb, x, y);
-   ASSERT(z32frb->DataType == GL_FLOAT);
-   ASSERT(dsrb->DataType == GL_FLOAT_32_UNSIGNED_INT_24_8_REV);
-   ASSERT(dsrb->Format == MESA_FORMAT_Z32_FLOAT_X24S8);
-   if (dst) {
-      /* direct access */
-      GLuint i;
-      const GLfloat val = *(GLfloat*)value;
-      for (i = 0; i < count; i++) {
-         if (!mask || mask[i]) {
-            dst[i*2] = val;
-         }
-      }
-   }
-   else {
-      /* get, modify, put */
-      GLfloat temp[MAX_WIDTH*2];
-      GLuint i;
-      const GLfloat val = *(GLfloat *)value;
-      dsrb->GetRow(ctx, dsrb, count, x, y, temp);
-      for (i = 0; i < count; i++) {
-         if (!mask || mask[i]) {
-            temp[i*2] = val;
-         }
-      }
-      dsrb->PutRow(ctx, dsrb, count, x, y, temp, mask);
-   }
-}
 
 static void
 put_values_z32f(struct gl_context *ctx, struct gl_renderbuffer *z32frb, GLuint count,
@@ -535,28 +417,6 @@ put_values_z32f(struct gl_context *ctx, struct gl_renderbuffer *z32frb, GLuint c
    }
 }
 
-static void
-put_mono_values_z32f(struct gl_context *ctx, struct gl_renderbuffer *z32frb,
-                     GLuint count, const GLint x[], const GLint y[],
-                     const void *value, const GLubyte *mask)
-{
-   struct gl_renderbuffer *dsrb = z32frb->Wrapped;
-   GLfloat temp[MAX_WIDTH*2];
-   GLuint i;
-   const GLfloat val = *(GLfloat *)value;
-   ASSERT(z32frb->DataType == GL_FLOAT);
-   ASSERT(dsrb->DataType == GL_FLOAT_32_UNSIGNED_INT_24_8_REV);
-   ASSERT(dsrb->Format == MESA_FORMAT_Z32_FLOAT_X24S8);
-   /* get, modify, put */
-   dsrb->GetValues(ctx, dsrb, count, x, y, temp);
-   for (i = 0; i < count; i++) {
-      if (!mask || mask[i]) {
-         temp[i*2] = val;
-      }
-   }
-   dsrb->PutValues(ctx, dsrb, count, x, y, temp, mask);
-}
-
 
 /**
  * Wrap the given GL_DEPTH_STENCIL renderbuffer so that it acts like
@@ -596,10 +456,7 @@ new_z32f_renderbuffer_wrapper(struct gl_context *ctx,
    z32frb->GetRow = get_row_z32f;
    z32frb->GetValues = get_values_z32f;
    z32frb->PutRow = put_row_z32f;
-   z32frb->PutRowRGB = NULL;
-   z32frb->PutMonoRow = put_mono_row_z32f;
    z32frb->PutValues = put_values_z32f;
-   z32frb->PutMonoValues = put_mono_values_z32f;
 
    return z32frb;
 }
@@ -739,71 +596,6 @@ put_row_s8(struct gl_context *ctx, struct gl_renderbuffer *s8rb, GLuint count,
    }
 }
 
-static void
-put_mono_row_s8(struct gl_context *ctx, struct gl_renderbuffer *s8rb, GLuint count,
-                GLint x, GLint y, const void *value, const GLubyte *mask)
-{
-   struct gl_renderbuffer *dsrb = s8rb->Wrapped;
-   const GLubyte val = *((GLubyte *) value);
-   GLuint *dst = (GLuint *) dsrb->GetPointer(ctx, dsrb, x, y);
-   ASSERT(s8rb->DataType == GL_UNSIGNED_BYTE);
-   ASSERT(dsrb->DataType == GL_UNSIGNED_INT_24_8_EXT ||
-          dsrb->DataType == GL_FLOAT_32_UNSIGNED_INT_24_8_REV);
-   if (dst) {
-      /* direct access */
-      GLuint i;
-      if (dsrb->Format == MESA_FORMAT_Z32_FLOAT_X24S8) {
-         for (i = 0; i < count; i++) {
-            if (!mask || mask[i]) {
-               dst[i*2+1] = val;
-            }
-         }
-      }
-      else if (dsrb->Format == MESA_FORMAT_Z24_S8) {
-         for (i = 0; i < count; i++) {
-            if (!mask || mask[i]) {
-               dst[i] = (dst[i] & 0xffffff00) | val;
-            }
-         }
-      }
-      else {
-         assert(dsrb->Format == MESA_FORMAT_S8_Z24);
-         for (i = 0; i < count; i++) {
-            if (!mask || mask[i]) {
-               dst[i] = (dst[i] & 0xffffff) | (val << 24);
-            }
-         }
-      }
-   }
-   else {
-      /* get, modify, put */
-      GLuint temp[MAX_WIDTH*2], i;
-      dsrb->GetRow(ctx, dsrb, count, x, y, temp);
-      if (dsrb->Format == MESA_FORMAT_Z32_FLOAT_X24S8) {
-         for (i = 0; i < count; i++) {
-            if (!mask || mask[i]) {
-               temp[i*2+1] = val;
-            }
-         }
-      }
-      else if (dsrb->Format == MESA_FORMAT_Z24_S8) {
-         for (i = 0; i < count; i++) {
-            if (!mask || mask[i]) {
-               temp[i] = (temp[i] & 0xffffff00) | val;
-            }
-         }
-      }
-      else {
-         assert(dsrb->Format == MESA_FORMAT_S8_Z24);
-         for (i = 0; i < count; i++) {
-            if (!mask || mask[i]) {
-               temp[i] = (temp[i] & 0xffffff) | (val << 24);
-            }
-         }
-      }
-      dsrb->PutRow(ctx, dsrb, count, x, y, temp, mask);
-   }
-}
 
 static void
 put_values_s8(struct gl_context *ctx, struct gl_renderbuffer *s8rb, GLuint count,
@@ -874,41 +666,6 @@ put_values_s8(struct gl_context *ctx, struct gl_renderbuffer *s8rb, GLuint count
    }
 }
 
-static void
-put_mono_values_s8(struct gl_context *ctx, struct gl_renderbuffer *s8rb, GLuint count,
-                   const GLint x[], const GLint y[],
-                   const void *value, const GLubyte *mask)
-{
-   struct gl_renderbuffer *dsrb = s8rb->Wrapped;
-   GLuint temp[MAX_WIDTH*2], i;
-   const GLubyte val = *((GLubyte *) value);
-   /* get, modify, put */
-   dsrb->GetValues(ctx, dsrb, count, x, y, temp);
-   if (dsrb->Format == MESA_FORMAT_Z32_FLOAT_X24S8) {
-      for (i = 0; i < count; i++) {
-         if (!mask || mask[i]) {
-            temp[i*2+1] = val;
-         }
-      }
-   }
-   else if (dsrb->Format == MESA_FORMAT_Z24_S8) {
-      for (i = 0; i < count; i++) {
-         if (!mask || mask[i]) {
-            temp[i] = (temp[i] & 0xffffff00) | val;
-         }
-      }
-   }
-   else {
-      assert(dsrb->Format == MESA_FORMAT_S8_Z24);
-      for (i = 0; i < count; i++) {
-         if (!mask || mask[i]) {
-            temp[i] = (temp[i] & 0xffffff) | (val << 24);
-         }
-      }
-   }
-   dsrb->PutValues(ctx, dsrb, count, x, y, temp, mask);
-}
-
 
 /**
  * Wrap the given GL_DEPTH_STENCIL renderbuffer so that it acts like
@@ -950,10 +707,7 @@ new_s8_renderbuffer_wrapper(struct gl_context *ctx, struct gl_renderbuffer *dsrb
    s8rb->GetRow = get_row_s8;
    s8rb->GetValues = get_values_s8;
    s8rb->PutRow = put_row_s8;
-   s8rb->PutRowRGB = NULL;
-   s8rb->PutMonoRow = put_mono_row_s8;
    s8rb->PutValues = put_values_s8;
-   s8rb->PutMonoValues = put_mono_values_s8;
 
    return s8rb;
 }
diff --git a/mesalib/src/mesa/swrast/s_drawpix.c b/mesalib/src/mesa/swrast/s_drawpix.c
index 7259881c1..4231bb4bd 100644
--- a/mesalib/src/mesa/swrast/s_drawpix.c
+++ b/mesalib/src/mesa/swrast/s_drawpix.c
@@ -25,8 +25,10 @@
 
 #include "main/glheader.h"
 #include "main/bufferobj.h"
+#include "main/colormac.h"
 #include "main/condrender.h"
 #include "main/context.h"
+#include "main/format_pack.h"
 #include "main/image.h"
 #include "main/imports.h"
 #include "main/macros.h"
@@ -41,6 +43,146 @@
 #include "s_zoom.h"
 
 
+/**
+ * Handle a common case of drawing GL_RGB/GL_UNSIGNED_BYTE into a
+ * MESA_FORMAT_XRGB888 or MESA_FORMAT_ARGB888 renderbuffer.
+ */
+static void
+fast_draw_rgb_ubyte_pixels(struct gl_context *ctx,
+                           struct gl_renderbuffer *rb,
+                           GLint x, GLint y,
+                           GLsizei width, GLsizei height,
+                           const struct gl_pixelstore_attrib *unpack,
+                           const GLvoid *pixels)
+{
+   const GLubyte *src = (const GLubyte *)
+      _mesa_image_address2d(unpack, pixels, width,
+                            height, GL_RGB, GL_UNSIGNED_BYTE, 0, 0);
+   const GLint srcRowStride = _mesa_image_row_stride(unpack, width,
+                                                     GL_RGB, GL_UNSIGNED_BYTE);
+   GLint i, j;
+   GLubyte *dst;
+   GLint dstRowStride;
+
+   ctx->Driver.MapRenderbuffer(ctx, rb, x, y, width, height,
+                               GL_MAP_WRITE_BIT, &dst, &dstRowStride);
+
+   if (!dst) {
+      _mesa_error(ctx, GL_OUT_OF_MEMORY, "glDrawPixels");
+      return;
+   }
+
+   if (ctx->Pixel.ZoomY == -1.0f) {
+      dst = dst + (height - 1) * dstRowStride;
+      dstRowStride = -dstRowStride;
+   }
+
+   for (i = 0; i < height; i++) {
+      GLuint *dst4 = (GLuint *) dst;
+      for (j = 0; j < width; j++) {
+         dst4[j] = PACK_COLOR_8888(0xff, src[j*3+0], src[j*3+1], src[j*3+2]);
+      }
+      dst += dstRowStride;
+      src += srcRowStride;
+   }
+
+   ctx->Driver.UnmapRenderbuffer(ctx, rb);
+}
+
+
+/**
+ * Handle a common case of drawing GL_RGBA/GL_UNSIGNED_BYTE into a
+ * MESA_FORMAT_ARGB888 or MESA_FORMAT_xRGB888 renderbuffer.
+ */
+static void
+fast_draw_rgba_ubyte_pixels(struct gl_context *ctx,
+                           struct gl_renderbuffer *rb,
+                           GLint x, GLint y,
+                           GLsizei width, GLsizei height,
+                           const struct gl_pixelstore_attrib *unpack,
+                           const GLvoid *pixels)
+{
+   const GLubyte *src = (const GLubyte *)
+      _mesa_image_address2d(unpack, pixels, width,
+                            height, GL_RGBA, GL_UNSIGNED_BYTE, 0, 0);
+   const GLint srcRowStride =
+      _mesa_image_row_stride(unpack, width, GL_RGBA, GL_UNSIGNED_BYTE);
+   GLint i, j;
+   GLubyte *dst;
+   GLint dstRowStride;
+
+   ctx->Driver.MapRenderbuffer(ctx, rb, x, y, width, height,
+                               GL_MAP_WRITE_BIT, &dst, &dstRowStride);
+
+   if (!dst) {
+      _mesa_error(ctx, GL_OUT_OF_MEMORY, "glDrawPixels");
+      return;
+   }
+
+   if (ctx->Pixel.ZoomY == -1.0f) {
+      dst = dst + (height - 1) * dstRowStride;
+      dstRowStride = -dstRowStride;
+   }
+
+   for (i = 0; i < height; i++) {
+      GLuint *dst4 = (GLuint *) dst;
+      for (j = 0; j < width; j++) {
+         dst4[j] = PACK_COLOR_8888(src[j*4+3], src[j*4+0],
+                                   src[j*4+1], src[j*4+2]);
+      }
+      dst += dstRowStride;
+      src += srcRowStride;
+   }
+
+   ctx->Driver.UnmapRenderbuffer(ctx, rb);
+}
+
+
+/**
+ * Handle a common case of drawing a format/type combination that
+ * exactly matches the renderbuffer format.
+ */
+static void
+fast_draw_generic_pixels(struct gl_context *ctx,
+                         struct gl_renderbuffer *rb,
+                         GLint x, GLint y,
+                         GLsizei width, GLsizei height,
+                         GLenum format, GLenum type,
+                         const struct gl_pixelstore_attrib *unpack,
+                         const GLvoid *pixels)
+{
+   const GLubyte *src = (const GLubyte *)
+      _mesa_image_address2d(unpack, pixels, width,
+                            height, format, type, 0, 0);
+   const GLint srcRowStride =
+      _mesa_image_row_stride(unpack, width, format, type);
+   const GLint rowLength = width * _mesa_get_format_bytes(rb->Format);
+   GLint i;
+   GLubyte *dst;
+   GLint dstRowStride;
+
+   ctx->Driver.MapRenderbuffer(ctx, rb, x, y, width, height,
+                               GL_MAP_WRITE_BIT, &dst, &dstRowStride);
+
+   if (!dst) {
+      _mesa_error(ctx, GL_OUT_OF_MEMORY, "glDrawPixels");
+      return;
+   }
+
+   if (ctx->Pixel.ZoomY == -1.0f) {
+      dst = dst + (height - 1) * dstRowStride;
+      dstRowStride = -dstRowStride;
+   }
+
+   for (i = 0; i < height; i++) {
+      memcpy(dst, src, rowLength);
+      dst += dstRowStride;
+      src += srcRowStride;
+   }
+
+   ctx->Driver.UnmapRenderbuffer(ctx, rb);
+}
+
 
 /**
  * Try to do a fast and simple RGB(a) glDrawPixels.
@@ -53,256 +195,56 @@ fast_draw_rgba_pixels(struct gl_context *ctx, GLint x, GLint y,
                       const struct gl_pixelstore_attrib *userUnpack,
                       const GLvoid *pixels)
 {
-   const GLint imgX = x, imgY = y;
    struct gl_renderbuffer *rb = ctx->DrawBuffer->_ColorDrawBuffers[0];
-   GLenum rbType;
    SWcontext *swrast = SWRAST_CONTEXT(ctx);
-   SWspan span;
-   GLboolean simpleZoom;
-   GLint yStep;  /* +1 or -1 */
    struct gl_pixelstore_attrib unpack;
-   GLint destX, destY, drawWidth, drawHeight; /* post clipping */
 
    if (!rb)
       return GL_TRUE; /* no-op */
 
-   rbType = rb->DataType;
-
-   if ((swrast->_RasterMask & ~CLIP_BIT) ||
+   if (ctx->DrawBuffer->_NumColorDrawBuffers > 1 ||
+       (swrast->_RasterMask & ~CLIP_BIT) ||
        ctx->Texture._EnabledCoordUnits ||
        userUnpack->SwapBytes ||
+       ctx->Pixel.ZoomX != 1.0f ||
+       fabsf(ctx->Pixel.ZoomY) != 1.0f ||
        ctx->_ImageTransferState) {
       /* can't handle any of those conditions */
       return GL_FALSE;
    }
 
-   INIT_SPAN(span, GL_BITMAP);
-   span.arrayMask = SPAN_RGBA;
-   span.arrayAttribs = FRAG_BIT_COL0;
-   _swrast_span_default_attribs(ctx, &span);
-
-   /* copy input params since clipping may change them */
    unpack = *userUnpack;
-   destX = x;
-   destY = y;
-   drawWidth = width;
-   drawHeight = height;
-
-   /* check for simple zooming and clipping */
-   if (ctx->Pixel.ZoomX == 1.0F &&
-       (ctx->Pixel.ZoomY == 1.0F || ctx->Pixel.ZoomY == -1.0F)) {
-      if (!_mesa_clip_drawpixels(ctx, &destX, &destY,
-                                 &drawWidth, &drawHeight, &unpack)) {
-         /* image was completely clipped: no-op, all done */
-         return GL_TRUE;
-      }
-      simpleZoom = GL_TRUE;
-      yStep = (GLint) ctx->Pixel.ZoomY;
-      ASSERT(yStep == 1 || yStep == -1);
-   }
-   else {
-      /* non-simple zooming */
-      simpleZoom = GL_FALSE;
-      yStep = 1;
-      if (unpack.RowLength == 0)
-         unpack.RowLength = width;
-   }
 
-   /*
-    * Ready to draw!
-    */
-
-   if (format == GL_RGBA && type == rbType) {
-      const GLubyte *src
-         = (const GLubyte *) _mesa_image_address2d(&unpack, pixels, width,
-                                                   height, format, type, 0, 0);
-      const GLint srcStride = _mesa_image_row_stride(&unpack, width,
-                                                     format, type);
-      if (simpleZoom) {
-         GLint row;
-         for (row = 0; row < drawHeight; row++) {
-            rb->PutRow(ctx, rb, drawWidth, destX, destY, src, NULL);
-            src += srcStride;
-            destY += yStep;
-         }
-      }
-      else {
-         /* with zooming */
-         GLint row;
-         for (row = 0; row < drawHeight; row++) {
-            span.x = destX;
-            span.y = destY + row;
-            span.end = drawWidth;
-            span.array->ChanType = rbType;
-            _swrast_write_zoomed_rgba_span(ctx, imgX, imgY, &span, src);
-            src += srcStride;
-         }
-         span.array->ChanType = CHAN_TYPE;
-      }
+   /* clipping */
+   if (!_mesa_clip_drawpixels(ctx, &x, &y, &width, &height, &unpack)) {
+      /* image was completely clipped: no-op, all done */
       return GL_TRUE;
    }
 
-   if (format == GL_RGB && type == rbType) {
-      const GLubyte *src
-         = (const GLubyte *) _mesa_image_address2d(&unpack, pixels, width,
-                                                   height, format, type, 0, 0);
-      const GLint srcStride = _mesa_image_row_stride(&unpack, width,
-                                                     format, type);
-      if (simpleZoom) {
-         GLint row;
-         for (row = 0; row < drawHeight; row++) {
-            rb->PutRowRGB(ctx, rb, drawWidth, destX, destY, src, NULL);
-            src += srcStride;
-            destY += yStep;
-         }
-      }
-      else {
-         /* with zooming */
-         GLint row;
-         for (row = 0; row < drawHeight; row++) {
-            span.x = destX;
-            span.y = destY;
-            span.end = drawWidth;
-            span.array->ChanType = rbType;
-            _swrast_write_zoomed_rgb_span(ctx, imgX, imgY, &span, src);
-            src += srcStride;
-            destY++;
-         }
-         span.array->ChanType = CHAN_TYPE;
-      }
+   if (format == GL_RGB &&
+       type == GL_UNSIGNED_BYTE &&
+       (rb->Format == MESA_FORMAT_XRGB8888 ||
+        rb->Format == MESA_FORMAT_ARGB8888)) {
+      fast_draw_rgb_ubyte_pixels(ctx, rb, x, y, width, height,
+                                 &unpack, pixels);
       return GL_TRUE;
    }
 
-   /* Remaining cases haven't been tested with alignment != 1 */
-   if (userUnpack->Alignment != 1)
-      return GL_FALSE;
-
-   if (format == GL_LUMINANCE && type == CHAN_TYPE && rbType == CHAN_TYPE) {
-      const GLchan *src = (const GLchan *) pixels
-         + (unpack.SkipRows * unpack.RowLength + unpack.SkipPixels);
-      if (simpleZoom) {
-         /* no zooming */
-         GLint row;
-         ASSERT(drawWidth <= MAX_WIDTH);
-         for (row = 0; row < drawHeight; row++) {
-            GLchan rgb[MAX_WIDTH][3];
-            GLint i;
-            for (i = 0;i<drawWidth;i++) {
-               rgb[i][0] = src[i];
-               rgb[i][1] = src[i];
-               rgb[i][2] = src[i];
-            }
-            rb->PutRowRGB(ctx, rb, drawWidth, destX, destY, rgb, NULL);
-            src += unpack.RowLength;
-            destY += yStep;
-         }
-      }
-      else {
-         /* with zooming */
-         GLint row;
-         ASSERT(drawWidth <= MAX_WIDTH);
-         for (row = 0; row < drawHeight; row++) {
-            GLchan rgb[MAX_WIDTH][3];
-            GLint i;
-            for (i = 0;i<drawWidth;i++) {
-               rgb[i][0] = src[i];
-               rgb[i][1] = src[i];
-               rgb[i][2] = src[i];
-            }
-            span.x = destX;
-            span.y = destY;
-            span.end = drawWidth;
-            _swrast_write_zoomed_rgb_span(ctx, imgX, imgY, &span, rgb);
-            src += unpack.RowLength;
-            destY++;
-         }
-      }
+   if (format == GL_RGBA &&
+       type == GL_UNSIGNED_BYTE &&
+       (rb->Format == MESA_FORMAT_XRGB8888 ||
+        rb->Format == MESA_FORMAT_ARGB8888)) {
+      fast_draw_rgba_ubyte_pixels(ctx, rb, x, y, width, height,
+                                  &unpack, pixels);
       return GL_TRUE;
    }
 
-   if (format == GL_LUMINANCE_ALPHA && type == CHAN_TYPE && rbType == CHAN_TYPE) {
-      const GLchan *src = (const GLchan *) pixels
-         + (unpack.SkipRows * unpack.RowLength + unpack.SkipPixels)*2;
-      if (simpleZoom) {
-         GLint row;
-         ASSERT(drawWidth <= MAX_WIDTH);
-         for (row = 0; row < drawHeight; row++) {
-            GLint i;
-            const GLchan *ptr = src;
-            for (i = 0;i<drawWidth;i++) {
-               span.array->rgba[i][0] = *ptr;
-               span.array->rgba[i][1] = *ptr;
-               span.array->rgba[i][2] = *ptr++;
-               span.array->rgba[i][3] = *ptr++;
-            }
-            rb->PutRow(ctx, rb, drawWidth, destX, destY,
-                       span.array->rgba, NULL);
-            src += unpack.RowLength*2;
-            destY += yStep;
-         }
-      }
-      else {
-         /* with zooming */
-         GLint row;
-         ASSERT(drawWidth <= MAX_WIDTH);
-         for (row = 0; row < drawHeight; row++) {
-            const GLchan *ptr = src;
-            GLint i;
-            for (i = 0;i<drawWidth;i++) {
-               span.array->rgba[i][0] = *ptr;
-               span.array->rgba[i][1] = *ptr;
-               span.array->rgba[i][2] = *ptr++;
-               span.array->rgba[i][3] = *ptr++;
-            }
-            span.x = destX;
-            span.y = destY;
-            span.end = drawWidth;
-            _swrast_write_zoomed_rgba_span(ctx, imgX, imgY, &span,
-                                           span.array->rgba);
-            src += unpack.RowLength*2;
-            destY++;
-         }
-      }
+   if (_mesa_format_matches_format_and_type(rb->Format, format, type)) {
+      fast_draw_generic_pixels(ctx, rb, x, y, width, height,
+                               format, type, &unpack, pixels);
       return GL_TRUE;
    }
 
-   if (format == GL_COLOR_INDEX && type == GL_UNSIGNED_BYTE) {
-      const GLubyte *src = (const GLubyte *) pixels
-         + unpack.SkipRows * unpack.RowLength + unpack.SkipPixels;
-      if (rbType == GL_UNSIGNED_BYTE) {
-         /* convert ubyte/CI data to ubyte/RGBA */
-         if (simpleZoom) {
-            GLint row;
-            for (row = 0; row < drawHeight; row++) {
-               ASSERT(drawWidth <= MAX_WIDTH);
-               _mesa_map_ci8_to_rgba8(ctx, drawWidth, src,
-                                      span.array->rgba8);
-               rb->PutRow(ctx, rb, drawWidth, destX, destY,
-                          span.array->rgba8, NULL);
-               src += unpack.RowLength;
-               destY += yStep;
-            }
-         }
-         else {
-            /* ubyte/CI to ubyte/RGBA with zooming */
-            GLint row;
-            for (row = 0; row < drawHeight; row++) {
-               ASSERT(drawWidth <= MAX_WIDTH);
-               _mesa_map_ci8_to_rgba8(ctx, drawWidth, src,
-                                      span.array->rgba8);
-               span.x = destX;
-               span.y = destY;
-               span.end = drawWidth;
-               _swrast_write_zoomed_rgba_span(ctx, imgX, imgY, &span,
-                                              span.array->rgba8);
-               src += unpack.RowLength;
-               destY++;
-            }
-         }
-         return GL_TRUE;
-      }
-   }
-
    /* can't handle this pixel format and/or data type */
    return GL_FALSE;
 }
@@ -480,6 +422,8 @@ draw_rgba_pixels( struct gl_context *ctx, GLint x, GLint y,
       return;
    }
 
+   swrast_render_start(ctx);
+
    INIT_SPAN(span, GL_BITMAP);
    _swrast_span_default_attribs(ctx, &span);
    span.arrayMask = SPAN_RGBA;
@@ -547,9 +491,46 @@ draw_rgba_pixels( struct gl_context *ctx, GLint x, GLint y,
    if (convImage) {
       free(convImage);
    }
+
+   swrast_render_finish(ctx);
+}
+
+
+/**
+ * Draw depth+stencil values into a MESA_FORAMT_Z24_S8 or MESA_FORMAT_S8_Z24
+ * renderbuffer.  No masking, zooming, scaling, etc.
+ */
+static void
+fast_draw_depth_stencil(struct gl_context *ctx, GLint x, GLint y,
+                        GLsizei width, GLsizei height,
+                        const struct gl_pixelstore_attrib *unpack,
+                        const GLvoid *pixels)
+{
+   const GLenum format = GL_DEPTH_STENCIL_EXT;
+   const GLenum type = GL_UNSIGNED_INT_24_8;
+   struct gl_renderbuffer *rb =
+      ctx->DrawBuffer->Attachment[BUFFER_DEPTH].Renderbuffer;
+   GLubyte *src, *dst;
+   GLint srcRowStride, dstRowStride;
+   GLint i;
+
+   src = _mesa_image_address2d(unpack, pixels, width, height,
+                               format, type, 0, 0);
+   srcRowStride = _mesa_image_row_stride(unpack, width, format, type);
+
+   dst = _swrast_pixel_address(rb, x, y);
+   dstRowStride = rb->RowStride * 4;
+
+   for (i = 0; i < height; i++) {
+      _mesa_pack_uint_24_8_depth_stencil_row(rb->Format, width,
+                                             (const GLuint *) src, dst);
+      dst += dstRowStride;
+      src += srcRowStride;
+   }
 }
 
 
+
 /**
  * This is a bit different from drawing GL_DEPTH_COMPONENT pixels.
  * The only per-pixel operations that apply are depth scale/bias,
@@ -567,7 +548,6 @@ draw_depth_stencil_pixels(struct gl_context *ctx, GLint x, GLint y,
    const GLint imgX = x, imgY = y;
    const GLboolean scaleOrBias
       = ctx->Pixel.DepthScale != 1.0 || ctx->Pixel.DepthBias != 0.0;
-   const GLuint depthMax = ctx->DrawBuffer->_DepthMax;
    const GLuint stencilMask = ctx->Stencil.WriteMask[0];
    const GLenum stencilType = GL_UNSIGNED_BYTE;
    const GLboolean zoom = ctx->Pixel.ZoomX != 1.0 || ctx->Pixel.ZoomY != 1.0;
@@ -587,27 +567,16 @@ draw_depth_stencil_pixels(struct gl_context *ctx, GLint x, GLint y,
    ASSERT(depthRb);
    ASSERT(stencilRb);
 
-   if (depthRb->_BaseFormat == GL_DEPTH_STENCIL_EXT &&
-       depthRb->Format == MESA_FORMAT_Z24_S8 &&
+   if (depthRb == stencilRb &&
+       (depthRb->Format == MESA_FORMAT_Z24_S8 ||
+        depthRb->Format == MESA_FORMAT_S8_Z24) &&
        type == GL_UNSIGNED_INT_24_8 &&
-       depthRb == stencilRb &&
-       depthRb->GetRow &&  /* May be null if depthRb is a wrapper around
-			    * separate depth and stencil buffers. */
        !scaleOrBias &&
        !zoom &&
        ctx->Depth.Mask &&
        (stencilMask & 0xff) == 0xff) {
-      /* This is the ideal case.
-       * Drawing GL_DEPTH_STENCIL pixels into a combined depth/stencil buffer.
-       * Plus, no pixel transfer ops, zooming, or masking needed.
-       */
-      GLint i;
-      for (i = 0; i < height; i++) {
-         const GLuint *src = (const GLuint *) 
-            _mesa_image_address2d(&clippedUnpack, pixels, width, height,
-                                  GL_DEPTH_STENCIL_EXT, type, i, 0);
-         depthRb->PutRow(ctx, depthRb, width, x, y + i, src, NULL);
-      }
+      fast_draw_depth_stencil(ctx, x, y, width, height,
+                              &clippedUnpack, pixels);
    }
    else {
       /* sub-optimal cases:
@@ -616,58 +585,27 @@ draw_depth_stencil_pixels(struct gl_context *ctx, GLint x, GLint y,
       /* XXX need to handle very wide images (skippixels) */
       GLint i;
 
-      depthRb = ctx->DrawBuffer->_DepthBuffer;
-      stencilRb = ctx->DrawBuffer->_StencilBuffer;
-
       for (i = 0; i < height; i++) {
          const GLuint *depthStencilSrc = (const GLuint *)
             _mesa_image_address2d(&clippedUnpack, pixels, width, height,
                                   GL_DEPTH_STENCIL_EXT, type, i, 0);
 
          if (ctx->Depth.Mask) {
-            if (!scaleOrBias && ctx->DrawBuffer->Visual.depthBits == 24 &&
-		type == GL_UNSIGNED_INT_24_8) {
-               /* fast path 24-bit zbuffer */
-               GLuint zValues[MAX_WIDTH];
-               GLint j;
-               ASSERT(depthRb->DataType == GL_UNSIGNED_INT);
-               for (j = 0; j < width; j++) {
-                  zValues[j] = depthStencilSrc[j] >> 8;
-               }
-               if (zoom)
-                  _swrast_write_zoomed_z_span(ctx, imgX, imgY, width,
-                                              x, y + i, zValues);
-               else
-                  depthRb->PutRow(ctx, depthRb, width, x, y + i, zValues,NULL);
-            }
-            else if (!scaleOrBias && ctx->DrawBuffer->Visual.depthBits == 16 &&
-		     type == GL_UNSIGNED_INT_24_8) {
-               /* fast path 16-bit zbuffer */
-               GLushort zValues[MAX_WIDTH];
-               GLint j;
-               ASSERT(depthRb->DataType == GL_UNSIGNED_SHORT);
-               for (j = 0; j < width; j++) {
-                  zValues[j] = depthStencilSrc[j] >> 16;
-               }
-               if (zoom)
-                  _swrast_write_zoomed_z_span(ctx, imgX, imgY, width,
-                                              x, y + i, zValues);
-               else
-                  depthRb->PutRow(ctx, depthRb, width, x, y + i, zValues,NULL);
+            GLuint zValues[MAX_WIDTH];  /* 32-bit Z values */
+            _mesa_unpack_depth_span(ctx, width,
+                                    GL_UNSIGNED_INT, /* dest type */
+                                    zValues,         /* dest addr */
+                                    0xffffffff,      /* depth max */
+                                    type,            /* src type */
+                                    depthStencilSrc, /* src addr */
+                                    &clippedUnpack);
+            if (zoom) {
+               _swrast_write_zoomed_z_span(ctx, imgX, imgY, width, x,
+                                           y + i, zValues);
             }
             else {
-               /* general case */
-               GLuint zValues[MAX_WIDTH];  /* 16 or 32-bit Z value storage */
-               _mesa_unpack_depth_span(ctx, width,
-                                       depthRb->DataType, zValues, depthMax,
-                                       type, depthStencilSrc, &clippedUnpack);
-               if (zoom) {
-                  _swrast_write_zoomed_z_span(ctx, imgX, imgY, width, x,
-                                              y + i, zValues);
-               }
-               else {
-                  depthRb->PutRow(ctx, depthRb, width, x, y + i, zValues,NULL);
-               }
+               GLubyte *dst = _swrast_pixel_address(depthRb, x, y + i);
+               _mesa_pack_uint_z_row(depthRb->Format, width, zValues, dst);
             }
          }
 
@@ -716,8 +654,6 @@ _swrast_DrawPixels( struct gl_context *ctx,
     */
    _mesa_set_vp_override(ctx, GL_TRUE);
 
-   swrast_render_start(ctx);
-
    if (ctx->NewState)
       _mesa_update_state(ctx);
 
@@ -726,7 +662,6 @@ _swrast_DrawPixels( struct gl_context *ctx,
 
    pixels = _mesa_map_pbo_source(ctx, unpack, pixels);
    if (!pixels) {
-      swrast_render_finish(ctx);
       _mesa_set_vp_override(ctx, save_vp_override);
       return;
    }
@@ -736,20 +671,25 @@ _swrast_DrawPixels( struct gl_context *ctx,
     */
    switch (format) {
    case GL_STENCIL_INDEX:
+      swrast_render_start(ctx);
       draw_stencil_pixels( ctx, x, y, width, height, type, unpack, pixels );
+      swrast_render_finish(ctx);
       break;
    case GL_DEPTH_COMPONENT:
+      swrast_render_start(ctx);
       draw_depth_pixels( ctx, x, y, width, height, type, unpack, pixels );
+      swrast_render_finish(ctx);
       break;
    case GL_DEPTH_STENCIL_EXT:
+      swrast_render_start(ctx);
       draw_depth_stencil_pixels(ctx, x, y, width, height, type, unpack, pixels);
+      swrast_render_finish(ctx);
       break;
    default:
       /* all other formats should be color formats */
       draw_rgba_pixels(ctx, x, y, width, height, format, type, unpack, pixels);
    }
 
-   swrast_render_finish(ctx);
    _mesa_set_vp_override(ctx, save_vp_override);
 
    _mesa_unmap_pbo_source(ctx, unpack);
diff --git a/mesalib/src/mesa/swrast/s_linetemp.h b/mesalib/src/mesa/swrast/s_linetemp.h
index bd94ca40a..1668f7db9 100644
--- a/mesalib/src/mesa/swrast/s_linetemp.h
+++ b/mesalib/src/mesa/swrast/s_linetemp.h
@@ -1,402 +1,402 @@
-/*

- * Mesa 3-D graphics library

- * Version:  7.0

- *

- * Copyright (C) 1999-2007  Brian Paul   All Rights Reserved.

- *

- * Permission is hereby granted, free of charge, to any person obtaining a

- * copy of this software and associated documentation files (the "Software"),

- * to deal in the Software without restriction, including without limitation

- * the rights to use, copy, modify, merge, publish, distribute, sublicense,

- * and/or sell copies of the Software, and to permit persons to whom the

- * Software is furnished to do so, subject to the following conditions:

- *

- * The above copyright notice and this permission notice shall be included

- * in all copies or substantial portions of the Software.

- *

- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS

- * OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,

- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL

- * BRIAN PAUL BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN

- * AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN

- * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.

- */

-

-

-/*

- * Line Rasterizer Template

- *

- * This file is #include'd to generate custom line rasterizers.

- *

- * The following macros may be defined to indicate what auxillary information

- * must be interplated along the line:

- *    INTERP_Z        - if defined, interpolate Z values

- *    INTERP_ATTRIBS  - if defined, interpolate attribs (texcoords, varying, etc)

- *

- * When one can directly address pixels in the color buffer the following

- * macros can be defined and used to directly compute pixel addresses during

- * rasterization (see pixelPtr):

- *    PIXEL_TYPE          - the datatype of a pixel (GLubyte, GLushort, GLuint)

- *    BYTES_PER_ROW       - number of bytes per row in the color buffer

- *    PIXEL_ADDRESS(X,Y)  - returns the address of pixel at (X,Y) where

- *                          Y==0 at bottom of screen and increases upward.

- *

- * Similarly, for direct depth buffer access, this type is used for depth

- * buffer addressing:

- *    DEPTH_TYPE          - either GLushort or GLuint

- *

- * Optionally, one may provide one-time setup code

- *    SETUP_CODE    - code which is to be executed once per line

- *

- * To actually "plot" each pixel the PLOT macro must be defined...

- *    PLOT(X,Y) - code to plot a pixel.  Example:

- *                if (Z < *zPtr) {

- *                   *zPtr = Z;

- *                   color = pack_rgb( FixedToInt(r0), FixedToInt(g0),

- *                                     FixedToInt(b0) );

- *                   put_pixel( X, Y, color );

- *                }

- *

- * This code was designed for the origin to be in the lower-left corner.

- *

- */

-

-

-static void

-NAME( struct gl_context *ctx, const SWvertex *vert0, const SWvertex *vert1 )

-{

-   const SWcontext *swrast = SWRAST_CONTEXT(ctx);

-   SWspan span;

-   GLuint interpFlags = 0;

-   GLint x0 = (GLint) vert0->attrib[FRAG_ATTRIB_WPOS][0];

-   GLint x1 = (GLint) vert1->attrib[FRAG_ATTRIB_WPOS][0];

-   GLint y0 = (GLint) vert0->attrib[FRAG_ATTRIB_WPOS][1];

-   GLint y1 = (GLint) vert1->attrib[FRAG_ATTRIB_WPOS][1];

-   GLint dx, dy;

-   GLint numPixels;

-   GLint xstep, ystep;

-#if defined(DEPTH_TYPE)

-   const GLint depthBits = ctx->DrawBuffer->Visual.depthBits;

-   const GLint fixedToDepthShift = depthBits <= 16 ? FIXED_SHIFT : 0;

-   struct gl_renderbuffer *zrb = ctx->DrawBuffer->Attachment[BUFFER_DEPTH].Renderbuffer;

-#define FixedToDepth(F)  ((F) >> fixedToDepthShift)

-   GLint zPtrXstep, zPtrYstep;

-   DEPTH_TYPE *zPtr;

-#elif defined(INTERP_Z)

-   const GLint depthBits = ctx->DrawBuffer->Visual.depthBits;

-#endif

-#ifdef PIXEL_ADDRESS

-   PIXEL_TYPE *pixelPtr;

-   GLint pixelXstep, pixelYstep;

-#endif

-

-#ifdef SETUP_CODE

-   SETUP_CODE

-#endif

-

-   (void) swrast;

-

-   /* Cull primitives with malformed coordinates.

-    */

-   {

-      GLfloat tmp = vert0->attrib[FRAG_ATTRIB_WPOS][0] + vert0->attrib[FRAG_ATTRIB_WPOS][1]

-                  + vert1->attrib[FRAG_ATTRIB_WPOS][0] + vert1->attrib[FRAG_ATTRIB_WPOS][1];

-      if (IS_INF_OR_NAN(tmp))

-	 return;

-   }

-

-   /*

-   printf("%s():\n", __FUNCTION__);

-   printf(" (%f, %f, %f) -> (%f, %f, %f)\n",

-          vert0->attrib[FRAG_ATTRIB_WPOS][0],

-          vert0->attrib[FRAG_ATTRIB_WPOS][1],

-          vert0->attrib[FRAG_ATTRIB_WPOS][2],

-          vert1->attrib[FRAG_ATTRIB_WPOS][0],

-          vert1->attrib[FRAG_ATTRIB_WPOS][1],

-          vert1->attrib[FRAG_ATTRIB_WPOS][2]);

-   printf(" (%d, %d, %d) -> (%d, %d, %d)\n",

-          vert0->color[0], vert0->color[1], vert0->color[2], 

-          vert1->color[0], vert1->color[1], vert1->color[2]);

-   printf(" (%d, %d, %d) -> (%d, %d, %d)\n",

-          vert0->specular[0], vert0->specular[1], vert0->specular[2], 

-          vert1->specular[0], vert1->specular[1], vert1->specular[2]);

-   */

-

-/*

- * Despite being clipped to the view volume, the line's window coordinates

- * may just lie outside the window bounds.  That is, if the legal window

- * coordinates are [0,W-1][0,H-1], it's possible for x==W and/or y==H.

- * This quick and dirty code nudges the endpoints inside the window if

- * necessary.

- */

-#ifdef CLIP_HACK

-   {

-      GLint w = ctx->DrawBuffer->Width;

-      GLint h = ctx->DrawBuffer->Height;

-      if ((x0==w) | (x1==w)) {

-         if ((x0==w) & (x1==w))

-           return;

-         x0 -= x0==w;

-         x1 -= x1==w;

-      }

-      if ((y0==h) | (y1==h)) {

-         if ((y0==h) & (y1==h))

-           return;

-         y0 -= y0==h;

-         y1 -= y1==h;

-      }

-   }

-#endif

-

-   dx = x1 - x0;

-   dy = y1 - y0;

-   if (dx == 0 && dy == 0)

-      return;

-

-   /*

-   printf("%s %d,%d  %g %g %g %g  %g %g %g %g\n", __FUNCTION__, dx, dy,

-          vert0->attrib[FRAG_ATTRIB_COL1][0],

-          vert0->attrib[FRAG_ATTRIB_COL1][1],

-          vert0->attrib[FRAG_ATTRIB_COL1][2],

-          vert0->attrib[FRAG_ATTRIB_COL1][3],

-          vert1->attrib[FRAG_ATTRIB_COL1][0],

-          vert1->attrib[FRAG_ATTRIB_COL1][1],

-          vert1->attrib[FRAG_ATTRIB_COL1][2],

-          vert1->attrib[FRAG_ATTRIB_COL1][3]);

-   */

-

-#ifdef DEPTH_TYPE

-   zPtr = (DEPTH_TYPE *) zrb->GetPointer(ctx, zrb, x0, y0);

-#endif

-#ifdef PIXEL_ADDRESS

-   pixelPtr = (PIXEL_TYPE *) PIXEL_ADDRESS(x0,y0);

-#endif

-

-   if (dx<0) {

-      dx = -dx;   /* make positive */

-      xstep = -1;

-#ifdef DEPTH_TYPE

-      zPtrXstep = -((GLint)sizeof(DEPTH_TYPE));

-#endif

-#ifdef PIXEL_ADDRESS

-      pixelXstep = -((GLint)sizeof(PIXEL_TYPE));

-#endif

-   }

-   else {

-      xstep = 1;

-#ifdef DEPTH_TYPE

-      zPtrXstep = ((GLint)sizeof(DEPTH_TYPE));

-#endif

-#ifdef PIXEL_ADDRESS

-      pixelXstep = ((GLint)sizeof(PIXEL_TYPE));

-#endif

-   }

-

-   if (dy<0) {

-      dy = -dy;   /* make positive */

-      ystep = -1;

-#ifdef DEPTH_TYPE

-      zPtrYstep = -((GLint) (ctx->DrawBuffer->Width * sizeof(DEPTH_TYPE)));

-#endif

-#ifdef PIXEL_ADDRESS

-      pixelYstep = BYTES_PER_ROW;

-#endif

-   }

-   else {

-      ystep = 1;

-#ifdef DEPTH_TYPE

-      zPtrYstep = (GLint) (ctx->DrawBuffer->Width * sizeof(DEPTH_TYPE));

-#endif

-#ifdef PIXEL_ADDRESS

-      pixelYstep = -(BYTES_PER_ROW);

-#endif

-   }

-

-   ASSERT(dx >= 0);

-   ASSERT(dy >= 0);

-

-   numPixels = MAX2(dx, dy);

-

-   /*

-    * Span setup: compute start and step values for all interpolated values.

-    */

-   interpFlags |= SPAN_RGBA;

-   if (ctx->Light.ShadeModel == GL_SMOOTH) {

-      span.red   = ChanToFixed(vert0->color[0]);

-      span.green = ChanToFixed(vert0->color[1]);

-      span.blue  = ChanToFixed(vert0->color[2]);

-      span.alpha = ChanToFixed(vert0->color[3]);

-      span.redStep   = (ChanToFixed(vert1->color[0]) - span.red  ) / numPixels;

-      span.greenStep = (ChanToFixed(vert1->color[1]) - span.green) / numPixels;

-      span.blueStep  = (ChanToFixed(vert1->color[2]) - span.blue ) / numPixels;

-      span.alphaStep = (ChanToFixed(vert1->color[3]) - span.alpha) / numPixels;

-   }

-   else {

-      span.red   = ChanToFixed(vert1->color[0]);

-      span.green = ChanToFixed(vert1->color[1]);

-      span.blue  = ChanToFixed(vert1->color[2]);

-      span.alpha = ChanToFixed(vert1->color[3]);

-      span.redStep   = 0;

-      span.greenStep = 0;

-      span.blueStep  = 0;

-      span.alphaStep = 0;

-   }

-#if defined(INTERP_Z) || defined(DEPTH_TYPE)

-   interpFlags |= SPAN_Z;

-   {

-      if (depthBits <= 16) {

-         span.z = FloatToFixed(vert0->attrib[FRAG_ATTRIB_WPOS][2]) + FIXED_HALF;

-         span.zStep = FloatToFixed(  vert1->attrib[FRAG_ATTRIB_WPOS][2]

-                                   - vert0->attrib[FRAG_ATTRIB_WPOS][2]) / numPixels;

-      }

-      else {

-         /* don't use fixed point */

-         span.z = (GLuint) vert0->attrib[FRAG_ATTRIB_WPOS][2];

-         span.zStep = (GLint) ((  vert1->attrib[FRAG_ATTRIB_WPOS][2]

-                                - vert0->attrib[FRAG_ATTRIB_WPOS][2]) / numPixels);

-      }

-   }

-#endif

-#if defined(INTERP_ATTRIBS)

-   {

-      const GLfloat invLen = 1.0F / numPixels;

-      const GLfloat invw0 = vert0->attrib[FRAG_ATTRIB_WPOS][3];

-      const GLfloat invw1 = vert1->attrib[FRAG_ATTRIB_WPOS][3];

-

-      span.attrStart[FRAG_ATTRIB_WPOS][3] = invw0;

-      span.attrStepX[FRAG_ATTRIB_WPOS][3] = (invw1 - invw0) * invLen;

-      span.attrStepY[FRAG_ATTRIB_WPOS][3] = 0.0;

-

-      ATTRIB_LOOP_BEGIN

-         if (swrast->_InterpMode[attr] == GL_FLAT) {

-            COPY_4V(span.attrStart[attr], vert1->attrib[attr]);

-            ASSIGN_4V(span.attrStepX[attr], 0.0, 0.0, 0.0, 0.0);

-         }

-         else {

-            GLuint c;

-            for (c = 0; c < 4; c++) {

-               float da;

-               span.attrStart[attr][c] = invw0 * vert0->attrib[attr][c];

-               da = (invw1 * vert1->attrib[attr][c]) - span.attrStart[attr][c];

-               span.attrStepX[attr][c] = da * invLen;

-            }

-         }

-         ASSIGN_4V(span.attrStepY[attr], 0.0, 0.0, 0.0, 0.0);

-      ATTRIB_LOOP_END

-   }

-#endif

-

-   INIT_SPAN(span, GL_LINE);

-   span.end = numPixels;

-   span.interpMask = interpFlags;

-   span.arrayMask = SPAN_XY;

-

-   span.facing = swrast->PointLineFacing;

-

-

-   /*

-    * Draw

-    */

-

-   if (dx > dy) {

-      /*** X-major line ***/

-      GLint i;

-      GLint errorInc = dy+dy;

-      GLint error = errorInc-dx;

-      GLint errorDec = error-dx;

-

-      for (i = 0; i < dx; i++) {

-#ifdef DEPTH_TYPE

-         GLuint Z = FixedToDepth(span.z);

-#endif

-#ifdef PLOT

-         PLOT( x0, y0 );

-#else

-         span.array->x[i] = x0;

-         span.array->y[i] = y0;

-#endif

-         x0 += xstep;

-#ifdef DEPTH_TYPE

-         zPtr = (DEPTH_TYPE *) ((GLubyte*) zPtr + zPtrXstep);

-         span.z += span.zStep;

-#endif

-#ifdef PIXEL_ADDRESS

-         pixelPtr = (PIXEL_TYPE*) ((GLubyte*) pixelPtr + pixelXstep);

-#endif

-         if (error < 0) {

-            error += errorInc;

-         }

-         else {

-            error += errorDec;

-            y0 += ystep;

-#ifdef DEPTH_TYPE

-            zPtr = (DEPTH_TYPE *) ((GLubyte*) zPtr + zPtrYstep);

-#endif

-#ifdef PIXEL_ADDRESS

-            pixelPtr = (PIXEL_TYPE*) ((GLubyte*) pixelPtr + pixelYstep);

-#endif

-         }

-      }

-   }

-   else {

-      /*** Y-major line ***/

-      GLint i;

-      GLint errorInc = dx+dx;

-      GLint error = errorInc-dy;

-      GLint errorDec = error-dy;

-

-      for (i=0;i<dy;i++) {

-#ifdef DEPTH_TYPE

-         GLuint Z = FixedToDepth(span.z);

-#endif

-#ifdef PLOT

-         PLOT( x0, y0 );

-#else

-         span.array->x[i] = x0;

-         span.array->y[i] = y0;

-#endif

-         y0 += ystep;

-#ifdef DEPTH_TYPE

-         zPtr = (DEPTH_TYPE *) ((GLubyte*) zPtr + zPtrYstep);

-         span.z += span.zStep;

-#endif

-#ifdef PIXEL_ADDRESS

-         pixelPtr = (PIXEL_TYPE*) ((GLubyte*) pixelPtr + pixelYstep);

-#endif

-         if (error<0) {

-            error += errorInc;

-         }

-         else {

-            error += errorDec;

-            x0 += xstep;

-#ifdef DEPTH_TYPE

-            zPtr = (DEPTH_TYPE *) ((GLubyte*) zPtr + zPtrXstep);

-#endif

-#ifdef PIXEL_ADDRESS

-            pixelPtr = (PIXEL_TYPE*) ((GLubyte*) pixelPtr + pixelXstep);

-#endif

-         }

-      }

-   }

-

-#ifdef RENDER_SPAN

-   RENDER_SPAN( span );

-#endif

-

-   (void)span;

-

-}

-

-

-#undef NAME

-#undef INTERP_Z

-#undef INTERP_ATTRIBS

-#undef PIXEL_ADDRESS

-#undef PIXEL_TYPE

-#undef DEPTH_TYPE

-#undef BYTES_PER_ROW

-#undef SETUP_CODE

-#undef PLOT

-#undef CLIP_HACK

-#undef FixedToDepth

-#undef RENDER_SPAN

+/*
+ * Mesa 3-D graphics library
+ * Version:  7.0
+ *
+ * Copyright (C) 1999-2007  Brian Paul   All Rights Reserved.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included
+ * in all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS
+ * OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * BRIAN PAUL BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN
+ * AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
+ */
+
+
+/*
+ * Line Rasterizer Template
+ *
+ * This file is #include'd to generate custom line rasterizers.
+ *
+ * The following macros may be defined to indicate what auxillary information
+ * must be interplated along the line:
+ *    INTERP_Z        - if defined, interpolate Z values
+ *    INTERP_ATTRIBS  - if defined, interpolate attribs (texcoords, varying, etc)
+ *
+ * When one can directly address pixels in the color buffer the following
+ * macros can be defined and used to directly compute pixel addresses during
+ * rasterization (see pixelPtr):
+ *    PIXEL_TYPE          - the datatype of a pixel (GLubyte, GLushort, GLuint)
+ *    BYTES_PER_ROW       - number of bytes per row in the color buffer
+ *    PIXEL_ADDRESS(X,Y)  - returns the address of pixel at (X,Y) where
+ *                          Y==0 at bottom of screen and increases upward.
+ *
+ * Similarly, for direct depth buffer access, this type is used for depth
+ * buffer addressing:
+ *    DEPTH_TYPE          - either GLushort or GLuint
+ *
+ * Optionally, one may provide one-time setup code
+ *    SETUP_CODE    - code which is to be executed once per line
+ *
+ * To actually "plot" each pixel the PLOT macro must be defined...
+ *    PLOT(X,Y) - code to plot a pixel.  Example:
+ *                if (Z < *zPtr) {
+ *                   *zPtr = Z;
+ *                   color = pack_rgb( FixedToInt(r0), FixedToInt(g0),
+ *                                     FixedToInt(b0) );
+ *                   put_pixel( X, Y, color );
+ *                }
+ *
+ * This code was designed for the origin to be in the lower-left corner.
+ *
+ */
+
+
+static void
+NAME( struct gl_context *ctx, const SWvertex *vert0, const SWvertex *vert1 )
+{
+   const SWcontext *swrast = SWRAST_CONTEXT(ctx);
+   SWspan span;
+   GLuint interpFlags = 0;
+   GLint x0 = (GLint) vert0->attrib[FRAG_ATTRIB_WPOS][0];
+   GLint x1 = (GLint) vert1->attrib[FRAG_ATTRIB_WPOS][0];
+   GLint y0 = (GLint) vert0->attrib[FRAG_ATTRIB_WPOS][1];
+   GLint y1 = (GLint) vert1->attrib[FRAG_ATTRIB_WPOS][1];
+   GLint dx, dy;
+   GLint numPixels;
+   GLint xstep, ystep;
+#if defined(DEPTH_TYPE)
+   const GLint depthBits = ctx->DrawBuffer->Visual.depthBits;
+   const GLint fixedToDepthShift = depthBits <= 16 ? FIXED_SHIFT : 0;
+   struct gl_renderbuffer *zrb = ctx->DrawBuffer->Attachment[BUFFER_DEPTH].Renderbuffer;
+#define FixedToDepth(F)  ((F) >> fixedToDepthShift)
+   GLint zPtrXstep, zPtrYstep;
+   DEPTH_TYPE *zPtr;
+#elif defined(INTERP_Z)
+   const GLint depthBits = ctx->DrawBuffer->Visual.depthBits;
+#endif
+#ifdef PIXEL_ADDRESS
+   PIXEL_TYPE *pixelPtr;
+   GLint pixelXstep, pixelYstep;
+#endif
+
+#ifdef SETUP_CODE
+   SETUP_CODE
+#endif
+
+   (void) swrast;
+
+   /* Cull primitives with malformed coordinates.
+    */
+   {
+      GLfloat tmp = vert0->attrib[FRAG_ATTRIB_WPOS][0] + vert0->attrib[FRAG_ATTRIB_WPOS][1]
+                  + vert1->attrib[FRAG_ATTRIB_WPOS][0] + vert1->attrib[FRAG_ATTRIB_WPOS][1];
+      if (IS_INF_OR_NAN(tmp))
+	 return;
+   }
+
+   /*
+   printf("%s():\n", __FUNCTION__);
+   printf(" (%f, %f, %f) -> (%f, %f, %f)\n",
+          vert0->attrib[FRAG_ATTRIB_WPOS][0],
+          vert0->attrib[FRAG_ATTRIB_WPOS][1],
+          vert0->attrib[FRAG_ATTRIB_WPOS][2],
+          vert1->attrib[FRAG_ATTRIB_WPOS][0],
+          vert1->attrib[FRAG_ATTRIB_WPOS][1],
+          vert1->attrib[FRAG_ATTRIB_WPOS][2]);
+   printf(" (%d, %d, %d) -> (%d, %d, %d)\n",
+          vert0->color[0], vert0->color[1], vert0->color[2], 
+          vert1->color[0], vert1->color[1], vert1->color[2]);
+   printf(" (%d, %d, %d) -> (%d, %d, %d)\n",
+          vert0->specular[0], vert0->specular[1], vert0->specular[2], 
+          vert1->specular[0], vert1->specular[1], vert1->specular[2]);
+   */
+
+/*
+ * Despite being clipped to the view volume, the line's window coordinates
+ * may just lie outside the window bounds.  That is, if the legal window
+ * coordinates are [0,W-1][0,H-1], it's possible for x==W and/or y==H.
+ * This quick and dirty code nudges the endpoints inside the window if
+ * necessary.
+ */
+#ifdef CLIP_HACK
+   {
+      GLint w = ctx->DrawBuffer->Width;
+      GLint h = ctx->DrawBuffer->Height;
+      if ((x0==w) | (x1==w)) {
+         if ((x0==w) & (x1==w))
+           return;
+         x0 -= x0==w;
+         x1 -= x1==w;
+      }
+      if ((y0==h) | (y1==h)) {
+         if ((y0==h) & (y1==h))
+           return;
+         y0 -= y0==h;
+         y1 -= y1==h;
+      }
+   }
+#endif
+
+   dx = x1 - x0;
+   dy = y1 - y0;
+   if (dx == 0 && dy == 0)
+      return;
+
+   /*
+   printf("%s %d,%d  %g %g %g %g  %g %g %g %g\n", __FUNCTION__, dx, dy,
+          vert0->attrib[FRAG_ATTRIB_COL1][0],
+          vert0->attrib[FRAG_ATTRIB_COL1][1],
+          vert0->attrib[FRAG_ATTRIB_COL1][2],
+          vert0->attrib[FRAG_ATTRIB_COL1][3],
+          vert1->attrib[FRAG_ATTRIB_COL1][0],
+          vert1->attrib[FRAG_ATTRIB_COL1][1],
+          vert1->attrib[FRAG_ATTRIB_COL1][2],
+          vert1->attrib[FRAG_ATTRIB_COL1][3]);
+   */
+
+#ifdef DEPTH_TYPE
+   zPtr = (DEPTH_TYPE *) _swrast_pixel_address(zrb, x0, y0);
+#endif
+#ifdef PIXEL_ADDRESS
+   pixelPtr = (PIXEL_TYPE *) PIXEL_ADDRESS(x0,y0);
+#endif
+
+   if (dx<0) {
+      dx = -dx;   /* make positive */
+      xstep = -1;
+#ifdef DEPTH_TYPE
+      zPtrXstep = -((GLint)sizeof(DEPTH_TYPE));
+#endif
+#ifdef PIXEL_ADDRESS
+      pixelXstep = -((GLint)sizeof(PIXEL_TYPE));
+#endif
+   }
+   else {
+      xstep = 1;
+#ifdef DEPTH_TYPE
+      zPtrXstep = ((GLint)sizeof(DEPTH_TYPE));
+#endif
+#ifdef PIXEL_ADDRESS
+      pixelXstep = ((GLint)sizeof(PIXEL_TYPE));
+#endif
+   }
+
+   if (dy<0) {
+      dy = -dy;   /* make positive */
+      ystep = -1;
+#ifdef DEPTH_TYPE
+      zPtrYstep = -((GLint) (ctx->DrawBuffer->Width * sizeof(DEPTH_TYPE)));
+#endif
+#ifdef PIXEL_ADDRESS
+      pixelYstep = BYTES_PER_ROW;
+#endif
+   }
+   else {
+      ystep = 1;
+#ifdef DEPTH_TYPE
+      zPtrYstep = (GLint) (ctx->DrawBuffer->Width * sizeof(DEPTH_TYPE));
+#endif
+#ifdef PIXEL_ADDRESS
+      pixelYstep = -(BYTES_PER_ROW);
+#endif
+   }
+
+   ASSERT(dx >= 0);
+   ASSERT(dy >= 0);
+
+   numPixels = MAX2(dx, dy);
+
+   /*
+    * Span setup: compute start and step values for all interpolated values.
+    */
+   interpFlags |= SPAN_RGBA;
+   if (ctx->Light.ShadeModel == GL_SMOOTH) {
+      span.red   = ChanToFixed(vert0->color[0]);
+      span.green = ChanToFixed(vert0->color[1]);
+      span.blue  = ChanToFixed(vert0->color[2]);
+      span.alpha = ChanToFixed(vert0->color[3]);
+      span.redStep   = (ChanToFixed(vert1->color[0]) - span.red  ) / numPixels;
+      span.greenStep = (ChanToFixed(vert1->color[1]) - span.green) / numPixels;
+      span.blueStep  = (ChanToFixed(vert1->color[2]) - span.blue ) / numPixels;
+      span.alphaStep = (ChanToFixed(vert1->color[3]) - span.alpha) / numPixels;
+   }
+   else {
+      span.red   = ChanToFixed(vert1->color[0]);
+      span.green = ChanToFixed(vert1->color[1]);
+      span.blue  = ChanToFixed(vert1->color[2]);
+      span.alpha = ChanToFixed(vert1->color[3]);
+      span.redStep   = 0;
+      span.greenStep = 0;
+      span.blueStep  = 0;
+      span.alphaStep = 0;
+   }
+#if defined(INTERP_Z) || defined(DEPTH_TYPE)
+   interpFlags |= SPAN_Z;
+   {
+      if (depthBits <= 16) {
+         span.z = FloatToFixed(vert0->attrib[FRAG_ATTRIB_WPOS][2]) + FIXED_HALF;
+         span.zStep = FloatToFixed(  vert1->attrib[FRAG_ATTRIB_WPOS][2]
+                                   - vert0->attrib[FRAG_ATTRIB_WPOS][2]) / numPixels;
+      }
+      else {
+         /* don't use fixed point */
+         span.z = (GLuint) vert0->attrib[FRAG_ATTRIB_WPOS][2];
+         span.zStep = (GLint) ((  vert1->attrib[FRAG_ATTRIB_WPOS][2]
+                                - vert0->attrib[FRAG_ATTRIB_WPOS][2]) / numPixels);
+      }
+   }
+#endif
+#if defined(INTERP_ATTRIBS)
+   {
+      const GLfloat invLen = 1.0F / numPixels;
+      const GLfloat invw0 = vert0->attrib[FRAG_ATTRIB_WPOS][3];
+      const GLfloat invw1 = vert1->attrib[FRAG_ATTRIB_WPOS][3];
+
+      span.attrStart[FRAG_ATTRIB_WPOS][3] = invw0;
+      span.attrStepX[FRAG_ATTRIB_WPOS][3] = (invw1 - invw0) * invLen;
+      span.attrStepY[FRAG_ATTRIB_WPOS][3] = 0.0;
+
+      ATTRIB_LOOP_BEGIN
+         if (swrast->_InterpMode[attr] == GL_FLAT) {
+            COPY_4V(span.attrStart[attr], vert1->attrib[attr]);
+            ASSIGN_4V(span.attrStepX[attr], 0.0, 0.0, 0.0, 0.0);
+         }
+         else {
+            GLuint c;
+            for (c = 0; c < 4; c++) {
+               float da;
+               span.attrStart[attr][c] = invw0 * vert0->attrib[attr][c];
+               da = (invw1 * vert1->attrib[attr][c]) - span.attrStart[attr][c];
+               span.attrStepX[attr][c] = da * invLen;
+            }
+         }
+         ASSIGN_4V(span.attrStepY[attr], 0.0, 0.0, 0.0, 0.0);
+      ATTRIB_LOOP_END
+   }
+#endif
+
+   INIT_SPAN(span, GL_LINE);
+   span.end = numPixels;
+   span.interpMask = interpFlags;
+   span.arrayMask = SPAN_XY;
+
+   span.facing = swrast->PointLineFacing;
+
+
+   /*
+    * Draw
+    */
+
+   if (dx > dy) {
+      /*** X-major line ***/
+      GLint i;
+      GLint errorInc = dy+dy;
+      GLint error = errorInc-dx;
+      GLint errorDec = error-dx;
+
+      for (i = 0; i < dx; i++) {
+#ifdef DEPTH_TYPE
+         GLuint Z = FixedToDepth(span.z);
+#endif
+#ifdef PLOT
+         PLOT( x0, y0 );
+#else
+         span.array->x[i] = x0;
+         span.array->y[i] = y0;
+#endif
+         x0 += xstep;
+#ifdef DEPTH_TYPE
+         zPtr = (DEPTH_TYPE *) ((GLubyte*) zPtr + zPtrXstep);
+         span.z += span.zStep;
+#endif
+#ifdef PIXEL_ADDRESS
+         pixelPtr = (PIXEL_TYPE*) ((GLubyte*) pixelPtr + pixelXstep);
+#endif
+         if (error < 0) {
+            error += errorInc;
+         }
+         else {
+            error += errorDec;
+            y0 += ystep;
+#ifdef DEPTH_TYPE
+            zPtr = (DEPTH_TYPE *) ((GLubyte*) zPtr + zPtrYstep);
+#endif
+#ifdef PIXEL_ADDRESS
+            pixelPtr = (PIXEL_TYPE*) ((GLubyte*) pixelPtr + pixelYstep);
+#endif
+         }
+      }
+   }
+   else {
+      /*** Y-major line ***/
+      GLint i;
+      GLint errorInc = dx+dx;
+      GLint error = errorInc-dy;
+      GLint errorDec = error-dy;
+
+      for (i=0;i<dy;i++) {
+#ifdef DEPTH_TYPE
+         GLuint Z = FixedToDepth(span.z);
+#endif
+#ifdef PLOT
+         PLOT( x0, y0 );
+#else
+         span.array->x[i] = x0;
+         span.array->y[i] = y0;
+#endif
+         y0 += ystep;
+#ifdef DEPTH_TYPE
+         zPtr = (DEPTH_TYPE *) ((GLubyte*) zPtr + zPtrYstep);
+         span.z += span.zStep;
+#endif
+#ifdef PIXEL_ADDRESS
+         pixelPtr = (PIXEL_TYPE*) ((GLubyte*) pixelPtr + pixelYstep);
+#endif
+         if (error<0) {
+            error += errorInc;
+         }
+         else {
+            error += errorDec;
+            x0 += xstep;
+#ifdef DEPTH_TYPE
+            zPtr = (DEPTH_TYPE *) ((GLubyte*) zPtr + zPtrXstep);
+#endif
+#ifdef PIXEL_ADDRESS
+            pixelPtr = (PIXEL_TYPE*) ((GLubyte*) pixelPtr + pixelXstep);
+#endif
+         }
+      }
+   }
+
+#ifdef RENDER_SPAN
+   RENDER_SPAN( span );
+#endif
+
+   (void)span;
+
+}
+
+
+#undef NAME
+#undef INTERP_Z
+#undef INTERP_ATTRIBS
+#undef PIXEL_ADDRESS
+#undef PIXEL_TYPE
+#undef DEPTH_TYPE
+#undef BYTES_PER_ROW
+#undef SETUP_CODE
+#undef PLOT
+#undef CLIP_HACK
+#undef FixedToDepth
+#undef RENDER_SPAN
diff --git a/mesalib/src/mesa/swrast/s_renderbuffer.c b/mesalib/src/mesa/swrast/s_renderbuffer.c
index bb7cdebfe..1a7cb36d9 100644
--- a/mesalib/src/mesa/swrast/s_renderbuffer.c
+++ b/mesalib/src/mesa/swrast/s_renderbuffer.c
@@ -158,31 +158,6 @@ put_row_generic(struct gl_context *ctx, struct gl_renderbuffer *rb,
    }
 }
 
-static void
-put_mono_row_generic(struct gl_context *ctx, struct gl_renderbuffer *rb,
-		     GLuint count, GLint x, GLint y,
-		     const void *value, const GLubyte *mask)
-{
-   void *row = rb->GetPointer(ctx, rb, x, y);
-   int format_bytes = _mesa_get_format_bytes(rb->Format) / sizeof(GLfloat);
-   unsigned int i;
-
-   if (mask) {
-      for (i = 0; i < count; i++) {
-	 char *dst = (char *) row + i * format_bytes;
-         if (mask[i]) {
-	    memcpy(dst, value, format_bytes);
-         }
-      }
-   }
-   else {
-      for (i = 0; i < count; i++) {
-	 char *dst = (char *) row + i * format_bytes;
-	 memcpy(dst, value, format_bytes);
-      }
-   }
-}
-
 
 static void
 put_values_generic(struct gl_context *ctx, struct gl_renderbuffer *rb,
@@ -203,22 +178,6 @@ put_values_generic(struct gl_context *ctx, struct gl_renderbuffer *rb,
 }
 
 
-static void
-put_mono_values_generic(struct gl_context *ctx,
-			struct gl_renderbuffer *rb,
-			GLuint count, const GLint x[], const GLint y[],
-			const void *value, const GLubyte *mask)
-{
-   int format_bytes = _mesa_get_format_bytes(rb->Format) / sizeof(GLfloat);
-   unsigned int i;
-
-   for (i = 0; i < count; i++) {
-      if (!mask || mask[i]) {
-	 void *dst = rb->GetPointer(ctx, rb, x[i], y[i]);
-	 memcpy(dst, value, format_bytes);
-      }
-   }
-}
 
 /**********************************************************************
  * Functions for buffers of 1 X GLubyte values.
@@ -261,30 +220,6 @@ put_row_ubyte(struct gl_context *ctx, struct gl_renderbuffer *rb, GLuint count,
 
 
 static void
-put_mono_row_ubyte(struct gl_context *ctx, struct gl_renderbuffer *rb, GLuint count,
-                   GLint x, GLint y, const void *value, const GLubyte *mask)
-{
-   const GLubyte val = *((const GLubyte *) value);
-   GLubyte *dst = (GLubyte *) rb->Data + y * rb->RowStride + x;
-   ASSERT(rb->DataType == GL_UNSIGNED_BYTE);
-   if (mask) {
-      GLuint i;
-      for (i = 0; i < count; i++) {
-         if (mask[i]) {
-            dst[i] = val;
-         }
-      }
-   }
-   else {
-      GLuint i;
-      for (i = 0; i < count; i++) {
-         dst[i] = val;
-      }
-   }
-}
-
-
-static void
 put_values_ubyte(struct gl_context *ctx, struct gl_renderbuffer *rb, GLuint count,
                  const GLint x[], const GLint y[],
                  const void *values, const GLubyte *mask)
@@ -301,23 +236,6 @@ put_values_ubyte(struct gl_context *ctx, struct gl_renderbuffer *rb, GLuint coun
 }
 
 
-static void
-put_mono_values_ubyte(struct gl_context *ctx, struct gl_renderbuffer *rb, GLuint count,
-                      const GLint x[], const GLint y[],
-                      const void *value, const GLubyte *mask)
-{
-   const GLubyte val = *((const GLubyte *) value);
-   GLuint i;
-   ASSERT(rb->DataType == GL_UNSIGNED_BYTE);
-   for (i = 0; i < count; i++) {
-      if (!mask || mask[i]) {
-         GLubyte *dst = (GLubyte *) rb->Data + y[i] * rb->RowStride + x[i];
-         *dst = val;
-      }
-   }
-}
-
-
 /**********************************************************************
  * Functions for buffers of 1 X GLushort values.
  * Typically depth/Z.
@@ -359,30 +277,6 @@ put_row_ushort(struct gl_context *ctx, struct gl_renderbuffer *rb, GLuint count,
 
 
 static void
-put_mono_row_ushort(struct gl_context *ctx, struct gl_renderbuffer *rb, GLuint count,
-                    GLint x, GLint y, const void *value, const GLubyte *mask)
-{
-   const GLushort val = *((const GLushort *) value);
-   GLushort *dst = (GLushort *) rb->Data + y * rb->RowStride + x;
-   ASSERT(rb->DataType == GL_UNSIGNED_SHORT);
-   if (mask) {
-      GLuint i;
-      for (i = 0; i < count; i++) {
-         if (mask[i]) {
-            dst[i] = val;
-         }
-      }
-   }
-   else {
-      GLuint i;
-      for (i = 0; i < count; i++) {
-         dst[i] = val;
-      }
-   }
-}
-
-
-static void
 put_values_ushort(struct gl_context *ctx, struct gl_renderbuffer *rb, GLuint count,
                   const GLint x[], const GLint y[], const void *values,
                   const GLubyte *mask)
@@ -399,32 +293,6 @@ put_values_ushort(struct gl_context *ctx, struct gl_renderbuffer *rb, GLuint cou
 }
  
 
-static void
-put_mono_values_ushort(struct gl_context *ctx, struct gl_renderbuffer *rb,
-                       GLuint count, const GLint x[], const GLint y[],
-                       const void *value, const GLubyte *mask)
-{
-   const GLushort val = *((const GLushort *) value);
-   ASSERT(rb->DataType == GL_UNSIGNED_SHORT);
-   if (mask) {
-      GLuint i;
-      for (i = 0; i < count; i++) {
-         if (mask[i]) {
-            GLushort *dst = (GLushort *) rb->Data + y[i] * rb->RowStride + x[i];
-            *dst = val;
-         }
-      }
-   }
-   else {
-      GLuint i;
-      for (i = 0; i < count; i++) {
-         GLushort *dst = (GLushort *) rb->Data + y[i] * rb->RowStride + x[i];
-         *dst = val;
-      }
-   }
-}
- 
-
 /**********************************************************************
  * Functions for buffers of 1 X GLuint values.
  * Typically depth/Z or color index.
@@ -468,31 +336,6 @@ put_row_uint(struct gl_context *ctx, struct gl_renderbuffer *rb, GLuint count,
 
 
 static void
-put_mono_row_uint(struct gl_context *ctx, struct gl_renderbuffer *rb, GLuint count,
-                  GLint x, GLint y, const void *value, const GLubyte *mask)
-{
-   const GLuint val = *((const GLuint *) value);
-   GLuint *dst = (GLuint *) rb->Data + y * rb->RowStride + x;
-   ASSERT(rb->DataType == GL_UNSIGNED_INT ||
-          rb->DataType == GL_UNSIGNED_INT_24_8_EXT);
-   if (mask) {
-      GLuint i;
-      for (i = 0; i < count; i++) {
-         if (mask[i]) {
-            dst[i] = val;
-         }
-      }
-   }
-   else {
-      GLuint i;
-      for (i = 0; i < count; i++) {
-         dst[i] = val;
-      }
-   }
-}
-
-
-static void
 put_values_uint(struct gl_context *ctx, struct gl_renderbuffer *rb, GLuint count,
                 const GLint x[], const GLint y[], const void *values,
                 const GLubyte *mask)
@@ -510,24 +353,6 @@ put_values_uint(struct gl_context *ctx, struct gl_renderbuffer *rb, GLuint count
 }
 
 
-static void
-put_mono_values_uint(struct gl_context *ctx, struct gl_renderbuffer *rb, GLuint count,
-                     const GLint x[], const GLint y[], const void *value,
-                     const GLubyte *mask)
-{
-   const GLuint val = *((const GLuint *) value);
-   GLuint i;
-   ASSERT(rb->DataType == GL_UNSIGNED_INT ||
-          rb->DataType == GL_UNSIGNED_INT_24_8_EXT);
-   for (i = 0; i < count; i++) {
-      if (!mask || mask[i]) {
-         GLuint *dst = (GLuint *) rb->Data + y[i] * rb->RowStride + x[i];
-         *dst = val;
-      }
-   }
-}
-
-
 /**********************************************************************
  * Functions for buffers of 3 X GLubyte (or GLbyte) values.
  * Typically color buffers.
@@ -606,54 +431,6 @@ put_row_ubyte3(struct gl_context *ctx, struct gl_renderbuffer *rb, GLuint count,
 
 
 static void
-put_row_rgb_ubyte3(struct gl_context *ctx, struct gl_renderbuffer *rb, GLuint count,
-                   GLint x, GLint y, const void *values, const GLubyte *mask)
-{
-   /* note: incoming values are RGB+A! */
-   const GLubyte *src = (const GLubyte *) values;
-   GLubyte *dst = (GLubyte *) rb->Data + 3 * (y * rb->RowStride + x);
-   GLuint i;
-   ASSERT(rb->Format == MESA_FORMAT_RGB888);
-   ASSERT(rb->DataType == GL_UNSIGNED_BYTE);
-   for (i = 0; i < count; i++) {
-      if (!mask || mask[i]) {
-         dst[i * 3 + 0] = src[i * 3 + 0];
-         dst[i * 3 + 1] = src[i * 3 + 1];
-         dst[i * 3 + 2] = src[i * 3 + 2];
-      }
-   }
-}
-
-
-static void
-put_mono_row_ubyte3(struct gl_context *ctx, struct gl_renderbuffer *rb, GLuint count,
-                    GLint x, GLint y, const void *value, const GLubyte *mask)
-{
-   /* note: incoming value is RGB+A! */
-   const GLubyte val0 = ((const GLubyte *) value)[0];
-   const GLubyte val1 = ((const GLubyte *) value)[1];
-   const GLubyte val2 = ((const GLubyte *) value)[2];
-   GLubyte *dst = (GLubyte *) rb->Data + 3 * (y * rb->RowStride + x);
-   ASSERT(rb->Format == MESA_FORMAT_RGB888);
-   ASSERT(rb->DataType == GL_UNSIGNED_BYTE);
-   if (!mask && val0 == val1 && val1 == val2) {
-      /* optimized case */
-      memset(dst, val0, 3 * count);
-   }
-   else {
-      GLuint i;
-      for (i = 0; i < count; i++) {
-         if (!mask || mask[i]) {
-            dst[i * 3 + 0] = val0;
-            dst[i * 3 + 1] = val1;
-            dst[i * 3 + 2] = val2;
-         }
-      }
-   }
-}
-
-
-static void
 put_values_ubyte3(struct gl_context *ctx, struct gl_renderbuffer *rb, GLuint count,
                   const GLint x[], const GLint y[], const void *values,
                   const GLubyte *mask)
@@ -674,30 +451,6 @@ put_values_ubyte3(struct gl_context *ctx, struct gl_renderbuffer *rb, GLuint cou
 }
 
 
-static void
-put_mono_values_ubyte3(struct gl_context *ctx, struct gl_renderbuffer *rb,
-                       GLuint count, const GLint x[], const GLint y[],
-                       const void *value, const GLubyte *mask)
-{
-   /* note: incoming value is RGB+A! */
-   const GLubyte val0 = ((const GLubyte *) value)[0];
-   const GLubyte val1 = ((const GLubyte *) value)[1];
-   const GLubyte val2 = ((const GLubyte *) value)[2];
-   GLuint i;
-   ASSERT(rb->Format == MESA_FORMAT_RGB888);
-   ASSERT(rb->DataType == GL_UNSIGNED_BYTE);
-   for (i = 0; i < count; i++) {
-      if (!mask || mask[i]) {
-         GLubyte *dst = ((GLubyte *) rb->Data) +
-				     3 * (y[i] * rb->RowStride + x[i]);
-         dst[0] = val0;
-         dst[1] = val1;
-         dst[2] = val2;
-      }
-   }
-}
-
-
 /**********************************************************************
  * Functions for buffers of 4 X GLubyte (or GLbyte) values.
  * Typically color buffers.
@@ -745,62 +498,6 @@ put_row_ubyte4(struct gl_context *ctx, struct gl_renderbuffer *rb, GLuint count,
 
 
 static void
-put_row_rgb_ubyte4(struct gl_context *ctx, struct gl_renderbuffer *rb, GLuint count,
-                   GLint x, GLint y, const void *values, const GLubyte *mask)
-{
-   /* Store RGB values in RGBA buffer */
-   const GLubyte *src = (const GLubyte *) values;
-   GLubyte *dst = (GLubyte *) rb->Data + 4 * (y * rb->RowStride + x);
-   GLuint i;
-   ASSERT(rb->DataType == GL_UNSIGNED_BYTE);
-   ASSERT(rb->Format == MESA_FORMAT_RGBA8888 ||
-          rb->Format == MESA_FORMAT_RGBA8888_REV);
-   for (i = 0; i < count; i++) {
-      if (!mask || mask[i]) {
-         dst[i * 4 + 0] = src[i * 3 + 0];
-         dst[i * 4 + 1] = src[i * 3 + 1];
-         dst[i * 4 + 2] = src[i * 3 + 2];
-         dst[i * 4 + 3] = 0xff;
-      }
-   }
-}
-
-
-static void
-put_mono_row_ubyte4(struct gl_context *ctx, struct gl_renderbuffer *rb, GLuint count,
-                    GLint x, GLint y, const void *value, const GLubyte *mask)
-{
-   /* treat 4*GLubyte as 1*GLuint */
-   const GLuint val = *((const GLuint *) value);
-   GLuint *dst = (GLuint *) rb->Data + (y * rb->RowStride + x);
-   ASSERT(rb->DataType == GL_UNSIGNED_BYTE);
-   ASSERT(rb->Format == MESA_FORMAT_RGBA8888 ||
-          rb->Format == MESA_FORMAT_RGBA8888_REV);
-   if (!mask && val == 0) {
-      /* common case */
-      memset(dst, 0, count * 4 * sizeof(GLubyte));
-   }
-   else {
-      /* general case */
-      if (mask) {
-         GLuint i;
-         for (i = 0; i < count; i++) {
-            if (mask[i]) {
-               dst[i] = val;
-            }
-         }
-      }
-      else {
-         GLuint i;
-         for (i = 0; i < count; i++) {
-            dst[i] = val;
-         }
-      }
-   }
-}
-
-
-static void
 put_values_ubyte4(struct gl_context *ctx, struct gl_renderbuffer *rb, GLuint count,
                   const GLint x[], const GLint y[], const void *values,
                   const GLubyte *mask)
@@ -820,26 +517,6 @@ put_values_ubyte4(struct gl_context *ctx, struct gl_renderbuffer *rb, GLuint cou
 }
 
 
-static void
-put_mono_values_ubyte4(struct gl_context *ctx, struct gl_renderbuffer *rb,
-                       GLuint count, const GLint x[], const GLint y[],
-                       const void *value, const GLubyte *mask)
-{
-   /* treat 4*GLubyte as 1*GLuint */
-   const GLuint val = *((const GLuint *) value);
-   GLuint i;
-   ASSERT(rb->DataType == GL_UNSIGNED_BYTE);
-   ASSERT(rb->Format == MESA_FORMAT_RGBA8888 ||
-          rb->Format == MESA_FORMAT_RGBA8888_REV);
-   for (i = 0; i < count; i++) {
-      if (!mask || mask[i]) {
-         GLuint *dst = (GLuint *) rb->Data + (y[i] * rb->RowStride + x[i]);
-         *dst = val;
-      }
-   }
-}
-
-
 /**********************************************************************
  * Functions for buffers of 4 X GLushort (or GLshort) values.
  * Typically accum buffer.
@@ -885,59 +562,6 @@ put_row_ushort4(struct gl_context *ctx, struct gl_renderbuffer *rb, GLuint count
 
 
 static void
-put_row_rgb_ushort4(struct gl_context *ctx, struct gl_renderbuffer *rb, GLuint count,
-                    GLint x, GLint y, const void *values, const GLubyte *mask)
-{
-   /* Put RGB values in RGBA buffer */
-   const GLushort *src = (const GLushort *) values;
-   GLushort *dst = (GLushort *) rb->Data + 4 * (y * rb->RowStride + x);
-   ASSERT(rb->DataType == GL_UNSIGNED_SHORT || rb->DataType == GL_SHORT);
-   if (mask) {
-      GLuint i;
-      for (i = 0; i < count; i++) {
-         if (mask[i]) {
-            dst[i * 4 + 0] = src[i * 3 + 0];
-            dst[i * 4 + 1] = src[i * 3 + 1];
-            dst[i * 4 + 2] = src[i * 3 + 2];
-            dst[i * 4 + 3] = 0xffff;
-         }
-      }
-   }
-   else {
-      memcpy(dst, src, 4 * count * sizeof(GLushort));
-   }
-}
-
-
-static void
-put_mono_row_ushort4(struct gl_context *ctx, struct gl_renderbuffer *rb, GLuint count,
-                     GLint x, GLint y, const void *value, const GLubyte *mask)
-{
-   const GLushort val0 = ((const GLushort *) value)[0];
-   const GLushort val1 = ((const GLushort *) value)[1];
-   const GLushort val2 = ((const GLushort *) value)[2];
-   const GLushort val3 = ((const GLushort *) value)[3];
-   GLushort *dst = (GLushort *) rb->Data + 4 * (y * rb->RowStride + x);
-   ASSERT(rb->DataType == GL_UNSIGNED_SHORT || rb->DataType == GL_SHORT);
-   if (!mask && val0 == 0 && val1 == 0 && val2 == 0 && val3 == 0) {
-      /* common case for clearing accum buffer */
-      memset(dst, 0, count * 4 * sizeof(GLushort));
-   }
-   else {
-      GLuint i;
-      for (i = 0; i < count; i++) {
-         if (!mask || mask[i]) {
-            dst[i * 4 + 0] = val0;
-            dst[i * 4 + 1] = val1;
-            dst[i * 4 + 2] = val2;
-            dst[i * 4 + 3] = val3;
-         }
-      }
-   }
-}
-
-
-static void
 put_values_ushort4(struct gl_context *ctx, struct gl_renderbuffer *rb, GLuint count,
                    const GLint x[], const GLint y[], const void *values,
                    const GLubyte *mask)
@@ -958,29 +582,6 @@ put_values_ushort4(struct gl_context *ctx, struct gl_renderbuffer *rb, GLuint co
 }
 
 
-static void
-put_mono_values_ushort4(struct gl_context *ctx, struct gl_renderbuffer *rb,
-                        GLuint count, const GLint x[], const GLint y[],
-                        const void *value, const GLubyte *mask)
-{
-   const GLushort val0 = ((const GLushort *) value)[0];
-   const GLushort val1 = ((const GLushort *) value)[1];
-   const GLushort val2 = ((const GLushort *) value)[2];
-   const GLushort val3 = ((const GLushort *) value)[3];
-   GLuint i;
-   ASSERT(rb->DataType == GL_UNSIGNED_SHORT || rb->DataType == GL_SHORT);
-   for (i = 0; i < count; i++) {
-      if (!mask || mask[i]) {
-         GLushort *dst = ((GLushort *) rb->Data) +
-				       4 * (y[i] * rb->RowStride + x[i]);
-         dst[0] = val0;
-         dst[1] = val1;
-         dst[2] = val2;
-         dst[3] = val3;
-      }
-   }
-}
-
 /**********************************************************************
  * Functions for MESA_FORMAT_R8.
  */
@@ -1241,29 +842,6 @@ put_row_a_float32(struct gl_context *ctx, struct gl_renderbuffer *rb,
 }
 
 static void
-put_mono_row_a_float32(struct gl_context *ctx, struct gl_renderbuffer *rb,
-		       GLuint count, GLint x, GLint y,
-		       const void *value, const GLubyte *mask)
-{
-   float *dst = rb->GetPointer(ctx, rb, x, y);
-   const float *src = value;
-   unsigned int i;
-
-   if (mask) {
-      for (i = 0; i < count; i++) {
-         if (mask[i]) {
-	    dst[i] = src[ACOMP];
-         }
-      }
-   }
-   else {
-      for (i = 0; i < count; i++) {
-	 dst[i] = src[ACOMP];
-      }
-   }
-}
-
-static void
 put_values_a_float32(struct gl_context *ctx, struct gl_renderbuffer *rb,
 		     GLuint count, const GLint x[], const GLint y[],
 		     const void *values, const GLubyte *mask)
@@ -1280,23 +858,6 @@ put_values_a_float32(struct gl_context *ctx, struct gl_renderbuffer *rb,
    }
 }
 
-static void
-put_mono_values_a_float32(struct gl_context *ctx,
-			  struct gl_renderbuffer *rb,
-			  GLuint count, const GLint x[], const GLint y[],
-			  const void *value, const GLubyte *mask)
-{
-   const float *src = value;
-   unsigned int i;
-
-   for (i = 0; i < count; i++) {
-      if (!mask || mask[i]) {
-	 float *dst = rb->GetPointer(ctx, rb, x[i], y[i]);
-	 *dst = src[ACOMP];
-      }
-   }
-}
-
 /**********************************************************************
  * Functions for MESA_FORMAT_R_FLOAT32.
  */
@@ -1390,10 +951,7 @@ _swrast_set_renderbuffer_accessors(struct gl_renderbuffer *rb)
       rb->GetRow = get_row_ubyte3;
       rb->GetValues = get_values_ubyte3;
       rb->PutRow = put_row_ubyte3;
-      rb->PutRowRGB = put_row_rgb_ubyte3;
-      rb->PutMonoRow = put_mono_row_ubyte3;
       rb->PutValues = put_values_ubyte3;
-      rb->PutMonoValues = put_mono_values_ubyte3;
       break;
 
    case MESA_FORMAT_RGBA8888:
@@ -1401,10 +959,7 @@ _swrast_set_renderbuffer_accessors(struct gl_renderbuffer *rb)
       rb->DataType = GL_UNSIGNED_BYTE;
       rb->GetValues = get_values_ubyte4;
       rb->PutRow = put_row_ubyte4;
-      rb->PutRowRGB = put_row_rgb_ubyte4;
-      rb->PutMonoRow = put_mono_row_ubyte4;
       rb->PutValues = put_values_ubyte4;
-      rb->PutMonoValues = put_mono_values_ubyte4;
       break;
 
    case MESA_FORMAT_R8:
@@ -1412,10 +967,7 @@ _swrast_set_renderbuffer_accessors(struct gl_renderbuffer *rb)
       rb->GetValues = get_values_r8;
       rb->GetRow = get_row_r8;
       rb->PutRow = put_row_generic;
-      rb->PutRowRGB = put_row_generic;
-      rb->PutMonoRow = put_mono_row_generic;
       rb->PutValues = put_values_generic;
-      rb->PutMonoValues = put_mono_values_generic;
       break;
 
    case MESA_FORMAT_GR88:
@@ -1423,10 +975,7 @@ _swrast_set_renderbuffer_accessors(struct gl_renderbuffer *rb)
       rb->GetValues = get_values_rg88;
       rb->GetRow = get_row_rg88;
       rb->PutRow = put_row_generic;
-      rb->PutRowRGB = put_row_generic;
-      rb->PutMonoRow = put_mono_row_generic;
       rb->PutValues = put_values_generic;
-      rb->PutMonoValues = put_mono_values_generic;
       break;
 
    case MESA_FORMAT_R16:
@@ -1434,10 +983,7 @@ _swrast_set_renderbuffer_accessors(struct gl_renderbuffer *rb)
       rb->GetValues = get_values_r16;
       rb->GetRow = get_row_r16;
       rb->PutRow = put_row_generic;
-      rb->PutRowRGB = put_row_generic;
-      rb->PutMonoRow = put_mono_row_generic;
       rb->PutValues = put_values_generic;
-      rb->PutMonoValues = put_mono_values_generic;
       break;
 
    case MESA_FORMAT_RG1616:
@@ -1445,40 +991,28 @@ _swrast_set_renderbuffer_accessors(struct gl_renderbuffer *rb)
       rb->GetValues = get_values_rg1616;
       rb->GetRow = get_row_rg1616;
       rb->PutRow = put_row_generic;
-      rb->PutRowRGB = put_row_generic;
-      rb->PutMonoRow = put_mono_row_generic;
       rb->PutValues = put_values_generic;
-      rb->PutMonoValues = put_mono_values_generic;
       break;
 
    case MESA_FORMAT_SIGNED_RGBA_16:
       rb->DataType = GL_SHORT;
       rb->GetValues = get_values_ushort4;
       rb->PutRow = put_row_ushort4;
-      rb->PutRowRGB = put_row_rgb_ushort4;
-      rb->PutMonoRow = put_mono_row_ushort4;
       rb->PutValues = put_values_ushort4;
-      rb->PutMonoValues = put_mono_values_ushort4;
       break;
 
    case MESA_FORMAT_S8:
       rb->DataType = GL_UNSIGNED_BYTE;
       rb->GetValues = get_values_ubyte;
       rb->PutRow = put_row_ubyte;
-      rb->PutRowRGB = NULL;
-      rb->PutMonoRow = put_mono_row_ubyte;
       rb->PutValues = put_values_ubyte;
-      rb->PutMonoValues = put_mono_values_ubyte;
       break;
 
    case MESA_FORMAT_Z16:
       rb->DataType = GL_UNSIGNED_SHORT;
       rb->GetValues = get_values_ushort;
       rb->PutRow = put_row_ushort;
-      rb->PutRowRGB = NULL;
-      rb->PutMonoRow = put_mono_row_ushort;
       rb->PutValues = put_values_ushort;
-      rb->PutMonoValues = put_mono_values_ushort;
       break;
 
    case MESA_FORMAT_Z32:
@@ -1487,10 +1021,7 @@ _swrast_set_renderbuffer_accessors(struct gl_renderbuffer *rb)
       rb->DataType = GL_UNSIGNED_INT;
       rb->GetValues = get_values_uint;
       rb->PutRow = put_row_uint;
-      rb->PutRowRGB = NULL;
-      rb->PutMonoRow = put_mono_row_uint;
       rb->PutValues = put_values_uint;
-      rb->PutMonoValues = put_mono_values_uint;
       break;
 
    case MESA_FORMAT_Z24_S8:
@@ -1498,70 +1029,49 @@ _swrast_set_renderbuffer_accessors(struct gl_renderbuffer *rb)
       rb->DataType = GL_UNSIGNED_INT_24_8_EXT;
       rb->GetValues = get_values_uint;
       rb->PutRow = put_row_uint;
-      rb->PutRowRGB = NULL;
-      rb->PutMonoRow = put_mono_row_uint;
       rb->PutValues = put_values_uint;
-      rb->PutMonoValues = put_mono_values_uint;
       break;
 
    case MESA_FORMAT_RGBA_FLOAT32:
       rb->GetRow = get_row_generic;
       rb->GetValues = get_values_generic;
       rb->PutRow = put_row_generic;
-      rb->PutRowRGB = NULL;
-      rb->PutMonoRow = put_mono_row_generic;
       rb->PutValues = put_values_generic;
-      rb->PutMonoValues = put_mono_values_generic;
       break;
 
    case MESA_FORMAT_INTENSITY_FLOAT32:
       rb->GetRow = get_row_i_float32;
       rb->GetValues = get_values_i_float32;
       rb->PutRow = put_row_generic;
-      rb->PutRowRGB = NULL;
-      rb->PutMonoRow = put_mono_row_generic;
       rb->PutValues = put_values_generic;
-      rb->PutMonoValues = put_mono_values_generic;
       break;
 
    case MESA_FORMAT_LUMINANCE_FLOAT32:
       rb->GetRow = get_row_l_float32;
       rb->GetValues = get_values_l_float32;
       rb->PutRow = put_row_generic;
-      rb->PutRowRGB = NULL;
-      rb->PutMonoRow = put_mono_row_generic;
       rb->PutValues = put_values_generic;
-      rb->PutMonoValues = put_mono_values_generic;
       break;
 
    case MESA_FORMAT_ALPHA_FLOAT32:
       rb->GetRow = get_row_a_float32;
       rb->GetValues = get_values_a_float32;
       rb->PutRow = put_row_a_float32;
-      rb->PutRowRGB = NULL;
-      rb->PutMonoRow = put_mono_row_a_float32;
       rb->PutValues = put_values_a_float32;
-      rb->PutMonoValues = put_mono_values_a_float32;
       break;
 
    case MESA_FORMAT_RG_FLOAT32:
       rb->GetRow = get_row_rg_float32;
       rb->GetValues = get_values_rg_float32;
       rb->PutRow = put_row_generic;
-      rb->PutRowRGB = NULL;
-      rb->PutMonoRow = put_mono_row_generic;
       rb->PutValues = put_values_generic;
-      rb->PutMonoValues = put_mono_values_generic;
       break;
 
    case MESA_FORMAT_R_FLOAT32:
       rb->GetRow = get_row_r_float32;
       rb->GetValues = get_values_r_float32;
       rb->PutRow = put_row_generic;
-      rb->PutRowRGB = NULL;
-      rb->PutMonoRow = put_mono_row_generic;
       rb->PutValues = put_values_generic;
-      rb->PutMonoValues = put_mono_values_generic;
       break;
 
    default:
@@ -1651,9 +1161,7 @@ soft_renderbuffer_storage(struct gl_context *ctx, struct gl_renderbuffer *rb,
    ASSERT(rb->GetRow);
    ASSERT(rb->GetValues);
    ASSERT(rb->PutRow);
-   ASSERT(rb->PutMonoRow);
    ASSERT(rb->PutValues);
-   ASSERT(rb->PutMonoValues);
 
    /* free old buffer storage */
    if (rb->Data) {
diff --git a/mesalib/src/mesa/swrast/s_spantemp.h b/mesalib/src/mesa/swrast/s_spantemp.h
index 517c2eb3f..2d2561b6f 100644
--- a/mesalib/src/mesa/swrast/s_spantemp.h
+++ b/mesalib/src/mesa/swrast/s_spantemp.h
@@ -25,8 +25,7 @@
 
 /*
  * Templates for the span/pixel-array write/read functions called via
- * the gl_renderbuffer's GetRow, GetValues, PutRow, PutMonoRow, PutValues
- * and PutMonoValues functions.
+ * the gl_renderbuffer's GetRow, GetValues, PutRow, and PutValues.
  *
  * Define the following macros before including this file:
  *   NAME(BASE)  to generate the function name (i.e. add prefix or suffix)
@@ -118,62 +117,6 @@ NAME(put_row)( struct gl_context *ctx, struct gl_renderbuffer *rb,
 
 
 static void
-NAME(put_row_rgb)( struct gl_context *ctx, struct gl_renderbuffer *rb,
-                   GLuint count, GLint x, GLint y,
-                   const void *values, const GLubyte mask[] )
-{
-#ifdef SPAN_VARS
-   SPAN_VARS
-#endif
-   const RB_TYPE (*src)[3] = (const RB_TYPE (*)[3]) values;
-   GLuint i;
-   INIT_PIXEL_PTR(pixel, x, y);
-   for (i = 0; i < count; i++) {
-      if (!mask || mask[i]) {
-#ifdef STORE_PIXEL_RGB
-         STORE_PIXEL_RGB(pixel, x + i, y, src[i]);
-#else
-         STORE_PIXEL(pixel, x + i, y, src[i]);
-#endif
-      }
-      INC_PIXEL_PTR(pixel);
-   }
-   (void) rb;
-   (void) ctx;
-}
-
-
-static void
-NAME(put_mono_row)( struct gl_context *ctx, struct gl_renderbuffer *rb,
-                    GLuint count, GLint x, GLint y,
-                    const void *value, const GLubyte mask[] )
-{
-#ifdef SPAN_VARS
-   SPAN_VARS
-#endif
-   const RB_TYPE *src = (const RB_TYPE *) value;
-   GLuint i;
-   INIT_PIXEL_PTR(pixel, x, y);
-   if (mask) {
-      for (i = 0; i < count; i++) {
-         if (mask[i]) {
-            STORE_PIXEL(pixel, x + i, y, src);
-         }
-         INC_PIXEL_PTR(pixel);
-      }
-   }
-   else {
-      for (i = 0; i < count; i++) {
-         STORE_PIXEL(pixel, x + i, y, src);
-         INC_PIXEL_PTR(pixel);
-      }
-   }
-   (void) rb;
-   (void) ctx;
-}
-
-
-static void
 NAME(put_values)( struct gl_context *ctx, struct gl_renderbuffer *rb,
                   GLuint count, const GLint x[], const GLint y[],
                   const void *values, const GLubyte mask[] )
@@ -195,28 +138,6 @@ NAME(put_values)( struct gl_context *ctx, struct gl_renderbuffer *rb,
 }
 
 
-static void
-NAME(put_mono_values)( struct gl_context *ctx, struct gl_renderbuffer *rb,
-                       GLuint count, const GLint x[], const GLint y[],
-                       const void *value, const GLubyte mask[] )
-{
-#ifdef SPAN_VARS
-   SPAN_VARS
-#endif
-   const RB_TYPE *src = (const RB_TYPE *) value;
-   GLuint i;
-   ASSERT(mask);
-   for (i = 0; i < count; i++) {
-      if (mask[i]) {
-         INIT_PIXEL_PTR(pixel, x[i], y[i]);
-         STORE_PIXEL(pixel, x[i], y[i], src);
-      }
-   }
-   (void) rb;
-   (void) ctx;
-}
-
-
 #undef NAME
 #undef RB_TYPE
 #undef RB_COMPONENTS
diff --git a/mesalib/src/mesa/swrast/s_stencil.c b/mesalib/src/mesa/swrast/s_stencil.c
index 101ee5056..dbcbd2be0 100644
--- a/mesalib/src/mesa/swrast/s_stencil.c
+++ b/mesalib/src/mesa/swrast/s_stencil.c
@@ -26,6 +26,8 @@
 #include "main/glheader.h"
 #include "main/context.h"
 #include "main/imports.h"
+#include "main/format_pack.h"
+#include "main/format_unpack.h"
 
 #include "s_context.h"
 #include "s_depth.h"
@@ -50,193 +52,171 @@ ENDIF
 */
 
 
+
+/**
+ * Compute/return the offset of the stencil value in a pixel.
+ * For example, if the format is Z24+S8, the position of the stencil bits
+ * within the 4-byte pixel will be either 0 or 3.
+ */
+static GLint
+get_stencil_offset(gl_format format)
+{
+   const GLubyte one = 1;
+   GLubyte pixel[MAX_PIXEL_BYTES];
+   GLint bpp = _mesa_get_format_bytes(format);
+   GLint i;
+
+   assert(_mesa_get_format_bits(format, GL_STENCIL_BITS) == 8);
+   memset(pixel, 0, sizeof(pixel));
+   _mesa_pack_ubyte_stencil_row(format, 1, &one, pixel);
+
+   for (i = 0; i < bpp; i++) {
+      if (pixel[i])
+         return i;
+   }
+
+   _mesa_problem(NULL, "get_stencil_offset() failed\n");
+   return 0;
+}
+
+
+/** Clamp the stencil value to [0, 255] */
+static inline GLubyte
+clamp(GLint val)
+{
+   if (val < 0)
+      return 0;
+   else if (val > 255)
+      return 255;
+   else
+      return val;
+}
+
+
+#define STENCIL_OP(NEW_VAL)                                                 \
+   if (invmask == 0) {                                                      \
+      for (i = j = 0; i < n; i++, j += stride) {                            \
+         if (mask[i]) {                                                     \
+            GLubyte s = stencil[j];                                         \
+            (void) s;                                                       \
+            stencil[j] = (GLubyte) (NEW_VAL);                               \
+         }                                                                  \
+      }                                                                     \
+   }                                                                        \
+   else {                                                                   \
+      for (i = j = 0; i < n; i++, j += stride) {                            \
+         if (mask[i]) {                                                     \
+            GLubyte s = stencil[j];                                         \
+            stencil[j] = (GLubyte) ((invmask & s) | (wrtmask & (NEW_VAL))); \
+         }                                                                  \
+      }                                                                     \
+   }
+
+
 /**
  * Apply the given stencil operator to the array of stencil values.
  * Don't touch stencil[i] if mask[i] is zero.
- * Input:  n - size of stencil array
- *         oper - the stencil buffer operator
- *         face - 0 or 1 for front or back face operation
- *         stencil - array of stencil values
- *         mask - array [n] of flag:  1=apply operator, 0=don't apply operator
- * Output:  stencil - modified values
+ * @param n   number of stencil values
+ * @param oper  the stencil buffer operator
+ * @param face  0 or 1 for front or back face operation
+ * @param stencil  array of stencil values (in/out)
+ * @param mask  array [n] of flag:  1=apply operator, 0=don't apply operator
+ * @param stride  stride between stencil values
  */
 static void
-apply_stencil_op( const struct gl_context *ctx, GLenum oper, GLuint face,
-                  GLuint n, GLubyte stencil[], const GLubyte mask[] )
+apply_stencil_op(const struct gl_context *ctx, GLenum oper, GLuint face,
+                 GLuint n, GLubyte stencil[], const GLubyte mask[],
+                 GLint stride)
 {
    const GLubyte ref = ctx->Stencil.Ref[face];
    const GLubyte wrtmask = ctx->Stencil.WriteMask[face];
    const GLubyte invmask = (GLubyte) (~wrtmask);
-   const GLubyte stencilMax = (1 << ctx->DrawBuffer->Visual.stencilBits) - 1;
-   GLuint i;
+   GLuint i, j;
 
    switch (oper) {
-      case GL_KEEP:
-         /* do nothing */
-         break;
-      case GL_ZERO:
-	 if (invmask==0) {
-	    for (i=0;i<n;i++) {
-	       if (mask[i]) {
-		  stencil[i] = 0;
-	       }
-	    }
-	 }
-	 else {
-	    for (i=0;i<n;i++) {
-	       if (mask[i]) {
-		  stencil[i] = (GLubyte) (stencil[i] & invmask);
-	       }
-	    }
-	 }
-	 break;
-      case GL_REPLACE:
-	 if (invmask==0) {
-	    for (i=0;i<n;i++) {
-	       if (mask[i]) {
-                  stencil[i] = ref;
-	       }
-	    }
-	 }
-	 else {
-	    for (i=0;i<n;i++) {
-	       if (mask[i]) {
-		  GLubyte s = stencil[i];
-		  stencil[i] = (GLubyte) ((invmask & s ) | (wrtmask & ref));
-	       }
-	    }
-	 }
-	 break;
-      case GL_INCR:
-	 if (invmask==0) {
-	    for (i=0;i<n;i++) {
-	       if (mask[i]) {
-		  GLubyte s = stencil[i];
-		  if (s < stencilMax) {
-		     stencil[i] = (GLubyte) (s+1);
-		  }
-	       }
-	    }
-	 }
-	 else {
-	    for (i=0;i<n;i++) {
-	       if (mask[i]) {
-		  /* VERIFY logic of adding 1 to a write-masked value */
-		  GLubyte s = stencil[i];
-		  if (s < stencilMax) {
-		     stencil[i] = (GLubyte) ((invmask & s) | (wrtmask & (s+1)));
-		  }
-	       }
-	    }
-	 }
-	 break;
-      case GL_DECR:
-	 if (invmask==0) {
-	    for (i=0;i<n;i++) {
-	       if (mask[i]) {
-		  GLubyte s = stencil[i];
-		  if (s>0) {
-		     stencil[i] = (GLubyte) (s-1);
-		  }
-	       }
-	    }
-	 }
-	 else {
-	    for (i=0;i<n;i++) {
-	       if (mask[i]) {
-		  /* VERIFY logic of subtracting 1 to a write-masked value */
-		  GLubyte s = stencil[i];
-		  if (s>0) {
-		     stencil[i] = (GLubyte) ((invmask & s) | (wrtmask & (s-1)));
-		  }
-	       }
-	    }
-	 }
-	 break;
-      case GL_INCR_WRAP_EXT:
-	 if (invmask==0) {
-	    for (i=0;i<n;i++) {
-	       if (mask[i]) {
-                  stencil[i]++;
-	       }
-	    }
-	 }
-	 else {
-	    for (i=0;i<n;i++) {
-	       if (mask[i]) {
-                  GLubyte s = stencil[i];
-                  stencil[i] = (GLubyte) ((invmask & s) | (wrtmask & (s+1)));
-	       }
-	    }
-	 }
-	 break;
-      case GL_DECR_WRAP_EXT:
-	 if (invmask==0) {
-	    for (i=0;i<n;i++) {
-	       if (mask[i]) {
-		  stencil[i]--;
-	       }
-	    }
-	 }
-	 else {
-	    for (i=0;i<n;i++) {
-	       if (mask[i]) {
-                  GLubyte s = stencil[i];
-                  stencil[i] = (GLubyte) ((invmask & s) | (wrtmask & (s-1)));
-	       }
-	    }
-	 }
-	 break;
-      case GL_INVERT:
-	 if (invmask==0) {
-	    for (i=0;i<n;i++) {
-	       if (mask[i]) {
-		  GLubyte s = stencil[i];
-		  stencil[i] = (GLubyte) ~s;
-	       }
-	    }
-	 }
-	 else {
-	    for (i=0;i<n;i++) {
-	       if (mask[i]) {
-		  GLubyte s = stencil[i];
-		  stencil[i] = (GLubyte) ((invmask & s) | (wrtmask & ~s));
-	       }
-	    }
-	 }
-	 break;
-      default:
-         _mesa_problem(ctx, "Bad stencil op in apply_stencil_op");
+   case GL_KEEP:
+      /* do nothing */
+      break;
+   case GL_ZERO:
+      /* replace stencil buf values with zero */
+      STENCIL_OP(0);
+      break;
+   case GL_REPLACE:
+      /* replace stencil buf values with ref value */
+      STENCIL_OP(ref);
+      break;
+   case GL_INCR:
+      /* increment stencil buf values, with clamping */
+      STENCIL_OP(clamp(s + 1));
+      break;
+   case GL_DECR:
+      /* increment stencil buf values, with clamping */
+      STENCIL_OP(clamp(s - 1));
+      break;
+   case GL_INCR_WRAP_EXT:
+      /* increment stencil buf values, without clamping */
+      STENCIL_OP(s + 1);
+      break;
+   case GL_DECR_WRAP_EXT:
+      /* increment stencil buf values, without clamping */
+      STENCIL_OP(s - 1);
+      break;
+   case GL_INVERT:
+      /* replace stencil buf values with inverted value */
+      STENCIL_OP(~s);
+      break;
+   default:
+      _mesa_problem(ctx, "Bad stencil op in apply_stencil_op");
    }
 }
 
 
 
+#define STENCIL_TEST(FUNC)                        \
+   for (i = j = 0; i < n; i++, j += stride) {     \
+      if (mask[i]) {                              \
+         s = (GLubyte) (stencil[j] & valueMask);  \
+         if (FUNC) {                              \
+            /* stencil pass */                    \
+            fail[i] = 0;                          \
+         }                                        \
+         else {                                   \
+            /* stencil fail */                    \
+            fail[i] = 1;                          \
+            mask[i] = 0;                          \
+         }                                        \
+      }                                           \
+      else {                                      \
+         fail[i] = 0;                             \
+      }                                           \
+   }
+
+
 
 /**
  * Apply stencil test to an array of stencil values (before depth buffering).
- * Input:  face - 0 or 1 for front or back-face polygons
- *         n - number of pixels in the array
- *         stencil - array of [n] stencil values
- *         mask - array [n] of flag:  0=skip the pixel, 1=stencil the pixel
- * Output:  mask - pixels which fail the stencil test will have their
- *                 mask flag set to 0.
- *          stencil - updated stencil values (where the test passed)
- * Return:  GL_FALSE = all pixels failed, GL_TRUE = zero or more pixels passed.
+ * For the values that fail, we'll apply the GL_STENCIL_FAIL operator to
+ * the stencil values.
+ *
+ * @param face  0 or 1 for front or back-face polygons
+ * @param n  number of pixels in the array
+ * @param stencil  array of [n] stencil values (in/out)
+ * @param mask  array [n] of flag:  0=skip the pixel, 1=stencil the pixel,
+ *              values are set to zero where the stencil test fails.
+ * @param stride  stride between stencil values
+ * @return GL_FALSE = all pixels failed, GL_TRUE = zero or more pixels passed.
  */
 static GLboolean
-do_stencil_test( struct gl_context *ctx, GLuint face, GLuint n, GLubyte stencil[],
-                 GLubyte mask[] )
+do_stencil_test(struct gl_context *ctx, GLuint face, GLuint n,
+                GLubyte stencil[], GLubyte mask[], GLint stride)
 {
    GLubyte fail[MAX_WIDTH];
    GLboolean allfail = GL_FALSE;
-   GLuint i;
+   GLuint i, j;
    const GLuint valueMask = ctx->Stencil.ValueMask[face];
-   const GLubyte r = (GLubyte) (ctx->Stencil.Ref[face] & valueMask);
+   const GLubyte ref = (GLubyte) (ctx->Stencil.Ref[face] & valueMask);
    GLubyte s;
 
-   ASSERT(n <= MAX_WIDTH);
-
    /*
     * Perform stencil test.  The results of this operation are stored
     * in the fail[] array:
@@ -247,140 +227,39 @@ do_stencil_test( struct gl_context *ctx, GLuint face, GLuint n, GLubyte stencil[
     *   ENDIF
     */
    switch (ctx->Stencil.Function[face]) {
-      case GL_NEVER:
-         /* never pass; always fail */
-         for (i=0;i<n;i++) {
-	    if (mask[i]) {
-	       mask[i] = 0;
-	       fail[i] = 1;
-	    }
-	    else {
-	       fail[i] = 0;
-	    }
-	 }
-	 allfail = GL_TRUE;
-	 break;
-      case GL_LESS:
-	 for (i=0;i<n;i++) {
-	    if (mask[i]) {
-	       s = (GLubyte) (stencil[i] & valueMask);
-	       if (r < s) {
-		  /* passed */
-		  fail[i] = 0;
-	       }
-	       else {
-		  fail[i] = 1;
-		  mask[i] = 0;
-	       }
-	    }
-	    else {
-	       fail[i] = 0;
-	    }
-	 }
-	 break;
-      case GL_LEQUAL:
-	 for (i=0;i<n;i++) {
-	    if (mask[i]) {
-	       s = (GLubyte) (stencil[i] & valueMask);
-	       if (r <= s) {
-		  /* pass */
-		  fail[i] = 0;
-	       }
-	       else {
-		  fail[i] = 1;
-		  mask[i] = 0;
-	       }
-	    }
-	    else {
-	       fail[i] = 0;
-	    }
-	 }
-	 break;
-      case GL_GREATER:
-	 for (i=0;i<n;i++) {
-	    if (mask[i]) {
-	       s = (GLubyte) (stencil[i] & valueMask);
-	       if (r > s) {
-		  /* passed */
-		  fail[i] = 0;
-	       }
-	       else {
-		  fail[i] = 1;
-		  mask[i] = 0;
-	       }
-	    }
-	    else {
-	       fail[i] = 0;
-	    }
-	 }
-	 break;
-      case GL_GEQUAL:
-	 for (i=0;i<n;i++) {
-	    if (mask[i]) {
-	       s = (GLubyte) (stencil[i] & valueMask);
-	       if (r >= s) {
-		  /* passed */
-		  fail[i] = 0;
-	       }
-	       else {
-		  fail[i] = 1;
-		  mask[i] = 0;
-	       }
-	    }
-	    else {
-	       fail[i] = 0;
-	    }
-	 }
-	 break;
-      case GL_EQUAL:
-	 for (i=0;i<n;i++) {
-	    if (mask[i]) {
-	       s = (GLubyte) (stencil[i] & valueMask);
-	       if (r == s) {
-		  /* passed */
-		  fail[i] = 0;
-	       }
-	       else {
-		  fail[i] = 1;
-		  mask[i] = 0;
-	       }
-	    }
-	    else {
-	       fail[i] = 0;
-	    }
-	 }
-	 break;
-      case GL_NOTEQUAL:
-	 for (i=0;i<n;i++) {
-	    if (mask[i]) {
-	       s = (GLubyte) (stencil[i] & valueMask);
-	       if (r != s) {
-		  /* passed */
-		  fail[i] = 0;
-	       }
-	       else {
-		  fail[i] = 1;
-		  mask[i] = 0;
-	       }
-	    }
-	    else {
-	       fail[i] = 0;
-	    }
-	 }
-	 break;
-      case GL_ALWAYS:
-	 /* always pass */
-	 for (i=0;i<n;i++) {
-	    fail[i] = 0;
-	 }
-	 break;
-      default:
-         _mesa_problem(ctx, "Bad stencil func in gl_stencil_span");
-         return 0;
+   case GL_NEVER:
+      STENCIL_TEST(0);
+      allfail = GL_TRUE;
+      break;
+   case GL_LESS:
+      STENCIL_TEST(ref < s);
+      break;
+   case GL_LEQUAL:
+      STENCIL_TEST(ref <= s);
+      break;
+   case GL_GREATER:
+      STENCIL_TEST(ref > s);
+      break;
+   case GL_GEQUAL:
+      STENCIL_TEST(ref >= s);
+      break;
+   case GL_EQUAL:
+      STENCIL_TEST(ref == s);
+      break;
+   case GL_NOTEQUAL:
+      STENCIL_TEST(ref != s);
+      break;
+   case GL_ALWAYS:
+      STENCIL_TEST(1);
+      break;
+   default:
+      _mesa_problem(ctx, "Bad stencil func in gl_stencil_span");
+      return 0;
    }
 
    if (ctx->Stencil.FailFunc[face] != GL_KEEP) {
-      apply_stencil_op( ctx, ctx->Stencil.FailFunc[face], face, n, stencil, fail );
+      apply_stencil_op(ctx, ctx->Stencil.FailFunc[face], face, n, stencil,
+                       fail, stride);
    }
 
    return !allfail;
@@ -406,54 +285,100 @@ compute_pass_fail_masks(GLuint n, const GLubyte origMask[],
 
 
 /**
- * Apply stencil and depth testing to the span of pixels.
- * Both software and hardware stencil buffers are acceptable.
- * Input:  n - number of pixels in the span
- *         x, y - location of leftmost pixel in span
- *         z - array [n] of z values
- *         mask - array [n] of flags  (1=test this pixel, 0=skip the pixel)
- * Output:  mask - array [n] of flags (1=stencil and depth test passed)
- * Return: GL_FALSE - all fragments failed the testing
- *         GL_TRUE - one or more fragments passed the testing
- *
+ * Get 8-bit stencil values from random locations in the stencil buffer.
  */
-static GLboolean
-stencil_and_ztest_span(struct gl_context *ctx, SWspan *span, GLuint face)
+static void
+get_s8_values(struct gl_context *ctx, struct gl_renderbuffer *rb,
+              GLuint count, const GLint x[], const GLint y[],
+              GLubyte stencil[])
 {
-   struct gl_framebuffer *fb = ctx->DrawBuffer;
-   struct gl_renderbuffer *rb = fb->_StencilBuffer;
-   GLubyte stencilRow[MAX_WIDTH];
-   GLubyte *stencil;
-   const GLuint n = span->end;
-   const GLint x = span->x;
-   const GLint y = span->y;
-   GLubyte *mask = span->array->mask;
+   const GLint w = rb->Width, h = rb->Height;
+   const GLubyte *map = (const GLubyte *) rb->Data;
+   GLuint i;
 
-   ASSERT((span->arrayMask & SPAN_XY) == 0);
-   ASSERT(ctx->Stencil.Enabled);
-   ASSERT(n <= MAX_WIDTH);
-#ifdef DEBUG
-   if (ctx->Depth.Test) {
-      ASSERT(span->arrayMask & SPAN_Z);
+   if (rb->Format == MESA_FORMAT_S8) {
+      const GLuint rowStride = rb->RowStride;
+      for (i = 0; i < count; i++) {
+         if (x[i] >= 0 && y[i] >= 0 && x[i] < w && y[i] < h) {
+            stencil[i] = *(map + y[i] * rowStride + x[i]);
+         }
+      }
+   }
+   else {
+      const GLuint bpp = _mesa_get_format_bytes(rb->Format);
+      const GLuint rowStride = rb->RowStride * bpp;
+      for (i = 0; i < count; i++) {
+         if (x[i] >= 0 && y[i] >= 0 && x[i] < w && y[i] < h) {
+            const GLubyte *src = map + y[i] * rowStride + x[i] * bpp;
+            _mesa_unpack_ubyte_stencil_row(rb->Format, 1, src, &stencil[i]);
+         }
+      }
    }
-#endif
+}
+
+
+/**
+ * Put 8-bit stencil values at random locations into the stencil buffer.
+ */
+static void
+put_s8_values(struct gl_context *ctx, struct gl_renderbuffer *rb,
+              GLuint count, const GLint x[], const GLint y[],
+              const GLubyte stencil[])
+{
+   const GLint w = rb->Width, h = rb->Height;
+   GLuint i;
 
-   stencil = (GLubyte *) rb->GetPointer(ctx, rb, x, y);
-   if (!stencil) {
-      rb->GetRow(ctx, rb, n, x, y, stencilRow);
-      stencil = stencilRow;
+   for (i = 0; i < count; i++) {
+      if (x[i] >= 0 && y[i] >= 0 && x[i] < w && y[i] < h) {
+         GLubyte *dst = _swrast_pixel_address(rb, x[i], y[i]);
+         _mesa_pack_ubyte_stencil_row(rb->Format, 1, &stencil[i], dst);
+      }
+   }
+}
+
+
+/**
+ * /return GL_TRUE = one or more fragments passed,
+ * GL_FALSE = all fragments failed.
+ */
+GLboolean
+_swrast_stencil_and_ztest_span(struct gl_context *ctx, SWspan *span)
+{
+   struct gl_framebuffer *fb = ctx->DrawBuffer;
+   struct gl_renderbuffer *rb = fb->Attachment[BUFFER_STENCIL].Renderbuffer;
+   const GLint stencilOffset = get_stencil_offset(rb->Format);
+   const GLint stencilStride = _mesa_get_format_bytes(rb->Format);
+   const GLuint face = (span->facing == 0) ? 0 : ctx->Stencil._BackFace;
+   const GLuint count = span->end;
+   GLubyte *mask = span->array->mask;
+   GLubyte stencilTemp[MAX_WIDTH];
+   GLubyte *stencilBuf;
+
+   if (span->arrayMask & SPAN_XY) {
+      /* read stencil values from random locations */
+      get_s8_values(ctx, rb, count, span->array->x, span->array->y,
+                    stencilTemp);
+      stencilBuf = stencilTemp;
+   }
+   else {
+      /* Processing a horizontal run of pixels.  Since stencil is always
+       * 8 bits for all MESA_FORMATs, we just need to use the right offset
+       * and stride to access them.
+       */
+      stencilBuf = _swrast_pixel_address(rb, span->x, span->y) + stencilOffset;
    }
 
    /*
     * Apply the stencil test to the fragments.
     * failMask[i] is 1 if the stencil test failed.
     */
-   if (do_stencil_test( ctx, face, n, stencil, mask ) == GL_FALSE) {
+   if (!do_stencil_test(ctx, face, count, stencilBuf, mask, stencilStride)) {
       /* all fragments failed the stencil test, we're done. */
       span->writeAll = GL_FALSE;
-      if (!rb->GetPointer(ctx, rb, 0, 0)) {
-         /* put updated stencil values into buffer */
-         rb->PutRow(ctx, rb, n, x, y, stencil, NULL);
+      if (span->arrayMask & SPAN_XY) {
+         /* need to write the updated stencil values back to the buffer */
+         put_s8_values(ctx, rb, count, span->array->x, span->array->y,
+                       stencilTemp);
       }
       return GL_FALSE;
    }
@@ -463,11 +388,12 @@ stencil_and_ztest_span(struct gl_context *ctx, SWspan *span, GLuint face)
     * and apply Zpass and Zfail stencil ops.
     */
    if (ctx->Depth.Test == GL_FALSE ||
-       ctx->DrawBuffer->_DepthBuffer == NULL) {
+       ctx->DrawBuffer->Attachment[BUFFER_DEPTH].Renderbuffer == NULL) {
       /*
        * No depth buffer, just apply zpass stencil function to active pixels.
        */
-      apply_stencil_op( ctx, ctx->Stencil.ZPassFunc[face], face, n, stencil, mask );
+      apply_stencil_op(ctx, ctx->Stencil.ZPassFunc[face], face, count,
+                       stencilBuf, mask, stencilStride);
    }
    else {
       /*
@@ -476,29 +402,28 @@ stencil_and_ztest_span(struct gl_context *ctx, SWspan *span, GLuint face)
       GLubyte passMask[MAX_WIDTH], failMask[MAX_WIDTH], origMask[MAX_WIDTH];
 
       /* save the current mask bits */
-      memcpy(origMask, mask, n * sizeof(GLubyte));
+      memcpy(origMask, mask, count * sizeof(GLubyte));
 
       /* apply the depth test */
       _swrast_depth_test_span(ctx, span);
 
-      compute_pass_fail_masks(n, origMask, mask, passMask, failMask);
+      compute_pass_fail_masks(count, origMask, mask, passMask, failMask);
 
       /* apply the pass and fail operations */
       if (ctx->Stencil.ZFailFunc[face] != GL_KEEP) {
-         apply_stencil_op( ctx, ctx->Stencil.ZFailFunc[face], face,
-                           n, stencil, failMask );
+         apply_stencil_op(ctx, ctx->Stencil.ZFailFunc[face], face,
+                          count, stencilBuf, failMask, stencilStride);
       }
       if (ctx->Stencil.ZPassFunc[face] != GL_KEEP) {
-         apply_stencil_op( ctx, ctx->Stencil.ZPassFunc[face], face,
-                           n, stencil, passMask );
+         apply_stencil_op(ctx, ctx->Stencil.ZPassFunc[face], face,
+                          count, stencilBuf, passMask, stencilStride);
       }
    }
 
-   /*
-    * Write updated stencil values back into hardware stencil buffer.
-    */
-   if (!rb->GetPointer(ctx, rb, 0, 0)) {
-      rb->PutRow(ctx, rb, n, x, y, stencil, NULL);
+   /* Write updated stencil values back into hardware stencil buffer */
+   if (span->arrayMask & SPAN_XY) {
+      put_s8_values(ctx, rb, count, span->array->x, span->array->y,
+                    stencilBuf);
    }
    
    span->writeAll = GL_FALSE;
@@ -508,530 +433,6 @@ stencil_and_ztest_span(struct gl_context *ctx, SWspan *span, GLuint face)
 
 
 
-/*
- * Return the address of a stencil buffer value given the window coords:
- */
-#define STENCIL_ADDRESS(X, Y)  (stencilStart + (Y) * stride + (X))
-
-
-
-/**
- * Apply the given stencil operator for each pixel in the array whose
- * mask flag is set.
- * \note  This is for software stencil buffers only.
- * Input:  n - number of pixels in the span
- *         x, y - array of [n] pixels
- *         operator - the stencil buffer operator
- *         mask - array [n] of flag:  1=apply operator, 0=don't apply operator
- */
-static void
-apply_stencil_op_to_pixels( struct gl_context *ctx,
-                            GLuint n, const GLint x[], const GLint y[],
-                            GLenum oper, GLuint face, const GLubyte mask[] )
-{
-   struct gl_framebuffer *fb = ctx->DrawBuffer;
-   struct gl_renderbuffer *rb = fb->_StencilBuffer;
-   const GLubyte stencilMax = (1 << fb->Visual.stencilBits) - 1;
-   const GLubyte ref = ctx->Stencil.Ref[face];
-   const GLubyte wrtmask = ctx->Stencil.WriteMask[face];
-   const GLubyte invmask = (GLubyte) (~wrtmask);
-   GLuint i;
-   GLubyte *stencilStart = (GLubyte *) rb->Data;
-   const GLuint stride = rb->Width;
-
-   ASSERT(rb->GetPointer(ctx, rb, 0, 0));
-
-   switch (oper) {
-      case GL_KEEP:
-         /* do nothing */
-         break;
-      case GL_ZERO:
-	 if (invmask==0) {
-	    for (i=0;i<n;i++) {
-	       if (mask[i]) {
-                  GLubyte *sptr = STENCIL_ADDRESS( x[i], y[i] );
-                  *sptr = 0;
-	       }
-	    }
-	 }
-	 else {
-	    for (i=0;i<n;i++) {
-	       if (mask[i]) {
-                  GLubyte *sptr = STENCIL_ADDRESS( x[i], y[i] );
-		  *sptr = (GLubyte) (invmask & *sptr);
-	       }
-	    }
-	 }
-	 break;
-      case GL_REPLACE:
-	 if (invmask==0) {
-	    for (i=0;i<n;i++) {
-	       if (mask[i]) {
-                  GLubyte *sptr = STENCIL_ADDRESS( x[i], y[i] );
-                  *sptr = ref;
-	       }
-	    }
-	 }
-	 else {
-	    for (i=0;i<n;i++) {
-	       if (mask[i]) {
-                  GLubyte *sptr = STENCIL_ADDRESS( x[i], y[i] );
-		  *sptr = (GLubyte) ((invmask & *sptr ) | (wrtmask & ref));
-	       }
-	    }
-	 }
-	 break;
-      case GL_INCR:
-	 if (invmask==0) {
-	    for (i=0;i<n;i++) {
-	       if (mask[i]) {
-                  GLubyte *sptr = STENCIL_ADDRESS( x[i], y[i] );
-		  if (*sptr < stencilMax) {
-		     *sptr = (GLubyte) (*sptr + 1);
-		  }
-	       }
-	    }
-	 }
-	 else {
-	    for (i=0;i<n;i++) {
-	       if (mask[i]) {
-                  GLubyte *sptr = STENCIL_ADDRESS( x[i], y[i] );
-		  if (*sptr < stencilMax) {
-		     *sptr = (GLubyte) ((invmask & *sptr) | (wrtmask & (*sptr+1)));
-		  }
-	       }
-	    }
-	 }
-	 break;
-      case GL_DECR:
-	 if (invmask==0) {
-	    for (i=0;i<n;i++) {
-	       if (mask[i]) {
-                  GLubyte *sptr = STENCIL_ADDRESS( x[i], y[i] );
-		  if (*sptr>0) {
-		     *sptr = (GLubyte) (*sptr - 1);
-		  }
-	       }
-	    }
-	 }
-	 else {
-	    for (i=0;i<n;i++) {
-	       if (mask[i]) {
-                  GLubyte *sptr = STENCIL_ADDRESS( x[i], y[i] );
-		  if (*sptr>0) {
-		     *sptr = (GLubyte) ((invmask & *sptr) | (wrtmask & (*sptr-1)));
-		  }
-	       }
-	    }
-	 }
-	 break;
-      case GL_INCR_WRAP_EXT:
-	 if (invmask==0) {
-	    for (i=0;i<n;i++) {
-	       if (mask[i]) {
-                  GLubyte *sptr = STENCIL_ADDRESS( x[i], y[i] );
-                  *sptr = (GLubyte) (*sptr + 1);
-	       }
-	    }
-	 }
-	 else {
-	    for (i=0;i<n;i++) {
-	       if (mask[i]) {
-                  GLubyte *sptr = STENCIL_ADDRESS( x[i], y[i] );
-                  *sptr = (GLubyte) ((invmask & *sptr) | (wrtmask & (*sptr+1)));
-	       }
-	    }
-	 }
-	 break;
-      case GL_DECR_WRAP_EXT:
-	 if (invmask==0) {
-	    for (i=0;i<n;i++) {
-	       if (mask[i]) {
-                  GLubyte *sptr = STENCIL_ADDRESS( x[i], y[i] );
-                  *sptr = (GLubyte) (*sptr - 1);
-	       }
-	    }
-	 }
-	 else {
-	    for (i=0;i<n;i++) {
-	       if (mask[i]) {
-                  GLubyte *sptr = STENCIL_ADDRESS( x[i], y[i] );
-                  *sptr = (GLubyte) ((invmask & *sptr) | (wrtmask & (*sptr-1)));
-	       }
-	    }
-	 }
-	 break;
-      case GL_INVERT:
-	 if (invmask==0) {
-	    for (i=0;i<n;i++) {
-	       if (mask[i]) {
-                  GLubyte *sptr = STENCIL_ADDRESS( x[i], y[i] );
-                  *sptr = (GLubyte) (~*sptr);
-	       }
-	    }
-	 }
-	 else {
-	    for (i=0;i<n;i++) {
-	       if (mask[i]) {
-                  GLubyte *sptr = STENCIL_ADDRESS( x[i], y[i] );
-                  *sptr = (GLubyte) ((invmask & *sptr) | (wrtmask & ~*sptr));
-	       }
-	    }
-	 }
-	 break;
-      default:
-         _mesa_problem(ctx, "Bad stencilop in apply_stencil_op_to_pixels");
-   }
-}
-
-
-
-/**
- * Apply stencil test to an array of pixels before depth buffering.
- *
- * \note Used for software stencil buffer only.
- * Input:  n - number of pixels in the span
- *         x, y - array of [n] pixels to stencil
- *         mask - array [n] of flag:  0=skip the pixel, 1=stencil the pixel
- * Output:  mask - pixels which fail the stencil test will have their
- *                 mask flag set to 0.
- * \return  GL_FALSE = all pixels failed, GL_TRUE = zero or more pixels passed.
- */
-static GLboolean
-stencil_test_pixels( struct gl_context *ctx, GLuint face, GLuint n,
-                     const GLint x[], const GLint y[], GLubyte mask[] )
-{
-   const struct gl_framebuffer *fb = ctx->DrawBuffer;
-   struct gl_renderbuffer *rb = fb->_StencilBuffer;
-   GLubyte fail[MAX_WIDTH];
-   GLubyte r, s;
-   GLuint i;
-   GLboolean allfail = GL_FALSE;
-   const GLuint valueMask = ctx->Stencil.ValueMask[face];
-   const GLubyte *stencilStart = (GLubyte *) rb->Data;
-   const GLuint stride = rb->Width;
-
-   ASSERT(rb->GetPointer(ctx, rb, 0, 0));
-
-   /*
-    * Perform stencil test.  The results of this operation are stored
-    * in the fail[] array:
-    *   IF fail[i] is non-zero THEN
-    *       the stencil fail operator is to be applied
-    *   ELSE
-    *       the stencil fail operator is not to be applied
-    *   ENDIF
-    */
-
-   switch (ctx->Stencil.Function[face]) {
-      case GL_NEVER:
-         /* always fail */
-         for (i=0;i<n;i++) {
-	    if (mask[i]) {
-	       mask[i] = 0;
-	       fail[i] = 1;
-	    }
-	    else {
-	       fail[i] = 0;
-	    }
-	 }
-	 allfail = GL_TRUE;
-	 break;
-      case GL_LESS:
-	 r = (GLubyte) (ctx->Stencil.Ref[face] & valueMask);
-	 for (i=0;i<n;i++) {
-	    if (mask[i]) {
-               const GLubyte *sptr = STENCIL_ADDRESS(x[i],y[i]);
-	       s = (GLubyte) (*sptr & valueMask);
-	       if (r < s) {
-		  /* passed */
-		  fail[i] = 0;
-	       }
-	       else {
-		  fail[i] = 1;
-		  mask[i] = 0;
-	       }
-	    }
-	    else {
-	       fail[i] = 0;
-	    }
-	 }
-	 break;
-      case GL_LEQUAL:
-	 r = (GLubyte) (ctx->Stencil.Ref[face] & valueMask);
-	 for (i=0;i<n;i++) {
-	    if (mask[i]) {
-               const GLubyte *sptr = STENCIL_ADDRESS(x[i],y[i]);
-	       s = (GLubyte) (*sptr & valueMask);
-	       if (r <= s) {
-		  /* pass */
-		  fail[i] = 0;
-	       }
-	       else {
-		  fail[i] = 1;
-		  mask[i] = 0;
-	       }
-	    }
-	    else {
-	       fail[i] = 0;
-	    }
-	 }
-	 break;
-      case GL_GREATER:
-	 r = (GLubyte) (ctx->Stencil.Ref[face] & valueMask);
-	 for (i=0;i<n;i++) {
-	    if (mask[i]) {
-               const GLubyte *sptr = STENCIL_ADDRESS(x[i],y[i]);
-	       s = (GLubyte) (*sptr & valueMask);
-	       if (r > s) {
-		  /* passed */
-		  fail[i] = 0;
-	       }
-	       else {
-		  fail[i] = 1;
-		  mask[i] = 0;
-	       }
-	    }
-	    else {
-	       fail[i] = 0;
-	    }
-	 }
-	 break;
-      case GL_GEQUAL:
-	 r = (GLubyte) (ctx->Stencil.Ref[face] & valueMask);
-	 for (i=0;i<n;i++) {
-	    if (mask[i]) {
-               const GLubyte *sptr = STENCIL_ADDRESS(x[i],y[i]);
-	       s = (GLubyte) (*sptr & valueMask);
-	       if (r >= s) {
-		  /* passed */
-		  fail[i] = 0;
-	       }
-	       else {
-		  fail[i] = 1;
-		  mask[i] = 0;
-	       }
-	    }
-	    else {
-	       fail[i] = 0;
-	    }
-	 }
-	 break;
-      case GL_EQUAL:
-	 r = (GLubyte) (ctx->Stencil.Ref[face] & valueMask);
-	 for (i=0;i<n;i++) {
-	    if (mask[i]) {
-               const GLubyte *sptr = STENCIL_ADDRESS(x[i],y[i]);
-	       s = (GLubyte) (*sptr & valueMask);
-	       if (r == s) {
-		  /* passed */
-		  fail[i] = 0;
-	       }
-	       else {
-		  fail[i] = 1;
-		  mask[i] = 0;
-	       }
-	    }
-	    else {
-	       fail[i] = 0;
-	    }
-	 }
-	 break;
-      case GL_NOTEQUAL:
-	 r = (GLubyte) (ctx->Stencil.Ref[face] & valueMask);
-	 for (i=0;i<n;i++) {
-	    if (mask[i]) {
-               const GLubyte *sptr = STENCIL_ADDRESS(x[i],y[i]);
-	       s = (GLubyte) (*sptr & valueMask);
-	       if (r != s) {
-		  /* passed */
-		  fail[i] = 0;
-	       }
-	       else {
-		  fail[i] = 1;
-		  mask[i] = 0;
-	       }
-	    }
-	    else {
-	       fail[i] = 0;
-	    }
-	 }
-	 break;
-      case GL_ALWAYS:
-	 /* always pass */
-	 for (i=0;i<n;i++) {
-	    fail[i] = 0;
-	 }
-	 break;
-      default:
-         _mesa_problem(ctx, "Bad stencil func in gl_stencil_pixels");
-         return 0;
-   }
-
-   if (ctx->Stencil.FailFunc[face] != GL_KEEP) {
-      apply_stencil_op_to_pixels( ctx, n, x, y, ctx->Stencil.FailFunc[face],
-                                  face, fail );
-   }
-
-   return !allfail;
-}
-
-
-
-
-/**
- * Apply stencil and depth testing to an array of pixels.
- * This is used both for software and hardware stencil buffers.
- *
- * The comments in this function are a bit sparse but the code is
- * almost identical to stencil_and_ztest_span(), which is well
- * commented.
- *
- * Input:  n - number of pixels in the array
- *         x, y - array of [n] pixel positions
- *         z - array [n] of z values
- *         mask - array [n] of flags  (1=test this pixel, 0=skip the pixel)
- * Output: mask - array [n] of flags (1=stencil and depth test passed)
- * Return: GL_FALSE - all fragments failed the testing
- *         GL_TRUE - one or more fragments passed the testing
- */
-static GLboolean
-stencil_and_ztest_pixels( struct gl_context *ctx, SWspan *span, GLuint face )
-{
-   GLubyte passMask[MAX_WIDTH], failMask[MAX_WIDTH], origMask[MAX_WIDTH];
-   struct gl_framebuffer *fb = ctx->DrawBuffer;
-   struct gl_renderbuffer *rb = fb->_StencilBuffer;
-   const GLuint n = span->end;
-   const GLint *x = span->array->x;
-   const GLint *y = span->array->y;
-   GLubyte *mask = span->array->mask;
-
-   ASSERT(span->arrayMask & SPAN_XY);
-   ASSERT(ctx->Stencil.Enabled);
-   ASSERT(n <= MAX_WIDTH);
-
-   if (!rb->GetPointer(ctx, rb, 0, 0)) {
-      /* No direct access */
-      GLubyte stencil[MAX_WIDTH];
-
-      ASSERT(rb->DataType == GL_UNSIGNED_BYTE);
-      _swrast_get_values(ctx, rb, n, x, y, stencil, sizeof(GLubyte));
-
-      memcpy(origMask, mask, n * sizeof(GLubyte));          
-
-      (void) do_stencil_test(ctx, face, n, stencil, mask);
-
-      if (ctx->Depth.Test == GL_FALSE) {
-         apply_stencil_op(ctx, ctx->Stencil.ZPassFunc[face], face,
-                          n, stencil, mask);
-      }
-      else {
-         GLubyte tmpMask[MAX_WIDTH]; 
-         memcpy(tmpMask, mask, n * sizeof(GLubyte));
-
-         _swrast_depth_test_span(ctx, span);
-
-         compute_pass_fail_masks(n, tmpMask, mask, passMask, failMask);
-
-         if (ctx->Stencil.ZFailFunc[face] != GL_KEEP) {
-            apply_stencil_op(ctx, ctx->Stencil.ZFailFunc[face], face,
-                             n, stencil, failMask);
-         }
-         if (ctx->Stencil.ZPassFunc[face] != GL_KEEP) {
-            apply_stencil_op(ctx, ctx->Stencil.ZPassFunc[face], face,
-                             n, stencil, passMask);
-         }
-      }
-
-      /* Write updated stencil values into hardware stencil buffer */
-      rb->PutValues(ctx, rb, n, x, y, stencil, origMask);
-
-      return GL_TRUE;
-   }
-   else {
-      /* Direct access to stencil buffer */
-
-      if (stencil_test_pixels(ctx, face, n, x, y, mask) == GL_FALSE) {
-         /* all fragments failed the stencil test, we're done. */
-         return GL_FALSE;
-      }
-
-      if (ctx->Depth.Test==GL_FALSE) {
-         apply_stencil_op_to_pixels(ctx, n, x, y,
-                                    ctx->Stencil.ZPassFunc[face], face, mask);
-      }
-      else {
-         memcpy(origMask, mask, n * sizeof(GLubyte));
-
-         _swrast_depth_test_span(ctx, span);
-
-         compute_pass_fail_masks(n, origMask, mask, passMask, failMask);
-
-         if (ctx->Stencil.ZFailFunc[face] != GL_KEEP) {
-            apply_stencil_op_to_pixels(ctx, n, x, y,
-                                       ctx->Stencil.ZFailFunc[face],
-                                       face, failMask);
-         }
-         if (ctx->Stencil.ZPassFunc[face] != GL_KEEP) {
-            apply_stencil_op_to_pixels(ctx, n, x, y,
-                                       ctx->Stencil.ZPassFunc[face],
-                                       face, passMask);
-         }
-      }
-
-      return GL_TRUE;  /* one or more fragments passed both tests */
-   }
-}
-
-
-/**
- * /return GL_TRUE = one or more fragments passed,
- * GL_FALSE = all fragments failed.
- */
-GLboolean
-_swrast_stencil_and_ztest_span(struct gl_context *ctx, SWspan *span)
-{
-   const GLuint face = (span->facing == 0) ? 0 : ctx->Stencil._BackFace;
-
-   if (span->arrayMask & SPAN_XY)
-      return stencil_and_ztest_pixels(ctx, span, face);
-   else
-      return stencil_and_ztest_span(ctx, span, face);
-}
-
-
-#if 0
-GLuint
-clip_span(GLuint bufferWidth, GLuint bufferHeight,
-          GLint x, GLint y, GLuint *count)
-{
-   GLuint n = *count;
-   GLuint skipPixels = 0;
-
-   if (y < 0 || y >= bufferHeight || x + n <= 0 || x >= bufferWidth) {
-      /* totally out of bounds */
-      n = 0;
-   }
-   else {
-      /* left clip */
-      if (x < 0) {
-         skipPixels = -x;
-         x = 0;
-         n -= skipPixels;
-      }
-      /* right clip */
-      if (x + n > bufferWidth) {
-         GLint dx = x + n - bufferWidth;
-         n -= dx;
-      }
-   }
-
-   *count = n;
-
-   return skipPixels;
-}
-#endif
-
 
 /**
  * Return a span of stencil values from the stencil buffer.
@@ -1044,6 +445,8 @@ void
 _swrast_read_stencil_span(struct gl_context *ctx, struct gl_renderbuffer *rb,
                           GLint n, GLint x, GLint y, GLubyte stencil[])
 {
+   GLubyte *src;
+
    if (y < 0 || y >= (GLint) rb->Height ||
        x + n <= 0 || x >= (GLint) rb->Width) {
       /* span is completely outside framebuffer */
@@ -1064,7 +467,8 @@ _swrast_read_stencil_span(struct gl_context *ctx, struct gl_renderbuffer *rb,
       return;
    }
 
-   rb->GetRow(ctx, rb, n, x, y, stencil);
+   src = _swrast_pixel_address(rb, x, y);
+   _mesa_unpack_ubyte_stencil_row(rb->Format, n, src, stencil);
 }
 
 
@@ -1082,9 +486,10 @@ _swrast_write_stencil_span(struct gl_context *ctx, GLint n, GLint x, GLint y,
                            const GLubyte stencil[] )
 {
    struct gl_framebuffer *fb = ctx->DrawBuffer;
-   struct gl_renderbuffer *rb = fb->_StencilBuffer;
+   struct gl_renderbuffer *rb = fb->Attachment[BUFFER_STENCIL].Renderbuffer;
    const GLuint stencilMax = (1 << fb->Visual.stencilBits) - 1;
    const GLuint stencilMask = ctx->Stencil.WriteMask[0];
+   GLubyte *stencilBuf;
 
    if (y < 0 || y >= (GLint) rb->Height ||
        x + n <= 0 || x >= (GLint) rb->Width) {
@@ -1105,140 +510,129 @@ _swrast_write_stencil_span(struct gl_context *ctx, GLint n, GLint x, GLint y,
       return;
    }
 
+   stencilBuf = _swrast_pixel_address(rb, x, y);
+
    if ((stencilMask & stencilMax) != stencilMax) {
       /* need to apply writemask */
       GLubyte destVals[MAX_WIDTH], newVals[MAX_WIDTH];
       GLint i;
-      rb->GetRow(ctx, rb, n, x, y, destVals);
+
+      _mesa_unpack_ubyte_stencil_row(rb->Format, n, stencilBuf, destVals);
       for (i = 0; i < n; i++) {
          newVals[i]
             = (stencil[i] & stencilMask) | (destVals[i] & ~stencilMask);
       }
-      rb->PutRow(ctx, rb, n, x, y, newVals, NULL);
+      _mesa_pack_ubyte_stencil_row(rb->Format, n, newVals, stencilBuf);
    }
    else {
-      rb->PutRow(ctx, rb, n, x, y, stencil, NULL);
+      _mesa_pack_ubyte_stencil_row(rb->Format, n, stencil, stencilBuf);
    }
 }
 
 
 
 /**
- * Clear the stencil buffer.
+ * Clear the stencil buffer.  If the buffer is a combined
+ * depth+stencil buffer, only the stencil bits will be touched.
  */
 void
-_swrast_clear_stencil_buffer( struct gl_context *ctx, struct gl_renderbuffer *rb )
+_swrast_clear_stencil_buffer(struct gl_context *ctx)
 {
+   struct gl_renderbuffer *rb =
+      ctx->DrawBuffer->Attachment[BUFFER_STENCIL].Renderbuffer;
    const GLubyte stencilBits = ctx->DrawBuffer->Visual.stencilBits;
-   const GLuint mask = ctx->Stencil.WriteMask[0];
-   const GLuint invMask = ~mask;
-   const GLuint clearVal = (ctx->Stencil.Clear & mask);
+   const GLuint writeMask = ctx->Stencil.WriteMask[0];
    const GLuint stencilMax = (1 << stencilBits) - 1;
    GLint x, y, width, height;
+   GLubyte *map;
+   GLint rowStride, i, j;
+   GLbitfield mapMode;
 
-   if (!rb || mask == 0)
+   if (!rb || writeMask == 0)
       return;
 
-   ASSERT(rb->DataType == GL_UNSIGNED_BYTE ||
-          rb->DataType == GL_UNSIGNED_SHORT);
-
-   ASSERT(rb->_BaseFormat == GL_STENCIL_INDEX);
-
    /* compute region to clear */
    x = ctx->DrawBuffer->_Xmin;
    y = ctx->DrawBuffer->_Ymin;
    width  = ctx->DrawBuffer->_Xmax - ctx->DrawBuffer->_Xmin;
    height = ctx->DrawBuffer->_Ymax - ctx->DrawBuffer->_Ymin;
 
-   if (rb->GetPointer(ctx, rb, 0, 0)) {
-      /* Direct buffer access */
-      if ((mask & stencilMax) != stencilMax) {
-         /* need to mask the clear */
-         if (rb->DataType == GL_UNSIGNED_BYTE) {
-            GLint i, j;
-            for (i = 0; i < height; i++) {
-               GLubyte *stencil = (GLubyte*) rb->GetPointer(ctx, rb, x, y + i);
-               for (j = 0; j < width; j++) {
-                  stencil[j] = (stencil[j] & invMask) | clearVal;
-               }
-            }
-         }
-         else {
-            GLint i, j;
+   mapMode = GL_MAP_WRITE_BIT;
+   if ((writeMask & stencilMax) != stencilMax) {
+      /* need to mask stencil values */
+      mapMode |= GL_MAP_READ_BIT;
+   }
+   else if (_mesa_get_format_bits(rb->Format, GL_DEPTH_BITS) > 0) {
+      /* combined depth+stencil, need to mask Z values */
+      mapMode |= GL_MAP_READ_BIT;
+   }
+
+   ctx->Driver.MapRenderbuffer(ctx, rb, x, y, width, height,
+                               mapMode, &map, &rowStride);
+   if (!map) {
+      _mesa_error(ctx, GL_OUT_OF_MEMORY, "glClear(stencil)");
+      return;
+   }
+
+   switch (rb->Format) {
+   case MESA_FORMAT_S8:
+      {
+         GLubyte clear = ctx->Stencil.Clear & writeMask & 0xff;
+         GLubyte mask = (~writeMask) & 0xff;
+         if (mask != 0) {
+            /* masked clear */
             for (i = 0; i < height; i++) {
-               GLushort *stencil = (GLushort*) rb->GetPointer(ctx, rb, x, y + i);
+               GLubyte *row = map;
                for (j = 0; j < width; j++) {
-                  stencil[j] = (stencil[j] & invMask) | clearVal;
+                  row[j] = (row[j] & mask) | clear;
                }
+               map += rowStride;
             }
          }
-      }
-      else {
-         /* no bit masking */
-         if (width == (GLint) rb->Width && rb->DataType == GL_UNSIGNED_BYTE) {
-            /* optimized case */
-            /* Note: bottom-to-top raster assumed! */
-            GLubyte *stencil = (GLubyte *) rb->GetPointer(ctx, rb, x, y);
-            GLuint len = width * height * sizeof(GLubyte);
-            memset(stencil, clearVal, len);
+         else if (rowStride == width) {
+            /* clear whole buffer */
+            memset(map, clear, width * height);
          }
          else {
-            /* general case */
-            GLint i;
+            /* clear scissored */
             for (i = 0; i < height; i++) {
-               GLvoid *stencil = rb->GetPointer(ctx, rb, x, y + i);
-               if (rb->DataType == GL_UNSIGNED_BYTE) {
-                  memset(stencil, clearVal, width);
-               }
-               else {
-                  _mesa_memset16((short unsigned int*) stencil, clearVal, width);
-               }
+               memset(map, clear, width);
+               map += rowStride;
             }
          }
       }
-   }
-   else {
-      /* no direct access */
-      if ((mask & stencilMax) != stencilMax) {
-         /* need to mask the clear */
-         if (rb->DataType == GL_UNSIGNED_BYTE) {
-            GLint i, j;
-            for (i = 0; i < height; i++) {
-               GLubyte stencil[MAX_WIDTH];
-               rb->GetRow(ctx, rb, width, x, y + i, stencil);
-               for (j = 0; j < width; j++) {
-                  stencil[j] = (stencil[j] & invMask) | clearVal;
-               }
-               rb->PutRow(ctx, rb, width, x, y + i, stencil, NULL);
-            }
-         }
-         else {
-            GLint i, j;
-            for (i = 0; i < height; i++) {
-               GLushort stencil[MAX_WIDTH];
-               rb->GetRow(ctx, rb, width, x, y + i, stencil);
-               for (j = 0; j < width; j++) {
-                  stencil[j] = (stencil[j] & invMask) | clearVal;
-               }
-               rb->PutRow(ctx, rb, width, x, y + i, stencil, NULL);
+      break;
+   case MESA_FORMAT_S8_Z24:
+      {
+         GLuint clear = (ctx->Stencil.Clear & writeMask & 0xff) << 24;
+         GLuint mask = (((~writeMask) & 0xff) << 24) | 0xffffff;
+         for (i = 0; i < height; i++) {
+            GLuint *row = (GLuint *) map;
+            for (j = 0; j < width; j++) {
+               row[j] = (row[j] & mask) | clear;
             }
+            map += rowStride;
          }
       }
-      else {
-         /* no bit masking */
-         const GLubyte clear8 = (GLubyte) clearVal;
-         const GLushort clear16 = (GLushort) clearVal;
-         const void *clear;
-         GLint i;
-         if (rb->DataType == GL_UNSIGNED_BYTE) {
-            clear = &clear8;
-         }
-         else {
-            clear = &clear16;
-         }
+      break;
+   case MESA_FORMAT_Z24_S8:
+      {
+         GLuint clear = ctx->Stencil.Clear & writeMask & 0xff;
+         GLuint mask = 0xffffff00 | ((~writeMask) & 0xff);
          for (i = 0; i < height; i++) {
-            rb->PutMonoRow(ctx, rb, width, x, y + i, clear, NULL);
+            GLuint *row = (GLuint *) map;
+            for (j = 0; j < width; j++) {
+               row[j] = (row[j] & mask) | clear;
+            }
+            map += rowStride;
          }
       }
+      break;
+   default:
+      _mesa_problem(ctx, "Unexpected stencil buffer format %s"
+                    " in _swrast_clear_stencil_buffer()",
+                    _mesa_get_format_name(rb->Format));
    }
+
+   ctx->Driver.UnmapRenderbuffer(ctx, rb);
 }
diff --git a/mesalib/src/mesa/swrast/s_stencil.h b/mesalib/src/mesa/swrast/s_stencil.h
index 37f3c8da1..113649a37 100644
--- a/mesalib/src/mesa/swrast/s_stencil.h
+++ b/mesalib/src/mesa/swrast/s_stencil.h
@@ -47,7 +47,7 @@ _swrast_write_stencil_span( struct gl_context *ctx, GLint n, GLint x, GLint y,
 
 
 extern void
-_swrast_clear_stencil_buffer( struct gl_context *ctx, struct gl_renderbuffer *rb );
+_swrast_clear_stencil_buffer(struct gl_context *ctx);
 
 
 #endif
diff --git a/mesalib/src/mesa/swrast/s_texrender.c b/mesalib/src/mesa/swrast/s_texrender.c
index 3734ce6d5..523420205 100644
--- a/mesalib/src/mesa/swrast/s_texrender.c
+++ b/mesalib/src/mesa/swrast/s_texrender.c
@@ -235,125 +235,6 @@ texture_put_row(struct gl_context *ctx, struct gl_renderbuffer *rb, GLuint count
    }
 }
 
-/**
- * Put row of RGB values into a renderbuffer that wraps a texture image.
- */
-static void
-texture_put_row_rgb(struct gl_context *ctx, struct gl_renderbuffer *rb, GLuint count,
-                GLint x, GLint y, const void *values, const GLubyte *mask)
-{
-   struct texture_renderbuffer *trb = texture_renderbuffer(rb);
-   const GLint z = trb->Zoffset;
-   GLuint i;
-
-   y += trb->Yoffset;
-
-   if (rb->DataType == CHAN_TYPE) {
-      const GLchan *rgb = (const GLchan *) values;
-      for (i = 0; i < count; i++) {
-         if (!mask || mask[i]) {
-            trb->Store(trb->TexImage, x + i, y, z, rgb);
-         }
-         rgb += 3;
-      }
-   }
-   else if (rb->DataType == GL_UNSIGNED_SHORT) {
-      const GLushort *zValues = (const GLushort *) values;
-      for (i = 0; i < count; i++) {
-         if (!mask || mask[i]) {
-            trb->Store(trb->TexImage, x + i, y, z, zValues + i);
-         }
-      }
-   }
-   else if (rb->DataType == GL_UNSIGNED_INT) {
-      const GLuint *zValues = (const GLuint *) values;
-      for (i = 0; i < count; i++) {
-         if (!mask || mask[i]) {
-            trb->Store(trb->TexImage, x + i, y, z, zValues + i);
-         }
-      }
-   }
-   else if (rb->DataType == GL_UNSIGNED_INT_24_8_EXT) {
-      const GLuint *zValues = (const GLuint *) values;
-      for (i = 0; i < count; i++) {
-         if (!mask || mask[i]) {
-            GLfloat flt = (GLfloat) ((zValues[i] >> 8) * (1.0 / 0xffffff));
-            trb->Store(trb->TexImage, x + i, y, z, &flt);
-         }
-      }
-   }
-   else if (rb->DataType == GL_UNSIGNED_INT_8_24_REV_MESA) {
-      const GLuint *zValues = (const GLuint *) values;
-      for (i = 0; i < count; i++) {
-         if (!mask || mask[i]) {
-            GLfloat flt = (GLfloat) ((zValues[i] & 0xffffff) * (1.0 / 0xffffff));
-            trb->Store(trb->TexImage, x + i, y, z, &flt);
-         }
-      }
-   }
-   else {
-      _mesa_problem(ctx, "invalid rb->DataType in texture_put_row");
-   }
-}
-
-
-static void
-texture_put_mono_row(struct gl_context *ctx, struct gl_renderbuffer *rb, GLuint count,
-                     GLint x, GLint y, const void *value, const GLubyte *mask)
-{
-   struct texture_renderbuffer *trb = texture_renderbuffer(rb);
-   const GLint z = trb->Zoffset;
-   GLuint i;
-
-   y += trb->Yoffset;
-
-   if (rb->DataType == CHAN_TYPE) {
-      const GLchan *rgba = (const GLchan *) value;
-      for (i = 0; i < count; i++) {
-         if (!mask || mask[i]) {
-            trb->Store(trb->TexImage, x + i, y, z, rgba);
-         }
-      }
-   }
-   else if (rb->DataType == GL_UNSIGNED_SHORT) {
-      const GLushort zValue = *((const GLushort *) value);
-      for (i = 0; i < count; i++) {
-         if (!mask || mask[i]) {
-            trb->Store(trb->TexImage, x + i, y, z, &zValue);
-         }
-      }
-   }
-   else if (rb->DataType == GL_UNSIGNED_INT) {
-      const GLuint zValue = *((const GLuint *) value);
-      for (i = 0; i < count; i++) {
-         if (!mask || mask[i]) {
-            trb->Store(trb->TexImage, x + i, y, z, &zValue);
-         }
-      }
-   }
-   else if (rb->DataType == GL_UNSIGNED_INT_24_8_EXT) {
-      const GLuint zValue = *((const GLuint *) value);
-      const GLfloat flt = (GLfloat) ((zValue >> 8) * (1.0 / 0xffffff));
-      for (i = 0; i < count; i++) {
-         if (!mask || mask[i]) {
-            trb->Store(trb->TexImage, x + i, y, z, &flt);
-         }
-      }
-   }
-   else if (rb->DataType == GL_UNSIGNED_INT_8_24_REV_MESA) {
-      const GLuint zValue = *((const GLuint *) value);
-      const GLfloat flt = (GLfloat) ((zValue & 0xffffff) * (1.0 / 0xffffff));
-      for (i = 0; i < count; i++) {
-         if (!mask || mask[i]) {
-            trb->Store(trb->TexImage, x + i, y, z, &flt);
-         }
-      }
-   }
-   else {
-      _mesa_problem(ctx, "invalid rb->DataType in texture_put_mono_row");
-   }
-}
-
 
 static void
 texture_put_values(struct gl_context *ctx, struct gl_renderbuffer *rb, GLuint count,
@@ -414,63 +295,6 @@ texture_put_values(struct gl_context *ctx, struct gl_renderbuffer *rb, GLuint co
 
 
 static void
-texture_put_mono_values(struct gl_context *ctx, struct gl_renderbuffer *rb,
-                        GLuint count, const GLint x[], const GLint y[],
-                        const void *value, const GLubyte *mask)
-{
-   struct texture_renderbuffer *trb = texture_renderbuffer(rb);
-   const GLint z = trb->Zoffset;
-   GLuint i;
-
-   if (rb->DataType == CHAN_TYPE) {
-      const GLchan *rgba = (const GLchan *) value;
-      for (i = 0; i < count; i++) {
-         if (!mask || mask[i]) {
-            trb->Store(trb->TexImage, x[i], y[i] + trb->Yoffset, z, rgba);
-         }
-      }
-   }
-   else if (rb->DataType == GL_UNSIGNED_INT) {
-      const GLuint zValue = *((const GLuint *) value);
-      for (i = 0; i < count; i++) {
-         if (!mask || mask[i]) {
-            trb->Store(trb->TexImage, x[i], y[i] + trb->Yoffset, z, &zValue);
-         }
-      }
-   }
-   else if (rb->DataType == GL_UNSIGNED_SHORT) {
-      const GLushort zValue = *((const GLushort *) value);
-      for (i = 0; i < count; i++) {
-         if (!mask || mask[i]) {
-            trb->Store(trb->TexImage, x[i], y[i] + trb->Yoffset, z, &zValue);
-         }
-      }
-   }
-   else if (rb->DataType == GL_UNSIGNED_INT_24_8_EXT) {
-      const GLuint zValue = *((const GLuint *) value);
-      const GLfloat flt = (GLfloat) ((zValue >> 8) * (1.0 / 0xffffff));
-      for (i = 0; i < count; i++) {
-         if (!mask || mask[i]) {
-            trb->Store(trb->TexImage, x[i], y[i] + trb->Yoffset, z, &flt);
-         }
-      }
-   }
-   else if (rb->DataType == GL_UNSIGNED_INT_8_24_REV_MESA) {
-      const GLuint zValue = *((const GLuint *) value);
-      const GLfloat flt = (GLfloat) ((zValue & 0xffffff) * (1.0 / 0xffffff));
-      for (i = 0; i < count; i++) {
-         if (!mask || mask[i]) {
-            trb->Store(trb->TexImage, x[i], y[i] + trb->Yoffset, z, &flt);
-         }
-      }
-   }
-   else {
-      _mesa_problem(ctx, "invalid rb->DataType in texture_put_mono_values");
-   }
-}
-
-
-static void
 store_nop(struct swrast_texture_image *texImage,
           GLint col, GLint row, GLint img,
           const void *texel)
@@ -514,10 +338,7 @@ wrap_texture(struct gl_context *ctx, struct gl_renderbuffer_attachment *att)
    trb->Base.GetRow = texture_get_row;
    trb->Base.GetValues = texture_get_values;
    trb->Base.PutRow = texture_put_row;
-   trb->Base.PutRowRGB = texture_put_row_rgb;
-   trb->Base.PutMonoRow = texture_put_mono_row;
    trb->Base.PutValues = texture_put_values;
-   trb->Base.PutMonoValues = texture_put_mono_values;
 
    /* update attachment point */
    _mesa_reference_renderbuffer(&att->Renderbuffer, &(trb->Base));
diff --git a/mesalib/src/mesa/swrast/s_triangle.c b/mesalib/src/mesa/swrast/s_triangle.c
index b4f8e7479..43deaf47b 100644
--- a/mesalib/src/mesa/swrast/s_triangle.c
+++ b/mesalib/src/mesa/swrast/s_triangle.c
@@ -142,7 +142,7 @@ _swrast_culltriangle( struct gl_context *ctx,
 
 #define RENDER_SPAN( span )						\
    GLuint i;								\
-   GLubyte rgb[MAX_WIDTH][3];						\
+   GLubyte rgba[MAX_WIDTH][4];						\
    span.intTex[0] -= FIXED_HALF; /* off-by-one error? */		\
    span.intTex[1] -= FIXED_HALF;					\
    for (i = 0; i < span.end; i++) {					\
@@ -150,13 +150,14 @@ _swrast_culltriangle( struct gl_context *ctx,
       GLint t = FixedToInt(span.intTex[1]) & tmask;			\
       GLint pos = (t << twidth_log2) + s;				\
       pos = pos + pos + pos;  /* multiply by 3 */			\
-      rgb[i][RCOMP] = texture[pos+2];					\
-      rgb[i][GCOMP] = texture[pos+1];					\
-      rgb[i][BCOMP] = texture[pos+0];					\
+      rgba[i][RCOMP] = texture[pos+2];					\
+      rgba[i][GCOMP] = texture[pos+1];					\
+      rgba[i][BCOMP] = texture[pos+0];					\
+      rgba[i][ACOMP] = 0xff;                                            \
       span.intTex[0] += span.intTexStep[0];				\
       span.intTex[1] += span.intTexStep[1];				\
    }									\
-   rb->PutRowRGB(ctx, rb, span.end, span.x, span.y, rgb, NULL);
+   rb->PutRow(ctx, rb, span.end, span.x, span.y, rgba, NULL);
 
 #include "s_tritemp.h"
 
@@ -198,7 +199,7 @@ _swrast_culltriangle( struct gl_context *ctx,
 
 #define RENDER_SPAN( span )						\
    GLuint i;				    				\
-   GLubyte rgb[MAX_WIDTH][3];						\
+   GLubyte rgba[MAX_WIDTH][4];						\
    span.intTex[0] -= FIXED_HALF; /* off-by-one error? */		\
    span.intTex[1] -= FIXED_HALF;					\
    for (i = 0; i < span.end; i++) {					\
@@ -208,9 +209,10 @@ _swrast_culltriangle( struct gl_context *ctx,
          GLint t = FixedToInt(span.intTex[1]) & tmask;			\
          GLint pos = (t << twidth_log2) + s;				\
          pos = pos + pos + pos;  /* multiply by 3 */			\
-         rgb[i][RCOMP] = texture[pos+2];				\
-         rgb[i][GCOMP] = texture[pos+1];				\
-         rgb[i][BCOMP] = texture[pos+0];				\
+         rgba[i][RCOMP] = texture[pos+2];				\
+         rgba[i][GCOMP] = texture[pos+1];				\
+         rgba[i][BCOMP] = texture[pos+0];				\
+         rgba[i][ACOMP] = 0xff;          				\
          zRow[i] = z;							\
          span.array->mask[i] = 1;					\
       }									\
@@ -221,7 +223,7 @@ _swrast_culltriangle( struct gl_context *ctx,
       span.intTex[1] += span.intTexStep[1];				\
       span.z += span.zStep;						\
    }									\
-   rb->PutRowRGB(ctx, rb, span.end, span.x, span.y, rgb, span.array->mask);
+   rb->PutRow(ctx, rb, span.end, span.x, span.y, rgba, span.array->mask);
 
 #include "s_tritemp.h"
 
@@ -872,22 +874,24 @@ fast_persp_span(struct gl_context *ctx, SWspan *span,
 /*
  * Special tri function for occlusion testing
  */
-#define NAME occlusion_zless_triangle
+#define NAME occlusion_zless_16_triangle
 #define INTERP_Z 1
 #define SETUP_CODE							\
-   struct gl_renderbuffer *rb = ctx->DrawBuffer->_DepthBuffer;		\
+   struct gl_renderbuffer *rb =                                         \
+      ctx->DrawBuffer->Attachment[BUFFER_DEPTH].Renderbuffer;           \
    struct gl_query_object *q = ctx->Query.CurrentOcclusionObject;	\
    ASSERT(ctx->Depth.Test);						\
    ASSERT(!ctx->Depth.Mask);						\
    ASSERT(ctx->Depth.Func == GL_LESS);					\
+   assert(rb->Format == MESA_FORMAT_Z16);                               \
    if (!q) {								\
       return;								\
    }
 #define RENDER_SPAN( span )						\
-   if (rb->Format == MESA_FORMAT_Z16) {					\
+   {                                                                    \
       GLuint i;								\
       const GLushort *zRow = (const GLushort *)				\
-         rb->GetPointer(ctx, rb, span.x, span.y);			\
+         _swrast_pixel_address(rb, span.x, span.y);                     \
       for (i = 0; i < span.end; i++) {					\
          GLuint z = FixedToDepth(span.z);				\
          if (z < zRow[i]) {						\
@@ -895,17 +899,6 @@ fast_persp_span(struct gl_context *ctx, SWspan *span,
          }								\
          span.z += span.zStep;						\
       }									\
-   }									\
-   else {								\
-      GLuint i;								\
-      const GLuint *zRow = (const GLuint *)				\
-         rb->GetPointer(ctx, rb, span.x, span.y);			\
-      for (i = 0; i < span.end; i++) {					\
-         if ((GLuint)span.z < zRow[i]) {				\
-            q->Result++;						\
-         }								\
-         span.z += span.zStep;						\
-      }									\
    }
 #include "s_tritemp.h"
 
@@ -1012,6 +1005,8 @@ _swrast_choose_triangle( struct gl_context *ctx )
    }
 
    if (ctx->RenderMode==GL_RENDER) {
+      struct gl_renderbuffer *depthRb =
+         ctx->DrawBuffer->Attachment[BUFFER_DEPTH].Renderbuffer;
 
       if (ctx->Polygon.SmoothFlag) {
          _swrast_set_aa_triangle_function(ctx);
@@ -1024,12 +1019,14 @@ _swrast_choose_triangle( struct gl_context *ctx )
           ctx->Depth.Test &&
           ctx->Depth.Mask == GL_FALSE &&
           ctx->Depth.Func == GL_LESS &&
-          !ctx->Stencil._Enabled) {
+          !ctx->Stencil._Enabled &&
+          depthRb &&
+          depthRb->Format == MESA_FORMAT_Z16) {
          if (ctx->Color.ColorMask[0][0] == 0 &&
 	     ctx->Color.ColorMask[0][1] == 0 &&
 	     ctx->Color.ColorMask[0][2] == 0 &&
 	     ctx->Color.ColorMask[0][3] == 0) {
-            USE(occlusion_zless_triangle);
+            USE(occlusion_zless_16_triangle);
             return;
          }
       }
diff --git a/mesalib/src/mesa/swrast/s_tritemp.h b/mesalib/src/mesa/swrast/s_tritemp.h
index 340c410ca..061759d26 100644
--- a/mesalib/src/mesa/swrast/s_tritemp.h
+++ b/mesalib/src/mesa/swrast/s_tritemp.h
@@ -1,929 +1,929 @@
-/*

- * Mesa 3-D graphics library

- * Version:  7.0

- *

- * Copyright (C) 1999-2007  Brian Paul   All Rights Reserved.

- *

- * Permission is hereby granted, free of charge, to any person obtaining a

- * copy of this software and associated documentation files (the "Software"),

- * to deal in the Software without restriction, including without limitation

- * the rights to use, copy, modify, merge, publish, distribute, sublicense,

- * and/or sell copies of the Software, and to permit persons to whom the

- * Software is furnished to do so, subject to the following conditions:

- *

- * The above copyright notice and this permission notice shall be included

- * in all copies or substantial portions of the Software.

- *

- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS

- * OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,

- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL

- * BRIAN PAUL BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN

- * AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN

- * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.

- */

-

-/*

- * Triangle Rasterizer Template

- *

- * This file is #include'd to generate custom triangle rasterizers.

- *

- * The following macros may be defined to indicate what auxillary information

- * must be interpolated across the triangle:

- *    INTERP_Z        - if defined, interpolate integer Z values

- *    INTERP_RGB      - if defined, interpolate integer RGB values

- *    INTERP_ALPHA    - if defined, interpolate integer Alpha values

- *    INTERP_INT_TEX  - if defined, interpolate integer ST texcoords

- *                         (fast, simple 2-D texture mapping, without

- *                         perspective correction)

- *    INTERP_ATTRIBS  - if defined, interpolate arbitrary attribs (texcoords,

- *                         varying vars, etc)  This also causes W to be

- *                         computed for perspective correction).

- *

- * When one can directly address pixels in the color buffer the following

- * macros can be defined and used to compute pixel addresses during

- * rasterization (see pRow):

- *    PIXEL_TYPE          - the datatype of a pixel (GLubyte, GLushort, GLuint)

- *    BYTES_PER_ROW       - number of bytes per row in the color buffer

- *    PIXEL_ADDRESS(X,Y)  - returns the address of pixel at (X,Y) where

- *                          Y==0 at bottom of screen and increases upward.

- *

- * Similarly, for direct depth buffer access, this type is used for depth

- * buffer addressing (see zRow):

- *    DEPTH_TYPE          - either GLushort or GLuint

- *

- * Optionally, one may provide one-time setup code per triangle:

- *    SETUP_CODE    - code which is to be executed once per triangle

- *

- * The following macro MUST be defined:

- *    RENDER_SPAN(span) - code to write a span of pixels.

- *

- * This code was designed for the origin to be in the lower-left corner.

- *

- * Inspired by triangle rasterizer code written by Allen Akin.  Thanks Allen!

- *

- *

- * Some notes on rasterization accuracy:

- *

- * This code uses fixed point arithmetic (the GLfixed type) to iterate

- * over the triangle edges and interpolate ancillary data (such as Z,

- * color, secondary color, etc).  The number of fractional bits in

- * GLfixed and the value of SUB_PIXEL_BITS has a direct bearing on the

- * accuracy of rasterization.

- *

- * If SUB_PIXEL_BITS=4 then we'll snap the vertices to the nearest

- * 1/16 of a pixel.  If we're walking up a long, nearly vertical edge

- * (dx=1/16, dy=1024) we'll need 4 + 10 = 14 fractional bits in

- * GLfixed to walk the edge without error.  If the maximum viewport

- * height is 4K pixels, then we'll need 4 + 12 = 16 fractional bits.

- *

- * Historically, Mesa has used 11 fractional bits in GLfixed, snaps

- * vertices to 1/16 pixel and allowed a maximum viewport height of 2K

- * pixels.  11 fractional bits is actually insufficient for accurately

- * rasterizing some triangles.  More recently, the maximum viewport

- * height was increased to 4K pixels.  Thus, Mesa should be using 16

- * fractional bits in GLfixed.  Unfortunately, there may be some issues

- * with setting FIXED_FRAC_BITS=16, such as multiplication overflow.

- * This will have to be examined in some detail...

- *

- * For now, if you find rasterization errors, particularly with tall,

- * sliver triangles, try increasing FIXED_FRAC_BITS and/or decreasing

- * SUB_PIXEL_BITS.

- */

-

-

-/*

- * Some code we unfortunately need to prevent negative interpolated colors.

- */

-#ifndef CLAMP_INTERPOLANT

-#define CLAMP_INTERPOLANT(CHANNEL, CHANNELSTEP, LEN)		\

-do {								\

-   GLfixed endVal = span.CHANNEL + (LEN) * span.CHANNELSTEP;	\

-   if (endVal < 0) {						\

-      span.CHANNEL -= endVal;					\

-   }								\

-   if (span.CHANNEL < 0) {					\

-      span.CHANNEL = 0;						\

-   }								\

-} while (0)

-#endif

-

-

-static void NAME(struct gl_context *ctx, const SWvertex *v0,

-                                 const SWvertex *v1,

-                                 const SWvertex *v2 )

-{

-   typedef struct {

-      const SWvertex *v0, *v1;   /* Y(v0) < Y(v1) */

-      GLfloat dx;	/* X(v1) - X(v0) */

-      GLfloat dy;	/* Y(v1) - Y(v0) */

-      GLfloat dxdy;	/* dx/dy */

-      GLfixed fdxdy;	/* dx/dy in fixed-point */

-      GLfloat adjy;	/* adjust from v[0]->fy to fsy, scaled */

-      GLfixed fsx;	/* first sample point x coord */

-      GLfixed fsy;

-      GLfixed fx0;	/* fixed pt X of lower endpoint */

-      GLint lines;	/* number of lines to be sampled on this edge */

-   } EdgeT;

-

-   const SWcontext *swrast = SWRAST_CONTEXT(ctx);

-#ifdef INTERP_Z

-   const GLint depthBits = ctx->DrawBuffer->Visual.depthBits;

-   const GLint fixedToDepthShift = depthBits <= 16 ? FIXED_SHIFT : 0;

-   const GLfloat maxDepth = ctx->DrawBuffer->_DepthMaxF;

-#define FixedToDepth(F)  ((F) >> fixedToDepthShift)

-#endif

-   EdgeT eMaj, eTop, eBot;

-   GLfloat oneOverArea;

-   const SWvertex *vMin, *vMid, *vMax;  /* Y(vMin)<=Y(vMid)<=Y(vMax) */

-   GLfloat bf = SWRAST_CONTEXT(ctx)->_BackfaceSign;

-   const GLint snapMask = ~((FIXED_ONE / (1 << SUB_PIXEL_BITS)) - 1); /* for x/y coord snapping */

-   GLfixed vMin_fx, vMin_fy, vMid_fx, vMid_fy, vMax_fx, vMax_fy;

-

-   SWspan span;

-

-   (void) swrast;

-

-   INIT_SPAN(span, GL_POLYGON);

-   span.y = 0; /* silence warnings */

-

-#ifdef INTERP_Z

-   (void) fixedToDepthShift;

-#endif

-

-   /*

-   printf("%s()\n", __FUNCTION__);

-   printf("  %g, %g, %g\n",

-          v0->attrib[FRAG_ATTRIB_WPOS][0],

-          v0->attrib[FRAG_ATTRIB_WPOS][1],

-          v0->attrib[FRAG_ATTRIB_WPOS][2]);

-   printf("  %g, %g, %g\n",

-          v1->attrib[FRAG_ATTRIB_WPOS][0],

-          v1->attrib[FRAG_ATTRIB_WPOS][1],

-          v1->attrib[FRAG_ATTRIB_WPOS][2]);

-   printf("  %g, %g, %g\n",

-          v2->attrib[FRAG_ATTRIB_WPOS][0],

-          v2->attrib[FRAG_ATTRIB_WPOS][1],

-          v2->attrib[FRAG_ATTRIB_WPOS][2]);

-   */

-

-   /* Compute fixed point x,y coords w/ half-pixel offsets and snapping.

-    * And find the order of the 3 vertices along the Y axis.

-    */

-   {

-      const GLfixed fy0 = FloatToFixed(v0->attrib[FRAG_ATTRIB_WPOS][1] - 0.5F) & snapMask;

-      const GLfixed fy1 = FloatToFixed(v1->attrib[FRAG_ATTRIB_WPOS][1] - 0.5F) & snapMask;

-      const GLfixed fy2 = FloatToFixed(v2->attrib[FRAG_ATTRIB_WPOS][1] - 0.5F) & snapMask;

-      if (fy0 <= fy1) {

-         if (fy1 <= fy2) {

-            /* y0 <= y1 <= y2 */

-            vMin = v0;   vMid = v1;   vMax = v2;

-            vMin_fy = fy0;  vMid_fy = fy1;  vMax_fy = fy2;

-         }

-         else if (fy2 <= fy0) {

-            /* y2 <= y0 <= y1 */

-            vMin = v2;   vMid = v0;   vMax = v1;

-            vMin_fy = fy2;  vMid_fy = fy0;  vMax_fy = fy1;

-         }

-         else {

-            /* y0 <= y2 <= y1 */

-            vMin = v0;   vMid = v2;   vMax = v1;

-            vMin_fy = fy0;  vMid_fy = fy2;  vMax_fy = fy1;

-            bf = -bf;

-         }

-      }

-      else {

-         if (fy0 <= fy2) {

-            /* y1 <= y0 <= y2 */

-            vMin = v1;   vMid = v0;   vMax = v2;

-            vMin_fy = fy1;  vMid_fy = fy0;  vMax_fy = fy2;

-            bf = -bf;

-         }

-         else if (fy2 <= fy1) {

-            /* y2 <= y1 <= y0 */

-            vMin = v2;   vMid = v1;   vMax = v0;

-            vMin_fy = fy2;  vMid_fy = fy1;  vMax_fy = fy0;

-            bf = -bf;

-         }

-         else {

-            /* y1 <= y2 <= y0 */

-            vMin = v1;   vMid = v2;   vMax = v0;

-            vMin_fy = fy1;  vMid_fy = fy2;  vMax_fy = fy0;

-         }

-      }

-

-      /* fixed point X coords */

-      vMin_fx = FloatToFixed(vMin->attrib[FRAG_ATTRIB_WPOS][0] + 0.5F) & snapMask;

-      vMid_fx = FloatToFixed(vMid->attrib[FRAG_ATTRIB_WPOS][0] + 0.5F) & snapMask;

-      vMax_fx = FloatToFixed(vMax->attrib[FRAG_ATTRIB_WPOS][0] + 0.5F) & snapMask;

-   }

-

-   /* vertex/edge relationship */

-   eMaj.v0 = vMin;   eMaj.v1 = vMax;   /*TODO: .v1's not needed */

-   eTop.v0 = vMid;   eTop.v1 = vMax;

-   eBot.v0 = vMin;   eBot.v1 = vMid;

-

-   /* compute deltas for each edge:  vertex[upper] - vertex[lower] */

-   eMaj.dx = FixedToFloat(vMax_fx - vMin_fx);

-   eMaj.dy = FixedToFloat(vMax_fy - vMin_fy);

-   eTop.dx = FixedToFloat(vMax_fx - vMid_fx);

-   eTop.dy = FixedToFloat(vMax_fy - vMid_fy);

-   eBot.dx = FixedToFloat(vMid_fx - vMin_fx);

-   eBot.dy = FixedToFloat(vMid_fy - vMin_fy);

-

-   /* compute area, oneOverArea and perform backface culling */

-   {

-      const GLfloat area = eMaj.dx * eBot.dy - eBot.dx * eMaj.dy;

-

-      if (IS_INF_OR_NAN(area) || area == 0.0F)

-         return;

-

-      if (area * bf * swrast->_BackfaceCullSign < 0.0)

-         return;

-

-      oneOverArea = 1.0F / area;

-

-      /* 0 = front, 1 = back */

-      span.facing = oneOverArea * bf > 0.0F;

-   }

-

-   /* Edge setup.  For a triangle strip these could be reused... */

-   {

-      eMaj.fsy = FixedCeil(vMin_fy);

-      eMaj.lines = FixedToInt(FixedCeil(vMax_fy - eMaj.fsy));

-      if (eMaj.lines > 0) {

-         eMaj.dxdy = eMaj.dx / eMaj.dy;

-         eMaj.fdxdy = SignedFloatToFixed(eMaj.dxdy);

-         eMaj.adjy = (GLfloat) (eMaj.fsy - vMin_fy);  /* SCALED! */

-         eMaj.fx0 = vMin_fx;

-         eMaj.fsx = eMaj.fx0 + (GLfixed) (eMaj.adjy * eMaj.dxdy);

-      }

-      else {

-         return;  /*CULLED*/

-      }

-

-      eTop.fsy = FixedCeil(vMid_fy);

-      eTop.lines = FixedToInt(FixedCeil(vMax_fy - eTop.fsy));

-      if (eTop.lines > 0) {

-         eTop.dxdy = eTop.dx / eTop.dy;

-         eTop.fdxdy = SignedFloatToFixed(eTop.dxdy);

-         eTop.adjy = (GLfloat) (eTop.fsy - vMid_fy); /* SCALED! */

-         eTop.fx0 = vMid_fx;

-         eTop.fsx = eTop.fx0 + (GLfixed) (eTop.adjy * eTop.dxdy);

-      }

-

-      eBot.fsy = FixedCeil(vMin_fy);

-      eBot.lines = FixedToInt(FixedCeil(vMid_fy - eBot.fsy));

-      if (eBot.lines > 0) {

-         eBot.dxdy = eBot.dx / eBot.dy;

-         eBot.fdxdy = SignedFloatToFixed(eBot.dxdy);

-         eBot.adjy = (GLfloat) (eBot.fsy - vMin_fy);  /* SCALED! */

-         eBot.fx0 = vMin_fx;

-         eBot.fsx = eBot.fx0 + (GLfixed) (eBot.adjy * eBot.dxdy);

-      }

-   }

-

-   /*

-    * Conceptually, we view a triangle as two subtriangles

-    * separated by a perfectly horizontal line.  The edge that is

-    * intersected by this line is one with maximal absolute dy; we

-    * call it a ``major'' edge.  The other two edges are the

-    * ``top'' edge (for the upper subtriangle) and the ``bottom''

-    * edge (for the lower subtriangle).  If either of these two

-    * edges is horizontal or very close to horizontal, the

-    * corresponding subtriangle might cover zero sample points;

-    * we take care to handle such cases, for performance as well

-    * as correctness.

-    *

-    * By stepping rasterization parameters along the major edge,

-    * we can avoid recomputing them at the discontinuity where

-    * the top and bottom edges meet.  However, this forces us to

-    * be able to scan both left-to-right and right-to-left.

-    * Also, we must determine whether the major edge is at the

-    * left or right side of the triangle.  We do this by

-    * computing the magnitude of the cross-product of the major

-    * and top edges.  Since this magnitude depends on the sine of

-    * the angle between the two edges, its sign tells us whether

-    * we turn to the left or to the right when travelling along

-    * the major edge to the top edge, and from this we infer

-    * whether the major edge is on the left or the right.

-    *

-    * Serendipitously, this cross-product magnitude is also a

-    * value we need to compute the iteration parameter

-    * derivatives for the triangle, and it can be used to perform

-    * backface culling because its sign tells us whether the

-    * triangle is clockwise or counterclockwise.  In this code we

-    * refer to it as ``area'' because it's also proportional to

-    * the pixel area of the triangle.

-    */

-

-   {

-      GLint scan_from_left_to_right;  /* true if scanning left-to-right */

-

-      /*

-       * Execute user-supplied setup code

-       */

-#ifdef SETUP_CODE

-      SETUP_CODE

-#endif

-

-      scan_from_left_to_right = (oneOverArea < 0.0F);

-

-

-      /* compute d?/dx and d?/dy derivatives */

-#ifdef INTERP_Z

-      span.interpMask |= SPAN_Z;

-      {

-         GLfloat eMaj_dz = vMax->attrib[FRAG_ATTRIB_WPOS][2] - vMin->attrib[FRAG_ATTRIB_WPOS][2];

-         GLfloat eBot_dz = vMid->attrib[FRAG_ATTRIB_WPOS][2] - vMin->attrib[FRAG_ATTRIB_WPOS][2];

-         span.attrStepX[FRAG_ATTRIB_WPOS][2] = oneOverArea * (eMaj_dz * eBot.dy - eMaj.dy * eBot_dz);

-         if (span.attrStepX[FRAG_ATTRIB_WPOS][2] > maxDepth ||

-             span.attrStepX[FRAG_ATTRIB_WPOS][2] < -maxDepth) {

-            /* probably a sliver triangle */

-            span.attrStepX[FRAG_ATTRIB_WPOS][2] = 0.0;

-            span.attrStepY[FRAG_ATTRIB_WPOS][2] = 0.0;

-         }

-         else {

-            span.attrStepY[FRAG_ATTRIB_WPOS][2] = oneOverArea * (eMaj.dx * eBot_dz - eMaj_dz * eBot.dx);

-         }

-         if (depthBits <= 16)

-            span.zStep = SignedFloatToFixed(span.attrStepX[FRAG_ATTRIB_WPOS][2]);

-         else

-            span.zStep = (GLint) span.attrStepX[FRAG_ATTRIB_WPOS][2];

-      }

-#endif

-#ifdef INTERP_RGB

-      span.interpMask |= SPAN_RGBA;

-      if (ctx->Light.ShadeModel == GL_SMOOTH) {

-         GLfloat eMaj_dr = (GLfloat) (vMax->color[RCOMP] - vMin->color[RCOMP]);

-         GLfloat eBot_dr = (GLfloat) (vMid->color[RCOMP] - vMin->color[RCOMP]);

-         GLfloat eMaj_dg = (GLfloat) (vMax->color[GCOMP] - vMin->color[GCOMP]);

-         GLfloat eBot_dg = (GLfloat) (vMid->color[GCOMP] - vMin->color[GCOMP]);

-         GLfloat eMaj_db = (GLfloat) (vMax->color[BCOMP] - vMin->color[BCOMP]);

-         GLfloat eBot_db = (GLfloat) (vMid->color[BCOMP] - vMin->color[BCOMP]);

-#  ifdef INTERP_ALPHA

-         GLfloat eMaj_da = (GLfloat) (vMax->color[ACOMP] - vMin->color[ACOMP]);

-         GLfloat eBot_da = (GLfloat) (vMid->color[ACOMP] - vMin->color[ACOMP]);

-#  endif

-         span.attrStepX[FRAG_ATTRIB_COL0][0] = oneOverArea * (eMaj_dr * eBot.dy - eMaj.dy * eBot_dr);

-         span.attrStepY[FRAG_ATTRIB_COL0][0] = oneOverArea * (eMaj.dx * eBot_dr - eMaj_dr * eBot.dx);

-         span.attrStepX[FRAG_ATTRIB_COL0][1] = oneOverArea * (eMaj_dg * eBot.dy - eMaj.dy * eBot_dg);

-         span.attrStepY[FRAG_ATTRIB_COL0][1] = oneOverArea * (eMaj.dx * eBot_dg - eMaj_dg * eBot.dx);

-         span.attrStepX[FRAG_ATTRIB_COL0][2] = oneOverArea * (eMaj_db * eBot.dy - eMaj.dy * eBot_db);

-         span.attrStepY[FRAG_ATTRIB_COL0][2] = oneOverArea * (eMaj.dx * eBot_db - eMaj_db * eBot.dx);

-         span.redStep   = SignedFloatToFixed(span.attrStepX[FRAG_ATTRIB_COL0][0]);

-         span.greenStep = SignedFloatToFixed(span.attrStepX[FRAG_ATTRIB_COL0][1]);

-         span.blueStep  = SignedFloatToFixed(span.attrStepX[FRAG_ATTRIB_COL0][2]);

-#  ifdef INTERP_ALPHA

-         span.attrStepX[FRAG_ATTRIB_COL0][3] = oneOverArea * (eMaj_da * eBot.dy - eMaj.dy * eBot_da);

-         span.attrStepY[FRAG_ATTRIB_COL0][3] = oneOverArea * (eMaj.dx * eBot_da - eMaj_da * eBot.dx);

-         span.alphaStep = SignedFloatToFixed(span.attrStepX[FRAG_ATTRIB_COL0][3]);

-#  endif /* INTERP_ALPHA */

-      }

-      else {

-         ASSERT(ctx->Light.ShadeModel == GL_FLAT);

-         span.interpMask |= SPAN_FLAT;

-         span.attrStepX[FRAG_ATTRIB_COL0][0] = span.attrStepY[FRAG_ATTRIB_COL0][0] = 0.0F;

-         span.attrStepX[FRAG_ATTRIB_COL0][1] = span.attrStepY[FRAG_ATTRIB_COL0][1] = 0.0F;

-         span.attrStepX[FRAG_ATTRIB_COL0][2] = span.attrStepY[FRAG_ATTRIB_COL0][2] = 0.0F;

-	 span.redStep   = 0;

-	 span.greenStep = 0;

-	 span.blueStep  = 0;

-#  ifdef INTERP_ALPHA

-         span.attrStepX[FRAG_ATTRIB_COL0][3] = span.attrStepY[FRAG_ATTRIB_COL0][3] = 0.0F;

-	 span.alphaStep = 0;

-#  endif

-      }

-#endif /* INTERP_RGB */

-#ifdef INTERP_INT_TEX

-      {

-         GLfloat eMaj_ds = (vMax->attrib[FRAG_ATTRIB_TEX0][0] - vMin->attrib[FRAG_ATTRIB_TEX0][0]) * S_SCALE;

-         GLfloat eBot_ds = (vMid->attrib[FRAG_ATTRIB_TEX0][0] - vMin->attrib[FRAG_ATTRIB_TEX0][0]) * S_SCALE;

-         GLfloat eMaj_dt = (vMax->attrib[FRAG_ATTRIB_TEX0][1] - vMin->attrib[FRAG_ATTRIB_TEX0][1]) * T_SCALE;

-         GLfloat eBot_dt = (vMid->attrib[FRAG_ATTRIB_TEX0][1] - vMin->attrib[FRAG_ATTRIB_TEX0][1]) * T_SCALE;

-         span.attrStepX[FRAG_ATTRIB_TEX0][0] = oneOverArea * (eMaj_ds * eBot.dy - eMaj.dy * eBot_ds);

-         span.attrStepY[FRAG_ATTRIB_TEX0][0] = oneOverArea * (eMaj.dx * eBot_ds - eMaj_ds * eBot.dx);

-         span.attrStepX[FRAG_ATTRIB_TEX0][1] = oneOverArea * (eMaj_dt * eBot.dy - eMaj.dy * eBot_dt);

-         span.attrStepY[FRAG_ATTRIB_TEX0][1] = oneOverArea * (eMaj.dx * eBot_dt - eMaj_dt * eBot.dx);

-         span.intTexStep[0] = SignedFloatToFixed(span.attrStepX[FRAG_ATTRIB_TEX0][0]);

-         span.intTexStep[1] = SignedFloatToFixed(span.attrStepX[FRAG_ATTRIB_TEX0][1]);

-      }

-#endif

-#ifdef INTERP_ATTRIBS

-      {

-         /* attrib[FRAG_ATTRIB_WPOS][3] is 1/W */

-         const GLfloat wMax = vMax->attrib[FRAG_ATTRIB_WPOS][3];

-         const GLfloat wMin = vMin->attrib[FRAG_ATTRIB_WPOS][3];

-         const GLfloat wMid = vMid->attrib[FRAG_ATTRIB_WPOS][3];

-         {

-            const GLfloat eMaj_dw = wMax - wMin;

-            const GLfloat eBot_dw = wMid - wMin;

-            span.attrStepX[FRAG_ATTRIB_WPOS][3] = oneOverArea * (eMaj_dw * eBot.dy - eMaj.dy * eBot_dw);

-            span.attrStepY[FRAG_ATTRIB_WPOS][3] = oneOverArea * (eMaj.dx * eBot_dw - eMaj_dw * eBot.dx);

-         }

-         ATTRIB_LOOP_BEGIN

-            if (swrast->_InterpMode[attr] == GL_FLAT) {

-               ASSIGN_4V(span.attrStepX[attr], 0.0, 0.0, 0.0, 0.0);

-               ASSIGN_4V(span.attrStepY[attr], 0.0, 0.0, 0.0, 0.0);

-            }

-            else {

-               GLuint c;

-               for (c = 0; c < 4; c++) {

-                  GLfloat eMaj_da = vMax->attrib[attr][c] * wMax - vMin->attrib[attr][c] * wMin;

-                  GLfloat eBot_da = vMid->attrib[attr][c] * wMid - vMin->attrib[attr][c] * wMin;

-                  span.attrStepX[attr][c] = oneOverArea * (eMaj_da * eBot.dy - eMaj.dy * eBot_da);

-                  span.attrStepY[attr][c] = oneOverArea * (eMaj.dx * eBot_da - eMaj_da * eBot.dx);

-               }

-            }

-         ATTRIB_LOOP_END

-      }

-#endif

-

-      /*

-       * We always sample at pixel centers.  However, we avoid

-       * explicit half-pixel offsets in this code by incorporating

-       * the proper offset in each of x and y during the

-       * transformation to window coordinates.

-       *

-       * We also apply the usual rasterization rules to prevent

-       * cracks and overlaps.  A pixel is considered inside a

-       * subtriangle if it meets all of four conditions: it is on or

-       * to the right of the left edge, strictly to the left of the

-       * right edge, on or below the top edge, and strictly above

-       * the bottom edge.  (Some edges may be degenerate.)

-       *

-       * The following discussion assumes left-to-right scanning

-       * (that is, the major edge is on the left); the right-to-left

-       * case is a straightforward variation.

-       *

-       * We start by finding the half-integral y coordinate that is

-       * at or below the top of the triangle.  This gives us the

-       * first scan line that could possibly contain pixels that are

-       * inside the triangle.

-       *

-       * Next we creep down the major edge until we reach that y,

-       * and compute the corresponding x coordinate on the edge.

-       * Then we find the half-integral x that lies on or just

-       * inside the edge.  This is the first pixel that might lie in

-       * the interior of the triangle.  (We won't know for sure

-       * until we check the other edges.)

-       *

-       * As we rasterize the triangle, we'll step down the major

-       * edge.  For each step in y, we'll move an integer number

-       * of steps in x.  There are two possible x step sizes, which

-       * we'll call the ``inner'' step (guaranteed to land on the

-       * edge or inside it) and the ``outer'' step (guaranteed to

-       * land on the edge or outside it).  The inner and outer steps

-       * differ by one.  During rasterization we maintain an error

-       * term that indicates our distance from the true edge, and

-       * select either the inner step or the outer step, whichever

-       * gets us to the first pixel that falls inside the triangle.

-       *

-       * All parameters (z, red, etc.) as well as the buffer

-       * addresses for color and z have inner and outer step values,

-       * so that we can increment them appropriately.  This method

-       * eliminates the need to adjust parameters by creeping a

-       * sub-pixel amount into the triangle at each scanline.

-       */

-

-      {

-         GLint subTriangle;

-         GLfixed fxLeftEdge = 0, fxRightEdge = 0;

-         GLfixed fdxLeftEdge = 0, fdxRightEdge = 0;

-         GLfixed fError = 0, fdError = 0;

-#ifdef PIXEL_ADDRESS

-         PIXEL_TYPE *pRow = NULL;

-         GLint dPRowOuter = 0, dPRowInner;  /* offset in bytes */

-#endif

-#ifdef INTERP_Z

-#  ifdef DEPTH_TYPE

-         struct gl_renderbuffer *zrb

-            = ctx->DrawBuffer->Attachment[BUFFER_DEPTH].Renderbuffer;

-         DEPTH_TYPE *zRow = NULL;

-         GLint dZRowOuter = 0, dZRowInner;  /* offset in bytes */

-#  endif

-         GLuint zLeft = 0;

-         GLfixed fdzOuter = 0, fdzInner;

-#endif

-#ifdef INTERP_RGB

-         GLint rLeft = 0, fdrOuter = 0, fdrInner;

-         GLint gLeft = 0, fdgOuter = 0, fdgInner;

-         GLint bLeft = 0, fdbOuter = 0, fdbInner;

-#endif

-#ifdef INTERP_ALPHA

-         GLint aLeft = 0, fdaOuter = 0, fdaInner;

-#endif

-#ifdef INTERP_INT_TEX

-         GLfixed sLeft=0, dsOuter=0, dsInner;

-         GLfixed tLeft=0, dtOuter=0, dtInner;

-#endif

-#ifdef INTERP_ATTRIBS

-         GLfloat wLeft = 0, dwOuter = 0, dwInner;

-         GLfloat attrLeft[FRAG_ATTRIB_MAX][4];

-         GLfloat daOuter[FRAG_ATTRIB_MAX][4], daInner[FRAG_ATTRIB_MAX][4];

-#endif

-

-         for (subTriangle=0; subTriangle<=1; subTriangle++) {

-            EdgeT *eLeft, *eRight;

-            int setupLeft, setupRight;

-            int lines;

-

-            if (subTriangle==0) {

-               /* bottom half */

-               if (scan_from_left_to_right) {

-                  eLeft = &eMaj;

-                  eRight = &eBot;

-                  lines = eRight->lines;

-                  setupLeft = 1;

-                  setupRight = 1;

-               }

-               else {

-                  eLeft = &eBot;

-                  eRight = &eMaj;

-                  lines = eLeft->lines;

-                  setupLeft = 1;

-                  setupRight = 1;

-               }

-            }

-            else {

-               /* top half */

-               if (scan_from_left_to_right) {

-                  eLeft = &eMaj;

-                  eRight = &eTop;

-                  lines = eRight->lines;

-                  setupLeft = 0;

-                  setupRight = 1;

-               }

-               else {

-                  eLeft = &eTop;

-                  eRight = &eMaj;

-                  lines = eLeft->lines;

-                  setupLeft = 1;

-                  setupRight = 0;

-               }

-               if (lines == 0)

-                  return;

-            }

-

-            if (setupLeft && eLeft->lines > 0) {

-               const SWvertex *vLower = eLeft->v0;

-               const GLfixed fsy = eLeft->fsy;

-               const GLfixed fsx = eLeft->fsx;  /* no fractional part */

-               const GLfixed fx = FixedCeil(fsx);  /* no fractional part */

-               const GLfixed adjx = (GLfixed) (fx - eLeft->fx0); /* SCALED! */

-               const GLfixed adjy = (GLfixed) eLeft->adjy;      /* SCALED! */

-               GLint idxOuter;

-               GLfloat dxOuter;

-               GLfixed fdxOuter;

-

-               fError = fx - fsx - FIXED_ONE;

-               fxLeftEdge = fsx - FIXED_EPSILON;

-               fdxLeftEdge = eLeft->fdxdy;

-               fdxOuter = FixedFloor(fdxLeftEdge - FIXED_EPSILON);

-               fdError = fdxOuter - fdxLeftEdge + FIXED_ONE;

-               idxOuter = FixedToInt(fdxOuter);

-               dxOuter = (GLfloat) idxOuter;

-               span.y = FixedToInt(fsy);

-

-               /* silence warnings on some compilers */

-               (void) dxOuter;

-               (void) adjx;

-               (void) adjy;

-               (void) vLower;

-

-#ifdef PIXEL_ADDRESS

-               {

-                  pRow = (PIXEL_TYPE *) PIXEL_ADDRESS(FixedToInt(fxLeftEdge), span.y);

-                  dPRowOuter = -((int)BYTES_PER_ROW) + idxOuter * sizeof(PIXEL_TYPE);

-                  /* negative because Y=0 at bottom and increases upward */

-               }

-#endif

-               /*

-                * Now we need the set of parameter (z, color, etc.) values at

-                * the point (fx, fsy).  This gives us properly-sampled parameter

-                * values that we can step from pixel to pixel.  Furthermore,

-                * although we might have intermediate results that overflow

-                * the normal parameter range when we step temporarily outside

-                * the triangle, we shouldn't overflow or underflow for any

-                * pixel that's actually inside the triangle.

-                */

-

-#ifdef INTERP_Z

-               {

-                  GLfloat z0 = vLower->attrib[FRAG_ATTRIB_WPOS][2];

-                  if (depthBits <= 16) {

-                     /* interpolate fixed-pt values */

-                     GLfloat tmp = (z0 * FIXED_SCALE

-                                    + span.attrStepX[FRAG_ATTRIB_WPOS][2] * adjx

-                                    + span.attrStepY[FRAG_ATTRIB_WPOS][2] * adjy) + FIXED_HALF;

-                     if (tmp < MAX_GLUINT / 2)

-                        zLeft = (GLfixed) tmp;

-                     else

-                        zLeft = MAX_GLUINT / 2;

-                     fdzOuter = SignedFloatToFixed(span.attrStepY[FRAG_ATTRIB_WPOS][2] +

-                                                   dxOuter * span.attrStepX[FRAG_ATTRIB_WPOS][2]);

-                  }

-                  else {

-                     /* interpolate depth values w/out scaling */

-                     zLeft = (GLuint) (z0 + span.attrStepX[FRAG_ATTRIB_WPOS][2] * FixedToFloat(adjx)

-                                          + span.attrStepY[FRAG_ATTRIB_WPOS][2] * FixedToFloat(adjy));

-                     fdzOuter = (GLint) (span.attrStepY[FRAG_ATTRIB_WPOS][2] +

-                                         dxOuter * span.attrStepX[FRAG_ATTRIB_WPOS][2]);

-                  }

-#  ifdef DEPTH_TYPE

-                  zRow = (DEPTH_TYPE *)

-                    zrb->GetPointer(ctx, zrb, FixedToInt(fxLeftEdge), span.y);

-                  dZRowOuter = (ctx->DrawBuffer->Width + idxOuter) * sizeof(DEPTH_TYPE);

-#  endif

-               }

-#endif

-#ifdef INTERP_RGB

-               if (ctx->Light.ShadeModel == GL_SMOOTH) {

-                  rLeft = (GLint)(ChanToFixed(vLower->color[RCOMP])

-                                  + span.attrStepX[FRAG_ATTRIB_COL0][0] * adjx

-                                  + span.attrStepY[FRAG_ATTRIB_COL0][0] * adjy) + FIXED_HALF;

-                  gLeft = (GLint)(ChanToFixed(vLower->color[GCOMP])

-                                  + span.attrStepX[FRAG_ATTRIB_COL0][1] * adjx

-                                  + span.attrStepY[FRAG_ATTRIB_COL0][1] * adjy) + FIXED_HALF;

-                  bLeft = (GLint)(ChanToFixed(vLower->color[BCOMP])

-                                  + span.attrStepX[FRAG_ATTRIB_COL0][2] * adjx

-                                  + span.attrStepY[FRAG_ATTRIB_COL0][2] * adjy) + FIXED_HALF;

-                  fdrOuter = SignedFloatToFixed(span.attrStepY[FRAG_ATTRIB_COL0][0]

-                                                + dxOuter * span.attrStepX[FRAG_ATTRIB_COL0][0]);

-                  fdgOuter = SignedFloatToFixed(span.attrStepY[FRAG_ATTRIB_COL0][1]

-                                                + dxOuter * span.attrStepX[FRAG_ATTRIB_COL0][1]);

-                  fdbOuter = SignedFloatToFixed(span.attrStepY[FRAG_ATTRIB_COL0][2]

-                                                + dxOuter * span.attrStepX[FRAG_ATTRIB_COL0][2]);

-#  ifdef INTERP_ALPHA

-                  aLeft = (GLint)(ChanToFixed(vLower->color[ACOMP])

-                                  + span.attrStepX[FRAG_ATTRIB_COL0][3] * adjx

-                                  + span.attrStepY[FRAG_ATTRIB_COL0][3] * adjy) + FIXED_HALF;

-                  fdaOuter = SignedFloatToFixed(span.attrStepY[FRAG_ATTRIB_COL0][3]

-                                                + dxOuter * span.attrStepX[FRAG_ATTRIB_COL0][3]);

-#  endif

-               }

-               else {

-                  ASSERT(ctx->Light.ShadeModel == GL_FLAT);

-                  rLeft = ChanToFixed(v2->color[RCOMP]);

-                  gLeft = ChanToFixed(v2->color[GCOMP]);

-                  bLeft = ChanToFixed(v2->color[BCOMP]);

-                  fdrOuter = fdgOuter = fdbOuter = 0;

-#  ifdef INTERP_ALPHA

-                  aLeft = ChanToFixed(v2->color[ACOMP]);

-                  fdaOuter = 0;

-#  endif

-               }

-#endif /* INTERP_RGB */

-

-

-#ifdef INTERP_INT_TEX

-               {

-                  GLfloat s0, t0;

-                  s0 = vLower->attrib[FRAG_ATTRIB_TEX0][0] * S_SCALE;

-                  sLeft = (GLfixed)(s0 * FIXED_SCALE + span.attrStepX[FRAG_ATTRIB_TEX0][0] * adjx

-                                 + span.attrStepY[FRAG_ATTRIB_TEX0][0] * adjy) + FIXED_HALF;

-                  dsOuter = SignedFloatToFixed(span.attrStepY[FRAG_ATTRIB_TEX0][0]

-                                               + dxOuter * span.attrStepX[FRAG_ATTRIB_TEX0][0]);

-

-                  t0 = vLower->attrib[FRAG_ATTRIB_TEX0][1] * T_SCALE;

-                  tLeft = (GLfixed)(t0 * FIXED_SCALE + span.attrStepX[FRAG_ATTRIB_TEX0][1] * adjx

-                                 + span.attrStepY[FRAG_ATTRIB_TEX0][1] * adjy) + FIXED_HALF;

-                  dtOuter = SignedFloatToFixed(span.attrStepY[FRAG_ATTRIB_TEX0][1]

-                                               + dxOuter * span.attrStepX[FRAG_ATTRIB_TEX0][1]);

-               }

-#endif

-#ifdef INTERP_ATTRIBS

-               {

-                  const GLuint attr = FRAG_ATTRIB_WPOS;

-                  wLeft = vLower->attrib[FRAG_ATTRIB_WPOS][3]

-                        + (span.attrStepX[attr][3] * adjx

-                           + span.attrStepY[attr][3] * adjy) * (1.0F/FIXED_SCALE);

-                  dwOuter = span.attrStepY[attr][3] + dxOuter * span.attrStepX[attr][3];

-               }

-               ATTRIB_LOOP_BEGIN

-                  const GLfloat invW = vLower->attrib[FRAG_ATTRIB_WPOS][3];

-                  if (swrast->_InterpMode[attr] == GL_FLAT) {

-                     GLuint c;

-                     for (c = 0; c < 4; c++) {

-                        attrLeft[attr][c] = v2->attrib[attr][c] * invW;

-                        daOuter[attr][c] = 0.0;

-                     }

-                  }

-                  else {

-                     GLuint c;

-                     for (c = 0; c < 4; c++) {

-                        const GLfloat a = vLower->attrib[attr][c] * invW;

-                        attrLeft[attr][c] = a + (  span.attrStepX[attr][c] * adjx

-                                                 + span.attrStepY[attr][c] * adjy) * (1.0F/FIXED_SCALE);

-                        daOuter[attr][c] = span.attrStepY[attr][c] + dxOuter * span.attrStepX[attr][c];

-                     }

-                  }

-               ATTRIB_LOOP_END

-#endif

-            } /*if setupLeft*/

-

-

-            if (setupRight && eRight->lines>0) {

-               fxRightEdge = eRight->fsx - FIXED_EPSILON;

-               fdxRightEdge = eRight->fdxdy;

-            }

-

-            if (lines==0) {

-               continue;

-            }

-

-

-            /* Rasterize setup */

-#ifdef PIXEL_ADDRESS

-            dPRowInner = dPRowOuter + sizeof(PIXEL_TYPE);

-#endif

-#ifdef INTERP_Z

-#  ifdef DEPTH_TYPE

-            dZRowInner = dZRowOuter + sizeof(DEPTH_TYPE);

-#  endif

-            fdzInner = fdzOuter + span.zStep;

-#endif

-#ifdef INTERP_RGB

-            fdrInner = fdrOuter + span.redStep;

-            fdgInner = fdgOuter + span.greenStep;

-            fdbInner = fdbOuter + span.blueStep;

-#endif

-#ifdef INTERP_ALPHA

-            fdaInner = fdaOuter + span.alphaStep;

-#endif

-#ifdef INTERP_INT_TEX

-            dsInner = dsOuter + span.intTexStep[0];

-            dtInner = dtOuter + span.intTexStep[1];

-#endif

-#ifdef INTERP_ATTRIBS

-            dwInner = dwOuter + span.attrStepX[FRAG_ATTRIB_WPOS][3];

-            ATTRIB_LOOP_BEGIN

-               GLuint c;

-               for (c = 0; c < 4; c++) {

-                  daInner[attr][c] = daOuter[attr][c] + span.attrStepX[attr][c];

-               }

-            ATTRIB_LOOP_END

-#endif

-

-            while (lines > 0) {

-               /* initialize the span interpolants to the leftmost value */

-               /* ff = fixed-pt fragment */

-               const GLint right = FixedToInt(fxRightEdge);

-               span.x = FixedToInt(fxLeftEdge);

-               if (right <= span.x)

-                  span.end = 0;

-               else

-                  span.end = right - span.x;

-

-#ifdef INTERP_Z

-               span.z = zLeft;

-#endif

-#ifdef INTERP_RGB

-               span.red = rLeft;

-               span.green = gLeft;

-               span.blue = bLeft;

-#endif

-#ifdef INTERP_ALPHA

-               span.alpha = aLeft;

-#endif

-#ifdef INTERP_INT_TEX

-               span.intTex[0] = sLeft;

-               span.intTex[1] = tLeft;

-#endif

-

-#ifdef INTERP_ATTRIBS

-               span.attrStart[FRAG_ATTRIB_WPOS][3] = wLeft;

-               ATTRIB_LOOP_BEGIN

-                  GLuint c;

-                  for (c = 0; c < 4; c++) {

-                     span.attrStart[attr][c] = attrLeft[attr][c];

-                  }

-               ATTRIB_LOOP_END

-#endif

-

-               /* This is where we actually generate fragments */

-               /* XXX the test for span.y > 0 _shouldn't_ be needed but

-                * it fixes a problem on 64-bit Opterons (bug 4842).

-                */

-               if (span.end > 0 && span.y >= 0) {

-                  const GLint len = span.end - 1;

-                  (void) len;

-#ifdef INTERP_RGB

-                  CLAMP_INTERPOLANT(red, redStep, len);

-                  CLAMP_INTERPOLANT(green, greenStep, len);

-                  CLAMP_INTERPOLANT(blue, blueStep, len);

-#endif

-#ifdef INTERP_ALPHA

-                  CLAMP_INTERPOLANT(alpha, alphaStep, len);

-#endif

-                  {

-                     RENDER_SPAN( span );

-                  }

-               }

-

-               /*

-                * Advance to the next scan line.  Compute the

-                * new edge coordinates, and adjust the

-                * pixel-center x coordinate so that it stays

-                * on or inside the major edge.

-                */

-               span.y++;

-               lines--;

-

-               fxLeftEdge += fdxLeftEdge;

-               fxRightEdge += fdxRightEdge;

-

-               fError += fdError;

-               if (fError >= 0) {

-                  fError -= FIXED_ONE;

-

-#ifdef PIXEL_ADDRESS

-                  pRow = (PIXEL_TYPE *) ((GLubyte *) pRow + dPRowOuter);

-#endif

-#ifdef INTERP_Z

-#  ifdef DEPTH_TYPE

-                  zRow = (DEPTH_TYPE *) ((GLubyte *) zRow + dZRowOuter);

-#  endif

-                  zLeft += fdzOuter;

-#endif

-#ifdef INTERP_RGB

-                  rLeft += fdrOuter;

-                  gLeft += fdgOuter;

-                  bLeft += fdbOuter;

-#endif

-#ifdef INTERP_ALPHA

-                  aLeft += fdaOuter;

-#endif

-#ifdef INTERP_INT_TEX

-                  sLeft += dsOuter;

-                  tLeft += dtOuter;

-#endif

-#ifdef INTERP_ATTRIBS

-                  wLeft += dwOuter;

-                  ATTRIB_LOOP_BEGIN

-                     GLuint c;

-                     for (c = 0; c < 4; c++) {

-                        attrLeft[attr][c] += daOuter[attr][c];

-                     }

-                  ATTRIB_LOOP_END

-#endif

-               }

-               else {

-#ifdef PIXEL_ADDRESS

-                  pRow = (PIXEL_TYPE *) ((GLubyte *) pRow + dPRowInner);

-#endif

-#ifdef INTERP_Z

-#  ifdef DEPTH_TYPE

-                  zRow = (DEPTH_TYPE *) ((GLubyte *) zRow + dZRowInner);

-#  endif

-                  zLeft += fdzInner;

-#endif

-#ifdef INTERP_RGB

-                  rLeft += fdrInner;

-                  gLeft += fdgInner;

-                  bLeft += fdbInner;

-#endif

-#ifdef INTERP_ALPHA

-                  aLeft += fdaInner;

-#endif

-#ifdef INTERP_INT_TEX

-                  sLeft += dsInner;

-                  tLeft += dtInner;

-#endif

-#ifdef INTERP_ATTRIBS

-                  wLeft += dwInner;

-                  ATTRIB_LOOP_BEGIN

-                     GLuint c;

-                     for (c = 0; c < 4; c++) {

-                        attrLeft[attr][c] += daInner[attr][c];

-                     }

-                  ATTRIB_LOOP_END

-#endif

-               }

-            } /*while lines>0*/

-

-         } /* for subTriangle */

-

-      }

-   }

-}

-

-#undef SETUP_CODE

-#undef RENDER_SPAN

-

-#undef PIXEL_TYPE

-#undef BYTES_PER_ROW

-#undef PIXEL_ADDRESS

-#undef DEPTH_TYPE

-

-#undef INTERP_Z

-#undef INTERP_RGB

-#undef INTERP_ALPHA

-#undef INTERP_INT_TEX

-#undef INTERP_ATTRIBS

-

-#undef S_SCALE

-#undef T_SCALE

-

-#undef FixedToDepth

-

-#undef NAME

+/*
+ * Mesa 3-D graphics library
+ * Version:  7.0
+ *
+ * Copyright (C) 1999-2007  Brian Paul   All Rights Reserved.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included
+ * in all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS
+ * OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * BRIAN PAUL BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN
+ * AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
+ */
+
+/*
+ * Triangle Rasterizer Template
+ *
+ * This file is #include'd to generate custom triangle rasterizers.
+ *
+ * The following macros may be defined to indicate what auxillary information
+ * must be interpolated across the triangle:
+ *    INTERP_Z        - if defined, interpolate integer Z values
+ *    INTERP_RGB      - if defined, interpolate integer RGB values
+ *    INTERP_ALPHA    - if defined, interpolate integer Alpha values
+ *    INTERP_INT_TEX  - if defined, interpolate integer ST texcoords
+ *                         (fast, simple 2-D texture mapping, without
+ *                         perspective correction)
+ *    INTERP_ATTRIBS  - if defined, interpolate arbitrary attribs (texcoords,
+ *                         varying vars, etc)  This also causes W to be
+ *                         computed for perspective correction).
+ *
+ * When one can directly address pixels in the color buffer the following
+ * macros can be defined and used to compute pixel addresses during
+ * rasterization (see pRow):
+ *    PIXEL_TYPE          - the datatype of a pixel (GLubyte, GLushort, GLuint)
+ *    BYTES_PER_ROW       - number of bytes per row in the color buffer
+ *    PIXEL_ADDRESS(X,Y)  - returns the address of pixel at (X,Y) where
+ *                          Y==0 at bottom of screen and increases upward.
+ *
+ * Similarly, for direct depth buffer access, this type is used for depth
+ * buffer addressing (see zRow):
+ *    DEPTH_TYPE          - either GLushort or GLuint
+ *
+ * Optionally, one may provide one-time setup code per triangle:
+ *    SETUP_CODE    - code which is to be executed once per triangle
+ *
+ * The following macro MUST be defined:
+ *    RENDER_SPAN(span) - code to write a span of pixels.
+ *
+ * This code was designed for the origin to be in the lower-left corner.
+ *
+ * Inspired by triangle rasterizer code written by Allen Akin.  Thanks Allen!
+ *
+ *
+ * Some notes on rasterization accuracy:
+ *
+ * This code uses fixed point arithmetic (the GLfixed type) to iterate
+ * over the triangle edges and interpolate ancillary data (such as Z,
+ * color, secondary color, etc).  The number of fractional bits in
+ * GLfixed and the value of SUB_PIXEL_BITS has a direct bearing on the
+ * accuracy of rasterization.
+ *
+ * If SUB_PIXEL_BITS=4 then we'll snap the vertices to the nearest
+ * 1/16 of a pixel.  If we're walking up a long, nearly vertical edge
+ * (dx=1/16, dy=1024) we'll need 4 + 10 = 14 fractional bits in
+ * GLfixed to walk the edge without error.  If the maximum viewport
+ * height is 4K pixels, then we'll need 4 + 12 = 16 fractional bits.
+ *
+ * Historically, Mesa has used 11 fractional bits in GLfixed, snaps
+ * vertices to 1/16 pixel and allowed a maximum viewport height of 2K
+ * pixels.  11 fractional bits is actually insufficient for accurately
+ * rasterizing some triangles.  More recently, the maximum viewport
+ * height was increased to 4K pixels.  Thus, Mesa should be using 16
+ * fractional bits in GLfixed.  Unfortunately, there may be some issues
+ * with setting FIXED_FRAC_BITS=16, such as multiplication overflow.
+ * This will have to be examined in some detail...
+ *
+ * For now, if you find rasterization errors, particularly with tall,
+ * sliver triangles, try increasing FIXED_FRAC_BITS and/or decreasing
+ * SUB_PIXEL_BITS.
+ */
+
+
+/*
+ * Some code we unfortunately need to prevent negative interpolated colors.
+ */
+#ifndef CLAMP_INTERPOLANT
+#define CLAMP_INTERPOLANT(CHANNEL, CHANNELSTEP, LEN)		\
+do {								\
+   GLfixed endVal = span.CHANNEL + (LEN) * span.CHANNELSTEP;	\
+   if (endVal < 0) {						\
+      span.CHANNEL -= endVal;					\
+   }								\
+   if (span.CHANNEL < 0) {					\
+      span.CHANNEL = 0;						\
+   }								\
+} while (0)
+#endif
+
+
+static void NAME(struct gl_context *ctx, const SWvertex *v0,
+                                 const SWvertex *v1,
+                                 const SWvertex *v2 )
+{
+   typedef struct {
+      const SWvertex *v0, *v1;   /* Y(v0) < Y(v1) */
+      GLfloat dx;	/* X(v1) - X(v0) */
+      GLfloat dy;	/* Y(v1) - Y(v0) */
+      GLfloat dxdy;	/* dx/dy */
+      GLfixed fdxdy;	/* dx/dy in fixed-point */
+      GLfloat adjy;	/* adjust from v[0]->fy to fsy, scaled */
+      GLfixed fsx;	/* first sample point x coord */
+      GLfixed fsy;
+      GLfixed fx0;	/* fixed pt X of lower endpoint */
+      GLint lines;	/* number of lines to be sampled on this edge */
+   } EdgeT;
+
+   const SWcontext *swrast = SWRAST_CONTEXT(ctx);
+#ifdef INTERP_Z
+   const GLint depthBits = ctx->DrawBuffer->Visual.depthBits;
+   const GLint fixedToDepthShift = depthBits <= 16 ? FIXED_SHIFT : 0;
+   const GLfloat maxDepth = ctx->DrawBuffer->_DepthMaxF;
+#define FixedToDepth(F)  ((F) >> fixedToDepthShift)
+#endif
+   EdgeT eMaj, eTop, eBot;
+   GLfloat oneOverArea;
+   const SWvertex *vMin, *vMid, *vMax;  /* Y(vMin)<=Y(vMid)<=Y(vMax) */
+   GLfloat bf = SWRAST_CONTEXT(ctx)->_BackfaceSign;
+   const GLint snapMask = ~((FIXED_ONE / (1 << SUB_PIXEL_BITS)) - 1); /* for x/y coord snapping */
+   GLfixed vMin_fx, vMin_fy, vMid_fx, vMid_fy, vMax_fx, vMax_fy;
+
+   SWspan span;
+
+   (void) swrast;
+
+   INIT_SPAN(span, GL_POLYGON);
+   span.y = 0; /* silence warnings */
+
+#ifdef INTERP_Z
+   (void) fixedToDepthShift;
+#endif
+
+   /*
+   printf("%s()\n", __FUNCTION__);
+   printf("  %g, %g, %g\n",
+          v0->attrib[FRAG_ATTRIB_WPOS][0],
+          v0->attrib[FRAG_ATTRIB_WPOS][1],
+          v0->attrib[FRAG_ATTRIB_WPOS][2]);
+   printf("  %g, %g, %g\n",
+          v1->attrib[FRAG_ATTRIB_WPOS][0],
+          v1->attrib[FRAG_ATTRIB_WPOS][1],
+          v1->attrib[FRAG_ATTRIB_WPOS][2]);
+   printf("  %g, %g, %g\n",
+          v2->attrib[FRAG_ATTRIB_WPOS][0],
+          v2->attrib[FRAG_ATTRIB_WPOS][1],
+          v2->attrib[FRAG_ATTRIB_WPOS][2]);
+   */
+
+   /* Compute fixed point x,y coords w/ half-pixel offsets and snapping.
+    * And find the order of the 3 vertices along the Y axis.
+    */
+   {
+      const GLfixed fy0 = FloatToFixed(v0->attrib[FRAG_ATTRIB_WPOS][1] - 0.5F) & snapMask;
+      const GLfixed fy1 = FloatToFixed(v1->attrib[FRAG_ATTRIB_WPOS][1] - 0.5F) & snapMask;
+      const GLfixed fy2 = FloatToFixed(v2->attrib[FRAG_ATTRIB_WPOS][1] - 0.5F) & snapMask;
+      if (fy0 <= fy1) {
+         if (fy1 <= fy2) {
+            /* y0 <= y1 <= y2 */
+            vMin = v0;   vMid = v1;   vMax = v2;
+            vMin_fy = fy0;  vMid_fy = fy1;  vMax_fy = fy2;
+         }
+         else if (fy2 <= fy0) {
+            /* y2 <= y0 <= y1 */
+            vMin = v2;   vMid = v0;   vMax = v1;
+            vMin_fy = fy2;  vMid_fy = fy0;  vMax_fy = fy1;
+         }
+         else {
+            /* y0 <= y2 <= y1 */
+            vMin = v0;   vMid = v2;   vMax = v1;
+            vMin_fy = fy0;  vMid_fy = fy2;  vMax_fy = fy1;
+            bf = -bf;
+         }
+      }
+      else {
+         if (fy0 <= fy2) {
+            /* y1 <= y0 <= y2 */
+            vMin = v1;   vMid = v0;   vMax = v2;
+            vMin_fy = fy1;  vMid_fy = fy0;  vMax_fy = fy2;
+            bf = -bf;
+         }
+         else if (fy2 <= fy1) {
+            /* y2 <= y1 <= y0 */
+            vMin = v2;   vMid = v1;   vMax = v0;
+            vMin_fy = fy2;  vMid_fy = fy1;  vMax_fy = fy0;
+            bf = -bf;
+         }
+         else {
+            /* y1 <= y2 <= y0 */
+            vMin = v1;   vMid = v2;   vMax = v0;
+            vMin_fy = fy1;  vMid_fy = fy2;  vMax_fy = fy0;
+         }
+      }
+
+      /* fixed point X coords */
+      vMin_fx = FloatToFixed(vMin->attrib[FRAG_ATTRIB_WPOS][0] + 0.5F) & snapMask;
+      vMid_fx = FloatToFixed(vMid->attrib[FRAG_ATTRIB_WPOS][0] + 0.5F) & snapMask;
+      vMax_fx = FloatToFixed(vMax->attrib[FRAG_ATTRIB_WPOS][0] + 0.5F) & snapMask;
+   }
+
+   /* vertex/edge relationship */
+   eMaj.v0 = vMin;   eMaj.v1 = vMax;   /*TODO: .v1's not needed */
+   eTop.v0 = vMid;   eTop.v1 = vMax;
+   eBot.v0 = vMin;   eBot.v1 = vMid;
+
+   /* compute deltas for each edge:  vertex[upper] - vertex[lower] */
+   eMaj.dx = FixedToFloat(vMax_fx - vMin_fx);
+   eMaj.dy = FixedToFloat(vMax_fy - vMin_fy);
+   eTop.dx = FixedToFloat(vMax_fx - vMid_fx);
+   eTop.dy = FixedToFloat(vMax_fy - vMid_fy);
+   eBot.dx = FixedToFloat(vMid_fx - vMin_fx);
+   eBot.dy = FixedToFloat(vMid_fy - vMin_fy);
+
+   /* compute area, oneOverArea and perform backface culling */
+   {
+      const GLfloat area = eMaj.dx * eBot.dy - eBot.dx * eMaj.dy;
+
+      if (IS_INF_OR_NAN(area) || area == 0.0F)
+         return;
+
+      if (area * bf * swrast->_BackfaceCullSign < 0.0)
+         return;
+
+      oneOverArea = 1.0F / area;
+
+      /* 0 = front, 1 = back */
+      span.facing = oneOverArea * bf > 0.0F;
+   }
+
+   /* Edge setup.  For a triangle strip these could be reused... */
+   {
+      eMaj.fsy = FixedCeil(vMin_fy);
+      eMaj.lines = FixedToInt(FixedCeil(vMax_fy - eMaj.fsy));
+      if (eMaj.lines > 0) {
+         eMaj.dxdy = eMaj.dx / eMaj.dy;
+         eMaj.fdxdy = SignedFloatToFixed(eMaj.dxdy);
+         eMaj.adjy = (GLfloat) (eMaj.fsy - vMin_fy);  /* SCALED! */
+         eMaj.fx0 = vMin_fx;
+         eMaj.fsx = eMaj.fx0 + (GLfixed) (eMaj.adjy * eMaj.dxdy);
+      }
+      else {
+         return;  /*CULLED*/
+      }
+
+      eTop.fsy = FixedCeil(vMid_fy);
+      eTop.lines = FixedToInt(FixedCeil(vMax_fy - eTop.fsy));
+      if (eTop.lines > 0) {
+         eTop.dxdy = eTop.dx / eTop.dy;
+         eTop.fdxdy = SignedFloatToFixed(eTop.dxdy);
+         eTop.adjy = (GLfloat) (eTop.fsy - vMid_fy); /* SCALED! */
+         eTop.fx0 = vMid_fx;
+         eTop.fsx = eTop.fx0 + (GLfixed) (eTop.adjy * eTop.dxdy);
+      }
+
+      eBot.fsy = FixedCeil(vMin_fy);
+      eBot.lines = FixedToInt(FixedCeil(vMid_fy - eBot.fsy));
+      if (eBot.lines > 0) {
+         eBot.dxdy = eBot.dx / eBot.dy;
+         eBot.fdxdy = SignedFloatToFixed(eBot.dxdy);
+         eBot.adjy = (GLfloat) (eBot.fsy - vMin_fy);  /* SCALED! */
+         eBot.fx0 = vMin_fx;
+         eBot.fsx = eBot.fx0 + (GLfixed) (eBot.adjy * eBot.dxdy);
+      }
+   }
+
+   /*
+    * Conceptually, we view a triangle as two subtriangles
+    * separated by a perfectly horizontal line.  The edge that is
+    * intersected by this line is one with maximal absolute dy; we
+    * call it a ``major'' edge.  The other two edges are the
+    * ``top'' edge (for the upper subtriangle) and the ``bottom''
+    * edge (for the lower subtriangle).  If either of these two
+    * edges is horizontal or very close to horizontal, the
+    * corresponding subtriangle might cover zero sample points;
+    * we take care to handle such cases, for performance as well
+    * as correctness.
+    *
+    * By stepping rasterization parameters along the major edge,
+    * we can avoid recomputing them at the discontinuity where
+    * the top and bottom edges meet.  However, this forces us to
+    * be able to scan both left-to-right and right-to-left.
+    * Also, we must determine whether the major edge is at the
+    * left or right side of the triangle.  We do this by
+    * computing the magnitude of the cross-product of the major
+    * and top edges.  Since this magnitude depends on the sine of
+    * the angle between the two edges, its sign tells us whether
+    * we turn to the left or to the right when travelling along
+    * the major edge to the top edge, and from this we infer
+    * whether the major edge is on the left or the right.
+    *
+    * Serendipitously, this cross-product magnitude is also a
+    * value we need to compute the iteration parameter
+    * derivatives for the triangle, and it can be used to perform
+    * backface culling because its sign tells us whether the
+    * triangle is clockwise or counterclockwise.  In this code we
+    * refer to it as ``area'' because it's also proportional to
+    * the pixel area of the triangle.
+    */
+
+   {
+      GLint scan_from_left_to_right;  /* true if scanning left-to-right */
+
+      /*
+       * Execute user-supplied setup code
+       */
+#ifdef SETUP_CODE
+      SETUP_CODE
+#endif
+
+      scan_from_left_to_right = (oneOverArea < 0.0F);
+
+
+      /* compute d?/dx and d?/dy derivatives */
+#ifdef INTERP_Z
+      span.interpMask |= SPAN_Z;
+      {
+         GLfloat eMaj_dz = vMax->attrib[FRAG_ATTRIB_WPOS][2] - vMin->attrib[FRAG_ATTRIB_WPOS][2];
+         GLfloat eBot_dz = vMid->attrib[FRAG_ATTRIB_WPOS][2] - vMin->attrib[FRAG_ATTRIB_WPOS][2];
+         span.attrStepX[FRAG_ATTRIB_WPOS][2] = oneOverArea * (eMaj_dz * eBot.dy - eMaj.dy * eBot_dz);
+         if (span.attrStepX[FRAG_ATTRIB_WPOS][2] > maxDepth ||
+             span.attrStepX[FRAG_ATTRIB_WPOS][2] < -maxDepth) {
+            /* probably a sliver triangle */
+            span.attrStepX[FRAG_ATTRIB_WPOS][2] = 0.0;
+            span.attrStepY[FRAG_ATTRIB_WPOS][2] = 0.0;
+         }
+         else {
+            span.attrStepY[FRAG_ATTRIB_WPOS][2] = oneOverArea * (eMaj.dx * eBot_dz - eMaj_dz * eBot.dx);
+         }
+         if (depthBits <= 16)
+            span.zStep = SignedFloatToFixed(span.attrStepX[FRAG_ATTRIB_WPOS][2]);
+         else
+            span.zStep = (GLint) span.attrStepX[FRAG_ATTRIB_WPOS][2];
+      }
+#endif
+#ifdef INTERP_RGB
+      span.interpMask |= SPAN_RGBA;
+      if (ctx->Light.ShadeModel == GL_SMOOTH) {
+         GLfloat eMaj_dr = (GLfloat) (vMax->color[RCOMP] - vMin->color[RCOMP]);
+         GLfloat eBot_dr = (GLfloat) (vMid->color[RCOMP] - vMin->color[RCOMP]);
+         GLfloat eMaj_dg = (GLfloat) (vMax->color[GCOMP] - vMin->color[GCOMP]);
+         GLfloat eBot_dg = (GLfloat) (vMid->color[GCOMP] - vMin->color[GCOMP]);
+         GLfloat eMaj_db = (GLfloat) (vMax->color[BCOMP] - vMin->color[BCOMP]);
+         GLfloat eBot_db = (GLfloat) (vMid->color[BCOMP] - vMin->color[BCOMP]);
+#  ifdef INTERP_ALPHA
+         GLfloat eMaj_da = (GLfloat) (vMax->color[ACOMP] - vMin->color[ACOMP]);
+         GLfloat eBot_da = (GLfloat) (vMid->color[ACOMP] - vMin->color[ACOMP]);
+#  endif
+         span.attrStepX[FRAG_ATTRIB_COL0][0] = oneOverArea * (eMaj_dr * eBot.dy - eMaj.dy * eBot_dr);
+         span.attrStepY[FRAG_ATTRIB_COL0][0] = oneOverArea * (eMaj.dx * eBot_dr - eMaj_dr * eBot.dx);
+         span.attrStepX[FRAG_ATTRIB_COL0][1] = oneOverArea * (eMaj_dg * eBot.dy - eMaj.dy * eBot_dg);
+         span.attrStepY[FRAG_ATTRIB_COL0][1] = oneOverArea * (eMaj.dx * eBot_dg - eMaj_dg * eBot.dx);
+         span.attrStepX[FRAG_ATTRIB_COL0][2] = oneOverArea * (eMaj_db * eBot.dy - eMaj.dy * eBot_db);
+         span.attrStepY[FRAG_ATTRIB_COL0][2] = oneOverArea * (eMaj.dx * eBot_db - eMaj_db * eBot.dx);
+         span.redStep   = SignedFloatToFixed(span.attrStepX[FRAG_ATTRIB_COL0][0]);
+         span.greenStep = SignedFloatToFixed(span.attrStepX[FRAG_ATTRIB_COL0][1]);
+         span.blueStep  = SignedFloatToFixed(span.attrStepX[FRAG_ATTRIB_COL0][2]);
+#  ifdef INTERP_ALPHA
+         span.attrStepX[FRAG_ATTRIB_COL0][3] = oneOverArea * (eMaj_da * eBot.dy - eMaj.dy * eBot_da);
+         span.attrStepY[FRAG_ATTRIB_COL0][3] = oneOverArea * (eMaj.dx * eBot_da - eMaj_da * eBot.dx);
+         span.alphaStep = SignedFloatToFixed(span.attrStepX[FRAG_ATTRIB_COL0][3]);
+#  endif /* INTERP_ALPHA */
+      }
+      else {
+         ASSERT(ctx->Light.ShadeModel == GL_FLAT);
+         span.interpMask |= SPAN_FLAT;
+         span.attrStepX[FRAG_ATTRIB_COL0][0] = span.attrStepY[FRAG_ATTRIB_COL0][0] = 0.0F;
+         span.attrStepX[FRAG_ATTRIB_COL0][1] = span.attrStepY[FRAG_ATTRIB_COL0][1] = 0.0F;
+         span.attrStepX[FRAG_ATTRIB_COL0][2] = span.attrStepY[FRAG_ATTRIB_COL0][2] = 0.0F;
+	 span.redStep   = 0;
+	 span.greenStep = 0;
+	 span.blueStep  = 0;
+#  ifdef INTERP_ALPHA
+         span.attrStepX[FRAG_ATTRIB_COL0][3] = span.attrStepY[FRAG_ATTRIB_COL0][3] = 0.0F;
+	 span.alphaStep = 0;
+#  endif
+      }
+#endif /* INTERP_RGB */
+#ifdef INTERP_INT_TEX
+      {
+         GLfloat eMaj_ds = (vMax->attrib[FRAG_ATTRIB_TEX0][0] - vMin->attrib[FRAG_ATTRIB_TEX0][0]) * S_SCALE;
+         GLfloat eBot_ds = (vMid->attrib[FRAG_ATTRIB_TEX0][0] - vMin->attrib[FRAG_ATTRIB_TEX0][0]) * S_SCALE;
+         GLfloat eMaj_dt = (vMax->attrib[FRAG_ATTRIB_TEX0][1] - vMin->attrib[FRAG_ATTRIB_TEX0][1]) * T_SCALE;
+         GLfloat eBot_dt = (vMid->attrib[FRAG_ATTRIB_TEX0][1] - vMin->attrib[FRAG_ATTRIB_TEX0][1]) * T_SCALE;
+         span.attrStepX[FRAG_ATTRIB_TEX0][0] = oneOverArea * (eMaj_ds * eBot.dy - eMaj.dy * eBot_ds);
+         span.attrStepY[FRAG_ATTRIB_TEX0][0] = oneOverArea * (eMaj.dx * eBot_ds - eMaj_ds * eBot.dx);
+         span.attrStepX[FRAG_ATTRIB_TEX0][1] = oneOverArea * (eMaj_dt * eBot.dy - eMaj.dy * eBot_dt);
+         span.attrStepY[FRAG_ATTRIB_TEX0][1] = oneOverArea * (eMaj.dx * eBot_dt - eMaj_dt * eBot.dx);
+         span.intTexStep[0] = SignedFloatToFixed(span.attrStepX[FRAG_ATTRIB_TEX0][0]);
+         span.intTexStep[1] = SignedFloatToFixed(span.attrStepX[FRAG_ATTRIB_TEX0][1]);
+      }
+#endif
+#ifdef INTERP_ATTRIBS
+      {
+         /* attrib[FRAG_ATTRIB_WPOS][3] is 1/W */
+         const GLfloat wMax = vMax->attrib[FRAG_ATTRIB_WPOS][3];
+         const GLfloat wMin = vMin->attrib[FRAG_ATTRIB_WPOS][3];
+         const GLfloat wMid = vMid->attrib[FRAG_ATTRIB_WPOS][3];
+         {
+            const GLfloat eMaj_dw = wMax - wMin;
+            const GLfloat eBot_dw = wMid - wMin;
+            span.attrStepX[FRAG_ATTRIB_WPOS][3] = oneOverArea * (eMaj_dw * eBot.dy - eMaj.dy * eBot_dw);
+            span.attrStepY[FRAG_ATTRIB_WPOS][3] = oneOverArea * (eMaj.dx * eBot_dw - eMaj_dw * eBot.dx);
+         }
+         ATTRIB_LOOP_BEGIN
+            if (swrast->_InterpMode[attr] == GL_FLAT) {
+               ASSIGN_4V(span.attrStepX[attr], 0.0, 0.0, 0.0, 0.0);
+               ASSIGN_4V(span.attrStepY[attr], 0.0, 0.0, 0.0, 0.0);
+            }
+            else {
+               GLuint c;
+               for (c = 0; c < 4; c++) {
+                  GLfloat eMaj_da = vMax->attrib[attr][c] * wMax - vMin->attrib[attr][c] * wMin;
+                  GLfloat eBot_da = vMid->attrib[attr][c] * wMid - vMin->attrib[attr][c] * wMin;
+                  span.attrStepX[attr][c] = oneOverArea * (eMaj_da * eBot.dy - eMaj.dy * eBot_da);
+                  span.attrStepY[attr][c] = oneOverArea * (eMaj.dx * eBot_da - eMaj_da * eBot.dx);
+               }
+            }
+         ATTRIB_LOOP_END
+      }
+#endif
+
+      /*
+       * We always sample at pixel centers.  However, we avoid
+       * explicit half-pixel offsets in this code by incorporating
+       * the proper offset in each of x and y during the
+       * transformation to window coordinates.
+       *
+       * We also apply the usual rasterization rules to prevent
+       * cracks and overlaps.  A pixel is considered inside a
+       * subtriangle if it meets all of four conditions: it is on or
+       * to the right of the left edge, strictly to the left of the
+       * right edge, on or below the top edge, and strictly above
+       * the bottom edge.  (Some edges may be degenerate.)
+       *
+       * The following discussion assumes left-to-right scanning
+       * (that is, the major edge is on the left); the right-to-left
+       * case is a straightforward variation.
+       *
+       * We start by finding the half-integral y coordinate that is
+       * at or below the top of the triangle.  This gives us the
+       * first scan line that could possibly contain pixels that are
+       * inside the triangle.
+       *
+       * Next we creep down the major edge until we reach that y,
+       * and compute the corresponding x coordinate on the edge.
+       * Then we find the half-integral x that lies on or just
+       * inside the edge.  This is the first pixel that might lie in
+       * the interior of the triangle.  (We won't know for sure
+       * until we check the other edges.)
+       *
+       * As we rasterize the triangle, we'll step down the major
+       * edge.  For each step in y, we'll move an integer number
+       * of steps in x.  There are two possible x step sizes, which
+       * we'll call the ``inner'' step (guaranteed to land on the
+       * edge or inside it) and the ``outer'' step (guaranteed to
+       * land on the edge or outside it).  The inner and outer steps
+       * differ by one.  During rasterization we maintain an error
+       * term that indicates our distance from the true edge, and
+       * select either the inner step or the outer step, whichever
+       * gets us to the first pixel that falls inside the triangle.
+       *
+       * All parameters (z, red, etc.) as well as the buffer
+       * addresses for color and z have inner and outer step values,
+       * so that we can increment them appropriately.  This method
+       * eliminates the need to adjust parameters by creeping a
+       * sub-pixel amount into the triangle at each scanline.
+       */
+
+      {
+         GLint subTriangle;
+         GLfixed fxLeftEdge = 0, fxRightEdge = 0;
+         GLfixed fdxLeftEdge = 0, fdxRightEdge = 0;
+         GLfixed fError = 0, fdError = 0;
+#ifdef PIXEL_ADDRESS
+         PIXEL_TYPE *pRow = NULL;
+         GLint dPRowOuter = 0, dPRowInner;  /* offset in bytes */
+#endif
+#ifdef INTERP_Z
+#  ifdef DEPTH_TYPE
+         struct gl_renderbuffer *zrb
+            = ctx->DrawBuffer->Attachment[BUFFER_DEPTH].Renderbuffer;
+         DEPTH_TYPE *zRow = NULL;
+         GLint dZRowOuter = 0, dZRowInner;  /* offset in bytes */
+#  endif
+         GLuint zLeft = 0;
+         GLfixed fdzOuter = 0, fdzInner;
+#endif
+#ifdef INTERP_RGB
+         GLint rLeft = 0, fdrOuter = 0, fdrInner;
+         GLint gLeft = 0, fdgOuter = 0, fdgInner;
+         GLint bLeft = 0, fdbOuter = 0, fdbInner;
+#endif
+#ifdef INTERP_ALPHA
+         GLint aLeft = 0, fdaOuter = 0, fdaInner;
+#endif
+#ifdef INTERP_INT_TEX
+         GLfixed sLeft=0, dsOuter=0, dsInner;
+         GLfixed tLeft=0, dtOuter=0, dtInner;
+#endif
+#ifdef INTERP_ATTRIBS
+         GLfloat wLeft = 0, dwOuter = 0, dwInner;
+         GLfloat attrLeft[FRAG_ATTRIB_MAX][4];
+         GLfloat daOuter[FRAG_ATTRIB_MAX][4], daInner[FRAG_ATTRIB_MAX][4];
+#endif
+
+         for (subTriangle=0; subTriangle<=1; subTriangle++) {
+            EdgeT *eLeft, *eRight;
+            int setupLeft, setupRight;
+            int lines;
+
+            if (subTriangle==0) {
+               /* bottom half */
+               if (scan_from_left_to_right) {
+                  eLeft = &eMaj;
+                  eRight = &eBot;
+                  lines = eRight->lines;
+                  setupLeft = 1;
+                  setupRight = 1;
+               }
+               else {
+                  eLeft = &eBot;
+                  eRight = &eMaj;
+                  lines = eLeft->lines;
+                  setupLeft = 1;
+                  setupRight = 1;
+               }
+            }
+            else {
+               /* top half */
+               if (scan_from_left_to_right) {
+                  eLeft = &eMaj;
+                  eRight = &eTop;
+                  lines = eRight->lines;
+                  setupLeft = 0;
+                  setupRight = 1;
+               }
+               else {
+                  eLeft = &eTop;
+                  eRight = &eMaj;
+                  lines = eLeft->lines;
+                  setupLeft = 1;
+                  setupRight = 0;
+               }
+               if (lines == 0)
+                  return;
+            }
+
+            if (setupLeft && eLeft->lines > 0) {
+               const SWvertex *vLower = eLeft->v0;
+               const GLfixed fsy = eLeft->fsy;
+               const GLfixed fsx = eLeft->fsx;  /* no fractional part */
+               const GLfixed fx = FixedCeil(fsx);  /* no fractional part */
+               const GLfixed adjx = (GLfixed) (fx - eLeft->fx0); /* SCALED! */
+               const GLfixed adjy = (GLfixed) eLeft->adjy;      /* SCALED! */
+               GLint idxOuter;
+               GLfloat dxOuter;
+               GLfixed fdxOuter;
+
+               fError = fx - fsx - FIXED_ONE;
+               fxLeftEdge = fsx - FIXED_EPSILON;
+               fdxLeftEdge = eLeft->fdxdy;
+               fdxOuter = FixedFloor(fdxLeftEdge - FIXED_EPSILON);
+               fdError = fdxOuter - fdxLeftEdge + FIXED_ONE;
+               idxOuter = FixedToInt(fdxOuter);
+               dxOuter = (GLfloat) idxOuter;
+               span.y = FixedToInt(fsy);
+
+               /* silence warnings on some compilers */
+               (void) dxOuter;
+               (void) adjx;
+               (void) adjy;
+               (void) vLower;
+
+#ifdef PIXEL_ADDRESS
+               {
+                  pRow = (PIXEL_TYPE *) PIXEL_ADDRESS(FixedToInt(fxLeftEdge), span.y);
+                  dPRowOuter = -((int)BYTES_PER_ROW) + idxOuter * sizeof(PIXEL_TYPE);
+                  /* negative because Y=0 at bottom and increases upward */
+               }
+#endif
+               /*
+                * Now we need the set of parameter (z, color, etc.) values at
+                * the point (fx, fsy).  This gives us properly-sampled parameter
+                * values that we can step from pixel to pixel.  Furthermore,
+                * although we might have intermediate results that overflow
+                * the normal parameter range when we step temporarily outside
+                * the triangle, we shouldn't overflow or underflow for any
+                * pixel that's actually inside the triangle.
+                */
+
+#ifdef INTERP_Z
+               {
+                  GLfloat z0 = vLower->attrib[FRAG_ATTRIB_WPOS][2];
+                  if (depthBits <= 16) {
+                     /* interpolate fixed-pt values */
+                     GLfloat tmp = (z0 * FIXED_SCALE
+                                    + span.attrStepX[FRAG_ATTRIB_WPOS][2] * adjx
+                                    + span.attrStepY[FRAG_ATTRIB_WPOS][2] * adjy) + FIXED_HALF;
+                     if (tmp < MAX_GLUINT / 2)
+                        zLeft = (GLfixed) tmp;
+                     else
+                        zLeft = MAX_GLUINT / 2;
+                     fdzOuter = SignedFloatToFixed(span.attrStepY[FRAG_ATTRIB_WPOS][2] +
+                                                   dxOuter * span.attrStepX[FRAG_ATTRIB_WPOS][2]);
+                  }
+                  else {
+                     /* interpolate depth values w/out scaling */
+                     zLeft = (GLuint) (z0 + span.attrStepX[FRAG_ATTRIB_WPOS][2] * FixedToFloat(adjx)
+                                          + span.attrStepY[FRAG_ATTRIB_WPOS][2] * FixedToFloat(adjy));
+                     fdzOuter = (GLint) (span.attrStepY[FRAG_ATTRIB_WPOS][2] +
+                                         dxOuter * span.attrStepX[FRAG_ATTRIB_WPOS][2]);
+                  }
+#  ifdef DEPTH_TYPE
+                  zRow = (DEPTH_TYPE *)
+                    _swrast_pixel_address(zrb, FixedToInt(fxLeftEdge), span.y);
+                  dZRowOuter = (ctx->DrawBuffer->Width + idxOuter) * sizeof(DEPTH_TYPE);
+#  endif
+               }
+#endif
+#ifdef INTERP_RGB
+               if (ctx->Light.ShadeModel == GL_SMOOTH) {
+                  rLeft = (GLint)(ChanToFixed(vLower->color[RCOMP])
+                                  + span.attrStepX[FRAG_ATTRIB_COL0][0] * adjx
+                                  + span.attrStepY[FRAG_ATTRIB_COL0][0] * adjy) + FIXED_HALF;
+                  gLeft = (GLint)(ChanToFixed(vLower->color[GCOMP])
+                                  + span.attrStepX[FRAG_ATTRIB_COL0][1] * adjx
+                                  + span.attrStepY[FRAG_ATTRIB_COL0][1] * adjy) + FIXED_HALF;
+                  bLeft = (GLint)(ChanToFixed(vLower->color[BCOMP])
+                                  + span.attrStepX[FRAG_ATTRIB_COL0][2] * adjx
+                                  + span.attrStepY[FRAG_ATTRIB_COL0][2] * adjy) + FIXED_HALF;
+                  fdrOuter = SignedFloatToFixed(span.attrStepY[FRAG_ATTRIB_COL0][0]
+                                                + dxOuter * span.attrStepX[FRAG_ATTRIB_COL0][0]);
+                  fdgOuter = SignedFloatToFixed(span.attrStepY[FRAG_ATTRIB_COL0][1]
+                                                + dxOuter * span.attrStepX[FRAG_ATTRIB_COL0][1]);
+                  fdbOuter = SignedFloatToFixed(span.attrStepY[FRAG_ATTRIB_COL0][2]
+                                                + dxOuter * span.attrStepX[FRAG_ATTRIB_COL0][2]);
+#  ifdef INTERP_ALPHA
+                  aLeft = (GLint)(ChanToFixed(vLower->color[ACOMP])
+                                  + span.attrStepX[FRAG_ATTRIB_COL0][3] * adjx
+                                  + span.attrStepY[FRAG_ATTRIB_COL0][3] * adjy) + FIXED_HALF;
+                  fdaOuter = SignedFloatToFixed(span.attrStepY[FRAG_ATTRIB_COL0][3]
+                                                + dxOuter * span.attrStepX[FRAG_ATTRIB_COL0][3]);
+#  endif
+               }
+               else {
+                  ASSERT(ctx->Light.ShadeModel == GL_FLAT);
+                  rLeft = ChanToFixed(v2->color[RCOMP]);
+                  gLeft = ChanToFixed(v2->color[GCOMP]);
+                  bLeft = ChanToFixed(v2->color[BCOMP]);
+                  fdrOuter = fdgOuter = fdbOuter = 0;
+#  ifdef INTERP_ALPHA
+                  aLeft = ChanToFixed(v2->color[ACOMP]);
+                  fdaOuter = 0;
+#  endif
+               }
+#endif /* INTERP_RGB */
+
+
+#ifdef INTERP_INT_TEX
+               {
+                  GLfloat s0, t0;
+                  s0 = vLower->attrib[FRAG_ATTRIB_TEX0][0] * S_SCALE;
+                  sLeft = (GLfixed)(s0 * FIXED_SCALE + span.attrStepX[FRAG_ATTRIB_TEX0][0] * adjx
+                                 + span.attrStepY[FRAG_ATTRIB_TEX0][0] * adjy) + FIXED_HALF;
+                  dsOuter = SignedFloatToFixed(span.attrStepY[FRAG_ATTRIB_TEX0][0]
+                                               + dxOuter * span.attrStepX[FRAG_ATTRIB_TEX0][0]);
+
+                  t0 = vLower->attrib[FRAG_ATTRIB_TEX0][1] * T_SCALE;
+                  tLeft = (GLfixed)(t0 * FIXED_SCALE + span.attrStepX[FRAG_ATTRIB_TEX0][1] * adjx
+                                 + span.attrStepY[FRAG_ATTRIB_TEX0][1] * adjy) + FIXED_HALF;
+                  dtOuter = SignedFloatToFixed(span.attrStepY[FRAG_ATTRIB_TEX0][1]
+                                               + dxOuter * span.attrStepX[FRAG_ATTRIB_TEX0][1]);
+               }
+#endif
+#ifdef INTERP_ATTRIBS
+               {
+                  const GLuint attr = FRAG_ATTRIB_WPOS;
+                  wLeft = vLower->attrib[FRAG_ATTRIB_WPOS][3]
+                        + (span.attrStepX[attr][3] * adjx
+                           + span.attrStepY[attr][3] * adjy) * (1.0F/FIXED_SCALE);
+                  dwOuter = span.attrStepY[attr][3] + dxOuter * span.attrStepX[attr][3];
+               }
+               ATTRIB_LOOP_BEGIN
+                  const GLfloat invW = vLower->attrib[FRAG_ATTRIB_WPOS][3];
+                  if (swrast->_InterpMode[attr] == GL_FLAT) {
+                     GLuint c;
+                     for (c = 0; c < 4; c++) {
+                        attrLeft[attr][c] = v2->attrib[attr][c] * invW;
+                        daOuter[attr][c] = 0.0;
+                     }
+                  }
+                  else {
+                     GLuint c;
+                     for (c = 0; c < 4; c++) {
+                        const GLfloat a = vLower->attrib[attr][c] * invW;
+                        attrLeft[attr][c] = a + (  span.attrStepX[attr][c] * adjx
+                                                 + span.attrStepY[attr][c] * adjy) * (1.0F/FIXED_SCALE);
+                        daOuter[attr][c] = span.attrStepY[attr][c] + dxOuter * span.attrStepX[attr][c];
+                     }
+                  }
+               ATTRIB_LOOP_END
+#endif
+            } /*if setupLeft*/
+
+
+            if (setupRight && eRight->lines>0) {
+               fxRightEdge = eRight->fsx - FIXED_EPSILON;
+               fdxRightEdge = eRight->fdxdy;
+            }
+
+            if (lines==0) {
+               continue;
+            }
+
+
+            /* Rasterize setup */
+#ifdef PIXEL_ADDRESS
+            dPRowInner = dPRowOuter + sizeof(PIXEL_TYPE);
+#endif
+#ifdef INTERP_Z
+#  ifdef DEPTH_TYPE
+            dZRowInner = dZRowOuter + sizeof(DEPTH_TYPE);
+#  endif
+            fdzInner = fdzOuter + span.zStep;
+#endif
+#ifdef INTERP_RGB
+            fdrInner = fdrOuter + span.redStep;
+            fdgInner = fdgOuter + span.greenStep;
+            fdbInner = fdbOuter + span.blueStep;
+#endif
+#ifdef INTERP_ALPHA
+            fdaInner = fdaOuter + span.alphaStep;
+#endif
+#ifdef INTERP_INT_TEX
+            dsInner = dsOuter + span.intTexStep[0];
+            dtInner = dtOuter + span.intTexStep[1];
+#endif
+#ifdef INTERP_ATTRIBS
+            dwInner = dwOuter + span.attrStepX[FRAG_ATTRIB_WPOS][3];
+            ATTRIB_LOOP_BEGIN
+               GLuint c;
+               for (c = 0; c < 4; c++) {
+                  daInner[attr][c] = daOuter[attr][c] + span.attrStepX[attr][c];
+               }
+            ATTRIB_LOOP_END
+#endif
+
+            while (lines > 0) {
+               /* initialize the span interpolants to the leftmost value */
+               /* ff = fixed-pt fragment */
+               const GLint right = FixedToInt(fxRightEdge);
+               span.x = FixedToInt(fxLeftEdge);
+               if (right <= span.x)
+                  span.end = 0;
+               else
+                  span.end = right - span.x;
+
+#ifdef INTERP_Z
+               span.z = zLeft;
+#endif
+#ifdef INTERP_RGB
+               span.red = rLeft;
+               span.green = gLeft;
+               span.blue = bLeft;
+#endif
+#ifdef INTERP_ALPHA
+               span.alpha = aLeft;
+#endif
+#ifdef INTERP_INT_TEX
+               span.intTex[0] = sLeft;
+               span.intTex[1] = tLeft;
+#endif
+
+#ifdef INTERP_ATTRIBS
+               span.attrStart[FRAG_ATTRIB_WPOS][3] = wLeft;
+               ATTRIB_LOOP_BEGIN
+                  GLuint c;
+                  for (c = 0; c < 4; c++) {
+                     span.attrStart[attr][c] = attrLeft[attr][c];
+                  }
+               ATTRIB_LOOP_END
+#endif
+
+               /* This is where we actually generate fragments */
+               /* XXX the test for span.y > 0 _shouldn't_ be needed but
+                * it fixes a problem on 64-bit Opterons (bug 4842).
+                */
+               if (span.end > 0 && span.y >= 0) {
+                  const GLint len = span.end - 1;
+                  (void) len;
+#ifdef INTERP_RGB
+                  CLAMP_INTERPOLANT(red, redStep, len);
+                  CLAMP_INTERPOLANT(green, greenStep, len);
+                  CLAMP_INTERPOLANT(blue, blueStep, len);
+#endif
+#ifdef INTERP_ALPHA
+                  CLAMP_INTERPOLANT(alpha, alphaStep, len);
+#endif
+                  {
+                     RENDER_SPAN( span );
+                  }
+               }
+
+               /*
+                * Advance to the next scan line.  Compute the
+                * new edge coordinates, and adjust the
+                * pixel-center x coordinate so that it stays
+                * on or inside the major edge.
+                */
+               span.y++;
+               lines--;
+
+               fxLeftEdge += fdxLeftEdge;
+               fxRightEdge += fdxRightEdge;
+
+               fError += fdError;
+               if (fError >= 0) {
+                  fError -= FIXED_ONE;
+
+#ifdef PIXEL_ADDRESS
+                  pRow = (PIXEL_TYPE *) ((GLubyte *) pRow + dPRowOuter);
+#endif
+#ifdef INTERP_Z
+#  ifdef DEPTH_TYPE
+                  zRow = (DEPTH_TYPE *) ((GLubyte *) zRow + dZRowOuter);
+#  endif
+                  zLeft += fdzOuter;
+#endif
+#ifdef INTERP_RGB
+                  rLeft += fdrOuter;
+                  gLeft += fdgOuter;
+                  bLeft += fdbOuter;
+#endif
+#ifdef INTERP_ALPHA
+                  aLeft += fdaOuter;
+#endif
+#ifdef INTERP_INT_TEX
+                  sLeft += dsOuter;
+                  tLeft += dtOuter;
+#endif
+#ifdef INTERP_ATTRIBS
+                  wLeft += dwOuter;
+                  ATTRIB_LOOP_BEGIN
+                     GLuint c;
+                     for (c = 0; c < 4; c++) {
+                        attrLeft[attr][c] += daOuter[attr][c];
+                     }
+                  ATTRIB_LOOP_END
+#endif
+               }
+               else {
+#ifdef PIXEL_ADDRESS
+                  pRow = (PIXEL_TYPE *) ((GLubyte *) pRow + dPRowInner);
+#endif
+#ifdef INTERP_Z
+#  ifdef DEPTH_TYPE
+                  zRow = (DEPTH_TYPE *) ((GLubyte *) zRow + dZRowInner);
+#  endif
+                  zLeft += fdzInner;
+#endif
+#ifdef INTERP_RGB
+                  rLeft += fdrInner;
+                  gLeft += fdgInner;
+                  bLeft += fdbInner;
+#endif
+#ifdef INTERP_ALPHA
+                  aLeft += fdaInner;
+#endif
+#ifdef INTERP_INT_TEX
+                  sLeft += dsInner;
+                  tLeft += dtInner;
+#endif
+#ifdef INTERP_ATTRIBS
+                  wLeft += dwInner;
+                  ATTRIB_LOOP_BEGIN
+                     GLuint c;
+                     for (c = 0; c < 4; c++) {
+                        attrLeft[attr][c] += daInner[attr][c];
+                     }
+                  ATTRIB_LOOP_END
+#endif
+               }
+            } /*while lines>0*/
+
+         } /* for subTriangle */
+
+      }
+   }
+}
+
+#undef SETUP_CODE
+#undef RENDER_SPAN
+
+#undef PIXEL_TYPE
+#undef BYTES_PER_ROW
+#undef PIXEL_ADDRESS
+#undef DEPTH_TYPE
+
+#undef INTERP_Z
+#undef INTERP_RGB
+#undef INTERP_ALPHA
+#undef INTERP_INT_TEX
+#undef INTERP_ATTRIBS
+
+#undef S_SCALE
+#undef T_SCALE
+
+#undef FixedToDepth
+
+#undef NAME
diff --git a/mesalib/src/mesa/swrast/s_zoom.c b/mesalib/src/mesa/swrast/s_zoom.c
index f407fdcf4..73bff482b 100644
--- a/mesalib/src/mesa/swrast/s_zoom.c
+++ b/mesalib/src/mesa/swrast/s_zoom.c
@@ -25,6 +25,7 @@
 #include "main/glheader.h"
 #include "main/macros.h"
 #include "main/imports.h"
+#include "main/format_pack.h"
 #include "main/colormac.h"
 
 #include "s_context.h"
@@ -390,17 +391,17 @@ _swrast_write_zoomed_stencil_span(struct gl_context *ctx, GLint imgX, GLint imgY
 
 
 /**
- * Zoom/write z values (16 or 32-bit).
+ * Zoom/write 32-bit Z values.
  * No per-fragment operations are applied.
  */
 void
 _swrast_write_zoomed_z_span(struct gl_context *ctx, GLint imgX, GLint imgY,
                             GLint width, GLint spanX, GLint spanY,
-                            const GLvoid *z)
+                            const GLuint *zVals)
 {
-   struct gl_renderbuffer *rb = ctx->DrawBuffer->_DepthBuffer;
-   GLushort zoomedVals16[MAX_WIDTH];
-   GLuint zoomedVals32[MAX_WIDTH];
+   struct gl_renderbuffer *rb =
+      ctx->DrawBuffer->Attachment[BUFFER_DEPTH].Renderbuffer;
+   GLuint zoomedVals[MAX_WIDTH];
    GLint x0, x1, y0, y1, y;
    GLint i, zoomedWidth;
 
@@ -414,28 +415,16 @@ _swrast_write_zoomed_z_span(struct gl_context *ctx, GLint imgX, GLint imgY,
    ASSERT(zoomedWidth <= MAX_WIDTH);
 
    /* zoom the span horizontally */
-   if (rb->DataType == GL_UNSIGNED_SHORT) {
-      for (i = 0; i < zoomedWidth; i++) {
-         GLint j = unzoom_x(ctx->Pixel.ZoomX, imgX, x0 + i) - spanX;
-         ASSERT(j >= 0);
-         ASSERT(j < width);
-         zoomedVals16[i] = ((GLushort *) z)[j];
-      }
-      z = zoomedVals16;
-   }
-   else {
-      ASSERT(rb->DataType == GL_UNSIGNED_INT);
-      for (i = 0; i < zoomedWidth; i++) {
-         GLint j = unzoom_x(ctx->Pixel.ZoomX, imgX, x0 + i) - spanX;
-         ASSERT(j >= 0);
-         ASSERT(j < width);
-         zoomedVals32[i] = ((GLuint *) z)[j];
-      }
-      z = zoomedVals32;
+   for (i = 0; i < zoomedWidth; i++) {
+      GLint j = unzoom_x(ctx->Pixel.ZoomX, imgX, x0 + i) - spanX;
+      ASSERT(j >= 0);
+      ASSERT(j < width);
+      zoomedVals[i] = zVals[j];
    }
 
    /* write the zoomed spans */
    for (y = y0; y < y1; y++) {
-      rb->PutRow(ctx, rb, zoomedWidth, x0, y, z, NULL);
+      GLubyte *dst = _swrast_pixel_address(rb, x0, y);
+      _mesa_pack_uint_z_row(rb->Format, zoomedWidth, zoomedVals, dst);
    }
 }
diff --git a/mesalib/src/mesa/swrast/s_zoom.h b/mesalib/src/mesa/swrast/s_zoom.h
index 0b82bb824..1955e7e88 100644
--- a/mesalib/src/mesa/swrast/s_zoom.h
+++ b/mesalib/src/mesa/swrast/s_zoom.h
@@ -50,7 +50,7 @@ _swrast_write_zoomed_stencil_span(struct gl_context *ctx, GLint imgX, GLint imgY
 extern void
 _swrast_write_zoomed_z_span(struct gl_context *ctx, GLint imgX, GLint imgY,
                             GLint width, GLint spanX, GLint spanY,
-                            const GLvoid *z);
+                            const GLuint *zVals);
 
 
 #endif