Merge commit 'origin/master' into glsl-pp-rework-2

Conflicts:
	src/gallium/winsys/gdi/SConscript

29 files changed, 717 insertions, 299 deletions

diff --git a/src/mesa/drivers/common/meta.c b/src/mesa/drivers/common/meta.c
index 05909cfa30..aa7c73ba36 100644
--- a/src/mesa/drivers/common/meta.c
+++ b/src/mesa/drivers/common/meta.c
@@ -68,6 +68,29 @@
 
 
 /**
+ * Flags passed to _mesa_meta_begin().
+ */
+/*@{*/
+#define META_ALL              ~0x0
+#define META_ALPHA_TEST        0x1
+#define META_BLEND             0x2  /**< includes logicop */
+#define META_COLOR_MASK        0x4
+#define META_DEPTH_TEST        0x8
+#define META_FOG              0x10
+#define META_PIXEL_STORE      0x20
+#define META_PIXEL_TRANSFER   0x40
+#define META_RASTERIZATION    0x80
+#define META_SCISSOR         0x100
+#define META_SHADER          0x200
+#define META_STENCIL_TEST    0x400
+#define META_TRANSFORM       0x800 /**< modelview, projection, clip planes */
+#define META_TEXTURE        0x1000
+#define META_VERTEX         0x2000
+#define META_VIEWPORT       0x4000
+/*@}*/
+
+
+/**
  * State which we may save/restore across meta ops.
  * XXX this may be incomplete...
  */
@@ -94,6 +117,17 @@ struct save_state
    /** META_PIXEL_STORE */
    struct gl_pixelstore_attrib Pack, Unpack;
 
+   /** META_PIXEL_TRANSFER */
+   GLfloat RedBias, RedScale;
+   GLfloat GreenBias, GreenScale;
+   GLfloat BlueBias, BlueScale;
+   GLfloat AlphaBias, AlphaScale;
+   GLfloat DepthBias, DepthScale;
+   GLboolean MapColorFlag;
+   GLboolean Convolution1DEnabled;
+   GLboolean Convolution2DEnabled;
+   GLboolean Separable2DEnabled;
+
    /** META_RASTERIZATION */
    GLenum FrontPolygonMode, BackPolygonMode;
    GLboolean PolygonOffset;
@@ -363,6 +397,35 @@ _mesa_meta_begin(GLcontext *ctx, GLbitfield state)
       ctx->Unpack = ctx->DefaultPacking;
    }
 
+   if (state & META_PIXEL_TRANSFER) {
+      save->RedScale = ctx->Pixel.RedScale;
+      save->RedBias = ctx->Pixel.RedBias;
+      save->GreenScale = ctx->Pixel.GreenScale;
+      save->GreenBias = ctx->Pixel.GreenBias;
+      save->BlueScale = ctx->Pixel.BlueScale;
+      save->BlueBias = ctx->Pixel.BlueBias;
+      save->AlphaScale = ctx->Pixel.AlphaScale;
+      save->AlphaBias = ctx->Pixel.AlphaBias;
+      save->MapColorFlag = ctx->Pixel.MapColorFlag;
+      save->Convolution1DEnabled = ctx->Pixel.Convolution1DEnabled;
+      save->Convolution2DEnabled = ctx->Pixel.Convolution2DEnabled;
+      save->Separable2DEnabled = ctx->Pixel.Separable2DEnabled;
+      ctx->Pixel.RedScale = 1.0F;
+      ctx->Pixel.RedBias = 0.0F;
+      ctx->Pixel.GreenScale = 1.0F;
+      ctx->Pixel.GreenBias = 0.0F;
+      ctx->Pixel.BlueScale = 1.0F;
+      ctx->Pixel.BlueBias = 0.0F;
+      ctx->Pixel.AlphaScale = 1.0F;
+      ctx->Pixel.AlphaBias = 0.0F;
+      ctx->Pixel.MapColorFlag = GL_FALSE;
+      ctx->Pixel.Convolution1DEnabled = GL_FALSE;
+      ctx->Pixel.Convolution2DEnabled = GL_FALSE;
+      ctx->Pixel.Separable2DEnabled = GL_FALSE;
+      /* XXX more state */
+      ctx->NewState |=_NEW_PIXEL;
+   }
+
    if (state & META_RASTERIZATION) {
       save->FrontPolygonMode = ctx->Polygon.FrontMode;
       save->BackPolygonMode = ctx->Polygon.BackMode;
@@ -415,11 +478,6 @@ _mesa_meta_begin(GLcontext *ctx, GLbitfield state)
       save->ClientActiveUnit = ctx->Array.ActiveTexture;
       save->EnvMode = ctx->Texture.Unit[0].EnvMode;
 
-      if (ctx->Texture._EnabledUnits |
-          ctx->Texture._EnabledCoordUnits |
-          ctx->Texture._TexGenEnabled |
-          ctx->Texture._TexMatEnabled) {
-
       /* Disable all texture units */
       for (u = 0; u < ctx->Const.MaxTextureUnits; u++) {
          save->TexEnabled[u] = ctx->Texture.Unit[u].Enabled;
@@ -438,7 +496,6 @@ _mesa_meta_begin(GLcontext *ctx, GLbitfield state)
             _mesa_set_enable(ctx, GL_TEXTURE_GEN_Q, GL_FALSE);
          }
       }
-      }
 
       /* save current texture objects for unit[0] only */
       for (tgt = 0; tgt < NUM_TEXTURE_TARGETS; tgt++) {
@@ -564,6 +621,23 @@ _mesa_meta_end(GLcontext *ctx)
       ctx->Unpack = save->Unpack;
    }
 
+   if (state & META_PIXEL_TRANSFER) {
+      ctx->Pixel.RedScale = save->RedScale;
+      ctx->Pixel.RedBias = save->RedBias;
+      ctx->Pixel.GreenScale = save->GreenScale;
+      ctx->Pixel.GreenBias = save->GreenBias;
+      ctx->Pixel.BlueScale = save->BlueScale;
+      ctx->Pixel.BlueBias = save->BlueBias;
+      ctx->Pixel.AlphaScale = save->AlphaScale;
+      ctx->Pixel.AlphaBias = save->AlphaBias;
+      ctx->Pixel.MapColorFlag = save->MapColorFlag;
+      ctx->Pixel.Convolution1DEnabled = save->Convolution1DEnabled;
+      ctx->Pixel.Convolution2DEnabled = save->Convolution2DEnabled;
+      ctx->Pixel.Separable2DEnabled = save->Separable2DEnabled;
+      /* XXX more state */
+      ctx->NewState |=_NEW_PIXEL;
+   }
+
    if (state & META_RASTERIZATION) {
       _mesa_PolygonMode(GL_FRONT, save->FrontPolygonMode);
       _mesa_PolygonMode(GL_BACK, save->BackPolygonMode);
@@ -2037,3 +2111,231 @@ _mesa_meta_generate_mipmap(GLcontext *ctx, GLenum target,
    /* restore (XXX add to meta_begin/end()? */
    _mesa_BindFramebufferEXT(GL_FRAMEBUFFER_EXT, fboSave);
 }
+
+
+/**
+ * Determine the GL data type to use for the temporary image read with
+ * ReadPixels() and passed to Tex[Sub]Image().
+ */
+static GLenum
+get_temp_image_type(GLcontext *ctx, GLenum baseFormat)
+{
+   switch (baseFormat) {
+   case GL_RGBA:
+   case GL_RGB:
+   case GL_ALPHA:
+   case GL_LUMINANCE:
+   case GL_LUMINANCE_ALPHA:
+   case GL_INTENSITY:
+      if (ctx->DrawBuffer->Visual.redBits <= 8)
+         return GL_UNSIGNED_BYTE;
+      else if (ctx->DrawBuffer->Visual.redBits <= 8)
+         return GL_UNSIGNED_SHORT;
+      else
+         return GL_FLOAT;
+   case GL_DEPTH_COMPONENT:
+      return GL_UNSIGNED_INT;
+   case GL_DEPTH_STENCIL:
+      return GL_UNSIGNED_INT_24_8;
+   default:
+      _mesa_problem(ctx, "Unexpected format in get_temp_image_type()");
+      return 0;
+   }
+}
+
+
+/**
+ * Helper for _mesa_meta_CopyTexImage1/2D() functions.
+ * Have to be careful with locking and meta state for pixel transfer.
+ */
+static void
+copy_tex_image(GLcontext *ctx, GLuint dims, GLenum target, GLint level,
+               GLenum internalFormat, GLint x, GLint y,
+               GLsizei width, GLsizei height, GLint border)
+{
+   struct gl_texture_unit *texUnit;
+   struct gl_texture_object *texObj;
+   struct gl_texture_image *texImage;
+   GLenum format, type;
+   GLint bpp;
+   void *buf;
+
+   texUnit = _mesa_get_current_tex_unit(ctx);
+   texObj = _mesa_select_tex_object(ctx, texUnit, target);
+   texImage = _mesa_get_tex_image(ctx, texObj, target, level);
+
+   format = _mesa_base_tex_format(ctx, internalFormat);
+   type = get_temp_image_type(ctx, format);
+   bpp = _mesa_bytes_per_pixel(format, type);
+   if (bpp <= 0) {
+      _mesa_problem(ctx, "Bad bpp in meta copy_tex_image()");
+      return;
+   }
+
+   /*
+    * Alloc image buffer (XXX could use a PBO)
+    */
+   buf = _mesa_malloc(width * height * bpp);
+   if (!buf) {
+      _mesa_error(ctx, GL_OUT_OF_MEMORY, "glCopyTexImage%uD", dims);
+      return;
+   }
+
+   _mesa_unlock_texture(ctx, texObj); /* need to unlock first */
+
+   /*
+    * Read image from framebuffer (disable pixel transfer ops)
+    */
+   _mesa_meta_begin(ctx, META_PIXEL_STORE | META_PIXEL_TRANSFER);
+   ctx->Driver.ReadPixels(ctx, x, y, width, height,
+			  format, type, &ctx->Pack, buf);
+   _mesa_meta_end(ctx);
+
+   /*
+    * Store texture data (with pixel transfer ops)
+    */
+   _mesa_meta_begin(ctx, META_PIXEL_STORE);
+   if (target == GL_TEXTURE_1D) {
+      ctx->Driver.TexImage1D(ctx, target, level, internalFormat,
+                             width, border, format, type,
+                             buf, &ctx->Unpack, texObj, texImage);
+   }
+   else {
+      ctx->Driver.TexImage2D(ctx, target, level, internalFormat,
+                             width, height, border, format, type,
+                             buf, &ctx->Unpack, texObj, texImage);
+   }
+   _mesa_meta_end(ctx);
+
+   _mesa_lock_texture(ctx, texObj); /* re-lock */
+
+   _mesa_free(buf);
+}
+
+
+void
+_mesa_meta_CopyTexImage1D(GLcontext *ctx, GLenum target, GLint level,
+                          GLenum internalFormat, GLint x, GLint y,
+                          GLsizei width, GLint border)
+{
+   copy_tex_image(ctx, 1, target, level, internalFormat, x, y,
+                  width, 1, border);
+}
+
+
+void
+_mesa_meta_CopyTexImage2D(GLcontext *ctx, GLenum target, GLint level,
+                          GLenum internalFormat, GLint x, GLint y,
+                          GLsizei width, GLsizei height, GLint border)
+{
+   copy_tex_image(ctx, 2, target, level, internalFormat, x, y,
+                  width, height, border);
+}
+
+
+
+/**
+ * Helper for _mesa_meta_CopyTexSubImage1/2/3D() functions.
+ * Have to be careful with locking and meta state for pixel transfer.
+ */
+static void
+copy_tex_sub_image(GLcontext *ctx, GLuint dims, GLenum target, GLint level,
+                   GLint xoffset, GLint yoffset, GLint zoffset,
+                   GLint x, GLint y,
+                   GLsizei width, GLsizei height)
+{
+   struct gl_texture_unit *texUnit;
+   struct gl_texture_object *texObj;
+   struct gl_texture_image *texImage;
+   GLenum format, type;
+   GLint bpp;
+   void *buf;
+
+   texUnit = _mesa_get_current_tex_unit(ctx);
+   texObj = _mesa_select_tex_object(ctx, texUnit, target);
+   texImage = _mesa_select_tex_image(ctx, texObj, target, level);
+
+   format = texImage->TexFormat->BaseFormat;
+   type = get_temp_image_type(ctx, format);
+   bpp = _mesa_bytes_per_pixel(format, type);
+   if (bpp <= 0) {
+      _mesa_problem(ctx, "Bad bpp in meta copy_tex_sub_image()");
+      return;
+   }
+
+   /*
+    * Alloc image buffer (XXX could use a PBO)
+    */
+   buf = _mesa_malloc(width * height * bpp);
+   if (!buf) {
+      _mesa_error(ctx, GL_OUT_OF_MEMORY, "glCopyTexSubImage%uD", dims);
+      return;
+   }
+
+   _mesa_unlock_texture(ctx, texObj); /* need to unlock first */
+
+   /*
+    * Read image from framebuffer (disable pixel transfer ops)
+    */
+   _mesa_meta_begin(ctx, META_PIXEL_STORE | META_PIXEL_TRANSFER);
+   ctx->Driver.ReadPixels(ctx, x, y, width, height,
+			  format, type, &ctx->Pack, buf);
+   _mesa_meta_end(ctx);
+
+   /*
+    * Store texture data (with pixel transfer ops)
+    */
+   _mesa_meta_begin(ctx, META_PIXEL_STORE);
+   if (target == GL_TEXTURE_1D) {
+      ctx->Driver.TexSubImage1D(ctx, target, level, xoffset,
+                                width, format, type, buf,
+                                &ctx->Unpack, texObj, texImage);
+   }
+   else if (target == GL_TEXTURE_3D) {
+      ctx->Driver.TexSubImage3D(ctx, target, level, xoffset, yoffset, zoffset,
+                                width, height, 1, format, type, buf,
+                                &ctx->Unpack, texObj, texImage);
+   }
+   else {
+      ctx->Driver.TexSubImage2D(ctx, target, level, xoffset, yoffset,
+                                width, height, format, type, buf,
+                                &ctx->Unpack, texObj, texImage);
+   }
+   _mesa_meta_end(ctx);
+
+   _mesa_lock_texture(ctx, texObj); /* re-lock */
+
+   _mesa_free(buf);
+}
+
+
+void
+_mesa_meta_CopyTexSubImage1D(GLcontext *ctx, GLenum target, GLint level,
+                             GLint xoffset,
+                             GLint x, GLint y, GLsizei width)
+{
+   copy_tex_sub_image(ctx, 1, target, level, xoffset, 0, 0,
+                      x, y, width, 1);
+}
+
+
+void
+_mesa_meta_CopyTexSubImage2D(GLcontext *ctx, GLenum target, GLint level,
+                             GLint xoffset, GLint yoffset,
+                             GLint x, GLint y,
+                             GLsizei width, GLsizei height)
+{
+   copy_tex_sub_image(ctx, 2, target, level, xoffset, yoffset, 0,
+                      x, y, width, height);
+}
+
+
+void
+_mesa_meta_CopyTexSubImage3D(GLcontext *ctx, GLenum target, GLint level,
+                             GLint xoffset, GLint yoffset, GLint zoffset,
+                             GLint x, GLint y,
+                             GLsizei width, GLsizei height)
+{
+   copy_tex_sub_image(ctx, 3, target, level, xoffset, yoffset, zoffset,
+                      x, y, width, height);
+}
diff --git a/src/mesa/drivers/common/meta.h b/src/mesa/drivers/common/meta.h
index 171ad27f26..9f7100f09c 100644
--- a/src/mesa/drivers/common/meta.h
+++ b/src/mesa/drivers/common/meta.h
@@ -27,29 +27,6 @@
 #define META_H
 
 
-/**
- * Flags passed to _mesa_meta_begin().
- * XXX these flags may evolve...
- */
-/*@{*/
-#define META_ALPHA_TEST      0x1
-#define META_BLEND           0x2  /**< includes logicop */
-#define META_COLOR_MASK      0x4
-#define META_DEPTH_TEST      0x8
-#define META_FOG            0x10
-#define META_RASTERIZATION  0x20
-#define META_SCISSOR        0x40
-#define META_SHADER         0x80
-#define META_STENCIL_TEST  0x100
-#define META_TRANSFORM     0x200 /**< modelview, projection */
-#define META_TEXTURE       0x400
-#define META_VERTEX        0x800
-#define META_VIEWPORT     0x1000
-#define META_PIXEL_STORE  0x2000
-#define META_ALL            ~0x0
-/*@}*/
-
-
 extern void
 _mesa_meta_init(GLcontext *ctx);
 
@@ -87,5 +64,32 @@ extern void
 _mesa_meta_generate_mipmap(GLcontext *ctx, GLenum target,
                            struct gl_texture_object *texObj);
 
+extern void
+_mesa_meta_CopyTexImage1D(GLcontext *ctx, GLenum target, GLint level,
+                          GLenum internalFormat, GLint x, GLint y,
+                          GLsizei width, GLint border);
+
+extern void
+_mesa_meta_CopyTexImage2D(GLcontext *ctx, GLenum target, GLint level,
+                          GLenum internalFormat, GLint x, GLint y,
+                          GLsizei width, GLsizei height, GLint border);
+
+extern void
+_mesa_meta_CopyTexSubImage1D(GLcontext *ctx, GLenum target, GLint level,
+                             GLint xoffset,
+                             GLint x, GLint y, GLsizei width);
+
+extern void
+_mesa_meta_CopyTexSubImage2D(GLcontext *ctx, GLenum target, GLint level,
+                             GLint xoffset, GLint yoffset,
+                             GLint x, GLint y,
+                             GLsizei width, GLsizei height);
+
+extern void
+_mesa_meta_CopyTexSubImage3D(GLcontext *ctx, GLenum target, GLint level,
+                             GLint xoffset, GLint yoffset, GLint zoffset,
+                             GLint x, GLint y,
+                             GLsizei width, GLsizei height);
+
 
 #endif /* META_H */
diff --git a/src/mesa/drivers/dri/ffb/ffb_tex.c b/src/mesa/drivers/dri/ffb/ffb_tex.c
index 69d30aedba..95058e9069 100644
--- a/src/mesa/drivers/dri/ffb/ffb_tex.c
+++ b/src/mesa/drivers/dri/ffb/ffb_tex.c
@@ -30,24 +30,6 @@
 #include "ffb_tex.h"
 
 /* No texture unit, all software. */
-/* XXX this function isn't needed since _mesa_init_driver_functions()
- * will make all these assignments.
- */
 void ffbDDInitTexFuncs(GLcontext *ctx)
 {
-        /*
-	ctx->Driver.ChooseTextureFormat = _mesa_choose_tex_format;
-	ctx->Driver.TexImage1D = _mesa_store_teximage1d;
-	ctx->Driver.TexImage2D = _mesa_store_teximage2d;
-	ctx->Driver.TexImage3D = _mesa_store_teximage3d;
-	ctx->Driver.TexSubImage1D = _mesa_store_texsubimage1d;
-	ctx->Driver.TexSubImage2D = _mesa_store_texsubimage2d;
-	ctx->Driver.TexSubImage3D = _mesa_store_texsubimage3d;
-	ctx->Driver.CopyTexImage1D = _swrast_copy_teximage1d;
-	ctx->Driver.CopyTexImage2D = _swrast_copy_teximage2d;
-	ctx->Driver.CopyTexSubImage1D = _swrast_copy_texsubimage1d;
-	ctx->Driver.CopyTexSubImage2D = _swrast_copy_texsubimage2d;
-	ctx->Driver.CopyTexSubImage3D = _swrast_copy_texsubimage3d;
-	ctx->Driver.TestProxyTexImage = _mesa_test_proxy_teximage;
-	*/
 }
diff --git a/src/mesa/drivers/dri/intel/intel_clear.c b/src/mesa/drivers/dri/intel/intel_clear.c
index bce23724b3..9efe6a277c 100644
--- a/src/mesa/drivers/dri/intel/intel_clear.c
+++ b/src/mesa/drivers/dri/intel/intel_clear.c
@@ -170,6 +170,7 @@ intelClear(GLcontext *ctx, GLbitfield mask)
 	 }
 	 DBG("\n");
       }
+      intelFlush(&intel->ctx);
       _mesa_meta_clear(&intel->ctx, tri_mask);
    }
 
diff --git a/src/mesa/drivers/dri/intel/intel_context.c b/src/mesa/drivers/dri/intel/intel_context.c
index 7e21b94acc..03db8b1c68 100644
--- a/src/mesa/drivers/dri/intel/intel_context.c
+++ b/src/mesa/drivers/dri/intel/intel_context.c
@@ -587,11 +587,6 @@ intelInitDriverFunctions(struct dd_function_table *functions)
    functions->GetString = intelGetString;
    functions->UpdateState = intelInvalidateState;
 
-   functions->CopyColorTable = _swrast_CopyColorTable;
-   functions->CopyColorSubTable = _swrast_CopyColorSubTable;
-   functions->CopyConvolutionFilter1D = _swrast_CopyConvolutionFilter1D;
-   functions->CopyConvolutionFilter2D = _swrast_CopyConvolutionFilter2D;
-
    intelInitTextureFuncs(functions);
    intelInitTextureImageFuncs(functions);
    intelInitTextureSubImageFuncs(functions);
@@ -922,6 +917,14 @@ intelDestroyContext(__DRIcontextPrivate * driContextPriv)
 GLboolean
 intelUnbindContext(__DRIcontextPrivate * driContextPriv)
 {
+   struct intel_context *intel =
+      (struct intel_context *) driContextPriv->driverPrivate;
+
+   /* Deassociate the context with the drawables.
+    */
+   intel->driDrawable = NULL;
+   intel->driReadDrawable = NULL;
+
    return GL_TRUE;
 }
 
diff --git a/src/mesa/drivers/dri/intel/intel_generatemipmap.c b/src/mesa/drivers/dri/intel/intel_generatemipmap.c
index fe986092db..237754d469 100644
--- a/src/mesa/drivers/dri/intel/intel_generatemipmap.c
+++ b/src/mesa/drivers/dri/intel/intel_generatemipmap.c
@@ -223,10 +223,6 @@ fail:
  *
  * The texture object's miptree must be mapped.
  *
- * It would be really nice if this was just called by Mesa whenever mipmaps
- * needed to be regenerated, rather than us having to remember to do so in
- * each texture image modification path.
- *
  * This function should also include an accelerated path.
  */
 void
diff --git a/src/mesa/drivers/dri/intel/intel_span.c b/src/mesa/drivers/dri/intel/intel_span.c
index 1e73943457..28eabbc005 100644
--- a/src/mesa/drivers/dri/intel/intel_span.c
+++ b/src/mesa/drivers/dri/intel/intel_span.c
@@ -448,24 +448,22 @@ intel_map_unmap_framebuffer(struct intel_context *intel,
 			    struct gl_framebuffer *fb,
 			    GLboolean map)
 {
-   GLcontext *ctx = &intel->ctx;
-   GLuint i, j;
-
-   /* color buffers */
-   if (fb == ctx->DrawBuffer) {
-      for (j = 0; j < fb->_NumColorDrawBuffers; j++) {
-	 if (map)
-	    intel_renderbuffer_map(intel, fb->_ColorDrawBuffers[j]);
-	 else
-	    intel_renderbuffer_unmap(intel, fb->_ColorDrawBuffers[j]);
-      }
-   } else {
+   GLuint i;
+
+   /* color draw buffers */
+   for (i = 0; i < fb->_NumColorDrawBuffers; i++) {
       if (map)
-	 intel_renderbuffer_map(intel, fb->_ColorReadBuffer);
+         intel_renderbuffer_map(intel, fb->_ColorDrawBuffers[i]);
       else
-	 intel_renderbuffer_unmap(intel, fb->_ColorReadBuffer);
+         intel_renderbuffer_unmap(intel, fb->_ColorDrawBuffers[i]);
    }
 
+   /* color read buffer */
+   if (map)
+      intel_renderbuffer_map(intel, fb->_ColorReadBuffer);
+   else
+      intel_renderbuffer_unmap(intel, fb->_ColorReadBuffer);
+
    /* check for render to textures */
    for (i = 0; i < BUFFER_COUNT; i++) {
       struct gl_renderbuffer_attachment *att =
@@ -484,20 +482,17 @@ intel_map_unmap_framebuffer(struct intel_context *intel,
    /* depth buffer (Note wrapper!) */
    if (fb->_DepthBuffer) {
       if (map)
-	 intel_renderbuffer_map(intel, fb->_DepthBuffer->Wrapped);
+         intel_renderbuffer_map(intel, fb->_DepthBuffer->Wrapped);
       else
-	 intel_renderbuffer_unmap(intel,
-				  fb->_DepthBuffer->Wrapped);
+         intel_renderbuffer_unmap(intel, fb->_DepthBuffer->Wrapped);
    }
 
    /* stencil buffer (Note wrapper!) */
    if (fb->_StencilBuffer) {
       if (map)
-	 intel_renderbuffer_map(intel,
-				fb->_StencilBuffer->Wrapped);
+         intel_renderbuffer_map(intel, fb->_StencilBuffer->Wrapped);
       else
-	 intel_renderbuffer_unmap(intel,
-				  fb->_StencilBuffer->Wrapped);
+         intel_renderbuffer_unmap(intel, fb->_StencilBuffer->Wrapped);
    }
 }
 
@@ -524,7 +519,8 @@ intelSpanRenderStart(GLcontext * ctx)
    }
 
    intel_map_unmap_framebuffer(intel, ctx->DrawBuffer, GL_TRUE);
-   intel_map_unmap_framebuffer(intel, ctx->ReadBuffer, GL_TRUE);
+   if (ctx->ReadBuffer != ctx->DrawBuffer)
+      intel_map_unmap_framebuffer(intel, ctx->ReadBuffer, GL_TRUE);
 }
 
 /**
@@ -547,7 +543,8 @@ intelSpanRenderFinish(GLcontext * ctx)
    }
 
    intel_map_unmap_framebuffer(intel, ctx->DrawBuffer, GL_FALSE);
-   intel_map_unmap_framebuffer(intel, ctx->ReadBuffer, GL_FALSE);
+   if (ctx->ReadBuffer != ctx->DrawBuffer)
+      intel_map_unmap_framebuffer(intel, ctx->ReadBuffer, GL_FALSE);
 
    UNLOCK_HARDWARE(intel);
 }
diff --git a/src/mesa/drivers/dri/intel/intel_tex_copy.c b/src/mesa/drivers/dri/intel/intel_tex_copy.c
index 028b49c14d..148e8c048c 100644
--- a/src/mesa/drivers/dri/intel/intel_tex_copy.c
+++ b/src/mesa/drivers/dri/intel/intel_tex_copy.c
@@ -90,7 +90,6 @@ do_copy_texsubimage(struct intel_context *intel,
                     GLint x, GLint y, GLsizei width, GLsizei height)
 {
    GLcontext *ctx = &intel->ctx;
-   struct gl_texture_object *texObj = intelImage->base.TexObject;
    const struct intel_region *src =
       get_teximage_source(intel, internalFormat);
 
@@ -170,11 +169,6 @@ do_copy_texsubimage(struct intel_context *intel,
 
    UNLOCK_HARDWARE(intel);
 
-   /* GL_SGIS_generate_mipmap */
-   if (intelImage->level == texObj->BaseLevel && texObj->GenerateMipmap) {
-      intel_generate_mipmap(ctx, target, texObj);
-   }
-
    return GL_TRUE;
 }
 
diff --git a/src/mesa/drivers/dri/intel/intel_tex_image.c b/src/mesa/drivers/dri/intel/intel_tex_image.c
index a206fe6805..66201b1f46 100644
--- a/src/mesa/drivers/dri/intel/intel_tex_image.c
+++ b/src/mesa/drivers/dri/intel/intel_tex_image.c
@@ -546,11 +546,6 @@ intelTexImage(GLcontext * ctx,
    }
 
    UNLOCK_HARDWARE(intel);
-
-   /* GL_SGIS_generate_mipmap */
-   if (level == texObj->BaseLevel && texObj->GenerateMipmap) {
-      intel_generate_mipmap(ctx, target, texObj);
-   }
 }
 
 
diff --git a/src/mesa/drivers/dri/intel/intel_tex_subimage.c b/src/mesa/drivers/dri/intel/intel_tex_subimage.c
index 89037073f8..751ec2c98c 100644
--- a/src/mesa/drivers/dri/intel/intel_tex_subimage.c
+++ b/src/mesa/drivers/dri/intel/intel_tex_subimage.c
@@ -129,11 +129,6 @@ intelTexSubimage(GLcontext * ctx,
    }
 
    UNLOCK_HARDWARE(intel);
-
-   /* GL_SGIS_generate_mipmap */
-   if (level == texObj->BaseLevel && texObj->GenerateMipmap) {
-      intel_generate_mipmap(ctx, target, texObj);
-   }
 }
 
 
diff --git a/src/mesa/drivers/dri/r200/Makefile b/src/mesa/drivers/dri/r200/Makefile
index 42635bf9d9..fbce70c37b 100644
--- a/src/mesa/drivers/dri/r200/Makefile
+++ b/src/mesa/drivers/dri/r200/Makefile
@@ -55,8 +55,7 @@ C_SOURCES = $(COMMON_SOURCES) $(DRIVER_SOURCES)
 
 X86_SOURCES = 
 
-DRIVER_DEFINES = -DRADEON_COMMON=1 -DRADEON_COMMON_FOR_R200 \
-				 -Wall
+DRIVER_DEFINES = -DRADEON_R200 -Wall
 
 DRI_LIB_DEPS += $(RADEON_LDFLAGS)
 
diff --git a/src/mesa/drivers/dri/r300/Makefile b/src/mesa/drivers/dri/r300/Makefile
index 5d8d6f6658..fe775eac99 100644
--- a/src/mesa/drivers/dri/r300/Makefile
+++ b/src/mesa/drivers/dri/r300/Makefile
@@ -62,8 +62,7 @@ DRIVER_SOURCES = \
 
 C_SOURCES = $(COMMON_SOURCES) $(DRIVER_SOURCES)
 
-DRIVER_DEFINES = -DCOMPILE_R300 -DR200_MERGED=0 \
-	-DRADEON_COMMON=1 -DRADEON_COMMON_FOR_R300 \
+DRIVER_DEFINES = -DRADEON_R300
 #	-DRADEON_BO_TRACK \
 	-Wall
 
diff --git a/src/mesa/drivers/dri/r300/radeon_context.h b/src/mesa/drivers/dri/r300/radeon_context.h
index 250570f6b8..da4812d323 100644
--- a/src/mesa/drivers/dri/r300/radeon_context.h
+++ b/src/mesa/drivers/dri/r300/radeon_context.h
@@ -51,26 +51,12 @@ WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
 
 #include "radeon_screen.h"
 
-#if R200_MERGED
-extern void radeonFallback(GLcontext * ctx, GLuint bit, GLboolean mode);
-
-#define FALLBACK( radeon, bit, mode ) do {			\
-   if ( 0 ) fprintf( stderr, "FALLBACK in %s: #%d=%d\n",	\
-		     __FUNCTION__, bit, mode );			\
-   radeonFallback( (radeon)->glCtx, bit, mode );		\
-} while (0)
-#else
 #define FALLBACK( radeon, bit, mode ) fprintf(stderr, "%s:%s\n", __LINE__, __FILE__);
-#endif
 
 /* TCL fallbacks */
 extern void radeonTclFallback(GLcontext * ctx, GLuint bit, GLboolean mode);
 
-#if R200_MERGED
-#define TCL_FALLBACK( ctx, bit, mode )	radeonTclFallback( ctx, bit, mode )
-#else
 #define TCL_FALLBACK( ctx, bit, mode )	;
-#endif
 
 
 #endif				/* __RADEON_CONTEXT_H__ */
diff --git a/src/mesa/drivers/dri/r600/Makefile b/src/mesa/drivers/dri/r600/Makefile
index d925a2dfe3..36bf773c05 100644
--- a/src/mesa/drivers/dri/r600/Makefile
+++ b/src/mesa/drivers/dri/r600/Makefile
@@ -65,8 +65,7 @@ DRIVER_SOURCES = \
 
 C_SOURCES = $(COMMON_SOURCES) $(DRIVER_SOURCES)
 
-DRIVER_DEFINES = -DCOMPILE_R600 -DR200_MERGED=0 \
-	-DRADEON_COMMON=1 -DRADEON_COMMON_FOR_R600 \
+DRIVER_DEFINES = -DRADEON_R600 \
 #	-DRADEON_BO_TRACK \
 	-Wall
 
diff --git a/src/mesa/drivers/dri/r600/r600_context.c b/src/mesa/drivers/dri/r600/r600_context.c
index 07a7bcf11f..f8fd9c13d7 100644
--- a/src/mesa/drivers/dri/r600/r600_context.c
+++ b/src/mesa/drivers/dri/r600/r600_context.c
@@ -385,9 +385,6 @@ GLboolean r600CreateContext(const __GLcontextModes * glVisual,
 
 	if (driQueryOptionb(&r600->radeon.optionCache, "no_rast")) {
 		radeon_warning("disabling 3D acceleration\n");
-#if R200_MERGED
-		FALLBACK(&r600->radeon, RADEON_FALLBACK_DISABLE, 1);
-#endif
 	}
 
 	return GL_TRUE;
diff --git a/src/mesa/drivers/dri/r600/r600_reg_auto_r6xx.h b/src/mesa/drivers/dri/r600/r600_reg_auto_r6xx.h
index 9d5aa3c7e4..edd85b0fac 100644
--- a/src/mesa/drivers/dri/r600/r600_reg_auto_r6xx.h
+++ b/src/mesa/drivers/dri/r600/r600_reg_auto_r6xx.h
@@ -1366,6 +1366,7 @@ enum {
 	DB_DEPTH_INFO__READ_SIZE_bit                      = 1 << 3,
 	DB_DEPTH_INFO__ARRAY_MODE_mask                    = 0x0f << 15,
 	DB_DEPTH_INFO__ARRAY_MODE_shift                   = 15,
+	    ARRAY_1D_TILED_THIN1                          = 0x02,
 	    ARRAY_2D_TILED_THIN1                          = 0x04,
 	TILE_SURFACE_ENABLE_bit                           = 1 << 25,
 	TILE_COMPACT_bit                                  = 1 << 26,
@@ -1449,6 +1450,7 @@ enum {
 	CB_COLOR0_INFO__ARRAY_MODE_shift                  = 8,
 	    ARRAY_LINEAR_GENERAL                          = 0x00,
 	    ARRAY_LINEAR_ALIGNED                          = 0x01,
+/*	    ARRAY_1D_TILED_THIN1                          = 0x02, */
 /* 	    ARRAY_2D_TILED_THIN1                          = 0x04, */
 	NUMBER_TYPE_mask                                  = 0x07 << 12,
 	NUMBER_TYPE_shift                                 = 12,
diff --git a/src/mesa/drivers/dri/r600/r700_assembler.c b/src/mesa/drivers/dri/r600/r700_assembler.c
index fda6692725..efeccb25f1 100644
--- a/src/mesa/drivers/dri/r600/r700_assembler.c
+++ b/src/mesa/drivers/dri/r600/r700_assembler.c
@@ -3387,6 +3387,7 @@ GLboolean assemble_TEX(r700_AssemblerBase *pAsm)
         break;
     case PROGRAM_TEMPORARY:
     case PROGRAM_INPUT:
+    default:
         src_const = GL_FALSE;
 	break;
     }
diff --git a/src/mesa/drivers/dri/r600/r700_chip.c b/src/mesa/drivers/dri/r600/r700_chip.c
index 1b56059197..06d7e9c9ab 100644
--- a/src/mesa/drivers/dri/r600/r700_chip.c
+++ b/src/mesa/drivers/dri/r600/r700_chip.c
@@ -351,7 +351,7 @@ static void r700SetDepthTarget(context_t *context)
         SETfield(r700->DB_DEPTH_INFO.u32All, DEPTH_16,
                      DB_DEPTH_INFO__FORMAT_shift, DB_DEPTH_INFO__FORMAT_mask);
     }
-    SETfield(r700->DB_DEPTH_INFO.u32All, ARRAY_2D_TILED_THIN1,
+    SETfield(r700->DB_DEPTH_INFO.u32All, ARRAY_1D_TILED_THIN1,
              DB_DEPTH_INFO__ARRAY_MODE_shift, DB_DEPTH_INFO__ARRAY_MODE_mask);
     /* r700->DB_PREFETCH_LIMIT.bits.DEPTH_HEIGHT_TILE_MAX = (context->currentDraw->h >> 3) - 1; */ /* z buffer sie may much bigger than what need, so use actual used h. */
 }
diff --git a/src/mesa/drivers/dri/r600/r700_vertprog.c b/src/mesa/drivers/dri/r600/r700_vertprog.c
index 8c2b0071df..9ee26286d9 100644
--- a/src/mesa/drivers/dri/r600/r700_vertprog.c
+++ b/src/mesa/drivers/dri/r600/r700_vertprog.c
@@ -42,6 +42,7 @@
 #include "radeon_debug.h"
 #include "r600_context.h"
 #include "r600_cmdbuf.h"
+#include "shader/programopt.c"
 
 #include "r700_debug.h"
 #include "r700_vertprog.h"
@@ -272,6 +273,11 @@ struct r700_vertex_program* r700TranslateVertexShader(GLcontext *ctx,
 	vp = _mesa_calloc(sizeof(*vp));
 	vp->mesa_program = (struct gl_vertex_program *)_mesa_clone_program(ctx, &mesa_vp->Base);
 
+	if (mesa_vp->IsPositionInvariant)
+	{
+                _mesa_insert_mvp_code(ctx, vp->mesa_program);
+        }
+
 	for(i=0; i<VERT_ATTRIB_MAX; i++)
 	{
 		unBit = 1 << i;
@@ -290,21 +296,21 @@ struct r700_vertex_program* r700TranslateVertexShader(GLcontext *ctx,
 
 	//Init_Program
 	Init_r700_AssemblerBase(SPT_VP, &(vp->r700AsmCode), &(vp->r700Shader) );
-	Map_Vertex_Program( vp, mesa_vp );
+	Map_Vertex_Program( vp, vp->mesa_program );
 
-	if(GL_FALSE == Find_Instruction_Dependencies_vp(vp, mesa_vp))
+	if(GL_FALSE == Find_Instruction_Dependencies_vp(vp, vp->mesa_program))
 	{
 		return NULL;
     }
 
-	if(GL_FALSE == AssembleInstr(mesa_vp->Base.NumInstructions,
-                                 &(mesa_vp->Base.Instructions[0]), 
+	if(GL_FALSE == AssembleInstr(vp->mesa_program->Base.NumInstructions,
+                                 &(vp->mesa_program->Base.Instructions[0]), 
                                  &(vp->r700AsmCode)) )
 	{
 		return NULL;
 	} 
 
-    if(GL_FALSE == Process_Vertex_Exports(&(vp->r700AsmCode), mesa_vp->Base.OutputsWritten) )
+    if(GL_FALSE == Process_Vertex_Exports(&(vp->r700AsmCode), vp->mesa_program->Base.OutputsWritten) )
     {
         return NULL;
     }
@@ -329,23 +335,23 @@ void r700SelectVertexShader(GLcontext *ctx)
     unsigned int unBit;
     unsigned int i;
     GLboolean match;
+    GLbitfield InputsRead;
 
     vpc = (struct r700_vertex_program_cont *)ctx->VertexProgram._Current;
 
-#if 0
-    if (context->radeon.NewGLState & (_NEW_PROGRAM_CONSTANTS|_NEW_PROGRAM))
+    InputsRead = vpc->mesa_program.Base.InputsRead;
+    if (vpc->mesa_program.IsPositionInvariant)
     {
-	vpc->needUpdateVF = 1;
-    }
-#endif
-
+	InputsRead |= VERT_BIT_POS;
+    } 
+    
     for (vp = vpc->progs; vp; vp = vp->next)
     {
 	match = GL_TRUE;	
 	for(i=0; i<VERT_ATTRIB_MAX; i++)
 	{
 		unBit = 1 << i;
-                if(vpc->mesa_program.Base.InputsRead & unBit)
+		if(InputsRead & unBit)
 		{
 			if (vp->aos_desc[i].size != vb->AttribPtr[i]->size)
 				match = GL_FALSE;
diff --git a/src/mesa/drivers/dri/radeon/Makefile b/src/mesa/drivers/dri/radeon/Makefile
index 1f286776b5..b1efc72872 100644
--- a/src/mesa/drivers/dri/radeon/Makefile
+++ b/src/mesa/drivers/dri/radeon/Makefile
@@ -47,8 +47,7 @@ C_SOURCES = \
 	$(DRIVER_SOURCES) \
 	$(CS_SOURCES)
 
-DRIVER_DEFINES = -DRADEON_COMMON=0 \
-				 -Wall
+DRIVER_DEFINES = -DRADEON_R100 -Wall
 
 DRI_LIB_DEPS += $(RADEON_LDFLAGS)
 
diff --git a/src/mesa/drivers/dri/radeon/radeon_common_context.c b/src/mesa/drivers/dri/radeon/radeon_common_context.c
index 71ee06d9a7..1c53c04da7 100644
--- a/src/mesa/drivers/dri/radeon/radeon_common_context.c
+++ b/src/mesa/drivers/dri/radeon/radeon_common_context.c
@@ -47,7 +47,7 @@ WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
 #include "swrast_setup/swrast_setup.h"
 #include "tnl/tnl.h"
 
-#if RADEON_COMMON && defined(RADEON_COMMON_FOR_R600) /* +r6/r7 */
+#if defined(RADEON_R600)
 #include "r600_context.h"
 #endif
 
diff --git a/src/mesa/drivers/dri/radeon/radeon_dma.c b/src/mesa/drivers/dri/radeon/radeon_dma.c
index 2eefa3f2b1..c9a32c808b 100644
--- a/src/mesa/drivers/dri/radeon/radeon_dma.c
+++ b/src/mesa/drivers/dri/radeon/radeon_dma.c
@@ -263,7 +263,7 @@ void radeonAllocDmaRegion(radeonContextPtr rmesa,
 
 void radeonFreeDmaRegions(radeonContextPtr rmesa)
 {
-	struct radeon_dma_bo *dma_bo = CALLOC_STRUCT(radeon_dma_bo);
+	struct radeon_dma_bo *dma_bo;
 	struct radeon_dma_bo *temp;
 	if (RADEON_DEBUG & RADEON_DMA)
 		fprintf(stderr, "%s\n", __FUNCTION__);
diff --git a/src/mesa/drivers/dri/radeon/radeon_screen.c b/src/mesa/drivers/dri/radeon/radeon_screen.c
index 5ffb55db5e..573eb6c9c1 100644
--- a/src/mesa/drivers/dri/radeon/radeon_screen.c
+++ b/src/mesa/drivers/dri/radeon/radeon_screen.c
@@ -48,17 +48,17 @@ WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
 #include "radeon_screen.h"
 #include "radeon_common.h"
 #include "radeon_span.h"
-#if !RADEON_COMMON
+#if defined(RADEON_R100)
 #include "radeon_context.h"
 #include "radeon_tex.h"
-#elif RADEON_COMMON && defined(RADEON_COMMON_FOR_R200)
+#elif defined(RADEON_R200)
 #include "r200_context.h"
 #include "r200_ioctl.h"
 #include "r200_tex.h"
-#elif RADEON_COMMON && defined(RADEON_COMMON_FOR_R300)
+#elif defined(RADEON_R300)
 #include "r300_context.h"
 #include "r300_tex.h"
-#elif RADEON_COMMON && defined(RADEON_COMMON_FOR_R600)
+#elif defined(RADEON_R600)
 #include "r600_context.h"
 #include "r700_driconf.h" /* +r6/r7 */
 #include "r600_tex.h"     /* +r6/r7 */
@@ -82,7 +82,7 @@ DRI_CONF_OPT_BEGIN_V(command_buffer_size,int,def, # min ":" # max ) \
         DRI_CONF_DESC(de,"Grösse des Befehlspuffers (in KB)") \
 DRI_CONF_OPT_END
 
-#if !RADEON_COMMON	/* R100 */
+#if defined(RADEON_R100)	/* R100 */
 PUBLIC const char __driConfigOptions[] =
 DRI_CONF_BEGIN
     DRI_CONF_SECTION_PERFORMANCE
@@ -109,7 +109,7 @@ DRI_CONF_BEGIN
 DRI_CONF_END;
 static const GLuint __driNConfigOptions = 15;
 
-#elif RADEON_COMMON && defined(RADEON_COMMON_FOR_R200)
+#elif defined(RADEON_R200)
 
 PUBLIC const char __driConfigOptions[] =
 DRI_CONF_BEGIN
@@ -147,7 +147,7 @@ extern const struct dri_extension NV_vp_extension[];
 extern const struct dri_extension ATI_fs_extension[];
 extern const struct dri_extension point_extensions[];
 
-#elif RADEON_COMMON && (defined(RADEON_COMMON_FOR_R300) || defined(RADEON_COMMON_FOR_R600))
+#elif defined(RADEON_R300) || defined(RADEON_R600)
 
 #define DRI_CONF_FP_OPTIMIZATION_SPEED   0
 #define DRI_CONF_FP_OPTIMIZATION_QUALITY 1
@@ -220,7 +220,7 @@ static const GLuint __driNConfigOptions = 17;
 
 extern const struct dri_extension gl_20_extension[];
 
-#endif /* RADEON_COMMON && defined(RADEON_COMMON_FOR_R300) */
+#endif
 
 extern const struct dri_extension card_extensions[];
 extern const struct dri_extension mm_extensions[];
@@ -337,7 +337,7 @@ radeonFillInModes( __DRIscreenPrivate *psp,
     return (const __DRIconfig **) configs;
 }
 
-#if !RADEON_COMMON
+#if defined(RADEON_R100)
 static const __DRItexOffsetExtension radeonTexOffsetExtension = {
     { __DRI_TEX_OFFSET, __DRI_TEX_OFFSET_VERSION },
     radeonSetTexOffset,
@@ -350,7 +350,7 @@ static const __DRItexBufferExtension radeonTexBufferExtension = {
 };
 #endif
 
-#if RADEON_COMMON && defined(RADEON_COMMON_FOR_R200)
+#if defined(RADEON_R200)
 static const __DRIallocateExtension r200AllocateExtension = {
     { __DRI_ALLOCATE, __DRI_ALLOCATE_VERSION },
     r200AllocateMemoryMESA,
@@ -370,7 +370,7 @@ static const __DRItexBufferExtension r200TexBufferExtension = {
 };
 #endif
 
-#if RADEON_COMMON && defined(RADEON_COMMON_FOR_R300)
+#if defined(RADEON_R300)
 static const __DRItexOffsetExtension r300texOffsetExtension = {
     { __DRI_TEX_OFFSET, __DRI_TEX_OFFSET_VERSION },
    r300SetTexOffset,
@@ -383,7 +383,7 @@ static const __DRItexBufferExtension r300TexBufferExtension = {
 };
 #endif
 
-#if RADEON_COMMON && defined(RADEON_COMMON_FOR_R600)
+#if defined(RADEON_R600)
 static const __DRItexOffsetExtension r600texOffsetExtension = {
     { __DRI_TEX_OFFSET, __DRI_TEX_OFFSET_VERSION },
    r600SetTexOffset, /* +r6/r7 */
@@ -1222,22 +1222,22 @@ radeonCreateScreen( __DRIscreenPrivate *sPriv )
        screen->extensions[i++] = &driMediaStreamCounterExtension.base;
    }
 
-#if !RADEON_COMMON
+#if defined(RADEON_R100)
    screen->extensions[i++] = &radeonTexOffsetExtension.base;
 #endif
 
-#if RADEON_COMMON && defined(RADEON_COMMON_FOR_R200)
+#if defined(RADEON_R200)
    if (IS_R200_CLASS(screen))
       screen->extensions[i++] = &r200AllocateExtension.base;
 
    screen->extensions[i++] = &r200texOffsetExtension.base;
 #endif
 
-#if RADEON_COMMON && defined(RADEON_COMMON_FOR_R300)
+#if defined(RADEON_R300)
    screen->extensions[i++] = &r300texOffsetExtension.base;
 #endif
 
-#if RADEON_COMMON && defined(RADEON_COMMON_FOR_R600)
+#if defined(RADEON_R600)
    screen->extensions[i++] = &r600texOffsetExtension.base;
 #endif
 
@@ -1376,22 +1376,22 @@ radeonCreateScreen2(__DRIscreenPrivate *sPriv)
        screen->extensions[i++] = &driMediaStreamCounterExtension.base;
    }
 
-#if !RADEON_COMMON
+#if defined(RADEON_R100)
    screen->extensions[i++] = &radeonTexBufferExtension.base;
 #endif
 
-#if RADEON_COMMON && defined(RADEON_COMMON_FOR_R200)
+#if defined(RADEON_R200)
    if (IS_R200_CLASS(screen))
        screen->extensions[i++] = &r200AllocateExtension.base;
 
    screen->extensions[i++] = &r200TexBufferExtension.base;
 #endif
 
-#if RADEON_COMMON && defined(RADEON_COMMON_FOR_R300)
+#if defined(RADEON_R300)
    screen->extensions[i++] = &r300TexBufferExtension.base;
 #endif
 
-#if RADEON_COMMON && defined(RADEON_COMMON_FOR_R600)
+#if defined(RADEON_R600)
    screen->extensions[i++] = &r600TexBufferExtension.base;
 #endif
 
@@ -1589,22 +1589,22 @@ radeonDestroyBuffer(__DRIdrawablePrivate *driDrawPriv)
 static const __DRIconfig **
 radeonInitScreen(__DRIscreenPrivate *psp)
 {
-#if !RADEON_COMMON
+#if defined(RADEON_R100)
    static const char *driver_name = "Radeon";
    static const __DRIutilversion2 ddx_expected = { 4, 5, 0, 0 };
    static const __DRIversion dri_expected = { 4, 0, 0 };
    static const __DRIversion drm_expected = { 1, 6, 0 };
-#elif RADEON_COMMON && defined(RADEON_COMMON_FOR_R200)
+#elif defined(RADEON_R200)
    static const char *driver_name = "R200";
    static const __DRIutilversion2 ddx_expected = { 4, 5, 0, 0 };
    static const __DRIversion dri_expected = { 4, 0, 0 };
    static const __DRIversion drm_expected = { 1, 6, 0 };
-#elif RADEON_COMMON && defined(RADEON_COMMON_FOR_R300)
+#elif defined(RADEON_R300)
    static const char *driver_name = "R300";
    static const __DRIutilversion2 ddx_expected = { 4, 5, 0, 0 };
    static const __DRIversion dri_expected = { 4, 0, 0 };
    static const __DRIversion drm_expected = { 1, 24, 0 };
-#elif RADEON_COMMON && defined(RADEON_COMMON_FOR_R600)
+#elif defined(RADEON_R600)
    static const char *driver_name = "R600";
    static const __DRIutilversion2 ddx_expected = { 4, 5, 0, 0 };
    static const __DRIversion dri_expected = { 4, 0, 0 };
@@ -1630,13 +1630,13 @@ radeonInitScreen(__DRIscreenPrivate *psp)
     * Hello chicken.  Hello egg.  How are you two today?
     */
    driInitExtensions( NULL, card_extensions, GL_FALSE );
-#if RADEON_COMMON && defined(RADEON_COMMON_FOR_R200)
+#if defined(RADEON_R200)
    driInitExtensions( NULL, blend_extensions, GL_FALSE );
    driInitSingleExtension( NULL, ARB_vp_extension );
    driInitSingleExtension( NULL, NV_vp_extension );
    driInitSingleExtension( NULL, ATI_fs_extension );
    driInitExtensions( NULL, point_extensions, GL_FALSE );
-#elif (defined(RADEON_COMMON_FOR_R300) || defined(RADEON_COMMON_FOR_R600))
+#elif (defined(RADEON_R300) || defined(RADEON_R600))
    driInitSingleExtension( NULL, gl_20_extension );
 #endif
 
@@ -1684,13 +1684,13 @@ __DRIconfig **radeonInitScreen2(__DRIscreenPrivate *psp)
     */
    driInitExtensions( NULL, card_extensions, GL_FALSE );
    driInitExtensions( NULL, mm_extensions, GL_FALSE );
-#if RADEON_COMMON && defined(RADEON_COMMON_FOR_R200)
+#if defined(RADEON_R200)
    driInitExtensions( NULL, blend_extensions, GL_FALSE );
    driInitSingleExtension( NULL, ARB_vp_extension );
    driInitSingleExtension( NULL, NV_vp_extension );
    driInitSingleExtension( NULL, ATI_fs_extension );
    driInitExtensions( NULL, point_extensions, GL_FALSE );
-#elif (defined(RADEON_COMMON_FOR_R300) || defined(RADEON_COMMON_FOR_R600))
+#elif (defined(RADEON_R300) || defined(RADEON_R600))
    driInitSingleExtension( NULL, gl_20_extension );
 #endif
 
@@ -1772,13 +1772,13 @@ getSwapInfo( __DRIdrawablePrivate *dPriv, __DRIswapInfo * sInfo )
 const struct __DriverAPIRec driDriverAPI = {
    .InitScreen      = radeonInitScreen,
    .DestroyScreen   = radeonDestroyScreen,
-#if RADEON_COMMON && defined(RADEON_COMMON_FOR_R200)
+#if defined(RADEON_R200)
    .CreateContext   = r200CreateContext,
    .DestroyContext  = r200DestroyContext,
-#elif RADEON_COMMON && defined(RADEON_COMMON_FOR_R600)
+#elif defined(RADEON_R600)
    .CreateContext   = r600CreateContext,
    .DestroyContext  = radeonDestroyContext,
-#elif RADEON_COMMON && defined(RADEON_COMMON_FOR_R300)
+#elif defined(RADEON_R300)
    .CreateContext   = r300CreateContext,
    .DestroyContext  = radeonDestroyContext,
 #else
diff --git a/src/mesa/drivers/dri/radeon/radeon_span.c b/src/mesa/drivers/dri/radeon/radeon_span.c
index 4e100d854e..0c49c3713a 100644
--- a/src/mesa/drivers/dri/radeon/radeon_span.c
+++ b/src/mesa/drivers/dri/radeon/radeon_span.c
@@ -55,7 +55,7 @@ static void radeonSetSpanFunctions(struct radeon_renderbuffer *rrb);
 /* r200 depth buffer is always tiled - this is the formula
    according to the docs unless I typo'ed in it
 */
-#if defined(RADEON_COMMON_FOR_R200)
+#if defined(RADEON_R200)
 static GLubyte *r200_depth_2byte(const struct radeon_renderbuffer * rrb,
 				 GLint x, GLint y)
 {
@@ -106,6 +106,141 @@ static GLubyte *r200_depth_4byte(const struct radeon_renderbuffer * rrb,
 }
 #endif
 
+/* r600 tiling
+ * two main types:
+ * - 1D (akin to macro-linear/micro-tiled on older asics)
+ * - 2D (akin to macro-tiled/micro-tiled on older asics)
+ * only 1D tiling is implemented below
+ */
+#if defined(RADEON_R600)
+static inline GLint r600_1d_tile_helper(const struct radeon_renderbuffer * rrb,
+					GLint x, GLint y, GLint is_depth, GLint is_stencil)
+{
+    GLint element_bytes = rrb->cpp;
+    GLint num_samples = 1;
+    GLint tile_width = 8;
+    GLint tile_height = 8;
+    GLint tile_thickness = 1;
+    GLint pitch_elements = rrb->pitch / element_bytes;
+    GLint height = rrb->base.Height;
+    GLint z = 0;
+    GLint sample_number = 0;
+    /* */
+    GLint tile_bytes;
+    GLint tiles_per_row;
+    GLint tiles_per_slice;
+    GLint slice_offset;
+    GLint tile_row_index;
+    GLint tile_column_index;
+    GLint tile_offset;
+    GLint pixel_number = 0;
+    GLint element_offset;
+    GLint offset = 0;
+
+    tile_bytes = tile_width * tile_height * tile_thickness * element_bytes * num_samples;
+    tiles_per_row = pitch_elements / tile_width;
+    tiles_per_slice = tiles_per_row * (height / tile_height);
+    slice_offset = (z / tile_thickness) * tiles_per_slice * tile_bytes;
+    tile_row_index = y / tile_height;
+    tile_column_index = x / tile_width;
+    tile_offset = ((tile_row_index * tiles_per_row) + tile_column_index) * tile_bytes;
+
+    if (is_depth) {
+	    GLint pixel_offset = 0;
+
+	    pixel_number |= ((x >> 0) & 1) << 0; // pn[0] = x[0]
+	    pixel_number |= ((y >> 0) & 1) << 1; // pn[1] = y[0]
+	    pixel_number |= ((x >> 1) & 1) << 2; // pn[2] = x[1]
+	    pixel_number |= ((y >> 1) & 1) << 3; // pn[3] = y[1]
+	    pixel_number |= ((x >> 2) & 1) << 4; // pn[4] = x[2]
+	    pixel_number |= ((y >> 2) & 1) << 5; // pn[5] = y[2]
+	    switch (element_bytes) {
+	    case 2:
+		    pixel_offset = pixel_number * element_bytes * num_samples;
+		    break;
+	    case 4:
+		    /* stencil and depth data are stored separately within a tile.
+		     * stencil is stored in a contiguous tile before the depth tile.
+		     * stencil element is 1 byte, depth element is 3 bytes.
+		     * stencil tile is 64 bytes.
+		     */
+		    if (is_stencil)
+			    pixel_offset = pixel_number * 1 * num_samples;
+		    else
+			    pixel_offset = (pixel_number * 3 * num_samples) + 64;
+		    break;
+	    }
+	    element_offset = pixel_offset + (sample_number * element_bytes);
+    } else {
+	    GLint sample_offset;
+
+	    switch (element_bytes) {
+	    case 1:
+		    pixel_number |= ((x >> 0) & 1) << 0; // pn[0] = x[0]
+		    pixel_number |= ((x >> 1) & 1) << 1; // pn[1] = x[1]
+		    pixel_number |= ((x >> 2) & 1) << 2; // pn[2] = x[2]
+		    pixel_number |= ((y >> 1) & 1) << 3; // pn[3] = y[1]
+		    pixel_number |= ((y >> 0) & 1) << 4; // pn[4] = y[0]
+		    pixel_number |= ((y >> 2) & 1) << 5; // pn[5] = y[2]
+		    break;
+	    case 2:
+		    pixel_number |= ((x >> 0) & 1) << 0; // pn[0] = x[0]
+		    pixel_number |= ((x >> 1) & 1) << 1; // pn[1] = x[1]
+		    pixel_number |= ((x >> 2) & 1) << 2; // pn[2] = x[2]
+		    pixel_number |= ((y >> 0) & 1) << 3; // pn[3] = y[0]
+		    pixel_number |= ((y >> 1) & 1) << 4; // pn[4] = y[1]
+		    pixel_number |= ((y >> 2) & 1) << 5; // pn[5] = y[2]
+		    break;
+	    case 4:
+		    pixel_number |= ((x >> 0) & 1) << 0; // pn[0] = x[0]
+		    pixel_number |= ((x >> 1) & 1) << 1; // pn[1] = x[1]
+		    pixel_number |= ((y >> 0) & 1) << 2; // pn[2] = y[0]
+		    pixel_number |= ((x >> 2) & 1) << 3; // pn[3] = x[2]
+		    pixel_number |= ((y >> 1) & 1) << 4; // pn[4] = y[1]
+		    pixel_number |= ((y >> 2) & 1) << 5; // pn[5] = y[2]
+		    break;
+	    }
+	    sample_offset = sample_number * (tile_bytes / num_samples);
+	    element_offset = sample_offset + (pixel_number * element_bytes);
+    }
+    offset = slice_offset + tile_offset + element_offset;
+    return offset;
+}
+
+/* depth buffers */
+static GLubyte *r600_ptr_depth(const struct radeon_renderbuffer * rrb,
+			       GLint x, GLint y)
+{
+    GLubyte *ptr = rrb->bo->ptr;
+    GLint offset = r600_1d_tile_helper(rrb, x, y, 1, 0);
+    return &ptr[offset];
+}
+
+static GLubyte *r600_ptr_stencil(const struct radeon_renderbuffer * rrb,
+				 GLint x, GLint y)
+{
+    GLubyte *ptr = rrb->bo->ptr;
+    GLint offset = r600_1d_tile_helper(rrb, x, y, 1, 1);
+    return &ptr[offset];
+}
+
+static GLubyte *r600_ptr_color(const struct radeon_renderbuffer * rrb,
+			       GLint x, GLint y)
+{
+    GLubyte *ptr = rrb->bo->ptr;
+    uint32_t mask = RADEON_BO_FLAGS_MACRO_TILE | RADEON_BO_FLAGS_MICRO_TILE;
+    GLint offset;
+
+    if (rrb->has_surface || !(rrb->bo->flags & mask)) {
+        offset = x * rrb->cpp + y * rrb->pitch;
+    } else {
+	    offset = r600_1d_tile_helper(rrb, x, y, 0, 0);
+    }
+    return &ptr[offset];
+}
+
+#else
+
 /* radeon tiling on r300-r500 has 4 states,
    macro-linear/micro-linear
    macro-linear/micro-tiled
@@ -197,7 +332,10 @@ static GLubyte *radeon_ptr_2byte_8x2(const struct radeon_renderbuffer * rrb,
     return &ptr[offset];
 }
 
-#ifndef COMPILE_R300
+#endif
+
+#ifndef RADEON_R300
+#ifndef RADEON_R600
 static uint32_t
 z24s8_to_s8z24(uint32_t val)
 {
@@ -210,6 +348,7 @@ s8z24_to_z24s8(uint32_t val)
    return (val >> 24) | (val << 8);
 }
 #endif
+#endif
 
 /*
  * Note that all information needed to access pixels in a renderbuffer
@@ -270,7 +409,11 @@ s8z24_to_z24s8(uint32_t val)
 
 #define TAG(x)    radeon##x##_RGB565
 #define TAG2(x,y) radeon##x##_RGB565##y
+#if defined(RADEON_R600)
+#define GET_PTR(X,Y) r600_ptr_color(rrb, (X) + x_off, (Y) + y_off)
+#else
 #define GET_PTR(X,Y) radeon_ptr_2byte_8x2(rrb, (X) + x_off, (Y) + y_off)
+#endif
 #include "spantmp2.h"
 
 /* 16 bit, ARGB1555 color spanline and pixel functions
@@ -280,7 +423,11 @@ s8z24_to_z24s8(uint32_t val)
 
 #define TAG(x)    radeon##x##_ARGB1555
 #define TAG2(x,y) radeon##x##_ARGB1555##y
+#if defined(RADEON_R600)
+#define GET_PTR(X,Y) r600_ptr_color(rrb, (X) + x_off, (Y) + y_off)
+#else
 #define GET_PTR(X,Y) radeon_ptr_2byte_8x2(rrb, (X) + x_off, (Y) + y_off)
+#endif
 #include "spantmp2.h"
 
 /* 16 bit, RGBA4 color spanline and pixel functions
@@ -290,7 +437,11 @@ s8z24_to_z24s8(uint32_t val)
 
 #define TAG(x)    radeon##x##_ARGB4444
 #define TAG2(x,y) radeon##x##_ARGB4444##y
+#if defined(RADEON_R600)
+#define GET_PTR(X,Y) r600_ptr_color(rrb, (X) + x_off, (Y) + y_off)
+#else
 #define GET_PTR(X,Y) radeon_ptr_2byte_8x2(rrb, (X) + x_off, (Y) + y_off)
+#endif
 #include "spantmp2.h"
 
 /* 32 bit, xRGB8888 color spanline and pixel functions
@@ -300,11 +451,19 @@ s8z24_to_z24s8(uint32_t val)
 
 #define TAG(x)    radeon##x##_xRGB8888
 #define TAG2(x,y) radeon##x##_xRGB8888##y
+#if defined(RADEON_R600)
+#define GET_VALUE(_x, _y) ((*(GLuint*)(r600_ptr_color(rrb, _x + x_off, _y + y_off)) | 0xff000000))
+#define PUT_VALUE(_x, _y, d) { \
+   GLuint *_ptr = (GLuint*)r600_ptr_color( rrb, _x + x_off, _y + y_off );		\
+   *_ptr = d;								\
+} while (0)
+#else
 #define GET_VALUE(_x, _y) ((*(GLuint*)(radeon_ptr_4byte(rrb, _x + x_off, _y + y_off)) | 0xff000000))
 #define PUT_VALUE(_x, _y, d) { \
    GLuint *_ptr = (GLuint*)radeon_ptr_4byte( rrb, _x + x_off, _y + y_off );		\
    *_ptr = d;								\
 } while (0)
+#endif
 #include "spantmp2.h"
 
 /* 32 bit, ARGB8888 color spanline and pixel functions
@@ -314,11 +473,19 @@ s8z24_to_z24s8(uint32_t val)
 
 #define TAG(x)    radeon##x##_ARGB8888
 #define TAG2(x,y) radeon##x##_ARGB8888##y
+#if defined(RADEON_R600)
+#define GET_VALUE(_x, _y) (*(GLuint*)(r600_ptr_color(rrb, _x + x_off, _y + y_off)))
+#define PUT_VALUE(_x, _y, d) { \
+   GLuint *_ptr = (GLuint*)r600_ptr_color( rrb, _x + x_off, _y + y_off );		\
+   *_ptr = d;								\
+} while (0)
+#else
 #define GET_VALUE(_x, _y) (*(GLuint*)(radeon_ptr_4byte(rrb, _x + x_off, _y + y_off)))
 #define PUT_VALUE(_x, _y, d) { \
    GLuint *_ptr = (GLuint*)radeon_ptr_4byte( rrb, _x + x_off, _y + y_off );		\
    *_ptr = d;								\
 } while (0)
+#endif
 #include "spantmp2.h"
 
 /* ================================================================
@@ -339,17 +506,23 @@ s8z24_to_z24s8(uint32_t val)
  */
 #define VALUE_TYPE GLushort
 
-#if defined(RADEON_COMMON_FOR_R200)
+#if defined(RADEON_R200)
 #define WRITE_DEPTH( _x, _y, d )					\
    *(GLushort *)r200_depth_2byte(rrb, _x + x_off, _y + y_off) = d
+#elif defined(RADEON_R600)
+#define WRITE_DEPTH( _x, _y, d )					\
+   *(GLushort *)r600_ptr_depth(rrb, _x + x_off, _y + y_off) = d
 #else
 #define WRITE_DEPTH( _x, _y, d )					\
    *(GLushort *)radeon_ptr_2byte_8x2(rrb, _x + x_off, _y + y_off) = d
 #endif
 
-#if defined(RADEON_COMMON_FOR_R200)
+#if defined(RADEON_R200)
 #define READ_DEPTH( d, _x, _y )						\
    d = *(GLushort *)r200_depth_2byte(rrb, _x + x_off, _y + y_off)
+#elif defined(RADEON_R600)
+#define READ_DEPTH( d, _x, _y )						\
+   d = *(GLushort *)r600_ptr_depth(rrb, _x + x_off, _y + y_off)
 #else
 #define READ_DEPTH( d, _x, _y )						\
    d = *(GLushort *)radeon_ptr_2byte_8x2(rrb, _x + x_off, _y + y_off)
@@ -365,7 +538,7 @@ s8z24_to_z24s8(uint32_t val)
  */
 #define VALUE_TYPE GLuint
 
-#if defined(COMPILE_R300)
+#if defined(RADEON_R300)
 #define WRITE_DEPTH( _x, _y, d )					\
 do {									\
    GLuint *_ptr = (GLuint*)radeon_ptr_4byte( rrb, _x + x_off, _y + y_off );		\
@@ -374,7 +547,16 @@ do {									\
    tmp |= ((d << 8) & 0xffffff00);					\
    *_ptr = tmp;					\
 } while (0)
-#elif defined(RADEON_COMMON_FOR_R200)
+#elif defined(RADEON_R600)
+#define WRITE_DEPTH( _x, _y, d )					\
+do {									\
+   GLuint *_ptr = (GLuint*)r600_ptr_depth( rrb, _x + x_off, _y + y_off );		\
+   GLuint tmp = *_ptr;				\
+   tmp &= 0xff000000;							\
+   tmp |= ((d) & 0x00ffffff);					\
+   *_ptr = tmp;					\
+} while (0)
+#elif defined(RADEON_R200)
 #define WRITE_DEPTH( _x, _y, d )					\
 do {									\
    GLuint *_ptr = (GLuint*)r200_depth_4byte( rrb, _x + x_off, _y + y_off );		\
@@ -394,12 +576,17 @@ do {									\
 } while (0)
 #endif
 
-#if defined(COMPILE_R300)
+#if defined(RADEON_R300)
 #define READ_DEPTH( d, _x, _y )						\
   do {									\
     d = (*(GLuint*)(radeon_ptr_4byte(rrb, _x + x_off, _y + y_off)) & 0xffffff00) >> 8; \
   }while(0)
-#elif defined(RADEON_COMMON_FOR_R200)
+#elif defined(RADEON_R600)
+#define READ_DEPTH( d, _x, _y )						\
+  do {									\
+    d = (*(GLuint*)(r600_ptr_depth(rrb, _x + x_off, _y + y_off)) & 0x00ffffff); \
+  }while(0)
+#elif defined(RADEON_R200)
 #define READ_DEPTH( d, _x, _y )						\
   do {									\
     d = *(GLuint*)(r200_depth_4byte(rrb, _x + x_off, _y + y_off)) & 0x00ffffff; \
@@ -420,13 +607,27 @@ do {									\
  */
 #define VALUE_TYPE GLuint
 
-#if defined(COMPILE_R300)
+#if defined(RADEON_R300)
 #define WRITE_DEPTH( _x, _y, d )					\
 do {									\
    GLuint *_ptr = (GLuint*)radeon_ptr_4byte( rrb, _x + x_off, _y + y_off );		\
    *_ptr = d;								\
 } while (0)
-#elif defined(RADEON_COMMON_FOR_R200)
+#elif defined(RADEON_R600)
+#define WRITE_DEPTH( _x, _y, d )					\
+do {									\
+   GLuint *_ptr = (GLuint*)r600_ptr_depth( rrb, _x + x_off, _y + y_off );		\
+   GLuint tmp = *_ptr;				\
+   tmp &= 0xff000000;							\
+   tmp |= (((d) >> 8) & 0x00ffffff);					\
+   *_ptr = tmp;					\
+   _ptr = (GLuint*)r600_ptr_stencil(rrb, _x + x_off, _y + y_off);		\
+   tmp = *_ptr;				\
+   tmp &= 0xffffff00;							\
+   tmp |= (d) & 0xff;							\
+   *_ptr = tmp;					\
+} while (0)
+#elif defined(RADEON_R200)
 #define WRITE_DEPTH( _x, _y, d )					\
 do {									\
    GLuint *_ptr = (GLuint*)r200_depth_4byte( rrb, _x + x_off, _y + y_off );		\
@@ -442,12 +643,18 @@ do {									\
 } while (0)
 #endif
 
-#if defined(COMPILE_R300)
+#if defined(RADEON_R300)
 #define READ_DEPTH( d, _x, _y )						\
   do { \
     d = (*(GLuint*)(radeon_ptr_4byte(rrb, _x + x_off, _y + y_off)));	\
   }while(0)
-#elif defined(RADEON_COMMON_FOR_R200)
+#elif defined(RADEON_R600)
+#define READ_DEPTH( d, _x, _y )						\
+  do { \
+    d = ((*(GLuint*)(r600_ptr_depth(rrb, _x + x_off, _y + y_off))) << 8) & 0xffffff00; \
+    d |= (*(GLuint*)(r600_ptr_stencil(rrb, _x + x_off, _y + y_off))) & 0x000000ff;	\
+  }while(0)
+#elif defined(RADEON_R200)
 #define READ_DEPTH( d, _x, _y )						\
   do { \
     d = s8z24_to_z24s8(*(GLuint*)(r200_depth_4byte(rrb, _x + x_off, _y + y_off)));	\
@@ -467,7 +674,7 @@ do {									\
 
 /* 24 bit depth, 8 bit stencil depthbuffer functions
  */
-#ifdef COMPILE_R300
+#ifdef RADEON_R300
 #define WRITE_STENCIL( _x, _y, d )					\
 do {									\
    GLuint *_ptr = (GLuint*)radeon_ptr_4byte(rrb, _x + x_off, _y + y_off);		\
@@ -476,7 +683,16 @@ do {									\
    tmp |= (d) & 0xff;							\
    *_ptr = tmp;					\
 } while (0)
-#elif defined(RADEON_COMMON_FOR_R200)
+#elif defined(RADEON_R600)
+#define WRITE_STENCIL( _x, _y, d )					\
+do {									\
+   GLuint *_ptr = (GLuint*)r600_ptr_stencil(rrb, _x + x_off, _y + y_off);		\
+   GLuint tmp = *_ptr;				\
+   tmp &= 0xffffff00;							\
+   tmp |= (d) & 0xff;							\
+   *_ptr = tmp;					\
+} while (0)
+#elif defined(RADEON_R200)
 #define WRITE_STENCIL( _x, _y, d )					\
 do {									\
    GLuint *_ptr = (GLuint*)r200_depth_4byte(rrb, _x + x_off, _y + y_off);		\
@@ -496,14 +712,21 @@ do {									\
 } while (0)
 #endif
 
-#ifdef COMPILE_R300
+#ifdef RADEON_R300
 #define READ_STENCIL( d, _x, _y )					\
 do {									\
    GLuint *_ptr = (GLuint*)radeon_ptr_4byte( rrb, _x + x_off, _y + y_off );		\
    GLuint tmp = *_ptr;				\
    d = tmp & 0x000000ff;						\
 } while (0)
-#elif defined(RADEON_COMMON_FOR_R200)
+#elif defined(RADEON_R600)
+#define READ_STENCIL( d, _x, _y )					\
+do {									\
+   GLuint *_ptr = (GLuint*)r600_ptr_stencil( rrb, _x + x_off, _y + y_off );		\
+   GLuint tmp = *_ptr;				\
+   d = tmp & 0x000000ff;						\
+} while (0)
+#elif defined(RADEON_R200)
 #define READ_STENCIL( d, _x, _y )					\
 do {									\
    GLuint *_ptr = (GLuint*)r200_depth_4byte( rrb, _x + x_off, _y + y_off );		\
diff --git a/src/mesa/drivers/dri/radeon/radeon_texstate.c b/src/mesa/drivers/dri/radeon/radeon_texstate.c
index 9d252aa74c..ae41b90efe 100644
--- a/src/mesa/drivers/dri/radeon/radeon_texstate.c
+++ b/src/mesa/drivers/dri/radeon/radeon_texstate.c
@@ -833,11 +833,14 @@ static void import_tex_obj_state( r100ContextPtr rmesa,
    cmd[TEX_PP_TXFORMAT] |= texobj->pp_txformat & TEXOBJ_TXFORMAT_MASK;
    cmd[TEX_PP_BORDER_COLOR] = texobj->pp_border_color;
 
-   if (texobj->base.Target == GL_TEXTURE_RECTANGLE_NV) {
-      GLuint *txr_cmd = RADEON_DB_STATE( txr[unit] );
+   if (texobj->pp_txformat & RADEON_TXFORMAT_NON_POWER2) {
+      uint32_t *txr_cmd = &rmesa->hw.txr[unit].cmd[TXR_CMD_0];
       txr_cmd[TXR_PP_TEX_SIZE] = texobj->pp_txsize; /* NPOT only! */
       txr_cmd[TXR_PP_TEX_PITCH] = texobj->pp_txpitch; /* NPOT only! */
-      RADEON_DB_STATECHANGE( rmesa, &rmesa->hw.txr[unit] );
+      RADEON_STATECHANGE( rmesa, txr[unit] );
+   }
+
+   if (texobj->base.Target == GL_TEXTURE_RECTANGLE_NV) {
       se_coord_fmt |= RADEON_VTX_ST0_NONPARAMETRIC << unit;
    }
    else {
@@ -1114,7 +1117,6 @@ static GLboolean radeon_validate_texture(GLcontext *ctx, struct gl_texture_objec
    RADEON_STATECHANGE( rmesa, ctx );
    rmesa->hw.ctx.cmd[CTX_PP_CNTL] |= 
      (RADEON_TEX_0_ENABLE | RADEON_TEX_BLEND_0_ENABLE) << unit;
-
    RADEON_STATECHANGE( rmesa, tcl );
    rmesa->hw.tcl.cmd[TCL_OUTPUT_VTXFMT] |= RADEON_ST_BIT(unit);
 
diff --git a/src/mesa/drivers/dri/radeon/radeon_texture.c b/src/mesa/drivers/dri/radeon/radeon_texture.c
index fad3d1ceda..049284ef8c 100644
--- a/src/mesa/drivers/dri/radeon/radeon_texture.c
+++ b/src/mesa/drivers/dri/radeon/radeon_texture.c
@@ -659,11 +659,6 @@ static void radeon_teximage(
 			if (dims == 3)
 				_mesa_free(dstImageOffsets);
 		}
-
-		/* SGIS_generate_mipmap */
-		if (level == texObj->BaseLevel && texObj->GenerateMipmap) {
-			radeon_generate_mipmap(ctx, target, texObj);
-		}
 	}
 
 	_mesa_unmap_teximage_pbo(ctx, packing);
@@ -792,11 +787,6 @@ static void radeon_texsubimage(GLcontext* ctx, int dims, GLenum target, int leve
 							     format, type, pixels, packing))
 				_mesa_error(ctx, GL_OUT_OF_MEMORY, "glTexSubImage");
 		}
-
-		/* GL_SGIS_generate_mipmap */
-		if (level == texObj->BaseLevel && texObj->GenerateMipmap) {
-			radeon_generate_mipmap(ctx, target, texObj);
-		}
 	}
 
 	radeon_teximage_unmap(image);
diff --git a/src/mesa/drivers/dri/s3v/s3v_tex.c b/src/mesa/drivers/dri/s3v/s3v_tex.c
index 9b92519862..ec1182f34f 100644
--- a/src/mesa/drivers/dri/s3v/s3v_tex.c
+++ b/src/mesa/drivers/dri/s3v/s3v_tex.c
@@ -536,24 +536,13 @@ void s3vInitTextureFuncs( GLcontext *ctx )
 #endif
 
 	ctx->Driver.TexEnv = s3vTexEnv;
-	ctx->Driver.ChooseTextureFormat = _mesa_choose_tex_format;
-	ctx->Driver.TexImage1D = _mesa_store_teximage1d;
 	ctx->Driver.TexImage2D = s3vTexImage2D;
-	ctx->Driver.TexImage3D = _mesa_store_teximage3d;
-	ctx->Driver.TexSubImage1D = _mesa_store_texsubimage1d;
 	ctx->Driver.TexSubImage2D = s3vTexSubImage2D;
-	ctx->Driver.TexSubImage3D = _mesa_store_texsubimage3d;
-	ctx->Driver.CopyTexImage1D = _swrast_copy_teximage1d;
-	ctx->Driver.CopyTexImage2D = _swrast_copy_teximage2d;
-	ctx->Driver.CopyTexSubImage1D = _swrast_copy_texsubimage1d;
-	ctx->Driver.CopyTexSubImage2D = _swrast_copy_texsubimage2d;
-	ctx->Driver.CopyTexSubImage3D = _swrast_copy_texsubimage3d;
 	ctx->Driver.BindTexture = s3vBindTexture;
 	ctx->Driver.DeleteTexture = s3vDeleteTexture;
 	ctx->Driver.TexParameter = s3vTexParameter;
 	ctx->Driver.UpdateTexturePalette = 0;
 	ctx->Driver.IsTextureResident = s3vIsTextureResident;
-	ctx->Driver.TestProxyTexImage = _mesa_test_proxy_teximage;
 
 	s3vInitTextureObjects( ctx );
 }
diff --git a/src/mesa/drivers/dri/tdfx/tdfx_tex.c b/src/mesa/drivers/dri/tdfx/tdfx_tex.c
index 1f7257eaea..f6a48b3ae1 100644
--- a/src/mesa/drivers/dri/tdfx/tdfx_tex.c
+++ b/src/mesa/drivers/dri/tdfx/tdfx_tex.c
@@ -176,6 +176,54 @@ logbase2(int n)
 }
 
 
+static void
+tdfxGenerateMipmap(GLcontext *ctx, GLenum target,
+                   struct gl_texture_object *texObj)
+{
+   GLint mipWidth, mipHeight;
+   tdfxMipMapLevel *mip;
+   struct gl_texture_image *mipImage; /* the new/next image */
+   struct gl_texture_image *texImage;
+   const GLint maxLevels = _mesa_max_texture_levels(ctx, texObj->Target);
+   GLint level = texObj->BaseLevel;
+   GLsizei width, height, texelBytes;
+   const tdfxMipMapLevel *mml;
+
+   texImage = _mesa_get_tex_image(ctx, texObj, target, level);
+   assert(!texImage->IsCompressed);
+
+   mml = TDFX_TEXIMAGE_DATA(texImage);
+
+   width = texImage->Width;
+   height = texImage->Height;
+   while (level < texObj->MaxLevel && level < maxLevels - 1) {
+      mipWidth = width / 2;
+      if (!mipWidth) {
+         mipWidth = 1;
+      }
+      mipHeight = height / 2;
+      if (!mipHeight) {
+         mipHeight = 1;
+      }
+      if ((mipWidth == width) && (mipHeight == height)) {
+         break;
+      }
+      ++level;
+      mipImage = _mesa_select_tex_image(ctx, texObj, target, level);
+      mip = TDFX_TEXIMAGE_DATA(mipImage);
+      _mesa_halve2x2_teximage2d(ctx,
+                                texImage,
+                                texelBytes,
+                                mml->width, mml->height,
+                                texImage->Data, mipImage->Data);
+      texImage = mipImage;
+      mml = mip;
+      width = mipWidth;
+      height = mipHeight;
+   }
+}
+
+
 /*
  * Compute various texture image parameters.
  * Input:  w, h - source texture width and height
@@ -1397,45 +1445,6 @@ tdfxTexImage2D(GLcontext *ctx, GLenum target, GLint level,
                                           width, height, 1,
                                           format, type, pixels, packing);
        }
-
-      /* GL_SGIS_generate_mipmap */
-      if (level == texObj->BaseLevel && texObj->GenerateMipmap) {
-         GLint mipWidth, mipHeight;
-         tdfxMipMapLevel *mip;
-         struct gl_texture_image *mipImage;
-         const GLint maxLevels = _mesa_max_texture_levels(ctx, texObj->Target);
-   
-         assert(!texImage->IsCompressed);
-   
-         while (level < texObj->MaxLevel && level < maxLevels - 1) {
-            mipWidth = width / 2;
-            if (!mipWidth) {
-               mipWidth = 1;
-            }
-            mipHeight = height / 2;
-            if (!mipHeight) {
-               mipHeight = 1;
-            }
-            if ((mipWidth == width) && (mipHeight == height)) {
-               break;
-            }
-            _mesa_TexImage2D(target, ++level, internalFormat,
-                             mipWidth, mipHeight, border,
-                             format, type,
-                             NULL);
-            mipImage = _mesa_select_tex_image(ctx, texObj, target, level);
-            mip = TDFX_TEXIMAGE_DATA(mipImage);
-            _mesa_halve2x2_teximage2d(ctx,
-                                      texImage,
-                                      texelBytes,
-                                      mml->width, mml->height,
-                                      texImage->Data, mipImage->Data);
-            texImage = mipImage;
-            mml = mip;
-            width = mipWidth;
-            height = mipHeight;
-         }
-      }
     }
 
     RevalidateTexture(ctx, texObj);
@@ -1507,44 +1516,6 @@ tdfxTexSubImage2D(GLcontext *ctx, GLenum target, GLint level,
                                         format, type, pixels, packing);
     }
 
-   /* GL_SGIS_generate_mipmap */
-   if (level == texObj->BaseLevel && texObj->GenerateMipmap) {
-      GLint mipWidth, mipHeight;
-      tdfxMipMapLevel *mip;
-      struct gl_texture_image *mipImage;
-      const GLint maxLevels = _mesa_max_texture_levels(ctx, texObj->Target);
-
-      assert(!texImage->IsCompressed);
-
-      width = texImage->Width;
-      height = texImage->Height;
-      while (level < texObj->MaxLevel && level < maxLevels - 1) {
-         mipWidth = width / 2;
-         if (!mipWidth) {
-            mipWidth = 1;
-         }
-         mipHeight = height / 2;
-         if (!mipHeight) {
-            mipHeight = 1;
-         }
-         if ((mipWidth == width) && (mipHeight == height)) {
-            break;
-         }
-         ++level;
-         mipImage = _mesa_select_tex_image(ctx, texObj, target, level);
-         mip = TDFX_TEXIMAGE_DATA(mipImage);
-         _mesa_halve2x2_teximage2d(ctx,
-                                   texImage,
-                                   texelBytes,
-                                   mml->width, mml->height,
-                                   texImage->Data, mipImage->Data);
-         texImage = mipImage;
-         mml = mip;
-         width = mipWidth;
-         height = mipHeight;
-      }
-   }
-
     ti->reloadImages = GL_TRUE; /* signal the image needs to be reloaded */
     fxMesa->new_state |= TDFX_NEW_TEXTURE;  /* XXX this might be a bit much */
 }
@@ -1703,11 +1674,6 @@ tdfxCompressedTexImage2D (GLcontext *ctx, GLenum target,
        MEMCPY(texImage->Data, data, texImage->CompressedSize);
     }
 
-    /* GL_SGIS_generate_mipmap */
-    if (level == texObj->BaseLevel && texObj->GenerateMipmap) {
-       assert(!texImage->IsCompressed);
-    }
-
     RevalidateTexture(ctx, texObj);
 
     ti->reloadImages = GL_TRUE;
@@ -1770,11 +1736,6 @@ tdfxCompressedTexSubImage2D( GLcontext *ctx, GLenum target,
                                 texImage->Data);
     }
 
-    /* GL_SGIS_generate_mipmap */
-    if (level == texObj->BaseLevel && texObj->GenerateMipmap) {
-       assert(!texImage->IsCompressed);
-    }
-
     RevalidateTexture(ctx, texObj);
 
     ti->reloadImages = GL_TRUE;
@@ -1914,4 +1875,5 @@ void tdfxInitTextureFuncs( struct dd_function_table *functions )
    functions->CompressedTexImage2D	= tdfxCompressedTexImage2D;
    functions->CompressedTexSubImage2D	= tdfxCompressedTexSubImage2D;
    functions->UpdateTexturePalette      = tdfxUpdateTexturePalette;
+   functions->GenerateMipmap            = tdfxGenerateMipmap;
 }
diff --git a/src/mesa/drivers/dri/unichrome/via_tex.c b/src/mesa/drivers/dri/unichrome/via_tex.c
index d2010f0907..54073e7691 100644
--- a/src/mesa/drivers/dri/unichrome/via_tex.c
+++ b/src/mesa/drivers/dri/unichrome/via_tex.c
@@ -818,11 +818,6 @@ static void viaTexImage(GLcontext *ctx,
       }
    }
 
-   /* GL_SGIS_generate_mipmap */
-   if (level == texObj->BaseLevel && texObj->GenerateMipmap) {
-      _mesa_generate_mipmap(ctx, target, texObj);
-   }
-
    _mesa_unmap_teximage_pbo(ctx, packing);
 }