Merge remote-tracking branch 'origin/released'

33 files changed, 358 insertions, 1010 deletions

diff --git a/mesalib/configure.ac b/mesalib/configure.ac
index e5a5e8842..511d147cb 100644
--- a/mesalib/configure.ac
+++ b/mesalib/configure.ac
@@ -402,7 +402,6 @@ fi
 
 GL_LIB_NAME='lib$(GL_LIB).'${LIB_EXTENSION}
 GLU_LIB_NAME='lib$(GLU_LIB).'${LIB_EXTENSION}
-GLUT_LIB_NAME='lib$(GLUT_LIB).'${LIB_EXTENSION}
 OSMESA_LIB_NAME='lib$(OSMESA_LIB).'${LIB_EXTENSION}
 EGL_LIB_NAME='lib$(EGL_LIB).'${LIB_EXTENSION}
 GLESv1_CM_LIB_NAME='lib$(GLESv1_CM_LIB).'${LIB_EXTENSION}
@@ -414,7 +413,6 @@ GBM_LIB_NAME='lib$(GBM_LIB).'${LIB_EXTENSION}
 
 GL_LIB_GLOB=${LIB_PREFIX_GLOB}'$(GL_LIB)'${LIB_VERSION_SEPARATOR}'*'${LIB_EXTENSION}'*'
 GLU_LIB_GLOB=${LIB_PREFIX_GLOB}'$(GLU_LIB)'${LIB_VERSION_SEPARATOR}'*'${LIB_EXTENSION}'*'
-GLUT_LIB_GLOB=${LIB_PREFIX_GLOB}'$(GLUT_LIB)'${LIB_VERSION_SEPARATOR}'*'${LIB_EXTENSION}'*'
 OSMESA_LIB_GLOB=${LIB_PREFIX_GLOB}'$(OSMESA_LIB)'${LIB_VERSION_SEPARATOR}'*'${LIB_EXTENSION}'*'
 EGL_LIB_GLOB=${LIB_PREFIX_GLOB}'$(EGL_LIB)'${LIB_VERSION_SEPARATOR}'*'${LIB_EXTENSION}'*'
 EGL_LIB_GLOB=${LIB_PREFIX_GLOB}'$(EGL_LIB)'${LIB_VERSION_SEPARATOR}'*'${LIB_EXTENSION}'*'
@@ -427,7 +425,6 @@ GBM_LIB_GLOB=${LIB_PREFIX_GLOB}'$(GBM_LIB)'${LIB_VERSION_SEPARATOR}'*'${LIB_EXTE
 
 AC_SUBST([GL_LIB_NAME])
 AC_SUBST([GLU_LIB_NAME])
-AC_SUBST([GLUT_LIB_NAME])
 AC_SUBST([OSMESA_LIB_NAME])
 AC_SUBST([EGL_LIB_NAME])
 AC_SUBST([GLESv1_CM_LIB_NAME])
@@ -439,7 +436,6 @@ AC_SUBST([GBM_LIB_NAME])
 
 AC_SUBST([GL_LIB_GLOB])
 AC_SUBST([GLU_LIB_GLOB])
-AC_SUBST([GLUT_LIB_GLOB])
 AC_SUBST([OSMESA_LIB_GLOB])
 AC_SUBST([EGL_LIB_GLOB])
 AC_SUBST([GLESv1_CM_LIB_GLOB])
diff --git a/mesalib/src/gallium/auxiliary/util/u_debug.h b/mesalib/src/gallium/auxiliary/util/u_debug.h
index cbea35830..ed19cda05 100644
--- a/mesalib/src/gallium/auxiliary/util/u_debug.h
+++ b/mesalib/src/gallium/auxiliary/util/u_debug.h
@@ -215,6 +215,25 @@ void _debug_assert_fail(const char *expr,
 
 
 /**
+ * Emit a warning message, but only once.
+ */
+#ifdef DEBUG
+#define debug_warn_once(__msg) \
+   do { \
+      static bool warned = FALSE; \
+      if (!warned) { \
+         _debug_printf("%s:%u:%s: one time warning: %s\n", \
+                       __FILE__, __LINE__, __FUNCTION__, __msg); \
+         warned = TRUE; \
+      } \
+   } while (0)
+#else
+#define debug_warn_once(__msg) \
+   ((void)0) 
+#endif
+
+
+/**
  * Output an error message. Not muted on release version.
  */
 #ifdef DEBUG
diff --git a/mesalib/src/gallium/auxiliary/util/u_format_r11g11b10f.h b/mesalib/src/gallium/auxiliary/util/u_format_r11g11b10f.h
index 6bb430296..bd64b2896 100644
--- a/mesalib/src/gallium/auxiliary/util/u_format_r11g11b10f.h
+++ b/mesalib/src/gallium/auxiliary/util/u_format_r11g11b10f.h
@@ -119,12 +119,12 @@ static INLINE float uf11_to_f32(uint16_t val)
       float scale, decimal;
       exponent -= 15;
       if (exponent < 0) {
-         scale = 1.0 / (1 << -exponent);
+         scale = 1.0f / (1 << -exponent);
       }
       else {
-         scale = 1 << exponent;
+         scale = (float) (1 << exponent);
       }
-      decimal = 1.0 + (float) mantissa / 64;
+      decimal = 1.0f + (float) mantissa / 64;
       f32.f = scale * decimal;
    }
 
@@ -208,9 +208,9 @@ static INLINE float uf10_to_f32(uint16_t val)
          scale = 1.0 / (1 << -exponent);
       }
       else {
-         scale = 1 << exponent;
+         scale = (float) (1 << exponent);
       }
-      decimal = 1.0 + (float) mantissa / 32;
+      decimal = 1.0f + (float) mantissa / 32;
       f32.f = scale * decimal;
    }
 
diff --git a/mesalib/src/gallium/auxiliary/util/u_vbuf.c b/mesalib/src/gallium/auxiliary/util/u_vbuf.c
index 711ad10e8..50c35afd5 100644
--- a/mesalib/src/gallium/auxiliary/util/u_vbuf.c
+++ b/mesalib/src/gallium/auxiliary/util/u_vbuf.c
@@ -912,6 +912,39 @@ static boolean u_vbuf_need_minmax_index(struct u_vbuf_priv *mgr)
    return FALSE;
 }
 
+static boolean u_vbuf_mapping_vertex_buffer_blocks(struct u_vbuf_priv *mgr)
+{
+   unsigned i, nr = mgr->ve->count;
+
+   for (i = 0; i < nr; i++) {
+      struct pipe_vertex_buffer *vb;
+      unsigned index;
+
+      /* Per-instance attribs are not per-vertex data. */
+      if (mgr->ve->ve[i].instance_divisor) {
+         continue;
+      }
+
+      index = mgr->ve->ve[i].vertex_buffer_index;
+      vb = &mgr->b.vertex_buffer[index];
+
+      /* Constant attribs are not per-vertex data. */
+      if (!vb->stride) {
+         continue;
+      }
+
+      /* Return true for the hw buffers which don't need to be translated. */
+      /* XXX we could use some kind of a is-busy query. */
+      if (!u_vbuf_resource(vb->buffer)->user_ptr &&
+          !mgr->ve->incompatible_layout_elem[i] &&
+          !mgr->incompatible_vb[index]) {
+         return TRUE;
+      }
+   }
+
+   return FALSE;
+}
+
 static void u_vbuf_get_minmax_index(struct pipe_context *pipe,
                                     struct pipe_index_buffer *ib,
                                     const struct pipe_draw_info *info,
@@ -1054,7 +1087,8 @@ u_vbuf_draw_begin(struct u_vbuf *mgrb,
           * performance. */
          if (!info->primitive_restart &&
              num_vertices > info->count*2 &&
-             num_vertices-info->count > 32) {
+             num_vertices-info->count > 32 &&
+             !u_vbuf_mapping_vertex_buffer_blocks(mgr)) {
             /*printf("num_vertices=%i count=%i\n", num_vertices, info->count);*/
             unroll_indices = true;
          }
diff --git a/mesalib/src/glsl/ast.h b/mesalib/src/glsl/ast.h
index 79054be86..93d36bf34 100644
--- a/mesalib/src/glsl/ast.h
+++ b/mesalib/src/glsl/ast.h
@@ -211,6 +211,7 @@ public:
       subexpressions[1] = NULL;
       subexpressions[2] = NULL;
       primary_expression.identifier = (char *) identifier;
+      this->non_lvalue_description = NULL;
    }
 
    static const char *operator_string(enum ast_operators op);
@@ -238,6 +239,18 @@ public:
     * \c ast_function_call
     */
    exec_list expressions;
+
+   /**
+    * For things that can't be l-values, this describes what it is.
+    *
+    * This text is used by the code that generates IR for assignments to
+    * detect and emit useful messages for assignments to some things that
+    * can't be l-values.  For example, pre- or post-incerement expressions.
+    *
+    * \note
+    * This pointer may be \c NULL.
+    */
+   const char *non_lvalue_description;
 };
 
 class ast_expression_bin : public ast_expression {
diff --git a/mesalib/src/glsl/ast_function.cpp b/mesalib/src/glsl/ast_function.cpp
index 126b610d1..1c2e8613c 100644
--- a/mesalib/src/glsl/ast_function.cpp
+++ b/mesalib/src/glsl/ast_function.cpp
@@ -93,14 +93,21 @@ prototype_string(const glsl_type *return_type, const char *name,
    return str;
 }
 
+/**
+ * If a function call is generated, \c call_ir will point to it on exit.
+ * Otherwise \c call_ir will be set to \c NULL.
+ */
 static ir_rvalue *
 generate_call(exec_list *instructions, ir_function_signature *sig,
 	      YYLTYPE *loc, exec_list *actual_parameters,
+	      ir_call **call_ir,
 	      struct _mesa_glsl_parse_state *state)
 {
    void *ctx = state;
    exec_list post_call_conversions;
 
+   *call_ir = NULL;
+
    /* Verify that 'out' and 'inout' actual parameters are lvalues.  This
     * isn't done in ir_function::matching_signature because that function
     * cannot generate the necessary diagnostics.
@@ -256,10 +263,12 @@ generate_call(exec_list *instructions, ir_function_signature *sig,
       deref = new(ctx) ir_dereference_variable(var);
       ir_assignment *assign = new(ctx) ir_assignment(deref, call, NULL);
       instructions->push_tail(assign);
+      *call_ir = call;
 
       deref = new(ctx) ir_dereference_variable(var);
    } else {
       instructions->push_tail(call);
+      *call_ir = call;
       deref = NULL;
    }
    instructions->append_list(&post_call_conversions);
@@ -269,6 +278,7 @@ generate_call(exec_list *instructions, ir_function_signature *sig,
 static ir_rvalue *
 match_function_by_name(exec_list *instructions, const char *name,
 		       YYLTYPE *loc, exec_list *actual_parameters,
+		       ir_call **call_ir,
 		       struct _mesa_glsl_parse_state *state)
 {
    void *ctx = state;
@@ -342,7 +352,8 @@ done:
       }
 
       /* Finally, generate a call instruction. */
-      return generate_call(instructions, sig, loc, actual_parameters, state);
+      return generate_call(instructions, sig, loc, actual_parameters,
+			   call_ir, state);
    } else {
       char *str = prototype_string(NULL, name, actual_parameters);
 
@@ -1442,9 +1453,53 @@ ast_function_expression::hir(exec_list *instructions,
       process_parameters(instructions, &actual_parameters, &this->expressions,
 			 state);
 
-      return match_function_by_name(instructions, 
-				    id->primary_expression.identifier, & loc,
-				    &actual_parameters, state);
+      ir_call *call = NULL;
+      ir_rvalue *const value =
+	 match_function_by_name(instructions,
+				id->primary_expression.identifier,
+				&loc, &actual_parameters, &call, state);
+
+      if (call != NULL) {
+	 /* If a function was found, make sure that none of the 'out' or 'inout'
+	  * parameters violate the extra l-value rules.
+	  */
+	 ir_function_signature *f = call->get_callee();
+	 assert(f != NULL);
+
+	 exec_node *formal_node = f->parameters.head;
+
+	 foreach_list (actual_node, &this->expressions) {
+	    /* Both parameter lists had better be the same length!
+	     */
+	    assert(!actual_node->is_tail_sentinel());
+
+	    const ir_variable *const formal_parameter =
+	       (ir_variable *) formal_node;
+	    const ast_expression *const actual_parameter =
+	       exec_node_data(ast_expression, actual_node, link);
+
+	    if ((formal_parameter->mode == ir_var_out
+		 || formal_parameter->mode == ir_var_inout)
+		&& actual_parameter->non_lvalue_description != NULL) {
+	       YYLTYPE loc = actual_parameter->get_location();
+
+	       _mesa_glsl_error(&loc, state,
+				"function parameter '%s %s' references a %s",
+				(formal_parameter->mode == ir_var_out)
+				? "out" : "inout",
+				formal_parameter->name,
+				actual_parameter->non_lvalue_description);
+	       return ir_call::get_error_instruction(ctx);
+	    }
+
+	    /* Only advance the formal_node pointer here because the
+	     * foreach_list business already advances actual_node.
+	     */
+	    formal_node = formal_node->next;
+	 }
+      }
+
+      return value;
    }
 
    return ir_call::get_error_instruction(ctx);
diff --git a/mesalib/src/glsl/ast_to_hir.cpp b/mesalib/src/glsl/ast_to_hir.cpp
index f0c921875..1aebca40f 100644
--- a/mesalib/src/glsl/ast_to_hir.cpp
+++ b/mesalib/src/glsl/ast_to_hir.cpp
@@ -664,6 +664,7 @@ mark_whole_array_access(ir_rvalue *access)
 
 ir_rvalue *
 do_assignment(exec_list *instructions, struct _mesa_glsl_parse_state *state,
+	      const char *non_lvalue_description,
 	      ir_rvalue *lhs, ir_rvalue *rhs, bool is_initializer,
 	      YYLTYPE lhs_loc)
 {
@@ -671,8 +672,13 @@ do_assignment(exec_list *instructions, struct _mesa_glsl_parse_state *state,
    bool error_emitted = (lhs->type->is_error() || rhs->type->is_error());
 
    if (!error_emitted) {
-      if (lhs->variable_referenced() != NULL
-          && lhs->variable_referenced()->read_only) {
+      if (non_lvalue_description != NULL) {
+         _mesa_glsl_error(&lhs_loc, state,
+                          "assignment to %s",
+			  non_lvalue_description);
+	 error_emitted = true;
+      } else if (lhs->variable_referenced() != NULL
+		 && lhs->variable_referenced()->read_only) {
          _mesa_glsl_error(&lhs_loc, state,
                           "assignment to read-only variable '%s'",
                           lhs->variable_referenced()->name);
@@ -769,11 +775,6 @@ get_lvalue_copy(exec_list *instructions, ir_rvalue *lvalue)
    instructions->push_tail(new(ctx) ir_assignment(new(ctx) ir_dereference_variable(var),
 						  lvalue, NULL));
 
-   /* Once we've created this temporary, mark it read only so it's no
-    * longer considered an lvalue.
-    */
-   var->read_only = true;
-
    return new(ctx) ir_dereference_variable(var);
 }
 
@@ -1030,7 +1031,9 @@ ast_expression::hir(exec_list *instructions,
       op[0] = this->subexpressions[0]->hir(instructions, state);
       op[1] = this->subexpressions[1]->hir(instructions, state);
 
-      result = do_assignment(instructions, state, op[0], op[1], false,
+      result = do_assignment(instructions, state,
+			     this->subexpressions[0]->non_lvalue_description,
+			     op[0], op[1], false,
 			     this->subexpressions[0]->get_location());
       error_emitted = result->type->is_error();
       break;
@@ -1310,6 +1313,7 @@ ast_expression::hir(exec_list *instructions,
 						   op[0], op[1]);
 
       result = do_assignment(instructions, state,
+			     this->subexpressions[0]->non_lvalue_description,
 			     op[0]->clone(ctx, NULL), temp_rhs, false,
 			     this->subexpressions[0]->get_location());
       error_emitted = (op[0]->type->is_error());
@@ -1335,6 +1339,7 @@ ast_expression::hir(exec_list *instructions,
 					op[0], op[1]);
 
       result = do_assignment(instructions, state,
+			     this->subexpressions[0]->non_lvalue_description,
 			     op[0]->clone(ctx, NULL), temp_rhs, false,
 			     this->subexpressions[0]->get_location());
       error_emitted = type->is_error();
@@ -1349,8 +1354,9 @@ ast_expression::hir(exec_list *instructions,
                                &loc);
       ir_rvalue *temp_rhs = new(ctx) ir_expression(operations[this->oper],
                                                    type, op[0], op[1]);
-      result = do_assignment(instructions, state, op[0]->clone(ctx, NULL),
-                             temp_rhs, false,
+      result = do_assignment(instructions, state,
+			     this->subexpressions[0]->non_lvalue_description,
+			     op[0]->clone(ctx, NULL), temp_rhs, false,
                              this->subexpressions[0]->get_location());
       error_emitted = op[0]->type->is_error() || op[1]->type->is_error();
       break;
@@ -1365,8 +1371,9 @@ ast_expression::hir(exec_list *instructions,
                                    state, &loc);
       ir_rvalue *temp_rhs = new(ctx) ir_expression(operations[this->oper],
                                                    type, op[0], op[1]);
-      result = do_assignment(instructions, state, op[0]->clone(ctx, NULL),
-                             temp_rhs, false,
+      result = do_assignment(instructions, state,
+			     this->subexpressions[0]->non_lvalue_description,
+			     op[0]->clone(ctx, NULL), temp_rhs, false,
                              this->subexpressions[0]->get_location());
       error_emitted = op[0]->type->is_error() || op[1]->type->is_error();
       break;
@@ -1463,6 +1470,9 @@ ast_expression::hir(exec_list *instructions,
 
    case ast_pre_inc:
    case ast_pre_dec: {
+      this->non_lvalue_description = (this->oper == ast_pre_inc)
+	 ? "pre-increment operation" : "pre-decrement operation";
+
       op[0] = this->subexpressions[0]->hir(instructions, state);
       op[1] = constant_one_for_inc_dec(ctx, op[0]->type);
 
@@ -1473,6 +1483,7 @@ ast_expression::hir(exec_list *instructions,
 					op[0], op[1]);
 
       result = do_assignment(instructions, state,
+			     this->subexpressions[0]->non_lvalue_description,
 			     op[0]->clone(ctx, NULL), temp_rhs, false,
 			     this->subexpressions[0]->get_location());
       error_emitted = op[0]->type->is_error();
@@ -1481,6 +1492,8 @@ ast_expression::hir(exec_list *instructions,
 
    case ast_post_inc:
    case ast_post_dec: {
+      this->non_lvalue_description = (this->oper == ast_post_inc)
+	 ? "post-increment operation" : "post-decrement operation";
       op[0] = this->subexpressions[0]->hir(instructions, state);
       op[1] = constant_one_for_inc_dec(ctx, op[0]->type);
 
@@ -1498,6 +1511,7 @@ ast_expression::hir(exec_list *instructions,
       result = get_lvalue_copy(instructions, op[0]->clone(ctx, NULL));
 
       (void)do_assignment(instructions, state,
+			  this->subexpressions[0]->non_lvalue_description,
 			  op[0]->clone(ctx, NULL), temp_rhs, false,
 			  this->subexpressions[0]->get_location());
 
@@ -2365,6 +2379,7 @@ process_initializer(ir_variable *var, ast_declaration *decl,
       const glsl_type *initializer_type;
       if (!type->qualifier.flags.q.uniform) {
 	 result = do_assignment(initializer_instructions, state,
+				NULL,
 				lhs, rhs, true,
 				type->get_location());
 	 initializer_type = result->type;
diff --git a/mesalib/src/glsl/glsl_parser_extras.cpp b/mesalib/src/glsl/glsl_parser_extras.cpp
index 0b4ccac53..0b53232e9 100644
--- a/mesalib/src/glsl/glsl_parser_extras.cpp
+++ b/mesalib/src/glsl/glsl_parser_extras.cpp
@@ -626,6 +626,7 @@ ast_expression::ast_expression(int oper,
    this->subexpressions[0] = ex0;
    this->subexpressions[1] = ex1;
    this->subexpressions[2] = ex2;
+   this->non_lvalue_description = NULL;
 }
 
 
diff --git a/mesalib/src/mesa/SConscript b/mesalib/src/mesa/SConscript
index bd9e94b82..4754fbfc8 100644
--- a/mesalib/src/mesa/SConscript
+++ b/mesalib/src/mesa/SConscript
@@ -161,7 +161,6 @@ swrast_sources = [
     'swrast/s_copypix.c',
     'swrast/s_context.c',
     'swrast/s_depth.c',
-    'swrast/s_depthstencil.c',
     'swrast/s_drawpix.c',
     'swrast/s_feedback.c',
     'swrast/s_fog.c',
diff --git a/mesalib/src/mesa/drivers/common/meta.c b/mesalib/src/mesa/drivers/common/meta.c
index 785a59531..ad289aa70 100644
--- a/mesalib/src/mesa/drivers/common/meta.c
+++ b/mesalib/src/mesa/drivers/common/meta.c
@@ -1026,7 +1026,7 @@ _mesa_meta_in_progress(struct gl_context *ctx)
 static INLINE GLfloat
 invert_z(GLfloat normZ)
 {
-   GLfloat objZ = 1.0 - 2.0 * normZ;
+   GLfloat objZ = 1.0f - 2.0f * normZ;
    return objZ;
 }
 
@@ -2920,7 +2920,7 @@ _mesa_meta_GenerateMipmap(struct gl_context *ctx, GLenum target,
 
    /* setup texcoords (XXX what about border?) */
    setup_texture_coords(faceTarget,
-                        0.0, 0.0, /* width, height never used here */
+                        0.0f, 0.0f, /* width, height never used here */
                         slice,
                         verts[0].tex,
                         verts[1].tex,
@@ -3098,20 +3098,19 @@ get_temp_image_type(struct gl_context *ctx, GLenum baseFormat)
  */
 static void
 copy_tex_sub_image(struct gl_context *ctx,
-                   GLuint dims, GLenum target, GLint level,
+                   GLuint dims,
+                   struct gl_texture_image *texImage,
                    GLint xoffset, GLint yoffset, GLint zoffset,
+                   struct gl_renderbuffer *rb,
                    GLint x, GLint y,
                    GLsizei width, GLsizei height)
 {
-   struct gl_texture_object *texObj;
-   struct gl_texture_image *texImage;
+   struct gl_texture_object *texObj = texImage->TexObject;
+   const GLenum target = texObj->Target;
    GLenum format, type;
    GLint bpp;
    void *buf;
 
-   texObj = _mesa_get_current_tex_object(ctx, target);
-   texImage = _mesa_select_tex_image(ctx, texObj, target, level);
-
    /* Choose format/type for temporary image buffer */
    format = _mesa_get_format_base_format(texImage->TexFormat);
    if (format == GL_LUMINANCE ||
@@ -3180,34 +3179,40 @@ copy_tex_sub_image(struct gl_context *ctx,
 
 
 void
-_mesa_meta_CopyTexSubImage1D(struct gl_context *ctx, GLenum target, GLint level,
+_mesa_meta_CopyTexSubImage1D(struct gl_context *ctx,
+                             struct gl_texture_image *texImage,
                              GLint xoffset,
+                             struct gl_renderbuffer *rb,
                              GLint x, GLint y, GLsizei width)
 {
-   copy_tex_sub_image(ctx, 1, target, level, xoffset, 0, 0,
-                      x, y, width, 1);
+   copy_tex_sub_image(ctx, 1, texImage, xoffset, 0, 0,
+                      rb, x, y, width, 1);
 }
 
 
 void
-_mesa_meta_CopyTexSubImage2D(struct gl_context *ctx, GLenum target, GLint level,
+_mesa_meta_CopyTexSubImage2D(struct gl_context *ctx,
+                             struct gl_texture_image *texImage,
                              GLint xoffset, GLint yoffset,
+                             struct gl_renderbuffer *rb,
                              GLint x, GLint y,
                              GLsizei width, GLsizei height)
 {
-   copy_tex_sub_image(ctx, 2, target, level, xoffset, yoffset, 0,
-                      x, y, width, height);
+   copy_tex_sub_image(ctx, 2, texImage, xoffset, yoffset, 0,
+                      rb, x, y, width, height);
 }
 
 
 void
-_mesa_meta_CopyTexSubImage3D(struct gl_context *ctx, GLenum target, GLint level,
+_mesa_meta_CopyTexSubImage3D(struct gl_context *ctx,
+                             struct gl_texture_image *texImage,
                              GLint xoffset, GLint yoffset, GLint zoffset,
+                             struct gl_renderbuffer *rb,
                              GLint x, GLint y,
                              GLsizei width, GLsizei height)
 {
-   copy_tex_sub_image(ctx, 3, target, level, xoffset, yoffset, zoffset,
-                      x, y, width, height);
+   copy_tex_sub_image(ctx, 3, texImage, xoffset, yoffset, zoffset,
+                      rb, x, y, width, height);
 }
 
 
diff --git a/mesalib/src/mesa/drivers/common/meta.h b/mesalib/src/mesa/drivers/common/meta.h
index d13796ebe..de039b575 100644
--- a/mesalib/src/mesa/drivers/common/meta.h
+++ b/mesalib/src/mesa/drivers/common/meta.h
@@ -111,19 +111,25 @@ _mesa_meta_GenerateMipmap(struct gl_context *ctx, GLenum target,
                           struct gl_texture_object *texObj);
 
 extern void
-_mesa_meta_CopyTexSubImage1D(struct gl_context *ctx, GLenum target, GLint level,
+_mesa_meta_CopyTexSubImage1D(struct gl_context *ctx,
+                             struct gl_texture_image *texImage,
                              GLint xoffset,
+                             struct gl_renderbuffer *rb,
                              GLint x, GLint y, GLsizei width);
 
 extern void
-_mesa_meta_CopyTexSubImage2D(struct gl_context *ctx, GLenum target, GLint level,
+_mesa_meta_CopyTexSubImage2D(struct gl_context *ctx,
+                             struct gl_texture_image *texImage,
                              GLint xoffset, GLint yoffset,
+                             struct gl_renderbuffer *rb,
                              GLint x, GLint y,
                              GLsizei width, GLsizei height);
 
 extern void
-_mesa_meta_CopyTexSubImage3D(struct gl_context *ctx, GLenum target, GLint level,
+_mesa_meta_CopyTexSubImage3D(struct gl_context *ctx,
+                             struct gl_texture_image *texImage,
                              GLint xoffset, GLint yoffset, GLint zoffset,
+                             struct gl_renderbuffer *rb,
                              GLint x, GLint y,
                              GLsizei width, GLsizei height);
 
diff --git a/mesalib/src/mesa/main/attrib.c b/mesalib/src/mesa/main/attrib.c
index 6f4a91b1f..1c1ee5dde 100644
--- a/mesalib/src/mesa/main/attrib.c
+++ b/mesalib/src/mesa/main/attrib.c
@@ -1068,7 +1068,7 @@ _mesa_PopAttrib(void)
                      p[0] = l->QuadraticAttenuation;
                      _mesa_light(ctx, i, GL_QUADRATIC_ATTENUATION, p);
                   }
-                }
+               }
                /* light model */
                _mesa_LightModelfv(GL_LIGHT_MODEL_AMBIENT,
                                   light->Model.Ambient);
diff --git a/mesalib/src/mesa/main/bufferobj.c b/mesalib/src/mesa/main/bufferobj.c
index 63eb6faaa..b2a74f322 100644
--- a/mesalib/src/mesa/main/bufferobj.c
+++ b/mesalib/src/mesa/main/bufferobj.c
@@ -504,7 +504,7 @@ _mesa_copy_buffer_subdata(struct gl_context *ctx,
 {
    void *srcPtr, *dstPtr;
 
-   /* buffer should not already be mapped */
+   /* the buffers should not be mapped */
    assert(!_mesa_bufferobj_mapped(src));
    assert(!_mesa_bufferobj_mapped(dst));
 
@@ -514,6 +514,9 @@ _mesa_copy_buffer_subdata(struct gl_context *ctx,
                                        (GL_MAP_WRITE_BIT |
                                         GL_MAP_INVALIDATE_RANGE_BIT), dst);
 
+   /* Note: the src and dst regions will never overlap.  Trying to do so
+    * would generate GL_INVALID_VALUE earlier.
+    */
    if (srcPtr && dstPtr)
       memcpy(dstPtr, srcPtr, size);
 
diff --git a/mesalib/src/mesa/main/dd.h b/mesalib/src/mesa/main/dd.h
index 6707e785d..24f3d4ca5 100644
--- a/mesalib/src/mesa/main/dd.h
+++ b/mesalib/src/mesa/main/dd.h
@@ -283,31 +283,33 @@ struct dd_function_table {
                         struct gl_texture_image *texImage );
 
    /**
-    * Called by glCopyTexSubImage1D().
-    * 
-    * Drivers should use a fallback routine from texstore.c if needed.
+    * Called by glCopyTexSubImage1D() and glCopyTexImage1D().
     */
-   void (*CopyTexSubImage1D)( struct gl_context *ctx, GLenum target, GLint level,
-                              GLint xoffset,
-                              GLint x, GLint y, GLsizei width );
+   void (*CopyTexSubImage1D)(struct gl_context *ctx,
+                             struct gl_texture_image *texImage,
+                             GLint xoffset,
+                             struct gl_renderbuffer *rb,
+                             GLint x, GLint y, GLsizei width);
+
    /**
-    * Called by glCopyTexSubImage2D().
-    * 
-    * Drivers should use a fallback routine from texstore.c if needed.
+    * Called by glCopyTexSubImage2D() and glCopyTexImage2D().
     */
-   void (*CopyTexSubImage2D)( struct gl_context *ctx, GLenum target, GLint level,
-                              GLint xoffset, GLint yoffset,
-                              GLint x, GLint y,
-                              GLsizei width, GLsizei height );
+   void (*CopyTexSubImage2D)(struct gl_context *ctx,
+                             struct gl_texture_image *texImage,
+                             GLint xoffset, GLint yoffset,
+                             struct gl_renderbuffer *rb,
+                             GLint x, GLint y,
+                             GLsizei width, GLsizei height);
+
    /**
-    * Called by glCopyTexSubImage3D().
-    * 
-    * Drivers should use a fallback routine from texstore.c if needed.
+    * Called by glCopyTexSubImage3D() and glCopyTexImage3D().
     */
-   void (*CopyTexSubImage3D)( struct gl_context *ctx, GLenum target, GLint level,
-                              GLint xoffset, GLint yoffset, GLint zoffset,
-                              GLint x, GLint y,
-                              GLsizei width, GLsizei height );
+   void (*CopyTexSubImage3D)(struct gl_context *ctx,
+                             struct gl_texture_image *texImage,
+                             GLint xoffset, GLint yoffset, GLint zoffset,
+                             struct gl_renderbuffer *rb,
+                             GLint x, GLint y,
+                             GLsizei width, GLsizei height);
 
    /**
     * Called by glGenerateMipmap() or when GL_GENERATE_MIPMAP_SGIS is enabled.
diff --git a/mesalib/src/mesa/main/format_pack.c b/mesalib/src/mesa/main/format_pack.c
index 43677117e..0982c9af4 100644
--- a/mesalib/src/mesa/main/format_pack.c
+++ b/mesalib/src/mesa/main/format_pack.c
@@ -2532,10 +2532,10 @@ _mesa_pack_colormask(gl_format format, const GLubyte colorMask[4], void *dst)
          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;
+         maskColor[0] = colorMask[0] ? -1.0f : 0.0f;
+         maskColor[1] = colorMask[1] ? -1.0f : 0.0f;
+         maskColor[2] = colorMask[2] ? -1.0f : 0.0f;
+         maskColor[3] = colorMask[3] ? -1.0f : 0.0f;
          _mesa_pack_float_rgba_row(format, 1,
                                    (const GLfloat (*)[4]) maskColor, dst);
 
diff --git a/mesalib/src/mesa/main/format_unpack.h b/mesalib/src/mesa/main/format_unpack.h
index 0d13a2d39..c5348d30d 100644
--- a/mesalib/src/mesa/main/format_unpack.h
+++ b/mesalib/src/mesa/main/format_unpack.h
@@ -40,11 +40,6 @@ _mesa_unpack_rgba_block(gl_format format,
                         GLuint x, GLuint y, GLuint width, GLuint height);
 
 extern void
-_mesa_unpack_uint_rgba_row(gl_format format, GLuint n,
-			   const void *src, GLuint dst[][4]);
-
-
-extern void
 _mesa_unpack_float_z_row(gl_format format, GLuint n,
                          const void *src, GLfloat *dst);
 
diff --git a/mesalib/src/mesa/main/framebuffer.c b/mesalib/src/mesa/main/framebuffer.c
index 6d5e4524a..730de6206 100644
--- a/mesalib/src/mesa/main/framebuffer.c
+++ b/mesalib/src/mesa/main/framebuffer.c
@@ -222,10 +222,6 @@ _mesa_free_framebuffer_data(struct gl_framebuffer *fb)
       ASSERT(!att->Texture);
       att->Type = GL_NONE;
    }
-
-   /* unbind _Depth/_StencilBuffer to decr ref counts */
-   _mesa_reference_renderbuffer(&fb->_DepthBuffer, NULL);
-   _mesa_reference_renderbuffer(&fb->_StencilBuffer, NULL);
 }
 
 
@@ -681,8 +677,6 @@ update_color_read_buffer(struct gl_context *ctx, struct gl_framebuffer *fb)
  *    _ColorDrawBuffers
  *    _NumColorDrawBuffers
  *    _ColorReadBuffer
- *    _DepthBuffer
- *    _StencilBuffer
  *
  * If the framebuffer is user-created, make sure it's complete.
  *
diff --git a/mesalib/src/mesa/main/mtypes.h b/mesalib/src/mesa/main/mtypes.h
index dcb987116..64d8c8d3f 100644
--- a/mesalib/src/mesa/main/mtypes.h
+++ b/mesalib/src/mesa/main/mtypes.h
@@ -2673,12 +2673,6 @@ struct gl_framebuffer
    struct gl_renderbuffer *_ColorDrawBuffers[MAX_DRAW_BUFFERS];
    struct gl_renderbuffer *_ColorReadBuffer;
 
-   /** Wrappers to make combined depth/stencil buffers look like separate
-    * buffers.  Only used by swrast.  Will be removed in the future.
-    */
-   struct gl_renderbuffer *_DepthBuffer;
-   struct gl_renderbuffer *_StencilBuffer;
-
    /** Delete this framebuffer */
    void (*Delete)(struct gl_framebuffer *fb);
 };
diff --git a/mesalib/src/mesa/main/shaderobj.h b/mesalib/src/mesa/main/shaderobj.h
index 941841d43..5470b51d7 100644
--- a/mesalib/src/mesa/main/shaderobj.h
+++ b/mesalib/src/mesa/main/shaderobj.h
@@ -32,9 +32,12 @@
 #include "main/mtypes.h"
 #include "program/ir_to_mesa.h"
 
+
 #ifdef __cplusplus
 extern "C" {
 #endif
+
+
 /**
  * Internal functions
  */
diff --git a/mesalib/src/mesa/main/teximage.c b/mesalib/src/mesa/main/teximage.c
index 6dd70b96c..9475e84f5 100644
--- a/mesalib/src/mesa/main/teximage.c
+++ b/mesalib/src/mesa/main/teximage.c
@@ -2752,6 +2752,25 @@ _mesa_TexSubImage3D( GLenum target, GLint level,
 
 
 /**
+ * For glCopyTexSubImage, return the source renderbuffer to copy texel data
+ * from.  This depends on whether the texture contains color or depth values.
+ */
+static struct gl_renderbuffer *
+get_copy_tex_image_source(struct gl_context *ctx, gl_format texFormat)
+{
+   if (_mesa_get_format_bits(texFormat, GL_DEPTH_BITS) > 0) {
+      /* reading from depth/stencil buffer */
+      return ctx->ReadBuffer->Attachment[BUFFER_DEPTH].Renderbuffer;
+   }
+   else {
+      /* copying from color buffer */
+      return ctx->ReadBuffer->_ColorReadBuffer;
+   }
+}
+
+
+
+/**
  * Implement the glCopyTexImage1/2D() functions.
  */
 static void
@@ -2828,13 +2847,16 @@ copyteximage(struct gl_context *ctx, GLuint dims,
 
             if (_mesa_clip_copytexsubimage(ctx, &dstX, &dstY, &srcX, &srcY,
                                            &width, &height)) {
+               struct gl_renderbuffer *srcRb =
+                  get_copy_tex_image_source(ctx, texImage->TexFormat);
+
                if (dims == 1)
-                  ctx->Driver.CopyTexSubImage1D(ctx, target, level, dstX,
-                                                srcX, srcY, width);
+                  ctx->Driver.CopyTexSubImage1D(ctx, texImage, dstX,
+                                                srcRb, srcX, srcY, width);
                                                 
                else
-                  ctx->Driver.CopyTexSubImage2D(ctx, target, level, dstX, dstY,
-                                                srcX, srcY, width, height);
+                  ctx->Driver.CopyTexSubImage2D(ctx, texImage, dstX, dstY,
+                                                srcRb, srcX, srcY, width, height);
             }
 
             check_gen_mipmap(ctx, target, texObj, level);
@@ -2930,20 +2952,22 @@ copytexsubimage(struct gl_context *ctx, GLuint dims, GLenum target, GLint level,
 
          if (_mesa_clip_copytexsubimage(ctx, &xoffset, &yoffset, &x, &y,
                                         &width, &height)) {
+            struct gl_renderbuffer *srcRb =
+               get_copy_tex_image_source(ctx, texImage->TexFormat);
+
             switch (dims) {
             case 1:
-               ctx->Driver.CopyTexSubImage1D(ctx, target, level,
-                                             xoffset, x, y, width);
+               ctx->Driver.CopyTexSubImage1D(ctx, texImage, xoffset,
+                                             srcRb, x, y, width);
                break;
             case 2:
-               ctx->Driver.CopyTexSubImage2D(ctx, target, level,
-                                             xoffset, yoffset,
-                                             x, y, width, height);
+               ctx->Driver.CopyTexSubImage2D(ctx, texImage, xoffset, yoffset,
+                                             srcRb, x, y, width, height);
                break;
             case 3:
-               ctx->Driver.CopyTexSubImage3D(ctx, target, level,
+               ctx->Driver.CopyTexSubImage3D(ctx, texImage,
                                              xoffset, yoffset, zoffset,
-                                             x, y, width, height);
+                                             srcRb, x, y, width, height);
                break;
             default:
                _mesa_problem(ctx, "bad dims in copytexsubimage()");
diff --git a/mesalib/src/mesa/sources.mak b/mesalib/src/mesa/sources.mak
index 09cdd261b..165a6c85e 100644
--- a/mesalib/src/mesa/sources.mak
+++ b/mesalib/src/mesa/sources.mak
@@ -133,7 +133,6 @@ SWRAST_SOURCES = \
 	swrast/s_copypix.c \
 	swrast/s_context.c \
 	swrast/s_depth.c \
-	swrast/s_depthstencil.c \
 	swrast/s_drawpix.c \
 	swrast/s_feedback.c \
 	swrast/s_fog.c \
diff --git a/mesalib/src/mesa/state_tracker/st_atom_clip.c b/mesalib/src/mesa/state_tracker/st_atom_clip.c
index 1330db843..236d3cf40 100644
--- a/mesalib/src/mesa/state_tracker/st_atom_clip.c
+++ b/mesalib/src/mesa/state_tracker/st_atom_clip.c
@@ -34,7 +34,7 @@
 #include "st_context.h"
 #include "pipe/p_context.h"
 #include "st_atom.h"
-
+#include "st_program.h"
 #include "cso_cache/cso_context.h"
 
 
@@ -45,15 +45,23 @@ static void update_clip( struct st_context *st )
    struct pipe_clip_state clip;
    const struct gl_context *ctx = st->ctx;
    GLuint i;
+   bool use_eye = FALSE;
 
    memset(&clip, 0, sizeof(clip));
 
+   /* if we have a vertex shader that writes clip vertex we need to pass
+      the pre-projection transformed coordinates into the driver. */
+   if (st->vp) {
+      if (ctx->Shader.CurrentVertexProgram)
+         use_eye = TRUE;
+   }
+
    for (i = 0; i < PIPE_MAX_CLIP_PLANES; i++) {
       if (ctx->Transform.ClipPlanesEnabled & (1 << i)) {
-	 memcpy(clip.ucp[clip.nr], 
-		ctx->Transform._ClipUserPlane[i], 
-		sizeof(clip.ucp[0]));
-	 clip.nr++;
+         memcpy(clip.ucp[clip.nr],
+                use_eye ? ctx->Transform.EyeUserPlane[i] : ctx->Transform._ClipUserPlane[i],
+                sizeof(clip.ucp[0]));
+         clip.nr++;
       }
    }
 
@@ -69,7 +77,7 @@ static void update_clip( struct st_context *st )
 const struct st_tracked_state st_update_clip = {
    "st_update_clip",					/* name */
    {							/* dirty */
-      (_NEW_TRANSFORM),					/* mesa */
+      (_NEW_TRANSFORM | _NEW_PROGRAM),			/* mesa */
       0,						/* st */
    },
    update_clip						/* update */
diff --git a/mesalib/src/mesa/state_tracker/st_atom_texture.c b/mesalib/src/mesa/state_tracker/st_atom_texture.c
index 3115a2511..008e9bd10 100644
--- a/mesalib/src/mesa/state_tracker/st_atom_texture.c
+++ b/mesalib/src/mesa/state_tracker/st_atom_texture.c
@@ -267,14 +267,13 @@ update_vertex_textures(struct st_context *st)
          GLboolean retval;
          GLuint texUnit;
 
-	 texUnit = vprog->Base.SamplerUnits[su];
+         texUnit = vprog->Base.SamplerUnits[su];
 
-	 retval = update_single_texture(st, &sampler_view, texUnit);
-	 if (retval == GL_FALSE)
-	    continue;
-
-	 st->state.num_vertex_textures = su + 1;
+         retval = update_single_texture(st, &sampler_view, texUnit);
+         if (retval == GL_FALSE)
+            continue;
 
+         st->state.num_vertex_textures = su + 1;
       }
       pipe_sampler_view_reference(&st->state.sampler_vertex_views[su], sampler_view);
    }
diff --git a/mesalib/src/mesa/state_tracker/st_cb_texture.c b/mesalib/src/mesa/state_tracker/st_cb_texture.c
index 592d04b92..ad4f23c7e 100644
--- a/mesalib/src/mesa/state_tracker/st_cb_texture.c
+++ b/mesalib/src/mesa/state_tracker/st_cb_texture.c
@@ -736,7 +736,7 @@ st_GetTexImage(struct gl_context * ctx,
  * Note: srcY=0=TOP of renderbuffer
  */
 static void
-fallback_copy_texsubimage(struct gl_context *ctx, GLenum target, GLint level,
+fallback_copy_texsubimage(struct gl_context *ctx,
                           struct st_renderbuffer *strb,
                           struct st_texture_image *stImage,
                           GLenum baseFormat,
@@ -931,17 +931,12 @@ compatible_src_dst_formats(struct gl_context *ctx,
  */
 static void
 st_copy_texsubimage(struct gl_context *ctx,
-                    GLenum target, GLint level,
+                    struct gl_texture_image *texImage,
                     GLint destX, GLint destY, GLint destZ,
+                    struct gl_renderbuffer *rb,
                     GLint srcX, GLint srcY,
                     GLsizei width, GLsizei height)
 {
-   struct gl_texture_unit *texUnit =
-      &ctx->Texture.Unit[ctx->Texture.CurrentUnit];
-   struct gl_texture_object *texObj =
-      _mesa_select_tex_object(ctx, texUnit, target);
-   struct gl_texture_image *texImage =
-      _mesa_select_tex_image(ctx, texObj, target, level);
    struct st_texture_image *stImage = st_texture_image(texImage);
    const GLenum texBaseFormat = texImage->_BaseFormat;
    struct gl_framebuffer *fb = ctx->ReadBuffer;
@@ -1096,7 +1091,7 @@ st_copy_texsubimage(struct gl_context *ctx,
 
 fallback:
    /* software fallback */
-   fallback_copy_texsubimage(ctx, target, level,
+   fallback_copy_texsubimage(ctx,
                              strb, stImage, texBaseFormat,
                              destX, destY, destZ,
                              srcX, srcY, width, height);
@@ -1105,37 +1100,44 @@ fallback:
 
 
 static void
-st_CopyTexSubImage1D(struct gl_context * ctx, GLenum target, GLint level,
-                     GLint xoffset, GLint x, GLint y, GLsizei width)
+st_CopyTexSubImage1D(struct gl_context *ctx,
+                     struct gl_texture_image *texImage,
+                     GLint xoffset,
+                     struct gl_renderbuffer *rb,
+                     GLint x, GLint y, GLsizei width)
 {
    const GLint yoffset = 0, zoffset = 0;
    const GLsizei height = 1;
-   st_copy_texsubimage(ctx, target, level,
+   st_copy_texsubimage(ctx, texImage,
                        xoffset, yoffset, zoffset,  /* destX,Y,Z */
-                       x, y, width, height);  /* src X, Y, size */
+                       rb, x, y, width, height);  /* src X, Y, size */
 }
 
 
 static void
-st_CopyTexSubImage2D(struct gl_context * ctx, GLenum target, GLint level,
+st_CopyTexSubImage2D(struct gl_context *ctx,
+                     struct gl_texture_image *texImage,
                      GLint xoffset, GLint yoffset,
+                     struct gl_renderbuffer *rb,
                      GLint x, GLint y, GLsizei width, GLsizei height)
 {
    const GLint zoffset = 0;
-   st_copy_texsubimage(ctx, target, level,
+   st_copy_texsubimage(ctx, texImage,
                        xoffset, yoffset, zoffset,  /* destX,Y,Z */
-                       x, y, width, height);  /* src X, Y, size */
+                       rb, x, y, width, height);  /* src X, Y, size */
 }
 
 
 static void
-st_CopyTexSubImage3D(struct gl_context * ctx, GLenum target, GLint level,
+st_CopyTexSubImage3D(struct gl_context *ctx,
+                     struct gl_texture_image *texImage,
                      GLint xoffset, GLint yoffset, GLint zoffset,
+                     struct gl_renderbuffer *rb,
                      GLint x, GLint y, GLsizei width, GLsizei height)
 {
-   st_copy_texsubimage(ctx, target, level,
+   st_copy_texsubimage(ctx, texImage,
                        xoffset, yoffset, zoffset,  /* destX,Y,Z */
-                       x, y, width, height);  /* src X, Y, size */
+                       rb, x, y, width, height);  /* src X, Y, size */
 }
 
 
diff --git a/mesalib/src/mesa/state_tracker/st_glsl_to_tgsi.cpp b/mesalib/src/mesa/state_tracker/st_glsl_to_tgsi.cpp
index 73d956ea1..3b8e2fe37 100644
--- a/mesalib/src/mesa/state_tracker/st_glsl_to_tgsi.cpp
+++ b/mesalib/src/mesa/state_tracker/st_glsl_to_tgsi.cpp
@@ -1402,8 +1402,7 @@ glsl_to_tgsi_visitor::visit(ir_expression *ir)
       }
       break;
    case ir_unop_neg:
-      assert(result_dst.type == GLSL_TYPE_FLOAT || result_dst.type == GLSL_TYPE_INT);
-      if (result_dst.type == GLSL_TYPE_INT)
+      if (result_dst.type == GLSL_TYPE_INT || result_dst.type == GLSL_TYPE_UINT)
          emit(ir, TGSI_OPCODE_INEG, result_dst, op[0]);
       else {
          op[0].negate = ~op[0].negate;
@@ -1411,8 +1410,10 @@ glsl_to_tgsi_visitor::visit(ir_expression *ir)
       }
       break;
    case ir_unop_abs:
-      assert(result_dst.type == GLSL_TYPE_FLOAT);
-      emit(ir, TGSI_OPCODE_ABS, result_dst, op[0]);
+      if (result_dst.type == GLSL_TYPE_INT || result_dst.type == GLSL_TYPE_UINT)
+         emit(ir, TGSI_OPCODE_IABS, result_dst, op[0]);
+      else
+         emit(ir, TGSI_OPCODE_ABS, result_dst, op[0]);
       break;
    case ir_unop_sign:
       emit(ir, TGSI_OPCODE_SSG, result_dst, op[0]);
@@ -2375,7 +2376,7 @@ glsl_to_tgsi_visitor::visit(ir_constant *ir)
       gl_type = native_integers ? GL_BOOL : GL_FLOAT;
       for (i = 0; i < ir->type->vector_elements; i++) {
          if (native_integers)
-            values[i].b = ir->value.b[i];
+            values[i].u = ir->value.b[i] ? ~0 : 0;
          else
             values[i].f = ir->value.b[i];
       }
diff --git a/mesalib/src/mesa/state_tracker/st_program.c b/mesalib/src/mesa/state_tracker/st_program.c
index aceaaf8c5..8d7469dfb 100644
--- a/mesalib/src/mesa/state_tracker/st_program.c
+++ b/mesalib/src/mesa/state_tracker/st_program.c
@@ -255,6 +255,10 @@ st_prepare_vertex_program(struct gl_context *ctx,
          case VERT_RESULT_EDGE:
             assert(0);
             break;
+         case VERT_RESULT_CLIP_VERTEX:
+            stvp->output_semantic_name[slot] = TGSI_SEMANTIC_CLIPVERTEX;
+            stvp->output_semantic_index[slot] = 0;
+            break;
 
          case VERT_RESULT_TEX0:
          case VERT_RESULT_TEX1:
@@ -432,10 +436,13 @@ st_get_vp_variant(struct st_context *st,
 
 
 static unsigned
-st_translate_interp(enum glsl_interp_qualifier glsl_qual)
+st_translate_interp(enum glsl_interp_qualifier glsl_qual, bool is_color)
 {
    switch (glsl_qual) {
    case INTERP_QUALIFIER_NONE:
+      if (is_color)
+         return TGSI_INTERPOLATE_LINEAR;
+      return TGSI_INTERPOLATE_PERSPECTIVE;
    case INTERP_QUALIFIER_SMOOTH:
       return TGSI_INTERPOLATE_PERSPECTIVE;
    case INTERP_QUALIFIER_FLAT:
@@ -538,12 +545,14 @@ st_translate_fragment_program(struct st_context *st,
             case FRAG_ATTRIB_COL0:
                input_semantic_name[slot] = TGSI_SEMANTIC_COLOR;
                input_semantic_index[slot] = 0;
-               interpMode[slot] = st_translate_interp(stfp->Base.InterpQualifier[attr]);
+               interpMode[slot] = st_translate_interp(stfp->Base.InterpQualifier[attr],
+                                                      TRUE);
                break;
             case FRAG_ATTRIB_COL1:
                input_semantic_name[slot] = TGSI_SEMANTIC_COLOR;
                input_semantic_index[slot] = 1;
-               interpMode[slot] = st_translate_interp(stfp->Base.InterpQualifier[attr]);
+               interpMode[slot] = st_translate_interp(stfp->Base.InterpQualifier[attr],
+                                                      TRUE);
                break;
             case FRAG_ATTRIB_FOGC:
                input_semantic_name[slot] = TGSI_SEMANTIC_FOG;
@@ -600,7 +609,8 @@ st_translate_fragment_program(struct st_context *st,
                if (attr == FRAG_ATTRIB_PNTC)
                   interpMode[slot] = TGSI_INTERPOLATE_LINEAR;
                else
-                  interpMode[slot] = st_translate_interp(stfp->Base.InterpQualifier[attr]);
+                  interpMode[slot] = st_translate_interp(stfp->Base.InterpQualifier[attr],
+                                                         FALSE);
                break;
             }
          }
diff --git a/mesalib/src/mesa/swrast/s_context.c b/mesalib/src/mesa/swrast/s_context.c
index 6850a09f3..94b7fe34d 100644
--- a/mesalib/src/mesa/swrast/s_context.c
+++ b/mesalib/src/mesa/swrast/s_context.c
@@ -36,7 +36,6 @@
 #include "swrast.h"
 #include "s_blend.h"
 #include "s_context.h"
-#include "s_depthstencil.h"
 #include "s_lines.h"
 #include "s_points.h"
 #include "s_span.h"
@@ -493,7 +492,7 @@ static void
 _swrast_update_active_attribs(struct gl_context *ctx)
 {
    SWcontext *swrast = SWRAST_CONTEXT(ctx);
-   GLuint attribsMask;
+   GLbitfield64 attribsMask;
 
    /*
     * Compute _ActiveAttribsMask = which fragment attributes are needed.
@@ -546,25 +545,6 @@ _swrast_update_active_attribs(struct gl_context *ctx)
 }
 
 
-/**
- * Update the depth/stencil renderbuffers, if needed.
- */
-static void
-_swrast_update_depth_stencil(struct gl_context *ctx)
-{
-   struct gl_framebuffer *drawFb = ctx->DrawBuffer;
-   struct gl_framebuffer *readFb = ctx->ReadBuffer;
-
-   _swrast_update_depth_buffer(ctx, drawFb);
-   _swrast_update_stencil_buffer(ctx, drawFb);
-
-   if (readFb != drawFb) {
-      _swrast_update_depth_buffer(ctx, readFb);
-      _swrast_update_stencil_buffer(ctx, readFb);
-   }
-}
-
-
 void
 _swrast_validate_derived( struct gl_context *ctx )
 {
@@ -609,9 +589,6 @@ _swrast_validate_derived( struct gl_context *ctx )
                               _NEW_TEXTURE))
          _swrast_update_specular_vertex_add(ctx);
 
-      if (swrast->NewState & _NEW_BUFFERS)
-         _swrast_update_depth_stencil(ctx);
-
       swrast->NewState = 0;
       swrast->StateChanges = 0;
       swrast->InvalidateState = _swrast_invalidate_state;
@@ -741,9 +718,9 @@ _swrast_CreateContext( struct gl_context *ctx )
    GLuint i;
    SWcontext *swrast = (SWcontext *)CALLOC(sizeof(SWcontext));
 #ifdef _OPENMP
-   const GLint maxThreads = omp_get_max_threads();
+   const GLuint maxThreads = omp_get_max_threads();
 #else
-   const GLint maxThreads = 1;
+   const GLuint maxThreads = 1;
 #endif
 
    if (SWRAST_DEBUG) {
diff --git a/mesalib/src/mesa/swrast/s_context.h b/mesalib/src/mesa/swrast/s_context.h
index af9e49ecb..0a383aa3b 100644
--- a/mesalib/src/mesa/swrast/s_context.h
+++ b/mesalib/src/mesa/swrast/s_context.h
@@ -196,8 +196,8 @@ typedef struct
 
    /** List/array of the fragment attributes to interpolate */
    GLuint _ActiveAttribs[FRAG_ATTRIB_MAX];
-   /** Same info, but as a bitmask */
-   GLbitfield _ActiveAttribMask;
+   /** Same info, but as a bitmask of FRAG_BIT_x bits */
+   GLbitfield64 _ActiveAttribMask;
    /** Number of fragment attributes to interpolate */
    GLuint _NumActiveAttribs;
    /** Indicates how each attrib is to be interpolated (lines/tris) */
diff --git a/mesalib/src/mesa/swrast/s_depth.c b/mesalib/src/mesa/swrast/s_depth.c
index 53f21cb69..42724c72b 100644
--- a/mesalib/src/mesa/swrast/s_depth.c
+++ b/mesalib/src/mesa/swrast/s_depth.c
@@ -205,6 +205,7 @@ _swrast_depth_clamp_span( struct gl_context *ctx, SWspan *span )
 
 /**
  * Get array of 32-bit z values from the depth buffer.  With clipping.
+ * Note: the returned values are always in the range [0, 2^32-1].
  */
 static void
 get_z32_values(struct gl_context *ctx, struct gl_renderbuffer *rb,
@@ -235,6 +236,11 @@ get_z32_values(struct gl_context *ctx, struct gl_renderbuffer *rb,
    }
 }
 
+
+/**
+ * Put an array of 32-bit z values into the depth buffer.
+ * Note: the z values are always in the range [0, 2^32-1].
+ */
 static void
 put_z32_values(struct gl_context *ctx, struct gl_renderbuffer *rb,
                GLuint count, const GLint x[], const GLint y[],
@@ -284,8 +290,8 @@ _swrast_depth_test_span(struct gl_context *ctx, SWspan *span)
    void *zBufferVals;
    GLuint *zBufferTemp = NULL;
    GLuint passed;
+   GLuint zBits = _mesa_get_format_bits(rb->Format, GL_DEPTH_BITS);
    GLboolean ztest16 = GL_FALSE;
-   GLboolean ztest24 = _mesa_get_format_bits(rb->Format, GL_DEPTH_BITS) == 24;
 
    if (rb->Format == MESA_FORMAT_Z16 && !(span->arrayMask & SPAN_XY)) {
       /* directly read/write row of 16-bit Z values */
@@ -310,7 +316,7 @@ _swrast_depth_test_span(struct gl_context *ctx, SWspan *span)
          _mesa_unpack_uint_z_row(rb->Format, count, zStart, zBufferTemp);
       }
 
-      if (ztest24) {
+      if (zBits == 24) {
          GLuint i;
          /* Convert depth buffer values from 32 to 24 bits to match the
           * fragment Z values generated by rasterization.
@@ -319,6 +325,16 @@ _swrast_depth_test_span(struct gl_context *ctx, SWspan *span)
             zBufferTemp[i] >>= 8;
          }
       }
+      else if (zBits == 16) {
+         GLuint i;
+         /* Convert depth buffer values from 32 to 16 bits */
+         for (i = 0; i < count; i++) {
+            zBufferTemp[i] >>= 16;
+         }
+      }
+      else {
+         assert(zBits == 32);
+      }
 
       zBufferVals = zBufferTemp;
    }
@@ -332,16 +348,22 @@ _swrast_depth_test_span(struct gl_context *ctx, SWspan *span)
    if (zBufferTemp) {
       /* need to write temp Z values back into the buffer */
 
-      if (ztest24) {
+      /* 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.
+       */
+      if (zBits == 24) {
          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);
          }
       }
+      else if (zBits == 16) {
+         GLuint i;
+         for (i = 0; i < count; i++) {
+            zBufferTemp[i] = zBufferTemp[i] << 16;
+         }
+      }
 
       if (span->arrayMask & SPAN_XY) {
          /* random locations */
diff --git a/mesalib/src/mesa/swrast/s_depthstencil.c b/mesalib/src/mesa/swrast/s_depthstencil.c
deleted file mode 100644
index e48177ec1..000000000
--- a/mesalib/src/mesa/swrast/s_depthstencil.c
+++ /dev/null
@@ -1,791 +0,0 @@
-/*
- * Mesa 3-D graphics library
- * Version:  6.5
- *
- * 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.
- */
-
-#include "main/glheader.h"
-#include "main/imports.h"
-#include "main/context.h"
-#include "main/formats.h"
-#include "main/mtypes.h"
-#include "main/renderbuffer.h"
-#include "swrast/s_depthstencil.h"
-
-
-/**
- * Adaptor/wrappers for GL_DEPTH_STENCIL renderbuffers.
- *
- * The problem with a GL_DEPTH_STENCIL renderbuffer is that sometimes we
- * want to treat it as a stencil buffer, other times we want to treat it
- * as a depth/z buffer and still other times when we want to treat it as
- * a combined Z+stencil buffer!  That implies we need three different sets
- * of Get/Put functions.
- *
- * We solve this by wrapping the Z24_S8 or S8_Z24 renderbuffer with depth and
- * stencil adaptors, each with the right kind of depth/stencil Get/Put functions.
- */
-
-
-static void *
-nop_get_pointer(struct gl_context *ctx, struct gl_renderbuffer *rb, GLint x, GLint y)
-{
-   (void) ctx;
-   (void) rb;
-   (void) x;
-   (void) y;
-   return NULL;
-}
-
-
-/**
- * Delete a depth or stencil wrapper renderbuffer.
- */
-static void
-delete_wrapper(struct gl_renderbuffer *rb)
-{
-   ASSERT(rb->Format == MESA_FORMAT_S8 ||
-          rb->Format == MESA_FORMAT_X8_Z24 ||
-          rb->Format == MESA_FORMAT_Z32_FLOAT);
-   _mesa_reference_renderbuffer(&rb->Wrapped, NULL);
-   free(rb);
-}
-
-
-/**
- * Realloc storage for wrapper.
- */
-static GLboolean
-alloc_wrapper_storage(struct gl_context *ctx, struct gl_renderbuffer *rb,
-                      GLenum internalFormat, GLuint width, GLuint height)
-{
-   /* just pass this on to the wrapped renderbuffer */
-   struct gl_renderbuffer *dsrb = rb->Wrapped;
-   GLboolean retVal;
-
-   (void) internalFormat;
-
-   ASSERT(dsrb->Format == MESA_FORMAT_Z24_S8 ||
-          dsrb->Format == MESA_FORMAT_Z24_X8 ||
-          dsrb->Format == MESA_FORMAT_S8_Z24 ||
-          dsrb->Format == MESA_FORMAT_X8_Z24);
-
-   retVal = dsrb->AllocStorage(ctx, dsrb, dsrb->InternalFormat, width, height);
-   if (retVal) {
-      rb->Width = width;
-      rb->Height = height;
-      rb->RowStride = dsrb->RowStride;
-   }
-   return retVal;
-}
-
-
-
-
-/*======================================================================
- * Depth wrapper around depth/stencil renderbuffer
- */
-
-static void
-get_row_z24(struct gl_context *ctx, struct gl_renderbuffer *z24rb, GLuint count,
-            GLint x, GLint y, void *values)
-{
-   struct gl_renderbuffer *dsrb = z24rb->Wrapped;
-   GLuint temp[MAX_WIDTH], i;
-   GLuint *dst = (GLuint *) values;
-   const GLuint *src = (const GLuint *) dsrb->GetPointer(ctx, dsrb, x, y);
-   ASSERT(z24rb->DataType == GL_UNSIGNED_INT);
-   ASSERT(dsrb->DataType == GL_UNSIGNED_INT_24_8_EXT);
-   if (!src) {
-      dsrb->GetRow(ctx, dsrb, count, x, y, temp);
-      src = temp;
-   }
-   if (dsrb->Format == MESA_FORMAT_Z24_S8) {
-      for (i = 0; i < count; i++) {
-         dst[i] = src[i] >> 8;
-      }
-   }
-   else {
-      assert(dsrb->Format == MESA_FORMAT_S8_Z24);
-      for (i = 0; i < count; i++) {
-         dst[i] = src[i] & 0xffffff;
-      }
-   }
-}
-
-static void
-get_values_z24(struct gl_context *ctx, struct gl_renderbuffer *z24rb, GLuint count,
-               const GLint x[], const GLint y[], void *values)
-{
-   struct gl_renderbuffer *dsrb = z24rb->Wrapped;
-   GLuint temp[MAX_WIDTH], i;
-   GLuint *dst = (GLuint *) values;
-   ASSERT(z24rb->DataType == GL_UNSIGNED_INT);
-   ASSERT(dsrb->DataType == GL_UNSIGNED_INT_24_8_EXT);
-   ASSERT(count <= MAX_WIDTH);
-   /* don't bother trying direct access */
-   dsrb->GetValues(ctx, dsrb, count, x, y, temp);
-   if (dsrb->Format == MESA_FORMAT_Z24_S8) {
-      for (i = 0; i < count; i++) {
-         dst[i] = temp[i] >> 8;
-      }
-   }
-   else {
-      assert(dsrb->Format == MESA_FORMAT_S8_Z24);
-      for (i = 0; i < count; i++) {
-         dst[i] = temp[i] & 0xffffff;
-      }
-   }
-}
-
-static void
-put_row_z24(struct gl_context *ctx, struct gl_renderbuffer *z24rb, GLuint count,
-            GLint x, GLint y, const void *values, const GLubyte *mask)
-{
-   struct gl_renderbuffer *dsrb = z24rb->Wrapped;
-   const GLuint *src = (const GLuint *) values;
-   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) {
-         for (i = 0; i < count; i++) {
-            if (!mask || mask[i]) {
-               dst[i] = (src[i] << 8) | (dst[i] & 0xff);
-            }
-         }
-      }
-      else {
-         assert(dsrb->Format == MESA_FORMAT_S8_Z24);
-         for (i = 0; i < count; i++) {
-            if (!mask || mask[i]) {
-               dst[i] = (src[i] & 0xffffff) | (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) {
-         for (i = 0; i < count; i++) {
-            if (!mask || mask[i]) {
-               temp[i] = (src[i] << 8) | (temp[i] & 0xff);
-            }
-         }
-      }
-      else {
-         assert(dsrb->Format == MESA_FORMAT_S8_Z24);
-         for (i = 0; i < count; i++) {
-            if (!mask || mask[i]) {
-               temp[i] = (src[i] & 0xffffff) | (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,
-               const GLint x[], const GLint y[],
-               const void *values, const GLubyte *mask)
-{
-   struct gl_renderbuffer *dsrb = z24rb->Wrapped;
-   const GLuint *src = (const GLuint *) values;
-   ASSERT(z24rb->DataType == GL_UNSIGNED_INT);
-   ASSERT(dsrb->DataType == GL_UNSIGNED_INT_24_8_EXT);
-   if (dsrb->GetPointer(ctx, dsrb, 0, 0)) {
-      /* direct access */
-      GLuint i;
-      if (dsrb->Format == MESA_FORMAT_Z24_S8) {
-         for (i = 0; i < count; i++) {
-            if (!mask || mask[i]) {
-               GLuint *dst = (GLuint *) dsrb->GetPointer(ctx, dsrb, x[i], y[i]);
-               *dst = (src[i] << 8) | (*dst & 0xff);
-            }
-         }
-      }
-      else {
-         assert(dsrb->Format == MESA_FORMAT_S8_Z24);
-         for (i = 0; i < count; i++) {
-            if (!mask || mask[i]) {
-               GLuint *dst = (GLuint *) dsrb->GetPointer(ctx, dsrb, x[i], y[i]);
-               *dst = (src[i] & 0xffffff) | (*dst & 0xff000000);
-            }
-         }
-      }
-   }
-   else {
-      /* get, modify, put */
-      GLuint temp[MAX_WIDTH], i;
-      dsrb->GetValues(ctx, dsrb, count, x, y, temp);
-      if (dsrb->Format == MESA_FORMAT_Z24_S8) {
-         for (i = 0; i < count; i++) {
-            if (!mask || mask[i]) {
-               temp[i] = (src[i] << 8) | (temp[i] & 0xff);
-            }
-         }
-      }
-      else {
-         assert(dsrb->Format == MESA_FORMAT_S8_Z24);
-         for (i = 0; i < count; i++) {
-            if (!mask || mask[i]) {
-               temp[i] = (src[i] & 0xffffff) | (temp[i] & 0xff000000);
-            }
-         }
-      }
-      dsrb->PutValues(ctx, dsrb, count, x, y, temp, mask);
-   }
-}
-
-
-/**
- * Wrap the given GL_DEPTH_STENCIL renderbuffer so that it acts like
- * a depth renderbuffer.
- * \return new depth renderbuffer
- */
-static struct gl_renderbuffer *
-new_z24_renderbuffer_wrapper(struct gl_context *ctx,
-                             struct gl_renderbuffer *dsrb)
-{
-   struct gl_renderbuffer *z24rb;
-
-   ASSERT(dsrb->Format == MESA_FORMAT_Z24_S8 ||
-          dsrb->Format == MESA_FORMAT_Z24_X8 ||
-          dsrb->Format == MESA_FORMAT_S8_Z24 ||
-          dsrb->Format == MESA_FORMAT_X8_Z24);
-   ASSERT(dsrb->DataType == GL_UNSIGNED_INT_24_8_EXT);
-
-   z24rb = ctx->Driver.NewRenderbuffer(ctx, 0);
-   if (!z24rb)
-      return NULL;
-
-   /* NOTE: need to do manual refcounting here */
-   z24rb->Wrapped = dsrb;
-   dsrb->RefCount++;
-
-   z24rb->Name = dsrb->Name;
-   z24rb->RefCount = 0;
-   z24rb->Width = dsrb->Width;
-   z24rb->Height = dsrb->Height;
-   z24rb->RowStride = dsrb->RowStride;
-   z24rb->InternalFormat = GL_DEPTH_COMPONENT24;
-   z24rb->Format = MESA_FORMAT_X8_Z24;
-   z24rb->_BaseFormat = GL_DEPTH_COMPONENT;
-   z24rb->DataType = GL_UNSIGNED_INT;
-   z24rb->Data = NULL;
-   z24rb->Delete = delete_wrapper;
-   z24rb->AllocStorage = alloc_wrapper_storage;
-   z24rb->GetPointer = nop_get_pointer;
-   z24rb->GetRow = get_row_z24;
-   z24rb->GetValues = get_values_z24;
-   z24rb->PutRow = put_row_z24;
-   z24rb->PutValues = put_values_z24;
-
-   return z24rb;
-}
-
-
-static void
-get_row_z32f(struct gl_context *ctx, struct gl_renderbuffer *z32frb, GLuint count,
-             GLint x, GLint y, void *values)
-{
-   struct gl_renderbuffer *dsrb = z32frb->Wrapped;
-   GLfloat temp[MAX_WIDTH*2];
-   GLfloat *dst = (GLfloat *) values;
-   const GLfloat *src = (const GLfloat *) dsrb->GetPointer(ctx, dsrb, x, y);
-   GLuint i;
-   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 (!src) {
-      dsrb->GetRow(ctx, dsrb, count, x, y, temp);
-      src = temp;
-   }
-   for (i = 0; i < count; i++) {
-      dst[i] = src[i*2];
-   }
-}
-
-static void
-get_values_z32f(struct gl_context *ctx, struct gl_renderbuffer *z32frb, GLuint count,
-                const GLint x[], const GLint y[], void *values)
-{
-   struct gl_renderbuffer *dsrb = z32frb->Wrapped;
-   GLfloat temp[MAX_WIDTH*2];
-   GLfloat *dst = (GLfloat *) values;
-   GLuint i;
-   ASSERT(z32frb->DataType == GL_FLOAT);
-   ASSERT(dsrb->DataType == GL_FLOAT_32_UNSIGNED_INT_24_8_REV);
-   ASSERT(dsrb->Format == MESA_FORMAT_Z32_FLOAT_X24S8);
-   ASSERT(count <= MAX_WIDTH);
-   /* don't bother trying direct access */
-   dsrb->GetValues(ctx, dsrb, count, x, y, temp);
-   for (i = 0; i < count; i++) {
-      dst[i] = temp[i*2];
-   }
-}
-
-static void
-put_row_z32f(struct gl_context *ctx, struct gl_renderbuffer *z32frb, GLuint count,
-             GLint x, GLint y, const void *values, const GLubyte *mask)
-{
-   struct gl_renderbuffer *dsrb = z32frb->Wrapped;
-   const GLfloat *src = (const GLfloat *) values;
-   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;
-      for (i = 0; i < count; i++) {
-         if (!mask || mask[i]) {
-            dst[i*2] = src[i];
-         }
-      }
-   }
-   else {
-      /* get, modify, put */
-      GLfloat temp[MAX_WIDTH*2];
-      GLuint i;
-      dsrb->GetRow(ctx, dsrb, count, x, y, temp);
-      for (i = 0; i < count; i++) {
-         if (!mask || mask[i]) {
-            temp[i*2] = src[i];
-         }
-      }
-      dsrb->PutRow(ctx, dsrb, count, x, y, temp, mask);
-   }
-}
-
-
-static void
-put_values_z32f(struct gl_context *ctx, struct gl_renderbuffer *z32frb, GLuint count,
-                const GLint x[], const GLint y[],
-                const void *values, const GLubyte *mask)
-{
-   struct gl_renderbuffer *dsrb = z32frb->Wrapped;
-   const GLfloat *src = (const GLfloat *) values;
-   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 (dsrb->GetPointer(ctx, dsrb, 0, 0)) {
-      /* direct access */
-      GLuint i;
-      for (i = 0; i < count; i++) {
-         if (!mask || mask[i]) {
-            GLfloat *dst = (GLfloat *) dsrb->GetPointer(ctx, dsrb, x[i], y[i]);
-            *dst = src[i];
-         }
-      }
-   }
-   else {
-      /* get, modify, put */
-      GLfloat temp[MAX_WIDTH*2];
-      GLuint i;
-      dsrb->GetValues(ctx, dsrb, count, x, y, temp);
-      for (i = 0; i < count; i++) {
-         if (!mask || mask[i]) {
-            temp[i*2] = src[i];
-         }
-      }
-      dsrb->PutValues(ctx, dsrb, count, x, y, temp, mask);
-   }
-}
-
-
-/**
- * Wrap the given GL_DEPTH_STENCIL renderbuffer so that it acts like
- * a depth renderbuffer.
- * \return new depth renderbuffer
- */
-static struct gl_renderbuffer *
-new_z32f_renderbuffer_wrapper(struct gl_context *ctx,
-                              struct gl_renderbuffer *dsrb)
-{
-   struct gl_renderbuffer *z32frb;
-
-   ASSERT(dsrb->Format == MESA_FORMAT_Z32_FLOAT_X24S8);
-   ASSERT(dsrb->DataType == GL_FLOAT_32_UNSIGNED_INT_24_8_REV);
-
-   z32frb = ctx->Driver.NewRenderbuffer(ctx, 0);
-   if (!z32frb)
-      return NULL;
-
-   /* NOTE: need to do manual refcounting here */
-   z32frb->Wrapped = dsrb;
-   dsrb->RefCount++;
-
-   z32frb->Name = dsrb->Name;
-   z32frb->RefCount = 0;
-   z32frb->Width = dsrb->Width;
-   z32frb->Height = dsrb->Height;
-   z32frb->RowStride = dsrb->RowStride;
-   z32frb->InternalFormat = GL_DEPTH_COMPONENT32F;
-   z32frb->Format = MESA_FORMAT_Z32_FLOAT;
-   z32frb->_BaseFormat = GL_DEPTH_COMPONENT;
-   z32frb->DataType = GL_FLOAT;
-   z32frb->Data = NULL;
-   z32frb->Delete = delete_wrapper;
-   z32frb->AllocStorage = alloc_wrapper_storage;
-   z32frb->GetPointer = nop_get_pointer;
-   z32frb->GetRow = get_row_z32f;
-   z32frb->GetValues = get_values_z32f;
-   z32frb->PutRow = put_row_z32f;
-   z32frb->PutValues = put_values_z32f;
-
-   return z32frb;
-}
-
-
-/*======================================================================
- * Stencil wrapper around depth/stencil renderbuffer
- */
-
-static void
-get_row_s8(struct gl_context *ctx, struct gl_renderbuffer *s8rb, GLuint count,
-           GLint x, GLint y, void *values)
-{
-   struct gl_renderbuffer *dsrb = s8rb->Wrapped;
-   GLuint temp[MAX_WIDTH*2], i;
-   GLubyte *dst = (GLubyte *) values;
-   const GLuint *src = (const 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 (!src) {
-      dsrb->GetRow(ctx, dsrb, count, x, y, temp);
-      src = temp;
-   }
-   if (dsrb->Format == MESA_FORMAT_Z32_FLOAT_X24S8) {
-      for (i = 0; i < count; i++) {
-         dst[i] = src[i*2+1] & 0xff;
-      }
-   }
-   else if (dsrb->Format == MESA_FORMAT_Z24_S8) {
-      for (i = 0; i < count; i++) {
-         dst[i] = src[i] & 0xff;
-      }
-   }
-   else {
-      assert(dsrb->Format == MESA_FORMAT_S8_Z24);
-      for (i = 0; i < count; i++) {
-         dst[i] = src[i] >> 24;
-      }
-   }
-}
-
-static void
-get_values_s8(struct gl_context *ctx, struct gl_renderbuffer *s8rb, GLuint count,
-              const GLint x[], const GLint y[], void *values)
-{
-   struct gl_renderbuffer *dsrb = s8rb->Wrapped;
-   GLuint temp[MAX_WIDTH*2], i;
-   GLubyte *dst = (GLubyte *) values;
-   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);
-   ASSERT(count <= MAX_WIDTH);
-   /* don't bother trying direct access */
-   dsrb->GetValues(ctx, dsrb, count, x, y, temp);
-   if (dsrb->Format == MESA_FORMAT_Z32_FLOAT_X24S8) {
-      for (i = 0; i < count; i++) {
-         dst[i] = temp[i*2+1] & 0xff;
-      }
-   }
-   else if (dsrb->Format == MESA_FORMAT_Z24_S8) {
-      for (i = 0; i < count; i++) {
-         dst[i] = temp[i] & 0xff;
-      }
-   }
-   else {
-      assert(dsrb->Format == MESA_FORMAT_S8_Z24);
-      for (i = 0; i < count; i++) {
-         dst[i] = temp[i] >> 24;
-      }
-   }
-}
-
-static void
-put_row_s8(struct gl_context *ctx, struct gl_renderbuffer *s8rb, GLuint count,
-           GLint x, GLint y, const void *values, const GLubyte *mask)
-{
-   struct gl_renderbuffer *dsrb = s8rb->Wrapped;
-   const GLubyte *src = (const GLubyte *) values;
-   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] = src[i];
-            }
-         }
-      }
-      else if (dsrb->Format == MESA_FORMAT_Z24_S8) {
-         for (i = 0; i < count; i++) {
-            if (!mask || mask[i]) {
-               dst[i] = (dst[i] & 0xffffff00) | src[i];
-            }
-         }
-      }
-      else {
-         assert(dsrb->Format == MESA_FORMAT_S8_Z24);
-         for (i = 0; i < count; i++) {
-            if (!mask || mask[i]) {
-               dst[i] = (dst[i] & 0xffffff) | (src[i] << 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] = src[i];
-            }
-         }
-      }
-      else if (dsrb->Format == MESA_FORMAT_Z24_S8) {
-         for (i = 0; i < count; i++) {
-            if (!mask || mask[i]) {
-               temp[i] = (temp[i] & 0xffffff00) | src[i];
-            }
-         }
-      }
-      else {
-         assert(dsrb->Format == MESA_FORMAT_S8_Z24);
-         for (i = 0; i < count; i++) {
-            if (!mask || mask[i]) {
-               temp[i] = (temp[i] & 0xffffff) | (src[i] << 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,
-              const GLint x[], const GLint y[],
-              const void *values, const GLubyte *mask)
-{
-   struct gl_renderbuffer *dsrb = s8rb->Wrapped;
-   const GLubyte *src = (const GLubyte *) values;
-   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 (dsrb->GetPointer(ctx, dsrb, 0, 0)) {
-      /* direct access */
-      GLuint i;
-      if (dsrb->Format == MESA_FORMAT_Z32_FLOAT_X24S8) {
-         for (i = 0; i < count; i++) {
-            if (!mask || mask[i]) {
-               GLuint *dst = (GLuint *) dsrb->GetPointer(ctx, dsrb, x[i], y[i]);
-               dst[1] = src[i];
-            }
-         }
-      }
-      else if (dsrb->Format == MESA_FORMAT_Z24_S8) {
-         for (i = 0; i < count; i++) {
-            if (!mask || mask[i]) {
-               GLuint *dst = (GLuint *) dsrb->GetPointer(ctx, dsrb, x[i], y[i]);
-               *dst = (*dst & 0xffffff00) | src[i];
-            }
-         }
-      }
-      else {
-         assert(dsrb->Format == MESA_FORMAT_S8_Z24);
-         for (i = 0; i < count; i++) {
-            if (!mask || mask[i]) {
-               GLuint *dst = (GLuint *) dsrb->GetPointer(ctx, dsrb, x[i], y[i]);
-               *dst = (*dst & 0xffffff) | (src[i] << 24);
-            }
-         }
-      }
-   }
-   else {
-      /* get, modify, put */
-      GLuint temp[MAX_WIDTH*2], i;
-      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] = src[i];
-            }
-         }
-      }
-      else if (dsrb->Format == MESA_FORMAT_Z24_S8) {
-         for (i = 0; i < count; i++) {
-            if (!mask || mask[i]) {
-               temp[i] = (temp[i] & 0xffffff00) | src[i];
-            }
-         }
-      }
-      else {
-         assert(dsrb->Format == MESA_FORMAT_S8_Z24);
-         for (i = 0; i < count; i++) {
-            if (!mask || mask[i]) {
-               temp[i] = (temp[i] & 0xffffff) | (src[i] << 24);
-            }
-         }
-      }
-      dsrb->PutValues(ctx, dsrb, count, x, y, temp, mask);
-   }
-}
-
-
-/**
- * Wrap the given GL_DEPTH_STENCIL renderbuffer so that it acts like
- * a stencil renderbuffer.
- * \return new stencil renderbuffer
- */
-static struct gl_renderbuffer *
-new_s8_renderbuffer_wrapper(struct gl_context *ctx, struct gl_renderbuffer *dsrb)
-{
-   struct gl_renderbuffer *s8rb;
-
-   ASSERT(dsrb->Format == MESA_FORMAT_Z24_S8 ||
-          dsrb->Format == MESA_FORMAT_S8_Z24 ||
-          dsrb->Format == MESA_FORMAT_Z32_FLOAT_X24S8);
-   ASSERT(dsrb->DataType == GL_UNSIGNED_INT_24_8_EXT ||
-          dsrb->DataType == GL_FLOAT_32_UNSIGNED_INT_24_8_REV);
-
-   s8rb = ctx->Driver.NewRenderbuffer(ctx, 0);
-   if (!s8rb)
-      return NULL;
-
-   /* NOTE: need to do manual refcounting here */
-   s8rb->Wrapped = dsrb;
-   dsrb->RefCount++;
-
-   s8rb->Name = dsrb->Name;
-   s8rb->RefCount = 0;
-   s8rb->Width = dsrb->Width;
-   s8rb->Height = dsrb->Height;
-   s8rb->RowStride = dsrb->RowStride;
-   s8rb->InternalFormat = GL_STENCIL_INDEX8_EXT;
-   s8rb->Format = MESA_FORMAT_S8;
-   s8rb->_BaseFormat = GL_STENCIL_INDEX;
-   s8rb->DataType = GL_UNSIGNED_BYTE;
-   s8rb->Data = NULL;
-   s8rb->Delete = delete_wrapper;
-   s8rb->AllocStorage = alloc_wrapper_storage;
-   s8rb->GetPointer = nop_get_pointer;
-   s8rb->GetRow = get_row_s8;
-   s8rb->GetValues = get_values_s8;
-   s8rb->PutRow = put_row_s8;
-   s8rb->PutValues = put_values_s8;
-
-   return s8rb;
-}
-
-
-/**
- * Update the framebuffer's _DepthBuffer field using the renderbuffer
- * found at the given attachment index.
- *
- * If that attachment points to a combined GL_DEPTH_STENCIL renderbuffer,
- * create and install a depth wrapper/adaptor.
- *
- * \param fb  the framebuffer whose _DepthBuffer field to update
- */
-void
-_swrast_update_depth_buffer(struct gl_context *ctx, struct gl_framebuffer *fb)
-{
-   struct gl_renderbuffer *depthRb =
-      fb->Attachment[BUFFER_DEPTH].Renderbuffer;
-
-   if (depthRb && _mesa_is_format_packed_depth_stencil(depthRb->Format)) {
-      /* The attached depth buffer is a GL_DEPTH_STENCIL renderbuffer */
-      if (!fb->_DepthBuffer
-          || fb->_DepthBuffer->Wrapped != depthRb
-          || _mesa_get_format_base_format(fb->_DepthBuffer->Format) != GL_DEPTH_COMPONENT) {
-         /* need to update wrapper */
-         struct gl_renderbuffer *wrapper;
-
-         if (depthRb->Format == MESA_FORMAT_Z32_FLOAT_X24S8) {
-            wrapper = new_z32f_renderbuffer_wrapper(ctx, depthRb);
-         }
-         else {
-            wrapper = new_z24_renderbuffer_wrapper(ctx, depthRb);
-         }
-         _mesa_reference_renderbuffer(&fb->_DepthBuffer, wrapper);
-      }
-      ASSERT(fb->_DepthBuffer->Wrapped == depthRb);
-      fb->_DepthBuffer->Width = depthRb->Width;
-      fb->_DepthBuffer->Height = depthRb->Height;
-   }
-   else {
-      /* depthRb may be null */
-      _mesa_reference_renderbuffer(&fb->_DepthBuffer, depthRb);
-   }
-}
-
-
-/**
- * Update the framebuffer's _StencilBuffer field using the renderbuffer
- * found at the given attachment index.
- *
- * If that attachment points to a combined GL_DEPTH_STENCIL renderbuffer,
- * create and install a stencil wrapper/adaptor.
- *
- * \param fb  the framebuffer whose _StencilBuffer field to update
- */
-void
-_swrast_update_stencil_buffer(struct gl_context *ctx, struct gl_framebuffer *fb)
-{
-   struct gl_renderbuffer *stencilRb =
-      fb->Attachment[BUFFER_STENCIL].Renderbuffer;
-
-   if (stencilRb && _mesa_is_format_packed_depth_stencil(stencilRb->Format)) {
-      /* The attached stencil buffer is a GL_DEPTH_STENCIL renderbuffer */
-      if (!fb->_StencilBuffer
-          || fb->_StencilBuffer->Wrapped != stencilRb
-          || _mesa_get_format_base_format(fb->_StencilBuffer->Format) != GL_STENCIL_INDEX) {
-         /* need to update wrapper */
-         struct gl_renderbuffer *wrapper
-            = new_s8_renderbuffer_wrapper(ctx, stencilRb);
-         _mesa_reference_renderbuffer(&fb->_StencilBuffer, wrapper);
-      }
-      ASSERT(fb->_StencilBuffer->Wrapped == stencilRb);
-      fb->_StencilBuffer->Width = stencilRb->Width;
-      fb->_StencilBuffer->Height = stencilRb->Height;
-   }
-   else {
-      /* stencilRb may be null */
-      _mesa_reference_renderbuffer(&fb->_StencilBuffer, stencilRb);
-   }
-}
diff --git a/mesalib/src/mesa/swrast/s_depthstencil.h b/mesalib/src/mesa/swrast/s_depthstencil.h
deleted file mode 100644
index 0bf624a7b..000000000
--- a/mesalib/src/mesa/swrast/s_depthstencil.h
+++ /dev/null
@@ -1,39 +0,0 @@
-/*
- * Mesa 3-D graphics library
- * Version:  6.5
- *
- * 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.
- */
-
-
-#ifndef S_DEPTHSTENCIL_H
-#define S_DEPTHSTENCIL_H
-
-struct gl_context;
-struct gl_framebuffer;
-
-void
-_swrast_update_depth_buffer(struct gl_context *ctx, struct gl_framebuffer *fb);
-
-void
-_swrast_update_stencil_buffer(struct gl_context *ctx, struct gl_framebuffer *fb);
-
-
-#endif /* S_DEPTHSTENCIL_H */
diff --git a/mesalib/src/mesa/swrast/s_span.c b/mesalib/src/mesa/swrast/s_span.c
index 8f02eeae7..689fe34ae 100644
--- a/mesalib/src/mesa/swrast/s_span.c
+++ b/mesalib/src/mesa/swrast/s_span.c
@@ -163,7 +163,8 @@ _swrast_span_default_attribs(struct gl_context *ctx, SWspan *span)
  * should have computed attrStart/Step values for FRAG_ATTRIB_WPOS[3]!
  */
 static inline void
-interpolate_active_attribs(struct gl_context *ctx, SWspan *span, GLbitfield attrMask)
+interpolate_active_attribs(struct gl_context *ctx, SWspan *span,
+                           GLbitfield64 attrMask)
 {
    const SWcontext *swrast = SWRAST_CONTEXT(ctx);
 
@@ -1038,7 +1039,7 @@ _swrast_write_rgba_span( struct gl_context *ctx, SWspan *span)
    const GLuint *colorMask = (GLuint *) ctx->Color.ColorMask;
    const GLbitfield origInterpMask = span->interpMask;
    const GLbitfield origArrayMask = span->arrayMask;
-   const GLbitfield origArrayAttribs = span->arrayAttribs;
+   const GLbitfield64 origArrayAttribs = span->arrayAttribs;
    const GLenum origChanType = span->array->ChanType;
    void * const origRgba = span->array->rgba;
    const GLboolean shader = (ctx->FragmentProgram._Current
diff --git a/mesalib/src/mesa/swrast/s_span.h b/mesalib/src/mesa/swrast/s_span.h
index 382c3d2eb..f4d32dd87 100644
--- a/mesalib/src/mesa/swrast/s_span.h
+++ b/mesalib/src/mesa/swrast/s_span.h
@@ -155,7 +155,8 @@ typedef struct sw_span
     */
    GLbitfield arrayMask;
 
-   GLbitfield arrayAttribs;
+   /** Mask of FRAG_BIT_x bits */
+   GLbitfield64 arrayAttribs;
 
    /**
     * We store the arrays of fragment values in a separate struct so