Updated to Mesa 7.7

7 files changed, 94 insertions, 137 deletions

diff --git a/mesalib/src/mesa/tnl/NOTES b/mesalib/src/mesa/tnl/NOTES
deleted file mode 100644
index c39b43485..000000000
--- a/mesalib/src/mesa/tnl/NOTES
+++ /dev/null
@@ -1,102 +0,0 @@
-INTRODUCTION
-
-A generic, configurable software implementation of GL transformation &
-lighting.
-
-This module provides an implementation of the routines required by the
-'vtxfmt' mechanism of core mesa for tnl functionality in all
-combinations of compile and execute modes.
-
-Most current drivers use the tnl module exclusively to provide this
-functionality, though there is an experimental alternate
-implementation provided by the tnl_dd/t_dd_imm_* files which can
-handle a small subset of GL states in execute mode only.
-
-
-STATE
-
-To create and destroy the module:
-
-	GLboolean _tnl_CreateContext( GLcontext *ctx );
-	void _tnl_DestroyContext( GLcontext *ctx );
-
-The module is not active by default, and must be installed by calling
-_tnl_Wakeup().  This function installs internal tnl functions into all
-the vtxfmt dispatch hooks, thus taking over the task of transformation
-and lighting entirely:
-
-        void _tnl_wakeup_exec( GLcontext *ctx );
-	void _tnl_wakeup_save_exec( GLcontext *ctx );
-
-   
-This module tracks state changes internally and maintains derived
-values based on the current state.  For this to work, the driver
-ensure the following funciton is called whenever the state changes and
-the swsetup module is 'awake':
-
-	void _tnl_InvalidateState( GLcontext *ctx, GLuint new_state );
-
-There is no explicit call to put the tnl module to sleep.  Simply
-install other function pointers into all the vtxfmt dispatch slots,
-and (optionally) cease calling _tnl_InvalidateState().
-
-CUSTOMIZATION
-
-The module provides customizability through several mechanisms.  The
-most important is by allowing drivers to specify the pipeline through
-which vertex data is passed, including its eventual transfer to
-rasterization hardware (or software).
-
-The default pipeline is specified in t_pipeline.c, and is usually a
-starting point for driver pipelines.  Some drivers will remove a stage
-where hardware provides support for the implemented operation (for
-instance fog where per-pixel hardware fog is available, as in the dri
-tdfx driver), or add stages to shortcircuit latter operations (for
-example taking advantage of hardware support for strips and other
-higher-level primitives (for example the radeon driver).
-
-In addition, the following functions provide further tweaks:
-
-extern void
-_tnl_need_projected_coords( GLcontext *ctx, GLboolean flag );
-
-	- Direct the default vertex transformation stage to
-          produce/not produce projected clip coordinates.
-	  
-extern void
-_tnl_need_dlist_loopback( GLcontext *ctx, GLboolean flag );
-      
-        - Direct the display list component of the tnl module to
-          replay display lists as 'glVertex' type calls, rather than
-          passing the display list data directly into the tnl pipeline
-          mechanism.  
-
-	  This allows display lists to be replayed by the tnl module
-	  even when the module is not strictly active.
-
-
-extern void
-_tnl_need_dlist_norm_lengths( GLcontext *ctx, GLboolean flag );
-
-	- Direct the display list component to enable/disable caching
-          1/length values for display list normals.  Doing so is
-          ususally helpful when lighting is performed in software, but
-          wasteful otherwise.
-
-
-DRIVER INTERFACE
-
-The module itself offers a minimal driver interface:
-
-	 void (*RunPipeline)( GLcontext *ctx );
-
-Normally this is set to _tnl_RunPipeline(), however the driver can use
-this hook to wrap checks or other code around this call. 
-
-In addition, the driver interface for the default render pipeline
-stage is housed in the tnl context struct (this could be cleaner).  
-
-
-RENDER DRIVER INTERFACE
-
-See t_context.h for the definition and explanation of this.
\ No newline at end of file
diff --git a/mesalib/src/mesa/tnl/t_context.c b/mesalib/src/mesa/tnl/t_context.c
index f2771cde0..db21b4589 100644
--- a/mesalib/src/mesa/tnl/t_context.c
+++ b/mesalib/src/mesa/tnl/t_context.c
@@ -171,7 +171,7 @@ _tnl_InvalidateState( GLcontext *ctx, GLuint new_state )
    if (vp) {
       GLuint i;
       for (i = 0; i < MAX_VARYING; i++) {
-         if (vp->Base.OutputsWritten & (1 << (VERT_RESULT_VAR0 + i))) {
+	 if (vp->Base.OutputsWritten & BITFIELD64_BIT(VERT_RESULT_VAR0 + i)) {
             RENDERINPUTS_SET(tnl->render_inputs_bitset,
                              _TNL_ATTRIB_GENERIC(i));
          }
diff --git a/mesalib/src/mesa/tnl/t_context.h b/mesalib/src/mesa/tnl/t_context.h
index c19eb3df3..6137c2d2f 100644
--- a/mesalib/src/mesa/tnl/t_context.h
+++ b/mesalib/src/mesa/tnl/t_context.h
@@ -548,4 +548,8 @@ typedef struct
 #define MAX_TYPES TYPE_IDX(GL_DOUBLE)+1      /* 0xa + 1 */
 
 
+extern void
+tnl_clip_prepare(GLcontext *ctx);
+
+
 #endif
diff --git a/mesalib/src/mesa/tnl/t_draw.c b/mesalib/src/mesa/tnl/t_draw.c
index c64c2c207..04fa10630 100644
--- a/mesalib/src/mesa/tnl/t_draw.c
+++ b/mesalib/src/mesa/tnl/t_draw.c
@@ -316,22 +316,27 @@ static void bind_indices( GLcontext *ctx,
 
    ptr = ADD_POINTERS(ib->obj->Pointer, ib->ptr);
 
-   if (ib->type == GL_UNSIGNED_INT) {
+   if (ib->type == GL_UNSIGNED_INT && VB->Primitive[0].basevertex == 0) {
       VB->Elts = (GLuint *) ptr;
    }
    else {
       GLuint *elts = (GLuint *)get_space(ctx, ib->count * sizeof(GLuint));
       VB->Elts = elts;
 
-      if (ib->type == GL_UNSIGNED_SHORT) {
+      if (ib->type == GL_UNSIGNED_INT) {
+	 const GLuint *in = (GLuint *)ptr;
+	 for (i = 0; i < ib->count; i++)
+	    *elts++ = (GLuint)(*in++) + VB->Primitive[0].basevertex;
+      }
+      else if (ib->type == GL_UNSIGNED_SHORT) {
 	 const GLushort *in = (GLushort *)ptr;
 	 for (i = 0; i < ib->count; i++) 
-	    *elts++ = (GLuint)(*in++);
+	    *elts++ = (GLuint)(*in++) + VB->Primitive[0].basevertex;
       }
       else {
 	 const GLubyte *in = (GLubyte *)ptr;
 	 for (i = 0; i < ib->count; i++) 
-	    *elts++ = (GLuint)(*in++);
+	    *elts++ = (GLuint)(*in++) + VB->Primitive[0].basevertex;
       }
    }
 }
@@ -390,10 +395,14 @@ void _tnl_draw_prims( GLcontext *ctx,
    TNLcontext *tnl = TNL_CONTEXT(ctx);
    const GLuint TEST_SPLIT = 0;
    const GLint max = TEST_SPLIT ? 8 : tnl->vb.Size - MAX_CLIPPED_VERTICES;
+   GLuint max_basevertex = prim->basevertex;
+   GLuint i;
+
+   for (i = 1; i < nr_prims; i++)
+      max_basevertex = MAX2(max_basevertex, prim[i].basevertex);
 
    if (0)
    {
-      GLuint i;
       _mesa_printf("%s %d..%d\n", __FUNCTION__, min_index, max_index);
       for (i = 0; i < nr_prims; i++)
 	 _mesa_printf("prim %d: %s start %d count %d\n", i, 
@@ -410,7 +419,7 @@ void _tnl_draw_prims( GLcontext *ctx,
 			_tnl_vbo_draw_prims );
       return;
    }
-   else if (max_index > max) {
+   else if (max_index + max_basevertex > max) {
       /* The software TNL pipeline has a fixed amount of storage for
        * vertices and it is necessary to split incoming drawing commands
        * if they exceed that limit.
@@ -424,7 +433,7 @@ void _tnl_draw_prims( GLcontext *ctx,
        * recursively call back into this function.
        */
       vbo_split_prims( ctx, arrays, prim, nr_prims, ib, 
-		       0, max_index,
+		       0, max_index + prim->basevertex,
 		       _tnl_vbo_draw_prims,
 		       &limits );
    }
@@ -435,17 +444,34 @@ void _tnl_draw_prims( GLcontext *ctx,
       struct gl_buffer_object *bo[VERT_ATTRIB_MAX + 1];
       GLuint nr_bo = 0;
 
-      /* Binding inputs may imply mapping some vertex buffer objects.
-       * They will need to be unmapped below.
-       */
-      bind_inputs(ctx, arrays, max_index+1, bo, &nr_bo);
-      bind_indices(ctx, ib, bo, &nr_bo);
-      bind_prims(ctx, prim, nr_prims );
+      for (i = 0; i < nr_prims;) {
+	 GLuint this_nr_prims;
+
+	 /* Our SW TNL pipeline doesn't handle basevertex yet, so bind_indices
+	  * will rebase the elements to the basevertex, and we'll only
+	  * emit strings of prims with the same basevertex in one draw call.
+	  */
+	 for (this_nr_prims = 1; i + this_nr_prims < nr_prims;
+	      this_nr_prims++) {
+	    if (prim[i].basevertex != prim[i + this_nr_prims].basevertex)
+	       break;
+	 }
+
+	 /* Binding inputs may imply mapping some vertex buffer objects.
+	  * They will need to be unmapped below.
+	  */
+	 bind_prims(ctx, &prim[i], this_nr_prims);
+	 bind_inputs(ctx, arrays, max_index + prim[i].basevertex + 1,
+		     bo, &nr_bo);
+	 bind_indices(ctx, ib, bo, &nr_bo);
 
-      TNL_CONTEXT(ctx)->Driver.RunPipeline(ctx);
+	 TNL_CONTEXT(ctx)->Driver.RunPipeline(ctx);
 
-      unmap_vbos(ctx, bo, nr_bo);
-      free_space(ctx);
+	 unmap_vbos(ctx, bo, nr_bo);
+	 free_space(ctx);
+
+	 i += this_nr_prims;
+      }
    }
 }
 
diff --git a/mesalib/src/mesa/tnl/t_rasterpos.c b/mesalib/src/mesa/tnl/t_rasterpos.c
index f1fdddf0f..99b678745 100644
--- a/mesalib/src/mesa/tnl/t_rasterpos.c
+++ b/mesalib/src/mesa/tnl/t_rasterpos.c
@@ -46,11 +46,10 @@
  * \return zero if outside view volume, or one if inside.
  */
 static GLuint
-viewclip_point( const GLfloat v[] )
+viewclip_point_xy( const GLfloat v[] )
 {
    if (   v[0] > v[3] || v[0] < -v[3]
-       || v[1] > v[3] || v[1] < -v[3]
-       || v[2] > v[3] || v[2] < -v[3] ) {
+       || v[1] > v[3] || v[1] < -v[3] ) {
       return 0;
    }
    else {
@@ -408,18 +407,18 @@ _tnl_RasterPos(GLcontext *ctx, const GLfloat vObj[4])
       /* apply projection matrix:  clip = Proj * eye */
       TRANSFORM_POINT( clip, ctx->ProjectionMatrixStack.Top->m, eye );
 
-      /* clip to view volume */
-      if (ctx->Transform.RasterPositionUnclipped) {
-         /* GL_IBM_rasterpos_clip: only clip against Z */
+      /* clip to view volume. */
+      if (!ctx->Transform.DepthClamp) {
          if (viewclip_point_z(clip) == 0) {
             ctx->Current.RasterPosValid = GL_FALSE;
             return;
          }
       }
-      else if (viewclip_point(clip) == 0) {
-         /* Normal OpenGL behaviour */
-         ctx->Current.RasterPosValid = GL_FALSE;
-         return;
+      if (!ctx->Transform.RasterPositionUnclipped) {
+         if (viewclip_point_xy(clip) == 0) {
+            ctx->Current.RasterPosValid = GL_FALSE;
+            return;
+         }
       }
 
       /* clip to user clipping planes */
@@ -443,6 +442,12 @@ _tnl_RasterPos(GLcontext *ctx, const GLfloat vObj[4])
                                   / ctx->DrawBuffer->_DepthMaxF;
       ctx->Current.RasterPos[3] = clip[3];
 
+      if (ctx->Transform.DepthClamp) {
+	 ctx->Current.RasterPos[3] = CLAMP(ctx->Current.RasterPos[3],
+					   ctx->Viewport.Near,
+					   ctx->Viewport.Far);
+      }
+
       /* compute raster distance */
       if (ctx->Fog.FogCoordinateSource == GL_FOG_COORDINATE_EXT)
          ctx->Current.RasterDistance = ctx->Current.Attrib[VERT_ATTRIB_FOG][0];
diff --git a/mesalib/src/mesa/tnl/t_vb_program.c b/mesalib/src/mesa/tnl/t_vb_program.c
index dc954bcba..e69f7d576 100644
--- a/mesalib/src/mesa/tnl/t_vb_program.c
+++ b/mesalib/src/mesa/tnl/t_vb_program.c
@@ -131,13 +131,16 @@ do_ndc_cliptest(GLcontext *ctx, struct vp_stage_data *store)
    store->ormask = 0;
    store->andmask = CLIP_FRUSTUM_BITS;
 
+   tnl_clip_prepare(ctx);
+
    if (tnl->NeedNdcCoords) {
       VB->NdcPtr =
          _mesa_clip_tab[VB->ClipPtr->size]( VB->ClipPtr,
                                             &store->ndcCoords,
                                             store->clipmask,
                                             &store->ormask,
-                                            &store->andmask );
+                                            &store->andmask,
+					    !ctx->Transform.DepthClamp );
    }
    else {
       VB->NdcPtr = NULL;
@@ -145,7 +148,8 @@ do_ndc_cliptest(GLcontext *ctx, struct vp_stage_data *store)
                                             NULL,
                                             store->clipmask,
                                             &store->ormask,
-                                            &store->andmask );
+                                            &store->andmask,
+					    !ctx->Transform.DepthClamp );
    }
 
    if (store->andmask) {
@@ -325,7 +329,7 @@ run_vp( GLcontext *ctx, struct tnl_pipeline_stage *stage )
    /* make list of outputs to save some time below */
    numOutputs = 0;
    for (i = 0; i < VERT_RESULT_MAX; i++) {
-      if (program->Base.OutputsWritten & (1 << i)) {
+      if (program->Base.OutputsWritten & BITFIELD64_BIT(i)) {
          outputs[numOutputs++] = i;
       }
    }
@@ -403,14 +407,14 @@ run_vp( GLcontext *ctx, struct tnl_pipeline_stage *stage )
    /* Fixup fog and point size results if needed */
    if (program->IsNVProgram) {
       if (ctx->Fog.Enabled &&
-          (program->Base.OutputsWritten & (1 << VERT_RESULT_FOGC)) == 0) {
+          (program->Base.OutputsWritten & BITFIELD64_BIT(VERT_RESULT_FOGC)) == 0) {
          for (i = 0; i < VB->Count; i++) {
             store->results[VERT_RESULT_FOGC].data[i][0] = 1.0;
          }
       }
 
       if (ctx->VertexProgram.PointSizeEnabled &&
-          (program->Base.OutputsWritten & (1 << VERT_RESULT_PSIZ)) == 0) {
+          (program->Base.OutputsWritten & BITFIELD64_BIT(VERT_RESULT_PSIZ)) == 0) {
          for (i = 0; i < VB->Count; i++) {
             store->results[VERT_RESULT_PSIZ].data[i][0] = ctx->Point.Size;
          }
@@ -468,7 +472,7 @@ run_vp( GLcontext *ctx, struct tnl_pipeline_stage *stage )
    }
 
    for (i = 0; i < ctx->Const.MaxVarying; i++) {
-      if (program->Base.OutputsWritten & (1 << (VERT_RESULT_VAR0 + i))) {
+      if (program->Base.OutputsWritten & BITFIELD64_BIT(VERT_RESULT_VAR0 + i)) {
          /* Note: varying results get put into the generic attributes */
 	 VB->AttribPtr[VERT_ATTRIB_GENERIC0+i]
             = &store->results[VERT_RESULT_VAR0 + i];
diff --git a/mesalib/src/mesa/tnl/t_vb_vertex.c b/mesalib/src/mesa/tnl/t_vb_vertex.c
index 30aa7c408..4734754ea 100644
--- a/mesalib/src/mesa/tnl/t_vb_vertex.c
+++ b/mesalib/src/mesa/tnl/t_vb_vertex.c
@@ -118,6 +118,22 @@ static void (*(usercliptab[5]))( GLcontext *,
 };
 
 
+void
+tnl_clip_prepare(GLcontext *ctx)
+{
+   /* Neither the x86 nor sparc asm cliptest functions have been updated
+    * for ARB_depth_clamp, so force the C paths.
+    */
+   if (ctx->Transform.DepthClamp) {
+      static GLboolean c_funcs_installed = GL_FALSE;
+      if (!c_funcs_installed) {
+         init_c_cliptest();
+         c_funcs_installed = GL_TRUE;
+      }
+   }
+}
+
+
 
 static GLboolean run_vertex_stage( GLcontext *ctx,
 				   struct tnl_pipeline_stage *stage )
@@ -129,6 +145,8 @@ static GLboolean run_vertex_stage( GLcontext *ctx,
    if (ctx->VertexProgram._Current) 
       return GL_TRUE;
 
+   tnl_clip_prepare(ctx);
+
    if (ctx->_NeedEyeCoords) {
       /* Separate modelview transformation:
        * Use combined ModelProject to avoid some depth artifacts
@@ -173,7 +191,8 @@ static GLboolean run_vertex_stage( GLcontext *ctx,
 					    &store->proj,
 					    store->clipmask,
 					    &store->ormask,
-					    &store->andmask );
+					    &store->andmask,
+					    !ctx->Transform.DepthClamp );
    }
    else {
       VB->NdcPtr = NULL;
@@ -181,7 +200,8 @@ static GLboolean run_vertex_stage( GLcontext *ctx,
 					    NULL,
 					    store->clipmask,
 					    &store->ormask,
-					    &store->andmask );
+					    &store->andmask,
+					    !ctx->Transform.DepthClamp );
    }
 
    if (store->andmask)