Merge branch 'master' of git+ssh://[email protected]/git/mesa/mesa

18 files changed, 151 insertions, 38 deletions

diff --git a/src/mesa/drivers/dri/i810/i810context.c b/src/mesa/drivers/dri/i810/i810context.c
index db8f7a19a23..f0332d9ae8b 100644
--- a/src/mesa/drivers/dri/i810/i810context.c
+++ b/src/mesa/drivers/dri/i810/i810context.c
@@ -65,6 +65,7 @@ SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
 
 #define need_GL_ARB_multisample
 #define need_GL_ARB_texture_compression
+#define need_GL_ARB_vertex_buffer_object
 #include "extension_helper.h"
 
 #ifndef I810_DEBUG
@@ -129,6 +130,7 @@ const struct dri_extension card_extensions[] =
     { "GL_ARB_texture_env_combine",        NULL },
     { "GL_ARB_texture_env_crossbar",       NULL },
     { "GL_ARB_texture_mirrored_repeat",    NULL },
+    { "GL_ARB_vertex_buffer_object",       GL_ARB_vertex_buffer_object_functions },
     { "GL_EXT_stencil_wrap",               NULL },
     { "GL_EXT_texture_edge_clamp",         NULL },
     { "GL_EXT_texture_env_combine",        NULL },
diff --git a/src/mesa/drivers/dri/i915tex/intel_buffers.c b/src/mesa/drivers/dri/i915tex/intel_buffers.c
index 45fd2fa8de0..62ff54b0075 100644
--- a/src/mesa/drivers/dri/i915tex/intel_buffers.c
+++ b/src/mesa/drivers/dri/i915tex/intel_buffers.c
@@ -327,6 +327,9 @@ intelWindowMoved(struct intel_context *intel)
 	 }
 
 	 for (i = 0; i < intel_fb->pf_num_pages; i++) {
+	    if (!intel_fb->color_rb[i])
+	       continue;
+
 	    vbl.request.sequence = intel_fb->color_rb[i]->vbl_pending;
 	    drmWaitVBlank(intel->driFd, &vbl);
 	 }
@@ -336,7 +339,8 @@ intelWindowMoved(struct intel_context *intel)
 	 intel_fb->vbl_waited = intel_fb->vbl_seq;
 
 	 for (i = 0; i < intel_fb->pf_num_pages; i++) {
-	    intel_fb->color_rb[i]->vbl_pending = intel_fb->vbl_waited;
+	    if (intel_fb->color_rb[i])
+	       intel_fb->color_rb[i]->vbl_pending = intel_fb->vbl_waited;
 	 }
       }
    } else {
diff --git a/src/mesa/drivers/dri/i965/brw_vs_tnl.c b/src/mesa/drivers/dri/i965/brw_vs_tnl.c
index c05a9b5ea10..e22a26b2917 100644
--- a/src/mesa/drivers/dri/i965/brw_vs_tnl.c
+++ b/src/mesa/drivers/dri/i965/brw_vs_tnl.c
@@ -850,14 +850,13 @@ static struct ureg calculate_light_attenuation( struct tnl_program *p,
    /* Calculate spot attenuation:
     */
    if (!p->state->unit[i].light_spotcutoff_is_180) {
-      struct ureg spot_dir = register_param3(p, STATE_LIGHT, i,
-					     STATE_SPOT_DIRECTION);
+      struct ureg spot_dir_norm = register_param3(p, STATE_INTERNAL,
+						  STATE_SPOT_DIR_NORMALIZED, i);
       struct ureg spot = get_temp(p);
       struct ureg slt = get_temp(p);
-	       
-      emit_normalize_vec3( p, spot, spot_dir ); /* XXX: precompute! */
-      emit_op2(p, OPCODE_DP3, spot, 0, ureg_negate(VPpli), spot);
-      emit_op2(p, OPCODE_SLT, slt, 0, swizzle1(spot_dir,W), spot);
+
+      emit_op2(p, OPCODE_DP3, spot, 0, ureg_negate(VPpli), spot_dir_norm);
+      emit_op2(p, OPCODE_SLT, slt, 0, swizzle1(spot_dir_norm,W), spot);
       emit_op2(p, OPCODE_POW, spot, 0, spot, swizzle1(attenuation, W));
       emit_op2(p, OPCODE_MUL, att, 0, slt, spot);
 
@@ -1166,7 +1165,8 @@ static void build_fog( struct tnl_program *p )
 
    if (p->state->fog_option &&
        p->state->tnl_do_vertex_fog) {
-      struct ureg params = register_param1(p, STATE_FOG_PARAMS);
+      struct ureg params = register_param2(p, STATE_INTERNAL,
+					   STATE_FOG_PARAMS_OPTIMIZED);
       struct ureg tmp = get_temp(p);
       struct ureg id = get_identity_param(p);
 
@@ -1174,8 +1174,7 @@ static void build_fog( struct tnl_program *p )
 
       switch (p->state->fog_option) {
       case FOG_LINEAR: {
-	 emit_op1(p, OPCODE_ABS, tmp, 0, input); 
-	 emit_op2(p, OPCODE_SUB, tmp, 0, swizzle1(params,Z), tmp); 
+	 emit_op3(p, OPCODE_MAD, tmp, 0, input, swizzle1(params,X), swizzle1(params,Y));
 	 emit_op2(p, OPCODE_MUL, tmp, 0, tmp, swizzle1(params,W)); 
 	 emit_op2(p, OPCODE_MAX, tmp, 0, tmp, swizzle1(id,X)); /* saturate */
 	 emit_op2(p, OPCODE_MIN, fog, WRITEMASK_X, tmp, swizzle1(id,W));
@@ -1183,15 +1182,13 @@ static void build_fog( struct tnl_program *p )
       }
       case FOG_EXP:
 	 emit_op1(p, OPCODE_ABS, tmp, 0, input); 
-	 emit_op2(p, OPCODE_MUL, tmp, 0, tmp, swizzle1(params,X)); 
-	 emit_op2(p, OPCODE_POW, fog, WRITEMASK_X, 
-		  register_const1f(p, M_E), ureg_negate(tmp)); 
+	 emit_op2(p, OPCODE_MUL, tmp, 0, tmp, swizzle1(params,Z));
+	 emit_op1(p, OPCODE_EX2, fog, WRITEMASK_X, ureg_negate(tmp));
 	 break;
       case FOG_EXP2:
-	 emit_op2(p, OPCODE_MUL, tmp, 0, input, swizzle1(params,X)); 
+	 emit_op2(p, OPCODE_MUL, tmp, 0, input, swizzle1(params,W));
 	 emit_op2(p, OPCODE_MUL, tmp, 0, tmp, tmp); 
-	 emit_op2(p, OPCODE_POW, fog, WRITEMASK_X, 
-		  register_const1f(p, M_E), ureg_negate(tmp)); 
+	 emit_op1(p, OPCODE_EX2, fog, WRITEMASK_X, ureg_negate(tmp));
 	 break;
       }
       
diff --git a/src/mesa/drivers/dri/mach64/mach64_context.c b/src/mesa/drivers/dri/mach64/mach64_context.c
index 5a6c301da2d..ad661e198cd 100644
--- a/src/mesa/drivers/dri/mach64/mach64_context.c
+++ b/src/mesa/drivers/dri/mach64/mach64_context.c
@@ -58,6 +58,7 @@
 #include "vblank.h"
 
 #define need_GL_ARB_multisample
+#define need_GL_ARB_vertex_buffer_object
 #include "extension_helper.h"
 
 #ifndef MACH64_DEBUG
@@ -83,6 +84,7 @@ const struct dri_extension card_extensions[] =
 {
     { "GL_ARB_multisample",                GL_ARB_multisample_functions },
     { "GL_ARB_multitexture",               NULL },
+    { "GL_ARB_vertex_buffer_object",       GL_ARB_vertex_buffer_object_functions },
     { "GL_EXT_texture_edge_clamp",         NULL },
     { "GL_MESA_ycbcr_texture",             NULL },
     { "GL_SGIS_generate_mipmap",           NULL },
diff --git a/src/mesa/drivers/dri/mga/mga_xmesa.c b/src/mesa/drivers/dri/mga/mga_xmesa.c
index 67a6f8bdf09..ca2c8faa183 100644
--- a/src/mesa/drivers/dri/mga/mga_xmesa.c
+++ b/src/mesa/drivers/dri/mga/mga_xmesa.c
@@ -72,6 +72,7 @@
 
 #define need_GL_ARB_multisample
 #define need_GL_ARB_texture_compression
+#define need_GL_ARB_vertex_buffer_object
 #define need_GL_ARB_vertex_program
 #define need_GL_EXT_fog_coord
 #define need_GL_EXT_multi_draw_arrays
@@ -401,6 +402,7 @@ static const struct dri_extension card_extensions[] =
    { "GL_ARB_multisample",            GL_ARB_multisample_functions },
    { "GL_ARB_texture_compression",    GL_ARB_texture_compression_functions },
    { "GL_ARB_texture_rectangle",      NULL },
+   { "GL_ARB_vertex_buffer_object",   GL_ARB_vertex_buffer_object_functions },
    { "GL_EXT_blend_logic_op",         NULL },
    { "GL_EXT_fog_coord",              GL_EXT_fog_coord_functions },
    { "GL_EXT_multi_draw_arrays",      GL_EXT_multi_draw_arrays_functions },
diff --git a/src/mesa/drivers/dri/r128/r128_context.c b/src/mesa/drivers/dri/r128/r128_context.c
index 89ddafa02a5..95e54a6af50 100644
--- a/src/mesa/drivers/dri/r128/r128_context.c
+++ b/src/mesa/drivers/dri/r128/r128_context.c
@@ -68,6 +68,7 @@ int R128_DEBUG = 0;
 
 #define need_GL_ARB_multisample
 #define need_GL_ARB_texture_compression
+#define need_GL_ARB_vertex_buffer_object
 #define need_GL_EXT_blend_minmax
 #define need_GL_EXT_fog_coord
 #define need_GL_EXT_secondary_color
@@ -80,6 +81,7 @@ const struct dri_extension card_extensions[] =
     { "GL_ARB_texture_compression",        GL_ARB_texture_compression_functions },
     { "GL_ARB_texture_env_add",            NULL },
     { "GL_ARB_texture_mirrored_repeat",    NULL },
+    { "GL_ARB_vertex_buffer_object",       GL_ARB_vertex_buffer_object_functions },
     { "GL_EXT_blend_subtract",             GL_EXT_blend_minmax_functions },
     { "GL_EXT_fog_coord",                  GL_EXT_fog_coord_functions },
     { "GL_EXT_texture_edge_clamp",         NULL },
diff --git a/src/mesa/drivers/dri/r300/r300_cmdbuf.c b/src/mesa/drivers/dri/r300/r300_cmdbuf.c
index 89725447f1d..e4511e0c21c 100644
--- a/src/mesa/drivers/dri/r300/r300_cmdbuf.c
+++ b/src/mesa/drivers/dri/r300/r300_cmdbuf.c
@@ -174,11 +174,12 @@ static __inline__ void r300DoEmitState(r300ContextPtr r300, GLboolean dirty)
 	dest ++;
 	r300->cmdbuf.count_used ++;
 
+	/* Emit cache flush */
 	*dest = cmdpacket0(R300_TX_CNTL, 1);
 	dest ++;
 	r300->cmdbuf.count_used ++;
 	
-	*dest = 0x0;
+	*dest = R300_TX_FLUSH;
 	dest ++;
 	r300->cmdbuf.count_used ++;
 	
diff --git a/src/mesa/drivers/dri/r300/r300_ioctl.c b/src/mesa/drivers/dri/r300/r300_ioctl.c
index 6fa34ee482a..f674fea7c9e 100644
--- a/src/mesa/drivers/dri/r300/r300_ioctl.c
+++ b/src/mesa/drivers/dri/r300/r300_ioctl.c
@@ -162,11 +162,11 @@ static void r300ClearBuffer(r300ContextPtr r300, int flags, int buffer)
 	cmd2[8].u = r300PackFloat32(ctx->Color.ClearColor[3]);
 
 	reg_start(R300_RB3D_DSTCACHE_CTLSTAT,0);
-	e32(0x0000000a);
+	e32(R300_RB3D_DSTCACHE_UNKNOWN_0A);
 	  
 
-	reg_start(0x4f18,0);
-	e32(0x00000003);
+	reg_start(R300_RB3D_ZCACHE_CTLSTAT,0);
+	e32(R300_RB3D_ZCACHE_UNKNOWN_03);
 	cp_wait(rmesa, R300_WAIT_3D | R300_WAIT_3D_CLEAN);
 }
 
diff --git a/src/mesa/drivers/dri/r300/r300_reg.h b/src/mesa/drivers/dri/r300/r300_reg.h
index 7bc832c871c..3de15752b16 100644
--- a/src/mesa/drivers/dri/r300/r300_reg.h
+++ b/src/mesa/drivers/dri/r300/r300_reg.h
@@ -497,6 +497,7 @@ USE OR OTHER DEALINGS IN THE SOFTWARE.
 
 /* Zero to flush caches. */
 #define R300_TX_CNTL                        0x4100
+#define R300_TX_FLUSH                       0x0
 
 /* The upper enable bits are guessed, based on fglrx reported limits. */
 #define R300_TX_ENABLE                      0x4104
@@ -1323,8 +1324,8 @@ USE OR OTHER DEALINGS IN THE SOFTWARE.
  * Set to 0A before 3D operations, set to 02 afterwards.
  */
 #define R300_RB3D_DSTCACHE_CTLSTAT          0x4E4C
-#       define R300_RB3D_DSTCACHE_02             0x00000002
-#       define R300_RB3D_DSTCACHE_0A             0x0000000A
+#       define R300_RB3D_DSTCACHE_UNKNOWN_02             0x00000002
+#       define R300_RB3D_DSTCACHE_UNKNOWN_0A             0x0000000A
 
 /* gap */
 /* There seems to be no "write only" setting, so use Z-test = ALWAYS
@@ -1395,6 +1396,12 @@ USE OR OTHER DEALINGS IN THE SOFTWARE.
 
 /* gap */
 
+#define R300_RB3D_ZCACHE_CTLSTAT            0x4F18 /* GUESS */
+#       define R300_RB3D_ZCACHE_UNKNOWN_01  0x1
+#       define R300_RB3D_ZCACHE_UNKNOWN_03  0x3
+
+/* gap */
+
 #define R300_RB3D_DEPTHOFFSET               0x4F20
 #define R300_RB3D_DEPTHPITCH                0x4F24
 #       define R300_DEPTHPITCH_MASK              0x00001FF8 /* GUESS */
diff --git a/src/mesa/drivers/dri/r300/r300_render.c b/src/mesa/drivers/dri/r300/r300_render.c
index 211c451f661..0864558e8dc 100644
--- a/src/mesa/drivers/dri/r300/r300_render.c
+++ b/src/mesa/drivers/dri/r300/r300_render.c
@@ -344,10 +344,10 @@ GLboolean r300_run_vb_render(GLcontext *ctx,
 	r300UpdateShaderStates(rmesa);
 	
 	reg_start(R300_RB3D_DSTCACHE_CTLSTAT,0);
-	e32(0x0000000a);
+	e32(R300_RB3D_DSTCACHE_UNKNOWN_0A);
 
-	reg_start(0x4f18,0);
-	e32(0x00000003);
+	reg_start(R300_RB3D_ZCACHE_CTLSTAT,0);
+	e32(R300_RB3D_ZCACHE_UNKNOWN_03);
 	
 	r300EmitState(rmesa);
 	
@@ -360,10 +360,10 @@ GLboolean r300_run_vb_render(GLcontext *ctx,
 	}
 
 	reg_start(R300_RB3D_DSTCACHE_CTLSTAT,0);
-	e32(0x0000000a/*0x2*/);
+	e32(R300_RB3D_DSTCACHE_UNKNOWN_0A /*R300_RB3D_DSTCACHE_UNKNOWN_02*/);
 
-	reg_start(0x4f18,0);
-	e32(0x00000003/*0x1*/);
+	reg_start(R300_RB3D_ZCACHE_CTLSTAT,0);
+	e32(R300_RB3D_ZCACHE_UNKNOWN_03 /*R300_RB3D_ZCACHE_UNKNOWN_01*/);
 
 #ifdef USER_BUFFERS
 	r300UseArrays(ctx);
diff --git a/src/mesa/drivers/dri/r300/radeon_mm.c b/src/mesa/drivers/dri/r300/radeon_mm.c
index 32ed1f43939..1502dac8fc3 100644
--- a/src/mesa/drivers/dri/r300/radeon_mm.c
+++ b/src/mesa/drivers/dri/r300/radeon_mm.c
@@ -284,7 +284,7 @@ static void emit_lin_cp(r300ContextPtr rmesa, unsigned long dst, unsigned long s
 	}
 	
 	reg_start(R300_RB3D_DSTCACHE_CTLSTAT,0);
-	e32(0x0000000a);
+	e32(R300_RB3D_DSTCACHE_UNKNOWN_0A);
 	
 	reg_start(0x342c,0);
 	e32(0x00000005);
diff --git a/src/mesa/drivers/dri/radeon/radeon_context.c b/src/mesa/drivers/dri/radeon/radeon_context.c
index 0d25951b648..d7c2d1407d6 100644
--- a/src/mesa/drivers/dri/radeon/radeon_context.c
+++ b/src/mesa/drivers/dri/radeon/radeon_context.c
@@ -64,6 +64,7 @@ WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
 
 #define need_GL_ARB_multisample
 #define need_GL_ARB_texture_compression
+#define need_GL_ARB_vertex_buffer_object
 #define need_GL_EXT_blend_minmax
 #define need_GL_EXT_fog_coord
 #define need_GL_EXT_secondary_color
@@ -122,6 +123,7 @@ const struct dri_extension card_extensions[] =
     { "GL_ARB_texture_env_crossbar",       NULL },
     { "GL_ARB_texture_env_dot3",           NULL },
     { "GL_ARB_texture_mirrored_repeat",    NULL },
+    { "GL_ARB_vertex_buffer_object",       GL_ARB_vertex_buffer_object_functions },
     { "GL_EXT_blend_logic_op",             NULL },
     { "GL_EXT_blend_subtract",             GL_EXT_blend_minmax_functions },
     { "GL_EXT_fog_coord",                  GL_EXT_fog_coord_functions },
diff --git a/src/mesa/drivers/dri/savage/savage_xmesa.c b/src/mesa/drivers/dri/savage/savage_xmesa.c
index f859217069c..43422db9a82 100644
--- a/src/mesa/drivers/dri/savage/savage_xmesa.c
+++ b/src/mesa/drivers/dri/savage/savage_xmesa.c
@@ -61,6 +61,7 @@
 
 #define need_GL_ARB_multisample
 #define need_GL_ARB_texture_compression
+#define need_GL_ARB_vertex_buffer_object
 #define need_GL_EXT_secondary_color
 #include "extension_helper.h"
 
@@ -135,6 +136,7 @@ static const struct dri_extension card_extensions[] =
     { "GL_ARB_multisample",                GL_ARB_multisample_functions },
     { "GL_ARB_multitexture",               NULL },
     { "GL_ARB_texture_compression",        GL_ARB_texture_compression_functions },
+    { "GL_ARB_vertex_buffer_object",       GL_ARB_vertex_buffer_object_functions },
     { "GL_EXT_stencil_wrap",               NULL },
     { "GL_EXT_texture_lod_bias",           NULL },
     { "GL_EXT_secondary_color",            GL_EXT_secondary_color_functions },
diff --git a/src/mesa/drivers/dri/sis/sis_context.c b/src/mesa/drivers/dri/sis/sis_context.c
index 89b81da347a..b21df0a61ea 100644
--- a/src/mesa/drivers/dri/sis/sis_context.c
+++ b/src/mesa/drivers/dri/sis/sis_context.c
@@ -60,6 +60,7 @@ USE OR OTHER DEALINGS IN THE SOFTWARE.
 
 #define need_GL_ARB_multisample
 #define need_GL_ARB_texture_compression
+#define need_GL_ARB_vertex_buffer_object
 #define need_GL_EXT_fog_coord
 #define need_GL_EXT_secondary_color
 #include "extension_helper.h"
@@ -79,6 +80,7 @@ struct dri_extension card_extensions[] =
     { "GL_ARB_texture_border_clamp",       NULL },
     { "GL_ARB_texture_compression",        GL_ARB_texture_compression_functions },
     { "GL_ARB_texture_mirrored_repeat",    NULL },
+    { "GL_ARB_vertex_buffer_object",       GL_ARB_vertex_buffer_object_functions },
     /*{ "GL_EXT_fog_coord",                  GL_EXT_fog_coord_functions },*/
     { "GL_EXT_texture_lod_bias",           NULL },
     { "GL_EXT_secondary_color",            GL_EXT_secondary_color_functions },
diff --git a/src/mesa/drivers/dri/unichrome/via_context.c b/src/mesa/drivers/dri/unichrome/via_context.c
index bc5a414df60..89533421b46 100644
--- a/src/mesa/drivers/dri/unichrome/via_context.c
+++ b/src/mesa/drivers/dri/unichrome/via_context.c
@@ -64,6 +64,7 @@
 
 #define need_GL_ARB_multisample
 #define need_GL_ARB_point_parameters
+#define need_GL_ARB_vertex_buffer_object
 #define need_GL_EXT_fog_coord
 #define need_GL_EXT_secondary_color
 #include "extension_helper.h"
@@ -372,6 +373,7 @@ const struct dri_extension card_extensions[] =
     { "GL_ARB_texture_env_combine",        NULL },
 /*    { "GL_ARB_texture_env_dot3",           NULL }, */
     { "GL_ARB_texture_mirrored_repeat",    NULL },
+    { "GL_ARB_vertex_buffer_object",       GL_ARB_vertex_buffer_object_functions },
     { "GL_EXT_fog_coord",                  GL_EXT_fog_coord_functions },
     { "GL_EXT_secondary_color",            GL_EXT_secondary_color_functions },
     { "GL_EXT_stencil_wrap",               NULL },
diff --git a/src/mesa/main/image.c b/src/mesa/main/image.c
index 7c0b63b1738..a60f4d3f4d8 100644
--- a/src/mesa/main/image.c
+++ b/src/mesa/main/image.c
@@ -60,6 +60,34 @@
 
 
 /**
+ * \return GL_TRUE if type is packed pixel type, GL_FALSE otherwise.
+ */
+static GLboolean
+_mesa_type_is_packed(GLenum type)
+    {
+        switch (type) {
+            case GL_UNSIGNED_BYTE_3_3_2:
+            case GL_UNSIGNED_BYTE_2_3_3_REV:
+            case GL_UNSIGNED_SHORT_5_6_5:
+            case GL_UNSIGNED_SHORT_5_6_5_REV:
+            case GL_UNSIGNED_SHORT_4_4_4_4:
+            case GL_UNSIGNED_SHORT_4_4_4_4_REV:
+            case GL_UNSIGNED_SHORT_5_5_5_1:
+            case GL_UNSIGNED_SHORT_1_5_5_5_REV:
+            case GL_UNSIGNED_INT_8_8_8_8:
+            case GL_UNSIGNED_INT_8_8_8_8_REV:
+            case GL_UNSIGNED_INT_10_10_10_2:
+            case GL_UNSIGNED_INT_2_10_10_10_REV:
+            case GL_UNSIGNED_SHORT_8_8_MESA:
+            case GL_UNSIGNED_SHORT_8_8_REV_MESA:
+            case GL_UNSIGNED_INT_24_8_EXT:
+                return GL_TRUE;
+         }
+
+        return GL_FALSE;
+}
+
+/**
  * Flip the 8 bits in each byte of the given array.
  *
  * \param p array.
@@ -4187,14 +4215,18 @@ _mesa_unpack_image( GLuint dimensions,
 
    if (type == GL_BITMAP) {
       bytesPerRow = (width + 7) >> 3;
-      flipBytes = !unpack->LsbFirst;
+      flipBytes = unpack->LsbFirst;
       swap2 = swap4 = GL_FALSE;
       compsPerRow = 0;
    }
    else {
       const GLint bytesPerPixel = _mesa_bytes_per_pixel(format, type);
-      const GLint components = _mesa_components_in_format(format);
+      GLint components = _mesa_components_in_format(format);
       GLint bytesPerComp;
+
+      if (_mesa_type_is_packed(type))
+          components = 1;
+
       if (bytesPerPixel <= 0 || components <= 0)
          return NULL;   /* bad format or type.  generate error later */
       bytesPerRow = bytesPerPixel * width;
@@ -4219,7 +4251,61 @@ _mesa_unpack_image( GLuint dimensions,
          for (row = 0; row < height; row++) {
             const GLvoid *src = _mesa_image_address(dimensions, unpack, pixels,
                                width, height, format, type, img, row, 0);
-            _mesa_memcpy(dst, src, bytesPerRow);
+
+                if ((type == GL_BITMAP) && (unpack->SkipPixels & 0x7)) {
+                    GLint i;
+                    flipBytes = GL_FALSE;
+                    if (unpack->LsbFirst) {
+                            GLubyte srcMask = 1 << (unpack->SkipPixels & 0x7);
+                            GLubyte dstMask = 128;
+                            const GLubyte *s = src;
+                            GLubyte *d = dst;
+                            *d = 0;
+                            for (i = 0; i < width; i++) {
+                                if (*s & srcMask) {
+                                    *d |= dstMask;
+                                }      
+                                if (srcMask == 128) {
+                                    srcMask = 1;
+                                    s++;
+                                } else {
+                                    srcMask = srcMask << 1;
+                                }
+                                if (dstMask == 1) {
+                                    dstMask = 128;
+                                    d++;
+                                    *d = 0;
+                                } else {
+                                    dstMask = dstMask >> 1;
+                                }
+                            }
+                    } else {
+                        GLubyte srcMask = 128 >> (unpack->SkipPixels & 0x7);
+                        GLubyte dstMask = 128;
+                        const GLubyte *s = src;
+                        GLubyte *d = dst;
+                        *d = 0;
+                        for (i = 0; i < width; i++) {
+                            if (*s & srcMask) {
+                                *d |= dstMask;
+                            }
+                            if (srcMask == 1) {
+                                srcMask = 128;
+                                s++;
+                            } else {
+                                srcMask = srcMask >> 1;
+                            }
+                            if (dstMask == 1) {
+                                dstMask = 128;
+                                d++;
+                                *d = 0;
+                            } else {
+                                dstMask = dstMask >> 1;
+                            }      
+                        }
+                    }
+                } else
+                    _mesa_memcpy(dst, src, bytesPerRow);
             /* byte flipping/swapping */
             if (flipBytes) {
                flip_bytes((GLubyte *) dst, bytesPerRow);
diff --git a/src/mesa/tnl/t_vp_build.c b/src/mesa/tnl/t_vp_build.c
index 2a97a1c8c76..2663d993049 100644
--- a/src/mesa/tnl/t_vp_build.c
+++ b/src/mesa/tnl/t_vp_build.c
@@ -1329,14 +1329,16 @@ static void build_pointsize( struct tnl_program *p )
    struct ureg out = register_output(p, VERT_RESULT_PSIZ);
    struct ureg ut = get_temp(p);
 
+   /* dist = |eyez| */
+   emit_op1(p, OPCODE_ABS, ut, WRITEMASK_Y, swizzle1(eye, Z));
    /* p1 + dist * (p2 + dist * p3); */
-   emit_op3(p, OPCODE_MAD, ut, 0, negate(swizzle1(eye, Z)),
+   emit_op3(p, OPCODE_MAD, ut, WRITEMASK_X, swizzle1(ut, Y),
 		swizzle1(state_attenuation, Z), swizzle1(state_attenuation, Y));
-   emit_op3(p, OPCODE_MAD, ut, 0, negate(swizzle1(eye, Z)),
+   emit_op3(p, OPCODE_MAD, ut, WRITEMASK_X, swizzle1(ut, Y),
 		ut, swizzle1(state_attenuation, X));
 
    /* 1 / sqrt(factor) */
-   emit_op1(p, OPCODE_RSQ, ut, 0, ut );
+   emit_op1(p, OPCODE_RSQ, ut, WRITEMASK_X, ut );
 
 #if 1
    /* out = pointSize / sqrt(factor) */
@@ -1344,8 +1346,8 @@ static void build_pointsize( struct tnl_program *p )
 #else
    /* not sure, might make sense to do clamping here,
       but it's not done in t_vb_points neither */
-   emit_op2(p, OPCODE_MUL, ut, 0, ut, state_size);
-   emit_op2(p, OPCODE_MAX, ut, 0, ut, swizzle1(state_size, Y));
+   emit_op2(p, OPCODE_MUL, ut, WRITEMASK_X, ut, state_size);
+   emit_op2(p, OPCODE_MAX, ut, WRITEMASK_X, ut, swizzle1(state_size, Y));
    emit_op2(p, OPCODE_MIN, out, WRITEMASK_X, ut, swizzle1(state_size, Z));
 #endif
 
diff --git a/src/mesa/x86/read_rgba_span_x86.S b/src/mesa/x86/read_rgba_span_x86.S
index a6902838164..3cbcd719960 100644
--- a/src/mesa/x86/read_rgba_span_x86.S
+++ b/src/mesa/x86/read_rgba_span_x86.S
@@ -369,7 +369,7 @@ _generic_read_RGBA_span_BGRA8888_REV_SSE2:
 	movdqa	mask, %xmm1
 	movdqa	mask+16, %xmm2
  */
-	LOAD_MASK(movdqa,%xmm1,%xmm2)
+	LOAD_MASK(movdqu,%xmm1,%xmm2)
 
 	movl	12(%esp), %ebx	/* source pointer */
 	movl	20(%esp), %edx	/* number of pixels to copy */