Merge remote branch 'origin/master' into gallium-array-texturesgallium-array-textures

Conflicts:
	src/gallium/drivers/i915/i915_resource_texture.c
	src/gallium/drivers/i915/i915_state_emit.c
	src/gallium/drivers/i915/i915_surface.c

56 files changed, 1249 insertions, 351 deletions

diff --git a/src/gallium/drivers/i915/TODO b/src/gallium/drivers/i915/TODO
new file mode 100644
index 00000000000..94c428bebf8
--- /dev/null
+++ b/src/gallium/drivers/i915/TODO
@@ -0,0 +1,25 @@
+Random list of problems with i915g:
+
+- Dies with BadDrawable on GLXFBconfig changes/destruction. Makes piglit totally
+  unusable :( Upgrading xserver helped here, it doesn't crash anymore. Still
+  broken, it doesn't update the viewport/get new buffers.
+
+- Tends to hang the chip after a few minutes of openarena. Looks tiling related,
+  at the last frame rendered has tiling corruption over the complete frame.
+
+- Kills the chip in 3D_PRIMITIVE LINELIST with mesa-demos/fbotexture in
+  wireframe mode.
+
+- Tiling is funny: If unlucky, it renders/samples all black. No clue yet what's
+  going on. Seems to depend on tiny details like whethever the sampler
+  relocation is fenced/unfenced (broken _with_ fenced reloc using tiling bits!).
+
+- Y-tiling is even more fun. i915c doesn't use it, maybe there's a reason?
+  Texture sampling from Y-tiled buffers seems to work, though (save above
+  problems).
+
+- Need to validate buffers before usage. Currently do_exec on the batchbuffer
+  can fail with -ENOSPC.
+
+Other bugs can be found here:
+https://bugs.freedesktop.org/buglist.cgi?bug_status=NEW&bug_status=ASSIGNED&bug_status=REOPENED&component=Drivers/Gallium/i915g
diff --git a/src/gallium/drivers/i915/i915_batch.h b/src/gallium/drivers/i915/i915_batch.h
index c411b84ccd4..6e93da76209 100644
--- a/src/gallium/drivers/i915/i915_batch.h
+++ b/src/gallium/drivers/i915/i915_batch.h
@@ -38,7 +38,10 @@
    i915_winsys_batchbuffer_dword(i915->batch, dword)
 
 #define OUT_RELOC(buf, usage, offset) \
-   i915_winsys_batchbuffer_reloc(i915->batch, buf, usage, offset)
+   i915_winsys_batchbuffer_reloc(i915->batch, buf, usage, offset, false)
+
+#define OUT_RELOC_FENCED(buf, usage, offset) \
+   i915_winsys_batchbuffer_reloc(i915->batch, buf, usage, offset, true)
 
 #define FLUSH_BATCH(fence) \
    i915_flush(i915, fence)
diff --git a/src/gallium/drivers/i915/i915_batchbuffer.h b/src/gallium/drivers/i915/i915_batchbuffer.h
index c1cd314e7b8..d92b2ccb31e 100644
--- a/src/gallium/drivers/i915/i915_batchbuffer.h
+++ b/src/gallium/drivers/i915/i915_batchbuffer.h
@@ -29,42 +29,47 @@
 #define I915_BATCHBUFFER_H
 
 #include "i915_winsys.h"
+#include "util/u_debug.h"
 
 struct i915_context;
 
+static INLINE size_t
+i915_winsys_batchbuffer_space(struct i915_winsys_batchbuffer *batch)
+{
+   return batch->size - (batch->ptr - batch->map);
+}
+
 static INLINE boolean
 i915_winsys_batchbuffer_check(struct i915_winsys_batchbuffer *batch,
                               size_t dwords,
                               size_t relocs)
 {
-   return dwords * 4 <= batch->size - (batch->ptr - batch->map) &&
+   return dwords * 4 <= i915_winsys_batchbuffer_space(batch) &&
           relocs <= (batch->max_relocs - batch->relocs);
 }
 
-static INLINE size_t
-i915_winsys_batchbuffer_space(struct i915_winsys_batchbuffer *batch)
+static INLINE void
+i915_winsys_batchbuffer_dword_unchecked(struct i915_winsys_batchbuffer *batch,
+                              unsigned dword)
 {
-   return batch->size - (batch->ptr - batch->map);
+   *(unsigned *)batch->ptr = dword;
+   batch->ptr += 4;
 }
 
 static INLINE void
 i915_winsys_batchbuffer_dword(struct i915_winsys_batchbuffer *batch,
                               unsigned dword)
 {
-   if (i915_winsys_batchbuffer_space(batch) < 4)
-      return;
-
-   *(unsigned *)batch->ptr = dword;
-   batch->ptr += 4;
+   assert (i915_winsys_batchbuffer_space(batch) >= 4);
+   i915_winsys_batchbuffer_dword_unchecked(batch, dword);
 }
 
 static INLINE void
 i915_winsys_batchbuffer_write(struct i915_winsys_batchbuffer *batch,
-                       void *data,
-                       size_t size)
+			      void *data,
+			      size_t size)
 {
-   if (i915_winsys_batchbuffer_space(batch) < size)
-      return;
+   assert (i915_winsys_batchbuffer_space(batch) >= size);
 
    memcpy(data, batch->ptr, size);
    batch->ptr += size;
@@ -74,9 +79,9 @@ static INLINE int
 i915_winsys_batchbuffer_reloc(struct i915_winsys_batchbuffer *batch,
                               struct i915_winsys_buffer *buffer,
                               enum i915_winsys_buffer_usage usage,
-                              size_t offset)
+                              size_t offset, bool fenced)
 {
-   return batch->iws->batchbuffer_reloc(batch, buffer, usage, offset);
+   return batch->iws->batchbuffer_reloc(batch, buffer, usage, offset, fenced);
 }
 
 #endif
diff --git a/src/gallium/drivers/i915/i915_blit.c b/src/gallium/drivers/i915/i915_blit.c
index cdf20c0055a..97c25665156 100644
--- a/src/gallium/drivers/i915/i915_blit.c
+++ b/src/gallium/drivers/i915/i915_blit.c
@@ -74,7 +74,7 @@ i915_fill_blit(struct i915_context *i915,
    OUT_BATCH(BR13);
    OUT_BATCH((y << 16) | x);
    OUT_BATCH(((y + h) << 16) | (x + w));
-   OUT_RELOC(dst_buffer, I915_USAGE_2D_TARGET, dst_offset);
+   OUT_RELOC_FENCED(dst_buffer, I915_USAGE_2D_TARGET, dst_offset);
    OUT_BATCH(color);
 }
 
@@ -138,8 +138,8 @@ i915_copy_blit(struct i915_context *i915,
    OUT_BATCH(BR13);
    OUT_BATCH((dst_y << 16) | dst_x);
    OUT_BATCH((dst_y2 << 16) | dst_x2);
-   OUT_RELOC(dst_buffer, I915_USAGE_2D_TARGET, dst_offset);
+   OUT_RELOC_FENCED(dst_buffer, I915_USAGE_2D_TARGET, dst_offset);
    OUT_BATCH((src_y << 16) | src_x);
    OUT_BATCH(((int) src_pitch & 0xffff));
-   OUT_RELOC(src_buffer, I915_USAGE_2D_SOURCE, src_offset);
+   OUT_RELOC_FENCED(src_buffer, I915_USAGE_2D_SOURCE, src_offset);
 }
diff --git a/src/gallium/drivers/i915/i915_context.h b/src/gallium/drivers/i915/i915_context.h
index 3ae61d0ea70..7103a1b8c16 100644
--- a/src/gallium/drivers/i915/i915_context.h
+++ b/src/gallium/drivers/i915/i915_context.h
@@ -193,8 +193,7 @@ struct i915_velems_state {
 };
 
 
-struct i915_context
-{
+struct i915_context {
    struct pipe_context base;
 
    struct i915_winsys *iws;
diff --git a/src/gallium/drivers/i915/i915_debug.c b/src/gallium/drivers/i915/i915_debug.c
index 57d3390dea3..d7150c99c4e 100644
--- a/src/gallium/drivers/i915/i915_debug.c
+++ b/src/gallium/drivers/i915/i915_debug.c
@@ -46,10 +46,12 @@ static const struct debug_named_value debug_options[] = {
 };
 
 unsigned i915_debug = 0;
+boolean i915_tiling = TRUE;
 
 void i915_debug_init(struct i915_screen *screen)
 {
    i915_debug = debug_get_flags_option("I915_DEBUG", debug_options, 0);
+   i915_tiling = !debug_get_bool_option("I915_NO_TILING", FALSE);
 }
 
 
diff --git a/src/gallium/drivers/i915/i915_debug.h b/src/gallium/drivers/i915/i915_debug.h
index fa60799d0c5..11af7662f0a 100644
--- a/src/gallium/drivers/i915/i915_debug.h
+++ b/src/gallium/drivers/i915/i915_debug.h
@@ -46,6 +46,7 @@ struct i915_winsys_batchbuffer;
 #define DBG_CONSTANTS 0x20
 
 extern unsigned i915_debug;
+extern boolean i915_tiling;
 
 #ifdef DEBUG
 static INLINE boolean
diff --git a/src/gallium/drivers/i915/i915_prim_vbuf.c b/src/gallium/drivers/i915/i915_prim_vbuf.c
index bd046bd9058..baebbc7bae3 100644
--- a/src/gallium/drivers/i915/i915_prim_vbuf.c
+++ b/src/gallium/drivers/i915/i915_prim_vbuf.c
@@ -172,6 +172,7 @@ i915_vbuf_render_reserve(struct i915_vbuf_render *i915_render, size_t size)
  *
  * Side effects:
  *    Updates hw_offset, sw_offset, index and allocates a new buffer.
+ *    Will set i915->vbo to null on buffer allocation.
  */
 static void
 i915_vbuf_render_new_buf(struct i915_vbuf_render *i915_render, size_t size)
@@ -179,8 +180,16 @@ i915_vbuf_render_new_buf(struct i915_vbuf_render *i915_render, size_t size)
    struct i915_context *i915 = i915_render->i915;
    struct i915_winsys *iws = i915->iws;
 
-   if (i915_render->vbo)
+   if (i915_render->vbo) {
       iws->buffer_destroy(iws, i915_render->vbo);
+      /*
+       * XXX If buffers where referenced then this should be done in
+       * update_vbo_state but since they arn't and malloc likes to reuse
+       * memory we need to set it to null
+       */
+      i915->vbo = NULL;
+      i915_render->vbo = NULL;
+   }
 
    i915->vbo_flushed = 0;
 
@@ -198,7 +207,7 @@ i915_vbuf_render_new_buf(struct i915_vbuf_render *i915_render, size_t size)
 #endif
 
    i915_render->vbo = iws->buffer_create(iws, i915_render->vbo_size,
-                                         64, I915_NEW_VERTEX);
+                                         I915_NEW_VERTEX);
 }
 
 /**
@@ -726,7 +735,7 @@ i915_vbuf_render_create(struct i915_context *i915)
    i915_render->pool_fifo = u_fifo_create(6);
    for (i = 0; i < 6; i++)
       u_fifo_add(i915_render->pool_fifo,
-                 iws->buffer_create(iws, i915_render->pool_buffer_size, 64,
+                 iws->buffer_create(iws, i915_render->pool_buffer_size,
                                     I915_NEW_VERTEX));
 #else
    (void)i;
diff --git a/src/gallium/drivers/i915/i915_reg.h b/src/gallium/drivers/i915/i915_reg.h
index cc28891e4ac..5e4e80ddf6b 100644
--- a/src/gallium/drivers/i915/i915_reg.h
+++ b/src/gallium/drivers/i915/i915_reg.h
@@ -753,7 +753,7 @@
 #define    MT_COMPRESS_DXT1_RGB		   (4<<3)
 #define MS3_USE_FENCE_REGS              (1<<2)
 #define MS3_TILED_SURFACE             (1<<1)
-#define MS3_TILE_WALK                 (1<<0)
+#define MS3_TILE_WALK_Y                (1<<0)
 
 #define MS4_PITCH_SHIFT                 21
 #define MS4_CUBE_FACE_ENA_NEGX          (1<<20)
@@ -851,6 +851,7 @@
 #define MI_FLUSH                   ((0<<29)|(4<<23))
 #define FLUSH_MAP_CACHE            (1<<0)
 #define INHIBIT_FLUSH_RENDER_CACHE (1<<2)
+#define MI_NOOP                    0
 
 
 #define CMD_3D (0x3<<29)
diff --git a/src/gallium/drivers/i915/i915_resource.h b/src/gallium/drivers/i915/i915_resource.h
index 753bd266b12..86620e6a123 100644
--- a/src/gallium/drivers/i915/i915_resource.h
+++ b/src/gallium/drivers/i915/i915_resource.h
@@ -49,6 +49,10 @@ struct i915_buffer {
 #define I915_MAX_TEXTURE_3D_LEVELS  8  /* max 128x128x128 */
 
 
+struct offset_pair {
+	unsigned short nblocksx;
+	unsigned short nblocksy;
+};
 
 struct i915_texture {
    struct u_resource b;
@@ -63,14 +67,18 @@ struct i915_texture {
 
    /* Explicitly store the offset of each image for each cube face or
     * depth value.
+    *
+    * Array [depth] off offsets.
     */
-   unsigned *image_offset[I915_MAX_TEXTURE_2D_LEVELS];   /**< array [depth] of offsets */
+   struct offset_pair *image_offset[I915_MAX_TEXTURE_2D_LEVELS];
 
    /* The data is held here:
     */
    struct i915_winsys_buffer *buffer;
 };
 
+unsigned i915_texture_offset(struct i915_texture *tex,
+                             unsigned level, unsigned layer);
 void i915_init_screen_resource_functions(struct i915_screen *is);
 void i915_init_resource_functions(struct i915_context *i915);
 
diff --git a/src/gallium/drivers/i915/i915_resource_texture.c b/src/gallium/drivers/i915/i915_resource_texture.c
index 67356abe208..f19106f3414 100644
--- a/src/gallium/drivers/i915/i915_resource_texture.c
+++ b/src/gallium/drivers/i915/i915_resource_texture.c
@@ -106,6 +106,23 @@ get_pot_stride(enum pipe_format format, unsigned width)
    return util_next_power_of_two(util_format_get_stride(format, width));
 }
 
+static INLINE const char*
+get_tiling_string(enum i915_winsys_buffer_tile tile)
+{
+   switch(tile) {
+   case I915_TILE_NONE:
+      return "none";
+   case I915_TILE_X:
+      return "x";
+   case I915_TILE_Y:
+      return "y";
+   default:
+      assert(FALSE);
+      return "?";
+   }
+}
+
+
 /*
  * More advanced helper funcs
  */
@@ -120,28 +137,56 @@ i915_texture_set_level_info(struct i915_texture *tex,
    assert(!tex->image_offset[level]);
 
    tex->nr_images[level] = nr_images;
-   tex->image_offset[level] = (unsigned *) MALLOC(nr_images * sizeof(unsigned));
-   tex->image_offset[level][0] = 0;
+   tex->image_offset[level] = MALLOC(nr_images * sizeof(struct offset_pair));
+   tex->image_offset[level][0].nblocksx = 0;
+   tex->image_offset[level][0].nblocksy = 0;
+}
+
+INLINE unsigned i915_texture_offset(struct i915_texture *tex,
+                                    unsigned level, unsigned layer)
+{
+   unsigned x, y;
+   x = tex->image_offset[level][layer].nblocksx
+      * util_format_get_blocksize(tex->b.b.format);
+   y = tex->image_offset[level][layer].nblocksy;
+
+   return y * tex->stride + x;
 }
 
 static void
 i915_texture_set_image_offset(struct i915_texture *tex,
                               unsigned level, unsigned img,
-                              unsigned x, unsigned y)
+                              unsigned nblocksx, unsigned nblocksy)
 {
    /* for the first image and level make sure offset is zero */
-   assert(!(img == 0 && level == 0) || (x == 0 && y == 0));
+   assert(!(img == 0 && level == 0) || (nblocksx == 0 && nblocksy == 0));
    assert(img < tex->nr_images[level]);
 
-   tex->image_offset[level][img] = y * tex->stride + x * util_format_get_blocksize(tex->b.b.format);
+   tex->image_offset[level][img].nblocksx = nblocksx;
+   tex->image_offset[level][img].nblocksy = nblocksy;
 
 #if DEBUG_TEXTURES
-   debug_printf("%s: %p level %u, img %u (%u, %u) %p\n", __FUNCTION__,
-                tex, level, img, x, y,
-                (void*)(uintptr_t)tex->image_offset[level][img]);
+   debug_printf("%s: %p level %u, img %u (%u, %u)\n", __FUNCTION__,
+                tex, level, img, x, y);
 #endif
 }
 
+static enum i915_winsys_buffer_tile
+i915_texture_tiling(struct pipe_resource *pt)
+{
+   if (!i915_tiling)
+      return I915_TILE_NONE;
+
+   if (pt->target == PIPE_TEXTURE_1D)
+      return I915_TILE_NONE;
+
+   if (util_format_is_s3tc(pt->format))
+      /* XXX X-tiling might make sense */
+      return I915_TILE_NONE;
+
+   return I915_TILE_X;
+}
+
 
 /*
  * Shared layout functions
@@ -163,9 +208,10 @@ i9x5_scanout_layout(struct i915_texture *tex)
    i915_texture_set_image_offset(tex, 0, 0, 0, 0);
 
    if (pt->width0 >= 240) {
-      tex->stride = get_pot_stride(pt->format, pt->width0);
+      tex->stride = align(util_format_get_stride(pt->format, pt->width0), 64);
       tex->total_nblocksy = align_nblocksy(pt->format, pt->height0, 8);
       tex->tiling = I915_TILE_X;
+   /* special case for cursors */
    } else if (pt->width0 == 64 && pt->height0 == 64) {
       tex->stride = get_pot_stride(pt->format, pt->width0);
       tex->total_nblocksy = align_nblocksy(pt->format, pt->height0, 8);
@@ -200,7 +246,7 @@ i9x5_display_target_layout(struct i915_texture *tex)
    i915_texture_set_level_info(tex, 0, 1);
    i915_texture_set_image_offset(tex, 0, 0, 0, 0);
 
-   tex->stride = get_pot_stride(pt->format, pt->width0);
+   tex->stride = align(util_format_get_stride(pt->format, pt->width0), 64);
    tex->total_nblocksy = align_nblocksy(pt->format, pt->height0, 8);
    tex->tiling = I915_TILE_X;
 
@@ -357,6 +403,8 @@ i915_texture_layout(struct i915_texture * tex)
 {
    struct pipe_resource *pt = &tex->b.b;
 
+   tex->tiling = i915_texture_tiling(pt);
+
    switch (pt->target) {
    case PIPE_TEXTURE_1D:
    case PIPE_TEXTURE_2D:
@@ -603,6 +651,8 @@ i945_texture_layout(struct i915_texture * tex)
 {
    struct pipe_resource *pt = &tex->b.b;
 
+   tex->tiling = i915_texture_tiling(pt);
+
    switch (pt->target) {
    case PIPE_TEXTURE_1D:
    case PIPE_TEXTURE_2D:
@@ -687,7 +737,6 @@ i915_texture_get_transfer(struct pipe_context *context,
    return transfer;
 }
 
-
 static void *
 i915_texture_transfer_map(struct pipe_context *pipe,
                           struct pipe_transfer *transfer)
@@ -703,7 +752,7 @@ i915_texture_transfer_map(struct pipe_context *pipe,
    if (resource->target != PIPE_TEXTURE_3D &&
        resource->target != PIPE_TEXTURE_CUBE)
       assert(box->z == 0);
-   offset = tex->image_offset[transfer->level][box->z];
+   offset = i915_texture_offset(tex, transfer->level, box->z);
 
    map = iws->buffer_map(iws, tex->buffer,
                          (transfer->usage & PIPE_TRANSFER_WRITE) ? TRUE : FALSE);
@@ -749,7 +798,6 @@ i915_texture_create(struct pipe_screen *screen,
    struct i915_screen *is = i915_screen(screen);
    struct i915_winsys *iws = is->iws;
    struct i915_texture *tex = CALLOC_STRUCT(i915_texture);
-   size_t tex_size;
    unsigned buf_usage = 0;
 
    if (!tex)
@@ -768,8 +816,6 @@ i915_texture_create(struct pipe_screen *screen,
          goto fail;
    }
 
-   tex_size = tex->stride * tex->total_nblocksy;
-
    /* for scanouts and cursors, cursors arn't scanouts */
 
    /* XXX: use a custom flag for cursors, don't rely on magically
@@ -780,27 +826,15 @@ i915_texture_create(struct pipe_screen *screen,
    else
       buf_usage = I915_NEW_TEXTURE;
 
-   tex->buffer = iws->buffer_create(iws, tex_size, 64, buf_usage);
+   tex->buffer = iws->buffer_create_tiled(iws, &tex->stride, tex->total_nblocksy,
+		   			  &tex->tiling, buf_usage);
    if (!tex->buffer)
       goto fail;
 
-   /* setup any hw fences */
-   if (tex->tiling) {
-      iws->buffer_set_fence_reg(iws, tex->buffer, tex->stride, tex->tiling);
-   }
-
-   
-#if 0
-   void *ptr = ws->buffer_map(ws, tex->buffer,
-      PIPE_BUFFER_USAGE_CPU_WRITE);
-   memset(ptr, 0x80, tex_size);
-   ws->buffer_unmap(ws, tex->buffer);
-#endif
-
-   I915_DBG(DBG_TEXTURE, "%s: %p size %u, stride %u, blocks (%u, %u)\n", __func__,
-            tex, (unsigned int)tex_size, tex->stride,
+   I915_DBG(DBG_TEXTURE, "%s: %p stride %u, blocks (%u, %u) tiling %s\n", __func__,
+            tex, tex->stride,
             tex->stride / util_format_get_blocksize(tex->b.b.format),
-            tex->total_nblocksy);
+            tex->total_nblocksy, get_tiling_string(tex->tiling));
 
    return &tex->b.b;
 
@@ -819,10 +853,11 @@ i915_texture_from_handle(struct pipe_screen * screen,
    struct i915_winsys *iws = is->iws;
    struct i915_winsys_buffer *buffer;
    unsigned stride;
+   enum i915_winsys_buffer_tile tiling;
 
    assert(screen);
 
-   buffer = iws->buffer_from_handle(iws, whandle, &stride);
+   buffer = iws->buffer_from_handle(iws, whandle, &tiling, &stride);
 
    /* Only supports one type */
    if ((template->target != PIPE_TEXTURE_2D &&
@@ -842,6 +877,7 @@ i915_texture_from_handle(struct pipe_screen * screen,
    tex->b.b.screen = screen;
 
    tex->stride = stride;
+   tex->tiling = tiling;
    tex->total_nblocksy = align_nblocksy(tex->b.b.format, tex->b.b.height0, 8);
 
    i915_texture_set_level_info(tex, 0, 1);
@@ -849,10 +885,10 @@ i915_texture_from_handle(struct pipe_screen * screen,
 
    tex->buffer = buffer;
 
-   I915_DBG(DBG_TEXTURE, "%s: %p stride %u, blocks (%ux%u)\n", __func__,
+   I915_DBG(DBG_TEXTURE, "%s: %p stride %u, blocks (%u, %u) tiling %s\n", __func__,
             tex, tex->stride,
             tex->stride / util_format_get_blocksize(tex->b.b.format),
-            tex->total_nblocksy);
+            tex->total_nblocksy, get_tiling_string(tex->tiling));
 
    return &tex->b.b;
 }
diff --git a/src/gallium/drivers/i915/i915_state_emit.c b/src/gallium/drivers/i915/i915_state_emit.c
index 9292fa00b43..c48d53ffbb2 100644
--- a/src/gallium/drivers/i915/i915_state_emit.c
+++ b/src/gallium/drivers/i915/i915_state_emit.c
@@ -86,6 +86,22 @@ framebuffer_size(const struct pipe_framebuffer_state *fb,
    }
 }
 
+static inline uint32_t
+buf_3d_tiling_bits(enum i915_winsys_buffer_tile tiling)
+{
+         uint32_t tiling_bits = 0;
+
+         switch (tiling) {
+         case I915_TILE_Y:
+            tiling_bits |= BUF_3D_TILE_WALK_Y;
+         case I915_TILE_X:
+            tiling_bits |= BUF_3D_TILED_SURFACE;
+         case I915_TILE_NONE:
+            break;
+         }
+
+         return tiling_bits;
+}
 
 /* Push the state into the sarea and/or texture memory.
  */
@@ -220,44 +236,39 @@ i915_emit_hardware_state(struct i915_context *i915 )
       struct pipe_surface *depth_surface = i915->framebuffer.zsbuf;
 
       if (cbuf_surface) {
-         unsigned ctile = BUF_3D_USE_FENCE;
          struct i915_texture *tex = i915_texture(cbuf_surface->texture);
-         unsigned offset;
          assert(tex);
 
-         offset = tex->image_offset[cbuf_surface->u.tex.level][cbuf_surface->u.tex.first_layer];
-
          OUT_BATCH(_3DSTATE_BUF_INFO_CMD);
 
          OUT_BATCH(BUF_3D_ID_COLOR_BACK |
                    BUF_3D_PITCH(tex->stride) |  /* pitch in bytes */
-                   ctile);
+                   buf_3d_tiling_bits(tex->tiling));
 
          OUT_RELOC(tex->buffer,
                    I915_USAGE_RENDER,
-                   offset);
+                   0);
       }
 
       /* What happens if no zbuf??
        */
       if (depth_surface) {
-         unsigned ztile = BUF_3D_USE_FENCE;
          struct i915_texture *tex = i915_texture(depth_surface->texture);
-         unsigned offset;
+         unsigned offset = i915_texture_offset(tex, depth_surface->u.tex.level,
+                                               depth_surface->u.tex.first_layer);
          assert(tex);
-
-         offset = tex->image_offset[depth_surface->u.tex.level][depth_surface->u.tex.first_layer];
+         assert(offset == 0);
 
          OUT_BATCH(_3DSTATE_BUF_INFO_CMD);
 
          assert(tex);
          OUT_BATCH(BUF_3D_ID_DEPTH |
                    BUF_3D_PITCH(tex->stride) |  /* pitch in bytes */
-                   ztile);
+                   buf_3d_tiling_bits(tex->tiling));
 
          OUT_RELOC(tex->buffer,
                    I915_USAGE_RENDER,
-                   offset);
+                   0);
       }
 
       {
@@ -299,12 +310,11 @@ i915_emit_hardware_state(struct i915_context *i915 )
                if (enabled & (1 << unit)) {
                   struct i915_texture *texture = i915_texture(i915->fragment_sampler_views[unit]->texture);
                   struct i915_winsys_buffer *buf = texture->buffer;
-                  uint offset = 0;
                   assert(buf);
 
                   count++;
 
-                  OUT_RELOC(buf, I915_USAGE_SAMPLER, offset);
+                  OUT_RELOC(buf, I915_USAGE_SAMPLER, 0);
                   OUT_BATCH(i915->current.texbuffer[unit][0]); /* MS3 */
                   OUT_BATCH(i915->current.texbuffer[unit][1]); /* MS4 */
                }
@@ -397,18 +407,33 @@ i915_emit_hardware_state(struct i915_context *i915 )
 #if 01
    /* drawing surface size */
    /* 6 dwords, 0 relocs */
+   if (i915->hardware_dirty & I915_HW_STATIC)
    {
       uint w, h;
-      boolean k = framebuffer_size(&i915->framebuffer, &w, &h);
-      (void)k;
-      assert(k);
+      struct pipe_surface *cbuf_surface = i915->framebuffer.cbufs[0];
+      struct i915_texture *tex = i915_texture(cbuf_surface->texture);
+      unsigned x, y;
+      int layer;
+      uint32_t draw_offset;
+      boolean ret;
 
+      ret = framebuffer_size(&i915->framebuffer, &w, &h);
+      assert(ret);
+
+      layer = cbuf_surface->u.tex.first_layer;
+
+      x = tex->image_offset[cbuf_surface->u.tex.level][layer].nblocksx;
+      y = tex->image_offset[cbuf_surface->u.tex.level][layer].nblocksy;
+
+      draw_offset = x | (y << 16);
+
+      /* XXX flush only required when the draw_offset changes! */
+      OUT_BATCH(MI_FLUSH | INHIBIT_FLUSH_RENDER_CACHE);
       OUT_BATCH(_3DSTATE_DRAW_RECT_CMD);
-      OUT_BATCH(0);
-      OUT_BATCH(0);
-      OUT_BATCH(((w - 1) & 0xffff) | ((h - 1) << 16));
-      OUT_BATCH(0);
-      OUT_BATCH(0);
+      OUT_BATCH(DRAW_RECT_DIS_DEPTH_OFS);
+      OUT_BATCH(draw_offset);
+      OUT_BATCH((w - 1 + x) | ((h - 1 + y) << 16));
+      OUT_BATCH(draw_offset);
    }
 #endif
 
diff --git a/src/gallium/drivers/i915/i915_state_sampler.c b/src/gallium/drivers/i915/i915_state_sampler.c
index 9771274ca11..916cb767536 100644
--- a/src/gallium/drivers/i915/i915_state_sampler.c
+++ b/src/gallium/drivers/i915/i915_state_sampler.c
@@ -243,6 +243,23 @@ static uint translate_texture_format(enum pipe_format pipeFormat)
    }
 }
 
+static inline uint32_t
+ms3_tiling_bits(enum i915_winsys_buffer_tile tiling)
+{
+         uint32_t tiling_bits = 0;
+
+         switch (tiling) {
+         case I915_TILE_Y:
+            tiling_bits |= MS3_TILE_WALK_Y;
+         case I915_TILE_X:
+            tiling_bits |= MS3_TILED_SURFACE;
+         case I915_TILE_NONE:
+            break;
+         }
+
+         return tiling_bits;
+}
+
 static void update_map(struct i915_context *i915,
                        uint unit,
                        const struct i915_texture *tex,
@@ -254,7 +271,6 @@ static void update_map(struct i915_context *i915,
    const uint width = pt->width0, height = pt->height0, depth = pt->depth0;
    const uint num_levels = pt->last_level;
    unsigned max_lod = num_levels * 4;
-   unsigned tiled = MS3_USE_FENCE_REGS;
 
    assert(tex);
    assert(width);
@@ -272,7 +288,7 @@ static void update_map(struct i915_context *i915,
       (((height - 1) << MS3_HEIGHT_SHIFT)
        | ((width - 1) << MS3_WIDTH_SHIFT)
        | format
-       | tiled);
+       | ms3_tiling_bits(tex->tiling));
 
    /*
     * XXX When min_filter != mag_filter and there's just one mipmap level,
diff --git a/src/gallium/drivers/i915/i915_surface.c b/src/gallium/drivers/i915/i915_surface.c
index 4ac1f90ef95..becc6e93c2d 100644
--- a/src/gallium/drivers/i915/i915_surface.c
+++ b/src/gallium/drivers/i915/i915_surface.c
@@ -59,13 +59,12 @@ i915_surface_copy(struct pipe_context *pipe,
    if (dst->target != PIPE_TEXTURE_CUBE &&
        dst->target != PIPE_TEXTURE_3D)
       assert(dstz == 0);
-   dst_offset = dst_tex->image_offset[dst_level][dstz];
+   dst_offset = i915_texture_offset(dst_tex, dst_level, dstz);
 
    if (src->target != PIPE_TEXTURE_CUBE &&
        src->target != PIPE_TEXTURE_3D)
       assert(src_box->z == 0);
-   src_offset = src_tex->image_offset[src_level][src_box->z];
-
+   src_offset = i915_texture_offset(src_tex, src_level, src_box->z);
 
    assert( dst != src );
    assert( util_format_get_blocksize(dpt->format) == util_format_get_blocksize(spt->format) );
@@ -93,7 +92,7 @@ i915_clear_render_target(struct pipe_context *pipe,
    struct i915_texture *tex = i915_texture(dst->texture);
    struct pipe_resource *pt = &tex->b.b;
    union util_color uc;
-   unsigned offset = tex->image_offset[dst->u.tex.level][dst->u.tex.first_layer];
+   unsigned offset = i915_texture_offset(tex, dst->u.tex.level, dst->u.tex.first_layer);
 
    assert(util_format_get_blockwidth(pt->format) == 1);
    assert(util_format_get_blockheight(pt->format) == 1);
@@ -122,7 +121,7 @@ i915_clear_depth_stencil(struct pipe_context *pipe,
    struct pipe_resource *pt = &tex->b.b;
    unsigned packedds;
    unsigned mask = 0;
-   unsigned offset = tex->image_offset[dst->u.tex.level][dst->u.tex.first_layer];
+   unsigned offset = i915_texture_offset(tex, dst->u.tex.level, dst->u.tex.first_layer);
 
    assert(util_format_get_blockwidth(pt->format) == 1);
    assert(util_format_get_blockheight(pt->format) == 1);
diff --git a/src/gallium/drivers/i915/i915_winsys.h b/src/gallium/drivers/i915/i915_winsys.h
index 5385e403d22..24ea416f015 100644
--- a/src/gallium/drivers/i915/i915_winsys.h
+++ b/src/gallium/drivers/i915/i915_winsys.h
@@ -53,6 +53,7 @@ enum i915_winsys_buffer_type
    I915_NEW_VERTEX
 };
 
+/* These need to be in sync with the definitions of libdrm-intel! */
 enum i915_winsys_buffer_tile
 {
    I915_TILE_NONE,
@@ -106,7 +107,7 @@ struct i915_winsys {
    int (*batchbuffer_reloc)(struct i915_winsys_batchbuffer *batch,
                             struct i915_winsys_buffer *reloc,
                             enum i915_winsys_buffer_usage usage,
-                            unsigned offset);
+                            unsigned offset, bool fenced);
 
    /**
     * Flush a bufferbatch.
@@ -130,10 +131,24 @@ struct i915_winsys {
     */
    struct i915_winsys_buffer *
       (*buffer_create)(struct i915_winsys *iws,
-                       unsigned size, unsigned alignment,
+                       unsigned size,
                        enum i915_winsys_buffer_type type);
 
    /**
+    * Create a tiled buffer.
+    *
+    * *stride, height are in bytes. The winsys tries to allocate the buffer with
+    * the tiling mode provide in *tiling. If tiling is no possible, *tiling will
+    * be set to I915_TILE_NONE. The calculated stride (incorporateing hw/kernel
+    * requirements) is always returned in *stride.
+    */
+   struct i915_winsys_buffer *
+      (*buffer_create_tiled)(struct i915_winsys *iws,
+                             unsigned *stride, unsigned height,
+                             enum i915_winsys_buffer_tile *tiling,
+                             enum i915_winsys_buffer_type type);
+
+   /**
     * Creates a buffer from a handle.
     * Used to implement pipe_screen::resource_from_handle.
     * Also provides the stride information needed for the
@@ -142,6 +157,7 @@ struct i915_winsys {
    struct i915_winsys_buffer *
       (*buffer_from_handle)(struct i915_winsys *iws,
                             struct winsys_handle *whandle,
+                            enum i915_winsys_buffer_tile *tiling,
                             unsigned *stride);
 
    /**
@@ -154,15 +170,6 @@ struct i915_winsys {
                                 unsigned stride);
 
    /**
-    * Fence a buffer with a fence reg.
-    * Not to be confused with pipe_fence_handle.
-    */
-   int (*buffer_set_fence_reg)(struct i915_winsys *iws,
-                               struct i915_winsys_buffer *buffer,
-                               unsigned stride,
-                               enum i915_winsys_buffer_tile tile);
-
-   /**
     * Map a buffer.
     */
    void *(*buffer_map)(struct i915_winsys *iws,
diff --git a/src/gallium/drivers/r300/r300_chipset.c b/src/gallium/drivers/r300/r300_chipset.c
index 48c24092114..583e981a4d2 100644
--- a/src/gallium/drivers/r300/r300_chipset.c
+++ b/src/gallium/drivers/r300/r300_chipset.c
@@ -424,4 +424,5 @@ void r300_parse_chipset(struct r300_capabilities* caps)
     }
 
     caps->is_rv350 = caps->family >= CHIP_FAMILY_RV350;
+    caps->dxtc_swizzle = caps->is_r400 || caps->is_r500;
 }
diff --git a/src/gallium/drivers/r300/r300_chipset.h b/src/gallium/drivers/r300/r300_chipset.h
index e7ca642b4f3..7ea4175dbee 100644
--- a/src/gallium/drivers/r300/r300_chipset.h
+++ b/src/gallium/drivers/r300/r300_chipset.h
@@ -79,6 +79,8 @@ struct r300_capabilities {
     boolean is_r500;
     /* Whether or not the second pixel pipe is accessed with the high bit */
     boolean high_second_pipe;
+    /* DXTC texture swizzling. */
+    boolean dxtc_swizzle;
 };
 
 /* Enumerations for legibility and telling which card we're running on. */
diff --git a/src/gallium/drivers/r300/r300_reg.h b/src/gallium/drivers/r300/r300_reg.h
index 6bea783f697..788c513be75 100644
--- a/src/gallium/drivers/r300/r300_reg.h
+++ b/src/gallium/drivers/r300/r300_reg.h
@@ -1520,11 +1520,11 @@ USE OR OTHER DEALINGS IN THE SOFTWARE.
 #	define R300_TX_TRI_PERF_3_8            (3<<15)
 #	define R300_ANISO_THRESHOLD_MASK       (7<<17)
 
+#       define R400_DXTC_SWIZZLE_ENABLE        (1<<21)
 #	define R500_MACRO_SWITCH               (1<<22)
 #       define R500_TX_MAX_ANISO(x)            ((x) << 23)
 #       define R500_TX_MAX_ANISO_MASK          (63 << 23)
 #       define R500_TX_ANISO_HIGH_QUALITY      (1 << 30)
-
 #	define R500_BORDER_FIX                 (1<<31)
 
 #define R300_TX_FORMAT0_0                   0x4480
diff --git a/src/gallium/drivers/r300/r300_screen.c b/src/gallium/drivers/r300/r300_screen.c
index 85de60df0f4..09981cb26b8 100644
--- a/src/gallium/drivers/r300/r300_screen.c
+++ b/src/gallium/drivers/r300/r300_screen.c
@@ -116,8 +116,9 @@ static int r300_get_param(struct pipe_screen* pscreen, enum pipe_cap param)
         case PIPE_CAP_TEXTURE_MIRROR_CLAMP:
         case PIPE_CAP_TEXTURE_MIRROR_REPEAT:
         case PIPE_CAP_BLEND_EQUATION_SEPARATE:
-        case PIPE_CAP_TEXTURE_SWIZZLE:
             return 1;
+        case PIPE_CAP_TEXTURE_SWIZZLE:
+            return util_format_s3tc_enabled ? r300screen->caps.dxtc_swizzle : 1;
 
         /* Unsupported features (boolean caps). */
         case PIPE_CAP_TIMER_QUERY:
diff --git a/src/gallium/drivers/r300/r300_state.c b/src/gallium/drivers/r300/r300_state.c
index 509e3106a45..0f563703c06 100644
--- a/src/gallium/drivers/r300/r300_state.c
+++ b/src/gallium/drivers/r300/r300_state.c
@@ -1347,6 +1347,7 @@ r300_create_sampler_view(struct pipe_context *pipe,
     struct r300_sampler_view *view = CALLOC_STRUCT(r300_sampler_view);
     struct r300_texture *tex = r300_texture(texture);
     boolean is_r500 = r300_screen(pipe->screen)->caps.is_r500;
+    boolean dxtc_swizzle = r300_screen(pipe->screen)->caps.dxtc_swizzle;
 
     if (view) {
         view->base = *templ;
@@ -1363,7 +1364,8 @@ r300_create_sampler_view(struct pipe_context *pipe,
         view->format = tex->tx_format;
         view->format.format1 |= r300_translate_texformat(templ->format,
                                                          view->swizzle,
-                                                         is_r500);
+                                                         is_r500,
+                                                         dxtc_swizzle);
         if (is_r500) {
             view->format.format2 |= r500_tx_format_msb_bit(templ->format);
         }
diff --git a/src/gallium/drivers/r300/r300_state_derived.c b/src/gallium/drivers/r300/r300_state_derived.c
index 5722c91ecec..a6d07760515 100644
--- a/src/gallium/drivers/r300/r300_state_derived.c
+++ b/src/gallium/drivers/r300/r300_state_derived.c
@@ -764,13 +764,18 @@ static void r300_merge_textures_and_samplers(struct r300_context* r300)
                 if (sampler->state.compare_mode == PIPE_TEX_COMPARE_NONE) {
                     texstate->format.format1 |=
                         r300_get_swizzle_combined(depth_swizzle,
-                                                  view->swizzle);
+                                                  view->swizzle, FALSE);
                 } else {
                     texstate->format.format1 |=
-                        r300_get_swizzle_combined(depth_swizzle, 0);
+                        r300_get_swizzle_combined(depth_swizzle, 0, FALSE);
                 }
             }
 
+            if (r300->screen->caps.dxtc_swizzle &&
+                util_format_is_compressed(tex->desc.b.b.format)) {
+                texstate->filter1 |= R400_DXTC_SWIZZLE_ENABLE;
+            }
+
             /* to emulate 1D textures through 2D ones correctly */
             if (tex->desc.b.b.target == PIPE_TEXTURE_1D) {
                 texstate->filter0 &= ~R300_TX_WRAP_T_MASK;
diff --git a/src/gallium/drivers/r300/r300_texture.c b/src/gallium/drivers/r300/r300_texture.c
index fe8859ab8f0..4b7b3e03564 100644
--- a/src/gallium/drivers/r300/r300_texture.c
+++ b/src/gallium/drivers/r300/r300_texture.c
@@ -40,7 +40,8 @@
 #include "pipe/p_screen.h"
 
 unsigned r300_get_swizzle_combined(const unsigned char *swizzle_format,
-                                   const unsigned char *swizzle_view)
+                                   const unsigned char *swizzle_view,
+                                   boolean dxtc_swizzle)
 {
     unsigned i;
     unsigned char swizzle[4];
@@ -51,10 +52,10 @@ unsigned r300_get_swizzle_combined(const unsigned char *swizzle_format,
         R300_TX_FORMAT_B_SHIFT,
         R300_TX_FORMAT_A_SHIFT
     };
-    const uint32_t swizzle_bit[4] = {
-        R300_TX_FORMAT_X,
+    uint32_t swizzle_bit[4] = {
+        dxtc_swizzle ? R300_TX_FORMAT_Z : R300_TX_FORMAT_X,
         R300_TX_FORMAT_Y,
-        R300_TX_FORMAT_Z,
+        dxtc_swizzle ? R300_TX_FORMAT_X : R300_TX_FORMAT_Z,
         R300_TX_FORMAT_W
     };
 
@@ -107,7 +108,8 @@ unsigned r300_get_swizzle_combined(const unsigned char *swizzle_format,
  * makes available X, Y, Z, W, ZERO, and ONE for swizzling. */
 uint32_t r300_translate_texformat(enum pipe_format format,
                                   const unsigned char *swizzle_view,
-                                  boolean is_r500)
+                                  boolean is_r500,
+                                  boolean dxtc_swizzle)
 {
     uint32_t result = 0;
     const struct util_format_description *desc;
@@ -169,7 +171,8 @@ uint32_t r300_translate_texformat(enum pipe_format format,
             }
     }
 
-    result |= r300_get_swizzle_combined(desc->swizzle, swizzle_view);
+    result |= r300_get_swizzle_combined(desc->swizzle, swizzle_view,
+                    util_format_is_compressed(format) && dxtc_swizzle);
 
     /* S3TC formats. */
     if (desc->layout == UTIL_FORMAT_LAYOUT_S3TC) {
@@ -571,7 +574,7 @@ boolean r300_is_zs_format_supported(enum pipe_format format)
 
 boolean r300_is_sampler_format_supported(enum pipe_format format)
 {
-    return r300_translate_texformat(format, 0, TRUE) != ~0;
+    return r300_translate_texformat(format, 0, TRUE, FALSE) != ~0;
 }
 
 void r300_texture_setup_format_state(struct r300_screen *screen,
diff --git a/src/gallium/drivers/r300/r300_texture.h b/src/gallium/drivers/r300/r300_texture.h
index be2740efc3b..0ab22f747e4 100644
--- a/src/gallium/drivers/r300/r300_texture.h
+++ b/src/gallium/drivers/r300/r300_texture.h
@@ -36,11 +36,13 @@ struct r300_texture;
 struct r300_screen;
 
 unsigned r300_get_swizzle_combined(const unsigned char *swizzle_format,
-                                   const unsigned char *swizzle_view);
+                                   const unsigned char *swizzle_view,
+                                   boolean dxtc_swizzle);
 
 uint32_t r300_translate_texformat(enum pipe_format format,
                                   const unsigned char *swizzle_view,
-                                  boolean is_r500);
+                                  boolean is_r500,
+                                  boolean dxtc_swizzle);
 
 uint32_t r500_tx_format_msb_bit(enum pipe_format format);
 
diff --git a/src/gallium/drivers/r300/r300_tgsi_to_rc.c b/src/gallium/drivers/r300/r300_tgsi_to_rc.c
index 33448bf0def..15a323989b2 100644
--- a/src/gallium/drivers/r300/r300_tgsi_to_rc.c
+++ b/src/gallium/drivers/r300/r300_tgsi_to_rc.c
@@ -57,7 +57,7 @@ static unsigned translate_opcode(unsigned opcode)
      /* case TGSI_OPCODE_DP2A: return RC_OPCODE_DP2A; */
                                         /* gap */
         case TGSI_OPCODE_FRC: return RC_OPCODE_FRC;
-     /* case TGSI_OPCODE_CLAMP: return RC_OPCODE_CLAMP; */
+        case TGSI_OPCODE_CLAMP: return RC_OPCODE_CLAMP;
         case TGSI_OPCODE_FLR: return RC_OPCODE_FLR;
      /* case TGSI_OPCODE_ROUND: return RC_OPCODE_ROUND; */
         case TGSI_OPCODE_EX2: return RC_OPCODE_EX2;
diff --git a/src/gallium/winsys/i915/drm/i915_drm_batchbuffer.c b/src/gallium/winsys/i915/drm/i915_drm_batchbuffer.c
index c6daa52a379..ebe86dcf196 100644
--- a/src/gallium/winsys/i915/drm/i915_drm_batchbuffer.c
+++ b/src/gallium/winsys/i915/drm/i915_drm_batchbuffer.c
@@ -14,9 +14,6 @@
 #define INTEL_BATCH_CLIPRECTS    0x2
 
 #undef INTEL_RUN_SYNC
-#undef INTEL_MAP_BATCHBUFFER
-#undef INTEL_MAP_GTT
-#define INTEL_ALWAYS_FLUSH
 
 struct i915_drm_batchbuffer
 {
@@ -72,11 +69,7 @@ i915_drm_batchbuffer_create(struct i915_winsys *iws)
 
    batch->actual_size = idws->max_batch_size;
 
-#ifdef INTEL_MAP_BATCHBUFFER
-   batch->base.map = NULL;
-#else
    batch->base.map = MALLOC(batch->actual_size);
-#endif
    batch->base.ptr = NULL;
    batch->base.size = 0;
 
@@ -94,7 +87,7 @@ static int
 i915_drm_batchbuffer_reloc(struct i915_winsys_batchbuffer *ibatch,
                             struct i915_winsys_buffer *buffer,
                             enum i915_winsys_buffer_usage usage,
-                            unsigned pre_add)
+                            unsigned pre_add, bool fenced)
 {
    struct i915_drm_batchbuffer *batch = i915_drm_batchbuffer(ibatch);
    unsigned write_domain = 0;
@@ -104,37 +97,44 @@ i915_drm_batchbuffer_reloc(struct i915_winsys_batchbuffer *ibatch,
 
    assert(batch->base.relocs < batch->base.max_relocs);
 
-   if (usage == I915_USAGE_SAMPLER) {
+   switch (usage) {
+   case I915_USAGE_SAMPLER:
       write_domain = 0;
       read_domain = I915_GEM_DOMAIN_SAMPLER;
-
-   } else if (usage == I915_USAGE_RENDER) {
+      break;
+   case I915_USAGE_RENDER:
       write_domain = I915_GEM_DOMAIN_RENDER;
       read_domain = I915_GEM_DOMAIN_RENDER;
-
-   } else if (usage == I915_USAGE_2D_TARGET) {
+      break;
+   case I915_USAGE_2D_TARGET:
       write_domain = I915_GEM_DOMAIN_RENDER;
       read_domain = I915_GEM_DOMAIN_RENDER;
-
-   } else if (usage == I915_USAGE_2D_SOURCE) {
+      break;
+   case I915_USAGE_2D_SOURCE:
       write_domain = 0;
       read_domain = I915_GEM_DOMAIN_RENDER;
-
-   } else if (usage == I915_USAGE_VERTEX) {
+      break;
+   case I915_USAGE_VERTEX:
       write_domain = 0;
       read_domain = I915_GEM_DOMAIN_VERTEX;
-
-   } else {
+      break;
+   default:
       assert(0);
       return -1;
    }
 
    offset = (unsigned)(batch->base.ptr - batch->base.map);
 
-   ret = drm_intel_bo_emit_reloc(batch->bo, offset,
-                                 intel_bo(buffer), pre_add,
-                                 read_domain,
-                                 write_domain);
+   if (fenced)
+      ret = drm_intel_bo_emit_reloc_fence(batch->bo, offset,
+				    intel_bo(buffer), pre_add,
+				    read_domain,
+				    write_domain);
+   else
+      ret = drm_intel_bo_emit_reloc(batch->bo, offset,
+				    intel_bo(buffer), pre_add,
+				    read_domain,
+				    write_domain);
 
    ((uint32_t*)batch->base.ptr)[0] = intel_bo(buffer)->offset + pre_add;
    batch->base.ptr += 4;
@@ -150,70 +150,32 @@ i915_drm_batchbuffer_flush(struct i915_winsys_batchbuffer *ibatch,
                             struct pipe_fence_handle **fence)
 {
    struct i915_drm_batchbuffer *batch = i915_drm_batchbuffer(ibatch);
-   unsigned used = 0;
-   int ret = 0;
+   unsigned used;
+   int ret;
 
-   assert(i915_winsys_batchbuffer_space(ibatch) >= 0);
+   /* MI_BATCH_BUFFER_END */
+   i915_winsys_batchbuffer_dword_unchecked(ibatch, (0xA<<23));
 
    used = batch->base.ptr - batch->base.map;
-   assert((used & 3) == 0);
-
-
-#ifdef INTEL_ALWAYS_FLUSH
-   /* MI_FLUSH | FLUSH_MAP_CACHE */
-   i915_winsys_batchbuffer_dword(ibatch, (0x4<<23)|(1<<0));
-   used += 4;
-#endif
-
-   if ((used & 4) == 0) {
+   if (used & 4) {
       /* MI_NOOP */
-      i915_winsys_batchbuffer_dword(ibatch, 0);
+      i915_winsys_batchbuffer_dword_unchecked(ibatch, 0);
+      used += 4;
    }
-   /* MI_BATCH_BUFFER_END */
-   i915_winsys_batchbuffer_dword(ibatch, (0xA<<23));
-
-   used = batch->base.ptr - batch->base.map;
-   assert((used & 4) == 0);
-
-#ifdef INTEL_MAP_BATCHBUFFER
-#ifdef INTEL_MAP_GTT
-   drm_intel_gem_bo_unmap_gtt(batch->bo);
-#else
-   drm_intel_bo_unmap(batch->bo);
-#endif
-#else
-   drm_intel_bo_subdata(batch->bo, 0, used, batch->base.map);
-#endif
 
    /* Do the sending to HW */
-   if (i915_drm_winsys(ibatch->iws)->send_cmd)
+   ret = drm_intel_bo_subdata(batch->bo, 0, used, batch->base.map);
+   if (ret == 0 && i915_drm_winsys(ibatch->iws)->send_cmd)
       ret = drm_intel_bo_exec(batch->bo, used, NULL, 0, 0);
-   else
-      ret = 0;
 
    if (ret != 0 || i915_drm_winsys(ibatch->iws)->dump_cmd) {
-#ifdef INTEL_MAP_BATCHBUFFER
-#ifdef INTEL_MAP_GTT
-      drm_intel_gem_bo_map_gtt(batch->bo);
-#else
-      drm_intel_bo_map(batch->bo, 0);
-#endif
-#endif
       i915_dump_batchbuffer(ibatch);
       assert(ret == 0);
-#ifdef INTEL_MAP_BATCHBUFFER
-#ifdef INTEL_MAP_GTT
-   drm_intel_gem_bo_unmap_gtt(batch->bo);
-#else
-   drm_intel_bo_unmap(batch->bo);
-#endif
-#endif
-   } else {
+   }
+
 #ifdef INTEL_RUN_SYNC
-      drm_intel_bo_map(batch->bo, FALSE);
-      drm_intel_bo_unmap(batch->bo);
+   drm_intel_bo_wait_rendering(batch->bo);
 #endif
-   }
 
    if (fence) {
       ibatch->iws->fence_reference(ibatch->iws, fence, NULL);
@@ -237,9 +199,7 @@ i915_drm_batchbuffer_destroy(struct i915_winsys_batchbuffer *ibatch)
    if (batch->bo)
       drm_intel_bo_unreference(batch->bo);
 
-#ifndef INTEL_MAP_BATCHBUFFER
    FREE(batch->base.map);
-#endif
    FREE(batch);
 }
 
diff --git a/src/gallium/winsys/i915/drm/i915_drm_buffer.c b/src/gallium/winsys/i915/drm/i915_drm_buffer.c
index 15ec4487457..01dd4bf062f 100644
--- a/src/gallium/winsys/i915/drm/i915_drm_buffer.c
+++ b/src/gallium/winsys/i915/drm/i915_drm_buffer.c
@@ -5,14 +5,31 @@
 
 #include "i915_drm.h"
 
+static char *i915_drm_type_to_name(enum i915_winsys_buffer_type type)
+{
+   char *name;
+
+   if (type == I915_NEW_TEXTURE) {
+      name = "gallium3d_texture";
+   } else if (type == I915_NEW_VERTEX) {
+      name = "gallium3d_vertex";
+   } else if (type == I915_NEW_SCANOUT) {
+      name = "gallium3d_scanout";
+   } else {
+      assert(0);
+      name = "gallium3d_unknown";
+   }
+
+   return name;
+}
+
 static struct i915_winsys_buffer *
 i915_drm_buffer_create(struct i915_winsys *iws,
-                        unsigned size, unsigned alignment,
+                        unsigned size,
                         enum i915_winsys_buffer_type type)
 {
    struct i915_drm_buffer *buf = CALLOC_STRUCT(i915_drm_buffer);
    struct i915_drm_winsys *idws = i915_drm_winsys(iws);
-   char *name;
 
    if (!buf)
       return NULL;
@@ -21,22 +38,48 @@ i915_drm_buffer_create(struct i915_winsys *iws,
    buf->flinked = FALSE;
    buf->flink = 0;
 
-   if (type == I915_NEW_TEXTURE) {
-      name = "gallium3d_texture";
-   } else if (type == I915_NEW_VERTEX) {
-      name = "gallium3d_vertex";
-   } else if (type == I915_NEW_SCANOUT) {
-      name = "gallium3d_scanout";
-   } else {
-      assert(0);
-      name = "gallium3d_unknown";
-   }
+   buf->bo = drm_intel_bo_alloc(idws->gem_manager,
+                                i915_drm_type_to_name(type), size, 0);
 
-   buf->bo = drm_intel_bo_alloc(idws->gem_manager, name, size, alignment);
+   if (!buf->bo)
+      goto err;
+
+   return (struct i915_winsys_buffer *)buf;
+
+err:
+   assert(0);
+   FREE(buf);
+   return NULL;
+}
+
+static struct i915_winsys_buffer *
+i915_drm_buffer_create_tiled(struct i915_winsys *iws,
+                             unsigned *stride, unsigned height, 
+                             enum i915_winsys_buffer_tile *tiling,
+                             enum i915_winsys_buffer_type type)
+{
+   struct i915_drm_buffer *buf = CALLOC_STRUCT(i915_drm_buffer);
+   struct i915_drm_winsys *idws = i915_drm_winsys(iws);
+   unsigned long pitch = 0;
+   uint32_t tiling_mode = *tiling;
+
+   if (!buf)
+      return NULL;
+
+   buf->magic = 0xDEAD1337;
+   buf->flinked = FALSE;
+   buf->flink = 0;
+
+   buf->bo = drm_intel_bo_alloc_tiled(idws->gem_manager,
+                                      i915_drm_type_to_name(type),
+		   		      *stride, height, 1,
+                                      &tiling_mode, &pitch, 0);
 
    if (!buf->bo)
       goto err;
 
+   *stride = pitch;
+   *tiling = tiling_mode;
    return (struct i915_winsys_buffer *)buf;
 
 err:
@@ -47,8 +90,9 @@ err:
 
 static struct i915_winsys_buffer *
 i915_drm_buffer_from_handle(struct i915_winsys *iws,
-                             struct winsys_handle *whandle,
-                             unsigned *stride)
+                            struct winsys_handle *whandle,
+                            enum i915_winsys_buffer_tile *tiling,
+                            unsigned *stride)
 {
    struct i915_drm_winsys *idws = i915_drm_winsys(iws);
    struct i915_drm_buffer *buf = CALLOC_STRUCT(i915_drm_buffer);
@@ -68,6 +112,7 @@ i915_drm_buffer_from_handle(struct i915_winsys *iws,
    drm_intel_bo_get_tiling(buf->bo, &tile, &swizzle);
 
    *stride = whandle->stride;
+   *tiling = tile;
 
    return (struct i915_winsys_buffer *)buf;
 
@@ -103,24 +148,6 @@ i915_drm_buffer_get_handle(struct i915_winsys *iws,
    return TRUE;
 }
 
-static int
-i915_drm_buffer_set_fence_reg(struct i915_winsys *iws,
-                               struct i915_winsys_buffer *buffer,
-                               unsigned stride,
-                               enum i915_winsys_buffer_tile tile)
-{
-   struct i915_drm_buffer *buf = i915_drm_buffer(buffer);
-   assert(I915_TILING_NONE == I915_TILE_NONE);
-   assert(I915_TILING_X == I915_TILE_X);
-   assert(I915_TILING_Y == I915_TILE_Y);
-
-   if (tile != I915_TILE_NONE) {
-      assert(buf->map_count == 0);
-   }
-
-   return drm_intel_bo_set_tiling(buf->bo, &tile, stride);
-}
-
 static void *
 i915_drm_buffer_map(struct i915_winsys *iws,
                      struct i915_winsys_buffer *buffer,
@@ -190,9 +217,9 @@ void
 i915_drm_winsys_init_buffer_functions(struct i915_drm_winsys *idws)
 {
    idws->base.buffer_create = i915_drm_buffer_create;
+   idws->base.buffer_create_tiled = i915_drm_buffer_create_tiled;
    idws->base.buffer_from_handle = i915_drm_buffer_from_handle;
    idws->base.buffer_get_handle = i915_drm_buffer_get_handle;
-   idws->base.buffer_set_fence_reg = i915_drm_buffer_set_fence_reg;
    idws->base.buffer_map = i915_drm_buffer_map;
    idws->base.buffer_unmap = i915_drm_buffer_unmap;
    idws->base.buffer_write = i915_drm_buffer_write;
diff --git a/src/gallium/winsys/i915/drm/i915_drm_winsys.c b/src/gallium/winsys/i915/drm/i915_drm_winsys.c
index cc0b6a99577..2288b48b2bd 100644
--- a/src/gallium/winsys/i915/drm/i915_drm_winsys.c
+++ b/src/gallium/winsys/i915/drm/i915_drm_winsys.c
@@ -69,6 +69,7 @@ i915_drm_winsys_create(int drmFD)
 
    idws->gem_manager = drm_intel_bufmgr_gem_init(idws->fd, idws->max_batch_size);
    drm_intel_bufmgr_gem_enable_reuse(idws->gem_manager);
+   drm_intel_bufmgr_gem_enable_fenced_relocs(idws->gem_manager);
 
    idws->dump_cmd = debug_get_bool_option("I915_DUMP_CMD", FALSE);
    idws->send_cmd = !debug_get_bool_option("I915_NO_HW", FALSE);
diff --git a/src/gallium/winsys/i915/sw/i915_sw_batchbuffer.c b/src/gallium/winsys/i915/sw/i915_sw_batchbuffer.c
index a480cfed57b..44773ae30e7 100644
--- a/src/gallium/winsys/i915/sw/i915_sw_batchbuffer.c
+++ b/src/gallium/winsys/i915/sw/i915_sw_batchbuffer.c
@@ -61,7 +61,7 @@ static int
 i915_sw_batchbuffer_reloc(struct i915_winsys_batchbuffer *ibatch,
                           struct i915_winsys_buffer *buffer,
                           enum i915_winsys_buffer_usage usage,
-                          unsigned pre_add)
+                          unsigned pre_add, bool fenced)
 {
    struct i915_sw_batchbuffer *batch = i915_sw_batchbuffer(ibatch);
    int ret = 0;
diff --git a/src/gallium/winsys/i915/sw/i915_sw_buffer.c b/src/gallium/winsys/i915/sw/i915_sw_buffer.c
index df175688861..834805e621d 100644
--- a/src/gallium/winsys/i915/sw/i915_sw_buffer.c
+++ b/src/gallium/winsys/i915/sw/i915_sw_buffer.c
@@ -4,28 +4,15 @@
 
 static struct i915_winsys_buffer *
 i915_sw_buffer_create(struct i915_winsys *iws,
-                      unsigned size, unsigned alignment,
+                      unsigned size,
                       enum i915_winsys_buffer_type type)
 {
    struct i915_sw_buffer *buf = CALLOC_STRUCT(i915_sw_buffer);
-   char *name;
 
    if (!buf)
       return NULL;
 
-   if (type == I915_NEW_TEXTURE) {
-      name = "gallium3d_texture";
-   } else if (type == I915_NEW_VERTEX) {
-      name = "gallium3d_vertex";
-   } else if (type == I915_NEW_SCANOUT) {
-      name = "gallium3d_scanout";
-   } else {
-      assert(0);
-      name = "gallium3d_unknown";
-   }
-
    buf->magic = 0xDEAD1337;
-   buf->name = name;
    buf->type = type;
    buf->ptr = CALLOC(size, 1);
 
@@ -40,21 +27,32 @@ err:
    return NULL;
 }
 
-static int
-i915_sw_buffer_set_fence_reg(struct i915_winsys *iws,
-                               struct i915_winsys_buffer *buffer,
-                               unsigned stride,
-                               enum i915_winsys_buffer_tile tile)
+static struct i915_winsys_buffer *
+i915_sw_buffer_create_tiled(struct i915_winsys *iws,
+                      unsigned *stride, unsigned height, 
+                      enum i915_winsys_buffer_tile *tiling,
+                      enum i915_winsys_buffer_type type)
 {
-   struct i915_sw_buffer *buf = i915_sw_buffer(buffer);
+   struct i915_sw_buffer *buf = CALLOC_STRUCT(i915_sw_buffer);
+
+   if (!buf)
+      return NULL;
+
+   buf->magic = 0xDEAD1337;
+   buf->type = type;
+   buf->ptr = CALLOC(*stride * height, 1);
+   buf->tiling = *tiling;
+   buf->stride = *stride;
 
-   if (tile != I915_TILE_NONE) {
-      assert(buf->map_count == 0);
-   }
+   if (!buf->ptr)
+      goto err;
 
-   buf->tile = tile;
+   return (struct i915_winsys_buffer *)buf;
 
-   return 0;
+err:
+   assert(0);
+   FREE(buf);
+   return NULL;
 }
 
 static void *
@@ -108,7 +106,7 @@ void
 i915_sw_winsys_init_buffer_functions(struct i915_sw_winsys *isws)
 {
    isws->base.buffer_create = i915_sw_buffer_create;
-   isws->base.buffer_set_fence_reg = i915_sw_buffer_set_fence_reg;
+   isws->base.buffer_create_tiled = i915_sw_buffer_create_tiled;
    isws->base.buffer_map = i915_sw_buffer_map;
    isws->base.buffer_unmap = i915_sw_buffer_unmap;
    isws->base.buffer_write = i915_sw_buffer_write;
diff --git a/src/gallium/winsys/i915/sw/i915_sw_winsys.h b/src/gallium/winsys/i915/sw/i915_sw_winsys.h
index b7b43669f30..3af2548419e 100644
--- a/src/gallium/winsys/i915/sw/i915_sw_winsys.h
+++ b/src/gallium/winsys/i915/sw/i915_sw_winsys.h
@@ -43,8 +43,8 @@ struct i915_sw_buffer {
    void *ptr;
    unsigned map_count;
    enum i915_winsys_buffer_type type;
-   enum i915_winsys_buffer_tile tile;
-   const char *name;
+   enum i915_winsys_buffer_tile tiling;
+   unsigned stride;
 };
 
 static INLINE struct i915_sw_buffer *
diff --git a/src/glsl/Makefile b/src/glsl/Makefile
index f5aadc347bd..2674c6ec485 100644
--- a/src/glsl/Makefile
+++ b/src/glsl/Makefile
@@ -52,6 +52,7 @@ CXX_SOURCES = \
 	loop_analysis.cpp \
 	loop_controls.cpp \
 	loop_unroll.cpp \
+	lower_discard.cpp \
 	lower_if_to_cond_assign.cpp \
 	lower_instructions.cpp \
 	lower_jumps.cpp \
@@ -70,6 +71,7 @@ CXX_SOURCES = \
 	opt_dead_code.cpp \
 	opt_dead_code_local.cpp \
 	opt_dead_functions.cpp \
+	opt_discard_simplification.cpp \
 	opt_function_inlining.cpp \
 	opt_if_simplification.cpp \
 	opt_noop_swizzle.cpp \
diff --git a/src/glsl/SConscript b/src/glsl/SConscript
index fd22f668631..b5b1728beef 100644
--- a/src/glsl/SConscript
+++ b/src/glsl/SConscript
@@ -49,6 +49,7 @@ sources = [
     'loop_analysis.cpp',
     'loop_controls.cpp',
     'loop_unroll.cpp',
+    'lower_discard.cpp',
     'lower_if_to_cond_assign.cpp',
     'lower_instructions.cpp',
     'lower_jumps.cpp',
@@ -66,6 +67,7 @@ sources = [
     'opt_dead_code.cpp',
     'opt_dead_code_local.cpp',
     'opt_dead_functions.cpp',
+    'opt_discard_simplification.cpp',
     'opt_function_inlining.cpp',
     'opt_if_simplification.cpp',
     'opt_noop_swizzle.cpp',
diff --git a/src/glsl/ast_to_hir.cpp b/src/glsl/ast_to_hir.cpp
index 04b221e9b8d..f5b1120f785 100644
--- a/src/glsl/ast_to_hir.cpp
+++ b/src/glsl/ast_to_hir.cpp
@@ -745,6 +745,16 @@ ast_node::hir(exec_list *instructions,
    return NULL;
 }
 
+static void
+mark_whole_array_access(ir_rvalue *access)
+{
+   ir_dereference_variable *deref = access->as_dereference_variable();
+
+   if (deref) {
+      deref->var->max_array_access = deref->type->length - 1;
+   }
+}
+
 static ir_rvalue *
 do_comparison(void *mem_ctx, int operation, ir_rvalue *op0, ir_rvalue *op1)
 {
@@ -780,6 +790,10 @@ do_comparison(void *mem_ctx, int operation, ir_rvalue *op0, ir_rvalue *op1)
 	    last = result;
 	 }
       }
+
+      mark_whole_array_access(op0);
+      mark_whole_array_access(op1);
+
       return last;
    }
 
diff --git a/src/glsl/glsl_parser_extras.cpp b/src/glsl/glsl_parser_extras.cpp
index 302cfbc5668..8dbe66927dc 100644
--- a/src/glsl/glsl_parser_extras.cpp
+++ b/src/glsl/glsl_parser_extras.cpp
@@ -716,6 +716,7 @@ do_common_optimization(exec_list *ir, bool linked, unsigned max_unroll_iteration
    }
    progress = do_structure_splitting(ir) || progress;
    progress = do_if_simplification(ir) || progress;
+   progress = do_discard_simplification(ir) || progress;
    progress = do_copy_propagation(ir) || progress;
    if (linked)
       progress = do_dead_code(ir) || progress;
diff --git a/src/glsl/ir_optimization.h b/src/glsl/ir_optimization.h
index fa497a45551..f264265f4b1 100644
--- a/src/glsl/ir_optimization.h
+++ b/src/glsl/ir_optimization.h
@@ -32,8 +32,9 @@
 #define SUB_TO_ADD_NEG 0x01
 #define DIV_TO_MUL_RCP 0x02
 #define EXP_TO_EXP2    0x04
-#define LOG_TO_LOG2    0x08
-#define MOD_TO_FRACT   0x10
+#define POW_TO_EXP2    0x08
+#define LOG_TO_LOG2    0x10
+#define MOD_TO_FRACT   0x20
 
 bool do_common_optimization(exec_list *ir, bool linked, unsigned max_unroll_iterations);
 
@@ -51,6 +52,7 @@ bool do_function_inlining(exec_list *instructions);
 bool do_lower_jumps(exec_list *instructions, bool pull_out_jumps = true, bool lower_sub_return = true, bool lower_main_return = false, bool lower_continue = false, bool lower_break = false);
 bool do_lower_texture_projection(exec_list *instructions);
 bool do_if_simplification(exec_list *instructions);
+bool do_discard_simplification(exec_list *instructions);
 bool do_if_to_cond_assign(exec_list *instructions);
 bool do_mat_op_to_vec(exec_list *instructions);
 bool do_mod_to_fract(exec_list *instructions);
@@ -61,6 +63,7 @@ bool do_swizzle_swizzle(exec_list *instructions);
 bool do_tree_grafting(exec_list *instructions);
 bool do_vec_index_to_cond_assign(exec_list *instructions);
 bool do_vec_index_to_swizzle(exec_list *instructions);
+bool lower_discard(exec_list *instructions);
 bool lower_instructions(exec_list *instructions, unsigned what_to_lower);
 bool lower_noise(exec_list *instructions);
 bool lower_variable_index_to_cond_assign(exec_list *instructions,
diff --git a/src/glsl/lower_discard.cpp b/src/glsl/lower_discard.cpp
new file mode 100644
index 00000000000..b95313df8c8
--- /dev/null
+++ b/src/glsl/lower_discard.cpp
@@ -0,0 +1,198 @@
+/*
+ * Copyright © 2010 Intel Corporation
+ *
+ * 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 (including the next
+ * paragraph) 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
+ * THE AUTHORS OR COPYRIGHT HOLDERS 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.
+ */
+
+/**
+ * \file lower_discard.cpp
+ *
+ * This pass moves discards out of if-statements.
+ *
+ * Case 1: The "then" branch contains a conditional discard:
+ * ---------------------------------------------------------
+ *
+ *    if (cond1) {
+ *	 s1;
+ *	 discard cond2;
+ *	 s2;
+ *    } else {
+ *	 s3;
+ *    }
+ *
+ * becomes:
+ *
+ *    temp = false;
+ *    if (cond1) {
+ *	 s1;
+ *	 temp = cond2;
+ *	 s2;
+ *    } else {
+ *	 s3;
+ *    }
+ *    discard temp;
+ *
+ * Case 2: The "else" branch contains a conditional discard:
+ * ---------------------------------------------------------
+ *
+ *    if (cond1) {
+ *	 s1;
+ *    } else {
+ *	 s2;
+ *	 discard cond2;
+ *	 s3;
+ *    }
+ *
+ * becomes:
+ *
+ *    temp = false;
+ *    if (cond1) {
+ *	 s1;
+ *    } else {
+ *	 s2;
+ *	 temp = cond2;
+ *	 s3;
+ *    }
+ *    discard temp;
+ *
+ * Case 3: Both branches contain a conditional discard:
+ * ----------------------------------------------------
+ *
+ *    if (cond1) {
+ *	 s1;
+ *	 discard cond2;
+ *	 s2;
+ *    } else {
+ *	 s3;
+ *	 discard cond3;
+ *	 s4;
+ *    }
+ *
+ * becomes:
+ *
+ *    temp = false;
+ *    if (cond1) {
+ *	 s1;
+ *	 temp = cond2;
+ *	 s2;
+ *    } else {
+ *	 s3;
+ *	 temp = cond3;
+ *	 s4;
+ *    }
+ *    discard temp;
+ *
+ * If there are multiple conditional discards, we need only deal with one of
+ * them.  Repeatedly applying this pass will take care of the others.
+ *
+ * Unconditional discards are treated as having a condition of "true".
+ */
+
+#include "glsl_types.h"
+#include "ir.h"
+
+class lower_discard_visitor : public ir_hierarchical_visitor {
+public:
+   lower_discard_visitor()
+   {
+      this->progress = false;
+   }
+
+   ir_visitor_status visit_leave(ir_if *);
+
+   bool progress;
+};
+
+
+bool
+lower_discard(exec_list *instructions)
+{
+   lower_discard_visitor v;
+
+   visit_list_elements(&v, instructions);
+
+   return v.progress;
+}
+
+
+static ir_discard *
+find_discard(exec_list &instructions)
+{
+   foreach_list(n, &instructions) {
+      ir_discard *ir = ((ir_instruction *) n)->as_discard();
+      if (ir != NULL)
+	 return ir;
+   }
+   return NULL;
+}
+
+
+static void
+replace_discard(void *mem_ctx, ir_variable *var, ir_discard *ir)
+{
+   ir_rvalue *condition = ir->condition;
+
+   /* For unconditional discards, use "true" as the condition. */
+   if (condition == NULL)
+      condition = new(mem_ctx) ir_constant(true);
+
+   ir_assignment *assignment =
+      new(mem_ctx) ir_assignment(new(mem_ctx) ir_dereference_variable(var),
+				 condition, NULL);
+
+   ir->replace_with(assignment);
+}
+
+
+ir_visitor_status
+lower_discard_visitor::visit_leave(ir_if *ir)
+{
+   ir_discard *then_discard = find_discard(ir->then_instructions);
+   ir_discard *else_discard = find_discard(ir->else_instructions);
+
+   if (then_discard == NULL && else_discard == NULL)
+      return visit_continue;
+
+   void *mem_ctx = talloc_parent(ir);
+
+   ir_variable *temp = new(mem_ctx) ir_variable(glsl_type::bool_type,
+						"discard_cond_temp",
+						ir_var_temporary);
+   ir_assignment *temp_initializer =
+      new(mem_ctx) ir_assignment(new(mem_ctx) ir_dereference_variable(temp),
+				 new(mem_ctx) ir_constant(false), NULL);
+
+   ir->insert_before(temp);
+   ir->insert_before(temp_initializer);
+
+   if (then_discard != NULL)
+      replace_discard(mem_ctx, temp, then_discard);
+
+   if (else_discard != NULL)
+      replace_discard(mem_ctx, temp, else_discard);
+
+   ir_discard *discard = then_discard != NULL ? then_discard : else_discard;
+   discard->condition = new(mem_ctx) ir_dereference_variable(temp);
+   ir->insert_after(discard);
+
+   this->progress = true;
+
+   return visit_continue;
+}
diff --git a/src/glsl/lower_instructions.cpp b/src/glsl/lower_instructions.cpp
index d460ba1a97a..a5f61f213d0 100644
--- a/src/glsl/lower_instructions.cpp
+++ b/src/glsl/lower_instructions.cpp
@@ -33,6 +33,7 @@
  * - SUB_TO_ADD_NEG
  * - DIV_TO_MUL_RCP
  * - EXP_TO_EXP2
+ * - POW_TO_EXP2
  * - LOG_TO_LOG2
  * - MOD_TO_FRACT
  *
@@ -61,6 +62,11 @@
  * do have base 2 versions, so this pass converts exp and log to exp2
  * and log2 operations.
  *
+ * POW_TO_EXP2:
+ * -----------
+ * Many older GPUs don't have an x**y instruction.  For these GPUs, convert
+ * x**y to 2**(y * log2(x)).
+ *
  * MOD_TO_FRACT:
  * -------------
  * Breaks an ir_unop_mod expression down to (op1 * fract(op0 / op1))
@@ -70,7 +76,7 @@
  * opportunity to do things like constant fold the (1.0 / op1) easily.
  */
 
-#include "main/core.h" /* for M_E */
+#include "main/core.h" /* for M_LOG2E */
 #include "glsl_types.h"
 #include "ir.h"
 #include "ir_optimization.h"
@@ -91,6 +97,7 @@ private:
    void div_to_mul_rcp(ir_expression *);
    void mod_to_fract(ir_expression *);
    void exp_to_exp2(ir_expression *);
+   void pow_to_exp2(ir_expression *);
    void log_to_log2(ir_expression *);
 };
 
@@ -172,7 +179,7 @@ lower_instructions_visitor::div_to_mul_rcp(ir_expression *ir)
 void
 lower_instructions_visitor::exp_to_exp2(ir_expression *ir)
 {
-   ir_constant *log2_e = new(ir) ir_constant(log2f(M_E));
+   ir_constant *log2_e = new(ir) ir_constant(float(M_LOG2E));
 
    ir->operation = ir_unop_exp2;
    ir->operands[0] = new(ir) ir_expression(ir_binop_mul, ir->operands[0]->type,
@@ -181,12 +188,26 @@ lower_instructions_visitor::exp_to_exp2(ir_expression *ir)
 }
 
 void
+lower_instructions_visitor::pow_to_exp2(ir_expression *ir)
+{
+   ir_expression *const log2_x =
+      new(ir) ir_expression(ir_unop_log2, ir->operands[0]->type,
+			    ir->operands[0]);
+
+   ir->operation = ir_unop_exp2;
+   ir->operands[0] = new(ir) ir_expression(ir_binop_mul, ir->operands[1]->type,
+					   ir->operands[1], log2_x);
+   ir->operands[1] = NULL;
+   this->progress = true;
+}
+
+void
 lower_instructions_visitor::log_to_log2(ir_expression *ir)
 {
    ir->operation = ir_binop_mul;
    ir->operands[0] = new(ir) ir_expression(ir_unop_log2, ir->operands[0]->type,
 					   ir->operands[0], NULL);
-   ir->operands[1] = new(ir) ir_constant(1.0f / log2f(M_E));
+   ir->operands[1] = new(ir) ir_constant(float(1.0 / M_LOG2E));
    this->progress = true;
 }
 
@@ -254,6 +275,11 @@ lower_instructions_visitor::visit_leave(ir_expression *ir)
 	 mod_to_fract(ir);
       break;
 
+   case ir_binop_pow:
+      if (lowering(POW_TO_EXP2))
+	 pow_to_exp2(ir);
+      break;
+
    default:
       return visit_continue;
    }
diff --git a/src/glsl/lower_jumps.cpp b/src/glsl/lower_jumps.cpp
index e1e7a5b0073..9cd15ef7366 100644
--- a/src/glsl/lower_jumps.cpp
+++ b/src/glsl/lower_jumps.cpp
@@ -23,6 +23,37 @@
 
 /**
  * \file lower_jumps.cpp
+ *
+ * This pass lowers jumps (break, continue, and return) to if/else structures.
+ *
+ * It can be asked to:
+ * 1. Pull jumps out of ifs where possible
+ * 2. Remove all "continue"s, replacing them with an "execute flag"
+ * 3. Replace all "break" with a single conditional one at the end of the loop
+ * 4. Replace all "return"s with a single return at the end of the function,
+ *    for the main function and/or other functions
+ *
+ * Applying this pass gives several benefits:
+ * 1. All functions can be inlined.
+ * 2. nv40 and other pre-DX10 chips without "continue" can be supported
+ * 3. nv30 and other pre-DX10 chips with no control flow at all are better
+ *    supported
+ *
+ * Continues are lowered by adding a per-loop "execute flag", initialized to
+ * true, that when cleared inhibits all execution until the end of the loop.
+ *
+ * Breaks are lowered to continues, plus setting a "break flag" that is checked
+ * at the end of the loop, and trigger the unique "break".
+ *
+ * Returns are lowered to breaks/continues, plus adding a "return flag" that
+ * causes loops to break again out of their enclosing loops until all the
+ * loops are exited: then the "execute flag" logic will ignore everything
+ * until the end of the function.
+ *
+ * Note that "continue" and "return" can also be implemented by adding
+ * a dummy loop and using break.
+ * However, this is bad for hardware with limited nesting depth, and
+ * prevents further optimization, and thus is not currently performed.
  */
 
 #include "glsl_types.h"
@@ -36,7 +67,6 @@ enum jump_strength
    strength_continue,
    strength_break,
    strength_return,
-   strength_discard
 };
 
 struct block_record
@@ -202,8 +232,6 @@ struct ir_lower_jumps_visitor : public ir_control_flow_visitor {
 
    virtual void visit(class ir_discard * ir)
    {
-      truncate_after_instruction(ir);
-      this->block.min_strength = strength_discard;
    }
 
    enum jump_strength get_jump_strength(ir_instruction* ir)
@@ -217,8 +245,6 @@ struct ir_lower_jumps_visitor : public ir_control_flow_visitor {
             return strength_continue;
       } else if(ir->ir_type == ir_type_return)
          return strength_return;
-      else if(ir->ir_type == ir_type_discard)
-         return strength_discard;
       else
          return strength_none;
    }
@@ -253,9 +279,6 @@ struct ir_lower_jumps_visitor : public ir_control_flow_visitor {
          else
             lower = lower_sub_return;
          break;
-      case strength_discard:
-         lower = false; /* probably nothing needs this lowered */
-         break;
       }
       return lower;
    }
@@ -313,9 +336,8 @@ retry: /* we get here if we put code after the if inside a branch */
             /* FINISHME: unify returns with identical expressions */
             else if(jump_strengths[0] == strength_return && this->function.signature->return_type->is_void())
                ir->insert_after(new(ir) ir_return(NULL));
-            /* FINISHME: unify discards */
-            else
-               unify = false;
+	    else
+	       unify = false;
 
             if(unify) {
                jumps[0]->remove();
diff --git a/src/glsl/opt_discard_simplification.cpp b/src/glsl/opt_discard_simplification.cpp
new file mode 100644
index 00000000000..0e577c478a9
--- /dev/null
+++ b/src/glsl/opt_discard_simplification.cpp
@@ -0,0 +1,180 @@
+/*
+ * Copyright © 2010 Intel Corporation
+ *
+ * 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 (including the next
+ * paragraph) 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
+ * THE AUTHORS OR COPYRIGHT HOLDERS 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.
+ */
+
+/**
+ * \file opt_discard_simplification.cpp
+ *
+ * This pass simplifies if-statements and loops containing unconditional
+ * discards.
+ *
+ * Case 1: Both branches contain unconditional discards:
+ * -----------------------------------------------------
+ *
+ *    if (cond) {
+ *	 s1;
+ *	 discard;
+ *	 s2;
+ *    } else {
+ *	 s3;
+ *	 discard;
+ *	 s4;
+ *    }
+ *
+ * becomes:
+ *
+ *    discard
+ *
+ * Case 2: The "then" clause contains an unconditional discard:
+ * ------------------------------------------------------------
+ *
+ *    if (cond) {
+ *       s1;
+ *       discard;
+ *       s2;
+ *    } else {
+ *	 s3;
+ *    }
+ *
+ * becomes:
+ *
+ *    if (cond) {
+ *	 discard;
+ *    } else {
+ *	 s3;
+ *    }
+ *
+ * Case 3: The "else" clause contains an unconditional discard:
+ * ------------------------------------------------------------
+ *
+ *    if (cond) {
+ *       s1;
+ *    } else {
+ *       s2;
+ *       discard;
+ *	 s3;
+ *    }
+ *
+ * becomes:
+ *
+ *    if (cond) {
+ *	 s1;
+ *    } else {
+ *	 discard;
+ *    }
+ */
+
+#include "glsl_types.h"
+#include "ir.h"
+
+class discard_simplifier : public ir_hierarchical_visitor {
+public:
+   discard_simplifier()
+   {
+      this->progress = false;
+   }
+
+   ir_visitor_status visit_enter(ir_if *);
+   ir_visitor_status visit_enter(ir_loop *);
+
+   bool progress;
+};
+
+static ir_discard *
+find_unconditional_discard(exec_list &instructions)
+{
+   foreach_list(n, &instructions) {
+      ir_discard *ir = ((ir_instruction *) n)->as_discard();
+      if (ir != NULL && ir->condition == NULL)
+	 return ir;
+   }
+   return NULL;
+}
+
+static bool
+is_only_instruction(ir_discard *discard)
+{
+   return (discard->prev->is_head_sentinel() &&
+	   discard->next->is_tail_sentinel());
+}
+
+ir_visitor_status
+discard_simplifier::visit_enter(ir_if *ir)
+{
+   ir_discard *then_discard = find_unconditional_discard(ir->then_instructions);
+   ir_discard *else_discard = find_unconditional_discard(ir->else_instructions);
+
+   if (then_discard == NULL && else_discard == NULL)
+      return visit_continue;
+
+   /* If both branches result in discard, replace whole if with discard. */
+   if (then_discard != NULL && else_discard != NULL) {
+      this->progress = true;
+      ir->replace_with(then_discard);
+      return visit_continue_with_parent;
+   }
+
+   /* Otherwise, one branch has a discard. */
+   if (then_discard != NULL && !is_only_instruction(then_discard)) {
+      this->progress = true;
+      ir->then_instructions.make_empty();
+      ir->then_instructions.push_tail(then_discard);
+   } else if (else_discard != NULL && !is_only_instruction(else_discard)) {
+      this->progress = true;
+      ir->else_instructions.make_empty();
+      ir->else_instructions.push_tail(else_discard);
+   }
+
+   visit_list_elements(this, &ir->then_instructions);
+   return visit_continue_with_parent;
+}
+
+ir_visitor_status
+discard_simplifier::visit_enter(ir_loop *ir)
+{
+   ir_discard *discard = find_unconditional_discard(ir->body_instructions);
+
+   if (discard) {
+      ir->replace_with(discard);
+      return visit_continue_with_parent;
+   }
+
+   return visit_continue;
+}
+
+bool
+do_discard_simplification(exec_list *instructions)
+{
+   /* Look for a top-level unconditional discard */
+   ir_discard *discard = find_unconditional_discard(*instructions);
+   if (discard != NULL) {
+      instructions->make_empty();
+      instructions->push_tail(discard);
+      return true;
+   }
+
+   discard_simplifier v;
+
+   visit_list_elements(&v, instructions);
+
+   return v.progress;
+}
diff --git a/src/mesa/drivers/dri/i915/i915_context.c b/src/mesa/drivers/dri/i915/i915_context.c
index f943f81dd05..f32f3cf6020 100644
--- a/src/mesa/drivers/dri/i915/i915_context.c
+++ b/src/mesa/drivers/dri/i915/i915_context.c
@@ -176,6 +176,7 @@ i915CreateContext(int api,
    ctx->ShaderCompilerOptions[MESA_SHADER_VERTEX].EmitCondCodes = GL_TRUE;
    ctx->ShaderCompilerOptions[MESA_SHADER_FRAGMENT].EmitNoIfs = GL_TRUE;
    ctx->ShaderCompilerOptions[MESA_SHADER_FRAGMENT].EmitNoNoise = GL_TRUE;
+   ctx->ShaderCompilerOptions[MESA_SHADER_FRAGMENT].EmitNoPow = GL_TRUE;
 
    ctx->Const.MaxDrawBuffers = 1;
 
diff --git a/src/mesa/drivers/dri/i915/i915_fragprog.c b/src/mesa/drivers/dri/i915/i915_fragprog.c
index c00ee415b6b..7a9fb7f088b 100644
--- a/src/mesa/drivers/dri/i915/i915_fragprog.c
+++ b/src/mesa/drivers/dri/i915/i915_fragprog.c
@@ -569,10 +569,14 @@ upload_program(struct i915_fragment_program *p)
 	 if (inst->DstReg.CondMask == COND_TR) {
 	    tmp = i915_get_utemp(p);
 
+	    /* The KIL instruction discards the fragment if any component of
+	     * the source is < 0.  Emit an immediate operand of {-1}.xywz.
+	     */
 	    i915_emit_texld(p, get_live_regs(p, inst),
 			    tmp, A0_DEST_CHANNEL_ALL,
 			    0, /* use a dummy dest reg */
-			    swizzle(tmp, ONE, ONE, ONE, ONE), /* always */
+			    negate(swizzle(tmp, ONE, ONE, ONE, ONE),
+				   1, 1, 1, 1),
 			    T0_TEXKILL);
 	 } else {
 	    p->error = 1;
diff --git a/src/mesa/drivers/dri/i965/brw_context.c b/src/mesa/drivers/dri/i965/brw_context.c
index cb0a8b96c9c..28549f2574a 100644
--- a/src/mesa/drivers/dri/i965/brw_context.c
+++ b/src/mesa/drivers/dri/i965/brw_context.c
@@ -122,9 +122,6 @@ GLboolean brwCreateContext( int api,
 	 (i == MESA_SHADER_FRAGMENT);
       ctx->ShaderCompilerOptions[i].EmitNoIndirectTemp =
 	 (i == MESA_SHADER_FRAGMENT);
-
-      if (intel->gen == 6)
-	 ctx->ShaderCompilerOptions[i].EmitNoIfs = (i == MESA_SHADER_VERTEX);
    }
 
    ctx->Const.VertexProgram.MaxNativeInstructions = (16 * 1024);
diff --git a/src/mesa/drivers/dri/i965/brw_disasm.c b/src/mesa/drivers/dri/i965/brw_disasm.c
index 962c04128b8..6b61f7af15d 100644
--- a/src/mesa/drivers/dri/i965/brw_disasm.c
+++ b/src/mesa/drivers/dri/i965/brw_disasm.c
@@ -899,7 +899,8 @@ int brw_disasm (FILE *file, struct brw_instruction *inst, int gen)
 	err |= dest (file, inst);
     } else if (gen >= 6 && (inst->header.opcode == BRW_OPCODE_IF ||
 			    inst->header.opcode == BRW_OPCODE_ELSE ||
-			    inst->header.opcode == BRW_OPCODE_ENDIF)) {
+			    inst->header.opcode == BRW_OPCODE_ENDIF ||
+			    inst->header.opcode == BRW_OPCODE_WHILE)) {
        format (file, " %d", inst->bits1.branch_gen6.jump_count);
     }
 
diff --git a/src/mesa/drivers/dri/i965/brw_eu.h b/src/mesa/drivers/dri/i965/brw_eu.h
index b4538e6e8a7..a4904b7098a 100644
--- a/src/mesa/drivers/dri/i965/brw_eu.h
+++ b/src/mesa/drivers/dri/i965/brw_eu.h
@@ -954,6 +954,8 @@ struct brw_instruction *brw_WHILE(struct brw_compile *p,
 	       struct brw_instruction *patch_insn);
 
 struct brw_instruction *brw_BREAK(struct brw_compile *p, int pop_count);
+struct brw_instruction *brw_CONT_gen6(struct brw_compile *p,
+				      struct brw_instruction *do_insn);
 struct brw_instruction *brw_CONT(struct brw_compile *p, int pop_count);
 /* Forward jumps:
  */
@@ -1009,6 +1011,7 @@ void brw_math_invert( struct brw_compile *p,
 void brw_set_src1( struct brw_instruction *insn,
                           struct brw_reg reg );
 
+void brw_set_uip_jip(struct brw_compile *p);
 
 /* brw_optimize.c */
 void brw_optimize(struct brw_compile *p);
diff --git a/src/mesa/drivers/dri/i965/brw_eu_emit.c b/src/mesa/drivers/dri/i965/brw_eu_emit.c
index 9cb941dacfd..945f50d1106 100644
--- a/src/mesa/drivers/dri/i965/brw_eu_emit.c
+++ b/src/mesa/drivers/dri/i965/brw_eu_emit.c
@@ -1029,16 +1029,44 @@ void brw_ENDIF(struct brw_compile *p,
 
 struct brw_instruction *brw_BREAK(struct brw_compile *p, int pop_count)
 {
+   struct intel_context *intel = &p->brw->intel;
    struct brw_instruction *insn;
+
    insn = next_insn(p, BRW_OPCODE_BREAK);
+   if (intel->gen >= 6) {
+      brw_set_dest(insn, retype(brw_null_reg(), BRW_REGISTER_TYPE_D));
+      brw_set_src0(insn, retype(brw_null_reg(), BRW_REGISTER_TYPE_D));
+      brw_set_src1(insn, brw_imm_d(0x0));
+   } else {
+      brw_set_dest(insn, brw_ip_reg());
+      brw_set_src0(insn, brw_ip_reg());
+      brw_set_src1(insn, brw_imm_d(0x0));
+      insn->bits3.if_else.pad0 = 0;
+      insn->bits3.if_else.pop_count = pop_count;
+   }
+   insn->header.compression_control = BRW_COMPRESSION_NONE;
+   insn->header.execution_size = BRW_EXECUTE_8;
+
+   return insn;
+}
+
+struct brw_instruction *brw_CONT_gen6(struct brw_compile *p,
+				      struct brw_instruction *do_insn)
+{
+   struct brw_instruction *insn;
+   int br = 2;
+
+   insn = next_insn(p, BRW_OPCODE_CONTINUE);
+   brw_set_dest(insn, retype(brw_null_reg(), BRW_REGISTER_TYPE_D));
+   brw_set_src0(insn, retype(brw_null_reg(), BRW_REGISTER_TYPE_D));
    brw_set_dest(insn, brw_ip_reg());
    brw_set_src0(insn, brw_ip_reg());
    brw_set_src1(insn, brw_imm_d(0x0));
+
+   insn->bits3.break_cont.uip = br * (do_insn - insn);
+
    insn->header.compression_control = BRW_COMPRESSION_NONE;
    insn->header.execution_size = BRW_EXECUTE_8;
-   /* insn->header.mask_control = BRW_MASK_DISABLE; */
-   insn->bits3.if_else.pad0 = 0;
-   insn->bits3.if_else.pop_count = pop_count;
    return insn;
 }
 
@@ -1058,10 +1086,26 @@ struct brw_instruction *brw_CONT(struct brw_compile *p, int pop_count)
 }
 
 /* DO/WHILE loop:
+ *
+ * The DO/WHILE is just an unterminated loop -- break or continue are
+ * used for control within the loop.  We have a few ways they can be
+ * done.
+ *
+ * For uniform control flow, the WHILE is just a jump, so ADD ip, ip,
+ * jip and no DO instruction.
+ *
+ * For non-uniform control flow pre-gen6, there's a DO instruction to
+ * push the mask, and a WHILE to jump back, and BREAK to get out and
+ * pop the mask.
+ *
+ * For gen6, there's no more mask stack, so no need for DO.  WHILE
+ * just points back to the first instruction of the loop.
  */
 struct brw_instruction *brw_DO(struct brw_compile *p, GLuint execute_size)
 {
-   if (p->single_program_flow) {
+   struct intel_context *intel = &p->brw->intel;
+
+   if (intel->gen >= 6 || p->single_program_flow) {
       return &p->store[p->nr_insn];
    } else {
       struct brw_instruction *insn = next_insn(p, BRW_OPCODE_DO);
@@ -1094,34 +1138,42 @@ struct brw_instruction *brw_WHILE(struct brw_compile *p,
    if (intel->gen >= 5)
       br = 2;
 
-   if (p->single_program_flow)
-      insn = next_insn(p, BRW_OPCODE_ADD);
-   else
+   if (intel->gen >= 6) {
       insn = next_insn(p, BRW_OPCODE_WHILE);
 
-   brw_set_dest(insn, brw_ip_reg());
-   brw_set_src0(insn, brw_ip_reg());
-   brw_set_src1(insn, brw_imm_d(0x0));
+      brw_set_dest(insn, brw_imm_w(0));
+      insn->bits1.branch_gen6.jump_count = br * (do_insn - insn);
+      brw_set_src0(insn, retype(brw_null_reg(), BRW_REGISTER_TYPE_D));
+      brw_set_src1(insn, retype(brw_null_reg(), BRW_REGISTER_TYPE_D));
 
-   insn->header.compression_control = BRW_COMPRESSION_NONE;
+      insn->header.execution_size = do_insn->header.execution_size;
+      assert(insn->header.execution_size == BRW_EXECUTE_8);
+   } else {
+      if (p->single_program_flow) {
+	 insn = next_insn(p, BRW_OPCODE_ADD);
 
-   if (p->single_program_flow) {
-      insn->header.execution_size = BRW_EXECUTE_1;
+	 brw_set_dest(insn, brw_ip_reg());
+	 brw_set_src0(insn, brw_ip_reg());
+	 brw_set_src1(insn, brw_imm_d((do_insn - insn) * 16));
+	 insn->header.execution_size = BRW_EXECUTE_1;
+      } else {
+	 insn = next_insn(p, BRW_OPCODE_WHILE);
 
-      insn->bits3.d = (do_insn - insn) * 16;
-   } else {
-      insn->header.execution_size = do_insn->header.execution_size;
+	 assert(do_insn->header.opcode == BRW_OPCODE_DO);
 
-      assert(do_insn->header.opcode == BRW_OPCODE_DO);
-      insn->bits3.if_else.jump_count = br * (do_insn - insn + 1);
-      insn->bits3.if_else.pop_count = 0;
-      insn->bits3.if_else.pad0 = 0;
-   }
+	 brw_set_dest(insn, brw_ip_reg());
+	 brw_set_src0(insn, brw_ip_reg());
+	 brw_set_src1(insn, brw_imm_d(0));
 
-/*    insn->header.mask_control = BRW_MASK_ENABLE; */
+	 insn->header.execution_size = do_insn->header.execution_size;
+	 insn->bits3.if_else.jump_count = br * (do_insn - insn + 1);
+	 insn->bits3.if_else.pop_count = 0;
+	 insn->bits3.if_else.pad0 = 0;
+      }
+   }
+   insn->header.compression_control = BRW_COMPRESSION_NONE;
+   p->current->header.predicate_control = BRW_PREDICATE_NONE;
 
-   /* insn->header.mask_control = BRW_MASK_DISABLE; */
-   p->current->header.predicate_control = BRW_PREDICATE_NONE;   
    return insn;
 }
 
@@ -1989,6 +2041,80 @@ void brw_urb_WRITE(struct brw_compile *p,
 		       swizzle);
 }
 
+static int
+brw_find_next_block_end(struct brw_compile *p, int start)
+{
+   int ip;
+
+   for (ip = start + 1; ip < p->nr_insn; ip++) {
+      struct brw_instruction *insn = &p->store[ip];
+
+      switch (insn->header.opcode) {
+      case BRW_OPCODE_ENDIF:
+      case BRW_OPCODE_ELSE:
+      case BRW_OPCODE_WHILE:
+	 return ip;
+      }
+   }
+   assert(!"not reached");
+   return start + 1;
+}
+
+/* There is no DO instruction on gen6, so to find the end of the loop
+ * we have to see if the loop is jumping back before our start
+ * instruction.
+ */
+static int
+brw_find_loop_end(struct brw_compile *p, int start)
+{
+   int ip;
+   int br = 2;
+
+   for (ip = start + 1; ip < p->nr_insn; ip++) {
+      struct brw_instruction *insn = &p->store[ip];
+
+      if (insn->header.opcode == BRW_OPCODE_WHILE) {
+	 if (ip + insn->bits1.branch_gen6.jump_count / br < start)
+	    return ip;
+      }
+   }
+   assert(!"not reached");
+   return start + 1;
+}
+
+/* After program generation, go back and update the UIP and JIP of
+ * BREAK and CONT instructions to their correct locations.
+ */
+void
+brw_set_uip_jip(struct brw_compile *p)
+{
+   struct intel_context *intel = &p->brw->intel;
+   int ip;
+   int br = 2;
+
+   if (intel->gen < 6)
+      return;
+
+   for (ip = 0; ip < p->nr_insn; ip++) {
+      struct brw_instruction *insn = &p->store[ip];
+
+      switch (insn->header.opcode) {
+      case BRW_OPCODE_BREAK:
+	 insn->bits3.break_cont.jip = br * (brw_find_next_block_end(p, ip) - ip);
+	 insn->bits3.break_cont.uip = br * (brw_find_loop_end(p, ip) - ip + 1);
+	 break;
+      case BRW_OPCODE_CONTINUE:
+	 /* JIP is set at CONTINUE emit time, since that's when we
+	  * know where the start of the loop is.
+	  */
+	 insn->bits3.break_cont.jip = br * (brw_find_next_block_end(p, ip) - ip);
+	 assert(insn->bits3.break_cont.uip != 0);
+	 assert(insn->bits3.break_cont.jip != 0);
+	 break;
+      }
+   }
+}
+
 void brw_ff_sync(struct brw_compile *p,
 		   struct brw_reg dest,
 		   GLuint msg_reg_nr,
diff --git a/src/mesa/drivers/dri/i965/brw_fs.cpp b/src/mesa/drivers/dri/i965/brw_fs.cpp
index 7a8e9812257..ee9ae160bdb 100644
--- a/src/mesa/drivers/dri/i965/brw_fs.cpp
+++ b/src/mesa/drivers/dri/i965/brw_fs.cpp
@@ -933,6 +933,10 @@ fs_visitor::visit(ir_expression *ir)
       assert(!"not reached: should be handled by lower_noise");
       break;
 
+   case ir_quadop_vector:
+      assert(!"not reached: should be handled by lower_quadop_vector");
+      break;
+
    case ir_unop_sqrt:
       emit_math(FS_OPCODE_SQRT, this->result, op[0]);
       break;
@@ -3375,10 +3379,6 @@ fs_visitor::generate_code()
 	 break;
 
       case BRW_OPCODE_DO:
-	 /* FINISHME: We need to write the loop instruction support still. */
-	 if (intel->gen >= 6)
-	    this->fail = true;
-
 	 loop_stack[loop_stack_depth++] = brw_DO(p, BRW_EXECUTE_8);
 	 if_depth_in_loop[loop_stack_depth] = 0;
 	 break;
@@ -3388,7 +3388,11 @@ fs_visitor::generate_code()
 	 brw_set_predicate_control(p, BRW_PREDICATE_NONE);
 	 break;
       case BRW_OPCODE_CONTINUE:
-	 brw_CONT(p, if_depth_in_loop[loop_stack_depth]);
+	 /* FINISHME: We need to write the loop instruction support still. */
+	 if (intel->gen >= 6)
+	    brw_CONT_gen6(p, loop_stack[loop_stack_depth - 1]);
+	 else
+	    brw_CONT(p, if_depth_in_loop[loop_stack_depth]);
 	 brw_set_predicate_control(p, BRW_PREDICATE_NONE);
 	 break;
 
@@ -3402,16 +3406,18 @@ fs_visitor::generate_code()
 	 assert(loop_stack_depth > 0);
 	 loop_stack_depth--;
 	 inst0 = inst1 = brw_WHILE(p, loop_stack[loop_stack_depth]);
-	 /* patch all the BREAK/CONT instructions from last BGNLOOP */
-	 while (inst0 > loop_stack[loop_stack_depth]) {
-	    inst0--;
-	    if (inst0->header.opcode == BRW_OPCODE_BREAK &&
-		inst0->bits3.if_else.jump_count == 0) {
-	       inst0->bits3.if_else.jump_count = br * (inst1 - inst0 + 1);
+	 if (intel->gen < 6) {
+	    /* patch all the BREAK/CONT instructions from last BGNLOOP */
+	    while (inst0 > loop_stack[loop_stack_depth]) {
+	       inst0--;
+	       if (inst0->header.opcode == BRW_OPCODE_BREAK &&
+		   inst0->bits3.if_else.jump_count == 0) {
+		  inst0->bits3.if_else.jump_count = br * (inst1 - inst0 + 1);
 	    }
-	    else if (inst0->header.opcode == BRW_OPCODE_CONTINUE &&
-		     inst0->bits3.if_else.jump_count == 0) {
-	       inst0->bits3.if_else.jump_count = br * (inst1 - inst0);
+	       else if (inst0->header.opcode == BRW_OPCODE_CONTINUE &&
+			inst0->bits3.if_else.jump_count == 0) {
+		  inst0->bits3.if_else.jump_count = br * (inst1 - inst0);
+	       }
 	    }
 	 }
       }
@@ -3488,6 +3494,26 @@ fs_visitor::generate_code()
 
       last_native_inst = p->nr_insn;
    }
+
+   brw_set_uip_jip(p);
+
+   /* OK, while the INTEL_DEBUG=wm above is very nice for debugging FS
+    * emit issues, it doesn't get the jump distances into the output,
+    * which is often something we want to debug.  So this is here in
+    * case you're doing that.
+    */
+   if (0) {
+      if (unlikely(INTEL_DEBUG & DEBUG_WM)) {
+	 for (unsigned int i = 0; i < p->nr_insn; i++) {
+	    printf("0x%08x 0x%08x 0x%08x 0x%08x ",
+		   ((uint32_t *)&p->store[i])[3],
+		   ((uint32_t *)&p->store[i])[2],
+		   ((uint32_t *)&p->store[i])[1],
+		   ((uint32_t *)&p->store[i])[0]);
+	    brw_disasm(stdout, &p->store[i], intel->gen);
+	 }
+      }
+   }
 }
 
 GLboolean
diff --git a/src/mesa/drivers/dri/i965/brw_structs.h b/src/mesa/drivers/dri/i965/brw_structs.h
index 8ce9af9c4fe..8f97bd136fd 100644
--- a/src/mesa/drivers/dri/i965/brw_structs.h
+++ b/src/mesa/drivers/dri/i965/brw_structs.h
@@ -1539,6 +1539,21 @@ struct brw_instruction
 	 GLuint  pad0:12;
       } if_else;
 
+      struct
+      {
+	 /* Signed jump distance to the ip to jump to if all channels
+	  * are disabled after the break or continue.  It should point
+	  * to the end of the innermost control flow block, as that's
+	  * where some channel could get re-enabled.
+	  */
+	 int jip:16;
+
+	 /* Signed jump distance to the location to resume execution
+	  * of this channel if it's enabled for the break or continue.
+	  */
+	 int uip:16;
+      } break_cont;
+
       struct {
 	 GLuint function:4;
 	 GLuint int_type:1;
diff --git a/src/mesa/drivers/dri/i965/brw_vs_emit.c b/src/mesa/drivers/dri/i965/brw_vs_emit.c
index b13e0c2a2c8..407358f498f 100644
--- a/src/mesa/drivers/dri/i965/brw_vs_emit.c
+++ b/src/mesa/drivers/dri/i965/brw_vs_emit.c
@@ -2032,35 +2032,42 @@ void brw_vs_emit(struct brw_vs_compile *c )
          break;
       case OPCODE_CONT:
 	 brw_set_predicate_control(p, get_predicate(inst));
-	 brw_CONT(p, if_depth_in_loop[loop_depth]);
+	 if (intel->gen >= 6) {
+	    brw_CONT_gen6(p, loop_inst[loop_depth - 1]);
+	 } else {
+	    brw_CONT(p, if_depth_in_loop[loop_depth]);
+	 }
          brw_set_predicate_control(p, BRW_PREDICATE_NONE);
          break;
-      case OPCODE_ENDLOOP: 
-         {
-	    clear_current_const(c);
-            struct brw_instruction *inst0, *inst1;
-	    GLuint br = 1;
-
-            loop_depth--;
-
-	    if (intel->gen == 5)
-	       br = 2;
-
-            inst0 = inst1 = brw_WHILE(p, loop_inst[loop_depth]);
-            /* patch all the BREAK/CONT instructions from last BEGINLOOP */
-            while (inst0 > loop_inst[loop_depth]) {
-               inst0--;
-               if (inst0->header.opcode == BRW_OPCODE_BREAK &&
+
+      case OPCODE_ENDLOOP: {
+	 clear_current_const(c);
+	 struct brw_instruction *inst0, *inst1;
+	 GLuint br = 1;
+
+	 loop_depth--;
+
+	 if (intel->gen == 5)
+	    br = 2;
+
+	 inst0 = inst1 = brw_WHILE(p, loop_inst[loop_depth]);
+
+	 if (intel->gen < 6) {
+	    /* patch all the BREAK/CONT instructions from last BEGINLOOP */
+	    while (inst0 > loop_inst[loop_depth]) {
+	       inst0--;
+	       if (inst0->header.opcode == BRW_OPCODE_BREAK &&
 		   inst0->bits3.if_else.jump_count == 0) {
-                  inst0->bits3.if_else.jump_count = br * (inst1 - inst0 + 1);
-               }
-               else if (inst0->header.opcode == BRW_OPCODE_CONTINUE &&
-			inst0->bits3.if_else.jump_count == 0) {
-                  inst0->bits3.if_else.jump_count = br * (inst1 - inst0);
-               }
-            }
-         }
+		  inst0->bits3.if_else.jump_count = br * (inst1 - inst0 + 1);
+	       } else if (inst0->header.opcode == BRW_OPCODE_CONTINUE &&
+			  inst0->bits3.if_else.jump_count == 0) {
+		  inst0->bits3.if_else.jump_count = br * (inst1 - inst0);
+	       }
+	    }
+	 }
+      }
          break;
+
       case OPCODE_BRA:
 	 brw_set_predicate_control(p, get_predicate(inst));
          brw_ADD(p, brw_ip_reg(), brw_ip_reg(), brw_imm_d(1*16));
@@ -2151,6 +2158,7 @@ void brw_vs_emit(struct brw_vs_compile *c )
    }
 
    brw_resolve_cals(p);
+   brw_set_uip_jip(p);
 
    brw_optimize(p);
 
diff --git a/src/mesa/drivers/dri/i965/gen6_sf_state.c b/src/mesa/drivers/dri/i965/gen6_sf_state.c
index 471067e8f02..06ac5d49a0f 100644
--- a/src/mesa/drivers/dri/i965/gen6_sf_state.c
+++ b/src/mesa/drivers/dri/i965/gen6_sf_state.c
@@ -99,6 +99,48 @@ upload_sf_state(struct brw_context *brw)
    if (ctx->Polygon.OffsetFill)
        dw2 |= GEN6_SF_GLOBAL_DEPTH_OFFSET_SOLID;
 
+   if (ctx->Polygon.OffsetLine)
+       dw2 |= GEN6_SF_GLOBAL_DEPTH_OFFSET_WIREFRAME;
+
+   if (ctx->Polygon.OffsetPoint)
+       dw2 |= GEN6_SF_GLOBAL_DEPTH_OFFSET_POINT;
+
+   switch (ctx->Polygon.FrontMode) {
+   case GL_FILL:
+       dw2 |= GEN6_SF_FRONT_SOLID;
+       break;
+
+   case GL_LINE:
+       dw2 |= GEN6_SF_FRONT_WIREFRAME;
+       break;
+
+   case GL_POINT:
+       dw2 |= GEN6_SF_FRONT_POINT;
+       break;
+
+   default:
+       assert(0);
+       break;
+   }
+
+   switch (ctx->Polygon.BackMode) {
+   case GL_FILL:
+       dw2 |= GEN6_SF_BACK_SOLID;
+       break;
+
+   case GL_LINE:
+       dw2 |= GEN6_SF_BACK_WIREFRAME;
+       break;
+
+   case GL_POINT:
+       dw2 |= GEN6_SF_BACK_POINT;
+       break;
+
+   default:
+       assert(0);
+       break;
+   }
+
    /* _NEW_SCISSOR */
    if (ctx->Scissor.Enabled)
       dw3 |= GEN6_SF_SCISSOR_ENABLE;
diff --git a/src/mesa/drivers/dri/r300/compiler/radeon_opcodes.c b/src/mesa/drivers/dri/r300/compiler/radeon_opcodes.c
index da495a3afaa..113b27632a2 100644
--- a/src/mesa/drivers/dri/r300/compiler/radeon_opcodes.c
+++ b/src/mesa/drivers/dri/r300/compiler/radeon_opcodes.c
@@ -67,6 +67,13 @@ struct rc_opcode_info rc_opcodes[MAX_RC_OPCODE] = {
 		.IsComponentwise = 1
 	},
 	{
+		.Opcode = RC_OPCODE_CLAMP,
+		.Name = "CLAMP",
+		.NumSrcRegs = 3,
+		.HasDstReg = 1,
+		.IsComponentwise = 1
+	},
+	{
 		.Opcode = RC_OPCODE_CMP,
 		.Name = "CMP",
 		.NumSrcRegs = 3,
diff --git a/src/mesa/drivers/dri/r300/compiler/radeon_opcodes.h b/src/mesa/drivers/dri/r300/compiler/radeon_opcodes.h
index d3f639c8701..7e666101276 100644
--- a/src/mesa/drivers/dri/r300/compiler/radeon_opcodes.h
+++ b/src/mesa/drivers/dri/r300/compiler/radeon_opcodes.h
@@ -50,6 +50,9 @@ typedef enum {
 	/** vec4 instruction: dst.c = ceil(src0.c) */
 	RC_OPCODE_CEIL,
 
+	/** vec4 instruction: dst.c = clamp(src0.c, src1.c, src2.c) */
+	RC_OPCODE_CLAMP,
+
 	/** vec4 instruction: dst.c = src0.c < 0.0 ? src1.c : src2.c */
 	RC_OPCODE_CMP,
 
diff --git a/src/mesa/drivers/dri/r300/compiler/radeon_program_alu.c b/src/mesa/drivers/dri/r300/compiler/radeon_program_alu.c
index 106e03495d8..01c2e74e7b3 100644
--- a/src/mesa/drivers/dri/r300/compiler/radeon_program_alu.c
+++ b/src/mesa/drivers/dri/r300/compiler/radeon_program_alu.c
@@ -217,6 +217,22 @@ static void transform_CEIL(struct radeon_compiler* c,
 	rc_remove_instruction(inst);
 }
 
+static void transform_CLAMP(struct radeon_compiler *c,
+	struct rc_instruction *inst)
+{
+	/* CLAMP dst, src, min, max
+	 *    into:
+	 * MIN tmp, src, max
+	 * MAX dst, tmp, min
+	 */
+	int tempreg = rc_find_free_temporary(c);
+	emit2(c, inst->Prev, RC_OPCODE_MIN, 0, dstreg(RC_FILE_TEMPORARY, tempreg),
+		inst->U.I.SrcReg[0], inst->U.I.SrcReg[2]);
+	emit2(c, inst->Prev, RC_OPCODE_MAX, inst->U.I.SaturateMode, inst->U.I.DstReg,
+		srcreg(RC_FILE_TEMPORARY, tempreg), inst->U.I.SrcReg[1]);
+	rc_remove_instruction(inst);
+}
+
 static void transform_DP2(struct radeon_compiler* c,
 	struct rc_instruction* inst)
 {
@@ -554,6 +570,7 @@ int radeonTransformALU(
 	switch(inst->U.I.Opcode) {
 	case RC_OPCODE_ABS: transform_ABS(c, inst); return 1;
 	case RC_OPCODE_CEIL: transform_CEIL(c, inst); return 1;
+	case RC_OPCODE_CLAMP: transform_CLAMP(c, inst); return 1;
 	case RC_OPCODE_DP2: transform_DP2(c, inst); return 1;
 	case RC_OPCODE_DPH: transform_DPH(c, inst); return 1;
 	case RC_OPCODE_DST: transform_DST(c, inst); return 1;
@@ -782,6 +799,7 @@ int r300_transform_vertex_alu(
 	switch(inst->U.I.Opcode) {
 	case RC_OPCODE_ABS: transform_r300_vertex_ABS(c, inst); return 1;
 	case RC_OPCODE_CEIL: transform_CEIL(c, inst); return 1;
+	case RC_OPCODE_CLAMP: transform_CLAMP(c, inst); return 1;
 	case RC_OPCODE_CMP: transform_r300_vertex_CMP(c, inst); return 1;
 	case RC_OPCODE_DP2: transform_r300_vertex_DP2(c, inst); return 1;
 	case RC_OPCODE_DP3: transform_r300_vertex_DP3(c, inst); return 1;
diff --git a/src/mesa/drivers/dri/r600/r700_assembler.c b/src/mesa/drivers/dri/r600/r700_assembler.c
index 2bf24096a0d..1fa559cec1a 100644
--- a/src/mesa/drivers/dri/r600/r700_assembler.c
+++ b/src/mesa/drivers/dri/r600/r700_assembler.c
@@ -3334,7 +3334,14 @@ GLboolean assemble_CMP(r700_AssemblerBase *pAsm)
         return GL_FALSE;
     }
 
-    pAsm->D.dst.opcode = SQ_OP3_INST_CNDGE;
+    if(8 == pAsm->unAsic)
+    {
+	pAsm->D.dst.opcode = EG_OP3_INST_CNDGE;
+    }
+    else
+    {
+	pAsm->D.dst.opcode = SQ_OP3_INST_CNDGE;
+    }
     pAsm->D.dst.op3     = 1;  
 
     tmp = (-1);
@@ -3416,8 +3423,14 @@ GLboolean assemble_TRIG(r700_AssemblerBase *pAsm, BITS opcode)
     checkop1(pAsm);
 
     tmp = gethelpr(pAsm);
-
-    pAsm->D.dst.opcode = SQ_OP3_INST_MULADD;
+    if(8 == pAsm->unAsic)
+    {
+        pAsm->D.dst.opcode = EG_OP3_INST_MULADD;
+    }
+    else
+    {
+        pAsm->D.dst.opcode = SQ_OP3_INST_MULADD;
+    }
     pAsm->D.dst.op3    = 1;
 
     setaddrmode_PVSDST(&(pAsm->D.dst), ADDR_ABSOLUTE);
@@ -3457,7 +3470,14 @@ GLboolean assemble_TRIG(r700_AssemblerBase *pAsm, BITS opcode)
     {
         return GL_FALSE;
     }
-    pAsm->D.dst.opcode = SQ_OP3_INST_MULADD;
+    if(8 == pAsm->unAsic)
+    {
+        pAsm->D.dst.opcode = EG_OP3_INST_MULADD;
+    }
+    else
+    {
+        pAsm->D.dst.opcode = SQ_OP3_INST_MULADD;
+    }
     pAsm->D.dst.op3    = 1;
 
     setaddrmode_PVSDST(&(pAsm->D.dst), ADDR_ABSOLUTE);
@@ -4742,7 +4762,14 @@ GLboolean assemble_SCS(r700_AssemblerBase *pAsm)
 
     tmp = gethelpr(pAsm);
 
-    pAsm->D.dst.opcode = SQ_OP3_INST_MULADD;
+    if(8 == pAsm->unAsic)
+    {
+        pAsm->D.dst.opcode = EG_OP3_INST_MULADD;
+    }
+    else
+    {
+        pAsm->D.dst.opcode = SQ_OP3_INST_MULADD;
+    }
     pAsm->D.dst.op3    = 1;
 
     setaddrmode_PVSDST(&(pAsm->D.dst), ADDR_ABSOLUTE);
@@ -4782,7 +4809,14 @@ GLboolean assemble_SCS(r700_AssemblerBase *pAsm)
     {
         return GL_FALSE;
     }
-    pAsm->D.dst.opcode = SQ_OP3_INST_MULADD;
+    if(8 == pAsm->unAsic)
+    {
+        pAsm->D.dst.opcode = EG_OP3_INST_MULADD;
+    }
+    else
+    {
+        pAsm->D.dst.opcode = SQ_OP3_INST_MULADD;
+    }
     pAsm->D.dst.op3    = 1;
 
     setaddrmode_PVSDST(&(pAsm->D.dst), ADDR_ABSOLUTE);
@@ -5010,7 +5044,14 @@ GLboolean assemble_SSG(r700_AssemblerBase *pAsm)
     
     GLuint tmp = gethelpr(pAsm);
     /* tmp = (src > 0 ? 1 : src) */
-    pAsm->D.dst.opcode = SQ_OP3_INST_CNDGT;
+    if(8 == pAsm->unAsic)
+    {
+        pAsm->D.dst.opcode = EG_OP3_INST_CNDGT;
+    }
+    else
+    {
+        pAsm->D.dst.opcode = SQ_OP3_INST_CNDGT;
+    }
     pAsm->D.dst.op3    = 1;
     pAsm->D.dst.rtype = DST_REG_TEMPORARY;
     pAsm->D.dst.reg   = tmp;
@@ -5033,7 +5074,14 @@ GLboolean assemble_SSG(r700_AssemblerBase *pAsm)
     }
 
     /* dst = (-tmp > 0 ? -1 : tmp) */
-    pAsm->D.dst.opcode = SQ_OP3_INST_CNDGT;
+    if(8 == pAsm->unAsic)
+    {
+        pAsm->D.dst.opcode = EG_OP3_INST_CNDGT;
+    }
+    else
+    {
+        pAsm->D.dst.opcode = SQ_OP3_INST_CNDGT;
+    }
     pAsm->D.dst.op3    = 1;
 
     if( GL_FALSE == assemble_dst(pAsm) )
diff --git a/src/mesa/main/compiler.h b/src/mesa/main/compiler.h
index 800eb839005..5557a3b5cb5 100644
--- a/src/mesa/main/compiler.h
+++ b/src/mesa/main/compiler.h
@@ -358,6 +358,10 @@ static INLINE GLuint CPU_TO_LE32(GLuint x)
 #define M_E (2.7182818284590452354)
 #endif
 
+#ifndef M_LOG2E
+#define M_LOG2E     (1.4426950408889634074)
+#endif
+
 #ifndef ONE_DIV_LN2
 #define ONE_DIV_LN2 (1.442695040888963456)
 #endif
diff --git a/src/mesa/main/mtypes.h b/src/mesa/main/mtypes.h
index 80c20e09d9a..82495714f21 100644
--- a/src/mesa/main/mtypes.h
+++ b/src/mesa/main/mtypes.h
@@ -2197,6 +2197,7 @@ struct gl_shader_compiler_options
    GLboolean EmitNoCont;                  /**< Emit CONT opcode? */
    GLboolean EmitNoMainReturn;            /**< Emit CONT/RET opcodes? */
    GLboolean EmitNoNoise;                 /**< Emit NOISE opcodes? */
+   GLboolean EmitNoPow;                   /**< Emit POW opcodes? */
 
    /**
     * \name Forms of indirect addressing the driver cannot do.
diff --git a/src/mesa/program/ir_to_mesa.cpp b/src/mesa/program/ir_to_mesa.cpp
index 8f75c82c3eb..b274a961b28 100644
--- a/src/mesa/program/ir_to_mesa.cpp
+++ b/src/mesa/program/ir_to_mesa.cpp
@@ -2166,9 +2166,14 @@ ir_to_mesa_visitor::visit(ir_discard *ir)
 {
    struct gl_fragment_program *fp = (struct gl_fragment_program *)this->prog;
 
-   assert(ir->condition == NULL); /* FINISHME */
+   if (ir->condition) {
+      ir->condition->accept(this);
+      this->result.negate = ~this->result.negate;
+      ir_to_mesa_emit_op1(ir, OPCODE_KIL, ir_to_mesa_undef_dst, this->result);
+   } else {
+      ir_to_mesa_emit_op0(ir, OPCODE_KIL_NV);
+   }
 
-   ir_to_mesa_emit_op0(ir, OPCODE_KIL_NV);
    fp->UsesKill = GL_TRUE;
 }
 
@@ -2844,8 +2849,9 @@ _mesa_ir_link_shader(struct gl_context *ctx, struct gl_shader_program *prog)
 
 	 /* Lowering */
 	 do_mat_op_to_vec(ir);
-	 lower_instructions(ir, MOD_TO_FRACT | DIV_TO_MUL_RCP | EXP_TO_EXP2
-			      | LOG_TO_LOG2);
+	 lower_instructions(ir, (MOD_TO_FRACT | DIV_TO_MUL_RCP | EXP_TO_EXP2
+				 | LOG_TO_LOG2
+				 | ((options->EmitNoPow) ? POW_TO_EXP2 : 0)));
 
 	 progress = do_lower_jumps(ir, true, true, options->EmitNoMainReturn, options->EmitNoCont, options->EmitNoLoops) || progress;
 
@@ -2853,8 +2859,10 @@ _mesa_ir_link_shader(struct gl_context *ctx, struct gl_shader_program *prog)
 
 	 progress = lower_quadop_vector(ir, true) || progress;
 
-	 if (options->EmitNoIfs)
+	 if (options->EmitNoIfs) {
+	    progress = lower_discard(ir) || progress;
 	    progress = do_if_to_cond_assign(ir) || progress;
+	 }
 
 	 if (options->EmitNoNoise)
 	    progress = lower_noise(ir) || progress;