r300g: implement MSAA compression and fast MSAA color clear

These are optimizations which make MSAA a lot faster.

The MSAA work is complete with this commit.  (except for enablement of AA
optimizations for RGBA16F, for which a patch is ready and waiting until
the kernel CS checker fix lands)

MSAA can't be made any faster as far as hw programming is concerned.

The catch is only one process and one colorbuffer can use the optimizations
at a time.  There usually is only one MSAA colorbuffer, so it shouldn't be
an issue.

Also, there is a limit on the size of MSAA colorbuffer resolution in terms
of megapixels.  If the limit is surpassed, the AA optimizations are disabled.
The limit is:
- 1 Mpix on low-end and some mid-level chipsets (1024x768 and 1280x720)
- 2 Mpix on some mid-level chipsets (1600x1200 and 1920x1080)
- 3 or 4 Mpix on high-end chipsets (2048x1536 or 2560x1600, respectively)
It corresponds to the number of raster pipes (= GB pipes) available, each pipe
can hold 1 Mpix of AA compression data.

If it's enabled, the driver prints to stdout:
  radeon: Acquired access to AA optimizations.

1 files changed, 35 insertions, 2 deletions

diff --git a/src/gallium/drivers/r300/r300_emit.c b/src/gallium/drivers/r300/r300_emit.c
index cb6c46e5e6c..9ea084fac2d 100644
--- a/src/gallium/drivers/r300/r300_emit.c
+++ b/src/gallium/drivers/r300/r300_emit.c
@@ -399,14 +399,17 @@ void r300_emit_fb_state(struct r300_context* r300, unsigned size, void* state)
 
     BEGIN_CS(size);
 
-    /* NUM_MULTIWRITES replicates COLOR[0] to all colorbuffers, which is not
-     * what we usually want. */
     if (r300->screen->caps.is_r500) {
         rb3d_cctl = R300_RB3D_CCTL_INDEPENDENT_COLORFORMAT_ENABLE_ENABLE;
     }
+    /* NUM_MULTIWRITES replicates COLOR[0] to all colorbuffers. */
     if (fb->nr_cbufs && r300->fb_multiwrite) {
         rb3d_cctl |= R300_RB3D_CCTL_NUM_MULTIWRITES(fb->nr_cbufs);
     }
+    if (r300->cmask_in_use) {
+        rb3d_cctl |= R300_RB3D_CCTL_AA_COMPRESSION_ENABLE |
+                     R300_RB3D_CCTL_CMASK_ENABLE;
+    }
 
     OUT_CS_REG(R300_RB3D_CCTL, rb3d_cctl);
 
@@ -419,6 +422,12 @@ void r300_emit_fb_state(struct r300_context* r300, unsigned size, void* state)
 
         OUT_CS_REG(R300_RB3D_COLORPITCH0 + (4 * i), surf->pitch);
         OUT_CS_RELOC(surf);
+
+        if (r300->cmask_in_use && i == 0) {
+            OUT_CS_REG(R300_RB3D_CMASK_OFFSET0, 0);
+            OUT_CS_REG(R300_RB3D_CMASK_PITCH0, surf->pitch_cmask);
+            OUT_CS_REG(R300_RB3D_COLOR_CLEAR_VALUE, r300->color_clear_value);
+        }
     }
 
     /* Set up the ZB part of the CBZB clear. */
@@ -1240,6 +1249,30 @@ void r300_emit_zmask_clear(struct r300_context *r300, unsigned size, void *state
     r300_mark_atom_dirty(r300, &r300->hyperz_state);
 }
 
+void r300_emit_cmask_clear(struct r300_context *r300, unsigned size, void *state)
+{
+    struct pipe_framebuffer_state *fb =
+        (struct pipe_framebuffer_state*)r300->fb_state.state;
+    struct r300_resource *tex;
+    CS_LOCALS(r300);
+
+    tex = r300_resource(fb->cbufs[0]->texture);
+
+    BEGIN_CS(size);
+    OUT_CS_REG(R300_RB3D_DSTCACHE_CTLSTAT,
+               R300_RB3D_DSTCACHE_CTLSTAT_DC_FREE_FREE_3D_TAGS |
+               R300_RB3D_DSTCACHE_CTLSTAT_DC_FLUSH_FLUSH_DIRTY_3D);
+    OUT_CS_PKT3(R300_PACKET3_3D_CLEAR_CMASK, 2);
+    OUT_CS(0);
+    OUT_CS(tex->tex.cmask_dwords);
+    OUT_CS(0);
+    END_CS;
+
+    /* Mark the current zbuffer's zmask as in use. */
+    r300->cmask_in_use = TRUE;
+    r300_mark_fb_state_dirty(r300, R300_CHANGED_CMASK_ENABLE);
+}
+
 void r300_emit_ztop_state(struct r300_context* r300,
                           unsigned size, void* state)
 {