[Mesa-dev] [PATCH] i965: Split Gen4-5 BlitFramebuffer code; prefer BLT over Meta.

[Kenneth Graunke]

A while back I switched intel_blit_framebuffer to prefer Meta over the
BLT.  This meant that Gen8 platforms would start using the 3D engine
for blits, just like we do on Gen6-7.5.

However, I hadn't considered Gen4-5 when making that change.  The BLT
engine appears to be substantially faster on 965GM than using Meta to
drive the 3D engine.  This isn't too surprising: original Gen4 doesn't
support tile offsets (that came on G45), and the level/layer fields
don't work for cubemap rendering, so for inconvenient miplevel
alignments, we end up blitting or copying data to/from temporaries
in order to render to it.  We may as well just use the blitter.

I chose to use the BLT on Gen4-5 because they use the same ring for
both 3D and BLT; Gen6+ splits it out.

Fixes regressions on 965GM due to botched tile offset code (we should
fix those properly as well, but they're longstanding bugs - for now,
put things back to the status quo).

Signed-off-by: Kenneth Graunke <kenneth at whitecape.org>
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=89430
Cc: "10.5" <mesa-stable at lists.freedesktop.org>
Cc: Mark Janes <mark.a.janes at intel.com>
---
 src/mesa/drivers/dri/i965/intel_fbo.c | 50 ++++++++++++++++++++++++++++++++++-
 1 file changed, 49 insertions(+), 1 deletion(-)

diff --git a/src/mesa/drivers/dri/i965/intel_fbo.c b/src/mesa/drivers/dri/i965/intel_fbo.c
index 04e5030..121c97f 100644
--- a/src/mesa/drivers/dri/i965/intel_fbo.c
+++ b/src/mesa/drivers/dri/i965/intel_fbo.c
@@ -916,6 +916,51 @@ intel_blit_framebuffer(struct gl_context *ctx,
 }
 
 /**
+ * Gen4-5 implementation of glBlitFrameBuffer().
+ *
+ * Tries BLT, Meta, then swrast.
+ *
+ * Gen4-5 have a single ring for both 3D and BLT operations, so there's no
+ * inter-ring synchronization issues like on Gen6+.  It is apparently faster
+ * than using the 3D pipeline.  Original Gen4 also has to rebase and copy
+ * miptree slices in order to render to unaligned locations.
+ */
+static void
+gen4_blit_framebuffer(struct gl_context *ctx,
+                      struct gl_framebuffer *readFb,
+                      struct gl_framebuffer *drawFb,
+                      GLint srcX0, GLint srcY0, GLint srcX1, GLint srcY1,
+                      GLint dstX0, GLint dstY0, GLint dstX1, GLint dstY1,
+                      GLbitfield mask, GLenum filter)
+{
+   /* Page 679 of OpenGL 4.4 spec says:
+    *    "Added BlitFramebuffer to commands affected by conditional rendering in
+    *     section 10.10 (Bug 9562)."
+    */
+   if (!_mesa_check_conditional_render(ctx))
+      return;
+
+   mask = intel_blit_framebuffer_with_blitter(ctx, readFb, drawFb,
+                                              srcX0, srcY0, srcX1, srcY1,
+                                              dstX0, dstY0, dstX1, dstY1,
+                                              mask, filter);
+   if (mask == 0x0)
+      return;
+
+   mask = _mesa_meta_BlitFramebuffer(ctx, readFb, drawFb,
+                                     srcX0, srcY0, srcX1, srcY1,
+                                     dstX0, dstY0, dstX1, dstY1,
+                                     mask, filter);
+   if (mask == 0x0)
+      return;
+
+   _swrast_BlitFramebuffer(ctx, readFb, drawFb,
+                           srcX0, srcY0, srcX1, srcY1,
+                           dstX0, dstY0, dstX1, dstY1,
+                           mask, filter);
+}
+
+/**
  * Does the renderbuffer have hiz enabled?
  */
 bool
@@ -1049,7 +1094,10 @@ intel_fbo_init(struct brw_context *brw)
    dd->UnmapRenderbuffer = intel_unmap_renderbuffer;
    dd->RenderTexture = intel_render_texture;
    dd->ValidateFramebuffer = intel_validate_framebuffer;
-   dd->BlitFramebuffer = intel_blit_framebuffer;
+   if (brw->gen >= 6)
+      dd->BlitFramebuffer = intel_blit_framebuffer;
+   else
+      dd->BlitFramebuffer = gen4_blit_framebuffer;
    dd->EGLImageTargetRenderbufferStorage =
       intel_image_target_renderbuffer_storage;
 
-- 
2.2.2