Texture Draw Calls Coalescing

cc/gl_renderer.cc

 // Copyright 2010 The Chromium Authors. All rights reserved.
 // Use of this source code is governed by a BSD-style license that can be
 // found in the LICENSE file.
 
 #include "cc/gl_renderer.h"
 
 #include "base/debug/trace_event.h"
 #include "base/logging.h"
 #include "base/string_split.h"
 #include "base/string_util.h"
@@ skipping 225 matching lines @@
     makeContextCurrent();
     // Bind the common vertex attributes used for drawing all the layers.
     m_sharedGeometry->prepareForDraw();
 
     GLC(m_context, m_context->disable(GL_DEPTH_TEST));
     GLC(m_context, m_context->disable(GL_CULL_FACE));
     GLC(m_context, m_context->colorMask(true, true, true, true));
     GLC(m_context, m_context->enable(GL_BLEND));
     GLC(m_context, m_context->blendFunc(GL_ONE, GL_ONE_MINUS_SRC_ALPHA));
     GLC(context(), context()->activeTexture(GL_TEXTURE0));
+    m_blendShadow = true;

[jamesr, 2012/11/29 11:17:11]

nit: could you move this right next to the enable(GL_BLEND) call (or move that
call down a few lines) just to be extra clear about what state this is
shadowing?

[whunt, 2012/11/29 23:55:06]

Done.

On 2012/11/29 11:17:11, jamesr wrote:

+    m_programShadow = 0;
 }
 
 void GLRenderer::doNoOp()
 {
     GLC(m_context, m_context->bindFramebuffer(GL_FRAMEBUFFER, 0));
     GLC(m_context, m_context->flush());
 }
 
 void GLRenderer::drawQuad(DrawingFrame& frame, const DrawQuad* quad)
 {
     DCHECK(quad->rect.Contains(quad->visible_rect));
-
-    if (quad->ShouldDrawWithBlending())
-        GLC(m_context, m_context->enable(GL_BLEND));
-    else
-        GLC(m_context, m_context->disable(GL_BLEND));
+    if (quad->material != DrawQuad::TEXTURE_CONTENT)
+    {
+        flushTextureQuadCache();
+        setBlendEnabled(quad->ShouldDrawWithBlending());
+    }
 
     switch (quad->material) {
     case DrawQuad::INVALID:
         NOTREACHED();
         break;
     case DrawQuad::CHECKERBOARD:
         drawCheckerboardQuad(frame, CheckerboardDrawQuad::MaterialCast(quad));
         break;
     case DrawQuad::DEBUG_BORDER:
         drawDebugBorderQuad(frame, DebugBorderDrawQuad::MaterialCast(quad));
         break;
     case DrawQuad::IO_SURFACE_CONTENT:
         drawIOSurfaceQuad(frame, IOSurfaceDrawQuad::MaterialCast(quad));
         break;
     case DrawQuad::RENDER_PASS:
         drawRenderPassQuad(frame, RenderPassDrawQuad::MaterialCast(quad));
         break;
     case DrawQuad::SOLID_COLOR:
         drawSolidColorQuad(frame, SolidColorDrawQuad::MaterialCast(quad));
         break;
     case DrawQuad::STREAM_VIDEO_CONTENT:
         drawStreamVideoQuad(frame, StreamVideoDrawQuad::MaterialCast(quad));
         break;
     case DrawQuad::TEXTURE_CONTENT:
-        drawTextureQuad(frame, TextureDrawQuad::MaterialCast(quad));
+        enqueueTextureQuad(frame, TextureDrawQuad::MaterialCast(quad));
         break;
     case DrawQuad::TILED_CONTENT:
         drawTileQuad(frame, TileDrawQuad::MaterialCast(quad));
         break;
     case DrawQuad::YUV_VIDEO_CONTENT:
         drawYUVVideoQuad(frame, YUVVideoDrawQuad::MaterialCast(quad));
         break;
     }
 }
 
 void GLRenderer::drawCheckerboardQuad(const DrawingFrame& frame, const CheckerboardDrawQuad* quad)
 {
     const TileCheckerboardProgram* program = tileCheckerboardProgram();
     DCHECK(program && (program->initialized() || isContextLost()));
-    GLC(context(), context()->useProgram(program->program()));
+    setUseProgram(program->program());
 
     SkColor color = quad->color;
     GLC(context(), context()->uniform4f(program->fragmentShader().colorLocation(), SkColorGetR(color) / 255.0, SkColorGetG(color) / 255.0, SkColorGetB(color) / 255.0, 1));
 
     const int checkerboardWidth = 16;
     float frequency = 1.0 / checkerboardWidth;
 
     gfx::Rect tileRect = quad->rect;
     float texOffsetX = tileRect.x() % checkerboardWidth;
     float texOffsetY = tileRect.y() % checkerboardWidth;
     float texScaleX = tileRect.width();
     float texScaleY = tileRect.height();
     GLC(context(), context()->uniform4f(program->fragmentShader().texTransformLocation(), texOffsetX, texOffsetY, texScaleX, texScaleY));
 
     GLC(context(), context()->uniform1f(program->fragmentShader().frequencyLocation(), frequency));
 
     setShaderOpacity(quad->opacity(), program->fragmentShader().alphaLocation());
     drawQuadGeometry(frame, quad->quadTransform(), quad->rect, program->vertexShader().matrixLocation());
 }
 
 void GLRenderer::drawDebugBorderQuad(const DrawingFrame& frame, const DebugBorderDrawQuad* quad)
 {
     static float glMatrix[16];
     const SolidColorProgram* program = solidColorProgram();
     DCHECK(program && (program->initialized() || isContextLost()));
-    GLC(context(), context()->useProgram(program->program()));
+    setUseProgram(program->program());
 
     // Use the full quadRect for debug quads to not move the edges based on partial swaps.
     const gfx::Rect& layerRect = quad->rect;
     gfx::Transform renderMatrix = quad->quadTransform();
     renderMatrix.Translate(0.5 * layerRect.width() + layerRect.x(), 0.5 * layerRect.height() + layerRect.y());
     renderMatrix.Scale(layerRect.width(), layerRect.height());
     GLRenderer::toGLMatrix(&glMatrix[0], frame.projectionMatrix * renderMatrix);
     GLC(context(), context()->uniformMatrix4fv(program->vertexShader().matrixLocation(), 1, false, &glMatrix[0]));
 
     SkColor color = quad->color;
     float alpha = SkColorGetA(color) / 255.0;
 
     GLC(context(), context()->uniform4f(program->fragmentShader().colorLocation(), (SkColorGetR(color) / 255.0) * alpha, (SkColorGetG(color) / 255.0) * alpha, (SkColorGetB(color) / 255.0) * alpha, alpha));
 
     GLC(context(), context()->lineWidth(quad->width));
 
     // The indices for the line are stored in the same array as the triangle indices.
-    GLC(context(), context()->drawElements(GL_LINE_LOOP, 4, GL_UNSIGNED_SHORT, 6 * sizeof(unsigned short)));
+    GLC(context(), context()->drawElements(GL_LINE_LOOP, 4, GL_UNSIGNED_SHORT, 0));
 }
 
 static WebGraphicsContext3D* getFilterContext(bool hasImplThread)
 {
     if (hasImplThread)
         return WebSharedGraphicsContext3D::compositorThreadContext();
     else
         return WebSharedGraphicsContext3D::mainThreadContext();
 }
 
@@ skipping 224 matching lines @@
 
     int shaderQuadLocation = -1;
     int shaderEdgeLocation = -1;
     int shaderMaskSamplerLocation = -1;
     int shaderMaskTexCoordScaleLocation = -1;
     int shaderMaskTexCoordOffsetLocation = -1;
     int shaderMatrixLocation = -1;
     int shaderAlphaLocation = -1;
     if (useAA && maskTextureId) {
         const RenderPassMaskProgramAA* program = renderPassMaskProgramAA();
-        GLC(context(), context()->useProgram(program->program()));
+        setUseProgram(program->program());
         GLC(context(), context()->uniform1i(program->fragmentShader().samplerLocation(), 0));
 
         shaderQuadLocation = program->vertexShader().pointLocation();
         shaderEdgeLocation = program->fragmentShader().edgeLocation();
         shaderMaskSamplerLocation = program->fragmentShader().maskSamplerLocation();
         shaderMaskTexCoordScaleLocation = program->fragmentShader().maskTexCoordScaleLocation();
         shaderMaskTexCoordOffsetLocation = program->fragmentShader().maskTexCoordOffsetLocation();
         shaderMatrixLocation = program->vertexShader().matrixLocation();
         shaderAlphaLocation = program->fragmentShader().alphaLocation();
     } else if (!useAA && maskTextureId) {
         const RenderPassMaskProgram* program = renderPassMaskProgram();
-        GLC(context(), context()->useProgram(program->program()));
+        setUseProgram(program->program());
         GLC(context(), context()->uniform1i(program->fragmentShader().samplerLocation(), 0));
 
         shaderMaskSamplerLocation = program->fragmentShader().maskSamplerLocation();
         shaderMaskTexCoordScaleLocation = program->fragmentShader().maskTexCoordScaleLocation();
         shaderMaskTexCoordOffsetLocation = program->fragmentShader().maskTexCoordOffsetLocation();
         shaderMatrixLocation = program->vertexShader().matrixLocation();
         shaderAlphaLocation = program->fragmentShader().alphaLocation();
     } else if (useAA && !maskTextureId) {
         const RenderPassProgramAA* program = renderPassProgramAA();
-        GLC(context(), context()->useProgram(program->program()));
+        setUseProgram(program->program());
         GLC(context(), context()->uniform1i(program->fragmentShader().samplerLocation(), 0));
 
         shaderQuadLocation = program->vertexShader().pointLocation();
         shaderEdgeLocation = program->fragmentShader().edgeLocation();
         shaderMatrixLocation = program->vertexShader().matrixLocation();
         shaderAlphaLocation = program->fragmentShader().alphaLocation();
     } else {
         const RenderPassProgram* program = renderPassProgram();
-        GLC(context(), context()->useProgram(program->program()));
+        setUseProgram(program->program());
         GLC(context(), context()->uniform1i(program->fragmentShader().samplerLocation(), 0));
 
         shaderMatrixLocation = program->vertexShader().matrixLocation();
         shaderAlphaLocation = program->fragmentShader().alphaLocation();
     }
 
     if (shaderMaskSamplerLocation != -1) {
         DCHECK(shaderMaskTexCoordScaleLocation != 1);
         DCHECK(shaderMaskTexCoordOffsetLocation != 1);
         GLC(context(), context()->activeTexture(GL_TEXTURE1));
@@ skipping 21 matching lines @@
 
     // Flush the compositor context before the filter bitmap goes out of
     // scope, so the draw gets processed before the filter texture gets deleted.
     if (filterBitmap.getTexture())
         m_context->flush();
 }
 
 void GLRenderer::drawSolidColorQuad(const DrawingFrame& frame, const SolidColorDrawQuad* quad)
 {
     const SolidColorProgram* program = solidColorProgram();
-    GLC(context(), context()->useProgram(program->program()));
+    setUseProgram(program->program());
 
     SkColor color = quad->color;
     float opacity = quad->opacity();
     float alpha = (SkColorGetA(color) / 255.0) * opacity;
 
     GLC(context(), context()->uniform4f(program->fragmentShader().colorLocation(), (SkColorGetR(color) / 255.0) * alpha, (SkColorGetG(color) / 255.0) * alpha, (SkColorGetB(color) / 255.0) * alpha, alpha));
 
     drawQuadGeometry(frame, quad->quadTransform(), quad->rect, program->vertexShader().matrixLocation());
 }
 
@@ skipping 97 matching lines @@
             else
                 tileUniformLocation(tileProgram(), uniforms);
         } else {
             if (quad->swizzle_contents)
                 tileUniformLocation(tileProgramSwizzleOpaque(), uniforms);
             else
                 tileUniformLocation(tileProgramOpaque(), uniforms);
         }
     }
 
-    GLC(context(), context()->useProgram(uniforms.program));
+    setUseProgram(uniforms.program);
     GLC(context(), context()->uniform1i(uniforms.samplerLocation, 0));
     ResourceProvider::ScopedReadLockGL quadResourceLock(m_resourceProvider, quad->resource_id);
     GLC(context(), context()->bindTexture(GL_TEXTURE_2D, quadResourceLock.textureId()));
 
     bool useAA = !clipped && quad->IsAntialiased();
     if (useAA) {
         LayerQuad deviceLayerBounds = LayerQuad(gfx::QuadF(deviceLayerQuad.BoundingBox()));
         deviceLayerBounds.inflateAntiAliasingDistance();
 
         LayerQuad deviceLayerEdges = LayerQuad(deviceLayerQuad);
@@ skipping 96 matching lines @@
     ResourceProvider::ScopedReadLockGL yPlaneLock(m_resourceProvider, yPlane.resourceId);
     ResourceProvider::ScopedReadLockGL uPlaneLock(m_resourceProvider, uPlane.resourceId);
     ResourceProvider::ScopedReadLockGL vPlaneLock(m_resourceProvider, vPlane.resourceId);
     GLC(context(), context()->activeTexture(GL_TEXTURE1));
     GLC(context(), context()->bindTexture(GL_TEXTURE_2D, yPlaneLock.textureId()));
     GLC(context(), context()->activeTexture(GL_TEXTURE2));
     GLC(context(), context()->bindTexture(GL_TEXTURE_2D, uPlaneLock.textureId()));
     GLC(context(), context()->activeTexture(GL_TEXTURE3));
     GLC(context(), context()->bindTexture(GL_TEXTURE_2D, vPlaneLock.textureId()));
 
-    GLC(context(), context()->useProgram(program->program()));
+    setUseProgram(program->program());
 
     GLC(context(), context()->uniform2f(program->vertexShader().texScaleLocation(), quad->tex_scale.width(), quad->tex_scale.height()));
     GLC(context(), context()->uniform1i(program->fragmentShader().yTextureLocation(), 1));
     GLC(context(), context()->uniform1i(program->fragmentShader().uTextureLocation(), 2));
     GLC(context(), context()->uniform1i(program->fragmentShader().vTextureLocation(), 3));
 
     // These values are magic numbers that are used in the transformation from YUV to RGB color values.
     // They are taken from the following webpage: http://www.fourcc.org/fccyvrgb.php
     float yuv2RGB[9] = {
         1.164f, 1.164f, 1.164f,
@@ skipping 22 matching lines @@
     GLC(context(), context()->activeTexture(GL_TEXTURE0));
 }
 
 void GLRenderer::drawStreamVideoQuad(const DrawingFrame& frame, const StreamVideoDrawQuad* quad)
 {
     static float glMatrix[16];
 
     DCHECK(m_capabilities.usingEglImage);
 
     const VideoStreamTextureProgram* program = videoStreamTextureProgram();
-    GLC(context(), context()->useProgram(program->program()));
+    setUseProgram(program->program());
 
     toGLMatrix(&glMatrix[0], quad->matrix);
     GLC(context(), context()->uniformMatrix4fv(program->vertexShader().texMatrixLocation(), 1, false, glMatrix));
 
     GLC(context(), context()->bindTexture(GL_TEXTURE_EXTERNAL_OES, quad->texture_id));
 
     GLC(context(), context()->uniform1i(program->fragmentShader().samplerLocation(), 0));
 
     setShaderOpacity(quad->opacity(), program->fragmentShader().alphaLocation());
     drawQuadGeometry(frame, quad->quadTransform(), quad->rect, program->vertexShader().matrixLocation());
@@ skipping 18 matching lines @@
 struct TexTransformTextureProgramBinding : TextureProgramBinding {
     template<class Program> void set(
         Program* program, WebKit::WebGraphicsContext3D* context)
     {
         TextureProgramBinding::set(program, context);
         texTransformLocation = program->vertexShader().texTransformLocation();
     }
     int texTransformLocation;
 };
 
+void GLRenderer::flushTextureQuadCache()
+{
+    // Check to see if we have anything to draw

[jamesr, 2012/11/29 11:17:11]

nit: end comments with a period as if they were a complete sentence, even if
it's not

[whunt, 2012/11/29 23:55:06]

Done, but Why?? They're not complete sentences...

On 2012/11/29 11:17:11, jamesr wrote:

+    if (m_drawCache.programID == 0)
+      return;
+
+    // Set the correct blending mode
+    setBlendEnabled(m_drawCache.needsBlending);
+
+    // Bind the program to the GL state
+    setUseProgram(m_drawCache.programID);
+
+    // Assume the current active textures is 0
+    //GLC(context(), context()->activeTexture(GL_TEXTURE0));

[jamesr, 2012/11/29 11:17:11]

I would prefer deleting this line to commenting it out.  It's fine to document
our assumptions in comments since we do enforce that the active texture unit is
0 between draw calls

+    ResourceProvider::ScopedReadLockGL lockedQuad(m_resourceProvider, m_drawCache.resourceID);
+    GLC(context(), context()->bindTexture(GL_TEXTURE_2D, lockedQuad.textureId()));
+
+    // set up premultiplied alpha
+    if (!m_drawCache.usePremultipliedAlpha) {
+      // As it turns out, the premultiplied alpha blending function (ONE, ONE_MINUS_SRC_ALPHA)
+        // will never cause the alpha channel to be set to anything less than 1.0 if it is
+        // initialized to that value! Therefore, premultipliedAlpha being false is the first
+        // situation we can generally see an alpha channel less than 1.0 coming out of the
+        // compositor. This is causing platform differences in some layout tests (see
+        // https://bugs.webkit.org/show_bug.cgi?id=82412), so in this situation, use a separate
+        // blend function for the alpha channel to avoid modifying it. Don't use colorMask for this
+        // as it has performance implications on some platforms.
+        GLC(context(), context()->blendFuncSeparate(GL_SRC_ALPHA, GL_ONE_MINUS_SRC_ALPHA, GL_ZERO, GL_ONE));
+    }
+
+    // Set the shader opacity
+    setShaderOpacity(m_drawCache.alpha, m_drawCache.alphaLocation);
+
+    DCHECK(sizeof(float4) == 4 * sizeof(float));

[jamesr, 2012/11/29 11:17:11]

You can do this with COMPILE_ASSERT() from base/basictypes.h (which is already
included in this file)

+    DCHECK(sizeof(float16) == 16 * sizeof(float));
+
+    // Upload the tranforms for both points and uvs
+    GLC(m_context, m_context->uniformMatrix4fv((int)m_drawCache.matrixLocation, (int)m_drawCache.matrixData.size(), false, (float*)&m_drawCache.matrixData.front()));
+    GLC(m_context, m_context->uniform4fv((int)m_drawCache.uvXformLocation, (int)m_drawCache.uvXformData.size(), (float*)&m_drawCache.uvXformData.front()));
+
+    // Draw the quads!
+    GLC(m_context, m_context->drawElements(GL_TRIANGLES, 6 * m_drawCache.matrixData.size(), GL_UNSIGNED_SHORT, 0));
+
+    // Clean up after ourselves (reset state set above)
+     if (!m_drawCache.usePremultipliedAlpha)
+        GLC(m_context, m_context->blendFunc(GL_ONE, GL_ONE_MINUS_SRC_ALPHA));
+
+    // Clear the cache
+    m_drawCache.programID = 0;
+    m_drawCache.uvXformData.resize(0);
+    m_drawCache.matrixData.resize(0);
+}
+
+void GLRenderer::enqueueTextureQuad(const DrawingFrame& frame, const TextureDrawQuad* quad)
+{
+    // Choose the correcte texture program binding
+    TexTransformTextureProgramBinding binding;
+    if (quad->flipped)
+        binding.set(textureProgramFlip(), context());
+    else
+        binding.set(textureProgram(), context());
+
+    int resourceID = quad->resource_id;
+
+    if (m_drawCache.programID             != binding.programId              ||

[jamesr, 2012/11/29 11:17:11]

I think this is a little more spaced out than we normally do - chromium style
would be to format this with one space around the "!=" operator and before the
||.  The "!="s won't line up, but it's still pretty readable

[whunt, 2012/11/29 23:55:06]

More importantly, it wouldn't be block-editable for those of us who edit
multiple lines at a time.  I can collapse it.
On 2012/11/29 11:17:11, jamesr wrote:

+        m_drawCache.resourceID            != resourceID                     ||
+        m_drawCache.alpha                 != quad->opacity()                ||
+        m_drawCache.usePremultipliedAlpha != quad->premultiplied_alpha      ||
+        m_drawCache.needsBlending         != quad->ShouldDrawWithBlending() ||
+        m_drawCache.matrixData.size() >= 8)
+    {
+        flushTextureQuadCache();
+        m_drawCache.programID = binding.programId;
+        m_drawCache.resourceID = resourceID;
+        m_drawCache.alpha = quad->opacity();
+        m_drawCache.usePremultipliedAlpha = quad->premultiplied_alpha;
+        m_drawCache.needsBlending = quad->ShouldDrawWithBlending();
+
+        m_drawCache.alphaLocation = binding.alphaLocation;
+        m_drawCache.uvXformLocation = binding.texTransformLocation;
+        m_drawCache.matrixLocation = binding.matrixLocation;
+    }
+
+    // Generate the uv-transform
+    const gfx::RectF& uvRect = quad->uv_rect;
+    float4 uv = {uvRect.x(), uvRect.y(), uvRect.width(), uvRect.height()};
+    m_drawCache.uvXformData.push_back(uv);
+
+    // Generate the transform matrix
+    gfx::Transform quadRectMatrix;
+    quadRectTransform(&quadRectMatrix, quad->quadTransform(), quad->rect);
+    quadRectMatrix = frame.projectionMatrix * quadRectMatrix;
+
+    float16 m;
+    quadRectMatrix.matrix().asColMajorf(m.data);
+    m_drawCache.matrixData.push_back(m);
+}
+
 void GLRenderer::drawTextureQuad(const DrawingFrame& frame, const TextureDrawQuad* quad)
 {
     TexTransformTextureProgramBinding binding;
     if (quad->flipped)
         binding.set(textureProgramFlip(), context());
     else
         binding.set(textureProgram(), context());
-    GLC(context(), context()->useProgram(binding.programId));
+    setUseProgram(binding.programId);
     GLC(context(), context()->uniform1i(binding.samplerLocation, 0));
     const gfx::RectF& uvRect = quad->uv_rect;
     GLC(context(), context()->uniform4f(binding.texTransformLocation, uvRect.x(), uvRect.y(), uvRect.width(), uvRect.height()));
 
     ResourceProvider::ScopedReadLockGL quadResourceLock(m_resourceProvider, quad->resource_id);
     GLC(context(), context()->bindTexture(GL_TEXTURE_2D, quadResourceLock.textureId()));
 
     if (!quad->premultiplied_alpha) {
         // As it turns out, the premultiplied alpha blending function (ONE, ONE_MINUS_SRC_ALPHA)
         // will never cause the alpha channel to be set to anything less than 1.0 if it is
@@ skipping 11 matching lines @@
 
     if (!quad->premultiplied_alpha)
         GLC(m_context, m_context->blendFunc(GL_ONE, GL_ONE_MINUS_SRC_ALPHA));
 }
 
 void GLRenderer::drawIOSurfaceQuad(const DrawingFrame& frame, const IOSurfaceDrawQuad* quad)
 {
     TexTransformTextureProgramBinding binding;
     binding.set(textureIOSurfaceProgram(), context());
 
-    GLC(context(), context()->useProgram(binding.programId));
+    setUseProgram(binding.programId);
     GLC(context(), context()->uniform1i(binding.samplerLocation, 0));
     if (quad->orientation == IOSurfaceDrawQuad::FLIPPED)
         GLC(context(), context()->uniform4f(binding.texTransformLocation, 0, quad->io_surface_size.height(), quad->io_surface_size.width(), quad->io_surface_size.height() * -1.0));
     else
         GLC(context(), context()->uniform4f(binding.texTransformLocation, 0, 0, quad->io_surface_size.width(), quad->io_surface_size.height()));
 
     GLC(context(), context()->bindTexture(GL_TEXTURE_RECTANGLE_ARB, quad->io_surface_texture_id));
 
     setShaderOpacity(quad->opacity(), binding.alphaLocation);
     drawQuadGeometry(frame, quad->quadTransform(), quad->rect, binding.matrixLocation);
 
     GLC(context(), context()->bindTexture(GL_TEXTURE_RECTANGLE_ARB, 0));
 }
 
 void GLRenderer::finishDrawingFrame(DrawingFrame& frame)
 {
     m_currentFramebufferLock.reset();
     m_swapBufferRect.Union(gfx::ToEnclosingRect(frame.rootDamageRect));
 
     GLC(m_context, m_context->disable(GL_BLEND));
+    m_blendShadow = false;
+}
+
+void GLRenderer::finishDrawingQuadList()
+{
+    flushTextureQuadCache();
 }
 
 bool GLRenderer::flippedFramebuffer() const
 {
     return true;
 }
 
 void GLRenderer::ensureScissorTestEnabled()
 {
     if (m_isScissorEnabled)
         return;
 
+    flushTextureQuadCache();
     GLC(m_context, m_context->enable(GL_SCISSOR_TEST));
     m_isScissorEnabled = true;
 }
 
 void GLRenderer::ensureScissorTestDisabled()
 {
     if (!m_isScissorEnabled)
         return;
 
+    flushTextureQuadCache();
     GLC(m_context, m_context->disable(GL_SCISSOR_TEST));
     m_isScissorEnabled = false;
 }
 
 void GLRenderer::toGLMatrix(float* glMatrix, const gfx::Transform& transform)
 {
     transform.matrix().asColMajorf(glMatrix);
 }
 
 void GLRenderer::setShaderQuadF(const gfx::QuadF& quad, int quadLocation)
@@ skipping 12 matching lines @@
     point[7] = quad.p4().y();
     GLC(m_context, m_context->uniform2fv(quadLocation, 4, point));
 }
 
 void GLRenderer::setShaderOpacity(float opacity, int alphaLocation)
 {
     if (alphaLocation != -1)
         GLC(m_context, m_context->uniform1f(alphaLocation, opacity));
 }
 
+void GLRenderer::setBlendEnabled(bool enabled)
+{
+    if (enabled != m_blendShadow)

[jamesr, 2012/11/29 11:17:11]

we generally prefer to early return so the indentation level isn't quite as
deep.  so this would be:

if (enabled == m_blendShadow)
   return;

... rest of function

+    {
+        if (enabled)
+            GLC(m_context, m_context->enable(GL_BLEND));
+        else
+            GLC(m_context, m_context->disable(GL_BLEND));
+        m_blendShadow = enabled;
+    }
+}
+
+void GLRenderer::setUseProgram(unsigned program)
+{
+    if (program != m_programShadow)

[jamesr, 2012/11/29 11:17:11]

same here

+    {
+        GLC(m_context, m_context->useProgram(program));
+        m_programShadow = program;
+    }
+}
+
 void GLRenderer::drawQuadGeometry(const DrawingFrame& frame, const gfx::Transform& drawTransform, const gfx::RectF& quadRect, int matrixLocation)
 {
     gfx::Transform quadRectMatrix;
     quadRectTransform(&quadRectMatrix, drawTransform, quadRect);
     static float glMatrix[16];
     toGLMatrix(&glMatrix[0], frame.projectionMatrix * quadRectMatrix);
     GLC(m_context, m_context->uniformMatrix4fv(matrixLocation, 1, false, &glMatrix[0]));
 
     GLC(m_context, m_context->drawElements(GL_TRIANGLES, 6, GL_UNSIGNED_SHORT, 0));
 }
 
 void GLRenderer::copyTextureToFramebuffer(const DrawingFrame& frame, int textureId, const gfx::Rect& rect, const gfx::Transform& drawMatrix)
 {
     const RenderPassProgram* program = renderPassProgram();
 
     GLC(context(), context()->bindTexture(GL_TEXTURE_2D, textureId));
 
-    GLC(context(), context()->useProgram(program->program()));
+    setUseProgram(program->program());
     GLC(context(), context()->uniform1i(program->fragmentShader().samplerLocation(), 0));
     setShaderOpacity(1, program->fragmentShader().alphaLocation());
     drawQuadGeometry(frame, drawMatrix, rect, program->vertexShader().matrixLocation());
 }
 
 void GLRenderer::finish()
 {
     TRACE_EVENT0("cc", "GLRenderer::finish");
     m_context->finish();
 }
@@ skipping 230 matching lines @@
 void GLRenderer::setScissorTestRect(const gfx::Rect& scissorRect)
 {
     ensureScissorTestEnabled();
 
     // Don't unnecessarily ask the context to change the scissor, because it
     // may cause undesired GPU pipeline flushes.
     if (scissorRect == m_scissorRect)
         return;
 
     m_scissorRect = scissorRect;
+    flushTextureQuadCache();
     GLC(m_context, m_context->scissor(scissorRect.x(), scissorRect.y(), scissorRect.width(), scissorRect.height()));
 }
 
 void GLRenderer::setDrawViewportSize(const gfx::Size& viewportSize)
 {
     GLC(m_context, m_context->viewport(0, 0, viewportSize.width(), viewportSize.height()));
 }
 
 bool GLRenderer::makeContextCurrent()
 {
@@ skipping 149 matching lines @@
     return m_tileProgramSwizzleAA.get();
 }
 
 const GLRenderer::TextureProgram* GLRenderer::textureProgram()
 {
     if (!m_textureProgram)
         m_textureProgram = make_scoped_ptr(new TextureProgram(m_context));
     if (!m_textureProgram->initialized()) {
         TRACE_EVENT0("cc", "GLRenderer::textureProgram::initialize");
         m_textureProgram->initialize(m_context, m_isUsingBindUniform);
+        GLC(context(), context()->uniform1i(m_textureProgram.get()->fragmentShader().samplerLocation(), 0));
     }
     return m_textureProgram.get();
 }
 
 const GLRenderer::TextureProgramFlip* GLRenderer::textureProgramFlip()
 {
     if (!m_textureProgramFlip)
         m_textureProgramFlip = make_scoped_ptr(new TextureProgramFlip(m_context));
     if (!m_textureProgramFlip->initialized()) {
         TRACE_EVENT0("cc", "GLRenderer::textureProgramFlip::initialize");
         m_textureProgramFlip->initialize(m_context, m_isUsingBindUniform);
+        GLC(context(), context()->uniform1i(m_textureProgramFlip.get()->fragmentShader().samplerLocation(), 0));
     }
     return m_textureProgramFlip.get();
 }
 
 const GLRenderer::TextureIOSurfaceProgram* GLRenderer::textureIOSurfaceProgram()
 {
     if (!m_textureIOSurfaceProgram)
         m_textureIOSurfaceProgram = make_scoped_ptr(new TextureIOSurfaceProgram(m_context));
     if (!m_textureIOSurfaceProgram->initialized()) {
         TRACE_EVENT0("cc", "GLRenderer::textureIOSurfaceProgram::initialize");
@@ skipping 74 matching lines @@
 
     releaseRenderPassTextures();
 }
 
 bool GLRenderer::isContextLost()
 {
     return (m_context->getGraphicsResetStatusARB() != GL_NO_ERROR);
 }
 
 }  // namespace cc