Isolate GrBufferAllocPools inside GrBatchTarget

Isolate GrBufferAllocPools inside GrBatchTarget

This CL refactors the location of the GrBufferAllocPools so they reside entirely inside the GrBatchTarget. This is in preparation for making them use scratch resources.

Committed: https://skia.googlesource.com/skia/+/e40d39725908d99d255496025411e5c9ca32663e

[robertphillips, 2015-05-07 12:52:13]

robertphillips@google.com changed reviewers:
+ bsalomon@google.com

[bsalomon, 2015-05-07 14:28:40]

Really nice to see the removal of the pools from all these layers!

https://codereview.chromium.org/1131553002/diff/80001/src/gpu/GrBatchTarget.h
File src/gpu/GrBatchTarget.h (right):

https://codereview.chromium.org/1131553002/diff/80001/src/gpu/GrBatchTarget.h...
src/gpu/GrBatchTarget.h:129: void unmap() {
Maybe a bit more descriptive name here? unmapBuffers()?
unmapVertexAndIndexBuffers()?

Also, wondering whether we want to expose these methods to GrBatchs.

[robertphillips, 2015-05-07 14:34:21]

https://codereview.chromium.org/1131553002/diff/80001/src/gpu/GrBatchTarget.h
File src/gpu/GrBatchTarget.h (right):

https://codereview.chromium.org/1131553002/diff/80001/src/gpu/GrBatchTarget.h...
src/gpu/GrBatchTarget.h:129: void unmap() {
On 2015/05/07 14:28:40, bsalomon wrote:
> Maybe a bit more descriptive name here? unmapBuffers()?
> unmapVertexAndIndexBuffers()?
> 
> Also, wondering whether we want to expose these methods to GrBatchs.

Done-ish. unmap has been renamed and made private. reset needs to be accessible
for GrTargetCommands::reset to call.

[bsalomon, 2015-05-07 14:44:34]

lgtm

[robertphillips, 2015-05-07 15:04:30]

The CQ bit was checked by robertphillips@google.com to run a CQ dry run

[commit-bot: I haz the power, 2015-05-07 15:04:48]

Dry run: CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1131553002/100001

[commit-bot: I haz the power, 2015-05-07 15:10:32]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2015-05-07 15:10:33]

Dry run: This issue passed the CQ dry run.

[robertphillips, 2015-05-07 15:24:44]

The CQ bit was checked by robertphillips@google.com

[commit-bot: I haz the power, 2015-05-07 15:25:01]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1131553002/100001

[commit-bot: I haz the power, 2015-05-07 15:25:15]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2015-05-07 15:25:16]

Failed to apply patch for src/gpu/GrDefaultPathRenderer.cpp:
While running git apply --index -3 -p1;
  error: patch failed: src/gpu/GrDefaultPathRenderer.cpp:9
  Falling back to three-way merge...
  Applied patch to 'src/gpu/GrDefaultPathRenderer.cpp' with conflicts.
  U src/gpu/GrDefaultPathRenderer.cpp

Patch:       src/gpu/GrDefaultPathRenderer.cpp
Index: src/gpu/GrDefaultPathRenderer.cpp
diff --git a/src/gpu/GrDefaultPathRenderer.cpp
b/src/gpu/GrDefaultPathRenderer.cpp
index
e1ba1c84518003e033c683dc8280906d28549126..c3883b9c281d7dc86d78057391aa35e77aa41173
100644
--- a/src/gpu/GrDefaultPathRenderer.cpp
+++ b/src/gpu/GrDefaultPathRenderer.cpp
@@ -9,7 +9,6 @@
 
 #include "GrBatch.h"
 #include "GrBatchTarget.h"
-#include "GrBufferAllocPool.h"
 #include "GrContext.h"
 #include "GrDefaultGeoProcFactory.h"
 #include "GrPathUtils.h"
@@ -319,10 +318,8 @@ public:
         const GrVertexBuffer* vertexBuffer;
         int firstVertex;
 
-        void* verts = batchTarget->vertexPool()->makeSpace(vertexStride,
-                                                           maxVertices,
-                                                           &vertexBuffer,
-                                                           &firstVertex);
+        void* verts = batchTarget->makeVertSpace(vertexStride, maxVertices,
+                                                 &vertexBuffer, &firstVertex);
 
         if (!verts) {
             SkDebugf("Could not allocate vertices\n");
@@ -334,9 +331,7 @@ public:
 
         void* indices = NULL;
         if (isIndexed) {
-            indices = batchTarget->indexPool()->makeSpace(maxIndices,
-                                                          &indexBuffer,
-                                                          &firstIndex);
+            indices = batchTarget->makeIndexSpace(maxIndices, &indexBuffer,
&firstIndex);
 
             if (!indices) {
                 SkDebugf("Could not allocate indices\n");

[robertphillips, 2015-05-07 15:28:06]

The CQ bit was checked by robertphillips@google.com to run a CQ dry run

[robertphillips, 2015-05-07 15:28:07]

The patchset sent to the CQ was uploaded after l-g-t-m from bsalomon@google.com
Link to the patchset: https://codereview.chromium.org/1131553002/#ps120001
(title: "update to ToT")

[commit-bot: I haz the power, 2015-05-07 15:28:21]

Dry run: CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1131553002/120001

[commit-bot: I haz the power, 2015-05-07 15:38:41]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2015-05-07 15:38:42]

Dry run: This issue passed the CQ dry run.

[robertphillips, 2015-05-07 16:51:01]

The CQ bit was checked by robertphillips@google.com

[commit-bot: I haz the power, 2015-05-07 16:51:12]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1131553002/120001

[commit-bot: I haz the power, 2015-05-07 16:51:47]

Committed patchset #7 (id:120001) as
https://skia.googlesource.com/skia/+/e40d39725908d99d255496025411e5c9ca32663e