include/private/GrSurfaceProxy.h - skia - Git at Google

 /*
  * Copyright 2016 Google Inc.
  *
  * Use of this source code is governed by a BSD-style license that can be
  * found in the LICENSE file.
  */

 #ifndef GrSurfaceProxy_DEFINED
 #define GrSurfaceProxy_DEFINED

 #include "GrGpuResource.h"
 #include "GrSurface.h"

 #include "SkRect.h"

 class GrBackendTexture;
 class GrCaps;
 class GrOpList;
 class GrRenderTargetOpList;
 class GrRenderTargetProxy;
 class GrResourceProvider;
 class GrSurfaceContext;
 class GrSurfaceProxyPriv;
 class GrTextureOpList;
 class GrTextureProxy;

 // This class replicates the functionality GrIORef<GrSurface> but tracks the
 // utilitization for later resource allocation (for the deferred case) and
 // forwards on the utilization in the wrapped case
 class GrIORefProxy : public SkNoncopyable {
 public:
     void ref() const {
         this->validate();

         ++fRefCnt;
         if (fTarget) {
             fTarget->ref();
         }
     }

     void unref() const {
         this->validate();

         if (fTarget) {
             fTarget->unref();
         }

         if (!(--fRefCnt)) {
             delete this;
             return;
         }

         this->validate();
     }

     void validate() const {
 #ifdef SK_DEBUG
         SkASSERT(fRefCnt >= 1);
         SkASSERT(fPendingReads >= 0);
         SkASSERT(fPendingWrites >= 0);
         SkASSERT(fRefCnt + fPendingReads + fPendingWrites >= 1);

         if (fTarget) {
             SkASSERT(!fPendingReads && !fPendingWrites);
             // The backing GrSurface can have more refs than the proxy if the proxy
             // started off wrapping an external resource (that came in with refs).
             // The GrSurface should never have fewer refs than the proxy however.
             SkASSERT(fTarget->fRefCnt >= fRefCnt);
         }
 #endif
     }

     int32_t getProxyRefCnt_TestOnly() const;
     int32_t getBackingRefCnt_TestOnly() const;
     int32_t getPendingReadCnt_TestOnly() const;
     int32_t getPendingWriteCnt_TestOnly() const;

 protected:
     GrIORefProxy() : fTarget(nullptr), fRefCnt(1), fPendingReads(0), fPendingWrites(0) {}
     GrIORefProxy(sk_sp<GrSurface> surface) : fRefCnt(1), fPendingReads(0), fPendingWrites(0) {
         // Since we're manually forwarding on refs & unrefs we don't want sk_sp doing
         // anything extra.
         fTarget = surface.release();
     }
     virtual ~GrIORefProxy() {
         // We don't unref 'fTarget' here since the 'unref' method will already
         // have forwarded on the unref call that got use here.
     }

     // This GrIORefProxy was deferred before but has just been instantiated. To
     // make all the reffing & unreffing work out we now need to transfer any deferred
     // refs & unrefs to the new GrSurface
     void transferRefs() {
         SkASSERT(fTarget);

         fTarget->fRefCnt += (fRefCnt-1); // don't xfer the proxy's creation ref
         fTarget->fPendingReads += fPendingReads;
         fTarget->fPendingWrites += fPendingWrites;

         fPendingReads = 0;
         fPendingWrites = 0;
     }

     bool internalHasPendingIO() const {
         if (fTarget) {
             return fTarget->internalHasPendingIO();
         }

         return SkToBool(fPendingWrites | fPendingReads);
     }

     bool internalHasPendingWrite() const {
         if (fTarget) {
             return fTarget->internalHasPendingWrite();
         }

         return SkToBool(fPendingWrites);
     }

     // For deferred proxies this will be null. For wrapped proxies it will point to the
     // wrapped resource.
     GrSurface* fTarget;

 private:
     // This class is used to manage conversion of refs to pending reads/writes.
     friend class GrGpuResourceRef;
     template <typename, GrIOType> friend class GrPendingIOResource;

     void addPendingRead() const {
         this->validate();

         if (fTarget) {
             fTarget->addPendingRead();
             return;
         }

         ++fPendingReads;
     }

     void completedRead() const {
         this->validate();

         if (fTarget) {
             fTarget->completedRead();
             return;
         }

         SkFAIL("How was the read completed if the Proxy hasn't been instantiated?");
     }

     void addPendingWrite() const {
         this->validate();

         if (fTarget) {
             fTarget->addPendingWrite();
             return;
         }

         ++fPendingWrites;
     }

     void completedWrite() const {
         this->validate();

         if (fTarget) {
             fTarget->completedWrite();
             return;
         }

         SkFAIL("How was the write completed if the Proxy hasn't been instantiated?");
     }

     mutable int32_t fRefCnt;
     mutable int32_t fPendingReads;
     mutable int32_t fPendingWrites;
 };

 class GrSurfaceProxy : public GrIORefProxy {
 public:
     static sk_sp<GrSurfaceProxy> MakeWrapped(sk_sp<GrSurface>);
     static sk_sp<GrTextureProxy> MakeWrapped(sk_sp<GrTexture>);

     static sk_sp<GrTextureProxy> MakeDeferred(GrResourceProvider*,
                                               const GrSurfaceDesc&, SkBackingFit,
                                               SkBudgeted, uint32_t flags = 0);

     // TODO: need to refine ownership semantics of 'srcData' if we're in completely
     // deferred mode
     static sk_sp<GrTextureProxy> MakeDeferred(GrResourceProvider*,
                                               const GrSurfaceDesc&, SkBudgeted,
                                               const void* srcData, size_t rowBytes);

     static sk_sp<GrTextureProxy> MakeWrappedBackend(GrContext*, GrBackendTexture&, GrSurfaceOrigin);

     const GrSurfaceDesc& desc() const { return fDesc; }

     GrSurfaceOrigin origin() const {
         SkASSERT(kTopLeft_GrSurfaceOrigin == fDesc.fOrigin ||
                  kBottomLeft_GrSurfaceOrigin == fDesc.fOrigin);
         return fDesc.fOrigin;
     }
     int width() const { return fDesc.fWidth; }
     int height() const { return fDesc.fHeight; }
     GrPixelConfig config() const { return fDesc.fConfig; }

     class UniqueID {
     public:
         static UniqueID InvalidID() {
             return UniqueID(uint32_t(SK_InvalidUniqueID));
         }

         // wrapped
         explicit UniqueID(const GrGpuResource::UniqueID& id) : fID(id.asUInt()) { }
         // deferred
         UniqueID() : fID(GrGpuResource::CreateUniqueID()) { }

         uint32_t asUInt() const { return fID; }

         bool operator==(const UniqueID& other) const {
             return fID == other.fID;
         }
         bool operator!=(const UniqueID& other) const {
             return !(*this == other);
         }

         void makeInvalid() { fID = SK_InvalidUniqueID; }
         bool isInvalid() const { return SK_InvalidUniqueID == fID; }

     private:
         explicit UniqueID(uint32_t id) : fID(id) {}

         uint32_t fID;
     };

     /*
      * The contract for the uniqueID is:
      *   for wrapped resources:
      *      the uniqueID will match that of the wrapped resource
      *
      *   for deferred resources:
      *      the uniqueID will be different from the real resource, when it is allocated
      *      the proxy's uniqueID will not change across the instantiate call
      *
      *    the uniqueIDs of the proxies and the resources draw from the same pool
      *
      * What this boils down to is that the uniqueID of a proxy can be used to consistently
      * track/identify a proxy but should never be used to distinguish between
      * resources and proxies - beware!
      */
     UniqueID uniqueID() const { return fUniqueID; }

     GrSurface* instantiate(GrResourceProvider* resourceProvider);

     /**
      * Helper that gets the width and height of the surface as a bounding rectangle.
      */
     SkRect getBoundsRect() const { return SkRect::MakeIWH(this->width(), this->height()); }

     int worstCaseWidth(const GrCaps& caps) const;
     int worstCaseHeight(const GrCaps& caps) const;

     /**
      * @return the texture proxy associated with the surface proxy, may be NULL.
      */
     virtual GrTextureProxy* asTextureProxy() { return nullptr; }
     virtual const GrTextureProxy* asTextureProxy() const { return nullptr; }

     /**
      * @return the render target proxy associated with the surface proxy, may be NULL.
      */
     virtual GrRenderTargetProxy* asRenderTargetProxy() { return nullptr; }
     virtual const GrRenderTargetProxy* asRenderTargetProxy() const { return nullptr; }

     /**
      * Does the resource count against the resource budget?
      */
     SkBudgeted isBudgeted() const { return fBudgeted; }

     void setLastOpList(GrOpList* opList);
     GrOpList* getLastOpList() { return fLastOpList; }

     GrRenderTargetOpList* getLastRenderTargetOpList();
     GrTextureOpList* getLastTextureOpList();

     /**
      * Retrieves the amount of GPU memory that will be or currently is used by this resource
      * in bytes. It is approximate since we aren't aware of additional padding or copies made
      * by the driver.
      *
      * @return the amount of GPU memory used in bytes
      */
     size_t gpuMemorySize() const {
         if (kInvalidGpuMemorySize == fGpuMemorySize) {
             fGpuMemorySize = this->onGpuMemorySize();
             SkASSERT(kInvalidGpuMemorySize != fGpuMemorySize);
         }
         return fGpuMemorySize;
     }

     // Helper function that creates a temporary SurfaceContext to perform the copy
     // It always returns a kExact-backed proxy bc it is used when converting an SkSpecialImage
     // to an SkImage.
     static sk_sp<GrTextureProxy> Copy(GrContext*, GrSurfaceProxy* src,
                                       SkIRect srcRect, SkBudgeted);

     // Copy the entire 'src'
     // It always returns a kExact-backed proxy bc it is used in SkGpuDevice::snapSpecial
     static sk_sp<GrTextureProxy> Copy(GrContext* context, GrSurfaceProxy* src,
                                       SkBudgeted budgeted);

     // Test-only entry point - should decrease in use as proxies propagate
     static sk_sp<GrSurfaceContext> TestCopy(GrContext* context, const GrSurfaceDesc& dstDesc,
                                             GrSurfaceProxy* srcProxy);

     bool isWrapped_ForTesting() const;

     SkDEBUGCODE(void validate(GrContext*) const;)

     // Provides access to functions that aren't part of the public API.
     GrSurfaceProxyPriv priv();
     const GrSurfaceProxyPriv priv() const;

 protected:
     // Deferred version
     GrSurfaceProxy(const GrSurfaceDesc& desc, SkBackingFit fit, SkBudgeted budgeted, uint32_t flags)
         : fDesc(desc)
         , fFit(fit)
         , fBudgeted(budgeted)
         , fFlags(flags)
         // fMipColorMode is only valid for texturable proxies
         , fMipColorMode(SkDestinationSurfaceColorMode::kLegacy)
         , fGpuMemorySize(kInvalidGpuMemorySize)
         , fLastOpList(nullptr) {
         // Note: this ctor pulls a new uniqueID from the same pool at the GrGpuResources
     }

     // Wrapped version
     GrSurfaceProxy(sk_sp<GrSurface> surface, SkBackingFit fit);

     virtual ~GrSurfaceProxy();

     friend class GrSurfaceProxyPriv;

     // Methods made available via GrSurfaceProxyPriv
     bool hasPendingIO() const {
         return this->internalHasPendingIO();
     }

     bool hasPendingWrite() const {
         return this->internalHasPendingWrite();
     }

     // For wrapped resources, 'fDesc' will always be filled in from the wrapped resource.
     GrSurfaceDesc        fDesc;
     SkBackingFit         fFit;      // always exact for wrapped resources
     mutable SkBudgeted   fBudgeted; // set from the backing resource for wrapped resources
                                     // mutable bc of SkSurface/SkImage wishy-washiness
     const uint32_t       fFlags;

     SkDestinationSurfaceColorMode fMipColorMode;

     const UniqueID       fUniqueID; // set from the backing resource for wrapped resources

     static const size_t kInvalidGpuMemorySize = ~static_cast<size_t>(0);
     SkDEBUGCODE(size_t getRawGpuMemorySize_debugOnly() const { return fGpuMemorySize; })

 private:
     virtual size_t onGpuMemorySize() const = 0;

     // This entry is lazily evaluated so, when the proxy wraps a resource, the resource
     // will be called but, when the proxy is deferred, it will compute the answer itself.
     // If the proxy computes its own answer that answer is checked (in debug mode) in
     // the instantiation method.
     mutable size_t      fGpuMemorySize;

     // The last opList that wrote to or is currently going to write to this surface
     // The opList can be closed (e.g., no render target context is currently bound
     // to this renderTarget).
     // This back-pointer is required so that we can add a dependancy between
     // the opList used to create the current contents of this surface
     // and the opList of a destination surface to which this one is being drawn or copied.
     GrOpList* fLastOpList;


     typedef GrIORefProxy INHERITED;
 };

 #endif
	/*
	* Copyright 2016 Google Inc.
	*
	* Use of this source code is governed by a BSD-style license that can be
	* found in the LICENSE file.
	*/

	#ifndef GrSurfaceProxy_DEFINED
	#define GrSurfaceProxy_DEFINED

	#include "GrGpuResource.h"
	#include "GrSurface.h"

	#include "SkRect.h"

	class GrBackendTexture;
	class GrCaps;
	class GrOpList;
	class GrRenderTargetOpList;
	class GrRenderTargetProxy;
	class GrResourceProvider;
	class GrSurfaceContext;
	class GrSurfaceProxyPriv;
	class GrTextureOpList;
	class GrTextureProxy;

	// This class replicates the functionality GrIORef<GrSurface> but tracks the
	// utilitization for later resource allocation (for the deferred case) and
	// forwards on the utilization in the wrapped case
	class GrIORefProxy : public SkNoncopyable {
	public:
	void ref() const {
	this->validate();

	++fRefCnt;
	if (fTarget) {
	fTarget->ref();
	}
	}

	void unref() const {
	this->validate();

	if (fTarget) {
	fTarget->unref();
	}

	if (!(--fRefCnt)) {
	delete this;
	return;
	}

	this->validate();
	}

	void validate() const {
	#ifdef SK_DEBUG
	SkASSERT(fRefCnt >= 1);
	SkASSERT(fPendingReads >= 0);
	SkASSERT(fPendingWrites >= 0);
	SkASSERT(fRefCnt + fPendingReads + fPendingWrites >= 1);

	if (fTarget) {
	SkASSERT(!fPendingReads && !fPendingWrites);
	// The backing GrSurface can have more refs than the proxy if the proxy
	// started off wrapping an external resource (that came in with refs).
	// The GrSurface should never have fewer refs than the proxy however.
	SkASSERT(fTarget->fRefCnt >= fRefCnt);
	}
	#endif
	}

	int32_t getProxyRefCnt_TestOnly() const;
	int32_t getBackingRefCnt_TestOnly() const;
	int32_t getPendingReadCnt_TestOnly() const;
	int32_t getPendingWriteCnt_TestOnly() const;

	protected:
	GrIORefProxy() : fTarget(nullptr), fRefCnt(1), fPendingReads(0), fPendingWrites(0) {}
	GrIORefProxy(sk_sp<GrSurface> surface) : fRefCnt(1), fPendingReads(0), fPendingWrites(0) {
	// Since we're manually forwarding on refs & unrefs we don't want sk_sp doing
	// anything extra.
	fTarget = surface.release();
	}
	virtual ~GrIORefProxy() {
	// We don't unref 'fTarget' here since the 'unref' method will already
	// have forwarded on the unref call that got use here.
	}

	// This GrIORefProxy was deferred before but has just been instantiated. To
	// make all the reffing & unreffing work out we now need to transfer any deferred
	// refs & unrefs to the new GrSurface
	void transferRefs() {
	SkASSERT(fTarget);

	fTarget->fRefCnt += (fRefCnt-1); // don't xfer the proxy's creation ref
	fTarget->fPendingReads += fPendingReads;
	fTarget->fPendingWrites += fPendingWrites;

	fPendingReads = 0;
	fPendingWrites = 0;
	}

	bool internalHasPendingIO() const {
	if (fTarget) {
	return fTarget->internalHasPendingIO();
	}

	return SkToBool(fPendingWrites \| fPendingReads);
	}

	bool internalHasPendingWrite() const {
	if (fTarget) {
	return fTarget->internalHasPendingWrite();
	}

	return SkToBool(fPendingWrites);
	}

	// For deferred proxies this will be null. For wrapped proxies it will point to the
	// wrapped resource.
	GrSurface* fTarget;

	private:
	// This class is used to manage conversion of refs to pending reads/writes.
	friend class GrGpuResourceRef;
	template <typename, GrIOType> friend class GrPendingIOResource;

	void addPendingRead() const {
	this->validate();

	if (fTarget) {
	fTarget->addPendingRead();
	return;
	}

	++fPendingReads;
	}

	void completedRead() const {
	this->validate();

	if (fTarget) {
	fTarget->completedRead();
	return;
	}

	SkFAIL("How was the read completed if the Proxy hasn't been instantiated?");
	}

	void addPendingWrite() const {
	this->validate();

	if (fTarget) {
	fTarget->addPendingWrite();
	return;
	}

	++fPendingWrites;
	}

	void completedWrite() const {
	this->validate();

	if (fTarget) {
	fTarget->completedWrite();
	return;
	}

	SkFAIL("How was the write completed if the Proxy hasn't been instantiated?");
	}

	mutable int32_t fRefCnt;
	mutable int32_t fPendingReads;
	mutable int32_t fPendingWrites;
	};

	class GrSurfaceProxy : public GrIORefProxy {
	public:
	static sk_sp<GrSurfaceProxy> MakeWrapped(sk_sp<GrSurface>);
	static sk_sp<GrTextureProxy> MakeWrapped(sk_sp<GrTexture>);

	static sk_sp<GrTextureProxy> MakeDeferred(GrResourceProvider*,
	const GrSurfaceDesc&, SkBackingFit,
	SkBudgeted, uint32_t flags = 0);

	// TODO: need to refine ownership semantics of 'srcData' if we're in completely
	// deferred mode
	static sk_sp<GrTextureProxy> MakeDeferred(GrResourceProvider*,
	const GrSurfaceDesc&, SkBudgeted,
	const void* srcData, size_t rowBytes);

	static sk_sp<GrTextureProxy> MakeWrappedBackend(GrContext*, GrBackendTexture&, GrSurfaceOrigin);

	const GrSurfaceDesc& desc() const { return fDesc; }

	GrSurfaceOrigin origin() const {
	SkASSERT(kTopLeft_GrSurfaceOrigin == fDesc.fOrigin \|\|
	kBottomLeft_GrSurfaceOrigin == fDesc.fOrigin);
	return fDesc.fOrigin;
	}
	int width() const { return fDesc.fWidth; }
	int height() const { return fDesc.fHeight; }
	GrPixelConfig config() const { return fDesc.fConfig; }

	class UniqueID {
	public:
	static UniqueID InvalidID() {
	return UniqueID(uint32_t(SK_InvalidUniqueID));
	}

	// wrapped
	explicit UniqueID(const GrGpuResource::UniqueID& id) : fID(id.asUInt()) { }
	// deferred
	UniqueID() : fID(GrGpuResource::CreateUniqueID()) { }

	uint32_t asUInt() const { return fID; }

	bool operator==(const UniqueID& other) const {
	return fID == other.fID;
	}
	bool operator!=(const UniqueID& other) const {
	return !(*this == other);
	}

	void makeInvalid() { fID = SK_InvalidUniqueID; }
	bool isInvalid() const { return SK_InvalidUniqueID == fID; }

	private:
	explicit UniqueID(uint32_t id) : fID(id) {}

	uint32_t fID;
	};

	/*
	* The contract for the uniqueID is:
	* for wrapped resources:
	* the uniqueID will match that of the wrapped resource
	*
	* for deferred resources:
	* the uniqueID will be different from the real resource, when it is allocated
	* the proxy's uniqueID will not change across the instantiate call
	*
	* the uniqueIDs of the proxies and the resources draw from the same pool
	*
	* What this boils down to is that the uniqueID of a proxy can be used to consistently
	* track/identify a proxy but should never be used to distinguish between
	* resources and proxies - beware!
	*/
	UniqueID uniqueID() const { return fUniqueID; }

	GrSurface* instantiate(GrResourceProvider* resourceProvider);

	/**
	* Helper that gets the width and height of the surface as a bounding rectangle.
	*/
	SkRect getBoundsRect() const { return SkRect::MakeIWH(this->width(), this->height()); }

	int worstCaseWidth(const GrCaps& caps) const;
	int worstCaseHeight(const GrCaps& caps) const;

	/**
	* @return the texture proxy associated with the surface proxy, may be NULL.
	*/
	virtual GrTextureProxy* asTextureProxy() { return nullptr; }
	virtual const GrTextureProxy* asTextureProxy() const { return nullptr; }

	/**
	* @return the render target proxy associated with the surface proxy, may be NULL.
	*/
	virtual GrRenderTargetProxy* asRenderTargetProxy() { return nullptr; }
	virtual const GrRenderTargetProxy* asRenderTargetProxy() const { return nullptr; }

	/**
	* Does the resource count against the resource budget?
	*/
	SkBudgeted isBudgeted() const { return fBudgeted; }

	void setLastOpList(GrOpList* opList);
	GrOpList* getLastOpList() { return fLastOpList; }

	GrRenderTargetOpList* getLastRenderTargetOpList();
	GrTextureOpList* getLastTextureOpList();

	/**
	* Retrieves the amount of GPU memory that will be or currently is used by this resource
	* in bytes. It is approximate since we aren't aware of additional padding or copies made
	* by the driver.
	*
	* @return the amount of GPU memory used in bytes
	*/
	size_t gpuMemorySize() const {
	if (kInvalidGpuMemorySize == fGpuMemorySize) {
	fGpuMemorySize = this->onGpuMemorySize();
	SkASSERT(kInvalidGpuMemorySize != fGpuMemorySize);
	}
	return fGpuMemorySize;
	}

	// Helper function that creates a temporary SurfaceContext to perform the copy
	// It always returns a kExact-backed proxy bc it is used when converting an SkSpecialImage
	// to an SkImage.
	static sk_sp<GrTextureProxy> Copy(GrContext, GrSurfaceProxy src,
	SkIRect srcRect, SkBudgeted);

	// Copy the entire 'src'
	// It always returns a kExact-backed proxy bc it is used in SkGpuDevice::snapSpecial
	static sk_sp<GrTextureProxy> Copy(GrContext* context, GrSurfaceProxy* src,
	SkBudgeted budgeted);

	// Test-only entry point - should decrease in use as proxies propagate
	static sk_sp<GrSurfaceContext> TestCopy(GrContext* context, const GrSurfaceDesc& dstDesc,
	GrSurfaceProxy* srcProxy);

	bool isWrapped_ForTesting() const;

	SkDEBUGCODE(void validate(GrContext*) const;)

	// Provides access to functions that aren't part of the public API.
	GrSurfaceProxyPriv priv();
	const GrSurfaceProxyPriv priv() const;

	protected:
	// Deferred version
	GrSurfaceProxy(const GrSurfaceDesc& desc, SkBackingFit fit, SkBudgeted budgeted, uint32_t flags)
	: fDesc(desc)
	, fFit(fit)
	, fBudgeted(budgeted)
	, fFlags(flags)
	// fMipColorMode is only valid for texturable proxies
	, fMipColorMode(SkDestinationSurfaceColorMode::kLegacy)
	, fGpuMemorySize(kInvalidGpuMemorySize)
	, fLastOpList(nullptr) {
	// Note: this ctor pulls a new uniqueID from the same pool at the GrGpuResources
	}

	// Wrapped version
	GrSurfaceProxy(sk_sp<GrSurface> surface, SkBackingFit fit);

	virtual ~GrSurfaceProxy();

	friend class GrSurfaceProxyPriv;

	// Methods made available via GrSurfaceProxyPriv
	bool hasPendingIO() const {
	return this->internalHasPendingIO();
	}

	bool hasPendingWrite() const {
	return this->internalHasPendingWrite();
	}

	// For wrapped resources, 'fDesc' will always be filled in from the wrapped resource.
	GrSurfaceDesc fDesc;
	SkBackingFit fFit; // always exact for wrapped resources
	mutable SkBudgeted fBudgeted; // set from the backing resource for wrapped resources
	// mutable bc of SkSurface/SkImage wishy-washiness
	const uint32_t fFlags;

	SkDestinationSurfaceColorMode fMipColorMode;

	const UniqueID fUniqueID; // set from the backing resource for wrapped resources

	static const size_t kInvalidGpuMemorySize = ~static_cast<size_t>(0);
	SkDEBUGCODE(size_t getRawGpuMemorySize_debugOnly() const { return fGpuMemorySize; })

	private:
	virtual size_t onGpuMemorySize() const = 0;

	// This entry is lazily evaluated so, when the proxy wraps a resource, the resource
	// will be called but, when the proxy is deferred, it will compute the answer itself.
	// If the proxy computes its own answer that answer is checked (in debug mode) in
	// the instantiation method.
	mutable size_t fGpuMemorySize;

	// The last opList that wrote to or is currently going to write to this surface
	// The opList can be closed (e.g., no render target context is currently bound
	// to this renderTarget).
	// This back-pointer is required so that we can add a dependancy between
	// the opList used to create the current contents of this surface
	// and the opList of a destination surface to which this one is being drawn or copied.
	GrOpList* fLastOpList;


	typedef GrIORefProxy INHERITED;
	};

	#endif