1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
|
/* -*- Mode: C++; tab-width: 4; indent-tabs-mode: nil; c-basic-offset: 2 -*- */
/* This Source Code Form is subject to the terms of the Mozilla Public
* License, v. 2.0. If a copy of the MPL was not distributed with this
* file, You can obtain one at http://mozilla.org/MPL/2.0/. */
#ifndef GPU_OBJECT_MODEL_H_
#define GPU_OBJECT_MODEL_H_
#include "nsString.h"
#include "nsWrapperCache.h"
class nsIGlobalObject;
namespace mozilla::webgpu {
class WebGPUChild;
template <typename T>
class ChildOf {
protected:
explicit ChildOf(T* const parent);
virtual ~ChildOf();
RefPtr<T> mParent;
public:
nsIGlobalObject* GetParentObject() const;
};
/// Most WebGPU DOM objects inherit from this class.
///
/// mValid should only be used in the destruction steps in Cleanup() to check
/// whether they have already run. This is because the destruction steps can be
/// triggered by either the object's destructor or the cycle collector
/// attempting to break a cycle. As a result, all methods accessible from JS can
/// assume that mValid is true.
///
/// Similarly, pointers to the device and the IPDL actor (bridge) can be assumed
/// to be non-null whenever the object is accessible from JS but not during
/// cleanup as they might have been snatched by cycle collection.
///
/// The general pattern is that all objects should implement Cleanup more or
/// less the same way. Cleanup should be the only function sending the
/// corresponding Drop message and cleanup should *never* be called by anything
/// other than the object destructor or the cycle collector.
///
/// These rules guarantee that:
/// - The Drop message is called only once and that no other IPC message
/// referring
/// to the same object is send after Drop.
/// - Any method outside of the destruction sequence can assume the pointers are
/// non-null. They only have to check that the IPDL actor can send messages
/// using `WebGPUChild::CanSend()`.
class ObjectBase : public nsWrapperCache {
protected:
virtual ~ObjectBase() = default;
/// False during the destruction sequence of the object. This is not the same
/// thing as an object's valid state in the WebGPU spec. In the case of
/// render/compute pass encoders, this can also be false if the pass encoder
/// was created by calling beginRenderPass/beginComputePass on a command
/// encoder in the "ended" state (i.e. the pass was never valid).
bool mValid = true;
public:
void GetLabel(nsAString& aValue) const;
void SetLabel(const nsAString& aLabel);
auto CLabel() const { return NS_ConvertUTF16toUTF8(mLabel); }
protected:
// Object label, initialized from GPUObjectDescriptorBase.label.
nsString mLabel;
};
} // namespace mozilla::webgpu
#define GPU_DECL_JS_WRAP(T) \
JSObject* WrapObject(JSContext* cx, JS::Handle<JSObject*> givenProto) \
override;
#define GPU_DECL_CYCLE_COLLECTION(T) \
NS_DECL_CYCLE_COLLECTION_NATIVE_WRAPPERCACHE_CLASS(T) \
NS_INLINE_DECL_CYCLE_COLLECTING_NATIVE_REFCOUNTING(T)
#define GPU_IMPL_JS_WRAP(T) \
JSObject* T::WrapObject(JSContext* cx, JS::Handle<JSObject*> givenProto) { \
return dom::GPU##T##_Binding::Wrap(cx, this, givenProto); \
}
// Note: we don't use `NS_IMPL_CYCLE_COLLECTION_WRAPPERCACHE` directly
// because there is a custom action we need to always do.
#define GPU_IMPL_CYCLE_COLLECTION_WRAPPERCACHE(T, ...) \
NS_IMPL_CYCLE_COLLECTION_WRAPPERCACHE_CLASS(T) \
NS_IMPL_CYCLE_COLLECTION_UNLINK_BEGIN(T) \
tmp->Cleanup(); \
NS_IMPL_CYCLE_COLLECTION_UNLINK(__VA_ARGS__) \
NS_IMPL_CYCLE_COLLECTION_UNLINK_PRESERVED_WRAPPER \
NS_IMPL_CYCLE_COLLECTION_UNLINK_END \
NS_IMPL_CYCLE_COLLECTION_TRAVERSE_BEGIN(T) \
NS_IMPL_CYCLE_COLLECTION_TRAVERSE(__VA_ARGS__) \
NS_IMPL_CYCLE_COLLECTION_TRAVERSE_END
#define GPU_IMPL_CYCLE_COLLECTION_WRAPPERCACHE_WEAK_PTR(T, ...) \
NS_IMPL_CYCLE_COLLECTION_WRAPPERCACHE_CLASS(T) \
NS_IMPL_CYCLE_COLLECTION_UNLINK_BEGIN(T) \
tmp->Cleanup(); \
NS_IMPL_CYCLE_COLLECTION_UNLINK(__VA_ARGS__) \
NS_IMPL_CYCLE_COLLECTION_UNLINK_PRESERVED_WRAPPER \
NS_IMPL_CYCLE_COLLECTION_UNLINK_WEAK_PTR \
NS_IMPL_CYCLE_COLLECTION_UNLINK_END \
NS_IMPL_CYCLE_COLLECTION_TRAVERSE_BEGIN(T) \
NS_IMPL_CYCLE_COLLECTION_TRAVERSE(__VA_ARGS__) \
NS_IMPL_CYCLE_COLLECTION_TRAVERSE_END
#define GPU_IMPL_CYCLE_COLLECTION_WRAPPERCACHE_INHERITED(T, P, ...) \
NS_IMPL_CYCLE_COLLECTION_CLASS(T) \
NS_IMPL_CYCLE_COLLECTION_UNLINK_BEGIN_INHERITED(T, P) \
tmp->Cleanup(); \
NS_IMPL_CYCLE_COLLECTION_UNLINK(__VA_ARGS__) \
NS_IMPL_CYCLE_COLLECTION_UNLINK_PRESERVED_WRAPPER \
NS_IMPL_CYCLE_COLLECTION_UNLINK_END \
NS_IMPL_CYCLE_COLLECTION_TRAVERSE_BEGIN_INHERITED(T, P) \
NS_IMPL_CYCLE_COLLECTION_TRAVERSE(__VA_ARGS__) \
NS_IMPL_CYCLE_COLLECTION_TRAVERSE_END
#define GPU_IMPL_CYCLE_COLLECTION(T, ...) \
GPU_IMPL_CYCLE_COLLECTION_WRAPPERCACHE(T, __VA_ARGS__)
template <typename T>
void ImplCycleCollectionTraverse(nsCycleCollectionTraversalCallback& callback,
nsTArray<RefPtr<const T>>& field,
const char* name, uint32_t flags) {
for (auto& element : field) {
CycleCollectionNoteChild(callback, const_cast<T*>(element.get()), name,
flags);
}
}
template <typename T>
void ImplCycleCollectionUnlink(nsTArray<RefPtr<const T>>& field) {
for (auto& element : field) {
ImplCycleCollectionUnlink(element);
}
field.Clear();
}
#endif // GPU_OBJECT_MODEL_H_
|
